diff options
Diffstat (limited to 'kernel')
193 files changed, 47826 insertions, 18833 deletions
diff --git a/kernel/Kconfig.locks b/kernel/Kconfig.locks new file mode 100644 index 00000000000..88c92fb4461 --- /dev/null +++ b/kernel/Kconfig.locks @@ -0,0 +1,202 @@ +# +# The ARCH_INLINE foo is necessary because select ignores "depends on" +# +config ARCH_INLINE_SPIN_TRYLOCK + bool + +config ARCH_INLINE_SPIN_TRYLOCK_BH + bool + +config ARCH_INLINE_SPIN_LOCK + bool + +config ARCH_INLINE_SPIN_LOCK_BH + bool + +config ARCH_INLINE_SPIN_LOCK_IRQ + bool + +config ARCH_INLINE_SPIN_LOCK_IRQSAVE + bool + +config ARCH_INLINE_SPIN_UNLOCK + bool + +config ARCH_INLINE_SPIN_UNLOCK_BH + bool + +config ARCH_INLINE_SPIN_UNLOCK_IRQ + bool + +config ARCH_INLINE_SPIN_UNLOCK_IRQRESTORE + bool + + +config ARCH_INLINE_READ_TRYLOCK + bool + +config ARCH_INLINE_READ_LOCK + bool + +config ARCH_INLINE_READ_LOCK_BH + bool + +config ARCH_INLINE_READ_LOCK_IRQ + bool + +config ARCH_INLINE_READ_LOCK_IRQSAVE + bool + +config ARCH_INLINE_READ_UNLOCK + bool + +config ARCH_INLINE_READ_UNLOCK_BH + bool + +config ARCH_INLINE_READ_UNLOCK_IRQ + bool + +config ARCH_INLINE_READ_UNLOCK_IRQRESTORE + bool + + +config ARCH_INLINE_WRITE_TRYLOCK + bool + +config ARCH_INLINE_WRITE_LOCK + bool + +config ARCH_INLINE_WRITE_LOCK_BH + bool + +config ARCH_INLINE_WRITE_LOCK_IRQ + bool + +config ARCH_INLINE_WRITE_LOCK_IRQSAVE + bool + +config ARCH_INLINE_WRITE_UNLOCK + bool + +config ARCH_INLINE_WRITE_UNLOCK_BH + bool + +config ARCH_INLINE_WRITE_UNLOCK_IRQ + bool + +config ARCH_INLINE_WRITE_UNLOCK_IRQRESTORE + bool + +# +# lock_* functions are inlined when: +# - DEBUG_SPINLOCK=n and GENERIC_LOCKBREAK=n and ARCH_INLINE_*LOCK=y +# +# trylock_* functions are inlined when: +# - DEBUG_SPINLOCK=n and ARCH_INLINE_*LOCK=y +# +# unlock and unlock_irq functions are inlined when: +# - DEBUG_SPINLOCK=n and ARCH_INLINE_*LOCK=y +# or +# - DEBUG_SPINLOCK=n and PREEMPT=n +# +# unlock_bh and unlock_irqrestore functions are inlined when: +# - DEBUG_SPINLOCK=n and ARCH_INLINE_*LOCK=y +# + +config INLINE_SPIN_TRYLOCK + def_bool !DEBUG_SPINLOCK && ARCH_INLINE_SPIN_TRYLOCK + +config INLINE_SPIN_TRYLOCK_BH + def_bool !DEBUG_SPINLOCK && ARCH_INLINE_SPIN_TRYLOCK_BH + +config INLINE_SPIN_LOCK + def_bool !DEBUG_SPINLOCK && !GENERIC_LOCKBREAK && ARCH_INLINE_SPIN_LOCK + +config INLINE_SPIN_LOCK_BH + def_bool !DEBUG_SPINLOCK && !GENERIC_LOCKBREAK && \ + ARCH_INLINE_SPIN_LOCK_BH + +config INLINE_SPIN_LOCK_IRQ + def_bool !DEBUG_SPINLOCK && !GENERIC_LOCKBREAK && \ + ARCH_INLINE_SPIN_LOCK_IRQ + +config INLINE_SPIN_LOCK_IRQSAVE + def_bool !DEBUG_SPINLOCK && !GENERIC_LOCKBREAK && \ + ARCH_INLINE_SPIN_LOCK_IRQSAVE + +config INLINE_SPIN_UNLOCK + def_bool !DEBUG_SPINLOCK && (!PREEMPT || ARCH_INLINE_SPIN_UNLOCK) + +config INLINE_SPIN_UNLOCK_BH + def_bool !DEBUG_SPINLOCK && ARCH_INLINE_SPIN_UNLOCK_BH + +config INLINE_SPIN_UNLOCK_IRQ + def_bool !DEBUG_SPINLOCK && (!PREEMPT || ARCH_INLINE_SPIN_UNLOCK_BH) + +config INLINE_SPIN_UNLOCK_IRQRESTORE + def_bool !DEBUG_SPINLOCK && ARCH_INLINE_SPIN_UNLOCK_IRQRESTORE + + +config INLINE_READ_TRYLOCK + def_bool !DEBUG_SPINLOCK && ARCH_INLINE_READ_TRYLOCK + +config INLINE_READ_LOCK + def_bool !DEBUG_SPINLOCK && !GENERIC_LOCKBREAK && ARCH_INLINE_READ_LOCK + +config INLINE_READ_LOCK_BH + def_bool !DEBUG_SPINLOCK && !GENERIC_LOCKBREAK && \ + ARCH_INLINE_READ_LOCK_BH + +config INLINE_READ_LOCK_IRQ + def_bool !DEBUG_SPINLOCK && !GENERIC_LOCKBREAK && \ + ARCH_INLINE_READ_LOCK_IRQ + +config INLINE_READ_LOCK_IRQSAVE + def_bool !DEBUG_SPINLOCK && !GENERIC_LOCKBREAK && \ + ARCH_INLINE_READ_LOCK_IRQSAVE + +config INLINE_READ_UNLOCK + def_bool !DEBUG_SPINLOCK && (!PREEMPT || ARCH_INLINE_READ_UNLOCK) + +config INLINE_READ_UNLOCK_BH + def_bool !DEBUG_SPINLOCK && ARCH_INLINE_READ_UNLOCK_BH + +config INLINE_READ_UNLOCK_IRQ + def_bool !DEBUG_SPINLOCK && (!PREEMPT || ARCH_INLINE_READ_UNLOCK_BH) + +config INLINE_READ_UNLOCK_IRQRESTORE + def_bool !DEBUG_SPINLOCK && ARCH_INLINE_READ_UNLOCK_IRQRESTORE + + +config INLINE_WRITE_TRYLOCK + def_bool !DEBUG_SPINLOCK && ARCH_INLINE_WRITE_TRYLOCK + +config INLINE_WRITE_LOCK + def_bool !DEBUG_SPINLOCK && !GENERIC_LOCKBREAK && ARCH_INLINE_WRITE_LOCK + +config INLINE_WRITE_LOCK_BH + def_bool !DEBUG_SPINLOCK && !GENERIC_LOCKBREAK && \ + ARCH_INLINE_WRITE_LOCK_BH + +config INLINE_WRITE_LOCK_IRQ + def_bool !DEBUG_SPINLOCK && !GENERIC_LOCKBREAK && \ + ARCH_INLINE_WRITE_LOCK_IRQ + +config INLINE_WRITE_LOCK_IRQSAVE + def_bool !DEBUG_SPINLOCK && !GENERIC_LOCKBREAK && \ + ARCH_INLINE_WRITE_LOCK_IRQSAVE + +config INLINE_WRITE_UNLOCK + def_bool !DEBUG_SPINLOCK && (!PREEMPT || ARCH_INLINE_WRITE_UNLOCK) + +config INLINE_WRITE_UNLOCK_BH + def_bool !DEBUG_SPINLOCK && ARCH_INLINE_WRITE_UNLOCK_BH + +config INLINE_WRITE_UNLOCK_IRQ + def_bool !DEBUG_SPINLOCK && (!PREEMPT || ARCH_INLINE_WRITE_UNLOCK_BH) + +config INLINE_WRITE_UNLOCK_IRQRESTORE + def_bool !DEBUG_SPINLOCK && ARCH_INLINE_WRITE_UNLOCK_IRQRESTORE + +config MUTEX_SPIN_ON_OWNER + def_bool SMP && !DEBUG_MUTEXES && !HAVE_DEFAULT_NO_SPIN_MUTEXES diff --git a/kernel/Makefile b/kernel/Makefile index e4791b3ba55..864ff75d65f 100644 --- a/kernel/Makefile +++ b/kernel/Makefile @@ -4,13 +4,14 @@ obj-y = sched.o fork.o exec_domain.o panic.o printk.o \ cpu.o exit.o itimer.o time.o softirq.o resource.o \ - sysctl.o capability.o ptrace.o timer.o user.o \ + sysctl.o sysctl_binary.o capability.o ptrace.o timer.o user.o \ signal.o sys.o kmod.o workqueue.o pid.o \ rcupdate.o extable.o params.o posix-timers.o \ kthread.o wait.o kfifo.o sys_ni.o posix-cpu-timers.o mutex.o \ hrtimer.o rwsem.o nsproxy.o srcu.o semaphore.o \ notifier.o ksysfs.o pm_qos_params.o sched_clock.o cred.o \ async.o +obj-y += groups.o ifdef CONFIG_FUNCTION_TRACER # Do not trace debug files and internal ftrace files @@ -20,6 +21,7 @@ CFLAGS_REMOVE_mutex-debug.o = -pg CFLAGS_REMOVE_rtmutex-debug.o = -pg CFLAGS_REMOVE_cgroup-debug.o = -pg CFLAGS_REMOVE_sched_clock.o = -pg +CFLAGS_REMOVE_perf_event.o = -pg endif obj-$(CONFIG_FREEZER) += freezer.o @@ -57,7 +59,6 @@ obj-$(CONFIG_KEXEC) += kexec.o obj-$(CONFIG_BACKTRACE_SELF_TEST) += backtracetest.o obj-$(CONFIG_COMPAT) += compat.o obj-$(CONFIG_CGROUPS) += cgroup.o -obj-$(CONFIG_CGROUP_DEBUG) += cgroup_debug.o obj-$(CONFIG_CGROUP_FREEZER) += cgroup_freezer.o obj-$(CONFIG_CPUSETS) += cpuset.o obj-$(CONFIG_CGROUP_NS) += ns_cgroup.o @@ -68,31 +69,37 @@ obj-$(CONFIG_IKCONFIG) += configs.o obj-$(CONFIG_RESOURCE_COUNTERS) += res_counter.o obj-$(CONFIG_STOP_MACHINE) += stop_machine.o obj-$(CONFIG_KPROBES_SANITY_TEST) += test_kprobes.o -obj-$(CONFIG_AUDIT) += audit.o auditfilter.o +obj-$(CONFIG_AUDIT) += audit.o auditfilter.o audit_watch.o obj-$(CONFIG_AUDITSYSCALL) += auditsc.o +obj-$(CONFIG_GCOV_KERNEL) += gcov/ obj-$(CONFIG_AUDIT_TREE) += audit_tree.o obj-$(CONFIG_KPROBES) += kprobes.o obj-$(CONFIG_KGDB) += kgdb.o obj-$(CONFIG_DETECT_SOFTLOCKUP) += softlockup.o +obj-$(CONFIG_DETECT_HUNG_TASK) += hung_task.o obj-$(CONFIG_GENERIC_HARDIRQS) += irq/ obj-$(CONFIG_SECCOMP) += seccomp.o obj-$(CONFIG_RCU_TORTURE_TEST) += rcutorture.o -obj-$(CONFIG_CLASSIC_RCU) += rcuclassic.o obj-$(CONFIG_TREE_RCU) += rcutree.o -obj-$(CONFIG_PREEMPT_RCU) += rcupreempt.o +obj-$(CONFIG_TREE_PREEMPT_RCU) += rcutree.o obj-$(CONFIG_TREE_RCU_TRACE) += rcutree_trace.o -obj-$(CONFIG_PREEMPT_RCU_TRACE) += rcupreempt_trace.o +obj-$(CONFIG_TINY_RCU) += rcutiny.o obj-$(CONFIG_RELAY) += relay.o obj-$(CONFIG_SYSCTL) += utsname_sysctl.o obj-$(CONFIG_TASK_DELAY_ACCT) += delayacct.o obj-$(CONFIG_TASKSTATS) += taskstats.o tsacct.o -obj-$(CONFIG_MARKERS) += marker.o obj-$(CONFIG_TRACEPOINTS) += tracepoint.o obj-$(CONFIG_LATENCYTOP) += latencytop.o -obj-$(CONFIG_HAVE_GENERIC_DMA_COHERENT) += dma-coherent.o obj-$(CONFIG_FUNCTION_TRACER) += trace/ obj-$(CONFIG_TRACING) += trace/ +obj-$(CONFIG_X86_DS) += trace/ +obj-$(CONFIG_RING_BUFFER) += trace/ obj-$(CONFIG_SMP) += sched_cpupri.o +obj-$(CONFIG_SLOW_WORK) += slow-work.o +obj-$(CONFIG_SLOW_WORK_DEBUG) += slow-work-debugfs.o +obj-$(CONFIG_PERF_EVENTS) += perf_event.o +obj-$(CONFIG_HAVE_HW_BREAKPOINT) += hw_breakpoint.o +obj-$(CONFIG_USER_RETURN_NOTIFIER) += user-return-notifier.o ifneq ($(CONFIG_SCHED_OMIT_FRAME_POINTER),y) # According to Alan Modra <alan@linuxcare.com.au>, the -fno-omit-frame-pointer is @@ -112,7 +119,7 @@ $(obj)/config_data.gz: .config FORCE $(call if_changed,gzip) quiet_cmd_ikconfiggz = IKCFG $@ - cmd_ikconfiggz = (echo "static const char kernel_config_data[] = MAGIC_START"; cat $< | scripts/bin2c; echo "MAGIC_END;") > $@ + cmd_ikconfiggz = (echo "static const char kernel_config_data[] __used = MAGIC_START"; cat $< | scripts/bin2c; echo "MAGIC_END;") > $@ targets += config_data.h $(obj)/config_data.h: $(obj)/config_data.gz FORCE $(call if_changed,ikconfiggz) diff --git a/kernel/acct.c b/kernel/acct.c index 7afa3156416..9a4715a2f6b 100644 --- a/kernel/acct.c +++ b/kernel/acct.c @@ -215,6 +215,7 @@ static void acct_file_reopen(struct bsd_acct_struct *acct, struct file *file, static int acct_on(char *name) { struct file *file; + struct vfsmount *mnt; int error; struct pid_namespace *ns; struct bsd_acct_struct *acct = NULL; @@ -256,11 +257,12 @@ static int acct_on(char *name) acct = NULL; } - mnt_pin(file->f_path.mnt); + mnt = file->f_path.mnt; + mnt_pin(mnt); acct_file_reopen(ns->bacct, file, ns); spin_unlock(&acct_lock); - mntput(file->f_path.mnt); /* it's pinned, now give up active reference */ + mntput(mnt); /* it's pinned, now give up active reference */ kfree(acct); return 0; @@ -489,13 +491,17 @@ static void do_acct_process(struct bsd_acct_struct *acct, u64 run_time; struct timespec uptime; struct tty_struct *tty; + const struct cred *orig_cred; + + /* Perform file operations on behalf of whoever enabled accounting */ + orig_cred = override_creds(file->f_cred); /* * First check to see if there is enough free_space to continue * the process accounting system. */ if (!check_free_space(acct, file)) - return; + goto out; /* * Fill the accounting struct with the needed info as recorded @@ -576,6 +582,8 @@ static void do_acct_process(struct bsd_acct_struct *acct, sizeof(acct_t), &file->f_pos); current->signal->rlim[RLIMIT_FSIZE].rlim_cur = flim; set_fs(fs); +out: + revert_creds(orig_cred); } /** diff --git a/kernel/async.c b/kernel/async.c index 968ef9457d4..27235f5de19 100644 --- a/kernel/async.c +++ b/kernel/async.c @@ -92,19 +92,18 @@ extern int initcall_debug; static async_cookie_t __lowest_in_progress(struct list_head *running) { struct async_entry *entry; + if (!list_empty(running)) { entry = list_first_entry(running, struct async_entry, list); return entry->cookie; - } else if (!list_empty(&async_pending)) { - entry = list_first_entry(&async_pending, - struct async_entry, list); - return entry->cookie; - } else { - /* nothing in progress... next_cookie is "infinity" */ - return next_cookie; } + list_for_each_entry(entry, &async_pending, list) + if (entry->running == running) + return entry->cookie; + + return next_cookie; /* "infinity" value */ } static async_cookie_t lowest_in_progress(struct list_head *running) diff --git a/kernel/audit.c b/kernel/audit.c index ce6d8ea3131..5feed232be9 100644 --- a/kernel/audit.c +++ b/kernel/audit.c @@ -115,9 +115,6 @@ static atomic_t audit_lost = ATOMIC_INIT(0); /* The netlink socket. */ static struct sock *audit_sock; -/* Inotify handle. */ -struct inotify_handle *audit_ih; - /* Hash for inode-based rules */ struct list_head audit_inode_hash[AUDIT_INODE_BUCKETS]; @@ -136,7 +133,7 @@ static DECLARE_WAIT_QUEUE_HEAD(kauditd_wait); static DECLARE_WAIT_QUEUE_HEAD(audit_backlog_wait); /* Serialize requests from userspace. */ -static DEFINE_MUTEX(audit_cmd_mutex); +DEFINE_MUTEX(audit_cmd_mutex); /* AUDIT_BUFSIZ is the size of the temporary buffer used for formatting * audit records. Since printk uses a 1024 byte buffer, this buffer @@ -375,6 +372,25 @@ static void audit_hold_skb(struct sk_buff *skb) kfree_skb(skb); } +/* + * For one reason or another this nlh isn't getting delivered to the userspace + * audit daemon, just send it to printk. + */ +static void audit_printk_skb(struct sk_buff *skb) +{ + struct nlmsghdr *nlh = nlmsg_hdr(skb); + char *data = NLMSG_DATA(nlh); + + if (nlh->nlmsg_type != AUDIT_EOE) { + if (printk_ratelimit()) + printk(KERN_NOTICE "type=%d %s\n", nlh->nlmsg_type, data); + else + audit_log_lost("printk limit exceeded\n"); + } + + audit_hold_skb(skb); +} + static void kauditd_send_skb(struct sk_buff *skb) { int err; @@ -427,14 +443,8 @@ static int kauditd_thread(void *dummy) if (skb) { if (audit_pid) kauditd_send_skb(skb); - else { - if (printk_ratelimit()) - printk(KERN_NOTICE "%s\n", skb->data + NLMSG_SPACE(0)); - else - audit_log_lost("printk limit exceeded\n"); - - audit_hold_skb(skb); - } + else + audit_printk_skb(skb); } else { DECLARE_WAITQUEUE(wait, current); set_current_state(TASK_INTERRUPTIBLE); @@ -495,42 +505,25 @@ int audit_send_list(void *_dest) return 0; } -#ifdef CONFIG_AUDIT_TREE -static int prune_tree_thread(void *unused) -{ - mutex_lock(&audit_cmd_mutex); - audit_prune_trees(); - mutex_unlock(&audit_cmd_mutex); - return 0; -} - -void audit_schedule_prune(void) -{ - kthread_run(prune_tree_thread, NULL, "audit_prune_tree"); -} -#endif - struct sk_buff *audit_make_reply(int pid, int seq, int type, int done, int multi, void *payload, int size) { struct sk_buff *skb; struct nlmsghdr *nlh; - int len = NLMSG_SPACE(size); void *data; int flags = multi ? NLM_F_MULTI : 0; int t = done ? NLMSG_DONE : type; - skb = alloc_skb(len, GFP_KERNEL); + skb = nlmsg_new(size, GFP_KERNEL); if (!skb) return NULL; - nlh = NLMSG_PUT(skb, pid, seq, t, size); - nlh->nlmsg_flags = flags; - data = NLMSG_DATA(nlh); + nlh = NLMSG_NEW(skb, pid, seq, t, size, flags); + data = NLMSG_DATA(nlh); memcpy(data, payload, size); return skb; -nlmsg_failure: /* Used by NLMSG_PUT */ +nlmsg_failure: /* Used by NLMSG_NEW */ if (skb) kfree_skb(skb); return NULL; @@ -766,6 +759,9 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh) audit_log_format(ab, " msg="); size = nlmsg_len(nlh); + if (size > 0 && + ((unsigned char *)data)[size - 1] == '\0') + size--; audit_log_n_untrustedstring(ab, data, size); } audit_set_pid(ab, pid); @@ -859,18 +855,24 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh) break; } case AUDIT_SIGNAL_INFO: - err = security_secid_to_secctx(audit_sig_sid, &ctx, &len); - if (err) - return err; + len = 0; + if (audit_sig_sid) { + err = security_secid_to_secctx(audit_sig_sid, &ctx, &len); + if (err) + return err; + } sig_data = kmalloc(sizeof(*sig_data) + len, GFP_KERNEL); if (!sig_data) { - security_release_secctx(ctx, len); + if (audit_sig_sid) + security_release_secctx(ctx, len); return -ENOMEM; } sig_data->uid = audit_sig_uid; sig_data->pid = audit_sig_pid; - memcpy(sig_data->ctx, ctx, len); - security_release_secctx(ctx, len); + if (audit_sig_sid) { + memcpy(sig_data->ctx, ctx, len); + security_release_secctx(ctx, len); + } audit_send_reply(NETLINK_CB(skb).pid, seq, AUDIT_SIGNAL_INFO, 0, 0, sig_data, sizeof(*sig_data) + len); kfree(sig_data); @@ -923,28 +925,29 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh) } /* - * Get message from skb (based on rtnetlink_rcv_skb). Each message is - * processed by audit_receive_msg. Malformed skbs with wrong length are - * discarded silently. + * Get message from skb. Each message is processed by audit_receive_msg. + * Malformed skbs with wrong length are discarded silently. */ static void audit_receive_skb(struct sk_buff *skb) { - int err; - struct nlmsghdr *nlh; - u32 rlen; + struct nlmsghdr *nlh; + /* + * len MUST be signed for NLMSG_NEXT to be able to dec it below 0 + * if the nlmsg_len was not aligned + */ + int len; + int err; - while (skb->len >= NLMSG_SPACE(0)) { - nlh = nlmsg_hdr(skb); - if (nlh->nlmsg_len < sizeof(*nlh) || skb->len < nlh->nlmsg_len) - return; - rlen = NLMSG_ALIGN(nlh->nlmsg_len); - if (rlen > skb->len) - rlen = skb->len; - if ((err = audit_receive_msg(skb, nlh))) { + nlh = nlmsg_hdr(skb); + len = skb->len; + + while (NLMSG_OK(nlh, len)) { + err = audit_receive_msg(skb, nlh); + /* if err or if this message says it wants a response */ + if (err || (nlh->nlmsg_flags & NLM_F_ACK)) netlink_ack(skb, nlh, err); - } else if (nlh->nlmsg_flags & NLM_F_ACK) - netlink_ack(skb, nlh, 0); - skb_pull(skb, rlen); + + nlh = NLMSG_NEXT(nlh, len); } } @@ -956,13 +959,6 @@ static void audit_receive(struct sk_buff *skb) mutex_unlock(&audit_cmd_mutex); } -#ifdef CONFIG_AUDITSYSCALL -static const struct inotify_operations audit_inotify_ops = { - .handle_event = audit_handle_ievent, - .destroy_watch = audit_free_parent, -}; -#endif - /* Initialize audit support at boot time. */ static int __init audit_init(void) { @@ -988,12 +984,6 @@ static int __init audit_init(void) audit_log(NULL, GFP_KERNEL, AUDIT_KERNEL, "initialized"); -#ifdef CONFIG_AUDITSYSCALL - audit_ih = inotify_init(&audit_inotify_ops); - if (IS_ERR(audit_ih)) - audit_panic("cannot initialize inotify handle"); -#endif - for (i = 0; i < AUDIT_INODE_BUCKETS; i++) INIT_LIST_HEAD(&audit_inode_hash[i]); @@ -1067,18 +1057,20 @@ static struct audit_buffer * audit_buffer_alloc(struct audit_context *ctx, goto err; } - ab->skb = alloc_skb(AUDIT_BUFSIZ, gfp_mask); - if (!ab->skb) - goto err; - ab->ctx = ctx; ab->gfp_mask = gfp_mask; - nlh = (struct nlmsghdr *)skb_put(ab->skb, NLMSG_SPACE(0)); - nlh->nlmsg_type = type; - nlh->nlmsg_flags = 0; - nlh->nlmsg_pid = 0; - nlh->nlmsg_seq = 0; + + ab->skb = nlmsg_new(AUDIT_BUFSIZ, gfp_mask); + if (!ab->skb) + goto nlmsg_failure; + + nlh = NLMSG_NEW(ab->skb, 0, 0, type, 0, 0); + return ab; + +nlmsg_failure: /* Used by NLMSG_NEW */ + kfree_skb(ab->skb); + ab->skb = NULL; err: audit_buffer_free(ab); return NULL; @@ -1382,7 +1374,7 @@ void audit_log_n_string(struct audit_buffer *ab, const char *string, int audit_string_contains_control(const char *string, size_t len) { const unsigned char *p; - for (p = string; p < (const unsigned char *)string + len && *p; p++) { + for (p = string; p < (const unsigned char *)string + len; p++) { if (*p == '"' || *p < 0x21 || *p > 0x7e) return 1; } @@ -1437,18 +1429,27 @@ void audit_log_d_path(struct audit_buffer *ab, const char *prefix, /* We will allow 11 spaces for ' (deleted)' to be appended */ pathname = kmalloc(PATH_MAX+11, ab->gfp_mask); if (!pathname) { - audit_log_format(ab, "<no memory>"); + audit_log_string(ab, "<no_memory>"); return; } p = d_path(path, pathname, PATH_MAX+11); if (IS_ERR(p)) { /* Should never happen since we send PATH_MAX */ /* FIXME: can we save some information here? */ - audit_log_format(ab, "<too long>"); + audit_log_string(ab, "<too_long>"); } else audit_log_untrustedstring(ab, p); kfree(pathname); } +void audit_log_key(struct audit_buffer *ab, char *key) +{ + audit_log_format(ab, " key="); + if (key) + audit_log_untrustedstring(ab, key); + else + audit_log_format(ab, "(null)"); +} + /** * audit_log_end - end one audit record * @ab: the audit_buffer @@ -1472,15 +1473,7 @@ void audit_log_end(struct audit_buffer *ab) skb_queue_tail(&audit_skb_queue, ab->skb); wake_up_interruptible(&kauditd_wait); } else { - if (nlh->nlmsg_type != AUDIT_EOE) { - if (printk_ratelimit()) { - printk(KERN_NOTICE "type=%d %s\n", - nlh->nlmsg_type, - ab->skb->data + NLMSG_SPACE(0)); - } else - audit_log_lost("printk limit exceeded\n"); - } - audit_hold_skb(ab->skb); + audit_printk_skb(ab->skb); } ab->skb = NULL; } diff --git a/kernel/audit.h b/kernel/audit.h index 16f18cac661..208687be4f3 100644 --- a/kernel/audit.h +++ b/kernel/audit.h @@ -53,18 +53,7 @@ enum audit_state { }; /* Rule lists */ -struct audit_parent; - -struct audit_watch { - atomic_t count; /* reference count */ - char *path; /* insertion path */ - dev_t dev; /* associated superblock device */ - unsigned long ino; /* associated inode number */ - struct audit_parent *parent; /* associated parent */ - struct list_head wlist; /* entry in parent->watches list */ - struct list_head rules; /* associated rules */ -}; - +struct audit_watch; struct audit_tree; struct audit_chunk; @@ -108,19 +97,28 @@ struct audit_netlink_list { int audit_send_list(void *); -struct inotify_watch; -/* Inotify handle */ -extern struct inotify_handle *audit_ih; - -extern void audit_free_parent(struct inotify_watch *); -extern void audit_handle_ievent(struct inotify_watch *, u32, u32, u32, - const char *, struct inode *); extern int selinux_audit_rule_update(void); extern struct mutex audit_filter_mutex; extern void audit_free_rule_rcu(struct rcu_head *); extern struct list_head audit_filter_list[]; +/* audit watch functions */ +extern unsigned long audit_watch_inode(struct audit_watch *watch); +extern dev_t audit_watch_dev(struct audit_watch *watch); +extern void audit_put_watch(struct audit_watch *watch); +extern void audit_get_watch(struct audit_watch *watch); +extern int audit_to_watch(struct audit_krule *krule, char *path, int len, u32 op); +extern int audit_add_watch(struct audit_krule *krule); +extern void audit_remove_watch(struct audit_watch *watch); +extern void audit_remove_watch_rule(struct audit_krule *krule, struct list_head *list); +extern void audit_inotify_unregister(struct list_head *in_list); +extern char *audit_watch_path(struct audit_watch *watch); +extern struct list_head *audit_watch_rules(struct audit_watch *watch); + +extern struct audit_entry *audit_dupe_rule(struct audit_krule *old, + struct audit_watch *watch); + #ifdef CONFIG_AUDIT_TREE extern struct audit_chunk *audit_tree_lookup(const struct inode *); extern void audit_put_chunk(struct audit_chunk *); @@ -130,10 +128,9 @@ extern int audit_add_tree_rule(struct audit_krule *); extern int audit_remove_tree_rule(struct audit_krule *); extern void audit_trim_trees(void); extern int audit_tag_tree(char *old, char *new); -extern void audit_schedule_prune(void); -extern void audit_prune_trees(void); extern const char *audit_tree_path(struct audit_tree *); extern void audit_put_tree(struct audit_tree *); +extern void audit_kill_trees(struct list_head *); #else #define audit_remove_tree_rule(rule) BUG() #define audit_add_tree_rule(rule) -EINVAL @@ -142,6 +139,7 @@ extern void audit_put_tree(struct audit_tree *); #define audit_put_tree(tree) (void)0 #define audit_tag_tree(old, new) -EINVAL #define audit_tree_path(rule) "" /* never called */ +#define audit_kill_trees(list) BUG() #endif extern char *audit_unpack_string(void **, size_t *, size_t); @@ -160,7 +158,10 @@ static inline int audit_signal_info(int sig, struct task_struct *t) return 0; } extern void audit_filter_inodes(struct task_struct *, struct audit_context *); +extern struct list_head *audit_killed_trees(void); #else #define audit_signal_info(s,t) AUDIT_DISABLED #define audit_filter_inodes(t,c) AUDIT_DISABLED #endif + +extern struct mutex audit_cmd_mutex; diff --git a/kernel/audit_tree.c b/kernel/audit_tree.c index 8ad9545b8db..2451dc6f328 100644 --- a/kernel/audit_tree.c +++ b/kernel/audit_tree.c @@ -2,6 +2,7 @@ #include <linux/inotify.h> #include <linux/namei.h> #include <linux/mount.h> +#include <linux/kthread.h> struct audit_tree; struct audit_chunk; @@ -385,6 +386,7 @@ static int tag_chunk(struct inode *inode, struct audit_tree *tree) mutex_lock(&inode->inotify_mutex); if (inotify_clone_watch(&old->watch, &chunk->watch) < 0) { mutex_unlock(&inode->inotify_mutex); + put_inotify_watch(&old->watch); free_chunk(chunk); return -ENOSPC; } @@ -394,6 +396,7 @@ static int tag_chunk(struct inode *inode, struct audit_tree *tree) chunk->dead = 1; inotify_evict_watch(&chunk->watch); mutex_unlock(&inode->inotify_mutex); + put_inotify_watch(&old->watch); put_inotify_watch(&chunk->watch); return 0; } @@ -439,13 +442,11 @@ static void kill_rules(struct audit_tree *tree) if (rule->tree) { /* not a half-baked one */ ab = audit_log_start(NULL, GFP_KERNEL, AUDIT_CONFIG_CHANGE); - audit_log_format(ab, "op=remove rule dir="); + audit_log_format(ab, "op="); + audit_log_string(ab, "remove rule"); + audit_log_format(ab, " dir="); audit_log_untrustedstring(ab, rule->tree->pathname); - if (rule->filterkey) { - audit_log_format(ab, " key="); - audit_log_untrustedstring(ab, rule->filterkey); - } else - audit_log_format(ab, " key=(null)"); + audit_log_key(ab, rule->filterkey); audit_log_format(ab, " list=%d res=1", rule->listnr); audit_log_end(ab); rule->tree = NULL; @@ -517,6 +518,8 @@ static void trim_marked(struct audit_tree *tree) } } +static void audit_schedule_prune(void); + /* called with audit_filter_mutex */ int audit_remove_tree_rule(struct audit_krule *rule) { @@ -566,7 +569,7 @@ void audit_trim_trees(void) if (err) goto skip_it; - root_mnt = collect_mounts(path.mnt, path.dentry); + root_mnt = collect_mounts(&path); path_put(&path); if (!root_mnt) goto skip_it; @@ -658,7 +661,7 @@ int audit_add_tree_rule(struct audit_krule *rule) err = kern_path(tree->pathname, 0, &path); if (err) goto Err; - mnt = collect_mounts(path.mnt, path.dentry); + mnt = collect_mounts(&path); path_put(&path); if (!mnt) { err = -ENOMEM; @@ -718,7 +721,7 @@ int audit_tag_tree(char *old, char *new) err = kern_path(new, 0, &path); if (err) return err; - tagged = collect_mounts(path.mnt, path.dentry); + tagged = collect_mounts(&path); path_put(&path); if (!tagged) return -ENOMEM; @@ -732,9 +735,6 @@ int audit_tag_tree(char *old, char *new) dentry = dget(path.dentry); path_put(&path); - if (dentry == tagged->mnt_root && dentry == mnt->mnt_root) - follow_up(&mnt, &dentry); - list_add_tail(&list, &tagged->mnt_list); mutex_lock(&audit_filter_mutex); @@ -825,10 +825,11 @@ int audit_tag_tree(char *old, char *new) /* * That gets run when evict_chunk() ends up needing to kill audit_tree. - * Runs from a separate thread, with audit_cmd_mutex held. + * Runs from a separate thread. */ -void audit_prune_trees(void) +static int prune_tree_thread(void *unused) { + mutex_lock(&audit_cmd_mutex); mutex_lock(&audit_filter_mutex); while (!list_empty(&prune_list)) { @@ -845,6 +846,40 @@ void audit_prune_trees(void) } mutex_unlock(&audit_filter_mutex); + mutex_unlock(&audit_cmd_mutex); + return 0; +} + +static void audit_schedule_prune(void) +{ + kthread_run(prune_tree_thread, NULL, "audit_prune_tree"); +} + +/* + * ... and that one is done if evict_chunk() decides to delay until the end + * of syscall. Runs synchronously. + */ +void audit_kill_trees(struct list_head *list) +{ + mutex_lock(&audit_cmd_mutex); + mutex_lock(&audit_filter_mutex); + + while (!list_empty(list)) { + struct audit_tree *victim; + + victim = list_entry(list->next, struct audit_tree, list); + kill_rules(victim); + list_del_init(&victim->list); + + mutex_unlock(&audit_filter_mutex); + + prune_one(victim); + + mutex_lock(&audit_filter_mutex); + } + + mutex_unlock(&audit_filter_mutex); + mutex_unlock(&audit_cmd_mutex); } /* @@ -855,6 +890,8 @@ void audit_prune_trees(void) static void evict_chunk(struct audit_chunk *chunk) { struct audit_tree *owner; + struct list_head *postponed = audit_killed_trees(); + int need_prune = 0; int n; if (chunk->dead) @@ -870,15 +907,21 @@ static void evict_chunk(struct audit_chunk *chunk) owner->root = NULL; list_del_init(&owner->same_root); spin_unlock(&hash_lock); - kill_rules(owner); - list_move(&owner->list, &prune_list); - audit_schedule_prune(); + if (!postponed) { + kill_rules(owner); + list_move(&owner->list, &prune_list); + need_prune = 1; + } else { + list_move(&owner->list, postponed); + } spin_lock(&hash_lock); } list_del_rcu(&chunk->hash); for (n = 0; n < chunk->count; n++) list_del_init(&chunk->owners[n].list); spin_unlock(&hash_lock); + if (need_prune) + audit_schedule_prune(); mutex_unlock(&audit_filter_mutex); } diff --git a/kernel/audit_watch.c b/kernel/audit_watch.c new file mode 100644 index 00000000000..cc7e87936cb --- /dev/null +++ b/kernel/audit_watch.c @@ -0,0 +1,543 @@ +/* audit_watch.c -- watching inodes + * + * Copyright 2003-2009 Red Hat, Inc. + * Copyright 2005 Hewlett-Packard Development Company, L.P. + * Copyright 2005 IBM Corporation + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + */ + +#include <linux/kernel.h> +#include <linux/audit.h> +#include <linux/kthread.h> +#include <linux/mutex.h> +#include <linux/fs.h> +#include <linux/namei.h> +#include <linux/netlink.h> +#include <linux/sched.h> +#include <linux/inotify.h> +#include <linux/security.h> +#include "audit.h" + +/* + * Reference counting: + * + * audit_parent: lifetime is from audit_init_parent() to receipt of an IN_IGNORED + * event. Each audit_watch holds a reference to its associated parent. + * + * audit_watch: if added to lists, lifetime is from audit_init_watch() to + * audit_remove_watch(). Additionally, an audit_watch may exist + * temporarily to assist in searching existing filter data. Each + * audit_krule holds a reference to its associated watch. + */ + +struct audit_watch { + atomic_t count; /* reference count */ + dev_t dev; /* associated superblock device */ + char *path; /* insertion path */ + unsigned long ino; /* associated inode number */ + struct audit_parent *parent; /* associated parent */ + struct list_head wlist; /* entry in parent->watches list */ + struct list_head rules; /* associated rules */ +}; + +struct audit_parent { + struct list_head ilist; /* entry in inotify registration list */ + struct list_head watches; /* associated watches */ + struct inotify_watch wdata; /* inotify watch data */ + unsigned flags; /* status flags */ +}; + +/* Inotify handle. */ +struct inotify_handle *audit_ih; + +/* + * audit_parent status flags: + * + * AUDIT_PARENT_INVALID - set anytime rules/watches are auto-removed due to + * a filesystem event to ensure we're adding audit watches to a valid parent. + * Technically not needed for IN_DELETE_SELF or IN_UNMOUNT events, as we cannot + * receive them while we have nameidata, but must be used for IN_MOVE_SELF which + * we can receive while holding nameidata. + */ +#define AUDIT_PARENT_INVALID 0x001 + +/* Inotify events we care about. */ +#define AUDIT_IN_WATCH IN_MOVE|IN_CREATE|IN_DELETE|IN_DELETE_SELF|IN_MOVE_SELF + +static void audit_free_parent(struct inotify_watch *i_watch) +{ + struct audit_parent *parent; + + parent = container_of(i_watch, struct audit_parent, wdata); + WARN_ON(!list_empty(&parent->watches)); + kfree(parent); +} + +void audit_get_watch(struct audit_watch *watch) +{ + atomic_inc(&watch->count); +} + +void audit_put_watch(struct audit_watch *watch) +{ + if (atomic_dec_and_test(&watch->count)) { + WARN_ON(watch->parent); + WARN_ON(!list_empty(&watch->rules)); + kfree(watch->path); + kfree(watch); + } +} + +void audit_remove_watch(struct audit_watch *watch) +{ + list_del(&watch->wlist); + put_inotify_watch(&watch->parent->wdata); + watch->parent = NULL; + audit_put_watch(watch); /* match initial get */ +} + +char *audit_watch_path(struct audit_watch *watch) +{ + return watch->path; +} + +struct list_head *audit_watch_rules(struct audit_watch *watch) +{ + return &watch->rules; +} + +unsigned long audit_watch_inode(struct audit_watch *watch) +{ + return watch->ino; +} + +dev_t audit_watch_dev(struct audit_watch *watch) +{ + return watch->dev; +} + +/* Initialize a parent watch entry. */ +static struct audit_parent *audit_init_parent(struct nameidata *ndp) +{ + struct audit_parent *parent; + s32 wd; + + parent = kzalloc(sizeof(*parent), GFP_KERNEL); + if (unlikely(!parent)) + return ERR_PTR(-ENOMEM); + + INIT_LIST_HEAD(&parent->watches); + parent->flags = 0; + + inotify_init_watch(&parent->wdata); + /* grab a ref so inotify watch hangs around until we take audit_filter_mutex */ + get_inotify_watch(&parent->wdata); + wd = inotify_add_watch(audit_ih, &parent->wdata, + ndp->path.dentry->d_inode, AUDIT_IN_WATCH); + if (wd < 0) { + audit_free_parent(&parent->wdata); + return ERR_PTR(wd); + } + + return parent; +} + +/* Initialize a watch entry. */ +static struct audit_watch *audit_init_watch(char *path) +{ + struct audit_watch *watch; + + watch = kzalloc(sizeof(*watch), GFP_KERNEL); + if (unlikely(!watch)) + return ERR_PTR(-ENOMEM); + + INIT_LIST_HEAD(&watch->rules); + atomic_set(&watch->count, 1); + watch->path = path; + watch->dev = (dev_t)-1; + watch->ino = (unsigned long)-1; + + return watch; +} + +/* Translate a watch string to kernel respresentation. */ +int audit_to_watch(struct audit_krule *krule, char *path, int len, u32 op) +{ + struct audit_watch *watch; + + if (!audit_ih) + return -EOPNOTSUPP; + + if (path[0] != '/' || path[len-1] == '/' || + krule->listnr != AUDIT_FILTER_EXIT || + op != Audit_equal || + krule->inode_f || krule->watch || krule->tree) + return -EINVAL; + + watch = audit_init_watch(path); + if (IS_ERR(watch)) + return PTR_ERR(watch); + + audit_get_watch(watch); + krule->watch = watch; + + return 0; +} + +/* Duplicate the given audit watch. The new watch's rules list is initialized + * to an empty list and wlist is undefined. */ +static struct audit_watch *audit_dupe_watch(struct audit_watch *old) +{ + char *path; + struct audit_watch *new; + + path = kstrdup(old->path, GFP_KERNEL); + if (unlikely(!path)) + return ERR_PTR(-ENOMEM); + + new = audit_init_watch(path); + if (IS_ERR(new)) { + kfree(path); + goto out; + } + + new->dev = old->dev; + new->ino = old->ino; + get_inotify_watch(&old->parent->wdata); + new->parent = old->parent; + +out: + return new; +} + +static void audit_watch_log_rule_change(struct audit_krule *r, struct audit_watch *w, char *op) +{ + if (audit_enabled) { + struct audit_buffer *ab; + ab = audit_log_start(NULL, GFP_NOFS, AUDIT_CONFIG_CHANGE); + audit_log_format(ab, "auid=%u ses=%u op=", + audit_get_loginuid(current), + audit_get_sessionid(current)); + audit_log_string(ab, op); + audit_log_format(ab, " path="); + audit_log_untrustedstring(ab, w->path); + audit_log_key(ab, r->filterkey); + audit_log_format(ab, " list=%d res=1", r->listnr); + audit_log_end(ab); + } +} + +/* Update inode info in audit rules based on filesystem event. */ +static void audit_update_watch(struct audit_parent *parent, + const char *dname, dev_t dev, + unsigned long ino, unsigned invalidating) +{ + struct audit_watch *owatch, *nwatch, *nextw; + struct audit_krule *r, *nextr; + struct audit_entry *oentry, *nentry; + + mutex_lock(&audit_filter_mutex); + list_for_each_entry_safe(owatch, nextw, &parent->watches, wlist) { + if (audit_compare_dname_path(dname, owatch->path, NULL)) + continue; + + /* If the update involves invalidating rules, do the inode-based + * filtering now, so we don't omit records. */ + if (invalidating && current->audit_context) + audit_filter_inodes(current, current->audit_context); + + nwatch = audit_dupe_watch(owatch); + if (IS_ERR(nwatch)) { + mutex_unlock(&audit_filter_mutex); + audit_panic("error updating watch, skipping"); + return; + } + nwatch->dev = dev; + nwatch->ino = ino; + + list_for_each_entry_safe(r, nextr, &owatch->rules, rlist) { + + oentry = container_of(r, struct audit_entry, rule); + list_del(&oentry->rule.rlist); + list_del_rcu(&oentry->list); + + nentry = audit_dupe_rule(&oentry->rule, nwatch); + if (IS_ERR(nentry)) { + list_del(&oentry->rule.list); + audit_panic("error updating watch, removing"); + } else { + int h = audit_hash_ino((u32)ino); + list_add(&nentry->rule.rlist, &nwatch->rules); + list_add_rcu(&nentry->list, &audit_inode_hash[h]); + list_replace(&oentry->rule.list, + &nentry->rule.list); + } + + audit_watch_log_rule_change(r, owatch, "updated rules"); + + call_rcu(&oentry->rcu, audit_free_rule_rcu); + } + + audit_remove_watch(owatch); + goto add_watch_to_parent; /* event applies to a single watch */ + } + mutex_unlock(&audit_filter_mutex); + return; + +add_watch_to_parent: + list_add(&nwatch->wlist, &parent->watches); + mutex_unlock(&audit_filter_mutex); + return; +} + +/* Remove all watches & rules associated with a parent that is going away. */ +static void audit_remove_parent_watches(struct audit_parent *parent) +{ + struct audit_watch *w, *nextw; + struct audit_krule *r, *nextr; + struct audit_entry *e; + + mutex_lock(&audit_filter_mutex); + parent->flags |= AUDIT_PARENT_INVALID; + list_for_each_entry_safe(w, nextw, &parent->watches, wlist) { + list_for_each_entry_safe(r, nextr, &w->rules, rlist) { + e = container_of(r, struct audit_entry, rule); + audit_watch_log_rule_change(r, w, "remove rule"); + list_del(&r->rlist); + list_del(&r->list); + list_del_rcu(&e->list); + call_rcu(&e->rcu, audit_free_rule_rcu); + } + audit_remove_watch(w); + } + mutex_unlock(&audit_filter_mutex); +} + +/* Unregister inotify watches for parents on in_list. + * Generates an IN_IGNORED event. */ +void audit_inotify_unregister(struct list_head *in_list) +{ + struct audit_parent *p, *n; + + list_for_each_entry_safe(p, n, in_list, ilist) { + list_del(&p->ilist); + inotify_rm_watch(audit_ih, &p->wdata); + /* the unpin matching the pin in audit_do_del_rule() */ + unpin_inotify_watch(&p->wdata); + } +} + +/* Get path information necessary for adding watches. */ +static int audit_get_nd(char *path, struct nameidata **ndp, struct nameidata **ndw) +{ + struct nameidata *ndparent, *ndwatch; + int err; + + ndparent = kmalloc(sizeof(*ndparent), GFP_KERNEL); + if (unlikely(!ndparent)) + return -ENOMEM; + + ndwatch = kmalloc(sizeof(*ndwatch), GFP_KERNEL); + if (unlikely(!ndwatch)) { + kfree(ndparent); + return -ENOMEM; + } + + err = path_lookup(path, LOOKUP_PARENT, ndparent); + if (err) { + kfree(ndparent); + kfree(ndwatch); + return err; + } + + err = path_lookup(path, 0, ndwatch); + if (err) { + kfree(ndwatch); + ndwatch = NULL; + } + + *ndp = ndparent; + *ndw = ndwatch; + + return 0; +} + +/* Release resources used for watch path information. */ +static void audit_put_nd(struct nameidata *ndp, struct nameidata *ndw) +{ + if (ndp) { + path_put(&ndp->path); + kfree(ndp); + } + if (ndw) { + path_put(&ndw->path); + kfree(ndw); + } +} + +/* Associate the given rule with an existing parent inotify_watch. + * Caller must hold audit_filter_mutex. */ +static void audit_add_to_parent(struct audit_krule *krule, + struct audit_parent *parent) +{ + struct audit_watch *w, *watch = krule->watch; + int watch_found = 0; + + list_for_each_entry(w, &parent->watches, wlist) { + if (strcmp(watch->path, w->path)) + continue; + + watch_found = 1; + + /* put krule's and initial refs to temporary watch */ + audit_put_watch(watch); + audit_put_watch(watch); + + audit_get_watch(w); + krule->watch = watch = w; + break; + } + + if (!watch_found) { + get_inotify_watch(&parent->wdata); + watch->parent = parent; + + list_add(&watch->wlist, &parent->watches); + } + list_add(&krule->rlist, &watch->rules); +} + +/* Find a matching watch entry, or add this one. + * Caller must hold audit_filter_mutex. */ +int audit_add_watch(struct audit_krule *krule) +{ + struct audit_watch *watch = krule->watch; + struct inotify_watch *i_watch; + struct audit_parent *parent; + struct nameidata *ndp = NULL, *ndw = NULL; + int ret = 0; + + mutex_unlock(&audit_filter_mutex); + + /* Avoid calling path_lookup under audit_filter_mutex. */ + ret = audit_get_nd(watch->path, &ndp, &ndw); + if (ret) { + /* caller expects mutex locked */ + mutex_lock(&audit_filter_mutex); + goto error; + } + + /* update watch filter fields */ + if (ndw) { + watch->dev = ndw->path.dentry->d_inode->i_sb->s_dev; + watch->ino = ndw->path.dentry->d_inode->i_ino; + } + + /* The audit_filter_mutex must not be held during inotify calls because + * we hold it during inotify event callback processing. If an existing + * inotify watch is found, inotify_find_watch() grabs a reference before + * returning. + */ + if (inotify_find_watch(audit_ih, ndp->path.dentry->d_inode, + &i_watch) < 0) { + parent = audit_init_parent(ndp); + if (IS_ERR(parent)) { + /* caller expects mutex locked */ + mutex_lock(&audit_filter_mutex); + ret = PTR_ERR(parent); + goto error; + } + } else + parent = container_of(i_watch, struct audit_parent, wdata); + + mutex_lock(&audit_filter_mutex); + + /* parent was moved before we took audit_filter_mutex */ + if (parent->flags & AUDIT_PARENT_INVALID) + ret = -ENOENT; + else + audit_add_to_parent(krule, parent); + + /* match get in audit_init_parent or inotify_find_watch */ + put_inotify_watch(&parent->wdata); + +error: + audit_put_nd(ndp, ndw); /* NULL args OK */ + return ret; + +} + +void audit_remove_watch_rule(struct audit_krule *krule, struct list_head *list) +{ + struct audit_watch *watch = krule->watch; + struct audit_parent *parent = watch->parent; + + list_del(&krule->rlist); + + if (list_empty(&watch->rules)) { + audit_remove_watch(watch); + + if (list_empty(&parent->watches)) { + /* Put parent on the inotify un-registration + * list. Grab a reference before releasing + * audit_filter_mutex, to be released in + * audit_inotify_unregister(). + * If filesystem is going away, just leave + * the sucker alone, eviction will take + * care of it. */ + if (pin_inotify_watch(&parent->wdata)) + list_add(&parent->ilist, list); + } + } +} + +/* Update watch data in audit rules based on inotify events. */ +static void audit_handle_ievent(struct inotify_watch *i_watch, u32 wd, u32 mask, + u32 cookie, const char *dname, struct inode *inode) +{ + struct audit_parent *parent; + + parent = container_of(i_watch, struct audit_parent, wdata); + + if (mask & (IN_CREATE|IN_MOVED_TO) && inode) + audit_update_watch(parent, dname, inode->i_sb->s_dev, + inode->i_ino, 0); + else if (mask & (IN_DELETE|IN_MOVED_FROM)) + audit_update_watch(parent, dname, (dev_t)-1, (unsigned long)-1, 1); + /* inotify automatically removes the watch and sends IN_IGNORED */ + else if (mask & (IN_DELETE_SELF|IN_UNMOUNT)) + audit_remove_parent_watches(parent); + /* inotify does not remove the watch, so remove it manually */ + else if(mask & IN_MOVE_SELF) { + audit_remove_parent_watches(parent); + inotify_remove_watch_locked(audit_ih, i_watch); + } else if (mask & IN_IGNORED) + put_inotify_watch(i_watch); +} + +static const struct inotify_operations audit_inotify_ops = { + .handle_event = audit_handle_ievent, + .destroy_watch = audit_free_parent, +}; + +static int __init audit_watch_init(void) +{ + audit_ih = inotify_init(&audit_inotify_ops); + if (IS_ERR(audit_ih)) + audit_panic("cannot initialize inotify handle"); + return 0; +} +subsys_initcall(audit_watch_init); diff --git a/kernel/auditfilter.c b/kernel/auditfilter.c index fbf24d121d9..a70604047f3 100644 --- a/kernel/auditfilter.c +++ b/kernel/auditfilter.c @@ -27,7 +27,6 @@ #include <linux/namei.h> #include <linux/netlink.h> #include <linux/sched.h> -#include <linux/inotify.h> #include <linux/security.h> #include "audit.h" @@ -44,36 +43,6 @@ * be written directly provided audit_filter_mutex is held. */ -/* - * Reference counting: - * - * audit_parent: lifetime is from audit_init_parent() to receipt of an IN_IGNORED - * event. Each audit_watch holds a reference to its associated parent. - * - * audit_watch: if added to lists, lifetime is from audit_init_watch() to - * audit_remove_watch(). Additionally, an audit_watch may exist - * temporarily to assist in searching existing filter data. Each - * audit_krule holds a reference to its associated watch. - */ - -struct audit_parent { - struct list_head ilist; /* entry in inotify registration list */ - struct list_head watches; /* associated watches */ - struct inotify_watch wdata; /* inotify watch data */ - unsigned flags; /* status flags */ -}; - -/* - * audit_parent status flags: - * - * AUDIT_PARENT_INVALID - set anytime rules/watches are auto-removed due to - * a filesystem event to ensure we're adding audit watches to a valid parent. - * Technically not needed for IN_DELETE_SELF or IN_UNMOUNT events, as we cannot - * receive them while we have nameidata, but must be used for IN_MOVE_SELF which - * we can receive while holding nameidata. - */ -#define AUDIT_PARENT_INVALID 0x001 - /* Audit filter lists, defined in <linux/audit.h> */ struct list_head audit_filter_list[AUDIT_NR_FILTERS] = { LIST_HEAD_INIT(audit_filter_list[0]), @@ -97,56 +66,21 @@ static struct list_head audit_rules_list[AUDIT_NR_FILTERS] = { DEFINE_MUTEX(audit_filter_mutex); -/* Inotify events we care about. */ -#define AUDIT_IN_WATCH IN_MOVE|IN_CREATE|IN_DELETE|IN_DELETE_SELF|IN_MOVE_SELF - -void audit_free_parent(struct inotify_watch *i_watch) -{ - struct audit_parent *parent; - - parent = container_of(i_watch, struct audit_parent, wdata); - WARN_ON(!list_empty(&parent->watches)); - kfree(parent); -} - -static inline void audit_get_watch(struct audit_watch *watch) -{ - atomic_inc(&watch->count); -} - -static void audit_put_watch(struct audit_watch *watch) -{ - if (atomic_dec_and_test(&watch->count)) { - WARN_ON(watch->parent); - WARN_ON(!list_empty(&watch->rules)); - kfree(watch->path); - kfree(watch); - } -} - -static void audit_remove_watch(struct audit_watch *watch) -{ - list_del(&watch->wlist); - put_inotify_watch(&watch->parent->wdata); - watch->parent = NULL; - audit_put_watch(watch); /* match initial get */ -} - static inline void audit_free_rule(struct audit_entry *e) { int i; - + struct audit_krule *erule = &e->rule; /* some rules don't have associated watches */ - if (e->rule.watch) - audit_put_watch(e->rule.watch); - if (e->rule.fields) - for (i = 0; i < e->rule.field_count; i++) { - struct audit_field *f = &e->rule.fields[i]; + if (erule->watch) + audit_put_watch(erule->watch); + if (erule->fields) + for (i = 0; i < erule->field_count; i++) { + struct audit_field *f = &erule->fields[i]; kfree(f->lsm_str); security_audit_rule_free(f->lsm_rule); } - kfree(e->rule.fields); - kfree(e->rule.filterkey); + kfree(erule->fields); + kfree(erule->filterkey); kfree(e); } @@ -156,50 +90,6 @@ void audit_free_rule_rcu(struct rcu_head *head) audit_free_rule(e); } -/* Initialize a parent watch entry. */ -static struct audit_parent *audit_init_parent(struct nameidata *ndp) -{ - struct audit_parent *parent; - s32 wd; - - parent = kzalloc(sizeof(*parent), GFP_KERNEL); - if (unlikely(!parent)) - return ERR_PTR(-ENOMEM); - - INIT_LIST_HEAD(&parent->watches); - parent->flags = 0; - - inotify_init_watch(&parent->wdata); - /* grab a ref so inotify watch hangs around until we take audit_filter_mutex */ - get_inotify_watch(&parent->wdata); - wd = inotify_add_watch(audit_ih, &parent->wdata, - ndp->path.dentry->d_inode, AUDIT_IN_WATCH); - if (wd < 0) { - audit_free_parent(&parent->wdata); - return ERR_PTR(wd); - } - - return parent; -} - -/* Initialize a watch entry. */ -static struct audit_watch *audit_init_watch(char *path) -{ - struct audit_watch *watch; - - watch = kzalloc(sizeof(*watch), GFP_KERNEL); - if (unlikely(!watch)) - return ERR_PTR(-ENOMEM); - - INIT_LIST_HEAD(&watch->rules); - atomic_set(&watch->count, 1); - watch->path = path; - watch->dev = (dev_t)-1; - watch->ino = (unsigned long)-1; - - return watch; -} - /* Initialize an audit filterlist entry. */ static inline struct audit_entry *audit_init_entry(u32 field_count) { @@ -260,31 +150,6 @@ static inline int audit_to_inode(struct audit_krule *krule, return 0; } -/* Translate a watch string to kernel respresentation. */ -static int audit_to_watch(struct audit_krule *krule, char *path, int len, - u32 op) -{ - struct audit_watch *watch; - - if (!audit_ih) - return -EOPNOTSUPP; - - if (path[0] != '/' || path[len-1] == '/' || - krule->listnr != AUDIT_FILTER_EXIT || - op != Audit_equal || - krule->inode_f || krule->watch || krule->tree) - return -EINVAL; - - watch = audit_init_watch(path); - if (IS_ERR(watch)) - return PTR_ERR(watch); - - audit_get_watch(watch); - krule->watch = watch; - - return 0; -} - static __u32 *classes[AUDIT_SYSCALL_CLASSES]; int __init audit_register_class(int class, unsigned *list) @@ -766,7 +631,8 @@ static struct audit_rule_data *audit_krule_to_data(struct audit_krule *krule) break; case AUDIT_WATCH: data->buflen += data->values[i] = - audit_pack_string(&bufp, krule->watch->path); + audit_pack_string(&bufp, + audit_watch_path(krule->watch)); break; case AUDIT_DIR: data->buflen += data->values[i] = @@ -818,7 +684,8 @@ static int audit_compare_rule(struct audit_krule *a, struct audit_krule *b) return 1; break; case AUDIT_WATCH: - if (strcmp(a->watch->path, b->watch->path)) + if (strcmp(audit_watch_path(a->watch), + audit_watch_path(b->watch))) return 1; break; case AUDIT_DIR: @@ -844,32 +711,6 @@ static int audit_compare_rule(struct audit_krule *a, struct audit_krule *b) return 0; } -/* Duplicate the given audit watch. The new watch's rules list is initialized - * to an empty list and wlist is undefined. */ -static struct audit_watch *audit_dupe_watch(struct audit_watch *old) -{ - char *path; - struct audit_watch *new; - - path = kstrdup(old->path, GFP_KERNEL); - if (unlikely(!path)) - return ERR_PTR(-ENOMEM); - - new = audit_init_watch(path); - if (IS_ERR(new)) { - kfree(path); - goto out; - } - - new->dev = old->dev; - new->ino = old->ino; - get_inotify_watch(&old->parent->wdata); - new->parent = old->parent; - -out: - return new; -} - /* Duplicate LSM field information. The lsm_rule is opaque, so must be * re-initialized. */ static inline int audit_dupe_lsm_field(struct audit_field *df, @@ -904,8 +745,8 @@ static inline int audit_dupe_lsm_field(struct audit_field *df, * rule with the new rule in the filterlist, then free the old rule. * The rlist element is undefined; list manipulations are handled apart from * the initial copy. */ -static struct audit_entry *audit_dupe_rule(struct audit_krule *old, - struct audit_watch *watch) +struct audit_entry *audit_dupe_rule(struct audit_krule *old, + struct audit_watch *watch) { u32 fcount = old->field_count; struct audit_entry *entry; @@ -977,137 +818,6 @@ static struct audit_entry *audit_dupe_rule(struct audit_krule *old, return entry; } -/* Update inode info in audit rules based on filesystem event. */ -static void audit_update_watch(struct audit_parent *parent, - const char *dname, dev_t dev, - unsigned long ino, unsigned invalidating) -{ - struct audit_watch *owatch, *nwatch, *nextw; - struct audit_krule *r, *nextr; - struct audit_entry *oentry, *nentry; - - mutex_lock(&audit_filter_mutex); - list_for_each_entry_safe(owatch, nextw, &parent->watches, wlist) { - if (audit_compare_dname_path(dname, owatch->path, NULL)) - continue; - - /* If the update involves invalidating rules, do the inode-based - * filtering now, so we don't omit records. */ - if (invalidating && current->audit_context) - audit_filter_inodes(current, current->audit_context); - - nwatch = audit_dupe_watch(owatch); - if (IS_ERR(nwatch)) { - mutex_unlock(&audit_filter_mutex); - audit_panic("error updating watch, skipping"); - return; - } - nwatch->dev = dev; - nwatch->ino = ino; - - list_for_each_entry_safe(r, nextr, &owatch->rules, rlist) { - - oentry = container_of(r, struct audit_entry, rule); - list_del(&oentry->rule.rlist); - list_del_rcu(&oentry->list); - - nentry = audit_dupe_rule(&oentry->rule, nwatch); - if (IS_ERR(nentry)) { - list_del(&oentry->rule.list); - audit_panic("error updating watch, removing"); - } else { - int h = audit_hash_ino((u32)ino); - list_add(&nentry->rule.rlist, &nwatch->rules); - list_add_rcu(&nentry->list, &audit_inode_hash[h]); - list_replace(&oentry->rule.list, - &nentry->rule.list); - } - - call_rcu(&oentry->rcu, audit_free_rule_rcu); - } - - if (audit_enabled) { - struct audit_buffer *ab; - ab = audit_log_start(NULL, GFP_KERNEL, - AUDIT_CONFIG_CHANGE); - audit_log_format(ab, "auid=%u ses=%u", - audit_get_loginuid(current), - audit_get_sessionid(current)); - audit_log_format(ab, - " op=updated rules specifying path="); - audit_log_untrustedstring(ab, owatch->path); - audit_log_format(ab, " with dev=%u ino=%lu\n", - dev, ino); - audit_log_format(ab, " list=%d res=1", r->listnr); - audit_log_end(ab); - } - audit_remove_watch(owatch); - goto add_watch_to_parent; /* event applies to a single watch */ - } - mutex_unlock(&audit_filter_mutex); - return; - -add_watch_to_parent: - list_add(&nwatch->wlist, &parent->watches); - mutex_unlock(&audit_filter_mutex); - return; -} - -/* Remove all watches & rules associated with a parent that is going away. */ -static void audit_remove_parent_watches(struct audit_parent *parent) -{ - struct audit_watch *w, *nextw; - struct audit_krule *r, *nextr; - struct audit_entry *e; - - mutex_lock(&audit_filter_mutex); - parent->flags |= AUDIT_PARENT_INVALID; - list_for_each_entry_safe(w, nextw, &parent->watches, wlist) { - list_for_each_entry_safe(r, nextr, &w->rules, rlist) { - e = container_of(r, struct audit_entry, rule); - if (audit_enabled) { - struct audit_buffer *ab; - ab = audit_log_start(NULL, GFP_KERNEL, - AUDIT_CONFIG_CHANGE); - audit_log_format(ab, "auid=%u ses=%u", - audit_get_loginuid(current), - audit_get_sessionid(current)); - audit_log_format(ab, " op=remove rule path="); - audit_log_untrustedstring(ab, w->path); - if (r->filterkey) { - audit_log_format(ab, " key="); - audit_log_untrustedstring(ab, - r->filterkey); - } else - audit_log_format(ab, " key=(null)"); - audit_log_format(ab, " list=%d res=1", - r->listnr); - audit_log_end(ab); - } - list_del(&r->rlist); - list_del(&r->list); - list_del_rcu(&e->list); - call_rcu(&e->rcu, audit_free_rule_rcu); - } - audit_remove_watch(w); - } - mutex_unlock(&audit_filter_mutex); -} - -/* Unregister inotify watches for parents on in_list. - * Generates an IN_IGNORED event. */ -static void audit_inotify_unregister(struct list_head *in_list) -{ - struct audit_parent *p, *n; - - list_for_each_entry_safe(p, n, in_list, ilist) { - list_del(&p->ilist); - inotify_rm_watch(audit_ih, &p->wdata); - /* the unpin matching the pin in audit_do_del_rule() */ - unpin_inotify_watch(&p->wdata); - } -} - /* Find an existing audit rule. * Caller must hold audit_filter_mutex to prevent stale rule data. */ static struct audit_entry *audit_find_rule(struct audit_entry *entry, @@ -1145,134 +855,6 @@ out: return found; } -/* Get path information necessary for adding watches. */ -static int audit_get_nd(char *path, struct nameidata **ndp, - struct nameidata **ndw) -{ - struct nameidata *ndparent, *ndwatch; - int err; - - ndparent = kmalloc(sizeof(*ndparent), GFP_KERNEL); - if (unlikely(!ndparent)) - return -ENOMEM; - - ndwatch = kmalloc(sizeof(*ndwatch), GFP_KERNEL); - if (unlikely(!ndwatch)) { - kfree(ndparent); - return -ENOMEM; - } - - err = path_lookup(path, LOOKUP_PARENT, ndparent); - if (err) { - kfree(ndparent); - kfree(ndwatch); - return err; - } - - err = path_lookup(path, 0, ndwatch); - if (err) { - kfree(ndwatch); - ndwatch = NULL; - } - - *ndp = ndparent; - *ndw = ndwatch; - - return 0; -} - -/* Release resources used for watch path information. */ -static void audit_put_nd(struct nameidata *ndp, struct nameidata *ndw) -{ - if (ndp) { - path_put(&ndp->path); - kfree(ndp); - } - if (ndw) { - path_put(&ndw->path); - kfree(ndw); - } -} - -/* Associate the given rule with an existing parent inotify_watch. - * Caller must hold audit_filter_mutex. */ -static void audit_add_to_parent(struct audit_krule *krule, - struct audit_parent *parent) -{ - struct audit_watch *w, *watch = krule->watch; - int watch_found = 0; - - list_for_each_entry(w, &parent->watches, wlist) { - if (strcmp(watch->path, w->path)) - continue; - - watch_found = 1; - - /* put krule's and initial refs to temporary watch */ - audit_put_watch(watch); - audit_put_watch(watch); - - audit_get_watch(w); - krule->watch = watch = w; - break; - } - - if (!watch_found) { - get_inotify_watch(&parent->wdata); - watch->parent = parent; - - list_add(&watch->wlist, &parent->watches); - } - list_add(&krule->rlist, &watch->rules); -} - -/* Find a matching watch entry, or add this one. - * Caller must hold audit_filter_mutex. */ -static int audit_add_watch(struct audit_krule *krule, struct nameidata *ndp, - struct nameidata *ndw) -{ - struct audit_watch *watch = krule->watch; - struct inotify_watch *i_watch; - struct audit_parent *parent; - int ret = 0; - - /* update watch filter fields */ - if (ndw) { - watch->dev = ndw->path.dentry->d_inode->i_sb->s_dev; - watch->ino = ndw->path.dentry->d_inode->i_ino; - } - - /* The audit_filter_mutex must not be held during inotify calls because - * we hold it during inotify event callback processing. If an existing - * inotify watch is found, inotify_find_watch() grabs a reference before - * returning. - */ - mutex_unlock(&audit_filter_mutex); - - if (inotify_find_watch(audit_ih, ndp->path.dentry->d_inode, - &i_watch) < 0) { - parent = audit_init_parent(ndp); - if (IS_ERR(parent)) { - /* caller expects mutex locked */ - mutex_lock(&audit_filter_mutex); - return PTR_ERR(parent); - } - } else - parent = container_of(i_watch, struct audit_parent, wdata); - - mutex_lock(&audit_filter_mutex); - - /* parent was moved before we took audit_filter_mutex */ - if (parent->flags & AUDIT_PARENT_INVALID) - ret = -ENOENT; - else - audit_add_to_parent(krule, parent); - - /* match get in audit_init_parent or inotify_find_watch */ - put_inotify_watch(&parent->wdata); - return ret; -} - static u64 prio_low = ~0ULL/2; static u64 prio_high = ~0ULL/2 - 1; @@ -1282,7 +864,6 @@ static inline int audit_add_rule(struct audit_entry *entry) struct audit_entry *e; struct audit_watch *watch = entry->rule.watch; struct audit_tree *tree = entry->rule.tree; - struct nameidata *ndp = NULL, *ndw = NULL; struct list_head *list; int h, err; #ifdef CONFIG_AUDITSYSCALL @@ -1296,8 +877,8 @@ static inline int audit_add_rule(struct audit_entry *entry) mutex_lock(&audit_filter_mutex); e = audit_find_rule(entry, &list); - mutex_unlock(&audit_filter_mutex); if (e) { + mutex_unlock(&audit_filter_mutex); err = -EEXIST; /* normally audit_add_tree_rule() will free it on failure */ if (tree) @@ -1305,22 +886,16 @@ static inline int audit_add_rule(struct audit_entry *entry) goto error; } - /* Avoid calling path_lookup under audit_filter_mutex. */ - if (watch) { - err = audit_get_nd(watch->path, &ndp, &ndw); - if (err) - goto error; - } - - mutex_lock(&audit_filter_mutex); if (watch) { /* audit_filter_mutex is dropped and re-taken during this call */ - err = audit_add_watch(&entry->rule, ndp, ndw); + err = audit_add_watch(&entry->rule); if (err) { mutex_unlock(&audit_filter_mutex); goto error; } - h = audit_hash_ino((u32)watch->ino); + /* entry->rule.watch may have changed during audit_add_watch() */ + watch = entry->rule.watch; + h = audit_hash_ino((u32)audit_watch_inode(watch)); list = &audit_inode_hash[h]; } if (tree) { @@ -1358,11 +933,9 @@ static inline int audit_add_rule(struct audit_entry *entry) #endif mutex_unlock(&audit_filter_mutex); - audit_put_nd(ndp, ndw); /* NULL args OK */ return 0; error: - audit_put_nd(ndp, ndw); /* NULL args OK */ if (watch) audit_put_watch(watch); /* tmp watch, matches initial get */ return err; @@ -1372,7 +945,7 @@ error: static inline int audit_del_rule(struct audit_entry *entry) { struct audit_entry *e; - struct audit_watch *watch, *tmp_watch = entry->rule.watch; + struct audit_watch *watch = entry->rule.watch; struct audit_tree *tree = entry->rule.tree; struct list_head *list; LIST_HEAD(inotify_list); @@ -1394,29 +967,8 @@ static inline int audit_del_rule(struct audit_entry *entry) goto out; } - watch = e->rule.watch; - if (watch) { - struct audit_parent *parent = watch->parent; - - list_del(&e->rule.rlist); - - if (list_empty(&watch->rules)) { - audit_remove_watch(watch); - - if (list_empty(&parent->watches)) { - /* Put parent on the inotify un-registration - * list. Grab a reference before releasing - * audit_filter_mutex, to be released in - * audit_inotify_unregister(). - * If filesystem is going away, just leave - * the sucker alone, eviction will take - * care of it. - */ - if (pin_inotify_watch(&parent->wdata)) - list_add(&parent->ilist, &inotify_list); - } - } - } + if (e->rule.watch) + audit_remove_watch_rule(&e->rule, &inotify_list); if (e->rule.tree) audit_remove_tree_rule(&e->rule); @@ -1438,8 +990,8 @@ static inline int audit_del_rule(struct audit_entry *entry) audit_inotify_unregister(&inotify_list); out: - if (tmp_watch) - audit_put_watch(tmp_watch); /* match initial get */ + if (watch) + audit_put_watch(watch); /* match initial get */ if (tree) audit_put_tree(tree); /* that's the temporary one */ @@ -1527,11 +1079,9 @@ static void audit_log_rule_change(uid_t loginuid, u32 sessionid, u32 sid, security_release_secctx(ctx, len); } } - audit_log_format(ab, " op=%s rule key=", action); - if (rule->filterkey) - audit_log_untrustedstring(ab, rule->filterkey); - else - audit_log_format(ab, "(null)"); + audit_log_format(ab, " op="); + audit_log_string(ab, action); + audit_log_key(ab, rule->filterkey); audit_log_format(ab, " list=%d res=%d", rule->listnr, res); audit_log_end(ab); } @@ -1595,7 +1145,7 @@ int audit_receive_filter(int type, int pid, int uid, int seq, void *data, return PTR_ERR(entry); err = audit_add_rule(entry); - audit_log_rule_change(loginuid, sessionid, sid, "add", + audit_log_rule_change(loginuid, sessionid, sid, "add rule", &entry->rule, !err); if (err) @@ -1611,7 +1161,7 @@ int audit_receive_filter(int type, int pid, int uid, int seq, void *data, return PTR_ERR(entry); err = audit_del_rule(entry); - audit_log_rule_change(loginuid, sessionid, sid, "remove", + audit_log_rule_change(loginuid, sessionid, sid, "remove rule", &entry->rule, !err); audit_free_rule(entry); @@ -1793,7 +1343,7 @@ static int update_lsm_rule(struct audit_krule *r) list_del(&r->list); } else { if (watch) { - list_add(&nentry->rule.rlist, &watch->rules); + list_add(&nentry->rule.rlist, audit_watch_rules(watch)); list_del(&r->rlist); } else if (tree) list_replace_init(&r->rlist, &nentry->rule.rlist); @@ -1829,27 +1379,3 @@ int audit_update_lsm_rules(void) return err; } - -/* Update watch data in audit rules based on inotify events. */ -void audit_handle_ievent(struct inotify_watch *i_watch, u32 wd, u32 mask, - u32 cookie, const char *dname, struct inode *inode) -{ - struct audit_parent *parent; - - parent = container_of(i_watch, struct audit_parent, wdata); - - if (mask & (IN_CREATE|IN_MOVED_TO) && inode) - audit_update_watch(parent, dname, inode->i_sb->s_dev, - inode->i_ino, 0); - else if (mask & (IN_DELETE|IN_MOVED_FROM)) - audit_update_watch(parent, dname, (dev_t)-1, (unsigned long)-1, 1); - /* inotify automatically removes the watch and sends IN_IGNORED */ - else if (mask & (IN_DELETE_SELF|IN_UNMOUNT)) - audit_remove_parent_watches(parent); - /* inotify does not remove the watch, so remove it manually */ - else if(mask & IN_MOVE_SELF) { - audit_remove_parent_watches(parent); - inotify_remove_watch_locked(audit_ih, i_watch); - } else if (mask & IN_IGNORED) - put_inotify_watch(i_watch); -} diff --git a/kernel/auditsc.c b/kernel/auditsc.c index 8cbddff6c28..267e484f019 100644 --- a/kernel/auditsc.c +++ b/kernel/auditsc.c @@ -66,6 +66,7 @@ #include <linux/syscalls.h> #include <linux/inotify.h> #include <linux/capability.h> +#include <linux/fs_struct.h> #include "audit.h" @@ -167,12 +168,12 @@ struct audit_context { int in_syscall; /* 1 if task is in a syscall */ enum audit_state state, current_state; unsigned int serial; /* serial number for record */ - struct timespec ctime; /* time of syscall entry */ int major; /* syscall number */ + struct timespec ctime; /* time of syscall entry */ unsigned long argv[4]; /* syscall arguments */ - int return_valid; /* return code is valid */ long return_code;/* syscall return code */ u64 prio; + int return_valid; /* return code is valid */ int name_count; struct audit_names names[AUDIT_NAMES]; char * filterkey; /* key for rule that triggered record */ @@ -197,6 +198,7 @@ struct audit_context { char target_comm[TASK_COMM_LEN]; struct audit_tree_refs *trees, *first_trees; + struct list_head killed_trees; int tree_count; int type; @@ -328,6 +330,14 @@ static int audit_match_filetype(struct audit_context *ctx, int which) */ #ifdef CONFIG_AUDIT_TREE +static void audit_set_auditable(struct audit_context *ctx) +{ + if (!ctx->prio) { + ctx->prio = 1; + ctx->current_state = AUDIT_RECORD_CONTEXT; + } +} + static int put_tree_ref(struct audit_context *ctx, struct audit_chunk *chunk) { struct audit_tree_refs *p = ctx->trees; @@ -539,9 +549,9 @@ static int audit_filter_rules(struct task_struct *tsk, } break; case AUDIT_WATCH: - if (name && rule->watch->ino != (unsigned long)-1) - result = (name->dev == rule->watch->dev && - name->ino == rule->watch->ino); + if (name && audit_watch_inode(rule->watch) != (unsigned long)-1) + result = (name->dev == audit_watch_dev(rule->watch) && + name->ino == audit_watch_inode(rule->watch)); break; case AUDIT_DIR: if (ctx) @@ -741,17 +751,9 @@ void audit_filter_inodes(struct task_struct *tsk, struct audit_context *ctx) rcu_read_unlock(); } -static void audit_set_auditable(struct audit_context *ctx) -{ - if (!ctx->prio) { - ctx->prio = 1; - ctx->current_state = AUDIT_RECORD_CONTEXT; - } -} - static inline struct audit_context *audit_get_context(struct task_struct *tsk, int return_valid, - int return_code) + long return_code) { struct audit_context *context = tsk->audit_context; @@ -852,6 +854,7 @@ static inline struct audit_context *audit_alloc_context(enum audit_state state) if (!(context = kmalloc(sizeof(*context), GFP_KERNEL))) return NULL; audit_zero_context(context, state); + INIT_LIST_HEAD(&context->killed_trees); return context; } @@ -1023,8 +1026,8 @@ static int audit_log_single_execve_arg(struct audit_context *context, { char arg_num_len_buf[12]; const char __user *tmp_p = p; - /* how many digits are in arg_num? 3 is the length of a=\n */ - size_t arg_num_len = snprintf(arg_num_len_buf, 12, "%d", arg_num) + 3; + /* how many digits are in arg_num? 5 is the length of ' a=""' */ + size_t arg_num_len = snprintf(arg_num_len_buf, 12, "%d", arg_num) + 5; size_t len, len_left, to_send; size_t max_execve_audit_len = MAX_EXECVE_AUDIT_LEN; unsigned int i, has_cntl = 0, too_long = 0; @@ -1109,7 +1112,7 @@ static int audit_log_single_execve_arg(struct audit_context *context, * so we can be sure nothing was lost. */ if ((i == 0) && (too_long)) - audit_log_format(*ab, "a%d_len=%zu ", arg_num, + audit_log_format(*ab, " a%d_len=%zu", arg_num, has_cntl ? 2*len : len); /* @@ -1129,15 +1132,14 @@ static int audit_log_single_execve_arg(struct audit_context *context, buf[to_send] = '\0'; /* actually log it */ - audit_log_format(*ab, "a%d", arg_num); + audit_log_format(*ab, " a%d", arg_num); if (too_long) audit_log_format(*ab, "[%d]", i); audit_log_format(*ab, "="); if (has_cntl) audit_log_n_hex(*ab, buf, to_send); else - audit_log_format(*ab, "\"%s\"", buf); - audit_log_format(*ab, "\n"); + audit_log_string(*ab, buf); p += to_send; len_left -= to_send; @@ -1165,7 +1167,7 @@ static void audit_log_execve_info(struct audit_context *context, p = (const char __user *)axi->mm->arg_start; - audit_log_format(*ab, "argc=%d ", axi->argc); + audit_log_format(*ab, "argc=%d", axi->argc); /* * we need some kernel buffer to hold the userspace args. Just @@ -1372,11 +1374,7 @@ static void audit_log_exit(struct audit_context *context, struct task_struct *ts audit_log_task_info(ab, tsk); - if (context->filterkey) { - audit_log_format(ab, " key="); - audit_log_untrustedstring(ab, context->filterkey); - } else - audit_log_format(ab, " key=(null)"); + audit_log_key(ab, context->filterkey); audit_log_end(ab); for (aux = context->aux; aux; aux = aux->next) { @@ -1478,7 +1476,7 @@ static void audit_log_exit(struct audit_context *context, struct task_struct *ts case 0: /* name was specified as a relative path and the * directory component is the cwd */ - audit_log_d_path(ab, " name=", &context->pwd); + audit_log_d_path(ab, "name=", &context->pwd); break; default: /* log the name's directory component */ @@ -1549,6 +1547,8 @@ void audit_free(struct task_struct *tsk) /* that can happen only if we are called from do_exit() */ if (context->in_syscall && context->current_state == AUDIT_RECORD_CONTEXT) audit_log_exit(context, tsk); + if (!list_empty(&context->killed_trees)) + audit_kill_trees(&context->killed_trees); audit_free_context(context); } @@ -1692,6 +1692,9 @@ void audit_syscall_exit(int valid, long return_code) context->in_syscall = 0; context->prio = context->state == AUDIT_RECORD_CONTEXT ? ~0ULL : 0; + if (!list_empty(&context->killed_trees)) + audit_kill_trees(&context->killed_trees); + if (context->previous) { struct audit_context *new_context = context->previous; context->previous = NULL; @@ -2149,7 +2152,7 @@ int audit_set_loginuid(struct task_struct *task, uid_t loginuid) * __audit_mq_open - record audit data for a POSIX MQ open * @oflag: open flag * @mode: mode bits - * @u_attr: queue attributes + * @attr: queue attributes * */ void __audit_mq_open(int oflag, mode_t mode, struct mq_attr *attr) @@ -2196,7 +2199,7 @@ void __audit_mq_sendrecv(mqd_t mqdes, size_t msg_len, unsigned int msg_prio, /** * __audit_mq_notify - record audit data for a POSIX MQ notify * @mqdes: MQ descriptor - * @u_notification: Notification event + * @notification: Notification event * */ @@ -2525,3 +2528,11 @@ void audit_core_dumps(long signr) audit_log_format(ab, " sig=%ld", signr); audit_log_end(ab); } + +struct list_head *audit_killed_trees(void) +{ + struct audit_context *ctx = current->audit_context; + if (likely(!ctx || !ctx->in_syscall)) + return NULL; + return &ctx->killed_trees; +} diff --git a/kernel/capability.c b/kernel/capability.c index 4e17041963f..7f876e60521 100644 --- a/kernel/capability.c +++ b/kernel/capability.c @@ -29,7 +29,6 @@ EXPORT_SYMBOL(__cap_empty_set); EXPORT_SYMBOL(__cap_full_set); EXPORT_SYMBOL(__cap_init_eff_set); -#ifdef CONFIG_SECURITY_FILE_CAPABILITIES int file_caps_enabled = 1; static int __init file_caps_disable(char *str) @@ -38,7 +37,6 @@ static int __init file_caps_disable(char *str) return 1; } __setup("no_file_caps", file_caps_disable); -#endif /* * More recent versions of libcap are available from: @@ -169,8 +167,8 @@ SYSCALL_DEFINE2(capget, cap_user_header_t, header, cap_user_data_t, dataptr) kernel_cap_t pE, pI, pP; ret = cap_validate_magic(header, &tocopy); - if (ret != 0) - return ret; + if ((dataptr == NULL) || (ret != 0)) + return ((dataptr == NULL) && (ret == -EINVAL)) ? 0 : ret; if (get_user(pid, &header->pid)) return -EFAULT; @@ -238,7 +236,7 @@ SYSCALL_DEFINE2(capget, cap_user_header_t, header, cap_user_data_t, dataptr) SYSCALL_DEFINE2(capset, cap_user_header_t, header, const cap_user_data_t, data) { struct __user_cap_data_struct kdata[_KERNEL_CAPABILITY_U32S]; - unsigned i, tocopy; + unsigned i, tocopy, copybytes; kernel_cap_t inheritable, permitted, effective; struct cred *new; int ret; @@ -255,8 +253,11 @@ SYSCALL_DEFINE2(capset, cap_user_header_t, header, const cap_user_data_t, data) if (pid != 0 && pid != task_pid_vnr(current)) return -EPERM; - if (copy_from_user(&kdata, data, - tocopy * sizeof(struct __user_cap_data_struct))) + copybytes = tocopy * sizeof(struct __user_cap_data_struct); + if (copybytes > sizeof(kdata)) + return -EFAULT; + + if (copy_from_user(&kdata, data, copybytes)) return -EFAULT; for (i = 0; i < tocopy; i++) { diff --git a/kernel/cgroup.c b/kernel/cgroup.c index c500ca7239b..0249f4be9b5 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -23,6 +23,7 @@ */ #include <linux/cgroup.h> +#include <linux/ctype.h> #include <linux/errno.h> #include <linux/fs.h> #include <linux/kernel.h> @@ -46,6 +47,10 @@ #include <linux/cgroupstats.h> #include <linux/hash.h> #include <linux/namei.h> +#include <linux/smp_lock.h> +#include <linux/pid_namespace.h> +#include <linux/idr.h> +#include <linux/vmalloc.h> /* TODO: replace with more sophisticated array */ #include <asm/atomic.h> @@ -58,6 +63,8 @@ static struct cgroup_subsys *subsys[] = { #include <linux/cgroup_subsys.h> }; +#define MAX_CGROUP_ROOT_NAMELEN 64 + /* * A cgroupfs_root represents the root of a cgroup hierarchy, * and may be associated with a superblock to form an active @@ -72,6 +79,9 @@ struct cgroupfs_root { */ unsigned long subsys_bits; + /* Unique id for this hierarchy. */ + int hierarchy_id; + /* The bitmask of subsystems currently attached to this hierarchy */ unsigned long actual_subsys_bits; @@ -92,8 +102,10 @@ struct cgroupfs_root { /* The path to use for release notifications. */ char release_agent_path[PATH_MAX]; -}; + /* The name for this hierarchy - may be empty */ + char name[MAX_CGROUP_ROOT_NAMELEN]; +}; /* * The "rootnode" hierarchy is the "dummy hierarchy", reserved for the @@ -102,11 +114,48 @@ struct cgroupfs_root { */ static struct cgroupfs_root rootnode; +/* + * CSS ID -- ID per subsys's Cgroup Subsys State(CSS). used only when + * cgroup_subsys->use_id != 0. + */ +#define CSS_ID_MAX (65535) +struct css_id { + /* + * The css to which this ID points. This pointer is set to valid value + * after cgroup is populated. If cgroup is removed, this will be NULL. + * This pointer is expected to be RCU-safe because destroy() + * is called after synchronize_rcu(). But for safe use, css_is_removed() + * css_tryget() should be used for avoiding race. + */ + struct cgroup_subsys_state *css; + /* + * ID of this css. + */ + unsigned short id; + /* + * Depth in hierarchy which this ID belongs to. + */ + unsigned short depth; + /* + * ID is freed by RCU. (and lookup routine is RCU safe.) + */ + struct rcu_head rcu_head; + /* + * Hierarchy of CSS ID belongs to. + */ + unsigned short stack[0]; /* Array of Length (depth+1) */ +}; + + /* The list of hierarchy roots */ static LIST_HEAD(roots); static int root_count; +static DEFINE_IDA(hierarchy_ida); +static int next_hierarchy_id; +static DEFINE_SPINLOCK(hierarchy_id_lock); + /* dummytop is a shorthand for the dummy hierarchy's top cgroup */ #define dummytop (&rootnode.top_cgroup) @@ -167,6 +216,7 @@ struct cg_cgroup_link { * cgroup, anchored on cgroup->css_sets */ struct list_head cgrp_link_list; + struct cgroup *cgrp; /* * List running through cg_cgroup_links pointing at a * single css_set object, anchored on css_set->cg_links @@ -185,14 +235,19 @@ struct cg_cgroup_link { static struct css_set init_css_set; static struct cg_cgroup_link init_css_set_link; +static int cgroup_subsys_init_idr(struct cgroup_subsys *ss); + /* css_set_lock protects the list of css_set objects, and the * chain of tasks off each css_set. Nests outside task->alloc_lock * due to cgroup_iter_start() */ static DEFINE_RWLOCK(css_set_lock); static int css_set_count; -/* hash table for cgroup groups. This improves the performance to - * find an existing css_set */ +/* + * hash table for cgroup groups. This improves the performance to find + * an existing css_set. This hash doesn't (currently) take into + * account cgroups in empty hierarchies. + */ #define CSS_SET_HASH_BITS 7 #define CSS_SET_TABLE_SIZE (1 << CSS_SET_HASH_BITS) static struct hlist_head css_set_table[CSS_SET_TABLE_SIZE]; @@ -212,48 +267,22 @@ static struct hlist_head *css_set_hash(struct cgroup_subsys_state *css[]) return &css_set_table[index]; } +static void free_css_set_rcu(struct rcu_head *obj) +{ + struct css_set *cg = container_of(obj, struct css_set, rcu_head); + kfree(cg); +} + /* We don't maintain the lists running through each css_set to its * task until after the first call to cgroup_iter_start(). This * reduces the fork()/exit() overhead for people who have cgroups * compiled into their kernel but not actually in use */ static int use_task_css_set_links __read_mostly; -/* When we create or destroy a css_set, the operation simply - * takes/releases a reference count on all the cgroups referenced - * by subsystems in this css_set. This can end up multiple-counting - * some cgroups, but that's OK - the ref-count is just a - * busy/not-busy indicator; ensuring that we only count each cgroup - * once would require taking a global lock to ensure that no - * subsystems moved between hierarchies while we were doing so. - * - * Possible TODO: decide at boot time based on the number of - * registered subsystems and the number of CPUs or NUMA nodes whether - * it's better for performance to ref-count every subsystem, or to - * take a global lock and only add one ref count to each hierarchy. - */ - -/* - * unlink a css_set from the list and free it - */ -static void unlink_css_set(struct css_set *cg) +static void __put_css_set(struct css_set *cg, int taskexit) { struct cg_cgroup_link *link; struct cg_cgroup_link *saved_link; - - hlist_del(&cg->hlist); - css_set_count--; - - list_for_each_entry_safe(link, saved_link, &cg->cg_links, - cg_link_list) { - list_del(&link->cg_link_list); - list_del(&link->cgrp_link_list); - kfree(link); - } -} - -static void __put_css_set(struct css_set *cg, int taskexit) -{ - int i; /* * Ensure that the refcount doesn't hit zero while any readers * can see it. Similar to atomic_dec_and_lock(), but for an @@ -266,21 +295,28 @@ static void __put_css_set(struct css_set *cg, int taskexit) write_unlock(&css_set_lock); return; } - unlink_css_set(cg); - write_unlock(&css_set_lock); - rcu_read_lock(); - for (i = 0; i < CGROUP_SUBSYS_COUNT; i++) { - struct cgroup *cgrp = rcu_dereference(cg->subsys[i]->cgroup); + /* This css_set is dead. unlink it and release cgroup refcounts */ + hlist_del(&cg->hlist); + css_set_count--; + + list_for_each_entry_safe(link, saved_link, &cg->cg_links, + cg_link_list) { + struct cgroup *cgrp = link->cgrp; + list_del(&link->cg_link_list); + list_del(&link->cgrp_link_list); if (atomic_dec_and_test(&cgrp->count) && notify_on_release(cgrp)) { if (taskexit) set_bit(CGRP_RELEASABLE, &cgrp->flags); check_for_release(cgrp); } + + kfree(link); } - rcu_read_unlock(); - kfree(cg); + + write_unlock(&css_set_lock); + call_rcu(&cg->rcu_head, free_css_set_rcu); } /* @@ -302,6 +338,78 @@ static inline void put_css_set_taskexit(struct css_set *cg) } /* + * compare_css_sets - helper function for find_existing_css_set(). + * @cg: candidate css_set being tested + * @old_cg: existing css_set for a task + * @new_cgrp: cgroup that's being entered by the task + * @template: desired set of css pointers in css_set (pre-calculated) + * + * Returns true if "cg" matches "old_cg" except for the hierarchy + * which "new_cgrp" belongs to, for which it should match "new_cgrp". + */ +static bool compare_css_sets(struct css_set *cg, + struct css_set *old_cg, + struct cgroup *new_cgrp, + struct cgroup_subsys_state *template[]) +{ + struct list_head *l1, *l2; + + if (memcmp(template, cg->subsys, sizeof(cg->subsys))) { + /* Not all subsystems matched */ + return false; + } + + /* + * Compare cgroup pointers in order to distinguish between + * different cgroups in heirarchies with no subsystems. We + * could get by with just this check alone (and skip the + * memcmp above) but on most setups the memcmp check will + * avoid the need for this more expensive check on almost all + * candidates. + */ + + l1 = &cg->cg_links; + l2 = &old_cg->cg_links; + while (1) { + struct cg_cgroup_link *cgl1, *cgl2; + struct cgroup *cg1, *cg2; + + l1 = l1->next; + l2 = l2->next; + /* See if we reached the end - both lists are equal length. */ + if (l1 == &cg->cg_links) { + BUG_ON(l2 != &old_cg->cg_links); + break; + } else { + BUG_ON(l2 == &old_cg->cg_links); + } + /* Locate the cgroups associated with these links. */ + cgl1 = list_entry(l1, struct cg_cgroup_link, cg_link_list); + cgl2 = list_entry(l2, struct cg_cgroup_link, cg_link_list); + cg1 = cgl1->cgrp; + cg2 = cgl2->cgrp; + /* Hierarchies should be linked in the same order. */ + BUG_ON(cg1->root != cg2->root); + + /* + * If this hierarchy is the hierarchy of the cgroup + * that's changing, then we need to check that this + * css_set points to the new cgroup; if it's any other + * hierarchy, then this css_set should point to the + * same cgroup as the old css_set. + */ + if (cg1->root == new_cgrp->root) { + if (cg1 != new_cgrp) + return false; + } else { + if (cg1 != cg2) + return false; + } + } + return true; +} + +/* * find_existing_css_set() is a helper for * find_css_set(), and checks to see whether an existing * css_set is suitable. @@ -342,10 +450,11 @@ static struct css_set *find_existing_css_set( hhead = css_set_hash(template); hlist_for_each_entry(cg, node, hhead, hlist) { - if (!memcmp(template, cg->subsys, sizeof(cg->subsys))) { - /* All subsystems matched */ - return cg; - } + if (!compare_css_sets(cg, oldcg, cgrp, template)) + continue; + + /* This css_set matches what we need */ + return cg; } /* No existing cgroup group matched */ @@ -399,8 +508,14 @@ static void link_css_set(struct list_head *tmp_cg_links, link = list_first_entry(tmp_cg_links, struct cg_cgroup_link, cgrp_link_list); link->cg = cg; + link->cgrp = cgrp; + atomic_inc(&cgrp->count); list_move(&link->cgrp_link_list, &cgrp->css_sets); - list_add(&link->cg_link_list, &cg->cg_links); + /* + * Always add links to the tail of the list so that the list + * is sorted by order of hierarchy creation + */ + list_add_tail(&link->cg_link_list, &cg->cg_links); } /* @@ -415,11 +530,11 @@ static struct css_set *find_css_set( { struct css_set *res; struct cgroup_subsys_state *template[CGROUP_SUBSYS_COUNT]; - int i; struct list_head tmp_cg_links; struct hlist_head *hhead; + struct cg_cgroup_link *link; /* First see if we already have a cgroup group that matches * the desired set */ @@ -453,20 +568,12 @@ static struct css_set *find_css_set( write_lock(&css_set_lock); /* Add reference counts and links from the new css_set. */ - for (i = 0; i < CGROUP_SUBSYS_COUNT; i++) { - struct cgroup *cgrp = res->subsys[i]->cgroup; - struct cgroup_subsys *ss = subsys[i]; - atomic_inc(&cgrp->count); - /* - * We want to add a link once per cgroup, so we - * only do it for the first subsystem in each - * hierarchy - */ - if (ss->root->subsys_list.next == &ss->sibling) - link_css_set(&tmp_cg_links, res, cgrp); + list_for_each_entry(link, &oldcg->cg_links, cg_link_list) { + struct cgroup *c = link->cgrp; + if (c->root == cgrp->root) + c = cgrp; + link_css_set(&tmp_cg_links, res, c); } - if (list_empty(&rootnode.subsys_list)) - link_css_set(&tmp_cg_links, res, dummytop); BUG_ON(!list_empty(&tmp_cg_links)); @@ -482,6 +589,41 @@ static struct css_set *find_css_set( } /* + * Return the cgroup for "task" from the given hierarchy. Must be + * called with cgroup_mutex held. + */ +static struct cgroup *task_cgroup_from_root(struct task_struct *task, + struct cgroupfs_root *root) +{ + struct css_set *css; + struct cgroup *res = NULL; + + BUG_ON(!mutex_is_locked(&cgroup_mutex)); + read_lock(&css_set_lock); + /* + * No need to lock the task - since we hold cgroup_mutex the + * task can't change groups, so the only thing that can happen + * is that it exits and its css is set back to init_css_set. + */ + css = task->cgroups; + if (css == &init_css_set) { + res = &root->top_cgroup; + } else { + struct cg_cgroup_link *link; + list_for_each_entry(link, &css->cg_links, cg_link_list) { + struct cgroup *c = link->cgrp; + if (c->root == root) { + res = c; + break; + } + } + } + read_unlock(&css_set_lock); + BUG_ON(!res); + return res; +} + +/* * There is one global cgroup mutex. We also require taking * task_lock() when dereferencing a task's cgroup subsys pointers. * See "The task_lock() exception", at the end of this comment. @@ -560,13 +702,17 @@ void cgroup_unlock(void) static int cgroup_mkdir(struct inode *dir, struct dentry *dentry, int mode); static int cgroup_rmdir(struct inode *unused_dir, struct dentry *dentry); static int cgroup_populate_dir(struct cgroup *cgrp); -static struct inode_operations cgroup_dir_inode_operations; -static struct file_operations proc_cgroupstats_operations; +static const struct inode_operations cgroup_dir_inode_operations; +static const struct file_operations proc_cgroupstats_operations; static struct backing_dev_info cgroup_backing_dev_info = { + .name = "cgroup", .capabilities = BDI_CAP_NO_ACCT_AND_WRITEBACK, }; +static int alloc_css_id(struct cgroup_subsys *ss, + struct cgroup *parent, struct cgroup *child); + static struct inode *cgroup_new_inode(mode_t mode, struct super_block *sb) { struct inode *inode = new_inode(sb); @@ -585,13 +731,18 @@ static struct inode *cgroup_new_inode(mode_t mode, struct super_block *sb) * Call subsys's pre_destroy handler. * This is called before css refcnt check. */ -static void cgroup_call_pre_destroy(struct cgroup *cgrp) +static int cgroup_call_pre_destroy(struct cgroup *cgrp) { struct cgroup_subsys *ss; + int ret = 0; + for_each_subsys(cgrp->root, ss) - if (ss->pre_destroy) - ss->pre_destroy(ss, cgrp); - return; + if (ss->pre_destroy) { + ret = ss->pre_destroy(ss, cgrp); + if (ret) + break; + } + return ret; } static void free_cgroup_rcu(struct rcu_head *obj) @@ -632,6 +783,12 @@ static void cgroup_diput(struct dentry *dentry, struct inode *inode) */ deactivate_super(cgrp->root->sb); + /* + * if we're getting rid of the cgroup, refcount should ensure + * that there are no pidlists left. + */ + BUG_ON(!list_empty(&cgrp->pidlists)); + call_rcu(&cgrp->rcu_head, free_cgroup_rcu); } iput(inode); @@ -685,6 +842,34 @@ static void cgroup_d_remove_dir(struct dentry *dentry) remove_dir(dentry); } +/* + * A queue for waiters to do rmdir() cgroup. A tasks will sleep when + * cgroup->count == 0 && list_empty(&cgroup->children) && subsys has some + * reference to css->refcnt. In general, this refcnt is expected to goes down + * to zero, soon. + * + * CGRP_WAIT_ON_RMDIR flag is set under cgroup's inode->i_mutex; + */ +DECLARE_WAIT_QUEUE_HEAD(cgroup_rmdir_waitq); + +static void cgroup_wakeup_rmdir_waiter(struct cgroup *cgrp) +{ + if (unlikely(test_and_clear_bit(CGRP_WAIT_ON_RMDIR, &cgrp->flags))) + wake_up_all(&cgroup_rmdir_waitq); +} + +void cgroup_exclude_rmdir(struct cgroup_subsys_state *css) +{ + css_get(css); +} + +void cgroup_release_and_wakeup_rmdir(struct cgroup_subsys_state *css) +{ + cgroup_wakeup_rmdir_waiter(css->cgroup); + css_put(css); +} + + static int rebind_subsystems(struct cgroupfs_root *root, unsigned long final_bits) { @@ -768,6 +953,8 @@ static int cgroup_show_options(struct seq_file *seq, struct vfsmount *vfs) seq_puts(seq, ",noprefix"); if (strlen(root->release_agent_path)) seq_printf(seq, ",release_agent=%s", root->release_agent_path); + if (strlen(root->name)) + seq_printf(seq, ",name=%s", root->name); mutex_unlock(&cgroup_mutex); return 0; } @@ -776,6 +963,12 @@ struct cgroup_sb_opts { unsigned long subsys_bits; unsigned long flags; char *release_agent; + char *name; + /* User explicitly requested empty subsystem */ + bool none; + + struct cgroupfs_root *new_root; + }; /* Convert a hierarchy specifier into a bitmask of subsystems and @@ -784,10 +977,13 @@ static int parse_cgroupfs_options(char *data, struct cgroup_sb_opts *opts) { char *token, *o = data ?: "all"; + unsigned long mask = (unsigned long)-1; - opts->subsys_bits = 0; - opts->flags = 0; - opts->release_agent = NULL; +#ifdef CONFIG_CPUSETS + mask = ~(1UL << cpuset_subsys_id); +#endif + + memset(opts, 0, sizeof(*opts)); while ((token = strsep(&o, ",")) != NULL) { if (!*token) @@ -801,17 +997,42 @@ static int parse_cgroupfs_options(char *data, if (!ss->disabled) opts->subsys_bits |= 1ul << i; } + } else if (!strcmp(token, "none")) { + /* Explicitly have no subsystems */ + opts->none = true; } else if (!strcmp(token, "noprefix")) { set_bit(ROOT_NOPREFIX, &opts->flags); } else if (!strncmp(token, "release_agent=", 14)) { /* Specifying two release agents is forbidden */ if (opts->release_agent) return -EINVAL; - opts->release_agent = kzalloc(PATH_MAX, GFP_KERNEL); + opts->release_agent = + kstrndup(token + 14, PATH_MAX, GFP_KERNEL); if (!opts->release_agent) return -ENOMEM; - strncpy(opts->release_agent, token + 14, PATH_MAX - 1); - opts->release_agent[PATH_MAX - 1] = 0; + } else if (!strncmp(token, "name=", 5)) { + int i; + const char *name = token + 5; + /* Can't specify an empty name */ + if (!strlen(name)) + return -EINVAL; + /* Must match [\w.-]+ */ + for (i = 0; i < strlen(name); i++) { + char c = name[i]; + if (isalnum(c)) + continue; + if ((c == '.') || (c == '-') || (c == '_')) + continue; + return -EINVAL; + } + /* Specifying two names is forbidden */ + if (opts->name) + return -EINVAL; + opts->name = kstrndup(name, + MAX_CGROUP_ROOT_NAMELEN, + GFP_KERNEL); + if (!opts->name) + return -ENOMEM; } else { struct cgroup_subsys *ss; int i; @@ -828,8 +1049,27 @@ static int parse_cgroupfs_options(char *data, } } - /* We can't have an empty hierarchy */ - if (!opts->subsys_bits) + /* Consistency checks */ + + /* + * Option noprefix was introduced just for backward compatibility + * with the old cpuset, so we allow noprefix only if mounting just + * the cpuset subsystem. + */ + if (test_bit(ROOT_NOPREFIX, &opts->flags) && + (opts->subsys_bits & mask)) + return -EINVAL; + + + /* Can't specify "none" and some subsystems */ + if (opts->subsys_bits && opts->none) + return -EINVAL; + + /* + * We either have to specify by name or by subsystems. (So all + * empty hierarchies must have a name). + */ + if (!opts->subsys_bits && !opts->name) return -EINVAL; return 0; @@ -842,6 +1082,7 @@ static int cgroup_remount(struct super_block *sb, int *flags, char *data) struct cgroup *cgrp = &root->top_cgroup; struct cgroup_sb_opts opts; + lock_kernel(); mutex_lock(&cgrp->dentry->d_inode->i_mutex); mutex_lock(&cgroup_mutex); @@ -856,23 +1097,31 @@ static int cgroup_remount(struct super_block *sb, int *flags, char *data) goto out_unlock; } + /* Don't allow name to change at remount */ + if (opts.name && strcmp(opts.name, root->name)) { + ret = -EINVAL; + goto out_unlock; + } + ret = rebind_subsystems(root, opts.subsys_bits); + if (ret) + goto out_unlock; /* (re)populate subsystem files */ - if (!ret) - cgroup_populate_dir(cgrp); + cgroup_populate_dir(cgrp); if (opts.release_agent) strcpy(root->release_agent_path, opts.release_agent); out_unlock: - if (opts.release_agent) - kfree(opts.release_agent); + kfree(opts.release_agent); + kfree(opts.name); mutex_unlock(&cgroup_mutex); mutex_unlock(&cgrp->dentry->d_inode->i_mutex); + unlock_kernel(); return ret; } -static struct super_operations cgroup_ops = { +static const struct super_operations cgroup_ops = { .statfs = simple_statfs, .drop_inode = generic_delete_inode, .show_options = cgroup_show_options, @@ -885,8 +1134,10 @@ static void init_cgroup_housekeeping(struct cgroup *cgrp) INIT_LIST_HEAD(&cgrp->children); INIT_LIST_HEAD(&cgrp->css_sets); INIT_LIST_HEAD(&cgrp->release_list); - init_rwsem(&cgrp->pids_mutex); + INIT_LIST_HEAD(&cgrp->pidlists); + mutex_init(&cgrp->pidlist_mutex); } + static void init_cgroup_root(struct cgroupfs_root *root) { struct cgroup *cgrp = &root->top_cgroup; @@ -898,33 +1149,106 @@ static void init_cgroup_root(struct cgroupfs_root *root) init_cgroup_housekeeping(cgrp); } +static bool init_root_id(struct cgroupfs_root *root) +{ + int ret = 0; + + do { + if (!ida_pre_get(&hierarchy_ida, GFP_KERNEL)) + return false; + spin_lock(&hierarchy_id_lock); + /* Try to allocate the next unused ID */ + ret = ida_get_new_above(&hierarchy_ida, next_hierarchy_id, + &root->hierarchy_id); + if (ret == -ENOSPC) + /* Try again starting from 0 */ + ret = ida_get_new(&hierarchy_ida, &root->hierarchy_id); + if (!ret) { + next_hierarchy_id = root->hierarchy_id + 1; + } else if (ret != -EAGAIN) { + /* Can only get here if the 31-bit IDR is full ... */ + BUG_ON(ret); + } + spin_unlock(&hierarchy_id_lock); + } while (ret); + return true; +} + static int cgroup_test_super(struct super_block *sb, void *data) { - struct cgroupfs_root *new = data; + struct cgroup_sb_opts *opts = data; struct cgroupfs_root *root = sb->s_fs_info; - /* First check subsystems */ - if (new->subsys_bits != root->subsys_bits) - return 0; + /* If we asked for a name then it must match */ + if (opts->name && strcmp(opts->name, root->name)) + return 0; - /* Next check flags */ - if (new->flags != root->flags) + /* + * If we asked for subsystems (or explicitly for no + * subsystems) then they must match + */ + if ((opts->subsys_bits || opts->none) + && (opts->subsys_bits != root->subsys_bits)) return 0; return 1; } +static struct cgroupfs_root *cgroup_root_from_opts(struct cgroup_sb_opts *opts) +{ + struct cgroupfs_root *root; + + if (!opts->subsys_bits && !opts->none) + return NULL; + + root = kzalloc(sizeof(*root), GFP_KERNEL); + if (!root) + return ERR_PTR(-ENOMEM); + + if (!init_root_id(root)) { + kfree(root); + return ERR_PTR(-ENOMEM); + } + init_cgroup_root(root); + + root->subsys_bits = opts->subsys_bits; + root->flags = opts->flags; + if (opts->release_agent) + strcpy(root->release_agent_path, opts->release_agent); + if (opts->name) + strcpy(root->name, opts->name); + return root; +} + +static void cgroup_drop_root(struct cgroupfs_root *root) +{ + if (!root) + return; + + BUG_ON(!root->hierarchy_id); + spin_lock(&hierarchy_id_lock); + ida_remove(&hierarchy_ida, root->hierarchy_id); + spin_unlock(&hierarchy_id_lock); + kfree(root); +} + static int cgroup_set_super(struct super_block *sb, void *data) { int ret; - struct cgroupfs_root *root = data; + struct cgroup_sb_opts *opts = data; + + /* If we don't have a new root, we can't set up a new sb */ + if (!opts->new_root) + return -EINVAL; + + BUG_ON(!opts->subsys_bits && !opts->none); ret = set_anon_super(sb, NULL); if (ret) return ret; - sb->s_fs_info = root; - root->sb = sb; + sb->s_fs_info = opts->new_root; + opts->new_root->sb = sb; sb->s_blocksize = PAGE_CACHE_SIZE; sb->s_blocksize_bits = PAGE_CACHE_SHIFT; @@ -961,50 +1285,43 @@ static int cgroup_get_sb(struct file_system_type *fs_type, void *data, struct vfsmount *mnt) { struct cgroup_sb_opts opts; + struct cgroupfs_root *root; int ret = 0; struct super_block *sb; - struct cgroupfs_root *root; - struct list_head tmp_cg_links; + struct cgroupfs_root *new_root; /* First find the desired set of subsystems */ ret = parse_cgroupfs_options(data, &opts); - if (ret) { - if (opts.release_agent) - kfree(opts.release_agent); - return ret; - } - - root = kzalloc(sizeof(*root), GFP_KERNEL); - if (!root) { - if (opts.release_agent) - kfree(opts.release_agent); - return -ENOMEM; - } + if (ret) + goto out_err; - init_cgroup_root(root); - root->subsys_bits = opts.subsys_bits; - root->flags = opts.flags; - if (opts.release_agent) { - strcpy(root->release_agent_path, opts.release_agent); - kfree(opts.release_agent); + /* + * Allocate a new cgroup root. We may not need it if we're + * reusing an existing hierarchy. + */ + new_root = cgroup_root_from_opts(&opts); + if (IS_ERR(new_root)) { + ret = PTR_ERR(new_root); + goto out_err; } + opts.new_root = new_root; - sb = sget(fs_type, cgroup_test_super, cgroup_set_super, root); - + /* Locate an existing or new sb for this hierarchy */ + sb = sget(fs_type, cgroup_test_super, cgroup_set_super, &opts); if (IS_ERR(sb)) { - kfree(root); - return PTR_ERR(sb); + ret = PTR_ERR(sb); + cgroup_drop_root(opts.new_root); + goto out_err; } - if (sb->s_fs_info != root) { - /* Reusing an existing superblock */ - BUG_ON(sb->s_root == NULL); - kfree(root); - root = NULL; - } else { - /* New superblock */ + root = sb->s_fs_info; + BUG_ON(!root); + if (root == opts.new_root) { + /* We used the new root structure, so this is a new hierarchy */ + struct list_head tmp_cg_links; struct cgroup *root_cgrp = &root->top_cgroup; struct inode *inode; + struct cgroupfs_root *existing_root; int i; BUG_ON(sb->s_root != NULL); @@ -1017,6 +1334,18 @@ static int cgroup_get_sb(struct file_system_type *fs_type, mutex_lock(&inode->i_mutex); mutex_lock(&cgroup_mutex); + if (strlen(root->name)) { + /* Check for name clashes with existing mounts */ + for_each_active_root(existing_root) { + if (!strcmp(existing_root->name, root->name)) { + ret = -EBUSY; + mutex_unlock(&cgroup_mutex); + mutex_unlock(&inode->i_mutex); + goto drop_new_super; + } + } + } + /* * We're accessing css_set_count without locking * css_set_lock here, but that's OK - it can only be @@ -1035,7 +1364,8 @@ static int cgroup_get_sb(struct file_system_type *fs_type, if (ret == -EBUSY) { mutex_unlock(&cgroup_mutex); mutex_unlock(&inode->i_mutex); - goto free_cg_links; + free_cg_links(&tmp_cg_links); + goto drop_new_super; } /* EBUSY should be the only error here */ @@ -1067,18 +1397,27 @@ static int cgroup_get_sb(struct file_system_type *fs_type, BUG_ON(root->number_of_cgroups != 1); cgroup_populate_dir(root_cgrp); - mutex_unlock(&inode->i_mutex); mutex_unlock(&cgroup_mutex); + mutex_unlock(&inode->i_mutex); + } else { + /* + * We re-used an existing hierarchy - the new root (if + * any) is not needed + */ + cgroup_drop_root(opts.new_root); } simple_set_mnt(mnt, sb); + kfree(opts.release_agent); + kfree(opts.name); return 0; - free_cg_links: - free_cg_links(&tmp_cg_links); drop_new_super: - up_write(&sb->s_umount); - deactivate_super(sb); + deactivate_locked_super(sb); + out_err: + kfree(opts.release_agent); + kfree(opts.name); + return ret; } @@ -1124,7 +1463,7 @@ static void cgroup_kill_sb(struct super_block *sb) { mutex_unlock(&cgroup_mutex); kill_litter_super(sb); - kfree(root); + cgroup_drop_root(root); } static struct file_system_type cgroup_fs_type = { @@ -1189,27 +1528,6 @@ int cgroup_path(const struct cgroup *cgrp, char *buf, int buflen) return 0; } -/* - * Return the first subsystem attached to a cgroup's hierarchy, and - * its subsystem id. - */ - -static void get_first_subsys(const struct cgroup *cgrp, - struct cgroup_subsys_state **css, int *subsys_id) -{ - const struct cgroupfs_root *root = cgrp->root; - const struct cgroup_subsys *test_ss; - BUG_ON(list_empty(&root->subsys_list)); - test_ss = list_entry(root->subsys_list.next, - struct cgroup_subsys, sibling); - if (css) { - *css = cgrp->subsys[test_ss->subsys_id]; - BUG_ON(!*css); - } - if (subsys_id) - *subsys_id = test_ss->subsys_id; -} - /** * cgroup_attach_task - attach task 'tsk' to cgroup 'cgrp' * @cgrp: the cgroup the task is attaching to @@ -1226,18 +1544,15 @@ int cgroup_attach_task(struct cgroup *cgrp, struct task_struct *tsk) struct css_set *cg; struct css_set *newcg; struct cgroupfs_root *root = cgrp->root; - int subsys_id; - - get_first_subsys(cgrp, NULL, &subsys_id); /* Nothing to do if the task is already in that cgroup */ - oldcgrp = task_cgroup(tsk, subsys_id); + oldcgrp = task_cgroup_from_root(tsk, root); if (cgrp == oldcgrp) return 0; for_each_subsys(root, ss) { if (ss->can_attach) { - retval = ss->can_attach(ss, cgrp, tsk); + retval = ss->can_attach(ss, cgrp, tsk, false); if (retval) return retval; } @@ -1275,11 +1590,17 @@ int cgroup_attach_task(struct cgroup *cgrp, struct task_struct *tsk) for_each_subsys(root, ss) { if (ss->attach) - ss->attach(ss, cgrp, oldcgrp, tsk); + ss->attach(ss, cgrp, oldcgrp, tsk, false); } set_bit(CGRP_RELEASABLE, &oldcgrp->flags); synchronize_rcu(); put_css_set(cg); + + /* + * wake up rmdir() waiter. the rmdir should fail since the cgroup + * is no longer empty. + */ + cgroup_wakeup_rmdir_waiter(cgrp); return 0; } @@ -1330,15 +1651,6 @@ static int cgroup_tasks_write(struct cgroup *cgrp, struct cftype *cft, u64 pid) return ret; } -/* The various types of files and directories in a cgroup file system */ -enum cgroup_filetype { - FILE_ROOT, - FILE_DIR, - FILE_TASKLIST, - FILE_NOTIFY_ON_RELEASE, - FILE_RELEASE_AGENT, -}; - /** * cgroup_lock_live_group - take cgroup_mutex and check that cgrp is alive. * @cgrp: the cgroup to be checked for liveness @@ -1398,14 +1710,13 @@ static ssize_t cgroup_write_X64(struct cgroup *cgrp, struct cftype *cft, return -EFAULT; buffer[nbytes] = 0; /* nul-terminate */ - strstrip(buffer); if (cft->write_u64) { - u64 val = simple_strtoull(buffer, &end, 0); + u64 val = simple_strtoull(strstrip(buffer), &end, 0); if (*end) return -EINVAL; retval = cft->write_u64(cgrp, cft, val); } else { - s64 val = simple_strtoll(buffer, &end, 0); + s64 val = simple_strtoll(strstrip(buffer), &end, 0); if (*end) return -EINVAL; retval = cft->write_s64(cgrp, cft, val); @@ -1441,8 +1752,7 @@ static ssize_t cgroup_write_string(struct cgroup *cgrp, struct cftype *cft, } buffer[nbytes] = 0; /* nul-terminate */ - strstrip(buffer); - retval = cft->write_string(cgrp, cft, buffer); + retval = cft->write_string(cgrp, cft, strstrip(buffer)); if (!retval) retval = nbytes; out: @@ -1551,7 +1861,7 @@ static int cgroup_seqfile_release(struct inode *inode, struct file *file) return single_release(inode, file); } -static struct file_operations cgroup_seqfile_operations = { +static const struct file_operations cgroup_seqfile_operations = { .read = seq_read, .write = cgroup_file_write, .llseek = seq_lseek, @@ -1610,7 +1920,7 @@ static int cgroup_rename(struct inode *old_dir, struct dentry *old_dentry, return simple_rename(old_dir, old_dentry, new_dir, new_dentry); } -static struct file_operations cgroup_file_operations = { +static const struct file_operations cgroup_file_operations = { .read = cgroup_file_read, .write = cgroup_file_write, .llseek = generic_file_llseek, @@ -1618,14 +1928,14 @@ static struct file_operations cgroup_file_operations = { .release = cgroup_file_release, }; -static struct inode_operations cgroup_dir_inode_operations = { +static const struct inode_operations cgroup_dir_inode_operations = { .lookup = simple_lookup, .mkdir = cgroup_mkdir, .rmdir = cgroup_rmdir, .rename = cgroup_rename, }; -static int cgroup_create_file(struct dentry *dentry, int mode, +static int cgroup_create_file(struct dentry *dentry, mode_t mode, struct super_block *sb) { static const struct dentry_operations cgroup_dops = { @@ -1671,7 +1981,7 @@ static int cgroup_create_file(struct dentry *dentry, int mode, * @mode: mode to set on new directory. */ static int cgroup_create_dir(struct cgroup *cgrp, struct dentry *dentry, - int mode) + mode_t mode) { struct dentry *parent; int error = 0; @@ -1689,6 +1999,33 @@ static int cgroup_create_dir(struct cgroup *cgrp, struct dentry *dentry, return error; } +/** + * cgroup_file_mode - deduce file mode of a control file + * @cft: the control file in question + * + * returns cft->mode if ->mode is not 0 + * returns S_IRUGO|S_IWUSR if it has both a read and a write handler + * returns S_IRUGO if it has only a read handler + * returns S_IWUSR if it has only a write hander + */ +static mode_t cgroup_file_mode(const struct cftype *cft) +{ + mode_t mode = 0; + + if (cft->mode) + return cft->mode; + + if (cft->read || cft->read_u64 || cft->read_s64 || + cft->read_map || cft->read_seq_string) + mode |= S_IRUGO; + + if (cft->write || cft->write_u64 || cft->write_s64 || + cft->write_string || cft->trigger) + mode |= S_IWUSR; + + return mode; +} + int cgroup_add_file(struct cgroup *cgrp, struct cgroup_subsys *subsys, const struct cftype *cft) @@ -1696,6 +2033,7 @@ int cgroup_add_file(struct cgroup *cgrp, struct dentry *dir = cgrp->dentry; struct dentry *dentry; int error; + mode_t mode; char name[MAX_CGROUP_TYPE_NAMELEN + MAX_CFTYPE_NAME + 2] = { 0 }; if (subsys && !test_bit(ROOT_NOPREFIX, &cgrp->root->flags)) { @@ -1706,7 +2044,8 @@ int cgroup_add_file(struct cgroup *cgrp, BUG_ON(!mutex_is_locked(&dir->d_inode->i_mutex)); dentry = lookup_one_len(name, dir, strlen(name)); if (!IS_ERR(dentry)) { - error = cgroup_create_file(dentry, 0644 | S_IFREG, + mode = cgroup_file_mode(cft); + error = cgroup_create_file(dentry, mode | S_IFREG, cgrp->root->sb); if (!error) dentry->d_fsdata = (void *)cft; @@ -1754,7 +2093,7 @@ int cgroup_task_count(const struct cgroup *cgrp) * the start of a css_set */ static void cgroup_advance_iter(struct cgroup *cgrp, - struct cgroup_iter *it) + struct cgroup_iter *it) { struct list_head *l = it->cg_link; struct cg_cgroup_link *link; @@ -2007,7 +2346,7 @@ int cgroup_scan_tasks(struct cgroup_scanner *scan) } /* - * Stuff for reading the 'tasks' file. + * Stuff for reading the 'tasks'/'procs' files. * * Reading this file can return large amounts of data if a cgroup has * *lots* of attached tasks. So it may need several calls to read(), @@ -2017,27 +2356,196 @@ int cgroup_scan_tasks(struct cgroup_scanner *scan) */ /* - * Load into 'pidarray' up to 'npids' of the tasks using cgroup - * 'cgrp'. Return actual number of pids loaded. No need to - * task_lock(p) when reading out p->cgroup, since we're in an RCU - * read section, so the css_set can't go away, and is - * immutable after creation. + * The following two functions "fix" the issue where there are more pids + * than kmalloc will give memory for; in such cases, we use vmalloc/vfree. + * TODO: replace with a kernel-wide solution to this problem + */ +#define PIDLIST_TOO_LARGE(c) ((c) * sizeof(pid_t) > (PAGE_SIZE * 2)) +static void *pidlist_allocate(int count) +{ + if (PIDLIST_TOO_LARGE(count)) + return vmalloc(count * sizeof(pid_t)); + else + return kmalloc(count * sizeof(pid_t), GFP_KERNEL); +} +static void pidlist_free(void *p) +{ + if (is_vmalloc_addr(p)) + vfree(p); + else + kfree(p); +} +static void *pidlist_resize(void *p, int newcount) +{ + void *newlist; + /* note: if new alloc fails, old p will still be valid either way */ + if (is_vmalloc_addr(p)) { + newlist = vmalloc(newcount * sizeof(pid_t)); + if (!newlist) + return NULL; + memcpy(newlist, p, newcount * sizeof(pid_t)); + vfree(p); + } else { + newlist = krealloc(p, newcount * sizeof(pid_t), GFP_KERNEL); + } + return newlist; +} + +/* + * pidlist_uniq - given a kmalloc()ed list, strip out all duplicate entries + * If the new stripped list is sufficiently smaller and there's enough memory + * to allocate a new buffer, will let go of the unneeded memory. Returns the + * number of unique elements. + */ +/* is the size difference enough that we should re-allocate the array? */ +#define PIDLIST_REALLOC_DIFFERENCE(old, new) ((old) - PAGE_SIZE >= (new)) +static int pidlist_uniq(pid_t **p, int length) +{ + int src, dest = 1; + pid_t *list = *p; + pid_t *newlist; + + /* + * we presume the 0th element is unique, so i starts at 1. trivial + * edge cases first; no work needs to be done for either + */ + if (length == 0 || length == 1) + return length; + /* src and dest walk down the list; dest counts unique elements */ + for (src = 1; src < length; src++) { + /* find next unique element */ + while (list[src] == list[src-1]) { + src++; + if (src == length) + goto after; + } + /* dest always points to where the next unique element goes */ + list[dest] = list[src]; + dest++; + } +after: + /* + * if the length difference is large enough, we want to allocate a + * smaller buffer to save memory. if this fails due to out of memory, + * we'll just stay with what we've got. + */ + if (PIDLIST_REALLOC_DIFFERENCE(length, dest)) { + newlist = pidlist_resize(list, dest); + if (newlist) + *p = newlist; + } + return dest; +} + +static int cmppid(const void *a, const void *b) +{ + return *(pid_t *)a - *(pid_t *)b; +} + +/* + * find the appropriate pidlist for our purpose (given procs vs tasks) + * returns with the lock on that pidlist already held, and takes care + * of the use count, or returns NULL with no locks held if we're out of + * memory. + */ +static struct cgroup_pidlist *cgroup_pidlist_find(struct cgroup *cgrp, + enum cgroup_filetype type) +{ + struct cgroup_pidlist *l; + /* don't need task_nsproxy() if we're looking at ourself */ + struct pid_namespace *ns = get_pid_ns(current->nsproxy->pid_ns); + /* + * We can't drop the pidlist_mutex before taking the l->mutex in case + * the last ref-holder is trying to remove l from the list at the same + * time. Holding the pidlist_mutex precludes somebody taking whichever + * list we find out from under us - compare release_pid_array(). + */ + mutex_lock(&cgrp->pidlist_mutex); + list_for_each_entry(l, &cgrp->pidlists, links) { + if (l->key.type == type && l->key.ns == ns) { + /* found a matching list - drop the extra refcount */ + put_pid_ns(ns); + /* make sure l doesn't vanish out from under us */ + down_write(&l->mutex); + mutex_unlock(&cgrp->pidlist_mutex); + l->use_count++; + return l; + } + } + /* entry not found; create a new one */ + l = kmalloc(sizeof(struct cgroup_pidlist), GFP_KERNEL); + if (!l) { + mutex_unlock(&cgrp->pidlist_mutex); + put_pid_ns(ns); + return l; + } + init_rwsem(&l->mutex); + down_write(&l->mutex); + l->key.type = type; + l->key.ns = ns; + l->use_count = 0; /* don't increment here */ + l->list = NULL; + l->owner = cgrp; + list_add(&l->links, &cgrp->pidlists); + mutex_unlock(&cgrp->pidlist_mutex); + return l; +} + +/* + * Load a cgroup's pidarray with either procs' tgids or tasks' pids */ -static int pid_array_load(pid_t *pidarray, int npids, struct cgroup *cgrp) +static int pidlist_array_load(struct cgroup *cgrp, enum cgroup_filetype type, + struct cgroup_pidlist **lp) { - int n = 0, pid; + pid_t *array; + int length; + int pid, n = 0; /* used for populating the array */ struct cgroup_iter it; struct task_struct *tsk; + struct cgroup_pidlist *l; + + /* + * If cgroup gets more users after we read count, we won't have + * enough space - tough. This race is indistinguishable to the + * caller from the case that the additional cgroup users didn't + * show up until sometime later on. + */ + length = cgroup_task_count(cgrp); + array = pidlist_allocate(length); + if (!array) + return -ENOMEM; + /* now, populate the array */ cgroup_iter_start(cgrp, &it); while ((tsk = cgroup_iter_next(cgrp, &it))) { - if (unlikely(n == npids)) + if (unlikely(n == length)) break; - pid = task_pid_vnr(tsk); - if (pid > 0) - pidarray[n++] = pid; + /* get tgid or pid for procs or tasks file respectively */ + if (type == CGROUP_FILE_PROCS) + pid = task_tgid_vnr(tsk); + else + pid = task_pid_vnr(tsk); + if (pid > 0) /* make sure to only use valid results */ + array[n++] = pid; } cgroup_iter_end(cgrp, &it); - return n; + length = n; + /* now sort & (if procs) strip out duplicates */ + sort(array, length, sizeof(pid_t), cmppid, NULL); + if (type == CGROUP_FILE_PROCS) + length = pidlist_uniq(&array, length); + l = cgroup_pidlist_find(cgrp, type); + if (!l) { + pidlist_free(array); + return -ENOMEM; + } + /* store array, freeing old if necessary - lock already held */ + pidlist_free(l->list); + l->list = array; + l->length = length; + l->use_count++; + up_write(&l->mutex); + *lp = l; + return 0; } /** @@ -2094,19 +2602,14 @@ err: return ret; } -static int cmppid(const void *a, const void *b) -{ - return *(pid_t *)a - *(pid_t *)b; -} - /* - * seq_file methods for the "tasks" file. The seq_file position is the + * seq_file methods for the tasks/procs files. The seq_file position is the * next pid to display; the seq_file iterator is a pointer to the pid - * in the cgroup->tasks_pids array. + * in the cgroup->l->list array. */ -static void *cgroup_tasks_start(struct seq_file *s, loff_t *pos) +static void *cgroup_pidlist_start(struct seq_file *s, loff_t *pos) { /* * Initially we receive a position value that corresponds to @@ -2114,46 +2617,45 @@ static void *cgroup_tasks_start(struct seq_file *s, loff_t *pos) * after a seek to the start). Use a binary-search to find the * next pid to display, if any */ - struct cgroup *cgrp = s->private; + struct cgroup_pidlist *l = s->private; int index = 0, pid = *pos; int *iter; - down_read(&cgrp->pids_mutex); + down_read(&l->mutex); if (pid) { - int end = cgrp->pids_length; + int end = l->length; while (index < end) { int mid = (index + end) / 2; - if (cgrp->tasks_pids[mid] == pid) { + if (l->list[mid] == pid) { index = mid; break; - } else if (cgrp->tasks_pids[mid] <= pid) + } else if (l->list[mid] <= pid) index = mid + 1; else end = mid; } } /* If we're off the end of the array, we're done */ - if (index >= cgrp->pids_length) + if (index >= l->length) return NULL; /* Update the abstract position to be the actual pid that we found */ - iter = cgrp->tasks_pids + index; + iter = l->list + index; *pos = *iter; return iter; } -static void cgroup_tasks_stop(struct seq_file *s, void *v) +static void cgroup_pidlist_stop(struct seq_file *s, void *v) { - struct cgroup *cgrp = s->private; - up_read(&cgrp->pids_mutex); + struct cgroup_pidlist *l = s->private; + up_read(&l->mutex); } -static void *cgroup_tasks_next(struct seq_file *s, void *v, loff_t *pos) +static void *cgroup_pidlist_next(struct seq_file *s, void *v, loff_t *pos) { - struct cgroup *cgrp = s->private; - int *p = v; - int *end = cgrp->tasks_pids + cgrp->pids_length; - + struct cgroup_pidlist *l = s->private; + pid_t *p = v; + pid_t *end = l->list + l->length; /* * Advance to the next pid in the array. If this goes off the * end, we're done @@ -2167,98 +2669,107 @@ static void *cgroup_tasks_next(struct seq_file *s, void *v, loff_t *pos) } } -static int cgroup_tasks_show(struct seq_file *s, void *v) +static int cgroup_pidlist_show(struct seq_file *s, void *v) { return seq_printf(s, "%d\n", *(int *)v); } -static struct seq_operations cgroup_tasks_seq_operations = { - .start = cgroup_tasks_start, - .stop = cgroup_tasks_stop, - .next = cgroup_tasks_next, - .show = cgroup_tasks_show, +/* + * seq_operations functions for iterating on pidlists through seq_file - + * independent of whether it's tasks or procs + */ +static const struct seq_operations cgroup_pidlist_seq_operations = { + .start = cgroup_pidlist_start, + .stop = cgroup_pidlist_stop, + .next = cgroup_pidlist_next, + .show = cgroup_pidlist_show, }; -static void release_cgroup_pid_array(struct cgroup *cgrp) +static void cgroup_release_pid_array(struct cgroup_pidlist *l) { - down_write(&cgrp->pids_mutex); - BUG_ON(!cgrp->pids_use_count); - if (!--cgrp->pids_use_count) { - kfree(cgrp->tasks_pids); - cgrp->tasks_pids = NULL; - cgrp->pids_length = 0; + /* + * the case where we're the last user of this particular pidlist will + * have us remove it from the cgroup's list, which entails taking the + * mutex. since in pidlist_find the pidlist->lock depends on cgroup-> + * pidlist_mutex, we have to take pidlist_mutex first. + */ + mutex_lock(&l->owner->pidlist_mutex); + down_write(&l->mutex); + BUG_ON(!l->use_count); + if (!--l->use_count) { + /* we're the last user if refcount is 0; remove and free */ + list_del(&l->links); + mutex_unlock(&l->owner->pidlist_mutex); + pidlist_free(l->list); + put_pid_ns(l->key.ns); + up_write(&l->mutex); + kfree(l); + return; } - up_write(&cgrp->pids_mutex); + mutex_unlock(&l->owner->pidlist_mutex); + up_write(&l->mutex); } -static int cgroup_tasks_release(struct inode *inode, struct file *file) +static int cgroup_pidlist_release(struct inode *inode, struct file *file) { - struct cgroup *cgrp = __d_cgrp(file->f_dentry->d_parent); - + struct cgroup_pidlist *l; if (!(file->f_mode & FMODE_READ)) return 0; - - release_cgroup_pid_array(cgrp); + /* + * the seq_file will only be initialized if the file was opened for + * reading; hence we check if it's not null only in that case. + */ + l = ((struct seq_file *)file->private_data)->private; + cgroup_release_pid_array(l); return seq_release(inode, file); } -static struct file_operations cgroup_tasks_operations = { +static const struct file_operations cgroup_pidlist_operations = { .read = seq_read, .llseek = seq_lseek, .write = cgroup_file_write, - .release = cgroup_tasks_release, + .release = cgroup_pidlist_release, }; /* - * Handle an open on 'tasks' file. Prepare an array containing the - * process id's of tasks currently attached to the cgroup being opened. + * The following functions handle opens on a file that displays a pidlist + * (tasks or procs). Prepare an array of the process/thread IDs of whoever's + * in the cgroup. */ - -static int cgroup_tasks_open(struct inode *unused, struct file *file) +/* helper function for the two below it */ +static int cgroup_pidlist_open(struct file *file, enum cgroup_filetype type) { struct cgroup *cgrp = __d_cgrp(file->f_dentry->d_parent); - pid_t *pidarray; - int npids; + struct cgroup_pidlist *l; int retval; /* Nothing to do for write-only files */ if (!(file->f_mode & FMODE_READ)) return 0; - /* - * If cgroup gets more users after we read count, we won't have - * enough space - tough. This race is indistinguishable to the - * caller from the case that the additional cgroup users didn't - * show up until sometime later on. - */ - npids = cgroup_task_count(cgrp); - pidarray = kmalloc(npids * sizeof(pid_t), GFP_KERNEL); - if (!pidarray) - return -ENOMEM; - npids = pid_array_load(pidarray, npids, cgrp); - sort(pidarray, npids, sizeof(pid_t), cmppid, NULL); - - /* - * Store the array in the cgroup, freeing the old - * array if necessary - */ - down_write(&cgrp->pids_mutex); - kfree(cgrp->tasks_pids); - cgrp->tasks_pids = pidarray; - cgrp->pids_length = npids; - cgrp->pids_use_count++; - up_write(&cgrp->pids_mutex); - - file->f_op = &cgroup_tasks_operations; + /* have the array populated */ + retval = pidlist_array_load(cgrp, type, &l); + if (retval) + return retval; + /* configure file information */ + file->f_op = &cgroup_pidlist_operations; - retval = seq_open(file, &cgroup_tasks_seq_operations); + retval = seq_open(file, &cgroup_pidlist_seq_operations); if (retval) { - release_cgroup_pid_array(cgrp); + cgroup_release_pid_array(l); return retval; } - ((struct seq_file *)file->private_data)->private = cgrp; + ((struct seq_file *)file->private_data)->private = l; return 0; } +static int cgroup_tasks_open(struct inode *unused, struct file *file) +{ + return cgroup_pidlist_open(file, CGROUP_FILE_TASKS); +} +static int cgroup_procs_open(struct inode *unused, struct file *file) +{ + return cgroup_pidlist_open(file, CGROUP_FILE_PROCS); +} static u64 cgroup_read_notify_on_release(struct cgroup *cgrp, struct cftype *cft) @@ -2281,20 +2792,27 @@ static int cgroup_write_notify_on_release(struct cgroup *cgrp, /* * for the common functions, 'private' gives the type of file */ +/* for hysterical raisins, we can't put this on the older files */ +#define CGROUP_FILE_GENERIC_PREFIX "cgroup." static struct cftype files[] = { { .name = "tasks", .open = cgroup_tasks_open, .write_u64 = cgroup_tasks_write, - .release = cgroup_tasks_release, - .private = FILE_TASKLIST, + .release = cgroup_pidlist_release, + .mode = S_IRUGO | S_IWUSR, + }, + { + .name = CGROUP_FILE_GENERIC_PREFIX "procs", + .open = cgroup_procs_open, + /* .write_u64 = cgroup_procs_write, TODO */ + .release = cgroup_pidlist_release, + .mode = S_IRUGO, }, - { .name = "notify_on_release", .read_u64 = cgroup_read_notify_on_release, .write_u64 = cgroup_write_notify_on_release, - .private = FILE_NOTIFY_ON_RELEASE, }, }; @@ -2303,7 +2821,6 @@ static struct cftype cft_release_agent = { .read_seq_string = cgroup_release_agent_show, .write_string = cgroup_release_agent_write, .max_write_len = PATH_MAX, - .private = FILE_RELEASE_AGENT, }; static int cgroup_populate_dir(struct cgroup *cgrp) @@ -2327,6 +2844,17 @@ static int cgroup_populate_dir(struct cgroup *cgrp) if (ss->populate && (err = ss->populate(ss, cgrp)) < 0) return err; } + /* This cgroup is ready now */ + for_each_subsys(cgrp->root, ss) { + struct cgroup_subsys_state *css = cgrp->subsys[ss->subsys_id]; + /* + * Update id->css pointer and make this css visible from + * CSS ID functions. This pointer will be dereferened + * from RCU-read-side without locks. + */ + if (css->id) + rcu_assign_pointer(css->id->css, css); + } return 0; } @@ -2338,6 +2866,7 @@ static void init_cgroup_css(struct cgroup_subsys_state *css, css->cgroup = cgrp; atomic_set(&css->refcnt, 1); css->flags = 0; + css->id = NULL; if (cgrp == dummytop) set_bit(CSS_ROOT, &css->flags); BUG_ON(cgrp->subsys[ss->subsys_id]); @@ -2376,7 +2905,7 @@ static void cgroup_unlock_hierarchy(struct cgroupfs_root *root) * Must be called with the mutex on the parent inode held */ static long cgroup_create(struct cgroup *parent, struct dentry *dentry, - int mode) + mode_t mode) { struct cgroup *cgrp; struct cgroupfs_root *root = parent->root; @@ -2413,6 +2942,10 @@ static long cgroup_create(struct cgroup *parent, struct dentry *dentry, goto err_destroy; } init_cgroup_css(css, ss, cgrp); + if (ss->use_id) + if (alloc_css_id(ss, parent, cgrp)) + goto err_destroy; + /* At error, ->destroy() callback has to free assigned ID. */ } cgroup_lock_hierarchy(root); @@ -2555,9 +3088,11 @@ static int cgroup_rmdir(struct inode *unused_dir, struct dentry *dentry) struct cgroup *cgrp = dentry->d_fsdata; struct dentry *d; struct cgroup *parent; + DEFINE_WAIT(wait); + int ret; /* the vfs holds both inode->i_mutex already */ - +again: mutex_lock(&cgroup_mutex); if (atomic_read(&cgrp->count) != 0) { mutex_unlock(&cgroup_mutex); @@ -2570,20 +3105,51 @@ static int cgroup_rmdir(struct inode *unused_dir, struct dentry *dentry) mutex_unlock(&cgroup_mutex); /* + * In general, subsystem has no css->refcnt after pre_destroy(). But + * in racy cases, subsystem may have to get css->refcnt after + * pre_destroy() and it makes rmdir return with -EBUSY. This sometimes + * make rmdir return -EBUSY too often. To avoid that, we use waitqueue + * for cgroup's rmdir. CGRP_WAIT_ON_RMDIR is for synchronizing rmdir + * and subsystem's reference count handling. Please see css_get/put + * and css_tryget() and cgroup_wakeup_rmdir_waiter() implementation. + */ + set_bit(CGRP_WAIT_ON_RMDIR, &cgrp->flags); + + /* * Call pre_destroy handlers of subsys. Notify subsystems * that rmdir() request comes. */ - cgroup_call_pre_destroy(cgrp); + ret = cgroup_call_pre_destroy(cgrp); + if (ret) { + clear_bit(CGRP_WAIT_ON_RMDIR, &cgrp->flags); + return ret; + } mutex_lock(&cgroup_mutex); parent = cgrp->parent; - - if (atomic_read(&cgrp->count) - || !list_empty(&cgrp->children) - || !cgroup_clear_css_refs(cgrp)) { + if (atomic_read(&cgrp->count) || !list_empty(&cgrp->children)) { + clear_bit(CGRP_WAIT_ON_RMDIR, &cgrp->flags); mutex_unlock(&cgroup_mutex); return -EBUSY; } + prepare_to_wait(&cgroup_rmdir_waitq, &wait, TASK_INTERRUPTIBLE); + if (!cgroup_clear_css_refs(cgrp)) { + mutex_unlock(&cgroup_mutex); + /* + * Because someone may call cgroup_wakeup_rmdir_waiter() before + * prepare_to_wait(), we need to check this flag. + */ + if (test_bit(CGRP_WAIT_ON_RMDIR, &cgrp->flags)) + schedule(); + finish_wait(&cgroup_rmdir_waitq, &wait); + clear_bit(CGRP_WAIT_ON_RMDIR, &cgrp->flags); + if (signal_pending(current)) + return -EINTR; + goto again; + } + /* NO css_tryget() can success after here. */ + finish_wait(&cgroup_rmdir_waitq, &wait); + clear_bit(CGRP_WAIT_ON_RMDIR, &cgrp->flags); spin_lock(&release_list_lock); set_bit(CGRP_REMOVED, &cgrp->flags); @@ -2661,6 +3227,7 @@ int __init cgroup_init_early(void) init_task.cgroups = &init_css_set; init_css_set_link.cg = &init_css_set; + init_css_set_link.cgrp = dummytop; list_add(&init_css_set_link.cgrp_link_list, &rootnode.top_cgroup.css_sets); list_add(&init_css_set_link.cg_link_list, @@ -2708,12 +3275,14 @@ int __init cgroup_init(void) struct cgroup_subsys *ss = subsys[i]; if (!ss->early_init) cgroup_init_subsys(ss); + if (ss->use_id) + cgroup_subsys_init_idr(ss); } /* Add init_css_set to the hash table */ hhead = css_set_hash(init_css_set.subsys); hlist_add_head(&init_css_set.hlist, hhead); - + BUG_ON(!init_root_id(&rootnode)); err = register_filesystem(&cgroup_fs_type); if (err < 0) goto out; @@ -2766,15 +3335,16 @@ static int proc_cgroup_show(struct seq_file *m, void *v) for_each_active_root(root) { struct cgroup_subsys *ss; struct cgroup *cgrp; - int subsys_id; int count = 0; - seq_printf(m, "%lu:", root->subsys_bits); + seq_printf(m, "%d:", root->hierarchy_id); for_each_subsys(root, ss) seq_printf(m, "%s%s", count++ ? "," : "", ss->name); + if (strlen(root->name)) + seq_printf(m, "%sname=%s", count ? "," : "", + root->name); seq_putc(m, ':'); - get_first_subsys(&root->top_cgroup, NULL, &subsys_id); - cgrp = task_cgroup(tsk, subsys_id); + cgrp = task_cgroup_from_root(tsk, root); retval = cgroup_path(cgrp, buf, PAGE_SIZE); if (retval < 0) goto out_unlock; @@ -2797,7 +3367,7 @@ static int cgroup_open(struct inode *inode, struct file *file) return single_open(file, proc_cgroup_show, pid); } -struct file_operations proc_cgroup_operations = { +const struct file_operations proc_cgroup_operations = { .open = cgroup_open, .read = seq_read, .llseek = seq_lseek, @@ -2813,8 +3383,8 @@ static int proc_cgroupstats_show(struct seq_file *m, void *v) mutex_lock(&cgroup_mutex); for (i = 0; i < CGROUP_SUBSYS_COUNT; i++) { struct cgroup_subsys *ss = subsys[i]; - seq_printf(m, "%s\t%lu\t%d\t%d\n", - ss->name, ss->root->subsys_bits, + seq_printf(m, "%s\t%d\t%d\t%d\n", + ss->name, ss->root->hierarchy_id, ss->root->number_of_cgroups, !ss->disabled); } mutex_unlock(&cgroup_mutex); @@ -2826,7 +3396,7 @@ static int cgroupstats_open(struct inode *inode, struct file *file) return single_open(file, proc_cgroupstats_show, NULL); } -static struct file_operations proc_cgroupstats_operations = { +static const struct file_operations proc_cgroupstats_operations = { .open = cgroupstats_open, .read = seq_read, .llseek = seq_lseek, @@ -3084,28 +3654,27 @@ int cgroup_clone(struct task_struct *tsk, struct cgroup_subsys *subsys, } /** - * cgroup_is_descendant - see if @cgrp is a descendant of current task's cgrp + * cgroup_is_descendant - see if @cgrp is a descendant of @task's cgrp * @cgrp: the cgroup in question + * @task: the task in question * - * See if @cgrp is a descendant of the current task's cgroup in - * the appropriate hierarchy. + * See if @cgrp is a descendant of @task's cgroup in the appropriate + * hierarchy. * * If we are sending in dummytop, then presumably we are creating * the top cgroup in the subsystem. * * Called only by the ns (nsproxy) cgroup. */ -int cgroup_is_descendant(const struct cgroup *cgrp) +int cgroup_is_descendant(const struct cgroup *cgrp, struct task_struct *task) { int ret; struct cgroup *target; - int subsys_id; if (cgrp == dummytop) return 1; - get_first_subsys(cgrp, NULL, &subsys_id); - target = task_cgroup(current, subsys_id); + target = task_cgroup_from_root(task, cgrp->root); while (cgrp != target && cgrp!= cgrp->top_cgroup) cgrp = cgrp->parent; ret = (cgrp == target); @@ -3137,13 +3706,18 @@ static void check_for_release(struct cgroup *cgrp) void __css_put(struct cgroup_subsys_state *css) { struct cgroup *cgrp = css->cgroup; + int val; rcu_read_lock(); - if ((atomic_dec_return(&css->refcnt) == 1) && - notify_on_release(cgrp)) { - set_bit(CGRP_RELEASABLE, &cgrp->flags); - check_for_release(cgrp); + val = atomic_dec_return(&css->refcnt); + if (val == 1) { + if (notify_on_release(cgrp)) { + set_bit(CGRP_RELEASABLE, &cgrp->flags); + check_for_release(cgrp); + } + cgroup_wakeup_rmdir_waiter(cgrp); } rcu_read_unlock(); + WARN_ON_ONCE(val < 1); } /* @@ -3241,3 +3815,383 @@ static int __init cgroup_disable(char *str) return 1; } __setup("cgroup_disable=", cgroup_disable); + +/* + * Functons for CSS ID. + */ + +/* + *To get ID other than 0, this should be called when !cgroup_is_removed(). + */ +unsigned short css_id(struct cgroup_subsys_state *css) +{ + struct css_id *cssid = rcu_dereference(css->id); + + if (cssid) + return cssid->id; + return 0; +} + +unsigned short css_depth(struct cgroup_subsys_state *css) +{ + struct css_id *cssid = rcu_dereference(css->id); + + if (cssid) + return cssid->depth; + return 0; +} + +bool css_is_ancestor(struct cgroup_subsys_state *child, + const struct cgroup_subsys_state *root) +{ + struct css_id *child_id = rcu_dereference(child->id); + struct css_id *root_id = rcu_dereference(root->id); + + if (!child_id || !root_id || (child_id->depth < root_id->depth)) + return false; + return child_id->stack[root_id->depth] == root_id->id; +} + +static void __free_css_id_cb(struct rcu_head *head) +{ + struct css_id *id; + + id = container_of(head, struct css_id, rcu_head); + kfree(id); +} + +void free_css_id(struct cgroup_subsys *ss, struct cgroup_subsys_state *css) +{ + struct css_id *id = css->id; + /* When this is called before css_id initialization, id can be NULL */ + if (!id) + return; + + BUG_ON(!ss->use_id); + + rcu_assign_pointer(id->css, NULL); + rcu_assign_pointer(css->id, NULL); + spin_lock(&ss->id_lock); + idr_remove(&ss->idr, id->id); + spin_unlock(&ss->id_lock); + call_rcu(&id->rcu_head, __free_css_id_cb); +} + +/* + * This is called by init or create(). Then, calls to this function are + * always serialized (By cgroup_mutex() at create()). + */ + +static struct css_id *get_new_cssid(struct cgroup_subsys *ss, int depth) +{ + struct css_id *newid; + int myid, error, size; + + BUG_ON(!ss->use_id); + + size = sizeof(*newid) + sizeof(unsigned short) * (depth + 1); + newid = kzalloc(size, GFP_KERNEL); + if (!newid) + return ERR_PTR(-ENOMEM); + /* get id */ + if (unlikely(!idr_pre_get(&ss->idr, GFP_KERNEL))) { + error = -ENOMEM; + goto err_out; + } + spin_lock(&ss->id_lock); + /* Don't use 0. allocates an ID of 1-65535 */ + error = idr_get_new_above(&ss->idr, newid, 1, &myid); + spin_unlock(&ss->id_lock); + + /* Returns error when there are no free spaces for new ID.*/ + if (error) { + error = -ENOSPC; + goto err_out; + } + if (myid > CSS_ID_MAX) + goto remove_idr; + + newid->id = myid; + newid->depth = depth; + return newid; +remove_idr: + error = -ENOSPC; + spin_lock(&ss->id_lock); + idr_remove(&ss->idr, myid); + spin_unlock(&ss->id_lock); +err_out: + kfree(newid); + return ERR_PTR(error); + +} + +static int __init cgroup_subsys_init_idr(struct cgroup_subsys *ss) +{ + struct css_id *newid; + struct cgroup_subsys_state *rootcss; + + spin_lock_init(&ss->id_lock); + idr_init(&ss->idr); + + rootcss = init_css_set.subsys[ss->subsys_id]; + newid = get_new_cssid(ss, 0); + if (IS_ERR(newid)) + return PTR_ERR(newid); + + newid->stack[0] = newid->id; + newid->css = rootcss; + rootcss->id = newid; + return 0; +} + +static int alloc_css_id(struct cgroup_subsys *ss, struct cgroup *parent, + struct cgroup *child) +{ + int subsys_id, i, depth = 0; + struct cgroup_subsys_state *parent_css, *child_css; + struct css_id *child_id, *parent_id = NULL; + + subsys_id = ss->subsys_id; + parent_css = parent->subsys[subsys_id]; + child_css = child->subsys[subsys_id]; + depth = css_depth(parent_css) + 1; + parent_id = parent_css->id; + + child_id = get_new_cssid(ss, depth); + if (IS_ERR(child_id)) + return PTR_ERR(child_id); + + for (i = 0; i < depth; i++) + child_id->stack[i] = parent_id->stack[i]; + child_id->stack[depth] = child_id->id; + /* + * child_id->css pointer will be set after this cgroup is available + * see cgroup_populate_dir() + */ + rcu_assign_pointer(child_css->id, child_id); + + return 0; +} + +/** + * css_lookup - lookup css by id + * @ss: cgroup subsys to be looked into. + * @id: the id + * + * Returns pointer to cgroup_subsys_state if there is valid one with id. + * NULL if not. Should be called under rcu_read_lock() + */ +struct cgroup_subsys_state *css_lookup(struct cgroup_subsys *ss, int id) +{ + struct css_id *cssid = NULL; + + BUG_ON(!ss->use_id); + cssid = idr_find(&ss->idr, id); + + if (unlikely(!cssid)) + return NULL; + + return rcu_dereference(cssid->css); +} + +/** + * css_get_next - lookup next cgroup under specified hierarchy. + * @ss: pointer to subsystem + * @id: current position of iteration. + * @root: pointer to css. search tree under this. + * @foundid: position of found object. + * + * Search next css under the specified hierarchy of rootid. Calling under + * rcu_read_lock() is necessary. Returns NULL if it reaches the end. + */ +struct cgroup_subsys_state * +css_get_next(struct cgroup_subsys *ss, int id, + struct cgroup_subsys_state *root, int *foundid) +{ + struct cgroup_subsys_state *ret = NULL; + struct css_id *tmp; + int tmpid; + int rootid = css_id(root); + int depth = css_depth(root); + + if (!rootid) + return NULL; + + BUG_ON(!ss->use_id); + /* fill start point for scan */ + tmpid = id; + while (1) { + /* + * scan next entry from bitmap(tree), tmpid is updated after + * idr_get_next(). + */ + spin_lock(&ss->id_lock); + tmp = idr_get_next(&ss->idr, &tmpid); + spin_unlock(&ss->id_lock); + + if (!tmp) + break; + if (tmp->depth >= depth && tmp->stack[depth] == rootid) { + ret = rcu_dereference(tmp->css); + if (ret) { + *foundid = tmpid; + break; + } + } + /* continue to scan from next id */ + tmpid = tmpid + 1; + } + return ret; +} + +#ifdef CONFIG_CGROUP_DEBUG +static struct cgroup_subsys_state *debug_create(struct cgroup_subsys *ss, + struct cgroup *cont) +{ + struct cgroup_subsys_state *css = kzalloc(sizeof(*css), GFP_KERNEL); + + if (!css) + return ERR_PTR(-ENOMEM); + + return css; +} + +static void debug_destroy(struct cgroup_subsys *ss, struct cgroup *cont) +{ + kfree(cont->subsys[debug_subsys_id]); +} + +static u64 cgroup_refcount_read(struct cgroup *cont, struct cftype *cft) +{ + return atomic_read(&cont->count); +} + +static u64 debug_taskcount_read(struct cgroup *cont, struct cftype *cft) +{ + return cgroup_task_count(cont); +} + +static u64 current_css_set_read(struct cgroup *cont, struct cftype *cft) +{ + return (u64)(unsigned long)current->cgroups; +} + +static u64 current_css_set_refcount_read(struct cgroup *cont, + struct cftype *cft) +{ + u64 count; + + rcu_read_lock(); + count = atomic_read(¤t->cgroups->refcount); + rcu_read_unlock(); + return count; +} + +static int current_css_set_cg_links_read(struct cgroup *cont, + struct cftype *cft, + struct seq_file *seq) +{ + struct cg_cgroup_link *link; + struct css_set *cg; + + read_lock(&css_set_lock); + rcu_read_lock(); + cg = rcu_dereference(current->cgroups); + list_for_each_entry(link, &cg->cg_links, cg_link_list) { + struct cgroup *c = link->cgrp; + const char *name; + + if (c->dentry) + name = c->dentry->d_name.name; + else + name = "?"; + seq_printf(seq, "Root %d group %s\n", + c->root->hierarchy_id, name); + } + rcu_read_unlock(); + read_unlock(&css_set_lock); + return 0; +} + +#define MAX_TASKS_SHOWN_PER_CSS 25 +static int cgroup_css_links_read(struct cgroup *cont, + struct cftype *cft, + struct seq_file *seq) +{ + struct cg_cgroup_link *link; + + read_lock(&css_set_lock); + list_for_each_entry(link, &cont->css_sets, cgrp_link_list) { + struct css_set *cg = link->cg; + struct task_struct *task; + int count = 0; + seq_printf(seq, "css_set %p\n", cg); + list_for_each_entry(task, &cg->tasks, cg_list) { + if (count++ > MAX_TASKS_SHOWN_PER_CSS) { + seq_puts(seq, " ...\n"); + break; + } else { + seq_printf(seq, " task %d\n", + task_pid_vnr(task)); + } + } + } + read_unlock(&css_set_lock); + return 0; +} + +static u64 releasable_read(struct cgroup *cgrp, struct cftype *cft) +{ + return test_bit(CGRP_RELEASABLE, &cgrp->flags); +} + +static struct cftype debug_files[] = { + { + .name = "cgroup_refcount", + .read_u64 = cgroup_refcount_read, + }, + { + .name = "taskcount", + .read_u64 = debug_taskcount_read, + }, + + { + .name = "current_css_set", + .read_u64 = current_css_set_read, + }, + + { + .name = "current_css_set_refcount", + .read_u64 = current_css_set_refcount_read, + }, + + { + .name = "current_css_set_cg_links", + .read_seq_string = current_css_set_cg_links_read, + }, + + { + .name = "cgroup_css_links", + .read_seq_string = cgroup_css_links_read, + }, + + { + .name = "releasable", + .read_u64 = releasable_read, + }, +}; + +static int debug_populate(struct cgroup_subsys *ss, struct cgroup *cont) +{ + return cgroup_add_files(cont, ss, debug_files, + ARRAY_SIZE(debug_files)); +} + +struct cgroup_subsys debug_subsys = { + .name = "debug", + .create = debug_create, + .destroy = debug_destroy, + .populate = debug_populate, + .subsys_id = debug_subsys_id, +}; +#endif /* CONFIG_CGROUP_DEBUG */ diff --git a/kernel/cgroup_debug.c b/kernel/cgroup_debug.c deleted file mode 100644 index daca6209202..00000000000 --- a/kernel/cgroup_debug.c +++ /dev/null @@ -1,107 +0,0 @@ -/* - * kernel/cgroup_debug.c - Example cgroup subsystem that - * exposes debug info - * - * Copyright (C) Google Inc, 2007 - * - * Developed by Paul Menage (menage@google.com) - * - */ - -#include <linux/cgroup.h> -#include <linux/fs.h> -#include <linux/slab.h> -#include <linux/rcupdate.h> - -#include <asm/atomic.h> - -static struct cgroup_subsys_state *debug_create(struct cgroup_subsys *ss, - struct cgroup *cont) -{ - struct cgroup_subsys_state *css = kzalloc(sizeof(*css), GFP_KERNEL); - - if (!css) - return ERR_PTR(-ENOMEM); - - return css; -} - -static void debug_destroy(struct cgroup_subsys *ss, struct cgroup *cont) -{ - kfree(cont->subsys[debug_subsys_id]); -} - -static u64 cgroup_refcount_read(struct cgroup *cont, struct cftype *cft) -{ - return atomic_read(&cont->count); -} - -static u64 taskcount_read(struct cgroup *cont, struct cftype *cft) -{ - u64 count; - - cgroup_lock(); - count = cgroup_task_count(cont); - cgroup_unlock(); - return count; -} - -static u64 current_css_set_read(struct cgroup *cont, struct cftype *cft) -{ - return (u64)(long)current->cgroups; -} - -static u64 current_css_set_refcount_read(struct cgroup *cont, - struct cftype *cft) -{ - u64 count; - - rcu_read_lock(); - count = atomic_read(¤t->cgroups->refcount); - rcu_read_unlock(); - return count; -} - -static u64 releasable_read(struct cgroup *cgrp, struct cftype *cft) -{ - return test_bit(CGRP_RELEASABLE, &cgrp->flags); -} - -static struct cftype files[] = { - { - .name = "cgroup_refcount", - .read_u64 = cgroup_refcount_read, - }, - { - .name = "taskcount", - .read_u64 = taskcount_read, - }, - - { - .name = "current_css_set", - .read_u64 = current_css_set_read, - }, - - { - .name = "current_css_set_refcount", - .read_u64 = current_css_set_refcount_read, - }, - - { - .name = "releasable", - .read_u64 = releasable_read, - }, -}; - -static int debug_populate(struct cgroup_subsys *ss, struct cgroup *cont) -{ - return cgroup_add_files(cont, ss, files, ARRAY_SIZE(files)); -} - -struct cgroup_subsys debug_subsys = { - .name = "debug", - .create = debug_create, - .destroy = debug_destroy, - .populate = debug_populate, - .subsys_id = debug_subsys_id, -}; diff --git a/kernel/cgroup_freezer.c b/kernel/cgroup_freezer.c index fb249e2bcad..59e9ef6aab4 100644 --- a/kernel/cgroup_freezer.c +++ b/kernel/cgroup_freezer.c @@ -159,7 +159,7 @@ static bool is_task_frozen_enough(struct task_struct *task) */ static int freezer_can_attach(struct cgroup_subsys *ss, struct cgroup *new_cgroup, - struct task_struct *task) + struct task_struct *task, bool threadgroup) { struct freezer *freezer; @@ -177,6 +177,19 @@ static int freezer_can_attach(struct cgroup_subsys *ss, if (freezer->state == CGROUP_FROZEN) return -EBUSY; + if (threadgroup) { + struct task_struct *c; + + rcu_read_lock(); + list_for_each_entry_rcu(c, &task->thread_group, thread_group) { + if (is_task_frozen_enough(c)) { + rcu_read_unlock(); + return -EBUSY; + } + } + rcu_read_unlock(); + } + return 0; } diff --git a/kernel/compat.c b/kernel/compat.c index 42d56544460..f6c204f07ea 100644 --- a/kernel/compat.c +++ b/kernel/compat.c @@ -882,6 +882,17 @@ compat_sys_rt_sigtimedwait (compat_sigset_t __user *uthese, } +asmlinkage long +compat_sys_rt_tgsigqueueinfo(compat_pid_t tgid, compat_pid_t pid, int sig, + struct compat_siginfo __user *uinfo) +{ + siginfo_t info; + + if (copy_siginfo_from_user32(&info, uinfo)) + return -EFAULT; + return do_rt_tgsigqueueinfo(tgid, pid, sig, &info); +} + #ifdef __ARCH_WANT_COMPAT_SYS_TIME /* compat_time_t is a 32 bit "long" and needs to get converted. */ diff --git a/kernel/cpu.c b/kernel/cpu.c index 395b6974dc8..7c4e2713df0 100644 --- a/kernel/cpu.c +++ b/kernel/cpu.c @@ -34,14 +34,11 @@ static struct { * an ongoing cpu hotplug operation. */ int refcount; -} cpu_hotplug; - -void __init cpu_hotplug_init(void) -{ - cpu_hotplug.active_writer = NULL; - mutex_init(&cpu_hotplug.lock); - cpu_hotplug.refcount = 0; -} +} cpu_hotplug = { + .active_writer = NULL, + .lock = __MUTEX_INITIALIZER(cpu_hotplug.lock), + .refcount = 0, +}; #ifdef CONFIG_HOTPLUG_CPU @@ -395,15 +392,15 @@ int disable_nonboot_cpus(void) if (cpu == first_cpu) continue; error = _cpu_down(cpu, 1); - if (!error) { + if (!error) cpumask_set_cpu(cpu, frozen_cpus); - printk("CPU%d is down\n", cpu); - } else { + else { printk(KERN_ERR "Error taking CPU%d down: %d\n", cpu, error); break; } } + if (!error) { BUG_ON(num_online_cpus() > 1); /* Make sure the CPUs won't be enabled by someone else */ @@ -416,6 +413,14 @@ int disable_nonboot_cpus(void) return error; } +void __weak arch_enable_nonboot_cpus_begin(void) +{ +} + +void __weak arch_enable_nonboot_cpus_end(void) +{ +} + void __ref enable_nonboot_cpus(void) { int cpu, error; @@ -427,6 +432,9 @@ void __ref enable_nonboot_cpus(void) goto out; printk("Enabling non-boot CPUs ...\n"); + + arch_enable_nonboot_cpus_begin(); + for_each_cpu(cpu, frozen_cpus) { error = _cpu_up(cpu, 1); if (!error) { @@ -435,6 +443,9 @@ void __ref enable_nonboot_cpus(void) } printk(KERN_WARNING "Error taking CPU%d up: %d\n", cpu, error); } + + arch_enable_nonboot_cpus_end(); + cpumask_clear(frozen_cpus); out: cpu_maps_update_done(); diff --git a/kernel/cpuset.c b/kernel/cpuset.c index f76db9dcaa0..3cf2183b472 100644 --- a/kernel/cpuset.c +++ b/kernel/cpuset.c @@ -97,12 +97,6 @@ struct cpuset { struct cpuset *parent; /* my parent */ - /* - * Copy of global cpuset_mems_generation as of the most - * recent time this cpuset changed its mems_allowed. - */ - int mems_generation; - struct fmeter fmeter; /* memory_pressure filter */ /* partition number for rebuild_sched_domains() */ @@ -128,10 +122,6 @@ static inline struct cpuset *task_cs(struct task_struct *task) return container_of(task_subsys_state(task, cpuset_subsys_id), struct cpuset, css); } -struct cpuset_hotplug_scanner { - struct cgroup_scanner scan; - struct cgroup *to; -}; /* bits in struct cpuset flags field */ typedef enum { @@ -180,27 +170,6 @@ static inline int is_spread_slab(const struct cpuset *cs) return test_bit(CS_SPREAD_SLAB, &cs->flags); } -/* - * Increment this integer everytime any cpuset changes its - * mems_allowed value. Users of cpusets can track this generation - * number, and avoid having to lock and reload mems_allowed unless - * the cpuset they're using changes generation. - * - * A single, global generation is needed because cpuset_attach_task() could - * reattach a task to a different cpuset, which must not have its - * generation numbers aliased with those of that tasks previous cpuset. - * - * Generations are needed for mems_allowed because one task cannot - * modify another's memory placement. So we must enable every task, - * on every visit to __alloc_pages(), to efficiently check whether - * its current->cpuset->mems_allowed has changed, requiring an update - * of its current->mems_allowed. - * - * Since writes to cpuset_mems_generation are guarded by the cgroup lock - * there is no need to mark it atomic. - */ -static int cpuset_mems_generation; - static struct cpuset top_cpuset = { .flags = ((1 << CS_CPU_EXCLUSIVE) | (1 << CS_MEM_EXCLUSIVE)), }; @@ -232,8 +201,9 @@ static struct cpuset top_cpuset = { * If a task is only holding callback_mutex, then it has read-only * access to cpusets. * - * The task_struct fields mems_allowed and mems_generation may only - * be accessed in the context of that task, so require no locks. + * Now, the task_struct fields mems_allowed and mempolicy may be changed + * by other task, we use alloc_lock in the task_struct fields to protect + * them. * * The cpuset_common_file_read() handlers only hold callback_mutex across * small pieces of code, such as when reading out possibly multi-word @@ -335,75 +305,22 @@ static void guarantee_online_mems(const struct cpuset *cs, nodemask_t *pmask) BUG_ON(!nodes_intersects(*pmask, node_states[N_HIGH_MEMORY])); } -/** - * cpuset_update_task_memory_state - update task memory placement - * - * If the current tasks cpusets mems_allowed changed behind our - * backs, update current->mems_allowed, mems_generation and task NUMA - * mempolicy to the new value. - * - * Task mempolicy is updated by rebinding it relative to the - * current->cpuset if a task has its memory placement changed. - * Do not call this routine if in_interrupt(). - * - * Call without callback_mutex or task_lock() held. May be - * called with or without cgroup_mutex held. Thanks in part to - * 'the_top_cpuset_hack', the task's cpuset pointer will never - * be NULL. This routine also might acquire callback_mutex during - * call. - * - * Reading current->cpuset->mems_generation doesn't need task_lock - * to guard the current->cpuset derefence, because it is guarded - * from concurrent freeing of current->cpuset using RCU. - * - * The rcu_dereference() is technically probably not needed, - * as I don't actually mind if I see a new cpuset pointer but - * an old value of mems_generation. However this really only - * matters on alpha systems using cpusets heavily. If I dropped - * that rcu_dereference(), it would save them a memory barrier. - * For all other arch's, rcu_dereference is a no-op anyway, and for - * alpha systems not using cpusets, another planned optimization, - * avoiding the rcu critical section for tasks in the root cpuset - * which is statically allocated, so can't vanish, will make this - * irrelevant. Better to use RCU as intended, than to engage in - * some cute trick to save a memory barrier that is impossible to - * test, for alpha systems using cpusets heavily, which might not - * even exist. - * - * This routine is needed to update the per-task mems_allowed data, - * within the tasks context, when it is trying to allocate memory - * (in various mm/mempolicy.c routines) and notices that some other - * task has been modifying its cpuset. +/* + * update task's spread flag if cpuset's page/slab spread flag is set + * + * Called with callback_mutex/cgroup_mutex held */ - -void cpuset_update_task_memory_state(void) +static void cpuset_update_task_spread_flag(struct cpuset *cs, + struct task_struct *tsk) { - int my_cpusets_mem_gen; - struct task_struct *tsk = current; - struct cpuset *cs; - - rcu_read_lock(); - my_cpusets_mem_gen = task_cs(tsk)->mems_generation; - rcu_read_unlock(); - - if (my_cpusets_mem_gen != tsk->cpuset_mems_generation) { - mutex_lock(&callback_mutex); - task_lock(tsk); - cs = task_cs(tsk); /* Maybe changed when task not locked */ - guarantee_online_mems(cs, &tsk->mems_allowed); - tsk->cpuset_mems_generation = cs->mems_generation; - if (is_spread_page(cs)) - tsk->flags |= PF_SPREAD_PAGE; - else - tsk->flags &= ~PF_SPREAD_PAGE; - if (is_spread_slab(cs)) - tsk->flags |= PF_SPREAD_SLAB; - else - tsk->flags &= ~PF_SPREAD_SLAB; - task_unlock(tsk); - mutex_unlock(&callback_mutex); - mpol_rebind_task(tsk, &tsk->mems_allowed); - } + if (is_spread_page(cs)) + tsk->flags |= PF_SPREAD_PAGE; + else + tsk->flags &= ~PF_SPREAD_PAGE; + if (is_spread_slab(cs)) + tsk->flags |= PF_SPREAD_SLAB; + else + tsk->flags &= ~PF_SPREAD_SLAB; } /* @@ -521,6 +438,7 @@ static int validate_change(const struct cpuset *cur, const struct cpuset *trial) return 0; } +#ifdef CONFIG_SMP /* * Helper routine for generate_sched_domains(). * Do cpusets a, b have overlapping cpus_allowed masks? @@ -619,8 +537,7 @@ update_domain_attr_tree(struct sched_domain_attr *dattr, struct cpuset *c) * element of the partition (one sched domain) to be passed to * partition_sched_domains(). */ -/* FIXME: see the FIXME in partition_sched_domains() */ -static int generate_sched_domains(struct cpumask **domains, +static int generate_sched_domains(cpumask_var_t **domains, struct sched_domain_attr **attributes) { LIST_HEAD(q); /* queue of cpusets to be scanned */ @@ -628,7 +545,7 @@ static int generate_sched_domains(struct cpumask **domains, struct cpuset **csa; /* array of all cpuset ptrs */ int csn; /* how many cpuset ptrs in csa so far */ int i, j, k; /* indices for partition finding loops */ - struct cpumask *doms; /* resulting partition; i.e. sched domains */ + cpumask_var_t *doms; /* resulting partition; i.e. sched domains */ struct sched_domain_attr *dattr; /* attributes for custom domains */ int ndoms = 0; /* number of sched domains in result */ int nslot; /* next empty doms[] struct cpumask slot */ @@ -639,7 +556,8 @@ static int generate_sched_domains(struct cpumask **domains, /* Special case for the 99% of systems with one, full, sched domain */ if (is_sched_load_balance(&top_cpuset)) { - doms = kmalloc(cpumask_size(), GFP_KERNEL); + ndoms = 1; + doms = alloc_sched_domains(ndoms); if (!doms) goto done; @@ -648,9 +566,8 @@ static int generate_sched_domains(struct cpumask **domains, *dattr = SD_ATTR_INIT; update_domain_attr_tree(dattr, &top_cpuset); } - cpumask_copy(doms, top_cpuset.cpus_allowed); + cpumask_copy(doms[0], top_cpuset.cpus_allowed); - ndoms = 1; goto done; } @@ -718,7 +635,7 @@ restart: * Now we know how many domains to create. * Convert <csn, csa> to <ndoms, doms> and populate cpu masks. */ - doms = kmalloc(ndoms * cpumask_size(), GFP_KERNEL); + doms = alloc_sched_domains(ndoms); if (!doms) goto done; @@ -738,7 +655,7 @@ restart: continue; } - dp = doms + nslot; + dp = doms[nslot]; if (nslot == ndoms) { static int warnings = 10; @@ -800,7 +717,7 @@ done: static void do_rebuild_sched_domains(struct work_struct *unused) { struct sched_domain_attr *attr; - struct cpumask *doms; + cpumask_var_t *doms; int ndoms; get_online_cpus(); @@ -815,6 +732,18 @@ static void do_rebuild_sched_domains(struct work_struct *unused) put_online_cpus(); } +#else /* !CONFIG_SMP */ +static void do_rebuild_sched_domains(struct work_struct *unused) +{ +} + +static int generate_sched_domains(struct cpumask **domains, + struct sched_domain_attr **attributes) +{ + *domains = NULL; + return 1; +} +#endif /* CONFIG_SMP */ static DECLARE_WORK(rebuild_sched_domains_work, do_rebuild_sched_domains); @@ -998,14 +927,6 @@ static int update_cpumask(struct cpuset *cs, struct cpuset *trialcs, * other task, the task_struct mems_allowed that we are hacking * is for our current task, which must allocate new pages for that * migrating memory region. - * - * We call cpuset_update_task_memory_state() before hacking - * our tasks mems_allowed, so that we are assured of being in - * sync with our tasks cpuset, and in particular, callbacks to - * cpuset_update_task_memory_state() from nested page allocations - * won't see any mismatch of our cpuset and task mems_generation - * values, so won't overwrite our hacked tasks mems_allowed - * nodemask. */ static void cpuset_migrate_mm(struct mm_struct *mm, const nodemask_t *from, @@ -1013,17 +934,64 @@ static void cpuset_migrate_mm(struct mm_struct *mm, const nodemask_t *from, { struct task_struct *tsk = current; - cpuset_update_task_memory_state(); - - mutex_lock(&callback_mutex); tsk->mems_allowed = *to; - mutex_unlock(&callback_mutex); do_migrate_pages(mm, from, to, MPOL_MF_MOVE_ALL); - mutex_lock(&callback_mutex); guarantee_online_mems(task_cs(tsk),&tsk->mems_allowed); - mutex_unlock(&callback_mutex); +} + +/* + * cpuset_change_task_nodemask - change task's mems_allowed and mempolicy + * @tsk: the task to change + * @newmems: new nodes that the task will be set + * + * In order to avoid seeing no nodes if the old and new nodes are disjoint, + * we structure updates as setting all new allowed nodes, then clearing newly + * disallowed ones. + * + * Called with task's alloc_lock held + */ +static void cpuset_change_task_nodemask(struct task_struct *tsk, + nodemask_t *newmems) +{ + nodes_or(tsk->mems_allowed, tsk->mems_allowed, *newmems); + mpol_rebind_task(tsk, &tsk->mems_allowed); + mpol_rebind_task(tsk, newmems); + tsk->mems_allowed = *newmems; +} + +/* + * Update task's mems_allowed and rebind its mempolicy and vmas' mempolicy + * of it to cpuset's new mems_allowed, and migrate pages to new nodes if + * memory_migrate flag is set. Called with cgroup_mutex held. + */ +static void cpuset_change_nodemask(struct task_struct *p, + struct cgroup_scanner *scan) +{ + struct mm_struct *mm; + struct cpuset *cs; + int migrate; + const nodemask_t *oldmem = scan->data; + nodemask_t newmems; + + cs = cgroup_cs(scan->cg); + guarantee_online_mems(cs, &newmems); + + task_lock(p); + cpuset_change_task_nodemask(p, &newmems); + task_unlock(p); + + mm = get_task_mm(p); + if (!mm) + return; + + migrate = is_memory_migrate(cs); + + mpol_rebind_mm(mm, &cs->mems_allowed); + if (migrate) + cpuset_migrate_mm(mm, oldmem, &cs->mems_allowed); + mmput(mm); } static void *cpuset_being_rebound; @@ -1032,104 +1000,48 @@ static void *cpuset_being_rebound; * update_tasks_nodemask - Update the nodemasks of tasks in the cpuset. * @cs: the cpuset in which each task's mems_allowed mask needs to be changed * @oldmem: old mems_allowed of cpuset cs + * @heap: if NULL, defer allocating heap memory to cgroup_scan_tasks() * * Called with cgroup_mutex held - * Return 0 if successful, -errno if not. + * No return value. It's guaranteed that cgroup_scan_tasks() always returns 0 + * if @heap != NULL. */ -static int update_tasks_nodemask(struct cpuset *cs, const nodemask_t *oldmem) +static void update_tasks_nodemask(struct cpuset *cs, const nodemask_t *oldmem, + struct ptr_heap *heap) { - struct task_struct *p; - struct mm_struct **mmarray; - int i, n, ntasks; - int migrate; - int fudge; - struct cgroup_iter it; - int retval; + struct cgroup_scanner scan; cpuset_being_rebound = cs; /* causes mpol_dup() rebind */ - fudge = 10; /* spare mmarray[] slots */ - fudge += cpumask_weight(cs->cpus_allowed);/* imagine 1 fork-bomb/cpu */ - retval = -ENOMEM; - - /* - * Allocate mmarray[] to hold mm reference for each task - * in cpuset cs. Can't kmalloc GFP_KERNEL while holding - * tasklist_lock. We could use GFP_ATOMIC, but with a - * few more lines of code, we can retry until we get a big - * enough mmarray[] w/o using GFP_ATOMIC. - */ - while (1) { - ntasks = cgroup_task_count(cs->css.cgroup); /* guess */ - ntasks += fudge; - mmarray = kmalloc(ntasks * sizeof(*mmarray), GFP_KERNEL); - if (!mmarray) - goto done; - read_lock(&tasklist_lock); /* block fork */ - if (cgroup_task_count(cs->css.cgroup) <= ntasks) - break; /* got enough */ - read_unlock(&tasklist_lock); /* try again */ - kfree(mmarray); - } - - n = 0; - - /* Load up mmarray[] with mm reference for each task in cpuset. */ - cgroup_iter_start(cs->css.cgroup, &it); - while ((p = cgroup_iter_next(cs->css.cgroup, &it))) { - struct mm_struct *mm; - - if (n >= ntasks) { - printk(KERN_WARNING - "Cpuset mempolicy rebind incomplete.\n"); - break; - } - mm = get_task_mm(p); - if (!mm) - continue; - mmarray[n++] = mm; - } - cgroup_iter_end(cs->css.cgroup, &it); - read_unlock(&tasklist_lock); + scan.cg = cs->css.cgroup; + scan.test_task = NULL; + scan.process_task = cpuset_change_nodemask; + scan.heap = heap; + scan.data = (nodemask_t *)oldmem; /* - * Now that we've dropped the tasklist spinlock, we can - * rebind the vma mempolicies of each mm in mmarray[] to their - * new cpuset, and release that mm. The mpol_rebind_mm() - * call takes mmap_sem, which we couldn't take while holding - * tasklist_lock. Forks can happen again now - the mpol_dup() - * cpuset_being_rebound check will catch such forks, and rebind - * their vma mempolicies too. Because we still hold the global - * cgroup_mutex, we know that no other rebind effort will - * be contending for the global variable cpuset_being_rebound. + * The mpol_rebind_mm() call takes mmap_sem, which we couldn't + * take while holding tasklist_lock. Forks can happen - the + * mpol_dup() cpuset_being_rebound check will catch such forks, + * and rebind their vma mempolicies too. Because we still hold + * the global cgroup_mutex, we know that no other rebind effort + * will be contending for the global variable cpuset_being_rebound. * It's ok if we rebind the same mm twice; mpol_rebind_mm() * is idempotent. Also migrate pages in each mm to new nodes. */ - migrate = is_memory_migrate(cs); - for (i = 0; i < n; i++) { - struct mm_struct *mm = mmarray[i]; - - mpol_rebind_mm(mm, &cs->mems_allowed); - if (migrate) - cpuset_migrate_mm(mm, oldmem, &cs->mems_allowed); - mmput(mm); - } + cgroup_scan_tasks(&scan); /* We're done rebinding vmas to this cpuset's new mems_allowed. */ - kfree(mmarray); cpuset_being_rebound = NULL; - retval = 0; -done: - return retval; } /* * Handle user request to change the 'mems' memory placement * of a cpuset. Needs to validate the request, update the - * cpusets mems_allowed and mems_generation, and for each - * task in the cpuset, rebind any vma mempolicies and if - * the cpuset is marked 'memory_migrate', migrate the tasks - * pages to the new memory. + * cpusets mems_allowed, and for each task in the cpuset, + * update mems_allowed and rebind task's mempolicy and any vma + * mempolicies and if the cpuset is marked 'memory_migrate', + * migrate the tasks pages to the new memory. * * Call with cgroup_mutex held. May take callback_mutex during call. * Will take tasklist_lock, scan tasklist for tasks in cpuset cs, @@ -1141,6 +1053,7 @@ static int update_nodemask(struct cpuset *cs, struct cpuset *trialcs, { nodemask_t oldmem; int retval; + struct ptr_heap heap; /* * top_cpuset.mems_allowed tracks node_stats[N_HIGH_MEMORY]; @@ -1175,12 +1088,17 @@ static int update_nodemask(struct cpuset *cs, struct cpuset *trialcs, if (retval < 0) goto done; + retval = heap_init(&heap, PAGE_SIZE, GFP_KERNEL, NULL); + if (retval < 0) + goto done; + mutex_lock(&callback_mutex); cs->mems_allowed = trialcs->mems_allowed; - cs->mems_generation = cpuset_mems_generation++; mutex_unlock(&callback_mutex); - retval = update_tasks_nodemask(cs, &oldmem); + update_tasks_nodemask(cs, &oldmem, &heap); + + heap_free(&heap); done: return retval; } @@ -1192,8 +1110,10 @@ int current_cpuset_is_being_rebound(void) static int update_relax_domain_level(struct cpuset *cs, s64 val) { +#ifdef CONFIG_SMP if (val < -1 || val >= SD_LV_MAX) return -EINVAL; +#endif if (val != cs->relax_domain_level) { cs->relax_domain_level = val; @@ -1206,6 +1126,46 @@ static int update_relax_domain_level(struct cpuset *cs, s64 val) } /* + * cpuset_change_flag - make a task's spread flags the same as its cpuset's + * @tsk: task to be updated + * @scan: struct cgroup_scanner containing the cgroup of the task + * + * Called by cgroup_scan_tasks() for each task in a cgroup. + * + * We don't need to re-check for the cgroup/cpuset membership, since we're + * holding cgroup_lock() at this point. + */ +static void cpuset_change_flag(struct task_struct *tsk, + struct cgroup_scanner *scan) +{ + cpuset_update_task_spread_flag(cgroup_cs(scan->cg), tsk); +} + +/* + * update_tasks_flags - update the spread flags of tasks in the cpuset. + * @cs: the cpuset in which each task's spread flags needs to be changed + * @heap: if NULL, defer allocating heap memory to cgroup_scan_tasks() + * + * Called with cgroup_mutex held + * + * The cgroup_scan_tasks() function will scan all the tasks in a cgroup, + * calling callback functions for each. + * + * No return value. It's guaranteed that cgroup_scan_tasks() always returns 0 + * if @heap != NULL. + */ +static void update_tasks_flags(struct cpuset *cs, struct ptr_heap *heap) +{ + struct cgroup_scanner scan; + + scan.cg = cs->css.cgroup; + scan.test_task = NULL; + scan.process_task = cpuset_change_flag; + scan.heap = heap; + cgroup_scan_tasks(&scan); +} + +/* * update_flag - read a 0 or a 1 in a file and update associated flag * bit: the bit to update (see cpuset_flagbits_t) * cs: the cpuset to update @@ -1218,8 +1178,10 @@ static int update_flag(cpuset_flagbits_t bit, struct cpuset *cs, int turning_on) { struct cpuset *trialcs; - int err; int balance_flag_changed; + int spread_flag_changed; + struct ptr_heap heap; + int err; trialcs = alloc_trial_cpuset(cs); if (!trialcs) @@ -1234,9 +1196,16 @@ static int update_flag(cpuset_flagbits_t bit, struct cpuset *cs, if (err < 0) goto out; + err = heap_init(&heap, PAGE_SIZE, GFP_KERNEL, NULL); + if (err < 0) + goto out; + balance_flag_changed = (is_sched_load_balance(cs) != is_sched_load_balance(trialcs)); + spread_flag_changed = ((is_spread_slab(cs) != is_spread_slab(trialcs)) + || (is_spread_page(cs) != is_spread_page(trialcs))); + mutex_lock(&callback_mutex); cs->flags = trialcs->flags; mutex_unlock(&callback_mutex); @@ -1244,6 +1213,9 @@ static int update_flag(cpuset_flagbits_t bit, struct cpuset *cs, if (!cpumask_empty(trialcs->cpus_allowed) && balance_flag_changed) async_rebuild_sched_domains(); + if (spread_flag_changed) + update_tasks_flags(cs, &heap); + heap_free(&heap); out: free_trial_cpuset(trialcs); return err; @@ -1351,46 +1323,92 @@ static int fmeter_getrate(struct fmeter *fmp) static cpumask_var_t cpus_attach; /* Called by cgroups to determine if a cpuset is usable; cgroup_mutex held */ -static int cpuset_can_attach(struct cgroup_subsys *ss, - struct cgroup *cont, struct task_struct *tsk) +static int cpuset_can_attach(struct cgroup_subsys *ss, struct cgroup *cont, + struct task_struct *tsk, bool threadgroup) { + int ret; struct cpuset *cs = cgroup_cs(cont); - int ret = 0; if (cpumask_empty(cs->cpus_allowed) || nodes_empty(cs->mems_allowed)) return -ENOSPC; - if (tsk->flags & PF_THREAD_BOUND) { - mutex_lock(&callback_mutex); - if (!cpumask_equal(&tsk->cpus_allowed, cs->cpus_allowed)) - ret = -EINVAL; - mutex_unlock(&callback_mutex); + /* + * Kthreads bound to specific cpus cannot be moved to a new cpuset; we + * cannot change their cpu affinity and isolating such threads by their + * set of allowed nodes is unnecessary. Thus, cpusets are not + * applicable for such threads. This prevents checking for success of + * set_cpus_allowed_ptr() on all attached tasks before cpus_allowed may + * be changed. + */ + if (tsk->flags & PF_THREAD_BOUND) + return -EINVAL; + + ret = security_task_setscheduler(tsk, 0, NULL); + if (ret) + return ret; + if (threadgroup) { + struct task_struct *c; + + rcu_read_lock(); + list_for_each_entry_rcu(c, &tsk->thread_group, thread_group) { + ret = security_task_setscheduler(c, 0, NULL); + if (ret) { + rcu_read_unlock(); + return ret; + } + } + rcu_read_unlock(); } + return 0; +} + +static void cpuset_attach_task(struct task_struct *tsk, nodemask_t *to, + struct cpuset *cs) +{ + int err; + /* + * can_attach beforehand should guarantee that this doesn't fail. + * TODO: have a better way to handle failure here + */ + err = set_cpus_allowed_ptr(tsk, cpus_attach); + WARN_ON_ONCE(err); + + task_lock(tsk); + cpuset_change_task_nodemask(tsk, to); + task_unlock(tsk); + cpuset_update_task_spread_flag(cs, tsk); - return ret < 0 ? ret : security_task_setscheduler(tsk, 0, NULL); } -static void cpuset_attach(struct cgroup_subsys *ss, - struct cgroup *cont, struct cgroup *oldcont, - struct task_struct *tsk) +static void cpuset_attach(struct cgroup_subsys *ss, struct cgroup *cont, + struct cgroup *oldcont, struct task_struct *tsk, + bool threadgroup) { nodemask_t from, to; struct mm_struct *mm; struct cpuset *cs = cgroup_cs(cont); struct cpuset *oldcs = cgroup_cs(oldcont); - int err; if (cs == &top_cpuset) { cpumask_copy(cpus_attach, cpu_possible_mask); + to = node_possible_map; } else { - mutex_lock(&callback_mutex); guarantee_online_cpus(cs, cpus_attach); - mutex_unlock(&callback_mutex); + guarantee_online_mems(cs, &to); + } + + /* do per-task migration stuff possibly for each in the threadgroup */ + cpuset_attach_task(tsk, &to, cs); + if (threadgroup) { + struct task_struct *c; + rcu_read_lock(); + list_for_each_entry_rcu(c, &tsk->thread_group, thread_group) { + cpuset_attach_task(c, &to, cs); + } + rcu_read_unlock(); } - err = set_cpus_allowed_ptr(tsk, cpus_attach); - if (err) - return; + /* change mm; only needs to be done once even if threadgroup */ from = oldcs->mems_allowed; to = cs->mems_allowed; mm = get_task_mm(tsk); @@ -1452,11 +1470,9 @@ static int cpuset_write_u64(struct cgroup *cgrp, struct cftype *cft, u64 val) break; case FILE_SPREAD_PAGE: retval = update_flag(CS_SPREAD_PAGE, cs, val); - cs->mems_generation = cpuset_mems_generation++; break; case FILE_SPREAD_SLAB: retval = update_flag(CS_SPREAD_SLAB, cs, val); - cs->mems_generation = cpuset_mems_generation++; break; default: retval = -EINVAL; @@ -1706,6 +1722,7 @@ static struct cftype files[] = { .read_u64 = cpuset_read_u64, .write_u64 = cpuset_write_u64, .private = FILE_MEMORY_PRESSURE, + .mode = S_IRUGO, }, { @@ -1795,8 +1812,6 @@ static struct cgroup_subsys_state *cpuset_create( struct cpuset *parent; if (!cont->parent) { - /* This is early initialization for the top cgroup */ - top_cpuset.mems_generation = cpuset_mems_generation++; return &top_cpuset.css; } parent = cgroup_cs(cont->parent); @@ -1808,7 +1823,6 @@ static struct cgroup_subsys_state *cpuset_create( return ERR_PTR(-ENOMEM); } - cpuset_update_task_memory_state(); cs->flags = 0; if (is_spread_page(parent)) set_bit(CS_SPREAD_PAGE, &cs->flags); @@ -1817,7 +1831,6 @@ static struct cgroup_subsys_state *cpuset_create( set_bit(CS_SCHED_LOAD_BALANCE, &cs->flags); cpumask_clear(cs->cpus_allowed); nodes_clear(cs->mems_allowed); - cs->mems_generation = cpuset_mems_generation++; fmeter_init(&cs->fmeter); cs->relax_domain_level = -1; @@ -1836,8 +1849,6 @@ static void cpuset_destroy(struct cgroup_subsys *ss, struct cgroup *cont) { struct cpuset *cs = cgroup_cs(cont); - cpuset_update_task_memory_state(); - if (is_sched_load_balance(cs)) update_flag(CS_SCHED_LOAD_BALANCE, cs, 0); @@ -1858,21 +1869,6 @@ struct cgroup_subsys cpuset_subsys = { .early_init = 1, }; -/* - * cpuset_init_early - just enough so that the calls to - * cpuset_update_task_memory_state() in early init code - * are harmless. - */ - -int __init cpuset_init_early(void) -{ - alloc_bootmem_cpumask_var(&top_cpuset.cpus_allowed); - - top_cpuset.mems_generation = cpuset_mems_generation++; - return 0; -} - - /** * cpuset_init - initialize cpusets at system boot * @@ -1883,11 +1879,13 @@ int __init cpuset_init(void) { int err = 0; + if (!alloc_cpumask_var(&top_cpuset.cpus_allowed, GFP_KERNEL)) + BUG(); + cpumask_setall(top_cpuset.cpus_allowed); nodes_setall(top_cpuset.mems_allowed); fmeter_init(&top_cpuset.fmeter); - top_cpuset.mems_generation = cpuset_mems_generation++; set_bit(CS_SCHED_LOAD_BALANCE, &top_cpuset.flags); top_cpuset.relax_domain_level = -1; @@ -1913,10 +1911,9 @@ int __init cpuset_init(void) static void cpuset_do_move_task(struct task_struct *tsk, struct cgroup_scanner *scan) { - struct cpuset_hotplug_scanner *chsp; + struct cgroup *new_cgroup = scan->data; - chsp = container_of(scan, struct cpuset_hotplug_scanner, scan); - cgroup_attach_task(chsp->to, tsk); + cgroup_attach_task(new_cgroup, tsk); } /** @@ -1932,15 +1929,15 @@ static void cpuset_do_move_task(struct task_struct *tsk, */ static void move_member_tasks_to_cpuset(struct cpuset *from, struct cpuset *to) { - struct cpuset_hotplug_scanner scan; + struct cgroup_scanner scan; - scan.scan.cg = from->css.cgroup; - scan.scan.test_task = NULL; /* select all tasks in cgroup */ - scan.scan.process_task = cpuset_do_move_task; - scan.scan.heap = NULL; - scan.to = to->css.cgroup; + scan.cg = from->css.cgroup; + scan.test_task = NULL; /* select all tasks in cgroup */ + scan.process_task = cpuset_do_move_task; + scan.heap = NULL; + scan.data = to->css.cgroup; - if (cgroup_scan_tasks(&scan.scan)) + if (cgroup_scan_tasks(&scan)) printk(KERN_ERR "move_member_tasks_to_cpuset: " "cgroup_scan_tasks failed\n"); } @@ -2033,7 +2030,7 @@ static void scan_for_empty_cpusets(struct cpuset *root) remove_tasks_in_empty_cpuset(cp); else { update_tasks_cpumask(cp, NULL); - update_tasks_nodemask(cp, &oldmems); + update_tasks_nodemask(cp, &oldmems, NULL); } } } @@ -2054,7 +2051,7 @@ static int cpuset_track_online_cpus(struct notifier_block *unused_nb, unsigned long phase, void *unused_cpu) { struct sched_domain_attr *attr; - struct cpumask *doms; + cpumask_var_t *doms; int ndoms; switch (phase) { @@ -2069,7 +2066,9 @@ static int cpuset_track_online_cpus(struct notifier_block *unused_nb, } cgroup_lock(); + mutex_lock(&callback_mutex); cpumask_copy(top_cpuset.cpus_allowed, cpu_online_mask); + mutex_unlock(&callback_mutex); scan_for_empty_cpusets(&top_cpuset); ndoms = generate_sched_domains(&doms, &attr); cgroup_unlock(); @@ -2092,11 +2091,12 @@ static int cpuset_track_online_nodes(struct notifier_block *self, cgroup_lock(); switch (action) { case MEM_ONLINE: - top_cpuset.mems_allowed = node_states[N_HIGH_MEMORY]; - break; case MEM_OFFLINE: + mutex_lock(&callback_mutex); top_cpuset.mems_allowed = node_states[N_HIGH_MEMORY]; - scan_for_empty_cpusets(&top_cpuset); + mutex_unlock(&callback_mutex); + if (action == MEM_OFFLINE) + scan_for_empty_cpusets(&top_cpuset); break; default: break; @@ -2206,26 +2206,24 @@ static const struct cpuset *nearest_hardwall_ancestor(const struct cpuset *cs) } /** - * cpuset_zone_allowed_softwall - Can we allocate on zone z's memory node? - * @z: is this zone on an allowed node? + * cpuset_node_allowed_softwall - Can we allocate on a memory node? + * @node: is this an allowed node? * @gfp_mask: memory allocation flags * - * If we're in interrupt, yes, we can always allocate. If - * __GFP_THISNODE is set, yes, we can always allocate. If zone - * z's node is in our tasks mems_allowed, yes. If it's not a - * __GFP_HARDWALL request and this zone's nodes is in the nearest - * hardwalled cpuset ancestor to this tasks cpuset, yes. - * If the task has been OOM killed and has access to memory reserves - * as specified by the TIF_MEMDIE flag, yes. + * If we're in interrupt, yes, we can always allocate. If __GFP_THISNODE is + * set, yes, we can always allocate. If node is in our task's mems_allowed, + * yes. If it's not a __GFP_HARDWALL request and this node is in the nearest + * hardwalled cpuset ancestor to this task's cpuset, yes. If the task has been + * OOM killed and has access to memory reserves as specified by the TIF_MEMDIE + * flag, yes. * Otherwise, no. * - * If __GFP_HARDWALL is set, cpuset_zone_allowed_softwall() - * reduces to cpuset_zone_allowed_hardwall(). Otherwise, - * cpuset_zone_allowed_softwall() might sleep, and might allow a zone - * from an enclosing cpuset. + * If __GFP_HARDWALL is set, cpuset_node_allowed_softwall() reduces to + * cpuset_node_allowed_hardwall(). Otherwise, cpuset_node_allowed_softwall() + * might sleep, and might allow a node from an enclosing cpuset. * - * cpuset_zone_allowed_hardwall() only handles the simpler case of - * hardwall cpusets, and never sleeps. + * cpuset_node_allowed_hardwall() only handles the simpler case of hardwall + * cpusets, and never sleeps. * * The __GFP_THISNODE placement logic is really handled elsewhere, * by forcibly using a zonelist starting at a specified node, and by @@ -2264,20 +2262,17 @@ static const struct cpuset *nearest_hardwall_ancestor(const struct cpuset *cs) * GFP_USER - only nodes in current tasks mems allowed ok. * * Rule: - * Don't call cpuset_zone_allowed_softwall if you can't sleep, unless you + * Don't call cpuset_node_allowed_softwall if you can't sleep, unless you * pass in the __GFP_HARDWALL flag set in gfp_flag, which disables * the code that might scan up ancestor cpusets and sleep. */ - -int __cpuset_zone_allowed_softwall(struct zone *z, gfp_t gfp_mask) +int __cpuset_node_allowed_softwall(int node, gfp_t gfp_mask) { - int node; /* node that zone z is on */ const struct cpuset *cs; /* current cpuset ancestors */ int allowed; /* is allocation in zone z allowed? */ if (in_interrupt() || (gfp_mask & __GFP_THISNODE)) return 1; - node = zone_to_nid(z); might_sleep_if(!(gfp_mask & __GFP_HARDWALL)); if (node_isset(node, current->mems_allowed)) return 1; @@ -2306,15 +2301,15 @@ int __cpuset_zone_allowed_softwall(struct zone *z, gfp_t gfp_mask) } /* - * cpuset_zone_allowed_hardwall - Can we allocate on zone z's memory node? - * @z: is this zone on an allowed node? + * cpuset_node_allowed_hardwall - Can we allocate on a memory node? + * @node: is this an allowed node? * @gfp_mask: memory allocation flags * - * If we're in interrupt, yes, we can always allocate. - * If __GFP_THISNODE is set, yes, we can always allocate. If zone - * z's node is in our tasks mems_allowed, yes. If the task has been - * OOM killed and has access to memory reserves as specified by the - * TIF_MEMDIE flag, yes. Otherwise, no. + * If we're in interrupt, yes, we can always allocate. If __GFP_THISNODE is + * set, yes, we can always allocate. If node is in our task's mems_allowed, + * yes. If the task has been OOM killed and has access to memory reserves as + * specified by the TIF_MEMDIE flag, yes. + * Otherwise, no. * * The __GFP_THISNODE placement logic is really handled elsewhere, * by forcibly using a zonelist starting at a specified node, and by @@ -2322,20 +2317,16 @@ int __cpuset_zone_allowed_softwall(struct zone *z, gfp_t gfp_mask) * any node on the zonelist except the first. By the time any such * calls get to this routine, we should just shut up and say 'yes'. * - * Unlike the cpuset_zone_allowed_softwall() variant, above, - * this variant requires that the zone be in the current tasks + * Unlike the cpuset_node_allowed_softwall() variant, above, + * this variant requires that the node be in the current task's * mems_allowed or that we're in interrupt. It does not scan up the * cpuset hierarchy for the nearest enclosing mem_exclusive cpuset. * It never sleeps. */ - -int __cpuset_zone_allowed_hardwall(struct zone *z, gfp_t gfp_mask) +int __cpuset_node_allowed_hardwall(int node, gfp_t gfp_mask) { - int node; /* node that zone z is on */ - if (in_interrupt() || (gfp_mask & __GFP_THISNODE)) return 1; - node = zone_to_nid(z); if (node_isset(node, current->mems_allowed)) return 1; /* @@ -2545,15 +2536,9 @@ const struct file_operations proc_cpuset_operations = { }; #endif /* CONFIG_PROC_PID_CPUSET */ -/* Display task cpus_allowed, mems_allowed in /proc/<pid>/status file. */ +/* Display task mems_allowed in /proc/<pid>/status file. */ void cpuset_task_status_allowed(struct seq_file *m, struct task_struct *task) { - seq_printf(m, "Cpus_allowed:\t"); - seq_cpumask(m, &task->cpus_allowed); - seq_printf(m, "\n"); - seq_printf(m, "Cpus_allowed_list:\t"); - seq_cpumask_list(m, &task->cpus_allowed); - seq_printf(m, "\n"); seq_printf(m, "Mems_allowed:\t"); seq_nodemask(m, &task->mems_allowed); seq_printf(m, "\n"); diff --git a/kernel/cred.c b/kernel/cred.c index 3a039189d70..dd76cfe5f5b 100644 --- a/kernel/cred.c +++ b/kernel/cred.c @@ -18,6 +18,18 @@ #include <linux/cn_proc.h> #include "cred-internals.h" +#if 0 +#define kdebug(FMT, ...) \ + printk("[%-5.5s%5u] "FMT"\n", current->comm, current->pid ,##__VA_ARGS__) +#else +static inline __attribute__((format(printf, 1, 2))) +void no_printk(const char *fmt, ...) +{ +} +#define kdebug(FMT, ...) \ + no_printk("[%-5.5s%5u] "FMT"\n", current->comm, current->pid ,##__VA_ARGS__) +#endif + static struct kmem_cache *cred_jar; /* @@ -36,6 +48,10 @@ static struct thread_group_cred init_tgcred = { */ struct cred init_cred = { .usage = ATOMIC_INIT(4), +#ifdef CONFIG_DEBUG_CREDENTIALS + .subscribers = ATOMIC_INIT(2), + .magic = CRED_MAGIC, +#endif .securebits = SECUREBITS_DEFAULT, .cap_inheritable = CAP_INIT_INH_SET, .cap_permitted = CAP_FULL_SET, @@ -48,6 +64,31 @@ struct cred init_cred = { #endif }; +static inline void set_cred_subscribers(struct cred *cred, int n) +{ +#ifdef CONFIG_DEBUG_CREDENTIALS + atomic_set(&cred->subscribers, n); +#endif +} + +static inline int read_cred_subscribers(const struct cred *cred) +{ +#ifdef CONFIG_DEBUG_CREDENTIALS + return atomic_read(&cred->subscribers); +#else + return 0; +#endif +} + +static inline void alter_cred_subscribers(const struct cred *_cred, int n) +{ +#ifdef CONFIG_DEBUG_CREDENTIALS + struct cred *cred = (struct cred *) _cred; + + atomic_add(n, &cred->subscribers); +#endif +} + /* * Dispose of the shared task group credentials */ @@ -85,15 +126,29 @@ static void put_cred_rcu(struct rcu_head *rcu) { struct cred *cred = container_of(rcu, struct cred, rcu); + kdebug("put_cred_rcu(%p)", cred); + +#ifdef CONFIG_DEBUG_CREDENTIALS + if (cred->magic != CRED_MAGIC_DEAD || + atomic_read(&cred->usage) != 0 || + read_cred_subscribers(cred) != 0) + panic("CRED: put_cred_rcu() sees %p with" + " mag %x, put %p, usage %d, subscr %d\n", + cred, cred->magic, cred->put_addr, + atomic_read(&cred->usage), + read_cred_subscribers(cred)); +#else if (atomic_read(&cred->usage) != 0) panic("CRED: put_cred_rcu() sees %p with usage %d\n", cred, atomic_read(&cred->usage)); +#endif security_cred_free(cred); key_put(cred->thread_keyring); key_put(cred->request_key_auth); release_tgcred(cred); - put_group_info(cred->group_info); + if (cred->group_info) + put_group_info(cred->group_info); free_uid(cred->user); kmem_cache_free(cred_jar, cred); } @@ -106,12 +161,90 @@ static void put_cred_rcu(struct rcu_head *rcu) */ void __put_cred(struct cred *cred) { + kdebug("__put_cred(%p{%d,%d})", cred, + atomic_read(&cred->usage), + read_cred_subscribers(cred)); + BUG_ON(atomic_read(&cred->usage) != 0); +#ifdef CONFIG_DEBUG_CREDENTIALS + BUG_ON(read_cred_subscribers(cred) != 0); + cred->magic = CRED_MAGIC_DEAD; + cred->put_addr = __builtin_return_address(0); +#endif + BUG_ON(cred == current->cred); + BUG_ON(cred == current->real_cred); call_rcu(&cred->rcu, put_cred_rcu); } EXPORT_SYMBOL(__put_cred); +/* + * Clean up a task's credentials when it exits + */ +void exit_creds(struct task_struct *tsk) +{ + struct cred *cred; + + kdebug("exit_creds(%u,%p,%p,{%d,%d})", tsk->pid, tsk->real_cred, tsk->cred, + atomic_read(&tsk->cred->usage), + read_cred_subscribers(tsk->cred)); + + cred = (struct cred *) tsk->real_cred; + tsk->real_cred = NULL; + validate_creds(cred); + alter_cred_subscribers(cred, -1); + put_cred(cred); + + cred = (struct cred *) tsk->cred; + tsk->cred = NULL; + validate_creds(cred); + alter_cred_subscribers(cred, -1); + put_cred(cred); + + cred = (struct cred *) tsk->replacement_session_keyring; + if (cred) { + tsk->replacement_session_keyring = NULL; + validate_creds(cred); + put_cred(cred); + } +} + +/* + * Allocate blank credentials, such that the credentials can be filled in at a + * later date without risk of ENOMEM. + */ +struct cred *cred_alloc_blank(void) +{ + struct cred *new; + + new = kmem_cache_zalloc(cred_jar, GFP_KERNEL); + if (!new) + return NULL; + +#ifdef CONFIG_KEYS + new->tgcred = kzalloc(sizeof(*new->tgcred), GFP_KERNEL); + if (!new->tgcred) { + kfree(new); + return NULL; + } + atomic_set(&new->tgcred->usage, 1); +#endif + + atomic_set(&new->usage, 1); + + if (security_cred_alloc_blank(new, GFP_KERNEL) < 0) + goto error; + +#ifdef CONFIG_DEBUG_CREDENTIALS + new->magic = CRED_MAGIC; +#endif + return new; + +error: + abort_creds(new); + return NULL; +} + /** * prepare_creds - Prepare a new set of credentials for modification * @@ -132,16 +265,19 @@ struct cred *prepare_creds(void) const struct cred *old; struct cred *new; - BUG_ON(atomic_read(&task->real_cred->usage) < 1); + validate_process_creds(); new = kmem_cache_alloc(cred_jar, GFP_KERNEL); if (!new) return NULL; + kdebug("prepare_creds() alloc %p", new); + old = task->cred; memcpy(new, old, sizeof(struct cred)); atomic_set(&new->usage, 1); + set_cred_subscribers(new, 0); get_group_info(new->group_info); get_uid(new->user); @@ -157,6 +293,7 @@ struct cred *prepare_creds(void) if (security_prepare_creds(new, old, GFP_KERNEL) < 0) goto error; + validate_creds(new); return new; error: @@ -167,7 +304,7 @@ EXPORT_SYMBOL(prepare_creds); /* * Prepare credentials for current to perform an execve() - * - The caller must hold current->cred_exec_mutex + * - The caller must hold current->cred_guard_mutex */ struct cred *prepare_exec_creds(void) { @@ -229,9 +366,12 @@ struct cred *prepare_usermodehelper_creds(void) if (!new) return NULL; + kdebug("prepare_usermodehelper_creds() alloc %p", new); + memcpy(new, &init_cred, sizeof(struct cred)); atomic_set(&new->usage, 1); + set_cred_subscribers(new, 0); get_group_info(new->group_info); get_uid(new->user); @@ -250,6 +390,7 @@ struct cred *prepare_usermodehelper_creds(void) #endif if (security_prepare_creds(new, &init_cred, GFP_ATOMIC) < 0) goto error; + validate_creds(new); BUG_ON(atomic_read(&new->usage) != 1); return new; @@ -276,7 +417,7 @@ int copy_creds(struct task_struct *p, unsigned long clone_flags) struct cred *new; int ret; - mutex_init(&p->cred_exec_mutex); + mutex_init(&p->cred_guard_mutex); if ( #ifdef CONFIG_KEYS @@ -286,6 +427,10 @@ int copy_creds(struct task_struct *p, unsigned long clone_flags) ) { p->real_cred = get_cred(p->cred); get_cred(p->cred); + alter_cred_subscribers(p->cred, 2); + kdebug("share_creds(%p{%d,%d})", + p->cred, atomic_read(&p->cred->usage), + read_cred_subscribers(p->cred)); atomic_inc(&p->cred->user->processes); return 0; } @@ -331,6 +476,8 @@ int copy_creds(struct task_struct *p, unsigned long clone_flags) atomic_inc(&new->user->processes); p->cred = p->real_cred = get_cred(new); + alter_cred_subscribers(new, 2); + validate_creds(new); return 0; error_put: @@ -355,13 +502,20 @@ error_put: int commit_creds(struct cred *new) { struct task_struct *task = current; - const struct cred *old; + const struct cred *old = task->real_cred; + + kdebug("commit_creds(%p{%d,%d})", new, + atomic_read(&new->usage), + read_cred_subscribers(new)); - BUG_ON(task->cred != task->real_cred); - BUG_ON(atomic_read(&task->real_cred->usage) < 2); + BUG_ON(task->cred != old); +#ifdef CONFIG_DEBUG_CREDENTIALS + BUG_ON(read_cred_subscribers(old) < 2); + validate_creds(old); + validate_creds(new); +#endif BUG_ON(atomic_read(&new->usage) < 1); - old = task->real_cred; security_commit_creds(new, old); get_cred(new); /* we will require a ref for the subj creds too */ @@ -390,12 +544,14 @@ int commit_creds(struct cred *new) * cheaply with the new uid cache, so if it matters * we should be checking for it. -DaveM */ + alter_cred_subscribers(new, 2); if (new->user != old->user) atomic_inc(&new->user->processes); rcu_assign_pointer(task->real_cred, new); rcu_assign_pointer(task->cred, new); if (new->user != old->user) atomic_dec(&old->user->processes); + alter_cred_subscribers(old, -2); sched_switch_user(task); @@ -428,6 +584,13 @@ EXPORT_SYMBOL(commit_creds); */ void abort_creds(struct cred *new) { + kdebug("abort_creds(%p{%d,%d})", new, + atomic_read(&new->usage), + read_cred_subscribers(new)); + +#ifdef CONFIG_DEBUG_CREDENTIALS + BUG_ON(read_cred_subscribers(new) != 0); +#endif BUG_ON(atomic_read(&new->usage) < 1); put_cred(new); } @@ -444,7 +607,20 @@ const struct cred *override_creds(const struct cred *new) { const struct cred *old = current->cred; - rcu_assign_pointer(current->cred, get_cred(new)); + kdebug("override_creds(%p{%d,%d})", new, + atomic_read(&new->usage), + read_cred_subscribers(new)); + + validate_creds(old); + validate_creds(new); + get_cred(new); + alter_cred_subscribers(new, 1); + rcu_assign_pointer(current->cred, new); + alter_cred_subscribers(old, -1); + + kdebug("override_creds() = %p{%d,%d}", old, + atomic_read(&old->usage), + read_cred_subscribers(old)); return old; } EXPORT_SYMBOL(override_creds); @@ -460,7 +636,15 @@ void revert_creds(const struct cred *old) { const struct cred *override = current->cred; + kdebug("revert_creds(%p{%d,%d})", old, + atomic_read(&old->usage), + read_cred_subscribers(old)); + + validate_creds(old); + validate_creds(override); + alter_cred_subscribers(old, 1); rcu_assign_pointer(current->cred, old); + alter_cred_subscribers(override, -1); put_cred(override); } EXPORT_SYMBOL(revert_creds); @@ -502,11 +686,15 @@ struct cred *prepare_kernel_cred(struct task_struct *daemon) if (!new) return NULL; + kdebug("prepare_kernel_cred() alloc %p", new); + if (daemon) old = get_task_cred(daemon); else old = get_cred(&init_cred); + validate_creds(old); + *new = *old; get_uid(new->user); get_group_info(new->group_info); @@ -526,7 +714,9 @@ struct cred *prepare_kernel_cred(struct task_struct *daemon) goto error; atomic_set(&new->usage, 1); + set_cred_subscribers(new, 0); put_cred(old); + validate_creds(new); return new; error: @@ -589,3 +779,114 @@ int set_create_files_as(struct cred *new, struct inode *inode) return security_kernel_create_files_as(new, inode); } EXPORT_SYMBOL(set_create_files_as); + +#ifdef CONFIG_DEBUG_CREDENTIALS + +bool creds_are_invalid(const struct cred *cred) +{ + if (cred->magic != CRED_MAGIC) + return true; + if (atomic_read(&cred->usage) < atomic_read(&cred->subscribers)) + return true; +#ifdef CONFIG_SECURITY_SELINUX + if (selinux_is_enabled()) { + if ((unsigned long) cred->security < PAGE_SIZE) + return true; + if ((*(u32 *)cred->security & 0xffffff00) == + (POISON_FREE << 24 | POISON_FREE << 16 | POISON_FREE << 8)) + return true; + } +#endif + return false; +} +EXPORT_SYMBOL(creds_are_invalid); + +/* + * dump invalid credentials + */ +static void dump_invalid_creds(const struct cred *cred, const char *label, + const struct task_struct *tsk) +{ + printk(KERN_ERR "CRED: %s credentials: %p %s%s%s\n", + label, cred, + cred == &init_cred ? "[init]" : "", + cred == tsk->real_cred ? "[real]" : "", + cred == tsk->cred ? "[eff]" : ""); + printk(KERN_ERR "CRED: ->magic=%x, put_addr=%p\n", + cred->magic, cred->put_addr); + printk(KERN_ERR "CRED: ->usage=%d, subscr=%d\n", + atomic_read(&cred->usage), + read_cred_subscribers(cred)); + printk(KERN_ERR "CRED: ->*uid = { %d,%d,%d,%d }\n", + cred->uid, cred->euid, cred->suid, cred->fsuid); + printk(KERN_ERR "CRED: ->*gid = { %d,%d,%d,%d }\n", + cred->gid, cred->egid, cred->sgid, cred->fsgid); +#ifdef CONFIG_SECURITY + printk(KERN_ERR "CRED: ->security is %p\n", cred->security); + if ((unsigned long) cred->security >= PAGE_SIZE && + (((unsigned long) cred->security & 0xffffff00) != + (POISON_FREE << 24 | POISON_FREE << 16 | POISON_FREE << 8))) + printk(KERN_ERR "CRED: ->security {%x, %x}\n", + ((u32*)cred->security)[0], + ((u32*)cred->security)[1]); +#endif +} + +/* + * report use of invalid credentials + */ +void __invalid_creds(const struct cred *cred, const char *file, unsigned line) +{ + printk(KERN_ERR "CRED: Invalid credentials\n"); + printk(KERN_ERR "CRED: At %s:%u\n", file, line); + dump_invalid_creds(cred, "Specified", current); + BUG(); +} +EXPORT_SYMBOL(__invalid_creds); + +/* + * check the credentials on a process + */ +void __validate_process_creds(struct task_struct *tsk, + const char *file, unsigned line) +{ + if (tsk->cred == tsk->real_cred) { + if (unlikely(read_cred_subscribers(tsk->cred) < 2 || + creds_are_invalid(tsk->cred))) + goto invalid_creds; + } else { + if (unlikely(read_cred_subscribers(tsk->real_cred) < 1 || + read_cred_subscribers(tsk->cred) < 1 || + creds_are_invalid(tsk->real_cred) || + creds_are_invalid(tsk->cred))) + goto invalid_creds; + } + return; + +invalid_creds: + printk(KERN_ERR "CRED: Invalid process credentials\n"); + printk(KERN_ERR "CRED: At %s:%u\n", file, line); + + dump_invalid_creds(tsk->real_cred, "Real", tsk); + if (tsk->cred != tsk->real_cred) + dump_invalid_creds(tsk->cred, "Effective", tsk); + else + printk(KERN_ERR "CRED: Effective creds == Real creds\n"); + BUG(); +} +EXPORT_SYMBOL(__validate_process_creds); + +/* + * check creds for do_exit() + */ +void validate_creds_for_do_exit(struct task_struct *tsk) +{ + kdebug("validate_creds_for_do_exit(%p,%p{%d,%d})", + tsk->real_cred, tsk->cred, + atomic_read(&tsk->cred->usage), + read_cred_subscribers(tsk->cred)); + + __validate_process_creds(tsk, __FILE__, __LINE__); +} + +#endif /* CONFIG_DEBUG_CREDENTIALS */ diff --git a/kernel/delayacct.c b/kernel/delayacct.c index abb6e17505e..ead9b610aa7 100644 --- a/kernel/delayacct.c +++ b/kernel/delayacct.c @@ -15,6 +15,7 @@ #include <linux/sched.h> #include <linux/slab.h> +#include <linux/taskstats.h> #include <linux/time.h> #include <linux/sysctl.h> #include <linux/delayacct.h> diff --git a/kernel/dma-coherent.c b/kernel/dma-coherent.c deleted file mode 100644 index 962a3b574f2..00000000000 --- a/kernel/dma-coherent.c +++ /dev/null @@ -1,176 +0,0 @@ -/* - * Coherent per-device memory handling. - * Borrowed from i386 - */ -#include <linux/kernel.h> -#include <linux/dma-mapping.h> - -struct dma_coherent_mem { - void *virt_base; - u32 device_base; - int size; - int flags; - unsigned long *bitmap; -}; - -int dma_declare_coherent_memory(struct device *dev, dma_addr_t bus_addr, - dma_addr_t device_addr, size_t size, int flags) -{ - void __iomem *mem_base = NULL; - int pages = size >> PAGE_SHIFT; - int bitmap_size = BITS_TO_LONGS(pages) * sizeof(long); - - if ((flags & (DMA_MEMORY_MAP | DMA_MEMORY_IO)) == 0) - goto out; - if (!size) - goto out; - if (dev->dma_mem) - goto out; - - /* FIXME: this routine just ignores DMA_MEMORY_INCLUDES_CHILDREN */ - - mem_base = ioremap(bus_addr, size); - if (!mem_base) - goto out; - - dev->dma_mem = kzalloc(sizeof(struct dma_coherent_mem), GFP_KERNEL); - if (!dev->dma_mem) - goto out; - dev->dma_mem->bitmap = kzalloc(bitmap_size, GFP_KERNEL); - if (!dev->dma_mem->bitmap) - goto free1_out; - - dev->dma_mem->virt_base = mem_base; - dev->dma_mem->device_base = device_addr; - dev->dma_mem->size = pages; - dev->dma_mem->flags = flags; - - if (flags & DMA_MEMORY_MAP) - return DMA_MEMORY_MAP; - - return DMA_MEMORY_IO; - - free1_out: - kfree(dev->dma_mem); - out: - if (mem_base) - iounmap(mem_base); - return 0; -} -EXPORT_SYMBOL(dma_declare_coherent_memory); - -void dma_release_declared_memory(struct device *dev) -{ - struct dma_coherent_mem *mem = dev->dma_mem; - - if (!mem) - return; - dev->dma_mem = NULL; - iounmap(mem->virt_base); - kfree(mem->bitmap); - kfree(mem); -} -EXPORT_SYMBOL(dma_release_declared_memory); - -void *dma_mark_declared_memory_occupied(struct device *dev, - dma_addr_t device_addr, size_t size) -{ - struct dma_coherent_mem *mem = dev->dma_mem; - int pos, err; - - size += device_addr & ~PAGE_MASK; - - if (!mem) - return ERR_PTR(-EINVAL); - - pos = (device_addr - mem->device_base) >> PAGE_SHIFT; - err = bitmap_allocate_region(mem->bitmap, pos, get_order(size)); - if (err != 0) - return ERR_PTR(err); - return mem->virt_base + (pos << PAGE_SHIFT); -} -EXPORT_SYMBOL(dma_mark_declared_memory_occupied); - -/** - * dma_alloc_from_coherent() - try to allocate memory from the per-device coherent area - * - * @dev: device from which we allocate memory - * @size: size of requested memory area - * @dma_handle: This will be filled with the correct dma handle - * @ret: This pointer will be filled with the virtual address - * to allocated area. - * - * This function should be only called from per-arch dma_alloc_coherent() - * to support allocation from per-device coherent memory pools. - * - * Returns 0 if dma_alloc_coherent should continue with allocating from - * generic memory areas, or !0 if dma_alloc_coherent should return @ret. - */ -int dma_alloc_from_coherent(struct device *dev, ssize_t size, - dma_addr_t *dma_handle, void **ret) -{ - struct dma_coherent_mem *mem; - int order = get_order(size); - int pageno; - - if (!dev) - return 0; - mem = dev->dma_mem; - if (!mem) - return 0; - - *ret = NULL; - - if (unlikely(size > (mem->size << PAGE_SHIFT))) - goto err; - - pageno = bitmap_find_free_region(mem->bitmap, mem->size, order); - if (unlikely(pageno < 0)) - goto err; - - /* - * Memory was found in the per-device area. - */ - *dma_handle = mem->device_base + (pageno << PAGE_SHIFT); - *ret = mem->virt_base + (pageno << PAGE_SHIFT); - memset(*ret, 0, size); - - return 1; - -err: - /* - * In the case where the allocation can not be satisfied from the - * per-device area, try to fall back to generic memory if the - * constraints allow it. - */ - return mem->flags & DMA_MEMORY_EXCLUSIVE; -} -EXPORT_SYMBOL(dma_alloc_from_coherent); - -/** - * dma_release_from_coherent() - try to free the memory allocated from per-device coherent memory pool - * @dev: device from which the memory was allocated - * @order: the order of pages allocated - * @vaddr: virtual address of allocated pages - * - * This checks whether the memory was allocated from the per-device - * coherent memory pool and if so, releases that memory. - * - * Returns 1 if we correctly released the memory, or 0 if - * dma_release_coherent() should proceed with releasing memory from - * generic pools. - */ -int dma_release_from_coherent(struct device *dev, int order, void *vaddr) -{ - struct dma_coherent_mem *mem = dev ? dev->dma_mem : NULL; - - if (mem && vaddr >= mem->virt_base && vaddr < - (mem->virt_base + (mem->size << PAGE_SHIFT))) { - int page = (vaddr - mem->virt_base) >> PAGE_SHIFT; - - bitmap_release_region(mem->bitmap, page, order); - return 1; - } - return 0; -} -EXPORT_SYMBOL(dma_release_from_coherent); diff --git a/kernel/exec_domain.c b/kernel/exec_domain.c index 667c841c295..c35452cadde 100644 --- a/kernel/exec_domain.c +++ b/kernel/exec_domain.c @@ -18,6 +18,7 @@ #include <linux/syscalls.h> #include <linux/sysctl.h> #include <linux/types.h> +#include <linux/fs_struct.h> static void default_handler(int, struct pt_regs *); @@ -145,28 +146,6 @@ __set_personality(u_long personality) return 0; } - if (atomic_read(¤t->fs->count) != 1) { - struct fs_struct *fsp, *ofsp; - - fsp = copy_fs_struct(current->fs); - if (fsp == NULL) { - module_put(ep->module); - return -ENOMEM; - } - - task_lock(current); - ofsp = current->fs; - current->fs = fsp; - task_unlock(current); - - put_fs_struct(ofsp); - } - - /* - * At that point we are guaranteed to be the sole owner of - * current->fs. - */ - current->personality = personality; oep = current_thread_info()->exec_domain; current_thread_info()->exec_domain = ep; diff --git a/kernel/exit.c b/kernel/exit.c index 167e1e3ad7c..1143012951e 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -12,7 +12,6 @@ #include <linux/completion.h> #include <linux/personality.h> #include <linux/tty.h> -#include <linux/mnt_namespace.h> #include <linux/iocontext.h> #include <linux/key.h> #include <linux/security.h> @@ -46,8 +45,11 @@ #include <linux/blkdev.h> #include <linux/task_io_accounting_ops.h> #include <linux/tracehook.h> +#include <linux/fs_struct.h> #include <linux/init_task.h> -#include <trace/sched.h> +#include <linux/perf_event.h> +#include <trace/events/sched.h> +#include <linux/hw_breakpoint.h> #include <asm/uaccess.h> #include <asm/unistd.h> @@ -55,17 +57,8 @@ #include <asm/mmu_context.h> #include "cred-internals.h" -DEFINE_TRACE(sched_process_free); -DEFINE_TRACE(sched_process_exit); -DEFINE_TRACE(sched_process_wait); - static void exit_mm(struct task_struct * tsk); -static inline int task_detached(struct task_struct *p) -{ - return p->exit_signal == -1; -} - static void __unhash_process(struct task_struct *p) { nr_threads--; @@ -118,9 +111,9 @@ static void __exit_signal(struct task_struct *tsk) * We won't ever get here for the group leader, since it * will have been the last reference on the signal_struct. */ - sig->utime = cputime_add(sig->utime, task_utime(tsk)); - sig->stime = cputime_add(sig->stime, task_stime(tsk)); - sig->gtime = cputime_add(sig->gtime, task_gtime(tsk)); + sig->utime = cputime_add(sig->utime, tsk->utime); + sig->stime = cputime_add(sig->stime, tsk->stime); + sig->gtime = cputime_add(sig->gtime, tsk->gtime); sig->min_flt += tsk->min_flt; sig->maj_flt += tsk->maj_flt; sig->nvcsw += tsk->nvcsw; @@ -162,6 +155,9 @@ static void delayed_put_task_struct(struct rcu_head *rhp) { struct task_struct *tsk = container_of(rhp, struct task_struct, rcu); +#ifdef CONFIG_PERF_EVENTS + WARN_ON_ONCE(tsk->perf_event_ctxp); +#endif trace_sched_process_free(tsk); put_task_struct(tsk); } @@ -178,6 +174,7 @@ repeat: atomic_dec(&__task_cred(p)->user->processes); proc_flush_task(p); + write_lock_irq(&tasklist_lock); tracehook_finish_release_task(p); __exit_signal(p); @@ -362,16 +359,12 @@ static void reparent_to_kthreadd(void) void __set_special_pids(struct pid *pid) { struct task_struct *curr = current->group_leader; - pid_t nr = pid_nr(pid); - if (task_session(curr) != pid) { + if (task_session(curr) != pid) change_pid(curr, PIDTYPE_SID, pid); - set_task_session(curr, nr); - } - if (task_pgrp(curr) != pid) { + + if (task_pgrp(curr) != pid) change_pid(curr, PIDTYPE_PGID, pid); - set_task_pgrp(curr, nr); - } } static void set_special_pids(struct pid *pid) @@ -382,9 +375,8 @@ static void set_special_pids(struct pid *pid) } /* - * Let kernel threads use this to say that they - * allow a certain signal (since daemonize() will - * have disabled all of them by default). + * Let kernel threads use this to say that they allow a certain signal. + * Must not be used if kthread was cloned with CLONE_SIGHAND. */ int allow_signal(int sig) { @@ -392,14 +384,14 @@ int allow_signal(int sig) return -EINVAL; spin_lock_irq(¤t->sighand->siglock); + /* This is only needed for daemonize()'ed kthreads */ sigdelset(¤t->blocked, sig); - if (!current->mm) { - /* Kernel threads handle their own signals. - Let the signal code know it'll be handled, so - that they don't get converted to SIGKILL or - just silently dropped */ - current->sighand->action[(sig)-1].sa.sa_handler = (void __user *)2; - } + /* + * Kernel threads handle their own signals. Let the signal code + * know it'll be handled, so that they don't get converted to + * SIGKILL or just silently dropped. + */ + current->sighand->action[(sig)-1].sa.sa_handler = (void __user *)2; recalc_sigpending(); spin_unlock_irq(¤t->sighand->siglock); return 0; @@ -429,7 +421,6 @@ EXPORT_SYMBOL(disallow_signal); void daemonize(const char *name, ...) { va_list args; - struct fs_struct *fs; sigset_t blocked; va_start(args, name); @@ -462,11 +453,7 @@ void daemonize(const char *name, ...) /* Become as one with the init task */ - exit_fs(current); /* current->fs->count--; */ - fs = init_task.fs; - current->fs = fs; - atomic_inc(&fs->count); - + daemonize_fs_struct(); exit_files(current); current->files = init_task.files; atomic_inc(¤t->files->count); @@ -565,30 +552,6 @@ void exit_files(struct task_struct *tsk) } } -void put_fs_struct(struct fs_struct *fs) -{ - /* No need to hold fs->lock if we are killing it */ - if (atomic_dec_and_test(&fs->count)) { - path_put(&fs->root); - path_put(&fs->pwd); - kmem_cache_free(fs_cachep, fs); - } -} - -void exit_fs(struct task_struct *tsk) -{ - struct fs_struct * fs = tsk->fs; - - if (fs) { - task_lock(tsk); - tsk->fs = NULL; - task_unlock(tsk); - put_fs_struct(fs); - } -} - -EXPORT_SYMBOL_GPL(exit_fs); - #ifdef CONFIG_MM_OWNER /* * Task p is exiting and it owned mm, lets find a new owner for it @@ -627,7 +590,7 @@ retry: /* * Search in the siblings */ - list_for_each_entry(c, &p->parent->children, sibling) { + list_for_each_entry(c, &p->real_parent->children, sibling) { if (c->mm == mm) goto assign_new_owner; } @@ -732,119 +695,6 @@ static void exit_mm(struct task_struct * tsk) } /* - * Return nonzero if @parent's children should reap themselves. - * - * Called with write_lock_irq(&tasklist_lock) held. - */ -static int ignoring_children(struct task_struct *parent) -{ - int ret; - struct sighand_struct *psig = parent->sighand; - unsigned long flags; - spin_lock_irqsave(&psig->siglock, flags); - ret = (psig->action[SIGCHLD-1].sa.sa_handler == SIG_IGN || - (psig->action[SIGCHLD-1].sa.sa_flags & SA_NOCLDWAIT)); - spin_unlock_irqrestore(&psig->siglock, flags); - return ret; -} - -/* - * Detach all tasks we were using ptrace on. - * Any that need to be release_task'd are put on the @dead list. - * - * Called with write_lock(&tasklist_lock) held. - */ -static void ptrace_exit(struct task_struct *parent, struct list_head *dead) -{ - struct task_struct *p, *n; - int ign = -1; - - list_for_each_entry_safe(p, n, &parent->ptraced, ptrace_entry) { - __ptrace_unlink(p); - - if (p->exit_state != EXIT_ZOMBIE) - continue; - - /* - * If it's a zombie, our attachedness prevented normal - * parent notification or self-reaping. Do notification - * now if it would have happened earlier. If it should - * reap itself, add it to the @dead list. We can't call - * release_task() here because we already hold tasklist_lock. - * - * If it's our own child, there is no notification to do. - * But if our normal children self-reap, then this child - * was prevented by ptrace and we must reap it now. - */ - if (!task_detached(p) && thread_group_empty(p)) { - if (!same_thread_group(p->real_parent, parent)) - do_notify_parent(p, p->exit_signal); - else { - if (ign < 0) - ign = ignoring_children(parent); - if (ign) - p->exit_signal = -1; - } - } - - if (task_detached(p)) { - /* - * Mark it as in the process of being reaped. - */ - p->exit_state = EXIT_DEAD; - list_add(&p->ptrace_entry, dead); - } - } -} - -/* - * Finish up exit-time ptrace cleanup. - * - * Called without locks. - */ -static void ptrace_exit_finish(struct task_struct *parent, - struct list_head *dead) -{ - struct task_struct *p, *n; - - BUG_ON(!list_empty(&parent->ptraced)); - - list_for_each_entry_safe(p, n, dead, ptrace_entry) { - list_del_init(&p->ptrace_entry); - release_task(p); - } -} - -static void reparent_thread(struct task_struct *p, struct task_struct *father) -{ - if (p->pdeath_signal) - /* We already hold the tasklist_lock here. */ - group_send_sig_info(p->pdeath_signal, SEND_SIG_NOINFO, p); - - list_move_tail(&p->sibling, &p->real_parent->children); - - /* If this is a threaded reparent there is no need to - * notify anyone anything has happened. - */ - if (same_thread_group(p->real_parent, father)) - return; - - /* We don't want people slaying init. */ - if (!task_detached(p)) - p->exit_signal = SIGCHLD; - - /* If we'd notified the old parent about this child's death, - * also notify the new parent. - */ - if (!ptrace_reparented(p) && - p->exit_state == EXIT_ZOMBIE && - !task_detached(p) && thread_group_empty(p)) - do_notify_parent(p, p->exit_signal); - - kill_orphaned_pgrp(p, father); -} - -/* * When we die, we re-parent all our children. * Try to give them to another thread in our thread * group, and if no such member exists, give it to @@ -883,31 +733,68 @@ static struct task_struct *find_new_reaper(struct task_struct *father) return pid_ns->child_reaper; } +/* +* Any that need to be release_task'd are put on the @dead list. + */ +static void reparent_thread(struct task_struct *father, struct task_struct *p, + struct list_head *dead) +{ + if (p->pdeath_signal) + group_send_sig_info(p->pdeath_signal, SEND_SIG_NOINFO, p); + + list_move_tail(&p->sibling, &p->real_parent->children); + + if (task_detached(p)) + return; + /* + * If this is a threaded reparent there is no need to + * notify anyone anything has happened. + */ + if (same_thread_group(p->real_parent, father)) + return; + + /* We don't want people slaying init. */ + p->exit_signal = SIGCHLD; + + /* If it has exited notify the new parent about this child's death. */ + if (!task_ptrace(p) && + p->exit_state == EXIT_ZOMBIE && thread_group_empty(p)) { + do_notify_parent(p, p->exit_signal); + if (task_detached(p)) { + p->exit_state = EXIT_DEAD; + list_move_tail(&p->sibling, dead); + } + } + + kill_orphaned_pgrp(p, father); +} + static void forget_original_parent(struct task_struct *father) { struct task_struct *p, *n, *reaper; - LIST_HEAD(ptrace_dead); + LIST_HEAD(dead_children); + + exit_ptrace(father); write_lock_irq(&tasklist_lock); reaper = find_new_reaper(father); - /* - * First clean up ptrace if we were using it. - */ - ptrace_exit(father, &ptrace_dead); list_for_each_entry_safe(p, n, &father->children, sibling) { p->real_parent = reaper; if (p->parent == father) { - BUG_ON(p->ptrace); + BUG_ON(task_ptrace(p)); p->parent = p->real_parent; } - reparent_thread(p, father); + reparent_thread(father, p, &dead_children); } - write_unlock_irq(&tasklist_lock); + BUG_ON(!list_empty(&father->children)); - ptrace_exit_finish(father, &ptrace_dead); + list_for_each_entry_safe(p, n, &dead_children, sibling) { + list_del_init(&p->sibling); + release_task(p); + } } /* @@ -950,8 +837,7 @@ static void exit_notify(struct task_struct *tsk, int group_dead) */ if (tsk->exit_signal != SIGCHLD && !task_detached(tsk) && (tsk->parent_exec_id != tsk->real_parent->self_exec_id || - tsk->self_exec_id != tsk->parent_exec_id) && - !capable(CAP_KILL)) + tsk->self_exec_id != tsk->parent_exec_id)) tsk->exit_signal = SIGCHLD; signal = tracehook_notify_death(tsk, &cookie, group_dead); @@ -1016,6 +902,8 @@ NORET_TYPE void do_exit(long code) tracehook_report_exit(&code); + validate_creds_for_do_exit(tsk); + /* * We're taking recursive faults here in do_exit. Safest is to just * leave this task alone and wait for reboot. @@ -1037,6 +925,8 @@ NORET_TYPE void do_exit(long code) schedule(); } + exit_irq_thread(); + exit_signals(tsk); /* sets PF_EXITING */ /* * tsk->flags are checked in the futex code to protect against @@ -1056,6 +946,8 @@ NORET_TYPE void do_exit(long code) if (group_dead) { hrtimer_cancel(&tsk->signal->real_timer); exit_itimers(tsk->signal); + if (tsk->mm) + setmax_mm_hiwater_rss(&tsk->signal->maxrss, tsk->mm); } acct_collect(code, group_dead); if (group_dead) @@ -1083,22 +975,25 @@ NORET_TYPE void do_exit(long code) disassociate_ctty(1); module_put(task_thread_info(tsk)->exec_domain->module); - if (tsk->binfmt) - module_put(tsk->binfmt->module); proc_exit_connector(tsk); + + /* + * FIXME: do that only when needed, using sched_exit tracepoint + */ + flush_ptrace_hw_breakpoint(tsk); + /* + * Flush inherited counters to the parent - before the parent + * gets woken up by child-exit notifications. + */ + perf_event_exit_task(tsk); + exit_notify(tsk, group_dead); #ifdef CONFIG_NUMA mpol_put(tsk->mempolicy); tsk->mempolicy = NULL; #endif #ifdef CONFIG_FUTEX - /* - * This must happen late, after the PID is not - * hashed anymore: - */ - if (unlikely(!list_empty(&tsk->pi_state_list))) - exit_pi_state_list(tsk); if (unlikely(current->pi_state_cache)) kfree(current->pi_state_cache); #endif @@ -1114,12 +1009,15 @@ NORET_TYPE void do_exit(long code) tsk->flags |= PF_EXITPIDONE; if (tsk->io_context) - exit_io_context(); + exit_io_context(tsk); if (tsk->splice_pipe) __free_pipe_info(tsk->splice_pipe); + validate_creds_for_do_exit(tsk); + preempt_disable(); + exit_rcu(); /* causes final put_task_struct in finish_task_switch(). */ tsk->state = TASK_DEAD; schedule(); @@ -1189,62 +1087,72 @@ SYSCALL_DEFINE1(exit_group, int, error_code) return 0; } -static struct pid *task_pid_type(struct task_struct *task, enum pid_type type) +struct wait_opts { + enum pid_type wo_type; + int wo_flags; + struct pid *wo_pid; + + struct siginfo __user *wo_info; + int __user *wo_stat; + struct rusage __user *wo_rusage; + + wait_queue_t child_wait; + int notask_error; +}; + +static inline +struct pid *task_pid_type(struct task_struct *task, enum pid_type type) { - struct pid *pid = NULL; - if (type == PIDTYPE_PID) - pid = task->pids[type].pid; - else if (type < PIDTYPE_MAX) - pid = task->group_leader->pids[type].pid; - return pid; + if (type != PIDTYPE_PID) + task = task->group_leader; + return task->pids[type].pid; } -static int eligible_child(enum pid_type type, struct pid *pid, int options, - struct task_struct *p) +static int eligible_pid(struct wait_opts *wo, struct task_struct *p) { - int err; - - if (type < PIDTYPE_MAX) { - if (task_pid_type(p, type) != pid) - return 0; - } + return wo->wo_type == PIDTYPE_MAX || + task_pid_type(p, wo->wo_type) == wo->wo_pid; +} +static int eligible_child(struct wait_opts *wo, struct task_struct *p) +{ + if (!eligible_pid(wo, p)) + return 0; /* Wait for all children (clone and not) if __WALL is set; * otherwise, wait for clone children *only* if __WCLONE is * set; otherwise, wait for non-clone children *only*. (Note: * A "clone" child here is one that reports to its parent * using a signal other than SIGCHLD.) */ - if (((p->exit_signal != SIGCHLD) ^ ((options & __WCLONE) != 0)) - && !(options & __WALL)) + if (((p->exit_signal != SIGCHLD) ^ !!(wo->wo_flags & __WCLONE)) + && !(wo->wo_flags & __WALL)) return 0; - err = security_task_wait(p); - if (err) - return err; - return 1; } -static int wait_noreap_copyout(struct task_struct *p, pid_t pid, uid_t uid, - int why, int status, - struct siginfo __user *infop, - struct rusage __user *rusagep) +static int wait_noreap_copyout(struct wait_opts *wo, struct task_struct *p, + pid_t pid, uid_t uid, int why, int status) { - int retval = rusagep ? getrusage(p, RUSAGE_BOTH, rusagep) : 0; + struct siginfo __user *infop; + int retval = wo->wo_rusage + ? getrusage(p, RUSAGE_BOTH, wo->wo_rusage) : 0; put_task_struct(p); - if (!retval) - retval = put_user(SIGCHLD, &infop->si_signo); - if (!retval) - retval = put_user(0, &infop->si_errno); - if (!retval) - retval = put_user((short)why, &infop->si_code); - if (!retval) - retval = put_user(pid, &infop->si_pid); - if (!retval) - retval = put_user(uid, &infop->si_uid); - if (!retval) - retval = put_user(status, &infop->si_status); + infop = wo->wo_info; + if (infop) { + if (!retval) + retval = put_user(SIGCHLD, &infop->si_signo); + if (!retval) + retval = put_user(0, &infop->si_errno); + if (!retval) + retval = put_user((short)why, &infop->si_code); + if (!retval) + retval = put_user(pid, &infop->si_pid); + if (!retval) + retval = put_user(uid, &infop->si_uid); + if (!retval) + retval = put_user(status, &infop->si_status); + } if (!retval) retval = pid; return retval; @@ -1256,19 +1164,18 @@ static int wait_noreap_copyout(struct task_struct *p, pid_t pid, uid_t uid, * the lock and this task is uninteresting. If we return nonzero, we have * released the lock and the system call should return. */ -static int wait_task_zombie(struct task_struct *p, int options, - struct siginfo __user *infop, - int __user *stat_addr, struct rusage __user *ru) +static int wait_task_zombie(struct wait_opts *wo, struct task_struct *p) { unsigned long state; int retval, status, traced; pid_t pid = task_pid_vnr(p); uid_t uid = __task_cred(p)->uid; + struct siginfo __user *infop; - if (!likely(options & WEXITED)) + if (!likely(wo->wo_flags & WEXITED)) return 0; - if (unlikely(options & WNOWAIT)) { + if (unlikely(wo->wo_flags & WNOWAIT)) { int exit_code = p->exit_code; int why, status; @@ -1281,8 +1188,7 @@ static int wait_task_zombie(struct task_struct *p, int options, why = (exit_code & 0x80) ? CLD_DUMPED : CLD_KILLED; status = exit_code & 0x7f; } - return wait_noreap_copyout(p, pid, uid, why, - status, infop, ru); + return wait_noreap_copyout(wo, p, pid, uid, why, status); } /* @@ -1296,11 +1202,15 @@ static int wait_task_zombie(struct task_struct *p, int options, } traced = ptrace_reparented(p); - - if (likely(!traced)) { + /* + * It can be ptraced but not reparented, check + * !task_detached() to filter out sub-threads. + */ + if (likely(!traced) && likely(!task_detached(p))) { struct signal_struct *psig; struct signal_struct *sig; - struct task_cputime cputime; + unsigned long maxrss; + cputime_t tgutime, tgstime; /* * The resource counters for the group leader are in its @@ -1313,25 +1223,25 @@ static int wait_task_zombie(struct task_struct *p, int options, * p->signal fields, because they are only touched by * __exit_signal, which runs with tasklist_lock * write-locked anyway, and so is excluded here. We do - * need to protect the access to p->parent->signal fields, + * need to protect the access to parent->signal fields, * as other threads in the parent group can be right * here reaping other children at the same time. * - * We use thread_group_cputime() to get times for the thread + * We use thread_group_times() to get times for the thread * group, which consolidates times for all threads in the * group including the group leader. */ - thread_group_cputime(p, &cputime); - spin_lock_irq(&p->parent->sighand->siglock); - psig = p->parent->signal; + thread_group_times(p, &tgutime, &tgstime); + spin_lock_irq(&p->real_parent->sighand->siglock); + psig = p->real_parent->signal; sig = p->signal; psig->cutime = cputime_add(psig->cutime, - cputime_add(cputime.utime, + cputime_add(tgutime, sig->cutime)); psig->cstime = cputime_add(psig->cstime, - cputime_add(cputime.stime, + cputime_add(tgstime, sig->cstime)); psig->cgtime = cputime_add(psig->cgtime, @@ -1352,9 +1262,12 @@ static int wait_task_zombie(struct task_struct *p, int options, psig->coublock += task_io_get_oublock(p) + sig->oublock + sig->coublock; + maxrss = max(sig->maxrss, sig->cmaxrss); + if (psig->cmaxrss < maxrss) + psig->cmaxrss = maxrss; task_io_accounting_add(&psig->ioac, &p->ioac); task_io_accounting_add(&psig->ioac, &sig->ioac); - spin_unlock_irq(&p->parent->sighand->siglock); + spin_unlock_irq(&p->real_parent->sighand->siglock); } /* @@ -1363,11 +1276,14 @@ static int wait_task_zombie(struct task_struct *p, int options, */ read_unlock(&tasklist_lock); - retval = ru ? getrusage(p, RUSAGE_BOTH, ru) : 0; + retval = wo->wo_rusage + ? getrusage(p, RUSAGE_BOTH, wo->wo_rusage) : 0; status = (p->signal->flags & SIGNAL_GROUP_EXIT) ? p->signal->group_exit_code : p->exit_code; - if (!retval && stat_addr) - retval = put_user(status, stat_addr); + if (!retval && wo->wo_stat) + retval = put_user(status, wo->wo_stat); + + infop = wo->wo_info; if (!retval && infop) retval = put_user(SIGCHLD, &infop->si_signo); if (!retval && infop) @@ -1417,42 +1333,51 @@ static int wait_task_zombie(struct task_struct *p, int options, return retval; } +static int *task_stopped_code(struct task_struct *p, bool ptrace) +{ + if (ptrace) { + if (task_is_stopped_or_traced(p)) + return &p->exit_code; + } else { + if (p->signal->flags & SIGNAL_STOP_STOPPED) + return &p->signal->group_exit_code; + } + return NULL; +} + /* * Handle sys_wait4 work for one task in state TASK_STOPPED. We hold * read_lock(&tasklist_lock) on entry. If we return zero, we still hold * the lock and this task is uninteresting. If we return nonzero, we have * released the lock and the system call should return. */ -static int wait_task_stopped(int ptrace, struct task_struct *p, - int options, struct siginfo __user *infop, - int __user *stat_addr, struct rusage __user *ru) +static int wait_task_stopped(struct wait_opts *wo, + int ptrace, struct task_struct *p) { - int retval, exit_code, why; + struct siginfo __user *infop; + int retval, exit_code, *p_code, why; uid_t uid = 0; /* unneeded, required by compiler */ pid_t pid; - if (!(options & WUNTRACED)) + /* + * Traditionally we see ptrace'd stopped tasks regardless of options. + */ + if (!ptrace && !(wo->wo_flags & WUNTRACED)) return 0; exit_code = 0; spin_lock_irq(&p->sighand->siglock); - if (unlikely(!task_is_stopped_or_traced(p))) - goto unlock_sig; - - if (!ptrace && p->signal->group_stop_count > 0) - /* - * A group stop is in progress and this is the group leader. - * We won't report until all threads have stopped. - */ + p_code = task_stopped_code(p, ptrace); + if (unlikely(!p_code)) goto unlock_sig; - exit_code = p->exit_code; + exit_code = *p_code; if (!exit_code) goto unlock_sig; - if (!unlikely(options & WNOWAIT)) - p->exit_code = 0; + if (!unlikely(wo->wo_flags & WNOWAIT)) + *p_code = 0; /* don't need the RCU readlock here as we're holding a spinlock */ uid = __task_cred(p)->uid; @@ -1473,14 +1398,15 @@ unlock_sig: why = ptrace ? CLD_TRAPPED : CLD_STOPPED; read_unlock(&tasklist_lock); - if (unlikely(options & WNOWAIT)) - return wait_noreap_copyout(p, pid, uid, - why, exit_code, - infop, ru); + if (unlikely(wo->wo_flags & WNOWAIT)) + return wait_noreap_copyout(wo, p, pid, uid, why, exit_code); + + retval = wo->wo_rusage + ? getrusage(p, RUSAGE_BOTH, wo->wo_rusage) : 0; + if (!retval && wo->wo_stat) + retval = put_user((exit_code << 8) | 0x7f, wo->wo_stat); - retval = ru ? getrusage(p, RUSAGE_BOTH, ru) : 0; - if (!retval && stat_addr) - retval = put_user((exit_code << 8) | 0x7f, stat_addr); + infop = wo->wo_info; if (!retval && infop) retval = put_user(SIGCHLD, &infop->si_signo); if (!retval && infop) @@ -1507,15 +1433,13 @@ unlock_sig: * the lock and this task is uninteresting. If we return nonzero, we have * released the lock and the system call should return. */ -static int wait_task_continued(struct task_struct *p, int options, - struct siginfo __user *infop, - int __user *stat_addr, struct rusage __user *ru) +static int wait_task_continued(struct wait_opts *wo, struct task_struct *p) { int retval; pid_t pid; uid_t uid; - if (!unlikely(options & WCONTINUED)) + if (!unlikely(wo->wo_flags & WCONTINUED)) return 0; if (!(p->signal->flags & SIGNAL_STOP_CONTINUED)) @@ -1527,7 +1451,7 @@ static int wait_task_continued(struct task_struct *p, int options, spin_unlock_irq(&p->sighand->siglock); return 0; } - if (!unlikely(options & WNOWAIT)) + if (!unlikely(wo->wo_flags & WNOWAIT)) p->signal->flags &= ~SIGNAL_STOP_CONTINUED; uid = __task_cred(p)->uid; spin_unlock_irq(&p->sighand->siglock); @@ -1536,17 +1460,17 @@ static int wait_task_continued(struct task_struct *p, int options, get_task_struct(p); read_unlock(&tasklist_lock); - if (!infop) { - retval = ru ? getrusage(p, RUSAGE_BOTH, ru) : 0; + if (!wo->wo_info) { + retval = wo->wo_rusage + ? getrusage(p, RUSAGE_BOTH, wo->wo_rusage) : 0; put_task_struct(p); - if (!retval && stat_addr) - retval = put_user(0xffff, stat_addr); + if (!retval && wo->wo_stat) + retval = put_user(0xffff, wo->wo_stat); if (!retval) retval = pid; } else { - retval = wait_noreap_copyout(p, pid, uid, - CLD_CONTINUED, SIGCONT, - infop, ru); + retval = wait_noreap_copyout(wo, p, pid, uid, + CLD_CONTINUED, SIGCONT); BUG_ON(retval == 0); } @@ -1556,22 +1480,20 @@ static int wait_task_continued(struct task_struct *p, int options, /* * Consider @p for a wait by @parent. * - * -ECHILD should be in *@notask_error before the first call. + * -ECHILD should be in ->notask_error before the first call. * Returns nonzero for a final return, when we have unlocked tasklist_lock. * Returns zero if the search for a child should continue; - * then *@notask_error is 0 if @p is an eligible child, + * then ->notask_error is 0 if @p is an eligible child, * or another error from security_task_wait(), or still -ECHILD. */ -static int wait_consider_task(struct task_struct *parent, int ptrace, - struct task_struct *p, int *notask_error, - enum pid_type type, struct pid *pid, int options, - struct siginfo __user *infop, - int __user *stat_addr, struct rusage __user *ru) +static int wait_consider_task(struct wait_opts *wo, int ptrace, + struct task_struct *p) { - int ret = eligible_child(type, pid, options, p); + int ret = eligible_child(wo, p); if (!ret) return ret; + ret = security_task_wait(p); if (unlikely(ret < 0)) { /* * If we have not yet seen any eligible child, @@ -1580,16 +1502,17 @@ static int wait_consider_task(struct task_struct *parent, int ptrace, * to look for security policy problems, rather * than for mysterious wait bugs. */ - if (*notask_error) - *notask_error = ret; + if (wo->notask_error) + wo->notask_error = ret; + return 0; } - if (likely(!ptrace) && unlikely(p->ptrace)) { + if (likely(!ptrace) && unlikely(task_ptrace(p))) { /* * This child is hidden by ptrace. * We aren't allowed to see it now, but eventually we will. */ - *notask_error = 0; + wo->notask_error = 0; return 0; } @@ -1600,34 +1523,30 @@ static int wait_consider_task(struct task_struct *parent, int ptrace, * We don't reap group leaders with subthreads. */ if (p->exit_state == EXIT_ZOMBIE && !delay_group_leader(p)) - return wait_task_zombie(p, options, infop, stat_addr, ru); + return wait_task_zombie(wo, p); /* * It's stopped or running now, so it might * later continue, exit, or stop again. */ - *notask_error = 0; + wo->notask_error = 0; - if (task_is_stopped_or_traced(p)) - return wait_task_stopped(ptrace, p, options, - infop, stat_addr, ru); + if (task_stopped_code(p, ptrace)) + return wait_task_stopped(wo, ptrace, p); - return wait_task_continued(p, options, infop, stat_addr, ru); + return wait_task_continued(wo, p); } /* * Do the work of do_wait() for one thread in the group, @tsk. * - * -ECHILD should be in *@notask_error before the first call. + * -ECHILD should be in ->notask_error before the first call. * Returns nonzero for a final return, when we have unlocked tasklist_lock. * Returns zero if the search for a child should continue; then - * *@notask_error is 0 if there were any eligible children, + * ->notask_error is 0 if there were any eligible children, * or another error from security_task_wait(), or still -ECHILD. */ -static int do_wait_thread(struct task_struct *tsk, int *notask_error, - enum pid_type type, struct pid *pid, int options, - struct siginfo __user *infop, int __user *stat_addr, - struct rusage __user *ru) +static int do_wait_thread(struct wait_opts *wo, struct task_struct *tsk) { struct task_struct *p; @@ -1636,9 +1555,7 @@ static int do_wait_thread(struct task_struct *tsk, int *notask_error, * Do not consider detached threads. */ if (!task_detached(p)) { - int ret = wait_consider_task(tsk, 0, p, notask_error, - type, pid, options, - infop, stat_addr, ru); + int ret = wait_consider_task(wo, 0, p); if (ret) return ret; } @@ -1647,22 +1564,12 @@ static int do_wait_thread(struct task_struct *tsk, int *notask_error, return 0; } -static int ptrace_do_wait(struct task_struct *tsk, int *notask_error, - enum pid_type type, struct pid *pid, int options, - struct siginfo __user *infop, int __user *stat_addr, - struct rusage __user *ru) +static int ptrace_do_wait(struct wait_opts *wo, struct task_struct *tsk) { struct task_struct *p; - /* - * Traditionally we see ptrace'd stopped tasks regardless of options. - */ - options |= WUNTRACED; - list_for_each_entry(p, &tsk->ptraced, ptrace_entry) { - int ret = wait_consider_task(tsk, 1, p, notask_error, - type, pid, options, - infop, stat_addr, ru); + int ret = wait_consider_task(wo, 1, p); if (ret) return ret; } @@ -1670,93 +1577,86 @@ static int ptrace_do_wait(struct task_struct *tsk, int *notask_error, return 0; } -static long do_wait(enum pid_type type, struct pid *pid, int options, - struct siginfo __user *infop, int __user *stat_addr, - struct rusage __user *ru) +static int child_wait_callback(wait_queue_t *wait, unsigned mode, + int sync, void *key) +{ + struct wait_opts *wo = container_of(wait, struct wait_opts, + child_wait); + struct task_struct *p = key; + + if (!eligible_pid(wo, p)) + return 0; + + if ((wo->wo_flags & __WNOTHREAD) && wait->private != p->parent) + return 0; + + return default_wake_function(wait, mode, sync, key); +} + +void __wake_up_parent(struct task_struct *p, struct task_struct *parent) +{ + __wake_up_sync_key(&parent->signal->wait_chldexit, + TASK_INTERRUPTIBLE, 1, p); +} + +static long do_wait(struct wait_opts *wo) { - DECLARE_WAITQUEUE(wait, current); struct task_struct *tsk; int retval; - trace_sched_process_wait(pid); + trace_sched_process_wait(wo->wo_pid); - add_wait_queue(¤t->signal->wait_chldexit,&wait); + init_waitqueue_func_entry(&wo->child_wait, child_wait_callback); + wo->child_wait.private = current; + add_wait_queue(¤t->signal->wait_chldexit, &wo->child_wait); repeat: /* * If there is nothing that can match our critiera just get out. - * We will clear @retval to zero if we see any child that might later - * match our criteria, even if we are not able to reap it yet. + * We will clear ->notask_error to zero if we see any child that + * might later match our criteria, even if we are not able to reap + * it yet. */ - retval = -ECHILD; - if ((type < PIDTYPE_MAX) && (!pid || hlist_empty(&pid->tasks[type]))) - goto end; + wo->notask_error = -ECHILD; + if ((wo->wo_type < PIDTYPE_MAX) && + (!wo->wo_pid || hlist_empty(&wo->wo_pid->tasks[wo->wo_type]))) + goto notask; - current->state = TASK_INTERRUPTIBLE; + set_current_state(TASK_INTERRUPTIBLE); read_lock(&tasklist_lock); tsk = current; do { - int tsk_result = do_wait_thread(tsk, &retval, - type, pid, options, - infop, stat_addr, ru); - if (!tsk_result) - tsk_result = ptrace_do_wait(tsk, &retval, - type, pid, options, - infop, stat_addr, ru); - if (tsk_result) { - /* - * tasklist_lock is unlocked and we have a final result. - */ - retval = tsk_result; + retval = do_wait_thread(wo, tsk); + if (retval) goto end; - } - if (options & __WNOTHREAD) + retval = ptrace_do_wait(wo, tsk); + if (retval) + goto end; + + if (wo->wo_flags & __WNOTHREAD) break; - tsk = next_thread(tsk); - BUG_ON(tsk->signal != current->signal); - } while (tsk != current); + } while_each_thread(current, tsk); read_unlock(&tasklist_lock); - if (!retval && !(options & WNOHANG)) { +notask: + retval = wo->notask_error; + if (!retval && !(wo->wo_flags & WNOHANG)) { retval = -ERESTARTSYS; if (!signal_pending(current)) { schedule(); goto repeat; } } - end: - current->state = TASK_RUNNING; - remove_wait_queue(¤t->signal->wait_chldexit,&wait); - if (infop) { - if (retval > 0) - retval = 0; - else { - /* - * For a WNOHANG return, clear out all the fields - * we would set so the user can easily tell the - * difference. - */ - if (!retval) - retval = put_user(0, &infop->si_signo); - if (!retval) - retval = put_user(0, &infop->si_errno); - if (!retval) - retval = put_user(0, &infop->si_code); - if (!retval) - retval = put_user(0, &infop->si_pid); - if (!retval) - retval = put_user(0, &infop->si_uid); - if (!retval) - retval = put_user(0, &infop->si_status); - } - } + __set_current_state(TASK_RUNNING); + remove_wait_queue(¤t->signal->wait_chldexit, &wo->child_wait); return retval; } SYSCALL_DEFINE5(waitid, int, which, pid_t, upid, struct siginfo __user *, infop, int, options, struct rusage __user *, ru) { + struct wait_opts wo; struct pid *pid = NULL; enum pid_type type; long ret; @@ -1786,7 +1686,37 @@ SYSCALL_DEFINE5(waitid, int, which, pid_t, upid, struct siginfo __user *, if (type < PIDTYPE_MAX) pid = find_get_pid(upid); - ret = do_wait(type, pid, options, infop, NULL, ru); + + wo.wo_type = type; + wo.wo_pid = pid; + wo.wo_flags = options; + wo.wo_info = infop; + wo.wo_stat = NULL; + wo.wo_rusage = ru; + ret = do_wait(&wo); + + if (ret > 0) { + ret = 0; + } else if (infop) { + /* + * For a WNOHANG return, clear out all the fields + * we would set so the user can easily tell the + * difference. + */ + if (!ret) + ret = put_user(0, &infop->si_signo); + if (!ret) + ret = put_user(0, &infop->si_errno); + if (!ret) + ret = put_user(0, &infop->si_code); + if (!ret) + ret = put_user(0, &infop->si_pid); + if (!ret) + ret = put_user(0, &infop->si_uid); + if (!ret) + ret = put_user(0, &infop->si_status); + } + put_pid(pid); /* avoid REGPARM breakage on x86: */ @@ -1797,6 +1727,7 @@ SYSCALL_DEFINE5(waitid, int, which, pid_t, upid, struct siginfo __user *, SYSCALL_DEFINE4(wait4, pid_t, upid, int __user *, stat_addr, int, options, struct rusage __user *, ru) { + struct wait_opts wo; struct pid *pid = NULL; enum pid_type type; long ret; @@ -1812,13 +1743,19 @@ SYSCALL_DEFINE4(wait4, pid_t, upid, int __user *, stat_addr, pid = find_get_pid(-upid); } else if (upid == 0) { type = PIDTYPE_PGID; - pid = get_pid(task_pgrp(current)); + pid = get_task_pid(current, PIDTYPE_PGID); } else /* upid > 0 */ { type = PIDTYPE_PID; pid = find_get_pid(upid); } - ret = do_wait(type, pid, options | WEXITED, NULL, stat_addr, ru); + wo.wo_type = type; + wo.wo_pid = pid; + wo.wo_flags = options | WEXITED; + wo.wo_info = NULL; + wo.wo_stat = stat_addr; + wo.wo_rusage = ru; + ret = do_wait(&wo); put_pid(pid); /* avoid REGPARM breakage on x86: */ diff --git a/kernel/extable.c b/kernel/extable.c index e136ed8d82b..7f8f263f852 100644 --- a/kernel/extable.c +++ b/kernel/extable.c @@ -15,11 +15,22 @@ along with this program; if not, write to the Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA */ +#include <linux/ftrace.h> +#include <linux/memory.h> #include <linux/module.h> +#include <linux/mutex.h> #include <linux/init.h> -#include <linux/ftrace.h> -#include <asm/uaccess.h> + #include <asm/sections.h> +#include <asm/uaccess.h> + +/* + * mutex protecting text section modification (dynamic code patching). + * some users need to sleep (allocating memory...) while they hold this lock. + * + * NOT exported to modules - patching kernel text is a really delicate matter. + */ +DEFINE_MUTEX(text_mutex); extern struct exception_table_entry __start___ex_table[]; extern struct exception_table_entry __stop___ex_table[]; @@ -41,31 +52,50 @@ const struct exception_table_entry *search_exception_tables(unsigned long addr) return e; } -__notrace_funcgraph int core_kernel_text(unsigned long addr) +static inline int init_kernel_text(unsigned long addr) +{ + if (addr >= (unsigned long)_sinittext && + addr <= (unsigned long)_einittext) + return 1; + return 0; +} + +int core_kernel_text(unsigned long addr) { if (addr >= (unsigned long)_stext && addr <= (unsigned long)_etext) return 1; if (system_state == SYSTEM_BOOTING && - addr >= (unsigned long)_sinittext && - addr <= (unsigned long)_einittext) + init_kernel_text(addr)) return 1; return 0; } -__notrace_funcgraph int __kernel_text_address(unsigned long addr) +int __kernel_text_address(unsigned long addr) { if (core_kernel_text(addr)) return 1; - return __module_text_address(addr) != NULL; + if (is_module_text_address(addr)) + return 1; + /* + * There might be init symbols in saved stacktraces. + * Give those symbols a chance to be printed in + * backtraces (such as lockdep traces). + * + * Since we are after the module-symbols check, there's + * no danger of address overlap: + */ + if (init_kernel_text(addr)) + return 1; + return 0; } int kernel_text_address(unsigned long addr) { if (core_kernel_text(addr)) return 1; - return module_text_address(addr) != NULL; + return is_module_text_address(addr); } /* @@ -81,5 +111,5 @@ int func_ptr_is_kernel_text(void *ptr) addr = (unsigned long) dereference_function_descriptor(ptr); if (core_kernel_text(addr)) return 1; - return module_text_address(addr) != NULL; + return is_module_text_address(addr); } diff --git a/kernel/fork.c b/kernel/fork.c index 47c15840a38..1415dc4598a 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -17,7 +17,6 @@ #include <linux/module.h> #include <linux/vmalloc.h> #include <linux/completion.h> -#include <linux/mnt_namespace.h> #include <linux/personality.h> #include <linux/mempolicy.h> #include <linux/sem.h> @@ -50,6 +49,7 @@ #include <linux/ftrace.h> #include <linux/profile.h> #include <linux/rmap.h> +#include <linux/ksm.h> #include <linux/acct.h> #include <linux/tsacct_kern.h> #include <linux/cn_proc.h> @@ -60,8 +60,11 @@ #include <linux/tty.h> #include <linux/proc_fs.h> #include <linux/blkdev.h> -#include <trace/sched.h> +#include <linux/fs_struct.h> #include <linux/magic.h> +#include <linux/perf_event.h> +#include <linux/posix-timers.h> +#include <linux/user-return-notifier.h> #include <asm/pgtable.h> #include <asm/pgalloc.h> @@ -70,6 +73,8 @@ #include <asm/cacheflush.h> #include <asm/tlbflush.h> +#include <trace/events/sched.h> + /* * Protected counters by write_lock_irq(&tasklist_lock) */ @@ -82,14 +87,12 @@ DEFINE_PER_CPU(unsigned long, process_counts) = 0; __cacheline_aligned DEFINE_RWLOCK(tasklist_lock); /* outer */ -DEFINE_TRACE(sched_process_fork); - int nr_processes(void) { int cpu; int total = 0; - for_each_online_cpu(cpu) + for_each_possible_cpu(cpu) total += per_cpu(process_counts, cpu); return total; @@ -136,9 +139,17 @@ struct kmem_cache *vm_area_cachep; /* SLAB cache for mm_struct structures (tsk->mm) */ static struct kmem_cache *mm_cachep; +static void account_kernel_stack(struct thread_info *ti, int account) +{ + struct zone *zone = page_zone(virt_to_page(ti)); + + mod_zone_page_state(zone, NR_KERNEL_STACK, account); +} + void free_task(struct task_struct *tsk) { prop_local_destroy_single(&tsk->dirties); + account_kernel_stack(tsk->stack, -1); free_thread_info(tsk->stack); rt_mutex_debug_task_free(tsk); ftrace_graph_exit_task(tsk); @@ -152,8 +163,7 @@ void __put_task_struct(struct task_struct *tsk) WARN_ON(atomic_read(&tsk->usage)); WARN_ON(tsk == current); - put_cred(tsk->real_cred); - put_cred(tsk->cred); + exit_creds(tsk); delayacct_tsk_free(tsk); if (!profile_handoff_task(tsk)) @@ -177,7 +187,7 @@ void __init fork_init(unsigned long mempages) /* create a slab on which task_structs can be allocated */ task_struct_cachep = kmem_cache_create("task_struct", sizeof(struct task_struct), - ARCH_MIN_TASKALIGN, SLAB_PANIC, NULL); + ARCH_MIN_TASKALIGN, SLAB_PANIC | SLAB_NOTRACK, NULL); #endif /* do the arch specific task caches init */ @@ -240,6 +250,7 @@ static struct task_struct *dup_task_struct(struct task_struct *orig) goto out; setup_thread_stack(tsk, orig); + clear_user_return_notifier(tsk); stackend = end_of_stack(tsk); *stackend = STACK_END_MAGIC; /* for overflow detection */ @@ -254,6 +265,9 @@ static struct task_struct *dup_task_struct(struct task_struct *orig) tsk->btrace_seq = 0; #endif tsk->splice_pipe = NULL; + + account_kernel_stack(ti, 1); + return tsk; out: @@ -289,6 +303,9 @@ static int dup_mmap(struct mm_struct *mm, struct mm_struct *oldmm) rb_link = &mm->mm_rb.rb_node; rb_parent = NULL; pprev = &mm->mmap; + retval = ksm_fork(mm, oldmm); + if (retval) + goto out; for (mpnt = oldmm->mmap; mpnt; mpnt = mpnt->vm_next) { struct file *file; @@ -419,22 +436,30 @@ __setup("coredump_filter=", coredump_filter_setup); #include <linux/init_task.h> +static void mm_init_aio(struct mm_struct *mm) +{ +#ifdef CONFIG_AIO + spin_lock_init(&mm->ioctx_lock); + INIT_HLIST_HEAD(&mm->ioctx_list); +#endif +} + static struct mm_struct * mm_init(struct mm_struct * mm, struct task_struct *p) { atomic_set(&mm->mm_users, 1); atomic_set(&mm->mm_count, 1); init_rwsem(&mm->mmap_sem); INIT_LIST_HEAD(&mm->mmlist); - mm->flags = (current->mm) ? current->mm->flags : default_dump_filter; + mm->flags = (current->mm) ? + (current->mm->flags & MMF_INIT_MASK) : default_dump_filter; mm->core_state = NULL; mm->nr_ptes = 0; set_mm_counter(mm, file_rss, 0); set_mm_counter(mm, anon_rss, 0); spin_lock_init(&mm->page_table_lock); - spin_lock_init(&mm->ioctx_lock); - INIT_HLIST_HEAD(&mm->ioctx_list); mm->free_area_cache = TASK_UNMAPPED_BASE; mm->cached_hole_size = ~0UL; + mm_init_aio(mm); mm_init_owner(mm, p); if (likely(!mm_alloc_pgd(mm))) { @@ -486,6 +511,7 @@ void mmput(struct mm_struct *mm) if (atomic_dec_and_test(&mm->mm_users)) { exit_aio(mm); + ksm_exit(mm); exit_mmap(mm); set_mm_exe_file(mm, NULL); if (!list_empty(&mm->mmlist)) { @@ -494,6 +520,8 @@ void mmput(struct mm_struct *mm) spin_unlock(&mmlist_lock); } put_swap_token(mm); + if (mm->binfmt) + module_put(mm->binfmt->module); mmdrop(mm); } } @@ -544,12 +572,18 @@ void mm_release(struct task_struct *tsk, struct mm_struct *mm) /* Get rid of any futexes when releasing the mm */ #ifdef CONFIG_FUTEX - if (unlikely(tsk->robust_list)) + if (unlikely(tsk->robust_list)) { exit_robust_list(tsk); + tsk->robust_list = NULL; + } #ifdef CONFIG_COMPAT - if (unlikely(tsk->compat_robust_list)) + if (unlikely(tsk->compat_robust_list)) { compat_exit_robust_list(tsk); + tsk->compat_robust_list = NULL; + } #endif + if (unlikely(!list_empty(&tsk->pi_state_list))) + exit_pi_state_list(tsk); #endif /* Get rid of any cached register state */ @@ -567,18 +601,18 @@ void mm_release(struct task_struct *tsk, struct mm_struct *mm) * the value intact in a core dump, and to save the unnecessary * trouble otherwise. Userland only wants this done for a sys_exit. */ - if (tsk->clear_child_tid - && !(tsk->flags & PF_SIGNALED) - && atomic_read(&mm->mm_users) > 1) { - u32 __user * tidptr = tsk->clear_child_tid; + if (tsk->clear_child_tid) { + if (!(tsk->flags & PF_SIGNALED) && + atomic_read(&mm->mm_users) > 1) { + /* + * We don't check the error code - if userspace has + * not set up a proper pointer then tough luck. + */ + put_user(0, tsk->clear_child_tid); + sys_futex(tsk->clear_child_tid, FUTEX_WAKE, + 1, NULL, NULL, 0); + } tsk->clear_child_tid = NULL; - - /* - * We don't check the error code - if userspace has - * not set up a proper pointer then tough luck. - */ - put_user(0, tidptr); - sys_futex(tidptr, FUTEX_WAKE, 1, NULL, NULL, 0); } } @@ -619,9 +653,14 @@ struct mm_struct *dup_mm(struct task_struct *tsk) mm->hiwater_rss = get_mm_rss(mm); mm->hiwater_vm = mm->total_vm; + if (mm->binfmt && !try_module_get(mm->binfmt->module)) + goto free_pt; + return mm; free_pt: + /* don't put binfmt in mmput, we haven't got module yet */ + mm->binfmt = NULL; mmput(mm); fail_nomem: @@ -644,6 +683,9 @@ static int copy_mm(unsigned long clone_flags, struct task_struct * tsk) tsk->min_flt = tsk->maj_flt = 0; tsk->nvcsw = tsk->nivcsw = 0; +#ifdef CONFIG_DETECT_HUNG_TASK + tsk->last_switch_count = tsk->nvcsw + tsk->nivcsw; +#endif tsk->mm = NULL; tsk->active_mm = NULL; @@ -681,38 +723,21 @@ fail_nomem: return retval; } -static struct fs_struct *__copy_fs_struct(struct fs_struct *old) -{ - struct fs_struct *fs = kmem_cache_alloc(fs_cachep, GFP_KERNEL); - /* We don't need to lock fs - think why ;-) */ - if (fs) { - atomic_set(&fs->count, 1); - rwlock_init(&fs->lock); - fs->umask = old->umask; - read_lock(&old->lock); - fs->root = old->root; - path_get(&old->root); - fs->pwd = old->pwd; - path_get(&old->pwd); - read_unlock(&old->lock); - } - return fs; -} - -struct fs_struct *copy_fs_struct(struct fs_struct *old) -{ - return __copy_fs_struct(old); -} - -EXPORT_SYMBOL_GPL(copy_fs_struct); - static int copy_fs(unsigned long clone_flags, struct task_struct *tsk) { + struct fs_struct *fs = current->fs; if (clone_flags & CLONE_FS) { - atomic_inc(¤t->fs->count); + /* tsk->fs is already what we want */ + write_lock(&fs->lock); + if (fs->in_exec) { + write_unlock(&fs->lock); + return -EAGAIN; + } + fs->users++; + write_unlock(&fs->lock); return 0; } - tsk->fs = __copy_fs_struct(current->fs); + tsk->fs = copy_fs_struct(fs); if (!tsk->fs) return -ENOMEM; return 0; @@ -803,16 +828,22 @@ static void posix_cpu_timers_init_group(struct signal_struct *sig) thread_group_cputime_init(sig); /* Expiration times and increments. */ - sig->it_virt_expires = cputime_zero; - sig->it_virt_incr = cputime_zero; - sig->it_prof_expires = cputime_zero; - sig->it_prof_incr = cputime_zero; + sig->it[CPUCLOCK_PROF].expires = cputime_zero; + sig->it[CPUCLOCK_PROF].incr = cputime_zero; + sig->it[CPUCLOCK_VIRT].expires = cputime_zero; + sig->it[CPUCLOCK_VIRT].incr = cputime_zero; /* Cached expiration times. */ sig->cputime_expires.prof_exp = cputime_zero; sig->cputime_expires.virt_exp = cputime_zero; sig->cputime_expires.sched_exp = 0; + if (sig->rlim[RLIMIT_CPU].rlim_cur != RLIM_INFINITY) { + sig->cputime_expires.prof_exp = + secs_to_cputime(sig->rlim[RLIMIT_CPU].rlim_cur); + sig->cputimer.running = 1; + } + /* The timer lists. */ INIT_LIST_HEAD(&sig->cpu_timers[0]); INIT_LIST_HEAD(&sig->cpu_timers[1]); @@ -823,16 +854,10 @@ static int copy_signal(unsigned long clone_flags, struct task_struct *tsk) { struct signal_struct *sig; - if (clone_flags & CLONE_THREAD) { - atomic_inc(¤t->signal->count); - atomic_inc(¤t->signal->live); + if (clone_flags & CLONE_THREAD) return 0; - } - sig = kmem_cache_alloc(signal_cachep, GFP_KERNEL); - - if (sig) - posix_cpu_timers_init_group(sig); + sig = kmem_cache_alloc(signal_cachep, GFP_KERNEL); tsk->signal = sig; if (!sig) return -ENOMEM; @@ -841,6 +866,8 @@ static int copy_signal(unsigned long clone_flags, struct task_struct *tsk) atomic_set(&sig->live, 1); init_waitqueue_head(&sig->wait_chldexit); sig->flags = 0; + if (clone_flags & CLONE_NEWPID) + sig->flags |= SIGNAL_UNKILLABLE; sig->group_exit_code = 0; sig->group_exit_task = NULL; sig->group_stop_count = 0; @@ -859,9 +886,13 @@ static int copy_signal(unsigned long clone_flags, struct task_struct *tsk) sig->utime = sig->stime = sig->cutime = sig->cstime = cputime_zero; sig->gtime = cputime_zero; sig->cgtime = cputime_zero; +#ifndef CONFIG_VIRT_CPU_ACCOUNTING + sig->prev_utime = sig->prev_stime = cputime_zero; +#endif sig->nvcsw = sig->nivcsw = sig->cnvcsw = sig->cnivcsw = 0; sig->min_flt = sig->maj_flt = sig->cmin_flt = sig->cmaj_flt = 0; sig->inblock = sig->oublock = sig->cinblock = sig->coublock = 0; + sig->maxrss = sig->cmaxrss = 0; task_io_accounting_init(&sig->ioac); sig->sum_sched_runtime = 0; taskstats_tgid_init(sig); @@ -870,10 +901,14 @@ static int copy_signal(unsigned long clone_flags, struct task_struct *tsk) memcpy(sig->rlim, current->signal->rlim, sizeof sig->rlim); task_unlock(current->group_leader); + posix_cpu_timers_init_group(sig); + acct_init_pacct(&sig->pacct); tty_audit_fork(sig); + sig->oom_adj = current->signal->oom_adj; + return 0; } @@ -884,16 +919,6 @@ void __cleanup_signal(struct signal_struct *sig) kmem_cache_free(signal_cachep, sig); } -static void cleanup_signal(struct task_struct *tsk) -{ - struct signal_struct *sig = tsk->signal; - - atomic_dec(&sig->live); - - if (atomic_dec_and_test(&sig->count)) - __cleanup_signal(sig); -} - static void copy_flags(unsigned long clone_flags, struct task_struct *p) { unsigned long new_flags = p->flags; @@ -979,6 +1004,16 @@ static struct task_struct *copy_process(unsigned long clone_flags, if ((clone_flags & CLONE_SIGHAND) && !(clone_flags & CLONE_VM)) return ERR_PTR(-EINVAL); + /* + * Siblings of global init remain as zombies on exit since they are + * not reaped by their parent (swapper). To solve this and to avoid + * multi-rooted process trees, prevent global and container-inits + * from creating siblings. + */ + if ((clone_flags & CLONE_PARENT) && + current->signal->flags & SIGNAL_UNKILLABLE) + return ERR_PTR(-EINVAL); + retval = security_task_create(clone_flags); if (retval) goto fork_out; @@ -988,6 +1023,8 @@ static struct task_struct *copy_process(unsigned long clone_flags, if (!p) goto fork_out; + ftrace_graph_init_task(p); + rt_mutex_init_task(p); #ifdef CONFIG_PROVE_LOCKING @@ -1018,22 +1055,15 @@ static struct task_struct *copy_process(unsigned long clone_flags, if (!try_module_get(task_thread_info(p)->exec_domain->module)) goto bad_fork_cleanup_count; - if (p->binfmt && !try_module_get(p->binfmt->module)) - goto bad_fork_cleanup_put_domain; - p->did_exec = 0; delayacct_tsk_init(p); /* Must remain after dup_task_struct() */ copy_flags(clone_flags, p); INIT_LIST_HEAD(&p->children); INIT_LIST_HEAD(&p->sibling); -#ifdef CONFIG_PREEMPT_RCU - p->rcu_read_lock_nesting = 0; - p->rcu_flipctr_idx = 0; -#endif /* #ifdef CONFIG_PREEMPT_RCU */ + rcu_copy_process(p); p->vfork_done = NULL; spin_lock_init(&p->alloc_lock); - clear_tsk_thread_flag(p, TIF_SIGPENDING); init_sigpending(&p->pending); p->utime = cputime_zero; @@ -1041,16 +1071,13 @@ static struct task_struct *copy_process(unsigned long clone_flags, p->gtime = cputime_zero; p->utimescaled = cputime_zero; p->stimescaled = cputime_zero; +#ifndef CONFIG_VIRT_CPU_ACCOUNTING p->prev_utime = cputime_zero; p->prev_stime = cputime_zero; +#endif p->default_timer_slack_ns = current->timer_slack_ns; -#ifdef CONFIG_DETECT_SOFTLOCKUP - p->last_switch_count = 0; - p->last_switch_timestamp = 0; -#endif - task_io_accounting_init(&p->ioac); acct_clear_integrals(p); @@ -1100,12 +1127,18 @@ static struct task_struct *copy_process(unsigned long clone_flags, #ifdef CONFIG_DEBUG_MUTEXES p->blocked_on = NULL; /* not blocked yet */ #endif - if (unlikely(current->ptrace)) - ptrace_fork(p, clone_flags); + + p->bts = NULL; + + p->stack_start = stack_start; /* Perform scheduler related setup. Assign this task to a CPU. */ sched_fork(p, clone_flags); + retval = perf_event_init_task(p); + if (retval) + goto bad_fork_cleanup_policy; + if ((retval = audit_alloc(p))) goto bad_fork_cleanup_policy; /* copy all the process information */ @@ -1125,7 +1158,7 @@ static struct task_struct *copy_process(unsigned long clone_flags, goto bad_fork_cleanup_mm; if ((retval = copy_io(clone_flags, p))) goto bad_fork_cleanup_namespaces; - retval = copy_thread(0, clone_flags, stack_start, stack_size, p, regs); + retval = copy_thread(clone_flags, stack_start, stack_size, p, regs); if (retval) goto bad_fork_cleanup_io; @@ -1142,8 +1175,6 @@ static struct task_struct *copy_process(unsigned long clone_flags, } } - ftrace_graph_init_task(p); - p->pid = pid_nr(pid); p->tgid = p->pid; if (clone_flags & CLONE_THREAD) @@ -1152,7 +1183,7 @@ static struct task_struct *copy_process(unsigned long clone_flags, if (current->nsproxy != p->nsproxy) { retval = ns_cgroup_clone(p, pid); if (retval) - goto bad_fork_free_graph; + goto bad_fork_free_pid; } p->set_child_tid = (clone_flags & CLONE_CHILD_SETTID) ? child_tidptr : NULL; @@ -1244,10 +1275,12 @@ static struct task_struct *copy_process(unsigned long clone_flags, spin_unlock(¤t->sighand->siglock); write_unlock_irq(&tasklist_lock); retval = -ERESTARTNOINTR; - goto bad_fork_free_graph; + goto bad_fork_free_pid; } if (clone_flags & CLONE_THREAD) { + atomic_inc(¤t->signal->count); + atomic_inc(¤t->signal->live); p->group_leader = current->group_leader; list_add_tail_rcu(&p->thread_group, &p->group_leader->thread_group); } @@ -1263,8 +1296,6 @@ static struct task_struct *copy_process(unsigned long clone_flags, p->signal->leader_pid = pid; tty_kref_put(p->signal->tty); p->signal->tty = tty_kref_get(current->signal->tty); - set_task_pgrp(p, task_pgrp_nr(current)); - set_task_session(p, task_session_nr(current)); attach_pid(p, PIDTYPE_PGID, task_pgrp(current)); attach_pid(p, PIDTYPE_SID, task_session(current)); list_add_tail_rcu(&p->tasks, &init_task.tasks); @@ -1279,22 +1310,23 @@ static struct task_struct *copy_process(unsigned long clone_flags, write_unlock_irq(&tasklist_lock); proc_fork_connector(p); cgroup_post_fork(p); + perf_event_fork(p); return p; -bad_fork_free_graph: - ftrace_graph_exit_task(p); bad_fork_free_pid: if (pid != &init_struct_pid) free_pid(pid); bad_fork_cleanup_io: - put_io_context(p->io_context); + if (p->io_context) + exit_io_context(p); bad_fork_cleanup_namespaces: exit_task_namespaces(p); bad_fork_cleanup_mm: if (p->mm) mmput(p->mm); bad_fork_cleanup_signal: - cleanup_signal(p); + if (!(clone_flags & CLONE_THREAD)) + __cleanup_signal(p->signal); bad_fork_cleanup_sighand: __cleanup_sighand(p->sighand); bad_fork_cleanup_fs: @@ -1306,20 +1338,17 @@ bad_fork_cleanup_semundo: bad_fork_cleanup_audit: audit_free(p); bad_fork_cleanup_policy: + perf_event_free_task(p); #ifdef CONFIG_NUMA mpol_put(p->mempolicy); bad_fork_cleanup_cgroup: #endif cgroup_exit(p, cgroup_callbacks_done); delayacct_tsk_free(p); - if (p->binfmt) - module_put(p->binfmt->module); -bad_fork_cleanup_put_domain: module_put(task_thread_info(p)->exec_domain->module); bad_fork_cleanup_count: atomic_dec(&p->cred->user->processes); - put_cred(p->real_cred); - put_cred(p->cred); + exit_creds(p); bad_fork_free: free_task(p); fork_out: @@ -1422,7 +1451,7 @@ long do_fork(unsigned long clone_flags, } audit_finish_fork(p); - tracehook_report_clone(trace, regs, clone_flags, nr, p); + tracehook_report_clone(regs, clone_flags, nr, p); /* * We set PF_STARTING at creation in case tracing wants to @@ -1474,20 +1503,21 @@ void __init proc_caches_init(void) { sighand_cachep = kmem_cache_create("sighand_cache", sizeof(struct sighand_struct), 0, - SLAB_HWCACHE_ALIGN|SLAB_PANIC|SLAB_DESTROY_BY_RCU, - sighand_ctor); + SLAB_HWCACHE_ALIGN|SLAB_PANIC|SLAB_DESTROY_BY_RCU| + SLAB_NOTRACK, sighand_ctor); signal_cachep = kmem_cache_create("signal_cache", sizeof(struct signal_struct), 0, - SLAB_HWCACHE_ALIGN|SLAB_PANIC, NULL); + SLAB_HWCACHE_ALIGN|SLAB_PANIC|SLAB_NOTRACK, NULL); files_cachep = kmem_cache_create("files_cache", sizeof(struct files_struct), 0, - SLAB_HWCACHE_ALIGN|SLAB_PANIC, NULL); + SLAB_HWCACHE_ALIGN|SLAB_PANIC|SLAB_NOTRACK, NULL); fs_cachep = kmem_cache_create("fs_cache", sizeof(struct fs_struct), 0, - SLAB_HWCACHE_ALIGN|SLAB_PANIC, NULL); + SLAB_HWCACHE_ALIGN|SLAB_PANIC|SLAB_NOTRACK, NULL); mm_cachep = kmem_cache_create("mm_struct", sizeof(struct mm_struct), ARCH_MIN_MMSTRUCT_ALIGN, - SLAB_HWCACHE_ALIGN|SLAB_PANIC, NULL); + SLAB_HWCACHE_ALIGN|SLAB_PANIC|SLAB_NOTRACK, NULL); + vm_area_cachep = KMEM_CACHE(vm_area_struct, SLAB_PANIC); mmap_init(); } @@ -1543,12 +1573,16 @@ static int unshare_fs(unsigned long unshare_flags, struct fs_struct **new_fsp) { struct fs_struct *fs = current->fs; - if ((unshare_flags & CLONE_FS) && - (fs && atomic_read(&fs->count) > 1)) { - *new_fsp = __copy_fs_struct(current->fs); - if (!*new_fsp) - return -ENOMEM; - } + if (!(unshare_flags & CLONE_FS) || !fs) + return 0; + + /* don't need lock here; in the worst case we'll do useless copy */ + if (fs->users == 1) + return 0; + + *new_fsp = copy_fs_struct(fs); + if (!*new_fsp) + return -ENOMEM; return 0; } @@ -1664,8 +1698,13 @@ SYSCALL_DEFINE1(unshare, unsigned long, unshare_flags) if (new_fs) { fs = current->fs; + write_lock(&fs->lock); current->fs = new_fs; - new_fs = fs; + if (--fs->users) + new_fs = NULL; + else + new_fs = fs; + write_unlock(&fs->lock); } if (new_mm) { @@ -1704,7 +1743,7 @@ bad_unshare_cleanup_sigh: bad_unshare_cleanup_fs: if (new_fs) - put_fs_struct(new_fs); + free_fs_struct(new_fs); bad_unshare_cleanup_thread: bad_unshare_out: diff --git a/kernel/freezer.c b/kernel/freezer.c index 2f4936cf708..bd1d42b17cb 100644 --- a/kernel/freezer.c +++ b/kernel/freezer.c @@ -44,12 +44,19 @@ void refrigerator(void) recalc_sigpending(); /* We sent fake signal, clean it up */ spin_unlock_irq(¤t->sighand->siglock); + /* prevent accounting of that task to load */ + current->flags |= PF_FREEZING; + for (;;) { set_current_state(TASK_UNINTERRUPTIBLE); if (!frozen(current)) break; schedule(); } + + /* Remove the accounting blocker */ + current->flags &= ~PF_FREEZING; + pr_debug("%s left refrigerator\n", current->comm); __set_current_state(save); } diff --git a/kernel/futex.c b/kernel/futex.c index 6b50a024bca..fb65e822fc4 100644 --- a/kernel/futex.c +++ b/kernel/futex.c @@ -19,6 +19,10 @@ * PRIVATE futexes by Eric Dumazet * Copyright (C) 2007 Eric Dumazet <dada1@cosmosbay.com> * + * Requeue-PI support by Darren Hart <dvhltc@us.ibm.com> + * Copyright (C) IBM Corporation, 2009 + * Thanks to Thomas Gleixner for conceptual design and careful reviews. + * * Thanks to Ben LaHaise for yelling "hashed waitqueues" loudly * enough at me, Linus for the original (flawed) idea, Matthew * Kirkwood for proof-of-concept implementation. @@ -85,31 +89,36 @@ struct futex_pi_state { union futex_key key; }; -/* - * We use this hashed waitqueue instead of a normal wait_queue_t, so +/** + * struct futex_q - The hashed futex queue entry, one per waiting task + * @task: the task waiting on the futex + * @lock_ptr: the hash bucket lock + * @key: the key the futex is hashed on + * @pi_state: optional priority inheritance state + * @rt_waiter: rt_waiter storage for use with requeue_pi + * @requeue_pi_key: the requeue_pi target futex key + * @bitset: bitset for the optional bitmasked wakeup + * + * We use this hashed waitqueue, instead of a normal wait_queue_t, so * we can wake only the relevant ones (hashed queues may be shared). * * A futex_q has a woken state, just like tasks have TASK_RUNNING. * It is considered woken when plist_node_empty(&q->list) || q->lock_ptr == 0. * The order of wakup is always to make the first condition true, then - * wake up q->waiter, then make the second condition true. + * the second. + * + * PI futexes are typically woken before they are removed from the hash list via + * the rt_mutex code. See unqueue_me_pi(). */ struct futex_q { struct plist_node list; - /* There can only be a single waiter */ - wait_queue_head_t waiter; - /* Which hash list lock to use: */ + struct task_struct *task; spinlock_t *lock_ptr; - - /* Key which the futex is hashed on: */ union futex_key key; - - /* Optional priority inheritance state: */ struct futex_pi_state *pi_state; - struct task_struct *task; - - /* Bitset for the optional bitmasked wakeup */ + struct rt_mutex_waiter *rt_waiter; + union futex_key *requeue_pi_key; u32 bitset; }; @@ -141,7 +150,8 @@ static struct futex_hash_bucket *hash_futex(union futex_key *key) */ static inline int match_futex(union futex_key *key1, union futex_key *key2) { - return (key1->both.word == key2->both.word + return (key1 && key2 + && key1->both.word == key2->both.word && key1->both.ptr == key2->both.ptr && key1->both.offset == key2->both.offset); } @@ -189,10 +199,12 @@ static void drop_futex_key_refs(union futex_key *key) } /** - * get_futex_key - Get parameters which are the keys for a futex. - * @uaddr: virtual address of the futex - * @fshared: 0 for a PROCESS_PRIVATE futex, 1 for PROCESS_SHARED - * @key: address where result is stored. + * get_futex_key() - Get parameters which are the keys for a futex + * @uaddr: virtual address of the futex + * @fshared: 0 for a PROCESS_PRIVATE futex, 1 for PROCESS_SHARED + * @key: address where result is stored. + * @rw: mapping needs to be read/write (values: VERIFY_READ, + * VERIFY_WRITE) * * Returns a negative error code or 0 * The key words are stored in *key on success. @@ -203,7 +215,8 @@ static void drop_futex_key_refs(union futex_key *key) * * lock_page() might sleep, the caller should not hold a spinlock. */ -static int get_futex_key(u32 __user *uaddr, int fshared, union futex_key *key) +static int +get_futex_key(u32 __user *uaddr, int fshared, union futex_key *key, int rw) { unsigned long address = (unsigned long)uaddr; struct mm_struct *mm = current->mm; @@ -226,7 +239,7 @@ static int get_futex_key(u32 __user *uaddr, int fshared, union futex_key *key) * but access_ok() should be faster than find_vma() */ if (!fshared) { - if (unlikely(!access_ok(VERIFY_WRITE, uaddr, sizeof(u32)))) + if (unlikely(!access_ok(rw, uaddr, sizeof(u32)))) return -EFAULT; key->private.mm = mm; key->private.address = address; @@ -235,10 +248,11 @@ static int get_futex_key(u32 __user *uaddr, int fshared, union futex_key *key) } again: - err = get_user_pages_fast(address, 1, 0, &page); + err = get_user_pages_fast(address, 1, rw == VERIFY_WRITE, &page); if (err < 0) return err; + page = compound_head(page); lock_page(page); if (!page->mapping) { unlock_page(page); @@ -276,6 +290,44 @@ void put_futex_key(int fshared, union futex_key *key) drop_futex_key_refs(key); } +/** + * fault_in_user_writeable() - Fault in user address and verify RW access + * @uaddr: pointer to faulting user space address + * + * Slow path to fixup the fault we just took in the atomic write + * access to @uaddr. + * + * We have no generic implementation of a non destructive write to the + * user address. We know that we faulted in the atomic pagefault + * disabled section so we can as well avoid the #PF overhead by + * calling get_user_pages() right away. + */ +static int fault_in_user_writeable(u32 __user *uaddr) +{ + int ret = get_user_pages(current, current->mm, (unsigned long)uaddr, + 1, 1, 0, NULL, NULL); + return ret < 0 ? ret : 0; +} + +/** + * futex_top_waiter() - Return the highest priority waiter on a futex + * @hb: the hash bucket the futex_q's reside in + * @key: the futex key (to distinguish it from other futex futex_q's) + * + * Must be called with the hb lock held. + */ +static struct futex_q *futex_top_waiter(struct futex_hash_bucket *hb, + union futex_key *key) +{ + struct futex_q *this; + + plist_for_each_entry(this, &hb->chain, list) { + if (match_futex(&this->key, key)) + return this; + } + return NULL; +} + static u32 cmpxchg_futex_value_locked(u32 __user *uaddr, u32 uval, u32 newval) { u32 curval; @@ -537,28 +589,160 @@ lookup_pi_state(u32 uval, struct futex_hash_bucket *hb, return 0; } +/** + * futex_lock_pi_atomic() - Atomic work required to acquire a pi aware futex + * @uaddr: the pi futex user address + * @hb: the pi futex hash bucket + * @key: the futex key associated with uaddr and hb + * @ps: the pi_state pointer where we store the result of the + * lookup + * @task: the task to perform the atomic lock work for. This will + * be "current" except in the case of requeue pi. + * @set_waiters: force setting the FUTEX_WAITERS bit (1) or not (0) + * + * Returns: + * 0 - ready to wait + * 1 - acquired the lock + * <0 - error + * + * The hb->lock and futex_key refs shall be held by the caller. + */ +static int futex_lock_pi_atomic(u32 __user *uaddr, struct futex_hash_bucket *hb, + union futex_key *key, + struct futex_pi_state **ps, + struct task_struct *task, int set_waiters) +{ + int lock_taken, ret, ownerdied = 0; + u32 uval, newval, curval; + +retry: + ret = lock_taken = 0; + + /* + * To avoid races, we attempt to take the lock here again + * (by doing a 0 -> TID atomic cmpxchg), while holding all + * the locks. It will most likely not succeed. + */ + newval = task_pid_vnr(task); + if (set_waiters) + newval |= FUTEX_WAITERS; + + curval = cmpxchg_futex_value_locked(uaddr, 0, newval); + + if (unlikely(curval == -EFAULT)) + return -EFAULT; + + /* + * Detect deadlocks. + */ + if ((unlikely((curval & FUTEX_TID_MASK) == task_pid_vnr(task)))) + return -EDEADLK; + + /* + * Surprise - we got the lock. Just return to userspace: + */ + if (unlikely(!curval)) + return 1; + + uval = curval; + + /* + * Set the FUTEX_WAITERS flag, so the owner will know it has someone + * to wake at the next unlock. + */ + newval = curval | FUTEX_WAITERS; + + /* + * There are two cases, where a futex might have no owner (the + * owner TID is 0): OWNER_DIED. We take over the futex in this + * case. We also do an unconditional take over, when the owner + * of the futex died. + * + * This is safe as we are protected by the hash bucket lock ! + */ + if (unlikely(ownerdied || !(curval & FUTEX_TID_MASK))) { + /* Keep the OWNER_DIED bit */ + newval = (curval & ~FUTEX_TID_MASK) | task_pid_vnr(task); + ownerdied = 0; + lock_taken = 1; + } + + curval = cmpxchg_futex_value_locked(uaddr, uval, newval); + + if (unlikely(curval == -EFAULT)) + return -EFAULT; + if (unlikely(curval != uval)) + goto retry; + + /* + * We took the lock due to owner died take over. + */ + if (unlikely(lock_taken)) + return 1; + + /* + * We dont have the lock. Look up the PI state (or create it if + * we are the first waiter): + */ + ret = lookup_pi_state(uval, hb, key, ps); + + if (unlikely(ret)) { + switch (ret) { + case -ESRCH: + /* + * No owner found for this futex. Check if the + * OWNER_DIED bit is set to figure out whether + * this is a robust futex or not. + */ + if (get_futex_value_locked(&curval, uaddr)) + return -EFAULT; + + /* + * We simply start over in case of a robust + * futex. The code above will take the futex + * and return happy. + */ + if (curval & FUTEX_OWNER_DIED) { + ownerdied = 1; + goto retry; + } + default: + break; + } + } + + return ret; +} + /* * The hash bucket lock must be held when this is called. * Afterwards, the futex_q must not be accessed. */ static void wake_futex(struct futex_q *q) { - plist_del(&q->list, &q->list.plist); + struct task_struct *p = q->task; + /* - * The lock in wake_up_all() is a crucial memory barrier after the - * plist_del() and also before assigning to q->lock_ptr. + * We set q->lock_ptr = NULL _before_ we wake up the task. If + * a non futex wake up happens on another CPU then the task + * might exit and p would dereference a non existing task + * struct. Prevent this by holding a reference on p across the + * wake up. */ - wake_up(&q->waiter); + get_task_struct(p); + + plist_del(&q->list, &q->list.plist); /* - * The waiting task can free the futex_q as soon as this is written, - * without taking any locks. This must come last. - * - * A memory barrier is required here to prevent the following store to - * lock_ptr from getting ahead of the wakeup. Clearing the lock at the - * end of wake_up() does not prevent this store from moving. + * The waiting task can free the futex_q as soon as + * q->lock_ptr = NULL is written, without taking any locks. A + * memory barrier is required here to prevent the following + * store to lock_ptr from getting ahead of the plist_del. */ smp_wmb(); q->lock_ptr = NULL; + + wake_up_state(p, TASK_NORMAL); + put_task_struct(p); } static int wake_futex_pi(u32 __user *uaddr, u32 uval, struct futex_q *this) @@ -677,7 +861,7 @@ static int futex_wake(u32 __user *uaddr, int fshared, int nr_wake, u32 bitset) if (!bitset) return -EINVAL; - ret = get_futex_key(uaddr, fshared, &key); + ret = get_futex_key(uaddr, fshared, &key, VERIFY_READ); if (unlikely(ret != 0)) goto out; @@ -687,7 +871,7 @@ static int futex_wake(u32 __user *uaddr, int fshared, int nr_wake, u32 bitset) plist_for_each_entry_safe(this, next, head, list) { if (match_futex (&this->key, &key)) { - if (this->pi_state) { + if (this->pi_state || this->rt_waiter) { ret = -EINVAL; break; } @@ -723,21 +907,20 @@ futex_wake_op(u32 __user *uaddr1, int fshared, u32 __user *uaddr2, int ret, op_ret; retry: - ret = get_futex_key(uaddr1, fshared, &key1); + ret = get_futex_key(uaddr1, fshared, &key1, VERIFY_READ); if (unlikely(ret != 0)) goto out; - ret = get_futex_key(uaddr2, fshared, &key2); + ret = get_futex_key(uaddr2, fshared, &key2, VERIFY_WRITE); if (unlikely(ret != 0)) goto out_put_key1; hb1 = hash_futex(&key1); hb2 = hash_futex(&key2); - double_lock_hb(hb1, hb2); retry_private: + double_lock_hb(hb1, hb2); op_ret = futex_atomic_op_inuser(op, uaddr2); if (unlikely(op_ret < 0)) { - u32 dummy; double_unlock_hb(hb1, hb2); @@ -755,7 +938,7 @@ retry_private: goto out_put_keys; } - ret = get_user(dummy, uaddr2); + ret = fault_in_user_writeable(uaddr2); if (ret) goto out_put_keys; @@ -800,24 +983,197 @@ out: return ret; } -/* - * Requeue all waiters hashed on one physical page to another - * physical page. +/** + * requeue_futex() - Requeue a futex_q from one hb to another + * @q: the futex_q to requeue + * @hb1: the source hash_bucket + * @hb2: the target hash_bucket + * @key2: the new key for the requeued futex_q + */ +static inline +void requeue_futex(struct futex_q *q, struct futex_hash_bucket *hb1, + struct futex_hash_bucket *hb2, union futex_key *key2) +{ + + /* + * If key1 and key2 hash to the same bucket, no need to + * requeue. + */ + if (likely(&hb1->chain != &hb2->chain)) { + plist_del(&q->list, &hb1->chain); + plist_add(&q->list, &hb2->chain); + q->lock_ptr = &hb2->lock; +#ifdef CONFIG_DEBUG_PI_LIST + q->list.plist.lock = &hb2->lock; +#endif + } + get_futex_key_refs(key2); + q->key = *key2; +} + +/** + * requeue_pi_wake_futex() - Wake a task that acquired the lock during requeue + * @q: the futex_q + * @key: the key of the requeue target futex + * @hb: the hash_bucket of the requeue target futex + * + * During futex_requeue, with requeue_pi=1, it is possible to acquire the + * target futex if it is uncontended or via a lock steal. Set the futex_q key + * to the requeue target futex so the waiter can detect the wakeup on the right + * futex, but remove it from the hb and NULL the rt_waiter so it can detect + * atomic lock acquisition. Set the q->lock_ptr to the requeue target hb->lock + * to protect access to the pi_state to fixup the owner later. Must be called + * with both q->lock_ptr and hb->lock held. + */ +static inline +void requeue_pi_wake_futex(struct futex_q *q, union futex_key *key, + struct futex_hash_bucket *hb) +{ + get_futex_key_refs(key); + q->key = *key; + + WARN_ON(plist_node_empty(&q->list)); + plist_del(&q->list, &q->list.plist); + + WARN_ON(!q->rt_waiter); + q->rt_waiter = NULL; + + q->lock_ptr = &hb->lock; +#ifdef CONFIG_DEBUG_PI_LIST + q->list.plist.lock = &hb->lock; +#endif + + wake_up_state(q->task, TASK_NORMAL); +} + +/** + * futex_proxy_trylock_atomic() - Attempt an atomic lock for the top waiter + * @pifutex: the user address of the to futex + * @hb1: the from futex hash bucket, must be locked by the caller + * @hb2: the to futex hash bucket, must be locked by the caller + * @key1: the from futex key + * @key2: the to futex key + * @ps: address to store the pi_state pointer + * @set_waiters: force setting the FUTEX_WAITERS bit (1) or not (0) + * + * Try and get the lock on behalf of the top waiter if we can do it atomically. + * Wake the top waiter if we succeed. If the caller specified set_waiters, + * then direct futex_lock_pi_atomic() to force setting the FUTEX_WAITERS bit. + * hb1 and hb2 must be held by the caller. + * + * Returns: + * 0 - failed to acquire the lock atomicly + * 1 - acquired the lock + * <0 - error + */ +static int futex_proxy_trylock_atomic(u32 __user *pifutex, + struct futex_hash_bucket *hb1, + struct futex_hash_bucket *hb2, + union futex_key *key1, union futex_key *key2, + struct futex_pi_state **ps, int set_waiters) +{ + struct futex_q *top_waiter = NULL; + u32 curval; + int ret; + + if (get_futex_value_locked(&curval, pifutex)) + return -EFAULT; + + /* + * Find the top_waiter and determine if there are additional waiters. + * If the caller intends to requeue more than 1 waiter to pifutex, + * force futex_lock_pi_atomic() to set the FUTEX_WAITERS bit now, + * as we have means to handle the possible fault. If not, don't set + * the bit unecessarily as it will force the subsequent unlock to enter + * the kernel. + */ + top_waiter = futex_top_waiter(hb1, key1); + + /* There are no waiters, nothing for us to do. */ + if (!top_waiter) + return 0; + + /* Ensure we requeue to the expected futex. */ + if (!match_futex(top_waiter->requeue_pi_key, key2)) + return -EINVAL; + + /* + * Try to take the lock for top_waiter. Set the FUTEX_WAITERS bit in + * the contended case or if set_waiters is 1. The pi_state is returned + * in ps in contended cases. + */ + ret = futex_lock_pi_atomic(pifutex, hb2, key2, ps, top_waiter->task, + set_waiters); + if (ret == 1) + requeue_pi_wake_futex(top_waiter, key2, hb2); + + return ret; +} + +/** + * futex_requeue() - Requeue waiters from uaddr1 to uaddr2 + * uaddr1: source futex user address + * uaddr2: target futex user address + * nr_wake: number of waiters to wake (must be 1 for requeue_pi) + * nr_requeue: number of waiters to requeue (0-INT_MAX) + * requeue_pi: if we are attempting to requeue from a non-pi futex to a + * pi futex (pi to pi requeue is not supported) + * + * Requeue waiters on uaddr1 to uaddr2. In the requeue_pi case, try to acquire + * uaddr2 atomically on behalf of the top waiter. + * + * Returns: + * >=0 - on success, the number of tasks requeued or woken + * <0 - on error */ static int futex_requeue(u32 __user *uaddr1, int fshared, u32 __user *uaddr2, - int nr_wake, int nr_requeue, u32 *cmpval) + int nr_wake, int nr_requeue, u32 *cmpval, + int requeue_pi) { union futex_key key1 = FUTEX_KEY_INIT, key2 = FUTEX_KEY_INIT; + int drop_count = 0, task_count = 0, ret; + struct futex_pi_state *pi_state = NULL; struct futex_hash_bucket *hb1, *hb2; struct plist_head *head1; struct futex_q *this, *next; - int ret, drop_count = 0; + u32 curval2; + + if (requeue_pi) { + /* + * requeue_pi requires a pi_state, try to allocate it now + * without any locks in case it fails. + */ + if (refill_pi_state_cache()) + return -ENOMEM; + /* + * requeue_pi must wake as many tasks as it can, up to nr_wake + * + nr_requeue, since it acquires the rt_mutex prior to + * returning to userspace, so as to not leave the rt_mutex with + * waiters and no owner. However, second and third wake-ups + * cannot be predicted as they involve race conditions with the + * first wake and a fault while looking up the pi_state. Both + * pthread_cond_signal() and pthread_cond_broadcast() should + * use nr_wake=1. + */ + if (nr_wake != 1) + return -EINVAL; + } retry: - ret = get_futex_key(uaddr1, fshared, &key1); + if (pi_state != NULL) { + /* + * We will have to lookup the pi_state again, so free this one + * to keep the accounting correct. + */ + free_pi_state(pi_state); + pi_state = NULL; + } + + ret = get_futex_key(uaddr1, fshared, &key1, VERIFY_READ); if (unlikely(ret != 0)) goto out; - ret = get_futex_key(uaddr2, fshared, &key2); + ret = get_futex_key(uaddr2, fshared, &key2, + requeue_pi ? VERIFY_WRITE : VERIFY_READ); if (unlikely(ret != 0)) goto out_put_key1; @@ -852,38 +1208,125 @@ retry_private: } } + if (requeue_pi && (task_count - nr_wake < nr_requeue)) { + /* + * Attempt to acquire uaddr2 and wake the top waiter. If we + * intend to requeue waiters, force setting the FUTEX_WAITERS + * bit. We force this here where we are able to easily handle + * faults rather in the requeue loop below. + */ + ret = futex_proxy_trylock_atomic(uaddr2, hb1, hb2, &key1, + &key2, &pi_state, nr_requeue); + + /* + * At this point the top_waiter has either taken uaddr2 or is + * waiting on it. If the former, then the pi_state will not + * exist yet, look it up one more time to ensure we have a + * reference to it. + */ + if (ret == 1) { + WARN_ON(pi_state); + drop_count++; + task_count++; + ret = get_futex_value_locked(&curval2, uaddr2); + if (!ret) + ret = lookup_pi_state(curval2, hb2, &key2, + &pi_state); + } + + switch (ret) { + case 0: + break; + case -EFAULT: + double_unlock_hb(hb1, hb2); + put_futex_key(fshared, &key2); + put_futex_key(fshared, &key1); + ret = fault_in_user_writeable(uaddr2); + if (!ret) + goto retry; + goto out; + case -EAGAIN: + /* The owner was exiting, try again. */ + double_unlock_hb(hb1, hb2); + put_futex_key(fshared, &key2); + put_futex_key(fshared, &key1); + cond_resched(); + goto retry; + default: + goto out_unlock; + } + } + head1 = &hb1->chain; plist_for_each_entry_safe(this, next, head1, list) { - if (!match_futex (&this->key, &key1)) + if (task_count - nr_wake >= nr_requeue) + break; + + if (!match_futex(&this->key, &key1)) continue; - if (++ret <= nr_wake) { + + /* + * FUTEX_WAIT_REQEUE_PI and FUTEX_CMP_REQUEUE_PI should always + * be paired with each other and no other futex ops. + */ + if ((requeue_pi && !this->rt_waiter) || + (!requeue_pi && this->rt_waiter)) { + ret = -EINVAL; + break; + } + + /* + * Wake nr_wake waiters. For requeue_pi, if we acquired the + * lock, we already woke the top_waiter. If not, it will be + * woken by futex_unlock_pi(). + */ + if (++task_count <= nr_wake && !requeue_pi) { wake_futex(this); - } else { - /* - * If key1 and key2 hash to the same bucket, no need to - * requeue. - */ - if (likely(head1 != &hb2->chain)) { - plist_del(&this->list, &hb1->chain); - plist_add(&this->list, &hb2->chain); - this->lock_ptr = &hb2->lock; -#ifdef CONFIG_DEBUG_PI_LIST - this->list.plist.lock = &hb2->lock; -#endif - } - this->key = key2; - get_futex_key_refs(&key2); - drop_count++; + continue; + } - if (ret - nr_wake >= nr_requeue) - break; + /* Ensure we requeue to the expected futex for requeue_pi. */ + if (requeue_pi && !match_futex(this->requeue_pi_key, &key2)) { + ret = -EINVAL; + break; } + + /* + * Requeue nr_requeue waiters and possibly one more in the case + * of requeue_pi if we couldn't acquire the lock atomically. + */ + if (requeue_pi) { + /* Prepare the waiter to take the rt_mutex. */ + atomic_inc(&pi_state->refcount); + this->pi_state = pi_state; + ret = rt_mutex_start_proxy_lock(&pi_state->pi_mutex, + this->rt_waiter, + this->task, 1); + if (ret == 1) { + /* We got the lock. */ + requeue_pi_wake_futex(this, &key2, hb2); + drop_count++; + continue; + } else if (ret) { + /* -EDEADLK */ + this->pi_state = NULL; + free_pi_state(pi_state); + goto out_unlock; + } + } + requeue_futex(this, hb1, hb2, &key2); + drop_count++; } out_unlock: double_unlock_hb(hb1, hb2); - /* drop_futex_key_refs() must be called outside the spinlocks. */ + /* + * drop_futex_key_refs() must be called outside the spinlocks. During + * the requeue we moved futex_q's from the hash bucket at key1 to the + * one at key2 and updated their key pointer. We no longer need to + * hold the references to key1. + */ while (--drop_count >= 0) drop_futex_key_refs(&key1); @@ -892,7 +1335,9 @@ out_put_keys: out_put_key1: put_futex_key(fshared, &key1); out: - return ret; + if (pi_state != NULL) + free_pi_state(pi_state); + return ret ? ret : task_count; } /* The key must be already stored in q->key. */ @@ -900,8 +1345,6 @@ static inline struct futex_hash_bucket *queue_lock(struct futex_q *q) { struct futex_hash_bucket *hb; - init_waitqueue_head(&q->waiter); - get_futex_key_refs(&q->key); hb = hash_futex(&q->key); q->lock_ptr = &hb->lock; @@ -910,6 +1353,25 @@ static inline struct futex_hash_bucket *queue_lock(struct futex_q *q) return hb; } +static inline void +queue_unlock(struct futex_q *q, struct futex_hash_bucket *hb) +{ + spin_unlock(&hb->lock); + drop_futex_key_refs(&q->key); +} + +/** + * queue_me() - Enqueue the futex_q on the futex_hash_bucket + * @q: The futex_q to enqueue + * @hb: The destination hash bucket + * + * The hb->lock must be held by the caller, and is released here. A call to + * queue_me() is typically paired with exactly one call to unqueue_me(). The + * exceptions involve the PI related operations, which may use unqueue_me_pi() + * or nothing if the unqueue is done as part of the wake process and the unqueue + * state is implicit in the state of woken task (see futex_wait_requeue_pi() for + * an example). + */ static inline void queue_me(struct futex_q *q, struct futex_hash_bucket *hb) { int prio; @@ -933,19 +1395,17 @@ static inline void queue_me(struct futex_q *q, struct futex_hash_bucket *hb) spin_unlock(&hb->lock); } -static inline void -queue_unlock(struct futex_q *q, struct futex_hash_bucket *hb) -{ - spin_unlock(&hb->lock); - drop_futex_key_refs(&q->key); -} - -/* - * queue_me and unqueue_me must be called as a pair, each - * exactly once. They are called with the hashed spinlock held. +/** + * unqueue_me() - Remove the futex_q from its futex_hash_bucket + * @q: The futex_q to unqueue + * + * The q->lock_ptr must not be held by the caller. A call to unqueue_me() must + * be paired with exactly one earlier call to queue_me(). + * + * Returns: + * 1 - if the futex_q was still queued (and we removed unqueued it) + * 0 - if the futex_q was already removed by the waking thread */ - -/* Return 1 if we were still queued (ie. 0 means we were woken) */ static int unqueue_me(struct futex_q *q) { spinlock_t *lock_ptr; @@ -1090,7 +1550,7 @@ retry: handle_fault: spin_unlock(q->lock_ptr); - ret = get_user(uval, uaddr); + ret = fault_in_user_writeable(uaddr); spin_lock(q->lock_ptr); @@ -1112,35 +1572,146 @@ handle_fault: */ #define FLAGS_SHARED 0x01 #define FLAGS_CLOCKRT 0x02 +#define FLAGS_HAS_TIMEOUT 0x04 static long futex_wait_restart(struct restart_block *restart); -static int futex_wait(u32 __user *uaddr, int fshared, - u32 val, ktime_t *abs_time, u32 bitset, int clockrt) +/** + * fixup_owner() - Post lock pi_state and corner case management + * @uaddr: user address of the futex + * @fshared: whether the futex is shared (1) or not (0) + * @q: futex_q (contains pi_state and access to the rt_mutex) + * @locked: if the attempt to take the rt_mutex succeeded (1) or not (0) + * + * After attempting to lock an rt_mutex, this function is called to cleanup + * the pi_state owner as well as handle race conditions that may allow us to + * acquire the lock. Must be called with the hb lock held. + * + * Returns: + * 1 - success, lock taken + * 0 - success, lock not taken + * <0 - on error (-EFAULT) + */ +static int fixup_owner(u32 __user *uaddr, int fshared, struct futex_q *q, + int locked) { - struct task_struct *curr = current; - struct restart_block *restart; - DECLARE_WAITQUEUE(wait, curr); - struct futex_hash_bucket *hb; - struct futex_q q; - u32 uval; - int ret; - struct hrtimer_sleeper t; - int rem = 0; + struct task_struct *owner; + int ret = 0; - if (!bitset) - return -EINVAL; + if (locked) { + /* + * Got the lock. We might not be the anticipated owner if we + * did a lock-steal - fix up the PI-state in that case: + */ + if (q->pi_state->owner != current) + ret = fixup_pi_state_owner(uaddr, q, current, fshared); + goto out; + } - q.pi_state = NULL; - q.bitset = bitset; -retry: - q.key = FUTEX_KEY_INIT; - ret = get_futex_key(uaddr, fshared, &q.key); - if (unlikely(ret != 0)) + /* + * Catch the rare case, where the lock was released when we were on the + * way back before we locked the hash bucket. + */ + if (q->pi_state->owner == current) { + /* + * Try to get the rt_mutex now. This might fail as some other + * task acquired the rt_mutex after we removed ourself from the + * rt_mutex waiters list. + */ + if (rt_mutex_trylock(&q->pi_state->pi_mutex)) { + locked = 1; + goto out; + } + + /* + * pi_state is incorrect, some other task did a lock steal and + * we returned due to timeout or signal without taking the + * rt_mutex. Too late. We can access the rt_mutex_owner without + * locking, as the other task is now blocked on the hash bucket + * lock. Fix the state up. + */ + owner = rt_mutex_owner(&q->pi_state->pi_mutex); + ret = fixup_pi_state_owner(uaddr, q, owner, fshared); goto out; + } -retry_private: - hb = queue_lock(&q); + /* + * Paranoia check. If we did not take the lock, then we should not be + * the owner, nor the pending owner, of the rt_mutex. + */ + if (rt_mutex_owner(&q->pi_state->pi_mutex) == current) + printk(KERN_ERR "fixup_owner: ret = %d pi-mutex: %p " + "pi-state %p\n", ret, + q->pi_state->pi_mutex.owner, + q->pi_state->owner); + +out: + return ret ? ret : locked; +} + +/** + * futex_wait_queue_me() - queue_me() and wait for wakeup, timeout, or signal + * @hb: the futex hash bucket, must be locked by the caller + * @q: the futex_q to queue up on + * @timeout: the prepared hrtimer_sleeper, or null for no timeout + */ +static void futex_wait_queue_me(struct futex_hash_bucket *hb, struct futex_q *q, + struct hrtimer_sleeper *timeout) +{ + /* + * The task state is guaranteed to be set before another task can + * wake it. set_current_state() is implemented using set_mb() and + * queue_me() calls spin_unlock() upon completion, both serializing + * access to the hash list and forcing another memory barrier. + */ + set_current_state(TASK_INTERRUPTIBLE); + queue_me(q, hb); + + /* Arm the timer */ + if (timeout) { + hrtimer_start_expires(&timeout->timer, HRTIMER_MODE_ABS); + if (!hrtimer_active(&timeout->timer)) + timeout->task = NULL; + } + + /* + * If we have been removed from the hash list, then another task + * has tried to wake us, and we can skip the call to schedule(). + */ + if (likely(!plist_node_empty(&q->list))) { + /* + * If the timer has already expired, current will already be + * flagged for rescheduling. Only call schedule if there + * is no timeout, or if it has yet to expire. + */ + if (!timeout || timeout->task) + schedule(); + } + __set_current_state(TASK_RUNNING); +} + +/** + * futex_wait_setup() - Prepare to wait on a futex + * @uaddr: the futex userspace address + * @val: the expected value + * @fshared: whether the futex is shared (1) or not (0) + * @q: the associated futex_q + * @hb: storage for hash_bucket pointer to be returned to caller + * + * Setup the futex_q and locate the hash_bucket. Get the futex value and + * compare it with the expected value. Handle atomic faults internally. + * Return with the hb lock held and a q.key reference on success, and unlocked + * with no q.key reference on failure. + * + * Returns: + * 0 - uaddr contains val and hb has been locked + * <1 - -EFAULT or -EWOULDBLOCK (uaddr does not contain val) and hb is unlcoked + */ +static int futex_wait_setup(u32 __user *uaddr, u32 val, int fshared, + struct futex_q *q, struct futex_hash_bucket **hb) +{ + u32 uval; + int ret; /* * Access the page AFTER the hash-bucket is locked. @@ -1158,101 +1729,96 @@ retry_private: * A consequence is that futex_wait() can return zero and absorb * a wakeup when *uaddr != val on entry to the syscall. This is * rare, but normal. - * - * For shared futexes, we hold the mmap semaphore, so the mapping - * cannot have changed since we looked it up in get_futex_key. */ +retry: + q->key = FUTEX_KEY_INIT; + ret = get_futex_key(uaddr, fshared, &q->key, VERIFY_READ); + if (unlikely(ret != 0)) + return ret; + +retry_private: + *hb = queue_lock(q); + ret = get_futex_value_locked(&uval, uaddr); - if (unlikely(ret)) { - queue_unlock(&q, hb); + if (ret) { + queue_unlock(q, *hb); ret = get_user(uval, uaddr); if (ret) - goto out_put_key; + goto out; if (!fshared) goto retry_private; - put_futex_key(fshared, &q.key); + put_futex_key(fshared, &q->key); goto retry; } - ret = -EWOULDBLOCK; - if (unlikely(uval != val)) { - queue_unlock(&q, hb); - goto out_put_key; - } - /* Only actually queue if *uaddr contained val. */ - queue_me(&q, hb); + if (uval != val) { + queue_unlock(q, *hb); + ret = -EWOULDBLOCK; + } - /* - * There might have been scheduling since the queue_me(), as we - * cannot hold a spinlock across the get_user() in case it - * faults, and we cannot just set TASK_INTERRUPTIBLE state when - * queueing ourselves into the futex hash. This code thus has to - * rely on the futex_wake() code removing us from hash when it - * wakes us up. - */ +out: + if (ret) + put_futex_key(fshared, &q->key); + return ret; +} - /* add_wait_queue is the barrier after __set_current_state. */ - __set_current_state(TASK_INTERRUPTIBLE); - add_wait_queue(&q.waiter, &wait); - /* - * !plist_node_empty() is safe here without any lock. - * q.lock_ptr != 0 is not safe, because of ordering against wakeup. - */ - if (likely(!plist_node_empty(&q.list))) { - if (!abs_time) - schedule(); - else { - hrtimer_init_on_stack(&t.timer, - clockrt ? CLOCK_REALTIME : - CLOCK_MONOTONIC, - HRTIMER_MODE_ABS); - hrtimer_init_sleeper(&t, current); - hrtimer_set_expires_range_ns(&t.timer, *abs_time, - current->timer_slack_ns); - - hrtimer_start_expires(&t.timer, HRTIMER_MODE_ABS); - if (!hrtimer_active(&t.timer)) - t.task = NULL; +static int futex_wait(u32 __user *uaddr, int fshared, + u32 val, ktime_t *abs_time, u32 bitset, int clockrt) +{ + struct hrtimer_sleeper timeout, *to = NULL; + struct restart_block *restart; + struct futex_hash_bucket *hb; + struct futex_q q; + int ret; - /* - * the timer could have already expired, in which - * case current would be flagged for rescheduling. - * Don't bother calling schedule. - */ - if (likely(t.task)) - schedule(); + if (!bitset) + return -EINVAL; - hrtimer_cancel(&t.timer); + q.pi_state = NULL; + q.bitset = bitset; + q.rt_waiter = NULL; + q.requeue_pi_key = NULL; - /* Flag if a timeout occured */ - rem = (t.task == NULL); + if (abs_time) { + to = &timeout; - destroy_hrtimer_on_stack(&t.timer); - } + hrtimer_init_on_stack(&to->timer, clockrt ? CLOCK_REALTIME : + CLOCK_MONOTONIC, HRTIMER_MODE_ABS); + hrtimer_init_sleeper(to, current); + hrtimer_set_expires_range_ns(&to->timer, *abs_time, + current->timer_slack_ns); } - __set_current_state(TASK_RUNNING); - /* - * NOTE: we don't remove ourselves from the waitqueue because - * we are the only user of it. - */ +retry: + /* Prepare to wait on uaddr. */ + ret = futex_wait_setup(uaddr, val, fshared, &q, &hb); + if (ret) + goto out; + + /* queue_me and wait for wakeup, timeout, or a signal. */ + futex_wait_queue_me(hb, &q, to); /* If we were woken (and unqueued), we succeeded, whatever. */ ret = 0; if (!unqueue_me(&q)) goto out_put_key; ret = -ETIMEDOUT; - if (rem) + if (to && !to->task) goto out_put_key; /* - * We expect signal_pending(current), but another thread may - * have handled it for us already. + * We expect signal_pending(current), but we might be the + * victim of a spurious wakeup as well. */ + if (!signal_pending(current)) { + put_futex_key(fshared, &q.key); + goto retry; + } + ret = -ERESTARTSYS; if (!abs_time) goto out_put_key; @@ -1263,7 +1829,7 @@ retry_private: restart->futex.val = val; restart->futex.time = abs_time->tv64; restart->futex.bitset = bitset; - restart->futex.flags = 0; + restart->futex.flags = FLAGS_HAS_TIMEOUT; if (fshared) restart->futex.flags |= FLAGS_SHARED; @@ -1275,6 +1841,10 @@ retry_private: out_put_key: put_futex_key(fshared, &q.key); out: + if (to) { + hrtimer_cancel(&to->timer); + destroy_hrtimer_on_stack(&to->timer); + } return ret; } @@ -1283,13 +1853,16 @@ static long futex_wait_restart(struct restart_block *restart) { u32 __user *uaddr = (u32 __user *)restart->futex.uaddr; int fshared = 0; - ktime_t t; + ktime_t t, *tp = NULL; - t.tv64 = restart->futex.time; + if (restart->futex.flags & FLAGS_HAS_TIMEOUT) { + t.tv64 = restart->futex.time; + tp = &t; + } restart->fn = do_no_restart_syscall; if (restart->futex.flags & FLAGS_SHARED) fshared = 1; - return (long)futex_wait(uaddr, fshared, restart->futex.val, &t, + return (long)futex_wait(uaddr, fshared, restart->futex.val, tp, restart->futex.bitset, restart->futex.flags & FLAGS_CLOCKRT); } @@ -1305,11 +1878,9 @@ static int futex_lock_pi(u32 __user *uaddr, int fshared, int detect, ktime_t *time, int trylock) { struct hrtimer_sleeper timeout, *to = NULL; - struct task_struct *curr = current; struct futex_hash_bucket *hb; - u32 uval, newval, curval; struct futex_q q; - int ret, lock_taken, ownerdied = 0; + int res, ret; if (refill_pi_state_cache()) return -ENOMEM; @@ -1323,90 +1894,26 @@ static int futex_lock_pi(u32 __user *uaddr, int fshared, } q.pi_state = NULL; + q.rt_waiter = NULL; + q.requeue_pi_key = NULL; retry: q.key = FUTEX_KEY_INIT; - ret = get_futex_key(uaddr, fshared, &q.key); + ret = get_futex_key(uaddr, fshared, &q.key, VERIFY_WRITE); if (unlikely(ret != 0)) goto out; retry_private: hb = queue_lock(&q); -retry_locked: - ret = lock_taken = 0; - - /* - * To avoid races, we attempt to take the lock here again - * (by doing a 0 -> TID atomic cmpxchg), while holding all - * the locks. It will most likely not succeed. - */ - newval = task_pid_vnr(current); - - curval = cmpxchg_futex_value_locked(uaddr, 0, newval); - - if (unlikely(curval == -EFAULT)) - goto uaddr_faulted; - - /* - * Detect deadlocks. In case of REQUEUE_PI this is a valid - * situation and we return success to user space. - */ - if (unlikely((curval & FUTEX_TID_MASK) == task_pid_vnr(current))) { - ret = -EDEADLK; - goto out_unlock_put_key; - } - - /* - * Surprise - we got the lock. Just return to userspace: - */ - if (unlikely(!curval)) - goto out_unlock_put_key; - - uval = curval; - - /* - * Set the WAITERS flag, so the owner will know it has someone - * to wake at next unlock - */ - newval = curval | FUTEX_WAITERS; - - /* - * There are two cases, where a futex might have no owner (the - * owner TID is 0): OWNER_DIED. We take over the futex in this - * case. We also do an unconditional take over, when the owner - * of the futex died. - * - * This is safe as we are protected by the hash bucket lock ! - */ - if (unlikely(ownerdied || !(curval & FUTEX_TID_MASK))) { - /* Keep the OWNER_DIED bit */ - newval = (curval & ~FUTEX_TID_MASK) | task_pid_vnr(current); - ownerdied = 0; - lock_taken = 1; - } - - curval = cmpxchg_futex_value_locked(uaddr, uval, newval); - - if (unlikely(curval == -EFAULT)) - goto uaddr_faulted; - if (unlikely(curval != uval)) - goto retry_locked; - - /* - * We took the lock due to owner died take over. - */ - if (unlikely(lock_taken)) - goto out_unlock_put_key; - - /* - * We dont have the lock. Look up the PI state (or create it if - * we are the first waiter): - */ - ret = lookup_pi_state(uval, hb, &q.key, &q.pi_state); - + ret = futex_lock_pi_atomic(uaddr, hb, &q.key, &q.pi_state, current, 0); if (unlikely(ret)) { switch (ret) { - + case 1: + /* We got the lock. */ + ret = 0; + goto out_unlock_put_key; + case -EFAULT: + goto uaddr_faulted; case -EAGAIN: /* * Task is exiting and we just wait for the @@ -1416,25 +1923,6 @@ retry_locked: put_futex_key(fshared, &q.key); cond_resched(); goto retry; - - case -ESRCH: - /* - * No owner found for this futex. Check if the - * OWNER_DIED bit is set to figure out whether - * this is a robust futex or not. - */ - if (get_futex_value_locked(&curval, uaddr)) - goto uaddr_faulted; - - /* - * We simply start over in case of a robust - * futex. The code above will take the futex - * and return happy. - */ - if (curval & FUTEX_OWNER_DIED) { - ownerdied = 1; - goto retry_locked; - } default: goto out_unlock_put_key; } @@ -1458,71 +1946,21 @@ retry_locked: } spin_lock(q.lock_ptr); - - if (!ret) { - /* - * Got the lock. We might not be the anticipated owner - * if we did a lock-steal - fix up the PI-state in - * that case: - */ - if (q.pi_state->owner != curr) - ret = fixup_pi_state_owner(uaddr, &q, curr, fshared); - } else { - /* - * Catch the rare case, where the lock was released - * when we were on the way back before we locked the - * hash bucket. - */ - if (q.pi_state->owner == curr) { - /* - * Try to get the rt_mutex now. This might - * fail as some other task acquired the - * rt_mutex after we removed ourself from the - * rt_mutex waiters list. - */ - if (rt_mutex_trylock(&q.pi_state->pi_mutex)) - ret = 0; - else { - /* - * pi_state is incorrect, some other - * task did a lock steal and we - * returned due to timeout or signal - * without taking the rt_mutex. Too - * late. We can access the - * rt_mutex_owner without locking, as - * the other task is now blocked on - * the hash bucket lock. Fix the state - * up. - */ - struct task_struct *owner; - int res; - - owner = rt_mutex_owner(&q.pi_state->pi_mutex); - res = fixup_pi_state_owner(uaddr, &q, owner, - fshared); - - /* propagate -EFAULT, if the fixup failed */ - if (res) - ret = res; - } - } else { - /* - * Paranoia check. If we did not take the lock - * in the trylock above, then we should not be - * the owner of the rtmutex, neither the real - * nor the pending one: - */ - if (rt_mutex_owner(&q.pi_state->pi_mutex) == curr) - printk(KERN_ERR "futex_lock_pi: ret = %d " - "pi-mutex: %p pi-state %p\n", ret, - q.pi_state->pi_mutex.owner, - q.pi_state->owner); - } - } + /* + * Fixup the pi_state owner and possibly acquire the lock if we + * haven't already. + */ + res = fixup_owner(uaddr, fshared, &q, !ret); + /* + * If fixup_owner() returned an error, proprogate that. If it acquired + * the lock, clear our -ETIMEDOUT or -EINTR. + */ + if (res) + ret = (res < 0) ? res : 0; /* - * If fixup_pi_state_owner() faulted and was unable to handle the - * fault, unlock it and return the fault to userspace. + * If fixup_owner() faulted and was unable to handle the fault, unlock + * it and return the fault to userspace. */ if (ret && (rt_mutex_owner(&q.pi_state->pi_mutex) == current)) rt_mutex_unlock(&q.pi_state->pi_mutex); @@ -1530,9 +1968,7 @@ retry_locked: /* Unqueue and drop the lock */ unqueue_me_pi(&q); - if (to) - destroy_hrtimer_on_stack(&to->timer); - return ret != -EINTR ? ret : -ERESTARTNOINTR; + goto out; out_unlock_put_key: queue_unlock(&q, hb); @@ -1542,19 +1978,12 @@ out_put_key: out: if (to) destroy_hrtimer_on_stack(&to->timer); - return ret; + return ret != -EINTR ? ret : -ERESTARTNOINTR; uaddr_faulted: - /* - * We have to r/w *(int __user *)uaddr, and we have to modify it - * atomically. Therefore, if we continue to fault after get_user() - * below, we need to handle the fault ourselves, while still holding - * the mmap_sem. This can occur if the uaddr is under contention as - * we have to drop the mmap_sem in order to call get_user(). - */ queue_unlock(&q, hb); - ret = get_user(uval, uaddr); + ret = fault_in_user_writeable(uaddr); if (ret) goto out_put_key; @@ -1565,7 +1994,6 @@ uaddr_faulted: goto retry; } - /* * Userspace attempted a TID -> 0 atomic transition, and failed. * This is the in-kernel slowpath: we look up the PI state (if any), @@ -1589,7 +2017,7 @@ retry: if ((uval & FUTEX_TID_MASK) != task_pid_vnr(current)) return -EPERM; - ret = get_futex_key(uaddr, fshared, &key); + ret = get_futex_key(uaddr, fshared, &key, VERIFY_WRITE); if (unlikely(ret != 0)) goto out; @@ -1650,23 +2078,238 @@ out: return ret; pi_faulted: - /* - * We have to r/w *(int __user *)uaddr, and we have to modify it - * atomically. Therefore, if we continue to fault after get_user() - * below, we need to handle the fault ourselves, while still holding - * the mmap_sem. This can occur if the uaddr is under contention as - * we have to drop the mmap_sem in order to call get_user(). - */ spin_unlock(&hb->lock); put_futex_key(fshared, &key); - ret = get_user(uval, uaddr); + ret = fault_in_user_writeable(uaddr); if (!ret) goto retry; return ret; } +/** + * handle_early_requeue_pi_wakeup() - Detect early wakeup on the initial futex + * @hb: the hash_bucket futex_q was original enqueued on + * @q: the futex_q woken while waiting to be requeued + * @key2: the futex_key of the requeue target futex + * @timeout: the timeout associated with the wait (NULL if none) + * + * Detect if the task was woken on the initial futex as opposed to the requeue + * target futex. If so, determine if it was a timeout or a signal that caused + * the wakeup and return the appropriate error code to the caller. Must be + * called with the hb lock held. + * + * Returns + * 0 - no early wakeup detected + * <0 - -ETIMEDOUT or -ERESTARTNOINTR + */ +static inline +int handle_early_requeue_pi_wakeup(struct futex_hash_bucket *hb, + struct futex_q *q, union futex_key *key2, + struct hrtimer_sleeper *timeout) +{ + int ret = 0; + + /* + * With the hb lock held, we avoid races while we process the wakeup. + * We only need to hold hb (and not hb2) to ensure atomicity as the + * wakeup code can't change q.key from uaddr to uaddr2 if we hold hb. + * It can't be requeued from uaddr2 to something else since we don't + * support a PI aware source futex for requeue. + */ + if (!match_futex(&q->key, key2)) { + WARN_ON(q->lock_ptr && (&hb->lock != q->lock_ptr)); + /* + * We were woken prior to requeue by a timeout or a signal. + * Unqueue the futex_q and determine which it was. + */ + plist_del(&q->list, &q->list.plist); + + /* Handle spurious wakeups gracefully */ + ret = -EWOULDBLOCK; + if (timeout && !timeout->task) + ret = -ETIMEDOUT; + else if (signal_pending(current)) + ret = -ERESTARTNOINTR; + } + return ret; +} + +/** + * futex_wait_requeue_pi() - Wait on uaddr and take uaddr2 + * @uaddr: the futex we initially wait on (non-pi) + * @fshared: whether the futexes are shared (1) or not (0). They must be + * the same type, no requeueing from private to shared, etc. + * @val: the expected value of uaddr + * @abs_time: absolute timeout + * @bitset: 32 bit wakeup bitset set by userspace, defaults to all + * @clockrt: whether to use CLOCK_REALTIME (1) or CLOCK_MONOTONIC (0) + * @uaddr2: the pi futex we will take prior to returning to user-space + * + * The caller will wait on uaddr and will be requeued by futex_requeue() to + * uaddr2 which must be PI aware. Normal wakeup will wake on uaddr2 and + * complete the acquisition of the rt_mutex prior to returning to userspace. + * This ensures the rt_mutex maintains an owner when it has waiters; without + * one, the pi logic wouldn't know which task to boost/deboost, if there was a + * need to. + * + * We call schedule in futex_wait_queue_me() when we enqueue and return there + * via the following: + * 1) wakeup on uaddr2 after an atomic lock acquisition by futex_requeue() + * 2) wakeup on uaddr2 after a requeue + * 3) signal + * 4) timeout + * + * If 3, cleanup and return -ERESTARTNOINTR. + * + * If 2, we may then block on trying to take the rt_mutex and return via: + * 5) successful lock + * 6) signal + * 7) timeout + * 8) other lock acquisition failure + * + * If 6, return -EWOULDBLOCK (restarting the syscall would do the same). + * + * If 4 or 7, we cleanup and return with -ETIMEDOUT. + * + * Returns: + * 0 - On success + * <0 - On error + */ +static int futex_wait_requeue_pi(u32 __user *uaddr, int fshared, + u32 val, ktime_t *abs_time, u32 bitset, + int clockrt, u32 __user *uaddr2) +{ + struct hrtimer_sleeper timeout, *to = NULL; + struct rt_mutex_waiter rt_waiter; + struct rt_mutex *pi_mutex = NULL; + struct futex_hash_bucket *hb; + union futex_key key2; + struct futex_q q; + int res, ret; + + if (!bitset) + return -EINVAL; + + if (abs_time) { + to = &timeout; + hrtimer_init_on_stack(&to->timer, clockrt ? CLOCK_REALTIME : + CLOCK_MONOTONIC, HRTIMER_MODE_ABS); + hrtimer_init_sleeper(to, current); + hrtimer_set_expires_range_ns(&to->timer, *abs_time, + current->timer_slack_ns); + } + + /* + * The waiter is allocated on our stack, manipulated by the requeue + * code while we sleep on uaddr. + */ + debug_rt_mutex_init_waiter(&rt_waiter); + rt_waiter.task = NULL; + + key2 = FUTEX_KEY_INIT; + ret = get_futex_key(uaddr2, fshared, &key2, VERIFY_WRITE); + if (unlikely(ret != 0)) + goto out; + + q.pi_state = NULL; + q.bitset = bitset; + q.rt_waiter = &rt_waiter; + q.requeue_pi_key = &key2; + + /* Prepare to wait on uaddr. */ + ret = futex_wait_setup(uaddr, val, fshared, &q, &hb); + if (ret) + goto out_key2; + + /* Queue the futex_q, drop the hb lock, wait for wakeup. */ + futex_wait_queue_me(hb, &q, to); + + spin_lock(&hb->lock); + ret = handle_early_requeue_pi_wakeup(hb, &q, &key2, to); + spin_unlock(&hb->lock); + if (ret) + goto out_put_keys; + + /* + * In order for us to be here, we know our q.key == key2, and since + * we took the hb->lock above, we also know that futex_requeue() has + * completed and we no longer have to concern ourselves with a wakeup + * race with the atomic proxy lock acquition by the requeue code. + */ + + /* Check if the requeue code acquired the second futex for us. */ + if (!q.rt_waiter) { + /* + * Got the lock. We might not be the anticipated owner if we + * did a lock-steal - fix up the PI-state in that case. + */ + if (q.pi_state && (q.pi_state->owner != current)) { + spin_lock(q.lock_ptr); + ret = fixup_pi_state_owner(uaddr2, &q, current, + fshared); + spin_unlock(q.lock_ptr); + } + } else { + /* + * We have been woken up by futex_unlock_pi(), a timeout, or a + * signal. futex_unlock_pi() will not destroy the lock_ptr nor + * the pi_state. + */ + WARN_ON(!&q.pi_state); + pi_mutex = &q.pi_state->pi_mutex; + ret = rt_mutex_finish_proxy_lock(pi_mutex, to, &rt_waiter, 1); + debug_rt_mutex_free_waiter(&rt_waiter); + + spin_lock(q.lock_ptr); + /* + * Fixup the pi_state owner and possibly acquire the lock if we + * haven't already. + */ + res = fixup_owner(uaddr2, fshared, &q, !ret); + /* + * If fixup_owner() returned an error, proprogate that. If it + * acquired the lock, clear -ETIMEDOUT or -EINTR. + */ + if (res) + ret = (res < 0) ? res : 0; + + /* Unqueue and drop the lock. */ + unqueue_me_pi(&q); + } + + /* + * If fixup_pi_state_owner() faulted and was unable to handle the + * fault, unlock the rt_mutex and return the fault to userspace. + */ + if (ret == -EFAULT) { + if (rt_mutex_owner(pi_mutex) == current) + rt_mutex_unlock(pi_mutex); + } else if (ret == -EINTR) { + /* + * We've already been requeued, but cannot restart by calling + * futex_lock_pi() directly. We could restart this syscall, but + * it would detect that the user space "val" changed and return + * -EWOULDBLOCK. Save the overhead of the restart and return + * -EWOULDBLOCK directly. + */ + ret = -EWOULDBLOCK; + } + +out_put_keys: + put_futex_key(fshared, &q.key); +out_key2: + put_futex_key(fshared, &key2); + +out: + if (to) { + hrtimer_cancel(&to->timer); + destroy_hrtimer_on_stack(&to->timer); + } + return ret; +} + /* * Support for robust futexes: the kernel cleans up held futexes at * thread exit time. @@ -1683,9 +2326,9 @@ pi_faulted: */ /** - * sys_set_robust_list - set the robust-futex list head of a task - * @head: pointer to the list-head - * @len: length of the list-head, as userspace expects + * sys_set_robust_list() - Set the robust-futex list head of a task + * @head: pointer to the list-head + * @len: length of the list-head, as userspace expects */ SYSCALL_DEFINE2(set_robust_list, struct robust_list_head __user *, head, size_t, len) @@ -1704,10 +2347,10 @@ SYSCALL_DEFINE2(set_robust_list, struct robust_list_head __user *, head, } /** - * sys_get_robust_list - get the robust-futex list head of a task - * @pid: pid of the process [zero for current task] - * @head_ptr: pointer to a list-head pointer, the kernel fills it in - * @len_ptr: pointer to a length field, the kernel fills in the header size + * sys_get_robust_list() - Get the robust-futex list head of a task + * @pid: pid of the process [zero for current task] + * @head_ptr: pointer to a list-head pointer, the kernel fills it in + * @len_ptr: pointer to a length field, the kernel fills in the header size */ SYSCALL_DEFINE3(get_robust_list, int, pid, struct robust_list_head __user * __user *, head_ptr, @@ -1889,7 +2532,7 @@ long do_futex(u32 __user *uaddr, int op, u32 val, ktime_t *timeout, fshared = 1; clockrt = op & FUTEX_CLOCK_REALTIME; - if (clockrt && cmd != FUTEX_WAIT_BITSET) + if (clockrt && cmd != FUTEX_WAIT_BITSET && cmd != FUTEX_WAIT_REQUEUE_PI) return -ENOSYS; switch (cmd) { @@ -1904,10 +2547,11 @@ long do_futex(u32 __user *uaddr, int op, u32 val, ktime_t *timeout, ret = futex_wake(uaddr, fshared, val, val3); break; case FUTEX_REQUEUE: - ret = futex_requeue(uaddr, fshared, uaddr2, val, val2, NULL); + ret = futex_requeue(uaddr, fshared, uaddr2, val, val2, NULL, 0); break; case FUTEX_CMP_REQUEUE: - ret = futex_requeue(uaddr, fshared, uaddr2, val, val2, &val3); + ret = futex_requeue(uaddr, fshared, uaddr2, val, val2, &val3, + 0); break; case FUTEX_WAKE_OP: ret = futex_wake_op(uaddr, fshared, uaddr2, val, val2, val3); @@ -1924,6 +2568,15 @@ long do_futex(u32 __user *uaddr, int op, u32 val, ktime_t *timeout, if (futex_cmpxchg_enabled) ret = futex_lock_pi(uaddr, fshared, 0, timeout, 1); break; + case FUTEX_WAIT_REQUEUE_PI: + val3 = FUTEX_BITSET_MATCH_ANY; + ret = futex_wait_requeue_pi(uaddr, fshared, val, timeout, val3, + clockrt, uaddr2); + break; + case FUTEX_CMP_REQUEUE_PI: + ret = futex_requeue(uaddr, fshared, uaddr2, val, val2, &val3, + 1); + break; default: ret = -ENOSYS; } @@ -1941,7 +2594,8 @@ SYSCALL_DEFINE6(futex, u32 __user *, uaddr, int, op, u32, val, int cmd = op & FUTEX_CMD_MASK; if (utime && (cmd == FUTEX_WAIT || cmd == FUTEX_LOCK_PI || - cmd == FUTEX_WAIT_BITSET)) { + cmd == FUTEX_WAIT_BITSET || + cmd == FUTEX_WAIT_REQUEUE_PI)) { if (copy_from_user(&ts, utime, sizeof(ts)) != 0) return -EFAULT; if (!timespec_valid(&ts)) @@ -1953,11 +2607,11 @@ SYSCALL_DEFINE6(futex, u32 __user *, uaddr, int, op, u32, val, tp = &t; } /* - * requeue parameter in 'utime' if cmd == FUTEX_REQUEUE. + * requeue parameter in 'utime' if cmd == FUTEX_*_REQUEUE_*. * number of waiters to wake in 'utime' if cmd == FUTEX_WAKE_OP. */ if (cmd == FUTEX_REQUEUE || cmd == FUTEX_CMP_REQUEUE || - cmd == FUTEX_WAKE_OP) + cmd == FUTEX_CMP_REQUEUE_PI || cmd == FUTEX_WAKE_OP) val2 = (u32) (unsigned long) utime; return do_futex(uaddr, op, val, tp, uaddr2, val2, val3); diff --git a/kernel/futex_compat.c b/kernel/futex_compat.c index d607a5b9ee2..235716556bf 100644 --- a/kernel/futex_compat.c +++ b/kernel/futex_compat.c @@ -180,7 +180,8 @@ asmlinkage long compat_sys_futex(u32 __user *uaddr, int op, u32 val, int cmd = op & FUTEX_CMD_MASK; if (utime && (cmd == FUTEX_WAIT || cmd == FUTEX_LOCK_PI || - cmd == FUTEX_WAIT_BITSET)) { + cmd == FUTEX_WAIT_BITSET || + cmd == FUTEX_WAIT_REQUEUE_PI)) { if (get_compat_timespec(&ts, utime)) return -EFAULT; if (!timespec_valid(&ts)) @@ -191,7 +192,8 @@ asmlinkage long compat_sys_futex(u32 __user *uaddr, int op, u32 val, t = ktime_add_safe(ktime_get(), t); tp = &t; } - if (cmd == FUTEX_REQUEUE || cmd == FUTEX_CMP_REQUEUE) + if (cmd == FUTEX_REQUEUE || cmd == FUTEX_CMP_REQUEUE || + cmd == FUTEX_CMP_REQUEUE_PI || cmd == FUTEX_WAKE_OP) val2 = (int) (unsigned long) utime; return do_futex(uaddr, op, val, tp, uaddr2, val2, val3); diff --git a/kernel/gcov/Kconfig b/kernel/gcov/Kconfig new file mode 100644 index 00000000000..70a298d6da7 --- /dev/null +++ b/kernel/gcov/Kconfig @@ -0,0 +1,48 @@ +menu "GCOV-based kernel profiling" + +config GCOV_KERNEL + bool "Enable gcov-based kernel profiling" + depends on DEBUG_FS && CONSTRUCTORS + default n + ---help--- + This option enables gcov-based code profiling (e.g. for code coverage + measurements). + + If unsure, say N. + + Additionally specify CONFIG_GCOV_PROFILE_ALL=y to get profiling data + for the entire kernel. To enable profiling for specific files or + directories, add a line similar to the following to the respective + Makefile: + + For a single file (e.g. main.o): + GCOV_PROFILE_main.o := y + + For all files in one directory: + GCOV_PROFILE := y + + To exclude files from being profiled even when CONFIG_GCOV_PROFILE_ALL + is specified, use: + + GCOV_PROFILE_main.o := n + and: + GCOV_PROFILE := n + + Note that the debugfs filesystem has to be mounted to access + profiling data. + +config GCOV_PROFILE_ALL + bool "Profile entire Kernel" + depends on GCOV_KERNEL + depends on S390 || X86 || (PPC && EXPERIMENTAL) || MICROBLAZE + default n + ---help--- + This options activates profiling for the entire kernel. + + If unsure, say N. + + Note that a kernel compiled with profiling flags will be significantly + larger and run slower. Also be sure to exclude files from profiling + which are not linked to the kernel image to prevent linker errors. + +endmenu diff --git a/kernel/gcov/Makefile b/kernel/gcov/Makefile new file mode 100644 index 00000000000..3f761001d51 --- /dev/null +++ b/kernel/gcov/Makefile @@ -0,0 +1,3 @@ +EXTRA_CFLAGS := -DSRCTREE='"$(srctree)"' -DOBJTREE='"$(objtree)"' + +obj-$(CONFIG_GCOV_KERNEL) := base.o fs.o gcc_3_4.o diff --git a/kernel/gcov/base.c b/kernel/gcov/base.c new file mode 100644 index 00000000000..9b22d03cc58 --- /dev/null +++ b/kernel/gcov/base.c @@ -0,0 +1,148 @@ +/* + * This code maintains a list of active profiling data structures. + * + * Copyright IBM Corp. 2009 + * Author(s): Peter Oberparleiter <oberpar@linux.vnet.ibm.com> + * + * Uses gcc-internal data definitions. + * Based on the gcov-kernel patch by: + * Hubertus Franke <frankeh@us.ibm.com> + * Nigel Hinds <nhinds@us.ibm.com> + * Rajan Ravindran <rajancr@us.ibm.com> + * Peter Oberparleiter <oberpar@linux.vnet.ibm.com> + * Paul Larson + */ + +#define pr_fmt(fmt) "gcov: " fmt + +#include <linux/init.h> +#include <linux/module.h> +#include <linux/mutex.h> +#include "gcov.h" + +static struct gcov_info *gcov_info_head; +static int gcov_events_enabled; +static DEFINE_MUTEX(gcov_lock); + +/* + * __gcov_init is called by gcc-generated constructor code for each object + * file compiled with -fprofile-arcs. + */ +void __gcov_init(struct gcov_info *info) +{ + static unsigned int gcov_version; + + mutex_lock(&gcov_lock); + if (gcov_version == 0) { + gcov_version = info->version; + /* + * Printing gcc's version magic may prove useful for debugging + * incompatibility reports. + */ + pr_info("version magic: 0x%x\n", gcov_version); + } + /* + * Add new profiling data structure to list and inform event + * listener. + */ + info->next = gcov_info_head; + gcov_info_head = info; + if (gcov_events_enabled) + gcov_event(GCOV_ADD, info); + mutex_unlock(&gcov_lock); +} +EXPORT_SYMBOL(__gcov_init); + +/* + * These functions may be referenced by gcc-generated profiling code but serve + * no function for kernel profiling. + */ +void __gcov_flush(void) +{ + /* Unused. */ +} +EXPORT_SYMBOL(__gcov_flush); + +void __gcov_merge_add(gcov_type *counters, unsigned int n_counters) +{ + /* Unused. */ +} +EXPORT_SYMBOL(__gcov_merge_add); + +void __gcov_merge_single(gcov_type *counters, unsigned int n_counters) +{ + /* Unused. */ +} +EXPORT_SYMBOL(__gcov_merge_single); + +void __gcov_merge_delta(gcov_type *counters, unsigned int n_counters) +{ + /* Unused. */ +} +EXPORT_SYMBOL(__gcov_merge_delta); + +/** + * gcov_enable_events - enable event reporting through gcov_event() + * + * Turn on reporting of profiling data load/unload-events through the + * gcov_event() callback. Also replay all previous events once. This function + * is needed because some events are potentially generated too early for the + * callback implementation to handle them initially. + */ +void gcov_enable_events(void) +{ + struct gcov_info *info; + + mutex_lock(&gcov_lock); + gcov_events_enabled = 1; + /* Perform event callback for previously registered entries. */ + for (info = gcov_info_head; info; info = info->next) + gcov_event(GCOV_ADD, info); + mutex_unlock(&gcov_lock); +} + +#ifdef CONFIG_MODULES +static inline int within(void *addr, void *start, unsigned long size) +{ + return ((addr >= start) && (addr < start + size)); +} + +/* Update list and generate events when modules are unloaded. */ +static int gcov_module_notifier(struct notifier_block *nb, unsigned long event, + void *data) +{ + struct module *mod = data; + struct gcov_info *info; + struct gcov_info *prev; + + if (event != MODULE_STATE_GOING) + return NOTIFY_OK; + mutex_lock(&gcov_lock); + prev = NULL; + /* Remove entries located in module from linked list. */ + for (info = gcov_info_head; info; info = info->next) { + if (within(info, mod->module_core, mod->core_size)) { + if (prev) + prev->next = info->next; + else + gcov_info_head = info->next; + if (gcov_events_enabled) + gcov_event(GCOV_REMOVE, info); + } else + prev = info; + } + mutex_unlock(&gcov_lock); + + return NOTIFY_OK; +} + +static struct notifier_block gcov_nb = { + .notifier_call = gcov_module_notifier, +}; + +static int __init gcov_init(void) +{ + return register_module_notifier(&gcov_nb); +} +device_initcall(gcov_init); +#endif /* CONFIG_MODULES */ diff --git a/kernel/gcov/fs.c b/kernel/gcov/fs.c new file mode 100644 index 00000000000..ef3c3f88a7a --- /dev/null +++ b/kernel/gcov/fs.c @@ -0,0 +1,673 @@ +/* + * This code exports profiling data as debugfs files to userspace. + * + * Copyright IBM Corp. 2009 + * Author(s): Peter Oberparleiter <oberpar@linux.vnet.ibm.com> + * + * Uses gcc-internal data definitions. + * Based on the gcov-kernel patch by: + * Hubertus Franke <frankeh@us.ibm.com> + * Nigel Hinds <nhinds@us.ibm.com> + * Rajan Ravindran <rajancr@us.ibm.com> + * Peter Oberparleiter <oberpar@linux.vnet.ibm.com> + * Paul Larson + * Yi CDL Yang + */ + +#define pr_fmt(fmt) "gcov: " fmt + +#include <linux/init.h> +#include <linux/module.h> +#include <linux/debugfs.h> +#include <linux/fs.h> +#include <linux/list.h> +#include <linux/string.h> +#include <linux/slab.h> +#include <linux/mutex.h> +#include <linux/seq_file.h> +#include "gcov.h" + +/** + * struct gcov_node - represents a debugfs entry + * @list: list head for child node list + * @children: child nodes + * @all: list head for list of all nodes + * @parent: parent node + * @info: associated profiling data structure if not a directory + * @ghost: when an object file containing profiling data is unloaded we keep a + * copy of the profiling data here to allow collecting coverage data + * for cleanup code. Such a node is called a "ghost". + * @dentry: main debugfs entry, either a directory or data file + * @links: associated symbolic links + * @name: data file basename + * + * struct gcov_node represents an entity within the gcov/ subdirectory + * of debugfs. There are directory and data file nodes. The latter represent + * the actual synthesized data file plus any associated symbolic links which + * are needed by the gcov tool to work correctly. + */ +struct gcov_node { + struct list_head list; + struct list_head children; + struct list_head all; + struct gcov_node *parent; + struct gcov_info *info; + struct gcov_info *ghost; + struct dentry *dentry; + struct dentry **links; + char name[0]; +}; + +static const char objtree[] = OBJTREE; +static const char srctree[] = SRCTREE; +static struct gcov_node root_node; +static struct dentry *reset_dentry; +static LIST_HEAD(all_head); +static DEFINE_MUTEX(node_lock); + +/* If non-zero, keep copies of profiling data for unloaded modules. */ +static int gcov_persist = 1; + +static int __init gcov_persist_setup(char *str) +{ + unsigned long val; + + if (strict_strtoul(str, 0, &val)) { + pr_warning("invalid gcov_persist parameter '%s'\n", str); + return 0; + } + gcov_persist = val; + pr_info("setting gcov_persist to %d\n", gcov_persist); + + return 1; +} +__setup("gcov_persist=", gcov_persist_setup); + +/* + * seq_file.start() implementation for gcov data files. Note that the + * gcov_iterator interface is designed to be more restrictive than seq_file + * (no start from arbitrary position, etc.), to simplify the iterator + * implementation. + */ +static void *gcov_seq_start(struct seq_file *seq, loff_t *pos) +{ + loff_t i; + + gcov_iter_start(seq->private); + for (i = 0; i < *pos; i++) { + if (gcov_iter_next(seq->private)) + return NULL; + } + return seq->private; +} + +/* seq_file.next() implementation for gcov data files. */ +static void *gcov_seq_next(struct seq_file *seq, void *data, loff_t *pos) +{ + struct gcov_iterator *iter = data; + + if (gcov_iter_next(iter)) + return NULL; + (*pos)++; + + return iter; +} + +/* seq_file.show() implementation for gcov data files. */ +static int gcov_seq_show(struct seq_file *seq, void *data) +{ + struct gcov_iterator *iter = data; + + if (gcov_iter_write(iter, seq)) + return -EINVAL; + return 0; +} + +static void gcov_seq_stop(struct seq_file *seq, void *data) +{ + /* Unused. */ +} + +static const struct seq_operations gcov_seq_ops = { + .start = gcov_seq_start, + .next = gcov_seq_next, + .show = gcov_seq_show, + .stop = gcov_seq_stop, +}; + +/* + * Return the profiling data set for a given node. This can either be the + * original profiling data structure or a duplicate (also called "ghost") + * in case the associated object file has been unloaded. + */ +static struct gcov_info *get_node_info(struct gcov_node *node) +{ + if (node->info) + return node->info; + + return node->ghost; +} + +/* + * open() implementation for gcov data files. Create a copy of the profiling + * data set and initialize the iterator and seq_file interface. + */ +static int gcov_seq_open(struct inode *inode, struct file *file) +{ + struct gcov_node *node = inode->i_private; + struct gcov_iterator *iter; + struct seq_file *seq; + struct gcov_info *info; + int rc = -ENOMEM; + + mutex_lock(&node_lock); + /* + * Read from a profiling data copy to minimize reference tracking + * complexity and concurrent access. + */ + info = gcov_info_dup(get_node_info(node)); + if (!info) + goto out_unlock; + iter = gcov_iter_new(info); + if (!iter) + goto err_free_info; + rc = seq_open(file, &gcov_seq_ops); + if (rc) + goto err_free_iter_info; + seq = file->private_data; + seq->private = iter; +out_unlock: + mutex_unlock(&node_lock); + return rc; + +err_free_iter_info: + gcov_iter_free(iter); +err_free_info: + gcov_info_free(info); + goto out_unlock; +} + +/* + * release() implementation for gcov data files. Release resources allocated + * by open(). + */ +static int gcov_seq_release(struct inode *inode, struct file *file) +{ + struct gcov_iterator *iter; + struct gcov_info *info; + struct seq_file *seq; + + seq = file->private_data; + iter = seq->private; + info = gcov_iter_get_info(iter); + gcov_iter_free(iter); + gcov_info_free(info); + seq_release(inode, file); + + return 0; +} + +/* + * Find a node by the associated data file name. Needs to be called with + * node_lock held. + */ +static struct gcov_node *get_node_by_name(const char *name) +{ + struct gcov_node *node; + struct gcov_info *info; + + list_for_each_entry(node, &all_head, all) { + info = get_node_info(node); + if (info && (strcmp(info->filename, name) == 0)) + return node; + } + + return NULL; +} + +static void remove_node(struct gcov_node *node); + +/* + * write() implementation for gcov data files. Reset profiling data for the + * associated file. If the object file has been unloaded (i.e. this is + * a "ghost" node), remove the debug fs node as well. + */ +static ssize_t gcov_seq_write(struct file *file, const char __user *addr, + size_t len, loff_t *pos) +{ + struct seq_file *seq; + struct gcov_info *info; + struct gcov_node *node; + + seq = file->private_data; + info = gcov_iter_get_info(seq->private); + mutex_lock(&node_lock); + node = get_node_by_name(info->filename); + if (node) { + /* Reset counts or remove node for unloaded modules. */ + if (node->ghost) + remove_node(node); + else + gcov_info_reset(node->info); + } + /* Reset counts for open file. */ + gcov_info_reset(info); + mutex_unlock(&node_lock); + + return len; +} + +/* + * Given a string <path> representing a file path of format: + * path/to/file.gcda + * construct and return a new string: + * <dir/>path/to/file.<ext> + */ +static char *link_target(const char *dir, const char *path, const char *ext) +{ + char *target; + char *old_ext; + char *copy; + + copy = kstrdup(path, GFP_KERNEL); + if (!copy) + return NULL; + old_ext = strrchr(copy, '.'); + if (old_ext) + *old_ext = '\0'; + if (dir) + target = kasprintf(GFP_KERNEL, "%s/%s.%s", dir, copy, ext); + else + target = kasprintf(GFP_KERNEL, "%s.%s", copy, ext); + kfree(copy); + + return target; +} + +/* + * Construct a string representing the symbolic link target for the given + * gcov data file name and link type. Depending on the link type and the + * location of the data file, the link target can either point to a + * subdirectory of srctree, objtree or in an external location. + */ +static char *get_link_target(const char *filename, const struct gcov_link *ext) +{ + const char *rel; + char *result; + + if (strncmp(filename, objtree, strlen(objtree)) == 0) { + rel = filename + strlen(objtree) + 1; + if (ext->dir == SRC_TREE) + result = link_target(srctree, rel, ext->ext); + else + result = link_target(objtree, rel, ext->ext); + } else { + /* External compilation. */ + result = link_target(NULL, filename, ext->ext); + } + + return result; +} + +#define SKEW_PREFIX ".tmp_" + +/* + * For a filename .tmp_filename.ext return filename.ext. Needed to compensate + * for filename skewing caused by the mod-versioning mechanism. + */ +static const char *deskew(const char *basename) +{ + if (strncmp(basename, SKEW_PREFIX, sizeof(SKEW_PREFIX) - 1) == 0) + return basename + sizeof(SKEW_PREFIX) - 1; + return basename; +} + +/* + * Create links to additional files (usually .c and .gcno files) which the + * gcov tool expects to find in the same directory as the gcov data file. + */ +static void add_links(struct gcov_node *node, struct dentry *parent) +{ + char *basename; + char *target; + int num; + int i; + + for (num = 0; gcov_link[num].ext; num++) + /* Nothing. */; + node->links = kcalloc(num, sizeof(struct dentry *), GFP_KERNEL); + if (!node->links) + return; + for (i = 0; i < num; i++) { + target = get_link_target(get_node_info(node)->filename, + &gcov_link[i]); + if (!target) + goto out_err; + basename = strrchr(target, '/'); + if (!basename) + goto out_err; + basename++; + node->links[i] = debugfs_create_symlink(deskew(basename), + parent, target); + if (!node->links[i]) + goto out_err; + kfree(target); + } + + return; +out_err: + kfree(target); + while (i-- > 0) + debugfs_remove(node->links[i]); + kfree(node->links); + node->links = NULL; +} + +static const struct file_operations gcov_data_fops = { + .open = gcov_seq_open, + .release = gcov_seq_release, + .read = seq_read, + .llseek = seq_lseek, + .write = gcov_seq_write, +}; + +/* Basic initialization of a new node. */ +static void init_node(struct gcov_node *node, struct gcov_info *info, + const char *name, struct gcov_node *parent) +{ + INIT_LIST_HEAD(&node->list); + INIT_LIST_HEAD(&node->children); + INIT_LIST_HEAD(&node->all); + node->info = info; + node->parent = parent; + if (name) + strcpy(node->name, name); +} + +/* + * Create a new node and associated debugfs entry. Needs to be called with + * node_lock held. + */ +static struct gcov_node *new_node(struct gcov_node *parent, + struct gcov_info *info, const char *name) +{ + struct gcov_node *node; + + node = kzalloc(sizeof(struct gcov_node) + strlen(name) + 1, GFP_KERNEL); + if (!node) { + pr_warning("out of memory\n"); + return NULL; + } + init_node(node, info, name, parent); + /* Differentiate between gcov data file nodes and directory nodes. */ + if (info) { + node->dentry = debugfs_create_file(deskew(node->name), 0600, + parent->dentry, node, &gcov_data_fops); + } else + node->dentry = debugfs_create_dir(node->name, parent->dentry); + if (!node->dentry) { + pr_warning("could not create file\n"); + kfree(node); + return NULL; + } + if (info) + add_links(node, parent->dentry); + list_add(&node->list, &parent->children); + list_add(&node->all, &all_head); + + return node; +} + +/* Remove symbolic links associated with node. */ +static void remove_links(struct gcov_node *node) +{ + int i; + + if (!node->links) + return; + for (i = 0; gcov_link[i].ext; i++) + debugfs_remove(node->links[i]); + kfree(node->links); + node->links = NULL; +} + +/* + * Remove node from all lists and debugfs and release associated resources. + * Needs to be called with node_lock held. + */ +static void release_node(struct gcov_node *node) +{ + list_del(&node->list); + list_del(&node->all); + debugfs_remove(node->dentry); + remove_links(node); + if (node->ghost) + gcov_info_free(node->ghost); + kfree(node); +} + +/* Release node and empty parents. Needs to be called with node_lock held. */ +static void remove_node(struct gcov_node *node) +{ + struct gcov_node *parent; + + while ((node != &root_node) && list_empty(&node->children)) { + parent = node->parent; + release_node(node); + node = parent; + } +} + +/* + * Find child node with given basename. Needs to be called with node_lock + * held. + */ +static struct gcov_node *get_child_by_name(struct gcov_node *parent, + const char *name) +{ + struct gcov_node *node; + + list_for_each_entry(node, &parent->children, list) { + if (strcmp(node->name, name) == 0) + return node; + } + + return NULL; +} + +/* + * write() implementation for reset file. Reset all profiling data to zero + * and remove ghost nodes. + */ +static ssize_t reset_write(struct file *file, const char __user *addr, + size_t len, loff_t *pos) +{ + struct gcov_node *node; + + mutex_lock(&node_lock); +restart: + list_for_each_entry(node, &all_head, all) { + if (node->info) + gcov_info_reset(node->info); + else if (list_empty(&node->children)) { + remove_node(node); + /* Several nodes may have gone - restart loop. */ + goto restart; + } + } + mutex_unlock(&node_lock); + + return len; +} + +/* read() implementation for reset file. Unused. */ +static ssize_t reset_read(struct file *file, char __user *addr, size_t len, + loff_t *pos) +{ + /* Allow read operation so that a recursive copy won't fail. */ + return 0; +} + +static const struct file_operations gcov_reset_fops = { + .write = reset_write, + .read = reset_read, +}; + +/* + * Create a node for a given profiling data set and add it to all lists and + * debugfs. Needs to be called with node_lock held. + */ +static void add_node(struct gcov_info *info) +{ + char *filename; + char *curr; + char *next; + struct gcov_node *parent; + struct gcov_node *node; + + filename = kstrdup(info->filename, GFP_KERNEL); + if (!filename) + return; + parent = &root_node; + /* Create directory nodes along the path. */ + for (curr = filename; (next = strchr(curr, '/')); curr = next + 1) { + if (curr == next) + continue; + *next = 0; + if (strcmp(curr, ".") == 0) + continue; + if (strcmp(curr, "..") == 0) { + if (!parent->parent) + goto err_remove; + parent = parent->parent; + continue; + } + node = get_child_by_name(parent, curr); + if (!node) { + node = new_node(parent, NULL, curr); + if (!node) + goto err_remove; + } + parent = node; + } + /* Create file node. */ + node = new_node(parent, info, curr); + if (!node) + goto err_remove; +out: + kfree(filename); + return; + +err_remove: + remove_node(parent); + goto out; +} + +/* + * The profiling data set associated with this node is being unloaded. Store a + * copy of the profiling data and turn this node into a "ghost". + */ +static int ghost_node(struct gcov_node *node) +{ + node->ghost = gcov_info_dup(node->info); + if (!node->ghost) { + pr_warning("could not save data for '%s' (out of memory)\n", + node->info->filename); + return -ENOMEM; + } + node->info = NULL; + + return 0; +} + +/* + * Profiling data for this node has been loaded again. Add profiling data + * from previous instantiation and turn this node into a regular node. + */ +static void revive_node(struct gcov_node *node, struct gcov_info *info) +{ + if (gcov_info_is_compatible(node->ghost, info)) + gcov_info_add(info, node->ghost); + else { + pr_warning("discarding saved data for '%s' (version changed)\n", + info->filename); + } + gcov_info_free(node->ghost); + node->ghost = NULL; + node->info = info; +} + +/* + * Callback to create/remove profiling files when code compiled with + * -fprofile-arcs is loaded/unloaded. + */ +void gcov_event(enum gcov_action action, struct gcov_info *info) +{ + struct gcov_node *node; + + mutex_lock(&node_lock); + node = get_node_by_name(info->filename); + switch (action) { + case GCOV_ADD: + /* Add new node or revive ghost. */ + if (!node) { + add_node(info); + break; + } + if (gcov_persist) + revive_node(node, info); + else { + pr_warning("could not add '%s' (already exists)\n", + info->filename); + } + break; + case GCOV_REMOVE: + /* Remove node or turn into ghost. */ + if (!node) { + pr_warning("could not remove '%s' (not found)\n", + info->filename); + break; + } + if (gcov_persist) { + if (!ghost_node(node)) + break; + } + remove_node(node); + break; + } + mutex_unlock(&node_lock); +} + +/* Create debugfs entries. */ +static __init int gcov_fs_init(void) +{ + int rc = -EIO; + + init_node(&root_node, NULL, NULL, NULL); + /* + * /sys/kernel/debug/gcov will be parent for the reset control file + * and all profiling files. + */ + root_node.dentry = debugfs_create_dir("gcov", NULL); + if (!root_node.dentry) + goto err_remove; + /* + * Create reset file which resets all profiling counts when written + * to. + */ + reset_dentry = debugfs_create_file("reset", 0600, root_node.dentry, + NULL, &gcov_reset_fops); + if (!reset_dentry) + goto err_remove; + /* Replay previous events to get our fs hierarchy up-to-date. */ + gcov_enable_events(); + return 0; + +err_remove: + pr_err("init failed\n"); + if (root_node.dentry) + debugfs_remove(root_node.dentry); + + return rc; +} +device_initcall(gcov_fs_init); diff --git a/kernel/gcov/gcc_3_4.c b/kernel/gcov/gcc_3_4.c new file mode 100644 index 00000000000..ae5bb426003 --- /dev/null +++ b/kernel/gcov/gcc_3_4.c @@ -0,0 +1,447 @@ +/* + * This code provides functions to handle gcc's profiling data format + * introduced with gcc 3.4. Future versions of gcc may change the gcov + * format (as happened before), so all format-specific information needs + * to be kept modular and easily exchangeable. + * + * This file is based on gcc-internal definitions. Functions and data + * structures are defined to be compatible with gcc counterparts. + * For a better understanding, refer to gcc source: gcc/gcov-io.h. + * + * Copyright IBM Corp. 2009 + * Author(s): Peter Oberparleiter <oberpar@linux.vnet.ibm.com> + * + * Uses gcc-internal data definitions. + */ + +#include <linux/errno.h> +#include <linux/slab.h> +#include <linux/string.h> +#include <linux/seq_file.h> +#include <linux/vmalloc.h> +#include "gcov.h" + +/* Symbolic links to be created for each profiling data file. */ +const struct gcov_link gcov_link[] = { + { OBJ_TREE, "gcno" }, /* Link to .gcno file in $(objtree). */ + { 0, NULL}, +}; + +/* + * Determine whether a counter is active. Based on gcc magic. Doesn't change + * at run-time. + */ +static int counter_active(struct gcov_info *info, unsigned int type) +{ + return (1 << type) & info->ctr_mask; +} + +/* Determine number of active counters. Based on gcc magic. */ +static unsigned int num_counter_active(struct gcov_info *info) +{ + unsigned int i; + unsigned int result = 0; + + for (i = 0; i < GCOV_COUNTERS; i++) { + if (counter_active(info, i)) + result++; + } + return result; +} + +/** + * gcov_info_reset - reset profiling data to zero + * @info: profiling data set + */ +void gcov_info_reset(struct gcov_info *info) +{ + unsigned int active = num_counter_active(info); + unsigned int i; + + for (i = 0; i < active; i++) { + memset(info->counts[i].values, 0, + info->counts[i].num * sizeof(gcov_type)); + } +} + +/** + * gcov_info_is_compatible - check if profiling data can be added + * @info1: first profiling data set + * @info2: second profiling data set + * + * Returns non-zero if profiling data can be added, zero otherwise. + */ +int gcov_info_is_compatible(struct gcov_info *info1, struct gcov_info *info2) +{ + return (info1->stamp == info2->stamp); +} + +/** + * gcov_info_add - add up profiling data + * @dest: profiling data set to which data is added + * @source: profiling data set which is added + * + * Adds profiling counts of @source to @dest. + */ +void gcov_info_add(struct gcov_info *dest, struct gcov_info *source) +{ + unsigned int i; + unsigned int j; + + for (i = 0; i < num_counter_active(dest); i++) { + for (j = 0; j < dest->counts[i].num; j++) { + dest->counts[i].values[j] += + source->counts[i].values[j]; + } + } +} + +/* Get size of function info entry. Based on gcc magic. */ +static size_t get_fn_size(struct gcov_info *info) +{ + size_t size; + + size = sizeof(struct gcov_fn_info) + num_counter_active(info) * + sizeof(unsigned int); + if (__alignof__(struct gcov_fn_info) > sizeof(unsigned int)) + size = ALIGN(size, __alignof__(struct gcov_fn_info)); + return size; +} + +/* Get address of function info entry. Based on gcc magic. */ +static struct gcov_fn_info *get_fn_info(struct gcov_info *info, unsigned int fn) +{ + return (struct gcov_fn_info *) + ((char *) info->functions + fn * get_fn_size(info)); +} + +/** + * gcov_info_dup - duplicate profiling data set + * @info: profiling data set to duplicate + * + * Return newly allocated duplicate on success, %NULL on error. + */ +struct gcov_info *gcov_info_dup(struct gcov_info *info) +{ + struct gcov_info *dup; + unsigned int i; + unsigned int active; + + /* Duplicate gcov_info. */ + active = num_counter_active(info); + dup = kzalloc(sizeof(struct gcov_info) + + sizeof(struct gcov_ctr_info) * active, GFP_KERNEL); + if (!dup) + return NULL; + dup->version = info->version; + dup->stamp = info->stamp; + dup->n_functions = info->n_functions; + dup->ctr_mask = info->ctr_mask; + /* Duplicate filename. */ + dup->filename = kstrdup(info->filename, GFP_KERNEL); + if (!dup->filename) + goto err_free; + /* Duplicate table of functions. */ + dup->functions = kmemdup(info->functions, info->n_functions * + get_fn_size(info), GFP_KERNEL); + if (!dup->functions) + goto err_free; + /* Duplicate counter arrays. */ + for (i = 0; i < active ; i++) { + struct gcov_ctr_info *ctr = &info->counts[i]; + size_t size = ctr->num * sizeof(gcov_type); + + dup->counts[i].num = ctr->num; + dup->counts[i].merge = ctr->merge; + dup->counts[i].values = vmalloc(size); + if (!dup->counts[i].values) + goto err_free; + memcpy(dup->counts[i].values, ctr->values, size); + } + return dup; + +err_free: + gcov_info_free(dup); + return NULL; +} + +/** + * gcov_info_free - release memory for profiling data set duplicate + * @info: profiling data set duplicate to free + */ +void gcov_info_free(struct gcov_info *info) +{ + unsigned int active = num_counter_active(info); + unsigned int i; + + for (i = 0; i < active ; i++) + vfree(info->counts[i].values); + kfree(info->functions); + kfree(info->filename); + kfree(info); +} + +/** + * struct type_info - iterator helper array + * @ctr_type: counter type + * @offset: index of the first value of the current function for this type + * + * This array is needed to convert the in-memory data format into the in-file + * data format: + * + * In-memory: + * for each counter type + * for each function + * values + * + * In-file: + * for each function + * for each counter type + * values + * + * See gcc source gcc/gcov-io.h for more information on data organization. + */ +struct type_info { + int ctr_type; + unsigned int offset; +}; + +/** + * struct gcov_iterator - specifies current file position in logical records + * @info: associated profiling data + * @record: record type + * @function: function number + * @type: counter type + * @count: index into values array + * @num_types: number of counter types + * @type_info: helper array to get values-array offset for current function + */ +struct gcov_iterator { + struct gcov_info *info; + + int record; + unsigned int function; + unsigned int type; + unsigned int count; + + int num_types; + struct type_info type_info[0]; +}; + +static struct gcov_fn_info *get_func(struct gcov_iterator *iter) +{ + return get_fn_info(iter->info, iter->function); +} + +static struct type_info *get_type(struct gcov_iterator *iter) +{ + return &iter->type_info[iter->type]; +} + +/** + * gcov_iter_new - allocate and initialize profiling data iterator + * @info: profiling data set to be iterated + * + * Return file iterator on success, %NULL otherwise. + */ +struct gcov_iterator *gcov_iter_new(struct gcov_info *info) +{ + struct gcov_iterator *iter; + + iter = kzalloc(sizeof(struct gcov_iterator) + + num_counter_active(info) * sizeof(struct type_info), + GFP_KERNEL); + if (iter) + iter->info = info; + + return iter; +} + +/** + * gcov_iter_free - release memory for iterator + * @iter: file iterator to free + */ +void gcov_iter_free(struct gcov_iterator *iter) +{ + kfree(iter); +} + +/** + * gcov_iter_get_info - return profiling data set for given file iterator + * @iter: file iterator + */ +struct gcov_info *gcov_iter_get_info(struct gcov_iterator *iter) +{ + return iter->info; +} + +/** + * gcov_iter_start - reset file iterator to starting position + * @iter: file iterator + */ +void gcov_iter_start(struct gcov_iterator *iter) +{ + int i; + + iter->record = 0; + iter->function = 0; + iter->type = 0; + iter->count = 0; + iter->num_types = 0; + for (i = 0; i < GCOV_COUNTERS; i++) { + if (counter_active(iter->info, i)) { + iter->type_info[iter->num_types].ctr_type = i; + iter->type_info[iter->num_types++].offset = 0; + } + } +} + +/* Mapping of logical record number to actual file content. */ +#define RECORD_FILE_MAGIC 0 +#define RECORD_GCOV_VERSION 1 +#define RECORD_TIME_STAMP 2 +#define RECORD_FUNCTION_TAG 3 +#define RECORD_FUNCTON_TAG_LEN 4 +#define RECORD_FUNCTION_IDENT 5 +#define RECORD_FUNCTION_CHECK 6 +#define RECORD_COUNT_TAG 7 +#define RECORD_COUNT_LEN 8 +#define RECORD_COUNT 9 + +/** + * gcov_iter_next - advance file iterator to next logical record + * @iter: file iterator + * + * Return zero if new position is valid, non-zero if iterator has reached end. + */ +int gcov_iter_next(struct gcov_iterator *iter) +{ + switch (iter->record) { + case RECORD_FILE_MAGIC: + case RECORD_GCOV_VERSION: + case RECORD_FUNCTION_TAG: + case RECORD_FUNCTON_TAG_LEN: + case RECORD_FUNCTION_IDENT: + case RECORD_COUNT_TAG: + /* Advance to next record */ + iter->record++; + break; + case RECORD_COUNT: + /* Advance to next count */ + iter->count++; + /* fall through */ + case RECORD_COUNT_LEN: + if (iter->count < get_func(iter)->n_ctrs[iter->type]) { + iter->record = 9; + break; + } + /* Advance to next counter type */ + get_type(iter)->offset += iter->count; + iter->count = 0; + iter->type++; + /* fall through */ + case RECORD_FUNCTION_CHECK: + if (iter->type < iter->num_types) { + iter->record = 7; + break; + } + /* Advance to next function */ + iter->type = 0; + iter->function++; + /* fall through */ + case RECORD_TIME_STAMP: + if (iter->function < iter->info->n_functions) + iter->record = 3; + else + iter->record = -1; + break; + } + /* Check for EOF. */ + if (iter->record == -1) + return -EINVAL; + else + return 0; +} + +/** + * seq_write_gcov_u32 - write 32 bit number in gcov format to seq_file + * @seq: seq_file handle + * @v: value to be stored + * + * Number format defined by gcc: numbers are recorded in the 32 bit + * unsigned binary form of the endianness of the machine generating the + * file. + */ +static int seq_write_gcov_u32(struct seq_file *seq, u32 v) +{ + return seq_write(seq, &v, sizeof(v)); +} + +/** + * seq_write_gcov_u64 - write 64 bit number in gcov format to seq_file + * @seq: seq_file handle + * @v: value to be stored + * + * Number format defined by gcc: numbers are recorded in the 32 bit + * unsigned binary form of the endianness of the machine generating the + * file. 64 bit numbers are stored as two 32 bit numbers, the low part + * first. + */ +static int seq_write_gcov_u64(struct seq_file *seq, u64 v) +{ + u32 data[2]; + + data[0] = (v & 0xffffffffUL); + data[1] = (v >> 32); + return seq_write(seq, data, sizeof(data)); +} + +/** + * gcov_iter_write - write data for current pos to seq_file + * @iter: file iterator + * @seq: seq_file handle + * + * Return zero on success, non-zero otherwise. + */ +int gcov_iter_write(struct gcov_iterator *iter, struct seq_file *seq) +{ + int rc = -EINVAL; + + switch (iter->record) { + case RECORD_FILE_MAGIC: + rc = seq_write_gcov_u32(seq, GCOV_DATA_MAGIC); + break; + case RECORD_GCOV_VERSION: + rc = seq_write_gcov_u32(seq, iter->info->version); + break; + case RECORD_TIME_STAMP: + rc = seq_write_gcov_u32(seq, iter->info->stamp); + break; + case RECORD_FUNCTION_TAG: + rc = seq_write_gcov_u32(seq, GCOV_TAG_FUNCTION); + break; + case RECORD_FUNCTON_TAG_LEN: + rc = seq_write_gcov_u32(seq, 2); + break; + case RECORD_FUNCTION_IDENT: + rc = seq_write_gcov_u32(seq, get_func(iter)->ident); + break; + case RECORD_FUNCTION_CHECK: + rc = seq_write_gcov_u32(seq, get_func(iter)->checksum); + break; + case RECORD_COUNT_TAG: + rc = seq_write_gcov_u32(seq, + GCOV_TAG_FOR_COUNTER(get_type(iter)->ctr_type)); + break; + case RECORD_COUNT_LEN: + rc = seq_write_gcov_u32(seq, + get_func(iter)->n_ctrs[iter->type] * 2); + break; + case RECORD_COUNT: + rc = seq_write_gcov_u64(seq, + iter->info->counts[iter->type]. + values[iter->count + get_type(iter)->offset]); + break; + } + return rc; +} diff --git a/kernel/gcov/gcov.h b/kernel/gcov/gcov.h new file mode 100644 index 00000000000..060073ebf7a --- /dev/null +++ b/kernel/gcov/gcov.h @@ -0,0 +1,128 @@ +/* + * Profiling infrastructure declarations. + * + * This file is based on gcc-internal definitions. Data structures are + * defined to be compatible with gcc counterparts. For a better + * understanding, refer to gcc source: gcc/gcov-io.h. + * + * Copyright IBM Corp. 2009 + * Author(s): Peter Oberparleiter <oberpar@linux.vnet.ibm.com> + * + * Uses gcc-internal data definitions. + */ + +#ifndef GCOV_H +#define GCOV_H GCOV_H + +#include <linux/types.h> + +/* + * Profiling data types used for gcc 3.4 and above - these are defined by + * gcc and need to be kept as close to the original definition as possible to + * remain compatible. + */ +#define GCOV_COUNTERS 5 +#define GCOV_DATA_MAGIC ((unsigned int) 0x67636461) +#define GCOV_TAG_FUNCTION ((unsigned int) 0x01000000) +#define GCOV_TAG_COUNTER_BASE ((unsigned int) 0x01a10000) +#define GCOV_TAG_FOR_COUNTER(count) \ + (GCOV_TAG_COUNTER_BASE + ((unsigned int) (count) << 17)) + +#if BITS_PER_LONG >= 64 +typedef long gcov_type; +#else +typedef long long gcov_type; +#endif + +/** + * struct gcov_fn_info - profiling meta data per function + * @ident: object file-unique function identifier + * @checksum: function checksum + * @n_ctrs: number of values per counter type belonging to this function + * + * This data is generated by gcc during compilation and doesn't change + * at run-time. + */ +struct gcov_fn_info { + unsigned int ident; + unsigned int checksum; + unsigned int n_ctrs[0]; +}; + +/** + * struct gcov_ctr_info - profiling data per counter type + * @num: number of counter values for this type + * @values: array of counter values for this type + * @merge: merge function for counter values of this type (unused) + * + * This data is generated by gcc during compilation and doesn't change + * at run-time with the exception of the values array. + */ +struct gcov_ctr_info { + unsigned int num; + gcov_type *values; + void (*merge)(gcov_type *, unsigned int); +}; + +/** + * struct gcov_info - profiling data per object file + * @version: gcov version magic indicating the gcc version used for compilation + * @next: list head for a singly-linked list + * @stamp: time stamp + * @filename: name of the associated gcov data file + * @n_functions: number of instrumented functions + * @functions: function data + * @ctr_mask: mask specifying which counter types are active + * @counts: counter data per counter type + * + * This data is generated by gcc during compilation and doesn't change + * at run-time with the exception of the next pointer. + */ +struct gcov_info { + unsigned int version; + struct gcov_info *next; + unsigned int stamp; + const char *filename; + unsigned int n_functions; + const struct gcov_fn_info *functions; + unsigned int ctr_mask; + struct gcov_ctr_info counts[0]; +}; + +/* Base interface. */ +enum gcov_action { + GCOV_ADD, + GCOV_REMOVE, +}; + +void gcov_event(enum gcov_action action, struct gcov_info *info); +void gcov_enable_events(void); + +/* Iterator control. */ +struct seq_file; +struct gcov_iterator; + +struct gcov_iterator *gcov_iter_new(struct gcov_info *info); +void gcov_iter_free(struct gcov_iterator *iter); +void gcov_iter_start(struct gcov_iterator *iter); +int gcov_iter_next(struct gcov_iterator *iter); +int gcov_iter_write(struct gcov_iterator *iter, struct seq_file *seq); +struct gcov_info *gcov_iter_get_info(struct gcov_iterator *iter); + +/* gcov_info control. */ +void gcov_info_reset(struct gcov_info *info); +int gcov_info_is_compatible(struct gcov_info *info1, struct gcov_info *info2); +void gcov_info_add(struct gcov_info *dest, struct gcov_info *source); +struct gcov_info *gcov_info_dup(struct gcov_info *info); +void gcov_info_free(struct gcov_info *info); + +struct gcov_link { + enum { + OBJ_TREE, + SRC_TREE, + } dir; + const char *ext; +}; +extern const struct gcov_link gcov_link[]; + +#endif /* GCOV_H */ diff --git a/kernel/groups.c b/kernel/groups.c new file mode 100644 index 00000000000..2b45b2ee396 --- /dev/null +++ b/kernel/groups.c @@ -0,0 +1,288 @@ +/* + * Supplementary group IDs + */ +#include <linux/cred.h> +#include <linux/module.h> +#include <linux/slab.h> +#include <linux/security.h> +#include <linux/syscalls.h> +#include <asm/uaccess.h> + +/* init to 2 - one for init_task, one to ensure it is never freed */ +struct group_info init_groups = { .usage = ATOMIC_INIT(2) }; + +struct group_info *groups_alloc(int gidsetsize) +{ + struct group_info *group_info; + int nblocks; + int i; + + nblocks = (gidsetsize + NGROUPS_PER_BLOCK - 1) / NGROUPS_PER_BLOCK; + /* Make sure we always allocate at least one indirect block pointer */ + nblocks = nblocks ? : 1; + group_info = kmalloc(sizeof(*group_info) + nblocks*sizeof(gid_t *), GFP_USER); + if (!group_info) + return NULL; + group_info->ngroups = gidsetsize; + group_info->nblocks = nblocks; + atomic_set(&group_info->usage, 1); + + if (gidsetsize <= NGROUPS_SMALL) + group_info->blocks[0] = group_info->small_block; + else { + for (i = 0; i < nblocks; i++) { + gid_t *b; + b = (void *)__get_free_page(GFP_USER); + if (!b) + goto out_undo_partial_alloc; + group_info->blocks[i] = b; + } + } + return group_info; + +out_undo_partial_alloc: + while (--i >= 0) { + free_page((unsigned long)group_info->blocks[i]); + } + kfree(group_info); + return NULL; +} + +EXPORT_SYMBOL(groups_alloc); + +void groups_free(struct group_info *group_info) +{ + if (group_info->blocks[0] != group_info->small_block) { + int i; + for (i = 0; i < group_info->nblocks; i++) + free_page((unsigned long)group_info->blocks[i]); + } + kfree(group_info); +} + +EXPORT_SYMBOL(groups_free); + +/* export the group_info to a user-space array */ +static int groups_to_user(gid_t __user *grouplist, + const struct group_info *group_info) +{ + int i; + unsigned int count = group_info->ngroups; + + for (i = 0; i < group_info->nblocks; i++) { + unsigned int cp_count = min(NGROUPS_PER_BLOCK, count); + unsigned int len = cp_count * sizeof(*grouplist); + + if (copy_to_user(grouplist, group_info->blocks[i], len)) + return -EFAULT; + + grouplist += NGROUPS_PER_BLOCK; + count -= cp_count; + } + return 0; +} + +/* fill a group_info from a user-space array - it must be allocated already */ +static int groups_from_user(struct group_info *group_info, + gid_t __user *grouplist) +{ + int i; + unsigned int count = group_info->ngroups; + + for (i = 0; i < group_info->nblocks; i++) { + unsigned int cp_count = min(NGROUPS_PER_BLOCK, count); + unsigned int len = cp_count * sizeof(*grouplist); + + if (copy_from_user(group_info->blocks[i], grouplist, len)) + return -EFAULT; + + grouplist += NGROUPS_PER_BLOCK; + count -= cp_count; + } + return 0; +} + +/* a simple Shell sort */ +static void groups_sort(struct group_info *group_info) +{ + int base, max, stride; + int gidsetsize = group_info->ngroups; + + for (stride = 1; stride < gidsetsize; stride = 3 * stride + 1) + ; /* nothing */ + stride /= 3; + + while (stride) { + max = gidsetsize - stride; + for (base = 0; base < max; base++) { + int left = base; + int right = left + stride; + gid_t tmp = GROUP_AT(group_info, right); + + while (left >= 0 && GROUP_AT(group_info, left) > tmp) { + GROUP_AT(group_info, right) = + GROUP_AT(group_info, left); + right = left; + left -= stride; + } + GROUP_AT(group_info, right) = tmp; + } + stride /= 3; + } +} + +/* a simple bsearch */ +int groups_search(const struct group_info *group_info, gid_t grp) +{ + unsigned int left, right; + + if (!group_info) + return 0; + + left = 0; + right = group_info->ngroups; + while (left < right) { + unsigned int mid = (left+right)/2; + int cmp = grp - GROUP_AT(group_info, mid); + if (cmp > 0) + left = mid + 1; + else if (cmp < 0) + right = mid; + else + return 1; + } + return 0; +} + +/** + * set_groups - Change a group subscription in a set of credentials + * @new: The newly prepared set of credentials to alter + * @group_info: The group list to install + * + * Validate a group subscription and, if valid, insert it into a set + * of credentials. + */ +int set_groups(struct cred *new, struct group_info *group_info) +{ + int retval; + + retval = security_task_setgroups(group_info); + if (retval) + return retval; + + put_group_info(new->group_info); + groups_sort(group_info); + get_group_info(group_info); + new->group_info = group_info; + return 0; +} + +EXPORT_SYMBOL(set_groups); + +/** + * set_current_groups - Change current's group subscription + * @group_info: The group list to impose + * + * Validate a group subscription and, if valid, impose it upon current's task + * security record. + */ +int set_current_groups(struct group_info *group_info) +{ + struct cred *new; + int ret; + + new = prepare_creds(); + if (!new) + return -ENOMEM; + + ret = set_groups(new, group_info); + if (ret < 0) { + abort_creds(new); + return ret; + } + + return commit_creds(new); +} + +EXPORT_SYMBOL(set_current_groups); + +SYSCALL_DEFINE2(getgroups, int, gidsetsize, gid_t __user *, grouplist) +{ + const struct cred *cred = current_cred(); + int i; + + if (gidsetsize < 0) + return -EINVAL; + + /* no need to grab task_lock here; it cannot change */ + i = cred->group_info->ngroups; + if (gidsetsize) { + if (i > gidsetsize) { + i = -EINVAL; + goto out; + } + if (groups_to_user(grouplist, cred->group_info)) { + i = -EFAULT; + goto out; + } + } +out: + return i; +} + +/* + * SMP: Our groups are copy-on-write. We can set them safely + * without another task interfering. + */ + +SYSCALL_DEFINE2(setgroups, int, gidsetsize, gid_t __user *, grouplist) +{ + struct group_info *group_info; + int retval; + + if (!capable(CAP_SETGID)) + return -EPERM; + if ((unsigned)gidsetsize > NGROUPS_MAX) + return -EINVAL; + + group_info = groups_alloc(gidsetsize); + if (!group_info) + return -ENOMEM; + retval = groups_from_user(group_info, grouplist); + if (retval) { + put_group_info(group_info); + return retval; + } + + retval = set_current_groups(group_info); + put_group_info(group_info); + + return retval; +} + +/* + * Check whether we're fsgid/egid or in the supplemental group.. + */ +int in_group_p(gid_t grp) +{ + const struct cred *cred = current_cred(); + int retval = 1; + + if (grp != cred->fsgid) + retval = groups_search(cred->group_info, grp); + return retval; +} + +EXPORT_SYMBOL(in_group_p); + +int in_egroup_p(gid_t grp) +{ + const struct cred *cred = current_cred(); + int retval = 1; + + if (grp != cred->egid) + retval = groups_search(cred->group_info, grp); + return retval; +} + +EXPORT_SYMBOL(in_egroup_p); diff --git a/kernel/hrtimer.c b/kernel/hrtimer.c index f394d2a42ca..ede52770812 100644 --- a/kernel/hrtimer.c +++ b/kernel/hrtimer.c @@ -43,39 +43,12 @@ #include <linux/seq_file.h> #include <linux/err.h> #include <linux/debugobjects.h> +#include <linux/sched.h> +#include <linux/timer.h> #include <asm/uaccess.h> -/** - * ktime_get - get the monotonic time in ktime_t format - * - * returns the time in ktime_t format - */ -ktime_t ktime_get(void) -{ - struct timespec now; - - ktime_get_ts(&now); - - return timespec_to_ktime(now); -} -EXPORT_SYMBOL_GPL(ktime_get); - -/** - * ktime_get_real - get the real (wall-) time in ktime_t format - * - * returns the time in ktime_t format - */ -ktime_t ktime_get_real(void) -{ - struct timespec now; - - getnstimeofday(&now); - - return timespec_to_ktime(now); -} - -EXPORT_SYMBOL_GPL(ktime_get_real); +#include <trace/events/timer.h> /* * The timer bases: @@ -104,31 +77,6 @@ DEFINE_PER_CPU(struct hrtimer_cpu_base, hrtimer_bases) = } }; -/** - * ktime_get_ts - get the monotonic clock in timespec format - * @ts: pointer to timespec variable - * - * The function calculates the monotonic clock from the realtime - * clock and the wall_to_monotonic offset and stores the result - * in normalized timespec format in the variable pointed to by @ts. - */ -void ktime_get_ts(struct timespec *ts) -{ - struct timespec tomono; - unsigned long seq; - - do { - seq = read_seqbegin(&xtime_lock); - getnstimeofday(ts); - tomono = wall_to_monotonic; - - } while (read_seqretry(&xtime_lock, seq)); - - set_normalized_timespec(ts, ts->tv_sec + tomono.tv_sec, - ts->tv_nsec + tomono.tv_nsec); -} -EXPORT_SYMBOL_GPL(ktime_get_ts); - /* * Get the coarse grained time at the softirq based on xtime and * wall_to_monotonic. @@ -189,21 +137,65 @@ struct hrtimer_clock_base *lock_hrtimer_base(const struct hrtimer *timer, } } + +/* + * Get the preferred target CPU for NOHZ + */ +static int hrtimer_get_target(int this_cpu, int pinned) +{ +#ifdef CONFIG_NO_HZ + if (!pinned && get_sysctl_timer_migration() && idle_cpu(this_cpu)) { + int preferred_cpu = get_nohz_load_balancer(); + + if (preferred_cpu >= 0) + return preferred_cpu; + } +#endif + return this_cpu; +} + +/* + * With HIGHRES=y we do not migrate the timer when it is expiring + * before the next event on the target cpu because we cannot reprogram + * the target cpu hardware and we would cause it to fire late. + * + * Called with cpu_base->lock of target cpu held. + */ +static int +hrtimer_check_target(struct hrtimer *timer, struct hrtimer_clock_base *new_base) +{ +#ifdef CONFIG_HIGH_RES_TIMERS + ktime_t expires; + + if (!new_base->cpu_base->hres_active) + return 0; + + expires = ktime_sub(hrtimer_get_expires(timer), new_base->offset); + return expires.tv64 <= new_base->cpu_base->expires_next.tv64; +#else + return 0; +#endif +} + /* * Switch the timer base to the current CPU when possible. */ static inline struct hrtimer_clock_base * -switch_hrtimer_base(struct hrtimer *timer, struct hrtimer_clock_base *base) +switch_hrtimer_base(struct hrtimer *timer, struct hrtimer_clock_base *base, + int pinned) { struct hrtimer_clock_base *new_base; struct hrtimer_cpu_base *new_cpu_base; + int this_cpu = smp_processor_id(); + int cpu = hrtimer_get_target(this_cpu, pinned); - new_cpu_base = &__get_cpu_var(hrtimer_bases); +again: + new_cpu_base = &per_cpu(hrtimer_bases, cpu); new_base = &new_cpu_base->clock_base[base->index]; if (base != new_base) { /* - * We are trying to schedule the timer on the local CPU. + * We are trying to move timer to new_base. * However we can't change timer's base while it is running, * so we keep it on the same CPU. No hassle vs. reprogramming * the event source in the high resolution case. The softirq @@ -218,6 +210,14 @@ switch_hrtimer_base(struct hrtimer *timer, struct hrtimer_clock_base *base) timer->base = NULL; spin_unlock(&base->cpu_base->lock); spin_lock(&new_base->cpu_base->lock); + + if (cpu != this_cpu && hrtimer_check_target(timer, new_base)) { + cpu = this_cpu; + spin_unlock(&new_base->cpu_base->lock); + spin_lock(&base->cpu_base->lock); + timer->base = base; + goto again; + } timer->base = new_base; } return new_base; @@ -235,7 +235,7 @@ lock_hrtimer_base(const struct hrtimer *timer, unsigned long *flags) return base; } -# define switch_hrtimer_base(t, b) (b) +# define switch_hrtimer_base(t, b, p) (b) #endif /* !CONFIG_SMP */ @@ -332,6 +332,8 @@ ktime_t ktime_add_safe(const ktime_t lhs, const ktime_t rhs) return res; } +EXPORT_SYMBOL_GPL(ktime_add_safe); + #ifdef CONFIG_DEBUG_OBJECTS_TIMERS static struct debug_obj_descr hrtimer_debug_descr; @@ -429,6 +431,7 @@ void hrtimer_init_on_stack(struct hrtimer *timer, clockid_t clock_id, debug_object_init_on_stack(timer, &hrtimer_debug_descr); __hrtimer_init(timer, clock_id, mode); } +EXPORT_SYMBOL_GPL(hrtimer_init_on_stack); void destroy_hrtimer_on_stack(struct hrtimer *timer) { @@ -441,6 +444,26 @@ static inline void debug_hrtimer_activate(struct hrtimer *timer) { } static inline void debug_hrtimer_deactivate(struct hrtimer *timer) { } #endif +static inline void +debug_init(struct hrtimer *timer, clockid_t clockid, + enum hrtimer_mode mode) +{ + debug_hrtimer_init(timer); + trace_hrtimer_init(timer, clockid, mode); +} + +static inline void debug_activate(struct hrtimer *timer) +{ + debug_hrtimer_activate(timer); + trace_hrtimer_start(timer); +} + +static inline void debug_deactivate(struct hrtimer *timer) +{ + debug_hrtimer_deactivate(timer); + trace_hrtimer_cancel(timer); +} + /* High resolution timer related functions */ #ifdef CONFIG_HIGH_RES_TIMERS @@ -486,13 +509,14 @@ static inline int hrtimer_hres_active(void) * next event * Called with interrupts disabled and base->lock held */ -static void hrtimer_force_reprogram(struct hrtimer_cpu_base *cpu_base) +static void +hrtimer_force_reprogram(struct hrtimer_cpu_base *cpu_base, int skip_equal) { int i; struct hrtimer_clock_base *base = cpu_base->clock_base; - ktime_t expires; + ktime_t expires, expires_next; - cpu_base->expires_next.tv64 = KTIME_MAX; + expires_next.tv64 = KTIME_MAX; for (i = 0; i < HRTIMER_MAX_CLOCK_BASES; i++, base++) { struct hrtimer *timer; @@ -508,10 +532,15 @@ static void hrtimer_force_reprogram(struct hrtimer_cpu_base *cpu_base) */ if (expires.tv64 < 0) expires.tv64 = 0; - if (expires.tv64 < cpu_base->expires_next.tv64) - cpu_base->expires_next = expires; + if (expires.tv64 < expires_next.tv64) + expires_next = expires; } + if (skip_equal && expires_next.tv64 == cpu_base->expires_next.tv64) + return; + + cpu_base->expires_next.tv64 = expires_next.tv64; + if (cpu_base->expires_next.tv64 != KTIME_MAX) tick_program_event(cpu_base->expires_next, 1); } @@ -594,7 +623,7 @@ static void retrigger_next_event(void *arg) base->clock_base[CLOCK_REALTIME].offset = timespec_to_ktime(realtime_offset); - hrtimer_force_reprogram(base); + hrtimer_force_reprogram(base, 0); spin_unlock(&base->lock); } @@ -651,14 +680,20 @@ static inline void hrtimer_init_timer_hres(struct hrtimer *timer) * and expiry check is done in the hrtimer_interrupt or in the softirq. */ static inline int hrtimer_enqueue_reprogram(struct hrtimer *timer, - struct hrtimer_clock_base *base) + struct hrtimer_clock_base *base, + int wakeup) { if (base->cpu_base->hres_active && hrtimer_reprogram(timer, base)) { - spin_unlock(&base->cpu_base->lock); - raise_softirq_irqoff(HRTIMER_SOFTIRQ); - spin_lock(&base->cpu_base->lock); + if (wakeup) { + spin_unlock(&base->cpu_base->lock); + raise_softirq_irqoff(HRTIMER_SOFTIRQ); + spin_lock(&base->cpu_base->lock); + } else + __raise_softirq_irqoff(HRTIMER_SOFTIRQ); + return 1; } + return 0; } @@ -691,8 +726,6 @@ static int hrtimer_switch_to_hres(void) /* "Retrigger" the interrupt to get things going */ retrigger_next_event(NULL); local_irq_restore(flags); - printk(KERN_DEBUG "Switched to high resolution mode on CPU %d\n", - smp_processor_id()); return 1; } @@ -701,9 +734,11 @@ static int hrtimer_switch_to_hres(void) static inline int hrtimer_hres_active(void) { return 0; } static inline int hrtimer_is_hres_enabled(void) { return 0; } static inline int hrtimer_switch_to_hres(void) { return 0; } -static inline void hrtimer_force_reprogram(struct hrtimer_cpu_base *base) { } +static inline void +hrtimer_force_reprogram(struct hrtimer_cpu_base *base, int skip_equal) { } static inline int hrtimer_enqueue_reprogram(struct hrtimer *timer, - struct hrtimer_clock_base *base) + struct hrtimer_clock_base *base, + int wakeup) { return 0; } @@ -790,7 +825,7 @@ static int enqueue_hrtimer(struct hrtimer *timer, struct hrtimer *entry; int leftmost = 1; - debug_hrtimer_activate(timer); + debug_activate(timer); /* * Find the right place in the rbtree: @@ -843,19 +878,29 @@ static void __remove_hrtimer(struct hrtimer *timer, struct hrtimer_clock_base *base, unsigned long newstate, int reprogram) { - if (timer->state & HRTIMER_STATE_ENQUEUED) { - /* - * Remove the timer from the rbtree and replace the - * first entry pointer if necessary. - */ - if (base->first == &timer->node) { - base->first = rb_next(&timer->node); - /* Reprogram the clock event device. if enabled */ - if (reprogram && hrtimer_hres_active()) - hrtimer_force_reprogram(base->cpu_base); + if (!(timer->state & HRTIMER_STATE_ENQUEUED)) + goto out; + + /* + * Remove the timer from the rbtree and replace the first + * entry pointer if necessary. + */ + if (base->first == &timer->node) { + base->first = rb_next(&timer->node); +#ifdef CONFIG_HIGH_RES_TIMERS + /* Reprogram the clock event device. if enabled */ + if (reprogram && hrtimer_hres_active()) { + ktime_t expires; + + expires = ktime_sub(hrtimer_get_expires(timer), + base->offset); + if (base->cpu_base->expires_next.tv64 == expires.tv64) + hrtimer_force_reprogram(base->cpu_base, 1); } - rb_erase(&timer->node, &base->active); +#endif } + rb_erase(&timer->node, &base->active); +out: timer->state = newstate; } @@ -876,7 +921,7 @@ remove_hrtimer(struct hrtimer *timer, struct hrtimer_clock_base *base) * reprogramming happens in the interrupt handler. This is a * rare case and less expensive than a smp call. */ - debug_hrtimer_deactivate(timer); + debug_deactivate(timer); timer_stats_hrtimer_clear_start_info(timer); reprogram = base->cpu_base == &__get_cpu_var(hrtimer_bases); __remove_hrtimer(timer, base, HRTIMER_STATE_INACTIVE, @@ -886,20 +931,9 @@ remove_hrtimer(struct hrtimer *timer, struct hrtimer_clock_base *base) return 0; } -/** - * hrtimer_start_range_ns - (re)start an hrtimer on the current CPU - * @timer: the timer to be added - * @tim: expiry time - * @delta_ns: "slack" range for the timer - * @mode: expiry mode: absolute (HRTIMER_ABS) or relative (HRTIMER_REL) - * - * Returns: - * 0 on success - * 1 when the timer was active - */ -int -hrtimer_start_range_ns(struct hrtimer *timer, ktime_t tim, unsigned long delta_ns, - const enum hrtimer_mode mode) +int __hrtimer_start_range_ns(struct hrtimer *timer, ktime_t tim, + unsigned long delta_ns, const enum hrtimer_mode mode, + int wakeup) { struct hrtimer_clock_base *base, *new_base; unsigned long flags; @@ -911,9 +945,9 @@ hrtimer_start_range_ns(struct hrtimer *timer, ktime_t tim, unsigned long delta_n ret = remove_hrtimer(timer, base); /* Switch the timer base, if necessary: */ - new_base = switch_hrtimer_base(timer, base); + new_base = switch_hrtimer_base(timer, base, mode & HRTIMER_MODE_PINNED); - if (mode == HRTIMER_MODE_REL) { + if (mode & HRTIMER_MODE_REL) { tim = ktime_add_safe(tim, new_base->get_time()); /* * CONFIG_TIME_LOW_RES is a temporary way for architectures @@ -940,12 +974,29 @@ hrtimer_start_range_ns(struct hrtimer *timer, ktime_t tim, unsigned long delta_n * XXX send_remote_softirq() ? */ if (leftmost && new_base->cpu_base == &__get_cpu_var(hrtimer_bases)) - hrtimer_enqueue_reprogram(timer, new_base); + hrtimer_enqueue_reprogram(timer, new_base, wakeup); unlock_hrtimer_base(timer, &flags); return ret; } + +/** + * hrtimer_start_range_ns - (re)start an hrtimer on the current CPU + * @timer: the timer to be added + * @tim: expiry time + * @delta_ns: "slack" range for the timer + * @mode: expiry mode: absolute (HRTIMER_ABS) or relative (HRTIMER_REL) + * + * Returns: + * 0 on success + * 1 when the timer was active + */ +int hrtimer_start_range_ns(struct hrtimer *timer, ktime_t tim, + unsigned long delta_ns, const enum hrtimer_mode mode) +{ + return __hrtimer_start_range_ns(timer, tim, delta_ns, mode, 1); +} EXPORT_SYMBOL_GPL(hrtimer_start_range_ns); /** @@ -961,7 +1012,7 @@ EXPORT_SYMBOL_GPL(hrtimer_start_range_ns); int hrtimer_start(struct hrtimer *timer, ktime_t tim, const enum hrtimer_mode mode) { - return hrtimer_start_range_ns(timer, tim, 0, mode); + return __hrtimer_start_range_ns(timer, tim, 0, mode, 1); } EXPORT_SYMBOL_GPL(hrtimer_start); @@ -1085,7 +1136,6 @@ static void __hrtimer_init(struct hrtimer *timer, clockid_t clock_id, clock_id = CLOCK_MONOTONIC; timer->base = &cpu_base->clock_base[clock_id]; - INIT_LIST_HEAD(&timer->cb_entry); hrtimer_init_timer_hres(timer); #ifdef CONFIG_TIMER_STATS @@ -1104,7 +1154,7 @@ static void __hrtimer_init(struct hrtimer *timer, clockid_t clock_id, void hrtimer_init(struct hrtimer *timer, clockid_t clock_id, enum hrtimer_mode mode) { - debug_hrtimer_init(timer); + debug_init(timer, clock_id, mode); __hrtimer_init(timer, clock_id, mode); } EXPORT_SYMBOL_GPL(hrtimer_init); @@ -1128,7 +1178,7 @@ int hrtimer_get_res(const clockid_t which_clock, struct timespec *tp) } EXPORT_SYMBOL_GPL(hrtimer_get_res); -static void __run_hrtimer(struct hrtimer *timer) +static void __run_hrtimer(struct hrtimer *timer, ktime_t *now) { struct hrtimer_clock_base *base = timer->base; struct hrtimer_cpu_base *cpu_base = base->cpu_base; @@ -1137,7 +1187,7 @@ static void __run_hrtimer(struct hrtimer *timer) WARN_ON(!irqs_disabled()); - debug_hrtimer_deactivate(timer); + debug_deactivate(timer); __remove_hrtimer(timer, base, HRTIMER_STATE_CALLBACK, 0); timer_stats_account_hrtimer(timer); fn = timer->function; @@ -1148,7 +1198,9 @@ static void __run_hrtimer(struct hrtimer *timer) * the timer base. */ spin_unlock(&cpu_base->lock); + trace_hrtimer_expire_entry(timer, now); restart = fn(timer); + trace_hrtimer_expire_exit(timer); spin_lock(&cpu_base->lock); /* @@ -1186,7 +1238,8 @@ hrtimer_interrupt_hanging(struct clock_event_device *dev, force_clock_reprogram = 1; dev->min_delta_ns = (unsigned long)try_time.tv64 * 3; printk(KERN_WARNING "hrtimer: interrupt too slow, " - "forcing clock min delta to %lu ns\n", dev->min_delta_ns); + "forcing clock min delta to %llu ns\n", + (unsigned long long) dev->min_delta_ns); } /* * High resolution timer interrupt @@ -1213,14 +1266,22 @@ void hrtimer_interrupt(struct clock_event_device *dev) expires_next.tv64 = KTIME_MAX; + spin_lock(&cpu_base->lock); + /* + * We set expires_next to KTIME_MAX here with cpu_base->lock + * held to prevent that a timer is enqueued in our queue via + * the migration code. This does not affect enqueueing of + * timers which run their callback and need to be requeued on + * this CPU. + */ + cpu_base->expires_next.tv64 = KTIME_MAX; + base = cpu_base->clock_base; for (i = 0; i < HRTIMER_MAX_CLOCK_BASES; i++) { ktime_t basenow; struct rb_node *node; - spin_lock(&cpu_base->lock); - basenow = ktime_add(now, base->offset); while ((node = base->first)) { @@ -1251,13 +1312,17 @@ void hrtimer_interrupt(struct clock_event_device *dev) break; } - __run_hrtimer(timer); + __run_hrtimer(timer, &basenow); } - spin_unlock(&cpu_base->lock); base++; } + /* + * Store the new expiry value so the migration code can verify + * against it. + */ cpu_base->expires_next = expires_next; + spin_unlock(&cpu_base->lock); /* Reprogramming necessary ? */ if (expires_next.tv64 != KTIME_MAX) { @@ -1369,7 +1434,7 @@ void hrtimer_run_queues(void) hrtimer_get_expires_tv64(timer)) break; - __run_hrtimer(timer); + __run_hrtimer(timer, &base->softirq_time); } spin_unlock(&cpu_base->lock); } @@ -1396,6 +1461,7 @@ void hrtimer_init_sleeper(struct hrtimer_sleeper *sl, struct task_struct *task) sl->timer.function = hrtimer_wakeup; sl->task = task; } +EXPORT_SYMBOL_GPL(hrtimer_init_sleeper); static int __sched do_nanosleep(struct hrtimer_sleeper *t, enum hrtimer_mode mode) { @@ -1545,7 +1611,7 @@ static void migrate_hrtimer_list(struct hrtimer_clock_base *old_base, while ((node = rb_first(&old_base->active))) { timer = rb_entry(node, struct hrtimer, node); BUG_ON(hrtimer_callback_running(timer)); - debug_hrtimer_deactivate(timer); + debug_deactivate(timer); /* * Mark it as STATE_MIGRATE not INACTIVE otherwise the diff --git a/kernel/hung_task.c b/kernel/hung_task.c new file mode 100644 index 00000000000..0c642d51aac --- /dev/null +++ b/kernel/hung_task.c @@ -0,0 +1,217 @@ +/* + * Detect Hung Task + * + * kernel/hung_task.c - kernel thread for detecting tasks stuck in D state + * + */ + +#include <linux/mm.h> +#include <linux/cpu.h> +#include <linux/nmi.h> +#include <linux/init.h> +#include <linux/delay.h> +#include <linux/freezer.h> +#include <linux/kthread.h> +#include <linux/lockdep.h> +#include <linux/module.h> +#include <linux/sysctl.h> + +/* + * The number of tasks checked: + */ +unsigned long __read_mostly sysctl_hung_task_check_count = PID_MAX_LIMIT; + +/* + * Limit number of tasks checked in a batch. + * + * This value controls the preemptibility of khungtaskd since preemption + * is disabled during the critical section. It also controls the size of + * the RCU grace period. So it needs to be upper-bound. + */ +#define HUNG_TASK_BATCHING 1024 + +/* + * Zero means infinite timeout - no checking done: + */ +unsigned long __read_mostly sysctl_hung_task_timeout_secs = 120; + +unsigned long __read_mostly sysctl_hung_task_warnings = 10; + +static int __read_mostly did_panic; + +static struct task_struct *watchdog_task; + +/* + * Should we panic (and reboot, if panic_timeout= is set) when a + * hung task is detected: + */ +unsigned int __read_mostly sysctl_hung_task_panic = + CONFIG_BOOTPARAM_HUNG_TASK_PANIC_VALUE; + +static int __init hung_task_panic_setup(char *str) +{ + sysctl_hung_task_panic = simple_strtoul(str, NULL, 0); + + return 1; +} +__setup("hung_task_panic=", hung_task_panic_setup); + +static int +hung_task_panic(struct notifier_block *this, unsigned long event, void *ptr) +{ + did_panic = 1; + + return NOTIFY_DONE; +} + +static struct notifier_block panic_block = { + .notifier_call = hung_task_panic, +}; + +static void check_hung_task(struct task_struct *t, unsigned long timeout) +{ + unsigned long switch_count = t->nvcsw + t->nivcsw; + + /* + * Ensure the task is not frozen. + * Also, when a freshly created task is scheduled once, changes + * its state to TASK_UNINTERRUPTIBLE without having ever been + * switched out once, it musn't be checked. + */ + if (unlikely(t->flags & PF_FROZEN || !switch_count)) + return; + + if (switch_count != t->last_switch_count) { + t->last_switch_count = switch_count; + return; + } + if (!sysctl_hung_task_warnings) + return; + sysctl_hung_task_warnings--; + + /* + * Ok, the task did not get scheduled for more than 2 minutes, + * complain: + */ + printk(KERN_ERR "INFO: task %s:%d blocked for more than " + "%ld seconds.\n", t->comm, t->pid, timeout); + printk(KERN_ERR "\"echo 0 > /proc/sys/kernel/hung_task_timeout_secs\"" + " disables this message.\n"); + sched_show_task(t); + __debug_show_held_locks(t); + + touch_nmi_watchdog(); + + if (sysctl_hung_task_panic) + panic("hung_task: blocked tasks"); +} + +/* + * To avoid extending the RCU grace period for an unbounded amount of time, + * periodically exit the critical section and enter a new one. + * + * For preemptible RCU it is sufficient to call rcu_read_unlock in order + * exit the grace period. For classic RCU, a reschedule is required. + */ +static void rcu_lock_break(struct task_struct *g, struct task_struct *t) +{ + get_task_struct(g); + get_task_struct(t); + rcu_read_unlock(); + cond_resched(); + rcu_read_lock(); + put_task_struct(t); + put_task_struct(g); +} + +/* + * Check whether a TASK_UNINTERRUPTIBLE does not get woken up for + * a really long time (120 seconds). If that happens, print out + * a warning. + */ +static void check_hung_uninterruptible_tasks(unsigned long timeout) +{ + int max_count = sysctl_hung_task_check_count; + int batch_count = HUNG_TASK_BATCHING; + struct task_struct *g, *t; + + /* + * If the system crashed already then all bets are off, + * do not report extra hung tasks: + */ + if (test_taint(TAINT_DIE) || did_panic) + return; + + rcu_read_lock(); + do_each_thread(g, t) { + if (!max_count--) + goto unlock; + if (!--batch_count) { + batch_count = HUNG_TASK_BATCHING; + rcu_lock_break(g, t); + /* Exit if t or g was unhashed during refresh. */ + if (t->state == TASK_DEAD || g->state == TASK_DEAD) + goto unlock; + } + /* use "==" to skip the TASK_KILLABLE tasks waiting on NFS */ + if (t->state == TASK_UNINTERRUPTIBLE) + check_hung_task(t, timeout); + } while_each_thread(g, t); + unlock: + rcu_read_unlock(); +} + +static unsigned long timeout_jiffies(unsigned long timeout) +{ + /* timeout of 0 will disable the watchdog */ + return timeout ? timeout * HZ : MAX_SCHEDULE_TIMEOUT; +} + +/* + * Process updating of timeout sysctl + */ +int proc_dohung_task_timeout_secs(struct ctl_table *table, int write, + void __user *buffer, + size_t *lenp, loff_t *ppos) +{ + int ret; + + ret = proc_doulongvec_minmax(table, write, buffer, lenp, ppos); + + if (ret || !write) + goto out; + + wake_up_process(watchdog_task); + + out: + return ret; +} + +/* + * kthread which checks for tasks stuck in D state + */ +static int watchdog(void *dummy) +{ + set_user_nice(current, 0); + + for ( ; ; ) { + unsigned long timeout = sysctl_hung_task_timeout_secs; + + while (schedule_timeout_interruptible(timeout_jiffies(timeout))) + timeout = sysctl_hung_task_timeout_secs; + + check_hung_uninterruptible_tasks(timeout); + } + + return 0; +} + +static int __init hung_task_init(void) +{ + atomic_notifier_chain_register(&panic_notifier_list, &panic_block); + watchdog_task = kthread_run(watchdog, NULL, "khungtaskd"); + + return 0; +} + +module_init(hung_task_init); diff --git a/kernel/hw_breakpoint.c b/kernel/hw_breakpoint.c new file mode 100644 index 00000000000..cf5ee162841 --- /dev/null +++ b/kernel/hw_breakpoint.c @@ -0,0 +1,423 @@ +/* + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + * + * Copyright (C) 2007 Alan Stern + * Copyright (C) IBM Corporation, 2009 + * Copyright (C) 2009, Frederic Weisbecker <fweisbec@gmail.com> + * + * Thanks to Ingo Molnar for his many suggestions. + * + * Authors: Alan Stern <stern@rowland.harvard.edu> + * K.Prasad <prasad@linux.vnet.ibm.com> + * Frederic Weisbecker <fweisbec@gmail.com> + */ + +/* + * HW_breakpoint: a unified kernel/user-space hardware breakpoint facility, + * using the CPU's debug registers. + * This file contains the arch-independent routines. + */ + +#include <linux/irqflags.h> +#include <linux/kallsyms.h> +#include <linux/notifier.h> +#include <linux/kprobes.h> +#include <linux/kdebug.h> +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/percpu.h> +#include <linux/sched.h> +#include <linux/init.h> +#include <linux/smp.h> + +#include <linux/hw_breakpoint.h> + +/* + * Constraints data + */ + +/* Number of pinned cpu breakpoints in a cpu */ +static DEFINE_PER_CPU(unsigned int, nr_cpu_bp_pinned); + +/* Number of pinned task breakpoints in a cpu */ +static DEFINE_PER_CPU(unsigned int, task_bp_pinned[HBP_NUM]); + +/* Number of non-pinned cpu/task breakpoints in a cpu */ +static DEFINE_PER_CPU(unsigned int, nr_bp_flexible); + +/* Gather the number of total pinned and un-pinned bp in a cpuset */ +struct bp_busy_slots { + unsigned int pinned; + unsigned int flexible; +}; + +/* Serialize accesses to the above constraints */ +static DEFINE_MUTEX(nr_bp_mutex); + +/* + * Report the maximum number of pinned breakpoints a task + * have in this cpu + */ +static unsigned int max_task_bp_pinned(int cpu) +{ + int i; + unsigned int *tsk_pinned = per_cpu(task_bp_pinned, cpu); + + for (i = HBP_NUM -1; i >= 0; i--) { + if (tsk_pinned[i] > 0) + return i + 1; + } + + return 0; +} + +/* + * Report the number of pinned/un-pinned breakpoints we have in + * a given cpu (cpu > -1) or in all of them (cpu = -1). + */ +static void fetch_bp_busy_slots(struct bp_busy_slots *slots, int cpu) +{ + if (cpu >= 0) { + slots->pinned = per_cpu(nr_cpu_bp_pinned, cpu); + slots->pinned += max_task_bp_pinned(cpu); + slots->flexible = per_cpu(nr_bp_flexible, cpu); + + return; + } + + for_each_online_cpu(cpu) { + unsigned int nr; + + nr = per_cpu(nr_cpu_bp_pinned, cpu); + nr += max_task_bp_pinned(cpu); + + if (nr > slots->pinned) + slots->pinned = nr; + + nr = per_cpu(nr_bp_flexible, cpu); + + if (nr > slots->flexible) + slots->flexible = nr; + } +} + +/* + * Add a pinned breakpoint for the given task in our constraint table + */ +static void toggle_bp_task_slot(struct task_struct *tsk, int cpu, bool enable) +{ + int count = 0; + struct perf_event *bp; + struct perf_event_context *ctx = tsk->perf_event_ctxp; + unsigned int *tsk_pinned; + struct list_head *list; + unsigned long flags; + + if (WARN_ONCE(!ctx, "No perf context for this task")) + return; + + list = &ctx->event_list; + + spin_lock_irqsave(&ctx->lock, flags); + + /* + * The current breakpoint counter is not included in the list + * at the open() callback time + */ + list_for_each_entry(bp, list, event_entry) { + if (bp->attr.type == PERF_TYPE_BREAKPOINT) + count++; + } + + spin_unlock_irqrestore(&ctx->lock, flags); + + if (WARN_ONCE(count < 0, "No breakpoint counter found in the counter list")) + return; + + tsk_pinned = per_cpu(task_bp_pinned, cpu); + if (enable) { + tsk_pinned[count]++; + if (count > 0) + tsk_pinned[count-1]--; + } else { + tsk_pinned[count]--; + if (count > 0) + tsk_pinned[count-1]++; + } +} + +/* + * Add/remove the given breakpoint in our constraint table + */ +static void toggle_bp_slot(struct perf_event *bp, bool enable) +{ + int cpu = bp->cpu; + struct task_struct *tsk = bp->ctx->task; + + /* Pinned counter task profiling */ + if (tsk) { + if (cpu >= 0) { + toggle_bp_task_slot(tsk, cpu, enable); + return; + } + + for_each_online_cpu(cpu) + toggle_bp_task_slot(tsk, cpu, enable); + return; + } + + /* Pinned counter cpu profiling */ + if (enable) + per_cpu(nr_cpu_bp_pinned, bp->cpu)++; + else + per_cpu(nr_cpu_bp_pinned, bp->cpu)--; +} + +/* + * Contraints to check before allowing this new breakpoint counter: + * + * == Non-pinned counter == (Considered as pinned for now) + * + * - If attached to a single cpu, check: + * + * (per_cpu(nr_bp_flexible, cpu) || (per_cpu(nr_cpu_bp_pinned, cpu) + * + max(per_cpu(task_bp_pinned, cpu)))) < HBP_NUM + * + * -> If there are already non-pinned counters in this cpu, it means + * there is already a free slot for them. + * Otherwise, we check that the maximum number of per task + * breakpoints (for this cpu) plus the number of per cpu breakpoint + * (for this cpu) doesn't cover every registers. + * + * - If attached to every cpus, check: + * + * (per_cpu(nr_bp_flexible, *) || (max(per_cpu(nr_cpu_bp_pinned, *)) + * + max(per_cpu(task_bp_pinned, *)))) < HBP_NUM + * + * -> This is roughly the same, except we check the number of per cpu + * bp for every cpu and we keep the max one. Same for the per tasks + * breakpoints. + * + * + * == Pinned counter == + * + * - If attached to a single cpu, check: + * + * ((per_cpu(nr_bp_flexible, cpu) > 1) + per_cpu(nr_cpu_bp_pinned, cpu) + * + max(per_cpu(task_bp_pinned, cpu))) < HBP_NUM + * + * -> Same checks as before. But now the nr_bp_flexible, if any, must keep + * one register at least (or they will never be fed). + * + * - If attached to every cpus, check: + * + * ((per_cpu(nr_bp_flexible, *) > 1) + max(per_cpu(nr_cpu_bp_pinned, *)) + * + max(per_cpu(task_bp_pinned, *))) < HBP_NUM + */ +int reserve_bp_slot(struct perf_event *bp) +{ + struct bp_busy_slots slots = {0}; + int ret = 0; + + mutex_lock(&nr_bp_mutex); + + fetch_bp_busy_slots(&slots, bp->cpu); + + /* Flexible counters need to keep at least one slot */ + if (slots.pinned + (!!slots.flexible) == HBP_NUM) { + ret = -ENOSPC; + goto end; + } + + toggle_bp_slot(bp, true); + +end: + mutex_unlock(&nr_bp_mutex); + + return ret; +} + +void release_bp_slot(struct perf_event *bp) +{ + mutex_lock(&nr_bp_mutex); + + toggle_bp_slot(bp, false); + + mutex_unlock(&nr_bp_mutex); +} + + +int __register_perf_hw_breakpoint(struct perf_event *bp) +{ + int ret; + + ret = reserve_bp_slot(bp); + if (ret) + return ret; + + /* + * Ptrace breakpoints can be temporary perf events only + * meant to reserve a slot. In this case, it is created disabled and + * we don't want to check the params right now (as we put a null addr) + * But perf tools create events as disabled and we want to check + * the params for them. + * This is a quick hack that will be removed soon, once we remove + * the tmp breakpoints from ptrace + */ + if (!bp->attr.disabled || bp->callback == perf_bp_event) + ret = arch_validate_hwbkpt_settings(bp, bp->ctx->task); + + return ret; +} + +int register_perf_hw_breakpoint(struct perf_event *bp) +{ + bp->callback = perf_bp_event; + + return __register_perf_hw_breakpoint(bp); +} + +/** + * register_user_hw_breakpoint - register a hardware breakpoint for user space + * @attr: breakpoint attributes + * @triggered: callback to trigger when we hit the breakpoint + * @tsk: pointer to 'task_struct' of the process to which the address belongs + */ +struct perf_event * +register_user_hw_breakpoint(struct perf_event_attr *attr, + perf_callback_t triggered, + struct task_struct *tsk) +{ + return perf_event_create_kernel_counter(attr, -1, tsk->pid, triggered); +} +EXPORT_SYMBOL_GPL(register_user_hw_breakpoint); + +/** + * modify_user_hw_breakpoint - modify a user-space hardware breakpoint + * @bp: the breakpoint structure to modify + * @attr: new breakpoint attributes + * @triggered: callback to trigger when we hit the breakpoint + * @tsk: pointer to 'task_struct' of the process to which the address belongs + */ +struct perf_event * +modify_user_hw_breakpoint(struct perf_event *bp, struct perf_event_attr *attr, + perf_callback_t triggered, + struct task_struct *tsk) +{ + /* + * FIXME: do it without unregistering + * - We don't want to lose our slot + * - If the new bp is incorrect, don't lose the older one + */ + unregister_hw_breakpoint(bp); + + return perf_event_create_kernel_counter(attr, -1, tsk->pid, triggered); +} +EXPORT_SYMBOL_GPL(modify_user_hw_breakpoint); + +/** + * unregister_hw_breakpoint - unregister a user-space hardware breakpoint + * @bp: the breakpoint structure to unregister + */ +void unregister_hw_breakpoint(struct perf_event *bp) +{ + if (!bp) + return; + perf_event_release_kernel(bp); +} +EXPORT_SYMBOL_GPL(unregister_hw_breakpoint); + +/** + * register_wide_hw_breakpoint - register a wide breakpoint in the kernel + * @attr: breakpoint attributes + * @triggered: callback to trigger when we hit the breakpoint + * + * @return a set of per_cpu pointers to perf events + */ +struct perf_event ** +register_wide_hw_breakpoint(struct perf_event_attr *attr, + perf_callback_t triggered) +{ + struct perf_event **cpu_events, **pevent, *bp; + long err; + int cpu; + + cpu_events = alloc_percpu(typeof(*cpu_events)); + if (!cpu_events) + return ERR_PTR(-ENOMEM); + + for_each_possible_cpu(cpu) { + pevent = per_cpu_ptr(cpu_events, cpu); + bp = perf_event_create_kernel_counter(attr, cpu, -1, triggered); + + *pevent = bp; + + if (IS_ERR(bp)) { + err = PTR_ERR(bp); + goto fail; + } + } + + return cpu_events; + +fail: + for_each_possible_cpu(cpu) { + pevent = per_cpu_ptr(cpu_events, cpu); + if (IS_ERR(*pevent)) + break; + unregister_hw_breakpoint(*pevent); + } + free_percpu(cpu_events); + /* return the error if any */ + return ERR_PTR(err); +} +EXPORT_SYMBOL_GPL(register_wide_hw_breakpoint); + +/** + * unregister_wide_hw_breakpoint - unregister a wide breakpoint in the kernel + * @cpu_events: the per cpu set of events to unregister + */ +void unregister_wide_hw_breakpoint(struct perf_event **cpu_events) +{ + int cpu; + struct perf_event **pevent; + + for_each_possible_cpu(cpu) { + pevent = per_cpu_ptr(cpu_events, cpu); + unregister_hw_breakpoint(*pevent); + } + free_percpu(cpu_events); +} +EXPORT_SYMBOL_GPL(unregister_wide_hw_breakpoint); + +static struct notifier_block hw_breakpoint_exceptions_nb = { + .notifier_call = hw_breakpoint_exceptions_notify, + /* we need to be notified first */ + .priority = 0x7fffffff +}; + +static int __init init_hw_breakpoint(void) +{ + return register_die_notifier(&hw_breakpoint_exceptions_nb); +} +core_initcall(init_hw_breakpoint); + + +struct pmu perf_ops_bp = { + .enable = arch_install_hw_breakpoint, + .disable = arch_uninstall_hw_breakpoint, + .read = hw_breakpoint_pmu_read, + .unthrottle = hw_breakpoint_pmu_unthrottle +}; diff --git a/kernel/irq/Makefile b/kernel/irq/Makefile index 3394f8f5296..7d047808419 100644 --- a/kernel/irq/Makefile +++ b/kernel/irq/Makefile @@ -3,5 +3,5 @@ obj-y := handle.o manage.o spurious.o resend.o chip.o devres.o obj-$(CONFIG_GENERIC_IRQ_PROBE) += autoprobe.o obj-$(CONFIG_PROC_FS) += proc.o obj-$(CONFIG_GENERIC_PENDING_IRQ) += migration.o -obj-$(CONFIG_NUMA_MIGRATE_IRQ_DESC) += numa_migrate.o +obj-$(CONFIG_NUMA_IRQ_DESC) += numa_migrate.o obj-$(CONFIG_PM_SLEEP) += pm.o diff --git a/kernel/irq/chip.c b/kernel/irq/chip.c index c687ba4363f..ba566c261ad 100644 --- a/kernel/irq/chip.c +++ b/kernel/irq/chip.c @@ -166,11 +166,11 @@ int set_irq_data(unsigned int irq, void *data) EXPORT_SYMBOL(set_irq_data); /** - * set_irq_data - set irq type data for an irq + * set_irq_msi - set MSI descriptor data for an irq * @irq: Interrupt number * @entry: Pointer to MSI descriptor data * - * Set the hardware irq controller data for an irq + * Set the MSI descriptor entry for an irq */ int set_irq_msi(unsigned int irq, struct msi_desc *entry) { @@ -222,6 +222,34 @@ int set_irq_chip_data(unsigned int irq, void *data) } EXPORT_SYMBOL(set_irq_chip_data); +/** + * set_irq_nested_thread - Set/Reset the IRQ_NESTED_THREAD flag of an irq + * + * @irq: Interrupt number + * @nest: 0 to clear / 1 to set the IRQ_NESTED_THREAD flag + * + * The IRQ_NESTED_THREAD flag indicates that on + * request_threaded_irq() no separate interrupt thread should be + * created for the irq as the handler are called nested in the + * context of a demultiplexing interrupt handler thread. + */ +void set_irq_nested_thread(unsigned int irq, int nest) +{ + struct irq_desc *desc = irq_to_desc(irq); + unsigned long flags; + + if (!desc) + return; + + spin_lock_irqsave(&desc->lock, flags); + if (nest) + desc->status |= IRQ_NESTED_THREAD; + else + desc->status &= ~IRQ_NESTED_THREAD; + spin_unlock_irqrestore(&desc->lock, flags); +} +EXPORT_SYMBOL_GPL(set_irq_nested_thread); + /* * default enable function */ @@ -299,6 +327,45 @@ static inline void mask_ack_irq(struct irq_desc *desc, int irq) } } +/* + * handle_nested_irq - Handle a nested irq from a irq thread + * @irq: the interrupt number + * + * Handle interrupts which are nested into a threaded interrupt + * handler. The handler function is called inside the calling + * threads context. + */ +void handle_nested_irq(unsigned int irq) +{ + struct irq_desc *desc = irq_to_desc(irq); + struct irqaction *action; + irqreturn_t action_ret; + + might_sleep(); + + spin_lock_irq(&desc->lock); + + kstat_incr_irqs_this_cpu(irq, desc); + + action = desc->action; + if (unlikely(!action || (desc->status & IRQ_DISABLED))) + goto out_unlock; + + desc->status |= IRQ_INPROGRESS; + spin_unlock_irq(&desc->lock); + + action_ret = action->thread_fn(action->irq, action->dev_id); + if (!noirqdebug) + note_interrupt(irq, desc, action_ret); + + spin_lock_irq(&desc->lock); + desc->status &= ~IRQ_INPROGRESS; + +out_unlock: + spin_unlock_irq(&desc->lock); +} +EXPORT_SYMBOL_GPL(handle_nested_irq); + /** * handle_simple_irq - Simple and software-decoded IRQs. * @irq: the interrupt number @@ -359,7 +426,6 @@ handle_level_irq(unsigned int irq, struct irq_desc *desc) spin_lock(&desc->lock); mask_ack_irq(desc, irq); - desc = irq_remap_to_desc(irq, desc); if (unlikely(desc->status & IRQ_INPROGRESS)) goto out_unlock; @@ -383,7 +449,10 @@ handle_level_irq(unsigned int irq, struct irq_desc *desc) spin_lock(&desc->lock); desc->status &= ~IRQ_INPROGRESS; - if (!(desc->status & IRQ_DISABLED) && desc->chip->unmask) + + if (unlikely(desc->status & IRQ_ONESHOT)) + desc->status |= IRQ_MASKED; + else if (!(desc->status & IRQ_DISABLED) && desc->chip->unmask) desc->chip->unmask(irq); out_unlock: spin_unlock(&desc->lock); @@ -438,7 +507,6 @@ handle_fasteoi_irq(unsigned int irq, struct irq_desc *desc) desc->status &= ~IRQ_INPROGRESS; out: desc->chip->eoi(irq); - desc = irq_remap_to_desc(irq, desc); spin_unlock(&desc->lock); } @@ -475,7 +543,6 @@ handle_edge_irq(unsigned int irq, struct irq_desc *desc) !desc->action)) { desc->status |= (IRQ_PENDING | IRQ_MASKED); mask_ack_irq(desc, irq); - desc = irq_remap_to_desc(irq, desc); goto out_unlock; } kstat_incr_irqs_this_cpu(irq, desc); @@ -483,7 +550,6 @@ handle_edge_irq(unsigned int irq, struct irq_desc *desc) /* Start handling the irq */ if (desc->chip->ack) desc->chip->ack(irq); - desc = irq_remap_to_desc(irq, desc); /* Mark the IRQ currently in progress.*/ desc->status |= IRQ_INPROGRESS; @@ -524,7 +590,7 @@ out_unlock: } /** - * handle_percpu_IRQ - Per CPU local irq handler + * handle_percpu_irq - Per CPU local irq handler * @irq: the interrupt number * @desc: the interrupt description structure for this irq * @@ -544,10 +610,8 @@ handle_percpu_irq(unsigned int irq, struct irq_desc *desc) if (!noirqdebug) note_interrupt(irq, desc, action_ret); - if (desc->chip->eoi) { + if (desc->chip->eoi) desc->chip->eoi(irq); - desc = irq_remap_to_desc(irq, desc); - } } void @@ -578,14 +642,13 @@ __set_irq_handler(unsigned int irq, irq_flow_handler_t handle, int is_chained, desc->chip = &dummy_irq_chip; } + chip_bus_lock(irq, desc); spin_lock_irqsave(&desc->lock, flags); /* Uninstall? */ if (handle == handle_bad_irq) { - if (desc->chip != &no_irq_chip) { + if (desc->chip != &no_irq_chip) mask_ack_irq(desc, irq); - desc = irq_remap_to_desc(irq, desc); - } desc->status |= IRQ_DISABLED; desc->depth = 1; } @@ -599,6 +662,7 @@ __set_irq_handler(unsigned int irq, irq_flow_handler_t handle, int is_chained, desc->chip->startup(irq); } spin_unlock_irqrestore(&desc->lock, flags); + chip_bus_sync_unlock(irq, desc); } EXPORT_SYMBOL_GPL(__set_irq_handler); diff --git a/kernel/irq/devres.c b/kernel/irq/devres.c index 38a25b8d8bf..d06df9c41cb 100644 --- a/kernel/irq/devres.c +++ b/kernel/irq/devres.c @@ -26,10 +26,12 @@ static int devm_irq_match(struct device *dev, void *res, void *data) } /** - * devm_request_irq - allocate an interrupt line for a managed device + * devm_request_threaded_irq - allocate an interrupt line for a managed device * @dev: device to request interrupt for * @irq: Interrupt line to allocate * @handler: Function to be called when the IRQ occurs + * @thread_fn: function to be called in a threaded interrupt context. NULL + * for devices which handle everything in @handler * @irqflags: Interrupt type flags * @devname: An ascii name for the claiming device * @dev_id: A cookie passed back to the handler function @@ -42,9 +44,10 @@ static int devm_irq_match(struct device *dev, void *res, void *data) * If an IRQ allocated with this function needs to be freed * separately, dev_free_irq() must be used. */ -int devm_request_irq(struct device *dev, unsigned int irq, - irq_handler_t handler, unsigned long irqflags, - const char *devname, void *dev_id) +int devm_request_threaded_irq(struct device *dev, unsigned int irq, + irq_handler_t handler, irq_handler_t thread_fn, + unsigned long irqflags, const char *devname, + void *dev_id) { struct irq_devres *dr; int rc; @@ -54,7 +57,8 @@ int devm_request_irq(struct device *dev, unsigned int irq, if (!dr) return -ENOMEM; - rc = request_irq(irq, handler, irqflags, devname, dev_id); + rc = request_threaded_irq(irq, handler, thread_fn, irqflags, devname, + dev_id); if (rc) { devres_free(dr); return rc; @@ -66,7 +70,7 @@ int devm_request_irq(struct device *dev, unsigned int irq, return 0; } -EXPORT_SYMBOL(devm_request_irq); +EXPORT_SYMBOL(devm_request_threaded_irq); /** * devm_free_irq - free an interrupt diff --git a/kernel/irq/handle.c b/kernel/irq/handle.c index 9ebf7796887..17c71bb565c 100644 --- a/kernel/irq/handle.c +++ b/kernel/irq/handle.c @@ -11,6 +11,8 @@ */ #include <linux/irq.h> +#include <linux/sched.h> +#include <linux/slab.h> #include <linux/module.h> #include <linux/random.h> #include <linux/interrupt.h> @@ -18,6 +20,7 @@ #include <linux/rculist.h> #include <linux/hash.h> #include <linux/bootmem.h> +#include <trace/events/irq.h> #include "internals.h" @@ -43,7 +46,7 @@ void handle_bad_irq(unsigned int irq, struct irq_desc *desc) #if defined(CONFIG_SMP) && defined(CONFIG_GENERIC_HARDIRQS) static void __init init_irq_default_affinity(void) { - alloc_bootmem_cpumask_var(&irq_default_affinity); + alloc_cpumask_var(&irq_default_affinity, GFP_NOWAIT); cpumask_setall(irq_default_affinity); } #else @@ -80,45 +83,48 @@ static struct irq_desc irq_desc_init = { .lock = __SPIN_LOCK_UNLOCKED(irq_desc_init.lock), }; -void init_kstat_irqs(struct irq_desc *desc, int cpu, int nr) +void __ref init_kstat_irqs(struct irq_desc *desc, int node, int nr) { - int node; void *ptr; - node = cpu_to_node(cpu); - ptr = kzalloc_node(nr * sizeof(*desc->kstat_irqs), GFP_ATOMIC, node); + if (slab_is_available()) + ptr = kzalloc_node(nr * sizeof(*desc->kstat_irqs), + GFP_ATOMIC, node); + else + ptr = alloc_bootmem_node(NODE_DATA(node), + nr * sizeof(*desc->kstat_irqs)); /* * don't overwite if can not get new one * init_copy_kstat_irqs() could still use old one */ if (ptr) { - printk(KERN_DEBUG " alloc kstat_irqs on cpu %d node %d\n", - cpu, node); + printk(KERN_DEBUG " alloc kstat_irqs on node %d\n", node); desc->kstat_irqs = ptr; } } -static void init_one_irq_desc(int irq, struct irq_desc *desc, int cpu) +static void init_one_irq_desc(int irq, struct irq_desc *desc, int node) { memcpy(desc, &irq_desc_init, sizeof(struct irq_desc)); spin_lock_init(&desc->lock); desc->irq = irq; #ifdef CONFIG_SMP - desc->cpu = cpu; + desc->node = node; #endif lockdep_set_class(&desc->lock, &irq_desc_lock_class); - init_kstat_irqs(desc, cpu, nr_cpu_ids); + init_kstat_irqs(desc, node, nr_cpu_ids); if (!desc->kstat_irqs) { printk(KERN_ERR "can not alloc kstat_irqs\n"); BUG_ON(1); } - if (!init_alloc_desc_masks(desc, cpu, false)) { + if (!alloc_desc_masks(desc, node, false)) { printk(KERN_ERR "can not alloc irq_desc cpumasks\n"); BUG_ON(1); } - arch_init_chip_data(desc, cpu); + init_desc_masks(desc); + arch_init_chip_data(desc, node); } /* @@ -145,6 +151,7 @@ int __init early_irq_init(void) { struct irq_desc *desc; int legacy_count; + int node; int i; init_irq_default_affinity(); @@ -155,20 +162,24 @@ int __init early_irq_init(void) desc = irq_desc_legacy; legacy_count = ARRAY_SIZE(irq_desc_legacy); + node = first_online_node; /* allocate irq_desc_ptrs array based on nr_irqs */ - irq_desc_ptrs = alloc_bootmem(nr_irqs * sizeof(void *)); + irq_desc_ptrs = kcalloc(nr_irqs, sizeof(void *), GFP_NOWAIT); /* allocate based on nr_cpu_ids */ - /* FIXME: invert kstat_irgs, and it'd be a per_cpu_alloc'd thing */ - kstat_irqs_legacy = alloc_bootmem(NR_IRQS_LEGACY * nr_cpu_ids * - sizeof(int)); + kstat_irqs_legacy = kzalloc_node(NR_IRQS_LEGACY * nr_cpu_ids * + sizeof(int), GFP_NOWAIT, node); for (i = 0; i < legacy_count; i++) { desc[i].irq = i; +#ifdef CONFIG_SMP + desc[i].node = node; +#endif desc[i].kstat_irqs = kstat_irqs_legacy + i * nr_cpu_ids; lockdep_set_class(&desc[i].lock, &irq_desc_lock_class); - init_alloc_desc_masks(&desc[i], 0, true); + alloc_desc_masks(&desc[i], node, true); + init_desc_masks(&desc[i]); irq_desc_ptrs[i] = desc + i; } @@ -186,11 +197,10 @@ struct irq_desc *irq_to_desc(unsigned int irq) return NULL; } -struct irq_desc *irq_to_desc_alloc_cpu(unsigned int irq, int cpu) +struct irq_desc * __ref irq_to_desc_alloc_node(unsigned int irq, int node) { struct irq_desc *desc; unsigned long flags; - int node; if (irq >= nr_irqs) { WARN(1, "irq (%d) >= nr_irqs (%d) in irq_to_desc_alloc\n", @@ -209,15 +219,17 @@ struct irq_desc *irq_to_desc_alloc_cpu(unsigned int irq, int cpu) if (desc) goto out_unlock; - node = cpu_to_node(cpu); - desc = kzalloc_node(sizeof(*desc), GFP_ATOMIC, node); - printk(KERN_DEBUG " alloc irq_desc for %d on cpu %d node %d\n", - irq, cpu, node); + if (slab_is_available()) + desc = kzalloc_node(sizeof(*desc), GFP_ATOMIC, node); + else + desc = alloc_bootmem_node(NODE_DATA(node), sizeof(*desc)); + + printk(KERN_DEBUG " alloc irq_desc for %d on node %d\n", irq, node); if (!desc) { printk(KERN_ERR "can not alloc irq_desc\n"); BUG_ON(1); } - init_one_irq_desc(irq, desc, cpu); + init_one_irq_desc(irq, desc, node); irq_desc_ptrs[irq] = desc; @@ -255,7 +267,8 @@ int __init early_irq_init(void) for (i = 0; i < count; i++) { desc[i].irq = i; - init_alloc_desc_masks(&desc[i], 0, true); + alloc_desc_masks(&desc[i], 0, true); + init_desc_masks(&desc[i]); desc[i].kstat_irqs = kstat_irqs_all[i]; } return arch_early_irq_init(); @@ -266,7 +279,7 @@ struct irq_desc *irq_to_desc(unsigned int irq) return (irq < NR_IRQS) ? irq_desc + irq : NULL; } -struct irq_desc *irq_to_desc_alloc_cpu(unsigned int irq, int cpu) +struct irq_desc *irq_to_desc_alloc_node(unsigned int irq, int node) { return irq_to_desc(irq); } @@ -338,6 +351,15 @@ irqreturn_t no_action(int cpl, void *dev_id) return IRQ_NONE; } +static void warn_no_thread(unsigned int irq, struct irqaction *action) +{ + if (test_and_set_bit(IRQTF_WARNED, &action->thread_flags)) + return; + + printk(KERN_WARNING "IRQ %d device %s returned IRQ_WAKE_THREAD " + "but no thread function available.", irq, action->name); +} + /** * handle_IRQ_event - irq action chain handler * @irq: the interrupt number @@ -350,15 +372,54 @@ irqreturn_t handle_IRQ_event(unsigned int irq, struct irqaction *action) irqreturn_t ret, retval = IRQ_NONE; unsigned int status = 0; - WARN_ONCE(!in_irq(), "BUG: IRQ handler called from non-hardirq context!"); - if (!(action->flags & IRQF_DISABLED)) local_irq_enable_in_hardirq(); do { + trace_irq_handler_entry(irq, action); ret = action->handler(irq, action->dev_id); - if (ret == IRQ_HANDLED) + trace_irq_handler_exit(irq, action, ret); + + switch (ret) { + case IRQ_WAKE_THREAD: + /* + * Set result to handled so the spurious check + * does not trigger. + */ + ret = IRQ_HANDLED; + + /* + * Catch drivers which return WAKE_THREAD but + * did not set up a thread function + */ + if (unlikely(!action->thread_fn)) { + warn_no_thread(irq, action); + break; + } + + /* + * Wake up the handler thread for this + * action. In case the thread crashed and was + * killed we just pretend that we handled the + * interrupt. The hardirq handler above has + * disabled the device interrupt, so no irq + * storm is lurking. + */ + if (likely(!test_bit(IRQTF_DIED, + &action->thread_flags))) { + set_bit(IRQTF_RUNTHREAD, &action->thread_flags); + wake_up_process(action->thread); + } + + /* Fall through to add to randomness */ + case IRQ_HANDLED: status |= action->flags; + break; + + default: + break; + } + retval |= ret; action = action->next; } while (action); @@ -401,11 +462,8 @@ unsigned int __do_IRQ(unsigned int irq) /* * No locking required for CPU-local interrupts: */ - if (desc->chip->ack) { + if (desc->chip->ack) desc->chip->ack(irq); - /* get new one */ - desc = irq_remap_to_desc(irq, desc); - } if (likely(!(desc->status & IRQ_DISABLED))) { action_ret = handle_IRQ_event(irq, desc->action); if (!noirqdebug) @@ -416,10 +474,8 @@ unsigned int __do_IRQ(unsigned int irq) } spin_lock(&desc->lock); - if (desc->chip->ack) { + if (desc->chip->ack) desc->chip->ack(irq); - desc = irq_remap_to_desc(irq, desc); - } /* * REPLAY is when Linux resends an IRQ that was dropped earlier * WAITING is used by probe to mark irqs that are being tested diff --git a/kernel/irq/internals.h b/kernel/irq/internals.h index 01ce20eab38..1b5d742c6a7 100644 --- a/kernel/irq/internals.h +++ b/kernel/irq/internals.h @@ -16,7 +16,7 @@ extern void __disable_irq(struct irq_desc *desc, unsigned int irq, bool susp); extern void __enable_irq(struct irq_desc *desc, unsigned int irq, bool resume); extern struct lock_class_key irq_desc_lock_class; -extern void init_kstat_irqs(struct irq_desc *desc, int cpu, int nr); +extern void init_kstat_irqs(struct irq_desc *desc, int node, int nr); extern void clear_kstat_irqs(struct irq_desc *desc); extern spinlock_t sparse_irq_lock; @@ -42,6 +42,21 @@ static inline void unregister_handler_proc(unsigned int irq, extern int irq_select_affinity_usr(unsigned int irq); +extern void irq_set_thread_affinity(struct irq_desc *desc); + +/* Inline functions for support of irq chips on slow busses */ +static inline void chip_bus_lock(unsigned int irq, struct irq_desc *desc) +{ + if (unlikely(desc->chip->bus_lock)) + desc->chip->bus_lock(irq); +} + +static inline void chip_bus_sync_unlock(unsigned int irq, struct irq_desc *desc) +{ + if (unlikely(desc->chip->bus_sync_unlock)) + desc->chip->bus_sync_unlock(irq); +} + /* * Debugging printout: */ diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c index 8c68d5b95d4..7305b297d1e 100644 --- a/kernel/irq/manage.c +++ b/kernel/irq/manage.c @@ -8,16 +8,15 @@ */ #include <linux/irq.h> +#include <linux/kthread.h> #include <linux/module.h> #include <linux/random.h> #include <linux/interrupt.h> #include <linux/slab.h> +#include <linux/sched.h> #include "internals.h" -#if defined(CONFIG_SMP) && defined(CONFIG_GENERIC_HARDIRQS) -cpumask_var_t irq_default_affinity; - /** * synchronize_irq - wait for pending IRQ handlers (on other CPUs) * @irq: interrupt number to wait for @@ -53,9 +52,18 @@ void synchronize_irq(unsigned int irq) /* Oops, that failed? */ } while (status & IRQ_INPROGRESS); + + /* + * We made sure that no hardirq handler is running. Now verify + * that no threaded handlers are active. + */ + wait_event(desc->wait_for_threads, !atomic_read(&desc->threads_active)); } EXPORT_SYMBOL(synchronize_irq); +#ifdef CONFIG_SMP +cpumask_var_t irq_default_affinity; + /** * irq_can_set_affinity - Check if the affinity of a given irq can be set * @irq: Interrupt to check @@ -73,6 +81,26 @@ int irq_can_set_affinity(unsigned int irq) } /** + * irq_set_thread_affinity - Notify irq threads to adjust affinity + * @desc: irq descriptor which has affitnity changed + * + * We just set IRQTF_AFFINITY and delegate the affinity setting + * to the interrupt thread itself. We can not call + * set_cpus_allowed_ptr() here as we hold desc->lock and this + * code can be called from hard interrupt context. + */ +void irq_set_thread_affinity(struct irq_desc *desc) +{ + struct irqaction *action = desc->action; + + while (action) { + if (action->thread) + set_bit(IRQTF_AFFINITY, &action->thread_flags); + action = action->next; + } +} + +/** * irq_set_affinity - Set the irq affinity of a given irq * @irq: Interrupt to set affinity * @cpumask: cpumask @@ -89,16 +117,21 @@ int irq_set_affinity(unsigned int irq, const struct cpumask *cpumask) spin_lock_irqsave(&desc->lock, flags); #ifdef CONFIG_GENERIC_PENDING_IRQ - if (desc->status & IRQ_MOVE_PCNTXT || desc->status & IRQ_DISABLED) { - cpumask_copy(desc->affinity, cpumask); - desc->chip->set_affinity(irq, cpumask); - } else { + if (desc->status & IRQ_MOVE_PCNTXT) { + if (!desc->chip->set_affinity(irq, cpumask)) { + cpumask_copy(desc->affinity, cpumask); + irq_set_thread_affinity(desc); + } + } + else { desc->status |= IRQ_MOVE_PENDING; cpumask_copy(desc->pending_mask, cpumask); } #else - cpumask_copy(desc->affinity, cpumask); - desc->chip->set_affinity(irq, cpumask); + if (!desc->chip->set_affinity(irq, cpumask)) { + cpumask_copy(desc->affinity, cpumask); + irq_set_thread_affinity(desc); + } #endif desc->status |= IRQ_AFFINITY_SET; spin_unlock_irqrestore(&desc->lock, flags); @@ -150,6 +183,8 @@ int irq_select_affinity_usr(unsigned int irq) spin_lock_irqsave(&desc->lock, flags); ret = setup_affinity(irq, desc); + if (!ret) + irq_set_thread_affinity(desc); spin_unlock_irqrestore(&desc->lock, flags); return ret; @@ -195,9 +230,11 @@ void disable_irq_nosync(unsigned int irq) if (!desc) return; + chip_bus_lock(irq, desc); spin_lock_irqsave(&desc->lock, flags); __disable_irq(desc, irq, false); spin_unlock_irqrestore(&desc->lock, flags); + chip_bus_sync_unlock(irq, desc); } EXPORT_SYMBOL(disable_irq_nosync); @@ -259,7 +296,8 @@ void __enable_irq(struct irq_desc *desc, unsigned int irq, bool resume) * matches the last disable, processing of interrupts on this * IRQ line is re-enabled. * - * This function may be called from IRQ context. + * This function may be called from IRQ context only when + * desc->chip->bus_lock and desc->chip->bus_sync_unlock are NULL ! */ void enable_irq(unsigned int irq) { @@ -269,9 +307,11 @@ void enable_irq(unsigned int irq) if (!desc) return; + chip_bus_lock(irq, desc); spin_lock_irqsave(&desc->lock, flags); __enable_irq(desc, irq, false); spin_unlock_irqrestore(&desc->lock, flags); + chip_bus_sync_unlock(irq, desc); } EXPORT_SYMBOL(enable_irq); @@ -402,6 +442,165 @@ int __irq_set_trigger(struct irq_desc *desc, unsigned int irq, } /* + * Default primary interrupt handler for threaded interrupts. Is + * assigned as primary handler when request_threaded_irq is called + * with handler == NULL. Useful for oneshot interrupts. + */ +static irqreturn_t irq_default_primary_handler(int irq, void *dev_id) +{ + return IRQ_WAKE_THREAD; +} + +/* + * Primary handler for nested threaded interrupts. Should never be + * called. + */ +static irqreturn_t irq_nested_primary_handler(int irq, void *dev_id) +{ + WARN(1, "Primary handler called for nested irq %d\n", irq); + return IRQ_NONE; +} + +static int irq_wait_for_interrupt(struct irqaction *action) +{ + while (!kthread_should_stop()) { + set_current_state(TASK_INTERRUPTIBLE); + + if (test_and_clear_bit(IRQTF_RUNTHREAD, + &action->thread_flags)) { + __set_current_state(TASK_RUNNING); + return 0; + } + schedule(); + } + return -1; +} + +/* + * Oneshot interrupts keep the irq line masked until the threaded + * handler finished. unmask if the interrupt has not been disabled and + * is marked MASKED. + */ +static void irq_finalize_oneshot(unsigned int irq, struct irq_desc *desc) +{ + chip_bus_lock(irq, desc); + spin_lock_irq(&desc->lock); + if (!(desc->status & IRQ_DISABLED) && (desc->status & IRQ_MASKED)) { + desc->status &= ~IRQ_MASKED; + desc->chip->unmask(irq); + } + spin_unlock_irq(&desc->lock); + chip_bus_sync_unlock(irq, desc); +} + +#ifdef CONFIG_SMP +/* + * Check whether we need to change the affinity of the interrupt thread. + */ +static void +irq_thread_check_affinity(struct irq_desc *desc, struct irqaction *action) +{ + cpumask_var_t mask; + + if (!test_and_clear_bit(IRQTF_AFFINITY, &action->thread_flags)) + return; + + /* + * In case we are out of memory we set IRQTF_AFFINITY again and + * try again next time + */ + if (!alloc_cpumask_var(&mask, GFP_KERNEL)) { + set_bit(IRQTF_AFFINITY, &action->thread_flags); + return; + } + + spin_lock_irq(&desc->lock); + cpumask_copy(mask, desc->affinity); + spin_unlock_irq(&desc->lock); + + set_cpus_allowed_ptr(current, mask); + free_cpumask_var(mask); +} +#else +static inline void +irq_thread_check_affinity(struct irq_desc *desc, struct irqaction *action) { } +#endif + +/* + * Interrupt handler thread + */ +static int irq_thread(void *data) +{ + struct sched_param param = { .sched_priority = MAX_USER_RT_PRIO/2, }; + struct irqaction *action = data; + struct irq_desc *desc = irq_to_desc(action->irq); + int wake, oneshot = desc->status & IRQ_ONESHOT; + + sched_setscheduler(current, SCHED_FIFO, ¶m); + current->irqaction = action; + + while (!irq_wait_for_interrupt(action)) { + + irq_thread_check_affinity(desc, action); + + atomic_inc(&desc->threads_active); + + spin_lock_irq(&desc->lock); + if (unlikely(desc->status & IRQ_DISABLED)) { + /* + * CHECKME: We might need a dedicated + * IRQ_THREAD_PENDING flag here, which + * retriggers the thread in check_irq_resend() + * but AFAICT IRQ_PENDING should be fine as it + * retriggers the interrupt itself --- tglx + */ + desc->status |= IRQ_PENDING; + spin_unlock_irq(&desc->lock); + } else { + spin_unlock_irq(&desc->lock); + + action->thread_fn(action->irq, action->dev_id); + + if (oneshot) + irq_finalize_oneshot(action->irq, desc); + } + + wake = atomic_dec_and_test(&desc->threads_active); + + if (wake && waitqueue_active(&desc->wait_for_threads)) + wake_up(&desc->wait_for_threads); + } + + /* + * Clear irqaction. Otherwise exit_irq_thread() would make + * fuzz about an active irq thread going into nirvana. + */ + current->irqaction = NULL; + return 0; +} + +/* + * Called from do_exit() + */ +void exit_irq_thread(void) +{ + struct task_struct *tsk = current; + + if (!tsk->irqaction) + return; + + printk(KERN_ERR + "exiting task \"%s\" (%d) is an active IRQ thread (irq %d)\n", + tsk->comm ? tsk->comm : "", tsk->pid, tsk->irqaction->irq); + + /* + * Set the THREAD DIED flag to prevent further wakeups of the + * soon to be gone threaded handler. + */ + set_bit(IRQTF_DIED, &tsk->irqaction->flags); +} + +/* * Internal function to register an irqaction - typically used to * allocate special interrupts that are part of the architecture. */ @@ -411,7 +610,7 @@ __setup_irq(unsigned int irq, struct irq_desc *desc, struct irqaction *new) struct irqaction *old, **old_ptr; const char *old_name = NULL; unsigned long flags; - int shared = 0; + int nested, shared = 0; int ret; if (!desc) @@ -436,6 +635,47 @@ __setup_irq(unsigned int irq, struct irq_desc *desc, struct irqaction *new) rand_initialize_irq(irq); } + /* Oneshot interrupts are not allowed with shared */ + if ((new->flags & IRQF_ONESHOT) && (new->flags & IRQF_SHARED)) + return -EINVAL; + + /* + * Check whether the interrupt nests into another interrupt + * thread. + */ + nested = desc->status & IRQ_NESTED_THREAD; + if (nested) { + if (!new->thread_fn) + return -EINVAL; + /* + * Replace the primary handler which was provided from + * the driver for non nested interrupt handling by the + * dummy function which warns when called. + */ + new->handler = irq_nested_primary_handler; + } + + /* + * Create a handler thread when a thread function is supplied + * and the interrupt does not nest into another interrupt + * thread. + */ + if (new->thread_fn && !nested) { + struct task_struct *t; + + t = kthread_create(irq_thread, new, "irq/%d-%s", irq, + new->name); + if (IS_ERR(t)) + return PTR_ERR(t); + /* + * We keep the reference to the task struct even if + * the thread dies to avoid that the interrupt code + * references an already freed task_struct. + */ + get_task_struct(t); + new->thread = t; + } + /* * The following block of code has to be executed atomically */ @@ -473,15 +713,15 @@ __setup_irq(unsigned int irq, struct irq_desc *desc, struct irqaction *new) if (!shared) { irq_chip_set_defaults(desc->chip); + init_waitqueue_head(&desc->wait_for_threads); + /* Setup the type (level, edge polarity) if configured: */ if (new->flags & IRQF_TRIGGER_MASK) { ret = __irq_set_trigger(desc, irq, new->flags & IRQF_TRIGGER_MASK); - if (ret) { - spin_unlock_irqrestore(&desc->lock, flags); - return ret; - } + if (ret) + goto out_thread; } else compat_irq_chip_set_default_handler(desc); #if defined(CONFIG_IRQ_PER_CPU) @@ -489,9 +729,12 @@ __setup_irq(unsigned int irq, struct irq_desc *desc, struct irqaction *new) desc->status |= IRQ_PER_CPU; #endif - desc->status &= ~(IRQ_AUTODETECT | IRQ_WAITING | + desc->status &= ~(IRQ_AUTODETECT | IRQ_WAITING | IRQ_ONESHOT | IRQ_INPROGRESS | IRQ_SPURIOUS_DISABLED); + if (new->flags & IRQF_ONESHOT) + desc->status |= IRQ_ONESHOT; + if (!(desc->status & IRQ_NOAUTOEN)) { desc->depth = 0; desc->status &= ~IRQ_DISABLED; @@ -516,6 +759,7 @@ __setup_irq(unsigned int irq, struct irq_desc *desc, struct irqaction *new) (int)(new->flags & IRQF_TRIGGER_MASK)); } + new->irq = irq; *old_ptr = new; /* Reset broken irq detection when installing new handler */ @@ -533,7 +777,13 @@ __setup_irq(unsigned int irq, struct irq_desc *desc, struct irqaction *new) spin_unlock_irqrestore(&desc->lock, flags); - new->irq = irq; + /* + * Strictly no need to wake it up, but hung_task complains + * when no hard interrupt wakes the thread up. + */ + if (new->thread) + wake_up_process(new->thread); + register_irq_proc(irq, desc); new->dir = NULL; register_handler_proc(irq, new); @@ -549,8 +799,19 @@ mismatch: dump_stack(); } #endif + ret = -EBUSY; + +out_thread: spin_unlock_irqrestore(&desc->lock, flags); - return -EBUSY; + if (new->thread) { + struct task_struct *t = new->thread; + + new->thread = NULL; + if (likely(!test_bit(IRQTF_DIED, &new->thread_flags))) + kthread_stop(t); + put_task_struct(t); + } + return ret; } /** @@ -622,6 +883,7 @@ static struct irqaction *__free_irq(unsigned int irq, void *dev_id) else desc->chip->disable(irq); } + spin_unlock_irqrestore(&desc->lock, flags); unregister_handler_proc(irq, action); @@ -644,6 +906,13 @@ static struct irqaction *__free_irq(unsigned int irq, void *dev_id) local_irq_restore(flags); } #endif + + if (action->thread) { + if (!test_bit(IRQTF_DIED, &action->thread_flags)) + kthread_stop(action->thread); + put_task_struct(action->thread); + } + return action; } @@ -676,14 +945,26 @@ EXPORT_SYMBOL_GPL(remove_irq); */ void free_irq(unsigned int irq, void *dev_id) { + struct irq_desc *desc = irq_to_desc(irq); + + if (!desc) + return; + + chip_bus_lock(irq, desc); kfree(__free_irq(irq, dev_id)); + chip_bus_sync_unlock(irq, desc); } EXPORT_SYMBOL(free_irq); /** - * request_irq - allocate an interrupt line + * request_threaded_irq - allocate an interrupt line * @irq: Interrupt line to allocate - * @handler: Function to be called when the IRQ occurs + * @handler: Function to be called when the IRQ occurs. + * Primary handler for threaded interrupts + * If NULL and thread_fn != NULL the default + * primary handler is installed + * @thread_fn: Function called from the irq handler thread + * If NULL, no irq thread is created * @irqflags: Interrupt type flags * @devname: An ascii name for the claiming device * @dev_id: A cookie passed back to the handler function @@ -695,6 +976,15 @@ EXPORT_SYMBOL(free_irq); * raises, you must take care both to initialise your hardware * and to set up the interrupt handler in the right order. * + * If you want to set up a threaded irq handler for your device + * then you need to supply @handler and @thread_fn. @handler ist + * still called in hard interrupt context and has to check + * whether the interrupt originates from the device. If yes it + * needs to disable the interrupt on the device and return + * IRQ_WAKE_THREAD which will wake up the handler thread and run + * @thread_fn. This split handler design is necessary to support + * shared interrupts. + * * Dev_id must be globally unique. Normally the address of the * device data structure is used as the cookie. Since the handler * receives this value it makes sense to use it. @@ -710,8 +1000,9 @@ EXPORT_SYMBOL(free_irq); * IRQF_TRIGGER_* Specify active edge(s) or level * */ -int request_irq(unsigned int irq, irq_handler_t handler, - unsigned long irqflags, const char *devname, void *dev_id) +int request_threaded_irq(unsigned int irq, irq_handler_t handler, + irq_handler_t thread_fn, unsigned long irqflags, + const char *devname, void *dev_id) { struct irqaction *action; struct irq_desc *desc; @@ -751,19 +1042,27 @@ int request_irq(unsigned int irq, irq_handler_t handler, if (desc->status & IRQ_NOREQUEST) return -EINVAL; - if (!handler) - return -EINVAL; + + if (!handler) { + if (!thread_fn) + return -EINVAL; + handler = irq_default_primary_handler; + } action = kzalloc(sizeof(struct irqaction), GFP_KERNEL); if (!action) return -ENOMEM; action->handler = handler; + action->thread_fn = thread_fn; action->flags = irqflags; action->name = devname; action->dev_id = dev_id; + chip_bus_lock(irq, desc); retval = __setup_irq(irq, desc, action); + chip_bus_sync_unlock(irq, desc); + if (retval) kfree(action); @@ -788,4 +1087,4 @@ int request_irq(unsigned int irq, irq_handler_t handler, #endif return retval; } -EXPORT_SYMBOL(request_irq); +EXPORT_SYMBOL(request_threaded_irq); diff --git a/kernel/irq/migration.c b/kernel/irq/migration.c index e05ad9be43b..fcb6c96f262 100644 --- a/kernel/irq/migration.c +++ b/kernel/irq/migration.c @@ -1,5 +1,8 @@ #include <linux/irq.h> +#include <linux/interrupt.h> + +#include "internals.h" void move_masked_irq(int irq) { @@ -39,11 +42,12 @@ void move_masked_irq(int irq) * masking the irqs. */ if (likely(cpumask_any_and(desc->pending_mask, cpu_online_mask) - < nr_cpu_ids)) { - cpumask_and(desc->affinity, - desc->pending_mask, cpu_online_mask); - desc->chip->set_affinity(irq, desc->affinity); - } + < nr_cpu_ids)) + if (!desc->chip->set_affinity(irq, desc->pending_mask)) { + cpumask_copy(desc->affinity, desc->pending_mask); + irq_set_thread_affinity(desc); + } + cpumask_clear(desc->pending_mask); } diff --git a/kernel/irq/numa_migrate.c b/kernel/irq/numa_migrate.c index 243d6121e50..3fd30197da2 100644 --- a/kernel/irq/numa_migrate.c +++ b/kernel/irq/numa_migrate.c @@ -15,9 +15,9 @@ static void init_copy_kstat_irqs(struct irq_desc *old_desc, struct irq_desc *desc, - int cpu, int nr) + int node, int nr) { - init_kstat_irqs(desc, cpu, nr); + init_kstat_irqs(desc, node, nr); if (desc->kstat_irqs != old_desc->kstat_irqs) memcpy(desc->kstat_irqs, old_desc->kstat_irqs, @@ -34,36 +34,36 @@ static void free_kstat_irqs(struct irq_desc *old_desc, struct irq_desc *desc) } static bool init_copy_one_irq_desc(int irq, struct irq_desc *old_desc, - struct irq_desc *desc, int cpu) + struct irq_desc *desc, int node) { memcpy(desc, old_desc, sizeof(struct irq_desc)); - if (!init_alloc_desc_masks(desc, cpu, false)) { + if (!alloc_desc_masks(desc, node, false)) { printk(KERN_ERR "irq %d: can not get new irq_desc cpumask " "for migration.\n", irq); return false; } spin_lock_init(&desc->lock); - desc->cpu = cpu; + desc->node = node; lockdep_set_class(&desc->lock, &irq_desc_lock_class); - init_copy_kstat_irqs(old_desc, desc, cpu, nr_cpu_ids); + init_copy_kstat_irqs(old_desc, desc, node, nr_cpu_ids); init_copy_desc_masks(old_desc, desc); - arch_init_copy_chip_data(old_desc, desc, cpu); + arch_init_copy_chip_data(old_desc, desc, node); return true; } static void free_one_irq_desc(struct irq_desc *old_desc, struct irq_desc *desc) { free_kstat_irqs(old_desc, desc); + free_desc_masks(old_desc, desc); arch_free_chip_data(old_desc, desc); } static struct irq_desc *__real_move_irq_desc(struct irq_desc *old_desc, - int cpu) + int node) { struct irq_desc *desc; unsigned int irq; unsigned long flags; - int node; irq = old_desc->irq; @@ -75,7 +75,6 @@ static struct irq_desc *__real_move_irq_desc(struct irq_desc *old_desc, if (desc && old_desc != desc) goto out_unlock; - node = cpu_to_node(cpu); desc = kzalloc_node(sizeof(*desc), GFP_ATOMIC, node); if (!desc) { printk(KERN_ERR "irq %d: can not get new irq_desc " @@ -84,7 +83,7 @@ static struct irq_desc *__real_move_irq_desc(struct irq_desc *old_desc, desc = old_desc; goto out_unlock; } - if (!init_copy_one_irq_desc(irq, old_desc, desc, cpu)) { + if (!init_copy_one_irq_desc(irq, old_desc, desc, node)) { /* still use old one */ kfree(desc); desc = old_desc; @@ -96,9 +95,7 @@ static struct irq_desc *__real_move_irq_desc(struct irq_desc *old_desc, /* free the old one */ free_one_irq_desc(old_desc, desc); - spin_unlock(&old_desc->lock); kfree(old_desc); - spin_lock(&desc->lock); return desc; @@ -108,24 +105,14 @@ out_unlock: return desc; } -struct irq_desc *move_irq_desc(struct irq_desc *desc, int cpu) +struct irq_desc *move_irq_desc(struct irq_desc *desc, int node) { - int old_cpu; - int node, old_node; - - /* those all static, do move them */ - if (desc->irq < NR_IRQS_LEGACY) + /* those static or target node is -1, do not move them */ + if (desc->irq < NR_IRQS_LEGACY || node == -1) return desc; - old_cpu = desc->cpu; - if (old_cpu != cpu) { - node = cpu_to_node(cpu); - old_node = cpu_to_node(old_cpu); - if (old_node != node) - desc = __real_move_irq_desc(desc, cpu); - else - desc->cpu = cpu; - } + if (desc->node != node) + desc = __real_move_irq_desc(desc, node); return desc; } diff --git a/kernel/irq/pm.c b/kernel/irq/pm.c index 638d8bedec1..a0bb09e7986 100644 --- a/kernel/irq/pm.c +++ b/kernel/irq/pm.c @@ -15,10 +15,10 @@ /** * suspend_device_irqs - disable all currently enabled interrupt lines * - * During system-wide suspend or hibernation device interrupts need to be - * disabled at the chip level and this function is provided for this purpose. - * It disables all interrupt lines that are enabled at the moment and sets the - * IRQ_SUSPENDED flag for them. + * During system-wide suspend or hibernation device drivers need to be prevented + * from receiving interrupts and this function is provided for this purpose. + * It marks all interrupt lines in use, except for the timer ones, as disabled + * and sets the IRQ_SUSPENDED flag for each of them. */ void suspend_device_irqs(void) { diff --git a/kernel/irq/proc.c b/kernel/irq/proc.c index 692363dd591..0832145fea9 100644 --- a/kernel/irq/proc.c +++ b/kernel/irq/proc.c @@ -136,7 +136,7 @@ out: static int default_affinity_open(struct inode *inode, struct file *file) { - return single_open(file, default_affinity_show, NULL); + return single_open(file, default_affinity_show, PDE(inode)->data); } static const struct file_operations default_affinity_proc_fops = { @@ -148,18 +148,28 @@ static const struct file_operations default_affinity_proc_fops = { }; #endif -static int irq_spurious_read(char *page, char **start, off_t off, - int count, int *eof, void *data) +static int irq_spurious_proc_show(struct seq_file *m, void *v) { - struct irq_desc *desc = irq_to_desc((long) data); - return sprintf(page, "count %u\n" - "unhandled %u\n" - "last_unhandled %u ms\n", - desc->irq_count, - desc->irqs_unhandled, - jiffies_to_msecs(desc->last_unhandled)); + struct irq_desc *desc = irq_to_desc((long) m->private); + + seq_printf(m, "count %u\n" "unhandled %u\n" "last_unhandled %u ms\n", + desc->irq_count, desc->irqs_unhandled, + jiffies_to_msecs(desc->last_unhandled)); + return 0; +} + +static int irq_spurious_proc_open(struct inode *inode, struct file *file) +{ + return single_open(file, irq_spurious_proc_show, NULL); } +static const struct file_operations irq_spurious_proc_fops = { + .open = irq_spurious_proc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; + #define MAX_NAMELEN 128 static int name_unique(unsigned int irq, struct irqaction *new_action) @@ -204,7 +214,6 @@ void register_handler_proc(unsigned int irq, struct irqaction *action) void register_irq_proc(unsigned int irq, struct irq_desc *desc) { char name [MAX_NAMELEN]; - struct proc_dir_entry *entry; if (!root_irq_dir || (desc->chip == &no_irq_chip) || desc->dir) return; @@ -214,6 +223,8 @@ void register_irq_proc(unsigned int irq, struct irq_desc *desc) /* create /proc/irq/1234 */ desc->dir = proc_mkdir(name, root_irq_dir); + if (!desc->dir) + return; #ifdef CONFIG_SMP /* create /proc/irq/<irq>/smp_affinity */ @@ -221,11 +232,8 @@ void register_irq_proc(unsigned int irq, struct irq_desc *desc) &irq_affinity_proc_fops, (void *)(long)irq); #endif - entry = create_proc_entry("spurious", 0444, desc->dir); - if (entry) { - entry->data = (void *)(long)irq; - entry->read_proc = irq_spurious_read; - } + proc_create_data("spurious", 0444, desc->dir, + &irq_spurious_proc_fops, (void *)(long)irq); } #undef MAX_NAMELEN diff --git a/kernel/irq/resend.c b/kernel/irq/resend.c index 89c7117acf2..090c3763f3a 100644 --- a/kernel/irq/resend.c +++ b/kernel/irq/resend.c @@ -70,8 +70,7 @@ void check_irq_resend(struct irq_desc *desc, unsigned int irq) if ((status & (IRQ_LEVEL | IRQ_PENDING | IRQ_REPLAY)) == IRQ_PENDING) { desc->status = (status & ~IRQ_PENDING) | IRQ_REPLAY; - if (!desc->chip || !desc->chip->retrigger || - !desc->chip->retrigger(irq)) { + if (!desc->chip->retrigger || !desc->chip->retrigger(irq)) { #ifdef CONFIG_HARDIRQS_SW_RESEND /* Set it pending and activate the softirq: */ set_bit(irq, irqs_resend); diff --git a/kernel/irq/spurious.c b/kernel/irq/spurious.c index 4d568294de3..22b0a6eedf2 100644 --- a/kernel/irq/spurious.c +++ b/kernel/irq/spurious.c @@ -104,7 +104,7 @@ static int misrouted_irq(int irq) return ok; } -static void poll_all_shared_irqs(void) +static void poll_spurious_irqs(unsigned long dummy) { struct irq_desc *desc; int i; @@ -121,25 +121,15 @@ static void poll_all_shared_irqs(void) if (!(status & IRQ_SPURIOUS_DISABLED)) continue; + local_irq_disable(); try_one_irq(i, desc); + local_irq_enable(); } -} - -static void poll_spurious_irqs(unsigned long dummy) -{ - poll_all_shared_irqs(); mod_timer(&poll_spurious_irq_timer, jiffies + POLL_SPURIOUS_IRQ_INTERVAL); } -#ifdef CONFIG_DEBUG_SHIRQ -void debug_poll_all_shared_irqs(void) -{ - poll_all_shared_irqs(); -} -#endif - /* * If 99,900 of the previous 100,000 interrupts have not been handled * then assume that the IRQ is stuck in some manner. Drop a diagnostic @@ -297,7 +287,6 @@ static int __init irqfixup_setup(char *str) __setup("irqfixup", irqfixup_setup); module_param(irqfixup, int, 0644); -MODULE_PARM_DESC("irqfixup", "0: No fixup, 1: irqfixup mode, 2: irqpoll mode"); static int __init irqpoll_setup(char *str) { diff --git a/kernel/itimer.c b/kernel/itimer.c index 58762f7077e..d802883153d 100644 --- a/kernel/itimer.c +++ b/kernel/itimer.c @@ -12,6 +12,7 @@ #include <linux/time.h> #include <linux/posix-timers.h> #include <linux/hrtimer.h> +#include <trace/events/timer.h> #include <asm/uaccess.h> @@ -41,10 +42,43 @@ static struct timeval itimer_get_remtime(struct hrtimer *timer) return ktime_to_timeval(rem); } +static void get_cpu_itimer(struct task_struct *tsk, unsigned int clock_id, + struct itimerval *const value) +{ + cputime_t cval, cinterval; + struct cpu_itimer *it = &tsk->signal->it[clock_id]; + + spin_lock_irq(&tsk->sighand->siglock); + + cval = it->expires; + cinterval = it->incr; + if (!cputime_eq(cval, cputime_zero)) { + struct task_cputime cputime; + cputime_t t; + + thread_group_cputimer(tsk, &cputime); + if (clock_id == CPUCLOCK_PROF) + t = cputime_add(cputime.utime, cputime.stime); + else + /* CPUCLOCK_VIRT */ + t = cputime.utime; + + if (cputime_le(cval, t)) + /* about to fire */ + cval = cputime_one_jiffy; + else + cval = cputime_sub(cval, t); + } + + spin_unlock_irq(&tsk->sighand->siglock); + + cputime_to_timeval(cval, &value->it_value); + cputime_to_timeval(cinterval, &value->it_interval); +} + int do_getitimer(int which, struct itimerval *value) { struct task_struct *tsk = current; - cputime_t cinterval, cval; switch (which) { case ITIMER_REAL: @@ -55,44 +89,10 @@ int do_getitimer(int which, struct itimerval *value) spin_unlock_irq(&tsk->sighand->siglock); break; case ITIMER_VIRTUAL: - spin_lock_irq(&tsk->sighand->siglock); - cval = tsk->signal->it_virt_expires; - cinterval = tsk->signal->it_virt_incr; - if (!cputime_eq(cval, cputime_zero)) { - struct task_cputime cputime; - cputime_t utime; - - thread_group_cputimer(tsk, &cputime); - utime = cputime.utime; - if (cputime_le(cval, utime)) { /* about to fire */ - cval = jiffies_to_cputime(1); - } else { - cval = cputime_sub(cval, utime); - } - } - spin_unlock_irq(&tsk->sighand->siglock); - cputime_to_timeval(cval, &value->it_value); - cputime_to_timeval(cinterval, &value->it_interval); + get_cpu_itimer(tsk, CPUCLOCK_VIRT, value); break; case ITIMER_PROF: - spin_lock_irq(&tsk->sighand->siglock); - cval = tsk->signal->it_prof_expires; - cinterval = tsk->signal->it_prof_incr; - if (!cputime_eq(cval, cputime_zero)) { - struct task_cputime times; - cputime_t ptime; - - thread_group_cputimer(tsk, ×); - ptime = cputime_add(times.utime, times.stime); - if (cputime_le(cval, ptime)) { /* about to fire */ - cval = jiffies_to_cputime(1); - } else { - cval = cputime_sub(cval, ptime); - } - } - spin_unlock_irq(&tsk->sighand->siglock); - cputime_to_timeval(cval, &value->it_value); - cputime_to_timeval(cinterval, &value->it_interval); + get_cpu_itimer(tsk, CPUCLOCK_PROF, value); break; default: return(-EINVAL); @@ -123,11 +123,65 @@ enum hrtimer_restart it_real_fn(struct hrtimer *timer) struct signal_struct *sig = container_of(timer, struct signal_struct, real_timer); + trace_itimer_expire(ITIMER_REAL, sig->leader_pid, 0); kill_pid_info(SIGALRM, SEND_SIG_PRIV, sig->leader_pid); return HRTIMER_NORESTART; } +static inline u32 cputime_sub_ns(cputime_t ct, s64 real_ns) +{ + struct timespec ts; + s64 cpu_ns; + + cputime_to_timespec(ct, &ts); + cpu_ns = timespec_to_ns(&ts); + + return (cpu_ns <= real_ns) ? 0 : cpu_ns - real_ns; +} + +static void set_cpu_itimer(struct task_struct *tsk, unsigned int clock_id, + const struct itimerval *const value, + struct itimerval *const ovalue) +{ + cputime_t cval, nval, cinterval, ninterval; + s64 ns_ninterval, ns_nval; + u32 error, incr_error; + struct cpu_itimer *it = &tsk->signal->it[clock_id]; + + nval = timeval_to_cputime(&value->it_value); + ns_nval = timeval_to_ns(&value->it_value); + ninterval = timeval_to_cputime(&value->it_interval); + ns_ninterval = timeval_to_ns(&value->it_interval); + + error = cputime_sub_ns(nval, ns_nval); + incr_error = cputime_sub_ns(ninterval, ns_ninterval); + + spin_lock_irq(&tsk->sighand->siglock); + + cval = it->expires; + cinterval = it->incr; + if (!cputime_eq(cval, cputime_zero) || + !cputime_eq(nval, cputime_zero)) { + if (cputime_gt(nval, cputime_zero)) + nval = cputime_add(nval, cputime_one_jiffy); + set_process_cpu_timer(tsk, clock_id, &nval, &cval); + } + it->expires = nval; + it->incr = ninterval; + it->error = error; + it->incr_error = incr_error; + trace_itimer_state(clock_id == CPUCLOCK_VIRT ? + ITIMER_VIRTUAL : ITIMER_PROF, value, nval); + + spin_unlock_irq(&tsk->sighand->siglock); + + if (ovalue) { + cputime_to_timeval(cval, &ovalue->it_value); + cputime_to_timeval(cinterval, &ovalue->it_interval); + } +} + /* * Returns true if the timeval is in canonical form */ @@ -139,7 +193,6 @@ int do_setitimer(int which, struct itimerval *value, struct itimerval *ovalue) struct task_struct *tsk = current; struct hrtimer *timer; ktime_t expires; - cputime_t cval, cinterval, nval, ninterval; /* * Validate the timevals in value. @@ -171,51 +224,14 @@ again: } else tsk->signal->it_real_incr.tv64 = 0; + trace_itimer_state(ITIMER_REAL, value, 0); spin_unlock_irq(&tsk->sighand->siglock); break; case ITIMER_VIRTUAL: - nval = timeval_to_cputime(&value->it_value); - ninterval = timeval_to_cputime(&value->it_interval); - spin_lock_irq(&tsk->sighand->siglock); - cval = tsk->signal->it_virt_expires; - cinterval = tsk->signal->it_virt_incr; - if (!cputime_eq(cval, cputime_zero) || - !cputime_eq(nval, cputime_zero)) { - if (cputime_gt(nval, cputime_zero)) - nval = cputime_add(nval, - jiffies_to_cputime(1)); - set_process_cpu_timer(tsk, CPUCLOCK_VIRT, - &nval, &cval); - } - tsk->signal->it_virt_expires = nval; - tsk->signal->it_virt_incr = ninterval; - spin_unlock_irq(&tsk->sighand->siglock); - if (ovalue) { - cputime_to_timeval(cval, &ovalue->it_value); - cputime_to_timeval(cinterval, &ovalue->it_interval); - } + set_cpu_itimer(tsk, CPUCLOCK_VIRT, value, ovalue); break; case ITIMER_PROF: - nval = timeval_to_cputime(&value->it_value); - ninterval = timeval_to_cputime(&value->it_interval); - spin_lock_irq(&tsk->sighand->siglock); - cval = tsk->signal->it_prof_expires; - cinterval = tsk->signal->it_prof_incr; - if (!cputime_eq(cval, cputime_zero) || - !cputime_eq(nval, cputime_zero)) { - if (cputime_gt(nval, cputime_zero)) - nval = cputime_add(nval, - jiffies_to_cputime(1)); - set_process_cpu_timer(tsk, CPUCLOCK_PROF, - &nval, &cval); - } - tsk->signal->it_prof_expires = nval; - tsk->signal->it_prof_incr = ninterval; - spin_unlock_irq(&tsk->sighand->siglock); - if (ovalue) { - cputime_to_timeval(cval, &ovalue->it_value); - cputime_to_timeval(cinterval, &ovalue->it_interval); - } + set_cpu_itimer(tsk, CPUCLOCK_PROF, value, ovalue); break; default: return -EINVAL; diff --git a/kernel/kallsyms.c b/kernel/kallsyms.c index 7b8b0f21a5b..8e5288a8a35 100644 --- a/kernel/kallsyms.c +++ b/kernel/kallsyms.c @@ -30,12 +30,16 @@ #define all_var 0 #endif -/* These will be re-linked against their real values during the second link stage */ +/* + * These will be re-linked against their real values + * during the second link stage. + */ extern const unsigned long kallsyms_addresses[] __attribute__((weak)); extern const u8 kallsyms_names[] __attribute__((weak)); -/* tell the compiler that the count isn't in the small data section if the arch - * has one (eg: FRV) +/* + * Tell the compiler that the count isn't in the small data section if the arch + * has one (eg: FRV). */ extern const unsigned long kallsyms_num_syms __attribute__((weak, section(".rodata"))); @@ -55,7 +59,8 @@ static inline int is_kernel_inittext(unsigned long addr) static inline int is_kernel_text(unsigned long addr) { - if (addr >= (unsigned long)_stext && addr <= (unsigned long)_etext) + if ((addr >= (unsigned long)_stext && addr <= (unsigned long)_etext) || + arch_is_kernel_text(addr)) return 1; return in_gate_area_no_task(addr); } @@ -75,31 +80,37 @@ static int is_ksym_addr(unsigned long addr) return is_kernel_text(addr) || is_kernel_inittext(addr); } -/* expand a compressed symbol data into the resulting uncompressed string, - given the offset to where the symbol is in the compressed stream */ +/* + * Expand a compressed symbol data into the resulting uncompressed string, + * given the offset to where the symbol is in the compressed stream. + */ static unsigned int kallsyms_expand_symbol(unsigned int off, char *result) { int len, skipped_first = 0; const u8 *tptr, *data; - /* get the compressed symbol length from the first symbol byte */ + /* Get the compressed symbol length from the first symbol byte. */ data = &kallsyms_names[off]; len = *data; data++; - /* update the offset to return the offset for the next symbol on - * the compressed stream */ + /* + * Update the offset to return the offset for the next symbol on + * the compressed stream. + */ off += len + 1; - /* for every byte on the compressed symbol data, copy the table - entry for that byte */ - while(len) { - tptr = &kallsyms_token_table[ kallsyms_token_index[*data] ]; + /* + * For every byte on the compressed symbol data, copy the table + * entry for that byte. + */ + while (len) { + tptr = &kallsyms_token_table[kallsyms_token_index[*data]]; data++; len--; while (*tptr) { - if(skipped_first) { + if (skipped_first) { *result = *tptr; result++; } else @@ -110,36 +121,46 @@ static unsigned int kallsyms_expand_symbol(unsigned int off, char *result) *result = '\0'; - /* return to offset to the next symbol */ + /* Return to offset to the next symbol. */ return off; } -/* get symbol type information. This is encoded as a single char at the - * begining of the symbol name */ +/* + * Get symbol type information. This is encoded as a single char at the + * beginning of the symbol name. + */ static char kallsyms_get_symbol_type(unsigned int off) { - /* get just the first code, look it up in the token table, and return the - * first char from this token */ - return kallsyms_token_table[ kallsyms_token_index[ kallsyms_names[off+1] ] ]; + /* + * Get just the first code, look it up in the token table, + * and return the first char from this token. + */ + return kallsyms_token_table[kallsyms_token_index[kallsyms_names[off + 1]]]; } -/* find the offset on the compressed stream given and index in the - * kallsyms array */ +/* + * Find the offset on the compressed stream given and index in the + * kallsyms array. + */ static unsigned int get_symbol_offset(unsigned long pos) { const u8 *name; int i; - /* use the closest marker we have. We have markers every 256 positions, - * so that should be close enough */ - name = &kallsyms_names[ kallsyms_markers[pos>>8] ]; + /* + * Use the closest marker we have. We have markers every 256 positions, + * so that should be close enough. + */ + name = &kallsyms_names[kallsyms_markers[pos >> 8]]; - /* sequentially scan all the symbols up to the point we're searching for. - * Every symbol is stored in a [<len>][<len> bytes of data] format, so we - * just need to add the len to the current pointer for every symbol we - * wish to skip */ - for(i = 0; i < (pos&0xFF); i++) + /* + * Sequentially scan all the symbols up to the point we're searching + * for. Every symbol is stored in a [<len>][<len> bytes of data] format, + * so we just need to add the len to the current pointer for every + * symbol we wish to skip. + */ + for (i = 0; i < (pos & 0xFF); i++) name = name + (*name) + 1; return name - kallsyms_names; @@ -160,6 +181,26 @@ unsigned long kallsyms_lookup_name(const char *name) } return module_kallsyms_lookup_name(name); } +EXPORT_SYMBOL_GPL(kallsyms_lookup_name); + +int kallsyms_on_each_symbol(int (*fn)(void *, const char *, struct module *, + unsigned long), + void *data) +{ + char namebuf[KSYM_NAME_LEN]; + unsigned long i; + unsigned int off; + int ret; + + for (i = 0, off = 0; i < kallsyms_num_syms; i++) { + off = kallsyms_expand_symbol(off, namebuf); + ret = fn(data, namebuf, NULL, kallsyms_addresses[i]); + if (ret != 0) + return ret; + } + return module_kallsyms_on_each_symbol(fn, data); +} +EXPORT_SYMBOL_GPL(kallsyms_on_each_symbol); static unsigned long get_symbol_pos(unsigned long addr, unsigned long *symbolsize, @@ -171,7 +212,7 @@ static unsigned long get_symbol_pos(unsigned long addr, /* This kernel should never had been booted. */ BUG_ON(!kallsyms_addresses); - /* do a binary search on the sorted kallsyms_addresses array */ + /* Do a binary search on the sorted kallsyms_addresses array. */ low = 0; high = kallsyms_num_syms; @@ -184,15 +225,15 @@ static unsigned long get_symbol_pos(unsigned long addr, } /* - * search for the first aliased symbol. Aliased - * symbols are symbols with the same address + * Search for the first aliased symbol. Aliased + * symbols are symbols with the same address. */ while (low && kallsyms_addresses[low-1] == kallsyms_addresses[low]) --low; symbol_start = kallsyms_addresses[low]; - /* Search for next non-aliased symbol */ + /* Search for next non-aliased symbol. */ for (i = low + 1; i < kallsyms_num_syms; i++) { if (kallsyms_addresses[i] > symbol_start) { symbol_end = kallsyms_addresses[i]; @@ -200,7 +241,7 @@ static unsigned long get_symbol_pos(unsigned long addr, } } - /* if we found no next symbol, we use the end of the section */ + /* If we found no next symbol, we use the end of the section. */ if (!symbol_end) { if (is_kernel_inittext(addr)) symbol_end = (unsigned long)_einittext; @@ -233,10 +274,10 @@ int kallsyms_lookup_size_offset(unsigned long addr, unsigned long *symbolsize, /* * Lookup an address - * - modname is set to NULL if it's in the kernel - * - we guarantee that the returned name is valid until we reschedule even if - * it resides in a module - * - we also guarantee that modname will be valid until rescheduled + * - modname is set to NULL if it's in the kernel. + * - We guarantee that the returned name is valid until we reschedule even if. + * It resides in a module. + * - We also guarantee that modname will be valid until rescheduled. */ const char *kallsyms_lookup(unsigned long addr, unsigned long *symbolsize, @@ -257,7 +298,7 @@ const char *kallsyms_lookup(unsigned long addr, return namebuf; } - /* see if it's in a module */ + /* See if it's in a module. */ return module_address_lookup(addr, symbolsize, offset, modname, namebuf); } @@ -275,7 +316,7 @@ int lookup_symbol_name(unsigned long addr, char *symname) kallsyms_expand_symbol(get_symbol_offset(pos), symname); return 0; } - /* see if it's in a module */ + /* See if it's in a module. */ return lookup_module_symbol_name(addr, symname); } @@ -294,7 +335,7 @@ int lookup_symbol_attrs(unsigned long addr, unsigned long *size, modname[0] = '\0'; return 0; } - /* see if it's in a module */ + /* See if it's in a module. */ return lookup_module_symbol_attrs(addr, size, offset, modname, name); } @@ -323,6 +364,7 @@ int sprint_symbol(char *buffer, unsigned long address) return len; } +EXPORT_SYMBOL_GPL(sprint_symbol); /* Look up a kernel symbol and print it to the kernel messages. */ void __print_symbol(const char *fmt, unsigned long address) @@ -333,13 +375,13 @@ void __print_symbol(const char *fmt, unsigned long address) printk(fmt, buffer); } +EXPORT_SYMBOL(__print_symbol); /* To avoid using get_symbol_offset for every symbol, we carry prefix along. */ -struct kallsym_iter -{ +struct kallsym_iter { loff_t pos; unsigned long value; - unsigned int nameoff; /* If iterating in core kernel symbols */ + unsigned int nameoff; /* If iterating in core kernel symbols. */ char type; char name[KSYM_NAME_LEN]; char module_name[MODULE_NAME_LEN]; @@ -385,7 +427,7 @@ static int update_iter(struct kallsym_iter *iter, loff_t pos) iter->pos = pos; return get_ksymbol_mod(iter); } - + /* If we're not on the desired position, reset to new position. */ if (pos != iter->pos) reset_iter(iter, pos); @@ -420,23 +462,25 @@ static int s_show(struct seq_file *m, void *p) { struct kallsym_iter *iter = m->private; - /* Some debugging symbols have no name. Ignore them. */ + /* Some debugging symbols have no name. Ignore them. */ if (!iter->name[0]) return 0; if (iter->module_name[0]) { char type; - /* Label it "global" if it is exported, - * "local" if not exported. */ + /* + * Label it "global" if it is exported, + * "local" if not exported. + */ type = iter->exported ? toupper(iter->type) : tolower(iter->type); seq_printf(m, "%0*lx %c %s\t[%s]\n", - (int)(2*sizeof(void*)), + (int)(2 * sizeof(void *)), iter->value, type, iter->name, iter->module_name); } else seq_printf(m, "%0*lx %c %s\n", - (int)(2*sizeof(void*)), + (int)(2 * sizeof(void *)), iter->value, iter->type, iter->name); return 0; } @@ -450,9 +494,11 @@ static const struct seq_operations kallsyms_op = { static int kallsyms_open(struct inode *inode, struct file *file) { - /* We keep iterator in m->private, since normal case is to + /* + * We keep iterator in m->private, since normal case is to * s_start from where we left off, so we avoid doing - * using get_symbol_offset for every symbol */ + * using get_symbol_offset for every symbol. + */ struct kallsym_iter *iter; int ret; @@ -481,7 +527,4 @@ static int __init kallsyms_init(void) proc_create("kallsyms", 0444, NULL, &kallsyms_operations); return 0; } -__initcall(kallsyms_init); - -EXPORT_SYMBOL(__print_symbol); -EXPORT_SYMBOL_GPL(sprint_symbol); +device_initcall(kallsyms_init); diff --git a/kernel/kexec.c b/kernel/kexec.c index 93eed85fe01..f336e2107f9 100644 --- a/kernel/kexec.c +++ b/kernel/kexec.c @@ -42,7 +42,7 @@ note_buf_t* crash_notes; /* vmcoreinfo stuff */ -unsigned char vmcoreinfo_data[VMCOREINFO_BYTES]; +static unsigned char vmcoreinfo_data[VMCOREINFO_BYTES]; u32 vmcoreinfo_note[VMCOREINFO_NOTE_SIZE/4]; size_t vmcoreinfo_size; size_t vmcoreinfo_max_size = sizeof(vmcoreinfo_data); @@ -1228,7 +1228,7 @@ static int __init parse_crashkernel_mem(char *cmdline, } while (*cur++ == ','); if (*crash_size > 0) { - while (*cur != ' ' && *cur != '@') + while (*cur && *cur != ' ' && *cur != '@') cur++; if (*cur == '@') { cur++; @@ -1409,6 +1409,7 @@ static int __init crash_save_vmcoreinfo_init(void) VMCOREINFO_OFFSET(list_head, prev); VMCOREINFO_OFFSET(vm_struct, addr); VMCOREINFO_LENGTH(zone.free_area, MAX_ORDER); + log_buf_kexec_setup(); VMCOREINFO_LENGTH(free_area.free_list, MIGRATE_TYPES); VMCOREINFO_NUMBER(NR_FREE_PAGES); VMCOREINFO_NUMBER(PG_lru); @@ -1447,18 +1448,17 @@ int kernel_kexec(void) goto Restore_console; } suspend_console(); - error = device_suspend(PMSG_FREEZE); + error = dpm_suspend_start(PMSG_FREEZE); if (error) goto Resume_console; - device_pm_lock(); - /* At this point, device_suspend() has been called, - * but *not* device_power_down(). We *must* - * device_power_down() now. Otherwise, drivers for + /* At this point, dpm_suspend_start() has been called, + * but *not* dpm_suspend_noirq(). We *must* call + * dpm_suspend_noirq() now. Otherwise, drivers for * some devices (e.g. interrupt controllers) become * desynchronized with the actual state of the * hardware at resume time, and evil weirdness ensues. */ - error = device_power_down(PMSG_FREEZE); + error = dpm_suspend_noirq(PMSG_FREEZE); if (error) goto Resume_devices; error = disable_nonboot_cpus(); @@ -1486,10 +1486,9 @@ int kernel_kexec(void) local_irq_enable(); Enable_cpus: enable_nonboot_cpus(); - device_power_up(PMSG_RESTORE); + dpm_resume_noirq(PMSG_RESTORE); Resume_devices: - device_pm_unlock(); - device_resume(PMSG_RESTORE); + dpm_resume_end(PMSG_RESTORE); Resume_console: resume_console(); thaw_processes(); diff --git a/kernel/kfifo.c b/kernel/kfifo.c index bc41ad0f24f..3765ff3c1bb 100644 --- a/kernel/kfifo.c +++ b/kernel/kfifo.c @@ -72,9 +72,9 @@ struct kfifo *kfifo_alloc(unsigned int size, gfp_t gfp_mask, spinlock_t *lock) /* * round up to the next power of 2, since our 'let the indices - * wrap' tachnique works only in this case. + * wrap' technique works only in this case. */ - if (size & (size - 1)) { + if (!is_power_of_2(size)) { BUG_ON(size > 0x80000000); size = roundup_pow_of_two(size); } @@ -117,7 +117,7 @@ EXPORT_SYMBOL(kfifo_free); * writer, you don't need extra locking to use these functions. */ unsigned int __kfifo_put(struct kfifo *fifo, - unsigned char *buffer, unsigned int len) + const unsigned char *buffer, unsigned int len) { unsigned int l; diff --git a/kernel/kgdb.c b/kernel/kgdb.c index e4dcfb2272a..7d701463402 100644 --- a/kernel/kgdb.c +++ b/kernel/kgdb.c @@ -870,7 +870,7 @@ static void gdb_cmd_getregs(struct kgdb_state *ks) /* * All threads that don't have debuggerinfo should be - * in __schedule() sleeping, since all other CPUs + * in schedule() sleeping, since all other CPUs * are in kgdb_wait, and thus have debuggerinfo. */ if (local_debuggerinfo) { @@ -1583,8 +1583,8 @@ static void sysrq_handle_gdb(int key, struct tty_struct *tty) static struct sysrq_key_op sysrq_gdb_op = { .handler = sysrq_handle_gdb, - .help_msg = "Gdb", - .action_msg = "GDB", + .help_msg = "debug(G)", + .action_msg = "DEBUG", }; #endif diff --git a/kernel/kmod.c b/kernel/kmod.c index f0c8f545180..25b10319036 100644 --- a/kernel/kmod.c +++ b/kernel/kmod.c @@ -24,7 +24,6 @@ #include <linux/unistd.h> #include <linux/kmod.h> #include <linux/slab.h> -#include <linux/mnt_namespace.h> #include <linux/completion.h> #include <linux/file.h> #include <linux/fdtable.h> @@ -38,6 +37,8 @@ #include <linux/suspend.h> #include <asm/uaccess.h> +#include <trace/events/module.h> + extern int max_threads; static struct workqueue_struct *khelper_wq; @@ -50,7 +51,8 @@ static struct workqueue_struct *khelper_wq; char modprobe_path[KMOD_PATH_LEN] = "/sbin/modprobe"; /** - * request_module - try to load a kernel module + * __request_module - try to load a kernel module + * @wait: wait (or not) for the operation to complete * @fmt: printf style format string for the name of the module * @...: arguments as specified in the format string * @@ -63,7 +65,7 @@ char modprobe_path[KMOD_PATH_LEN] = "/sbin/modprobe"; * If module auto-loading support is disabled then this function * becomes a no-operation. */ -int request_module(const char *fmt, ...) +int __request_module(bool wait, const char *fmt, ...) { va_list args; char module_name[MODULE_NAME_LEN]; @@ -84,6 +86,10 @@ int request_module(const char *fmt, ...) if (ret >= MODULE_NAME_LEN) return -ENAMETOOLONG; + ret = security_kernel_module_request(module_name); + if (ret) + return ret; + /* If modprobe needs a service that is in a module, we get a recursive * loop. Limit the number of running kmod threads to max_threads/2 or * MAX_KMOD_CONCURRENT, whichever is the smaller. A cleaner method @@ -108,11 +114,14 @@ int request_module(const char *fmt, ...) return -ENOMEM; } - ret = call_usermodehelper(modprobe_path, argv, envp, 1); + trace_module_request(module_name, wait, _RET_IP_); + + ret = call_usermodehelper(modprobe_path, argv, envp, + wait ? UMH_WAIT_PROC : UMH_WAIT_EXEC); atomic_dec(&kmod_concurrent); return ret; } -EXPORT_SYMBOL(request_module); +EXPORT_SYMBOL(__request_module); #endif /* CONFIG_MODULES */ struct subprocess_info { @@ -368,8 +377,10 @@ struct subprocess_info *call_usermodehelper_setup(char *path, char **argv, sub_info->argv = argv; sub_info->envp = envp; sub_info->cred = prepare_usermodehelper_creds(); - if (!sub_info->cred) + if (!sub_info->cred) { + kfree(sub_info); return NULL; + } out: return sub_info; @@ -459,6 +470,7 @@ int call_usermodehelper_exec(struct subprocess_info *sub_info, int retval = 0; BUG_ON(atomic_read(&sub_info->cred->usage) != 1); + validate_creds(sub_info->cred); helper_lock(); if (sub_info->path[0] == '\0') diff --git a/kernel/kprobes.c b/kernel/kprobes.c index 7ba8cd9845c..e5342a344c4 100644 --- a/kernel/kprobes.c +++ b/kernel/kprobes.c @@ -43,6 +43,7 @@ #include <linux/seq_file.h> #include <linux/debugfs.h> #include <linux/kdebug.h> +#include <linux/memory.h> #include <asm-generic/sections.h> #include <asm/cacheflush.h> @@ -67,7 +68,7 @@ static struct hlist_head kprobe_table[KPROBE_TABLE_SIZE]; static struct hlist_head kretprobe_inst_table[KPROBE_TABLE_SIZE]; /* NOTE: change this value only with kprobe_mutex held */ -static bool kprobe_enabled; +static bool kprobes_all_disarmed; static DEFINE_MUTEX(kprobe_mutex); /* Protects kprobe_table */ static DEFINE_PER_CPU(struct kprobe *, kprobe_instance) = NULL; @@ -89,6 +90,9 @@ static spinlock_t *kretprobe_table_lock_ptr(unsigned long hash) */ static struct kprobe_blackpoint kprobe_blacklist[] = { {"preempt_schedule",}, + {"native_get_debugreg",}, + {"irq_entries_start",}, + {"common_interrupt",}, {NULL} /* Terminator */ }; @@ -102,7 +106,7 @@ static struct kprobe_blackpoint kprobe_blacklist[] = { #define INSNS_PER_PAGE (PAGE_SIZE/(MAX_INSN_SIZE * sizeof(kprobe_opcode_t))) struct kprobe_insn_page { - struct hlist_node hlist; + struct list_head list; kprobe_opcode_t *insns; /* Page of instruction slots */ char slot_used[INSNS_PER_PAGE]; int nused; @@ -116,7 +120,7 @@ enum kprobe_slot_state { }; static DEFINE_MUTEX(kprobe_insn_mutex); /* Protects kprobe_insn_pages */ -static struct hlist_head kprobe_insn_pages; +static LIST_HEAD(kprobe_insn_pages); static int kprobe_garbage_slots; static int collect_garbage_slots(void); @@ -151,10 +155,9 @@ loop_end: static kprobe_opcode_t __kprobes *__get_insn_slot(void) { struct kprobe_insn_page *kip; - struct hlist_node *pos; retry: - hlist_for_each_entry(kip, pos, &kprobe_insn_pages, hlist) { + list_for_each_entry(kip, &kprobe_insn_pages, list) { if (kip->nused < INSNS_PER_PAGE) { int i; for (i = 0; i < INSNS_PER_PAGE; i++) { @@ -188,8 +191,8 @@ static kprobe_opcode_t __kprobes *__get_insn_slot(void) kfree(kip); return NULL; } - INIT_HLIST_NODE(&kip->hlist); - hlist_add_head(&kip->hlist, &kprobe_insn_pages); + INIT_LIST_HEAD(&kip->list); + list_add(&kip->list, &kprobe_insn_pages); memset(kip->slot_used, SLOT_CLEAN, INSNS_PER_PAGE); kip->slot_used[0] = SLOT_USED; kip->nused = 1; @@ -218,12 +221,8 @@ static int __kprobes collect_one_slot(struct kprobe_insn_page *kip, int idx) * so as not to have to set it up again the * next time somebody inserts a probe. */ - hlist_del(&kip->hlist); - if (hlist_empty(&kprobe_insn_pages)) { - INIT_HLIST_NODE(&kip->hlist); - hlist_add_head(&kip->hlist, - &kprobe_insn_pages); - } else { + if (!list_is_singular(&kprobe_insn_pages)) { + list_del(&kip->list); module_free(NULL, kip->insns); kfree(kip); } @@ -234,18 +233,13 @@ static int __kprobes collect_one_slot(struct kprobe_insn_page *kip, int idx) static int __kprobes collect_garbage_slots(void) { - struct kprobe_insn_page *kip; - struct hlist_node *pos, *next; - int safety; + struct kprobe_insn_page *kip, *next; /* Ensure no-one is preepmted on the garbages */ - mutex_unlock(&kprobe_insn_mutex); - safety = check_safety(); - mutex_lock(&kprobe_insn_mutex); - if (safety != 0) + if (check_safety()) return -EAGAIN; - hlist_for_each_entry_safe(kip, pos, next, &kprobe_insn_pages, hlist) { + list_for_each_entry_safe(kip, next, &kprobe_insn_pages, list) { int i; if (kip->ngarbage == 0) continue; @@ -263,19 +257,17 @@ static int __kprobes collect_garbage_slots(void) void __kprobes free_insn_slot(kprobe_opcode_t * slot, int dirty) { struct kprobe_insn_page *kip; - struct hlist_node *pos; mutex_lock(&kprobe_insn_mutex); - hlist_for_each_entry(kip, pos, &kprobe_insn_pages, hlist) { + list_for_each_entry(kip, &kprobe_insn_pages, list) { if (kip->insns <= slot && slot < kip->insns + (INSNS_PER_PAGE * MAX_INSN_SIZE)) { int i = (slot - kip->insns) / MAX_INSN_SIZE; if (dirty) { kip->slot_used[i] = SLOT_DIRTY; kip->ngarbage++; - } else { + } else collect_one_slot(kip, i); - } break; } } @@ -318,6 +310,22 @@ struct kprobe __kprobes *get_kprobe(void *addr) return NULL; } +/* Arm a kprobe with text_mutex */ +static void __kprobes arm_kprobe(struct kprobe *kp) +{ + mutex_lock(&text_mutex); + arch_arm_kprobe(kp); + mutex_unlock(&text_mutex); +} + +/* Disarm a kprobe with text_mutex */ +static void __kprobes disarm_kprobe(struct kprobe *kp) +{ + mutex_lock(&text_mutex); + arch_disarm_kprobe(kp); + mutex_unlock(&text_mutex); +} + /* * Aggregate handlers for multiple kprobes support - these handlers * take care of invoking the individual kprobe handlers on p->list @@ -327,7 +335,7 @@ static int __kprobes aggr_pre_handler(struct kprobe *p, struct pt_regs *regs) struct kprobe *kp; list_for_each_entry_rcu(kp, &p->list, list) { - if (kp->pre_handler && !kprobe_gone(kp)) { + if (kp->pre_handler && likely(!kprobe_disabled(kp))) { set_kprobe_instance(kp); if (kp->pre_handler(kp, regs)) return 1; @@ -343,7 +351,7 @@ static void __kprobes aggr_post_handler(struct kprobe *p, struct pt_regs *regs, struct kprobe *kp; list_for_each_entry_rcu(kp, &p->list, list) { - if (kp->post_handler && !kprobe_gone(kp)) { + if (kp->post_handler && likely(!kprobe_disabled(kp))) { set_kprobe_instance(kp); kp->post_handler(kp, regs, flags); reset_kprobe_instance(); @@ -517,20 +525,28 @@ static inline void copy_kprobe(struct kprobe *old_p, struct kprobe *p) } /* -* Add the new probe to old_p->list. Fail if this is the +* Add the new probe to ap->list. Fail if this is the * second jprobe at the address - two jprobes can't coexist */ -static int __kprobes add_new_kprobe(struct kprobe *old_p, struct kprobe *p) +static int __kprobes add_new_kprobe(struct kprobe *ap, struct kprobe *p) { + BUG_ON(kprobe_gone(ap) || kprobe_gone(p)); if (p->break_handler) { - if (old_p->break_handler) + if (ap->break_handler) return -EEXIST; - list_add_tail_rcu(&p->list, &old_p->list); - old_p->break_handler = aggr_break_handler; + list_add_tail_rcu(&p->list, &ap->list); + ap->break_handler = aggr_break_handler; } else - list_add_rcu(&p->list, &old_p->list); - if (p->post_handler && !old_p->post_handler) - old_p->post_handler = aggr_post_handler; + list_add_rcu(&p->list, &ap->list); + if (p->post_handler && !ap->post_handler) + ap->post_handler = aggr_post_handler; + + if (kprobe_disabled(ap) && !kprobe_disabled(p)) { + ap->flags &= ~KPROBE_FLAG_DISABLED; + if (!kprobes_all_disarmed) + /* Arm the breakpoint again. */ + arm_kprobe(ap); + } return 0; } @@ -543,6 +559,7 @@ static inline void add_aggr_kprobe(struct kprobe *ap, struct kprobe *p) copy_kprobe(p, ap); flush_insn_slot(ap); ap->addr = p->addr; + ap->flags = p->flags; ap->pre_handler = aggr_pre_handler; ap->fault_handler = aggr_fault_handler; /* We don't care the kprobe which has gone. */ @@ -565,44 +582,59 @@ static int __kprobes register_aggr_kprobe(struct kprobe *old_p, struct kprobe *p) { int ret = 0; - struct kprobe *ap; + struct kprobe *ap = old_p; + + if (old_p->pre_handler != aggr_pre_handler) { + /* If old_p is not an aggr_probe, create new aggr_kprobe. */ + ap = kzalloc(sizeof(struct kprobe), GFP_KERNEL); + if (!ap) + return -ENOMEM; + add_aggr_kprobe(ap, old_p); + } - if (kprobe_gone(old_p)) { + if (kprobe_gone(ap)) { /* * Attempting to insert new probe at the same location that * had a probe in the module vaddr area which already * freed. So, the instruction slot has already been * released. We need a new slot for the new probe. */ - ret = arch_prepare_kprobe(old_p); + ret = arch_prepare_kprobe(ap); if (ret) + /* + * Even if fail to allocate new slot, don't need to + * free aggr_probe. It will be used next time, or + * freed by unregister_kprobe. + */ return ret; - } - if (old_p->pre_handler == aggr_pre_handler) { - copy_kprobe(old_p, p); - ret = add_new_kprobe(old_p, p); - ap = old_p; - } else { - ap = kzalloc(sizeof(struct kprobe), GFP_KERNEL); - if (!ap) { - if (kprobe_gone(old_p)) - arch_remove_kprobe(old_p); - return -ENOMEM; - } - add_aggr_kprobe(ap, old_p); - copy_kprobe(ap, p); - ret = add_new_kprobe(ap, p); - } - if (kprobe_gone(old_p)) { + /* - * If the old_p has gone, its breakpoint has been disarmed. - * We have to arm it again after preparing real kprobes. + * Clear gone flag to prevent allocating new slot again, and + * set disabled flag because it is not armed yet. */ - ap->flags &= ~KPROBE_FLAG_GONE; - if (kprobe_enabled) - arch_arm_kprobe(ap); + ap->flags = (ap->flags & ~KPROBE_FLAG_GONE) + | KPROBE_FLAG_DISABLED; } - return ret; + + copy_kprobe(ap, p); + return add_new_kprobe(ap, p); +} + +/* Try to disable aggr_kprobe, and return 1 if succeeded.*/ +static int __kprobes try_to_disable_aggr_kprobe(struct kprobe *p) +{ + struct kprobe *kp; + + list_for_each_entry_rcu(kp, &p->list, list) { + if (!kprobe_disabled(kp)) + /* + * There is an active probe on the list. + * We can't disable aggr_kprobe. + */ + return 0; + } + p->flags |= KPROBE_FLAG_DISABLED; + return 1; } static int __kprobes in_kprobes_functions(unsigned long addr) @@ -644,6 +676,40 @@ static kprobe_opcode_t __kprobes *kprobe_addr(struct kprobe *p) return (kprobe_opcode_t *)(((char *)addr) + p->offset); } +/* Check passed kprobe is valid and return kprobe in kprobe_table. */ +static struct kprobe * __kprobes __get_valid_kprobe(struct kprobe *p) +{ + struct kprobe *old_p, *list_p; + + old_p = get_kprobe(p->addr); + if (unlikely(!old_p)) + return NULL; + + if (p != old_p) { + list_for_each_entry_rcu(list_p, &old_p->list, list) + if (list_p == p) + /* kprobe p is a valid probe */ + goto valid; + return NULL; + } +valid: + return old_p; +} + +/* Return error if the kprobe is being re-registered */ +static inline int check_kprobe_rereg(struct kprobe *p) +{ + int ret = 0; + struct kprobe *old_p; + + mutex_lock(&kprobe_mutex); + old_p = __get_valid_kprobe(p); + if (old_p) + ret = -EINVAL; + mutex_unlock(&kprobe_mutex); + return ret; +} + int __kprobes register_kprobe(struct kprobe *p) { int ret = 0; @@ -656,14 +722,20 @@ int __kprobes register_kprobe(struct kprobe *p) return -EINVAL; p->addr = addr; + ret = check_kprobe_rereg(p); + if (ret) + return ret; + preempt_disable(); - if (!__kernel_text_address((unsigned long) p->addr) || + if (!kernel_text_address((unsigned long) p->addr) || in_kprobes_functions((unsigned long) p->addr)) { preempt_enable(); return -EINVAL; } - p->flags = 0; + /* User can pass only KPROBE_FLAG_DISABLED to register_kprobe */ + p->flags &= KPROBE_FLAG_DISABLED; + /* * Check if are we probing a module. */ @@ -699,17 +771,20 @@ int __kprobes register_kprobe(struct kprobe *p) goto out; } + mutex_lock(&text_mutex); ret = arch_prepare_kprobe(p); if (ret) - goto out; + goto out_unlock_text; INIT_HLIST_NODE(&p->hlist); hlist_add_head_rcu(&p->hlist, &kprobe_table[hash_ptr(p->addr, KPROBE_HASH_BITS)]); - if (kprobe_enabled) + if (!kprobes_all_disarmed && !kprobe_disabled(p)) arch_arm_kprobe(p); +out_unlock_text: + mutex_unlock(&text_mutex); out: mutex_unlock(&kprobe_mutex); @@ -718,6 +793,7 @@ out: return ret; } +EXPORT_SYMBOL_GPL(register_kprobe); /* * Unregister a kprobe without a scheduler synchronization. @@ -726,18 +802,10 @@ static int __kprobes __unregister_kprobe_top(struct kprobe *p) { struct kprobe *old_p, *list_p; - old_p = get_kprobe(p->addr); - if (unlikely(!old_p)) + old_p = __get_valid_kprobe(p); + if (old_p == NULL) return -EINVAL; - if (p != old_p) { - list_for_each_entry_rcu(list_p, &old_p->list, list) - if (list_p == p) - /* kprobe p is a valid probe */ - goto valid_p; - return -EINVAL; - } -valid_p: if (old_p == p || (old_p->pre_handler == aggr_pre_handler && list_is_singular(&old_p->list))) { @@ -746,8 +814,8 @@ valid_p: * enabled and not gone - otherwise, the breakpoint would * already have been removed. We save on flushing icache. */ - if (kprobe_enabled && !kprobe_gone(old_p)) - arch_disarm_kprobe(p); + if (!kprobes_all_disarmed && !kprobe_disabled(old_p)) + disarm_kprobe(p); hlist_del_rcu(&old_p->hlist); } else { if (p->break_handler && !kprobe_gone(p)) @@ -761,6 +829,11 @@ valid_p: } noclean: list_del_rcu(&p->list); + if (!kprobe_disabled(old_p)) { + try_to_disable_aggr_kprobe(old_p); + if (!kprobes_all_disarmed && kprobe_disabled(old_p)) + disarm_kprobe(old_p); + } } return 0; } @@ -796,11 +869,13 @@ int __kprobes register_kprobes(struct kprobe **kps, int num) } return ret; } +EXPORT_SYMBOL_GPL(register_kprobes); void __kprobes unregister_kprobe(struct kprobe *p) { unregister_kprobes(&p, 1); } +EXPORT_SYMBOL_GPL(unregister_kprobe); void __kprobes unregister_kprobes(struct kprobe **kps, int num) { @@ -819,6 +894,7 @@ void __kprobes unregister_kprobes(struct kprobe **kps, int num) if (kps[i]->addr) __unregister_kprobe_bottom(kps[i]); } +EXPORT_SYMBOL_GPL(unregister_kprobes); static struct notifier_block kprobe_exceptions_nb = { .notifier_call = kprobe_exceptions_notify, @@ -858,16 +934,19 @@ int __kprobes register_jprobes(struct jprobe **jps, int num) } return ret; } +EXPORT_SYMBOL_GPL(register_jprobes); int __kprobes register_jprobe(struct jprobe *jp) { return register_jprobes(&jp, 1); } +EXPORT_SYMBOL_GPL(register_jprobe); void __kprobes unregister_jprobe(struct jprobe *jp) { unregister_jprobes(&jp, 1); } +EXPORT_SYMBOL_GPL(unregister_jprobe); void __kprobes unregister_jprobes(struct jprobe **jps, int num) { @@ -887,6 +966,7 @@ void __kprobes unregister_jprobes(struct jprobe **jps, int num) __unregister_kprobe_bottom(&jps[i]->kp); } } +EXPORT_SYMBOL_GPL(unregister_jprobes); #ifdef CONFIG_KRETPROBES /* @@ -912,10 +992,8 @@ static int __kprobes pre_handler_kretprobe(struct kprobe *p, ri->rp = rp; ri->task = current; - if (rp->entry_handler && rp->entry_handler(ri, regs)) { - spin_unlock_irqrestore(&rp->lock, flags); + if (rp->entry_handler && rp->entry_handler(ri, regs)) return 0; - } arch_prepare_kretprobe(ri, regs); @@ -957,9 +1035,9 @@ int __kprobes register_kretprobe(struct kretprobe *rp) /* Pre-allocate memory for max kretprobe instances */ if (rp->maxactive <= 0) { #ifdef CONFIG_PREEMPT - rp->maxactive = max(10, 2 * NR_CPUS); + rp->maxactive = max(10, 2 * num_possible_cpus()); #else - rp->maxactive = NR_CPUS; + rp->maxactive = num_possible_cpus(); #endif } spin_lock_init(&rp->lock); @@ -982,6 +1060,7 @@ int __kprobes register_kretprobe(struct kretprobe *rp) free_rp_inst(rp); return ret; } +EXPORT_SYMBOL_GPL(register_kretprobe); int __kprobes register_kretprobes(struct kretprobe **rps, int num) { @@ -999,11 +1078,13 @@ int __kprobes register_kretprobes(struct kretprobe **rps, int num) } return ret; } +EXPORT_SYMBOL_GPL(register_kretprobes); void __kprobes unregister_kretprobe(struct kretprobe *rp) { unregister_kretprobes(&rp, 1); } +EXPORT_SYMBOL_GPL(unregister_kretprobe); void __kprobes unregister_kretprobes(struct kretprobe **rps, int num) { @@ -1025,24 +1106,30 @@ void __kprobes unregister_kretprobes(struct kretprobe **rps, int num) } } } +EXPORT_SYMBOL_GPL(unregister_kretprobes); #else /* CONFIG_KRETPROBES */ int __kprobes register_kretprobe(struct kretprobe *rp) { return -ENOSYS; } +EXPORT_SYMBOL_GPL(register_kretprobe); int __kprobes register_kretprobes(struct kretprobe **rps, int num) { return -ENOSYS; } +EXPORT_SYMBOL_GPL(register_kretprobes); + void __kprobes unregister_kretprobe(struct kretprobe *rp) { } +EXPORT_SYMBOL_GPL(unregister_kretprobe); void __kprobes unregister_kretprobes(struct kretprobe **rps, int num) { } +EXPORT_SYMBOL_GPL(unregister_kretprobes); static int __kprobes pre_handler_kretprobe(struct kprobe *p, struct pt_regs *regs) @@ -1056,6 +1143,7 @@ static int __kprobes pre_handler_kretprobe(struct kprobe *p, static void __kprobes kill_kprobe(struct kprobe *p) { struct kprobe *kp; + p->flags |= KPROBE_FLAG_GONE; if (p->pre_handler == aggr_pre_handler) { /* @@ -1074,6 +1162,13 @@ static void __kprobes kill_kprobe(struct kprobe *p) arch_remove_kprobe(p); } +void __kprobes dump_kprobe(struct kprobe *kp) +{ + printk(KERN_WARNING "Dumping kprobe:\n"); + printk(KERN_WARNING "Name: %s\nAddress: %p\nOffset: %x\n", + kp->symbol_name, kp->addr, kp->offset); +} + /* Module notifier call back, checking kprobes on the module */ static int __kprobes kprobes_module_callback(struct notifier_block *nb, unsigned long val, void *data) @@ -1168,8 +1263,8 @@ static int __init init_kprobes(void) } } - /* By default, kprobes are enabled */ - kprobe_enabled = true; + /* By default, kprobes are armed */ + kprobes_all_disarmed = false; err = arch_init_kprobes(); if (!err) @@ -1197,12 +1292,18 @@ static void __kprobes report_probe(struct seq_file *pi, struct kprobe *p, else kprobe_type = "k"; if (sym) - seq_printf(pi, "%p %s %s+0x%x %s %s\n", p->addr, kprobe_type, - sym, offset, (modname ? modname : " "), - (kprobe_gone(p) ? "[GONE]" : "")); + seq_printf(pi, "%p %s %s+0x%x %s %s%s\n", + p->addr, kprobe_type, sym, offset, + (modname ? modname : " "), + (kprobe_gone(p) ? "[GONE]" : ""), + ((kprobe_disabled(p) && !kprobe_gone(p)) ? + "[DISABLED]" : "")); else - seq_printf(pi, "%p %s %p %s\n", p->addr, kprobe_type, p->addr, - (kprobe_gone(p) ? "[GONE]" : "")); + seq_printf(pi, "%p %s %p %s%s\n", + p->addr, kprobe_type, p->addr, + (kprobe_gone(p) ? "[GONE]" : ""), + ((kprobe_disabled(p) && !kprobe_gone(p)) ? + "[DISABLED]" : "")); } static void __kprobes *kprobe_seq_start(struct seq_file *f, loff_t *pos) @@ -1248,7 +1349,7 @@ static int __kprobes show_kprobe_addr(struct seq_file *pi, void *v) return 0; } -static struct seq_operations kprobes_seq_ops = { +static const struct seq_operations kprobes_seq_ops = { .start = kprobe_seq_start, .next = kprobe_seq_next, .stop = kprobe_seq_stop, @@ -1260,14 +1361,79 @@ static int __kprobes kprobes_open(struct inode *inode, struct file *filp) return seq_open(filp, &kprobes_seq_ops); } -static struct file_operations debugfs_kprobes_operations = { +static const struct file_operations debugfs_kprobes_operations = { .open = kprobes_open, .read = seq_read, .llseek = seq_lseek, .release = seq_release, }; -static void __kprobes enable_all_kprobes(void) +/* Disable one kprobe */ +int __kprobes disable_kprobe(struct kprobe *kp) +{ + int ret = 0; + struct kprobe *p; + + mutex_lock(&kprobe_mutex); + + /* Check whether specified probe is valid. */ + p = __get_valid_kprobe(kp); + if (unlikely(p == NULL)) { + ret = -EINVAL; + goto out; + } + + /* If the probe is already disabled (or gone), just return */ + if (kprobe_disabled(kp)) + goto out; + + kp->flags |= KPROBE_FLAG_DISABLED; + if (p != kp) + /* When kp != p, p is always enabled. */ + try_to_disable_aggr_kprobe(p); + + if (!kprobes_all_disarmed && kprobe_disabled(p)) + disarm_kprobe(p); +out: + mutex_unlock(&kprobe_mutex); + return ret; +} +EXPORT_SYMBOL_GPL(disable_kprobe); + +/* Enable one kprobe */ +int __kprobes enable_kprobe(struct kprobe *kp) +{ + int ret = 0; + struct kprobe *p; + + mutex_lock(&kprobe_mutex); + + /* Check whether specified probe is valid. */ + p = __get_valid_kprobe(kp); + if (unlikely(p == NULL)) { + ret = -EINVAL; + goto out; + } + + if (kprobe_gone(kp)) { + /* This kprobe has gone, we couldn't enable it. */ + ret = -EINVAL; + goto out; + } + + if (!kprobes_all_disarmed && kprobe_disabled(p)) + arm_kprobe(p); + + p->flags &= ~KPROBE_FLAG_DISABLED; + if (p != kp) + kp->flags &= ~KPROBE_FLAG_DISABLED; +out: + mutex_unlock(&kprobe_mutex); + return ret; +} +EXPORT_SYMBOL_GPL(enable_kprobe); + +static void __kprobes arm_all_kprobes(void) { struct hlist_head *head; struct hlist_node *node; @@ -1276,18 +1442,20 @@ static void __kprobes enable_all_kprobes(void) mutex_lock(&kprobe_mutex); - /* If kprobes are already enabled, just return */ - if (kprobe_enabled) + /* If kprobes are armed, just return */ + if (!kprobes_all_disarmed) goto already_enabled; + mutex_lock(&text_mutex); for (i = 0; i < KPROBE_TABLE_SIZE; i++) { head = &kprobe_table[i]; hlist_for_each_entry_rcu(p, node, head, hlist) - if (!kprobe_gone(p)) + if (!kprobe_disabled(p)) arch_arm_kprobe(p); } + mutex_unlock(&text_mutex); - kprobe_enabled = true; + kprobes_all_disarmed = false; printk(KERN_INFO "Kprobes globally enabled\n"); already_enabled: @@ -1295,7 +1463,7 @@ already_enabled: return; } -static void __kprobes disable_all_kprobes(void) +static void __kprobes disarm_all_kprobes(void) { struct hlist_head *head; struct hlist_node *node; @@ -1304,20 +1472,22 @@ static void __kprobes disable_all_kprobes(void) mutex_lock(&kprobe_mutex); - /* If kprobes are already disabled, just return */ - if (!kprobe_enabled) + /* If kprobes are already disarmed, just return */ + if (kprobes_all_disarmed) goto already_disabled; - kprobe_enabled = false; + kprobes_all_disarmed = true; printk(KERN_INFO "Kprobes globally disabled\n"); + mutex_lock(&text_mutex); for (i = 0; i < KPROBE_TABLE_SIZE; i++) { head = &kprobe_table[i]; hlist_for_each_entry_rcu(p, node, head, hlist) { - if (!arch_trampoline_kprobe(p) && !kprobe_gone(p)) + if (!arch_trampoline_kprobe(p) && !kprobe_disabled(p)) arch_disarm_kprobe(p); } } + mutex_unlock(&text_mutex); mutex_unlock(&kprobe_mutex); /* Allow all currently running kprobes to complete */ synchronize_sched(); @@ -1338,7 +1508,7 @@ static ssize_t read_enabled_file_bool(struct file *file, { char buf[3]; - if (kprobe_enabled) + if (!kprobes_all_disarmed) buf[0] = '1'; else buf[0] = '0'; @@ -1361,19 +1531,19 @@ static ssize_t write_enabled_file_bool(struct file *file, case 'y': case 'Y': case '1': - enable_all_kprobes(); + arm_all_kprobes(); break; case 'n': case 'N': case '0': - disable_all_kprobes(); + disarm_all_kprobes(); break; } return count; } -static struct file_operations fops_kp = { +static const struct file_operations fops_kp = { .read = read_enabled_file_bool, .write = write_enabled_file_bool, }; @@ -1409,16 +1579,5 @@ late_initcall(debugfs_kprobe_init); module_init(init_kprobes); -EXPORT_SYMBOL_GPL(register_kprobe); -EXPORT_SYMBOL_GPL(unregister_kprobe); -EXPORT_SYMBOL_GPL(register_kprobes); -EXPORT_SYMBOL_GPL(unregister_kprobes); -EXPORT_SYMBOL_GPL(register_jprobe); -EXPORT_SYMBOL_GPL(unregister_jprobe); -EXPORT_SYMBOL_GPL(register_jprobes); -EXPORT_SYMBOL_GPL(unregister_jprobes); +/* defined in arch/.../kernel/kprobes.c */ EXPORT_SYMBOL_GPL(jprobe_return); -EXPORT_SYMBOL_GPL(register_kretprobe); -EXPORT_SYMBOL_GPL(unregister_kretprobe); -EXPORT_SYMBOL_GPL(register_kretprobes); -EXPORT_SYMBOL_GPL(unregister_kretprobes); diff --git a/kernel/kthread.c b/kernel/kthread.c index 84bbadd4d02..ab7ae57773e 100644 --- a/kernel/kthread.c +++ b/kernel/kthread.c @@ -9,27 +9,22 @@ #include <linux/kthread.h> #include <linux/completion.h> #include <linux/err.h> +#include <linux/cpuset.h> #include <linux/unistd.h> #include <linux/file.h> #include <linux/module.h> #include <linux/mutex.h> -#include <trace/sched.h> - -#define KTHREAD_NICE_LEVEL (-5) +#include <trace/events/sched.h> static DEFINE_SPINLOCK(kthread_create_lock); static LIST_HEAD(kthread_create_list); struct task_struct *kthreadd_task; -DEFINE_TRACE(sched_kthread_stop); -DEFINE_TRACE(sched_kthread_stop_ret); - struct kthread_create_info { /* Information passed to kthread() from kthreadd. */ int (*threadfn)(void *data); void *data; - struct completion started; /* Result passed back to kthread_create() from kthreadd. */ struct task_struct *result; @@ -38,17 +33,13 @@ struct kthread_create_info struct list_head list; }; -struct kthread_stop_info -{ - struct task_struct *k; - int err; - struct completion done; +struct kthread { + int should_stop; + struct completion exited; }; -/* Thread stopping is done by setthing this var: lock serializes - * multiple kthread_stop calls. */ -static DEFINE_MUTEX(kthread_stop_lock); -static struct kthread_stop_info kthread_stop_info; +#define to_kthread(tsk) \ + container_of((tsk)->vfork_done, struct kthread, exited) /** * kthread_should_stop - should this kthread return now? @@ -59,35 +50,35 @@ static struct kthread_stop_info kthread_stop_info; */ int kthread_should_stop(void) { - return (kthread_stop_info.k == current); + return to_kthread(current)->should_stop; } EXPORT_SYMBOL(kthread_should_stop); static int kthread(void *_create) { + /* Copy data: it's on kthread's stack */ struct kthread_create_info *create = _create; - int (*threadfn)(void *data); - void *data; - int ret = -EINTR; + int (*threadfn)(void *data) = create->threadfn; + void *data = create->data; + struct kthread self; + int ret; - /* Copy data: it's on kthread's stack */ - threadfn = create->threadfn; - data = create->data; + self.should_stop = 0; + init_completion(&self.exited); + current->vfork_done = &self.exited; /* OK, tell user we're spawned, wait for stop or wakeup */ __set_current_state(TASK_UNINTERRUPTIBLE); - complete(&create->started); + create->result = current; + complete(&create->done); schedule(); - if (!kthread_should_stop()) + ret = -EINTR; + if (!self.should_stop) ret = threadfn(data); - /* It might have exited on its own, w/o kthread_stop. Check. */ - if (kthread_should_stop()) { - kthread_stop_info.err = ret; - complete(&kthread_stop_info.done); - } - return 0; + /* we can't just return, we must preserve "self" on stack */ + do_exit(ret); } static void create_kthread(struct kthread_create_info *create) @@ -98,21 +89,8 @@ static void create_kthread(struct kthread_create_info *create) pid = kernel_thread(kthread, create, CLONE_FS | CLONE_FILES | SIGCHLD); if (pid < 0) { create->result = ERR_PTR(pid); - } else { - struct sched_param param = { .sched_priority = 0 }; - wait_for_completion(&create->started); - read_lock(&tasklist_lock); - create->result = find_task_by_pid_ns(pid, &init_pid_ns); - read_unlock(&tasklist_lock); - /* - * root may have changed our (kthreadd's) priority or CPU mask. - * The kernel thread should not inherit these properties. - */ - sched_setscheduler(create->result, SCHED_NORMAL, ¶m); - set_user_nice(create->result, KTHREAD_NICE_LEVEL); - set_cpus_allowed_ptr(create->result, cpu_all_mask); + complete(&create->done); } - complete(&create->done); } /** @@ -143,7 +121,6 @@ struct task_struct *kthread_create(int (*threadfn)(void *data), create.threadfn = threadfn; create.data = data; - init_completion(&create.started); init_completion(&create.done); spin_lock(&kthread_create_lock); @@ -154,78 +131,57 @@ struct task_struct *kthread_create(int (*threadfn)(void *data), wait_for_completion(&create.done); if (!IS_ERR(create.result)) { + struct sched_param param = { .sched_priority = 0 }; va_list args; + va_start(args, namefmt); vsnprintf(create.result->comm, sizeof(create.result->comm), namefmt, args); va_end(args); + /* + * root may have changed our (kthreadd's) priority or CPU mask. + * The kernel thread should not inherit these properties. + */ + sched_setscheduler_nocheck(create.result, SCHED_NORMAL, ¶m); + set_cpus_allowed_ptr(create.result, cpu_all_mask); } return create.result; } EXPORT_SYMBOL(kthread_create); /** - * kthread_bind - bind a just-created kthread to a cpu. - * @k: thread created by kthread_create(). - * @cpu: cpu (might not be online, must be possible) for @k to run on. - * - * Description: This function is equivalent to set_cpus_allowed(), - * except that @cpu doesn't need to be online, and the thread must be - * stopped (i.e., just returned from kthread_create()). - */ -void kthread_bind(struct task_struct *k, unsigned int cpu) -{ - /* Must have done schedule() in kthread() before we set_task_cpu */ - if (!wait_task_inactive(k, TASK_UNINTERRUPTIBLE)) { - WARN_ON(1); - return; - } - set_task_cpu(k, cpu); - k->cpus_allowed = cpumask_of_cpu(cpu); - k->rt.nr_cpus_allowed = 1; - k->flags |= PF_THREAD_BOUND; -} -EXPORT_SYMBOL(kthread_bind); - -/** * kthread_stop - stop a thread created by kthread_create(). * @k: thread created by kthread_create(). * * Sets kthread_should_stop() for @k to return true, wakes it, and - * waits for it to exit. Your threadfn() must not call do_exit() - * itself if you use this function! This can also be called after - * kthread_create() instead of calling wake_up_process(): the thread - * will exit without calling threadfn(). + * waits for it to exit. This can also be called after kthread_create() + * instead of calling wake_up_process(): the thread will exit without + * calling threadfn(). + * + * If threadfn() may call do_exit() itself, the caller must ensure + * task_struct can't go away. * * Returns the result of threadfn(), or %-EINTR if wake_up_process() * was never called. */ int kthread_stop(struct task_struct *k) { + struct kthread *kthread; int ret; - mutex_lock(&kthread_stop_lock); - - /* It could exit after stop_info.k set, but before wake_up_process. */ - get_task_struct(k); - trace_sched_kthread_stop(k); + get_task_struct(k); - /* Must init completion *before* thread sees kthread_stop_info.k */ - init_completion(&kthread_stop_info.done); - smp_wmb(); + kthread = to_kthread(k); + barrier(); /* it might have exited */ + if (k->vfork_done != NULL) { + kthread->should_stop = 1; + wake_up_process(k); + wait_for_completion(&kthread->exited); + } + ret = k->exit_code; - /* Now set kthread_should_stop() to true, and wake it up. */ - kthread_stop_info.k = k; - wake_up_process(k); put_task_struct(k); - - /* Once it dies, reset stop ptr, gather result and we're done. */ - wait_for_completion(&kthread_stop_info.done); - kthread_stop_info.k = NULL; - ret = kthread_stop_info.err; - mutex_unlock(&kthread_stop_lock); - trace_sched_kthread_stop_ret(ret); return ret; @@ -239,8 +195,8 @@ int kthreadd(void *unused) /* Setup a clean context for our children to inherit. */ set_task_comm(tsk, "kthreadd"); ignore_signals(tsk); - set_user_nice(tsk, KTHREAD_NICE_LEVEL); set_cpus_allowed_ptr(tsk, cpu_all_mask); + set_mems_allowed(node_possible_map); current->flags |= PF_NOFREEZE | PF_FREEZER_NOSIG; diff --git a/kernel/lockdep.c b/kernel/lockdep.c index 3673a3f44d9..f5dcd36d315 100644 --- a/kernel/lockdep.c +++ b/kernel/lockdep.c @@ -42,11 +42,15 @@ #include <linux/hash.h> #include <linux/ftrace.h> #include <linux/stringify.h> +#include <linux/bitops.h> #include <asm/sections.h> #include "lockdep_internals.h" +#define CREATE_TRACE_POINTS +#include <trace/events/lock.h> + #ifdef CONFIG_PROVE_LOCKING int prove_locking = 1; module_param(prove_locking, int, 0644); @@ -138,6 +142,11 @@ static inline struct lock_class *hlock_class(struct held_lock *hlock) #ifdef CONFIG_LOCK_STAT static DEFINE_PER_CPU(struct lock_class_stats[MAX_LOCKDEP_KEYS], lock_stats); +static inline u64 lockstat_clock(void) +{ + return cpu_clock(smp_processor_id()); +} + static int lock_point(unsigned long points[], unsigned long ip) { int i; @@ -154,7 +163,7 @@ static int lock_point(unsigned long points[], unsigned long ip) return i; } -static void lock_time_inc(struct lock_time *lt, s64 time) +static void lock_time_inc(struct lock_time *lt, u64 time) { if (time > lt->max) lt->max = time; @@ -230,12 +239,12 @@ static void put_lock_stats(struct lock_class_stats *stats) static void lock_release_holdtime(struct held_lock *hlock) { struct lock_class_stats *stats; - s64 holdtime; + u64 holdtime; if (!lock_stat) return; - holdtime = sched_clock() - hlock->holdtime_stamp; + holdtime = lockstat_clock() - hlock->holdtime_stamp; stats = get_lock_stats(hlock_class(hlock)); if (hlock->read) @@ -363,11 +372,21 @@ static int save_trace(struct stack_trace *trace) save_stack_trace(trace); + /* + * Some daft arches put -1 at the end to indicate its a full trace. + * + * <rant> this is buggy anyway, since it takes a whole extra entry so a + * complete trace that maxes out the entries provided will be reported + * as incomplete, friggin useless </rant> + */ + if (trace->entries[trace->nr_entries-1] == ULONG_MAX) + trace->nr_entries--; + trace->max_entries = trace->nr_entries; nr_stack_trace_entries += trace->nr_entries; - if (nr_stack_trace_entries == MAX_STACK_TRACE_ENTRIES) { + if (nr_stack_trace_entries >= MAX_STACK_TRACE_ENTRIES-1) { if (!debug_locks_off_graph_unlock()) return 0; @@ -385,20 +404,6 @@ unsigned int nr_hardirq_chains; unsigned int nr_softirq_chains; unsigned int nr_process_chains; unsigned int max_lockdep_depth; -unsigned int max_recursion_depth; - -static unsigned int lockdep_dependency_gen_id; - -static bool lockdep_dependency_visit(struct lock_class *source, - unsigned int depth) -{ - if (!depth) - lockdep_dependency_gen_id++; - if (source->dep_gen_id == lockdep_dependency_gen_id) - return true; - source->dep_gen_id = lockdep_dependency_gen_id; - return false; -} #ifdef CONFIG_DEBUG_LOCKDEP /* @@ -428,18 +433,8 @@ atomic_t redundant_softirqs_on; atomic_t redundant_softirqs_off; atomic_t nr_unused_locks; atomic_t nr_cyclic_checks; -atomic_t nr_cyclic_check_recursions; atomic_t nr_find_usage_forwards_checks; -atomic_t nr_find_usage_forwards_recursions; atomic_t nr_find_usage_backwards_checks; -atomic_t nr_find_usage_backwards_recursions; -# define debug_atomic_inc(ptr) atomic_inc(ptr) -# define debug_atomic_dec(ptr) atomic_dec(ptr) -# define debug_atomic_read(ptr) atomic_read(ptr) -#else -# define debug_atomic_inc(ptr) do { } while (0) -# define debug_atomic_dec(ptr) do { } while (0) -# define debug_atomic_read(ptr) 0 #endif /* @@ -555,58 +550,6 @@ static void lockdep_print_held_locks(struct task_struct *curr) } } -static void print_lock_class_header(struct lock_class *class, int depth) -{ - int bit; - - printk("%*s->", depth, ""); - print_lock_name(class); - printk(" ops: %lu", class->ops); - printk(" {\n"); - - for (bit = 0; bit < LOCK_USAGE_STATES; bit++) { - if (class->usage_mask & (1 << bit)) { - int len = depth; - - len += printk("%*s %s", depth, "", usage_str[bit]); - len += printk(" at:\n"); - print_stack_trace(class->usage_traces + bit, len); - } - } - printk("%*s }\n", depth, ""); - - printk("%*s ... key at: ",depth,""); - print_ip_sym((unsigned long)class->key); -} - -/* - * printk all lock dependencies starting at <entry>: - */ -static void __used -print_lock_dependencies(struct lock_class *class, int depth) -{ - struct lock_list *entry; - - if (lockdep_dependency_visit(class, depth)) - return; - - if (DEBUG_LOCKS_WARN_ON(depth >= 20)) - return; - - print_lock_class_header(class, depth); - - list_for_each_entry(entry, &class->locks_after, entry) { - if (DEBUG_LOCKS_WARN_ON(!entry->class)) - return; - - print_lock_dependencies(entry->class, depth + 1); - - printk("%*s ... acquired at:\n",depth,""); - print_stack_trace(&entry->trace, 2); - printk("\n"); - } -} - static void print_kernel_version(void) { printk("%s %.*s\n", init_utsname()->release, @@ -640,6 +583,9 @@ static int static_obj(void *obj) if ((addr >= start) && (addr < end)) return 1; + if (arch_is_kernel_data(addr)) + return 1; + #ifdef CONFIG_SMP /* * percpu var? @@ -799,6 +745,7 @@ register_lock_class(struct lockdep_map *lock, unsigned int subclass, int force) printk("BUG: MAX_LOCKDEP_KEYS too low!\n"); printk("turning off the locking correctness validator.\n"); + dump_stack(); return NULL; } class = lock_classes + nr_lock_classes++; @@ -862,6 +809,7 @@ static struct lock_list *alloc_list_entry(void) printk("BUG: MAX_LOCKDEP_ENTRIES too low!\n"); printk("turning off the locking correctness validator.\n"); + dump_stack(); return NULL; } return list_entries + nr_list_entries++; @@ -900,22 +848,203 @@ static int add_lock_to_list(struct lock_class *class, struct lock_class *this, } /* + * For good efficiency of modular, we use power of 2 + */ +#define MAX_CIRCULAR_QUEUE_SIZE 4096UL +#define CQ_MASK (MAX_CIRCULAR_QUEUE_SIZE-1) + +/* + * The circular_queue and helpers is used to implement the + * breadth-first search(BFS)algorithem, by which we can build + * the shortest path from the next lock to be acquired to the + * previous held lock if there is a circular between them. + */ +struct circular_queue { + unsigned long element[MAX_CIRCULAR_QUEUE_SIZE]; + unsigned int front, rear; +}; + +static struct circular_queue lock_cq; + +unsigned int max_bfs_queue_depth; + +static unsigned int lockdep_dependency_gen_id; + +static inline void __cq_init(struct circular_queue *cq) +{ + cq->front = cq->rear = 0; + lockdep_dependency_gen_id++; +} + +static inline int __cq_empty(struct circular_queue *cq) +{ + return (cq->front == cq->rear); +} + +static inline int __cq_full(struct circular_queue *cq) +{ + return ((cq->rear + 1) & CQ_MASK) == cq->front; +} + +static inline int __cq_enqueue(struct circular_queue *cq, unsigned long elem) +{ + if (__cq_full(cq)) + return -1; + + cq->element[cq->rear] = elem; + cq->rear = (cq->rear + 1) & CQ_MASK; + return 0; +} + +static inline int __cq_dequeue(struct circular_queue *cq, unsigned long *elem) +{ + if (__cq_empty(cq)) + return -1; + + *elem = cq->element[cq->front]; + cq->front = (cq->front + 1) & CQ_MASK; + return 0; +} + +static inline unsigned int __cq_get_elem_count(struct circular_queue *cq) +{ + return (cq->rear - cq->front) & CQ_MASK; +} + +static inline void mark_lock_accessed(struct lock_list *lock, + struct lock_list *parent) +{ + unsigned long nr; + + nr = lock - list_entries; + WARN_ON(nr >= nr_list_entries); + lock->parent = parent; + lock->class->dep_gen_id = lockdep_dependency_gen_id; +} + +static inline unsigned long lock_accessed(struct lock_list *lock) +{ + unsigned long nr; + + nr = lock - list_entries; + WARN_ON(nr >= nr_list_entries); + return lock->class->dep_gen_id == lockdep_dependency_gen_id; +} + +static inline struct lock_list *get_lock_parent(struct lock_list *child) +{ + return child->parent; +} + +static inline int get_lock_depth(struct lock_list *child) +{ + int depth = 0; + struct lock_list *parent; + + while ((parent = get_lock_parent(child))) { + child = parent; + depth++; + } + return depth; +} + +static int __bfs(struct lock_list *source_entry, + void *data, + int (*match)(struct lock_list *entry, void *data), + struct lock_list **target_entry, + int forward) +{ + struct lock_list *entry; + struct list_head *head; + struct circular_queue *cq = &lock_cq; + int ret = 1; + + if (match(source_entry, data)) { + *target_entry = source_entry; + ret = 0; + goto exit; + } + + if (forward) + head = &source_entry->class->locks_after; + else + head = &source_entry->class->locks_before; + + if (list_empty(head)) + goto exit; + + __cq_init(cq); + __cq_enqueue(cq, (unsigned long)source_entry); + + while (!__cq_empty(cq)) { + struct lock_list *lock; + + __cq_dequeue(cq, (unsigned long *)&lock); + + if (!lock->class) { + ret = -2; + goto exit; + } + + if (forward) + head = &lock->class->locks_after; + else + head = &lock->class->locks_before; + + list_for_each_entry(entry, head, entry) { + if (!lock_accessed(entry)) { + unsigned int cq_depth; + mark_lock_accessed(entry, lock); + if (match(entry, data)) { + *target_entry = entry; + ret = 0; + goto exit; + } + + if (__cq_enqueue(cq, (unsigned long)entry)) { + ret = -1; + goto exit; + } + cq_depth = __cq_get_elem_count(cq); + if (max_bfs_queue_depth < cq_depth) + max_bfs_queue_depth = cq_depth; + } + } + } +exit: + return ret; +} + +static inline int __bfs_forwards(struct lock_list *src_entry, + void *data, + int (*match)(struct lock_list *entry, void *data), + struct lock_list **target_entry) +{ + return __bfs(src_entry, data, match, target_entry, 1); + +} + +static inline int __bfs_backwards(struct lock_list *src_entry, + void *data, + int (*match)(struct lock_list *entry, void *data), + struct lock_list **target_entry) +{ + return __bfs(src_entry, data, match, target_entry, 0); + +} + +/* * Recursive, forwards-direction lock-dependency checking, used for * both noncyclic checking and for hardirq-unsafe/softirq-unsafe * checking. - * - * (to keep the stackframe of the recursive functions small we - * use these global variables, and we also mark various helper - * functions as noinline.) */ -static struct held_lock *check_source, *check_target; /* * Print a dependency chain entry (this is only done when a deadlock * has been detected): */ static noinline int -print_circular_bug_entry(struct lock_list *target, unsigned int depth) +print_circular_bug_entry(struct lock_list *target, int depth) { if (debug_locks_silent) return 0; @@ -932,11 +1061,13 @@ print_circular_bug_entry(struct lock_list *target, unsigned int depth) * header first: */ static noinline int -print_circular_bug_header(struct lock_list *entry, unsigned int depth) +print_circular_bug_header(struct lock_list *entry, unsigned int depth, + struct held_lock *check_src, + struct held_lock *check_tgt) { struct task_struct *curr = current; - if (!debug_locks_off_graph_unlock() || debug_locks_silent) + if (debug_locks_silent) return 0; printk("\n=======================================================\n"); @@ -945,9 +1076,9 @@ print_circular_bug_header(struct lock_list *entry, unsigned int depth) printk( "-------------------------------------------------------\n"); printk("%s/%d is trying to acquire lock:\n", curr->comm, task_pid_nr(curr)); - print_lock(check_source); + print_lock(check_src); printk("\nbut task is already holding lock:\n"); - print_lock(check_target); + print_lock(check_tgt); printk("\nwhich lock already depends on the new lock.\n\n"); printk("\nthe existing dependency chain (in reverse order) is:\n"); @@ -956,19 +1087,36 @@ print_circular_bug_header(struct lock_list *entry, unsigned int depth) return 0; } -static noinline int print_circular_bug_tail(void) +static inline int class_equal(struct lock_list *entry, void *data) +{ + return entry->class == data; +} + +static noinline int print_circular_bug(struct lock_list *this, + struct lock_list *target, + struct held_lock *check_src, + struct held_lock *check_tgt) { struct task_struct *curr = current; - struct lock_list this; + struct lock_list *parent; + int depth; - if (debug_locks_silent) + if (!debug_locks_off_graph_unlock() || debug_locks_silent) return 0; - this.class = hlock_class(check_source); - if (!save_trace(&this.trace)) + if (!save_trace(&this->trace)) return 0; - print_circular_bug_entry(&this, 0); + depth = get_lock_depth(target); + + print_circular_bug_header(target, depth, check_src, check_tgt); + + parent = get_lock_parent(target); + + while (parent) { + print_circular_bug_entry(parent, --depth); + parent = get_lock_parent(parent); + } printk("\nother info that might help us debug this:\n\n"); lockdep_print_held_locks(curr); @@ -979,73 +1127,69 @@ static noinline int print_circular_bug_tail(void) return 0; } -#define RECURSION_LIMIT 40 - -static int noinline print_infinite_recursion_bug(void) +static noinline int print_bfs_bug(int ret) { if (!debug_locks_off_graph_unlock()) return 0; - WARN_ON(1); + WARN(1, "lockdep bfs error:%d\n", ret); return 0; } -unsigned long __lockdep_count_forward_deps(struct lock_class *class, - unsigned int depth) +static int noop_count(struct lock_list *entry, void *data) { - struct lock_list *entry; - unsigned long ret = 1; + (*(unsigned long *)data)++; + return 0; +} - if (lockdep_dependency_visit(class, depth)) - return 0; +unsigned long __lockdep_count_forward_deps(struct lock_list *this) +{ + unsigned long count = 0; + struct lock_list *uninitialized_var(target_entry); - /* - * Recurse this class's dependency list: - */ - list_for_each_entry(entry, &class->locks_after, entry) - ret += __lockdep_count_forward_deps(entry->class, depth + 1); + __bfs_forwards(this, (void *)&count, noop_count, &target_entry); - return ret; + return count; } - unsigned long lockdep_count_forward_deps(struct lock_class *class) { unsigned long ret, flags; + struct lock_list this; + + this.parent = NULL; + this.class = class; local_irq_save(flags); __raw_spin_lock(&lockdep_lock); - ret = __lockdep_count_forward_deps(class, 0); + ret = __lockdep_count_forward_deps(&this); __raw_spin_unlock(&lockdep_lock); local_irq_restore(flags); return ret; } -unsigned long __lockdep_count_backward_deps(struct lock_class *class, - unsigned int depth) +unsigned long __lockdep_count_backward_deps(struct lock_list *this) { - struct lock_list *entry; - unsigned long ret = 1; + unsigned long count = 0; + struct lock_list *uninitialized_var(target_entry); - if (lockdep_dependency_visit(class, depth)) - return 0; - /* - * Recurse this class's dependency list: - */ - list_for_each_entry(entry, &class->locks_before, entry) - ret += __lockdep_count_backward_deps(entry->class, depth + 1); + __bfs_backwards(this, (void *)&count, noop_count, &target_entry); - return ret; + return count; } unsigned long lockdep_count_backward_deps(struct lock_class *class) { unsigned long ret, flags; + struct lock_list this; + + this.parent = NULL; + this.class = class; local_irq_save(flags); __raw_spin_lock(&lockdep_lock); - ret = __lockdep_count_backward_deps(class, 0); + ret = __lockdep_count_backward_deps(&this); __raw_spin_unlock(&lockdep_lock); local_irq_restore(flags); @@ -1057,29 +1201,16 @@ unsigned long lockdep_count_backward_deps(struct lock_class *class) * lead to <target>. Print an error and return 0 if it does. */ static noinline int -check_noncircular(struct lock_class *source, unsigned int depth) +check_noncircular(struct lock_list *root, struct lock_class *target, + struct lock_list **target_entry) { - struct lock_list *entry; + int result; - if (lockdep_dependency_visit(source, depth)) - return 1; + debug_atomic_inc(&nr_cyclic_checks); - debug_atomic_inc(&nr_cyclic_check_recursions); - if (depth > max_recursion_depth) - max_recursion_depth = depth; - if (depth >= RECURSION_LIMIT) - return print_infinite_recursion_bug(); - /* - * Check this lock's dependency list: - */ - list_for_each_entry(entry, &source->locks_after, entry) { - if (entry->class == hlock_class(check_target)) - return print_circular_bug_header(entry, depth+1); - debug_atomic_inc(&nr_cyclic_checks); - if (!check_noncircular(entry->class, depth+1)) - return print_circular_bug_entry(entry, depth+1); - } - return 1; + result = __bfs_forwards(root, target, class_equal, target_entry); + + return result; } #if defined(CONFIG_TRACE_IRQFLAGS) && defined(CONFIG_PROVE_LOCKING) @@ -1088,103 +1219,121 @@ check_noncircular(struct lock_class *source, unsigned int depth) * proving that two subgraphs can be connected by a new dependency * without creating any illegal irq-safe -> irq-unsafe lock dependency. */ -static enum lock_usage_bit find_usage_bit; -static struct lock_class *forwards_match, *backwards_match; + +static inline int usage_match(struct lock_list *entry, void *bit) +{ + return entry->class->usage_mask & (1 << (enum lock_usage_bit)bit); +} + + /* * Find a node in the forwards-direction dependency sub-graph starting - * at <source> that matches <find_usage_bit>. + * at @root->class that matches @bit. * - * Return 2 if such a node exists in the subgraph, and put that node - * into <forwards_match>. + * Return 0 if such a node exists in the subgraph, and put that node + * into *@target_entry. * - * Return 1 otherwise and keep <forwards_match> unchanged. - * Return 0 on error. + * Return 1 otherwise and keep *@target_entry unchanged. + * Return <0 on error. */ -static noinline int -find_usage_forwards(struct lock_class *source, unsigned int depth) +static int +find_usage_forwards(struct lock_list *root, enum lock_usage_bit bit, + struct lock_list **target_entry) { - struct lock_list *entry; - int ret; - - if (lockdep_dependency_visit(source, depth)) - return 1; - - if (depth > max_recursion_depth) - max_recursion_depth = depth; - if (depth >= RECURSION_LIMIT) - return print_infinite_recursion_bug(); + int result; debug_atomic_inc(&nr_find_usage_forwards_checks); - if (source->usage_mask & (1 << find_usage_bit)) { - forwards_match = source; - return 2; - } - /* - * Check this lock's dependency list: - */ - list_for_each_entry(entry, &source->locks_after, entry) { - debug_atomic_inc(&nr_find_usage_forwards_recursions); - ret = find_usage_forwards(entry->class, depth+1); - if (ret == 2 || ret == 0) - return ret; - } - return 1; + result = __bfs_forwards(root, (void *)bit, usage_match, target_entry); + + return result; } /* * Find a node in the backwards-direction dependency sub-graph starting - * at <source> that matches <find_usage_bit>. + * at @root->class that matches @bit. * - * Return 2 if such a node exists in the subgraph, and put that node - * into <backwards_match>. + * Return 0 if such a node exists in the subgraph, and put that node + * into *@target_entry. * - * Return 1 otherwise and keep <backwards_match> unchanged. - * Return 0 on error. + * Return 1 otherwise and keep *@target_entry unchanged. + * Return <0 on error. */ -static noinline int -find_usage_backwards(struct lock_class *source, unsigned int depth) +static int +find_usage_backwards(struct lock_list *root, enum lock_usage_bit bit, + struct lock_list **target_entry) { - struct lock_list *entry; - int ret; + int result; - if (lockdep_dependency_visit(source, depth)) - return 1; + debug_atomic_inc(&nr_find_usage_backwards_checks); - if (!__raw_spin_is_locked(&lockdep_lock)) - return DEBUG_LOCKS_WARN_ON(1); + result = __bfs_backwards(root, (void *)bit, usage_match, target_entry); - if (depth > max_recursion_depth) - max_recursion_depth = depth; - if (depth >= RECURSION_LIMIT) - return print_infinite_recursion_bug(); + return result; +} - debug_atomic_inc(&nr_find_usage_backwards_checks); - if (source->usage_mask & (1 << find_usage_bit)) { - backwards_match = source; - return 2; - } +static void print_lock_class_header(struct lock_class *class, int depth) +{ + int bit; - if (!source && debug_locks_off_graph_unlock()) { - WARN_ON(1); - return 0; - } + printk("%*s->", depth, ""); + print_lock_name(class); + printk(" ops: %lu", class->ops); + printk(" {\n"); - /* - * Check this lock's dependency list: - */ - list_for_each_entry(entry, &source->locks_before, entry) { - debug_atomic_inc(&nr_find_usage_backwards_recursions); - ret = find_usage_backwards(entry->class, depth+1); - if (ret == 2 || ret == 0) - return ret; + for (bit = 0; bit < LOCK_USAGE_STATES; bit++) { + if (class->usage_mask & (1 << bit)) { + int len = depth; + + len += printk("%*s %s", depth, "", usage_str[bit]); + len += printk(" at:\n"); + print_stack_trace(class->usage_traces + bit, len); + } } - return 1; + printk("%*s }\n", depth, ""); + + printk("%*s ... key at: ",depth,""); + print_ip_sym((unsigned long)class->key); +} + +/* + * printk the shortest lock dependencies from @start to @end in reverse order: + */ +static void __used +print_shortest_lock_dependencies(struct lock_list *leaf, + struct lock_list *root) +{ + struct lock_list *entry = leaf; + int depth; + + /*compute depth from generated tree by BFS*/ + depth = get_lock_depth(leaf); + + do { + print_lock_class_header(entry->class, depth); + printk("%*s ... acquired at:\n", depth, ""); + print_stack_trace(&entry->trace, 2); + printk("\n"); + + if (depth == 0 && (entry != root)) { + printk("lockdep:%s bad BFS generated tree\n", __func__); + break; + } + + entry = get_lock_parent(entry); + depth--; + } while (entry && (depth >= 0)); + + return; } static int print_bad_irq_dependency(struct task_struct *curr, + struct lock_list *prev_root, + struct lock_list *next_root, + struct lock_list *backwards_entry, + struct lock_list *forwards_entry, struct held_lock *prev, struct held_lock *next, enum lock_usage_bit bit1, @@ -1217,26 +1366,32 @@ print_bad_irq_dependency(struct task_struct *curr, printk("\nbut this new dependency connects a %s-irq-safe lock:\n", irqclass); - print_lock_name(backwards_match); + print_lock_name(backwards_entry->class); printk("\n... which became %s-irq-safe at:\n", irqclass); - print_stack_trace(backwards_match->usage_traces + bit1, 1); + print_stack_trace(backwards_entry->class->usage_traces + bit1, 1); printk("\nto a %s-irq-unsafe lock:\n", irqclass); - print_lock_name(forwards_match); + print_lock_name(forwards_entry->class); printk("\n... which became %s-irq-unsafe at:\n", irqclass); printk("..."); - print_stack_trace(forwards_match->usage_traces + bit2, 1); + print_stack_trace(forwards_entry->class->usage_traces + bit2, 1); printk("\nother info that might help us debug this:\n\n"); lockdep_print_held_locks(curr); - printk("\nthe %s-irq-safe lock's dependencies:\n", irqclass); - print_lock_dependencies(backwards_match, 0); + printk("\nthe dependencies between %s-irq-safe lock", irqclass); + printk(" and the holding lock:\n"); + if (!save_trace(&prev_root->trace)) + return 0; + print_shortest_lock_dependencies(backwards_entry, prev_root); - printk("\nthe %s-irq-unsafe lock's dependencies:\n", irqclass); - print_lock_dependencies(forwards_match, 0); + printk("\nthe dependencies between the lock to be acquired"); + printk(" and %s-irq-unsafe lock:\n", irqclass); + if (!save_trace(&next_root->trace)) + return 0; + print_shortest_lock_dependencies(forwards_entry, next_root); printk("\nstack backtrace:\n"); dump_stack(); @@ -1250,19 +1405,30 @@ check_usage(struct task_struct *curr, struct held_lock *prev, enum lock_usage_bit bit_forwards, const char *irqclass) { int ret; + struct lock_list this, that; + struct lock_list *uninitialized_var(target_entry); + struct lock_list *uninitialized_var(target_entry1); + + this.parent = NULL; - find_usage_bit = bit_backwards; - /* fills in <backwards_match> */ - ret = find_usage_backwards(hlock_class(prev), 0); - if (!ret || ret == 1) + this.class = hlock_class(prev); + ret = find_usage_backwards(&this, bit_backwards, &target_entry); + if (ret < 0) + return print_bfs_bug(ret); + if (ret == 1) return ret; - find_usage_bit = bit_forwards; - ret = find_usage_forwards(hlock_class(next), 0); - if (!ret || ret == 1) + that.parent = NULL; + that.class = hlock_class(next); + ret = find_usage_forwards(&that, bit_forwards, &target_entry1); + if (ret < 0) + return print_bfs_bug(ret); + if (ret == 1) return ret; - /* ret == 2 */ - return print_bad_irq_dependency(curr, prev, next, + + return print_bad_irq_dependency(curr, &this, &that, + target_entry, target_entry1, + prev, next, bit_backwards, bit_forwards, irqclass); } @@ -1474,6 +1640,8 @@ check_prev_add(struct task_struct *curr, struct held_lock *prev, { struct lock_list *entry; int ret; + struct lock_list this; + struct lock_list *uninitialized_var(target_entry); /* * Prove that the new <prev> -> <next> dependency would not @@ -1484,10 +1652,13 @@ check_prev_add(struct task_struct *curr, struct held_lock *prev, * We are using global variables to control the recursion, to * keep the stackframe size of the recursive functions low: */ - check_source = next; - check_target = prev; - if (!(check_noncircular(hlock_class(next), 0))) - return print_circular_bug_tail(); + this.class = hlock_class(next); + this.parent = NULL; + ret = check_noncircular(&this, hlock_class(prev), &target_entry); + if (unlikely(!ret)) + return print_circular_bug(&this, target_entry, next, prev); + else if (unlikely(ret < 0)) + return print_bfs_bug(ret); if (!check_prev_add_irq(curr, prev, next)) return 0; @@ -1688,6 +1859,7 @@ cache_hit: printk("BUG: MAX_LOCKDEP_CHAINS too low!\n"); printk("turning off the locking correctness validator.\n"); + dump_stack(); return 0; } chain = lock_chains + nr_lock_chains++; @@ -1885,7 +2057,8 @@ static int mark_lock(struct task_struct *curr, struct held_lock *this, * print irq inversion bug: */ static int -print_irq_inversion_bug(struct task_struct *curr, struct lock_class *other, +print_irq_inversion_bug(struct task_struct *curr, + struct lock_list *root, struct lock_list *other, struct held_lock *this, int forwards, const char *irqclass) { @@ -1900,20 +2073,19 @@ print_irq_inversion_bug(struct task_struct *curr, struct lock_class *other, curr->comm, task_pid_nr(curr)); print_lock(this); if (forwards) - printk("but this lock took another, %s-irq-unsafe lock in the past:\n", irqclass); + printk("but this lock took another, %s-unsafe lock in the past:\n", irqclass); else - printk("but this lock was taken by another, %s-irq-safe lock in the past:\n", irqclass); - print_lock_name(other); + printk("but this lock was taken by another, %s-safe lock in the past:\n", irqclass); + print_lock_name(other->class); printk("\n\nand interrupts could create inverse lock ordering between them.\n\n"); printk("\nother info that might help us debug this:\n"); lockdep_print_held_locks(curr); - printk("\nthe first lock's dependencies:\n"); - print_lock_dependencies(hlock_class(this), 0); - - printk("\nthe second lock's dependencies:\n"); - print_lock_dependencies(other, 0); + printk("\nthe shortest dependencies between 2nd lock and 1st lock:\n"); + if (!save_trace(&root->trace)) + return 0; + print_shortest_lock_dependencies(other, root); printk("\nstack backtrace:\n"); dump_stack(); @@ -1930,14 +2102,19 @@ check_usage_forwards(struct task_struct *curr, struct held_lock *this, enum lock_usage_bit bit, const char *irqclass) { int ret; - - find_usage_bit = bit; - /* fills in <forwards_match> */ - ret = find_usage_forwards(hlock_class(this), 0); - if (!ret || ret == 1) + struct lock_list root; + struct lock_list *uninitialized_var(target_entry); + + root.parent = NULL; + root.class = hlock_class(this); + ret = find_usage_forwards(&root, bit, &target_entry); + if (ret < 0) + return print_bfs_bug(ret); + if (ret == 1) return ret; - return print_irq_inversion_bug(curr, forwards_match, this, 1, irqclass); + return print_irq_inversion_bug(curr, &root, target_entry, + this, 1, irqclass); } /* @@ -1949,14 +2126,19 @@ check_usage_backwards(struct task_struct *curr, struct held_lock *this, enum lock_usage_bit bit, const char *irqclass) { int ret; - - find_usage_bit = bit; - /* fills in <backwards_match> */ - ret = find_usage_backwards(hlock_class(this), 0); - if (!ret || ret == 1) + struct lock_list root; + struct lock_list *uninitialized_var(target_entry); + + root.parent = NULL; + root.class = hlock_class(this); + ret = find_usage_backwards(&root, bit, &target_entry); + if (ret < 0) + return print_bfs_bug(ret); + if (ret == 1) return ret; - return print_irq_inversion_bug(curr, backwards_match, this, 0, irqclass); + return print_irq_inversion_bug(curr, &root, target_entry, + this, 1, irqclass); } void print_irqtrace_events(struct task_struct *curr) @@ -2015,7 +2197,8 @@ typedef int (*check_usage_f)(struct task_struct *, struct held_lock *, enum lock_usage_bit bit, const char *name); static int -mark_lock_irq(struct task_struct *curr, struct held_lock *this, int new_bit) +mark_lock_irq(struct task_struct *curr, struct held_lock *this, + enum lock_usage_bit new_bit) { int excl_bit = exclusive_bit(new_bit); int read = new_bit & 1; @@ -2043,7 +2226,7 @@ mark_lock_irq(struct task_struct *curr, struct held_lock *this, int new_bit) * states. */ if ((!read || !dir || STRICT_READ_CHECKS) && - !usage(curr, this, excl_bit, state_name(new_bit))) + !usage(curr, this, excl_bit, state_name(new_bit & ~1))) return 0; /* @@ -2492,13 +2675,20 @@ static int mark_lock(struct task_struct *curr, struct held_lock *this, void lockdep_init_map(struct lockdep_map *lock, const char *name, struct lock_class_key *key, int subclass) { - if (unlikely(!debug_locks)) + lock->class_cache = NULL; +#ifdef CONFIG_LOCK_STAT + lock->cpu = raw_smp_processor_id(); +#endif + + if (DEBUG_LOCKS_WARN_ON(!name)) { + lock->name = "NULL"; return; + } + + lock->name = name; if (DEBUG_LOCKS_WARN_ON(!key)) return; - if (DEBUG_LOCKS_WARN_ON(!name)) - return; /* * Sanity check, the lock-class key must be persistent: */ @@ -2507,12 +2697,11 @@ void lockdep_init_map(struct lockdep_map *lock, const char *name, DEBUG_LOCKS_WARN_ON(1); return; } - lock->name = name; lock->key = key; - lock->class_cache = NULL; -#ifdef CONFIG_LOCK_STAT - lock->cpu = raw_smp_processor_id(); -#endif + + if (unlikely(!debug_locks)) + return; + if (subclass) register_lock_class(lock, subclass, 1); } @@ -2524,13 +2713,15 @@ EXPORT_SYMBOL_GPL(lockdep_init_map); */ static int __lock_acquire(struct lockdep_map *lock, unsigned int subclass, int trylock, int read, int check, int hardirqs_off, - struct lockdep_map *nest_lock, unsigned long ip) + struct lockdep_map *nest_lock, unsigned long ip, + int references) { struct task_struct *curr = current; struct lock_class *class = NULL; struct held_lock *hlock; unsigned int depth, id; int chain_head = 0; + int class_idx; u64 chain_key; if (!prove_locking) @@ -2546,6 +2737,7 @@ static int __lock_acquire(struct lockdep_map *lock, unsigned int subclass, debug_locks_off(); printk("BUG: MAX_LOCKDEP_SUBCLASSES too low!\n"); printk("turning off the locking correctness validator.\n"); + dump_stack(); return 0; } @@ -2577,10 +2769,24 @@ static int __lock_acquire(struct lockdep_map *lock, unsigned int subclass, if (DEBUG_LOCKS_WARN_ON(depth >= MAX_LOCK_DEPTH)) return 0; + class_idx = class - lock_classes + 1; + + if (depth) { + hlock = curr->held_locks + depth - 1; + if (hlock->class_idx == class_idx && nest_lock) { + if (hlock->references) + hlock->references++; + else + hlock->references = 2; + + return 1; + } + } + hlock = curr->held_locks + depth; if (DEBUG_LOCKS_WARN_ON(!class)) return 0; - hlock->class_idx = class - lock_classes + 1; + hlock->class_idx = class_idx; hlock->acquire_ip = ip; hlock->instance = lock; hlock->nest_lock = nest_lock; @@ -2588,9 +2794,10 @@ static int __lock_acquire(struct lockdep_map *lock, unsigned int subclass, hlock->read = read; hlock->check = check; hlock->hardirqs_off = !!hardirqs_off; + hlock->references = references; #ifdef CONFIG_LOCK_STAT hlock->waittime_stamp = 0; - hlock->holdtime_stamp = sched_clock(); + hlock->holdtime_stamp = lockstat_clock(); #endif if (check == 2 && !mark_irqflags(curr, hlock)) @@ -2642,6 +2849,7 @@ static int __lock_acquire(struct lockdep_map *lock, unsigned int subclass, debug_locks_off(); printk("BUG: MAX_LOCK_DEPTH too low!\n"); printk("turning off the locking correctness validator.\n"); + dump_stack(); return 0; } @@ -2695,6 +2903,30 @@ static int check_unlock(struct task_struct *curr, struct lockdep_map *lock, return 1; } +static int match_held_lock(struct held_lock *hlock, struct lockdep_map *lock) +{ + if (hlock->instance == lock) + return 1; + + if (hlock->references) { + struct lock_class *class = lock->class_cache; + + if (!class) + class = look_up_lock_class(lock, 0); + + if (DEBUG_LOCKS_WARN_ON(!class)) + return 0; + + if (DEBUG_LOCKS_WARN_ON(!hlock->nest_lock)) + return 0; + + if (hlock->class_idx == class - lock_classes + 1) + return 1; + } + + return 0; +} + static int __lock_set_class(struct lockdep_map *lock, const char *name, struct lock_class_key *key, unsigned int subclass, @@ -2718,7 +2950,7 @@ __lock_set_class(struct lockdep_map *lock, const char *name, */ if (prev_hlock && prev_hlock->irq_context != hlock->irq_context) break; - if (hlock->instance == lock) + if (match_held_lock(hlock, lock)) goto found_it; prev_hlock = hlock; } @@ -2737,7 +2969,8 @@ found_it: if (!__lock_acquire(hlock->instance, hlock_class(hlock)->subclass, hlock->trylock, hlock->read, hlock->check, hlock->hardirqs_off, - hlock->nest_lock, hlock->acquire_ip)) + hlock->nest_lock, hlock->acquire_ip, + hlock->references)) return 0; } @@ -2776,20 +3009,34 @@ lock_release_non_nested(struct task_struct *curr, */ if (prev_hlock && prev_hlock->irq_context != hlock->irq_context) break; - if (hlock->instance == lock) + if (match_held_lock(hlock, lock)) goto found_it; prev_hlock = hlock; } return print_unlock_inbalance_bug(curr, lock, ip); found_it: - lock_release_holdtime(hlock); + if (hlock->instance == lock) + lock_release_holdtime(hlock); + + if (hlock->references) { + hlock->references--; + if (hlock->references) { + /* + * We had, and after removing one, still have + * references, the current lock stack is still + * valid. We're done! + */ + return 1; + } + } /* * We have the right lock to unlock, 'hlock' points to it. * Now we remove it from the stack, and add back the other * entries (if any), recalculating the hash along the way: */ + curr->lockdep_depth = i; curr->curr_chain_key = hlock->prev_chain_key; @@ -2798,7 +3045,8 @@ found_it: if (!__lock_acquire(hlock->instance, hlock_class(hlock)->subclass, hlock->trylock, hlock->read, hlock->check, hlock->hardirqs_off, - hlock->nest_lock, hlock->acquire_ip)) + hlock->nest_lock, hlock->acquire_ip, + hlock->references)) return 0; } @@ -2828,7 +3076,7 @@ static int lock_release_nested(struct task_struct *curr, /* * Is the unlock non-nested: */ - if (hlock->instance != lock) + if (hlock->instance != lock || hlock->references) return lock_release_non_nested(curr, lock, ip); curr->lockdep_depth--; @@ -2873,6 +3121,21 @@ __lock_release(struct lockdep_map *lock, int nested, unsigned long ip) check_chain_key(curr); } +static int __lock_is_held(struct lockdep_map *lock) +{ + struct task_struct *curr = current; + int i; + + for (i = 0; i < curr->lockdep_depth; i++) { + struct held_lock *hlock = curr->held_locks + i; + + if (match_held_lock(hlock, lock)) + return 1; + } + + return 0; +} + /* * Check whether we follow the irq-flags state precisely: */ @@ -2939,6 +3202,8 @@ void lock_acquire(struct lockdep_map *lock, unsigned int subclass, { unsigned long flags; + trace_lock_acquire(lock, subclass, trylock, read, check, nest_lock, ip); + if (unlikely(current->lockdep_recursion)) return; @@ -2947,7 +3212,7 @@ void lock_acquire(struct lockdep_map *lock, unsigned int subclass, current->lockdep_recursion = 1; __lock_acquire(lock, subclass, trylock, read, check, - irqs_disabled_flags(flags), nest_lock, ip); + irqs_disabled_flags(flags), nest_lock, ip, 0); current->lockdep_recursion = 0; raw_local_irq_restore(flags); } @@ -2958,6 +3223,8 @@ void lock_release(struct lockdep_map *lock, int nested, { unsigned long flags; + trace_lock_release(lock, nested, ip); + if (unlikely(current->lockdep_recursion)) return; @@ -2970,6 +3237,26 @@ void lock_release(struct lockdep_map *lock, int nested, } EXPORT_SYMBOL_GPL(lock_release); +int lock_is_held(struct lockdep_map *lock) +{ + unsigned long flags; + int ret = 0; + + if (unlikely(current->lockdep_recursion)) + return ret; + + raw_local_irq_save(flags); + check_flags(flags); + + current->lockdep_recursion = 1; + ret = __lock_is_held(lock); + current->lockdep_recursion = 0; + raw_local_irq_restore(flags); + + return ret; +} +EXPORT_SYMBOL_GPL(lock_is_held); + void lockdep_set_current_reclaim_state(gfp_t gfp_mask) { current->lockdep_reclaim_gfp = gfp_mask; @@ -3029,7 +3316,7 @@ __lock_contended(struct lockdep_map *lock, unsigned long ip) */ if (prev_hlock && prev_hlock->irq_context != hlock->irq_context) break; - if (hlock->instance == lock) + if (match_held_lock(hlock, lock)) goto found_it; prev_hlock = hlock; } @@ -3037,7 +3324,10 @@ __lock_contended(struct lockdep_map *lock, unsigned long ip) return; found_it: - hlock->waittime_stamp = sched_clock(); + if (hlock->instance != lock) + return; + + hlock->waittime_stamp = lockstat_clock(); contention_point = lock_point(hlock_class(hlock)->contention_point, ip); contending_point = lock_point(hlock_class(hlock)->contending_point, @@ -3060,8 +3350,7 @@ __lock_acquired(struct lockdep_map *lock, unsigned long ip) struct held_lock *hlock, *prev_hlock; struct lock_class_stats *stats; unsigned int depth; - u64 now; - s64 waittime = 0; + u64 now, waittime = 0; int i, cpu; depth = curr->lockdep_depth; @@ -3076,7 +3365,7 @@ __lock_acquired(struct lockdep_map *lock, unsigned long ip) */ if (prev_hlock && prev_hlock->irq_context != hlock->irq_context) break; - if (hlock->instance == lock) + if (match_held_lock(hlock, lock)) goto found_it; prev_hlock = hlock; } @@ -3084,13 +3373,18 @@ __lock_acquired(struct lockdep_map *lock, unsigned long ip) return; found_it: + if (hlock->instance != lock) + return; + cpu = smp_processor_id(); if (hlock->waittime_stamp) { - now = sched_clock(); + now = lockstat_clock(); waittime = now - hlock->waittime_stamp; hlock->holdtime_stamp = now; } + trace_lock_acquired(lock, ip, waittime); + stats = get_lock_stats(hlock_class(hlock)); if (waittime) { if (hlock->read) @@ -3110,6 +3404,8 @@ void lock_contended(struct lockdep_map *lock, unsigned long ip) { unsigned long flags; + trace_lock_contended(lock, ip); + if (unlikely(!lock_stat)) return; @@ -3310,7 +3606,12 @@ void __init lockdep_info(void) sizeof(struct list_head) * CLASSHASH_SIZE + sizeof(struct lock_list) * MAX_LOCKDEP_ENTRIES + sizeof(struct lock_chain) * MAX_LOCKDEP_CHAINS + - sizeof(struct list_head) * CHAINHASH_SIZE) / 1024); + sizeof(struct list_head) * CHAINHASH_SIZE +#ifdef CONFIG_PROVE_LOCKING + + sizeof(struct circular_queue) +#endif + ) / 1024 + ); printk(" per task-struct memory footprint: %lu bytes\n", sizeof(struct held_lock) * MAX_LOCK_DEPTH); diff --git a/kernel/lockdep_internals.h b/kernel/lockdep_internals.h index a2cc7e9a6e8..a2ee95ad131 100644 --- a/kernel/lockdep_internals.h +++ b/kernel/lockdep_internals.h @@ -54,9 +54,9 @@ enum { * table (if it's not there yet), and we check it for lock order * conflicts and deadlocks. */ -#define MAX_LOCKDEP_ENTRIES 8192UL +#define MAX_LOCKDEP_ENTRIES 16384UL -#define MAX_LOCKDEP_CHAINS_BITS 14 +#define MAX_LOCKDEP_CHAINS_BITS 15 #define MAX_LOCKDEP_CHAINS (1UL << MAX_LOCKDEP_CHAINS_BITS) #define MAX_LOCKDEP_CHAIN_HLOCKS (MAX_LOCKDEP_CHAINS*5) @@ -91,6 +91,8 @@ extern unsigned int nr_process_chains; extern unsigned int max_lockdep_depth; extern unsigned int max_recursion_depth; +extern unsigned int max_bfs_queue_depth; + #ifdef CONFIG_PROVE_LOCKING extern unsigned long lockdep_count_forward_deps(struct lock_class *); extern unsigned long lockdep_count_backward_deps(struct lock_class *); diff --git a/kernel/lockdep_proc.c b/kernel/lockdep_proc.c index d7135aa2d2c..d4aba4f3584 100644 --- a/kernel/lockdep_proc.c +++ b/kernel/lockdep_proc.c @@ -25,38 +25,12 @@ static void *l_next(struct seq_file *m, void *v, loff_t *pos) { - struct lock_class *class; - - (*pos)++; - - if (v == SEQ_START_TOKEN) - class = m->private; - else { - class = v; - - if (class->lock_entry.next != &all_lock_classes) - class = list_entry(class->lock_entry.next, - struct lock_class, lock_entry); - else - class = NULL; - } - - return class; + return seq_list_next(v, &all_lock_classes, pos); } static void *l_start(struct seq_file *m, loff_t *pos) { - struct lock_class *class; - loff_t i = 0; - - if (*pos == 0) - return SEQ_START_TOKEN; - - list_for_each_entry(class, &all_lock_classes, lock_entry) { - if (++i == *pos) - return class; - } - return NULL; + return seq_list_start_head(&all_lock_classes, *pos); } static void l_stop(struct seq_file *m, void *v) @@ -82,11 +56,11 @@ static void print_name(struct seq_file *m, struct lock_class *class) static int l_show(struct seq_file *m, void *v) { - struct lock_class *class = v; + struct lock_class *class = list_entry(v, struct lock_class, lock_entry); struct lock_list *entry; char usage[LOCK_USAGE_CHARS]; - if (v == SEQ_START_TOKEN) { + if (v == &all_lock_classes) { seq_printf(m, "all lock classes:\n"); return 0; } @@ -128,17 +102,7 @@ static const struct seq_operations lockdep_ops = { static int lockdep_open(struct inode *inode, struct file *file) { - int res = seq_open(file, &lockdep_ops); - if (!res) { - struct seq_file *m = file->private_data; - - if (!list_empty(&all_lock_classes)) - m->private = list_entry(all_lock_classes.next, - struct lock_class, lock_entry); - else - m->private = NULL; - } - return res; + return seq_open(file, &lockdep_ops); } static const struct file_operations proc_lockdep_operations = { @@ -149,37 +113,23 @@ static const struct file_operations proc_lockdep_operations = { }; #ifdef CONFIG_PROVE_LOCKING -static void *lc_next(struct seq_file *m, void *v, loff_t *pos) -{ - struct lock_chain *chain; - - (*pos)++; - - if (v == SEQ_START_TOKEN) - chain = m->private; - else { - chain = v; - - if (*pos < nr_lock_chains) - chain = lock_chains + *pos; - else - chain = NULL; - } - - return chain; -} - static void *lc_start(struct seq_file *m, loff_t *pos) { if (*pos == 0) return SEQ_START_TOKEN; - if (*pos < nr_lock_chains) - return lock_chains + *pos; + if (*pos - 1 < nr_lock_chains) + return lock_chains + (*pos - 1); return NULL; } +static void *lc_next(struct seq_file *m, void *v, loff_t *pos) +{ + (*pos)++; + return lc_start(m, pos); +} + static void lc_stop(struct seq_file *m, void *v) { } @@ -220,16 +170,7 @@ static const struct seq_operations lockdep_chains_ops = { static int lockdep_chains_open(struct inode *inode, struct file *file) { - int res = seq_open(file, &lockdep_chains_ops); - if (!res) { - struct seq_file *m = file->private_data; - - if (nr_lock_chains) - m->private = lock_chains; - else - m->private = NULL; - } - return res; + return seq_open(file, &lockdep_chains_ops); } static const struct file_operations proc_lockdep_chains_operations = { @@ -258,16 +199,10 @@ static void lockdep_stats_debug_show(struct seq_file *m) debug_atomic_read(&chain_lookup_hits)); seq_printf(m, " cyclic checks: %11u\n", debug_atomic_read(&nr_cyclic_checks)); - seq_printf(m, " cyclic-check recursions: %11u\n", - debug_atomic_read(&nr_cyclic_check_recursions)); seq_printf(m, " find-mask forwards checks: %11u\n", debug_atomic_read(&nr_find_usage_forwards_checks)); - seq_printf(m, " find-mask forwards recursions: %11u\n", - debug_atomic_read(&nr_find_usage_forwards_recursions)); seq_printf(m, " find-mask backwards checks: %11u\n", debug_atomic_read(&nr_find_usage_backwards_checks)); - seq_printf(m, " find-mask backwards recursions:%11u\n", - debug_atomic_read(&nr_find_usage_backwards_recursions)); seq_printf(m, " hardirq on events: %11u\n", hi1); seq_printf(m, " hardirq off events: %11u\n", hi2); @@ -409,8 +344,10 @@ static int lockdep_stats_show(struct seq_file *m, void *v) nr_unused); seq_printf(m, " max locking depth: %11u\n", max_lockdep_depth); - seq_printf(m, " max recursion depth: %11u\n", - max_recursion_depth); +#ifdef CONFIG_PROVE_LOCKING + seq_printf(m, " max bfs queue depth: %11u\n", + max_bfs_queue_depth); +#endif lockdep_stats_debug_show(m); seq_printf(m, " debug_locks: %11u\n", debug_locks); @@ -438,7 +375,6 @@ struct lock_stat_data { }; struct lock_stat_seq { - struct lock_stat_data *iter; struct lock_stat_data *iter_end; struct lock_stat_data stats[MAX_LOCKDEP_KEYS]; }; @@ -626,34 +562,22 @@ static void seq_header(struct seq_file *m) static void *ls_start(struct seq_file *m, loff_t *pos) { struct lock_stat_seq *data = m->private; + struct lock_stat_data *iter; if (*pos == 0) return SEQ_START_TOKEN; - data->iter = data->stats + *pos; - if (data->iter >= data->iter_end) - data->iter = NULL; + iter = data->stats + (*pos - 1); + if (iter >= data->iter_end) + iter = NULL; - return data->iter; + return iter; } static void *ls_next(struct seq_file *m, void *v, loff_t *pos) { - struct lock_stat_seq *data = m->private; - (*pos)++; - - if (v == SEQ_START_TOKEN) - data->iter = data->stats; - else { - data->iter = v; - data->iter++; - } - - if (data->iter == data->iter_end) - data->iter = NULL; - - return data->iter; + return ls_start(m, pos); } static void ls_stop(struct seq_file *m, void *v) @@ -670,7 +594,7 @@ static int ls_show(struct seq_file *m, void *v) return 0; } -static struct seq_operations lockstat_ops = { +static const struct seq_operations lockstat_ops = { .start = ls_start, .next = ls_next, .stop = ls_stop, @@ -691,7 +615,6 @@ static int lock_stat_open(struct inode *inode, struct file *file) struct lock_stat_data *iter = data->stats; struct seq_file *m = file->private_data; - data->iter = iter; list_for_each_entry(class, &all_lock_classes, lock_entry) { iter->class = class; iter->stats = lock_stats(class); @@ -699,7 +622,7 @@ static int lock_stat_open(struct inode *inode, struct file *file) } data->iter_end = iter; - sort(data->stats, data->iter_end - data->iter, + sort(data->stats, data->iter_end - data->stats, sizeof(struct lock_stat_data), lock_stat_cmp, NULL); @@ -734,7 +657,6 @@ static int lock_stat_release(struct inode *inode, struct file *file) struct seq_file *seq = file->private_data; vfree(seq->private); - seq->private = NULL; return seq_release(inode, file); } @@ -758,7 +680,8 @@ static int __init lockdep_proc_init(void) &proc_lockdep_stats_operations); #ifdef CONFIG_LOCK_STAT - proc_create("lock_stat", S_IRUSR, NULL, &proc_lock_stat_operations); + proc_create("lock_stat", S_IRUSR | S_IWUSR, NULL, + &proc_lock_stat_operations); #endif return 0; diff --git a/kernel/marker.c b/kernel/marker.c deleted file mode 100644 index ea54f264786..00000000000 --- a/kernel/marker.c +++ /dev/null @@ -1,930 +0,0 @@ -/* - * Copyright (C) 2007 Mathieu Desnoyers - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or - * (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software - * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. - */ -#include <linux/module.h> -#include <linux/mutex.h> -#include <linux/types.h> -#include <linux/jhash.h> -#include <linux/list.h> -#include <linux/rcupdate.h> -#include <linux/marker.h> -#include <linux/err.h> -#include <linux/slab.h> - -extern struct marker __start___markers[]; -extern struct marker __stop___markers[]; - -/* Set to 1 to enable marker debug output */ -static const int marker_debug; - -/* - * markers_mutex nests inside module_mutex. Markers mutex protects the builtin - * and module markers and the hash table. - */ -static DEFINE_MUTEX(markers_mutex); - -/* - * Marker hash table, containing the active markers. - * Protected by module_mutex. - */ -#define MARKER_HASH_BITS 6 -#define MARKER_TABLE_SIZE (1 << MARKER_HASH_BITS) -static struct hlist_head marker_table[MARKER_TABLE_SIZE]; - -/* - * Note about RCU : - * It is used to make sure every handler has finished using its private data - * between two consecutive operation (add or remove) on a given marker. It is - * also used to delay the free of multiple probes array until a quiescent state - * is reached. - * marker entries modifications are protected by the markers_mutex. - */ -struct marker_entry { - struct hlist_node hlist; - char *format; - /* Probe wrapper */ - void (*call)(const struct marker *mdata, void *call_private, ...); - struct marker_probe_closure single; - struct marker_probe_closure *multi; - int refcount; /* Number of times armed. 0 if disarmed. */ - struct rcu_head rcu; - void *oldptr; - int rcu_pending; - unsigned char ptype:1; - unsigned char format_allocated:1; - char name[0]; /* Contains name'\0'format'\0' */ -}; - -/** - * __mark_empty_function - Empty probe callback - * @probe_private: probe private data - * @call_private: call site private data - * @fmt: format string - * @...: variable argument list - * - * Empty callback provided as a probe to the markers. By providing this to a - * disabled marker, we make sure the execution flow is always valid even - * though the function pointer change and the marker enabling are two distinct - * operations that modifies the execution flow of preemptible code. - */ -notrace void __mark_empty_function(void *probe_private, void *call_private, - const char *fmt, va_list *args) -{ -} -EXPORT_SYMBOL_GPL(__mark_empty_function); - -/* - * marker_probe_cb Callback that prepares the variable argument list for probes. - * @mdata: pointer of type struct marker - * @call_private: caller site private data - * @...: Variable argument list. - * - * Since we do not use "typical" pointer based RCU in the 1 argument case, we - * need to put a full smp_rmb() in this branch. This is why we do not use - * rcu_dereference() for the pointer read. - */ -notrace void marker_probe_cb(const struct marker *mdata, - void *call_private, ...) -{ - va_list args; - char ptype; - - /* - * rcu_read_lock_sched does two things : disabling preemption to make - * sure the teardown of the callbacks can be done correctly when they - * are in modules and they insure RCU read coherency. - */ - rcu_read_lock_sched_notrace(); - ptype = mdata->ptype; - if (likely(!ptype)) { - marker_probe_func *func; - /* Must read the ptype before ptr. They are not data dependant, - * so we put an explicit smp_rmb() here. */ - smp_rmb(); - func = mdata->single.func; - /* Must read the ptr before private data. They are not data - * dependant, so we put an explicit smp_rmb() here. */ - smp_rmb(); - va_start(args, call_private); - func(mdata->single.probe_private, call_private, mdata->format, - &args); - va_end(args); - } else { - struct marker_probe_closure *multi; - int i; - /* - * Read mdata->ptype before mdata->multi. - */ - smp_rmb(); - multi = mdata->multi; - /* - * multi points to an array, therefore accessing the array - * depends on reading multi. However, even in this case, - * we must insure that the pointer is read _before_ the array - * data. Same as rcu_dereference, but we need a full smp_rmb() - * in the fast path, so put the explicit barrier here. - */ - smp_read_barrier_depends(); - for (i = 0; multi[i].func; i++) { - va_start(args, call_private); - multi[i].func(multi[i].probe_private, call_private, - mdata->format, &args); - va_end(args); - } - } - rcu_read_unlock_sched_notrace(); -} -EXPORT_SYMBOL_GPL(marker_probe_cb); - -/* - * marker_probe_cb Callback that does not prepare the variable argument list. - * @mdata: pointer of type struct marker - * @call_private: caller site private data - * @...: Variable argument list. - * - * Should be connected to markers "MARK_NOARGS". - */ -static notrace void marker_probe_cb_noarg(const struct marker *mdata, - void *call_private, ...) -{ - va_list args; /* not initialized */ - char ptype; - - rcu_read_lock_sched_notrace(); - ptype = mdata->ptype; - if (likely(!ptype)) { - marker_probe_func *func; - /* Must read the ptype before ptr. They are not data dependant, - * so we put an explicit smp_rmb() here. */ - smp_rmb(); - func = mdata->single.func; - /* Must read the ptr before private data. They are not data - * dependant, so we put an explicit smp_rmb() here. */ - smp_rmb(); - func(mdata->single.probe_private, call_private, mdata->format, - &args); - } else { - struct marker_probe_closure *multi; - int i; - /* - * Read mdata->ptype before mdata->multi. - */ - smp_rmb(); - multi = mdata->multi; - /* - * multi points to an array, therefore accessing the array - * depends on reading multi. However, even in this case, - * we must insure that the pointer is read _before_ the array - * data. Same as rcu_dereference, but we need a full smp_rmb() - * in the fast path, so put the explicit barrier here. - */ - smp_read_barrier_depends(); - for (i = 0; multi[i].func; i++) - multi[i].func(multi[i].probe_private, call_private, - mdata->format, &args); - } - rcu_read_unlock_sched_notrace(); -} - -static void free_old_closure(struct rcu_head *head) -{ - struct marker_entry *entry = container_of(head, - struct marker_entry, rcu); - kfree(entry->oldptr); - /* Make sure we free the data before setting the pending flag to 0 */ - smp_wmb(); - entry->rcu_pending = 0; -} - -static void debug_print_probes(struct marker_entry *entry) -{ - int i; - - if (!marker_debug) - return; - - if (!entry->ptype) { - printk(KERN_DEBUG "Single probe : %p %p\n", - entry->single.func, - entry->single.probe_private); - } else { - for (i = 0; entry->multi[i].func; i++) - printk(KERN_DEBUG "Multi probe %d : %p %p\n", i, - entry->multi[i].func, - entry->multi[i].probe_private); - } -} - -static struct marker_probe_closure * -marker_entry_add_probe(struct marker_entry *entry, - marker_probe_func *probe, void *probe_private) -{ - int nr_probes = 0; - struct marker_probe_closure *old, *new; - - WARN_ON(!probe); - - debug_print_probes(entry); - old = entry->multi; - if (!entry->ptype) { - if (entry->single.func == probe && - entry->single.probe_private == probe_private) - return ERR_PTR(-EBUSY); - if (entry->single.func == __mark_empty_function) { - /* 0 -> 1 probes */ - entry->single.func = probe; - entry->single.probe_private = probe_private; - entry->refcount = 1; - entry->ptype = 0; - debug_print_probes(entry); - return NULL; - } else { - /* 1 -> 2 probes */ - nr_probes = 1; - old = NULL; - } - } else { - /* (N -> N+1), (N != 0, 1) probes */ - for (nr_probes = 0; old[nr_probes].func; nr_probes++) - if (old[nr_probes].func == probe - && old[nr_probes].probe_private - == probe_private) - return ERR_PTR(-EBUSY); - } - /* + 2 : one for new probe, one for NULL func */ - new = kzalloc((nr_probes + 2) * sizeof(struct marker_probe_closure), - GFP_KERNEL); - if (new == NULL) - return ERR_PTR(-ENOMEM); - if (!old) - new[0] = entry->single; - else - memcpy(new, old, - nr_probes * sizeof(struct marker_probe_closure)); - new[nr_probes].func = probe; - new[nr_probes].probe_private = probe_private; - entry->refcount = nr_probes + 1; - entry->multi = new; - entry->ptype = 1; - debug_print_probes(entry); - return old; -} - -static struct marker_probe_closure * -marker_entry_remove_probe(struct marker_entry *entry, - marker_probe_func *probe, void *probe_private) -{ - int nr_probes = 0, nr_del = 0, i; - struct marker_probe_closure *old, *new; - - old = entry->multi; - - debug_print_probes(entry); - if (!entry->ptype) { - /* 0 -> N is an error */ - WARN_ON(entry->single.func == __mark_empty_function); - /* 1 -> 0 probes */ - WARN_ON(probe && entry->single.func != probe); - WARN_ON(entry->single.probe_private != probe_private); - entry->single.func = __mark_empty_function; - entry->refcount = 0; - entry->ptype = 0; - debug_print_probes(entry); - return NULL; - } else { - /* (N -> M), (N > 1, M >= 0) probes */ - for (nr_probes = 0; old[nr_probes].func; nr_probes++) { - if ((!probe || old[nr_probes].func == probe) - && old[nr_probes].probe_private - == probe_private) - nr_del++; - } - } - - if (nr_probes - nr_del == 0) { - /* N -> 0, (N > 1) */ - entry->single.func = __mark_empty_function; - entry->refcount = 0; - entry->ptype = 0; - } else if (nr_probes - nr_del == 1) { - /* N -> 1, (N > 1) */ - for (i = 0; old[i].func; i++) - if ((probe && old[i].func != probe) || - old[i].probe_private != probe_private) - entry->single = old[i]; - entry->refcount = 1; - entry->ptype = 0; - } else { - int j = 0; - /* N -> M, (N > 1, M > 1) */ - /* + 1 for NULL */ - new = kzalloc((nr_probes - nr_del + 1) - * sizeof(struct marker_probe_closure), GFP_KERNEL); - if (new == NULL) - return ERR_PTR(-ENOMEM); - for (i = 0; old[i].func; i++) - if ((probe && old[i].func != probe) || - old[i].probe_private != probe_private) - new[j++] = old[i]; - entry->refcount = nr_probes - nr_del; - entry->ptype = 1; - entry->multi = new; - } - debug_print_probes(entry); - return old; -} - -/* - * Get marker if the marker is present in the marker hash table. - * Must be called with markers_mutex held. - * Returns NULL if not present. - */ -static struct marker_entry *get_marker(const char *name) -{ - struct hlist_head *head; - struct hlist_node *node; - struct marker_entry *e; - u32 hash = jhash(name, strlen(name), 0); - - head = &marker_table[hash & ((1 << MARKER_HASH_BITS)-1)]; - hlist_for_each_entry(e, node, head, hlist) { - if (!strcmp(name, e->name)) - return e; - } - return NULL; -} - -/* - * Add the marker to the marker hash table. Must be called with markers_mutex - * held. - */ -static struct marker_entry *add_marker(const char *name, const char *format) -{ - struct hlist_head *head; - struct hlist_node *node; - struct marker_entry *e; - size_t name_len = strlen(name) + 1; - size_t format_len = 0; - u32 hash = jhash(name, name_len-1, 0); - - if (format) - format_len = strlen(format) + 1; - head = &marker_table[hash & ((1 << MARKER_HASH_BITS)-1)]; - hlist_for_each_entry(e, node, head, hlist) { - if (!strcmp(name, e->name)) { - printk(KERN_NOTICE - "Marker %s busy\n", name); - return ERR_PTR(-EBUSY); /* Already there */ - } - } - /* - * Using kmalloc here to allocate a variable length element. Could - * cause some memory fragmentation if overused. - */ - e = kmalloc(sizeof(struct marker_entry) + name_len + format_len, - GFP_KERNEL); - if (!e) - return ERR_PTR(-ENOMEM); - memcpy(&e->name[0], name, name_len); - if (format) { - e->format = &e->name[name_len]; - memcpy(e->format, format, format_len); - if (strcmp(e->format, MARK_NOARGS) == 0) - e->call = marker_probe_cb_noarg; - else - e->call = marker_probe_cb; - trace_mark(core_marker_format, "name %s format %s", - e->name, e->format); - } else { - e->format = NULL; - e->call = marker_probe_cb; - } - e->single.func = __mark_empty_function; - e->single.probe_private = NULL; - e->multi = NULL; - e->ptype = 0; - e->format_allocated = 0; - e->refcount = 0; - e->rcu_pending = 0; - hlist_add_head(&e->hlist, head); - return e; -} - -/* - * Remove the marker from the marker hash table. Must be called with mutex_lock - * held. - */ -static int remove_marker(const char *name) -{ - struct hlist_head *head; - struct hlist_node *node; - struct marker_entry *e; - int found = 0; - size_t len = strlen(name) + 1; - u32 hash = jhash(name, len-1, 0); - - head = &marker_table[hash & ((1 << MARKER_HASH_BITS)-1)]; - hlist_for_each_entry(e, node, head, hlist) { - if (!strcmp(name, e->name)) { - found = 1; - break; - } - } - if (!found) - return -ENOENT; - if (e->single.func != __mark_empty_function) - return -EBUSY; - hlist_del(&e->hlist); - if (e->format_allocated) - kfree(e->format); - /* Make sure the call_rcu has been executed */ - if (e->rcu_pending) - rcu_barrier_sched(); - kfree(e); - return 0; -} - -/* - * Set the mark_entry format to the format found in the element. - */ -static int marker_set_format(struct marker_entry *entry, const char *format) -{ - entry->format = kstrdup(format, GFP_KERNEL); - if (!entry->format) - return -ENOMEM; - entry->format_allocated = 1; - - trace_mark(core_marker_format, "name %s format %s", - entry->name, entry->format); - return 0; -} - -/* - * Sets the probe callback corresponding to one marker. - */ -static int set_marker(struct marker_entry *entry, struct marker *elem, - int active) -{ - int ret = 0; - WARN_ON(strcmp(entry->name, elem->name) != 0); - - if (entry->format) { - if (strcmp(entry->format, elem->format) != 0) { - printk(KERN_NOTICE - "Format mismatch for probe %s " - "(%s), marker (%s)\n", - entry->name, - entry->format, - elem->format); - return -EPERM; - } - } else { - ret = marker_set_format(entry, elem->format); - if (ret) - return ret; - } - - /* - * probe_cb setup (statically known) is done here. It is - * asynchronous with the rest of execution, therefore we only - * pass from a "safe" callback (with argument) to an "unsafe" - * callback (does not set arguments). - */ - elem->call = entry->call; - /* - * Sanity check : - * We only update the single probe private data when the ptr is - * set to a _non_ single probe! (0 -> 1 and N -> 1, N != 1) - */ - WARN_ON(elem->single.func != __mark_empty_function - && elem->single.probe_private != entry->single.probe_private - && !elem->ptype); - elem->single.probe_private = entry->single.probe_private; - /* - * Make sure the private data is valid when we update the - * single probe ptr. - */ - smp_wmb(); - elem->single.func = entry->single.func; - /* - * We also make sure that the new probe callbacks array is consistent - * before setting a pointer to it. - */ - rcu_assign_pointer(elem->multi, entry->multi); - /* - * Update the function or multi probe array pointer before setting the - * ptype. - */ - smp_wmb(); - elem->ptype = entry->ptype; - - if (elem->tp_name && (active ^ elem->state)) { - WARN_ON(!elem->tp_cb); - /* - * It is ok to directly call the probe registration because type - * checking has been done in the __trace_mark_tp() macro. - */ - - if (active) { - /* - * try_module_get should always succeed because we hold - * lock_module() to get the tp_cb address. - */ - ret = try_module_get(__module_text_address( - (unsigned long)elem->tp_cb)); - BUG_ON(!ret); - ret = tracepoint_probe_register_noupdate( - elem->tp_name, - elem->tp_cb); - } else { - ret = tracepoint_probe_unregister_noupdate( - elem->tp_name, - elem->tp_cb); - /* - * tracepoint_probe_update_all() must be called - * before the module containing tp_cb is unloaded. - */ - module_put(__module_text_address( - (unsigned long)elem->tp_cb)); - } - } - elem->state = active; - - return ret; -} - -/* - * Disable a marker and its probe callback. - * Note: only waiting an RCU period after setting elem->call to the empty - * function insures that the original callback is not used anymore. This insured - * by rcu_read_lock_sched around the call site. - */ -static void disable_marker(struct marker *elem) -{ - int ret; - - /* leave "call" as is. It is known statically. */ - if (elem->tp_name && elem->state) { - WARN_ON(!elem->tp_cb); - /* - * It is ok to directly call the probe registration because type - * checking has been done in the __trace_mark_tp() macro. - */ - ret = tracepoint_probe_unregister_noupdate(elem->tp_name, - elem->tp_cb); - WARN_ON(ret); - /* - * tracepoint_probe_update_all() must be called - * before the module containing tp_cb is unloaded. - */ - module_put(__module_text_address((unsigned long)elem->tp_cb)); - } - elem->state = 0; - elem->single.func = __mark_empty_function; - /* Update the function before setting the ptype */ - smp_wmb(); - elem->ptype = 0; /* single probe */ - /* - * Leave the private data and id there, because removal is racy and - * should be done only after an RCU period. These are never used until - * the next initialization anyway. - */ -} - -/** - * marker_update_probe_range - Update a probe range - * @begin: beginning of the range - * @end: end of the range - * - * Updates the probe callback corresponding to a range of markers. - */ -void marker_update_probe_range(struct marker *begin, - struct marker *end) -{ - struct marker *iter; - struct marker_entry *mark_entry; - - mutex_lock(&markers_mutex); - for (iter = begin; iter < end; iter++) { - mark_entry = get_marker(iter->name); - if (mark_entry) { - set_marker(mark_entry, iter, !!mark_entry->refcount); - /* - * ignore error, continue - */ - } else { - disable_marker(iter); - } - } - mutex_unlock(&markers_mutex); -} - -/* - * Update probes, removing the faulty probes. - * - * Internal callback only changed before the first probe is connected to it. - * Single probe private data can only be changed on 0 -> 1 and 2 -> 1 - * transitions. All other transitions will leave the old private data valid. - * This makes the non-atomicity of the callback/private data updates valid. - * - * "special case" updates : - * 0 -> 1 callback - * 1 -> 0 callback - * 1 -> 2 callbacks - * 2 -> 1 callbacks - * Other updates all behave the same, just like the 2 -> 3 or 3 -> 2 updates. - * Site effect : marker_set_format may delete the marker entry (creating a - * replacement). - */ -static void marker_update_probes(void) -{ - /* Core kernel markers */ - marker_update_probe_range(__start___markers, __stop___markers); - /* Markers in modules. */ - module_update_markers(); - tracepoint_probe_update_all(); -} - -/** - * marker_probe_register - Connect a probe to a marker - * @name: marker name - * @format: format string - * @probe: probe handler - * @probe_private: probe private data - * - * private data must be a valid allocated memory address, or NULL. - * Returns 0 if ok, error value on error. - * The probe address must at least be aligned on the architecture pointer size. - */ -int marker_probe_register(const char *name, const char *format, - marker_probe_func *probe, void *probe_private) -{ - struct marker_entry *entry; - int ret = 0; - struct marker_probe_closure *old; - - mutex_lock(&markers_mutex); - entry = get_marker(name); - if (!entry) { - entry = add_marker(name, format); - if (IS_ERR(entry)) - ret = PTR_ERR(entry); - } else if (format) { - if (!entry->format) - ret = marker_set_format(entry, format); - else if (strcmp(entry->format, format)) - ret = -EPERM; - } - if (ret) - goto end; - - /* - * If we detect that a call_rcu is pending for this marker, - * make sure it's executed now. - */ - if (entry->rcu_pending) - rcu_barrier_sched(); - old = marker_entry_add_probe(entry, probe, probe_private); - if (IS_ERR(old)) { - ret = PTR_ERR(old); - goto end; - } - mutex_unlock(&markers_mutex); - marker_update_probes(); - mutex_lock(&markers_mutex); - entry = get_marker(name); - if (!entry) - goto end; - if (entry->rcu_pending) - rcu_barrier_sched(); - entry->oldptr = old; - entry->rcu_pending = 1; - /* write rcu_pending before calling the RCU callback */ - smp_wmb(); - call_rcu_sched(&entry->rcu, free_old_closure); -end: - mutex_unlock(&markers_mutex); - return ret; -} -EXPORT_SYMBOL_GPL(marker_probe_register); - -/** - * marker_probe_unregister - Disconnect a probe from a marker - * @name: marker name - * @probe: probe function pointer - * @probe_private: probe private data - * - * Returns the private data given to marker_probe_register, or an ERR_PTR(). - * We do not need to call a synchronize_sched to make sure the probes have - * finished running before doing a module unload, because the module unload - * itself uses stop_machine(), which insures that every preempt disabled section - * have finished. - */ -int marker_probe_unregister(const char *name, - marker_probe_func *probe, void *probe_private) -{ - struct marker_entry *entry; - struct marker_probe_closure *old; - int ret = -ENOENT; - - mutex_lock(&markers_mutex); - entry = get_marker(name); - if (!entry) - goto end; - if (entry->rcu_pending) - rcu_barrier_sched(); - old = marker_entry_remove_probe(entry, probe, probe_private); - mutex_unlock(&markers_mutex); - marker_update_probes(); - mutex_lock(&markers_mutex); - entry = get_marker(name); - if (!entry) - goto end; - if (entry->rcu_pending) - rcu_barrier_sched(); - entry->oldptr = old; - entry->rcu_pending = 1; - /* write rcu_pending before calling the RCU callback */ - smp_wmb(); - call_rcu_sched(&entry->rcu, free_old_closure); - remove_marker(name); /* Ignore busy error message */ - ret = 0; -end: - mutex_unlock(&markers_mutex); - return ret; -} -EXPORT_SYMBOL_GPL(marker_probe_unregister); - -static struct marker_entry * -get_marker_from_private_data(marker_probe_func *probe, void *probe_private) -{ - struct marker_entry *entry; - unsigned int i; - struct hlist_head *head; - struct hlist_node *node; - - for (i = 0; i < MARKER_TABLE_SIZE; i++) { - head = &marker_table[i]; - hlist_for_each_entry(entry, node, head, hlist) { - if (!entry->ptype) { - if (entry->single.func == probe - && entry->single.probe_private - == probe_private) - return entry; - } else { - struct marker_probe_closure *closure; - closure = entry->multi; - for (i = 0; closure[i].func; i++) { - if (closure[i].func == probe && - closure[i].probe_private - == probe_private) - return entry; - } - } - } - } - return NULL; -} - -/** - * marker_probe_unregister_private_data - Disconnect a probe from a marker - * @probe: probe function - * @probe_private: probe private data - * - * Unregister a probe by providing the registered private data. - * Only removes the first marker found in hash table. - * Return 0 on success or error value. - * We do not need to call a synchronize_sched to make sure the probes have - * finished running before doing a module unload, because the module unload - * itself uses stop_machine(), which insures that every preempt disabled section - * have finished. - */ -int marker_probe_unregister_private_data(marker_probe_func *probe, - void *probe_private) -{ - struct marker_entry *entry; - int ret = 0; - struct marker_probe_closure *old; - - mutex_lock(&markers_mutex); - entry = get_marker_from_private_data(probe, probe_private); - if (!entry) { - ret = -ENOENT; - goto end; - } - if (entry->rcu_pending) - rcu_barrier_sched(); - old = marker_entry_remove_probe(entry, NULL, probe_private); - mutex_unlock(&markers_mutex); - marker_update_probes(); - mutex_lock(&markers_mutex); - entry = get_marker_from_private_data(probe, probe_private); - if (!entry) - goto end; - if (entry->rcu_pending) - rcu_barrier_sched(); - entry->oldptr = old; - entry->rcu_pending = 1; - /* write rcu_pending before calling the RCU callback */ - smp_wmb(); - call_rcu_sched(&entry->rcu, free_old_closure); - remove_marker(entry->name); /* Ignore busy error message */ -end: - mutex_unlock(&markers_mutex); - return ret; -} -EXPORT_SYMBOL_GPL(marker_probe_unregister_private_data); - -/** - * marker_get_private_data - Get a marker's probe private data - * @name: marker name - * @probe: probe to match - * @num: get the nth matching probe's private data - * - * Returns the nth private data pointer (starting from 0) matching, or an - * ERR_PTR. - * Returns the private data pointer, or an ERR_PTR. - * The private data pointer should _only_ be dereferenced if the caller is the - * owner of the data, or its content could vanish. This is mostly used to - * confirm that a caller is the owner of a registered probe. - */ -void *marker_get_private_data(const char *name, marker_probe_func *probe, - int num) -{ - struct hlist_head *head; - struct hlist_node *node; - struct marker_entry *e; - size_t name_len = strlen(name) + 1; - u32 hash = jhash(name, name_len-1, 0); - int i; - - head = &marker_table[hash & ((1 << MARKER_HASH_BITS)-1)]; - hlist_for_each_entry(e, node, head, hlist) { - if (!strcmp(name, e->name)) { - if (!e->ptype) { - if (num == 0 && e->single.func == probe) - return e->single.probe_private; - } else { - struct marker_probe_closure *closure; - int match = 0; - closure = e->multi; - for (i = 0; closure[i].func; i++) { - if (closure[i].func != probe) - continue; - if (match++ == num) - return closure[i].probe_private; - } - } - break; - } - } - return ERR_PTR(-ENOENT); -} -EXPORT_SYMBOL_GPL(marker_get_private_data); - -#ifdef CONFIG_MODULES - -int marker_module_notify(struct notifier_block *self, - unsigned long val, void *data) -{ - struct module *mod = data; - - switch (val) { - case MODULE_STATE_COMING: - marker_update_probe_range(mod->markers, - mod->markers + mod->num_markers); - break; - case MODULE_STATE_GOING: - marker_update_probe_range(mod->markers, - mod->markers + mod->num_markers); - break; - } - return 0; -} - -struct notifier_block marker_module_nb = { - .notifier_call = marker_module_notify, - .priority = 0, -}; - -static int init_markers(void) -{ - return register_module_notifier(&marker_module_nb); -} -__initcall(init_markers); - -#endif /* CONFIG_MODULES */ diff --git a/kernel/module.c b/kernel/module.c index f77ac320d0b..5842a71cf05 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -18,6 +18,7 @@ */ #include <linux/module.h> #include <linux/moduleloader.h> +#include <linux/ftrace_event.h> #include <linux/init.h> #include <linux/kallsyms.h> #include <linux/fs.h> @@ -46,12 +47,19 @@ #include <linux/rculist.h> #include <asm/uaccess.h> #include <asm/cacheflush.h> +#include <asm/mmu_context.h> #include <linux/license.h> #include <asm/sections.h> #include <linux/tracepoint.h> #include <linux/ftrace.h> #include <linux/async.h> #include <linux/percpu.h> +#include <linux/kmemleak.h> + +#define CREATE_TRACE_POINTS +#include <trace/events/module.h> + +EXPORT_TRACEPOINT_SYMBOL(module_get); #if 0 #define DEBUGP printk @@ -68,15 +76,19 @@ /* List of modules, protected by module_mutex or preempt_disable * (delete uses stop_machine/add uses RCU list operations). */ -static DEFINE_MUTEX(module_mutex); +DEFINE_MUTEX(module_mutex); +EXPORT_SYMBOL_GPL(module_mutex); static LIST_HEAD(modules); +/* Block module loading/unloading? */ +int modules_disabled = 0; + /* Waiting for a module to finish initializing? */ static DECLARE_WAIT_QUEUE_HEAD(module_wq); static BLOCKING_NOTIFIER_HEAD(module_notify_list); -/* Bounds of module allocation, for speeding __module_text_address */ +/* Bounds of module allocation, for speeding __module_address */ static unsigned long module_addr_min = -1UL, module_addr_max = 0; int register_module_notifier(struct notifier_block * nb) @@ -186,17 +198,6 @@ extern const unsigned long __start___kcrctab_unused_gpl[]; #define symversion(base, idx) ((base != NULL) ? ((base) + (idx)) : NULL) #endif -struct symsearch { - const struct kernel_symbol *start, *stop; - const unsigned long *crcs; - enum { - NOT_GPL_ONLY, - GPL_ONLY, - WILL_BE_GPL_ONLY, - } licence; - bool unused; -}; - static bool each_symbol_in_section(const struct symsearch *arr, unsigned int arrsize, struct module *owner, @@ -217,10 +218,8 @@ static bool each_symbol_in_section(const struct symsearch *arr, } /* Returns true as soon as fn returns true, otherwise false. */ -static bool each_symbol(bool (*fn)(const struct symsearch *arr, - struct module *owner, - unsigned int symnum, void *data), - void *data) +bool each_symbol(bool (*fn)(const struct symsearch *arr, struct module *owner, + unsigned int symnum, void *data), void *data) { struct module *mod; const struct symsearch arr[] = { @@ -273,6 +272,7 @@ static bool each_symbol(bool (*fn)(const struct symsearch *arr, } return false; } +EXPORT_SYMBOL_GPL(each_symbol); struct find_symbol_arg { /* Input */ @@ -283,7 +283,7 @@ struct find_symbol_arg { /* Output */ struct module *owner; const unsigned long *crc; - unsigned long value; + const struct kernel_symbol *sym; }; static bool find_symbol_in_section(const struct symsearch *syms, @@ -324,17 +324,17 @@ static bool find_symbol_in_section(const struct symsearch *syms, fsa->owner = owner; fsa->crc = symversion(syms->crcs, symnum); - fsa->value = syms->start[symnum].value; + fsa->sym = &syms->start[symnum]; return true; } -/* Find a symbol, return value, (optional) crc and (optional) module - * which owns it */ -static unsigned long find_symbol(const char *name, - struct module **owner, - const unsigned long **crc, - bool gplok, - bool warn) +/* Find a symbol and return it, along with, (optional) crc and + * (optional) module which owns it */ +const struct kernel_symbol *find_symbol(const char *name, + struct module **owner, + const unsigned long **crc, + bool gplok, + bool warn) { struct find_symbol_arg fsa; @@ -347,15 +347,16 @@ static unsigned long find_symbol(const char *name, *owner = fsa.owner; if (crc) *crc = fsa.crc; - return fsa.value; + return fsa.sym; } DEBUGP("Failed to find symbol %s\n", name); - return -ENOENT; + return NULL; } +EXPORT_SYMBOL_GPL(find_symbol); /* Search for module by name: must hold module_mutex. */ -static struct module *find_module(const char *name) +struct module *find_module(const char *name) { struct module *mod; @@ -365,10 +366,11 @@ static struct module *find_module(const char *name) } return NULL; } +EXPORT_SYMBOL_GPL(find_module); #ifdef CONFIG_SMP -#ifdef CONFIG_HAVE_DYNAMIC_PER_CPU_AREA +#ifndef CONFIG_HAVE_LEGACY_PER_CPU_AREA static void *percpu_modalloc(unsigned long size, unsigned long align, const char *name) @@ -393,7 +395,7 @@ static void percpu_modfree(void *freeme) free_percpu(freeme); } -#else /* ... !CONFIG_HAVE_DYNAMIC_PER_CPU_AREA */ +#else /* ... CONFIG_HAVE_LEGACY_PER_CPU_AREA */ /* Number of blocks used and allocated. */ static unsigned int pcpu_num_used, pcpu_num_allocated; @@ -438,6 +440,7 @@ static void *percpu_modalloc(unsigned long size, unsigned long align, unsigned long extra; unsigned int i; void *ptr; + int cpu; if (align > PAGE_SIZE) { printk(KERN_WARNING "%s: per-cpu alignment %li > %li\n", @@ -467,6 +470,11 @@ static void *percpu_modalloc(unsigned long size, unsigned long align, if (!split_block(i, size)) return NULL; + /* add the per-cpu scanning areas */ + for_each_possible_cpu(cpu) + kmemleak_alloc(ptr + per_cpu_offset(cpu), size, 0, + GFP_KERNEL); + /* Mark allocated */ pcpu_size[i] = -pcpu_size[i]; return ptr; @@ -481,6 +489,7 @@ static void percpu_modfree(void *freeme) { unsigned int i; void *ptr = __per_cpu_start + block_size(pcpu_size[0]); + int cpu; /* First entry is core kernel percpu data. */ for (i = 1; i < pcpu_num_used; ptr += block_size(pcpu_size[i]), i++) { @@ -492,6 +501,10 @@ static void percpu_modfree(void *freeme) BUG(); free: + /* remove the per-cpu scanning areas */ + for_each_possible_cpu(cpu) + kmemleak_free(freeme + per_cpu_offset(cpu)); + /* Merge with previous? */ if (pcpu_size[i-1] >= 0) { pcpu_size[i-1] += pcpu_size[i]; @@ -528,7 +541,7 @@ static int percpu_modinit(void) } __initcall(percpu_modinit); -#endif /* CONFIG_HAVE_DYNAMIC_PER_CPU_AREA */ +#endif /* CONFIG_HAVE_LEGACY_PER_CPU_AREA */ static unsigned int find_pcpusec(Elf_Ehdr *hdr, Elf_Shdr *sechdrs, @@ -641,7 +654,7 @@ static int already_uses(struct module *a, struct module *b) } /* Module a uses b */ -static int use_module(struct module *a, struct module *b) +int use_module(struct module *a, struct module *b) { struct module_use *use; int no_warn, err; @@ -674,6 +687,7 @@ static int use_module(struct module *a, struct module *b) no_warn = sysfs_create_link(b->holders_dir, &a->mkobj.kobj, a->name); return 1; } +EXPORT_SYMBOL_GPL(use_module); /* Clear the unload stuff of the module. */ static void module_unload_free(struct module *mod) @@ -785,7 +799,7 @@ SYSCALL_DEFINE2(delete_module, const char __user *, name_user, char name[MODULE_NAME_LEN]; int ret, forced = 0; - if (!capable(CAP_SYS_MODULE)) + if (!capable(CAP_SYS_MODULE) || modules_disabled) return -EPERM; if (strncpy_from_user(name, name_user, MODULE_NAME_LEN-1) < 0) @@ -894,22 +908,26 @@ void __symbol_put(const char *symbol) struct module *owner; preempt_disable(); - if (IS_ERR_VALUE(find_symbol(symbol, &owner, NULL, true, false))) + if (!find_symbol(symbol, &owner, NULL, true, false)) BUG(); module_put(owner); preempt_enable(); } EXPORT_SYMBOL(__symbol_put); +/* Note this assumes addr is a function, which it currently always is. */ void symbol_put_addr(void *addr) { struct module *modaddr; + unsigned long a = (unsigned long)dereference_function_descriptor(addr); - if (core_kernel_text((unsigned long)addr)) + if (core_kernel_text(a)) return; - if (!(modaddr = module_text_address((unsigned long)addr))) - BUG(); + /* module_text_address is safe here: we're supposed to have reference + * to module from symbol_get, so it can't go away. */ + modaddr = __module_text_address(a); + BUG_ON(!modaddr); module_put(modaddr); } EXPORT_SYMBOL_GPL(symbol_put_addr); @@ -930,6 +948,8 @@ void module_put(struct module *module) if (module) { unsigned int cpu = get_cpu(); local_dec(__module_ref_addr(module, cpu)); + trace_module_put(module, _RET_IP_, + local_read(__module_ref_addr(module, cpu))); /* Maybe they're waiting for us to drop reference? */ if (unlikely(!module_is_live(module))) wake_up_process(module->waiter); @@ -949,10 +969,11 @@ static inline void module_unload_free(struct module *mod) { } -static inline int use_module(struct module *a, struct module *b) +int use_module(struct module *a, struct module *b) { return strong_try_module_get(b) == 0; } +EXPORT_SYMBOL_GPL(use_module); static inline void module_unload_init(struct module *mod) { @@ -995,12 +1016,12 @@ static struct module_attribute *modinfo_attrs[] = { static const char vermagic[] = VERMAGIC_STRING; -static int try_to_force_load(struct module *mod, const char *symname) +static int try_to_force_load(struct module *mod, const char *reason) { #ifdef CONFIG_MODULE_FORCE_LOAD if (!test_taint(TAINT_FORCED_MODULE)) - printk("%s: no version for \"%s\" found: kernel tainted.\n", - mod->name, symname); + printk(KERN_WARNING "%s: %s: kernel tainted.\n", + mod->name, reason); add_taint_module(mod, TAINT_FORCED_MODULE); return 0; #else @@ -1057,9 +1078,10 @@ static inline int check_modstruct_version(Elf_Shdr *sechdrs, { const unsigned long *crc; - if (IS_ERR_VALUE(find_symbol("struct_module", NULL, &crc, true, false))) + if (!find_symbol(MODULE_SYMBOL_PREFIX "module_layout", NULL, + &crc, true, false)) BUG(); - return check_version(sechdrs, versindex, "struct_module", mod, crc); + return check_version(sechdrs, versindex, "module_layout", mod, crc); } /* First part is kernel version, which we ignore if module has crcs. */ @@ -1098,25 +1120,25 @@ static inline int same_magic(const char *amagic, const char *bmagic, /* Resolve a symbol for this module. I.e. if we find one, record usage. Must be holding module_mutex. */ -static unsigned long resolve_symbol(Elf_Shdr *sechdrs, - unsigned int versindex, - const char *name, - struct module *mod) +static const struct kernel_symbol *resolve_symbol(Elf_Shdr *sechdrs, + unsigned int versindex, + const char *name, + struct module *mod) { struct module *owner; - unsigned long ret; + const struct kernel_symbol *sym; const unsigned long *crc; - ret = find_symbol(name, &owner, &crc, + sym = find_symbol(name, &owner, &crc, !(mod->taints & (1 << TAINT_PROPRIETARY_MODULE)), true); - if (!IS_ERR_VALUE(ret)) { - /* use_module can fail due to OOM, - or module initialization or unloading */ + /* use_module can fail due to OOM, + or module initialization or unloading */ + if (sym) { if (!check_version(sechdrs, versindex, name, mod, crc) || !use_module(mod, owner)) - ret = -EINVAL; + sym = NULL; } - return ret; + return sym; } /* @@ -1165,7 +1187,8 @@ static void add_sect_attrs(struct module *mod, unsigned int nsect, /* Count loaded sections and allocate structures */ for (i = 0; i < nsect; i++) - if (sechdrs[i].sh_flags & SHF_ALLOC) + if (sechdrs[i].sh_flags & SHF_ALLOC + && sechdrs[i].sh_size) nloaded++; size[0] = ALIGN(sizeof(*sect_attrs) + nloaded * sizeof(sect_attrs->attrs[0]), @@ -1185,6 +1208,8 @@ static void add_sect_attrs(struct module *mod, unsigned int nsect, for (i = 0; i < nsect; i++) { if (! (sechdrs[i].sh_flags & SHF_ALLOC)) continue; + if (!sechdrs[i].sh_size) + continue; sattr->address = sechdrs[i].sh_addr; sattr->name = kstrdup(secstrings + sechdrs[i].sh_name, GFP_KERNEL); @@ -1260,6 +1285,10 @@ static void add_notes_attrs(struct module *mod, unsigned int nsect, struct module_notes_attrs *notes_attrs; struct bin_attribute *nattr; + /* failed to create section attributes, so can't create notes */ + if (!mod->sect_attrs) + return; + /* Count notes sections and allocate structures. */ notes = 0; for (i = 0; i < nsect; i++) @@ -1479,6 +1508,8 @@ static int __unlink_module(void *_mod) /* Free a module, remove from lists, etc (must hold module_mutex). */ static void free_module(struct module *mod) { + trace_module_free(mod); + /* Delete from various lists */ stop_machine(__unlink_module, mod, NULL); remove_notes_attrs(mod); @@ -1491,8 +1522,8 @@ static void free_module(struct module *mod) /* Module unload stuff */ module_unload_free(mod); - /* release any pointers to mcount in this module */ - ftrace_release(mod->module_core, mod->core_size); + /* Free any allocated parameters. */ + destroy_params(mod->kp, mod->num_kp); /* This may be NULL, but that's OK */ module_free(mod, mod->module_init); @@ -1508,22 +1539,24 @@ static void free_module(struct module *mod) /* Finally, free the core (containing the module structure) */ module_free(mod, mod->module_core); + +#ifdef CONFIG_MPU + update_protections(current->mm); +#endif } void *__symbol_get(const char *symbol) { struct module *owner; - unsigned long value; + const struct kernel_symbol *sym; preempt_disable(); - value = find_symbol(symbol, &owner, NULL, true, true); - if (IS_ERR_VALUE(value)) - value = 0; - else if (strong_try_module_get(owner)) - value = 0; + sym = find_symbol(symbol, &owner, NULL, true, true); + if (sym && strong_try_module_get(owner)) + sym = NULL; preempt_enable(); - return (void *)value; + return sym ? (void *)sym->value : NULL; } EXPORT_SYMBOL_GPL(__symbol_get); @@ -1551,8 +1584,7 @@ static int verify_export_symbols(struct module *mod) for (i = 0; i < ARRAY_SIZE(arr); i++) { for (s = arr[i].sym; s < arr[i].sym + arr[i].num; s++) { - if (!IS_ERR_VALUE(find_symbol(s->name, &owner, - NULL, true, false))) { + if (find_symbol(s->name, &owner, NULL, true, false)) { printk(KERN_ERR "%s: exports duplicate symbol %s" " (owned by %s)\n", @@ -1576,6 +1608,7 @@ static int simplify_symbols(Elf_Shdr *sechdrs, unsigned long secbase; unsigned int i, n = sechdrs[symindex].sh_size / sizeof(Elf_Sym); int ret = 0; + const struct kernel_symbol *ksym; for (i = 1; i < n; i++) { switch (sym[i].st_shndx) { @@ -1595,13 +1628,14 @@ static int simplify_symbols(Elf_Shdr *sechdrs, break; case SHN_UNDEF: - sym[i].st_value - = resolve_symbol(sechdrs, versindex, - strtab + sym[i].st_name, mod); - + ksym = resolve_symbol(sechdrs, versindex, + strtab + sym[i].st_name, mod); /* Ok if resolved. */ - if (!IS_ERR_VALUE(sym[i].st_value)) + if (ksym) { + sym[i].st_value = ksym->value; break; + } + /* Ok if weak. */ if (ELF_ST_BIND(sym[i].st_info) == STB_WEAK) break; @@ -1676,8 +1710,7 @@ static void layout_sections(struct module *mod, if ((s->sh_flags & masks[m][0]) != masks[m][0] || (s->sh_flags & masks[m][1]) || s->sh_entsize != ~0UL - || strncmp(secstrings + s->sh_name, - ".init", 5) == 0) + || strstarts(secstrings + s->sh_name, ".init")) continue; s->sh_entsize = get_offset(mod, &mod->core_size, s, i); DEBUGP("\t%s\n", secstrings + s->sh_name); @@ -1694,8 +1727,7 @@ static void layout_sections(struct module *mod, if ((s->sh_flags & masks[m][0]) != masks[m][0] || (s->sh_flags & masks[m][1]) || s->sh_entsize != ~0UL - || strncmp(secstrings + s->sh_name, - ".init", 5) != 0) + || !strstarts(secstrings + s->sh_name, ".init")) continue; s->sh_entsize = (get_offset(mod, &mod->init_size, s, i) | INIT_OFFSET_MASK); @@ -1768,6 +1800,17 @@ static void setup_modinfo(struct module *mod, Elf_Shdr *sechdrs, } } +static void free_modinfo(struct module *mod) +{ + struct module_attribute *attr; + int i; + + for (i = 0; (attr = modinfo_attrs[i]); i++) { + if (attr->free) + attr->free(mod); + } +} + #ifdef CONFIG_KALLSYMS /* lookup symbol in given range of kernel_symbols */ @@ -1828,19 +1871,98 @@ static char elf_type(const Elf_Sym *sym, else return 'b'; } - if (strncmp(secstrings + sechdrs[sym->st_shndx].sh_name, - ".debug", strlen(".debug")) == 0) + if (strstarts(secstrings + sechdrs[sym->st_shndx].sh_name, ".debug")) return 'n'; return '?'; } +static bool is_core_symbol(const Elf_Sym *src, const Elf_Shdr *sechdrs, + unsigned int shnum) +{ + const Elf_Shdr *sec; + + if (src->st_shndx == SHN_UNDEF + || src->st_shndx >= shnum + || !src->st_name) + return false; + + sec = sechdrs + src->st_shndx; + if (!(sec->sh_flags & SHF_ALLOC) +#ifndef CONFIG_KALLSYMS_ALL + || !(sec->sh_flags & SHF_EXECINSTR) +#endif + || (sec->sh_entsize & INIT_OFFSET_MASK)) + return false; + + return true; +} + +static unsigned long layout_symtab(struct module *mod, + Elf_Shdr *sechdrs, + unsigned int symindex, + unsigned int strindex, + const Elf_Ehdr *hdr, + const char *secstrings, + unsigned long *pstroffs, + unsigned long *strmap) +{ + unsigned long symoffs; + Elf_Shdr *symsect = sechdrs + symindex; + Elf_Shdr *strsect = sechdrs + strindex; + const Elf_Sym *src; + const char *strtab; + unsigned int i, nsrc, ndst; + + /* Put symbol section at end of init part of module. */ + symsect->sh_flags |= SHF_ALLOC; + symsect->sh_entsize = get_offset(mod, &mod->init_size, symsect, + symindex) | INIT_OFFSET_MASK; + DEBUGP("\t%s\n", secstrings + symsect->sh_name); + + src = (void *)hdr + symsect->sh_offset; + nsrc = symsect->sh_size / sizeof(*src); + strtab = (void *)hdr + strsect->sh_offset; + for (ndst = i = 1; i < nsrc; ++i, ++src) + if (is_core_symbol(src, sechdrs, hdr->e_shnum)) { + unsigned int j = src->st_name; + + while(!__test_and_set_bit(j, strmap) && strtab[j]) + ++j; + ++ndst; + } + + /* Append room for core symbols at end of core part. */ + symoffs = ALIGN(mod->core_size, symsect->sh_addralign ?: 1); + mod->core_size = symoffs + ndst * sizeof(Elf_Sym); + + /* Put string table section at end of init part of module. */ + strsect->sh_flags |= SHF_ALLOC; + strsect->sh_entsize = get_offset(mod, &mod->init_size, strsect, + strindex) | INIT_OFFSET_MASK; + DEBUGP("\t%s\n", secstrings + strsect->sh_name); + + /* Append room for core symbols' strings at end of core part. */ + *pstroffs = mod->core_size; + __set_bit(0, strmap); + mod->core_size += bitmap_weight(strmap, strsect->sh_size); + + return symoffs; +} + static void add_kallsyms(struct module *mod, Elf_Shdr *sechdrs, + unsigned int shnum, unsigned int symindex, unsigned int strindex, - const char *secstrings) + unsigned long symoffs, + unsigned long stroffs, + const char *secstrings, + unsigned long *strmap) { - unsigned int i; + unsigned int i, ndst; + const Elf_Sym *src; + Elf_Sym *dst; + char *s; mod->symtab = (void *)sechdrs[symindex].sh_addr; mod->num_symtab = sechdrs[symindex].sh_size / sizeof(Elf_Sym); @@ -1850,13 +1972,46 @@ static void add_kallsyms(struct module *mod, for (i = 0; i < mod->num_symtab; i++) mod->symtab[i].st_info = elf_type(&mod->symtab[i], sechdrs, secstrings, mod); + + mod->core_symtab = dst = mod->module_core + symoffs; + src = mod->symtab; + *dst = *src; + for (ndst = i = 1; i < mod->num_symtab; ++i, ++src) { + if (!is_core_symbol(src, sechdrs, shnum)) + continue; + dst[ndst] = *src; + dst[ndst].st_name = bitmap_weight(strmap, dst[ndst].st_name); + ++ndst; + } + mod->core_num_syms = ndst; + + mod->core_strtab = s = mod->module_core + stroffs; + for (*s = 0, i = 1; i < sechdrs[strindex].sh_size; ++i) + if (test_bit(i, strmap)) + *++s = mod->strtab[i]; } #else +static inline unsigned long layout_symtab(struct module *mod, + Elf_Shdr *sechdrs, + unsigned int symindex, + unsigned int strindex, + const Elf_Ehdr *hdr, + const char *secstrings, + unsigned long *pstroffs, + unsigned long *strmap) +{ + return 0; +} + static inline void add_kallsyms(struct module *mod, Elf_Shdr *sechdrs, + unsigned int shnum, unsigned int symindex, unsigned int strindex, - const char *secstrings) + unsigned long symoffs, + unsigned long stroffs, + const char *secstrings, + const unsigned long *strmap) { } #endif /* CONFIG_KALLSYMS */ @@ -1884,6 +2039,36 @@ static void *module_alloc_update_bounds(unsigned long size) return ret; } +#ifdef CONFIG_DEBUG_KMEMLEAK +static void kmemleak_load_module(struct module *mod, Elf_Ehdr *hdr, + Elf_Shdr *sechdrs, char *secstrings) +{ + unsigned int i; + + /* only scan the sections containing data */ + kmemleak_scan_area(mod->module_core, (unsigned long)mod - + (unsigned long)mod->module_core, + sizeof(struct module), GFP_KERNEL); + + for (i = 1; i < hdr->e_shnum; i++) { + if (!(sechdrs[i].sh_flags & SHF_ALLOC)) + continue; + if (strncmp(secstrings + sechdrs[i].sh_name, ".data", 5) != 0 + && strncmp(secstrings + sechdrs[i].sh_name, ".bss", 4) != 0) + continue; + + kmemleak_scan_area(mod->module_core, sechdrs[i].sh_addr - + (unsigned long)mod->module_core, + sechdrs[i].sh_size, GFP_KERNEL); + } +} +#else +static inline void kmemleak_load_module(struct module *mod, Elf_Ehdr *hdr, + Elf_Shdr *sechdrs, char *secstrings) +{ +} +#endif + /* Allocate and load the module: note that size of section 0 is always zero, and we rely on this for optional sections. */ static noinline struct module *load_module(void __user *umod, @@ -1898,12 +2083,11 @@ static noinline struct module *load_module(void __user *umod, unsigned int symindex = 0; unsigned int strindex = 0; unsigned int modindex, versindex, infoindex, pcpuindex; - unsigned int num_kp, num_mcount; - struct kernel_param *kp; struct module *mod; long err = 0; void *percpu = NULL, *ptr = NULL; /* Stops spurious gcc warning */ - unsigned long *mseg; + unsigned long symoffs, stroffs, *strmap; + mm_segment_t old_fs; DEBUGP("load_module: umod=%p, len=%lu, uargs=%p\n", @@ -1916,12 +2100,6 @@ static noinline struct module *load_module(void __user *umod, if (len > 64 * 1024 * 1024 || (hdr = vmalloc(len)) == NULL) return ERR_PTR(-ENOMEM); - /* Create stop_machine threads since the error path relies on - * a non-failing stop_machine call. */ - err = stop_machine_create(); - if (err) - goto free_hdr; - if (copy_from_user(hdr, umod, len) != 0) { err = -EFAULT; goto free_hdr; @@ -1962,7 +2140,7 @@ static noinline struct module *load_module(void __user *umod, } #ifndef CONFIG_MODULE_UNLOAD /* Don't load .exit sections */ - if (strncmp(secstrings+sechdrs[i].sh_name, ".exit", 5) == 0) + if (strstarts(secstrings+sechdrs[i].sh_name, ".exit")) sechdrs[i].sh_flags &= ~(unsigned long)SHF_ALLOC; #endif } @@ -1991,11 +2169,6 @@ static noinline struct module *load_module(void __user *umod, /* Don't keep modinfo and version sections. */ sechdrs[infoindex].sh_flags &= ~(unsigned long)SHF_ALLOC; sechdrs[versindex].sh_flags &= ~(unsigned long)SHF_ALLOC; -#ifdef CONFIG_KALLSYMS - /* Keep symbol and string tables for decoding later. */ - sechdrs[symindex].sh_flags |= SHF_ALLOC; - sechdrs[strindex].sh_flags |= SHF_ALLOC; -#endif /* Check module struct version now, before we try to use module. */ if (!check_modstruct_version(sechdrs, versindex, mod)) { @@ -2006,7 +2179,7 @@ static noinline struct module *load_module(void __user *umod, modmagic = get_modinfo(sechdrs, infoindex, "vermagic"); /* This is allowed: modprobe --force will invalidate it. */ if (!modmagic) { - err = try_to_force_load(mod, "magic"); + err = try_to_force_load(mod, "bad vermagic"); if (err) goto free_hdr; } else if (!same_magic(modmagic, vermagic, versindex)) { @@ -2031,6 +2204,13 @@ static noinline struct module *load_module(void __user *umod, goto free_hdr; } + strmap = kzalloc(BITS_TO_LONGS(sechdrs[strindex].sh_size) + * sizeof(long), GFP_KERNEL); + if (!strmap) { + err = -ENOMEM; + goto free_mod; + } + if (find_module(mod->name)) { err = -EEXIST; goto free_mod; @@ -2060,9 +2240,17 @@ static noinline struct module *load_module(void __user *umod, this is done generically; there doesn't appear to be any special cases for the architectures. */ layout_sections(mod, hdr, sechdrs, secstrings); + symoffs = layout_symtab(mod, sechdrs, symindex, strindex, hdr, + secstrings, &stroffs, strmap); /* Do the allocs. */ ptr = module_alloc_update_bounds(mod->core_size); + /* + * The pointer to this block is stored in the module structure + * which is inside the block. Just mark it as not being a + * leak. + */ + kmemleak_not_leak(ptr); if (!ptr) { err = -ENOMEM; goto free_percpu; @@ -2071,6 +2259,13 @@ static noinline struct module *load_module(void __user *umod, mod->module_core = ptr; ptr = module_alloc_update_bounds(mod->init_size); + /* + * The pointer to this block is stored in the module structure + * which is inside the block. This block doesn't need to be + * scanned as it contains data and code that will be freed + * after the module is initialized. + */ + kmemleak_ignore(ptr); if (!ptr && mod->init_size) { err = -ENOMEM; goto free_core; @@ -2101,6 +2296,7 @@ static noinline struct module *load_module(void __user *umod, } /* Module has been moved. */ mod = (void *)sechdrs[modindex].sh_addr; + kmemleak_load_module(mod, hdr, sechdrs, secstrings); #if defined(CONFIG_MODULE_UNLOAD) && defined(CONFIG_SMP) mod->refptr = percpu_modalloc(sizeof(local_t), __alignof__(local_t), @@ -2144,8 +2340,8 @@ static noinline struct module *load_module(void __user *umod, /* Now we've got everything in the final locations, we can * find optional sections. */ - kp = section_objs(hdr, sechdrs, secstrings, "__param", sizeof(*kp), - &num_kp); + mod->kp = section_objs(hdr, sechdrs, secstrings, "__param", + sizeof(*mod->kp), &mod->num_kp); mod->syms = section_objs(hdr, sechdrs, secstrings, "__ksymtab", sizeof(*mod->syms), &mod->num_syms); mod->crcs = section_addr(hdr, sechdrs, secstrings, "__kcrctab"); @@ -2174,18 +2370,30 @@ static noinline struct module *load_module(void __user *umod, mod->unused_gpl_crcs = section_addr(hdr, sechdrs, secstrings, "__kcrctab_unused_gpl"); #endif - -#ifdef CONFIG_MARKERS - mod->markers = section_objs(hdr, sechdrs, secstrings, "__markers", - sizeof(*mod->markers), &mod->num_markers); +#ifdef CONFIG_CONSTRUCTORS + mod->ctors = section_objs(hdr, sechdrs, secstrings, ".ctors", + sizeof(*mod->ctors), &mod->num_ctors); #endif + #ifdef CONFIG_TRACEPOINTS mod->tracepoints = section_objs(hdr, sechdrs, secstrings, "__tracepoints", sizeof(*mod->tracepoints), &mod->num_tracepoints); #endif - +#ifdef CONFIG_EVENT_TRACING + mod->trace_events = section_objs(hdr, sechdrs, secstrings, + "_ftrace_events", + sizeof(*mod->trace_events), + &mod->num_trace_events); +#endif +#ifdef CONFIG_FTRACE_MCOUNT_RECORD + /* sechdrs[0].sh_size is always zero */ + mod->ftrace_callsites = section_objs(hdr, sechdrs, secstrings, + "__mcount_loc", + sizeof(*mod->ftrace_callsites), + &mod->num_ftrace_callsites); +#endif #ifdef CONFIG_MODVERSIONS if ((mod->num_syms && !mod->crcs) || (mod->num_gpl_syms && !mod->gpl_crcs) @@ -2195,8 +2403,8 @@ static noinline struct module *load_module(void __user *umod, || (mod->num_unused_gpl_syms && !mod->unused_gpl_crcs) #endif ) { - printk(KERN_WARNING "%s: No versions for exported symbols.\n", mod->name); - err = try_to_force_load(mod, "nocrc"); + err = try_to_force_load(mod, + "no versions for exported symbols"); if (err) goto cleanup; } @@ -2238,7 +2446,10 @@ static noinline struct module *load_module(void __user *umod, percpu_modcopy(mod->percpu, (void *)sechdrs[pcpuindex].sh_addr, sechdrs[pcpuindex].sh_size); - add_kallsyms(mod, sechdrs, symindex, strindex, secstrings); + add_kallsyms(mod, sechdrs, hdr->e_shnum, symindex, strindex, + symoffs, stroffs, secstrings, strmap); + kfree(strmap); + strmap = NULL; if (!mod->taints) { struct _ddebug *debug; @@ -2250,11 +2461,6 @@ static noinline struct module *load_module(void __user *umod, dynamic_debug_setup(debug, num_debug); } - /* sechdrs[0].sh_size is always zero */ - mseg = section_objs(hdr, sechdrs, secstrings, "__mcount_loc", - sizeof(*mseg), &num_mcount); - ftrace_init_module(mod, mseg, mseg + num_mcount); - err = module_finalize(hdr, sechdrs, mod); if (err < 0) goto cleanup; @@ -2291,11 +2497,11 @@ static noinline struct module *load_module(void __user *umod, */ list_add_rcu(&mod->list, &modules); - err = parse_args(mod->name, mod->args, kp, num_kp, NULL); + err = parse_args(mod->name, mod->args, mod->kp, mod->num_kp, NULL); if (err < 0) goto unlink; - err = mod_sysfs_setup(mod, kp, num_kp); + err = mod_sysfs_setup(mod, mod->kp, mod->num_kp); if (err < 0) goto unlink; add_sect_attrs(mod, hdr->e_shnum, secstrings, sechdrs); @@ -2304,22 +2510,25 @@ static noinline struct module *load_module(void __user *umod, /* Get rid of temporary copy */ vfree(hdr); - stop_machine_destroy(); + trace_module_load(mod); + /* Done! */ return mod; unlink: - stop_machine(__unlink_module, mod, NULL); + /* Unlink carefully: kallsyms could be walking list. */ + list_del_rcu(&mod->list); + synchronize_sched(); module_arch_cleanup(mod); cleanup: + free_modinfo(mod); kobject_del(&mod->mkobj.kobj); kobject_put(&mod->mkobj.kobj); - ftrace_release(mod->module_core, mod->core_size); free_unload: module_unload_free(mod); - free_init: #if defined(CONFIG_MODULE_UNLOAD) && defined(CONFIG_SMP) percpu_modfree(mod->refptr); + free_init: #endif module_free(mod, mod->module_init); free_core: @@ -2330,9 +2539,9 @@ static noinline struct module *load_module(void __user *umod, percpu_modfree(percpu); free_mod: kfree(args); + kfree(strmap); free_hdr: vfree(hdr); - stop_machine_destroy(); return ERR_PTR(err); truncated: @@ -2341,6 +2550,17 @@ static noinline struct module *load_module(void __user *umod, goto free_hdr; } +/* Call module constructors. */ +static void do_mod_ctors(struct module *mod) +{ +#ifdef CONFIG_CONSTRUCTORS + unsigned long i; + + for (i = 0; i < mod->num_ctors; i++) + mod->ctors[i](); +#endif +} + /* This is where the real work happens */ SYSCALL_DEFINE3(init_module, void __user *, umod, unsigned long, len, const char __user *, uargs) @@ -2349,7 +2569,7 @@ SYSCALL_DEFINE3(init_module, void __user *, umod, int ret = 0; /* Must have permission */ - if (!capable(CAP_SYS_MODULE)) + if (!capable(CAP_SYS_MODULE) || modules_disabled) return -EPERM; /* Only one module load at a time, please */ @@ -2369,6 +2589,7 @@ SYSCALL_DEFINE3(init_module, void __user *, umod, blocking_notifier_call_chain(&module_notify_list, MODULE_STATE_COMING, mod); + do_mod_ctors(mod); /* Start the module */ if (mod->init != NULL) ret = do_one_initcall(mod->init); @@ -2387,9 +2608,9 @@ SYSCALL_DEFINE3(init_module, void __user *, umod, return ret; } if (ret > 0) { - printk(KERN_WARNING "%s: '%s'->init suspiciously returned %d, " - "it should follow 0/-E convention\n" - KERN_WARNING "%s: loading module anyway...\n", + printk(KERN_WARNING +"%s: '%s'->init suspiciously returned %d, it should follow 0/-E convention\n" +"%s: loading module anyway...\n", __func__, mod->name, ret, __func__); dump_stack(); @@ -2401,9 +2622,18 @@ SYSCALL_DEFINE3(init_module, void __user *, umod, blocking_notifier_call_chain(&module_notify_list, MODULE_STATE_LIVE, mod); + /* We need to finish all async code before the module init sequence is done */ + async_synchronize_full(); + mutex_lock(&module_mutex); /* Drop initial reference. */ module_put(mod); + trim_init_extable(mod); +#ifdef CONFIG_KALLSYMS + mod->num_symtab = mod->core_num_syms; + mod->symtab = mod->core_symtab; + mod->strtab = mod->core_strtab; +#endif module_free(mod, mod->module_init); mod->module_init = NULL; mod->init_size = 0; @@ -2609,6 +2839,25 @@ unsigned long module_kallsyms_lookup_name(const char *name) preempt_enable(); return ret; } + +int module_kallsyms_on_each_symbol(int (*fn)(void *, const char *, + struct module *, unsigned long), + void *data) +{ + struct module *mod; + unsigned int i; + int ret; + + list_for_each_entry(mod, &modules, list) { + for (i = 0; i < mod->num_symtab; i++) { + ret = fn(data, mod->strtab + mod->symtab[i].st_name, + mod, mod->symtab[i].st_value); + if (ret != 0) + return ret; + } + } + return 0; +} #endif /* CONFIG_KALLSYMS */ static char *module_flags(struct module *mod, char *buf) @@ -2744,29 +2993,31 @@ const struct exception_table_entry *search_module_extables(unsigned long addr) } /* - * Is this a valid module address? + * is_module_address - is this address inside a module? + * @addr: the address to check. + * + * See is_module_text_address() if you simply want to see if the address + * is code (not data). */ -int is_module_address(unsigned long addr) +bool is_module_address(unsigned long addr) { - struct module *mod; + bool ret; preempt_disable(); - - list_for_each_entry_rcu(mod, &modules, list) { - if (within_module_core(addr, mod)) { - preempt_enable(); - return 1; - } - } - + ret = __module_address(addr) != NULL; preempt_enable(); - return 0; + return ret; } - -/* Is this a valid kernel address? */ -__notrace_funcgraph struct module *__module_text_address(unsigned long addr) +/* + * __module_address - get the module which contains an address. + * @addr: the address. + * + * Must be called with preempt disabled or module mutex held so that + * module doesn't get freed during this. + */ +struct module *__module_address(unsigned long addr) { struct module *mod; @@ -2774,22 +3025,51 @@ __notrace_funcgraph struct module *__module_text_address(unsigned long addr) return NULL; list_for_each_entry_rcu(mod, &modules, list) - if (within(addr, mod->module_init, mod->init_text_size) - || within(addr, mod->module_core, mod->core_text_size)) + if (within_module_core(addr, mod) + || within_module_init(addr, mod)) return mod; return NULL; } +EXPORT_SYMBOL_GPL(__module_address); -struct module *module_text_address(unsigned long addr) +/* + * is_module_text_address - is this address inside module code? + * @addr: the address to check. + * + * See is_module_address() if you simply want to see if the address is + * anywhere in a module. See kernel_text_address() for testing if an + * address corresponds to kernel or module code. + */ +bool is_module_text_address(unsigned long addr) { - struct module *mod; + bool ret; preempt_disable(); - mod = __module_text_address(addr); + ret = __module_text_address(addr) != NULL; preempt_enable(); + return ret; +} + +/* + * __module_text_address - get the module whose code contains an address. + * @addr: the address. + * + * Must be called with preempt disabled or module mutex held so that + * module doesn't get freed during this. + */ +struct module *__module_text_address(unsigned long addr) +{ + struct module *mod = __module_address(addr); + if (mod) { + /* Make sure it's within the text section. */ + if (!within(addr, mod->module_init, mod->init_text_size) + && !within(addr, mod->module_core, mod->core_text_size)) + mod = NULL; + } return mod; } +EXPORT_SYMBOL_GPL(__module_text_address); /* Don't grab lock, we're oopsing. */ void print_modules(void) @@ -2797,7 +3077,7 @@ void print_modules(void) struct module *mod; char buf[8]; - printk("Modules linked in:"); + printk(KERN_DEFAULT "Modules linked in:"); /* Most callers should already have preempt disabled, but make sure */ preempt_disable(); list_for_each_entry_rcu(mod, &modules, list) @@ -2809,23 +3089,16 @@ void print_modules(void) } #ifdef CONFIG_MODVERSIONS -/* Generate the signature for struct module here, too, for modversions. */ -void struct_module(struct module *mod) { return; } -EXPORT_SYMBOL(struct_module); -#endif - -#ifdef CONFIG_MARKERS -void module_update_markers(void) +/* Generate the signature for all relevant module structures here. + * If these change, we don't want to try to parse the module. */ +void module_layout(struct module *mod, + struct modversion_info *ver, + struct kernel_param *kp, + struct kernel_symbol *ks, + struct tracepoint *tp) { - struct module *mod; - - mutex_lock(&module_mutex); - list_for_each_entry(mod, &modules, list) - if (!mod->taints) - marker_update_probe_range(mod->markers, - mod->markers + mod->num_markers); - mutex_unlock(&module_mutex); } +EXPORT_SYMBOL(module_layout); #endif #ifdef CONFIG_TRACEPOINTS diff --git a/kernel/mutex-debug.c b/kernel/mutex-debug.c index 50d022e5a56..ec815a960b5 100644 --- a/kernel/mutex-debug.c +++ b/kernel/mutex-debug.c @@ -16,6 +16,7 @@ #include <linux/delay.h> #include <linux/module.h> #include <linux/poison.h> +#include <linux/sched.h> #include <linux/spinlock.h> #include <linux/kallsyms.h> #include <linux/interrupt.h> diff --git a/kernel/mutex.c b/kernel/mutex.c index 5d79781394a..632f04c57d8 100644 --- a/kernel/mutex.c +++ b/kernel/mutex.c @@ -89,7 +89,7 @@ __mutex_lock_slowpath(atomic_t *lock_count); * * This function is similar to (but not equivalent to) down(). */ -void inline __sched mutex_lock(struct mutex *lock) +void __sched mutex_lock(struct mutex *lock) { might_sleep(); /* @@ -148,7 +148,8 @@ __mutex_lock_common(struct mutex *lock, long state, unsigned int subclass, preempt_disable(); mutex_acquire(&lock->dep_map, subclass, 0, ip); -#if defined(CONFIG_SMP) && !defined(CONFIG_DEBUG_MUTEXES) + +#ifdef CONFIG_MUTEX_SPIN_ON_OWNER /* * Optimistic spinning. * @@ -248,7 +249,9 @@ __mutex_lock_common(struct mutex *lock, long state, unsigned int subclass, /* didnt get the lock, go to sleep: */ spin_unlock_mutex(&lock->wait_lock, flags); - __schedule(); + preempt_enable_no_resched(); + schedule(); + preempt_disable(); spin_lock_mutex(&lock->wait_lock, flags); } @@ -470,5 +473,28 @@ int __sched mutex_trylock(struct mutex *lock) return ret; } - EXPORT_SYMBOL(mutex_trylock); + +/** + * atomic_dec_and_mutex_lock - return holding mutex if we dec to 0 + * @cnt: the atomic which we are to dec + * @lock: the mutex to return holding if we dec to 0 + * + * return true and hold lock if we dec to 0, return false otherwise + */ +int atomic_dec_and_mutex_lock(atomic_t *cnt, struct mutex *lock) +{ + /* dec if we can't possibly hit 0 */ + if (atomic_add_unless(cnt, -1, 1)) + return 0; + /* we might hit 0, so take the lock */ + mutex_lock(lock); + if (!atomic_dec_and_test(cnt)) { + /* when we actually did the dec, we didn't hit 0 */ + mutex_unlock(lock); + return 0; + } + /* we hit 0, and we hold the lock */ + return 1; +} +EXPORT_SYMBOL(atomic_dec_and_mutex_lock); diff --git a/kernel/notifier.c b/kernel/notifier.c index 61d5aa5eced..acd24e7643e 100644 --- a/kernel/notifier.c +++ b/kernel/notifier.c @@ -558,7 +558,7 @@ EXPORT_SYMBOL(unregister_reboot_notifier); static ATOMIC_NOTIFIER_HEAD(die_chain); -int notrace notify_die(enum die_val val, const char *str, +int notrace __kprobes notify_die(enum die_val val, const char *str, struct pt_regs *regs, long err, int trap, int sig) { struct die_args args = { diff --git a/kernel/ns_cgroup.c b/kernel/ns_cgroup.c index 78bc3fdac0d..2a5dfec8efe 100644 --- a/kernel/ns_cgroup.c +++ b/kernel/ns_cgroup.c @@ -34,7 +34,7 @@ int ns_cgroup_clone(struct task_struct *task, struct pid *pid) /* * Rules: - * 1. you can only enter a cgroup which is a child of your current + * 1. you can only enter a cgroup which is a descendant of your current * cgroup * 2. you can only place another process into a cgroup if * a. you have CAP_SYS_ADMIN @@ -42,25 +42,31 @@ int ns_cgroup_clone(struct task_struct *task, struct pid *pid) * (hence either you are in the same cgroup as task, or in an * ancestor cgroup thereof) */ -static int ns_can_attach(struct cgroup_subsys *ss, - struct cgroup *new_cgroup, struct task_struct *task) +static int ns_can_attach(struct cgroup_subsys *ss, struct cgroup *new_cgroup, + struct task_struct *task, bool threadgroup) { - struct cgroup *orig; - if (current != task) { if (!capable(CAP_SYS_ADMIN)) return -EPERM; - if (!cgroup_is_descendant(new_cgroup)) + if (!cgroup_is_descendant(new_cgroup, current)) return -EPERM; } - if (atomic_read(&new_cgroup->count) != 0) + if (!cgroup_is_descendant(new_cgroup, task)) return -EPERM; - orig = task_cgroup(task, ns_subsys_id); - if (orig && orig != new_cgroup->parent) - return -EPERM; + if (threadgroup) { + struct task_struct *c; + rcu_read_lock(); + list_for_each_entry_rcu(c, &task->thread_group, thread_group) { + if (!cgroup_is_descendant(new_cgroup, c)) { + rcu_read_unlock(); + return -EPERM; + } + } + rcu_read_unlock(); + } return 0; } @@ -77,7 +83,7 @@ static struct cgroup_subsys_state *ns_create(struct cgroup_subsys *ss, if (!capable(CAP_SYS_ADMIN)) return ERR_PTR(-EPERM); - if (!cgroup_is_descendant(cgroup)) + if (!cgroup_is_descendant(cgroup, current)) return ERR_PTR(-EPERM); ns_cgroup = kzalloc(sizeof(*ns_cgroup), GFP_KERNEL); diff --git a/kernel/nsproxy.c b/kernel/nsproxy.c index 63598dca2d0..09b4ff9711b 100644 --- a/kernel/nsproxy.c +++ b/kernel/nsproxy.c @@ -26,19 +26,14 @@ static struct kmem_cache *nsproxy_cachep; struct nsproxy init_nsproxy = INIT_NSPROXY(init_nsproxy); -/* - * creates a copy of "orig" with refcount 1. - */ -static inline struct nsproxy *clone_nsproxy(struct nsproxy *orig) +static inline struct nsproxy *create_nsproxy(void) { - struct nsproxy *ns; + struct nsproxy *nsproxy; - ns = kmem_cache_alloc(nsproxy_cachep, GFP_KERNEL); - if (ns) { - memcpy(ns, orig, sizeof(struct nsproxy)); - atomic_set(&ns->count, 1); - } - return ns; + nsproxy = kmem_cache_alloc(nsproxy_cachep, GFP_KERNEL); + if (nsproxy) + atomic_set(&nsproxy->count, 1); + return nsproxy; } /* @@ -52,7 +47,7 @@ static struct nsproxy *create_new_namespaces(unsigned long flags, struct nsproxy *new_nsp; int err; - new_nsp = clone_nsproxy(tsk->nsproxy); + new_nsp = create_nsproxy(); if (!new_nsp) return ERR_PTR(-ENOMEM); diff --git a/kernel/panic.c b/kernel/panic.c index 32fe4eff1b8..96b45d0b4ba 100644 --- a/kernel/panic.c +++ b/kernel/panic.c @@ -8,19 +8,19 @@ * This function is used through-out the kernel (including mm and fs) * to indicate a major problem. */ +#include <linux/debug_locks.h> +#include <linux/interrupt.h> +#include <linux/kallsyms.h> +#include <linux/notifier.h> #include <linux/module.h> -#include <linux/sched.h> -#include <linux/delay.h> +#include <linux/random.h> #include <linux/reboot.h> -#include <linux/notifier.h> -#include <linux/init.h> +#include <linux/delay.h> +#include <linux/kexec.h> +#include <linux/sched.h> #include <linux/sysrq.h> -#include <linux/interrupt.h> +#include <linux/init.h> #include <linux/nmi.h> -#include <linux/kexec.h> -#include <linux/debug_locks.h> -#include <linux/random.h> -#include <linux/kallsyms.h> #include <linux/dmi.h> int panic_on_oops; @@ -52,19 +52,15 @@ EXPORT_SYMBOL(panic_blink); * * This function never returns. */ - NORET_TYPE void panic(const char * fmt, ...) { - long i; static char buf[1024]; va_list args; -#if defined(CONFIG_S390) - unsigned long caller = (unsigned long) __builtin_return_address(0); -#endif + long i; /* - * It's possible to come here directly from a panic-assertion and not - * have preempt disabled. Some functions called from here want + * It's possible to come here directly from a panic-assertion and + * not have preempt disabled. Some functions called from here want * preempt to be disabled. No point enabling it later though... */ preempt_disable(); @@ -77,7 +73,6 @@ NORET_TYPE void panic(const char * fmt, ...) #ifdef CONFIG_DEBUG_BUGVERBOSE dump_stack(); #endif - bust_spinlocks(0); /* * If we have crashed and we have a crash kernel loaded let it handle @@ -86,35 +81,37 @@ NORET_TYPE void panic(const char * fmt, ...) */ crash_kexec(NULL); -#ifdef CONFIG_SMP /* * Note smp_send_stop is the usual smp shutdown function, which * unfortunately means it may not be hardened to work in a panic * situation. */ smp_send_stop(); -#endif atomic_notifier_call_chain(&panic_notifier_list, 0, buf); + bust_spinlocks(0); + if (!panic_blink) panic_blink = no_blink; if (panic_timeout > 0) { /* - * Delay timeout seconds before rebooting the machine. - * We can't use the "normal" timers since we just panicked.. - */ - printk(KERN_EMERG "Rebooting in %d seconds..",panic_timeout); + * Delay timeout seconds before rebooting the machine. + * We can't use the "normal" timers since we just panicked. + */ + printk(KERN_EMERG "Rebooting in %d seconds..", panic_timeout); + for (i = 0; i < panic_timeout*1000; ) { touch_nmi_watchdog(); i += panic_blink(i); mdelay(1); i++; } - /* This will not be a clean reboot, with everything - * shutting down. But if there is a chance of - * rebooting the system it will be rebooted. + /* + * This will not be a clean reboot, with everything + * shutting down. But if there is a chance of + * rebooting the system it will be rebooted. */ emergency_restart(); } @@ -127,10 +124,15 @@ NORET_TYPE void panic(const char * fmt, ...) } #endif #if defined(CONFIG_S390) - disabled_wait(caller); + { + unsigned long caller; + + caller = (unsigned long)__builtin_return_address(0); + disabled_wait(caller); + } #endif local_irq_enable(); - for (i = 0;;) { + for (i = 0; ; ) { touch_softlockup_watchdog(); i += panic_blink(i); mdelay(1); @@ -142,23 +144,23 @@ EXPORT_SYMBOL(panic); struct tnt { - u8 bit; - char true; - char false; + u8 bit; + char true; + char false; }; static const struct tnt tnts[] = { - { TAINT_PROPRIETARY_MODULE, 'P', 'G' }, - { TAINT_FORCED_MODULE, 'F', ' ' }, - { TAINT_UNSAFE_SMP, 'S', ' ' }, - { TAINT_FORCED_RMMOD, 'R', ' ' }, - { TAINT_MACHINE_CHECK, 'M', ' ' }, - { TAINT_BAD_PAGE, 'B', ' ' }, - { TAINT_USER, 'U', ' ' }, - { TAINT_DIE, 'D', ' ' }, - { TAINT_OVERRIDDEN_ACPI_TABLE, 'A', ' ' }, - { TAINT_WARN, 'W', ' ' }, - { TAINT_CRAP, 'C', ' ' }, + { TAINT_PROPRIETARY_MODULE, 'P', 'G' }, + { TAINT_FORCED_MODULE, 'F', ' ' }, + { TAINT_UNSAFE_SMP, 'S', ' ' }, + { TAINT_FORCED_RMMOD, 'R', ' ' }, + { TAINT_MACHINE_CHECK, 'M', ' ' }, + { TAINT_BAD_PAGE, 'B', ' ' }, + { TAINT_USER, 'U', ' ' }, + { TAINT_DIE, 'D', ' ' }, + { TAINT_OVERRIDDEN_ACPI_TABLE, 'A', ' ' }, + { TAINT_WARN, 'W', ' ' }, + { TAINT_CRAP, 'C', ' ' }, }; /** @@ -176,7 +178,7 @@ static const struct tnt tnts[] = { * 'W' - Taint on warning. * 'C' - modules from drivers/staging are loaded. * - * The string is overwritten by the next call to print_taint(). + * The string is overwritten by the next call to print_tainted(). */ const char *print_tainted(void) { @@ -195,7 +197,8 @@ const char *print_tainted(void) *s = 0; } else snprintf(buf, sizeof(buf), "Not tainted"); - return(buf); + + return buf; } int test_taint(unsigned flag) @@ -211,7 +214,16 @@ unsigned long get_taint(void) void add_taint(unsigned flag) { - debug_locks = 0; /* can't trust the integrity of the kernel anymore */ + /* + * Can't trust the integrity of the kernel anymore. + * We don't call directly debug_locks_off() because the issue + * is not necessarily serious enough to set oops_in_progress to 1 + * Also we want to keep up lockdep for staging development and + * post-warning case. + */ + if (flag != TAINT_CRAP && flag != TAINT_WARN && __debug_locks_off()) + printk(KERN_WARNING "Disabling lock debugging due to kernel taint\n"); + set_bit(flag, &tainted_mask); } EXPORT_SYMBOL(add_taint); @@ -266,8 +278,8 @@ static void do_oops_enter_exit(void) } /* - * Return true if the calling CPU is allowed to print oops-related info. This - * is a bit racy.. + * Return true if the calling CPU is allowed to print oops-related info. + * This is a bit racy.. */ int oops_may_print(void) { @@ -276,20 +288,23 @@ int oops_may_print(void) /* * Called when the architecture enters its oops handler, before it prints - * anything. If this is the first CPU to oops, and it's oopsing the first time - * then let it proceed. + * anything. If this is the first CPU to oops, and it's oopsing the first + * time then let it proceed. * - * This is all enabled by the pause_on_oops kernel boot option. We do all this - * to ensure that oopses don't scroll off the screen. It has the side-effect - * of preventing later-oopsing CPUs from mucking up the display, too. + * This is all enabled by the pause_on_oops kernel boot option. We do all + * this to ensure that oopses don't scroll off the screen. It has the + * side-effect of preventing later-oopsing CPUs from mucking up the display, + * too. * - * It turns out that the CPU which is allowed to print ends up pausing for the - * right duration, whereas all the other CPUs pause for twice as long: once in - * oops_enter(), once in oops_exit(). + * It turns out that the CPU which is allowed to print ends up pausing for + * the right duration, whereas all the other CPUs pause for twice as long: + * once in oops_enter(), once in oops_exit(). */ void oops_enter(void) { - debug_locks_off(); /* can't trust the integrity of the kernel anymore */ + tracing_off(); + /* can't trust the integrity of the kernel anymore: */ + debug_locks_off(); do_oops_enter_exit(); } @@ -327,34 +342,46 @@ void oops_exit(void) } #ifdef WANT_WARN_ON_SLOWPATH -void warn_slowpath(const char *file, int line, const char *fmt, ...) -{ +struct slowpath_args { + const char *fmt; va_list args; - char function[KSYM_SYMBOL_LEN]; - unsigned long caller = (unsigned long)__builtin_return_address(0); - const char *board; +}; - sprint_symbol(function, caller); +static void warn_slowpath_common(const char *file, int line, void *caller, struct slowpath_args *args) +{ + const char *board; printk(KERN_WARNING "------------[ cut here ]------------\n"); - printk(KERN_WARNING "WARNING: at %s:%d %s()\n", file, - line, function); + printk(KERN_WARNING "WARNING: at %s:%d %pS()\n", file, line, caller); board = dmi_get_system_info(DMI_PRODUCT_NAME); if (board) printk(KERN_WARNING "Hardware name: %s\n", board); - if (fmt) { - va_start(args, fmt); - vprintk(fmt, args); - va_end(args); - } + if (args) + vprintk(args->fmt, args->args); print_modules(); dump_stack(); print_oops_end_marker(); add_taint(TAINT_WARN); } -EXPORT_SYMBOL(warn_slowpath); + +void warn_slowpath_fmt(const char *file, int line, const char *fmt, ...) +{ + struct slowpath_args args; + + args.fmt = fmt; + va_start(args.args, fmt); + warn_slowpath_common(file, line, __builtin_return_address(0), &args); + va_end(args.args); +} +EXPORT_SYMBOL(warn_slowpath_fmt); + +void warn_slowpath_null(const char *file, int line) +{ + warn_slowpath_common(file, line, __builtin_return_address(0), NULL); +} +EXPORT_SYMBOL(warn_slowpath_null); #endif #ifdef CONFIG_CC_STACKPROTECTOR diff --git a/kernel/params.c b/kernel/params.c index a1e3025b19a..d656c276508 100644 --- a/kernel/params.c +++ b/kernel/params.c @@ -23,6 +23,7 @@ #include <linux/device.h> #include <linux/err.h> #include <linux/slab.h> +#include <linux/ctype.h> #if 0 #define DEBUGP printk @@ -87,7 +88,7 @@ static char *next_arg(char *args, char **param, char **val) } for (i = 0; args[i]; i++) { - if (args[i] == ' ' && !in_quote) + if (isspace(args[i]) && !in_quote) break; if (equals == 0) { if (args[i] == '=') @@ -121,7 +122,7 @@ static char *next_arg(char *args, char **param, char **val) next = args + i; /* Chew up trailing spaces. */ - while (*next == ' ') + while (isspace(*next)) next++; return next; } @@ -138,7 +139,7 @@ int parse_args(const char *name, DEBUGP("Parsing ARGS: %s\n", args); /* Chew leading spaces */ - while (*args == ' ') + while (isspace(*args)) args++; while (*args) { @@ -217,7 +218,15 @@ int param_set_charp(const char *val, struct kernel_param *kp) return -ENOSPC; } - *(char **)kp->arg = (char *)val; + /* This is a hack. We can't need to strdup in early boot, and we + * don't need to; this mangled commandline is preserved. */ + if (slab_is_available()) { + *(char **)kp->arg = kstrdup(val, GFP_KERNEL); + if (!*(char **)kp->arg) + return -ENOMEM; + } else + *(const char **)kp->arg = val; + return 0; } @@ -226,44 +235,63 @@ int param_get_charp(char *buffer, struct kernel_param *kp) return sprintf(buffer, "%s", *((char **)kp->arg)); } +/* Actually could be a bool or an int, for historical reasons. */ int param_set_bool(const char *val, struct kernel_param *kp) { + bool v; + /* No equals means "set"... */ if (!val) val = "1"; /* One of =[yYnN01] */ switch (val[0]) { case 'y': case 'Y': case '1': - *(int *)kp->arg = 1; - return 0; + v = true; + break; case 'n': case 'N': case '0': - *(int *)kp->arg = 0; - return 0; + v = false; + break; + default: + return -EINVAL; } - return -EINVAL; + + if (kp->flags & KPARAM_ISBOOL) + *(bool *)kp->arg = v; + else + *(int *)kp->arg = v; + return 0; } int param_get_bool(char *buffer, struct kernel_param *kp) { + bool val; + if (kp->flags & KPARAM_ISBOOL) + val = *(bool *)kp->arg; + else + val = *(int *)kp->arg; + /* Y and N chosen as being relatively non-coder friendly */ - return sprintf(buffer, "%c", (*(int *)kp->arg) ? 'Y' : 'N'); + return sprintf(buffer, "%c", val ? 'Y' : 'N'); } +/* This one must be bool. */ int param_set_invbool(const char *val, struct kernel_param *kp) { - int boolval, ret; + int ret; + bool boolval; struct kernel_param dummy; dummy.arg = &boolval; + dummy.flags = KPARAM_ISBOOL; ret = param_set_bool(val, &dummy); if (ret == 0) - *(int *)kp->arg = !boolval; + *(bool *)kp->arg = !boolval; return ret; } int param_get_invbool(char *buffer, struct kernel_param *kp) { - return sprintf(buffer, "%c", (*(int *)kp->arg) ? 'N' : 'Y'); + return sprintf(buffer, "%c", (*(bool *)kp->arg) ? 'N' : 'Y'); } /* We break the rule and mangle the string. */ @@ -272,6 +300,7 @@ static int param_array(const char *name, unsigned int min, unsigned int max, void *elem, int elemsize, int (*set)(const char *, struct kernel_param *kp), + u16 flags, unsigned int *num) { int ret; @@ -281,6 +310,7 @@ static int param_array(const char *name, /* Get the name right for errors. */ kp.name = name; kp.arg = elem; + kp.flags = flags; /* No equals sign? */ if (!val) { @@ -326,7 +356,8 @@ int param_array_set(const char *val, struct kernel_param *kp) unsigned int temp_num; return param_array(kp->name, val, 1, arr->max, arr->elem, - arr->elemsize, arr->set, arr->num ?: &temp_num); + arr->elemsize, arr->set, kp->flags, + arr->num ?: &temp_num); } int param_array_get(char *buffer, struct kernel_param *kp) @@ -571,6 +602,11 @@ void module_param_sysfs_remove(struct module *mod) } #endif +void destroy_params(const struct kernel_param *params, unsigned num) +{ + /* FIXME: This should free kmalloced charp parameters. It doesn't. */ +} + static void __init kernel_add_sysfs_param(const char *name, struct kernel_param *kparam, unsigned int name_skip) diff --git a/kernel/perf_event.c b/kernel/perf_event.c new file mode 100644 index 00000000000..6b7ddba1dd6 --- /dev/null +++ b/kernel/perf_event.c @@ -0,0 +1,5359 @@ +/* + * Performance events core code: + * + * Copyright (C) 2008 Thomas Gleixner <tglx@linutronix.de> + * Copyright (C) 2008-2009 Red Hat, Inc., Ingo Molnar + * Copyright (C) 2008-2009 Red Hat, Inc., Peter Zijlstra <pzijlstr@redhat.com> + * Copyright © 2009 Paul Mackerras, IBM Corp. <paulus@au1.ibm.com> + * + * For licensing details see kernel-base/COPYING + */ + +#include <linux/fs.h> +#include <linux/mm.h> +#include <linux/cpu.h> +#include <linux/smp.h> +#include <linux/file.h> +#include <linux/poll.h> +#include <linux/sysfs.h> +#include <linux/dcache.h> +#include <linux/percpu.h> +#include <linux/ptrace.h> +#include <linux/vmstat.h> +#include <linux/vmalloc.h> +#include <linux/hardirq.h> +#include <linux/rculist.h> +#include <linux/uaccess.h> +#include <linux/syscalls.h> +#include <linux/anon_inodes.h> +#include <linux/kernel_stat.h> +#include <linux/perf_event.h> +#include <linux/ftrace_event.h> +#include <linux/hw_breakpoint.h> + +#include <asm/irq_regs.h> + +/* + * Each CPU has a list of per CPU events: + */ +DEFINE_PER_CPU(struct perf_cpu_context, perf_cpu_context); + +int perf_max_events __read_mostly = 1; +static int perf_reserved_percpu __read_mostly; +static int perf_overcommit __read_mostly = 1; + +static atomic_t nr_events __read_mostly; +static atomic_t nr_mmap_events __read_mostly; +static atomic_t nr_comm_events __read_mostly; +static atomic_t nr_task_events __read_mostly; + +/* + * perf event paranoia level: + * -1 - not paranoid at all + * 0 - disallow raw tracepoint access for unpriv + * 1 - disallow cpu events for unpriv + * 2 - disallow kernel profiling for unpriv + */ +int sysctl_perf_event_paranoid __read_mostly = 1; + +static inline bool perf_paranoid_tracepoint_raw(void) +{ + return sysctl_perf_event_paranoid > -1; +} + +static inline bool perf_paranoid_cpu(void) +{ + return sysctl_perf_event_paranoid > 0; +} + +static inline bool perf_paranoid_kernel(void) +{ + return sysctl_perf_event_paranoid > 1; +} + +int sysctl_perf_event_mlock __read_mostly = 512; /* 'free' kb per user */ + +/* + * max perf event sample rate + */ +int sysctl_perf_event_sample_rate __read_mostly = 100000; + +static atomic64_t perf_event_id; + +/* + * Lock for (sysadmin-configurable) event reservations: + */ +static DEFINE_SPINLOCK(perf_resource_lock); + +/* + * Architecture provided APIs - weak aliases: + */ +extern __weak const struct pmu *hw_perf_event_init(struct perf_event *event) +{ + return NULL; +} + +void __weak hw_perf_disable(void) { barrier(); } +void __weak hw_perf_enable(void) { barrier(); } + +void __weak hw_perf_event_setup(int cpu) { barrier(); } +void __weak hw_perf_event_setup_online(int cpu) { barrier(); } + +int __weak +hw_perf_group_sched_in(struct perf_event *group_leader, + struct perf_cpu_context *cpuctx, + struct perf_event_context *ctx, int cpu) +{ + return 0; +} + +void __weak perf_event_print_debug(void) { } + +static DEFINE_PER_CPU(int, perf_disable_count); + +void __perf_disable(void) +{ + __get_cpu_var(perf_disable_count)++; +} + +bool __perf_enable(void) +{ + return !--__get_cpu_var(perf_disable_count); +} + +void perf_disable(void) +{ + __perf_disable(); + hw_perf_disable(); +} + +void perf_enable(void) +{ + if (__perf_enable()) + hw_perf_enable(); +} + +static void get_ctx(struct perf_event_context *ctx) +{ + WARN_ON(!atomic_inc_not_zero(&ctx->refcount)); +} + +static void free_ctx(struct rcu_head *head) +{ + struct perf_event_context *ctx; + + ctx = container_of(head, struct perf_event_context, rcu_head); + kfree(ctx); +} + +static void put_ctx(struct perf_event_context *ctx) +{ + if (atomic_dec_and_test(&ctx->refcount)) { + if (ctx->parent_ctx) + put_ctx(ctx->parent_ctx); + if (ctx->task) + put_task_struct(ctx->task); + call_rcu(&ctx->rcu_head, free_ctx); + } +} + +static void unclone_ctx(struct perf_event_context *ctx) +{ + if (ctx->parent_ctx) { + put_ctx(ctx->parent_ctx); + ctx->parent_ctx = NULL; + } +} + +/* + * If we inherit events we want to return the parent event id + * to userspace. + */ +static u64 primary_event_id(struct perf_event *event) +{ + u64 id = event->id; + + if (event->parent) + id = event->parent->id; + + return id; +} + +/* + * Get the perf_event_context for a task and lock it. + * This has to cope with with the fact that until it is locked, + * the context could get moved to another task. + */ +static struct perf_event_context * +perf_lock_task_context(struct task_struct *task, unsigned long *flags) +{ + struct perf_event_context *ctx; + + rcu_read_lock(); + retry: + ctx = rcu_dereference(task->perf_event_ctxp); + if (ctx) { + /* + * If this context is a clone of another, it might + * get swapped for another underneath us by + * perf_event_task_sched_out, though the + * rcu_read_lock() protects us from any context + * getting freed. Lock the context and check if it + * got swapped before we could get the lock, and retry + * if so. If we locked the right context, then it + * can't get swapped on us any more. + */ + spin_lock_irqsave(&ctx->lock, *flags); + if (ctx != rcu_dereference(task->perf_event_ctxp)) { + spin_unlock_irqrestore(&ctx->lock, *flags); + goto retry; + } + + if (!atomic_inc_not_zero(&ctx->refcount)) { + spin_unlock_irqrestore(&ctx->lock, *flags); + ctx = NULL; + } + } + rcu_read_unlock(); + return ctx; +} + +/* + * Get the context for a task and increment its pin_count so it + * can't get swapped to another task. This also increments its + * reference count so that the context can't get freed. + */ +static struct perf_event_context *perf_pin_task_context(struct task_struct *task) +{ + struct perf_event_context *ctx; + unsigned long flags; + + ctx = perf_lock_task_context(task, &flags); + if (ctx) { + ++ctx->pin_count; + spin_unlock_irqrestore(&ctx->lock, flags); + } + return ctx; +} + +static void perf_unpin_context(struct perf_event_context *ctx) +{ + unsigned long flags; + + spin_lock_irqsave(&ctx->lock, flags); + --ctx->pin_count; + spin_unlock_irqrestore(&ctx->lock, flags); + put_ctx(ctx); +} + +static inline u64 perf_clock(void) +{ + return cpu_clock(smp_processor_id()); +} + +/* + * Update the record of the current time in a context. + */ +static void update_context_time(struct perf_event_context *ctx) +{ + u64 now = perf_clock(); + + ctx->time += now - ctx->timestamp; + ctx->timestamp = now; +} + +/* + * Update the total_time_enabled and total_time_running fields for a event. + */ +static void update_event_times(struct perf_event *event) +{ + struct perf_event_context *ctx = event->ctx; + u64 run_end; + + if (event->state < PERF_EVENT_STATE_INACTIVE || + event->group_leader->state < PERF_EVENT_STATE_INACTIVE) + return; + + if (ctx->is_active) + run_end = ctx->time; + else + run_end = event->tstamp_stopped; + + event->total_time_enabled = run_end - event->tstamp_enabled; + + if (event->state == PERF_EVENT_STATE_INACTIVE) + run_end = event->tstamp_stopped; + else + run_end = ctx->time; + + event->total_time_running = run_end - event->tstamp_running; +} + +/* + * Add a event from the lists for its context. + * Must be called with ctx->mutex and ctx->lock held. + */ +static void +list_add_event(struct perf_event *event, struct perf_event_context *ctx) +{ + struct perf_event *group_leader = event->group_leader; + + /* + * Depending on whether it is a standalone or sibling event, + * add it straight to the context's event list, or to the group + * leader's sibling list: + */ + if (group_leader == event) + list_add_tail(&event->group_entry, &ctx->group_list); + else { + list_add_tail(&event->group_entry, &group_leader->sibling_list); + group_leader->nr_siblings++; + } + + list_add_rcu(&event->event_entry, &ctx->event_list); + ctx->nr_events++; + if (event->attr.inherit_stat) + ctx->nr_stat++; +} + +/* + * Remove a event from the lists for its context. + * Must be called with ctx->mutex and ctx->lock held. + */ +static void +list_del_event(struct perf_event *event, struct perf_event_context *ctx) +{ + struct perf_event *sibling, *tmp; + + if (list_empty(&event->group_entry)) + return; + ctx->nr_events--; + if (event->attr.inherit_stat) + ctx->nr_stat--; + + list_del_init(&event->group_entry); + list_del_rcu(&event->event_entry); + + if (event->group_leader != event) + event->group_leader->nr_siblings--; + + update_event_times(event); + + /* + * If event was in error state, then keep it + * that way, otherwise bogus counts will be + * returned on read(). The only way to get out + * of error state is by explicit re-enabling + * of the event + */ + if (event->state > PERF_EVENT_STATE_OFF) + event->state = PERF_EVENT_STATE_OFF; + + /* + * If this was a group event with sibling events then + * upgrade the siblings to singleton events by adding them + * to the context list directly: + */ + list_for_each_entry_safe(sibling, tmp, &event->sibling_list, group_entry) { + + list_move_tail(&sibling->group_entry, &ctx->group_list); + sibling->group_leader = sibling; + } +} + +static void +event_sched_out(struct perf_event *event, + struct perf_cpu_context *cpuctx, + struct perf_event_context *ctx) +{ + if (event->state != PERF_EVENT_STATE_ACTIVE) + return; + + event->state = PERF_EVENT_STATE_INACTIVE; + if (event->pending_disable) { + event->pending_disable = 0; + event->state = PERF_EVENT_STATE_OFF; + } + event->tstamp_stopped = ctx->time; + event->pmu->disable(event); + event->oncpu = -1; + + if (!is_software_event(event)) + cpuctx->active_oncpu--; + ctx->nr_active--; + if (event->attr.exclusive || !cpuctx->active_oncpu) + cpuctx->exclusive = 0; +} + +static void +group_sched_out(struct perf_event *group_event, + struct perf_cpu_context *cpuctx, + struct perf_event_context *ctx) +{ + struct perf_event *event; + + if (group_event->state != PERF_EVENT_STATE_ACTIVE) + return; + + event_sched_out(group_event, cpuctx, ctx); + + /* + * Schedule out siblings (if any): + */ + list_for_each_entry(event, &group_event->sibling_list, group_entry) + event_sched_out(event, cpuctx, ctx); + + if (group_event->attr.exclusive) + cpuctx->exclusive = 0; +} + +/* + * Cross CPU call to remove a performance event + * + * We disable the event on the hardware level first. After that we + * remove it from the context list. + */ +static void __perf_event_remove_from_context(void *info) +{ + struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context); + struct perf_event *event = info; + struct perf_event_context *ctx = event->ctx; + + /* + * If this is a task context, we need to check whether it is + * the current task context of this cpu. If not it has been + * scheduled out before the smp call arrived. + */ + if (ctx->task && cpuctx->task_ctx != ctx) + return; + + spin_lock(&ctx->lock); + /* + * Protect the list operation against NMI by disabling the + * events on a global level. + */ + perf_disable(); + + event_sched_out(event, cpuctx, ctx); + + list_del_event(event, ctx); + + if (!ctx->task) { + /* + * Allow more per task events with respect to the + * reservation: + */ + cpuctx->max_pertask = + min(perf_max_events - ctx->nr_events, + perf_max_events - perf_reserved_percpu); + } + + perf_enable(); + spin_unlock(&ctx->lock); +} + + +/* + * Remove the event from a task's (or a CPU's) list of events. + * + * Must be called with ctx->mutex held. + * + * CPU events are removed with a smp call. For task events we only + * call when the task is on a CPU. + * + * If event->ctx is a cloned context, callers must make sure that + * every task struct that event->ctx->task could possibly point to + * remains valid. This is OK when called from perf_release since + * that only calls us on the top-level context, which can't be a clone. + * When called from perf_event_exit_task, it's OK because the + * context has been detached from its task. + */ +static void perf_event_remove_from_context(struct perf_event *event) +{ + struct perf_event_context *ctx = event->ctx; + struct task_struct *task = ctx->task; + + if (!task) { + /* + * Per cpu events are removed via an smp call and + * the removal is always sucessful. + */ + smp_call_function_single(event->cpu, + __perf_event_remove_from_context, + event, 1); + return; + } + +retry: + task_oncpu_function_call(task, __perf_event_remove_from_context, + event); + + spin_lock_irq(&ctx->lock); + /* + * If the context is active we need to retry the smp call. + */ + if (ctx->nr_active && !list_empty(&event->group_entry)) { + spin_unlock_irq(&ctx->lock); + goto retry; + } + + /* + * The lock prevents that this context is scheduled in so we + * can remove the event safely, if the call above did not + * succeed. + */ + if (!list_empty(&event->group_entry)) + list_del_event(event, ctx); + spin_unlock_irq(&ctx->lock); +} + +/* + * Update total_time_enabled and total_time_running for all events in a group. + */ +static void update_group_times(struct perf_event *leader) +{ + struct perf_event *event; + + update_event_times(leader); + list_for_each_entry(event, &leader->sibling_list, group_entry) + update_event_times(event); +} + +/* + * Cross CPU call to disable a performance event + */ +static void __perf_event_disable(void *info) +{ + struct perf_event *event = info; + struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context); + struct perf_event_context *ctx = event->ctx; + + /* + * If this is a per-task event, need to check whether this + * event's task is the current task on this cpu. + */ + if (ctx->task && cpuctx->task_ctx != ctx) + return; + + spin_lock(&ctx->lock); + + /* + * If the event is on, turn it off. + * If it is in error state, leave it in error state. + */ + if (event->state >= PERF_EVENT_STATE_INACTIVE) { + update_context_time(ctx); + update_group_times(event); + if (event == event->group_leader) + group_sched_out(event, cpuctx, ctx); + else + event_sched_out(event, cpuctx, ctx); + event->state = PERF_EVENT_STATE_OFF; + } + + spin_unlock(&ctx->lock); +} + +/* + * Disable a event. + * + * If event->ctx is a cloned context, callers must make sure that + * every task struct that event->ctx->task could possibly point to + * remains valid. This condition is satisifed when called through + * perf_event_for_each_child or perf_event_for_each because they + * hold the top-level event's child_mutex, so any descendant that + * goes to exit will block in sync_child_event. + * When called from perf_pending_event it's OK because event->ctx + * is the current context on this CPU and preemption is disabled, + * hence we can't get into perf_event_task_sched_out for this context. + */ +static void perf_event_disable(struct perf_event *event) +{ + struct perf_event_context *ctx = event->ctx; + struct task_struct *task = ctx->task; + + if (!task) { + /* + * Disable the event on the cpu that it's on + */ + smp_call_function_single(event->cpu, __perf_event_disable, + event, 1); + return; + } + + retry: + task_oncpu_function_call(task, __perf_event_disable, event); + + spin_lock_irq(&ctx->lock); + /* + * If the event is still active, we need to retry the cross-call. + */ + if (event->state == PERF_EVENT_STATE_ACTIVE) { + spin_unlock_irq(&ctx->lock); + goto retry; + } + + /* + * Since we have the lock this context can't be scheduled + * in, so we can change the state safely. + */ + if (event->state == PERF_EVENT_STATE_INACTIVE) { + update_group_times(event); + event->state = PERF_EVENT_STATE_OFF; + } + + spin_unlock_irq(&ctx->lock); +} + +static int +event_sched_in(struct perf_event *event, + struct perf_cpu_context *cpuctx, + struct perf_event_context *ctx, + int cpu) +{ + if (event->state <= PERF_EVENT_STATE_OFF) + return 0; + + event->state = PERF_EVENT_STATE_ACTIVE; + event->oncpu = cpu; /* TODO: put 'cpu' into cpuctx->cpu */ + /* + * The new state must be visible before we turn it on in the hardware: + */ + smp_wmb(); + + if (event->pmu->enable(event)) { + event->state = PERF_EVENT_STATE_INACTIVE; + event->oncpu = -1; + return -EAGAIN; + } + + event->tstamp_running += ctx->time - event->tstamp_stopped; + + if (!is_software_event(event)) + cpuctx->active_oncpu++; + ctx->nr_active++; + + if (event->attr.exclusive) + cpuctx->exclusive = 1; + + return 0; +} + +static int +group_sched_in(struct perf_event *group_event, + struct perf_cpu_context *cpuctx, + struct perf_event_context *ctx, + int cpu) +{ + struct perf_event *event, *partial_group; + int ret; + + if (group_event->state == PERF_EVENT_STATE_OFF) + return 0; + + ret = hw_perf_group_sched_in(group_event, cpuctx, ctx, cpu); + if (ret) + return ret < 0 ? ret : 0; + + if (event_sched_in(group_event, cpuctx, ctx, cpu)) + return -EAGAIN; + + /* + * Schedule in siblings as one group (if any): + */ + list_for_each_entry(event, &group_event->sibling_list, group_entry) { + if (event_sched_in(event, cpuctx, ctx, cpu)) { + partial_group = event; + goto group_error; + } + } + + return 0; + +group_error: + /* + * Groups can be scheduled in as one unit only, so undo any + * partial group before returning: + */ + list_for_each_entry(event, &group_event->sibling_list, group_entry) { + if (event == partial_group) + break; + event_sched_out(event, cpuctx, ctx); + } + event_sched_out(group_event, cpuctx, ctx); + + return -EAGAIN; +} + +/* + * Return 1 for a group consisting entirely of software events, + * 0 if the group contains any hardware events. + */ +static int is_software_only_group(struct perf_event *leader) +{ + struct perf_event *event; + + if (!is_software_event(leader)) + return 0; + + list_for_each_entry(event, &leader->sibling_list, group_entry) + if (!is_software_event(event)) + return 0; + + return 1; +} + +/* + * Work out whether we can put this event group on the CPU now. + */ +static int group_can_go_on(struct perf_event *event, + struct perf_cpu_context *cpuctx, + int can_add_hw) +{ + /* + * Groups consisting entirely of software events can always go on. + */ + if (is_software_only_group(event)) + return 1; + /* + * If an exclusive group is already on, no other hardware + * events can go on. + */ + if (cpuctx->exclusive) + return 0; + /* + * If this group is exclusive and there are already + * events on the CPU, it can't go on. + */ + if (event->attr.exclusive && cpuctx->active_oncpu) + return 0; + /* + * Otherwise, try to add it if all previous groups were able + * to go on. + */ + return can_add_hw; +} + +static void add_event_to_ctx(struct perf_event *event, + struct perf_event_context *ctx) +{ + list_add_event(event, ctx); + event->tstamp_enabled = ctx->time; + event->tstamp_running = ctx->time; + event->tstamp_stopped = ctx->time; +} + +/* + * Cross CPU call to install and enable a performance event + * + * Must be called with ctx->mutex held + */ +static void __perf_install_in_context(void *info) +{ + struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context); + struct perf_event *event = info; + struct perf_event_context *ctx = event->ctx; + struct perf_event *leader = event->group_leader; + int cpu = smp_processor_id(); + int err; + + /* + * If this is a task context, we need to check whether it is + * the current task context of this cpu. If not it has been + * scheduled out before the smp call arrived. + * Or possibly this is the right context but it isn't + * on this cpu because it had no events. + */ + if (ctx->task && cpuctx->task_ctx != ctx) { + if (cpuctx->task_ctx || ctx->task != current) + return; + cpuctx->task_ctx = ctx; + } + + spin_lock(&ctx->lock); + ctx->is_active = 1; + update_context_time(ctx); + + /* + * Protect the list operation against NMI by disabling the + * events on a global level. NOP for non NMI based events. + */ + perf_disable(); + + add_event_to_ctx(event, ctx); + + /* + * Don't put the event on if it is disabled or if + * it is in a group and the group isn't on. + */ + if (event->state != PERF_EVENT_STATE_INACTIVE || + (leader != event && leader->state != PERF_EVENT_STATE_ACTIVE)) + goto unlock; + + /* + * An exclusive event can't go on if there are already active + * hardware events, and no hardware event can go on if there + * is already an exclusive event on. + */ + if (!group_can_go_on(event, cpuctx, 1)) + err = -EEXIST; + else + err = event_sched_in(event, cpuctx, ctx, cpu); + + if (err) { + /* + * This event couldn't go on. If it is in a group + * then we have to pull the whole group off. + * If the event group is pinned then put it in error state. + */ + if (leader != event) + group_sched_out(leader, cpuctx, ctx); + if (leader->attr.pinned) { + update_group_times(leader); + leader->state = PERF_EVENT_STATE_ERROR; + } + } + + if (!err && !ctx->task && cpuctx->max_pertask) + cpuctx->max_pertask--; + + unlock: + perf_enable(); + + spin_unlock(&ctx->lock); +} + +/* + * Attach a performance event to a context + * + * First we add the event to the list with the hardware enable bit + * in event->hw_config cleared. + * + * If the event is attached to a task which is on a CPU we use a smp + * call to enable it in the task context. The task might have been + * scheduled away, but we check this in the smp call again. + * + * Must be called with ctx->mutex held. + */ +static void +perf_install_in_context(struct perf_event_context *ctx, + struct perf_event *event, + int cpu) +{ + struct task_struct *task = ctx->task; + + if (!task) { + /* + * Per cpu events are installed via an smp call and + * the install is always sucessful. + */ + smp_call_function_single(cpu, __perf_install_in_context, + event, 1); + return; + } + +retry: + task_oncpu_function_call(task, __perf_install_in_context, + event); + + spin_lock_irq(&ctx->lock); + /* + * we need to retry the smp call. + */ + if (ctx->is_active && list_empty(&event->group_entry)) { + spin_unlock_irq(&ctx->lock); + goto retry; + } + + /* + * The lock prevents that this context is scheduled in so we + * can add the event safely, if it the call above did not + * succeed. + */ + if (list_empty(&event->group_entry)) + add_event_to_ctx(event, ctx); + spin_unlock_irq(&ctx->lock); +} + +/* + * Put a event into inactive state and update time fields. + * Enabling the leader of a group effectively enables all + * the group members that aren't explicitly disabled, so we + * have to update their ->tstamp_enabled also. + * Note: this works for group members as well as group leaders + * since the non-leader members' sibling_lists will be empty. + */ +static void __perf_event_mark_enabled(struct perf_event *event, + struct perf_event_context *ctx) +{ + struct perf_event *sub; + + event->state = PERF_EVENT_STATE_INACTIVE; + event->tstamp_enabled = ctx->time - event->total_time_enabled; + list_for_each_entry(sub, &event->sibling_list, group_entry) + if (sub->state >= PERF_EVENT_STATE_INACTIVE) + sub->tstamp_enabled = + ctx->time - sub->total_time_enabled; +} + +/* + * Cross CPU call to enable a performance event + */ +static void __perf_event_enable(void *info) +{ + struct perf_event *event = info; + struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context); + struct perf_event_context *ctx = event->ctx; + struct perf_event *leader = event->group_leader; + int err; + + /* + * If this is a per-task event, need to check whether this + * event's task is the current task on this cpu. + */ + if (ctx->task && cpuctx->task_ctx != ctx) { + if (cpuctx->task_ctx || ctx->task != current) + return; + cpuctx->task_ctx = ctx; + } + + spin_lock(&ctx->lock); + ctx->is_active = 1; + update_context_time(ctx); + + if (event->state >= PERF_EVENT_STATE_INACTIVE) + goto unlock; + __perf_event_mark_enabled(event, ctx); + + /* + * If the event is in a group and isn't the group leader, + * then don't put it on unless the group is on. + */ + if (leader != event && leader->state != PERF_EVENT_STATE_ACTIVE) + goto unlock; + + if (!group_can_go_on(event, cpuctx, 1)) { + err = -EEXIST; + } else { + perf_disable(); + if (event == leader) + err = group_sched_in(event, cpuctx, ctx, + smp_processor_id()); + else + err = event_sched_in(event, cpuctx, ctx, + smp_processor_id()); + perf_enable(); + } + + if (err) { + /* + * If this event can't go on and it's part of a + * group, then the whole group has to come off. + */ + if (leader != event) + group_sched_out(leader, cpuctx, ctx); + if (leader->attr.pinned) { + update_group_times(leader); + leader->state = PERF_EVENT_STATE_ERROR; + } + } + + unlock: + spin_unlock(&ctx->lock); +} + +/* + * Enable a event. + * + * If event->ctx is a cloned context, callers must make sure that + * every task struct that event->ctx->task could possibly point to + * remains valid. This condition is satisfied when called through + * perf_event_for_each_child or perf_event_for_each as described + * for perf_event_disable. + */ +static void perf_event_enable(struct perf_event *event) +{ + struct perf_event_context *ctx = event->ctx; + struct task_struct *task = ctx->task; + + if (!task) { + /* + * Enable the event on the cpu that it's on + */ + smp_call_function_single(event->cpu, __perf_event_enable, + event, 1); + return; + } + + spin_lock_irq(&ctx->lock); + if (event->state >= PERF_EVENT_STATE_INACTIVE) + goto out; + + /* + * If the event is in error state, clear that first. + * That way, if we see the event in error state below, we + * know that it has gone back into error state, as distinct + * from the task having been scheduled away before the + * cross-call arrived. + */ + if (event->state == PERF_EVENT_STATE_ERROR) + event->state = PERF_EVENT_STATE_OFF; + + retry: + spin_unlock_irq(&ctx->lock); + task_oncpu_function_call(task, __perf_event_enable, event); + + spin_lock_irq(&ctx->lock); + + /* + * If the context is active and the event is still off, + * we need to retry the cross-call. + */ + if (ctx->is_active && event->state == PERF_EVENT_STATE_OFF) + goto retry; + + /* + * Since we have the lock this context can't be scheduled + * in, so we can change the state safely. + */ + if (event->state == PERF_EVENT_STATE_OFF) + __perf_event_mark_enabled(event, ctx); + + out: + spin_unlock_irq(&ctx->lock); +} + +static int perf_event_refresh(struct perf_event *event, int refresh) +{ + /* + * not supported on inherited events + */ + if (event->attr.inherit) + return -EINVAL; + + atomic_add(refresh, &event->event_limit); + perf_event_enable(event); + + return 0; +} + +void __perf_event_sched_out(struct perf_event_context *ctx, + struct perf_cpu_context *cpuctx) +{ + struct perf_event *event; + + spin_lock(&ctx->lock); + ctx->is_active = 0; + if (likely(!ctx->nr_events)) + goto out; + update_context_time(ctx); + + perf_disable(); + if (ctx->nr_active) { + list_for_each_entry(event, &ctx->group_list, group_entry) + group_sched_out(event, cpuctx, ctx); + } + perf_enable(); + out: + spin_unlock(&ctx->lock); +} + +/* + * Test whether two contexts are equivalent, i.e. whether they + * have both been cloned from the same version of the same context + * and they both have the same number of enabled events. + * If the number of enabled events is the same, then the set + * of enabled events should be the same, because these are both + * inherited contexts, therefore we can't access individual events + * in them directly with an fd; we can only enable/disable all + * events via prctl, or enable/disable all events in a family + * via ioctl, which will have the same effect on both contexts. + */ +static int context_equiv(struct perf_event_context *ctx1, + struct perf_event_context *ctx2) +{ + return ctx1->parent_ctx && ctx1->parent_ctx == ctx2->parent_ctx + && ctx1->parent_gen == ctx2->parent_gen + && !ctx1->pin_count && !ctx2->pin_count; +} + +static void __perf_event_sync_stat(struct perf_event *event, + struct perf_event *next_event) +{ + u64 value; + + if (!event->attr.inherit_stat) + return; + + /* + * Update the event value, we cannot use perf_event_read() + * because we're in the middle of a context switch and have IRQs + * disabled, which upsets smp_call_function_single(), however + * we know the event must be on the current CPU, therefore we + * don't need to use it. + */ + switch (event->state) { + case PERF_EVENT_STATE_ACTIVE: + event->pmu->read(event); + /* fall-through */ + + case PERF_EVENT_STATE_INACTIVE: + update_event_times(event); + break; + + default: + break; + } + + /* + * In order to keep per-task stats reliable we need to flip the event + * values when we flip the contexts. + */ + value = atomic64_read(&next_event->count); + value = atomic64_xchg(&event->count, value); + atomic64_set(&next_event->count, value); + + swap(event->total_time_enabled, next_event->total_time_enabled); + swap(event->total_time_running, next_event->total_time_running); + + /* + * Since we swizzled the values, update the user visible data too. + */ + perf_event_update_userpage(event); + perf_event_update_userpage(next_event); +} + +#define list_next_entry(pos, member) \ + list_entry(pos->member.next, typeof(*pos), member) + +static void perf_event_sync_stat(struct perf_event_context *ctx, + struct perf_event_context *next_ctx) +{ + struct perf_event *event, *next_event; + + if (!ctx->nr_stat) + return; + + update_context_time(ctx); + + event = list_first_entry(&ctx->event_list, + struct perf_event, event_entry); + + next_event = list_first_entry(&next_ctx->event_list, + struct perf_event, event_entry); + + while (&event->event_entry != &ctx->event_list && + &next_event->event_entry != &next_ctx->event_list) { + + __perf_event_sync_stat(event, next_event); + + event = list_next_entry(event, event_entry); + next_event = list_next_entry(next_event, event_entry); + } +} + +/* + * Called from scheduler to remove the events of the current task, + * with interrupts disabled. + * + * We stop each event and update the event value in event->count. + * + * This does not protect us against NMI, but disable() + * sets the disabled bit in the control field of event _before_ + * accessing the event control register. If a NMI hits, then it will + * not restart the event. + */ +void perf_event_task_sched_out(struct task_struct *task, + struct task_struct *next, int cpu) +{ + struct perf_cpu_context *cpuctx = &per_cpu(perf_cpu_context, cpu); + struct perf_event_context *ctx = task->perf_event_ctxp; + struct perf_event_context *next_ctx; + struct perf_event_context *parent; + struct pt_regs *regs; + int do_switch = 1; + + regs = task_pt_regs(task); + perf_sw_event(PERF_COUNT_SW_CONTEXT_SWITCHES, 1, 1, regs, 0); + + if (likely(!ctx || !cpuctx->task_ctx)) + return; + + rcu_read_lock(); + parent = rcu_dereference(ctx->parent_ctx); + next_ctx = next->perf_event_ctxp; + if (parent && next_ctx && + rcu_dereference(next_ctx->parent_ctx) == parent) { + /* + * Looks like the two contexts are clones, so we might be + * able to optimize the context switch. We lock both + * contexts and check that they are clones under the + * lock (including re-checking that neither has been + * uncloned in the meantime). It doesn't matter which + * order we take the locks because no other cpu could + * be trying to lock both of these tasks. + */ + spin_lock(&ctx->lock); + spin_lock_nested(&next_ctx->lock, SINGLE_DEPTH_NESTING); + if (context_equiv(ctx, next_ctx)) { + /* + * XXX do we need a memory barrier of sorts + * wrt to rcu_dereference() of perf_event_ctxp + */ + task->perf_event_ctxp = next_ctx; + next->perf_event_ctxp = ctx; + ctx->task = next; + next_ctx->task = task; + do_switch = 0; + + perf_event_sync_stat(ctx, next_ctx); + } + spin_unlock(&next_ctx->lock); + spin_unlock(&ctx->lock); + } + rcu_read_unlock(); + + if (do_switch) { + __perf_event_sched_out(ctx, cpuctx); + cpuctx->task_ctx = NULL; + } +} + +/* + * Called with IRQs disabled + */ +static void __perf_event_task_sched_out(struct perf_event_context *ctx) +{ + struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context); + + if (!cpuctx->task_ctx) + return; + + if (WARN_ON_ONCE(ctx != cpuctx->task_ctx)) + return; + + __perf_event_sched_out(ctx, cpuctx); + cpuctx->task_ctx = NULL; +} + +/* + * Called with IRQs disabled + */ +static void perf_event_cpu_sched_out(struct perf_cpu_context *cpuctx) +{ + __perf_event_sched_out(&cpuctx->ctx, cpuctx); +} + +static void +__perf_event_sched_in(struct perf_event_context *ctx, + struct perf_cpu_context *cpuctx, int cpu) +{ + struct perf_event *event; + int can_add_hw = 1; + + spin_lock(&ctx->lock); + ctx->is_active = 1; + if (likely(!ctx->nr_events)) + goto out; + + ctx->timestamp = perf_clock(); + + perf_disable(); + + /* + * First go through the list and put on any pinned groups + * in order to give them the best chance of going on. + */ + list_for_each_entry(event, &ctx->group_list, group_entry) { + if (event->state <= PERF_EVENT_STATE_OFF || + !event->attr.pinned) + continue; + if (event->cpu != -1 && event->cpu != cpu) + continue; + + if (group_can_go_on(event, cpuctx, 1)) + group_sched_in(event, cpuctx, ctx, cpu); + + /* + * If this pinned group hasn't been scheduled, + * put it in error state. + */ + if (event->state == PERF_EVENT_STATE_INACTIVE) { + update_group_times(event); + event->state = PERF_EVENT_STATE_ERROR; + } + } + + list_for_each_entry(event, &ctx->group_list, group_entry) { + /* + * Ignore events in OFF or ERROR state, and + * ignore pinned events since we did them already. + */ + if (event->state <= PERF_EVENT_STATE_OFF || + event->attr.pinned) + continue; + + /* + * Listen to the 'cpu' scheduling filter constraint + * of events: + */ + if (event->cpu != -1 && event->cpu != cpu) + continue; + + if (group_can_go_on(event, cpuctx, can_add_hw)) + if (group_sched_in(event, cpuctx, ctx, cpu)) + can_add_hw = 0; + } + perf_enable(); + out: + spin_unlock(&ctx->lock); +} + +/* + * Called from scheduler to add the events of the current task + * with interrupts disabled. + * + * We restore the event value and then enable it. + * + * This does not protect us against NMI, but enable() + * sets the enabled bit in the control field of event _before_ + * accessing the event control register. If a NMI hits, then it will + * keep the event running. + */ +void perf_event_task_sched_in(struct task_struct *task, int cpu) +{ + struct perf_cpu_context *cpuctx = &per_cpu(perf_cpu_context, cpu); + struct perf_event_context *ctx = task->perf_event_ctxp; + + if (likely(!ctx)) + return; + if (cpuctx->task_ctx == ctx) + return; + __perf_event_sched_in(ctx, cpuctx, cpu); + cpuctx->task_ctx = ctx; +} + +static void perf_event_cpu_sched_in(struct perf_cpu_context *cpuctx, int cpu) +{ + struct perf_event_context *ctx = &cpuctx->ctx; + + __perf_event_sched_in(ctx, cpuctx, cpu); +} + +#define MAX_INTERRUPTS (~0ULL) + +static void perf_log_throttle(struct perf_event *event, int enable); + +static void perf_adjust_period(struct perf_event *event, u64 events) +{ + struct hw_perf_event *hwc = &event->hw; + u64 period, sample_period; + s64 delta; + + events *= hwc->sample_period; + period = div64_u64(events, event->attr.sample_freq); + + delta = (s64)(period - hwc->sample_period); + delta = (delta + 7) / 8; /* low pass filter */ + + sample_period = hwc->sample_period + delta; + + if (!sample_period) + sample_period = 1; + + hwc->sample_period = sample_period; +} + +static void perf_ctx_adjust_freq(struct perf_event_context *ctx) +{ + struct perf_event *event; + struct hw_perf_event *hwc; + u64 interrupts, freq; + + spin_lock(&ctx->lock); + list_for_each_entry_rcu(event, &ctx->event_list, event_entry) { + if (event->state != PERF_EVENT_STATE_ACTIVE) + continue; + + hwc = &event->hw; + + interrupts = hwc->interrupts; + hwc->interrupts = 0; + + /* + * unthrottle events on the tick + */ + if (interrupts == MAX_INTERRUPTS) { + perf_log_throttle(event, 1); + event->pmu->unthrottle(event); + interrupts = 2*sysctl_perf_event_sample_rate/HZ; + } + + if (!event->attr.freq || !event->attr.sample_freq) + continue; + + /* + * if the specified freq < HZ then we need to skip ticks + */ + if (event->attr.sample_freq < HZ) { + freq = event->attr.sample_freq; + + hwc->freq_count += freq; + hwc->freq_interrupts += interrupts; + + if (hwc->freq_count < HZ) + continue; + + interrupts = hwc->freq_interrupts; + hwc->freq_interrupts = 0; + hwc->freq_count -= HZ; + } else + freq = HZ; + + perf_adjust_period(event, freq * interrupts); + + /* + * In order to avoid being stalled by an (accidental) huge + * sample period, force reset the sample period if we didn't + * get any events in this freq period. + */ + if (!interrupts) { + perf_disable(); + event->pmu->disable(event); + atomic64_set(&hwc->period_left, 0); + event->pmu->enable(event); + perf_enable(); + } + } + spin_unlock(&ctx->lock); +} + +/* + * Round-robin a context's events: + */ +static void rotate_ctx(struct perf_event_context *ctx) +{ + struct perf_event *event; + + if (!ctx->nr_events) + return; + + spin_lock(&ctx->lock); + /* + * Rotate the first entry last (works just fine for group events too): + */ + perf_disable(); + list_for_each_entry(event, &ctx->group_list, group_entry) { + list_move_tail(&event->group_entry, &ctx->group_list); + break; + } + perf_enable(); + + spin_unlock(&ctx->lock); +} + +void perf_event_task_tick(struct task_struct *curr, int cpu) +{ + struct perf_cpu_context *cpuctx; + struct perf_event_context *ctx; + + if (!atomic_read(&nr_events)) + return; + + cpuctx = &per_cpu(perf_cpu_context, cpu); + ctx = curr->perf_event_ctxp; + + perf_ctx_adjust_freq(&cpuctx->ctx); + if (ctx) + perf_ctx_adjust_freq(ctx); + + perf_event_cpu_sched_out(cpuctx); + if (ctx) + __perf_event_task_sched_out(ctx); + + rotate_ctx(&cpuctx->ctx); + if (ctx) + rotate_ctx(ctx); + + perf_event_cpu_sched_in(cpuctx, cpu); + if (ctx) + perf_event_task_sched_in(curr, cpu); +} + +/* + * Enable all of a task's events that have been marked enable-on-exec. + * This expects task == current. + */ +static void perf_event_enable_on_exec(struct task_struct *task) +{ + struct perf_event_context *ctx; + struct perf_event *event; + unsigned long flags; + int enabled = 0; + + local_irq_save(flags); + ctx = task->perf_event_ctxp; + if (!ctx || !ctx->nr_events) + goto out; + + __perf_event_task_sched_out(ctx); + + spin_lock(&ctx->lock); + + list_for_each_entry(event, &ctx->group_list, group_entry) { + if (!event->attr.enable_on_exec) + continue; + event->attr.enable_on_exec = 0; + if (event->state >= PERF_EVENT_STATE_INACTIVE) + continue; + __perf_event_mark_enabled(event, ctx); + enabled = 1; + } + + /* + * Unclone this context if we enabled any event. + */ + if (enabled) + unclone_ctx(ctx); + + spin_unlock(&ctx->lock); + + perf_event_task_sched_in(task, smp_processor_id()); + out: + local_irq_restore(flags); +} + +/* + * Cross CPU call to read the hardware event + */ +static void __perf_event_read(void *info) +{ + struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context); + struct perf_event *event = info; + struct perf_event_context *ctx = event->ctx; + + /* + * If this is a task context, we need to check whether it is + * the current task context of this cpu. If not it has been + * scheduled out before the smp call arrived. In that case + * event->count would have been updated to a recent sample + * when the event was scheduled out. + */ + if (ctx->task && cpuctx->task_ctx != ctx) + return; + + spin_lock(&ctx->lock); + update_context_time(ctx); + update_event_times(event); + spin_unlock(&ctx->lock); + + event->pmu->read(event); +} + +static u64 perf_event_read(struct perf_event *event) +{ + /* + * If event is enabled and currently active on a CPU, update the + * value in the event structure: + */ + if (event->state == PERF_EVENT_STATE_ACTIVE) { + smp_call_function_single(event->oncpu, + __perf_event_read, event, 1); + } else if (event->state == PERF_EVENT_STATE_INACTIVE) { + struct perf_event_context *ctx = event->ctx; + unsigned long flags; + + spin_lock_irqsave(&ctx->lock, flags); + update_context_time(ctx); + update_event_times(event); + spin_unlock_irqrestore(&ctx->lock, flags); + } + + return atomic64_read(&event->count); +} + +/* + * Initialize the perf_event context in a task_struct: + */ +static void +__perf_event_init_context(struct perf_event_context *ctx, + struct task_struct *task) +{ + memset(ctx, 0, sizeof(*ctx)); + spin_lock_init(&ctx->lock); + mutex_init(&ctx->mutex); + INIT_LIST_HEAD(&ctx->group_list); + INIT_LIST_HEAD(&ctx->event_list); + atomic_set(&ctx->refcount, 1); + ctx->task = task; +} + +static struct perf_event_context *find_get_context(pid_t pid, int cpu) +{ + struct perf_event_context *ctx; + struct perf_cpu_context *cpuctx; + struct task_struct *task; + unsigned long flags; + int err; + + /* + * If cpu is not a wildcard then this is a percpu event: + */ + if (cpu != -1) { + /* Must be root to operate on a CPU event: */ + if (perf_paranoid_cpu() && !capable(CAP_SYS_ADMIN)) + return ERR_PTR(-EACCES); + + if (cpu < 0 || cpu > num_possible_cpus()) + return ERR_PTR(-EINVAL); + + /* + * We could be clever and allow to attach a event to an + * offline CPU and activate it when the CPU comes up, but + * that's for later. + */ + if (!cpu_isset(cpu, cpu_online_map)) + return ERR_PTR(-ENODEV); + + cpuctx = &per_cpu(perf_cpu_context, cpu); + ctx = &cpuctx->ctx; + get_ctx(ctx); + + return ctx; + } + + rcu_read_lock(); + if (!pid) + task = current; + else + task = find_task_by_vpid(pid); + if (task) + get_task_struct(task); + rcu_read_unlock(); + + if (!task) + return ERR_PTR(-ESRCH); + + /* + * Can't attach events to a dying task. + */ + err = -ESRCH; + if (task->flags & PF_EXITING) + goto errout; + + /* Reuse ptrace permission checks for now. */ + err = -EACCES; + if (!ptrace_may_access(task, PTRACE_MODE_READ)) + goto errout; + + retry: + ctx = perf_lock_task_context(task, &flags); + if (ctx) { + unclone_ctx(ctx); + spin_unlock_irqrestore(&ctx->lock, flags); + } + + if (!ctx) { + ctx = kmalloc(sizeof(struct perf_event_context), GFP_KERNEL); + err = -ENOMEM; + if (!ctx) + goto errout; + __perf_event_init_context(ctx, task); + get_ctx(ctx); + if (cmpxchg(&task->perf_event_ctxp, NULL, ctx)) { + /* + * We raced with some other task; use + * the context they set. + */ + kfree(ctx); + goto retry; + } + get_task_struct(task); + } + + put_task_struct(task); + return ctx; + + errout: + put_task_struct(task); + return ERR_PTR(err); +} + +static void perf_event_free_filter(struct perf_event *event); + +static void free_event_rcu(struct rcu_head *head) +{ + struct perf_event *event; + + event = container_of(head, struct perf_event, rcu_head); + if (event->ns) + put_pid_ns(event->ns); + perf_event_free_filter(event); + kfree(event); +} + +static void perf_pending_sync(struct perf_event *event); + +static void free_event(struct perf_event *event) +{ + perf_pending_sync(event); + + if (!event->parent) { + atomic_dec(&nr_events); + if (event->attr.mmap) + atomic_dec(&nr_mmap_events); + if (event->attr.comm) + atomic_dec(&nr_comm_events); + if (event->attr.task) + atomic_dec(&nr_task_events); + } + + if (event->output) { + fput(event->output->filp); + event->output = NULL; + } + + if (event->destroy) + event->destroy(event); + + put_ctx(event->ctx); + call_rcu(&event->rcu_head, free_event_rcu); +} + +int perf_event_release_kernel(struct perf_event *event) +{ + struct perf_event_context *ctx = event->ctx; + + WARN_ON_ONCE(ctx->parent_ctx); + mutex_lock(&ctx->mutex); + perf_event_remove_from_context(event); + mutex_unlock(&ctx->mutex); + + mutex_lock(&event->owner->perf_event_mutex); + list_del_init(&event->owner_entry); + mutex_unlock(&event->owner->perf_event_mutex); + put_task_struct(event->owner); + + free_event(event); + + return 0; +} +EXPORT_SYMBOL_GPL(perf_event_release_kernel); + +/* + * Called when the last reference to the file is gone. + */ +static int perf_release(struct inode *inode, struct file *file) +{ + struct perf_event *event = file->private_data; + + file->private_data = NULL; + + return perf_event_release_kernel(event); +} + +static int perf_event_read_size(struct perf_event *event) +{ + int entry = sizeof(u64); /* value */ + int size = 0; + int nr = 1; + + if (event->attr.read_format & PERF_FORMAT_TOTAL_TIME_ENABLED) + size += sizeof(u64); + + if (event->attr.read_format & PERF_FORMAT_TOTAL_TIME_RUNNING) + size += sizeof(u64); + + if (event->attr.read_format & PERF_FORMAT_ID) + entry += sizeof(u64); + + if (event->attr.read_format & PERF_FORMAT_GROUP) { + nr += event->group_leader->nr_siblings; + size += sizeof(u64); + } + + size += entry * nr; + + return size; +} + +u64 perf_event_read_value(struct perf_event *event, u64 *enabled, u64 *running) +{ + struct perf_event *child; + u64 total = 0; + + *enabled = 0; + *running = 0; + + mutex_lock(&event->child_mutex); + total += perf_event_read(event); + *enabled += event->total_time_enabled + + atomic64_read(&event->child_total_time_enabled); + *running += event->total_time_running + + atomic64_read(&event->child_total_time_running); + + list_for_each_entry(child, &event->child_list, child_list) { + total += perf_event_read(child); + *enabled += child->total_time_enabled; + *running += child->total_time_running; + } + mutex_unlock(&event->child_mutex); + + return total; +} +EXPORT_SYMBOL_GPL(perf_event_read_value); + +static int perf_event_read_group(struct perf_event *event, + u64 read_format, char __user *buf) +{ + struct perf_event *leader = event->group_leader, *sub; + int n = 0, size = 0, ret = -EFAULT; + struct perf_event_context *ctx = leader->ctx; + u64 values[5]; + u64 count, enabled, running; + + mutex_lock(&ctx->mutex); + count = perf_event_read_value(leader, &enabled, &running); + + values[n++] = 1 + leader->nr_siblings; + if (read_format & PERF_FORMAT_TOTAL_TIME_ENABLED) + values[n++] = enabled; + if (read_format & PERF_FORMAT_TOTAL_TIME_RUNNING) + values[n++] = running; + values[n++] = count; + if (read_format & PERF_FORMAT_ID) + values[n++] = primary_event_id(leader); + + size = n * sizeof(u64); + + if (copy_to_user(buf, values, size)) + goto unlock; + + ret = size; + + list_for_each_entry(sub, &leader->sibling_list, group_entry) { + n = 0; + + values[n++] = perf_event_read_value(sub, &enabled, &running); + if (read_format & PERF_FORMAT_ID) + values[n++] = primary_event_id(sub); + + size = n * sizeof(u64); + + if (copy_to_user(buf + ret, values, size)) { + ret = -EFAULT; + goto unlock; + } + + ret += size; + } +unlock: + mutex_unlock(&ctx->mutex); + + return ret; +} + +static int perf_event_read_one(struct perf_event *event, + u64 read_format, char __user *buf) +{ + u64 enabled, running; + u64 values[4]; + int n = 0; + + values[n++] = perf_event_read_value(event, &enabled, &running); + if (read_format & PERF_FORMAT_TOTAL_TIME_ENABLED) + values[n++] = enabled; + if (read_format & PERF_FORMAT_TOTAL_TIME_RUNNING) + values[n++] = running; + if (read_format & PERF_FORMAT_ID) + values[n++] = primary_event_id(event); + + if (copy_to_user(buf, values, n * sizeof(u64))) + return -EFAULT; + + return n * sizeof(u64); +} + +/* + * Read the performance event - simple non blocking version for now + */ +static ssize_t +perf_read_hw(struct perf_event *event, char __user *buf, size_t count) +{ + u64 read_format = event->attr.read_format; + int ret; + + /* + * Return end-of-file for a read on a event that is in + * error state (i.e. because it was pinned but it couldn't be + * scheduled on to the CPU at some point). + */ + if (event->state == PERF_EVENT_STATE_ERROR) + return 0; + + if (count < perf_event_read_size(event)) + return -ENOSPC; + + WARN_ON_ONCE(event->ctx->parent_ctx); + if (read_format & PERF_FORMAT_GROUP) + ret = perf_event_read_group(event, read_format, buf); + else + ret = perf_event_read_one(event, read_format, buf); + + return ret; +} + +static ssize_t +perf_read(struct file *file, char __user *buf, size_t count, loff_t *ppos) +{ + struct perf_event *event = file->private_data; + + return perf_read_hw(event, buf, count); +} + +static unsigned int perf_poll(struct file *file, poll_table *wait) +{ + struct perf_event *event = file->private_data; + struct perf_mmap_data *data; + unsigned int events = POLL_HUP; + + rcu_read_lock(); + data = rcu_dereference(event->data); + if (data) + events = atomic_xchg(&data->poll, 0); + rcu_read_unlock(); + + poll_wait(file, &event->waitq, wait); + + return events; +} + +static void perf_event_reset(struct perf_event *event) +{ + (void)perf_event_read(event); + atomic64_set(&event->count, 0); + perf_event_update_userpage(event); +} + +/* + * Holding the top-level event's child_mutex means that any + * descendant process that has inherited this event will block + * in sync_child_event if it goes to exit, thus satisfying the + * task existence requirements of perf_event_enable/disable. + */ +static void perf_event_for_each_child(struct perf_event *event, + void (*func)(struct perf_event *)) +{ + struct perf_event *child; + + WARN_ON_ONCE(event->ctx->parent_ctx); + mutex_lock(&event->child_mutex); + func(event); + list_for_each_entry(child, &event->child_list, child_list) + func(child); + mutex_unlock(&event->child_mutex); +} + +static void perf_event_for_each(struct perf_event *event, + void (*func)(struct perf_event *)) +{ + struct perf_event_context *ctx = event->ctx; + struct perf_event *sibling; + + WARN_ON_ONCE(ctx->parent_ctx); + mutex_lock(&ctx->mutex); + event = event->group_leader; + + perf_event_for_each_child(event, func); + func(event); + list_for_each_entry(sibling, &event->sibling_list, group_entry) + perf_event_for_each_child(event, func); + mutex_unlock(&ctx->mutex); +} + +static int perf_event_period(struct perf_event *event, u64 __user *arg) +{ + struct perf_event_context *ctx = event->ctx; + unsigned long size; + int ret = 0; + u64 value; + + if (!event->attr.sample_period) + return -EINVAL; + + size = copy_from_user(&value, arg, sizeof(value)); + if (size != sizeof(value)) + return -EFAULT; + + if (!value) + return -EINVAL; + + spin_lock_irq(&ctx->lock); + if (event->attr.freq) { + if (value > sysctl_perf_event_sample_rate) { + ret = -EINVAL; + goto unlock; + } + + event->attr.sample_freq = value; + } else { + event->attr.sample_period = value; + event->hw.sample_period = value; + } +unlock: + spin_unlock_irq(&ctx->lock); + + return ret; +} + +static int perf_event_set_output(struct perf_event *event, int output_fd); +static int perf_event_set_filter(struct perf_event *event, void __user *arg); + +static long perf_ioctl(struct file *file, unsigned int cmd, unsigned long arg) +{ + struct perf_event *event = file->private_data; + void (*func)(struct perf_event *); + u32 flags = arg; + + switch (cmd) { + case PERF_EVENT_IOC_ENABLE: + func = perf_event_enable; + break; + case PERF_EVENT_IOC_DISABLE: + func = perf_event_disable; + break; + case PERF_EVENT_IOC_RESET: + func = perf_event_reset; + break; + + case PERF_EVENT_IOC_REFRESH: + return perf_event_refresh(event, arg); + + case PERF_EVENT_IOC_PERIOD: + return perf_event_period(event, (u64 __user *)arg); + + case PERF_EVENT_IOC_SET_OUTPUT: + return perf_event_set_output(event, arg); + + case PERF_EVENT_IOC_SET_FILTER: + return perf_event_set_filter(event, (void __user *)arg); + + default: + return -ENOTTY; + } + + if (flags & PERF_IOC_FLAG_GROUP) + perf_event_for_each(event, func); + else + perf_event_for_each_child(event, func); + + return 0; +} + +int perf_event_task_enable(void) +{ + struct perf_event *event; + + mutex_lock(¤t->perf_event_mutex); + list_for_each_entry(event, ¤t->perf_event_list, owner_entry) + perf_event_for_each_child(event, perf_event_enable); + mutex_unlock(¤t->perf_event_mutex); + + return 0; +} + +int perf_event_task_disable(void) +{ + struct perf_event *event; + + mutex_lock(¤t->perf_event_mutex); + list_for_each_entry(event, ¤t->perf_event_list, owner_entry) + perf_event_for_each_child(event, perf_event_disable); + mutex_unlock(¤t->perf_event_mutex); + + return 0; +} + +#ifndef PERF_EVENT_INDEX_OFFSET +# define PERF_EVENT_INDEX_OFFSET 0 +#endif + +static int perf_event_index(struct perf_event *event) +{ + if (event->state != PERF_EVENT_STATE_ACTIVE) + return 0; + + return event->hw.idx + 1 - PERF_EVENT_INDEX_OFFSET; +} + +/* + * Callers need to ensure there can be no nesting of this function, otherwise + * the seqlock logic goes bad. We can not serialize this because the arch + * code calls this from NMI context. + */ +void perf_event_update_userpage(struct perf_event *event) +{ + struct perf_event_mmap_page *userpg; + struct perf_mmap_data *data; + + rcu_read_lock(); + data = rcu_dereference(event->data); + if (!data) + goto unlock; + + userpg = data->user_page; + + /* + * Disable preemption so as to not let the corresponding user-space + * spin too long if we get preempted. + */ + preempt_disable(); + ++userpg->lock; + barrier(); + userpg->index = perf_event_index(event); + userpg->offset = atomic64_read(&event->count); + if (event->state == PERF_EVENT_STATE_ACTIVE) + userpg->offset -= atomic64_read(&event->hw.prev_count); + + userpg->time_enabled = event->total_time_enabled + + atomic64_read(&event->child_total_time_enabled); + + userpg->time_running = event->total_time_running + + atomic64_read(&event->child_total_time_running); + + barrier(); + ++userpg->lock; + preempt_enable(); +unlock: + rcu_read_unlock(); +} + +static unsigned long perf_data_size(struct perf_mmap_data *data) +{ + return data->nr_pages << (PAGE_SHIFT + data->data_order); +} + +#ifndef CONFIG_PERF_USE_VMALLOC + +/* + * Back perf_mmap() with regular GFP_KERNEL-0 pages. + */ + +static struct page * +perf_mmap_to_page(struct perf_mmap_data *data, unsigned long pgoff) +{ + if (pgoff > data->nr_pages) + return NULL; + + if (pgoff == 0) + return virt_to_page(data->user_page); + + return virt_to_page(data->data_pages[pgoff - 1]); +} + +static struct perf_mmap_data * +perf_mmap_data_alloc(struct perf_event *event, int nr_pages) +{ + struct perf_mmap_data *data; + unsigned long size; + int i; + + WARN_ON(atomic_read(&event->mmap_count)); + + size = sizeof(struct perf_mmap_data); + size += nr_pages * sizeof(void *); + + data = kzalloc(size, GFP_KERNEL); + if (!data) + goto fail; + + data->user_page = (void *)get_zeroed_page(GFP_KERNEL); + if (!data->user_page) + goto fail_user_page; + + for (i = 0; i < nr_pages; i++) { + data->data_pages[i] = (void *)get_zeroed_page(GFP_KERNEL); + if (!data->data_pages[i]) + goto fail_data_pages; + } + + data->data_order = 0; + data->nr_pages = nr_pages; + + return data; + +fail_data_pages: + for (i--; i >= 0; i--) + free_page((unsigned long)data->data_pages[i]); + + free_page((unsigned long)data->user_page); + +fail_user_page: + kfree(data); + +fail: + return NULL; +} + +static void perf_mmap_free_page(unsigned long addr) +{ + struct page *page = virt_to_page((void *)addr); + + page->mapping = NULL; + __free_page(page); +} + +static void perf_mmap_data_free(struct perf_mmap_data *data) +{ + int i; + + perf_mmap_free_page((unsigned long)data->user_page); + for (i = 0; i < data->nr_pages; i++) + perf_mmap_free_page((unsigned long)data->data_pages[i]); + kfree(data); +} + +#else + +/* + * Back perf_mmap() with vmalloc memory. + * + * Required for architectures that have d-cache aliasing issues. + */ + +static struct page * +perf_mmap_to_page(struct perf_mmap_data *data, unsigned long pgoff) +{ + if (pgoff > (1UL << data->data_order)) + return NULL; + + return vmalloc_to_page((void *)data->user_page + pgoff * PAGE_SIZE); +} + +static void perf_mmap_unmark_page(void *addr) +{ + struct page *page = vmalloc_to_page(addr); + + page->mapping = NULL; +} + +static void perf_mmap_data_free_work(struct work_struct *work) +{ + struct perf_mmap_data *data; + void *base; + int i, nr; + + data = container_of(work, struct perf_mmap_data, work); + nr = 1 << data->data_order; + + base = data->user_page; + for (i = 0; i < nr + 1; i++) + perf_mmap_unmark_page(base + (i * PAGE_SIZE)); + + vfree(base); + kfree(data); +} + +static void perf_mmap_data_free(struct perf_mmap_data *data) +{ + schedule_work(&data->work); +} + +static struct perf_mmap_data * +perf_mmap_data_alloc(struct perf_event *event, int nr_pages) +{ + struct perf_mmap_data *data; + unsigned long size; + void *all_buf; + + WARN_ON(atomic_read(&event->mmap_count)); + + size = sizeof(struct perf_mmap_data); + size += sizeof(void *); + + data = kzalloc(size, GFP_KERNEL); + if (!data) + goto fail; + + INIT_WORK(&data->work, perf_mmap_data_free_work); + + all_buf = vmalloc_user((nr_pages + 1) * PAGE_SIZE); + if (!all_buf) + goto fail_all_buf; + + data->user_page = all_buf; + data->data_pages[0] = all_buf + PAGE_SIZE; + data->data_order = ilog2(nr_pages); + data->nr_pages = 1; + + return data; + +fail_all_buf: + kfree(data); + +fail: + return NULL; +} + +#endif + +static int perf_mmap_fault(struct vm_area_struct *vma, struct vm_fault *vmf) +{ + struct perf_event *event = vma->vm_file->private_data; + struct perf_mmap_data *data; + int ret = VM_FAULT_SIGBUS; + + if (vmf->flags & FAULT_FLAG_MKWRITE) { + if (vmf->pgoff == 0) + ret = 0; + return ret; + } + + rcu_read_lock(); + data = rcu_dereference(event->data); + if (!data) + goto unlock; + + if (vmf->pgoff && (vmf->flags & FAULT_FLAG_WRITE)) + goto unlock; + + vmf->page = perf_mmap_to_page(data, vmf->pgoff); + if (!vmf->page) + goto unlock; + + get_page(vmf->page); + vmf->page->mapping = vma->vm_file->f_mapping; + vmf->page->index = vmf->pgoff; + + ret = 0; +unlock: + rcu_read_unlock(); + + return ret; +} + +static void +perf_mmap_data_init(struct perf_event *event, struct perf_mmap_data *data) +{ + long max_size = perf_data_size(data); + + atomic_set(&data->lock, -1); + + if (event->attr.watermark) { + data->watermark = min_t(long, max_size, + event->attr.wakeup_watermark); + } + + if (!data->watermark) + data->watermark = max_size / 2; + + + rcu_assign_pointer(event->data, data); +} + +static void perf_mmap_data_free_rcu(struct rcu_head *rcu_head) +{ + struct perf_mmap_data *data; + + data = container_of(rcu_head, struct perf_mmap_data, rcu_head); + perf_mmap_data_free(data); +} + +static void perf_mmap_data_release(struct perf_event *event) +{ + struct perf_mmap_data *data = event->data; + + WARN_ON(atomic_read(&event->mmap_count)); + + rcu_assign_pointer(event->data, NULL); + call_rcu(&data->rcu_head, perf_mmap_data_free_rcu); +} + +static void perf_mmap_open(struct vm_area_struct *vma) +{ + struct perf_event *event = vma->vm_file->private_data; + + atomic_inc(&event->mmap_count); +} + +static void perf_mmap_close(struct vm_area_struct *vma) +{ + struct perf_event *event = vma->vm_file->private_data; + + WARN_ON_ONCE(event->ctx->parent_ctx); + if (atomic_dec_and_mutex_lock(&event->mmap_count, &event->mmap_mutex)) { + unsigned long size = perf_data_size(event->data); + struct user_struct *user = current_user(); + + atomic_long_sub((size >> PAGE_SHIFT) + 1, &user->locked_vm); + vma->vm_mm->locked_vm -= event->data->nr_locked; + perf_mmap_data_release(event); + mutex_unlock(&event->mmap_mutex); + } +} + +static const struct vm_operations_struct perf_mmap_vmops = { + .open = perf_mmap_open, + .close = perf_mmap_close, + .fault = perf_mmap_fault, + .page_mkwrite = perf_mmap_fault, +}; + +static int perf_mmap(struct file *file, struct vm_area_struct *vma) +{ + struct perf_event *event = file->private_data; + unsigned long user_locked, user_lock_limit; + struct user_struct *user = current_user(); + unsigned long locked, lock_limit; + struct perf_mmap_data *data; + unsigned long vma_size; + unsigned long nr_pages; + long user_extra, extra; + int ret = 0; + + if (!(vma->vm_flags & VM_SHARED)) + return -EINVAL; + + vma_size = vma->vm_end - vma->vm_start; + nr_pages = (vma_size / PAGE_SIZE) - 1; + + /* + * If we have data pages ensure they're a power-of-two number, so we + * can do bitmasks instead of modulo. + */ + if (nr_pages != 0 && !is_power_of_2(nr_pages)) + return -EINVAL; + + if (vma_size != PAGE_SIZE * (1 + nr_pages)) + return -EINVAL; + + if (vma->vm_pgoff != 0) + return -EINVAL; + + WARN_ON_ONCE(event->ctx->parent_ctx); + mutex_lock(&event->mmap_mutex); + if (event->output) { + ret = -EINVAL; + goto unlock; + } + + if (atomic_inc_not_zero(&event->mmap_count)) { + if (nr_pages != event->data->nr_pages) + ret = -EINVAL; + goto unlock; + } + + user_extra = nr_pages + 1; + user_lock_limit = sysctl_perf_event_mlock >> (PAGE_SHIFT - 10); + + /* + * Increase the limit linearly with more CPUs: + */ + user_lock_limit *= num_online_cpus(); + + user_locked = atomic_long_read(&user->locked_vm) + user_extra; + + extra = 0; + if (user_locked > user_lock_limit) + extra = user_locked - user_lock_limit; + + lock_limit = current->signal->rlim[RLIMIT_MEMLOCK].rlim_cur; + lock_limit >>= PAGE_SHIFT; + locked = vma->vm_mm->locked_vm + extra; + + if ((locked > lock_limit) && perf_paranoid_tracepoint_raw() && + !capable(CAP_IPC_LOCK)) { + ret = -EPERM; + goto unlock; + } + + WARN_ON(event->data); + + data = perf_mmap_data_alloc(event, nr_pages); + ret = -ENOMEM; + if (!data) + goto unlock; + + ret = 0; + perf_mmap_data_init(event, data); + + atomic_set(&event->mmap_count, 1); + atomic_long_add(user_extra, &user->locked_vm); + vma->vm_mm->locked_vm += extra; + event->data->nr_locked = extra; + if (vma->vm_flags & VM_WRITE) + event->data->writable = 1; + +unlock: + mutex_unlock(&event->mmap_mutex); + + vma->vm_flags |= VM_RESERVED; + vma->vm_ops = &perf_mmap_vmops; + + return ret; +} + +static int perf_fasync(int fd, struct file *filp, int on) +{ + struct inode *inode = filp->f_path.dentry->d_inode; + struct perf_event *event = filp->private_data; + int retval; + + mutex_lock(&inode->i_mutex); + retval = fasync_helper(fd, filp, on, &event->fasync); + mutex_unlock(&inode->i_mutex); + + if (retval < 0) + return retval; + + return 0; +} + +static const struct file_operations perf_fops = { + .release = perf_release, + .read = perf_read, + .poll = perf_poll, + .unlocked_ioctl = perf_ioctl, + .compat_ioctl = perf_ioctl, + .mmap = perf_mmap, + .fasync = perf_fasync, +}; + +/* + * Perf event wakeup + * + * If there's data, ensure we set the poll() state and publish everything + * to user-space before waking everybody up. + */ + +void perf_event_wakeup(struct perf_event *event) +{ + wake_up_all(&event->waitq); + + if (event->pending_kill) { + kill_fasync(&event->fasync, SIGIO, event->pending_kill); + event->pending_kill = 0; + } +} + +/* + * Pending wakeups + * + * Handle the case where we need to wakeup up from NMI (or rq->lock) context. + * + * The NMI bit means we cannot possibly take locks. Therefore, maintain a + * single linked list and use cmpxchg() to add entries lockless. + */ + +static void perf_pending_event(struct perf_pending_entry *entry) +{ + struct perf_event *event = container_of(entry, + struct perf_event, pending); + + if (event->pending_disable) { + event->pending_disable = 0; + __perf_event_disable(event); + } + + if (event->pending_wakeup) { + event->pending_wakeup = 0; + perf_event_wakeup(event); + } +} + +#define PENDING_TAIL ((struct perf_pending_entry *)-1UL) + +static DEFINE_PER_CPU(struct perf_pending_entry *, perf_pending_head) = { + PENDING_TAIL, +}; + +static void perf_pending_queue(struct perf_pending_entry *entry, + void (*func)(struct perf_pending_entry *)) +{ + struct perf_pending_entry **head; + + if (cmpxchg(&entry->next, NULL, PENDING_TAIL) != NULL) + return; + + entry->func = func; + + head = &get_cpu_var(perf_pending_head); + + do { + entry->next = *head; + } while (cmpxchg(head, entry->next, entry) != entry->next); + + set_perf_event_pending(); + + put_cpu_var(perf_pending_head); +} + +static int __perf_pending_run(void) +{ + struct perf_pending_entry *list; + int nr = 0; + + list = xchg(&__get_cpu_var(perf_pending_head), PENDING_TAIL); + while (list != PENDING_TAIL) { + void (*func)(struct perf_pending_entry *); + struct perf_pending_entry *entry = list; + + list = list->next; + + func = entry->func; + entry->next = NULL; + /* + * Ensure we observe the unqueue before we issue the wakeup, + * so that we won't be waiting forever. + * -- see perf_not_pending(). + */ + smp_wmb(); + + func(entry); + nr++; + } + + return nr; +} + +static inline int perf_not_pending(struct perf_event *event) +{ + /* + * If we flush on whatever cpu we run, there is a chance we don't + * need to wait. + */ + get_cpu(); + __perf_pending_run(); + put_cpu(); + + /* + * Ensure we see the proper queue state before going to sleep + * so that we do not miss the wakeup. -- see perf_pending_handle() + */ + smp_rmb(); + return event->pending.next == NULL; +} + +static void perf_pending_sync(struct perf_event *event) +{ + wait_event(event->waitq, perf_not_pending(event)); +} + +void perf_event_do_pending(void) +{ + __perf_pending_run(); +} + +/* + * Callchain support -- arch specific + */ + +__weak struct perf_callchain_entry *perf_callchain(struct pt_regs *regs) +{ + return NULL; +} + +/* + * Output + */ +static bool perf_output_space(struct perf_mmap_data *data, unsigned long tail, + unsigned long offset, unsigned long head) +{ + unsigned long mask; + + if (!data->writable) + return true; + + mask = perf_data_size(data) - 1; + + offset = (offset - tail) & mask; + head = (head - tail) & mask; + + if ((int)(head - offset) < 0) + return false; + + return true; +} + +static void perf_output_wakeup(struct perf_output_handle *handle) +{ + atomic_set(&handle->data->poll, POLL_IN); + + if (handle->nmi) { + handle->event->pending_wakeup = 1; + perf_pending_queue(&handle->event->pending, + perf_pending_event); + } else + perf_event_wakeup(handle->event); +} + +/* + * Curious locking construct. + * + * We need to ensure a later event_id doesn't publish a head when a former + * event_id isn't done writing. However since we need to deal with NMIs we + * cannot fully serialize things. + * + * What we do is serialize between CPUs so we only have to deal with NMI + * nesting on a single CPU. + * + * We only publish the head (and generate a wakeup) when the outer-most + * event_id completes. + */ +static void perf_output_lock(struct perf_output_handle *handle) +{ + struct perf_mmap_data *data = handle->data; + int cur, cpu = get_cpu(); + + handle->locked = 0; + + for (;;) { + cur = atomic_cmpxchg(&data->lock, -1, cpu); + if (cur == -1) { + handle->locked = 1; + break; + } + if (cur == cpu) + break; + + cpu_relax(); + } +} + +static void perf_output_unlock(struct perf_output_handle *handle) +{ + struct perf_mmap_data *data = handle->data; + unsigned long head; + int cpu; + + data->done_head = data->head; + + if (!handle->locked) + goto out; + +again: + /* + * The xchg implies a full barrier that ensures all writes are done + * before we publish the new head, matched by a rmb() in userspace when + * reading this position. + */ + while ((head = atomic_long_xchg(&data->done_head, 0))) + data->user_page->data_head = head; + + /* + * NMI can happen here, which means we can miss a done_head update. + */ + + cpu = atomic_xchg(&data->lock, -1); + WARN_ON_ONCE(cpu != smp_processor_id()); + + /* + * Therefore we have to validate we did not indeed do so. + */ + if (unlikely(atomic_long_read(&data->done_head))) { + /* + * Since we had it locked, we can lock it again. + */ + while (atomic_cmpxchg(&data->lock, -1, cpu) != -1) + cpu_relax(); + + goto again; + } + + if (atomic_xchg(&data->wakeup, 0)) + perf_output_wakeup(handle); +out: + put_cpu(); +} + +void perf_output_copy(struct perf_output_handle *handle, + const void *buf, unsigned int len) +{ + unsigned int pages_mask; + unsigned long offset; + unsigned int size; + void **pages; + + offset = handle->offset; + pages_mask = handle->data->nr_pages - 1; + pages = handle->data->data_pages; + + do { + unsigned long page_offset; + unsigned long page_size; + int nr; + + nr = (offset >> PAGE_SHIFT) & pages_mask; + page_size = 1UL << (handle->data->data_order + PAGE_SHIFT); + page_offset = offset & (page_size - 1); + size = min_t(unsigned int, page_size - page_offset, len); + + memcpy(pages[nr] + page_offset, buf, size); + + len -= size; + buf += size; + offset += size; + } while (len); + + handle->offset = offset; + + /* + * Check we didn't copy past our reservation window, taking the + * possible unsigned int wrap into account. + */ + WARN_ON_ONCE(((long)(handle->head - handle->offset)) < 0); +} + +int perf_output_begin(struct perf_output_handle *handle, + struct perf_event *event, unsigned int size, + int nmi, int sample) +{ + struct perf_event *output_event; + struct perf_mmap_data *data; + unsigned long tail, offset, head; + int have_lost; + struct { + struct perf_event_header header; + u64 id; + u64 lost; + } lost_event; + + rcu_read_lock(); + /* + * For inherited events we send all the output towards the parent. + */ + if (event->parent) + event = event->parent; + + output_event = rcu_dereference(event->output); + if (output_event) + event = output_event; + + data = rcu_dereference(event->data); + if (!data) + goto out; + + handle->data = data; + handle->event = event; + handle->nmi = nmi; + handle->sample = sample; + + if (!data->nr_pages) + goto fail; + + have_lost = atomic_read(&data->lost); + if (have_lost) + size += sizeof(lost_event); + + perf_output_lock(handle); + + do { + /* + * Userspace could choose to issue a mb() before updating the + * tail pointer. So that all reads will be completed before the + * write is issued. + */ + tail = ACCESS_ONCE(data->user_page->data_tail); + smp_rmb(); + offset = head = atomic_long_read(&data->head); + head += size; + if (unlikely(!perf_output_space(data, tail, offset, head))) + goto fail; + } while (atomic_long_cmpxchg(&data->head, offset, head) != offset); + + handle->offset = offset; + handle->head = head; + + if (head - tail > data->watermark) + atomic_set(&data->wakeup, 1); + + if (have_lost) { + lost_event.header.type = PERF_RECORD_LOST; + lost_event.header.misc = 0; + lost_event.header.size = sizeof(lost_event); + lost_event.id = event->id; + lost_event.lost = atomic_xchg(&data->lost, 0); + + perf_output_put(handle, lost_event); + } + + return 0; + +fail: + atomic_inc(&data->lost); + perf_output_unlock(handle); +out: + rcu_read_unlock(); + + return -ENOSPC; +} + +void perf_output_end(struct perf_output_handle *handle) +{ + struct perf_event *event = handle->event; + struct perf_mmap_data *data = handle->data; + + int wakeup_events = event->attr.wakeup_events; + + if (handle->sample && wakeup_events) { + int events = atomic_inc_return(&data->events); + if (events >= wakeup_events) { + atomic_sub(wakeup_events, &data->events); + atomic_set(&data->wakeup, 1); + } + } + + perf_output_unlock(handle); + rcu_read_unlock(); +} + +static u32 perf_event_pid(struct perf_event *event, struct task_struct *p) +{ + /* + * only top level events have the pid namespace they were created in + */ + if (event->parent) + event = event->parent; + + return task_tgid_nr_ns(p, event->ns); +} + +static u32 perf_event_tid(struct perf_event *event, struct task_struct *p) +{ + /* + * only top level events have the pid namespace they were created in + */ + if (event->parent) + event = event->parent; + + return task_pid_nr_ns(p, event->ns); +} + +static void perf_output_read_one(struct perf_output_handle *handle, + struct perf_event *event) +{ + u64 read_format = event->attr.read_format; + u64 values[4]; + int n = 0; + + values[n++] = atomic64_read(&event->count); + if (read_format & PERF_FORMAT_TOTAL_TIME_ENABLED) { + values[n++] = event->total_time_enabled + + atomic64_read(&event->child_total_time_enabled); + } + if (read_format & PERF_FORMAT_TOTAL_TIME_RUNNING) { + values[n++] = event->total_time_running + + atomic64_read(&event->child_total_time_running); + } + if (read_format & PERF_FORMAT_ID) + values[n++] = primary_event_id(event); + + perf_output_copy(handle, values, n * sizeof(u64)); +} + +/* + * XXX PERF_FORMAT_GROUP vs inherited events seems difficult. + */ +static void perf_output_read_group(struct perf_output_handle *handle, + struct perf_event *event) +{ + struct perf_event *leader = event->group_leader, *sub; + u64 read_format = event->attr.read_format; + u64 values[5]; + int n = 0; + + values[n++] = 1 + leader->nr_siblings; + + if (read_format & PERF_FORMAT_TOTAL_TIME_ENABLED) + values[n++] = leader->total_time_enabled; + + if (read_format & PERF_FORMAT_TOTAL_TIME_RUNNING) + values[n++] = leader->total_time_running; + + if (leader != event) + leader->pmu->read(leader); + + values[n++] = atomic64_read(&leader->count); + if (read_format & PERF_FORMAT_ID) + values[n++] = primary_event_id(leader); + + perf_output_copy(handle, values, n * sizeof(u64)); + + list_for_each_entry(sub, &leader->sibling_list, group_entry) { + n = 0; + + if (sub != event) + sub->pmu->read(sub); + + values[n++] = atomic64_read(&sub->count); + if (read_format & PERF_FORMAT_ID) + values[n++] = primary_event_id(sub); + + perf_output_copy(handle, values, n * sizeof(u64)); + } +} + +static void perf_output_read(struct perf_output_handle *handle, + struct perf_event *event) +{ + if (event->attr.read_format & PERF_FORMAT_GROUP) + perf_output_read_group(handle, event); + else + perf_output_read_one(handle, event); +} + +void perf_output_sample(struct perf_output_handle *handle, + struct perf_event_header *header, + struct perf_sample_data *data, + struct perf_event *event) +{ + u64 sample_type = data->type; + + perf_output_put(handle, *header); + + if (sample_type & PERF_SAMPLE_IP) + perf_output_put(handle, data->ip); + + if (sample_type & PERF_SAMPLE_TID) + perf_output_put(handle, data->tid_entry); + + if (sample_type & PERF_SAMPLE_TIME) + perf_output_put(handle, data->time); + + if (sample_type & PERF_SAMPLE_ADDR) + perf_output_put(handle, data->addr); + + if (sample_type & PERF_SAMPLE_ID) + perf_output_put(handle, data->id); + + if (sample_type & PERF_SAMPLE_STREAM_ID) + perf_output_put(handle, data->stream_id); + + if (sample_type & PERF_SAMPLE_CPU) + perf_output_put(handle, data->cpu_entry); + + if (sample_type & PERF_SAMPLE_PERIOD) + perf_output_put(handle, data->period); + + if (sample_type & PERF_SAMPLE_READ) + perf_output_read(handle, event); + + if (sample_type & PERF_SAMPLE_CALLCHAIN) { + if (data->callchain) { + int size = 1; + + if (data->callchain) + size += data->callchain->nr; + + size *= sizeof(u64); + + perf_output_copy(handle, data->callchain, size); + } else { + u64 nr = 0; + perf_output_put(handle, nr); + } + } + + if (sample_type & PERF_SAMPLE_RAW) { + if (data->raw) { + perf_output_put(handle, data->raw->size); + perf_output_copy(handle, data->raw->data, + data->raw->size); + } else { + struct { + u32 size; + u32 data; + } raw = { + .size = sizeof(u32), + .data = 0, + }; + perf_output_put(handle, raw); + } + } +} + +void perf_prepare_sample(struct perf_event_header *header, + struct perf_sample_data *data, + struct perf_event *event, + struct pt_regs *regs) +{ + u64 sample_type = event->attr.sample_type; + + data->type = sample_type; + + header->type = PERF_RECORD_SAMPLE; + header->size = sizeof(*header); + + header->misc = 0; + header->misc |= perf_misc_flags(regs); + + if (sample_type & PERF_SAMPLE_IP) { + data->ip = perf_instruction_pointer(regs); + + header->size += sizeof(data->ip); + } + + if (sample_type & PERF_SAMPLE_TID) { + /* namespace issues */ + data->tid_entry.pid = perf_event_pid(event, current); + data->tid_entry.tid = perf_event_tid(event, current); + + header->size += sizeof(data->tid_entry); + } + + if (sample_type & PERF_SAMPLE_TIME) { + data->time = perf_clock(); + + header->size += sizeof(data->time); + } + + if (sample_type & PERF_SAMPLE_ADDR) + header->size += sizeof(data->addr); + + if (sample_type & PERF_SAMPLE_ID) { + data->id = primary_event_id(event); + + header->size += sizeof(data->id); + } + + if (sample_type & PERF_SAMPLE_STREAM_ID) { + data->stream_id = event->id; + + header->size += sizeof(data->stream_id); + } + + if (sample_type & PERF_SAMPLE_CPU) { + data->cpu_entry.cpu = raw_smp_processor_id(); + data->cpu_entry.reserved = 0; + + header->size += sizeof(data->cpu_entry); + } + + if (sample_type & PERF_SAMPLE_PERIOD) + header->size += sizeof(data->period); + + if (sample_type & PERF_SAMPLE_READ) + header->size += perf_event_read_size(event); + + if (sample_type & PERF_SAMPLE_CALLCHAIN) { + int size = 1; + + data->callchain = perf_callchain(regs); + + if (data->callchain) + size += data->callchain->nr; + + header->size += size * sizeof(u64); + } + + if (sample_type & PERF_SAMPLE_RAW) { + int size = sizeof(u32); + + if (data->raw) + size += data->raw->size; + else + size += sizeof(u32); + + WARN_ON_ONCE(size & (sizeof(u64)-1)); + header->size += size; + } +} + +static void perf_event_output(struct perf_event *event, int nmi, + struct perf_sample_data *data, + struct pt_regs *regs) +{ + struct perf_output_handle handle; + struct perf_event_header header; + + perf_prepare_sample(&header, data, event, regs); + + if (perf_output_begin(&handle, event, header.size, nmi, 1)) + return; + + perf_output_sample(&handle, &header, data, event); + + perf_output_end(&handle); +} + +/* + * read event_id + */ + +struct perf_read_event { + struct perf_event_header header; + + u32 pid; + u32 tid; +}; + +static void +perf_event_read_event(struct perf_event *event, + struct task_struct *task) +{ + struct perf_output_handle handle; + struct perf_read_event read_event = { + .header = { + .type = PERF_RECORD_READ, + .misc = 0, + .size = sizeof(read_event) + perf_event_read_size(event), + }, + .pid = perf_event_pid(event, task), + .tid = perf_event_tid(event, task), + }; + int ret; + + ret = perf_output_begin(&handle, event, read_event.header.size, 0, 0); + if (ret) + return; + + perf_output_put(&handle, read_event); + perf_output_read(&handle, event); + + perf_output_end(&handle); +} + +/* + * task tracking -- fork/exit + * + * enabled by: attr.comm | attr.mmap | attr.task + */ + +struct perf_task_event { + struct task_struct *task; + struct perf_event_context *task_ctx; + + struct { + struct perf_event_header header; + + u32 pid; + u32 ppid; + u32 tid; + u32 ptid; + u64 time; + } event_id; +}; + +static void perf_event_task_output(struct perf_event *event, + struct perf_task_event *task_event) +{ + struct perf_output_handle handle; + int size; + struct task_struct *task = task_event->task; + int ret; + + size = task_event->event_id.header.size; + ret = perf_output_begin(&handle, event, size, 0, 0); + + if (ret) + return; + + task_event->event_id.pid = perf_event_pid(event, task); + task_event->event_id.ppid = perf_event_pid(event, current); + + task_event->event_id.tid = perf_event_tid(event, task); + task_event->event_id.ptid = perf_event_tid(event, current); + + task_event->event_id.time = perf_clock(); + + perf_output_put(&handle, task_event->event_id); + + perf_output_end(&handle); +} + +static int perf_event_task_match(struct perf_event *event) +{ + if (event->attr.comm || event->attr.mmap || event->attr.task) + return 1; + + return 0; +} + +static void perf_event_task_ctx(struct perf_event_context *ctx, + struct perf_task_event *task_event) +{ + struct perf_event *event; + + list_for_each_entry_rcu(event, &ctx->event_list, event_entry) { + if (perf_event_task_match(event)) + perf_event_task_output(event, task_event); + } +} + +static void perf_event_task_event(struct perf_task_event *task_event) +{ + struct perf_cpu_context *cpuctx; + struct perf_event_context *ctx = task_event->task_ctx; + + rcu_read_lock(); + cpuctx = &get_cpu_var(perf_cpu_context); + perf_event_task_ctx(&cpuctx->ctx, task_event); + put_cpu_var(perf_cpu_context); + + if (!ctx) + ctx = rcu_dereference(task_event->task->perf_event_ctxp); + if (ctx) + perf_event_task_ctx(ctx, task_event); + rcu_read_unlock(); +} + +static void perf_event_task(struct task_struct *task, + struct perf_event_context *task_ctx, + int new) +{ + struct perf_task_event task_event; + + if (!atomic_read(&nr_comm_events) && + !atomic_read(&nr_mmap_events) && + !atomic_read(&nr_task_events)) + return; + + task_event = (struct perf_task_event){ + .task = task, + .task_ctx = task_ctx, + .event_id = { + .header = { + .type = new ? PERF_RECORD_FORK : PERF_RECORD_EXIT, + .misc = 0, + .size = sizeof(task_event.event_id), + }, + /* .pid */ + /* .ppid */ + /* .tid */ + /* .ptid */ + }, + }; + + perf_event_task_event(&task_event); +} + +void perf_event_fork(struct task_struct *task) +{ + perf_event_task(task, NULL, 1); +} + +/* + * comm tracking + */ + +struct perf_comm_event { + struct task_struct *task; + char *comm; + int comm_size; + + struct { + struct perf_event_header header; + + u32 pid; + u32 tid; + } event_id; +}; + +static void perf_event_comm_output(struct perf_event *event, + struct perf_comm_event *comm_event) +{ + struct perf_output_handle handle; + int size = comm_event->event_id.header.size; + int ret = perf_output_begin(&handle, event, size, 0, 0); + + if (ret) + return; + + comm_event->event_id.pid = perf_event_pid(event, comm_event->task); + comm_event->event_id.tid = perf_event_tid(event, comm_event->task); + + perf_output_put(&handle, comm_event->event_id); + perf_output_copy(&handle, comm_event->comm, + comm_event->comm_size); + perf_output_end(&handle); +} + +static int perf_event_comm_match(struct perf_event *event) +{ + if (event->attr.comm) + return 1; + + return 0; +} + +static void perf_event_comm_ctx(struct perf_event_context *ctx, + struct perf_comm_event *comm_event) +{ + struct perf_event *event; + + list_for_each_entry_rcu(event, &ctx->event_list, event_entry) { + if (perf_event_comm_match(event)) + perf_event_comm_output(event, comm_event); + } +} + +static void perf_event_comm_event(struct perf_comm_event *comm_event) +{ + struct perf_cpu_context *cpuctx; + struct perf_event_context *ctx; + unsigned int size; + char comm[TASK_COMM_LEN]; + + memset(comm, 0, sizeof(comm)); + strlcpy(comm, comm_event->task->comm, sizeof(comm)); + size = ALIGN(strlen(comm)+1, sizeof(u64)); + + comm_event->comm = comm; + comm_event->comm_size = size; + + comm_event->event_id.header.size = sizeof(comm_event->event_id) + size; + + rcu_read_lock(); + cpuctx = &get_cpu_var(perf_cpu_context); + perf_event_comm_ctx(&cpuctx->ctx, comm_event); + put_cpu_var(perf_cpu_context); + + /* + * doesn't really matter which of the child contexts the + * events ends up in. + */ + ctx = rcu_dereference(current->perf_event_ctxp); + if (ctx) + perf_event_comm_ctx(ctx, comm_event); + rcu_read_unlock(); +} + +void perf_event_comm(struct task_struct *task) +{ + struct perf_comm_event comm_event; + + if (task->perf_event_ctxp) + perf_event_enable_on_exec(task); + + if (!atomic_read(&nr_comm_events)) + return; + + comm_event = (struct perf_comm_event){ + .task = task, + /* .comm */ + /* .comm_size */ + .event_id = { + .header = { + .type = PERF_RECORD_COMM, + .misc = 0, + /* .size */ + }, + /* .pid */ + /* .tid */ + }, + }; + + perf_event_comm_event(&comm_event); +} + +/* + * mmap tracking + */ + +struct perf_mmap_event { + struct vm_area_struct *vma; + + const char *file_name; + int file_size; + + struct { + struct perf_event_header header; + + u32 pid; + u32 tid; + u64 start; + u64 len; + u64 pgoff; + } event_id; +}; + +static void perf_event_mmap_output(struct perf_event *event, + struct perf_mmap_event *mmap_event) +{ + struct perf_output_handle handle; + int size = mmap_event->event_id.header.size; + int ret = perf_output_begin(&handle, event, size, 0, 0); + + if (ret) + return; + + mmap_event->event_id.pid = perf_event_pid(event, current); + mmap_event->event_id.tid = perf_event_tid(event, current); + + perf_output_put(&handle, mmap_event->event_id); + perf_output_copy(&handle, mmap_event->file_name, + mmap_event->file_size); + perf_output_end(&handle); +} + +static int perf_event_mmap_match(struct perf_event *event, + struct perf_mmap_event *mmap_event) +{ + if (event->attr.mmap) + return 1; + + return 0; +} + +static void perf_event_mmap_ctx(struct perf_event_context *ctx, + struct perf_mmap_event *mmap_event) +{ + struct perf_event *event; + + list_for_each_entry_rcu(event, &ctx->event_list, event_entry) { + if (perf_event_mmap_match(event, mmap_event)) + perf_event_mmap_output(event, mmap_event); + } +} + +static void perf_event_mmap_event(struct perf_mmap_event *mmap_event) +{ + struct perf_cpu_context *cpuctx; + struct perf_event_context *ctx; + struct vm_area_struct *vma = mmap_event->vma; + struct file *file = vma->vm_file; + unsigned int size; + char tmp[16]; + char *buf = NULL; + const char *name; + + memset(tmp, 0, sizeof(tmp)); + + if (file) { + /* + * d_path works from the end of the buffer backwards, so we + * need to add enough zero bytes after the string to handle + * the 64bit alignment we do later. + */ + buf = kzalloc(PATH_MAX + sizeof(u64), GFP_KERNEL); + if (!buf) { + name = strncpy(tmp, "//enomem", sizeof(tmp)); + goto got_name; + } + name = d_path(&file->f_path, buf, PATH_MAX); + if (IS_ERR(name)) { + name = strncpy(tmp, "//toolong", sizeof(tmp)); + goto got_name; + } + } else { + if (arch_vma_name(mmap_event->vma)) { + name = strncpy(tmp, arch_vma_name(mmap_event->vma), + sizeof(tmp)); + goto got_name; + } + + if (!vma->vm_mm) { + name = strncpy(tmp, "[vdso]", sizeof(tmp)); + goto got_name; + } + + name = strncpy(tmp, "//anon", sizeof(tmp)); + goto got_name; + } + +got_name: + size = ALIGN(strlen(name)+1, sizeof(u64)); + + mmap_event->file_name = name; + mmap_event->file_size = size; + + mmap_event->event_id.header.size = sizeof(mmap_event->event_id) + size; + + rcu_read_lock(); + cpuctx = &get_cpu_var(perf_cpu_context); + perf_event_mmap_ctx(&cpuctx->ctx, mmap_event); + put_cpu_var(perf_cpu_context); + + /* + * doesn't really matter which of the child contexts the + * events ends up in. + */ + ctx = rcu_dereference(current->perf_event_ctxp); + if (ctx) + perf_event_mmap_ctx(ctx, mmap_event); + rcu_read_unlock(); + + kfree(buf); +} + +void __perf_event_mmap(struct vm_area_struct *vma) +{ + struct perf_mmap_event mmap_event; + + if (!atomic_read(&nr_mmap_events)) + return; + + mmap_event = (struct perf_mmap_event){ + .vma = vma, + /* .file_name */ + /* .file_size */ + .event_id = { + .header = { + .type = PERF_RECORD_MMAP, + .misc = 0, + /* .size */ + }, + /* .pid */ + /* .tid */ + .start = vma->vm_start, + .len = vma->vm_end - vma->vm_start, + .pgoff = vma->vm_pgoff, + }, + }; + + perf_event_mmap_event(&mmap_event); +} + +/* + * IRQ throttle logging + */ + +static void perf_log_throttle(struct perf_event *event, int enable) +{ + struct perf_output_handle handle; + int ret; + + struct { + struct perf_event_header header; + u64 time; + u64 id; + u64 stream_id; + } throttle_event = { + .header = { + .type = PERF_RECORD_THROTTLE, + .misc = 0, + .size = sizeof(throttle_event), + }, + .time = perf_clock(), + .id = primary_event_id(event), + .stream_id = event->id, + }; + + if (enable) + throttle_event.header.type = PERF_RECORD_UNTHROTTLE; + + ret = perf_output_begin(&handle, event, sizeof(throttle_event), 1, 0); + if (ret) + return; + + perf_output_put(&handle, throttle_event); + perf_output_end(&handle); +} + +/* + * Generic event overflow handling, sampling. + */ + +static int __perf_event_overflow(struct perf_event *event, int nmi, + int throttle, struct perf_sample_data *data, + struct pt_regs *regs) +{ + int events = atomic_read(&event->event_limit); + struct hw_perf_event *hwc = &event->hw; + int ret = 0; + + throttle = (throttle && event->pmu->unthrottle != NULL); + + if (!throttle) { + hwc->interrupts++; + } else { + if (hwc->interrupts != MAX_INTERRUPTS) { + hwc->interrupts++; + if (HZ * hwc->interrupts > + (u64)sysctl_perf_event_sample_rate) { + hwc->interrupts = MAX_INTERRUPTS; + perf_log_throttle(event, 0); + ret = 1; + } + } else { + /* + * Keep re-disabling events even though on the previous + * pass we disabled it - just in case we raced with a + * sched-in and the event got enabled again: + */ + ret = 1; + } + } + + if (event->attr.freq) { + u64 now = perf_clock(); + s64 delta = now - hwc->freq_stamp; + + hwc->freq_stamp = now; + + if (delta > 0 && delta < TICK_NSEC) + perf_adjust_period(event, NSEC_PER_SEC / (int)delta); + } + + /* + * XXX event_limit might not quite work as expected on inherited + * events + */ + + event->pending_kill = POLL_IN; + if (events && atomic_dec_and_test(&event->event_limit)) { + ret = 1; + event->pending_kill = POLL_HUP; + if (nmi) { + event->pending_disable = 1; + perf_pending_queue(&event->pending, + perf_pending_event); + } else + perf_event_disable(event); + } + + if (event->overflow_handler) + event->overflow_handler(event, nmi, data, regs); + else + perf_event_output(event, nmi, data, regs); + + return ret; +} + +int perf_event_overflow(struct perf_event *event, int nmi, + struct perf_sample_data *data, + struct pt_regs *regs) +{ + return __perf_event_overflow(event, nmi, 1, data, regs); +} + +/* + * Generic software event infrastructure + */ + +/* + * We directly increment event->count and keep a second value in + * event->hw.period_left to count intervals. This period event + * is kept in the range [-sample_period, 0] so that we can use the + * sign as trigger. + */ + +static u64 perf_swevent_set_period(struct perf_event *event) +{ + struct hw_perf_event *hwc = &event->hw; + u64 period = hwc->last_period; + u64 nr, offset; + s64 old, val; + + hwc->last_period = hwc->sample_period; + +again: + old = val = atomic64_read(&hwc->period_left); + if (val < 0) + return 0; + + nr = div64_u64(period + val, period); + offset = nr * period; + val -= offset; + if (atomic64_cmpxchg(&hwc->period_left, old, val) != old) + goto again; + + return nr; +} + +static void perf_swevent_overflow(struct perf_event *event, u64 overflow, + int nmi, struct perf_sample_data *data, + struct pt_regs *regs) +{ + struct hw_perf_event *hwc = &event->hw; + int throttle = 0; + + data->period = event->hw.last_period; + if (!overflow) + overflow = perf_swevent_set_period(event); + + if (hwc->interrupts == MAX_INTERRUPTS) + return; + + for (; overflow; overflow--) { + if (__perf_event_overflow(event, nmi, throttle, + data, regs)) { + /* + * We inhibit the overflow from happening when + * hwc->interrupts == MAX_INTERRUPTS. + */ + break; + } + throttle = 1; + } +} + +static void perf_swevent_unthrottle(struct perf_event *event) +{ + /* + * Nothing to do, we already reset hwc->interrupts. + */ +} + +static void perf_swevent_add(struct perf_event *event, u64 nr, + int nmi, struct perf_sample_data *data, + struct pt_regs *regs) +{ + struct hw_perf_event *hwc = &event->hw; + + atomic64_add(nr, &event->count); + + if (!regs) + return; + + if (!hwc->sample_period) + return; + + if (nr == 1 && hwc->sample_period == 1 && !event->attr.freq) + return perf_swevent_overflow(event, 1, nmi, data, regs); + + if (atomic64_add_negative(nr, &hwc->period_left)) + return; + + perf_swevent_overflow(event, 0, nmi, data, regs); +} + +static int perf_swevent_is_counting(struct perf_event *event) +{ + /* + * The event is active, we're good! + */ + if (event->state == PERF_EVENT_STATE_ACTIVE) + return 1; + + /* + * The event is off/error, not counting. + */ + if (event->state != PERF_EVENT_STATE_INACTIVE) + return 0; + + /* + * The event is inactive, if the context is active + * we're part of a group that didn't make it on the 'pmu', + * not counting. + */ + if (event->ctx->is_active) + return 0; + + /* + * We're inactive and the context is too, this means the + * task is scheduled out, we're counting events that happen + * to us, like migration events. + */ + return 1; +} + +static int perf_tp_event_match(struct perf_event *event, + struct perf_sample_data *data); + +static int perf_exclude_event(struct perf_event *event, + struct pt_regs *regs) +{ + if (regs) { + if (event->attr.exclude_user && user_mode(regs)) + return 1; + + if (event->attr.exclude_kernel && !user_mode(regs)) + return 1; + } + + return 0; +} + +static int perf_swevent_match(struct perf_event *event, + enum perf_type_id type, + u32 event_id, + struct perf_sample_data *data, + struct pt_regs *regs) +{ + if (!perf_swevent_is_counting(event)) + return 0; + + if (event->attr.type != type) + return 0; + + if (event->attr.config != event_id) + return 0; + + if (perf_exclude_event(event, regs)) + return 0; + + if (event->attr.type == PERF_TYPE_TRACEPOINT && + !perf_tp_event_match(event, data)) + return 0; + + return 1; +} + +static void perf_swevent_ctx_event(struct perf_event_context *ctx, + enum perf_type_id type, + u32 event_id, u64 nr, int nmi, + struct perf_sample_data *data, + struct pt_regs *regs) +{ + struct perf_event *event; + + list_for_each_entry_rcu(event, &ctx->event_list, event_entry) { + if (perf_swevent_match(event, type, event_id, data, regs)) + perf_swevent_add(event, nr, nmi, data, regs); + } +} + +int perf_swevent_get_recursion_context(void) +{ + struct perf_cpu_context *cpuctx = &get_cpu_var(perf_cpu_context); + int rctx; + + if (in_nmi()) + rctx = 3; + else if (in_irq()) + rctx = 2; + else if (in_softirq()) + rctx = 1; + else + rctx = 0; + + if (cpuctx->recursion[rctx]) { + put_cpu_var(perf_cpu_context); + return -1; + } + + cpuctx->recursion[rctx]++; + barrier(); + + return rctx; +} +EXPORT_SYMBOL_GPL(perf_swevent_get_recursion_context); + +void perf_swevent_put_recursion_context(int rctx) +{ + struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context); + barrier(); + cpuctx->recursion[rctx]--; + put_cpu_var(perf_cpu_context); +} +EXPORT_SYMBOL_GPL(perf_swevent_put_recursion_context); + +static void do_perf_sw_event(enum perf_type_id type, u32 event_id, + u64 nr, int nmi, + struct perf_sample_data *data, + struct pt_regs *regs) +{ + struct perf_cpu_context *cpuctx; + struct perf_event_context *ctx; + + cpuctx = &__get_cpu_var(perf_cpu_context); + rcu_read_lock(); + perf_swevent_ctx_event(&cpuctx->ctx, type, event_id, + nr, nmi, data, regs); + /* + * doesn't really matter which of the child contexts the + * events ends up in. + */ + ctx = rcu_dereference(current->perf_event_ctxp); + if (ctx) + perf_swevent_ctx_event(ctx, type, event_id, nr, nmi, data, regs); + rcu_read_unlock(); +} + +void __perf_sw_event(u32 event_id, u64 nr, int nmi, + struct pt_regs *regs, u64 addr) +{ + struct perf_sample_data data; + int rctx; + + rctx = perf_swevent_get_recursion_context(); + if (rctx < 0) + return; + + data.addr = addr; + data.raw = NULL; + + do_perf_sw_event(PERF_TYPE_SOFTWARE, event_id, nr, nmi, &data, regs); + + perf_swevent_put_recursion_context(rctx); +} + +static void perf_swevent_read(struct perf_event *event) +{ +} + +static int perf_swevent_enable(struct perf_event *event) +{ + struct hw_perf_event *hwc = &event->hw; + + if (hwc->sample_period) { + hwc->last_period = hwc->sample_period; + perf_swevent_set_period(event); + } + return 0; +} + +static void perf_swevent_disable(struct perf_event *event) +{ +} + +static const struct pmu perf_ops_generic = { + .enable = perf_swevent_enable, + .disable = perf_swevent_disable, + .read = perf_swevent_read, + .unthrottle = perf_swevent_unthrottle, +}; + +/* + * hrtimer based swevent callback + */ + +static enum hrtimer_restart perf_swevent_hrtimer(struct hrtimer *hrtimer) +{ + enum hrtimer_restart ret = HRTIMER_RESTART; + struct perf_sample_data data; + struct pt_regs *regs; + struct perf_event *event; + u64 period; + + event = container_of(hrtimer, struct perf_event, hw.hrtimer); + event->pmu->read(event); + + data.addr = 0; + data.period = event->hw.last_period; + regs = get_irq_regs(); + /* + * In case we exclude kernel IPs or are somehow not in interrupt + * context, provide the next best thing, the user IP. + */ + if ((event->attr.exclude_kernel || !regs) && + !event->attr.exclude_user) + regs = task_pt_regs(current); + + if (regs) { + if (!(event->attr.exclude_idle && current->pid == 0)) + if (perf_event_overflow(event, 0, &data, regs)) + ret = HRTIMER_NORESTART; + } + + period = max_t(u64, 10000, event->hw.sample_period); + hrtimer_forward_now(hrtimer, ns_to_ktime(period)); + + return ret; +} + +static void perf_swevent_start_hrtimer(struct perf_event *event) +{ + struct hw_perf_event *hwc = &event->hw; + + hrtimer_init(&hwc->hrtimer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); + hwc->hrtimer.function = perf_swevent_hrtimer; + if (hwc->sample_period) { + u64 period; + + if (hwc->remaining) { + if (hwc->remaining < 0) + period = 10000; + else + period = hwc->remaining; + hwc->remaining = 0; + } else { + period = max_t(u64, 10000, hwc->sample_period); + } + __hrtimer_start_range_ns(&hwc->hrtimer, + ns_to_ktime(period), 0, + HRTIMER_MODE_REL, 0); + } +} + +static void perf_swevent_cancel_hrtimer(struct perf_event *event) +{ + struct hw_perf_event *hwc = &event->hw; + + if (hwc->sample_period) { + ktime_t remaining = hrtimer_get_remaining(&hwc->hrtimer); + hwc->remaining = ktime_to_ns(remaining); + + hrtimer_cancel(&hwc->hrtimer); + } +} + +/* + * Software event: cpu wall time clock + */ + +static void cpu_clock_perf_event_update(struct perf_event *event) +{ + int cpu = raw_smp_processor_id(); + s64 prev; + u64 now; + + now = cpu_clock(cpu); + prev = atomic64_read(&event->hw.prev_count); + atomic64_set(&event->hw.prev_count, now); + atomic64_add(now - prev, &event->count); +} + +static int cpu_clock_perf_event_enable(struct perf_event *event) +{ + struct hw_perf_event *hwc = &event->hw; + int cpu = raw_smp_processor_id(); + + atomic64_set(&hwc->prev_count, cpu_clock(cpu)); + perf_swevent_start_hrtimer(event); + + return 0; +} + +static void cpu_clock_perf_event_disable(struct perf_event *event) +{ + perf_swevent_cancel_hrtimer(event); + cpu_clock_perf_event_update(event); +} + +static void cpu_clock_perf_event_read(struct perf_event *event) +{ + cpu_clock_perf_event_update(event); +} + +static const struct pmu perf_ops_cpu_clock = { + .enable = cpu_clock_perf_event_enable, + .disable = cpu_clock_perf_event_disable, + .read = cpu_clock_perf_event_read, +}; + +/* + * Software event: task time clock + */ + +static void task_clock_perf_event_update(struct perf_event *event, u64 now) +{ + u64 prev; + s64 delta; + + prev = atomic64_xchg(&event->hw.prev_count, now); + delta = now - prev; + atomic64_add(delta, &event->count); +} + +static int task_clock_perf_event_enable(struct perf_event *event) +{ + struct hw_perf_event *hwc = &event->hw; + u64 now; + + now = event->ctx->time; + + atomic64_set(&hwc->prev_count, now); + + perf_swevent_start_hrtimer(event); + + return 0; +} + +static void task_clock_perf_event_disable(struct perf_event *event) +{ + perf_swevent_cancel_hrtimer(event); + task_clock_perf_event_update(event, event->ctx->time); + +} + +static void task_clock_perf_event_read(struct perf_event *event) +{ + u64 time; + + if (!in_nmi()) { + update_context_time(event->ctx); + time = event->ctx->time; + } else { + u64 now = perf_clock(); + u64 delta = now - event->ctx->timestamp; + time = event->ctx->time + delta; + } + + task_clock_perf_event_update(event, time); +} + +static const struct pmu perf_ops_task_clock = { + .enable = task_clock_perf_event_enable, + .disable = task_clock_perf_event_disable, + .read = task_clock_perf_event_read, +}; + +#ifdef CONFIG_EVENT_PROFILE + +void perf_tp_event(int event_id, u64 addr, u64 count, void *record, + int entry_size) +{ + struct perf_raw_record raw = { + .size = entry_size, + .data = record, + }; + + struct perf_sample_data data = { + .addr = addr, + .raw = &raw, + }; + + struct pt_regs *regs = get_irq_regs(); + + if (!regs) + regs = task_pt_regs(current); + + /* Trace events already protected against recursion */ + do_perf_sw_event(PERF_TYPE_TRACEPOINT, event_id, count, 1, + &data, regs); +} +EXPORT_SYMBOL_GPL(perf_tp_event); + +static int perf_tp_event_match(struct perf_event *event, + struct perf_sample_data *data) +{ + void *record = data->raw->data; + + if (likely(!event->filter) || filter_match_preds(event->filter, record)) + return 1; + return 0; +} + +static void tp_perf_event_destroy(struct perf_event *event) +{ + ftrace_profile_disable(event->attr.config); +} + +static const struct pmu *tp_perf_event_init(struct perf_event *event) +{ + /* + * Raw tracepoint data is a severe data leak, only allow root to + * have these. + */ + if ((event->attr.sample_type & PERF_SAMPLE_RAW) && + perf_paranoid_tracepoint_raw() && + !capable(CAP_SYS_ADMIN)) + return ERR_PTR(-EPERM); + + if (ftrace_profile_enable(event->attr.config)) + return NULL; + + event->destroy = tp_perf_event_destroy; + + return &perf_ops_generic; +} + +static int perf_event_set_filter(struct perf_event *event, void __user *arg) +{ + char *filter_str; + int ret; + + if (event->attr.type != PERF_TYPE_TRACEPOINT) + return -EINVAL; + + filter_str = strndup_user(arg, PAGE_SIZE); + if (IS_ERR(filter_str)) + return PTR_ERR(filter_str); + + ret = ftrace_profile_set_filter(event, event->attr.config, filter_str); + + kfree(filter_str); + return ret; +} + +static void perf_event_free_filter(struct perf_event *event) +{ + ftrace_profile_free_filter(event); +} + +#else + +static int perf_tp_event_match(struct perf_event *event, + struct perf_sample_data *data) +{ + return 1; +} + +static const struct pmu *tp_perf_event_init(struct perf_event *event) +{ + return NULL; +} + +static int perf_event_set_filter(struct perf_event *event, void __user *arg) +{ + return -ENOENT; +} + +static void perf_event_free_filter(struct perf_event *event) +{ +} + +#endif /* CONFIG_EVENT_PROFILE */ + +#ifdef CONFIG_HAVE_HW_BREAKPOINT +static void bp_perf_event_destroy(struct perf_event *event) +{ + release_bp_slot(event); +} + +static const struct pmu *bp_perf_event_init(struct perf_event *bp) +{ + int err; + /* + * The breakpoint is already filled if we haven't created the counter + * through perf syscall + * FIXME: manage to get trigerred to NULL if it comes from syscalls + */ + if (!bp->callback) + err = register_perf_hw_breakpoint(bp); + else + err = __register_perf_hw_breakpoint(bp); + if (err) + return ERR_PTR(err); + + bp->destroy = bp_perf_event_destroy; + + return &perf_ops_bp; +} + +void perf_bp_event(struct perf_event *bp, void *data) +{ + struct perf_sample_data sample; + struct pt_regs *regs = data; + + sample.addr = bp->attr.bp_addr; + + if (!perf_exclude_event(bp, regs)) + perf_swevent_add(bp, 1, 1, &sample, regs); +} +#else +static const struct pmu *bp_perf_event_init(struct perf_event *bp) +{ + return NULL; +} + +void perf_bp_event(struct perf_event *bp, void *regs) +{ +} +#endif + +atomic_t perf_swevent_enabled[PERF_COUNT_SW_MAX]; + +static void sw_perf_event_destroy(struct perf_event *event) +{ + u64 event_id = event->attr.config; + + WARN_ON(event->parent); + + atomic_dec(&perf_swevent_enabled[event_id]); +} + +static const struct pmu *sw_perf_event_init(struct perf_event *event) +{ + const struct pmu *pmu = NULL; + u64 event_id = event->attr.config; + + /* + * Software events (currently) can't in general distinguish + * between user, kernel and hypervisor events. + * However, context switches and cpu migrations are considered + * to be kernel events, and page faults are never hypervisor + * events. + */ + switch (event_id) { + case PERF_COUNT_SW_CPU_CLOCK: + pmu = &perf_ops_cpu_clock; + + break; + case PERF_COUNT_SW_TASK_CLOCK: + /* + * If the user instantiates this as a per-cpu event, + * use the cpu_clock event instead. + */ + if (event->ctx->task) + pmu = &perf_ops_task_clock; + else + pmu = &perf_ops_cpu_clock; + + break; + case PERF_COUNT_SW_PAGE_FAULTS: + case PERF_COUNT_SW_PAGE_FAULTS_MIN: + case PERF_COUNT_SW_PAGE_FAULTS_MAJ: + case PERF_COUNT_SW_CONTEXT_SWITCHES: + case PERF_COUNT_SW_CPU_MIGRATIONS: + case PERF_COUNT_SW_ALIGNMENT_FAULTS: + case PERF_COUNT_SW_EMULATION_FAULTS: + if (!event->parent) { + atomic_inc(&perf_swevent_enabled[event_id]); + event->destroy = sw_perf_event_destroy; + } + pmu = &perf_ops_generic; + break; + } + + return pmu; +} + +/* + * Allocate and initialize a event structure + */ +static struct perf_event * +perf_event_alloc(struct perf_event_attr *attr, + int cpu, + struct perf_event_context *ctx, + struct perf_event *group_leader, + struct perf_event *parent_event, + perf_callback_t callback, + gfp_t gfpflags) +{ + const struct pmu *pmu; + struct perf_event *event; + struct hw_perf_event *hwc; + long err; + + event = kzalloc(sizeof(*event), gfpflags); + if (!event) + return ERR_PTR(-ENOMEM); + + /* + * Single events are their own group leaders, with an + * empty sibling list: + */ + if (!group_leader) + group_leader = event; + + mutex_init(&event->child_mutex); + INIT_LIST_HEAD(&event->child_list); + + INIT_LIST_HEAD(&event->group_entry); + INIT_LIST_HEAD(&event->event_entry); + INIT_LIST_HEAD(&event->sibling_list); + init_waitqueue_head(&event->waitq); + + mutex_init(&event->mmap_mutex); + + event->cpu = cpu; + event->attr = *attr; + event->group_leader = group_leader; + event->pmu = NULL; + event->ctx = ctx; + event->oncpu = -1; + + event->parent = parent_event; + + event->ns = get_pid_ns(current->nsproxy->pid_ns); + event->id = atomic64_inc_return(&perf_event_id); + + event->state = PERF_EVENT_STATE_INACTIVE; + + if (!callback && parent_event) + callback = parent_event->callback; + + event->callback = callback; + + if (attr->disabled) + event->state = PERF_EVENT_STATE_OFF; + + pmu = NULL; + + hwc = &event->hw; + hwc->sample_period = attr->sample_period; + if (attr->freq && attr->sample_freq) + hwc->sample_period = 1; + hwc->last_period = hwc->sample_period; + + atomic64_set(&hwc->period_left, hwc->sample_period); + + /* + * we currently do not support PERF_FORMAT_GROUP on inherited events + */ + if (attr->inherit && (attr->read_format & PERF_FORMAT_GROUP)) + goto done; + + switch (attr->type) { + case PERF_TYPE_RAW: + case PERF_TYPE_HARDWARE: + case PERF_TYPE_HW_CACHE: + pmu = hw_perf_event_init(event); + break; + + case PERF_TYPE_SOFTWARE: + pmu = sw_perf_event_init(event); + break; + + case PERF_TYPE_TRACEPOINT: + pmu = tp_perf_event_init(event); + break; + + case PERF_TYPE_BREAKPOINT: + pmu = bp_perf_event_init(event); + break; + + + default: + break; + } +done: + err = 0; + if (!pmu) + err = -EINVAL; + else if (IS_ERR(pmu)) + err = PTR_ERR(pmu); + + if (err) { + if (event->ns) + put_pid_ns(event->ns); + kfree(event); + return ERR_PTR(err); + } + + event->pmu = pmu; + + if (!event->parent) { + atomic_inc(&nr_events); + if (event->attr.mmap) + atomic_inc(&nr_mmap_events); + if (event->attr.comm) + atomic_inc(&nr_comm_events); + if (event->attr.task) + atomic_inc(&nr_task_events); + } + + return event; +} + +static int perf_copy_attr(struct perf_event_attr __user *uattr, + struct perf_event_attr *attr) +{ + u32 size; + int ret; + + if (!access_ok(VERIFY_WRITE, uattr, PERF_ATTR_SIZE_VER0)) + return -EFAULT; + + /* + * zero the full structure, so that a short copy will be nice. + */ + memset(attr, 0, sizeof(*attr)); + + ret = get_user(size, &uattr->size); + if (ret) + return ret; + + if (size > PAGE_SIZE) /* silly large */ + goto err_size; + + if (!size) /* abi compat */ + size = PERF_ATTR_SIZE_VER0; + + if (size < PERF_ATTR_SIZE_VER0) + goto err_size; + + /* + * If we're handed a bigger struct than we know of, + * ensure all the unknown bits are 0 - i.e. new + * user-space does not rely on any kernel feature + * extensions we dont know about yet. + */ + if (size > sizeof(*attr)) { + unsigned char __user *addr; + unsigned char __user *end; + unsigned char val; + + addr = (void __user *)uattr + sizeof(*attr); + end = (void __user *)uattr + size; + + for (; addr < end; addr++) { + ret = get_user(val, addr); + if (ret) + return ret; + if (val) + goto err_size; + } + size = sizeof(*attr); + } + + ret = copy_from_user(attr, uattr, size); + if (ret) + return -EFAULT; + + /* + * If the type exists, the corresponding creation will verify + * the attr->config. + */ + if (attr->type >= PERF_TYPE_MAX) + return -EINVAL; + + if (attr->__reserved_1 || attr->__reserved_2 || attr->__reserved_3) + return -EINVAL; + + if (attr->sample_type & ~(PERF_SAMPLE_MAX-1)) + return -EINVAL; + + if (attr->read_format & ~(PERF_FORMAT_MAX-1)) + return -EINVAL; + +out: + return ret; + +err_size: + put_user(sizeof(*attr), &uattr->size); + ret = -E2BIG; + goto out; +} + +static int perf_event_set_output(struct perf_event *event, int output_fd) +{ + struct perf_event *output_event = NULL; + struct file *output_file = NULL; + struct perf_event *old_output; + int fput_needed = 0; + int ret = -EINVAL; + + if (!output_fd) + goto set; + + output_file = fget_light(output_fd, &fput_needed); + if (!output_file) + return -EBADF; + + if (output_file->f_op != &perf_fops) + goto out; + + output_event = output_file->private_data; + + /* Don't chain output fds */ + if (output_event->output) + goto out; + + /* Don't set an output fd when we already have an output channel */ + if (event->data) + goto out; + + atomic_long_inc(&output_file->f_count); + +set: + mutex_lock(&event->mmap_mutex); + old_output = event->output; + rcu_assign_pointer(event->output, output_event); + mutex_unlock(&event->mmap_mutex); + + if (old_output) { + /* + * we need to make sure no existing perf_output_*() + * is still referencing this event. + */ + synchronize_rcu(); + fput(old_output->filp); + } + + ret = 0; +out: + fput_light(output_file, fput_needed); + return ret; +} + +/** + * sys_perf_event_open - open a performance event, associate it to a task/cpu + * + * @attr_uptr: event_id type attributes for monitoring/sampling + * @pid: target pid + * @cpu: target cpu + * @group_fd: group leader event fd + */ +SYSCALL_DEFINE5(perf_event_open, + struct perf_event_attr __user *, attr_uptr, + pid_t, pid, int, cpu, int, group_fd, unsigned long, flags) +{ + struct perf_event *event, *group_leader; + struct perf_event_attr attr; + struct perf_event_context *ctx; + struct file *event_file = NULL; + struct file *group_file = NULL; + int fput_needed = 0; + int fput_needed2 = 0; + int err; + + /* for future expandability... */ + if (flags & ~(PERF_FLAG_FD_NO_GROUP | PERF_FLAG_FD_OUTPUT)) + return -EINVAL; + + err = perf_copy_attr(attr_uptr, &attr); + if (err) + return err; + + if (!attr.exclude_kernel) { + if (perf_paranoid_kernel() && !capable(CAP_SYS_ADMIN)) + return -EACCES; + } + + if (attr.freq) { + if (attr.sample_freq > sysctl_perf_event_sample_rate) + return -EINVAL; + } + + /* + * Get the target context (task or percpu): + */ + ctx = find_get_context(pid, cpu); + if (IS_ERR(ctx)) + return PTR_ERR(ctx); + + /* + * Look up the group leader (we will attach this event to it): + */ + group_leader = NULL; + if (group_fd != -1 && !(flags & PERF_FLAG_FD_NO_GROUP)) { + err = -EINVAL; + group_file = fget_light(group_fd, &fput_needed); + if (!group_file) + goto err_put_context; + if (group_file->f_op != &perf_fops) + goto err_put_context; + + group_leader = group_file->private_data; + /* + * Do not allow a recursive hierarchy (this new sibling + * becoming part of another group-sibling): + */ + if (group_leader->group_leader != group_leader) + goto err_put_context; + /* + * Do not allow to attach to a group in a different + * task or CPU context: + */ + if (group_leader->ctx != ctx) + goto err_put_context; + /* + * Only a group leader can be exclusive or pinned + */ + if (attr.exclusive || attr.pinned) + goto err_put_context; + } + + event = perf_event_alloc(&attr, cpu, ctx, group_leader, + NULL, NULL, GFP_KERNEL); + err = PTR_ERR(event); + if (IS_ERR(event)) + goto err_put_context; + + err = anon_inode_getfd("[perf_event]", &perf_fops, event, 0); + if (err < 0) + goto err_free_put_context; + + event_file = fget_light(err, &fput_needed2); + if (!event_file) + goto err_free_put_context; + + if (flags & PERF_FLAG_FD_OUTPUT) { + err = perf_event_set_output(event, group_fd); + if (err) + goto err_fput_free_put_context; + } + + event->filp = event_file; + WARN_ON_ONCE(ctx->parent_ctx); + mutex_lock(&ctx->mutex); + perf_install_in_context(ctx, event, cpu); + ++ctx->generation; + mutex_unlock(&ctx->mutex); + + event->owner = current; + get_task_struct(current); + mutex_lock(¤t->perf_event_mutex); + list_add_tail(&event->owner_entry, ¤t->perf_event_list); + mutex_unlock(¤t->perf_event_mutex); + +err_fput_free_put_context: + fput_light(event_file, fput_needed2); + +err_free_put_context: + if (err < 0) + kfree(event); + +err_put_context: + if (err < 0) + put_ctx(ctx); + + fput_light(group_file, fput_needed); + + return err; +} + +/** + * perf_event_create_kernel_counter + * + * @attr: attributes of the counter to create + * @cpu: cpu in which the counter is bound + * @pid: task to profile + */ +struct perf_event * +perf_event_create_kernel_counter(struct perf_event_attr *attr, int cpu, + pid_t pid, perf_callback_t callback) +{ + struct perf_event *event; + struct perf_event_context *ctx; + int err; + + /* + * Get the target context (task or percpu): + */ + + ctx = find_get_context(pid, cpu); + if (IS_ERR(ctx)) { + err = PTR_ERR(ctx); + goto err_exit; + } + + event = perf_event_alloc(attr, cpu, ctx, NULL, + NULL, callback, GFP_KERNEL); + if (IS_ERR(event)) { + err = PTR_ERR(event); + goto err_put_context; + } + + event->filp = NULL; + WARN_ON_ONCE(ctx->parent_ctx); + mutex_lock(&ctx->mutex); + perf_install_in_context(ctx, event, cpu); + ++ctx->generation; + mutex_unlock(&ctx->mutex); + + event->owner = current; + get_task_struct(current); + mutex_lock(¤t->perf_event_mutex); + list_add_tail(&event->owner_entry, ¤t->perf_event_list); + mutex_unlock(¤t->perf_event_mutex); + + return event; + + err_put_context: + put_ctx(ctx); + err_exit: + return ERR_PTR(err); +} +EXPORT_SYMBOL_GPL(perf_event_create_kernel_counter); + +/* + * inherit a event from parent task to child task: + */ +static struct perf_event * +inherit_event(struct perf_event *parent_event, + struct task_struct *parent, + struct perf_event_context *parent_ctx, + struct task_struct *child, + struct perf_event *group_leader, + struct perf_event_context *child_ctx) +{ + struct perf_event *child_event; + + /* + * Instead of creating recursive hierarchies of events, + * we link inherited events back to the original parent, + * which has a filp for sure, which we use as the reference + * count: + */ + if (parent_event->parent) + parent_event = parent_event->parent; + + child_event = perf_event_alloc(&parent_event->attr, + parent_event->cpu, child_ctx, + group_leader, parent_event, + NULL, GFP_KERNEL); + if (IS_ERR(child_event)) + return child_event; + get_ctx(child_ctx); + + /* + * Make the child state follow the state of the parent event, + * not its attr.disabled bit. We hold the parent's mutex, + * so we won't race with perf_event_{en, dis}able_family. + */ + if (parent_event->state >= PERF_EVENT_STATE_INACTIVE) + child_event->state = PERF_EVENT_STATE_INACTIVE; + else + child_event->state = PERF_EVENT_STATE_OFF; + + if (parent_event->attr.freq) + child_event->hw.sample_period = parent_event->hw.sample_period; + + child_event->overflow_handler = parent_event->overflow_handler; + + /* + * Link it up in the child's context: + */ + add_event_to_ctx(child_event, child_ctx); + + /* + * Get a reference to the parent filp - we will fput it + * when the child event exits. This is safe to do because + * we are in the parent and we know that the filp still + * exists and has a nonzero count: + */ + atomic_long_inc(&parent_event->filp->f_count); + + /* + * Link this into the parent event's child list + */ + WARN_ON_ONCE(parent_event->ctx->parent_ctx); + mutex_lock(&parent_event->child_mutex); + list_add_tail(&child_event->child_list, &parent_event->child_list); + mutex_unlock(&parent_event->child_mutex); + + return child_event; +} + +static int inherit_group(struct perf_event *parent_event, + struct task_struct *parent, + struct perf_event_context *parent_ctx, + struct task_struct *child, + struct perf_event_context *child_ctx) +{ + struct perf_event *leader; + struct perf_event *sub; + struct perf_event *child_ctr; + + leader = inherit_event(parent_event, parent, parent_ctx, + child, NULL, child_ctx); + if (IS_ERR(leader)) + return PTR_ERR(leader); + list_for_each_entry(sub, &parent_event->sibling_list, group_entry) { + child_ctr = inherit_event(sub, parent, parent_ctx, + child, leader, child_ctx); + if (IS_ERR(child_ctr)) + return PTR_ERR(child_ctr); + } + return 0; +} + +static void sync_child_event(struct perf_event *child_event, + struct task_struct *child) +{ + struct perf_event *parent_event = child_event->parent; + u64 child_val; + + if (child_event->attr.inherit_stat) + perf_event_read_event(child_event, child); + + child_val = atomic64_read(&child_event->count); + + /* + * Add back the child's count to the parent's count: + */ + atomic64_add(child_val, &parent_event->count); + atomic64_add(child_event->total_time_enabled, + &parent_event->child_total_time_enabled); + atomic64_add(child_event->total_time_running, + &parent_event->child_total_time_running); + + /* + * Remove this event from the parent's list + */ + WARN_ON_ONCE(parent_event->ctx->parent_ctx); + mutex_lock(&parent_event->child_mutex); + list_del_init(&child_event->child_list); + mutex_unlock(&parent_event->child_mutex); + + /* + * Release the parent event, if this was the last + * reference to it. + */ + fput(parent_event->filp); +} + +static void +__perf_event_exit_task(struct perf_event *child_event, + struct perf_event_context *child_ctx, + struct task_struct *child) +{ + struct perf_event *parent_event; + + perf_event_remove_from_context(child_event); + + parent_event = child_event->parent; + /* + * It can happen that parent exits first, and has events + * that are still around due to the child reference. These + * events need to be zapped - but otherwise linger. + */ + if (parent_event) { + sync_child_event(child_event, child); + free_event(child_event); + } +} + +/* + * When a child task exits, feed back event values to parent events. + */ +void perf_event_exit_task(struct task_struct *child) +{ + struct perf_event *child_event, *tmp; + struct perf_event_context *child_ctx; + unsigned long flags; + + if (likely(!child->perf_event_ctxp)) { + perf_event_task(child, NULL, 0); + return; + } + + local_irq_save(flags); + /* + * We can't reschedule here because interrupts are disabled, + * and either child is current or it is a task that can't be + * scheduled, so we are now safe from rescheduling changing + * our context. + */ + child_ctx = child->perf_event_ctxp; + __perf_event_task_sched_out(child_ctx); + + /* + * Take the context lock here so that if find_get_context is + * reading child->perf_event_ctxp, we wait until it has + * incremented the context's refcount before we do put_ctx below. + */ + spin_lock(&child_ctx->lock); + child->perf_event_ctxp = NULL; + /* + * If this context is a clone; unclone it so it can't get + * swapped to another process while we're removing all + * the events from it. + */ + unclone_ctx(child_ctx); + update_context_time(child_ctx); + spin_unlock_irqrestore(&child_ctx->lock, flags); + + /* + * Report the task dead after unscheduling the events so that we + * won't get any samples after PERF_RECORD_EXIT. We can however still + * get a few PERF_RECORD_READ events. + */ + perf_event_task(child, child_ctx, 0); + + /* + * We can recurse on the same lock type through: + * + * __perf_event_exit_task() + * sync_child_event() + * fput(parent_event->filp) + * perf_release() + * mutex_lock(&ctx->mutex) + * + * But since its the parent context it won't be the same instance. + */ + mutex_lock_nested(&child_ctx->mutex, SINGLE_DEPTH_NESTING); + +again: + list_for_each_entry_safe(child_event, tmp, &child_ctx->group_list, + group_entry) + __perf_event_exit_task(child_event, child_ctx, child); + + /* + * If the last event was a group event, it will have appended all + * its siblings to the list, but we obtained 'tmp' before that which + * will still point to the list head terminating the iteration. + */ + if (!list_empty(&child_ctx->group_list)) + goto again; + + mutex_unlock(&child_ctx->mutex); + + put_ctx(child_ctx); +} + +/* + * free an unexposed, unused context as created by inheritance by + * init_task below, used by fork() in case of fail. + */ +void perf_event_free_task(struct task_struct *task) +{ + struct perf_event_context *ctx = task->perf_event_ctxp; + struct perf_event *event, *tmp; + + if (!ctx) + return; + + mutex_lock(&ctx->mutex); +again: + list_for_each_entry_safe(event, tmp, &ctx->group_list, group_entry) { + struct perf_event *parent = event->parent; + + if (WARN_ON_ONCE(!parent)) + continue; + + mutex_lock(&parent->child_mutex); + list_del_init(&event->child_list); + mutex_unlock(&parent->child_mutex); + + fput(parent->filp); + + list_del_event(event, ctx); + free_event(event); + } + + if (!list_empty(&ctx->group_list)) + goto again; + + mutex_unlock(&ctx->mutex); + + put_ctx(ctx); +} + +/* + * Initialize the perf_event context in task_struct + */ +int perf_event_init_task(struct task_struct *child) +{ + struct perf_event_context *child_ctx, *parent_ctx; + struct perf_event_context *cloned_ctx; + struct perf_event *event; + struct task_struct *parent = current; + int inherited_all = 1; + int ret = 0; + + child->perf_event_ctxp = NULL; + + mutex_init(&child->perf_event_mutex); + INIT_LIST_HEAD(&child->perf_event_list); + + if (likely(!parent->perf_event_ctxp)) + return 0; + + /* + * This is executed from the parent task context, so inherit + * events that have been marked for cloning. + * First allocate and initialize a context for the child. + */ + + child_ctx = kmalloc(sizeof(struct perf_event_context), GFP_KERNEL); + if (!child_ctx) + return -ENOMEM; + + __perf_event_init_context(child_ctx, child); + child->perf_event_ctxp = child_ctx; + get_task_struct(child); + + /* + * If the parent's context is a clone, pin it so it won't get + * swapped under us. + */ + parent_ctx = perf_pin_task_context(parent); + + /* + * No need to check if parent_ctx != NULL here; since we saw + * it non-NULL earlier, the only reason for it to become NULL + * is if we exit, and since we're currently in the middle of + * a fork we can't be exiting at the same time. + */ + + /* + * Lock the parent list. No need to lock the child - not PID + * hashed yet and not running, so nobody can access it. + */ + mutex_lock(&parent_ctx->mutex); + + /* + * We dont have to disable NMIs - we are only looking at + * the list, not manipulating it: + */ + list_for_each_entry(event, &parent_ctx->group_list, group_entry) { + + if (!event->attr.inherit) { + inherited_all = 0; + continue; + } + + ret = inherit_group(event, parent, parent_ctx, + child, child_ctx); + if (ret) { + inherited_all = 0; + break; + } + } + + if (inherited_all) { + /* + * Mark the child context as a clone of the parent + * context, or of whatever the parent is a clone of. + * Note that if the parent is a clone, it could get + * uncloned at any point, but that doesn't matter + * because the list of events and the generation + * count can't have changed since we took the mutex. + */ + cloned_ctx = rcu_dereference(parent_ctx->parent_ctx); + if (cloned_ctx) { + child_ctx->parent_ctx = cloned_ctx; + child_ctx->parent_gen = parent_ctx->parent_gen; + } else { + child_ctx->parent_ctx = parent_ctx; + child_ctx->parent_gen = parent_ctx->generation; + } + get_ctx(child_ctx->parent_ctx); + } + + mutex_unlock(&parent_ctx->mutex); + + perf_unpin_context(parent_ctx); + + return ret; +} + +static void __cpuinit perf_event_init_cpu(int cpu) +{ + struct perf_cpu_context *cpuctx; + + cpuctx = &per_cpu(perf_cpu_context, cpu); + __perf_event_init_context(&cpuctx->ctx, NULL); + + spin_lock(&perf_resource_lock); + cpuctx->max_pertask = perf_max_events - perf_reserved_percpu; + spin_unlock(&perf_resource_lock); + + hw_perf_event_setup(cpu); +} + +#ifdef CONFIG_HOTPLUG_CPU +static void __perf_event_exit_cpu(void *info) +{ + struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context); + struct perf_event_context *ctx = &cpuctx->ctx; + struct perf_event *event, *tmp; + + list_for_each_entry_safe(event, tmp, &ctx->group_list, group_entry) + __perf_event_remove_from_context(event); +} +static void perf_event_exit_cpu(int cpu) +{ + struct perf_cpu_context *cpuctx = &per_cpu(perf_cpu_context, cpu); + struct perf_event_context *ctx = &cpuctx->ctx; + + mutex_lock(&ctx->mutex); + smp_call_function_single(cpu, __perf_event_exit_cpu, NULL, 1); + mutex_unlock(&ctx->mutex); +} +#else +static inline void perf_event_exit_cpu(int cpu) { } +#endif + +static int __cpuinit +perf_cpu_notify(struct notifier_block *self, unsigned long action, void *hcpu) +{ + unsigned int cpu = (long)hcpu; + + switch (action) { + + case CPU_UP_PREPARE: + case CPU_UP_PREPARE_FROZEN: + perf_event_init_cpu(cpu); + break; + + case CPU_ONLINE: + case CPU_ONLINE_FROZEN: + hw_perf_event_setup_online(cpu); + break; + + case CPU_DOWN_PREPARE: + case CPU_DOWN_PREPARE_FROZEN: + perf_event_exit_cpu(cpu); + break; + + default: + break; + } + + return NOTIFY_OK; +} + +/* + * This has to have a higher priority than migration_notifier in sched.c. + */ +static struct notifier_block __cpuinitdata perf_cpu_nb = { + .notifier_call = perf_cpu_notify, + .priority = 20, +}; + +void __init perf_event_init(void) +{ + perf_cpu_notify(&perf_cpu_nb, (unsigned long)CPU_UP_PREPARE, + (void *)(long)smp_processor_id()); + perf_cpu_notify(&perf_cpu_nb, (unsigned long)CPU_ONLINE, + (void *)(long)smp_processor_id()); + register_cpu_notifier(&perf_cpu_nb); +} + +static ssize_t perf_show_reserve_percpu(struct sysdev_class *class, char *buf) +{ + return sprintf(buf, "%d\n", perf_reserved_percpu); +} + +static ssize_t +perf_set_reserve_percpu(struct sysdev_class *class, + const char *buf, + size_t count) +{ + struct perf_cpu_context *cpuctx; + unsigned long val; + int err, cpu, mpt; + + err = strict_strtoul(buf, 10, &val); + if (err) + return err; + if (val > perf_max_events) + return -EINVAL; + + spin_lock(&perf_resource_lock); + perf_reserved_percpu = val; + for_each_online_cpu(cpu) { + cpuctx = &per_cpu(perf_cpu_context, cpu); + spin_lock_irq(&cpuctx->ctx.lock); + mpt = min(perf_max_events - cpuctx->ctx.nr_events, + perf_max_events - perf_reserved_percpu); + cpuctx->max_pertask = mpt; + spin_unlock_irq(&cpuctx->ctx.lock); + } + spin_unlock(&perf_resource_lock); + + return count; +} + +static ssize_t perf_show_overcommit(struct sysdev_class *class, char *buf) +{ + return sprintf(buf, "%d\n", perf_overcommit); +} + +static ssize_t +perf_set_overcommit(struct sysdev_class *class, const char *buf, size_t count) +{ + unsigned long val; + int err; + + err = strict_strtoul(buf, 10, &val); + if (err) + return err; + if (val > 1) + return -EINVAL; + + spin_lock(&perf_resource_lock); + perf_overcommit = val; + spin_unlock(&perf_resource_lock); + + return count; +} + +static SYSDEV_CLASS_ATTR( + reserve_percpu, + 0644, + perf_show_reserve_percpu, + perf_set_reserve_percpu + ); + +static SYSDEV_CLASS_ATTR( + overcommit, + 0644, + perf_show_overcommit, + perf_set_overcommit + ); + +static struct attribute *perfclass_attrs[] = { + &attr_reserve_percpu.attr, + &attr_overcommit.attr, + NULL +}; + +static struct attribute_group perfclass_attr_group = { + .attrs = perfclass_attrs, + .name = "perf_events", +}; + +static int __init perf_event_sysfs_init(void) +{ + return sysfs_create_group(&cpu_sysdev_class.kset.kobj, + &perfclass_attr_group); +} +device_initcall(perf_event_sysfs_init); diff --git a/kernel/pid.c b/kernel/pid.c index 1b3586fe753..d3f722d20f9 100644 --- a/kernel/pid.c +++ b/kernel/pid.c @@ -40,7 +40,7 @@ #define pid_hashfn(nr, ns) \ hash_long((unsigned long)nr + (unsigned long)ns, pidhash_shift) static struct hlist_head *pid_hash; -static int pidhash_shift; +static unsigned int pidhash_shift = 4; struct pid init_struct_pid = INIT_STRUCT_PID; int pid_max = PID_MAX_DEFAULT; @@ -378,31 +378,22 @@ EXPORT_SYMBOL(pid_task); /* * Must be called under rcu_read_lock() or with tasklist_lock read-held. */ -struct task_struct *find_task_by_pid_type_ns(int type, int nr, - struct pid_namespace *ns) +struct task_struct *find_task_by_pid_ns(pid_t nr, struct pid_namespace *ns) { - return pid_task(find_pid_ns(nr, ns), type); + return pid_task(find_pid_ns(nr, ns), PIDTYPE_PID); } -EXPORT_SYMBOL(find_task_by_pid_type_ns); - struct task_struct *find_task_by_vpid(pid_t vnr) { - return find_task_by_pid_type_ns(PIDTYPE_PID, vnr, - current->nsproxy->pid_ns); -} -EXPORT_SYMBOL(find_task_by_vpid); - -struct task_struct *find_task_by_pid_ns(pid_t nr, struct pid_namespace *ns) -{ - return find_task_by_pid_type_ns(PIDTYPE_PID, nr, ns); + return find_task_by_pid_ns(vnr, current->nsproxy->pid_ns); } -EXPORT_SYMBOL(find_task_by_pid_ns); struct pid *get_task_pid(struct task_struct *task, enum pid_type type) { struct pid *pid; rcu_read_lock(); + if (type != PIDTYPE_PID) + task = task->group_leader; pid = get_pid(task->pids[type].pid); rcu_read_unlock(); return pid; @@ -450,11 +441,24 @@ pid_t pid_vnr(struct pid *pid) } EXPORT_SYMBOL_GPL(pid_vnr); -pid_t task_pid_nr_ns(struct task_struct *tsk, struct pid_namespace *ns) +pid_t __task_pid_nr_ns(struct task_struct *task, enum pid_type type, + struct pid_namespace *ns) { - return pid_nr_ns(task_pid(tsk), ns); + pid_t nr = 0; + + rcu_read_lock(); + if (!ns) + ns = current->nsproxy->pid_ns; + if (likely(pid_alive(task))) { + if (type != PIDTYPE_PID) + task = task->group_leader; + nr = pid_nr_ns(task->pids[type].pid, ns); + } + rcu_read_unlock(); + + return nr; } -EXPORT_SYMBOL(task_pid_nr_ns); +EXPORT_SYMBOL(__task_pid_nr_ns); pid_t task_tgid_nr_ns(struct task_struct *tsk, struct pid_namespace *ns) { @@ -462,18 +466,6 @@ pid_t task_tgid_nr_ns(struct task_struct *tsk, struct pid_namespace *ns) } EXPORT_SYMBOL(task_tgid_nr_ns); -pid_t task_pgrp_nr_ns(struct task_struct *tsk, struct pid_namespace *ns) -{ - return pid_nr_ns(task_pgrp(tsk), ns); -} -EXPORT_SYMBOL(task_pgrp_nr_ns); - -pid_t task_session_nr_ns(struct task_struct *tsk, struct pid_namespace *ns) -{ - return pid_nr_ns(task_session(tsk), ns); -} -EXPORT_SYMBOL(task_session_nr_ns); - struct pid_namespace *task_active_pid_ns(struct task_struct *tsk) { return ns_of_pid(task_pid(tsk)); @@ -507,19 +499,12 @@ struct pid *find_ge_pid(int nr, struct pid_namespace *ns) void __init pidhash_init(void) { int i, pidhash_size; - unsigned long megabytes = nr_kernel_pages >> (20 - PAGE_SHIFT); - pidhash_shift = max(4, fls(megabytes * 4)); - pidhash_shift = min(12, pidhash_shift); + pid_hash = alloc_large_system_hash("PID", sizeof(*pid_hash), 0, 18, + HASH_EARLY | HASH_SMALL, + &pidhash_shift, NULL, 4096); pidhash_size = 1 << pidhash_shift; - printk("PID hash table entries: %d (order: %d, %Zd bytes)\n", - pidhash_size, pidhash_shift, - pidhash_size * sizeof(struct hlist_head)); - - pid_hash = alloc_bootmem(pidhash_size * sizeof(*(pid_hash))); - if (!pid_hash) - panic("Could not alloc pidhash!\n"); for (i = 0; i < pidhash_size; i++) INIT_HLIST_HEAD(&pid_hash[i]); } diff --git a/kernel/pid_namespace.c b/kernel/pid_namespace.c index fab8ea86fac..86b3796b043 100644 --- a/kernel/pid_namespace.c +++ b/kernel/pid_namespace.c @@ -67,9 +67,10 @@ err_alloc: return NULL; } -static struct pid_namespace *create_pid_namespace(unsigned int level) +static struct pid_namespace *create_pid_namespace(struct pid_namespace *parent_pid_ns) { struct pid_namespace *ns; + unsigned int level = parent_pid_ns->level + 1; int i; ns = kmem_cache_zalloc(pid_ns_cachep, GFP_KERNEL); @@ -86,6 +87,7 @@ static struct pid_namespace *create_pid_namespace(unsigned int level) kref_init(&ns->kref); ns->level = level; + ns->parent = get_pid_ns(parent_pid_ns); set_bit(0, ns->pidmap[0].page); atomic_set(&ns->pidmap[0].nr_free, BITS_PER_PAGE - 1); @@ -114,25 +116,11 @@ static void destroy_pid_namespace(struct pid_namespace *ns) struct pid_namespace *copy_pid_ns(unsigned long flags, struct pid_namespace *old_ns) { - struct pid_namespace *new_ns; - - BUG_ON(!old_ns); - new_ns = get_pid_ns(old_ns); if (!(flags & CLONE_NEWPID)) - goto out; - - new_ns = ERR_PTR(-EINVAL); - if (flags & CLONE_THREAD) - goto out_put; - - new_ns = create_pid_namespace(old_ns->level + 1); - if (!IS_ERR(new_ns)) - new_ns->parent = get_pid_ns(old_ns); - -out_put: - put_pid_ns(old_ns); -out: - return new_ns; + return get_pid_ns(old_ns); + if (flags & (CLONE_THREAD|CLONE_PARENT)) + return ERR_PTR(-EINVAL); + return create_pid_namespace(old_ns); } void free_pid_ns(struct kref *kref) @@ -152,6 +140,7 @@ void zap_pid_ns_processes(struct pid_namespace *pid_ns) { int nr; int rc; + struct task_struct *task; /* * The last thread in the cgroup-init thread group is terminating. @@ -169,7 +158,19 @@ void zap_pid_ns_processes(struct pid_namespace *pid_ns) read_lock(&tasklist_lock); nr = next_pidmap(pid_ns, 1); while (nr > 0) { - kill_proc_info(SIGKILL, SEND_SIG_PRIV, nr); + rcu_read_lock(); + + /* + * Use force_sig() since it clears SIGNAL_UNKILLABLE ensuring + * any nested-container's init processes don't ignore the + * signal + */ + task = pid_task(find_vpid(nr), PIDTYPE_PID); + if (task) + force_sig(SIGKILL, task); + + rcu_read_unlock(); + nr = next_pidmap(pid_ns, nr); } read_unlock(&tasklist_lock); diff --git a/kernel/posix-cpu-timers.c b/kernel/posix-cpu-timers.c index 8e5d9a68b02..438ff452351 100644 --- a/kernel/posix-cpu-timers.c +++ b/kernel/posix-cpu-timers.c @@ -8,17 +8,18 @@ #include <linux/math64.h> #include <asm/uaccess.h> #include <linux/kernel_stat.h> +#include <trace/events/timer.h> /* * Called after updating RLIMIT_CPU to set timer expiration if necessary. */ void update_rlimit_cpu(unsigned long rlim_new) { - cputime_t cputime; + cputime_t cputime = secs_to_cputime(rlim_new); + struct signal_struct *const sig = current->signal; - cputime = secs_to_cputime(rlim_new); - if (cputime_eq(current->signal->it_prof_expires, cputime_zero) || - cputime_lt(current->signal->it_prof_expires, cputime)) { + if (cputime_eq(sig->it[CPUCLOCK_PROF].expires, cputime_zero) || + cputime_gt(sig->it[CPUCLOCK_PROF].expires, cputime)) { spin_lock_irq(¤t->sighand->siglock); set_process_cpu_timer(current, CPUCLOCK_PROF, &cputime, NULL); spin_unlock_irq(¤t->sighand->siglock); @@ -224,7 +225,7 @@ static int cpu_clock_sample(const clockid_t which_clock, struct task_struct *p, cpu->cpu = virt_ticks(p); break; case CPUCLOCK_SCHED: - cpu->sched = p->se.sum_exec_runtime + task_delta_exec(p); + cpu->sched = task_sched_runtime(p); break; } return 0; @@ -305,18 +306,19 @@ static int cpu_clock_sample_group(const clockid_t which_clock, { struct task_cputime cputime; - thread_group_cputime(p, &cputime); switch (CPUCLOCK_WHICH(which_clock)) { default: return -EINVAL; case CPUCLOCK_PROF: + thread_group_cputime(p, &cputime); cpu->cpu = cputime_add(cputime.utime, cputime.stime); break; case CPUCLOCK_VIRT: + thread_group_cputime(p, &cputime); cpu->cpu = cputime.utime; break; case CPUCLOCK_SCHED: - cpu->sched = cputime.sum_exec_runtime + task_delta_exec(p); + cpu->sched = thread_group_sched_runtime(p); break; } return 0; @@ -382,7 +384,8 @@ int posix_cpu_clock_get(const clockid_t which_clock, struct timespec *tp) /* * Validate the clockid_t for a new CPU-clock timer, and initialize the timer. - * This is called from sys_timer_create with the new timer already locked. + * This is called from sys_timer_create() and do_cpu_nanosleep() with the + * new timer already all-zeros initialized. */ int posix_cpu_timer_create(struct k_itimer *new_timer) { @@ -394,8 +397,6 @@ int posix_cpu_timer_create(struct k_itimer *new_timer) return -EINVAL; INIT_LIST_HEAD(&new_timer->it.cpu.entry); - new_timer->it.cpu.incr.sched = 0; - new_timer->it.cpu.expires.sched = 0; read_lock(&tasklist_lock); if (CPUCLOCK_PERTHREAD(new_timer->it_clock)) { @@ -520,11 +521,12 @@ void posix_cpu_timers_exit(struct task_struct *tsk) } void posix_cpu_timers_exit_group(struct task_struct *tsk) { - struct task_cputime cputime; + struct signal_struct *const sig = tsk->signal; - thread_group_cputimer(tsk, &cputime); cleanup_timers(tsk->signal->cpu_timers, - cputime.utime, cputime.stime, cputime.sum_exec_runtime); + cputime_add(tsk->utime, sig->utime), + cputime_add(tsk->stime, sig->stime), + tsk->se.sum_exec_runtime + sig->sum_sched_runtime); } static void clear_dead_task(struct k_itimer *timer, union cpu_time_count now) @@ -540,6 +542,17 @@ static void clear_dead_task(struct k_itimer *timer, union cpu_time_count now) now); } +static inline int expires_gt(cputime_t expires, cputime_t new_exp) +{ + return cputime_eq(expires, cputime_zero) || + cputime_gt(expires, new_exp); +} + +static inline int expires_le(cputime_t expires, cputime_t new_exp) +{ + return !cputime_eq(expires, cputime_zero) && + cputime_le(expires, new_exp); +} /* * Insert the timer on the appropriate list before any timers that * expire later. This must be called with the tasklist_lock held @@ -584,34 +597,32 @@ static void arm_timer(struct k_itimer *timer, union cpu_time_count now) */ if (CPUCLOCK_PERTHREAD(timer->it_clock)) { + union cpu_time_count *exp = &nt->expires; + switch (CPUCLOCK_WHICH(timer->it_clock)) { default: BUG(); case CPUCLOCK_PROF: - if (cputime_eq(p->cputime_expires.prof_exp, - cputime_zero) || - cputime_gt(p->cputime_expires.prof_exp, - nt->expires.cpu)) - p->cputime_expires.prof_exp = - nt->expires.cpu; + if (expires_gt(p->cputime_expires.prof_exp, + exp->cpu)) + p->cputime_expires.prof_exp = exp->cpu; break; case CPUCLOCK_VIRT: - if (cputime_eq(p->cputime_expires.virt_exp, - cputime_zero) || - cputime_gt(p->cputime_expires.virt_exp, - nt->expires.cpu)) - p->cputime_expires.virt_exp = - nt->expires.cpu; + if (expires_gt(p->cputime_expires.virt_exp, + exp->cpu)) + p->cputime_expires.virt_exp = exp->cpu; break; case CPUCLOCK_SCHED: if (p->cputime_expires.sched_exp == 0 || - p->cputime_expires.sched_exp > - nt->expires.sched) + p->cputime_expires.sched_exp > exp->sched) p->cputime_expires.sched_exp = - nt->expires.sched; + exp->sched; break; } } else { + struct signal_struct *const sig = p->signal; + union cpu_time_count *exp = &timer->it.cpu.expires; + /* * For a process timer, set the cached expiration time. */ @@ -619,30 +630,23 @@ static void arm_timer(struct k_itimer *timer, union cpu_time_count now) default: BUG(); case CPUCLOCK_VIRT: - if (!cputime_eq(p->signal->it_virt_expires, - cputime_zero) && - cputime_lt(p->signal->it_virt_expires, - timer->it.cpu.expires.cpu)) + if (expires_le(sig->it[CPUCLOCK_VIRT].expires, + exp->cpu)) break; - p->signal->cputime_expires.virt_exp = - timer->it.cpu.expires.cpu; + sig->cputime_expires.virt_exp = exp->cpu; break; case CPUCLOCK_PROF: - if (!cputime_eq(p->signal->it_prof_expires, - cputime_zero) && - cputime_lt(p->signal->it_prof_expires, - timer->it.cpu.expires.cpu)) + if (expires_le(sig->it[CPUCLOCK_PROF].expires, + exp->cpu)) break; - i = p->signal->rlim[RLIMIT_CPU].rlim_cur; + i = sig->rlim[RLIMIT_CPU].rlim_cur; if (i != RLIM_INFINITY && - i <= cputime_to_secs(timer->it.cpu.expires.cpu)) + i <= cputime_to_secs(exp->cpu)) break; - p->signal->cputime_expires.prof_exp = - timer->it.cpu.expires.cpu; + sig->cputime_expires.prof_exp = exp->cpu; break; case CPUCLOCK_SCHED: - p->signal->cputime_expires.sched_exp = - timer->it.cpu.expires.sched; + sig->cputime_expires.sched_exp = exp->sched; break; } } @@ -1069,6 +1073,40 @@ static void stop_process_timers(struct task_struct *tsk) spin_unlock_irqrestore(&cputimer->lock, flags); } +static u32 onecputick; + +static void check_cpu_itimer(struct task_struct *tsk, struct cpu_itimer *it, + cputime_t *expires, cputime_t cur_time, int signo) +{ + if (cputime_eq(it->expires, cputime_zero)) + return; + + if (cputime_ge(cur_time, it->expires)) { + if (!cputime_eq(it->incr, cputime_zero)) { + it->expires = cputime_add(it->expires, it->incr); + it->error += it->incr_error; + if (it->error >= onecputick) { + it->expires = cputime_sub(it->expires, + cputime_one_jiffy); + it->error -= onecputick; + } + } else { + it->expires = cputime_zero; + } + + trace_itimer_expire(signo == SIGPROF ? + ITIMER_PROF : ITIMER_VIRTUAL, + tsk->signal->leader_pid, cur_time); + __group_send_sig_info(signo, SEND_SIG_PRIV, tsk); + } + + if (!cputime_eq(it->expires, cputime_zero) && + (cputime_eq(*expires, cputime_zero) || + cputime_lt(it->expires, *expires))) { + *expires = it->expires; + } +} + /* * Check for any per-thread CPU timers that have fired and move them * off the tsk->*_timers list onto the firing list. Per-thread timers @@ -1088,10 +1126,10 @@ static void check_process_timers(struct task_struct *tsk, * Don't sample the current process CPU clocks if there are no timers. */ if (list_empty(&timers[CPUCLOCK_PROF]) && - cputime_eq(sig->it_prof_expires, cputime_zero) && + cputime_eq(sig->it[CPUCLOCK_PROF].expires, cputime_zero) && sig->rlim[RLIMIT_CPU].rlim_cur == RLIM_INFINITY && list_empty(&timers[CPUCLOCK_VIRT]) && - cputime_eq(sig->it_virt_expires, cputime_zero) && + cputime_eq(sig->it[CPUCLOCK_VIRT].expires, cputime_zero) && list_empty(&timers[CPUCLOCK_SCHED])) { stop_process_timers(tsk); return; @@ -1151,38 +1189,11 @@ static void check_process_timers(struct task_struct *tsk, /* * Check for the special case process timers. */ - if (!cputime_eq(sig->it_prof_expires, cputime_zero)) { - if (cputime_ge(ptime, sig->it_prof_expires)) { - /* ITIMER_PROF fires and reloads. */ - sig->it_prof_expires = sig->it_prof_incr; - if (!cputime_eq(sig->it_prof_expires, cputime_zero)) { - sig->it_prof_expires = cputime_add( - sig->it_prof_expires, ptime); - } - __group_send_sig_info(SIGPROF, SEND_SIG_PRIV, tsk); - } - if (!cputime_eq(sig->it_prof_expires, cputime_zero) && - (cputime_eq(prof_expires, cputime_zero) || - cputime_lt(sig->it_prof_expires, prof_expires))) { - prof_expires = sig->it_prof_expires; - } - } - if (!cputime_eq(sig->it_virt_expires, cputime_zero)) { - if (cputime_ge(utime, sig->it_virt_expires)) { - /* ITIMER_VIRTUAL fires and reloads. */ - sig->it_virt_expires = sig->it_virt_incr; - if (!cputime_eq(sig->it_virt_expires, cputime_zero)) { - sig->it_virt_expires = cputime_add( - sig->it_virt_expires, utime); - } - __group_send_sig_info(SIGVTALRM, SEND_SIG_PRIV, tsk); - } - if (!cputime_eq(sig->it_virt_expires, cputime_zero) && - (cputime_eq(virt_expires, cputime_zero) || - cputime_lt(sig->it_virt_expires, virt_expires))) { - virt_expires = sig->it_virt_expires; - } - } + check_cpu_itimer(tsk, &sig->it[CPUCLOCK_PROF], &prof_expires, ptime, + SIGPROF); + check_cpu_itimer(tsk, &sig->it[CPUCLOCK_VIRT], &virt_expires, utime, + SIGVTALRM); + if (sig->rlim[RLIMIT_CPU].rlim_cur != RLIM_INFINITY) { unsigned long psecs = cputime_to_secs(ptime); cputime_t x; @@ -1419,19 +1430,19 @@ void run_posix_cpu_timers(struct task_struct *tsk) * timer call will interfere. */ list_for_each_entry_safe(timer, next, &firing, it.cpu.entry) { - int firing; + int cpu_firing; + spin_lock(&timer->it_lock); list_del_init(&timer->it.cpu.entry); - firing = timer->it.cpu.firing; + cpu_firing = timer->it.cpu.firing; timer->it.cpu.firing = 0; /* * The firing flag is -1 if we collided with a reset * of the timer, which already reported this * almost-firing as an overrun. So don't generate an event. */ - if (likely(firing >= 0)) { + if (likely(cpu_firing >= 0)) cpu_timer_fire(timer); - } spin_unlock(&timer->it_lock); } } @@ -1455,7 +1466,7 @@ void set_process_cpu_timer(struct task_struct *tsk, unsigned int clock_idx, if (!cputime_eq(*oldval, cputime_zero)) { if (cputime_le(*oldval, now.cpu)) { /* Just about to fire. */ - *oldval = jiffies_to_cputime(1); + *oldval = cputime_one_jiffy; } else { *oldval = cputime_sub(*oldval, now.cpu); } @@ -1701,10 +1712,15 @@ static __init int init_posix_cpu_timers(void) .nsleep = thread_cpu_nsleep, .nsleep_restart = thread_cpu_nsleep_restart, }; + struct timespec ts; register_posix_clock(CLOCK_PROCESS_CPUTIME_ID, &process); register_posix_clock(CLOCK_THREAD_CPUTIME_ID, &thread); + cputime_to_timespec(cputime_one_jiffy, &ts); + onecputick = ts.tv_nsec; + WARN_ON(ts.tv_sec != 0); + return 0; } __initcall(init_posix_cpu_timers); diff --git a/kernel/posix-timers.c b/kernel/posix-timers.c index 052ec4d195c..495440779ce 100644 --- a/kernel/posix-timers.c +++ b/kernel/posix-timers.c @@ -202,6 +202,12 @@ static int no_timer_create(struct k_itimer *new_timer) return -EOPNOTSUPP; } +static int no_nsleep(const clockid_t which_clock, int flags, + struct timespec *tsave, struct timespec __user *rmtp) +{ + return -EOPNOTSUPP; +} + /* * Return nonzero if we know a priori this clockid_t value is bogus. */ @@ -236,6 +242,25 @@ static int posix_get_monotonic_raw(clockid_t which_clock, struct timespec *tp) return 0; } + +static int posix_get_realtime_coarse(clockid_t which_clock, struct timespec *tp) +{ + *tp = current_kernel_time(); + return 0; +} + +static int posix_get_monotonic_coarse(clockid_t which_clock, + struct timespec *tp) +{ + *tp = get_monotonic_coarse(); + return 0; +} + +int posix_get_coarse_res(const clockid_t which_clock, struct timespec *tp) +{ + *tp = ktime_to_timespec(KTIME_LOW_RES); + return 0; +} /* * Initialize everything, well, just everything in Posix clocks/timers ;) */ @@ -254,11 +279,28 @@ static __init int init_posix_timers(void) .clock_get = posix_get_monotonic_raw, .clock_set = do_posix_clock_nosettime, .timer_create = no_timer_create, + .nsleep = no_nsleep, + }; + struct k_clock clock_realtime_coarse = { + .clock_getres = posix_get_coarse_res, + .clock_get = posix_get_realtime_coarse, + .clock_set = do_posix_clock_nosettime, + .timer_create = no_timer_create, + .nsleep = no_nsleep, + }; + struct k_clock clock_monotonic_coarse = { + .clock_getres = posix_get_coarse_res, + .clock_get = posix_get_monotonic_coarse, + .clock_set = do_posix_clock_nosettime, + .timer_create = no_timer_create, + .nsleep = no_nsleep, }; register_posix_clock(CLOCK_REALTIME, &clock_realtime); register_posix_clock(CLOCK_MONOTONIC, &clock_monotonic); register_posix_clock(CLOCK_MONOTONIC_RAW, &clock_monotonic_raw); + register_posix_clock(CLOCK_REALTIME_COARSE, &clock_realtime_coarse); + register_posix_clock(CLOCK_MONOTONIC_COARSE, &clock_monotonic_coarse); posix_timers_cache = kmem_cache_create("posix_timers_cache", sizeof (struct k_itimer), 0, SLAB_PANIC, diff --git a/kernel/power/Kconfig b/kernel/power/Kconfig index 23bd4daeb96..91e09d3b2eb 100644 --- a/kernel/power/Kconfig +++ b/kernel/power/Kconfig @@ -116,9 +116,13 @@ config SUSPEND_FREEZER Turning OFF this setting is NOT recommended! If in doubt, say Y. +config HIBERNATION_NVS + bool + config HIBERNATION bool "Hibernation (aka 'suspend to disk')" depends on PM && SWAP && ARCH_HIBERNATION_POSSIBLE + select HIBERNATION_NVS if HAS_IOMEM ---help--- Enable the suspend to disk (STD) functionality, which is usually called "hibernation" in user interfaces. STD checkpoints the @@ -204,3 +208,17 @@ config APM_EMULATION random kernel OOPSes or reboots that don't seem to be related to anything, try disabling/enabling this option (or disabling/enabling APM in your BIOS). + +config PM_RUNTIME + bool "Run-time PM core functionality" + depends on PM + ---help--- + Enable functionality allowing I/O devices to be put into energy-saving + (low power) states at run time (or autosuspended) after a specified + period of inactivity and woken up in response to a hardware-generated + wake-up event or a driver's request. + + Hardware support is generally required for this functionality to work + and the bus type drivers of the buses the devices are on are + responsible for the actual handling of the autosuspend requests and + wake-up events. diff --git a/kernel/power/Makefile b/kernel/power/Makefile index 720ea4f781b..43191815f87 100644 --- a/kernel/power/Makefile +++ b/kernel/power/Makefile @@ -6,6 +6,9 @@ endif obj-$(CONFIG_PM) += main.o obj-$(CONFIG_PM_SLEEP) += console.o obj-$(CONFIG_FREEZER) += process.o -obj-$(CONFIG_HIBERNATION) += swsusp.o disk.o snapshot.o swap.o user.o +obj-$(CONFIG_SUSPEND) += suspend.o +obj-$(CONFIG_PM_TEST_SUSPEND) += suspend_test.o +obj-$(CONFIG_HIBERNATION) += hibernate.o snapshot.o swap.o user.o +obj-$(CONFIG_HIBERNATION_NVS) += hibernate_nvs.o obj-$(CONFIG_MAGIC_SYSRQ) += poweroff.o diff --git a/kernel/power/console.c b/kernel/power/console.c index a3961b205de..5187136fe1d 100644 --- a/kernel/power/console.c +++ b/kernel/power/console.c @@ -14,56 +14,13 @@ #define SUSPEND_CONSOLE (MAX_NR_CONSOLES-1) static int orig_fgconsole, orig_kmsg; -static int disable_vt_switch; - -/* - * Normally during a suspend, we allocate a new console and switch to it. - * When we resume, we switch back to the original console. This switch - * can be slow, so on systems where the framebuffer can handle restoration - * of video registers anyways, there's little point in doing the console - * switch. This function allows you to disable it by passing it '0'. - */ -void pm_set_vt_switch(int do_switch) -{ - acquire_console_sem(); - disable_vt_switch = !do_switch; - release_console_sem(); -} -EXPORT_SYMBOL(pm_set_vt_switch); int pm_prepare_console(void) { - acquire_console_sem(); - - if (disable_vt_switch) { - release_console_sem(); - return 0; - } - - orig_fgconsole = fg_console; - - if (vc_allocate(SUSPEND_CONSOLE)) { - /* we can't have a free VC for now. Too bad, - * we don't want to mess the screen for now. */ - release_console_sem(); + orig_fgconsole = vt_move_to_console(SUSPEND_CONSOLE, 1); + if (orig_fgconsole < 0) return 1; - } - if (set_console(SUSPEND_CONSOLE)) { - /* - * We're unable to switch to the SUSPEND_CONSOLE. - * Let the calling function know so it can decide - * what to do. - */ - release_console_sem(); - return 1; - } - release_console_sem(); - - if (vt_waitactive(SUSPEND_CONSOLE)) { - pr_debug("Suspend: Can't switch VCs."); - return 1; - } orig_kmsg = kmsg_redirect; kmsg_redirect = SUSPEND_CONSOLE; return 0; @@ -71,19 +28,9 @@ int pm_prepare_console(void) void pm_restore_console(void) { - acquire_console_sem(); - if (disable_vt_switch) { - release_console_sem(); - return; - } - set_console(orig_fgconsole); - release_console_sem(); - - if (vt_waitactive(orig_fgconsole)) { - pr_debug("Resume: Can't switch VCs."); - return; + if (orig_fgconsole >= 0) { + vt_move_to_console(orig_fgconsole, 0); + kmsg_redirect = orig_kmsg; } - - kmsg_redirect = orig_kmsg; } #endif diff --git a/kernel/power/disk.c b/kernel/power/hibernate.c index f3db382c2b2..bbfe472d752 100644 --- a/kernel/power/disk.c +++ b/kernel/power/hibernate.c @@ -1,12 +1,12 @@ /* - * kernel/power/disk.c - Suspend-to-disk support. + * kernel/power/hibernate.c - Hibernation (a.k.a suspend-to-disk) support. * * Copyright (c) 2003 Patrick Mochel * Copyright (c) 2003 Open Source Development Lab * Copyright (c) 2004 Pavel Machek <pavel@suse.cz> + * Copyright (c) 2009 Rafael J. Wysocki, Novell Inc. * * This file is released under the GPLv2. - * */ #include <linux/suspend.h> @@ -22,6 +22,7 @@ #include <linux/console.h> #include <linux/cpu.h> #include <linux/freezer.h> +#include <scsi/scsi_scan.h> #include <asm/suspend.h> #include "power.h" @@ -31,6 +32,7 @@ static int noresume = 0; static char resume_file[256] = CONFIG_PM_STD_PARTITION; dev_t swsusp_resume_device; sector_t swsusp_resume_block; +int in_suspend __nosavedata = 0; enum { HIBERNATION_INVALID, @@ -201,6 +203,35 @@ static void platform_recover(int platform_mode) } /** + * swsusp_show_speed - print the time elapsed between two events. + * @start: Starting event. + * @stop: Final event. + * @nr_pages - number of pages processed between @start and @stop + * @msg - introductory message to print + */ + +void swsusp_show_speed(struct timeval *start, struct timeval *stop, + unsigned nr_pages, char *msg) +{ + s64 elapsed_centisecs64; + int centisecs; + int k; + int kps; + + elapsed_centisecs64 = timeval_to_ns(stop) - timeval_to_ns(start); + do_div(elapsed_centisecs64, NSEC_PER_SEC / 100); + centisecs = elapsed_centisecs64; + if (centisecs == 0) + centisecs = 1; /* avoid div-by-zero */ + k = nr_pages * (PAGE_SIZE / 1024); + kps = (k * 100) / centisecs; + printk(KERN_INFO "PM: %s %d kbytes in %d.%02d seconds (%d.%02d MB/s)\n", + msg, k, + centisecs / 100, centisecs % 100, + kps / 1000, (kps % 1000) / 10); +} + +/** * create_image - freeze devices that need to be frozen with interrupts * off, create the hibernation image and thaw those devices. Control * reappears in this routine after a restore. @@ -214,19 +245,17 @@ static int create_image(int platform_mode) if (error) return error; - device_pm_lock(); - - /* At this point, device_suspend() has been called, but *not* - * device_power_down(). We *must* call device_power_down() now. + /* At this point, dpm_suspend_start() has been called, but *not* + * dpm_suspend_noirq(). We *must* call dpm_suspend_noirq() now. * Otherwise, drivers for some devices (e.g. interrupt controllers) * become desynchronized with the actual state of the hardware * at resume time, and evil weirdness ensues. */ - error = device_power_down(PMSG_FREEZE); + error = dpm_suspend_noirq(PMSG_FREEZE); if (error) { printk(KERN_ERR "PM: Some devices failed to power down, " "aborting hibernation\n"); - goto Unlock; + return error; } error = platform_pre_snapshot(platform_mode); @@ -240,9 +269,9 @@ static int create_image(int platform_mode) local_irq_disable(); - sysdev_suspend(PMSG_FREEZE); + error = sysdev_suspend(PMSG_FREEZE); if (error) { - printk(KERN_ERR "PM: Some devices failed to power down, " + printk(KERN_ERR "PM: Some system devices failed to power down, " "aborting hibernation\n"); goto Enable_irqs; } @@ -263,7 +292,7 @@ static int create_image(int platform_mode) Power_up: sysdev_resume(); - /* NOTE: device_power_up() is just a resume() for devices + /* NOTE: dpm_resume_noirq() is just a resume() for devices * that suspended with irqs off ... no overall powerup. */ @@ -276,12 +305,9 @@ static int create_image(int platform_mode) Platform_finish: platform_finish(platform_mode); - device_power_up(in_suspend ? + dpm_resume_noirq(in_suspend ? (error ? PMSG_RECOVER : PMSG_THAW) : PMSG_RESTORE); - Unlock: - device_pm_unlock(); - return error; } @@ -289,7 +315,7 @@ static int create_image(int platform_mode) * hibernation_snapshot - quiesce devices and create the hibernation * snapshot image. * @platform_mode - if set, use the platform driver, if available, to - * prepare the platform frimware for the power transition. + * prepare the platform firmware for the power transition. * * Must be called with pm_mutex held */ @@ -302,13 +328,13 @@ int hibernation_snapshot(int platform_mode) if (error) return error; - /* Free memory before shutting down devices. */ - error = swsusp_shrink_memory(); + /* Preallocate image memory before shutting down devices. */ + error = hibernate_preallocate_memory(); if (error) goto Close; suspend_console(); - error = device_suspend(PMSG_FREEZE); + error = dpm_suspend_start(PMSG_FREEZE); if (error) goto Recover_platform; @@ -319,7 +345,11 @@ int hibernation_snapshot(int platform_mode) /* Control returns here after successful restore */ Resume_devices: - device_resume(in_suspend ? + /* We may need to release the preallocated image pages here. */ + if (error || !in_suspend) + swsusp_free(); + + dpm_resume_end(in_suspend ? (error ? PMSG_RECOVER : PMSG_THAW) : PMSG_RESTORE); resume_console(); Close: @@ -343,13 +373,11 @@ static int resume_target_kernel(bool platform_mode) { int error; - device_pm_lock(); - - error = device_power_down(PMSG_QUIESCE); + error = dpm_suspend_noirq(PMSG_QUIESCE); if (error) { printk(KERN_ERR "PM: Some devices failed to power down, " "aborting resume\n"); - goto Unlock; + return error; } error = platform_pre_restore(platform_mode); @@ -400,10 +428,7 @@ static int resume_target_kernel(bool platform_mode) Cleanup: platform_restore_cleanup(platform_mode); - device_power_up(PMSG_RECOVER); - - Unlock: - device_pm_unlock(); + dpm_resume_noirq(PMSG_RECOVER); return error; } @@ -412,7 +437,7 @@ static int resume_target_kernel(bool platform_mode) * hibernation_restore - quiesce devices and restore the hibernation * snapshot image. If successful, control returns in hibernation_snaphot() * @platform_mode - if set, use the platform driver, if available, to - * prepare the platform frimware for the transition. + * prepare the platform firmware for the transition. * * Must be called with pm_mutex held */ @@ -423,10 +448,10 @@ int hibernation_restore(int platform_mode) pm_prepare_console(); suspend_console(); - error = device_suspend(PMSG_QUIESCE); + error = dpm_suspend_start(PMSG_QUIESCE); if (!error) { error = resume_target_kernel(platform_mode); - device_resume(PMSG_RECOVER); + dpm_resume_end(PMSG_RECOVER); } resume_console(); pm_restore_console(); @@ -456,26 +481,24 @@ int hibernation_platform_enter(void) entering_platform_hibernation = true; suspend_console(); - error = device_suspend(PMSG_HIBERNATE); + error = dpm_suspend_start(PMSG_HIBERNATE); if (error) { if (hibernation_ops->recover) hibernation_ops->recover(); goto Resume_devices; } - device_pm_lock(); - - error = device_power_down(PMSG_HIBERNATE); + error = dpm_suspend_noirq(PMSG_HIBERNATE); if (error) - goto Unlock; + goto Resume_devices; error = hibernation_ops->prepare(); if (error) - goto Platofrm_finish; + goto Platform_finish; error = disable_nonboot_cpus(); if (error) - goto Platofrm_finish; + goto Platform_finish; local_irq_disable(); sysdev_suspend(PMSG_HIBERNATE); @@ -487,17 +510,14 @@ int hibernation_platform_enter(void) * We don't need to reenable the nonboot CPUs or resume consoles, since * the system is going to be halted anyway. */ - Platofrm_finish: + Platform_finish: hibernation_ops->finish(); - device_power_up(PMSG_RESTORE); - - Unlock: - device_pm_unlock(); + dpm_suspend_noirq(PMSG_RESTORE); Resume_devices: entering_platform_hibernation = false; - device_resume(PMSG_RESTORE); + dpm_resume_end(PMSG_RESTORE); resume_console(); Close: @@ -592,7 +612,10 @@ int hibernate(void) goto Thaw; error = hibernation_snapshot(hibernation_mode == HIBERNATION_PLATFORM); - if (in_suspend && !error) { + if (error) + goto Thaw; + + if (in_suspend) { unsigned int flags = 0; if (hibernation_mode == HIBERNATION_PLATFORM) @@ -604,8 +627,8 @@ int hibernate(void) power_down(); } else { pr_debug("PM: Image restored successfully.\n"); - swsusp_free(); } + Thaw: thaw_processes(); Finish: @@ -655,32 +678,42 @@ static int software_resume(void) * here to avoid lockdep complaining. */ mutex_lock_nested(&pm_mutex, SINGLE_DEPTH_NESTING); + + if (swsusp_resume_device) + goto Check_image; + + if (!strlen(resume_file)) { + error = -ENOENT; + goto Unlock; + } + + pr_debug("PM: Checking image partition %s\n", resume_file); + + /* Check if the device is there */ + swsusp_resume_device = name_to_dev_t(resume_file); if (!swsusp_resume_device) { - if (!strlen(resume_file)) { - mutex_unlock(&pm_mutex); - return -ENOENT; - } /* * Some device discovery might still be in progress; we need * to wait for this to finish. */ wait_for_device_probe(); + /* + * We can't depend on SCSI devices being available after loading + * one of their modules until scsi_complete_async_scans() is + * called and the resume device usually is a SCSI one. + */ + scsi_complete_async_scans(); + swsusp_resume_device = name_to_dev_t(resume_file); - pr_debug("PM: Resume from partition %s\n", resume_file); - } else { - pr_debug("PM: Resume from partition %d:%d\n", - MAJOR(swsusp_resume_device), - MINOR(swsusp_resume_device)); + if (!swsusp_resume_device) { + error = -ENODEV; + goto Unlock; + } } - if (noresume) { - /** - * FIXME: If noresume is specified, we need to find the - * partition and reset it back to normal swap space. - */ - mutex_unlock(&pm_mutex); - return 0; - } + Check_image: + pr_debug("PM: Resume from partition %d:%d\n", + MAJOR(swsusp_resume_device), MINOR(swsusp_resume_device)); pr_debug("PM: Checking hibernation image.\n"); error = swsusp_check(); @@ -690,21 +723,22 @@ static int software_resume(void) /* The snapshot device should not be opened while we're running */ if (!atomic_add_unless(&snapshot_device_available, -1, 0)) { error = -EBUSY; + swsusp_close(FMODE_READ); goto Unlock; } pm_prepare_console(); error = pm_notifier_call_chain(PM_RESTORE_PREPARE); if (error) - goto Finish; + goto close_finish; error = usermodehelper_disable(); if (error) - goto Finish; + goto close_finish; error = create_basic_memory_bitmaps(); if (error) - goto Finish; + goto close_finish; pr_debug("PM: Preparing processes for restore.\n"); error = prepare_processes(); @@ -716,6 +750,7 @@ static int software_resume(void) pr_debug("PM: Reading hibernation image.\n"); error = swsusp_read(&flags); + swsusp_close(FMODE_READ); if (!error) hibernation_restore(flags & SF_PLATFORM_MODE); @@ -734,6 +769,9 @@ static int software_resume(void) mutex_unlock(&pm_mutex); pr_debug("PM: Resume from disk failed.\n"); return error; +close_finish: + swsusp_close(FMODE_READ); + goto Finish; } late_initcall(software_resume); diff --git a/kernel/power/hibernate_nvs.c b/kernel/power/hibernate_nvs.c new file mode 100644 index 00000000000..39ac698ef83 --- /dev/null +++ b/kernel/power/hibernate_nvs.c @@ -0,0 +1,135 @@ +/* + * linux/kernel/power/hibernate_nvs.c - Routines for handling NVS memory + * + * Copyright (C) 2008,2009 Rafael J. Wysocki <rjw@sisk.pl>, Novell Inc. + * + * This file is released under the GPLv2. + */ + +#include <linux/io.h> +#include <linux/kernel.h> +#include <linux/list.h> +#include <linux/mm.h> +#include <linux/suspend.h> + +/* + * Platforms, like ACPI, may want us to save some memory used by them during + * hibernation and to restore the contents of this memory during the subsequent + * resume. The code below implements a mechanism allowing us to do that. + */ + +struct nvs_page { + unsigned long phys_start; + unsigned int size; + void *kaddr; + void *data; + struct list_head node; +}; + +static LIST_HEAD(nvs_list); + +/** + * hibernate_nvs_register - register platform NVS memory region to save + * @start - physical address of the region + * @size - size of the region + * + * The NVS region need not be page-aligned (both ends) and we arrange + * things so that the data from page-aligned addresses in this region will + * be copied into separate RAM pages. + */ +int hibernate_nvs_register(unsigned long start, unsigned long size) +{ + struct nvs_page *entry, *next; + + while (size > 0) { + unsigned int nr_bytes; + + entry = kzalloc(sizeof(struct nvs_page), GFP_KERNEL); + if (!entry) + goto Error; + + list_add_tail(&entry->node, &nvs_list); + entry->phys_start = start; + nr_bytes = PAGE_SIZE - (start & ~PAGE_MASK); + entry->size = (size < nr_bytes) ? size : nr_bytes; + + start += entry->size; + size -= entry->size; + } + return 0; + + Error: + list_for_each_entry_safe(entry, next, &nvs_list, node) { + list_del(&entry->node); + kfree(entry); + } + return -ENOMEM; +} + +/** + * hibernate_nvs_free - free data pages allocated for saving NVS regions + */ +void hibernate_nvs_free(void) +{ + struct nvs_page *entry; + + list_for_each_entry(entry, &nvs_list, node) + if (entry->data) { + free_page((unsigned long)entry->data); + entry->data = NULL; + if (entry->kaddr) { + iounmap(entry->kaddr); + entry->kaddr = NULL; + } + } +} + +/** + * hibernate_nvs_alloc - allocate memory necessary for saving NVS regions + */ +int hibernate_nvs_alloc(void) +{ + struct nvs_page *entry; + + list_for_each_entry(entry, &nvs_list, node) { + entry->data = (void *)__get_free_page(GFP_KERNEL); + if (!entry->data) { + hibernate_nvs_free(); + return -ENOMEM; + } + } + return 0; +} + +/** + * hibernate_nvs_save - save NVS memory regions + */ +void hibernate_nvs_save(void) +{ + struct nvs_page *entry; + + printk(KERN_INFO "PM: Saving platform NVS memory\n"); + + list_for_each_entry(entry, &nvs_list, node) + if (entry->data) { + entry->kaddr = ioremap(entry->phys_start, entry->size); + memcpy(entry->data, entry->kaddr, entry->size); + } +} + +/** + * hibernate_nvs_restore - restore NVS memory regions + * + * This function is going to be called with interrupts disabled, so it + * cannot iounmap the virtual addresses used to access the NVS region. + */ +void hibernate_nvs_restore(void) +{ + struct nvs_page *entry; + + printk(KERN_INFO "PM: Restoring platform NVS memory\n"); + + list_for_each_entry(entry, &nvs_list, node) + if (entry->data) + memcpy(entry->kaddr, entry->data, entry->size); +} diff --git a/kernel/power/main.c b/kernel/power/main.c index f172f41858b..0998c713905 100644 --- a/kernel/power/main.c +++ b/kernel/power/main.c @@ -8,20 +8,10 @@ * */ -#include <linux/module.h> -#include <linux/suspend.h> #include <linux/kobject.h> #include <linux/string.h> -#include <linux/delay.h> -#include <linux/errno.h> -#include <linux/kmod.h> -#include <linux/init.h> -#include <linux/console.h> -#include <linux/cpu.h> #include <linux/resume-trace.h> -#include <linux/freezer.h> -#include <linux/vmstat.h> -#include <linux/syscalls.h> +#include <linux/workqueue.h> #include "power.h" @@ -119,368 +109,6 @@ power_attr(pm_test); #endif /* CONFIG_PM_SLEEP */ -#ifdef CONFIG_SUSPEND - -static int suspend_test(int level) -{ -#ifdef CONFIG_PM_DEBUG - if (pm_test_level == level) { - printk(KERN_INFO "suspend debug: Waiting for 5 seconds.\n"); - mdelay(5000); - return 1; - } -#endif /* !CONFIG_PM_DEBUG */ - return 0; -} - -#ifdef CONFIG_PM_TEST_SUSPEND - -/* - * We test the system suspend code by setting an RTC wakealarm a short - * time in the future, then suspending. Suspending the devices won't - * normally take long ... some systems only need a few milliseconds. - * - * The time it takes is system-specific though, so when we test this - * during system bootup we allow a LOT of time. - */ -#define TEST_SUSPEND_SECONDS 5 - -static unsigned long suspend_test_start_time; - -static void suspend_test_start(void) -{ - /* FIXME Use better timebase than "jiffies", ideally a clocksource. - * What we want is a hardware counter that will work correctly even - * during the irqs-are-off stages of the suspend/resume cycle... - */ - suspend_test_start_time = jiffies; -} - -static void suspend_test_finish(const char *label) -{ - long nj = jiffies - suspend_test_start_time; - unsigned msec; - - msec = jiffies_to_msecs(abs(nj)); - pr_info("PM: %s took %d.%03d seconds\n", label, - msec / 1000, msec % 1000); - - /* Warning on suspend means the RTC alarm period needs to be - * larger -- the system was sooo slooowwww to suspend that the - * alarm (should have) fired before the system went to sleep! - * - * Warning on either suspend or resume also means the system - * has some performance issues. The stack dump of a WARN_ON - * is more likely to get the right attention than a printk... - */ - WARN(msec > (TEST_SUSPEND_SECONDS * 1000), "Component: %s\n", label); -} - -#else - -static void suspend_test_start(void) -{ -} - -static void suspend_test_finish(const char *label) -{ -} - -#endif - -/* This is just an arbitrary number */ -#define FREE_PAGE_NUMBER (100) - -static struct platform_suspend_ops *suspend_ops; - -/** - * suspend_set_ops - Set the global suspend method table. - * @ops: Pointer to ops structure. - */ - -void suspend_set_ops(struct platform_suspend_ops *ops) -{ - mutex_lock(&pm_mutex); - suspend_ops = ops; - mutex_unlock(&pm_mutex); -} - -/** - * suspend_valid_only_mem - generic memory-only valid callback - * - * Platform drivers that implement mem suspend only and only need - * to check for that in their .valid callback can use this instead - * of rolling their own .valid callback. - */ -int suspend_valid_only_mem(suspend_state_t state) -{ - return state == PM_SUSPEND_MEM; -} - -/** - * suspend_prepare - Do prep work before entering low-power state. - * - * This is common code that is called for each state that we're entering. - * Run suspend notifiers, allocate a console and stop all processes. - */ -static int suspend_prepare(void) -{ - int error; - unsigned int free_pages; - - if (!suspend_ops || !suspend_ops->enter) - return -EPERM; - - pm_prepare_console(); - - error = pm_notifier_call_chain(PM_SUSPEND_PREPARE); - if (error) - goto Finish; - - error = usermodehelper_disable(); - if (error) - goto Finish; - - if (suspend_freeze_processes()) { - error = -EAGAIN; - goto Thaw; - } - - free_pages = global_page_state(NR_FREE_PAGES); - if (free_pages < FREE_PAGE_NUMBER) { - pr_debug("PM: free some memory\n"); - shrink_all_memory(FREE_PAGE_NUMBER - free_pages); - if (nr_free_pages() < FREE_PAGE_NUMBER) { - error = -ENOMEM; - printk(KERN_ERR "PM: No enough memory\n"); - } - } - if (!error) - return 0; - - Thaw: - suspend_thaw_processes(); - usermodehelper_enable(); - Finish: - pm_notifier_call_chain(PM_POST_SUSPEND); - pm_restore_console(); - return error; -} - -/* default implementation */ -void __attribute__ ((weak)) arch_suspend_disable_irqs(void) -{ - local_irq_disable(); -} - -/* default implementation */ -void __attribute__ ((weak)) arch_suspend_enable_irqs(void) -{ - local_irq_enable(); -} - -/** - * suspend_enter - enter the desired system sleep state. - * @state: state to enter - * - * This function should be called after devices have been suspended. - */ -static int suspend_enter(suspend_state_t state) -{ - int error; - - device_pm_lock(); - - error = device_power_down(PMSG_SUSPEND); - if (error) { - printk(KERN_ERR "PM: Some devices failed to power down\n"); - goto Done; - } - - if (suspend_ops->prepare) { - error = suspend_ops->prepare(); - if (error) - goto Power_up_devices; - } - - if (suspend_test(TEST_PLATFORM)) - goto Platfrom_finish; - - error = disable_nonboot_cpus(); - if (error || suspend_test(TEST_CPUS)) - goto Enable_cpus; - - arch_suspend_disable_irqs(); - BUG_ON(!irqs_disabled()); - - error = sysdev_suspend(PMSG_SUSPEND); - if (!error) { - if (!suspend_test(TEST_CORE)) - error = suspend_ops->enter(state); - sysdev_resume(); - } - - arch_suspend_enable_irqs(); - BUG_ON(irqs_disabled()); - - Enable_cpus: - enable_nonboot_cpus(); - - Platfrom_finish: - if (suspend_ops->finish) - suspend_ops->finish(); - - Power_up_devices: - device_power_up(PMSG_RESUME); - - Done: - device_pm_unlock(); - - return error; -} - -/** - * suspend_devices_and_enter - suspend devices and enter the desired system - * sleep state. - * @state: state to enter - */ -int suspend_devices_and_enter(suspend_state_t state) -{ - int error; - - if (!suspend_ops) - return -ENOSYS; - - if (suspend_ops->begin) { - error = suspend_ops->begin(state); - if (error) - goto Close; - } - suspend_console(); - suspend_test_start(); - error = device_suspend(PMSG_SUSPEND); - if (error) { - printk(KERN_ERR "PM: Some devices failed to suspend\n"); - goto Recover_platform; - } - suspend_test_finish("suspend devices"); - if (suspend_test(TEST_DEVICES)) - goto Recover_platform; - - suspend_enter(state); - - Resume_devices: - suspend_test_start(); - device_resume(PMSG_RESUME); - suspend_test_finish("resume devices"); - resume_console(); - Close: - if (suspend_ops->end) - suspend_ops->end(); - return error; - - Recover_platform: - if (suspend_ops->recover) - suspend_ops->recover(); - goto Resume_devices; -} - -/** - * suspend_finish - Do final work before exiting suspend sequence. - * - * Call platform code to clean up, restart processes, and free the - * console that we've allocated. This is not called for suspend-to-disk. - */ -static void suspend_finish(void) -{ - suspend_thaw_processes(); - usermodehelper_enable(); - pm_notifier_call_chain(PM_POST_SUSPEND); - pm_restore_console(); -} - - - - -static const char * const pm_states[PM_SUSPEND_MAX] = { - [PM_SUSPEND_STANDBY] = "standby", - [PM_SUSPEND_MEM] = "mem", -}; - -static inline int valid_state(suspend_state_t state) -{ - /* All states need lowlevel support and need to be valid - * to the lowlevel implementation, no valid callback - * implies that none are valid. */ - if (!suspend_ops || !suspend_ops->valid || !suspend_ops->valid(state)) - return 0; - return 1; -} - - -/** - * enter_state - Do common work of entering low-power state. - * @state: pm_state structure for state we're entering. - * - * Make sure we're the only ones trying to enter a sleep state. Fail - * if someone has beat us to it, since we don't want anything weird to - * happen when we wake up. - * Then, do the setup for suspend, enter the state, and cleaup (after - * we've woken up). - */ -static int enter_state(suspend_state_t state) -{ - int error; - - if (!valid_state(state)) - return -ENODEV; - - if (!mutex_trylock(&pm_mutex)) - return -EBUSY; - - printk(KERN_INFO "PM: Syncing filesystems ... "); - sys_sync(); - printk("done.\n"); - - pr_debug("PM: Preparing system for %s sleep\n", pm_states[state]); - error = suspend_prepare(); - if (error) - goto Unlock; - - if (suspend_test(TEST_FREEZER)) - goto Finish; - - pr_debug("PM: Entering %s sleep\n", pm_states[state]); - error = suspend_devices_and_enter(state); - - Finish: - pr_debug("PM: Finishing wakeup.\n"); - suspend_finish(); - Unlock: - mutex_unlock(&pm_mutex); - return error; -} - - -/** - * pm_suspend - Externally visible function for suspending system. - * @state: Enumerated value of state to enter. - * - * Determine whether or not value is within range, get state - * structure, and enter (above). - */ - -int pm_suspend(suspend_state_t state) -{ - if (state > PM_SUSPEND_ON && state <= PM_SUSPEND_MAX) - return enter_state(state); - return -EINVAL; -} - -EXPORT_SYMBOL(pm_suspend); - -#endif /* CONFIG_SUSPEND */ - struct kobject *power_kobj; /** @@ -493,7 +121,6 @@ struct kobject *power_kobj; * store() accepts one of those strings, translates it into the * proper enumerated value, and initiates a suspend transition. */ - static ssize_t state_show(struct kobject *kobj, struct kobj_attribute *attr, char *buf) { @@ -591,9 +218,25 @@ static struct attribute_group attr_group = { .attrs = g, }; +#ifdef CONFIG_PM_RUNTIME +struct workqueue_struct *pm_wq; +EXPORT_SYMBOL_GPL(pm_wq); + +static int __init pm_start_workqueue(void) +{ + pm_wq = create_freezeable_workqueue("pm"); + + return pm_wq ? 0 : -ENOMEM; +} +#else +static inline int pm_start_workqueue(void) { return 0; } +#endif static int __init pm_init(void) { + int error = pm_start_workqueue(); + if (error) + return error; power_kobj = kobject_create_and_add("power", NULL); if (!power_kobj) return -ENOMEM; @@ -601,144 +244,3 @@ static int __init pm_init(void) } core_initcall(pm_init); - - -#ifdef CONFIG_PM_TEST_SUSPEND - -#include <linux/rtc.h> - -/* - * To test system suspend, we need a hands-off mechanism to resume the - * system. RTCs wake alarms are a common self-contained mechanism. - */ - -static void __init test_wakealarm(struct rtc_device *rtc, suspend_state_t state) -{ - static char err_readtime[] __initdata = - KERN_ERR "PM: can't read %s time, err %d\n"; - static char err_wakealarm [] __initdata = - KERN_ERR "PM: can't set %s wakealarm, err %d\n"; - static char err_suspend[] __initdata = - KERN_ERR "PM: suspend test failed, error %d\n"; - static char info_test[] __initdata = - KERN_INFO "PM: test RTC wakeup from '%s' suspend\n"; - - unsigned long now; - struct rtc_wkalrm alm; - int status; - - /* this may fail if the RTC hasn't been initialized */ - status = rtc_read_time(rtc, &alm.time); - if (status < 0) { - printk(err_readtime, dev_name(&rtc->dev), status); - return; - } - rtc_tm_to_time(&alm.time, &now); - - memset(&alm, 0, sizeof alm); - rtc_time_to_tm(now + TEST_SUSPEND_SECONDS, &alm.time); - alm.enabled = true; - - status = rtc_set_alarm(rtc, &alm); - if (status < 0) { - printk(err_wakealarm, dev_name(&rtc->dev), status); - return; - } - - if (state == PM_SUSPEND_MEM) { - printk(info_test, pm_states[state]); - status = pm_suspend(state); - if (status == -ENODEV) - state = PM_SUSPEND_STANDBY; - } - if (state == PM_SUSPEND_STANDBY) { - printk(info_test, pm_states[state]); - status = pm_suspend(state); - } - if (status < 0) - printk(err_suspend, status); - - /* Some platforms can't detect that the alarm triggered the - * wakeup, or (accordingly) disable it after it afterwards. - * It's supposed to give oneshot behavior; cope. - */ - alm.enabled = false; - rtc_set_alarm(rtc, &alm); -} - -static int __init has_wakealarm(struct device *dev, void *name_ptr) -{ - struct rtc_device *candidate = to_rtc_device(dev); - - if (!candidate->ops->set_alarm) - return 0; - if (!device_may_wakeup(candidate->dev.parent)) - return 0; - - *(const char **)name_ptr = dev_name(dev); - return 1; -} - -/* - * Kernel options like "test_suspend=mem" force suspend/resume sanity tests - * at startup time. They're normally disabled, for faster boot and because - * we can't know which states really work on this particular system. - */ -static suspend_state_t test_state __initdata = PM_SUSPEND_ON; - -static char warn_bad_state[] __initdata = - KERN_WARNING "PM: can't test '%s' suspend state\n"; - -static int __init setup_test_suspend(char *value) -{ - unsigned i; - - /* "=mem" ==> "mem" */ - value++; - for (i = 0; i < PM_SUSPEND_MAX; i++) { - if (!pm_states[i]) - continue; - if (strcmp(pm_states[i], value) != 0) - continue; - test_state = (__force suspend_state_t) i; - return 0; - } - printk(warn_bad_state, value); - return 0; -} -__setup("test_suspend", setup_test_suspend); - -static int __init test_suspend(void) -{ - static char warn_no_rtc[] __initdata = - KERN_WARNING "PM: no wakealarm-capable RTC driver is ready\n"; - - char *pony = NULL; - struct rtc_device *rtc = NULL; - - /* PM is initialized by now; is that state testable? */ - if (test_state == PM_SUSPEND_ON) - goto done; - if (!valid_state(test_state)) { - printk(warn_bad_state, pm_states[test_state]); - goto done; - } - - /* RTCs have initialized by now too ... can we use one? */ - class_find_device(rtc_class, NULL, &pony, has_wakealarm); - if (pony) - rtc = rtc_class_open(pony); - if (!rtc) { - printk(warn_no_rtc); - goto done; - } - - /* go for it */ - test_wakealarm(rtc, test_state); - rtc_class_close(rtc); -done: - return 0; -} -late_initcall(test_suspend); - -#endif /* CONFIG_PM_TEST_SUSPEND */ diff --git a/kernel/power/power.h b/kernel/power/power.h index 46b5ec7a3af..46c5a26630a 100644 --- a/kernel/power/power.h +++ b/kernel/power/power.h @@ -45,7 +45,7 @@ static inline char *check_image_kernel(struct swsusp_info *info) */ #define SPARE_PAGES ((1024 * 1024) >> PAGE_SHIFT) -/* kernel/power/disk.c */ +/* kernel/power/hibernate.c */ extern int hibernation_snapshot(int platform_mode); extern int hibernation_restore(int platform_mode); extern int hibernation_platform_enter(void); @@ -74,7 +74,7 @@ extern asmlinkage int swsusp_arch_resume(void); extern int create_basic_memory_bitmaps(void); extern void free_basic_memory_bitmaps(void); -extern unsigned int count_data_pages(void); +extern int hibernate_preallocate_memory(void); /** * Auxiliary structure used for reading the snapshot image data and @@ -147,9 +147,8 @@ extern int swsusp_swap_in_use(void); */ #define SF_PLATFORM_MODE 1 -/* kernel/power/disk.c */ +/* kernel/power/hibernate.c */ extern int swsusp_check(void); -extern int swsusp_shrink_memory(void); extern void swsusp_free(void); extern int swsusp_read(unsigned int *flags_p); extern int swsusp_write(unsigned int flags); @@ -161,22 +160,36 @@ extern void swsusp_show_speed(struct timeval *, struct timeval *, unsigned int, char *); #ifdef CONFIG_SUSPEND -/* kernel/power/main.c */ +/* kernel/power/suspend.c */ +extern const char *const pm_states[]; + +extern bool valid_state(suspend_state_t state); extern int suspend_devices_and_enter(suspend_state_t state); +extern int enter_state(suspend_state_t state); #else /* !CONFIG_SUSPEND */ static inline int suspend_devices_and_enter(suspend_state_t state) { return -ENOSYS; } +static inline int enter_state(suspend_state_t state) { return -ENOSYS; } +static inline bool valid_state(suspend_state_t state) { return false; } #endif /* !CONFIG_SUSPEND */ +#ifdef CONFIG_PM_TEST_SUSPEND +/* kernel/power/suspend_test.c */ +extern void suspend_test_start(void); +extern void suspend_test_finish(const char *label); +#else /* !CONFIG_PM_TEST_SUSPEND */ +static inline void suspend_test_start(void) {} +static inline void suspend_test_finish(const char *label) {} +#endif /* !CONFIG_PM_TEST_SUSPEND */ + #ifdef CONFIG_PM_SLEEP /* kernel/power/main.c */ extern int pm_notifier_call_chain(unsigned long val); #endif #ifdef CONFIG_HIGHMEM -unsigned int count_highmem_pages(void); int restore_highmem(void); #else static inline unsigned int count_highmem_pages(void) { return 0; } diff --git a/kernel/power/poweroff.c b/kernel/power/poweroff.c index 97890831e1b..e8b33700627 100644 --- a/kernel/power/poweroff.c +++ b/kernel/power/poweroff.c @@ -34,7 +34,7 @@ static struct sysrq_key_op sysrq_poweroff_op = { .handler = handle_poweroff, .help_msg = "powerOff", .action_msg = "Power Off", - .enable_mask = SYSRQ_ENABLE_BOOT, + .enable_mask = SYSRQ_ENABLE_BOOT, }; static int pm_sysrq_init(void) diff --git a/kernel/power/process.c b/kernel/power/process.c index ca634019497..5ade1bdcf36 100644 --- a/kernel/power/process.c +++ b/kernel/power/process.c @@ -9,10 +9,12 @@ #undef DEBUG #include <linux/interrupt.h> +#include <linux/oom.h> #include <linux/suspend.h> #include <linux/module.h> #include <linux/syscalls.h> #include <linux/freezer.h> +#include <linux/delay.h> /* * Timeout for stopping processes @@ -40,7 +42,7 @@ static int try_to_freeze_tasks(bool sig_only) do_gettimeofday(&start); end_time = jiffies + TIMEOUT; - do { + while (true) { todo = 0; read_lock(&tasklist_lock); do_each_thread(g, p) { @@ -61,10 +63,15 @@ static int try_to_freeze_tasks(bool sig_only) todo++; } while_each_thread(g, p); read_unlock(&tasklist_lock); - yield(); /* Yield is okay here */ - if (time_after(jiffies, end_time)) + if (!todo || time_after(jiffies, end_time)) break; - } while (todo); + + /* + * We need to retry, but first give the freezing tasks some + * time to enter the regrigerator. + */ + msleep(10); + } do_gettimeofday(&end); elapsed_csecs64 = timeval_to_ns(&end) - timeval_to_ns(&start); @@ -117,9 +124,12 @@ int freeze_processes(void) if (error) goto Exit; printk("done."); + + oom_killer_disable(); Exit: BUG_ON(in_atomic()); printk("\n"); + return error; } @@ -145,6 +155,8 @@ static void thaw_tasks(bool nosig_only) void thaw_processes(void) { + oom_killer_enable(); + printk("Restarting tasks ... "); thaw_tasks(true); thaw_tasks(false); diff --git a/kernel/power/snapshot.c b/kernel/power/snapshot.c index 33e2e4a819f..36cb168e433 100644 --- a/kernel/power/snapshot.c +++ b/kernel/power/snapshot.c @@ -39,6 +39,14 @@ static int swsusp_page_is_free(struct page *); static void swsusp_set_page_forbidden(struct page *); static void swsusp_unset_page_forbidden(struct page *); +/* + * Preferred image size in bytes (tunable via /sys/power/image_size). + * When it is set to N, swsusp will do its best to ensure the image + * size will not exceed N bytes, but if that is impossible, it will + * try to create the smallest image possible. + */ +unsigned long image_size = 500 * 1024 * 1024; + /* List of PBEs needed for restoring the pages that were allocated before * the suspend and included in the suspend image, but have also been * allocated by the "resume" kernel, so their contents cannot be written @@ -225,7 +233,7 @@ static void *chain_alloc(struct chain_allocator *ca, unsigned int size) #define BM_END_OF_MAP (~0UL) -#define BM_BITS_PER_BLOCK (PAGE_SIZE << 3) +#define BM_BITS_PER_BLOCK (PAGE_SIZE * BITS_PER_BYTE) struct bm_block { struct list_head hook; /* hook into a list of bitmap blocks */ @@ -267,7 +275,7 @@ static void memory_bm_free(struct memory_bitmap *bm, int clear_nosave_free); /** * create_bm_block_list - create a list of block bitmap objects - * @nr_blocks - number of blocks to allocate + * @pages - number of pages to track * @list - list to put the allocated blocks into * @ca - chain allocator to be used for allocating memory */ @@ -611,7 +619,7 @@ __register_nosave_region(unsigned long start_pfn, unsigned long end_pfn, BUG_ON(!region); } else /* This allocation cannot fail */ - region = alloc_bootmem_low(sizeof(struct nosave_region)); + region = alloc_bootmem(sizeof(struct nosave_region)); region->start_pfn = start_pfn; region->end_pfn = end_pfn; list_add_tail(®ion->list, &nosave_regions); @@ -840,12 +848,12 @@ static struct page *saveable_highmem_page(struct zone *zone, unsigned long pfn) * pages. */ -unsigned int count_highmem_pages(void) +static unsigned int count_highmem_pages(void) { struct zone *zone; unsigned int n = 0; - for_each_zone(zone) { + for_each_populated_zone(zone) { unsigned long pfn, max_zone_pfn; if (!is_highmem(zone)) @@ -902,13 +910,13 @@ static struct page *saveable_page(struct zone *zone, unsigned long pfn) * pages. */ -unsigned int count_data_pages(void) +static unsigned int count_data_pages(void) { struct zone *zone; unsigned long pfn, max_zone_pfn; unsigned int n = 0; - for_each_zone(zone) { + for_each_populated_zone(zone) { if (is_highmem(zone)) continue; @@ -1002,7 +1010,7 @@ copy_data_pages(struct memory_bitmap *copy_bm, struct memory_bitmap *orig_bm) struct zone *zone; unsigned long pfn; - for_each_zone(zone) { + for_each_populated_zone(zone) { unsigned long max_zone_pfn; mark_free_pages(zone); @@ -1025,6 +1033,25 @@ copy_data_pages(struct memory_bitmap *copy_bm, struct memory_bitmap *orig_bm) static unsigned int nr_copy_pages; /* Number of pages needed for saving the original pfns of the image pages */ static unsigned int nr_meta_pages; +/* + * Numbers of normal and highmem page frames allocated for hibernation image + * before suspending devices. + */ +unsigned int alloc_normal, alloc_highmem; +/* + * Memory bitmap used for marking saveable pages (during hibernation) or + * hibernation image pages (during restore) + */ +static struct memory_bitmap orig_bm; +/* + * Memory bitmap used during hibernation for marking allocated page frames that + * will contain copies of saveable pages. During restore it is initially used + * for marking hibernation image pages, but then the set bits from it are + * duplicated in @orig_bm and it is released. On highmem systems it is next + * used for marking "safe" highmem pages, but it has to be reinitialized for + * this purpose. + */ +static struct memory_bitmap copy_bm; /** * swsusp_free - free pages allocated for the suspend. @@ -1038,7 +1065,7 @@ void swsusp_free(void) struct zone *zone; unsigned long pfn, max_zone_pfn; - for_each_zone(zone) { + for_each_populated_zone(zone) { max_zone_pfn = zone->zone_start_pfn + zone->spanned_pages; for (pfn = zone->zone_start_pfn; pfn < max_zone_pfn; pfn++) if (pfn_valid(pfn)) { @@ -1056,6 +1083,286 @@ void swsusp_free(void) nr_meta_pages = 0; restore_pblist = NULL; buffer = NULL; + alloc_normal = 0; + alloc_highmem = 0; +} + +/* Helper functions used for the shrinking of memory. */ + +#define GFP_IMAGE (GFP_KERNEL | __GFP_NOWARN) + +/** + * preallocate_image_pages - Allocate a number of pages for hibernation image + * @nr_pages: Number of page frames to allocate. + * @mask: GFP flags to use for the allocation. + * + * Return value: Number of page frames actually allocated + */ +static unsigned long preallocate_image_pages(unsigned long nr_pages, gfp_t mask) +{ + unsigned long nr_alloc = 0; + + while (nr_pages > 0) { + struct page *page; + + page = alloc_image_page(mask); + if (!page) + break; + memory_bm_set_bit(©_bm, page_to_pfn(page)); + if (PageHighMem(page)) + alloc_highmem++; + else + alloc_normal++; + nr_pages--; + nr_alloc++; + } + + return nr_alloc; +} + +static unsigned long preallocate_image_memory(unsigned long nr_pages) +{ + return preallocate_image_pages(nr_pages, GFP_IMAGE); +} + +#ifdef CONFIG_HIGHMEM +static unsigned long preallocate_image_highmem(unsigned long nr_pages) +{ + return preallocate_image_pages(nr_pages, GFP_IMAGE | __GFP_HIGHMEM); +} + +/** + * __fraction - Compute (an approximation of) x * (multiplier / base) + */ +static unsigned long __fraction(u64 x, u64 multiplier, u64 base) +{ + x *= multiplier; + do_div(x, base); + return (unsigned long)x; +} + +static unsigned long preallocate_highmem_fraction(unsigned long nr_pages, + unsigned long highmem, + unsigned long total) +{ + unsigned long alloc = __fraction(nr_pages, highmem, total); + + return preallocate_image_pages(alloc, GFP_IMAGE | __GFP_HIGHMEM); +} +#else /* CONFIG_HIGHMEM */ +static inline unsigned long preallocate_image_highmem(unsigned long nr_pages) +{ + return 0; +} + +static inline unsigned long preallocate_highmem_fraction(unsigned long nr_pages, + unsigned long highmem, + unsigned long total) +{ + return 0; +} +#endif /* CONFIG_HIGHMEM */ + +/** + * free_unnecessary_pages - Release preallocated pages not needed for the image + */ +static void free_unnecessary_pages(void) +{ + unsigned long save_highmem, to_free_normal, to_free_highmem; + + to_free_normal = alloc_normal - count_data_pages(); + save_highmem = count_highmem_pages(); + if (alloc_highmem > save_highmem) { + to_free_highmem = alloc_highmem - save_highmem; + } else { + to_free_highmem = 0; + to_free_normal -= save_highmem - alloc_highmem; + } + + memory_bm_position_reset(©_bm); + + while (to_free_normal > 0 && to_free_highmem > 0) { + unsigned long pfn = memory_bm_next_pfn(©_bm); + struct page *page = pfn_to_page(pfn); + + if (PageHighMem(page)) { + if (!to_free_highmem) + continue; + to_free_highmem--; + alloc_highmem--; + } else { + if (!to_free_normal) + continue; + to_free_normal--; + alloc_normal--; + } + memory_bm_clear_bit(©_bm, pfn); + swsusp_unset_page_forbidden(page); + swsusp_unset_page_free(page); + __free_page(page); + } +} + +/** + * minimum_image_size - Estimate the minimum acceptable size of an image + * @saveable: Number of saveable pages in the system. + * + * We want to avoid attempting to free too much memory too hard, so estimate the + * minimum acceptable size of a hibernation image to use as the lower limit for + * preallocating memory. + * + * We assume that the minimum image size should be proportional to + * + * [number of saveable pages] - [number of pages that can be freed in theory] + * + * where the second term is the sum of (1) reclaimable slab pages, (2) active + * and (3) inactive anonymouns pages, (4) active and (5) inactive file pages, + * minus mapped file pages. + */ +static unsigned long minimum_image_size(unsigned long saveable) +{ + unsigned long size; + + size = global_page_state(NR_SLAB_RECLAIMABLE) + + global_page_state(NR_ACTIVE_ANON) + + global_page_state(NR_INACTIVE_ANON) + + global_page_state(NR_ACTIVE_FILE) + + global_page_state(NR_INACTIVE_FILE) + - global_page_state(NR_FILE_MAPPED); + + return saveable <= size ? 0 : saveable - size; +} + +/** + * hibernate_preallocate_memory - Preallocate memory for hibernation image + * + * To create a hibernation image it is necessary to make a copy of every page + * frame in use. We also need a number of page frames to be free during + * hibernation for allocations made while saving the image and for device + * drivers, in case they need to allocate memory from their hibernation + * callbacks (these two numbers are given by PAGES_FOR_IO and SPARE_PAGES, + * respectively, both of which are rough estimates). To make this happen, we + * compute the total number of available page frames and allocate at least + * + * ([page frames total] + PAGES_FOR_IO + [metadata pages]) / 2 + 2 * SPARE_PAGES + * + * of them, which corresponds to the maximum size of a hibernation image. + * + * If image_size is set below the number following from the above formula, + * the preallocation of memory is continued until the total number of saveable + * pages in the system is below the requested image size or the minimum + * acceptable image size returned by minimum_image_size(), whichever is greater. + */ +int hibernate_preallocate_memory(void) +{ + struct zone *zone; + unsigned long saveable, size, max_size, count, highmem, pages = 0; + unsigned long alloc, save_highmem, pages_highmem; + struct timeval start, stop; + int error; + + printk(KERN_INFO "PM: Preallocating image memory... "); + do_gettimeofday(&start); + + error = memory_bm_create(&orig_bm, GFP_IMAGE, PG_ANY); + if (error) + goto err_out; + + error = memory_bm_create(©_bm, GFP_IMAGE, PG_ANY); + if (error) + goto err_out; + + alloc_normal = 0; + alloc_highmem = 0; + + /* Count the number of saveable data pages. */ + save_highmem = count_highmem_pages(); + saveable = count_data_pages(); + + /* + * Compute the total number of page frames we can use (count) and the + * number of pages needed for image metadata (size). + */ + count = saveable; + saveable += save_highmem; + highmem = save_highmem; + size = 0; + for_each_populated_zone(zone) { + size += snapshot_additional_pages(zone); + if (is_highmem(zone)) + highmem += zone_page_state(zone, NR_FREE_PAGES); + else + count += zone_page_state(zone, NR_FREE_PAGES); + } + count += highmem; + count -= totalreserve_pages; + + /* Compute the maximum number of saveable pages to leave in memory. */ + max_size = (count - (size + PAGES_FOR_IO)) / 2 - 2 * SPARE_PAGES; + size = DIV_ROUND_UP(image_size, PAGE_SIZE); + if (size > max_size) + size = max_size; + /* + * If the maximum is not less than the current number of saveable pages + * in memory, allocate page frames for the image and we're done. + */ + if (size >= saveable) { + pages = preallocate_image_highmem(save_highmem); + pages += preallocate_image_memory(saveable - pages); + goto out; + } + + /* Estimate the minimum size of the image. */ + pages = minimum_image_size(saveable); + if (size < pages) + size = min_t(unsigned long, pages, max_size); + + /* + * Let the memory management subsystem know that we're going to need a + * large number of page frames to allocate and make it free some memory. + * NOTE: If this is not done, performance will be hurt badly in some + * test cases. + */ + shrink_all_memory(saveable - size); + + /* + * The number of saveable pages in memory was too high, so apply some + * pressure to decrease it. First, make room for the largest possible + * image and fail if that doesn't work. Next, try to decrease the size + * of the image as much as indicated by 'size' using allocations from + * highmem and non-highmem zones separately. + */ + pages_highmem = preallocate_image_highmem(highmem / 2); + alloc = (count - max_size) - pages_highmem; + pages = preallocate_image_memory(alloc); + if (pages < alloc) + goto err_out; + size = max_size - size; + alloc = size; + size = preallocate_highmem_fraction(size, highmem, count); + pages_highmem += size; + alloc -= size; + pages += preallocate_image_memory(alloc); + pages += pages_highmem; + + /* + * We only need as many page frames for the image as there are saveable + * pages in memory, but we have allocated more. Release the excessive + * ones now. + */ + free_unnecessary_pages(); + + out: + do_gettimeofday(&stop); + printk(KERN_CONT "done (allocated %lu pages)\n", pages); + swsusp_show_speed(&start, &stop, pages, "Allocated"); + + return 0; + + err_out: + printk(KERN_CONT "\n"); + swsusp_free(); + return -ENOMEM; } #ifdef CONFIG_HIGHMEM @@ -1066,7 +1373,7 @@ void swsusp_free(void) static unsigned int count_pages_for_highmem(unsigned int nr_highmem) { - unsigned int free_highmem = count_free_highmem_pages(); + unsigned int free_highmem = count_free_highmem_pages() + alloc_highmem; if (free_highmem >= nr_highmem) nr_highmem = 0; @@ -1088,19 +1395,17 @@ count_pages_for_highmem(unsigned int nr_highmem) { return 0; } static int enough_free_mem(unsigned int nr_pages, unsigned int nr_highmem) { struct zone *zone; - unsigned int free = 0, meta = 0; + unsigned int free = alloc_normal; - for_each_zone(zone) { - meta += snapshot_additional_pages(zone); + for_each_populated_zone(zone) if (!is_highmem(zone)) free += zone_page_state(zone, NR_FREE_PAGES); - } nr_pages += count_pages_for_highmem(nr_highmem); - pr_debug("PM: Normal pages needed: %u + %u + %u, available pages: %u\n", - nr_pages, PAGES_FOR_IO, meta, free); + pr_debug("PM: Normal pages needed: %u + %u, available pages: %u\n", + nr_pages, PAGES_FOR_IO, free); - return free > nr_pages + PAGES_FOR_IO + meta; + return free > nr_pages + PAGES_FOR_IO; } #ifdef CONFIG_HIGHMEM @@ -1122,7 +1427,7 @@ static inline int get_highmem_buffer(int safe_needed) */ static inline unsigned int -alloc_highmem_image_pages(struct memory_bitmap *bm, unsigned int nr_highmem) +alloc_highmem_pages(struct memory_bitmap *bm, unsigned int nr_highmem) { unsigned int to_alloc = count_free_highmem_pages(); @@ -1142,7 +1447,7 @@ alloc_highmem_image_pages(struct memory_bitmap *bm, unsigned int nr_highmem) static inline int get_highmem_buffer(int safe_needed) { return 0; } static inline unsigned int -alloc_highmem_image_pages(struct memory_bitmap *bm, unsigned int n) { return 0; } +alloc_highmem_pages(struct memory_bitmap *bm, unsigned int n) { return 0; } #endif /* CONFIG_HIGHMEM */ /** @@ -1161,51 +1466,36 @@ static int swsusp_alloc(struct memory_bitmap *orig_bm, struct memory_bitmap *copy_bm, unsigned int nr_pages, unsigned int nr_highmem) { - int error; - - error = memory_bm_create(orig_bm, GFP_ATOMIC | __GFP_COLD, PG_ANY); - if (error) - goto Free; - - error = memory_bm_create(copy_bm, GFP_ATOMIC | __GFP_COLD, PG_ANY); - if (error) - goto Free; + int error = 0; if (nr_highmem > 0) { error = get_highmem_buffer(PG_ANY); if (error) - goto Free; - - nr_pages += alloc_highmem_image_pages(copy_bm, nr_highmem); + goto err_out; + if (nr_highmem > alloc_highmem) { + nr_highmem -= alloc_highmem; + nr_pages += alloc_highmem_pages(copy_bm, nr_highmem); + } } - while (nr_pages-- > 0) { - struct page *page = alloc_image_page(GFP_ATOMIC | __GFP_COLD); - - if (!page) - goto Free; + if (nr_pages > alloc_normal) { + nr_pages -= alloc_normal; + while (nr_pages-- > 0) { + struct page *page; - memory_bm_set_bit(copy_bm, page_to_pfn(page)); + page = alloc_image_page(GFP_ATOMIC | __GFP_COLD); + if (!page) + goto err_out; + memory_bm_set_bit(copy_bm, page_to_pfn(page)); + } } + return 0; - Free: + err_out: swsusp_free(); - return -ENOMEM; + return error; } -/* Memory bitmap used for marking saveable pages (during suspend) or the - * suspend image pages (during resume) - */ -static struct memory_bitmap orig_bm; -/* Memory bitmap used on suspend for marking allocated pages that will contain - * the copies of saveable pages. During resume it is initially used for - * marking the suspend image pages, but then its set bits are duplicated in - * @orig_bm and it is released. Next, on systems with high memory, it may be - * used for marking "safe" highmem pages, but it has to be reinitialized for - * this purpose. - */ -static struct memory_bitmap copy_bm; - asmlinkage int swsusp_save(void) { unsigned int nr_pages, nr_highmem; @@ -1398,7 +1688,7 @@ static int mark_unsafe_pages(struct memory_bitmap *bm) unsigned long pfn, max_zone_pfn; /* Clear page flags */ - for_each_zone(zone) { + for_each_populated_zone(zone) { max_zone_pfn = zone->zone_start_pfn + zone->spanned_pages; for (pfn = zone->zone_start_pfn; pfn < max_zone_pfn; pfn++) if (pfn_valid(pfn)) diff --git a/kernel/power/suspend.c b/kernel/power/suspend.c new file mode 100644 index 00000000000..6f10dfc2d3e --- /dev/null +++ b/kernel/power/suspend.c @@ -0,0 +1,300 @@ +/* + * kernel/power/suspend.c - Suspend to RAM and standby functionality. + * + * Copyright (c) 2003 Patrick Mochel + * Copyright (c) 2003 Open Source Development Lab + * Copyright (c) 2009 Rafael J. Wysocki <rjw@sisk.pl>, Novell Inc. + * + * This file is released under the GPLv2. + */ + +#include <linux/string.h> +#include <linux/delay.h> +#include <linux/errno.h> +#include <linux/init.h> +#include <linux/console.h> +#include <linux/cpu.h> +#include <linux/syscalls.h> + +#include "power.h" + +const char *const pm_states[PM_SUSPEND_MAX] = { + [PM_SUSPEND_STANDBY] = "standby", + [PM_SUSPEND_MEM] = "mem", +}; + +static struct platform_suspend_ops *suspend_ops; + +/** + * suspend_set_ops - Set the global suspend method table. + * @ops: Pointer to ops structure. + */ +void suspend_set_ops(struct platform_suspend_ops *ops) +{ + mutex_lock(&pm_mutex); + suspend_ops = ops; + mutex_unlock(&pm_mutex); +} + +bool valid_state(suspend_state_t state) +{ + /* + * All states need lowlevel support and need to be valid to the lowlevel + * implementation, no valid callback implies that none are valid. + */ + return suspend_ops && suspend_ops->valid && suspend_ops->valid(state); +} + +/** + * suspend_valid_only_mem - generic memory-only valid callback + * + * Platform drivers that implement mem suspend only and only need + * to check for that in their .valid callback can use this instead + * of rolling their own .valid callback. + */ +int suspend_valid_only_mem(suspend_state_t state) +{ + return state == PM_SUSPEND_MEM; +} + +static int suspend_test(int level) +{ +#ifdef CONFIG_PM_DEBUG + if (pm_test_level == level) { + printk(KERN_INFO "suspend debug: Waiting for 5 seconds.\n"); + mdelay(5000); + return 1; + } +#endif /* !CONFIG_PM_DEBUG */ + return 0; +} + +/** + * suspend_prepare - Do prep work before entering low-power state. + * + * This is common code that is called for each state that we're entering. + * Run suspend notifiers, allocate a console and stop all processes. + */ +static int suspend_prepare(void) +{ + int error; + + if (!suspend_ops || !suspend_ops->enter) + return -EPERM; + + pm_prepare_console(); + + error = pm_notifier_call_chain(PM_SUSPEND_PREPARE); + if (error) + goto Finish; + + error = usermodehelper_disable(); + if (error) + goto Finish; + + error = suspend_freeze_processes(); + if (!error) + return 0; + + suspend_thaw_processes(); + usermodehelper_enable(); + Finish: + pm_notifier_call_chain(PM_POST_SUSPEND); + pm_restore_console(); + return error; +} + +/* default implementation */ +void __attribute__ ((weak)) arch_suspend_disable_irqs(void) +{ + local_irq_disable(); +} + +/* default implementation */ +void __attribute__ ((weak)) arch_suspend_enable_irqs(void) +{ + local_irq_enable(); +} + +/** + * suspend_enter - enter the desired system sleep state. + * @state: state to enter + * + * This function should be called after devices have been suspended. + */ +static int suspend_enter(suspend_state_t state) +{ + int error; + + if (suspend_ops->prepare) { + error = suspend_ops->prepare(); + if (error) + return error; + } + + error = dpm_suspend_noirq(PMSG_SUSPEND); + if (error) { + printk(KERN_ERR "PM: Some devices failed to power down\n"); + goto Platfrom_finish; + } + + if (suspend_ops->prepare_late) { + error = suspend_ops->prepare_late(); + if (error) + goto Power_up_devices; + } + + if (suspend_test(TEST_PLATFORM)) + goto Platform_wake; + + error = disable_nonboot_cpus(); + if (error || suspend_test(TEST_CPUS)) + goto Enable_cpus; + + arch_suspend_disable_irqs(); + BUG_ON(!irqs_disabled()); + + error = sysdev_suspend(PMSG_SUSPEND); + if (!error) { + if (!suspend_test(TEST_CORE)) + error = suspend_ops->enter(state); + sysdev_resume(); + } + + arch_suspend_enable_irqs(); + BUG_ON(irqs_disabled()); + + Enable_cpus: + enable_nonboot_cpus(); + + Platform_wake: + if (suspend_ops->wake) + suspend_ops->wake(); + + Power_up_devices: + dpm_resume_noirq(PMSG_RESUME); + + Platfrom_finish: + if (suspend_ops->finish) + suspend_ops->finish(); + + return error; +} + +/** + * suspend_devices_and_enter - suspend devices and enter the desired system + * sleep state. + * @state: state to enter + */ +int suspend_devices_and_enter(suspend_state_t state) +{ + int error; + + if (!suspend_ops) + return -ENOSYS; + + if (suspend_ops->begin) { + error = suspend_ops->begin(state); + if (error) + goto Close; + } + suspend_console(); + suspend_test_start(); + error = dpm_suspend_start(PMSG_SUSPEND); + if (error) { + printk(KERN_ERR "PM: Some devices failed to suspend\n"); + goto Recover_platform; + } + suspend_test_finish("suspend devices"); + if (suspend_test(TEST_DEVICES)) + goto Recover_platform; + + suspend_enter(state); + + Resume_devices: + suspend_test_start(); + dpm_resume_end(PMSG_RESUME); + suspend_test_finish("resume devices"); + resume_console(); + Close: + if (suspend_ops->end) + suspend_ops->end(); + return error; + + Recover_platform: + if (suspend_ops->recover) + suspend_ops->recover(); + goto Resume_devices; +} + +/** + * suspend_finish - Do final work before exiting suspend sequence. + * + * Call platform code to clean up, restart processes, and free the + * console that we've allocated. This is not called for suspend-to-disk. + */ +static void suspend_finish(void) +{ + suspend_thaw_processes(); + usermodehelper_enable(); + pm_notifier_call_chain(PM_POST_SUSPEND); + pm_restore_console(); +} + +/** + * enter_state - Do common work of entering low-power state. + * @state: pm_state structure for state we're entering. + * + * Make sure we're the only ones trying to enter a sleep state. Fail + * if someone has beat us to it, since we don't want anything weird to + * happen when we wake up. + * Then, do the setup for suspend, enter the state, and cleaup (after + * we've woken up). + */ +int enter_state(suspend_state_t state) +{ + int error; + + if (!valid_state(state)) + return -ENODEV; + + if (!mutex_trylock(&pm_mutex)) + return -EBUSY; + + printk(KERN_INFO "PM: Syncing filesystems ... "); + sys_sync(); + printk("done.\n"); + + pr_debug("PM: Preparing system for %s sleep\n", pm_states[state]); + error = suspend_prepare(); + if (error) + goto Unlock; + + if (suspend_test(TEST_FREEZER)) + goto Finish; + + pr_debug("PM: Entering %s sleep\n", pm_states[state]); + error = suspend_devices_and_enter(state); + + Finish: + pr_debug("PM: Finishing wakeup.\n"); + suspend_finish(); + Unlock: + mutex_unlock(&pm_mutex); + return error; +} + +/** + * pm_suspend - Externally visible function for suspending system. + * @state: Enumerated value of state to enter. + * + * Determine whether or not value is within range, get state + * structure, and enter (above). + */ +int pm_suspend(suspend_state_t state) +{ + if (state > PM_SUSPEND_ON && state <= PM_SUSPEND_MAX) + return enter_state(state); + return -EINVAL; +} +EXPORT_SYMBOL(pm_suspend); diff --git a/kernel/power/suspend_test.c b/kernel/power/suspend_test.c new file mode 100644 index 00000000000..25596e450ac --- /dev/null +++ b/kernel/power/suspend_test.c @@ -0,0 +1,188 @@ +/* + * kernel/power/suspend_test.c - Suspend to RAM and standby test facility. + * + * Copyright (c) 2009 Pavel Machek <pavel@ucw.cz> + * + * This file is released under the GPLv2. + */ + +#include <linux/init.h> +#include <linux/rtc.h> + +#include "power.h" + +/* + * We test the system suspend code by setting an RTC wakealarm a short + * time in the future, then suspending. Suspending the devices won't + * normally take long ... some systems only need a few milliseconds. + * + * The time it takes is system-specific though, so when we test this + * during system bootup we allow a LOT of time. + */ +#define TEST_SUSPEND_SECONDS 10 + +static unsigned long suspend_test_start_time; + +void suspend_test_start(void) +{ + /* FIXME Use better timebase than "jiffies", ideally a clocksource. + * What we want is a hardware counter that will work correctly even + * during the irqs-are-off stages of the suspend/resume cycle... + */ + suspend_test_start_time = jiffies; +} + +void suspend_test_finish(const char *label) +{ + long nj = jiffies - suspend_test_start_time; + unsigned msec; + + msec = jiffies_to_msecs(abs(nj)); + pr_info("PM: %s took %d.%03d seconds\n", label, + msec / 1000, msec % 1000); + + /* Warning on suspend means the RTC alarm period needs to be + * larger -- the system was sooo slooowwww to suspend that the + * alarm (should have) fired before the system went to sleep! + * + * Warning on either suspend or resume also means the system + * has some performance issues. The stack dump of a WARN_ON + * is more likely to get the right attention than a printk... + */ + WARN(msec > (TEST_SUSPEND_SECONDS * 1000), + "Component: %s, time: %u\n", label, msec); +} + +/* + * To test system suspend, we need a hands-off mechanism to resume the + * system. RTCs wake alarms are a common self-contained mechanism. + */ + +static void __init test_wakealarm(struct rtc_device *rtc, suspend_state_t state) +{ + static char err_readtime[] __initdata = + KERN_ERR "PM: can't read %s time, err %d\n"; + static char err_wakealarm [] __initdata = + KERN_ERR "PM: can't set %s wakealarm, err %d\n"; + static char err_suspend[] __initdata = + KERN_ERR "PM: suspend test failed, error %d\n"; + static char info_test[] __initdata = + KERN_INFO "PM: test RTC wakeup from '%s' suspend\n"; + + unsigned long now; + struct rtc_wkalrm alm; + int status; + + /* this may fail if the RTC hasn't been initialized */ + status = rtc_read_time(rtc, &alm.time); + if (status < 0) { + printk(err_readtime, dev_name(&rtc->dev), status); + return; + } + rtc_tm_to_time(&alm.time, &now); + + memset(&alm, 0, sizeof alm); + rtc_time_to_tm(now + TEST_SUSPEND_SECONDS, &alm.time); + alm.enabled = true; + + status = rtc_set_alarm(rtc, &alm); + if (status < 0) { + printk(err_wakealarm, dev_name(&rtc->dev), status); + return; + } + + if (state == PM_SUSPEND_MEM) { + printk(info_test, pm_states[state]); + status = pm_suspend(state); + if (status == -ENODEV) + state = PM_SUSPEND_STANDBY; + } + if (state == PM_SUSPEND_STANDBY) { + printk(info_test, pm_states[state]); + status = pm_suspend(state); + } + if (status < 0) + printk(err_suspend, status); + + /* Some platforms can't detect that the alarm triggered the + * wakeup, or (accordingly) disable it after it afterwards. + * It's supposed to give oneshot behavior; cope. + */ + alm.enabled = false; + rtc_set_alarm(rtc, &alm); +} + +static int __init has_wakealarm(struct device *dev, void *name_ptr) +{ + struct rtc_device *candidate = to_rtc_device(dev); + + if (!candidate->ops->set_alarm) + return 0; + if (!device_may_wakeup(candidate->dev.parent)) + return 0; + + *(const char **)name_ptr = dev_name(dev); + return 1; +} + +/* + * Kernel options like "test_suspend=mem" force suspend/resume sanity tests + * at startup time. They're normally disabled, for faster boot and because + * we can't know which states really work on this particular system. + */ +static suspend_state_t test_state __initdata = PM_SUSPEND_ON; + +static char warn_bad_state[] __initdata = + KERN_WARNING "PM: can't test '%s' suspend state\n"; + +static int __init setup_test_suspend(char *value) +{ + unsigned i; + + /* "=mem" ==> "mem" */ + value++; + for (i = 0; i < PM_SUSPEND_MAX; i++) { + if (!pm_states[i]) + continue; + if (strcmp(pm_states[i], value) != 0) + continue; + test_state = (__force suspend_state_t) i; + return 0; + } + printk(warn_bad_state, value); + return 0; +} +__setup("test_suspend", setup_test_suspend); + +static int __init test_suspend(void) +{ + static char warn_no_rtc[] __initdata = + KERN_WARNING "PM: no wakealarm-capable RTC driver is ready\n"; + + char *pony = NULL; + struct rtc_device *rtc = NULL; + + /* PM is initialized by now; is that state testable? */ + if (test_state == PM_SUSPEND_ON) + goto done; + if (!valid_state(test_state)) { + printk(warn_bad_state, pm_states[test_state]); + goto done; + } + + /* RTCs have initialized by now too ... can we use one? */ + class_find_device(rtc_class, NULL, &pony, has_wakealarm); + if (pony) + rtc = rtc_class_open(pony); + if (!rtc) { + printk(warn_no_rtc); + goto done; + } + + /* go for it */ + test_wakealarm(rtc, test_state); + rtc_class_close(rtc); +done: + return 0; +} +late_initcall(test_suspend); diff --git a/kernel/power/swap.c b/kernel/power/swap.c index 505f319e489..09b2b0ae9e9 100644 --- a/kernel/power/swap.c +++ b/kernel/power/swap.c @@ -13,7 +13,6 @@ #include <linux/module.h> #include <linux/file.h> -#include <linux/utsname.h> #include <linux/delay.h> #include <linux/bitops.h> #include <linux/genhd.h> @@ -39,6 +38,107 @@ struct swsusp_header { static struct swsusp_header *swsusp_header; +/** + * The following functions are used for tracing the allocated + * swap pages, so that they can be freed in case of an error. + */ + +struct swsusp_extent { + struct rb_node node; + unsigned long start; + unsigned long end; +}; + +static struct rb_root swsusp_extents = RB_ROOT; + +static int swsusp_extents_insert(unsigned long swap_offset) +{ + struct rb_node **new = &(swsusp_extents.rb_node); + struct rb_node *parent = NULL; + struct swsusp_extent *ext; + + /* Figure out where to put the new node */ + while (*new) { + ext = container_of(*new, struct swsusp_extent, node); + parent = *new; + if (swap_offset < ext->start) { + /* Try to merge */ + if (swap_offset == ext->start - 1) { + ext->start--; + return 0; + } + new = &((*new)->rb_left); + } else if (swap_offset > ext->end) { + /* Try to merge */ + if (swap_offset == ext->end + 1) { + ext->end++; + return 0; + } + new = &((*new)->rb_right); + } else { + /* It already is in the tree */ + return -EINVAL; + } + } + /* Add the new node and rebalance the tree. */ + ext = kzalloc(sizeof(struct swsusp_extent), GFP_KERNEL); + if (!ext) + return -ENOMEM; + + ext->start = swap_offset; + ext->end = swap_offset; + rb_link_node(&ext->node, parent, new); + rb_insert_color(&ext->node, &swsusp_extents); + return 0; +} + +/** + * alloc_swapdev_block - allocate a swap page and register that it has + * been allocated, so that it can be freed in case of an error. + */ + +sector_t alloc_swapdev_block(int swap) +{ + unsigned long offset; + + offset = swp_offset(get_swap_page_of_type(swap)); + if (offset) { + if (swsusp_extents_insert(offset)) + swap_free(swp_entry(swap, offset)); + else + return swapdev_block(swap, offset); + } + return 0; +} + +/** + * free_all_swap_pages - free swap pages allocated for saving image data. + * It also frees the extents used to register which swap entres had been + * allocated. + */ + +void free_all_swap_pages(int swap) +{ + struct rb_node *node; + + while ((node = swsusp_extents.rb_node)) { + struct swsusp_extent *ext; + unsigned long offset; + + ext = container_of(node, struct swsusp_extent, node); + rb_erase(node, &swsusp_extents); + for (offset = ext->start; offset <= ext->end; offset++) + swap_free(swp_entry(swap, offset)); + + kfree(ext); + } +} + +int swsusp_swap_in_use(void) +{ + return (swsusp_extents.rb_node != NULL); +} + /* * General things */ @@ -64,8 +164,6 @@ static int submit(int rw, pgoff_t page_off, struct page *page, struct bio *bio; bio = bio_alloc(__GFP_WAIT | __GFP_HIGH, 1); - if (!bio) - return -ENOMEM; bio->bi_sector = page_off * (PAGE_SIZE >> 9); bio->bi_bdev = resume_bdev; bio->bi_end_io = end_swap_bio_read; @@ -317,7 +415,6 @@ static int save_image(struct swap_map_handle *handle, { unsigned int m; int ret; - int error = 0; int nr_pages; int err2; struct bio *bio; @@ -332,26 +429,27 @@ static int save_image(struct swap_map_handle *handle, nr_pages = 0; bio = NULL; do_gettimeofday(&start); - do { + while (1) { ret = snapshot_read_next(snapshot, PAGE_SIZE); - if (ret > 0) { - error = swap_write_page(handle, data_of(*snapshot), - &bio); - if (error) - break; - if (!(nr_pages % m)) - printk("\b\b\b\b%3d%%", nr_pages / m); - nr_pages++; - } - } while (ret > 0); + if (ret <= 0) + break; + ret = swap_write_page(handle, data_of(*snapshot), &bio); + if (ret) + break; + if (!(nr_pages % m)) + printk(KERN_CONT "\b\b\b\b%3d%%", nr_pages / m); + nr_pages++; + } err2 = wait_on_bio_chain(&bio); do_gettimeofday(&stop); - if (!error) - error = err2; - if (!error) - printk("\b\b\b\bdone\n"); + if (!ret) + ret = err2; + if (!ret) + printk(KERN_CONT "\b\b\b\bdone\n"); + else + printk(KERN_CONT "\n"); swsusp_show_speed(&start, &stop, nr_to_write, "Wrote"); - return error; + return ret; } /** @@ -539,7 +637,8 @@ static int load_image(struct swap_map_handle *handle, snapshot_write_finalize(snapshot); if (!snapshot_image_loaded(snapshot)) error = -ENODATA; - } + } else + printk("\n"); swsusp_show_speed(&start, &stop, nr_to_read, "Read"); return error; } @@ -575,8 +674,6 @@ int swsusp_read(unsigned int *flags_p) error = load_image(&handle, &snapshot, header->pages - 1); release_swap_reader(&handle); - blkdev_put(resume_bdev, FMODE_READ); - if (!error) pr_debug("PM: Image successfully loaded\n"); else @@ -599,7 +696,7 @@ int swsusp_check(void) error = bio_read_page(swsusp_resume_block, swsusp_header, NULL); if (error) - return error; + goto put; if (!memcmp(SWSUSP_SIG, swsusp_header->sig, 10)) { memcpy(swsusp_header->sig, swsusp_header->orig_sig, 10); @@ -607,8 +704,10 @@ int swsusp_check(void) error = bio_write_page(swsusp_resume_block, swsusp_header, NULL); } else { - return -EINVAL; + error = -EINVAL; } + +put: if (error) blkdev_put(resume_bdev, FMODE_READ); else diff --git a/kernel/power/swsusp.c b/kernel/power/swsusp.c index 78c35047586..5b3601bd189 100644 --- a/kernel/power/swsusp.c +++ b/kernel/power/swsusp.c @@ -55,332 +55,4 @@ #include "power.h" -/* - * Preferred image size in bytes (tunable via /sys/power/image_size). - * When it is set to N, swsusp will do its best to ensure the image - * size will not exceed N bytes, but if that is impossible, it will - * try to create the smallest image possible. - */ -unsigned long image_size = 500 * 1024 * 1024; - int in_suspend __nosavedata = 0; - -/** - * The following functions are used for tracing the allocated - * swap pages, so that they can be freed in case of an error. - */ - -struct swsusp_extent { - struct rb_node node; - unsigned long start; - unsigned long end; -}; - -static struct rb_root swsusp_extents = RB_ROOT; - -static int swsusp_extents_insert(unsigned long swap_offset) -{ - struct rb_node **new = &(swsusp_extents.rb_node); - struct rb_node *parent = NULL; - struct swsusp_extent *ext; - - /* Figure out where to put the new node */ - while (*new) { - ext = container_of(*new, struct swsusp_extent, node); - parent = *new; - if (swap_offset < ext->start) { - /* Try to merge */ - if (swap_offset == ext->start - 1) { - ext->start--; - return 0; - } - new = &((*new)->rb_left); - } else if (swap_offset > ext->end) { - /* Try to merge */ - if (swap_offset == ext->end + 1) { - ext->end++; - return 0; - } - new = &((*new)->rb_right); - } else { - /* It already is in the tree */ - return -EINVAL; - } - } - /* Add the new node and rebalance the tree. */ - ext = kzalloc(sizeof(struct swsusp_extent), GFP_KERNEL); - if (!ext) - return -ENOMEM; - - ext->start = swap_offset; - ext->end = swap_offset; - rb_link_node(&ext->node, parent, new); - rb_insert_color(&ext->node, &swsusp_extents); - return 0; -} - -/** - * alloc_swapdev_block - allocate a swap page and register that it has - * been allocated, so that it can be freed in case of an error. - */ - -sector_t alloc_swapdev_block(int swap) -{ - unsigned long offset; - - offset = swp_offset(get_swap_page_of_type(swap)); - if (offset) { - if (swsusp_extents_insert(offset)) - swap_free(swp_entry(swap, offset)); - else - return swapdev_block(swap, offset); - } - return 0; -} - -/** - * free_all_swap_pages - free swap pages allocated for saving image data. - * It also frees the extents used to register which swap entres had been - * allocated. - */ - -void free_all_swap_pages(int swap) -{ - struct rb_node *node; - - while ((node = swsusp_extents.rb_node)) { - struct swsusp_extent *ext; - unsigned long offset; - - ext = container_of(node, struct swsusp_extent, node); - rb_erase(node, &swsusp_extents); - for (offset = ext->start; offset <= ext->end; offset++) - swap_free(swp_entry(swap, offset)); - - kfree(ext); - } -} - -int swsusp_swap_in_use(void) -{ - return (swsusp_extents.rb_node != NULL); -} - -/** - * swsusp_show_speed - print the time elapsed between two events represented by - * @start and @stop - * - * @nr_pages - number of pages processed between @start and @stop - * @msg - introductory message to print - */ - -void swsusp_show_speed(struct timeval *start, struct timeval *stop, - unsigned nr_pages, char *msg) -{ - s64 elapsed_centisecs64; - int centisecs; - int k; - int kps; - - elapsed_centisecs64 = timeval_to_ns(stop) - timeval_to_ns(start); - do_div(elapsed_centisecs64, NSEC_PER_SEC / 100); - centisecs = elapsed_centisecs64; - if (centisecs == 0) - centisecs = 1; /* avoid div-by-zero */ - k = nr_pages * (PAGE_SIZE / 1024); - kps = (k * 100) / centisecs; - printk(KERN_INFO "PM: %s %d kbytes in %d.%02d seconds (%d.%02d MB/s)\n", - msg, k, - centisecs / 100, centisecs % 100, - kps / 1000, (kps % 1000) / 10); -} - -/** - * swsusp_shrink_memory - Try to free as much memory as needed - * - * ... but do not OOM-kill anyone - * - * Notice: all userland should be stopped before it is called, or - * livelock is possible. - */ - -#define SHRINK_BITE 10000 -static inline unsigned long __shrink_memory(long tmp) -{ - if (tmp > SHRINK_BITE) - tmp = SHRINK_BITE; - return shrink_all_memory(tmp); -} - -int swsusp_shrink_memory(void) -{ - long tmp; - struct zone *zone; - unsigned long pages = 0; - unsigned int i = 0; - char *p = "-\\|/"; - struct timeval start, stop; - - printk(KERN_INFO "PM: Shrinking memory... "); - do_gettimeofday(&start); - do { - long size, highmem_size; - - highmem_size = count_highmem_pages(); - size = count_data_pages() + PAGES_FOR_IO + SPARE_PAGES; - tmp = size; - size += highmem_size; - for_each_populated_zone(zone) { - tmp += snapshot_additional_pages(zone); - if (is_highmem(zone)) { - highmem_size -= - zone_page_state(zone, NR_FREE_PAGES); - } else { - tmp -= zone_page_state(zone, NR_FREE_PAGES); - tmp += zone->lowmem_reserve[ZONE_NORMAL]; - } - } - - if (highmem_size < 0) - highmem_size = 0; - - tmp += highmem_size; - if (tmp > 0) { - tmp = __shrink_memory(tmp); - if (!tmp) - return -ENOMEM; - pages += tmp; - } else if (size > image_size / PAGE_SIZE) { - tmp = __shrink_memory(size - (image_size / PAGE_SIZE)); - pages += tmp; - } - printk("\b%c", p[i++%4]); - } while (tmp > 0); - do_gettimeofday(&stop); - printk("\bdone (%lu pages freed)\n", pages); - swsusp_show_speed(&start, &stop, pages, "Freed"); - - return 0; -} - -/* - * Platforms, like ACPI, may want us to save some memory used by them during - * hibernation and to restore the contents of this memory during the subsequent - * resume. The code below implements a mechanism allowing us to do that. - */ - -struct nvs_page { - unsigned long phys_start; - unsigned int size; - void *kaddr; - void *data; - struct list_head node; -}; - -static LIST_HEAD(nvs_list); - -/** - * hibernate_nvs_register - register platform NVS memory region to save - * @start - physical address of the region - * @size - size of the region - * - * The NVS region need not be page-aligned (both ends) and we arrange - * things so that the data from page-aligned addresses in this region will - * be copied into separate RAM pages. - */ -int hibernate_nvs_register(unsigned long start, unsigned long size) -{ - struct nvs_page *entry, *next; - - while (size > 0) { - unsigned int nr_bytes; - - entry = kzalloc(sizeof(struct nvs_page), GFP_KERNEL); - if (!entry) - goto Error; - - list_add_tail(&entry->node, &nvs_list); - entry->phys_start = start; - nr_bytes = PAGE_SIZE - (start & ~PAGE_MASK); - entry->size = (size < nr_bytes) ? size : nr_bytes; - - start += entry->size; - size -= entry->size; - } - return 0; - - Error: - list_for_each_entry_safe(entry, next, &nvs_list, node) { - list_del(&entry->node); - kfree(entry); - } - return -ENOMEM; -} - -/** - * hibernate_nvs_free - free data pages allocated for saving NVS regions - */ -void hibernate_nvs_free(void) -{ - struct nvs_page *entry; - - list_for_each_entry(entry, &nvs_list, node) - if (entry->data) { - free_page((unsigned long)entry->data); - entry->data = NULL; - if (entry->kaddr) { - iounmap(entry->kaddr); - entry->kaddr = NULL; - } - } -} - -/** - * hibernate_nvs_alloc - allocate memory necessary for saving NVS regions - */ -int hibernate_nvs_alloc(void) -{ - struct nvs_page *entry; - - list_for_each_entry(entry, &nvs_list, node) { - entry->data = (void *)__get_free_page(GFP_KERNEL); - if (!entry->data) { - hibernate_nvs_free(); - return -ENOMEM; - } - } - return 0; -} - -/** - * hibernate_nvs_save - save NVS memory regions - */ -void hibernate_nvs_save(void) -{ - struct nvs_page *entry; - - printk(KERN_INFO "PM: Saving platform NVS memory\n"); - - list_for_each_entry(entry, &nvs_list, node) - if (entry->data) { - entry->kaddr = ioremap(entry->phys_start, entry->size); - memcpy(entry->data, entry->kaddr, entry->size); - } -} - -/** - * hibernate_nvs_restore - restore NVS memory regions - * - * This function is going to be called with interrupts disabled, so it - * cannot iounmap the virtual addresses used to access the NVS region. - */ -void hibernate_nvs_restore(void) -{ - struct nvs_page *entry; - - printk(KERN_INFO "PM: Restoring platform NVS memory\n"); - - list_for_each_entry(entry, &nvs_list, node) - if (entry->data) - memcpy(entry->kaddr, entry->data, entry->size); -} diff --git a/kernel/power/user.c b/kernel/power/user.c index 6c85359364f..bf0014d6a5f 100644 --- a/kernel/power/user.c +++ b/kernel/power/user.c @@ -23,7 +23,7 @@ #include <linux/console.h> #include <linux/cpu.h> #include <linux/freezer.h> -#include <linux/smp_lock.h> +#include <scsi/scsi_scan.h> #include <asm/uaccess.h> @@ -92,6 +92,7 @@ static int snapshot_open(struct inode *inode, struct file *filp) filp->private_data = data; memset(&data->handle, 0, sizeof(struct snapshot_handle)); if ((filp->f_flags & O_ACCMODE) == O_RDONLY) { + /* Hibernating. The image device should be accessible. */ data->swap = swsusp_resume_device ? swap_type_of(swsusp_resume_device, 0, NULL) : -1; data->mode = O_RDONLY; @@ -99,6 +100,13 @@ static int snapshot_open(struct inode *inode, struct file *filp) if (error) pm_notifier_call_chain(PM_POST_HIBERNATION); } else { + /* + * Resuming. We may need to wait for the image device to + * appear. + */ + wait_for_device_probe(); + scsi_complete_async_scans(); + data->swap = -1; data->mode = O_WRONLY; error = pm_notifier_call_chain(PM_RESTORE_PREPARE); diff --git a/kernel/printk.c b/kernel/printk.c index e3602d0755b..b5ac4d99c66 100644 --- a/kernel/printk.c +++ b/kernel/printk.c @@ -32,10 +32,18 @@ #include <linux/security.h> #include <linux/bootmem.h> #include <linux/syscalls.h> +#include <linux/kexec.h> +#include <linux/ratelimit.h> #include <asm/uaccess.h> /* + * for_each_console() allows you to iterate on each console + */ +#define for_each_console(con) \ + for (con = console_drivers; con != NULL; con = con->next) + +/* * Architectures can override it: */ void asmlinkage __attribute__((weak)) early_printk(const char *fmt, ...) @@ -60,6 +68,8 @@ int console_printk[4] = { DEFAULT_CONSOLE_LOGLEVEL, /* default_console_loglevel */ }; +static int saved_console_loglevel = -1; + /* * Low level drivers may need that to know if they can schedule in * their unblank() callback or not. So let's export it. @@ -135,6 +145,24 @@ static char *log_buf = __log_buf; static int log_buf_len = __LOG_BUF_LEN; static unsigned logged_chars; /* Number of chars produced since last read+clear operation */ +#ifdef CONFIG_KEXEC +/* + * This appends the listed symbols to /proc/vmcoreinfo + * + * /proc/vmcoreinfo is used by various utiilties, like crash and makedumpfile to + * obtain access to symbols that are otherwise very difficult to locate. These + * symbols are specifically used so that utilities can access and extract the + * dmesg log from a vmcore file after a crash. + */ +void log_buf_kexec_setup(void) +{ + VMCOREINFO_SYMBOL(log_buf); + VMCOREINFO_SYMBOL(log_end); + VMCOREINFO_SYMBOL(log_buf_len); + VMCOREINFO_SYMBOL(logged_chars); +} +#endif + static int __init log_buf_len_setup(char *str) { unsigned size = memparse(str, &str); @@ -179,12 +207,11 @@ __setup("log_buf_len=", log_buf_len_setup); #ifdef CONFIG_BOOT_PRINTK_DELAY static unsigned int boot_delay; /* msecs delay after each printk during bootup */ -static unsigned long long printk_delay_msec; /* per msec, based on boot_delay */ +static unsigned long long loops_per_msec; /* based on boot_delay */ static int __init boot_delay_setup(char *str) { unsigned long lpj; - unsigned long long loops_per_msec; lpj = preset_lpj ? preset_lpj : 1000000; /* some guess */ loops_per_msec = (unsigned long long)lpj / 1000 * HZ; @@ -193,10 +220,9 @@ static int __init boot_delay_setup(char *str) if (boot_delay > 10 * 1000) boot_delay = 0; - printk_delay_msec = loops_per_msec; - printk(KERN_DEBUG "boot_delay: %u, preset_lpj: %ld, lpj: %lu, " - "HZ: %d, printk_delay_msec: %llu\n", - boot_delay, preset_lpj, lpj, HZ, printk_delay_msec); + pr_debug("boot_delay: %u, preset_lpj: %ld, lpj: %lu, " + "HZ: %d, loops_per_msec: %llu\n", + boot_delay, preset_lpj, lpj, HZ, loops_per_msec); return 1; } __setup("boot_delay=", boot_delay_setup); @@ -209,7 +235,7 @@ static void boot_delay_msec(void) if (boot_delay == 0 || system_state != SYSTEM_BOOTING) return; - k = (unsigned long long)printk_delay_msec * boot_delay; + k = (unsigned long long)loops_per_msec * boot_delay; timeout = jiffies + msecs_to_jiffies(boot_delay); while (k) { @@ -353,10 +379,15 @@ int do_syslog(int type, char __user *buf, int len) logged_chars = 0; break; case 6: /* Disable logging to console */ + if (saved_console_loglevel == -1) + saved_console_loglevel = console_loglevel; console_loglevel = minimum_console_loglevel; break; case 7: /* Enable logging to console */ - console_loglevel = default_console_loglevel; + if (saved_console_loglevel != -1) { + console_loglevel = saved_console_loglevel; + saved_console_loglevel = -1; + } break; case 8: /* Set level of messages printed to console */ error = -EINVAL; @@ -365,6 +396,8 @@ int do_syslog(int type, char __user *buf, int len) if (len < minimum_console_loglevel) len = minimum_console_loglevel; console_loglevel = len; + /* Implicitly re-enable logging to console */ + saved_console_loglevel = -1; error = 0; break; case 9: /* Number of chars in the log buffer */ @@ -393,7 +426,7 @@ static void __call_console_drivers(unsigned start, unsigned end) { struct console *con; - for (con = console_drivers; con; con = con->next) { + for_each_console(con) { if ((con->flags & CON_ENABLED) && con->write && (cpu_online(smp_processor_id()) || (con->flags & CON_ANYTIME))) @@ -525,7 +558,7 @@ static int have_callable_console(void) { struct console *con; - for (con = console_drivers; con; con = con->next) + for_each_console(con) if (con->flags & CON_ANYTIME) return 1; @@ -621,6 +654,20 @@ static int recursion_bug; static int new_text_line = 1; static char printk_buf[1024]; +int printk_delay_msec __read_mostly; + +static inline void printk_delay(void) +{ + if (unlikely(printk_delay_msec)) { + int m = printk_delay_msec; + + while (m--) { + mdelay(1); + touch_nmi_watchdog(); + } + } +} + asmlinkage int vprintk(const char *fmt, va_list args) { int printed_len = 0; @@ -630,6 +677,7 @@ asmlinkage int vprintk(const char *fmt, va_list args) char *p; boot_delay_msec(); + printk_delay(); preempt_disable(); /* This stops the holder of console_sem just where we want him */ @@ -668,20 +716,35 @@ asmlinkage int vprintk(const char *fmt, va_list args) sizeof(printk_buf) - printed_len, fmt, args); + p = printk_buf; + + /* Do we have a loglevel in the string? */ + if (p[0] == '<') { + unsigned char c = p[1]; + if (c && p[2] == '>') { + switch (c) { + case '0' ... '7': /* loglevel */ + current_log_level = c - '0'; + /* Fallthrough - make sure we're on a new line */ + case 'd': /* KERN_DEFAULT */ + if (!new_text_line) { + emit_log_char('\n'); + new_text_line = 1; + } + /* Fallthrough - skip the loglevel */ + case 'c': /* KERN_CONT */ + p += 3; + break; + } + } + } + /* * Copy the output into log_buf. If the caller didn't provide * appropriate log level tags, we insert them here */ - for (p = printk_buf; *p; p++) { + for ( ; *p; p++) { if (new_text_line) { - /* If a token, set current_log_level and skip over */ - if (p[0] == '<' && p[1] >= '0' && p[1] <= '7' && - p[2] == '>') { - current_log_level = p[1] - '0'; - p += 3; - printed_len -= 3; - } - /* Always output the token */ emit_log_char('<'); emit_log_char(current_log_level + '0'); @@ -1026,12 +1089,6 @@ void __sched console_conditional_schedule(void) } EXPORT_SYMBOL(console_conditional_schedule); -void console_print(const char *s) -{ - printk(KERN_EMERG "%s", s); -} -EXPORT_SYMBOL(console_print); - void console_unblank(void) { struct console *c; @@ -1048,7 +1105,7 @@ void console_unblank(void) console_locked = 1; console_may_schedule = 0; - for (c = console_drivers; c != NULL; c = c->next) + for_each_console(c) if ((c->flags & CON_ENABLED) && c->unblank) c->unblank(); release_console_sem(); @@ -1063,7 +1120,7 @@ struct tty_driver *console_device(int *index) struct tty_driver *driver = NULL; acquire_console_sem(); - for (c = console_drivers; c != NULL; c = c->next) { + for_each_console(c) { if (!c->device) continue; driver = c->device(c, index); @@ -1100,25 +1157,49 @@ EXPORT_SYMBOL(console_start); * to register the console printing procedure with printk() and to * print any messages that were printed by the kernel before the * console driver was initialized. + * + * This can happen pretty early during the boot process (because of + * early_printk) - sometimes before setup_arch() completes - be careful + * of what kernel features are used - they may not be initialised yet. + * + * There are two types of consoles - bootconsoles (early_printk) and + * "real" consoles (everything which is not a bootconsole) which are + * handled differently. + * - Any number of bootconsoles can be registered at any time. + * - As soon as a "real" console is registered, all bootconsoles + * will be unregistered automatically. + * - Once a "real" console is registered, any attempt to register a + * bootconsoles will be rejected */ -void register_console(struct console *console) +void register_console(struct console *newcon) { int i; unsigned long flags; - struct console *bootconsole = NULL; + struct console *bcon = NULL; - if (console_drivers) { - if (console->flags & CON_BOOT) - return; - if (console_drivers->flags & CON_BOOT) - bootconsole = console_drivers; + /* + * before we register a new CON_BOOT console, make sure we don't + * already have a valid console + */ + if (console_drivers && newcon->flags & CON_BOOT) { + /* find the last or real console */ + for_each_console(bcon) { + if (!(bcon->flags & CON_BOOT)) { + printk(KERN_INFO "Too late to register bootconsole %s%d\n", + newcon->name, newcon->index); + return; + } + } } - if (preferred_console < 0 || bootconsole || !console_drivers) + if (console_drivers && console_drivers->flags & CON_BOOT) + bcon = console_drivers; + + if (preferred_console < 0 || bcon || !console_drivers) preferred_console = selected_console; - if (console->early_setup) - console->early_setup(); + if (newcon->early_setup) + newcon->early_setup(); /* * See if we want to use this console driver. If we @@ -1126,13 +1207,13 @@ void register_console(struct console *console) * that registers here. */ if (preferred_console < 0) { - if (console->index < 0) - console->index = 0; - if (console->setup == NULL || - console->setup(console, NULL) == 0) { - console->flags |= CON_ENABLED; - if (console->device) { - console->flags |= CON_CONSDEV; + if (newcon->index < 0) + newcon->index = 0; + if (newcon->setup == NULL || + newcon->setup(newcon, NULL) == 0) { + newcon->flags |= CON_ENABLED; + if (newcon->device) { + newcon->flags |= CON_CONSDEV; preferred_console = 0; } } @@ -1144,64 +1225,62 @@ void register_console(struct console *console) */ for (i = 0; i < MAX_CMDLINECONSOLES && console_cmdline[i].name[0]; i++) { - if (strcmp(console_cmdline[i].name, console->name) != 0) + if (strcmp(console_cmdline[i].name, newcon->name) != 0) continue; - if (console->index >= 0 && - console->index != console_cmdline[i].index) + if (newcon->index >= 0 && + newcon->index != console_cmdline[i].index) continue; - if (console->index < 0) - console->index = console_cmdline[i].index; + if (newcon->index < 0) + newcon->index = console_cmdline[i].index; #ifdef CONFIG_A11Y_BRAILLE_CONSOLE if (console_cmdline[i].brl_options) { - console->flags |= CON_BRL; - braille_register_console(console, + newcon->flags |= CON_BRL; + braille_register_console(newcon, console_cmdline[i].index, console_cmdline[i].options, console_cmdline[i].brl_options); return; } #endif - if (console->setup && - console->setup(console, console_cmdline[i].options) != 0) + if (newcon->setup && + newcon->setup(newcon, console_cmdline[i].options) != 0) break; - console->flags |= CON_ENABLED; - console->index = console_cmdline[i].index; + newcon->flags |= CON_ENABLED; + newcon->index = console_cmdline[i].index; if (i == selected_console) { - console->flags |= CON_CONSDEV; + newcon->flags |= CON_CONSDEV; preferred_console = selected_console; } break; } - if (!(console->flags & CON_ENABLED)) + if (!(newcon->flags & CON_ENABLED)) return; - if (bootconsole && (console->flags & CON_CONSDEV)) { - printk(KERN_INFO "console handover: boot [%s%d] -> real [%s%d]\n", - bootconsole->name, bootconsole->index, - console->name, console->index); - unregister_console(bootconsole); - console->flags &= ~CON_PRINTBUFFER; - } else { - printk(KERN_INFO "console [%s%d] enabled\n", - console->name, console->index); - } + /* + * If we have a bootconsole, and are switching to a real console, + * don't print everything out again, since when the boot console, and + * the real console are the same physical device, it's annoying to + * see the beginning boot messages twice + */ + if (bcon && ((newcon->flags & (CON_CONSDEV | CON_BOOT)) == CON_CONSDEV)) + newcon->flags &= ~CON_PRINTBUFFER; /* * Put this console in the list - keep the * preferred driver at the head of the list. */ acquire_console_sem(); - if ((console->flags & CON_CONSDEV) || console_drivers == NULL) { - console->next = console_drivers; - console_drivers = console; - if (console->next) - console->next->flags &= ~CON_CONSDEV; + if ((newcon->flags & CON_CONSDEV) || console_drivers == NULL) { + newcon->next = console_drivers; + console_drivers = newcon; + if (newcon->next) + newcon->next->flags &= ~CON_CONSDEV; } else { - console->next = console_drivers->next; - console_drivers->next = console; + newcon->next = console_drivers->next; + console_drivers->next = newcon; } - if (console->flags & CON_PRINTBUFFER) { + if (newcon->flags & CON_PRINTBUFFER) { /* * release_console_sem() will print out the buffered messages * for us. @@ -1211,6 +1290,28 @@ void register_console(struct console *console) spin_unlock_irqrestore(&logbuf_lock, flags); } release_console_sem(); + + /* + * By unregistering the bootconsoles after we enable the real console + * we get the "console xxx enabled" message on all the consoles - + * boot consoles, real consoles, etc - this is to ensure that end + * users know there might be something in the kernel's log buffer that + * went to the bootconsole (that they do not see on the real console) + */ + if (bcon && ((newcon->flags & (CON_CONSDEV | CON_BOOT)) == CON_CONSDEV)) { + /* we need to iterate through twice, to make sure we print + * everything out, before we unregister the console(s) + */ + printk(KERN_INFO "console [%s%d] enabled, bootconsole disabled\n", + newcon->name, newcon->index); + for_each_console(bcon) + if (bcon->flags & CON_BOOT) + unregister_console(bcon); + } else { + printk(KERN_INFO "%sconsole [%s%d] enabled\n", + (newcon->flags & CON_BOOT) ? "boot" : "" , + newcon->name, newcon->index); + } } EXPORT_SYMBOL(register_console); @@ -1253,11 +1354,13 @@ EXPORT_SYMBOL(unregister_console); static int __init disable_boot_consoles(void) { - if (console_drivers != NULL) { - if (console_drivers->flags & CON_BOOT) { + struct console *con; + + for_each_console(con) { + if (con->flags & CON_BOOT) { printk(KERN_INFO "turn off boot console %s%d\n", - console_drivers->name, console_drivers->index); - return unregister_console(console_drivers); + con->name, con->index); + unregister_console(con); } } return 0; @@ -1274,11 +1377,11 @@ late_initcall(disable_boot_consoles); */ DEFINE_RATELIMIT_STATE(printk_ratelimit_state, 5 * HZ, 10); -int printk_ratelimit(void) +int __printk_ratelimit(const char *func) { - return __ratelimit(&printk_ratelimit_state); + return ___ratelimit(&printk_ratelimit_state, func); } -EXPORT_SYMBOL(printk_ratelimit); +EXPORT_SYMBOL(__printk_ratelimit); /** * printk_timed_ratelimit - caller-controlled printk ratelimiting @@ -1292,8 +1395,11 @@ EXPORT_SYMBOL(printk_ratelimit); bool printk_timed_ratelimit(unsigned long *caller_jiffies, unsigned int interval_msecs) { - if (*caller_jiffies == 0 || time_after(jiffies, *caller_jiffies)) { - *caller_jiffies = jiffies + msecs_to_jiffies(interval_msecs); + if (*caller_jiffies == 0 + || !time_in_range(jiffies, *caller_jiffies, + *caller_jiffies + + msecs_to_jiffies(interval_msecs))) { + *caller_jiffies = jiffies; return true; } return false; diff --git a/kernel/profile.c b/kernel/profile.c index 7724e0409ba..a55d3a367ae 100644 --- a/kernel/profile.c +++ b/kernel/profile.c @@ -111,23 +111,18 @@ int __ref profile_init(void) /* only text is profiled */ prof_len = (_etext - _stext) >> prof_shift; buffer_bytes = prof_len*sizeof(atomic_t); - if (!slab_is_available()) { - prof_buffer = alloc_bootmem(buffer_bytes); - alloc_bootmem_cpumask_var(&prof_cpu_mask); - cpumask_copy(prof_cpu_mask, cpu_possible_mask); - return 0; - } if (!alloc_cpumask_var(&prof_cpu_mask, GFP_KERNEL)) return -ENOMEM; cpumask_copy(prof_cpu_mask, cpu_possible_mask); - prof_buffer = kzalloc(buffer_bytes, GFP_KERNEL); + prof_buffer = kzalloc(buffer_bytes, GFP_KERNEL|__GFP_NOWARN); if (prof_buffer) return 0; - prof_buffer = alloc_pages_exact(buffer_bytes, GFP_KERNEL|__GFP_ZERO); + prof_buffer = alloc_pages_exact(buffer_bytes, + GFP_KERNEL|__GFP_ZERO|__GFP_NOWARN); if (prof_buffer) return 0; @@ -371,7 +366,7 @@ static int __cpuinit profile_cpu_callback(struct notifier_block *info, node = cpu_to_node(cpu); per_cpu(cpu_profile_flip, cpu) = 0; if (!per_cpu(cpu_profile_hits, cpu)[1]) { - page = alloc_pages_node(node, + page = alloc_pages_exact_node(node, GFP_KERNEL | __GFP_ZERO, 0); if (!page) @@ -379,7 +374,7 @@ static int __cpuinit profile_cpu_callback(struct notifier_block *info, per_cpu(cpu_profile_hits, cpu)[1] = page_address(page); } if (!per_cpu(cpu_profile_hits, cpu)[0]) { - page = alloc_pages_node(node, + page = alloc_pages_exact_node(node, GFP_KERNEL | __GFP_ZERO, 0); if (!page) @@ -447,48 +442,51 @@ void profile_tick(int type) #ifdef CONFIG_PROC_FS #include <linux/proc_fs.h> +#include <linux/seq_file.h> #include <asm/uaccess.h> -static int prof_cpu_mask_read_proc(char *page, char **start, off_t off, - int count, int *eof, void *data) +static int prof_cpu_mask_proc_show(struct seq_file *m, void *v) +{ + seq_cpumask(m, prof_cpu_mask); + seq_putc(m, '\n'); + return 0; +} + +static int prof_cpu_mask_proc_open(struct inode *inode, struct file *file) { - int len = cpumask_scnprintf(page, count, data); - if (count - len < 2) - return -EINVAL; - len += sprintf(page + len, "\n"); - return len; + return single_open(file, prof_cpu_mask_proc_show, NULL); } -static int prof_cpu_mask_write_proc(struct file *file, - const char __user *buffer, unsigned long count, void *data) +static ssize_t prof_cpu_mask_proc_write(struct file *file, + const char __user *buffer, size_t count, loff_t *pos) { - struct cpumask *mask = data; - unsigned long full_count = count, err; cpumask_var_t new_value; + int err; if (!alloc_cpumask_var(&new_value, GFP_KERNEL)) return -ENOMEM; err = cpumask_parse_user(buffer, count, new_value); if (!err) { - cpumask_copy(mask, new_value); - err = full_count; + cpumask_copy(prof_cpu_mask, new_value); + err = count; } free_cpumask_var(new_value); return err; } +static const struct file_operations prof_cpu_mask_proc_fops = { + .open = prof_cpu_mask_proc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, + .write = prof_cpu_mask_proc_write, +}; + void create_prof_cpu_mask(struct proc_dir_entry *root_irq_dir) { - struct proc_dir_entry *entry; - /* create /proc/irq/prof_cpu_mask */ - entry = create_proc_entry("prof_cpu_mask", 0600, root_irq_dir); - if (!entry) - return; - entry->data = prof_cpu_mask; - entry->read_proc = prof_cpu_mask_read_proc; - entry->write_proc = prof_cpu_mask_write_proc; + proc_create("prof_cpu_mask", 0600, root_irq_dir, &prof_cpu_mask_proc_fops); } /* @@ -570,14 +568,14 @@ static int create_hash_tables(void) int node = cpu_to_node(cpu); struct page *page; - page = alloc_pages_node(node, + page = alloc_pages_exact_node(node, GFP_KERNEL | __GFP_ZERO | GFP_THISNODE, 0); if (!page) goto out_cleanup; per_cpu(cpu_profile_hits, cpu)[1] = (struct profile_hit *)page_address(page); - page = alloc_pages_node(node, + page = alloc_pages_exact_node(node, GFP_KERNEL | __GFP_ZERO | GFP_THISNODE, 0); if (!page) diff --git a/kernel/ptrace.c b/kernel/ptrace.c index c9cf48b21f0..23bd09cd042 100644 --- a/kernel/ptrace.c +++ b/kernel/ptrace.c @@ -21,22 +21,10 @@ #include <linux/audit.h> #include <linux/pid_namespace.h> #include <linux/syscalls.h> - -#include <asm/pgtable.h> -#include <asm/uaccess.h> +#include <linux/uaccess.h> /* - * Initialize a new task whose father had been ptraced. - * - * Called from copy_process(). - */ -void ptrace_fork(struct task_struct *child, unsigned long clone_flags) -{ - arch_ptrace_fork(child, clone_flags); -} - -/* * ptrace a task: make the debugger its new parent and * move it to the ptrace list. * @@ -48,7 +36,7 @@ void __ptrace_link(struct task_struct *child, struct task_struct *new_parent) list_add(&child->ptrace_entry, &new_parent->ptraced); child->parent = new_parent; } - + /* * Turn a tracing stop into a normal stop now, since with no tracer there * would be no way to wake it up with SIGCONT or SIGKILL. If there was a @@ -60,11 +48,15 @@ static void ptrace_untrace(struct task_struct *child) { spin_lock(&child->sighand->siglock); if (task_is_traced(child)) { - if (child->signal->flags & SIGNAL_STOP_STOPPED) { + /* + * If the group stop is completed or in progress, + * this thread was already counted as stopped. + */ + if (child->signal->flags & SIGNAL_STOP_STOPPED || + child->signal->group_stop_count) __set_task_state(child, TASK_STOPPED); - } else { + else signal_wake_up(child, 1); - } } spin_unlock(&child->sighand->siglock); } @@ -160,7 +152,7 @@ int __ptrace_may_access(struct task_struct *task, unsigned int mode) if (!dumpable && !capable(CAP_SYS_PTRACE)) return -EPERM; - return security_ptrace_may_access(task, mode); + return security_ptrace_access_check(task, mode); } bool ptrace_may_access(struct task_struct *task, unsigned int mode) @@ -169,84 +161,143 @@ bool ptrace_may_access(struct task_struct *task, unsigned int mode) task_lock(task); err = __ptrace_may_access(task, mode); task_unlock(task); - return (!err ? true : false); + return !err; } int ptrace_attach(struct task_struct *task) { int retval; - unsigned long flags; audit_ptrace(task); retval = -EPERM; + if (unlikely(task->flags & PF_KTHREAD)) + goto out; if (same_thread_group(task, current)) goto out; - /* Protect exec's credential calculations against our interference; - * SUID, SGID and LSM creds get determined differently under ptrace. + /* + * Protect exec's credential calculations against our interference; + * interference; SUID, SGID and LSM creds get determined differently + * under ptrace. */ - retval = mutex_lock_interruptible(¤t->cred_exec_mutex); - if (retval < 0) + retval = -ERESTARTNOINTR; + if (mutex_lock_interruptible(&task->cred_guard_mutex)) goto out; - retval = -EPERM; -repeat: - /* - * Nasty, nasty. - * - * We want to hold both the task-lock and the - * tasklist_lock for writing at the same time. - * But that's against the rules (tasklist_lock - * is taken for reading by interrupts on other - * cpu's that may have task_lock). - */ task_lock(task); - if (!write_trylock_irqsave(&tasklist_lock, flags)) { - task_unlock(task); - do { - cpu_relax(); - } while (!write_can_lock(&tasklist_lock)); - goto repeat; - } - - if (!task->mm) - goto bad; - /* the same process cannot be attached many times */ - if (task->ptrace & PT_PTRACED) - goto bad; retval = __ptrace_may_access(task, PTRACE_MODE_ATTACH); + task_unlock(task); if (retval) - goto bad; + goto unlock_creds; - /* Go */ - task->ptrace |= PT_PTRACED; + write_lock_irq(&tasklist_lock); + retval = -EPERM; + if (unlikely(task->exit_state)) + goto unlock_tasklist; + if (task->ptrace) + goto unlock_tasklist; + + task->ptrace = PT_PTRACED; if (capable(CAP_SYS_PTRACE)) task->ptrace |= PT_PTRACE_CAP; __ptrace_link(task, current); - send_sig_info(SIGSTOP, SEND_SIG_FORCED, task); -bad: - write_unlock_irqrestore(&tasklist_lock, flags); - task_unlock(task); - mutex_unlock(¤t->cred_exec_mutex); + + retval = 0; +unlock_tasklist: + write_unlock_irq(&tasklist_lock); +unlock_creds: + mutex_unlock(&task->cred_guard_mutex); out: return retval; } -static inline void __ptrace_detach(struct task_struct *child, unsigned int data) +/** + * ptrace_traceme -- helper for PTRACE_TRACEME + * + * Performs checks and sets PT_PTRACED. + * Should be used by all ptrace implementations for PTRACE_TRACEME. + */ +int ptrace_traceme(void) { - child->exit_code = data; - /* .. re-parent .. */ - __ptrace_unlink(child); - /* .. and wake it up. */ - if (child->exit_state != EXIT_ZOMBIE) - wake_up_process(child); + int ret = -EPERM; + + write_lock_irq(&tasklist_lock); + /* Are we already being traced? */ + if (!current->ptrace) { + ret = security_ptrace_traceme(current->parent); + /* + * Check PF_EXITING to ensure ->real_parent has not passed + * exit_ptrace(). Otherwise we don't report the error but + * pretend ->real_parent untraces us right after return. + */ + if (!ret && !(current->real_parent->flags & PF_EXITING)) { + current->ptrace = PT_PTRACED; + __ptrace_link(current, current->real_parent); + } + } + write_unlock_irq(&tasklist_lock); + + return ret; +} + +/* + * Called with irqs disabled, returns true if childs should reap themselves. + */ +static int ignoring_children(struct sighand_struct *sigh) +{ + int ret; + spin_lock(&sigh->siglock); + ret = (sigh->action[SIGCHLD-1].sa.sa_handler == SIG_IGN) || + (sigh->action[SIGCHLD-1].sa.sa_flags & SA_NOCLDWAIT); + spin_unlock(&sigh->siglock); + return ret; +} + +/* + * Called with tasklist_lock held for writing. + * Unlink a traced task, and clean it up if it was a traced zombie. + * Return true if it needs to be reaped with release_task(). + * (We can't call release_task() here because we already hold tasklist_lock.) + * + * If it's a zombie, our attachedness prevented normal parent notification + * or self-reaping. Do notification now if it would have happened earlier. + * If it should reap itself, return true. + * + * If it's our own child, there is no notification to do. But if our normal + * children self-reap, then this child was prevented by ptrace and we must + * reap it now, in that case we must also wake up sub-threads sleeping in + * do_wait(). + */ +static bool __ptrace_detach(struct task_struct *tracer, struct task_struct *p) +{ + __ptrace_unlink(p); + + if (p->exit_state == EXIT_ZOMBIE) { + if (!task_detached(p) && thread_group_empty(p)) { + if (!same_thread_group(p->real_parent, tracer)) + do_notify_parent(p, p->exit_signal); + else if (ignoring_children(tracer->sighand)) { + __wake_up_parent(p, tracer); + p->exit_signal = -1; + } + } + if (task_detached(p)) { + /* Mark it as in the process of being reaped. */ + p->exit_state = EXIT_DEAD; + return true; + } + } + + return false; } int ptrace_detach(struct task_struct *child, unsigned int data) { + bool dead = false; + if (!valid_signal(data)) return -EIO; @@ -255,14 +306,47 @@ int ptrace_detach(struct task_struct *child, unsigned int data) clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE); write_lock_irq(&tasklist_lock); - /* protect against de_thread()->release_task() */ - if (child->ptrace) - __ptrace_detach(child, data); + /* + * This child can be already killed. Make sure de_thread() or + * our sub-thread doing do_wait() didn't do release_task() yet. + */ + if (child->ptrace) { + child->exit_code = data; + dead = __ptrace_detach(current, child); + if (!child->exit_state) + wake_up_process(child); + } write_unlock_irq(&tasklist_lock); + if (unlikely(dead)) + release_task(child); + return 0; } +/* + * Detach all tasks we were using ptrace on. + */ +void exit_ptrace(struct task_struct *tracer) +{ + struct task_struct *p, *n; + LIST_HEAD(ptrace_dead); + + write_lock_irq(&tasklist_lock); + list_for_each_entry_safe(p, n, &tracer->ptraced, ptrace_entry) { + if (__ptrace_detach(tracer, p)) + list_add(&p->ptrace_entry, &ptrace_dead); + } + write_unlock_irq(&tasklist_lock); + + BUG_ON(!list_empty(&tracer->ptraced)); + + list_for_each_entry_safe(p, n, &ptrace_dead, ptrace_entry) { + list_del_init(&p->ptrace_entry); + release_task(p); + } +} + int ptrace_readdata(struct task_struct *tsk, unsigned long src, char __user *dst, int len) { int copied = 0; @@ -283,7 +367,7 @@ int ptrace_readdata(struct task_struct *tsk, unsigned long src, char __user *dst copied += retval; src += retval; dst += retval; - len -= retval; + len -= retval; } return copied; } @@ -308,7 +392,7 @@ int ptrace_writedata(struct task_struct *tsk, char __user *src, unsigned long ds copied += retval; src += retval; dst += retval; - len -= retval; + len -= retval; } return copied; } @@ -343,37 +427,33 @@ static int ptrace_setoptions(struct task_struct *child, long data) static int ptrace_getsiginfo(struct task_struct *child, siginfo_t *info) { + unsigned long flags; int error = -ESRCH; - read_lock(&tasklist_lock); - if (likely(child->sighand != NULL)) { + if (lock_task_sighand(child, &flags)) { error = -EINVAL; - spin_lock_irq(&child->sighand->siglock); if (likely(child->last_siginfo != NULL)) { *info = *child->last_siginfo; error = 0; } - spin_unlock_irq(&child->sighand->siglock); + unlock_task_sighand(child, &flags); } - read_unlock(&tasklist_lock); return error; } static int ptrace_setsiginfo(struct task_struct *child, const siginfo_t *info) { + unsigned long flags; int error = -ESRCH; - read_lock(&tasklist_lock); - if (likely(child->sighand != NULL)) { + if (lock_task_sighand(child, &flags)) { error = -EINVAL; - spin_lock_irq(&child->sighand->siglock); if (likely(child->last_siginfo != NULL)) { *child->last_siginfo = *info; error = 0; } - spin_unlock_irq(&child->sighand->siglock); + unlock_task_sighand(child, &flags); } - read_unlock(&tasklist_lock); return error; } @@ -421,9 +501,9 @@ static int ptrace_resume(struct task_struct *child, long request, long data) if (unlikely(!arch_has_single_step())) return -EIO; user_enable_single_step(child); - } - else + } else { user_disable_single_step(child); + } child->exit_code = data; wake_up_process(child); @@ -500,71 +580,16 @@ int ptrace_request(struct task_struct *child, long request, return ret; } -/** - * ptrace_traceme -- helper for PTRACE_TRACEME - * - * Performs checks and sets PT_PTRACED. - * Should be used by all ptrace implementations for PTRACE_TRACEME. - */ -int ptrace_traceme(void) -{ - int ret = -EPERM; - - /* - * Are we already being traced? - */ -repeat: - task_lock(current); - if (!(current->ptrace & PT_PTRACED)) { - /* - * See ptrace_attach() comments about the locking here. - */ - unsigned long flags; - if (!write_trylock_irqsave(&tasklist_lock, flags)) { - task_unlock(current); - do { - cpu_relax(); - } while (!write_can_lock(&tasklist_lock)); - goto repeat; - } - - ret = security_ptrace_traceme(current->parent); - - /* - * Set the ptrace bit in the process ptrace flags. - * Then link us on our parent's ptraced list. - */ - if (!ret) { - current->ptrace |= PT_PTRACED; - __ptrace_link(current, current->real_parent); - } - - write_unlock_irqrestore(&tasklist_lock, flags); - } - task_unlock(current); - return ret; -} - -/** - * ptrace_get_task_struct -- grab a task struct reference for ptrace - * @pid: process id to grab a task_struct reference of - * - * This function is a helper for ptrace implementations. It checks - * permissions and then grabs a task struct for use of the actual - * ptrace implementation. - * - * Returns the task_struct for @pid or an ERR_PTR() on failure. - */ -struct task_struct *ptrace_get_task_struct(pid_t pid) +static struct task_struct *ptrace_get_task_struct(pid_t pid) { struct task_struct *child; - read_lock(&tasklist_lock); + rcu_read_lock(); child = find_task_by_vpid(pid); if (child) get_task_struct(child); + rcu_read_unlock(); - read_unlock(&tasklist_lock); if (!child) return ERR_PTR(-ESRCH); return child; @@ -612,8 +637,6 @@ SYSCALL_DEFINE4(ptrace, long, request, long, pid, long, addr, long, data) goto out_put_task_struct; ret = arch_ptrace(child, request, addr, data); - if (ret < 0) - goto out_put_task_struct; out_put_task_struct: put_task_struct(child); diff --git a/kernel/rcuclassic.c b/kernel/rcuclassic.c deleted file mode 100644 index 654c640a6b9..00000000000 --- a/kernel/rcuclassic.c +++ /dev/null @@ -1,788 +0,0 @@ -/* - * Read-Copy Update mechanism for mutual exclusion - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or - * (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software - * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. - * - * Copyright IBM Corporation, 2001 - * - * Authors: Dipankar Sarma <dipankar@in.ibm.com> - * Manfred Spraul <manfred@colorfullife.com> - * - * Based on the original work by Paul McKenney <paulmck@us.ibm.com> - * and inputs from Rusty Russell, Andrea Arcangeli and Andi Kleen. - * Papers: - * http://www.rdrop.com/users/paulmck/paper/rclockpdcsproof.pdf - * http://lse.sourceforge.net/locking/rclock_OLS.2001.05.01c.sc.pdf (OLS2001) - * - * For detailed explanation of Read-Copy Update mechanism see - - * Documentation/RCU - * - */ -#include <linux/types.h> -#include <linux/kernel.h> -#include <linux/init.h> -#include <linux/spinlock.h> -#include <linux/smp.h> -#include <linux/rcupdate.h> -#include <linux/interrupt.h> -#include <linux/sched.h> -#include <asm/atomic.h> -#include <linux/bitops.h> -#include <linux/module.h> -#include <linux/completion.h> -#include <linux/moduleparam.h> -#include <linux/percpu.h> -#include <linux/notifier.h> -#include <linux/cpu.h> -#include <linux/mutex.h> -#include <linux/time.h> - -#ifdef CONFIG_DEBUG_LOCK_ALLOC -static struct lock_class_key rcu_lock_key; -struct lockdep_map rcu_lock_map = - STATIC_LOCKDEP_MAP_INIT("rcu_read_lock", &rcu_lock_key); -EXPORT_SYMBOL_GPL(rcu_lock_map); -#endif - - -/* Definition for rcupdate control block. */ -static struct rcu_ctrlblk rcu_ctrlblk = { - .cur = -300, - .completed = -300, - .pending = -300, - .lock = __SPIN_LOCK_UNLOCKED(&rcu_ctrlblk.lock), - .cpumask = CPU_BITS_NONE, -}; -static struct rcu_ctrlblk rcu_bh_ctrlblk = { - .cur = -300, - .completed = -300, - .pending = -300, - .lock = __SPIN_LOCK_UNLOCKED(&rcu_bh_ctrlblk.lock), - .cpumask = CPU_BITS_NONE, -}; - -DEFINE_PER_CPU(struct rcu_data, rcu_data) = { 0L }; -DEFINE_PER_CPU(struct rcu_data, rcu_bh_data) = { 0L }; - -static int blimit = 10; -static int qhimark = 10000; -static int qlowmark = 100; - -#ifdef CONFIG_SMP -static void force_quiescent_state(struct rcu_data *rdp, - struct rcu_ctrlblk *rcp) -{ - int cpu; - unsigned long flags; - - set_need_resched(); - spin_lock_irqsave(&rcp->lock, flags); - if (unlikely(!rcp->signaled)) { - rcp->signaled = 1; - /* - * Don't send IPI to itself. With irqs disabled, - * rdp->cpu is the current cpu. - * - * cpu_online_mask is updated by the _cpu_down() - * using __stop_machine(). Since we're in irqs disabled - * section, __stop_machine() is not exectuting, hence - * the cpu_online_mask is stable. - * - * However, a cpu might have been offlined _just_ before - * we disabled irqs while entering here. - * And rcu subsystem might not yet have handled the CPU_DEAD - * notification, leading to the offlined cpu's bit - * being set in the rcp->cpumask. - * - * Hence cpumask = (rcp->cpumask & cpu_online_mask) to prevent - * sending smp_reschedule() to an offlined CPU. - */ - for_each_cpu_and(cpu, - to_cpumask(rcp->cpumask), cpu_online_mask) { - if (cpu != rdp->cpu) - smp_send_reschedule(cpu); - } - } - spin_unlock_irqrestore(&rcp->lock, flags); -} -#else -static inline void force_quiescent_state(struct rcu_data *rdp, - struct rcu_ctrlblk *rcp) -{ - set_need_resched(); -} -#endif - -static void __call_rcu(struct rcu_head *head, struct rcu_ctrlblk *rcp, - struct rcu_data *rdp) -{ - long batch; - - head->next = NULL; - smp_mb(); /* Read of rcu->cur must happen after any change by caller. */ - - /* - * Determine the batch number of this callback. - * - * Using ACCESS_ONCE to avoid the following error when gcc eliminates - * local variable "batch" and emits codes like this: - * 1) rdp->batch = rcp->cur + 1 # gets old value - * ...... - * 2)rcu_batch_after(rcp->cur + 1, rdp->batch) # gets new value - * then [*nxttail[0], *nxttail[1]) may contain callbacks - * that batch# = rdp->batch, see the comment of struct rcu_data. - */ - batch = ACCESS_ONCE(rcp->cur) + 1; - - if (rdp->nxtlist && rcu_batch_after(batch, rdp->batch)) { - /* process callbacks */ - rdp->nxttail[0] = rdp->nxttail[1]; - rdp->nxttail[1] = rdp->nxttail[2]; - if (rcu_batch_after(batch - 1, rdp->batch)) - rdp->nxttail[0] = rdp->nxttail[2]; - } - - rdp->batch = batch; - *rdp->nxttail[2] = head; - rdp->nxttail[2] = &head->next; - - if (unlikely(++rdp->qlen > qhimark)) { - rdp->blimit = INT_MAX; - force_quiescent_state(rdp, &rcu_ctrlblk); - } -} - -#ifdef CONFIG_RCU_CPU_STALL_DETECTOR - -static void record_gp_stall_check_time(struct rcu_ctrlblk *rcp) -{ - rcp->gp_start = jiffies; - rcp->jiffies_stall = jiffies + RCU_SECONDS_TILL_STALL_CHECK; -} - -static void print_other_cpu_stall(struct rcu_ctrlblk *rcp) -{ - int cpu; - long delta; - unsigned long flags; - - /* Only let one CPU complain about others per time interval. */ - - spin_lock_irqsave(&rcp->lock, flags); - delta = jiffies - rcp->jiffies_stall; - if (delta < 2 || rcp->cur != rcp->completed) { - spin_unlock_irqrestore(&rcp->lock, flags); - return; - } - rcp->jiffies_stall = jiffies + RCU_SECONDS_TILL_STALL_RECHECK; - spin_unlock_irqrestore(&rcp->lock, flags); - - /* OK, time to rat on our buddy... */ - - printk(KERN_ERR "INFO: RCU detected CPU stalls:"); - for_each_possible_cpu(cpu) { - if (cpumask_test_cpu(cpu, to_cpumask(rcp->cpumask))) - printk(" %d", cpu); - } - printk(" (detected by %d, t=%ld jiffies)\n", - smp_processor_id(), (long)(jiffies - rcp->gp_start)); -} - -static void print_cpu_stall(struct rcu_ctrlblk *rcp) -{ - unsigned long flags; - - printk(KERN_ERR "INFO: RCU detected CPU %d stall (t=%lu/%lu jiffies)\n", - smp_processor_id(), jiffies, - jiffies - rcp->gp_start); - dump_stack(); - spin_lock_irqsave(&rcp->lock, flags); - if ((long)(jiffies - rcp->jiffies_stall) >= 0) - rcp->jiffies_stall = - jiffies + RCU_SECONDS_TILL_STALL_RECHECK; - spin_unlock_irqrestore(&rcp->lock, flags); - set_need_resched(); /* kick ourselves to get things going. */ -} - -static void check_cpu_stall(struct rcu_ctrlblk *rcp) -{ - long delta; - - delta = jiffies - rcp->jiffies_stall; - if (cpumask_test_cpu(smp_processor_id(), to_cpumask(rcp->cpumask)) && - delta >= 0) { - - /* We haven't checked in, so go dump stack. */ - print_cpu_stall(rcp); - - } else if (rcp->cur != rcp->completed && delta >= 2) { - - /* They had two seconds to dump stack, so complain. */ - print_other_cpu_stall(rcp); - } -} - -#else /* #ifdef CONFIG_RCU_CPU_STALL_DETECTOR */ - -static void record_gp_stall_check_time(struct rcu_ctrlblk *rcp) -{ -} - -static inline void check_cpu_stall(struct rcu_ctrlblk *rcp) -{ -} - -#endif /* #else #ifdef CONFIG_RCU_CPU_STALL_DETECTOR */ - -/** - * call_rcu - Queue an RCU callback for invocation after a grace period. - * @head: structure to be used for queueing the RCU updates. - * @func: actual update function to be invoked after the grace period - * - * The update function will be invoked some time after a full grace - * period elapses, in other words after all currently executing RCU - * read-side critical sections have completed. RCU read-side critical - * sections are delimited by rcu_read_lock() and rcu_read_unlock(), - * and may be nested. - */ -void call_rcu(struct rcu_head *head, - void (*func)(struct rcu_head *rcu)) -{ - unsigned long flags; - - head->func = func; - local_irq_save(flags); - __call_rcu(head, &rcu_ctrlblk, &__get_cpu_var(rcu_data)); - local_irq_restore(flags); -} -EXPORT_SYMBOL_GPL(call_rcu); - -/** - * call_rcu_bh - Queue an RCU for invocation after a quicker grace period. - * @head: structure to be used for queueing the RCU updates. - * @func: actual update function to be invoked after the grace period - * - * The update function will be invoked some time after a full grace - * period elapses, in other words after all currently executing RCU - * read-side critical sections have completed. call_rcu_bh() assumes - * that the read-side critical sections end on completion of a softirq - * handler. This means that read-side critical sections in process - * context must not be interrupted by softirqs. This interface is to be - * used when most of the read-side critical sections are in softirq context. - * RCU read-side critical sections are delimited by rcu_read_lock() and - * rcu_read_unlock(), * if in interrupt context or rcu_read_lock_bh() - * and rcu_read_unlock_bh(), if in process context. These may be nested. - */ -void call_rcu_bh(struct rcu_head *head, - void (*func)(struct rcu_head *rcu)) -{ - unsigned long flags; - - head->func = func; - local_irq_save(flags); - __call_rcu(head, &rcu_bh_ctrlblk, &__get_cpu_var(rcu_bh_data)); - local_irq_restore(flags); -} -EXPORT_SYMBOL_GPL(call_rcu_bh); - -/* - * Return the number of RCU batches processed thus far. Useful - * for debug and statistics. - */ -long rcu_batches_completed(void) -{ - return rcu_ctrlblk.completed; -} -EXPORT_SYMBOL_GPL(rcu_batches_completed); - -/* - * Return the number of RCU batches processed thus far. Useful - * for debug and statistics. - */ -long rcu_batches_completed_bh(void) -{ - return rcu_bh_ctrlblk.completed; -} -EXPORT_SYMBOL_GPL(rcu_batches_completed_bh); - -/* Raises the softirq for processing rcu_callbacks. */ -static inline void raise_rcu_softirq(void) -{ - raise_softirq(RCU_SOFTIRQ); -} - -/* - * Invoke the completed RCU callbacks. They are expected to be in - * a per-cpu list. - */ -static void rcu_do_batch(struct rcu_data *rdp) -{ - unsigned long flags; - struct rcu_head *next, *list; - int count = 0; - - list = rdp->donelist; - while (list) { - next = list->next; - prefetch(next); - list->func(list); - list = next; - if (++count >= rdp->blimit) - break; - } - rdp->donelist = list; - - local_irq_save(flags); - rdp->qlen -= count; - local_irq_restore(flags); - if (rdp->blimit == INT_MAX && rdp->qlen <= qlowmark) - rdp->blimit = blimit; - - if (!rdp->donelist) - rdp->donetail = &rdp->donelist; - else - raise_rcu_softirq(); -} - -/* - * Grace period handling: - * The grace period handling consists out of two steps: - * - A new grace period is started. - * This is done by rcu_start_batch. The start is not broadcasted to - * all cpus, they must pick this up by comparing rcp->cur with - * rdp->quiescbatch. All cpus are recorded in the - * rcu_ctrlblk.cpumask bitmap. - * - All cpus must go through a quiescent state. - * Since the start of the grace period is not broadcasted, at least two - * calls to rcu_check_quiescent_state are required: - * The first call just notices that a new grace period is running. The - * following calls check if there was a quiescent state since the beginning - * of the grace period. If so, it updates rcu_ctrlblk.cpumask. If - * the bitmap is empty, then the grace period is completed. - * rcu_check_quiescent_state calls rcu_start_batch(0) to start the next grace - * period (if necessary). - */ - -/* - * Register a new batch of callbacks, and start it up if there is currently no - * active batch and the batch to be registered has not already occurred. - * Caller must hold rcu_ctrlblk.lock. - */ -static void rcu_start_batch(struct rcu_ctrlblk *rcp) -{ - if (rcp->cur != rcp->pending && - rcp->completed == rcp->cur) { - rcp->cur++; - record_gp_stall_check_time(rcp); - - /* - * Accessing nohz_cpu_mask before incrementing rcp->cur needs a - * Barrier Otherwise it can cause tickless idle CPUs to be - * included in rcp->cpumask, which will extend graceperiods - * unnecessarily. - */ - smp_mb(); - cpumask_andnot(to_cpumask(rcp->cpumask), - cpu_online_mask, nohz_cpu_mask); - - rcp->signaled = 0; - } -} - -/* - * cpu went through a quiescent state since the beginning of the grace period. - * Clear it from the cpu mask and complete the grace period if it was the last - * cpu. Start another grace period if someone has further entries pending - */ -static void cpu_quiet(int cpu, struct rcu_ctrlblk *rcp) -{ - cpumask_clear_cpu(cpu, to_cpumask(rcp->cpumask)); - if (cpumask_empty(to_cpumask(rcp->cpumask))) { - /* batch completed ! */ - rcp->completed = rcp->cur; - rcu_start_batch(rcp); - } -} - -/* - * Check if the cpu has gone through a quiescent state (say context - * switch). If so and if it already hasn't done so in this RCU - * quiescent cycle, then indicate that it has done so. - */ -static void rcu_check_quiescent_state(struct rcu_ctrlblk *rcp, - struct rcu_data *rdp) -{ - unsigned long flags; - - if (rdp->quiescbatch != rcp->cur) { - /* start new grace period: */ - rdp->qs_pending = 1; - rdp->passed_quiesc = 0; - rdp->quiescbatch = rcp->cur; - return; - } - - /* Grace period already completed for this cpu? - * qs_pending is checked instead of the actual bitmap to avoid - * cacheline trashing. - */ - if (!rdp->qs_pending) - return; - - /* - * Was there a quiescent state since the beginning of the grace - * period? If no, then exit and wait for the next call. - */ - if (!rdp->passed_quiesc) - return; - rdp->qs_pending = 0; - - spin_lock_irqsave(&rcp->lock, flags); - /* - * rdp->quiescbatch/rcp->cur and the cpu bitmap can come out of sync - * during cpu startup. Ignore the quiescent state. - */ - if (likely(rdp->quiescbatch == rcp->cur)) - cpu_quiet(rdp->cpu, rcp); - - spin_unlock_irqrestore(&rcp->lock, flags); -} - - -#ifdef CONFIG_HOTPLUG_CPU - -/* warning! helper for rcu_offline_cpu. do not use elsewhere without reviewing - * locking requirements, the list it's pulling from has to belong to a cpu - * which is dead and hence not processing interrupts. - */ -static void rcu_move_batch(struct rcu_data *this_rdp, struct rcu_head *list, - struct rcu_head **tail, long batch) -{ - unsigned long flags; - - if (list) { - local_irq_save(flags); - this_rdp->batch = batch; - *this_rdp->nxttail[2] = list; - this_rdp->nxttail[2] = tail; - local_irq_restore(flags); - } -} - -static void __rcu_offline_cpu(struct rcu_data *this_rdp, - struct rcu_ctrlblk *rcp, struct rcu_data *rdp) -{ - unsigned long flags; - - /* - * if the cpu going offline owns the grace period - * we can block indefinitely waiting for it, so flush - * it here - */ - spin_lock_irqsave(&rcp->lock, flags); - if (rcp->cur != rcp->completed) - cpu_quiet(rdp->cpu, rcp); - rcu_move_batch(this_rdp, rdp->donelist, rdp->donetail, rcp->cur + 1); - rcu_move_batch(this_rdp, rdp->nxtlist, rdp->nxttail[2], rcp->cur + 1); - spin_unlock(&rcp->lock); - - this_rdp->qlen += rdp->qlen; - local_irq_restore(flags); -} - -static void rcu_offline_cpu(int cpu) -{ - struct rcu_data *this_rdp = &get_cpu_var(rcu_data); - struct rcu_data *this_bh_rdp = &get_cpu_var(rcu_bh_data); - - __rcu_offline_cpu(this_rdp, &rcu_ctrlblk, - &per_cpu(rcu_data, cpu)); - __rcu_offline_cpu(this_bh_rdp, &rcu_bh_ctrlblk, - &per_cpu(rcu_bh_data, cpu)); - put_cpu_var(rcu_data); - put_cpu_var(rcu_bh_data); -} - -#else - -static void rcu_offline_cpu(int cpu) -{ -} - -#endif - -/* - * This does the RCU processing work from softirq context. - */ -static void __rcu_process_callbacks(struct rcu_ctrlblk *rcp, - struct rcu_data *rdp) -{ - unsigned long flags; - long completed_snap; - - if (rdp->nxtlist) { - local_irq_save(flags); - completed_snap = ACCESS_ONCE(rcp->completed); - - /* - * move the other grace-period-completed entries to - * [rdp->nxtlist, *rdp->nxttail[0]) temporarily - */ - if (!rcu_batch_before(completed_snap, rdp->batch)) - rdp->nxttail[0] = rdp->nxttail[1] = rdp->nxttail[2]; - else if (!rcu_batch_before(completed_snap, rdp->batch - 1)) - rdp->nxttail[0] = rdp->nxttail[1]; - - /* - * the grace period for entries in - * [rdp->nxtlist, *rdp->nxttail[0]) has completed and - * move these entries to donelist - */ - if (rdp->nxttail[0] != &rdp->nxtlist) { - *rdp->donetail = rdp->nxtlist; - rdp->donetail = rdp->nxttail[0]; - rdp->nxtlist = *rdp->nxttail[0]; - *rdp->donetail = NULL; - - if (rdp->nxttail[1] == rdp->nxttail[0]) - rdp->nxttail[1] = &rdp->nxtlist; - if (rdp->nxttail[2] == rdp->nxttail[0]) - rdp->nxttail[2] = &rdp->nxtlist; - rdp->nxttail[0] = &rdp->nxtlist; - } - - local_irq_restore(flags); - - if (rcu_batch_after(rdp->batch, rcp->pending)) { - unsigned long flags2; - - /* and start it/schedule start if it's a new batch */ - spin_lock_irqsave(&rcp->lock, flags2); - if (rcu_batch_after(rdp->batch, rcp->pending)) { - rcp->pending = rdp->batch; - rcu_start_batch(rcp); - } - spin_unlock_irqrestore(&rcp->lock, flags2); - } - } - - rcu_check_quiescent_state(rcp, rdp); - if (rdp->donelist) - rcu_do_batch(rdp); -} - -static void rcu_process_callbacks(struct softirq_action *unused) -{ - /* - * Memory references from any prior RCU read-side critical sections - * executed by the interrupted code must be see before any RCU - * grace-period manupulations below. - */ - - smp_mb(); /* See above block comment. */ - - __rcu_process_callbacks(&rcu_ctrlblk, &__get_cpu_var(rcu_data)); - __rcu_process_callbacks(&rcu_bh_ctrlblk, &__get_cpu_var(rcu_bh_data)); - - /* - * Memory references from any later RCU read-side critical sections - * executed by the interrupted code must be see after any RCU - * grace-period manupulations above. - */ - - smp_mb(); /* See above block comment. */ -} - -static int __rcu_pending(struct rcu_ctrlblk *rcp, struct rcu_data *rdp) -{ - /* Check for CPU stalls, if enabled. */ - check_cpu_stall(rcp); - - if (rdp->nxtlist) { - long completed_snap = ACCESS_ONCE(rcp->completed); - - /* - * This cpu has pending rcu entries and the grace period - * for them has completed. - */ - if (!rcu_batch_before(completed_snap, rdp->batch)) - return 1; - if (!rcu_batch_before(completed_snap, rdp->batch - 1) && - rdp->nxttail[0] != rdp->nxttail[1]) - return 1; - if (rdp->nxttail[0] != &rdp->nxtlist) - return 1; - - /* - * This cpu has pending rcu entries and the new batch - * for then hasn't been started nor scheduled start - */ - if (rcu_batch_after(rdp->batch, rcp->pending)) - return 1; - } - - /* This cpu has finished callbacks to invoke */ - if (rdp->donelist) - return 1; - - /* The rcu core waits for a quiescent state from the cpu */ - if (rdp->quiescbatch != rcp->cur || rdp->qs_pending) - return 1; - - /* nothing to do */ - return 0; -} - -/* - * Check to see if there is any immediate RCU-related work to be done - * by the current CPU, returning 1 if so. This function is part of the - * RCU implementation; it is -not- an exported member of the RCU API. - */ -int rcu_pending(int cpu) -{ - return __rcu_pending(&rcu_ctrlblk, &per_cpu(rcu_data, cpu)) || - __rcu_pending(&rcu_bh_ctrlblk, &per_cpu(rcu_bh_data, cpu)); -} - -/* - * Check to see if any future RCU-related work will need to be done - * by the current CPU, even if none need be done immediately, returning - * 1 if so. This function is part of the RCU implementation; it is -not- - * an exported member of the RCU API. - */ -int rcu_needs_cpu(int cpu) -{ - struct rcu_data *rdp = &per_cpu(rcu_data, cpu); - struct rcu_data *rdp_bh = &per_cpu(rcu_bh_data, cpu); - - return !!rdp->nxtlist || !!rdp_bh->nxtlist || rcu_pending(cpu); -} - -/* - * Top-level function driving RCU grace-period detection, normally - * invoked from the scheduler-clock interrupt. This function simply - * increments counters that are read only from softirq by this same - * CPU, so there are no memory barriers required. - */ -void rcu_check_callbacks(int cpu, int user) -{ - if (user || - (idle_cpu(cpu) && rcu_scheduler_active && - !in_softirq() && hardirq_count() <= (1 << HARDIRQ_SHIFT))) { - - /* - * Get here if this CPU took its interrupt from user - * mode or from the idle loop, and if this is not a - * nested interrupt. In this case, the CPU is in - * a quiescent state, so count it. - * - * Also do a memory barrier. This is needed to handle - * the case where writes from a preempt-disable section - * of code get reordered into schedule() by this CPU's - * write buffer. The memory barrier makes sure that - * the rcu_qsctr_inc() and rcu_bh_qsctr_inc() are see - * by other CPUs to happen after any such write. - */ - - smp_mb(); /* See above block comment. */ - rcu_qsctr_inc(cpu); - rcu_bh_qsctr_inc(cpu); - - } else if (!in_softirq()) { - - /* - * Get here if this CPU did not take its interrupt from - * softirq, in other words, if it is not interrupting - * a rcu_bh read-side critical section. This is an _bh - * critical section, so count it. The memory barrier - * is needed for the same reason as is the above one. - */ - - smp_mb(); /* See above block comment. */ - rcu_bh_qsctr_inc(cpu); - } - raise_rcu_softirq(); -} - -static void __cpuinit rcu_init_percpu_data(int cpu, struct rcu_ctrlblk *rcp, - struct rcu_data *rdp) -{ - unsigned long flags; - - spin_lock_irqsave(&rcp->lock, flags); - memset(rdp, 0, sizeof(*rdp)); - rdp->nxttail[0] = rdp->nxttail[1] = rdp->nxttail[2] = &rdp->nxtlist; - rdp->donetail = &rdp->donelist; - rdp->quiescbatch = rcp->completed; - rdp->qs_pending = 0; - rdp->cpu = cpu; - rdp->blimit = blimit; - spin_unlock_irqrestore(&rcp->lock, flags); -} - -static void __cpuinit rcu_online_cpu(int cpu) -{ - struct rcu_data *rdp = &per_cpu(rcu_data, cpu); - struct rcu_data *bh_rdp = &per_cpu(rcu_bh_data, cpu); - - rcu_init_percpu_data(cpu, &rcu_ctrlblk, rdp); - rcu_init_percpu_data(cpu, &rcu_bh_ctrlblk, bh_rdp); - open_softirq(RCU_SOFTIRQ, rcu_process_callbacks); -} - -static int __cpuinit rcu_cpu_notify(struct notifier_block *self, - unsigned long action, void *hcpu) -{ - long cpu = (long)hcpu; - - switch (action) { - case CPU_UP_PREPARE: - case CPU_UP_PREPARE_FROZEN: - rcu_online_cpu(cpu); - break; - case CPU_DEAD: - case CPU_DEAD_FROZEN: - rcu_offline_cpu(cpu); - break; - default: - break; - } - return NOTIFY_OK; -} - -static struct notifier_block __cpuinitdata rcu_nb = { - .notifier_call = rcu_cpu_notify, -}; - -/* - * Initializes rcu mechanism. Assumed to be called early. - * That is before local timer(SMP) or jiffie timer (uniproc) is setup. - * Note that rcu_qsctr and friends are implicitly - * initialized due to the choice of ``0'' for RCU_CTR_INVALID. - */ -void __init __rcu_init(void) -{ -#ifdef CONFIG_RCU_CPU_STALL_DETECTOR - printk(KERN_INFO "RCU-based detection of stalled CPUs is enabled.\n"); -#endif /* #ifdef CONFIG_RCU_CPU_STALL_DETECTOR */ - rcu_cpu_notify(&rcu_nb, CPU_UP_PREPARE, - (void *)(long)smp_processor_id()); - /* Register notifier for non-boot CPUs */ - register_cpu_notifier(&rcu_nb); -} - -module_param(blimit, int, 0); -module_param(qhimark, int, 0); -module_param(qlowmark, int, 0); diff --git a/kernel/rcupdate.c b/kernel/rcupdate.c index cae8a059cf4..9b7fd472387 100644 --- a/kernel/rcupdate.c +++ b/kernel/rcupdate.c @@ -19,7 +19,7 @@ * * Authors: Dipankar Sarma <dipankar@in.ibm.com> * Manfred Spraul <manfred@colorfullife.com> - * + * * Based on the original work by Paul McKenney <paulmck@us.ibm.com> * and inputs from Rusty Russell, Andrea Arcangeli and Andi Kleen. * Papers: @@ -27,7 +27,7 @@ * http://lse.sourceforge.net/locking/rclock_OLS.2001.05.01c.sc.pdf (OLS2001) * * For detailed explanation of Read-Copy Update mechanism see - - * http://lse.sourceforge.net/locking/rcupdate.html + * http://lse.sourceforge.net/locking/rcupdate.html * */ #include <linux/types.h> @@ -44,19 +44,13 @@ #include <linux/cpu.h> #include <linux/mutex.h> #include <linux/module.h> -#include <linux/kernel_stat.h> - -enum rcu_barrier { - RCU_BARRIER_STD, - RCU_BARRIER_BH, - RCU_BARRIER_SCHED, -}; -static DEFINE_PER_CPU(struct rcu_head, rcu_barrier_head) = {NULL}; -static atomic_t rcu_barrier_cpu_count; -static DEFINE_MUTEX(rcu_barrier_mutex); -static struct completion rcu_barrier_completion; -int rcu_scheduler_active __read_mostly; +#ifdef CONFIG_DEBUG_LOCK_ALLOC +static struct lock_class_key rcu_lock_key; +struct lockdep_map rcu_lock_map = + STATIC_LOCKDEP_MAP_INIT("rcu_read_lock", &rcu_lock_key); +EXPORT_SYMBOL_GPL(rcu_lock_map); +#endif /* * Awaken the corresponding synchronize_rcu() instance now that a @@ -69,121 +63,3 @@ void wakeme_after_rcu(struct rcu_head *head) rcu = container_of(head, struct rcu_synchronize, head); complete(&rcu->completion); } - -/** - * synchronize_rcu - wait until a grace period has elapsed. - * - * Control will return to the caller some time after a full grace - * period has elapsed, in other words after all currently executing RCU - * read-side critical sections have completed. RCU read-side critical - * sections are delimited by rcu_read_lock() and rcu_read_unlock(), - * and may be nested. - */ -void synchronize_rcu(void) -{ - struct rcu_synchronize rcu; - - if (rcu_blocking_is_gp()) - return; - - init_completion(&rcu.completion); - /* Will wake me after RCU finished. */ - call_rcu(&rcu.head, wakeme_after_rcu); - /* Wait for it. */ - wait_for_completion(&rcu.completion); -} -EXPORT_SYMBOL_GPL(synchronize_rcu); - -static void rcu_barrier_callback(struct rcu_head *notused) -{ - if (atomic_dec_and_test(&rcu_barrier_cpu_count)) - complete(&rcu_barrier_completion); -} - -/* - * Called with preemption disabled, and from cross-cpu IRQ context. - */ -static void rcu_barrier_func(void *type) -{ - int cpu = smp_processor_id(); - struct rcu_head *head = &per_cpu(rcu_barrier_head, cpu); - - atomic_inc(&rcu_barrier_cpu_count); - switch ((enum rcu_barrier)type) { - case RCU_BARRIER_STD: - call_rcu(head, rcu_barrier_callback); - break; - case RCU_BARRIER_BH: - call_rcu_bh(head, rcu_barrier_callback); - break; - case RCU_BARRIER_SCHED: - call_rcu_sched(head, rcu_barrier_callback); - break; - } -} - -/* - * Orchestrate the specified type of RCU barrier, waiting for all - * RCU callbacks of the specified type to complete. - */ -static void _rcu_barrier(enum rcu_barrier type) -{ - BUG_ON(in_interrupt()); - /* Take cpucontrol mutex to protect against CPU hotplug */ - mutex_lock(&rcu_barrier_mutex); - init_completion(&rcu_barrier_completion); - /* - * Initialize rcu_barrier_cpu_count to 1, then invoke - * rcu_barrier_func() on each CPU, so that each CPU also has - * incremented rcu_barrier_cpu_count. Only then is it safe to - * decrement rcu_barrier_cpu_count -- otherwise the first CPU - * might complete its grace period before all of the other CPUs - * did their increment, causing this function to return too - * early. - */ - atomic_set(&rcu_barrier_cpu_count, 1); - on_each_cpu(rcu_barrier_func, (void *)type, 1); - if (atomic_dec_and_test(&rcu_barrier_cpu_count)) - complete(&rcu_barrier_completion); - wait_for_completion(&rcu_barrier_completion); - mutex_unlock(&rcu_barrier_mutex); -} - -/** - * rcu_barrier - Wait until all in-flight call_rcu() callbacks complete. - */ -void rcu_barrier(void) -{ - _rcu_barrier(RCU_BARRIER_STD); -} -EXPORT_SYMBOL_GPL(rcu_barrier); - -/** - * rcu_barrier_bh - Wait until all in-flight call_rcu_bh() callbacks complete. - */ -void rcu_barrier_bh(void) -{ - _rcu_barrier(RCU_BARRIER_BH); -} -EXPORT_SYMBOL_GPL(rcu_barrier_bh); - -/** - * rcu_barrier_sched - Wait for in-flight call_rcu_sched() callbacks. - */ -void rcu_barrier_sched(void) -{ - _rcu_barrier(RCU_BARRIER_SCHED); -} -EXPORT_SYMBOL_GPL(rcu_barrier_sched); - -void __init rcu_init(void) -{ - __rcu_init(); -} - -void rcu_scheduler_starting(void) -{ - WARN_ON(num_online_cpus() != 1); - WARN_ON(nr_context_switches() > 0); - rcu_scheduler_active = 1; -} diff --git a/kernel/rcupreempt.c b/kernel/rcupreempt.c deleted file mode 100644 index 5d59e850fb7..00000000000 --- a/kernel/rcupreempt.c +++ /dev/null @@ -1,1505 +0,0 @@ -/* - * Read-Copy Update mechanism for mutual exclusion, realtime implementation - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or - * (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software - * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. - * - * Copyright IBM Corporation, 2006 - * - * Authors: Paul E. McKenney <paulmck@us.ibm.com> - * With thanks to Esben Nielsen, Bill Huey, and Ingo Molnar - * for pushing me away from locks and towards counters, and - * to Suparna Bhattacharya for pushing me completely away - * from atomic instructions on the read side. - * - * - Added handling of Dynamic Ticks - * Copyright 2007 - Paul E. Mckenney <paulmck@us.ibm.com> - * - Steven Rostedt <srostedt@redhat.com> - * - * Papers: http://www.rdrop.com/users/paulmck/RCU - * - * Design Document: http://lwn.net/Articles/253651/ - * - * For detailed explanation of Read-Copy Update mechanism see - - * Documentation/RCU/ *.txt - * - */ -#include <linux/types.h> -#include <linux/kernel.h> -#include <linux/init.h> -#include <linux/spinlock.h> -#include <linux/smp.h> -#include <linux/rcupdate.h> -#include <linux/interrupt.h> -#include <linux/sched.h> -#include <asm/atomic.h> -#include <linux/bitops.h> -#include <linux/module.h> -#include <linux/kthread.h> -#include <linux/completion.h> -#include <linux/moduleparam.h> -#include <linux/percpu.h> -#include <linux/notifier.h> -#include <linux/cpu.h> -#include <linux/random.h> -#include <linux/delay.h> -#include <linux/cpumask.h> -#include <linux/rcupreempt_trace.h> -#include <asm/byteorder.h> - -/* - * PREEMPT_RCU data structures. - */ - -/* - * GP_STAGES specifies the number of times the state machine has - * to go through the all the rcu_try_flip_states (see below) - * in a single Grace Period. - * - * GP in GP_STAGES stands for Grace Period ;) - */ -#define GP_STAGES 2 -struct rcu_data { - spinlock_t lock; /* Protect rcu_data fields. */ - long completed; /* Number of last completed batch. */ - int waitlistcount; - struct rcu_head *nextlist; - struct rcu_head **nexttail; - struct rcu_head *waitlist[GP_STAGES]; - struct rcu_head **waittail[GP_STAGES]; - struct rcu_head *donelist; /* from waitlist & waitschedlist */ - struct rcu_head **donetail; - long rcu_flipctr[2]; - struct rcu_head *nextschedlist; - struct rcu_head **nextschedtail; - struct rcu_head *waitschedlist; - struct rcu_head **waitschedtail; - int rcu_sched_sleeping; -#ifdef CONFIG_RCU_TRACE - struct rcupreempt_trace trace; -#endif /* #ifdef CONFIG_RCU_TRACE */ -}; - -/* - * States for rcu_try_flip() and friends. - */ - -enum rcu_try_flip_states { - - /* - * Stay here if nothing is happening. Flip the counter if somthing - * starts happening. Denoted by "I" - */ - rcu_try_flip_idle_state, - - /* - * Wait here for all CPUs to notice that the counter has flipped. This - * prevents the old set of counters from ever being incremented once - * we leave this state, which in turn is necessary because we cannot - * test any individual counter for zero -- we can only check the sum. - * Denoted by "A". - */ - rcu_try_flip_waitack_state, - - /* - * Wait here for the sum of the old per-CPU counters to reach zero. - * Denoted by "Z". - */ - rcu_try_flip_waitzero_state, - - /* - * Wait here for each of the other CPUs to execute a memory barrier. - * This is necessary to ensure that these other CPUs really have - * completed executing their RCU read-side critical sections, despite - * their CPUs wildly reordering memory. Denoted by "M". - */ - rcu_try_flip_waitmb_state, -}; - -/* - * States for rcu_ctrlblk.rcu_sched_sleep. - */ - -enum rcu_sched_sleep_states { - rcu_sched_not_sleeping, /* Not sleeping, callbacks need GP. */ - rcu_sched_sleep_prep, /* Thinking of sleeping, rechecking. */ - rcu_sched_sleeping, /* Sleeping, awaken if GP needed. */ -}; - -struct rcu_ctrlblk { - spinlock_t fliplock; /* Protect state-machine transitions. */ - long completed; /* Number of last completed batch. */ - enum rcu_try_flip_states rcu_try_flip_state; /* The current state of - the rcu state machine */ - spinlock_t schedlock; /* Protect rcu_sched sleep state. */ - enum rcu_sched_sleep_states sched_sleep; /* rcu_sched state. */ - wait_queue_head_t sched_wq; /* Place for rcu_sched to sleep. */ -}; - -static DEFINE_PER_CPU(struct rcu_data, rcu_data); -static struct rcu_ctrlblk rcu_ctrlblk = { - .fliplock = __SPIN_LOCK_UNLOCKED(rcu_ctrlblk.fliplock), - .completed = 0, - .rcu_try_flip_state = rcu_try_flip_idle_state, - .schedlock = __SPIN_LOCK_UNLOCKED(rcu_ctrlblk.schedlock), - .sched_sleep = rcu_sched_not_sleeping, - .sched_wq = __WAIT_QUEUE_HEAD_INITIALIZER(rcu_ctrlblk.sched_wq), -}; - -static struct task_struct *rcu_sched_grace_period_task; - -#ifdef CONFIG_RCU_TRACE -static char *rcu_try_flip_state_names[] = - { "idle", "waitack", "waitzero", "waitmb" }; -#endif /* #ifdef CONFIG_RCU_TRACE */ - -static DECLARE_BITMAP(rcu_cpu_online_map, NR_CPUS) __read_mostly - = CPU_BITS_NONE; - -/* - * Enum and per-CPU flag to determine when each CPU has seen - * the most recent counter flip. - */ - -enum rcu_flip_flag_values { - rcu_flip_seen, /* Steady/initial state, last flip seen. */ - /* Only GP detector can update. */ - rcu_flipped /* Flip just completed, need confirmation. */ - /* Only corresponding CPU can update. */ -}; -static DEFINE_PER_CPU_SHARED_ALIGNED(enum rcu_flip_flag_values, rcu_flip_flag) - = rcu_flip_seen; - -/* - * Enum and per-CPU flag to determine when each CPU has executed the - * needed memory barrier to fence in memory references from its last RCU - * read-side critical section in the just-completed grace period. - */ - -enum rcu_mb_flag_values { - rcu_mb_done, /* Steady/initial state, no mb()s required. */ - /* Only GP detector can update. */ - rcu_mb_needed /* Flip just completed, need an mb(). */ - /* Only corresponding CPU can update. */ -}; -static DEFINE_PER_CPU_SHARED_ALIGNED(enum rcu_mb_flag_values, rcu_mb_flag) - = rcu_mb_done; - -/* - * RCU_DATA_ME: find the current CPU's rcu_data structure. - * RCU_DATA_CPU: find the specified CPU's rcu_data structure. - */ -#define RCU_DATA_ME() (&__get_cpu_var(rcu_data)) -#define RCU_DATA_CPU(cpu) (&per_cpu(rcu_data, cpu)) - -/* - * Helper macro for tracing when the appropriate rcu_data is not - * cached in a local variable, but where the CPU number is so cached. - */ -#define RCU_TRACE_CPU(f, cpu) RCU_TRACE(f, &(RCU_DATA_CPU(cpu)->trace)); - -/* - * Helper macro for tracing when the appropriate rcu_data is not - * cached in a local variable. - */ -#define RCU_TRACE_ME(f) RCU_TRACE(f, &(RCU_DATA_ME()->trace)); - -/* - * Helper macro for tracing when the appropriate rcu_data is pointed - * to by a local variable. - */ -#define RCU_TRACE_RDP(f, rdp) RCU_TRACE(f, &((rdp)->trace)); - -#define RCU_SCHED_BATCH_TIME (HZ / 50) - -/* - * Return the number of RCU batches processed thus far. Useful - * for debug and statistics. - */ -long rcu_batches_completed(void) -{ - return rcu_ctrlblk.completed; -} -EXPORT_SYMBOL_GPL(rcu_batches_completed); - -void __rcu_read_lock(void) -{ - int idx; - struct task_struct *t = current; - int nesting; - - nesting = ACCESS_ONCE(t->rcu_read_lock_nesting); - if (nesting != 0) { - - /* An earlier rcu_read_lock() covers us, just count it. */ - - t->rcu_read_lock_nesting = nesting + 1; - - } else { - unsigned long flags; - - /* - * We disable interrupts for the following reasons: - * - If we get scheduling clock interrupt here, and we - * end up acking the counter flip, it's like a promise - * that we will never increment the old counter again. - * Thus we will break that promise if that - * scheduling clock interrupt happens between the time - * we pick the .completed field and the time that we - * increment our counter. - * - * - We don't want to be preempted out here. - * - * NMIs can still occur, of course, and might themselves - * contain rcu_read_lock(). - */ - - local_irq_save(flags); - - /* - * Outermost nesting of rcu_read_lock(), so increment - * the current counter for the current CPU. Use volatile - * casts to prevent the compiler from reordering. - */ - - idx = ACCESS_ONCE(rcu_ctrlblk.completed) & 0x1; - ACCESS_ONCE(RCU_DATA_ME()->rcu_flipctr[idx])++; - - /* - * Now that the per-CPU counter has been incremented, we - * are protected from races with rcu_read_lock() invoked - * from NMI handlers on this CPU. We can therefore safely - * increment the nesting counter, relieving further NMIs - * of the need to increment the per-CPU counter. - */ - - ACCESS_ONCE(t->rcu_read_lock_nesting) = nesting + 1; - - /* - * Now that we have preventing any NMIs from storing - * to the ->rcu_flipctr_idx, we can safely use it to - * remember which counter to decrement in the matching - * rcu_read_unlock(). - */ - - ACCESS_ONCE(t->rcu_flipctr_idx) = idx; - local_irq_restore(flags); - } -} -EXPORT_SYMBOL_GPL(__rcu_read_lock); - -void __rcu_read_unlock(void) -{ - int idx; - struct task_struct *t = current; - int nesting; - - nesting = ACCESS_ONCE(t->rcu_read_lock_nesting); - if (nesting > 1) { - - /* - * We are still protected by the enclosing rcu_read_lock(), - * so simply decrement the counter. - */ - - t->rcu_read_lock_nesting = nesting - 1; - - } else { - unsigned long flags; - - /* - * Disable local interrupts to prevent the grace-period - * detection state machine from seeing us half-done. - * NMIs can still occur, of course, and might themselves - * contain rcu_read_lock() and rcu_read_unlock(). - */ - - local_irq_save(flags); - - /* - * Outermost nesting of rcu_read_unlock(), so we must - * decrement the current counter for the current CPU. - * This must be done carefully, because NMIs can - * occur at any point in this code, and any rcu_read_lock() - * and rcu_read_unlock() pairs in the NMI handlers - * must interact non-destructively with this code. - * Lots of volatile casts, and -very- careful ordering. - * - * Changes to this code, including this one, must be - * inspected, validated, and tested extremely carefully!!! - */ - - /* - * First, pick up the index. - */ - - idx = ACCESS_ONCE(t->rcu_flipctr_idx); - - /* - * Now that we have fetched the counter index, it is - * safe to decrement the per-task RCU nesting counter. - * After this, any interrupts or NMIs will increment and - * decrement the per-CPU counters. - */ - ACCESS_ONCE(t->rcu_read_lock_nesting) = nesting - 1; - - /* - * It is now safe to decrement this task's nesting count. - * NMIs that occur after this statement will route their - * rcu_read_lock() calls through this "else" clause, and - * will thus start incrementing the per-CPU counter on - * their own. They will also clobber ->rcu_flipctr_idx, - * but that is OK, since we have already fetched it. - */ - - ACCESS_ONCE(RCU_DATA_ME()->rcu_flipctr[idx])--; - local_irq_restore(flags); - } -} -EXPORT_SYMBOL_GPL(__rcu_read_unlock); - -/* - * If a global counter flip has occurred since the last time that we - * advanced callbacks, advance them. Hardware interrupts must be - * disabled when calling this function. - */ -static void __rcu_advance_callbacks(struct rcu_data *rdp) -{ - int cpu; - int i; - int wlc = 0; - - if (rdp->completed != rcu_ctrlblk.completed) { - if (rdp->waitlist[GP_STAGES - 1] != NULL) { - *rdp->donetail = rdp->waitlist[GP_STAGES - 1]; - rdp->donetail = rdp->waittail[GP_STAGES - 1]; - RCU_TRACE_RDP(rcupreempt_trace_move2done, rdp); - } - for (i = GP_STAGES - 2; i >= 0; i--) { - if (rdp->waitlist[i] != NULL) { - rdp->waitlist[i + 1] = rdp->waitlist[i]; - rdp->waittail[i + 1] = rdp->waittail[i]; - wlc++; - } else { - rdp->waitlist[i + 1] = NULL; - rdp->waittail[i + 1] = - &rdp->waitlist[i + 1]; - } - } - if (rdp->nextlist != NULL) { - rdp->waitlist[0] = rdp->nextlist; - rdp->waittail[0] = rdp->nexttail; - wlc++; - rdp->nextlist = NULL; - rdp->nexttail = &rdp->nextlist; - RCU_TRACE_RDP(rcupreempt_trace_move2wait, rdp); - } else { - rdp->waitlist[0] = NULL; - rdp->waittail[0] = &rdp->waitlist[0]; - } - rdp->waitlistcount = wlc; - rdp->completed = rcu_ctrlblk.completed; - } - - /* - * Check to see if this CPU needs to report that it has seen - * the most recent counter flip, thereby declaring that all - * subsequent rcu_read_lock() invocations will respect this flip. - */ - - cpu = raw_smp_processor_id(); - if (per_cpu(rcu_flip_flag, cpu) == rcu_flipped) { - smp_mb(); /* Subsequent counter accesses must see new value */ - per_cpu(rcu_flip_flag, cpu) = rcu_flip_seen; - smp_mb(); /* Subsequent RCU read-side critical sections */ - /* seen -after- acknowledgement. */ - } -} - -DEFINE_PER_CPU_SHARED_ALIGNED(struct rcu_dyntick_sched, rcu_dyntick_sched) = { - .dynticks = 1, -}; - -#ifdef CONFIG_NO_HZ -static DEFINE_PER_CPU(int, rcu_update_flag); - -/** - * rcu_irq_enter - Called from Hard irq handlers and NMI/SMI. - * - * If the CPU was idle with dynamic ticks active, this updates the - * rcu_dyntick_sched.dynticks to let the RCU handling know that the - * CPU is active. - */ -void rcu_irq_enter(void) -{ - int cpu = smp_processor_id(); - struct rcu_dyntick_sched *rdssp = &per_cpu(rcu_dyntick_sched, cpu); - - if (per_cpu(rcu_update_flag, cpu)) - per_cpu(rcu_update_flag, cpu)++; - - /* - * Only update if we are coming from a stopped ticks mode - * (rcu_dyntick_sched.dynticks is even). - */ - if (!in_interrupt() && - (rdssp->dynticks & 0x1) == 0) { - /* - * The following might seem like we could have a race - * with NMI/SMIs. But this really isn't a problem. - * Here we do a read/modify/write, and the race happens - * when an NMI/SMI comes in after the read and before - * the write. But NMI/SMIs will increment this counter - * twice before returning, so the zero bit will not - * be corrupted by the NMI/SMI which is the most important - * part. - * - * The only thing is that we would bring back the counter - * to a postion that it was in during the NMI/SMI. - * But the zero bit would be set, so the rest of the - * counter would again be ignored. - * - * On return from the IRQ, the counter may have the zero - * bit be 0 and the counter the same as the return from - * the NMI/SMI. If the state machine was so unlucky to - * see that, it still doesn't matter, since all - * RCU read-side critical sections on this CPU would - * have already completed. - */ - rdssp->dynticks++; - /* - * The following memory barrier ensures that any - * rcu_read_lock() primitives in the irq handler - * are seen by other CPUs to follow the above - * increment to rcu_dyntick_sched.dynticks. This is - * required in order for other CPUs to correctly - * determine when it is safe to advance the RCU - * grace-period state machine. - */ - smp_mb(); /* see above block comment. */ - /* - * Since we can't determine the dynamic tick mode from - * the rcu_dyntick_sched.dynticks after this routine, - * we use a second flag to acknowledge that we came - * from an idle state with ticks stopped. - */ - per_cpu(rcu_update_flag, cpu)++; - /* - * If we take an NMI/SMI now, they will also increment - * the rcu_update_flag, and will not update the - * rcu_dyntick_sched.dynticks on exit. That is for - * this IRQ to do. - */ - } -} - -/** - * rcu_irq_exit - Called from exiting Hard irq context. - * - * If the CPU was idle with dynamic ticks active, update the - * rcu_dyntick_sched.dynticks to put let the RCU handling be - * aware that the CPU is going back to idle with no ticks. - */ -void rcu_irq_exit(void) -{ - int cpu = smp_processor_id(); - struct rcu_dyntick_sched *rdssp = &per_cpu(rcu_dyntick_sched, cpu); - - /* - * rcu_update_flag is set if we interrupted the CPU - * when it was idle with ticks stopped. - * Once this occurs, we keep track of interrupt nesting - * because a NMI/SMI could also come in, and we still - * only want the IRQ that started the increment of the - * rcu_dyntick_sched.dynticks to be the one that modifies - * it on exit. - */ - if (per_cpu(rcu_update_flag, cpu)) { - if (--per_cpu(rcu_update_flag, cpu)) - return; - - /* This must match the interrupt nesting */ - WARN_ON(in_interrupt()); - - /* - * If an NMI/SMI happens now we are still - * protected by the rcu_dyntick_sched.dynticks being odd. - */ - - /* - * The following memory barrier ensures that any - * rcu_read_unlock() primitives in the irq handler - * are seen by other CPUs to preceed the following - * increment to rcu_dyntick_sched.dynticks. This - * is required in order for other CPUs to determine - * when it is safe to advance the RCU grace-period - * state machine. - */ - smp_mb(); /* see above block comment. */ - rdssp->dynticks++; - WARN_ON(rdssp->dynticks & 0x1); - } -} - -void rcu_nmi_enter(void) -{ - rcu_irq_enter(); -} - -void rcu_nmi_exit(void) -{ - rcu_irq_exit(); -} - -static void dyntick_save_progress_counter(int cpu) -{ - struct rcu_dyntick_sched *rdssp = &per_cpu(rcu_dyntick_sched, cpu); - - rdssp->dynticks_snap = rdssp->dynticks; -} - -static inline int -rcu_try_flip_waitack_needed(int cpu) -{ - long curr; - long snap; - struct rcu_dyntick_sched *rdssp = &per_cpu(rcu_dyntick_sched, cpu); - - curr = rdssp->dynticks; - snap = rdssp->dynticks_snap; - smp_mb(); /* force ordering with cpu entering/leaving dynticks. */ - - /* - * If the CPU remained in dynticks mode for the entire time - * and didn't take any interrupts, NMIs, SMIs, or whatever, - * then it cannot be in the middle of an rcu_read_lock(), so - * the next rcu_read_lock() it executes must use the new value - * of the counter. So we can safely pretend that this CPU - * already acknowledged the counter. - */ - - if ((curr == snap) && ((curr & 0x1) == 0)) - return 0; - - /* - * If the CPU passed through or entered a dynticks idle phase with - * no active irq handlers, then, as above, we can safely pretend - * that this CPU already acknowledged the counter. - */ - - if ((curr - snap) > 2 || (curr & 0x1) == 0) - return 0; - - /* We need this CPU to explicitly acknowledge the counter flip. */ - - return 1; -} - -static inline int -rcu_try_flip_waitmb_needed(int cpu) -{ - long curr; - long snap; - struct rcu_dyntick_sched *rdssp = &per_cpu(rcu_dyntick_sched, cpu); - - curr = rdssp->dynticks; - snap = rdssp->dynticks_snap; - smp_mb(); /* force ordering with cpu entering/leaving dynticks. */ - - /* - * If the CPU remained in dynticks mode for the entire time - * and didn't take any interrupts, NMIs, SMIs, or whatever, - * then it cannot have executed an RCU read-side critical section - * during that time, so there is no need for it to execute a - * memory barrier. - */ - - if ((curr == snap) && ((curr & 0x1) == 0)) - return 0; - - /* - * If the CPU either entered or exited an outermost interrupt, - * SMI, NMI, or whatever handler, then we know that it executed - * a memory barrier when doing so. So we don't need another one. - */ - if (curr != snap) - return 0; - - /* We need the CPU to execute a memory barrier. */ - - return 1; -} - -static void dyntick_save_progress_counter_sched(int cpu) -{ - struct rcu_dyntick_sched *rdssp = &per_cpu(rcu_dyntick_sched, cpu); - - rdssp->sched_dynticks_snap = rdssp->dynticks; -} - -static int rcu_qsctr_inc_needed_dyntick(int cpu) -{ - long curr; - long snap; - struct rcu_dyntick_sched *rdssp = &per_cpu(rcu_dyntick_sched, cpu); - - curr = rdssp->dynticks; - snap = rdssp->sched_dynticks_snap; - smp_mb(); /* force ordering with cpu entering/leaving dynticks. */ - - /* - * If the CPU remained in dynticks mode for the entire time - * and didn't take any interrupts, NMIs, SMIs, or whatever, - * then it cannot be in the middle of an rcu_read_lock(), so - * the next rcu_read_lock() it executes must use the new value - * of the counter. Therefore, this CPU has been in a quiescent - * state the entire time, and we don't need to wait for it. - */ - - if ((curr == snap) && ((curr & 0x1) == 0)) - return 0; - - /* - * If the CPU passed through or entered a dynticks idle phase with - * no active irq handlers, then, as above, this CPU has already - * passed through a quiescent state. - */ - - if ((curr - snap) > 2 || (snap & 0x1) == 0) - return 0; - - /* We need this CPU to go through a quiescent state. */ - - return 1; -} - -#else /* !CONFIG_NO_HZ */ - -# define dyntick_save_progress_counter(cpu) do { } while (0) -# define rcu_try_flip_waitack_needed(cpu) (1) -# define rcu_try_flip_waitmb_needed(cpu) (1) - -# define dyntick_save_progress_counter_sched(cpu) do { } while (0) -# define rcu_qsctr_inc_needed_dyntick(cpu) (1) - -#endif /* CONFIG_NO_HZ */ - -static void save_qsctr_sched(int cpu) -{ - struct rcu_dyntick_sched *rdssp = &per_cpu(rcu_dyntick_sched, cpu); - - rdssp->sched_qs_snap = rdssp->sched_qs; -} - -static inline int rcu_qsctr_inc_needed(int cpu) -{ - struct rcu_dyntick_sched *rdssp = &per_cpu(rcu_dyntick_sched, cpu); - - /* - * If there has been a quiescent state, no more need to wait - * on this CPU. - */ - - if (rdssp->sched_qs != rdssp->sched_qs_snap) { - smp_mb(); /* force ordering with cpu entering schedule(). */ - return 0; - } - - /* We need this CPU to go through a quiescent state. */ - - return 1; -} - -/* - * Get here when RCU is idle. Decide whether we need to - * move out of idle state, and return non-zero if so. - * "Straightforward" approach for the moment, might later - * use callback-list lengths, grace-period duration, or - * some such to determine when to exit idle state. - * Might also need a pre-idle test that does not acquire - * the lock, but let's get the simple case working first... - */ - -static int -rcu_try_flip_idle(void) -{ - int cpu; - - RCU_TRACE_ME(rcupreempt_trace_try_flip_i1); - if (!rcu_pending(smp_processor_id())) { - RCU_TRACE_ME(rcupreempt_trace_try_flip_ie1); - return 0; - } - - /* - * Do the flip. - */ - - RCU_TRACE_ME(rcupreempt_trace_try_flip_g1); - rcu_ctrlblk.completed++; /* stands in for rcu_try_flip_g2 */ - - /* - * Need a memory barrier so that other CPUs see the new - * counter value before they see the subsequent change of all - * the rcu_flip_flag instances to rcu_flipped. - */ - - smp_mb(); /* see above block comment. */ - - /* Now ask each CPU for acknowledgement of the flip. */ - - for_each_cpu(cpu, to_cpumask(rcu_cpu_online_map)) { - per_cpu(rcu_flip_flag, cpu) = rcu_flipped; - dyntick_save_progress_counter(cpu); - } - - return 1; -} - -/* - * Wait for CPUs to acknowledge the flip. - */ - -static int -rcu_try_flip_waitack(void) -{ - int cpu; - - RCU_TRACE_ME(rcupreempt_trace_try_flip_a1); - for_each_cpu(cpu, to_cpumask(rcu_cpu_online_map)) - if (rcu_try_flip_waitack_needed(cpu) && - per_cpu(rcu_flip_flag, cpu) != rcu_flip_seen) { - RCU_TRACE_ME(rcupreempt_trace_try_flip_ae1); - return 0; - } - - /* - * Make sure our checks above don't bleed into subsequent - * waiting for the sum of the counters to reach zero. - */ - - smp_mb(); /* see above block comment. */ - RCU_TRACE_ME(rcupreempt_trace_try_flip_a2); - return 1; -} - -/* - * Wait for collective ``last'' counter to reach zero, - * then tell all CPUs to do an end-of-grace-period memory barrier. - */ - -static int -rcu_try_flip_waitzero(void) -{ - int cpu; - int lastidx = !(rcu_ctrlblk.completed & 0x1); - int sum = 0; - - /* Check to see if the sum of the "last" counters is zero. */ - - RCU_TRACE_ME(rcupreempt_trace_try_flip_z1); - for_each_cpu(cpu, to_cpumask(rcu_cpu_online_map)) - sum += RCU_DATA_CPU(cpu)->rcu_flipctr[lastidx]; - if (sum != 0) { - RCU_TRACE_ME(rcupreempt_trace_try_flip_ze1); - return 0; - } - - /* - * This ensures that the other CPUs see the call for - * memory barriers -after- the sum to zero has been - * detected here - */ - smp_mb(); /* ^^^^^^^^^^^^ */ - - /* Call for a memory barrier from each CPU. */ - for_each_cpu(cpu, to_cpumask(rcu_cpu_online_map)) { - per_cpu(rcu_mb_flag, cpu) = rcu_mb_needed; - dyntick_save_progress_counter(cpu); - } - - RCU_TRACE_ME(rcupreempt_trace_try_flip_z2); - return 1; -} - -/* - * Wait for all CPUs to do their end-of-grace-period memory barrier. - * Return 0 once all CPUs have done so. - */ - -static int -rcu_try_flip_waitmb(void) -{ - int cpu; - - RCU_TRACE_ME(rcupreempt_trace_try_flip_m1); - for_each_cpu(cpu, to_cpumask(rcu_cpu_online_map)) - if (rcu_try_flip_waitmb_needed(cpu) && - per_cpu(rcu_mb_flag, cpu) != rcu_mb_done) { - RCU_TRACE_ME(rcupreempt_trace_try_flip_me1); - return 0; - } - - smp_mb(); /* Ensure that the above checks precede any following flip. */ - RCU_TRACE_ME(rcupreempt_trace_try_flip_m2); - return 1; -} - -/* - * Attempt a single flip of the counters. Remember, a single flip does - * -not- constitute a grace period. Instead, the interval between - * at least GP_STAGES consecutive flips is a grace period. - * - * If anyone is nuts enough to run this CONFIG_PREEMPT_RCU implementation - * on a large SMP, they might want to use a hierarchical organization of - * the per-CPU-counter pairs. - */ -static void rcu_try_flip(void) -{ - unsigned long flags; - - RCU_TRACE_ME(rcupreempt_trace_try_flip_1); - if (unlikely(!spin_trylock_irqsave(&rcu_ctrlblk.fliplock, flags))) { - RCU_TRACE_ME(rcupreempt_trace_try_flip_e1); - return; - } - - /* - * Take the next transition(s) through the RCU grace-period - * flip-counter state machine. - */ - - switch (rcu_ctrlblk.rcu_try_flip_state) { - case rcu_try_flip_idle_state: - if (rcu_try_flip_idle()) - rcu_ctrlblk.rcu_try_flip_state = - rcu_try_flip_waitack_state; - break; - case rcu_try_flip_waitack_state: - if (rcu_try_flip_waitack()) - rcu_ctrlblk.rcu_try_flip_state = - rcu_try_flip_waitzero_state; - break; - case rcu_try_flip_waitzero_state: - if (rcu_try_flip_waitzero()) - rcu_ctrlblk.rcu_try_flip_state = - rcu_try_flip_waitmb_state; - break; - case rcu_try_flip_waitmb_state: - if (rcu_try_flip_waitmb()) - rcu_ctrlblk.rcu_try_flip_state = - rcu_try_flip_idle_state; - } - spin_unlock_irqrestore(&rcu_ctrlblk.fliplock, flags); -} - -/* - * Check to see if this CPU needs to do a memory barrier in order to - * ensure that any prior RCU read-side critical sections have committed - * their counter manipulations and critical-section memory references - * before declaring the grace period to be completed. - */ -static void rcu_check_mb(int cpu) -{ - if (per_cpu(rcu_mb_flag, cpu) == rcu_mb_needed) { - smp_mb(); /* Ensure RCU read-side accesses are visible. */ - per_cpu(rcu_mb_flag, cpu) = rcu_mb_done; - } -} - -void rcu_check_callbacks(int cpu, int user) -{ - unsigned long flags; - struct rcu_data *rdp = RCU_DATA_CPU(cpu); - - /* - * If this CPU took its interrupt from user mode or from the - * idle loop, and this is not a nested interrupt, then - * this CPU has to have exited all prior preept-disable - * sections of code. So increment the counter to note this. - * - * The memory barrier is needed to handle the case where - * writes from a preempt-disable section of code get reordered - * into schedule() by this CPU's write buffer. So the memory - * barrier makes sure that the rcu_qsctr_inc() is seen by other - * CPUs to happen after any such write. - */ - - if (user || - (idle_cpu(cpu) && !in_softirq() && - hardirq_count() <= (1 << HARDIRQ_SHIFT))) { - smp_mb(); /* Guard against aggressive schedule(). */ - rcu_qsctr_inc(cpu); - } - - rcu_check_mb(cpu); - if (rcu_ctrlblk.completed == rdp->completed) - rcu_try_flip(); - spin_lock_irqsave(&rdp->lock, flags); - RCU_TRACE_RDP(rcupreempt_trace_check_callbacks, rdp); - __rcu_advance_callbacks(rdp); - if (rdp->donelist == NULL) { - spin_unlock_irqrestore(&rdp->lock, flags); - } else { - spin_unlock_irqrestore(&rdp->lock, flags); - raise_softirq(RCU_SOFTIRQ); - } -} - -/* - * Needed by dynticks, to make sure all RCU processing has finished - * when we go idle: - */ -void rcu_advance_callbacks(int cpu, int user) -{ - unsigned long flags; - struct rcu_data *rdp = RCU_DATA_CPU(cpu); - - if (rcu_ctrlblk.completed == rdp->completed) { - rcu_try_flip(); - if (rcu_ctrlblk.completed == rdp->completed) - return; - } - spin_lock_irqsave(&rdp->lock, flags); - RCU_TRACE_RDP(rcupreempt_trace_check_callbacks, rdp); - __rcu_advance_callbacks(rdp); - spin_unlock_irqrestore(&rdp->lock, flags); -} - -#ifdef CONFIG_HOTPLUG_CPU -#define rcu_offline_cpu_enqueue(srclist, srctail, dstlist, dsttail) do { \ - *dsttail = srclist; \ - if (srclist != NULL) { \ - dsttail = srctail; \ - srclist = NULL; \ - srctail = &srclist;\ - } \ - } while (0) - -void rcu_offline_cpu(int cpu) -{ - int i; - struct rcu_head *list = NULL; - unsigned long flags; - struct rcu_data *rdp = RCU_DATA_CPU(cpu); - struct rcu_head *schedlist = NULL; - struct rcu_head **schedtail = &schedlist; - struct rcu_head **tail = &list; - - /* - * Remove all callbacks from the newly dead CPU, retaining order. - * Otherwise rcu_barrier() will fail - */ - - spin_lock_irqsave(&rdp->lock, flags); - rcu_offline_cpu_enqueue(rdp->donelist, rdp->donetail, list, tail); - for (i = GP_STAGES - 1; i >= 0; i--) - rcu_offline_cpu_enqueue(rdp->waitlist[i], rdp->waittail[i], - list, tail); - rcu_offline_cpu_enqueue(rdp->nextlist, rdp->nexttail, list, tail); - rcu_offline_cpu_enqueue(rdp->waitschedlist, rdp->waitschedtail, - schedlist, schedtail); - rcu_offline_cpu_enqueue(rdp->nextschedlist, rdp->nextschedtail, - schedlist, schedtail); - rdp->rcu_sched_sleeping = 0; - spin_unlock_irqrestore(&rdp->lock, flags); - rdp->waitlistcount = 0; - - /* Disengage the newly dead CPU from the grace-period computation. */ - - spin_lock_irqsave(&rcu_ctrlblk.fliplock, flags); - rcu_check_mb(cpu); - if (per_cpu(rcu_flip_flag, cpu) == rcu_flipped) { - smp_mb(); /* Subsequent counter accesses must see new value */ - per_cpu(rcu_flip_flag, cpu) = rcu_flip_seen; - smp_mb(); /* Subsequent RCU read-side critical sections */ - /* seen -after- acknowledgement. */ - } - - RCU_DATA_ME()->rcu_flipctr[0] += RCU_DATA_CPU(cpu)->rcu_flipctr[0]; - RCU_DATA_ME()->rcu_flipctr[1] += RCU_DATA_CPU(cpu)->rcu_flipctr[1]; - - RCU_DATA_CPU(cpu)->rcu_flipctr[0] = 0; - RCU_DATA_CPU(cpu)->rcu_flipctr[1] = 0; - - cpumask_clear_cpu(cpu, to_cpumask(rcu_cpu_online_map)); - - spin_unlock_irqrestore(&rcu_ctrlblk.fliplock, flags); - - /* - * Place the removed callbacks on the current CPU's queue. - * Make them all start a new grace period: simple approach, - * in theory could starve a given set of callbacks, but - * you would need to be doing some serious CPU hotplugging - * to make this happen. If this becomes a problem, adding - * a synchronize_rcu() to the hotplug path would be a simple - * fix. - */ - - local_irq_save(flags); /* disable preempt till we know what lock. */ - rdp = RCU_DATA_ME(); - spin_lock(&rdp->lock); - *rdp->nexttail = list; - if (list) - rdp->nexttail = tail; - *rdp->nextschedtail = schedlist; - if (schedlist) - rdp->nextschedtail = schedtail; - spin_unlock_irqrestore(&rdp->lock, flags); -} - -#else /* #ifdef CONFIG_HOTPLUG_CPU */ - -void rcu_offline_cpu(int cpu) -{ -} - -#endif /* #else #ifdef CONFIG_HOTPLUG_CPU */ - -void __cpuinit rcu_online_cpu(int cpu) -{ - unsigned long flags; - struct rcu_data *rdp; - - spin_lock_irqsave(&rcu_ctrlblk.fliplock, flags); - cpumask_set_cpu(cpu, to_cpumask(rcu_cpu_online_map)); - spin_unlock_irqrestore(&rcu_ctrlblk.fliplock, flags); - - /* - * The rcu_sched grace-period processing might have bypassed - * this CPU, given that it was not in the rcu_cpu_online_map - * when the grace-period scan started. This means that the - * grace-period task might sleep. So make sure that if this - * should happen, the first callback posted to this CPU will - * wake up the grace-period task if need be. - */ - - rdp = RCU_DATA_CPU(cpu); - spin_lock_irqsave(&rdp->lock, flags); - rdp->rcu_sched_sleeping = 1; - spin_unlock_irqrestore(&rdp->lock, flags); -} - -static void rcu_process_callbacks(struct softirq_action *unused) -{ - unsigned long flags; - struct rcu_head *next, *list; - struct rcu_data *rdp; - - local_irq_save(flags); - rdp = RCU_DATA_ME(); - spin_lock(&rdp->lock); - list = rdp->donelist; - if (list == NULL) { - spin_unlock_irqrestore(&rdp->lock, flags); - return; - } - rdp->donelist = NULL; - rdp->donetail = &rdp->donelist; - RCU_TRACE_RDP(rcupreempt_trace_done_remove, rdp); - spin_unlock_irqrestore(&rdp->lock, flags); - while (list) { - next = list->next; - list->func(list); - list = next; - RCU_TRACE_ME(rcupreempt_trace_invoke); - } -} - -void call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu)) -{ - unsigned long flags; - struct rcu_data *rdp; - - head->func = func; - head->next = NULL; - local_irq_save(flags); - rdp = RCU_DATA_ME(); - spin_lock(&rdp->lock); - __rcu_advance_callbacks(rdp); - *rdp->nexttail = head; - rdp->nexttail = &head->next; - RCU_TRACE_RDP(rcupreempt_trace_next_add, rdp); - spin_unlock_irqrestore(&rdp->lock, flags); -} -EXPORT_SYMBOL_GPL(call_rcu); - -void call_rcu_sched(struct rcu_head *head, void (*func)(struct rcu_head *rcu)) -{ - unsigned long flags; - struct rcu_data *rdp; - int wake_gp = 0; - - head->func = func; - head->next = NULL; - local_irq_save(flags); - rdp = RCU_DATA_ME(); - spin_lock(&rdp->lock); - *rdp->nextschedtail = head; - rdp->nextschedtail = &head->next; - if (rdp->rcu_sched_sleeping) { - - /* Grace-period processing might be sleeping... */ - - rdp->rcu_sched_sleeping = 0; - wake_gp = 1; - } - spin_unlock_irqrestore(&rdp->lock, flags); - if (wake_gp) { - - /* Wake up grace-period processing, unless someone beat us. */ - - spin_lock_irqsave(&rcu_ctrlblk.schedlock, flags); - if (rcu_ctrlblk.sched_sleep != rcu_sched_sleeping) - wake_gp = 0; - rcu_ctrlblk.sched_sleep = rcu_sched_not_sleeping; - spin_unlock_irqrestore(&rcu_ctrlblk.schedlock, flags); - if (wake_gp) - wake_up_interruptible(&rcu_ctrlblk.sched_wq); - } -} -EXPORT_SYMBOL_GPL(call_rcu_sched); - -/* - * Wait until all currently running preempt_disable() code segments - * (including hardware-irq-disable segments) complete. Note that - * in -rt this does -not- necessarily result in all currently executing - * interrupt -handlers- having completed. - */ -void __synchronize_sched(void) -{ - struct rcu_synchronize rcu; - - if (num_online_cpus() == 1) - return; /* blocking is gp if only one CPU! */ - - init_completion(&rcu.completion); - /* Will wake me after RCU finished. */ - call_rcu_sched(&rcu.head, wakeme_after_rcu); - /* Wait for it. */ - wait_for_completion(&rcu.completion); -} -EXPORT_SYMBOL_GPL(__synchronize_sched); - -/* - * kthread function that manages call_rcu_sched grace periods. - */ -static int rcu_sched_grace_period(void *arg) -{ - int couldsleep; /* might sleep after current pass. */ - int couldsleepnext = 0; /* might sleep after next pass. */ - int cpu; - unsigned long flags; - struct rcu_data *rdp; - int ret; - - /* - * Each pass through the following loop handles one - * rcu_sched grace period cycle. - */ - do { - /* Save each CPU's current state. */ - - for_each_online_cpu(cpu) { - dyntick_save_progress_counter_sched(cpu); - save_qsctr_sched(cpu); - } - - /* - * Sleep for about an RCU grace-period's worth to - * allow better batching and to consume less CPU. - */ - schedule_timeout_interruptible(RCU_SCHED_BATCH_TIME); - - /* - * If there was nothing to do last time, prepare to - * sleep at the end of the current grace period cycle. - */ - couldsleep = couldsleepnext; - couldsleepnext = 1; - if (couldsleep) { - spin_lock_irqsave(&rcu_ctrlblk.schedlock, flags); - rcu_ctrlblk.sched_sleep = rcu_sched_sleep_prep; - spin_unlock_irqrestore(&rcu_ctrlblk.schedlock, flags); - } - - /* - * Wait on each CPU in turn to have either visited - * a quiescent state or been in dynticks-idle mode. - */ - for_each_online_cpu(cpu) { - while (rcu_qsctr_inc_needed(cpu) && - rcu_qsctr_inc_needed_dyntick(cpu)) { - /* resched_cpu(cpu); @@@ */ - schedule_timeout_interruptible(1); - } - } - - /* Advance callbacks for each CPU. */ - - for_each_online_cpu(cpu) { - - rdp = RCU_DATA_CPU(cpu); - spin_lock_irqsave(&rdp->lock, flags); - - /* - * We are running on this CPU irq-disabled, so no - * CPU can go offline until we re-enable irqs. - * The current CPU might have already gone - * offline (between the for_each_offline_cpu and - * the spin_lock_irqsave), but in that case all its - * callback lists will be empty, so no harm done. - * - * Advance the callbacks! We share normal RCU's - * donelist, since callbacks are invoked the - * same way in either case. - */ - if (rdp->waitschedlist != NULL) { - *rdp->donetail = rdp->waitschedlist; - rdp->donetail = rdp->waitschedtail; - - /* - * Next rcu_check_callbacks() will - * do the required raise_softirq(). - */ - } - if (rdp->nextschedlist != NULL) { - rdp->waitschedlist = rdp->nextschedlist; - rdp->waitschedtail = rdp->nextschedtail; - couldsleep = 0; - couldsleepnext = 0; - } else { - rdp->waitschedlist = NULL; - rdp->waitschedtail = &rdp->waitschedlist; - } - rdp->nextschedlist = NULL; - rdp->nextschedtail = &rdp->nextschedlist; - - /* Mark sleep intention. */ - - rdp->rcu_sched_sleeping = couldsleep; - - spin_unlock_irqrestore(&rdp->lock, flags); - } - - /* If we saw callbacks on the last scan, go deal with them. */ - - if (!couldsleep) - continue; - - /* Attempt to block... */ - - spin_lock_irqsave(&rcu_ctrlblk.schedlock, flags); - if (rcu_ctrlblk.sched_sleep != rcu_sched_sleep_prep) { - - /* - * Someone posted a callback after we scanned. - * Go take care of it. - */ - spin_unlock_irqrestore(&rcu_ctrlblk.schedlock, flags); - couldsleepnext = 0; - continue; - } - - /* Block until the next person posts a callback. */ - - rcu_ctrlblk.sched_sleep = rcu_sched_sleeping; - spin_unlock_irqrestore(&rcu_ctrlblk.schedlock, flags); - ret = 0; - __wait_event_interruptible(rcu_ctrlblk.sched_wq, - rcu_ctrlblk.sched_sleep != rcu_sched_sleeping, - ret); - - /* - * Signals would prevent us from sleeping, and we cannot - * do much with them in any case. So flush them. - */ - if (ret) - flush_signals(current); - couldsleepnext = 0; - - } while (!kthread_should_stop()); - - return (0); -} - -/* - * Check to see if any future RCU-related work will need to be done - * by the current CPU, even if none need be done immediately, returning - * 1 if so. Assumes that notifiers would take care of handling any - * outstanding requests from the RCU core. - * - * This function is part of the RCU implementation; it is -not- - * an exported member of the RCU API. - */ -int rcu_needs_cpu(int cpu) -{ - struct rcu_data *rdp = RCU_DATA_CPU(cpu); - - return (rdp->donelist != NULL || - !!rdp->waitlistcount || - rdp->nextlist != NULL || - rdp->nextschedlist != NULL || - rdp->waitschedlist != NULL); -} - -int rcu_pending(int cpu) -{ - struct rcu_data *rdp = RCU_DATA_CPU(cpu); - - /* The CPU has at least one callback queued somewhere. */ - - if (rdp->donelist != NULL || - !!rdp->waitlistcount || - rdp->nextlist != NULL || - rdp->nextschedlist != NULL || - rdp->waitschedlist != NULL) - return 1; - - /* The RCU core needs an acknowledgement from this CPU. */ - - if ((per_cpu(rcu_flip_flag, cpu) == rcu_flipped) || - (per_cpu(rcu_mb_flag, cpu) == rcu_mb_needed)) - return 1; - - /* This CPU has fallen behind the global grace-period number. */ - - if (rdp->completed != rcu_ctrlblk.completed) - return 1; - - /* Nothing needed from this CPU. */ - - return 0; -} - -static int __cpuinit rcu_cpu_notify(struct notifier_block *self, - unsigned long action, void *hcpu) -{ - long cpu = (long)hcpu; - - switch (action) { - case CPU_UP_PREPARE: - case CPU_UP_PREPARE_FROZEN: - rcu_online_cpu(cpu); - break; - case CPU_UP_CANCELED: - case CPU_UP_CANCELED_FROZEN: - case CPU_DEAD: - case CPU_DEAD_FROZEN: - rcu_offline_cpu(cpu); - break; - default: - break; - } - return NOTIFY_OK; -} - -static struct notifier_block __cpuinitdata rcu_nb = { - .notifier_call = rcu_cpu_notify, -}; - -void __init __rcu_init(void) -{ - int cpu; - int i; - struct rcu_data *rdp; - - printk(KERN_NOTICE "Preemptible RCU implementation.\n"); - for_each_possible_cpu(cpu) { - rdp = RCU_DATA_CPU(cpu); - spin_lock_init(&rdp->lock); - rdp->completed = 0; - rdp->waitlistcount = 0; - rdp->nextlist = NULL; - rdp->nexttail = &rdp->nextlist; - for (i = 0; i < GP_STAGES; i++) { - rdp->waitlist[i] = NULL; - rdp->waittail[i] = &rdp->waitlist[i]; - } - rdp->donelist = NULL; - rdp->donetail = &rdp->donelist; - rdp->rcu_flipctr[0] = 0; - rdp->rcu_flipctr[1] = 0; - rdp->nextschedlist = NULL; - rdp->nextschedtail = &rdp->nextschedlist; - rdp->waitschedlist = NULL; - rdp->waitschedtail = &rdp->waitschedlist; - rdp->rcu_sched_sleeping = 0; - } - register_cpu_notifier(&rcu_nb); - - /* - * We don't need protection against CPU-Hotplug here - * since - * a) If a CPU comes online while we are iterating over the - * cpu_online_mask below, we would only end up making a - * duplicate call to rcu_online_cpu() which sets the corresponding - * CPU's mask in the rcu_cpu_online_map. - * - * b) A CPU cannot go offline at this point in time since the user - * does not have access to the sysfs interface, nor do we - * suspend the system. - */ - for_each_online_cpu(cpu) - rcu_cpu_notify(&rcu_nb, CPU_UP_PREPARE, (void *)(long) cpu); - - open_softirq(RCU_SOFTIRQ, rcu_process_callbacks); -} - -/* - * Late-boot-time RCU initialization that must wait until after scheduler - * has been initialized. - */ -void __init rcu_init_sched(void) -{ - rcu_sched_grace_period_task = kthread_run(rcu_sched_grace_period, - NULL, - "rcu_sched_grace_period"); - WARN_ON(IS_ERR(rcu_sched_grace_period_task)); -} - -#ifdef CONFIG_RCU_TRACE -long *rcupreempt_flipctr(int cpu) -{ - return &RCU_DATA_CPU(cpu)->rcu_flipctr[0]; -} -EXPORT_SYMBOL_GPL(rcupreempt_flipctr); - -int rcupreempt_flip_flag(int cpu) -{ - return per_cpu(rcu_flip_flag, cpu); -} -EXPORT_SYMBOL_GPL(rcupreempt_flip_flag); - -int rcupreempt_mb_flag(int cpu) -{ - return per_cpu(rcu_mb_flag, cpu); -} -EXPORT_SYMBOL_GPL(rcupreempt_mb_flag); - -char *rcupreempt_try_flip_state_name(void) -{ - return rcu_try_flip_state_names[rcu_ctrlblk.rcu_try_flip_state]; -} -EXPORT_SYMBOL_GPL(rcupreempt_try_flip_state_name); - -struct rcupreempt_trace *rcupreempt_trace_cpu(int cpu) -{ - struct rcu_data *rdp = RCU_DATA_CPU(cpu); - - return &rdp->trace; -} -EXPORT_SYMBOL_GPL(rcupreempt_trace_cpu); - -#endif /* #ifdef RCU_TRACE */ diff --git a/kernel/rcupreempt_trace.c b/kernel/rcupreempt_trace.c deleted file mode 100644 index 7c2665cac17..00000000000 --- a/kernel/rcupreempt_trace.c +++ /dev/null @@ -1,334 +0,0 @@ -/* - * Read-Copy Update tracing for realtime implementation - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or - * (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software - * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. - * - * Copyright IBM Corporation, 2006 - * - * Papers: http://www.rdrop.com/users/paulmck/RCU - * - * For detailed explanation of Read-Copy Update mechanism see - - * Documentation/RCU/ *.txt - * - */ -#include <linux/types.h> -#include <linux/kernel.h> -#include <linux/init.h> -#include <linux/spinlock.h> -#include <linux/smp.h> -#include <linux/rcupdate.h> -#include <linux/interrupt.h> -#include <linux/sched.h> -#include <asm/atomic.h> -#include <linux/bitops.h> -#include <linux/module.h> -#include <linux/completion.h> -#include <linux/moduleparam.h> -#include <linux/percpu.h> -#include <linux/notifier.h> -#include <linux/cpu.h> -#include <linux/mutex.h> -#include <linux/rcupreempt_trace.h> -#include <linux/debugfs.h> - -static struct mutex rcupreempt_trace_mutex; -static char *rcupreempt_trace_buf; -#define RCUPREEMPT_TRACE_BUF_SIZE 4096 - -void rcupreempt_trace_move2done(struct rcupreempt_trace *trace) -{ - trace->done_length += trace->wait_length; - trace->done_add += trace->wait_length; - trace->wait_length = 0; -} -void rcupreempt_trace_move2wait(struct rcupreempt_trace *trace) -{ - trace->wait_length += trace->next_length; - trace->wait_add += trace->next_length; - trace->next_length = 0; -} -void rcupreempt_trace_try_flip_1(struct rcupreempt_trace *trace) -{ - atomic_inc(&trace->rcu_try_flip_1); -} -void rcupreempt_trace_try_flip_e1(struct rcupreempt_trace *trace) -{ - atomic_inc(&trace->rcu_try_flip_e1); -} -void rcupreempt_trace_try_flip_i1(struct rcupreempt_trace *trace) -{ - trace->rcu_try_flip_i1++; -} -void rcupreempt_trace_try_flip_ie1(struct rcupreempt_trace *trace) -{ - trace->rcu_try_flip_ie1++; -} -void rcupreempt_trace_try_flip_g1(struct rcupreempt_trace *trace) -{ - trace->rcu_try_flip_g1++; -} -void rcupreempt_trace_try_flip_a1(struct rcupreempt_trace *trace) -{ - trace->rcu_try_flip_a1++; -} -void rcupreempt_trace_try_flip_ae1(struct rcupreempt_trace *trace) -{ - trace->rcu_try_flip_ae1++; -} -void rcupreempt_trace_try_flip_a2(struct rcupreempt_trace *trace) -{ - trace->rcu_try_flip_a2++; -} -void rcupreempt_trace_try_flip_z1(struct rcupreempt_trace *trace) -{ - trace->rcu_try_flip_z1++; -} -void rcupreempt_trace_try_flip_ze1(struct rcupreempt_trace *trace) -{ - trace->rcu_try_flip_ze1++; -} -void rcupreempt_trace_try_flip_z2(struct rcupreempt_trace *trace) -{ - trace->rcu_try_flip_z2++; -} -void rcupreempt_trace_try_flip_m1(struct rcupreempt_trace *trace) -{ - trace->rcu_try_flip_m1++; -} -void rcupreempt_trace_try_flip_me1(struct rcupreempt_trace *trace) -{ - trace->rcu_try_flip_me1++; -} -void rcupreempt_trace_try_flip_m2(struct rcupreempt_trace *trace) -{ - trace->rcu_try_flip_m2++; -} -void rcupreempt_trace_check_callbacks(struct rcupreempt_trace *trace) -{ - trace->rcu_check_callbacks++; -} -void rcupreempt_trace_done_remove(struct rcupreempt_trace *trace) -{ - trace->done_remove += trace->done_length; - trace->done_length = 0; -} -void rcupreempt_trace_invoke(struct rcupreempt_trace *trace) -{ - atomic_inc(&trace->done_invoked); -} -void rcupreempt_trace_next_add(struct rcupreempt_trace *trace) -{ - trace->next_add++; - trace->next_length++; -} - -static void rcupreempt_trace_sum(struct rcupreempt_trace *sp) -{ - struct rcupreempt_trace *cp; - int cpu; - - memset(sp, 0, sizeof(*sp)); - for_each_possible_cpu(cpu) { - cp = rcupreempt_trace_cpu(cpu); - sp->next_length += cp->next_length; - sp->next_add += cp->next_add; - sp->wait_length += cp->wait_length; - sp->wait_add += cp->wait_add; - sp->done_length += cp->done_length; - sp->done_add += cp->done_add; - sp->done_remove += cp->done_remove; - atomic_add(atomic_read(&cp->done_invoked), &sp->done_invoked); - sp->rcu_check_callbacks += cp->rcu_check_callbacks; - atomic_add(atomic_read(&cp->rcu_try_flip_1), - &sp->rcu_try_flip_1); - atomic_add(atomic_read(&cp->rcu_try_flip_e1), - &sp->rcu_try_flip_e1); - sp->rcu_try_flip_i1 += cp->rcu_try_flip_i1; - sp->rcu_try_flip_ie1 += cp->rcu_try_flip_ie1; - sp->rcu_try_flip_g1 += cp->rcu_try_flip_g1; - sp->rcu_try_flip_a1 += cp->rcu_try_flip_a1; - sp->rcu_try_flip_ae1 += cp->rcu_try_flip_ae1; - sp->rcu_try_flip_a2 += cp->rcu_try_flip_a2; - sp->rcu_try_flip_z1 += cp->rcu_try_flip_z1; - sp->rcu_try_flip_ze1 += cp->rcu_try_flip_ze1; - sp->rcu_try_flip_z2 += cp->rcu_try_flip_z2; - sp->rcu_try_flip_m1 += cp->rcu_try_flip_m1; - sp->rcu_try_flip_me1 += cp->rcu_try_flip_me1; - sp->rcu_try_flip_m2 += cp->rcu_try_flip_m2; - } -} - -static ssize_t rcustats_read(struct file *filp, char __user *buffer, - size_t count, loff_t *ppos) -{ - struct rcupreempt_trace trace; - ssize_t bcount; - int cnt = 0; - - rcupreempt_trace_sum(&trace); - mutex_lock(&rcupreempt_trace_mutex); - snprintf(&rcupreempt_trace_buf[cnt], RCUPREEMPT_TRACE_BUF_SIZE - cnt, - "ggp=%ld rcc=%ld\n", - rcu_batches_completed(), - trace.rcu_check_callbacks); - snprintf(&rcupreempt_trace_buf[cnt], RCUPREEMPT_TRACE_BUF_SIZE - cnt, - "na=%ld nl=%ld wa=%ld wl=%ld da=%ld dl=%ld dr=%ld di=%d\n" - "1=%d e1=%d i1=%ld ie1=%ld g1=%ld a1=%ld ae1=%ld a2=%ld\n" - "z1=%ld ze1=%ld z2=%ld m1=%ld me1=%ld m2=%ld\n", - - trace.next_add, trace.next_length, - trace.wait_add, trace.wait_length, - trace.done_add, trace.done_length, - trace.done_remove, atomic_read(&trace.done_invoked), - atomic_read(&trace.rcu_try_flip_1), - atomic_read(&trace.rcu_try_flip_e1), - trace.rcu_try_flip_i1, trace.rcu_try_flip_ie1, - trace.rcu_try_flip_g1, - trace.rcu_try_flip_a1, trace.rcu_try_flip_ae1, - trace.rcu_try_flip_a2, - trace.rcu_try_flip_z1, trace.rcu_try_flip_ze1, - trace.rcu_try_flip_z2, - trace.rcu_try_flip_m1, trace.rcu_try_flip_me1, - trace.rcu_try_flip_m2); - bcount = simple_read_from_buffer(buffer, count, ppos, - rcupreempt_trace_buf, strlen(rcupreempt_trace_buf)); - mutex_unlock(&rcupreempt_trace_mutex); - return bcount; -} - -static ssize_t rcugp_read(struct file *filp, char __user *buffer, - size_t count, loff_t *ppos) -{ - long oldgp = rcu_batches_completed(); - ssize_t bcount; - - mutex_lock(&rcupreempt_trace_mutex); - synchronize_rcu(); - snprintf(rcupreempt_trace_buf, RCUPREEMPT_TRACE_BUF_SIZE, - "oldggp=%ld newggp=%ld\n", oldgp, rcu_batches_completed()); - bcount = simple_read_from_buffer(buffer, count, ppos, - rcupreempt_trace_buf, strlen(rcupreempt_trace_buf)); - mutex_unlock(&rcupreempt_trace_mutex); - return bcount; -} - -static ssize_t rcuctrs_read(struct file *filp, char __user *buffer, - size_t count, loff_t *ppos) -{ - int cnt = 0; - int cpu; - int f = rcu_batches_completed() & 0x1; - ssize_t bcount; - - mutex_lock(&rcupreempt_trace_mutex); - - cnt += snprintf(&rcupreempt_trace_buf[cnt], RCUPREEMPT_TRACE_BUF_SIZE, - "CPU last cur F M\n"); - for_each_online_cpu(cpu) { - long *flipctr = rcupreempt_flipctr(cpu); - cnt += snprintf(&rcupreempt_trace_buf[cnt], - RCUPREEMPT_TRACE_BUF_SIZE - cnt, - "%3d %4ld %3ld %d %d\n", - cpu, - flipctr[!f], - flipctr[f], - rcupreempt_flip_flag(cpu), - rcupreempt_mb_flag(cpu)); - } - cnt += snprintf(&rcupreempt_trace_buf[cnt], - RCUPREEMPT_TRACE_BUF_SIZE - cnt, - "ggp = %ld, state = %s\n", - rcu_batches_completed(), - rcupreempt_try_flip_state_name()); - cnt += snprintf(&rcupreempt_trace_buf[cnt], - RCUPREEMPT_TRACE_BUF_SIZE - cnt, - "\n"); - bcount = simple_read_from_buffer(buffer, count, ppos, - rcupreempt_trace_buf, strlen(rcupreempt_trace_buf)); - mutex_unlock(&rcupreempt_trace_mutex); - return bcount; -} - -static struct file_operations rcustats_fops = { - .owner = THIS_MODULE, - .read = rcustats_read, -}; - -static struct file_operations rcugp_fops = { - .owner = THIS_MODULE, - .read = rcugp_read, -}; - -static struct file_operations rcuctrs_fops = { - .owner = THIS_MODULE, - .read = rcuctrs_read, -}; - -static struct dentry *rcudir, *statdir, *ctrsdir, *gpdir; -static int rcupreempt_debugfs_init(void) -{ - rcudir = debugfs_create_dir("rcu", NULL); - if (!rcudir) - goto out; - statdir = debugfs_create_file("rcustats", 0444, rcudir, - NULL, &rcustats_fops); - if (!statdir) - goto free_out; - - gpdir = debugfs_create_file("rcugp", 0444, rcudir, NULL, &rcugp_fops); - if (!gpdir) - goto free_out; - - ctrsdir = debugfs_create_file("rcuctrs", 0444, rcudir, - NULL, &rcuctrs_fops); - if (!ctrsdir) - goto free_out; - return 0; -free_out: - if (statdir) - debugfs_remove(statdir); - if (gpdir) - debugfs_remove(gpdir); - debugfs_remove(rcudir); -out: - return 1; -} - -static int __init rcupreempt_trace_init(void) -{ - int ret; - - mutex_init(&rcupreempt_trace_mutex); - rcupreempt_trace_buf = kmalloc(RCUPREEMPT_TRACE_BUF_SIZE, GFP_KERNEL); - if (!rcupreempt_trace_buf) - return 1; - ret = rcupreempt_debugfs_init(); - if (ret) - kfree(rcupreempt_trace_buf); - return ret; -} - -static void __exit rcupreempt_trace_cleanup(void) -{ - debugfs_remove(statdir); - debugfs_remove(gpdir); - debugfs_remove(ctrsdir); - debugfs_remove(rcudir); - kfree(rcupreempt_trace_buf); -} - - -module_init(rcupreempt_trace_init); -module_exit(rcupreempt_trace_cleanup); diff --git a/kernel/rcutiny.c b/kernel/rcutiny.c new file mode 100644 index 00000000000..9f6d9ff2572 --- /dev/null +++ b/kernel/rcutiny.c @@ -0,0 +1,282 @@ +/* + * Read-Copy Update mechanism for mutual exclusion, the Bloatwatch edition. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + * + * Copyright IBM Corporation, 2008 + * + * Author: Paul E. McKenney <paulmck@linux.vnet.ibm.com> + * + * For detailed explanation of Read-Copy Update mechanism see - + * Documentation/RCU + */ +#include <linux/moduleparam.h> +#include <linux/completion.h> +#include <linux/interrupt.h> +#include <linux/notifier.h> +#include <linux/rcupdate.h> +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/mutex.h> +#include <linux/sched.h> +#include <linux/types.h> +#include <linux/init.h> +#include <linux/time.h> +#include <linux/cpu.h> + +/* Global control variables for rcupdate callback mechanism. */ +struct rcu_ctrlblk { + struct rcu_head *rcucblist; /* List of pending callbacks (CBs). */ + struct rcu_head **donetail; /* ->next pointer of last "done" CB. */ + struct rcu_head **curtail; /* ->next pointer of last CB. */ +}; + +/* Definition for rcupdate control block. */ +static struct rcu_ctrlblk rcu_ctrlblk = { + .donetail = &rcu_ctrlblk.rcucblist, + .curtail = &rcu_ctrlblk.rcucblist, +}; + +static struct rcu_ctrlblk rcu_bh_ctrlblk = { + .donetail = &rcu_bh_ctrlblk.rcucblist, + .curtail = &rcu_bh_ctrlblk.rcucblist, +}; + +#ifdef CONFIG_NO_HZ + +static long rcu_dynticks_nesting = 1; + +/* + * Enter dynticks-idle mode, which is an extended quiescent state + * if we have fully entered that mode (i.e., if the new value of + * dynticks_nesting is zero). + */ +void rcu_enter_nohz(void) +{ + if (--rcu_dynticks_nesting == 0) + rcu_sched_qs(0); /* implies rcu_bh_qsctr_inc(0) */ +} + +/* + * Exit dynticks-idle mode, so that we are no longer in an extended + * quiescent state. + */ +void rcu_exit_nohz(void) +{ + rcu_dynticks_nesting++; +} + +#endif /* #ifdef CONFIG_NO_HZ */ + +/* + * Helper function for rcu_qsctr_inc() and rcu_bh_qsctr_inc(). + * Also disable irqs to avoid confusion due to interrupt handlers + * invoking call_rcu(). + */ +static int rcu_qsctr_help(struct rcu_ctrlblk *rcp) +{ + unsigned long flags; + + local_irq_save(flags); + if (rcp->rcucblist != NULL && + rcp->donetail != rcp->curtail) { + rcp->donetail = rcp->curtail; + local_irq_restore(flags); + return 1; + } + local_irq_restore(flags); + + return 0; +} + +/* + * Record an rcu quiescent state. And an rcu_bh quiescent state while we + * are at it, given that any rcu quiescent state is also an rcu_bh + * quiescent state. Use "+" instead of "||" to defeat short circuiting. + */ +void rcu_sched_qs(int cpu) +{ + if (rcu_qsctr_help(&rcu_ctrlblk) + rcu_qsctr_help(&rcu_bh_ctrlblk)) + raise_softirq(RCU_SOFTIRQ); +} + +/* + * Record an rcu_bh quiescent state. + */ +void rcu_bh_qs(int cpu) +{ + if (rcu_qsctr_help(&rcu_bh_ctrlblk)) + raise_softirq(RCU_SOFTIRQ); +} + +/* + * Check to see if the scheduling-clock interrupt came from an extended + * quiescent state, and, if so, tell RCU about it. + */ +void rcu_check_callbacks(int cpu, int user) +{ + if (user || + (idle_cpu(cpu) && + !in_softirq() && + hardirq_count() <= (1 << HARDIRQ_SHIFT))) + rcu_sched_qs(cpu); + else if (!in_softirq()) + rcu_bh_qs(cpu); +} + +/* + * Helper function for rcu_process_callbacks() that operates on the + * specified rcu_ctrlkblk structure. + */ +static void __rcu_process_callbacks(struct rcu_ctrlblk *rcp) +{ + struct rcu_head *next, *list; + unsigned long flags; + + /* If no RCU callbacks ready to invoke, just return. */ + if (&rcp->rcucblist == rcp->donetail) + return; + + /* Move the ready-to-invoke callbacks to a local list. */ + local_irq_save(flags); + list = rcp->rcucblist; + rcp->rcucblist = *rcp->donetail; + *rcp->donetail = NULL; + if (rcp->curtail == rcp->donetail) + rcp->curtail = &rcp->rcucblist; + rcp->donetail = &rcp->rcucblist; + local_irq_restore(flags); + + /* Invoke the callbacks on the local list. */ + while (list) { + next = list->next; + prefetch(next); + list->func(list); + list = next; + } +} + +/* + * Invoke any callbacks whose grace period has completed. + */ +static void rcu_process_callbacks(struct softirq_action *unused) +{ + __rcu_process_callbacks(&rcu_ctrlblk); + __rcu_process_callbacks(&rcu_bh_ctrlblk); +} + +/* + * Wait for a grace period to elapse. But it is illegal to invoke + * synchronize_sched() from within an RCU read-side critical section. + * Therefore, any legal call to synchronize_sched() is a quiescent + * state, and so on a UP system, synchronize_sched() need do nothing. + * Ditto for synchronize_rcu_bh(). (But Lai Jiangshan points out the + * benefits of doing might_sleep() to reduce latency.) + * + * Cool, huh? (Due to Josh Triplett.) + * + * But we want to make this a static inline later. + */ +void synchronize_sched(void) +{ + cond_resched(); +} +EXPORT_SYMBOL_GPL(synchronize_sched); + +void synchronize_rcu_bh(void) +{ + synchronize_sched(); +} +EXPORT_SYMBOL_GPL(synchronize_rcu_bh); + +/* + * Helper function for call_rcu() and call_rcu_bh(). + */ +static void __call_rcu(struct rcu_head *head, + void (*func)(struct rcu_head *rcu), + struct rcu_ctrlblk *rcp) +{ + unsigned long flags; + + head->func = func; + head->next = NULL; + + local_irq_save(flags); + *rcp->curtail = head; + rcp->curtail = &head->next; + local_irq_restore(flags); +} + +/* + * Post an RCU callback to be invoked after the end of an RCU grace + * period. But since we have but one CPU, that would be after any + * quiescent state. + */ +void call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu)) +{ + __call_rcu(head, func, &rcu_ctrlblk); +} +EXPORT_SYMBOL_GPL(call_rcu); + +/* + * Post an RCU bottom-half callback to be invoked after any subsequent + * quiescent state. + */ +void call_rcu_bh(struct rcu_head *head, void (*func)(struct rcu_head *rcu)) +{ + __call_rcu(head, func, &rcu_bh_ctrlblk); +} +EXPORT_SYMBOL_GPL(call_rcu_bh); + +void rcu_barrier(void) +{ + struct rcu_synchronize rcu; + + init_completion(&rcu.completion); + /* Will wake me after RCU finished. */ + call_rcu(&rcu.head, wakeme_after_rcu); + /* Wait for it. */ + wait_for_completion(&rcu.completion); +} +EXPORT_SYMBOL_GPL(rcu_barrier); + +void rcu_barrier_bh(void) +{ + struct rcu_synchronize rcu; + + init_completion(&rcu.completion); + /* Will wake me after RCU finished. */ + call_rcu_bh(&rcu.head, wakeme_after_rcu); + /* Wait for it. */ + wait_for_completion(&rcu.completion); +} +EXPORT_SYMBOL_GPL(rcu_barrier_bh); + +void rcu_barrier_sched(void) +{ + struct rcu_synchronize rcu; + + init_completion(&rcu.completion); + /* Will wake me after RCU finished. */ + call_rcu_sched(&rcu.head, wakeme_after_rcu); + /* Wait for it. */ + wait_for_completion(&rcu.completion); +} +EXPORT_SYMBOL_GPL(rcu_barrier_sched); + +void __init rcu_init(void) +{ + open_softirq(RCU_SOFTIRQ, rcu_process_callbacks); +} diff --git a/kernel/rcutorture.c b/kernel/rcutorture.c index 9b4a975a4b4..a621a67ef4e 100644 --- a/kernel/rcutorture.c +++ b/kernel/rcutorture.c @@ -18,7 +18,7 @@ * Copyright (C) IBM Corporation, 2005, 2006 * * Authors: Paul E. McKenney <paulmck@us.ibm.com> - * Josh Triplett <josh@freedesktop.org> + * Josh Triplett <josh@freedesktop.org> * * See also: Documentation/RCU/torture.txt */ @@ -50,7 +50,7 @@ MODULE_LICENSE("GPL"); MODULE_AUTHOR("Paul E. McKenney <paulmck@us.ibm.com> and " - "Josh Triplett <josh@freedesktop.org>"); + "Josh Triplett <josh@freedesktop.org>"); static int nreaders = -1; /* # reader threads, defaults to 2*ncpus */ static int nfakewriters = 4; /* # fake writer threads */ @@ -110,8 +110,8 @@ struct rcu_torture { }; static LIST_HEAD(rcu_torture_freelist); -static struct rcu_torture *rcu_torture_current = NULL; -static long rcu_torture_current_version = 0; +static struct rcu_torture *rcu_torture_current; +static long rcu_torture_current_version; static struct rcu_torture rcu_tortures[10 * RCU_TORTURE_PIPE_LEN]; static DEFINE_SPINLOCK(rcu_torture_lock); static DEFINE_PER_CPU(long [RCU_TORTURE_PIPE_LEN + 1], rcu_torture_count) = @@ -124,11 +124,11 @@ static atomic_t n_rcu_torture_alloc_fail; static atomic_t n_rcu_torture_free; static atomic_t n_rcu_torture_mberror; static atomic_t n_rcu_torture_error; -static long n_rcu_torture_timers = 0; +static long n_rcu_torture_timers; static struct list_head rcu_torture_removed; static cpumask_var_t shuffle_tmp_mask; -static int stutter_pause_test = 0; +static int stutter_pause_test; #if defined(MODULE) || defined(CONFIG_RCU_TORTURE_TEST_RUNNABLE) #define RCUTORTURE_RUNNABLE_INIT 1 @@ -257,17 +257,18 @@ struct rcu_torture_ops { void (*init)(void); void (*cleanup)(void); int (*readlock)(void); - void (*readdelay)(struct rcu_random_state *rrsp); + void (*read_delay)(struct rcu_random_state *rrsp); void (*readunlock)(int idx); int (*completed)(void); - void (*deferredfree)(struct rcu_torture *p); + void (*deferred_free)(struct rcu_torture *p); void (*sync)(void); void (*cb_barrier)(void); int (*stats)(char *page); - int irqcapable; + int irq_capable; char *name; }; -static struct rcu_torture_ops *cur_ops = NULL; + +static struct rcu_torture_ops *cur_ops; /* * Definitions for rcu torture testing. @@ -281,14 +282,17 @@ static int rcu_torture_read_lock(void) __acquires(RCU) static void rcu_read_delay(struct rcu_random_state *rrsp) { - long delay; - const long longdelay = 200; + const unsigned long shortdelay_us = 200; + const unsigned long longdelay_ms = 50; - /* We want there to be long-running readers, but not all the time. */ + /* We want a short delay sometimes to make a reader delay the grace + * period, and we want a long delay occasionally to trigger + * force_quiescent_state. */ - delay = rcu_random(rrsp) % (nrealreaders * 2 * longdelay); - if (!delay) - udelay(longdelay); + if (!(rcu_random(rrsp) % (nrealreaders * 2000 * longdelay_ms))) + mdelay(longdelay_ms); + if (!(rcu_random(rrsp) % (nrealreaders * 2 * shortdelay_us))) + udelay(shortdelay_us); } static void rcu_torture_read_unlock(int idx) __releases(RCU) @@ -320,7 +324,12 @@ rcu_torture_cb(struct rcu_head *p) rp->rtort_mbtest = 0; rcu_torture_free(rp); } else - cur_ops->deferredfree(rp); + cur_ops->deferred_free(rp); +} + +static int rcu_no_completed(void) +{ + return 0; } static void rcu_torture_deferred_free(struct rcu_torture *p) @@ -329,18 +338,18 @@ static void rcu_torture_deferred_free(struct rcu_torture *p) } static struct rcu_torture_ops rcu_ops = { - .init = NULL, - .cleanup = NULL, - .readlock = rcu_torture_read_lock, - .readdelay = rcu_read_delay, - .readunlock = rcu_torture_read_unlock, - .completed = rcu_torture_completed, - .deferredfree = rcu_torture_deferred_free, - .sync = synchronize_rcu, - .cb_barrier = rcu_barrier, - .stats = NULL, - .irqcapable = 1, - .name = "rcu" + .init = NULL, + .cleanup = NULL, + .readlock = rcu_torture_read_lock, + .read_delay = rcu_read_delay, + .readunlock = rcu_torture_read_unlock, + .completed = rcu_torture_completed, + .deferred_free = rcu_torture_deferred_free, + .sync = synchronize_rcu, + .cb_barrier = rcu_barrier, + .stats = NULL, + .irq_capable = 1, + .name = "rcu" }; static void rcu_sync_torture_deferred_free(struct rcu_torture *p) @@ -370,18 +379,33 @@ static void rcu_sync_torture_init(void) } static struct rcu_torture_ops rcu_sync_ops = { - .init = rcu_sync_torture_init, - .cleanup = NULL, - .readlock = rcu_torture_read_lock, - .readdelay = rcu_read_delay, - .readunlock = rcu_torture_read_unlock, - .completed = rcu_torture_completed, - .deferredfree = rcu_sync_torture_deferred_free, - .sync = synchronize_rcu, - .cb_barrier = NULL, - .stats = NULL, - .irqcapable = 1, - .name = "rcu_sync" + .init = rcu_sync_torture_init, + .cleanup = NULL, + .readlock = rcu_torture_read_lock, + .read_delay = rcu_read_delay, + .readunlock = rcu_torture_read_unlock, + .completed = rcu_torture_completed, + .deferred_free = rcu_sync_torture_deferred_free, + .sync = synchronize_rcu, + .cb_barrier = NULL, + .stats = NULL, + .irq_capable = 1, + .name = "rcu_sync" +}; + +static struct rcu_torture_ops rcu_expedited_ops = { + .init = rcu_sync_torture_init, + .cleanup = NULL, + .readlock = rcu_torture_read_lock, + .read_delay = rcu_read_delay, /* just reuse rcu's version. */ + .readunlock = rcu_torture_read_unlock, + .completed = rcu_no_completed, + .deferred_free = rcu_sync_torture_deferred_free, + .sync = synchronize_rcu_expedited, + .cb_barrier = NULL, + .stats = NULL, + .irq_capable = 1, + .name = "rcu_expedited" }; /* @@ -432,33 +456,33 @@ static void rcu_bh_torture_synchronize(void) } static struct rcu_torture_ops rcu_bh_ops = { - .init = NULL, - .cleanup = NULL, - .readlock = rcu_bh_torture_read_lock, - .readdelay = rcu_read_delay, /* just reuse rcu's version. */ - .readunlock = rcu_bh_torture_read_unlock, - .completed = rcu_bh_torture_completed, - .deferredfree = rcu_bh_torture_deferred_free, - .sync = rcu_bh_torture_synchronize, - .cb_barrier = rcu_barrier_bh, - .stats = NULL, - .irqcapable = 1, - .name = "rcu_bh" + .init = NULL, + .cleanup = NULL, + .readlock = rcu_bh_torture_read_lock, + .read_delay = rcu_read_delay, /* just reuse rcu's version. */ + .readunlock = rcu_bh_torture_read_unlock, + .completed = rcu_bh_torture_completed, + .deferred_free = rcu_bh_torture_deferred_free, + .sync = rcu_bh_torture_synchronize, + .cb_barrier = rcu_barrier_bh, + .stats = NULL, + .irq_capable = 1, + .name = "rcu_bh" }; static struct rcu_torture_ops rcu_bh_sync_ops = { - .init = rcu_sync_torture_init, - .cleanup = NULL, - .readlock = rcu_bh_torture_read_lock, - .readdelay = rcu_read_delay, /* just reuse rcu's version. */ - .readunlock = rcu_bh_torture_read_unlock, - .completed = rcu_bh_torture_completed, - .deferredfree = rcu_sync_torture_deferred_free, - .sync = rcu_bh_torture_synchronize, - .cb_barrier = NULL, - .stats = NULL, - .irqcapable = 1, - .name = "rcu_bh_sync" + .init = rcu_sync_torture_init, + .cleanup = NULL, + .readlock = rcu_bh_torture_read_lock, + .read_delay = rcu_read_delay, /* just reuse rcu's version. */ + .readunlock = rcu_bh_torture_read_unlock, + .completed = rcu_bh_torture_completed, + .deferred_free = rcu_sync_torture_deferred_free, + .sync = rcu_bh_torture_synchronize, + .cb_barrier = NULL, + .stats = NULL, + .irq_capable = 1, + .name = "rcu_bh_sync" }; /* @@ -530,17 +554,36 @@ static int srcu_torture_stats(char *page) } static struct rcu_torture_ops srcu_ops = { - .init = srcu_torture_init, - .cleanup = srcu_torture_cleanup, - .readlock = srcu_torture_read_lock, - .readdelay = srcu_read_delay, - .readunlock = srcu_torture_read_unlock, - .completed = srcu_torture_completed, - .deferredfree = rcu_sync_torture_deferred_free, - .sync = srcu_torture_synchronize, - .cb_barrier = NULL, - .stats = srcu_torture_stats, - .name = "srcu" + .init = srcu_torture_init, + .cleanup = srcu_torture_cleanup, + .readlock = srcu_torture_read_lock, + .read_delay = srcu_read_delay, + .readunlock = srcu_torture_read_unlock, + .completed = srcu_torture_completed, + .deferred_free = rcu_sync_torture_deferred_free, + .sync = srcu_torture_synchronize, + .cb_barrier = NULL, + .stats = srcu_torture_stats, + .name = "srcu" +}; + +static void srcu_torture_synchronize_expedited(void) +{ + synchronize_srcu_expedited(&srcu_ctl); +} + +static struct rcu_torture_ops srcu_expedited_ops = { + .init = srcu_torture_init, + .cleanup = srcu_torture_cleanup, + .readlock = srcu_torture_read_lock, + .read_delay = srcu_read_delay, + .readunlock = srcu_torture_read_unlock, + .completed = srcu_torture_completed, + .deferred_free = rcu_sync_torture_deferred_free, + .sync = srcu_torture_synchronize_expedited, + .cb_barrier = NULL, + .stats = srcu_torture_stats, + .name = "srcu_expedited" }; /* @@ -558,11 +601,6 @@ static void sched_torture_read_unlock(int idx) preempt_enable(); } -static int sched_torture_completed(void) -{ - return 0; -} - static void rcu_sched_torture_deferred_free(struct rcu_torture *p) { call_rcu_sched(&p->rtort_rcu, rcu_torture_cb); @@ -574,32 +612,47 @@ static void sched_torture_synchronize(void) } static struct rcu_torture_ops sched_ops = { - .init = rcu_sync_torture_init, - .cleanup = NULL, - .readlock = sched_torture_read_lock, - .readdelay = rcu_read_delay, /* just reuse rcu's version. */ - .readunlock = sched_torture_read_unlock, - .completed = sched_torture_completed, - .deferredfree = rcu_sched_torture_deferred_free, - .sync = sched_torture_synchronize, - .cb_barrier = rcu_barrier_sched, - .stats = NULL, - .irqcapable = 1, - .name = "sched" + .init = rcu_sync_torture_init, + .cleanup = NULL, + .readlock = sched_torture_read_lock, + .read_delay = rcu_read_delay, /* just reuse rcu's version. */ + .readunlock = sched_torture_read_unlock, + .completed = rcu_no_completed, + .deferred_free = rcu_sched_torture_deferred_free, + .sync = sched_torture_synchronize, + .cb_barrier = rcu_barrier_sched, + .stats = NULL, + .irq_capable = 1, + .name = "sched" +}; + +static struct rcu_torture_ops sched_sync_ops = { + .init = rcu_sync_torture_init, + .cleanup = NULL, + .readlock = sched_torture_read_lock, + .read_delay = rcu_read_delay, /* just reuse rcu's version. */ + .readunlock = sched_torture_read_unlock, + .completed = rcu_no_completed, + .deferred_free = rcu_sync_torture_deferred_free, + .sync = sched_torture_synchronize, + .cb_barrier = NULL, + .stats = NULL, + .name = "sched_sync" }; -static struct rcu_torture_ops sched_ops_sync = { - .init = rcu_sync_torture_init, - .cleanup = NULL, - .readlock = sched_torture_read_lock, - .readdelay = rcu_read_delay, /* just reuse rcu's version. */ - .readunlock = sched_torture_read_unlock, - .completed = sched_torture_completed, - .deferredfree = rcu_sync_torture_deferred_free, - .sync = sched_torture_synchronize, - .cb_barrier = NULL, - .stats = NULL, - .name = "sched_sync" +static struct rcu_torture_ops sched_expedited_ops = { + .init = rcu_sync_torture_init, + .cleanup = NULL, + .readlock = sched_torture_read_lock, + .read_delay = rcu_read_delay, /* just reuse rcu's version. */ + .readunlock = sched_torture_read_unlock, + .completed = rcu_no_completed, + .deferred_free = rcu_sync_torture_deferred_free, + .sync = synchronize_sched_expedited, + .cb_barrier = NULL, + .stats = rcu_expedited_torture_stats, + .irq_capable = 1, + .name = "sched_expedited" }; /* @@ -621,21 +674,22 @@ rcu_torture_writer(void *arg) do { schedule_timeout_uninterruptible(1); - if ((rp = rcu_torture_alloc()) == NULL) + rp = rcu_torture_alloc(); + if (rp == NULL) continue; rp->rtort_pipe_count = 0; udelay(rcu_random(&rand) & 0x3ff); old_rp = rcu_torture_current; rp->rtort_mbtest = 1; rcu_assign_pointer(rcu_torture_current, rp); - smp_wmb(); + smp_wmb(); /* Mods to old_rp must follow rcu_assign_pointer() */ if (old_rp) { i = old_rp->rtort_pipe_count; if (i > RCU_TORTURE_PIPE_LEN) i = RCU_TORTURE_PIPE_LEN; atomic_inc(&rcu_torture_wcount[i]); old_rp->rtort_pipe_count++; - cur_ops->deferredfree(old_rp); + cur_ops->deferred_free(old_rp); } rcu_torture_current_version++; oldbatch = cur_ops->completed(); @@ -700,7 +754,7 @@ static void rcu_torture_timer(unsigned long unused) if (p->rtort_mbtest == 0) atomic_inc(&n_rcu_torture_mberror); spin_lock(&rand_lock); - cur_ops->readdelay(&rand); + cur_ops->read_delay(&rand); n_rcu_torture_timers++; spin_unlock(&rand_lock); preempt_disable(); @@ -738,11 +792,11 @@ rcu_torture_reader(void *arg) VERBOSE_PRINTK_STRING("rcu_torture_reader task started"); set_user_nice(current, 19); - if (irqreader && cur_ops->irqcapable) + if (irqreader && cur_ops->irq_capable) setup_timer_on_stack(&t, rcu_torture_timer, 0); do { - if (irqreader && cur_ops->irqcapable) { + if (irqreader && cur_ops->irq_capable) { if (!timer_pending(&t)) mod_timer(&t, 1); } @@ -757,7 +811,7 @@ rcu_torture_reader(void *arg) } if (p->rtort_mbtest == 0) atomic_inc(&n_rcu_torture_mberror); - cur_ops->readdelay(&rand); + cur_ops->read_delay(&rand); preempt_disable(); pipe_count = p->rtort_pipe_count; if (pipe_count > RCU_TORTURE_PIPE_LEN) { @@ -778,7 +832,7 @@ rcu_torture_reader(void *arg) } while (!kthread_should_stop() && fullstop == FULLSTOP_DONTSTOP); VERBOSE_PRINTK_STRING("rcu_torture_reader task stopping"); rcutorture_shutdown_absorb("rcu_torture_reader"); - if (irqreader && cur_ops->irqcapable) + if (irqreader && cur_ops->irq_capable) del_timer_sync(&t); while (!kthread_should_stop()) schedule_timeout_uninterruptible(1); @@ -1077,8 +1131,10 @@ rcu_torture_init(void) int cpu; int firsterr = 0; static struct rcu_torture_ops *torture_ops[] = - { &rcu_ops, &rcu_sync_ops, &rcu_bh_ops, &rcu_bh_sync_ops, - &srcu_ops, &sched_ops, &sched_ops_sync, }; + { &rcu_ops, &rcu_sync_ops, &rcu_expedited_ops, + &rcu_bh_ops, &rcu_bh_sync_ops, + &srcu_ops, &srcu_expedited_ops, + &sched_ops, &sched_sync_ops, &sched_expedited_ops, }; mutex_lock(&fullstop_mutex); @@ -1089,10 +1145,14 @@ rcu_torture_init(void) break; } if (i == ARRAY_SIZE(torture_ops)) { - printk(KERN_ALERT "rcutorture: invalid torture type: \"%s\"\n", + printk(KERN_ALERT "rcu-torture: invalid torture type: \"%s\"\n", torture_type); + printk(KERN_ALERT "rcu-torture types:"); + for (i = 0; i < ARRAY_SIZE(torture_ops); i++) + printk(KERN_ALERT " %s", torture_ops[i]->name); + printk(KERN_ALERT "\n"); mutex_unlock(&fullstop_mutex); - return (-EINVAL); + return -EINVAL; } if (cur_ops->init) cur_ops->init(); /* no "goto unwind" prior to this point!!! */ @@ -1143,7 +1203,7 @@ rcu_torture_init(void) goto unwind; } fakewriter_tasks = kzalloc(nfakewriters * sizeof(fakewriter_tasks[0]), - GFP_KERNEL); + GFP_KERNEL); if (fakewriter_tasks == NULL) { VERBOSE_PRINTK_ERRSTRING("out of memory"); firsterr = -ENOMEM; @@ -1152,7 +1212,7 @@ rcu_torture_init(void) for (i = 0; i < nfakewriters; i++) { VERBOSE_PRINTK_STRING("Creating rcu_torture_fakewriter task"); fakewriter_tasks[i] = kthread_run(rcu_torture_fakewriter, NULL, - "rcu_torture_fakewriter"); + "rcu_torture_fakewriter"); if (IS_ERR(fakewriter_tasks[i])) { firsterr = PTR_ERR(fakewriter_tasks[i]); VERBOSE_PRINTK_ERRSTRING("Failed to create fakewriter"); diff --git a/kernel/rcutree.c b/kernel/rcutree.c index 97ce31579ec..53ae9598f79 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -25,7 +25,7 @@ * and inputs from Rusty Russell, Andrea Arcangeli and Andi Kleen. * * For detailed explanation of Read-Copy Update mechanism see - - * Documentation/RCU + * Documentation/RCU */ #include <linux/types.h> #include <linux/kernel.h> @@ -35,6 +35,7 @@ #include <linux/rcupdate.h> #include <linux/interrupt.h> #include <linux/sched.h> +#include <linux/nmi.h> #include <asm/atomic.h> #include <linux/bitops.h> #include <linux/module.h> @@ -45,39 +46,80 @@ #include <linux/cpu.h> #include <linux/mutex.h> #include <linux/time.h> +#include <linux/kernel_stat.h> -#ifdef CONFIG_DEBUG_LOCK_ALLOC -static struct lock_class_key rcu_lock_key; -struct lockdep_map rcu_lock_map = - STATIC_LOCKDEP_MAP_INIT("rcu_read_lock", &rcu_lock_key); -EXPORT_SYMBOL_GPL(rcu_lock_map); -#endif +#include "rcutree.h" /* Data structures. */ +static struct lock_class_key rcu_node_class[NUM_RCU_LVLS]; + #define RCU_STATE_INITIALIZER(name) { \ .level = { &name.node[0] }, \ .levelcnt = { \ NUM_RCU_LVL_0, /* root of hierarchy. */ \ NUM_RCU_LVL_1, \ NUM_RCU_LVL_2, \ - NUM_RCU_LVL_3, /* == MAX_RCU_LVLS */ \ + NUM_RCU_LVL_3, \ + NUM_RCU_LVL_4, /* == MAX_RCU_LVLS */ \ }, \ - .signaled = RCU_SIGNAL_INIT, \ + .signaled = RCU_GP_IDLE, \ .gpnum = -300, \ .completed = -300, \ .onofflock = __SPIN_LOCK_UNLOCKED(&name.onofflock), \ + .orphan_cbs_list = NULL, \ + .orphan_cbs_tail = &name.orphan_cbs_list, \ + .orphan_qlen = 0, \ .fqslock = __SPIN_LOCK_UNLOCKED(&name.fqslock), \ .n_force_qs = 0, \ .n_force_qs_ngp = 0, \ } -struct rcu_state rcu_state = RCU_STATE_INITIALIZER(rcu_state); -DEFINE_PER_CPU(struct rcu_data, rcu_data); +struct rcu_state rcu_sched_state = RCU_STATE_INITIALIZER(rcu_sched_state); +DEFINE_PER_CPU(struct rcu_data, rcu_sched_data); struct rcu_state rcu_bh_state = RCU_STATE_INITIALIZER(rcu_bh_state); DEFINE_PER_CPU(struct rcu_data, rcu_bh_data); +static int rcu_scheduler_active __read_mostly; + + +/* + * Return true if an RCU grace period is in progress. The ACCESS_ONCE()s + * permit this function to be invoked without holding the root rcu_node + * structure's ->lock, but of course results can be subject to change. + */ +static int rcu_gp_in_progress(struct rcu_state *rsp) +{ + return ACCESS_ONCE(rsp->completed) != ACCESS_ONCE(rsp->gpnum); +} + +/* + * Note a quiescent state. Because we do not need to know + * how many quiescent states passed, just if there was at least + * one since the start of the grace period, this just sets a flag. + */ +void rcu_sched_qs(int cpu) +{ + struct rcu_data *rdp; + + rdp = &per_cpu(rcu_sched_data, cpu); + rdp->passed_quiesc_completed = rdp->gpnum - 1; + barrier(); + rdp->passed_quiesc = 1; + rcu_preempt_note_context_switch(cpu); +} + +void rcu_bh_qs(int cpu) +{ + struct rcu_data *rdp; + + rdp = &per_cpu(rcu_bh_data, cpu); + rdp->passed_quiesc_completed = rdp->gpnum - 1; + barrier(); + rdp->passed_quiesc = 1; +} + #ifdef CONFIG_NO_HZ DEFINE_PER_CPU(struct rcu_dynticks, rcu_dynticks) = { .dynticks_nesting = 1, @@ -89,16 +131,21 @@ static int blimit = 10; /* Maximum callbacks per softirq. */ static int qhimark = 10000; /* If this many pending, ignore blimit. */ static int qlowmark = 100; /* Once only this many pending, use blimit. */ +module_param(blimit, int, 0); +module_param(qhimark, int, 0); +module_param(qlowmark, int, 0); + static void force_quiescent_state(struct rcu_state *rsp, int relaxed); +static int rcu_pending(int cpu); /* - * Return the number of RCU batches processed thus far for debug & stats. + * Return the number of RCU-sched batches processed thus far for debug & stats. */ -long rcu_batches_completed(void) +long rcu_batches_completed_sched(void) { - return rcu_state.completed; + return rcu_sched_state.completed; } -EXPORT_SYMBOL_GPL(rcu_batches_completed); +EXPORT_SYMBOL_GPL(rcu_batches_completed_sched); /* * Return the number of RCU BH batches processed thus far for debug & stats. @@ -124,9 +171,7 @@ cpu_has_callbacks_ready_to_invoke(struct rcu_data *rdp) static int cpu_needs_another_gp(struct rcu_state *rsp, struct rcu_data *rdp) { - /* ACCESS_ONCE() because we are accessing outside of lock. */ - return *rdp->nxttail[RCU_DONE_TAIL] && - ACCESS_ONCE(rsp->completed) == ACCESS_ONCE(rsp->gpnum); + return *rdp->nxttail[RCU_DONE_TAIL] && !rcu_gp_in_progress(rsp); } /* @@ -161,6 +206,10 @@ static int rcu_implicit_offline_qs(struct rcu_data *rdp) return 1; } + /* If preemptable RCU, no point in sending reschedule IPI. */ + if (rdp->preemptable) + return 0; + /* The CPU is online, so send it a reschedule IPI. */ if (rdp->cpu != smp_processor_id()) smp_send_reschedule(rdp->cpu); @@ -173,7 +222,6 @@ static int rcu_implicit_offline_qs(struct rcu_data *rdp) #endif /* #ifdef CONFIG_SMP */ #ifdef CONFIG_NO_HZ -static DEFINE_RATELIMIT_STATE(rcu_rs, 10 * HZ, 5); /** * rcu_enter_nohz - inform RCU that current CPU is entering nohz @@ -193,7 +241,7 @@ void rcu_enter_nohz(void) rdtp = &__get_cpu_var(rcu_dynticks); rdtp->dynticks++; rdtp->dynticks_nesting--; - WARN_ON_RATELIMIT(rdtp->dynticks & 0x1, &rcu_rs); + WARN_ON_ONCE(rdtp->dynticks & 0x1); local_irq_restore(flags); } @@ -212,7 +260,7 @@ void rcu_exit_nohz(void) rdtp = &__get_cpu_var(rcu_dynticks); rdtp->dynticks++; rdtp->dynticks_nesting++; - WARN_ON_RATELIMIT(!(rdtp->dynticks & 0x1), &rcu_rs); + WARN_ON_ONCE(!(rdtp->dynticks & 0x1)); local_irq_restore(flags); smp_mb(); /* CPUs seeing ++ must see later RCU read-side crit sects */ } @@ -231,7 +279,7 @@ void rcu_nmi_enter(void) if (rdtp->dynticks & 0x1) return; rdtp->dynticks_nmi++; - WARN_ON_RATELIMIT(!(rdtp->dynticks_nmi & 0x1), &rcu_rs); + WARN_ON_ONCE(!(rdtp->dynticks_nmi & 0x1)); smp_mb(); /* CPUs seeing ++ must see later RCU read-side crit sects */ } @@ -250,7 +298,7 @@ void rcu_nmi_exit(void) return; smp_mb(); /* CPUs seeing ++ must see prior RCU read-side crit sects */ rdtp->dynticks_nmi++; - WARN_ON_RATELIMIT(rdtp->dynticks_nmi & 0x1, &rcu_rs); + WARN_ON_ONCE(rdtp->dynticks_nmi & 0x1); } /** @@ -266,7 +314,7 @@ void rcu_irq_enter(void) if (rdtp->dynticks_nesting++) return; rdtp->dynticks++; - WARN_ON_RATELIMIT(!(rdtp->dynticks & 0x1), &rcu_rs); + WARN_ON_ONCE(!(rdtp->dynticks & 0x1)); smp_mb(); /* CPUs seeing ++ must see later RCU read-side crit sects */ } @@ -285,39 +333,20 @@ void rcu_irq_exit(void) return; smp_mb(); /* CPUs seeing ++ must see prior RCU read-side crit sects */ rdtp->dynticks++; - WARN_ON_RATELIMIT(rdtp->dynticks & 0x1, &rcu_rs); + WARN_ON_ONCE(rdtp->dynticks & 0x1); /* If the interrupt queued a callback, get out of dyntick mode. */ - if (__get_cpu_var(rcu_data).nxtlist || + if (__get_cpu_var(rcu_sched_data).nxtlist || __get_cpu_var(rcu_bh_data).nxtlist) set_need_resched(); } -/* - * Record the specified "completed" value, which is later used to validate - * dynticks counter manipulations. Specify "rsp->completed - 1" to - * unconditionally invalidate any future dynticks manipulations (which is - * useful at the beginning of a grace period). - */ -static void dyntick_record_completed(struct rcu_state *rsp, long comp) -{ - rsp->dynticks_completed = comp; -} - #ifdef CONFIG_SMP /* - * Recall the previously recorded value of the completion for dynticks. - */ -static long dyntick_recall_completed(struct rcu_state *rsp) -{ - return rsp->dynticks_completed; -} - -/* * Snapshot the specified CPU's dynticks counter so that we can later * credit them with an implicit quiescent state. Return 1 if this CPU - * is already in a quiescent state courtesy of dynticks idle mode. + * is in dynticks idle mode, which is an extended quiescent state. */ static int dyntick_save_progress_counter(struct rcu_data *rdp) { @@ -377,24 +406,8 @@ static int rcu_implicit_dynticks_qs(struct rcu_data *rdp) #else /* #ifdef CONFIG_NO_HZ */ -static void dyntick_record_completed(struct rcu_state *rsp, long comp) -{ -} - #ifdef CONFIG_SMP -/* - * If there are no dynticks, then the only way that a CPU can passively - * be in a quiescent state is to be offline. Unlike dynticks idle, which - * is a point in time during the prior (already finished) grace period, - * an offline CPU is always in a quiescent state, and thus can be - * unconditionally applied. So just return the current value of completed. - */ -static long dyntick_recall_completed(struct rcu_state *rsp) -{ - return rsp->completed; -} - static int dyntick_save_progress_counter(struct rcu_data *rdp) { return 0; @@ -423,32 +436,39 @@ static void print_other_cpu_stall(struct rcu_state *rsp) long delta; unsigned long flags; struct rcu_node *rnp = rcu_get_root(rsp); - struct rcu_node *rnp_cur = rsp->level[NUM_RCU_LVLS - 1]; - struct rcu_node *rnp_end = &rsp->node[NUM_RCU_NODES]; /* Only let one CPU complain about others per time interval. */ spin_lock_irqsave(&rnp->lock, flags); delta = jiffies - rsp->jiffies_stall; - if (delta < RCU_STALL_RAT_DELAY || rsp->gpnum == rsp->completed) { + if (delta < RCU_STALL_RAT_DELAY || !rcu_gp_in_progress(rsp)) { spin_unlock_irqrestore(&rnp->lock, flags); return; } rsp->jiffies_stall = jiffies + RCU_SECONDS_TILL_STALL_RECHECK; + + /* + * Now rat on any tasks that got kicked up to the root rcu_node + * due to CPU offlining. + */ + rcu_print_task_stall(rnp); spin_unlock_irqrestore(&rnp->lock, flags); /* OK, time to rat on our buddy... */ printk(KERN_ERR "INFO: RCU detected CPU stalls:"); - for (; rnp_cur < rnp_end; rnp_cur++) { - if (rnp_cur->qsmask == 0) + rcu_for_each_leaf_node(rsp, rnp) { + rcu_print_task_stall(rnp); + if (rnp->qsmask == 0) continue; - for (cpu = 0; cpu <= rnp_cur->grphi - rnp_cur->grplo; cpu++) - if (rnp_cur->qsmask & (1UL << cpu)) - printk(" %d", rnp_cur->grplo + cpu); + for (cpu = 0; cpu <= rnp->grphi - rnp->grplo; cpu++) + if (rnp->qsmask & (1UL << cpu)) + printk(" %d", rnp->grplo + cpu); } printk(" (detected by %d, t=%ld jiffies)\n", smp_processor_id(), (long)(jiffies - rsp->gp_start)); + trigger_all_cpu_backtrace(); + force_quiescent_state(rsp, 0); /* Kick them all. */ } @@ -459,12 +479,14 @@ static void print_cpu_stall(struct rcu_state *rsp) printk(KERN_ERR "INFO: RCU detected CPU %d stall (t=%lu jiffies)\n", smp_processor_id(), jiffies - rsp->gp_start); - dump_stack(); + trigger_all_cpu_backtrace(); + spin_lock_irqsave(&rnp->lock, flags); if ((long)(jiffies - rsp->jiffies_stall) >= 0) rsp->jiffies_stall = jiffies + RCU_SECONDS_TILL_STALL_RECHECK; spin_unlock_irqrestore(&rnp->lock, flags); + set_need_resched(); /* kick ourselves to get things going. */ } @@ -480,8 +502,7 @@ static void check_cpu_stall(struct rcu_state *rsp, struct rcu_data *rdp) /* We haven't checked in, so go dump stack. */ print_cpu_stall(rsp); - } else if (rsp->gpnum != rsp->completed && - delta >= RCU_STALL_RAT_DELAY) { + } else if (rcu_gp_in_progress(rsp) && delta >= RCU_STALL_RAT_DELAY) { /* They had two time units to dump stack, so complain. */ print_other_cpu_stall(rsp); @@ -503,15 +524,33 @@ static void check_cpu_stall(struct rcu_state *rsp, struct rcu_data *rdp) /* * Update CPU-local rcu_data state to record the newly noticed grace period. * This is used both when we started the grace period and when we notice - * that someone else started the grace period. + * that someone else started the grace period. The caller must hold the + * ->lock of the leaf rcu_node structure corresponding to the current CPU, + * and must have irqs disabled. */ +static void __note_new_gpnum(struct rcu_state *rsp, struct rcu_node *rnp, struct rcu_data *rdp) +{ + if (rdp->gpnum != rnp->gpnum) { + rdp->qs_pending = 1; + rdp->passed_quiesc = 0; + rdp->gpnum = rnp->gpnum; + } +} + static void note_new_gpnum(struct rcu_state *rsp, struct rcu_data *rdp) { - rdp->qs_pending = 1; - rdp->passed_quiesc = 0; - rdp->gpnum = rsp->gpnum; - rdp->n_rcu_pending_force_qs = rdp->n_rcu_pending + - RCU_JIFFIES_TILL_FORCE_QS; + unsigned long flags; + struct rcu_node *rnp; + + local_irq_save(flags); + rnp = rdp->mynode; + if (rdp->gpnum == ACCESS_ONCE(rnp->gpnum) || /* outside lock. */ + !spin_trylock(&rnp->lock)) { /* irqs already off, retry later. */ + local_irq_restore(flags); + return; + } + __note_new_gpnum(rsp, rnp, rdp); + spin_unlock_irqrestore(&rnp->lock, flags); } /* @@ -535,6 +574,79 @@ check_for_new_grace_period(struct rcu_state *rsp, struct rcu_data *rdp) } /* + * Advance this CPU's callbacks, but only if the current grace period + * has ended. This may be called only from the CPU to whom the rdp + * belongs. In addition, the corresponding leaf rcu_node structure's + * ->lock must be held by the caller, with irqs disabled. + */ +static void +__rcu_process_gp_end(struct rcu_state *rsp, struct rcu_node *rnp, struct rcu_data *rdp) +{ + /* Did another grace period end? */ + if (rdp->completed != rnp->completed) { + + /* Advance callbacks. No harm if list empty. */ + rdp->nxttail[RCU_DONE_TAIL] = rdp->nxttail[RCU_WAIT_TAIL]; + rdp->nxttail[RCU_WAIT_TAIL] = rdp->nxttail[RCU_NEXT_READY_TAIL]; + rdp->nxttail[RCU_NEXT_READY_TAIL] = rdp->nxttail[RCU_NEXT_TAIL]; + + /* Remember that we saw this grace-period completion. */ + rdp->completed = rnp->completed; + } +} + +/* + * Advance this CPU's callbacks, but only if the current grace period + * has ended. This may be called only from the CPU to whom the rdp + * belongs. + */ +static void +rcu_process_gp_end(struct rcu_state *rsp, struct rcu_data *rdp) +{ + unsigned long flags; + struct rcu_node *rnp; + + local_irq_save(flags); + rnp = rdp->mynode; + if (rdp->completed == ACCESS_ONCE(rnp->completed) || /* outside lock. */ + !spin_trylock(&rnp->lock)) { /* irqs already off, retry later. */ + local_irq_restore(flags); + return; + } + __rcu_process_gp_end(rsp, rnp, rdp); + spin_unlock_irqrestore(&rnp->lock, flags); +} + +/* + * Do per-CPU grace-period initialization for running CPU. The caller + * must hold the lock of the leaf rcu_node structure corresponding to + * this CPU. + */ +static void +rcu_start_gp_per_cpu(struct rcu_state *rsp, struct rcu_node *rnp, struct rcu_data *rdp) +{ + /* Prior grace period ended, so advance callbacks for current CPU. */ + __rcu_process_gp_end(rsp, rnp, rdp); + + /* + * Because this CPU just now started the new grace period, we know + * that all of its callbacks will be covered by this upcoming grace + * period, even the ones that were registered arbitrarily recently. + * Therefore, advance all outstanding callbacks to RCU_WAIT_TAIL. + * + * Other CPUs cannot be sure exactly when the grace period started. + * Therefore, their recently registered callbacks must pass through + * an additional RCU_NEXT_READY stage, so that they will be handled + * by the next RCU grace period. + */ + rdp->nxttail[RCU_NEXT_READY_TAIL] = rdp->nxttail[RCU_NEXT_TAIL]; + rdp->nxttail[RCU_WAIT_TAIL] = rdp->nxttail[RCU_NEXT_TAIL]; + + /* Set state so that this CPU will detect the next quiescent state. */ + __note_new_gpnum(rsp, rnp, rdp); +} + +/* * Start a new RCU grace period if warranted, re-initializing the hierarchy * in preparation for detecting the next grace period. The caller must hold * the root node's ->lock, which is released before return. Hard irqs must @@ -546,36 +658,43 @@ rcu_start_gp(struct rcu_state *rsp, unsigned long flags) { struct rcu_data *rdp = rsp->rda[smp_processor_id()]; struct rcu_node *rnp = rcu_get_root(rsp); - struct rcu_node *rnp_cur; - struct rcu_node *rnp_end; if (!cpu_needs_another_gp(rsp, rdp)) { - spin_unlock_irqrestore(&rnp->lock, flags); + if (rnp->completed == rsp->completed) { + spin_unlock_irqrestore(&rnp->lock, flags); + return; + } + spin_unlock(&rnp->lock); /* irqs remain disabled. */ + + /* + * Propagate new ->completed value to rcu_node structures + * so that other CPUs don't have to wait until the start + * of the next grace period to process their callbacks. + */ + rcu_for_each_node_breadth_first(rsp, rnp) { + spin_lock(&rnp->lock); /* irqs already disabled. */ + rnp->completed = rsp->completed; + spin_unlock(&rnp->lock); /* irqs remain disabled. */ + } + local_irq_restore(flags); return; } /* Advance to a new grace period and initialize state. */ rsp->gpnum++; + WARN_ON_ONCE(rsp->signaled == RCU_GP_INIT); rsp->signaled = RCU_GP_INIT; /* Hold off force_quiescent_state. */ rsp->jiffies_force_qs = jiffies + RCU_JIFFIES_TILL_FORCE_QS; - rdp->n_rcu_pending_force_qs = rdp->n_rcu_pending + - RCU_JIFFIES_TILL_FORCE_QS; record_gp_stall_check_time(rsp); - dyntick_record_completed(rsp, rsp->completed - 1); - note_new_gpnum(rsp, rdp); - - /* - * Because we are first, we know that all our callbacks will - * be covered by this upcoming grace period, even the ones - * that were registered arbitrarily recently. - */ - rdp->nxttail[RCU_NEXT_READY_TAIL] = rdp->nxttail[RCU_NEXT_TAIL]; - rdp->nxttail[RCU_WAIT_TAIL] = rdp->nxttail[RCU_NEXT_TAIL]; /* Special-case the common single-level case. */ if (NUM_RCU_NODES == 1) { + rcu_preempt_check_blocked_tasks(rnp); rnp->qsmask = rnp->qsmaskinit; + rnp->gpnum = rsp->gpnum; + rnp->completed = rsp->completed; rsp->signaled = RCU_SIGNAL_INIT; /* force_quiescent_state OK. */ + rcu_start_gp_per_cpu(rsp, rnp, rdp); spin_unlock_irqrestore(&rnp->lock, flags); return; } @@ -587,88 +706,71 @@ rcu_start_gp(struct rcu_state *rsp, unsigned long flags) spin_lock(&rsp->onofflock); /* irqs already disabled. */ /* - * Set the quiescent-state-needed bits in all the non-leaf RCU - * nodes for all currently online CPUs. This operation relies - * on the layout of the hierarchy within the rsp->node[] array. - * Note that other CPUs will access only the leaves of the - * hierarchy, which still indicate that no grace period is in - * progress. In addition, we have excluded CPU-hotplug operations. - * - * We therefore do not need to hold any locks. Any required - * memory barriers will be supplied by the locks guarding the - * leaf rcu_nodes in the hierarchy. - */ - - rnp_end = rsp->level[NUM_RCU_LVLS - 1]; - for (rnp_cur = &rsp->node[0]; rnp_cur < rnp_end; rnp_cur++) - rnp_cur->qsmask = rnp_cur->qsmaskinit; - - /* - * Now set up the leaf nodes. Here we must be careful. First, - * we need to hold the lock in order to exclude other CPUs, which - * might be contending for the leaf nodes' locks. Second, as - * soon as we initialize a given leaf node, its CPUs might run - * up the rest of the hierarchy. We must therefore acquire locks - * for each node that we touch during this stage. (But we still - * are excluding CPU-hotplug operations.) + * Set the quiescent-state-needed bits in all the rcu_node + * structures for all currently online CPUs in breadth-first + * order, starting from the root rcu_node structure. This + * operation relies on the layout of the hierarchy within the + * rsp->node[] array. Note that other CPUs will access only + * the leaves of the hierarchy, which still indicate that no + * grace period is in progress, at least until the corresponding + * leaf node has been initialized. In addition, we have excluded + * CPU-hotplug operations. * * Note that the grace period cannot complete until we finish * the initialization process, as there will be at least one * qsmask bit set in the root node until that time, namely the - * one corresponding to this CPU. + * one corresponding to this CPU, due to the fact that we have + * irqs disabled. */ - rnp_end = &rsp->node[NUM_RCU_NODES]; - rnp_cur = rsp->level[NUM_RCU_LVLS - 1]; - for (; rnp_cur < rnp_end; rnp_cur++) { - spin_lock(&rnp_cur->lock); /* irqs already disabled. */ - rnp_cur->qsmask = rnp_cur->qsmaskinit; - spin_unlock(&rnp_cur->lock); /* irqs already disabled. */ + rcu_for_each_node_breadth_first(rsp, rnp) { + spin_lock(&rnp->lock); /* irqs already disabled. */ + rcu_preempt_check_blocked_tasks(rnp); + rnp->qsmask = rnp->qsmaskinit; + rnp->gpnum = rsp->gpnum; + rnp->completed = rsp->completed; + if (rnp == rdp->mynode) + rcu_start_gp_per_cpu(rsp, rnp, rdp); + spin_unlock(&rnp->lock); /* irqs remain disabled. */ } + rnp = rcu_get_root(rsp); + spin_lock(&rnp->lock); /* irqs already disabled. */ rsp->signaled = RCU_SIGNAL_INIT; /* force_quiescent_state now OK. */ + spin_unlock(&rnp->lock); /* irqs remain disabled. */ spin_unlock_irqrestore(&rsp->onofflock, flags); } /* - * Advance this CPU's callbacks, but only if the current grace period - * has ended. This may be called only from the CPU to whom the rdp - * belongs. + * Report a full set of quiescent states to the specified rcu_state + * data structure. This involves cleaning up after the prior grace + * period and letting rcu_start_gp() start up the next grace period + * if one is needed. Note that the caller must hold rnp->lock, as + * required by rcu_start_gp(), which will release it. */ -static void -rcu_process_gp_end(struct rcu_state *rsp, struct rcu_data *rdp) +static void rcu_report_qs_rsp(struct rcu_state *rsp, unsigned long flags) + __releases(rcu_get_root(rsp)->lock) { - long completed_snap; - unsigned long flags; - - local_irq_save(flags); - completed_snap = ACCESS_ONCE(rsp->completed); /* outside of lock. */ - - /* Did another grace period end? */ - if (rdp->completed != completed_snap) { - - /* Advance callbacks. No harm if list empty. */ - rdp->nxttail[RCU_DONE_TAIL] = rdp->nxttail[RCU_WAIT_TAIL]; - rdp->nxttail[RCU_WAIT_TAIL] = rdp->nxttail[RCU_NEXT_READY_TAIL]; - rdp->nxttail[RCU_NEXT_READY_TAIL] = rdp->nxttail[RCU_NEXT_TAIL]; - - /* Remember that we saw this grace-period completion. */ - rdp->completed = completed_snap; - } - local_irq_restore(flags); + WARN_ON_ONCE(!rcu_gp_in_progress(rsp)); + rsp->completed = rsp->gpnum; + rsp->signaled = RCU_GP_IDLE; + rcu_start_gp(rsp, flags); /* releases root node's rnp->lock. */ } /* - * Similar to cpu_quiet(), for which it is a helper function. Allows - * a group of CPUs to be quieted at one go, though all the CPUs in the - * group must be represented by the same leaf rcu_node structure. - * That structure's lock must be held upon entry, and it is released - * before return. + * Similar to rcu_report_qs_rdp(), for which it is a helper function. + * Allows quiescent states for a group of CPUs to be reported at one go + * to the specified rcu_node structure, though all the CPUs in the group + * must be represented by the same rcu_node structure (which need not be + * a leaf rcu_node structure, though it often will be). That structure's + * lock must be held upon entry, and it is released before return. */ static void -cpu_quiet_msk(unsigned long mask, struct rcu_state *rsp, struct rcu_node *rnp, - unsigned long flags) +rcu_report_qs_rnp(unsigned long mask, struct rcu_state *rsp, + struct rcu_node *rnp, unsigned long flags) __releases(rnp->lock) { + struct rcu_node *rnp_c; + /* Walk up the rcu_node hierarchy. */ for (;;) { if (!(rnp->qsmask & mask)) { @@ -678,7 +780,7 @@ cpu_quiet_msk(unsigned long mask, struct rcu_state *rsp, struct rcu_node *rnp, return; } rnp->qsmask &= ~mask; - if (rnp->qsmask != 0) { + if (rnp->qsmask != 0 || rcu_preempted_readers(rnp)) { /* Other bits still set at this level, so done. */ spin_unlock_irqrestore(&rnp->lock, flags); @@ -692,31 +794,31 @@ cpu_quiet_msk(unsigned long mask, struct rcu_state *rsp, struct rcu_node *rnp, break; } spin_unlock_irqrestore(&rnp->lock, flags); + rnp_c = rnp; rnp = rnp->parent; spin_lock_irqsave(&rnp->lock, flags); + WARN_ON_ONCE(rnp_c->qsmask); } /* * Get here if we are the last CPU to pass through a quiescent - * state for this grace period. Clean up and let rcu_start_gp() - * start up the next grace period if one is needed. Note that - * we still hold rnp->lock, as required by rcu_start_gp(), which - * will release it. + * state for this grace period. Invoke rcu_report_qs_rsp() + * to clean up and start the next grace period if one is needed. */ - rsp->completed = rsp->gpnum; - rcu_process_gp_end(rsp, rsp->rda[smp_processor_id()]); - rcu_start_gp(rsp, flags); /* releases rnp->lock. */ + rcu_report_qs_rsp(rsp, flags); /* releases rnp->lock. */ } /* - * Record a quiescent state for the specified CPU, which must either be - * the current CPU or an offline CPU. The lastcomp argument is used to - * make sure we are still in the grace period of interest. We don't want - * to end the current grace period based on quiescent states detected in - * an earlier grace period! + * Record a quiescent state for the specified CPU to that CPU's rcu_data + * structure. This must be either called from the specified CPU, or + * called when the specified CPU is known to be offline (and when it is + * also known that no other CPU is concurrently trying to help the offline + * CPU). The lastcomp argument is used to make sure we are still in the + * grace period of interest. We don't want to end the current grace period + * based on quiescent states detected in an earlier grace period! */ static void -cpu_quiet(int cpu, struct rcu_state *rsp, struct rcu_data *rdp, long lastcomp) +rcu_report_qs_rdp(int cpu, struct rcu_state *rsp, struct rcu_data *rdp, long lastcomp) { unsigned long flags; unsigned long mask; @@ -724,15 +826,15 @@ cpu_quiet(int cpu, struct rcu_state *rsp, struct rcu_data *rdp, long lastcomp) rnp = rdp->mynode; spin_lock_irqsave(&rnp->lock, flags); - if (lastcomp != ACCESS_ONCE(rsp->completed)) { + if (lastcomp != rnp->completed) { /* * Someone beat us to it for this grace period, so leave. * The race with GP start is resolved by the fact that we * hold the leaf rcu_node lock, so that the per-CPU bits * cannot yet be initialized -- so we would simply find our - * CPU's bit already cleared in cpu_quiet_msk() if this race - * occurred. + * CPU's bit already cleared in rcu_report_qs_rnp() if this + * race occurred. */ rdp->passed_quiesc = 0; /* try again later! */ spin_unlock_irqrestore(&rnp->lock, flags); @@ -748,10 +850,9 @@ cpu_quiet(int cpu, struct rcu_state *rsp, struct rcu_data *rdp, long lastcomp) * This GP can't end until cpu checks in, so all of our * callbacks can be processed during the next GP. */ - rdp = rsp->rda[smp_processor_id()]; rdp->nxttail[RCU_NEXT_READY_TAIL] = rdp->nxttail[RCU_NEXT_TAIL]; - cpu_quiet_msk(mask, rsp, rnp, flags); /* releases rnp->lock */ + rcu_report_qs_rnp(mask, rsp, rnp, flags); /* rlses rnp->lock */ } } @@ -782,74 +883,113 @@ rcu_check_quiescent_state(struct rcu_state *rsp, struct rcu_data *rdp) if (!rdp->passed_quiesc) return; - /* Tell RCU we are done (but cpu_quiet() will be the judge of that). */ - cpu_quiet(rdp->cpu, rsp, rdp, rdp->passed_quiesc_completed); + /* + * Tell RCU we are done (but rcu_report_qs_rdp() will be the + * judge of that). + */ + rcu_report_qs_rdp(rdp->cpu, rsp, rdp, rdp->passed_quiesc_completed); } #ifdef CONFIG_HOTPLUG_CPU /* + * Move a dying CPU's RCU callbacks to the ->orphan_cbs_list for the + * specified flavor of RCU. The callbacks will be adopted by the next + * _rcu_barrier() invocation or by the CPU_DEAD notifier, whichever + * comes first. Because this is invoked from the CPU_DYING notifier, + * irqs are already disabled. + */ +static void rcu_send_cbs_to_orphanage(struct rcu_state *rsp) +{ + int i; + struct rcu_data *rdp = rsp->rda[smp_processor_id()]; + + if (rdp->nxtlist == NULL) + return; /* irqs disabled, so comparison is stable. */ + spin_lock(&rsp->onofflock); /* irqs already disabled. */ + *rsp->orphan_cbs_tail = rdp->nxtlist; + rsp->orphan_cbs_tail = rdp->nxttail[RCU_NEXT_TAIL]; + rdp->nxtlist = NULL; + for (i = 0; i < RCU_NEXT_SIZE; i++) + rdp->nxttail[i] = &rdp->nxtlist; + rsp->orphan_qlen += rdp->qlen; + rdp->qlen = 0; + spin_unlock(&rsp->onofflock); /* irqs remain disabled. */ +} + +/* + * Adopt previously orphaned RCU callbacks. + */ +static void rcu_adopt_orphan_cbs(struct rcu_state *rsp) +{ + unsigned long flags; + struct rcu_data *rdp; + + spin_lock_irqsave(&rsp->onofflock, flags); + rdp = rsp->rda[smp_processor_id()]; + if (rsp->orphan_cbs_list == NULL) { + spin_unlock_irqrestore(&rsp->onofflock, flags); + return; + } + *rdp->nxttail[RCU_NEXT_TAIL] = rsp->orphan_cbs_list; + rdp->nxttail[RCU_NEXT_TAIL] = rsp->orphan_cbs_tail; + rdp->qlen += rsp->orphan_qlen; + rsp->orphan_cbs_list = NULL; + rsp->orphan_cbs_tail = &rsp->orphan_cbs_list; + rsp->orphan_qlen = 0; + spin_unlock_irqrestore(&rsp->onofflock, flags); +} + +/* * Remove the outgoing CPU from the bitmasks in the rcu_node hierarchy * and move all callbacks from the outgoing CPU to the current one. */ static void __rcu_offline_cpu(int cpu, struct rcu_state *rsp) { - int i; unsigned long flags; - long lastcomp; unsigned long mask; + int need_report = 0; struct rcu_data *rdp = rsp->rda[cpu]; - struct rcu_data *rdp_me; struct rcu_node *rnp; /* Exclude any attempts to start a new grace period. */ spin_lock_irqsave(&rsp->onofflock, flags); /* Remove the outgoing CPU from the masks in the rcu_node hierarchy. */ - rnp = rdp->mynode; + rnp = rdp->mynode; /* this is the outgoing CPU's rnp. */ mask = rdp->grpmask; /* rnp->grplo is constant. */ do { spin_lock(&rnp->lock); /* irqs already disabled. */ rnp->qsmaskinit &= ~mask; if (rnp->qsmaskinit != 0) { - spin_unlock(&rnp->lock); /* irqs already disabled. */ + if (rnp != rdp->mynode) + spin_unlock(&rnp->lock); /* irqs remain disabled. */ break; } + if (rnp == rdp->mynode) + need_report = rcu_preempt_offline_tasks(rsp, rnp, rdp); + else + spin_unlock(&rnp->lock); /* irqs remain disabled. */ mask = rnp->grpmask; - spin_unlock(&rnp->lock); /* irqs already disabled. */ rnp = rnp->parent; } while (rnp != NULL); - lastcomp = rsp->completed; - - spin_unlock(&rsp->onofflock); /* irqs remain disabled. */ - - /* Being offline is a quiescent state, so go record it. */ - cpu_quiet(cpu, rsp, rdp, lastcomp); /* - * Move callbacks from the outgoing CPU to the running CPU. - * Note that the outgoing CPU is now quiscent, so it is now - * (uncharacteristically) safe to access it rcu_data structure. - * Note also that we must carefully retain the order of the - * outgoing CPU's callbacks in order for rcu_barrier() to work - * correctly. Finally, note that we start all the callbacks - * afresh, even those that have passed through a grace period - * and are therefore ready to invoke. The theory is that hotplug - * events are rare, and that if they are frequent enough to - * indefinitely delay callbacks, you have far worse things to - * be worrying about. + * We still hold the leaf rcu_node structure lock here, and + * irqs are still disabled. The reason for this subterfuge is + * because invoking rcu_report_unblock_qs_rnp() with ->onofflock + * held leads to deadlock. */ - rdp_me = rsp->rda[smp_processor_id()]; - if (rdp->nxtlist != NULL) { - *rdp_me->nxttail[RCU_NEXT_TAIL] = rdp->nxtlist; - rdp_me->nxttail[RCU_NEXT_TAIL] = rdp->nxttail[RCU_NEXT_TAIL]; - rdp->nxtlist = NULL; - for (i = 0; i < RCU_NEXT_SIZE; i++) - rdp->nxttail[i] = &rdp->nxtlist; - rdp_me->qlen += rdp->qlen; - rdp->qlen = 0; - } - local_irq_restore(flags); + spin_unlock(&rsp->onofflock); /* irqs remain disabled. */ + rnp = rdp->mynode; + if (need_report & RCU_OFL_TASKS_NORM_GP) + rcu_report_unblock_qs_rnp(rnp, flags); + else + spin_unlock_irqrestore(&rnp->lock, flags); + if (need_report & RCU_OFL_TASKS_EXP_GP) + rcu_report_exp_rnp(rsp, rnp); + + rcu_adopt_orphan_cbs(rsp); } /* @@ -860,12 +1000,21 @@ static void __rcu_offline_cpu(int cpu, struct rcu_state *rsp) */ static void rcu_offline_cpu(int cpu) { - __rcu_offline_cpu(cpu, &rcu_state); + __rcu_offline_cpu(cpu, &rcu_sched_state); __rcu_offline_cpu(cpu, &rcu_bh_state); + rcu_preempt_offline_cpu(cpu); } #else /* #ifdef CONFIG_HOTPLUG_CPU */ +static void rcu_send_cbs_to_orphanage(struct rcu_state *rsp) +{ +} + +static void rcu_adopt_orphan_cbs(struct rcu_state *rsp) +{ +} + static void rcu_offline_cpu(int cpu) { } @@ -876,7 +1025,7 @@ static void rcu_offline_cpu(int cpu) * Invoke any RCU callbacks that have made it to the end of their grace * period. Thottle as specified by rdp->blimit. */ -static void rcu_do_batch(struct rcu_data *rdp) +static void rcu_do_batch(struct rcu_state *rsp, struct rcu_data *rdp) { unsigned long flags; struct rcu_head *next, *list, **tail; @@ -929,6 +1078,13 @@ static void rcu_do_batch(struct rcu_data *rdp) if (rdp->blimit == LONG_MAX && rdp->qlen <= qlowmark) rdp->blimit = blimit; + /* Reset ->qlen_last_fqs_check trigger if enough CBs have drained. */ + if (rdp->qlen == 0 && rdp->qlen_last_fqs_check != 0) { + rdp->qlen_last_fqs_check = 0; + rdp->n_force_qs_snap = rsp->n_force_qs; + } else if (rdp->qlen < rdp->qlen_last_fqs_check - qhimark) + rdp->qlen_last_fqs_check = rdp->qlen; + local_irq_restore(flags); /* Re-raise the RCU softirq if there are callbacks remaining. */ @@ -947,6 +1103,8 @@ static void rcu_do_batch(struct rcu_data *rdp) */ void rcu_check_callbacks(int cpu, int user) { + if (!rcu_pending(cpu)) + return; /* if nothing for RCU to do. */ if (user || (idle_cpu(cpu) && rcu_scheduler_active && !in_softirq() && hardirq_count() <= (1 << HARDIRQ_SHIFT))) { @@ -955,17 +1113,16 @@ void rcu_check_callbacks(int cpu, int user) * Get here if this CPU took its interrupt from user * mode or from the idle loop, and if this is not a * nested interrupt. In this case, the CPU is in - * a quiescent state, so count it. + * a quiescent state, so note it. * * No memory barrier is required here because both - * rcu_qsctr_inc() and rcu_bh_qsctr_inc() reference - * only CPU-local variables that other CPUs neither - * access nor modify, at least not while the corresponding - * CPU is online. + * rcu_sched_qs() and rcu_bh_qs() reference only CPU-local + * variables that other CPUs neither access nor modify, + * at least not while the corresponding CPU is online. */ - rcu_qsctr_inc(cpu); - rcu_bh_qsctr_inc(cpu); + rcu_sched_qs(cpu); + rcu_bh_qs(cpu); } else if (!in_softirq()) { @@ -973,11 +1130,12 @@ void rcu_check_callbacks(int cpu, int user) * Get here if this CPU did not take its interrupt from * softirq, in other words, if it is not interrupting * a rcu_bh read-side critical section. This is an _bh - * critical section, so count it. + * critical section, so note it. */ - rcu_bh_qsctr_inc(cpu); + rcu_bh_qs(cpu); } + rcu_preempt_check_callbacks(cpu); raise_softirq(RCU_SOFTIRQ); } @@ -996,33 +1154,32 @@ static int rcu_process_dyntick(struct rcu_state *rsp, long lastcomp, int cpu; unsigned long flags; unsigned long mask; - struct rcu_node *rnp_cur = rsp->level[NUM_RCU_LVLS - 1]; - struct rcu_node *rnp_end = &rsp->node[NUM_RCU_NODES]; + struct rcu_node *rnp; - for (; rnp_cur < rnp_end; rnp_cur++) { + rcu_for_each_leaf_node(rsp, rnp) { mask = 0; - spin_lock_irqsave(&rnp_cur->lock, flags); - if (rsp->completed != lastcomp) { - spin_unlock_irqrestore(&rnp_cur->lock, flags); + spin_lock_irqsave(&rnp->lock, flags); + if (rnp->completed != lastcomp) { + spin_unlock_irqrestore(&rnp->lock, flags); return 1; } - if (rnp_cur->qsmask == 0) { - spin_unlock_irqrestore(&rnp_cur->lock, flags); + if (rnp->qsmask == 0) { + spin_unlock_irqrestore(&rnp->lock, flags); continue; } - cpu = rnp_cur->grplo; + cpu = rnp->grplo; bit = 1; - for (; cpu <= rnp_cur->grphi; cpu++, bit <<= 1) { - if ((rnp_cur->qsmask & bit) != 0 && f(rsp->rda[cpu])) + for (; cpu <= rnp->grphi; cpu++, bit <<= 1) { + if ((rnp->qsmask & bit) != 0 && f(rsp->rda[cpu])) mask |= bit; } - if (mask != 0 && rsp->completed == lastcomp) { + if (mask != 0 && rnp->completed == lastcomp) { - /* cpu_quiet_msk() releases rnp_cur->lock. */ - cpu_quiet_msk(mask, rsp, rnp_cur, flags); + /* rcu_report_qs_rnp() releases rnp->lock. */ + rcu_report_qs_rnp(mask, rsp, rnp, flags); continue; } - spin_unlock_irqrestore(&rnp_cur->lock, flags); + spin_unlock_irqrestore(&rnp->lock, flags); } return 0; } @@ -1035,37 +1192,35 @@ static void force_quiescent_state(struct rcu_state *rsp, int relaxed) { unsigned long flags; long lastcomp; - struct rcu_data *rdp = rsp->rda[smp_processor_id()]; struct rcu_node *rnp = rcu_get_root(rsp); u8 signaled; + u8 forcenow; - if (ACCESS_ONCE(rsp->completed) == ACCESS_ONCE(rsp->gpnum)) + if (!rcu_gp_in_progress(rsp)) return; /* No grace period in progress, nothing to force. */ if (!spin_trylock_irqsave(&rsp->fqslock, flags)) { rsp->n_force_qs_lh++; /* Inexact, can lose counts. Tough! */ return; /* Someone else is already on the job. */ } if (relaxed && - (long)(rsp->jiffies_force_qs - jiffies) >= 0 && - (rdp->n_rcu_pending_force_qs - rdp->n_rcu_pending) >= 0) + (long)(rsp->jiffies_force_qs - jiffies) >= 0) goto unlock_ret; /* no emergency and done recently. */ rsp->n_force_qs++; spin_lock(&rnp->lock); - lastcomp = rsp->completed; + lastcomp = rsp->gpnum - 1; signaled = rsp->signaled; rsp->jiffies_force_qs = jiffies + RCU_JIFFIES_TILL_FORCE_QS; - rdp->n_rcu_pending_force_qs = rdp->n_rcu_pending + - RCU_JIFFIES_TILL_FORCE_QS; - if (lastcomp == rsp->gpnum) { + if(!rcu_gp_in_progress(rsp)) { rsp->n_force_qs_ngp++; spin_unlock(&rnp->lock); goto unlock_ret; /* no GP in progress, time updated. */ } spin_unlock(&rnp->lock); switch (signaled) { + case RCU_GP_IDLE: case RCU_GP_INIT: - break; /* grace period still initializing, ignore. */ + break; /* grace period idle or initializing, ignore. */ case RCU_SAVE_DYNTICK: @@ -1076,20 +1231,29 @@ static void force_quiescent_state(struct rcu_state *rsp, int relaxed) if (rcu_process_dyntick(rsp, lastcomp, dyntick_save_progress_counter)) goto unlock_ret; + /* fall into next case. */ + + case RCU_SAVE_COMPLETED: /* Update state, record completion counter. */ + forcenow = 0; spin_lock(&rnp->lock); - if (lastcomp == rsp->completed) { + if (lastcomp + 1 == rsp->gpnum && + lastcomp == rsp->completed && + rsp->signaled == signaled) { rsp->signaled = RCU_FORCE_QS; - dyntick_record_completed(rsp, lastcomp); + rsp->completed_fqs = lastcomp; + forcenow = signaled == RCU_SAVE_COMPLETED; } spin_unlock(&rnp->lock); - break; + if (!forcenow) + break; + /* fall into next case. */ case RCU_FORCE_QS: /* Check dyntick-idle state, send IPI to laggarts. */ - if (rcu_process_dyntick(rsp, dyntick_recall_completed(rsp), + if (rcu_process_dyntick(rsp, rsp->completed_fqs, rcu_implicit_dynticks_qs)) goto unlock_ret; @@ -1120,12 +1284,13 @@ __rcu_process_callbacks(struct rcu_state *rsp, struct rcu_data *rdp) { unsigned long flags; + WARN_ON_ONCE(rdp->beenonline == 0); + /* * If an RCU GP has gone long enough, go check for dyntick * idle CPUs and, if needed, send resched IPIs. */ - if ((long)(ACCESS_ONCE(rsp->jiffies_force_qs) - jiffies) < 0 || - (rdp->n_rcu_pending_force_qs - rdp->n_rcu_pending) < 0) + if ((long)(ACCESS_ONCE(rsp->jiffies_force_qs) - jiffies) < 0) force_quiescent_state(rsp, 1); /* @@ -1144,7 +1309,7 @@ __rcu_process_callbacks(struct rcu_state *rsp, struct rcu_data *rdp) } /* If there are callbacks ready, invoke them. */ - rcu_do_batch(rdp); + rcu_do_batch(rsp, rdp); } /* @@ -1159,8 +1324,10 @@ static void rcu_process_callbacks(struct softirq_action *unused) */ smp_mb(); /* See above block comment. */ - __rcu_process_callbacks(&rcu_state, &__get_cpu_var(rcu_data)); + __rcu_process_callbacks(&rcu_sched_state, + &__get_cpu_var(rcu_sched_data)); __rcu_process_callbacks(&rcu_bh_state, &__get_cpu_var(rcu_bh_data)); + rcu_preempt_process_callbacks(); /* * Memory references from any later RCU read-side critical sections @@ -1198,7 +1365,7 @@ __call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu), rdp->nxttail[RCU_NEXT_TAIL] = &head->next; /* Start a new grace period if one not already started. */ - if (ACCESS_ONCE(rsp->completed) == ACCESS_ONCE(rsp->gpnum)) { + if (!rcu_gp_in_progress(rsp)) { unsigned long nestflag; struct rcu_node *rnp_root = rcu_get_root(rsp); @@ -1206,24 +1373,33 @@ __call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu), rcu_start_gp(rsp, nestflag); /* releases rnp_root->lock. */ } - /* Force the grace period if too many callbacks or too long waiting. */ - if (unlikely(++rdp->qlen > qhimark)) { + /* + * Force the grace period if too many callbacks or too long waiting. + * Enforce hysteresis, and don't invoke force_quiescent_state() + * if some other CPU has recently done so. Also, don't bother + * invoking force_quiescent_state() if the newly enqueued callback + * is the only one waiting for a grace period to complete. + */ + if (unlikely(++rdp->qlen > rdp->qlen_last_fqs_check + qhimark)) { rdp->blimit = LONG_MAX; - force_quiescent_state(rsp, 0); - } else if ((long)(ACCESS_ONCE(rsp->jiffies_force_qs) - jiffies) < 0 || - (rdp->n_rcu_pending_force_qs - rdp->n_rcu_pending) < 0) + if (rsp->n_force_qs == rdp->n_force_qs_snap && + *rdp->nxttail[RCU_DONE_TAIL] != head) + force_quiescent_state(rsp, 0); + rdp->n_force_qs_snap = rsp->n_force_qs; + rdp->qlen_last_fqs_check = rdp->qlen; + } else if ((long)(ACCESS_ONCE(rsp->jiffies_force_qs) - jiffies) < 0) force_quiescent_state(rsp, 1); local_irq_restore(flags); } /* - * Queue an RCU callback for invocation after a grace period. + * Queue an RCU-sched callback for invocation after a grace period. */ -void call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu)) +void call_rcu_sched(struct rcu_head *head, void (*func)(struct rcu_head *rcu)) { - __call_rcu(head, func, &rcu_state); + __call_rcu(head, func, &rcu_sched_state); } -EXPORT_SYMBOL_GPL(call_rcu); +EXPORT_SYMBOL_GPL(call_rcu_sched); /* * Queue an RCU for invocation after a quicker grace period. @@ -1234,6 +1410,68 @@ void call_rcu_bh(struct rcu_head *head, void (*func)(struct rcu_head *rcu)) } EXPORT_SYMBOL_GPL(call_rcu_bh); +/** + * synchronize_sched - wait until an rcu-sched grace period has elapsed. + * + * Control will return to the caller some time after a full rcu-sched + * grace period has elapsed, in other words after all currently executing + * rcu-sched read-side critical sections have completed. These read-side + * critical sections are delimited by rcu_read_lock_sched() and + * rcu_read_unlock_sched(), and may be nested. Note that preempt_disable(), + * local_irq_disable(), and so on may be used in place of + * rcu_read_lock_sched(). + * + * This means that all preempt_disable code sequences, including NMI and + * hardware-interrupt handlers, in progress on entry will have completed + * before this primitive returns. However, this does not guarantee that + * softirq handlers will have completed, since in some kernels, these + * handlers can run in process context, and can block. + * + * This primitive provides the guarantees made by the (now removed) + * synchronize_kernel() API. In contrast, synchronize_rcu() only + * guarantees that rcu_read_lock() sections will have completed. + * In "classic RCU", these two guarantees happen to be one and + * the same, but can differ in realtime RCU implementations. + */ +void synchronize_sched(void) +{ + struct rcu_synchronize rcu; + + if (rcu_blocking_is_gp()) + return; + + init_completion(&rcu.completion); + /* Will wake me after RCU finished. */ + call_rcu_sched(&rcu.head, wakeme_after_rcu); + /* Wait for it. */ + wait_for_completion(&rcu.completion); +} +EXPORT_SYMBOL_GPL(synchronize_sched); + +/** + * synchronize_rcu_bh - wait until an rcu_bh grace period has elapsed. + * + * Control will return to the caller some time after a full rcu_bh grace + * period has elapsed, in other words after all currently executing rcu_bh + * read-side critical sections have completed. RCU read-side critical + * sections are delimited by rcu_read_lock_bh() and rcu_read_unlock_bh(), + * and may be nested. + */ +void synchronize_rcu_bh(void) +{ + struct rcu_synchronize rcu; + + if (rcu_blocking_is_gp()) + return; + + init_completion(&rcu.completion); + /* Will wake me after RCU finished. */ + call_rcu_bh(&rcu.head, wakeme_after_rcu); + /* Wait for it. */ + wait_for_completion(&rcu.completion); +} +EXPORT_SYMBOL_GPL(synchronize_rcu_bh); + /* * Check to see if there is any immediate RCU-related work to be done * by the current CPU, for the specified type of RCU, returning 1 if so. @@ -1243,38 +1481,52 @@ EXPORT_SYMBOL_GPL(call_rcu_bh); */ static int __rcu_pending(struct rcu_state *rsp, struct rcu_data *rdp) { + struct rcu_node *rnp = rdp->mynode; + rdp->n_rcu_pending++; /* Check for CPU stalls, if enabled. */ check_cpu_stall(rsp, rdp); /* Is the RCU core waiting for a quiescent state from this CPU? */ - if (rdp->qs_pending) + if (rdp->qs_pending) { + rdp->n_rp_qs_pending++; return 1; + } /* Does this CPU have callbacks ready to invoke? */ - if (cpu_has_callbacks_ready_to_invoke(rdp)) + if (cpu_has_callbacks_ready_to_invoke(rdp)) { + rdp->n_rp_cb_ready++; return 1; + } /* Has RCU gone idle with this CPU needing another grace period? */ - if (cpu_needs_another_gp(rsp, rdp)) + if (cpu_needs_another_gp(rsp, rdp)) { + rdp->n_rp_cpu_needs_gp++; return 1; + } /* Has another RCU grace period completed? */ - if (ACCESS_ONCE(rsp->completed) != rdp->completed) /* outside of lock */ + if (ACCESS_ONCE(rnp->completed) != rdp->completed) { /* outside lock */ + rdp->n_rp_gp_completed++; return 1; + } /* Has a new RCU grace period started? */ - if (ACCESS_ONCE(rsp->gpnum) != rdp->gpnum) /* outside of lock */ + if (ACCESS_ONCE(rnp->gpnum) != rdp->gpnum) { /* outside lock */ + rdp->n_rp_gp_started++; return 1; + } /* Has an RCU GP gone long enough to send resched IPIs &c? */ - if (ACCESS_ONCE(rsp->completed) != ACCESS_ONCE(rsp->gpnum) && - ((long)(ACCESS_ONCE(rsp->jiffies_force_qs) - jiffies) < 0 || - (rdp->n_rcu_pending_force_qs - rdp->n_rcu_pending) < 0)) + if (rcu_gp_in_progress(rsp) && + ((long)(ACCESS_ONCE(rsp->jiffies_force_qs) - jiffies) < 0)) { + rdp->n_rp_need_fqs++; return 1; + } /* nothing to do */ + rdp->n_rp_need_nothing++; return 0; } @@ -1283,10 +1535,11 @@ static int __rcu_pending(struct rcu_state *rsp, struct rcu_data *rdp) * by the current CPU, returning 1 if so. This function is part of the * RCU implementation; it is -not- an exported member of the RCU API. */ -int rcu_pending(int cpu) +static int rcu_pending(int cpu) { - return __rcu_pending(&rcu_state, &per_cpu(rcu_data, cpu)) || - __rcu_pending(&rcu_bh_state, &per_cpu(rcu_bh_data, cpu)); + return __rcu_pending(&rcu_sched_state, &per_cpu(rcu_sched_data, cpu)) || + __rcu_pending(&rcu_bh_state, &per_cpu(rcu_bh_data, cpu)) || + rcu_preempt_pending(cpu); } /* @@ -1298,51 +1551,150 @@ int rcu_pending(int cpu) int rcu_needs_cpu(int cpu) { /* RCU callbacks either ready or pending? */ - return per_cpu(rcu_data, cpu).nxtlist || - per_cpu(rcu_bh_data, cpu).nxtlist; + return per_cpu(rcu_sched_data, cpu).nxtlist || + per_cpu(rcu_bh_data, cpu).nxtlist || + rcu_preempt_needs_cpu(cpu); } /* - * Initialize a CPU's per-CPU RCU data. We take this "scorched earth" - * approach so that we don't have to worry about how long the CPU has - * been gone, or whether it ever was online previously. We do trust the - * ->mynode field, as it is constant for a given struct rcu_data and - * initialized during early boot. - * - * Note that only one online or offline event can be happening at a given - * time. Note also that we can accept some slop in the rsp->completed - * access due to the fact that this CPU cannot possibly have any RCU - * callbacks in flight yet. + * This function is invoked towards the end of the scheduler's initialization + * process. Before this is called, the idle task might contain + * RCU read-side critical sections (during which time, this idle + * task is booting the system). After this function is called, the + * idle tasks are prohibited from containing RCU read-side critical + * sections. */ -static void __cpuinit -rcu_init_percpu_data(int cpu, struct rcu_state *rsp) +void rcu_scheduler_starting(void) +{ + WARN_ON(num_online_cpus() != 1); + WARN_ON(nr_context_switches() > 0); + rcu_scheduler_active = 1; +} + +static DEFINE_PER_CPU(struct rcu_head, rcu_barrier_head) = {NULL}; +static atomic_t rcu_barrier_cpu_count; +static DEFINE_MUTEX(rcu_barrier_mutex); +static struct completion rcu_barrier_completion; + +static void rcu_barrier_callback(struct rcu_head *notused) +{ + if (atomic_dec_and_test(&rcu_barrier_cpu_count)) + complete(&rcu_barrier_completion); +} + +/* + * Called with preemption disabled, and from cross-cpu IRQ context. + */ +static void rcu_barrier_func(void *type) +{ + int cpu = smp_processor_id(); + struct rcu_head *head = &per_cpu(rcu_barrier_head, cpu); + void (*call_rcu_func)(struct rcu_head *head, + void (*func)(struct rcu_head *head)); + + atomic_inc(&rcu_barrier_cpu_count); + call_rcu_func = type; + call_rcu_func(head, rcu_barrier_callback); +} + +/* + * Orchestrate the specified type of RCU barrier, waiting for all + * RCU callbacks of the specified type to complete. + */ +static void _rcu_barrier(struct rcu_state *rsp, + void (*call_rcu_func)(struct rcu_head *head, + void (*func)(struct rcu_head *head))) +{ + BUG_ON(in_interrupt()); + /* Take mutex to serialize concurrent rcu_barrier() requests. */ + mutex_lock(&rcu_barrier_mutex); + init_completion(&rcu_barrier_completion); + /* + * Initialize rcu_barrier_cpu_count to 1, then invoke + * rcu_barrier_func() on each CPU, so that each CPU also has + * incremented rcu_barrier_cpu_count. Only then is it safe to + * decrement rcu_barrier_cpu_count -- otherwise the first CPU + * might complete its grace period before all of the other CPUs + * did their increment, causing this function to return too + * early. + */ + atomic_set(&rcu_barrier_cpu_count, 1); + preempt_disable(); /* stop CPU_DYING from filling orphan_cbs_list */ + rcu_adopt_orphan_cbs(rsp); + on_each_cpu(rcu_barrier_func, (void *)call_rcu_func, 1); + preempt_enable(); /* CPU_DYING can again fill orphan_cbs_list */ + if (atomic_dec_and_test(&rcu_barrier_cpu_count)) + complete(&rcu_barrier_completion); + wait_for_completion(&rcu_barrier_completion); + mutex_unlock(&rcu_barrier_mutex); +} + +/** + * rcu_barrier_bh - Wait until all in-flight call_rcu_bh() callbacks complete. + */ +void rcu_barrier_bh(void) +{ + _rcu_barrier(&rcu_bh_state, call_rcu_bh); +} +EXPORT_SYMBOL_GPL(rcu_barrier_bh); + +/** + * rcu_barrier_sched - Wait for in-flight call_rcu_sched() callbacks. + */ +void rcu_barrier_sched(void) +{ + _rcu_barrier(&rcu_sched_state, call_rcu_sched); +} +EXPORT_SYMBOL_GPL(rcu_barrier_sched); + +/* + * Do boot-time initialization of a CPU's per-CPU RCU data. + */ +static void __init +rcu_boot_init_percpu_data(int cpu, struct rcu_state *rsp) { unsigned long flags; int i; - long lastcomp; - unsigned long mask; struct rcu_data *rdp = rsp->rda[cpu]; struct rcu_node *rnp = rcu_get_root(rsp); /* Set up local state, ensuring consistent view of global state. */ spin_lock_irqsave(&rnp->lock, flags); - lastcomp = rsp->completed; - rdp->completed = lastcomp; - rdp->gpnum = lastcomp; - rdp->passed_quiesc = 0; /* We could be racing with new GP, */ - rdp->qs_pending = 1; /* so set up to respond to current GP. */ - rdp->beenonline = 1; /* We have now been online. */ - rdp->passed_quiesc_completed = lastcomp - 1; rdp->grpmask = 1UL << (cpu - rdp->mynode->grplo); rdp->nxtlist = NULL; for (i = 0; i < RCU_NEXT_SIZE; i++) rdp->nxttail[i] = &rdp->nxtlist; rdp->qlen = 0; - rdp->blimit = blimit; #ifdef CONFIG_NO_HZ rdp->dynticks = &per_cpu(rcu_dynticks, cpu); #endif /* #ifdef CONFIG_NO_HZ */ rdp->cpu = cpu; + spin_unlock_irqrestore(&rnp->lock, flags); +} + +/* + * Initialize a CPU's per-CPU RCU data. Note that only one online or + * offline event can be happening at a given time. Note also that we + * can accept some slop in the rsp->completed access due to the fact + * that this CPU cannot possibly have any RCU callbacks in flight yet. + */ +static void __cpuinit +rcu_init_percpu_data(int cpu, struct rcu_state *rsp, int preemptable) +{ + unsigned long flags; + unsigned long mask; + struct rcu_data *rdp = rsp->rda[cpu]; + struct rcu_node *rnp = rcu_get_root(rsp); + + /* Set up local state, ensuring consistent view of global state. */ + spin_lock_irqsave(&rnp->lock, flags); + rdp->passed_quiesc = 0; /* We could be racing with new GP, */ + rdp->qs_pending = 1; /* so set up to respond to current GP. */ + rdp->beenonline = 1; /* We have now been online. */ + rdp->preemptable = preemptable; + rdp->qlen_last_fqs_check = 0; + rdp->n_force_qs_snap = rsp->n_force_qs; + rdp->blimit = blimit; spin_unlock(&rnp->lock); /* irqs remain disabled. */ /* @@ -1361,38 +1713,30 @@ rcu_init_percpu_data(int cpu, struct rcu_state *rsp) spin_lock(&rnp->lock); /* irqs already disabled. */ rnp->qsmaskinit |= mask; mask = rnp->grpmask; + if (rnp == rdp->mynode) { + rdp->gpnum = rnp->completed; /* if GP in progress... */ + rdp->completed = rnp->completed; + rdp->passed_quiesc_completed = rnp->completed - 1; + } spin_unlock(&rnp->lock); /* irqs already disabled. */ rnp = rnp->parent; } while (rnp != NULL && !(rnp->qsmaskinit & mask)); - spin_unlock(&rsp->onofflock); /* irqs remain disabled. */ - - /* - * A new grace period might start here. If so, we will be part of - * it, and its gpnum will be greater than ours, so we will - * participate. It is also possible for the gpnum to have been - * incremented before this function was called, and the bitmasks - * to not be filled out until now, in which case we will also - * participate due to our gpnum being behind. - */ - - /* Since it is coming online, the CPU is in a quiescent state. */ - cpu_quiet(cpu, rsp, rdp, lastcomp); - local_irq_restore(flags); + spin_unlock_irqrestore(&rsp->onofflock, flags); } static void __cpuinit rcu_online_cpu(int cpu) { - rcu_init_percpu_data(cpu, &rcu_state); - rcu_init_percpu_data(cpu, &rcu_bh_state); - open_softirq(RCU_SOFTIRQ, rcu_process_callbacks); + rcu_init_percpu_data(cpu, &rcu_sched_state, 0); + rcu_init_percpu_data(cpu, &rcu_bh_state, 0); + rcu_preempt_init_percpu_data(cpu); } /* - * Handle CPU online/offline notifcation events. + * Handle CPU online/offline notification events. */ static int __cpuinit rcu_cpu_notify(struct notifier_block *self, - unsigned long action, void *hcpu) + unsigned long action, void *hcpu) { long cpu = (long)hcpu; @@ -1401,6 +1745,22 @@ static int __cpuinit rcu_cpu_notify(struct notifier_block *self, case CPU_UP_PREPARE_FROZEN: rcu_online_cpu(cpu); break; + case CPU_DYING: + case CPU_DYING_FROZEN: + /* + * preempt_disable() in _rcu_barrier() prevents stop_machine(), + * so when "on_each_cpu(rcu_barrier_func, (void *)type, 1);" + * returns, all online cpus have queued rcu_barrier_func(). + * The dying CPU clears its cpu_online_mask bit and + * moves all of its RCU callbacks to ->orphan_cbs_list + * in the context of stop_machine(), so subsequent calls + * to _rcu_barrier() will adopt these callbacks and only + * then queue rcu_barrier_func() on all remaining CPUs. + */ + rcu_send_cbs_to_orphanage(&rcu_bh_state); + rcu_send_cbs_to_orphanage(&rcu_sched_state); + rcu_preempt_send_cbs_to_orphanage(); + break; case CPU_DEAD: case CPU_DEAD_FROZEN: case CPU_UP_CANCELED: @@ -1464,6 +1824,8 @@ static void __init rcu_init_one(struct rcu_state *rsp) rnp = rsp->level[i]; for (j = 0; j < rsp->levelcnt[i]; j++, rnp++) { spin_lock_init(&rnp->lock); + lockdep_set_class(&rnp->lock, &rcu_node_class[i]); + rnp->gpnum = 0; rnp->qsmask = 0; rnp->qsmaskinit = 0; rnp->grplo = j * cpustride; @@ -1481,16 +1843,26 @@ static void __init rcu_init_one(struct rcu_state *rsp) j / rsp->levelspread[i - 1]; } rnp->level = i; + INIT_LIST_HEAD(&rnp->blocked_tasks[0]); + INIT_LIST_HEAD(&rnp->blocked_tasks[1]); + INIT_LIST_HEAD(&rnp->blocked_tasks[2]); + INIT_LIST_HEAD(&rnp->blocked_tasks[3]); } } } /* - * Helper macro for __rcu_init(). To be used nowhere else! - * Assigns leaf node pointers into each CPU's rcu_data structure. + * Helper macro for __rcu_init() and __rcu_init_preempt(). To be used + * nowhere else! Assigns leaf node pointers into each CPU's rcu_data + * structure. */ -#define RCU_DATA_PTR_INIT(rsp, rcu_data) \ +#define RCU_INIT_FLAVOR(rsp, rcu_data) \ do { \ + int i; \ + int j; \ + struct rcu_node *rnp; \ + \ + rcu_init_one(rsp); \ rnp = (rsp)->level[NUM_RCU_LVLS - 1]; \ j = 0; \ for_each_possible_cpu(i) { \ @@ -1498,35 +1870,34 @@ do { \ j++; \ per_cpu(rcu_data, i).mynode = &rnp[j]; \ (rsp)->rda[i] = &per_cpu(rcu_data, i); \ + rcu_boot_init_percpu_data(i, rsp); \ } \ } while (0) -static struct notifier_block __cpuinitdata rcu_nb = { - .notifier_call = rcu_cpu_notify, -}; - -void __init __rcu_init(void) +void __init rcu_init(void) { - int i; /* All used by RCU_DATA_PTR_INIT(). */ - int j; - struct rcu_node *rnp; + int i; - printk(KERN_WARNING "Experimental hierarchical RCU implementation.\n"); + rcu_bootup_announce(); #ifdef CONFIG_RCU_CPU_STALL_DETECTOR printk(KERN_INFO "RCU-based detection of stalled CPUs is enabled.\n"); #endif /* #ifdef CONFIG_RCU_CPU_STALL_DETECTOR */ - rcu_init_one(&rcu_state); - RCU_DATA_PTR_INIT(&rcu_state, rcu_data); - rcu_init_one(&rcu_bh_state); - RCU_DATA_PTR_INIT(&rcu_bh_state, rcu_bh_data); +#if NUM_RCU_LVL_4 != 0 + printk(KERN_INFO "Experimental four-level hierarchy is enabled.\n"); +#endif /* #if NUM_RCU_LVL_4 != 0 */ + RCU_INIT_FLAVOR(&rcu_sched_state, rcu_sched_data); + RCU_INIT_FLAVOR(&rcu_bh_state, rcu_bh_data); + __rcu_init_preempt(); + open_softirq(RCU_SOFTIRQ, rcu_process_callbacks); + /* + * We don't need protection against CPU-hotplug here because + * this is called early in boot, before either interrupts + * or the scheduler are operational. + */ + cpu_notifier(rcu_cpu_notify, 0); for_each_online_cpu(i) - rcu_cpu_notify(&rcu_nb, CPU_UP_PREPARE, (void *)(long)i); - /* Register notifier for non-boot CPUs */ - register_cpu_notifier(&rcu_nb); - printk(KERN_WARNING "Experimental hierarchical RCU init done.\n"); + rcu_cpu_notify(NULL, CPU_UP_PREPARE, (void *)(long)i); } -module_param(blimit, int, 0); -module_param(qhimark, int, 0); -module_param(qlowmark, int, 0); +#include "rcutree_plugin.h" diff --git a/kernel/rcutree.h b/kernel/rcutree.h new file mode 100644 index 00000000000..d2a0046f63b --- /dev/null +++ b/kernel/rcutree.h @@ -0,0 +1,371 @@ +/* + * Read-Copy Update mechanism for mutual exclusion (tree-based version) + * Internal non-public definitions. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + * + * Copyright IBM Corporation, 2008 + * + * Author: Ingo Molnar <mingo@elte.hu> + * Paul E. McKenney <paulmck@linux.vnet.ibm.com> + */ + +#include <linux/cache.h> +#include <linux/spinlock.h> +#include <linux/threads.h> +#include <linux/cpumask.h> +#include <linux/seqlock.h> + +/* + * Define shape of hierarchy based on NR_CPUS and CONFIG_RCU_FANOUT. + * In theory, it should be possible to add more levels straightforwardly. + * In practice, this has not been tested, so there is probably some + * bug somewhere. + */ +#define MAX_RCU_LVLS 4 +#define RCU_FANOUT (CONFIG_RCU_FANOUT) +#define RCU_FANOUT_SQ (RCU_FANOUT * RCU_FANOUT) +#define RCU_FANOUT_CUBE (RCU_FANOUT_SQ * RCU_FANOUT) +#define RCU_FANOUT_FOURTH (RCU_FANOUT_CUBE * RCU_FANOUT) + +#if NR_CPUS <= RCU_FANOUT +# define NUM_RCU_LVLS 1 +# define NUM_RCU_LVL_0 1 +# define NUM_RCU_LVL_1 (NR_CPUS) +# define NUM_RCU_LVL_2 0 +# define NUM_RCU_LVL_3 0 +# define NUM_RCU_LVL_4 0 +#elif NR_CPUS <= RCU_FANOUT_SQ +# define NUM_RCU_LVLS 2 +# define NUM_RCU_LVL_0 1 +# define NUM_RCU_LVL_1 DIV_ROUND_UP(NR_CPUS, RCU_FANOUT) +# define NUM_RCU_LVL_2 (NR_CPUS) +# define NUM_RCU_LVL_3 0 +# define NUM_RCU_LVL_4 0 +#elif NR_CPUS <= RCU_FANOUT_CUBE +# define NUM_RCU_LVLS 3 +# define NUM_RCU_LVL_0 1 +# define NUM_RCU_LVL_1 DIV_ROUND_UP(NR_CPUS, RCU_FANOUT_SQ) +# define NUM_RCU_LVL_2 DIV_ROUND_UP(NR_CPUS, RCU_FANOUT) +# define NUM_RCU_LVL_3 NR_CPUS +# define NUM_RCU_LVL_4 0 +#elif NR_CPUS <= RCU_FANOUT_FOURTH +# define NUM_RCU_LVLS 4 +# define NUM_RCU_LVL_0 1 +# define NUM_RCU_LVL_1 DIV_ROUND_UP(NR_CPUS, RCU_FANOUT_CUBE) +# define NUM_RCU_LVL_2 DIV_ROUND_UP(NR_CPUS, RCU_FANOUT_SQ) +# define NUM_RCU_LVL_3 DIV_ROUND_UP(NR_CPUS, RCU_FANOUT) +# define NUM_RCU_LVL_4 NR_CPUS +#else +# error "CONFIG_RCU_FANOUT insufficient for NR_CPUS" +#endif /* #if (NR_CPUS) <= RCU_FANOUT */ + +#define RCU_SUM (NUM_RCU_LVL_0 + NUM_RCU_LVL_1 + NUM_RCU_LVL_2 + NUM_RCU_LVL_3 + NUM_RCU_LVL_4) +#define NUM_RCU_NODES (RCU_SUM - NR_CPUS) + +/* + * Dynticks per-CPU state. + */ +struct rcu_dynticks { + int dynticks_nesting; /* Track nesting level, sort of. */ + int dynticks; /* Even value for dynticks-idle, else odd. */ + int dynticks_nmi; /* Even value for either dynticks-idle or */ + /* not in nmi handler, else odd. So this */ + /* remains even for nmi from irq handler. */ +}; + +/* + * Definition for node within the RCU grace-period-detection hierarchy. + */ +struct rcu_node { + spinlock_t lock; /* Root rcu_node's lock protects some */ + /* rcu_state fields as well as following. */ + long gpnum; /* Current grace period for this node. */ + /* This will either be equal to or one */ + /* behind the root rcu_node's gpnum. */ + long completed; /* Last grace period completed for this node. */ + /* This will either be equal to or one */ + /* behind the root rcu_node's gpnum. */ + unsigned long qsmask; /* CPUs or groups that need to switch in */ + /* order for current grace period to proceed.*/ + /* In leaf rcu_node, each bit corresponds to */ + /* an rcu_data structure, otherwise, each */ + /* bit corresponds to a child rcu_node */ + /* structure. */ + unsigned long expmask; /* Groups that have ->blocked_tasks[] */ + /* elements that need to drain to allow the */ + /* current expedited grace period to */ + /* complete (only for TREE_PREEMPT_RCU). */ + unsigned long qsmaskinit; + /* Per-GP initial value for qsmask & expmask. */ + unsigned long grpmask; /* Mask to apply to parent qsmask. */ + /* Only one bit will be set in this mask. */ + int grplo; /* lowest-numbered CPU or group here. */ + int grphi; /* highest-numbered CPU or group here. */ + u8 grpnum; /* CPU/group number for next level up. */ + u8 level; /* root is at level 0. */ + struct rcu_node *parent; + struct list_head blocked_tasks[4]; + /* Tasks blocked in RCU read-side critsect. */ + /* Grace period number (->gpnum) x blocked */ + /* by tasks on the (x & 0x1) element of the */ + /* blocked_tasks[] array. */ +} ____cacheline_internodealigned_in_smp; + +/* + * Do a full breadth-first scan of the rcu_node structures for the + * specified rcu_state structure. + */ +#define rcu_for_each_node_breadth_first(rsp, rnp) \ + for ((rnp) = &(rsp)->node[0]; \ + (rnp) < &(rsp)->node[NUM_RCU_NODES]; (rnp)++) + +/* + * Do a breadth-first scan of the non-leaf rcu_node structures for the + * specified rcu_state structure. Note that if there is a singleton + * rcu_node tree with but one rcu_node structure, this loop is a no-op. + */ +#define rcu_for_each_nonleaf_node_breadth_first(rsp, rnp) \ + for ((rnp) = &(rsp)->node[0]; \ + (rnp) < (rsp)->level[NUM_RCU_LVLS - 1]; (rnp)++) + +/* + * Scan the leaves of the rcu_node hierarchy for the specified rcu_state + * structure. Note that if there is a singleton rcu_node tree with but + * one rcu_node structure, this loop -will- visit the rcu_node structure. + * It is still a leaf node, even if it is also the root node. + */ +#define rcu_for_each_leaf_node(rsp, rnp) \ + for ((rnp) = (rsp)->level[NUM_RCU_LVLS - 1]; \ + (rnp) < &(rsp)->node[NUM_RCU_NODES]; (rnp)++) + +/* Index values for nxttail array in struct rcu_data. */ +#define RCU_DONE_TAIL 0 /* Also RCU_WAIT head. */ +#define RCU_WAIT_TAIL 1 /* Also RCU_NEXT_READY head. */ +#define RCU_NEXT_READY_TAIL 2 /* Also RCU_NEXT head. */ +#define RCU_NEXT_TAIL 3 +#define RCU_NEXT_SIZE 4 + +/* Per-CPU data for read-copy update. */ +struct rcu_data { + /* 1) quiescent-state and grace-period handling : */ + long completed; /* Track rsp->completed gp number */ + /* in order to detect GP end. */ + long gpnum; /* Highest gp number that this CPU */ + /* is aware of having started. */ + long passed_quiesc_completed; + /* Value of completed at time of qs. */ + bool passed_quiesc; /* User-mode/idle loop etc. */ + bool qs_pending; /* Core waits for quiesc state. */ + bool beenonline; /* CPU online at least once. */ + bool preemptable; /* Preemptable RCU? */ + struct rcu_node *mynode; /* This CPU's leaf of hierarchy */ + unsigned long grpmask; /* Mask to apply to leaf qsmask. */ + + /* 2) batch handling */ + /* + * If nxtlist is not NULL, it is partitioned as follows. + * Any of the partitions might be empty, in which case the + * pointer to that partition will be equal to the pointer for + * the following partition. When the list is empty, all of + * the nxttail elements point to the ->nxtlist pointer itself, + * which in that case is NULL. + * + * [nxtlist, *nxttail[RCU_DONE_TAIL]): + * Entries that batch # <= ->completed + * The grace period for these entries has completed, and + * the other grace-period-completed entries may be moved + * here temporarily in rcu_process_callbacks(). + * [*nxttail[RCU_DONE_TAIL], *nxttail[RCU_WAIT_TAIL]): + * Entries that batch # <= ->completed - 1: waiting for current GP + * [*nxttail[RCU_WAIT_TAIL], *nxttail[RCU_NEXT_READY_TAIL]): + * Entries known to have arrived before current GP ended + * [*nxttail[RCU_NEXT_READY_TAIL], *nxttail[RCU_NEXT_TAIL]): + * Entries that might have arrived after current GP ended + * Note that the value of *nxttail[RCU_NEXT_TAIL] will + * always be NULL, as this is the end of the list. + */ + struct rcu_head *nxtlist; + struct rcu_head **nxttail[RCU_NEXT_SIZE]; + long qlen; /* # of queued callbacks */ + long qlen_last_fqs_check; + /* qlen at last check for QS forcing */ + unsigned long n_force_qs_snap; + /* did other CPU force QS recently? */ + long blimit; /* Upper limit on a processed batch */ + +#ifdef CONFIG_NO_HZ + /* 3) dynticks interface. */ + struct rcu_dynticks *dynticks; /* Shared per-CPU dynticks state. */ + int dynticks_snap; /* Per-GP tracking for dynticks. */ + int dynticks_nmi_snap; /* Per-GP tracking for dynticks_nmi. */ +#endif /* #ifdef CONFIG_NO_HZ */ + + /* 4) reasons this CPU needed to be kicked by force_quiescent_state */ +#ifdef CONFIG_NO_HZ + unsigned long dynticks_fqs; /* Kicked due to dynticks idle. */ +#endif /* #ifdef CONFIG_NO_HZ */ + unsigned long offline_fqs; /* Kicked due to being offline. */ + unsigned long resched_ipi; /* Sent a resched IPI. */ + + /* 5) __rcu_pending() statistics. */ + long n_rcu_pending; /* rcu_pending() calls since boot. */ + long n_rp_qs_pending; + long n_rp_cb_ready; + long n_rp_cpu_needs_gp; + long n_rp_gp_completed; + long n_rp_gp_started; + long n_rp_need_fqs; + long n_rp_need_nothing; + + int cpu; +}; + +/* Values for signaled field in struct rcu_state. */ +#define RCU_GP_IDLE 0 /* No grace period in progress. */ +#define RCU_GP_INIT 1 /* Grace period being initialized. */ +#define RCU_SAVE_DYNTICK 2 /* Need to scan dyntick state. */ +#define RCU_SAVE_COMPLETED 3 /* Need to save rsp->completed. */ +#define RCU_FORCE_QS 4 /* Need to force quiescent state. */ +#ifdef CONFIG_NO_HZ +#define RCU_SIGNAL_INIT RCU_SAVE_DYNTICK +#else /* #ifdef CONFIG_NO_HZ */ +#define RCU_SIGNAL_INIT RCU_SAVE_COMPLETED +#endif /* #else #ifdef CONFIG_NO_HZ */ + +#define RCU_JIFFIES_TILL_FORCE_QS 3 /* for rsp->jiffies_force_qs */ +#ifdef CONFIG_RCU_CPU_STALL_DETECTOR +#define RCU_SECONDS_TILL_STALL_CHECK (10 * HZ) /* for rsp->jiffies_stall */ +#define RCU_SECONDS_TILL_STALL_RECHECK (30 * HZ) /* for rsp->jiffies_stall */ +#define RCU_STALL_RAT_DELAY 2 /* Allow other CPUs time */ + /* to take at least one */ + /* scheduling clock irq */ + /* before ratting on them. */ + +#endif /* #ifdef CONFIG_RCU_CPU_STALL_DETECTOR */ + +/* + * RCU global state, including node hierarchy. This hierarchy is + * represented in "heap" form in a dense array. The root (first level) + * of the hierarchy is in ->node[0] (referenced by ->level[0]), the second + * level in ->node[1] through ->node[m] (->node[1] referenced by ->level[1]), + * and the third level in ->node[m+1] and following (->node[m+1] referenced + * by ->level[2]). The number of levels is determined by the number of + * CPUs and by CONFIG_RCU_FANOUT. Small systems will have a "hierarchy" + * consisting of a single rcu_node. + */ +struct rcu_state { + struct rcu_node node[NUM_RCU_NODES]; /* Hierarchy. */ + struct rcu_node *level[NUM_RCU_LVLS]; /* Hierarchy levels. */ + u32 levelcnt[MAX_RCU_LVLS + 1]; /* # nodes in each level. */ + u8 levelspread[NUM_RCU_LVLS]; /* kids/node in each level. */ + struct rcu_data *rda[NR_CPUS]; /* array of rdp pointers. */ + + /* The following fields are guarded by the root rcu_node's lock. */ + + u8 signaled ____cacheline_internodealigned_in_smp; + /* Force QS state. */ + long gpnum; /* Current gp number. */ + long completed; /* # of last completed gp. */ + + /* End of fields guarded by root rcu_node's lock. */ + + spinlock_t onofflock; /* exclude on/offline and */ + /* starting new GP. Also */ + /* protects the following */ + /* orphan_cbs fields. */ + struct rcu_head *orphan_cbs_list; /* list of rcu_head structs */ + /* orphaned by all CPUs in */ + /* a given leaf rcu_node */ + /* going offline. */ + struct rcu_head **orphan_cbs_tail; /* And tail pointer. */ + long orphan_qlen; /* Number of orphaned cbs. */ + spinlock_t fqslock; /* Only one task forcing */ + /* quiescent states. */ + long completed_fqs; /* Value of completed @ snap. */ + /* Protected by fqslock. */ + unsigned long jiffies_force_qs; /* Time at which to invoke */ + /* force_quiescent_state(). */ + unsigned long n_force_qs; /* Number of calls to */ + /* force_quiescent_state(). */ + unsigned long n_force_qs_lh; /* ~Number of calls leaving */ + /* due to lock unavailable. */ + unsigned long n_force_qs_ngp; /* Number of calls leaving */ + /* due to no GP active. */ +#ifdef CONFIG_RCU_CPU_STALL_DETECTOR + unsigned long gp_start; /* Time at which GP started, */ + /* but in jiffies. */ + unsigned long jiffies_stall; /* Time at which to check */ + /* for CPU stalls. */ +#endif /* #ifdef CONFIG_RCU_CPU_STALL_DETECTOR */ +}; + +/* Return values for rcu_preempt_offline_tasks(). */ + +#define RCU_OFL_TASKS_NORM_GP 0x1 /* Tasks blocking normal */ + /* GP were moved to root. */ +#define RCU_OFL_TASKS_EXP_GP 0x2 /* Tasks blocking expedited */ + /* GP were moved to root. */ + +#ifdef RCU_TREE_NONCORE + +/* + * RCU implementation internal declarations: + */ +extern struct rcu_state rcu_sched_state; +DECLARE_PER_CPU(struct rcu_data, rcu_sched_data); + +extern struct rcu_state rcu_bh_state; +DECLARE_PER_CPU(struct rcu_data, rcu_bh_data); + +#ifdef CONFIG_TREE_PREEMPT_RCU +extern struct rcu_state rcu_preempt_state; +DECLARE_PER_CPU(struct rcu_data, rcu_preempt_data); +#endif /* #ifdef CONFIG_TREE_PREEMPT_RCU */ + +#else /* #ifdef RCU_TREE_NONCORE */ + +/* Forward declarations for rcutree_plugin.h */ +static void rcu_bootup_announce(void); +long rcu_batches_completed(void); +static void rcu_preempt_note_context_switch(int cpu); +static int rcu_preempted_readers(struct rcu_node *rnp); +#ifdef CONFIG_HOTPLUG_CPU +static void rcu_report_unblock_qs_rnp(struct rcu_node *rnp, + unsigned long flags); +#endif /* #ifdef CONFIG_HOTPLUG_CPU */ +#ifdef CONFIG_RCU_CPU_STALL_DETECTOR +static void rcu_print_task_stall(struct rcu_node *rnp); +#endif /* #ifdef CONFIG_RCU_CPU_STALL_DETECTOR */ +static void rcu_preempt_check_blocked_tasks(struct rcu_node *rnp); +#ifdef CONFIG_HOTPLUG_CPU +static int rcu_preempt_offline_tasks(struct rcu_state *rsp, + struct rcu_node *rnp, + struct rcu_data *rdp); +static void rcu_preempt_offline_cpu(int cpu); +#endif /* #ifdef CONFIG_HOTPLUG_CPU */ +static void rcu_preempt_check_callbacks(int cpu); +static void rcu_preempt_process_callbacks(void); +void call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu)); +#if defined(CONFIG_HOTPLUG_CPU) || defined(CONFIG_TREE_PREEMPT_RCU) +static void rcu_report_exp_rnp(struct rcu_state *rsp, struct rcu_node *rnp); +#endif /* #if defined(CONFIG_HOTPLUG_CPU) || defined(CONFIG_TREE_PREEMPT_RCU) */ +static int rcu_preempt_pending(int cpu); +static int rcu_preempt_needs_cpu(int cpu); +static void __cpuinit rcu_preempt_init_percpu_data(int cpu); +static void rcu_preempt_send_cbs_to_orphanage(void); +static void __init __rcu_init_preempt(void); + +#endif /* #else #ifdef RCU_TREE_NONCORE */ diff --git a/kernel/rcutree_plugin.h b/kernel/rcutree_plugin.h new file mode 100644 index 00000000000..37fbccdf41d --- /dev/null +++ b/kernel/rcutree_plugin.h @@ -0,0 +1,886 @@ +/* + * Read-Copy Update mechanism for mutual exclusion (tree-based version) + * Internal non-public definitions that provide either classic + * or preemptable semantics. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + * + * Copyright Red Hat, 2009 + * Copyright IBM Corporation, 2009 + * + * Author: Ingo Molnar <mingo@elte.hu> + * Paul E. McKenney <paulmck@linux.vnet.ibm.com> + */ + +#include <linux/delay.h> + +#ifdef CONFIG_TREE_PREEMPT_RCU + +struct rcu_state rcu_preempt_state = RCU_STATE_INITIALIZER(rcu_preempt_state); +DEFINE_PER_CPU(struct rcu_data, rcu_preempt_data); + +static int rcu_preempted_readers_exp(struct rcu_node *rnp); + +/* + * Tell them what RCU they are running. + */ +static void __init rcu_bootup_announce(void) +{ + printk(KERN_INFO + "Experimental preemptable hierarchical RCU implementation.\n"); +} + +/* + * Return the number of RCU-preempt batches processed thus far + * for debug and statistics. + */ +long rcu_batches_completed_preempt(void) +{ + return rcu_preempt_state.completed; +} +EXPORT_SYMBOL_GPL(rcu_batches_completed_preempt); + +/* + * Return the number of RCU batches processed thus far for debug & stats. + */ +long rcu_batches_completed(void) +{ + return rcu_batches_completed_preempt(); +} +EXPORT_SYMBOL_GPL(rcu_batches_completed); + +/* + * Record a preemptable-RCU quiescent state for the specified CPU. Note + * that this just means that the task currently running on the CPU is + * not in a quiescent state. There might be any number of tasks blocked + * while in an RCU read-side critical section. + */ +static void rcu_preempt_qs(int cpu) +{ + struct rcu_data *rdp = &per_cpu(rcu_preempt_data, cpu); + rdp->passed_quiesc_completed = rdp->gpnum - 1; + barrier(); + rdp->passed_quiesc = 1; +} + +/* + * We have entered the scheduler, and the current task might soon be + * context-switched away from. If this task is in an RCU read-side + * critical section, we will no longer be able to rely on the CPU to + * record that fact, so we enqueue the task on the appropriate entry + * of the blocked_tasks[] array. The task will dequeue itself when + * it exits the outermost enclosing RCU read-side critical section. + * Therefore, the current grace period cannot be permitted to complete + * until the blocked_tasks[] entry indexed by the low-order bit of + * rnp->gpnum empties. + * + * Caller must disable preemption. + */ +static void rcu_preempt_note_context_switch(int cpu) +{ + struct task_struct *t = current; + unsigned long flags; + int phase; + struct rcu_data *rdp; + struct rcu_node *rnp; + + if (t->rcu_read_lock_nesting && + (t->rcu_read_unlock_special & RCU_READ_UNLOCK_BLOCKED) == 0) { + + /* Possibly blocking in an RCU read-side critical section. */ + rdp = rcu_preempt_state.rda[cpu]; + rnp = rdp->mynode; + spin_lock_irqsave(&rnp->lock, flags); + t->rcu_read_unlock_special |= RCU_READ_UNLOCK_BLOCKED; + t->rcu_blocked_node = rnp; + + /* + * If this CPU has already checked in, then this task + * will hold up the next grace period rather than the + * current grace period. Queue the task accordingly. + * If the task is queued for the current grace period + * (i.e., this CPU has not yet passed through a quiescent + * state for the current grace period), then as long + * as that task remains queued, the current grace period + * cannot end. + * + * But first, note that the current CPU must still be + * on line! + */ + WARN_ON_ONCE((rdp->grpmask & rnp->qsmaskinit) == 0); + WARN_ON_ONCE(!list_empty(&t->rcu_node_entry)); + phase = (rnp->gpnum + !(rnp->qsmask & rdp->grpmask)) & 0x1; + list_add(&t->rcu_node_entry, &rnp->blocked_tasks[phase]); + spin_unlock_irqrestore(&rnp->lock, flags); + } + + /* + * Either we were not in an RCU read-side critical section to + * begin with, or we have now recorded that critical section + * globally. Either way, we can now note a quiescent state + * for this CPU. Again, if we were in an RCU read-side critical + * section, and if that critical section was blocking the current + * grace period, then the fact that the task has been enqueued + * means that we continue to block the current grace period. + */ + rcu_preempt_qs(cpu); + local_irq_save(flags); + t->rcu_read_unlock_special &= ~RCU_READ_UNLOCK_NEED_QS; + local_irq_restore(flags); +} + +/* + * Tree-preemptable RCU implementation for rcu_read_lock(). + * Just increment ->rcu_read_lock_nesting, shared state will be updated + * if we block. + */ +void __rcu_read_lock(void) +{ + ACCESS_ONCE(current->rcu_read_lock_nesting)++; + barrier(); /* needed if we ever invoke rcu_read_lock in rcutree.c */ +} +EXPORT_SYMBOL_GPL(__rcu_read_lock); + +/* + * Check for preempted RCU readers blocking the current grace period + * for the specified rcu_node structure. If the caller needs a reliable + * answer, it must hold the rcu_node's ->lock. + */ +static int rcu_preempted_readers(struct rcu_node *rnp) +{ + int phase = rnp->gpnum & 0x1; + + return !list_empty(&rnp->blocked_tasks[phase]) || + !list_empty(&rnp->blocked_tasks[phase + 2]); +} + +/* + * Record a quiescent state for all tasks that were previously queued + * on the specified rcu_node structure and that were blocking the current + * RCU grace period. The caller must hold the specified rnp->lock with + * irqs disabled, and this lock is released upon return, but irqs remain + * disabled. + */ +static void rcu_report_unblock_qs_rnp(struct rcu_node *rnp, unsigned long flags) + __releases(rnp->lock) +{ + unsigned long mask; + struct rcu_node *rnp_p; + + if (rnp->qsmask != 0 || rcu_preempted_readers(rnp)) { + spin_unlock_irqrestore(&rnp->lock, flags); + return; /* Still need more quiescent states! */ + } + + rnp_p = rnp->parent; + if (rnp_p == NULL) { + /* + * Either there is only one rcu_node in the tree, + * or tasks were kicked up to root rcu_node due to + * CPUs going offline. + */ + rcu_report_qs_rsp(&rcu_preempt_state, flags); + return; + } + + /* Report up the rest of the hierarchy. */ + mask = rnp->grpmask; + spin_unlock(&rnp->lock); /* irqs remain disabled. */ + spin_lock(&rnp_p->lock); /* irqs already disabled. */ + rcu_report_qs_rnp(mask, &rcu_preempt_state, rnp_p, flags); +} + +/* + * Handle special cases during rcu_read_unlock(), such as needing to + * notify RCU core processing or task having blocked during the RCU + * read-side critical section. + */ +static void rcu_read_unlock_special(struct task_struct *t) +{ + int empty; + int empty_exp; + unsigned long flags; + struct rcu_node *rnp; + int special; + + /* NMI handlers cannot block and cannot safely manipulate state. */ + if (in_nmi()) + return; + + local_irq_save(flags); + + /* + * If RCU core is waiting for this CPU to exit critical section, + * let it know that we have done so. + */ + special = t->rcu_read_unlock_special; + if (special & RCU_READ_UNLOCK_NEED_QS) { + t->rcu_read_unlock_special &= ~RCU_READ_UNLOCK_NEED_QS; + rcu_preempt_qs(smp_processor_id()); + } + + /* Hardware IRQ handlers cannot block. */ + if (in_irq()) { + local_irq_restore(flags); + return; + } + + /* Clean up if blocked during RCU read-side critical section. */ + if (special & RCU_READ_UNLOCK_BLOCKED) { + t->rcu_read_unlock_special &= ~RCU_READ_UNLOCK_BLOCKED; + + /* + * Remove this task from the list it blocked on. The + * task can migrate while we acquire the lock, but at + * most one time. So at most two passes through loop. + */ + for (;;) { + rnp = t->rcu_blocked_node; + spin_lock(&rnp->lock); /* irqs already disabled. */ + if (rnp == t->rcu_blocked_node) + break; + spin_unlock(&rnp->lock); /* irqs remain disabled. */ + } + empty = !rcu_preempted_readers(rnp); + empty_exp = !rcu_preempted_readers_exp(rnp); + smp_mb(); /* ensure expedited fastpath sees end of RCU c-s. */ + list_del_init(&t->rcu_node_entry); + t->rcu_blocked_node = NULL; + + /* + * If this was the last task on the current list, and if + * we aren't waiting on any CPUs, report the quiescent state. + * Note that rcu_report_unblock_qs_rnp() releases rnp->lock. + */ + if (empty) + spin_unlock_irqrestore(&rnp->lock, flags); + else + rcu_report_unblock_qs_rnp(rnp, flags); + + /* + * If this was the last task on the expedited lists, + * then we need to report up the rcu_node hierarchy. + */ + if (!empty_exp && !rcu_preempted_readers_exp(rnp)) + rcu_report_exp_rnp(&rcu_preempt_state, rnp); + } else { + local_irq_restore(flags); + } +} + +/* + * Tree-preemptable RCU implementation for rcu_read_unlock(). + * Decrement ->rcu_read_lock_nesting. If the result is zero (outermost + * rcu_read_unlock()) and ->rcu_read_unlock_special is non-zero, then + * invoke rcu_read_unlock_special() to clean up after a context switch + * in an RCU read-side critical section and other special cases. + */ +void __rcu_read_unlock(void) +{ + struct task_struct *t = current; + + barrier(); /* needed if we ever invoke rcu_read_unlock in rcutree.c */ + if (--ACCESS_ONCE(t->rcu_read_lock_nesting) == 0 && + unlikely(ACCESS_ONCE(t->rcu_read_unlock_special))) + rcu_read_unlock_special(t); +} +EXPORT_SYMBOL_GPL(__rcu_read_unlock); + +#ifdef CONFIG_RCU_CPU_STALL_DETECTOR + +/* + * Scan the current list of tasks blocked within RCU read-side critical + * sections, printing out the tid of each. + */ +static void rcu_print_task_stall(struct rcu_node *rnp) +{ + unsigned long flags; + struct list_head *lp; + int phase; + struct task_struct *t; + + if (rcu_preempted_readers(rnp)) { + spin_lock_irqsave(&rnp->lock, flags); + phase = rnp->gpnum & 0x1; + lp = &rnp->blocked_tasks[phase]; + list_for_each_entry(t, lp, rcu_node_entry) + printk(" P%d", t->pid); + spin_unlock_irqrestore(&rnp->lock, flags); + } +} + +#endif /* #ifdef CONFIG_RCU_CPU_STALL_DETECTOR */ + +/* + * Check that the list of blocked tasks for the newly completed grace + * period is in fact empty. It is a serious bug to complete a grace + * period that still has RCU readers blocked! This function must be + * invoked -before- updating this rnp's ->gpnum, and the rnp's ->lock + * must be held by the caller. + */ +static void rcu_preempt_check_blocked_tasks(struct rcu_node *rnp) +{ + WARN_ON_ONCE(rcu_preempted_readers(rnp)); + WARN_ON_ONCE(rnp->qsmask); +} + +#ifdef CONFIG_HOTPLUG_CPU + +/* + * Handle tasklist migration for case in which all CPUs covered by the + * specified rcu_node have gone offline. Move them up to the root + * rcu_node. The reason for not just moving them to the immediate + * parent is to remove the need for rcu_read_unlock_special() to + * make more than two attempts to acquire the target rcu_node's lock. + * Returns true if there were tasks blocking the current RCU grace + * period. + * + * Returns 1 if there was previously a task blocking the current grace + * period on the specified rcu_node structure. + * + * The caller must hold rnp->lock with irqs disabled. + */ +static int rcu_preempt_offline_tasks(struct rcu_state *rsp, + struct rcu_node *rnp, + struct rcu_data *rdp) +{ + int i; + struct list_head *lp; + struct list_head *lp_root; + int retval = 0; + struct rcu_node *rnp_root = rcu_get_root(rsp); + struct task_struct *tp; + + if (rnp == rnp_root) { + WARN_ONCE(1, "Last CPU thought to be offlined?"); + return 0; /* Shouldn't happen: at least one CPU online. */ + } + WARN_ON_ONCE(rnp != rdp->mynode && + (!list_empty(&rnp->blocked_tasks[0]) || + !list_empty(&rnp->blocked_tasks[1]) || + !list_empty(&rnp->blocked_tasks[2]) || + !list_empty(&rnp->blocked_tasks[3]))); + + /* + * Move tasks up to root rcu_node. Rely on the fact that the + * root rcu_node can be at most one ahead of the rest of the + * rcu_nodes in terms of gp_num value. This fact allows us to + * move the blocked_tasks[] array directly, element by element. + */ + if (rcu_preempted_readers(rnp)) + retval |= RCU_OFL_TASKS_NORM_GP; + if (rcu_preempted_readers_exp(rnp)) + retval |= RCU_OFL_TASKS_EXP_GP; + for (i = 0; i < 4; i++) { + lp = &rnp->blocked_tasks[i]; + lp_root = &rnp_root->blocked_tasks[i]; + while (!list_empty(lp)) { + tp = list_entry(lp->next, typeof(*tp), rcu_node_entry); + spin_lock(&rnp_root->lock); /* irqs already disabled */ + list_del(&tp->rcu_node_entry); + tp->rcu_blocked_node = rnp_root; + list_add(&tp->rcu_node_entry, lp_root); + spin_unlock(&rnp_root->lock); /* irqs remain disabled */ + } + } + return retval; +} + +/* + * Do CPU-offline processing for preemptable RCU. + */ +static void rcu_preempt_offline_cpu(int cpu) +{ + __rcu_offline_cpu(cpu, &rcu_preempt_state); +} + +#endif /* #ifdef CONFIG_HOTPLUG_CPU */ + +/* + * Check for a quiescent state from the current CPU. When a task blocks, + * the task is recorded in the corresponding CPU's rcu_node structure, + * which is checked elsewhere. + * + * Caller must disable hard irqs. + */ +static void rcu_preempt_check_callbacks(int cpu) +{ + struct task_struct *t = current; + + if (t->rcu_read_lock_nesting == 0) { + t->rcu_read_unlock_special &= ~RCU_READ_UNLOCK_NEED_QS; + rcu_preempt_qs(cpu); + return; + } + if (per_cpu(rcu_preempt_data, cpu).qs_pending) + t->rcu_read_unlock_special |= RCU_READ_UNLOCK_NEED_QS; +} + +/* + * Process callbacks for preemptable RCU. + */ +static void rcu_preempt_process_callbacks(void) +{ + __rcu_process_callbacks(&rcu_preempt_state, + &__get_cpu_var(rcu_preempt_data)); +} + +/* + * Queue a preemptable-RCU callback for invocation after a grace period. + */ +void call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu)) +{ + __call_rcu(head, func, &rcu_preempt_state); +} +EXPORT_SYMBOL_GPL(call_rcu); + +/** + * synchronize_rcu - wait until a grace period has elapsed. + * + * Control will return to the caller some time after a full grace + * period has elapsed, in other words after all currently executing RCU + * read-side critical sections have completed. RCU read-side critical + * sections are delimited by rcu_read_lock() and rcu_read_unlock(), + * and may be nested. + */ +void synchronize_rcu(void) +{ + struct rcu_synchronize rcu; + + if (!rcu_scheduler_active) + return; + + init_completion(&rcu.completion); + /* Will wake me after RCU finished. */ + call_rcu(&rcu.head, wakeme_after_rcu); + /* Wait for it. */ + wait_for_completion(&rcu.completion); +} +EXPORT_SYMBOL_GPL(synchronize_rcu); + +static DECLARE_WAIT_QUEUE_HEAD(sync_rcu_preempt_exp_wq); +static long sync_rcu_preempt_exp_count; +static DEFINE_MUTEX(sync_rcu_preempt_exp_mutex); + +/* + * Return non-zero if there are any tasks in RCU read-side critical + * sections blocking the current preemptible-RCU expedited grace period. + * If there is no preemptible-RCU expedited grace period currently in + * progress, returns zero unconditionally. + */ +static int rcu_preempted_readers_exp(struct rcu_node *rnp) +{ + return !list_empty(&rnp->blocked_tasks[2]) || + !list_empty(&rnp->blocked_tasks[3]); +} + +/* + * return non-zero if there is no RCU expedited grace period in progress + * for the specified rcu_node structure, in other words, if all CPUs and + * tasks covered by the specified rcu_node structure have done their bit + * for the current expedited grace period. Works only for preemptible + * RCU -- other RCU implementation use other means. + * + * Caller must hold sync_rcu_preempt_exp_mutex. + */ +static int sync_rcu_preempt_exp_done(struct rcu_node *rnp) +{ + return !rcu_preempted_readers_exp(rnp) && + ACCESS_ONCE(rnp->expmask) == 0; +} + +/* + * Report the exit from RCU read-side critical section for the last task + * that queued itself during or before the current expedited preemptible-RCU + * grace period. This event is reported either to the rcu_node structure on + * which the task was queued or to one of that rcu_node structure's ancestors, + * recursively up the tree. (Calm down, calm down, we do the recursion + * iteratively!) + * + * Caller must hold sync_rcu_preempt_exp_mutex. + */ +static void rcu_report_exp_rnp(struct rcu_state *rsp, struct rcu_node *rnp) +{ + unsigned long flags; + unsigned long mask; + + spin_lock_irqsave(&rnp->lock, flags); + for (;;) { + if (!sync_rcu_preempt_exp_done(rnp)) + break; + if (rnp->parent == NULL) { + wake_up(&sync_rcu_preempt_exp_wq); + break; + } + mask = rnp->grpmask; + spin_unlock(&rnp->lock); /* irqs remain disabled */ + rnp = rnp->parent; + spin_lock(&rnp->lock); /* irqs already disabled */ + rnp->expmask &= ~mask; + } + spin_unlock_irqrestore(&rnp->lock, flags); +} + +/* + * Snapshot the tasks blocking the newly started preemptible-RCU expedited + * grace period for the specified rcu_node structure. If there are no such + * tasks, report it up the rcu_node hierarchy. + * + * Caller must hold sync_rcu_preempt_exp_mutex and rsp->onofflock. + */ +static void +sync_rcu_preempt_exp_init(struct rcu_state *rsp, struct rcu_node *rnp) +{ + int must_wait; + + spin_lock(&rnp->lock); /* irqs already disabled */ + list_splice_init(&rnp->blocked_tasks[0], &rnp->blocked_tasks[2]); + list_splice_init(&rnp->blocked_tasks[1], &rnp->blocked_tasks[3]); + must_wait = rcu_preempted_readers_exp(rnp); + spin_unlock(&rnp->lock); /* irqs remain disabled */ + if (!must_wait) + rcu_report_exp_rnp(rsp, rnp); +} + +/* + * Wait for an rcu-preempt grace period, but expedite it. The basic idea + * is to invoke synchronize_sched_expedited() to push all the tasks to + * the ->blocked_tasks[] lists, move all entries from the first set of + * ->blocked_tasks[] lists to the second set, and finally wait for this + * second set to drain. + */ +void synchronize_rcu_expedited(void) +{ + unsigned long flags; + struct rcu_node *rnp; + struct rcu_state *rsp = &rcu_preempt_state; + long snap; + int trycount = 0; + + smp_mb(); /* Caller's modifications seen first by other CPUs. */ + snap = ACCESS_ONCE(sync_rcu_preempt_exp_count) + 1; + smp_mb(); /* Above access cannot bleed into critical section. */ + + /* + * Acquire lock, falling back to synchronize_rcu() if too many + * lock-acquisition failures. Of course, if someone does the + * expedited grace period for us, just leave. + */ + while (!mutex_trylock(&sync_rcu_preempt_exp_mutex)) { + if (trycount++ < 10) + udelay(trycount * num_online_cpus()); + else { + synchronize_rcu(); + return; + } + if ((ACCESS_ONCE(sync_rcu_preempt_exp_count) - snap) > 0) + goto mb_ret; /* Others did our work for us. */ + } + if ((ACCESS_ONCE(sync_rcu_preempt_exp_count) - snap) > 0) + goto unlock_mb_ret; /* Others did our work for us. */ + + /* force all RCU readers onto blocked_tasks[]. */ + synchronize_sched_expedited(); + + spin_lock_irqsave(&rsp->onofflock, flags); + + /* Initialize ->expmask for all non-leaf rcu_node structures. */ + rcu_for_each_nonleaf_node_breadth_first(rsp, rnp) { + spin_lock(&rnp->lock); /* irqs already disabled. */ + rnp->expmask = rnp->qsmaskinit; + spin_unlock(&rnp->lock); /* irqs remain disabled. */ + } + + /* Snapshot current state of ->blocked_tasks[] lists. */ + rcu_for_each_leaf_node(rsp, rnp) + sync_rcu_preempt_exp_init(rsp, rnp); + if (NUM_RCU_NODES > 1) + sync_rcu_preempt_exp_init(rsp, rcu_get_root(rsp)); + + spin_unlock_irqrestore(&rsp->onofflock, flags); + + /* Wait for snapshotted ->blocked_tasks[] lists to drain. */ + rnp = rcu_get_root(rsp); + wait_event(sync_rcu_preempt_exp_wq, + sync_rcu_preempt_exp_done(rnp)); + + /* Clean up and exit. */ + smp_mb(); /* ensure expedited GP seen before counter increment. */ + ACCESS_ONCE(sync_rcu_preempt_exp_count)++; +unlock_mb_ret: + mutex_unlock(&sync_rcu_preempt_exp_mutex); +mb_ret: + smp_mb(); /* ensure subsequent action seen after grace period. */ +} +EXPORT_SYMBOL_GPL(synchronize_rcu_expedited); + +/* + * Check to see if there is any immediate preemptable-RCU-related work + * to be done. + */ +static int rcu_preempt_pending(int cpu) +{ + return __rcu_pending(&rcu_preempt_state, + &per_cpu(rcu_preempt_data, cpu)); +} + +/* + * Does preemptable RCU need the CPU to stay out of dynticks mode? + */ +static int rcu_preempt_needs_cpu(int cpu) +{ + return !!per_cpu(rcu_preempt_data, cpu).nxtlist; +} + +/** + * rcu_barrier - Wait until all in-flight call_rcu() callbacks complete. + */ +void rcu_barrier(void) +{ + _rcu_barrier(&rcu_preempt_state, call_rcu); +} +EXPORT_SYMBOL_GPL(rcu_barrier); + +/* + * Initialize preemptable RCU's per-CPU data. + */ +static void __cpuinit rcu_preempt_init_percpu_data(int cpu) +{ + rcu_init_percpu_data(cpu, &rcu_preempt_state, 1); +} + +/* + * Move preemptable RCU's callbacks to ->orphan_cbs_list. + */ +static void rcu_preempt_send_cbs_to_orphanage(void) +{ + rcu_send_cbs_to_orphanage(&rcu_preempt_state); +} + +/* + * Initialize preemptable RCU's state structures. + */ +static void __init __rcu_init_preempt(void) +{ + RCU_INIT_FLAVOR(&rcu_preempt_state, rcu_preempt_data); +} + +/* + * Check for a task exiting while in a preemptable-RCU read-side + * critical section, clean up if so. No need to issue warnings, + * as debug_check_no_locks_held() already does this if lockdep + * is enabled. + */ +void exit_rcu(void) +{ + struct task_struct *t = current; + + if (t->rcu_read_lock_nesting == 0) + return; + t->rcu_read_lock_nesting = 1; + rcu_read_unlock(); +} + +#else /* #ifdef CONFIG_TREE_PREEMPT_RCU */ + +/* + * Tell them what RCU they are running. + */ +static void __init rcu_bootup_announce(void) +{ + printk(KERN_INFO "Hierarchical RCU implementation.\n"); +} + +/* + * Return the number of RCU batches processed thus far for debug & stats. + */ +long rcu_batches_completed(void) +{ + return rcu_batches_completed_sched(); +} +EXPORT_SYMBOL_GPL(rcu_batches_completed); + +/* + * Because preemptable RCU does not exist, we never have to check for + * CPUs being in quiescent states. + */ +static void rcu_preempt_note_context_switch(int cpu) +{ +} + +/* + * Because preemptable RCU does not exist, there are never any preempted + * RCU readers. + */ +static int rcu_preempted_readers(struct rcu_node *rnp) +{ + return 0; +} + +#ifdef CONFIG_HOTPLUG_CPU + +/* Because preemptible RCU does not exist, no quieting of tasks. */ +static void rcu_report_unblock_qs_rnp(struct rcu_node *rnp, unsigned long flags) +{ + spin_unlock_irqrestore(&rnp->lock, flags); +} + +#endif /* #ifdef CONFIG_HOTPLUG_CPU */ + +#ifdef CONFIG_RCU_CPU_STALL_DETECTOR + +/* + * Because preemptable RCU does not exist, we never have to check for + * tasks blocked within RCU read-side critical sections. + */ +static void rcu_print_task_stall(struct rcu_node *rnp) +{ +} + +#endif /* #ifdef CONFIG_RCU_CPU_STALL_DETECTOR */ + +/* + * Because there is no preemptable RCU, there can be no readers blocked, + * so there is no need to check for blocked tasks. So check only for + * bogus qsmask values. + */ +static void rcu_preempt_check_blocked_tasks(struct rcu_node *rnp) +{ + WARN_ON_ONCE(rnp->qsmask); +} + +#ifdef CONFIG_HOTPLUG_CPU + +/* + * Because preemptable RCU does not exist, it never needs to migrate + * tasks that were blocked within RCU read-side critical sections, and + * such non-existent tasks cannot possibly have been blocking the current + * grace period. + */ +static int rcu_preempt_offline_tasks(struct rcu_state *rsp, + struct rcu_node *rnp, + struct rcu_data *rdp) +{ + return 0; +} + +/* + * Because preemptable RCU does not exist, it never needs CPU-offline + * processing. + */ +static void rcu_preempt_offline_cpu(int cpu) +{ +} + +#endif /* #ifdef CONFIG_HOTPLUG_CPU */ + +/* + * Because preemptable RCU does not exist, it never has any callbacks + * to check. + */ +static void rcu_preempt_check_callbacks(int cpu) +{ +} + +/* + * Because preemptable RCU does not exist, it never has any callbacks + * to process. + */ +static void rcu_preempt_process_callbacks(void) +{ +} + +/* + * In classic RCU, call_rcu() is just call_rcu_sched(). + */ +void call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu)) +{ + call_rcu_sched(head, func); +} +EXPORT_SYMBOL_GPL(call_rcu); + +/* + * Wait for an rcu-preempt grace period, but make it happen quickly. + * But because preemptable RCU does not exist, map to rcu-sched. + */ +void synchronize_rcu_expedited(void) +{ + synchronize_sched_expedited(); +} +EXPORT_SYMBOL_GPL(synchronize_rcu_expedited); + +#ifdef CONFIG_HOTPLUG_CPU + +/* + * Because preemptable RCU does not exist, there is never any need to + * report on tasks preempted in RCU read-side critical sections during + * expedited RCU grace periods. + */ +static void rcu_report_exp_rnp(struct rcu_state *rsp, struct rcu_node *rnp) +{ + return; +} + +#endif /* #ifdef CONFIG_HOTPLUG_CPU */ + +/* + * Because preemptable RCU does not exist, it never has any work to do. + */ +static int rcu_preempt_pending(int cpu) +{ + return 0; +} + +/* + * Because preemptable RCU does not exist, it never needs any CPU. + */ +static int rcu_preempt_needs_cpu(int cpu) +{ + return 0; +} + +/* + * Because preemptable RCU does not exist, rcu_barrier() is just + * another name for rcu_barrier_sched(). + */ +void rcu_barrier(void) +{ + rcu_barrier_sched(); +} +EXPORT_SYMBOL_GPL(rcu_barrier); + +/* + * Because preemptable RCU does not exist, there is no per-CPU + * data to initialize. + */ +static void __cpuinit rcu_preempt_init_percpu_data(int cpu) +{ +} + +/* + * Because there is no preemptable RCU, there are no callbacks to move. + */ +static void rcu_preempt_send_cbs_to_orphanage(void) +{ +} + +/* + * Because preemptable RCU does not exist, it need not be initialized. + */ +static void __init __rcu_init_preempt(void) +{ +} + +#endif /* #else #ifdef CONFIG_TREE_PREEMPT_RCU */ diff --git a/kernel/rcutree_trace.c b/kernel/rcutree_trace.c index d6db3e83782..9d2c88423b3 100644 --- a/kernel/rcutree_trace.c +++ b/kernel/rcutree_trace.c @@ -20,7 +20,7 @@ * Papers: http://www.rdrop.com/users/paulmck/RCU * * For detailed explanation of Read-Copy Update mechanism see - - * Documentation/RCU + * Documentation/RCU * */ #include <linux/types.h> @@ -43,18 +43,19 @@ #include <linux/debugfs.h> #include <linux/seq_file.h> +#define RCU_TREE_NONCORE +#include "rcutree.h" + static void print_one_rcu_data(struct seq_file *m, struct rcu_data *rdp) { if (!rdp->beenonline) return; - seq_printf(m, "%3d%cc=%ld g=%ld pq=%d pqc=%ld qp=%d rpfq=%ld rp=%x", + seq_printf(m, "%3d%cc=%ld g=%ld pq=%d pqc=%ld qp=%d", rdp->cpu, cpu_is_offline(rdp->cpu) ? '!' : ' ', rdp->completed, rdp->gpnum, rdp->passed_quiesc, rdp->passed_quiesc_completed, - rdp->qs_pending, - rdp->n_rcu_pending_force_qs - rdp->n_rcu_pending, - (int)(rdp->n_rcu_pending & 0xffff)); + rdp->qs_pending); #ifdef CONFIG_NO_HZ seq_printf(m, " dt=%d/%d dn=%d df=%lu", rdp->dynticks->dynticks, @@ -76,8 +77,12 @@ static void print_one_rcu_data(struct seq_file *m, struct rcu_data *rdp) static int show_rcudata(struct seq_file *m, void *unused) { - seq_puts(m, "rcu:\n"); - PRINT_RCU_DATA(rcu_data, print_one_rcu_data, m); +#ifdef CONFIG_TREE_PREEMPT_RCU + seq_puts(m, "rcu_preempt:\n"); + PRINT_RCU_DATA(rcu_preempt_data, print_one_rcu_data, m); +#endif /* #ifdef CONFIG_TREE_PREEMPT_RCU */ + seq_puts(m, "rcu_sched:\n"); + PRINT_RCU_DATA(rcu_sched_data, print_one_rcu_data, m); seq_puts(m, "rcu_bh:\n"); PRINT_RCU_DATA(rcu_bh_data, print_one_rcu_data, m); return 0; @@ -88,7 +93,7 @@ static int rcudata_open(struct inode *inode, struct file *file) return single_open(file, show_rcudata, NULL); } -static struct file_operations rcudata_fops = { +static const struct file_operations rcudata_fops = { .owner = THIS_MODULE, .open = rcudata_open, .read = seq_read, @@ -100,14 +105,12 @@ static void print_one_rcu_data_csv(struct seq_file *m, struct rcu_data *rdp) { if (!rdp->beenonline) return; - seq_printf(m, "%d,%s,%ld,%ld,%d,%ld,%d,%ld,%ld", + seq_printf(m, "%d,%s,%ld,%ld,%d,%ld,%d", rdp->cpu, - cpu_is_offline(rdp->cpu) ? "\"Y\"" : "\"N\"", + cpu_is_offline(rdp->cpu) ? "\"N\"" : "\"Y\"", rdp->completed, rdp->gpnum, rdp->passed_quiesc, rdp->passed_quiesc_completed, - rdp->qs_pending, - rdp->n_rcu_pending_force_qs - rdp->n_rcu_pending, - rdp->n_rcu_pending); + rdp->qs_pending); #ifdef CONFIG_NO_HZ seq_printf(m, ",%d,%d,%d,%lu", rdp->dynticks->dynticks, @@ -121,13 +124,17 @@ static void print_one_rcu_data_csv(struct seq_file *m, struct rcu_data *rdp) static int show_rcudata_csv(struct seq_file *m, void *unused) { - seq_puts(m, "\"CPU\",\"Online?\",\"c\",\"g\",\"pq\",\"pqc\",\"pq\",\"rpfq\",\"rp\","); + seq_puts(m, "\"CPU\",\"Online?\",\"c\",\"g\",\"pq\",\"pqc\",\"pq\","); #ifdef CONFIG_NO_HZ seq_puts(m, "\"dt\",\"dt nesting\",\"dn\",\"df\","); #endif /* #ifdef CONFIG_NO_HZ */ seq_puts(m, "\"of\",\"ri\",\"ql\",\"b\"\n"); - seq_puts(m, "\"rcu:\"\n"); - PRINT_RCU_DATA(rcu_data, print_one_rcu_data_csv, m); +#ifdef CONFIG_TREE_PREEMPT_RCU + seq_puts(m, "\"rcu_preempt:\"\n"); + PRINT_RCU_DATA(rcu_preempt_data, print_one_rcu_data_csv, m); +#endif /* #ifdef CONFIG_TREE_PREEMPT_RCU */ + seq_puts(m, "\"rcu_sched:\"\n"); + PRINT_RCU_DATA(rcu_sched_data, print_one_rcu_data_csv, m); seq_puts(m, "\"rcu_bh:\"\n"); PRINT_RCU_DATA(rcu_bh_data, print_one_rcu_data_csv, m); return 0; @@ -138,7 +145,7 @@ static int rcudata_csv_open(struct inode *inode, struct file *file) return single_open(file, show_rcudata_csv, NULL); } -static struct file_operations rcudata_csv_fops = { +static const struct file_operations rcudata_csv_fops = { .owner = THIS_MODULE, .open = rcudata_csv_open, .read = seq_read, @@ -148,24 +155,32 @@ static struct file_operations rcudata_csv_fops = { static void print_one_rcu_state(struct seq_file *m, struct rcu_state *rsp) { + long gpnum; int level = 0; + int phase; struct rcu_node *rnp; + gpnum = rsp->gpnum; seq_printf(m, "c=%ld g=%ld s=%d jfq=%ld j=%x " - "nfqs=%lu/nfqsng=%lu(%lu) fqlh=%lu\n", - rsp->completed, rsp->gpnum, rsp->signaled, + "nfqs=%lu/nfqsng=%lu(%lu) fqlh=%lu oqlen=%ld\n", + rsp->completed, gpnum, rsp->signaled, (long)(rsp->jiffies_force_qs - jiffies), (int)(jiffies & 0xffff), rsp->n_force_qs, rsp->n_force_qs_ngp, rsp->n_force_qs - rsp->n_force_qs_ngp, - rsp->n_force_qs_lh); + rsp->n_force_qs_lh, rsp->orphan_qlen); for (rnp = &rsp->node[0]; rnp - &rsp->node[0] < NUM_RCU_NODES; rnp++) { if (rnp->level != level) { seq_puts(m, "\n"); level = rnp->level; } - seq_printf(m, "%lx/%lx %d:%d ^%d ", + phase = gpnum & 0x1; + seq_printf(m, "%lx/%lx %c%c>%c%c %d:%d ^%d ", rnp->qsmask, rnp->qsmaskinit, + "T."[list_empty(&rnp->blocked_tasks[phase])], + "E."[list_empty(&rnp->blocked_tasks[phase + 2])], + "T."[list_empty(&rnp->blocked_tasks[!phase])], + "E."[list_empty(&rnp->blocked_tasks[!phase + 2])], rnp->grplo, rnp->grphi, rnp->grpnum); } seq_puts(m, "\n"); @@ -173,8 +188,12 @@ static void print_one_rcu_state(struct seq_file *m, struct rcu_state *rsp) static int show_rcuhier(struct seq_file *m, void *unused) { - seq_puts(m, "rcu:\n"); - print_one_rcu_state(m, &rcu_state); +#ifdef CONFIG_TREE_PREEMPT_RCU + seq_puts(m, "rcu_preempt:\n"); + print_one_rcu_state(m, &rcu_preempt_state); +#endif /* #ifdef CONFIG_TREE_PREEMPT_RCU */ + seq_puts(m, "rcu_sched:\n"); + print_one_rcu_state(m, &rcu_sched_state); seq_puts(m, "rcu_bh:\n"); print_one_rcu_state(m, &rcu_bh_state); return 0; @@ -185,7 +204,7 @@ static int rcuhier_open(struct inode *inode, struct file *file) return single_open(file, show_rcuhier, NULL); } -static struct file_operations rcuhier_fops = { +static const struct file_operations rcuhier_fops = { .owner = THIS_MODULE, .open = rcuhier_open, .read = seq_read, @@ -195,8 +214,12 @@ static struct file_operations rcuhier_fops = { static int show_rcugp(struct seq_file *m, void *unused) { - seq_printf(m, "rcu: completed=%ld gpnum=%ld\n", - rcu_state.completed, rcu_state.gpnum); +#ifdef CONFIG_TREE_PREEMPT_RCU + seq_printf(m, "rcu_preempt: completed=%ld gpnum=%ld\n", + rcu_preempt_state.completed, rcu_preempt_state.gpnum); +#endif /* #ifdef CONFIG_TREE_PREEMPT_RCU */ + seq_printf(m, "rcu_sched: completed=%ld gpnum=%ld\n", + rcu_sched_state.completed, rcu_sched_state.gpnum); seq_printf(m, "rcu_bh: completed=%ld gpnum=%ld\n", rcu_bh_state.completed, rcu_bh_state.gpnum); return 0; @@ -207,7 +230,7 @@ static int rcugp_open(struct inode *inode, struct file *file) return single_open(file, show_rcugp, NULL); } -static struct file_operations rcugp_fops = { +static const struct file_operations rcugp_fops = { .owner = THIS_MODULE, .open = rcugp_open, .read = seq_read, @@ -215,51 +238,102 @@ static struct file_operations rcugp_fops = { .release = single_release, }; -static struct dentry *rcudir, *datadir, *datadir_csv, *hierdir, *gpdir; +static void print_one_rcu_pending(struct seq_file *m, struct rcu_data *rdp) +{ + seq_printf(m, "%3d%cnp=%ld " + "qsp=%ld cbr=%ld cng=%ld gpc=%ld gps=%ld nf=%ld nn=%ld\n", + rdp->cpu, + cpu_is_offline(rdp->cpu) ? '!' : ' ', + rdp->n_rcu_pending, + rdp->n_rp_qs_pending, + rdp->n_rp_cb_ready, + rdp->n_rp_cpu_needs_gp, + rdp->n_rp_gp_completed, + rdp->n_rp_gp_started, + rdp->n_rp_need_fqs, + rdp->n_rp_need_nothing); +} + +static void print_rcu_pendings(struct seq_file *m, struct rcu_state *rsp) +{ + int cpu; + struct rcu_data *rdp; + + for_each_possible_cpu(cpu) { + rdp = rsp->rda[cpu]; + if (rdp->beenonline) + print_one_rcu_pending(m, rdp); + } +} + +static int show_rcu_pending(struct seq_file *m, void *unused) +{ +#ifdef CONFIG_TREE_PREEMPT_RCU + seq_puts(m, "rcu_preempt:\n"); + print_rcu_pendings(m, &rcu_preempt_state); +#endif /* #ifdef CONFIG_TREE_PREEMPT_RCU */ + seq_puts(m, "rcu_sched:\n"); + print_rcu_pendings(m, &rcu_sched_state); + seq_puts(m, "rcu_bh:\n"); + print_rcu_pendings(m, &rcu_bh_state); + return 0; +} + +static int rcu_pending_open(struct inode *inode, struct file *file) +{ + return single_open(file, show_rcu_pending, NULL); +} + +static const struct file_operations rcu_pending_fops = { + .owner = THIS_MODULE, + .open = rcu_pending_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; + +static struct dentry *rcudir; + static int __init rcuclassic_trace_init(void) { + struct dentry *retval; + rcudir = debugfs_create_dir("rcu", NULL); if (!rcudir) - goto out; + goto free_out; - datadir = debugfs_create_file("rcudata", 0444, rcudir, + retval = debugfs_create_file("rcudata", 0444, rcudir, NULL, &rcudata_fops); - if (!datadir) + if (!retval) goto free_out; - datadir_csv = debugfs_create_file("rcudata.csv", 0444, rcudir, + retval = debugfs_create_file("rcudata.csv", 0444, rcudir, NULL, &rcudata_csv_fops); - if (!datadir_csv) + if (!retval) goto free_out; - gpdir = debugfs_create_file("rcugp", 0444, rcudir, NULL, &rcugp_fops); - if (!gpdir) + retval = debugfs_create_file("rcugp", 0444, rcudir, NULL, &rcugp_fops); + if (!retval) goto free_out; - hierdir = debugfs_create_file("rcuhier", 0444, rcudir, + retval = debugfs_create_file("rcuhier", 0444, rcudir, NULL, &rcuhier_fops); - if (!hierdir) + if (!retval) + goto free_out; + + retval = debugfs_create_file("rcu_pending", 0444, rcudir, + NULL, &rcu_pending_fops); + if (!retval) goto free_out; return 0; free_out: - if (datadir) - debugfs_remove(datadir); - if (datadir_csv) - debugfs_remove(datadir_csv); - if (gpdir) - debugfs_remove(gpdir); - debugfs_remove(rcudir); -out: + debugfs_remove_recursive(rcudir); return 1; } static void __exit rcuclassic_trace_cleanup(void) { - debugfs_remove(datadir); - debugfs_remove(datadir_csv); - debugfs_remove(gpdir); - debugfs_remove(hierdir); - debugfs_remove(rcudir); + debugfs_remove_recursive(rcudir); } diff --git a/kernel/relay.c b/kernel/relay.c index 8f2179c8056..760c26209a3 100644 --- a/kernel/relay.c +++ b/kernel/relay.c @@ -60,7 +60,7 @@ static int relay_buf_fault(struct vm_area_struct *vma, struct vm_fault *vmf) /* * vm_ops for relay file mappings. */ -static struct vm_operations_struct relay_file_mmap_ops = { +static const struct vm_operations_struct relay_file_mmap_ops = { .fault = relay_buf_fault, .close = relay_file_mmap_close, }; @@ -677,9 +677,7 @@ int relay_late_setup_files(struct rchan *chan, */ for_each_online_cpu(i) { if (unlikely(!chan->buf[i])) { - printk(KERN_ERR "relay_late_setup_files: CPU %u " - "has no buffer, it must have!\n", i); - BUG(); + WARN_ONCE(1, KERN_ERR "CPU has no buffer!\n"); err = -EINVAL; break; } @@ -797,13 +795,15 @@ void relay_subbufs_consumed(struct rchan *chan, if (!chan) return; - if (cpu >= NR_CPUS || !chan->buf[cpu]) + if (cpu >= NR_CPUS || !chan->buf[cpu] || + subbufs_consumed > chan->n_subbufs) return; buf = chan->buf[cpu]; - buf->subbufs_consumed += subbufs_consumed; - if (buf->subbufs_consumed > buf->subbufs_produced) + if (subbufs_consumed > buf->subbufs_produced - buf->subbufs_consumed) buf->subbufs_consumed = buf->subbufs_produced; + else + buf->subbufs_consumed += subbufs_consumed; } EXPORT_SYMBOL_GPL(relay_subbufs_consumed); diff --git a/kernel/res_counter.c b/kernel/res_counter.c index bf8e7534c80..bcdabf37c40 100644 --- a/kernel/res_counter.c +++ b/kernel/res_counter.c @@ -18,7 +18,8 @@ void res_counter_init(struct res_counter *counter, struct res_counter *parent) { spin_lock_init(&counter->lock); - counter->limit = (unsigned long long)LLONG_MAX; + counter->limit = RESOURCE_MAX; + counter->soft_limit = RESOURCE_MAX; counter->parent = parent; } @@ -101,6 +102,8 @@ res_counter_member(struct res_counter *counter, int member) return &counter->limit; case RES_FAILCNT: return &counter->failcnt; + case RES_SOFT_LIMIT: + return &counter->soft_limit; }; BUG(); @@ -133,6 +136,16 @@ int res_counter_memparse_write_strategy(const char *buf, unsigned long long *res) { char *end; + + /* return RESOURCE_MAX(unlimited) if "-1" is specified */ + if (*buf == '-') { + *res = simple_strtoull(buf + 1, &end, 10); + if (*res != 1 || *end != '\0') + return -EINVAL; + *res = RESOURCE_MAX; + return 0; + } + /* FIXME - make memparse() take const char* args */ *res = memparse((char *)buf, &end); if (*end != '\0') diff --git a/kernel/resource.c b/kernel/resource.c index fd5d7d574bb..fb11a58b959 100644 --- a/kernel/resource.c +++ b/kernel/resource.c @@ -223,13 +223,13 @@ int release_resource(struct resource *old) EXPORT_SYMBOL(release_resource); -#if defined(CONFIG_MEMORY_HOTPLUG) && !defined(CONFIG_ARCH_HAS_WALK_MEMORY) +#if !defined(CONFIG_ARCH_HAS_WALK_MEMORY) /* * Finds the lowest memory reosurce exists within [res->start.res->end) - * the caller must specify res->start, res->end, res->flags. + * the caller must specify res->start, res->end, res->flags and "name". * If found, returns 0, res is overwritten, if not found, returns -1. */ -static int find_next_system_ram(struct resource *res) +static int find_next_system_ram(struct resource *res, char *name) { resource_size_t start, end; struct resource *p; @@ -245,6 +245,8 @@ static int find_next_system_ram(struct resource *res) /* system ram is just marked as IORESOURCE_MEM */ if (p->flags != res->flags) continue; + if (name && strcmp(p->name, name)) + continue; if (p->start > end) { p = NULL; break; @@ -262,19 +264,26 @@ static int find_next_system_ram(struct resource *res) res->end = p->end; return 0; } -int -walk_memory_resource(unsigned long start_pfn, unsigned long nr_pages, void *arg, - int (*func)(unsigned long, unsigned long, void *)) + +/* + * This function calls callback against all memory range of "System RAM" + * which are marked as IORESOURCE_MEM and IORESOUCE_BUSY. + * Now, this function is only for "System RAM". + */ +int walk_system_ram_range(unsigned long start_pfn, unsigned long nr_pages, + void *arg, int (*func)(unsigned long, unsigned long, void *)) { struct resource res; unsigned long pfn, len; u64 orig_end; int ret = -1; + res.start = (u64) start_pfn << PAGE_SHIFT; res.end = ((u64)(start_pfn + nr_pages) << PAGE_SHIFT) - 1; res.flags = IORESOURCE_MEM | IORESOURCE_BUSY; orig_end = res.end; - while ((res.start < res.end) && (find_next_system_ram(&res) >= 0)) { + while ((res.start < res.end) && + (find_next_system_ram(&res, "System RAM") >= 0)) { pfn = (unsigned long)(res.start >> PAGE_SHIFT); len = (unsigned long)((res.end + 1 - res.start) >> PAGE_SHIFT); ret = (*func)(pfn, len, arg); @@ -533,43 +542,21 @@ static void __init __reserve_region_with_split(struct resource *root, res->end = end; res->flags = IORESOURCE_BUSY; - for (;;) { - conflict = __request_resource(parent, res); - if (!conflict) - break; - if (conflict != parent) { - parent = conflict; - if (!(conflict->flags & IORESOURCE_BUSY)) - continue; - } - - /* Uhhuh, that didn't work out.. */ - kfree(res); - res = NULL; - break; - } - - if (!res) { - /* failed, split and try again */ - - /* conflict covered whole area */ - if (conflict->start <= start && conflict->end >= end) - return; + conflict = __request_resource(parent, res); + if (!conflict) + return; - if (conflict->start > start) - __reserve_region_with_split(root, start, conflict->start-1, name); - if (!(conflict->flags & IORESOURCE_BUSY)) { - resource_size_t common_start, common_end; + /* failed, split and try again */ + kfree(res); - common_start = max(conflict->start, start); - common_end = min(conflict->end, end); - if (common_start < common_end) - __reserve_region_with_split(root, common_start, common_end, name); - } - if (conflict->end < end) - __reserve_region_with_split(root, conflict->end+1, end, name); - } + /* conflict covered whole area */ + if (conflict->start <= start && conflict->end >= end) + return; + if (conflict->start > start) + __reserve_region_with_split(root, start, conflict->start-1, name); + if (conflict->end < end) + __reserve_region_with_split(root, conflict->end+1, end, name); } void __init reserve_region_with_split(struct resource *root, @@ -809,7 +796,7 @@ static int __init reserve_setup(char *str) static struct resource reserve[MAXRESERVE]; for (;;) { - int io_start, io_num; + unsigned int io_start, io_num; int x = reserved; if (get_option (&str, &io_start) != 2) diff --git a/kernel/rtmutex.c b/kernel/rtmutex.c index 69d9cb921ff..29bd4baf9e7 100644 --- a/kernel/rtmutex.c +++ b/kernel/rtmutex.c @@ -300,7 +300,8 @@ static int rt_mutex_adjust_prio_chain(struct task_struct *task, * assigned pending owner [which might not have taken the * lock yet]: */ -static inline int try_to_steal_lock(struct rt_mutex *lock) +static inline int try_to_steal_lock(struct rt_mutex *lock, + struct task_struct *task) { struct task_struct *pendowner = rt_mutex_owner(lock); struct rt_mutex_waiter *next; @@ -309,11 +310,11 @@ static inline int try_to_steal_lock(struct rt_mutex *lock) if (!rt_mutex_owner_pending(lock)) return 0; - if (pendowner == current) + if (pendowner == task) return 1; spin_lock_irqsave(&pendowner->pi_lock, flags); - if (current->prio >= pendowner->prio) { + if (task->prio >= pendowner->prio) { spin_unlock_irqrestore(&pendowner->pi_lock, flags); return 0; } @@ -338,21 +339,21 @@ static inline int try_to_steal_lock(struct rt_mutex *lock) * We are going to steal the lock and a waiter was * enqueued on the pending owners pi_waiters queue. So * we have to enqueue this waiter into - * current->pi_waiters list. This covers the case, - * where current is boosted because it holds another + * task->pi_waiters list. This covers the case, + * where task is boosted because it holds another * lock and gets unboosted because the booster is * interrupted, so we would delay a waiter with higher - * priority as current->normal_prio. + * priority as task->normal_prio. * * Note: in the rare case of a SCHED_OTHER task changing * its priority and thus stealing the lock, next->task - * might be current: + * might be task: */ - if (likely(next->task != current)) { - spin_lock_irqsave(¤t->pi_lock, flags); - plist_add(&next->pi_list_entry, ¤t->pi_waiters); - __rt_mutex_adjust_prio(current); - spin_unlock_irqrestore(¤t->pi_lock, flags); + if (likely(next->task != task)) { + spin_lock_irqsave(&task->pi_lock, flags); + plist_add(&next->pi_list_entry, &task->pi_waiters); + __rt_mutex_adjust_prio(task); + spin_unlock_irqrestore(&task->pi_lock, flags); } return 1; } @@ -389,7 +390,7 @@ static int try_to_take_rt_mutex(struct rt_mutex *lock) */ mark_rt_mutex_waiters(lock); - if (rt_mutex_owner(lock) && !try_to_steal_lock(lock)) + if (rt_mutex_owner(lock) && !try_to_steal_lock(lock, current)) return 0; /* We got the lock. */ @@ -411,6 +412,7 @@ static int try_to_take_rt_mutex(struct rt_mutex *lock) */ static int task_blocks_on_rt_mutex(struct rt_mutex *lock, struct rt_mutex_waiter *waiter, + struct task_struct *task, int detect_deadlock) { struct task_struct *owner = rt_mutex_owner(lock); @@ -418,21 +420,21 @@ static int task_blocks_on_rt_mutex(struct rt_mutex *lock, unsigned long flags; int chain_walk = 0, res; - spin_lock_irqsave(¤t->pi_lock, flags); - __rt_mutex_adjust_prio(current); - waiter->task = current; + spin_lock_irqsave(&task->pi_lock, flags); + __rt_mutex_adjust_prio(task); + waiter->task = task; waiter->lock = lock; - plist_node_init(&waiter->list_entry, current->prio); - plist_node_init(&waiter->pi_list_entry, current->prio); + plist_node_init(&waiter->list_entry, task->prio); + plist_node_init(&waiter->pi_list_entry, task->prio); /* Get the top priority waiter on the lock */ if (rt_mutex_has_waiters(lock)) top_waiter = rt_mutex_top_waiter(lock); plist_add(&waiter->list_entry, &lock->wait_list); - current->pi_blocked_on = waiter; + task->pi_blocked_on = waiter; - spin_unlock_irqrestore(¤t->pi_lock, flags); + spin_unlock_irqrestore(&task->pi_lock, flags); if (waiter == rt_mutex_top_waiter(lock)) { spin_lock_irqsave(&owner->pi_lock, flags); @@ -460,7 +462,7 @@ static int task_blocks_on_rt_mutex(struct rt_mutex *lock, spin_unlock(&lock->wait_lock); res = rt_mutex_adjust_prio_chain(owner, detect_deadlock, lock, waiter, - current); + task); spin_lock(&lock->wait_lock); @@ -605,37 +607,25 @@ void rt_mutex_adjust_pi(struct task_struct *task) rt_mutex_adjust_prio_chain(task, 0, NULL, NULL, task); } -/* - * Slow path lock function: +/** + * __rt_mutex_slowlock() - Perform the wait-wake-try-to-take loop + * @lock: the rt_mutex to take + * @state: the state the task should block in (TASK_INTERRUPTIBLE + * or TASK_UNINTERRUPTIBLE) + * @timeout: the pre-initialized and started timer, or NULL for none + * @waiter: the pre-initialized rt_mutex_waiter + * @detect_deadlock: passed to task_blocks_on_rt_mutex + * + * lock->wait_lock must be held by the caller. */ static int __sched -rt_mutex_slowlock(struct rt_mutex *lock, int state, - struct hrtimer_sleeper *timeout, - int detect_deadlock) +__rt_mutex_slowlock(struct rt_mutex *lock, int state, + struct hrtimer_sleeper *timeout, + struct rt_mutex_waiter *waiter, + int detect_deadlock) { - struct rt_mutex_waiter waiter; int ret = 0; - debug_rt_mutex_init_waiter(&waiter); - waiter.task = NULL; - - spin_lock(&lock->wait_lock); - - /* Try to acquire the lock again: */ - if (try_to_take_rt_mutex(lock)) { - spin_unlock(&lock->wait_lock); - return 0; - } - - set_current_state(state); - - /* Setup the timer, when timeout != NULL */ - if (unlikely(timeout)) { - hrtimer_start_expires(&timeout->timer, HRTIMER_MODE_ABS); - if (!hrtimer_active(&timeout->timer)) - timeout->task = NULL; - } - for (;;) { /* Try to acquire the lock: */ if (try_to_take_rt_mutex(lock)) @@ -656,19 +646,19 @@ rt_mutex_slowlock(struct rt_mutex *lock, int state, } /* - * waiter.task is NULL the first time we come here and + * waiter->task is NULL the first time we come here and * when we have been woken up by the previous owner * but the lock got stolen by a higher prio task. */ - if (!waiter.task) { - ret = task_blocks_on_rt_mutex(lock, &waiter, + if (!waiter->task) { + ret = task_blocks_on_rt_mutex(lock, waiter, current, detect_deadlock); /* * If we got woken up by the owner then start loop * all over without going into schedule to try * to get the lock now: */ - if (unlikely(!waiter.task)) { + if (unlikely(!waiter->task)) { /* * Reset the return value. We might * have returned with -EDEADLK and the @@ -684,15 +674,52 @@ rt_mutex_slowlock(struct rt_mutex *lock, int state, spin_unlock(&lock->wait_lock); - debug_rt_mutex_print_deadlock(&waiter); + debug_rt_mutex_print_deadlock(waiter); - if (waiter.task) + if (waiter->task) schedule_rt_mutex(lock); spin_lock(&lock->wait_lock); set_current_state(state); } + return ret; +} + +/* + * Slow path lock function: + */ +static int __sched +rt_mutex_slowlock(struct rt_mutex *lock, int state, + struct hrtimer_sleeper *timeout, + int detect_deadlock) +{ + struct rt_mutex_waiter waiter; + int ret = 0; + + debug_rt_mutex_init_waiter(&waiter); + waiter.task = NULL; + + spin_lock(&lock->wait_lock); + + /* Try to acquire the lock again: */ + if (try_to_take_rt_mutex(lock)) { + spin_unlock(&lock->wait_lock); + return 0; + } + + set_current_state(state); + + /* Setup the timer, when timeout != NULL */ + if (unlikely(timeout)) { + hrtimer_start_expires(&timeout->timer, HRTIMER_MODE_ABS); + if (!hrtimer_active(&timeout->timer)) + timeout->task = NULL; + } + + ret = __rt_mutex_slowlock(lock, state, timeout, &waiter, + detect_deadlock); + set_current_state(TASK_RUNNING); if (unlikely(waiter.task)) @@ -864,9 +891,9 @@ int __sched rt_mutex_lock_interruptible(struct rt_mutex *lock, EXPORT_SYMBOL_GPL(rt_mutex_lock_interruptible); /** - * rt_mutex_lock_interruptible_ktime - lock a rt_mutex interruptible - * the timeout structure is provided - * by the caller + * rt_mutex_timed_lock - lock a rt_mutex interruptible + * the timeout structure is provided + * by the caller * * @lock: the rt_mutex to be locked * @timeout: timeout structure or NULL (no timeout) @@ -875,7 +902,7 @@ EXPORT_SYMBOL_GPL(rt_mutex_lock_interruptible); * Returns: * 0 on success * -EINTR when interrupted by a signal - * -ETIMEOUT when the timeout expired + * -ETIMEDOUT when the timeout expired * -EDEADLK when the lock would deadlock (when deadlock detection is on) */ int @@ -913,7 +940,7 @@ void __sched rt_mutex_unlock(struct rt_mutex *lock) } EXPORT_SYMBOL_GPL(rt_mutex_unlock); -/*** +/** * rt_mutex_destroy - mark a mutex unusable * @lock: the mutex to be destroyed * @@ -986,6 +1013,57 @@ void rt_mutex_proxy_unlock(struct rt_mutex *lock, } /** + * rt_mutex_start_proxy_lock() - Start lock acquisition for another task + * @lock: the rt_mutex to take + * @waiter: the pre-initialized rt_mutex_waiter + * @task: the task to prepare + * @detect_deadlock: perform deadlock detection (1) or not (0) + * + * Returns: + * 0 - task blocked on lock + * 1 - acquired the lock for task, caller should wake it up + * <0 - error + * + * Special API call for FUTEX_REQUEUE_PI support. + */ +int rt_mutex_start_proxy_lock(struct rt_mutex *lock, + struct rt_mutex_waiter *waiter, + struct task_struct *task, int detect_deadlock) +{ + int ret; + + spin_lock(&lock->wait_lock); + + mark_rt_mutex_waiters(lock); + + if (!rt_mutex_owner(lock) || try_to_steal_lock(lock, task)) { + /* We got the lock for task. */ + debug_rt_mutex_lock(lock); + rt_mutex_set_owner(lock, task, 0); + spin_unlock(&lock->wait_lock); + rt_mutex_deadlock_account_lock(lock, task); + return 1; + } + + ret = task_blocks_on_rt_mutex(lock, waiter, task, detect_deadlock); + + if (ret && !waiter->task) { + /* + * Reset the return value. We might have + * returned with -EDEADLK and the owner + * released the lock while we were walking the + * pi chain. Let the waiter sort it out. + */ + ret = 0; + } + spin_unlock(&lock->wait_lock); + + debug_rt_mutex_print_deadlock(waiter); + + return ret; +} + +/** * rt_mutex_next_owner - return the next owner of the lock * * @lock: the rt lock query @@ -1004,3 +1082,57 @@ struct task_struct *rt_mutex_next_owner(struct rt_mutex *lock) return rt_mutex_top_waiter(lock)->task; } + +/** + * rt_mutex_finish_proxy_lock() - Complete lock acquisition + * @lock: the rt_mutex we were woken on + * @to: the timeout, null if none. hrtimer should already have + * been started. + * @waiter: the pre-initialized rt_mutex_waiter + * @detect_deadlock: perform deadlock detection (1) or not (0) + * + * Complete the lock acquisition started our behalf by another thread. + * + * Returns: + * 0 - success + * <0 - error, one of -EINTR, -ETIMEDOUT, or -EDEADLK + * + * Special API call for PI-futex requeue support + */ +int rt_mutex_finish_proxy_lock(struct rt_mutex *lock, + struct hrtimer_sleeper *to, + struct rt_mutex_waiter *waiter, + int detect_deadlock) +{ + int ret; + + spin_lock(&lock->wait_lock); + + set_current_state(TASK_INTERRUPTIBLE); + + ret = __rt_mutex_slowlock(lock, TASK_INTERRUPTIBLE, to, waiter, + detect_deadlock); + + set_current_state(TASK_RUNNING); + + if (unlikely(waiter->task)) + remove_waiter(lock, waiter); + + /* + * try_to_take_rt_mutex() sets the waiter bit unconditionally. We might + * have to fix that up. + */ + fixup_rt_mutex_waiters(lock); + + spin_unlock(&lock->wait_lock); + + /* + * Readjust priority, when we did not get the lock. We might have been + * the pending owner and boosted. Since we did not take the lock, the + * PI boost has to go. + */ + if (unlikely(ret)) + rt_mutex_adjust_prio(current); + + return ret; +} diff --git a/kernel/rtmutex_common.h b/kernel/rtmutex_common.h index e124bf5800e..97a2f81866a 100644 --- a/kernel/rtmutex_common.h +++ b/kernel/rtmutex_common.h @@ -120,6 +120,14 @@ extern void rt_mutex_init_proxy_locked(struct rt_mutex *lock, struct task_struct *proxy_owner); extern void rt_mutex_proxy_unlock(struct rt_mutex *lock, struct task_struct *proxy_owner); +extern int rt_mutex_start_proxy_lock(struct rt_mutex *lock, + struct rt_mutex_waiter *waiter, + struct task_struct *task, + int detect_deadlock); +extern int rt_mutex_finish_proxy_lock(struct rt_mutex *lock, + struct hrtimer_sleeper *to, + struct rt_mutex_waiter *waiter, + int detect_deadlock); #ifdef CONFIG_DEBUG_RT_MUTEXES # include "rtmutex-debug.h" diff --git a/kernel/sched.c b/kernel/sched.c index 73513f4e19d..e7f2cfa6a25 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -39,6 +39,7 @@ #include <linux/completion.h> #include <linux/kernel_stat.h> #include <linux/debug_locks.h> +#include <linux/perf_event.h> #include <linux/security.h> #include <linux/notifier.h> #include <linux/profile.h> @@ -63,22 +64,22 @@ #include <linux/tsacct_kern.h> #include <linux/kprobes.h> #include <linux/delayacct.h> -#include <linux/reciprocal_div.h> #include <linux/unistd.h> #include <linux/pagemap.h> #include <linux/hrtimer.h> #include <linux/tick.h> -#include <linux/bootmem.h> #include <linux/debugfs.h> #include <linux/ctype.h> #include <linux/ftrace.h> -#include <trace/sched.h> #include <asm/tlb.h> #include <asm/irq_regs.h> #include "sched_cpupri.h" +#define CREATE_TRACE_POINTS +#include <trace/events/sched.h> + /* * Convert user-nice values [ -20 ... 0 ... 19 ] * to static priority [ MAX_RT_PRIO..MAX_PRIO-1 ], @@ -118,36 +119,6 @@ */ #define RUNTIME_INF ((u64)~0ULL) -DEFINE_TRACE(sched_wait_task); -DEFINE_TRACE(sched_wakeup); -DEFINE_TRACE(sched_wakeup_new); -DEFINE_TRACE(sched_switch); -DEFINE_TRACE(sched_migrate_task); - -#ifdef CONFIG_SMP - -static void double_rq_lock(struct rq *rq1, struct rq *rq2); - -/* - * Divide a load by a sched group cpu_power : (load / sg->__cpu_power) - * Since cpu_power is a 'constant', we can use a reciprocal divide. - */ -static inline u32 sg_div_cpu_power(const struct sched_group *sg, u32 load) -{ - return reciprocal_divide(load, sg->reciprocal_cpu_power); -} - -/* - * Each time a sched group cpu_power is changed, - * we must compute its reciprocal value - */ -static inline void sg_inc_cpu_power(struct sched_group *sg, u32 val) -{ - sg->__cpu_power += val; - sg->reciprocal_cpu_power = reciprocal_value(sg->__cpu_power); -} -#endif - static inline int rt_policy(int policy) { if (unlikely(policy == SCHED_FIFO || policy == SCHED_RR)) @@ -231,13 +202,20 @@ static void start_rt_bandwidth(struct rt_bandwidth *rt_b) spin_lock(&rt_b->rt_runtime_lock); for (;;) { + unsigned long delta; + ktime_t soft, hard; + if (hrtimer_active(&rt_b->rt_period_timer)) break; now = hrtimer_cb_get_time(&rt_b->rt_period_timer); hrtimer_forward(&rt_b->rt_period_timer, now, rt_b->rt_period); - hrtimer_start_expires(&rt_b->rt_period_timer, - HRTIMER_MODE_ABS); + + soft = hrtimer_get_softexpires(&rt_b->rt_period_timer); + hard = hrtimer_get_expires(&rt_b->rt_period_timer); + delta = ktime_to_ns(ktime_sub(hard, soft)); + __hrtimer_start_range_ns(&rt_b->rt_period_timer, soft, delta, + HRTIMER_MODE_ABS_PINNED, 0); } spin_unlock(&rt_b->rt_runtime_lock); } @@ -306,8 +284,8 @@ void set_tg_uid(struct user_struct *user) /* * Root task group. - * Every UID task group (including init_task_group aka UID-0) will - * be a child to this group. + * Every UID task group (including init_task_group aka UID-0) will + * be a child to this group. */ struct task_group root_task_group; @@ -315,12 +293,12 @@ struct task_group root_task_group; /* Default task group's sched entity on each cpu */ static DEFINE_PER_CPU(struct sched_entity, init_sched_entity); /* Default task group's cfs_rq on each cpu */ -static DEFINE_PER_CPU(struct cfs_rq, init_cfs_rq) ____cacheline_aligned_in_smp; +static DEFINE_PER_CPU_SHARED_ALIGNED(struct cfs_rq, init_tg_cfs_rq); #endif /* CONFIG_FAIR_GROUP_SCHED */ #ifdef CONFIG_RT_GROUP_SCHED static DEFINE_PER_CPU(struct sched_rt_entity, init_sched_rt_entity); -static DEFINE_PER_CPU(struct rt_rq, init_rt_rq) ____cacheline_aligned_in_smp; +static DEFINE_PER_CPU_SHARED_ALIGNED(struct rt_rq, init_rt_rq); #endif /* CONFIG_RT_GROUP_SCHED */ #else /* !CONFIG_USER_SCHED */ #define root_task_group init_task_group @@ -331,6 +309,8 @@ static DEFINE_PER_CPU(struct rt_rq, init_rt_rq) ____cacheline_aligned_in_smp; */ static DEFINE_SPINLOCK(task_group_lock); +#ifdef CONFIG_FAIR_GROUP_SCHED + #ifdef CONFIG_SMP static int root_task_group_empty(void) { @@ -338,7 +318,6 @@ static int root_task_group_empty(void) } #endif -#ifdef CONFIG_FAIR_GROUP_SCHED #ifdef CONFIG_USER_SCHED # define INIT_TASK_GROUP_LOAD (2*NICE_0_LOAD) #else /* !CONFIG_USER_SCHED */ @@ -398,13 +377,6 @@ static inline void set_task_rq(struct task_struct *p, unsigned int cpu) #else -#ifdef CONFIG_SMP -static int root_task_group_empty(void) -{ - return 1; -} -#endif - static inline void set_task_rq(struct task_struct *p, unsigned int cpu) { } static inline struct task_group *task_group(struct task_struct *p) { @@ -490,6 +462,7 @@ struct rt_rq { #endif #ifdef CONFIG_SMP unsigned long rt_nr_migratory; + unsigned long rt_nr_total; int overloaded; struct plist_head pushable_tasks; #endif @@ -533,14 +506,6 @@ struct root_domain { #ifdef CONFIG_SMP struct cpupri cpupri; #endif -#if defined(CONFIG_SCHED_MC) || defined(CONFIG_SCHED_SMT) - /* - * Preferred wake up cpu nominated by sched_mc balance that will be - * used when most cpus are idle in the system indicating overall very - * low system utilisation. Triggered at POWERSAVINGS_BALANCE_WAKEUP(2) - */ - unsigned int sched_mc_preferred_wakeup_cpu; -#endif }; /* @@ -570,7 +535,6 @@ struct rq { #define CPU_LOAD_IDX_MAX 5 unsigned long cpu_load[CPU_LOAD_IDX_MAX]; #ifdef CONFIG_NO_HZ - unsigned long last_tick_seen; unsigned char in_nohz_recently; #endif /* capture load from *all* tasks on this cpu: */ @@ -611,6 +575,7 @@ struct rq { unsigned char idle_at_tick; /* For active balancing */ + int post_schedule; int active_balance; int push_cpu; /* cpu of this runqueue: */ @@ -621,8 +586,17 @@ struct rq { struct task_struct *migration_thread; struct list_head migration_queue; + + u64 rt_avg; + u64 age_stamp; + u64 idle_stamp; + u64 avg_idle; #endif + /* calc_load related fields */ + unsigned long calc_load_update; + long calc_load_active; + #ifdef CONFIG_SCHED_HRTICK #ifdef CONFIG_SMP int hrtick_csd_pending; @@ -656,9 +630,10 @@ struct rq { static DEFINE_PER_CPU_SHARED_ALIGNED(struct rq, runqueues); -static inline void check_preempt_curr(struct rq *rq, struct task_struct *p, int sync) +static inline +void check_preempt_curr(struct rq *rq, struct task_struct *p, int flags) { - rq->curr->sched_class->check_preempt_curr(rq, p, sync); + rq->curr->sched_class->check_preempt_curr(rq, p, flags); } static inline int cpu_of(struct rq *rq) @@ -684,8 +659,9 @@ static inline int cpu_of(struct rq *rq) #define this_rq() (&__get_cpu_var(runqueues)) #define task_rq(p) cpu_rq(task_cpu(p)) #define cpu_curr(cpu) (cpu_rq(cpu)->curr) +#define raw_rq() (&__raw_get_cpu_var(runqueues)) -static inline void update_rq_clock(struct rq *rq) +inline void update_rq_clock(struct rq *rq) { rq->clock = sched_clock_cpu(cpu_of(rq)); } @@ -701,20 +677,15 @@ static inline void update_rq_clock(struct rq *rq) /** * runqueue_is_locked + * @cpu: the processor in question. * * Returns true if the current cpu runqueue is locked. * This interface allows printk to be called with the runqueue lock * held and know whether or not it is OK to wake up the klogd. */ -int runqueue_is_locked(void) +int runqueue_is_locked(int cpu) { - int cpu = get_cpu(); - struct rq *rq = cpu_rq(cpu); - int ret; - - ret = spin_is_locked(&rq->lock); - put_cpu(); - return ret; + return spin_is_locked(&cpu_rq(cpu)->lock); } /* @@ -801,7 +772,7 @@ sched_feat_write(struct file *filp, const char __user *ubuf, if (!sched_feat_names[i]) return -EINVAL; - filp->f_pos += cnt; + *ppos += cnt; return cnt; } @@ -811,7 +782,7 @@ static int sched_feat_open(struct inode *inode, struct file *filp) return single_open(filp, sched_feat_show, NULL); } -static struct file_operations sched_feat_fops = { +static const struct file_operations sched_feat_fops = { .open = sched_feat_open, .write = sched_feat_write, .read = seq_read, @@ -852,6 +823,14 @@ unsigned int sysctl_sched_shares_ratelimit = 250000; unsigned int sysctl_sched_shares_thresh = 4; /* + * period over which we average the RT time consumption, measured + * in ms. + * + * default: 1s + */ +const_debug unsigned int sysctl_sched_time_avg = MSEC_PER_SEC; + +/* * period over which we measure -rt task cpu usage in us. * default: 1s */ @@ -1110,7 +1089,7 @@ static void hrtick_start(struct rq *rq, u64 delay) if (rq == this_rq()) { hrtimer_restart(timer); } else if (!rq->hrtick_csd_pending) { - __smp_call_function_single(cpu_of(rq), &rq->hrtick_csd); + __smp_call_function_single(cpu_of(rq), &rq->hrtick_csd, 0); rq->hrtick_csd_pending = 1; } } @@ -1146,7 +1125,8 @@ static __init void init_hrtick(void) */ static void hrtick_start(struct rq *rq, u64 delay) { - hrtimer_start(&rq->hrtick_timer, ns_to_ktime(delay), HRTIMER_MODE_REL); + __hrtimer_start_range_ns(&rq->hrtick_timer, ns_to_ktime(delay), 0, + HRTIMER_MODE_REL_PINNED, 0); } static inline void init_hrtick(void) @@ -1268,12 +1248,37 @@ void wake_up_idle_cpu(int cpu) } #endif /* CONFIG_NO_HZ */ +static u64 sched_avg_period(void) +{ + return (u64)sysctl_sched_time_avg * NSEC_PER_MSEC / 2; +} + +static void sched_avg_update(struct rq *rq) +{ + s64 period = sched_avg_period(); + + while ((s64)(rq->clock - rq->age_stamp) > period) { + rq->age_stamp += period; + rq->rt_avg /= 2; + } +} + +static void sched_rt_avg_update(struct rq *rq, u64 rt_delta) +{ + rq->rt_avg += rt_delta; + sched_avg_update(rq); +} + #else /* !CONFIG_SMP */ static void resched_task(struct task_struct *p) { assert_spin_locked(&task_rq(p)->lock); set_tsk_need_resched(p); } + +static void sched_rt_avg_update(struct rq *rq, u64 rt_delta) +{ +} #endif /* CONFIG_SMP */ #if BITS_PER_LONG == 32 @@ -1410,10 +1415,22 @@ iter_move_one_task(struct rq *this_rq, int this_cpu, struct rq *busiest, struct rq_iterator *iterator); #endif +/* Time spent by the tasks of the cpu accounting group executing in ... */ +enum cpuacct_stat_index { + CPUACCT_STAT_USER, /* ... user mode */ + CPUACCT_STAT_SYSTEM, /* ... kernel mode */ + + CPUACCT_STAT_NSTATS, +}; + #ifdef CONFIG_CGROUP_CPUACCT static void cpuacct_charge(struct task_struct *tsk, u64 cputime); +static void cpuacct_update_stats(struct task_struct *tsk, + enum cpuacct_stat_index idx, cputime_t val); #else static inline void cpuacct_charge(struct task_struct *tsk, u64 cputime) {} +static inline void cpuacct_update_stats(struct task_struct *tsk, + enum cpuacct_stat_index idx, cputime_t val) {} #endif static inline void inc_cpu_load(struct rq *rq, unsigned long load) @@ -1472,8 +1489,65 @@ static int tg_nop(struct task_group *tg, void *data) #endif #ifdef CONFIG_SMP -static unsigned long source_load(int cpu, int type); -static unsigned long target_load(int cpu, int type); +/* Used instead of source_load when we know the type == 0 */ +static unsigned long weighted_cpuload(const int cpu) +{ + return cpu_rq(cpu)->load.weight; +} + +/* + * Return a low guess at the load of a migration-source cpu weighted + * according to the scheduling class and "nice" value. + * + * We want to under-estimate the load of migration sources, to + * balance conservatively. + */ +static unsigned long source_load(int cpu, int type) +{ + struct rq *rq = cpu_rq(cpu); + unsigned long total = weighted_cpuload(cpu); + + if (type == 0 || !sched_feat(LB_BIAS)) + return total; + + return min(rq->cpu_load[type-1], total); +} + +/* + * Return a high guess at the load of a migration-target cpu weighted + * according to the scheduling class and "nice" value. + */ +static unsigned long target_load(int cpu, int type) +{ + struct rq *rq = cpu_rq(cpu); + unsigned long total = weighted_cpuload(cpu); + + if (type == 0 || !sched_feat(LB_BIAS)) + return total; + + return max(rq->cpu_load[type-1], total); +} + +static struct sched_group *group_of(int cpu) +{ + struct sched_domain *sd = rcu_dereference(cpu_rq(cpu)->sd); + + if (!sd) + return NULL; + + return sd->groups; +} + +static unsigned long power_of(int cpu) +{ + struct sched_group *group = group_of(cpu); + + if (!group) + return SCHED_LOAD_SCALE; + + return group->cpu_power; +} + static int task_hot(struct task_struct *p, u64 now, struct sched_domain *sd); static unsigned long cpu_avg_load_per_task(int cpu) @@ -1491,28 +1565,31 @@ static unsigned long cpu_avg_load_per_task(int cpu) #ifdef CONFIG_FAIR_GROUP_SCHED +static __read_mostly unsigned long *update_shares_data; + static void __set_se_shares(struct sched_entity *se, unsigned long shares); /* * Calculate and set the cpu's group shares. */ -static void -update_group_shares_cpu(struct task_group *tg, int cpu, - unsigned long sd_shares, unsigned long sd_rq_weight) +static void update_group_shares_cpu(struct task_group *tg, int cpu, + unsigned long sd_shares, + unsigned long sd_rq_weight, + unsigned long *usd_rq_weight) { - unsigned long shares; - unsigned long rq_weight; - - if (!tg->se[cpu]) - return; + unsigned long shares, rq_weight; + int boost = 0; - rq_weight = tg->cfs_rq[cpu]->rq_weight; + rq_weight = usd_rq_weight[cpu]; + if (!rq_weight) { + boost = 1; + rq_weight = NICE_0_LOAD; + } /* - * \Sum shares * rq_weight - * shares = ----------------------- - * \Sum rq_weight - * + * \Sum_j shares_j * rq_weight_i + * shares_i = ----------------------------- + * \Sum_j rq_weight_j */ shares = (sd_shares * rq_weight) / sd_rq_weight; shares = clamp_t(unsigned long, shares, MIN_SHARES, MAX_SHARES); @@ -1523,8 +1600,8 @@ update_group_shares_cpu(struct task_group *tg, int cpu, unsigned long flags; spin_lock_irqsave(&rq->lock, flags); - tg->cfs_rq[cpu]->shares = shares; - + tg->cfs_rq[cpu]->rq_weight = boost ? 0 : rq_weight; + tg->cfs_rq[cpu]->shares = boost ? 0 : shares; __set_se_shares(tg->se[cpu], shares); spin_unlock_irqrestore(&rq->lock, flags); } @@ -1537,22 +1614,30 @@ update_group_shares_cpu(struct task_group *tg, int cpu, */ static int tg_shares_up(struct task_group *tg, void *data) { - unsigned long weight, rq_weight = 0; - unsigned long shares = 0; + unsigned long weight, rq_weight = 0, shares = 0; + unsigned long *usd_rq_weight; struct sched_domain *sd = data; + unsigned long flags; int i; + if (!tg->se[0]) + return 0; + + local_irq_save(flags); + usd_rq_weight = per_cpu_ptr(update_shares_data, smp_processor_id()); + for_each_cpu(i, sched_domain_span(sd)) { + weight = tg->cfs_rq[i]->load.weight; + usd_rq_weight[i] = weight; + /* * If there are currently no tasks on the cpu pretend there * is one of average load so that when a new task gets to * run here it will not get delayed by group starvation. */ - weight = tg->cfs_rq[i]->load.weight; if (!weight) weight = NICE_0_LOAD; - tg->cfs_rq[i]->rq_weight = weight; rq_weight += weight; shares += tg->cfs_rq[i]->shares; } @@ -1564,7 +1649,9 @@ static int tg_shares_up(struct task_group *tg, void *data) shares = tg->shares; for_each_cpu(i, sched_domain_span(sd)) - update_group_shares_cpu(tg, i, shares, rq_weight); + update_group_shares_cpu(tg, i, shares, rq_weight, usd_rq_weight); + + local_irq_restore(flags); return 0; } @@ -1594,8 +1681,14 @@ static int tg_load_down(struct task_group *tg, void *data) static void update_shares(struct sched_domain *sd) { - u64 now = cpu_clock(raw_smp_processor_id()); - s64 elapsed = now - sd->last_update; + s64 elapsed; + u64 now; + + if (root_task_group_empty()) + return; + + now = cpu_clock(raw_smp_processor_id()); + elapsed = now - sd->last_update; if (elapsed >= (s64)(u64)sysctl_sched_shares_ratelimit) { sd->last_update = now; @@ -1605,6 +1698,9 @@ static void update_shares(struct sched_domain *sd) static void update_shares_locked(struct rq *rq, struct sched_domain *sd) { + if (root_task_group_empty()) + return; + spin_unlock(&rq->lock); update_shares(sd); spin_lock(&rq->lock); @@ -1612,6 +1708,9 @@ static void update_shares_locked(struct rq *rq, struct sched_domain *sd) static void update_h_load(long cpu) { + if (root_task_group_empty()) + return; + walk_tg_tree(tg_load_down, tg_nop, (void *)cpu); } @@ -1629,6 +1728,8 @@ static inline void update_shares_locked(struct rq *rq, struct sched_domain *sd) #ifdef CONFIG_PREEMPT +static void double_rq_lock(struct rq *rq1, struct rq *rq2); + /* * fair double_lock_balance: Safely acquires both rq->locks in a fair * way at the expense of forcing extra atomic operations in all @@ -1708,6 +1809,8 @@ static void cfs_rq_set_shares(struct cfs_rq *cfs_rq, unsigned long shares) } #endif +static void calc_load_account_active(struct rq *this_rq); + #include "sched_stats.h" #include "sched_idletask.c" #include "sched_fair.c" @@ -1890,14 +1993,40 @@ static inline void check_class_changed(struct rq *rq, struct task_struct *p, p->sched_class->prio_changed(rq, p, oldprio, running); } -#ifdef CONFIG_SMP - -/* Used instead of source_load when we know the type == 0 */ -static unsigned long weighted_cpuload(const int cpu) +/** + * kthread_bind - bind a just-created kthread to a cpu. + * @p: thread created by kthread_create(). + * @cpu: cpu (might not be online, must be possible) for @k to run on. + * + * Description: This function is equivalent to set_cpus_allowed(), + * except that @cpu doesn't need to be online, and the thread must be + * stopped (i.e., just returned from kthread_create()). + * + * Function lives here instead of kthread.c because it messes with + * scheduler internals which require locking. + */ +void kthread_bind(struct task_struct *p, unsigned int cpu) { - return cpu_rq(cpu)->load.weight; + struct rq *rq = cpu_rq(cpu); + unsigned long flags; + + /* Must have done schedule() in kthread() before we set_task_cpu */ + if (!wait_task_inactive(p, TASK_UNINTERRUPTIBLE)) { + WARN_ON(1); + return; + } + + spin_lock_irqsave(&rq->lock, flags); + update_rq_clock(rq); + set_task_cpu(p, cpu); + p->cpus_allowed = cpumask_of_cpu(cpu); + p->rt.nr_cpus_allowed = 1; + p->flags |= PF_THREAD_BOUND; + spin_unlock_irqrestore(&rq->lock, flags); } +EXPORT_SYMBOL(kthread_bind); +#ifdef CONFIG_SMP /* * Is this task likely cache-hot: */ @@ -1909,7 +2038,7 @@ task_hot(struct task_struct *p, u64 now, struct sched_domain *sd) /* * Buddy candidates are cache hot: */ - if (sched_feat(CACHE_HOT_BUDDY) && + if (sched_feat(CACHE_HOT_BUDDY) && this_rq()->nr_running && (&p->se == cfs_rq_of(&p->se)->next || &p->se == cfs_rq_of(&p->se)->last)) return 1; @@ -1938,7 +2067,7 @@ void set_task_cpu(struct task_struct *p, unsigned int new_cpu) clock_offset = old_rq->clock - new_rq->clock; - trace_sched_migrate_task(p, task_cpu(p), new_cpu); + trace_sched_migrate_task(p, new_cpu); #ifdef CONFIG_SCHEDSTATS if (p->se.wait_start) @@ -1947,12 +2076,16 @@ void set_task_cpu(struct task_struct *p, unsigned int new_cpu) p->se.sleep_start -= clock_offset; if (p->se.block_start) p->se.block_start -= clock_offset; +#endif if (old_cpu != new_cpu) { - schedstat_inc(p, se.nr_migrations); + p->se.nr_migrations++; +#ifdef CONFIG_SCHEDSTATS if (task_hot(p, old_rq->clock, NULL)) schedstat_inc(p, se.nr_forced2_migrations); - } #endif + perf_sw_event(PERF_COUNT_SW_CPU_MIGRATIONS, + 1, 1, NULL, 0); + } p->se.vruntime -= old_cfsrq->min_vruntime - new_cfsrq->min_vruntime; @@ -1982,6 +2115,7 @@ migrate_task(struct task_struct *p, int dest_cpu, struct migration_req *req) * it is sufficient to simply update the task's cpu field. */ if (!p->se.on_rq && !task_running(rq, p)) { + update_rq_clock(rq); set_task_cpu(p, dest_cpu); return 0; } @@ -1995,6 +2129,49 @@ migrate_task(struct task_struct *p, int dest_cpu, struct migration_req *req) } /* + * wait_task_context_switch - wait for a thread to complete at least one + * context switch. + * + * @p must not be current. + */ +void wait_task_context_switch(struct task_struct *p) +{ + unsigned long nvcsw, nivcsw, flags; + int running; + struct rq *rq; + + nvcsw = p->nvcsw; + nivcsw = p->nivcsw; + for (;;) { + /* + * The runqueue is assigned before the actual context + * switch. We need to take the runqueue lock. + * + * We could check initially without the lock but it is + * very likely that we need to take the lock in every + * iteration. + */ + rq = task_rq_lock(p, &flags); + running = task_running(rq, p); + task_rq_unlock(rq, &flags); + + if (likely(!running)) + break; + /* + * The switch count is incremented before the actual + * context switch. We thus wait for two switches to be + * sure at least one completed. + */ + if ((p->nvcsw - nvcsw) > 1) + break; + if ((p->nivcsw - nivcsw) > 1) + break; + + cpu_relax(); + } +} + +/* * wait_task_inactive - wait for a thread to unschedule. * * If @match_state is nonzero, it's the @p->state value just checked and @@ -2122,188 +2299,30 @@ void kick_process(struct task_struct *p) smp_send_reschedule(cpu); preempt_enable(); } +EXPORT_SYMBOL_GPL(kick_process); +#endif /* CONFIG_SMP */ -/* - * Return a low guess at the load of a migration-source cpu weighted - * according to the scheduling class and "nice" value. - * - * We want to under-estimate the load of migration sources, to - * balance conservatively. - */ -static unsigned long source_load(int cpu, int type) -{ - struct rq *rq = cpu_rq(cpu); - unsigned long total = weighted_cpuload(cpu); - - if (type == 0 || !sched_feat(LB_BIAS)) - return total; - - return min(rq->cpu_load[type-1], total); -} - -/* - * Return a high guess at the load of a migration-target cpu weighted - * according to the scheduling class and "nice" value. - */ -static unsigned long target_load(int cpu, int type) -{ - struct rq *rq = cpu_rq(cpu); - unsigned long total = weighted_cpuload(cpu); - - if (type == 0 || !sched_feat(LB_BIAS)) - return total; - - return max(rq->cpu_load[type-1], total); -} - -/* - * find_idlest_group finds and returns the least busy CPU group within the - * domain. - */ -static struct sched_group * -find_idlest_group(struct sched_domain *sd, struct task_struct *p, int this_cpu) -{ - struct sched_group *idlest = NULL, *this = NULL, *group = sd->groups; - unsigned long min_load = ULONG_MAX, this_load = 0; - int load_idx = sd->forkexec_idx; - int imbalance = 100 + (sd->imbalance_pct-100)/2; - - do { - unsigned long load, avg_load; - int local_group; - int i; - - /* Skip over this group if it has no CPUs allowed */ - if (!cpumask_intersects(sched_group_cpus(group), - &p->cpus_allowed)) - continue; - - local_group = cpumask_test_cpu(this_cpu, - sched_group_cpus(group)); - - /* Tally up the load of all CPUs in the group */ - avg_load = 0; - - for_each_cpu(i, sched_group_cpus(group)) { - /* Bias balancing toward cpus of our domain */ - if (local_group) - load = source_load(i, load_idx); - else - load = target_load(i, load_idx); - - avg_load += load; - } - - /* Adjust by relative CPU power of the group */ - avg_load = sg_div_cpu_power(group, - avg_load * SCHED_LOAD_SCALE); - - if (local_group) { - this_load = avg_load; - this = group; - } else if (avg_load < min_load) { - min_load = avg_load; - idlest = group; - } - } while (group = group->next, group != sd->groups); - - if (!idlest || 100*this_load < imbalance*min_load) - return NULL; - return idlest; -} - -/* - * find_idlest_cpu - find the idlest cpu among the cpus in group. - */ -static int -find_idlest_cpu(struct sched_group *group, struct task_struct *p, int this_cpu) -{ - unsigned long load, min_load = ULONG_MAX; - int idlest = -1; - int i; - - /* Traverse only the allowed CPUs */ - for_each_cpu_and(i, sched_group_cpus(group), &p->cpus_allowed) { - load = weighted_cpuload(i); - - if (load < min_load || (load == min_load && i == this_cpu)) { - min_load = load; - idlest = i; - } - } - - return idlest; -} - -/* - * sched_balance_self: balance the current task (running on cpu) in domains - * that have the 'flag' flag set. In practice, this is SD_BALANCE_FORK and - * SD_BALANCE_EXEC. - * - * Balance, ie. select the least loaded group. - * - * Returns the target CPU number, or the same CPU if no balancing is needed. +/** + * task_oncpu_function_call - call a function on the cpu on which a task runs + * @p: the task to evaluate + * @func: the function to be called + * @info: the function call argument * - * preempt must be disabled. + * Calls the function @func when the task is currently running. This might + * be on the current CPU, which just calls the function directly */ -static int sched_balance_self(int cpu, int flag) +void task_oncpu_function_call(struct task_struct *p, + void (*func) (void *info), void *info) { - struct task_struct *t = current; - struct sched_domain *tmp, *sd = NULL; - - for_each_domain(cpu, tmp) { - /* - * If power savings logic is enabled for a domain, stop there. - */ - if (tmp->flags & SD_POWERSAVINGS_BALANCE) - break; - if (tmp->flags & flag) - sd = tmp; - } - - if (sd) - update_shares(sd); - - while (sd) { - struct sched_group *group; - int new_cpu, weight; - - if (!(sd->flags & flag)) { - sd = sd->child; - continue; - } - - group = find_idlest_group(sd, t, cpu); - if (!group) { - sd = sd->child; - continue; - } - - new_cpu = find_idlest_cpu(group, t, cpu); - if (new_cpu == -1 || new_cpu == cpu) { - /* Now try balancing at a lower domain level of cpu */ - sd = sd->child; - continue; - } - - /* Now try balancing at a lower domain level of new_cpu */ - cpu = new_cpu; - weight = cpumask_weight(sched_domain_span(sd)); - sd = NULL; - for_each_domain(cpu, tmp) { - if (weight <= cpumask_weight(sched_domain_span(tmp))) - break; - if (tmp->flags & flag) - sd = tmp; - } - /* while loop will break here if sd == NULL */ - } + int cpu; - return cpu; + preempt_disable(); + cpu = task_cpu(p); + if (task_curr(p)) + smp_call_function_single(cpu, func, info, 1); + preempt_enable(); } -#endif /* CONFIG_SMP */ - /*** * try_to_wake_up - wake up a thread * @p: the to-be-woken-up thread @@ -2318,37 +2337,22 @@ static int sched_balance_self(int cpu, int flag) * * returns failure only if the task is already active. */ -static int try_to_wake_up(struct task_struct *p, unsigned int state, int sync) +static int try_to_wake_up(struct task_struct *p, unsigned int state, + int wake_flags) { int cpu, orig_cpu, this_cpu, success = 0; unsigned long flags; - long old_state; - struct rq *rq; + struct rq *rq, *orig_rq; if (!sched_feat(SYNC_WAKEUPS)) - sync = 0; - -#ifdef CONFIG_SMP - if (sched_feat(LB_WAKEUP_UPDATE) && !root_task_group_empty()) { - struct sched_domain *sd; + wake_flags &= ~WF_SYNC; - this_cpu = raw_smp_processor_id(); - cpu = task_cpu(p); - - for_each_domain(this_cpu, sd) { - if (cpumask_test_cpu(cpu, sched_domain_span(sd))) { - update_shares(sd); - break; - } - } - } -#endif + this_cpu = get_cpu(); smp_wmb(); - rq = task_rq_lock(p, &flags); + rq = orig_rq = task_rq_lock(p, &flags); update_rq_clock(rq); - old_state = p->state; - if (!(old_state & state)) + if (!(p->state & state)) goto out; if (p->se.on_rq) @@ -2356,27 +2360,34 @@ static int try_to_wake_up(struct task_struct *p, unsigned int state, int sync) cpu = task_cpu(p); orig_cpu = cpu; - this_cpu = smp_processor_id(); #ifdef CONFIG_SMP if (unlikely(task_running(rq, p))) goto out_activate; - cpu = p->sched_class->select_task_rq(p, sync); + /* + * In order to handle concurrent wakeups and release the rq->lock + * we put the task in TASK_WAKING state. + * + * First fix up the nr_uninterruptible count: + */ + if (task_contributes_to_load(p)) + rq->nr_uninterruptible--; + p->state = TASK_WAKING; + task_rq_unlock(rq, &flags); + + cpu = p->sched_class->select_task_rq(p, SD_BALANCE_WAKE, wake_flags); if (cpu != orig_cpu) { + local_irq_save(flags); + rq = cpu_rq(cpu); + update_rq_clock(rq); set_task_cpu(p, cpu); - task_rq_unlock(rq, &flags); - /* might preempt at this point */ - rq = task_rq_lock(p, &flags); - old_state = p->state; - if (!(old_state & state)) - goto out; - if (p->se.on_rq) - goto out_running; - - this_cpu = smp_processor_id(); - cpu = task_cpu(p); + local_irq_restore(flags); } + rq = task_rq_lock(p, &flags); + + WARN_ON(p->state != TASK_WAKING); + cpu = task_cpu(p); #ifdef CONFIG_SCHEDSTATS schedstat_inc(rq, ttwu_count); @@ -2396,7 +2407,7 @@ static int try_to_wake_up(struct task_struct *p, unsigned int state, int sync) out_activate: #endif /* CONFIG_SMP */ schedstat_inc(p, se.nr_wakeups); - if (sync) + if (wake_flags & WF_SYNC) schedstat_inc(p, se.nr_wakeups_sync); if (orig_cpu != cpu) schedstat_inc(p, se.nr_wakeups_migrate); @@ -2425,19 +2436,42 @@ out_activate: out_running: trace_sched_wakeup(rq, p, success); - check_preempt_curr(rq, p, sync); + check_preempt_curr(rq, p, wake_flags); p->state = TASK_RUNNING; #ifdef CONFIG_SMP if (p->sched_class->task_wake_up) p->sched_class->task_wake_up(rq, p); + + if (unlikely(rq->idle_stamp)) { + u64 delta = rq->clock - rq->idle_stamp; + u64 max = 2*sysctl_sched_migration_cost; + + if (delta > max) + rq->avg_idle = max; + else + update_avg(&rq->avg_idle, delta); + rq->idle_stamp = 0; + } #endif out: task_rq_unlock(rq, &flags); + put_cpu(); return success; } +/** + * wake_up_process - Wake up a specific process + * @p: The process to be woken up. + * + * Attempt to wake up the nominated process and move it to the set of runnable + * processes. Returns 1 if the process was woken up, 0 if it was already + * running. + * + * It may be assumed that this function implies a write memory barrier before + * changing the task state if and only if any tasks are woken up. + */ int wake_up_process(struct task_struct *p) { return try_to_wake_up(p, TASK_ALL, 0); @@ -2460,21 +2494,45 @@ static void __sched_fork(struct task_struct *p) p->se.exec_start = 0; p->se.sum_exec_runtime = 0; p->se.prev_sum_exec_runtime = 0; + p->se.nr_migrations = 0; p->se.last_wakeup = 0; p->se.avg_overlap = 0; p->se.start_runtime = 0; p->se.avg_wakeup = sysctl_sched_wakeup_granularity; + p->se.avg_running = 0; #ifdef CONFIG_SCHEDSTATS - p->se.wait_start = 0; - p->se.sum_sleep_runtime = 0; - p->se.sleep_start = 0; - p->se.block_start = 0; - p->se.sleep_max = 0; - p->se.block_max = 0; - p->se.exec_max = 0; - p->se.slice_max = 0; - p->se.wait_max = 0; + p->se.wait_start = 0; + p->se.wait_max = 0; + p->se.wait_count = 0; + p->se.wait_sum = 0; + + p->se.sleep_start = 0; + p->se.sleep_max = 0; + p->se.sum_sleep_runtime = 0; + + p->se.block_start = 0; + p->se.block_max = 0; + p->se.exec_max = 0; + p->se.slice_max = 0; + + p->se.nr_migrations_cold = 0; + p->se.nr_failed_migrations_affine = 0; + p->se.nr_failed_migrations_running = 0; + p->se.nr_failed_migrations_hot = 0; + p->se.nr_forced_migrations = 0; + p->se.nr_forced2_migrations = 0; + + p->se.nr_wakeups = 0; + p->se.nr_wakeups_sync = 0; + p->se.nr_wakeups_migrate = 0; + p->se.nr_wakeups_local = 0; + p->se.nr_wakeups_remote = 0; + p->se.nr_wakeups_affine = 0; + p->se.nr_wakeups_affine_attempts = 0; + p->se.nr_wakeups_passive = 0; + p->se.nr_wakeups_idle = 0; + #endif INIT_LIST_HEAD(&p->rt.run_list); @@ -2500,21 +2558,48 @@ static void __sched_fork(struct task_struct *p) void sched_fork(struct task_struct *p, int clone_flags) { int cpu = get_cpu(); + unsigned long flags; __sched_fork(p); -#ifdef CONFIG_SMP - cpu = sched_balance_self(cpu, SD_BALANCE_FORK); -#endif - set_task_cpu(p, cpu); + /* + * Revert to default priority/policy on fork if requested. + */ + if (unlikely(p->sched_reset_on_fork)) { + if (p->policy == SCHED_FIFO || p->policy == SCHED_RR) { + p->policy = SCHED_NORMAL; + p->normal_prio = p->static_prio; + } + + if (PRIO_TO_NICE(p->static_prio) < 0) { + p->static_prio = NICE_TO_PRIO(0); + p->normal_prio = p->static_prio; + set_load_weight(p); + } + + /* + * We don't need the reset flag anymore after the fork. It has + * fulfilled its duty: + */ + p->sched_reset_on_fork = 0; + } /* - * Make sure we do not leak PI boosting priority to the child: + * Make sure we do not leak PI boosting priority to the child. */ p->prio = current->normal_prio; + if (!rt_prio(p->prio)) p->sched_class = &fair_sched_class; +#ifdef CONFIG_SMP + cpu = p->sched_class->select_task_rq(p, SD_BALANCE_FORK, 0); +#endif + local_irq_save(flags); + update_rq_clock(cpu_rq(cpu)); + set_task_cpu(p, cpu); + local_irq_restore(flags); + #if defined(CONFIG_SCHEDSTATS) || defined(CONFIG_TASK_DELAY_ACCT) if (likely(sched_info_on())) memset(&p->sched_info, 0, sizeof(p->sched_info)); @@ -2547,8 +2632,6 @@ void wake_up_new_task(struct task_struct *p, unsigned long clone_flags) BUG_ON(p->state != TASK_RUNNING); update_rq_clock(rq); - p->prio = effective_prio(p); - if (!p->sched_class->task_new || !current->se.on_rq) { activate_task(rq, p, 0); } else { @@ -2560,7 +2643,7 @@ void wake_up_new_task(struct task_struct *p, unsigned long clone_flags) inc_nr_running(rq); } trace_sched_wakeup_new(rq, p, 1); - check_preempt_curr(rq, p, 0); + check_preempt_curr(rq, p, WF_FORK); #ifdef CONFIG_SMP if (p->sched_class->task_wake_up) p->sched_class->task_wake_up(rq, p); @@ -2668,12 +2751,6 @@ static void finish_task_switch(struct rq *rq, struct task_struct *prev) { struct mm_struct *mm = rq->prev_mm; long prev_state; -#ifdef CONFIG_SMP - int post_schedule = 0; - - if (current->sched_class->needs_post_schedule) - post_schedule = current->sched_class->needs_post_schedule(rq); -#endif rq->prev_mm = NULL; @@ -2690,11 +2767,8 @@ static void finish_task_switch(struct rq *rq, struct task_struct *prev) */ prev_state = prev->state; finish_arch_switch(prev); + perf_event_task_sched_in(current, cpu_of(rq)); finish_lock_switch(rq, prev); -#ifdef CONFIG_SMP - if (post_schedule) - current->sched_class->post_schedule(rq); -#endif fire_sched_in_preempt_notifiers(current); if (mm) @@ -2709,6 +2783,42 @@ static void finish_task_switch(struct rq *rq, struct task_struct *prev) } } +#ifdef CONFIG_SMP + +/* assumes rq->lock is held */ +static inline void pre_schedule(struct rq *rq, struct task_struct *prev) +{ + if (prev->sched_class->pre_schedule) + prev->sched_class->pre_schedule(rq, prev); +} + +/* rq->lock is NOT held, but preemption is disabled */ +static inline void post_schedule(struct rq *rq) +{ + if (rq->post_schedule) { + unsigned long flags; + + spin_lock_irqsave(&rq->lock, flags); + if (rq->curr->sched_class->post_schedule) + rq->curr->sched_class->post_schedule(rq); + spin_unlock_irqrestore(&rq->lock, flags); + + rq->post_schedule = 0; + } +} + +#else + +static inline void pre_schedule(struct rq *rq, struct task_struct *p) +{ +} + +static inline void post_schedule(struct rq *rq) +{ +} + +#endif + /** * schedule_tail - first thing a freshly forked thread must call. * @prev: the thread we just switched away from. @@ -2719,6 +2829,13 @@ asmlinkage void schedule_tail(struct task_struct *prev) struct rq *rq = this_rq(); finish_task_switch(rq, prev); + + /* + * FIXME: do we need to worry about rq being invalidated by the + * task_switch? + */ + post_schedule(rq); + #ifdef __ARCH_WANT_UNLOCKED_CTXSW /* In this case, finish_task_switch does not reenable preemption */ preempt_enable(); @@ -2746,16 +2863,16 @@ context_switch(struct rq *rq, struct task_struct *prev, * combine the page table reload and the switch backend into * one hypercall. */ - arch_enter_lazy_cpu_mode(); + arch_start_context_switch(prev); - if (unlikely(!mm)) { + if (likely(!mm)) { next->active_mm = oldmm; atomic_inc(&oldmm->mm_count); enter_lazy_tlb(oldmm, next); } else switch_mm(oldmm, mm, next); - if (unlikely(!prev->mm)) { + if (likely(!prev->mm)) { prev->active_mm = NULL; rq->prev_mm = oldmm; } @@ -2836,19 +2953,85 @@ unsigned long nr_iowait(void) return sum; } -unsigned long nr_active(void) +unsigned long nr_iowait_cpu(void) +{ + struct rq *this = this_rq(); + return atomic_read(&this->nr_iowait); +} + +unsigned long this_cpu_load(void) { - unsigned long i, running = 0, uninterruptible = 0; + struct rq *this = this_rq(); + return this->cpu_load[0]; +} - for_each_online_cpu(i) { - running += cpu_rq(i)->nr_running; - uninterruptible += cpu_rq(i)->nr_uninterruptible; - } - if (unlikely((long)uninterruptible < 0)) - uninterruptible = 0; +/* Variables and functions for calc_load */ +static atomic_long_t calc_load_tasks; +static unsigned long calc_load_update; +unsigned long avenrun[3]; +EXPORT_SYMBOL(avenrun); - return running + uninterruptible; +/** + * get_avenrun - get the load average array + * @loads: pointer to dest load array + * @offset: offset to add + * @shift: shift count to shift the result left + * + * These values are estimates at best, so no need for locking. + */ +void get_avenrun(unsigned long *loads, unsigned long offset, int shift) +{ + loads[0] = (avenrun[0] + offset) << shift; + loads[1] = (avenrun[1] + offset) << shift; + loads[2] = (avenrun[2] + offset) << shift; +} + +static unsigned long +calc_load(unsigned long load, unsigned long exp, unsigned long active) +{ + load *= exp; + load += active * (FIXED_1 - exp); + return load >> FSHIFT; +} + +/* + * calc_load - update the avenrun load estimates 10 ticks after the + * CPUs have updated calc_load_tasks. + */ +void calc_global_load(void) +{ + unsigned long upd = calc_load_update + 10; + long active; + + if (time_before(jiffies, upd)) + return; + + active = atomic_long_read(&calc_load_tasks); + active = active > 0 ? active * FIXED_1 : 0; + + avenrun[0] = calc_load(avenrun[0], EXP_1, active); + avenrun[1] = calc_load(avenrun[1], EXP_5, active); + avenrun[2] = calc_load(avenrun[2], EXP_15, active); + + calc_load_update += LOAD_FREQ; +} + +/* + * Either called from update_cpu_load() or from a cpu going idle + */ +static void calc_load_account_active(struct rq *this_rq) +{ + long nr_active, delta; + + nr_active = this_rq->nr_running; + nr_active += (long) this_rq->nr_uninterruptible; + + if (nr_active != this_rq->calc_load_active) { + delta = nr_active - this_rq->calc_load_active; + this_rq->calc_load_active = nr_active; + atomic_long_add(delta, &calc_load_tasks); + } } /* @@ -2879,6 +3062,11 @@ static void update_cpu_load(struct rq *this_rq) new_load += scale-1; this_rq->cpu_load[i] = (old_load*(scale-1) + new_load) >> i; } + + if (time_after_eq(jiffies, this_rq->calc_load_update)) { + this_rq->calc_load_update += LOAD_FREQ; + calc_load_account_active(this_rq); + } } #ifdef CONFIG_SMP @@ -2968,7 +3156,7 @@ out: void sched_exec(void) { int new_cpu, this_cpu = get_cpu(); - new_cpu = sched_balance_self(this_cpu, SD_BALANCE_EXEC); + new_cpu = current->sched_class->select_task_rq(current, SD_BALANCE_EXEC, 0); put_cpu(); if (new_cpu != this_cpu) sched_migrate_task(current, new_cpu); @@ -3183,9 +3371,10 @@ static int move_one_task(struct rq *this_rq, int this_cpu, struct rq *busiest, { const struct sched_class *class; - for (class = sched_class_highest; class; class = class->next) + for_each_class(class) { if (class->move_one_task(this_rq, this_cpu, busiest, sd, idle)) return 1; + } return 0; } @@ -3348,7 +3537,7 @@ static inline void update_sd_power_savings_stats(struct sched_group *group, * capacity but still has some space to pick up some load * from other group and save more power */ - if (sgs->sum_nr_running > sgs->group_capacity - 1) + if (sgs->sum_nr_running + 1 > sgs->group_capacity) return; if (sgs->sum_nr_running > sds->leader_nr_running || @@ -3387,11 +3576,6 @@ static inline int check_power_save_busiest_group(struct sd_lb_stats *sds, *imbalance = sds->min_load_per_task; sds->busiest = sds->group_min; - if (sched_mc_power_savings >= POWERSAVINGS_BALANCE_WAKEUP) { - cpu_rq(this_cpu)->rd->sched_mc_preferred_wakeup_cpu = - group_first_cpu(sds->group_leader); - } - return 1; } @@ -3416,8 +3600,105 @@ static inline int check_power_save_busiest_group(struct sd_lb_stats *sds, #endif /* CONFIG_SCHED_MC || CONFIG_SCHED_SMT */ +unsigned long default_scale_freq_power(struct sched_domain *sd, int cpu) +{ + return SCHED_LOAD_SCALE; +} + +unsigned long __weak arch_scale_freq_power(struct sched_domain *sd, int cpu) +{ + return default_scale_freq_power(sd, cpu); +} + +unsigned long default_scale_smt_power(struct sched_domain *sd, int cpu) +{ + unsigned long weight = cpumask_weight(sched_domain_span(sd)); + unsigned long smt_gain = sd->smt_gain; + + smt_gain /= weight; + + return smt_gain; +} + +unsigned long __weak arch_scale_smt_power(struct sched_domain *sd, int cpu) +{ + return default_scale_smt_power(sd, cpu); +} + +unsigned long scale_rt_power(int cpu) +{ + struct rq *rq = cpu_rq(cpu); + u64 total, available; + + sched_avg_update(rq); + + total = sched_avg_period() + (rq->clock - rq->age_stamp); + available = total - rq->rt_avg; + + if (unlikely((s64)total < SCHED_LOAD_SCALE)) + total = SCHED_LOAD_SCALE; + + total >>= SCHED_LOAD_SHIFT; + + return div_u64(available, total); +} + +static void update_cpu_power(struct sched_domain *sd, int cpu) +{ + unsigned long weight = cpumask_weight(sched_domain_span(sd)); + unsigned long power = SCHED_LOAD_SCALE; + struct sched_group *sdg = sd->groups; + + if (sched_feat(ARCH_POWER)) + power *= arch_scale_freq_power(sd, cpu); + else + power *= default_scale_freq_power(sd, cpu); + + power >>= SCHED_LOAD_SHIFT; + + if ((sd->flags & SD_SHARE_CPUPOWER) && weight > 1) { + if (sched_feat(ARCH_POWER)) + power *= arch_scale_smt_power(sd, cpu); + else + power *= default_scale_smt_power(sd, cpu); + + power >>= SCHED_LOAD_SHIFT; + } + + power *= scale_rt_power(cpu); + power >>= SCHED_LOAD_SHIFT; + + if (!power) + power = 1; + + sdg->cpu_power = power; +} + +static void update_group_power(struct sched_domain *sd, int cpu) +{ + struct sched_domain *child = sd->child; + struct sched_group *group, *sdg = sd->groups; + unsigned long power; + + if (!child) { + update_cpu_power(sd, cpu); + return; + } + + power = 0; + + group = child->groups; + do { + power += group->cpu_power; + group = group->next; + } while (group != child->groups); + + sdg->cpu_power = power; +} + /** * update_sg_lb_stats - Update sched_group's statistics for load balancing. + * @sd: The sched_domain whose statistics are to be updated. * @group: sched_group whose statistics are to be updated. * @this_cpu: Cpu for which load balance is currently performed. * @idle: Idle status of this_cpu @@ -3428,7 +3709,8 @@ static inline int check_power_save_busiest_group(struct sd_lb_stats *sds, * @balance: Should we balance. * @sgs: variable to hold the statistics for this group. */ -static inline void update_sg_lb_stats(struct sched_group *group, int this_cpu, +static inline void update_sg_lb_stats(struct sched_domain *sd, + struct sched_group *group, int this_cpu, enum cpu_idle_type idle, int load_idx, int *sd_idle, int local_group, const struct cpumask *cpus, int *balance, struct sg_lb_stats *sgs) @@ -3439,8 +3721,11 @@ static inline void update_sg_lb_stats(struct sched_group *group, int this_cpu, unsigned long sum_avg_load_per_task; unsigned long avg_load_per_task; - if (local_group) + if (local_group) { balance_cpu = group_first_cpu(group); + if (balance_cpu == this_cpu) + update_group_power(sd, this_cpu); + } /* Tally up the load of all CPUs in the group */ sum_avg_load_per_task = avg_load_per_task = 0; @@ -3489,8 +3774,7 @@ static inline void update_sg_lb_stats(struct sched_group *group, int this_cpu, } /* Adjust by relative CPU power of the group */ - sgs->avg_load = sg_div_cpu_power(group, - sgs->group_load * SCHED_LOAD_SCALE); + sgs->avg_load = (sgs->group_load * SCHED_LOAD_SCALE) / group->cpu_power; /* @@ -3502,14 +3786,14 @@ static inline void update_sg_lb_stats(struct sched_group *group, int this_cpu, * normalized nr_running number somewhere that negates * the hierarchy? */ - avg_load_per_task = sg_div_cpu_power(group, - sum_avg_load_per_task * SCHED_LOAD_SCALE); + avg_load_per_task = (sum_avg_load_per_task * SCHED_LOAD_SCALE) / + group->cpu_power; if ((max_cpu_load - min_cpu_load) > 2*avg_load_per_task) sgs->group_imb = 1; - sgs->group_capacity = group->__cpu_power / SCHED_LOAD_SCALE; - + sgs->group_capacity = + DIV_ROUND_CLOSEST(group->cpu_power, SCHED_LOAD_SCALE); } /** @@ -3527,9 +3811,13 @@ static inline void update_sd_lb_stats(struct sched_domain *sd, int this_cpu, const struct cpumask *cpus, int *balance, struct sd_lb_stats *sds) { + struct sched_domain *child = sd->child; struct sched_group *group = sd->groups; struct sg_lb_stats sgs; - int load_idx; + int load_idx, prefer_sibling = 0; + + if (child && child->flags & SD_PREFER_SIBLING) + prefer_sibling = 1; init_sd_power_savings_stats(sd, sds, idle); load_idx = get_sd_load_idx(sd, idle); @@ -3540,14 +3828,22 @@ static inline void update_sd_lb_stats(struct sched_domain *sd, int this_cpu, local_group = cpumask_test_cpu(this_cpu, sched_group_cpus(group)); memset(&sgs, 0, sizeof(sgs)); - update_sg_lb_stats(group, this_cpu, idle, load_idx, sd_idle, + update_sg_lb_stats(sd, group, this_cpu, idle, load_idx, sd_idle, local_group, cpus, balance, &sgs); if (local_group && balance && !(*balance)) return; sds->total_load += sgs.group_load; - sds->total_pwr += group->__cpu_power; + sds->total_pwr += group->cpu_power; + + /* + * In case the child domain prefers tasks go to siblings + * first, lower the group capacity to one so that we'll try + * and move all the excess tasks away. + */ + if (prefer_sibling) + sgs.group_capacity = min(sgs.group_capacity, 1UL); if (local_group) { sds->this_load = sgs.avg_load; @@ -3567,7 +3863,6 @@ static inline void update_sd_lb_stats(struct sched_domain *sd, int this_cpu, update_sd_power_savings_stats(group, sds, local_group, &sgs); group = group->next; } while (group != sd->groups); - } /** @@ -3605,28 +3900,28 @@ static inline void fix_small_imbalance(struct sd_lb_stats *sds, * moving them. */ - pwr_now += sds->busiest->__cpu_power * + pwr_now += sds->busiest->cpu_power * min(sds->busiest_load_per_task, sds->max_load); - pwr_now += sds->this->__cpu_power * + pwr_now += sds->this->cpu_power * min(sds->this_load_per_task, sds->this_load); pwr_now /= SCHED_LOAD_SCALE; /* Amount of load we'd subtract */ - tmp = sg_div_cpu_power(sds->busiest, - sds->busiest_load_per_task * SCHED_LOAD_SCALE); + tmp = (sds->busiest_load_per_task * SCHED_LOAD_SCALE) / + sds->busiest->cpu_power; if (sds->max_load > tmp) - pwr_move += sds->busiest->__cpu_power * + pwr_move += sds->busiest->cpu_power * min(sds->busiest_load_per_task, sds->max_load - tmp); /* Amount of load we'd add */ - if (sds->max_load * sds->busiest->__cpu_power < + if (sds->max_load * sds->busiest->cpu_power < sds->busiest_load_per_task * SCHED_LOAD_SCALE) - tmp = sg_div_cpu_power(sds->this, - sds->max_load * sds->busiest->__cpu_power); + tmp = (sds->max_load * sds->busiest->cpu_power) / + sds->this->cpu_power; else - tmp = sg_div_cpu_power(sds->this, - sds->busiest_load_per_task * SCHED_LOAD_SCALE); - pwr_move += sds->this->__cpu_power * + tmp = (sds->busiest_load_per_task * SCHED_LOAD_SCALE) / + sds->this->cpu_power; + pwr_move += sds->this->cpu_power * min(sds->this_load_per_task, sds->this_load + tmp); pwr_move /= SCHED_LOAD_SCALE; @@ -3661,8 +3956,8 @@ static inline void calculate_imbalance(struct sd_lb_stats *sds, int this_cpu, sds->max_load - sds->busiest_load_per_task); /* How much load to actually move to equalise the imbalance */ - *imbalance = min(max_pull * sds->busiest->__cpu_power, - (sds->avg_load - sds->this_load) * sds->this->__cpu_power) + *imbalance = min(max_pull * sds->busiest->cpu_power, + (sds->avg_load - sds->this_load) * sds->this->cpu_power) / SCHED_LOAD_SCALE; /* @@ -3792,15 +4087,18 @@ find_busiest_queue(struct sched_group *group, enum cpu_idle_type idle, int i; for_each_cpu(i, sched_group_cpus(group)) { + unsigned long power = power_of(i); + unsigned long capacity = DIV_ROUND_CLOSEST(power, SCHED_LOAD_SCALE); unsigned long wl; if (!cpumask_test_cpu(i, cpus)) continue; rq = cpu_rq(i); - wl = weighted_cpuload(i); + wl = weighted_cpuload(i) * SCHED_LOAD_SCALE; + wl /= power; - if (rq->nr_running == 1 && wl > imbalance) + if (capacity && rq->nr_running == 1 && wl > imbalance) continue; if (wl > max_load) { @@ -3818,21 +4116,25 @@ find_busiest_queue(struct sched_group *group, enum cpu_idle_type idle, */ #define MAX_PINNED_INTERVAL 512 +/* Working cpumask for load_balance and load_balance_newidle. */ +static DEFINE_PER_CPU(cpumask_var_t, load_balance_tmpmask); + /* * Check this_cpu to ensure it is balanced within domain. Attempt to move * tasks if there is an imbalance. */ static int load_balance(int this_cpu, struct rq *this_rq, struct sched_domain *sd, enum cpu_idle_type idle, - int *balance, struct cpumask *cpus) + int *balance) { int ld_moved, all_pinned = 0, active_balance = 0, sd_idle = 0; struct sched_group *group; unsigned long imbalance; struct rq *busiest; unsigned long flags; + struct cpumask *cpus = __get_cpu_var(load_balance_tmpmask); - cpumask_setall(cpus); + cpumask_copy(cpus, cpu_online_mask); /* * When power savings policy is enabled for the parent domain, idle @@ -3985,8 +4287,7 @@ out: * this_rq is locked. */ static int -load_balance_newidle(int this_cpu, struct rq *this_rq, struct sched_domain *sd, - struct cpumask *cpus) +load_balance_newidle(int this_cpu, struct rq *this_rq, struct sched_domain *sd) { struct sched_group *group; struct rq *busiest = NULL; @@ -3994,8 +4295,9 @@ load_balance_newidle(int this_cpu, struct rq *this_rq, struct sched_domain *sd, int ld_moved = 0; int sd_idle = 0; int all_pinned = 0; + struct cpumask *cpus = __get_cpu_var(load_balance_tmpmask); - cpumask_setall(cpus); + cpumask_copy(cpus, cpu_online_mask); /* * When power savings policy is enabled for the parent domain, idle @@ -4134,9 +4436,10 @@ static void idle_balance(int this_cpu, struct rq *this_rq) struct sched_domain *sd; int pulled_task = 0; unsigned long next_balance = jiffies + HZ; - cpumask_var_t tmpmask; - if (!alloc_cpumask_var(&tmpmask, GFP_ATOMIC)) + this_rq->idle_stamp = this_rq->clock; + + if (this_rq->avg_idle < sysctl_sched_migration_cost) return; for_each_domain(this_cpu, sd) { @@ -4148,13 +4451,15 @@ static void idle_balance(int this_cpu, struct rq *this_rq) if (sd->flags & SD_BALANCE_NEWIDLE) /* If we've pulled tasks over stop searching: */ pulled_task = load_balance_newidle(this_cpu, this_rq, - sd, tmpmask); + sd); interval = msecs_to_jiffies(sd->balance_interval); if (time_after(next_balance, sd->last_balance + interval)) next_balance = sd->last_balance + interval; - if (pulled_task) + if (pulled_task) { + this_rq->idle_stamp = 0; break; + } } if (pulled_task || time_after(jiffies, this_rq->next_balance)) { /* @@ -4163,7 +4468,6 @@ static void idle_balance(int this_cpu, struct rq *this_rq) */ this_rq->next_balance = next_balance; } - free_cpumask_var(tmpmask); } /* @@ -4221,10 +4525,131 @@ static void active_load_balance(struct rq *busiest_rq, int busiest_cpu) static struct { atomic_t load_balancer; cpumask_var_t cpu_mask; + cpumask_var_t ilb_grp_nohz_mask; } nohz ____cacheline_aligned = { .load_balancer = ATOMIC_INIT(-1), }; +int get_nohz_load_balancer(void) +{ + return atomic_read(&nohz.load_balancer); +} + +#if defined(CONFIG_SCHED_MC) || defined(CONFIG_SCHED_SMT) +/** + * lowest_flag_domain - Return lowest sched_domain containing flag. + * @cpu: The cpu whose lowest level of sched domain is to + * be returned. + * @flag: The flag to check for the lowest sched_domain + * for the given cpu. + * + * Returns the lowest sched_domain of a cpu which contains the given flag. + */ +static inline struct sched_domain *lowest_flag_domain(int cpu, int flag) +{ + struct sched_domain *sd; + + for_each_domain(cpu, sd) + if (sd && (sd->flags & flag)) + break; + + return sd; +} + +/** + * for_each_flag_domain - Iterates over sched_domains containing the flag. + * @cpu: The cpu whose domains we're iterating over. + * @sd: variable holding the value of the power_savings_sd + * for cpu. + * @flag: The flag to filter the sched_domains to be iterated. + * + * Iterates over all the scheduler domains for a given cpu that has the 'flag' + * set, starting from the lowest sched_domain to the highest. + */ +#define for_each_flag_domain(cpu, sd, flag) \ + for (sd = lowest_flag_domain(cpu, flag); \ + (sd && (sd->flags & flag)); sd = sd->parent) + +/** + * is_semi_idle_group - Checks if the given sched_group is semi-idle. + * @ilb_group: group to be checked for semi-idleness + * + * Returns: 1 if the group is semi-idle. 0 otherwise. + * + * We define a sched_group to be semi idle if it has atleast one idle-CPU + * and atleast one non-idle CPU. This helper function checks if the given + * sched_group is semi-idle or not. + */ +static inline int is_semi_idle_group(struct sched_group *ilb_group) +{ + cpumask_and(nohz.ilb_grp_nohz_mask, nohz.cpu_mask, + sched_group_cpus(ilb_group)); + + /* + * A sched_group is semi-idle when it has atleast one busy cpu + * and atleast one idle cpu. + */ + if (cpumask_empty(nohz.ilb_grp_nohz_mask)) + return 0; + + if (cpumask_equal(nohz.ilb_grp_nohz_mask, sched_group_cpus(ilb_group))) + return 0; + + return 1; +} +/** + * find_new_ilb - Finds the optimum idle load balancer for nomination. + * @cpu: The cpu which is nominating a new idle_load_balancer. + * + * Returns: Returns the id of the idle load balancer if it exists, + * Else, returns >= nr_cpu_ids. + * + * This algorithm picks the idle load balancer such that it belongs to a + * semi-idle powersavings sched_domain. The idea is to try and avoid + * completely idle packages/cores just for the purpose of idle load balancing + * when there are other idle cpu's which are better suited for that job. + */ +static int find_new_ilb(int cpu) +{ + struct sched_domain *sd; + struct sched_group *ilb_group; + + /* + * Have idle load balancer selection from semi-idle packages only + * when power-aware load balancing is enabled + */ + if (!(sched_smt_power_savings || sched_mc_power_savings)) + goto out_done; + + /* + * Optimize for the case when we have no idle CPUs or only one + * idle CPU. Don't walk the sched_domain hierarchy in such cases + */ + if (cpumask_weight(nohz.cpu_mask) < 2) + goto out_done; + + for_each_flag_domain(cpu, sd, SD_POWERSAVINGS_BALANCE) { + ilb_group = sd->groups; + + do { + if (is_semi_idle_group(ilb_group)) + return cpumask_first(nohz.ilb_grp_nohz_mask); + + ilb_group = ilb_group->next; + + } while (ilb_group != sd->groups); + } + +out_done: + return cpumask_first(nohz.cpu_mask); +} +#else /* (CONFIG_SCHED_MC || CONFIG_SCHED_SMT) */ +static inline int find_new_ilb(int call_cpu) +{ + return cpumask_first(nohz.cpu_mask); +} +#endif + /* * This routine will try to nominate the ilb (idle load balancing) * owner among the cpus whose ticks are stopped. ilb owner will do the idle @@ -4279,8 +4704,24 @@ int select_nohz_load_balancer(int stop_tick) /* make me the ilb owner */ if (atomic_cmpxchg(&nohz.load_balancer, -1, cpu) == -1) return 1; - } else if (atomic_read(&nohz.load_balancer) == cpu) + } else if (atomic_read(&nohz.load_balancer) == cpu) { + int new_ilb; + + if (!(sched_smt_power_savings || + sched_mc_power_savings)) + return 1; + /* + * Check to see if there is a more power-efficient + * ilb. + */ + new_ilb = find_new_ilb(cpu); + if (new_ilb < nr_cpu_ids && new_ilb != cpu) { + atomic_set(&nohz.load_balancer, -1); + resched_cpu(new_ilb); + return 0; + } return 1; + } } else { if (!cpumask_test_cpu(cpu, nohz.cpu_mask)) return 0; @@ -4313,11 +4754,6 @@ static void rebalance_domains(int cpu, enum cpu_idle_type idle) unsigned long next_balance = jiffies + 60*HZ; int update_next_balance = 0; int need_serialize; - cpumask_var_t tmp; - - /* Fails alloc? Rebalancing probably not a priority right now. */ - if (!alloc_cpumask_var(&tmp, GFP_ATOMIC)) - return; for_each_domain(cpu, sd) { if (!(sd->flags & SD_LOAD_BALANCE)) @@ -4342,7 +4778,7 @@ static void rebalance_domains(int cpu, enum cpu_idle_type idle) } if (time_after_eq(jiffies, sd->last_balance + interval)) { - if (load_balance(cpu, rq, sd, idle, &balance, tmp)) { + if (load_balance(cpu, rq, sd, idle, &balance)) { /* * We've pulled tasks over so either we're no * longer idle, or one of our SMT siblings is @@ -4376,8 +4812,6 @@ out: */ if (likely(update_next_balance)) rq->next_balance = next_balance; - - free_cpumask_var(tmp); } /* @@ -4456,15 +4890,7 @@ static inline void trigger_load_balance(struct rq *rq, int cpu) } if (atomic_read(&nohz.load_balancer) == -1) { - /* - * simple selection for now: Nominate the - * first cpu in the nohz list to be the next - * ilb owner. - * - * TBD: Traverse the sched domains and nominate - * the nearest cpu in the nohz.cpu_mask. - */ - int ilb = cpumask_first(nohz.cpu_mask); + int ilb = find_new_ilb(cpu); if (ilb < nr_cpu_ids) resched_cpu(ilb); @@ -4511,9 +4937,25 @@ DEFINE_PER_CPU(struct kernel_stat, kstat); EXPORT_PER_CPU_SYMBOL(kstat); /* - * Return any ns on the sched_clock that have not yet been banked in + * Return any ns on the sched_clock that have not yet been accounted in * @p in case that task is currently running. + * + * Called with task_rq_lock() held on @rq. */ +static u64 do_task_delta_exec(struct task_struct *p, struct rq *rq) +{ + u64 ns = 0; + + if (task_current(rq, p)) { + update_rq_clock(rq); + ns = rq->clock - p->se.exec_start; + if ((s64)ns < 0) + ns = 0; + } + + return ns; +} + unsigned long long task_delta_exec(struct task_struct *p) { unsigned long flags; @@ -4521,16 +4963,49 @@ unsigned long long task_delta_exec(struct task_struct *p) u64 ns = 0; rq = task_rq_lock(p, &flags); + ns = do_task_delta_exec(p, rq); + task_rq_unlock(rq, &flags); - if (task_current(rq, p)) { - u64 delta_exec; + return ns; +} - update_rq_clock(rq); - delta_exec = rq->clock - p->se.exec_start; - if ((s64)delta_exec > 0) - ns = delta_exec; - } +/* + * Return accounted runtime for the task. + * In case the task is currently running, return the runtime plus current's + * pending runtime that have not been accounted yet. + */ +unsigned long long task_sched_runtime(struct task_struct *p) +{ + unsigned long flags; + struct rq *rq; + u64 ns = 0; + + rq = task_rq_lock(p, &flags); + ns = p->se.sum_exec_runtime + do_task_delta_exec(p, rq); + task_rq_unlock(rq, &flags); + + return ns; +} + +/* + * Return sum_exec_runtime for the thread group. + * In case the task is currently running, return the sum plus current's + * pending runtime that have not been accounted yet. + * + * Note that the thread group might have other running tasks as well, + * so the return value not includes other pending runtime that other + * running tasks might have. + */ +unsigned long long thread_group_sched_runtime(struct task_struct *p) +{ + struct task_cputime totals; + unsigned long flags; + struct rq *rq; + u64 ns; + rq = task_rq_lock(p, &flags); + thread_group_cputime(p, &totals); + ns = totals.sum_exec_runtime + do_task_delta_exec(p, rq); task_rq_unlock(rq, &flags); return ns; @@ -4559,6 +5034,8 @@ void account_user_time(struct task_struct *p, cputime_t cputime, cpustat->nice = cputime64_add(cpustat->nice, tmp); else cpustat->user = cputime64_add(cpustat->user, tmp); + + cpuacct_update_stats(p, CPUACCT_STAT_USER, cputime); /* Account for user time used */ acct_update_integrals(p); } @@ -4584,8 +5061,13 @@ static void account_guest_time(struct task_struct *p, cputime_t cputime, p->gtime = cputime_add(p->gtime, cputime); /* Add guest time to cpustat. */ - cpustat->user = cputime64_add(cpustat->user, tmp); - cpustat->guest = cputime64_add(cpustat->guest, tmp); + if (TASK_NICE(p) > 0) { + cpustat->nice = cputime64_add(cpustat->nice, tmp); + cpustat->guest_nice = cputime64_add(cpustat->guest_nice, tmp); + } else { + cpustat->user = cputime64_add(cpustat->user, tmp); + cpustat->guest = cputime64_add(cpustat->guest, tmp); + } } /* @@ -4620,6 +5102,8 @@ void account_system_time(struct task_struct *p, int hardirq_offset, else cpustat->system = cputime64_add(cpustat->system, tmp); + cpuacct_update_stats(p, CPUACCT_STAT_SYSTEM, cputime); + /* Account for system time used */ acct_update_integrals(p); } @@ -4661,17 +5145,16 @@ void account_idle_time(cputime_t cputime) */ void account_process_tick(struct task_struct *p, int user_tick) { - cputime_t one_jiffy = jiffies_to_cputime(1); - cputime_t one_jiffy_scaled = cputime_to_scaled(one_jiffy); + cputime_t one_jiffy_scaled = cputime_to_scaled(cputime_one_jiffy); struct rq *rq = this_rq(); if (user_tick) - account_user_time(p, one_jiffy, one_jiffy_scaled); - else if (p != rq->idle) - account_system_time(p, HARDIRQ_OFFSET, one_jiffy, + account_user_time(p, cputime_one_jiffy, one_jiffy_scaled); + else if ((p != rq->idle) || (irq_count() != HARDIRQ_OFFSET)) + account_system_time(p, HARDIRQ_OFFSET, cputime_one_jiffy, one_jiffy_scaled); else - account_idle_time(one_jiffy); + account_idle_time(cputime_one_jiffy); } /* @@ -4699,60 +5182,86 @@ void account_idle_ticks(unsigned long ticks) * Use precise platform statistics if available: */ #ifdef CONFIG_VIRT_CPU_ACCOUNTING -cputime_t task_utime(struct task_struct *p) +void task_times(struct task_struct *p, cputime_t *ut, cputime_t *st) { - return p->utime; + *ut = p->utime; + *st = p->stime; } -cputime_t task_stime(struct task_struct *p) +void thread_group_times(struct task_struct *p, cputime_t *ut, cputime_t *st) { - return p->stime; + struct task_cputime cputime; + + thread_group_cputime(p, &cputime); + + *ut = cputime.utime; + *st = cputime.stime; } #else -cputime_t task_utime(struct task_struct *p) + +#ifndef nsecs_to_cputime +# define nsecs_to_cputime(__nsecs) nsecs_to_jiffies(__nsecs) +#endif + +void task_times(struct task_struct *p, cputime_t *ut, cputime_t *st) { - clock_t utime = cputime_to_clock_t(p->utime), - total = utime + cputime_to_clock_t(p->stime); - u64 temp; + cputime_t rtime, utime = p->utime, total = cputime_add(utime, p->stime); /* * Use CFS's precise accounting: */ - temp = (u64)nsec_to_clock_t(p->se.sum_exec_runtime); + rtime = nsecs_to_cputime(p->se.sum_exec_runtime); if (total) { - temp *= utime; + u64 temp; + + temp = (u64)(rtime * utime); do_div(temp, total); - } - utime = (clock_t)temp; + utime = (cputime_t)temp; + } else + utime = rtime; + + /* + * Compare with previous values, to keep monotonicity: + */ + p->prev_utime = max(p->prev_utime, utime); + p->prev_stime = max(p->prev_stime, cputime_sub(rtime, p->prev_utime)); - p->prev_utime = max(p->prev_utime, clock_t_to_cputime(utime)); - return p->prev_utime; + *ut = p->prev_utime; + *st = p->prev_stime; } -cputime_t task_stime(struct task_struct *p) +/* + * Must be called with siglock held. + */ +void thread_group_times(struct task_struct *p, cputime_t *ut, cputime_t *st) { - clock_t stime; + struct signal_struct *sig = p->signal; + struct task_cputime cputime; + cputime_t rtime, utime, total; - /* - * Use CFS's precise accounting. (we subtract utime from - * the total, to make sure the total observed by userspace - * grows monotonically - apps rely on that): - */ - stime = nsec_to_clock_t(p->se.sum_exec_runtime) - - cputime_to_clock_t(task_utime(p)); + thread_group_cputime(p, &cputime); - if (stime >= 0) - p->prev_stime = max(p->prev_stime, clock_t_to_cputime(stime)); + total = cputime_add(cputime.utime, cputime.stime); + rtime = nsecs_to_cputime(cputime.sum_exec_runtime); - return p->prev_stime; -} -#endif + if (total) { + u64 temp; -inline cputime_t task_gtime(struct task_struct *p) -{ - return p->gtime; + temp = (u64)(rtime * cputime.utime); + do_div(temp, total); + utime = (cputime_t)temp; + } else + utime = rtime; + + sig->prev_utime = max(sig->prev_utime, utime); + sig->prev_stime = max(sig->prev_stime, + cputime_sub(rtime, sig->prev_utime)); + + *ut = sig->prev_utime; + *st = sig->prev_stime; } +#endif /* * This function gets called by the timer code, with HZ frequency. @@ -4775,16 +5284,15 @@ void scheduler_tick(void) curr->sched_class->task_tick(rq, curr, 0); spin_unlock(&rq->lock); + perf_event_task_tick(curr, cpu); + #ifdef CONFIG_SMP rq->idle_at_tick = idle_cpu(cpu); trigger_load_balance(rq, cpu); #endif } -#if defined(CONFIG_PREEMPT) && (defined(CONFIG_DEBUG_PREEMPT) || \ - defined(CONFIG_PREEMPT_TRACER)) - -static inline unsigned long get_parent_ip(unsigned long addr) +notrace unsigned long get_parent_ip(unsigned long addr) { if (in_lock_functions(addr)) { addr = CALLER_ADDR2; @@ -4794,6 +5302,9 @@ static inline unsigned long get_parent_ip(unsigned long addr) return addr; } +#if defined(CONFIG_PREEMPT) && (defined(CONFIG_DEBUG_PREEMPT) || \ + defined(CONFIG_PREEMPT_TRACER)) + void __kprobes add_preempt_count(int val) { #ifdef CONFIG_DEBUG_PREEMPT @@ -4885,14 +5396,13 @@ static inline void schedule_debug(struct task_struct *prev) #endif } -static void put_prev_task(struct rq *rq, struct task_struct *prev) +static void put_prev_task(struct rq *rq, struct task_struct *p) { - if (prev->state == TASK_RUNNING) { - u64 runtime = prev->se.sum_exec_runtime; + u64 runtime = p->se.sum_exec_runtime - p->se.prev_sum_exec_runtime; - runtime -= prev->se.prev_sum_exec_runtime; - runtime = min_t(u64, runtime, 2*sysctl_sched_migration_cost); + update_avg(&p->se.avg_running, runtime); + if (p->state == TASK_RUNNING) { /* * In order to avoid avg_overlap growing stale when we are * indeed overlapping and hence not getting put to sleep, grow @@ -4902,9 +5412,12 @@ static void put_prev_task(struct rq *rq, struct task_struct *prev) * correlates to the amount of cache footprint a task can * build up. */ - update_avg(&prev->se.avg_overlap, runtime); + runtime = min_t(u64, runtime, 2*sysctl_sched_migration_cost); + update_avg(&p->se.avg_overlap, runtime); + } else { + update_avg(&p->se.avg_running, 0); } - prev->sched_class->put_prev_task(rq, prev); + p->sched_class->put_prev_task(rq, p); } /* @@ -4942,16 +5455,18 @@ pick_next_task(struct rq *rq) /* * schedule() is the main scheduler function. */ -asmlinkage void __sched __schedule(void) +asmlinkage void __sched schedule(void) { struct task_struct *prev, *next; unsigned long *switch_count; struct rq *rq; int cpu; +need_resched: + preempt_disable(); cpu = smp_processor_id(); rq = cpu_rq(cpu); - rcu_qsctr_inc(cpu); + rcu_sched_qs(cpu); prev = rq->curr; switch_count = &prev->nivcsw; @@ -4975,10 +5490,7 @@ need_resched_nonpreemptible: switch_count = &prev->nvcsw; } -#ifdef CONFIG_SMP - if (prev->sched_class->pre_schedule) - prev->sched_class->pre_schedule(rq, prev); -#endif + pre_schedule(rq, prev); if (unlikely(!rq->nr_running)) idle_balance(cpu, rq); @@ -4988,6 +5500,7 @@ need_resched_nonpreemptible: if (likely(prev != next)) { sched_info_switch(prev, next); + perf_event_task_sched_out(prev, next, cpu); rq->nr_switches++; rq->curr = next; @@ -5003,22 +5516,18 @@ need_resched_nonpreemptible: } else spin_unlock_irq(&rq->lock); + post_schedule(rq); + if (unlikely(reacquire_kernel_lock(current) < 0)) goto need_resched_nonpreemptible; -} -asmlinkage void __sched schedule(void) -{ -need_resched: - preempt_disable(); - __schedule(); preempt_enable_no_resched(); - if (unlikely(test_thread_flag(TIF_NEED_RESCHED))) + if (need_resched()) goto need_resched; } EXPORT_SYMBOL(schedule); -#ifdef CONFIG_SMP +#ifdef CONFIG_MUTEX_SPIN_ON_OWNER /* * Look out! "owner" is an entirely speculative pointer * access and not reliable. @@ -5140,10 +5649,10 @@ asmlinkage void __sched preempt_schedule_irq(void) #endif /* CONFIG_PREEMPT */ -int default_wake_function(wait_queue_t *curr, unsigned mode, int sync, +int default_wake_function(wait_queue_t *curr, unsigned mode, int wake_flags, void *key) { - return try_to_wake_up(curr->private, mode, sync); + return try_to_wake_up(curr->private, mode, wake_flags); } EXPORT_SYMBOL(default_wake_function); @@ -5156,15 +5665,15 @@ EXPORT_SYMBOL(default_wake_function); * started to run but is not in state TASK_RUNNING. try_to_wake_up() returns * zero in this (rare) case, and we handle it by continuing to scan the queue. */ -void __wake_up_common(wait_queue_head_t *q, unsigned int mode, - int nr_exclusive, int sync, void *key) +static void __wake_up_common(wait_queue_head_t *q, unsigned int mode, + int nr_exclusive, int wake_flags, void *key) { wait_queue_t *curr, *next; list_for_each_entry_safe(curr, next, &q->task_list, task_list) { unsigned flags = curr->flags; - if (curr->func(curr, mode, sync, key) && + if (curr->func(curr, mode, wake_flags, key) && (flags & WQ_FLAG_EXCLUSIVE) && !--nr_exclusive) break; } @@ -5176,6 +5685,9 @@ void __wake_up_common(wait_queue_head_t *q, unsigned int mode, * @mode: which threads * @nr_exclusive: how many wake-one or wake-many threads to wake up * @key: is directly passed to the wakeup function + * + * It may be assumed that this function implies a write memory barrier before + * changing the task state if and only if any tasks are woken up. */ void __wake_up(wait_queue_head_t *q, unsigned int mode, int nr_exclusive, void *key) @@ -5214,21 +5726,24 @@ void __wake_up_locked_key(wait_queue_head_t *q, unsigned int mode, void *key) * with each other. This can prevent needless bouncing between CPUs. * * On UP it can prevent extra preemption. + * + * It may be assumed that this function implies a write memory barrier before + * changing the task state if and only if any tasks are woken up. */ void __wake_up_sync_key(wait_queue_head_t *q, unsigned int mode, int nr_exclusive, void *key) { unsigned long flags; - int sync = 1; + int wake_flags = WF_SYNC; if (unlikely(!q)) return; if (unlikely(!nr_exclusive)) - sync = 0; + wake_flags = 0; spin_lock_irqsave(&q->lock, flags); - __wake_up_common(q, mode, nr_exclusive, sync, key); + __wake_up_common(q, mode, nr_exclusive, wake_flags, key); spin_unlock_irqrestore(&q->lock, flags); } EXPORT_SYMBOL_GPL(__wake_up_sync_key); @@ -5250,6 +5765,9 @@ EXPORT_SYMBOL_GPL(__wake_up_sync); /* For internal use only */ * awakened in the same order in which they were queued. * * See also complete_all(), wait_for_completion() and related routines. + * + * It may be assumed that this function implies a write memory barrier before + * changing the task state if and only if any tasks are woken up. */ void complete(struct completion *x) { @@ -5267,6 +5785,9 @@ EXPORT_SYMBOL(complete); * @x: holds the state of this particular completion * * This will wake up all threads waiting on this particular completion event. + * + * It may be assumed that this function implies a write memory barrier before + * changing the task state if and only if any tasks are woken up. */ void complete_all(struct completion *x) { @@ -5700,22 +6221,14 @@ __setscheduler(struct rq *rq, struct task_struct *p, int policy, int prio) BUG_ON(p->se.on_rq); p->policy = policy; - switch (p->policy) { - case SCHED_NORMAL: - case SCHED_BATCH: - case SCHED_IDLE: - p->sched_class = &fair_sched_class; - break; - case SCHED_FIFO: - case SCHED_RR: - p->sched_class = &rt_sched_class; - break; - } - p->rt_priority = prio; p->normal_prio = normal_prio(p); /* we are holding p->pi_lock already */ p->prio = rt_mutex_getprio(p); + if (rt_prio(p->prio)) + p->sched_class = &rt_sched_class; + else + p->sched_class = &fair_sched_class; set_load_weight(p); } @@ -5742,17 +6255,25 @@ static int __sched_setscheduler(struct task_struct *p, int policy, unsigned long flags; const struct sched_class *prev_class = p->sched_class; struct rq *rq; + int reset_on_fork; /* may grab non-irq protected spin_locks */ BUG_ON(in_interrupt()); recheck: /* double check policy once rq lock held */ - if (policy < 0) + if (policy < 0) { + reset_on_fork = p->sched_reset_on_fork; policy = oldpolicy = p->policy; - else if (policy != SCHED_FIFO && policy != SCHED_RR && - policy != SCHED_NORMAL && policy != SCHED_BATCH && - policy != SCHED_IDLE) - return -EINVAL; + } else { + reset_on_fork = !!(policy & SCHED_RESET_ON_FORK); + policy &= ~SCHED_RESET_ON_FORK; + + if (policy != SCHED_FIFO && policy != SCHED_RR && + policy != SCHED_NORMAL && policy != SCHED_BATCH && + policy != SCHED_IDLE) + return -EINVAL; + } + /* * Valid priorities for SCHED_FIFO and SCHED_RR are * 1..MAX_USER_RT_PRIO-1, valid priority for SCHED_NORMAL, @@ -5796,6 +6317,10 @@ recheck: /* can't change other user's priorities */ if (!check_same_owner(p)) return -EPERM; + + /* Normal users shall not reset the sched_reset_on_fork flag */ + if (p->sched_reset_on_fork && !reset_on_fork) + return -EPERM; } if (user) { @@ -5839,6 +6364,8 @@ recheck: if (running) p->sched_class->put_prev_task(rq, p); + p->sched_reset_on_fork = reset_on_fork; + oldprio = p->prio; __setscheduler(rq, p, policy, param->sched_priority); @@ -5955,14 +6482,15 @@ SYSCALL_DEFINE1(sched_getscheduler, pid_t, pid) if (p) { retval = security_task_getscheduler(p); if (!retval) - retval = p->policy; + retval = p->policy + | (p->sched_reset_on_fork ? SCHED_RESET_ON_FORK : 0); } read_unlock(&tasklist_lock); return retval; } /** - * sys_sched_getscheduler - get the RT priority of a thread + * sys_sched_getparam - get the RT priority of a thread * @pid: the pid in question. * @param: structure containing the RT priority. */ @@ -6183,27 +6711,21 @@ SYSCALL_DEFINE0(sched_yield) return 0; } +static inline int should_resched(void) +{ + return need_resched() && !(preempt_count() & PREEMPT_ACTIVE); +} + static void __cond_resched(void) { -#ifdef CONFIG_DEBUG_SPINLOCK_SLEEP - __might_sleep(__FILE__, __LINE__); -#endif - /* - * The BKS might be reacquired before we have dropped - * PREEMPT_ACTIVE, which could trigger a second - * cond_resched() call. - */ - do { - add_preempt_count(PREEMPT_ACTIVE); - schedule(); - sub_preempt_count(PREEMPT_ACTIVE); - } while (need_resched()); + add_preempt_count(PREEMPT_ACTIVE); + schedule(); + sub_preempt_count(PREEMPT_ACTIVE); } int __sched _cond_resched(void) { - if (need_resched() && !(preempt_count() & PREEMPT_ACTIVE) && - system_state == SYSTEM_RUNNING) { + if (should_resched()) { __cond_resched(); return 1; } @@ -6212,21 +6734,23 @@ int __sched _cond_resched(void) EXPORT_SYMBOL(_cond_resched); /* - * cond_resched_lock() - if a reschedule is pending, drop the given lock, + * __cond_resched_lock() - if a reschedule is pending, drop the given lock, * call schedule, and on return reacquire the lock. * * This works OK both with and without CONFIG_PREEMPT. We do strange low-level * operations here to prevent schedule() from being called twice (once via * spin_unlock(), once by hand). */ -int cond_resched_lock(spinlock_t *lock) +int __cond_resched_lock(spinlock_t *lock) { - int resched = need_resched() && system_state == SYSTEM_RUNNING; + int resched = should_resched(); int ret = 0; + lockdep_assert_held(lock); + if (spin_needbreak(lock) || resched) { spin_unlock(lock); - if (resched && need_resched()) + if (resched) __cond_resched(); else cpu_relax(); @@ -6235,13 +6759,13 @@ int cond_resched_lock(spinlock_t *lock) } return ret; } -EXPORT_SYMBOL(cond_resched_lock); +EXPORT_SYMBOL(__cond_resched_lock); -int __sched cond_resched_softirq(void) +int __sched __cond_resched_softirq(void) { BUG_ON(!in_softirq()); - if (need_resched() && system_state == SYSTEM_RUNNING) { + if (should_resched()) { local_bh_enable(); __cond_resched(); local_bh_disable(); @@ -6249,7 +6773,7 @@ int __sched cond_resched_softirq(void) } return 0; } -EXPORT_SYMBOL(cond_resched_softirq); +EXPORT_SYMBOL(__cond_resched_softirq); /** * yield - yield the current processor to other threads. @@ -6267,17 +6791,16 @@ EXPORT_SYMBOL(yield); /* * This task is about to go to sleep on IO. Increment rq->nr_iowait so * that process accounting knows that this is a task in IO wait state. - * - * But don't do that if it is a deliberate, throttling IO wait (this task - * has set its backing_dev_info: the queue against which it should throttle) */ void __sched io_schedule(void) { - struct rq *rq = &__raw_get_cpu_var(runqueues); + struct rq *rq = raw_rq(); delayacct_blkio_start(); atomic_inc(&rq->nr_iowait); + current->in_iowait = 1; schedule(); + current->in_iowait = 0; atomic_dec(&rq->nr_iowait); delayacct_blkio_end(); } @@ -6285,12 +6808,14 @@ EXPORT_SYMBOL(io_schedule); long __sched io_schedule_timeout(long timeout) { - struct rq *rq = &__raw_get_cpu_var(runqueues); + struct rq *rq = raw_rq(); long ret; delayacct_blkio_start(); atomic_inc(&rq->nr_iowait); + current->in_iowait = 1; ret = schedule_timeout(timeout); + current->in_iowait = 0; atomic_dec(&rq->nr_iowait); delayacct_blkio_end(); return ret; @@ -6374,23 +6899,8 @@ SYSCALL_DEFINE2(sched_rr_get_interval, pid_t, pid, if (retval) goto out_unlock; - /* - * Time slice is 0 for SCHED_FIFO tasks and for SCHED_OTHER - * tasks that are on an otherwise idle runqueue: - */ - time_slice = 0; - if (p->policy == SCHED_RR) { - time_slice = DEF_TIMESLICE; - } else if (p->policy != SCHED_FIFO) { - struct sched_entity *se = &p->se; - unsigned long flags; - struct rq *rq; + time_slice = p->sched_class->get_rr_interval(p); - rq = task_rq_lock(p, &flags); - if (rq->cfs.load.weight) - time_slice = NS_TO_JIFFIES(sched_slice(&rq->cfs, se)); - task_rq_unlock(rq, &flags); - } read_unlock(&tasklist_lock); jiffies_to_timespec(time_slice, &t); retval = copy_to_user(interval, &t, sizeof(t)) ? -EFAULT : 0; @@ -6425,8 +6935,9 @@ void sched_show_task(struct task_struct *p) #ifdef CONFIG_DEBUG_STACK_USAGE free = stack_not_used(p); #endif - printk(KERN_CONT "%5lu %5d %6d\n", free, - task_pid_nr(p), task_pid_nr(p->real_parent)); + printk(KERN_CONT "%5lu %5d %6d 0x%08lx\n", free, + task_pid_nr(p), task_pid_nr(p->real_parent), + (unsigned long)task_thread_info(p)->flags); show_stack(p, NULL); } @@ -6462,7 +6973,7 @@ void show_state_filter(unsigned long state_filter) /* * Only show locks if all tasks are dumped: */ - if (state_filter == -1) + if (!state_filter) debug_show_all_locks(); } @@ -6606,8 +7117,12 @@ int set_cpus_allowed_ptr(struct task_struct *p, const struct cpumask *new_mask) if (migrate_task(p, cpumask_any_and(cpu_online_mask, new_mask), &req)) { /* Need help from migration thread: drop lock and wait. */ + struct task_struct *mt = rq->migration_thread; + + get_task_struct(mt); task_rq_unlock(rq, &flags); wake_up_process(rq->migration_thread); + put_task_struct(mt); wait_for_completion(&req.done); tlb_migrate_finish(p->mm); return 0; @@ -6665,6 +7180,11 @@ fail: return ret; } +#define RCU_MIGRATION_IDLE 0 +#define RCU_MIGRATION_NEED_QS 1 +#define RCU_MIGRATION_GOT_QS 2 +#define RCU_MIGRATION_MUST_SYNC 3 + /* * migration_thread - this is a highprio system thread that performs * thread migration by bumping thread off CPU then 'pushing' onto @@ -6672,6 +7192,7 @@ fail: */ static int migration_thread(void *data) { + int badcpu; int cpu = (long)data; struct rq *rq; @@ -6687,7 +7208,7 @@ static int migration_thread(void *data) if (cpu_is_offline(cpu)) { spin_unlock_irq(&rq->lock); - goto wait_to_die; + break; } if (rq->active_balance) { @@ -6706,23 +7227,23 @@ static int migration_thread(void *data) req = list_entry(head->next, struct migration_req, list); list_del_init(head->next); - spin_unlock(&rq->lock); - __migrate_task(req->task, cpu, req->dest_cpu); + if (req->task != NULL) { + spin_unlock(&rq->lock); + __migrate_task(req->task, cpu, req->dest_cpu); + } else if (likely(cpu == (badcpu = smp_processor_id()))) { + req->dest_cpu = RCU_MIGRATION_GOT_QS; + spin_unlock(&rq->lock); + } else { + req->dest_cpu = RCU_MIGRATION_MUST_SYNC; + spin_unlock(&rq->lock); + WARN_ONCE(1, "migration_thread() on CPU %d, expected %d\n", badcpu, cpu); + } local_irq_enable(); complete(&req->done); } __set_current_state(TASK_RUNNING); - return 0; -wait_to_die: - /* Wait for kthread_stop */ - set_current_state(TASK_INTERRUPTIBLE); - while (!kthread_should_stop()) { - schedule(); - set_current_state(TASK_INTERRUPTIBLE); - } - __set_current_state(TASK_RUNNING); return 0; } @@ -6905,6 +7426,15 @@ static void migrate_dead_tasks(unsigned int dead_cpu) } } + +/* + * remove the tasks which were accounted by rq from calc_load_tasks. + */ +static void calc_global_load_remove(struct rq *rq) +{ + atomic_long_sub(rq->calc_load_active, &calc_load_tasks); + rq->calc_load_active = 0; +} #endif /* CONFIG_HOTPLUG_CPU */ #if defined(CONFIG_SCHED_DEBUG) && defined(CONFIG_SYSCTL) @@ -6914,17 +7444,16 @@ static struct ctl_table sd_ctl_dir[] = { .procname = "sched_domain", .mode = 0555, }, - {0, }, + {} }; static struct ctl_table sd_ctl_root[] = { { - .ctl_name = CTL_KERN, .procname = "kernel", .mode = 0555, .child = sd_ctl_dir, }, - {0, }, + {} }; static struct ctl_table *sd_alloc_ctl_entry(int n) @@ -7128,7 +7657,9 @@ migration_call(struct notifier_block *nfb, unsigned long action, void *hcpu) rq = task_rq_lock(p, &flags); __setscheduler(rq, p, SCHED_FIFO, MAX_RT_PRIO-1); task_rq_unlock(rq, &flags); + get_task_struct(p); cpu_rq(cpu)->migration_thread = p; + rq->calc_load_update = calc_load_update; break; case CPU_ONLINE: @@ -7156,6 +7687,7 @@ migration_call(struct notifier_block *nfb, unsigned long action, void *hcpu) kthread_bind(cpu_rq(cpu)->migration_thread, cpumask_any(cpu_online_mask)); kthread_stop(cpu_rq(cpu)->migration_thread); + put_task_struct(cpu_rq(cpu)->migration_thread); cpu_rq(cpu)->migration_thread = NULL; break; @@ -7165,6 +7697,7 @@ migration_call(struct notifier_block *nfb, unsigned long action, void *hcpu) migrate_live_tasks(cpu); rq = cpu_rq(cpu); kthread_stop(rq->migration_thread); + put_task_struct(rq->migration_thread); rq->migration_thread = NULL; /* Idle task back to normal (off runqueue, low prio) */ spin_lock_irq(&rq->lock); @@ -7178,7 +7711,7 @@ migration_call(struct notifier_block *nfb, unsigned long action, void *hcpu) cpuset_unlock(); migrate_nr_uninterruptible(rq); BUG_ON(rq->nr_running != 0); - + calc_global_load_remove(rq); /* * No need to migrate the tasks: it was best-effort if * they didn't take sched_hotcpu_mutex. Just wake up @@ -7214,8 +7747,10 @@ migration_call(struct notifier_block *nfb, unsigned long action, void *hcpu) return NOTIFY_OK; } -/* Register at highest priority so that task migration (migrate_all_tasks) - * happens before everything else. +/* + * Register at high priority so that task migration (migrate_all_tasks) + * happens before everything else. This has to be lower priority than + * the notifier in the perf_event subsystem, though. */ static struct notifier_block __cpuinitdata migration_notifier = { .notifier_call = migration_call, @@ -7233,7 +7768,7 @@ static int __init migration_init(void) migration_call(&migration_notifier, CPU_ONLINE, cpu); register_cpu_notifier(&migration_notifier); - return err; + return 0; } early_initcall(migration_init); #endif @@ -7242,6 +7777,16 @@ early_initcall(migration_init); #ifdef CONFIG_SCHED_DEBUG +static __read_mostly int sched_domain_debug_enabled; + +static int __init sched_domain_debug_setup(char *str) +{ + sched_domain_debug_enabled = 1; + + return 0; +} +early_param("sched_debug", sched_domain_debug_setup); + static int sched_domain_debug_one(struct sched_domain *sd, int cpu, int level, struct cpumask *groupmask) { @@ -7280,7 +7825,7 @@ static int sched_domain_debug_one(struct sched_domain *sd, int cpu, int level, break; } - if (!group->__cpu_power) { + if (!group->cpu_power) { printk(KERN_CONT "\n"); printk(KERN_ERR "ERROR: domain->cpu_power not " "set\n"); @@ -7302,7 +7847,12 @@ static int sched_domain_debug_one(struct sched_domain *sd, int cpu, int level, cpumask_or(groupmask, groupmask, sched_group_cpus(group)); cpulist_scnprintf(str, sizeof(str), sched_group_cpus(group)); + printk(KERN_CONT " %s", str); + if (group->cpu_power != SCHED_LOAD_SCALE) { + printk(KERN_CONT " (cpu_power = %d)", + group->cpu_power); + } group = group->next; } while (group != sd->groups); @@ -7323,6 +7873,9 @@ static void sched_domain_debug(struct sched_domain *sd, int cpu) cpumask_var_t groupmask; int level = 0; + if (!sched_domain_debug_enabled) + return; + if (!sd) { printk(KERN_DEBUG "CPU%d attaching NULL sched-domain.\n", cpu); return; @@ -7366,9 +7919,7 @@ static int sd_degenerate(struct sched_domain *sd) } /* Following flags don't use groups */ - if (sd->flags & (SD_WAKE_IDLE | - SD_WAKE_AFFINE | - SD_WAKE_BALANCE)) + if (sd->flags & (SD_WAKE_AFFINE)) return 0; return 1; @@ -7385,10 +7936,6 @@ sd_parent_degenerate(struct sched_domain *sd, struct sched_domain *parent) if (!cpumask_equal(sched_domain_span(sd), sched_domain_span(parent))) return 0; - /* Does parent contain flags not in child? */ - /* WAKE_BALANCE is a subset of WAKE_AFFINE */ - if (cflags & SD_WAKE_AFFINE) - pflags &= ~SD_WAKE_BALANCE; /* Flags needing groups don't count if only 1 group in parent */ if (parent->groups == parent->groups->next) { pflags &= ~(SD_LOAD_BALANCE | @@ -7408,6 +7955,8 @@ sd_parent_degenerate(struct sched_domain *sd, struct sched_domain *parent) static void free_rootdomain(struct root_domain *rd) { + synchronize_sched(); + cpupri_cleanup(&rd->cpupri); free_cpumask_var(rd->rto_mask); @@ -7444,7 +7993,7 @@ static void rq_attach_root(struct rq *rq, struct root_domain *rd) rq->rd = rd; cpumask_set_cpu(rq->cpu, rd->span); - if (cpumask_test_cpu(rq->cpu, cpu_online_mask)) + if (cpumask_test_cpu(rq->cpu, cpu_active_mask)) set_rq_online(rq); spin_unlock_irqrestore(&rq->lock, flags); @@ -7453,26 +8002,23 @@ static void rq_attach_root(struct rq *rq, struct root_domain *rd) free_rootdomain(old_rd); } -static int __init_refok init_rootdomain(struct root_domain *rd, bool bootmem) +static int init_rootdomain(struct root_domain *rd, bool bootmem) { + gfp_t gfp = GFP_KERNEL; + memset(rd, 0, sizeof(*rd)); - if (bootmem) { - alloc_bootmem_cpumask_var(&def_root_domain.span); - alloc_bootmem_cpumask_var(&def_root_domain.online); - alloc_bootmem_cpumask_var(&def_root_domain.rto_mask); - cpupri_init(&rd->cpupri, true); - return 0; - } + if (bootmem) + gfp = GFP_NOWAIT; - if (!alloc_cpumask_var(&rd->span, GFP_KERNEL)) + if (!alloc_cpumask_var(&rd->span, gfp)) goto out; - if (!alloc_cpumask_var(&rd->online, GFP_KERNEL)) + if (!alloc_cpumask_var(&rd->online, gfp)) goto free_span; - if (!alloc_cpumask_var(&rd->rto_mask, GFP_KERNEL)) + if (!alloc_cpumask_var(&rd->rto_mask, gfp)) goto free_online; - if (cpupri_init(&rd->cpupri, false) != 0) + if (cpupri_init(&rd->cpupri, bootmem) != 0) goto free_rto_mask; return 0; @@ -7551,6 +8097,7 @@ static cpumask_var_t cpu_isolated_map; /* Setup the mask of cpus configured for isolated domains */ static int __init isolated_cpu_setup(char *str) { + alloc_bootmem_cpumask_var(&cpu_isolated_map); cpulist_parse(str, cpu_isolated_map); return 1; } @@ -7589,7 +8136,7 @@ init_sched_build_groups(const struct cpumask *span, continue; cpumask_clear(sched_group_cpus(sg)); - sg->__cpu_power = 0; + sg->cpu_power = 0; for_each_cpu(j, span) { if (group_fn(j, cpu_map, NULL, tmpmask) != group) @@ -7683,8 +8230,9 @@ int sched_smt_power_savings = 0, sched_mc_power_savings = 0; /* * The cpus mask in sched_group and sched_domain hangs off the end. - * FIXME: use cpumask_var_t or dynamic percpu alloc to avoid wasting space - * for nr_cpu_ids < CONFIG_NR_CPUS. + * + * ( See the the comments in include/linux/sched.h:struct sched_group + * and struct sched_domain. ) */ struct static_sched_group { struct sched_group sg; @@ -7696,6 +8244,39 @@ struct static_sched_domain { DECLARE_BITMAP(span, CONFIG_NR_CPUS); }; +struct s_data { +#ifdef CONFIG_NUMA + int sd_allnodes; + cpumask_var_t domainspan; + cpumask_var_t covered; + cpumask_var_t notcovered; +#endif + cpumask_var_t nodemask; + cpumask_var_t this_sibling_map; + cpumask_var_t this_core_map; + cpumask_var_t send_covered; + cpumask_var_t tmpmask; + struct sched_group **sched_group_nodes; + struct root_domain *rd; +}; + +enum s_alloc { + sa_sched_groups = 0, + sa_rootdomain, + sa_tmpmask, + sa_send_covered, + sa_this_core_map, + sa_this_sibling_map, + sa_nodemask, + sa_sched_group_nodes, +#ifdef CONFIG_NUMA + sa_notcovered, + sa_covered, + sa_domainspan, +#endif + sa_none, +}; + /* * SMT sched-domains: */ @@ -7728,7 +8309,7 @@ cpu_to_core_group(int cpu, const struct cpumask *cpu_map, { int group; - cpumask_and(mask, &per_cpu(cpu_sibling_map, cpu), cpu_map); + cpumask_and(mask, topology_thread_cpumask(cpu), cpu_map); group = cpumask_first(mask); if (sg) *sg = &per_cpu(sched_group_core, group).sg; @@ -7757,7 +8338,7 @@ cpu_to_phys_group(int cpu, const struct cpumask *cpu_map, cpumask_and(mask, cpu_coregroup_mask(cpu), cpu_map); group = cpumask_first(mask); #elif defined(CONFIG_SCHED_SMT) - cpumask_and(mask, &per_cpu(cpu_sibling_map, cpu), cpu_map); + cpumask_and(mask, topology_thread_cpumask(cpu), cpu_map); group = cpumask_first(mask); #else group = cpu; @@ -7805,7 +8386,7 @@ static void init_numa_sched_groups_power(struct sched_group *group_head) struct sched_domain *sd; sd = &per_cpu(phys_domains, j).sd; - if (j != cpumask_first(sched_group_cpus(sd->groups))) { + if (j != group_first_cpu(sd->groups)) { /* * Only add "power" once for each * physical package. @@ -7813,11 +8394,76 @@ static void init_numa_sched_groups_power(struct sched_group *group_head) continue; } - sg_inc_cpu_power(sg, sd->groups->__cpu_power); + sg->cpu_power += sd->groups->cpu_power; } sg = sg->next; } while (sg != group_head); } + +static int build_numa_sched_groups(struct s_data *d, + const struct cpumask *cpu_map, int num) +{ + struct sched_domain *sd; + struct sched_group *sg, *prev; + int n, j; + + cpumask_clear(d->covered); + cpumask_and(d->nodemask, cpumask_of_node(num), cpu_map); + if (cpumask_empty(d->nodemask)) { + d->sched_group_nodes[num] = NULL; + goto out; + } + + sched_domain_node_span(num, d->domainspan); + cpumask_and(d->domainspan, d->domainspan, cpu_map); + + sg = kmalloc_node(sizeof(struct sched_group) + cpumask_size(), + GFP_KERNEL, num); + if (!sg) { + printk(KERN_WARNING "Can not alloc domain group for node %d\n", + num); + return -ENOMEM; + } + d->sched_group_nodes[num] = sg; + + for_each_cpu(j, d->nodemask) { + sd = &per_cpu(node_domains, j).sd; + sd->groups = sg; + } + + sg->cpu_power = 0; + cpumask_copy(sched_group_cpus(sg), d->nodemask); + sg->next = sg; + cpumask_or(d->covered, d->covered, d->nodemask); + + prev = sg; + for (j = 0; j < nr_node_ids; j++) { + n = (num + j) % nr_node_ids; + cpumask_complement(d->notcovered, d->covered); + cpumask_and(d->tmpmask, d->notcovered, cpu_map); + cpumask_and(d->tmpmask, d->tmpmask, d->domainspan); + if (cpumask_empty(d->tmpmask)) + break; + cpumask_and(d->tmpmask, d->tmpmask, cpumask_of_node(n)); + if (cpumask_empty(d->tmpmask)) + continue; + sg = kmalloc_node(sizeof(struct sched_group) + cpumask_size(), + GFP_KERNEL, num); + if (!sg) { + printk(KERN_WARNING + "Can not alloc domain group for node %d\n", j); + return -ENOMEM; + } + sg->cpu_power = 0; + cpumask_copy(sched_group_cpus(sg), d->tmpmask); + sg->next = prev->next; + cpumask_or(d->covered, d->covered, d->tmpmask); + prev->next = sg; + prev = sg; + } +out: + return 0; +} #endif /* CONFIG_NUMA */ #ifdef CONFIG_NUMA @@ -7871,45 +8517,47 @@ static void free_sched_groups(const struct cpumask *cpu_map, * there are asymmetries in the topology. If there are asymmetries, group * having more cpu_power will pickup more load compared to the group having * less cpu_power. - * - * cpu_power will be a multiple of SCHED_LOAD_SCALE. This multiple represents - * the maximum number of tasks a group can handle in the presence of other idle - * or lightly loaded groups in the same sched domain. */ static void init_sched_groups_power(int cpu, struct sched_domain *sd) { struct sched_domain *child; struct sched_group *group; + long power; + int weight; WARN_ON(!sd || !sd->groups); - if (cpu != cpumask_first(sched_group_cpus(sd->groups))) + if (cpu != group_first_cpu(sd->groups)) return; child = sd->child; - sd->groups->__cpu_power = 0; + sd->groups->cpu_power = 0; - /* - * For perf policy, if the groups in child domain share resources - * (for example cores sharing some portions of the cache hierarchy - * or SMT), then set this domain groups cpu_power such that each group - * can handle only one task, when there are other idle groups in the - * same sched domain. - */ - if (!child || (!(sd->flags & SD_POWERSAVINGS_BALANCE) && - (child->flags & - (SD_SHARE_CPUPOWER | SD_SHARE_PKG_RESOURCES)))) { - sg_inc_cpu_power(sd->groups, SCHED_LOAD_SCALE); + if (!child) { + power = SCHED_LOAD_SCALE; + weight = cpumask_weight(sched_domain_span(sd)); + /* + * SMT siblings share the power of a single core. + * Usually multiple threads get a better yield out of + * that one core than a single thread would have, + * reflect that in sd->smt_gain. + */ + if ((sd->flags & SD_SHARE_CPUPOWER) && weight > 1) { + power *= sd->smt_gain; + power /= weight; + power >>= SCHED_LOAD_SHIFT; + } + sd->groups->cpu_power += power; return; } /* - * add cpu_power of each child group to this groups cpu_power + * Add cpu_power of each child group to this groups cpu_power. */ group = child->groups; do { - sg_inc_cpu_power(sd->groups, group->__cpu_power); + sd->groups->cpu_power += group->cpu_power; group = group->next; } while (group != child->groups); } @@ -7976,287 +8624,292 @@ static void set_domain_attribute(struct sched_domain *sd, request = attr->relax_domain_level; if (request < sd->level) { /* turn off idle balance on this domain */ - sd->flags &= ~(SD_WAKE_IDLE|SD_BALANCE_NEWIDLE); + sd->flags &= ~(SD_BALANCE_WAKE|SD_BALANCE_NEWIDLE); } else { /* turn on idle balance on this domain */ - sd->flags |= (SD_WAKE_IDLE_FAR|SD_BALANCE_NEWIDLE); + sd->flags |= (SD_BALANCE_WAKE|SD_BALANCE_NEWIDLE); + } +} + +static void __free_domain_allocs(struct s_data *d, enum s_alloc what, + const struct cpumask *cpu_map) +{ + switch (what) { + case sa_sched_groups: + free_sched_groups(cpu_map, d->tmpmask); /* fall through */ + d->sched_group_nodes = NULL; + case sa_rootdomain: + free_rootdomain(d->rd); /* fall through */ + case sa_tmpmask: + free_cpumask_var(d->tmpmask); /* fall through */ + case sa_send_covered: + free_cpumask_var(d->send_covered); /* fall through */ + case sa_this_core_map: + free_cpumask_var(d->this_core_map); /* fall through */ + case sa_this_sibling_map: + free_cpumask_var(d->this_sibling_map); /* fall through */ + case sa_nodemask: + free_cpumask_var(d->nodemask); /* fall through */ + case sa_sched_group_nodes: +#ifdef CONFIG_NUMA + kfree(d->sched_group_nodes); /* fall through */ + case sa_notcovered: + free_cpumask_var(d->notcovered); /* fall through */ + case sa_covered: + free_cpumask_var(d->covered); /* fall through */ + case sa_domainspan: + free_cpumask_var(d->domainspan); /* fall through */ +#endif + case sa_none: + break; } } -/* - * Build sched domains for a given set of cpus and attach the sched domains - * to the individual cpus - */ -static int __build_sched_domains(const struct cpumask *cpu_map, - struct sched_domain_attr *attr) +static enum s_alloc __visit_domain_allocation_hell(struct s_data *d, + const struct cpumask *cpu_map) { - int i, err = -ENOMEM; - struct root_domain *rd; - cpumask_var_t nodemask, this_sibling_map, this_core_map, send_covered, - tmpmask; -#ifdef CONFIG_NUMA - cpumask_var_t domainspan, covered, notcovered; - struct sched_group **sched_group_nodes = NULL; - int sd_allnodes = 0; - - if (!alloc_cpumask_var(&domainspan, GFP_KERNEL)) - goto out; - if (!alloc_cpumask_var(&covered, GFP_KERNEL)) - goto free_domainspan; - if (!alloc_cpumask_var(¬covered, GFP_KERNEL)) - goto free_covered; -#endif - - if (!alloc_cpumask_var(&nodemask, GFP_KERNEL)) - goto free_notcovered; - if (!alloc_cpumask_var(&this_sibling_map, GFP_KERNEL)) - goto free_nodemask; - if (!alloc_cpumask_var(&this_core_map, GFP_KERNEL)) - goto free_this_sibling_map; - if (!alloc_cpumask_var(&send_covered, GFP_KERNEL)) - goto free_this_core_map; - if (!alloc_cpumask_var(&tmpmask, GFP_KERNEL)) - goto free_send_covered; - #ifdef CONFIG_NUMA - /* - * Allocate the per-node list of sched groups - */ - sched_group_nodes = kcalloc(nr_node_ids, sizeof(struct sched_group *), - GFP_KERNEL); - if (!sched_group_nodes) { + if (!alloc_cpumask_var(&d->domainspan, GFP_KERNEL)) + return sa_none; + if (!alloc_cpumask_var(&d->covered, GFP_KERNEL)) + return sa_domainspan; + if (!alloc_cpumask_var(&d->notcovered, GFP_KERNEL)) + return sa_covered; + /* Allocate the per-node list of sched groups */ + d->sched_group_nodes = kcalloc(nr_node_ids, + sizeof(struct sched_group *), GFP_KERNEL); + if (!d->sched_group_nodes) { printk(KERN_WARNING "Can not alloc sched group node list\n"); - goto free_tmpmask; - } -#endif - - rd = alloc_rootdomain(); - if (!rd) { + return sa_notcovered; + } + sched_group_nodes_bycpu[cpumask_first(cpu_map)] = d->sched_group_nodes; +#endif + if (!alloc_cpumask_var(&d->nodemask, GFP_KERNEL)) + return sa_sched_group_nodes; + if (!alloc_cpumask_var(&d->this_sibling_map, GFP_KERNEL)) + return sa_nodemask; + if (!alloc_cpumask_var(&d->this_core_map, GFP_KERNEL)) + return sa_this_sibling_map; + if (!alloc_cpumask_var(&d->send_covered, GFP_KERNEL)) + return sa_this_core_map; + if (!alloc_cpumask_var(&d->tmpmask, GFP_KERNEL)) + return sa_send_covered; + d->rd = alloc_rootdomain(); + if (!d->rd) { printk(KERN_WARNING "Cannot alloc root domain\n"); - goto free_sched_groups; + return sa_tmpmask; } + return sa_rootdomain; +} +static struct sched_domain *__build_numa_sched_domains(struct s_data *d, + const struct cpumask *cpu_map, struct sched_domain_attr *attr, int i) +{ + struct sched_domain *sd = NULL; #ifdef CONFIG_NUMA - sched_group_nodes_bycpu[cpumask_first(cpu_map)] = sched_group_nodes; -#endif - - /* - * Set up domains for cpus specified by the cpu_map. - */ - for_each_cpu(i, cpu_map) { - struct sched_domain *sd = NULL, *p; - - cpumask_and(nodemask, cpumask_of_node(cpu_to_node(i)), cpu_map); - -#ifdef CONFIG_NUMA - if (cpumask_weight(cpu_map) > - SD_NODES_PER_DOMAIN*cpumask_weight(nodemask)) { - sd = &per_cpu(allnodes_domains, i).sd; - SD_INIT(sd, ALLNODES); - set_domain_attribute(sd, attr); - cpumask_copy(sched_domain_span(sd), cpu_map); - cpu_to_allnodes_group(i, cpu_map, &sd->groups, tmpmask); - p = sd; - sd_allnodes = 1; - } else - p = NULL; + struct sched_domain *parent; - sd = &per_cpu(node_domains, i).sd; - SD_INIT(sd, NODE); + d->sd_allnodes = 0; + if (cpumask_weight(cpu_map) > + SD_NODES_PER_DOMAIN * cpumask_weight(d->nodemask)) { + sd = &per_cpu(allnodes_domains, i).sd; + SD_INIT(sd, ALLNODES); set_domain_attribute(sd, attr); - sched_domain_node_span(cpu_to_node(i), sched_domain_span(sd)); - sd->parent = p; - if (p) - p->child = sd; - cpumask_and(sched_domain_span(sd), - sched_domain_span(sd), cpu_map); + cpumask_copy(sched_domain_span(sd), cpu_map); + cpu_to_allnodes_group(i, cpu_map, &sd->groups, d->tmpmask); + d->sd_allnodes = 1; + } + parent = sd; + + sd = &per_cpu(node_domains, i).sd; + SD_INIT(sd, NODE); + set_domain_attribute(sd, attr); + sched_domain_node_span(cpu_to_node(i), sched_domain_span(sd)); + sd->parent = parent; + if (parent) + parent->child = sd; + cpumask_and(sched_domain_span(sd), sched_domain_span(sd), cpu_map); #endif + return sd; +} - p = sd; - sd = &per_cpu(phys_domains, i).sd; - SD_INIT(sd, CPU); - set_domain_attribute(sd, attr); - cpumask_copy(sched_domain_span(sd), nodemask); - sd->parent = p; - if (p) - p->child = sd; - cpu_to_phys_group(i, cpu_map, &sd->groups, tmpmask); +static struct sched_domain *__build_cpu_sched_domain(struct s_data *d, + const struct cpumask *cpu_map, struct sched_domain_attr *attr, + struct sched_domain *parent, int i) +{ + struct sched_domain *sd; + sd = &per_cpu(phys_domains, i).sd; + SD_INIT(sd, CPU); + set_domain_attribute(sd, attr); + cpumask_copy(sched_domain_span(sd), d->nodemask); + sd->parent = parent; + if (parent) + parent->child = sd; + cpu_to_phys_group(i, cpu_map, &sd->groups, d->tmpmask); + return sd; +} +static struct sched_domain *__build_mc_sched_domain(struct s_data *d, + const struct cpumask *cpu_map, struct sched_domain_attr *attr, + struct sched_domain *parent, int i) +{ + struct sched_domain *sd = parent; #ifdef CONFIG_SCHED_MC - p = sd; - sd = &per_cpu(core_domains, i).sd; - SD_INIT(sd, MC); - set_domain_attribute(sd, attr); - cpumask_and(sched_domain_span(sd), cpu_map, - cpu_coregroup_mask(i)); - sd->parent = p; - p->child = sd; - cpu_to_core_group(i, cpu_map, &sd->groups, tmpmask); + sd = &per_cpu(core_domains, i).sd; + SD_INIT(sd, MC); + set_domain_attribute(sd, attr); + cpumask_and(sched_domain_span(sd), cpu_map, cpu_coregroup_mask(i)); + sd->parent = parent; + parent->child = sd; + cpu_to_core_group(i, cpu_map, &sd->groups, d->tmpmask); #endif + return sd; +} +static struct sched_domain *__build_smt_sched_domain(struct s_data *d, + const struct cpumask *cpu_map, struct sched_domain_attr *attr, + struct sched_domain *parent, int i) +{ + struct sched_domain *sd = parent; #ifdef CONFIG_SCHED_SMT - p = sd; - sd = &per_cpu(cpu_domains, i).sd; - SD_INIT(sd, SIBLING); - set_domain_attribute(sd, attr); - cpumask_and(sched_domain_span(sd), - &per_cpu(cpu_sibling_map, i), cpu_map); - sd->parent = p; - p->child = sd; - cpu_to_cpu_group(i, cpu_map, &sd->groups, tmpmask); + sd = &per_cpu(cpu_domains, i).sd; + SD_INIT(sd, SIBLING); + set_domain_attribute(sd, attr); + cpumask_and(sched_domain_span(sd), cpu_map, topology_thread_cpumask(i)); + sd->parent = parent; + parent->child = sd; + cpu_to_cpu_group(i, cpu_map, &sd->groups, d->tmpmask); #endif - } + return sd; +} +static void build_sched_groups(struct s_data *d, enum sched_domain_level l, + const struct cpumask *cpu_map, int cpu) +{ + switch (l) { #ifdef CONFIG_SCHED_SMT - /* Set up CPU (sibling) groups */ - for_each_cpu(i, cpu_map) { - cpumask_and(this_sibling_map, - &per_cpu(cpu_sibling_map, i), cpu_map); - if (i != cpumask_first(this_sibling_map)) - continue; - - init_sched_build_groups(this_sibling_map, cpu_map, - &cpu_to_cpu_group, - send_covered, tmpmask); - } + case SD_LV_SIBLING: /* set up CPU (sibling) groups */ + cpumask_and(d->this_sibling_map, cpu_map, + topology_thread_cpumask(cpu)); + if (cpu == cpumask_first(d->this_sibling_map)) + init_sched_build_groups(d->this_sibling_map, cpu_map, + &cpu_to_cpu_group, + d->send_covered, d->tmpmask); + break; #endif - #ifdef CONFIG_SCHED_MC - /* Set up multi-core groups */ - for_each_cpu(i, cpu_map) { - cpumask_and(this_core_map, cpu_coregroup_mask(i), cpu_map); - if (i != cpumask_first(this_core_map)) - continue; - - init_sched_build_groups(this_core_map, cpu_map, - &cpu_to_core_group, - send_covered, tmpmask); - } + case SD_LV_MC: /* set up multi-core groups */ + cpumask_and(d->this_core_map, cpu_map, cpu_coregroup_mask(cpu)); + if (cpu == cpumask_first(d->this_core_map)) + init_sched_build_groups(d->this_core_map, cpu_map, + &cpu_to_core_group, + d->send_covered, d->tmpmask); + break; #endif - - /* Set up physical groups */ - for (i = 0; i < nr_node_ids; i++) { - cpumask_and(nodemask, cpumask_of_node(i), cpu_map); - if (cpumask_empty(nodemask)) - continue; - - init_sched_build_groups(nodemask, cpu_map, - &cpu_to_phys_group, - send_covered, tmpmask); - } - + case SD_LV_CPU: /* set up physical groups */ + cpumask_and(d->nodemask, cpumask_of_node(cpu), cpu_map); + if (!cpumask_empty(d->nodemask)) + init_sched_build_groups(d->nodemask, cpu_map, + &cpu_to_phys_group, + d->send_covered, d->tmpmask); + break; #ifdef CONFIG_NUMA - /* Set up node groups */ - if (sd_allnodes) { - init_sched_build_groups(cpu_map, cpu_map, - &cpu_to_allnodes_group, - send_covered, tmpmask); + case SD_LV_ALLNODES: + init_sched_build_groups(cpu_map, cpu_map, &cpu_to_allnodes_group, + d->send_covered, d->tmpmask); + break; +#endif + default: + break; } +} - for (i = 0; i < nr_node_ids; i++) { - /* Set up node groups */ - struct sched_group *sg, *prev; - int j; - - cpumask_clear(covered); - cpumask_and(nodemask, cpumask_of_node(i), cpu_map); - if (cpumask_empty(nodemask)) { - sched_group_nodes[i] = NULL; - continue; - } +/* + * Build sched domains for a given set of cpus and attach the sched domains + * to the individual cpus + */ +static int __build_sched_domains(const struct cpumask *cpu_map, + struct sched_domain_attr *attr) +{ + enum s_alloc alloc_state = sa_none; + struct s_data d; + struct sched_domain *sd; + int i; +#ifdef CONFIG_NUMA + d.sd_allnodes = 0; +#endif - sched_domain_node_span(i, domainspan); - cpumask_and(domainspan, domainspan, cpu_map); + alloc_state = __visit_domain_allocation_hell(&d, cpu_map); + if (alloc_state != sa_rootdomain) + goto error; + alloc_state = sa_sched_groups; - sg = kmalloc_node(sizeof(struct sched_group) + cpumask_size(), - GFP_KERNEL, i); - if (!sg) { - printk(KERN_WARNING "Can not alloc domain group for " - "node %d\n", i); - goto error; - } - sched_group_nodes[i] = sg; - for_each_cpu(j, nodemask) { - struct sched_domain *sd; + /* + * Set up domains for cpus specified by the cpu_map. + */ + for_each_cpu(i, cpu_map) { + cpumask_and(d.nodemask, cpumask_of_node(cpu_to_node(i)), + cpu_map); - sd = &per_cpu(node_domains, j).sd; - sd->groups = sg; - } - sg->__cpu_power = 0; - cpumask_copy(sched_group_cpus(sg), nodemask); - sg->next = sg; - cpumask_or(covered, covered, nodemask); - prev = sg; + sd = __build_numa_sched_domains(&d, cpu_map, attr, i); + sd = __build_cpu_sched_domain(&d, cpu_map, attr, sd, i); + sd = __build_mc_sched_domain(&d, cpu_map, attr, sd, i); + sd = __build_smt_sched_domain(&d, cpu_map, attr, sd, i); + } - for (j = 0; j < nr_node_ids; j++) { - int n = (i + j) % nr_node_ids; + for_each_cpu(i, cpu_map) { + build_sched_groups(&d, SD_LV_SIBLING, cpu_map, i); + build_sched_groups(&d, SD_LV_MC, cpu_map, i); + } - cpumask_complement(notcovered, covered); - cpumask_and(tmpmask, notcovered, cpu_map); - cpumask_and(tmpmask, tmpmask, domainspan); - if (cpumask_empty(tmpmask)) - break; + /* Set up physical groups */ + for (i = 0; i < nr_node_ids; i++) + build_sched_groups(&d, SD_LV_CPU, cpu_map, i); - cpumask_and(tmpmask, tmpmask, cpumask_of_node(n)); - if (cpumask_empty(tmpmask)) - continue; +#ifdef CONFIG_NUMA + /* Set up node groups */ + if (d.sd_allnodes) + build_sched_groups(&d, SD_LV_ALLNODES, cpu_map, 0); - sg = kmalloc_node(sizeof(struct sched_group) + - cpumask_size(), - GFP_KERNEL, i); - if (!sg) { - printk(KERN_WARNING - "Can not alloc domain group for node %d\n", j); - goto error; - } - sg->__cpu_power = 0; - cpumask_copy(sched_group_cpus(sg), tmpmask); - sg->next = prev->next; - cpumask_or(covered, covered, tmpmask); - prev->next = sg; - prev = sg; - } - } + for (i = 0; i < nr_node_ids; i++) + if (build_numa_sched_groups(&d, cpu_map, i)) + goto error; #endif /* Calculate CPU power for physical packages and nodes */ #ifdef CONFIG_SCHED_SMT for_each_cpu(i, cpu_map) { - struct sched_domain *sd = &per_cpu(cpu_domains, i).sd; - + sd = &per_cpu(cpu_domains, i).sd; init_sched_groups_power(i, sd); } #endif #ifdef CONFIG_SCHED_MC for_each_cpu(i, cpu_map) { - struct sched_domain *sd = &per_cpu(core_domains, i).sd; - + sd = &per_cpu(core_domains, i).sd; init_sched_groups_power(i, sd); } #endif for_each_cpu(i, cpu_map) { - struct sched_domain *sd = &per_cpu(phys_domains, i).sd; - + sd = &per_cpu(phys_domains, i).sd; init_sched_groups_power(i, sd); } #ifdef CONFIG_NUMA for (i = 0; i < nr_node_ids; i++) - init_numa_sched_groups_power(sched_group_nodes[i]); + init_numa_sched_groups_power(d.sched_group_nodes[i]); - if (sd_allnodes) { + if (d.sd_allnodes) { struct sched_group *sg; cpu_to_allnodes_group(cpumask_first(cpu_map), cpu_map, &sg, - tmpmask); + d.tmpmask); init_numa_sched_groups_power(sg); } #endif /* Attach the domains */ for_each_cpu(i, cpu_map) { - struct sched_domain *sd; #ifdef CONFIG_SCHED_SMT sd = &per_cpu(cpu_domains, i).sd; #elif defined(CONFIG_SCHED_MC) @@ -8264,44 +8917,16 @@ static int __build_sched_domains(const struct cpumask *cpu_map, #else sd = &per_cpu(phys_domains, i).sd; #endif - cpu_attach_domain(sd, rd, i); + cpu_attach_domain(sd, d.rd, i); } - err = 0; - -free_tmpmask: - free_cpumask_var(tmpmask); -free_send_covered: - free_cpumask_var(send_covered); -free_this_core_map: - free_cpumask_var(this_core_map); -free_this_sibling_map: - free_cpumask_var(this_sibling_map); -free_nodemask: - free_cpumask_var(nodemask); -free_notcovered: -#ifdef CONFIG_NUMA - free_cpumask_var(notcovered); -free_covered: - free_cpumask_var(covered); -free_domainspan: - free_cpumask_var(domainspan); -out: -#endif - return err; - -free_sched_groups: -#ifdef CONFIG_NUMA - kfree(sched_group_nodes); -#endif - goto free_tmpmask; + d.sched_group_nodes = NULL; /* don't free this we still need it */ + __free_domain_allocs(&d, sa_tmpmask, cpu_map); + return 0; -#ifdef CONFIG_NUMA error: - free_sched_groups(cpu_map, tmpmask); - free_rootdomain(rd); - goto free_tmpmask; -#endif + __free_domain_allocs(&d, alloc_state, cpu_map); + return -ENOMEM; } static int build_sched_domains(const struct cpumask *cpu_map) @@ -8309,7 +8934,7 @@ static int build_sched_domains(const struct cpumask *cpu_map) return __build_sched_domains(cpu_map, NULL); } -static struct cpumask *doms_cur; /* current sched domains */ +static cpumask_var_t *doms_cur; /* current sched domains */ static int ndoms_cur; /* number of sched domains in 'doms_cur' */ static struct sched_domain_attr *dattr_cur; /* attribues of custom domains in 'doms_cur' */ @@ -8331,6 +8956,31 @@ int __attribute__((weak)) arch_update_cpu_topology(void) return 0; } +cpumask_var_t *alloc_sched_domains(unsigned int ndoms) +{ + int i; + cpumask_var_t *doms; + + doms = kmalloc(sizeof(*doms) * ndoms, GFP_KERNEL); + if (!doms) + return NULL; + for (i = 0; i < ndoms; i++) { + if (!alloc_cpumask_var(&doms[i], GFP_KERNEL)) { + free_sched_domains(doms, i); + return NULL; + } + } + return doms; +} + +void free_sched_domains(cpumask_var_t doms[], unsigned int ndoms) +{ + unsigned int i; + for (i = 0; i < ndoms; i++) + free_cpumask_var(doms[i]); + kfree(doms); +} + /* * Set up scheduler domains and groups. Callers must hold the hotplug lock. * For now this just excludes isolated cpus, but could be used to @@ -8342,12 +8992,12 @@ static int arch_init_sched_domains(const struct cpumask *cpu_map) arch_update_cpu_topology(); ndoms_cur = 1; - doms_cur = kmalloc(cpumask_size(), GFP_KERNEL); + doms_cur = alloc_sched_domains(ndoms_cur); if (!doms_cur) - doms_cur = fallback_doms; - cpumask_andnot(doms_cur, cpu_map, cpu_isolated_map); + doms_cur = &fallback_doms; + cpumask_andnot(doms_cur[0], cpu_map, cpu_isolated_map); dattr_cur = NULL; - err = build_sched_domains(doms_cur); + err = build_sched_domains(doms_cur[0]); register_sched_domain_sysctl(); return err; @@ -8397,19 +9047,19 @@ static int dattrs_equal(struct sched_domain_attr *cur, int idx_cur, * doms_new[] to the current sched domain partitioning, doms_cur[]. * It destroys each deleted domain and builds each new domain. * - * 'doms_new' is an array of cpumask's of length 'ndoms_new'. + * 'doms_new' is an array of cpumask_var_t's of length 'ndoms_new'. * The masks don't intersect (don't overlap.) We should setup one * sched domain for each mask. CPUs not in any of the cpumasks will * not be load balanced. If the same cpumask appears both in the * current 'doms_cur' domains and in the new 'doms_new', we can leave * it as it is. * - * The passed in 'doms_new' should be kmalloc'd. This routine takes - * ownership of it and will kfree it when done with it. If the caller - * failed the kmalloc call, then it can pass in doms_new == NULL && - * ndoms_new == 1, and partition_sched_domains() will fallback to - * the single partition 'fallback_doms', it also forces the domains - * to be rebuilt. + * The passed in 'doms_new' should be allocated using + * alloc_sched_domains. This routine takes ownership of it and will + * free_sched_domains it when done with it. If the caller failed the + * alloc call, then it can pass in doms_new == NULL && ndoms_new == 1, + * and partition_sched_domains() will fallback to the single partition + * 'fallback_doms', it also forces the domains to be rebuilt. * * If doms_new == NULL it will be replaced with cpu_online_mask. * ndoms_new == 0 is a special case for destroying existing domains, @@ -8417,8 +9067,7 @@ static int dattrs_equal(struct sched_domain_attr *cur, int idx_cur, * * Call with hotplug lock held */ -/* FIXME: Change to struct cpumask *doms_new[] */ -void partition_sched_domains(int ndoms_new, struct cpumask *doms_new, +void partition_sched_domains(int ndoms_new, cpumask_var_t doms_new[], struct sched_domain_attr *dattr_new) { int i, j, n; @@ -8437,40 +9086,40 @@ void partition_sched_domains(int ndoms_new, struct cpumask *doms_new, /* Destroy deleted domains */ for (i = 0; i < ndoms_cur; i++) { for (j = 0; j < n && !new_topology; j++) { - if (cpumask_equal(&doms_cur[i], &doms_new[j]) + if (cpumask_equal(doms_cur[i], doms_new[j]) && dattrs_equal(dattr_cur, i, dattr_new, j)) goto match1; } /* no match - a current sched domain not in new doms_new[] */ - detach_destroy_domains(doms_cur + i); + detach_destroy_domains(doms_cur[i]); match1: ; } if (doms_new == NULL) { ndoms_cur = 0; - doms_new = fallback_doms; - cpumask_andnot(&doms_new[0], cpu_online_mask, cpu_isolated_map); + doms_new = &fallback_doms; + cpumask_andnot(doms_new[0], cpu_online_mask, cpu_isolated_map); WARN_ON_ONCE(dattr_new); } /* Build new domains */ for (i = 0; i < ndoms_new; i++) { for (j = 0; j < ndoms_cur && !new_topology; j++) { - if (cpumask_equal(&doms_new[i], &doms_cur[j]) + if (cpumask_equal(doms_new[i], doms_cur[j]) && dattrs_equal(dattr_new, i, dattr_cur, j)) goto match2; } /* no match - add a new doms_new */ - __build_sched_domains(doms_new + i, + __build_sched_domains(doms_new[i], dattr_new ? dattr_new + i : NULL); match2: ; } /* Remember the new sched domains */ - if (doms_cur != fallback_doms) - kfree(doms_cur); + if (doms_cur != &fallback_doms) + free_sched_domains(doms_cur, ndoms_cur); kfree(dattr_cur); /* kfree(NULL) is safe */ doms_cur = doms_new; dattr_cur = dattr_new; @@ -8620,6 +9269,7 @@ void __init sched_init_smp(void) cpumask_var_t non_isolated_cpus; alloc_cpumask_var(&non_isolated_cpus, GFP_KERNEL); + alloc_cpumask_var(&fallback_doms, GFP_KERNEL); #if defined(CONFIG_NUMA) sched_group_nodes_bycpu = kzalloc(nr_cpu_ids * sizeof(void **), @@ -8651,7 +9301,6 @@ void __init sched_init_smp(void) sched_init_granularity(); free_cpumask_var(non_isolated_cpus); - alloc_cpumask_var(&fallback_doms, GFP_KERNEL); init_sched_rt_class(); } #else @@ -8661,6 +9310,8 @@ void __init sched_init_smp(void) } #endif /* CONFIG_SMP */ +const_debug unsigned int sysctl_timer_migration = 1; + int in_sched_functions(unsigned long addr) { return in_lock_functions(addr) || @@ -8700,7 +9351,7 @@ static void init_rt_rq(struct rt_rq *rt_rq, struct rq *rq) #ifdef CONFIG_SMP rt_rq->rt_nr_migratory = 0; rt_rq->overloaded = 0; - plist_head_init(&rq->rt.pushable_tasks, &rq->lock); + plist_head_init(&rt_rq->pushable_tasks, &rq->lock); #endif rt_rq->rt_time = 0; @@ -8787,12 +9438,11 @@ void __init sched_init(void) #ifdef CONFIG_USER_SCHED alloc_size *= 2; #endif - /* - * As sched_init() is called before page_alloc is setup, - * we use alloc_bootmem(). - */ +#ifdef CONFIG_CPUMASK_OFFSTACK + alloc_size += num_possible_cpus() * cpumask_size(); +#endif if (alloc_size) { - ptr = (unsigned long)alloc_bootmem(alloc_size); + ptr = (unsigned long)kzalloc(alloc_size, GFP_NOWAIT); #ifdef CONFIG_FAIR_GROUP_SCHED init_task_group.se = (struct sched_entity **)ptr; @@ -8824,6 +9474,12 @@ void __init sched_init(void) ptr += nr_cpu_ids * sizeof(void **); #endif /* CONFIG_USER_SCHED */ #endif /* CONFIG_RT_GROUP_SCHED */ +#ifdef CONFIG_CPUMASK_OFFSTACK + for_each_possible_cpu(i) { + per_cpu(load_balance_tmpmask, i) = (void *)ptr; + ptr += cpumask_size(); + } +#endif /* CONFIG_CPUMASK_OFFSTACK */ } #ifdef CONFIG_SMP @@ -8853,12 +9509,18 @@ void __init sched_init(void) #endif /* CONFIG_USER_SCHED */ #endif /* CONFIG_GROUP_SCHED */ +#if defined CONFIG_FAIR_GROUP_SCHED && defined CONFIG_SMP + update_shares_data = __alloc_percpu(nr_cpu_ids * sizeof(unsigned long), + __alignof__(unsigned long)); +#endif for_each_possible_cpu(i) { struct rq *rq; rq = cpu_rq(i); spin_lock_init(&rq->lock); rq->nr_running = 0; + rq->calc_load_active = 0; + rq->calc_load_update = jiffies + LOAD_FREQ; init_cfs_rq(&rq->cfs, rq); init_rt_rq(&rq->rt, rq); #ifdef CONFIG_FAIR_GROUP_SCHED @@ -8879,7 +9541,7 @@ void __init sched_init(void) * 1024) and two child groups A0 and A1 (of weight 1024 each), * then A0's share of the cpu resource is: * - * A0's bandwidth = 1024 / (10*1024 + 1024 + 1024) = 8.33% + * A0's bandwidth = 1024 / (10*1024 + 1024 + 1024) = 8.33% * * We achieve this by letting init_task_group's tasks sit * directly in rq->cfs (i.e init_task_group->se[] = NULL). @@ -8896,11 +9558,11 @@ void __init sched_init(void) * system cpu resource, based on the weight assigned to root * user's cpu share (INIT_TASK_GROUP_LOAD). This is accomplished * by letting tasks of init_task_group sit in a separate cfs_rq - * (init_cfs_rq) and having one entity represent this group of + * (init_tg_cfs_rq) and having one entity represent this group of * tasks in rq->cfs (i.e init_task_group->se[] != NULL). */ init_tg_cfs_entry(&init_task_group, - &per_cpu(init_cfs_rq, i), + &per_cpu(init_tg_cfs_rq, i), &per_cpu(init_sched_entity, i), i, 1, root_task_group.se[i]); @@ -8926,12 +9588,15 @@ void __init sched_init(void) #ifdef CONFIG_SMP rq->sd = NULL; rq->rd = NULL; + rq->post_schedule = 0; rq->active_balance = 0; rq->next_balance = jiffies; rq->push_cpu = 0; rq->cpu = i; rq->online = 0; rq->migration_thread = NULL; + rq->idle_stamp = 0; + rq->avg_idle = 2*sysctl_sched_migration_cost; INIT_LIST_HEAD(&rq->migration_queue); rq_attach_root(rq, &def_root_domain); #endif @@ -8966,31 +9631,46 @@ void __init sched_init(void) * when this runqueue becomes "idle". */ init_idle(current, smp_processor_id()); + + calc_load_update = jiffies + LOAD_FREQ; + /* * During early bootup we pretend to be a normal task: */ current->sched_class = &fair_sched_class; /* Allocate the nohz_cpu_mask if CONFIG_CPUMASK_OFFSTACK */ - alloc_bootmem_cpumask_var(&nohz_cpu_mask); + zalloc_cpumask_var(&nohz_cpu_mask, GFP_NOWAIT); #ifdef CONFIG_SMP #ifdef CONFIG_NO_HZ - alloc_bootmem_cpumask_var(&nohz.cpu_mask); + zalloc_cpumask_var(&nohz.cpu_mask, GFP_NOWAIT); + alloc_cpumask_var(&nohz.ilb_grp_nohz_mask, GFP_NOWAIT); #endif - alloc_bootmem_cpumask_var(&cpu_isolated_map); + /* May be allocated at isolcpus cmdline parse time */ + if (cpu_isolated_map == NULL) + zalloc_cpumask_var(&cpu_isolated_map, GFP_NOWAIT); #endif /* SMP */ + perf_event_init(); + scheduler_running = 1; } #ifdef CONFIG_DEBUG_SPINLOCK_SLEEP -void __might_sleep(char *file, int line) +static inline int preempt_count_equals(int preempt_offset) +{ + int nested = preempt_count() & ~PREEMPT_ACTIVE; + + return (nested == PREEMPT_INATOMIC_BASE + preempt_offset); +} + +void __might_sleep(char *file, int line, int preempt_offset) { #ifdef in_atomic static unsigned long prev_jiffy; /* ratelimiting */ - if ((!in_atomic() && !irqs_disabled()) || - system_state != SYSTEM_RUNNING || oops_in_progress) + if ((preempt_count_equals(preempt_offset) && !irqs_disabled()) || + system_state != SYSTEM_RUNNING || oops_in_progress) return; if (time_before(jiffies, prev_jiffy + HZ) && prev_jiffy) return; @@ -9721,6 +10401,13 @@ static int sched_rt_global_constraints(void) if (sysctl_sched_rt_period <= 0) return -EINVAL; + /* + * There's always some RT tasks in the root group + * -- migration, kstopmachine etc.. + */ + if (sysctl_sched_rt_runtime == 0) + return -EBUSY; + spin_lock_irqsave(&def_rt_bandwidth.rt_runtime_lock, flags); for_each_possible_cpu(i) { struct rt_rq *rt_rq = &cpu_rq(i)->rt; @@ -9736,7 +10423,7 @@ static int sched_rt_global_constraints(void) #endif /* CONFIG_RT_GROUP_SCHED */ int sched_rt_handler(struct ctl_table *table, int write, - struct file *filp, void __user *buffer, size_t *lenp, + void __user *buffer, size_t *lenp, loff_t *ppos) { int ret; @@ -9747,7 +10434,7 @@ int sched_rt_handler(struct ctl_table *table, int write, old_period = sysctl_sched_rt_period; old_runtime = sysctl_sched_rt_runtime; - ret = proc_dointvec(table, write, filp, buffer, lenp, ppos); + ret = proc_dointvec(table, write, buffer, lenp, ppos); if (!ret && write) { ret = sched_rt_global_constraints(); @@ -9801,8 +10488,7 @@ cpu_cgroup_destroy(struct cgroup_subsys *ss, struct cgroup *cgrp) } static int -cpu_cgroup_can_attach(struct cgroup_subsys *ss, struct cgroup *cgrp, - struct task_struct *tsk) +cpu_cgroup_can_attach_task(struct cgroup *cgrp, struct task_struct *tsk) { #ifdef CONFIG_RT_GROUP_SCHED if (!sched_rt_can_attach(cgroup_tg(cgrp), tsk)) @@ -9812,15 +10498,45 @@ cpu_cgroup_can_attach(struct cgroup_subsys *ss, struct cgroup *cgrp, if (tsk->sched_class != &fair_sched_class) return -EINVAL; #endif + return 0; +} +static int +cpu_cgroup_can_attach(struct cgroup_subsys *ss, struct cgroup *cgrp, + struct task_struct *tsk, bool threadgroup) +{ + int retval = cpu_cgroup_can_attach_task(cgrp, tsk); + if (retval) + return retval; + if (threadgroup) { + struct task_struct *c; + rcu_read_lock(); + list_for_each_entry_rcu(c, &tsk->thread_group, thread_group) { + retval = cpu_cgroup_can_attach_task(cgrp, c); + if (retval) { + rcu_read_unlock(); + return retval; + } + } + rcu_read_unlock(); + } return 0; } static void cpu_cgroup_attach(struct cgroup_subsys *ss, struct cgroup *cgrp, - struct cgroup *old_cont, struct task_struct *tsk) + struct cgroup *old_cont, struct task_struct *tsk, + bool threadgroup) { sched_move_task(tsk); + if (threadgroup) { + struct task_struct *c; + rcu_read_lock(); + list_for_each_entry_rcu(c, &tsk->thread_group, thread_group) { + sched_move_task(c); + } + rcu_read_unlock(); + } } #ifdef CONFIG_FAIR_GROUP_SCHED @@ -9916,6 +10632,7 @@ struct cpuacct { struct cgroup_subsys_state css; /* cpuusage holds pointer to a u64-type object on every cpu */ u64 *cpuusage; + struct percpu_counter cpustat[CPUACCT_STAT_NSTATS]; struct cpuacct *parent; }; @@ -9940,20 +10657,32 @@ static struct cgroup_subsys_state *cpuacct_create( struct cgroup_subsys *ss, struct cgroup *cgrp) { struct cpuacct *ca = kzalloc(sizeof(*ca), GFP_KERNEL); + int i; if (!ca) - return ERR_PTR(-ENOMEM); + goto out; ca->cpuusage = alloc_percpu(u64); - if (!ca->cpuusage) { - kfree(ca); - return ERR_PTR(-ENOMEM); - } + if (!ca->cpuusage) + goto out_free_ca; + + for (i = 0; i < CPUACCT_STAT_NSTATS; i++) + if (percpu_counter_init(&ca->cpustat[i], 0)) + goto out_free_counters; if (cgrp->parent) ca->parent = cgroup_ca(cgrp->parent); return &ca->css; + +out_free_counters: + while (--i >= 0) + percpu_counter_destroy(&ca->cpustat[i]); + free_percpu(ca->cpuusage); +out_free_ca: + kfree(ca); +out: + return ERR_PTR(-ENOMEM); } /* destroy an existing cpu accounting group */ @@ -9961,7 +10690,10 @@ static void cpuacct_destroy(struct cgroup_subsys *ss, struct cgroup *cgrp) { struct cpuacct *ca = cgroup_ca(cgrp); + int i; + for (i = 0; i < CPUACCT_STAT_NSTATS; i++) + percpu_counter_destroy(&ca->cpustat[i]); free_percpu(ca->cpuusage); kfree(ca); } @@ -10048,6 +10780,25 @@ static int cpuacct_percpu_seq_read(struct cgroup *cgroup, struct cftype *cft, return 0; } +static const char *cpuacct_stat_desc[] = { + [CPUACCT_STAT_USER] = "user", + [CPUACCT_STAT_SYSTEM] = "system", +}; + +static int cpuacct_stats_show(struct cgroup *cgrp, struct cftype *cft, + struct cgroup_map_cb *cb) +{ + struct cpuacct *ca = cgroup_ca(cgrp); + int i; + + for (i = 0; i < CPUACCT_STAT_NSTATS; i++) { + s64 val = percpu_counter_read(&ca->cpustat[i]); + val = cputime64_to_clock_t(val); + cb->fill(cb, cpuacct_stat_desc[i], val); + } + return 0; +} + static struct cftype files[] = { { .name = "usage", @@ -10058,7 +10809,10 @@ static struct cftype files[] = { .name = "usage_percpu", .read_seq_string = cpuacct_percpu_seq_read, }, - + { + .name = "stat", + .read_map = cpuacct_stats_show, + }, }; static int cpuacct_populate(struct cgroup_subsys *ss, struct cgroup *cgrp) @@ -10080,12 +10834,38 @@ static void cpuacct_charge(struct task_struct *tsk, u64 cputime) return; cpu = task_cpu(tsk); + + rcu_read_lock(); + ca = task_ca(tsk); for (; ca; ca = ca->parent) { u64 *cpuusage = per_cpu_ptr(ca->cpuusage, cpu); *cpuusage += cputime; } + + rcu_read_unlock(); +} + +/* + * Charge the system/user time to the task's accounting group. + */ +static void cpuacct_update_stats(struct task_struct *tsk, + enum cpuacct_stat_index idx, cputime_t val) +{ + struct cpuacct *ca; + + if (unlikely(!cpuacct_subsys.active)) + return; + + rcu_read_lock(); + ca = task_ca(tsk); + + do { + percpu_counter_add(&ca->cpustat[idx], val); + ca = ca->parent; + } while (ca); + rcu_read_unlock(); } struct cgroup_subsys cpuacct_subsys = { @@ -10096,3 +10876,114 @@ struct cgroup_subsys cpuacct_subsys = { .subsys_id = cpuacct_subsys_id, }; #endif /* CONFIG_CGROUP_CPUACCT */ + +#ifndef CONFIG_SMP + +int rcu_expedited_torture_stats(char *page) +{ + return 0; +} +EXPORT_SYMBOL_GPL(rcu_expedited_torture_stats); + +void synchronize_sched_expedited(void) +{ +} +EXPORT_SYMBOL_GPL(synchronize_sched_expedited); + +#else /* #ifndef CONFIG_SMP */ + +static DEFINE_PER_CPU(struct migration_req, rcu_migration_req); +static DEFINE_MUTEX(rcu_sched_expedited_mutex); + +#define RCU_EXPEDITED_STATE_POST -2 +#define RCU_EXPEDITED_STATE_IDLE -1 + +static int rcu_expedited_state = RCU_EXPEDITED_STATE_IDLE; + +int rcu_expedited_torture_stats(char *page) +{ + int cnt = 0; + int cpu; + + cnt += sprintf(&page[cnt], "state: %d /", rcu_expedited_state); + for_each_online_cpu(cpu) { + cnt += sprintf(&page[cnt], " %d:%d", + cpu, per_cpu(rcu_migration_req, cpu).dest_cpu); + } + cnt += sprintf(&page[cnt], "\n"); + return cnt; +} +EXPORT_SYMBOL_GPL(rcu_expedited_torture_stats); + +static long synchronize_sched_expedited_count; + +/* + * Wait for an rcu-sched grace period to elapse, but use "big hammer" + * approach to force grace period to end quickly. This consumes + * significant time on all CPUs, and is thus not recommended for + * any sort of common-case code. + * + * Note that it is illegal to call this function while holding any + * lock that is acquired by a CPU-hotplug notifier. Failing to + * observe this restriction will result in deadlock. + */ +void synchronize_sched_expedited(void) +{ + int cpu; + unsigned long flags; + bool need_full_sync = 0; + struct rq *rq; + struct migration_req *req; + long snap; + int trycount = 0; + + smp_mb(); /* ensure prior mod happens before capturing snap. */ + snap = ACCESS_ONCE(synchronize_sched_expedited_count) + 1; + get_online_cpus(); + while (!mutex_trylock(&rcu_sched_expedited_mutex)) { + put_online_cpus(); + if (trycount++ < 10) + udelay(trycount * num_online_cpus()); + else { + synchronize_sched(); + return; + } + if (ACCESS_ONCE(synchronize_sched_expedited_count) - snap > 0) { + smp_mb(); /* ensure test happens before caller kfree */ + return; + } + get_online_cpus(); + } + rcu_expedited_state = RCU_EXPEDITED_STATE_POST; + for_each_online_cpu(cpu) { + rq = cpu_rq(cpu); + req = &per_cpu(rcu_migration_req, cpu); + init_completion(&req->done); + req->task = NULL; + req->dest_cpu = RCU_MIGRATION_NEED_QS; + spin_lock_irqsave(&rq->lock, flags); + list_add(&req->list, &rq->migration_queue); + spin_unlock_irqrestore(&rq->lock, flags); + wake_up_process(rq->migration_thread); + } + for_each_online_cpu(cpu) { + rcu_expedited_state = cpu; + req = &per_cpu(rcu_migration_req, cpu); + rq = cpu_rq(cpu); + wait_for_completion(&req->done); + spin_lock_irqsave(&rq->lock, flags); + if (unlikely(req->dest_cpu == RCU_MIGRATION_MUST_SYNC)) + need_full_sync = 1; + req->dest_cpu = RCU_MIGRATION_IDLE; + spin_unlock_irqrestore(&rq->lock, flags); + } + rcu_expedited_state = RCU_EXPEDITED_STATE_IDLE; + synchronize_sched_expedited_count++; + mutex_unlock(&rcu_sched_expedited_mutex); + put_online_cpus(); + if (need_full_sync) + synchronize_sched(); +} +EXPORT_SYMBOL_GPL(synchronize_sched_expedited); + +#endif /* #else #ifndef CONFIG_SMP */ diff --git a/kernel/sched_clock.c b/kernel/sched_clock.c index 390f33234bd..479ce5682d7 100644 --- a/kernel/sched_clock.c +++ b/kernel/sched_clock.c @@ -25,6 +25,7 @@ * consistent between cpus (never more than 2 jiffies difference). */ #include <linux/spinlock.h> +#include <linux/hardirq.h> #include <linux/module.h> #include <linux/percpu.h> #include <linux/ktime.h> @@ -37,7 +38,8 @@ */ unsigned long long __attribute__((weak)) sched_clock(void) { - return (unsigned long long)jiffies * (NSEC_PER_SEC / HZ); + return (unsigned long long)(jiffies - INITIAL_JIFFIES) + * (NSEC_PER_SEC / HZ); } static __read_mostly int sched_clock_running; @@ -46,13 +48,6 @@ static __read_mostly int sched_clock_running; __read_mostly int sched_clock_stable; struct sched_clock_data { - /* - * Raw spinlock - this is a special case: this might be called - * from within instrumentation code so we dont want to do any - * instrumentation ourselves. - */ - raw_spinlock_t lock; - u64 tick_raw; u64 tick_gtod; u64 clock; @@ -78,7 +73,6 @@ void sched_clock_init(void) for_each_possible_cpu(cpu) { struct sched_clock_data *scd = cpu_sdc(cpu); - scd->lock = (raw_spinlock_t)__RAW_SPIN_LOCK_UNLOCKED; scd->tick_raw = 0; scd->tick_gtod = ktime_now; scd->clock = ktime_now; @@ -107,14 +101,19 @@ static inline u64 wrap_max(u64 x, u64 y) * - filter out backward motion * - use the GTOD tick value to create a window to filter crazy TSC values */ -static u64 __update_sched_clock(struct sched_clock_data *scd, u64 now) +static u64 sched_clock_local(struct sched_clock_data *scd) { - s64 delta = now - scd->tick_raw; - u64 clock, min_clock, max_clock; + u64 now, clock, old_clock, min_clock, max_clock; + s64 delta; +again: + now = sched_clock(); + delta = now - scd->tick_raw; if (unlikely(delta < 0)) delta = 0; + old_clock = scd->clock; + /* * scd->clock = clamp(scd->tick_gtod + delta, * max(scd->tick_gtod, scd->clock), @@ -122,73 +121,73 @@ static u64 __update_sched_clock(struct sched_clock_data *scd, u64 now) */ clock = scd->tick_gtod + delta; - min_clock = wrap_max(scd->tick_gtod, scd->clock); - max_clock = wrap_max(scd->clock, scd->tick_gtod + TICK_NSEC); + min_clock = wrap_max(scd->tick_gtod, old_clock); + max_clock = wrap_max(old_clock, scd->tick_gtod + TICK_NSEC); clock = wrap_max(clock, min_clock); clock = wrap_min(clock, max_clock); - scd->clock = clock; + if (cmpxchg64(&scd->clock, old_clock, clock) != old_clock) + goto again; - return scd->clock; + return clock; } -static void lock_double_clock(struct sched_clock_data *data1, - struct sched_clock_data *data2) +static u64 sched_clock_remote(struct sched_clock_data *scd) { - if (data1 < data2) { - __raw_spin_lock(&data1->lock); - __raw_spin_lock(&data2->lock); + struct sched_clock_data *my_scd = this_scd(); + u64 this_clock, remote_clock; + u64 *ptr, old_val, val; + + sched_clock_local(my_scd); +again: + this_clock = my_scd->clock; + remote_clock = scd->clock; + + /* + * Use the opportunity that we have both locks + * taken to couple the two clocks: we take the + * larger time as the latest time for both + * runqueues. (this creates monotonic movement) + */ + if (likely((s64)(remote_clock - this_clock) < 0)) { + ptr = &scd->clock; + old_val = remote_clock; + val = this_clock; } else { - __raw_spin_lock(&data2->lock); - __raw_spin_lock(&data1->lock); + /* + * Should be rare, but possible: + */ + ptr = &my_scd->clock; + old_val = this_clock; + val = remote_clock; } + + if (cmpxchg64(ptr, old_val, val) != old_val) + goto again; + + return val; } u64 sched_clock_cpu(int cpu) { - u64 now, clock, this_clock, remote_clock; struct sched_clock_data *scd; + u64 clock; - if (sched_clock_stable) - return sched_clock(); - - scd = cpu_sdc(cpu); WARN_ON_ONCE(!irqs_disabled()); - now = sched_clock(); - - if (cpu != raw_smp_processor_id()) { - struct sched_clock_data *my_scd = this_scd(); - lock_double_clock(scd, my_scd); + if (sched_clock_stable) + return sched_clock(); - this_clock = __update_sched_clock(my_scd, now); - remote_clock = scd->clock; + if (unlikely(!sched_clock_running)) + return 0ull; - /* - * Use the opportunity that we have both locks - * taken to couple the two clocks: we take the - * larger time as the latest time for both - * runqueues. (this creates monotonic movement) - */ - if (likely((s64)(remote_clock - this_clock) < 0)) { - clock = this_clock; - scd->clock = clock; - } else { - /* - * Should be rare, but possible: - */ - clock = remote_clock; - my_scd->clock = remote_clock; - } - - __raw_spin_unlock(&my_scd->lock); - } else { - __raw_spin_lock(&scd->lock); - clock = __update_sched_clock(scd, now); - } + scd = cpu_sdc(cpu); - __raw_spin_unlock(&scd->lock); + if (cpu != smp_processor_id()) + clock = sched_clock_remote(scd); + else + clock = sched_clock_local(scd); return clock; } @@ -210,11 +209,9 @@ void sched_clock_tick(void) now_gtod = ktime_to_ns(ktime_get()); now = sched_clock(); - __raw_spin_lock(&scd->lock); scd->tick_raw = now; scd->tick_gtod = now_gtod; - __update_sched_clock(scd, now); - __raw_spin_unlock(&scd->lock); + sched_clock_local(scd); } /* diff --git a/kernel/sched_cpupri.c b/kernel/sched_cpupri.c index 1e00bfacf9b..0f052fc674d 100644 --- a/kernel/sched_cpupri.c +++ b/kernel/sched_cpupri.c @@ -55,7 +55,7 @@ static int convert_prio(int prio) * cpupri_find - find the best (lowest-pri) CPU in the system * @cp: The cpupri context * @p: The task - * @lowest_mask: A mask to fill in with selected CPUs + * @lowest_mask: A mask to fill in with selected CPUs (or NULL) * * Note: This function returns the recommended CPUs as calculated during the * current invokation. By the time the call returns, the CPUs may have in @@ -81,7 +81,21 @@ int cpupri_find(struct cpupri *cp, struct task_struct *p, if (cpumask_any_and(&p->cpus_allowed, vec->mask) >= nr_cpu_ids) continue; - cpumask_and(lowest_mask, &p->cpus_allowed, vec->mask); + if (lowest_mask) { + cpumask_and(lowest_mask, &p->cpus_allowed, vec->mask); + + /* + * We have to ensure that we have at least one bit + * still set in the array, since the map could have + * been concurrently emptied between the first and + * second reads of vec->mask. If we hit this + * condition, simply act as though we never hit this + * priority level and continue on. + */ + if (cpumask_any(lowest_mask) >= nr_cpu_ids) + continue; + } + return 1; } @@ -113,21 +127,11 @@ void cpupri_set(struct cpupri *cp, int cpu, int newpri) /* * If the cpu was currently mapped to a different value, we - * first need to unmap the old value + * need to map it to the new value then remove the old value. + * Note, we must add the new value first, otherwise we risk the + * cpu being cleared from pri_active, and this cpu could be + * missed for a push or pull. */ - if (likely(oldpri != CPUPRI_INVALID)) { - struct cpupri_vec *vec = &cp->pri_to_cpu[oldpri]; - - spin_lock_irqsave(&vec->lock, flags); - - vec->count--; - if (!vec->count) - clear_bit(oldpri, cp->pri_active); - cpumask_clear_cpu(cpu, vec->mask); - - spin_unlock_irqrestore(&vec->lock, flags); - } - if (likely(newpri != CPUPRI_INVALID)) { struct cpupri_vec *vec = &cp->pri_to_cpu[newpri]; @@ -140,6 +144,18 @@ void cpupri_set(struct cpupri *cp, int cpu, int newpri) spin_unlock_irqrestore(&vec->lock, flags); } + if (likely(oldpri != CPUPRI_INVALID)) { + struct cpupri_vec *vec = &cp->pri_to_cpu[oldpri]; + + spin_lock_irqsave(&vec->lock, flags); + + vec->count--; + if (!vec->count) + clear_bit(oldpri, cp->pri_active); + cpumask_clear_cpu(cpu, vec->mask); + + spin_unlock_irqrestore(&vec->lock, flags); + } *currpri = newpri; } @@ -151,10 +167,14 @@ void cpupri_set(struct cpupri *cp, int cpu, int newpri) * * Returns: -ENOMEM if memory fails. */ -int __init_refok cpupri_init(struct cpupri *cp, bool bootmem) +int cpupri_init(struct cpupri *cp, bool bootmem) { + gfp_t gfp = GFP_KERNEL; int i; + if (bootmem) + gfp = GFP_NOWAIT; + memset(cp, 0, sizeof(*cp)); for (i = 0; i < CPUPRI_NR_PRIORITIES; i++) { @@ -162,9 +182,7 @@ int __init_refok cpupri_init(struct cpupri *cp, bool bootmem) spin_lock_init(&vec->lock); vec->count = 0; - if (bootmem) - alloc_bootmem_cpumask_var(&vec->mask); - else if (!alloc_cpumask_var(&vec->mask, GFP_KERNEL)) + if (!zalloc_cpumask_var(&vec->mask, gfp)) goto cleanup; } diff --git a/kernel/sched_debug.c b/kernel/sched_debug.c index 467ca72f165..6988cf08f70 100644 --- a/kernel/sched_debug.c +++ b/kernel/sched_debug.c @@ -162,7 +162,7 @@ void print_cfs_rq(struct seq_file *m, int cpu, struct cfs_rq *cfs_rq) { s64 MIN_vruntime = -1, min_vruntime, max_vruntime = -1, spread, rq0_min_vruntime, spread0; - struct rq *rq = &per_cpu(runqueues, cpu); + struct rq *rq = cpu_rq(cpu); struct sched_entity *last; unsigned long flags; @@ -191,7 +191,7 @@ void print_cfs_rq(struct seq_file *m, int cpu, struct cfs_rq *cfs_rq) if (last) max_vruntime = last->vruntime; min_vruntime = cfs_rq->min_vruntime; - rq0_min_vruntime = per_cpu(runqueues, 0).cfs.min_vruntime; + rq0_min_vruntime = cpu_rq(0)->cfs.min_vruntime; spin_unlock_irqrestore(&rq->lock, flags); SEQ_printf(m, " .%-30s: %Ld.%06ld\n", "MIN_vruntime", SPLIT_NS(MIN_vruntime)); @@ -248,7 +248,7 @@ void print_rt_rq(struct seq_file *m, int cpu, struct rt_rq *rt_rq) static void print_cpu(struct seq_file *m, int cpu) { - struct rq *rq = &per_cpu(runqueues, cpu); + struct rq *rq = cpu_rq(cpu); #ifdef CONFIG_X86 { @@ -285,12 +285,16 @@ static void print_cpu(struct seq_file *m, int cpu) #ifdef CONFIG_SCHEDSTATS #define P(n) SEQ_printf(m, " .%-30s: %d\n", #n, rq->n); +#define P64(n) SEQ_printf(m, " .%-30s: %Ld\n", #n, rq->n); P(yld_count); P(sched_switch); P(sched_count); P(sched_goidle); +#ifdef CONFIG_SMP + P64(avg_idle); +#endif P(ttwu_count); P(ttwu_local); @@ -395,6 +399,7 @@ void proc_sched_show_task(struct task_struct *p, struct seq_file *m) PN(se.sum_exec_runtime); PN(se.avg_overlap); PN(se.avg_wakeup); + PN(se.avg_running); nr_switches = p->nvcsw + p->nivcsw; @@ -409,6 +414,8 @@ void proc_sched_show_task(struct task_struct *p, struct seq_file *m) PN(se.wait_max); PN(se.wait_sum); P(se.wait_count); + PN(se.iowait_sum); + P(se.iowait_count); P(sched_info.bkl_count); P(se.nr_migrations); P(se.nr_migrations_cold); @@ -479,6 +486,8 @@ void proc_sched_set_task(struct task_struct *p) p->se.wait_max = 0; p->se.wait_sum = 0; p->se.wait_count = 0; + p->se.iowait_sum = 0; + p->se.iowait_count = 0; p->se.sleep_max = 0; p->se.sum_sleep_runtime = 0; p->se.block_max = 0; diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c index 3816f217f11..f61837ad336 100644 --- a/kernel/sched_fair.c +++ b/kernel/sched_fair.c @@ -24,7 +24,7 @@ /* * Targeted preemption latency for CPU-bound tasks: - * (default: 20ms * (1 + ilog(ncpus)), units: nanoseconds) + * (default: 5ms * (1 + ilog(ncpus)), units: nanoseconds) * * NOTE: this latency value is not the same as the concept of * 'timeslice length' - timeslices in CFS are of variable length @@ -34,13 +34,13 @@ * (to see the precise effective timeslice length of your workload, * run vmstat and monitor the context-switches (cs) field) */ -unsigned int sysctl_sched_latency = 20000000ULL; +unsigned int sysctl_sched_latency = 5000000ULL; /* * Minimal preemption granularity for CPU-bound tasks: - * (default: 4 msec * (1 + ilog(ncpus)), units: nanoseconds) + * (default: 1 msec * (1 + ilog(ncpus)), units: nanoseconds) */ -unsigned int sysctl_sched_min_granularity = 4000000ULL; +unsigned int sysctl_sched_min_granularity = 1000000ULL; /* * is kept at sysctl_sched_latency / sysctl_sched_min_granularity @@ -48,10 +48,10 @@ unsigned int sysctl_sched_min_granularity = 4000000ULL; static unsigned int sched_nr_latency = 5; /* - * After fork, child runs first. (default) If set to 0 then + * After fork, child runs first. If set to 0 (default) then * parent will (try to) run first. */ -const_debug unsigned int sysctl_sched_child_runs_first = 1; +unsigned int sysctl_sched_child_runs_first __read_mostly; /* * sys_sched_yield() compat mode @@ -63,13 +63,13 @@ unsigned int __read_mostly sysctl_sched_compat_yield; /* * SCHED_OTHER wake-up granularity. - * (default: 5 msec * (1 + ilog(ncpus)), units: nanoseconds) + * (default: 1 msec * (1 + ilog(ncpus)), units: nanoseconds) * * This option delays the preemption effects of decoupled workloads * and reduces their over-scheduling. Synchronous workloads will still * have immediate wakeup/sleep latencies. */ -unsigned int sysctl_sched_wakeup_granularity = 5000000UL; +unsigned int sysctl_sched_wakeup_granularity = 1000000UL; const_debug unsigned int sysctl_sched_migration_cost = 500000UL; @@ -79,11 +79,6 @@ static const struct sched_class fair_sched_class; * CFS operations on generic schedulable entities: */ -static inline struct task_struct *task_of(struct sched_entity *se) -{ - return container_of(se, struct task_struct, se); -} - #ifdef CONFIG_FAIR_GROUP_SCHED /* cpu runqueue to which this cfs_rq is attached */ @@ -95,6 +90,14 @@ static inline struct rq *rq_of(struct cfs_rq *cfs_rq) /* An entity is a task if it doesn't "own" a runqueue */ #define entity_is_task(se) (!se->my_q) +static inline struct task_struct *task_of(struct sched_entity *se) +{ +#ifdef CONFIG_SCHED_DEBUG + WARN_ON_ONCE(!entity_is_task(se)); +#endif + return container_of(se, struct task_struct, se); +} + /* Walk up scheduling entities hierarchy */ #define for_each_sched_entity(se) \ for (; se; se = se->parent) @@ -186,7 +189,12 @@ find_matching_se(struct sched_entity **se, struct sched_entity **pse) } } -#else /* CONFIG_FAIR_GROUP_SCHED */ +#else /* !CONFIG_FAIR_GROUP_SCHED */ + +static inline struct task_struct *task_of(struct sched_entity *se) +{ + return container_of(se, struct task_struct, se); +} static inline struct rq *rq_of(struct cfs_rq *cfs_rq) { @@ -266,6 +274,12 @@ static inline u64 min_vruntime(u64 min_vruntime, u64 vruntime) return min_vruntime; } +static inline int entity_before(struct sched_entity *a, + struct sched_entity *b) +{ + return (s64)(a->vruntime - b->vruntime) < 0; +} + static inline s64 entity_key(struct cfs_rq *cfs_rq, struct sched_entity *se) { return se->vruntime - cfs_rq->min_vruntime; @@ -370,10 +384,10 @@ static struct sched_entity *__pick_last_entity(struct cfs_rq *cfs_rq) #ifdef CONFIG_SCHED_DEBUG int sched_nr_latency_handler(struct ctl_table *table, int write, - struct file *filp, void __user *buffer, size_t *lenp, + void __user *buffer, size_t *lenp, loff_t *ppos) { - int ret = proc_dointvec_minmax(table, write, filp, buffer, lenp, ppos); + int ret = proc_dointvec_minmax(table, write, buffer, lenp, ppos); if (ret || !write) return ret; @@ -430,12 +444,13 @@ static u64 sched_slice(struct cfs_rq *cfs_rq, struct sched_entity *se) for_each_sched_entity(se) { struct load_weight *load; + struct load_weight lw; cfs_rq = cfs_rq_of(se); load = &cfs_rq->load; if (unlikely(!se->on_rq)) { - struct load_weight lw = cfs_rq->load; + lw = cfs_rq->load; update_load_add(&lw, se->load.weight); load = &lw; @@ -498,6 +513,7 @@ static void update_curr(struct cfs_rq *cfs_rq) if (entity_is_task(curr)) { struct task_struct *curtask = task_of(curr); + trace_sched_stat_runtime(curtask, delta_exec, curr->vruntime); cpuacct_charge(curtask, delta_exec); account_group_exec_runtime(curtask, delta_exec); } @@ -530,6 +546,12 @@ update_stats_wait_end(struct cfs_rq *cfs_rq, struct sched_entity *se) schedstat_set(se->wait_count, se->wait_count + 1); schedstat_set(se->wait_sum, se->wait_sum + rq_of(cfs_rq)->clock - se->wait_start); +#ifdef CONFIG_SCHEDSTATS + if (entity_is_task(se)) { + trace_sched_stat_wait(task_of(se), + rq_of(cfs_rq)->clock - se->wait_start); + } +#endif schedstat_set(se->wait_start, 0); } @@ -604,9 +626,13 @@ account_entity_dequeue(struct cfs_rq *cfs_rq, struct sched_entity *se) static void enqueue_sleeper(struct cfs_rq *cfs_rq, struct sched_entity *se) { #ifdef CONFIG_SCHEDSTATS + struct task_struct *tsk = NULL; + + if (entity_is_task(se)) + tsk = task_of(se); + if (se->sleep_start) { u64 delta = rq_of(cfs_rq)->clock - se->sleep_start; - struct task_struct *tsk = task_of(se); if ((s64)delta < 0) delta = 0; @@ -617,11 +643,13 @@ static void enqueue_sleeper(struct cfs_rq *cfs_rq, struct sched_entity *se) se->sleep_start = 0; se->sum_sleep_runtime += delta; - account_scheduler_latency(tsk, delta >> 10, 1); + if (tsk) { + account_scheduler_latency(tsk, delta >> 10, 1); + trace_sched_stat_sleep(tsk, delta); + } } if (se->block_start) { u64 delta = rq_of(cfs_rq)->clock - se->block_start; - struct task_struct *tsk = task_of(se); if ((s64)delta < 0) delta = 0; @@ -632,17 +660,25 @@ static void enqueue_sleeper(struct cfs_rq *cfs_rq, struct sched_entity *se) se->block_start = 0; se->sum_sleep_runtime += delta; - /* - * Blocking time is in units of nanosecs, so shift by 20 to - * get a milliseconds-range estimation of the amount of - * time that the task spent sleeping: - */ - if (unlikely(prof_on == SLEEP_PROFILING)) { + if (tsk) { + if (tsk->in_iowait) { + se->iowait_sum += delta; + se->iowait_count++; + trace_sched_stat_iowait(tsk, delta); + } - profile_hits(SLEEP_PROFILING, (void *)get_wchan(tsk), - delta >> 20); + /* + * Blocking time is in units of nanosecs, so shift by + * 20 to get a milliseconds-range estimation of the + * amount of time that the task spent sleeping: + */ + if (unlikely(prof_on == SLEEP_PROFILING)) { + profile_hits(SLEEP_PROFILING, + (void *)get_wchan(tsk), + delta >> 20); + } + account_scheduler_latency(tsk, delta >> 10, 0); } - account_scheduler_latency(tsk, delta >> 10, 0); } #endif } @@ -674,28 +710,33 @@ place_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int initial) if (initial && sched_feat(START_DEBIT)) vruntime += sched_vslice(cfs_rq, se); - if (!initial) { - /* sleeps upto a single latency don't count. */ - if (sched_feat(NEW_FAIR_SLEEPERS)) { - unsigned long thresh = sysctl_sched_latency; + /* sleeps up to a single latency don't count. */ + if (!initial && sched_feat(FAIR_SLEEPERS)) { + unsigned long thresh = sysctl_sched_latency; - /* - * Convert the sleeper threshold into virtual time. - * SCHED_IDLE is a special sub-class. We care about - * fairness only relative to other SCHED_IDLE tasks, - * all of which have the same weight. - */ - if (sched_feat(NORMALIZED_SLEEPER) && - task_of(se)->policy != SCHED_IDLE) - thresh = calc_delta_fair(thresh, se); + /* + * Convert the sleeper threshold into virtual time. + * SCHED_IDLE is a special sub-class. We care about + * fairness only relative to other SCHED_IDLE tasks, + * all of which have the same weight. + */ + if (sched_feat(NORMALIZED_SLEEPER) && (!entity_is_task(se) || + task_of(se)->policy != SCHED_IDLE)) + thresh = calc_delta_fair(thresh, se); - vruntime -= thresh; - } + /* + * Halve their sleep time's effect, to allow + * for a gentler effect of sleepers: + */ + if (sched_feat(GENTLE_FAIR_SLEEPERS)) + thresh >>= 1; - /* ensure we never gain time by being placed backwards. */ - vruntime = max_vruntime(se->vruntime, vruntime); + vruntime -= thresh; } + /* ensure we never gain time by being placed backwards. */ + vruntime = max_vruntime(se->vruntime, vruntime); + se->vruntime = vruntime; } @@ -721,10 +762,10 @@ enqueue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int wakeup) static void __clear_buddies(struct cfs_rq *cfs_rq, struct sched_entity *se) { - if (cfs_rq->last == se) + if (!se || cfs_rq->last == se) cfs_rq->last = NULL; - if (cfs_rq->next == se) + if (!se || cfs_rq->next == se) cfs_rq->next = NULL; } @@ -781,6 +822,26 @@ check_preempt_tick(struct cfs_rq *cfs_rq, struct sched_entity *curr) * re-elected due to buddy favours. */ clear_buddies(cfs_rq, curr); + return; + } + + /* + * Ensure that a task that missed wakeup preemption by a + * narrow margin doesn't have to wait for a full slice. + * This also mitigates buddy induced latencies under load. + */ + if (!sched_feat(WAKEUP_PREEMPT)) + return; + + if (delta_exec < sysctl_sched_min_granularity) + return; + + if (cfs_rq->nr_running > 1) { + struct sched_entity *se = __pick_next_entity(cfs_rq); + s64 delta = curr->vruntime - se->vruntime; + + if (delta > ideal_runtime) + resched_task(rq_of(cfs_rq)->curr); } } @@ -820,12 +881,18 @@ wakeup_preempt_entity(struct sched_entity *curr, struct sched_entity *se); static struct sched_entity *pick_next_entity(struct cfs_rq *cfs_rq) { struct sched_entity *se = __pick_next_entity(cfs_rq); + struct sched_entity *left = se; - if (cfs_rq->next && wakeup_preempt_entity(cfs_rq->next, se) < 1) - return cfs_rq->next; + if (cfs_rq->next && wakeup_preempt_entity(cfs_rq->next, left) < 1) + se = cfs_rq->next; - if (cfs_rq->last && wakeup_preempt_entity(cfs_rq->last, se) < 1) - return cfs_rq->last; + /* + * Prefer last buddy, try to return the CPU to a preempted task. + */ + if (cfs_rq->last && wakeup_preempt_entity(cfs_rq->last, left) < 1) + se = cfs_rq->last; + + clear_buddies(cfs_rq, se); return se; } @@ -1015,7 +1082,7 @@ static void yield_task_fair(struct rq *rq) /* * Already in the rightmost position? */ - if (unlikely(!rightmost || rightmost->vruntime < se->vruntime)) + if (unlikely(!rightmost || entity_before(rightmost, se))) return; /* @@ -1026,79 +1093,6 @@ static void yield_task_fair(struct rq *rq) se->vruntime = rightmost->vruntime + 1; } -/* - * wake_idle() will wake a task on an idle cpu if task->cpu is - * not idle and an idle cpu is available. The span of cpus to - * search starts with cpus closest then further out as needed, - * so we always favor a closer, idle cpu. - * Domains may include CPUs that are not usable for migration, - * hence we need to mask them out (cpu_active_mask) - * - * Returns the CPU we should wake onto. - */ -#if defined(ARCH_HAS_SCHED_WAKE_IDLE) -static int wake_idle(int cpu, struct task_struct *p) -{ - struct sched_domain *sd; - int i; - unsigned int chosen_wakeup_cpu; - int this_cpu; - - /* - * At POWERSAVINGS_BALANCE_WAKEUP level, if both this_cpu and prev_cpu - * are idle and this is not a kernel thread and this task's affinity - * allows it to be moved to preferred cpu, then just move! - */ - - this_cpu = smp_processor_id(); - chosen_wakeup_cpu = - cpu_rq(this_cpu)->rd->sched_mc_preferred_wakeup_cpu; - - if (sched_mc_power_savings >= POWERSAVINGS_BALANCE_WAKEUP && - idle_cpu(cpu) && idle_cpu(this_cpu) && - p->mm && !(p->flags & PF_KTHREAD) && - cpu_isset(chosen_wakeup_cpu, p->cpus_allowed)) - return chosen_wakeup_cpu; - - /* - * If it is idle, then it is the best cpu to run this task. - * - * This cpu is also the best, if it has more than one task already. - * Siblings must be also busy(in most cases) as they didn't already - * pickup the extra load from this cpu and hence we need not check - * sibling runqueue info. This will avoid the checks and cache miss - * penalities associated with that. - */ - if (idle_cpu(cpu) || cpu_rq(cpu)->cfs.nr_running > 1) - return cpu; - - for_each_domain(cpu, sd) { - if ((sd->flags & SD_WAKE_IDLE) - || ((sd->flags & SD_WAKE_IDLE_FAR) - && !task_hot(p, task_rq(p)->clock, sd))) { - for_each_cpu_and(i, sched_domain_span(sd), - &p->cpus_allowed) { - if (cpu_active(i) && idle_cpu(i)) { - if (i != task_cpu(p)) { - schedstat_inc(p, - se.nr_wakeups_idle); - } - return i; - } - } - } else { - break; - } - } - return cpu; -} -#else /* !ARCH_HAS_SCHED_WAKE_IDLE*/ -static inline int wake_idle(int cpu, struct task_struct *p) -{ - return cpu; -} -#endif - #ifdef CONFIG_SMP #ifdef CONFIG_FAIR_GROUP_SCHED @@ -1185,25 +1179,34 @@ static inline unsigned long effective_load(struct task_group *tg, int cpu, #endif -static int -wake_affine(struct sched_domain *this_sd, struct rq *this_rq, - struct task_struct *p, int prev_cpu, int this_cpu, int sync, - int idx, unsigned long load, unsigned long this_load, - unsigned int imbalance) +static int wake_affine(struct sched_domain *sd, struct task_struct *p, int sync) { - struct task_struct *curr = this_rq->curr; - struct task_group *tg; - unsigned long tl = this_load; + struct task_struct *curr = current; + unsigned long this_load, load; + int idx, this_cpu, prev_cpu; unsigned long tl_per_task; + unsigned int imbalance; + struct task_group *tg; unsigned long weight; int balanced; - if (!(this_sd->flags & SD_WAKE_AFFINE) || !sched_feat(AFFINE_WAKEUPS)) - return 0; + idx = sd->wake_idx; + this_cpu = smp_processor_id(); + prev_cpu = task_cpu(p); + load = source_load(prev_cpu, idx); + this_load = target_load(this_cpu, idx); - if (sync && (curr->se.avg_overlap > sysctl_sched_migration_cost || - p->se.avg_overlap > sysctl_sched_migration_cost)) - sync = 0; + if (sync) { + if (sched_feat(SYNC_LESS) && + (curr->se.avg_overlap > sysctl_sched_migration_cost || + p->se.avg_overlap > sysctl_sched_migration_cost)) + sync = 0; + } else { + if (sched_feat(SYNC_MORE) && + (curr->se.avg_overlap < sysctl_sched_migration_cost && + p->se.avg_overlap < sysctl_sched_migration_cost)) + sync = 1; + } /* * If sync wakeup then subtract the (maximum possible) @@ -1214,14 +1217,26 @@ wake_affine(struct sched_domain *this_sd, struct rq *this_rq, tg = task_group(current); weight = current->se.load.weight; - tl += effective_load(tg, this_cpu, -weight, -weight); + this_load += effective_load(tg, this_cpu, -weight, -weight); load += effective_load(tg, prev_cpu, 0, -weight); } tg = task_group(p); weight = p->se.load.weight; - balanced = 100*(tl + effective_load(tg, this_cpu, weight, weight)) <= + imbalance = 100 + (sd->imbalance_pct - 100) / 2; + + /* + * In low-load situations, where prev_cpu is idle and this_cpu is idle + * due to the sync cause above having dropped this_load to 0, we'll + * always have an imbalance, but there's really nothing you can do + * about that, so that's good too. + * + * Otherwise check if either cpus are near enough in load to allow this + * task to be woken on this_cpu. + */ + balanced = !this_load || + 100*(this_load + effective_load(tg, this_cpu, weight, weight)) <= imbalance*(load + effective_load(tg, prev_cpu, 0, weight)); /* @@ -1235,14 +1250,15 @@ wake_affine(struct sched_domain *this_sd, struct rq *this_rq, schedstat_inc(p, se.nr_wakeups_affine_attempts); tl_per_task = cpu_avg_load_per_task(this_cpu); - if (balanced || (tl <= load && tl + target_load(prev_cpu, idx) <= - tl_per_task)) { + if (balanced || + (this_load <= load && + this_load + target_load(prev_cpu, idx) <= tl_per_task)) { /* * This domain has SD_WAKE_AFFINE and * p is cache cold in this domain, and * there is no bad imbalance. */ - schedstat_inc(this_sd, ttwu_move_affine); + schedstat_inc(sd, ttwu_move_affine); schedstat_inc(p, se.nr_wakeups_affine); return 1; @@ -1250,67 +1266,271 @@ wake_affine(struct sched_domain *this_sd, struct rq *this_rq, return 0; } -static int select_task_rq_fair(struct task_struct *p, int sync) +/* + * find_idlest_group finds and returns the least busy CPU group within the + * domain. + */ +static struct sched_group * +find_idlest_group(struct sched_domain *sd, struct task_struct *p, + int this_cpu, int load_idx) { - struct sched_domain *sd, *this_sd = NULL; - int prev_cpu, this_cpu, new_cpu; - unsigned long load, this_load; - struct rq *this_rq; - unsigned int imbalance; - int idx; + struct sched_group *idlest = NULL, *this = NULL, *group = sd->groups; + unsigned long min_load = ULONG_MAX, this_load = 0; + int imbalance = 100 + (sd->imbalance_pct-100)/2; - prev_cpu = task_cpu(p); - this_cpu = smp_processor_id(); - this_rq = cpu_rq(this_cpu); - new_cpu = prev_cpu; + do { + unsigned long load, avg_load; + int local_group; + int i; + + /* Skip over this group if it has no CPUs allowed */ + if (!cpumask_intersects(sched_group_cpus(group), + &p->cpus_allowed)) + continue; + + local_group = cpumask_test_cpu(this_cpu, + sched_group_cpus(group)); + + /* Tally up the load of all CPUs in the group */ + avg_load = 0; + + for_each_cpu(i, sched_group_cpus(group)) { + /* Bias balancing toward cpus of our domain */ + if (local_group) + load = source_load(i, load_idx); + else + load = target_load(i, load_idx); + + avg_load += load; + } + + /* Adjust by relative CPU power of the group */ + avg_load = (avg_load * SCHED_LOAD_SCALE) / group->cpu_power; + + if (local_group) { + this_load = avg_load; + this = group; + } else if (avg_load < min_load) { + min_load = avg_load; + idlest = group; + } + } while (group = group->next, group != sd->groups); + + if (!idlest || 100*this_load < imbalance*min_load) + return NULL; + return idlest; +} + +/* + * find_idlest_cpu - find the idlest cpu among the cpus in group. + */ +static int +find_idlest_cpu(struct sched_group *group, struct task_struct *p, int this_cpu) +{ + unsigned long load, min_load = ULONG_MAX; + int idlest = -1; + int i; + + /* Traverse only the allowed CPUs */ + for_each_cpu_and(i, sched_group_cpus(group), &p->cpus_allowed) { + load = weighted_cpuload(i); + + if (load < min_load || (load == min_load && i == this_cpu)) { + min_load = load; + idlest = i; + } + } + + return idlest; +} + +/* + * Try and locate an idle CPU in the sched_domain. + */ +static int +select_idle_sibling(struct task_struct *p, struct sched_domain *sd, int target) +{ + int cpu = smp_processor_id(); + int prev_cpu = task_cpu(p); + int i; + + /* + * If this domain spans both cpu and prev_cpu (see the SD_WAKE_AFFINE + * test in select_task_rq_fair) and the prev_cpu is idle then that's + * always a better target than the current cpu. + */ + if (target == cpu && !cpu_rq(prev_cpu)->cfs.nr_running) + return prev_cpu; - if (prev_cpu == this_cpu) - goto out; /* - * 'this_sd' is the first domain that both - * this_cpu and prev_cpu are present in: + * Otherwise, iterate the domain and find an elegible idle cpu. */ - for_each_domain(this_cpu, sd) { - if (cpumask_test_cpu(prev_cpu, sched_domain_span(sd))) { - this_sd = sd; + for_each_cpu_and(i, sched_domain_span(sd), &p->cpus_allowed) { + if (!cpu_rq(i)->cfs.nr_running) { + target = i; break; } } - if (unlikely(!cpumask_test_cpu(this_cpu, &p->cpus_allowed))) - goto out; + return target; +} - /* - * Check for affine wakeup and passive balancing possibilities. - */ - if (!this_sd) +/* + * sched_balance_self: balance the current task (running on cpu) in domains + * that have the 'flag' flag set. In practice, this is SD_BALANCE_FORK and + * SD_BALANCE_EXEC. + * + * Balance, ie. select the least loaded group. + * + * Returns the target CPU number, or the same CPU if no balancing is needed. + * + * preempt must be disabled. + */ +static int select_task_rq_fair(struct task_struct *p, int sd_flag, int wake_flags) +{ + struct sched_domain *tmp, *affine_sd = NULL, *sd = NULL; + int cpu = smp_processor_id(); + int prev_cpu = task_cpu(p); + int new_cpu = cpu; + int want_affine = 0; + int want_sd = 1; + int sync = wake_flags & WF_SYNC; + + if (sd_flag & SD_BALANCE_WAKE) { + if (sched_feat(AFFINE_WAKEUPS) && + cpumask_test_cpu(cpu, &p->cpus_allowed)) + want_affine = 1; + new_cpu = prev_cpu; + } + + rcu_read_lock(); + for_each_domain(cpu, tmp) { + /* + * If power savings logic is enabled for a domain, see if we + * are not overloaded, if so, don't balance wider. + */ + if (tmp->flags & (SD_POWERSAVINGS_BALANCE|SD_PREFER_LOCAL)) { + unsigned long power = 0; + unsigned long nr_running = 0; + unsigned long capacity; + int i; + + for_each_cpu(i, sched_domain_span(tmp)) { + power += power_of(i); + nr_running += cpu_rq(i)->cfs.nr_running; + } + + capacity = DIV_ROUND_CLOSEST(power, SCHED_LOAD_SCALE); + + if (tmp->flags & SD_POWERSAVINGS_BALANCE) + nr_running /= 2; + + if (nr_running < capacity) + want_sd = 0; + } + + /* + * While iterating the domains looking for a spanning + * WAKE_AFFINE domain, adjust the affine target to any idle cpu + * in cache sharing domains along the way. + */ + if (want_affine) { + int target = -1; + + /* + * If both cpu and prev_cpu are part of this domain, + * cpu is a valid SD_WAKE_AFFINE target. + */ + if (cpumask_test_cpu(prev_cpu, sched_domain_span(tmp))) + target = cpu; + + /* + * If there's an idle sibling in this domain, make that + * the wake_affine target instead of the current cpu. + */ + if (tmp->flags & SD_PREFER_SIBLING) + target = select_idle_sibling(p, tmp, target); + + if (target >= 0) { + if (tmp->flags & SD_WAKE_AFFINE) { + affine_sd = tmp; + want_affine = 0; + } + cpu = target; + } + } + + if (!want_sd && !want_affine) + break; + + if (!(tmp->flags & sd_flag)) + continue; + + if (want_sd) + sd = tmp; + } + + if (sched_feat(LB_SHARES_UPDATE)) { + /* + * Pick the largest domain to update shares over + */ + tmp = sd; + if (affine_sd && (!tmp || + cpumask_weight(sched_domain_span(affine_sd)) > + cpumask_weight(sched_domain_span(sd)))) + tmp = affine_sd; + + if (tmp) + update_shares(tmp); + } + + if (affine_sd && wake_affine(affine_sd, p, sync)) { + new_cpu = cpu; goto out; + } - idx = this_sd->wake_idx; + while (sd) { + int load_idx = sd->forkexec_idx; + struct sched_group *group; + int weight; - imbalance = 100 + (this_sd->imbalance_pct - 100) / 2; + if (!(sd->flags & sd_flag)) { + sd = sd->child; + continue; + } - load = source_load(prev_cpu, idx); - this_load = target_load(this_cpu, idx); + if (sd_flag & SD_BALANCE_WAKE) + load_idx = sd->wake_idx; - if (wake_affine(this_sd, this_rq, p, prev_cpu, this_cpu, sync, idx, - load, this_load, imbalance)) - return this_cpu; + group = find_idlest_group(sd, p, cpu, load_idx); + if (!group) { + sd = sd->child; + continue; + } - /* - * Start passive balancing when half the imbalance_pct - * limit is reached. - */ - if (this_sd->flags & SD_WAKE_BALANCE) { - if (imbalance*this_load <= 100*load) { - schedstat_inc(this_sd, ttwu_move_balance); - schedstat_inc(p, se.nr_wakeups_passive); - return this_cpu; + new_cpu = find_idlest_cpu(group, p, cpu); + if (new_cpu == -1 || new_cpu == cpu) { + /* Now try balancing at a lower domain level of cpu */ + sd = sd->child; + continue; + } + + /* Now try balancing at a lower domain level of new_cpu */ + cpu = new_cpu; + weight = cpumask_weight(sched_domain_span(sd)); + sd = NULL; + for_each_domain(cpu, tmp) { + if (weight <= cpumask_weight(sched_domain_span(tmp))) + break; + if (tmp->flags & sd_flag) + sd = tmp; } + /* while loop will break here if sd == NULL */ } out: - return wake_idle(new_cpu, p); + rcu_read_unlock(); + return new_cpu; } #endif /* CONFIG_SMP */ @@ -1423,11 +1643,13 @@ static void set_next_buddy(struct sched_entity *se) /* * Preempt the current task with a newly woken task if needed: */ -static void check_preempt_wakeup(struct rq *rq, struct task_struct *p, int sync) +static void check_preempt_wakeup(struct rq *rq, struct task_struct *p, int wake_flags) { struct task_struct *curr = rq->curr; struct sched_entity *se = &curr->se, *pse = &p->se; struct cfs_rq *cfs_rq = task_cfs_rq(curr); + int sync = wake_flags & WF_SYNC; + int scale = cfs_rq->nr_running >= sched_nr_latency; update_curr(cfs_rq); @@ -1442,18 +1664,8 @@ static void check_preempt_wakeup(struct rq *rq, struct task_struct *p, int sync) if (unlikely(se == pse)) return; - /* - * Only set the backward buddy when the current task is still on the - * rq. This can happen when a wakeup gets interleaved with schedule on - * the ->pre_schedule() or idle_balance() point, either of which can - * drop the rq lock. - * - * Also, during early boot the idle thread is in the fair class, for - * obvious reasons its a bad idea to schedule back to the idle thread. - */ - if (sched_feat(LAST_BUDDY) && likely(se->on_rq && curr != rq->idle)) - set_last_buddy(se); - set_next_buddy(pse); + if (sched_feat(NEXT_BUDDY) && scale && !(wake_flags & WF_FORK)) + set_next_buddy(pse); /* * We can come here with TIF_NEED_RESCHED already set from new task @@ -1475,28 +1687,44 @@ static void check_preempt_wakeup(struct rq *rq, struct task_struct *p, int sync) return; } - if (!sched_feat(WAKEUP_PREEMPT)) - return; - - if (sched_feat(WAKEUP_OVERLAP) && (sync || - (se->avg_overlap < sysctl_sched_migration_cost && - pse->avg_overlap < sysctl_sched_migration_cost))) { + if ((sched_feat(WAKEUP_SYNC) && sync) || + (sched_feat(WAKEUP_OVERLAP) && + (se->avg_overlap < sysctl_sched_migration_cost && + pse->avg_overlap < sysctl_sched_migration_cost))) { resched_task(curr); return; } - find_matching_se(&se, &pse); - - while (se) { - BUG_ON(!pse); - - if (wakeup_preempt_entity(se, pse) == 1) { + if (sched_feat(WAKEUP_RUNNING)) { + if (pse->avg_running < se->avg_running) { + set_next_buddy(pse); resched_task(curr); - break; + return; } + } + + if (!sched_feat(WAKEUP_PREEMPT)) + return; + + find_matching_se(&se, &pse); - se = parent_entity(se); - pse = parent_entity(pse); + BUG_ON(!pse); + + if (wakeup_preempt_entity(se, pse) == 1) { + resched_task(curr); + /* + * Only set the backward buddy when the current task is still + * on the rq. This can happen when a wakeup gets interleaved + * with schedule on the ->pre_schedule() or idle_balance() + * point, either of which can * drop the rq lock. + * + * Also, during early boot the idle thread is in the fair class, + * for obvious reasons its a bad idea to schedule back to it. + */ + if (unlikely(!se->on_rq || curr == rq->idle)) + return; + if (sched_feat(LAST_BUDDY) && scale && entity_is_task(se)) + set_last_buddy(se); } } @@ -1506,16 +1734,11 @@ static struct task_struct *pick_next_task_fair(struct rq *rq) struct cfs_rq *cfs_rq = &rq->cfs; struct sched_entity *se; - if (unlikely(!cfs_rq->nr_running)) + if (!cfs_rq->nr_running) return NULL; do { se = pick_next_entity(cfs_rq); - /* - * If se was a buddy, clear it so that it will have to earn - * the favour again. - */ - __clear_buddies(cfs_rq, se); set_next_entity(cfs_rq, se); cfs_rq = group_cfs_rq(se); } while (cfs_rq); @@ -1714,11 +1937,13 @@ static void task_new_fair(struct rq *rq, struct task_struct *p) sched_info_queued(p); update_curr(cfs_rq); + if (curr) + se->vruntime = curr->vruntime; place_entity(cfs_rq, se, 1); /* 'curr' will be NULL if the child belongs to a different group */ if (sysctl_sched_child_runs_first && this_cpu == task_cpu(p) && - curr && curr->vruntime < se->vruntime) { + curr && entity_before(curr, se)) { /* * Upon rescheduling, sched_class::put_prev_task() will place * 'current' within the tree based on its new key value. @@ -1789,6 +2014,25 @@ static void moved_group_fair(struct task_struct *p) } #endif +unsigned int get_rr_interval_fair(struct task_struct *task) +{ + struct sched_entity *se = &task->se; + unsigned long flags; + struct rq *rq; + unsigned int rr_interval = 0; + + /* + * Time slice is 0 for SCHED_OTHER tasks that are on an otherwise + * idle runqueue: + */ + rq = task_rq_lock(task, &flags); + if (rq->cfs.load.weight) + rr_interval = NS_TO_JIFFIES(sched_slice(&rq->cfs, se)); + task_rq_unlock(rq, &flags); + + return rr_interval; +} + /* * All the scheduling class methods: */ @@ -1817,6 +2061,8 @@ static const struct sched_class fair_sched_class = { .prio_changed = prio_changed_fair, .switched_to = switched_to_fair, + .get_rr_interval = get_rr_interval_fair, + #ifdef CONFIG_FAIR_GROUP_SCHED .moved_group = moved_group_fair, #endif diff --git a/kernel/sched_features.h b/kernel/sched_features.h index 4569bfa7df9..0d94083582c 100644 --- a/kernel/sched_features.h +++ b/kernel/sched_features.h @@ -1,17 +1,123 @@ -SCHED_FEAT(NEW_FAIR_SLEEPERS, 1) +/* + * Disregards a certain amount of sleep time (sched_latency_ns) and + * considers the task to be running during that period. This gives it + * a service deficit on wakeup, allowing it to run sooner. + */ +SCHED_FEAT(FAIR_SLEEPERS, 1) + +/* + * Only give sleepers 50% of their service deficit. This allows + * them to run sooner, but does not allow tons of sleepers to + * rip the spread apart. + */ +SCHED_FEAT(GENTLE_FAIR_SLEEPERS, 1) + +/* + * By not normalizing the sleep time, heavy tasks get an effective + * longer period, and lighter task an effective shorter period they + * are considered running. + */ SCHED_FEAT(NORMALIZED_SLEEPER, 0) -SCHED_FEAT(ADAPTIVE_GRAN, 1) -SCHED_FEAT(WAKEUP_PREEMPT, 1) + +/* + * Place new tasks ahead so that they do not starve already running + * tasks + */ SCHED_FEAT(START_DEBIT, 1) + +/* + * Should wakeups try to preempt running tasks. + */ +SCHED_FEAT(WAKEUP_PREEMPT, 1) + +/* + * Compute wakeup_gran based on task behaviour, clipped to + * [0, sched_wakeup_gran_ns] + */ +SCHED_FEAT(ADAPTIVE_GRAN, 1) + +/* + * When converting the wakeup granularity to virtual time, do it such + * that heavier tasks preempting a lighter task have an edge. + */ +SCHED_FEAT(ASYM_GRAN, 1) + +/* + * Always wakeup-preempt SYNC wakeups, see SYNC_WAKEUPS. + */ +SCHED_FEAT(WAKEUP_SYNC, 0) + +/* + * Wakeup preempt based on task behaviour. Tasks that do not overlap + * don't get preempted. + */ +SCHED_FEAT(WAKEUP_OVERLAP, 0) + +/* + * Wakeup preemption towards tasks that run short + */ +SCHED_FEAT(WAKEUP_RUNNING, 0) + +/* + * Use the SYNC wakeup hint, pipes and the likes use this to indicate + * the remote end is likely to consume the data we just wrote, and + * therefore has cache benefit from being placed on the same cpu, see + * also AFFINE_WAKEUPS. + */ +SCHED_FEAT(SYNC_WAKEUPS, 1) + +/* + * Based on load and program behaviour, see if it makes sense to place + * a newly woken task on the same cpu as the task that woke it -- + * improve cache locality. Typically used with SYNC wakeups as + * generated by pipes and the like, see also SYNC_WAKEUPS. + */ SCHED_FEAT(AFFINE_WAKEUPS, 1) + +/* + * Weaken SYNC hint based on overlap + */ +SCHED_FEAT(SYNC_LESS, 1) + +/* + * Add SYNC hint based on overlap + */ +SCHED_FEAT(SYNC_MORE, 0) + +/* + * Prefer to schedule the task we woke last (assuming it failed + * wakeup-preemption), since its likely going to consume data we + * touched, increases cache locality. + */ +SCHED_FEAT(NEXT_BUDDY, 0) + +/* + * Prefer to schedule the task that ran last (when we did + * wake-preempt) as that likely will touch the same data, increases + * cache locality. + */ +SCHED_FEAT(LAST_BUDDY, 1) + +/* + * Consider buddies to be cache hot, decreases the likelyness of a + * cache buddy being migrated away, increases cache locality. + */ SCHED_FEAT(CACHE_HOT_BUDDY, 1) -SCHED_FEAT(SYNC_WAKEUPS, 1) + +/* + * Use arch dependent cpu power functions + */ +SCHED_FEAT(ARCH_POWER, 0) + SCHED_FEAT(HRTICK, 0) SCHED_FEAT(DOUBLE_TICK, 0) -SCHED_FEAT(ASYM_GRAN, 1) SCHED_FEAT(LB_BIAS, 1) -SCHED_FEAT(LB_WAKEUP_UPDATE, 1) +SCHED_FEAT(LB_SHARES_UPDATE, 1) SCHED_FEAT(ASYM_EFF_LOAD, 1) -SCHED_FEAT(WAKEUP_OVERLAP, 0) -SCHED_FEAT(LAST_BUDDY, 1) + +/* + * Spin-wait on mutex acquisition when the mutex owner is running on + * another cpu -- assumes that when the owner is running, it will soon + * release the lock. Decreases scheduling overhead. + */ SCHED_FEAT(OWNER_SPIN, 1) diff --git a/kernel/sched_idletask.c b/kernel/sched_idletask.c index 8a21a2e28c1..b133a28fcde 100644 --- a/kernel/sched_idletask.c +++ b/kernel/sched_idletask.c @@ -6,7 +6,7 @@ */ #ifdef CONFIG_SMP -static int select_task_rq_idle(struct task_struct *p, int sync) +static int select_task_rq_idle(struct task_struct *p, int sd_flag, int flags) { return task_cpu(p); /* IDLE tasks as never migrated */ } @@ -14,7 +14,7 @@ static int select_task_rq_idle(struct task_struct *p, int sync) /* * Idle tasks are unconditionally rescheduled: */ -static void check_preempt_curr_idle(struct rq *rq, struct task_struct *p, int sync) +static void check_preempt_curr_idle(struct rq *rq, struct task_struct *p, int flags) { resched_task(rq->idle); } @@ -22,7 +22,8 @@ static void check_preempt_curr_idle(struct rq *rq, struct task_struct *p, int sy static struct task_struct *pick_next_task_idle(struct rq *rq) { schedstat_inc(rq, sched_goidle); - + /* adjust the active tasks as we might go into a long sleep */ + calc_load_account_active(rq); return rq->idle; } @@ -96,6 +97,11 @@ static void prio_changed_idle(struct rq *rq, struct task_struct *p, check_preempt_curr(rq, p, 0); } +unsigned int get_rr_interval_idle(struct task_struct *task) +{ + return 0; +} + /* * Simple, special scheduling class for the per-CPU idle tasks: */ @@ -121,6 +127,8 @@ static const struct sched_class idle_sched_class = { .set_curr_task = set_curr_task_idle, .task_tick = task_tick_idle, + .get_rr_interval = get_rr_interval_idle, + .prio_changed = prio_changed_idle, .switched_to = switched_to_idle, diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c index 299d012b439..5c5fef37841 100644 --- a/kernel/sched_rt.c +++ b/kernel/sched_rt.c @@ -3,13 +3,18 @@ * policies) */ +#ifdef CONFIG_RT_GROUP_SCHED + +#define rt_entity_is_task(rt_se) (!(rt_se)->my_q) + static inline struct task_struct *rt_task_of(struct sched_rt_entity *rt_se) { +#ifdef CONFIG_SCHED_DEBUG + WARN_ON_ONCE(!rt_entity_is_task(rt_se)); +#endif return container_of(rt_se, struct task_struct, rt); } -#ifdef CONFIG_RT_GROUP_SCHED - static inline struct rq *rq_of_rt_rq(struct rt_rq *rt_rq) { return rt_rq->rq; @@ -22,6 +27,13 @@ static inline struct rt_rq *rt_rq_of_se(struct sched_rt_entity *rt_se) #else /* CONFIG_RT_GROUP_SCHED */ +#define rt_entity_is_task(rt_se) (1) + +static inline struct task_struct *rt_task_of(struct sched_rt_entity *rt_se) +{ + return container_of(rt_se, struct task_struct, rt); +} + static inline struct rq *rq_of_rt_rq(struct rt_rq *rt_rq) { return container_of(rt_rq, struct rq, rt); @@ -73,7 +85,7 @@ static inline void rt_clear_overload(struct rq *rq) static void update_rt_migration(struct rt_rq *rt_rq) { - if (rt_rq->rt_nr_migratory && (rt_rq->rt_nr_running > 1)) { + if (rt_rq->rt_nr_migratory && rt_rq->rt_nr_total > 1) { if (!rt_rq->overloaded) { rt_set_overload(rq_of_rt_rq(rt_rq)); rt_rq->overloaded = 1; @@ -86,6 +98,12 @@ static void update_rt_migration(struct rt_rq *rt_rq) static void inc_rt_migration(struct sched_rt_entity *rt_se, struct rt_rq *rt_rq) { + if (!rt_entity_is_task(rt_se)) + return; + + rt_rq = &rq_of_rt_rq(rt_rq)->rt; + + rt_rq->rt_nr_total++; if (rt_se->nr_cpus_allowed > 1) rt_rq->rt_nr_migratory++; @@ -94,6 +112,12 @@ static void inc_rt_migration(struct sched_rt_entity *rt_se, struct rt_rq *rt_rq) static void dec_rt_migration(struct sched_rt_entity *rt_se, struct rt_rq *rt_rq) { + if (!rt_entity_is_task(rt_se)) + return; + + rt_rq = &rq_of_rt_rq(rt_rq)->rt; + + rt_rq->rt_nr_total--; if (rt_se->nr_cpus_allowed > 1) rt_rq->rt_nr_migratory--; @@ -112,6 +136,11 @@ static void dequeue_pushable_task(struct rq *rq, struct task_struct *p) plist_del(&p->pushable_tasks, &rq->rt.pushable_tasks); } +static inline int has_pushable_tasks(struct rq *rq) +{ + return !plist_head_empty(&rq->rt.pushable_tasks); +} + #else static inline void enqueue_pushable_task(struct rq *rq, struct task_struct *p) @@ -586,6 +615,8 @@ static void update_curr_rt(struct rq *rq) curr->se.exec_start = rq->clock; cpuacct_charge(curr, delta_exec); + sched_rt_avg_update(rq, delta_exec); + if (!rt_bandwidth_enabled()) return; @@ -858,8 +889,6 @@ static void enqueue_task_rt(struct rq *rq, struct task_struct *p, int wakeup) if (!task_current(rq, p) && p->rt.nr_cpus_allowed > 1) enqueue_pushable_task(rq, p); - - inc_cpu_load(rq, p->se.load.weight); } static void dequeue_task_rt(struct rq *rq, struct task_struct *p, int sleep) @@ -870,8 +899,6 @@ static void dequeue_task_rt(struct rq *rq, struct task_struct *p, int sleep) dequeue_rt_entity(rt_se); dequeue_pushable_task(rq, p); - - dec_cpu_load(rq, p->se.load.weight); } /* @@ -911,10 +938,13 @@ static void yield_task_rt(struct rq *rq) #ifdef CONFIG_SMP static int find_lowest_rq(struct task_struct *task); -static int select_task_rq_rt(struct task_struct *p, int sync) +static int select_task_rq_rt(struct task_struct *p, int sd_flag, int flags) { struct rq *rq = task_rq(p); + if (sd_flag != SD_BALANCE_WAKE) + return smp_processor_id(); + /* * If the current task is an RT task, then * try to see if we can wake this RT task up on another @@ -948,20 +978,15 @@ static int select_task_rq_rt(struct task_struct *p, int sync) static void check_preempt_equal_prio(struct rq *rq, struct task_struct *p) { - cpumask_var_t mask; - if (rq->curr->rt.nr_cpus_allowed == 1) return; - if (!alloc_cpumask_var(&mask, GFP_ATOMIC)) - return; - if (p->rt.nr_cpus_allowed != 1 - && cpupri_find(&rq->rd->cpupri, p, mask)) - goto free; + && cpupri_find(&rq->rd->cpupri, p, NULL)) + return; - if (!cpupri_find(&rq->rd->cpupri, rq->curr, mask)) - goto free; + if (!cpupri_find(&rq->rd->cpupri, rq->curr, NULL)) + return; /* * There appears to be other cpus that can accept @@ -970,8 +995,6 @@ static void check_preempt_equal_prio(struct rq *rq, struct task_struct *p) */ requeue_task_rt(rq, p, 1); resched_task(rq->curr); -free: - free_cpumask_var(mask); } #endif /* CONFIG_SMP */ @@ -979,7 +1002,7 @@ free: /* * Preempt the current task with a newly woken task if needed: */ -static void check_preempt_curr_rt(struct rq *rq, struct task_struct *p, int sync) +static void check_preempt_curr_rt(struct rq *rq, struct task_struct *p, int flags) { if (p->prio < rq->curr->prio) { resched_task(rq->curr); @@ -1055,6 +1078,14 @@ static struct task_struct *pick_next_task_rt(struct rq *rq) if (p) dequeue_pushable_task(rq, p); +#ifdef CONFIG_SMP + /* + * We detect this state here so that we can avoid taking the RQ + * lock again later if there is no need to push + */ + rq->post_schedule = has_pushable_tasks(rq); +#endif + return p; } @@ -1122,29 +1153,12 @@ static struct task_struct *pick_next_highest_task_rt(struct rq *rq, int cpu) static DEFINE_PER_CPU(cpumask_var_t, local_cpu_mask); -static inline int pick_optimal_cpu(int this_cpu, - const struct cpumask *mask) -{ - int first; - - /* "this_cpu" is cheaper to preempt than a remote processor */ - if ((this_cpu != -1) && cpumask_test_cpu(this_cpu, mask)) - return this_cpu; - - first = cpumask_first(mask); - if (first < nr_cpu_ids) - return first; - - return -1; -} - static int find_lowest_rq(struct task_struct *task) { struct sched_domain *sd; struct cpumask *lowest_mask = __get_cpu_var(local_cpu_mask); int this_cpu = smp_processor_id(); int cpu = task_cpu(task); - cpumask_var_t domain_mask; if (task->rt.nr_cpus_allowed == 1) return -1; /* No other targets possible */ @@ -1153,13 +1167,6 @@ static int find_lowest_rq(struct task_struct *task) return -1; /* No targets found */ /* - * Only consider CPUs that are usable for migration. - * I guess we might want to change cpupri_find() to ignore those - * in the first place. - */ - cpumask_and(lowest_mask, lowest_mask, cpu_active_mask); - - /* * At this point we have built a mask of cpus representing the * lowest priority tasks in the system. Now we want to elect * the best one based on our affinity and topology. @@ -1174,28 +1181,26 @@ static int find_lowest_rq(struct task_struct *task) * Otherwise, we consult the sched_domains span maps to figure * out which cpu is logically closest to our hot cache data. */ - if (this_cpu == cpu) - this_cpu = -1; /* Skip this_cpu opt if the same */ + if (!cpumask_test_cpu(this_cpu, lowest_mask)) + this_cpu = -1; /* Skip this_cpu opt if not among lowest */ - if (alloc_cpumask_var(&domain_mask, GFP_ATOMIC)) { - for_each_domain(cpu, sd) { - if (sd->flags & SD_WAKE_AFFINE) { - int best_cpu; + for_each_domain(cpu, sd) { + if (sd->flags & SD_WAKE_AFFINE) { + int best_cpu; - cpumask_and(domain_mask, - sched_domain_span(sd), - lowest_mask); - - best_cpu = pick_optimal_cpu(this_cpu, - domain_mask); - - if (best_cpu != -1) { - free_cpumask_var(domain_mask); - return best_cpu; - } - } + /* + * "this_cpu" is cheaper to preempt than a + * remote processor. + */ + if (this_cpu != -1 && + cpumask_test_cpu(this_cpu, sched_domain_span(sd))) + return this_cpu; + + best_cpu = cpumask_first_and(lowest_mask, + sched_domain_span(sd)); + if (best_cpu < nr_cpu_ids) + return best_cpu; } - free_cpumask_var(domain_mask); } /* @@ -1203,7 +1208,13 @@ static int find_lowest_rq(struct task_struct *task) * just give the caller *something* to work with from the compatible * locations. */ - return pick_optimal_cpu(this_cpu, lowest_mask); + if (this_cpu != -1) + return this_cpu; + + cpu = cpumask_any(lowest_mask); + if (cpu < nr_cpu_ids) + return cpu; + return -1; } /* Will lock the rq it finds */ @@ -1253,11 +1264,6 @@ static struct rq *find_lock_lowest_rq(struct task_struct *task, struct rq *rq) return lowest_rq; } -static inline int has_pushable_tasks(struct rq *rq) -{ - return !plist_head_empty(&rq->rt.pushable_tasks); -} - static struct task_struct *pick_next_pushable_task(struct rq *rq) { struct task_struct *p; @@ -1457,23 +1463,9 @@ static void pre_schedule_rt(struct rq *rq, struct task_struct *prev) pull_rt_task(rq); } -/* - * assumes rq->lock is held - */ -static int needs_post_schedule_rt(struct rq *rq) -{ - return has_pushable_tasks(rq); -} - static void post_schedule_rt(struct rq *rq) { - /* - * This is only called if needs_post_schedule_rt() indicates that - * we need to push tasks away - */ - spin_lock_irq(&rq->lock); push_rt_tasks(rq); - spin_unlock_irq(&rq->lock); } /* @@ -1598,7 +1590,7 @@ static inline void init_sched_rt_class(void) unsigned int i; for_each_possible_cpu(i) - alloc_cpumask_var_node(&per_cpu(local_cpu_mask, i), + zalloc_cpumask_var_node(&per_cpu(local_cpu_mask, i), GFP_KERNEL, cpu_to_node(i)); } #endif /* CONFIG_SMP */ @@ -1729,6 +1721,17 @@ static void set_curr_task_rt(struct rq *rq) dequeue_pushable_task(rq, p); } +unsigned int get_rr_interval_rt(struct task_struct *task) +{ + /* + * Time slice is 0 for SCHED_FIFO tasks + */ + if (task->policy == SCHED_RR) + return DEF_TIMESLICE; + else + return 0; +} + static const struct sched_class rt_sched_class = { .next = &fair_sched_class, .enqueue_task = enqueue_task_rt, @@ -1749,7 +1752,6 @@ static const struct sched_class rt_sched_class = { .rq_online = rq_online_rt, .rq_offline = rq_offline_rt, .pre_schedule = pre_schedule_rt, - .needs_post_schedule = needs_post_schedule_rt, .post_schedule = post_schedule_rt, .task_wake_up = task_wake_up_rt, .switched_from = switched_from_rt, @@ -1758,6 +1760,8 @@ static const struct sched_class rt_sched_class = { .set_curr_task = set_curr_task_rt, .task_tick = task_tick_rt, + .get_rr_interval = get_rr_interval_rt, + .prio_changed = prio_changed_rt, .switched_to = switched_to_rt, }; diff --git a/kernel/signal.c b/kernel/signal.c index 1c8814481a1..6b982f2cf52 100644 --- a/kernel/signal.c +++ b/kernel/signal.c @@ -22,12 +22,14 @@ #include <linux/ptrace.h> #include <linux/signal.h> #include <linux/signalfd.h> +#include <linux/ratelimit.h> #include <linux/tracehook.h> #include <linux/capability.h> #include <linux/freezer.h> #include <linux/pid_namespace.h> #include <linux/nsproxy.h> -#include <trace/sched.h> +#define CREATE_TRACE_POINTS +#include <trace/events/signal.h> #include <asm/param.h> #include <asm/uaccess.h> @@ -41,7 +43,7 @@ static struct kmem_cache *sigqueue_cachep; -DEFINE_TRACE(sched_signal_send); +int print_fatal_signals __read_mostly; static void __user *sig_handler(struct task_struct *t, int sig) { @@ -55,10 +57,22 @@ static int sig_handler_ignored(void __user *handler, int sig) (handler == SIG_DFL && sig_kernel_ignore(sig)); } -static int sig_ignored(struct task_struct *t, int sig) +static int sig_task_ignored(struct task_struct *t, int sig, + int from_ancestor_ns) { void __user *handler; + handler = sig_handler(t, sig); + + if (unlikely(t->signal->flags & SIGNAL_UNKILLABLE) && + handler == SIG_DFL && !from_ancestor_ns) + return 1; + + return sig_handler_ignored(handler, sig); +} + +static int sig_ignored(struct task_struct *t, int sig, int from_ancestor_ns) +{ /* * Blocked signals are never ignored, since the * signal handler may change by the time it is @@ -67,14 +81,13 @@ static int sig_ignored(struct task_struct *t, int sig) if (sigismember(&t->blocked, sig) || sigismember(&t->real_blocked, sig)) return 0; - handler = sig_handler(t, sig); - if (!sig_handler_ignored(handler, sig)) + if (!sig_task_ignored(t, sig, from_ancestor_ns)) return 0; /* * Tracers may want to know about even ignored signals. */ - return !tracehook_consider_ignored_signal(t, sig, handler); + return !tracehook_consider_ignored_signal(t, sig); } /* @@ -150,7 +163,7 @@ int next_signal(struct sigpending *pending, sigset_t *mask) { unsigned long i, *s, *m, x; int sig = 0; - + s = pending->signal.sig; m = mask->sig; switch (_NSIG_WORDS) { @@ -175,17 +188,31 @@ int next_signal(struct sigpending *pending, sigset_t *mask) sig = ffz(~x) + 1; break; } - + return sig; } +static inline void print_dropped_signal(int sig) +{ + static DEFINE_RATELIMIT_STATE(ratelimit_state, 5 * HZ, 10); + + if (!print_fatal_signals) + return; + + if (!__ratelimit(&ratelimit_state)) + return; + + printk(KERN_INFO "%s/%d: reached RLIMIT_SIGPENDING, dropped signal %d\n", + current->comm, current->pid, sig); +} + /* * allocate a new signal queue record * - this may be called without locks if and only if t == current, otherwise an * appopriate lock must be held to stop the target task from exiting */ -static struct sigqueue *__sigqueue_alloc(struct task_struct *t, gfp_t flags, - int override_rlimit) +static struct sigqueue * +__sigqueue_alloc(int sig, struct task_struct *t, gfp_t flags, int override_rlimit) { struct sigqueue *q = NULL; struct user_struct *user; @@ -198,10 +225,15 @@ static struct sigqueue *__sigqueue_alloc(struct task_struct *t, gfp_t flags, */ user = get_uid(__task_cred(t)->user); atomic_inc(&user->sigpending); + if (override_rlimit || atomic_read(&user->sigpending) <= - t->signal->rlim[RLIMIT_SIGPENDING].rlim_cur) + t->signal->rlim[RLIMIT_SIGPENDING].rlim_cur) { q = kmem_cache_alloc(sigqueue_cachep, flags); + } else { + print_dropped_signal(sig); + } + if (unlikely(q == NULL)) { atomic_dec(&user->sigpending); free_uid(user); @@ -238,14 +270,19 @@ void flush_sigqueue(struct sigpending *queue) /* * Flush all pending signals for a task. */ +void __flush_signals(struct task_struct *t) +{ + clear_tsk_thread_flag(t, TIF_SIGPENDING); + flush_sigqueue(&t->pending); + flush_sigqueue(&t->signal->shared_pending); +} + void flush_signals(struct task_struct *t) { unsigned long flags; spin_lock_irqsave(&t->sighand->siglock, flags); - clear_tsk_thread_flag(t, TIF_SIGPENDING); - flush_sigqueue(&t->pending); - flush_sigqueue(&t->signal->shared_pending); + __flush_signals(t); spin_unlock_irqrestore(&t->sighand->siglock, flags); } @@ -318,7 +355,7 @@ int unhandled_signal(struct task_struct *tsk, int sig) return 1; if (handler != SIG_IGN && handler != SIG_DFL) return 0; - return !tracehook_consider_fatal_signal(tsk, sig, handler); + return !tracehook_consider_fatal_signal(tsk, sig); } @@ -624,7 +661,7 @@ static int check_kill_permission(int sig, struct siginfo *info, * Returns true if the signal should be actually delivered, otherwise * it should be dropped. */ -static int prepare_signal(int sig, struct task_struct *p) +static int prepare_signal(int sig, struct task_struct *p, int from_ancestor_ns) { struct signal_struct *signal = p->signal; struct task_struct *t; @@ -691,7 +728,7 @@ static int prepare_signal(int sig, struct task_struct *p) if (why) { /* - * The first thread which returns from finish_stop() + * The first thread which returns from do_signal_stop() * will take ->siglock, notice SIGNAL_CLD_MASK, and * notify its parent. See get_signal_to_deliver(). */ @@ -708,7 +745,7 @@ static int prepare_signal(int sig, struct task_struct *p) } } - return !sig_ignored(p, sig); + return !sig_ignored(p, sig, from_ancestor_ns); } /* @@ -777,7 +814,7 @@ static void complete_signal(int sig, struct task_struct *p, int group) !(signal->flags & (SIGNAL_UNKILLABLE | SIGNAL_GROUP_EXIT)) && !sigismember(&t->real_blocked, sig) && (sig == SIGKILL || - !tracehook_consider_fatal_signal(t, sig, SIG_DFL))) { + !tracehook_consider_fatal_signal(t, sig))) { /* * This signal will be fatal to the whole group. */ @@ -813,16 +850,18 @@ static inline int legacy_queue(struct sigpending *signals, int sig) return (sig < SIGRTMIN) && sigismember(&signals->signal, sig); } -static int send_signal(int sig, struct siginfo *info, struct task_struct *t, - int group) +static int __send_signal(int sig, struct siginfo *info, struct task_struct *t, + int group, int from_ancestor_ns) { struct sigpending *pending; struct sigqueue *q; + int override_rlimit; - trace_sched_signal_send(sig, t); + trace_signal_generate(sig, info, t); assert_spin_locked(&t->sighand->siglock); - if (!prepare_signal(sig, t)) + + if (!prepare_signal(sig, t, from_ancestor_ns)) return 0; pending = group ? &t->signal->shared_pending : &t->pending; @@ -848,9 +887,13 @@ static int send_signal(int sig, struct siginfo *info, struct task_struct *t, make sure at least one signal gets delivered and don't pass on the info struct. */ - q = __sigqueue_alloc(t, GFP_ATOMIC, (sig < SIGRTMIN && - (is_si_special(info) || - info->si_code >= 0))); + if (sig < SIGRTMIN) + override_rlimit = (is_si_special(info) || info->si_code >= 0); + else + override_rlimit = 0; + + q = __sigqueue_alloc(sig, t, GFP_ATOMIC | __GFP_NOTRACK_FALSE_POSITIVE, + override_rlimit); if (q) { list_add_tail(&q->list, &pending->list); switch ((unsigned long) info) { @@ -871,15 +914,26 @@ static int send_signal(int sig, struct siginfo *info, struct task_struct *t, break; default: copy_siginfo(&q->info, info); + if (from_ancestor_ns) + q->info.si_pid = 0; break; } } else if (!is_si_special(info)) { - if (sig >= SIGRTMIN && info->si_code != SI_USER) - /* - * Queue overflow, abort. We may abort if the signal was rt - * and sent by user using something other than kill(). - */ + if (sig >= SIGRTMIN && info->si_code != SI_USER) { + /* + * Queue overflow, abort. We may abort if the + * signal was rt and sent by user using something + * other than kill(). + */ + trace_signal_overflow_fail(sig, group, info); return -EAGAIN; + } else { + /* + * This is a silent loss of information. We still + * send the signal, but the *info bits are lost. + */ + trace_signal_lose_info(sig, group, info); + } } out_set: @@ -889,7 +943,19 @@ out_set: return 0; } -int print_fatal_signals; +static int send_signal(int sig, struct siginfo *info, struct task_struct *t, + int group) +{ + int from_ancestor_ns = 0; + +#ifdef CONFIG_PID_NS + if (!is_si_special(info) && SI_FROMUSER(info) && + task_pid_nr_ns(current, task_active_pid_ns(t)) <= 0) + from_ancestor_ns = 1; +#endif + + return __send_signal(sig, info, t, group, from_ancestor_ns); +} static void print_fatal_signal(struct pt_regs *regs, int signr) { @@ -935,6 +1001,20 @@ specific_send_sig_info(int sig, struct siginfo *info, struct task_struct *t) return send_signal(sig, info, t, 0); } +int do_send_sig_info(int sig, struct siginfo *info, struct task_struct *p, + bool group) +{ + unsigned long flags; + int ret = -ESRCH; + + if (lock_task_sighand(p, &flags)) { + ret = send_signal(sig, info, p, group); + unlock_task_sighand(p, &flags); + } + + return ret; +} + /* * Force a signal that the process can't ignore: if necessary * we unblock the signal and change any SIG_IGN to SIG_DFL. @@ -1000,12 +1080,6 @@ void zap_other_threads(struct task_struct *p) } } -int __fatal_signal_pending(struct task_struct *tsk) -{ - return sigismember(&tsk->pending.signal, SIGKILL); -} -EXPORT_SYMBOL(__fatal_signal_pending); - struct sighand_struct *lock_task_sighand(struct task_struct *tsk, unsigned long *flags) { struct sighand_struct *sighand; @@ -1032,18 +1106,10 @@ struct sighand_struct *lock_task_sighand(struct task_struct *tsk, unsigned long */ int group_send_sig_info(int sig, struct siginfo *info, struct task_struct *p) { - unsigned long flags; - int ret; + int ret = check_kill_permission(sig, info, p); - ret = check_kill_permission(sig, info, p); - - if (!ret && sig) { - ret = -ESRCH; - if (lock_task_sighand(p, &flags)) { - ret = __group_send_sig_info(sig, info, p); - unlock_task_sighand(p, &flags); - } - } + if (!ret && sig) + ret = do_send_sig_info(sig, info, p, true); return ret; } @@ -1133,7 +1199,7 @@ int kill_pid_info_as_uid(int sig, struct siginfo *info, struct pid *pid, if (sig && p->sighand) { unsigned long flags; spin_lock_irqsave(&p->sighand->siglock, flags); - ret = __group_send_sig_info(sig, info, p); + ret = __send_signal(sig, info, p, 1, 0); spin_unlock_irqrestore(&p->sighand->siglock, flags); } out_unlock: @@ -1188,15 +1254,9 @@ static int kill_something_info(int sig, struct siginfo *info, pid_t pid) * These are for backward compatibility with the rest of the kernel source. */ -/* - * The caller must ensure the task can't exit. - */ int send_sig_info(int sig, struct siginfo *info, struct task_struct *p) { - int ret; - unsigned long flags; - /* * Make sure legacy kernel users don't send in bad values * (normal paths check this in check_kill_permission). @@ -1204,10 +1264,7 @@ send_sig_info(int sig, struct siginfo *info, struct task_struct *p) if (!valid_signal(sig)) return -EINVAL; - spin_lock_irqsave(&p->sighand->siglock, flags); - ret = specific_send_sig_info(sig, info, p); - spin_unlock_irqrestore(&p->sighand->siglock, flags); - return ret; + return do_send_sig_info(sig, info, p, false); } #define __si_special(priv) \ @@ -1266,19 +1323,19 @@ EXPORT_SYMBOL(kill_pid); * These functions support sending signals using preallocated sigqueue * structures. This is needed "because realtime applications cannot * afford to lose notifications of asynchronous events, like timer - * expirations or I/O completions". In the case of Posix Timers + * expirations or I/O completions". In the case of Posix Timers * we allocate the sigqueue structure from the timer_create. If this * allocation fails we are able to report the failure to the application * with an EAGAIN error. */ - struct sigqueue *sigqueue_alloc(void) { - struct sigqueue *q; + struct sigqueue *q = __sigqueue_alloc(-1, current, GFP_KERNEL, 0); - if ((q = __sigqueue_alloc(current, GFP_KERNEL, 0))) + if (q) q->flags |= SIGQUEUE_PREALLOC; - return(q); + + return q; } void sigqueue_free(struct sigqueue *q) @@ -1320,7 +1377,7 @@ int send_sigqueue(struct sigqueue *q, struct task_struct *t, int group) goto ret; ret = 1; /* the signal is ignored */ - if (!prepare_signal(sig, t)) + if (!prepare_signal(sig, t, 0)) goto out; ret = 0; @@ -1347,15 +1404,6 @@ ret: } /* - * Wake up any threads in the parent blocked in wait* syscalls. - */ -static inline void __wake_up_parent(struct task_struct *p, - struct task_struct *parent) -{ - wake_up_interruptible_sync(&parent->signal->wait_chldexit); -} - -/* * Let a parent know about the death of a child. * For a stopped/continued status change, use do_notify_parent_cldstop instead. * @@ -1374,7 +1422,7 @@ int do_notify_parent(struct task_struct *tsk, int sig) /* do_notify_parent_cldstop should have been called instead. */ BUG_ON(task_is_stopped_or_traced(tsk)); - BUG_ON(!tsk->ptrace && + BUG_ON(!task_ptrace(tsk) && (tsk->group_leader != tsk || !thread_group_empty(tsk))); info.si_signo = sig; @@ -1413,7 +1461,7 @@ int do_notify_parent(struct task_struct *tsk, int sig) psig = tsk->parent->sighand; spin_lock_irqsave(&psig->siglock, flags); - if (!tsk->ptrace && sig == SIGCHLD && + if (!task_ptrace(tsk) && sig == SIGCHLD && (psig->action[SIGCHLD-1].sa.sa_handler == SIG_IGN || (psig->action[SIGCHLD-1].sa.sa_flags & SA_NOCLDWAIT))) { /* @@ -1450,7 +1498,7 @@ static void do_notify_parent_cldstop(struct task_struct *tsk, int why) struct task_struct *parent; struct sighand_struct *sighand; - if (tsk->ptrace & PT_PTRACED) + if (task_ptrace(tsk)) parent = tsk->parent; else { tsk = tsk->group_leader; @@ -1463,7 +1511,7 @@ static void do_notify_parent_cldstop(struct task_struct *tsk, int why) * see comment in do_notify_parent() abot the following 3 lines */ rcu_read_lock(); - info.si_pid = task_pid_nr_ns(tsk, tsk->parent->nsproxy->pid_ns); + info.si_pid = task_pid_nr_ns(tsk, parent->nsproxy->pid_ns); info.si_uid = __task_cred(tsk)->uid; rcu_read_unlock(); @@ -1499,7 +1547,7 @@ static void do_notify_parent_cldstop(struct task_struct *tsk, int why) static inline int may_ptrace_stop(void) { - if (!likely(current->ptrace & PT_PTRACED)) + if (!likely(task_ptrace(current))) return 0; /* * Are we in the middle of do_coredump? @@ -1637,29 +1685,6 @@ void ptrace_notify(int exit_code) spin_unlock_irq(¤t->sighand->siglock); } -static void -finish_stop(int stop_count) -{ - /* - * If there are no other threads in the group, or if there is - * a group stop in progress and we are the last to stop, - * report to the parent. When ptraced, every thread reports itself. - */ - if (tracehook_notify_jctl(stop_count == 0, CLD_STOPPED)) { - read_lock(&tasklist_lock); - do_notify_parent_cldstop(current, CLD_STOPPED); - read_unlock(&tasklist_lock); - } - - do { - schedule(); - } while (try_to_freeze()); - /* - * Now we don't run again until continued. - */ - current->exit_code = 0; -} - /* * This performs the stopping for SIGSTOP and other stop signals. * We have to stop all threads in the thread group. @@ -1669,15 +1694,9 @@ finish_stop(int stop_count) static int do_signal_stop(int signr) { struct signal_struct *sig = current->signal; - int stop_count; + int notify; - if (sig->group_stop_count > 0) { - /* - * There is a group stop in progress. We don't need to - * start another one. - */ - stop_count = --sig->group_stop_count; - } else { + if (!sig->group_stop_count) { struct task_struct *t; if (!likely(sig->flags & SIGNAL_STOP_DEQUEUED) || @@ -1689,7 +1708,7 @@ static int do_signal_stop(int signr) */ sig->group_exit_code = signr; - stop_count = 0; + sig->group_stop_count = 1; for (t = next_thread(current); t != current; t = next_thread(t)) /* * Setting state to TASK_STOPPED for a group @@ -1698,26 +1717,51 @@ static int do_signal_stop(int signr) */ if (!(t->flags & PF_EXITING) && !task_is_stopped_or_traced(t)) { - stop_count++; + sig->group_stop_count++; signal_wake_up(t, 0); } - sig->group_stop_count = stop_count; } + /* + * If there are no other threads in the group, or if there is + * a group stop in progress and we are the last to stop, report + * to the parent. When ptraced, every thread reports itself. + */ + notify = sig->group_stop_count == 1 ? CLD_STOPPED : 0; + notify = tracehook_notify_jctl(notify, CLD_STOPPED); + /* + * tracehook_notify_jctl() can drop and reacquire siglock, so + * we keep ->group_stop_count != 0 before the call. If SIGCONT + * or SIGKILL comes in between ->group_stop_count == 0. + */ + if (sig->group_stop_count) { + if (!--sig->group_stop_count) + sig->flags = SIGNAL_STOP_STOPPED; + current->exit_code = sig->group_exit_code; + __set_current_state(TASK_STOPPED); + } + spin_unlock_irq(¤t->sighand->siglock); - if (stop_count == 0) - sig->flags = SIGNAL_STOP_STOPPED; - current->exit_code = sig->group_exit_code; - __set_current_state(TASK_STOPPED); + if (notify) { + read_lock(&tasklist_lock); + do_notify_parent_cldstop(current, notify); + read_unlock(&tasklist_lock); + } + + /* Now we don't run again until woken by SIGCONT or SIGKILL */ + do { + schedule(); + } while (try_to_freeze()); + + tracehook_finish_jctl(); + current->exit_code = 0; - spin_unlock_irq(¤t->sighand->siglock); - finish_stop(stop_count); return 1; } static int ptrace_signal(int signr, siginfo_t *info, struct pt_regs *regs, void *cookie) { - if (!(current->ptrace & PT_PTRACED)) + if (!task_ptrace(current)) return signr; ptrace_signal_deliver(regs, cookie); @@ -1779,14 +1823,15 @@ relock: int why = (signal->flags & SIGNAL_STOP_CONTINUED) ? CLD_CONTINUED : CLD_STOPPED; signal->flags &= ~SIGNAL_CLD_MASK; - spin_unlock_irq(&sighand->siglock); - if (unlikely(!tracehook_notify_jctl(1, why))) - goto relock; + why = tracehook_notify_jctl(why, CLD_CONTINUED); + spin_unlock_irq(&sighand->siglock); - read_lock(&tasklist_lock); - do_notify_parent_cldstop(current->group_leader, why); - read_unlock(&tasklist_lock); + if (why) { + read_lock(&tasklist_lock); + do_notify_parent_cldstop(current->group_leader, why); + read_unlock(&tasklist_lock); + } goto relock; } @@ -1824,6 +1869,9 @@ relock: ka = &sighand->action[signr-1]; } + /* Trace actually delivered signals. */ + trace_signal_deliver(signr, info, ka); + if (ka->sa.sa_handler == SIG_IGN) /* Do nothing. */ continue; if (ka->sa.sa_handler != SIG_DFL) { @@ -1844,9 +1892,16 @@ relock: /* * Global init gets no signals it doesn't want. + * Container-init gets no signals it doesn't want from same + * container. + * + * Note that if global/container-init sees a sig_kernel_only() + * signal here, the signal must have been generated internally + * or must have come from an ancestor namespace. In either + * case, the signal cannot be dropped. */ if (unlikely(signal->flags & SIGNAL_UNKILLABLE) && - !signal_group_exit(signal)) + !sig_kernel_only(signr)) continue; if (sig_kernel_stop(signr)) { @@ -1944,14 +1999,14 @@ void exit_signals(struct task_struct *tsk) if (unlikely(tsk->signal->group_stop_count) && !--tsk->signal->group_stop_count) { tsk->signal->flags = SIGNAL_STOP_STOPPED; - group_stop = 1; + group_stop = tracehook_notify_jctl(CLD_STOPPED, CLD_STOPPED); } out: spin_unlock_irq(&tsk->sighand->siglock); - if (unlikely(group_stop) && tracehook_notify_jctl(1, CLD_STOPPED)) { + if (unlikely(group_stop)) { read_lock(&tasklist_lock); - do_notify_parent_cldstop(tsk, CLD_STOPPED); + do_notify_parent_cldstop(tsk, group_stop); read_unlock(&tasklist_lock); } } @@ -2243,35 +2298,29 @@ SYSCALL_DEFINE2(kill, pid_t, pid, int, sig) return kill_something_info(sig, &info, pid); } -static int do_tkill(pid_t tgid, pid_t pid, int sig) +static int +do_send_specific(pid_t tgid, pid_t pid, int sig, struct siginfo *info) { - int error; - struct siginfo info; struct task_struct *p; - unsigned long flags; - - error = -ESRCH; - info.si_signo = sig; - info.si_errno = 0; - info.si_code = SI_TKILL; - info.si_pid = task_tgid_vnr(current); - info.si_uid = current_uid(); + int error = -ESRCH; rcu_read_lock(); p = find_task_by_vpid(pid); if (p && (tgid <= 0 || task_tgid_vnr(p) == tgid)) { - error = check_kill_permission(sig, &info, p); + error = check_kill_permission(sig, info, p); /* * The null signal is a permissions and process existence * probe. No signal is actually delivered. - * - * If lock_task_sighand() fails we pretend the task dies - * after receiving the signal. The window is tiny, and the - * signal is private anyway. */ - if (!error && sig && lock_task_sighand(p, &flags)) { - error = specific_send_sig_info(sig, &info, p); - unlock_task_sighand(p, &flags); + if (!error && sig) { + error = do_send_sig_info(sig, info, p, false); + /* + * If lock_task_sighand() failed we pretend the task + * dies after receiving the signal. The window is tiny, + * and the signal is private anyway. + */ + if (unlikely(error == -ESRCH)) + error = 0; } } rcu_read_unlock(); @@ -2279,6 +2328,19 @@ static int do_tkill(pid_t tgid, pid_t pid, int sig) return error; } +static int do_tkill(pid_t tgid, pid_t pid, int sig) +{ + struct siginfo info; + + info.si_signo = sig; + info.si_errno = 0; + info.si_code = SI_TKILL; + info.si_pid = task_tgid_vnr(current); + info.si_uid = current_uid(); + + return do_send_specific(tgid, pid, sig, &info); +} + /** * sys_tgkill - send signal to one specific thread * @tgid: the thread group ID of the thread @@ -2328,6 +2390,32 @@ SYSCALL_DEFINE3(rt_sigqueueinfo, pid_t, pid, int, sig, return kill_proc_info(sig, &info, pid); } +long do_rt_tgsigqueueinfo(pid_t tgid, pid_t pid, int sig, siginfo_t *info) +{ + /* This is only valid for single tasks */ + if (pid <= 0 || tgid <= 0) + return -EINVAL; + + /* Not even root can pretend to send signals from the kernel. + Nor can they impersonate a kill(), which adds source info. */ + if (info->si_code >= 0) + return -EPERM; + info->si_signo = sig; + + return do_send_specific(tgid, pid, sig, info); +} + +SYSCALL_DEFINE4(rt_tgsigqueueinfo, pid_t, tgid, pid_t, pid, int, sig, + siginfo_t __user *, uinfo) +{ + siginfo_t info; + + if (copy_from_user(&info, uinfo, sizeof(siginfo_t))) + return -EFAULT; + + return do_rt_tgsigqueueinfo(tgid, pid, sig, &info); +} + int do_sigaction(int sig, struct k_sigaction *act, struct k_sigaction *oact) { struct task_struct *t = current; @@ -2379,11 +2467,9 @@ do_sigaltstack (const stack_t __user *uss, stack_t __user *uoss, unsigned long s stack_t oss; int error; - if (uoss) { - oss.ss_sp = (void __user *) current->sas_ss_sp; - oss.ss_size = current->sas_ss_size; - oss.ss_flags = sas_ss_flags(sp); - } + oss.ss_sp = (void __user *) current->sas_ss_sp; + oss.ss_size = current->sas_ss_size; + oss.ss_flags = sas_ss_flags(sp); if (uss) { void __user *ss_sp; @@ -2391,10 +2477,12 @@ do_sigaltstack (const stack_t __user *uss, stack_t __user *uoss, unsigned long s int ss_flags; error = -EFAULT; - if (!access_ok(VERIFY_READ, uss, sizeof(*uss)) - || __get_user(ss_sp, &uss->ss_sp) - || __get_user(ss_flags, &uss->ss_flags) - || __get_user(ss_size, &uss->ss_size)) + if (!access_ok(VERIFY_READ, uss, sizeof(*uss))) + goto out; + error = __get_user(ss_sp, &uss->ss_sp) | + __get_user(ss_flags, &uss->ss_flags) | + __get_user(ss_size, &uss->ss_size); + if (error) goto out; error = -EPERM; @@ -2426,13 +2514,16 @@ do_sigaltstack (const stack_t __user *uss, stack_t __user *uoss, unsigned long s current->sas_ss_size = ss_size; } + error = 0; if (uoss) { error = -EFAULT; - if (copy_to_user(uoss, &oss, sizeof(oss))) + if (!access_ok(VERIFY_WRITE, uoss, sizeof(*uoss))) goto out; + error = __put_user(oss.ss_sp, &uoss->ss_sp) | + __put_user(oss.ss_size, &uoss->ss_size) | + __put_user(oss.ss_flags, &uoss->ss_flags); } - error = 0; out: return error; } diff --git a/kernel/slow-work-debugfs.c b/kernel/slow-work-debugfs.c new file mode 100644 index 00000000000..e45c4364529 --- /dev/null +++ b/kernel/slow-work-debugfs.c @@ -0,0 +1,227 @@ +/* Slow work debugging + * + * Copyright (C) 2009 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public Licence + * as published by the Free Software Foundation; either version + * 2 of the Licence, or (at your option) any later version. + */ + +#include <linux/module.h> +#include <linux/slow-work.h> +#include <linux/fs.h> +#include <linux/time.h> +#include <linux/seq_file.h> +#include "slow-work.h" + +#define ITERATOR_SHIFT (BITS_PER_LONG - 4) +#define ITERATOR_SELECTOR (0xfUL << ITERATOR_SHIFT) +#define ITERATOR_COUNTER (~ITERATOR_SELECTOR) + +void slow_work_new_thread_desc(struct slow_work *work, struct seq_file *m) +{ + seq_puts(m, "Slow-work: New thread"); +} + +/* + * Render the time mark field on a work item into a 5-char time with units plus + * a space + */ +static void slow_work_print_mark(struct seq_file *m, struct slow_work *work) +{ + struct timespec now, diff; + + now = CURRENT_TIME; + diff = timespec_sub(now, work->mark); + + if (diff.tv_sec < 0) + seq_puts(m, " -ve "); + else if (diff.tv_sec == 0 && diff.tv_nsec < 1000) + seq_printf(m, "%3luns ", diff.tv_nsec); + else if (diff.tv_sec == 0 && diff.tv_nsec < 1000000) + seq_printf(m, "%3luus ", diff.tv_nsec / 1000); + else if (diff.tv_sec == 0 && diff.tv_nsec < 1000000000) + seq_printf(m, "%3lums ", diff.tv_nsec / 1000000); + else if (diff.tv_sec <= 1) + seq_puts(m, " 1s "); + else if (diff.tv_sec < 60) + seq_printf(m, "%4lus ", diff.tv_sec); + else if (diff.tv_sec < 60 * 60) + seq_printf(m, "%4lum ", diff.tv_sec / 60); + else if (diff.tv_sec < 60 * 60 * 24) + seq_printf(m, "%4luh ", diff.tv_sec / 3600); + else + seq_puts(m, "exces "); +} + +/* + * Describe a slow work item for debugfs + */ +static int slow_work_runqueue_show(struct seq_file *m, void *v) +{ + struct slow_work *work; + struct list_head *p = v; + unsigned long id; + + switch ((unsigned long) v) { + case 1: + seq_puts(m, "THR PID ITEM ADDR FL MARK DESC\n"); + return 0; + case 2: + seq_puts(m, "=== ===== ================ == ===== ==========\n"); + return 0; + + case 3 ... 3 + SLOW_WORK_THREAD_LIMIT - 1: + id = (unsigned long) v - 3; + + read_lock(&slow_work_execs_lock); + work = slow_work_execs[id]; + if (work) { + smp_read_barrier_depends(); + + seq_printf(m, "%3lu %5d %16p %2lx ", + id, slow_work_pids[id], work, work->flags); + slow_work_print_mark(m, work); + + if (work->ops->desc) + work->ops->desc(work, m); + seq_putc(m, '\n'); + } + read_unlock(&slow_work_execs_lock); + return 0; + + default: + work = list_entry(p, struct slow_work, link); + seq_printf(m, "%3s - %16p %2lx ", + work->flags & SLOW_WORK_VERY_SLOW ? "vsq" : "sq", + work, work->flags); + slow_work_print_mark(m, work); + + if (work->ops->desc) + work->ops->desc(work, m); + seq_putc(m, '\n'); + return 0; + } +} + +/* + * map the iterator to a work item + */ +static void *slow_work_runqueue_index(struct seq_file *m, loff_t *_pos) +{ + struct list_head *p; + unsigned long count, id; + + switch (*_pos >> ITERATOR_SHIFT) { + case 0x0: + if (*_pos == 0) + *_pos = 1; + if (*_pos < 3) + return (void *)(unsigned long) *_pos; + if (*_pos < 3 + SLOW_WORK_THREAD_LIMIT) + for (id = *_pos - 3; + id < SLOW_WORK_THREAD_LIMIT; + id++, (*_pos)++) + if (slow_work_execs[id]) + return (void *)(unsigned long) *_pos; + *_pos = 0x1UL << ITERATOR_SHIFT; + + case 0x1: + count = *_pos & ITERATOR_COUNTER; + list_for_each(p, &slow_work_queue) { + if (count == 0) + return p; + count--; + } + *_pos = 0x2UL << ITERATOR_SHIFT; + + case 0x2: + count = *_pos & ITERATOR_COUNTER; + list_for_each(p, &vslow_work_queue) { + if (count == 0) + return p; + count--; + } + *_pos = 0x3UL << ITERATOR_SHIFT; + + default: + return NULL; + } +} + +/* + * set up the iterator to start reading from the first line + */ +static void *slow_work_runqueue_start(struct seq_file *m, loff_t *_pos) +{ + spin_lock_irq(&slow_work_queue_lock); + return slow_work_runqueue_index(m, _pos); +} + +/* + * move to the next line + */ +static void *slow_work_runqueue_next(struct seq_file *m, void *v, loff_t *_pos) +{ + struct list_head *p = v; + unsigned long selector = *_pos >> ITERATOR_SHIFT; + + (*_pos)++; + switch (selector) { + case 0x0: + return slow_work_runqueue_index(m, _pos); + + case 0x1: + if (*_pos >> ITERATOR_SHIFT == 0x1) { + p = p->next; + if (p != &slow_work_queue) + return p; + } + *_pos = 0x2UL << ITERATOR_SHIFT; + p = &vslow_work_queue; + + case 0x2: + if (*_pos >> ITERATOR_SHIFT == 0x2) { + p = p->next; + if (p != &vslow_work_queue) + return p; + } + *_pos = 0x3UL << ITERATOR_SHIFT; + + default: + return NULL; + } +} + +/* + * clean up after reading + */ +static void slow_work_runqueue_stop(struct seq_file *m, void *v) +{ + spin_unlock_irq(&slow_work_queue_lock); +} + +static const struct seq_operations slow_work_runqueue_ops = { + .start = slow_work_runqueue_start, + .stop = slow_work_runqueue_stop, + .next = slow_work_runqueue_next, + .show = slow_work_runqueue_show, +}; + +/* + * open "/sys/kernel/debug/slow_work/runqueue" to list queue contents + */ +static int slow_work_runqueue_open(struct inode *inode, struct file *file) +{ + return seq_open(file, &slow_work_runqueue_ops); +} + +const struct file_operations slow_work_runqueue_fops = { + .owner = THIS_MODULE, + .open = slow_work_runqueue_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release, +}; diff --git a/kernel/slow-work.c b/kernel/slow-work.c new file mode 100644 index 00000000000..7494bbf5a27 --- /dev/null +++ b/kernel/slow-work.c @@ -0,0 +1,1068 @@ +/* Worker thread pool for slow items, such as filesystem lookups or mkdirs + * + * Copyright (C) 2008 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public Licence + * as published by the Free Software Foundation; either version + * 2 of the Licence, or (at your option) any later version. + * + * See Documentation/slow-work.txt + */ + +#include <linux/module.h> +#include <linux/slow-work.h> +#include <linux/kthread.h> +#include <linux/freezer.h> +#include <linux/wait.h> +#include <linux/debugfs.h> +#include "slow-work.h" + +static void slow_work_cull_timeout(unsigned long); +static void slow_work_oom_timeout(unsigned long); + +#ifdef CONFIG_SYSCTL +static int slow_work_min_threads_sysctl(struct ctl_table *, int, + void __user *, size_t *, loff_t *); + +static int slow_work_max_threads_sysctl(struct ctl_table *, int , + void __user *, size_t *, loff_t *); +#endif + +/* + * The pool of threads has at least min threads in it as long as someone is + * using the facility, and may have as many as max. + * + * A portion of the pool may be processing very slow operations. + */ +static unsigned slow_work_min_threads = 2; +static unsigned slow_work_max_threads = 4; +static unsigned vslow_work_proportion = 50; /* % of threads that may process + * very slow work */ + +#ifdef CONFIG_SYSCTL +static const int slow_work_min_min_threads = 2; +static int slow_work_max_max_threads = SLOW_WORK_THREAD_LIMIT; +static const int slow_work_min_vslow = 1; +static const int slow_work_max_vslow = 99; + +ctl_table slow_work_sysctls[] = { + { + .procname = "min-threads", + .data = &slow_work_min_threads, + .maxlen = sizeof(unsigned), + .mode = 0644, + .proc_handler = slow_work_min_threads_sysctl, + .extra1 = (void *) &slow_work_min_min_threads, + .extra2 = &slow_work_max_threads, + }, + { + .procname = "max-threads", + .data = &slow_work_max_threads, + .maxlen = sizeof(unsigned), + .mode = 0644, + .proc_handler = slow_work_max_threads_sysctl, + .extra1 = &slow_work_min_threads, + .extra2 = (void *) &slow_work_max_max_threads, + }, + { + .procname = "vslow-percentage", + .data = &vslow_work_proportion, + .maxlen = sizeof(unsigned), + .mode = 0644, + .proc_handler = proc_dointvec_minmax, + .extra1 = (void *) &slow_work_min_vslow, + .extra2 = (void *) &slow_work_max_vslow, + }, + {} +}; +#endif + +/* + * The active state of the thread pool + */ +static atomic_t slow_work_thread_count; +static atomic_t vslow_work_executing_count; + +static bool slow_work_may_not_start_new_thread; +static bool slow_work_cull; /* cull a thread due to lack of activity */ +static DEFINE_TIMER(slow_work_cull_timer, slow_work_cull_timeout, 0, 0); +static DEFINE_TIMER(slow_work_oom_timer, slow_work_oom_timeout, 0, 0); +static struct slow_work slow_work_new_thread; /* new thread starter */ + +/* + * slow work ID allocation (use slow_work_queue_lock) + */ +static DECLARE_BITMAP(slow_work_ids, SLOW_WORK_THREAD_LIMIT); + +/* + * Unregistration tracking to prevent put_ref() from disappearing during module + * unload + */ +#ifdef CONFIG_MODULES +static struct module *slow_work_thread_processing[SLOW_WORK_THREAD_LIMIT]; +static struct module *slow_work_unreg_module; +static struct slow_work *slow_work_unreg_work_item; +static DECLARE_WAIT_QUEUE_HEAD(slow_work_unreg_wq); +static DEFINE_MUTEX(slow_work_unreg_sync_lock); + +static void slow_work_set_thread_processing(int id, struct slow_work *work) +{ + if (work) + slow_work_thread_processing[id] = work->owner; +} +static void slow_work_done_thread_processing(int id, struct slow_work *work) +{ + struct module *module = slow_work_thread_processing[id]; + + slow_work_thread_processing[id] = NULL; + smp_mb(); + if (slow_work_unreg_work_item == work || + slow_work_unreg_module == module) + wake_up_all(&slow_work_unreg_wq); +} +static void slow_work_clear_thread_processing(int id) +{ + slow_work_thread_processing[id] = NULL; +} +#else +static void slow_work_set_thread_processing(int id, struct slow_work *work) {} +static void slow_work_done_thread_processing(int id, struct slow_work *work) {} +static void slow_work_clear_thread_processing(int id) {} +#endif + +/* + * Data for tracking currently executing items for indication through /proc + */ +#ifdef CONFIG_SLOW_WORK_DEBUG +struct slow_work *slow_work_execs[SLOW_WORK_THREAD_LIMIT]; +pid_t slow_work_pids[SLOW_WORK_THREAD_LIMIT]; +DEFINE_RWLOCK(slow_work_execs_lock); +#endif + +/* + * The queues of work items and the lock governing access to them. These are + * shared between all the CPUs. It doesn't make sense to have per-CPU queues + * as the number of threads bears no relation to the number of CPUs. + * + * There are two queues of work items: one for slow work items, and one for + * very slow work items. + */ +LIST_HEAD(slow_work_queue); +LIST_HEAD(vslow_work_queue); +DEFINE_SPINLOCK(slow_work_queue_lock); + +/* + * The following are two wait queues that get pinged when a work item is placed + * on an empty queue. These allow work items that are hogging a thread by + * sleeping in a way that could be deferred to yield their thread and enqueue + * themselves. + */ +static DECLARE_WAIT_QUEUE_HEAD(slow_work_queue_waits_for_occupation); +static DECLARE_WAIT_QUEUE_HEAD(vslow_work_queue_waits_for_occupation); + +/* + * The thread controls. A variable used to signal to the threads that they + * should exit when the queue is empty, a waitqueue used by the threads to wait + * for signals, and a completion set by the last thread to exit. + */ +static bool slow_work_threads_should_exit; +static DECLARE_WAIT_QUEUE_HEAD(slow_work_thread_wq); +static DECLARE_COMPLETION(slow_work_last_thread_exited); + +/* + * The number of users of the thread pool and its lock. Whilst this is zero we + * have no threads hanging around, and when this reaches zero, we wait for all + * active or queued work items to complete and kill all the threads we do have. + */ +static int slow_work_user_count; +static DEFINE_MUTEX(slow_work_user_lock); + +static inline int slow_work_get_ref(struct slow_work *work) +{ + if (work->ops->get_ref) + return work->ops->get_ref(work); + + return 0; +} + +static inline void slow_work_put_ref(struct slow_work *work) +{ + if (work->ops->put_ref) + work->ops->put_ref(work); +} + +/* + * Calculate the maximum number of active threads in the pool that are + * permitted to process very slow work items. + * + * The answer is rounded up to at least 1, but may not equal or exceed the + * maximum number of the threads in the pool. This means we always have at + * least one thread that can process slow work items, and we always have at + * least one thread that won't get tied up doing so. + */ +static unsigned slow_work_calc_vsmax(void) +{ + unsigned vsmax; + + vsmax = atomic_read(&slow_work_thread_count) * vslow_work_proportion; + vsmax /= 100; + vsmax = max(vsmax, 1U); + return min(vsmax, slow_work_max_threads - 1); +} + +/* + * Attempt to execute stuff queued on a slow thread. Return true if we managed + * it, false if there was nothing to do. + */ +static noinline bool slow_work_execute(int id) +{ + struct slow_work *work = NULL; + unsigned vsmax; + bool very_slow; + + vsmax = slow_work_calc_vsmax(); + + /* see if we can schedule a new thread to be started if we're not + * keeping up with the work */ + if (!waitqueue_active(&slow_work_thread_wq) && + (!list_empty(&slow_work_queue) || !list_empty(&vslow_work_queue)) && + atomic_read(&slow_work_thread_count) < slow_work_max_threads && + !slow_work_may_not_start_new_thread) + slow_work_enqueue(&slow_work_new_thread); + + /* find something to execute */ + spin_lock_irq(&slow_work_queue_lock); + if (!list_empty(&vslow_work_queue) && + atomic_read(&vslow_work_executing_count) < vsmax) { + work = list_entry(vslow_work_queue.next, + struct slow_work, link); + if (test_and_set_bit_lock(SLOW_WORK_EXECUTING, &work->flags)) + BUG(); + list_del_init(&work->link); + atomic_inc(&vslow_work_executing_count); + very_slow = true; + } else if (!list_empty(&slow_work_queue)) { + work = list_entry(slow_work_queue.next, + struct slow_work, link); + if (test_and_set_bit_lock(SLOW_WORK_EXECUTING, &work->flags)) + BUG(); + list_del_init(&work->link); + very_slow = false; + } else { + very_slow = false; /* avoid the compiler warning */ + } + + slow_work_set_thread_processing(id, work); + if (work) { + slow_work_mark_time(work); + slow_work_begin_exec(id, work); + } + + spin_unlock_irq(&slow_work_queue_lock); + + if (!work) + return false; + + if (!test_and_clear_bit(SLOW_WORK_PENDING, &work->flags)) + BUG(); + + /* don't execute if the work is in the process of being cancelled */ + if (!test_bit(SLOW_WORK_CANCELLING, &work->flags)) + work->ops->execute(work); + + if (very_slow) + atomic_dec(&vslow_work_executing_count); + clear_bit_unlock(SLOW_WORK_EXECUTING, &work->flags); + + /* wake up anyone waiting for this work to be complete */ + wake_up_bit(&work->flags, SLOW_WORK_EXECUTING); + + slow_work_end_exec(id, work); + + /* if someone tried to enqueue the item whilst we were executing it, + * then it'll be left unenqueued to avoid multiple threads trying to + * execute it simultaneously + * + * there is, however, a race between us testing the pending flag and + * getting the spinlock, and between the enqueuer setting the pending + * flag and getting the spinlock, so we use a deferral bit to tell us + * if the enqueuer got there first + */ + if (test_bit(SLOW_WORK_PENDING, &work->flags)) { + spin_lock_irq(&slow_work_queue_lock); + + if (!test_bit(SLOW_WORK_EXECUTING, &work->flags) && + test_and_clear_bit(SLOW_WORK_ENQ_DEFERRED, &work->flags)) + goto auto_requeue; + + spin_unlock_irq(&slow_work_queue_lock); + } + + /* sort out the race between module unloading and put_ref() */ + slow_work_put_ref(work); + slow_work_done_thread_processing(id, work); + + return true; + +auto_requeue: + /* we must complete the enqueue operation + * - we transfer our ref on the item back to the appropriate queue + * - don't wake another thread up as we're awake already + */ + slow_work_mark_time(work); + if (test_bit(SLOW_WORK_VERY_SLOW, &work->flags)) + list_add_tail(&work->link, &vslow_work_queue); + else + list_add_tail(&work->link, &slow_work_queue); + spin_unlock_irq(&slow_work_queue_lock); + slow_work_clear_thread_processing(id); + return true; +} + +/** + * slow_work_sleep_till_thread_needed - Sleep till thread needed by other work + * work: The work item under execution that wants to sleep + * _timeout: Scheduler sleep timeout + * + * Allow a requeueable work item to sleep on a slow-work processor thread until + * that thread is needed to do some other work or the sleep is interrupted by + * some other event. + * + * The caller must set up a wake up event before calling this and must have set + * the appropriate sleep mode (such as TASK_UNINTERRUPTIBLE) and tested its own + * condition before calling this function as no test is made here. + * + * False is returned if there is nothing on the queue; true is returned if the + * work item should be requeued + */ +bool slow_work_sleep_till_thread_needed(struct slow_work *work, + signed long *_timeout) +{ + wait_queue_head_t *wfo_wq; + struct list_head *queue; + + DEFINE_WAIT(wait); + + if (test_bit(SLOW_WORK_VERY_SLOW, &work->flags)) { + wfo_wq = &vslow_work_queue_waits_for_occupation; + queue = &vslow_work_queue; + } else { + wfo_wq = &slow_work_queue_waits_for_occupation; + queue = &slow_work_queue; + } + + if (!list_empty(queue)) + return true; + + add_wait_queue_exclusive(wfo_wq, &wait); + if (list_empty(queue)) + *_timeout = schedule_timeout(*_timeout); + finish_wait(wfo_wq, &wait); + + return !list_empty(queue); +} +EXPORT_SYMBOL(slow_work_sleep_till_thread_needed); + +/** + * slow_work_enqueue - Schedule a slow work item for processing + * @work: The work item to queue + * + * Schedule a slow work item for processing. If the item is already undergoing + * execution, this guarantees not to re-enter the execution routine until the + * first execution finishes. + * + * The item is pinned by this function as it retains a reference to it, managed + * through the item operations. The item is unpinned once it has been + * executed. + * + * An item may hog the thread that is running it for a relatively large amount + * of time, sufficient, for example, to perform several lookup, mkdir, create + * and setxattr operations. It may sleep on I/O and may sleep to obtain locks. + * + * Conversely, if a number of items are awaiting processing, it may take some + * time before any given item is given attention. The number of threads in the + * pool may be increased to deal with demand, but only up to a limit. + * + * If SLOW_WORK_VERY_SLOW is set on the work item, then it will be placed in + * the very slow queue, from which only a portion of the threads will be + * allowed to pick items to execute. This ensures that very slow items won't + * overly block ones that are just ordinarily slow. + * + * Returns 0 if successful, -EAGAIN if not (or -ECANCELED if cancelled work is + * attempted queued) + */ +int slow_work_enqueue(struct slow_work *work) +{ + wait_queue_head_t *wfo_wq; + struct list_head *queue; + unsigned long flags; + int ret; + + if (test_bit(SLOW_WORK_CANCELLING, &work->flags)) + return -ECANCELED; + + BUG_ON(slow_work_user_count <= 0); + BUG_ON(!work); + BUG_ON(!work->ops); + + /* when honouring an enqueue request, we only promise that we will run + * the work function in the future; we do not promise to run it once + * per enqueue request + * + * we use the PENDING bit to merge together repeat requests without + * having to disable IRQs and take the spinlock, whilst still + * maintaining our promise + */ + if (!test_and_set_bit_lock(SLOW_WORK_PENDING, &work->flags)) { + if (test_bit(SLOW_WORK_VERY_SLOW, &work->flags)) { + wfo_wq = &vslow_work_queue_waits_for_occupation; + queue = &vslow_work_queue; + } else { + wfo_wq = &slow_work_queue_waits_for_occupation; + queue = &slow_work_queue; + } + + spin_lock_irqsave(&slow_work_queue_lock, flags); + + if (unlikely(test_bit(SLOW_WORK_CANCELLING, &work->flags))) + goto cancelled; + + /* we promise that we will not attempt to execute the work + * function in more than one thread simultaneously + * + * this, however, leaves us with a problem if we're asked to + * enqueue the work whilst someone is executing the work + * function as simply queueing the work immediately means that + * another thread may try executing it whilst it is already + * under execution + * + * to deal with this, we set the ENQ_DEFERRED bit instead of + * enqueueing, and the thread currently executing the work + * function will enqueue the work item when the work function + * returns and it has cleared the EXECUTING bit + */ + if (test_bit(SLOW_WORK_EXECUTING, &work->flags)) { + set_bit(SLOW_WORK_ENQ_DEFERRED, &work->flags); + } else { + ret = slow_work_get_ref(work); + if (ret < 0) + goto failed; + slow_work_mark_time(work); + list_add_tail(&work->link, queue); + wake_up(&slow_work_thread_wq); + + /* if someone who could be requeued is sleeping on a + * thread, then ask them to yield their thread */ + if (work->link.prev == queue) + wake_up(wfo_wq); + } + + spin_unlock_irqrestore(&slow_work_queue_lock, flags); + } + return 0; + +cancelled: + ret = -ECANCELED; +failed: + spin_unlock_irqrestore(&slow_work_queue_lock, flags); + return ret; +} +EXPORT_SYMBOL(slow_work_enqueue); + +static int slow_work_wait(void *word) +{ + schedule(); + return 0; +} + +/** + * slow_work_cancel - Cancel a slow work item + * @work: The work item to cancel + * + * This function will cancel a previously enqueued work item. If we cannot + * cancel the work item, it is guarenteed to have run when this function + * returns. + */ +void slow_work_cancel(struct slow_work *work) +{ + bool wait = true, put = false; + + set_bit(SLOW_WORK_CANCELLING, &work->flags); + smp_mb(); + + /* if the work item is a delayed work item with an active timer, we + * need to wait for the timer to finish _before_ getting the spinlock, + * lest we deadlock against the timer routine + * + * the timer routine will leave DELAYED set if it notices the + * CANCELLING flag in time + */ + if (test_bit(SLOW_WORK_DELAYED, &work->flags)) { + struct delayed_slow_work *dwork = + container_of(work, struct delayed_slow_work, work); + del_timer_sync(&dwork->timer); + } + + spin_lock_irq(&slow_work_queue_lock); + + if (test_bit(SLOW_WORK_DELAYED, &work->flags)) { + /* the timer routine aborted or never happened, so we are left + * holding the timer's reference on the item and should just + * drop the pending flag and wait for any ongoing execution to + * finish */ + struct delayed_slow_work *dwork = + container_of(work, struct delayed_slow_work, work); + + BUG_ON(timer_pending(&dwork->timer)); + BUG_ON(!list_empty(&work->link)); + + clear_bit(SLOW_WORK_DELAYED, &work->flags); + put = true; + clear_bit(SLOW_WORK_PENDING, &work->flags); + + } else if (test_bit(SLOW_WORK_PENDING, &work->flags) && + !list_empty(&work->link)) { + /* the link in the pending queue holds a reference on the item + * that we will need to release */ + list_del_init(&work->link); + wait = false; + put = true; + clear_bit(SLOW_WORK_PENDING, &work->flags); + + } else if (test_and_clear_bit(SLOW_WORK_ENQ_DEFERRED, &work->flags)) { + /* the executor is holding our only reference on the item, so + * we merely need to wait for it to finish executing */ + clear_bit(SLOW_WORK_PENDING, &work->flags); + } + + spin_unlock_irq(&slow_work_queue_lock); + + /* the EXECUTING flag is set by the executor whilst the spinlock is set + * and before the item is dequeued - so assuming the above doesn't + * actually dequeue it, simply waiting for the EXECUTING flag to be + * released here should be sufficient */ + if (wait) + wait_on_bit(&work->flags, SLOW_WORK_EXECUTING, slow_work_wait, + TASK_UNINTERRUPTIBLE); + + clear_bit(SLOW_WORK_CANCELLING, &work->flags); + if (put) + slow_work_put_ref(work); +} +EXPORT_SYMBOL(slow_work_cancel); + +/* + * Handle expiry of the delay timer, indicating that a delayed slow work item + * should now be queued if not cancelled + */ +static void delayed_slow_work_timer(unsigned long data) +{ + wait_queue_head_t *wfo_wq; + struct list_head *queue; + struct slow_work *work = (struct slow_work *) data; + unsigned long flags; + bool queued = false, put = false, first = false; + + if (test_bit(SLOW_WORK_VERY_SLOW, &work->flags)) { + wfo_wq = &vslow_work_queue_waits_for_occupation; + queue = &vslow_work_queue; + } else { + wfo_wq = &slow_work_queue_waits_for_occupation; + queue = &slow_work_queue; + } + + spin_lock_irqsave(&slow_work_queue_lock, flags); + if (likely(!test_bit(SLOW_WORK_CANCELLING, &work->flags))) { + clear_bit(SLOW_WORK_DELAYED, &work->flags); + + if (test_bit(SLOW_WORK_EXECUTING, &work->flags)) { + /* we discard the reference the timer was holding in + * favour of the one the executor holds */ + set_bit(SLOW_WORK_ENQ_DEFERRED, &work->flags); + put = true; + } else { + slow_work_mark_time(work); + list_add_tail(&work->link, queue); + queued = true; + if (work->link.prev == queue) + first = true; + } + } + + spin_unlock_irqrestore(&slow_work_queue_lock, flags); + if (put) + slow_work_put_ref(work); + if (first) + wake_up(wfo_wq); + if (queued) + wake_up(&slow_work_thread_wq); +} + +/** + * delayed_slow_work_enqueue - Schedule a delayed slow work item for processing + * @dwork: The delayed work item to queue + * @delay: When to start executing the work, in jiffies from now + * + * This is similar to slow_work_enqueue(), but it adds a delay before the work + * is actually queued for processing. + * + * The item can have delayed processing requested on it whilst it is being + * executed. The delay will begin immediately, and if it expires before the + * item finishes executing, the item will be placed back on the queue when it + * has done executing. + */ +int delayed_slow_work_enqueue(struct delayed_slow_work *dwork, + unsigned long delay) +{ + struct slow_work *work = &dwork->work; + unsigned long flags; + int ret; + + if (delay == 0) + return slow_work_enqueue(&dwork->work); + + BUG_ON(slow_work_user_count <= 0); + BUG_ON(!work); + BUG_ON(!work->ops); + + if (test_bit(SLOW_WORK_CANCELLING, &work->flags)) + return -ECANCELED; + + if (!test_and_set_bit_lock(SLOW_WORK_PENDING, &work->flags)) { + spin_lock_irqsave(&slow_work_queue_lock, flags); + + if (test_bit(SLOW_WORK_CANCELLING, &work->flags)) + goto cancelled; + + /* the timer holds a reference whilst it is pending */ + ret = work->ops->get_ref(work); + if (ret < 0) + goto cant_get_ref; + + if (test_and_set_bit(SLOW_WORK_DELAYED, &work->flags)) + BUG(); + dwork->timer.expires = jiffies + delay; + dwork->timer.data = (unsigned long) work; + dwork->timer.function = delayed_slow_work_timer; + add_timer(&dwork->timer); + + spin_unlock_irqrestore(&slow_work_queue_lock, flags); + } + + return 0; + +cancelled: + ret = -ECANCELED; +cant_get_ref: + spin_unlock_irqrestore(&slow_work_queue_lock, flags); + return ret; +} +EXPORT_SYMBOL(delayed_slow_work_enqueue); + +/* + * Schedule a cull of the thread pool at some time in the near future + */ +static void slow_work_schedule_cull(void) +{ + mod_timer(&slow_work_cull_timer, + round_jiffies(jiffies + SLOW_WORK_CULL_TIMEOUT)); +} + +/* + * Worker thread culling algorithm + */ +static bool slow_work_cull_thread(void) +{ + unsigned long flags; + bool do_cull = false; + + spin_lock_irqsave(&slow_work_queue_lock, flags); + + if (slow_work_cull) { + slow_work_cull = false; + + if (list_empty(&slow_work_queue) && + list_empty(&vslow_work_queue) && + atomic_read(&slow_work_thread_count) > + slow_work_min_threads) { + slow_work_schedule_cull(); + do_cull = true; + } + } + + spin_unlock_irqrestore(&slow_work_queue_lock, flags); + return do_cull; +} + +/* + * Determine if there is slow work available for dispatch + */ +static inline bool slow_work_available(int vsmax) +{ + return !list_empty(&slow_work_queue) || + (!list_empty(&vslow_work_queue) && + atomic_read(&vslow_work_executing_count) < vsmax); +} + +/* + * Worker thread dispatcher + */ +static int slow_work_thread(void *_data) +{ + int vsmax, id; + + DEFINE_WAIT(wait); + + set_freezable(); + set_user_nice(current, -5); + + /* allocate ourselves an ID */ + spin_lock_irq(&slow_work_queue_lock); + id = find_first_zero_bit(slow_work_ids, SLOW_WORK_THREAD_LIMIT); + BUG_ON(id < 0 || id >= SLOW_WORK_THREAD_LIMIT); + __set_bit(id, slow_work_ids); + slow_work_set_thread_pid(id, current->pid); + spin_unlock_irq(&slow_work_queue_lock); + + sprintf(current->comm, "kslowd%03u", id); + + for (;;) { + vsmax = vslow_work_proportion; + vsmax *= atomic_read(&slow_work_thread_count); + vsmax /= 100; + + prepare_to_wait_exclusive(&slow_work_thread_wq, &wait, + TASK_INTERRUPTIBLE); + if (!freezing(current) && + !slow_work_threads_should_exit && + !slow_work_available(vsmax) && + !slow_work_cull) + schedule(); + finish_wait(&slow_work_thread_wq, &wait); + + try_to_freeze(); + + vsmax = vslow_work_proportion; + vsmax *= atomic_read(&slow_work_thread_count); + vsmax /= 100; + + if (slow_work_available(vsmax) && slow_work_execute(id)) { + cond_resched(); + if (list_empty(&slow_work_queue) && + list_empty(&vslow_work_queue) && + atomic_read(&slow_work_thread_count) > + slow_work_min_threads) + slow_work_schedule_cull(); + continue; + } + + if (slow_work_threads_should_exit) + break; + + if (slow_work_cull && slow_work_cull_thread()) + break; + } + + spin_lock_irq(&slow_work_queue_lock); + slow_work_set_thread_pid(id, 0); + __clear_bit(id, slow_work_ids); + spin_unlock_irq(&slow_work_queue_lock); + + if (atomic_dec_and_test(&slow_work_thread_count)) + complete_and_exit(&slow_work_last_thread_exited, 0); + return 0; +} + +/* + * Handle thread cull timer expiration + */ +static void slow_work_cull_timeout(unsigned long data) +{ + slow_work_cull = true; + wake_up(&slow_work_thread_wq); +} + +/* + * Start a new slow work thread + */ +static void slow_work_new_thread_execute(struct slow_work *work) +{ + struct task_struct *p; + + if (slow_work_threads_should_exit) + return; + + if (atomic_read(&slow_work_thread_count) >= slow_work_max_threads) + return; + + if (!mutex_trylock(&slow_work_user_lock)) + return; + + slow_work_may_not_start_new_thread = true; + atomic_inc(&slow_work_thread_count); + p = kthread_run(slow_work_thread, NULL, "kslowd"); + if (IS_ERR(p)) { + printk(KERN_DEBUG "Slow work thread pool: OOM\n"); + if (atomic_dec_and_test(&slow_work_thread_count)) + BUG(); /* we're running on a slow work thread... */ + mod_timer(&slow_work_oom_timer, + round_jiffies(jiffies + SLOW_WORK_OOM_TIMEOUT)); + } else { + /* ratelimit the starting of new threads */ + mod_timer(&slow_work_oom_timer, jiffies + 1); + } + + mutex_unlock(&slow_work_user_lock); +} + +static const struct slow_work_ops slow_work_new_thread_ops = { + .owner = THIS_MODULE, + .execute = slow_work_new_thread_execute, +#ifdef CONFIG_SLOW_WORK_DEBUG + .desc = slow_work_new_thread_desc, +#endif +}; + +/* + * post-OOM new thread start suppression expiration + */ +static void slow_work_oom_timeout(unsigned long data) +{ + slow_work_may_not_start_new_thread = false; +} + +#ifdef CONFIG_SYSCTL +/* + * Handle adjustment of the minimum number of threads + */ +static int slow_work_min_threads_sysctl(struct ctl_table *table, int write, + void __user *buffer, + size_t *lenp, loff_t *ppos) +{ + int ret = proc_dointvec_minmax(table, write, buffer, lenp, ppos); + int n; + + if (ret == 0) { + mutex_lock(&slow_work_user_lock); + if (slow_work_user_count > 0) { + /* see if we need to start or stop threads */ + n = atomic_read(&slow_work_thread_count) - + slow_work_min_threads; + + if (n < 0 && !slow_work_may_not_start_new_thread) + slow_work_enqueue(&slow_work_new_thread); + else if (n > 0) + slow_work_schedule_cull(); + } + mutex_unlock(&slow_work_user_lock); + } + + return ret; +} + +/* + * Handle adjustment of the maximum number of threads + */ +static int slow_work_max_threads_sysctl(struct ctl_table *table, int write, + void __user *buffer, + size_t *lenp, loff_t *ppos) +{ + int ret = proc_dointvec_minmax(table, write, buffer, lenp, ppos); + int n; + + if (ret == 0) { + mutex_lock(&slow_work_user_lock); + if (slow_work_user_count > 0) { + /* see if we need to stop threads */ + n = slow_work_max_threads - + atomic_read(&slow_work_thread_count); + + if (n < 0) + slow_work_schedule_cull(); + } + mutex_unlock(&slow_work_user_lock); + } + + return ret; +} +#endif /* CONFIG_SYSCTL */ + +/** + * slow_work_register_user - Register a user of the facility + * @module: The module about to make use of the facility + * + * Register a user of the facility, starting up the initial threads if there + * aren't any other users at this point. This will return 0 if successful, or + * an error if not. + */ +int slow_work_register_user(struct module *module) +{ + struct task_struct *p; + int loop; + + mutex_lock(&slow_work_user_lock); + + if (slow_work_user_count == 0) { + printk(KERN_NOTICE "Slow work thread pool: Starting up\n"); + init_completion(&slow_work_last_thread_exited); + + slow_work_threads_should_exit = false; + slow_work_init(&slow_work_new_thread, + &slow_work_new_thread_ops); + slow_work_may_not_start_new_thread = false; + slow_work_cull = false; + + /* start the minimum number of threads */ + for (loop = 0; loop < slow_work_min_threads; loop++) { + atomic_inc(&slow_work_thread_count); + p = kthread_run(slow_work_thread, NULL, "kslowd"); + if (IS_ERR(p)) + goto error; + } + printk(KERN_NOTICE "Slow work thread pool: Ready\n"); + } + + slow_work_user_count++; + mutex_unlock(&slow_work_user_lock); + return 0; + +error: + if (atomic_dec_and_test(&slow_work_thread_count)) + complete(&slow_work_last_thread_exited); + if (loop > 0) { + printk(KERN_ERR "Slow work thread pool:" + " Aborting startup on ENOMEM\n"); + slow_work_threads_should_exit = true; + wake_up_all(&slow_work_thread_wq); + wait_for_completion(&slow_work_last_thread_exited); + printk(KERN_ERR "Slow work thread pool: Aborted\n"); + } + mutex_unlock(&slow_work_user_lock); + return PTR_ERR(p); +} +EXPORT_SYMBOL(slow_work_register_user); + +/* + * wait for all outstanding items from the calling module to complete + * - note that more items may be queued whilst we're waiting + */ +static void slow_work_wait_for_items(struct module *module) +{ +#ifdef CONFIG_MODULES + DECLARE_WAITQUEUE(myself, current); + struct slow_work *work; + int loop; + + mutex_lock(&slow_work_unreg_sync_lock); + add_wait_queue(&slow_work_unreg_wq, &myself); + + for (;;) { + spin_lock_irq(&slow_work_queue_lock); + + /* first of all, we wait for the last queued item in each list + * to be processed */ + list_for_each_entry_reverse(work, &vslow_work_queue, link) { + if (work->owner == module) { + set_current_state(TASK_UNINTERRUPTIBLE); + slow_work_unreg_work_item = work; + goto do_wait; + } + } + list_for_each_entry_reverse(work, &slow_work_queue, link) { + if (work->owner == module) { + set_current_state(TASK_UNINTERRUPTIBLE); + slow_work_unreg_work_item = work; + goto do_wait; + } + } + + /* then we wait for the items being processed to finish */ + slow_work_unreg_module = module; + smp_mb(); + for (loop = 0; loop < SLOW_WORK_THREAD_LIMIT; loop++) { + if (slow_work_thread_processing[loop] == module) + goto do_wait; + } + spin_unlock_irq(&slow_work_queue_lock); + break; /* okay, we're done */ + + do_wait: + spin_unlock_irq(&slow_work_queue_lock); + schedule(); + slow_work_unreg_work_item = NULL; + slow_work_unreg_module = NULL; + } + + remove_wait_queue(&slow_work_unreg_wq, &myself); + mutex_unlock(&slow_work_unreg_sync_lock); +#endif /* CONFIG_MODULES */ +} + +/** + * slow_work_unregister_user - Unregister a user of the facility + * @module: The module whose items should be cleared + * + * Unregister a user of the facility, killing all the threads if this was the + * last one. + * + * This waits for all the work items belonging to the nominated module to go + * away before proceeding. + */ +void slow_work_unregister_user(struct module *module) +{ + /* first of all, wait for all outstanding items from the calling module + * to complete */ + if (module) + slow_work_wait_for_items(module); + + /* then we can actually go about shutting down the facility if need + * be */ + mutex_lock(&slow_work_user_lock); + + BUG_ON(slow_work_user_count <= 0); + + slow_work_user_count--; + if (slow_work_user_count == 0) { + printk(KERN_NOTICE "Slow work thread pool: Shutting down\n"); + slow_work_threads_should_exit = true; + del_timer_sync(&slow_work_cull_timer); + del_timer_sync(&slow_work_oom_timer); + wake_up_all(&slow_work_thread_wq); + wait_for_completion(&slow_work_last_thread_exited); + printk(KERN_NOTICE "Slow work thread pool:" + " Shut down complete\n"); + } + + mutex_unlock(&slow_work_user_lock); +} +EXPORT_SYMBOL(slow_work_unregister_user); + +/* + * Initialise the slow work facility + */ +static int __init init_slow_work(void) +{ + unsigned nr_cpus = num_possible_cpus(); + + if (slow_work_max_threads < nr_cpus) + slow_work_max_threads = nr_cpus; +#ifdef CONFIG_SYSCTL + if (slow_work_max_max_threads < nr_cpus * 2) + slow_work_max_max_threads = nr_cpus * 2; +#endif +#ifdef CONFIG_SLOW_WORK_DEBUG + { + struct dentry *dbdir; + + dbdir = debugfs_create_dir("slow_work", NULL); + if (dbdir && !IS_ERR(dbdir)) + debugfs_create_file("runqueue", S_IFREG | 0400, dbdir, + NULL, &slow_work_runqueue_fops); + } +#endif + return 0; +} + +subsys_initcall(init_slow_work); diff --git a/kernel/slow-work.h b/kernel/slow-work.h new file mode 100644 index 00000000000..321f3c59d73 --- /dev/null +++ b/kernel/slow-work.h @@ -0,0 +1,72 @@ +/* Slow work private definitions + * + * Copyright (C) 2009 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public Licence + * as published by the Free Software Foundation; either version + * 2 of the Licence, or (at your option) any later version. + */ + +#define SLOW_WORK_CULL_TIMEOUT (5 * HZ) /* cull threads 5s after running out of + * things to do */ +#define SLOW_WORK_OOM_TIMEOUT (5 * HZ) /* can't start new threads for 5s after + * OOM */ + +#define SLOW_WORK_THREAD_LIMIT 255 /* abs maximum number of slow-work threads */ + +/* + * slow-work.c + */ +#ifdef CONFIG_SLOW_WORK_DEBUG +extern struct slow_work *slow_work_execs[]; +extern pid_t slow_work_pids[]; +extern rwlock_t slow_work_execs_lock; +#endif + +extern struct list_head slow_work_queue; +extern struct list_head vslow_work_queue; +extern spinlock_t slow_work_queue_lock; + +/* + * slow-work-debugfs.c + */ +#ifdef CONFIG_SLOW_WORK_DEBUG +extern const struct file_operations slow_work_runqueue_fops; + +extern void slow_work_new_thread_desc(struct slow_work *, struct seq_file *); +#endif + +/* + * Helper functions + */ +static inline void slow_work_set_thread_pid(int id, pid_t pid) +{ +#ifdef CONFIG_SLOW_WORK_PROC + slow_work_pids[id] = pid; +#endif +} + +static inline void slow_work_mark_time(struct slow_work *work) +{ +#ifdef CONFIG_SLOW_WORK_PROC + work->mark = CURRENT_TIME; +#endif +} + +static inline void slow_work_begin_exec(int id, struct slow_work *work) +{ +#ifdef CONFIG_SLOW_WORK_PROC + slow_work_execs[id] = work; +#endif +} + +static inline void slow_work_end_exec(int id, struct slow_work *work) +{ +#ifdef CONFIG_SLOW_WORK_PROC + write_lock(&slow_work_execs_lock); + slow_work_execs[id] = NULL; + write_unlock(&slow_work_execs_lock); +#endif +} diff --git a/kernel/smp.c b/kernel/smp.c index bbedbb7efe3..a8c76069cf5 100644 --- a/kernel/smp.c +++ b/kernel/smp.c @@ -2,40 +2,79 @@ * Generic helpers for smp ipi calls * * (C) Jens Axboe <jens.axboe@oracle.com> 2008 - * */ -#include <linux/init.h> -#include <linux/module.h> -#include <linux/percpu.h> #include <linux/rcupdate.h> #include <linux/rculist.h> +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/percpu.h> +#include <linux/init.h> #include <linux/smp.h> +#include <linux/cpu.h> static DEFINE_PER_CPU(struct call_single_queue, call_single_queue); -static LIST_HEAD(call_function_queue); -__cacheline_aligned_in_smp DEFINE_SPINLOCK(call_function_lock); + +static struct { + struct list_head queue; + spinlock_t lock; +} call_function __cacheline_aligned_in_smp = + { + .queue = LIST_HEAD_INIT(call_function.queue), + .lock = __SPIN_LOCK_UNLOCKED(call_function.lock), + }; enum { - CSD_FLAG_WAIT = 0x01, - CSD_FLAG_ALLOC = 0x02, - CSD_FLAG_LOCK = 0x04, + CSD_FLAG_LOCK = 0x01, }; struct call_function_data { - struct call_single_data csd; - spinlock_t lock; - unsigned int refs; - struct rcu_head rcu_head; - unsigned long cpumask_bits[]; + struct call_single_data csd; + atomic_t refs; + cpumask_var_t cpumask; }; struct call_single_queue { - struct list_head list; - spinlock_t lock; + struct list_head list; + spinlock_t lock; +}; + +static DEFINE_PER_CPU(struct call_function_data, cfd_data); + +static int +hotplug_cfd(struct notifier_block *nfb, unsigned long action, void *hcpu) +{ + long cpu = (long)hcpu; + struct call_function_data *cfd = &per_cpu(cfd_data, cpu); + + switch (action) { + case CPU_UP_PREPARE: + case CPU_UP_PREPARE_FROZEN: + if (!zalloc_cpumask_var_node(&cfd->cpumask, GFP_KERNEL, + cpu_to_node(cpu))) + return NOTIFY_BAD; + break; + +#ifdef CONFIG_HOTPLUG_CPU + case CPU_UP_CANCELED: + case CPU_UP_CANCELED_FROZEN: + + case CPU_DEAD: + case CPU_DEAD_FROZEN: + free_cpumask_var(cfd->cpumask); + break; +#endif + }; + + return NOTIFY_OK; +} + +static struct notifier_block __cpuinitdata hotplug_cfd_notifier = { + .notifier_call = hotplug_cfd, }; static int __cpuinit init_call_single_data(void) { + void *cpu = (void *)(long)smp_processor_id(); int i; for_each_possible_cpu(i) { @@ -44,29 +83,63 @@ static int __cpuinit init_call_single_data(void) spin_lock_init(&q->lock); INIT_LIST_HEAD(&q->list); } + + hotplug_cfd(&hotplug_cfd_notifier, CPU_UP_PREPARE, cpu); + register_cpu_notifier(&hotplug_cfd_notifier); + return 0; } early_initcall(init_call_single_data); -static void csd_flag_wait(struct call_single_data *data) +/* + * csd_lock/csd_unlock used to serialize access to per-cpu csd resources + * + * For non-synchronous ipi calls the csd can still be in use by the + * previous function call. For multi-cpu calls its even more interesting + * as we'll have to ensure no other cpu is observing our csd. + */ +static void csd_lock_wait(struct call_single_data *data) { - /* Wait for response */ - do { - if (!(data->flags & CSD_FLAG_WAIT)) - break; + while (data->flags & CSD_FLAG_LOCK) cpu_relax(); - } while (1); +} + +static void csd_lock(struct call_single_data *data) +{ + csd_lock_wait(data); + data->flags = CSD_FLAG_LOCK; + + /* + * prevent CPU from reordering the above assignment + * to ->flags with any subsequent assignments to other + * fields of the specified call_single_data structure: + */ + smp_mb(); +} + +static void csd_unlock(struct call_single_data *data) +{ + WARN_ON(!(data->flags & CSD_FLAG_LOCK)); + + /* + * ensure we're all done before releasing data: + */ + smp_mb(); + + data->flags &= ~CSD_FLAG_LOCK; } /* - * Insert a previously allocated call_single_data element for execution - * on the given CPU. data must already have ->func, ->info, and ->flags set. + * Insert a previously allocated call_single_data element + * for execution on the given CPU. data must already have + * ->func, ->info, and ->flags set. */ -static void generic_exec_single(int cpu, struct call_single_data *data) +static +void generic_exec_single(int cpu, struct call_single_data *data, int wait) { struct call_single_queue *dst = &per_cpu(call_single_queue, cpu); - int wait = data->flags & CSD_FLAG_WAIT, ipi; unsigned long flags; + int ipi; spin_lock_irqsave(&dst->lock, flags); ipi = list_empty(&dst->list); @@ -74,24 +147,21 @@ static void generic_exec_single(int cpu, struct call_single_data *data) spin_unlock_irqrestore(&dst->lock, flags); /* - * Make the list addition visible before sending the ipi. + * The list addition should be visible before sending the IPI + * handler locks the list to pull the entry off it because of + * normal cache coherency rules implied by spinlocks. + * + * If IPIs can go out of order to the cache coherency protocol + * in an architecture, sufficient synchronisation should be added + * to arch code to make it appear to obey cache coherency WRT + * locking and barrier primitives. Generic code isn't really + * equipped to do the right thing... */ - smp_mb(); - if (ipi) arch_send_call_function_single_ipi(cpu); if (wait) - csd_flag_wait(data); -} - -static void rcu_free_call_data(struct rcu_head *head) -{ - struct call_function_data *data; - - data = container_of(head, struct call_function_data, rcu_head); - - kfree(data); + csd_lock_wait(data); } /* @@ -104,99 +174,86 @@ void generic_smp_call_function_interrupt(void) int cpu = get_cpu(); /* - * It's ok to use list_for_each_rcu() here even though we may delete - * 'pos', since list_del_rcu() doesn't clear ->next + * Shouldn't receive this interrupt on a cpu that is not yet online. + */ + WARN_ON_ONCE(!cpu_online(cpu)); + + /* + * Ensure entry is visible on call_function_queue after we have + * entered the IPI. See comment in smp_call_function_many. + * If we don't have this, then we may miss an entry on the list + * and never get another IPI to process it. + */ + smp_mb(); + + /* + * It's ok to use list_for_each_rcu() here even though we may + * delete 'pos', since list_del_rcu() doesn't clear ->next */ - rcu_read_lock(); - list_for_each_entry_rcu(data, &call_function_queue, csd.list) { + list_for_each_entry_rcu(data, &call_function.queue, csd.list) { int refs; - if (!cpumask_test_cpu(cpu, to_cpumask(data->cpumask_bits))) + if (!cpumask_test_and_clear_cpu(cpu, data->cpumask)) continue; data->csd.func(data->csd.info); - spin_lock(&data->lock); - cpumask_clear_cpu(cpu, to_cpumask(data->cpumask_bits)); - WARN_ON(data->refs == 0); - data->refs--; - refs = data->refs; - spin_unlock(&data->lock); + refs = atomic_dec_return(&data->refs); + WARN_ON(refs < 0); + if (!refs) { + spin_lock(&call_function.lock); + list_del_rcu(&data->csd.list); + spin_unlock(&call_function.lock); + } if (refs) continue; - spin_lock(&call_function_lock); - list_del_rcu(&data->csd.list); - spin_unlock(&call_function_lock); - - if (data->csd.flags & CSD_FLAG_WAIT) { - /* - * serialize stores to data with the flag clear - * and wakeup - */ - smp_wmb(); - data->csd.flags &= ~CSD_FLAG_WAIT; - } - if (data->csd.flags & CSD_FLAG_ALLOC) - call_rcu(&data->rcu_head, rcu_free_call_data); + csd_unlock(&data->csd); } - rcu_read_unlock(); put_cpu(); } /* - * Invoked by arch to handle an IPI for call function single. Must be called - * from the arch with interrupts disabled. + * Invoked by arch to handle an IPI for call function single. Must be + * called from the arch with interrupts disabled. */ void generic_smp_call_function_single_interrupt(void) { struct call_single_queue *q = &__get_cpu_var(call_single_queue); + unsigned int data_flags; LIST_HEAD(list); /* - * Need to see other stores to list head for checking whether - * list is empty without holding q->lock + * Shouldn't receive this interrupt on a cpu that is not yet online. */ - smp_read_barrier_depends(); - while (!list_empty(&q->list)) { - unsigned int data_flags; - - spin_lock(&q->lock); - list_replace_init(&q->list, &list); - spin_unlock(&q->lock); - - while (!list_empty(&list)) { - struct call_single_data *data; - - data = list_entry(list.next, struct call_single_data, - list); - list_del(&data->list); - - /* - * 'data' can be invalid after this call if - * flags == 0 (when called through - * generic_exec_single(), so save them away before - * making the call. - */ - data_flags = data->flags; - - data->func(data->info); - - if (data_flags & CSD_FLAG_WAIT) { - smp_wmb(); - data->flags &= ~CSD_FLAG_WAIT; - } else if (data_flags & CSD_FLAG_LOCK) { - smp_wmb(); - data->flags &= ~CSD_FLAG_LOCK; - } else if (data_flags & CSD_FLAG_ALLOC) - kfree(data); - } + WARN_ON_ONCE(!cpu_online(smp_processor_id())); + + spin_lock(&q->lock); + list_replace_init(&q->list, &list); + spin_unlock(&q->lock); + + while (!list_empty(&list)) { + struct call_single_data *data; + + data = list_entry(list.next, struct call_single_data, list); + list_del(&data->list); + + /* + * 'data' can be invalid after this call if flags == 0 + * (when called through generic_exec_single()), + * so save them away before making the call: + */ + data_flags = data->flags; + + data->func(data->info); + /* - * See comment on outer loop + * Unlocked CSDs are valid through generic_exec_single(): */ - smp_read_barrier_depends(); + if (data_flags & CSD_FLAG_LOCK) + csd_unlock(data); } } @@ -208,137 +265,173 @@ static DEFINE_PER_CPU(struct call_single_data, csd_data); * @info: An arbitrary pointer to pass to the function. * @wait: If true, wait until function has completed on other CPUs. * - * Returns 0 on success, else a negative status code. Note that @wait - * will be implicitly turned on in case of allocation failures, since - * we fall back to on-stack allocation. + * Returns 0 on success, else a negative status code. */ int smp_call_function_single(int cpu, void (*func) (void *info), void *info, int wait) { - struct call_single_data d; + struct call_single_data d = { + .flags = 0, + }; unsigned long flags; - /* prevent preemption and reschedule on another processor, - as well as CPU removal */ - int me = get_cpu(); + int this_cpu; int err = 0; - /* Can deadlock when called with interrupts disabled */ - WARN_ON(irqs_disabled()); + /* + * prevent preemption and reschedule on another processor, + * as well as CPU removal + */ + this_cpu = get_cpu(); + + /* + * Can deadlock when called with interrupts disabled. + * We allow cpu's that are not yet online though, as no one else can + * send smp call function interrupt to this cpu and as such deadlocks + * can't happen. + */ + WARN_ON_ONCE(cpu_online(this_cpu) && irqs_disabled() + && !oops_in_progress); - if (cpu == me) { + if (cpu == this_cpu) { local_irq_save(flags); func(info); local_irq_restore(flags); - } else if ((unsigned)cpu < nr_cpu_ids && cpu_online(cpu)) { - struct call_single_data *data; + } else { + if ((unsigned)cpu < nr_cpu_ids && cpu_online(cpu)) { + struct call_single_data *data = &d; + + if (!wait) + data = &__get_cpu_var(csd_data); - if (!wait) { - /* - * We are calling a function on a single CPU - * and we are not going to wait for it to finish. - * We first try to allocate the data, but if we - * fail, we fall back to use a per cpu data to pass - * the information to that CPU. Since all callers - * of this code will use the same data, we must - * synchronize the callers to prevent a new caller - * from corrupting the data before the callee - * can access it. - * - * The CSD_FLAG_LOCK is used to let us know when - * the IPI handler is done with the data. - * The first caller will set it, and the callee - * will clear it. The next caller must wait for - * it to clear before we set it again. This - * will make sure the callee is done with the - * data before a new caller will use it. - */ - data = kmalloc(sizeof(*data), GFP_ATOMIC); - if (data) - data->flags = CSD_FLAG_ALLOC; - else { - data = &per_cpu(csd_data, me); - while (data->flags & CSD_FLAG_LOCK) - cpu_relax(); - data->flags = CSD_FLAG_LOCK; - } + csd_lock(data); + + data->func = func; + data->info = info; + generic_exec_single(cpu, data, wait); } else { - data = &d; - data->flags = CSD_FLAG_WAIT; + err = -ENXIO; /* CPU not online */ } - - data->func = func; - data->info = info; - generic_exec_single(cpu, data); - } else { - err = -ENXIO; /* CPU not online */ } put_cpu(); + return err; } EXPORT_SYMBOL(smp_call_function_single); +/* + * smp_call_function_any - Run a function on any of the given cpus + * @mask: The mask of cpus it can run on. + * @func: The function to run. This must be fast and non-blocking. + * @info: An arbitrary pointer to pass to the function. + * @wait: If true, wait until function has completed. + * + * Returns 0 on success, else a negative status code (if no cpus were online). + * Note that @wait will be implicitly turned on in case of allocation failures, + * since we fall back to on-stack allocation. + * + * Selection preference: + * 1) current cpu if in @mask + * 2) any cpu of current node if in @mask + * 3) any other online cpu in @mask + */ +int smp_call_function_any(const struct cpumask *mask, + void (*func)(void *info), void *info, int wait) +{ + unsigned int cpu; + const struct cpumask *nodemask; + int ret; + + /* Try for same CPU (cheapest) */ + cpu = get_cpu(); + if (cpumask_test_cpu(cpu, mask)) + goto call; + + /* Try for same node. */ + nodemask = cpumask_of_node(cpu); + for (cpu = cpumask_first_and(nodemask, mask); cpu < nr_cpu_ids; + cpu = cpumask_next_and(cpu, nodemask, mask)) { + if (cpu_online(cpu)) + goto call; + } + + /* Any online will do: smp_call_function_single handles nr_cpu_ids. */ + cpu = cpumask_any_and(mask, cpu_online_mask); +call: + ret = smp_call_function_single(cpu, func, info, wait); + put_cpu(); + return ret; +} +EXPORT_SYMBOL_GPL(smp_call_function_any); + /** * __smp_call_function_single(): Run a function on another CPU * @cpu: The CPU to run on. * @data: Pre-allocated and setup data structure * - * Like smp_call_function_single(), but allow caller to pass in a pre-allocated - * data structure. Useful for embedding @data inside other structures, for - * instance. - * + * Like smp_call_function_single(), but allow caller to pass in a + * pre-allocated data structure. Useful for embedding @data inside + * other structures, for instance. */ -void __smp_call_function_single(int cpu, struct call_single_data *data) +void __smp_call_function_single(int cpu, struct call_single_data *data, + int wait) { - /* Can deadlock when called with interrupts disabled */ - WARN_ON((data->flags & CSD_FLAG_WAIT) && irqs_disabled()); + csd_lock(data); - generic_exec_single(cpu, data); -} + /* + * Can deadlock when called with interrupts disabled. + * We allow cpu's that are not yet online though, as no one else can + * send smp call function interrupt to this cpu and as such deadlocks + * can't happen. + */ + WARN_ON_ONCE(cpu_online(smp_processor_id()) && wait && irqs_disabled() + && !oops_in_progress); -/* FIXME: Shim for archs using old arch_send_call_function_ipi API. */ -#ifndef arch_send_call_function_ipi_mask -#define arch_send_call_function_ipi_mask(maskp) \ - arch_send_call_function_ipi(*(maskp)) -#endif + generic_exec_single(cpu, data, wait); +} /** * smp_call_function_many(): Run a function on a set of other CPUs. * @mask: The set of cpus to run on (only runs on online subset). * @func: The function to run. This must be fast and non-blocking. * @info: An arbitrary pointer to pass to the function. - * @wait: If true, wait (atomically) until function has completed on other CPUs. + * @wait: If true, wait (atomically) until function has completed + * on other CPUs. * - * If @wait is true, then returns once @func has returned. Note that @wait - * will be implicitly turned on in case of allocation failures, since - * we fall back to on-stack allocation. + * If @wait is true, then returns once @func has returned. * * You must not call this function with disabled interrupts or from a * hardware interrupt handler or from a bottom half handler. Preemption * must be disabled when calling this function. */ void smp_call_function_many(const struct cpumask *mask, - void (*func)(void *), void *info, - bool wait) + void (*func)(void *), void *info, bool wait) { struct call_function_data *data; unsigned long flags; - int cpu, next_cpu; + int cpu, next_cpu, this_cpu = smp_processor_id(); - /* Can deadlock when called with interrupts disabled */ - WARN_ON(irqs_disabled()); + /* + * Can deadlock when called with interrupts disabled. + * We allow cpu's that are not yet online though, as no one else can + * send smp call function interrupt to this cpu and as such deadlocks + * can't happen. + */ + WARN_ON_ONCE(cpu_online(this_cpu) && irqs_disabled() + && !oops_in_progress); - /* So, what's a CPU they want? Ignoring this one. */ + /* So, what's a CPU they want? Ignoring this one. */ cpu = cpumask_first_and(mask, cpu_online_mask); - if (cpu == smp_processor_id()) + if (cpu == this_cpu) cpu = cpumask_next_and(cpu, mask, cpu_online_mask); + /* No online cpus? We're done. */ if (cpu >= nr_cpu_ids) return; /* Do we have another CPU which isn't us? */ next_cpu = cpumask_next_and(cpu, mask, cpu_online_mask); - if (next_cpu == smp_processor_id()) + if (next_cpu == this_cpu) next_cpu = cpumask_next_and(next_cpu, mask, cpu_online_mask); /* Fastpath: do that cpu by itself. */ @@ -347,43 +440,37 @@ void smp_call_function_many(const struct cpumask *mask, return; } - data = kmalloc(sizeof(*data) + cpumask_size(), GFP_ATOMIC); - if (unlikely(!data)) { - /* Slow path. */ - for_each_online_cpu(cpu) { - if (cpu == smp_processor_id()) - continue; - if (cpumask_test_cpu(cpu, mask)) - smp_call_function_single(cpu, func, info, wait); - } - return; - } + data = &__get_cpu_var(cfd_data); + csd_lock(&data->csd); - spin_lock_init(&data->lock); - data->csd.flags = CSD_FLAG_ALLOC; - if (wait) - data->csd.flags |= CSD_FLAG_WAIT; data->csd.func = func; data->csd.info = info; - cpumask_and(to_cpumask(data->cpumask_bits), mask, cpu_online_mask); - cpumask_clear_cpu(smp_processor_id(), to_cpumask(data->cpumask_bits)); - data->refs = cpumask_weight(to_cpumask(data->cpumask_bits)); + cpumask_and(data->cpumask, mask, cpu_online_mask); + cpumask_clear_cpu(this_cpu, data->cpumask); + atomic_set(&data->refs, cpumask_weight(data->cpumask)); - spin_lock_irqsave(&call_function_lock, flags); - list_add_tail_rcu(&data->csd.list, &call_function_queue); - spin_unlock_irqrestore(&call_function_lock, flags); + spin_lock_irqsave(&call_function.lock, flags); + /* + * Place entry at the _HEAD_ of the list, so that any cpu still + * observing the entry in generic_smp_call_function_interrupt() + * will not miss any other list entries: + */ + list_add_rcu(&data->csd.list, &call_function.queue); + spin_unlock_irqrestore(&call_function.lock, flags); /* * Make the list addition visible before sending the ipi. + * (IPIs must obey or appear to obey normal Linux cache + * coherency rules -- see comment in generic_exec_single). */ smp_mb(); /* Send a message to all CPUs in the map */ - arch_send_call_function_ipi_mask(to_cpumask(data->cpumask_bits)); + arch_send_call_function_ipi_mask(data->cpumask); - /* optionally wait for the CPUs to complete */ + /* Optionally wait for the CPUs to complete */ if (wait) - csd_flag_wait(&data->csd); + csd_lock_wait(&data->csd); } EXPORT_SYMBOL(smp_call_function_many); @@ -391,13 +478,13 @@ EXPORT_SYMBOL(smp_call_function_many); * smp_call_function(): Run a function on all other CPUs. * @func: The function to run. This must be fast and non-blocking. * @info: An arbitrary pointer to pass to the function. - * @wait: If true, wait (atomically) until function has completed on other CPUs. + * @wait: If true, wait (atomically) until function has completed + * on other CPUs. * * Returns 0. * * If @wait is true, then returns once @func has returned; otherwise - * it returns just before the target cpu calls @func. In case of allocation - * failure, @wait will be implicitly turned on. + * it returns just before the target cpu calls @func. * * You must not call this function with disabled interrupts or from a * hardware interrupt handler or from a bottom half handler. @@ -407,26 +494,27 @@ int smp_call_function(void (*func)(void *), void *info, int wait) preempt_disable(); smp_call_function_many(cpu_online_mask, func, info, wait); preempt_enable(); + return 0; } EXPORT_SYMBOL(smp_call_function); void ipi_call_lock(void) { - spin_lock(&call_function_lock); + spin_lock(&call_function.lock); } void ipi_call_unlock(void) { - spin_unlock(&call_function_lock); + spin_unlock(&call_function.lock); } void ipi_call_lock_irq(void) { - spin_lock_irq(&call_function_lock); + spin_lock_irq(&call_function.lock); } void ipi_call_unlock_irq(void) { - spin_unlock_irq(&call_function_lock); + spin_unlock_irq(&call_function.lock); } diff --git a/kernel/softirq.c b/kernel/softirq.c index 57d3f67f6f3..21939d9e830 100644 --- a/kernel/softirq.c +++ b/kernel/softirq.c @@ -21,9 +21,13 @@ #include <linux/freezer.h> #include <linux/kthread.h> #include <linux/rcupdate.h> +#include <linux/ftrace.h> #include <linux/smp.h> #include <linux/tick.h> +#define CREATE_TRACE_POINTS +#include <trace/events/irq.h> + #include <asm/irq.h> /* - No shared variables, all the data are CPU local. @@ -52,13 +56,18 @@ static struct softirq_action softirq_vec[NR_SOFTIRQS] __cacheline_aligned_in_smp static DEFINE_PER_CPU(struct task_struct *, ksoftirqd); +char *softirq_to_name[NR_SOFTIRQS] = { + "HI", "TIMER", "NET_TX", "NET_RX", "BLOCK", "BLOCK_IOPOLL", + "TASKLET", "SCHED", "HRTIMER", "RCU" +}; + /* * we cannot loop indefinitely here to avoid userspace starvation, * but we also don't want to introduce a worst case 1/HZ latency * to the pending events, so lets the scheduler to balance * the softirq load for us. */ -static inline void wakeup_softirqd(void) +void wakeup_softirqd(void) { /* Interrupts are disabled: no need to stop preemption */ struct task_struct *tsk = __get_cpu_var(ksoftirqd); @@ -79,13 +88,23 @@ static void __local_bh_disable(unsigned long ip) WARN_ON_ONCE(in_irq()); raw_local_irq_save(flags); - add_preempt_count(SOFTIRQ_OFFSET); + /* + * The preempt tracer hooks into add_preempt_count and will break + * lockdep because it calls back into lockdep after SOFTIRQ_OFFSET + * is set and before current->softirq_enabled is cleared. + * We must manually increment preempt_count here and manually + * call the trace_preempt_off later. + */ + preempt_count() += SOFTIRQ_OFFSET; /* * Were softirqs turned off above: */ if (softirq_count() == SOFTIRQ_OFFSET) trace_softirqs_off(ip); raw_local_irq_restore(flags); + + if (preempt_count() == SOFTIRQ_OFFSET) + trace_preempt_off(CALLER_ADDR0, get_parent_ip(CALLER_ADDR1)); } #else /* !CONFIG_TRACE_IRQFLAGS */ static inline void __local_bh_disable(unsigned long ip) @@ -180,7 +199,7 @@ asmlinkage void __do_softirq(void) account_system_vtime(current); __local_bh_disable((unsigned long)__builtin_return_address(0)); - trace_softirq_enter(); + lockdep_softirq_enter(); cpu = smp_processor_id(); restart: @@ -194,18 +213,21 @@ restart: do { if (pending & 1) { int prev_count = preempt_count(); + kstat_incr_softirqs_this_cpu(h - softirq_vec); + trace_softirq_entry(h, softirq_vec); h->action(h); - + trace_softirq_exit(h, softirq_vec); if (unlikely(prev_count != preempt_count())) { - printk(KERN_ERR "huh, entered softirq %td %p" + printk(KERN_ERR "huh, entered softirq %td %s %p" "with preempt_count %08x," " exited with %08x?\n", h - softirq_vec, + softirq_to_name[h - softirq_vec], h->action, prev_count, preempt_count()); preempt_count() = prev_count; } - rcu_bh_qsctr_inc(cpu); + rcu_bh_qs(cpu); } h++; pending >>= 1; @@ -220,7 +242,7 @@ restart: if (pending) wakeup_softirqd(); - trace_softirq_exit(); + lockdep_softirq_exit(); account_system_vtime(current); _local_bh_enable(); @@ -280,9 +302,9 @@ void irq_exit(void) if (!in_interrupt() && local_softirq_pending()) invoke_softirq(); + rcu_irq_exit(); #ifdef CONFIG_NO_HZ /* Make sure that timer wheel updates are propagated */ - rcu_irq_exit(); if (idle_cpu(smp_processor_id()) && !in_interrupt() && !need_resched()) tick_nohz_stop_sched_tick(0); #endif @@ -323,7 +345,9 @@ void open_softirq(int nr, void (*action)(struct softirq_action *)) softirq_vec[nr].action = action; } -/* Tasklets */ +/* + * Tasklets + */ struct tasklet_head { struct tasklet_struct *head; @@ -361,6 +385,17 @@ void __tasklet_hi_schedule(struct tasklet_struct *t) EXPORT_SYMBOL(__tasklet_hi_schedule); +void __tasklet_hi_schedule_first(struct tasklet_struct *t) +{ + BUG_ON(!irqs_disabled()); + + t->next = __get_cpu_var(tasklet_hi_vec).head; + __get_cpu_var(tasklet_hi_vec).head = t; + __raise_softirq_irqoff(HI_SOFTIRQ); +} + +EXPORT_SYMBOL(__tasklet_hi_schedule_first); + static void tasklet_action(struct softirq_action *a) { struct tasklet_struct *list; @@ -450,9 +485,9 @@ void tasklet_kill(struct tasklet_struct *t) printk("Attempt to kill tasklet from interrupt\n"); while (test_and_set_bit(TASKLET_STATE_SCHED, &t->state)) { - do + do { yield(); - while (test_bit(TASKLET_STATE_SCHED, &t->state)); + } while (test_bit(TASKLET_STATE_SCHED, &t->state)); } tasklet_unlock_wait(t); clear_bit(TASKLET_STATE_SCHED, &t->state); @@ -460,6 +495,66 @@ void tasklet_kill(struct tasklet_struct *t) EXPORT_SYMBOL(tasklet_kill); +/* + * tasklet_hrtimer + */ + +/* + * The trampoline is called when the hrtimer expires. If this is + * called from the hrtimer interrupt then we schedule the tasklet as + * the timer callback function expects to run in softirq context. If + * it's called in softirq context anyway (i.e. high resolution timers + * disabled) then the hrtimer callback is called right away. + */ +static enum hrtimer_restart __hrtimer_tasklet_trampoline(struct hrtimer *timer) +{ + struct tasklet_hrtimer *ttimer = + container_of(timer, struct tasklet_hrtimer, timer); + + if (hrtimer_is_hres_active(timer)) { + tasklet_hi_schedule(&ttimer->tasklet); + return HRTIMER_NORESTART; + } + return ttimer->function(timer); +} + +/* + * Helper function which calls the hrtimer callback from + * tasklet/softirq context + */ +static void __tasklet_hrtimer_trampoline(unsigned long data) +{ + struct tasklet_hrtimer *ttimer = (void *)data; + enum hrtimer_restart restart; + + restart = ttimer->function(&ttimer->timer); + if (restart != HRTIMER_NORESTART) + hrtimer_restart(&ttimer->timer); +} + +/** + * tasklet_hrtimer_init - Init a tasklet/hrtimer combo for softirq callbacks + * @ttimer: tasklet_hrtimer which is initialized + * @function: hrtimer callback funtion which gets called from softirq context + * @which_clock: clock id (CLOCK_MONOTONIC/CLOCK_REALTIME) + * @mode: hrtimer mode (HRTIMER_MODE_ABS/HRTIMER_MODE_REL) + */ +void tasklet_hrtimer_init(struct tasklet_hrtimer *ttimer, + enum hrtimer_restart (*function)(struct hrtimer *), + clockid_t which_clock, enum hrtimer_mode mode) +{ + hrtimer_init(&ttimer->timer, which_clock, mode); + ttimer->timer.function = __hrtimer_tasklet_trampoline; + tasklet_init(&ttimer->tasklet, __tasklet_hrtimer_trampoline, + (unsigned long)ttimer); + ttimer->function = function; +} +EXPORT_SYMBOL_GPL(tasklet_hrtimer_init); + +/* + * Remote softirq bits + */ + DEFINE_PER_CPU(struct list_head [NR_SOFTIRQS], softirq_work_list); EXPORT_PER_CPU_SYMBOL(softirq_work_list); @@ -496,7 +591,7 @@ static int __try_remote_softirq(struct call_single_data *cp, int cpu, int softir cp->flags = 0; cp->priv = softirq; - __smp_call_function_single(cpu, cp); + __smp_call_function_single(cpu, cp, 0); return 0; } return 1; @@ -626,7 +721,7 @@ static int ksoftirqd(void * __bind_cpu) preempt_enable_no_resched(); cond_resched(); preempt_disable(); - rcu_qsctr_inc((long)__bind_cpu); + rcu_sched_qs((long)__bind_cpu); } preempt_enable(); set_current_state(TASK_INTERRUPTIBLE); @@ -806,7 +901,7 @@ int __init __weak arch_early_irq_init(void) return 0; } -int __weak arch_init_chip_data(struct irq_desc *desc, int cpu) +int __weak arch_init_chip_data(struct irq_desc *desc, int node) { return 0; } diff --git a/kernel/softlockup.c b/kernel/softlockup.c index 85d5a245510..81324d12eb3 100644 --- a/kernel/softlockup.c +++ b/kernel/softlockup.c @@ -90,11 +90,11 @@ void touch_all_softlockup_watchdogs(void) EXPORT_SYMBOL(touch_all_softlockup_watchdogs); int proc_dosoftlockup_thresh(struct ctl_table *table, int write, - struct file *filp, void __user *buffer, + void __user *buffer, size_t *lenp, loff_t *ppos) { touch_all_softlockup_watchdogs(); - return proc_dointvec_minmax(table, write, filp, buffer, lenp, ppos); + return proc_dointvec_minmax(table, write, buffer, lenp, ppos); } /* @@ -166,97 +166,11 @@ void softlockup_tick(void) } /* - * Have a reasonable limit on the number of tasks checked: - */ -unsigned long __read_mostly sysctl_hung_task_check_count = 1024; - -/* - * Zero means infinite timeout - no checking done: - */ -unsigned long __read_mostly sysctl_hung_task_timeout_secs = 480; - -unsigned long __read_mostly sysctl_hung_task_warnings = 10; - -/* - * Only do the hung-tasks check on one CPU: - */ -static int check_cpu __read_mostly = -1; - -static void check_hung_task(struct task_struct *t, unsigned long now) -{ - unsigned long switch_count = t->nvcsw + t->nivcsw; - - if (t->flags & PF_FROZEN) - return; - - if (switch_count != t->last_switch_count || !t->last_switch_timestamp) { - t->last_switch_count = switch_count; - t->last_switch_timestamp = now; - return; - } - if ((long)(now - t->last_switch_timestamp) < - sysctl_hung_task_timeout_secs) - return; - if (!sysctl_hung_task_warnings) - return; - sysctl_hung_task_warnings--; - - /* - * Ok, the task did not get scheduled for more than 2 minutes, - * complain: - */ - printk(KERN_ERR "INFO: task %s:%d blocked for more than " - "%ld seconds.\n", t->comm, t->pid, - sysctl_hung_task_timeout_secs); - printk(KERN_ERR "\"echo 0 > /proc/sys/kernel/hung_task_timeout_secs\"" - " disables this message.\n"); - sched_show_task(t); - __debug_show_held_locks(t); - - t->last_switch_timestamp = now; - touch_nmi_watchdog(); - - if (softlockup_panic) - panic("softlockup: blocked tasks"); -} - -/* - * Check whether a TASK_UNINTERRUPTIBLE does not get woken up for - * a really long time (120 seconds). If that happens, print out - * a warning. - */ -static void check_hung_uninterruptible_tasks(int this_cpu) -{ - int max_count = sysctl_hung_task_check_count; - unsigned long now = get_timestamp(this_cpu); - struct task_struct *g, *t; - - /* - * If the system crashed already then all bets are off, - * do not report extra hung tasks: - */ - if (test_taint(TAINT_DIE) || did_panic) - return; - - read_lock(&tasklist_lock); - do_each_thread(g, t) { - if (!--max_count) - goto unlock; - /* use "==" to skip the TASK_KILLABLE tasks waiting on NFS */ - if (t->state == TASK_UNINTERRUPTIBLE) - check_hung_task(t, now); - } while_each_thread(g, t); - unlock: - read_unlock(&tasklist_lock); -} - -/* * The watchdog thread - runs every second and touches the timestamp. */ static int watchdog(void *__bind_cpu) { struct sched_param param = { .sched_priority = MAX_RT_PRIO-1 }; - int this_cpu = (long)__bind_cpu; sched_setscheduler(current, SCHED_FIFO, ¶m); @@ -276,11 +190,6 @@ static int watchdog(void *__bind_cpu) if (kthread_should_stop()) break; - if (this_cpu == check_cpu) { - if (sysctl_hung_task_timeout_secs) - check_hung_uninterruptible_tasks(this_cpu); - } - set_current_state(TASK_INTERRUPTIBLE); } __set_current_state(TASK_RUNNING); @@ -312,18 +221,9 @@ cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu) break; case CPU_ONLINE: case CPU_ONLINE_FROZEN: - check_cpu = cpumask_any(cpu_online_mask); wake_up_process(per_cpu(watchdog_task, hotcpu)); break; #ifdef CONFIG_HOTPLUG_CPU - case CPU_DOWN_PREPARE: - case CPU_DOWN_PREPARE_FROZEN: - if (hotcpu == check_cpu) { - /* Pick any other online cpu. */ - check_cpu = cpumask_any_but(cpu_online_mask, hotcpu); - } - break; - case CPU_UP_CANCELED: case CPU_UP_CANCELED_FROZEN: if (!per_cpu(watchdog_task, hotcpu)) diff --git a/kernel/spinlock.c b/kernel/spinlock.c index 29ab20749dd..41e042219ff 100644 --- a/kernel/spinlock.c +++ b/kernel/spinlock.c @@ -21,189 +21,28 @@ #include <linux/debug_locks.h> #include <linux/module.h> -int __lockfunc _spin_trylock(spinlock_t *lock) -{ - preempt_disable(); - if (_raw_spin_trylock(lock)) { - spin_acquire(&lock->dep_map, 0, 1, _RET_IP_); - return 1; - } - - preempt_enable(); - return 0; -} -EXPORT_SYMBOL(_spin_trylock); - -int __lockfunc _read_trylock(rwlock_t *lock) -{ - preempt_disable(); - if (_raw_read_trylock(lock)) { - rwlock_acquire_read(&lock->dep_map, 0, 1, _RET_IP_); - return 1; - } - - preempt_enable(); - return 0; -} -EXPORT_SYMBOL(_read_trylock); - -int __lockfunc _write_trylock(rwlock_t *lock) -{ - preempt_disable(); - if (_raw_write_trylock(lock)) { - rwlock_acquire(&lock->dep_map, 0, 1, _RET_IP_); - return 1; - } - - preempt_enable(); - return 0; -} -EXPORT_SYMBOL(_write_trylock); - /* * If lockdep is enabled then we use the non-preemption spin-ops * even on CONFIG_PREEMPT, because lockdep assumes that interrupts are * not re-enabled during lock-acquire (which the preempt-spin-ops do): */ #if !defined(CONFIG_GENERIC_LOCKBREAK) || defined(CONFIG_DEBUG_LOCK_ALLOC) - -void __lockfunc _read_lock(rwlock_t *lock) -{ - preempt_disable(); - rwlock_acquire_read(&lock->dep_map, 0, 0, _RET_IP_); - LOCK_CONTENDED(lock, _raw_read_trylock, _raw_read_lock); -} -EXPORT_SYMBOL(_read_lock); - -unsigned long __lockfunc _spin_lock_irqsave(spinlock_t *lock) -{ - unsigned long flags; - - local_irq_save(flags); - preempt_disable(); - spin_acquire(&lock->dep_map, 0, 0, _RET_IP_); - /* - * On lockdep we dont want the hand-coded irq-enable of - * _raw_spin_lock_flags() code, because lockdep assumes - * that interrupts are not re-enabled during lock-acquire: - */ -#ifdef CONFIG_LOCKDEP - LOCK_CONTENDED(lock, _raw_spin_trylock, _raw_spin_lock); +/* + * The __lock_function inlines are taken from + * include/linux/spinlock_api_smp.h + */ #else - _raw_spin_lock_flags(lock, &flags); -#endif - return flags; -} -EXPORT_SYMBOL(_spin_lock_irqsave); - -void __lockfunc _spin_lock_irq(spinlock_t *lock) -{ - local_irq_disable(); - preempt_disable(); - spin_acquire(&lock->dep_map, 0, 0, _RET_IP_); - LOCK_CONTENDED(lock, _raw_spin_trylock, _raw_spin_lock); -} -EXPORT_SYMBOL(_spin_lock_irq); - -void __lockfunc _spin_lock_bh(spinlock_t *lock) -{ - local_bh_disable(); - preempt_disable(); - spin_acquire(&lock->dep_map, 0, 0, _RET_IP_); - LOCK_CONTENDED(lock, _raw_spin_trylock, _raw_spin_lock); -} -EXPORT_SYMBOL(_spin_lock_bh); - -unsigned long __lockfunc _read_lock_irqsave(rwlock_t *lock) -{ - unsigned long flags; - - local_irq_save(flags); - preempt_disable(); - rwlock_acquire_read(&lock->dep_map, 0, 0, _RET_IP_); - LOCK_CONTENDED(lock, _raw_read_trylock, _raw_read_lock); - return flags; -} -EXPORT_SYMBOL(_read_lock_irqsave); - -void __lockfunc _read_lock_irq(rwlock_t *lock) -{ - local_irq_disable(); - preempt_disable(); - rwlock_acquire_read(&lock->dep_map, 0, 0, _RET_IP_); - LOCK_CONTENDED(lock, _raw_read_trylock, _raw_read_lock); -} -EXPORT_SYMBOL(_read_lock_irq); - -void __lockfunc _read_lock_bh(rwlock_t *lock) -{ - local_bh_disable(); - preempt_disable(); - rwlock_acquire_read(&lock->dep_map, 0, 0, _RET_IP_); - LOCK_CONTENDED(lock, _raw_read_trylock, _raw_read_lock); -} -EXPORT_SYMBOL(_read_lock_bh); - -unsigned long __lockfunc _write_lock_irqsave(rwlock_t *lock) -{ - unsigned long flags; - - local_irq_save(flags); - preempt_disable(); - rwlock_acquire(&lock->dep_map, 0, 0, _RET_IP_); - LOCK_CONTENDED(lock, _raw_write_trylock, _raw_write_lock); - return flags; -} -EXPORT_SYMBOL(_write_lock_irqsave); - -void __lockfunc _write_lock_irq(rwlock_t *lock) -{ - local_irq_disable(); - preempt_disable(); - rwlock_acquire(&lock->dep_map, 0, 0, _RET_IP_); - LOCK_CONTENDED(lock, _raw_write_trylock, _raw_write_lock); -} -EXPORT_SYMBOL(_write_lock_irq); - -void __lockfunc _write_lock_bh(rwlock_t *lock) -{ - local_bh_disable(); - preempt_disable(); - rwlock_acquire(&lock->dep_map, 0, 0, _RET_IP_); - LOCK_CONTENDED(lock, _raw_write_trylock, _raw_write_lock); -} -EXPORT_SYMBOL(_write_lock_bh); - -void __lockfunc _spin_lock(spinlock_t *lock) -{ - preempt_disable(); - spin_acquire(&lock->dep_map, 0, 0, _RET_IP_); - LOCK_CONTENDED(lock, _raw_spin_trylock, _raw_spin_lock); -} - -EXPORT_SYMBOL(_spin_lock); - -void __lockfunc _write_lock(rwlock_t *lock) -{ - preempt_disable(); - rwlock_acquire(&lock->dep_map, 0, 0, _RET_IP_); - LOCK_CONTENDED(lock, _raw_write_trylock, _raw_write_lock); -} - -EXPORT_SYMBOL(_write_lock); - -#else /* CONFIG_PREEMPT: */ - /* + * We build the __lock_function inlines here. They are too large for + * inlining all over the place, but here is only one user per function + * which embedds them into the calling _lock_function below. + * * This could be a long-held lock. We both prepare to spin for a long * time (making _this_ CPU preemptable if possible), and we also signal * towards that other CPU that it should break the lock ASAP. - * - * (We do this in a function because inlining it would be excessive.) */ - #define BUILD_LOCK_OPS(op, locktype) \ -void __lockfunc _##op##_lock(locktype##_t *lock) \ +void __lockfunc __##op##_lock(locktype##_t *lock) \ { \ for (;;) { \ preempt_disable(); \ @@ -219,9 +58,7 @@ void __lockfunc _##op##_lock(locktype##_t *lock) \ (lock)->break_lock = 0; \ } \ \ -EXPORT_SYMBOL(_##op##_lock); \ - \ -unsigned long __lockfunc _##op##_lock_irqsave(locktype##_t *lock) \ +unsigned long __lockfunc __##op##_lock_irqsave(locktype##_t *lock) \ { \ unsigned long flags; \ \ @@ -242,16 +79,12 @@ unsigned long __lockfunc _##op##_lock_irqsave(locktype##_t *lock) \ return flags; \ } \ \ -EXPORT_SYMBOL(_##op##_lock_irqsave); \ - \ -void __lockfunc _##op##_lock_irq(locktype##_t *lock) \ +void __lockfunc __##op##_lock_irq(locktype##_t *lock) \ { \ _##op##_lock_irqsave(lock); \ } \ \ -EXPORT_SYMBOL(_##op##_lock_irq); \ - \ -void __lockfunc _##op##_lock_bh(locktype##_t *lock) \ +void __lockfunc __##op##_lock_bh(locktype##_t *lock) \ { \ unsigned long flags; \ \ @@ -264,23 +97,21 @@ void __lockfunc _##op##_lock_bh(locktype##_t *lock) \ local_bh_disable(); \ local_irq_restore(flags); \ } \ - \ -EXPORT_SYMBOL(_##op##_lock_bh) /* * Build preemption-friendly versions of the following * lock-spinning functions: * - * _[spin|read|write]_lock() - * _[spin|read|write]_lock_irq() - * _[spin|read|write]_lock_irqsave() - * _[spin|read|write]_lock_bh() + * __[spin|read|write]_lock() + * __[spin|read|write]_lock_irq() + * __[spin|read|write]_lock_irqsave() + * __[spin|read|write]_lock_bh() */ BUILD_LOCK_OPS(spin, spinlock); BUILD_LOCK_OPS(read, rwlock); BUILD_LOCK_OPS(write, rwlock); -#endif /* CONFIG_PREEMPT */ +#endif #ifdef CONFIG_DEBUG_LOCK_ALLOC @@ -292,23 +123,16 @@ void __lockfunc _spin_lock_nested(spinlock_t *lock, int subclass) } EXPORT_SYMBOL(_spin_lock_nested); -unsigned long __lockfunc _spin_lock_irqsave_nested(spinlock_t *lock, int subclass) +unsigned long __lockfunc _spin_lock_irqsave_nested(spinlock_t *lock, + int subclass) { unsigned long flags; local_irq_save(flags); preempt_disable(); spin_acquire(&lock->dep_map, subclass, 0, _RET_IP_); - /* - * On lockdep we dont want the hand-coded irq-enable of - * _raw_spin_lock_flags() code, because lockdep assumes - * that interrupts are not re-enabled during lock-acquire: - */ -#ifdef CONFIG_LOCKDEP - LOCK_CONTENDED(lock, _raw_spin_trylock, _raw_spin_lock); -#else - _raw_spin_lock_flags(lock, &flags); -#endif + LOCK_CONTENDED_FLAGS(lock, _raw_spin_trylock, _raw_spin_lock, + _raw_spin_lock_flags, &flags); return flags; } EXPORT_SYMBOL(_spin_lock_irqsave_nested); @@ -324,125 +148,229 @@ EXPORT_SYMBOL(_spin_lock_nest_lock); #endif +#ifndef CONFIG_INLINE_SPIN_TRYLOCK +int __lockfunc _spin_trylock(spinlock_t *lock) +{ + return __spin_trylock(lock); +} +EXPORT_SYMBOL(_spin_trylock); +#endif + +#ifndef CONFIG_INLINE_READ_TRYLOCK +int __lockfunc _read_trylock(rwlock_t *lock) +{ + return __read_trylock(lock); +} +EXPORT_SYMBOL(_read_trylock); +#endif + +#ifndef CONFIG_INLINE_WRITE_TRYLOCK +int __lockfunc _write_trylock(rwlock_t *lock) +{ + return __write_trylock(lock); +} +EXPORT_SYMBOL(_write_trylock); +#endif + +#ifndef CONFIG_INLINE_READ_LOCK +void __lockfunc _read_lock(rwlock_t *lock) +{ + __read_lock(lock); +} +EXPORT_SYMBOL(_read_lock); +#endif + +#ifndef CONFIG_INLINE_SPIN_LOCK_IRQSAVE +unsigned long __lockfunc _spin_lock_irqsave(spinlock_t *lock) +{ + return __spin_lock_irqsave(lock); +} +EXPORT_SYMBOL(_spin_lock_irqsave); +#endif + +#ifndef CONFIG_INLINE_SPIN_LOCK_IRQ +void __lockfunc _spin_lock_irq(spinlock_t *lock) +{ + __spin_lock_irq(lock); +} +EXPORT_SYMBOL(_spin_lock_irq); +#endif + +#ifndef CONFIG_INLINE_SPIN_LOCK_BH +void __lockfunc _spin_lock_bh(spinlock_t *lock) +{ + __spin_lock_bh(lock); +} +EXPORT_SYMBOL(_spin_lock_bh); +#endif + +#ifndef CONFIG_INLINE_READ_LOCK_IRQSAVE +unsigned long __lockfunc _read_lock_irqsave(rwlock_t *lock) +{ + return __read_lock_irqsave(lock); +} +EXPORT_SYMBOL(_read_lock_irqsave); +#endif + +#ifndef CONFIG_INLINE_READ_LOCK_IRQ +void __lockfunc _read_lock_irq(rwlock_t *lock) +{ + __read_lock_irq(lock); +} +EXPORT_SYMBOL(_read_lock_irq); +#endif + +#ifndef CONFIG_INLINE_READ_LOCK_BH +void __lockfunc _read_lock_bh(rwlock_t *lock) +{ + __read_lock_bh(lock); +} +EXPORT_SYMBOL(_read_lock_bh); +#endif + +#ifndef CONFIG_INLINE_WRITE_LOCK_IRQSAVE +unsigned long __lockfunc _write_lock_irqsave(rwlock_t *lock) +{ + return __write_lock_irqsave(lock); +} +EXPORT_SYMBOL(_write_lock_irqsave); +#endif + +#ifndef CONFIG_INLINE_WRITE_LOCK_IRQ +void __lockfunc _write_lock_irq(rwlock_t *lock) +{ + __write_lock_irq(lock); +} +EXPORT_SYMBOL(_write_lock_irq); +#endif + +#ifndef CONFIG_INLINE_WRITE_LOCK_BH +void __lockfunc _write_lock_bh(rwlock_t *lock) +{ + __write_lock_bh(lock); +} +EXPORT_SYMBOL(_write_lock_bh); +#endif + +#ifndef CONFIG_INLINE_SPIN_LOCK +void __lockfunc _spin_lock(spinlock_t *lock) +{ + __spin_lock(lock); +} +EXPORT_SYMBOL(_spin_lock); +#endif + +#ifndef CONFIG_INLINE_WRITE_LOCK +void __lockfunc _write_lock(rwlock_t *lock) +{ + __write_lock(lock); +} +EXPORT_SYMBOL(_write_lock); +#endif + +#ifndef CONFIG_INLINE_SPIN_UNLOCK void __lockfunc _spin_unlock(spinlock_t *lock) { - spin_release(&lock->dep_map, 1, _RET_IP_); - _raw_spin_unlock(lock); - preempt_enable(); + __spin_unlock(lock); } EXPORT_SYMBOL(_spin_unlock); +#endif +#ifndef CONFIG_INLINE_WRITE_UNLOCK void __lockfunc _write_unlock(rwlock_t *lock) { - rwlock_release(&lock->dep_map, 1, _RET_IP_); - _raw_write_unlock(lock); - preempt_enable(); + __write_unlock(lock); } EXPORT_SYMBOL(_write_unlock); +#endif +#ifndef CONFIG_INLINE_READ_UNLOCK void __lockfunc _read_unlock(rwlock_t *lock) { - rwlock_release(&lock->dep_map, 1, _RET_IP_); - _raw_read_unlock(lock); - preempt_enable(); + __read_unlock(lock); } EXPORT_SYMBOL(_read_unlock); +#endif +#ifndef CONFIG_INLINE_SPIN_UNLOCK_IRQRESTORE void __lockfunc _spin_unlock_irqrestore(spinlock_t *lock, unsigned long flags) { - spin_release(&lock->dep_map, 1, _RET_IP_); - _raw_spin_unlock(lock); - local_irq_restore(flags); - preempt_enable(); + __spin_unlock_irqrestore(lock, flags); } EXPORT_SYMBOL(_spin_unlock_irqrestore); +#endif +#ifndef CONFIG_INLINE_SPIN_UNLOCK_IRQ void __lockfunc _spin_unlock_irq(spinlock_t *lock) { - spin_release(&lock->dep_map, 1, _RET_IP_); - _raw_spin_unlock(lock); - local_irq_enable(); - preempt_enable(); + __spin_unlock_irq(lock); } EXPORT_SYMBOL(_spin_unlock_irq); +#endif +#ifndef CONFIG_INLINE_SPIN_UNLOCK_BH void __lockfunc _spin_unlock_bh(spinlock_t *lock) { - spin_release(&lock->dep_map, 1, _RET_IP_); - _raw_spin_unlock(lock); - preempt_enable_no_resched(); - local_bh_enable_ip((unsigned long)__builtin_return_address(0)); + __spin_unlock_bh(lock); } EXPORT_SYMBOL(_spin_unlock_bh); +#endif +#ifndef CONFIG_INLINE_READ_UNLOCK_IRQRESTORE void __lockfunc _read_unlock_irqrestore(rwlock_t *lock, unsigned long flags) { - rwlock_release(&lock->dep_map, 1, _RET_IP_); - _raw_read_unlock(lock); - local_irq_restore(flags); - preempt_enable(); + __read_unlock_irqrestore(lock, flags); } EXPORT_SYMBOL(_read_unlock_irqrestore); +#endif +#ifndef CONFIG_INLINE_READ_UNLOCK_IRQ void __lockfunc _read_unlock_irq(rwlock_t *lock) { - rwlock_release(&lock->dep_map, 1, _RET_IP_); - _raw_read_unlock(lock); - local_irq_enable(); - preempt_enable(); + __read_unlock_irq(lock); } EXPORT_SYMBOL(_read_unlock_irq); +#endif +#ifndef CONFIG_INLINE_READ_UNLOCK_BH void __lockfunc _read_unlock_bh(rwlock_t *lock) { - rwlock_release(&lock->dep_map, 1, _RET_IP_); - _raw_read_unlock(lock); - preempt_enable_no_resched(); - local_bh_enable_ip((unsigned long)__builtin_return_address(0)); + __read_unlock_bh(lock); } EXPORT_SYMBOL(_read_unlock_bh); +#endif +#ifndef CONFIG_INLINE_WRITE_UNLOCK_IRQRESTORE void __lockfunc _write_unlock_irqrestore(rwlock_t *lock, unsigned long flags) { - rwlock_release(&lock->dep_map, 1, _RET_IP_); - _raw_write_unlock(lock); - local_irq_restore(flags); - preempt_enable(); + __write_unlock_irqrestore(lock, flags); } EXPORT_SYMBOL(_write_unlock_irqrestore); +#endif +#ifndef CONFIG_INLINE_WRITE_UNLOCK_IRQ void __lockfunc _write_unlock_irq(rwlock_t *lock) { - rwlock_release(&lock->dep_map, 1, _RET_IP_); - _raw_write_unlock(lock); - local_irq_enable(); - preempt_enable(); + __write_unlock_irq(lock); } EXPORT_SYMBOL(_write_unlock_irq); +#endif +#ifndef CONFIG_INLINE_WRITE_UNLOCK_BH void __lockfunc _write_unlock_bh(rwlock_t *lock) { - rwlock_release(&lock->dep_map, 1, _RET_IP_); - _raw_write_unlock(lock); - preempt_enable_no_resched(); - local_bh_enable_ip((unsigned long)__builtin_return_address(0)); + __write_unlock_bh(lock); } EXPORT_SYMBOL(_write_unlock_bh); +#endif +#ifndef CONFIG_INLINE_SPIN_TRYLOCK_BH int __lockfunc _spin_trylock_bh(spinlock_t *lock) { - local_bh_disable(); - preempt_disable(); - if (_raw_spin_trylock(lock)) { - spin_acquire(&lock->dep_map, 0, 1, _RET_IP_); - return 1; - } - - preempt_enable_no_resched(); - local_bh_enable_ip((unsigned long)__builtin_return_address(0)); - return 0; + return __spin_trylock_bh(lock); } EXPORT_SYMBOL(_spin_trylock_bh); +#endif notrace int in_lock_functions(unsigned long addr) { diff --git a/kernel/srcu.c b/kernel/srcu.c index b0aeeaf22ce..818d7d9aa03 100644 --- a/kernel/srcu.c +++ b/kernel/srcu.c @@ -49,6 +49,7 @@ int init_srcu_struct(struct srcu_struct *sp) sp->per_cpu_ref = alloc_percpu(struct srcu_struct_array); return (sp->per_cpu_ref ? 0 : -ENOMEM); } +EXPORT_SYMBOL_GPL(init_srcu_struct); /* * srcu_readers_active_idx -- returns approximate number of readers @@ -97,6 +98,7 @@ void cleanup_srcu_struct(struct srcu_struct *sp) free_percpu(sp->per_cpu_ref); sp->per_cpu_ref = NULL; } +EXPORT_SYMBOL_GPL(cleanup_srcu_struct); /** * srcu_read_lock - register a new reader for an SRCU-protected structure. @@ -118,6 +120,7 @@ int srcu_read_lock(struct srcu_struct *sp) preempt_enable(); return idx; } +EXPORT_SYMBOL_GPL(srcu_read_lock); /** * srcu_read_unlock - unregister a old reader from an SRCU-protected structure. @@ -136,22 +139,12 @@ void srcu_read_unlock(struct srcu_struct *sp, int idx) per_cpu_ptr(sp->per_cpu_ref, smp_processor_id())->c[idx]--; preempt_enable(); } +EXPORT_SYMBOL_GPL(srcu_read_unlock); -/** - * synchronize_srcu - wait for prior SRCU read-side critical-section completion - * @sp: srcu_struct with which to synchronize. - * - * Flip the completed counter, and wait for the old count to drain to zero. - * As with classic RCU, the updater must use some separate means of - * synchronizing concurrent updates. Can block; must be called from - * process context. - * - * Note that it is illegal to call synchornize_srcu() from the corresponding - * SRCU read-side critical section; doing so will result in deadlock. - * However, it is perfectly legal to call synchronize_srcu() on one - * srcu_struct from some other srcu_struct's read-side critical section. +/* + * Helper function for synchronize_srcu() and synchronize_srcu_expedited(). */ -void synchronize_srcu(struct srcu_struct *sp) +void __synchronize_srcu(struct srcu_struct *sp, void (*sync_func)(void)) { int idx; @@ -173,7 +166,7 @@ void synchronize_srcu(struct srcu_struct *sp) return; } - synchronize_sched(); /* Force memory barrier on all CPUs. */ + sync_func(); /* Force memory barrier on all CPUs. */ /* * The preceding synchronize_sched() ensures that any CPU that @@ -190,7 +183,7 @@ void synchronize_srcu(struct srcu_struct *sp) idx = sp->completed & 0x1; sp->completed++; - synchronize_sched(); /* Force memory barrier on all CPUs. */ + sync_func(); /* Force memory barrier on all CPUs. */ /* * At this point, because of the preceding synchronize_sched(), @@ -203,7 +196,7 @@ void synchronize_srcu(struct srcu_struct *sp) while (srcu_readers_active_idx(sp, idx)) schedule_timeout_interruptible(1); - synchronize_sched(); /* Force memory barrier on all CPUs. */ + sync_func(); /* Force memory barrier on all CPUs. */ /* * The preceding synchronize_sched() forces all srcu_read_unlock() @@ -237,6 +230,47 @@ void synchronize_srcu(struct srcu_struct *sp) } /** + * synchronize_srcu - wait for prior SRCU read-side critical-section completion + * @sp: srcu_struct with which to synchronize. + * + * Flip the completed counter, and wait for the old count to drain to zero. + * As with classic RCU, the updater must use some separate means of + * synchronizing concurrent updates. Can block; must be called from + * process context. + * + * Note that it is illegal to call synchronize_srcu() from the corresponding + * SRCU read-side critical section; doing so will result in deadlock. + * However, it is perfectly legal to call synchronize_srcu() on one + * srcu_struct from some other srcu_struct's read-side critical section. + */ +void synchronize_srcu(struct srcu_struct *sp) +{ + __synchronize_srcu(sp, synchronize_sched); +} +EXPORT_SYMBOL_GPL(synchronize_srcu); + +/** + * synchronize_srcu_expedited - like synchronize_srcu, but less patient + * @sp: srcu_struct with which to synchronize. + * + * Flip the completed counter, and wait for the old count to drain to zero. + * As with classic RCU, the updater must use some separate means of + * synchronizing concurrent updates. Can block; must be called from + * process context. + * + * Note that it is illegal to call synchronize_srcu_expedited() + * from the corresponding SRCU read-side critical section; doing so + * will result in deadlock. However, it is perfectly legal to call + * synchronize_srcu_expedited() on one srcu_struct from some other + * srcu_struct's read-side critical section. + */ +void synchronize_srcu_expedited(struct srcu_struct *sp) +{ + __synchronize_srcu(sp, synchronize_sched_expedited); +} +EXPORT_SYMBOL_GPL(synchronize_srcu_expedited); + +/** * srcu_batches_completed - return batches completed. * @sp: srcu_struct on which to report batch completion. * @@ -248,10 +282,4 @@ long srcu_batches_completed(struct srcu_struct *sp) { return sp->completed; } - -EXPORT_SYMBOL_GPL(init_srcu_struct); -EXPORT_SYMBOL_GPL(cleanup_srcu_struct); -EXPORT_SYMBOL_GPL(srcu_read_lock); -EXPORT_SYMBOL_GPL(srcu_read_unlock); -EXPORT_SYMBOL_GPL(synchronize_srcu); EXPORT_SYMBOL_GPL(srcu_batches_completed); diff --git a/kernel/sys.c b/kernel/sys.c index 37f458e6882..9968c5fb55b 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -14,6 +14,7 @@ #include <linux/prctl.h> #include <linux/highuid.h> #include <linux/fs.h> +#include <linux/perf_event.h> #include <linux/resource.h> #include <linux/kernel.h> #include <linux/kexec.h> @@ -34,6 +35,7 @@ #include <linux/seccomp.h> #include <linux/cpu.h> #include <linux/ptrace.h> +#include <linux/fs_struct.h> #include <linux/compat.h> #include <linux/syscalls.h> @@ -359,6 +361,7 @@ SYSCALL_DEFINE4(reboot, int, magic1, int, magic2, unsigned int, cmd, void __user *, arg) { char buffer[256]; + int ret = 0; /* We only trust the superuser with rebooting the system. */ if (!capable(CAP_SYS_BOOT)) @@ -396,7 +399,7 @@ SYSCALL_DEFINE4(reboot, int, magic1, int, magic2, unsigned int, cmd, kernel_halt(); unlock_kernel(); do_exit(0); - break; + panic("cannot halt"); case LINUX_REBOOT_CMD_POWER_OFF: kernel_power_off(); @@ -416,29 +419,22 @@ SYSCALL_DEFINE4(reboot, int, magic1, int, magic2, unsigned int, cmd, #ifdef CONFIG_KEXEC case LINUX_REBOOT_CMD_KEXEC: - { - int ret; - ret = kernel_kexec(); - unlock_kernel(); - return ret; - } + ret = kernel_kexec(); + break; #endif #ifdef CONFIG_HIBERNATION case LINUX_REBOOT_CMD_SW_SUSPEND: - { - int ret = hibernate(); - unlock_kernel(); - return ret; - } + ret = hibernate(); + break; #endif default: - unlock_kernel(); - return -EINVAL; + ret = -EINVAL; + break; } unlock_kernel(); - return 0; + return ret; } static void deferred_cad(struct work_struct *dummy) @@ -915,16 +911,15 @@ change_okay: void do_sys_times(struct tms *tms) { - struct task_cputime cputime; - cputime_t cutime, cstime; + cputime_t tgutime, tgstime, cutime, cstime; - thread_group_cputime(current, &cputime); spin_lock_irq(¤t->sighand->siglock); + thread_group_times(current, &tgutime, &tgstime); cutime = current->signal->cutime; cstime = current->signal->cstime; spin_unlock_irq(¤t->sighand->siglock); - tms->tms_utime = cputime_to_clock_t(cputime.utime); - tms->tms_stime = cputime_to_clock_t(cputime.stime); + tms->tms_utime = cputime_to_clock_t(tgutime); + tms->tms_stime = cputime_to_clock_t(tgstime); tms->tms_cutime = cputime_to_clock_t(cutime); tms->tms_cstime = cputime_to_clock_t(cstime); } @@ -1013,10 +1008,8 @@ SYSCALL_DEFINE2(setpgid, pid_t, pid, pid_t, pgid) if (err) goto out; - if (task_pgrp(p) != pgrp) { + if (task_pgrp(p) != pgrp) change_pid(p, PIDTYPE_PGID, pgrp); - set_task_pgrp(p, pid_nr(pgrp)); - } err = 0; out: @@ -1116,292 +1109,11 @@ SYSCALL_DEFINE0(setsid) err = session; out: write_unlock_irq(&tasklist_lock); + if (err > 0) + proc_sid_connector(group_leader); return err; } -/* - * Supplementary group IDs - */ - -/* init to 2 - one for init_task, one to ensure it is never freed */ -struct group_info init_groups = { .usage = ATOMIC_INIT(2) }; - -struct group_info *groups_alloc(int gidsetsize) -{ - struct group_info *group_info; - int nblocks; - int i; - - nblocks = (gidsetsize + NGROUPS_PER_BLOCK - 1) / NGROUPS_PER_BLOCK; - /* Make sure we always allocate at least one indirect block pointer */ - nblocks = nblocks ? : 1; - group_info = kmalloc(sizeof(*group_info) + nblocks*sizeof(gid_t *), GFP_USER); - if (!group_info) - return NULL; - group_info->ngroups = gidsetsize; - group_info->nblocks = nblocks; - atomic_set(&group_info->usage, 1); - - if (gidsetsize <= NGROUPS_SMALL) - group_info->blocks[0] = group_info->small_block; - else { - for (i = 0; i < nblocks; i++) { - gid_t *b; - b = (void *)__get_free_page(GFP_USER); - if (!b) - goto out_undo_partial_alloc; - group_info->blocks[i] = b; - } - } - return group_info; - -out_undo_partial_alloc: - while (--i >= 0) { - free_page((unsigned long)group_info->blocks[i]); - } - kfree(group_info); - return NULL; -} - -EXPORT_SYMBOL(groups_alloc); - -void groups_free(struct group_info *group_info) -{ - if (group_info->blocks[0] != group_info->small_block) { - int i; - for (i = 0; i < group_info->nblocks; i++) - free_page((unsigned long)group_info->blocks[i]); - } - kfree(group_info); -} - -EXPORT_SYMBOL(groups_free); - -/* export the group_info to a user-space array */ -static int groups_to_user(gid_t __user *grouplist, - const struct group_info *group_info) -{ - int i; - unsigned int count = group_info->ngroups; - - for (i = 0; i < group_info->nblocks; i++) { - unsigned int cp_count = min(NGROUPS_PER_BLOCK, count); - unsigned int len = cp_count * sizeof(*grouplist); - - if (copy_to_user(grouplist, group_info->blocks[i], len)) - return -EFAULT; - - grouplist += NGROUPS_PER_BLOCK; - count -= cp_count; - } - return 0; -} - -/* fill a group_info from a user-space array - it must be allocated already */ -static int groups_from_user(struct group_info *group_info, - gid_t __user *grouplist) -{ - int i; - unsigned int count = group_info->ngroups; - - for (i = 0; i < group_info->nblocks; i++) { - unsigned int cp_count = min(NGROUPS_PER_BLOCK, count); - unsigned int len = cp_count * sizeof(*grouplist); - - if (copy_from_user(group_info->blocks[i], grouplist, len)) - return -EFAULT; - - grouplist += NGROUPS_PER_BLOCK; - count -= cp_count; - } - return 0; -} - -/* a simple Shell sort */ -static void groups_sort(struct group_info *group_info) -{ - int base, max, stride; - int gidsetsize = group_info->ngroups; - - for (stride = 1; stride < gidsetsize; stride = 3 * stride + 1) - ; /* nothing */ - stride /= 3; - - while (stride) { - max = gidsetsize - stride; - for (base = 0; base < max; base++) { - int left = base; - int right = left + stride; - gid_t tmp = GROUP_AT(group_info, right); - - while (left >= 0 && GROUP_AT(group_info, left) > tmp) { - GROUP_AT(group_info, right) = - GROUP_AT(group_info, left); - right = left; - left -= stride; - } - GROUP_AT(group_info, right) = tmp; - } - stride /= 3; - } -} - -/* a simple bsearch */ -int groups_search(const struct group_info *group_info, gid_t grp) -{ - unsigned int left, right; - - if (!group_info) - return 0; - - left = 0; - right = group_info->ngroups; - while (left < right) { - unsigned int mid = (left+right)/2; - int cmp = grp - GROUP_AT(group_info, mid); - if (cmp > 0) - left = mid + 1; - else if (cmp < 0) - right = mid; - else - return 1; - } - return 0; -} - -/** - * set_groups - Change a group subscription in a set of credentials - * @new: The newly prepared set of credentials to alter - * @group_info: The group list to install - * - * Validate a group subscription and, if valid, insert it into a set - * of credentials. - */ -int set_groups(struct cred *new, struct group_info *group_info) -{ - int retval; - - retval = security_task_setgroups(group_info); - if (retval) - return retval; - - put_group_info(new->group_info); - groups_sort(group_info); - get_group_info(group_info); - new->group_info = group_info; - return 0; -} - -EXPORT_SYMBOL(set_groups); - -/** - * set_current_groups - Change current's group subscription - * @group_info: The group list to impose - * - * Validate a group subscription and, if valid, impose it upon current's task - * security record. - */ -int set_current_groups(struct group_info *group_info) -{ - struct cred *new; - int ret; - - new = prepare_creds(); - if (!new) - return -ENOMEM; - - ret = set_groups(new, group_info); - if (ret < 0) { - abort_creds(new); - return ret; - } - - return commit_creds(new); -} - -EXPORT_SYMBOL(set_current_groups); - -SYSCALL_DEFINE2(getgroups, int, gidsetsize, gid_t __user *, grouplist) -{ - const struct cred *cred = current_cred(); - int i; - - if (gidsetsize < 0) - return -EINVAL; - - /* no need to grab task_lock here; it cannot change */ - i = cred->group_info->ngroups; - if (gidsetsize) { - if (i > gidsetsize) { - i = -EINVAL; - goto out; - } - if (groups_to_user(grouplist, cred->group_info)) { - i = -EFAULT; - goto out; - } - } -out: - return i; -} - -/* - * SMP: Our groups are copy-on-write. We can set them safely - * without another task interfering. - */ - -SYSCALL_DEFINE2(setgroups, int, gidsetsize, gid_t __user *, grouplist) -{ - struct group_info *group_info; - int retval; - - if (!capable(CAP_SETGID)) - return -EPERM; - if ((unsigned)gidsetsize > NGROUPS_MAX) - return -EINVAL; - - group_info = groups_alloc(gidsetsize); - if (!group_info) - return -ENOMEM; - retval = groups_from_user(group_info, grouplist); - if (retval) { - put_group_info(group_info); - return retval; - } - - retval = set_current_groups(group_info); - put_group_info(group_info); - - return retval; -} - -/* - * Check whether we're fsgid/egid or in the supplemental group.. - */ -int in_group_p(gid_t grp) -{ - const struct cred *cred = current_cred(); - int retval = 1; - - if (grp != cred->fsgid) - retval = groups_search(cred->group_info, grp); - return retval; -} - -EXPORT_SYMBOL(in_group_p); - -int in_egroup_p(gid_t grp) -{ - const struct cred *cred = current_cred(); - int retval = 1; - - if (grp != cred->egid) - retval = groups_search(cred->group_info, grp); - return retval; -} - -EXPORT_SYMBOL(in_egroup_p); - DECLARE_RWSEM(uts_sem); SYSCALL_DEFINE1(newuname, struct new_utsname __user *, name) @@ -1625,16 +1337,16 @@ static void k_getrusage(struct task_struct *p, int who, struct rusage *r) { struct task_struct *t; unsigned long flags; - cputime_t utime, stime; - struct task_cputime cputime; + cputime_t tgutime, tgstime, utime, stime; + unsigned long maxrss = 0; memset((char *) r, 0, sizeof *r); utime = stime = cputime_zero; if (who == RUSAGE_THREAD) { - utime = task_utime(current); - stime = task_stime(current); + task_times(current, &utime, &stime); accumulate_thread_rusage(p, r); + maxrss = p->signal->maxrss; goto out; } @@ -1652,20 +1364,23 @@ static void k_getrusage(struct task_struct *p, int who, struct rusage *r) r->ru_majflt = p->signal->cmaj_flt; r->ru_inblock = p->signal->cinblock; r->ru_oublock = p->signal->coublock; + maxrss = p->signal->cmaxrss; if (who == RUSAGE_CHILDREN) break; case RUSAGE_SELF: - thread_group_cputime(p, &cputime); - utime = cputime_add(utime, cputime.utime); - stime = cputime_add(stime, cputime.stime); + thread_group_times(p, &tgutime, &tgstime); + utime = cputime_add(utime, tgutime); + stime = cputime_add(stime, tgstime); r->ru_nvcsw += p->signal->nvcsw; r->ru_nivcsw += p->signal->nivcsw; r->ru_minflt += p->signal->min_flt; r->ru_majflt += p->signal->maj_flt; r->ru_inblock += p->signal->inblock; r->ru_oublock += p->signal->oublock; + if (maxrss < p->signal->maxrss) + maxrss = p->signal->maxrss; t = p; do { accumulate_thread_rusage(t, r); @@ -1681,6 +1396,15 @@ static void k_getrusage(struct task_struct *p, int who, struct rusage *r) out: cputime_to_timeval(utime, &r->ru_utime); cputime_to_timeval(stime, &r->ru_stime); + + if (who != RUSAGE_CHILDREN) { + struct mm_struct *mm = get_task_mm(p); + if (mm) { + setmax_mm_hiwater_rss(&maxrss, mm); + mmput(mm); + } + } + r->ru_maxrss = maxrss * (PAGE_SIZE / 1024); /* convert pages to KBs */ } int getrusage(struct task_struct *p, int who, struct rusage __user *ru) @@ -1800,6 +1524,12 @@ SYSCALL_DEFINE5(prctl, int, option, unsigned long, arg2, unsigned long, arg3, case PR_SET_TSC: error = SET_TSC_CTL(arg2); break; + case PR_TASK_PERF_EVENTS_DISABLE: + error = perf_event_task_disable(); + break; + case PR_TASK_PERF_EVENTS_ENABLE: + error = perf_event_task_enable(); + break; case PR_GET_TIMERSLACK: error = current->timer_slack_ns; break; @@ -1811,6 +1541,41 @@ SYSCALL_DEFINE5(prctl, int, option, unsigned long, arg2, unsigned long, arg3, current->timer_slack_ns = arg2; error = 0; break; + case PR_MCE_KILL: + if (arg4 | arg5) + return -EINVAL; + switch (arg2) { + case PR_MCE_KILL_CLEAR: + if (arg3 != 0) + return -EINVAL; + current->flags &= ~PF_MCE_PROCESS; + break; + case PR_MCE_KILL_SET: + current->flags |= PF_MCE_PROCESS; + if (arg3 == PR_MCE_KILL_EARLY) + current->flags |= PF_MCE_EARLY; + else if (arg3 == PR_MCE_KILL_LATE) + current->flags &= ~PF_MCE_EARLY; + else if (arg3 == PR_MCE_KILL_DEFAULT) + current->flags &= + ~(PF_MCE_EARLY|PF_MCE_PROCESS); + else + return -EINVAL; + break; + default: + return -EINVAL; + } + error = 0; + break; + case PR_MCE_KILL_GET: + if (arg2 | arg3 | arg4 | arg5) + return -EINVAL; + if (current->flags & PF_MCE_PROCESS) + error = (current->flags & PF_MCE_EARLY) ? + PR_MCE_KILL_EARLY : PR_MCE_KILL_LATE; + else + error = PR_MCE_KILL_DEFAULT; + break; default: error = -EINVAL; break; diff --git a/kernel/sys_ni.c b/kernel/sys_ni.c index 27dad296738..695384f12a7 100644 --- a/kernel/sys_ni.c +++ b/kernel/sys_ni.c @@ -48,7 +48,10 @@ cond_syscall(sys_shutdown); cond_syscall(sys_sendmsg); cond_syscall(compat_sys_sendmsg); cond_syscall(sys_recvmsg); +cond_syscall(sys_recvmmsg); cond_syscall(compat_sys_recvmsg); +cond_syscall(compat_sys_recvfrom); +cond_syscall(compat_sys_recvmmsg); cond_syscall(sys_socketcall); cond_syscall(sys_futex); cond_syscall(compat_sys_futex); @@ -138,7 +141,6 @@ cond_syscall(sys_pciconfig_read); cond_syscall(sys_pciconfig_write); cond_syscall(sys_pciconfig_iobase); cond_syscall(sys32_ipc); -cond_syscall(sys32_sysctl); cond_syscall(ppc_rtas); cond_syscall(sys_spu_run); cond_syscall(sys_spu_create); @@ -175,3 +177,6 @@ cond_syscall(compat_sys_timerfd_settime); cond_syscall(compat_sys_timerfd_gettime); cond_syscall(sys_eventfd); cond_syscall(sys_eventfd2); + +/* performance counters: */ +cond_syscall(sys_perf_event_open); diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 2e490a389dd..9327a26765c 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -26,8 +26,7 @@ #include <linux/proc_fs.h> #include <linux/security.h> #include <linux/ctype.h> -#include <linux/utsname.h> -#include <linux/smp_lock.h> +#include <linux/kmemcheck.h> #include <linux/fs.h> #include <linux/init.h> #include <linux/kernel.h> @@ -36,6 +35,7 @@ #include <linux/sysrq.h> #include <linux/highuid.h> #include <linux/writeback.h> +#include <linux/ratelimit.h> #include <linux/hugetlb.h> #include <linux/initrd.h> #include <linux/key.h> @@ -48,6 +48,8 @@ #include <linux/acpi.h> #include <linux/reboot.h> #include <linux/ftrace.h> +#include <linux/slow-work.h> +#include <linux/perf_event.h> #include <asm/uaccess.h> #include <asm/processor.h> @@ -58,7 +60,6 @@ #include <asm/io.h> #endif -static int deprecated_sysctl_warning(struct __sysctl_args *args); #if defined(CONFIG_SYSCTL) @@ -74,6 +75,7 @@ extern int max_threads; extern int core_uses_pid; extern int suid_dumpable; extern char core_pattern[]; +extern unsigned int core_pipe_limit; extern int pid_max; extern int min_free_kbytes; extern int pid_max_min, pid_max_max; @@ -88,6 +90,9 @@ extern int sysctl_nr_trim_pages; #ifdef CONFIG_RCU_TORTURE_TEST extern int rcutorture_runnable; #endif /* #ifdef CONFIG_RCU_TORTURE_TEST */ +#ifdef CONFIG_BLOCK +extern int blk_iopoll_enabled; +#endif /* Constants used for minimum and maximum */ #ifdef CONFIG_DETECT_SOFTLOCKUP @@ -95,14 +100,17 @@ static int sixty = 60; static int neg_one = -1; #endif -#if defined(CONFIG_MMU) && defined(CONFIG_FILE_LOCKING) -static int two = 2; -#endif - static int zero; -static int one = 1; +static int __maybe_unused one = 1; +static int __maybe_unused two = 2; static unsigned long one_ul = 1; static int one_hundred = 100; +#ifdef CONFIG_PRINTK +static int ten_thousand = 10000; +#endif + +/* this is needed for the proc_doulongvec_minmax of vm_dirty_bytes */ +static unsigned long dirty_bytes_min = 2 * PAGE_SIZE; /* this is needed for the proc_dointvec_minmax for [fs_]overflow UID and GID */ static int maxolduid = 65535; @@ -113,6 +121,7 @@ static int ngroups_max = NGROUPS_MAX; #ifdef CONFIG_MODULES extern char modprobe_path[]; +extern int modules_disabled; #endif #ifdef CONFIG_CHR_DEV_SG extern int sg_big_buff; @@ -148,14 +157,16 @@ extern int no_unaligned_warning; extern int unaligned_dump_stack; #endif +extern struct ratelimit_state printk_ratelimit_state; + #ifdef CONFIG_RT_MUTEXES extern int max_lock_depth; #endif #ifdef CONFIG_PROC_SYSCTL -static int proc_do_cad_pid(struct ctl_table *table, int write, struct file *filp, +static int proc_do_cad_pid(struct ctl_table *table, int write, void __user *buffer, size_t *lenp, loff_t *ppos); -static int proc_taint(struct ctl_table *table, int write, struct file *filp, +static int proc_taint(struct ctl_table *table, int write, void __user *buffer, size_t *lenp, loff_t *ppos); #endif @@ -197,31 +208,26 @@ extern int lock_stat; static struct ctl_table root_table[] = { { - .ctl_name = CTL_KERN, .procname = "kernel", .mode = 0555, .child = kern_table, }, { - .ctl_name = CTL_VM, .procname = "vm", .mode = 0555, .child = vm_table, }, { - .ctl_name = CTL_FS, .procname = "fs", .mode = 0555, .child = fs_table, }, { - .ctl_name = CTL_DEBUG, .procname = "debug", .mode = 0555, .child = debug_table, }, { - .ctl_name = CTL_DEV, .procname = "dev", .mode = 0555, .child = dev_table, @@ -230,7 +236,7 @@ static struct ctl_table root_table[] = { * NOTE: do not add new entries to this table unless you have read * Documentation/sysctl/ctl_unnumbered.txt */ - { .ctl_name = 0 } + { } }; #ifdef CONFIG_SCHED_DEBUG @@ -241,166 +247,167 @@ static int max_wakeup_granularity_ns = NSEC_PER_SEC; /* 1 second */ #endif static struct ctl_table kern_table[] = { + { + .procname = "sched_child_runs_first", + .data = &sysctl_sched_child_runs_first, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = proc_dointvec, + }, #ifdef CONFIG_SCHED_DEBUG { - .ctl_name = CTL_UNNUMBERED, .procname = "sched_min_granularity_ns", .data = &sysctl_sched_min_granularity, .maxlen = sizeof(unsigned int), .mode = 0644, - .proc_handler = &sched_nr_latency_handler, - .strategy = &sysctl_intvec, + .proc_handler = sched_nr_latency_handler, .extra1 = &min_sched_granularity_ns, .extra2 = &max_sched_granularity_ns, }, { - .ctl_name = CTL_UNNUMBERED, .procname = "sched_latency_ns", .data = &sysctl_sched_latency, .maxlen = sizeof(unsigned int), .mode = 0644, - .proc_handler = &sched_nr_latency_handler, - .strategy = &sysctl_intvec, + .proc_handler = sched_nr_latency_handler, .extra1 = &min_sched_granularity_ns, .extra2 = &max_sched_granularity_ns, }, { - .ctl_name = CTL_UNNUMBERED, .procname = "sched_wakeup_granularity_ns", .data = &sysctl_sched_wakeup_granularity, .maxlen = sizeof(unsigned int), .mode = 0644, - .proc_handler = &proc_dointvec_minmax, - .strategy = &sysctl_intvec, + .proc_handler = proc_dointvec_minmax, .extra1 = &min_wakeup_granularity_ns, .extra2 = &max_wakeup_granularity_ns, }, { - .ctl_name = CTL_UNNUMBERED, .procname = "sched_shares_ratelimit", .data = &sysctl_sched_shares_ratelimit, .maxlen = sizeof(unsigned int), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, { - .ctl_name = CTL_UNNUMBERED, .procname = "sched_shares_thresh", .data = &sysctl_sched_shares_thresh, .maxlen = sizeof(unsigned int), .mode = 0644, - .proc_handler = &proc_dointvec_minmax, - .strategy = &sysctl_intvec, + .proc_handler = proc_dointvec_minmax, .extra1 = &zero, }, { - .ctl_name = CTL_UNNUMBERED, - .procname = "sched_child_runs_first", - .data = &sysctl_sched_child_runs_first, - .maxlen = sizeof(unsigned int), - .mode = 0644, - .proc_handler = &proc_dointvec, - }, - { - .ctl_name = CTL_UNNUMBERED, .procname = "sched_features", .data = &sysctl_sched_features, .maxlen = sizeof(unsigned int), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, { - .ctl_name = CTL_UNNUMBERED, .procname = "sched_migration_cost", .data = &sysctl_sched_migration_cost, .maxlen = sizeof(unsigned int), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, { - .ctl_name = CTL_UNNUMBERED, .procname = "sched_nr_migrate", .data = &sysctl_sched_nr_migrate, .maxlen = sizeof(unsigned int), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, + }, + { + .procname = "sched_time_avg", + .data = &sysctl_sched_time_avg, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = proc_dointvec, + }, + { + .procname = "timer_migration", + .data = &sysctl_timer_migration, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = proc_dointvec_minmax, + .extra1 = &zero, + .extra2 = &one, }, #endif { - .ctl_name = CTL_UNNUMBERED, .procname = "sched_rt_period_us", .data = &sysctl_sched_rt_period, .maxlen = sizeof(unsigned int), .mode = 0644, - .proc_handler = &sched_rt_handler, + .proc_handler = sched_rt_handler, }, { - .ctl_name = CTL_UNNUMBERED, .procname = "sched_rt_runtime_us", .data = &sysctl_sched_rt_runtime, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = &sched_rt_handler, + .proc_handler = sched_rt_handler, }, { - .ctl_name = CTL_UNNUMBERED, .procname = "sched_compat_yield", .data = &sysctl_sched_compat_yield, .maxlen = sizeof(unsigned int), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, #ifdef CONFIG_PROVE_LOCKING { - .ctl_name = CTL_UNNUMBERED, .procname = "prove_locking", .data = &prove_locking, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, #endif #ifdef CONFIG_LOCK_STAT { - .ctl_name = CTL_UNNUMBERED, .procname = "lock_stat", .data = &lock_stat, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, #endif { - .ctl_name = KERN_PANIC, .procname = "panic", .data = &panic_timeout, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, { - .ctl_name = KERN_CORE_USES_PID, .procname = "core_uses_pid", .data = &core_uses_pid, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, { - .ctl_name = KERN_CORE_PATTERN, .procname = "core_pattern", .data = core_pattern, .maxlen = CORENAME_MAX_SIZE, .mode = 0644, - .proc_handler = &proc_dostring, - .strategy = &sysctl_string, + .proc_handler = proc_dostring, + }, + { + .procname = "core_pipe_limit", + .data = &core_pipe_limit, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = proc_dointvec, }, #ifdef CONFIG_PROC_SYSCTL { .procname = "tainted", .maxlen = sizeof(long), .mode = 0644, - .proc_handler = &proc_taint, + .proc_handler = proc_taint, }, #endif #ifdef CONFIG_LATENCYTOP @@ -409,170 +416,160 @@ static struct ctl_table kern_table[] = { .data = &latencytop_enabled, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, #endif #ifdef CONFIG_BLK_DEV_INITRD { - .ctl_name = KERN_REALROOTDEV, .procname = "real-root-dev", .data = &real_root_dev, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, #endif { - .ctl_name = CTL_UNNUMBERED, .procname = "print-fatal-signals", .data = &print_fatal_signals, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, #ifdef CONFIG_SPARC { - .ctl_name = KERN_SPARC_REBOOT, .procname = "reboot-cmd", .data = reboot_command, .maxlen = 256, .mode = 0644, - .proc_handler = &proc_dostring, - .strategy = &sysctl_string, + .proc_handler = proc_dostring, }, { - .ctl_name = KERN_SPARC_STOP_A, .procname = "stop-a", .data = &stop_a_enabled, .maxlen = sizeof (int), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, { - .ctl_name = KERN_SPARC_SCONS_PWROFF, .procname = "scons-poweroff", .data = &scons_pwroff, .maxlen = sizeof (int), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, #endif #ifdef CONFIG_SPARC64 { - .ctl_name = CTL_UNNUMBERED, .procname = "tsb-ratio", .data = &sysctl_tsb_ratio, .maxlen = sizeof (int), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, #endif #ifdef __hppa__ { - .ctl_name = KERN_HPPA_PWRSW, .procname = "soft-power", .data = &pwrsw_enabled, .maxlen = sizeof (int), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, { - .ctl_name = KERN_HPPA_UNALIGNED, .procname = "unaligned-trap", .data = &unaligned_enabled, .maxlen = sizeof (int), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, #endif { - .ctl_name = KERN_CTLALTDEL, .procname = "ctrl-alt-del", .data = &C_A_D, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, #ifdef CONFIG_FUNCTION_TRACER { - .ctl_name = CTL_UNNUMBERED, .procname = "ftrace_enabled", .data = &ftrace_enabled, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = &ftrace_enable_sysctl, + .proc_handler = ftrace_enable_sysctl, }, #endif #ifdef CONFIG_STACK_TRACER { - .ctl_name = CTL_UNNUMBERED, .procname = "stack_tracer_enabled", .data = &stack_tracer_enabled, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = &stack_trace_sysctl, + .proc_handler = stack_trace_sysctl, }, #endif #ifdef CONFIG_TRACING { - .ctl_name = CTL_UNNUMBERED, .procname = "ftrace_dump_on_oops", .data = &ftrace_dump_on_oops, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, #endif #ifdef CONFIG_MODULES { - .ctl_name = KERN_MODPROBE, .procname = "modprobe", .data = &modprobe_path, .maxlen = KMOD_PATH_LEN, .mode = 0644, - .proc_handler = &proc_dostring, - .strategy = &sysctl_string, + .proc_handler = proc_dostring, + }, + { + .procname = "modules_disabled", + .data = &modules_disabled, + .maxlen = sizeof(int), + .mode = 0644, + /* only handle a transition from default "0" to "1" */ + .proc_handler = proc_dointvec_minmax, + .extra1 = &one, + .extra2 = &one, }, #endif #if defined(CONFIG_HOTPLUG) && defined(CONFIG_NET) { - .ctl_name = KERN_HOTPLUG, .procname = "hotplug", .data = &uevent_helper, .maxlen = UEVENT_HELPER_PATH_LEN, .mode = 0644, - .proc_handler = &proc_dostring, - .strategy = &sysctl_string, + .proc_handler = proc_dostring, }, #endif #ifdef CONFIG_CHR_DEV_SG { - .ctl_name = KERN_SG_BIG_BUFF, .procname = "sg-big-buff", .data = &sg_big_buff, .maxlen = sizeof (int), .mode = 0444, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, #endif #ifdef CONFIG_BSD_PROCESS_ACCT { - .ctl_name = KERN_ACCT, .procname = "acct", .data = &acct_parm, .maxlen = 3*sizeof(int), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, #endif #ifdef CONFIG_MAGIC_SYSRQ { - .ctl_name = KERN_SYSRQ, .procname = "sysrq", .data = &__sysrq_enabled, .maxlen = sizeof (int), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, #endif #ifdef CONFIG_PROC_SYSCTL @@ -581,188 +578,188 @@ static struct ctl_table kern_table[] = { .data = NULL, .maxlen = sizeof (int), .mode = 0600, - .proc_handler = &proc_do_cad_pid, + .proc_handler = proc_do_cad_pid, }, #endif { - .ctl_name = KERN_MAX_THREADS, .procname = "threads-max", .data = &max_threads, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, { - .ctl_name = KERN_RANDOM, .procname = "random", .mode = 0555, .child = random_table, }, { - .ctl_name = KERN_OVERFLOWUID, .procname = "overflowuid", .data = &overflowuid, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = &proc_dointvec_minmax, - .strategy = &sysctl_intvec, + .proc_handler = proc_dointvec_minmax, .extra1 = &minolduid, .extra2 = &maxolduid, }, { - .ctl_name = KERN_OVERFLOWGID, .procname = "overflowgid", .data = &overflowgid, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = &proc_dointvec_minmax, - .strategy = &sysctl_intvec, + .proc_handler = proc_dointvec_minmax, .extra1 = &minolduid, .extra2 = &maxolduid, }, #ifdef CONFIG_S390 #ifdef CONFIG_MATHEMU { - .ctl_name = KERN_IEEE_EMULATION_WARNINGS, .procname = "ieee_emulation_warnings", .data = &sysctl_ieee_emulation_warnings, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, #endif { - .ctl_name = KERN_S390_USER_DEBUG_LOGGING, .procname = "userprocess_debug", .data = &sysctl_userprocess_debug, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, #endif { - .ctl_name = KERN_PIDMAX, .procname = "pid_max", .data = &pid_max, .maxlen = sizeof (int), .mode = 0644, - .proc_handler = &proc_dointvec_minmax, - .strategy = sysctl_intvec, + .proc_handler = proc_dointvec_minmax, .extra1 = &pid_max_min, .extra2 = &pid_max_max, }, { - .ctl_name = KERN_PANIC_ON_OOPS, .procname = "panic_on_oops", .data = &panic_on_oops, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, #if defined CONFIG_PRINTK { - .ctl_name = KERN_PRINTK, .procname = "printk", .data = &console_loglevel, .maxlen = 4*sizeof(int), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, { - .ctl_name = KERN_PRINTK_RATELIMIT, .procname = "printk_ratelimit", .data = &printk_ratelimit_state.interval, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = &proc_dointvec_jiffies, - .strategy = &sysctl_jiffies, + .proc_handler = proc_dointvec_jiffies, }, { - .ctl_name = KERN_PRINTK_RATELIMIT_BURST, .procname = "printk_ratelimit_burst", .data = &printk_ratelimit_state.burst, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, + }, + { + .procname = "printk_delay", + .data = &printk_delay_msec, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec_minmax, + .extra1 = &zero, + .extra2 = &ten_thousand, }, #endif { - .ctl_name = KERN_NGROUPS_MAX, .procname = "ngroups_max", .data = &ngroups_max, .maxlen = sizeof (int), .mode = 0444, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, #if defined(CONFIG_X86_LOCAL_APIC) && defined(CONFIG_X86) { - .ctl_name = KERN_UNKNOWN_NMI_PANIC, .procname = "unknown_nmi_panic", .data = &unknown_nmi_panic, .maxlen = sizeof (int), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, { .procname = "nmi_watchdog", .data = &nmi_watchdog_enabled, .maxlen = sizeof (int), .mode = 0644, - .proc_handler = &proc_nmi_enabled, + .proc_handler = proc_nmi_enabled, }, #endif #if defined(CONFIG_X86) { - .ctl_name = KERN_PANIC_ON_NMI, .procname = "panic_on_unrecovered_nmi", .data = &panic_on_unrecovered_nmi, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, + }, + { + .procname = "panic_on_io_nmi", + .data = &panic_on_io_nmi, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec, }, { - .ctl_name = KERN_BOOTLOADER_TYPE, .procname = "bootloader_type", .data = &bootloader_type, .maxlen = sizeof (int), .mode = 0444, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, + }, + { + .procname = "bootloader_version", + .data = &bootloader_version, + .maxlen = sizeof (int), + .mode = 0444, + .proc_handler = proc_dointvec, }, { - .ctl_name = CTL_UNNUMBERED, .procname = "kstack_depth_to_print", .data = &kstack_depth_to_print, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, { - .ctl_name = CTL_UNNUMBERED, .procname = "io_delay_type", .data = &io_delay_type, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, #endif #if defined(CONFIG_MMU) { - .ctl_name = KERN_RANDOMIZE, .procname = "randomize_va_space", .data = &randomize_va_space, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, #endif #if defined(CONFIG_S390) && defined(CONFIG_SMP) { - .ctl_name = KERN_SPIN_RETRY, .procname = "spin_retry", .data = &spin_retry, .maxlen = sizeof (int), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, #endif #if defined(CONFIG_ACPI_SLEEP) && defined(CONFIG_X86) @@ -771,110 +768,104 @@ static struct ctl_table kern_table[] = { .data = &acpi_realmode_flags, .maxlen = sizeof (unsigned long), .mode = 0644, - .proc_handler = &proc_doulongvec_minmax, + .proc_handler = proc_doulongvec_minmax, }, #endif #ifdef CONFIG_IA64 { - .ctl_name = KERN_IA64_UNALIGNED, .procname = "ignore-unaligned-usertrap", .data = &no_unaligned_warning, .maxlen = sizeof (int), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, { - .ctl_name = CTL_UNNUMBERED, .procname = "unaligned-dump-stack", .data = &unaligned_dump_stack, .maxlen = sizeof (int), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, #endif #ifdef CONFIG_DETECT_SOFTLOCKUP { - .ctl_name = CTL_UNNUMBERED, .procname = "softlockup_panic", .data = &softlockup_panic, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = &proc_dointvec_minmax, - .strategy = &sysctl_intvec, + .proc_handler = proc_dointvec_minmax, .extra1 = &zero, .extra2 = &one, }, { - .ctl_name = CTL_UNNUMBERED, .procname = "softlockup_thresh", .data = &softlockup_thresh, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = &proc_dosoftlockup_thresh, - .strategy = &sysctl_intvec, + .proc_handler = proc_dosoftlockup_thresh, .extra1 = &neg_one, .extra2 = &sixty, }, +#endif +#ifdef CONFIG_DETECT_HUNG_TASK + { + .procname = "hung_task_panic", + .data = &sysctl_hung_task_panic, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec_minmax, + .extra1 = &zero, + .extra2 = &one, + }, { - .ctl_name = CTL_UNNUMBERED, .procname = "hung_task_check_count", .data = &sysctl_hung_task_check_count, .maxlen = sizeof(unsigned long), .mode = 0644, - .proc_handler = &proc_doulongvec_minmax, - .strategy = &sysctl_intvec, + .proc_handler = proc_doulongvec_minmax, }, { - .ctl_name = CTL_UNNUMBERED, .procname = "hung_task_timeout_secs", .data = &sysctl_hung_task_timeout_secs, .maxlen = sizeof(unsigned long), .mode = 0644, - .proc_handler = &proc_doulongvec_minmax, - .strategy = &sysctl_intvec, + .proc_handler = proc_dohung_task_timeout_secs, }, { - .ctl_name = CTL_UNNUMBERED, .procname = "hung_task_warnings", .data = &sysctl_hung_task_warnings, .maxlen = sizeof(unsigned long), .mode = 0644, - .proc_handler = &proc_doulongvec_minmax, - .strategy = &sysctl_intvec, + .proc_handler = proc_doulongvec_minmax, }, #endif #ifdef CONFIG_COMPAT { - .ctl_name = KERN_COMPAT_LOG, .procname = "compat-log", .data = &compat_log, .maxlen = sizeof (int), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, #endif #ifdef CONFIG_RT_MUTEXES { - .ctl_name = KERN_MAX_LOCK_DEPTH, .procname = "max_lock_depth", .data = &max_lock_depth, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, #endif { - .ctl_name = CTL_UNNUMBERED, .procname = "poweroff_cmd", .data = &poweroff_cmd, .maxlen = POWEROFF_CMD_PATH_LEN, .mode = 0644, - .proc_handler = &proc_dostring, - .strategy = &sysctl_string, + .proc_handler = proc_dostring, }, #ifdef CONFIG_KEYS { - .ctl_name = CTL_UNNUMBERED, .procname = "keys", .mode = 0555, .child = key_sysctls, @@ -882,152 +873,172 @@ static struct ctl_table kern_table[] = { #endif #ifdef CONFIG_RCU_TORTURE_TEST { - .ctl_name = CTL_UNNUMBERED, .procname = "rcutorture_runnable", .data = &rcutorture_runnable, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, #endif -#ifdef CONFIG_UNEVICTABLE_LRU +#ifdef CONFIG_SLOW_WORK { - .ctl_name = CTL_UNNUMBERED, - .procname = "scan_unevictable_pages", - .data = &scan_unevictable_pages, - .maxlen = sizeof(scan_unevictable_pages), + .procname = "slow-work", + .mode = 0555, + .child = slow_work_sysctls, + }, +#endif +#ifdef CONFIG_PERF_EVENTS + { + .procname = "perf_event_paranoid", + .data = &sysctl_perf_event_paranoid, + .maxlen = sizeof(sysctl_perf_event_paranoid), + .mode = 0644, + .proc_handler = proc_dointvec, + }, + { + .procname = "perf_event_mlock_kb", + .data = &sysctl_perf_event_mlock, + .maxlen = sizeof(sysctl_perf_event_mlock), + .mode = 0644, + .proc_handler = proc_dointvec, + }, + { + .procname = "perf_event_max_sample_rate", + .data = &sysctl_perf_event_sample_rate, + .maxlen = sizeof(sysctl_perf_event_sample_rate), + .mode = 0644, + .proc_handler = proc_dointvec, + }, +#endif +#ifdef CONFIG_KMEMCHECK + { + .procname = "kmemcheck", + .data = &kmemcheck_enabled, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec, + }, +#endif +#ifdef CONFIG_BLOCK + { + .procname = "blk_iopoll", + .data = &blk_iopoll_enabled, + .maxlen = sizeof(int), .mode = 0644, - .proc_handler = &scan_unevictable_handler, + .proc_handler = proc_dointvec, }, #endif /* * NOTE: do not add new entries to this table unless you have read * Documentation/sysctl/ctl_unnumbered.txt */ - { .ctl_name = 0 } + { } }; static struct ctl_table vm_table[] = { { - .ctl_name = VM_OVERCOMMIT_MEMORY, .procname = "overcommit_memory", .data = &sysctl_overcommit_memory, .maxlen = sizeof(sysctl_overcommit_memory), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, { - .ctl_name = VM_PANIC_ON_OOM, .procname = "panic_on_oom", .data = &sysctl_panic_on_oom, .maxlen = sizeof(sysctl_panic_on_oom), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, { - .ctl_name = CTL_UNNUMBERED, .procname = "oom_kill_allocating_task", .data = &sysctl_oom_kill_allocating_task, .maxlen = sizeof(sysctl_oom_kill_allocating_task), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, { - .ctl_name = CTL_UNNUMBERED, .procname = "oom_dump_tasks", .data = &sysctl_oom_dump_tasks, .maxlen = sizeof(sysctl_oom_dump_tasks), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, { - .ctl_name = VM_OVERCOMMIT_RATIO, .procname = "overcommit_ratio", .data = &sysctl_overcommit_ratio, .maxlen = sizeof(sysctl_overcommit_ratio), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, { - .ctl_name = VM_PAGE_CLUSTER, .procname = "page-cluster", .data = &page_cluster, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, { - .ctl_name = VM_DIRTY_BACKGROUND, .procname = "dirty_background_ratio", .data = &dirty_background_ratio, .maxlen = sizeof(dirty_background_ratio), .mode = 0644, - .proc_handler = &dirty_background_ratio_handler, - .strategy = &sysctl_intvec, + .proc_handler = dirty_background_ratio_handler, .extra1 = &zero, .extra2 = &one_hundred, }, { - .ctl_name = CTL_UNNUMBERED, .procname = "dirty_background_bytes", .data = &dirty_background_bytes, .maxlen = sizeof(dirty_background_bytes), .mode = 0644, - .proc_handler = &dirty_background_bytes_handler, - .strategy = &sysctl_intvec, + .proc_handler = dirty_background_bytes_handler, .extra1 = &one_ul, }, { - .ctl_name = VM_DIRTY_RATIO, .procname = "dirty_ratio", .data = &vm_dirty_ratio, .maxlen = sizeof(vm_dirty_ratio), .mode = 0644, - .proc_handler = &dirty_ratio_handler, - .strategy = &sysctl_intvec, + .proc_handler = dirty_ratio_handler, .extra1 = &zero, .extra2 = &one_hundred, }, { - .ctl_name = CTL_UNNUMBERED, .procname = "dirty_bytes", .data = &vm_dirty_bytes, .maxlen = sizeof(vm_dirty_bytes), .mode = 0644, - .proc_handler = &dirty_bytes_handler, - .strategy = &sysctl_intvec, - .extra1 = &one_ul, + .proc_handler = dirty_bytes_handler, + .extra1 = &dirty_bytes_min, }, { .procname = "dirty_writeback_centisecs", .data = &dirty_writeback_interval, .maxlen = sizeof(dirty_writeback_interval), .mode = 0644, - .proc_handler = &dirty_writeback_centisecs_handler, + .proc_handler = dirty_writeback_centisecs_handler, }, { .procname = "dirty_expire_centisecs", .data = &dirty_expire_interval, .maxlen = sizeof(dirty_expire_interval), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, { - .ctl_name = VM_NR_PDFLUSH_THREADS, .procname = "nr_pdflush_threads", .data = &nr_pdflush_threads, .maxlen = sizeof nr_pdflush_threads, .mode = 0444 /* read-only*/, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, { - .ctl_name = VM_SWAPPINESS, .procname = "swappiness", .data = &vm_swappiness, .maxlen = sizeof(vm_swappiness), .mode = 0644, - .proc_handler = &proc_dointvec_minmax, - .strategy = &sysctl_intvec, + .proc_handler = proc_dointvec_minmax, .extra1 = &zero, .extra2 = &one_hundred, }, @@ -1037,346 +1048,320 @@ static struct ctl_table vm_table[] = { .data = NULL, .maxlen = sizeof(unsigned long), .mode = 0644, - .proc_handler = &hugetlb_sysctl_handler, + .proc_handler = hugetlb_sysctl_handler, .extra1 = (void *)&hugetlb_zero, .extra2 = (void *)&hugetlb_infinity, }, { - .ctl_name = VM_HUGETLB_GROUP, .procname = "hugetlb_shm_group", .data = &sysctl_hugetlb_shm_group, .maxlen = sizeof(gid_t), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, { - .ctl_name = CTL_UNNUMBERED, .procname = "hugepages_treat_as_movable", .data = &hugepages_treat_as_movable, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = &hugetlb_treat_movable_handler, + .proc_handler = hugetlb_treat_movable_handler, }, { - .ctl_name = CTL_UNNUMBERED, .procname = "nr_overcommit_hugepages", .data = NULL, .maxlen = sizeof(unsigned long), .mode = 0644, - .proc_handler = &hugetlb_overcommit_handler, + .proc_handler = hugetlb_overcommit_handler, .extra1 = (void *)&hugetlb_zero, .extra2 = (void *)&hugetlb_infinity, }, #endif { - .ctl_name = VM_LOWMEM_RESERVE_RATIO, .procname = "lowmem_reserve_ratio", .data = &sysctl_lowmem_reserve_ratio, .maxlen = sizeof(sysctl_lowmem_reserve_ratio), .mode = 0644, - .proc_handler = &lowmem_reserve_ratio_sysctl_handler, - .strategy = &sysctl_intvec, + .proc_handler = lowmem_reserve_ratio_sysctl_handler, }, { - .ctl_name = VM_DROP_PAGECACHE, .procname = "drop_caches", .data = &sysctl_drop_caches, .maxlen = sizeof(int), .mode = 0644, .proc_handler = drop_caches_sysctl_handler, - .strategy = &sysctl_intvec, }, { - .ctl_name = VM_MIN_FREE_KBYTES, .procname = "min_free_kbytes", .data = &min_free_kbytes, .maxlen = sizeof(min_free_kbytes), .mode = 0644, - .proc_handler = &min_free_kbytes_sysctl_handler, - .strategy = &sysctl_intvec, + .proc_handler = min_free_kbytes_sysctl_handler, .extra1 = &zero, }, { - .ctl_name = VM_PERCPU_PAGELIST_FRACTION, .procname = "percpu_pagelist_fraction", .data = &percpu_pagelist_fraction, .maxlen = sizeof(percpu_pagelist_fraction), .mode = 0644, - .proc_handler = &percpu_pagelist_fraction_sysctl_handler, - .strategy = &sysctl_intvec, + .proc_handler = percpu_pagelist_fraction_sysctl_handler, .extra1 = &min_percpu_pagelist_fract, }, #ifdef CONFIG_MMU { - .ctl_name = VM_MAX_MAP_COUNT, .procname = "max_map_count", .data = &sysctl_max_map_count, .maxlen = sizeof(sysctl_max_map_count), .mode = 0644, - .proc_handler = &proc_dointvec + .proc_handler = proc_dointvec }, #else { - .ctl_name = CTL_UNNUMBERED, .procname = "nr_trim_pages", .data = &sysctl_nr_trim_pages, .maxlen = sizeof(sysctl_nr_trim_pages), .mode = 0644, - .proc_handler = &proc_dointvec_minmax, - .strategy = &sysctl_intvec, + .proc_handler = proc_dointvec_minmax, .extra1 = &zero, }, #endif { - .ctl_name = VM_LAPTOP_MODE, .procname = "laptop_mode", .data = &laptop_mode, .maxlen = sizeof(laptop_mode), .mode = 0644, - .proc_handler = &proc_dointvec_jiffies, - .strategy = &sysctl_jiffies, + .proc_handler = proc_dointvec_jiffies, }, { - .ctl_name = VM_BLOCK_DUMP, .procname = "block_dump", .data = &block_dump, .maxlen = sizeof(block_dump), .mode = 0644, - .proc_handler = &proc_dointvec, - .strategy = &sysctl_intvec, + .proc_handler = proc_dointvec, .extra1 = &zero, }, { - .ctl_name = VM_VFS_CACHE_PRESSURE, .procname = "vfs_cache_pressure", .data = &sysctl_vfs_cache_pressure, .maxlen = sizeof(sysctl_vfs_cache_pressure), .mode = 0644, - .proc_handler = &proc_dointvec, - .strategy = &sysctl_intvec, + .proc_handler = proc_dointvec, .extra1 = &zero, }, #ifdef HAVE_ARCH_PICK_MMAP_LAYOUT { - .ctl_name = VM_LEGACY_VA_LAYOUT, .procname = "legacy_va_layout", .data = &sysctl_legacy_va_layout, .maxlen = sizeof(sysctl_legacy_va_layout), .mode = 0644, - .proc_handler = &proc_dointvec, - .strategy = &sysctl_intvec, + .proc_handler = proc_dointvec, .extra1 = &zero, }, #endif #ifdef CONFIG_NUMA { - .ctl_name = VM_ZONE_RECLAIM_MODE, .procname = "zone_reclaim_mode", .data = &zone_reclaim_mode, .maxlen = sizeof(zone_reclaim_mode), .mode = 0644, - .proc_handler = &proc_dointvec, - .strategy = &sysctl_intvec, + .proc_handler = proc_dointvec, .extra1 = &zero, }, { - .ctl_name = VM_MIN_UNMAPPED, .procname = "min_unmapped_ratio", .data = &sysctl_min_unmapped_ratio, .maxlen = sizeof(sysctl_min_unmapped_ratio), .mode = 0644, - .proc_handler = &sysctl_min_unmapped_ratio_sysctl_handler, - .strategy = &sysctl_intvec, + .proc_handler = sysctl_min_unmapped_ratio_sysctl_handler, .extra1 = &zero, .extra2 = &one_hundred, }, { - .ctl_name = VM_MIN_SLAB, .procname = "min_slab_ratio", .data = &sysctl_min_slab_ratio, .maxlen = sizeof(sysctl_min_slab_ratio), .mode = 0644, - .proc_handler = &sysctl_min_slab_ratio_sysctl_handler, - .strategy = &sysctl_intvec, + .proc_handler = sysctl_min_slab_ratio_sysctl_handler, .extra1 = &zero, .extra2 = &one_hundred, }, #endif #ifdef CONFIG_SMP { - .ctl_name = CTL_UNNUMBERED, .procname = "stat_interval", .data = &sysctl_stat_interval, .maxlen = sizeof(sysctl_stat_interval), .mode = 0644, - .proc_handler = &proc_dointvec_jiffies, - .strategy = &sysctl_jiffies, + .proc_handler = proc_dointvec_jiffies, }, #endif -#ifdef CONFIG_SECURITY { - .ctl_name = CTL_UNNUMBERED, .procname = "mmap_min_addr", - .data = &mmap_min_addr, - .maxlen = sizeof(unsigned long), + .data = &dac_mmap_min_addr, + .maxlen = sizeof(unsigned long), .mode = 0644, - .proc_handler = &proc_doulongvec_minmax, + .proc_handler = mmap_min_addr_handler, }, -#endif #ifdef CONFIG_NUMA { - .ctl_name = CTL_UNNUMBERED, .procname = "numa_zonelist_order", .data = &numa_zonelist_order, .maxlen = NUMA_ZONELIST_ORDER_LEN, .mode = 0644, - .proc_handler = &numa_zonelist_order_handler, - .strategy = &sysctl_string, + .proc_handler = numa_zonelist_order_handler, }, #endif #if (defined(CONFIG_X86_32) && !defined(CONFIG_UML))|| \ (defined(CONFIG_SUPERH) && defined(CONFIG_VSYSCALL)) { - .ctl_name = VM_VDSO_ENABLED, .procname = "vdso_enabled", .data = &vdso_enabled, .maxlen = sizeof(vdso_enabled), .mode = 0644, - .proc_handler = &proc_dointvec, - .strategy = &sysctl_intvec, + .proc_handler = proc_dointvec, .extra1 = &zero, }, #endif #ifdef CONFIG_HIGHMEM { - .ctl_name = CTL_UNNUMBERED, .procname = "highmem_is_dirtyable", .data = &vm_highmem_is_dirtyable, .maxlen = sizeof(vm_highmem_is_dirtyable), .mode = 0644, - .proc_handler = &proc_dointvec_minmax, - .strategy = &sysctl_intvec, + .proc_handler = proc_dointvec_minmax, .extra1 = &zero, .extra2 = &one, }, #endif + { + .procname = "scan_unevictable_pages", + .data = &scan_unevictable_pages, + .maxlen = sizeof(scan_unevictable_pages), + .mode = 0644, + .proc_handler = scan_unevictable_handler, + }, +#ifdef CONFIG_MEMORY_FAILURE + { + .procname = "memory_failure_early_kill", + .data = &sysctl_memory_failure_early_kill, + .maxlen = sizeof(sysctl_memory_failure_early_kill), + .mode = 0644, + .proc_handler = proc_dointvec_minmax, + .extra1 = &zero, + .extra2 = &one, + }, + { + .procname = "memory_failure_recovery", + .data = &sysctl_memory_failure_recovery, + .maxlen = sizeof(sysctl_memory_failure_recovery), + .mode = 0644, + .proc_handler = proc_dointvec_minmax, + .extra1 = &zero, + .extra2 = &one, + }, +#endif + /* * NOTE: do not add new entries to this table unless you have read * Documentation/sysctl/ctl_unnumbered.txt */ - { .ctl_name = 0 } + { } }; #if defined(CONFIG_BINFMT_MISC) || defined(CONFIG_BINFMT_MISC_MODULE) static struct ctl_table binfmt_misc_table[] = { - { .ctl_name = 0 } + { } }; #endif static struct ctl_table fs_table[] = { { - .ctl_name = FS_NRINODE, .procname = "inode-nr", .data = &inodes_stat, .maxlen = 2*sizeof(int), .mode = 0444, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, { - .ctl_name = FS_STATINODE, .procname = "inode-state", .data = &inodes_stat, .maxlen = 7*sizeof(int), .mode = 0444, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, { .procname = "file-nr", .data = &files_stat, .maxlen = 3*sizeof(int), .mode = 0444, - .proc_handler = &proc_nr_files, + .proc_handler = proc_nr_files, }, { - .ctl_name = FS_MAXFILE, .procname = "file-max", .data = &files_stat.max_files, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, { - .ctl_name = CTL_UNNUMBERED, .procname = "nr_open", .data = &sysctl_nr_open, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = &proc_dointvec_minmax, + .proc_handler = proc_dointvec_minmax, .extra1 = &sysctl_nr_open_min, .extra2 = &sysctl_nr_open_max, }, { - .ctl_name = FS_DENTRY, .procname = "dentry-state", .data = &dentry_stat, .maxlen = 6*sizeof(int), .mode = 0444, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, { - .ctl_name = FS_OVERFLOWUID, .procname = "overflowuid", .data = &fs_overflowuid, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = &proc_dointvec_minmax, - .strategy = &sysctl_intvec, + .proc_handler = proc_dointvec_minmax, .extra1 = &minolduid, .extra2 = &maxolduid, }, { - .ctl_name = FS_OVERFLOWGID, .procname = "overflowgid", .data = &fs_overflowgid, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = &proc_dointvec_minmax, - .strategy = &sysctl_intvec, + .proc_handler = proc_dointvec_minmax, .extra1 = &minolduid, .extra2 = &maxolduid, }, #ifdef CONFIG_FILE_LOCKING { - .ctl_name = FS_LEASES, .procname = "leases-enable", .data = &leases_enable, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, #endif #ifdef CONFIG_DNOTIFY { - .ctl_name = FS_DIR_NOTIFY, .procname = "dir-notify-enable", .data = &dir_notify_enable, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, #endif #ifdef CONFIG_MMU #ifdef CONFIG_FILE_LOCKING { - .ctl_name = FS_LEASE_TIME, .procname = "lease-break-time", .data = &lease_break_time, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = &proc_dointvec_minmax, - .strategy = &sysctl_intvec, - .extra1 = &zero, - .extra2 = &two, + .proc_handler = proc_dointvec, }, #endif #ifdef CONFIG_AIO @@ -1385,19 +1370,18 @@ static struct ctl_table fs_table[] = { .data = &aio_nr, .maxlen = sizeof(aio_nr), .mode = 0444, - .proc_handler = &proc_doulongvec_minmax, + .proc_handler = proc_doulongvec_minmax, }, { .procname = "aio-max-nr", .data = &aio_max_nr, .maxlen = sizeof(aio_max_nr), .mode = 0644, - .proc_handler = &proc_doulongvec_minmax, + .proc_handler = proc_doulongvec_minmax, }, #endif /* CONFIG_AIO */ #ifdef CONFIG_INOTIFY_USER { - .ctl_name = FS_INOTIFY, .procname = "inotify", .mode = 0555, .child = inotify_table, @@ -1412,16 +1396,16 @@ static struct ctl_table fs_table[] = { #endif #endif { - .ctl_name = KERN_SETUID_DUMPABLE, .procname = "suid_dumpable", .data = &suid_dumpable, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec_minmax, + .extra1 = &zero, + .extra2 = &two, }, #if defined(CONFIG_BINFMT_MISC) || defined(CONFIG_BINFMT_MISC_MODULE) { - .ctl_name = CTL_UNNUMBERED, .procname = "binfmt_misc", .mode = 0555, .child = binfmt_misc_table, @@ -1431,13 +1415,12 @@ static struct ctl_table fs_table[] = { * NOTE: do not add new entries to this table unless you have read * Documentation/sysctl/ctl_unnumbered.txt */ - { .ctl_name = 0 } + { } }; static struct ctl_table debug_table[] = { #if defined(CONFIG_X86) || defined(CONFIG_PPC) { - .ctl_name = CTL_UNNUMBERED, .procname = "exception-trace", .data = &show_unhandled_signals, .maxlen = sizeof(int), @@ -1445,11 +1428,11 @@ static struct ctl_table debug_table[] = { .proc_handler = proc_dointvec }, #endif - { .ctl_name = 0 } + { } }; static struct ctl_table dev_table[] = { - { .ctl_name = 0 } + { } }; static DEFINE_SPINLOCK(sysctl_lock); @@ -1603,122 +1586,6 @@ void register_sysctl_root(struct ctl_table_root *root) spin_unlock(&sysctl_lock); } -#ifdef CONFIG_SYSCTL_SYSCALL -/* Perform the actual read/write of a sysctl table entry. */ -static int do_sysctl_strategy(struct ctl_table_root *root, - struct ctl_table *table, - void __user *oldval, size_t __user *oldlenp, - void __user *newval, size_t newlen) -{ - int op = 0, rc; - - if (oldval) - op |= MAY_READ; - if (newval) - op |= MAY_WRITE; - if (sysctl_perm(root, table, op)) - return -EPERM; - - if (table->strategy) { - rc = table->strategy(table, oldval, oldlenp, newval, newlen); - if (rc < 0) - return rc; - if (rc > 0) - return 0; - } - - /* If there is no strategy routine, or if the strategy returns - * zero, proceed with automatic r/w */ - if (table->data && table->maxlen) { - rc = sysctl_data(table, oldval, oldlenp, newval, newlen); - if (rc < 0) - return rc; - } - return 0; -} - -static int parse_table(int __user *name, int nlen, - void __user *oldval, size_t __user *oldlenp, - void __user *newval, size_t newlen, - struct ctl_table_root *root, - struct ctl_table *table) -{ - int n; -repeat: - if (!nlen) - return -ENOTDIR; - if (get_user(n, name)) - return -EFAULT; - for ( ; table->ctl_name || table->procname; table++) { - if (!table->ctl_name) - continue; - if (n == table->ctl_name) { - int error; - if (table->child) { - if (sysctl_perm(root, table, MAY_EXEC)) - return -EPERM; - name++; - nlen--; - table = table->child; - goto repeat; - } - error = do_sysctl_strategy(root, table, - oldval, oldlenp, - newval, newlen); - return error; - } - } - return -ENOTDIR; -} - -int do_sysctl(int __user *name, int nlen, void __user *oldval, size_t __user *oldlenp, - void __user *newval, size_t newlen) -{ - struct ctl_table_header *head; - int error = -ENOTDIR; - - if (nlen <= 0 || nlen >= CTL_MAXNAME) - return -ENOTDIR; - if (oldval) { - int old_len; - if (!oldlenp || get_user(old_len, oldlenp)) - return -EFAULT; - } - - for (head = sysctl_head_next(NULL); head; - head = sysctl_head_next(head)) { - error = parse_table(name, nlen, oldval, oldlenp, - newval, newlen, - head->root, head->ctl_table); - if (error != -ENOTDIR) { - sysctl_head_finish(head); - break; - } - } - return error; -} - -SYSCALL_DEFINE1(sysctl, struct __sysctl_args __user *, args) -{ - struct __sysctl_args tmp; - int error; - - if (copy_from_user(&tmp, args, sizeof(tmp))) - return -EFAULT; - - error = deprecated_sysctl_warning(&tmp); - if (error) - goto out; - - lock_kernel(); - error = do_sysctl(tmp.name, tmp.nlen, tmp.oldval, tmp.oldlenp, - tmp.newval, tmp.newlen); - unlock_kernel(); -out: - return error; -} -#endif /* CONFIG_SYSCTL_SYSCALL */ - /* * sysctl_perm does NOT grant the superuser all rights automatically, because * some sysctl variables are readonly even to root. @@ -1754,7 +1621,7 @@ int sysctl_perm(struct ctl_table_root *root, struct ctl_table *table, int op) static void sysctl_set_parent(struct ctl_table *parent, struct ctl_table *table) { - for (; table->ctl_name || table->procname; table++) { + for (; table->procname; table++) { table->parent = parent; if (table->child) sysctl_set_parent(table, table->child); @@ -1786,11 +1653,11 @@ static struct ctl_table *is_branch_in(struct ctl_table *branch, return NULL; /* ... and nothing else */ - if (branch[1].procname || branch[1].ctl_name) + if (branch[1].procname) return NULL; /* table should contain subdirectory with the same name */ - for (p = table; p->procname || p->ctl_name; p++) { + for (p = table; p->procname; p++) { if (!p->child) continue; if (p->procname && strcmp(p->procname, s) == 0) @@ -1835,9 +1702,6 @@ static void try_attach(struct ctl_table_header *p, struct ctl_table_header *q) * * The members of the &struct ctl_table structure are used as follows: * - * ctl_name - This is the numeric sysctl value used by sysctl(2). The number - * must be unique within that level of sysctl - * * procname - the name of the sysctl file under /proc/sys. Set to %NULL to not * enter a sysctl file * @@ -1852,8 +1716,6 @@ static void try_attach(struct ctl_table_header *p, struct ctl_table_header *q) * * proc_handler - the text handler routine (described below) * - * strategy - the strategy routine (described below) - * * de - for internal use by the sysctl routines * * extra1, extra2 - extra pointers usable by the proc handler routines @@ -1866,19 +1728,6 @@ static void try_attach(struct ctl_table_header *p, struct ctl_table_header *q) * struct enable minimal validation of the values being written to be * performed, and the mode field allows minimal authentication. * - * More sophisticated management can be enabled by the provision of a - * strategy routine with the table entry. This will be called before - * any automatic read or write of the data is performed. - * - * The strategy routine may return - * - * < 0 - Error occurred (error is passed to user process) - * - * 0 - OK - proceed with automatic read or write. - * - * > 0 - OK - read or write has been done by the strategy routine, so - * return immediately. - * * There must be a proc_handler routine for any terminal nodes * mirrored under /proc/sys (non-terminals are handled by a built-in * directory handler). Several default handlers are available to @@ -1905,13 +1754,13 @@ struct ctl_table_header *__register_sysctl_paths( struct ctl_table_set *set; /* Count the path components */ - for (npath = 0; path[npath].ctl_name || path[npath].procname; ++npath) + for (npath = 0; path[npath].procname; ++npath) ; /* * For each path component, allocate a 2-element ctl_table array. * The first array element will be filled with the sysctl entry - * for this, the second will be the sentinel (ctl_name == 0). + * for this, the second will be the sentinel (procname == 0). * * We allocate everything in one go so that we don't have to * worry about freeing additional memory in unregister_sysctl_table. @@ -1928,7 +1777,6 @@ struct ctl_table_header *__register_sysctl_paths( for (n = 0; n < npath; ++n, ++path) { /* Copy the procname */ new->procname = path->procname; - new->ctl_name = path->ctl_name; new->mode = 0555; *prevp = new; @@ -2088,7 +1936,7 @@ void sysctl_head_put(struct ctl_table_header *head) #ifdef CONFIG_PROC_SYSCTL static int _proc_do_string(void* data, int maxlen, int write, - struct file *filp, void __user *buffer, + void __user *buffer, size_t *lenp, loff_t *ppos) { size_t len; @@ -2149,7 +1997,6 @@ static int _proc_do_string(void* data, int maxlen, int write, * proc_dostring - read a string sysctl * @table: the sysctl table * @write: %TRUE if this is a write to the sysctl file - * @filp: the file structure * @buffer: the user buffer * @lenp: the size of the user buffer * @ppos: file position @@ -2163,10 +2010,10 @@ static int _proc_do_string(void* data, int maxlen, int write, * * Returns 0 on success. */ -int proc_dostring(struct ctl_table *table, int write, struct file *filp, +int proc_dostring(struct ctl_table *table, int write, void __user *buffer, size_t *lenp, loff_t *ppos) { - return _proc_do_string(table->data, table->maxlen, write, filp, + return _proc_do_string(table->data, table->maxlen, write, buffer, lenp, ppos); } @@ -2191,14 +2038,14 @@ static int do_proc_dointvec_conv(int *negp, unsigned long *lvalp, } static int __do_proc_dointvec(void *tbl_data, struct ctl_table *table, - int write, struct file *filp, void __user *buffer, + int write, void __user *buffer, size_t *lenp, loff_t *ppos, int (*conv)(int *negp, unsigned long *lvalp, int *valp, int write, void *data), void *data) { #define TMPBUFLEN 21 - int *i, vleft, first=1, neg, val; + int *i, vleft, first = 1, neg; unsigned long lval; size_t left, len; @@ -2251,8 +2098,6 @@ static int __do_proc_dointvec(void *tbl_data, struct ctl_table *table, len = p-buf; if ((len < left) && *p && !isspace(*p)) break; - if (neg) - val = -val; s += len; left -= len; @@ -2300,13 +2145,13 @@ static int __do_proc_dointvec(void *tbl_data, struct ctl_table *table, #undef TMPBUFLEN } -static int do_proc_dointvec(struct ctl_table *table, int write, struct file *filp, +static int do_proc_dointvec(struct ctl_table *table, int write, void __user *buffer, size_t *lenp, loff_t *ppos, int (*conv)(int *negp, unsigned long *lvalp, int *valp, int write, void *data), void *data) { - return __do_proc_dointvec(table->data, table, write, filp, + return __do_proc_dointvec(table->data, table, write, buffer, lenp, ppos, conv, data); } @@ -2314,7 +2159,6 @@ static int do_proc_dointvec(struct ctl_table *table, int write, struct file *fil * proc_dointvec - read a vector of integers * @table: the sysctl table * @write: %TRUE if this is a write to the sysctl file - * @filp: the file structure * @buffer: the user buffer * @lenp: the size of the user buffer * @ppos: file position @@ -2324,10 +2168,10 @@ static int do_proc_dointvec(struct ctl_table *table, int write, struct file *fil * * Returns 0 on success. */ -int proc_dointvec(struct ctl_table *table, int write, struct file *filp, +int proc_dointvec(struct ctl_table *table, int write, void __user *buffer, size_t *lenp, loff_t *ppos) { - return do_proc_dointvec(table,write,filp,buffer,lenp,ppos, + return do_proc_dointvec(table,write,buffer,lenp,ppos, NULL,NULL); } @@ -2335,7 +2179,7 @@ int proc_dointvec(struct ctl_table *table, int write, struct file *filp, * Taint values can only be increased * This means we can safely use a temporary. */ -static int proc_taint(struct ctl_table *table, int write, struct file *filp, +static int proc_taint(struct ctl_table *table, int write, void __user *buffer, size_t *lenp, loff_t *ppos) { struct ctl_table t; @@ -2347,7 +2191,7 @@ static int proc_taint(struct ctl_table *table, int write, struct file *filp, t = *table; t.data = &tmptaint; - err = proc_doulongvec_minmax(&t, write, filp, buffer, lenp, ppos); + err = proc_doulongvec_minmax(&t, write, buffer, lenp, ppos); if (err < 0) return err; @@ -2399,7 +2243,6 @@ static int do_proc_dointvec_minmax_conv(int *negp, unsigned long *lvalp, * proc_dointvec_minmax - read a vector of integers with min/max values * @table: the sysctl table * @write: %TRUE if this is a write to the sysctl file - * @filp: the file structure * @buffer: the user buffer * @lenp: the size of the user buffer * @ppos: file position @@ -2412,19 +2255,18 @@ static int do_proc_dointvec_minmax_conv(int *negp, unsigned long *lvalp, * * Returns 0 on success. */ -int proc_dointvec_minmax(struct ctl_table *table, int write, struct file *filp, +int proc_dointvec_minmax(struct ctl_table *table, int write, void __user *buffer, size_t *lenp, loff_t *ppos) { struct do_proc_dointvec_minmax_conv_param param = { .min = (int *) table->extra1, .max = (int *) table->extra2, }; - return do_proc_dointvec(table, write, filp, buffer, lenp, ppos, + return do_proc_dointvec(table, write, buffer, lenp, ppos, do_proc_dointvec_minmax_conv, ¶m); } static int __do_proc_doulongvec_minmax(void *data, struct ctl_table *table, int write, - struct file *filp, void __user *buffer, size_t *lenp, loff_t *ppos, unsigned long convmul, @@ -2529,21 +2371,19 @@ static int __do_proc_doulongvec_minmax(void *data, struct ctl_table *table, int } static int do_proc_doulongvec_minmax(struct ctl_table *table, int write, - struct file *filp, void __user *buffer, size_t *lenp, loff_t *ppos, unsigned long convmul, unsigned long convdiv) { return __do_proc_doulongvec_minmax(table->data, table, write, - filp, buffer, lenp, ppos, convmul, convdiv); + buffer, lenp, ppos, convmul, convdiv); } /** * proc_doulongvec_minmax - read a vector of long integers with min/max values * @table: the sysctl table * @write: %TRUE if this is a write to the sysctl file - * @filp: the file structure * @buffer: the user buffer * @lenp: the size of the user buffer * @ppos: file position @@ -2556,17 +2396,16 @@ static int do_proc_doulongvec_minmax(struct ctl_table *table, int write, * * Returns 0 on success. */ -int proc_doulongvec_minmax(struct ctl_table *table, int write, struct file *filp, +int proc_doulongvec_minmax(struct ctl_table *table, int write, void __user *buffer, size_t *lenp, loff_t *ppos) { - return do_proc_doulongvec_minmax(table, write, filp, buffer, lenp, ppos, 1l, 1l); + return do_proc_doulongvec_minmax(table, write, buffer, lenp, ppos, 1l, 1l); } /** * proc_doulongvec_ms_jiffies_minmax - read a vector of millisecond values with min/max values * @table: the sysctl table * @write: %TRUE if this is a write to the sysctl file - * @filp: the file structure * @buffer: the user buffer * @lenp: the size of the user buffer * @ppos: file position @@ -2581,11 +2420,10 @@ int proc_doulongvec_minmax(struct ctl_table *table, int write, struct file *filp * Returns 0 on success. */ int proc_doulongvec_ms_jiffies_minmax(struct ctl_table *table, int write, - struct file *filp, void __user *buffer, size_t *lenp, loff_t *ppos) { - return do_proc_doulongvec_minmax(table, write, filp, buffer, + return do_proc_doulongvec_minmax(table, write, buffer, lenp, ppos, HZ, 1000l); } @@ -2661,7 +2499,6 @@ static int do_proc_dointvec_ms_jiffies_conv(int *negp, unsigned long *lvalp, * proc_dointvec_jiffies - read a vector of integers as seconds * @table: the sysctl table * @write: %TRUE if this is a write to the sysctl file - * @filp: the file structure * @buffer: the user buffer * @lenp: the size of the user buffer * @ppos: file position @@ -2673,10 +2510,10 @@ static int do_proc_dointvec_ms_jiffies_conv(int *negp, unsigned long *lvalp, * * Returns 0 on success. */ -int proc_dointvec_jiffies(struct ctl_table *table, int write, struct file *filp, +int proc_dointvec_jiffies(struct ctl_table *table, int write, void __user *buffer, size_t *lenp, loff_t *ppos) { - return do_proc_dointvec(table,write,filp,buffer,lenp,ppos, + return do_proc_dointvec(table,write,buffer,lenp,ppos, do_proc_dointvec_jiffies_conv,NULL); } @@ -2684,7 +2521,6 @@ int proc_dointvec_jiffies(struct ctl_table *table, int write, struct file *filp, * proc_dointvec_userhz_jiffies - read a vector of integers as 1/USER_HZ seconds * @table: the sysctl table * @write: %TRUE if this is a write to the sysctl file - * @filp: the file structure * @buffer: the user buffer * @lenp: the size of the user buffer * @ppos: pointer to the file position @@ -2696,10 +2532,10 @@ int proc_dointvec_jiffies(struct ctl_table *table, int write, struct file *filp, * * Returns 0 on success. */ -int proc_dointvec_userhz_jiffies(struct ctl_table *table, int write, struct file *filp, +int proc_dointvec_userhz_jiffies(struct ctl_table *table, int write, void __user *buffer, size_t *lenp, loff_t *ppos) { - return do_proc_dointvec(table,write,filp,buffer,lenp,ppos, + return do_proc_dointvec(table,write,buffer,lenp,ppos, do_proc_dointvec_userhz_jiffies_conv,NULL); } @@ -2707,7 +2543,6 @@ int proc_dointvec_userhz_jiffies(struct ctl_table *table, int write, struct file * proc_dointvec_ms_jiffies - read a vector of integers as 1 milliseconds * @table: the sysctl table * @write: %TRUE if this is a write to the sysctl file - * @filp: the file structure * @buffer: the user buffer * @lenp: the size of the user buffer * @ppos: file position @@ -2720,14 +2555,14 @@ int proc_dointvec_userhz_jiffies(struct ctl_table *table, int write, struct file * * Returns 0 on success. */ -int proc_dointvec_ms_jiffies(struct ctl_table *table, int write, struct file *filp, +int proc_dointvec_ms_jiffies(struct ctl_table *table, int write, void __user *buffer, size_t *lenp, loff_t *ppos) { - return do_proc_dointvec(table, write, filp, buffer, lenp, ppos, + return do_proc_dointvec(table, write, buffer, lenp, ppos, do_proc_dointvec_ms_jiffies_conv, NULL); } -static int proc_do_cad_pid(struct ctl_table *table, int write, struct file *filp, +static int proc_do_cad_pid(struct ctl_table *table, int write, void __user *buffer, size_t *lenp, loff_t *ppos) { struct pid *new_pid; @@ -2736,7 +2571,7 @@ static int proc_do_cad_pid(struct ctl_table *table, int write, struct file *filp tmp = pid_vnr(cad_pid); - r = __do_proc_dointvec(&tmp, table, write, filp, buffer, + r = __do_proc_dointvec(&tmp, table, write, buffer, lenp, ppos, NULL, NULL); if (r || !write) return r; @@ -2751,50 +2586,49 @@ static int proc_do_cad_pid(struct ctl_table *table, int write, struct file *filp #else /* CONFIG_PROC_FS */ -int proc_dostring(struct ctl_table *table, int write, struct file *filp, +int proc_dostring(struct ctl_table *table, int write, void __user *buffer, size_t *lenp, loff_t *ppos) { return -ENOSYS; } -int proc_dointvec(struct ctl_table *table, int write, struct file *filp, +int proc_dointvec(struct ctl_table *table, int write, void __user *buffer, size_t *lenp, loff_t *ppos) { return -ENOSYS; } -int proc_dointvec_minmax(struct ctl_table *table, int write, struct file *filp, +int proc_dointvec_minmax(struct ctl_table *table, int write, void __user *buffer, size_t *lenp, loff_t *ppos) { return -ENOSYS; } -int proc_dointvec_jiffies(struct ctl_table *table, int write, struct file *filp, +int proc_dointvec_jiffies(struct ctl_table *table, int write, void __user *buffer, size_t *lenp, loff_t *ppos) { return -ENOSYS; } -int proc_dointvec_userhz_jiffies(struct ctl_table *table, int write, struct file *filp, +int proc_dointvec_userhz_jiffies(struct ctl_table *table, int write, void __user *buffer, size_t *lenp, loff_t *ppos) { return -ENOSYS; } -int proc_dointvec_ms_jiffies(struct ctl_table *table, int write, struct file *filp, +int proc_dointvec_ms_jiffies(struct ctl_table *table, int write, void __user *buffer, size_t *lenp, loff_t *ppos) { return -ENOSYS; } -int proc_doulongvec_minmax(struct ctl_table *table, int write, struct file *filp, +int proc_doulongvec_minmax(struct ctl_table *table, int write, void __user *buffer, size_t *lenp, loff_t *ppos) { return -ENOSYS; } int proc_doulongvec_ms_jiffies_minmax(struct ctl_table *table, int write, - struct file *filp, void __user *buffer, size_t *lenp, loff_t *ppos) { @@ -2804,286 +2638,6 @@ int proc_doulongvec_ms_jiffies_minmax(struct ctl_table *table, int write, #endif /* CONFIG_PROC_FS */ - -#ifdef CONFIG_SYSCTL_SYSCALL -/* - * General sysctl support routines - */ - -/* The generic sysctl data routine (used if no strategy routine supplied) */ -int sysctl_data(struct ctl_table *table, - void __user *oldval, size_t __user *oldlenp, - void __user *newval, size_t newlen) -{ - size_t len; - - /* Get out of I don't have a variable */ - if (!table->data || !table->maxlen) - return -ENOTDIR; - - if (oldval && oldlenp) { - if (get_user(len, oldlenp)) - return -EFAULT; - if (len) { - if (len > table->maxlen) - len = table->maxlen; - if (copy_to_user(oldval, table->data, len)) - return -EFAULT; - if (put_user(len, oldlenp)) - return -EFAULT; - } - } - - if (newval && newlen) { - if (newlen > table->maxlen) - newlen = table->maxlen; - - if (copy_from_user(table->data, newval, newlen)) - return -EFAULT; - } - return 1; -} - -/* The generic string strategy routine: */ -int sysctl_string(struct ctl_table *table, - void __user *oldval, size_t __user *oldlenp, - void __user *newval, size_t newlen) -{ - if (!table->data || !table->maxlen) - return -ENOTDIR; - - if (oldval && oldlenp) { - size_t bufsize; - if (get_user(bufsize, oldlenp)) - return -EFAULT; - if (bufsize) { - size_t len = strlen(table->data), copied; - - /* This shouldn't trigger for a well-formed sysctl */ - if (len > table->maxlen) - len = table->maxlen; - - /* Copy up to a max of bufsize-1 bytes of the string */ - copied = (len >= bufsize) ? bufsize - 1 : len; - - if (copy_to_user(oldval, table->data, copied) || - put_user(0, (char __user *)(oldval + copied))) - return -EFAULT; - if (put_user(len, oldlenp)) - return -EFAULT; - } - } - if (newval && newlen) { - size_t len = newlen; - if (len > table->maxlen) - len = table->maxlen; - if(copy_from_user(table->data, newval, len)) - return -EFAULT; - if (len == table->maxlen) - len--; - ((char *) table->data)[len] = 0; - } - return 1; -} - -/* - * This function makes sure that all of the integers in the vector - * are between the minimum and maximum values given in the arrays - * table->extra1 and table->extra2, respectively. - */ -int sysctl_intvec(struct ctl_table *table, - void __user *oldval, size_t __user *oldlenp, - void __user *newval, size_t newlen) -{ - - if (newval && newlen) { - int __user *vec = (int __user *) newval; - int *min = (int *) table->extra1; - int *max = (int *) table->extra2; - size_t length; - int i; - - if (newlen % sizeof(int) != 0) - return -EINVAL; - - if (!table->extra1 && !table->extra2) - return 0; - - if (newlen > table->maxlen) - newlen = table->maxlen; - length = newlen / sizeof(int); - - for (i = 0; i < length; i++) { - int value; - if (get_user(value, vec + i)) - return -EFAULT; - if (min && value < min[i]) - return -EINVAL; - if (max && value > max[i]) - return -EINVAL; - } - } - return 0; -} - -/* Strategy function to convert jiffies to seconds */ -int sysctl_jiffies(struct ctl_table *table, - void __user *oldval, size_t __user *oldlenp, - void __user *newval, size_t newlen) -{ - if (oldval && oldlenp) { - size_t olen; - - if (get_user(olen, oldlenp)) - return -EFAULT; - if (olen) { - int val; - - if (olen < sizeof(int)) - return -EINVAL; - - val = *(int *)(table->data) / HZ; - if (put_user(val, (int __user *)oldval)) - return -EFAULT; - if (put_user(sizeof(int), oldlenp)) - return -EFAULT; - } - } - if (newval && newlen) { - int new; - if (newlen != sizeof(int)) - return -EINVAL; - if (get_user(new, (int __user *)newval)) - return -EFAULT; - *(int *)(table->data) = new*HZ; - } - return 1; -} - -/* Strategy function to convert jiffies to seconds */ -int sysctl_ms_jiffies(struct ctl_table *table, - void __user *oldval, size_t __user *oldlenp, - void __user *newval, size_t newlen) -{ - if (oldval && oldlenp) { - size_t olen; - - if (get_user(olen, oldlenp)) - return -EFAULT; - if (olen) { - int val; - - if (olen < sizeof(int)) - return -EINVAL; - - val = jiffies_to_msecs(*(int *)(table->data)); - if (put_user(val, (int __user *)oldval)) - return -EFAULT; - if (put_user(sizeof(int), oldlenp)) - return -EFAULT; - } - } - if (newval && newlen) { - int new; - if (newlen != sizeof(int)) - return -EINVAL; - if (get_user(new, (int __user *)newval)) - return -EFAULT; - *(int *)(table->data) = msecs_to_jiffies(new); - } - return 1; -} - - - -#else /* CONFIG_SYSCTL_SYSCALL */ - - -SYSCALL_DEFINE1(sysctl, struct __sysctl_args __user *, args) -{ - struct __sysctl_args tmp; - int error; - - if (copy_from_user(&tmp, args, sizeof(tmp))) - return -EFAULT; - - error = deprecated_sysctl_warning(&tmp); - - /* If no error reading the parameters then just -ENOSYS ... */ - if (!error) - error = -ENOSYS; - - return error; -} - -int sysctl_data(struct ctl_table *table, - void __user *oldval, size_t __user *oldlenp, - void __user *newval, size_t newlen) -{ - return -ENOSYS; -} - -int sysctl_string(struct ctl_table *table, - void __user *oldval, size_t __user *oldlenp, - void __user *newval, size_t newlen) -{ - return -ENOSYS; -} - -int sysctl_intvec(struct ctl_table *table, - void __user *oldval, size_t __user *oldlenp, - void __user *newval, size_t newlen) -{ - return -ENOSYS; -} - -int sysctl_jiffies(struct ctl_table *table, - void __user *oldval, size_t __user *oldlenp, - void __user *newval, size_t newlen) -{ - return -ENOSYS; -} - -int sysctl_ms_jiffies(struct ctl_table *table, - void __user *oldval, size_t __user *oldlenp, - void __user *newval, size_t newlen) -{ - return -ENOSYS; -} - -#endif /* CONFIG_SYSCTL_SYSCALL */ - -static int deprecated_sysctl_warning(struct __sysctl_args *args) -{ - static int msg_count; - int name[CTL_MAXNAME]; - int i; - - /* Check args->nlen. */ - if (args->nlen < 0 || args->nlen > CTL_MAXNAME) - return -ENOTDIR; - - /* Read in the sysctl name for better debug message logging */ - for (i = 0; i < args->nlen; i++) - if (get_user(name[i], args->name + i)) - return -EFAULT; - - /* Ignore accesses to kernel.version */ - if ((args->nlen == 2) && (name[0] == CTL_KERN) && (name[1] == KERN_VERSION)) - return 0; - - if (msg_count < 5) { - msg_count++; - printk(KERN_INFO - "warning: process `%s' used the deprecated sysctl " - "system call with ", current->comm); - for (i = 0; i < args->nlen; i++) - printk("%d.", name[i]); - printk("\n"); - } - return 0; -} - /* * No sense putting this after each symbol definition, twice, * exception granted :-) @@ -3098,9 +2652,4 @@ EXPORT_SYMBOL(proc_doulongvec_minmax); EXPORT_SYMBOL(proc_doulongvec_ms_jiffies_minmax); EXPORT_SYMBOL(register_sysctl_table); EXPORT_SYMBOL(register_sysctl_paths); -EXPORT_SYMBOL(sysctl_intvec); -EXPORT_SYMBOL(sysctl_jiffies); -EXPORT_SYMBOL(sysctl_ms_jiffies); -EXPORT_SYMBOL(sysctl_string); -EXPORT_SYMBOL(sysctl_data); EXPORT_SYMBOL(unregister_sysctl_table); diff --git a/kernel/sysctl_binary.c b/kernel/sysctl_binary.c new file mode 100644 index 00000000000..b75dbf40f57 --- /dev/null +++ b/kernel/sysctl_binary.c @@ -0,0 +1,1507 @@ +#include <linux/stat.h> +#include <linux/sysctl.h> +#include "../fs/xfs/linux-2.6/xfs_sysctl.h" +#include <linux/sunrpc/debug.h> +#include <linux/string.h> +#include <net/ip_vs.h> +#include <linux/syscalls.h> +#include <linux/namei.h> +#include <linux/mount.h> +#include <linux/fs.h> +#include <linux/nsproxy.h> +#include <linux/pid_namespace.h> +#include <linux/file.h> +#include <linux/ctype.h> +#include <linux/netdevice.h> + +#ifdef CONFIG_SYSCTL_SYSCALL + +struct bin_table; +typedef ssize_t bin_convert_t(struct file *file, + void __user *oldval, size_t oldlen, void __user *newval, size_t newlen); + +static bin_convert_t bin_dir; +static bin_convert_t bin_string; +static bin_convert_t bin_intvec; +static bin_convert_t bin_ulongvec; +static bin_convert_t bin_uuid; +static bin_convert_t bin_dn_node_address; + +#define CTL_DIR bin_dir +#define CTL_STR bin_string +#define CTL_INT bin_intvec +#define CTL_ULONG bin_ulongvec +#define CTL_UUID bin_uuid +#define CTL_DNADR bin_dn_node_address + +#define BUFSZ 256 + +struct bin_table { + bin_convert_t *convert; + int ctl_name; + const char *procname; + const struct bin_table *child; +}; + +static const struct bin_table bin_random_table[] = { + { CTL_INT, RANDOM_POOLSIZE, "poolsize" }, + { CTL_INT, RANDOM_ENTROPY_COUNT, "entropy_avail" }, + { CTL_INT, RANDOM_READ_THRESH, "read_wakeup_threshold" }, + { CTL_INT, RANDOM_WRITE_THRESH, "write_wakeup_threshold" }, + { CTL_UUID, RANDOM_BOOT_ID, "boot_id" }, + { CTL_UUID, RANDOM_UUID, "uuid" }, + {} +}; + +static const struct bin_table bin_pty_table[] = { + { CTL_INT, PTY_MAX, "max" }, + { CTL_INT, PTY_NR, "nr" }, + {} +}; + +static const struct bin_table bin_kern_table[] = { + { CTL_STR, KERN_OSTYPE, "ostype" }, + { CTL_STR, KERN_OSRELEASE, "osrelease" }, + /* KERN_OSREV not used */ + { CTL_STR, KERN_VERSION, "version" }, + /* KERN_SECUREMASK not used */ + /* KERN_PROF not used */ + { CTL_STR, KERN_NODENAME, "hostname" }, + { CTL_STR, KERN_DOMAINNAME, "domainname" }, + + { CTL_INT, KERN_PANIC, "panic" }, + { CTL_INT, KERN_REALROOTDEV, "real-root-dev" }, + + { CTL_STR, KERN_SPARC_REBOOT, "reboot-cmd" }, + { CTL_INT, KERN_CTLALTDEL, "ctrl-alt-del" }, + { CTL_INT, KERN_PRINTK, "printk" }, + + /* KERN_NAMETRANS not used */ + /* KERN_PPC_HTABRECLAIM not used */ + /* KERN_PPC_ZEROPAGED not used */ + { CTL_INT, KERN_PPC_POWERSAVE_NAP, "powersave-nap" }, + + { CTL_STR, KERN_MODPROBE, "modprobe" }, + { CTL_INT, KERN_SG_BIG_BUFF, "sg-big-buff" }, + { CTL_INT, KERN_ACCT, "acct" }, + /* KERN_PPC_L2CR "l2cr" no longer used */ + + /* KERN_RTSIGNR not used */ + /* KERN_RTSIGMAX not used */ + + { CTL_ULONG, KERN_SHMMAX, "shmmax" }, + { CTL_INT, KERN_MSGMAX, "msgmax" }, + { CTL_INT, KERN_MSGMNB, "msgmnb" }, + /* KERN_MSGPOOL not used*/ + { CTL_INT, KERN_SYSRQ, "sysrq" }, + { CTL_INT, KERN_MAX_THREADS, "threads-max" }, + { CTL_DIR, KERN_RANDOM, "random", bin_random_table }, + { CTL_ULONG, KERN_SHMALL, "shmall" }, + { CTL_INT, KERN_MSGMNI, "msgmni" }, + { CTL_INT, KERN_SEM, "sem" }, + { CTL_INT, KERN_SPARC_STOP_A, "stop-a" }, + { CTL_INT, KERN_SHMMNI, "shmmni" }, + + { CTL_INT, KERN_OVERFLOWUID, "overflowuid" }, + { CTL_INT, KERN_OVERFLOWGID, "overflowgid" }, + + { CTL_STR, KERN_HOTPLUG, "hotplug", }, + { CTL_INT, KERN_IEEE_EMULATION_WARNINGS, "ieee_emulation_warnings" }, + + { CTL_INT, KERN_S390_USER_DEBUG_LOGGING, "userprocess_debug" }, + { CTL_INT, KERN_CORE_USES_PID, "core_uses_pid" }, + /* KERN_TAINTED "tainted" no longer used */ + { CTL_INT, KERN_CADPID, "cad_pid" }, + { CTL_INT, KERN_PIDMAX, "pid_max" }, + { CTL_STR, KERN_CORE_PATTERN, "core_pattern" }, + { CTL_INT, KERN_PANIC_ON_OOPS, "panic_on_oops" }, + { CTL_INT, KERN_HPPA_PWRSW, "soft-power" }, + { CTL_INT, KERN_HPPA_UNALIGNED, "unaligned-trap" }, + + { CTL_INT, KERN_PRINTK_RATELIMIT, "printk_ratelimit" }, + { CTL_INT, KERN_PRINTK_RATELIMIT_BURST, "printk_ratelimit_burst" }, + + { CTL_DIR, KERN_PTY, "pty", bin_pty_table }, + { CTL_INT, KERN_NGROUPS_MAX, "ngroups_max" }, + { CTL_INT, KERN_SPARC_SCONS_PWROFF, "scons-poweroff" }, + /* KERN_HZ_TIMER "hz_timer" no longer used */ + { CTL_INT, KERN_UNKNOWN_NMI_PANIC, "unknown_nmi_panic" }, + { CTL_INT, KERN_BOOTLOADER_TYPE, "bootloader_type" }, + { CTL_INT, KERN_RANDOMIZE, "randomize_va_space" }, + + { CTL_INT, KERN_SPIN_RETRY, "spin_retry" }, + /* KERN_ACPI_VIDEO_FLAGS "acpi_video_flags" no longer used */ + { CTL_INT, KERN_IA64_UNALIGNED, "ignore-unaligned-usertrap" }, + { CTL_INT, KERN_COMPAT_LOG, "compat-log" }, + { CTL_INT, KERN_MAX_LOCK_DEPTH, "max_lock_depth" }, + { CTL_INT, KERN_NMI_WATCHDOG, "nmi_watchdog" }, + { CTL_INT, KERN_PANIC_ON_NMI, "panic_on_unrecovered_nmi" }, + {} +}; + +static const struct bin_table bin_vm_table[] = { + { CTL_INT, VM_OVERCOMMIT_MEMORY, "overcommit_memory" }, + { CTL_INT, VM_PAGE_CLUSTER, "page-cluster" }, + { CTL_INT, VM_DIRTY_BACKGROUND, "dirty_background_ratio" }, + { CTL_INT, VM_DIRTY_RATIO, "dirty_ratio" }, + /* VM_DIRTY_WB_CS "dirty_writeback_centisecs" no longer used */ + /* VM_DIRTY_EXPIRE_CS "dirty_expire_centisecs" no longer used */ + { CTL_INT, VM_NR_PDFLUSH_THREADS, "nr_pdflush_threads" }, + { CTL_INT, VM_OVERCOMMIT_RATIO, "overcommit_ratio" }, + /* VM_PAGEBUF unused */ + /* VM_HUGETLB_PAGES "nr_hugepages" no longer used */ + { CTL_INT, VM_SWAPPINESS, "swappiness" }, + { CTL_INT, VM_LOWMEM_RESERVE_RATIO, "lowmem_reserve_ratio" }, + { CTL_INT, VM_MIN_FREE_KBYTES, "min_free_kbytes" }, + { CTL_INT, VM_MAX_MAP_COUNT, "max_map_count" }, + { CTL_INT, VM_LAPTOP_MODE, "laptop_mode" }, + { CTL_INT, VM_BLOCK_DUMP, "block_dump" }, + { CTL_INT, VM_HUGETLB_GROUP, "hugetlb_shm_group" }, + { CTL_INT, VM_VFS_CACHE_PRESSURE, "vfs_cache_pressure" }, + { CTL_INT, VM_LEGACY_VA_LAYOUT, "legacy_va_layout" }, + /* VM_SWAP_TOKEN_TIMEOUT unused */ + { CTL_INT, VM_DROP_PAGECACHE, "drop_caches" }, + { CTL_INT, VM_PERCPU_PAGELIST_FRACTION, "percpu_pagelist_fraction" }, + { CTL_INT, VM_ZONE_RECLAIM_MODE, "zone_reclaim_mode" }, + { CTL_INT, VM_MIN_UNMAPPED, "min_unmapped_ratio" }, + { CTL_INT, VM_PANIC_ON_OOM, "panic_on_oom" }, + { CTL_INT, VM_VDSO_ENABLED, "vdso_enabled" }, + { CTL_INT, VM_MIN_SLAB, "min_slab_ratio" }, + + {} +}; + +static const struct bin_table bin_net_core_table[] = { + { CTL_INT, NET_CORE_WMEM_MAX, "wmem_max" }, + { CTL_INT, NET_CORE_RMEM_MAX, "rmem_max" }, + { CTL_INT, NET_CORE_WMEM_DEFAULT, "wmem_default" }, + { CTL_INT, NET_CORE_RMEM_DEFAULT, "rmem_default" }, + /* NET_CORE_DESTROY_DELAY unused */ + { CTL_INT, NET_CORE_MAX_BACKLOG, "netdev_max_backlog" }, + /* NET_CORE_FASTROUTE unused */ + { CTL_INT, NET_CORE_MSG_COST, "message_cost" }, + { CTL_INT, NET_CORE_MSG_BURST, "message_burst" }, + { CTL_INT, NET_CORE_OPTMEM_MAX, "optmem_max" }, + /* NET_CORE_HOT_LIST_LENGTH unused */ + /* NET_CORE_DIVERT_VERSION unused */ + /* NET_CORE_NO_CONG_THRESH unused */ + /* NET_CORE_NO_CONG unused */ + /* NET_CORE_LO_CONG unused */ + /* NET_CORE_MOD_CONG unused */ + { CTL_INT, NET_CORE_DEV_WEIGHT, "dev_weight" }, + { CTL_INT, NET_CORE_SOMAXCONN, "somaxconn" }, + { CTL_INT, NET_CORE_BUDGET, "netdev_budget" }, + { CTL_INT, NET_CORE_AEVENT_ETIME, "xfrm_aevent_etime" }, + { CTL_INT, NET_CORE_AEVENT_RSEQTH, "xfrm_aevent_rseqth" }, + { CTL_INT, NET_CORE_WARNINGS, "warnings" }, + {}, +}; + +static const struct bin_table bin_net_unix_table[] = { + /* NET_UNIX_DESTROY_DELAY unused */ + /* NET_UNIX_DELETE_DELAY unused */ + { CTL_INT, NET_UNIX_MAX_DGRAM_QLEN, "max_dgram_qlen" }, + {} +}; + +static const struct bin_table bin_net_ipv4_route_table[] = { + { CTL_INT, NET_IPV4_ROUTE_FLUSH, "flush" }, + /* NET_IPV4_ROUTE_MIN_DELAY "min_delay" no longer used */ + /* NET_IPV4_ROUTE_MAX_DELAY "max_delay" no longer used */ + { CTL_INT, NET_IPV4_ROUTE_GC_THRESH, "gc_thresh" }, + { CTL_INT, NET_IPV4_ROUTE_MAX_SIZE, "max_size" }, + { CTL_INT, NET_IPV4_ROUTE_GC_MIN_INTERVAL, "gc_min_interval" }, + { CTL_INT, NET_IPV4_ROUTE_GC_MIN_INTERVAL_MS, "gc_min_interval_ms" }, + { CTL_INT, NET_IPV4_ROUTE_GC_TIMEOUT, "gc_timeout" }, + { CTL_INT, NET_IPV4_ROUTE_GC_INTERVAL, "gc_interval" }, + { CTL_INT, NET_IPV4_ROUTE_REDIRECT_LOAD, "redirect_load" }, + { CTL_INT, NET_IPV4_ROUTE_REDIRECT_NUMBER, "redirect_number" }, + { CTL_INT, NET_IPV4_ROUTE_REDIRECT_SILENCE, "redirect_silence" }, + { CTL_INT, NET_IPV4_ROUTE_ERROR_COST, "error_cost" }, + { CTL_INT, NET_IPV4_ROUTE_ERROR_BURST, "error_burst" }, + { CTL_INT, NET_IPV4_ROUTE_GC_ELASTICITY, "gc_elasticity" }, + { CTL_INT, NET_IPV4_ROUTE_MTU_EXPIRES, "mtu_expires" }, + { CTL_INT, NET_IPV4_ROUTE_MIN_PMTU, "min_pmtu" }, + { CTL_INT, NET_IPV4_ROUTE_MIN_ADVMSS, "min_adv_mss" }, + { CTL_INT, NET_IPV4_ROUTE_SECRET_INTERVAL, "secret_interval" }, + {} +}; + +static const struct bin_table bin_net_ipv4_conf_vars_table[] = { + { CTL_INT, NET_IPV4_CONF_FORWARDING, "forwarding" }, + { CTL_INT, NET_IPV4_CONF_MC_FORWARDING, "mc_forwarding" }, + + { CTL_INT, NET_IPV4_CONF_ACCEPT_REDIRECTS, "accept_redirects" }, + { CTL_INT, NET_IPV4_CONF_SECURE_REDIRECTS, "secure_redirects" }, + { CTL_INT, NET_IPV4_CONF_SEND_REDIRECTS, "send_redirects" }, + { CTL_INT, NET_IPV4_CONF_SHARED_MEDIA, "shared_media" }, + { CTL_INT, NET_IPV4_CONF_RP_FILTER, "rp_filter" }, + { CTL_INT, NET_IPV4_CONF_ACCEPT_SOURCE_ROUTE, "accept_source_route" }, + { CTL_INT, NET_IPV4_CONF_PROXY_ARP, "proxy_arp" }, + { CTL_INT, NET_IPV4_CONF_MEDIUM_ID, "medium_id" }, + { CTL_INT, NET_IPV4_CONF_BOOTP_RELAY, "bootp_relay" }, + { CTL_INT, NET_IPV4_CONF_LOG_MARTIANS, "log_martians" }, + { CTL_INT, NET_IPV4_CONF_TAG, "tag" }, + { CTL_INT, NET_IPV4_CONF_ARPFILTER, "arp_filter" }, + { CTL_INT, NET_IPV4_CONF_ARP_ANNOUNCE, "arp_announce" }, + { CTL_INT, NET_IPV4_CONF_ARP_IGNORE, "arp_ignore" }, + { CTL_INT, NET_IPV4_CONF_ARP_ACCEPT, "arp_accept" }, + { CTL_INT, NET_IPV4_CONF_ARP_NOTIFY, "arp_notify" }, + + { CTL_INT, NET_IPV4_CONF_NOXFRM, "disable_xfrm" }, + { CTL_INT, NET_IPV4_CONF_NOPOLICY, "disable_policy" }, + { CTL_INT, NET_IPV4_CONF_FORCE_IGMP_VERSION, "force_igmp_version" }, + { CTL_INT, NET_IPV4_CONF_PROMOTE_SECONDARIES, "promote_secondaries" }, + {} +}; + +static const struct bin_table bin_net_ipv4_conf_table[] = { + { CTL_DIR, NET_PROTO_CONF_ALL, "all", bin_net_ipv4_conf_vars_table }, + { CTL_DIR, NET_PROTO_CONF_DEFAULT, "default", bin_net_ipv4_conf_vars_table }, + { CTL_DIR, 0, NULL, bin_net_ipv4_conf_vars_table }, + {} +}; + +static const struct bin_table bin_net_neigh_vars_table[] = { + { CTL_INT, NET_NEIGH_MCAST_SOLICIT, "mcast_solicit" }, + { CTL_INT, NET_NEIGH_UCAST_SOLICIT, "ucast_solicit" }, + { CTL_INT, NET_NEIGH_APP_SOLICIT, "app_solicit" }, + /* NET_NEIGH_RETRANS_TIME "retrans_time" no longer used */ + { CTL_INT, NET_NEIGH_REACHABLE_TIME, "base_reachable_time" }, + { CTL_INT, NET_NEIGH_DELAY_PROBE_TIME, "delay_first_probe_time" }, + { CTL_INT, NET_NEIGH_GC_STALE_TIME, "gc_stale_time" }, + { CTL_INT, NET_NEIGH_UNRES_QLEN, "unres_qlen" }, + { CTL_INT, NET_NEIGH_PROXY_QLEN, "proxy_qlen" }, + /* NET_NEIGH_ANYCAST_DELAY "anycast_delay" no longer used */ + /* NET_NEIGH_PROXY_DELAY "proxy_delay" no longer used */ + /* NET_NEIGH_LOCKTIME "locktime" no longer used */ + { CTL_INT, NET_NEIGH_GC_INTERVAL, "gc_interval" }, + { CTL_INT, NET_NEIGH_GC_THRESH1, "gc_thresh1" }, + { CTL_INT, NET_NEIGH_GC_THRESH2, "gc_thresh2" }, + { CTL_INT, NET_NEIGH_GC_THRESH3, "gc_thresh3" }, + { CTL_INT, NET_NEIGH_RETRANS_TIME_MS, "retrans_time_ms" }, + { CTL_INT, NET_NEIGH_REACHABLE_TIME_MS, "base_reachable_time_ms" }, + {} +}; + +static const struct bin_table bin_net_neigh_table[] = { + { CTL_DIR, NET_PROTO_CONF_DEFAULT, "default", bin_net_neigh_vars_table }, + { CTL_DIR, 0, NULL, bin_net_neigh_vars_table }, + {} +}; + +static const struct bin_table bin_net_ipv4_netfilter_table[] = { + { CTL_INT, NET_IPV4_NF_CONNTRACK_MAX, "ip_conntrack_max" }, + + /* NET_IPV4_NF_CONNTRACK_TCP_TIMEOUT_SYN_SENT "ip_conntrack_tcp_timeout_syn_sent" no longer used */ + /* NET_IPV4_NF_CONNTRACK_TCP_TIMEOUT_SYN_RECV "ip_conntrack_tcp_timeout_syn_recv" no longer used */ + /* NET_IPV4_NF_CONNTRACK_TCP_TIMEOUT_ESTABLISHED "ip_conntrack_tcp_timeout_established" no longer used */ + /* NET_IPV4_NF_CONNTRACK_TCP_TIMEOUT_FIN_WAIT "ip_conntrack_tcp_timeout_fin_wait" no longer used */ + /* NET_IPV4_NF_CONNTRACK_TCP_TIMEOUT_CLOSE_WAIT "ip_conntrack_tcp_timeout_close_wait" no longer used */ + /* NET_IPV4_NF_CONNTRACK_TCP_TIMEOUT_LAST_ACK "ip_conntrack_tcp_timeout_last_ack" no longer used */ + /* NET_IPV4_NF_CONNTRACK_TCP_TIMEOUT_TIME_WAIT "ip_conntrack_tcp_timeout_time_wait" no longer used */ + /* NET_IPV4_NF_CONNTRACK_TCP_TIMEOUT_CLOSE "ip_conntrack_tcp_timeout_close" no longer used */ + + /* NET_IPV4_NF_CONNTRACK_UDP_TIMEOUT "ip_conntrack_udp_timeout" no longer used */ + /* NET_IPV4_NF_CONNTRACK_UDP_TIMEOUT_STREAM "ip_conntrack_udp_timeout_stream" no longer used */ + /* NET_IPV4_NF_CONNTRACK_ICMP_TIMEOUT "ip_conntrack_icmp_timeout" no longer used */ + /* NET_IPV4_NF_CONNTRACK_GENERIC_TIMEOUT "ip_conntrack_generic_timeout" no longer used */ + + { CTL_INT, NET_IPV4_NF_CONNTRACK_BUCKETS, "ip_conntrack_buckets" }, + { CTL_INT, NET_IPV4_NF_CONNTRACK_LOG_INVALID, "ip_conntrack_log_invalid" }, + /* NET_IPV4_NF_CONNTRACK_TCP_TIMEOUT_MAX_RETRANS "ip_conntrack_tcp_timeout_max_retrans" no longer used */ + { CTL_INT, NET_IPV4_NF_CONNTRACK_TCP_LOOSE, "ip_conntrack_tcp_loose" }, + { CTL_INT, NET_IPV4_NF_CONNTRACK_TCP_BE_LIBERAL, "ip_conntrack_tcp_be_liberal" }, + { CTL_INT, NET_IPV4_NF_CONNTRACK_TCP_MAX_RETRANS, "ip_conntrack_tcp_max_retrans" }, + + /* NET_IPV4_NF_CONNTRACK_SCTP_TIMEOUT_CLOSED "ip_conntrack_sctp_timeout_closed" no longer used */ + /* NET_IPV4_NF_CONNTRACK_SCTP_TIMEOUT_COOKIE_WAIT "ip_conntrack_sctp_timeout_cookie_wait" no longer used */ + /* NET_IPV4_NF_CONNTRACK_SCTP_TIMEOUT_COOKIE_ECHOED "ip_conntrack_sctp_timeout_cookie_echoed" no longer used */ + /* NET_IPV4_NF_CONNTRACK_SCTP_TIMEOUT_ESTABLISHED "ip_conntrack_sctp_timeout_established" no longer used */ + /* NET_IPV4_NF_CONNTRACK_SCTP_TIMEOUT_SHUTDOWN_SENT "ip_conntrack_sctp_timeout_shutdown_sent" no longer used */ + /* NET_IPV4_NF_CONNTRACK_SCTP_TIMEOUT_SHUTDOWN_RECD "ip_conntrack_sctp_timeout_shutdown_recd" no longer used */ + /* NET_IPV4_NF_CONNTRACK_SCTP_TIMEOUT_SHUTDOWN_ACK_SENT "ip_conntrack_sctp_timeout_shutdown_ack_sent" no longer used */ + + { CTL_INT, NET_IPV4_NF_CONNTRACK_COUNT, "ip_conntrack_count" }, + { CTL_INT, NET_IPV4_NF_CONNTRACK_CHECKSUM, "ip_conntrack_checksum" }, + {} +}; + +static const struct bin_table bin_net_ipv4_table[] = { + {CTL_INT, NET_IPV4_FORWARD, "ip_forward" }, + + { CTL_DIR, NET_IPV4_CONF, "conf", bin_net_ipv4_conf_table }, + { CTL_DIR, NET_IPV4_NEIGH, "neigh", bin_net_neigh_table }, + { CTL_DIR, NET_IPV4_ROUTE, "route", bin_net_ipv4_route_table }, + /* NET_IPV4_FIB_HASH unused */ + { CTL_DIR, NET_IPV4_NETFILTER, "netfilter", bin_net_ipv4_netfilter_table }, + + { CTL_INT, NET_IPV4_TCP_TIMESTAMPS, "tcp_timestamps" }, + { CTL_INT, NET_IPV4_TCP_WINDOW_SCALING, "tcp_window_scaling" }, + { CTL_INT, NET_IPV4_TCP_SACK, "tcp_sack" }, + { CTL_INT, NET_IPV4_TCP_RETRANS_COLLAPSE, "tcp_retrans_collapse" }, + { CTL_INT, NET_IPV4_DEFAULT_TTL, "ip_default_ttl" }, + /* NET_IPV4_AUTOCONFIG unused */ + { CTL_INT, NET_IPV4_NO_PMTU_DISC, "ip_no_pmtu_disc" }, + { CTL_INT, NET_IPV4_NONLOCAL_BIND, "ip_nonlocal_bind" }, + { CTL_INT, NET_IPV4_TCP_SYN_RETRIES, "tcp_syn_retries" }, + { CTL_INT, NET_TCP_SYNACK_RETRIES, "tcp_synack_retries" }, + { CTL_INT, NET_TCP_MAX_ORPHANS, "tcp_max_orphans" }, + { CTL_INT, NET_TCP_MAX_TW_BUCKETS, "tcp_max_tw_buckets" }, + { CTL_INT, NET_IPV4_DYNADDR, "ip_dynaddr" }, + { CTL_INT, NET_IPV4_TCP_KEEPALIVE_TIME, "tcp_keepalive_time" }, + { CTL_INT, NET_IPV4_TCP_KEEPALIVE_PROBES, "tcp_keepalive_probes" }, + { CTL_INT, NET_IPV4_TCP_KEEPALIVE_INTVL, "tcp_keepalive_intvl" }, + { CTL_INT, NET_IPV4_TCP_RETRIES1, "tcp_retries1" }, + { CTL_INT, NET_IPV4_TCP_RETRIES2, "tcp_retries2" }, + { CTL_INT, NET_IPV4_TCP_FIN_TIMEOUT, "tcp_fin_timeout" }, + { CTL_INT, NET_TCP_SYNCOOKIES, "tcp_syncookies" }, + { CTL_INT, NET_TCP_TW_RECYCLE, "tcp_tw_recycle" }, + { CTL_INT, NET_TCP_ABORT_ON_OVERFLOW, "tcp_abort_on_overflow" }, + { CTL_INT, NET_TCP_STDURG, "tcp_stdurg" }, + { CTL_INT, NET_TCP_RFC1337, "tcp_rfc1337" }, + { CTL_INT, NET_TCP_MAX_SYN_BACKLOG, "tcp_max_syn_backlog" }, + { CTL_INT, NET_IPV4_LOCAL_PORT_RANGE, "ip_local_port_range" }, + { CTL_INT, NET_IPV4_IGMP_MAX_MEMBERSHIPS, "igmp_max_memberships" }, + { CTL_INT, NET_IPV4_IGMP_MAX_MSF, "igmp_max_msf" }, + { CTL_INT, NET_IPV4_INET_PEER_THRESHOLD, "inet_peer_threshold" }, + { CTL_INT, NET_IPV4_INET_PEER_MINTTL, "inet_peer_minttl" }, + { CTL_INT, NET_IPV4_INET_PEER_MAXTTL, "inet_peer_maxttl" }, + { CTL_INT, NET_IPV4_INET_PEER_GC_MINTIME, "inet_peer_gc_mintime" }, + { CTL_INT, NET_IPV4_INET_PEER_GC_MAXTIME, "inet_peer_gc_maxtime" }, + { CTL_INT, NET_TCP_ORPHAN_RETRIES, "tcp_orphan_retries" }, + { CTL_INT, NET_TCP_FACK, "tcp_fack" }, + { CTL_INT, NET_TCP_REORDERING, "tcp_reordering" }, + { CTL_INT, NET_TCP_ECN, "tcp_ecn" }, + { CTL_INT, NET_TCP_DSACK, "tcp_dsack" }, + { CTL_INT, NET_TCP_MEM, "tcp_mem" }, + { CTL_INT, NET_TCP_WMEM, "tcp_wmem" }, + { CTL_INT, NET_TCP_RMEM, "tcp_rmem" }, + { CTL_INT, NET_TCP_APP_WIN, "tcp_app_win" }, + { CTL_INT, NET_TCP_ADV_WIN_SCALE, "tcp_adv_win_scale" }, + { CTL_INT, NET_TCP_TW_REUSE, "tcp_tw_reuse" }, + { CTL_INT, NET_TCP_FRTO, "tcp_frto" }, + { CTL_INT, NET_TCP_FRTO_RESPONSE, "tcp_frto_response" }, + { CTL_INT, NET_TCP_LOW_LATENCY, "tcp_low_latency" }, + { CTL_INT, NET_TCP_NO_METRICS_SAVE, "tcp_no_metrics_save" }, + { CTL_INT, NET_TCP_MODERATE_RCVBUF, "tcp_moderate_rcvbuf" }, + { CTL_INT, NET_TCP_TSO_WIN_DIVISOR, "tcp_tso_win_divisor" }, + { CTL_STR, NET_TCP_CONG_CONTROL, "tcp_congestion_control" }, + { CTL_INT, NET_TCP_ABC, "tcp_abc" }, + { CTL_INT, NET_TCP_MTU_PROBING, "tcp_mtu_probing" }, + { CTL_INT, NET_TCP_BASE_MSS, "tcp_base_mss" }, + { CTL_INT, NET_IPV4_TCP_WORKAROUND_SIGNED_WINDOWS, "tcp_workaround_signed_windows" }, + { CTL_INT, NET_TCP_DMA_COPYBREAK, "tcp_dma_copybreak" }, + { CTL_INT, NET_TCP_SLOW_START_AFTER_IDLE, "tcp_slow_start_after_idle" }, + { CTL_INT, NET_CIPSOV4_CACHE_ENABLE, "cipso_cache_enable" }, + { CTL_INT, NET_CIPSOV4_CACHE_BUCKET_SIZE, "cipso_cache_bucket_size" }, + { CTL_INT, NET_CIPSOV4_RBM_OPTFMT, "cipso_rbm_optfmt" }, + { CTL_INT, NET_CIPSOV4_RBM_STRICTVALID, "cipso_rbm_strictvalid" }, + /* NET_TCP_AVAIL_CONG_CONTROL "tcp_available_congestion_control" no longer used */ + { CTL_STR, NET_TCP_ALLOWED_CONG_CONTROL, "tcp_allowed_congestion_control" }, + { CTL_INT, NET_TCP_MAX_SSTHRESH, "tcp_max_ssthresh" }, + + { CTL_INT, NET_IPV4_ICMP_ECHO_IGNORE_ALL, "icmp_echo_ignore_all" }, + { CTL_INT, NET_IPV4_ICMP_ECHO_IGNORE_BROADCASTS, "icmp_echo_ignore_broadcasts" }, + { CTL_INT, NET_IPV4_ICMP_IGNORE_BOGUS_ERROR_RESPONSES, "icmp_ignore_bogus_error_responses" }, + { CTL_INT, NET_IPV4_ICMP_ERRORS_USE_INBOUND_IFADDR, "icmp_errors_use_inbound_ifaddr" }, + { CTL_INT, NET_IPV4_ICMP_RATELIMIT, "icmp_ratelimit" }, + { CTL_INT, NET_IPV4_ICMP_RATEMASK, "icmp_ratemask" }, + + { CTL_INT, NET_IPV4_IPFRAG_HIGH_THRESH, "ipfrag_high_thresh" }, + { CTL_INT, NET_IPV4_IPFRAG_LOW_THRESH, "ipfrag_low_thresh" }, + { CTL_INT, NET_IPV4_IPFRAG_TIME, "ipfrag_time" }, + + { CTL_INT, NET_IPV4_IPFRAG_SECRET_INTERVAL, "ipfrag_secret_interval" }, + /* NET_IPV4_IPFRAG_MAX_DIST "ipfrag_max_dist" no longer used */ + + { CTL_INT, 2088 /* NET_IPQ_QMAX */, "ip_queue_maxlen" }, + + /* NET_TCP_DEFAULT_WIN_SCALE unused */ + /* NET_TCP_BIC_BETA unused */ + /* NET_IPV4_TCP_MAX_KA_PROBES unused */ + /* NET_IPV4_IP_MASQ_DEBUG unused */ + /* NET_TCP_SYN_TAILDROP unused */ + /* NET_IPV4_ICMP_SOURCEQUENCH_RATE unused */ + /* NET_IPV4_ICMP_DESTUNREACH_RATE unused */ + /* NET_IPV4_ICMP_TIMEEXCEED_RATE unused */ + /* NET_IPV4_ICMP_PARAMPROB_RATE unused */ + /* NET_IPV4_ICMP_ECHOREPLY_RATE unused */ + /* NET_IPV4_ALWAYS_DEFRAG unused */ + {} +}; + +static const struct bin_table bin_net_ipx_table[] = { + { CTL_INT, NET_IPX_PPROP_BROADCASTING, "ipx_pprop_broadcasting" }, + /* NET_IPX_FORWARDING unused */ + {} +}; + +static const struct bin_table bin_net_atalk_table[] = { + { CTL_INT, NET_ATALK_AARP_EXPIRY_TIME, "aarp-expiry-time" }, + { CTL_INT, NET_ATALK_AARP_TICK_TIME, "aarp-tick-time" }, + { CTL_INT, NET_ATALK_AARP_RETRANSMIT_LIMIT, "aarp-retransmit-limit" }, + { CTL_INT, NET_ATALK_AARP_RESOLVE_TIME, "aarp-resolve-time" }, + {}, +}; + +static const struct bin_table bin_net_netrom_table[] = { + { CTL_INT, NET_NETROM_DEFAULT_PATH_QUALITY, "default_path_quality" }, + { CTL_INT, NET_NETROM_OBSOLESCENCE_COUNT_INITIALISER, "obsolescence_count_initialiser" }, + { CTL_INT, NET_NETROM_NETWORK_TTL_INITIALISER, "network_ttl_initialiser" }, + { CTL_INT, NET_NETROM_TRANSPORT_TIMEOUT, "transport_timeout" }, + { CTL_INT, NET_NETROM_TRANSPORT_MAXIMUM_TRIES, "transport_maximum_tries" }, + { CTL_INT, NET_NETROM_TRANSPORT_ACKNOWLEDGE_DELAY, "transport_acknowledge_delay" }, + { CTL_INT, NET_NETROM_TRANSPORT_BUSY_DELAY, "transport_busy_delay" }, + { CTL_INT, NET_NETROM_TRANSPORT_REQUESTED_WINDOW_SIZE, "transport_requested_window_size" }, + { CTL_INT, NET_NETROM_TRANSPORT_NO_ACTIVITY_TIMEOUT, "transport_no_activity_timeout" }, + { CTL_INT, NET_NETROM_ROUTING_CONTROL, "routing_control" }, + { CTL_INT, NET_NETROM_LINK_FAILS_COUNT, "link_fails_count" }, + { CTL_INT, NET_NETROM_RESET, "reset" }, + {} +}; + +static const struct bin_table bin_net_ax25_param_table[] = { + { CTL_INT, NET_AX25_IP_DEFAULT_MODE, "ip_default_mode" }, + { CTL_INT, NET_AX25_DEFAULT_MODE, "ax25_default_mode" }, + { CTL_INT, NET_AX25_BACKOFF_TYPE, "backoff_type" }, + { CTL_INT, NET_AX25_CONNECT_MODE, "connect_mode" }, + { CTL_INT, NET_AX25_STANDARD_WINDOW, "standard_window_size" }, + { CTL_INT, NET_AX25_EXTENDED_WINDOW, "extended_window_size" }, + { CTL_INT, NET_AX25_T1_TIMEOUT, "t1_timeout" }, + { CTL_INT, NET_AX25_T2_TIMEOUT, "t2_timeout" }, + { CTL_INT, NET_AX25_T3_TIMEOUT, "t3_timeout" }, + { CTL_INT, NET_AX25_IDLE_TIMEOUT, "idle_timeout" }, + { CTL_INT, NET_AX25_N2, "maximum_retry_count" }, + { CTL_INT, NET_AX25_PACLEN, "maximum_packet_length" }, + { CTL_INT, NET_AX25_PROTOCOL, "protocol" }, + { CTL_INT, NET_AX25_DAMA_SLAVE_TIMEOUT, "dama_slave_timeout" }, + {} +}; + +static const struct bin_table bin_net_ax25_table[] = { + { CTL_DIR, 0, NULL, bin_net_ax25_param_table }, + {} +}; + +static const struct bin_table bin_net_rose_table[] = { + { CTL_INT, NET_ROSE_RESTART_REQUEST_TIMEOUT, "restart_request_timeout" }, + { CTL_INT, NET_ROSE_CALL_REQUEST_TIMEOUT, "call_request_timeout" }, + { CTL_INT, NET_ROSE_RESET_REQUEST_TIMEOUT, "reset_request_timeout" }, + { CTL_INT, NET_ROSE_CLEAR_REQUEST_TIMEOUT, "clear_request_timeout" }, + { CTL_INT, NET_ROSE_ACK_HOLD_BACK_TIMEOUT, "acknowledge_hold_back_timeout" }, + { CTL_INT, NET_ROSE_ROUTING_CONTROL, "routing_control" }, + { CTL_INT, NET_ROSE_LINK_FAIL_TIMEOUT, "link_fail_timeout" }, + { CTL_INT, NET_ROSE_MAX_VCS, "maximum_virtual_circuits" }, + { CTL_INT, NET_ROSE_WINDOW_SIZE, "window_size" }, + { CTL_INT, NET_ROSE_NO_ACTIVITY_TIMEOUT, "no_activity_timeout" }, + {} +}; + +static const struct bin_table bin_net_ipv6_conf_var_table[] = { + { CTL_INT, NET_IPV6_FORWARDING, "forwarding" }, + { CTL_INT, NET_IPV6_HOP_LIMIT, "hop_limit" }, + { CTL_INT, NET_IPV6_MTU, "mtu" }, + { CTL_INT, NET_IPV6_ACCEPT_RA, "accept_ra" }, + { CTL_INT, NET_IPV6_ACCEPT_REDIRECTS, "accept_redirects" }, + { CTL_INT, NET_IPV6_AUTOCONF, "autoconf" }, + { CTL_INT, NET_IPV6_DAD_TRANSMITS, "dad_transmits" }, + { CTL_INT, NET_IPV6_RTR_SOLICITS, "router_solicitations" }, + { CTL_INT, NET_IPV6_RTR_SOLICIT_INTERVAL, "router_solicitation_interval" }, + { CTL_INT, NET_IPV6_RTR_SOLICIT_DELAY, "router_solicitation_delay" }, + { CTL_INT, NET_IPV6_USE_TEMPADDR, "use_tempaddr" }, + { CTL_INT, NET_IPV6_TEMP_VALID_LFT, "temp_valid_lft" }, + { CTL_INT, NET_IPV6_TEMP_PREFERED_LFT, "temp_prefered_lft" }, + { CTL_INT, NET_IPV6_REGEN_MAX_RETRY, "regen_max_retry" }, + { CTL_INT, NET_IPV6_MAX_DESYNC_FACTOR, "max_desync_factor" }, + { CTL_INT, NET_IPV6_MAX_ADDRESSES, "max_addresses" }, + { CTL_INT, NET_IPV6_FORCE_MLD_VERSION, "force_mld_version" }, + { CTL_INT, NET_IPV6_ACCEPT_RA_DEFRTR, "accept_ra_defrtr" }, + { CTL_INT, NET_IPV6_ACCEPT_RA_PINFO, "accept_ra_pinfo" }, + { CTL_INT, NET_IPV6_ACCEPT_RA_RTR_PREF, "accept_ra_rtr_pref" }, + { CTL_INT, NET_IPV6_RTR_PROBE_INTERVAL, "router_probe_interval" }, + { CTL_INT, NET_IPV6_ACCEPT_RA_RT_INFO_MAX_PLEN, "accept_ra_rt_info_max_plen" }, + { CTL_INT, NET_IPV6_PROXY_NDP, "proxy_ndp" }, + { CTL_INT, NET_IPV6_ACCEPT_SOURCE_ROUTE, "accept_source_route" }, + {} +}; + +static const struct bin_table bin_net_ipv6_conf_table[] = { + { CTL_DIR, NET_PROTO_CONF_ALL, "all", bin_net_ipv6_conf_var_table }, + { CTL_DIR, NET_PROTO_CONF_DEFAULT, "default", bin_net_ipv6_conf_var_table }, + { CTL_DIR, 0, NULL, bin_net_ipv6_conf_var_table }, + {} +}; + +static const struct bin_table bin_net_ipv6_route_table[] = { + /* NET_IPV6_ROUTE_FLUSH "flush" no longer used */ + { CTL_INT, NET_IPV6_ROUTE_GC_THRESH, "gc_thresh" }, + { CTL_INT, NET_IPV6_ROUTE_MAX_SIZE, "max_size" }, + { CTL_INT, NET_IPV6_ROUTE_GC_MIN_INTERVAL, "gc_min_interval" }, + { CTL_INT, NET_IPV6_ROUTE_GC_TIMEOUT, "gc_timeout" }, + { CTL_INT, NET_IPV6_ROUTE_GC_INTERVAL, "gc_interval" }, + { CTL_INT, NET_IPV6_ROUTE_GC_ELASTICITY, "gc_elasticity" }, + { CTL_INT, NET_IPV6_ROUTE_MTU_EXPIRES, "mtu_expires" }, + { CTL_INT, NET_IPV6_ROUTE_MIN_ADVMSS, "min_adv_mss" }, + { CTL_INT, NET_IPV6_ROUTE_GC_MIN_INTERVAL_MS, "gc_min_interval_ms" }, + {} +}; + +static const struct bin_table bin_net_ipv6_icmp_table[] = { + { CTL_INT, NET_IPV6_ICMP_RATELIMIT, "ratelimit" }, + {} +}; + +static const struct bin_table bin_net_ipv6_table[] = { + { CTL_DIR, NET_IPV6_CONF, "conf", bin_net_ipv6_conf_table }, + { CTL_DIR, NET_IPV6_NEIGH, "neigh", bin_net_neigh_table }, + { CTL_DIR, NET_IPV6_ROUTE, "route", bin_net_ipv6_route_table }, + { CTL_DIR, NET_IPV6_ICMP, "icmp", bin_net_ipv6_icmp_table }, + { CTL_INT, NET_IPV6_BINDV6ONLY, "bindv6only" }, + { CTL_INT, NET_IPV6_IP6FRAG_HIGH_THRESH, "ip6frag_high_thresh" }, + { CTL_INT, NET_IPV6_IP6FRAG_LOW_THRESH, "ip6frag_low_thresh" }, + { CTL_INT, NET_IPV6_IP6FRAG_TIME, "ip6frag_time" }, + { CTL_INT, NET_IPV6_IP6FRAG_SECRET_INTERVAL, "ip6frag_secret_interval" }, + { CTL_INT, NET_IPV6_MLD_MAX_MSF, "mld_max_msf" }, + { CTL_INT, 2088 /* IPQ_QMAX */, "ip6_queue_maxlen" }, + {} +}; + +static const struct bin_table bin_net_x25_table[] = { + { CTL_INT, NET_X25_RESTART_REQUEST_TIMEOUT, "restart_request_timeout" }, + { CTL_INT, NET_X25_CALL_REQUEST_TIMEOUT, "call_request_timeout" }, + { CTL_INT, NET_X25_RESET_REQUEST_TIMEOUT, "reset_request_timeout" }, + { CTL_INT, NET_X25_CLEAR_REQUEST_TIMEOUT, "clear_request_timeout" }, + { CTL_INT, NET_X25_ACK_HOLD_BACK_TIMEOUT, "acknowledgement_hold_back_timeout" }, + { CTL_INT, NET_X25_FORWARD, "x25_forward" }, + {} +}; + +static const struct bin_table bin_net_tr_table[] = { + { CTL_INT, NET_TR_RIF_TIMEOUT, "rif_timeout" }, + {} +}; + + +static const struct bin_table bin_net_decnet_conf_vars[] = { + { CTL_INT, NET_DECNET_CONF_DEV_FORWARDING, "forwarding" }, + { CTL_INT, NET_DECNET_CONF_DEV_PRIORITY, "priority" }, + { CTL_INT, NET_DECNET_CONF_DEV_T2, "t2" }, + { CTL_INT, NET_DECNET_CONF_DEV_T3, "t3" }, + {} +}; + +static const struct bin_table bin_net_decnet_conf[] = { + { CTL_DIR, NET_DECNET_CONF_ETHER, "ethernet", bin_net_decnet_conf_vars }, + { CTL_DIR, NET_DECNET_CONF_GRE, "ipgre", bin_net_decnet_conf_vars }, + { CTL_DIR, NET_DECNET_CONF_X25, "x25", bin_net_decnet_conf_vars }, + { CTL_DIR, NET_DECNET_CONF_PPP, "ppp", bin_net_decnet_conf_vars }, + { CTL_DIR, NET_DECNET_CONF_DDCMP, "ddcmp", bin_net_decnet_conf_vars }, + { CTL_DIR, NET_DECNET_CONF_LOOPBACK, "loopback", bin_net_decnet_conf_vars }, + { CTL_DIR, 0, NULL, bin_net_decnet_conf_vars }, + {} +}; + +static const struct bin_table bin_net_decnet_table[] = { + { CTL_DIR, NET_DECNET_CONF, "conf", bin_net_decnet_conf }, + { CTL_DNADR, NET_DECNET_NODE_ADDRESS, "node_address" }, + { CTL_STR, NET_DECNET_NODE_NAME, "node_name" }, + { CTL_STR, NET_DECNET_DEFAULT_DEVICE, "default_device" }, + { CTL_INT, NET_DECNET_TIME_WAIT, "time_wait" }, + { CTL_INT, NET_DECNET_DN_COUNT, "dn_count" }, + { CTL_INT, NET_DECNET_DI_COUNT, "di_count" }, + { CTL_INT, NET_DECNET_DR_COUNT, "dr_count" }, + { CTL_INT, NET_DECNET_DST_GC_INTERVAL, "dst_gc_interval" }, + { CTL_INT, NET_DECNET_NO_FC_MAX_CWND, "no_fc_max_cwnd" }, + { CTL_INT, NET_DECNET_MEM, "decnet_mem" }, + { CTL_INT, NET_DECNET_RMEM, "decnet_rmem" }, + { CTL_INT, NET_DECNET_WMEM, "decnet_wmem" }, + { CTL_INT, NET_DECNET_DEBUG_LEVEL, "debug" }, + {} +}; + +static const struct bin_table bin_net_sctp_table[] = { + { CTL_INT, NET_SCTP_RTO_INITIAL, "rto_initial" }, + { CTL_INT, NET_SCTP_RTO_MIN, "rto_min" }, + { CTL_INT, NET_SCTP_RTO_MAX, "rto_max" }, + { CTL_INT, NET_SCTP_RTO_ALPHA, "rto_alpha_exp_divisor" }, + { CTL_INT, NET_SCTP_RTO_BETA, "rto_beta_exp_divisor" }, + { CTL_INT, NET_SCTP_VALID_COOKIE_LIFE, "valid_cookie_life" }, + { CTL_INT, NET_SCTP_ASSOCIATION_MAX_RETRANS, "association_max_retrans" }, + { CTL_INT, NET_SCTP_PATH_MAX_RETRANS, "path_max_retrans" }, + { CTL_INT, NET_SCTP_MAX_INIT_RETRANSMITS, "max_init_retransmits" }, + { CTL_INT, NET_SCTP_HB_INTERVAL, "hb_interval" }, + { CTL_INT, NET_SCTP_PRESERVE_ENABLE, "cookie_preserve_enable" }, + { CTL_INT, NET_SCTP_MAX_BURST, "max_burst" }, + { CTL_INT, NET_SCTP_ADDIP_ENABLE, "addip_enable" }, + { CTL_INT, NET_SCTP_PRSCTP_ENABLE, "prsctp_enable" }, + { CTL_INT, NET_SCTP_SNDBUF_POLICY, "sndbuf_policy" }, + { CTL_INT, NET_SCTP_SACK_TIMEOUT, "sack_timeout" }, + { CTL_INT, NET_SCTP_RCVBUF_POLICY, "rcvbuf_policy" }, + {} +}; + +static const struct bin_table bin_net_llc_llc2_timeout_table[] = { + { CTL_INT, NET_LLC2_ACK_TIMEOUT, "ack" }, + { CTL_INT, NET_LLC2_P_TIMEOUT, "p" }, + { CTL_INT, NET_LLC2_REJ_TIMEOUT, "rej" }, + { CTL_INT, NET_LLC2_BUSY_TIMEOUT, "busy" }, + {} +}; + +static const struct bin_table bin_net_llc_station_table[] = { + { CTL_INT, NET_LLC_STATION_ACK_TIMEOUT, "ack_timeout" }, + {} +}; + +static const struct bin_table bin_net_llc_llc2_table[] = { + { CTL_DIR, NET_LLC2, "timeout", bin_net_llc_llc2_timeout_table }, + {} +}; + +static const struct bin_table bin_net_llc_table[] = { + { CTL_DIR, NET_LLC2, "llc2", bin_net_llc_llc2_table }, + { CTL_DIR, NET_LLC_STATION, "station", bin_net_llc_station_table }, + {} +}; + +static const struct bin_table bin_net_netfilter_table[] = { + { CTL_INT, NET_NF_CONNTRACK_MAX, "nf_conntrack_max" }, + /* NET_NF_CONNTRACK_TCP_TIMEOUT_SYN_SENT "nf_conntrack_tcp_timeout_syn_sent" no longer used */ + /* NET_NF_CONNTRACK_TCP_TIMEOUT_SYN_RECV "nf_conntrack_tcp_timeout_syn_recv" no longer used */ + /* NET_NF_CONNTRACK_TCP_TIMEOUT_ESTABLISHED "nf_conntrack_tcp_timeout_established" no longer used */ + /* NET_NF_CONNTRACK_TCP_TIMEOUT_FIN_WAIT "nf_conntrack_tcp_timeout_fin_wait" no longer used */ + /* NET_NF_CONNTRACK_TCP_TIMEOUT_CLOSE_WAIT "nf_conntrack_tcp_timeout_close_wait" no longer used */ + /* NET_NF_CONNTRACK_TCP_TIMEOUT_LAST_ACK "nf_conntrack_tcp_timeout_last_ack" no longer used */ + /* NET_NF_CONNTRACK_TCP_TIMEOUT_TIME_WAIT "nf_conntrack_tcp_timeout_time_wait" no longer used */ + /* NET_NF_CONNTRACK_TCP_TIMEOUT_CLOSE "nf_conntrack_tcp_timeout_close" no longer used */ + /* NET_NF_CONNTRACK_UDP_TIMEOUT "nf_conntrack_udp_timeout" no longer used */ + /* NET_NF_CONNTRACK_UDP_TIMEOUT_STREAM "nf_conntrack_udp_timeout_stream" no longer used */ + /* NET_NF_CONNTRACK_ICMP_TIMEOUT "nf_conntrack_icmp_timeout" no longer used */ + /* NET_NF_CONNTRACK_GENERIC_TIMEOUT "nf_conntrack_generic_timeout" no longer used */ + { CTL_INT, NET_NF_CONNTRACK_BUCKETS, "nf_conntrack_buckets" }, + { CTL_INT, NET_NF_CONNTRACK_LOG_INVALID, "nf_conntrack_log_invalid" }, + /* NET_NF_CONNTRACK_TCP_TIMEOUT_MAX_RETRANS "nf_conntrack_tcp_timeout_max_retrans" no longer used */ + { CTL_INT, NET_NF_CONNTRACK_TCP_LOOSE, "nf_conntrack_tcp_loose" }, + { CTL_INT, NET_NF_CONNTRACK_TCP_BE_LIBERAL, "nf_conntrack_tcp_be_liberal" }, + { CTL_INT, NET_NF_CONNTRACK_TCP_MAX_RETRANS, "nf_conntrack_tcp_max_retrans" }, + /* NET_NF_CONNTRACK_SCTP_TIMEOUT_CLOSED "nf_conntrack_sctp_timeout_closed" no longer used */ + /* NET_NF_CONNTRACK_SCTP_TIMEOUT_COOKIE_WAIT "nf_conntrack_sctp_timeout_cookie_wait" no longer used */ + /* NET_NF_CONNTRACK_SCTP_TIMEOUT_COOKIE_ECHOED "nf_conntrack_sctp_timeout_cookie_echoed" no longer used */ + /* NET_NF_CONNTRACK_SCTP_TIMEOUT_ESTABLISHED "nf_conntrack_sctp_timeout_established" no longer used */ + /* NET_NF_CONNTRACK_SCTP_TIMEOUT_SHUTDOWN_SENT "nf_conntrack_sctp_timeout_shutdown_sent" no longer used */ + /* NET_NF_CONNTRACK_SCTP_TIMEOUT_SHUTDOWN_RECD "nf_conntrack_sctp_timeout_shutdown_recd" no longer used */ + /* NET_NF_CONNTRACK_SCTP_TIMEOUT_SHUTDOWN_ACK_SENT "nf_conntrack_sctp_timeout_shutdown_ack_sent" no longer used */ + { CTL_INT, NET_NF_CONNTRACK_COUNT, "nf_conntrack_count" }, + /* NET_NF_CONNTRACK_ICMPV6_TIMEOUT "nf_conntrack_icmpv6_timeout" no longer used */ + /* NET_NF_CONNTRACK_FRAG6_TIMEOUT "nf_conntrack_frag6_timeout" no longer used */ + { CTL_INT, NET_NF_CONNTRACK_FRAG6_LOW_THRESH, "nf_conntrack_frag6_low_thresh" }, + { CTL_INT, NET_NF_CONNTRACK_FRAG6_HIGH_THRESH, "nf_conntrack_frag6_high_thresh" }, + { CTL_INT, NET_NF_CONNTRACK_CHECKSUM, "nf_conntrack_checksum" }, + + {} +}; + +static const struct bin_table bin_net_irda_table[] = { + { CTL_INT, NET_IRDA_DISCOVERY, "discovery" }, + { CTL_STR, NET_IRDA_DEVNAME, "devname" }, + { CTL_INT, NET_IRDA_DEBUG, "debug" }, + { CTL_INT, NET_IRDA_FAST_POLL, "fast_poll_increase" }, + { CTL_INT, NET_IRDA_DISCOVERY_SLOTS, "discovery_slots" }, + { CTL_INT, NET_IRDA_DISCOVERY_TIMEOUT, "discovery_timeout" }, + { CTL_INT, NET_IRDA_SLOT_TIMEOUT, "slot_timeout" }, + { CTL_INT, NET_IRDA_MAX_BAUD_RATE, "max_baud_rate" }, + { CTL_INT, NET_IRDA_MIN_TX_TURN_TIME, "min_tx_turn_time" }, + { CTL_INT, NET_IRDA_MAX_TX_DATA_SIZE, "max_tx_data_size" }, + { CTL_INT, NET_IRDA_MAX_TX_WINDOW, "max_tx_window" }, + { CTL_INT, NET_IRDA_MAX_NOREPLY_TIME, "max_noreply_time" }, + { CTL_INT, NET_IRDA_WARN_NOREPLY_TIME, "warn_noreply_time" }, + { CTL_INT, NET_IRDA_LAP_KEEPALIVE_TIME, "lap_keepalive_time" }, + {} +}; + +static const struct bin_table bin_net_table[] = { + { CTL_DIR, NET_CORE, "core", bin_net_core_table }, + /* NET_ETHER not used */ + /* NET_802 not used */ + { CTL_DIR, NET_UNIX, "unix", bin_net_unix_table }, + { CTL_DIR, NET_IPV4, "ipv4", bin_net_ipv4_table }, + { CTL_DIR, NET_IPX, "ipx", bin_net_ipx_table }, + { CTL_DIR, NET_ATALK, "appletalk", bin_net_atalk_table }, + { CTL_DIR, NET_NETROM, "netrom", bin_net_netrom_table }, + { CTL_DIR, NET_AX25, "ax25", bin_net_ax25_table }, + /* NET_BRIDGE "bridge" no longer used */ + { CTL_DIR, NET_ROSE, "rose", bin_net_rose_table }, + { CTL_DIR, NET_IPV6, "ipv6", bin_net_ipv6_table }, + { CTL_DIR, NET_X25, "x25", bin_net_x25_table }, + { CTL_DIR, NET_TR, "token-ring", bin_net_tr_table }, + { CTL_DIR, NET_DECNET, "decnet", bin_net_decnet_table }, + /* NET_ECONET not used */ + { CTL_DIR, NET_SCTP, "sctp", bin_net_sctp_table }, + { CTL_DIR, NET_LLC, "llc", bin_net_llc_table }, + { CTL_DIR, NET_NETFILTER, "netfilter", bin_net_netfilter_table }, + /* NET_DCCP "dccp" no longer used */ + { CTL_DIR, NET_IRDA, "irda", bin_net_irda_table }, + { CTL_INT, 2089, "nf_conntrack_max" }, + {} +}; + +static const struct bin_table bin_fs_quota_table[] = { + { CTL_INT, FS_DQ_LOOKUPS, "lookups" }, + { CTL_INT, FS_DQ_DROPS, "drops" }, + { CTL_INT, FS_DQ_READS, "reads" }, + { CTL_INT, FS_DQ_WRITES, "writes" }, + { CTL_INT, FS_DQ_CACHE_HITS, "cache_hits" }, + { CTL_INT, FS_DQ_ALLOCATED, "allocated_dquots" }, + { CTL_INT, FS_DQ_FREE, "free_dquots" }, + { CTL_INT, FS_DQ_SYNCS, "syncs" }, + { CTL_INT, FS_DQ_WARNINGS, "warnings" }, + {} +}; + +static const struct bin_table bin_fs_xfs_table[] = { + { CTL_INT, XFS_SGID_INHERIT, "irix_sgid_inherit" }, + { CTL_INT, XFS_SYMLINK_MODE, "irix_symlink_mode" }, + { CTL_INT, XFS_PANIC_MASK, "panic_mask" }, + + { CTL_INT, XFS_ERRLEVEL, "error_level" }, + { CTL_INT, XFS_SYNCD_TIMER, "xfssyncd_centisecs" }, + { CTL_INT, XFS_INHERIT_SYNC, "inherit_sync" }, + { CTL_INT, XFS_INHERIT_NODUMP, "inherit_nodump" }, + { CTL_INT, XFS_INHERIT_NOATIME, "inherit_noatime" }, + { CTL_INT, XFS_BUF_TIMER, "xfsbufd_centisecs" }, + { CTL_INT, XFS_BUF_AGE, "age_buffer_centisecs" }, + { CTL_INT, XFS_INHERIT_NOSYM, "inherit_nosymlinks" }, + { CTL_INT, XFS_ROTORSTEP, "rotorstep" }, + { CTL_INT, XFS_INHERIT_NODFRG, "inherit_nodefrag" }, + { CTL_INT, XFS_FILESTREAM_TIMER, "filestream_centisecs" }, + { CTL_INT, XFS_STATS_CLEAR, "stats_clear" }, + {} +}; + +static const struct bin_table bin_fs_ocfs2_nm_table[] = { + { CTL_STR, 1, "hb_ctl_path" }, + {} +}; + +static const struct bin_table bin_fs_ocfs2_table[] = { + { CTL_DIR, 1, "nm", bin_fs_ocfs2_nm_table }, + {} +}; + +static const struct bin_table bin_inotify_table[] = { + { CTL_INT, INOTIFY_MAX_USER_INSTANCES, "max_user_instances" }, + { CTL_INT, INOTIFY_MAX_USER_WATCHES, "max_user_watches" }, + { CTL_INT, INOTIFY_MAX_QUEUED_EVENTS, "max_queued_events" }, + {} +}; + +static const struct bin_table bin_fs_table[] = { + { CTL_INT, FS_NRINODE, "inode-nr" }, + { CTL_INT, FS_STATINODE, "inode-state" }, + /* FS_MAXINODE unused */ + /* FS_NRDQUOT unused */ + /* FS_MAXDQUOT unused */ + /* FS_NRFILE "file-nr" no longer used */ + { CTL_INT, FS_MAXFILE, "file-max" }, + { CTL_INT, FS_DENTRY, "dentry-state" }, + /* FS_NRSUPER unused */ + /* FS_MAXUPSER unused */ + { CTL_INT, FS_OVERFLOWUID, "overflowuid" }, + { CTL_INT, FS_OVERFLOWGID, "overflowgid" }, + { CTL_INT, FS_LEASES, "leases-enable" }, + { CTL_INT, FS_DIR_NOTIFY, "dir-notify-enable" }, + { CTL_INT, FS_LEASE_TIME, "lease-break-time" }, + { CTL_DIR, FS_DQSTATS, "quota", bin_fs_quota_table }, + { CTL_DIR, FS_XFS, "xfs", bin_fs_xfs_table }, + { CTL_ULONG, FS_AIO_NR, "aio-nr" }, + { CTL_ULONG, FS_AIO_MAX_NR, "aio-max-nr" }, + { CTL_DIR, FS_INOTIFY, "inotify", bin_inotify_table }, + { CTL_DIR, FS_OCFS2, "ocfs2", bin_fs_ocfs2_table }, + { CTL_INT, KERN_SETUID_DUMPABLE, "suid_dumpable" }, + {} +}; + +static const struct bin_table bin_ipmi_table[] = { + { CTL_INT, DEV_IPMI_POWEROFF_POWERCYCLE, "poweroff_powercycle" }, + {} +}; + +static const struct bin_table bin_mac_hid_files[] = { + /* DEV_MAC_HID_KEYBOARD_SENDS_LINUX_KEYCODES unused */ + /* DEV_MAC_HID_KEYBOARD_LOCK_KEYCODES unused */ + { CTL_INT, DEV_MAC_HID_MOUSE_BUTTON_EMULATION, "mouse_button_emulation" }, + { CTL_INT, DEV_MAC_HID_MOUSE_BUTTON2_KEYCODE, "mouse_button2_keycode" }, + { CTL_INT, DEV_MAC_HID_MOUSE_BUTTON3_KEYCODE, "mouse_button3_keycode" }, + /* DEV_MAC_HID_ADB_MOUSE_SENDS_KEYCODES unused */ + {} +}; + +static const struct bin_table bin_raid_table[] = { + { CTL_INT, DEV_RAID_SPEED_LIMIT_MIN, "speed_limit_min" }, + { CTL_INT, DEV_RAID_SPEED_LIMIT_MAX, "speed_limit_max" }, + {} +}; + +static const struct bin_table bin_scsi_table[] = { + { CTL_INT, DEV_SCSI_LOGGING_LEVEL, "logging_level" }, + {} +}; + +static const struct bin_table bin_dev_table[] = { + /* DEV_CDROM "cdrom" no longer used */ + /* DEV_HWMON unused */ + /* DEV_PARPORT "parport" no longer used */ + { CTL_DIR, DEV_RAID, "raid", bin_raid_table }, + { CTL_DIR, DEV_MAC_HID, "mac_hid", bin_mac_hid_files }, + { CTL_DIR, DEV_SCSI, "scsi", bin_scsi_table }, + { CTL_DIR, DEV_IPMI, "ipmi", bin_ipmi_table }, + {} +}; + +static const struct bin_table bin_bus_isa_table[] = { + { CTL_INT, BUS_ISA_MEM_BASE, "membase" }, + { CTL_INT, BUS_ISA_PORT_BASE, "portbase" }, + { CTL_INT, BUS_ISA_PORT_SHIFT, "portshift" }, + {} +}; + +static const struct bin_table bin_bus_table[] = { + { CTL_DIR, CTL_BUS_ISA, "isa", bin_bus_isa_table }, + {} +}; + + +static const struct bin_table bin_s390dbf_table[] = { + { CTL_INT, 5678 /* CTL_S390DBF_STOPPABLE */, "debug_stoppable" }, + { CTL_INT, 5679 /* CTL_S390DBF_ACTIVE */, "debug_active" }, + {} +}; + +static const struct bin_table bin_sunrpc_table[] = { + /* CTL_RPCDEBUG "rpc_debug" no longer used */ + /* CTL_NFSDEBUG "nfs_debug" no longer used */ + /* CTL_NFSDDEBUG "nfsd_debug" no longer used */ + /* CTL_NLMDEBUG "nlm_debug" no longer used */ + + { CTL_INT, CTL_SLOTTABLE_UDP, "udp_slot_table_entries" }, + { CTL_INT, CTL_SLOTTABLE_TCP, "tcp_slot_table_entries" }, + { CTL_INT, CTL_MIN_RESVPORT, "min_resvport" }, + { CTL_INT, CTL_MAX_RESVPORT, "max_resvport" }, + {} +}; + +static const struct bin_table bin_pm_table[] = { + /* frv specific */ + /* 1 == CTL_PM_SUSPEND "suspend" no longer used" */ + { CTL_INT, 2 /* CTL_PM_CMODE */, "cmode" }, + { CTL_INT, 3 /* CTL_PM_P0 */, "p0" }, + { CTL_INT, 4 /* CTL_PM_CM */, "cm" }, + {} +}; + +static const struct bin_table bin_root_table[] = { + { CTL_DIR, CTL_KERN, "kernel", bin_kern_table }, + { CTL_DIR, CTL_VM, "vm", bin_vm_table }, + { CTL_DIR, CTL_NET, "net", bin_net_table }, + /* CTL_PROC not used */ + { CTL_DIR, CTL_FS, "fs", bin_fs_table }, + /* CTL_DEBUG "debug" no longer used */ + { CTL_DIR, CTL_DEV, "dev", bin_dev_table }, + { CTL_DIR, CTL_BUS, "bus", bin_bus_table }, + { CTL_DIR, CTL_ABI, "abi" }, + /* CTL_CPU not used */ + /* CTL_ARLAN "arlan" no longer used */ + { CTL_DIR, CTL_S390DBF, "s390dbf", bin_s390dbf_table }, + { CTL_DIR, CTL_SUNRPC, "sunrpc", bin_sunrpc_table }, + { CTL_DIR, CTL_PM, "pm", bin_pm_table }, + {} +}; + +static ssize_t bin_dir(struct file *file, + void __user *oldval, size_t oldlen, void __user *newval, size_t newlen) +{ + return -ENOTDIR; +} + + +static ssize_t bin_string(struct file *file, + void __user *oldval, size_t oldlen, void __user *newval, size_t newlen) +{ + ssize_t result, copied = 0; + + if (oldval && oldlen) { + char __user *lastp; + loff_t pos = 0; + int ch; + + result = vfs_read(file, oldval, oldlen, &pos); + if (result < 0) + goto out; + + copied = result; + lastp = oldval + copied - 1; + + result = -EFAULT; + if (get_user(ch, lastp)) + goto out; + + /* Trim off the trailing newline */ + if (ch == '\n') { + result = -EFAULT; + if (put_user('\0', lastp)) + goto out; + copied -= 1; + } + } + + if (newval && newlen) { + loff_t pos = 0; + + result = vfs_write(file, newval, newlen, &pos); + if (result < 0) + goto out; + } + + result = copied; +out: + return result; +} + +static ssize_t bin_intvec(struct file *file, + void __user *oldval, size_t oldlen, void __user *newval, size_t newlen) +{ + mm_segment_t old_fs = get_fs(); + ssize_t copied = 0; + char *buffer; + ssize_t result; + + result = -ENOMEM; + buffer = kmalloc(BUFSZ, GFP_KERNEL); + if (!buffer) + goto out; + + if (oldval && oldlen) { + unsigned __user *vec = oldval; + size_t length = oldlen / sizeof(*vec); + loff_t pos = 0; + char *str, *end; + int i; + + set_fs(KERNEL_DS); + result = vfs_read(file, buffer, BUFSZ - 1, &pos); + set_fs(old_fs); + if (result < 0) + goto out_kfree; + + str = buffer; + end = str + result; + *end++ = '\0'; + for (i = 0; i < length; i++) { + unsigned long value; + + value = simple_strtoul(str, &str, 10); + while (isspace(*str)) + str++; + + result = -EFAULT; + if (put_user(value, vec + i)) + goto out_kfree; + + copied += sizeof(*vec); + if (!isdigit(*str)) + break; + } + } + + if (newval && newlen) { + unsigned __user *vec = newval; + size_t length = newlen / sizeof(*vec); + loff_t pos = 0; + char *str, *end; + int i; + + str = buffer; + end = str + BUFSZ; + for (i = 0; i < length; i++) { + unsigned long value; + + result = -EFAULT; + if (get_user(value, vec + i)) + goto out_kfree; + + str += snprintf(str, end - str, "%lu\t", value); + } + + set_fs(KERNEL_DS); + result = vfs_write(file, buffer, str - buffer, &pos); + set_fs(old_fs); + if (result < 0) + goto out_kfree; + } + result = copied; +out_kfree: + kfree(buffer); +out: + return result; +} + +static ssize_t bin_ulongvec(struct file *file, + void __user *oldval, size_t oldlen, void __user *newval, size_t newlen) +{ + mm_segment_t old_fs = get_fs(); + ssize_t copied = 0; + char *buffer; + ssize_t result; + + result = -ENOMEM; + buffer = kmalloc(BUFSZ, GFP_KERNEL); + if (!buffer) + goto out; + + if (oldval && oldlen) { + unsigned long __user *vec = oldval; + size_t length = oldlen / sizeof(*vec); + loff_t pos = 0; + char *str, *end; + int i; + + set_fs(KERNEL_DS); + result = vfs_read(file, buffer, BUFSZ - 1, &pos); + set_fs(old_fs); + if (result < 0) + goto out_kfree; + + str = buffer; + end = str + result; + *end++ = '\0'; + for (i = 0; i < length; i++) { + unsigned long value; + + value = simple_strtoul(str, &str, 10); + while (isspace(*str)) + str++; + + result = -EFAULT; + if (put_user(value, vec + i)) + goto out_kfree; + + copied += sizeof(*vec); + if (!isdigit(*str)) + break; + } + } + + if (newval && newlen) { + unsigned long __user *vec = newval; + size_t length = newlen / sizeof(*vec); + loff_t pos = 0; + char *str, *end; + int i; + + str = buffer; + end = str + BUFSZ; + for (i = 0; i < length; i++) { + unsigned long value; + + result = -EFAULT; + if (get_user(value, vec + i)) + goto out_kfree; + + str += snprintf(str, end - str, "%lu\t", value); + } + + set_fs(KERNEL_DS); + result = vfs_write(file, buffer, str - buffer, &pos); + set_fs(old_fs); + if (result < 0) + goto out_kfree; + } + result = copied; +out_kfree: + kfree(buffer); +out: + return result; +} + +static unsigned hex_value(int ch) +{ + return isdigit(ch) ? ch - '0' : ((ch | 0x20) - 'a') + 10; +} + +static ssize_t bin_uuid(struct file *file, + void __user *oldval, size_t oldlen, void __user *newval, size_t newlen) +{ + mm_segment_t old_fs = get_fs(); + ssize_t result, copied = 0; + + /* Only supports reads */ + if (oldval && oldlen) { + loff_t pos = 0; + char buf[40], *str = buf; + unsigned char uuid[16]; + int i; + + set_fs(KERNEL_DS); + result = vfs_read(file, buf, sizeof(buf) - 1, &pos); + set_fs(old_fs); + if (result < 0) + goto out; + + buf[result] = '\0'; + + /* Convert the uuid to from a string to binary */ + for (i = 0; i < 16; i++) { + result = -EIO; + if (!isxdigit(str[0]) || !isxdigit(str[1])) + goto out; + + uuid[i] = (hex_value(str[0]) << 4) | hex_value(str[1]); + str += 2; + if (*str == '-') + str++; + } + + if (oldlen > 16) + oldlen = 16; + + result = -EFAULT; + if (copy_to_user(oldval, uuid, oldlen)) + goto out; + + copied = oldlen; + } + result = copied; +out: + return result; +} + +static ssize_t bin_dn_node_address(struct file *file, + void __user *oldval, size_t oldlen, void __user *newval, size_t newlen) +{ + mm_segment_t old_fs = get_fs(); + ssize_t result, copied = 0; + + if (oldval && oldlen) { + loff_t pos = 0; + char buf[15], *nodep; + unsigned long area, node; + __le16 dnaddr; + + set_fs(KERNEL_DS); + result = vfs_read(file, buf, sizeof(buf) - 1, &pos); + set_fs(old_fs); + if (result < 0) + goto out; + + buf[result] = '\0'; + + /* Convert the decnet addresss to binary */ + result = -EIO; + nodep = strchr(buf, '.') + 1; + if (!nodep) + goto out; + + area = simple_strtoul(buf, NULL, 10); + node = simple_strtoul(nodep, NULL, 10); + + result = -EIO; + if ((area > 63)||(node > 1023)) + goto out; + + dnaddr = cpu_to_le16((area << 10) | node); + + result = -EFAULT; + if (put_user(dnaddr, (__le16 __user *)oldval)) + goto out; + + copied = sizeof(dnaddr); + } + + if (newval && newlen) { + loff_t pos = 0; + __le16 dnaddr; + char buf[15]; + int len; + + result = -EINVAL; + if (newlen != sizeof(dnaddr)) + goto out; + + result = -EFAULT; + if (get_user(dnaddr, (__le16 __user *)newval)) + goto out; + + len = snprintf(buf, sizeof(buf), "%hu.%hu", + le16_to_cpu(dnaddr) >> 10, + le16_to_cpu(dnaddr) & 0x3ff); + + set_fs(KERNEL_DS); + result = vfs_write(file, buf, len, &pos); + set_fs(old_fs); + if (result < 0) + goto out; + } + + result = copied; +out: + return result; +} + +static const struct bin_table *get_sysctl(const int *name, int nlen, char *path) +{ + const struct bin_table *table = &bin_root_table[0]; + int ctl_name; + + /* The binary sysctl tables have a small maximum depth so + * there is no danger of overflowing our path as it PATH_MAX + * bytes long. + */ + memcpy(path, "sys/", 4); + path += 4; + +repeat: + if (!nlen) + return ERR_PTR(-ENOTDIR); + ctl_name = *name; + name++; + nlen--; + for ( ; table->convert; table++) { + int len = 0; + + /* + * For a wild card entry map from ifindex to network + * device name. + */ + if (!table->ctl_name) { +#ifdef CONFIG_NET + struct net *net = current->nsproxy->net_ns; + struct net_device *dev; + dev = dev_get_by_index(net, ctl_name); + if (dev) { + len = strlen(dev->name); + memcpy(path, dev->name, len); + dev_put(dev); + } +#endif + /* Use the well known sysctl number to proc name mapping */ + } else if (ctl_name == table->ctl_name) { + len = strlen(table->procname); + memcpy(path, table->procname, len); + } + if (len) { + path += len; + if (table->child) { + *path++ = '/'; + table = table->child; + goto repeat; + } + *path = '\0'; + return table; + } + } + return ERR_PTR(-ENOTDIR); +} + +static char *sysctl_getname(const int *name, int nlen, const struct bin_table **tablep) +{ + char *tmp, *result; + + result = ERR_PTR(-ENOMEM); + tmp = __getname(); + if (tmp) { + const struct bin_table *table = get_sysctl(name, nlen, tmp); + result = tmp; + *tablep = table; + if (IS_ERR(table)) { + __putname(tmp); + result = ERR_CAST(table); + } + } + return result; +} + +static ssize_t binary_sysctl(const int *name, int nlen, + void __user *oldval, size_t oldlen, void __user *newval, size_t newlen) +{ + const struct bin_table *table = NULL; + struct nameidata nd; + struct vfsmount *mnt; + struct file *file; + ssize_t result; + char *pathname; + int flags; + int acc_mode, fmode; + + pathname = sysctl_getname(name, nlen, &table); + result = PTR_ERR(pathname); + if (IS_ERR(pathname)) + goto out; + + /* How should the sysctl be accessed? */ + if (oldval && oldlen && newval && newlen) { + flags = O_RDWR; + acc_mode = MAY_READ | MAY_WRITE; + fmode = FMODE_READ | FMODE_WRITE; + } else if (newval && newlen) { + flags = O_WRONLY; + acc_mode = MAY_WRITE; + fmode = FMODE_WRITE; + } else if (oldval && oldlen) { + flags = O_RDONLY; + acc_mode = MAY_READ; + fmode = FMODE_READ; + } else { + result = 0; + goto out_putname; + } + + mnt = current->nsproxy->pid_ns->proc_mnt; + result = vfs_path_lookup(mnt->mnt_root, mnt, pathname, 0, &nd); + if (result) + goto out_putname; + + result = may_open(&nd.path, acc_mode, fmode); + if (result) + goto out_putpath; + + file = dentry_open(nd.path.dentry, nd.path.mnt, flags, current_cred()); + result = PTR_ERR(file); + if (IS_ERR(file)) + goto out_putname; + + result = table->convert(file, oldval, oldlen, newval, newlen); + + fput(file); +out_putname: + putname(pathname); +out: + return result; + +out_putpath: + path_put(&nd.path); + goto out_putname; +} + + +#else /* CONFIG_SYSCTL_SYSCALL */ + +static ssize_t binary_sysctl(const int *name, int nlen, + void __user *oldval, size_t oldlen, void __user *newval, size_t newlen) +{ + return -ENOSYS; +} + +#endif /* CONFIG_SYSCTL_SYSCALL */ + + +static void deprecated_sysctl_warning(const int *name, int nlen) +{ + int i; + + if (printk_ratelimit()) { + printk(KERN_INFO + "warning: process `%s' used the deprecated sysctl " + "system call with ", current->comm); + for (i = 0; i < nlen; i++) + printk("%d.", name[i]); + printk("\n"); + } + return; +} + +static ssize_t do_sysctl(int __user *args_name, int nlen, + void __user *oldval, size_t oldlen, void __user *newval, size_t newlen) +{ + int name[CTL_MAXNAME]; + int i; + + /* Check args->nlen. */ + if (nlen < 0 || nlen > CTL_MAXNAME) + return -ENOTDIR; + /* Read in the sysctl name for simplicity */ + for (i = 0; i < nlen; i++) + if (get_user(name[i], args_name + i)) + return -EFAULT; + + deprecated_sysctl_warning(name, nlen); + + return binary_sysctl(name, nlen, oldval, oldlen, newval, newlen); +} + +SYSCALL_DEFINE1(sysctl, struct __sysctl_args __user *, args) +{ + struct __sysctl_args tmp; + size_t oldlen = 0; + ssize_t result; + + if (copy_from_user(&tmp, args, sizeof(tmp))) + return -EFAULT; + + if (tmp.oldval && !tmp.oldlenp) + return -EFAULT; + + if (tmp.oldlenp && get_user(oldlen, tmp.oldlenp)) + return -EFAULT; + + result = do_sysctl(tmp.name, tmp.nlen, tmp.oldval, oldlen, + tmp.newval, tmp.newlen); + + if (result >= 0) { + oldlen = result; + result = 0; + } + + if (tmp.oldlenp && put_user(oldlen, tmp.oldlenp)) + return -EFAULT; + + return result; +} + + +#ifdef CONFIG_COMPAT +#include <asm/compat.h> + +struct compat_sysctl_args { + compat_uptr_t name; + int nlen; + compat_uptr_t oldval; + compat_uptr_t oldlenp; + compat_uptr_t newval; + compat_size_t newlen; + compat_ulong_t __unused[4]; +}; + +asmlinkage long compat_sys_sysctl(struct compat_sysctl_args __user *args) +{ + struct compat_sysctl_args tmp; + compat_size_t __user *compat_oldlenp; + size_t oldlen = 0; + ssize_t result; + + if (copy_from_user(&tmp, args, sizeof(tmp))) + return -EFAULT; + + if (tmp.oldval && !tmp.oldlenp) + return -EFAULT; + + compat_oldlenp = compat_ptr(tmp.oldlenp); + if (compat_oldlenp && get_user(oldlen, compat_oldlenp)) + return -EFAULT; + + result = do_sysctl(compat_ptr(tmp.name), tmp.nlen, + compat_ptr(tmp.oldval), oldlen, + compat_ptr(tmp.newval), tmp.newlen); + + if (result >= 0) { + oldlen = result; + result = 0; + } + + if (compat_oldlenp && put_user(oldlen, compat_oldlenp)) + return -EFAULT; + + return result; +} + +#endif /* CONFIG_COMPAT */ diff --git a/kernel/sysctl_check.c b/kernel/sysctl_check.c index b38423ca711..04cdcf72c82 100644 --- a/kernel/sysctl_check.c +++ b/kernel/sysctl_check.c @@ -5,1239 +5,6 @@ #include <linux/string.h> #include <net/ip_vs.h> -struct trans_ctl_table { - int ctl_name; - const char *procname; - const struct trans_ctl_table *child; -}; - -static const struct trans_ctl_table trans_random_table[] = { - { RANDOM_POOLSIZE, "poolsize" }, - { RANDOM_ENTROPY_COUNT, "entropy_avail" }, - { RANDOM_READ_THRESH, "read_wakeup_threshold" }, - { RANDOM_WRITE_THRESH, "write_wakeup_threshold" }, - { RANDOM_BOOT_ID, "boot_id" }, - { RANDOM_UUID, "uuid" }, - {} -}; - -static const struct trans_ctl_table trans_pty_table[] = { - { PTY_MAX, "max" }, - { PTY_NR, "nr" }, - {} -}; - -static const struct trans_ctl_table trans_kern_table[] = { - { KERN_OSTYPE, "ostype" }, - { KERN_OSRELEASE, "osrelease" }, - /* KERN_OSREV not used */ - { KERN_VERSION, "version" }, - /* KERN_SECUREMASK not used */ - /* KERN_PROF not used */ - { KERN_NODENAME, "hostname" }, - { KERN_DOMAINNAME, "domainname" }, - - { KERN_PANIC, "panic" }, - { KERN_REALROOTDEV, "real-root-dev" }, - - { KERN_SPARC_REBOOT, "reboot-cmd" }, - { KERN_CTLALTDEL, "ctrl-alt-del" }, - { KERN_PRINTK, "printk" }, - - /* KERN_NAMETRANS not used */ - /* KERN_PPC_HTABRECLAIM not used */ - /* KERN_PPC_ZEROPAGED not used */ - { KERN_PPC_POWERSAVE_NAP, "powersave-nap" }, - - { KERN_MODPROBE, "modprobe" }, - { KERN_SG_BIG_BUFF, "sg-big-buff" }, - { KERN_ACCT, "acct" }, - { KERN_PPC_L2CR, "l2cr" }, - - /* KERN_RTSIGNR not used */ - /* KERN_RTSIGMAX not used */ - - { KERN_SHMMAX, "shmmax" }, - { KERN_MSGMAX, "msgmax" }, - { KERN_MSGMNB, "msgmnb" }, - /* KERN_MSGPOOL not used*/ - { KERN_SYSRQ, "sysrq" }, - { KERN_MAX_THREADS, "threads-max" }, - { KERN_RANDOM, "random", trans_random_table }, - { KERN_SHMALL, "shmall" }, - { KERN_MSGMNI, "msgmni" }, - { KERN_SEM, "sem" }, - { KERN_SPARC_STOP_A, "stop-a" }, - { KERN_SHMMNI, "shmmni" }, - - { KERN_OVERFLOWUID, "overflowuid" }, - { KERN_OVERFLOWGID, "overflowgid" }, - - { KERN_HOTPLUG, "hotplug", }, - { KERN_IEEE_EMULATION_WARNINGS, "ieee_emulation_warnings" }, - - { KERN_S390_USER_DEBUG_LOGGING, "userprocess_debug" }, - { KERN_CORE_USES_PID, "core_uses_pid" }, - { KERN_TAINTED, "tainted" }, - { KERN_CADPID, "cad_pid" }, - { KERN_PIDMAX, "pid_max" }, - { KERN_CORE_PATTERN, "core_pattern" }, - { KERN_PANIC_ON_OOPS, "panic_on_oops" }, - { KERN_HPPA_PWRSW, "soft-power" }, - { KERN_HPPA_UNALIGNED, "unaligned-trap" }, - - { KERN_PRINTK_RATELIMIT, "printk_ratelimit" }, - { KERN_PRINTK_RATELIMIT_BURST, "printk_ratelimit_burst" }, - - { KERN_PTY, "pty", trans_pty_table }, - { KERN_NGROUPS_MAX, "ngroups_max" }, - { KERN_SPARC_SCONS_PWROFF, "scons-poweroff" }, - { KERN_HZ_TIMER, "hz_timer" }, - { KERN_UNKNOWN_NMI_PANIC, "unknown_nmi_panic" }, - { KERN_BOOTLOADER_TYPE, "bootloader_type" }, - { KERN_RANDOMIZE, "randomize_va_space" }, - - { KERN_SPIN_RETRY, "spin_retry" }, - { KERN_ACPI_VIDEO_FLAGS, "acpi_video_flags" }, - { KERN_IA64_UNALIGNED, "ignore-unaligned-usertrap" }, - { KERN_COMPAT_LOG, "compat-log" }, - { KERN_MAX_LOCK_DEPTH, "max_lock_depth" }, - { KERN_NMI_WATCHDOG, "nmi_watchdog" }, - { KERN_PANIC_ON_NMI, "panic_on_unrecovered_nmi" }, - {} -}; - -static const struct trans_ctl_table trans_vm_table[] = { - { VM_OVERCOMMIT_MEMORY, "overcommit_memory" }, - { VM_PAGE_CLUSTER, "page-cluster" }, - { VM_DIRTY_BACKGROUND, "dirty_background_ratio" }, - { VM_DIRTY_RATIO, "dirty_ratio" }, - { VM_DIRTY_WB_CS, "dirty_writeback_centisecs" }, - { VM_DIRTY_EXPIRE_CS, "dirty_expire_centisecs" }, - { VM_NR_PDFLUSH_THREADS, "nr_pdflush_threads" }, - { VM_OVERCOMMIT_RATIO, "overcommit_ratio" }, - /* VM_PAGEBUF unused */ - { VM_HUGETLB_PAGES, "nr_hugepages" }, - { VM_SWAPPINESS, "swappiness" }, - { VM_LOWMEM_RESERVE_RATIO, "lowmem_reserve_ratio" }, - { VM_MIN_FREE_KBYTES, "min_free_kbytes" }, - { VM_MAX_MAP_COUNT, "max_map_count" }, - { VM_LAPTOP_MODE, "laptop_mode" }, - { VM_BLOCK_DUMP, "block_dump" }, - { VM_HUGETLB_GROUP, "hugetlb_shm_group" }, - { VM_VFS_CACHE_PRESSURE, "vfs_cache_pressure" }, - { VM_LEGACY_VA_LAYOUT, "legacy_va_layout" }, - /* VM_SWAP_TOKEN_TIMEOUT unused */ - { VM_DROP_PAGECACHE, "drop_caches" }, - { VM_PERCPU_PAGELIST_FRACTION, "percpu_pagelist_fraction" }, - { VM_ZONE_RECLAIM_MODE, "zone_reclaim_mode" }, - { VM_MIN_UNMAPPED, "min_unmapped_ratio" }, - { VM_PANIC_ON_OOM, "panic_on_oom" }, - { VM_VDSO_ENABLED, "vdso_enabled" }, - { VM_MIN_SLAB, "min_slab_ratio" }, - - {} -}; - -static const struct trans_ctl_table trans_net_core_table[] = { - { NET_CORE_WMEM_MAX, "wmem_max" }, - { NET_CORE_RMEM_MAX, "rmem_max" }, - { NET_CORE_WMEM_DEFAULT, "wmem_default" }, - { NET_CORE_RMEM_DEFAULT, "rmem_default" }, - /* NET_CORE_DESTROY_DELAY unused */ - { NET_CORE_MAX_BACKLOG, "netdev_max_backlog" }, - /* NET_CORE_FASTROUTE unused */ - { NET_CORE_MSG_COST, "message_cost" }, - { NET_CORE_MSG_BURST, "message_burst" }, - { NET_CORE_OPTMEM_MAX, "optmem_max" }, - /* NET_CORE_HOT_LIST_LENGTH unused */ - /* NET_CORE_DIVERT_VERSION unused */ - /* NET_CORE_NO_CONG_THRESH unused */ - /* NET_CORE_NO_CONG unused */ - /* NET_CORE_LO_CONG unused */ - /* NET_CORE_MOD_CONG unused */ - { NET_CORE_DEV_WEIGHT, "dev_weight" }, - { NET_CORE_SOMAXCONN, "somaxconn" }, - { NET_CORE_BUDGET, "netdev_budget" }, - { NET_CORE_AEVENT_ETIME, "xfrm_aevent_etime" }, - { NET_CORE_AEVENT_RSEQTH, "xfrm_aevent_rseqth" }, - { NET_CORE_WARNINGS, "warnings" }, - {}, -}; - -static const struct trans_ctl_table trans_net_unix_table[] = { - /* NET_UNIX_DESTROY_DELAY unused */ - /* NET_UNIX_DELETE_DELAY unused */ - { NET_UNIX_MAX_DGRAM_QLEN, "max_dgram_qlen" }, - {} -}; - -static const struct trans_ctl_table trans_net_ipv4_route_table[] = { - { NET_IPV4_ROUTE_FLUSH, "flush" }, - { NET_IPV4_ROUTE_MIN_DELAY, "min_delay" }, - { NET_IPV4_ROUTE_MAX_DELAY, "max_delay" }, - { NET_IPV4_ROUTE_GC_THRESH, "gc_thresh" }, - { NET_IPV4_ROUTE_MAX_SIZE, "max_size" }, - { NET_IPV4_ROUTE_GC_MIN_INTERVAL, "gc_min_interval" }, - { NET_IPV4_ROUTE_GC_TIMEOUT, "gc_timeout" }, - { NET_IPV4_ROUTE_GC_INTERVAL, "gc_interval" }, - { NET_IPV4_ROUTE_REDIRECT_LOAD, "redirect_load" }, - { NET_IPV4_ROUTE_REDIRECT_NUMBER, "redirect_number" }, - { NET_IPV4_ROUTE_REDIRECT_SILENCE, "redirect_silence" }, - { NET_IPV4_ROUTE_ERROR_COST, "error_cost" }, - { NET_IPV4_ROUTE_ERROR_BURST, "error_burst" }, - { NET_IPV4_ROUTE_GC_ELASTICITY, "gc_elasticity" }, - { NET_IPV4_ROUTE_MTU_EXPIRES, "mtu_expires" }, - { NET_IPV4_ROUTE_MIN_PMTU, "min_pmtu" }, - { NET_IPV4_ROUTE_MIN_ADVMSS, "min_adv_mss" }, - { NET_IPV4_ROUTE_SECRET_INTERVAL, "secret_interval" }, - { NET_IPV4_ROUTE_GC_MIN_INTERVAL_MS, "gc_min_interval_ms" }, - {} -}; - -static const struct trans_ctl_table trans_net_ipv4_conf_vars_table[] = { - { NET_IPV4_CONF_FORWARDING, "forwarding" }, - { NET_IPV4_CONF_MC_FORWARDING, "mc_forwarding" }, - - { NET_IPV4_CONF_PROXY_ARP, "proxy_arp" }, - { NET_IPV4_CONF_ACCEPT_REDIRECTS, "accept_redirects" }, - { NET_IPV4_CONF_SECURE_REDIRECTS, "secure_redirects" }, - { NET_IPV4_CONF_SEND_REDIRECTS, "send_redirects" }, - { NET_IPV4_CONF_SHARED_MEDIA, "shared_media" }, - { NET_IPV4_CONF_RP_FILTER, "rp_filter" }, - { NET_IPV4_CONF_ACCEPT_SOURCE_ROUTE, "accept_source_route" }, - { NET_IPV4_CONF_BOOTP_RELAY, "bootp_relay" }, - { NET_IPV4_CONF_LOG_MARTIANS, "log_martians" }, - { NET_IPV4_CONF_TAG, "tag" }, - { NET_IPV4_CONF_ARPFILTER, "arp_filter" }, - { NET_IPV4_CONF_MEDIUM_ID, "medium_id" }, - { NET_IPV4_CONF_NOXFRM, "disable_xfrm" }, - { NET_IPV4_CONF_NOPOLICY, "disable_policy" }, - { NET_IPV4_CONF_FORCE_IGMP_VERSION, "force_igmp_version" }, - - { NET_IPV4_CONF_ARP_ANNOUNCE, "arp_announce" }, - { NET_IPV4_CONF_ARP_IGNORE, "arp_ignore" }, - { NET_IPV4_CONF_PROMOTE_SECONDARIES, "promote_secondaries" }, - { NET_IPV4_CONF_ARP_ACCEPT, "arp_accept" }, - { NET_IPV4_CONF_ARP_NOTIFY, "arp_notify" }, - {} -}; - -static const struct trans_ctl_table trans_net_ipv4_conf_table[] = { - { NET_PROTO_CONF_ALL, "all", trans_net_ipv4_conf_vars_table }, - { NET_PROTO_CONF_DEFAULT, "default", trans_net_ipv4_conf_vars_table }, - { 0, NULL, trans_net_ipv4_conf_vars_table }, - {} -}; - -static const struct trans_ctl_table trans_net_neigh_vars_table[] = { - { NET_NEIGH_MCAST_SOLICIT, "mcast_solicit" }, - { NET_NEIGH_UCAST_SOLICIT, "ucast_solicit" }, - { NET_NEIGH_APP_SOLICIT, "app_solicit" }, - { NET_NEIGH_RETRANS_TIME, "retrans_time" }, - { NET_NEIGH_REACHABLE_TIME, "base_reachable_time" }, - { NET_NEIGH_DELAY_PROBE_TIME, "delay_first_probe_time" }, - { NET_NEIGH_GC_STALE_TIME, "gc_stale_time" }, - { NET_NEIGH_UNRES_QLEN, "unres_qlen" }, - { NET_NEIGH_PROXY_QLEN, "proxy_qlen" }, - { NET_NEIGH_ANYCAST_DELAY, "anycast_delay" }, - { NET_NEIGH_PROXY_DELAY, "proxy_delay" }, - { NET_NEIGH_LOCKTIME, "locktime" }, - { NET_NEIGH_GC_INTERVAL, "gc_interval" }, - { NET_NEIGH_GC_THRESH1, "gc_thresh1" }, - { NET_NEIGH_GC_THRESH2, "gc_thresh2" }, - { NET_NEIGH_GC_THRESH3, "gc_thresh3" }, - { NET_NEIGH_RETRANS_TIME_MS, "retrans_time_ms" }, - { NET_NEIGH_REACHABLE_TIME_MS, "base_reachable_time_ms" }, - {} -}; - -static const struct trans_ctl_table trans_net_neigh_table[] = { - { NET_PROTO_CONF_DEFAULT, "default", trans_net_neigh_vars_table }, - { 0, NULL, trans_net_neigh_vars_table }, - {} -}; - -static const struct trans_ctl_table trans_net_ipv4_netfilter_table[] = { - { NET_IPV4_NF_CONNTRACK_MAX, "ip_conntrack_max" }, - - { NET_IPV4_NF_CONNTRACK_TCP_TIMEOUT_SYN_SENT, "ip_conntrack_tcp_timeout_syn_sent" }, - { NET_IPV4_NF_CONNTRACK_TCP_TIMEOUT_SYN_RECV, "ip_conntrack_tcp_timeout_syn_recv" }, - { NET_IPV4_NF_CONNTRACK_TCP_TIMEOUT_ESTABLISHED, "ip_conntrack_tcp_timeout_established" }, - { NET_IPV4_NF_CONNTRACK_TCP_TIMEOUT_FIN_WAIT, "ip_conntrack_tcp_timeout_fin_wait" }, - { NET_IPV4_NF_CONNTRACK_TCP_TIMEOUT_CLOSE_WAIT, "ip_conntrack_tcp_timeout_close_wait" }, - { NET_IPV4_NF_CONNTRACK_TCP_TIMEOUT_LAST_ACK, "ip_conntrack_tcp_timeout_last_ack" }, - { NET_IPV4_NF_CONNTRACK_TCP_TIMEOUT_TIME_WAIT, "ip_conntrack_tcp_timeout_time_wait" }, - { NET_IPV4_NF_CONNTRACK_TCP_TIMEOUT_CLOSE, "ip_conntrack_tcp_timeout_close" }, - - { NET_IPV4_NF_CONNTRACK_UDP_TIMEOUT, "ip_conntrack_udp_timeout" }, - { NET_IPV4_NF_CONNTRACK_UDP_TIMEOUT_STREAM, "ip_conntrack_udp_timeout_stream" }, - { NET_IPV4_NF_CONNTRACK_ICMP_TIMEOUT, "ip_conntrack_icmp_timeout" }, - { NET_IPV4_NF_CONNTRACK_GENERIC_TIMEOUT, "ip_conntrack_generic_timeout" }, - - { NET_IPV4_NF_CONNTRACK_BUCKETS, "ip_conntrack_buckets" }, - { NET_IPV4_NF_CONNTRACK_LOG_INVALID, "ip_conntrack_log_invalid" }, - { NET_IPV4_NF_CONNTRACK_TCP_TIMEOUT_MAX_RETRANS, "ip_conntrack_tcp_timeout_max_retrans" }, - { NET_IPV4_NF_CONNTRACK_TCP_LOOSE, "ip_conntrack_tcp_loose" }, - { NET_IPV4_NF_CONNTRACK_TCP_BE_LIBERAL, "ip_conntrack_tcp_be_liberal" }, - { NET_IPV4_NF_CONNTRACK_TCP_MAX_RETRANS, "ip_conntrack_tcp_max_retrans" }, - - { NET_IPV4_NF_CONNTRACK_SCTP_TIMEOUT_CLOSED, "ip_conntrack_sctp_timeout_closed" }, - { NET_IPV4_NF_CONNTRACK_SCTP_TIMEOUT_COOKIE_WAIT, "ip_conntrack_sctp_timeout_cookie_wait" }, - { NET_IPV4_NF_CONNTRACK_SCTP_TIMEOUT_COOKIE_ECHOED, "ip_conntrack_sctp_timeout_cookie_echoed" }, - { NET_IPV4_NF_CONNTRACK_SCTP_TIMEOUT_ESTABLISHED, "ip_conntrack_sctp_timeout_established" }, - { NET_IPV4_NF_CONNTRACK_SCTP_TIMEOUT_SHUTDOWN_SENT, "ip_conntrack_sctp_timeout_shutdown_sent" }, - { NET_IPV4_NF_CONNTRACK_SCTP_TIMEOUT_SHUTDOWN_RECD, "ip_conntrack_sctp_timeout_shutdown_recd" }, - { NET_IPV4_NF_CONNTRACK_SCTP_TIMEOUT_SHUTDOWN_ACK_SENT, "ip_conntrack_sctp_timeout_shutdown_ack_sent" }, - - { NET_IPV4_NF_CONNTRACK_COUNT, "ip_conntrack_count" }, - { NET_IPV4_NF_CONNTRACK_CHECKSUM, "ip_conntrack_checksum" }, - {} -}; - -static const struct trans_ctl_table trans_net_ipv4_table[] = { - { NET_IPV4_FORWARD, "ip_forward" }, - { NET_IPV4_DYNADDR, "ip_dynaddr" }, - - { NET_IPV4_CONF, "conf", trans_net_ipv4_conf_table }, - { NET_IPV4_NEIGH, "neigh", trans_net_neigh_table }, - { NET_IPV4_ROUTE, "route", trans_net_ipv4_route_table }, - /* NET_IPV4_FIB_HASH unused */ - { NET_IPV4_NETFILTER, "netfilter", trans_net_ipv4_netfilter_table }, - - { NET_IPV4_TCP_TIMESTAMPS, "tcp_timestamps" }, - { NET_IPV4_TCP_WINDOW_SCALING, "tcp_window_scaling" }, - { NET_IPV4_TCP_SACK, "tcp_sack" }, - { NET_IPV4_TCP_RETRANS_COLLAPSE, "tcp_retrans_collapse" }, - { NET_IPV4_DEFAULT_TTL, "ip_default_ttl" }, - /* NET_IPV4_AUTOCONFIG unused */ - { NET_IPV4_NO_PMTU_DISC, "ip_no_pmtu_disc" }, - { NET_IPV4_TCP_SYN_RETRIES, "tcp_syn_retries" }, - { NET_IPV4_IPFRAG_HIGH_THRESH, "ipfrag_high_thresh" }, - { NET_IPV4_IPFRAG_LOW_THRESH, "ipfrag_low_thresh" }, - { NET_IPV4_IPFRAG_TIME, "ipfrag_time" }, - /* NET_IPV4_TCP_MAX_KA_PROBES unused */ - { NET_IPV4_TCP_KEEPALIVE_TIME, "tcp_keepalive_time" }, - { NET_IPV4_TCP_KEEPALIVE_PROBES, "tcp_keepalive_probes" }, - { NET_IPV4_TCP_RETRIES1, "tcp_retries1" }, - { NET_IPV4_TCP_RETRIES2, "tcp_retries2" }, - { NET_IPV4_TCP_FIN_TIMEOUT, "tcp_fin_timeout" }, - /* NET_IPV4_IP_MASQ_DEBUG unused */ - { NET_TCP_SYNCOOKIES, "tcp_syncookies" }, - { NET_TCP_STDURG, "tcp_stdurg" }, - { NET_TCP_RFC1337, "tcp_rfc1337" }, - /* NET_TCP_SYN_TAILDROP unused */ - { NET_TCP_MAX_SYN_BACKLOG, "tcp_max_syn_backlog" }, - { NET_IPV4_LOCAL_PORT_RANGE, "ip_local_port_range" }, - { NET_IPV4_ICMP_ECHO_IGNORE_ALL, "icmp_echo_ignore_all" }, - { NET_IPV4_ICMP_ECHO_IGNORE_BROADCASTS, "icmp_echo_ignore_broadcasts" }, - /* NET_IPV4_ICMP_SOURCEQUENCH_RATE unused */ - /* NET_IPV4_ICMP_DESTUNREACH_RATE unused */ - /* NET_IPV4_ICMP_TIMEEXCEED_RATE unused */ - /* NET_IPV4_ICMP_PARAMPROB_RATE unused */ - /* NET_IPV4_ICMP_ECHOREPLY_RATE unused */ - { NET_IPV4_ICMP_IGNORE_BOGUS_ERROR_RESPONSES, "icmp_ignore_bogus_error_responses" }, - { NET_IPV4_IGMP_MAX_MEMBERSHIPS, "igmp_max_memberships" }, - { NET_TCP_TW_RECYCLE, "tcp_tw_recycle" }, - /* NET_IPV4_ALWAYS_DEFRAG unused */ - { NET_IPV4_TCP_KEEPALIVE_INTVL, "tcp_keepalive_intvl" }, - { NET_IPV4_INET_PEER_THRESHOLD, "inet_peer_threshold" }, - { NET_IPV4_INET_PEER_MINTTL, "inet_peer_minttl" }, - { NET_IPV4_INET_PEER_MAXTTL, "inet_peer_maxttl" }, - { NET_IPV4_INET_PEER_GC_MINTIME, "inet_peer_gc_mintime" }, - { NET_IPV4_INET_PEER_GC_MAXTIME, "inet_peer_gc_maxtime" }, - { NET_TCP_ORPHAN_RETRIES, "tcp_orphan_retries" }, - { NET_TCP_ABORT_ON_OVERFLOW, "tcp_abort_on_overflow" }, - { NET_TCP_SYNACK_RETRIES, "tcp_synack_retries" }, - { NET_TCP_MAX_ORPHANS, "tcp_max_orphans" }, - { NET_TCP_MAX_TW_BUCKETS, "tcp_max_tw_buckets" }, - { NET_TCP_FACK, "tcp_fack" }, - { NET_TCP_REORDERING, "tcp_reordering" }, - { NET_TCP_ECN, "tcp_ecn" }, - { NET_TCP_DSACK, "tcp_dsack" }, - { NET_TCP_MEM, "tcp_mem" }, - { NET_TCP_WMEM, "tcp_wmem" }, - { NET_TCP_RMEM, "tcp_rmem" }, - { NET_TCP_APP_WIN, "tcp_app_win" }, - { NET_TCP_ADV_WIN_SCALE, "tcp_adv_win_scale" }, - { NET_IPV4_NONLOCAL_BIND, "ip_nonlocal_bind" }, - { NET_IPV4_ICMP_RATELIMIT, "icmp_ratelimit" }, - { NET_IPV4_ICMP_RATEMASK, "icmp_ratemask" }, - { NET_TCP_TW_REUSE, "tcp_tw_reuse" }, - { NET_TCP_FRTO, "tcp_frto" }, - { NET_TCP_LOW_LATENCY, "tcp_low_latency" }, - { NET_IPV4_IPFRAG_SECRET_INTERVAL, "ipfrag_secret_interval" }, - { NET_IPV4_IGMP_MAX_MSF, "igmp_max_msf" }, - { NET_TCP_NO_METRICS_SAVE, "tcp_no_metrics_save" }, - /* NET_TCP_DEFAULT_WIN_SCALE unused */ - { NET_TCP_MODERATE_RCVBUF, "tcp_moderate_rcvbuf" }, - { NET_TCP_TSO_WIN_DIVISOR, "tcp_tso_win_divisor" }, - /* NET_TCP_BIC_BETA unused */ - { NET_IPV4_ICMP_ERRORS_USE_INBOUND_IFADDR, "icmp_errors_use_inbound_ifaddr" }, - { NET_TCP_CONG_CONTROL, "tcp_congestion_control" }, - { NET_TCP_ABC, "tcp_abc" }, - { NET_IPV4_IPFRAG_MAX_DIST, "ipfrag_max_dist" }, - { NET_TCP_MTU_PROBING, "tcp_mtu_probing" }, - { NET_TCP_BASE_MSS, "tcp_base_mss" }, - { NET_IPV4_TCP_WORKAROUND_SIGNED_WINDOWS, "tcp_workaround_signed_windows" }, - { NET_TCP_DMA_COPYBREAK, "tcp_dma_copybreak" }, - { NET_TCP_SLOW_START_AFTER_IDLE, "tcp_slow_start_after_idle" }, - { NET_CIPSOV4_CACHE_ENABLE, "cipso_cache_enable" }, - { NET_CIPSOV4_CACHE_BUCKET_SIZE, "cipso_cache_bucket_size" }, - { NET_CIPSOV4_RBM_OPTFMT, "cipso_rbm_optfmt" }, - { NET_CIPSOV4_RBM_STRICTVALID, "cipso_rbm_strictvalid" }, - { NET_TCP_AVAIL_CONG_CONTROL, "tcp_available_congestion_control" }, - { NET_TCP_ALLOWED_CONG_CONTROL, "tcp_allowed_congestion_control" }, - { NET_TCP_MAX_SSTHRESH, "tcp_max_ssthresh" }, - { NET_TCP_FRTO_RESPONSE, "tcp_frto_response" }, - { 2088 /* NET_IPQ_QMAX */, "ip_queue_maxlen" }, - {} -}; - -static const struct trans_ctl_table trans_net_ipx_table[] = { - { NET_IPX_PPROP_BROADCASTING, "ipx_pprop_broadcasting" }, - /* NET_IPX_FORWARDING unused */ - {} -}; - -static const struct trans_ctl_table trans_net_atalk_table[] = { - { NET_ATALK_AARP_EXPIRY_TIME, "aarp-expiry-time" }, - { NET_ATALK_AARP_TICK_TIME, "aarp-tick-time" }, - { NET_ATALK_AARP_RETRANSMIT_LIMIT, "aarp-retransmit-limit" }, - { NET_ATALK_AARP_RESOLVE_TIME, "aarp-resolve-time" }, - {}, -}; - -static const struct trans_ctl_table trans_net_netrom_table[] = { - { NET_NETROM_DEFAULT_PATH_QUALITY, "default_path_quality" }, - { NET_NETROM_OBSOLESCENCE_COUNT_INITIALISER, "obsolescence_count_initialiser" }, - { NET_NETROM_NETWORK_TTL_INITIALISER, "network_ttl_initialiser" }, - { NET_NETROM_TRANSPORT_TIMEOUT, "transport_timeout" }, - { NET_NETROM_TRANSPORT_MAXIMUM_TRIES, "transport_maximum_tries" }, - { NET_NETROM_TRANSPORT_ACKNOWLEDGE_DELAY, "transport_acknowledge_delay" }, - { NET_NETROM_TRANSPORT_BUSY_DELAY, "transport_busy_delay" }, - { NET_NETROM_TRANSPORT_REQUESTED_WINDOW_SIZE, "transport_requested_window_size" }, - { NET_NETROM_TRANSPORT_NO_ACTIVITY_TIMEOUT, "transport_no_activity_timeout" }, - { NET_NETROM_ROUTING_CONTROL, "routing_control" }, - { NET_NETROM_LINK_FAILS_COUNT, "link_fails_count" }, - { NET_NETROM_RESET, "reset" }, - {} -}; - -static const struct trans_ctl_table trans_net_ax25_param_table[] = { - { NET_AX25_IP_DEFAULT_MODE, "ip_default_mode" }, - { NET_AX25_DEFAULT_MODE, "ax25_default_mode" }, - { NET_AX25_BACKOFF_TYPE, "backoff_type" }, - { NET_AX25_CONNECT_MODE, "connect_mode" }, - { NET_AX25_STANDARD_WINDOW, "standard_window_size" }, - { NET_AX25_EXTENDED_WINDOW, "extended_window_size" }, - { NET_AX25_T1_TIMEOUT, "t1_timeout" }, - { NET_AX25_T2_TIMEOUT, "t2_timeout" }, - { NET_AX25_T3_TIMEOUT, "t3_timeout" }, - { NET_AX25_IDLE_TIMEOUT, "idle_timeout" }, - { NET_AX25_N2, "maximum_retry_count" }, - { NET_AX25_PACLEN, "maximum_packet_length" }, - { NET_AX25_PROTOCOL, "protocol" }, - { NET_AX25_DAMA_SLAVE_TIMEOUT, "dama_slave_timeout" }, - {} -}; - -static const struct trans_ctl_table trans_net_ax25_table[] = { - { 0, NULL, trans_net_ax25_param_table }, - {} -}; - -static const struct trans_ctl_table trans_net_bridge_table[] = { - { NET_BRIDGE_NF_CALL_ARPTABLES, "bridge-nf-call-arptables" }, - { NET_BRIDGE_NF_CALL_IPTABLES, "bridge-nf-call-iptables" }, - { NET_BRIDGE_NF_CALL_IP6TABLES, "bridge-nf-call-ip6tables" }, - { NET_BRIDGE_NF_FILTER_VLAN_TAGGED, "bridge-nf-filter-vlan-tagged" }, - { NET_BRIDGE_NF_FILTER_PPPOE_TAGGED, "bridge-nf-filter-pppoe-tagged" }, - {} -}; - -static const struct trans_ctl_table trans_net_rose_table[] = { - { NET_ROSE_RESTART_REQUEST_TIMEOUT, "restart_request_timeout" }, - { NET_ROSE_CALL_REQUEST_TIMEOUT, "call_request_timeout" }, - { NET_ROSE_RESET_REQUEST_TIMEOUT, "reset_request_timeout" }, - { NET_ROSE_CLEAR_REQUEST_TIMEOUT, "clear_request_timeout" }, - { NET_ROSE_ACK_HOLD_BACK_TIMEOUT, "acknowledge_hold_back_timeout" }, - { NET_ROSE_ROUTING_CONTROL, "routing_control" }, - { NET_ROSE_LINK_FAIL_TIMEOUT, "link_fail_timeout" }, - { NET_ROSE_MAX_VCS, "maximum_virtual_circuits" }, - { NET_ROSE_WINDOW_SIZE, "window_size" }, - { NET_ROSE_NO_ACTIVITY_TIMEOUT, "no_activity_timeout" }, - {} -}; - -static const struct trans_ctl_table trans_net_ipv6_conf_var_table[] = { - { NET_IPV6_FORWARDING, "forwarding" }, - { NET_IPV6_HOP_LIMIT, "hop_limit" }, - { NET_IPV6_MTU, "mtu" }, - { NET_IPV6_ACCEPT_RA, "accept_ra" }, - { NET_IPV6_ACCEPT_REDIRECTS, "accept_redirects" }, - { NET_IPV6_AUTOCONF, "autoconf" }, - { NET_IPV6_DAD_TRANSMITS, "dad_transmits" }, - { NET_IPV6_RTR_SOLICITS, "router_solicitations" }, - { NET_IPV6_RTR_SOLICIT_INTERVAL, "router_solicitation_interval" }, - { NET_IPV6_RTR_SOLICIT_DELAY, "router_solicitation_delay" }, - { NET_IPV6_USE_TEMPADDR, "use_tempaddr" }, - { NET_IPV6_TEMP_VALID_LFT, "temp_valid_lft" }, - { NET_IPV6_TEMP_PREFERED_LFT, "temp_prefered_lft" }, - { NET_IPV6_REGEN_MAX_RETRY, "regen_max_retry" }, - { NET_IPV6_MAX_DESYNC_FACTOR, "max_desync_factor" }, - { NET_IPV6_MAX_ADDRESSES, "max_addresses" }, - { NET_IPV6_FORCE_MLD_VERSION, "force_mld_version" }, - { NET_IPV6_ACCEPT_RA_DEFRTR, "accept_ra_defrtr" }, - { NET_IPV6_ACCEPT_RA_PINFO, "accept_ra_pinfo" }, - { NET_IPV6_ACCEPT_RA_RTR_PREF, "accept_ra_rtr_pref" }, - { NET_IPV6_RTR_PROBE_INTERVAL, "router_probe_interval" }, - { NET_IPV6_ACCEPT_RA_RT_INFO_MAX_PLEN, "accept_ra_rt_info_max_plen" }, - { NET_IPV6_PROXY_NDP, "proxy_ndp" }, - { NET_IPV6_ACCEPT_SOURCE_ROUTE, "accept_source_route" }, - {} -}; - -static const struct trans_ctl_table trans_net_ipv6_conf_table[] = { - { NET_PROTO_CONF_ALL, "all", trans_net_ipv6_conf_var_table }, - { NET_PROTO_CONF_DEFAULT, "default", trans_net_ipv6_conf_var_table }, - { 0, NULL, trans_net_ipv6_conf_var_table }, - {} -}; - -static const struct trans_ctl_table trans_net_ipv6_route_table[] = { - { NET_IPV6_ROUTE_FLUSH, "flush" }, - { NET_IPV6_ROUTE_GC_THRESH, "gc_thresh" }, - { NET_IPV6_ROUTE_MAX_SIZE, "max_size" }, - { NET_IPV6_ROUTE_GC_MIN_INTERVAL, "gc_min_interval" }, - { NET_IPV6_ROUTE_GC_TIMEOUT, "gc_timeout" }, - { NET_IPV6_ROUTE_GC_INTERVAL, "gc_interval" }, - { NET_IPV6_ROUTE_GC_ELASTICITY, "gc_elasticity" }, - { NET_IPV6_ROUTE_MTU_EXPIRES, "mtu_expires" }, - { NET_IPV6_ROUTE_MIN_ADVMSS, "min_adv_mss" }, - { NET_IPV6_ROUTE_GC_MIN_INTERVAL_MS, "gc_min_interval_ms" }, - {} -}; - -static const struct trans_ctl_table trans_net_ipv6_icmp_table[] = { - { NET_IPV6_ICMP_RATELIMIT, "ratelimit" }, - {} -}; - -static const struct trans_ctl_table trans_net_ipv6_table[] = { - { NET_IPV6_CONF, "conf", trans_net_ipv6_conf_table }, - { NET_IPV6_NEIGH, "neigh", trans_net_neigh_table }, - { NET_IPV6_ROUTE, "route", trans_net_ipv6_route_table }, - { NET_IPV6_ICMP, "icmp", trans_net_ipv6_icmp_table }, - { NET_IPV6_BINDV6ONLY, "bindv6only" }, - { NET_IPV6_IP6FRAG_HIGH_THRESH, "ip6frag_high_thresh" }, - { NET_IPV6_IP6FRAG_LOW_THRESH, "ip6frag_low_thresh" }, - { NET_IPV6_IP6FRAG_TIME, "ip6frag_time" }, - { NET_IPV6_IP6FRAG_SECRET_INTERVAL, "ip6frag_secret_interval" }, - { NET_IPV6_MLD_MAX_MSF, "mld_max_msf" }, - { 2088 /* IPQ_QMAX */, "ip6_queue_maxlen" }, - {} -}; - -static const struct trans_ctl_table trans_net_x25_table[] = { - { NET_X25_RESTART_REQUEST_TIMEOUT, "restart_request_timeout" }, - { NET_X25_CALL_REQUEST_TIMEOUT, "call_request_timeout" }, - { NET_X25_RESET_REQUEST_TIMEOUT, "reset_request_timeout" }, - { NET_X25_CLEAR_REQUEST_TIMEOUT, "clear_request_timeout" }, - { NET_X25_ACK_HOLD_BACK_TIMEOUT, "acknowledgement_hold_back_timeout" }, - { NET_X25_FORWARD, "x25_forward" }, - {} -}; - -static const struct trans_ctl_table trans_net_tr_table[] = { - { NET_TR_RIF_TIMEOUT, "rif_timeout" }, - {} -}; - - -static const struct trans_ctl_table trans_net_decnet_conf_vars[] = { - { NET_DECNET_CONF_DEV_FORWARDING, "forwarding" }, - { NET_DECNET_CONF_DEV_PRIORITY, "priority" }, - { NET_DECNET_CONF_DEV_T2, "t2" }, - { NET_DECNET_CONF_DEV_T3, "t3" }, - {} -}; - -static const struct trans_ctl_table trans_net_decnet_conf[] = { - { 0, NULL, trans_net_decnet_conf_vars }, - {} -}; - -static const struct trans_ctl_table trans_net_decnet_table[] = { - { NET_DECNET_CONF, "conf", trans_net_decnet_conf }, - { NET_DECNET_NODE_ADDRESS, "node_address" }, - { NET_DECNET_NODE_NAME, "node_name" }, - { NET_DECNET_DEFAULT_DEVICE, "default_device" }, - { NET_DECNET_TIME_WAIT, "time_wait" }, - { NET_DECNET_DN_COUNT, "dn_count" }, - { NET_DECNET_DI_COUNT, "di_count" }, - { NET_DECNET_DR_COUNT, "dr_count" }, - { NET_DECNET_DST_GC_INTERVAL, "dst_gc_interval" }, - { NET_DECNET_NO_FC_MAX_CWND, "no_fc_max_cwnd" }, - { NET_DECNET_MEM, "decnet_mem" }, - { NET_DECNET_RMEM, "decnet_rmem" }, - { NET_DECNET_WMEM, "decnet_wmem" }, - { NET_DECNET_DEBUG_LEVEL, "debug" }, - {} -}; - -static const struct trans_ctl_table trans_net_sctp_table[] = { - { NET_SCTP_RTO_INITIAL, "rto_initial" }, - { NET_SCTP_RTO_MIN, "rto_min" }, - { NET_SCTP_RTO_MAX, "rto_max" }, - { NET_SCTP_RTO_ALPHA, "rto_alpha_exp_divisor" }, - { NET_SCTP_RTO_BETA, "rto_beta_exp_divisor" }, - { NET_SCTP_VALID_COOKIE_LIFE, "valid_cookie_life" }, - { NET_SCTP_ASSOCIATION_MAX_RETRANS, "association_max_retrans" }, - { NET_SCTP_PATH_MAX_RETRANS, "path_max_retrans" }, - { NET_SCTP_MAX_INIT_RETRANSMITS, "max_init_retransmits" }, - { NET_SCTP_HB_INTERVAL, "hb_interval" }, - { NET_SCTP_PRESERVE_ENABLE, "cookie_preserve_enable" }, - { NET_SCTP_MAX_BURST, "max_burst" }, - { NET_SCTP_ADDIP_ENABLE, "addip_enable" }, - { NET_SCTP_PRSCTP_ENABLE, "prsctp_enable" }, - { NET_SCTP_SNDBUF_POLICY, "sndbuf_policy" }, - { NET_SCTP_SACK_TIMEOUT, "sack_timeout" }, - { NET_SCTP_RCVBUF_POLICY, "rcvbuf_policy" }, - {} -}; - -static const struct trans_ctl_table trans_net_llc_llc2_timeout_table[] = { - { NET_LLC2_ACK_TIMEOUT, "ack" }, - { NET_LLC2_P_TIMEOUT, "p" }, - { NET_LLC2_REJ_TIMEOUT, "rej" }, - { NET_LLC2_BUSY_TIMEOUT, "busy" }, - {} -}; - -static const struct trans_ctl_table trans_net_llc_station_table[] = { - { NET_LLC_STATION_ACK_TIMEOUT, "ack_timeout" }, - {} -}; - -static const struct trans_ctl_table trans_net_llc_llc2_table[] = { - { NET_LLC2, "timeout", trans_net_llc_llc2_timeout_table }, - {} -}; - -static const struct trans_ctl_table trans_net_llc_table[] = { - { NET_LLC2, "llc2", trans_net_llc_llc2_table }, - { NET_LLC_STATION, "station", trans_net_llc_station_table }, - {} -}; - -static const struct trans_ctl_table trans_net_netfilter_table[] = { - { NET_NF_CONNTRACK_MAX, "nf_conntrack_max" }, - { NET_NF_CONNTRACK_TCP_TIMEOUT_SYN_SENT, "nf_conntrack_tcp_timeout_syn_sent" }, - { NET_NF_CONNTRACK_TCP_TIMEOUT_SYN_RECV, "nf_conntrack_tcp_timeout_syn_recv" }, - { NET_NF_CONNTRACK_TCP_TIMEOUT_ESTABLISHED, "nf_conntrack_tcp_timeout_established" }, - { NET_NF_CONNTRACK_TCP_TIMEOUT_FIN_WAIT, "nf_conntrack_tcp_timeout_fin_wait" }, - { NET_NF_CONNTRACK_TCP_TIMEOUT_CLOSE_WAIT, "nf_conntrack_tcp_timeout_close_wait" }, - { NET_NF_CONNTRACK_TCP_TIMEOUT_LAST_ACK, "nf_conntrack_tcp_timeout_last_ack" }, - { NET_NF_CONNTRACK_TCP_TIMEOUT_TIME_WAIT, "nf_conntrack_tcp_timeout_time_wait" }, - { NET_NF_CONNTRACK_TCP_TIMEOUT_CLOSE, "nf_conntrack_tcp_timeout_close" }, - { NET_NF_CONNTRACK_UDP_TIMEOUT, "nf_conntrack_udp_timeout" }, - { NET_NF_CONNTRACK_UDP_TIMEOUT_STREAM, "nf_conntrack_udp_timeout_stream" }, - { NET_NF_CONNTRACK_ICMP_TIMEOUT, "nf_conntrack_icmp_timeout" }, - { NET_NF_CONNTRACK_GENERIC_TIMEOUT, "nf_conntrack_generic_timeout" }, - { NET_NF_CONNTRACK_BUCKETS, "nf_conntrack_buckets" }, - { NET_NF_CONNTRACK_LOG_INVALID, "nf_conntrack_log_invalid" }, - { NET_NF_CONNTRACK_TCP_TIMEOUT_MAX_RETRANS, "nf_conntrack_tcp_timeout_max_retrans" }, - { NET_NF_CONNTRACK_TCP_LOOSE, "nf_conntrack_tcp_loose" }, - { NET_NF_CONNTRACK_TCP_BE_LIBERAL, "nf_conntrack_tcp_be_liberal" }, - { NET_NF_CONNTRACK_TCP_MAX_RETRANS, "nf_conntrack_tcp_max_retrans" }, - { NET_NF_CONNTRACK_SCTP_TIMEOUT_CLOSED, "nf_conntrack_sctp_timeout_closed" }, - { NET_NF_CONNTRACK_SCTP_TIMEOUT_COOKIE_WAIT, "nf_conntrack_sctp_timeout_cookie_wait" }, - { NET_NF_CONNTRACK_SCTP_TIMEOUT_COOKIE_ECHOED, "nf_conntrack_sctp_timeout_cookie_echoed" }, - { NET_NF_CONNTRACK_SCTP_TIMEOUT_ESTABLISHED, "nf_conntrack_sctp_timeout_established" }, - { NET_NF_CONNTRACK_SCTP_TIMEOUT_SHUTDOWN_SENT, "nf_conntrack_sctp_timeout_shutdown_sent" }, - { NET_NF_CONNTRACK_SCTP_TIMEOUT_SHUTDOWN_RECD, "nf_conntrack_sctp_timeout_shutdown_recd" }, - { NET_NF_CONNTRACK_SCTP_TIMEOUT_SHUTDOWN_ACK_SENT, "nf_conntrack_sctp_timeout_shutdown_ack_sent" }, - { NET_NF_CONNTRACK_COUNT, "nf_conntrack_count" }, - { NET_NF_CONNTRACK_ICMPV6_TIMEOUT, "nf_conntrack_icmpv6_timeout" }, - { NET_NF_CONNTRACK_FRAG6_TIMEOUT, "nf_conntrack_frag6_timeout" }, - { NET_NF_CONNTRACK_FRAG6_LOW_THRESH, "nf_conntrack_frag6_low_thresh" }, - { NET_NF_CONNTRACK_FRAG6_HIGH_THRESH, "nf_conntrack_frag6_high_thresh" }, - { NET_NF_CONNTRACK_CHECKSUM, "nf_conntrack_checksum" }, - - {} -}; - -static const struct trans_ctl_table trans_net_dccp_table[] = { - { NET_DCCP_DEFAULT, "default" }, - {} -}; - -static const struct trans_ctl_table trans_net_irda_table[] = { - { NET_IRDA_DISCOVERY, "discovery" }, - { NET_IRDA_DEVNAME, "devname" }, - { NET_IRDA_DEBUG, "debug" }, - { NET_IRDA_FAST_POLL, "fast_poll_increase" }, - { NET_IRDA_DISCOVERY_SLOTS, "discovery_slots" }, - { NET_IRDA_DISCOVERY_TIMEOUT, "discovery_timeout" }, - { NET_IRDA_SLOT_TIMEOUT, "slot_timeout" }, - { NET_IRDA_MAX_BAUD_RATE, "max_baud_rate" }, - { NET_IRDA_MIN_TX_TURN_TIME, "min_tx_turn_time" }, - { NET_IRDA_MAX_TX_DATA_SIZE, "max_tx_data_size" }, - { NET_IRDA_MAX_TX_WINDOW, "max_tx_window" }, - { NET_IRDA_MAX_NOREPLY_TIME, "max_noreply_time" }, - { NET_IRDA_WARN_NOREPLY_TIME, "warn_noreply_time" }, - { NET_IRDA_LAP_KEEPALIVE_TIME, "lap_keepalive_time" }, - {} -}; - -static const struct trans_ctl_table trans_net_table[] = { - { NET_CORE, "core", trans_net_core_table }, - /* NET_ETHER not used */ - /* NET_802 not used */ - { NET_UNIX, "unix", trans_net_unix_table }, - { NET_IPV4, "ipv4", trans_net_ipv4_table }, - { NET_IPX, "ipx", trans_net_ipx_table }, - { NET_ATALK, "appletalk", trans_net_atalk_table }, - { NET_NETROM, "netrom", trans_net_netrom_table }, - { NET_AX25, "ax25", trans_net_ax25_table }, - { NET_BRIDGE, "bridge", trans_net_bridge_table }, - { NET_ROSE, "rose", trans_net_rose_table }, - { NET_IPV6, "ipv6", trans_net_ipv6_table }, - { NET_X25, "x25", trans_net_x25_table }, - { NET_TR, "token-ring", trans_net_tr_table }, - { NET_DECNET, "decnet", trans_net_decnet_table }, - /* NET_ECONET not used */ - { NET_SCTP, "sctp", trans_net_sctp_table }, - { NET_LLC, "llc", trans_net_llc_table }, - { NET_NETFILTER, "netfilter", trans_net_netfilter_table }, - { NET_DCCP, "dccp", trans_net_dccp_table }, - { NET_IRDA, "irda", trans_net_irda_table }, - { 2089, "nf_conntrack_max" }, - {} -}; - -static const struct trans_ctl_table trans_fs_quota_table[] = { - { FS_DQ_LOOKUPS, "lookups" }, - { FS_DQ_DROPS, "drops" }, - { FS_DQ_READS, "reads" }, - { FS_DQ_WRITES, "writes" }, - { FS_DQ_CACHE_HITS, "cache_hits" }, - { FS_DQ_ALLOCATED, "allocated_dquots" }, - { FS_DQ_FREE, "free_dquots" }, - { FS_DQ_SYNCS, "syncs" }, - { FS_DQ_WARNINGS, "warnings" }, - {} -}; - -static const struct trans_ctl_table trans_fs_xfs_table[] = { - { XFS_SGID_INHERIT, "irix_sgid_inherit" }, - { XFS_SYMLINK_MODE, "irix_symlink_mode" }, - { XFS_PANIC_MASK, "panic_mask" }, - - { XFS_ERRLEVEL, "error_level" }, - { XFS_SYNCD_TIMER, "xfssyncd_centisecs" }, - { XFS_INHERIT_SYNC, "inherit_sync" }, - { XFS_INHERIT_NODUMP, "inherit_nodump" }, - { XFS_INHERIT_NOATIME, "inherit_noatime" }, - { XFS_BUF_TIMER, "xfsbufd_centisecs" }, - { XFS_BUF_AGE, "age_buffer_centisecs" }, - { XFS_INHERIT_NOSYM, "inherit_nosymlinks" }, - { XFS_ROTORSTEP, "rotorstep" }, - { XFS_INHERIT_NODFRG, "inherit_nodefrag" }, - { XFS_FILESTREAM_TIMER, "filestream_centisecs" }, - { XFS_STATS_CLEAR, "stats_clear" }, - {} -}; - -static const struct trans_ctl_table trans_fs_ocfs2_nm_table[] = { - { 1, "hb_ctl_path" }, - {} -}; - -static const struct trans_ctl_table trans_fs_ocfs2_table[] = { - { 1, "nm", trans_fs_ocfs2_nm_table }, - {} -}; - -static const struct trans_ctl_table trans_inotify_table[] = { - { INOTIFY_MAX_USER_INSTANCES, "max_user_instances" }, - { INOTIFY_MAX_USER_WATCHES, "max_user_watches" }, - { INOTIFY_MAX_QUEUED_EVENTS, "max_queued_events" }, - {} -}; - -static const struct trans_ctl_table trans_fs_table[] = { - { FS_NRINODE, "inode-nr" }, - { FS_STATINODE, "inode-state" }, - /* FS_MAXINODE unused */ - /* FS_NRDQUOT unused */ - /* FS_MAXDQUOT unused */ - { FS_NRFILE, "file-nr" }, - { FS_MAXFILE, "file-max" }, - { FS_DENTRY, "dentry-state" }, - /* FS_NRSUPER unused */ - /* FS_MAXUPSER unused */ - { FS_OVERFLOWUID, "overflowuid" }, - { FS_OVERFLOWGID, "overflowgid" }, - { FS_LEASES, "leases-enable" }, - { FS_DIR_NOTIFY, "dir-notify-enable" }, - { FS_LEASE_TIME, "lease-break-time" }, - { FS_DQSTATS, "quota", trans_fs_quota_table }, - { FS_XFS, "xfs", trans_fs_xfs_table }, - { FS_AIO_NR, "aio-nr" }, - { FS_AIO_MAX_NR, "aio-max-nr" }, - { FS_INOTIFY, "inotify", trans_inotify_table }, - { FS_OCFS2, "ocfs2", trans_fs_ocfs2_table }, - { KERN_SETUID_DUMPABLE, "suid_dumpable" }, - {} -}; - -static const struct trans_ctl_table trans_debug_table[] = { - {} -}; - -static const struct trans_ctl_table trans_cdrom_table[] = { - { DEV_CDROM_INFO, "info" }, - { DEV_CDROM_AUTOCLOSE, "autoclose" }, - { DEV_CDROM_AUTOEJECT, "autoeject" }, - { DEV_CDROM_DEBUG, "debug" }, - { DEV_CDROM_LOCK, "lock" }, - { DEV_CDROM_CHECK_MEDIA, "check_media" }, - {} -}; - -static const struct trans_ctl_table trans_ipmi_table[] = { - { DEV_IPMI_POWEROFF_POWERCYCLE, "poweroff_powercycle" }, - {} -}; - -static const struct trans_ctl_table trans_mac_hid_files[] = { - /* DEV_MAC_HID_KEYBOARD_SENDS_LINUX_KEYCODES unused */ - /* DEV_MAC_HID_KEYBOARD_LOCK_KEYCODES unused */ - { DEV_MAC_HID_MOUSE_BUTTON_EMULATION, "mouse_button_emulation" }, - { DEV_MAC_HID_MOUSE_BUTTON2_KEYCODE, "mouse_button2_keycode" }, - { DEV_MAC_HID_MOUSE_BUTTON3_KEYCODE, "mouse_button3_keycode" }, - /* DEV_MAC_HID_ADB_MOUSE_SENDS_KEYCODES unused */ - {} -}; - -static const struct trans_ctl_table trans_raid_table[] = { - { DEV_RAID_SPEED_LIMIT_MIN, "speed_limit_min" }, - { DEV_RAID_SPEED_LIMIT_MAX, "speed_limit_max" }, - {} -}; - -static const struct trans_ctl_table trans_scsi_table[] = { - { DEV_SCSI_LOGGING_LEVEL, "logging_level" }, - {} -}; - -static const struct trans_ctl_table trans_parport_default_table[] = { - { DEV_PARPORT_DEFAULT_TIMESLICE, "timeslice" }, - { DEV_PARPORT_DEFAULT_SPINTIME, "spintime" }, - {} -}; - -static const struct trans_ctl_table trans_parport_device_table[] = { - { DEV_PARPORT_DEVICE_TIMESLICE, "timeslice" }, - {} -}; - -static const struct trans_ctl_table trans_parport_devices_table[] = { - { DEV_PARPORT_DEVICES_ACTIVE, "active" }, - { 0, NULL, trans_parport_device_table }, - {} -}; - -static const struct trans_ctl_table trans_parport_parport_table[] = { - { DEV_PARPORT_SPINTIME, "spintime" }, - { DEV_PARPORT_BASE_ADDR, "base-addr" }, - { DEV_PARPORT_IRQ, "irq" }, - { DEV_PARPORT_DMA, "dma" }, - { DEV_PARPORT_MODES, "modes" }, - { DEV_PARPORT_DEVICES, "devices", trans_parport_devices_table }, - { DEV_PARPORT_AUTOPROBE, "autoprobe" }, - { DEV_PARPORT_AUTOPROBE + 1, "autoprobe0" }, - { DEV_PARPORT_AUTOPROBE + 2, "autoprobe1" }, - { DEV_PARPORT_AUTOPROBE + 3, "autoprobe2" }, - { DEV_PARPORT_AUTOPROBE + 4, "autoprobe3" }, - {} -}; -static const struct trans_ctl_table trans_parport_table[] = { - { DEV_PARPORT_DEFAULT, "default", trans_parport_default_table }, - { 0, NULL, trans_parport_parport_table }, - {} -}; - -static const struct trans_ctl_table trans_dev_table[] = { - { DEV_CDROM, "cdrom", trans_cdrom_table }, - /* DEV_HWMON unused */ - { DEV_PARPORT, "parport", trans_parport_table }, - { DEV_RAID, "raid", trans_raid_table }, - { DEV_MAC_HID, "mac_hid", trans_mac_hid_files }, - { DEV_SCSI, "scsi", trans_scsi_table }, - { DEV_IPMI, "ipmi", trans_ipmi_table }, - {} -}; - -static const struct trans_ctl_table trans_bus_isa_table[] = { - { BUS_ISA_MEM_BASE, "membase" }, - { BUS_ISA_PORT_BASE, "portbase" }, - { BUS_ISA_PORT_SHIFT, "portshift" }, - {} -}; - -static const struct trans_ctl_table trans_bus_table[] = { - { CTL_BUS_ISA, "isa", trans_bus_isa_table }, - {} -}; - -static const struct trans_ctl_table trans_arlan_conf_table0[] = { - { 1, "spreadingCode" }, - { 2, "channelNumber" }, - { 3, "scramblingDisable" }, - { 4, "txAttenuation" }, - { 5, "systemId" }, - { 6, "maxDatagramSize" }, - { 7, "maxFrameSize" }, - { 8, "maxRetries" }, - { 9, "receiveMode" }, - { 10, "priority" }, - { 11, "rootOrRepeater" }, - { 12, "SID" }, - { 13, "registrationMode" }, - { 14, "registrationFill" }, - { 15, "localTalkAddress" }, - { 16, "codeFormat" }, - { 17, "numChannels" }, - { 18, "channel1" }, - { 19, "channel2" }, - { 20, "channel3" }, - { 21, "channel4" }, - { 22, "txClear" }, - { 23, "txRetries" }, - { 24, "txRouting" }, - { 25, "txScrambled" }, - { 26, "rxParameter" }, - { 27, "txTimeoutMs" }, - { 28, "waitCardTimeout" }, - { 29, "channelSet" }, - { 30, "name" }, - { 31, "waitTime" }, - { 32, "lParameter" }, - { 33, "_15" }, - { 34, "headerSize" }, - { 36, "tx_delay_ms" }, - { 37, "retries" }, - { 38, "ReTransmitPacketMaxSize" }, - { 39, "waitReTransmitPacketMaxSize" }, - { 40, "fastReTransCount" }, - { 41, "driverRetransmissions" }, - { 42, "txAckTimeoutMs" }, - { 43, "registrationInterrupts" }, - { 44, "hardwareType" }, - { 45, "radioType" }, - { 46, "writeEEPROM" }, - { 47, "writeRadioType" }, - { 48, "entry_exit_debug" }, - { 49, "debug" }, - { 50, "in_speed" }, - { 51, "out_speed" }, - { 52, "in_speed10" }, - { 53, "out_speed10" }, - { 54, "in_speed_max" }, - { 55, "out_speed_max" }, - { 56, "measure_rate" }, - { 57, "pre_Command_Wait" }, - { 58, "rx_tweak1" }, - { 59, "rx_tweak2" }, - { 60, "tx_queue_len" }, - - { 150, "arlan0-txRing" }, - { 151, "arlan0-rxRing" }, - { 152, "arlan0-18" }, - { 153, "arlan0-ring" }, - { 154, "arlan0-shm-cpy" }, - { 155, "config0" }, - { 156, "reset0" }, - {} -}; - -static const struct trans_ctl_table trans_arlan_conf_table1[] = { - { 1, "spreadingCode" }, - { 2, "channelNumber" }, - { 3, "scramblingDisable" }, - { 4, "txAttenuation" }, - { 5, "systemId" }, - { 6, "maxDatagramSize" }, - { 7, "maxFrameSize" }, - { 8, "maxRetries" }, - { 9, "receiveMode" }, - { 10, "priority" }, - { 11, "rootOrRepeater" }, - { 12, "SID" }, - { 13, "registrationMode" }, - { 14, "registrationFill" }, - { 15, "localTalkAddress" }, - { 16, "codeFormat" }, - { 17, "numChannels" }, - { 18, "channel1" }, - { 19, "channel2" }, - { 20, "channel3" }, - { 21, "channel4" }, - { 22, "txClear" }, - { 23, "txRetries" }, - { 24, "txRouting" }, - { 25, "txScrambled" }, - { 26, "rxParameter" }, - { 27, "txTimeoutMs" }, - { 28, "waitCardTimeout" }, - { 29, "channelSet" }, - { 30, "name" }, - { 31, "waitTime" }, - { 32, "lParameter" }, - { 33, "_15" }, - { 34, "headerSize" }, - { 36, "tx_delay_ms" }, - { 37, "retries" }, - { 38, "ReTransmitPacketMaxSize" }, - { 39, "waitReTransmitPacketMaxSize" }, - { 40, "fastReTransCount" }, - { 41, "driverRetransmissions" }, - { 42, "txAckTimeoutMs" }, - { 43, "registrationInterrupts" }, - { 44, "hardwareType" }, - { 45, "radioType" }, - { 46, "writeEEPROM" }, - { 47, "writeRadioType" }, - { 48, "entry_exit_debug" }, - { 49, "debug" }, - { 50, "in_speed" }, - { 51, "out_speed" }, - { 52, "in_speed10" }, - { 53, "out_speed10" }, - { 54, "in_speed_max" }, - { 55, "out_speed_max" }, - { 56, "measure_rate" }, - { 57, "pre_Command_Wait" }, - { 58, "rx_tweak1" }, - { 59, "rx_tweak2" }, - { 60, "tx_queue_len" }, - - { 150, "arlan1-txRing" }, - { 151, "arlan1-rxRing" }, - { 152, "arlan1-18" }, - { 153, "arlan1-ring" }, - { 154, "arlan1-shm-cpy" }, - { 155, "config1" }, - { 156, "reset1" }, - {} -}; - -static const struct trans_ctl_table trans_arlan_conf_table2[] = { - { 1, "spreadingCode" }, - { 2, "channelNumber" }, - { 3, "scramblingDisable" }, - { 4, "txAttenuation" }, - { 5, "systemId" }, - { 6, "maxDatagramSize" }, - { 7, "maxFrameSize" }, - { 8, "maxRetries" }, - { 9, "receiveMode" }, - { 10, "priority" }, - { 11, "rootOrRepeater" }, - { 12, "SID" }, - { 13, "registrationMode" }, - { 14, "registrationFill" }, - { 15, "localTalkAddress" }, - { 16, "codeFormat" }, - { 17, "numChannels" }, - { 18, "channel1" }, - { 19, "channel2" }, - { 20, "channel3" }, - { 21, "channel4" }, - { 22, "txClear" }, - { 23, "txRetries" }, - { 24, "txRouting" }, - { 25, "txScrambled" }, - { 26, "rxParameter" }, - { 27, "txTimeoutMs" }, - { 28, "waitCardTimeout" }, - { 29, "channelSet" }, - { 30, "name" }, - { 31, "waitTime" }, - { 32, "lParameter" }, - { 33, "_15" }, - { 34, "headerSize" }, - { 36, "tx_delay_ms" }, - { 37, "retries" }, - { 38, "ReTransmitPacketMaxSize" }, - { 39, "waitReTransmitPacketMaxSize" }, - { 40, "fastReTransCount" }, - { 41, "driverRetransmissions" }, - { 42, "txAckTimeoutMs" }, - { 43, "registrationInterrupts" }, - { 44, "hardwareType" }, - { 45, "radioType" }, - { 46, "writeEEPROM" }, - { 47, "writeRadioType" }, - { 48, "entry_exit_debug" }, - { 49, "debug" }, - { 50, "in_speed" }, - { 51, "out_speed" }, - { 52, "in_speed10" }, - { 53, "out_speed10" }, - { 54, "in_speed_max" }, - { 55, "out_speed_max" }, - { 56, "measure_rate" }, - { 57, "pre_Command_Wait" }, - { 58, "rx_tweak1" }, - { 59, "rx_tweak2" }, - { 60, "tx_queue_len" }, - - { 150, "arlan2-txRing" }, - { 151, "arlan2-rxRing" }, - { 152, "arlan2-18" }, - { 153, "arlan2-ring" }, - { 154, "arlan2-shm-cpy" }, - { 155, "config2" }, - { 156, "reset2" }, - {} -}; - -static const struct trans_ctl_table trans_arlan_conf_table3[] = { - { 1, "spreadingCode" }, - { 2, "channelNumber" }, - { 3, "scramblingDisable" }, - { 4, "txAttenuation" }, - { 5, "systemId" }, - { 6, "maxDatagramSize" }, - { 7, "maxFrameSize" }, - { 8, "maxRetries" }, - { 9, "receiveMode" }, - { 10, "priority" }, - { 11, "rootOrRepeater" }, - { 12, "SID" }, - { 13, "registrationMode" }, - { 14, "registrationFill" }, - { 15, "localTalkAddress" }, - { 16, "codeFormat" }, - { 17, "numChannels" }, - { 18, "channel1" }, - { 19, "channel2" }, - { 20, "channel3" }, - { 21, "channel4" }, - { 22, "txClear" }, - { 23, "txRetries" }, - { 24, "txRouting" }, - { 25, "txScrambled" }, - { 26, "rxParameter" }, - { 27, "txTimeoutMs" }, - { 28, "waitCardTimeout" }, - { 29, "channelSet" }, - { 30, "name" }, - { 31, "waitTime" }, - { 32, "lParameter" }, - { 33, "_15" }, - { 34, "headerSize" }, - { 36, "tx_delay_ms" }, - { 37, "retries" }, - { 38, "ReTransmitPacketMaxSize" }, - { 39, "waitReTransmitPacketMaxSize" }, - { 40, "fastReTransCount" }, - { 41, "driverRetransmissions" }, - { 42, "txAckTimeoutMs" }, - { 43, "registrationInterrupts" }, - { 44, "hardwareType" }, - { 45, "radioType" }, - { 46, "writeEEPROM" }, - { 47, "writeRadioType" }, - { 48, "entry_exit_debug" }, - { 49, "debug" }, - { 50, "in_speed" }, - { 51, "out_speed" }, - { 52, "in_speed10" }, - { 53, "out_speed10" }, - { 54, "in_speed_max" }, - { 55, "out_speed_max" }, - { 56, "measure_rate" }, - { 57, "pre_Command_Wait" }, - { 58, "rx_tweak1" }, - { 59, "rx_tweak2" }, - { 60, "tx_queue_len" }, - - { 150, "arlan3-txRing" }, - { 151, "arlan3-rxRing" }, - { 152, "arlan3-18" }, - { 153, "arlan3-ring" }, - { 154, "arlan3-shm-cpy" }, - { 155, "config3" }, - { 156, "reset3" }, - {} -}; - -static const struct trans_ctl_table trans_arlan_table[] = { - { 1, "arlan0", trans_arlan_conf_table0 }, - { 2, "arlan1", trans_arlan_conf_table1 }, - { 3, "arlan2", trans_arlan_conf_table2 }, - { 4, "arlan3", trans_arlan_conf_table3 }, - {} -}; - -static const struct trans_ctl_table trans_s390dbf_table[] = { - { 5678 /* CTL_S390DBF_STOPPABLE */, "debug_stoppable" }, - { 5679 /* CTL_S390DBF_ACTIVE */, "debug_active" }, - {} -}; - -static const struct trans_ctl_table trans_sunrpc_table[] = { - { CTL_RPCDEBUG, "rpc_debug" }, - { CTL_NFSDEBUG, "nfs_debug" }, - { CTL_NFSDDEBUG, "nfsd_debug" }, - { CTL_NLMDEBUG, "nlm_debug" }, - { CTL_SLOTTABLE_UDP, "udp_slot_table_entries" }, - { CTL_SLOTTABLE_TCP, "tcp_slot_table_entries" }, - { CTL_MIN_RESVPORT, "min_resvport" }, - { CTL_MAX_RESVPORT, "max_resvport" }, - {} -}; - -static const struct trans_ctl_table trans_pm_table[] = { - { 1 /* CTL_PM_SUSPEND */, "suspend" }, - { 2 /* CTL_PM_CMODE */, "cmode" }, - { 3 /* CTL_PM_P0 */, "p0" }, - { 4 /* CTL_PM_CM */, "cm" }, - {} -}; - -static const struct trans_ctl_table trans_frv_table[] = { - { 1, "cache-mode" }, - { 2, "pin-cxnr" }, - {} -}; - -static const struct trans_ctl_table trans_root_table[] = { - { CTL_KERN, "kernel", trans_kern_table }, - { CTL_VM, "vm", trans_vm_table }, - { CTL_NET, "net", trans_net_table }, - /* CTL_PROC not used */ - { CTL_FS, "fs", trans_fs_table }, - { CTL_DEBUG, "debug", trans_debug_table }, - { CTL_DEV, "dev", trans_dev_table }, - { CTL_BUS, "bus", trans_bus_table }, - { CTL_ABI, "abi" }, - /* CTL_CPU not used */ - { CTL_ARLAN, "arlan", trans_arlan_table }, - { CTL_S390DBF, "s390dbf", trans_s390dbf_table }, - { CTL_SUNRPC, "sunrpc", trans_sunrpc_table }, - { CTL_PM, "pm", trans_pm_table }, - { CTL_FRV, "frv", trans_frv_table }, - {} -}; - - - static int sysctl_depth(struct ctl_table *table) { @@ -1261,47 +28,6 @@ static struct ctl_table *sysctl_parent(struct ctl_table *table, int n) return table; } -static const struct trans_ctl_table *sysctl_binary_lookup(struct ctl_table *table) -{ - struct ctl_table *test; - const struct trans_ctl_table *ref; - int cur_depth; - - cur_depth = sysctl_depth(table); - - ref = trans_root_table; -repeat: - test = sysctl_parent(table, cur_depth); - for (; ref->ctl_name || ref->procname || ref->child; ref++) { - int match = 0; - - if (cur_depth && !ref->child) - continue; - - if (test->procname && ref->procname && - (strcmp(test->procname, ref->procname) == 0)) - match++; - - if (test->ctl_name && ref->ctl_name && - (test->ctl_name == ref->ctl_name)) - match++; - - if (!ref->ctl_name && !ref->procname) - match++; - - if (match) { - if (cur_depth != 0) { - cur_depth--; - ref = ref->child; - goto repeat; - } - goto out; - } - } - ref = NULL; -out: - return ref; -} static void sysctl_print_path(struct ctl_table *table) { @@ -1315,26 +41,6 @@ static void sysctl_print_path(struct ctl_table *table) } } printk(" "); - if (table->ctl_name) { - for (i = depth; i >= 0; i--) { - tmp = sysctl_parent(table, i); - printk(".%d", tmp->ctl_name); - } - } -} - -static void sysctl_repair_table(struct ctl_table *table) -{ - /* Don't complain about the classic default - * sysctl strategy routine. Maybe later we - * can get the tables fixed and complain about - * this. - */ - if (table->ctl_name && table->procname && - (table->proc_handler == proc_dointvec) && - (!table->strategy)) { - table->strategy = sysctl_data; - } } static struct ctl_table *sysctl_check_lookup(struct nsproxy *namespaces, @@ -1352,7 +58,7 @@ static struct ctl_table *sysctl_check_lookup(struct nsproxy *namespaces, ref = head->ctl_table; repeat: test = sysctl_parent(table, cur_depth); - for (; ref->ctl_name || ref->procname; ref++) { + for (; ref->procname; ref++) { int match = 0; if (cur_depth && !ref->child) continue; @@ -1361,10 +67,6 @@ repeat: (strcmp(test->procname, ref->procname) == 0)) match++; - if (test->ctl_name && ref->ctl_name && - (test->ctl_name == ref->ctl_name)) - match++; - if (match) { if (cur_depth != 0) { cur_depth--; @@ -1392,38 +94,6 @@ static void set_fail(const char **fail, struct ctl_table *table, const char *str *fail = str; } -static int sysctl_check_dir(struct nsproxy *namespaces, - struct ctl_table *table) -{ - struct ctl_table *ref; - int error; - - error = 0; - ref = sysctl_check_lookup(namespaces, table); - if (ref) { - int match = 0; - if ((!table->procname && !ref->procname) || - (table->procname && ref->procname && - (strcmp(table->procname, ref->procname) == 0))) - match++; - - if ((!table->ctl_name && !ref->ctl_name) || - (table->ctl_name && ref->ctl_name && - (table->ctl_name == ref->ctl_name))) - match++; - - if (match != 2) { - printk(KERN_ERR "%s: failed: ", __func__); - sysctl_print_path(table); - printk(" ref: "); - sysctl_print_path(ref); - printk("\n"); - error = -EINVAL; - } - } - return error; -} - static void sysctl_check_leaf(struct nsproxy *namespaces, struct ctl_table *table, const char **fail) { @@ -1434,37 +104,15 @@ static void sysctl_check_leaf(struct nsproxy *namespaces, set_fail(fail, table, "Sysctl already exists"); } -static void sysctl_check_bin_path(struct ctl_table *table, const char **fail) -{ - const struct trans_ctl_table *ref; - - ref = sysctl_binary_lookup(table); - if (table->ctl_name && !ref) - set_fail(fail, table, "Unknown sysctl binary path"); - if (ref) { - if (ref->procname && - (!table->procname || - (strcmp(table->procname, ref->procname) != 0))) - set_fail(fail, table, "procname does not match binary path procname"); - - if (ref->ctl_name && table->ctl_name && - (table->ctl_name != ref->ctl_name)) - set_fail(fail, table, "ctl_name does not match binary path ctl_name"); - } -} - int sysctl_check_table(struct nsproxy *namespaces, struct ctl_table *table) { int error = 0; - for (; table->ctl_name || table->procname; table++) { + for (; table->procname; table++) { const char *fail = NULL; - sysctl_repair_table(table); if (table->parent) { if (table->procname && !table->parent->procname) set_fail(&fail, table, "Parent without procname"); - if (table->ctl_name && !table->parent->ctl_name) - set_fail(&fail, table, "Parent without ctl_name"); } if (!table->procname) set_fail(&fail, table, "No procname"); @@ -1477,21 +125,12 @@ int sysctl_check_table(struct nsproxy *namespaces, struct ctl_table *table) set_fail(&fail, table, "Writable sysctl directory"); if (table->proc_handler) set_fail(&fail, table, "Directory with proc_handler"); - if (table->strategy) - set_fail(&fail, table, "Directory with strategy"); if (table->extra1) set_fail(&fail, table, "Directory with extra1"); if (table->extra2) set_fail(&fail, table, "Directory with extra2"); - if (sysctl_check_dir(namespaces, table)) - set_fail(&fail, table, "Inconsistent directory names"); } else { - if ((table->strategy == sysctl_data) || - (table->strategy == sysctl_string) || - (table->strategy == sysctl_intvec) || - (table->strategy == sysctl_jiffies) || - (table->strategy == sysctl_ms_jiffies) || - (table->proc_handler == proc_dostring) || + if ((table->proc_handler == proc_dostring) || (table->proc_handler == proc_dointvec) || (table->proc_handler == proc_dointvec_minmax) || (table->proc_handler == proc_dointvec_jiffies) || @@ -1513,15 +152,7 @@ int sysctl_check_table(struct nsproxy *namespaces, struct ctl_table *table) set_fail(&fail, table, "No max"); } } -#ifdef CONFIG_SYSCTL_SYSCALL - if (table->ctl_name && !table->strategy) - set_fail(&fail, table, "Missing strategy"); -#endif -#if 0 - if (!table->ctl_name && table->strategy) - set_fail(&fail, table, "Strategy without ctl_name"); -#endif -#ifdef CONFIG_PROC_FS +#ifdef CONFIG_PROC_SYSCTL if (table->procname && !table->proc_handler) set_fail(&fail, table, "No proc_handler"); #endif @@ -1531,7 +162,6 @@ int sysctl_check_table(struct nsproxy *namespaces, struct ctl_table *table) #endif sysctl_check_leaf(namespaces, table, &fail); } - sysctl_check_bin_path(table, &fail); if (table->mode > 0777) set_fail(&fail, table, "bogus .mode"); if (fail) { diff --git a/kernel/taskstats.c b/kernel/taskstats.c index 888adbcca30..ea8384d3caa 100644 --- a/kernel/taskstats.c +++ b/kernel/taskstats.c @@ -108,7 +108,7 @@ static int prepare_reply(struct genl_info *info, u8 cmd, struct sk_buff **skbp, /* * Send taskstats data in @skb to listener with nl_pid @pid */ -static int send_reply(struct sk_buff *skb, pid_t pid) +static int send_reply(struct sk_buff *skb, struct genl_info *info) { struct genlmsghdr *genlhdr = nlmsg_data(nlmsg_hdr(skb)); void *reply = genlmsg_data(genlhdr); @@ -120,7 +120,7 @@ static int send_reply(struct sk_buff *skb, pid_t pid) return rc; } - return genlmsg_unicast(skb, pid); + return genlmsg_reply(skb, info); } /* @@ -150,7 +150,7 @@ static void send_cpu_listeners(struct sk_buff *skb, if (!skb_next) break; } - rc = genlmsg_unicast(skb_cur, s->pid); + rc = genlmsg_unicast(&init_net, skb_cur, s->pid); if (rc == -ECONNREFUSED) { s->valid = 0; delcount++; @@ -418,7 +418,7 @@ static int cgroupstats_user_cmd(struct sk_buff *skb, struct genl_info *info) goto err; } - rc = send_reply(rep_skb, info->snd_pid); + rc = send_reply(rep_skb, info); err: fput_light(file, fput_needed); @@ -487,7 +487,7 @@ free_return_rc: } else goto err; - return send_reply(rep_skb, info->snd_pid); + return send_reply(rep_skb, info); err: nlmsg_free(rep_skb); return rc; diff --git a/kernel/time.c b/kernel/time.c index 29511943871..c6324d96009 100644 --- a/kernel/time.c +++ b/kernel/time.c @@ -136,7 +136,6 @@ static inline void warp_clock(void) write_seqlock_irq(&xtime_lock); wall_to_monotonic.tv_sec -= sys_tz.tz_minuteswest * 60; xtime.tv_sec += sys_tz.tz_minuteswest * 60; - update_xtime_cache(0); write_sequnlock_irq(&xtime_lock); clock_was_set(); } @@ -370,13 +369,20 @@ EXPORT_SYMBOL(mktime); * 0 <= tv_nsec < NSEC_PER_SEC * For negative values only the tv_sec field is negative ! */ -void set_normalized_timespec(struct timespec *ts, time_t sec, long nsec) +void set_normalized_timespec(struct timespec *ts, time_t sec, s64 nsec) { while (nsec >= NSEC_PER_SEC) { + /* + * The following asm() prevents the compiler from + * optimising this loop into a modulo operation. See + * also __iter_div_u64_rem() in include/linux/time.h + */ + asm("" : "+rm"(nsec)); nsec -= NSEC_PER_SEC; ++sec; } while (nsec < 0) { + asm("" : "+rm"(nsec)); nsec += NSEC_PER_SEC; --sec; } @@ -655,6 +661,36 @@ u64 nsec_to_clock_t(u64 x) #endif } +/** + * nsecs_to_jiffies - Convert nsecs in u64 to jiffies + * + * @n: nsecs in u64 + * + * Unlike {m,u}secs_to_jiffies, type of input is not unsigned int but u64. + * And this doesn't return MAX_JIFFY_OFFSET since this function is designed + * for scheduler, not for use in device drivers to calculate timeout value. + * + * note: + * NSEC_PER_SEC = 10^9 = (5^9 * 2^9) = (1953125 * 512) + * ULLONG_MAX ns = 18446744073.709551615 secs = about 584 years + */ +unsigned long nsecs_to_jiffies(u64 n) +{ +#if (NSEC_PER_SEC % HZ) == 0 + /* Common case, HZ = 100, 128, 200, 250, 256, 500, 512, 1000 etc. */ + return div_u64(n, NSEC_PER_SEC / HZ); +#elif (HZ % 512) == 0 + /* overflow after 292 years if HZ = 1024 */ + return div_u64(n * HZ / 512, NSEC_PER_SEC / 512); +#else + /* + * Generic case - optimized for cases where HZ is a multiple of 3. + * overflow after 64.99 years, exact for HZ = 60, 72, 90, 120 etc. + */ + return div_u64(n * 9, (9ull * NSEC_PER_SEC + HZ / 2) / HZ); +#endif +} + #if (BITS_PER_LONG < 64) u64 get_jiffies_64(void) { diff --git a/kernel/time/Makefile b/kernel/time/Makefile index 0b0a6366c9d..ee266620b06 100644 --- a/kernel/time/Makefile +++ b/kernel/time/Makefile @@ -1,4 +1,4 @@ -obj-y += timekeeping.o ntp.o clocksource.o jiffies.o timer_list.o timecompare.o +obj-y += timekeeping.o ntp.o clocksource.o jiffies.o timer_list.o timecompare.o timeconv.o obj-$(CONFIG_GENERIC_CLOCKEVENTS_BUILD) += clockevents.o obj-$(CONFIG_GENERIC_CLOCKEVENTS) += tick-common.o diff --git a/kernel/time/clockevents.c b/kernel/time/clockevents.c index d13be216a79..20a8920029e 100644 --- a/kernel/time/clockevents.c +++ b/kernel/time/clockevents.c @@ -18,6 +18,9 @@ #include <linux/notifier.h> #include <linux/smp.h> #include <linux/sysdev.h> +#include <linux/tick.h> + +#include "tick-internal.h" /* The registered clock event devices */ static LIST_HEAD(clockevent_devices); @@ -36,10 +39,9 @@ static DEFINE_SPINLOCK(clockevents_lock); * * Math helper, returns latch value converted to nanoseconds (bound checked) */ -unsigned long clockevent_delta2ns(unsigned long latch, - struct clock_event_device *evt) +u64 clockevent_delta2ns(unsigned long latch, struct clock_event_device *evt) { - u64 clc = ((u64) latch << evt->shift); + u64 clc = (u64) latch << evt->shift; if (unlikely(!evt->mult)) { evt->mult = 1; @@ -49,11 +51,12 @@ unsigned long clockevent_delta2ns(unsigned long latch, do_div(clc, evt->mult); if (clc < 1000) clc = 1000; - if (clc > LONG_MAX) - clc = LONG_MAX; + if (clc > KTIME_MAX) + clc = KTIME_MAX; - return (unsigned long) clc; + return clc; } +EXPORT_SYMBOL_GPL(clockevent_delta2ns); /** * clockevents_set_mode - set the operating mode of a clock event device @@ -135,11 +138,12 @@ int clockevents_program_event(struct clock_event_device *dev, ktime_t expires, */ int clockevents_register_notifier(struct notifier_block *nb) { + unsigned long flags; int ret; - spin_lock(&clockevents_lock); + spin_lock_irqsave(&clockevents_lock, flags); ret = raw_notifier_chain_register(&clockevents_chain, nb); - spin_unlock(&clockevents_lock); + spin_unlock_irqrestore(&clockevents_lock, flags); return ret; } @@ -176,17 +180,20 @@ static void clockevents_notify_released(void) */ void clockevents_register_device(struct clock_event_device *dev) { + unsigned long flags; + BUG_ON(dev->mode != CLOCK_EVT_MODE_UNUSED); BUG_ON(!dev->cpumask); - spin_lock(&clockevents_lock); + spin_lock_irqsave(&clockevents_lock, flags); list_add(&dev->list, &clockevent_devices); clockevents_do_notify(CLOCK_EVT_NOTIFY_ADD, dev); clockevents_notify_released(); - spin_unlock(&clockevents_lock); + spin_unlock_irqrestore(&clockevents_lock, flags); } +EXPORT_SYMBOL_GPL(clockevents_register_device); /* * Noop handler when we shut down an event device @@ -232,8 +239,9 @@ void clockevents_exchange_device(struct clock_event_device *old, void clockevents_notify(unsigned long reason, void *arg) { struct list_head *node, *tmp; + unsigned long flags; - spin_lock(&clockevents_lock); + spin_lock_irqsave(&clockevents_lock, flags); clockevents_do_notify(reason, arg); switch (reason) { @@ -248,7 +256,7 @@ void clockevents_notify(unsigned long reason, void *arg) default: break; } - spin_unlock(&clockevents_lock); + spin_unlock_irqrestore(&clockevents_lock, flags); } EXPORT_SYMBOL_GPL(clockevents_notify); #endif diff --git a/kernel/time/clocksource.c b/kernel/time/clocksource.c index c46c931a7fe..d422c7b2236 100644 --- a/kernel/time/clocksource.c +++ b/kernel/time/clocksource.c @@ -21,7 +21,6 @@ * * TODO WishList: * o Allow clocksource drivers to be unregistered - * o get rid of clocksource_jiffies extern */ #include <linux/clocksource.h> @@ -30,6 +29,7 @@ #include <linux/module.h> #include <linux/sched.h> /* for spin_unlock_irq() using preempt_count() m68k */ #include <linux/tick.h> +#include <linux/kthread.h> void timecounter_init(struct timecounter *tc, const struct cyclecounter *cc, @@ -39,7 +39,7 @@ void timecounter_init(struct timecounter *tc, tc->cycle_last = cc->read(cc); tc->nsec = start_tstamp; } -EXPORT_SYMBOL(timecounter_init); +EXPORT_SYMBOL_GPL(timecounter_init); /** * timecounter_read_delta - get nanoseconds since last call of this function @@ -83,7 +83,7 @@ u64 timecounter_read(struct timecounter *tc) return nsec; } -EXPORT_SYMBOL(timecounter_read); +EXPORT_SYMBOL_GPL(timecounter_read); u64 timecounter_cyc2time(struct timecounter *tc, cycle_t cycle_tstamp) @@ -105,52 +105,90 @@ u64 timecounter_cyc2time(struct timecounter *tc, return nsec; } -EXPORT_SYMBOL(timecounter_cyc2time); +EXPORT_SYMBOL_GPL(timecounter_cyc2time); -/* XXX - Would like a better way for initializing curr_clocksource */ -extern struct clocksource clocksource_jiffies; +/** + * clocks_calc_mult_shift - calculate mult/shift factors for scaled math of clocks + * @mult: pointer to mult variable + * @shift: pointer to shift variable + * @from: frequency to convert from + * @to: frequency to convert to + * @minsec: guaranteed runtime conversion range in seconds + * + * The function evaluates the shift/mult pair for the scaled math + * operations of clocksources and clockevents. + * + * @to and @from are frequency values in HZ. For clock sources @to is + * NSEC_PER_SEC == 1GHz and @from is the counter frequency. For clock + * event @to is the counter frequency and @from is NSEC_PER_SEC. + * + * The @minsec conversion range argument controls the time frame in + * seconds which must be covered by the runtime conversion with the + * calculated mult and shift factors. This guarantees that no 64bit + * overflow happens when the input value of the conversion is + * multiplied with the calculated mult factor. Larger ranges may + * reduce the conversion accuracy by chosing smaller mult and shift + * factors. + */ +void +clocks_calc_mult_shift(u32 *mult, u32 *shift, u32 from, u32 to, u32 minsec) +{ + u64 tmp; + u32 sft, sftacc= 32; + + /* + * Calculate the shift factor which is limiting the conversion + * range: + */ + tmp = ((u64)minsec * from) >> 32; + while (tmp) { + tmp >>=1; + sftacc--; + } + + /* + * Find the conversion shift/mult pair which has the best + * accuracy and fits the maxsec conversion range: + */ + for (sft = 32; sft > 0; sft--) { + tmp = (u64) to << sft; + do_div(tmp, from); + if ((tmp >> sftacc) == 0) + break; + } + *mult = tmp; + *shift = sft; +} /*[Clocksource internal variables]--------- * curr_clocksource: - * currently selected clocksource. Initialized to clocksource_jiffies. - * next_clocksource: - * pending next selected clocksource. + * currently selected clocksource. * clocksource_list: * linked list with the registered clocksources - * clocksource_lock: - * protects manipulations to curr_clocksource and next_clocksource - * and the clocksource_list + * clocksource_mutex: + * protects manipulations to curr_clocksource and the clocksource_list * override_name: * Name of the user-specified clocksource. */ -static struct clocksource *curr_clocksource = &clocksource_jiffies; -static struct clocksource *next_clocksource; -static struct clocksource *clocksource_override; +static struct clocksource *curr_clocksource; static LIST_HEAD(clocksource_list); -static DEFINE_SPINLOCK(clocksource_lock); +static DEFINE_MUTEX(clocksource_mutex); static char override_name[32]; static int finished_booting; -/* clocksource_done_booting - Called near the end of core bootup - * - * Hack to avoid lots of clocksource churn at boot time. - * We use fs_initcall because we want this to start before - * device_initcall but after subsys_initcall. - */ -static int __init clocksource_done_booting(void) -{ - finished_booting = 1; - return 0; -} -fs_initcall(clocksource_done_booting); - #ifdef CONFIG_CLOCKSOURCE_WATCHDOG +static void clocksource_watchdog_work(struct work_struct *work); + static LIST_HEAD(watchdog_list); static struct clocksource *watchdog; static struct timer_list watchdog_timer; +static DECLARE_WORK(watchdog_work, clocksource_watchdog_work); static DEFINE_SPINLOCK(watchdog_lock); static cycle_t watchdog_last; -static unsigned long watchdog_resumed; +static int watchdog_running; + +static int clocksource_watchdog_kthread(void *data); +static void __clocksource_change_rating(struct clocksource *cs, int rating); /* * Interval: 0.5sec Threshold: 0.0625s @@ -158,135 +196,249 @@ static unsigned long watchdog_resumed; #define WATCHDOG_INTERVAL (HZ >> 1) #define WATCHDOG_THRESHOLD (NSEC_PER_SEC >> 4) -static void clocksource_ratewd(struct clocksource *cs, int64_t delta) +static void clocksource_watchdog_work(struct work_struct *work) { - if (delta > -WATCHDOG_THRESHOLD && delta < WATCHDOG_THRESHOLD) - return; + /* + * If kthread_run fails the next watchdog scan over the + * watchdog_list will find the unstable clock again. + */ + kthread_run(clocksource_watchdog_kthread, NULL, "kwatchdog"); +} +static void __clocksource_unstable(struct clocksource *cs) +{ + cs->flags &= ~(CLOCK_SOURCE_VALID_FOR_HRES | CLOCK_SOURCE_WATCHDOG); + cs->flags |= CLOCK_SOURCE_UNSTABLE; + if (finished_booting) + schedule_work(&watchdog_work); +} + +static void clocksource_unstable(struct clocksource *cs, int64_t delta) +{ printk(KERN_WARNING "Clocksource %s unstable (delta = %Ld ns)\n", cs->name, delta); - cs->flags &= ~(CLOCK_SOURCE_VALID_FOR_HRES | CLOCK_SOURCE_WATCHDOG); - clocksource_change_rating(cs, 0); - list_del(&cs->wd_list); + __clocksource_unstable(cs); +} + +/** + * clocksource_mark_unstable - mark clocksource unstable via watchdog + * @cs: clocksource to be marked unstable + * + * This function is called instead of clocksource_change_rating from + * cpu hotplug code to avoid a deadlock between the clocksource mutex + * and the cpu hotplug mutex. It defers the update of the clocksource + * to the watchdog thread. + */ +void clocksource_mark_unstable(struct clocksource *cs) +{ + unsigned long flags; + + spin_lock_irqsave(&watchdog_lock, flags); + if (!(cs->flags & CLOCK_SOURCE_UNSTABLE)) { + if (list_empty(&cs->wd_list)) + list_add(&cs->wd_list, &watchdog_list); + __clocksource_unstable(cs); + } + spin_unlock_irqrestore(&watchdog_lock, flags); } static void clocksource_watchdog(unsigned long data) { - struct clocksource *cs, *tmp; + struct clocksource *cs; cycle_t csnow, wdnow; int64_t wd_nsec, cs_nsec; - int resumed; + int next_cpu; spin_lock(&watchdog_lock); + if (!watchdog_running) + goto out; - resumed = test_and_clear_bit(0, &watchdog_resumed); - - wdnow = watchdog->read(); - wd_nsec = cyc2ns(watchdog, (wdnow - watchdog_last) & watchdog->mask); + wdnow = watchdog->read(watchdog); + wd_nsec = clocksource_cyc2ns((wdnow - watchdog_last) & watchdog->mask, + watchdog->mult, watchdog->shift); watchdog_last = wdnow; - list_for_each_entry_safe(cs, tmp, &watchdog_list, wd_list) { - csnow = cs->read(); + list_for_each_entry(cs, &watchdog_list, wd_list) { - if (unlikely(resumed)) { - cs->wd_last = csnow; + /* Clocksource already marked unstable? */ + if (cs->flags & CLOCK_SOURCE_UNSTABLE) { + if (finished_booting) + schedule_work(&watchdog_work); continue; } - /* Initialized ? */ + csnow = cs->read(cs); + + /* Clocksource initialized ? */ if (!(cs->flags & CLOCK_SOURCE_WATCHDOG)) { - if ((cs->flags & CLOCK_SOURCE_IS_CONTINUOUS) && - (watchdog->flags & CLOCK_SOURCE_IS_CONTINUOUS)) { - cs->flags |= CLOCK_SOURCE_VALID_FOR_HRES; - /* - * We just marked the clocksource as - * highres-capable, notify the rest of the - * system as well so that we transition - * into high-res mode: - */ - tick_clock_notify(); - } cs->flags |= CLOCK_SOURCE_WATCHDOG; cs->wd_last = csnow; - } else { - cs_nsec = cyc2ns(cs, (csnow - cs->wd_last) & cs->mask); - cs->wd_last = csnow; - /* Check the delta. Might remove from the list ! */ - clocksource_ratewd(cs, cs_nsec - wd_nsec); + continue; } - } - if (!list_empty(&watchdog_list)) { - /* - * Cycle through CPUs to check if the CPUs stay - * synchronized to each other. - */ - int next_cpu = cpumask_next(raw_smp_processor_id(), - cpu_online_mask); + /* Check the deviation from the watchdog clocksource. */ + cs_nsec = clocksource_cyc2ns((csnow - cs->wd_last) & + cs->mask, cs->mult, cs->shift); + cs->wd_last = csnow; + if (abs(cs_nsec - wd_nsec) > WATCHDOG_THRESHOLD) { + clocksource_unstable(cs, cs_nsec - wd_nsec); + continue; + } - if (next_cpu >= nr_cpu_ids) - next_cpu = cpumask_first(cpu_online_mask); - watchdog_timer.expires += WATCHDOG_INTERVAL; - add_timer_on(&watchdog_timer, next_cpu); + if (!(cs->flags & CLOCK_SOURCE_VALID_FOR_HRES) && + (cs->flags & CLOCK_SOURCE_IS_CONTINUOUS) && + (watchdog->flags & CLOCK_SOURCE_IS_CONTINUOUS)) { + cs->flags |= CLOCK_SOURCE_VALID_FOR_HRES; + /* + * We just marked the clocksource as highres-capable, + * notify the rest of the system as well so that we + * transition into high-res mode: + */ + tick_clock_notify(); + } } + + /* + * Cycle through CPUs to check if the CPUs stay synchronized + * to each other. + */ + next_cpu = cpumask_next(raw_smp_processor_id(), cpu_online_mask); + if (next_cpu >= nr_cpu_ids) + next_cpu = cpumask_first(cpu_online_mask); + watchdog_timer.expires += WATCHDOG_INTERVAL; + add_timer_on(&watchdog_timer, next_cpu); +out: spin_unlock(&watchdog_lock); } + +static inline void clocksource_start_watchdog(void) +{ + if (watchdog_running || !watchdog || list_empty(&watchdog_list)) + return; + init_timer(&watchdog_timer); + watchdog_timer.function = clocksource_watchdog; + watchdog_last = watchdog->read(watchdog); + watchdog_timer.expires = jiffies + WATCHDOG_INTERVAL; + add_timer_on(&watchdog_timer, cpumask_first(cpu_online_mask)); + watchdog_running = 1; +} + +static inline void clocksource_stop_watchdog(void) +{ + if (!watchdog_running || (watchdog && !list_empty(&watchdog_list))) + return; + del_timer(&watchdog_timer); + watchdog_running = 0; +} + +static inline void clocksource_reset_watchdog(void) +{ + struct clocksource *cs; + + list_for_each_entry(cs, &watchdog_list, wd_list) + cs->flags &= ~CLOCK_SOURCE_WATCHDOG; +} + static void clocksource_resume_watchdog(void) { - set_bit(0, &watchdog_resumed); + unsigned long flags; + + spin_lock_irqsave(&watchdog_lock, flags); + clocksource_reset_watchdog(); + spin_unlock_irqrestore(&watchdog_lock, flags); } -static void clocksource_check_watchdog(struct clocksource *cs) +static void clocksource_enqueue_watchdog(struct clocksource *cs) { - struct clocksource *cse; unsigned long flags; spin_lock_irqsave(&watchdog_lock, flags); if (cs->flags & CLOCK_SOURCE_MUST_VERIFY) { - int started = !list_empty(&watchdog_list); - + /* cs is a clocksource to be watched. */ list_add(&cs->wd_list, &watchdog_list); - if (!started && watchdog) { - watchdog_last = watchdog->read(); - watchdog_timer.expires = jiffies + WATCHDOG_INTERVAL; - add_timer_on(&watchdog_timer, - cpumask_first(cpu_online_mask)); - } + cs->flags &= ~CLOCK_SOURCE_WATCHDOG; } else { + /* cs is a watchdog. */ if (cs->flags & CLOCK_SOURCE_IS_CONTINUOUS) cs->flags |= CLOCK_SOURCE_VALID_FOR_HRES; - + /* Pick the best watchdog. */ if (!watchdog || cs->rating > watchdog->rating) { - if (watchdog) - del_timer(&watchdog_timer); watchdog = cs; - init_timer(&watchdog_timer); - watchdog_timer.function = clocksource_watchdog; - /* Reset watchdog cycles */ - list_for_each_entry(cse, &watchdog_list, wd_list) - cse->flags &= ~CLOCK_SOURCE_WATCHDOG; - /* Start if list is not empty */ - if (!list_empty(&watchdog_list)) { - watchdog_last = watchdog->read(); - watchdog_timer.expires = - jiffies + WATCHDOG_INTERVAL; - add_timer_on(&watchdog_timer, - cpumask_first(cpu_online_mask)); - } + clocksource_reset_watchdog(); + } + } + /* Check if the watchdog timer needs to be started. */ + clocksource_start_watchdog(); + spin_unlock_irqrestore(&watchdog_lock, flags); +} + +static void clocksource_dequeue_watchdog(struct clocksource *cs) +{ + struct clocksource *tmp; + unsigned long flags; + + spin_lock_irqsave(&watchdog_lock, flags); + if (cs->flags & CLOCK_SOURCE_MUST_VERIFY) { + /* cs is a watched clocksource. */ + list_del_init(&cs->wd_list); + } else if (cs == watchdog) { + /* Reset watchdog cycles */ + clocksource_reset_watchdog(); + /* Current watchdog is removed. Find an alternative. */ + watchdog = NULL; + list_for_each_entry(tmp, &clocksource_list, list) { + if (tmp == cs || tmp->flags & CLOCK_SOURCE_MUST_VERIFY) + continue; + if (!watchdog || tmp->rating > watchdog->rating) + watchdog = tmp; } } + cs->flags &= ~CLOCK_SOURCE_WATCHDOG; + /* Check if the watchdog timer needs to be stopped. */ + clocksource_stop_watchdog(); + spin_unlock_irqrestore(&watchdog_lock, flags); +} + +static int clocksource_watchdog_kthread(void *data) +{ + struct clocksource *cs, *tmp; + unsigned long flags; + LIST_HEAD(unstable); + + mutex_lock(&clocksource_mutex); + spin_lock_irqsave(&watchdog_lock, flags); + list_for_each_entry_safe(cs, tmp, &watchdog_list, wd_list) + if (cs->flags & CLOCK_SOURCE_UNSTABLE) { + list_del_init(&cs->wd_list); + list_add(&cs->wd_list, &unstable); + } + /* Check if the watchdog timer needs to be stopped. */ + clocksource_stop_watchdog(); spin_unlock_irqrestore(&watchdog_lock, flags); + + /* Needs to be done outside of watchdog lock */ + list_for_each_entry_safe(cs, tmp, &unstable, wd_list) { + list_del_init(&cs->wd_list); + __clocksource_change_rating(cs, 0); + } + mutex_unlock(&clocksource_mutex); + return 0; } -#else -static void clocksource_check_watchdog(struct clocksource *cs) + +#else /* CONFIG_CLOCKSOURCE_WATCHDOG */ + +static void clocksource_enqueue_watchdog(struct clocksource *cs) { if (cs->flags & CLOCK_SOURCE_IS_CONTINUOUS) cs->flags |= CLOCK_SOURCE_VALID_FOR_HRES; } +static inline void clocksource_dequeue_watchdog(struct clocksource *cs) { } static inline void clocksource_resume_watchdog(void) { } -#endif +static inline int clocksource_watchdog_kthread(void *data) { return 0; } + +#endif /* CONFIG_CLOCKSOURCE_WATCHDOG */ /** * clocksource_resume - resume the clocksource(s) @@ -294,18 +446,12 @@ static inline void clocksource_resume_watchdog(void) { } void clocksource_resume(void) { struct clocksource *cs; - unsigned long flags; - - spin_lock_irqsave(&clocksource_lock, flags); - list_for_each_entry(cs, &clocksource_list, list) { + list_for_each_entry(cs, &clocksource_list, list) if (cs->resume) cs->resume(); - } clocksource_resume_watchdog(); - - spin_unlock_irqrestore(&clocksource_lock, flags); } /** @@ -321,74 +467,134 @@ void clocksource_touch_watchdog(void) } /** - * clocksource_get_next - Returns the selected clocksource + * clocksource_max_deferment - Returns max time the clocksource can be deferred + * @cs: Pointer to clocksource * */ -struct clocksource *clocksource_get_next(void) +static u64 clocksource_max_deferment(struct clocksource *cs) { - unsigned long flags; + u64 max_nsecs, max_cycles; - spin_lock_irqsave(&clocksource_lock, flags); - if (next_clocksource && finished_booting) { - curr_clocksource = next_clocksource; - next_clocksource = NULL; - } - spin_unlock_irqrestore(&clocksource_lock, flags); + /* + * Calculate the maximum number of cycles that we can pass to the + * cyc2ns function without overflowing a 64-bit signed result. The + * maximum number of cycles is equal to ULLONG_MAX/cs->mult which + * is equivalent to the below. + * max_cycles < (2^63)/cs->mult + * max_cycles < 2^(log2((2^63)/cs->mult)) + * max_cycles < 2^(log2(2^63) - log2(cs->mult)) + * max_cycles < 2^(63 - log2(cs->mult)) + * max_cycles < 1 << (63 - log2(cs->mult)) + * Please note that we add 1 to the result of the log2 to account for + * any rounding errors, ensure the above inequality is satisfied and + * no overflow will occur. + */ + max_cycles = 1ULL << (63 - (ilog2(cs->mult) + 1)); - return curr_clocksource; + /* + * The actual maximum number of cycles we can defer the clocksource is + * determined by the minimum of max_cycles and cs->mask. + */ + max_cycles = min_t(u64, max_cycles, (u64) cs->mask); + max_nsecs = clocksource_cyc2ns(max_cycles, cs->mult, cs->shift); + + /* + * To ensure that the clocksource does not wrap whilst we are idle, + * limit the time the clocksource can be deferred by 12.5%. Please + * note a margin of 12.5% is used because this can be computed with + * a shift, versus say 10% which would require division. + */ + return max_nsecs - (max_nsecs >> 5); } +#ifdef CONFIG_GENERIC_TIME + /** - * select_clocksource - Selects the best registered clocksource. + * clocksource_select - Select the best clocksource available * - * Private function. Must hold clocksource_lock when called. + * Private function. Must hold clocksource_mutex when called. * * Select the clocksource with the best rating, or the clocksource, * which is selected by userspace override. */ -static struct clocksource *select_clocksource(void) +static void clocksource_select(void) { - struct clocksource *next; + struct clocksource *best, *cs; - if (list_empty(&clocksource_list)) - return NULL; + if (!finished_booting || list_empty(&clocksource_list)) + return; + /* First clocksource on the list has the best rating. */ + best = list_first_entry(&clocksource_list, struct clocksource, list); + /* Check for the override clocksource. */ + list_for_each_entry(cs, &clocksource_list, list) { + if (strcmp(cs->name, override_name) != 0) + continue; + /* + * Check to make sure we don't switch to a non-highres + * capable clocksource if the tick code is in oneshot + * mode (highres or nohz) + */ + if (!(cs->flags & CLOCK_SOURCE_VALID_FOR_HRES) && + tick_oneshot_mode_active()) { + /* Override clocksource cannot be used. */ + printk(KERN_WARNING "Override clocksource %s is not " + "HRT compatible. Cannot switch while in " + "HRT/NOHZ mode\n", cs->name); + override_name[0] = 0; + } else + /* Override clocksource can be used. */ + best = cs; + break; + } + if (curr_clocksource != best) { + printk(KERN_INFO "Switching to clocksource %s\n", best->name); + curr_clocksource = best; + timekeeping_notify(curr_clocksource); + } +} - if (clocksource_override) - next = clocksource_override; - else - next = list_entry(clocksource_list.next, struct clocksource, - list); +#else /* CONFIG_GENERIC_TIME */ - if (next == curr_clocksource) - return NULL; +static inline void clocksource_select(void) { } - return next; -} +#endif /* - * Enqueue the clocksource sorted by rating + * clocksource_done_booting - Called near the end of core bootup + * + * Hack to avoid lots of clocksource churn at boot time. + * We use fs_initcall because we want this to start before + * device_initcall but after subsys_initcall. */ -static int clocksource_enqueue(struct clocksource *c) +static int __init clocksource_done_booting(void) { - struct list_head *tmp, *entry = &clocksource_list; + finished_booting = 1; - list_for_each(tmp, &clocksource_list) { - struct clocksource *cs; + /* + * Run the watchdog first to eliminate unstable clock sources + */ + clocksource_watchdog_kthread(NULL); - cs = list_entry(tmp, struct clocksource, list); - if (cs == c) - return -EBUSY; - /* Keep track of the place, where to insert */ - if (cs->rating >= c->rating) - entry = tmp; - } - list_add(&c->list, entry); + mutex_lock(&clocksource_mutex); + clocksource_select(); + mutex_unlock(&clocksource_mutex); + return 0; +} +fs_initcall(clocksource_done_booting); - if (strlen(c->name) == strlen(override_name) && - !strcmp(c->name, override_name)) - clocksource_override = c; +/* + * Enqueue the clocksource sorted by rating + */ +static void clocksource_enqueue(struct clocksource *cs) +{ + struct list_head *entry = &clocksource_list; + struct clocksource *tmp; - return 0; + list_for_each_entry(tmp, &clocksource_list, list) + /* Keep track of the place, where to insert */ + if (tmp->rating >= cs->rating) + entry = &tmp->list; + list_add(&cs->list, entry); } /** @@ -397,55 +603,51 @@ static int clocksource_enqueue(struct clocksource *c) * * Returns -EBUSY if registration fails, zero otherwise. */ -int clocksource_register(struct clocksource *c) +int clocksource_register(struct clocksource *cs) { - unsigned long flags; - int ret; - - /* save mult_orig on registration */ - c->mult_orig = c->mult; - - spin_lock_irqsave(&clocksource_lock, flags); - ret = clocksource_enqueue(c); - if (!ret) - next_clocksource = select_clocksource(); - spin_unlock_irqrestore(&clocksource_lock, flags); - if (!ret) - clocksource_check_watchdog(c); - return ret; + /* calculate max idle time permitted for this clocksource */ + cs->max_idle_ns = clocksource_max_deferment(cs); + + mutex_lock(&clocksource_mutex); + clocksource_enqueue(cs); + clocksource_select(); + clocksource_enqueue_watchdog(cs); + mutex_unlock(&clocksource_mutex); + return 0; } EXPORT_SYMBOL(clocksource_register); +static void __clocksource_change_rating(struct clocksource *cs, int rating) +{ + list_del(&cs->list); + cs->rating = rating; + clocksource_enqueue(cs); + clocksource_select(); +} + /** * clocksource_change_rating - Change the rating of a registered clocksource - * */ void clocksource_change_rating(struct clocksource *cs, int rating) { - unsigned long flags; - - spin_lock_irqsave(&clocksource_lock, flags); - list_del(&cs->list); - cs->rating = rating; - clocksource_enqueue(cs); - next_clocksource = select_clocksource(); - spin_unlock_irqrestore(&clocksource_lock, flags); + mutex_lock(&clocksource_mutex); + __clocksource_change_rating(cs, rating); + mutex_unlock(&clocksource_mutex); } +EXPORT_SYMBOL(clocksource_change_rating); /** * clocksource_unregister - remove a registered clocksource */ void clocksource_unregister(struct clocksource *cs) { - unsigned long flags; - - spin_lock_irqsave(&clocksource_lock, flags); + mutex_lock(&clocksource_mutex); + clocksource_dequeue_watchdog(cs); list_del(&cs->list); - if (clocksource_override == cs) - clocksource_override = NULL; - next_clocksource = select_clocksource(); - spin_unlock_irqrestore(&clocksource_lock, flags); + clocksource_select(); + mutex_unlock(&clocksource_mutex); } +EXPORT_SYMBOL(clocksource_unregister); #ifdef CONFIG_SYSFS /** @@ -461,9 +663,9 @@ sysfs_show_current_clocksources(struct sys_device *dev, { ssize_t count = 0; - spin_lock_irq(&clocksource_lock); + mutex_lock(&clocksource_mutex); count = snprintf(buf, PAGE_SIZE, "%s\n", curr_clocksource->name); - spin_unlock_irq(&clocksource_lock); + mutex_unlock(&clocksource_mutex); return count; } @@ -481,9 +683,7 @@ static ssize_t sysfs_override_clocksource(struct sys_device *dev, struct sysdev_attribute *attr, const char *buf, size_t count) { - struct clocksource *ovr = NULL; size_t ret = count; - int len; /* strings from sysfs write are not 0 terminated! */ if (count >= sizeof(override_name)) @@ -493,32 +693,14 @@ static ssize_t sysfs_override_clocksource(struct sys_device *dev, if (buf[count-1] == '\n') count--; - spin_lock_irq(&clocksource_lock); + mutex_lock(&clocksource_mutex); if (count > 0) memcpy(override_name, buf, count); override_name[count] = 0; + clocksource_select(); - len = strlen(override_name); - if (len) { - struct clocksource *cs; - - ovr = clocksource_override; - /* try to select it: */ - list_for_each_entry(cs, &clocksource_list, list) { - if (strlen(cs->name) == len && - !strcmp(cs->name, override_name)) - ovr = cs; - } - } - - /* Reselect, when the override name has changed */ - if (ovr != clocksource_override) { - clocksource_override = ovr; - next_clocksource = select_clocksource(); - } - - spin_unlock_irq(&clocksource_lock); + mutex_unlock(&clocksource_mutex); return ret; } @@ -538,13 +720,19 @@ sysfs_show_available_clocksources(struct sys_device *dev, struct clocksource *src; ssize_t count = 0; - spin_lock_irq(&clocksource_lock); + mutex_lock(&clocksource_mutex); list_for_each_entry(src, &clocksource_list, list) { - count += snprintf(buf + count, + /* + * Don't show non-HRES clocksource if the tick code is + * in one shot mode (highres=on or nohz=on) + */ + if (!tick_oneshot_mode_active() || + (src->flags & CLOCK_SOURCE_VALID_FOR_HRES)) + count += snprintf(buf + count, max((ssize_t)PAGE_SIZE - count, (ssize_t)0), "%s ", src->name); } - spin_unlock_irq(&clocksource_lock); + mutex_unlock(&clocksource_mutex); count += snprintf(buf + count, max((ssize_t)PAGE_SIZE - count, (ssize_t)0), "\n"); @@ -599,11 +787,10 @@ device_initcall(init_clocksource_sysfs); */ static int __init boot_override_clocksource(char* str) { - unsigned long flags; - spin_lock_irqsave(&clocksource_lock, flags); + mutex_lock(&clocksource_mutex); if (str) strlcpy(override_name, str, sizeof(override_name)); - spin_unlock_irqrestore(&clocksource_lock, flags); + mutex_unlock(&clocksource_mutex); return 1; } diff --git a/kernel/time/jiffies.c b/kernel/time/jiffies.c index 06f197560f3..5404a845690 100644 --- a/kernel/time/jiffies.c +++ b/kernel/time/jiffies.c @@ -50,7 +50,7 @@ */ #define JIFFIES_SHIFT 8 -static cycle_t jiffies_read(void) +static cycle_t jiffies_read(struct clocksource *cs) { return (cycle_t) jiffies; } @@ -61,7 +61,6 @@ struct clocksource clocksource_jiffies = { .read = jiffies_read, .mask = 0xffffffff, /*32bits*/ .mult = NSEC_PER_JIFFY << JIFFIES_SHIFT, /* details above */ - .mult_orig = NSEC_PER_JIFFY << JIFFIES_SHIFT, .shift = JIFFIES_SHIFT, }; @@ -71,3 +70,8 @@ static int __init init_jiffies_clocksource(void) } core_initcall(init_jiffies_clocksource); + +struct clocksource * __init __weak clocksource_default_clock(void) +{ + return &clocksource_jiffies; +} diff --git a/kernel/time/ntp.c b/kernel/time/ntp.c index 7fc64375ff4..4800f933910 100644 --- a/kernel/time/ntp.c +++ b/kernel/time/ntp.c @@ -194,8 +194,7 @@ static enum hrtimer_restart ntp_leap_second(struct hrtimer *timer) case TIME_OK: break; case TIME_INS: - xtime.tv_sec--; - wall_to_monotonic.tv_sec++; + timekeeping_leap_insert(-1); time_state = TIME_OOP; printk(KERN_NOTICE "Clock: inserting leap second 23:59:60 UTC\n"); @@ -203,9 +202,8 @@ static enum hrtimer_restart ntp_leap_second(struct hrtimer *timer) res = HRTIMER_RESTART; break; case TIME_DEL: - xtime.tv_sec++; + timekeeping_leap_insert(1); time_tai--; - wall_to_monotonic.tv_sec--; time_state = TIME_WAIT; printk(KERN_NOTICE "Clock: deleting leap second 23:59:59 UTC\n"); @@ -219,7 +217,6 @@ static enum hrtimer_restart ntp_leap_second(struct hrtimer *timer) time_state = TIME_OK; break; } - update_vsyscall(&xtime, clock); write_sequnlock(&xtime_lock); diff --git a/kernel/time/tick-broadcast.c b/kernel/time/tick-broadcast.c index 118a3b3b3f9..c2ec25087a3 100644 --- a/kernel/time/tick-broadcast.c +++ b/kernel/time/tick-broadcast.c @@ -27,7 +27,7 @@ * timer stops in C3 state. */ -struct tick_device tick_broadcast_device; +static struct tick_device tick_broadcast_device; /* FIXME: Use cpumask_var_t. */ static DECLARE_BITMAP(tick_broadcast_mask, NR_CPUS); static DECLARE_BITMAP(tmpmask, NR_CPUS); @@ -205,11 +205,11 @@ static void tick_handle_periodic_broadcast(struct clock_event_device *dev) * Powerstate information: The system enters/leaves a state, where * affected devices might stop */ -static void tick_do_broadcast_on_off(void *why) +static void tick_do_broadcast_on_off(unsigned long *reason) { struct clock_event_device *bc, *dev; struct tick_device *td; - unsigned long flags, *reason = why; + unsigned long flags; int cpu, bc_stopped; spin_lock_irqsave(&tick_broadcast_lock, flags); @@ -276,8 +276,7 @@ void tick_broadcast_on_off(unsigned long reason, int *oncpu) printk(KERN_ERR "tick-broadcast: ignoring broadcast for " "offline CPU #%d\n", *oncpu); else - smp_call_function_single(*oncpu, tick_do_broadcast_on_off, - &reason, 1); + tick_do_broadcast_on_off(&reason); } /* diff --git a/kernel/time/tick-common.c b/kernel/time/tick-common.c index 21a5ca84951..83c4417b6a3 100644 --- a/kernel/time/tick-common.c +++ b/kernel/time/tick-common.c @@ -93,7 +93,17 @@ void tick_handle_periodic(struct clock_event_device *dev) for (;;) { if (!clockevents_program_event(dev, next, ktime_get())) return; - tick_periodic(cpu); + /* + * Have to be careful here. If we're in oneshot mode, + * before we call tick_periodic() in a loop, we need + * to be sure we're using a real hardware clocksource. + * Otherwise we could get trapped in an infinite + * loop, as the tick_periodic() increments jiffies, + * when then will increment time, posibly causing + * the loop to trigger again and again. + */ + if (timekeeping_valid_for_hres()) + tick_periodic(cpu); next = ktime_add(next, tick_period); } } diff --git a/kernel/time/tick-oneshot.c b/kernel/time/tick-oneshot.c index 2e8de678e76..0a8a213016f 100644 --- a/kernel/time/tick-oneshot.c +++ b/kernel/time/tick-oneshot.c @@ -50,9 +50,9 @@ int tick_dev_program_event(struct clock_event_device *dev, ktime_t expires, dev->min_delta_ns += dev->min_delta_ns >> 1; printk(KERN_WARNING - "CE: %s increasing min_delta_ns to %lu nsec\n", + "CE: %s increasing min_delta_ns to %llu nsec\n", dev->name ? dev->name : "?", - dev->min_delta_ns << 1); + (unsigned long long) dev->min_delta_ns << 1); i = 0; } @@ -128,6 +128,23 @@ int tick_switch_to_oneshot(void (*handler)(struct clock_event_device *)) return 0; } +/** + * tick_check_oneshot_mode - check whether the system is in oneshot mode + * + * returns 1 when either nohz or highres are enabled. otherwise 0. + */ +int tick_oneshot_mode_active(void) +{ + unsigned long flags; + int ret; + + local_irq_save(flags); + ret = __get_cpu_var(tick_cpu_device).mode == TICKDEV_MODE_ONESHOT; + local_irq_restore(flags); + + return ret; +} + #ifdef CONFIG_HIGH_RES_TIMERS /** * tick_init_highres - switch to high resolution mode diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c index d3f1ef4d5cb..f992762d7f5 100644 --- a/kernel/time/tick-sched.c +++ b/kernel/time/tick-sched.c @@ -134,18 +134,13 @@ __setup("nohz=", setup_tick_nohz); * value. We do this unconditionally on any cpu, as we don't know whether the * cpu, which has the update task assigned is in a long sleep. */ -static void tick_nohz_update_jiffies(void) +static void tick_nohz_update_jiffies(ktime_t now) { int cpu = smp_processor_id(); struct tick_sched *ts = &per_cpu(tick_cpu_sched, cpu); unsigned long flags; - ktime_t now; - - if (!ts->tick_stopped) - return; cpumask_clear_cpu(cpu, nohz_cpu_mask); - now = ktime_get(); ts->idle_waketime = now; local_irq_save(flags); @@ -155,20 +150,17 @@ static void tick_nohz_update_jiffies(void) touch_softlockup_watchdog(); } -static void tick_nohz_stop_idle(int cpu) +static void tick_nohz_stop_idle(int cpu, ktime_t now) { struct tick_sched *ts = &per_cpu(tick_cpu_sched, cpu); + ktime_t delta; - if (ts->idle_active) { - ktime_t now, delta; - now = ktime_get(); - delta = ktime_sub(now, ts->idle_entrytime); - ts->idle_lastupdate = now; - ts->idle_sleeptime = ktime_add(ts->idle_sleeptime, delta); - ts->idle_active = 0; + delta = ktime_sub(now, ts->idle_entrytime); + ts->idle_lastupdate = now; + ts->idle_sleeptime = ktime_add(ts->idle_sleeptime, delta); + ts->idle_active = 0; - sched_clock_idle_wakeup_event(0); - } + sched_clock_idle_wakeup_event(0); } static ktime_t tick_nohz_start_idle(struct tick_sched *ts) @@ -216,12 +208,29 @@ void tick_nohz_stop_sched_tick(int inidle) struct tick_sched *ts; ktime_t last_update, expires, now; struct clock_event_device *dev = __get_cpu_var(tick_cpu_device).evtdev; + u64 time_delta; int cpu; local_irq_save(flags); cpu = smp_processor_id(); ts = &per_cpu(tick_cpu_sched, cpu); + + /* + * Call to tick_nohz_start_idle stops the last_update_time from being + * updated. Thus, it must not be called in the event we are called from + * irq_exit() with the prior state different than idle. + */ + if (!inidle && !ts->inidle) + goto end; + + /* + * Set ts->inidle unconditionally. Even if the system did not + * switch to NOHZ mode the cpu frequency governers rely on the + * update of the idle time accounting in tick_nohz_start_idle(). + */ + ts->inidle = 1; + now = tick_nohz_start_idle(ts); /* @@ -239,11 +248,6 @@ void tick_nohz_stop_sched_tick(int inidle) if (unlikely(ts->nohz_mode == NOHZ_MODE_INACTIVE)) goto end; - if (!inidle && !ts->inidle) - goto end; - - ts->inidle = 1; - if (need_resched()) goto end; @@ -252,7 +256,7 @@ void tick_nohz_stop_sched_tick(int inidle) if (ratelimit < 10) { printk(KERN_ERR "NOHZ: local_softirq_pending %02x\n", - local_softirq_pending()); + (unsigned int) local_softirq_pending()); ratelimit++; } goto end; @@ -264,14 +268,18 @@ void tick_nohz_stop_sched_tick(int inidle) seq = read_seqbegin(&xtime_lock); last_update = last_jiffies_update; last_jiffies = jiffies; + time_delta = timekeeping_max_deferment(); } while (read_seqretry(&xtime_lock, seq)); - /* Get the next timer wheel timer */ - next_jiffies = get_next_timer_interrupt(last_jiffies); - delta_jiffies = next_jiffies - last_jiffies; - - if (rcu_needs_cpu(cpu) || printk_needs_cpu(cpu)) + if (rcu_needs_cpu(cpu) || printk_needs_cpu(cpu) || + arch_needs_cpu(cpu)) { + next_jiffies = last_jiffies + 1; delta_jiffies = 1; + } else { + /* Get the next timer wheel timer */ + next_jiffies = get_next_timer_interrupt(last_jiffies); + delta_jiffies = next_jiffies - last_jiffies; + } /* * Do not stop the tick, if we are only one off * or if the cpu is required for rcu @@ -283,22 +291,51 @@ void tick_nohz_stop_sched_tick(int inidle) if ((long)delta_jiffies >= 1) { /* - * calculate the expiry time for the next timer wheel - * timer - */ - expires = ktime_add_ns(last_update, tick_period.tv64 * - delta_jiffies); - - /* * If this cpu is the one which updates jiffies, then * give up the assignment and let it be taken by the * cpu which runs the tick timer next, which might be * this cpu as well. If we don't drop this here the * jiffies might be stale and do_timer() never - * invoked. + * invoked. Keep track of the fact that it was the one + * which had the do_timer() duty last. If this cpu is + * the one which had the do_timer() duty last, we + * limit the sleep time to the timekeeping + * max_deferement value which we retrieved + * above. Otherwise we can sleep as long as we want. */ - if (cpu == tick_do_timer_cpu) + if (cpu == tick_do_timer_cpu) { tick_do_timer_cpu = TICK_DO_TIMER_NONE; + ts->do_timer_last = 1; + } else if (tick_do_timer_cpu != TICK_DO_TIMER_NONE) { + time_delta = KTIME_MAX; + ts->do_timer_last = 0; + } else if (!ts->do_timer_last) { + time_delta = KTIME_MAX; + } + + /* + * calculate the expiry time for the next timer wheel + * timer. delta_jiffies >= NEXT_TIMER_MAX_DELTA signals + * that there is no timer pending or at least extremely + * far into the future (12 days for HZ=1000). In this + * case we set the expiry to the end of time. + */ + if (likely(delta_jiffies < NEXT_TIMER_MAX_DELTA)) { + /* + * Calculate the time delta for the next timer event. + * If the time delta exceeds the maximum time delta + * permitted by the current clocksource then adjust + * the time delta accordingly to ensure the + * clocksource does not wrap. + */ + time_delta = min_t(u64, time_delta, + tick_period.tv64 * delta_jiffies); + } + + if (time_delta < KTIME_MAX) + expires = ktime_add_ns(last_update, time_delta); + else + expires.tv64 = KTIME_MAX; if (delta_jiffies > 1) cpumask_set_cpu(cpu, nohz_cpu_mask); @@ -331,25 +368,22 @@ void tick_nohz_stop_sched_tick(int inidle) ts->idle_sleeps++; + /* Mark expires */ + ts->idle_expires = expires; + /* - * delta_jiffies >= NEXT_TIMER_MAX_DELTA signals that - * there is no timer pending or at least extremly far - * into the future (12 days for HZ=1000). In this case - * we simply stop the tick timer: + * If the expiration time == KTIME_MAX, then + * in this case we simply stop the tick timer. */ - if (unlikely(delta_jiffies >= NEXT_TIMER_MAX_DELTA)) { - ts->idle_expires.tv64 = KTIME_MAX; + if (unlikely(expires.tv64 == KTIME_MAX)) { if (ts->nohz_mode == NOHZ_MODE_HIGHRES) hrtimer_cancel(&ts->sched_timer); goto out; } - /* Mark expiries */ - ts->idle_expires = expires; - if (ts->nohz_mode == NOHZ_MODE_HIGHRES) { hrtimer_start(&ts->sched_timer, expires, - HRTIMER_MODE_ABS); + HRTIMER_MODE_ABS_PINNED); /* Check, if the timer was already in the past */ if (hrtimer_active(&ts->sched_timer)) goto out; @@ -395,7 +429,7 @@ static void tick_nohz_restart(struct tick_sched *ts, ktime_t now) if (ts->nohz_mode == NOHZ_MODE_HIGHRES) { hrtimer_start_expires(&ts->sched_timer, - HRTIMER_MODE_ABS); + HRTIMER_MODE_ABS_PINNED); /* Check, if the timer was already in the past */ if (hrtimer_active(&ts->sched_timer)) break; @@ -425,7 +459,11 @@ void tick_nohz_restart_sched_tick(void) ktime_t now; local_irq_disable(); - tick_nohz_stop_idle(cpu); + if (ts->idle_active || (ts->inidle && ts->tick_stopped)) + now = ktime_get(); + + if (ts->idle_active) + tick_nohz_stop_idle(cpu, now); if (!ts->inidle || !ts->tick_stopped) { ts->inidle = 0; @@ -439,7 +477,6 @@ void tick_nohz_restart_sched_tick(void) /* Update jiffies first */ select_nohz_load_balancer(0); - now = ktime_get(); tick_do_update_jiffies64(now); cpumask_clear_cpu(cpu, nohz_cpu_mask); @@ -573,22 +610,18 @@ static void tick_nohz_switch_to_nohz(void) * timer and do not touch the other magic bits which need to be done * when idle is left. */ -static void tick_nohz_kick_tick(int cpu) +static void tick_nohz_kick_tick(int cpu, ktime_t now) { #if 0 /* Switch back to 2.6.27 behaviour */ struct tick_sched *ts = &per_cpu(tick_cpu_sched, cpu); - ktime_t delta, now; - - if (!ts->tick_stopped) - return; + ktime_t delta; /* * Do not touch the tick device, when the next expiry is either * already reached or less/equal than the tick period. */ - now = ktime_get(); delta = ktime_sub(hrtimer_get_expires(&ts->sched_timer), now); if (delta.tv64 <= tick_period.tv64) return; @@ -597,9 +630,26 @@ static void tick_nohz_kick_tick(int cpu) #endif } +static inline void tick_check_nohz(int cpu) +{ + struct tick_sched *ts = &per_cpu(tick_cpu_sched, cpu); + ktime_t now; + + if (!ts->idle_active && !ts->tick_stopped) + return; + now = ktime_get(); + if (ts->idle_active) + tick_nohz_stop_idle(cpu, now); + if (ts->tick_stopped) { + tick_nohz_update_jiffies(now); + tick_nohz_kick_tick(cpu, now); + } +} + #else static inline void tick_nohz_switch_to_nohz(void) { } +static inline void tick_check_nohz(int cpu) { } #endif /* NO_HZ */ @@ -609,11 +659,7 @@ static inline void tick_nohz_switch_to_nohz(void) { } void tick_check_idle(int cpu) { tick_check_oneshot_broadcast(cpu); -#ifdef CONFIG_NO_HZ - tick_nohz_stop_idle(cpu); - tick_nohz_update_jiffies(); - tick_nohz_kick_tick(cpu); -#endif + tick_check_nohz(cpu); } /* @@ -698,7 +744,8 @@ void tick_setup_sched_timer(void) for (;;) { hrtimer_forward(&ts->sched_timer, now, tick_period); - hrtimer_start_expires(&ts->sched_timer, HRTIMER_MODE_ABS); + hrtimer_start_expires(&ts->sched_timer, + HRTIMER_MODE_ABS_PINNED); /* Check, if the timer was already in the past */ if (hrtimer_active(&ts->sched_timer)) break; diff --git a/kernel/time/timecompare.c b/kernel/time/timecompare.c index 71e7f1a1915..96ff643a5a5 100644 --- a/kernel/time/timecompare.c +++ b/kernel/time/timecompare.c @@ -40,7 +40,7 @@ ktime_t timecompare_transform(struct timecompare *sync, return ns_to_ktime(nsec); } -EXPORT_SYMBOL(timecompare_transform); +EXPORT_SYMBOL_GPL(timecompare_transform); int timecompare_offset(struct timecompare *sync, s64 *offset, @@ -131,7 +131,7 @@ int timecompare_offset(struct timecompare *sync, return used; } -EXPORT_SYMBOL(timecompare_offset); +EXPORT_SYMBOL_GPL(timecompare_offset); void __timecompare_update(struct timecompare *sync, u64 source_tstamp) @@ -188,4 +188,4 @@ void __timecompare_update(struct timecompare *sync, } } } -EXPORT_SYMBOL(__timecompare_update); +EXPORT_SYMBOL_GPL(__timecompare_update); diff --git a/kernel/time/timeconv.c b/kernel/time/timeconv.c new file mode 100644 index 00000000000..86628e755f3 --- /dev/null +++ b/kernel/time/timeconv.c @@ -0,0 +1,127 @@ +/* + * Copyright (C) 1993, 1994, 1995, 1996, 1997 Free Software Foundation, Inc. + * This file is part of the GNU C Library. + * Contributed by Paul Eggert (eggert@twinsun.com). + * + * The GNU C Library is free software; you can redistribute it and/or + * modify it under the terms of the GNU Library General Public License as + * published by the Free Software Foundation; either version 2 of the + * License, or (at your option) any later version. + * + * The GNU C Library is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * Library General Public License for more details. + * + * You should have received a copy of the GNU Library General Public + * License along with the GNU C Library; see the file COPYING.LIB. If not, + * write to the Free Software Foundation, Inc., 59 Temple Place - Suite 330, + * Boston, MA 02111-1307, USA. + */ + +/* + * Converts the calendar time to broken-down time representation + * Based on code from glibc-2.6 + * + * 2009-7-14: + * Moved from glibc-2.6 to kernel by Zhaolei<zhaolei@cn.fujitsu.com> + */ + +#include <linux/time.h> +#include <linux/module.h> + +/* + * Nonzero if YEAR is a leap year (every 4 years, + * except every 100th isn't, and every 400th is). + */ +static int __isleap(long year) +{ + return (year) % 4 == 0 && ((year) % 100 != 0 || (year) % 400 == 0); +} + +/* do a mathdiv for long type */ +static long math_div(long a, long b) +{ + return a / b - (a % b < 0); +} + +/* How many leap years between y1 and y2, y1 must less or equal to y2 */ +static long leaps_between(long y1, long y2) +{ + long leaps1 = math_div(y1 - 1, 4) - math_div(y1 - 1, 100) + + math_div(y1 - 1, 400); + long leaps2 = math_div(y2 - 1, 4) - math_div(y2 - 1, 100) + + math_div(y2 - 1, 400); + return leaps2 - leaps1; +} + +/* How many days come before each month (0-12). */ +static const unsigned short __mon_yday[2][13] = { + /* Normal years. */ + {0, 31, 59, 90, 120, 151, 181, 212, 243, 273, 304, 334, 365}, + /* Leap years. */ + {0, 31, 60, 91, 121, 152, 182, 213, 244, 274, 305, 335, 366} +}; + +#define SECS_PER_HOUR (60 * 60) +#define SECS_PER_DAY (SECS_PER_HOUR * 24) + +/** + * time_to_tm - converts the calendar time to local broken-down time + * + * @totalsecs the number of seconds elapsed since 00:00:00 on January 1, 1970, + * Coordinated Universal Time (UTC). + * @offset offset seconds adding to totalsecs. + * @result pointer to struct tm variable to receive broken-down time + */ +void time_to_tm(time_t totalsecs, int offset, struct tm *result) +{ + long days, rem, y; + const unsigned short *ip; + + days = totalsecs / SECS_PER_DAY; + rem = totalsecs % SECS_PER_DAY; + rem += offset; + while (rem < 0) { + rem += SECS_PER_DAY; + --days; + } + while (rem >= SECS_PER_DAY) { + rem -= SECS_PER_DAY; + ++days; + } + + result->tm_hour = rem / SECS_PER_HOUR; + rem %= SECS_PER_HOUR; + result->tm_min = rem / 60; + result->tm_sec = rem % 60; + + /* January 1, 1970 was a Thursday. */ + result->tm_wday = (4 + days) % 7; + if (result->tm_wday < 0) + result->tm_wday += 7; + + y = 1970; + + while (days < 0 || days >= (__isleap(y) ? 366 : 365)) { + /* Guess a corrected year, assuming 365 days per year. */ + long yg = y + math_div(days, 365); + + /* Adjust DAYS and Y to match the guessed year. */ + days -= (yg - y) * 365 + leaps_between(y, yg); + y = yg; + } + + result->tm_year = y - 1900; + + result->tm_yday = days; + + ip = __mon_yday[__isleap(y)]; + for (y = 11; days < ip[y]; y--) + continue; + days -= ip[y]; + + result->tm_mon = y; + result->tm_mday = days + 1; +} +EXPORT_SYMBOL(time_to_tm); diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c index 900f1b6598d..af4135f0582 100644 --- a/kernel/time/timekeeping.c +++ b/kernel/time/timekeeping.c @@ -13,16 +13,127 @@ #include <linux/percpu.h> #include <linux/init.h> #include <linux/mm.h> +#include <linux/sched.h> #include <linux/sysdev.h> #include <linux/clocksource.h> #include <linux/jiffies.h> #include <linux/time.h> #include <linux/tick.h> +#include <linux/stop_machine.h> + +/* Structure holding internal timekeeping values. */ +struct timekeeper { + /* Current clocksource used for timekeeping. */ + struct clocksource *clock; + /* The shift value of the current clocksource. */ + int shift; + + /* Number of clock cycles in one NTP interval. */ + cycle_t cycle_interval; + /* Number of clock shifted nano seconds in one NTP interval. */ + u64 xtime_interval; + /* Raw nano seconds accumulated per NTP interval. */ + u32 raw_interval; + + /* Clock shifted nano seconds remainder not stored in xtime.tv_nsec. */ + u64 xtime_nsec; + /* Difference between accumulated time and NTP time in ntp + * shifted nano seconds. */ + s64 ntp_error; + /* Shift conversion between clock shifted nano seconds and + * ntp shifted nano seconds. */ + int ntp_error_shift; + /* NTP adjusted clock multiplier */ + u32 mult; +}; + +struct timekeeper timekeeper; + +/** + * timekeeper_setup_internals - Set up internals to use clocksource clock. + * + * @clock: Pointer to clocksource. + * + * Calculates a fixed cycle/nsec interval for a given clocksource/adjustment + * pair and interval request. + * + * Unless you're the timekeeping code, you should not be using this! + */ +static void timekeeper_setup_internals(struct clocksource *clock) +{ + cycle_t interval; + u64 tmp; + + timekeeper.clock = clock; + clock->cycle_last = clock->read(clock); + + /* Do the ns -> cycle conversion first, using original mult */ + tmp = NTP_INTERVAL_LENGTH; + tmp <<= clock->shift; + tmp += clock->mult/2; + do_div(tmp, clock->mult); + if (tmp == 0) + tmp = 1; + + interval = (cycle_t) tmp; + timekeeper.cycle_interval = interval; + + /* Go back from cycles -> shifted ns */ + timekeeper.xtime_interval = (u64) interval * clock->mult; + timekeeper.raw_interval = + ((u64) interval * clock->mult) >> clock->shift; + + timekeeper.xtime_nsec = 0; + timekeeper.shift = clock->shift; + + timekeeper.ntp_error = 0; + timekeeper.ntp_error_shift = NTP_SCALE_SHIFT - clock->shift; + + /* + * The timekeeper keeps its own mult values for the currently + * active clocksource. These value will be adjusted via NTP + * to counteract clock drifting. + */ + timekeeper.mult = clock->mult; +} + +/* Timekeeper helper functions. */ +static inline s64 timekeeping_get_ns(void) +{ + cycle_t cycle_now, cycle_delta; + struct clocksource *clock; + /* read clocksource: */ + clock = timekeeper.clock; + cycle_now = clock->read(clock); + + /* calculate the delta since the last update_wall_time: */ + cycle_delta = (cycle_now - clock->cycle_last) & clock->mask; + + /* return delta convert to nanoseconds using ntp adjusted mult. */ + return clocksource_cyc2ns(cycle_delta, timekeeper.mult, + timekeeper.shift); +} + +static inline s64 timekeeping_get_ns_raw(void) +{ + cycle_t cycle_now, cycle_delta; + struct clocksource *clock; + + /* read clocksource: */ + clock = timekeeper.clock; + cycle_now = clock->read(clock); + + /* calculate the delta since the last update_wall_time: */ + cycle_delta = (cycle_now - clock->cycle_last) & clock->mask; + + /* return delta convert to nanoseconds using ntp adjusted mult. */ + return clocksource_cyc2ns(cycle_delta, clock->mult, clock->shift); +} /* * This read-write spinlock protects us from races in SMP while - * playing with xtime and avenrun. + * playing with xtime. */ __cacheline_aligned_in_smp DEFINE_SEQLOCK(xtime_lock); @@ -44,43 +155,54 @@ __cacheline_aligned_in_smp DEFINE_SEQLOCK(xtime_lock); */ struct timespec xtime __attribute__ ((aligned (16))); struct timespec wall_to_monotonic __attribute__ ((aligned (16))); -static unsigned long total_sleep_time; /* seconds */ +static struct timespec total_sleep_time; + +/* + * The raw monotonic time for the CLOCK_MONOTONIC_RAW posix clock. + */ +struct timespec raw_time; /* flag for if timekeeping is suspended */ int __read_mostly timekeeping_suspended; -static struct timespec xtime_cache __attribute__ ((aligned (16))); -void update_xtime_cache(u64 nsec) +/* must hold xtime_lock */ +void timekeeping_leap_insert(int leapsecond) { - xtime_cache = xtime; - timespec_add_ns(&xtime_cache, nsec); + xtime.tv_sec += leapsecond; + wall_to_monotonic.tv_sec -= leapsecond; + update_vsyscall(&xtime, timekeeper.clock, timekeeper.mult); } -struct clocksource *clock; - - #ifdef CONFIG_GENERIC_TIME + /** - * clocksource_forward_now - update clock to the current time + * timekeeping_forward_now - update clock to the current time * * Forward the current clock to update its state since the last call to * update_wall_time(). This is useful before significant clock changes, * as it avoids having to deal with this time offset explicitly. */ -static void clocksource_forward_now(void) +static void timekeeping_forward_now(void) { cycle_t cycle_now, cycle_delta; + struct clocksource *clock; s64 nsec; - cycle_now = clocksource_read(clock); + clock = timekeeper.clock; + cycle_now = clock->read(clock); cycle_delta = (cycle_now - clock->cycle_last) & clock->mask; clock->cycle_last = cycle_now; - nsec = cyc2ns(clock, cycle_delta); + nsec = clocksource_cyc2ns(cycle_delta, timekeeper.mult, + timekeeper.shift); + + /* If arch requires, add in gettimeoffset() */ + nsec += arch_gettimeoffset(); + timespec_add_ns(&xtime, nsec); - nsec = ((s64)cycle_delta * clock->mult_orig) >> clock->shift; - clock->raw_time.tv_nsec += nsec; + nsec = clocksource_cyc2ns(cycle_delta, clock->mult, clock->shift); + timespec_add_ns(&raw_time, nsec); } /** @@ -91,7 +213,6 @@ static void clocksource_forward_now(void) */ void getnstimeofday(struct timespec *ts) { - cycle_t cycle_now, cycle_delta; unsigned long seq; s64 nsecs; @@ -101,15 +222,10 @@ void getnstimeofday(struct timespec *ts) seq = read_seqbegin(&xtime_lock); *ts = xtime; + nsecs = timekeeping_get_ns(); - /* read clocksource: */ - cycle_now = clocksource_read(clock); - - /* calculate the delta since the last update_wall_time: */ - cycle_delta = (cycle_now - clock->cycle_last) & clock->mask; - - /* convert to nanoseconds: */ - nsecs = cyc2ns(clock, cycle_delta); + /* If arch requires, add in gettimeoffset() */ + nsecs += arch_gettimeoffset(); } while (read_seqretry(&xtime_lock, seq)); @@ -118,6 +234,57 @@ void getnstimeofday(struct timespec *ts) EXPORT_SYMBOL(getnstimeofday); +ktime_t ktime_get(void) +{ + unsigned int seq; + s64 secs, nsecs; + + WARN_ON(timekeeping_suspended); + + do { + seq = read_seqbegin(&xtime_lock); + secs = xtime.tv_sec + wall_to_monotonic.tv_sec; + nsecs = xtime.tv_nsec + wall_to_monotonic.tv_nsec; + nsecs += timekeeping_get_ns(); + + } while (read_seqretry(&xtime_lock, seq)); + /* + * Use ktime_set/ktime_add_ns to create a proper ktime on + * 32-bit architectures without CONFIG_KTIME_SCALAR. + */ + return ktime_add_ns(ktime_set(secs, 0), nsecs); +} +EXPORT_SYMBOL_GPL(ktime_get); + +/** + * ktime_get_ts - get the monotonic clock in timespec format + * @ts: pointer to timespec variable + * + * The function calculates the monotonic clock from the realtime + * clock and the wall_to_monotonic offset and stores the result + * in normalized timespec format in the variable pointed to by @ts. + */ +void ktime_get_ts(struct timespec *ts) +{ + struct timespec tomono; + unsigned int seq; + s64 nsecs; + + WARN_ON(timekeeping_suspended); + + do { + seq = read_seqbegin(&xtime_lock); + *ts = xtime; + tomono = wall_to_monotonic; + nsecs = timekeeping_get_ns(); + + } while (read_seqretry(&xtime_lock, seq)); + + set_normalized_timespec(ts, ts->tv_sec + tomono.tv_sec, + ts->tv_nsec + tomono.tv_nsec + nsecs); +} +EXPORT_SYMBOL_GPL(ktime_get_ts); + /** * do_gettimeofday - Returns the time of day in a timeval * @tv: pointer to the timeval to be set @@ -150,7 +317,7 @@ int do_settimeofday(struct timespec *tv) write_seqlock_irqsave(&xtime_lock, flags); - clocksource_forward_now(); + timekeeping_forward_now(); ts_delta.tv_sec = tv->tv_sec - xtime.tv_sec; ts_delta.tv_nsec = tv->tv_nsec - xtime.tv_nsec; @@ -158,12 +325,10 @@ int do_settimeofday(struct timespec *tv) xtime = *tv; - update_xtime_cache(0); - - clock->error = 0; + timekeeper.ntp_error = 0; ntp_clear(); - update_vsyscall(&xtime, clock); + update_vsyscall(&xtime, timekeeper.clock, timekeeper.mult); write_sequnlock_irqrestore(&xtime_lock, flags); @@ -180,39 +345,97 @@ EXPORT_SYMBOL(do_settimeofday); * * Accumulates current time interval and initializes new clocksource */ -static void change_clocksource(void) +static int change_clocksource(void *data) { - struct clocksource *new; + struct clocksource *new, *old; + + new = (struct clocksource *) data; - new = clocksource_get_next(); + timekeeping_forward_now(); + if (!new->enable || new->enable(new) == 0) { + old = timekeeper.clock; + timekeeper_setup_internals(new); + if (old->disable) + old->disable(old); + } + return 0; +} - if (clock == new) +/** + * timekeeping_notify - Install a new clock source + * @clock: pointer to the clock source + * + * This function is called from clocksource.c after a new, better clock + * source has been registered. The caller holds the clocksource_mutex. + */ +void timekeeping_notify(struct clocksource *clock) +{ + if (timekeeper.clock == clock) return; + stop_machine(change_clocksource, clock, NULL); + tick_clock_notify(); +} - clocksource_forward_now(); +#else /* GENERIC_TIME */ - new->raw_time = clock->raw_time; +static inline void timekeeping_forward_now(void) { } - clock = new; - clock->cycle_last = 0; - clock->cycle_last = clocksource_read(new); - clock->error = 0; - clock->xtime_nsec = 0; - clocksource_calculate_interval(clock, NTP_INTERVAL_LENGTH); +/** + * ktime_get - get the monotonic time in ktime_t format + * + * returns the time in ktime_t format + */ +ktime_t ktime_get(void) +{ + struct timespec now; - tick_clock_notify(); + ktime_get_ts(&now); - /* - * We're holding xtime lock and waking up klogd would deadlock - * us on enqueue. So no printing! - printk(KERN_INFO "Time: %s clocksource has been installed.\n", - clock->name); - */ + return timespec_to_ktime(now); } -#else -static inline void clocksource_forward_now(void) { } -static inline void change_clocksource(void) { } -#endif +EXPORT_SYMBOL_GPL(ktime_get); + +/** + * ktime_get_ts - get the monotonic clock in timespec format + * @ts: pointer to timespec variable + * + * The function calculates the monotonic clock from the realtime + * clock and the wall_to_monotonic offset and stores the result + * in normalized timespec format in the variable pointed to by @ts. + */ +void ktime_get_ts(struct timespec *ts) +{ + struct timespec tomono; + unsigned long seq; + + do { + seq = read_seqbegin(&xtime_lock); + getnstimeofday(ts); + tomono = wall_to_monotonic; + + } while (read_seqretry(&xtime_lock, seq)); + + set_normalized_timespec(ts, ts->tv_sec + tomono.tv_sec, + ts->tv_nsec + tomono.tv_nsec); +} +EXPORT_SYMBOL_GPL(ktime_get_ts); + +#endif /* !GENERIC_TIME */ + +/** + * ktime_get_real - get the real (wall-) time in ktime_t format + * + * returns the time in ktime_t format + */ +ktime_t ktime_get_real(void) +{ + struct timespec now; + + getnstimeofday(&now); + + return timespec_to_ktime(now); +} +EXPORT_SYMBOL_GPL(ktime_get_real); /** * getrawmonotonic - Returns the raw monotonic time in a timespec @@ -224,21 +447,11 @@ void getrawmonotonic(struct timespec *ts) { unsigned long seq; s64 nsecs; - cycle_t cycle_now, cycle_delta; do { seq = read_seqbegin(&xtime_lock); - - /* read clocksource: */ - cycle_now = clocksource_read(clock); - - /* calculate the delta since the last update_wall_time: */ - cycle_delta = (cycle_now - clock->cycle_last) & clock->mask; - - /* convert to nanoseconds: */ - nsecs = ((s64)cycle_delta * clock->mult_orig) >> clock->shift; - - *ts = clock->raw_time; + nsecs = timekeeping_get_ns_raw(); + *ts = raw_time; } while (read_seqretry(&xtime_lock, seq)); @@ -258,7 +471,7 @@ int timekeeping_valid_for_hres(void) do { seq = read_seqbegin(&xtime_lock); - ret = clock->flags & CLOCK_SOURCE_VALID_FOR_HRES; + ret = timekeeper.clock->flags & CLOCK_SOURCE_VALID_FOR_HRES; } while (read_seqretry(&xtime_lock, seq)); @@ -266,17 +479,44 @@ int timekeeping_valid_for_hres(void) } /** - * read_persistent_clock - Return time in seconds from the persistent clock. + * timekeeping_max_deferment - Returns max time the clocksource can be deferred + * + * Caller must observe xtime_lock via read_seqbegin/read_seqretry to + * ensure that the clocksource does not change! + */ +u64 timekeeping_max_deferment(void) +{ + return timekeeper.clock->max_idle_ns; +} + +/** + * read_persistent_clock - Return time from the persistent clock. * * Weak dummy function for arches that do not yet support it. - * Returns seconds from epoch using the battery backed persistent clock. - * Returns zero if unsupported. + * Reads the time from the battery backed persistent clock. + * Returns a timespec with tv_sec=0 and tv_nsec=0 if unsupported. * * XXX - Do be sure to remove it once all arches implement it. */ -unsigned long __attribute__((weak)) read_persistent_clock(void) +void __attribute__((weak)) read_persistent_clock(struct timespec *ts) { - return 0; + ts->tv_sec = 0; + ts->tv_nsec = 0; +} + +/** + * read_boot_clock - Return time of the system start. + * + * Weak dummy function for arches that do not yet support it. + * Function to read the exact time the system has been started. + * Returns a timespec with tv_sec=0 and tv_nsec=0 if unsupported. + * + * XXX - Do be sure to remove it once all arches implement it. + */ +void __attribute__((weak)) read_boot_clock(struct timespec *ts) +{ + ts->tv_sec = 0; + ts->tv_nsec = 0; } /* @@ -284,28 +524,39 @@ unsigned long __attribute__((weak)) read_persistent_clock(void) */ void __init timekeeping_init(void) { + struct clocksource *clock; unsigned long flags; - unsigned long sec = read_persistent_clock(); + struct timespec now, boot; + + read_persistent_clock(&now); + read_boot_clock(&boot); write_seqlock_irqsave(&xtime_lock, flags); ntp_init(); - clock = clocksource_get_next(); - clocksource_calculate_interval(clock, NTP_INTERVAL_LENGTH); - clock->cycle_last = clocksource_read(clock); - - xtime.tv_sec = sec; - xtime.tv_nsec = 0; + clock = clocksource_default_clock(); + if (clock->enable) + clock->enable(clock); + timekeeper_setup_internals(clock); + + xtime.tv_sec = now.tv_sec; + xtime.tv_nsec = now.tv_nsec; + raw_time.tv_sec = 0; + raw_time.tv_nsec = 0; + if (boot.tv_sec == 0 && boot.tv_nsec == 0) { + boot.tv_sec = xtime.tv_sec; + boot.tv_nsec = xtime.tv_nsec; + } set_normalized_timespec(&wall_to_monotonic, - -xtime.tv_sec, -xtime.tv_nsec); - update_xtime_cache(0); - total_sleep_time = 0; + -boot.tv_sec, -boot.tv_nsec); + total_sleep_time.tv_sec = 0; + total_sleep_time.tv_nsec = 0; write_sequnlock_irqrestore(&xtime_lock, flags); } /* time in seconds when suspend began */ -static unsigned long timekeeping_suspend_time; +static struct timespec timekeeping_suspend_time; /** * timekeeping_resume - Resumes the generic timekeeping subsystem. @@ -318,24 +569,23 @@ static unsigned long timekeeping_suspend_time; static int timekeeping_resume(struct sys_device *dev) { unsigned long flags; - unsigned long now = read_persistent_clock(); + struct timespec ts; + + read_persistent_clock(&ts); clocksource_resume(); write_seqlock_irqsave(&xtime_lock, flags); - if (now && (now > timekeeping_suspend_time)) { - unsigned long sleep_length = now - timekeeping_suspend_time; - - xtime.tv_sec += sleep_length; - wall_to_monotonic.tv_sec -= sleep_length; - total_sleep_time += sleep_length; + if (timespec_compare(&ts, &timekeeping_suspend_time) > 0) { + ts = timespec_sub(ts, timekeeping_suspend_time); + xtime = timespec_add_safe(xtime, ts); + wall_to_monotonic = timespec_sub(wall_to_monotonic, ts); + total_sleep_time = timespec_add_safe(total_sleep_time, ts); } - update_xtime_cache(0); /* re-base the last cycle value */ - clock->cycle_last = 0; - clock->cycle_last = clocksource_read(clock); - clock->error = 0; + timekeeper.clock->cycle_last = timekeeper.clock->read(timekeeper.clock); + timekeeper.ntp_error = 0; timekeeping_suspended = 0; write_sequnlock_irqrestore(&xtime_lock, flags); @@ -353,10 +603,10 @@ static int timekeeping_suspend(struct sys_device *dev, pm_message_t state) { unsigned long flags; - timekeeping_suspend_time = read_persistent_clock(); + read_persistent_clock(&timekeeping_suspend_time); write_seqlock_irqsave(&xtime_lock, flags); - clocksource_forward_now(); + timekeeping_forward_now(); timekeeping_suspended = 1; write_sequnlock_irqrestore(&xtime_lock, flags); @@ -391,7 +641,7 @@ device_initcall(timekeeping_init_device); * If the error is already larger, we look ahead even further * to compensate for late or lost adjustments. */ -static __always_inline int clocksource_bigadjust(s64 error, s64 *interval, +static __always_inline int timekeeping_bigadjust(s64 error, s64 *interval, s64 *offset) { s64 tick_error, i; @@ -407,7 +657,7 @@ static __always_inline int clocksource_bigadjust(s64 error, s64 *interval, * here. This is tuned so that an error of about 1 msec is adjusted * within about 1 sec (or 2^20 nsec in 2^SHIFT_HZ ticks). */ - error2 = clock->error >> (NTP_SCALE_SHIFT + 22 - 2 * SHIFT_HZ); + error2 = timekeeper.ntp_error >> (NTP_SCALE_SHIFT + 22 - 2 * SHIFT_HZ); error2 = abs(error2); for (look_ahead = 0; error2 > 0; look_ahead++) error2 >>= 2; @@ -416,8 +666,8 @@ static __always_inline int clocksource_bigadjust(s64 error, s64 *interval, * Now calculate the error in (1 << look_ahead) ticks, but first * remove the single look ahead already included in the error. */ - tick_error = tick_length >> (NTP_SCALE_SHIFT - clock->shift + 1); - tick_error -= clock->xtime_interval >> 1; + tick_error = tick_length >> (timekeeper.ntp_error_shift + 1); + tick_error -= timekeeper.xtime_interval >> 1; error = ((error - tick_error) >> look_ahead) + tick_error; /* Finally calculate the adjustment shift value. */ @@ -442,18 +692,18 @@ static __always_inline int clocksource_bigadjust(s64 error, s64 *interval, * this is optimized for the most common adjustments of -1,0,1, * for other values we can do a bit more work. */ -static void clocksource_adjust(s64 offset) +static void timekeeping_adjust(s64 offset) { - s64 error, interval = clock->cycle_interval; + s64 error, interval = timekeeper.cycle_interval; int adj; - error = clock->error >> (NTP_SCALE_SHIFT - clock->shift - 1); + error = timekeeper.ntp_error >> (timekeeper.ntp_error_shift - 1); if (error > interval) { error >>= 2; if (likely(error <= interval)) adj = 1; else - adj = clocksource_bigadjust(error, &interval, &offset); + adj = timekeeping_bigadjust(error, &interval, &offset); } else if (error < -interval) { error >>= 2; if (likely(error >= -interval)) { @@ -461,15 +711,58 @@ static void clocksource_adjust(s64 offset) interval = -interval; offset = -offset; } else - adj = clocksource_bigadjust(error, &interval, &offset); + adj = timekeeping_bigadjust(error, &interval, &offset); } else return; - clock->mult += adj; - clock->xtime_interval += interval; - clock->xtime_nsec -= offset; - clock->error -= (interval - offset) << - (NTP_SCALE_SHIFT - clock->shift); + timekeeper.mult += adj; + timekeeper.xtime_interval += interval; + timekeeper.xtime_nsec -= offset; + timekeeper.ntp_error -= (interval - offset) << + timekeeper.ntp_error_shift; +} + +/** + * logarithmic_accumulation - shifted accumulation of cycles + * + * This functions accumulates a shifted interval of cycles into + * into a shifted interval nanoseconds. Allows for O(log) accumulation + * loop. + * + * Returns the unconsumed cycles. + */ +static cycle_t logarithmic_accumulation(cycle_t offset, int shift) +{ + u64 nsecps = (u64)NSEC_PER_SEC << timekeeper.shift; + + /* If the offset is smaller then a shifted interval, do nothing */ + if (offset < timekeeper.cycle_interval<<shift) + return offset; + + /* Accumulate one shifted interval */ + offset -= timekeeper.cycle_interval << shift; + timekeeper.clock->cycle_last += timekeeper.cycle_interval << shift; + + timekeeper.xtime_nsec += timekeeper.xtime_interval << shift; + while (timekeeper.xtime_nsec >= nsecps) { + timekeeper.xtime_nsec -= nsecps; + xtime.tv_sec++; + second_overflow(); + } + + /* Accumulate into raw time */ + raw_time.tv_nsec += timekeeper.raw_interval << shift;; + while (raw_time.tv_nsec >= NSEC_PER_SEC) { + raw_time.tv_nsec -= NSEC_PER_SEC; + raw_time.tv_sec++; + } + + /* Accumulate error between NTP and clock interval */ + timekeeper.ntp_error += tick_length << shift; + timekeeper.ntp_error -= timekeeper.xtime_interval << + (timekeeper.ntp_error_shift + shift); + + return offset; } /** @@ -479,53 +772,48 @@ static void clocksource_adjust(s64 offset) */ void update_wall_time(void) { + struct clocksource *clock; cycle_t offset; + int shift = 0, maxshift; /* Make sure we're fully resumed: */ if (unlikely(timekeeping_suspended)) return; + clock = timekeeper.clock; #ifdef CONFIG_GENERIC_TIME - offset = (clocksource_read(clock) - clock->cycle_last) & clock->mask; + offset = (clock->read(clock) - clock->cycle_last) & clock->mask; #else - offset = clock->cycle_interval; + offset = timekeeper.cycle_interval; #endif - clock->xtime_nsec = (s64)xtime.tv_nsec << clock->shift; + timekeeper.xtime_nsec = (s64)xtime.tv_nsec << timekeeper.shift; - /* normally this loop will run just once, however in the - * case of lost or late ticks, it will accumulate correctly. + /* + * With NO_HZ we may have to accumulate many cycle_intervals + * (think "ticks") worth of time at once. To do this efficiently, + * we calculate the largest doubling multiple of cycle_intervals + * that is smaller then the offset. We then accumulate that + * chunk in one go, and then try to consume the next smaller + * doubled multiple. */ - while (offset >= clock->cycle_interval) { - /* accumulate one interval */ - offset -= clock->cycle_interval; - clock->cycle_last += clock->cycle_interval; - - clock->xtime_nsec += clock->xtime_interval; - if (clock->xtime_nsec >= (u64)NSEC_PER_SEC << clock->shift) { - clock->xtime_nsec -= (u64)NSEC_PER_SEC << clock->shift; - xtime.tv_sec++; - second_overflow(); - } - - clock->raw_time.tv_nsec += clock->raw_interval; - if (clock->raw_time.tv_nsec >= NSEC_PER_SEC) { - clock->raw_time.tv_nsec -= NSEC_PER_SEC; - clock->raw_time.tv_sec++; - } - - /* accumulate error between NTP and clock interval */ - clock->error += tick_length; - clock->error -= clock->xtime_interval << (NTP_SCALE_SHIFT - clock->shift); + shift = ilog2(offset) - ilog2(timekeeper.cycle_interval); + shift = max(0, shift); + /* Bound shift to one less then what overflows tick_length */ + maxshift = (8*sizeof(tick_length) - (ilog2(tick_length)+1)) - 1; + shift = min(shift, maxshift); + while (offset >= timekeeper.cycle_interval) { + offset = logarithmic_accumulation(offset, shift); + shift--; } /* correct the clock when NTP error is too big */ - clocksource_adjust(offset); + timekeeping_adjust(offset); /* * Since in the loop above, we accumulate any amount of time * in xtime_nsec over a second into xtime.tv_sec, its possible for * xtime_nsec to be fairly small after the loop. Further, if we're - * slightly speeding the clocksource up in clocksource_adjust(), + * slightly speeding the clocksource up in timekeeping_adjust(), * its possible the required corrective factor to xtime_nsec could * cause it to underflow. * @@ -537,24 +825,22 @@ void update_wall_time(void) * We'll correct this error next time through this function, when * xtime_nsec is not as small. */ - if (unlikely((s64)clock->xtime_nsec < 0)) { - s64 neg = -(s64)clock->xtime_nsec; - clock->xtime_nsec = 0; - clock->error += neg << (NTP_SCALE_SHIFT - clock->shift); + if (unlikely((s64)timekeeper.xtime_nsec < 0)) { + s64 neg = -(s64)timekeeper.xtime_nsec; + timekeeper.xtime_nsec = 0; + timekeeper.ntp_error += neg << timekeeper.ntp_error_shift; } /* store full nanoseconds into xtime after rounding it up and * add the remainder to the error difference. */ - xtime.tv_nsec = ((s64)clock->xtime_nsec >> clock->shift) + 1; - clock->xtime_nsec -= (s64)xtime.tv_nsec << clock->shift; - clock->error += clock->xtime_nsec << (NTP_SCALE_SHIFT - clock->shift); - - update_xtime_cache(cyc2ns(clock, offset)); + xtime.tv_nsec = ((s64) timekeeper.xtime_nsec >> timekeeper.shift) + 1; + timekeeper.xtime_nsec -= (s64) xtime.tv_nsec << timekeeper.shift; + timekeeper.ntp_error += timekeeper.xtime_nsec << + timekeeper.ntp_error_shift; /* check to see if there is a new clocksource to use */ - change_clocksource(); - update_vsyscall(&xtime, clock); + update_vsyscall(&xtime, timekeeper.clock, timekeeper.mult); } /** @@ -570,9 +856,12 @@ void update_wall_time(void) */ void getboottime(struct timespec *ts) { - set_normalized_timespec(ts, - - (wall_to_monotonic.tv_sec + total_sleep_time), - - wall_to_monotonic.tv_nsec); + struct timespec boottime = { + .tv_sec = wall_to_monotonic.tv_sec + total_sleep_time.tv_sec, + .tv_nsec = wall_to_monotonic.tv_nsec + total_sleep_time.tv_nsec + }; + + set_normalized_timespec(ts, -boottime.tv_sec, -boottime.tv_nsec); } /** @@ -581,15 +870,19 @@ void getboottime(struct timespec *ts) */ void monotonic_to_bootbased(struct timespec *ts) { - ts->tv_sec += total_sleep_time; + *ts = timespec_add_safe(*ts, total_sleep_time); } unsigned long get_seconds(void) { - return xtime_cache.tv_sec; + return xtime.tv_sec; } EXPORT_SYMBOL(get_seconds); +struct timespec __current_kernel_time(void) +{ + return xtime; +} struct timespec current_kernel_time(void) { @@ -598,10 +891,25 @@ struct timespec current_kernel_time(void) do { seq = read_seqbegin(&xtime_lock); - - now = xtime_cache; + now = xtime; } while (read_seqretry(&xtime_lock, seq)); return now; } EXPORT_SYMBOL(current_kernel_time); + +struct timespec get_monotonic_coarse(void) +{ + struct timespec now, mono; + unsigned long seq; + + do { + seq = read_seqbegin(&xtime_lock); + now = xtime; + mono = wall_to_monotonic; + } while (read_seqretry(&xtime_lock, seq)); + + set_normalized_timespec(&now, now.tv_sec + mono.tv_sec, + now.tv_nsec + mono.tv_nsec); + return now; +} diff --git a/kernel/time/timer_list.c b/kernel/time/timer_list.c index a999b92a127..665c76edbf1 100644 --- a/kernel/time/timer_list.c +++ b/kernel/time/timer_list.c @@ -204,10 +204,12 @@ print_tickdevice(struct seq_file *m, struct tick_device *td, int cpu) return; } SEQ_printf(m, "%s\n", dev->name); - SEQ_printf(m, " max_delta_ns: %lu\n", dev->max_delta_ns); - SEQ_printf(m, " min_delta_ns: %lu\n", dev->min_delta_ns); - SEQ_printf(m, " mult: %lu\n", dev->mult); - SEQ_printf(m, " shift: %d\n", dev->shift); + SEQ_printf(m, " max_delta_ns: %llu\n", + (unsigned long long) dev->max_delta_ns); + SEQ_printf(m, " min_delta_ns: %llu\n", + (unsigned long long) dev->min_delta_ns); + SEQ_printf(m, " mult: %u\n", dev->mult); + SEQ_printf(m, " shift: %u\n", dev->shift); SEQ_printf(m, " mode: %d\n", dev->mode); SEQ_printf(m, " next_event: %Ld nsecs\n", (unsigned long long) ktime_to_ns(dev->next_event)); @@ -275,7 +277,7 @@ static int timer_list_open(struct inode *inode, struct file *filp) return single_open(filp, timer_list_show, NULL); } -static struct file_operations timer_list_fops = { +static const struct file_operations timer_list_fops = { .open = timer_list_open, .read = seq_read, .llseek = seq_lseek, @@ -286,7 +288,7 @@ static int __init init_timer_list_procfs(void) { struct proc_dir_entry *pe; - pe = proc_create("timer_list", 0644, NULL, &timer_list_fops); + pe = proc_create("timer_list", 0444, NULL, &timer_list_fops); if (!pe) return -ENOMEM; return 0; diff --git a/kernel/time/timer_stats.c b/kernel/time/timer_stats.c index c994530d166..ee5681f8d7e 100644 --- a/kernel/time/timer_stats.c +++ b/kernel/time/timer_stats.c @@ -96,7 +96,7 @@ static DEFINE_MUTEX(show_mutex); /* * Collection status, active/inactive: */ -static int __read_mostly active; +int __read_mostly timer_stats_active; /* * Beginning/end timestamps of measurement: @@ -242,7 +242,7 @@ void timer_stats_update_stats(void *timer, pid_t pid, void *startf, struct entry *entry, input; unsigned long flags; - if (likely(!active)) + if (likely(!timer_stats_active)) return; lock = &per_cpu(lookup_lock, raw_smp_processor_id()); @@ -254,7 +254,7 @@ void timer_stats_update_stats(void *timer, pid_t pid, void *startf, input.timer_flag = timer_flag; spin_lock_irqsave(lock, flags); - if (!active) + if (!timer_stats_active) goto out_unlock; entry = tstat_lookup(&input, comm); @@ -290,7 +290,7 @@ static int tstats_show(struct seq_file *m, void *v) /* * If still active then calculate up to now: */ - if (active) + if (timer_stats_active) time_stop = ktime_get(); time = ktime_sub(time_stop, time_start); @@ -368,18 +368,18 @@ static ssize_t tstats_write(struct file *file, const char __user *buf, mutex_lock(&show_mutex); switch (ctl[0]) { case '0': - if (active) { - active = 0; + if (timer_stats_active) { + timer_stats_active = 0; time_stop = ktime_get(); sync_access(); } break; case '1': - if (!active) { + if (!timer_stats_active) { reset_entries(); time_start = ktime_get(); smp_mb(); - active = 1; + timer_stats_active = 1; } break; default: @@ -395,7 +395,7 @@ static int tstats_open(struct inode *inode, struct file *filp) return single_open(filp, tstats_show, NULL); } -static struct file_operations tstats_fops = { +static const struct file_operations tstats_fops = { .open = tstats_open, .read = seq_read, .write = tstats_write, diff --git a/kernel/timer.c b/kernel/timer.c index b4555568b4e..5db5a8d2681 100644 --- a/kernel/timer.c +++ b/kernel/timer.c @@ -37,6 +37,8 @@ #include <linux/delay.h> #include <linux/tick.h> #include <linux/kallsyms.h> +#include <linux/perf_event.h> +#include <linux/sched.h> #include <asm/uaccess.h> #include <asm/unistd.h> @@ -44,6 +46,9 @@ #include <asm/timex.h> #include <asm/io.h> +#define CREATE_TRACE_POINTS +#include <trace/events/timer.h> + u64 jiffies_64 __cacheline_aligned_in_smp = INITIAL_JIFFIES; EXPORT_SYMBOL(jiffies_64); @@ -70,6 +75,7 @@ struct tvec_base { spinlock_t lock; struct timer_list *running_timer; unsigned long timer_jiffies; + unsigned long next_timer; struct tvec_root tv1; struct tvec tv2; struct tvec tv3; @@ -378,6 +384,8 @@ static void timer_stats_account_timer(struct timer_list *timer) { unsigned int flag = 0; + if (likely(!timer->start_site)) + return; if (unlikely(tbase_get_deferrable(timer->base))) flag |= TIMER_STATS_FLAG_DEFERRABLE; @@ -516,6 +524,25 @@ static inline void debug_timer_activate(struct timer_list *timer) { } static inline void debug_timer_deactivate(struct timer_list *timer) { } #endif +static inline void debug_init(struct timer_list *timer) +{ + debug_timer_init(timer); + trace_timer_init(timer); +} + +static inline void +debug_activate(struct timer_list *timer, unsigned long expires) +{ + debug_timer_activate(timer); + trace_timer_start(timer, expires); +} + +static inline void debug_deactivate(struct timer_list *timer) +{ + debug_timer_deactivate(timer); + trace_timer_cancel(timer); +} + static void __init_timer(struct timer_list *timer, const char *name, struct lock_class_key *key) @@ -531,17 +558,20 @@ static void __init_timer(struct timer_list *timer, } /** - * init_timer - initialize a timer. + * init_timer_key - initialize a timer * @timer: the timer to be initialized + * @name: name of the timer + * @key: lockdep class key of the fake lock used for tracking timer + * sync lock dependencies * - * init_timer() must be done to a timer prior calling *any* of the + * init_timer_key() must be done to a timer prior calling *any* of the * other timer functions. */ void init_timer_key(struct timer_list *timer, const char *name, struct lock_class_key *key) { - debug_timer_init(timer); + debug_init(timer); __init_timer(timer, name, key); } EXPORT_SYMBOL(init_timer_key); @@ -560,7 +590,7 @@ static inline void detach_timer(struct timer_list *timer, { struct list_head *entry = &timer->entry; - debug_timer_deactivate(timer); + debug_deactivate(timer); __list_del(entry->prev, entry->next); if (clear_pending) @@ -601,13 +631,12 @@ static struct tvec_base *lock_timer_base(struct timer_list *timer, } static inline int -__mod_timer(struct timer_list *timer, unsigned long expires, bool pending_only) +__mod_timer(struct timer_list *timer, unsigned long expires, + bool pending_only, int pinned) { struct tvec_base *base, *new_base; unsigned long flags; - int ret; - - ret = 0; + int ret = 0 , cpu; timer_stats_timer_set_start_info(timer); BUG_ON(!timer->function); @@ -616,16 +645,31 @@ __mod_timer(struct timer_list *timer, unsigned long expires, bool pending_only) if (timer_pending(timer)) { detach_timer(timer, 0); + if (timer->expires == base->next_timer && + !tbase_get_deferrable(timer->base)) + base->next_timer = base->timer_jiffies; ret = 1; } else { if (pending_only) goto out_unlock; } - debug_timer_activate(timer); + debug_activate(timer, expires); new_base = __get_cpu_var(tvec_bases); + cpu = smp_processor_id(); + +#if defined(CONFIG_NO_HZ) && defined(CONFIG_SMP) + if (!pinned && get_sysctl_timer_migration() && idle_cpu(cpu)) { + int preferred_cpu = get_nohz_load_balancer(); + + if (preferred_cpu >= 0) + cpu = preferred_cpu; + } +#endif + new_base = per_cpu(tvec_bases, cpu); + if (base != new_base) { /* * We are trying to schedule the timer on the local CPU. @@ -645,6 +689,9 @@ __mod_timer(struct timer_list *timer, unsigned long expires, bool pending_only) } timer->expires = expires; + if (time_before(timer->expires, base->next_timer) && + !tbase_get_deferrable(timer->base)) + base->next_timer = timer->expires; internal_add_timer(base, timer); out_unlock: @@ -665,7 +712,7 @@ out_unlock: */ int mod_timer_pending(struct timer_list *timer, unsigned long expires) { - return __mod_timer(timer, expires, true); + return __mod_timer(timer, expires, true, TIMER_NOT_PINNED); } EXPORT_SYMBOL(mod_timer_pending); @@ -696,14 +743,36 @@ int mod_timer(struct timer_list *timer, unsigned long expires) * networking code - if the timer is re-modified * to be the same thing then just return: */ - if (timer->expires == expires && timer_pending(timer)) + if (timer_pending(timer) && timer->expires == expires) return 1; - return __mod_timer(timer, expires, false); + return __mod_timer(timer, expires, false, TIMER_NOT_PINNED); } EXPORT_SYMBOL(mod_timer); /** + * mod_timer_pinned - modify a timer's timeout + * @timer: the timer to be modified + * @expires: new timeout in jiffies + * + * mod_timer_pinned() is a way to update the expire field of an + * active timer (if the timer is inactive it will be activated) + * and not allow the timer to be migrated to a different CPU. + * + * mod_timer_pinned(timer, expires) is equivalent to: + * + * del_timer(timer); timer->expires = expires; add_timer(timer); + */ +int mod_timer_pinned(struct timer_list *timer, unsigned long expires) +{ + if (timer->expires == expires && timer_pending(timer)) + return 1; + + return __mod_timer(timer, expires, false, TIMER_PINNED); +} +EXPORT_SYMBOL(mod_timer_pinned); + +/** * add_timer - start a timer * @timer: the timer to be added * @@ -740,7 +809,10 @@ void add_timer_on(struct timer_list *timer, int cpu) BUG_ON(timer_pending(timer) || !timer->function); spin_lock_irqsave(&base->lock, flags); timer_set_base(timer, base); - debug_timer_activate(timer); + debug_activate(timer, timer->expires); + if (time_before(timer->expires, base->next_timer) && + !tbase_get_deferrable(timer->base)) + base->next_timer = timer->expires; internal_add_timer(base, timer); /* * Check whether the other CPU is idle and needs to be @@ -753,6 +825,7 @@ void add_timer_on(struct timer_list *timer, int cpu) wake_up_idle_cpu(cpu); spin_unlock_irqrestore(&base->lock, flags); } +EXPORT_SYMBOL_GPL(add_timer_on); /** * del_timer - deactive a timer. @@ -776,6 +849,9 @@ int del_timer(struct timer_list *timer) base = lock_timer_base(timer, &flags); if (timer_pending(timer)) { detach_timer(timer, 1); + if (timer->expires == base->next_timer && + !tbase_get_deferrable(timer->base)) + base->next_timer = base->timer_jiffies; ret = 1; } spin_unlock_irqrestore(&base->lock, flags); @@ -809,6 +885,9 @@ int try_to_del_timer_sync(struct timer_list *timer) ret = 0; if (timer_pending(timer)) { detach_timer(timer, 1); + if (timer->expires == base->next_timer && + !tbase_get_deferrable(timer->base)) + base->next_timer = base->timer_jiffies; ret = 1; } out: @@ -943,7 +1022,9 @@ static inline void __run_timers(struct tvec_base *base) */ lock_map_acquire(&lockdep_map); + trace_timer_expire_entry(timer); fn(data); + trace_timer_expire_exit(timer); lock_map_release(&lockdep_map); @@ -966,8 +1047,8 @@ static inline void __run_timers(struct tvec_base *base) #ifdef CONFIG_NO_HZ /* * Find out when the next timer event is due to happen. This - * is used on S/390 to stop all activity when a cpus is idle. - * This functions needs to be called disabled. + * is used on S/390 to stop all activity when a CPU is idle. + * This function needs to be called with interrupts disabled. */ static unsigned long __next_timer_interrupt(struct tvec_base *base) { @@ -1012,6 +1093,9 @@ cascade: index = slot = timer_jiffies & TVN_MASK; do { list_for_each_entry(nte, varp->vec + slot, entry) { + if (tbase_get_deferrable(nte->base)) + continue; + found = 1; if (time_before(nte->expires, expires)) expires = nte->expires; @@ -1090,7 +1174,9 @@ unsigned long get_next_timer_interrupt(unsigned long now) unsigned long expires; spin_lock(&base->lock); - expires = __next_timer_interrupt(base); + if (time_before_eq(base->next_timer, base->timer_jiffies)) + base->next_timer = __next_timer_interrupt(base); + expires = base->next_timer; spin_unlock(&base->lock); if (time_before_eq(expires, now)) @@ -1112,61 +1198,21 @@ void update_process_times(int user_tick) /* Note: this timer irq context must be accounted for as well. */ account_process_tick(p, user_tick); run_local_timers(); - if (rcu_pending(cpu)) - rcu_check_callbacks(cpu, user_tick); + rcu_check_callbacks(cpu, user_tick); printk_tick(); scheduler_tick(); run_posix_cpu_timers(p); } /* - * Nr of active tasks - counted in fixed-point numbers - */ -static unsigned long count_active_tasks(void) -{ - return nr_active() * FIXED_1; -} - -/* - * Hmm.. Changed this, as the GNU make sources (load.c) seems to - * imply that avenrun[] is the standard name for this kind of thing. - * Nothing else seems to be standardized: the fractional size etc - * all seem to differ on different machines. - * - * Requires xtime_lock to access. - */ -unsigned long avenrun[3]; - -EXPORT_SYMBOL(avenrun); - -/* - * calc_load - given tick count, update the avenrun load estimates. - * This is called while holding a write_lock on xtime_lock. - */ -static inline void calc_load(unsigned long ticks) -{ - unsigned long active_tasks; /* fixed-point */ - static int count = LOAD_FREQ; - - count -= ticks; - if (unlikely(count < 0)) { - active_tasks = count_active_tasks(); - do { - CALC_LOAD(avenrun[0], EXP_1, active_tasks); - CALC_LOAD(avenrun[1], EXP_5, active_tasks); - CALC_LOAD(avenrun[2], EXP_15, active_tasks); - count += LOAD_FREQ; - } while (count < 0); - } -} - -/* * This function runs timers and the timer-tq in bottom half context. */ static void run_timer_softirq(struct softirq_action *h) { struct tvec_base *base = __get_cpu_var(tvec_bases); + perf_event_do_pending(); + hrtimer_run_pending(); if (time_after_eq(jiffies, base->timer_jiffies)) @@ -1184,16 +1230,6 @@ void run_local_timers(void) } /* - * Called by the timer interrupt. xtime_lock must already be taken - * by the timer IRQ! - */ -static inline void update_times(unsigned long ticks) -{ - update_wall_time(); - calc_load(ticks); -} - -/* * The 64-bit jiffies value is not atomic - you MUST NOT read it * without sampling the sequence number in xtime_lock. * jiffies is defined in the linker script... @@ -1202,7 +1238,8 @@ static inline void update_times(unsigned long ticks) void do_timer(unsigned long ticks) { jiffies_64 += ticks; - update_times(ticks); + update_wall_time(); + calc_global_load(); } #ifdef __ARCH_WANT_SYS_ALARM @@ -1350,7 +1387,7 @@ signed long __sched schedule_timeout(signed long timeout) expire = timeout + jiffies; setup_timer_on_stack(&timer, process_timeout, (unsigned long)current); - __mod_timer(&timer, expire, false); + __mod_timer(&timer, expire, false, TIMER_NOT_PINNED); schedule(); del_singleshot_timer_sync(&timer); @@ -1403,37 +1440,17 @@ int do_sysinfo(struct sysinfo *info) { unsigned long mem_total, sav_total; unsigned int mem_unit, bitcount; - unsigned long seq; + struct timespec tp; memset(info, 0, sizeof(struct sysinfo)); - do { - struct timespec tp; - seq = read_seqbegin(&xtime_lock); - - /* - * This is annoying. The below is the same thing - * posix_get_clock_monotonic() does, but it wants to - * take the lock which we want to cover the loads stuff - * too. - */ - - getnstimeofday(&tp); - tp.tv_sec += wall_to_monotonic.tv_sec; - tp.tv_nsec += wall_to_monotonic.tv_nsec; - monotonic_to_bootbased(&tp); - if (tp.tv_nsec - NSEC_PER_SEC >= 0) { - tp.tv_nsec = tp.tv_nsec - NSEC_PER_SEC; - tp.tv_sec++; - } - info->uptime = tp.tv_sec + (tp.tv_nsec ? 1 : 0); + ktime_get_ts(&tp); + monotonic_to_bootbased(&tp); + info->uptime = tp.tv_sec + (tp.tv_nsec ? 1 : 0); - info->loads[0] = avenrun[0] << (SI_LOAD_SHIFT - FSHIFT); - info->loads[1] = avenrun[1] << (SI_LOAD_SHIFT - FSHIFT); - info->loads[2] = avenrun[2] << (SI_LOAD_SHIFT - FSHIFT); + get_avenrun(info->loads, 0, SI_LOAD_SHIFT - FSHIFT); - info->procs = nr_threads; - } while (read_seqretry(&xtime_lock, seq)); + info->procs = nr_threads; si_meminfo(info); si_swapinfo(info); @@ -1547,6 +1564,7 @@ static int __cpuinit init_timers_cpu(int cpu) INIT_LIST_HEAD(base->tv1.vec + j); base->timer_jiffies = jiffies; + base->next_timer = base->timer_jiffies; return 0; } @@ -1559,6 +1577,9 @@ static void migrate_timer_list(struct tvec_base *new_base, struct list_head *hea timer = list_first_entry(head, struct timer_list, entry); detach_timer(timer, 0); timer_set_base(timer, new_base); + if (time_before(timer->expires, new_base->next_timer) && + !tbase_get_deferrable(timer->base)) + new_base->next_timer = timer->expires; internal_add_timer(new_base, timer); } } diff --git a/kernel/trace/Kconfig b/kernel/trace/Kconfig index 34e707e5ab8..d006554888d 100644 --- a/kernel/trace/Kconfig +++ b/kernel/trace/Kconfig @@ -9,34 +9,82 @@ config USER_STACKTRACE_SUPPORT config NOP_TRACER bool +config HAVE_FTRACE_NMI_ENTER + bool + help + See Documentation/trace/ftrace-implementation.txt + config HAVE_FUNCTION_TRACER bool + help + See Documentation/trace/ftrace-implementation.txt config HAVE_FUNCTION_GRAPH_TRACER bool + help + See Documentation/trace/ftrace-implementation.txt + +config HAVE_FUNCTION_GRAPH_FP_TEST + bool + help + An arch may pass in a unique value (frame pointer) to both the + entering and exiting of a function. On exit, the value is compared + and if it does not match, then it will panic the kernel. config HAVE_FUNCTION_TRACE_MCOUNT_TEST bool help - This gets selected when the arch tests the function_trace_stop - variable at the mcount call site. Otherwise, this variable - is tested by the called function. + See Documentation/trace/ftrace-implementation.txt config HAVE_DYNAMIC_FTRACE bool + help + See Documentation/trace/ftrace-implementation.txt config HAVE_FTRACE_MCOUNT_RECORD bool + help + See Documentation/trace/ftrace-implementation.txt config HAVE_HW_BRANCH_TRACER bool +config HAVE_SYSCALL_TRACEPOINTS + bool + help + See Documentation/trace/ftrace-implementation.txt + config TRACER_MAX_TRACE bool config RING_BUFFER bool +config FTRACE_NMI_ENTER + bool + depends on HAVE_FTRACE_NMI_ENTER + default y + +config EVENT_TRACING + select CONTEXT_SWITCH_TRACER + bool + +config CONTEXT_SWITCH_TRACER + bool + +config RING_BUFFER_ALLOW_SWAP + bool + help + Allow the use of ring_buffer_swap_cpu. + Adds a very slight overhead to tracing when enabled. + +# All tracer options should select GENERIC_TRACER. For those options that are +# enabled by all tracers (context switch and event tracer) they select TRACING. +# This allows those options to appear when no other tracer is selected. But the +# options do not appear when something else selects it. We need the two options +# GENERIC_TRACER and TRACING to avoid circular dependencies to accomplish the +# hidding of the automatic options. + config TRACING bool select DEBUG_FS @@ -44,16 +92,43 @@ config TRACING select STACKTRACE if STACKTRACE_SUPPORT select TRACEPOINTS select NOP_TRACER + select BINARY_PRINTF + select EVENT_TRACING -menu "Tracers" +config GENERIC_TRACER + bool + select TRACING + +# +# Minimum requirements an architecture has to meet for us to +# be able to offer generic tracing facilities: +# +config TRACING_SUPPORT + bool + # PPC32 has no irqflags tracing support, but it can use most of the + # tracers anyway, they were tested to build and work. Note that new + # exceptions to this list aren't welcomed, better implement the + # irqflags tracing for your architecture. + depends on TRACE_IRQFLAGS_SUPPORT || PPC32 + depends on STACKTRACE_SUPPORT + default y + +if TRACING_SUPPORT + +menuconfig FTRACE + bool "Tracers" + default y if DEBUG_KERNEL + help + Enable the kernel tracing infrastructure. + +if FTRACE config FUNCTION_TRACER bool "Kernel Function Tracer" depends on HAVE_FUNCTION_TRACER - depends on DEBUG_KERNEL select FRAME_POINTER select KALLSYMS - select TRACING + select GENERIC_TRACER select CONTEXT_SWITCH_TRACER help Enable the kernel to trace every kernel function. This is done @@ -68,25 +143,26 @@ config FUNCTION_GRAPH_TRACER bool "Kernel Function Graph Tracer" depends on HAVE_FUNCTION_GRAPH_TRACER depends on FUNCTION_TRACER + depends on !X86_32 || !CC_OPTIMIZE_FOR_SIZE default y help Enable the kernel to trace a function at both its return and its entry. - It's first purpose is to trace the duration of functions and - draw a call graph for each thread with some informations like - the return value. - This is done by setting the current return address on the current - task structure into a stack of calls. + Its first purpose is to trace the duration of functions and + draw a call graph for each thread with some information like + the return value. This is done by setting the current return + address on the current task structure into a stack of calls. + config IRQSOFF_TRACER bool "Interrupts-off Latency Tracer" default n depends on TRACE_IRQFLAGS_SUPPORT depends on GENERIC_TIME - depends on DEBUG_KERNEL select TRACE_IRQFLAGS - select TRACING + select GENERIC_TRACER select TRACER_MAX_TRACE + select RING_BUFFER_ALLOW_SWAP help This option measures the time spent in irqs-off critical sections, with microsecond accuracy. @@ -95,7 +171,7 @@ config IRQSOFF_TRACER disabled by default and can be runtime (re-)started via: - echo 0 > /debugfs/tracing/tracing_max_latency + echo 0 > /sys/kernel/debug/tracing/tracing_max_latency (Note that kernel size and overhead increases with this option enabled. This option and the preempt-off timing option can be @@ -106,9 +182,9 @@ config PREEMPT_TRACER default n depends on GENERIC_TIME depends on PREEMPT - depends on DEBUG_KERNEL - select TRACING + select GENERIC_TRACER select TRACER_MAX_TRACE + select RING_BUFFER_ALLOW_SWAP help This option measures the time spent in preemption off critical sections, with microsecond accuracy. @@ -117,7 +193,7 @@ config PREEMPT_TRACER disabled by default and can be runtime (re-)started via: - echo 0 > /debugfs/tracing/tracing_max_latency + echo 0 > /sys/kernel/debug/tracing/tracing_max_latency (Note that kernel size and overhead increases with this option enabled. This option and the irqs-off timing option can be @@ -126,79 +202,111 @@ config PREEMPT_TRACER config SYSPROF_TRACER bool "Sysprof Tracer" depends on X86 - select TRACING + select GENERIC_TRACER + select CONTEXT_SWITCH_TRACER help This tracer provides the trace needed by the 'Sysprof' userspace tool. config SCHED_TRACER bool "Scheduling Latency Tracer" - depends on DEBUG_KERNEL - select TRACING + select GENERIC_TRACER select CONTEXT_SWITCH_TRACER select TRACER_MAX_TRACE help This tracer tracks the latency of the highest priority task to be scheduled in, starting from the point it has woken up. -config CONTEXT_SWITCH_TRACER - bool "Trace process context switches" - depends on DEBUG_KERNEL +config ENABLE_DEFAULT_TRACERS + bool "Trace process context switches and events" + depends on !GENERIC_TRACER select TRACING - select MARKERS help - This tracer gets called from the context switch and records - all switching of tasks. + This tracer hooks to various trace points in the kernel + allowing the user to pick and choose which trace point they + want to trace. It also includes the sched_switch tracer plugin. + +config FTRACE_SYSCALLS + bool "Trace syscalls" + depends on HAVE_SYSCALL_TRACEPOINTS + select GENERIC_TRACER + select KALLSYMS + help + Basic tracer to catch the syscall entry and exit events. config BOOT_TRACER bool "Trace boot initcalls" - depends on DEBUG_KERNEL - select TRACING + select GENERIC_TRACER select CONTEXT_SWITCH_TRACER help This tracer helps developers to optimize boot times: it records the timings of the initcalls and traces key events and the identity of tasks that can cause boot delays, such as context-switches. - Its aim is to be parsed by the /scripts/bootgraph.pl tool to + Its aim is to be parsed by the scripts/bootgraph.pl tool to produce pretty graphics about boot inefficiencies, giving a visual representation of the delays during initcalls - but the raw /debug/tracing/trace text output is readable too. - ( Note that tracing self tests can't be enabled if this tracer is - selected, because the self-tests are an initcall as well and that - would invalidate the boot trace. ) + You must pass in initcall_debug and ftrace=initcall to the kernel + command line to enable this on bootup. config TRACE_BRANCH_PROFILING + bool + select GENERIC_TRACER + +choice + prompt "Branch Profiling" + default BRANCH_PROFILE_NONE + help + The branch profiling is a software profiler. It will add hooks + into the C conditionals to test which path a branch takes. + + The likely/unlikely profiler only looks at the conditions that + are annotated with a likely or unlikely macro. + + The "all branch" profiler will profile every if statement in the + kernel. This profiler will also enable the likely/unlikely + profiler as well. + + Either of the above profilers add a bit of overhead to the system. + If unsure choose "No branch profiling". + +config BRANCH_PROFILE_NONE + bool "No branch profiling" + help + No branch profiling. Branch profiling adds a bit of overhead. + Only enable it if you want to analyse the branching behavior. + Otherwise keep it disabled. + +config PROFILE_ANNOTATED_BRANCHES bool "Trace likely/unlikely profiler" - depends on DEBUG_KERNEL - select TRACING + select TRACE_BRANCH_PROFILING help This tracer profiles all the the likely and unlikely macros in the kernel. It will display the results in: - /debugfs/tracing/profile_annotated_branch + /sys/kernel/debug/tracing/profile_annotated_branch Note: this will add a significant overhead, only turn this on if you need to profile the system's use of these macros. - Say N if unsure. - config PROFILE_ALL_BRANCHES bool "Profile all if conditionals" - depends on TRACE_BRANCH_PROFILING + select TRACE_BRANCH_PROFILING help This tracer profiles all branch conditions. Every if () taken in the kernel is recorded whether it hit or miss. The results will be displayed in: - /debugfs/tracing/profile_branch + /sys/kernel/debug/tracing/profile_branch + + This option also enables the likely/unlikely profiler. This configuration, when enabled, will impose a great overhead on the system. This should only be enabled when the system is to be analyzed - - Say N if unsure. +endchoice config TRACING_BRANCHES bool @@ -224,25 +332,44 @@ config BRANCH_TRACER config POWER_TRACER bool "Trace power consumption behavior" - depends on DEBUG_KERNEL depends on X86 - select TRACING + select GENERIC_TRACER help This tracer helps developers to analyze and optimize the kernels power management decisions, specifically the C-state and P-state behavior. +config KSYM_TRACER + bool "Trace read and write access on kernel memory locations" + depends on HAVE_HW_BREAKPOINT + select TRACING + help + This tracer helps find read and write operations on any given kernel + symbol i.e. /proc/kallsyms. + +config PROFILE_KSYM_TRACER + bool "Profile all kernel memory accesses on 'watched' variables" + depends on KSYM_TRACER + help + This tracer profiles kernel accesses on variables watched through the + ksym tracer ftrace plugin. Depending upon the hardware, all read + and write operations on kernel variables can be monitored for + accesses. + + The results will be displayed in: + /debugfs/tracing/profile_ksym + + Say N if unsure. config STACK_TRACER bool "Trace max stack" depends on HAVE_FUNCTION_TRACER - depends on DEBUG_KERNEL select FUNCTION_TRACER select STACKTRACE select KALLSYMS help This special tracer records the maximum stack footprint of the - kernel and displays it in debugfs/tracing/stack_trace. + kernel and displays it in /sys/kernel/debug/tracing/stack_trace. This tracer works by hooking into every function call that the kernel executes, and keeping a maximum stack depth value and @@ -261,16 +388,88 @@ config STACK_TRACER config HW_BRANCH_TRACER depends on HAVE_HW_BRANCH_TRACER bool "Trace hw branches" - select TRACING + select GENERIC_TRACER help This tracer records all branches on the system in a circular buffer giving access to the last N branches for each cpu. +config KMEMTRACE + bool "Trace SLAB allocations" + select GENERIC_TRACER + help + kmemtrace provides tracing for slab allocator functions, such as + kmalloc, kfree, kmem_cache_alloc, kmem_cache_free etc.. Collected + data is then fed to the userspace application in order to analyse + allocation hotspots, internal fragmentation and so on, making it + possible to see how well an allocator performs, as well as debug + and profile kernel code. + + This requires an userspace application to use. See + Documentation/trace/kmemtrace.txt for more information. + + Saying Y will make the kernel somewhat larger and slower. However, + if you disable kmemtrace at run-time or boot-time, the performance + impact is minimal (depending on the arch the kernel is built for). + + If unsure, say N. + +config WORKQUEUE_TRACER + bool "Trace workqueues" + select GENERIC_TRACER + help + The workqueue tracer provides some statistical informations + about each cpu workqueue thread such as the number of the + works inserted and executed since their creation. It can help + to evaluate the amount of work each of them have to perform. + For example it can help a developer to decide whether he should + choose a per cpu workqueue instead of a singlethreaded one. + +config BLK_DEV_IO_TRACE + bool "Support for tracing block io actions" + depends on SYSFS + depends on BLOCK + select RELAY + select DEBUG_FS + select TRACEPOINTS + select GENERIC_TRACER + select STACKTRACE + help + Say Y here if you want to be able to trace the block layer actions + on a given queue. Tracing allows you to see any traffic happening + on a block device queue. For more information (and the userspace + support tools needed), fetch the blktrace tools from: + + git://git.kernel.dk/blktrace.git + + Tracing also is possible using the ftrace interface, e.g.: + + echo 1 > /sys/block/sda/sda1/trace/enable + echo blk > /sys/kernel/debug/tracing/current_tracer + cat /sys/kernel/debug/tracing/trace_pipe + + If unsure, say N. + +config KPROBE_EVENT + depends on KPROBES + depends on X86 + bool "Enable kprobes-based dynamic events" + select TRACING + default y + help + This allows the user to add tracing events (similar to tracepoints) on the fly + via the ftrace interface. See Documentation/trace/kprobetrace.txt + for more details. + + Those events can be inserted wherever kprobes can probe, and record + various register and memory values. + + This option is also required by perf-probe subcommand of perf tools. If + you want to use perf tools, this option is strongly recommended. + config DYNAMIC_FTRACE bool "enable/disable ftrace tracepoints dynamically" depends on FUNCTION_TRACER depends on HAVE_DYNAMIC_FTRACE - depends on DEBUG_KERNEL default y help This option will modify all the calls to ftrace dynamically @@ -286,6 +485,20 @@ config DYNAMIC_FTRACE were made. If so, it runs stop_machine (stops all CPUS) and modifies the code to jump over the call to ftrace. +config FUNCTION_PROFILER + bool "Kernel function profiler" + depends on FUNCTION_TRACER + default n + help + This option enables the kernel function profiler. A file is created + in debugfs called function_profile_enabled which defaults to zero. + When a 1 is echoed into this file profiling begins, and when a + zero is entered, profiling stops. A file in the trace_stats + directory called functions, that show the list of functions that + have been hit and their counters. + + If in doubt, say N + config FTRACE_MCOUNT_RECORD def_bool y depends on DYNAMIC_FTRACE @@ -296,7 +509,7 @@ config FTRACE_SELFTEST config FTRACE_STARTUP_TEST bool "Perform a startup test on ftrace" - depends on TRACING && DEBUG_KERNEL && !BOOT_TRACER + depends on GENERIC_TRACER select FTRACE_SELFTEST help This option performs a series of startup tests on ftrace. On bootup @@ -304,17 +517,29 @@ config FTRACE_STARTUP_TEST functioning properly. It will do tests on all the configured tracers of ftrace. +config EVENT_TRACE_TEST_SYSCALLS + bool "Run selftest on syscall events" + depends on FTRACE_STARTUP_TEST + help + This option will also enable testing every syscall event. + It only enables the event and disables it and runs various loads + with the event enabled. This adds a bit more time for kernel boot + up since it runs this on every system call defined. + + TBD - enable a way to actually call the syscalls as we test their + events + config MMIOTRACE bool "Memory mapped IO tracing" - depends on HAVE_MMIOTRACE_SUPPORT && DEBUG_KERNEL && PCI - select TRACING + depends on HAVE_MMIOTRACE_SUPPORT && PCI + select GENERIC_TRACER help Mmiotrace traces Memory Mapped I/O access and is meant for debugging and reverse engineering. It is called from the ioremap implementation and works via page faults. Tracing is disabled by default and can be enabled at run-time. - See Documentation/tracers/mmiotrace.txt. + See Documentation/trace/mmiotrace.txt. If you are not helping to develop drivers, say N. config MMIOTRACE_TEST @@ -327,4 +552,23 @@ config MMIOTRACE_TEST Say N, unless you absolutely know what you are doing. -endmenu +config RING_BUFFER_BENCHMARK + tristate "Ring buffer benchmark stress tester" + depends on RING_BUFFER + help + This option creates a test to stress the ring buffer and bench mark it. + It creates its own ring buffer such that it will not interfer with + any other users of the ring buffer (such as ftrace). It then creates + a producer and consumer that will run for 10 seconds and sleep for + 10 seconds. Each interval it will print out the number of events + it recorded and give a rough estimate of how long each iteration took. + + It does not disable interrupts or raise its priority, so it may be + affected by processes that are running. + + If unsure, say N + +endif # FTRACE + +endif # TRACING_SUPPORT + diff --git a/kernel/trace/Makefile b/kernel/trace/Makefile index 349d5a93653..cd9ecd89ec7 100644 --- a/kernel/trace/Makefile +++ b/kernel/trace/Makefile @@ -15,10 +15,20 @@ ifdef CONFIG_TRACING_BRANCHES KBUILD_CFLAGS += -DDISABLE_BRANCH_PROFILING endif +# +# Make the trace clocks available generally: it's infrastructure +# relied on by ptrace for example: +# +obj-y += trace_clock.o + obj-$(CONFIG_FUNCTION_TRACER) += libftrace.o obj-$(CONFIG_RING_BUFFER) += ring_buffer.o +obj-$(CONFIG_RING_BUFFER_BENCHMARK) += ring_buffer_benchmark.o obj-$(CONFIG_TRACING) += trace.o +obj-$(CONFIG_TRACING) += trace_output.o +obj-$(CONFIG_TRACING) += trace_stat.o +obj-$(CONFIG_TRACING) += trace_printk.o obj-$(CONFIG_CONTEXT_SWITCH_TRACER) += trace_sched_switch.o obj-$(CONFIG_SYSPROF_TRACER) += trace_sysprof.o obj-$(CONFIG_FUNCTION_TRACER) += trace_functions.o @@ -32,6 +42,19 @@ obj-$(CONFIG_BOOT_TRACER) += trace_boot.o obj-$(CONFIG_FUNCTION_GRAPH_TRACER) += trace_functions_graph.o obj-$(CONFIG_TRACE_BRANCH_PROFILING) += trace_branch.o obj-$(CONFIG_HW_BRANCH_TRACER) += trace_hw_branches.o -obj-$(CONFIG_POWER_TRACER) += trace_power.o +obj-$(CONFIG_KMEMTRACE) += kmemtrace.o +obj-$(CONFIG_WORKQUEUE_TRACER) += trace_workqueue.o +obj-$(CONFIG_BLK_DEV_IO_TRACE) += blktrace.o +ifeq ($(CONFIG_BLOCK),y) +obj-$(CONFIG_EVENT_TRACING) += blktrace.o +endif +obj-$(CONFIG_EVENT_TRACING) += trace_events.o +obj-$(CONFIG_EVENT_TRACING) += trace_export.o +obj-$(CONFIG_FTRACE_SYSCALLS) += trace_syscalls.o +obj-$(CONFIG_EVENT_PROFILE) += trace_event_profile.o +obj-$(CONFIG_EVENT_TRACING) += trace_events_filter.o +obj-$(CONFIG_KPROBE_EVENT) += trace_kprobe.o +obj-$(CONFIG_KSYM_TRACER) += trace_ksym.o +obj-$(CONFIG_EVENT_TRACING) += power-traces.o libftrace-y := ftrace.o diff --git a/kernel/trace/blktrace.c b/kernel/trace/blktrace.c new file mode 100644 index 00000000000..d9d6206e0b1 --- /dev/null +++ b/kernel/trace/blktrace.c @@ -0,0 +1,1767 @@ +/* + * Copyright (C) 2006 Jens Axboe <axboe@kernel.dk> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + * + */ +#include <linux/kernel.h> +#include <linux/blkdev.h> +#include <linux/blktrace_api.h> +#include <linux/percpu.h> +#include <linux/init.h> +#include <linux/mutex.h> +#include <linux/debugfs.h> +#include <linux/smp_lock.h> +#include <linux/time.h> +#include <linux/uaccess.h> + +#include <trace/events/block.h> + +#include "trace_output.h" + +#ifdef CONFIG_BLK_DEV_IO_TRACE + +static unsigned int blktrace_seq __read_mostly = 1; + +static struct trace_array *blk_tr; +static bool blk_tracer_enabled __read_mostly; + +/* Select an alternative, minimalistic output than the original one */ +#define TRACE_BLK_OPT_CLASSIC 0x1 + +static struct tracer_opt blk_tracer_opts[] = { + /* Default disable the minimalistic output */ + { TRACER_OPT(blk_classic, TRACE_BLK_OPT_CLASSIC) }, + { } +}; + +static struct tracer_flags blk_tracer_flags = { + .val = 0, + .opts = blk_tracer_opts, +}; + +/* Global reference count of probes */ +static atomic_t blk_probes_ref = ATOMIC_INIT(0); + +static void blk_register_tracepoints(void); +static void blk_unregister_tracepoints(void); + +/* + * Send out a notify message. + */ +static void trace_note(struct blk_trace *bt, pid_t pid, int action, + const void *data, size_t len) +{ + struct blk_io_trace *t; + struct ring_buffer_event *event = NULL; + struct ring_buffer *buffer = NULL; + int pc = 0; + int cpu = smp_processor_id(); + bool blk_tracer = blk_tracer_enabled; + + if (blk_tracer) { + buffer = blk_tr->buffer; + pc = preempt_count(); + event = trace_buffer_lock_reserve(buffer, TRACE_BLK, + sizeof(*t) + len, + 0, pc); + if (!event) + return; + t = ring_buffer_event_data(event); + goto record_it; + } + + if (!bt->rchan) + return; + + t = relay_reserve(bt->rchan, sizeof(*t) + len); + if (t) { + t->magic = BLK_IO_TRACE_MAGIC | BLK_IO_TRACE_VERSION; + t->time = ktime_to_ns(ktime_get()); +record_it: + t->device = bt->dev; + t->action = action; + t->pid = pid; + t->cpu = cpu; + t->pdu_len = len; + memcpy((void *) t + sizeof(*t), data, len); + + if (blk_tracer) + trace_buffer_unlock_commit(buffer, event, 0, pc); + } +} + +/* + * Send out a notify for this process, if we haven't done so since a trace + * started + */ +static void trace_note_tsk(struct blk_trace *bt, struct task_struct *tsk) +{ + tsk->btrace_seq = blktrace_seq; + trace_note(bt, tsk->pid, BLK_TN_PROCESS, tsk->comm, sizeof(tsk->comm)); +} + +static void trace_note_time(struct blk_trace *bt) +{ + struct timespec now; + unsigned long flags; + u32 words[2]; + + getnstimeofday(&now); + words[0] = now.tv_sec; + words[1] = now.tv_nsec; + + local_irq_save(flags); + trace_note(bt, 0, BLK_TN_TIMESTAMP, words, sizeof(words)); + local_irq_restore(flags); +} + +void __trace_note_message(struct blk_trace *bt, const char *fmt, ...) +{ + int n; + va_list args; + unsigned long flags; + char *buf; + + if (unlikely(bt->trace_state != Blktrace_running && + !blk_tracer_enabled)) + return; + + local_irq_save(flags); + buf = per_cpu_ptr(bt->msg_data, smp_processor_id()); + va_start(args, fmt); + n = vscnprintf(buf, BLK_TN_MAX_MSG, fmt, args); + va_end(args); + + trace_note(bt, 0, BLK_TN_MESSAGE, buf, n); + local_irq_restore(flags); +} +EXPORT_SYMBOL_GPL(__trace_note_message); + +static int act_log_check(struct blk_trace *bt, u32 what, sector_t sector, + pid_t pid) +{ + if (((bt->act_mask << BLK_TC_SHIFT) & what) == 0) + return 1; + if (sector && (sector < bt->start_lba || sector > bt->end_lba)) + return 1; + if (bt->pid && pid != bt->pid) + return 1; + + return 0; +} + +/* + * Data direction bit lookup + */ +static const u32 ddir_act[2] = { BLK_TC_ACT(BLK_TC_READ), + BLK_TC_ACT(BLK_TC_WRITE) }; + +/* The ilog2() calls fall out because they're constant */ +#define MASK_TC_BIT(rw, __name) ((rw & (1 << BIO_RW_ ## __name)) << \ + (ilog2(BLK_TC_ ## __name) + BLK_TC_SHIFT - BIO_RW_ ## __name)) + +/* + * The worker for the various blk_add_trace*() types. Fills out a + * blk_io_trace structure and places it in a per-cpu subbuffer. + */ +static void __blk_add_trace(struct blk_trace *bt, sector_t sector, int bytes, + int rw, u32 what, int error, int pdu_len, void *pdu_data) +{ + struct task_struct *tsk = current; + struct ring_buffer_event *event = NULL; + struct ring_buffer *buffer = NULL; + struct blk_io_trace *t; + unsigned long flags = 0; + unsigned long *sequence; + pid_t pid; + int cpu, pc = 0; + bool blk_tracer = blk_tracer_enabled; + + if (unlikely(bt->trace_state != Blktrace_running && !blk_tracer)) + return; + + what |= ddir_act[rw & WRITE]; + what |= MASK_TC_BIT(rw, BARRIER); + what |= MASK_TC_BIT(rw, SYNCIO); + what |= MASK_TC_BIT(rw, AHEAD); + what |= MASK_TC_BIT(rw, META); + what |= MASK_TC_BIT(rw, DISCARD); + + pid = tsk->pid; + if (act_log_check(bt, what, sector, pid)) + return; + cpu = raw_smp_processor_id(); + + if (blk_tracer) { + tracing_record_cmdline(current); + + buffer = blk_tr->buffer; + pc = preempt_count(); + event = trace_buffer_lock_reserve(buffer, TRACE_BLK, + sizeof(*t) + pdu_len, + 0, pc); + if (!event) + return; + t = ring_buffer_event_data(event); + goto record_it; + } + + /* + * A word about the locking here - we disable interrupts to reserve + * some space in the relay per-cpu buffer, to prevent an irq + * from coming in and stepping on our toes. + */ + local_irq_save(flags); + + if (unlikely(tsk->btrace_seq != blktrace_seq)) + trace_note_tsk(bt, tsk); + + t = relay_reserve(bt->rchan, sizeof(*t) + pdu_len); + if (t) { + sequence = per_cpu_ptr(bt->sequence, cpu); + + t->magic = BLK_IO_TRACE_MAGIC | BLK_IO_TRACE_VERSION; + t->sequence = ++(*sequence); + t->time = ktime_to_ns(ktime_get()); +record_it: + /* + * These two are not needed in ftrace as they are in the + * generic trace_entry, filled by tracing_generic_entry_update, + * but for the trace_event->bin() synthesizer benefit we do it + * here too. + */ + t->cpu = cpu; + t->pid = pid; + + t->sector = sector; + t->bytes = bytes; + t->action = what; + t->device = bt->dev; + t->error = error; + t->pdu_len = pdu_len; + + if (pdu_len) + memcpy((void *) t + sizeof(*t), pdu_data, pdu_len); + + if (blk_tracer) { + trace_buffer_unlock_commit(buffer, event, 0, pc); + return; + } + } + + local_irq_restore(flags); +} + +static struct dentry *blk_tree_root; +static DEFINE_MUTEX(blk_tree_mutex); + +static void blk_trace_free(struct blk_trace *bt) +{ + debugfs_remove(bt->msg_file); + debugfs_remove(bt->dropped_file); + relay_close(bt->rchan); + debugfs_remove(bt->dir); + free_percpu(bt->sequence); + free_percpu(bt->msg_data); + kfree(bt); +} + +static void blk_trace_cleanup(struct blk_trace *bt) +{ + blk_trace_free(bt); + if (atomic_dec_and_test(&blk_probes_ref)) + blk_unregister_tracepoints(); +} + +int blk_trace_remove(struct request_queue *q) +{ + struct blk_trace *bt; + + bt = xchg(&q->blk_trace, NULL); + if (!bt) + return -EINVAL; + + if (bt->trace_state != Blktrace_running) + blk_trace_cleanup(bt); + + return 0; +} +EXPORT_SYMBOL_GPL(blk_trace_remove); + +static int blk_dropped_open(struct inode *inode, struct file *filp) +{ + filp->private_data = inode->i_private; + + return 0; +} + +static ssize_t blk_dropped_read(struct file *filp, char __user *buffer, + size_t count, loff_t *ppos) +{ + struct blk_trace *bt = filp->private_data; + char buf[16]; + + snprintf(buf, sizeof(buf), "%u\n", atomic_read(&bt->dropped)); + + return simple_read_from_buffer(buffer, count, ppos, buf, strlen(buf)); +} + +static const struct file_operations blk_dropped_fops = { + .owner = THIS_MODULE, + .open = blk_dropped_open, + .read = blk_dropped_read, +}; + +static int blk_msg_open(struct inode *inode, struct file *filp) +{ + filp->private_data = inode->i_private; + + return 0; +} + +static ssize_t blk_msg_write(struct file *filp, const char __user *buffer, + size_t count, loff_t *ppos) +{ + char *msg; + struct blk_trace *bt; + + if (count >= BLK_TN_MAX_MSG) + return -EINVAL; + + msg = kmalloc(count + 1, GFP_KERNEL); + if (msg == NULL) + return -ENOMEM; + + if (copy_from_user(msg, buffer, count)) { + kfree(msg); + return -EFAULT; + } + + msg[count] = '\0'; + bt = filp->private_data; + __trace_note_message(bt, "%s", msg); + kfree(msg); + + return count; +} + +static const struct file_operations blk_msg_fops = { + .owner = THIS_MODULE, + .open = blk_msg_open, + .write = blk_msg_write, +}; + +/* + * Keep track of how many times we encountered a full subbuffer, to aid + * the user space app in telling how many lost events there were. + */ +static int blk_subbuf_start_callback(struct rchan_buf *buf, void *subbuf, + void *prev_subbuf, size_t prev_padding) +{ + struct blk_trace *bt; + + if (!relay_buf_full(buf)) + return 1; + + bt = buf->chan->private_data; + atomic_inc(&bt->dropped); + return 0; +} + +static int blk_remove_buf_file_callback(struct dentry *dentry) +{ + debugfs_remove(dentry); + + return 0; +} + +static struct dentry *blk_create_buf_file_callback(const char *filename, + struct dentry *parent, + int mode, + struct rchan_buf *buf, + int *is_global) +{ + return debugfs_create_file(filename, mode, parent, buf, + &relay_file_operations); +} + +static struct rchan_callbacks blk_relay_callbacks = { + .subbuf_start = blk_subbuf_start_callback, + .create_buf_file = blk_create_buf_file_callback, + .remove_buf_file = blk_remove_buf_file_callback, +}; + +static void blk_trace_setup_lba(struct blk_trace *bt, + struct block_device *bdev) +{ + struct hd_struct *part = NULL; + + if (bdev) + part = bdev->bd_part; + + if (part) { + bt->start_lba = part->start_sect; + bt->end_lba = part->start_sect + part->nr_sects; + } else { + bt->start_lba = 0; + bt->end_lba = -1ULL; + } +} + +/* + * Setup everything required to start tracing + */ +int do_blk_trace_setup(struct request_queue *q, char *name, dev_t dev, + struct block_device *bdev, + struct blk_user_trace_setup *buts) +{ + struct blk_trace *old_bt, *bt = NULL; + struct dentry *dir = NULL; + int ret, i; + + if (!buts->buf_size || !buts->buf_nr) + return -EINVAL; + + strncpy(buts->name, name, BLKTRACE_BDEV_SIZE); + buts->name[BLKTRACE_BDEV_SIZE - 1] = '\0'; + + /* + * some device names have larger paths - convert the slashes + * to underscores for this to work as expected + */ + for (i = 0; i < strlen(buts->name); i++) + if (buts->name[i] == '/') + buts->name[i] = '_'; + + bt = kzalloc(sizeof(*bt), GFP_KERNEL); + if (!bt) + return -ENOMEM; + + ret = -ENOMEM; + bt->sequence = alloc_percpu(unsigned long); + if (!bt->sequence) + goto err; + + bt->msg_data = __alloc_percpu(BLK_TN_MAX_MSG, __alignof__(char)); + if (!bt->msg_data) + goto err; + + ret = -ENOENT; + + mutex_lock(&blk_tree_mutex); + if (!blk_tree_root) { + blk_tree_root = debugfs_create_dir("block", NULL); + if (!blk_tree_root) { + mutex_unlock(&blk_tree_mutex); + goto err; + } + } + mutex_unlock(&blk_tree_mutex); + + dir = debugfs_create_dir(buts->name, blk_tree_root); + + if (!dir) + goto err; + + bt->dir = dir; + bt->dev = dev; + atomic_set(&bt->dropped, 0); + + ret = -EIO; + bt->dropped_file = debugfs_create_file("dropped", 0444, dir, bt, + &blk_dropped_fops); + if (!bt->dropped_file) + goto err; + + bt->msg_file = debugfs_create_file("msg", 0222, dir, bt, &blk_msg_fops); + if (!bt->msg_file) + goto err; + + bt->rchan = relay_open("trace", dir, buts->buf_size, + buts->buf_nr, &blk_relay_callbacks, bt); + if (!bt->rchan) + goto err; + + bt->act_mask = buts->act_mask; + if (!bt->act_mask) + bt->act_mask = (u16) -1; + + blk_trace_setup_lba(bt, bdev); + + /* overwrite with user settings */ + if (buts->start_lba) + bt->start_lba = buts->start_lba; + if (buts->end_lba) + bt->end_lba = buts->end_lba; + + bt->pid = buts->pid; + bt->trace_state = Blktrace_setup; + + ret = -EBUSY; + old_bt = xchg(&q->blk_trace, bt); + if (old_bt) { + (void) xchg(&q->blk_trace, old_bt); + goto err; + } + + if (atomic_inc_return(&blk_probes_ref) == 1) + blk_register_tracepoints(); + + return 0; +err: + blk_trace_free(bt); + return ret; +} + +int blk_trace_setup(struct request_queue *q, char *name, dev_t dev, + struct block_device *bdev, + char __user *arg) +{ + struct blk_user_trace_setup buts; + int ret; + + ret = copy_from_user(&buts, arg, sizeof(buts)); + if (ret) + return -EFAULT; + + ret = do_blk_trace_setup(q, name, dev, bdev, &buts); + if (ret) + return ret; + + if (copy_to_user(arg, &buts, sizeof(buts))) + return -EFAULT; + + return 0; +} +EXPORT_SYMBOL_GPL(blk_trace_setup); + +int blk_trace_startstop(struct request_queue *q, int start) +{ + int ret; + struct blk_trace *bt = q->blk_trace; + + if (bt == NULL) + return -EINVAL; + + /* + * For starting a trace, we can transition from a setup or stopped + * trace. For stopping a trace, the state must be running + */ + ret = -EINVAL; + if (start) { + if (bt->trace_state == Blktrace_setup || + bt->trace_state == Blktrace_stopped) { + blktrace_seq++; + smp_mb(); + bt->trace_state = Blktrace_running; + + trace_note_time(bt); + ret = 0; + } + } else { + if (bt->trace_state == Blktrace_running) { + bt->trace_state = Blktrace_stopped; + relay_flush(bt->rchan); + ret = 0; + } + } + + return ret; +} +EXPORT_SYMBOL_GPL(blk_trace_startstop); + +/** + * blk_trace_ioctl: - handle the ioctls associated with tracing + * @bdev: the block device + * @cmd: the ioctl cmd + * @arg: the argument data, if any + * + **/ +int blk_trace_ioctl(struct block_device *bdev, unsigned cmd, char __user *arg) +{ + struct request_queue *q; + int ret, start = 0; + char b[BDEVNAME_SIZE]; + + q = bdev_get_queue(bdev); + if (!q) + return -ENXIO; + + mutex_lock(&bdev->bd_mutex); + + switch (cmd) { + case BLKTRACESETUP: + bdevname(bdev, b); + ret = blk_trace_setup(q, b, bdev->bd_dev, bdev, arg); + break; + case BLKTRACESTART: + start = 1; + case BLKTRACESTOP: + ret = blk_trace_startstop(q, start); + break; + case BLKTRACETEARDOWN: + ret = blk_trace_remove(q); + break; + default: + ret = -ENOTTY; + break; + } + + mutex_unlock(&bdev->bd_mutex); + return ret; +} + +/** + * blk_trace_shutdown: - stop and cleanup trace structures + * @q: the request queue associated with the device + * + **/ +void blk_trace_shutdown(struct request_queue *q) +{ + if (q->blk_trace) { + blk_trace_startstop(q, 0); + blk_trace_remove(q); + } +} + +/* + * blktrace probes + */ + +/** + * blk_add_trace_rq - Add a trace for a request oriented action + * @q: queue the io is for + * @rq: the source request + * @what: the action + * + * Description: + * Records an action against a request. Will log the bio offset + size. + * + **/ +static void blk_add_trace_rq(struct request_queue *q, struct request *rq, + u32 what) +{ + struct blk_trace *bt = q->blk_trace; + int rw = rq->cmd_flags & 0x03; + + if (likely(!bt)) + return; + + if (blk_discard_rq(rq)) + rw |= (1 << BIO_RW_DISCARD); + + if (blk_pc_request(rq)) { + what |= BLK_TC_ACT(BLK_TC_PC); + __blk_add_trace(bt, 0, blk_rq_bytes(rq), rw, + what, rq->errors, rq->cmd_len, rq->cmd); + } else { + what |= BLK_TC_ACT(BLK_TC_FS); + __blk_add_trace(bt, blk_rq_pos(rq), blk_rq_bytes(rq), rw, + what, rq->errors, 0, NULL); + } +} + +static void blk_add_trace_rq_abort(struct request_queue *q, struct request *rq) +{ + blk_add_trace_rq(q, rq, BLK_TA_ABORT); +} + +static void blk_add_trace_rq_insert(struct request_queue *q, struct request *rq) +{ + blk_add_trace_rq(q, rq, BLK_TA_INSERT); +} + +static void blk_add_trace_rq_issue(struct request_queue *q, struct request *rq) +{ + blk_add_trace_rq(q, rq, BLK_TA_ISSUE); +} + +static void blk_add_trace_rq_requeue(struct request_queue *q, + struct request *rq) +{ + blk_add_trace_rq(q, rq, BLK_TA_REQUEUE); +} + +static void blk_add_trace_rq_complete(struct request_queue *q, + struct request *rq) +{ + blk_add_trace_rq(q, rq, BLK_TA_COMPLETE); +} + +/** + * blk_add_trace_bio - Add a trace for a bio oriented action + * @q: queue the io is for + * @bio: the source bio + * @what: the action + * + * Description: + * Records an action against a bio. Will log the bio offset + size. + * + **/ +static void blk_add_trace_bio(struct request_queue *q, struct bio *bio, + u32 what) +{ + struct blk_trace *bt = q->blk_trace; + + if (likely(!bt)) + return; + + __blk_add_trace(bt, bio->bi_sector, bio->bi_size, bio->bi_rw, what, + !bio_flagged(bio, BIO_UPTODATE), 0, NULL); +} + +static void blk_add_trace_bio_bounce(struct request_queue *q, struct bio *bio) +{ + blk_add_trace_bio(q, bio, BLK_TA_BOUNCE); +} + +static void blk_add_trace_bio_complete(struct request_queue *q, struct bio *bio) +{ + blk_add_trace_bio(q, bio, BLK_TA_COMPLETE); +} + +static void blk_add_trace_bio_backmerge(struct request_queue *q, + struct bio *bio) +{ + blk_add_trace_bio(q, bio, BLK_TA_BACKMERGE); +} + +static void blk_add_trace_bio_frontmerge(struct request_queue *q, + struct bio *bio) +{ + blk_add_trace_bio(q, bio, BLK_TA_FRONTMERGE); +} + +static void blk_add_trace_bio_queue(struct request_queue *q, struct bio *bio) +{ + blk_add_trace_bio(q, bio, BLK_TA_QUEUE); +} + +static void blk_add_trace_getrq(struct request_queue *q, + struct bio *bio, int rw) +{ + if (bio) + blk_add_trace_bio(q, bio, BLK_TA_GETRQ); + else { + struct blk_trace *bt = q->blk_trace; + + if (bt) + __blk_add_trace(bt, 0, 0, rw, BLK_TA_GETRQ, 0, 0, NULL); + } +} + + +static void blk_add_trace_sleeprq(struct request_queue *q, + struct bio *bio, int rw) +{ + if (bio) + blk_add_trace_bio(q, bio, BLK_TA_SLEEPRQ); + else { + struct blk_trace *bt = q->blk_trace; + + if (bt) + __blk_add_trace(bt, 0, 0, rw, BLK_TA_SLEEPRQ, + 0, 0, NULL); + } +} + +static void blk_add_trace_plug(struct request_queue *q) +{ + struct blk_trace *bt = q->blk_trace; + + if (bt) + __blk_add_trace(bt, 0, 0, 0, BLK_TA_PLUG, 0, 0, NULL); +} + +static void blk_add_trace_unplug_io(struct request_queue *q) +{ + struct blk_trace *bt = q->blk_trace; + + if (bt) { + unsigned int pdu = q->rq.count[READ] + q->rq.count[WRITE]; + __be64 rpdu = cpu_to_be64(pdu); + + __blk_add_trace(bt, 0, 0, 0, BLK_TA_UNPLUG_IO, 0, + sizeof(rpdu), &rpdu); + } +} + +static void blk_add_trace_unplug_timer(struct request_queue *q) +{ + struct blk_trace *bt = q->blk_trace; + + if (bt) { + unsigned int pdu = q->rq.count[READ] + q->rq.count[WRITE]; + __be64 rpdu = cpu_to_be64(pdu); + + __blk_add_trace(bt, 0, 0, 0, BLK_TA_UNPLUG_TIMER, 0, + sizeof(rpdu), &rpdu); + } +} + +static void blk_add_trace_split(struct request_queue *q, struct bio *bio, + unsigned int pdu) +{ + struct blk_trace *bt = q->blk_trace; + + if (bt) { + __be64 rpdu = cpu_to_be64(pdu); + + __blk_add_trace(bt, bio->bi_sector, bio->bi_size, bio->bi_rw, + BLK_TA_SPLIT, !bio_flagged(bio, BIO_UPTODATE), + sizeof(rpdu), &rpdu); + } +} + +/** + * blk_add_trace_remap - Add a trace for a remap operation + * @q: queue the io is for + * @bio: the source bio + * @dev: target device + * @from: source sector + * + * Description: + * Device mapper or raid target sometimes need to split a bio because + * it spans a stripe (or similar). Add a trace for that action. + * + **/ +static void blk_add_trace_remap(struct request_queue *q, struct bio *bio, + dev_t dev, sector_t from) +{ + struct blk_trace *bt = q->blk_trace; + struct blk_io_trace_remap r; + + if (likely(!bt)) + return; + + r.device_from = cpu_to_be32(dev); + r.device_to = cpu_to_be32(bio->bi_bdev->bd_dev); + r.sector_from = cpu_to_be64(from); + + __blk_add_trace(bt, bio->bi_sector, bio->bi_size, bio->bi_rw, + BLK_TA_REMAP, !bio_flagged(bio, BIO_UPTODATE), + sizeof(r), &r); +} + +/** + * blk_add_trace_rq_remap - Add a trace for a request-remap operation + * @q: queue the io is for + * @rq: the source request + * @dev: target device + * @from: source sector + * + * Description: + * Device mapper remaps request to other devices. + * Add a trace for that action. + * + **/ +static void blk_add_trace_rq_remap(struct request_queue *q, + struct request *rq, dev_t dev, + sector_t from) +{ + struct blk_trace *bt = q->blk_trace; + struct blk_io_trace_remap r; + + if (likely(!bt)) + return; + + r.device_from = cpu_to_be32(dev); + r.device_to = cpu_to_be32(disk_devt(rq->rq_disk)); + r.sector_from = cpu_to_be64(from); + + __blk_add_trace(bt, blk_rq_pos(rq), blk_rq_bytes(rq), + rq_data_dir(rq), BLK_TA_REMAP, !!rq->errors, + sizeof(r), &r); +} + +/** + * blk_add_driver_data - Add binary message with driver-specific data + * @q: queue the io is for + * @rq: io request + * @data: driver-specific data + * @len: length of driver-specific data + * + * Description: + * Some drivers might want to write driver-specific data per request. + * + **/ +void blk_add_driver_data(struct request_queue *q, + struct request *rq, + void *data, size_t len) +{ + struct blk_trace *bt = q->blk_trace; + + if (likely(!bt)) + return; + + if (blk_pc_request(rq)) + __blk_add_trace(bt, 0, blk_rq_bytes(rq), 0, + BLK_TA_DRV_DATA, rq->errors, len, data); + else + __blk_add_trace(bt, blk_rq_pos(rq), blk_rq_bytes(rq), 0, + BLK_TA_DRV_DATA, rq->errors, len, data); +} +EXPORT_SYMBOL_GPL(blk_add_driver_data); + +static void blk_register_tracepoints(void) +{ + int ret; + + ret = register_trace_block_rq_abort(blk_add_trace_rq_abort); + WARN_ON(ret); + ret = register_trace_block_rq_insert(blk_add_trace_rq_insert); + WARN_ON(ret); + ret = register_trace_block_rq_issue(blk_add_trace_rq_issue); + WARN_ON(ret); + ret = register_trace_block_rq_requeue(blk_add_trace_rq_requeue); + WARN_ON(ret); + ret = register_trace_block_rq_complete(blk_add_trace_rq_complete); + WARN_ON(ret); + ret = register_trace_block_bio_bounce(blk_add_trace_bio_bounce); + WARN_ON(ret); + ret = register_trace_block_bio_complete(blk_add_trace_bio_complete); + WARN_ON(ret); + ret = register_trace_block_bio_backmerge(blk_add_trace_bio_backmerge); + WARN_ON(ret); + ret = register_trace_block_bio_frontmerge(blk_add_trace_bio_frontmerge); + WARN_ON(ret); + ret = register_trace_block_bio_queue(blk_add_trace_bio_queue); + WARN_ON(ret); + ret = register_trace_block_getrq(blk_add_trace_getrq); + WARN_ON(ret); + ret = register_trace_block_sleeprq(blk_add_trace_sleeprq); + WARN_ON(ret); + ret = register_trace_block_plug(blk_add_trace_plug); + WARN_ON(ret); + ret = register_trace_block_unplug_timer(blk_add_trace_unplug_timer); + WARN_ON(ret); + ret = register_trace_block_unplug_io(blk_add_trace_unplug_io); + WARN_ON(ret); + ret = register_trace_block_split(blk_add_trace_split); + WARN_ON(ret); + ret = register_trace_block_remap(blk_add_trace_remap); + WARN_ON(ret); + ret = register_trace_block_rq_remap(blk_add_trace_rq_remap); + WARN_ON(ret); +} + +static void blk_unregister_tracepoints(void) +{ + unregister_trace_block_rq_remap(blk_add_trace_rq_remap); + unregister_trace_block_remap(blk_add_trace_remap); + unregister_trace_block_split(blk_add_trace_split); + unregister_trace_block_unplug_io(blk_add_trace_unplug_io); + unregister_trace_block_unplug_timer(blk_add_trace_unplug_timer); + unregister_trace_block_plug(blk_add_trace_plug); + unregister_trace_block_sleeprq(blk_add_trace_sleeprq); + unregister_trace_block_getrq(blk_add_trace_getrq); + unregister_trace_block_bio_queue(blk_add_trace_bio_queue); + unregister_trace_block_bio_frontmerge(blk_add_trace_bio_frontmerge); + unregister_trace_block_bio_backmerge(blk_add_trace_bio_backmerge); + unregister_trace_block_bio_complete(blk_add_trace_bio_complete); + unregister_trace_block_bio_bounce(blk_add_trace_bio_bounce); + unregister_trace_block_rq_complete(blk_add_trace_rq_complete); + unregister_trace_block_rq_requeue(blk_add_trace_rq_requeue); + unregister_trace_block_rq_issue(blk_add_trace_rq_issue); + unregister_trace_block_rq_insert(blk_add_trace_rq_insert); + unregister_trace_block_rq_abort(blk_add_trace_rq_abort); + + tracepoint_synchronize_unregister(); +} + +/* + * struct blk_io_tracer formatting routines + */ + +static void fill_rwbs(char *rwbs, const struct blk_io_trace *t) +{ + int i = 0; + int tc = t->action >> BLK_TC_SHIFT; + + if (t->action == BLK_TN_MESSAGE) { + rwbs[i++] = 'N'; + goto out; + } + + if (tc & BLK_TC_DISCARD) + rwbs[i++] = 'D'; + else if (tc & BLK_TC_WRITE) + rwbs[i++] = 'W'; + else if (t->bytes) + rwbs[i++] = 'R'; + else + rwbs[i++] = 'N'; + + if (tc & BLK_TC_AHEAD) + rwbs[i++] = 'A'; + if (tc & BLK_TC_BARRIER) + rwbs[i++] = 'B'; + if (tc & BLK_TC_SYNC) + rwbs[i++] = 'S'; + if (tc & BLK_TC_META) + rwbs[i++] = 'M'; +out: + rwbs[i] = '\0'; +} + +static inline +const struct blk_io_trace *te_blk_io_trace(const struct trace_entry *ent) +{ + return (const struct blk_io_trace *)ent; +} + +static inline const void *pdu_start(const struct trace_entry *ent) +{ + return te_blk_io_trace(ent) + 1; +} + +static inline u32 t_action(const struct trace_entry *ent) +{ + return te_blk_io_trace(ent)->action; +} + +static inline u32 t_bytes(const struct trace_entry *ent) +{ + return te_blk_io_trace(ent)->bytes; +} + +static inline u32 t_sec(const struct trace_entry *ent) +{ + return te_blk_io_trace(ent)->bytes >> 9; +} + +static inline unsigned long long t_sector(const struct trace_entry *ent) +{ + return te_blk_io_trace(ent)->sector; +} + +static inline __u16 t_error(const struct trace_entry *ent) +{ + return te_blk_io_trace(ent)->error; +} + +static __u64 get_pdu_int(const struct trace_entry *ent) +{ + const __u64 *val = pdu_start(ent); + return be64_to_cpu(*val); +} + +static void get_pdu_remap(const struct trace_entry *ent, + struct blk_io_trace_remap *r) +{ + const struct blk_io_trace_remap *__r = pdu_start(ent); + __u64 sector_from = __r->sector_from; + + r->device_from = be32_to_cpu(__r->device_from); + r->device_to = be32_to_cpu(__r->device_to); + r->sector_from = be64_to_cpu(sector_from); +} + +typedef int (blk_log_action_t) (struct trace_iterator *iter, const char *act); + +static int blk_log_action_classic(struct trace_iterator *iter, const char *act) +{ + char rwbs[6]; + unsigned long long ts = iter->ts; + unsigned long nsec_rem = do_div(ts, NSEC_PER_SEC); + unsigned secs = (unsigned long)ts; + const struct blk_io_trace *t = te_blk_io_trace(iter->ent); + + fill_rwbs(rwbs, t); + + return trace_seq_printf(&iter->seq, + "%3d,%-3d %2d %5d.%09lu %5u %2s %3s ", + MAJOR(t->device), MINOR(t->device), iter->cpu, + secs, nsec_rem, iter->ent->pid, act, rwbs); +} + +static int blk_log_action(struct trace_iterator *iter, const char *act) +{ + char rwbs[6]; + const struct blk_io_trace *t = te_blk_io_trace(iter->ent); + + fill_rwbs(rwbs, t); + return trace_seq_printf(&iter->seq, "%3d,%-3d %2s %3s ", + MAJOR(t->device), MINOR(t->device), act, rwbs); +} + +static int blk_log_dump_pdu(struct trace_seq *s, const struct trace_entry *ent) +{ + const unsigned char *pdu_buf; + int pdu_len; + int i, end, ret; + + pdu_buf = pdu_start(ent); + pdu_len = te_blk_io_trace(ent)->pdu_len; + + if (!pdu_len) + return 1; + + /* find the last zero that needs to be printed */ + for (end = pdu_len - 1; end >= 0; end--) + if (pdu_buf[end]) + break; + end++; + + if (!trace_seq_putc(s, '(')) + return 0; + + for (i = 0; i < pdu_len; i++) { + + ret = trace_seq_printf(s, "%s%02x", + i == 0 ? "" : " ", pdu_buf[i]); + if (!ret) + return ret; + + /* + * stop when the rest is just zeroes and indicate so + * with a ".." appended + */ + if (i == end && end != pdu_len - 1) + return trace_seq_puts(s, " ..) "); + } + + return trace_seq_puts(s, ") "); +} + +static int blk_log_generic(struct trace_seq *s, const struct trace_entry *ent) +{ + char cmd[TASK_COMM_LEN]; + + trace_find_cmdline(ent->pid, cmd); + + if (t_action(ent) & BLK_TC_ACT(BLK_TC_PC)) { + int ret; + + ret = trace_seq_printf(s, "%u ", t_bytes(ent)); + if (!ret) + return 0; + ret = blk_log_dump_pdu(s, ent); + if (!ret) + return 0; + return trace_seq_printf(s, "[%s]\n", cmd); + } else { + if (t_sec(ent)) + return trace_seq_printf(s, "%llu + %u [%s]\n", + t_sector(ent), t_sec(ent), cmd); + return trace_seq_printf(s, "[%s]\n", cmd); + } +} + +static int blk_log_with_error(struct trace_seq *s, + const struct trace_entry *ent) +{ + if (t_action(ent) & BLK_TC_ACT(BLK_TC_PC)) { + int ret; + + ret = blk_log_dump_pdu(s, ent); + if (ret) + return trace_seq_printf(s, "[%d]\n", t_error(ent)); + return 0; + } else { + if (t_sec(ent)) + return trace_seq_printf(s, "%llu + %u [%d]\n", + t_sector(ent), + t_sec(ent), t_error(ent)); + return trace_seq_printf(s, "%llu [%d]\n", + t_sector(ent), t_error(ent)); + } +} + +static int blk_log_remap(struct trace_seq *s, const struct trace_entry *ent) +{ + struct blk_io_trace_remap r = { .device_from = 0, }; + + get_pdu_remap(ent, &r); + return trace_seq_printf(s, "%llu + %u <- (%d,%d) %llu\n", + t_sector(ent), t_sec(ent), + MAJOR(r.device_from), MINOR(r.device_from), + (unsigned long long)r.sector_from); +} + +static int blk_log_plug(struct trace_seq *s, const struct trace_entry *ent) +{ + char cmd[TASK_COMM_LEN]; + + trace_find_cmdline(ent->pid, cmd); + + return trace_seq_printf(s, "[%s]\n", cmd); +} + +static int blk_log_unplug(struct trace_seq *s, const struct trace_entry *ent) +{ + char cmd[TASK_COMM_LEN]; + + trace_find_cmdline(ent->pid, cmd); + + return trace_seq_printf(s, "[%s] %llu\n", cmd, get_pdu_int(ent)); +} + +static int blk_log_split(struct trace_seq *s, const struct trace_entry *ent) +{ + char cmd[TASK_COMM_LEN]; + + trace_find_cmdline(ent->pid, cmd); + + return trace_seq_printf(s, "%llu / %llu [%s]\n", t_sector(ent), + get_pdu_int(ent), cmd); +} + +static int blk_log_msg(struct trace_seq *s, const struct trace_entry *ent) +{ + int ret; + const struct blk_io_trace *t = te_blk_io_trace(ent); + + ret = trace_seq_putmem(s, t + 1, t->pdu_len); + if (ret) + return trace_seq_putc(s, '\n'); + return ret; +} + +/* + * struct tracer operations + */ + +static void blk_tracer_print_header(struct seq_file *m) +{ + if (!(blk_tracer_flags.val & TRACE_BLK_OPT_CLASSIC)) + return; + seq_puts(m, "# DEV CPU TIMESTAMP PID ACT FLG\n" + "# | | | | | |\n"); +} + +static void blk_tracer_start(struct trace_array *tr) +{ + blk_tracer_enabled = true; +} + +static int blk_tracer_init(struct trace_array *tr) +{ + blk_tr = tr; + blk_tracer_start(tr); + return 0; +} + +static void blk_tracer_stop(struct trace_array *tr) +{ + blk_tracer_enabled = false; +} + +static void blk_tracer_reset(struct trace_array *tr) +{ + blk_tracer_stop(tr); +} + +static const struct { + const char *act[2]; + int (*print)(struct trace_seq *s, const struct trace_entry *ent); +} what2act[] = { + [__BLK_TA_QUEUE] = {{ "Q", "queue" }, blk_log_generic }, + [__BLK_TA_BACKMERGE] = {{ "M", "backmerge" }, blk_log_generic }, + [__BLK_TA_FRONTMERGE] = {{ "F", "frontmerge" }, blk_log_generic }, + [__BLK_TA_GETRQ] = {{ "G", "getrq" }, blk_log_generic }, + [__BLK_TA_SLEEPRQ] = {{ "S", "sleeprq" }, blk_log_generic }, + [__BLK_TA_REQUEUE] = {{ "R", "requeue" }, blk_log_with_error }, + [__BLK_TA_ISSUE] = {{ "D", "issue" }, blk_log_generic }, + [__BLK_TA_COMPLETE] = {{ "C", "complete" }, blk_log_with_error }, + [__BLK_TA_PLUG] = {{ "P", "plug" }, blk_log_plug }, + [__BLK_TA_UNPLUG_IO] = {{ "U", "unplug_io" }, blk_log_unplug }, + [__BLK_TA_UNPLUG_TIMER] = {{ "UT", "unplug_timer" }, blk_log_unplug }, + [__BLK_TA_INSERT] = {{ "I", "insert" }, blk_log_generic }, + [__BLK_TA_SPLIT] = {{ "X", "split" }, blk_log_split }, + [__BLK_TA_BOUNCE] = {{ "B", "bounce" }, blk_log_generic }, + [__BLK_TA_REMAP] = {{ "A", "remap" }, blk_log_remap }, +}; + +static enum print_line_t print_one_line(struct trace_iterator *iter, + bool classic) +{ + struct trace_seq *s = &iter->seq; + const struct blk_io_trace *t; + u16 what; + int ret; + bool long_act; + blk_log_action_t *log_action; + + t = te_blk_io_trace(iter->ent); + what = t->action & ((1 << BLK_TC_SHIFT) - 1); + long_act = !!(trace_flags & TRACE_ITER_VERBOSE); + log_action = classic ? &blk_log_action_classic : &blk_log_action; + + if (t->action == BLK_TN_MESSAGE) { + ret = log_action(iter, long_act ? "message" : "m"); + if (ret) + ret = blk_log_msg(s, iter->ent); + goto out; + } + + if (unlikely(what == 0 || what >= ARRAY_SIZE(what2act))) + ret = trace_seq_printf(s, "Unknown action %x\n", what); + else { + ret = log_action(iter, what2act[what].act[long_act]); + if (ret) + ret = what2act[what].print(s, iter->ent); + } +out: + return ret ? TRACE_TYPE_HANDLED : TRACE_TYPE_PARTIAL_LINE; +} + +static enum print_line_t blk_trace_event_print(struct trace_iterator *iter, + int flags) +{ + return print_one_line(iter, false); +} + +static int blk_trace_synthesize_old_trace(struct trace_iterator *iter) +{ + struct trace_seq *s = &iter->seq; + struct blk_io_trace *t = (struct blk_io_trace *)iter->ent; + const int offset = offsetof(struct blk_io_trace, sector); + struct blk_io_trace old = { + .magic = BLK_IO_TRACE_MAGIC | BLK_IO_TRACE_VERSION, + .time = iter->ts, + }; + + if (!trace_seq_putmem(s, &old, offset)) + return 0; + return trace_seq_putmem(s, &t->sector, + sizeof(old) - offset + t->pdu_len); +} + +static enum print_line_t +blk_trace_event_print_binary(struct trace_iterator *iter, int flags) +{ + return blk_trace_synthesize_old_trace(iter) ? + TRACE_TYPE_HANDLED : TRACE_TYPE_PARTIAL_LINE; +} + +static enum print_line_t blk_tracer_print_line(struct trace_iterator *iter) +{ + if (!(blk_tracer_flags.val & TRACE_BLK_OPT_CLASSIC)) + return TRACE_TYPE_UNHANDLED; + + return print_one_line(iter, true); +} + +static int blk_tracer_set_flag(u32 old_flags, u32 bit, int set) +{ + /* don't output context-info for blk_classic output */ + if (bit == TRACE_BLK_OPT_CLASSIC) { + if (set) + trace_flags &= ~TRACE_ITER_CONTEXT_INFO; + else + trace_flags |= TRACE_ITER_CONTEXT_INFO; + } + return 0; +} + +static struct tracer blk_tracer __read_mostly = { + .name = "blk", + .init = blk_tracer_init, + .reset = blk_tracer_reset, + .start = blk_tracer_start, + .stop = blk_tracer_stop, + .print_header = blk_tracer_print_header, + .print_line = blk_tracer_print_line, + .flags = &blk_tracer_flags, + .set_flag = blk_tracer_set_flag, +}; + +static struct trace_event trace_blk_event = { + .type = TRACE_BLK, + .trace = blk_trace_event_print, + .binary = blk_trace_event_print_binary, +}; + +static int __init init_blk_tracer(void) +{ + if (!register_ftrace_event(&trace_blk_event)) { + pr_warning("Warning: could not register block events\n"); + return 1; + } + + if (register_tracer(&blk_tracer) != 0) { + pr_warning("Warning: could not register the block tracer\n"); + unregister_ftrace_event(&trace_blk_event); + return 1; + } + + return 0; +} + +device_initcall(init_blk_tracer); + +static int blk_trace_remove_queue(struct request_queue *q) +{ + struct blk_trace *bt; + + bt = xchg(&q->blk_trace, NULL); + if (bt == NULL) + return -EINVAL; + + if (atomic_dec_and_test(&blk_probes_ref)) + blk_unregister_tracepoints(); + + blk_trace_free(bt); + return 0; +} + +/* + * Setup everything required to start tracing + */ +static int blk_trace_setup_queue(struct request_queue *q, + struct block_device *bdev) +{ + struct blk_trace *old_bt, *bt = NULL; + int ret = -ENOMEM; + + bt = kzalloc(sizeof(*bt), GFP_KERNEL); + if (!bt) + return -ENOMEM; + + bt->msg_data = __alloc_percpu(BLK_TN_MAX_MSG, __alignof__(char)); + if (!bt->msg_data) + goto free_bt; + + bt->dev = bdev->bd_dev; + bt->act_mask = (u16)-1; + + blk_trace_setup_lba(bt, bdev); + + old_bt = xchg(&q->blk_trace, bt); + if (old_bt != NULL) { + (void)xchg(&q->blk_trace, old_bt); + ret = -EBUSY; + goto free_bt; + } + + if (atomic_inc_return(&blk_probes_ref) == 1) + blk_register_tracepoints(); + return 0; + +free_bt: + blk_trace_free(bt); + return ret; +} + +/* + * sysfs interface to enable and configure tracing + */ + +static ssize_t sysfs_blk_trace_attr_show(struct device *dev, + struct device_attribute *attr, + char *buf); +static ssize_t sysfs_blk_trace_attr_store(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count); +#define BLK_TRACE_DEVICE_ATTR(_name) \ + DEVICE_ATTR(_name, S_IRUGO | S_IWUSR, \ + sysfs_blk_trace_attr_show, \ + sysfs_blk_trace_attr_store) + +static BLK_TRACE_DEVICE_ATTR(enable); +static BLK_TRACE_DEVICE_ATTR(act_mask); +static BLK_TRACE_DEVICE_ATTR(pid); +static BLK_TRACE_DEVICE_ATTR(start_lba); +static BLK_TRACE_DEVICE_ATTR(end_lba); + +static struct attribute *blk_trace_attrs[] = { + &dev_attr_enable.attr, + &dev_attr_act_mask.attr, + &dev_attr_pid.attr, + &dev_attr_start_lba.attr, + &dev_attr_end_lba.attr, + NULL +}; + +struct attribute_group blk_trace_attr_group = { + .name = "trace", + .attrs = blk_trace_attrs, +}; + +static const struct { + int mask; + const char *str; +} mask_maps[] = { + { BLK_TC_READ, "read" }, + { BLK_TC_WRITE, "write" }, + { BLK_TC_BARRIER, "barrier" }, + { BLK_TC_SYNC, "sync" }, + { BLK_TC_QUEUE, "queue" }, + { BLK_TC_REQUEUE, "requeue" }, + { BLK_TC_ISSUE, "issue" }, + { BLK_TC_COMPLETE, "complete" }, + { BLK_TC_FS, "fs" }, + { BLK_TC_PC, "pc" }, + { BLK_TC_AHEAD, "ahead" }, + { BLK_TC_META, "meta" }, + { BLK_TC_DISCARD, "discard" }, + { BLK_TC_DRV_DATA, "drv_data" }, +}; + +static int blk_trace_str2mask(const char *str) +{ + int i; + int mask = 0; + char *buf, *s, *token; + + buf = kstrdup(str, GFP_KERNEL); + if (buf == NULL) + return -ENOMEM; + s = strstrip(buf); + + while (1) { + token = strsep(&s, ","); + if (token == NULL) + break; + + if (*token == '\0') + continue; + + for (i = 0; i < ARRAY_SIZE(mask_maps); i++) { + if (strcasecmp(token, mask_maps[i].str) == 0) { + mask |= mask_maps[i].mask; + break; + } + } + if (i == ARRAY_SIZE(mask_maps)) { + mask = -EINVAL; + break; + } + } + kfree(buf); + + return mask; +} + +static ssize_t blk_trace_mask2str(char *buf, int mask) +{ + int i; + char *p = buf; + + for (i = 0; i < ARRAY_SIZE(mask_maps); i++) { + if (mask & mask_maps[i].mask) { + p += sprintf(p, "%s%s", + (p == buf) ? "" : ",", mask_maps[i].str); + } + } + *p++ = '\n'; + + return p - buf; +} + +static struct request_queue *blk_trace_get_queue(struct block_device *bdev) +{ + if (bdev->bd_disk == NULL) + return NULL; + + return bdev_get_queue(bdev); +} + +static ssize_t sysfs_blk_trace_attr_show(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + struct hd_struct *p = dev_to_part(dev); + struct request_queue *q; + struct block_device *bdev; + ssize_t ret = -ENXIO; + + lock_kernel(); + bdev = bdget(part_devt(p)); + if (bdev == NULL) + goto out_unlock_kernel; + + q = blk_trace_get_queue(bdev); + if (q == NULL) + goto out_bdput; + + mutex_lock(&bdev->bd_mutex); + + if (attr == &dev_attr_enable) { + ret = sprintf(buf, "%u\n", !!q->blk_trace); + goto out_unlock_bdev; + } + + if (q->blk_trace == NULL) + ret = sprintf(buf, "disabled\n"); + else if (attr == &dev_attr_act_mask) + ret = blk_trace_mask2str(buf, q->blk_trace->act_mask); + else if (attr == &dev_attr_pid) + ret = sprintf(buf, "%u\n", q->blk_trace->pid); + else if (attr == &dev_attr_start_lba) + ret = sprintf(buf, "%llu\n", q->blk_trace->start_lba); + else if (attr == &dev_attr_end_lba) + ret = sprintf(buf, "%llu\n", q->blk_trace->end_lba); + +out_unlock_bdev: + mutex_unlock(&bdev->bd_mutex); +out_bdput: + bdput(bdev); +out_unlock_kernel: + unlock_kernel(); + return ret; +} + +static ssize_t sysfs_blk_trace_attr_store(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count) +{ + struct block_device *bdev; + struct request_queue *q; + struct hd_struct *p; + u64 value; + ssize_t ret = -EINVAL; + + if (count == 0) + goto out; + + if (attr == &dev_attr_act_mask) { + if (sscanf(buf, "%llx", &value) != 1) { + /* Assume it is a list of trace category names */ + ret = blk_trace_str2mask(buf); + if (ret < 0) + goto out; + value = ret; + } + } else if (sscanf(buf, "%llu", &value) != 1) + goto out; + + ret = -ENXIO; + + lock_kernel(); + p = dev_to_part(dev); + bdev = bdget(part_devt(p)); + if (bdev == NULL) + goto out_unlock_kernel; + + q = blk_trace_get_queue(bdev); + if (q == NULL) + goto out_bdput; + + mutex_lock(&bdev->bd_mutex); + + if (attr == &dev_attr_enable) { + if (value) + ret = blk_trace_setup_queue(q, bdev); + else + ret = blk_trace_remove_queue(q); + goto out_unlock_bdev; + } + + ret = 0; + if (q->blk_trace == NULL) + ret = blk_trace_setup_queue(q, bdev); + + if (ret == 0) { + if (attr == &dev_attr_act_mask) + q->blk_trace->act_mask = value; + else if (attr == &dev_attr_pid) + q->blk_trace->pid = value; + else if (attr == &dev_attr_start_lba) + q->blk_trace->start_lba = value; + else if (attr == &dev_attr_end_lba) + q->blk_trace->end_lba = value; + } + +out_unlock_bdev: + mutex_unlock(&bdev->bd_mutex); +out_bdput: + bdput(bdev); +out_unlock_kernel: + unlock_kernel(); +out: + return ret ? ret : count; +} + +int blk_trace_init_sysfs(struct device *dev) +{ + return sysfs_create_group(&dev->kobj, &blk_trace_attr_group); +} + +void blk_trace_remove_sysfs(struct device *dev) +{ + sysfs_remove_group(&dev->kobj, &blk_trace_attr_group); +} + +#endif /* CONFIG_BLK_DEV_IO_TRACE */ + +#ifdef CONFIG_EVENT_TRACING + +void blk_dump_cmd(char *buf, struct request *rq) +{ + int i, end; + int len = rq->cmd_len; + unsigned char *cmd = rq->cmd; + + if (!blk_pc_request(rq)) { + buf[0] = '\0'; + return; + } + + for (end = len - 1; end >= 0; end--) + if (cmd[end]) + break; + end++; + + for (i = 0; i < len; i++) { + buf += sprintf(buf, "%s%02x", i == 0 ? "" : " ", cmd[i]); + if (i == end && end != len - 1) { + sprintf(buf, " .."); + break; + } + } +} + +void blk_fill_rwbs(char *rwbs, u32 rw, int bytes) +{ + int i = 0; + + if (rw & WRITE) + rwbs[i++] = 'W'; + else if (rw & 1 << BIO_RW_DISCARD) + rwbs[i++] = 'D'; + else if (bytes) + rwbs[i++] = 'R'; + else + rwbs[i++] = 'N'; + + if (rw & 1 << BIO_RW_AHEAD) + rwbs[i++] = 'A'; + if (rw & 1 << BIO_RW_BARRIER) + rwbs[i++] = 'B'; + if (rw & 1 << BIO_RW_SYNCIO) + rwbs[i++] = 'S'; + if (rw & 1 << BIO_RW_META) + rwbs[i++] = 'M'; + + rwbs[i] = '\0'; +} + +void blk_fill_rwbs_rq(char *rwbs, struct request *rq) +{ + int rw = rq->cmd_flags & 0x03; + int bytes; + + if (blk_discard_rq(rq)) + rw |= (1 << BIO_RW_DISCARD); + + bytes = blk_rq_bytes(rq); + + blk_fill_rwbs(rwbs, rw, bytes); +} + +#endif /* CONFIG_EVENT_TRACING */ + diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index fdf913dfc7e..e51a1bcb7be 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -27,10 +27,15 @@ #include <linux/sysctl.h> #include <linux/ctype.h> #include <linux/list.h> +#include <linux/hash.h> + +#include <trace/events/sched.h> #include <asm/ftrace.h> +#include <asm/setup.h> -#include "trace.h" +#include "trace_output.h" +#include "trace_stat.h" #define FTRACE_WARN_ON(cond) \ do { \ @@ -44,30 +49,35 @@ ftrace_kill(); \ } while (0) +/* hash bits for specific function selection */ +#define FTRACE_HASH_BITS 7 +#define FTRACE_FUNC_HASHSIZE (1 << FTRACE_HASH_BITS) + /* ftrace_enabled is a method to turn ftrace on or off */ int ftrace_enabled __read_mostly; static int last_ftrace_enabled; -/* set when tracing only a pid */ -struct pid *ftrace_pid_trace; -static struct pid * const ftrace_swapper_pid = &init_struct_pid; - /* Quick disabling of function tracer. */ int function_trace_stop; +/* List for set_ftrace_pid's pids. */ +LIST_HEAD(ftrace_pids); +struct ftrace_pid { + struct list_head list; + struct pid *pid; +}; + /* * ftrace_disabled is set when an anomaly is discovered. * ftrace_disabled is much stronger than ftrace_enabled. */ static int ftrace_disabled __read_mostly; -static DEFINE_SPINLOCK(ftrace_lock); -static DEFINE_MUTEX(ftrace_sysctl_lock); -static DEFINE_MUTEX(ftrace_start_lock); +static DEFINE_MUTEX(ftrace_lock); static struct ftrace_ops ftrace_list_end __read_mostly = { - .func = ftrace_stub, + .func = ftrace_stub, }; static struct ftrace_ops *ftrace_list __read_mostly = &ftrace_list_end; @@ -75,6 +85,10 @@ ftrace_func_t ftrace_trace_function __read_mostly = ftrace_stub; ftrace_func_t __ftrace_trace_function __read_mostly = ftrace_stub; ftrace_func_t ftrace_pid_function __read_mostly = ftrace_stub; +#ifdef CONFIG_FUNCTION_GRAPH_TRACER +static int ftrace_set_func(unsigned long *array, int *idx, char *buffer); +#endif + static void ftrace_list_func(unsigned long ip, unsigned long parent_ip) { struct ftrace_ops *op = ftrace_list; @@ -134,9 +148,6 @@ static void ftrace_test_stop_func(unsigned long ip, unsigned long parent_ip) static int __register_ftrace_function(struct ftrace_ops *ops) { - /* should not be called from interrupt context */ - spin_lock(&ftrace_lock); - ops->next = ftrace_list; /* * We are entering ops into the ftrace_list but another @@ -155,7 +166,7 @@ static int __register_ftrace_function(struct ftrace_ops *ops) else func = ftrace_list_func; - if (ftrace_pid_trace) { + if (!list_empty(&ftrace_pids)) { set_ftrace_pid_function(func); func = ftrace_pid_func; } @@ -172,18 +183,12 @@ static int __register_ftrace_function(struct ftrace_ops *ops) #endif } - spin_unlock(&ftrace_lock); - return 0; } static int __unregister_ftrace_function(struct ftrace_ops *ops) { struct ftrace_ops **p; - int ret = 0; - - /* should not be called from interrupt context */ - spin_lock(&ftrace_lock); /* * If we are removing the last function, then simply point @@ -192,17 +197,15 @@ static int __unregister_ftrace_function(struct ftrace_ops *ops) if (ftrace_list == ops && ops->next == &ftrace_list_end) { ftrace_trace_function = ftrace_stub; ftrace_list = &ftrace_list_end; - goto out; + return 0; } for (p = &ftrace_list; *p != &ftrace_list_end; p = &(*p)->next) if (*p == ops) break; - if (*p != ops) { - ret = -1; - goto out; - } + if (*p != ops) + return -1; *p = (*p)->next; @@ -211,7 +214,7 @@ static int __unregister_ftrace_function(struct ftrace_ops *ops) if (ftrace_list->next == &ftrace_list_end) { ftrace_func_t func = ftrace_list->func; - if (ftrace_pid_trace) { + if (!list_empty(&ftrace_pids)) { set_ftrace_pid_function(func); func = ftrace_pid_func; } @@ -223,25 +226,23 @@ static int __unregister_ftrace_function(struct ftrace_ops *ops) } } - out: - spin_unlock(&ftrace_lock); - - return ret; + return 0; } static void ftrace_update_pid_func(void) { ftrace_func_t func; - /* should not be called from interrupt context */ - spin_lock(&ftrace_lock); - if (ftrace_trace_function == ftrace_stub) - goto out; + return; +#ifdef CONFIG_HAVE_FUNCTION_TRACE_MCOUNT_TEST func = ftrace_trace_function; +#else + func = __ftrace_trace_function; +#endif - if (ftrace_pid_trace) { + if (!list_empty(&ftrace_pids)) { set_ftrace_pid_function(func); func = ftrace_pid_func; } else { @@ -254,23 +255,601 @@ static void ftrace_update_pid_func(void) #else __ftrace_trace_function = func; #endif +} + +#ifdef CONFIG_FUNCTION_PROFILER +struct ftrace_profile { + struct hlist_node node; + unsigned long ip; + unsigned long counter; +#ifdef CONFIG_FUNCTION_GRAPH_TRACER + unsigned long long time; +#endif +}; + +struct ftrace_profile_page { + struct ftrace_profile_page *next; + unsigned long index; + struct ftrace_profile records[]; +}; + +struct ftrace_profile_stat { + atomic_t disabled; + struct hlist_head *hash; + struct ftrace_profile_page *pages; + struct ftrace_profile_page *start; + struct tracer_stat stat; +}; + +#define PROFILE_RECORDS_SIZE \ + (PAGE_SIZE - offsetof(struct ftrace_profile_page, records)) + +#define PROFILES_PER_PAGE \ + (PROFILE_RECORDS_SIZE / sizeof(struct ftrace_profile)) + +static int ftrace_profile_bits __read_mostly; +static int ftrace_profile_enabled __read_mostly; + +/* ftrace_profile_lock - synchronize the enable and disable of the profiler */ +static DEFINE_MUTEX(ftrace_profile_lock); + +static DEFINE_PER_CPU(struct ftrace_profile_stat, ftrace_profile_stats); + +#define FTRACE_PROFILE_HASH_SIZE 1024 /* must be power of 2 */ + +static void * +function_stat_next(void *v, int idx) +{ + struct ftrace_profile *rec = v; + struct ftrace_profile_page *pg; + + pg = (struct ftrace_profile_page *)((unsigned long)rec & PAGE_MASK); + + again: + if (idx != 0) + rec++; + + if ((void *)rec >= (void *)&pg->records[pg->index]) { + pg = pg->next; + if (!pg) + return NULL; + rec = &pg->records[0]; + if (!rec->counter) + goto again; + } + + return rec; +} + +static void *function_stat_start(struct tracer_stat *trace) +{ + struct ftrace_profile_stat *stat = + container_of(trace, struct ftrace_profile_stat, stat); + + if (!stat || !stat->start) + return NULL; + + return function_stat_next(&stat->start->records[0], 0); +} + +#ifdef CONFIG_FUNCTION_GRAPH_TRACER +/* function graph compares on total time */ +static int function_stat_cmp(void *p1, void *p2) +{ + struct ftrace_profile *a = p1; + struct ftrace_profile *b = p2; + + if (a->time < b->time) + return -1; + if (a->time > b->time) + return 1; + else + return 0; +} +#else +/* not function graph compares against hits */ +static int function_stat_cmp(void *p1, void *p2) +{ + struct ftrace_profile *a = p1; + struct ftrace_profile *b = p2; + + if (a->counter < b->counter) + return -1; + if (a->counter > b->counter) + return 1; + else + return 0; +} +#endif + +static int function_stat_headers(struct seq_file *m) +{ +#ifdef CONFIG_FUNCTION_GRAPH_TRACER + seq_printf(m, " Function " + "Hit Time Avg\n" + " -------- " + "--- ---- ---\n"); +#else + seq_printf(m, " Function Hit\n" + " -------- ---\n"); +#endif + return 0; +} + +static int function_stat_show(struct seq_file *m, void *v) +{ + struct ftrace_profile *rec = v; + char str[KSYM_SYMBOL_LEN]; +#ifdef CONFIG_FUNCTION_GRAPH_TRACER + static DEFINE_MUTEX(mutex); + static struct trace_seq s; + unsigned long long avg; +#endif + + kallsyms_lookup(rec->ip, NULL, NULL, NULL, str); + seq_printf(m, " %-30.30s %10lu", str, rec->counter); + +#ifdef CONFIG_FUNCTION_GRAPH_TRACER + seq_printf(m, " "); + avg = rec->time; + do_div(avg, rec->counter); + + mutex_lock(&mutex); + trace_seq_init(&s); + trace_print_graph_duration(rec->time, &s); + trace_seq_puts(&s, " "); + trace_print_graph_duration(avg, &s); + trace_print_seq(m, &s); + mutex_unlock(&mutex); +#endif + seq_putc(m, '\n'); + + return 0; +} + +static void ftrace_profile_reset(struct ftrace_profile_stat *stat) +{ + struct ftrace_profile_page *pg; + + pg = stat->pages = stat->start; + + while (pg) { + memset(pg->records, 0, PROFILE_RECORDS_SIZE); + pg->index = 0; + pg = pg->next; + } + + memset(stat->hash, 0, + FTRACE_PROFILE_HASH_SIZE * sizeof(struct hlist_head)); +} + +int ftrace_profile_pages_init(struct ftrace_profile_stat *stat) +{ + struct ftrace_profile_page *pg; + int functions; + int pages; + int i; + + /* If we already allocated, do nothing */ + if (stat->pages) + return 0; + + stat->pages = (void *)get_zeroed_page(GFP_KERNEL); + if (!stat->pages) + return -ENOMEM; + +#ifdef CONFIG_DYNAMIC_FTRACE + functions = ftrace_update_tot_cnt; +#else + /* + * We do not know the number of functions that exist because + * dynamic tracing is what counts them. With past experience + * we have around 20K functions. That should be more than enough. + * It is highly unlikely we will execute every function in + * the kernel. + */ + functions = 20000; +#endif + + pg = stat->start = stat->pages; + + pages = DIV_ROUND_UP(functions, PROFILES_PER_PAGE); + + for (i = 0; i < pages; i++) { + pg->next = (void *)get_zeroed_page(GFP_KERNEL); + if (!pg->next) + goto out_free; + pg = pg->next; + } + + return 0; + + out_free: + pg = stat->start; + while (pg) { + unsigned long tmp = (unsigned long)pg; + + pg = pg->next; + free_page(tmp); + } + + free_page((unsigned long)stat->pages); + stat->pages = NULL; + stat->start = NULL; + + return -ENOMEM; +} + +static int ftrace_profile_init_cpu(int cpu) +{ + struct ftrace_profile_stat *stat; + int size; + + stat = &per_cpu(ftrace_profile_stats, cpu); + + if (stat->hash) { + /* If the profile is already created, simply reset it */ + ftrace_profile_reset(stat); + return 0; + } + + /* + * We are profiling all functions, but usually only a few thousand + * functions are hit. We'll make a hash of 1024 items. + */ + size = FTRACE_PROFILE_HASH_SIZE; + + stat->hash = kzalloc(sizeof(struct hlist_head) * size, GFP_KERNEL); + + if (!stat->hash) + return -ENOMEM; + + if (!ftrace_profile_bits) { + size--; + + for (; size; size >>= 1) + ftrace_profile_bits++; + } + + /* Preallocate the function profiling pages */ + if (ftrace_profile_pages_init(stat) < 0) { + kfree(stat->hash); + stat->hash = NULL; + return -ENOMEM; + } + + return 0; +} + +static int ftrace_profile_init(void) +{ + int cpu; + int ret = 0; + + for_each_online_cpu(cpu) { + ret = ftrace_profile_init_cpu(cpu); + if (ret) + break; + } + + return ret; +} + +/* interrupts must be disabled */ +static struct ftrace_profile * +ftrace_find_profiled_func(struct ftrace_profile_stat *stat, unsigned long ip) +{ + struct ftrace_profile *rec; + struct hlist_head *hhd; + struct hlist_node *n; + unsigned long key; + + key = hash_long(ip, ftrace_profile_bits); + hhd = &stat->hash[key]; + + if (hlist_empty(hhd)) + return NULL; + + hlist_for_each_entry_rcu(rec, n, hhd, node) { + if (rec->ip == ip) + return rec; + } + + return NULL; +} + +static void ftrace_add_profile(struct ftrace_profile_stat *stat, + struct ftrace_profile *rec) +{ + unsigned long key; + + key = hash_long(rec->ip, ftrace_profile_bits); + hlist_add_head_rcu(&rec->node, &stat->hash[key]); +} + +/* + * The memory is already allocated, this simply finds a new record to use. + */ +static struct ftrace_profile * +ftrace_profile_alloc(struct ftrace_profile_stat *stat, unsigned long ip) +{ + struct ftrace_profile *rec = NULL; + + /* prevent recursion (from NMIs) */ + if (atomic_inc_return(&stat->disabled) != 1) + goto out; + + /* + * Try to find the function again since an NMI + * could have added it + */ + rec = ftrace_find_profiled_func(stat, ip); + if (rec) + goto out; + + if (stat->pages->index == PROFILES_PER_PAGE) { + if (!stat->pages->next) + goto out; + stat->pages = stat->pages->next; + } + + rec = &stat->pages->records[stat->pages->index++]; + rec->ip = ip; + ftrace_add_profile(stat, rec); + + out: + atomic_dec(&stat->disabled); + + return rec; +} + +static void +function_profile_call(unsigned long ip, unsigned long parent_ip) +{ + struct ftrace_profile_stat *stat; + struct ftrace_profile *rec; + unsigned long flags; + + if (!ftrace_profile_enabled) + return; + + local_irq_save(flags); + + stat = &__get_cpu_var(ftrace_profile_stats); + if (!stat->hash || !ftrace_profile_enabled) + goto out; + + rec = ftrace_find_profiled_func(stat, ip); + if (!rec) { + rec = ftrace_profile_alloc(stat, ip); + if (!rec) + goto out; + } + + rec->counter++; + out: + local_irq_restore(flags); +} + +#ifdef CONFIG_FUNCTION_GRAPH_TRACER +static int profile_graph_entry(struct ftrace_graph_ent *trace) +{ + function_profile_call(trace->func, 0); + return 1; +} + +static void profile_graph_return(struct ftrace_graph_ret *trace) +{ + struct ftrace_profile_stat *stat; + unsigned long long calltime; + struct ftrace_profile *rec; + unsigned long flags; + + local_irq_save(flags); + stat = &__get_cpu_var(ftrace_profile_stats); + if (!stat->hash || !ftrace_profile_enabled) + goto out; + + calltime = trace->rettime - trace->calltime; + + if (!(trace_flags & TRACE_ITER_GRAPH_TIME)) { + int index; + + index = trace->depth; + + /* Append this call time to the parent time to subtract */ + if (index) + current->ret_stack[index - 1].subtime += calltime; + + if (current->ret_stack[index].subtime < calltime) + calltime -= current->ret_stack[index].subtime; + else + calltime = 0; + } + + rec = ftrace_find_profiled_func(stat, trace->func); + if (rec) + rec->time += calltime; + + out: + local_irq_restore(flags); +} + +static int register_ftrace_profiler(void) +{ + return register_ftrace_graph(&profile_graph_return, + &profile_graph_entry); +} + +static void unregister_ftrace_profiler(void) +{ + unregister_ftrace_graph(); +} +#else +static struct ftrace_ops ftrace_profile_ops __read_mostly = +{ + .func = function_profile_call, +}; + +static int register_ftrace_profiler(void) +{ + return register_ftrace_function(&ftrace_profile_ops); +} + +static void unregister_ftrace_profiler(void) +{ + unregister_ftrace_function(&ftrace_profile_ops); +} +#endif /* CONFIG_FUNCTION_GRAPH_TRACER */ + +static ssize_t +ftrace_profile_write(struct file *filp, const char __user *ubuf, + size_t cnt, loff_t *ppos) +{ + unsigned long val; + char buf[64]; /* big enough to hold a number */ + int ret; + + if (cnt >= sizeof(buf)) + return -EINVAL; + + if (copy_from_user(&buf, ubuf, cnt)) + return -EFAULT; + + buf[cnt] = 0; + + ret = strict_strtoul(buf, 10, &val); + if (ret < 0) + return ret; + val = !!val; + + mutex_lock(&ftrace_profile_lock); + if (ftrace_profile_enabled ^ val) { + if (val) { + ret = ftrace_profile_init(); + if (ret < 0) { + cnt = ret; + goto out; + } + + ret = register_ftrace_profiler(); + if (ret < 0) { + cnt = ret; + goto out; + } + ftrace_profile_enabled = 1; + } else { + ftrace_profile_enabled = 0; + /* + * unregister_ftrace_profiler calls stop_machine + * so this acts like an synchronize_sched. + */ + unregister_ftrace_profiler(); + } + } out: - spin_unlock(&ftrace_lock); + mutex_unlock(&ftrace_profile_lock); + + *ppos += cnt; + + return cnt; } +static ssize_t +ftrace_profile_read(struct file *filp, char __user *ubuf, + size_t cnt, loff_t *ppos) +{ + char buf[64]; /* big enough to hold a number */ + int r; + + r = sprintf(buf, "%u\n", ftrace_profile_enabled); + return simple_read_from_buffer(ubuf, cnt, ppos, buf, r); +} + +static const struct file_operations ftrace_profile_fops = { + .open = tracing_open_generic, + .read = ftrace_profile_read, + .write = ftrace_profile_write, +}; + +/* used to initialize the real stat files */ +static struct tracer_stat function_stats __initdata = { + .name = "functions", + .stat_start = function_stat_start, + .stat_next = function_stat_next, + .stat_cmp = function_stat_cmp, + .stat_headers = function_stat_headers, + .stat_show = function_stat_show +}; + +static __init void ftrace_profile_debugfs(struct dentry *d_tracer) +{ + struct ftrace_profile_stat *stat; + struct dentry *entry; + char *name; + int ret; + int cpu; + + for_each_possible_cpu(cpu) { + stat = &per_cpu(ftrace_profile_stats, cpu); + + /* allocate enough for function name + cpu number */ + name = kmalloc(32, GFP_KERNEL); + if (!name) { + /* + * The files created are permanent, if something happens + * we still do not free memory. + */ + WARN(1, + "Could not allocate stat file for cpu %d\n", + cpu); + return; + } + stat->stat = function_stats; + snprintf(name, 32, "function%d", cpu); + stat->stat.name = name; + ret = register_stat_tracer(&stat->stat); + if (ret) { + WARN(1, + "Could not register function stat for cpu %d\n", + cpu); + kfree(name); + return; + } + } + + entry = debugfs_create_file("function_profile_enabled", 0644, + d_tracer, NULL, &ftrace_profile_fops); + if (!entry) + pr_warning("Could not create debugfs " + "'function_profile_enabled' entry\n"); +} + +#else /* CONFIG_FUNCTION_PROFILER */ +static __init void ftrace_profile_debugfs(struct dentry *d_tracer) +{ +} +#endif /* CONFIG_FUNCTION_PROFILER */ + +static struct pid * const ftrace_swapper_pid = &init_struct_pid; + #ifdef CONFIG_DYNAMIC_FTRACE + #ifndef CONFIG_FTRACE_MCOUNT_RECORD # error Dynamic ftrace depends on MCOUNT_RECORD #endif -/* - * Since MCOUNT_ADDR may point to mcount itself, we do not want - * to get it confused by reading a reference in the code as we - * are parsing on objcopy output of text. Use a variable for - * it instead. - */ -static unsigned long mcount_addr = MCOUNT_ADDR; +static struct hlist_head ftrace_func_hash[FTRACE_FUNC_HASHSIZE] __read_mostly; + +struct ftrace_func_probe { + struct hlist_node node; + struct ftrace_probe_ops *ops; + unsigned long flags; + unsigned long ip; + void *data; + struct rcu_head rcu; +}; enum { FTRACE_ENABLE_CALLS = (1 << 0), @@ -284,13 +863,13 @@ enum { static int ftrace_filtered; -static LIST_HEAD(ftrace_new_addrs); +static struct dyn_ftrace *ftrace_new_addrs; static DEFINE_MUTEX(ftrace_regex_lock); struct ftrace_page { struct ftrace_page *next; - unsigned long index; + int index; struct dyn_ftrace records[]; }; @@ -305,6 +884,19 @@ static struct ftrace_page *ftrace_pages; static struct dyn_ftrace *ftrace_free_records; +/* + * This is a double for. Do not use 'break' to break out of the loop, + * you must use a goto. + */ +#define do_for_each_ftrace_rec(pg, rec) \ + for (pg = ftrace_pages_start; pg; pg = pg->next) { \ + int _____i; \ + for (_____i = 0; _____i < pg->index; _____i++) { \ + rec = &pg->records[_____i]; + +#define while_for_each_ftrace_rec() \ + } \ + } #ifdef CONFIG_KPROBES @@ -338,36 +930,11 @@ static inline int record_frozen(struct dyn_ftrace *rec) static void ftrace_free_rec(struct dyn_ftrace *rec) { - rec->ip = (unsigned long)ftrace_free_records; + rec->freelist = ftrace_free_records; ftrace_free_records = rec; rec->flags |= FTRACE_FL_FREE; } -void ftrace_release(void *start, unsigned long size) -{ - struct dyn_ftrace *rec; - struct ftrace_page *pg; - unsigned long s = (unsigned long)start; - unsigned long e = s + size; - int i; - - if (ftrace_disabled || !start) - return; - - /* should not be called from interrupt context */ - spin_lock(&ftrace_lock); - - for (pg = ftrace_pages_start; pg; pg = pg->next) { - for (i = 0; i < pg->index; i++) { - rec = &pg->records[i]; - - if ((rec->ip >= s) && (rec->ip < e)) - ftrace_free_rec(rec); - } - } - spin_unlock(&ftrace_lock); -} - static struct dyn_ftrace *ftrace_alloc_dyn_node(unsigned long ip) { struct dyn_ftrace *rec; @@ -382,7 +949,7 @@ static struct dyn_ftrace *ftrace_alloc_dyn_node(unsigned long ip) return NULL; } - ftrace_free_records = (void *)rec->ip; + ftrace_free_records = rec->freelist; memset(rec, 0, sizeof(*rec)); return rec; } @@ -414,8 +981,8 @@ ftrace_record_ip(unsigned long ip) return NULL; rec->ip = ip; - - list_add(&rec->list, &ftrace_new_addrs); + rec->newlist = ftrace_new_addrs; + ftrace_new_addrs = rec; return rec; } @@ -461,111 +1028,70 @@ static void ftrace_bug(int failed, unsigned long ip) static int __ftrace_replace_code(struct dyn_ftrace *rec, int enable) { - unsigned long ip, fl; unsigned long ftrace_addr; + unsigned long flag = 0UL; - ftrace_addr = (unsigned long)ftrace_caller; - - ip = rec->ip; + ftrace_addr = (unsigned long)FTRACE_ADDR; /* - * If this record is not to be traced and - * it is not enabled then do nothing. + * If this record is not to be traced or we want to disable it, + * then disable it. * - * If this record is not to be traced and - * it is enabled then disabled it. + * If we want to enable it and filtering is off, then enable it. * + * If we want to enable it and filtering is on, enable it only if + * it's filtered */ - if (rec->flags & FTRACE_FL_NOTRACE) { - if (rec->flags & FTRACE_FL_ENABLED) - rec->flags &= ~FTRACE_FL_ENABLED; - else - return 0; - - } else if (ftrace_filtered && enable) { - /* - * Filtering is on: - */ - - fl = rec->flags & (FTRACE_FL_FILTER | FTRACE_FL_ENABLED); - - /* Record is filtered and enabled, do nothing */ - if (fl == (FTRACE_FL_FILTER | FTRACE_FL_ENABLED)) - return 0; - - /* Record is not filtered and is not enabled do nothing */ - if (!fl) - return 0; - - /* Record is not filtered but enabled, disable it */ - if (fl == FTRACE_FL_ENABLED) - rec->flags &= ~FTRACE_FL_ENABLED; - else - /* Otherwise record is filtered but not enabled, enable it */ - rec->flags |= FTRACE_FL_ENABLED; - } else { - /* Disable or not filtered */ - - if (enable) { - /* if record is enabled, do nothing */ - if (rec->flags & FTRACE_FL_ENABLED) - return 0; - - rec->flags |= FTRACE_FL_ENABLED; - - } else { + if (enable && !(rec->flags & FTRACE_FL_NOTRACE)) { + if (!ftrace_filtered || (rec->flags & FTRACE_FL_FILTER)) + flag = FTRACE_FL_ENABLED; + } - /* if record is not enabled do nothing */ - if (!(rec->flags & FTRACE_FL_ENABLED)) - return 0; + /* If the state of this record hasn't changed, then do nothing */ + if ((rec->flags & FTRACE_FL_ENABLED) == flag) + return 0; - rec->flags &= ~FTRACE_FL_ENABLED; - } + if (flag) { + rec->flags |= FTRACE_FL_ENABLED; + return ftrace_make_call(rec, ftrace_addr); } - if (rec->flags & FTRACE_FL_ENABLED) - return ftrace_make_call(rec, ftrace_addr); - else - return ftrace_make_nop(NULL, rec, ftrace_addr); + rec->flags &= ~FTRACE_FL_ENABLED; + return ftrace_make_nop(NULL, rec, ftrace_addr); } static void ftrace_replace_code(int enable) { - int i, failed; struct dyn_ftrace *rec; struct ftrace_page *pg; + int failed; - for (pg = ftrace_pages_start; pg; pg = pg->next) { - for (i = 0; i < pg->index; i++) { - rec = &pg->records[i]; - - /* - * Skip over free records and records that have - * failed. - */ - if (rec->flags & FTRACE_FL_FREE || - rec->flags & FTRACE_FL_FAILED) - continue; + do_for_each_ftrace_rec(pg, rec) { + /* + * Skip over free records, records that have + * failed and not converted. + */ + if (rec->flags & FTRACE_FL_FREE || + rec->flags & FTRACE_FL_FAILED || + !(rec->flags & FTRACE_FL_CONVERTED)) + continue; - /* ignore updates to this record's mcount site */ - if (get_kprobe((void *)rec->ip)) { - freeze_record(rec); - continue; - } else { - unfreeze_record(rec); - } + /* ignore updates to this record's mcount site */ + if (get_kprobe((void *)rec->ip)) { + freeze_record(rec); + continue; + } else { + unfreeze_record(rec); + } - failed = __ftrace_replace_code(rec, enable); - if (failed && (rec->flags & FTRACE_FL_CONVERTED)) { - rec->flags |= FTRACE_FL_FAILED; - if ((system_state == SYSTEM_BOOTING) || - !core_kernel_text(rec->ip)) { - ftrace_free_rec(rec); - } else - ftrace_bug(failed, rec->ip); - } + failed = __ftrace_replace_code(rec, enable); + if (failed) { + rec->flags |= FTRACE_FL_FAILED; + ftrace_bug(failed, rec->ip); + /* Stop processing */ + return; } - } + } while_for_each_ftrace_rec(); } static int @@ -576,7 +1102,7 @@ ftrace_code_disable(struct module *mod, struct dyn_ftrace *rec) ip = rec->ip; - ret = ftrace_make_nop(mod, rec, mcount_addr); + ret = ftrace_make_nop(mod, rec, MCOUNT_ADDR); if (ret) { ftrace_bug(ret, ip); rec->flags |= FTRACE_FL_FAILED; @@ -585,6 +1111,24 @@ ftrace_code_disable(struct module *mod, struct dyn_ftrace *rec) return 1; } +/* + * archs can override this function if they must do something + * before the modifying code is performed. + */ +int __weak ftrace_arch_code_modify_prepare(void) +{ + return 0; +} + +/* + * archs can override this function if they must do something + * after the modifying code is performed. + */ +int __weak ftrace_arch_code_modify_post_process(void) +{ + return 0; +} + static int __ftrace_modify_code(void *data) { int *command = data; @@ -607,7 +1151,17 @@ static int __ftrace_modify_code(void *data) static void ftrace_run_update_code(int command) { + int ret; + + ret = ftrace_arch_code_modify_prepare(); + FTRACE_WARN_ON(ret); + if (ret) + return; + stop_machine(__ftrace_modify_code, &command, NULL); + + ret = ftrace_arch_code_modify_post_process(); + FTRACE_WARN_ON(ret); } static ftrace_func_t saved_ftrace_func; @@ -631,13 +1185,10 @@ static void ftrace_startup(int command) if (unlikely(ftrace_disabled)) return; - mutex_lock(&ftrace_start_lock); ftrace_start_up++; command |= FTRACE_ENABLE_CALLS; ftrace_startup_enable(command); - - mutex_unlock(&ftrace_start_lock); } static void ftrace_shutdown(int command) @@ -645,8 +1196,14 @@ static void ftrace_shutdown(int command) if (unlikely(ftrace_disabled)) return; - mutex_lock(&ftrace_start_lock); ftrace_start_up--; + /* + * Just warn in case of unbalance, no need to kill ftrace, it's not + * critical but the ftrace_call callers may be never nopped again after + * further ftrace uses. + */ + WARN_ON_ONCE(ftrace_start_up < 0); + if (!ftrace_start_up) command |= FTRACE_DISABLE_CALLS; @@ -656,11 +1213,9 @@ static void ftrace_shutdown(int command) } if (!command || !ftrace_enabled) - goto out; + return; ftrace_run_update_code(command); - out: - mutex_unlock(&ftrace_start_lock); } static void ftrace_startup_sysctl(void) @@ -670,7 +1225,6 @@ static void ftrace_startup_sysctl(void) if (unlikely(ftrace_disabled)) return; - mutex_lock(&ftrace_start_lock); /* Force update next time */ saved_ftrace_func = NULL; /* ftrace_start_up is true if we want ftrace running */ @@ -678,7 +1232,6 @@ static void ftrace_startup_sysctl(void) command |= FTRACE_ENABLE_CALLS; ftrace_run_update_code(command); - mutex_unlock(&ftrace_start_lock); } static void ftrace_shutdown_sysctl(void) @@ -688,13 +1241,11 @@ static void ftrace_shutdown_sysctl(void) if (unlikely(ftrace_disabled)) return; - mutex_lock(&ftrace_start_lock); /* ftrace_start_up is true if ftrace is running */ if (ftrace_start_up) command |= FTRACE_DISABLE_CALLS; ftrace_run_update_code(command); - mutex_unlock(&ftrace_start_lock); } static cycle_t ftrace_update_time; @@ -703,26 +1254,50 @@ unsigned long ftrace_update_tot_cnt; static int ftrace_update_code(struct module *mod) { - struct dyn_ftrace *p, *t; + struct dyn_ftrace *p; cycle_t start, stop; start = ftrace_now(raw_smp_processor_id()); ftrace_update_cnt = 0; - list_for_each_entry_safe(p, t, &ftrace_new_addrs, list) { + while (ftrace_new_addrs) { /* If something went wrong, bail without enabling anything */ if (unlikely(ftrace_disabled)) return -1; - list_del_init(&p->list); + p = ftrace_new_addrs; + ftrace_new_addrs = p->newlist; + p->flags = 0L; - /* convert record (i.e, patch mcount-call with NOP) */ - if (ftrace_code_disable(mod, p)) { - p->flags |= FTRACE_FL_CONVERTED; - ftrace_update_cnt++; - } else + /* + * Do the initial record convertion from mcount jump + * to the NOP instructions. + */ + if (!ftrace_code_disable(mod, p)) { ftrace_free_rec(p); + continue; + } + + p->flags |= FTRACE_FL_CONVERTED; + ftrace_update_cnt++; + + /* + * If the tracing is enabled, go ahead and enable the record. + * + * The reason not to enable the record immediatelly is the + * inherent check of ftrace_make_nop/ftrace_make_call for + * correct previous instructions. Making first the NOP + * conversion puts the module to the correct state, thus + * passing the ftrace_make_call check. + */ + if (ftrace_start_up) { + int failed = __ftrace_replace_code(p, 1); + if (failed) { + ftrace_bug(failed, p->ip); + ftrace_free_rec(p); + } + } } stop = ftrace_now(raw_smp_processor_id()); @@ -778,40 +1353,117 @@ static int __init ftrace_dyn_table_alloc(unsigned long num_to_init) enum { FTRACE_ITER_FILTER = (1 << 0), - FTRACE_ITER_CONT = (1 << 1), - FTRACE_ITER_NOTRACE = (1 << 2), - FTRACE_ITER_FAILURES = (1 << 3), + FTRACE_ITER_NOTRACE = (1 << 1), + FTRACE_ITER_FAILURES = (1 << 2), + FTRACE_ITER_PRINTALL = (1 << 3), + FTRACE_ITER_HASH = (1 << 4), }; #define FTRACE_BUFF_MAX (KSYM_SYMBOL_LEN+4) /* room for wildcards */ struct ftrace_iterator { struct ftrace_page *pg; - unsigned idx; + int hidx; + int idx; unsigned flags; - unsigned char buffer[FTRACE_BUFF_MAX+1]; - unsigned buffer_idx; - unsigned filtered; + struct trace_parser parser; }; static void * +t_hash_next(struct seq_file *m, void *v, loff_t *pos) +{ + struct ftrace_iterator *iter = m->private; + struct hlist_node *hnd = v; + struct hlist_head *hhd; + + WARN_ON(!(iter->flags & FTRACE_ITER_HASH)); + + (*pos)++; + + retry: + if (iter->hidx >= FTRACE_FUNC_HASHSIZE) + return NULL; + + hhd = &ftrace_func_hash[iter->hidx]; + + if (hlist_empty(hhd)) { + iter->hidx++; + hnd = NULL; + goto retry; + } + + if (!hnd) + hnd = hhd->first; + else { + hnd = hnd->next; + if (!hnd) { + iter->hidx++; + goto retry; + } + } + + return hnd; +} + +static void *t_hash_start(struct seq_file *m, loff_t *pos) +{ + struct ftrace_iterator *iter = m->private; + void *p = NULL; + loff_t l; + + if (!(iter->flags & FTRACE_ITER_HASH)) + *pos = 0; + + iter->flags |= FTRACE_ITER_HASH; + + iter->hidx = 0; + for (l = 0; l <= *pos; ) { + p = t_hash_next(m, p, &l); + if (!p) + break; + } + return p; +} + +static int t_hash_show(struct seq_file *m, void *v) +{ + struct ftrace_func_probe *rec; + struct hlist_node *hnd = v; + + rec = hlist_entry(hnd, struct ftrace_func_probe, node); + + if (rec->ops->print) + return rec->ops->print(m, rec->ip, rec->ops, rec->data); + + seq_printf(m, "%ps:%ps", (void *)rec->ip, (void *)rec->ops->func); + + if (rec->data) + seq_printf(m, ":%p", rec->data); + seq_putc(m, '\n'); + + return 0; +} + +static void * t_next(struct seq_file *m, void *v, loff_t *pos) { struct ftrace_iterator *iter = m->private; struct dyn_ftrace *rec = NULL; + if (iter->flags & FTRACE_ITER_HASH) + return t_hash_next(m, v, pos); + (*pos)++; - /* should not be called from interrupt context */ - spin_lock(&ftrace_lock); + if (iter->flags & FTRACE_ITER_PRINTALL) + return NULL; + retry: if (iter->idx >= iter->pg->index) { if (iter->pg->next) { iter->pg = iter->pg->next; iter->idx = 0; goto retry; - } else { - iter->idx = -1; } } else { rec = &iter->pg->records[iter->idx++]; @@ -832,7 +1484,6 @@ t_next(struct seq_file *m, void *v, loff_t *pos) goto retry; } } - spin_unlock(&ftrace_lock); return rec; } @@ -841,39 +1492,65 @@ static void *t_start(struct seq_file *m, loff_t *pos) { struct ftrace_iterator *iter = m->private; void *p = NULL; + loff_t l; + + mutex_lock(&ftrace_lock); + /* + * For set_ftrace_filter reading, if we have the filter + * off, we can short cut and just print out that all + * functions are enabled. + */ + if (iter->flags & FTRACE_ITER_FILTER && !ftrace_filtered) { + if (*pos > 0) + return t_hash_start(m, pos); + iter->flags |= FTRACE_ITER_PRINTALL; + return iter; + } + + if (iter->flags & FTRACE_ITER_HASH) + return t_hash_start(m, pos); - if (*pos > 0) { - if (iter->idx < 0) - return p; - (*pos)--; - iter->idx--; + iter->pg = ftrace_pages_start; + iter->idx = 0; + for (l = 0; l <= *pos; ) { + p = t_next(m, p, &l); + if (!p) + break; } - p = t_next(m, p, pos); + if (!p && iter->flags & FTRACE_ITER_FILTER) + return t_hash_start(m, pos); return p; } static void t_stop(struct seq_file *m, void *p) { + mutex_unlock(&ftrace_lock); } static int t_show(struct seq_file *m, void *v) { + struct ftrace_iterator *iter = m->private; struct dyn_ftrace *rec = v; - char str[KSYM_SYMBOL_LEN]; - if (!rec) + if (iter->flags & FTRACE_ITER_HASH) + return t_hash_show(m, v); + + if (iter->flags & FTRACE_ITER_PRINTALL) { + seq_printf(m, "#### all functions enabled ####\n"); return 0; + } - kallsyms_lookup(rec->ip, NULL, NULL, NULL, str); + if (!rec) + return 0; - seq_printf(m, "%s\n", str); + seq_printf(m, "%ps\n", (void *)rec->ip); return 0; } -static struct seq_operations show_ftrace_seq_ops = { +static const struct seq_operations show_ftrace_seq_ops = { .start = t_start, .next = t_next, .stop = t_stop, @@ -907,17 +1584,6 @@ ftrace_avail_open(struct inode *inode, struct file *file) return ret; } -int ftrace_avail_release(struct inode *inode, struct file *file) -{ - struct seq_file *m = (struct seq_file *)file->private_data; - struct ftrace_iterator *iter = m->private; - - seq_release(inode, file); - kfree(iter); - - return 0; -} - static int ftrace_failures_open(struct inode *inode, struct file *file) { @@ -941,23 +1607,16 @@ static void ftrace_filter_reset(int enable) struct ftrace_page *pg; struct dyn_ftrace *rec; unsigned long type = enable ? FTRACE_FL_FILTER : FTRACE_FL_NOTRACE; - unsigned i; - /* should not be called from interrupt context */ - spin_lock(&ftrace_lock); + mutex_lock(&ftrace_lock); if (enable) ftrace_filtered = 0; - pg = ftrace_pages_start; - while (pg) { - for (i = 0; i < pg->index; i++) { - rec = &pg->records[i]; - if (rec->flags & FTRACE_FL_FAILED) - continue; - rec->flags &= ~type; - } - pg = pg->next; - } - spin_unlock(&ftrace_lock); + do_for_each_ftrace_rec(pg, rec) { + if (rec->flags & FTRACE_FL_FAILED) + continue; + rec->flags &= ~type; + } while_for_each_ftrace_rec(); + mutex_unlock(&ftrace_lock); } static int @@ -973,9 +1632,14 @@ ftrace_regex_open(struct inode *inode, struct file *file, int enable) if (!iter) return -ENOMEM; + if (trace_parser_get_init(&iter->parser, FTRACE_BUFF_MAX)) { + kfree(iter); + return -ENOMEM; + } + mutex_lock(&ftrace_regex_lock); if ((file->f_mode & FMODE_WRITE) && - !(file->f_flags & O_APPEND)) + (file->f_flags & O_TRUNC)) ftrace_filter_reset(enable); if (file->f_mode & FMODE_READ) { @@ -987,8 +1651,10 @@ ftrace_regex_open(struct inode *inode, struct file *file, int enable) if (!ret) { struct seq_file *m = file->private_data; m->private = iter; - } else + } else { + trace_parser_put(&iter->parser); kfree(iter); + } } else file->private_data = iter; mutex_unlock(&ftrace_regex_lock); @@ -1008,16 +1674,6 @@ ftrace_notrace_open(struct inode *inode, struct file *file) return ftrace_regex_open(inode, file, 0); } -static ssize_t -ftrace_regex_read(struct file *file, char __user *ubuf, - size_t cnt, loff_t *ppos) -{ - if (file->f_mode & FMODE_READ) - return seq_read(file, ubuf, cnt, ppos); - else - return -EPERM; -} - static loff_t ftrace_regex_lseek(struct file *file, loff_t offset, int origin) { @@ -1031,93 +1687,489 @@ ftrace_regex_lseek(struct file *file, loff_t offset, int origin) return ret; } -enum { - MATCH_FULL, - MATCH_FRONT_ONLY, - MATCH_MIDDLE_ONLY, - MATCH_END_ONLY, -}; +static int ftrace_match(char *str, char *regex, int len, int type) +{ + int matched = 0; + char *ptr; -static void -ftrace_match(unsigned char *buff, int len, int enable) + switch (type) { + case MATCH_FULL: + if (strcmp(str, regex) == 0) + matched = 1; + break; + case MATCH_FRONT_ONLY: + if (strncmp(str, regex, len) == 0) + matched = 1; + break; + case MATCH_MIDDLE_ONLY: + if (strstr(str, regex)) + matched = 1; + break; + case MATCH_END_ONLY: + ptr = strstr(str, regex); + if (ptr && (ptr[len] == 0)) + matched = 1; + break; + } + + return matched; +} + +static int +ftrace_match_record(struct dyn_ftrace *rec, char *regex, int len, int type) { char str[KSYM_SYMBOL_LEN]; - char *search = NULL; + + kallsyms_lookup(rec->ip, NULL, NULL, NULL, str); + return ftrace_match(str, regex, len, type); +} + +static void ftrace_match_records(char *buff, int len, int enable) +{ + unsigned int search_len; + struct ftrace_page *pg; + struct dyn_ftrace *rec; + unsigned long flag; + char *search; + int type; + int not; + + flag = enable ? FTRACE_FL_FILTER : FTRACE_FL_NOTRACE; + type = filter_parse_regex(buff, len, &search, ¬); + + search_len = strlen(search); + + mutex_lock(&ftrace_lock); + do_for_each_ftrace_rec(pg, rec) { + + if (rec->flags & FTRACE_FL_FAILED) + continue; + + if (ftrace_match_record(rec, search, search_len, type)) { + if (not) + rec->flags &= ~flag; + else + rec->flags |= flag; + } + /* + * Only enable filtering if we have a function that + * is filtered on. + */ + if (enable && (rec->flags & FTRACE_FL_FILTER)) + ftrace_filtered = 1; + } while_for_each_ftrace_rec(); + mutex_unlock(&ftrace_lock); +} + +static int +ftrace_match_module_record(struct dyn_ftrace *rec, char *mod, + char *regex, int len, int type) +{ + char str[KSYM_SYMBOL_LEN]; + char *modname; + + kallsyms_lookup(rec->ip, NULL, NULL, &modname, str); + + if (!modname || strcmp(modname, mod)) + return 0; + + /* blank search means to match all funcs in the mod */ + if (len) + return ftrace_match(str, regex, len, type); + else + return 1; +} + +static void ftrace_match_module_records(char *buff, char *mod, int enable) +{ + unsigned search_len = 0; struct ftrace_page *pg; struct dyn_ftrace *rec; int type = MATCH_FULL; - unsigned long flag = enable ? FTRACE_FL_FILTER : FTRACE_FL_NOTRACE; - unsigned i, match = 0, search_len = 0; + char *search = buff; + unsigned long flag; int not = 0; - if (buff[0] == '!') { + flag = enable ? FTRACE_FL_FILTER : FTRACE_FL_NOTRACE; + + /* blank or '*' mean the same */ + if (strcmp(buff, "*") == 0) + buff[0] = 0; + + /* handle the case of 'dont filter this module' */ + if (strcmp(buff, "!") == 0 || strcmp(buff, "!*") == 0) { + buff[0] = 0; not = 1; - buff++; - len--; - } - - for (i = 0; i < len; i++) { - if (buff[i] == '*') { - if (!i) { - search = buff + i + 1; - type = MATCH_END_ONLY; - search_len = len - (i + 1); - } else { - if (type == MATCH_END_ONLY) { - type = MATCH_MIDDLE_ONLY; - } else { - match = i; - type = MATCH_FRONT_ONLY; - } - buff[i] = 0; - break; - } + } + + if (strlen(buff)) { + type = filter_parse_regex(buff, strlen(buff), &search, ¬); + search_len = strlen(search); + } + + mutex_lock(&ftrace_lock); + do_for_each_ftrace_rec(pg, rec) { + + if (rec->flags & FTRACE_FL_FAILED) + continue; + + if (ftrace_match_module_record(rec, mod, + search, search_len, type)) { + if (not) + rec->flags &= ~flag; + else + rec->flags |= flag; } + if (enable && (rec->flags & FTRACE_FL_FILTER)) + ftrace_filtered = 1; + + } while_for_each_ftrace_rec(); + mutex_unlock(&ftrace_lock); +} + +/* + * We register the module command as a template to show others how + * to register the a command as well. + */ + +static int +ftrace_mod_callback(char *func, char *cmd, char *param, int enable) +{ + char *mod; + + /* + * cmd == 'mod' because we only registered this func + * for the 'mod' ftrace_func_command. + * But if you register one func with multiple commands, + * you can tell which command was used by the cmd + * parameter. + */ + + /* we must have a module name */ + if (!param) + return -EINVAL; + + mod = strsep(¶m, ":"); + if (!strlen(mod)) + return -EINVAL; + + ftrace_match_module_records(func, mod, enable); + return 0; +} + +static struct ftrace_func_command ftrace_mod_cmd = { + .name = "mod", + .func = ftrace_mod_callback, +}; + +static int __init ftrace_mod_cmd_init(void) +{ + return register_ftrace_command(&ftrace_mod_cmd); +} +device_initcall(ftrace_mod_cmd_init); + +static void +function_trace_probe_call(unsigned long ip, unsigned long parent_ip) +{ + struct ftrace_func_probe *entry; + struct hlist_head *hhd; + struct hlist_node *n; + unsigned long key; + int resched; + + key = hash_long(ip, FTRACE_HASH_BITS); + + hhd = &ftrace_func_hash[key]; + + if (hlist_empty(hhd)) + return; + + /* + * Disable preemption for these calls to prevent a RCU grace + * period. This syncs the hash iteration and freeing of items + * on the hash. rcu_read_lock is too dangerous here. + */ + resched = ftrace_preempt_disable(); + hlist_for_each_entry_rcu(entry, n, hhd, node) { + if (entry->ip == ip) + entry->ops->func(ip, parent_ip, &entry->data); } + ftrace_preempt_enable(resched); +} - /* should not be called from interrupt context */ - spin_lock(&ftrace_lock); - if (enable) - ftrace_filtered = 1; - pg = ftrace_pages_start; - while (pg) { - for (i = 0; i < pg->index; i++) { - int matched = 0; - char *ptr; +static struct ftrace_ops trace_probe_ops __read_mostly = +{ + .func = function_trace_probe_call, +}; + +static int ftrace_probe_registered; + +static void __enable_ftrace_function_probe(void) +{ + int i; + + if (ftrace_probe_registered) + return; + + for (i = 0; i < FTRACE_FUNC_HASHSIZE; i++) { + struct hlist_head *hhd = &ftrace_func_hash[i]; + if (hhd->first) + break; + } + /* Nothing registered? */ + if (i == FTRACE_FUNC_HASHSIZE) + return; + + __register_ftrace_function(&trace_probe_ops); + ftrace_startup(0); + ftrace_probe_registered = 1; +} + +static void __disable_ftrace_function_probe(void) +{ + int i; + + if (!ftrace_probe_registered) + return; + + for (i = 0; i < FTRACE_FUNC_HASHSIZE; i++) { + struct hlist_head *hhd = &ftrace_func_hash[i]; + if (hhd->first) + return; + } + + /* no more funcs left */ + __unregister_ftrace_function(&trace_probe_ops); + ftrace_shutdown(0); + ftrace_probe_registered = 0; +} + + +static void ftrace_free_entry_rcu(struct rcu_head *rhp) +{ + struct ftrace_func_probe *entry = + container_of(rhp, struct ftrace_func_probe, rcu); + + if (entry->ops->free) + entry->ops->free(&entry->data); + kfree(entry); +} + + +int +register_ftrace_function_probe(char *glob, struct ftrace_probe_ops *ops, + void *data) +{ + struct ftrace_func_probe *entry; + struct ftrace_page *pg; + struct dyn_ftrace *rec; + int type, len, not; + unsigned long key; + int count = 0; + char *search; + + type = filter_parse_regex(glob, strlen(glob), &search, ¬); + len = strlen(search); + + /* we do not support '!' for function probes */ + if (WARN_ON(not)) + return -EINVAL; + + mutex_lock(&ftrace_lock); + do_for_each_ftrace_rec(pg, rec) { + + if (rec->flags & FTRACE_FL_FAILED) + continue; - rec = &pg->records[i]; - if (rec->flags & FTRACE_FL_FAILED) + if (!ftrace_match_record(rec, search, len, type)) + continue; + + entry = kmalloc(sizeof(*entry), GFP_KERNEL); + if (!entry) { + /* If we did not process any, then return error */ + if (!count) + count = -ENOMEM; + goto out_unlock; + } + + count++; + + entry->data = data; + + /* + * The caller might want to do something special + * for each function we find. We call the callback + * to give the caller an opportunity to do so. + */ + if (ops->callback) { + if (ops->callback(rec->ip, &entry->data) < 0) { + /* caller does not like this func */ + kfree(entry); continue; - kallsyms_lookup(rec->ip, NULL, NULL, NULL, str); - switch (type) { - case MATCH_FULL: - if (strcmp(str, buff) == 0) - matched = 1; - break; - case MATCH_FRONT_ONLY: - if (memcmp(str, buff, match) == 0) - matched = 1; - break; - case MATCH_MIDDLE_ONLY: - if (strstr(str, search)) - matched = 1; - break; - case MATCH_END_ONLY: - ptr = strstr(str, search); - if (ptr && (ptr[search_len] == 0)) - matched = 1; - break; } - if (matched) { - if (not) - rec->flags &= ~flag; - else - rec->flags |= flag; + } + + entry->ops = ops; + entry->ip = rec->ip; + + key = hash_long(entry->ip, FTRACE_HASH_BITS); + hlist_add_head_rcu(&entry->node, &ftrace_func_hash[key]); + + } while_for_each_ftrace_rec(); + __enable_ftrace_function_probe(); + + out_unlock: + mutex_unlock(&ftrace_lock); + + return count; +} + +enum { + PROBE_TEST_FUNC = 1, + PROBE_TEST_DATA = 2 +}; + +static void +__unregister_ftrace_function_probe(char *glob, struct ftrace_probe_ops *ops, + void *data, int flags) +{ + struct ftrace_func_probe *entry; + struct hlist_node *n, *tmp; + char str[KSYM_SYMBOL_LEN]; + int type = MATCH_FULL; + int i, len = 0; + char *search; + + if (glob && (strcmp(glob, "*") == 0 || !strlen(glob))) + glob = NULL; + else if (glob) { + int not; + + type = filter_parse_regex(glob, strlen(glob), &search, ¬); + len = strlen(search); + + /* we do not support '!' for function probes */ + if (WARN_ON(not)) + return; + } + + mutex_lock(&ftrace_lock); + for (i = 0; i < FTRACE_FUNC_HASHSIZE; i++) { + struct hlist_head *hhd = &ftrace_func_hash[i]; + + hlist_for_each_entry_safe(entry, n, tmp, hhd, node) { + + /* break up if statements for readability */ + if ((flags & PROBE_TEST_FUNC) && entry->ops != ops) + continue; + + if ((flags & PROBE_TEST_DATA) && entry->data != data) + continue; + + /* do this last, since it is the most expensive */ + if (glob) { + kallsyms_lookup(entry->ip, NULL, NULL, + NULL, str); + if (!ftrace_match(str, glob, len, type)) + continue; } + + hlist_del(&entry->node); + call_rcu(&entry->rcu, ftrace_free_entry_rcu); } - pg = pg->next; } - spin_unlock(&ftrace_lock); + __disable_ftrace_function_probe(); + mutex_unlock(&ftrace_lock); +} + +void +unregister_ftrace_function_probe(char *glob, struct ftrace_probe_ops *ops, + void *data) +{ + __unregister_ftrace_function_probe(glob, ops, data, + PROBE_TEST_FUNC | PROBE_TEST_DATA); +} + +void +unregister_ftrace_function_probe_func(char *glob, struct ftrace_probe_ops *ops) +{ + __unregister_ftrace_function_probe(glob, ops, NULL, PROBE_TEST_FUNC); +} + +void unregister_ftrace_function_probe_all(char *glob) +{ + __unregister_ftrace_function_probe(glob, NULL, NULL, 0); +} + +static LIST_HEAD(ftrace_commands); +static DEFINE_MUTEX(ftrace_cmd_mutex); + +int register_ftrace_command(struct ftrace_func_command *cmd) +{ + struct ftrace_func_command *p; + int ret = 0; + + mutex_lock(&ftrace_cmd_mutex); + list_for_each_entry(p, &ftrace_commands, list) { + if (strcmp(cmd->name, p->name) == 0) { + ret = -EBUSY; + goto out_unlock; + } + } + list_add(&cmd->list, &ftrace_commands); + out_unlock: + mutex_unlock(&ftrace_cmd_mutex); + + return ret; +} + +int unregister_ftrace_command(struct ftrace_func_command *cmd) +{ + struct ftrace_func_command *p, *n; + int ret = -ENODEV; + + mutex_lock(&ftrace_cmd_mutex); + list_for_each_entry_safe(p, n, &ftrace_commands, list) { + if (strcmp(cmd->name, p->name) == 0) { + ret = 0; + list_del_init(&p->list); + goto out_unlock; + } + } + out_unlock: + mutex_unlock(&ftrace_cmd_mutex); + + return ret; +} + +static int ftrace_process_regex(char *buff, int len, int enable) +{ + char *func, *command, *next = buff; + struct ftrace_func_command *p; + int ret = -EINVAL; + + func = strsep(&next, ":"); + + if (!next) { + ftrace_match_records(func, len, enable); + return 0; + } + + /* command found */ + + command = strsep(&next, ":"); + + mutex_lock(&ftrace_cmd_mutex); + list_for_each_entry(p, &ftrace_commands, list) { + if (strcmp(p->name, command) == 0) { + ret = p->func(func, command, next, enable); + goto out_unlock; + } + } + out_unlock: + mutex_unlock(&ftrace_cmd_mutex); + + return ret; } static ssize_t @@ -1125,11 +2177,10 @@ ftrace_regex_write(struct file *file, const char __user *ubuf, size_t cnt, loff_t *ppos, int enable) { struct ftrace_iterator *iter; - char ch; - size_t read = 0; - ssize_t ret; + struct trace_parser *parser; + ssize_t ret, read; - if (!cnt || cnt < 0) + if (!cnt) return 0; mutex_lock(&ftrace_regex_lock); @@ -1140,63 +2191,21 @@ ftrace_regex_write(struct file *file, const char __user *ubuf, } else iter = file->private_data; - if (!*ppos) { - iter->flags &= ~FTRACE_ITER_CONT; - iter->buffer_idx = 0; - } - - ret = get_user(ch, ubuf++); - if (ret) - goto out; - read++; - cnt--; - - if (!(iter->flags & ~FTRACE_ITER_CONT)) { - /* skip white space */ - while (cnt && isspace(ch)) { - ret = get_user(ch, ubuf++); - if (ret) - goto out; - read++; - cnt--; - } - - if (isspace(ch)) { - file->f_pos += read; - ret = read; - goto out; - } - - iter->buffer_idx = 0; - } + parser = &iter->parser; + read = trace_get_user(parser, ubuf, cnt, ppos); - while (cnt && !isspace(ch)) { - if (iter->buffer_idx < FTRACE_BUFF_MAX) - iter->buffer[iter->buffer_idx++] = ch; - else { - ret = -EINVAL; - goto out; - } - ret = get_user(ch, ubuf++); + if (read >= 0 && trace_parser_loaded(parser) && + !trace_parser_cont(parser)) { + ret = ftrace_process_regex(parser->buffer, + parser->idx, enable); if (ret) - goto out; - read++; - cnt--; - } - - if (isspace(ch)) { - iter->filtered++; - iter->buffer[iter->buffer_idx] = 0; - ftrace_match(iter->buffer, iter->buffer_idx, enable); - iter->buffer_idx = 0; - } else - iter->flags |= FTRACE_ITER_CONT; - + goto out_unlock; - file->f_pos += read; + trace_parser_clear(parser); + } ret = read; - out: +out_unlock: mutex_unlock(&ftrace_regex_lock); return ret; @@ -1226,7 +2235,7 @@ ftrace_set_regex(unsigned char *buf, int len, int reset, int enable) if (reset) ftrace_filter_reset(enable); if (buf) - ftrace_match(buf, len, enable); + ftrace_match_records(buf, len, enable); mutex_unlock(&ftrace_regex_lock); } @@ -1259,11 +2268,81 @@ void ftrace_set_notrace(unsigned char *buf, int len, int reset) ftrace_set_regex(buf, len, reset, 0); } +/* + * command line interface to allow users to set filters on boot up. + */ +#define FTRACE_FILTER_SIZE COMMAND_LINE_SIZE +static char ftrace_notrace_buf[FTRACE_FILTER_SIZE] __initdata; +static char ftrace_filter_buf[FTRACE_FILTER_SIZE] __initdata; + +static int __init set_ftrace_notrace(char *str) +{ + strncpy(ftrace_notrace_buf, str, FTRACE_FILTER_SIZE); + return 1; +} +__setup("ftrace_notrace=", set_ftrace_notrace); + +static int __init set_ftrace_filter(char *str) +{ + strncpy(ftrace_filter_buf, str, FTRACE_FILTER_SIZE); + return 1; +} +__setup("ftrace_filter=", set_ftrace_filter); + +#ifdef CONFIG_FUNCTION_GRAPH_TRACER +static char ftrace_graph_buf[FTRACE_FILTER_SIZE] __initdata; +static int __init set_graph_function(char *str) +{ + strlcpy(ftrace_graph_buf, str, FTRACE_FILTER_SIZE); + return 1; +} +__setup("ftrace_graph_filter=", set_graph_function); + +static void __init set_ftrace_early_graph(char *buf) +{ + int ret; + char *func; + + while (buf) { + func = strsep(&buf, ","); + /* we allow only one expression at a time */ + ret = ftrace_set_func(ftrace_graph_funcs, &ftrace_graph_count, + func); + if (ret) + printk(KERN_DEBUG "ftrace: function %s not " + "traceable\n", func); + } +} +#endif /* CONFIG_FUNCTION_GRAPH_TRACER */ + +static void __init set_ftrace_early_filter(char *buf, int enable) +{ + char *func; + + while (buf) { + func = strsep(&buf, ","); + ftrace_set_regex(func, strlen(func), 0, enable); + } +} + +static void __init set_ftrace_early_filters(void) +{ + if (ftrace_filter_buf[0]) + set_ftrace_early_filter(ftrace_filter_buf, 1); + if (ftrace_notrace_buf[0]) + set_ftrace_early_filter(ftrace_notrace_buf, 0); +#ifdef CONFIG_FUNCTION_GRAPH_TRACER + if (ftrace_graph_buf[0]) + set_ftrace_early_graph(ftrace_graph_buf); +#endif /* CONFIG_FUNCTION_GRAPH_TRACER */ +} + static int ftrace_regex_release(struct inode *inode, struct file *file, int enable) { struct seq_file *m = (struct seq_file *)file->private_data; struct ftrace_iterator *iter; + struct trace_parser *parser; mutex_lock(&ftrace_regex_lock); if (file->f_mode & FMODE_READ) { @@ -1273,20 +2352,20 @@ ftrace_regex_release(struct inode *inode, struct file *file, int enable) } else iter = file->private_data; - if (iter->buffer_idx) { - iter->filtered++; - iter->buffer[iter->buffer_idx] = 0; - ftrace_match(iter->buffer, iter->buffer_idx, enable); + parser = &iter->parser; + if (trace_parser_loaded(parser)) { + parser->buffer[parser->idx] = 0; + ftrace_match_records(parser->buffer, parser->idx, enable); } - mutex_lock(&ftrace_sysctl_lock); - mutex_lock(&ftrace_start_lock); + mutex_lock(&ftrace_lock); if (ftrace_start_up && ftrace_enabled) ftrace_run_update_code(FTRACE_ENABLE_CALLS); - mutex_unlock(&ftrace_start_lock); - mutex_unlock(&ftrace_sysctl_lock); + mutex_unlock(&ftrace_lock); + trace_parser_put(parser); kfree(iter); + mutex_unlock(&ftrace_regex_lock); return 0; } @@ -1303,31 +2382,31 @@ ftrace_notrace_release(struct inode *inode, struct file *file) return ftrace_regex_release(inode, file, 0); } -static struct file_operations ftrace_avail_fops = { +static const struct file_operations ftrace_avail_fops = { .open = ftrace_avail_open, .read = seq_read, .llseek = seq_lseek, - .release = ftrace_avail_release, + .release = seq_release_private, }; -static struct file_operations ftrace_failures_fops = { +static const struct file_operations ftrace_failures_fops = { .open = ftrace_failures_open, .read = seq_read, .llseek = seq_lseek, - .release = ftrace_avail_release, + .release = seq_release_private, }; -static struct file_operations ftrace_filter_fops = { +static const struct file_operations ftrace_filter_fops = { .open = ftrace_filter_open, - .read = ftrace_regex_read, + .read = seq_read, .write = ftrace_filter_write, .llseek = ftrace_regex_lseek, .release = ftrace_filter_release, }; -static struct file_operations ftrace_notrace_fops = { +static const struct file_operations ftrace_notrace_fops = { .open = ftrace_notrace_open, - .read = ftrace_regex_read, + .read = seq_read, .write = ftrace_notrace_write, .llseek = ftrace_regex_lseek, .release = ftrace_notrace_release, @@ -1341,28 +2420,29 @@ int ftrace_graph_count; unsigned long ftrace_graph_funcs[FTRACE_GRAPH_MAX_FUNCS] __read_mostly; static void * -g_next(struct seq_file *m, void *v, loff_t *pos) +__g_next(struct seq_file *m, loff_t *pos) { - unsigned long *array = m->private; - int index = *pos; - - (*pos)++; - - if (index >= ftrace_graph_count) + if (*pos >= ftrace_graph_count) return NULL; + return &ftrace_graph_funcs[*pos]; +} - return &array[index]; +static void * +g_next(struct seq_file *m, void *v, loff_t *pos) +{ + (*pos)++; + return __g_next(m, pos); } static void *g_start(struct seq_file *m, loff_t *pos) { - void *p = NULL; - mutex_lock(&graph_lock); - p = g_next(m, p, pos); + /* Nothing, tell g_show to print all functions are enabled */ + if (!ftrace_graph_count && !*pos) + return (void *)1; - return p; + return __g_next(m, pos); } static void g_stop(struct seq_file *m, void *p) @@ -1373,19 +2453,21 @@ static void g_stop(struct seq_file *m, void *p) static int g_show(struct seq_file *m, void *v) { unsigned long *ptr = v; - char str[KSYM_SYMBOL_LEN]; if (!ptr) return 0; - kallsyms_lookup(*ptr, NULL, NULL, NULL, str); + if (ptr == (unsigned long *)1) { + seq_printf(m, "#### all functions enabled ####\n"); + return 0; + } - seq_printf(m, "%s\n", str); + seq_printf(m, "%ps\n", (void *)*ptr); return 0; } -static struct seq_operations ftrace_graph_seq_ops = { +static const struct seq_operations ftrace_graph_seq_ops = { .start = g_start, .next = g_next, .stop = g_stop, @@ -1402,71 +2484,73 @@ ftrace_graph_open(struct inode *inode, struct file *file) mutex_lock(&graph_lock); if ((file->f_mode & FMODE_WRITE) && - !(file->f_flags & O_APPEND)) { + (file->f_flags & O_TRUNC)) { ftrace_graph_count = 0; memset(ftrace_graph_funcs, 0, sizeof(ftrace_graph_funcs)); } + mutex_unlock(&graph_lock); - if (file->f_mode & FMODE_READ) { + if (file->f_mode & FMODE_READ) ret = seq_open(file, &ftrace_graph_seq_ops); - if (!ret) { - struct seq_file *m = file->private_data; - m->private = ftrace_graph_funcs; - } - } else - file->private_data = ftrace_graph_funcs; - mutex_unlock(&graph_lock); return ret; } -static ssize_t -ftrace_graph_read(struct file *file, char __user *ubuf, - size_t cnt, loff_t *ppos) +static int +ftrace_graph_release(struct inode *inode, struct file *file) { if (file->f_mode & FMODE_READ) - return seq_read(file, ubuf, cnt, ppos); - else - return -EPERM; + seq_release(inode, file); + return 0; } static int -ftrace_set_func(unsigned long *array, int idx, char *buffer) +ftrace_set_func(unsigned long *array, int *idx, char *buffer) { - char str[KSYM_SYMBOL_LEN]; struct dyn_ftrace *rec; struct ftrace_page *pg; + int search_len; int found = 0; - int i, j; + int type, not; + char *search; + bool exists; + int i; if (ftrace_disabled) return -ENODEV; - /* should not be called from interrupt context */ - spin_lock(&ftrace_lock); + /* decode regex */ + type = filter_parse_regex(buffer, strlen(buffer), &search, ¬); + if (not) + return -EINVAL; - for (pg = ftrace_pages_start; pg; pg = pg->next) { - for (i = 0; i < pg->index; i++) { - rec = &pg->records[i]; + search_len = strlen(search); - if (rec->flags & (FTRACE_FL_FAILED | FTRACE_FL_FREE)) - continue; + mutex_lock(&ftrace_lock); + do_for_each_ftrace_rec(pg, rec) { + + if (*idx >= FTRACE_GRAPH_MAX_FUNCS) + break; - kallsyms_lookup(rec->ip, NULL, NULL, NULL, str); - if (strcmp(str, buffer) == 0) { + if (rec->flags & (FTRACE_FL_FAILED | FTRACE_FL_FREE)) + continue; + + if (ftrace_match_record(rec, search, search_len, type)) { + /* ensure it is not already in the array */ + exists = false; + for (i = 0; i < *idx; i++) + if (array[i] == rec->ip) { + exists = true; + break; + } + if (!exists) { + array[(*idx)++] = rec->ip; found = 1; - for (j = 0; j < idx; j++) - if (array[j] == rec->ip) { - found = 0; - break; - } - if (found) - array[idx] = rec->ip; - break; } } - } - spin_unlock(&ftrace_lock); + } while_for_each_ftrace_rec(); + + mutex_unlock(&ftrace_lock); return found ? 0 : -EINVAL; } @@ -1475,12 +2559,8 @@ static ssize_t ftrace_graph_write(struct file *file, const char __user *ubuf, size_t cnt, loff_t *ppos) { - unsigned char buffer[FTRACE_BUFF_MAX+1]; - unsigned long *array; - size_t read = 0; - ssize_t ret; - int index = 0; - char ch; + struct trace_parser parser; + ssize_t read, ret; if (!cnt || cnt < 0) return 0; @@ -1489,114 +2569,69 @@ ftrace_graph_write(struct file *file, const char __user *ubuf, if (ftrace_graph_count >= FTRACE_GRAPH_MAX_FUNCS) { ret = -EBUSY; - goto out; + goto out_unlock; } - if (file->f_mode & FMODE_READ) { - struct seq_file *m = file->private_data; - array = m->private; - } else - array = file->private_data; - - ret = get_user(ch, ubuf++); - if (ret) - goto out; - read++; - cnt--; - - /* skip white space */ - while (cnt && isspace(ch)) { - ret = get_user(ch, ubuf++); - if (ret) - goto out; - read++; - cnt--; + if (trace_parser_get_init(&parser, FTRACE_BUFF_MAX)) { + ret = -ENOMEM; + goto out_unlock; } - if (isspace(ch)) { - *ppos += read; - ret = read; - goto out; - } + read = trace_get_user(&parser, ubuf, cnt, ppos); - while (cnt && !isspace(ch)) { - if (index < FTRACE_BUFF_MAX) - buffer[index++] = ch; - else { - ret = -EINVAL; - goto out; - } - ret = get_user(ch, ubuf++); + if (read >= 0 && trace_parser_loaded((&parser))) { + parser.buffer[parser.idx] = 0; + + /* we allow only one expression at a time */ + ret = ftrace_set_func(ftrace_graph_funcs, &ftrace_graph_count, + parser.buffer); if (ret) - goto out; - read++; - cnt--; + goto out_free; } - buffer[index] = 0; - - /* we allow only one at a time */ - ret = ftrace_set_func(array, ftrace_graph_count, buffer); - if (ret) - goto out; - - ftrace_graph_count++; - - file->f_pos += read; ret = read; - out: + +out_free: + trace_parser_put(&parser); +out_unlock: mutex_unlock(&graph_lock); return ret; } static const struct file_operations ftrace_graph_fops = { - .open = ftrace_graph_open, - .read = ftrace_graph_read, - .write = ftrace_graph_write, + .open = ftrace_graph_open, + .read = seq_read, + .write = ftrace_graph_write, + .release = ftrace_graph_release, }; #endif /* CONFIG_FUNCTION_GRAPH_TRACER */ static __init int ftrace_init_dyn_debugfs(struct dentry *d_tracer) { - struct dentry *entry; - entry = debugfs_create_file("available_filter_functions", 0444, - d_tracer, NULL, &ftrace_avail_fops); - if (!entry) - pr_warning("Could not create debugfs " - "'available_filter_functions' entry\n"); + trace_create_file("available_filter_functions", 0444, + d_tracer, NULL, &ftrace_avail_fops); - entry = debugfs_create_file("failures", 0444, - d_tracer, NULL, &ftrace_failures_fops); - if (!entry) - pr_warning("Could not create debugfs 'failures' entry\n"); + trace_create_file("failures", 0444, + d_tracer, NULL, &ftrace_failures_fops); - entry = debugfs_create_file("set_ftrace_filter", 0644, d_tracer, - NULL, &ftrace_filter_fops); - if (!entry) - pr_warning("Could not create debugfs " - "'set_ftrace_filter' entry\n"); + trace_create_file("set_ftrace_filter", 0644, d_tracer, + NULL, &ftrace_filter_fops); - entry = debugfs_create_file("set_ftrace_notrace", 0644, d_tracer, + trace_create_file("set_ftrace_notrace", 0644, d_tracer, NULL, &ftrace_notrace_fops); - if (!entry) - pr_warning("Could not create debugfs " - "'set_ftrace_notrace' entry\n"); #ifdef CONFIG_FUNCTION_GRAPH_TRACER - entry = debugfs_create_file("set_graph_function", 0444, d_tracer, + trace_create_file("set_graph_function", 0444, d_tracer, NULL, &ftrace_graph_fops); - if (!entry) - pr_warning("Could not create debugfs " - "'set_graph_function' entry\n"); #endif /* CONFIG_FUNCTION_GRAPH_TRACER */ return 0; } -static int ftrace_convert_nops(struct module *mod, +static int ftrace_process_locs(struct module *mod, unsigned long *start, unsigned long *end) { @@ -1604,7 +2639,7 @@ static int ftrace_convert_nops(struct module *mod, unsigned long addr; unsigned long flags; - mutex_lock(&ftrace_start_lock); + mutex_lock(&ftrace_lock); p = start; while (p < end) { addr = ftrace_call_adjust(*p++); @@ -1623,18 +2658,72 @@ static int ftrace_convert_nops(struct module *mod, local_irq_save(flags); ftrace_update_code(mod); local_irq_restore(flags); - mutex_unlock(&ftrace_start_lock); + mutex_unlock(&ftrace_lock); return 0; } -void ftrace_init_module(struct module *mod, - unsigned long *start, unsigned long *end) +#ifdef CONFIG_MODULES +void ftrace_release_mod(struct module *mod) +{ + struct dyn_ftrace *rec; + struct ftrace_page *pg; + + if (ftrace_disabled) + return; + + mutex_lock(&ftrace_lock); + do_for_each_ftrace_rec(pg, rec) { + if (within_module_core(rec->ip, mod)) { + /* + * rec->ip is changed in ftrace_free_rec() + * It should not between s and e if record was freed. + */ + FTRACE_WARN_ON(rec->flags & FTRACE_FL_FREE); + ftrace_free_rec(rec); + } + } while_for_each_ftrace_rec(); + mutex_unlock(&ftrace_lock); +} + +static void ftrace_init_module(struct module *mod, + unsigned long *start, unsigned long *end) { if (ftrace_disabled || start == end) return; - ftrace_convert_nops(mod, start, end); + ftrace_process_locs(mod, start, end); +} + +static int ftrace_module_notify(struct notifier_block *self, + unsigned long val, void *data) +{ + struct module *mod = data; + + switch (val) { + case MODULE_STATE_COMING: + ftrace_init_module(mod, mod->ftrace_callsites, + mod->ftrace_callsites + + mod->num_ftrace_callsites); + break; + case MODULE_STATE_GOING: + ftrace_release_mod(mod); + break; + } + + return 0; +} +#else +static int ftrace_module_notify(struct notifier_block *self, + unsigned long val, void *data) +{ + return 0; } +#endif /* CONFIG_MODULES */ + +struct notifier_block ftrace_module_nb = { + .notifier_call = ftrace_module_notify, + .priority = 0, +}; extern unsigned long __start_mcount_loc[]; extern unsigned long __stop_mcount_loc[]; @@ -1663,10 +2752,16 @@ void __init ftrace_init(void) last_ftrace_enabled = ftrace_enabled = 1; - ret = ftrace_convert_nops(NULL, + ret = ftrace_process_locs(NULL, __start_mcount_loc, __stop_mcount_loc); + ret = register_module_notifier(&ftrace_module_nb); + if (ret) + pr_warning("Failed to register trace ftrace module notifier\n"); + + set_ftrace_early_filters(); + return; failed: ftrace_disabled = 1; @@ -1690,23 +2785,6 @@ static inline void ftrace_startup_enable(int command) { } # define ftrace_shutdown_sysctl() do { } while (0) #endif /* CONFIG_DYNAMIC_FTRACE */ -static ssize_t -ftrace_pid_read(struct file *file, char __user *ubuf, - size_t cnt, loff_t *ppos) -{ - char buf[64]; - int r; - - if (ftrace_pid_trace == ftrace_swapper_pid) - r = sprintf(buf, "swapper tasks\n"); - else if (ftrace_pid_trace) - r = sprintf(buf, "%u\n", pid_nr(ftrace_pid_trace)); - else - r = sprintf(buf, "no pid\n"); - - return simple_read_from_buffer(ubuf, cnt, ppos, buf, r); -} - static void clear_ftrace_swapper(void) { struct task_struct *p; @@ -1757,14 +2835,12 @@ static void set_ftrace_pid(struct pid *pid) rcu_read_unlock(); } -static void clear_ftrace_pid_task(struct pid **pid) +static void clear_ftrace_pid_task(struct pid *pid) { - if (*pid == ftrace_swapper_pid) + if (pid == ftrace_swapper_pid) clear_ftrace_swapper(); else - clear_ftrace_pid(*pid); - - *pid = NULL; + clear_ftrace_pid(pid); } static void set_ftrace_pid_task(struct pid *pid) @@ -1775,12 +2851,141 @@ static void set_ftrace_pid_task(struct pid *pid) set_ftrace_pid(pid); } +static int ftrace_pid_add(int p) +{ + struct pid *pid; + struct ftrace_pid *fpid; + int ret = -EINVAL; + + mutex_lock(&ftrace_lock); + + if (!p) + pid = ftrace_swapper_pid; + else + pid = find_get_pid(p); + + if (!pid) + goto out; + + ret = 0; + + list_for_each_entry(fpid, &ftrace_pids, list) + if (fpid->pid == pid) + goto out_put; + + ret = -ENOMEM; + + fpid = kmalloc(sizeof(*fpid), GFP_KERNEL); + if (!fpid) + goto out_put; + + list_add(&fpid->list, &ftrace_pids); + fpid->pid = pid; + + set_ftrace_pid_task(pid); + + ftrace_update_pid_func(); + ftrace_startup_enable(0); + + mutex_unlock(&ftrace_lock); + return 0; + +out_put: + if (pid != ftrace_swapper_pid) + put_pid(pid); + +out: + mutex_unlock(&ftrace_lock); + return ret; +} + +static void ftrace_pid_reset(void) +{ + struct ftrace_pid *fpid, *safe; + + mutex_lock(&ftrace_lock); + list_for_each_entry_safe(fpid, safe, &ftrace_pids, list) { + struct pid *pid = fpid->pid; + + clear_ftrace_pid_task(pid); + + list_del(&fpid->list); + kfree(fpid); + } + + ftrace_update_pid_func(); + ftrace_startup_enable(0); + + mutex_unlock(&ftrace_lock); +} + +static void *fpid_start(struct seq_file *m, loff_t *pos) +{ + mutex_lock(&ftrace_lock); + + if (list_empty(&ftrace_pids) && (!*pos)) + return (void *) 1; + + return seq_list_start(&ftrace_pids, *pos); +} + +static void *fpid_next(struct seq_file *m, void *v, loff_t *pos) +{ + if (v == (void *)1) + return NULL; + + return seq_list_next(v, &ftrace_pids, pos); +} + +static void fpid_stop(struct seq_file *m, void *p) +{ + mutex_unlock(&ftrace_lock); +} + +static int fpid_show(struct seq_file *m, void *v) +{ + const struct ftrace_pid *fpid = list_entry(v, struct ftrace_pid, list); + + if (v == (void *)1) { + seq_printf(m, "no pid\n"); + return 0; + } + + if (fpid->pid == ftrace_swapper_pid) + seq_printf(m, "swapper tasks\n"); + else + seq_printf(m, "%u\n", pid_vnr(fpid->pid)); + + return 0; +} + +static const struct seq_operations ftrace_pid_sops = { + .start = fpid_start, + .next = fpid_next, + .stop = fpid_stop, + .show = fpid_show, +}; + +static int +ftrace_pid_open(struct inode *inode, struct file *file) +{ + int ret = 0; + + if ((file->f_mode & FMODE_WRITE) && + (file->f_flags & O_TRUNC)) + ftrace_pid_reset(); + + if (file->f_mode & FMODE_READ) + ret = seq_open(file, &ftrace_pid_sops); + + return ret; +} + static ssize_t ftrace_pid_write(struct file *filp, const char __user *ubuf, size_t cnt, loff_t *ppos) { - struct pid *pid; - char buf[64]; + char buf[64], *tmp; long val; int ret; @@ -1792,63 +2997,43 @@ ftrace_pid_write(struct file *filp, const char __user *ubuf, buf[cnt] = 0; - ret = strict_strtol(buf, 10, &val); + /* + * Allow "echo > set_ftrace_pid" or "echo -n '' > set_ftrace_pid" + * to clean the filter quietly. + */ + tmp = strstrip(buf); + if (strlen(tmp) == 0) + return 1; + + ret = strict_strtol(tmp, 10, &val); if (ret < 0) return ret; - mutex_lock(&ftrace_start_lock); - if (val < 0) { - /* disable pid tracing */ - if (!ftrace_pid_trace) - goto out; - - clear_ftrace_pid_task(&ftrace_pid_trace); - - } else { - /* swapper task is special */ - if (!val) { - pid = ftrace_swapper_pid; - if (pid == ftrace_pid_trace) - goto out; - } else { - pid = find_get_pid(val); + ret = ftrace_pid_add(val); - if (pid == ftrace_pid_trace) { - put_pid(pid); - goto out; - } - } - - if (ftrace_pid_trace) - clear_ftrace_pid_task(&ftrace_pid_trace); - - if (!pid) - goto out; - - ftrace_pid_trace = pid; - - set_ftrace_pid_task(ftrace_pid_trace); - } - - /* update the function call */ - ftrace_update_pid_func(); - ftrace_startup_enable(0); + return ret ? ret : cnt; +} - out: - mutex_unlock(&ftrace_start_lock); +static int +ftrace_pid_release(struct inode *inode, struct file *file) +{ + if (file->f_mode & FMODE_READ) + seq_release(inode, file); - return cnt; + return 0; } -static struct file_operations ftrace_pid_fops = { - .read = ftrace_pid_read, - .write = ftrace_pid_write, +static const struct file_operations ftrace_pid_fops = { + .open = ftrace_pid_open, + .write = ftrace_pid_write, + .read = seq_read, + .llseek = seq_lseek, + .release = ftrace_pid_release, }; static __init int ftrace_init_debugfs(void) { struct dentry *d_tracer; - struct dentry *entry; d_tracer = tracing_init_dentry(); if (!d_tracer) @@ -1856,14 +3041,13 @@ static __init int ftrace_init_debugfs(void) ftrace_init_dyn_debugfs(d_tracer); - entry = debugfs_create_file("set_ftrace_pid", 0644, d_tracer, - NULL, &ftrace_pid_fops); - if (!entry) - pr_warning("Could not create debugfs " - "'set_ftrace_pid' entry\n"); + trace_create_file("set_ftrace_pid", 0644, d_tracer, + NULL, &ftrace_pid_fops); + + ftrace_profile_debugfs(d_tracer); + return 0; } - fs_initcall(ftrace_init_debugfs); /** @@ -1898,17 +3082,17 @@ int register_ftrace_function(struct ftrace_ops *ops) if (unlikely(ftrace_disabled)) return -1; - mutex_lock(&ftrace_sysctl_lock); + mutex_lock(&ftrace_lock); ret = __register_ftrace_function(ops); ftrace_startup(0); - mutex_unlock(&ftrace_sysctl_lock); + mutex_unlock(&ftrace_lock); return ret; } /** - * unregister_ftrace_function - unresgister a function for profiling. + * unregister_ftrace_function - unregister a function for profiling. * @ops - ops structure that holds the function to unregister * * Unregister a function that was added to be called by ftrace profiling. @@ -1917,17 +3101,17 @@ int unregister_ftrace_function(struct ftrace_ops *ops) { int ret; - mutex_lock(&ftrace_sysctl_lock); + mutex_lock(&ftrace_lock); ret = __unregister_ftrace_function(ops); ftrace_shutdown(0); - mutex_unlock(&ftrace_sysctl_lock); + mutex_unlock(&ftrace_lock); return ret; } int ftrace_enable_sysctl(struct ctl_table *table, int write, - struct file *file, void __user *buffer, size_t *lenp, + void __user *buffer, size_t *lenp, loff_t *ppos) { int ret; @@ -1935,14 +3119,14 @@ ftrace_enable_sysctl(struct ctl_table *table, int write, if (unlikely(ftrace_disabled)) return -ENODEV; - mutex_lock(&ftrace_sysctl_lock); + mutex_lock(&ftrace_lock); - ret = proc_dointvec(table, write, file, buffer, lenp, ppos); + ret = proc_dointvec(table, write, buffer, lenp, ppos); - if (ret || !write || (last_ftrace_enabled == ftrace_enabled)) + if (ret || !write || (last_ftrace_enabled == !!ftrace_enabled)) goto out; - last_ftrace_enabled = ftrace_enabled; + last_ftrace_enabled = !!ftrace_enabled; if (ftrace_enabled) { @@ -1964,13 +3148,13 @@ ftrace_enable_sysctl(struct ctl_table *table, int write, } out: - mutex_unlock(&ftrace_sysctl_lock); + mutex_unlock(&ftrace_lock); return ret; } #ifdef CONFIG_FUNCTION_GRAPH_TRACER -static atomic_t ftrace_graph_active; +static int ftrace_graph_active; static struct notifier_block ftrace_suspend_notifier; int ftrace_graph_entry_stub(struct ftrace_graph_ent *trace) @@ -2012,12 +3196,12 @@ static int alloc_retstack_tasklist(struct ftrace_ret_stack **ret_stack_list) } if (t->ret_stack == NULL) { - t->curr_ret_stack = -1; - /* Make sure IRQs see the -1 first: */ - barrier(); - t->ret_stack = ret_stack_list[start++]; atomic_set(&t->tracing_graph_pause, 0); atomic_set(&t->trace_overrun, 0); + t->curr_ret_stack = -1; + /* Make sure the tasks see the -1 first: */ + smp_wmb(); + t->ret_stack = ret_stack_list[start++]; } } while_each_thread(g, t); @@ -2029,6 +3213,38 @@ free: return ret; } +static void +ftrace_graph_probe_sched_switch(struct rq *__rq, struct task_struct *prev, + struct task_struct *next) +{ + unsigned long long timestamp; + int index; + + /* + * Does the user want to count the time a function was asleep. + * If so, do not update the time stamps. + */ + if (trace_flags & TRACE_ITER_SLEEP_TIME) + return; + + timestamp = trace_clock_local(); + + prev->ftrace_timestamp = timestamp; + + /* only process tasks that we timestamped */ + if (!next->ftrace_timestamp) + return; + + /* + * Update all the counters in next to make up for the + * time next was sleeping. + */ + timestamp -= next->ftrace_timestamp; + + for (index = next->curr_ret_stack; index >= 0; index--) + next->ret_stack[index].calltime += timestamp; +} + /* Allocate a return stack for each task */ static int start_graph_tracing(void) { @@ -2043,13 +3259,22 @@ static int start_graph_tracing(void) return -ENOMEM; /* The cpu_boot init_task->ret_stack will never be freed */ - for_each_online_cpu(cpu) - ftrace_graph_init_task(idle_task(cpu)); + for_each_online_cpu(cpu) { + if (!idle_task(cpu)->ret_stack) + ftrace_graph_init_task(idle_task(cpu)); + } do { ret = alloc_retstack_tasklist(ret_stack_list); } while (ret == -EAGAIN); + if (!ret) { + ret = register_trace_sched_switch(ftrace_graph_probe_sched_switch); + if (ret) + pr_info("ftrace_graph: Couldn't activate tracepoint" + " probe to kernel_sched_switch\n"); + } + kfree(ret_stack_list); return ret; } @@ -2080,15 +3305,21 @@ int register_ftrace_graph(trace_func_graph_ret_t retfunc, { int ret = 0; - mutex_lock(&ftrace_sysctl_lock); + mutex_lock(&ftrace_lock); + + /* we currently allow only one tracer registered at a time */ + if (ftrace_graph_active) { + ret = -EBUSY; + goto out; + } ftrace_suspend_notifier.notifier_call = ftrace_suspend_notifier_call; register_pm_notifier(&ftrace_suspend_notifier); - atomic_inc(&ftrace_graph_active); + ftrace_graph_active++; ret = start_graph_tracing(); if (ret) { - atomic_dec(&ftrace_graph_active); + ftrace_graph_active--; goto out; } @@ -2098,37 +3329,50 @@ int register_ftrace_graph(trace_func_graph_ret_t retfunc, ftrace_startup(FTRACE_START_FUNC_RET); out: - mutex_unlock(&ftrace_sysctl_lock); + mutex_unlock(&ftrace_lock); return ret; } void unregister_ftrace_graph(void) { - mutex_lock(&ftrace_sysctl_lock); + mutex_lock(&ftrace_lock); + + if (unlikely(!ftrace_graph_active)) + goto out; - atomic_dec(&ftrace_graph_active); + ftrace_graph_active--; + unregister_trace_sched_switch(ftrace_graph_probe_sched_switch); ftrace_graph_return = (trace_func_graph_ret_t)ftrace_stub; ftrace_graph_entry = ftrace_graph_entry_stub; ftrace_shutdown(FTRACE_STOP_FUNC_RET); unregister_pm_notifier(&ftrace_suspend_notifier); - mutex_unlock(&ftrace_sysctl_lock); + out: + mutex_unlock(&ftrace_lock); } /* Allocate a return stack for newly created task */ void ftrace_graph_init_task(struct task_struct *t) { - if (atomic_read(&ftrace_graph_active)) { - t->ret_stack = kmalloc(FTRACE_RETFUNC_DEPTH + /* Make sure we do not use the parent ret_stack */ + t->ret_stack = NULL; + + if (ftrace_graph_active) { + struct ftrace_ret_stack *ret_stack; + + ret_stack = kmalloc(FTRACE_RETFUNC_DEPTH * sizeof(struct ftrace_ret_stack), GFP_KERNEL); - if (!t->ret_stack) + if (!ret_stack) return; t->curr_ret_stack = -1; atomic_set(&t->tracing_graph_pause, 0); atomic_set(&t->trace_overrun, 0); - } else - t->ret_stack = NULL; + t->ftrace_timestamp = 0; + /* make curr_ret_stack visable before we add the ret_stack */ + smp_wmb(); + t->ret_stack = ret_stack; + } } void ftrace_graph_exit_task(struct task_struct *t) @@ -2147,4 +3391,3 @@ void ftrace_graph_stop(void) ftrace_stop(); } #endif - diff --git a/kernel/trace/kmemtrace.c b/kernel/trace/kmemtrace.c new file mode 100644 index 00000000000..a91da69f153 --- /dev/null +++ b/kernel/trace/kmemtrace.c @@ -0,0 +1,511 @@ +/* + * Memory allocator tracing + * + * Copyright (C) 2008 Eduard - Gabriel Munteanu + * Copyright (C) 2008 Pekka Enberg <penberg@cs.helsinki.fi> + * Copyright (C) 2008 Frederic Weisbecker <fweisbec@gmail.com> + */ + +#include <linux/tracepoint.h> +#include <linux/seq_file.h> +#include <linux/debugfs.h> +#include <linux/dcache.h> +#include <linux/fs.h> + +#include <linux/kmemtrace.h> + +#include "trace_output.h" +#include "trace.h" + +/* Select an alternative, minimalistic output than the original one */ +#define TRACE_KMEM_OPT_MINIMAL 0x1 + +static struct tracer_opt kmem_opts[] = { + /* Default disable the minimalistic output */ + { TRACER_OPT(kmem_minimalistic, TRACE_KMEM_OPT_MINIMAL) }, + { } +}; + +static struct tracer_flags kmem_tracer_flags = { + .val = 0, + .opts = kmem_opts +}; + +static struct trace_array *kmemtrace_array; + +/* Trace allocations */ +static inline void kmemtrace_alloc(enum kmemtrace_type_id type_id, + unsigned long call_site, + const void *ptr, + size_t bytes_req, + size_t bytes_alloc, + gfp_t gfp_flags, + int node) +{ + struct ftrace_event_call *call = &event_kmem_alloc; + struct trace_array *tr = kmemtrace_array; + struct kmemtrace_alloc_entry *entry; + struct ring_buffer_event *event; + + event = ring_buffer_lock_reserve(tr->buffer, sizeof(*entry)); + if (!event) + return; + + entry = ring_buffer_event_data(event); + tracing_generic_entry_update(&entry->ent, 0, 0); + + entry->ent.type = TRACE_KMEM_ALLOC; + entry->type_id = type_id; + entry->call_site = call_site; + entry->ptr = ptr; + entry->bytes_req = bytes_req; + entry->bytes_alloc = bytes_alloc; + entry->gfp_flags = gfp_flags; + entry->node = node; + + if (!filter_check_discard(call, entry, tr->buffer, event)) + ring_buffer_unlock_commit(tr->buffer, event); + + trace_wake_up(); +} + +static inline void kmemtrace_free(enum kmemtrace_type_id type_id, + unsigned long call_site, + const void *ptr) +{ + struct ftrace_event_call *call = &event_kmem_free; + struct trace_array *tr = kmemtrace_array; + struct kmemtrace_free_entry *entry; + struct ring_buffer_event *event; + + event = ring_buffer_lock_reserve(tr->buffer, sizeof(*entry)); + if (!event) + return; + entry = ring_buffer_event_data(event); + tracing_generic_entry_update(&entry->ent, 0, 0); + + entry->ent.type = TRACE_KMEM_FREE; + entry->type_id = type_id; + entry->call_site = call_site; + entry->ptr = ptr; + + if (!filter_check_discard(call, entry, tr->buffer, event)) + ring_buffer_unlock_commit(tr->buffer, event); + + trace_wake_up(); +} + +static void kmemtrace_kmalloc(unsigned long call_site, + const void *ptr, + size_t bytes_req, + size_t bytes_alloc, + gfp_t gfp_flags) +{ + kmemtrace_alloc(KMEMTRACE_TYPE_KMALLOC, call_site, ptr, + bytes_req, bytes_alloc, gfp_flags, -1); +} + +static void kmemtrace_kmem_cache_alloc(unsigned long call_site, + const void *ptr, + size_t bytes_req, + size_t bytes_alloc, + gfp_t gfp_flags) +{ + kmemtrace_alloc(KMEMTRACE_TYPE_CACHE, call_site, ptr, + bytes_req, bytes_alloc, gfp_flags, -1); +} + +static void kmemtrace_kmalloc_node(unsigned long call_site, + const void *ptr, + size_t bytes_req, + size_t bytes_alloc, + gfp_t gfp_flags, + int node) +{ + kmemtrace_alloc(KMEMTRACE_TYPE_KMALLOC, call_site, ptr, + bytes_req, bytes_alloc, gfp_flags, node); +} + +static void kmemtrace_kmem_cache_alloc_node(unsigned long call_site, + const void *ptr, + size_t bytes_req, + size_t bytes_alloc, + gfp_t gfp_flags, + int node) +{ + kmemtrace_alloc(KMEMTRACE_TYPE_CACHE, call_site, ptr, + bytes_req, bytes_alloc, gfp_flags, node); +} + +static void kmemtrace_kfree(unsigned long call_site, const void *ptr) +{ + kmemtrace_free(KMEMTRACE_TYPE_KMALLOC, call_site, ptr); +} + +static void kmemtrace_kmem_cache_free(unsigned long call_site, const void *ptr) +{ + kmemtrace_free(KMEMTRACE_TYPE_CACHE, call_site, ptr); +} + +static int kmemtrace_start_probes(void) +{ + int err; + + err = register_trace_kmalloc(kmemtrace_kmalloc); + if (err) + return err; + err = register_trace_kmem_cache_alloc(kmemtrace_kmem_cache_alloc); + if (err) + return err; + err = register_trace_kmalloc_node(kmemtrace_kmalloc_node); + if (err) + return err; + err = register_trace_kmem_cache_alloc_node(kmemtrace_kmem_cache_alloc_node); + if (err) + return err; + err = register_trace_kfree(kmemtrace_kfree); + if (err) + return err; + err = register_trace_kmem_cache_free(kmemtrace_kmem_cache_free); + + return err; +} + +static void kmemtrace_stop_probes(void) +{ + unregister_trace_kmalloc(kmemtrace_kmalloc); + unregister_trace_kmem_cache_alloc(kmemtrace_kmem_cache_alloc); + unregister_trace_kmalloc_node(kmemtrace_kmalloc_node); + unregister_trace_kmem_cache_alloc_node(kmemtrace_kmem_cache_alloc_node); + unregister_trace_kfree(kmemtrace_kfree); + unregister_trace_kmem_cache_free(kmemtrace_kmem_cache_free); +} + +static int kmem_trace_init(struct trace_array *tr) +{ + kmemtrace_array = tr; + + tracing_reset_online_cpus(tr); + + kmemtrace_start_probes(); + + return 0; +} + +static void kmem_trace_reset(struct trace_array *tr) +{ + kmemtrace_stop_probes(); +} + +static void kmemtrace_headers(struct seq_file *s) +{ + /* Don't need headers for the original kmemtrace output */ + if (!(kmem_tracer_flags.val & TRACE_KMEM_OPT_MINIMAL)) + return; + + seq_printf(s, "#\n"); + seq_printf(s, "# ALLOC TYPE REQ GIVEN FLAGS " + " POINTER NODE CALLER\n"); + seq_printf(s, "# FREE | | | | " + " | | | |\n"); + seq_printf(s, "# |\n\n"); +} + +/* + * The following functions give the original output from kmemtrace, + * plus the origin CPU, since reordering occurs in-kernel now. + */ + +#define KMEMTRACE_USER_ALLOC 0 +#define KMEMTRACE_USER_FREE 1 + +struct kmemtrace_user_event { + u8 event_id; + u8 type_id; + u16 event_size; + u32 cpu; + u64 timestamp; + unsigned long call_site; + unsigned long ptr; +}; + +struct kmemtrace_user_event_alloc { + size_t bytes_req; + size_t bytes_alloc; + unsigned gfp_flags; + int node; +}; + +static enum print_line_t +kmemtrace_print_alloc(struct trace_iterator *iter, int flags) +{ + struct trace_seq *s = &iter->seq; + struct kmemtrace_alloc_entry *entry; + int ret; + + trace_assign_type(entry, iter->ent); + + ret = trace_seq_printf(s, "type_id %d call_site %pF ptr %lu " + "bytes_req %lu bytes_alloc %lu gfp_flags %lu node %d\n", + entry->type_id, (void *)entry->call_site, (unsigned long)entry->ptr, + (unsigned long)entry->bytes_req, (unsigned long)entry->bytes_alloc, + (unsigned long)entry->gfp_flags, entry->node); + + if (!ret) + return TRACE_TYPE_PARTIAL_LINE; + return TRACE_TYPE_HANDLED; +} + +static enum print_line_t +kmemtrace_print_free(struct trace_iterator *iter, int flags) +{ + struct trace_seq *s = &iter->seq; + struct kmemtrace_free_entry *entry; + int ret; + + trace_assign_type(entry, iter->ent); + + ret = trace_seq_printf(s, "type_id %d call_site %pF ptr %lu\n", + entry->type_id, (void *)entry->call_site, + (unsigned long)entry->ptr); + + if (!ret) + return TRACE_TYPE_PARTIAL_LINE; + return TRACE_TYPE_HANDLED; +} + +static enum print_line_t +kmemtrace_print_alloc_user(struct trace_iterator *iter, int flags) +{ + struct trace_seq *s = &iter->seq; + struct kmemtrace_alloc_entry *entry; + struct kmemtrace_user_event *ev; + struct kmemtrace_user_event_alloc *ev_alloc; + + trace_assign_type(entry, iter->ent); + + ev = trace_seq_reserve(s, sizeof(*ev)); + if (!ev) + return TRACE_TYPE_PARTIAL_LINE; + + ev->event_id = KMEMTRACE_USER_ALLOC; + ev->type_id = entry->type_id; + ev->event_size = sizeof(*ev) + sizeof(*ev_alloc); + ev->cpu = iter->cpu; + ev->timestamp = iter->ts; + ev->call_site = entry->call_site; + ev->ptr = (unsigned long)entry->ptr; + + ev_alloc = trace_seq_reserve(s, sizeof(*ev_alloc)); + if (!ev_alloc) + return TRACE_TYPE_PARTIAL_LINE; + + ev_alloc->bytes_req = entry->bytes_req; + ev_alloc->bytes_alloc = entry->bytes_alloc; + ev_alloc->gfp_flags = entry->gfp_flags; + ev_alloc->node = entry->node; + + return TRACE_TYPE_HANDLED; +} + +static enum print_line_t +kmemtrace_print_free_user(struct trace_iterator *iter, int flags) +{ + struct trace_seq *s = &iter->seq; + struct kmemtrace_free_entry *entry; + struct kmemtrace_user_event *ev; + + trace_assign_type(entry, iter->ent); + + ev = trace_seq_reserve(s, sizeof(*ev)); + if (!ev) + return TRACE_TYPE_PARTIAL_LINE; + + ev->event_id = KMEMTRACE_USER_FREE; + ev->type_id = entry->type_id; + ev->event_size = sizeof(*ev); + ev->cpu = iter->cpu; + ev->timestamp = iter->ts; + ev->call_site = entry->call_site; + ev->ptr = (unsigned long)entry->ptr; + + return TRACE_TYPE_HANDLED; +} + +/* The two other following provide a more minimalistic output */ +static enum print_line_t +kmemtrace_print_alloc_compress(struct trace_iterator *iter) +{ + struct kmemtrace_alloc_entry *entry; + struct trace_seq *s = &iter->seq; + int ret; + + trace_assign_type(entry, iter->ent); + + /* Alloc entry */ + ret = trace_seq_printf(s, " + "); + if (!ret) + return TRACE_TYPE_PARTIAL_LINE; + + /* Type */ + switch (entry->type_id) { + case KMEMTRACE_TYPE_KMALLOC: + ret = trace_seq_printf(s, "K "); + break; + case KMEMTRACE_TYPE_CACHE: + ret = trace_seq_printf(s, "C "); + break; + case KMEMTRACE_TYPE_PAGES: + ret = trace_seq_printf(s, "P "); + break; + default: + ret = trace_seq_printf(s, "? "); + } + + if (!ret) + return TRACE_TYPE_PARTIAL_LINE; + + /* Requested */ + ret = trace_seq_printf(s, "%4zu ", entry->bytes_req); + if (!ret) + return TRACE_TYPE_PARTIAL_LINE; + + /* Allocated */ + ret = trace_seq_printf(s, "%4zu ", entry->bytes_alloc); + if (!ret) + return TRACE_TYPE_PARTIAL_LINE; + + /* Flags + * TODO: would be better to see the name of the GFP flag names + */ + ret = trace_seq_printf(s, "%08x ", entry->gfp_flags); + if (!ret) + return TRACE_TYPE_PARTIAL_LINE; + + /* Pointer to allocated */ + ret = trace_seq_printf(s, "0x%tx ", (ptrdiff_t)entry->ptr); + if (!ret) + return TRACE_TYPE_PARTIAL_LINE; + + /* Node and call site*/ + ret = trace_seq_printf(s, "%4d %pf\n", entry->node, + (void *)entry->call_site); + if (!ret) + return TRACE_TYPE_PARTIAL_LINE; + + return TRACE_TYPE_HANDLED; +} + +static enum print_line_t +kmemtrace_print_free_compress(struct trace_iterator *iter) +{ + struct kmemtrace_free_entry *entry; + struct trace_seq *s = &iter->seq; + int ret; + + trace_assign_type(entry, iter->ent); + + /* Free entry */ + ret = trace_seq_printf(s, " - "); + if (!ret) + return TRACE_TYPE_PARTIAL_LINE; + + /* Type */ + switch (entry->type_id) { + case KMEMTRACE_TYPE_KMALLOC: + ret = trace_seq_printf(s, "K "); + break; + case KMEMTRACE_TYPE_CACHE: + ret = trace_seq_printf(s, "C "); + break; + case KMEMTRACE_TYPE_PAGES: + ret = trace_seq_printf(s, "P "); + break; + default: + ret = trace_seq_printf(s, "? "); + } + + if (!ret) + return TRACE_TYPE_PARTIAL_LINE; + + /* Skip requested/allocated/flags */ + ret = trace_seq_printf(s, " "); + if (!ret) + return TRACE_TYPE_PARTIAL_LINE; + + /* Pointer to allocated */ + ret = trace_seq_printf(s, "0x%tx ", (ptrdiff_t)entry->ptr); + if (!ret) + return TRACE_TYPE_PARTIAL_LINE; + + /* Skip node and print call site*/ + ret = trace_seq_printf(s, " %pf\n", (void *)entry->call_site); + if (!ret) + return TRACE_TYPE_PARTIAL_LINE; + + return TRACE_TYPE_HANDLED; +} + +static enum print_line_t kmemtrace_print_line(struct trace_iterator *iter) +{ + struct trace_entry *entry = iter->ent; + + if (!(kmem_tracer_flags.val & TRACE_KMEM_OPT_MINIMAL)) + return TRACE_TYPE_UNHANDLED; + + switch (entry->type) { + case TRACE_KMEM_ALLOC: + return kmemtrace_print_alloc_compress(iter); + case TRACE_KMEM_FREE: + return kmemtrace_print_free_compress(iter); + default: + return TRACE_TYPE_UNHANDLED; + } +} + +static struct trace_event kmem_trace_alloc = { + .type = TRACE_KMEM_ALLOC, + .trace = kmemtrace_print_alloc, + .binary = kmemtrace_print_alloc_user, +}; + +static struct trace_event kmem_trace_free = { + .type = TRACE_KMEM_FREE, + .trace = kmemtrace_print_free, + .binary = kmemtrace_print_free_user, +}; + +static struct tracer kmem_tracer __read_mostly = { + .name = "kmemtrace", + .init = kmem_trace_init, + .reset = kmem_trace_reset, + .print_line = kmemtrace_print_line, + .print_header = kmemtrace_headers, + .flags = &kmem_tracer_flags +}; + +void kmemtrace_init(void) +{ + /* earliest opportunity to start kmem tracing */ +} + +static int __init init_kmem_tracer(void) +{ + if (!register_ftrace_event(&kmem_trace_alloc)) { + pr_warning("Warning: could not register kmem events\n"); + return 1; + } + + if (!register_ftrace_event(&kmem_trace_free)) { + pr_warning("Warning: could not register kmem events\n"); + return 1; + } + + if (register_tracer(&kmem_tracer) != 0) { + pr_warning("Warning: could not register the kmem tracer\n"); + return 1; + } + + return 0; +} +device_initcall(init_kmem_tracer); diff --git a/kernel/trace/power-traces.c b/kernel/trace/power-traces.c new file mode 100644 index 00000000000..e06c6e3d56a --- /dev/null +++ b/kernel/trace/power-traces.c @@ -0,0 +1,20 @@ +/* + * Power trace points + * + * Copyright (C) 2009 Arjan van de Ven <arjan@linux.intel.com> + */ + +#include <linux/string.h> +#include <linux/types.h> +#include <linux/workqueue.h> +#include <linux/sched.h> +#include <linux/module.h> +#include <linux/slab.h> + +#define CREATE_TRACE_POINTS +#include <trace/events/power.h> + +EXPORT_TRACEPOINT_SYMBOL_GPL(power_start); +EXPORT_TRACEPOINT_SYMBOL_GPL(power_end); +EXPORT_TRACEPOINT_SYMBOL_GPL(power_frequency); + diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c index bd38c5cfd8a..a1ca4956ab5 100644 --- a/kernel/trace/ring_buffer.c +++ b/kernel/trace/ring_buffer.c @@ -4,21 +4,115 @@ * Copyright (C) 2008 Steven Rostedt <srostedt@redhat.com> */ #include <linux/ring_buffer.h> +#include <linux/trace_clock.h> +#include <linux/ftrace_irq.h> #include <linux/spinlock.h> #include <linux/debugfs.h> #include <linux/uaccess.h> +#include <linux/hardirq.h> +#include <linux/kmemcheck.h> #include <linux/module.h> #include <linux/percpu.h> #include <linux/mutex.h> -#include <linux/sched.h> /* used for sched_clock() (for now) */ #include <linux/init.h> #include <linux/hash.h> #include <linux/list.h> +#include <linux/cpu.h> #include <linux/fs.h> #include "trace.h" /* + * The ring buffer header is special. We must manually up keep it. + */ +int ring_buffer_print_entry_header(struct trace_seq *s) +{ + int ret; + + ret = trace_seq_printf(s, "# compressed entry header\n"); + ret = trace_seq_printf(s, "\ttype_len : 5 bits\n"); + ret = trace_seq_printf(s, "\ttime_delta : 27 bits\n"); + ret = trace_seq_printf(s, "\tarray : 32 bits\n"); + ret = trace_seq_printf(s, "\n"); + ret = trace_seq_printf(s, "\tpadding : type == %d\n", + RINGBUF_TYPE_PADDING); + ret = trace_seq_printf(s, "\ttime_extend : type == %d\n", + RINGBUF_TYPE_TIME_EXTEND); + ret = trace_seq_printf(s, "\tdata max type_len == %d\n", + RINGBUF_TYPE_DATA_TYPE_LEN_MAX); + + return ret; +} + +/* + * The ring buffer is made up of a list of pages. A separate list of pages is + * allocated for each CPU. A writer may only write to a buffer that is + * associated with the CPU it is currently executing on. A reader may read + * from any per cpu buffer. + * + * The reader is special. For each per cpu buffer, the reader has its own + * reader page. When a reader has read the entire reader page, this reader + * page is swapped with another page in the ring buffer. + * + * Now, as long as the writer is off the reader page, the reader can do what + * ever it wants with that page. The writer will never write to that page + * again (as long as it is out of the ring buffer). + * + * Here's some silly ASCII art. + * + * +------+ + * |reader| RING BUFFER + * |page | + * +------+ +---+ +---+ +---+ + * | |-->| |-->| | + * +---+ +---+ +---+ + * ^ | + * | | + * +---------------+ + * + * + * +------+ + * |reader| RING BUFFER + * |page |------------------v + * +------+ +---+ +---+ +---+ + * | |-->| |-->| | + * +---+ +---+ +---+ + * ^ | + * | | + * +---------------+ + * + * + * +------+ + * |reader| RING BUFFER + * |page |------------------v + * +------+ +---+ +---+ +---+ + * ^ | |-->| |-->| | + * | +---+ +---+ +---+ + * | | + * | | + * +------------------------------+ + * + * + * +------+ + * |buffer| RING BUFFER + * |page |------------------v + * +------+ +---+ +---+ +---+ + * ^ | | | |-->| | + * | New +---+ +---+ +---+ + * | Reader------^ | + * | page | + * +------------------------------+ + * + * + * After we make this swap, the reader can hand this page off to the splice + * code and be done with it. It can even allocate a new page if it needs to + * and swap that into the ring buffer. + * + * We will be using cmpxchg soon to make all this lockless. + * + */ + +/* * A fast way to enable or disable all ring buffers is to * call tracing_on or tracing_off. Turning off the ring buffers * prevents all ring buffers from being recorded to. @@ -57,7 +151,9 @@ enum { RB_BUFFERS_DISABLED = 1 << RB_BUFFERS_DISABLED_BIT, }; -static long ring_buffer_flags __read_mostly = RB_BUFFERS_ON; +static unsigned long ring_buffer_flags __read_mostly = RB_BUFFERS_ON; + +#define BUF_PAGE_HDR_SIZE offsetof(struct buffer_data_page, data) /** * tracing_on - enable all tracing buffers @@ -89,59 +185,69 @@ EXPORT_SYMBOL_GPL(tracing_off); * tracing_off_permanent - permanently disable ring buffers * * This function, once called, will disable all ring buffers - * permanenty. + * permanently. */ void tracing_off_permanent(void) { set_bit(RB_BUFFERS_DISABLED_BIT, &ring_buffer_flags); } -#include "trace.h" +/** + * tracing_is_on - show state of ring buffers enabled + */ +int tracing_is_on(void) +{ + return ring_buffer_flags == RB_BUFFERS_ON; +} +EXPORT_SYMBOL_GPL(tracing_is_on); -/* Up this if you want to test the TIME_EXTENTS and normalization */ -#define DEBUG_SHIFT 0 +#define RB_EVNT_HDR_SIZE (offsetof(struct ring_buffer_event, array)) +#define RB_ALIGNMENT 4U +#define RB_MAX_SMALL_DATA (RB_ALIGNMENT * RINGBUF_TYPE_DATA_TYPE_LEN_MAX) +#define RB_EVNT_MIN_SIZE 8U /* two 32bit words */ -/* FIXME!!! */ -u64 ring_buffer_time_stamp(int cpu) -{ - u64 time; +/* define RINGBUF_TYPE_DATA for 'case RINGBUF_TYPE_DATA:' */ +#define RINGBUF_TYPE_DATA 0 ... RINGBUF_TYPE_DATA_TYPE_LEN_MAX - preempt_disable_notrace(); - /* shift to debug/test normalization and TIME_EXTENTS */ - time = sched_clock() << DEBUG_SHIFT; - preempt_enable_no_resched_notrace(); +enum { + RB_LEN_TIME_EXTEND = 8, + RB_LEN_TIME_STAMP = 16, +}; - return time; +static inline int rb_null_event(struct ring_buffer_event *event) +{ + return event->type_len == RINGBUF_TYPE_PADDING && !event->time_delta; } -EXPORT_SYMBOL_GPL(ring_buffer_time_stamp); -void ring_buffer_normalize_time_stamp(int cpu, u64 *ts) +static void rb_event_set_padding(struct ring_buffer_event *event) { - /* Just stupid testing the normalize function and deltas */ - *ts >>= DEBUG_SHIFT; + /* padding has a NULL time_delta */ + event->type_len = RINGBUF_TYPE_PADDING; + event->time_delta = 0; } -EXPORT_SYMBOL_GPL(ring_buffer_normalize_time_stamp); -#define RB_EVNT_HDR_SIZE (sizeof(struct ring_buffer_event)) -#define RB_ALIGNMENT_SHIFT 2 -#define RB_ALIGNMENT (1 << RB_ALIGNMENT_SHIFT) -#define RB_MAX_SMALL_DATA 28 +static unsigned +rb_event_data_length(struct ring_buffer_event *event) +{ + unsigned length; -enum { - RB_LEN_TIME_EXTEND = 8, - RB_LEN_TIME_STAMP = 16, -}; + if (event->type_len) + length = event->type_len * RB_ALIGNMENT; + else + length = event->array[0]; + return length + RB_EVNT_HDR_SIZE; +} /* inline for ring buffer fast paths */ -static inline unsigned +static unsigned rb_event_length(struct ring_buffer_event *event) { - unsigned length; - - switch (event->type) { + switch (event->type_len) { case RINGBUF_TYPE_PADDING: - /* undefined */ - return -1; + if (rb_null_event(event)) + /* undefined */ + return -1; + return event->array[0] + RB_EVNT_HDR_SIZE; case RINGBUF_TYPE_TIME_EXTEND: return RB_LEN_TIME_EXTEND; @@ -150,11 +256,7 @@ rb_event_length(struct ring_buffer_event *event) return RB_LEN_TIME_STAMP; case RINGBUF_TYPE_DATA: - if (event->len) - length = event->len << RB_ALIGNMENT_SHIFT; - else - length = event->array[0]; - return length + RB_EVNT_HDR_SIZE; + return rb_event_data_length(event); default: BUG(); } @@ -169,7 +271,7 @@ rb_event_length(struct ring_buffer_event *event) unsigned ring_buffer_event_length(struct ring_buffer_event *event) { unsigned length = rb_event_length(event); - if (event->type != RINGBUF_TYPE_DATA) + if (event->type_len > RINGBUF_TYPE_DATA_TYPE_LEN_MAX) return length; length -= RB_EVNT_HDR_SIZE; if (length > RB_MAX_SMALL_DATA + sizeof(event->array[0])) @@ -179,12 +281,12 @@ unsigned ring_buffer_event_length(struct ring_buffer_event *event) EXPORT_SYMBOL_GPL(ring_buffer_event_length); /* inline for ring buffer fast paths */ -static inline void * +static void * rb_event_data(struct ring_buffer_event *event) { - BUG_ON(event->type != RINGBUF_TYPE_DATA); + BUG_ON(event->type_len > RINGBUF_TYPE_DATA_TYPE_LEN_MAX); /* If length is in len field, then array[0] has the data */ - if (event->len) + if (event->type_len) return (void *)&event->array[0]; /* Otherwise length is in array[0] and array[1] has the data */ return (void *)&event->array[1]; @@ -209,30 +311,65 @@ EXPORT_SYMBOL_GPL(ring_buffer_event_data); struct buffer_data_page { u64 time_stamp; /* page time stamp */ - local_t commit; /* write commited index */ + local_t commit; /* write committed index */ unsigned char data[]; /* data of buffer page */ }; +/* + * Note, the buffer_page list must be first. The buffer pages + * are allocated in cache lines, which means that each buffer + * page will be at the beginning of a cache line, and thus + * the least significant bits will be zero. We use this to + * add flags in the list struct pointers, to make the ring buffer + * lockless. + */ struct buffer_page { + struct list_head list; /* list of buffer pages */ local_t write; /* index for next write */ unsigned read; /* index for next read */ - struct list_head list; /* list of free pages */ + local_t entries; /* entries on this page */ struct buffer_data_page *page; /* Actual data page */ }; +/* + * The buffer page counters, write and entries, must be reset + * atomically when crossing page boundaries. To synchronize this + * update, two counters are inserted into the number. One is + * the actual counter for the write position or count on the page. + * + * The other is a counter of updaters. Before an update happens + * the update partition of the counter is incremented. This will + * allow the updater to update the counter atomically. + * + * The counter is 20 bits, and the state data is 12. + */ +#define RB_WRITE_MASK 0xfffff +#define RB_WRITE_INTCNT (1 << 20) + static void rb_init_page(struct buffer_data_page *bpage) { local_set(&bpage->commit, 0); } +/** + * ring_buffer_page_len - the size of data on the page. + * @page: The page to read + * + * Returns the amount of data on the page, including buffer page header. + */ +size_t ring_buffer_page_len(void *page) +{ + return local_read(&((struct buffer_data_page *)page)->commit) + + BUF_PAGE_HDR_SIZE; +} + /* * Also stolen from mm/slob.c. Thanks to Mathieu Desnoyers for pointing * this issue out. */ -static inline void free_buffer_page(struct buffer_page *bpage) +static void free_buffer_page(struct buffer_page *bpage) { - if (bpage->page) - free_page((unsigned long)bpage->page); + free_page((unsigned long)bpage->page); kfree(bpage); } @@ -246,7 +383,38 @@ static inline int test_time_stamp(u64 delta) return 0; } -#define BUF_PAGE_SIZE (PAGE_SIZE - offsetof(struct buffer_data_page, data)) +#define BUF_PAGE_SIZE (PAGE_SIZE - BUF_PAGE_HDR_SIZE) + +/* Max payload is BUF_PAGE_SIZE - header (8bytes) */ +#define BUF_MAX_DATA_SIZE (BUF_PAGE_SIZE - (sizeof(u32) * 2)) + +/* Max number of timestamps that can fit on a page */ +#define RB_TIMESTAMPS_PER_PAGE (BUF_PAGE_SIZE / RB_LEN_TIME_STAMP) + +int ring_buffer_print_page_header(struct trace_seq *s) +{ + struct buffer_data_page field; + int ret; + + ret = trace_seq_printf(s, "\tfield: u64 timestamp;\t" + "offset:0;\tsize:%u;\tsigned:%u;\n", + (unsigned int)sizeof(field.time_stamp), + (unsigned int)is_signed_type(u64)); + + ret = trace_seq_printf(s, "\tfield: local_t commit;\t" + "offset:%u;\tsize:%u;\tsigned:%u;\n", + (unsigned int)offsetof(typeof(field), commit), + (unsigned int)sizeof(field.commit), + (unsigned int)is_signed_type(long)); + + ret = trace_seq_printf(s, "\tfield: char data;\t" + "offset:%u;\tsize:%u;\tsigned:%u;\n", + (unsigned int)offsetof(typeof(field), data), + (unsigned int)BUF_PAGE_SIZE, + (unsigned int)is_signed_type(char)); + + return ret; +} /* * head_page == tail_page && head == tail then buffer is empty. @@ -254,16 +422,20 @@ static inline int test_time_stamp(u64 delta) struct ring_buffer_per_cpu { int cpu; struct ring_buffer *buffer; - spinlock_t reader_lock; /* serialize readers */ + spinlock_t reader_lock; /* serialize readers */ raw_spinlock_t lock; struct lock_class_key lock_key; - struct list_head pages; + struct list_head *pages; struct buffer_page *head_page; /* read from head */ struct buffer_page *tail_page; /* write to tail */ - struct buffer_page *commit_page; /* commited pages */ + struct buffer_page *commit_page; /* committed pages */ struct buffer_page *reader_page; - unsigned long overrun; - unsigned long entries; + local_t commit_overrun; + local_t overrun; + local_t entries; + local_t committing; + local_t commits; + unsigned long read; u64 write_stamp; u64 read_stamp; atomic_t record_disabled; @@ -273,12 +445,19 @@ struct ring_buffer { unsigned pages; unsigned flags; int cpus; - cpumask_var_t cpumask; atomic_t record_disabled; + cpumask_var_t cpumask; + + struct lock_class_key *reader_lock_key; struct mutex mutex; struct ring_buffer_per_cpu **buffers; + +#ifdef CONFIG_HOTPLUG_CPU + struct notifier_block cpu_notify; +#endif + u64 (*clock)(void); }; struct ring_buffer_iter { @@ -289,33 +468,456 @@ struct ring_buffer_iter { }; /* buffer may be either ring_buffer or ring_buffer_per_cpu */ -#define RB_WARN_ON(buffer, cond) \ - ({ \ - int _____ret = unlikely(cond); \ - if (_____ret) { \ - atomic_inc(&buffer->record_disabled); \ - WARN_ON(1); \ - } \ - _____ret; \ +#define RB_WARN_ON(b, cond) \ + ({ \ + int _____ret = unlikely(cond); \ + if (_____ret) { \ + if (__same_type(*(b), struct ring_buffer_per_cpu)) { \ + struct ring_buffer_per_cpu *__b = \ + (void *)b; \ + atomic_inc(&__b->buffer->record_disabled); \ + } else \ + atomic_inc(&b->record_disabled); \ + WARN_ON(1); \ + } \ + _____ret; \ }) +/* Up this if you want to test the TIME_EXTENTS and normalization */ +#define DEBUG_SHIFT 0 + +static inline u64 rb_time_stamp(struct ring_buffer *buffer) +{ + /* shift to debug/test normalization and TIME_EXTENTS */ + return buffer->clock() << DEBUG_SHIFT; +} + +u64 ring_buffer_time_stamp(struct ring_buffer *buffer, int cpu) +{ + u64 time; + + preempt_disable_notrace(); + time = rb_time_stamp(buffer); + preempt_enable_no_resched_notrace(); + + return time; +} +EXPORT_SYMBOL_GPL(ring_buffer_time_stamp); + +void ring_buffer_normalize_time_stamp(struct ring_buffer *buffer, + int cpu, u64 *ts) +{ + /* Just stupid testing the normalize function and deltas */ + *ts >>= DEBUG_SHIFT; +} +EXPORT_SYMBOL_GPL(ring_buffer_normalize_time_stamp); + +/* + * Making the ring buffer lockless makes things tricky. + * Although writes only happen on the CPU that they are on, + * and they only need to worry about interrupts. Reads can + * happen on any CPU. + * + * The reader page is always off the ring buffer, but when the + * reader finishes with a page, it needs to swap its page with + * a new one from the buffer. The reader needs to take from + * the head (writes go to the tail). But if a writer is in overwrite + * mode and wraps, it must push the head page forward. + * + * Here lies the problem. + * + * The reader must be careful to replace only the head page, and + * not another one. As described at the top of the file in the + * ASCII art, the reader sets its old page to point to the next + * page after head. It then sets the page after head to point to + * the old reader page. But if the writer moves the head page + * during this operation, the reader could end up with the tail. + * + * We use cmpxchg to help prevent this race. We also do something + * special with the page before head. We set the LSB to 1. + * + * When the writer must push the page forward, it will clear the + * bit that points to the head page, move the head, and then set + * the bit that points to the new head page. + * + * We also don't want an interrupt coming in and moving the head + * page on another writer. Thus we use the second LSB to catch + * that too. Thus: + * + * head->list->prev->next bit 1 bit 0 + * ------- ------- + * Normal page 0 0 + * Points to head page 0 1 + * New head page 1 0 + * + * Note we can not trust the prev pointer of the head page, because: + * + * +----+ +-----+ +-----+ + * | |------>| T |---X--->| N | + * | |<------| | | | + * +----+ +-----+ +-----+ + * ^ ^ | + * | +-----+ | | + * +----------| R |----------+ | + * | |<-----------+ + * +-----+ + * + * Key: ---X--> HEAD flag set in pointer + * T Tail page + * R Reader page + * N Next page + * + * (see __rb_reserve_next() to see where this happens) + * + * What the above shows is that the reader just swapped out + * the reader page with a page in the buffer, but before it + * could make the new header point back to the new page added + * it was preempted by a writer. The writer moved forward onto + * the new page added by the reader and is about to move forward + * again. + * + * You can see, it is legitimate for the previous pointer of + * the head (or any page) not to point back to itself. But only + * temporarially. + */ + +#define RB_PAGE_NORMAL 0UL +#define RB_PAGE_HEAD 1UL +#define RB_PAGE_UPDATE 2UL + + +#define RB_FLAG_MASK 3UL + +/* PAGE_MOVED is not part of the mask */ +#define RB_PAGE_MOVED 4UL + +/* + * rb_list_head - remove any bit + */ +static struct list_head *rb_list_head(struct list_head *list) +{ + unsigned long val = (unsigned long)list; + + return (struct list_head *)(val & ~RB_FLAG_MASK); +} + +/* + * rb_is_head_page - test if the given page is the head page + * + * Because the reader may move the head_page pointer, we can + * not trust what the head page is (it may be pointing to + * the reader page). But if the next page is a header page, + * its flags will be non zero. + */ +static int inline +rb_is_head_page(struct ring_buffer_per_cpu *cpu_buffer, + struct buffer_page *page, struct list_head *list) +{ + unsigned long val; + + val = (unsigned long)list->next; + + if ((val & ~RB_FLAG_MASK) != (unsigned long)&page->list) + return RB_PAGE_MOVED; + + return val & RB_FLAG_MASK; +} + +/* + * rb_is_reader_page + * + * The unique thing about the reader page, is that, if the + * writer is ever on it, the previous pointer never points + * back to the reader page. + */ +static int rb_is_reader_page(struct buffer_page *page) +{ + struct list_head *list = page->list.prev; + + return rb_list_head(list->next) != &page->list; +} + +/* + * rb_set_list_to_head - set a list_head to be pointing to head. + */ +static void rb_set_list_to_head(struct ring_buffer_per_cpu *cpu_buffer, + struct list_head *list) +{ + unsigned long *ptr; + + ptr = (unsigned long *)&list->next; + *ptr |= RB_PAGE_HEAD; + *ptr &= ~RB_PAGE_UPDATE; +} + +/* + * rb_head_page_activate - sets up head page + */ +static void rb_head_page_activate(struct ring_buffer_per_cpu *cpu_buffer) +{ + struct buffer_page *head; + + head = cpu_buffer->head_page; + if (!head) + return; + + /* + * Set the previous list pointer to have the HEAD flag. + */ + rb_set_list_to_head(cpu_buffer, head->list.prev); +} + +static void rb_list_head_clear(struct list_head *list) +{ + unsigned long *ptr = (unsigned long *)&list->next; + + *ptr &= ~RB_FLAG_MASK; +} + +/* + * rb_head_page_dactivate - clears head page ptr (for free list) + */ +static void +rb_head_page_deactivate(struct ring_buffer_per_cpu *cpu_buffer) +{ + struct list_head *hd; + + /* Go through the whole list and clear any pointers found. */ + rb_list_head_clear(cpu_buffer->pages); + + list_for_each(hd, cpu_buffer->pages) + rb_list_head_clear(hd); +} + +static int rb_head_page_set(struct ring_buffer_per_cpu *cpu_buffer, + struct buffer_page *head, + struct buffer_page *prev, + int old_flag, int new_flag) +{ + struct list_head *list; + unsigned long val = (unsigned long)&head->list; + unsigned long ret; + + list = &prev->list; + + val &= ~RB_FLAG_MASK; + + ret = cmpxchg((unsigned long *)&list->next, + val | old_flag, val | new_flag); + + /* check if the reader took the page */ + if ((ret & ~RB_FLAG_MASK) != val) + return RB_PAGE_MOVED; + + return ret & RB_FLAG_MASK; +} + +static int rb_head_page_set_update(struct ring_buffer_per_cpu *cpu_buffer, + struct buffer_page *head, + struct buffer_page *prev, + int old_flag) +{ + return rb_head_page_set(cpu_buffer, head, prev, + old_flag, RB_PAGE_UPDATE); +} + +static int rb_head_page_set_head(struct ring_buffer_per_cpu *cpu_buffer, + struct buffer_page *head, + struct buffer_page *prev, + int old_flag) +{ + return rb_head_page_set(cpu_buffer, head, prev, + old_flag, RB_PAGE_HEAD); +} + +static int rb_head_page_set_normal(struct ring_buffer_per_cpu *cpu_buffer, + struct buffer_page *head, + struct buffer_page *prev, + int old_flag) +{ + return rb_head_page_set(cpu_buffer, head, prev, + old_flag, RB_PAGE_NORMAL); +} + +static inline void rb_inc_page(struct ring_buffer_per_cpu *cpu_buffer, + struct buffer_page **bpage) +{ + struct list_head *p = rb_list_head((*bpage)->list.next); + + *bpage = list_entry(p, struct buffer_page, list); +} + +static struct buffer_page * +rb_set_head_page(struct ring_buffer_per_cpu *cpu_buffer) +{ + struct buffer_page *head; + struct buffer_page *page; + struct list_head *list; + int i; + + if (RB_WARN_ON(cpu_buffer, !cpu_buffer->head_page)) + return NULL; + + /* sanity check */ + list = cpu_buffer->pages; + if (RB_WARN_ON(cpu_buffer, rb_list_head(list->prev->next) != list)) + return NULL; + + page = head = cpu_buffer->head_page; + /* + * It is possible that the writer moves the header behind + * where we started, and we miss in one loop. + * A second loop should grab the header, but we'll do + * three loops just because I'm paranoid. + */ + for (i = 0; i < 3; i++) { + do { + if (rb_is_head_page(cpu_buffer, page, page->list.prev)) { + cpu_buffer->head_page = page; + return page; + } + rb_inc_page(cpu_buffer, &page); + } while (page != head); + } + + RB_WARN_ON(cpu_buffer, 1); + + return NULL; +} + +static int rb_head_page_replace(struct buffer_page *old, + struct buffer_page *new) +{ + unsigned long *ptr = (unsigned long *)&old->list.prev->next; + unsigned long val; + unsigned long ret; + + val = *ptr & ~RB_FLAG_MASK; + val |= RB_PAGE_HEAD; + + ret = cmpxchg(ptr, val, (unsigned long)&new->list); + + return ret == val; +} + +/* + * rb_tail_page_update - move the tail page forward + * + * Returns 1 if moved tail page, 0 if someone else did. + */ +static int rb_tail_page_update(struct ring_buffer_per_cpu *cpu_buffer, + struct buffer_page *tail_page, + struct buffer_page *next_page) +{ + struct buffer_page *old_tail; + unsigned long old_entries; + unsigned long old_write; + int ret = 0; + + /* + * The tail page now needs to be moved forward. + * + * We need to reset the tail page, but without messing + * with possible erasing of data brought in by interrupts + * that have moved the tail page and are currently on it. + * + * We add a counter to the write field to denote this. + */ + old_write = local_add_return(RB_WRITE_INTCNT, &next_page->write); + old_entries = local_add_return(RB_WRITE_INTCNT, &next_page->entries); + + /* + * Just make sure we have seen our old_write and synchronize + * with any interrupts that come in. + */ + barrier(); + + /* + * If the tail page is still the same as what we think + * it is, then it is up to us to update the tail + * pointer. + */ + if (tail_page == cpu_buffer->tail_page) { + /* Zero the write counter */ + unsigned long val = old_write & ~RB_WRITE_MASK; + unsigned long eval = old_entries & ~RB_WRITE_MASK; + + /* + * This will only succeed if an interrupt did + * not come in and change it. In which case, we + * do not want to modify it. + * + * We add (void) to let the compiler know that we do not care + * about the return value of these functions. We use the + * cmpxchg to only update if an interrupt did not already + * do it for us. If the cmpxchg fails, we don't care. + */ + (void)local_cmpxchg(&next_page->write, old_write, val); + (void)local_cmpxchg(&next_page->entries, old_entries, eval); + + /* + * No need to worry about races with clearing out the commit. + * it only can increment when a commit takes place. But that + * only happens in the outer most nested commit. + */ + local_set(&next_page->page->commit, 0); + + old_tail = cmpxchg(&cpu_buffer->tail_page, + tail_page, next_page); + + if (old_tail == tail_page) + ret = 1; + } + + return ret; +} + +static int rb_check_bpage(struct ring_buffer_per_cpu *cpu_buffer, + struct buffer_page *bpage) +{ + unsigned long val = (unsigned long)bpage; + + if (RB_WARN_ON(cpu_buffer, val & RB_FLAG_MASK)) + return 1; + + return 0; +} + +/** + * rb_check_list - make sure a pointer to a list has the last bits zero + */ +static int rb_check_list(struct ring_buffer_per_cpu *cpu_buffer, + struct list_head *list) +{ + if (RB_WARN_ON(cpu_buffer, rb_list_head(list->prev) != list->prev)) + return 1; + if (RB_WARN_ON(cpu_buffer, rb_list_head(list->next) != list->next)) + return 1; + return 0; +} + /** * check_pages - integrity check of buffer pages * @cpu_buffer: CPU buffer with pages to test * - * As a safty measure we check to make sure the data pages have not + * As a safety measure we check to make sure the data pages have not * been corrupted. */ static int rb_check_pages(struct ring_buffer_per_cpu *cpu_buffer) { - struct list_head *head = &cpu_buffer->pages; + struct list_head *head = cpu_buffer->pages; struct buffer_page *bpage, *tmp; + rb_head_page_deactivate(cpu_buffer); + if (RB_WARN_ON(cpu_buffer, head->next->prev != head)) return -1; if (RB_WARN_ON(cpu_buffer, head->prev->next != head)) return -1; + if (rb_check_list(cpu_buffer, head)) + return -1; + list_for_each_entry_safe(bpage, tmp, head, list) { if (RB_WARN_ON(cpu_buffer, bpage->list.next->prev != &bpage->list)) @@ -323,25 +925,33 @@ static int rb_check_pages(struct ring_buffer_per_cpu *cpu_buffer) if (RB_WARN_ON(cpu_buffer, bpage->list.prev->next != &bpage->list)) return -1; + if (rb_check_list(cpu_buffer, &bpage->list)) + return -1; } + rb_head_page_activate(cpu_buffer); + return 0; } static int rb_allocate_pages(struct ring_buffer_per_cpu *cpu_buffer, unsigned nr_pages) { - struct list_head *head = &cpu_buffer->pages; struct buffer_page *bpage, *tmp; unsigned long addr; LIST_HEAD(pages); unsigned i; + WARN_ON(!nr_pages); + for (i = 0; i < nr_pages; i++) { bpage = kzalloc_node(ALIGN(sizeof(*bpage), cache_line_size()), GFP_KERNEL, cpu_to_node(cpu_buffer->cpu)); if (!bpage) goto free_pages; + + rb_check_bpage(cpu_buffer, bpage); + list_add(&bpage->list, &pages); addr = __get_free_page(GFP_KERNEL); @@ -351,7 +961,13 @@ static int rb_allocate_pages(struct ring_buffer_per_cpu *cpu_buffer, rb_init_page(bpage->page); } - list_splice(&pages, head); + /* + * The ring buffer page list is a circular list that does not + * start and end with a list head. All page list items point to + * other pages. + */ + cpu_buffer->pages = pages.next; + list_del(&pages); rb_check_pages(cpu_buffer); @@ -381,14 +997,16 @@ rb_allocate_cpu_buffer(struct ring_buffer *buffer, int cpu) cpu_buffer->cpu = cpu; cpu_buffer->buffer = buffer; spin_lock_init(&cpu_buffer->reader_lock); + lockdep_set_class(&cpu_buffer->reader_lock, buffer->reader_lock_key); cpu_buffer->lock = (raw_spinlock_t)__RAW_SPIN_LOCK_UNLOCKED; - INIT_LIST_HEAD(&cpu_buffer->pages); bpage = kzalloc_node(ALIGN(sizeof(*bpage), cache_line_size()), GFP_KERNEL, cpu_to_node(cpu)); if (!bpage) goto fail_free_buffer; + rb_check_bpage(cpu_buffer, bpage); + cpu_buffer->reader_page = bpage; addr = __get_free_page(GFP_KERNEL); if (!addr) @@ -403,9 +1021,11 @@ rb_allocate_cpu_buffer(struct ring_buffer *buffer, int cpu) goto fail_free_reader; cpu_buffer->head_page - = list_entry(cpu_buffer->pages.next, struct buffer_page, list); + = list_entry(cpu_buffer->pages, struct buffer_page, list); cpu_buffer->tail_page = cpu_buffer->commit_page = cpu_buffer->head_page; + rb_head_page_activate(cpu_buffer); + return cpu_buffer; fail_free_reader: @@ -418,24 +1038,29 @@ rb_allocate_cpu_buffer(struct ring_buffer *buffer, int cpu) static void rb_free_cpu_buffer(struct ring_buffer_per_cpu *cpu_buffer) { - struct list_head *head = &cpu_buffer->pages; + struct list_head *head = cpu_buffer->pages; struct buffer_page *bpage, *tmp; - list_del_init(&cpu_buffer->reader_page->list); free_buffer_page(cpu_buffer->reader_page); - list_for_each_entry_safe(bpage, tmp, head, list) { - list_del_init(&bpage->list); + rb_head_page_deactivate(cpu_buffer); + + if (head) { + list_for_each_entry_safe(bpage, tmp, head, list) { + list_del_init(&bpage->list); + free_buffer_page(bpage); + } + bpage = list_entry(head, struct buffer_page, list); free_buffer_page(bpage); } + kfree(cpu_buffer); } -/* - * Causes compile errors if the struct buffer_page gets bigger - * than the struct page. - */ -extern int ring_buffer_page_too_big(void); +#ifdef CONFIG_HOTPLUG_CPU +static int rb_cpu_notify(struct notifier_block *self, + unsigned long action, void *hcpu); +#endif /** * ring_buffer_alloc - allocate a new ring_buffer @@ -447,17 +1072,13 @@ extern int ring_buffer_page_too_big(void); * when the buffer wraps. If this flag is not set, the buffer will * drop data when the tail hits the head. */ -struct ring_buffer *ring_buffer_alloc(unsigned long size, unsigned flags) +struct ring_buffer *__ring_buffer_alloc(unsigned long size, unsigned flags, + struct lock_class_key *key) { struct ring_buffer *buffer; int bsize; int cpu; - /* Paranoid! Optimizes out when all is well */ - if (sizeof(struct buffer_page) > sizeof(struct page)) - ring_buffer_page_too_big(); - - /* keep it in its own cache line */ buffer = kzalloc(ALIGN(sizeof(*buffer), cache_line_size()), GFP_KERNEL); @@ -469,12 +1090,24 @@ struct ring_buffer *ring_buffer_alloc(unsigned long size, unsigned flags) buffer->pages = DIV_ROUND_UP(size, BUF_PAGE_SIZE); buffer->flags = flags; + buffer->clock = trace_clock_local; + buffer->reader_lock_key = key; /* need at least two pages */ - if (buffer->pages == 1) - buffer->pages++; + if (buffer->pages < 2) + buffer->pages = 2; + /* + * In case of non-hotplug cpu, if the ring-buffer is allocated + * in early initcall, it will not be notified of secondary cpus. + * In that off case, we need to allocate for all possible cpus. + */ +#ifdef CONFIG_HOTPLUG_CPU + get_online_cpus(); + cpumask_copy(buffer->cpumask, cpu_online_mask); +#else cpumask_copy(buffer->cpumask, cpu_possible_mask); +#endif buffer->cpus = nr_cpu_ids; bsize = sizeof(void *) * nr_cpu_ids; @@ -490,6 +1123,13 @@ struct ring_buffer *ring_buffer_alloc(unsigned long size, unsigned flags) goto fail_free_buffers; } +#ifdef CONFIG_HOTPLUG_CPU + buffer->cpu_notify.notifier_call = rb_cpu_notify; + buffer->cpu_notify.priority = 0; + register_cpu_notifier(&buffer->cpu_notify); +#endif + + put_online_cpus(); mutex_init(&buffer->mutex); return buffer; @@ -503,12 +1143,13 @@ struct ring_buffer *ring_buffer_alloc(unsigned long size, unsigned flags) fail_free_cpumask: free_cpumask_var(buffer->cpumask); + put_online_cpus(); fail_free_buffer: kfree(buffer); return NULL; } -EXPORT_SYMBOL_GPL(ring_buffer_alloc); +EXPORT_SYMBOL_GPL(__ring_buffer_alloc); /** * ring_buffer_free - free a ring buffer. @@ -519,15 +1160,30 @@ ring_buffer_free(struct ring_buffer *buffer) { int cpu; + get_online_cpus(); + +#ifdef CONFIG_HOTPLUG_CPU + unregister_cpu_notifier(&buffer->cpu_notify); +#endif + for_each_buffer_cpu(buffer, cpu) rb_free_cpu_buffer(buffer->buffers[cpu]); + put_online_cpus(); + + kfree(buffer->buffers); free_cpumask_var(buffer->cpumask); kfree(buffer); } EXPORT_SYMBOL_GPL(ring_buffer_free); +void ring_buffer_set_clock(struct ring_buffer *buffer, + u64 (*clock)(void)) +{ + buffer->clock = clock; +} + static void rb_reset_cpu(struct ring_buffer_per_cpu *cpu_buffer); static void @@ -540,18 +1196,22 @@ rb_remove_pages(struct ring_buffer_per_cpu *cpu_buffer, unsigned nr_pages) atomic_inc(&cpu_buffer->record_disabled); synchronize_sched(); + spin_lock_irq(&cpu_buffer->reader_lock); + rb_head_page_deactivate(cpu_buffer); + for (i = 0; i < nr_pages; i++) { - if (RB_WARN_ON(cpu_buffer, list_empty(&cpu_buffer->pages))) + if (RB_WARN_ON(cpu_buffer, list_empty(cpu_buffer->pages))) return; - p = cpu_buffer->pages.next; + p = cpu_buffer->pages->next; bpage = list_entry(p, struct buffer_page, list); list_del_init(&bpage->list); free_buffer_page(bpage); } - if (RB_WARN_ON(cpu_buffer, list_empty(&cpu_buffer->pages))) + if (RB_WARN_ON(cpu_buffer, list_empty(cpu_buffer->pages))) return; rb_reset_cpu(cpu_buffer); + spin_unlock_irq(&cpu_buffer->reader_lock); rb_check_pages(cpu_buffer); @@ -570,15 +1230,19 @@ rb_insert_pages(struct ring_buffer_per_cpu *cpu_buffer, atomic_inc(&cpu_buffer->record_disabled); synchronize_sched(); + spin_lock_irq(&cpu_buffer->reader_lock); + rb_head_page_deactivate(cpu_buffer); + for (i = 0; i < nr_pages; i++) { if (RB_WARN_ON(cpu_buffer, list_empty(pages))) return; p = pages->next; bpage = list_entry(p, struct buffer_page, list); list_del_init(&bpage->list); - list_add_tail(&bpage->list, &cpu_buffer->pages); + list_add_tail(&bpage->list, cpu_buffer->pages); } rb_reset_cpu(cpu_buffer); + spin_unlock_irq(&cpu_buffer->reader_lock); rb_check_pages(cpu_buffer); @@ -627,16 +1291,15 @@ int ring_buffer_resize(struct ring_buffer *buffer, unsigned long size) return size; mutex_lock(&buffer->mutex); + get_online_cpus(); nr_pages = DIV_ROUND_UP(size, BUF_PAGE_SIZE); if (size < buffer_size) { /* easy case, just free pages */ - if (RB_WARN_ON(buffer, nr_pages >= buffer->pages)) { - mutex_unlock(&buffer->mutex); - return -1; - } + if (RB_WARN_ON(buffer, nr_pages >= buffer->pages)) + goto out_fail; rm_pages = buffer->pages - nr_pages; @@ -655,10 +1318,8 @@ int ring_buffer_resize(struct ring_buffer *buffer, unsigned long size) * add these pages to the cpu_buffers. Otherwise we just free * them all and return -ENOMEM; */ - if (RB_WARN_ON(buffer, nr_pages <= buffer->pages)) { - mutex_unlock(&buffer->mutex); - return -1; - } + if (RB_WARN_ON(buffer, nr_pages <= buffer->pages)) + goto out_fail; new_pages = nr_pages - buffer->pages; @@ -683,13 +1344,12 @@ int ring_buffer_resize(struct ring_buffer *buffer, unsigned long size) rb_insert_pages(cpu_buffer, &pages, new_pages); } - if (RB_WARN_ON(buffer, !list_empty(&pages))) { - mutex_unlock(&buffer->mutex); - return -1; - } + if (RB_WARN_ON(buffer, !list_empty(&pages))) + goto out_fail; out: buffer->pages = nr_pages; + put_online_cpus(); mutex_unlock(&buffer->mutex); return size; @@ -699,15 +1359,20 @@ int ring_buffer_resize(struct ring_buffer *buffer, unsigned long size) list_del_init(&bpage->list); free_buffer_page(bpage); } + put_online_cpus(); mutex_unlock(&buffer->mutex); return -ENOMEM; -} -EXPORT_SYMBOL_GPL(ring_buffer_resize); -static inline int rb_null_event(struct ring_buffer_event *event) -{ - return event->type == RINGBUF_TYPE_PADDING; + /* + * Something went totally wrong, and we are too paranoid + * to even clean up the mess. + */ + out_fail: + put_online_cpus(); + mutex_unlock(&buffer->mutex); + return -1; } +EXPORT_SYMBOL_GPL(ring_buffer_resize); static inline void * __rb_data_page_index(struct buffer_data_page *bpage, unsigned index) @@ -728,21 +1393,14 @@ rb_reader_event(struct ring_buffer_per_cpu *cpu_buffer) } static inline struct ring_buffer_event * -rb_head_event(struct ring_buffer_per_cpu *cpu_buffer) -{ - return __rb_page_index(cpu_buffer->head_page, - cpu_buffer->head_page->read); -} - -static inline struct ring_buffer_event * rb_iter_head_event(struct ring_buffer_iter *iter) { return __rb_page_index(iter->head_page, iter->head); } -static inline unsigned rb_page_write(struct buffer_page *bpage) +static inline unsigned long rb_page_write(struct buffer_page *bpage) { - return local_read(&bpage->write); + return local_read(&bpage->write) & RB_WRITE_MASK; } static inline unsigned rb_page_commit(struct buffer_page *bpage) @@ -750,6 +1408,11 @@ static inline unsigned rb_page_commit(struct buffer_page *bpage) return local_read(&bpage->page->commit); } +static inline unsigned long rb_page_entries(struct buffer_page *bpage) +{ + return local_read(&bpage->entries) & RB_WRITE_MASK; +} + /* Size is determined by what has been commited */ static inline unsigned rb_page_size(struct buffer_page *bpage) { @@ -762,58 +1425,17 @@ rb_commit_index(struct ring_buffer_per_cpu *cpu_buffer) return rb_page_commit(cpu_buffer->commit_page); } -static inline unsigned rb_head_size(struct ring_buffer_per_cpu *cpu_buffer) -{ - return rb_page_commit(cpu_buffer->head_page); -} - -/* - * When the tail hits the head and the buffer is in overwrite mode, - * the head jumps to the next page and all content on the previous - * page is discarded. But before doing so, we update the overrun - * variable of the buffer. - */ -static void rb_update_overflow(struct ring_buffer_per_cpu *cpu_buffer) -{ - struct ring_buffer_event *event; - unsigned long head; - - for (head = 0; head < rb_head_size(cpu_buffer); - head += rb_event_length(event)) { - - event = __rb_page_index(cpu_buffer->head_page, head); - if (RB_WARN_ON(cpu_buffer, rb_null_event(event))) - return; - /* Only count data entries */ - if (event->type != RINGBUF_TYPE_DATA) - continue; - cpu_buffer->overrun++; - cpu_buffer->entries--; - } -} - -static inline void rb_inc_page(struct ring_buffer_per_cpu *cpu_buffer, - struct buffer_page **bpage) -{ - struct list_head *p = (*bpage)->list.next; - - if (p == &cpu_buffer->pages) - p = p->next; - - *bpage = list_entry(p, struct buffer_page, list); -} - static inline unsigned rb_event_index(struct ring_buffer_event *event) { unsigned long addr = (unsigned long)event; - return (addr & ~PAGE_MASK) - (PAGE_SIZE - BUF_PAGE_SIZE); + return (addr & ~PAGE_MASK) - BUF_PAGE_HDR_SIZE; } static inline int -rb_is_commit(struct ring_buffer_per_cpu *cpu_buffer, - struct ring_buffer_event *event) +rb_event_is_commit(struct ring_buffer_per_cpu *cpu_buffer, + struct ring_buffer_event *event) { unsigned long addr = (unsigned long)event; unsigned long index; @@ -825,34 +1447,11 @@ rb_is_commit(struct ring_buffer_per_cpu *cpu_buffer, rb_commit_index(cpu_buffer) == index; } -static inline void -rb_set_commit_event(struct ring_buffer_per_cpu *cpu_buffer, - struct ring_buffer_event *event) -{ - unsigned long addr = (unsigned long)event; - unsigned long index; - - index = rb_event_index(event); - addr &= PAGE_MASK; - - while (cpu_buffer->commit_page->page != (void *)addr) { - if (RB_WARN_ON(cpu_buffer, - cpu_buffer->commit_page == cpu_buffer->tail_page)) - return; - cpu_buffer->commit_page->page->commit = - cpu_buffer->commit_page->write; - rb_inc_page(cpu_buffer, &cpu_buffer->commit_page); - cpu_buffer->write_stamp = - cpu_buffer->commit_page->page->time_stamp; - } - - /* Now set the commit to the event's index */ - local_set(&cpu_buffer->commit_page->page->commit, index); -} - -static inline void +static void rb_set_commit_to_write(struct ring_buffer_per_cpu *cpu_buffer) { + unsigned long max_count; + /* * We only race with interrupts and NMIs on this CPU. * If we own the commit event, then we can commit @@ -862,9 +1461,16 @@ rb_set_commit_to_write(struct ring_buffer_per_cpu *cpu_buffer) * assign the commit to the tail. */ again: + max_count = cpu_buffer->buffer->pages * 100; + while (cpu_buffer->commit_page != cpu_buffer->tail_page) { - cpu_buffer->commit_page->page->commit = - cpu_buffer->commit_page->write; + if (RB_WARN_ON(cpu_buffer, !(--max_count))) + return; + if (RB_WARN_ON(cpu_buffer, + rb_is_reader_page(cpu_buffer->tail_page))) + return; + local_set(&cpu_buffer->commit_page->page->commit, + rb_page_write(cpu_buffer->commit_page)); rb_inc_page(cpu_buffer, &cpu_buffer->commit_page); cpu_buffer->write_stamp = cpu_buffer->commit_page->page->time_stamp; @@ -873,8 +1479,12 @@ rb_set_commit_to_write(struct ring_buffer_per_cpu *cpu_buffer) } while (rb_commit_index(cpu_buffer) != rb_page_write(cpu_buffer->commit_page)) { - cpu_buffer->commit_page->page->commit = - cpu_buffer->commit_page->write; + + local_set(&cpu_buffer->commit_page->page->commit, + rb_page_write(cpu_buffer->commit_page)); + RB_WARN_ON(cpu_buffer, + local_read(&cpu_buffer->commit_page->page->commit) & + ~RB_WRITE_MASK); barrier(); } @@ -896,7 +1506,7 @@ static void rb_reset_reader_page(struct ring_buffer_per_cpu *cpu_buffer) cpu_buffer->reader_page->read = 0; } -static inline void rb_inc_iter(struct ring_buffer_iter *iter) +static void rb_inc_iter(struct ring_buffer_iter *iter) { struct ring_buffer_per_cpu *cpu_buffer = iter->cpu_buffer; @@ -907,7 +1517,7 @@ static inline void rb_inc_iter(struct ring_buffer_iter *iter) * to the head page instead of next. */ if (iter->head_page == cpu_buffer->reader_page) - iter->head_page = cpu_buffer->head_page; + iter->head_page = rb_set_head_page(cpu_buffer); else rb_inc_page(cpu_buffer, &iter->head_page); @@ -926,45 +1536,189 @@ static inline void rb_inc_iter(struct ring_buffer_iter *iter) * and with this, we can determine what to place into the * data field. */ -static inline void +static void rb_update_event(struct ring_buffer_event *event, unsigned type, unsigned length) { - event->type = type; + event->type_len = type; switch (type) { case RINGBUF_TYPE_PADDING: - break; - case RINGBUF_TYPE_TIME_EXTEND: - event->len = - (RB_LEN_TIME_EXTEND + (RB_ALIGNMENT-1)) - >> RB_ALIGNMENT_SHIFT; - break; - case RINGBUF_TYPE_TIME_STAMP: - event->len = - (RB_LEN_TIME_STAMP + (RB_ALIGNMENT-1)) - >> RB_ALIGNMENT_SHIFT; break; - case RINGBUF_TYPE_DATA: + case 0: length -= RB_EVNT_HDR_SIZE; - if (length > RB_MAX_SMALL_DATA) { - event->len = 0; + if (length > RB_MAX_SMALL_DATA) event->array[0] = length; - } else - event->len = - (length + (RB_ALIGNMENT-1)) - >> RB_ALIGNMENT_SHIFT; + else + event->type_len = DIV_ROUND_UP(length, RB_ALIGNMENT); break; default: BUG(); } } -static inline unsigned rb_calculate_event_length(unsigned length) +/* + * rb_handle_head_page - writer hit the head page + * + * Returns: +1 to retry page + * 0 to continue + * -1 on error + */ +static int +rb_handle_head_page(struct ring_buffer_per_cpu *cpu_buffer, + struct buffer_page *tail_page, + struct buffer_page *next_page) +{ + struct buffer_page *new_head; + int entries; + int type; + int ret; + + entries = rb_page_entries(next_page); + + /* + * The hard part is here. We need to move the head + * forward, and protect against both readers on + * other CPUs and writers coming in via interrupts. + */ + type = rb_head_page_set_update(cpu_buffer, next_page, tail_page, + RB_PAGE_HEAD); + + /* + * type can be one of four: + * NORMAL - an interrupt already moved it for us + * HEAD - we are the first to get here. + * UPDATE - we are the interrupt interrupting + * a current move. + * MOVED - a reader on another CPU moved the next + * pointer to its reader page. Give up + * and try again. + */ + + switch (type) { + case RB_PAGE_HEAD: + /* + * We changed the head to UPDATE, thus + * it is our responsibility to update + * the counters. + */ + local_add(entries, &cpu_buffer->overrun); + + /* + * The entries will be zeroed out when we move the + * tail page. + */ + + /* still more to do */ + break; + + case RB_PAGE_UPDATE: + /* + * This is an interrupt that interrupt the + * previous update. Still more to do. + */ + break; + case RB_PAGE_NORMAL: + /* + * An interrupt came in before the update + * and processed this for us. + * Nothing left to do. + */ + return 1; + case RB_PAGE_MOVED: + /* + * The reader is on another CPU and just did + * a swap with our next_page. + * Try again. + */ + return 1; + default: + RB_WARN_ON(cpu_buffer, 1); /* WTF??? */ + return -1; + } + + /* + * Now that we are here, the old head pointer is + * set to UPDATE. This will keep the reader from + * swapping the head page with the reader page. + * The reader (on another CPU) will spin till + * we are finished. + * + * We just need to protect against interrupts + * doing the job. We will set the next pointer + * to HEAD. After that, we set the old pointer + * to NORMAL, but only if it was HEAD before. + * otherwise we are an interrupt, and only + * want the outer most commit to reset it. + */ + new_head = next_page; + rb_inc_page(cpu_buffer, &new_head); + + ret = rb_head_page_set_head(cpu_buffer, new_head, next_page, + RB_PAGE_NORMAL); + + /* + * Valid returns are: + * HEAD - an interrupt came in and already set it. + * NORMAL - One of two things: + * 1) We really set it. + * 2) A bunch of interrupts came in and moved + * the page forward again. + */ + switch (ret) { + case RB_PAGE_HEAD: + case RB_PAGE_NORMAL: + /* OK */ + break; + default: + RB_WARN_ON(cpu_buffer, 1); + return -1; + } + + /* + * It is possible that an interrupt came in, + * set the head up, then more interrupts came in + * and moved it again. When we get back here, + * the page would have been set to NORMAL but we + * just set it back to HEAD. + * + * How do you detect this? Well, if that happened + * the tail page would have moved. + */ + if (ret == RB_PAGE_NORMAL) { + /* + * If the tail had moved passed next, then we need + * to reset the pointer. + */ + if (cpu_buffer->tail_page != tail_page && + cpu_buffer->tail_page != next_page) + rb_head_page_set_normal(cpu_buffer, new_head, + next_page, + RB_PAGE_HEAD); + } + + /* + * If this was the outer most commit (the one that + * changed the original pointer from HEAD to UPDATE), + * then it is up to us to reset it to NORMAL. + */ + if (type == RB_PAGE_HEAD) { + ret = rb_head_page_set_normal(cpu_buffer, next_page, + tail_page, + RB_PAGE_UPDATE); + if (RB_WARN_ON(cpu_buffer, + ret != RB_PAGE_UPDATE)) + return -1; + } + + return 0; +} + +static unsigned rb_calculate_event_length(unsigned length) { struct ring_buffer_event event; /* Used only for sizeof array */ @@ -981,133 +1735,236 @@ static inline unsigned rb_calculate_event_length(unsigned length) return length; } -static struct ring_buffer_event * -__rb_reserve_next(struct ring_buffer_per_cpu *cpu_buffer, - unsigned type, unsigned long length, u64 *ts) +static inline void +rb_reset_tail(struct ring_buffer_per_cpu *cpu_buffer, + struct buffer_page *tail_page, + unsigned long tail, unsigned long length) { - struct buffer_page *tail_page, *head_page, *reader_page, *commit_page; - unsigned long tail, write; - struct ring_buffer *buffer = cpu_buffer->buffer; struct ring_buffer_event *event; - unsigned long flags; - commit_page = cpu_buffer->commit_page; - /* we just need to protect against interrupts */ - barrier(); - tail_page = cpu_buffer->tail_page; - write = local_add_return(length, &tail_page->write); - tail = write - length; + /* + * Only the event that crossed the page boundary + * must fill the old tail_page with padding. + */ + if (tail >= BUF_PAGE_SIZE) { + local_sub(length, &tail_page->write); + return; + } - /* See if we shot pass the end of this buffer page */ - if (write > BUF_PAGE_SIZE) { - struct buffer_page *next_page = tail_page; + event = __rb_page_index(tail_page, tail); + kmemcheck_annotate_bitfield(event, bitfield); - local_irq_save(flags); - __raw_spin_lock(&cpu_buffer->lock); + /* + * If this event is bigger than the minimum size, then + * we need to be careful that we don't subtract the + * write counter enough to allow another writer to slip + * in on this page. + * We put in a discarded commit instead, to make sure + * that this space is not used again. + * + * If we are less than the minimum size, we don't need to + * worry about it. + */ + if (tail > (BUF_PAGE_SIZE - RB_EVNT_MIN_SIZE)) { + /* No room for any events */ - rb_inc_page(cpu_buffer, &next_page); + /* Mark the rest of the page with padding */ + rb_event_set_padding(event); - head_page = cpu_buffer->head_page; - reader_page = cpu_buffer->reader_page; + /* Set the write back to the previous setting */ + local_sub(length, &tail_page->write); + return; + } - /* we grabbed the lock before incrementing */ - if (RB_WARN_ON(cpu_buffer, next_page == reader_page)) - goto out_unlock; + /* Put in a discarded event */ + event->array[0] = (BUF_PAGE_SIZE - tail) - RB_EVNT_HDR_SIZE; + event->type_len = RINGBUF_TYPE_PADDING; + /* time delta must be non zero */ + event->time_delta = 1; - /* - * If for some reason, we had an interrupt storm that made - * it all the way around the buffer, bail, and warn - * about it. - */ - if (unlikely(next_page == commit_page)) { - WARN_ON_ONCE(1); - goto out_unlock; - } + /* Set write to end of buffer */ + length = (tail + length) - BUF_PAGE_SIZE; + local_sub(length, &tail_page->write); +} - if (next_page == head_page) { - if (!(buffer->flags & RB_FL_OVERWRITE)) - goto out_unlock; +static struct ring_buffer_event * +rb_move_tail(struct ring_buffer_per_cpu *cpu_buffer, + unsigned long length, unsigned long tail, + struct buffer_page *tail_page, u64 *ts) +{ + struct buffer_page *commit_page = cpu_buffer->commit_page; + struct ring_buffer *buffer = cpu_buffer->buffer; + struct buffer_page *next_page; + int ret; - /* tail_page has not moved yet? */ - if (tail_page == cpu_buffer->tail_page) { - /* count overflows */ - rb_update_overflow(cpu_buffer); + next_page = tail_page; - rb_inc_page(cpu_buffer, &head_page); - cpu_buffer->head_page = head_page; - cpu_buffer->head_page->read = 0; - } - } + rb_inc_page(cpu_buffer, &next_page); + + /* + * If for some reason, we had an interrupt storm that made + * it all the way around the buffer, bail, and warn + * about it. + */ + if (unlikely(next_page == commit_page)) { + local_inc(&cpu_buffer->commit_overrun); + goto out_reset; + } + + /* + * This is where the fun begins! + * + * We are fighting against races between a reader that + * could be on another CPU trying to swap its reader + * page with the buffer head. + * + * We are also fighting against interrupts coming in and + * moving the head or tail on us as well. + * + * If the next page is the head page then we have filled + * the buffer, unless the commit page is still on the + * reader page. + */ + if (rb_is_head_page(cpu_buffer, next_page, &tail_page->list)) { /* - * If the tail page is still the same as what we think - * it is, then it is up to us to update the tail - * pointer. + * If the commit is not on the reader page, then + * move the header page. */ - if (tail_page == cpu_buffer->tail_page) { - local_set(&next_page->write, 0); - local_set(&next_page->page->commit, 0); - cpu_buffer->tail_page = next_page; - - /* reread the time stamp */ - *ts = ring_buffer_time_stamp(cpu_buffer->cpu); - cpu_buffer->tail_page->page->time_stamp = *ts; + if (!rb_is_reader_page(cpu_buffer->commit_page)) { + /* + * If we are not in overwrite mode, + * this is easy, just stop here. + */ + if (!(buffer->flags & RB_FL_OVERWRITE)) + goto out_reset; + + ret = rb_handle_head_page(cpu_buffer, + tail_page, + next_page); + if (ret < 0) + goto out_reset; + if (ret) + goto out_again; + } else { + /* + * We need to be careful here too. The + * commit page could still be on the reader + * page. We could have a small buffer, and + * have filled up the buffer with events + * from interrupts and such, and wrapped. + * + * Note, if the tail page is also the on the + * reader_page, we let it move out. + */ + if (unlikely((cpu_buffer->commit_page != + cpu_buffer->tail_page) && + (cpu_buffer->commit_page == + cpu_buffer->reader_page))) { + local_inc(&cpu_buffer->commit_overrun); + goto out_reset; + } } + } + ret = rb_tail_page_update(cpu_buffer, tail_page, next_page); + if (ret) { /* - * The actual tail page has moved forward. + * Nested commits always have zero deltas, so + * just reread the time stamp */ - if (tail < BUF_PAGE_SIZE) { - /* Mark the rest of the page with padding */ - event = __rb_page_index(tail_page, tail); - event->type = RINGBUF_TYPE_PADDING; - } + *ts = rb_time_stamp(buffer); + next_page->page->time_stamp = *ts; + } - if (tail <= BUF_PAGE_SIZE) - /* Set the write back to the previous setting */ - local_set(&tail_page->write, tail); + out_again: - /* - * If this was a commit entry that failed, - * increment that too - */ - if (tail_page == cpu_buffer->commit_page && - tail == rb_commit_index(cpu_buffer)) { - rb_set_commit_to_write(cpu_buffer); - } + rb_reset_tail(cpu_buffer, tail_page, tail, length); - __raw_spin_unlock(&cpu_buffer->lock); - local_irq_restore(flags); + /* fail and let the caller try again */ + return ERR_PTR(-EAGAIN); - /* fail and let the caller try again */ - return ERR_PTR(-EAGAIN); - } + out_reset: + /* reset write */ + rb_reset_tail(cpu_buffer, tail_page, tail, length); - /* We reserved something on the buffer */ + return NULL; +} - if (RB_WARN_ON(cpu_buffer, write > BUF_PAGE_SIZE)) - return NULL; +static struct ring_buffer_event * +__rb_reserve_next(struct ring_buffer_per_cpu *cpu_buffer, + unsigned type, unsigned long length, u64 *ts) +{ + struct buffer_page *tail_page; + struct ring_buffer_event *event; + unsigned long tail, write; + + tail_page = cpu_buffer->tail_page; + write = local_add_return(length, &tail_page->write); + + /* set write to only the index of the write */ + write &= RB_WRITE_MASK; + tail = write - length; + + /* See if we shot pass the end of this buffer page */ + if (write > BUF_PAGE_SIZE) + return rb_move_tail(cpu_buffer, length, tail, + tail_page, ts); + + /* We reserved something on the buffer */ event = __rb_page_index(tail_page, tail); + kmemcheck_annotate_bitfield(event, bitfield); rb_update_event(event, type, length); + /* The passed in type is zero for DATA */ + if (likely(!type)) + local_inc(&tail_page->entries); + /* - * If this is a commit and the tail is zero, then update - * this page's time stamp. + * If this is the first commit on the page, then update + * its timestamp. */ - if (!tail && rb_is_commit(cpu_buffer, event)) - cpu_buffer->commit_page->page->time_stamp = *ts; + if (!tail) + tail_page->page->time_stamp = *ts; return event; +} - out_unlock: - /* reset write */ - if (tail <= BUF_PAGE_SIZE) - local_set(&tail_page->write, tail); +static inline int +rb_try_to_discard(struct ring_buffer_per_cpu *cpu_buffer, + struct ring_buffer_event *event) +{ + unsigned long new_index, old_index; + struct buffer_page *bpage; + unsigned long index; + unsigned long addr; - __raw_spin_unlock(&cpu_buffer->lock); - local_irq_restore(flags); - return NULL; + new_index = rb_event_index(event); + old_index = new_index + rb_event_length(event); + addr = (unsigned long)event; + addr &= PAGE_MASK; + + bpage = cpu_buffer->tail_page; + + if (bpage->page == (void *)addr && rb_page_write(bpage) == old_index) { + unsigned long write_mask = + local_read(&bpage->write) & ~RB_WRITE_MASK; + /* + * This is on the tail page. It is possible that + * a write could come in and move the tail page + * and write to the next page. That is fine + * because we just shorten what is on this page. + */ + old_index += write_mask; + new_index += write_mask; + index = local_cmpxchg(&bpage->write, old_index, new_index); + if (index == old_index) + return 1; + } + + /* could not discard */ + return 0; } static int @@ -1142,26 +1999,33 @@ rb_add_time_stamp(struct ring_buffer_per_cpu *cpu_buffer, return -EAGAIN; /* Only a commited time event can update the write stamp */ - if (rb_is_commit(cpu_buffer, event)) { + if (rb_event_is_commit(cpu_buffer, event)) { /* - * If this is the first on the page, then we need to - * update the page itself, and just put in a zero. + * If this is the first on the page, then it was + * updated with the page itself. Try to discard it + * and if we can't just make it zero. */ if (rb_event_index(event)) { event->time_delta = *delta & TS_MASK; event->array[0] = *delta >> TS_SHIFT; } else { - cpu_buffer->commit_page->page->time_stamp = *ts; - event->time_delta = 0; - event->array[0] = 0; + /* try to discard, since we do not need this */ + if (!rb_try_to_discard(cpu_buffer, event)) { + /* nope, just zero it */ + event->time_delta = 0; + event->array[0] = 0; + } } cpu_buffer->write_stamp = *ts; /* let the caller know this was the commit */ ret = 1; } else { - /* Darn, this is just wasted space */ - event->time_delta = 0; - event->array[0] = 0; + /* Try to discard the event */ + if (!rb_try_to_discard(cpu_buffer, event)) { + /* Darn, this is just wasted space */ + event->time_delta = 0; + event->array[0] = 0; + } ret = 0; } @@ -1170,15 +2034,72 @@ rb_add_time_stamp(struct ring_buffer_per_cpu *cpu_buffer, return ret; } +static void rb_start_commit(struct ring_buffer_per_cpu *cpu_buffer) +{ + local_inc(&cpu_buffer->committing); + local_inc(&cpu_buffer->commits); +} + +static void rb_end_commit(struct ring_buffer_per_cpu *cpu_buffer) +{ + unsigned long commits; + + if (RB_WARN_ON(cpu_buffer, + !local_read(&cpu_buffer->committing))) + return; + + again: + commits = local_read(&cpu_buffer->commits); + /* synchronize with interrupts */ + barrier(); + if (local_read(&cpu_buffer->committing) == 1) + rb_set_commit_to_write(cpu_buffer); + + local_dec(&cpu_buffer->committing); + + /* synchronize with interrupts */ + barrier(); + + /* + * Need to account for interrupts coming in between the + * updating of the commit page and the clearing of the + * committing counter. + */ + if (unlikely(local_read(&cpu_buffer->commits) != commits) && + !local_read(&cpu_buffer->committing)) { + local_inc(&cpu_buffer->committing); + goto again; + } +} + static struct ring_buffer_event * -rb_reserve_next_event(struct ring_buffer_per_cpu *cpu_buffer, - unsigned type, unsigned long length) +rb_reserve_next_event(struct ring_buffer *buffer, + struct ring_buffer_per_cpu *cpu_buffer, + unsigned long length) { struct ring_buffer_event *event; - u64 ts, delta; + u64 ts, delta = 0; int commit = 0; int nr_loops = 0; + rb_start_commit(cpu_buffer); + +#ifdef CONFIG_RING_BUFFER_ALLOW_SWAP + /* + * Due to the ability to swap a cpu buffer from a buffer + * it is possible it was swapped before we committed. + * (committing stops a swap). We check for it here and + * if it happened, we have to fail the write. + */ + barrier(); + if (unlikely(ACCESS_ONCE(cpu_buffer->buffer) != buffer)) { + local_dec(&cpu_buffer->committing); + local_dec(&cpu_buffer->commits); + return NULL; + } +#endif + + length = rb_calculate_event_length(length); again: /* * We allow for interrupts to reenter here and do a trace. @@ -1190,9 +2111,9 @@ rb_reserve_next_event(struct ring_buffer_per_cpu *cpu_buffer, * Bail! */ if (RB_WARN_ON(cpu_buffer, ++nr_loops > 1000)) - return NULL; + goto out_fail; - ts = ring_buffer_time_stamp(cpu_buffer->cpu); + ts = rb_time_stamp(cpu_buffer->buffer); /* * Only the first commit can update the timestamp. @@ -1202,70 +2123,99 @@ rb_reserve_next_event(struct ring_buffer_per_cpu *cpu_buffer, * also be made. But only the entry that did the actual * commit will be something other than zero. */ - if (cpu_buffer->tail_page == cpu_buffer->commit_page && - rb_page_write(cpu_buffer->tail_page) == - rb_commit_index(cpu_buffer)) { + if (likely(cpu_buffer->tail_page == cpu_buffer->commit_page && + rb_page_write(cpu_buffer->tail_page) == + rb_commit_index(cpu_buffer))) { + u64 diff; - delta = ts - cpu_buffer->write_stamp; + diff = ts - cpu_buffer->write_stamp; - /* make sure this delta is calculated here */ + /* make sure this diff is calculated here */ barrier(); /* Did the write stamp get updated already? */ if (unlikely(ts < cpu_buffer->write_stamp)) - delta = 0; + goto get_event; - if (test_time_stamp(delta)) { + delta = diff; + if (unlikely(test_time_stamp(delta))) { commit = rb_add_time_stamp(cpu_buffer, &ts, &delta); - if (commit == -EBUSY) - return NULL; + goto out_fail; if (commit == -EAGAIN) goto again; RB_WARN_ON(cpu_buffer, commit < 0); } - } else - /* Non commits have zero deltas */ - delta = 0; + } - event = __rb_reserve_next(cpu_buffer, type, length, &ts); - if (PTR_ERR(event) == -EAGAIN) + get_event: + event = __rb_reserve_next(cpu_buffer, 0, length, &ts); + if (unlikely(PTR_ERR(event) == -EAGAIN)) goto again; - if (!event) { - if (unlikely(commit)) - /* - * Ouch! We needed a timestamp and it was commited. But - * we didn't get our event reserved. - */ - rb_set_commit_to_write(cpu_buffer); - return NULL; - } + if (!event) + goto out_fail; - /* - * If the timestamp was commited, make the commit our entry - * now so that we will update it when needed. - */ - if (commit) - rb_set_commit_event(cpu_buffer, event); - else if (!rb_is_commit(cpu_buffer, event)) + if (!rb_event_is_commit(cpu_buffer, event)) delta = 0; event->time_delta = delta; return event; + + out_fail: + rb_end_commit(cpu_buffer); + return NULL; } +#ifdef CONFIG_TRACING + +#define TRACE_RECURSIVE_DEPTH 16 + +static int trace_recursive_lock(void) +{ + current->trace_recursion++; + + if (likely(current->trace_recursion < TRACE_RECURSIVE_DEPTH)) + return 0; + + /* Disable all tracing before we do anything else */ + tracing_off_permanent(); + + printk_once(KERN_WARNING "Tracing recursion: depth[%ld]:" + "HC[%lu]:SC[%lu]:NMI[%lu]\n", + current->trace_recursion, + hardirq_count() >> HARDIRQ_SHIFT, + softirq_count() >> SOFTIRQ_SHIFT, + in_nmi()); + + WARN_ON_ONCE(1); + return -1; +} + +static void trace_recursive_unlock(void) +{ + WARN_ON_ONCE(!current->trace_recursion); + + current->trace_recursion--; +} + +#else + +#define trace_recursive_lock() (0) +#define trace_recursive_unlock() do { } while (0) + +#endif + static DEFINE_PER_CPU(int, rb_need_resched); /** * ring_buffer_lock_reserve - reserve a part of the buffer * @buffer: the ring buffer to reserve from * @length: the length of the data to reserve (excluding event header) - * @flags: a pointer to save the interrupt flags * * Returns a reseverd event on the ring buffer to copy directly to. * The user of this interface will need to get the body to write into @@ -1278,9 +2228,7 @@ static DEFINE_PER_CPU(int, rb_need_resched); * If NULL is returned, then nothing has been allocated or locked. */ struct ring_buffer_event * -ring_buffer_lock_reserve(struct ring_buffer *buffer, - unsigned long length, - unsigned long *flags) +ring_buffer_lock_reserve(struct ring_buffer *buffer, unsigned long length) { struct ring_buffer_per_cpu *cpu_buffer; struct ring_buffer_event *event; @@ -1295,6 +2243,9 @@ ring_buffer_lock_reserve(struct ring_buffer *buffer, /* If we are tracing schedule, we don't want to recurse */ resched = ftrace_preempt_disable(); + if (trace_recursive_lock()) + goto out_nocheck; + cpu = raw_smp_processor_id(); if (!cpumask_test_cpu(cpu, buffer->cpumask)) @@ -1305,11 +2256,10 @@ ring_buffer_lock_reserve(struct ring_buffer *buffer, if (atomic_read(&cpu_buffer->record_disabled)) goto out; - length = rb_calculate_event_length(length); - if (length > BUF_PAGE_SIZE) + if (length > BUF_MAX_DATA_SIZE) goto out; - event = rb_reserve_next_event(cpu_buffer, RINGBUF_TYPE_DATA, length); + event = rb_reserve_next_event(buffer, cpu_buffer, length); if (!event) goto out; @@ -1324,38 +2274,45 @@ ring_buffer_lock_reserve(struct ring_buffer *buffer, return event; out: + trace_recursive_unlock(); + + out_nocheck: ftrace_preempt_enable(resched); return NULL; } EXPORT_SYMBOL_GPL(ring_buffer_lock_reserve); -static void rb_commit(struct ring_buffer_per_cpu *cpu_buffer, +static void +rb_update_write_stamp(struct ring_buffer_per_cpu *cpu_buffer, struct ring_buffer_event *event) { - cpu_buffer->entries++; - - /* Only process further if we own the commit */ - if (!rb_is_commit(cpu_buffer, event)) - return; - - cpu_buffer->write_stamp += event->time_delta; + /* + * The event first in the commit queue updates the + * time stamp. + */ + if (rb_event_is_commit(cpu_buffer, event)) + cpu_buffer->write_stamp += event->time_delta; +} - rb_set_commit_to_write(cpu_buffer); +static void rb_commit(struct ring_buffer_per_cpu *cpu_buffer, + struct ring_buffer_event *event) +{ + local_inc(&cpu_buffer->entries); + rb_update_write_stamp(cpu_buffer, event); + rb_end_commit(cpu_buffer); } /** * ring_buffer_unlock_commit - commit a reserved * @buffer: The buffer to commit to * @event: The event pointer to commit. - * @flags: the interrupt flags received from ring_buffer_lock_reserve. * * This commits the data to the ring buffer, and releases any locks held. * * Must be paired with ring_buffer_lock_reserve. */ int ring_buffer_unlock_commit(struct ring_buffer *buffer, - struct ring_buffer_event *event, - unsigned long flags) + struct ring_buffer_event *event) { struct ring_buffer_per_cpu *cpu_buffer; int cpu = raw_smp_processor_id(); @@ -1364,6 +2321,8 @@ int ring_buffer_unlock_commit(struct ring_buffer *buffer, rb_commit(cpu_buffer, event); + trace_recursive_unlock(); + /* * Only the last preempt count needs to restore preemption. */ @@ -1376,6 +2335,119 @@ int ring_buffer_unlock_commit(struct ring_buffer *buffer, } EXPORT_SYMBOL_GPL(ring_buffer_unlock_commit); +static inline void rb_event_discard(struct ring_buffer_event *event) +{ + /* array[0] holds the actual length for the discarded event */ + event->array[0] = rb_event_data_length(event) - RB_EVNT_HDR_SIZE; + event->type_len = RINGBUF_TYPE_PADDING; + /* time delta must be non zero */ + if (!event->time_delta) + event->time_delta = 1; +} + +/* + * Decrement the entries to the page that an event is on. + * The event does not even need to exist, only the pointer + * to the page it is on. This may only be called before the commit + * takes place. + */ +static inline void +rb_decrement_entry(struct ring_buffer_per_cpu *cpu_buffer, + struct ring_buffer_event *event) +{ + unsigned long addr = (unsigned long)event; + struct buffer_page *bpage = cpu_buffer->commit_page; + struct buffer_page *start; + + addr &= PAGE_MASK; + + /* Do the likely case first */ + if (likely(bpage->page == (void *)addr)) { + local_dec(&bpage->entries); + return; + } + + /* + * Because the commit page may be on the reader page we + * start with the next page and check the end loop there. + */ + rb_inc_page(cpu_buffer, &bpage); + start = bpage; + do { + if (bpage->page == (void *)addr) { + local_dec(&bpage->entries); + return; + } + rb_inc_page(cpu_buffer, &bpage); + } while (bpage != start); + + /* commit not part of this buffer?? */ + RB_WARN_ON(cpu_buffer, 1); +} + +/** + * ring_buffer_commit_discard - discard an event that has not been committed + * @buffer: the ring buffer + * @event: non committed event to discard + * + * Sometimes an event that is in the ring buffer needs to be ignored. + * This function lets the user discard an event in the ring buffer + * and then that event will not be read later. + * + * This function only works if it is called before the the item has been + * committed. It will try to free the event from the ring buffer + * if another event has not been added behind it. + * + * If another event has been added behind it, it will set the event + * up as discarded, and perform the commit. + * + * If this function is called, do not call ring_buffer_unlock_commit on + * the event. + */ +void ring_buffer_discard_commit(struct ring_buffer *buffer, + struct ring_buffer_event *event) +{ + struct ring_buffer_per_cpu *cpu_buffer; + int cpu; + + /* The event is discarded regardless */ + rb_event_discard(event); + + cpu = smp_processor_id(); + cpu_buffer = buffer->buffers[cpu]; + + /* + * This must only be called if the event has not been + * committed yet. Thus we can assume that preemption + * is still disabled. + */ + RB_WARN_ON(buffer, !local_read(&cpu_buffer->committing)); + + rb_decrement_entry(cpu_buffer, event); + if (rb_try_to_discard(cpu_buffer, event)) + goto out; + + /* + * The commit is still visible by the reader, so we + * must still update the timestamp. + */ + rb_update_write_stamp(cpu_buffer, event); + out: + rb_end_commit(cpu_buffer); + + trace_recursive_unlock(); + + /* + * Only the last preempt count needs to restore preemption. + */ + if (preempt_count() == 1) + ftrace_preempt_enable(per_cpu(rb_need_resched, cpu)); + else + preempt_enable_no_resched_notrace(); + +} +EXPORT_SYMBOL_GPL(ring_buffer_discard_commit); + /** * ring_buffer_write - write data to the buffer without reserving * @buffer: The ring buffer to write to. @@ -1395,7 +2467,6 @@ int ring_buffer_write(struct ring_buffer *buffer, { struct ring_buffer_per_cpu *cpu_buffer; struct ring_buffer_event *event; - unsigned long event_length; void *body; int ret = -EBUSY; int cpu, resched; @@ -1418,9 +2489,10 @@ int ring_buffer_write(struct ring_buffer *buffer, if (atomic_read(&cpu_buffer->record_disabled)) goto out; - event_length = rb_calculate_event_length(length); - event = rb_reserve_next_event(cpu_buffer, - RINGBUF_TYPE_DATA, event_length); + if (length > BUF_MAX_DATA_SIZE) + goto out; + + event = rb_reserve_next_event(buffer, cpu_buffer, length); if (!event) goto out; @@ -1438,12 +2510,16 @@ int ring_buffer_write(struct ring_buffer *buffer, } EXPORT_SYMBOL_GPL(ring_buffer_write); -static inline int rb_per_cpu_empty(struct ring_buffer_per_cpu *cpu_buffer) +static int rb_per_cpu_empty(struct ring_buffer_per_cpu *cpu_buffer) { struct buffer_page *reader = cpu_buffer->reader_page; - struct buffer_page *head = cpu_buffer->head_page; + struct buffer_page *head = rb_set_head_page(cpu_buffer); struct buffer_page *commit = cpu_buffer->commit_page; + /* In case of error, head will be NULL */ + if (unlikely(!head)) + return 1; + return reader->read == rb_page_commit(reader) && (commit == reader || (commit == head && @@ -1528,12 +2604,16 @@ EXPORT_SYMBOL_GPL(ring_buffer_record_enable_cpu); unsigned long ring_buffer_entries_cpu(struct ring_buffer *buffer, int cpu) { struct ring_buffer_per_cpu *cpu_buffer; + unsigned long ret; if (!cpumask_test_cpu(cpu, buffer->cpumask)) return 0; cpu_buffer = buffer->buffers[cpu]; - return cpu_buffer->entries; + ret = (local_read(&cpu_buffer->entries) - local_read(&cpu_buffer->overrun)) + - cpu_buffer->read; + + return ret; } EXPORT_SYMBOL_GPL(ring_buffer_entries_cpu); @@ -1545,16 +2625,40 @@ EXPORT_SYMBOL_GPL(ring_buffer_entries_cpu); unsigned long ring_buffer_overrun_cpu(struct ring_buffer *buffer, int cpu) { struct ring_buffer_per_cpu *cpu_buffer; + unsigned long ret; if (!cpumask_test_cpu(cpu, buffer->cpumask)) return 0; cpu_buffer = buffer->buffers[cpu]; - return cpu_buffer->overrun; + ret = local_read(&cpu_buffer->overrun); + + return ret; } EXPORT_SYMBOL_GPL(ring_buffer_overrun_cpu); /** + * ring_buffer_commit_overrun_cpu - get the number of overruns caused by commits + * @buffer: The ring buffer + * @cpu: The per CPU buffer to get the number of overruns from + */ +unsigned long +ring_buffer_commit_overrun_cpu(struct ring_buffer *buffer, int cpu) +{ + struct ring_buffer_per_cpu *cpu_buffer; + unsigned long ret; + + if (!cpumask_test_cpu(cpu, buffer->cpumask)) + return 0; + + cpu_buffer = buffer->buffers[cpu]; + ret = local_read(&cpu_buffer->commit_overrun); + + return ret; +} +EXPORT_SYMBOL_GPL(ring_buffer_commit_overrun_cpu); + +/** * ring_buffer_entries - get the number of entries in a buffer * @buffer: The ring buffer * @@ -1570,7 +2674,8 @@ unsigned long ring_buffer_entries(struct ring_buffer *buffer) /* if you care about this being correct, lock the buffer */ for_each_buffer_cpu(buffer, cpu) { cpu_buffer = buffer->buffers[cpu]; - entries += cpu_buffer->entries; + entries += (local_read(&cpu_buffer->entries) - + local_read(&cpu_buffer->overrun)) - cpu_buffer->read; } return entries; @@ -1578,7 +2683,7 @@ unsigned long ring_buffer_entries(struct ring_buffer *buffer) EXPORT_SYMBOL_GPL(ring_buffer_entries); /** - * ring_buffer_overrun_cpu - get the number of overruns in buffer + * ring_buffer_overruns - get the number of overruns in buffer * @buffer: The ring buffer * * Returns the total number of overruns in the ring buffer @@ -1593,7 +2698,7 @@ unsigned long ring_buffer_overruns(struct ring_buffer *buffer) /* if you care about this being correct, lock the buffer */ for_each_buffer_cpu(buffer, cpu) { cpu_buffer = buffer->buffers[cpu]; - overruns += cpu_buffer->overrun; + overruns += local_read(&cpu_buffer->overrun); } return overruns; @@ -1606,8 +2711,10 @@ static void rb_iter_reset(struct ring_buffer_iter *iter) /* Iterator usage is expected to have record disabled */ if (list_empty(&cpu_buffer->reader_page->list)) { - iter->head_page = cpu_buffer->head_page; - iter->head = cpu_buffer->head_page->read; + iter->head_page = rb_set_head_page(cpu_buffer); + if (unlikely(!iter->head_page)) + return; + iter->head = iter->head_page->read; } else { iter->head_page = cpu_buffer->reader_page; iter->head = cpu_buffer->reader_page->read; @@ -1627,9 +2734,14 @@ static void rb_iter_reset(struct ring_buffer_iter *iter) */ void ring_buffer_iter_reset(struct ring_buffer_iter *iter) { - struct ring_buffer_per_cpu *cpu_buffer = iter->cpu_buffer; + struct ring_buffer_per_cpu *cpu_buffer; unsigned long flags; + if (!iter) + return; + + cpu_buffer = iter->cpu_buffer; + spin_lock_irqsave(&cpu_buffer->reader_lock, flags); rb_iter_reset(iter); spin_unlock_irqrestore(&cpu_buffer->reader_lock, flags); @@ -1657,7 +2769,7 @@ rb_update_read_stamp(struct ring_buffer_per_cpu *cpu_buffer, { u64 delta; - switch (event->type) { + switch (event->type_len) { case RINGBUF_TYPE_PADDING: return; @@ -1688,7 +2800,7 @@ rb_update_iter_read_stamp(struct ring_buffer_iter *iter, { u64 delta; - switch (event->type) { + switch (event->type_len) { case RINGBUF_TYPE_PADDING: return; @@ -1719,6 +2831,7 @@ rb_get_reader_page(struct ring_buffer_per_cpu *cpu_buffer) struct buffer_page *reader = NULL; unsigned long flags; int nr_loops = 0; + int ret; local_irq_save(flags); __raw_spin_lock(&cpu_buffer->lock); @@ -1752,29 +2865,56 @@ rb_get_reader_page(struct ring_buffer_per_cpu *cpu_buffer) goto out; /* - * Splice the empty reader page into the list around the head. * Reset the reader page to size zero. */ + local_set(&cpu_buffer->reader_page->write, 0); + local_set(&cpu_buffer->reader_page->entries, 0); + local_set(&cpu_buffer->reader_page->page->commit, 0); - reader = cpu_buffer->head_page; + spin: + /* + * Splice the empty reader page into the list around the head. + */ + reader = rb_set_head_page(cpu_buffer); cpu_buffer->reader_page->list.next = reader->list.next; cpu_buffer->reader_page->list.prev = reader->list.prev; - local_set(&cpu_buffer->reader_page->write, 0); - local_set(&cpu_buffer->reader_page->page->commit, 0); + /* + * cpu_buffer->pages just needs to point to the buffer, it + * has no specific buffer page to point to. Lets move it out + * of our way so we don't accidently swap it. + */ + cpu_buffer->pages = reader->list.prev; - /* Make the reader page now replace the head */ - reader->list.prev->next = &cpu_buffer->reader_page->list; - reader->list.next->prev = &cpu_buffer->reader_page->list; + /* The reader page will be pointing to the new head */ + rb_set_list_to_head(cpu_buffer, &cpu_buffer->reader_page->list); /* - * If the tail is on the reader, then we must set the head - * to the inserted page, otherwise we set it one before. + * Here's the tricky part. + * + * We need to move the pointer past the header page. + * But we can only do that if a writer is not currently + * moving it. The page before the header page has the + * flag bit '1' set if it is pointing to the page we want. + * but if the writer is in the process of moving it + * than it will be '2' or already moved '0'. */ - cpu_buffer->head_page = cpu_buffer->reader_page; - if (cpu_buffer->commit_page != reader) - rb_inc_page(cpu_buffer, &cpu_buffer->head_page); + ret = rb_head_page_replace(reader, cpu_buffer->reader_page); + + /* + * If we did not convert it, then we must try again. + */ + if (!ret) + goto spin; + + /* + * Yeah! We succeeded in replacing the page. + * + * Now make the new head point back to the reader page. + */ + reader->list.next->prev = &cpu_buffer->reader_page->list; + rb_inc_page(cpu_buffer, &cpu_buffer->head_page); /* Finally update the reader page to the new head */ cpu_buffer->reader_page = reader; @@ -1803,8 +2943,8 @@ static void rb_advance_reader(struct ring_buffer_per_cpu *cpu_buffer) event = rb_reader_event(cpu_buffer); - if (event->type == RINGBUF_TYPE_DATA) - cpu_buffer->entries--; + if (event->type_len <= RINGBUF_TYPE_DATA_TYPE_LEN_MAX) + cpu_buffer->read++; rb_update_read_stamp(cpu_buffer, event); @@ -1826,8 +2966,8 @@ static void rb_advance_iter(struct ring_buffer_iter *iter) * Check if we are at the end of the buffer. */ if (iter->head >= rb_page_size(iter->head_page)) { - if (RB_WARN_ON(buffer, - iter->head_page == cpu_buffer->commit_page)) + /* discarded commits can make the page empty */ + if (iter->head_page == cpu_buffer->commit_page) return; rb_inc_iter(iter); return; @@ -1857,28 +2997,20 @@ static void rb_advance_iter(struct ring_buffer_iter *iter) } static struct ring_buffer_event * -rb_buffer_peek(struct ring_buffer *buffer, int cpu, u64 *ts) +rb_buffer_peek(struct ring_buffer_per_cpu *cpu_buffer, u64 *ts) { - struct ring_buffer_per_cpu *cpu_buffer; struct ring_buffer_event *event; struct buffer_page *reader; int nr_loops = 0; - if (!cpumask_test_cpu(cpu, buffer->cpumask)) - return NULL; - - cpu_buffer = buffer->buffers[cpu]; - again: /* * We repeat when a timestamp is encountered. It is possible * to get multiple timestamps from an interrupt entering just - * as one timestamp is about to be written. The max times - * that this can happen is the number of nested interrupts we - * can have. Nesting 10 deep of interrupts is clearly - * an anomaly. + * as one timestamp is about to be written, or from discarded + * commits. The most that we can have is the number on a single page. */ - if (RB_WARN_ON(cpu_buffer, ++nr_loops > 10)) + if (RB_WARN_ON(cpu_buffer, ++nr_loops > RB_TIMESTAMPS_PER_PAGE)) return NULL; reader = rb_get_reader_page(cpu_buffer); @@ -1887,11 +3019,19 @@ rb_buffer_peek(struct ring_buffer *buffer, int cpu, u64 *ts) event = rb_reader_event(cpu_buffer); - switch (event->type) { + switch (event->type_len) { case RINGBUF_TYPE_PADDING: - RB_WARN_ON(cpu_buffer, 1); - rb_advance_reader(cpu_buffer); - return NULL; + if (rb_null_event(event)) + RB_WARN_ON(cpu_buffer, 1); + /* + * Because the writer could be discarding every + * event it creates (which would probably be bad) + * if we were to go back to "again" then we may never + * catch up, and will trigger the warn on, or lock + * the box. Return the padding, and we will release + * the current locks, and try again. + */ + return event; case RINGBUF_TYPE_TIME_EXTEND: /* Internal data, OK to advance */ @@ -1906,7 +3046,8 @@ rb_buffer_peek(struct ring_buffer *buffer, int cpu, u64 *ts) case RINGBUF_TYPE_DATA: if (ts) { *ts = cpu_buffer->read_stamp + event->time_delta; - ring_buffer_normalize_time_stamp(cpu_buffer->cpu, ts); + ring_buffer_normalize_time_stamp(cpu_buffer->buffer, + cpu_buffer->cpu, ts); } return event; @@ -1934,14 +3075,14 @@ rb_iter_peek(struct ring_buffer_iter *iter, u64 *ts) again: /* - * We repeat when a timestamp is encountered. It is possible - * to get multiple timestamps from an interrupt entering just - * as one timestamp is about to be written. The max times - * that this can happen is the number of nested interrupts we - * can have. Nesting 10 deep of interrupts is clearly - * an anomaly. + * We repeat when a timestamp is encountered. + * We can get multiple timestamps by nested interrupts or also + * if filtering is on (discarding commits). Since discarding + * commits can be frequent we can get a lot of timestamps. + * But we limit them by not adding timestamps if they begin + * at the start of a page. */ - if (RB_WARN_ON(cpu_buffer, ++nr_loops > 10)) + if (RB_WARN_ON(cpu_buffer, ++nr_loops > RB_TIMESTAMPS_PER_PAGE)) return NULL; if (rb_per_cpu_empty(cpu_buffer)) @@ -1949,10 +3090,14 @@ rb_iter_peek(struct ring_buffer_iter *iter, u64 *ts) event = rb_iter_head_event(iter); - switch (event->type) { + switch (event->type_len) { case RINGBUF_TYPE_PADDING: - rb_inc_iter(iter); - goto again; + if (rb_null_event(event)) { + rb_inc_iter(iter); + goto again; + } + rb_advance_iter(iter); + return event; case RINGBUF_TYPE_TIME_EXTEND: /* Internal data, OK to advance */ @@ -1967,7 +3112,8 @@ rb_iter_peek(struct ring_buffer_iter *iter, u64 *ts) case RINGBUF_TYPE_DATA: if (ts) { *ts = iter->read_stamp + event->time_delta; - ring_buffer_normalize_time_stamp(cpu_buffer->cpu, ts); + ring_buffer_normalize_time_stamp(buffer, + cpu_buffer->cpu, ts); } return event; @@ -1979,6 +3125,21 @@ rb_iter_peek(struct ring_buffer_iter *iter, u64 *ts) } EXPORT_SYMBOL_GPL(ring_buffer_iter_peek); +static inline int rb_ok_to_lock(void) +{ + /* + * If an NMI die dumps out the content of the ring buffer + * do not grab locks. We also permanently disable the ring + * buffer too. A one time deal is all you get from reading + * the ring buffer from an NMI. + */ + if (likely(!in_nmi())) + return 1; + + tracing_off_permanent(); + return 0; +} + /** * ring_buffer_peek - peek at the next event to be read * @buffer: The ring buffer to read @@ -1994,10 +3155,25 @@ ring_buffer_peek(struct ring_buffer *buffer, int cpu, u64 *ts) struct ring_buffer_per_cpu *cpu_buffer = buffer->buffers[cpu]; struct ring_buffer_event *event; unsigned long flags; + int dolock; - spin_lock_irqsave(&cpu_buffer->reader_lock, flags); - event = rb_buffer_peek(buffer, cpu, ts); - spin_unlock_irqrestore(&cpu_buffer->reader_lock, flags); + if (!cpumask_test_cpu(cpu, buffer->cpumask)) + return NULL; + + dolock = rb_ok_to_lock(); + again: + local_irq_save(flags); + if (dolock) + spin_lock(&cpu_buffer->reader_lock); + event = rb_buffer_peek(cpu_buffer, ts); + if (event && event->type_len == RINGBUF_TYPE_PADDING) + rb_advance_reader(cpu_buffer); + if (dolock) + spin_unlock(&cpu_buffer->reader_lock); + local_irq_restore(flags); + + if (event && event->type_len == RINGBUF_TYPE_PADDING) + goto again; return event; } @@ -2017,10 +3193,14 @@ ring_buffer_iter_peek(struct ring_buffer_iter *iter, u64 *ts) struct ring_buffer_event *event; unsigned long flags; + again: spin_lock_irqsave(&cpu_buffer->reader_lock, flags); event = rb_iter_peek(iter, ts); spin_unlock_irqrestore(&cpu_buffer->reader_lock, flags); + if (event && event->type_len == RINGBUF_TYPE_PADDING) + goto again; + return event; } @@ -2035,23 +3215,38 @@ ring_buffer_iter_peek(struct ring_buffer_iter *iter, u64 *ts) struct ring_buffer_event * ring_buffer_consume(struct ring_buffer *buffer, int cpu, u64 *ts) { - struct ring_buffer_per_cpu *cpu_buffer = buffer->buffers[cpu]; - struct ring_buffer_event *event; + struct ring_buffer_per_cpu *cpu_buffer; + struct ring_buffer_event *event = NULL; unsigned long flags; + int dolock; - if (!cpumask_test_cpu(cpu, buffer->cpumask)) - return NULL; + dolock = rb_ok_to_lock(); - spin_lock_irqsave(&cpu_buffer->reader_lock, flags); + again: + /* might be called in atomic */ + preempt_disable(); - event = rb_buffer_peek(buffer, cpu, ts); - if (!event) + if (!cpumask_test_cpu(cpu, buffer->cpumask)) goto out; - rb_advance_reader(cpu_buffer); + cpu_buffer = buffer->buffers[cpu]; + local_irq_save(flags); + if (dolock) + spin_lock(&cpu_buffer->reader_lock); + + event = rb_buffer_peek(cpu_buffer, ts); + if (event) + rb_advance_reader(cpu_buffer); + + if (dolock) + spin_unlock(&cpu_buffer->reader_lock); + local_irq_restore(flags); out: - spin_unlock_irqrestore(&cpu_buffer->reader_lock, flags); + preempt_enable(); + + if (event && event->type_len == RINGBUF_TYPE_PADDING) + goto again; return event; } @@ -2132,10 +3327,14 @@ ring_buffer_read(struct ring_buffer_iter *iter, u64 *ts) unsigned long flags; spin_lock_irqsave(&cpu_buffer->reader_lock, flags); + again: event = rb_iter_peek(iter, ts); if (!event) goto out; + if (event->type_len == RINGBUF_TYPE_PADDING) + goto again; + rb_advance_iter(iter); out: spin_unlock_irqrestore(&cpu_buffer->reader_lock, flags); @@ -2157,9 +3356,12 @@ EXPORT_SYMBOL_GPL(ring_buffer_size); static void rb_reset_cpu(struct ring_buffer_per_cpu *cpu_buffer) { + rb_head_page_deactivate(cpu_buffer); + cpu_buffer->head_page - = list_entry(cpu_buffer->pages.next, struct buffer_page, list); + = list_entry(cpu_buffer->pages, struct buffer_page, list); local_set(&cpu_buffer->head_page->write, 0); + local_set(&cpu_buffer->head_page->entries, 0); local_set(&cpu_buffer->head_page->page->commit, 0); cpu_buffer->head_page->read = 0; @@ -2169,14 +3371,21 @@ rb_reset_cpu(struct ring_buffer_per_cpu *cpu_buffer) INIT_LIST_HEAD(&cpu_buffer->reader_page->list); local_set(&cpu_buffer->reader_page->write, 0); + local_set(&cpu_buffer->reader_page->entries, 0); local_set(&cpu_buffer->reader_page->page->commit, 0); cpu_buffer->reader_page->read = 0; - cpu_buffer->overrun = 0; - cpu_buffer->entries = 0; + local_set(&cpu_buffer->commit_overrun, 0); + local_set(&cpu_buffer->overrun, 0); + local_set(&cpu_buffer->entries, 0); + local_set(&cpu_buffer->committing, 0); + local_set(&cpu_buffer->commits, 0); + cpu_buffer->read = 0; cpu_buffer->write_stamp = 0; cpu_buffer->read_stamp = 0; + + rb_head_page_activate(cpu_buffer); } /** @@ -2192,15 +3401,23 @@ void ring_buffer_reset_cpu(struct ring_buffer *buffer, int cpu) if (!cpumask_test_cpu(cpu, buffer->cpumask)) return; + atomic_inc(&cpu_buffer->record_disabled); + spin_lock_irqsave(&cpu_buffer->reader_lock, flags); + if (RB_WARN_ON(cpu_buffer, local_read(&cpu_buffer->committing))) + goto out; + __raw_spin_lock(&cpu_buffer->lock); rb_reset_cpu(cpu_buffer); __raw_spin_unlock(&cpu_buffer->lock); + out: spin_unlock_irqrestore(&cpu_buffer->reader_lock, flags); + + atomic_dec(&cpu_buffer->record_disabled); } EXPORT_SYMBOL_GPL(ring_buffer_reset_cpu); @@ -2224,14 +3441,28 @@ EXPORT_SYMBOL_GPL(ring_buffer_reset); int ring_buffer_empty(struct ring_buffer *buffer) { struct ring_buffer_per_cpu *cpu_buffer; + unsigned long flags; + int dolock; int cpu; + int ret; + + dolock = rb_ok_to_lock(); /* yes this is racy, but if you don't like the race, lock the buffer */ for_each_buffer_cpu(buffer, cpu) { cpu_buffer = buffer->buffers[cpu]; - if (!rb_per_cpu_empty(cpu_buffer)) + local_irq_save(flags); + if (dolock) + spin_lock(&cpu_buffer->reader_lock); + ret = rb_per_cpu_empty(cpu_buffer); + if (dolock) + spin_unlock(&cpu_buffer->reader_lock); + local_irq_restore(flags); + + if (!ret) return 0; } + return 1; } EXPORT_SYMBOL_GPL(ring_buffer_empty); @@ -2244,15 +3475,29 @@ EXPORT_SYMBOL_GPL(ring_buffer_empty); int ring_buffer_empty_cpu(struct ring_buffer *buffer, int cpu) { struct ring_buffer_per_cpu *cpu_buffer; + unsigned long flags; + int dolock; + int ret; if (!cpumask_test_cpu(cpu, buffer->cpumask)) return 1; + dolock = rb_ok_to_lock(); + cpu_buffer = buffer->buffers[cpu]; - return rb_per_cpu_empty(cpu_buffer); + local_irq_save(flags); + if (dolock) + spin_lock(&cpu_buffer->reader_lock); + ret = rb_per_cpu_empty(cpu_buffer); + if (dolock) + spin_unlock(&cpu_buffer->reader_lock); + local_irq_restore(flags); + + return ret; } EXPORT_SYMBOL_GPL(ring_buffer_empty_cpu); +#ifdef CONFIG_RING_BUFFER_ALLOW_SWAP /** * ring_buffer_swap_cpu - swap a CPU buffer between two ring buffers * @buffer_a: One buffer to swap with @@ -2268,18 +3513,36 @@ int ring_buffer_swap_cpu(struct ring_buffer *buffer_a, { struct ring_buffer_per_cpu *cpu_buffer_a; struct ring_buffer_per_cpu *cpu_buffer_b; + int ret = -EINVAL; if (!cpumask_test_cpu(cpu, buffer_a->cpumask) || !cpumask_test_cpu(cpu, buffer_b->cpumask)) - return -EINVAL; + goto out; /* At least make sure the two buffers are somewhat the same */ if (buffer_a->pages != buffer_b->pages) - return -EINVAL; + goto out; + + ret = -EAGAIN; + + if (ring_buffer_flags != RB_BUFFERS_ON) + goto out; + + if (atomic_read(&buffer_a->record_disabled)) + goto out; + + if (atomic_read(&buffer_b->record_disabled)) + goto out; cpu_buffer_a = buffer_a->buffers[cpu]; cpu_buffer_b = buffer_b->buffers[cpu]; + if (atomic_read(&cpu_buffer_a->record_disabled)) + goto out; + + if (atomic_read(&cpu_buffer_b->record_disabled)) + goto out; + /* * We can't do a synchronize_sched here because this * function can be called in atomic context. @@ -2289,39 +3552,28 @@ int ring_buffer_swap_cpu(struct ring_buffer *buffer_a, atomic_inc(&cpu_buffer_a->record_disabled); atomic_inc(&cpu_buffer_b->record_disabled); + ret = -EBUSY; + if (local_read(&cpu_buffer_a->committing)) + goto out_dec; + if (local_read(&cpu_buffer_b->committing)) + goto out_dec; + buffer_a->buffers[cpu] = cpu_buffer_b; buffer_b->buffers[cpu] = cpu_buffer_a; cpu_buffer_b->buffer = buffer_a; cpu_buffer_a->buffer = buffer_b; + ret = 0; + +out_dec: atomic_dec(&cpu_buffer_a->record_disabled); atomic_dec(&cpu_buffer_b->record_disabled); - - return 0; +out: + return ret; } EXPORT_SYMBOL_GPL(ring_buffer_swap_cpu); - -static void rb_remove_entries(struct ring_buffer_per_cpu *cpu_buffer, - struct buffer_data_page *bpage) -{ - struct ring_buffer_event *event; - unsigned long head; - - __raw_spin_lock(&cpu_buffer->lock); - for (head = 0; head < local_read(&bpage->commit); - head += rb_event_length(event)) { - - event = __rb_data_page_index(bpage, head); - if (RB_WARN_ON(cpu_buffer, rb_null_event(event))) - return; - /* Only count data entries */ - if (event->type != RINGBUF_TYPE_DATA) - continue; - cpu_buffer->entries--; - } - __raw_spin_unlock(&cpu_buffer->lock); -} +#endif /* CONFIG_RING_BUFFER_ALLOW_SWAP */ /** * ring_buffer_alloc_read_page - allocate a page to read from buffer @@ -2340,8 +3592,8 @@ static void rb_remove_entries(struct ring_buffer_per_cpu *cpu_buffer, */ void *ring_buffer_alloc_read_page(struct ring_buffer *buffer) { - unsigned long addr; struct buffer_data_page *bpage; + unsigned long addr; addr = __get_free_page(GFP_KERNEL); if (!addr) @@ -2349,8 +3601,11 @@ void *ring_buffer_alloc_read_page(struct ring_buffer *buffer) bpage = (void *)addr; + rb_init_page(bpage); + return bpage; } +EXPORT_SYMBOL_GPL(ring_buffer_alloc_read_page); /** * ring_buffer_free_read_page - free an allocated read page @@ -2363,11 +3618,13 @@ void ring_buffer_free_read_page(struct ring_buffer *buffer, void *data) { free_page((unsigned long)data); } +EXPORT_SYMBOL_GPL(ring_buffer_free_read_page); /** * ring_buffer_read_page - extract a page from the ring buffer * @buffer: buffer to extract from * @data_page: the page to use allocated from ring_buffer_alloc_read_page + * @len: amount to extract * @cpu: the cpu of the buffer to extract * @full: should the extraction only happen when the page is full. * @@ -2377,12 +3634,12 @@ void ring_buffer_free_read_page(struct ring_buffer *buffer, void *data) * to swap with a page in the ring buffer. * * for example: - * rpage = ring_buffer_alloc_page(buffer); + * rpage = ring_buffer_alloc_read_page(buffer); * if (!rpage) * return error; - * ret = ring_buffer_read_page(buffer, &rpage, cpu, 0); - * if (ret) - * process_page(rpage); + * ret = ring_buffer_read_page(buffer, &rpage, len, cpu, 0); + * if (ret >= 0) + * process_page(rpage, ret); * * When @full is set, the function will not return true unless * the writer is off the reader page. @@ -2393,80 +3650,129 @@ void ring_buffer_free_read_page(struct ring_buffer *buffer, void *data) * responsible for that. * * Returns: - * 1 if data has been transferred - * 0 if no data has been transferred. + * >=0 if data has been transferred, returns the offset of consumed data. + * <0 if no data has been transferred. */ int ring_buffer_read_page(struct ring_buffer *buffer, - void **data_page, int cpu, int full) + void **data_page, size_t len, int cpu, int full) { struct ring_buffer_per_cpu *cpu_buffer = buffer->buffers[cpu]; struct ring_buffer_event *event; struct buffer_data_page *bpage; + struct buffer_page *reader; unsigned long flags; - int ret = 0; + unsigned int commit; + unsigned int read; + u64 save_timestamp; + int ret = -1; + + if (!cpumask_test_cpu(cpu, buffer->cpumask)) + goto out; + + /* + * If len is not big enough to hold the page header, then + * we can not copy anything. + */ + if (len <= BUF_PAGE_HDR_SIZE) + goto out; + + len -= BUF_PAGE_HDR_SIZE; if (!data_page) - return 0; + goto out; bpage = *data_page; if (!bpage) - return 0; + goto out; spin_lock_irqsave(&cpu_buffer->reader_lock, flags); - /* - * rb_buffer_peek will get the next ring buffer if - * the current reader page is empty. - */ - event = rb_buffer_peek(buffer, cpu, NULL); - if (!event) - goto out; + reader = rb_get_reader_page(cpu_buffer); + if (!reader) + goto out_unlock; + + event = rb_reader_event(cpu_buffer); + + read = reader->read; + commit = rb_page_commit(reader); - /* check for data */ - if (!local_read(&cpu_buffer->reader_page->page->commit)) - goto out; /* - * If the writer is already off of the read page, then simply - * switch the read page with the given page. Otherwise - * we need to copy the data from the reader to the writer. + * If this page has been partially read or + * if len is not big enough to read the rest of the page or + * a writer is still on the page, then + * we must copy the data from the page to the buffer. + * Otherwise, we can simply swap the page with the one passed in. */ - if (cpu_buffer->reader_page == cpu_buffer->commit_page) { - unsigned int read = cpu_buffer->reader_page->read; + if (read || (len < (commit - read)) || + cpu_buffer->reader_page == cpu_buffer->commit_page) { + struct buffer_data_page *rpage = cpu_buffer->reader_page->page; + unsigned int rpos = read; + unsigned int pos = 0; + unsigned int size; if (full) - goto out; - /* The writer is still on the reader page, we must copy */ - bpage = cpu_buffer->reader_page->page; - memcpy(bpage->data, - cpu_buffer->reader_page->page->data + read, - local_read(&bpage->commit) - read); + goto out_unlock; - /* consume what was read */ - cpu_buffer->reader_page += read; + if (len > (commit - read)) + len = (commit - read); + size = rb_event_length(event); + + if (len < size) + goto out_unlock; + + /* save the current timestamp, since the user will need it */ + save_timestamp = cpu_buffer->read_stamp; + + /* Need to copy one event at a time */ + do { + memcpy(bpage->data + pos, rpage->data + rpos, size); + + len -= size; + + rb_advance_reader(cpu_buffer); + rpos = reader->read; + pos += size; + + event = rb_reader_event(cpu_buffer); + size = rb_event_length(event); + } while (len > size); + + /* update bpage */ + local_set(&bpage->commit, pos); + bpage->time_stamp = save_timestamp; + + /* we copied everything to the beginning */ + read = 0; } else { + /* update the entry counter */ + cpu_buffer->read += rb_page_entries(reader); + /* swap the pages */ rb_init_page(bpage); - bpage = cpu_buffer->reader_page->page; - cpu_buffer->reader_page->page = *data_page; - cpu_buffer->reader_page->read = 0; + bpage = reader->page; + reader->page = *data_page; + local_set(&reader->write, 0); + local_set(&reader->entries, 0); + reader->read = 0; *data_page = bpage; } - ret = 1; + ret = read; - /* update the entry counter */ - rb_remove_entries(cpu_buffer, bpage); - out: + out_unlock: spin_unlock_irqrestore(&cpu_buffer->reader_lock, flags); + out: return ret; } +EXPORT_SYMBOL_GPL(ring_buffer_read_page); +#ifdef CONFIG_TRACING static ssize_t rb_simple_read(struct file *filp, char __user *ubuf, size_t cnt, loff_t *ppos) { - long *p = filp->private_data; + unsigned long *p = filp->private_data; char buf[64]; int r; @@ -2482,9 +3788,9 @@ static ssize_t rb_simple_write(struct file *filp, const char __user *ubuf, size_t cnt, loff_t *ppos) { - long *p = filp->private_data; + unsigned long *p = filp->private_data; char buf[64]; - long val; + unsigned long val; int ret; if (cnt >= sizeof(buf)) @@ -2509,7 +3815,7 @@ rb_simple_write(struct file *filp, const char __user *ubuf, return cnt; } -static struct file_operations rb_simple_fops = { +static const struct file_operations rb_simple_fops = { .open = tracing_open_generic, .read = rb_simple_read, .write = rb_simple_write, @@ -2519,16 +3825,53 @@ static struct file_operations rb_simple_fops = { static __init int rb_init_debugfs(void) { struct dentry *d_tracer; - struct dentry *entry; d_tracer = tracing_init_dentry(); - entry = debugfs_create_file("tracing_on", 0644, d_tracer, - &ring_buffer_flags, &rb_simple_fops); - if (!entry) - pr_warning("Could not create debugfs 'tracing_on' entry\n"); + trace_create_file("tracing_on", 0644, d_tracer, + &ring_buffer_flags, &rb_simple_fops); return 0; } fs_initcall(rb_init_debugfs); +#endif + +#ifdef CONFIG_HOTPLUG_CPU +static int rb_cpu_notify(struct notifier_block *self, + unsigned long action, void *hcpu) +{ + struct ring_buffer *buffer = + container_of(self, struct ring_buffer, cpu_notify); + long cpu = (long)hcpu; + + switch (action) { + case CPU_UP_PREPARE: + case CPU_UP_PREPARE_FROZEN: + if (cpumask_test_cpu(cpu, buffer->cpumask)) + return NOTIFY_OK; + + buffer->buffers[cpu] = + rb_allocate_cpu_buffer(buffer, cpu); + if (!buffer->buffers[cpu]) { + WARN(1, "failed to allocate ring buffer on CPU %ld\n", + cpu); + return NOTIFY_OK; + } + smp_wmb(); + cpumask_set_cpu(cpu, buffer->cpumask); + break; + case CPU_DOWN_PREPARE: + case CPU_DOWN_PREPARE_FROZEN: + /* + * Do nothing. + * If we were to free the buffer, then the user would + * lose any trace that was in the buffer. + */ + break; + default: + break; + } + return NOTIFY_OK; +} +#endif diff --git a/kernel/trace/ring_buffer_benchmark.c b/kernel/trace/ring_buffer_benchmark.c new file mode 100644 index 00000000000..b2477caf09c --- /dev/null +++ b/kernel/trace/ring_buffer_benchmark.c @@ -0,0 +1,486 @@ +/* + * ring buffer tester and benchmark + * + * Copyright (C) 2009 Steven Rostedt <srostedt@redhat.com> + */ +#include <linux/ring_buffer.h> +#include <linux/completion.h> +#include <linux/kthread.h> +#include <linux/module.h> +#include <linux/time.h> + +struct rb_page { + u64 ts; + local_t commit; + char data[4080]; +}; + +/* run time and sleep time in seconds */ +#define RUN_TIME 10 +#define SLEEP_TIME 10 + +/* number of events for writer to wake up the reader */ +static int wakeup_interval = 100; + +static int reader_finish; +static struct completion read_start; +static struct completion read_done; + +static struct ring_buffer *buffer; +static struct task_struct *producer; +static struct task_struct *consumer; +static unsigned long read; + +static int disable_reader; +module_param(disable_reader, uint, 0644); +MODULE_PARM_DESC(disable_reader, "only run producer"); + +static int write_iteration = 50; +module_param(write_iteration, uint, 0644); +MODULE_PARM_DESC(write_iteration, "# of writes between timestamp readings"); + +static int producer_nice = 19; +static int consumer_nice = 19; + +static int producer_fifo = -1; +static int consumer_fifo = -1; + +module_param(producer_nice, uint, 0644); +MODULE_PARM_DESC(producer_nice, "nice prio for producer"); + +module_param(consumer_nice, uint, 0644); +MODULE_PARM_DESC(consumer_nice, "nice prio for consumer"); + +module_param(producer_fifo, uint, 0644); +MODULE_PARM_DESC(producer_fifo, "fifo prio for producer"); + +module_param(consumer_fifo, uint, 0644); +MODULE_PARM_DESC(consumer_fifo, "fifo prio for consumer"); + +static int read_events; + +static int kill_test; + +#define KILL_TEST() \ + do { \ + if (!kill_test) { \ + kill_test = 1; \ + WARN_ON(1); \ + } \ + } while (0) + +enum event_status { + EVENT_FOUND, + EVENT_DROPPED, +}; + +static enum event_status read_event(int cpu) +{ + struct ring_buffer_event *event; + int *entry; + u64 ts; + + event = ring_buffer_consume(buffer, cpu, &ts); + if (!event) + return EVENT_DROPPED; + + entry = ring_buffer_event_data(event); + if (*entry != cpu) { + KILL_TEST(); + return EVENT_DROPPED; + } + + read++; + return EVENT_FOUND; +} + +static enum event_status read_page(int cpu) +{ + struct ring_buffer_event *event; + struct rb_page *rpage; + unsigned long commit; + void *bpage; + int *entry; + int ret; + int inc; + int i; + + bpage = ring_buffer_alloc_read_page(buffer); + if (!bpage) + return EVENT_DROPPED; + + ret = ring_buffer_read_page(buffer, &bpage, PAGE_SIZE, cpu, 1); + if (ret >= 0) { + rpage = bpage; + commit = local_read(&rpage->commit); + for (i = 0; i < commit && !kill_test; i += inc) { + + if (i >= (PAGE_SIZE - offsetof(struct rb_page, data))) { + KILL_TEST(); + break; + } + + inc = -1; + event = (void *)&rpage->data[i]; + switch (event->type_len) { + case RINGBUF_TYPE_PADDING: + /* failed writes may be discarded events */ + if (!event->time_delta) + KILL_TEST(); + inc = event->array[0] + 4; + break; + case RINGBUF_TYPE_TIME_EXTEND: + inc = 8; + break; + case 0: + entry = ring_buffer_event_data(event); + if (*entry != cpu) { + KILL_TEST(); + break; + } + read++; + if (!event->array[0]) { + KILL_TEST(); + break; + } + inc = event->array[0] + 4; + break; + default: + entry = ring_buffer_event_data(event); + if (*entry != cpu) { + KILL_TEST(); + break; + } + read++; + inc = ((event->type_len + 1) * 4); + } + if (kill_test) + break; + + if (inc <= 0) { + KILL_TEST(); + break; + } + } + } + ring_buffer_free_read_page(buffer, bpage); + + if (ret < 0) + return EVENT_DROPPED; + return EVENT_FOUND; +} + +static void ring_buffer_consumer(void) +{ + /* toggle between reading pages and events */ + read_events ^= 1; + + read = 0; + while (!reader_finish && !kill_test) { + int found; + + do { + int cpu; + + found = 0; + for_each_online_cpu(cpu) { + enum event_status stat; + + if (read_events) + stat = read_event(cpu); + else + stat = read_page(cpu); + + if (kill_test) + break; + if (stat == EVENT_FOUND) + found = 1; + } + } while (found && !kill_test); + + set_current_state(TASK_INTERRUPTIBLE); + if (reader_finish) + break; + + schedule(); + __set_current_state(TASK_RUNNING); + } + reader_finish = 0; + complete(&read_done); +} + +static void ring_buffer_producer(void) +{ + struct timeval start_tv; + struct timeval end_tv; + unsigned long long time; + unsigned long long entries; + unsigned long long overruns; + unsigned long missed = 0; + unsigned long hit = 0; + unsigned long avg; + int cnt = 0; + + /* + * Hammer the buffer for 10 secs (this may + * make the system stall) + */ + trace_printk("Starting ring buffer hammer\n"); + do_gettimeofday(&start_tv); + do { + struct ring_buffer_event *event; + int *entry; + int i; + + for (i = 0; i < write_iteration; i++) { + event = ring_buffer_lock_reserve(buffer, 10); + if (!event) { + missed++; + } else { + hit++; + entry = ring_buffer_event_data(event); + *entry = smp_processor_id(); + ring_buffer_unlock_commit(buffer, event); + } + } + do_gettimeofday(&end_tv); + + cnt++; + if (consumer && !(cnt % wakeup_interval)) + wake_up_process(consumer); + +#ifndef CONFIG_PREEMPT + /* + * If we are a non preempt kernel, the 10 second run will + * stop everything while it runs. Instead, we will call + * cond_resched and also add any time that was lost by a + * rescedule. + * + * Do a cond resched at the same frequency we would wake up + * the reader. + */ + if (cnt % wakeup_interval) + cond_resched(); +#endif + + } while (end_tv.tv_sec < (start_tv.tv_sec + RUN_TIME) && !kill_test); + trace_printk("End ring buffer hammer\n"); + + if (consumer) { + /* Init both completions here to avoid races */ + init_completion(&read_start); + init_completion(&read_done); + /* the completions must be visible before the finish var */ + smp_wmb(); + reader_finish = 1; + /* finish var visible before waking up the consumer */ + smp_wmb(); + wake_up_process(consumer); + wait_for_completion(&read_done); + } + + time = end_tv.tv_sec - start_tv.tv_sec; + time *= USEC_PER_SEC; + time += (long long)((long)end_tv.tv_usec - (long)start_tv.tv_usec); + + entries = ring_buffer_entries(buffer); + overruns = ring_buffer_overruns(buffer); + + if (kill_test) + trace_printk("ERROR!\n"); + + if (!disable_reader) { + if (consumer_fifo < 0) + trace_printk("Running Consumer at nice: %d\n", + consumer_nice); + else + trace_printk("Running Consumer at SCHED_FIFO %d\n", + consumer_fifo); + } + if (producer_fifo < 0) + trace_printk("Running Producer at nice: %d\n", + producer_nice); + else + trace_printk("Running Producer at SCHED_FIFO %d\n", + producer_fifo); + + /* Let the user know that the test is running at low priority */ + if (producer_fifo < 0 && consumer_fifo < 0 && + producer_nice == 19 && consumer_nice == 19) + trace_printk("WARNING!!! This test is running at lowest priority.\n"); + + trace_printk("Time: %lld (usecs)\n", time); + trace_printk("Overruns: %lld\n", overruns); + if (disable_reader) + trace_printk("Read: (reader disabled)\n"); + else + trace_printk("Read: %ld (by %s)\n", read, + read_events ? "events" : "pages"); + trace_printk("Entries: %lld\n", entries); + trace_printk("Total: %lld\n", entries + overruns + read); + trace_printk("Missed: %ld\n", missed); + trace_printk("Hit: %ld\n", hit); + + /* Convert time from usecs to millisecs */ + do_div(time, USEC_PER_MSEC); + if (time) + hit /= (long)time; + else + trace_printk("TIME IS ZERO??\n"); + + trace_printk("Entries per millisec: %ld\n", hit); + + if (hit) { + /* Calculate the average time in nanosecs */ + avg = NSEC_PER_MSEC / hit; + trace_printk("%ld ns per entry\n", avg); + } + + if (missed) { + if (time) + missed /= (long)time; + + trace_printk("Total iterations per millisec: %ld\n", + hit + missed); + + /* it is possible that hit + missed will overflow and be zero */ + if (!(hit + missed)) { + trace_printk("hit + missed overflowed and totalled zero!\n"); + hit--; /* make it non zero */ + } + + /* Caculate the average time in nanosecs */ + avg = NSEC_PER_MSEC / (hit + missed); + trace_printk("%ld ns per entry\n", avg); + } +} + +static void wait_to_die(void) +{ + set_current_state(TASK_INTERRUPTIBLE); + while (!kthread_should_stop()) { + schedule(); + set_current_state(TASK_INTERRUPTIBLE); + } + __set_current_state(TASK_RUNNING); +} + +static int ring_buffer_consumer_thread(void *arg) +{ + while (!kthread_should_stop() && !kill_test) { + complete(&read_start); + + ring_buffer_consumer(); + + set_current_state(TASK_INTERRUPTIBLE); + if (kthread_should_stop() || kill_test) + break; + + schedule(); + __set_current_state(TASK_RUNNING); + } + __set_current_state(TASK_RUNNING); + + if (kill_test) + wait_to_die(); + + return 0; +} + +static int ring_buffer_producer_thread(void *arg) +{ + init_completion(&read_start); + + while (!kthread_should_stop() && !kill_test) { + ring_buffer_reset(buffer); + + if (consumer) { + smp_wmb(); + wake_up_process(consumer); + wait_for_completion(&read_start); + } + + ring_buffer_producer(); + + trace_printk("Sleeping for 10 secs\n"); + set_current_state(TASK_INTERRUPTIBLE); + schedule_timeout(HZ * SLEEP_TIME); + __set_current_state(TASK_RUNNING); + } + + if (kill_test) + wait_to_die(); + + return 0; +} + +static int __init ring_buffer_benchmark_init(void) +{ + int ret; + + /* make a one meg buffer in overwite mode */ + buffer = ring_buffer_alloc(1000000, RB_FL_OVERWRITE); + if (!buffer) + return -ENOMEM; + + if (!disable_reader) { + consumer = kthread_create(ring_buffer_consumer_thread, + NULL, "rb_consumer"); + ret = PTR_ERR(consumer); + if (IS_ERR(consumer)) + goto out_fail; + } + + producer = kthread_run(ring_buffer_producer_thread, + NULL, "rb_producer"); + ret = PTR_ERR(producer); + + if (IS_ERR(producer)) + goto out_kill; + + /* + * Run them as low-prio background tasks by default: + */ + if (!disable_reader) { + if (consumer_fifo >= 0) { + struct sched_param param = { + .sched_priority = consumer_fifo + }; + sched_setscheduler(consumer, SCHED_FIFO, ¶m); + } else + set_user_nice(consumer, consumer_nice); + } + + if (producer_fifo >= 0) { + struct sched_param param = { + .sched_priority = consumer_fifo + }; + sched_setscheduler(producer, SCHED_FIFO, ¶m); + } else + set_user_nice(producer, producer_nice); + + return 0; + + out_kill: + if (consumer) + kthread_stop(consumer); + + out_fail: + ring_buffer_free(buffer); + return ret; +} + +static void __exit ring_buffer_benchmark_exit(void) +{ + kthread_stop(producer); + if (consumer) + kthread_stop(consumer); + ring_buffer_free(buffer); +} + +module_init(ring_buffer_benchmark_init); +module_exit(ring_buffer_benchmark_exit); + +MODULE_AUTHOR("Steven Rostedt"); +MODULE_DESCRIPTION("ring_buffer_benchmark"); +MODULE_LICENSE("GPL"); diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 17bb88d86ac..874f2893cff 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -11,47 +11,58 @@ * Copyright (C) 2004-2006 Ingo Molnar * Copyright (C) 2004 William Lee Irwin III */ +#include <linux/ring_buffer.h> #include <linux/utsrelease.h> +#include <linux/stacktrace.h> +#include <linux/writeback.h> #include <linux/kallsyms.h> #include <linux/seq_file.h> +#include <linux/smp_lock.h> #include <linux/notifier.h> +#include <linux/irqflags.h> #include <linux/debugfs.h> #include <linux/pagemap.h> #include <linux/hardirq.h> #include <linux/linkage.h> #include <linux/uaccess.h> +#include <linux/kprobes.h> #include <linux/ftrace.h> #include <linux/module.h> #include <linux/percpu.h> +#include <linux/splice.h> #include <linux/kdebug.h> +#include <linux/string.h> #include <linux/ctype.h> #include <linux/init.h> #include <linux/poll.h> #include <linux/gfp.h> #include <linux/fs.h> -#include <linux/kprobes.h> -#include <linux/writeback.h> - -#include <linux/stacktrace.h> -#include <linux/ring_buffer.h> -#include <linux/irqflags.h> #include "trace.h" +#include "trace_output.h" #define TRACE_BUFFER_FLAGS (RB_FL_OVERWRITE) -unsigned long __read_mostly tracing_max_latency; -unsigned long __read_mostly tracing_thresh; +/* + * On boot up, the ring buffer is set to the minimum size, so that + * we do not waste memory on systems that are not using tracing. + */ +int ring_buffer_expanded; /* * We need to change this state when a selftest is running. * A selftest will lurk into the ring-buffer to count the * entries inserted during the selftest although some concurrent - * insertions into the ring-buffer such as ftrace_printk could occurred + * insertions into the ring-buffer such as trace_printk could occurred * at the same time, giving false positive or negative results. */ static bool __read_mostly tracing_selftest_running; +/* + * If a tracer is running, we do not want to run SELFTEST. + */ +bool __read_mostly tracing_selftest_disabled; + /* For tracers that don't implement custom flags */ static struct tracer_opt dummy_tracer_opt[] = { { } @@ -73,9 +84,9 @@ static int dummy_set_flag(u32 old_flags, u32 bit, int set) * of the tracer is successful. But that is the only place that sets * this back to zero. */ -int tracing_disabled = 1; +static int tracing_disabled = 1; -static DEFINE_PER_CPU(local_t, ftrace_cpu_disabled); +DEFINE_PER_CPU(local_t, ftrace_cpu_disabled); static inline void ftrace_disable_cpu(void) { @@ -91,6 +102,9 @@ static inline void ftrace_enable_cpu(void) static cpumask_var_t __read_mostly tracing_buffer_mask; +/* Define which cpu buffers are currently read in trace_pipe */ +static cpumask_var_t tracing_reader_cpumask; + #define for_each_tracing_cpu(cpu) \ for_each_cpu(cpu, tracing_buffer_mask) @@ -109,14 +123,21 @@ static cpumask_var_t __read_mostly tracing_buffer_mask; */ int ftrace_dump_on_oops; -static int tracing_set_tracer(char *buf); +static int tracing_set_tracer(const char *buf); + +#define MAX_TRACER_SIZE 100 +static char bootup_tracer_buf[MAX_TRACER_SIZE] __initdata; +static char *default_bootup_tracer; -static int __init set_ftrace(char *str) +static int __init set_cmdline_ftrace(char *str) { - tracing_set_tracer(str); + strncpy(bootup_tracer_buf, str, MAX_TRACER_SIZE); + default_bootup_tracer = bootup_tracer_buf; + /* We are using ftrace early, expand it */ + ring_buffer_expanded = 1; return 1; } -__setup("ftrace", set_ftrace); +__setup("ftrace=", set_cmdline_ftrace); static int __init set_ftrace_dump_on_oops(char *str) { @@ -125,21 +146,13 @@ static int __init set_ftrace_dump_on_oops(char *str) } __setup("ftrace_dump_on_oops", set_ftrace_dump_on_oops); -long -ns2usecs(cycle_t nsec) +unsigned long long ns2usecs(cycle_t nsec) { nsec += 500; do_div(nsec, 1000); return nsec; } -cycle_t ftrace_now(int cpu) -{ - u64 ts = ring_buffer_time_stamp(cpu); - ring_buffer_normalize_time_stamp(cpu, &ts); - return ts; -} - /* * The global_trace is the descriptor that holds the tracing * buffers for the live tracing. For each CPU, it contains @@ -156,6 +169,28 @@ static struct trace_array global_trace; static DEFINE_PER_CPU(struct trace_array_cpu, global_trace_cpu); +int filter_current_check_discard(struct ring_buffer *buffer, + struct ftrace_event_call *call, void *rec, + struct ring_buffer_event *event) +{ + return filter_check_discard(call, rec, buffer, event); +} +EXPORT_SYMBOL_GPL(filter_current_check_discard); + +cycle_t ftrace_now(int cpu) +{ + u64 ts; + + /* Early boot up does not have a buffer yet */ + if (!global_trace.buffer) + return trace_clock_local(); + + ts = ring_buffer_time_stamp(global_trace.buffer, cpu); + ring_buffer_normalize_time_stamp(global_trace.buffer, cpu, &ts); + + return ts; +} + /* * The max_tr is used to snapshot the global_trace when a maximum * latency is reached. Some tracers will use this to store a maximum @@ -186,9 +221,6 @@ int tracing_is_enabled(void) return tracer_enabled; } -/* function tracing enabled */ -int ftrace_function_enabled; - /* * trace_buf_size is the size in bytes that is allocated * for a buffer. Note, the number of bytes is always rounded @@ -210,13 +242,6 @@ static struct tracer *trace_types __read_mostly; static struct tracer *current_trace __read_mostly; /* - * max_tracer_type_len is used to simplify the allocating of - * buffers to read userspace tracer names. We keep track of - * the longest tracer name registered. - */ -static int max_tracer_type_len; - -/* * trace_types_lock is used to protect the trace_types list. * This lock is also used to keep user access serialized. * Accesses from userspace will grab this lock while userspace @@ -229,7 +254,11 @@ static DECLARE_WAIT_QUEUE_HEAD(trace_wait); /* trace_flags holds trace_options default values */ unsigned long trace_flags = TRACE_ITER_PRINT_PARENT | TRACE_ITER_PRINTK | - TRACE_ITER_ANNOTATE; + TRACE_ITER_ANNOTATE | TRACE_ITER_CONTEXT_INFO | TRACE_ITER_SLEEP_TIME | + TRACE_ITER_GRAPH_TIME; + +static int trace_stop_count; +static DEFINE_SPINLOCK(tracing_start_lock); /** * trace_wake_up - wake up tasks waiting for trace input @@ -239,24 +268,29 @@ unsigned long trace_flags = TRACE_ITER_PRINT_PARENT | TRACE_ITER_PRINTK | */ void trace_wake_up(void) { + int cpu; + + if (trace_flags & TRACE_ITER_BLOCK) + return; /* * The runqueue_is_locked() can fail, but this is the best we * have for now: */ - if (!(trace_flags & TRACE_ITER_BLOCK) && !runqueue_is_locked()) + cpu = get_cpu(); + if (!runqueue_is_locked(cpu)) wake_up(&trace_wait); + put_cpu(); } static int __init set_buf_size(char *str) { unsigned long buf_size; - int ret; if (!str) return 0; - ret = strict_strtoul(str, 0, &buf_size); + buf_size = memparse(str, &str); /* nr_entries can not be zero */ - if (ret < 0 || buf_size == 0) + if (buf_size == 0) return 0; trace_buf_size = buf_size; return 1; @@ -280,192 +314,164 @@ static const char *trace_options[] = { "block", "stacktrace", "sched-tree", - "ftrace_printk", + "trace_printk", "ftrace_preempt", "branch", "annotate", "userstacktrace", "sym-userobj", "printk-msg-only", + "context-info", + "latency-format", + "sleep-time", + "graph-time", NULL }; -/* - * ftrace_max_lock is used to protect the swapping of buffers - * when taking a max snapshot. The buffers themselves are - * protected by per_cpu spinlocks. But the action of the swap - * needs its own lock. - * - * This is defined as a raw_spinlock_t in order to help - * with performance when lockdep debugging is enabled. - */ -static raw_spinlock_t ftrace_max_lock = - (raw_spinlock_t)__RAW_SPIN_LOCK_UNLOCKED; +static struct { + u64 (*func)(void); + const char *name; +} trace_clocks[] = { + { trace_clock_local, "local" }, + { trace_clock_global, "global" }, +}; + +int trace_clock_id; /* - * Copy the new maximum trace into the separate maximum-trace - * structure. (this way the maximum trace is permanently saved, - * for later retrieval via /debugfs/tracing/latency_trace) + * trace_parser_get_init - gets the buffer for trace parser */ -static void -__update_max_tr(struct trace_array *tr, struct task_struct *tsk, int cpu) +int trace_parser_get_init(struct trace_parser *parser, int size) { - struct trace_array_cpu *data = tr->data[cpu]; + memset(parser, 0, sizeof(*parser)); - max_tr.cpu = cpu; - max_tr.time_start = data->preempt_timestamp; - - data = max_tr.data[cpu]; - data->saved_latency = tracing_max_latency; - - memcpy(data->comm, tsk->comm, TASK_COMM_LEN); - data->pid = tsk->pid; - data->uid = task_uid(tsk); - data->nice = tsk->static_prio - 20 - MAX_RT_PRIO; - data->policy = tsk->policy; - data->rt_priority = tsk->rt_priority; + parser->buffer = kmalloc(size, GFP_KERNEL); + if (!parser->buffer) + return 1; - /* record this tasks comm */ - tracing_record_cmdline(current); + parser->size = size; + return 0; } -/** - * trace_seq_printf - sequence printing of trace information - * @s: trace sequence descriptor - * @fmt: printf format string - * - * The tracer may use either sequence operations or its own - * copy to user routines. To simplify formating of a trace - * trace_seq_printf is used to store strings into a special - * buffer (@s). Then the output may be either used by - * the sequencer or pulled into another buffer. +/* + * trace_parser_put - frees the buffer for trace parser */ -int -trace_seq_printf(struct trace_seq *s, const char *fmt, ...) +void trace_parser_put(struct trace_parser *parser) { - int len = (PAGE_SIZE - 1) - s->len; - va_list ap; - int ret; - - if (!len) - return 0; - - va_start(ap, fmt); - ret = vsnprintf(s->buffer + s->len, len, fmt, ap); - va_end(ap); - - /* If we can't write it all, don't bother writing anything */ - if (ret >= len) - return 0; - - s->len += ret; - - return len; + kfree(parser->buffer); } -/** - * trace_seq_puts - trace sequence printing of simple string - * @s: trace sequence descriptor - * @str: simple string to record +/* + * trace_get_user - reads the user input string separated by space + * (matched by isspace(ch)) + * + * For each string found the 'struct trace_parser' is updated, + * and the function returns. + * + * Returns number of bytes read. * - * The tracer may use either the sequence operations or its own - * copy to user routines. This function records a simple string - * into a special buffer (@s) for later retrieval by a sequencer - * or other mechanism. + * See kernel/trace/trace.h for 'struct trace_parser' details. */ -static int -trace_seq_puts(struct trace_seq *s, const char *str) -{ - int len = strlen(str); - - if (len > ((PAGE_SIZE - 1) - s->len)) - return 0; - - memcpy(s->buffer + s->len, str, len); - s->len += len; - - return len; -} - -static int -trace_seq_putc(struct trace_seq *s, unsigned char c) +int trace_get_user(struct trace_parser *parser, const char __user *ubuf, + size_t cnt, loff_t *ppos) { - if (s->len >= (PAGE_SIZE - 1)) - return 0; + char ch; + size_t read = 0; + ssize_t ret; - s->buffer[s->len++] = c; + if (!*ppos) + trace_parser_clear(parser); - return 1; -} + ret = get_user(ch, ubuf++); + if (ret) + goto out; -static int -trace_seq_putmem(struct trace_seq *s, void *mem, size_t len) -{ - if (len > ((PAGE_SIZE - 1) - s->len)) - return 0; + read++; + cnt--; - memcpy(s->buffer + s->len, mem, len); - s->len += len; + /* + * The parser is not finished with the last write, + * continue reading the user input without skipping spaces. + */ + if (!parser->cont) { + /* skip white space */ + while (cnt && isspace(ch)) { + ret = get_user(ch, ubuf++); + if (ret) + goto out; + read++; + cnt--; + } - return len; -} + /* only spaces were written */ + if (isspace(ch)) { + *ppos += read; + ret = read; + goto out; + } -#define MAX_MEMHEX_BYTES 8 -#define HEX_CHARS (MAX_MEMHEX_BYTES*2 + 1) + parser->idx = 0; + } -static int -trace_seq_putmem_hex(struct trace_seq *s, void *mem, size_t len) -{ - unsigned char hex[HEX_CHARS]; - unsigned char *data = mem; - int i, j; + /* read the non-space input */ + while (cnt && !isspace(ch)) { + if (parser->idx < parser->size - 1) + parser->buffer[parser->idx++] = ch; + else { + ret = -EINVAL; + goto out; + } + ret = get_user(ch, ubuf++); + if (ret) + goto out; + read++; + cnt--; + } -#ifdef __BIG_ENDIAN - for (i = 0, j = 0; i < len; i++) { -#else - for (i = len-1, j = 0; i >= 0; i--) { -#endif - hex[j++] = hex_asc_hi(data[i]); - hex[j++] = hex_asc_lo(data[i]); + /* We either got finished input or we have to wait for another call. */ + if (isspace(ch)) { + parser->buffer[parser->idx] = 0; + parser->cont = false; + } else { + parser->cont = true; + parser->buffer[parser->idx++] = ch; } - hex[j++] = ' '; - return trace_seq_putmem(s, hex, j); + *ppos += read; + ret = read; + +out: + return ret; } -static int -trace_seq_path(struct trace_seq *s, struct path *path) +ssize_t trace_seq_to_user(struct trace_seq *s, char __user *ubuf, size_t cnt) { - unsigned char *p; + int len; + int ret; - if (s->len >= (PAGE_SIZE - 1)) + if (!cnt) return 0; - p = d_path(path, s->buffer + s->len, PAGE_SIZE - s->len); - if (!IS_ERR(p)) { - p = mangle_path(s->buffer + s->len, p, "\n"); - if (p) { - s->len = p - s->buffer; - return 1; - } - } else { - s->buffer[s->len++] = '?'; - return 1; - } - return 0; -} + if (s->len <= s->readpos) + return -EBUSY; -static void -trace_seq_reset(struct trace_seq *s) -{ - s->len = 0; - s->readpos = 0; + len = s->len - s->readpos; + if (cnt > len) + cnt = len; + ret = copy_to_user(ubuf, s->buffer + s->readpos, cnt); + if (ret == cnt) + return -EFAULT; + + cnt -= ret; + + s->readpos += cnt; + return cnt; } -ssize_t trace_seq_to_user(struct trace_seq *s, char __user *ubuf, size_t cnt) +static ssize_t trace_seq_to_buffer(struct trace_seq *s, void *buf, size_t cnt) { int len; - int ret; + void *ret; if (s->len <= s->readpos) return -EBUSY; @@ -473,23 +479,62 @@ ssize_t trace_seq_to_user(struct trace_seq *s, char __user *ubuf, size_t cnt) len = s->len - s->readpos; if (cnt > len) cnt = len; - ret = copy_to_user(ubuf, s->buffer + s->readpos, cnt); - if (ret) + ret = memcpy(buf, s->buffer + s->readpos, cnt); + if (!ret) return -EFAULT; - s->readpos += len; + s->readpos += cnt; return cnt; } +/* + * ftrace_max_lock is used to protect the swapping of buffers + * when taking a max snapshot. The buffers themselves are + * protected by per_cpu spinlocks. But the action of the swap + * needs its own lock. + * + * This is defined as a raw_spinlock_t in order to help + * with performance when lockdep debugging is enabled. + * + * It is also used in other places outside the update_max_tr + * so it needs to be defined outside of the + * CONFIG_TRACER_MAX_TRACE. + */ +static raw_spinlock_t ftrace_max_lock = + (raw_spinlock_t)__RAW_SPIN_LOCK_UNLOCKED; + +#ifdef CONFIG_TRACER_MAX_TRACE +unsigned long __read_mostly tracing_max_latency; +unsigned long __read_mostly tracing_thresh; + +/* + * Copy the new maximum trace into the separate maximum-trace + * structure. (this way the maximum trace is permanently saved, + * for later retrieval via /sys/kernel/debug/tracing/latency_trace) + */ static void -trace_print_seq(struct seq_file *m, struct trace_seq *s) +__update_max_tr(struct trace_array *tr, struct task_struct *tsk, int cpu) { - int len = s->len >= PAGE_SIZE ? PAGE_SIZE - 1 : s->len; + struct trace_array_cpu *data = tr->data[cpu]; + struct trace_array_cpu *max_data = tr->data[cpu]; - s->buffer[len] = 0; - seq_puts(m, s->buffer); + max_tr.cpu = cpu; + max_tr.time_start = data->preempt_timestamp; + + max_data = max_tr.data[cpu]; + max_data->saved_latency = tracing_max_latency; + max_data->critical_start = data->critical_start; + max_data->critical_end = data->critical_end; - trace_seq_reset(s); + memcpy(data->comm, tsk->comm, TASK_COMM_LEN); + max_data->pid = tsk->pid; + max_data->uid = task_uid(tsk); + max_data->nice = tsk->static_prio - 20 - MAX_RT_PRIO; + max_data->policy = tsk->policy; + max_data->rt_priority = tsk->rt_priority; + + /* record this tasks comm */ + tracing_record_cmdline(tsk); } /** @@ -506,16 +551,15 @@ update_max_tr(struct trace_array *tr, struct task_struct *tsk, int cpu) { struct ring_buffer *buf = tr->buffer; + if (trace_stop_count) + return; + WARN_ON_ONCE(!irqs_disabled()); __raw_spin_lock(&ftrace_max_lock); tr->buffer = max_tr.buffer; max_tr.buffer = buf; - ftrace_disable_cpu(); - ring_buffer_reset(tr->buffer); - ftrace_enable_cpu(); - __update_max_tr(tr, tsk, cpu); __raw_spin_unlock(&ftrace_max_lock); } @@ -533,21 +577,35 @@ update_max_tr_single(struct trace_array *tr, struct task_struct *tsk, int cpu) { int ret; + if (trace_stop_count) + return; + WARN_ON_ONCE(!irqs_disabled()); __raw_spin_lock(&ftrace_max_lock); ftrace_disable_cpu(); - ring_buffer_reset(max_tr.buffer); ret = ring_buffer_swap_cpu(max_tr.buffer, tr->buffer, cpu); + if (ret == -EBUSY) { + /* + * We failed to swap the buffer due to a commit taking + * place on this CPU. We fail to record, but we reset + * the max trace buffer (no one writes directly to it) + * and flag that it failed. + */ + trace_array_printk(&max_tr, _THIS_IP_, + "Failed to swap buffers due to commit in progress\n"); + } + ftrace_enable_cpu(); - WARN_ON_ONCE(ret); + WARN_ON_ONCE(ret && ret != -EAGAIN && ret != -EBUSY); __update_max_tr(tr, tsk, cpu); __raw_spin_unlock(&ftrace_max_lock); } +#endif /* CONFIG_TRACER_MAX_TRACE */ /** * register_tracer - register a tracer with the ftrace system. @@ -556,9 +614,10 @@ update_max_tr_single(struct trace_array *tr, struct task_struct *tsk, int cpu) * Register a new plugin tracer. */ int register_tracer(struct tracer *type) +__releases(kernel_lock) +__acquires(kernel_lock) { struct tracer *t; - int len; int ret = 0; if (!type->name) { @@ -566,6 +625,11 @@ int register_tracer(struct tracer *type) return -1; } + if (strlen(type->name) > MAX_TRACER_SIZE) { + pr_info("Tracer has a name longer than %d\n", MAX_TRACER_SIZE); + return -1; + } + /* * When this gets called we hold the BKL which means that * preemption is disabled. Various trace selftests however @@ -580,7 +644,7 @@ int register_tracer(struct tracer *type) for (t = trace_types; t; t = t->next) { if (strcmp(type->name, t->name) == 0) { /* already found */ - pr_info("Trace %s already registered\n", + pr_info("Tracer %s already registered\n", type->name); ret = -1; goto out; @@ -594,12 +658,14 @@ int register_tracer(struct tracer *type) else if (!type->flags->opts) type->flags->opts = dummy_tracer_opt; + if (!type->wait_pipe) + type->wait_pipe = default_wait_pipe; + #ifdef CONFIG_FTRACE_STARTUP_TEST - if (type->selftest) { + if (type->selftest && !tracing_selftest_disabled) { struct tracer *saved_tracer = current_trace; struct trace_array *tr = &global_trace; - int i; /* * Run a selftest on this tracer. @@ -608,8 +674,7 @@ int register_tracer(struct tracer *type) * internal tracing to verify that everything is in order. * If we fail, we do not register this tracer. */ - for_each_tracing_cpu(i) - tracing_reset(tr, i); + tracing_reset_online_cpus(tr); current_trace = type; /* the test is responsible for initializing and enabling */ @@ -622,8 +687,7 @@ int register_tracer(struct tracer *type) goto out; } /* Only reset on passing, to avoid touching corrupted buffers */ - for_each_tracing_cpu(i) - tracing_reset(tr, i); + tracing_reset_online_cpus(tr); printk(KERN_CONT "PASSED\n"); } @@ -631,82 +695,129 @@ int register_tracer(struct tracer *type) type->next = trace_types; trace_types = type; - len = strlen(type->name); - if (len > max_tracer_type_len) - max_tracer_type_len = len; out: tracing_selftest_running = false; mutex_unlock(&trace_types_lock); - lock_kernel(); + if (ret || !default_bootup_tracer) + goto out_unlock; + + if (strncmp(default_bootup_tracer, type->name, MAX_TRACER_SIZE)) + goto out_unlock; + + printk(KERN_INFO "Starting tracer '%s'\n", type->name); + /* Do we want this tracer to start on bootup? */ + tracing_set_tracer(type->name); + default_bootup_tracer = NULL; + /* disable other selftests, since this will break it. */ + tracing_selftest_disabled = 1; +#ifdef CONFIG_FTRACE_STARTUP_TEST + printk(KERN_INFO "Disabling FTRACE selftests due to running tracer '%s'\n", + type->name); +#endif + + out_unlock: + lock_kernel(); return ret; } void unregister_tracer(struct tracer *type) { struct tracer **t; - int len; mutex_lock(&trace_types_lock); for (t = &trace_types; *t; t = &(*t)->next) { if (*t == type) goto found; } - pr_info("Trace %s not registered\n", type->name); + pr_info("Tracer %s not registered\n", type->name); goto out; found: *t = (*t)->next; - if (strlen(type->name) != max_tracer_type_len) - goto out; - max_tracer_type_len = 0; - for (t = &trace_types; *t; t = &(*t)->next) { - len = strlen((*t)->name); - if (len > max_tracer_type_len) - max_tracer_type_len = len; + if (type == current_trace && tracer_enabled) { + tracer_enabled = 0; + tracing_stop(); + if (current_trace->stop) + current_trace->stop(&global_trace); + current_trace = &nop_trace; } - out: +out: mutex_unlock(&trace_types_lock); } -void tracing_reset(struct trace_array *tr, int cpu) +static void __tracing_reset(struct trace_array *tr, int cpu) { ftrace_disable_cpu(); ring_buffer_reset_cpu(tr->buffer, cpu); ftrace_enable_cpu(); } +void tracing_reset(struct trace_array *tr, int cpu) +{ + struct ring_buffer *buffer = tr->buffer; + + ring_buffer_record_disable(buffer); + + /* Make sure all commits have finished */ + synchronize_sched(); + __tracing_reset(tr, cpu); + + ring_buffer_record_enable(buffer); +} + void tracing_reset_online_cpus(struct trace_array *tr) { + struct ring_buffer *buffer = tr->buffer; int cpu; + ring_buffer_record_disable(buffer); + + /* Make sure all commits have finished */ + synchronize_sched(); + tr->time_start = ftrace_now(tr->cpu); for_each_online_cpu(cpu) - tracing_reset(tr, cpu); + __tracing_reset(tr, cpu); + + ring_buffer_record_enable(buffer); +} + +void tracing_reset_current(int cpu) +{ + tracing_reset(&global_trace, cpu); +} + +void tracing_reset_current_online_cpus(void) +{ + tracing_reset_online_cpus(&global_trace); } #define SAVED_CMDLINES 128 +#define NO_CMDLINE_MAP UINT_MAX static unsigned map_pid_to_cmdline[PID_MAX_DEFAULT+1]; static unsigned map_cmdline_to_pid[SAVED_CMDLINES]; static char saved_cmdlines[SAVED_CMDLINES][TASK_COMM_LEN]; static int cmdline_idx; -static DEFINE_SPINLOCK(trace_cmdline_lock); +static raw_spinlock_t trace_cmdline_lock = __RAW_SPIN_LOCK_UNLOCKED; /* temporary disable recording */ -atomic_t trace_record_cmdline_disabled __read_mostly; +static atomic_t trace_record_cmdline_disabled __read_mostly; static void trace_init_cmdlines(void) { - memset(&map_pid_to_cmdline, -1, sizeof(map_pid_to_cmdline)); - memset(&map_cmdline_to_pid, -1, sizeof(map_cmdline_to_pid)); + memset(&map_pid_to_cmdline, NO_CMDLINE_MAP, sizeof(map_pid_to_cmdline)); + memset(&map_cmdline_to_pid, NO_CMDLINE_MAP, sizeof(map_cmdline_to_pid)); cmdline_idx = 0; } -static int trace_stop_count; -static DEFINE_SPINLOCK(tracing_start_lock); +int is_tracing_stopped(void) +{ + return trace_stop_count; +} /** * ftrace_off_permanent - disable all ftrace code permanently @@ -738,13 +849,12 @@ void tracing_start(void) return; spin_lock_irqsave(&tracing_start_lock, flags); - if (--trace_stop_count) - goto out; - - if (trace_stop_count < 0) { - /* Someone screwed up their debugging */ - WARN_ON_ONCE(1); - trace_stop_count = 0; + if (--trace_stop_count) { + if (trace_stop_count < 0) { + /* Someone screwed up their debugging */ + WARN_ON_ONCE(1); + trace_stop_count = 0; + } goto out; } @@ -794,8 +904,7 @@ void trace_stop_cmdline_recording(void); static void trace_save_cmdline(struct task_struct *tsk) { - unsigned map; - unsigned idx; + unsigned pid, idx; if (!tsk->pid || unlikely(tsk->pid > PID_MAX_DEFAULT)) return; @@ -806,17 +915,24 @@ static void trace_save_cmdline(struct task_struct *tsk) * nor do we want to disable interrupts, * so if we miss here, then better luck next time. */ - if (!spin_trylock(&trace_cmdline_lock)) + if (!__raw_spin_trylock(&trace_cmdline_lock)) return; idx = map_pid_to_cmdline[tsk->pid]; - if (idx >= SAVED_CMDLINES) { + if (idx == NO_CMDLINE_MAP) { idx = (cmdline_idx + 1) % SAVED_CMDLINES; - map = map_cmdline_to_pid[idx]; - if (map <= PID_MAX_DEFAULT) - map_pid_to_cmdline[map] = (unsigned)-1; + /* + * Check whether the cmdline buffer at idx has a pid + * mapped. We are going to overwrite that entry so we + * need to clear the map_pid_to_cmdline. Otherwise we + * would read the new comm for the old pid. + */ + pid = map_cmdline_to_pid[idx]; + if (pid != NO_CMDLINE_MAP) + map_pid_to_cmdline[pid] = NO_CMDLINE_MAP; + map_cmdline_to_pid[idx] = tsk->pid; map_pid_to_cmdline[tsk->pid] = idx; cmdline_idx = idx; @@ -824,33 +940,39 @@ static void trace_save_cmdline(struct task_struct *tsk) memcpy(&saved_cmdlines[idx], tsk->comm, TASK_COMM_LEN); - spin_unlock(&trace_cmdline_lock); + __raw_spin_unlock(&trace_cmdline_lock); } -char *trace_find_cmdline(int pid) +void trace_find_cmdline(int pid, char comm[]) { - char *cmdline = "<...>"; unsigned map; - if (!pid) - return "<idle>"; + if (!pid) { + strcpy(comm, "<idle>"); + return; + } - if (pid > PID_MAX_DEFAULT) - goto out; + if (pid > PID_MAX_DEFAULT) { + strcpy(comm, "<...>"); + return; + } + preempt_disable(); + __raw_spin_lock(&trace_cmdline_lock); map = map_pid_to_cmdline[pid]; - if (map >= SAVED_CMDLINES) - goto out; - - cmdline = saved_cmdlines[map]; + if (map != NO_CMDLINE_MAP) + strcpy(comm, saved_cmdlines[map]); + else + strcpy(comm, "<...>"); - out: - return cmdline; + __raw_spin_unlock(&trace_cmdline_lock); + preempt_enable(); } void tracing_record_cmdline(struct task_struct *tsk) { - if (atomic_read(&trace_record_cmdline_disabled)) + if (atomic_read(&trace_record_cmdline_disabled) || !tracer_enabled || + !tracing_is_on()) return; trace_save_cmdline(tsk); @@ -864,7 +986,7 @@ tracing_generic_entry_update(struct trace_entry *entry, unsigned long flags, entry->preempt_count = pc & 0xff; entry->pid = (tsk) ? tsk->pid : 0; - entry->tgid = (tsk) ? tsk->tgid : 0; + entry->lock_depth = (tsk) ? tsk->lock_depth : 0; entry->flags = #ifdef CONFIG_TRACE_IRQFLAGS_SUPPORT (irqs_disabled_flags(flags) ? TRACE_FLAG_IRQS_OFF : 0) | @@ -875,81 +997,108 @@ tracing_generic_entry_update(struct trace_entry *entry, unsigned long flags, ((pc & SOFTIRQ_MASK) ? TRACE_FLAG_SOFTIRQ : 0) | (need_resched() ? TRACE_FLAG_NEED_RESCHED : 0); } +EXPORT_SYMBOL_GPL(tracing_generic_entry_update); -void -trace_function(struct trace_array *tr, struct trace_array_cpu *data, - unsigned long ip, unsigned long parent_ip, unsigned long flags, - int pc) +struct ring_buffer_event * +trace_buffer_lock_reserve(struct ring_buffer *buffer, + int type, + unsigned long len, + unsigned long flags, int pc) { struct ring_buffer_event *event; - struct ftrace_entry *entry; - unsigned long irq_flags; - /* If we are reading the ring buffer, don't trace */ - if (unlikely(local_read(&__get_cpu_var(ftrace_cpu_disabled)))) - return; + event = ring_buffer_lock_reserve(buffer, len); + if (event != NULL) { + struct trace_entry *ent = ring_buffer_event_data(event); - event = ring_buffer_lock_reserve(tr->buffer, sizeof(*entry), - &irq_flags); - if (!event) - return; - entry = ring_buffer_event_data(event); - tracing_generic_entry_update(&entry->ent, flags, pc); - entry->ent.type = TRACE_FN; - entry->ip = ip; - entry->parent_ip = parent_ip; - ring_buffer_unlock_commit(tr->buffer, event, irq_flags); + tracing_generic_entry_update(ent, flags, pc); + ent->type = type; + } + + return event; } -#ifdef CONFIG_FUNCTION_GRAPH_TRACER -static void __trace_graph_entry(struct trace_array *tr, - struct trace_array_cpu *data, - struct ftrace_graph_ent *trace, - unsigned long flags, - int pc) +static inline void +__trace_buffer_unlock_commit(struct ring_buffer *buffer, + struct ring_buffer_event *event, + unsigned long flags, int pc, + int wake) { - struct ring_buffer_event *event; - struct ftrace_graph_ent_entry *entry; - unsigned long irq_flags; + ring_buffer_unlock_commit(buffer, event); - if (unlikely(local_read(&__get_cpu_var(ftrace_cpu_disabled)))) - return; + ftrace_trace_stack(buffer, flags, 6, pc); + ftrace_trace_userstack(buffer, flags, pc); - event = ring_buffer_lock_reserve(global_trace.buffer, sizeof(*entry), - &irq_flags); - if (!event) - return; - entry = ring_buffer_event_data(event); - tracing_generic_entry_update(&entry->ent, flags, pc); - entry->ent.type = TRACE_GRAPH_ENT; - entry->graph_ent = *trace; - ring_buffer_unlock_commit(global_trace.buffer, event, irq_flags); + if (wake) + trace_wake_up(); +} + +void trace_buffer_unlock_commit(struct ring_buffer *buffer, + struct ring_buffer_event *event, + unsigned long flags, int pc) +{ + __trace_buffer_unlock_commit(buffer, event, flags, pc, 1); +} + +struct ring_buffer_event * +trace_current_buffer_lock_reserve(struct ring_buffer **current_rb, + int type, unsigned long len, + unsigned long flags, int pc) +{ + *current_rb = global_trace.buffer; + return trace_buffer_lock_reserve(*current_rb, + type, len, flags, pc); +} +EXPORT_SYMBOL_GPL(trace_current_buffer_lock_reserve); + +void trace_current_buffer_unlock_commit(struct ring_buffer *buffer, + struct ring_buffer_event *event, + unsigned long flags, int pc) +{ + __trace_buffer_unlock_commit(buffer, event, flags, pc, 1); } +EXPORT_SYMBOL_GPL(trace_current_buffer_unlock_commit); -static void __trace_graph_return(struct trace_array *tr, - struct trace_array_cpu *data, - struct ftrace_graph_ret *trace, - unsigned long flags, - int pc) +void trace_nowake_buffer_unlock_commit(struct ring_buffer *buffer, + struct ring_buffer_event *event, + unsigned long flags, int pc) { + __trace_buffer_unlock_commit(buffer, event, flags, pc, 0); +} +EXPORT_SYMBOL_GPL(trace_nowake_buffer_unlock_commit); + +void trace_current_buffer_discard_commit(struct ring_buffer *buffer, + struct ring_buffer_event *event) +{ + ring_buffer_discard_commit(buffer, event); +} +EXPORT_SYMBOL_GPL(trace_current_buffer_discard_commit); + +void +trace_function(struct trace_array *tr, + unsigned long ip, unsigned long parent_ip, unsigned long flags, + int pc) +{ + struct ftrace_event_call *call = &event_function; + struct ring_buffer *buffer = tr->buffer; struct ring_buffer_event *event; - struct ftrace_graph_ret_entry *entry; - unsigned long irq_flags; + struct ftrace_entry *entry; + /* If we are reading the ring buffer, don't trace */ if (unlikely(local_read(&__get_cpu_var(ftrace_cpu_disabled)))) return; - event = ring_buffer_lock_reserve(global_trace.buffer, sizeof(*entry), - &irq_flags); + event = trace_buffer_lock_reserve(buffer, TRACE_FN, sizeof(*entry), + flags, pc); if (!event) return; entry = ring_buffer_event_data(event); - tracing_generic_entry_update(&entry->ent, flags, pc); - entry->ent.type = TRACE_GRAPH_RET; - entry->ret = *trace; - ring_buffer_unlock_commit(global_trace.buffer, event, irq_flags); + entry->ip = ip; + entry->parent_ip = parent_ip; + + if (!filter_check_discard(call, entry, buffer, event)) + ring_buffer_unlock_commit(buffer, event); } -#endif void ftrace(struct trace_array *tr, struct trace_array_cpu *data, @@ -957,31 +1106,24 @@ ftrace(struct trace_array *tr, struct trace_array_cpu *data, int pc) { if (likely(!atomic_read(&data->disabled))) - trace_function(tr, data, ip, parent_ip, flags, pc); + trace_function(tr, ip, parent_ip, flags, pc); } -static void ftrace_trace_stack(struct trace_array *tr, - struct trace_array_cpu *data, - unsigned long flags, - int skip, int pc) -{ #ifdef CONFIG_STACKTRACE +static void __ftrace_trace_stack(struct ring_buffer *buffer, + unsigned long flags, + int skip, int pc) +{ + struct ftrace_event_call *call = &event_kernel_stack; struct ring_buffer_event *event; struct stack_entry *entry; struct stack_trace trace; - unsigned long irq_flags; - - if (!(trace_flags & TRACE_ITER_STACKTRACE)) - return; - event = ring_buffer_lock_reserve(tr->buffer, sizeof(*entry), - &irq_flags); + event = trace_buffer_lock_reserve(buffer, TRACE_STACK, + sizeof(*entry), flags, pc); if (!event) return; entry = ring_buffer_event_data(event); - tracing_generic_entry_update(&entry->ent, flags, pc); - entry->ent.type = TRACE_STACK; - memset(&entry->caller, 0, sizeof(entry->caller)); trace.nr_entries = 0; @@ -990,39 +1132,43 @@ static void ftrace_trace_stack(struct trace_array *tr, trace.entries = entry->caller; save_stack_trace(&trace); - ring_buffer_unlock_commit(tr->buffer, event, irq_flags); -#endif + if (!filter_check_discard(call, entry, buffer, event)) + ring_buffer_unlock_commit(buffer, event); } -void __trace_stack(struct trace_array *tr, - struct trace_array_cpu *data, - unsigned long flags, - int skip) +void ftrace_trace_stack(struct ring_buffer *buffer, unsigned long flags, + int skip, int pc) { - ftrace_trace_stack(tr, data, flags, skip, preempt_count()); + if (!(trace_flags & TRACE_ITER_STACKTRACE)) + return; + + __ftrace_trace_stack(buffer, flags, skip, pc); } -static void ftrace_trace_userstack(struct trace_array *tr, - struct trace_array_cpu *data, - unsigned long flags, int pc) +void __trace_stack(struct trace_array *tr, unsigned long flags, int skip, + int pc) { -#ifdef CONFIG_STACKTRACE + __ftrace_trace_stack(tr->buffer, flags, skip, pc); +} + +void +ftrace_trace_userstack(struct ring_buffer *buffer, unsigned long flags, int pc) +{ + struct ftrace_event_call *call = &event_user_stack; struct ring_buffer_event *event; struct userstack_entry *entry; struct stack_trace trace; - unsigned long irq_flags; if (!(trace_flags & TRACE_ITER_USERSTACKTRACE)) return; - event = ring_buffer_lock_reserve(tr->buffer, sizeof(*entry), - &irq_flags); + event = trace_buffer_lock_reserve(buffer, TRACE_USER_STACK, + sizeof(*entry), flags, pc); if (!event) return; entry = ring_buffer_event_data(event); - tracing_generic_entry_update(&entry->ent, flags, pc); - entry->ent.type = TRACE_USER_STACK; + entry->tgid = current->tgid; memset(&entry->caller, 0, sizeof(entry->caller)); trace.nr_entries = 0; @@ -1031,112 +1177,48 @@ static void ftrace_trace_userstack(struct trace_array *tr, trace.entries = entry->caller; save_stack_trace_user(&trace); - ring_buffer_unlock_commit(tr->buffer, event, irq_flags); -#endif + if (!filter_check_discard(call, entry, buffer, event)) + ring_buffer_unlock_commit(buffer, event); } -void __trace_userstack(struct trace_array *tr, - struct trace_array_cpu *data, - unsigned long flags) +#ifdef UNUSED +static void __trace_userstack(struct trace_array *tr, unsigned long flags) { - ftrace_trace_userstack(tr, data, flags, preempt_count()); + ftrace_trace_userstack(tr, flags, preempt_count()); } +#endif /* UNUSED */ + +#endif /* CONFIG_STACKTRACE */ static void -ftrace_trace_special(void *__tr, void *__data, +ftrace_trace_special(void *__tr, unsigned long arg1, unsigned long arg2, unsigned long arg3, int pc) { + struct ftrace_event_call *call = &event_special; struct ring_buffer_event *event; - struct trace_array_cpu *data = __data; struct trace_array *tr = __tr; + struct ring_buffer *buffer = tr->buffer; struct special_entry *entry; - unsigned long irq_flags; - event = ring_buffer_lock_reserve(tr->buffer, sizeof(*entry), - &irq_flags); + event = trace_buffer_lock_reserve(buffer, TRACE_SPECIAL, + sizeof(*entry), 0, pc); if (!event) return; entry = ring_buffer_event_data(event); - tracing_generic_entry_update(&entry->ent, 0, pc); - entry->ent.type = TRACE_SPECIAL; entry->arg1 = arg1; entry->arg2 = arg2; entry->arg3 = arg3; - ring_buffer_unlock_commit(tr->buffer, event, irq_flags); - ftrace_trace_stack(tr, data, irq_flags, 4, pc); - ftrace_trace_userstack(tr, data, irq_flags, pc); - trace_wake_up(); + if (!filter_check_discard(call, entry, buffer, event)) + trace_buffer_unlock_commit(buffer, event, 0, pc); } void __trace_special(void *__tr, void *__data, unsigned long arg1, unsigned long arg2, unsigned long arg3) { - ftrace_trace_special(__tr, __data, arg1, arg2, arg3, preempt_count()); -} - -void -tracing_sched_switch_trace(struct trace_array *tr, - struct trace_array_cpu *data, - struct task_struct *prev, - struct task_struct *next, - unsigned long flags, int pc) -{ - struct ring_buffer_event *event; - struct ctx_switch_entry *entry; - unsigned long irq_flags; - - event = ring_buffer_lock_reserve(tr->buffer, sizeof(*entry), - &irq_flags); - if (!event) - return; - entry = ring_buffer_event_data(event); - tracing_generic_entry_update(&entry->ent, flags, pc); - entry->ent.type = TRACE_CTX; - entry->prev_pid = prev->pid; - entry->prev_prio = prev->prio; - entry->prev_state = prev->state; - entry->next_pid = next->pid; - entry->next_prio = next->prio; - entry->next_state = next->state; - entry->next_cpu = task_cpu(next); - ring_buffer_unlock_commit(tr->buffer, event, irq_flags); - ftrace_trace_stack(tr, data, flags, 5, pc); - ftrace_trace_userstack(tr, data, flags, pc); -} - -void -tracing_sched_wakeup_trace(struct trace_array *tr, - struct trace_array_cpu *data, - struct task_struct *wakee, - struct task_struct *curr, - unsigned long flags, int pc) -{ - struct ring_buffer_event *event; - struct ctx_switch_entry *entry; - unsigned long irq_flags; - - event = ring_buffer_lock_reserve(tr->buffer, sizeof(*entry), - &irq_flags); - if (!event) - return; - entry = ring_buffer_event_data(event); - tracing_generic_entry_update(&entry->ent, flags, pc); - entry->ent.type = TRACE_WAKE; - entry->prev_pid = curr->pid; - entry->prev_prio = curr->prio; - entry->prev_state = curr->state; - entry->next_pid = wakee->pid; - entry->next_prio = wakee->prio; - entry->next_state = wakee->state; - entry->next_cpu = task_cpu(wakee); - ring_buffer_unlock_commit(tr->buffer, event, irq_flags); - ftrace_trace_stack(tr, data, flags, 6, pc); - ftrace_trace_userstack(tr, data, flags, pc); - - trace_wake_up(); + ftrace_trace_special(__tr, arg1, arg2, arg3, preempt_count()); } void @@ -1157,152 +1239,164 @@ ftrace_special(unsigned long arg1, unsigned long arg2, unsigned long arg3) data = tr->data[cpu]; if (likely(atomic_inc_return(&data->disabled) == 1)) - ftrace_trace_special(tr, data, arg1, arg2, arg3, pc); + ftrace_trace_special(tr, arg1, arg2, arg3, pc); atomic_dec(&data->disabled); local_irq_restore(flags); } -#ifdef CONFIG_FUNCTION_TRACER -static void -function_trace_call_preempt_only(unsigned long ip, unsigned long parent_ip) +/** + * trace_vbprintk - write binary msg to tracing buffer + * + */ +int trace_vbprintk(unsigned long ip, const char *fmt, va_list args) { + static raw_spinlock_t trace_buf_lock = + (raw_spinlock_t)__RAW_SPIN_LOCK_UNLOCKED; + static u32 trace_buf[TRACE_BUF_SIZE]; + + struct ftrace_event_call *call = &event_bprint; + struct ring_buffer_event *event; + struct ring_buffer *buffer; struct trace_array *tr = &global_trace; struct trace_array_cpu *data; + struct bprint_entry *entry; unsigned long flags; - long disabled; - int cpu, resched; - int pc; + int disable; + int resched; + int cpu, len = 0, size, pc; - if (unlikely(!ftrace_function_enabled)) - return; + if (unlikely(tracing_selftest_running || tracing_disabled)) + return 0; + + /* Don't pollute graph traces with trace_vprintk internals */ + pause_graph_tracing(); pc = preempt_count(); resched = ftrace_preempt_disable(); - local_save_flags(flags); cpu = raw_smp_processor_id(); data = tr->data[cpu]; - disabled = atomic_inc_return(&data->disabled); - if (likely(disabled == 1)) - trace_function(tr, data, ip, parent_ip, flags, pc); - - atomic_dec(&data->disabled); - ftrace_preempt_enable(resched); -} + disable = atomic_inc_return(&data->disabled); + if (unlikely(disable != 1)) + goto out; -static void -function_trace_call(unsigned long ip, unsigned long parent_ip) -{ - struct trace_array *tr = &global_trace; - struct trace_array_cpu *data; - unsigned long flags; - long disabled; - int cpu; - int pc; + /* Lockdep uses trace_printk for lock tracing */ + local_irq_save(flags); + __raw_spin_lock(&trace_buf_lock); + len = vbin_printf(trace_buf, TRACE_BUF_SIZE, fmt, args); - if (unlikely(!ftrace_function_enabled)) - return; + if (len > TRACE_BUF_SIZE || len < 0) + goto out_unlock; - /* - * Need to use raw, since this must be called before the - * recursive protection is performed. - */ - local_irq_save(flags); - cpu = raw_smp_processor_id(); - data = tr->data[cpu]; - disabled = atomic_inc_return(&data->disabled); + size = sizeof(*entry) + sizeof(u32) * len; + buffer = tr->buffer; + event = trace_buffer_lock_reserve(buffer, TRACE_BPRINT, size, + flags, pc); + if (!event) + goto out_unlock; + entry = ring_buffer_event_data(event); + entry->ip = ip; + entry->fmt = fmt; - if (likely(disabled == 1)) { - pc = preempt_count(); - trace_function(tr, data, ip, parent_ip, flags, pc); - } + memcpy(entry->buf, trace_buf, sizeof(u32) * len); + if (!filter_check_discard(call, entry, buffer, event)) + ring_buffer_unlock_commit(buffer, event); - atomic_dec(&data->disabled); +out_unlock: + __raw_spin_unlock(&trace_buf_lock); local_irq_restore(flags); + +out: + atomic_dec_return(&data->disabled); + ftrace_preempt_enable(resched); + unpause_graph_tracing(); + + return len; } +EXPORT_SYMBOL_GPL(trace_vbprintk); -#ifdef CONFIG_FUNCTION_GRAPH_TRACER -int trace_graph_entry(struct ftrace_graph_ent *trace) +int trace_array_printk(struct trace_array *tr, + unsigned long ip, const char *fmt, ...) { - struct trace_array *tr = &global_trace; - struct trace_array_cpu *data; - unsigned long flags; - long disabled; - int cpu; - int pc; - - if (!ftrace_trace_task(current)) - return 0; + int ret; + va_list ap; - if (!ftrace_graph_addr(trace->func)) + if (!(trace_flags & TRACE_ITER_PRINTK)) return 0; - local_irq_save(flags); - cpu = raw_smp_processor_id(); - data = tr->data[cpu]; - disabled = atomic_inc_return(&data->disabled); - if (likely(disabled == 1)) { - pc = preempt_count(); - __trace_graph_entry(tr, data, trace, flags, pc); - } - /* Only do the atomic if it is not already set */ - if (!test_tsk_trace_graph(current)) - set_tsk_trace_graph(current); - atomic_dec(&data->disabled); - local_irq_restore(flags); - - return 1; + va_start(ap, fmt); + ret = trace_array_vprintk(tr, ip, fmt, ap); + va_end(ap); + return ret; } -void trace_graph_return(struct ftrace_graph_ret *trace) +int trace_array_vprintk(struct trace_array *tr, + unsigned long ip, const char *fmt, va_list args) { - struct trace_array *tr = &global_trace; + static raw_spinlock_t trace_buf_lock = __RAW_SPIN_LOCK_UNLOCKED; + static char trace_buf[TRACE_BUF_SIZE]; + + struct ftrace_event_call *call = &event_print; + struct ring_buffer_event *event; + struct ring_buffer *buffer; struct trace_array_cpu *data; - unsigned long flags; - long disabled; - int cpu; - int pc; + int cpu, len = 0, size, pc; + struct print_entry *entry; + unsigned long irq_flags; + int disable; - local_irq_save(flags); + if (tracing_disabled || tracing_selftest_running) + return 0; + + pc = preempt_count(); + preempt_disable_notrace(); cpu = raw_smp_processor_id(); data = tr->data[cpu]; - disabled = atomic_inc_return(&data->disabled); - if (likely(disabled == 1)) { - pc = preempt_count(); - __trace_graph_return(tr, data, trace, flags, pc); - } - if (!trace->depth) - clear_tsk_trace_graph(current); - atomic_dec(&data->disabled); - local_irq_restore(flags); -} -#endif /* CONFIG_FUNCTION_GRAPH_TRACER */ -static struct ftrace_ops trace_ops __read_mostly = -{ - .func = function_trace_call, -}; + disable = atomic_inc_return(&data->disabled); + if (unlikely(disable != 1)) + goto out; -void tracing_start_function_trace(void) -{ - ftrace_function_enabled = 0; + pause_graph_tracing(); + raw_local_irq_save(irq_flags); + __raw_spin_lock(&trace_buf_lock); + if (args == NULL) { + strncpy(trace_buf, fmt, TRACE_BUF_SIZE); + len = strlen(trace_buf); + } else + len = vsnprintf(trace_buf, TRACE_BUF_SIZE, fmt, args); - if (trace_flags & TRACE_ITER_PREEMPTONLY) - trace_ops.func = function_trace_call_preempt_only; - else - trace_ops.func = function_trace_call; + size = sizeof(*entry) + len + 1; + buffer = tr->buffer; + event = trace_buffer_lock_reserve(buffer, TRACE_PRINT, size, + irq_flags, pc); + if (!event) + goto out_unlock; + entry = ring_buffer_event_data(event); + entry->ip = ip; + + memcpy(&entry->buf, trace_buf, len); + entry->buf[len] = '\0'; + if (!filter_check_discard(call, entry, buffer, event)) + ring_buffer_unlock_commit(buffer, event); + + out_unlock: + __raw_spin_unlock(&trace_buf_lock); + raw_local_irq_restore(irq_flags); + unpause_graph_tracing(); + out: + atomic_dec_return(&data->disabled); + preempt_enable_notrace(); - register_ftrace_function(&trace_ops); - ftrace_function_enabled = 1; + return len; } -void tracing_stop_function_trace(void) +int trace_vprintk(unsigned long ip, const char *fmt, va_list args) { - ftrace_function_enabled = 0; - unregister_ftrace_function(&trace_ops); + return trace_array_vprintk(&global_trace, ip, fmt, args); } -#endif +EXPORT_SYMBOL_GPL(trace_vprintk); enum trace_file_type { TRACE_FILE_LAT_FMT = 1, @@ -1345,10 +1439,25 @@ __find_next_entry(struct trace_iterator *iter, int *ent_cpu, u64 *ent_ts) { struct ring_buffer *buffer = iter->tr->buffer; struct trace_entry *ent, *next = NULL; + int cpu_file = iter->cpu_file; u64 next_ts = 0, ts; int next_cpu = -1; int cpu; + /* + * If we are in a per_cpu trace file, don't bother by iterating over + * all cpu and peek directly. + */ + if (cpu_file > TRACE_PIPE_ALL_CPU) { + if (ring_buffer_empty_cpu(buffer, cpu_file)) + return NULL; + ent = peek_next_entry(iter, cpu_file, ent_ts); + if (ent_cpu) + *ent_cpu = cpu_file; + + return ent; + } + for_each_tracing_cpu(cpu) { if (ring_buffer_empty_cpu(buffer, cpu)) @@ -1376,8 +1485,8 @@ __find_next_entry(struct trace_iterator *iter, int *ent_cpu, u64 *ent_ts) } /* Find the next real entry, without updating the iterator itself */ -static struct trace_entry * -find_next_entry(struct trace_iterator *iter, int *ent_cpu, u64 *ent_ts) +struct trace_entry *trace_find_next_entry(struct trace_iterator *iter, + int *ent_cpu, u64 *ent_ts) { return __find_next_entry(iter, ent_cpu, ent_ts); } @@ -1426,19 +1535,63 @@ static void *s_next(struct seq_file *m, void *v, loff_t *pos) return ent; } +static void tracing_iter_reset(struct trace_iterator *iter, int cpu) +{ + struct trace_array *tr = iter->tr; + struct ring_buffer_event *event; + struct ring_buffer_iter *buf_iter; + unsigned long entries = 0; + u64 ts; + + tr->data[cpu]->skipped_entries = 0; + + if (!iter->buffer_iter[cpu]) + return; + + buf_iter = iter->buffer_iter[cpu]; + ring_buffer_iter_reset(buf_iter); + + /* + * We could have the case with the max latency tracers + * that a reset never took place on a cpu. This is evident + * by the timestamp being before the start of the buffer. + */ + while ((event = ring_buffer_iter_peek(buf_iter, &ts))) { + if (ts >= iter->tr->time_start) + break; + entries++; + ring_buffer_read(buf_iter, NULL); + } + + tr->data[cpu]->skipped_entries = entries; +} + +/* + * No necessary locking here. The worst thing which can + * happen is loosing events consumed at the same time + * by a trace_pipe reader. + * Other than that, we don't risk to crash the ring buffer + * because it serializes the readers. + * + * The current tracer is copied to avoid a global locking + * all around. + */ static void *s_start(struct seq_file *m, loff_t *pos) { struct trace_iterator *iter = m->private; + static struct tracer *old_tracer; + int cpu_file = iter->cpu_file; void *p = NULL; loff_t l = 0; int cpu; + /* copy the tracer to avoid using a global lock all around */ mutex_lock(&trace_types_lock); - - if (!current_trace || current_trace != iter->trace) { - mutex_unlock(&trace_types_lock); - return NULL; + if (unlikely(old_tracer != current_trace && current_trace)) { + old_tracer = current_trace; + *iter->trace = *current_trace; } + mutex_unlock(&trace_types_lock); atomic_inc(&trace_record_cmdline_disabled); @@ -1449,9 +1602,11 @@ static void *s_start(struct seq_file *m, loff_t *pos) ftrace_disable_cpu(); - for_each_tracing_cpu(cpu) { - ring_buffer_iter_reset(iter->buffer_iter[cpu]); - } + if (cpu_file == TRACE_PIPE_ALL_CPU) { + for_each_tracing_cpu(cpu) + tracing_iter_reset(iter, cpu); + } else + tracing_iter_reset(iter, cpu_file); ftrace_enable_cpu(); @@ -1463,161 +1618,14 @@ static void *s_start(struct seq_file *m, loff_t *pos) p = s_next(m, p, &l); } + trace_event_read_lock(); return p; } static void s_stop(struct seq_file *m, void *p) { atomic_dec(&trace_record_cmdline_disabled); - mutex_unlock(&trace_types_lock); -} - -#ifdef CONFIG_KRETPROBES -static inline const char *kretprobed(const char *name) -{ - static const char tramp_name[] = "kretprobe_trampoline"; - int size = sizeof(tramp_name); - - if (strncmp(tramp_name, name, size) == 0) - return "[unknown/kretprobe'd]"; - return name; -} -#else -static inline const char *kretprobed(const char *name) -{ - return name; -} -#endif /* CONFIG_KRETPROBES */ - -static int -seq_print_sym_short(struct trace_seq *s, const char *fmt, unsigned long address) -{ -#ifdef CONFIG_KALLSYMS - char str[KSYM_SYMBOL_LEN]; - const char *name; - - kallsyms_lookup(address, NULL, NULL, NULL, str); - - name = kretprobed(str); - - return trace_seq_printf(s, fmt, name); -#endif - return 1; -} - -static int -seq_print_sym_offset(struct trace_seq *s, const char *fmt, - unsigned long address) -{ -#ifdef CONFIG_KALLSYMS - char str[KSYM_SYMBOL_LEN]; - const char *name; - - sprint_symbol(str, address); - name = kretprobed(str); - - return trace_seq_printf(s, fmt, name); -#endif - return 1; -} - -#ifndef CONFIG_64BIT -# define IP_FMT "%08lx" -#else -# define IP_FMT "%016lx" -#endif - -int -seq_print_ip_sym(struct trace_seq *s, unsigned long ip, unsigned long sym_flags) -{ - int ret; - - if (!ip) - return trace_seq_printf(s, "0"); - - if (sym_flags & TRACE_ITER_SYM_OFFSET) - ret = seq_print_sym_offset(s, "%s", ip); - else - ret = seq_print_sym_short(s, "%s", ip); - - if (!ret) - return 0; - - if (sym_flags & TRACE_ITER_SYM_ADDR) - ret = trace_seq_printf(s, " <" IP_FMT ">", ip); - return ret; -} - -static inline int seq_print_user_ip(struct trace_seq *s, struct mm_struct *mm, - unsigned long ip, unsigned long sym_flags) -{ - struct file *file = NULL; - unsigned long vmstart = 0; - int ret = 1; - - if (mm) { - const struct vm_area_struct *vma; - - down_read(&mm->mmap_sem); - vma = find_vma(mm, ip); - if (vma) { - file = vma->vm_file; - vmstart = vma->vm_start; - } - if (file) { - ret = trace_seq_path(s, &file->f_path); - if (ret) - ret = trace_seq_printf(s, "[+0x%lx]", ip - vmstart); - } - up_read(&mm->mmap_sem); - } - if (ret && ((sym_flags & TRACE_ITER_SYM_ADDR) || !file)) - ret = trace_seq_printf(s, " <" IP_FMT ">", ip); - return ret; -} - -static int -seq_print_userip_objs(const struct userstack_entry *entry, struct trace_seq *s, - unsigned long sym_flags) -{ - struct mm_struct *mm = NULL; - int ret = 1; - unsigned int i; - - if (trace_flags & TRACE_ITER_SYM_USEROBJ) { - struct task_struct *task; - /* - * we do the lookup on the thread group leader, - * since individual threads might have already quit! - */ - rcu_read_lock(); - task = find_task_by_vpid(entry->ent.tgid); - if (task) - mm = get_task_mm(task); - rcu_read_unlock(); - } - - for (i = 0; i < FTRACE_STACK_ENTRIES; i++) { - unsigned long ip = entry->caller[i]; - - if (ip == ULONG_MAX || !ret) - break; - if (i && ret) - ret = trace_seq_puts(s, " <- "); - if (!ip) { - if (ret) - ret = trace_seq_puts(s, "??"); - continue; - } - if (!ret) - break; - if (ret) - ret = seq_print_user_ip(s, mm, ip, sym_flags); - } - - if (mm) - mmput(mm); - return ret; + trace_event_read_unlock(); } static void print_lat_help_header(struct seq_file *m) @@ -1627,10 +1635,10 @@ static void print_lat_help_header(struct seq_file *m) seq_puts(m, "# | / _----=> need-resched \n"); seq_puts(m, "# || / _---=> hardirq/softirq \n"); seq_puts(m, "# ||| / _--=> preempt-depth \n"); - seq_puts(m, "# |||| / \n"); - seq_puts(m, "# ||||| delay \n"); - seq_puts(m, "# cmd pid ||||| time | caller \n"); - seq_puts(m, "# \\ / ||||| \\ | / \n"); + seq_puts(m, "# |||| /_--=> lock-depth \n"); + seq_puts(m, "# |||||/ delay \n"); + seq_puts(m, "# cmd pid |||||| time | caller \n"); + seq_puts(m, "# \\ / |||||| \\ | / \n"); } static void print_func_help_header(struct seq_file *m) @@ -1647,22 +1655,38 @@ print_trace_header(struct seq_file *m, struct trace_iterator *iter) struct trace_array *tr = iter->tr; struct trace_array_cpu *data = tr->data[tr->cpu]; struct tracer *type = current_trace; - unsigned long total; - unsigned long entries; + unsigned long entries = 0; + unsigned long total = 0; + unsigned long count; const char *name = "preemption"; + int cpu; if (type) name = type->name; - entries = ring_buffer_entries(iter->tr->buffer); - total = entries + - ring_buffer_overruns(iter->tr->buffer); - seq_printf(m, "%s latency trace v1.1.5 on %s\n", + for_each_tracing_cpu(cpu) { + count = ring_buffer_entries_cpu(tr->buffer, cpu); + /* + * If this buffer has skipped entries, then we hold all + * entries for the trace and we need to ignore the + * ones before the time stamp. + */ + if (tr->data[cpu]->skipped_entries) { + count -= tr->data[cpu]->skipped_entries; + /* total is the same as the entries */ + total += count; + } else + total += count + + ring_buffer_overrun_cpu(tr->buffer, cpu); + entries += count; + } + + seq_printf(m, "# %s latency trace v1.1.5 on %s\n", name, UTS_RELEASE); - seq_puts(m, "-----------------------------------" + seq_puts(m, "# -----------------------------------" "---------------------------------\n"); - seq_printf(m, " latency: %lu us, #%lu/%lu, CPU#%d |" + seq_printf(m, "# latency: %lu us, #%lu/%lu, CPU#%d |" " (M:%s VP:%d, KP:%d, SP:%d HP:%d", nsecs_to_usecs(data->saved_latency), entries, @@ -1684,121 +1708,24 @@ print_trace_header(struct seq_file *m, struct trace_iterator *iter) #else seq_puts(m, ")\n"); #endif - seq_puts(m, " -----------------\n"); - seq_printf(m, " | task: %.16s-%d " + seq_puts(m, "# -----------------\n"); + seq_printf(m, "# | task: %.16s-%d " "(uid:%d nice:%ld policy:%ld rt_prio:%ld)\n", data->comm, data->pid, data->uid, data->nice, data->policy, data->rt_priority); - seq_puts(m, " -----------------\n"); + seq_puts(m, "# -----------------\n"); if (data->critical_start) { - seq_puts(m, " => started at: "); + seq_puts(m, "# => started at: "); seq_print_ip_sym(&iter->seq, data->critical_start, sym_flags); trace_print_seq(m, &iter->seq); - seq_puts(m, "\n => ended at: "); + seq_puts(m, "\n# => ended at: "); seq_print_ip_sym(&iter->seq, data->critical_end, sym_flags); trace_print_seq(m, &iter->seq); - seq_puts(m, "\n"); + seq_puts(m, "\n#\n"); } - seq_puts(m, "\n"); -} - -static void -lat_print_generic(struct trace_seq *s, struct trace_entry *entry, int cpu) -{ - int hardirq, softirq; - char *comm; - - comm = trace_find_cmdline(entry->pid); - - trace_seq_printf(s, "%8.8s-%-5d ", comm, entry->pid); - trace_seq_printf(s, "%3d", cpu); - trace_seq_printf(s, "%c%c", - (entry->flags & TRACE_FLAG_IRQS_OFF) ? 'd' : - (entry->flags & TRACE_FLAG_IRQS_NOSUPPORT) ? 'X' : '.', - ((entry->flags & TRACE_FLAG_NEED_RESCHED) ? 'N' : '.')); - - hardirq = entry->flags & TRACE_FLAG_HARDIRQ; - softirq = entry->flags & TRACE_FLAG_SOFTIRQ; - if (hardirq && softirq) { - trace_seq_putc(s, 'H'); - } else { - if (hardirq) { - trace_seq_putc(s, 'h'); - } else { - if (softirq) - trace_seq_putc(s, 's'); - else - trace_seq_putc(s, '.'); - } - } - - if (entry->preempt_count) - trace_seq_printf(s, "%x", entry->preempt_count); - else - trace_seq_puts(s, "."); -} - -unsigned long preempt_mark_thresh = 100; - -static void -lat_print_timestamp(struct trace_seq *s, u64 abs_usecs, - unsigned long rel_usecs) -{ - trace_seq_printf(s, " %4lldus", abs_usecs); - if (rel_usecs > preempt_mark_thresh) - trace_seq_puts(s, "!: "); - else if (rel_usecs > 1) - trace_seq_puts(s, "+: "); - else - trace_seq_puts(s, " : "); -} - -static const char state_to_char[] = TASK_STATE_TO_CHAR_STR; - -static int task_state_char(unsigned long state) -{ - int bit = state ? __ffs(state) + 1 : 0; - - return bit < sizeof(state_to_char) - 1 ? state_to_char[bit] : '?'; -} - -/* - * The message is supposed to contain an ending newline. - * If the printing stops prematurely, try to add a newline of our own. - */ -void trace_seq_print_cont(struct trace_seq *s, struct trace_iterator *iter) -{ - struct trace_entry *ent; - struct trace_field_cont *cont; - bool ok = true; - - ent = peek_next_entry(iter, iter->cpu, NULL); - if (!ent || ent->type != TRACE_CONT) { - trace_seq_putc(s, '\n'); - return; - } - - do { - cont = (struct trace_field_cont *)ent; - if (ok) - ok = (trace_seq_printf(s, "%s", cont->buf) > 0); - - ftrace_disable_cpu(); - - if (iter->buffer_iter[iter->cpu]) - ring_buffer_read(iter->buffer_iter[iter->cpu], NULL); - else - ring_buffer_consume(iter->tr->buffer, iter->cpu, NULL); - - ftrace_enable_cpu(); - - ent = peek_next_entry(iter, iter->cpu, NULL); - } while (ent && ent->type == TRACE_CONT); - - if (!ok) - trace_seq_putc(s, '\n'); + seq_puts(m, "#\n"); } static void test_cpu_buff_start(struct trace_iterator *iter) @@ -1814,142 +1741,15 @@ static void test_cpu_buff_start(struct trace_iterator *iter) if (cpumask_test_cpu(iter->cpu, iter->started)) return; - cpumask_set_cpu(iter->cpu, iter->started); - trace_seq_printf(s, "##### CPU %u buffer started ####\n", iter->cpu); -} - -static enum print_line_t -print_lat_fmt(struct trace_iterator *iter, unsigned int trace_idx, int cpu) -{ - struct trace_seq *s = &iter->seq; - unsigned long sym_flags = (trace_flags & TRACE_ITER_SYM_MASK); - struct trace_entry *next_entry; - unsigned long verbose = (trace_flags & TRACE_ITER_VERBOSE); - struct trace_entry *entry = iter->ent; - unsigned long abs_usecs; - unsigned long rel_usecs; - u64 next_ts; - char *comm; - int S, T; - int i; - - if (entry->type == TRACE_CONT) - return TRACE_TYPE_HANDLED; - - test_cpu_buff_start(iter); - - next_entry = find_next_entry(iter, NULL, &next_ts); - if (!next_entry) - next_ts = iter->ts; - rel_usecs = ns2usecs(next_ts - iter->ts); - abs_usecs = ns2usecs(iter->ts - iter->tr->time_start); - - if (verbose) { - comm = trace_find_cmdline(entry->pid); - trace_seq_printf(s, "%16s %5d %3d %d %08x %08x [%08lx]" - " %ld.%03ldms (+%ld.%03ldms): ", - comm, - entry->pid, cpu, entry->flags, - entry->preempt_count, trace_idx, - ns2usecs(iter->ts), - abs_usecs/1000, - abs_usecs % 1000, rel_usecs/1000, - rel_usecs % 1000); - } else { - lat_print_generic(s, entry, cpu); - lat_print_timestamp(s, abs_usecs, rel_usecs); - } - switch (entry->type) { - case TRACE_FN: { - struct ftrace_entry *field; - - trace_assign_type(field, entry); - - seq_print_ip_sym(s, field->ip, sym_flags); - trace_seq_puts(s, " ("); - seq_print_ip_sym(s, field->parent_ip, sym_flags); - trace_seq_puts(s, ")\n"); - break; - } - case TRACE_CTX: - case TRACE_WAKE: { - struct ctx_switch_entry *field; - - trace_assign_type(field, entry); - - T = task_state_char(field->next_state); - S = task_state_char(field->prev_state); - comm = trace_find_cmdline(field->next_pid); - trace_seq_printf(s, " %5d:%3d:%c %s [%03d] %5d:%3d:%c %s\n", - field->prev_pid, - field->prev_prio, - S, entry->type == TRACE_CTX ? "==>" : " +", - field->next_cpu, - field->next_pid, - field->next_prio, - T, comm); - break; - } - case TRACE_SPECIAL: { - struct special_entry *field; - - trace_assign_type(field, entry); - - trace_seq_printf(s, "# %ld %ld %ld\n", - field->arg1, - field->arg2, - field->arg3); - break; - } - case TRACE_STACK: { - struct stack_entry *field; - - trace_assign_type(field, entry); - - for (i = 0; i < FTRACE_STACK_ENTRIES; i++) { - if (i) - trace_seq_puts(s, " <= "); - seq_print_ip_sym(s, field->caller[i], sym_flags); - } - trace_seq_puts(s, "\n"); - break; - } - case TRACE_PRINT: { - struct print_entry *field; - - trace_assign_type(field, entry); - - seq_print_ip_sym(s, field->ip, sym_flags); - trace_seq_printf(s, ": %s", field->buf); - if (entry->flags & TRACE_FLAG_CONT) - trace_seq_print_cont(s, iter); - break; - } - case TRACE_BRANCH: { - struct trace_branch *field; - - trace_assign_type(field, entry); - - trace_seq_printf(s, "[%s] %s:%s:%d\n", - field->correct ? " ok " : " MISS ", - field->func, - field->file, - field->line); - break; - } - case TRACE_USER_STACK: { - struct userstack_entry *field; + if (iter->tr->data[iter->cpu]->skipped_entries) + return; - trace_assign_type(field, entry); + cpumask_set_cpu(iter->cpu, iter->started); - seq_print_userip_objs(field, s, sym_flags); - trace_seq_putc(s, '\n'); - break; - } - default: - trace_seq_printf(s, "Unknown type %d\n", entry->type); - } - return TRACE_TYPE_HANDLED; + /* Don't print started cpu buffer for the first entry of the trace */ + if (iter->idx > 1) + trace_seq_printf(s, "##### CPU %u buffer started ####\n", + iter->cpu); } static enum print_line_t print_trace_fmt(struct trace_iterator *iter) @@ -1957,333 +1757,84 @@ static enum print_line_t print_trace_fmt(struct trace_iterator *iter) struct trace_seq *s = &iter->seq; unsigned long sym_flags = (trace_flags & TRACE_ITER_SYM_MASK); struct trace_entry *entry; - unsigned long usec_rem; - unsigned long long t; - unsigned long secs; - char *comm; - int ret; - int S, T; - int i; + struct trace_event *event; entry = iter->ent; - if (entry->type == TRACE_CONT) - return TRACE_TYPE_HANDLED; - test_cpu_buff_start(iter); - comm = trace_find_cmdline(iter->ent->pid); - - t = ns2usecs(iter->ts); - usec_rem = do_div(t, 1000000ULL); - secs = (unsigned long)t; + event = ftrace_find_event(entry->type); - ret = trace_seq_printf(s, "%16s-%-5d ", comm, entry->pid); - if (!ret) - return TRACE_TYPE_PARTIAL_LINE; - ret = trace_seq_printf(s, "[%03d] ", iter->cpu); - if (!ret) - return TRACE_TYPE_PARTIAL_LINE; - ret = trace_seq_printf(s, "%5lu.%06lu: ", secs, usec_rem); - if (!ret) - return TRACE_TYPE_PARTIAL_LINE; - - switch (entry->type) { - case TRACE_FN: { - struct ftrace_entry *field; - - trace_assign_type(field, entry); - - ret = seq_print_ip_sym(s, field->ip, sym_flags); - if (!ret) - return TRACE_TYPE_PARTIAL_LINE; - if ((sym_flags & TRACE_ITER_PRINT_PARENT) && - field->parent_ip) { - ret = trace_seq_printf(s, " <-"); - if (!ret) - return TRACE_TYPE_PARTIAL_LINE; - ret = seq_print_ip_sym(s, - field->parent_ip, - sym_flags); - if (!ret) - return TRACE_TYPE_PARTIAL_LINE; - } - ret = trace_seq_printf(s, "\n"); - if (!ret) - return TRACE_TYPE_PARTIAL_LINE; - break; - } - case TRACE_CTX: - case TRACE_WAKE: { - struct ctx_switch_entry *field; - - trace_assign_type(field, entry); - - T = task_state_char(field->next_state); - S = task_state_char(field->prev_state); - ret = trace_seq_printf(s, " %5d:%3d:%c %s [%03d] %5d:%3d:%c\n", - field->prev_pid, - field->prev_prio, - S, - entry->type == TRACE_CTX ? "==>" : " +", - field->next_cpu, - field->next_pid, - field->next_prio, - T); - if (!ret) - return TRACE_TYPE_PARTIAL_LINE; - break; - } - case TRACE_SPECIAL: { - struct special_entry *field; - - trace_assign_type(field, entry); - - ret = trace_seq_printf(s, "# %ld %ld %ld\n", - field->arg1, - field->arg2, - field->arg3); - if (!ret) - return TRACE_TYPE_PARTIAL_LINE; - break; - } - case TRACE_STACK: { - struct stack_entry *field; - - trace_assign_type(field, entry); - - for (i = 0; i < FTRACE_STACK_ENTRIES; i++) { - if (i) { - ret = trace_seq_puts(s, " <= "); - if (!ret) - return TRACE_TYPE_PARTIAL_LINE; - } - ret = seq_print_ip_sym(s, field->caller[i], - sym_flags); - if (!ret) - return TRACE_TYPE_PARTIAL_LINE; + if (trace_flags & TRACE_ITER_CONTEXT_INFO) { + if (iter->iter_flags & TRACE_FILE_LAT_FMT) { + if (!trace_print_lat_context(iter)) + goto partial; + } else { + if (!trace_print_context(iter)) + goto partial; } - ret = trace_seq_puts(s, "\n"); - if (!ret) - return TRACE_TYPE_PARTIAL_LINE; - break; } - case TRACE_PRINT: { - struct print_entry *field; - trace_assign_type(field, entry); + if (event) + return event->trace(iter, sym_flags); - seq_print_ip_sym(s, field->ip, sym_flags); - trace_seq_printf(s, ": %s", field->buf); - if (entry->flags & TRACE_FLAG_CONT) - trace_seq_print_cont(s, iter); - break; - } - case TRACE_GRAPH_RET: { - return print_graph_function(iter); - } - case TRACE_GRAPH_ENT: { - return print_graph_function(iter); - } - case TRACE_BRANCH: { - struct trace_branch *field; - - trace_assign_type(field, entry); - - trace_seq_printf(s, "[%s] %s:%s:%d\n", - field->correct ? " ok " : " MISS ", - field->func, - field->file, - field->line); - break; - } - case TRACE_USER_STACK: { - struct userstack_entry *field; + if (!trace_seq_printf(s, "Unknown type %d\n", entry->type)) + goto partial; - trace_assign_type(field, entry); - - ret = seq_print_userip_objs(field, s, sym_flags); - if (!ret) - return TRACE_TYPE_PARTIAL_LINE; - ret = trace_seq_putc(s, '\n'); - if (!ret) - return TRACE_TYPE_PARTIAL_LINE; - break; - } - } return TRACE_TYPE_HANDLED; +partial: + return TRACE_TYPE_PARTIAL_LINE; } static enum print_line_t print_raw_fmt(struct trace_iterator *iter) { struct trace_seq *s = &iter->seq; struct trace_entry *entry; - int ret; - int S, T; + struct trace_event *event; entry = iter->ent; - if (entry->type == TRACE_CONT) - return TRACE_TYPE_HANDLED; - - ret = trace_seq_printf(s, "%d %d %llu ", - entry->pid, iter->cpu, iter->ts); - if (!ret) - return TRACE_TYPE_PARTIAL_LINE; - - switch (entry->type) { - case TRACE_FN: { - struct ftrace_entry *field; - - trace_assign_type(field, entry); - - ret = trace_seq_printf(s, "%x %x\n", - field->ip, - field->parent_ip); - if (!ret) - return TRACE_TYPE_PARTIAL_LINE; - break; + if (trace_flags & TRACE_ITER_CONTEXT_INFO) { + if (!trace_seq_printf(s, "%d %d %llu ", + entry->pid, iter->cpu, iter->ts)) + goto partial; } - case TRACE_CTX: - case TRACE_WAKE: { - struct ctx_switch_entry *field; - - trace_assign_type(field, entry); - - T = task_state_char(field->next_state); - S = entry->type == TRACE_WAKE ? '+' : - task_state_char(field->prev_state); - ret = trace_seq_printf(s, "%d %d %c %d %d %d %c\n", - field->prev_pid, - field->prev_prio, - S, - field->next_cpu, - field->next_pid, - field->next_prio, - T); - if (!ret) - return TRACE_TYPE_PARTIAL_LINE; - break; - } - case TRACE_SPECIAL: - case TRACE_USER_STACK: - case TRACE_STACK: { - struct special_entry *field; - - trace_assign_type(field, entry); - ret = trace_seq_printf(s, "# %ld %ld %ld\n", - field->arg1, - field->arg2, - field->arg3); - if (!ret) - return TRACE_TYPE_PARTIAL_LINE; - break; - } - case TRACE_PRINT: { - struct print_entry *field; + event = ftrace_find_event(entry->type); + if (event) + return event->raw(iter, 0); - trace_assign_type(field, entry); + if (!trace_seq_printf(s, "%d ?\n", entry->type)) + goto partial; - trace_seq_printf(s, "# %lx %s", field->ip, field->buf); - if (entry->flags & TRACE_FLAG_CONT) - trace_seq_print_cont(s, iter); - break; - } - } return TRACE_TYPE_HANDLED; +partial: + return TRACE_TYPE_PARTIAL_LINE; } -#define SEQ_PUT_FIELD_RET(s, x) \ -do { \ - if (!trace_seq_putmem(s, &(x), sizeof(x))) \ - return 0; \ -} while (0) - -#define SEQ_PUT_HEX_FIELD_RET(s, x) \ -do { \ - BUILD_BUG_ON(sizeof(x) > MAX_MEMHEX_BYTES); \ - if (!trace_seq_putmem_hex(s, &(x), sizeof(x))) \ - return 0; \ -} while (0) - static enum print_line_t print_hex_fmt(struct trace_iterator *iter) { struct trace_seq *s = &iter->seq; unsigned char newline = '\n'; struct trace_entry *entry; - int S, T; + struct trace_event *event; entry = iter->ent; - if (entry->type == TRACE_CONT) - return TRACE_TYPE_HANDLED; - - SEQ_PUT_HEX_FIELD_RET(s, entry->pid); - SEQ_PUT_HEX_FIELD_RET(s, iter->cpu); - SEQ_PUT_HEX_FIELD_RET(s, iter->ts); - - switch (entry->type) { - case TRACE_FN: { - struct ftrace_entry *field; - - trace_assign_type(field, entry); - - SEQ_PUT_HEX_FIELD_RET(s, field->ip); - SEQ_PUT_HEX_FIELD_RET(s, field->parent_ip); - break; + if (trace_flags & TRACE_ITER_CONTEXT_INFO) { + SEQ_PUT_HEX_FIELD_RET(s, entry->pid); + SEQ_PUT_HEX_FIELD_RET(s, iter->cpu); + SEQ_PUT_HEX_FIELD_RET(s, iter->ts); } - case TRACE_CTX: - case TRACE_WAKE: { - struct ctx_switch_entry *field; - - trace_assign_type(field, entry); - - T = task_state_char(field->next_state); - S = entry->type == TRACE_WAKE ? '+' : - task_state_char(field->prev_state); - SEQ_PUT_HEX_FIELD_RET(s, field->prev_pid); - SEQ_PUT_HEX_FIELD_RET(s, field->prev_prio); - SEQ_PUT_HEX_FIELD_RET(s, S); - SEQ_PUT_HEX_FIELD_RET(s, field->next_cpu); - SEQ_PUT_HEX_FIELD_RET(s, field->next_pid); - SEQ_PUT_HEX_FIELD_RET(s, field->next_prio); - SEQ_PUT_HEX_FIELD_RET(s, T); - break; - } - case TRACE_SPECIAL: - case TRACE_USER_STACK: - case TRACE_STACK: { - struct special_entry *field; - - trace_assign_type(field, entry); - SEQ_PUT_HEX_FIELD_RET(s, field->arg1); - SEQ_PUT_HEX_FIELD_RET(s, field->arg2); - SEQ_PUT_HEX_FIELD_RET(s, field->arg3); - break; - } + event = ftrace_find_event(entry->type); + if (event) { + enum print_line_t ret = event->hex(iter, 0); + if (ret != TRACE_TYPE_HANDLED) + return ret; } - SEQ_PUT_FIELD_RET(s, newline); - - return TRACE_TYPE_HANDLED; -} - -static enum print_line_t print_printk_msg_only(struct trace_iterator *iter) -{ - struct trace_seq *s = &iter->seq; - struct trace_entry *entry = iter->ent; - struct print_entry *field; - int ret; - - trace_assign_type(field, entry); - ret = trace_seq_printf(s, field->buf); - if (!ret) - return TRACE_TYPE_PARTIAL_LINE; - - if (entry->flags & TRACE_FLAG_CONT) - trace_seq_print_cont(s, iter); + SEQ_PUT_FIELD_RET(s, newline); return TRACE_TYPE_HANDLED; } @@ -2292,59 +1843,37 @@ static enum print_line_t print_bin_fmt(struct trace_iterator *iter) { struct trace_seq *s = &iter->seq; struct trace_entry *entry; + struct trace_event *event; entry = iter->ent; - if (entry->type == TRACE_CONT) - return TRACE_TYPE_HANDLED; - - SEQ_PUT_FIELD_RET(s, entry->pid); - SEQ_PUT_FIELD_RET(s, entry->cpu); - SEQ_PUT_FIELD_RET(s, iter->ts); - - switch (entry->type) { - case TRACE_FN: { - struct ftrace_entry *field; - - trace_assign_type(field, entry); - - SEQ_PUT_FIELD_RET(s, field->ip); - SEQ_PUT_FIELD_RET(s, field->parent_ip); - break; - } - case TRACE_CTX: { - struct ctx_switch_entry *field; - - trace_assign_type(field, entry); - - SEQ_PUT_FIELD_RET(s, field->prev_pid); - SEQ_PUT_FIELD_RET(s, field->prev_prio); - SEQ_PUT_FIELD_RET(s, field->prev_state); - SEQ_PUT_FIELD_RET(s, field->next_pid); - SEQ_PUT_FIELD_RET(s, field->next_prio); - SEQ_PUT_FIELD_RET(s, field->next_state); - break; + if (trace_flags & TRACE_ITER_CONTEXT_INFO) { + SEQ_PUT_FIELD_RET(s, entry->pid); + SEQ_PUT_FIELD_RET(s, iter->cpu); + SEQ_PUT_FIELD_RET(s, iter->ts); } - case TRACE_SPECIAL: - case TRACE_USER_STACK: - case TRACE_STACK: { - struct special_entry *field; - - trace_assign_type(field, entry); - SEQ_PUT_FIELD_RET(s, field->arg1); - SEQ_PUT_FIELD_RET(s, field->arg2); - SEQ_PUT_FIELD_RET(s, field->arg3); - break; - } - } - return 1; + event = ftrace_find_event(entry->type); + return event ? event->binary(iter, 0) : TRACE_TYPE_HANDLED; } static int trace_empty(struct trace_iterator *iter) { int cpu; + /* If we are looking at one CPU buffer, only check that one */ + if (iter->cpu_file != TRACE_PIPE_ALL_CPU) { + cpu = iter->cpu_file; + if (iter->buffer_iter[cpu]) { + if (!ring_buffer_iter_empty(iter->buffer_iter[cpu])) + return 0; + } else { + if (!ring_buffer_empty_cpu(iter->tr->buffer, cpu)) + return 0; + } + return 1; + } + for_each_tracing_cpu(cpu) { if (iter->buffer_iter[cpu]) { if (!ring_buffer_iter_empty(iter->buffer_iter[cpu])) @@ -2358,6 +1887,7 @@ static int trace_empty(struct trace_iterator *iter) return 1; } +/* Called with trace_event_read_lock() held. */ static enum print_line_t print_trace_line(struct trace_iterator *iter) { enum print_line_t ret; @@ -2368,10 +1898,15 @@ static enum print_line_t print_trace_line(struct trace_iterator *iter) return ret; } + if (iter->ent->type == TRACE_BPRINT && + trace_flags & TRACE_ITER_PRINTK && + trace_flags & TRACE_ITER_PRINTK_MSGONLY) + return trace_print_bprintk_msg_only(iter); + if (iter->ent->type == TRACE_PRINT && trace_flags & TRACE_ITER_PRINTK && trace_flags & TRACE_ITER_PRINTK_MSGONLY) - return print_printk_msg_only(iter); + return trace_print_printk_msg_only(iter); if (trace_flags & TRACE_ITER_BIN) return print_bin_fmt(iter); @@ -2382,9 +1917,6 @@ static enum print_line_t print_trace_line(struct trace_iterator *iter) if (trace_flags & TRACE_ITER_RAW) return print_raw_fmt(iter); - if (iter->iter_flags & TRACE_FILE_LAT_FMT) - return print_lat_fmt(iter, iter->idx, iter->cpu); - return print_trace_fmt(iter); } @@ -2418,7 +1950,7 @@ static int s_show(struct seq_file *m, void *v) return 0; } -static struct seq_operations tracer_seq_ops = { +static const struct seq_operations tracer_seq_ops = { .start = s_start, .next = s_next, .stop = s_stop, @@ -2426,30 +1958,43 @@ static struct seq_operations tracer_seq_ops = { }; static struct trace_iterator * -__tracing_open(struct inode *inode, struct file *file, int *ret) +__tracing_open(struct inode *inode, struct file *file) { + long cpu_file = (long) inode->i_private; + void *fail_ret = ERR_PTR(-ENOMEM); struct trace_iterator *iter; struct seq_file *m; - int cpu; + int cpu, ret; - if (tracing_disabled) { - *ret = -ENODEV; - return NULL; - } + if (tracing_disabled) + return ERR_PTR(-ENODEV); iter = kzalloc(sizeof(*iter), GFP_KERNEL); - if (!iter) { - *ret = -ENOMEM; - goto out; - } + if (!iter) + return ERR_PTR(-ENOMEM); + /* + * We make a copy of the current tracer to avoid concurrent + * changes on it while we are reading. + */ mutex_lock(&trace_types_lock); + iter->trace = kzalloc(sizeof(*iter->trace), GFP_KERNEL); + if (!iter->trace) + goto fail; + + if (current_trace) + *iter->trace = *current_trace; + + if (!zalloc_cpumask_var(&iter->started, GFP_KERNEL)) + goto fail; + if (current_trace && current_trace->print_max) iter->tr = &max_tr; else - iter->tr = inode->i_private; - iter->trace = current_trace; + iter->tr = &global_trace; iter->pos = -1; + mutex_init(&iter->mutex); + iter->cpu_file = cpu_file; /* Notify the tracer early; before we stop tracing. */ if (iter->trace && iter->trace->open) @@ -2459,30 +2004,34 @@ __tracing_open(struct inode *inode, struct file *file, int *ret) if (ring_buffer_overruns(iter->tr->buffer)) iter->iter_flags |= TRACE_FILE_ANNOTATE; + /* stop the trace while dumping */ + tracing_stop(); - for_each_tracing_cpu(cpu) { + if (iter->cpu_file == TRACE_PIPE_ALL_CPU) { + for_each_tracing_cpu(cpu) { + iter->buffer_iter[cpu] = + ring_buffer_read_start(iter->tr->buffer, cpu); + tracing_iter_reset(iter, cpu); + } + } else { + cpu = iter->cpu_file; iter->buffer_iter[cpu] = - ring_buffer_read_start(iter->tr->buffer, cpu); - - if (!iter->buffer_iter[cpu]) - goto fail_buffer; + ring_buffer_read_start(iter->tr->buffer, cpu); + tracing_iter_reset(iter, cpu); } - /* TODO stop tracer */ - *ret = seq_open(file, &tracer_seq_ops); - if (*ret) + ret = seq_open(file, &tracer_seq_ops); + if (ret < 0) { + fail_ret = ERR_PTR(ret); goto fail_buffer; + } m = file->private_data; m->private = iter; - /* stop the trace while dumping */ - tracing_stop(); - mutex_unlock(&trace_types_lock); - out: return iter; fail_buffer: @@ -2490,10 +2039,14 @@ __tracing_open(struct inode *inode, struct file *file, int *ret) if (iter->buffer_iter[cpu]) ring_buffer_read_finish(iter->buffer_iter[cpu]); } + free_cpumask_var(iter->started); + tracing_start(); + fail: mutex_unlock(&trace_types_lock); + kfree(iter->trace); kfree(iter); - return ERR_PTR(-ENOMEM); + return fail_ret; } int tracing_open_generic(struct inode *inode, struct file *filp) @@ -2505,12 +2058,17 @@ int tracing_open_generic(struct inode *inode, struct file *filp) return 0; } -int tracing_release(struct inode *inode, struct file *file) +static int tracing_release(struct inode *inode, struct file *file) { struct seq_file *m = (struct seq_file *)file->private_data; - struct trace_iterator *iter = m->private; + struct trace_iterator *iter; int cpu; + if (!(file->f_mode & FMODE_READ)) + return 0; + + iter = m->private; + mutex_lock(&trace_types_lock); for_each_tracing_cpu(cpu) { if (iter->buffer_iter[cpu]) @@ -2525,55 +2083,59 @@ int tracing_release(struct inode *inode, struct file *file) mutex_unlock(&trace_types_lock); seq_release(inode, file); + mutex_destroy(&iter->mutex); + free_cpumask_var(iter->started); + kfree(iter->trace); kfree(iter); return 0; } static int tracing_open(struct inode *inode, struct file *file) { - int ret; - - __tracing_open(inode, file, &ret); - - return ret; -} - -static int tracing_lt_open(struct inode *inode, struct file *file) -{ struct trace_iterator *iter; - int ret; + int ret = 0; - iter = __tracing_open(inode, file, &ret); + /* If this file was open for write, then erase contents */ + if ((file->f_mode & FMODE_WRITE) && + (file->f_flags & O_TRUNC)) { + long cpu = (long) inode->i_private; - if (!ret) - iter->iter_flags |= TRACE_FILE_LAT_FMT; + if (cpu == TRACE_PIPE_ALL_CPU) + tracing_reset_online_cpus(&global_trace); + else + tracing_reset(&global_trace, cpu); + } + if (file->f_mode & FMODE_READ) { + iter = __tracing_open(inode, file); + if (IS_ERR(iter)) + ret = PTR_ERR(iter); + else if (trace_flags & TRACE_ITER_LATENCY_FMT) + iter->iter_flags |= TRACE_FILE_LAT_FMT; + } return ret; } - static void * t_next(struct seq_file *m, void *v, loff_t *pos) { - struct tracer *t = m->private; + struct tracer *t = v; (*pos)++; if (t) t = t->next; - m->private = t; - return t; } static void *t_start(struct seq_file *m, loff_t *pos) { - struct tracer *t = m->private; + struct tracer *t; loff_t l = 0; mutex_lock(&trace_types_lock); - for (; t && l < *pos; t = t_next(m, t, &l)) + for (t = trace_types; t && l < *pos; t = t_next(m, t, &l)) ; return t; @@ -2600,7 +2162,7 @@ static int t_show(struct seq_file *m, void *v) return 0; } -static struct seq_operations show_traces_seq_ops = { +static const struct seq_operations show_traces_seq_ops = { .start = t_start, .next = t_next, .stop = t_stop, @@ -2609,35 +2171,28 @@ static struct seq_operations show_traces_seq_ops = { static int show_traces_open(struct inode *inode, struct file *file) { - int ret; - if (tracing_disabled) return -ENODEV; - ret = seq_open(file, &show_traces_seq_ops); - if (!ret) { - struct seq_file *m = file->private_data; - m->private = trace_types; - } + return seq_open(file, &show_traces_seq_ops); +} - return ret; +static ssize_t +tracing_write_stub(struct file *filp, const char __user *ubuf, + size_t count, loff_t *ppos) +{ + return count; } -static struct file_operations tracing_fops = { +static const struct file_operations tracing_fops = { .open = tracing_open, .read = seq_read, + .write = tracing_write_stub, .llseek = seq_lseek, .release = tracing_release, }; -static struct file_operations tracing_lt_fops = { - .open = tracing_lt_open, - .read = seq_read, - .llseek = seq_lseek, - .release = tracing_release, -}; - -static struct file_operations show_traces_fops = { +static const struct file_operations show_traces_fops = { .open = show_traces_open, .read = seq_read, .release = seq_release, @@ -2692,11 +2247,12 @@ tracing_cpumask_write(struct file *filp, const char __user *ubuf, if (!alloc_cpumask_var(&tracing_cpumask_new, GFP_KERNEL)) return -ENOMEM; - mutex_lock(&tracing_cpumask_update_lock); err = cpumask_parse_user(ubuf, count, tracing_cpumask_new); if (err) goto err_unlock; + mutex_lock(&tracing_cpumask_update_lock); + local_irq_disable(); __raw_spin_lock(&ftrace_max_lock); for_each_tracing_cpu(cpu) { @@ -2724,13 +2280,12 @@ tracing_cpumask_write(struct file *filp, const char __user *ubuf, return count; err_unlock: - mutex_unlock(&tracing_cpumask_update_lock); - free_cpumask_var(tracing_cpumask); + free_cpumask_var(tracing_cpumask_new); return err; } -static struct file_operations tracing_cpumask_fops = { +static const struct file_operations tracing_cpumask_fops = { .open = tracing_open_generic, .read = tracing_cpumask_read, .write = tracing_cpumask_write, @@ -2740,80 +2295,85 @@ static ssize_t tracing_trace_options_read(struct file *filp, char __user *ubuf, size_t cnt, loff_t *ppos) { - int i; + struct tracer_opt *trace_opts; + u32 tracer_flags; + int len = 0; char *buf; int r = 0; - int len = 0; - u32 tracer_flags = current_trace->flags->val; - struct tracer_opt *trace_opts = current_trace->flags->opts; + int i; - /* calulate max size */ + /* calculate max size */ for (i = 0; trace_options[i]; i++) { len += strlen(trace_options[i]); - len += 3; /* "no" and space */ + len += 3; /* "no" and newline */ } + mutex_lock(&trace_types_lock); + tracer_flags = current_trace->flags->val; + trace_opts = current_trace->flags->opts; + /* * Increase the size with names of options specific * of the current tracer. */ for (i = 0; trace_opts[i].name; i++) { len += strlen(trace_opts[i].name); - len += 3; /* "no" and space */ + len += 3; /* "no" and newline */ } - /* +2 for \n and \0 */ - buf = kmalloc(len + 2, GFP_KERNEL); - if (!buf) + /* +1 for \0 */ + buf = kmalloc(len + 1, GFP_KERNEL); + if (!buf) { + mutex_unlock(&trace_types_lock); return -ENOMEM; + } for (i = 0; trace_options[i]; i++) { if (trace_flags & (1 << i)) - r += sprintf(buf + r, "%s ", trace_options[i]); + r += sprintf(buf + r, "%s\n", trace_options[i]); else - r += sprintf(buf + r, "no%s ", trace_options[i]); + r += sprintf(buf + r, "no%s\n", trace_options[i]); } for (i = 0; trace_opts[i].name; i++) { if (tracer_flags & trace_opts[i].bit) - r += sprintf(buf + r, "%s ", + r += sprintf(buf + r, "%s\n", trace_opts[i].name); else - r += sprintf(buf + r, "no%s ", + r += sprintf(buf + r, "no%s\n", trace_opts[i].name); } + mutex_unlock(&trace_types_lock); - r += sprintf(buf + r, "\n"); - WARN_ON(r >= len + 2); + WARN_ON(r >= len + 1); r = simple_read_from_buffer(ubuf, cnt, ppos, buf, r); kfree(buf); - return r; } /* Try to assign a tracer specific option */ static int set_tracer_option(struct tracer *trace, char *cmp, int neg) { - struct tracer_flags *trace_flags = trace->flags; + struct tracer_flags *tracer_flags = trace->flags; struct tracer_opt *opts = NULL; int ret = 0, i = 0; int len; - for (i = 0; trace_flags->opts[i].name; i++) { - opts = &trace_flags->opts[i]; + for (i = 0; tracer_flags->opts[i].name; i++) { + opts = &tracer_flags->opts[i]; len = strlen(opts->name); if (strncmp(cmp, opts->name, len) == 0) { - ret = trace->set_flag(trace_flags->val, + ret = trace->set_flag(tracer_flags->val, opts->bit, !neg); break; } } /* Not found */ - if (!trace_flags->opts[i].name) + if (!tracer_flags->opts[i].name) return -EINVAL; /* Refused to handle */ @@ -2821,13 +2381,25 @@ static int set_tracer_option(struct tracer *trace, char *cmp, int neg) return ret; if (neg) - trace_flags->val &= ~opts->bit; + tracer_flags->val &= ~opts->bit; else - trace_flags->val |= opts->bit; + tracer_flags->val |= opts->bit; return 0; } +static void set_tracer_flags(unsigned int mask, int enabled) +{ + /* do nothing if flag is already set */ + if (!!(trace_flags & mask) == !!enabled) + return; + + if (enabled) + trace_flags |= mask; + else + trace_flags &= ~mask; +} + static ssize_t tracing_trace_options_write(struct file *filp, const char __user *ubuf, size_t cnt, loff_t *ppos) @@ -2855,27 +2427,26 @@ tracing_trace_options_write(struct file *filp, const char __user *ubuf, int len = strlen(trace_options[i]); if (strncmp(cmp, trace_options[i], len) == 0) { - if (neg) - trace_flags &= ~(1 << i); - else - trace_flags |= (1 << i); + set_tracer_flags(1 << i, !neg); break; } } /* If no option could be set, test the specific tracer options */ if (!trace_options[i]) { + mutex_lock(&trace_types_lock); ret = set_tracer_option(current_trace, cmp, neg); + mutex_unlock(&trace_types_lock); if (ret) return ret; } - filp->f_pos += cnt; + *ppos += cnt; return cnt; } -static struct file_operations tracing_iter_fops = { +static const struct file_operations tracing_iter_fops = { .open = tracing_open_generic, .read = tracing_trace_options_read, .write = tracing_trace_options_write, @@ -2883,21 +2454,20 @@ static struct file_operations tracing_iter_fops = { static const char readme_msg[] = "tracing mini-HOWTO:\n\n" - "# mkdir /debug\n" - "# mount -t debugfs nodev /debug\n\n" - "# cat /debug/tracing/available_tracers\n" - "wakeup preemptirqsoff preemptoff irqsoff ftrace sched_switch none\n\n" - "# cat /debug/tracing/current_tracer\n" - "none\n" - "# echo sched_switch > /debug/tracing/current_tracer\n" - "# cat /debug/tracing/current_tracer\n" + "# mount -t debugfs nodev /sys/kernel/debug\n\n" + "# cat /sys/kernel/debug/tracing/available_tracers\n" + "wakeup preemptirqsoff preemptoff irqsoff function sched_switch nop\n\n" + "# cat /sys/kernel/debug/tracing/current_tracer\n" + "nop\n" + "# echo sched_switch > /sys/kernel/debug/tracing/current_tracer\n" + "# cat /sys/kernel/debug/tracing/current_tracer\n" "sched_switch\n" - "# cat /debug/tracing/trace_options\n" + "# cat /sys/kernel/debug/tracing/trace_options\n" "noprint-parent nosym-offset nosym-addr noverbose\n" - "# echo print-parent > /debug/tracing/trace_options\n" - "# echo 1 > /debug/tracing/tracing_enabled\n" - "# cat /debug/tracing/trace > /tmp/trace.txt\n" - "echo 0 > /debug/tracing/tracing_enabled\n" + "# echo print-parent > /sys/kernel/debug/tracing/trace_options\n" + "# echo 1 > /sys/kernel/debug/tracing/tracing_enabled\n" + "# cat /sys/kernel/debug/tracing/trace > /tmp/trace.txt\n" + "# echo 0 > /sys/kernel/debug/tracing/tracing_enabled\n" ; static ssize_t @@ -2908,12 +2478,62 @@ tracing_readme_read(struct file *filp, char __user *ubuf, readme_msg, strlen(readme_msg)); } -static struct file_operations tracing_readme_fops = { +static const struct file_operations tracing_readme_fops = { .open = tracing_open_generic, .read = tracing_readme_read, }; static ssize_t +tracing_saved_cmdlines_read(struct file *file, char __user *ubuf, + size_t cnt, loff_t *ppos) +{ + char *buf_comm; + char *file_buf; + char *buf; + int len = 0; + int pid; + int i; + + file_buf = kmalloc(SAVED_CMDLINES*(16+TASK_COMM_LEN), GFP_KERNEL); + if (!file_buf) + return -ENOMEM; + + buf_comm = kmalloc(TASK_COMM_LEN, GFP_KERNEL); + if (!buf_comm) { + kfree(file_buf); + return -ENOMEM; + } + + buf = file_buf; + + for (i = 0; i < SAVED_CMDLINES; i++) { + int r; + + pid = map_cmdline_to_pid[i]; + if (pid == -1 || pid == NO_CMDLINE_MAP) + continue; + + trace_find_cmdline(pid, buf_comm); + r = sprintf(buf, "%d %s\n", pid, buf_comm); + buf += r; + len += r; + } + + len = simple_read_from_buffer(ubuf, cnt, ppos, + file_buf, len); + + kfree(file_buf); + kfree(buf_comm); + + return len; +} + +static const struct file_operations tracing_saved_cmdlines_fops = { + .open = tracing_open_generic, + .read = tracing_saved_cmdlines_read, +}; + +static ssize_t tracing_ctrl_read(struct file *filp, char __user *ubuf, size_t cnt, loff_t *ppos) { @@ -2930,7 +2550,7 @@ tracing_ctrl_write(struct file *filp, const char __user *ubuf, { struct trace_array *tr = filp->private_data; char buf[64]; - long val; + unsigned long val; int ret; if (cnt >= sizeof(buf)) @@ -2963,7 +2583,7 @@ tracing_ctrl_write(struct file *filp, const char __user *ubuf, } mutex_unlock(&trace_types_lock); - filp->f_pos += cnt; + *ppos += cnt; return cnt; } @@ -2972,7 +2592,7 @@ static ssize_t tracing_set_trace_read(struct file *filp, char __user *ubuf, size_t cnt, loff_t *ppos) { - char buf[max_tracer_type_len+2]; + char buf[MAX_TRACER_SIZE+2]; int r; mutex_lock(&trace_types_lock); @@ -2985,13 +2605,105 @@ tracing_set_trace_read(struct file *filp, char __user *ubuf, return simple_read_from_buffer(ubuf, cnt, ppos, buf, r); } -static int tracing_set_tracer(char *buf) +int tracer_init(struct tracer *t, struct trace_array *tr) +{ + tracing_reset_online_cpus(tr); + return t->init(tr); +} + +static int tracing_resize_ring_buffer(unsigned long size) +{ + int ret; + + /* + * If kernel or user changes the size of the ring buffer + * we use the size that was given, and we can forget about + * expanding it later. + */ + ring_buffer_expanded = 1; + + ret = ring_buffer_resize(global_trace.buffer, size); + if (ret < 0) + return ret; + + ret = ring_buffer_resize(max_tr.buffer, size); + if (ret < 0) { + int r; + + r = ring_buffer_resize(global_trace.buffer, + global_trace.entries); + if (r < 0) { + /* + * AARGH! We are left with different + * size max buffer!!!! + * The max buffer is our "snapshot" buffer. + * When a tracer needs a snapshot (one of the + * latency tracers), it swaps the max buffer + * with the saved snap shot. We succeeded to + * update the size of the main buffer, but failed to + * update the size of the max buffer. But when we tried + * to reset the main buffer to the original size, we + * failed there too. This is very unlikely to + * happen, but if it does, warn and kill all + * tracing. + */ + WARN_ON(1); + tracing_disabled = 1; + } + return ret; + } + + global_trace.entries = size; + + return ret; +} + +/** + * tracing_update_buffers - used by tracing facility to expand ring buffers + * + * To save on memory when the tracing is never used on a system with it + * configured in. The ring buffers are set to a minimum size. But once + * a user starts to use the tracing facility, then they need to grow + * to their default size. + * + * This function is to be called when a tracer is about to be used. + */ +int tracing_update_buffers(void) +{ + int ret = 0; + + mutex_lock(&trace_types_lock); + if (!ring_buffer_expanded) + ret = tracing_resize_ring_buffer(trace_buf_size); + mutex_unlock(&trace_types_lock); + + return ret; +} + +struct trace_option_dentry; + +static struct trace_option_dentry * +create_trace_option_files(struct tracer *tracer); + +static void +destroy_trace_option_files(struct trace_option_dentry *topts); + +static int tracing_set_tracer(const char *buf) { + static struct trace_option_dentry *topts; struct trace_array *tr = &global_trace; struct tracer *t; int ret = 0; mutex_lock(&trace_types_lock); + + if (!ring_buffer_expanded) { + ret = tracing_resize_ring_buffer(trace_buf_size); + if (ret < 0) + goto out; + ret = 0; + } + for (t = trace_types; t; t = t->next) { if (strcmp(t->name, buf) == 0) break; @@ -3007,9 +2719,14 @@ static int tracing_set_tracer(char *buf) if (current_trace && current_trace->reset) current_trace->reset(tr); + destroy_trace_option_files(topts); + current_trace = t; + + topts = create_trace_option_files(current_trace); + if (t->init) { - ret = t->init(tr); + ret = tracer_init(t, tr); if (ret) goto out; } @@ -3025,15 +2742,15 @@ static ssize_t tracing_set_trace_write(struct file *filp, const char __user *ubuf, size_t cnt, loff_t *ppos) { - char buf[max_tracer_type_len+1]; + char buf[MAX_TRACER_SIZE+1]; int i; size_t ret; int err; ret = cnt; - if (cnt > max_tracer_type_len) - cnt = max_tracer_type_len; + if (cnt > MAX_TRACER_SIZE) + cnt = MAX_TRACER_SIZE; if (copy_from_user(&buf, ubuf, cnt)) return -EFAULT; @@ -3048,7 +2765,7 @@ tracing_set_trace_write(struct file *filp, const char __user *ubuf, if (err) return err; - filp->f_pos += ret; + *ppos += ret; return ret; } @@ -3072,9 +2789,9 @@ static ssize_t tracing_max_lat_write(struct file *filp, const char __user *ubuf, size_t cnt, loff_t *ppos) { - long *ptr = filp->private_data; + unsigned long *ptr = filp->private_data; char buf[64]; - long val; + unsigned long val; int ret; if (cnt >= sizeof(buf)) @@ -3094,54 +2811,99 @@ tracing_max_lat_write(struct file *filp, const char __user *ubuf, return cnt; } -static atomic_t tracing_reader; - static int tracing_open_pipe(struct inode *inode, struct file *filp) { + long cpu_file = (long) inode->i_private; struct trace_iterator *iter; + int ret = 0; if (tracing_disabled) return -ENODEV; - /* We only allow for reader of the pipe */ - if (atomic_inc_return(&tracing_reader) != 1) { - atomic_dec(&tracing_reader); - return -EBUSY; + mutex_lock(&trace_types_lock); + + /* We only allow one reader per cpu */ + if (cpu_file == TRACE_PIPE_ALL_CPU) { + if (!cpumask_empty(tracing_reader_cpumask)) { + ret = -EBUSY; + goto out; + } + cpumask_setall(tracing_reader_cpumask); + } else { + if (!cpumask_test_cpu(cpu_file, tracing_reader_cpumask)) + cpumask_set_cpu(cpu_file, tracing_reader_cpumask); + else { + ret = -EBUSY; + goto out; + } } /* create a buffer to store the information to pass to userspace */ iter = kzalloc(sizeof(*iter), GFP_KERNEL); - if (!iter) - return -ENOMEM; + if (!iter) { + ret = -ENOMEM; + goto out; + } - if (!alloc_cpumask_var(&iter->started, GFP_KERNEL)) { - kfree(iter); - return -ENOMEM; + /* + * We make a copy of the current tracer to avoid concurrent + * changes on it while we are reading. + */ + iter->trace = kmalloc(sizeof(*iter->trace), GFP_KERNEL); + if (!iter->trace) { + ret = -ENOMEM; + goto fail; } + if (current_trace) + *iter->trace = *current_trace; - mutex_lock(&trace_types_lock); + if (!alloc_cpumask_var(&iter->started, GFP_KERNEL)) { + ret = -ENOMEM; + goto fail; + } /* trace pipe does not show start of buffer */ cpumask_setall(iter->started); + if (trace_flags & TRACE_ITER_LATENCY_FMT) + iter->iter_flags |= TRACE_FILE_LAT_FMT; + + iter->cpu_file = cpu_file; iter->tr = &global_trace; - iter->trace = current_trace; + mutex_init(&iter->mutex); filp->private_data = iter; if (iter->trace->pipe_open) iter->trace->pipe_open(iter); + +out: mutex_unlock(&trace_types_lock); + return ret; - return 0; +fail: + kfree(iter->trace); + kfree(iter); + mutex_unlock(&trace_types_lock); + return ret; } static int tracing_release_pipe(struct inode *inode, struct file *file) { struct trace_iterator *iter = file->private_data; + mutex_lock(&trace_types_lock); + + if (iter->cpu_file == TRACE_PIPE_ALL_CPU) + cpumask_clear(tracing_reader_cpumask); + else + cpumask_clear_cpu(iter->cpu_file, tracing_reader_cpumask); + + mutex_unlock(&trace_types_lock); + free_cpumask_var(iter->started); + mutex_destroy(&iter->mutex); + kfree(iter->trace); kfree(iter); - atomic_dec(&tracing_reader); return 0; } @@ -3167,67 +2929,57 @@ tracing_poll_pipe(struct file *filp, poll_table *poll_table) } } -/* - * Consumer reader. - */ -static ssize_t -tracing_read_pipe(struct file *filp, char __user *ubuf, - size_t cnt, loff_t *ppos) + +void default_wait_pipe(struct trace_iterator *iter) { - struct trace_iterator *iter = filp->private_data; - ssize_t sret; + DEFINE_WAIT(wait); - /* return any leftover data */ - sret = trace_seq_to_user(&iter->seq, ubuf, cnt); - if (sret != -EBUSY) - return sret; + prepare_to_wait(&trace_wait, &wait, TASK_INTERRUPTIBLE); - trace_seq_reset(&iter->seq); + if (trace_empty(iter)) + schedule(); - mutex_lock(&trace_types_lock); - if (iter->trace->read) { - sret = iter->trace->read(iter, filp, ubuf, cnt, ppos); - if (sret) - goto out; - } + finish_wait(&trace_wait, &wait); +} + +/* + * This is a make-shift waitqueue. + * A tracer might use this callback on some rare cases: + * + * 1) the current tracer might hold the runqueue lock when it wakes up + * a reader, hence a deadlock (sched, function, and function graph tracers) + * 2) the function tracers, trace all functions, we don't want + * the overhead of calling wake_up and friends + * (and tracing them too) + * + * Anyway, this is really very primitive wakeup. + */ +void poll_wait_pipe(struct trace_iterator *iter) +{ + set_current_state(TASK_INTERRUPTIBLE); + /* sleep for 100 msecs, and try again. */ + schedule_timeout(HZ / 10); +} + +/* Must be called with trace_types_lock mutex held. */ +static int tracing_wait_pipe(struct file *filp) +{ + struct trace_iterator *iter = filp->private_data; -waitagain: - sret = 0; while (trace_empty(iter)) { if ((filp->f_flags & O_NONBLOCK)) { - sret = -EAGAIN; - goto out; + return -EAGAIN; } - /* - * This is a make-shift waitqueue. The reason we don't use - * an actual wait queue is because: - * 1) we only ever have one waiter - * 2) the tracing, traces all functions, we don't want - * the overhead of calling wake_up and friends - * (and tracing them too) - * Anyway, this is really very primitive wakeup. - */ - set_current_state(TASK_INTERRUPTIBLE); - iter->tr->waiter = current; - - mutex_unlock(&trace_types_lock); - - /* sleep for 100 msecs, and try again. */ - schedule_timeout(HZ/10); - - mutex_lock(&trace_types_lock); + mutex_unlock(&iter->mutex); - iter->tr->waiter = NULL; + iter->trace->wait_pipe(iter); - if (signal_pending(current)) { - sret = -EINTR; - goto out; - } + mutex_lock(&iter->mutex); - if (iter->trace != current_trace) - goto out; + if (signal_pending(current)) + return -EINTR; /* * We block until we read something and tracing is disabled. @@ -3240,13 +2992,59 @@ waitagain: */ if (!tracer_enabled && iter->pos) break; + } + + return 1; +} + +/* + * Consumer reader. + */ +static ssize_t +tracing_read_pipe(struct file *filp, char __user *ubuf, + size_t cnt, loff_t *ppos) +{ + struct trace_iterator *iter = filp->private_data; + static struct tracer *old_tracer; + ssize_t sret; + + /* return any leftover data */ + sret = trace_seq_to_user(&iter->seq, ubuf, cnt); + if (sret != -EBUSY) + return sret; + + trace_seq_init(&iter->seq); - continue; + /* copy the tracer to avoid using a global lock all around */ + mutex_lock(&trace_types_lock); + if (unlikely(old_tracer != current_trace && current_trace)) { + old_tracer = current_trace; + *iter->trace = *current_trace; } + mutex_unlock(&trace_types_lock); + + /* + * Avoid more than one consumer on a single file descriptor + * This is just a matter of traces coherency, the ring buffer itself + * is protected. + */ + mutex_lock(&iter->mutex); + if (iter->trace->read) { + sret = iter->trace->read(iter, filp, ubuf, cnt, ppos); + if (sret) + goto out; + } + +waitagain: + sret = tracing_wait_pipe(filp); + if (sret <= 0) + goto out; /* stop when tracing is finished */ - if (trace_empty(iter)) + if (trace_empty(iter)) { + sret = 0; goto out; + } if (cnt >= PAGE_SIZE) cnt = PAGE_SIZE - 1; @@ -3257,6 +3055,7 @@ waitagain: offsetof(struct trace_iterator, seq)); iter->pos = -1; + trace_event_read_lock(); while (find_next_entry_inc(iter) != NULL) { enum print_line_t ret; int len = iter->seq.len; @@ -3267,17 +3066,18 @@ waitagain: iter->seq.len = len; break; } - - trace_consume(iter); + if (ret != TRACE_TYPE_NO_CONSUME) + trace_consume(iter); if (iter->seq.len >= cnt) break; } + trace_event_read_unlock(); /* Now copy what we have to the user */ sret = trace_seq_to_user(&iter->seq, ubuf, cnt); if (iter->seq.readpos >= iter->seq.len) - trace_seq_reset(&iter->seq); + trace_seq_init(&iter->seq); /* * If there was nothing to send to user, inspite of consuming trace @@ -3287,20 +3087,169 @@ waitagain: goto waitagain; out: - mutex_unlock(&trace_types_lock); + mutex_unlock(&iter->mutex); return sret; } +static void tracing_pipe_buf_release(struct pipe_inode_info *pipe, + struct pipe_buffer *buf) +{ + __free_page(buf->page); +} + +static void tracing_spd_release_pipe(struct splice_pipe_desc *spd, + unsigned int idx) +{ + __free_page(spd->pages[idx]); +} + +static struct pipe_buf_operations tracing_pipe_buf_ops = { + .can_merge = 0, + .map = generic_pipe_buf_map, + .unmap = generic_pipe_buf_unmap, + .confirm = generic_pipe_buf_confirm, + .release = tracing_pipe_buf_release, + .steal = generic_pipe_buf_steal, + .get = generic_pipe_buf_get, +}; + +static size_t +tracing_fill_pipe_page(size_t rem, struct trace_iterator *iter) +{ + size_t count; + int ret; + + /* Seq buffer is page-sized, exactly what we need. */ + for (;;) { + count = iter->seq.len; + ret = print_trace_line(iter); + count = iter->seq.len - count; + if (rem < count) { + rem = 0; + iter->seq.len -= count; + break; + } + if (ret == TRACE_TYPE_PARTIAL_LINE) { + iter->seq.len -= count; + break; + } + + if (ret != TRACE_TYPE_NO_CONSUME) + trace_consume(iter); + rem -= count; + if (!find_next_entry_inc(iter)) { + rem = 0; + iter->ent = NULL; + break; + } + } + + return rem; +} + +static ssize_t tracing_splice_read_pipe(struct file *filp, + loff_t *ppos, + struct pipe_inode_info *pipe, + size_t len, + unsigned int flags) +{ + struct page *pages[PIPE_BUFFERS]; + struct partial_page partial[PIPE_BUFFERS]; + struct trace_iterator *iter = filp->private_data; + struct splice_pipe_desc spd = { + .pages = pages, + .partial = partial, + .nr_pages = 0, /* This gets updated below. */ + .flags = flags, + .ops = &tracing_pipe_buf_ops, + .spd_release = tracing_spd_release_pipe, + }; + static struct tracer *old_tracer; + ssize_t ret; + size_t rem; + unsigned int i; + + /* copy the tracer to avoid using a global lock all around */ + mutex_lock(&trace_types_lock); + if (unlikely(old_tracer != current_trace && current_trace)) { + old_tracer = current_trace; + *iter->trace = *current_trace; + } + mutex_unlock(&trace_types_lock); + + mutex_lock(&iter->mutex); + + if (iter->trace->splice_read) { + ret = iter->trace->splice_read(iter, filp, + ppos, pipe, len, flags); + if (ret) + goto out_err; + } + + ret = tracing_wait_pipe(filp); + if (ret <= 0) + goto out_err; + + if (!iter->ent && !find_next_entry_inc(iter)) { + ret = -EFAULT; + goto out_err; + } + + trace_event_read_lock(); + + /* Fill as many pages as possible. */ + for (i = 0, rem = len; i < PIPE_BUFFERS && rem; i++) { + pages[i] = alloc_page(GFP_KERNEL); + if (!pages[i]) + break; + + rem = tracing_fill_pipe_page(rem, iter); + + /* Copy the data into the page, so we can start over. */ + ret = trace_seq_to_buffer(&iter->seq, + page_address(pages[i]), + iter->seq.len); + if (ret < 0) { + __free_page(pages[i]); + break; + } + partial[i].offset = 0; + partial[i].len = iter->seq.len; + + trace_seq_init(&iter->seq); + } + + trace_event_read_unlock(); + mutex_unlock(&iter->mutex); + + spd.nr_pages = i; + + return splice_to_pipe(pipe, &spd); + +out_err: + mutex_unlock(&iter->mutex); + + return ret; +} + static ssize_t tracing_entries_read(struct file *filp, char __user *ubuf, size_t cnt, loff_t *ppos) { struct trace_array *tr = filp->private_data; - char buf[64]; + char buf[96]; int r; - r = sprintf(buf, "%lu\n", tr->entries >> 10); + mutex_lock(&trace_types_lock); + if (!ring_buffer_expanded) + r = sprintf(buf, "%lu (expanded: %lu)\n", + tr->entries >> 10, + trace_buf_size >> 10); + else + r = sprintf(buf, "%lu\n", tr->entries >> 10); + mutex_unlock(&trace_types_lock); + return simple_read_from_buffer(ubuf, cnt, ppos, buf, r); } @@ -3344,31 +3293,14 @@ tracing_entries_write(struct file *filp, const char __user *ubuf, val <<= 10; if (val != global_trace.entries) { - ret = ring_buffer_resize(global_trace.buffer, val); + ret = tracing_resize_ring_buffer(val); if (ret < 0) { cnt = ret; goto out; } - - ret = ring_buffer_resize(max_tr.buffer, val); - if (ret < 0) { - int r; - cnt = ret; - r = ring_buffer_resize(global_trace.buffer, - global_trace.entries); - if (r < 0) { - /* AARGH! We are left with different - * size max buffer!!!! */ - WARN_ON(1); - tracing_disabled = 1; - } - goto out; - } - - global_trace.entries = val; } - filp->f_pos += cnt; + *ppos += cnt; /* If check pages failed, return ENOMEM */ if (tracing_disabled) @@ -3388,22 +3320,11 @@ tracing_entries_write(struct file *filp, const char __user *ubuf, return cnt; } -static int mark_printk(const char *fmt, ...) -{ - int ret; - va_list args; - va_start(args, fmt); - ret = trace_vprintk(0, -1, fmt, args); - va_end(args); - return ret; -} - static ssize_t tracing_mark_write(struct file *filp, const char __user *ubuf, size_t cnt, loff_t *fpos) { char *buf; - char *end; if (tracing_disabled) return -EINVAL; @@ -3411,7 +3332,7 @@ tracing_mark_write(struct file *filp, const char __user *ubuf, if (cnt > TRACE_BUF_SIZE) cnt = TRACE_BUF_SIZE; - buf = kmalloc(cnt + 1, GFP_KERNEL); + buf = kmalloc(cnt + 2, GFP_KERNEL); if (buf == NULL) return -ENOMEM; @@ -3419,56 +3340,410 @@ tracing_mark_write(struct file *filp, const char __user *ubuf, kfree(buf); return -EFAULT; } + if (buf[cnt-1] != '\n') { + buf[cnt] = '\n'; + buf[cnt+1] = '\0'; + } else + buf[cnt] = '\0'; - /* Cut from the first nil or newline. */ - buf[cnt] = '\0'; - end = strchr(buf, '\n'); - if (end) - *end = '\0'; - - cnt = mark_printk("%s\n", buf); + cnt = trace_vprintk(0, buf, NULL); kfree(buf); *fpos += cnt; return cnt; } -static struct file_operations tracing_max_lat_fops = { +static ssize_t tracing_clock_read(struct file *filp, char __user *ubuf, + size_t cnt, loff_t *ppos) +{ + char buf[64]; + int bufiter = 0; + int i; + + for (i = 0; i < ARRAY_SIZE(trace_clocks); i++) + bufiter += snprintf(buf + bufiter, sizeof(buf) - bufiter, + "%s%s%s%s", i ? " " : "", + i == trace_clock_id ? "[" : "", trace_clocks[i].name, + i == trace_clock_id ? "]" : ""); + bufiter += snprintf(buf + bufiter, sizeof(buf) - bufiter, "\n"); + + return simple_read_from_buffer(ubuf, cnt, ppos, buf, bufiter); +} + +static ssize_t tracing_clock_write(struct file *filp, const char __user *ubuf, + size_t cnt, loff_t *fpos) +{ + char buf[64]; + const char *clockstr; + int i; + + if (cnt >= sizeof(buf)) + return -EINVAL; + + if (copy_from_user(&buf, ubuf, cnt)) + return -EFAULT; + + buf[cnt] = 0; + + clockstr = strstrip(buf); + + for (i = 0; i < ARRAY_SIZE(trace_clocks); i++) { + if (strcmp(trace_clocks[i].name, clockstr) == 0) + break; + } + if (i == ARRAY_SIZE(trace_clocks)) + return -EINVAL; + + trace_clock_id = i; + + mutex_lock(&trace_types_lock); + + ring_buffer_set_clock(global_trace.buffer, trace_clocks[i].func); + if (max_tr.buffer) + ring_buffer_set_clock(max_tr.buffer, trace_clocks[i].func); + + mutex_unlock(&trace_types_lock); + + *fpos += cnt; + + return cnt; +} + +static const struct file_operations tracing_max_lat_fops = { .open = tracing_open_generic, .read = tracing_max_lat_read, .write = tracing_max_lat_write, }; -static struct file_operations tracing_ctrl_fops = { +static const struct file_operations tracing_ctrl_fops = { .open = tracing_open_generic, .read = tracing_ctrl_read, .write = tracing_ctrl_write, }; -static struct file_operations set_tracer_fops = { +static const struct file_operations set_tracer_fops = { .open = tracing_open_generic, .read = tracing_set_trace_read, .write = tracing_set_trace_write, }; -static struct file_operations tracing_pipe_fops = { +static const struct file_operations tracing_pipe_fops = { .open = tracing_open_pipe, .poll = tracing_poll_pipe, .read = tracing_read_pipe, + .splice_read = tracing_splice_read_pipe, .release = tracing_release_pipe, }; -static struct file_operations tracing_entries_fops = { +static const struct file_operations tracing_entries_fops = { .open = tracing_open_generic, .read = tracing_entries_read, .write = tracing_entries_write, }; -static struct file_operations tracing_mark_fops = { +static const struct file_operations tracing_mark_fops = { .open = tracing_open_generic, .write = tracing_mark_write, }; +static const struct file_operations trace_clock_fops = { + .open = tracing_open_generic, + .read = tracing_clock_read, + .write = tracing_clock_write, +}; + +struct ftrace_buffer_info { + struct trace_array *tr; + void *spare; + int cpu; + unsigned int read; +}; + +static int tracing_buffers_open(struct inode *inode, struct file *filp) +{ + int cpu = (int)(long)inode->i_private; + struct ftrace_buffer_info *info; + + if (tracing_disabled) + return -ENODEV; + + info = kzalloc(sizeof(*info), GFP_KERNEL); + if (!info) + return -ENOMEM; + + info->tr = &global_trace; + info->cpu = cpu; + info->spare = NULL; + /* Force reading ring buffer for first read */ + info->read = (unsigned int)-1; + + filp->private_data = info; + + return nonseekable_open(inode, filp); +} + +static ssize_t +tracing_buffers_read(struct file *filp, char __user *ubuf, + size_t count, loff_t *ppos) +{ + struct ftrace_buffer_info *info = filp->private_data; + unsigned int pos; + ssize_t ret; + size_t size; + + if (!count) + return 0; + + if (!info->spare) + info->spare = ring_buffer_alloc_read_page(info->tr->buffer); + if (!info->spare) + return -ENOMEM; + + /* Do we have previous read data to read? */ + if (info->read < PAGE_SIZE) + goto read; + + info->read = 0; + + ret = ring_buffer_read_page(info->tr->buffer, + &info->spare, + count, + info->cpu, 0); + if (ret < 0) + return 0; + + pos = ring_buffer_page_len(info->spare); + + if (pos < PAGE_SIZE) + memset(info->spare + pos, 0, PAGE_SIZE - pos); + +read: + size = PAGE_SIZE - info->read; + if (size > count) + size = count; + + ret = copy_to_user(ubuf, info->spare + info->read, size); + if (ret == size) + return -EFAULT; + size -= ret; + + *ppos += size; + info->read += size; + + return size; +} + +static int tracing_buffers_release(struct inode *inode, struct file *file) +{ + struct ftrace_buffer_info *info = file->private_data; + + if (info->spare) + ring_buffer_free_read_page(info->tr->buffer, info->spare); + kfree(info); + + return 0; +} + +struct buffer_ref { + struct ring_buffer *buffer; + void *page; + int ref; +}; + +static void buffer_pipe_buf_release(struct pipe_inode_info *pipe, + struct pipe_buffer *buf) +{ + struct buffer_ref *ref = (struct buffer_ref *)buf->private; + + if (--ref->ref) + return; + + ring_buffer_free_read_page(ref->buffer, ref->page); + kfree(ref); + buf->private = 0; +} + +static int buffer_pipe_buf_steal(struct pipe_inode_info *pipe, + struct pipe_buffer *buf) +{ + return 1; +} + +static void buffer_pipe_buf_get(struct pipe_inode_info *pipe, + struct pipe_buffer *buf) +{ + struct buffer_ref *ref = (struct buffer_ref *)buf->private; + + ref->ref++; +} + +/* Pipe buffer operations for a buffer. */ +static struct pipe_buf_operations buffer_pipe_buf_ops = { + .can_merge = 0, + .map = generic_pipe_buf_map, + .unmap = generic_pipe_buf_unmap, + .confirm = generic_pipe_buf_confirm, + .release = buffer_pipe_buf_release, + .steal = buffer_pipe_buf_steal, + .get = buffer_pipe_buf_get, +}; + +/* + * Callback from splice_to_pipe(), if we need to release some pages + * at the end of the spd in case we error'ed out in filling the pipe. + */ +static void buffer_spd_release(struct splice_pipe_desc *spd, unsigned int i) +{ + struct buffer_ref *ref = + (struct buffer_ref *)spd->partial[i].private; + + if (--ref->ref) + return; + + ring_buffer_free_read_page(ref->buffer, ref->page); + kfree(ref); + spd->partial[i].private = 0; +} + +static ssize_t +tracing_buffers_splice_read(struct file *file, loff_t *ppos, + struct pipe_inode_info *pipe, size_t len, + unsigned int flags) +{ + struct ftrace_buffer_info *info = file->private_data; + struct partial_page partial[PIPE_BUFFERS]; + struct page *pages[PIPE_BUFFERS]; + struct splice_pipe_desc spd = { + .pages = pages, + .partial = partial, + .flags = flags, + .ops = &buffer_pipe_buf_ops, + .spd_release = buffer_spd_release, + }; + struct buffer_ref *ref; + int entries, size, i; + size_t ret; + + if (*ppos & (PAGE_SIZE - 1)) { + WARN_ONCE(1, "Ftrace: previous read must page-align\n"); + return -EINVAL; + } + + if (len & (PAGE_SIZE - 1)) { + WARN_ONCE(1, "Ftrace: splice_read should page-align\n"); + if (len < PAGE_SIZE) + return -EINVAL; + len &= PAGE_MASK; + } + + entries = ring_buffer_entries_cpu(info->tr->buffer, info->cpu); + + for (i = 0; i < PIPE_BUFFERS && len && entries; i++, len -= PAGE_SIZE) { + struct page *page; + int r; + + ref = kzalloc(sizeof(*ref), GFP_KERNEL); + if (!ref) + break; + + ref->ref = 1; + ref->buffer = info->tr->buffer; + ref->page = ring_buffer_alloc_read_page(ref->buffer); + if (!ref->page) { + kfree(ref); + break; + } + + r = ring_buffer_read_page(ref->buffer, &ref->page, + len, info->cpu, 1); + if (r < 0) { + ring_buffer_free_read_page(ref->buffer, + ref->page); + kfree(ref); + break; + } + + /* + * zero out any left over data, this is going to + * user land. + */ + size = ring_buffer_page_len(ref->page); + if (size < PAGE_SIZE) + memset(ref->page + size, 0, PAGE_SIZE - size); + + page = virt_to_page(ref->page); + + spd.pages[i] = page; + spd.partial[i].len = PAGE_SIZE; + spd.partial[i].offset = 0; + spd.partial[i].private = (unsigned long)ref; + spd.nr_pages++; + *ppos += PAGE_SIZE; + + entries = ring_buffer_entries_cpu(info->tr->buffer, info->cpu); + } + + spd.nr_pages = i; + + /* did we read anything? */ + if (!spd.nr_pages) { + if (flags & SPLICE_F_NONBLOCK) + ret = -EAGAIN; + else + ret = 0; + /* TODO: block */ + return ret; + } + + ret = splice_to_pipe(pipe, &spd); + + return ret; +} + +static const struct file_operations tracing_buffers_fops = { + .open = tracing_buffers_open, + .read = tracing_buffers_read, + .release = tracing_buffers_release, + .splice_read = tracing_buffers_splice_read, + .llseek = no_llseek, +}; + +static ssize_t +tracing_stats_read(struct file *filp, char __user *ubuf, + size_t count, loff_t *ppos) +{ + unsigned long cpu = (unsigned long)filp->private_data; + struct trace_array *tr = &global_trace; + struct trace_seq *s; + unsigned long cnt; + + s = kmalloc(sizeof(*s), GFP_KERNEL); + if (!s) + return -ENOMEM; + + trace_seq_init(s); + + cnt = ring_buffer_entries_cpu(tr->buffer, cpu); + trace_seq_printf(s, "entries: %ld\n", cnt); + + cnt = ring_buffer_overrun_cpu(tr->buffer, cpu); + trace_seq_printf(s, "overrun: %ld\n", cnt); + + cnt = ring_buffer_commit_overrun_cpu(tr->buffer, cpu); + trace_seq_printf(s, "commit overrun: %ld\n", cnt); + + count = simple_read_from_buffer(ubuf, count, ppos, s->buffer, s->len); + + kfree(s); + + return count; +} + +static const struct file_operations tracing_stats_fops = { + .open = tracing_open_generic, + .read = tracing_stats_read, +}; + #ifdef CONFIG_DYNAMIC_FTRACE int __weak ftrace_arch_read_dyn_info(char *buf, int size) @@ -3500,7 +3775,7 @@ tracing_read_dyn_info(struct file *filp, char __user *ubuf, return r; } -static struct file_operations tracing_dyn_info_fops = { +static const struct file_operations tracing_dyn_info_fops = { .open = tracing_open_generic, .read = tracing_read_dyn_info, }; @@ -3515,6 +3790,9 @@ struct dentry *tracing_init_dentry(void) if (d_tracer) return d_tracer; + if (!debugfs_initialized()) + return NULL; + d_tracer = debugfs_create_dir("tracing", NULL); if (!d_tracer && !once) { @@ -3526,170 +3804,405 @@ struct dentry *tracing_init_dentry(void) return d_tracer; } +static struct dentry *d_percpu; + +struct dentry *tracing_dentry_percpu(void) +{ + static int once; + struct dentry *d_tracer; + + if (d_percpu) + return d_percpu; + + d_tracer = tracing_init_dentry(); + + if (!d_tracer) + return NULL; + + d_percpu = debugfs_create_dir("per_cpu", d_tracer); + + if (!d_percpu && !once) { + once = 1; + pr_warning("Could not create debugfs directory 'per_cpu'\n"); + return NULL; + } + + return d_percpu; +} + +static void tracing_init_debugfs_percpu(long cpu) +{ + struct dentry *d_percpu = tracing_dentry_percpu(); + struct dentry *d_cpu; + /* strlen(cpu) + MAX(log10(cpu)) + '\0' */ + char cpu_dir[7]; + + if (cpu > 999 || cpu < 0) + return; + + sprintf(cpu_dir, "cpu%ld", cpu); + d_cpu = debugfs_create_dir(cpu_dir, d_percpu); + if (!d_cpu) { + pr_warning("Could not create debugfs '%s' entry\n", cpu_dir); + return; + } + + /* per cpu trace_pipe */ + trace_create_file("trace_pipe", 0444, d_cpu, + (void *) cpu, &tracing_pipe_fops); + + /* per cpu trace */ + trace_create_file("trace", 0644, d_cpu, + (void *) cpu, &tracing_fops); + + trace_create_file("trace_pipe_raw", 0444, d_cpu, + (void *) cpu, &tracing_buffers_fops); + + trace_create_file("stats", 0444, d_cpu, + (void *) cpu, &tracing_stats_fops); +} + #ifdef CONFIG_FTRACE_SELFTEST /* Let selftest have access to static functions in this file */ #include "trace_selftest.c" #endif -static __init int tracer_init_debugfs(void) +struct trace_option_dentry { + struct tracer_opt *opt; + struct tracer_flags *flags; + struct dentry *entry; +}; + +static ssize_t +trace_options_read(struct file *filp, char __user *ubuf, size_t cnt, + loff_t *ppos) +{ + struct trace_option_dentry *topt = filp->private_data; + char *buf; + + if (topt->flags->val & topt->opt->bit) + buf = "1\n"; + else + buf = "0\n"; + + return simple_read_from_buffer(ubuf, cnt, ppos, buf, 2); +} + +static ssize_t +trace_options_write(struct file *filp, const char __user *ubuf, size_t cnt, + loff_t *ppos) +{ + struct trace_option_dentry *topt = filp->private_data; + unsigned long val; + char buf[64]; + int ret; + + if (cnt >= sizeof(buf)) + return -EINVAL; + + if (copy_from_user(&buf, ubuf, cnt)) + return -EFAULT; + + buf[cnt] = 0; + + ret = strict_strtoul(buf, 10, &val); + if (ret < 0) + return ret; + + ret = 0; + switch (val) { + case 0: + /* do nothing if already cleared */ + if (!(topt->flags->val & topt->opt->bit)) + break; + + mutex_lock(&trace_types_lock); + if (current_trace->set_flag) + ret = current_trace->set_flag(topt->flags->val, + topt->opt->bit, 0); + mutex_unlock(&trace_types_lock); + if (ret) + return ret; + topt->flags->val &= ~topt->opt->bit; + break; + case 1: + /* do nothing if already set */ + if (topt->flags->val & topt->opt->bit) + break; + + mutex_lock(&trace_types_lock); + if (current_trace->set_flag) + ret = current_trace->set_flag(topt->flags->val, + topt->opt->bit, 1); + mutex_unlock(&trace_types_lock); + if (ret) + return ret; + topt->flags->val |= topt->opt->bit; + break; + + default: + return -EINVAL; + } + + *ppos += cnt; + + return cnt; +} + + +static const struct file_operations trace_options_fops = { + .open = tracing_open_generic, + .read = trace_options_read, + .write = trace_options_write, +}; + +static ssize_t +trace_options_core_read(struct file *filp, char __user *ubuf, size_t cnt, + loff_t *ppos) +{ + long index = (long)filp->private_data; + char *buf; + + if (trace_flags & (1 << index)) + buf = "1\n"; + else + buf = "0\n"; + + return simple_read_from_buffer(ubuf, cnt, ppos, buf, 2); +} + +static ssize_t +trace_options_core_write(struct file *filp, const char __user *ubuf, size_t cnt, + loff_t *ppos) +{ + long index = (long)filp->private_data; + char buf[64]; + unsigned long val; + int ret; + + if (cnt >= sizeof(buf)) + return -EINVAL; + + if (copy_from_user(&buf, ubuf, cnt)) + return -EFAULT; + + buf[cnt] = 0; + + ret = strict_strtoul(buf, 10, &val); + if (ret < 0) + return ret; + + if (val != 0 && val != 1) + return -EINVAL; + set_tracer_flags(1 << index, val); + + *ppos += cnt; + + return cnt; +} + +static const struct file_operations trace_options_core_fops = { + .open = tracing_open_generic, + .read = trace_options_core_read, + .write = trace_options_core_write, +}; + +struct dentry *trace_create_file(const char *name, + mode_t mode, + struct dentry *parent, + void *data, + const struct file_operations *fops) +{ + struct dentry *ret; + + ret = debugfs_create_file(name, mode, parent, data, fops); + if (!ret) + pr_warning("Could not create debugfs '%s' entry\n", name); + + return ret; +} + + +static struct dentry *trace_options_init_dentry(void) { struct dentry *d_tracer; - struct dentry *entry; + static struct dentry *t_options; + + if (t_options) + return t_options; d_tracer = tracing_init_dentry(); + if (!d_tracer) + return NULL; - entry = debugfs_create_file("tracing_enabled", 0644, d_tracer, - &global_trace, &tracing_ctrl_fops); - if (!entry) - pr_warning("Could not create debugfs 'tracing_enabled' entry\n"); - - entry = debugfs_create_file("trace_options", 0644, d_tracer, - NULL, &tracing_iter_fops); - if (!entry) - pr_warning("Could not create debugfs 'trace_options' entry\n"); - - entry = debugfs_create_file("tracing_cpumask", 0644, d_tracer, - NULL, &tracing_cpumask_fops); - if (!entry) - pr_warning("Could not create debugfs 'tracing_cpumask' entry\n"); - - entry = debugfs_create_file("latency_trace", 0444, d_tracer, - &global_trace, &tracing_lt_fops); - if (!entry) - pr_warning("Could not create debugfs 'latency_trace' entry\n"); - - entry = debugfs_create_file("trace", 0444, d_tracer, - &global_trace, &tracing_fops); - if (!entry) - pr_warning("Could not create debugfs 'trace' entry\n"); - - entry = debugfs_create_file("available_tracers", 0444, d_tracer, - &global_trace, &show_traces_fops); - if (!entry) - pr_warning("Could not create debugfs 'available_tracers' entry\n"); - - entry = debugfs_create_file("current_tracer", 0444, d_tracer, - &global_trace, &set_tracer_fops); - if (!entry) - pr_warning("Could not create debugfs 'current_tracer' entry\n"); - - entry = debugfs_create_file("tracing_max_latency", 0644, d_tracer, - &tracing_max_latency, - &tracing_max_lat_fops); - if (!entry) - pr_warning("Could not create debugfs " - "'tracing_max_latency' entry\n"); - - entry = debugfs_create_file("tracing_thresh", 0644, d_tracer, - &tracing_thresh, &tracing_max_lat_fops); - if (!entry) - pr_warning("Could not create debugfs " - "'tracing_thresh' entry\n"); - entry = debugfs_create_file("README", 0644, d_tracer, - NULL, &tracing_readme_fops); - if (!entry) - pr_warning("Could not create debugfs 'README' entry\n"); - - entry = debugfs_create_file("trace_pipe", 0644, d_tracer, - NULL, &tracing_pipe_fops); - if (!entry) - pr_warning("Could not create debugfs " - "'trace_pipe' entry\n"); - - entry = debugfs_create_file("buffer_size_kb", 0644, d_tracer, - &global_trace, &tracing_entries_fops); - if (!entry) - pr_warning("Could not create debugfs " - "'buffer_size_kb' entry\n"); - - entry = debugfs_create_file("trace_marker", 0220, d_tracer, - NULL, &tracing_mark_fops); - if (!entry) - pr_warning("Could not create debugfs " - "'trace_marker' entry\n"); + t_options = debugfs_create_dir("options", d_tracer); + if (!t_options) { + pr_warning("Could not create debugfs directory 'options'\n"); + return NULL; + } -#ifdef CONFIG_DYNAMIC_FTRACE - entry = debugfs_create_file("dyn_ftrace_total_info", 0444, d_tracer, - &ftrace_update_tot_cnt, - &tracing_dyn_info_fops); - if (!entry) - pr_warning("Could not create debugfs " - "'dyn_ftrace_total_info' entry\n"); -#endif -#ifdef CONFIG_SYSPROF_TRACER - init_tracer_sysprof_debugfs(d_tracer); -#endif - return 0; + return t_options; } -int trace_vprintk(unsigned long ip, int depth, const char *fmt, va_list args) +static void +create_trace_option_file(struct trace_option_dentry *topt, + struct tracer_flags *flags, + struct tracer_opt *opt) { - static DEFINE_SPINLOCK(trace_buf_lock); - static char trace_buf[TRACE_BUF_SIZE]; + struct dentry *t_options; - struct ring_buffer_event *event; - struct trace_array *tr = &global_trace; - struct trace_array_cpu *data; - int cpu, len = 0, size, pc; - struct print_entry *entry; - unsigned long irq_flags; + t_options = trace_options_init_dentry(); + if (!t_options) + return; - if (tracing_disabled || tracing_selftest_running) - return 0; + topt->flags = flags; + topt->opt = opt; - pc = preempt_count(); - preempt_disable_notrace(); - cpu = raw_smp_processor_id(); - data = tr->data[cpu]; + topt->entry = trace_create_file(opt->name, 0644, t_options, topt, + &trace_options_fops); - if (unlikely(atomic_read(&data->disabled))) - goto out; +} - pause_graph_tracing(); - spin_lock_irqsave(&trace_buf_lock, irq_flags); - len = vsnprintf(trace_buf, TRACE_BUF_SIZE, fmt, args); +static struct trace_option_dentry * +create_trace_option_files(struct tracer *tracer) +{ + struct trace_option_dentry *topts; + struct tracer_flags *flags; + struct tracer_opt *opts; + int cnt; - len = min(len, TRACE_BUF_SIZE-1); - trace_buf[len] = 0; + if (!tracer) + return NULL; - size = sizeof(*entry) + len + 1; - event = ring_buffer_lock_reserve(tr->buffer, size, &irq_flags); - if (!event) - goto out_unlock; - entry = ring_buffer_event_data(event); - tracing_generic_entry_update(&entry->ent, irq_flags, pc); - entry->ent.type = TRACE_PRINT; - entry->ip = ip; - entry->depth = depth; + flags = tracer->flags; - memcpy(&entry->buf, trace_buf, len); - entry->buf[len] = 0; - ring_buffer_unlock_commit(tr->buffer, event, irq_flags); + if (!flags || !flags->opts) + return NULL; - out_unlock: - spin_unlock_irqrestore(&trace_buf_lock, irq_flags); - unpause_graph_tracing(); - out: - preempt_enable_notrace(); + opts = flags->opts; - return len; + for (cnt = 0; opts[cnt].name; cnt++) + ; + + topts = kcalloc(cnt + 1, sizeof(*topts), GFP_KERNEL); + if (!topts) + return NULL; + + for (cnt = 0; opts[cnt].name; cnt++) + create_trace_option_file(&topts[cnt], flags, + &opts[cnt]); + + return topts; } -EXPORT_SYMBOL_GPL(trace_vprintk); -int __ftrace_printk(unsigned long ip, const char *fmt, ...) +static void +destroy_trace_option_files(struct trace_option_dentry *topts) { - int ret; - va_list ap; + int cnt; - if (!(trace_flags & TRACE_ITER_PRINTK)) - return 0; + if (!topts) + return; - va_start(ap, fmt); - ret = trace_vprintk(ip, task_curr_ret_stack(current), fmt, ap); - va_end(ap); - return ret; + for (cnt = 0; topts[cnt].opt; cnt++) { + if (topts[cnt].entry) + debugfs_remove(topts[cnt].entry); + } + + kfree(topts); +} + +static struct dentry * +create_trace_option_core_file(const char *option, long index) +{ + struct dentry *t_options; + + t_options = trace_options_init_dentry(); + if (!t_options) + return NULL; + + return trace_create_file(option, 0644, t_options, (void *)index, + &trace_options_core_fops); +} + +static __init void create_trace_options_dir(void) +{ + struct dentry *t_options; + int i; + + t_options = trace_options_init_dentry(); + if (!t_options) + return; + + for (i = 0; trace_options[i]; i++) + create_trace_option_core_file(trace_options[i], i); +} + +static __init int tracer_init_debugfs(void) +{ + struct dentry *d_tracer; + int cpu; + + d_tracer = tracing_init_dentry(); + + trace_create_file("tracing_enabled", 0644, d_tracer, + &global_trace, &tracing_ctrl_fops); + + trace_create_file("trace_options", 0644, d_tracer, + NULL, &tracing_iter_fops); + + trace_create_file("tracing_cpumask", 0644, d_tracer, + NULL, &tracing_cpumask_fops); + + trace_create_file("trace", 0644, d_tracer, + (void *) TRACE_PIPE_ALL_CPU, &tracing_fops); + + trace_create_file("available_tracers", 0444, d_tracer, + &global_trace, &show_traces_fops); + + trace_create_file("current_tracer", 0644, d_tracer, + &global_trace, &set_tracer_fops); + +#ifdef CONFIG_TRACER_MAX_TRACE + trace_create_file("tracing_max_latency", 0644, d_tracer, + &tracing_max_latency, &tracing_max_lat_fops); + + trace_create_file("tracing_thresh", 0644, d_tracer, + &tracing_thresh, &tracing_max_lat_fops); +#endif + + trace_create_file("README", 0444, d_tracer, + NULL, &tracing_readme_fops); + + trace_create_file("trace_pipe", 0444, d_tracer, + (void *) TRACE_PIPE_ALL_CPU, &tracing_pipe_fops); + + trace_create_file("buffer_size_kb", 0644, d_tracer, + &global_trace, &tracing_entries_fops); + + trace_create_file("trace_marker", 0220, d_tracer, + NULL, &tracing_mark_fops); + + trace_create_file("saved_cmdlines", 0444, d_tracer, + NULL, &tracing_saved_cmdlines_fops); + + trace_create_file("trace_clock", 0644, d_tracer, NULL, + &trace_clock_fops); + +#ifdef CONFIG_DYNAMIC_FTRACE + trace_create_file("dyn_ftrace_total_info", 0444, d_tracer, + &ftrace_update_tot_cnt, &tracing_dyn_info_fops); +#endif +#ifdef CONFIG_SYSPROF_TRACER + init_tracer_sysprof_debugfs(d_tracer); +#endif + + create_trace_options_dir(); + + for_each_tracing_cpu(cpu) + tracing_init_debugfs_percpu(cpu); + + return 0; } -EXPORT_SYMBOL_GPL(__ftrace_printk); static int trace_panic_handler(struct notifier_block *this, unsigned long event, void *unused) @@ -3750,40 +4263,48 @@ trace_printk_seq(struct trace_seq *s) printk(KERN_TRACE "%s", s->buffer); - trace_seq_reset(s); + trace_seq_init(s); } -void ftrace_dump(void) +static void __ftrace_dump(bool disable_tracing) { - static DEFINE_SPINLOCK(ftrace_dump_lock); + static raw_spinlock_t ftrace_dump_lock = + (raw_spinlock_t)__RAW_SPIN_LOCK_UNLOCKED; /* use static because iter can be a bit big for the stack */ static struct trace_iterator iter; + unsigned int old_userobj; static int dump_ran; unsigned long flags; int cnt = 0, cpu; /* only one dump */ - spin_lock_irqsave(&ftrace_dump_lock, flags); + local_irq_save(flags); + __raw_spin_lock(&ftrace_dump_lock); if (dump_ran) goto out; dump_ran = 1; - /* No turning back! */ tracing_off(); - ftrace_kill(); + + if (disable_tracing) + ftrace_kill(); for_each_tracing_cpu(cpu) { atomic_inc(&global_trace.data[cpu]->disabled); } + old_userobj = trace_flags & TRACE_ITER_SYM_USEROBJ; + /* don't look at user memory in panic mode */ trace_flags &= ~TRACE_ITER_SYM_USEROBJ; printk(KERN_TRACE "Dumping ftrace buffer:\n"); + /* Simulate the iterator */ iter.tr = &global_trace; iter.trace = current_trace; + iter.cpu_file = TRACE_PIPE_ALL_CPU; /* * We need to stop all tracing on all CPUS to read the @@ -3807,8 +4328,11 @@ void ftrace_dump(void) iter.pos = -1; if (find_next_entry_inc(&iter) != NULL) { - print_trace_line(&iter); - trace_consume(&iter); + int ret; + + ret = print_trace_line(&iter); + if (ret != TRACE_TYPE_NO_CONSUME) + trace_consume(&iter); } trace_printk_seq(&iter.seq); @@ -3819,13 +4343,30 @@ void ftrace_dump(void) else printk(KERN_TRACE "---------------------------------\n"); + /* Re-enable tracing if requested */ + if (!disable_tracing) { + trace_flags |= old_userobj; + + for_each_tracing_cpu(cpu) { + atomic_dec(&global_trace.data[cpu]->disabled); + } + tracing_on(); + } + out: - spin_unlock_irqrestore(&ftrace_dump_lock, flags); + __raw_spin_unlock(&ftrace_dump_lock); + local_irq_restore(flags); +} + +/* By default: disable tracing after the dump */ +void ftrace_dump(void) +{ + __ftrace_dump(true); } __init static int tracer_alloc_buffers(void) { - struct trace_array_cpu *data; + int ring_buf_size; int i; int ret = -ENOMEM; @@ -3835,11 +4376,20 @@ __init static int tracer_alloc_buffers(void) if (!alloc_cpumask_var(&tracing_cpumask, GFP_KERNEL)) goto out_free_buffer_mask; + if (!zalloc_cpumask_var(&tracing_reader_cpumask, GFP_KERNEL)) + goto out_free_tracing_cpumask; + + /* To save memory, keep the ring buffer size to its minimum */ + if (ring_buffer_expanded) + ring_buf_size = trace_buf_size; + else + ring_buf_size = 1; + cpumask_copy(tracing_buffer_mask, cpu_possible_mask); cpumask_copy(tracing_cpumask, cpu_all_mask); /* TODO: make the number of buffers hot pluggable with CPUS */ - global_trace.buffer = ring_buffer_alloc(trace_buf_size, + global_trace.buffer = ring_buffer_alloc(ring_buf_size, TRACE_BUFFER_FLAGS); if (!global_trace.buffer) { printk(KERN_ERR "tracer: failed to allocate ring buffer!\n"); @@ -3850,7 +4400,7 @@ __init static int tracer_alloc_buffers(void) #ifdef CONFIG_TRACER_MAX_TRACE - max_tr.buffer = ring_buffer_alloc(trace_buf_size, + max_tr.buffer = ring_buffer_alloc(ring_buf_size, TRACE_BUFFER_FLAGS); if (!max_tr.buffer) { printk(KERN_ERR "tracer: failed to allocate max ring buffer!\n"); @@ -3864,21 +4414,17 @@ __init static int tracer_alloc_buffers(void) /* Allocate the first page for all buffers */ for_each_tracing_cpu(i) { - data = global_trace.data[i] = &per_cpu(global_trace_cpu, i); + global_trace.data[i] = &per_cpu(global_trace_cpu, i); max_tr.data[i] = &per_cpu(max_data, i); } trace_init_cmdlines(); register_tracer(&nop_trace); + current_trace = &nop_trace; #ifdef CONFIG_BOOT_TRACER register_tracer(&boot_tracer); - current_trace = &boot_tracer; - current_trace->init(&global_trace); -#else - current_trace = &nop_trace; #endif - /* All seems OK, enable tracing */ tracing_disabled = 0; @@ -3886,14 +4432,38 @@ __init static int tracer_alloc_buffers(void) &trace_panic_notifier); register_die_notifier(&trace_die_notifier); - ret = 0; + + return 0; out_free_cpumask: + free_cpumask_var(tracing_reader_cpumask); +out_free_tracing_cpumask: free_cpumask_var(tracing_cpumask); out_free_buffer_mask: free_cpumask_var(tracing_buffer_mask); out: return ret; } + +__init static int clear_boot_tracer(void) +{ + /* + * The default tracer at boot buffer is an init section. + * This function is called in lateinit. If we did not + * find the boot tracer, then clear it out, to prevent + * later registration from accessing the buffer that is + * about to be freed. + */ + if (!default_bootup_tracer) + return 0; + + printk(KERN_INFO "ftrace bootup tracer '%s' not registered.\n", + default_bootup_tracer); + default_bootup_tracer = NULL; + + return 0; +} + early_initcall(tracer_alloc_buffers); fs_initcall(tracer_init_debugfs); +late_initcall(clear_boot_tracer); diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h index 4d3d381bfd9..1d7f4830a80 100644 --- a/kernel/trace/trace.h +++ b/kernel/trace/trace.h @@ -7,8 +7,14 @@ #include <linux/clocksource.h> #include <linux/ring_buffer.h> #include <linux/mmiotrace.h> +#include <linux/tracepoint.h> #include <linux/ftrace.h> #include <trace/boot.h> +#include <linux/kmemtrace.h> +#include <linux/hw_breakpoint.h> + +#include <linux/trace_seq.h> +#include <linux/ftrace_event.h> enum trace_type { __TRACE_FIRST_TYPE = 0, @@ -16,9 +22,9 @@ enum trace_type { TRACE_FN, TRACE_CTX, TRACE_WAKE, - TRACE_CONT, TRACE_STACK, TRACE_PRINT, + TRACE_BPRINT, TRACE_SPECIAL, TRACE_MMIO_RW, TRACE_MMIO_MAP, @@ -29,156 +35,105 @@ enum trace_type { TRACE_GRAPH_ENT, TRACE_USER_STACK, TRACE_HW_BRANCHES, - TRACE_POWER, - - __TRACE_LAST_TYPE -}; + TRACE_KMEM_ALLOC, + TRACE_KMEM_FREE, + TRACE_BLK, + TRACE_KSYM, -/* - * The trace entry - the most basic unit of tracing. This is what - * is printed in the end as a single line in the trace output, such as: - * - * bash-15816 [01] 235.197585: idle_cpu <- irq_enter - */ -struct trace_entry { - unsigned char type; - unsigned char cpu; - unsigned char flags; - unsigned char preempt_count; - int pid; - int tgid; -}; - -/* - * Function trace entry - function address and parent function addres: - */ -struct ftrace_entry { - struct trace_entry ent; - unsigned long ip; - unsigned long parent_ip; + __TRACE_LAST_TYPE, }; -/* Function call entry */ -struct ftrace_graph_ent_entry { - struct trace_entry ent; - struct ftrace_graph_ent graph_ent; +enum kmemtrace_type_id { + KMEMTRACE_TYPE_KMALLOC = 0, /* kmalloc() or kfree(). */ + KMEMTRACE_TYPE_CACHE, /* kmem_cache_*(). */ + KMEMTRACE_TYPE_PAGES, /* __get_free_pages() and friends. */ }; -/* Function return entry */ -struct ftrace_graph_ret_entry { - struct trace_entry ent; - struct ftrace_graph_ret ret; -}; extern struct tracer boot_tracer; -/* - * Context switch trace entry - which task (and prio) we switched from/to: - */ -struct ctx_switch_entry { - struct trace_entry ent; - unsigned int prev_pid; - unsigned char prev_prio; - unsigned char prev_state; - unsigned int next_pid; - unsigned char next_prio; - unsigned char next_state; - unsigned int next_cpu; -}; +#undef __field +#define __field(type, item) type item; -/* - * Special (free-form) trace entry: - */ -struct special_entry { - struct trace_entry ent; - unsigned long arg1; - unsigned long arg2; - unsigned long arg3; -}; +#undef __field_struct +#define __field_struct(type, item) __field(type, item) -/* - * Stack-trace entry: - */ +#undef __field_desc +#define __field_desc(type, container, item) -#define FTRACE_STACK_ENTRIES 8 +#undef __array +#define __array(type, item, size) type item[size]; -struct stack_entry { - struct trace_entry ent; - unsigned long caller[FTRACE_STACK_ENTRIES]; -}; +#undef __array_desc +#define __array_desc(type, container, item, size) -struct userstack_entry { - struct trace_entry ent; - unsigned long caller[FTRACE_STACK_ENTRIES]; -}; +#undef __dynamic_array +#define __dynamic_array(type, item) type item[]; -/* - * ftrace_printk entry: - */ -struct print_entry { - struct trace_entry ent; - unsigned long ip; - int depth; - char buf[]; -}; +#undef F_STRUCT +#define F_STRUCT(args...) args -#define TRACE_OLD_SIZE 88 +#undef FTRACE_ENTRY +#define FTRACE_ENTRY(name, struct_name, id, tstruct, print) \ + struct struct_name { \ + struct trace_entry ent; \ + tstruct \ + } -struct trace_field_cont { - unsigned char type; - /* Temporary till we get rid of this completely */ - char buf[TRACE_OLD_SIZE - 1]; -}; +#undef TP_ARGS +#define TP_ARGS(args...) args -struct trace_mmiotrace_rw { - struct trace_entry ent; - struct mmiotrace_rw rw; -}; +#undef FTRACE_ENTRY_DUP +#define FTRACE_ENTRY_DUP(name, name_struct, id, tstruct, printk) -struct trace_mmiotrace_map { - struct trace_entry ent; - struct mmiotrace_map map; -}; +#include "trace_entries.h" -struct trace_boot_call { +/* + * syscalls are special, and need special handling, this is why + * they are not included in trace_entries.h + */ +struct syscall_trace_enter { struct trace_entry ent; - struct boot_trace_call boot_call; + int nr; + unsigned long args[]; }; -struct trace_boot_ret { +struct syscall_trace_exit { struct trace_entry ent; - struct boot_trace_ret boot_ret; + int nr; + long ret; }; -#define TRACE_FUNC_SIZE 30 -#define TRACE_FILE_SIZE 20 -struct trace_branch { +struct kprobe_trace_entry { struct trace_entry ent; - unsigned line; - char func[TRACE_FUNC_SIZE+1]; - char file[TRACE_FILE_SIZE+1]; - char correct; + unsigned long ip; + int nargs; + unsigned long args[]; }; -struct hw_branch_entry { - struct trace_entry ent; - u64 from; - u64 to; -}; +#define SIZEOF_KPROBE_TRACE_ENTRY(n) \ + (offsetof(struct kprobe_trace_entry, args) + \ + (sizeof(unsigned long) * (n))) -struct trace_power { +struct kretprobe_trace_entry { struct trace_entry ent; - struct power_trace state_data; + unsigned long func; + unsigned long ret_ip; + int nargs; + unsigned long args[]; }; +#define SIZEOF_KRETPROBE_TRACE_ENTRY(n) \ + (offsetof(struct kretprobe_trace_entry, args) + \ + (sizeof(unsigned long) * (n))) + /* * trace_flag_type is an enumeration that holds different * states when a trace occurs. These are: * IRQS_OFF - interrupts were disabled - * IRQS_NOSUPPORT - arch does not support irqs_disabled_flags - * NEED_RESCED - reschedule is requested + * IRQS_NOSUPPORT - arch does not support irqs_disabled_flags + * NEED_RESCHED - reschedule is requested * HARDIRQ - inside an interrupt handler * SOFTIRQ - inside a softirq handler - * CONT - multiple entries hold the trace item */ enum trace_flag_type { TRACE_FLAG_IRQS_OFF = 0x01, @@ -186,7 +141,6 @@ enum trace_flag_type { TRACE_FLAG_NEED_RESCHED = 0x04, TRACE_FLAG_HARDIRQ = 0x08, TRACE_FLAG_SOFTIRQ = 0x10, - TRACE_FLAG_CONT = 0x20, }; #define TRACE_BUF_SIZE 1024 @@ -198,10 +152,8 @@ enum trace_flag_type { */ struct trace_array_cpu { atomic_t disabled; + void *buffer_page; /* ring buffer spare */ - /* these fields get copied into max-trace: */ - unsigned long trace_idx; - unsigned long overrun; unsigned long saved_latency; unsigned long critical_start; unsigned long critical_end; @@ -209,14 +161,13 @@ struct trace_array_cpu { unsigned long nice; unsigned long policy; unsigned long rt_priority; + unsigned long skipped_entries; cycle_t preempt_timestamp; pid_t pid; uid_t uid; char comm[TASK_COMM_LEN]; }; -struct trace_iterator; - /* * The trace array - an array of per-CPU trace arrays. This is the * highest level data structure that individual tracers deal with. @@ -262,10 +213,10 @@ extern void __ftrace_bad_type(void); do { \ IF_ASSIGN(var, ent, struct ftrace_entry, TRACE_FN); \ IF_ASSIGN(var, ent, struct ctx_switch_entry, 0); \ - IF_ASSIGN(var, ent, struct trace_field_cont, TRACE_CONT); \ IF_ASSIGN(var, ent, struct stack_entry, TRACE_STACK); \ IF_ASSIGN(var, ent, struct userstack_entry, TRACE_USER_STACK);\ IF_ASSIGN(var, ent, struct print_entry, TRACE_PRINT); \ + IF_ASSIGN(var, ent, struct bprint_entry, TRACE_BPRINT); \ IF_ASSIGN(var, ent, struct special_entry, 0); \ IF_ASSIGN(var, ent, struct trace_mmiotrace_rw, \ TRACE_MMIO_RW); \ @@ -279,26 +230,22 @@ extern void __ftrace_bad_type(void); IF_ASSIGN(var, ent, struct ftrace_graph_ret_entry, \ TRACE_GRAPH_RET); \ IF_ASSIGN(var, ent, struct hw_branch_entry, TRACE_HW_BRANCHES);\ - IF_ASSIGN(var, ent, struct trace_power, TRACE_POWER); \ + IF_ASSIGN(var, ent, struct kmemtrace_alloc_entry, \ + TRACE_KMEM_ALLOC); \ + IF_ASSIGN(var, ent, struct kmemtrace_free_entry, \ + TRACE_KMEM_FREE); \ + IF_ASSIGN(var, ent, struct ksym_trace_entry, TRACE_KSYM);\ __ftrace_bad_type(); \ } while (0) -/* Return values for print_line callback */ -enum print_line_t { - TRACE_TYPE_PARTIAL_LINE = 0, /* Retry after flushing the seq */ - TRACE_TYPE_HANDLED = 1, - TRACE_TYPE_UNHANDLED = 2 /* Relay to other output functions */ -}; - - /* * An option specific to a tracer. This is a boolean value. * The bit is the bit index that sets its value on the * flags value in struct tracer_flags. */ struct tracer_opt { - const char *name; /* Will appear on the trace_options file */ - u32 bit; /* Mask assigned in val field in tracer_flags */ + const char *name; /* Will appear on the trace_options file */ + u32 bit; /* Mask assigned in val field in tracer_flags */ }; /* @@ -307,28 +254,51 @@ struct tracer_opt { */ struct tracer_flags { u32 val; - struct tracer_opt *opts; + struct tracer_opt *opts; }; /* Makes more easy to define a tracer opt */ #define TRACER_OPT(s, b) .name = #s, .bit = b -/* - * A specific tracer, represented by methods that operate on a trace array: + +/** + * struct tracer - a specific tracer and its callbacks to interact with debugfs + * @name: the name chosen to select it on the available_tracers file + * @init: called when one switches to this tracer (echo name > current_tracer) + * @reset: called when one switches to another tracer + * @start: called when tracing is unpaused (echo 1 > tracing_enabled) + * @stop: called when tracing is paused (echo 0 > tracing_enabled) + * @open: called when the trace file is opened + * @pipe_open: called when the trace_pipe file is opened + * @wait_pipe: override how the user waits for traces on trace_pipe + * @close: called when the trace file is released + * @read: override the default read callback on trace_pipe + * @splice_read: override the default splice_read callback on trace_pipe + * @selftest: selftest to run on boot (see trace_selftest.c) + * @print_headers: override the first lines that describe your columns + * @print_line: callback that prints a trace + * @set_flag: signals one of your private flags changed (trace_options file) + * @flags: your private flags */ struct tracer { const char *name; - /* Your tracer should raise a warning if init fails */ int (*init)(struct trace_array *tr); void (*reset)(struct trace_array *tr); void (*start)(struct trace_array *tr); void (*stop)(struct trace_array *tr); void (*open)(struct trace_iterator *iter); void (*pipe_open)(struct trace_iterator *iter); + void (*wait_pipe)(struct trace_iterator *iter); void (*close)(struct trace_iterator *iter); ssize_t (*read)(struct trace_iterator *iter, struct file *filp, char __user *ubuf, size_t cnt, loff_t *ppos); + ssize_t (*splice_read)(struct trace_iterator *iter, + struct file *filp, + loff_t *ppos, + struct pipe_inode_info *pipe, + size_t len, + unsigned int flags); #ifdef CONFIG_FTRACE_STARTUP_TEST int (*selftest)(struct tracer *trace, struct trace_array *tr); @@ -339,51 +309,49 @@ struct tracer { int (*set_flag)(u32 old_flags, u32 bit, int set); struct tracer *next; int print_max; - struct tracer_flags *flags; + struct tracer_flags *flags; }; -struct trace_seq { - unsigned char buffer[PAGE_SIZE]; - unsigned int len; - unsigned int readpos; -}; -/* - * Trace iterator - used by printout routines who present trace - * results to users and which routines might sleep, etc: - */ -struct trace_iterator { - struct trace_array *tr; - struct tracer *trace; - void *private; - struct ring_buffer_iter *buffer_iter[NR_CPUS]; - - /* The below is zeroed out in pipe_read */ - struct trace_seq seq; - struct trace_entry *ent; - int cpu; - u64 ts; - - unsigned long iter_flags; - loff_t pos; - long idx; - - cpumask_var_t started; -}; +#define TRACE_PIPE_ALL_CPU -1 +int tracer_init(struct tracer *t, struct trace_array *tr); int tracing_is_enabled(void); void trace_wake_up(void); void tracing_reset(struct trace_array *tr, int cpu); void tracing_reset_online_cpus(struct trace_array *tr); +void tracing_reset_current(int cpu); +void tracing_reset_current_online_cpus(void); int tracing_open_generic(struct inode *inode, struct file *filp); +struct dentry *trace_create_file(const char *name, + mode_t mode, + struct dentry *parent, + void *data, + const struct file_operations *fops); + struct dentry *tracing_init_dentry(void); void init_tracer_sysprof_debugfs(struct dentry *d_tracer); +struct ring_buffer_event; + +struct ring_buffer_event * +trace_buffer_lock_reserve(struct ring_buffer *buffer, + int type, + unsigned long len, + unsigned long flags, + int pc); +void trace_buffer_unlock_commit(struct ring_buffer *buffer, + struct ring_buffer_event *event, + unsigned long flags, int pc); + struct trace_entry *tracing_get_trace_entry(struct trace_array *tr, struct trace_array_cpu *data); -void tracing_generic_entry_update(struct trace_entry *entry, - unsigned long flags, - int pc); + +struct trace_entry *trace_find_next_entry(struct trace_iterator *iter, + int *ent_cpu, u64 *ent_ts); + +void default_wait_pipe(struct trace_iterator *iter); +void poll_wait_pipe(struct trace_iterator *iter); void ftrace(struct trace_array *tr, struct trace_array_cpu *data, @@ -391,14 +359,11 @@ void ftrace(struct trace_array *tr, unsigned long parent_ip, unsigned long flags, int pc); void tracing_sched_switch_trace(struct trace_array *tr, - struct trace_array_cpu *data, struct task_struct *prev, struct task_struct *next, unsigned long flags, int pc); -void tracing_record_cmdline(struct task_struct *tsk); void tracing_sched_wakeup_trace(struct trace_array *tr, - struct trace_array_cpu *data, struct task_struct *wakee, struct task_struct *cur, unsigned long flags, int pc); @@ -408,14 +373,13 @@ void trace_special(struct trace_array *tr, unsigned long arg2, unsigned long arg3, int pc); void trace_function(struct trace_array *tr, - struct trace_array_cpu *data, unsigned long ip, unsigned long parent_ip, unsigned long flags, int pc); void trace_graph_return(struct ftrace_graph_ret *trace); int trace_graph_entry(struct ftrace_graph_ent *trace); -void trace_hw_branch(struct trace_array *tr, u64 from, u64 to); +void set_graph_array(struct trace_array *tr); void tracing_start_cmdline_record(void); void tracing_stop_cmdline_record(void); @@ -424,41 +388,50 @@ void tracing_stop_sched_switch_record(void); void tracing_start_sched_switch_record(void); int register_tracer(struct tracer *type); void unregister_tracer(struct tracer *type); +int is_tracing_stopped(void); + +extern int process_new_ksym_entry(char *ksymname, int op, unsigned long addr); extern unsigned long nsecs_to_usecs(unsigned long nsecs); +#ifdef CONFIG_TRACER_MAX_TRACE extern unsigned long tracing_max_latency; extern unsigned long tracing_thresh; void update_max_tr(struct trace_array *tr, struct task_struct *tsk, int cpu); void update_max_tr_single(struct trace_array *tr, struct task_struct *tsk, int cpu); +#endif /* CONFIG_TRACER_MAX_TRACE */ -extern cycle_t ftrace_now(int cpu); +#ifdef CONFIG_STACKTRACE +void ftrace_trace_stack(struct ring_buffer *buffer, unsigned long flags, + int skip, int pc); -#ifdef CONFIG_FUNCTION_TRACER -void tracing_start_function_trace(void); -void tracing_stop_function_trace(void); +void ftrace_trace_userstack(struct ring_buffer *buffer, unsigned long flags, + int pc); + +void __trace_stack(struct trace_array *tr, unsigned long flags, int skip, + int pc); #else -# define tracing_start_function_trace() do { } while (0) -# define tracing_stop_function_trace() do { } while (0) -#endif +static inline void ftrace_trace_stack(struct trace_array *tr, + unsigned long flags, int skip, int pc) +{ +} -#ifdef CONFIG_CONTEXT_SWITCH_TRACER -typedef void -(*tracer_switch_func_t)(void *private, - void *__rq, - struct task_struct *prev, - struct task_struct *next); - -struct tracer_switch_ops { - tracer_switch_func_t func; - void *private; - struct tracer_switch_ops *next; -}; +static inline void ftrace_trace_userstack(struct trace_array *tr, + unsigned long flags, int pc) +{ +} + +static inline void __trace_stack(struct trace_array *tr, unsigned long flags, + int skip, int pc) +{ +} +#endif /* CONFIG_STACKTRACE */ + +extern cycle_t ftrace_now(int cpu); -char *trace_find_cmdline(int pid); -#endif /* CONFIG_CONTEXT_SWITCH_TRACER */ +extern void trace_find_cmdline(int pid, char comm[]); #ifdef CONFIG_DYNAMIC_FTRACE extern unsigned long ftrace_update_tot_cnt; @@ -466,9 +439,15 @@ extern unsigned long ftrace_update_tot_cnt; extern int DYN_FTRACE_TEST_NAME(void); #endif +extern int ring_buffer_expanded; +extern bool tracing_selftest_disabled; +DECLARE_PER_CPU(local_t, ftrace_cpu_disabled); + #ifdef CONFIG_FTRACE_STARTUP_TEST extern int trace_selftest_startup_function(struct tracer *trace, struct trace_array *tr); +extern int trace_selftest_startup_function_graph(struct tracer *trace, + struct trace_array *tr); extern int trace_selftest_startup_irqsoff(struct tracer *trace, struct trace_array *tr); extern int trace_selftest_startup_preemptoff(struct tracer *trace, @@ -485,27 +464,33 @@ extern int trace_selftest_startup_sysprof(struct tracer *trace, struct trace_array *tr); extern int trace_selftest_startup_branch(struct tracer *trace, struct trace_array *tr); +extern int trace_selftest_startup_hw_branches(struct tracer *trace, + struct trace_array *tr); +extern int trace_selftest_startup_ksym(struct tracer *trace, + struct trace_array *tr); #endif /* CONFIG_FTRACE_STARTUP_TEST */ extern void *head_page(struct trace_array_cpu *data); -extern int trace_seq_printf(struct trace_seq *s, const char *fmt, ...); -extern void trace_seq_print_cont(struct trace_seq *s, - struct trace_iterator *iter); - +extern unsigned long long ns2usecs(cycle_t nsec); +extern int +trace_vbprintk(unsigned long ip, const char *fmt, va_list args); extern int -seq_print_ip_sym(struct trace_seq *s, unsigned long ip, - unsigned long sym_flags); -extern ssize_t trace_seq_to_user(struct trace_seq *s, char __user *ubuf, - size_t cnt); -extern long ns2usecs(cycle_t nsec); +trace_vprintk(unsigned long ip, const char *fmt, va_list args); extern int -trace_vprintk(unsigned long ip, int depth, const char *fmt, va_list args); +trace_array_vprintk(struct trace_array *tr, + unsigned long ip, const char *fmt, va_list args); +int trace_array_printk(struct trace_array *tr, + unsigned long ip, const char *fmt, ...); extern unsigned long trace_flags; +extern int trace_clock_id; + /* Standard output formatting function used for function return traces */ #ifdef CONFIG_FUNCTION_GRAPH_TRACER extern enum print_line_t print_graph_function(struct trace_iterator *iter); +extern enum print_line_t +trace_print_graph_duration(unsigned long long duration, struct trace_seq *s); #ifdef CONFIG_DYNAMIC_FTRACE /* TODO: make this variable */ @@ -528,16 +513,11 @@ static inline int ftrace_graph_addr(unsigned long addr) return 0; } #else -static inline int ftrace_trace_addr(unsigned long addr) -{ - return 1; -} static inline int ftrace_graph_addr(unsigned long addr) { return 1; } #endif /* CONFIG_DYNAMIC_FTRACE */ - #else /* CONFIG_FUNCTION_GRAPH_TRACER */ static inline enum print_line_t print_graph_function(struct trace_iterator *iter) @@ -546,15 +526,57 @@ print_graph_function(struct trace_iterator *iter) } #endif /* CONFIG_FUNCTION_GRAPH_TRACER */ -extern struct pid *ftrace_pid_trace; +extern struct list_head ftrace_pids; +#ifdef CONFIG_FUNCTION_TRACER static inline int ftrace_trace_task(struct task_struct *task) { - if (!ftrace_pid_trace) + if (list_empty(&ftrace_pids)) return 1; return test_tsk_trace_trace(task); } +#else +static inline int ftrace_trace_task(struct task_struct *task) +{ + return 1; +} +#endif + +/* + * struct trace_parser - servers for reading the user input separated by spaces + * @cont: set if the input is not complete - no final space char was found + * @buffer: holds the parsed user input + * @idx: user input lenght + * @size: buffer size + */ +struct trace_parser { + bool cont; + char *buffer; + unsigned idx; + unsigned size; +}; + +static inline bool trace_parser_loaded(struct trace_parser *parser) +{ + return (parser->idx != 0); +} + +static inline bool trace_parser_cont(struct trace_parser *parser) +{ + return parser->cont; +} + +static inline void trace_parser_clear(struct trace_parser *parser) +{ + parser->cont = false; + parser->idx = 0; +} + +extern int trace_parser_get_init(struct trace_parser *parser, int size); +extern void trace_parser_put(struct trace_parser *parser); +extern int trace_get_user(struct trace_parser *parser, const char __user *ubuf, + size_t cnt, loff_t *ppos); /* * trace_iterator_flags is an enumeration that defines bit @@ -580,7 +602,11 @@ enum trace_iterator_flags { TRACE_ITER_ANNOTATE = 0x2000, TRACE_ITER_USERSTACKTRACE = 0x4000, TRACE_ITER_SYM_USEROBJ = 0x8000, - TRACE_ITER_PRINTK_MSGONLY = 0x10000 + TRACE_ITER_PRINTK_MSGONLY = 0x10000, + TRACE_ITER_CONTEXT_INFO = 0x20000, /* Print pid/cpu/time */ + TRACE_ITER_LATENCY_FMT = 0x40000, + TRACE_ITER_SLEEP_TIME = 0x80000, + TRACE_ITER_GRAPH_TIME = 0x100000, }; /* @@ -601,12 +627,12 @@ extern struct tracer nop_trace; * preempt_enable (after a disable), a schedule might take place * causing an infinite recursion. * - * To prevent this, we read the need_recshed flag before + * To prevent this, we read the need_resched flag before * disabling preemption. When we want to enable preemption we * check the flag, if it is set, then we call preempt_enable_no_resched. * Otherwise, we call preempt_enable. * - * The rational for doing the above is that if need resched is set + * The rational for doing the above is that if need_resched is set * and we have yet to reschedule, we are either in an atomic location * (where we do not need to check for scheduling) or we are inside * the scheduler and do not want to resched. @@ -627,7 +653,7 @@ static inline int ftrace_preempt_disable(void) * * This is a scheduler safe way to enable preemption and not miss * any preemption checks. The disabled saved the state of preemption. - * If resched is set, then we were either inside an atomic or + * If resched is set, then we are either inside an atomic or * are inside the scheduler (we would have already scheduled * otherwise). In this case, we do not want to call normal * preempt_enable, but preempt_enable_no_resched instead. @@ -664,4 +690,110 @@ static inline void trace_branch_disable(void) } #endif /* CONFIG_BRANCH_TRACER */ +/* set ring buffers to default size if not already done so */ +int tracing_update_buffers(void); + +/* trace event type bit fields, not numeric */ +enum { + TRACE_EVENT_TYPE_PRINTF = 1, + TRACE_EVENT_TYPE_RAW = 2, +}; + +struct ftrace_event_field { + struct list_head link; + char *name; + char *type; + int filter_type; + int offset; + int size; + int is_signed; +}; + +struct event_filter { + int n_preds; + struct filter_pred **preds; + char *filter_string; +}; + +struct event_subsystem { + struct list_head list; + const char *name; + struct dentry *entry; + struct event_filter *filter; + int nr_events; +}; + +struct filter_pred; +struct regex; + +typedef int (*filter_pred_fn_t) (struct filter_pred *pred, void *event, + int val1, int val2); + +typedef int (*regex_match_func)(char *str, struct regex *r, int len); + +enum regex_type { + MATCH_FULL = 0, + MATCH_FRONT_ONLY, + MATCH_MIDDLE_ONLY, + MATCH_END_ONLY, +}; + +struct regex { + char pattern[MAX_FILTER_STR_VAL]; + int len; + int field_len; + regex_match_func match; +}; + +struct filter_pred { + filter_pred_fn_t fn; + u64 val; + struct regex regex; + char *field_name; + int offset; + int not; + int op; + int pop_n; +}; + +extern enum regex_type +filter_parse_regex(char *buff, int len, char **search, int *not); +extern void print_event_filter(struct ftrace_event_call *call, + struct trace_seq *s); +extern int apply_event_filter(struct ftrace_event_call *call, + char *filter_string); +extern int apply_subsystem_event_filter(struct event_subsystem *system, + char *filter_string); +extern void print_subsystem_event_filter(struct event_subsystem *system, + struct trace_seq *s); +extern int filter_assign_type(const char *type); + +static inline int +filter_check_discard(struct ftrace_event_call *call, void *rec, + struct ring_buffer *buffer, + struct ring_buffer_event *event) +{ + if (unlikely(call->filter_active) && + !filter_match_preds(call->filter, rec)) { + ring_buffer_discard_commit(buffer, event); + return 1; + } + + return 0; +} + +extern struct mutex event_mutex; +extern struct list_head ftrace_events; + +extern const char *__start___trace_bprintk_fmt[]; +extern const char *__stop___trace_bprintk_fmt[]; + +#undef FTRACE_ENTRY +#define FTRACE_ENTRY(call, struct_name, id, tstruct, print) \ + extern struct ftrace_event_call event_##call; +#undef FTRACE_ENTRY_DUP +#define FTRACE_ENTRY_DUP(call, struct_name, id, tstruct, print) \ + FTRACE_ENTRY(call, struct_name, id, PARAMS(tstruct), PARAMS(print)) +#include "trace_entries.h" + #endif /* _LINUX_KERNEL_TRACE_H */ diff --git a/kernel/trace/trace_boot.c b/kernel/trace/trace_boot.c index 366c8c333e1..c21d5f3956a 100644 --- a/kernel/trace/trace_boot.c +++ b/kernel/trace/trace_boot.c @@ -9,8 +9,10 @@ #include <linux/debugfs.h> #include <linux/ftrace.h> #include <linux/kallsyms.h> +#include <linux/time.h> #include "trace.h" +#include "trace_output.h" static struct trace_array *boot_trace; static bool pre_initcalls_finished; @@ -27,23 +29,24 @@ void start_boot_trace(void) void enable_boot_trace(void) { - if (pre_initcalls_finished) + if (boot_trace && pre_initcalls_finished) tracing_start_sched_switch_record(); } void disable_boot_trace(void) { - if (pre_initcalls_finished) + if (boot_trace && pre_initcalls_finished) tracing_stop_sched_switch_record(); } static int boot_trace_init(struct trace_array *tr) { - int cpu; boot_trace = tr; - for_each_cpu(cpu, cpu_possible_mask) - tracing_reset(tr, cpu); + if (!tr) + return 0; + + tracing_reset_online_cpus(tr); tracing_sched_switch_assign_trace(tr); return 0; @@ -63,7 +66,7 @@ initcall_call_print_line(struct trace_iterator *iter) trace_assign_type(field, entry); call = &field->boot_call; ts = iter->ts; - nsec_rem = do_div(ts, 1000000000); + nsec_rem = do_div(ts, NSEC_PER_SEC); ret = trace_seq_printf(s, "[%5ld.%09ld] calling %s @ %i\n", (unsigned long)ts, nsec_rem, call->func, call->caller); @@ -88,7 +91,7 @@ initcall_ret_print_line(struct trace_iterator *iter) trace_assign_type(field, entry); init_ret = &field->boot_ret; ts = iter->ts; - nsec_rem = do_div(ts, 1000000000); + nsec_rem = do_div(ts, NSEC_PER_SEC); ret = trace_seq_printf(s, "[%5ld.%09ld] initcall %s " "returned %d after %llu msecs\n", @@ -126,12 +129,13 @@ struct tracer boot_tracer __read_mostly = void trace_boot_call(struct boot_trace_call *bt, initcall_t fn) { + struct ftrace_event_call *call = &event_boot_call; struct ring_buffer_event *event; + struct ring_buffer *buffer; struct trace_boot_call *entry; - unsigned long irq_flags; struct trace_array *tr = boot_trace; - if (!pre_initcalls_finished) + if (!tr || !pre_initcalls_finished) return; /* Get its name now since this function could @@ -140,47 +144,42 @@ void trace_boot_call(struct boot_trace_call *bt, initcall_t fn) sprint_symbol(bt->func, (unsigned long)fn); preempt_disable(); - event = ring_buffer_lock_reserve(tr->buffer, sizeof(*entry), - &irq_flags); + buffer = tr->buffer; + event = trace_buffer_lock_reserve(buffer, TRACE_BOOT_CALL, + sizeof(*entry), 0, 0); if (!event) goto out; entry = ring_buffer_event_data(event); - tracing_generic_entry_update(&entry->ent, 0, 0); - entry->ent.type = TRACE_BOOT_CALL; entry->boot_call = *bt; - ring_buffer_unlock_commit(tr->buffer, event, irq_flags); - - trace_wake_up(); - + if (!filter_check_discard(call, entry, buffer, event)) + trace_buffer_unlock_commit(buffer, event, 0, 0); out: preempt_enable(); } void trace_boot_ret(struct boot_trace_ret *bt, initcall_t fn) { + struct ftrace_event_call *call = &event_boot_ret; struct ring_buffer_event *event; + struct ring_buffer *buffer; struct trace_boot_ret *entry; - unsigned long irq_flags; struct trace_array *tr = boot_trace; - if (!pre_initcalls_finished) + if (!tr || !pre_initcalls_finished) return; sprint_symbol(bt->func, (unsigned long)fn); preempt_disable(); - event = ring_buffer_lock_reserve(tr->buffer, sizeof(*entry), - &irq_flags); + buffer = tr->buffer; + event = trace_buffer_lock_reserve(buffer, TRACE_BOOT_RET, + sizeof(*entry), 0, 0); if (!event) goto out; entry = ring_buffer_event_data(event); - tracing_generic_entry_update(&entry->ent, 0, 0); - entry->ent.type = TRACE_BOOT_RET; entry->boot_ret = *bt; - ring_buffer_unlock_commit(tr->buffer, event, irq_flags); - - trace_wake_up(); - + if (!filter_check_discard(call, entry, buffer, event)) + trace_buffer_unlock_commit(buffer, event, 0, 0); out: preempt_enable(); } diff --git a/kernel/trace/trace_branch.c b/kernel/trace/trace_branch.c index 6c00feb3bac..4a194f08f88 100644 --- a/kernel/trace/trace_branch.c +++ b/kernel/trace/trace_branch.c @@ -14,21 +14,28 @@ #include <linux/hash.h> #include <linux/fs.h> #include <asm/local.h> + #include "trace.h" +#include "trace_stat.h" +#include "trace_output.h" #ifdef CONFIG_BRANCH_TRACER +static struct tracer branch_trace; static int branch_tracing_enabled __read_mostly; static DEFINE_MUTEX(branch_tracing_mutex); + static struct trace_array *branch_tracer; static void probe_likely_condition(struct ftrace_branch_data *f, int val, int expect) { + struct ftrace_event_call *call = &event_branch; struct trace_array *tr = branch_tracer; struct ring_buffer_event *event; struct trace_branch *entry; - unsigned long flags, irq_flags; + struct ring_buffer *buffer; + unsigned long flags; int cpu, pc; const char *p; @@ -47,15 +54,14 @@ probe_likely_condition(struct ftrace_branch_data *f, int val, int expect) if (atomic_inc_return(&tr->data[cpu]->disabled) != 1) goto out; - event = ring_buffer_lock_reserve(tr->buffer, sizeof(*entry), - &irq_flags); + pc = preempt_count(); + buffer = tr->buffer; + event = trace_buffer_lock_reserve(buffer, TRACE_BRANCH, + sizeof(*entry), flags, pc); if (!event) goto out; - pc = preempt_count(); entry = ring_buffer_event_data(event); - tracing_generic_entry_update(&entry->ent, flags, pc); - entry->ent.type = TRACE_BRANCH; /* Strip off the path, only save the file */ p = f->file + strlen(f->file); @@ -70,7 +76,8 @@ probe_likely_condition(struct ftrace_branch_data *f, int val, int expect) entry->line = f->line; entry->correct = val == expect; - ring_buffer_unlock_commit(tr->buffer, event, irq_flags); + if (!filter_check_discard(call, entry, buffer, event)) + ring_buffer_unlock_commit(buffer, event); out: atomic_dec(&tr->data[cpu]->disabled); @@ -88,8 +95,6 @@ void trace_likely_condition(struct ftrace_branch_data *f, int val, int expect) int enable_branch_tracing(struct trace_array *tr) { - int ret = 0; - mutex_lock(&branch_tracing_mutex); branch_tracer = tr; /* @@ -100,7 +105,7 @@ int enable_branch_tracing(struct trace_array *tr) branch_tracing_enabled++; mutex_unlock(&branch_tracing_mutex); - return ret; + return 0; } void disable_branch_tracing(void) @@ -128,11 +133,6 @@ static void stop_branch_trace(struct trace_array *tr) static int branch_trace_init(struct trace_array *tr) { - int cpu; - - for_each_online_cpu(cpu) - tracing_reset(tr, cpu); - start_branch_trace(tr); return 0; } @@ -142,22 +142,61 @@ static void branch_trace_reset(struct trace_array *tr) stop_branch_trace(tr); } -struct tracer branch_trace __read_mostly = +static enum print_line_t trace_branch_print(struct trace_iterator *iter, + int flags) +{ + struct trace_branch *field; + + trace_assign_type(field, iter->ent); + + if (trace_seq_printf(&iter->seq, "[%s] %s:%s:%d\n", + field->correct ? " ok " : " MISS ", + field->func, + field->file, + field->line)) + return TRACE_TYPE_PARTIAL_LINE; + + return TRACE_TYPE_HANDLED; +} + +static void branch_print_header(struct seq_file *s) +{ + seq_puts(s, "# TASK-PID CPU# TIMESTAMP CORRECT" + " FUNC:FILE:LINE\n"); + seq_puts(s, "# | | | | | " + " |\n"); +} + +static struct trace_event trace_branch_event = { + .type = TRACE_BRANCH, + .trace = trace_branch_print, +}; + +static struct tracer branch_trace __read_mostly = { .name = "branch", .init = branch_trace_init, .reset = branch_trace_reset, #ifdef CONFIG_FTRACE_SELFTEST .selftest = trace_selftest_startup_branch, -#endif +#endif /* CONFIG_FTRACE_SELFTEST */ + .print_header = branch_print_header, }; -__init static int init_branch_trace(void) +__init static int init_branch_tracer(void) { + int ret; + + ret = register_ftrace_event(&trace_branch_event); + if (!ret) { + printk(KERN_WARNING "Warning: could not register " + "branch events\n"); + return 1; + } return register_tracer(&branch_trace); } +device_initcall(init_branch_tracer); -device_initcall(init_branch_trace); #else static inline void trace_likely_condition(struct ftrace_branch_data *f, int val, int expect) @@ -183,66 +222,39 @@ void ftrace_likely_update(struct ftrace_branch_data *f, int val, int expect) } EXPORT_SYMBOL(ftrace_likely_update); -struct ftrace_pointer { - void *start; - void *stop; - int hit; -}; +extern unsigned long __start_annotated_branch_profile[]; +extern unsigned long __stop_annotated_branch_profile[]; -static void * -t_next(struct seq_file *m, void *v, loff_t *pos) +static int annotated_branch_stat_headers(struct seq_file *m) { - const struct ftrace_pointer *f = m->private; - struct ftrace_branch_data *p = v; - - (*pos)++; - - if (v == (void *)1) - return f->start; - - ++p; - - if ((void *)p >= (void *)f->stop) - return NULL; - - return p; + seq_printf(m, " correct incorrect %% "); + seq_printf(m, " Function " + " File Line\n" + " ------- --------- - " + " -------- " + " ---- ----\n"); + return 0; } -static void *t_start(struct seq_file *m, loff_t *pos) +static inline long get_incorrect_percent(struct ftrace_branch_data *p) { - void *t = (void *)1; - loff_t l = 0; - - for (; t && l < *pos; t = t_next(m, t, &l)) - ; + long percent; - return t; -} + if (p->correct) { + percent = p->incorrect * 100; + percent /= p->correct + p->incorrect; + } else + percent = p->incorrect ? 100 : -1; -static void t_stop(struct seq_file *m, void *p) -{ + return percent; } -static int t_show(struct seq_file *m, void *v) +static int branch_stat_show(struct seq_file *m, void *v) { - const struct ftrace_pointer *fp = m->private; struct ftrace_branch_data *p = v; const char *f; long percent; - if (v == (void *)1) { - if (fp->hit) - seq_printf(m, " miss hit %% "); - else - seq_printf(m, " correct incorrect %% "); - seq_printf(m, " Function " - " File Line\n" - " ------- --------- - " - " -------- " - " ---- ----\n"); - return 0; - } - /* Only print the file, not the path */ f = p->file + strlen(p->file); while (f >= p->file && *f != '/') @@ -252,11 +264,7 @@ static int t_show(struct seq_file *m, void *v) /* * The miss is overlayed on correct, and hit on incorrect. */ - if (p->correct) { - percent = p->incorrect * 100; - percent /= p->correct + p->incorrect; - } else - percent = p->incorrect ? 100 : -1; + percent = get_incorrect_percent(p); seq_printf(m, "%8lu %8lu ", p->correct, p->incorrect); if (percent < 0) @@ -267,76 +275,118 @@ static int t_show(struct seq_file *m, void *v) return 0; } -static struct seq_operations tracing_likely_seq_ops = { - .start = t_start, - .next = t_next, - .stop = t_stop, - .show = t_show, +static void *annotated_branch_stat_start(struct tracer_stat *trace) +{ + return __start_annotated_branch_profile; +} + +static void * +annotated_branch_stat_next(void *v, int idx) +{ + struct ftrace_branch_data *p = v; + + ++p; + + if ((void *)p >= (void *)__stop_annotated_branch_profile) + return NULL; + + return p; +} + +static int annotated_branch_stat_cmp(void *p1, void *p2) +{ + struct ftrace_branch_data *a = p1; + struct ftrace_branch_data *b = p2; + + long percent_a, percent_b; + + percent_a = get_incorrect_percent(a); + percent_b = get_incorrect_percent(b); + + if (percent_a < percent_b) + return -1; + if (percent_a > percent_b) + return 1; + else + return 0; +} + +static struct tracer_stat annotated_branch_stats = { + .name = "branch_annotated", + .stat_start = annotated_branch_stat_start, + .stat_next = annotated_branch_stat_next, + .stat_cmp = annotated_branch_stat_cmp, + .stat_headers = annotated_branch_stat_headers, + .stat_show = branch_stat_show }; -static int tracing_branch_open(struct inode *inode, struct file *file) +__init static int init_annotated_branch_stats(void) { int ret; - ret = seq_open(file, &tracing_likely_seq_ops); + ret = register_stat_tracer(&annotated_branch_stats); if (!ret) { - struct seq_file *m = file->private_data; - m->private = (void *)inode->i_private; + printk(KERN_WARNING "Warning: could not register " + "annotated branches stats\n"); + return 1; } - - return ret; + return 0; } - -static const struct file_operations tracing_branch_fops = { - .open = tracing_branch_open, - .read = seq_read, - .llseek = seq_lseek, -}; +fs_initcall(init_annotated_branch_stats); #ifdef CONFIG_PROFILE_ALL_BRANCHES + extern unsigned long __start_branch_profile[]; extern unsigned long __stop_branch_profile[]; -static const struct ftrace_pointer ftrace_branch_pos = { - .start = __start_branch_profile, - .stop = __stop_branch_profile, - .hit = 1, -}; +static int all_branch_stat_headers(struct seq_file *m) +{ + seq_printf(m, " miss hit %% "); + seq_printf(m, " Function " + " File Line\n" + " ------- --------- - " + " -------- " + " ---- ----\n"); + return 0; +} -#endif /* CONFIG_PROFILE_ALL_BRANCHES */ +static void *all_branch_stat_start(struct tracer_stat *trace) +{ + return __start_branch_profile; +} -extern unsigned long __start_annotated_branch_profile[]; -extern unsigned long __stop_annotated_branch_profile[]; +static void * +all_branch_stat_next(void *v, int idx) +{ + struct ftrace_branch_data *p = v; -static const struct ftrace_pointer ftrace_annotated_branch_pos = { - .start = __start_annotated_branch_profile, - .stop = __stop_annotated_branch_profile, -}; + ++p; -static __init int ftrace_branch_init(void) -{ - struct dentry *d_tracer; - struct dentry *entry; + if ((void *)p >= (void *)__stop_branch_profile) + return NULL; - d_tracer = tracing_init_dentry(); + return p; +} - entry = debugfs_create_file("profile_annotated_branch", 0444, d_tracer, - (void *)&ftrace_annotated_branch_pos, - &tracing_branch_fops); - if (!entry) - pr_warning("Could not create debugfs " - "'profile_annotatet_branch' entry\n"); +static struct tracer_stat all_branch_stats = { + .name = "branch_all", + .stat_start = all_branch_stat_start, + .stat_next = all_branch_stat_next, + .stat_headers = all_branch_stat_headers, + .stat_show = branch_stat_show +}; -#ifdef CONFIG_PROFILE_ALL_BRANCHES - entry = debugfs_create_file("profile_branch", 0444, d_tracer, - (void *)&ftrace_branch_pos, - &tracing_branch_fops); - if (!entry) - pr_warning("Could not create debugfs" - " 'profile_branch' entry\n"); -#endif +__init static int all_annotated_branch_stats(void) +{ + int ret; + ret = register_stat_tracer(&all_branch_stats); + if (!ret) { + printk(KERN_WARNING "Warning: could not register " + "all branches stats\n"); + return 1; + } return 0; } - -device_initcall(ftrace_branch_init); +fs_initcall(all_annotated_branch_stats); +#endif /* CONFIG_PROFILE_ALL_BRANCHES */ diff --git a/kernel/trace/trace_clock.c b/kernel/trace/trace_clock.c new file mode 100644 index 00000000000..878c03f386b --- /dev/null +++ b/kernel/trace/trace_clock.c @@ -0,0 +1,115 @@ +/* + * tracing clocks + * + * Copyright (C) 2009 Red Hat, Inc., Ingo Molnar <mingo@redhat.com> + * + * Implements 3 trace clock variants, with differing scalability/precision + * tradeoffs: + * + * - local: CPU-local trace clock + * - medium: scalable global clock with some jitter + * - global: globally monotonic, serialized clock + * + * Tracer plugins will chose a default from these clocks. + */ +#include <linux/spinlock.h> +#include <linux/hardirq.h> +#include <linux/module.h> +#include <linux/percpu.h> +#include <linux/sched.h> +#include <linux/ktime.h> +#include <linux/trace_clock.h> + +#include "trace.h" + +/* + * trace_clock_local(): the simplest and least coherent tracing clock. + * + * Useful for tracing that does not cross to other CPUs nor + * does it go through idle events. + */ +u64 notrace trace_clock_local(void) +{ + u64 clock; + int resched; + + /* + * sched_clock() is an architecture implemented, fast, scalable, + * lockless clock. It is not guaranteed to be coherent across + * CPUs, nor across CPU idle events. + */ + resched = ftrace_preempt_disable(); + clock = sched_clock(); + ftrace_preempt_enable(resched); + + return clock; +} + +/* + * trace_clock(): 'inbetween' trace clock. Not completely serialized, + * but not completely incorrect when crossing CPUs either. + * + * This is based on cpu_clock(), which will allow at most ~1 jiffy of + * jitter between CPUs. So it's a pretty scalable clock, but there + * can be offsets in the trace data. + */ +u64 notrace trace_clock(void) +{ + return cpu_clock(raw_smp_processor_id()); +} + + +/* + * trace_clock_global(): special globally coherent trace clock + * + * It has higher overhead than the other trace clocks but is still + * an order of magnitude faster than GTOD derived hardware clocks. + * + * Used by plugins that need globally coherent timestamps. + */ + +/* keep prev_time and lock in the same cacheline. */ +static struct { + u64 prev_time; + raw_spinlock_t lock; +} trace_clock_struct ____cacheline_aligned_in_smp = + { + .lock = (raw_spinlock_t)__RAW_SPIN_LOCK_UNLOCKED, + }; + +u64 notrace trace_clock_global(void) +{ + unsigned long flags; + int this_cpu; + u64 now; + + raw_local_irq_save(flags); + + this_cpu = raw_smp_processor_id(); + now = cpu_clock(this_cpu); + /* + * If in an NMI context then dont risk lockups and return the + * cpu_clock() time: + */ + if (unlikely(in_nmi())) + goto out; + + __raw_spin_lock(&trace_clock_struct.lock); + + /* + * TODO: if this happens often then maybe we should reset + * my_scd->clock to prev_time+1, to make sure + * we start ticking with the local clock from now on? + */ + if ((s64)(now - trace_clock_struct.prev_time) < 0) + now = trace_clock_struct.prev_time + 1; + + trace_clock_struct.prev_time = now; + + __raw_spin_unlock(&trace_clock_struct.lock); + + out: + raw_local_irq_restore(flags); + + return now; +} diff --git a/kernel/trace/trace_entries.h b/kernel/trace/trace_entries.h new file mode 100644 index 00000000000..c16a08f399d --- /dev/null +++ b/kernel/trace/trace_entries.h @@ -0,0 +1,382 @@ +/* + * This file defines the trace event structures that go into the ring + * buffer directly. They are created via macros so that changes for them + * appear in the format file. Using macros will automate this process. + * + * The macro used to create a ftrace data structure is: + * + * FTRACE_ENTRY( name, struct_name, id, structure, print ) + * + * @name: the name used the event name, as well as the name of + * the directory that holds the format file. + * + * @struct_name: the name of the structure that is created. + * + * @id: The event identifier that is used to detect what event + * this is from the ring buffer. + * + * @structure: the structure layout + * + * - __field( type, item ) + * This is equivalent to declaring + * type item; + * in the structure. + * - __array( type, item, size ) + * This is equivalent to declaring + * type item[size]; + * in the structure. + * + * * for structures within structures, the format of the internal + * structure is layed out. This allows the internal structure + * to be deciphered for the format file. Although these macros + * may become out of sync with the internal structure, they + * will create a compile error if it happens. Since the + * internel structures are just tracing helpers, this is not + * an issue. + * + * When an internal structure is used, it should use: + * + * __field_struct( type, item ) + * + * instead of __field. This will prevent it from being shown in + * the output file. The fields in the structure should use. + * + * __field_desc( type, container, item ) + * __array_desc( type, container, item, len ) + * + * type, item and len are the same as __field and __array, but + * container is added. This is the name of the item in + * __field_struct that this is describing. + * + * + * @print: the print format shown to users in the format file. + */ + +/* + * Function trace entry - function address and parent function addres: + */ +FTRACE_ENTRY(function, ftrace_entry, + + TRACE_FN, + + F_STRUCT( + __field( unsigned long, ip ) + __field( unsigned long, parent_ip ) + ), + + F_printk(" %lx <-- %lx", __entry->ip, __entry->parent_ip) +); + +/* Function call entry */ +FTRACE_ENTRY(funcgraph_entry, ftrace_graph_ent_entry, + + TRACE_GRAPH_ENT, + + F_STRUCT( + __field_struct( struct ftrace_graph_ent, graph_ent ) + __field_desc( unsigned long, graph_ent, func ) + __field_desc( int, graph_ent, depth ) + ), + + F_printk("--> %lx (%d)", __entry->func, __entry->depth) +); + +/* Function return entry */ +FTRACE_ENTRY(funcgraph_exit, ftrace_graph_ret_entry, + + TRACE_GRAPH_RET, + + F_STRUCT( + __field_struct( struct ftrace_graph_ret, ret ) + __field_desc( unsigned long, ret, func ) + __field_desc( unsigned long long, ret, calltime) + __field_desc( unsigned long long, ret, rettime ) + __field_desc( unsigned long, ret, overrun ) + __field_desc( int, ret, depth ) + ), + + F_printk("<-- %lx (%d) (start: %llx end: %llx) over: %d", + __entry->func, __entry->depth, + __entry->calltime, __entry->rettime, + __entry->depth) +); + +/* + * Context switch trace entry - which task (and prio) we switched from/to: + * + * This is used for both wakeup and context switches. We only want + * to create one structure, but we need two outputs for it. + */ +#define FTRACE_CTX_FIELDS \ + __field( unsigned int, prev_pid ) \ + __field( unsigned char, prev_prio ) \ + __field( unsigned char, prev_state ) \ + __field( unsigned int, next_pid ) \ + __field( unsigned char, next_prio ) \ + __field( unsigned char, next_state ) \ + __field( unsigned int, next_cpu ) + +FTRACE_ENTRY(context_switch, ctx_switch_entry, + + TRACE_CTX, + + F_STRUCT( + FTRACE_CTX_FIELDS + ), + + F_printk("%u:%u:%u ==> %u:%u:%u [%03u]", + __entry->prev_pid, __entry->prev_prio, __entry->prev_state, + __entry->next_pid, __entry->next_prio, __entry->next_state, + __entry->next_cpu + ) +); + +/* + * FTRACE_ENTRY_DUP only creates the format file, it will not + * create another structure. + */ +FTRACE_ENTRY_DUP(wakeup, ctx_switch_entry, + + TRACE_WAKE, + + F_STRUCT( + FTRACE_CTX_FIELDS + ), + + F_printk("%u:%u:%u ==+ %u:%u:%u [%03u]", + __entry->prev_pid, __entry->prev_prio, __entry->prev_state, + __entry->next_pid, __entry->next_prio, __entry->next_state, + __entry->next_cpu + ) +); + +/* + * Special (free-form) trace entry: + */ +FTRACE_ENTRY(special, special_entry, + + TRACE_SPECIAL, + + F_STRUCT( + __field( unsigned long, arg1 ) + __field( unsigned long, arg2 ) + __field( unsigned long, arg3 ) + ), + + F_printk("(%08lx) (%08lx) (%08lx)", + __entry->arg1, __entry->arg2, __entry->arg3) +); + +/* + * Stack-trace entry: + */ + +#define FTRACE_STACK_ENTRIES 8 + +FTRACE_ENTRY(kernel_stack, stack_entry, + + TRACE_STACK, + + F_STRUCT( + __array( unsigned long, caller, FTRACE_STACK_ENTRIES ) + ), + + F_printk("\t=> (%08lx)\n\t=> (%08lx)\n\t=> (%08lx)\n\t=> (%08lx)\n" + "\t=> (%08lx)\n\t=> (%08lx)\n\t=> (%08lx)\n\t=> (%08lx)\n", + __entry->caller[0], __entry->caller[1], __entry->caller[2], + __entry->caller[3], __entry->caller[4], __entry->caller[5], + __entry->caller[6], __entry->caller[7]) +); + +FTRACE_ENTRY(user_stack, userstack_entry, + + TRACE_USER_STACK, + + F_STRUCT( + __field( unsigned int, tgid ) + __array( unsigned long, caller, FTRACE_STACK_ENTRIES ) + ), + + F_printk("\t=> (%08lx)\n\t=> (%08lx)\n\t=> (%08lx)\n\t=> (%08lx)\n" + "\t=> (%08lx)\n\t=> (%08lx)\n\t=> (%08lx)\n\t=> (%08lx)\n", + __entry->caller[0], __entry->caller[1], __entry->caller[2], + __entry->caller[3], __entry->caller[4], __entry->caller[5], + __entry->caller[6], __entry->caller[7]) +); + +/* + * trace_printk entry: + */ +FTRACE_ENTRY(bprint, bprint_entry, + + TRACE_BPRINT, + + F_STRUCT( + __field( unsigned long, ip ) + __field( const char *, fmt ) + __dynamic_array( u32, buf ) + ), + + F_printk("%08lx fmt:%p", + __entry->ip, __entry->fmt) +); + +FTRACE_ENTRY(print, print_entry, + + TRACE_PRINT, + + F_STRUCT( + __field( unsigned long, ip ) + __dynamic_array( char, buf ) + ), + + F_printk("%08lx %s", + __entry->ip, __entry->buf) +); + +FTRACE_ENTRY(mmiotrace_rw, trace_mmiotrace_rw, + + TRACE_MMIO_RW, + + F_STRUCT( + __field_struct( struct mmiotrace_rw, rw ) + __field_desc( resource_size_t, rw, phys ) + __field_desc( unsigned long, rw, value ) + __field_desc( unsigned long, rw, pc ) + __field_desc( int, rw, map_id ) + __field_desc( unsigned char, rw, opcode ) + __field_desc( unsigned char, rw, width ) + ), + + F_printk("%lx %lx %lx %d %x %x", + (unsigned long)__entry->phys, __entry->value, __entry->pc, + __entry->map_id, __entry->opcode, __entry->width) +); + +FTRACE_ENTRY(mmiotrace_map, trace_mmiotrace_map, + + TRACE_MMIO_MAP, + + F_STRUCT( + __field_struct( struct mmiotrace_map, map ) + __field_desc( resource_size_t, map, phys ) + __field_desc( unsigned long, map, virt ) + __field_desc( unsigned long, map, len ) + __field_desc( int, map, map_id ) + __field_desc( unsigned char, map, opcode ) + ), + + F_printk("%lx %lx %lx %d %x", + (unsigned long)__entry->phys, __entry->virt, __entry->len, + __entry->map_id, __entry->opcode) +); + +FTRACE_ENTRY(boot_call, trace_boot_call, + + TRACE_BOOT_CALL, + + F_STRUCT( + __field_struct( struct boot_trace_call, boot_call ) + __field_desc( pid_t, boot_call, caller ) + __array_desc( char, boot_call, func, KSYM_SYMBOL_LEN) + ), + + F_printk("%d %s", __entry->caller, __entry->func) +); + +FTRACE_ENTRY(boot_ret, trace_boot_ret, + + TRACE_BOOT_RET, + + F_STRUCT( + __field_struct( struct boot_trace_ret, boot_ret ) + __array_desc( char, boot_ret, func, KSYM_SYMBOL_LEN) + __field_desc( int, boot_ret, result ) + __field_desc( unsigned long, boot_ret, duration ) + ), + + F_printk("%s %d %lx", + __entry->func, __entry->result, __entry->duration) +); + +#define TRACE_FUNC_SIZE 30 +#define TRACE_FILE_SIZE 20 + +FTRACE_ENTRY(branch, trace_branch, + + TRACE_BRANCH, + + F_STRUCT( + __field( unsigned int, line ) + __array( char, func, TRACE_FUNC_SIZE+1 ) + __array( char, file, TRACE_FILE_SIZE+1 ) + __field( char, correct ) + ), + + F_printk("%u:%s:%s (%u)", + __entry->line, + __entry->func, __entry->file, __entry->correct) +); + +FTRACE_ENTRY(hw_branch, hw_branch_entry, + + TRACE_HW_BRANCHES, + + F_STRUCT( + __field( u64, from ) + __field( u64, to ) + ), + + F_printk("from: %llx to: %llx", __entry->from, __entry->to) +); + +FTRACE_ENTRY(kmem_alloc, kmemtrace_alloc_entry, + + TRACE_KMEM_ALLOC, + + F_STRUCT( + __field( enum kmemtrace_type_id, type_id ) + __field( unsigned long, call_site ) + __field( const void *, ptr ) + __field( size_t, bytes_req ) + __field( size_t, bytes_alloc ) + __field( gfp_t, gfp_flags ) + __field( int, node ) + ), + + F_printk("type:%u call_site:%lx ptr:%p req:%zi alloc:%zi" + " flags:%x node:%d", + __entry->type_id, __entry->call_site, __entry->ptr, + __entry->bytes_req, __entry->bytes_alloc, + __entry->gfp_flags, __entry->node) +); + +FTRACE_ENTRY(kmem_free, kmemtrace_free_entry, + + TRACE_KMEM_FREE, + + F_STRUCT( + __field( enum kmemtrace_type_id, type_id ) + __field( unsigned long, call_site ) + __field( const void *, ptr ) + ), + + F_printk("type:%u call_site:%lx ptr:%p", + __entry->type_id, __entry->call_site, __entry->ptr) +); + +FTRACE_ENTRY(ksym_trace, ksym_trace_entry, + + TRACE_KSYM, + + F_STRUCT( + __field( unsigned long, ip ) + __field( unsigned char, type ) + __array( char , cmd, TASK_COMM_LEN ) + __field( unsigned long, addr ) + ), + + F_printk("ip: %pF type: %d ksym_name: %pS cmd: %s", + (void *)__entry->ip, (unsigned int)__entry->type, + (void *)__entry->addr, __entry->cmd) +); diff --git a/kernel/trace/trace_event_profile.c b/kernel/trace/trace_event_profile.c new file mode 100644 index 00000000000..d9c60f80aa0 --- /dev/null +++ b/kernel/trace/trace_event_profile.c @@ -0,0 +1,122 @@ +/* + * trace event based perf counter profiling + * + * Copyright (C) 2009 Red Hat Inc, Peter Zijlstra <pzijlstr@redhat.com> + * + */ + +#include <linux/module.h> +#include "trace.h" + + +char *perf_trace_buf; +EXPORT_SYMBOL_GPL(perf_trace_buf); + +char *perf_trace_buf_nmi; +EXPORT_SYMBOL_GPL(perf_trace_buf_nmi); + +typedef typeof(char [FTRACE_MAX_PROFILE_SIZE]) perf_trace_t ; + +/* Count the events in use (per event id, not per instance) */ +static int total_profile_count; + +static int ftrace_profile_enable_event(struct ftrace_event_call *event) +{ + char *buf; + int ret = -ENOMEM; + + if (atomic_inc_return(&event->profile_count)) + return 0; + + if (!total_profile_count) { + buf = (char *)alloc_percpu(perf_trace_t); + if (!buf) + goto fail_buf; + + rcu_assign_pointer(perf_trace_buf, buf); + + buf = (char *)alloc_percpu(perf_trace_t); + if (!buf) + goto fail_buf_nmi; + + rcu_assign_pointer(perf_trace_buf_nmi, buf); + } + + ret = event->profile_enable(event); + if (!ret) { + total_profile_count++; + return 0; + } + +fail_buf_nmi: + if (!total_profile_count) { + free_percpu(perf_trace_buf_nmi); + free_percpu(perf_trace_buf); + perf_trace_buf_nmi = NULL; + perf_trace_buf = NULL; + } +fail_buf: + atomic_dec(&event->profile_count); + + return ret; +} + +int ftrace_profile_enable(int event_id) +{ + struct ftrace_event_call *event; + int ret = -EINVAL; + + mutex_lock(&event_mutex); + list_for_each_entry(event, &ftrace_events, list) { + if (event->id == event_id && event->profile_enable && + try_module_get(event->mod)) { + ret = ftrace_profile_enable_event(event); + break; + } + } + mutex_unlock(&event_mutex); + + return ret; +} + +static void ftrace_profile_disable_event(struct ftrace_event_call *event) +{ + char *buf, *nmi_buf; + + if (!atomic_add_negative(-1, &event->profile_count)) + return; + + event->profile_disable(event); + + if (!--total_profile_count) { + buf = perf_trace_buf; + rcu_assign_pointer(perf_trace_buf, NULL); + + nmi_buf = perf_trace_buf_nmi; + rcu_assign_pointer(perf_trace_buf_nmi, NULL); + + /* + * Ensure every events in profiling have finished before + * releasing the buffers + */ + synchronize_sched(); + + free_percpu(buf); + free_percpu(nmi_buf); + } +} + +void ftrace_profile_disable(int event_id) +{ + struct ftrace_event_call *event; + + mutex_lock(&event_mutex); + list_for_each_entry(event, &ftrace_events, list) { + if (event->id == event_id) { + ftrace_profile_disable_event(event); + module_put(event->mod); + break; + } + } + mutex_unlock(&event_mutex); +} diff --git a/kernel/trace/trace_events.c b/kernel/trace/trace_events.c new file mode 100644 index 00000000000..1d18315dc83 --- /dev/null +++ b/kernel/trace/trace_events.c @@ -0,0 +1,1521 @@ +/* + * event tracer + * + * Copyright (C) 2008 Red Hat Inc, Steven Rostedt <srostedt@redhat.com> + * + * - Added format output of fields of the trace point. + * This was based off of work by Tom Zanussi <tzanussi@gmail.com>. + * + */ + +#include <linux/workqueue.h> +#include <linux/spinlock.h> +#include <linux/kthread.h> +#include <linux/debugfs.h> +#include <linux/uaccess.h> +#include <linux/module.h> +#include <linux/ctype.h> +#include <linux/delay.h> + +#include <asm/setup.h> + +#include "trace_output.h" + +#undef TRACE_SYSTEM +#define TRACE_SYSTEM "TRACE_SYSTEM" + +DEFINE_MUTEX(event_mutex); + +LIST_HEAD(ftrace_events); + +int trace_define_field(struct ftrace_event_call *call, const char *type, + const char *name, int offset, int size, int is_signed, + int filter_type) +{ + struct ftrace_event_field *field; + + field = kzalloc(sizeof(*field), GFP_KERNEL); + if (!field) + goto err; + + field->name = kstrdup(name, GFP_KERNEL); + if (!field->name) + goto err; + + field->type = kstrdup(type, GFP_KERNEL); + if (!field->type) + goto err; + + if (filter_type == FILTER_OTHER) + field->filter_type = filter_assign_type(type); + else + field->filter_type = filter_type; + + field->offset = offset; + field->size = size; + field->is_signed = is_signed; + + list_add(&field->link, &call->fields); + + return 0; + +err: + if (field) { + kfree(field->name); + kfree(field->type); + } + kfree(field); + + return -ENOMEM; +} +EXPORT_SYMBOL_GPL(trace_define_field); + +#define __common_field(type, item) \ + ret = trace_define_field(call, #type, "common_" #item, \ + offsetof(typeof(ent), item), \ + sizeof(ent.item), \ + is_signed_type(type), FILTER_OTHER); \ + if (ret) \ + return ret; + +int trace_define_common_fields(struct ftrace_event_call *call) +{ + int ret; + struct trace_entry ent; + + __common_field(unsigned short, type); + __common_field(unsigned char, flags); + __common_field(unsigned char, preempt_count); + __common_field(int, pid); + __common_field(int, lock_depth); + + return ret; +} +EXPORT_SYMBOL_GPL(trace_define_common_fields); + +void trace_destroy_fields(struct ftrace_event_call *call) +{ + struct ftrace_event_field *field, *next; + + list_for_each_entry_safe(field, next, &call->fields, link) { + list_del(&field->link); + kfree(field->type); + kfree(field->name); + kfree(field); + } +} + +static void ftrace_event_enable_disable(struct ftrace_event_call *call, + int enable) +{ + switch (enable) { + case 0: + if (call->enabled) { + call->enabled = 0; + tracing_stop_cmdline_record(); + call->unregfunc(call); + } + break; + case 1: + if (!call->enabled) { + call->enabled = 1; + tracing_start_cmdline_record(); + call->regfunc(call); + } + break; + } +} + +static void ftrace_clear_events(void) +{ + struct ftrace_event_call *call; + + mutex_lock(&event_mutex); + list_for_each_entry(call, &ftrace_events, list) { + ftrace_event_enable_disable(call, 0); + } + mutex_unlock(&event_mutex); +} + +/* + * __ftrace_set_clr_event(NULL, NULL, NULL, set) will set/unset all events. + */ +static int __ftrace_set_clr_event(const char *match, const char *sub, + const char *event, int set) +{ + struct ftrace_event_call *call; + int ret = -EINVAL; + + mutex_lock(&event_mutex); + list_for_each_entry(call, &ftrace_events, list) { + + if (!call->name || !call->regfunc) + continue; + + if (match && + strcmp(match, call->name) != 0 && + strcmp(match, call->system) != 0) + continue; + + if (sub && strcmp(sub, call->system) != 0) + continue; + + if (event && strcmp(event, call->name) != 0) + continue; + + ftrace_event_enable_disable(call, set); + + ret = 0; + } + mutex_unlock(&event_mutex); + + return ret; +} + +static int ftrace_set_clr_event(char *buf, int set) +{ + char *event = NULL, *sub = NULL, *match; + + /* + * The buf format can be <subsystem>:<event-name> + * *:<event-name> means any event by that name. + * :<event-name> is the same. + * + * <subsystem>:* means all events in that subsystem + * <subsystem>: means the same. + * + * <name> (no ':') means all events in a subsystem with + * the name <name> or any event that matches <name> + */ + + match = strsep(&buf, ":"); + if (buf) { + sub = match; + event = buf; + match = NULL; + + if (!strlen(sub) || strcmp(sub, "*") == 0) + sub = NULL; + if (!strlen(event) || strcmp(event, "*") == 0) + event = NULL; + } + + return __ftrace_set_clr_event(match, sub, event, set); +} + +/** + * trace_set_clr_event - enable or disable an event + * @system: system name to match (NULL for any system) + * @event: event name to match (NULL for all events, within system) + * @set: 1 to enable, 0 to disable + * + * This is a way for other parts of the kernel to enable or disable + * event recording. + * + * Returns 0 on success, -EINVAL if the parameters do not match any + * registered events. + */ +int trace_set_clr_event(const char *system, const char *event, int set) +{ + return __ftrace_set_clr_event(NULL, system, event, set); +} + +/* 128 should be much more than enough */ +#define EVENT_BUF_SIZE 127 + +static ssize_t +ftrace_event_write(struct file *file, const char __user *ubuf, + size_t cnt, loff_t *ppos) +{ + struct trace_parser parser; + ssize_t read, ret; + + if (!cnt) + return 0; + + ret = tracing_update_buffers(); + if (ret < 0) + return ret; + + if (trace_parser_get_init(&parser, EVENT_BUF_SIZE + 1)) + return -ENOMEM; + + read = trace_get_user(&parser, ubuf, cnt, ppos); + + if (read >= 0 && trace_parser_loaded((&parser))) { + int set = 1; + + if (*parser.buffer == '!') + set = 0; + + parser.buffer[parser.idx] = 0; + + ret = ftrace_set_clr_event(parser.buffer + !set, set); + if (ret) + goto out_put; + } + + ret = read; + + out_put: + trace_parser_put(&parser); + + return ret; +} + +static void * +t_next(struct seq_file *m, void *v, loff_t *pos) +{ + struct ftrace_event_call *call = v; + + (*pos)++; + + list_for_each_entry_continue(call, &ftrace_events, list) { + /* + * The ftrace subsystem is for showing formats only. + * They can not be enabled or disabled via the event files. + */ + if (call->regfunc) + return call; + } + + return NULL; +} + +static void *t_start(struct seq_file *m, loff_t *pos) +{ + struct ftrace_event_call *call; + loff_t l; + + mutex_lock(&event_mutex); + + call = list_entry(&ftrace_events, struct ftrace_event_call, list); + for (l = 0; l <= *pos; ) { + call = t_next(m, call, &l); + if (!call) + break; + } + return call; +} + +static void * +s_next(struct seq_file *m, void *v, loff_t *pos) +{ + struct ftrace_event_call *call = v; + + (*pos)++; + + list_for_each_entry_continue(call, &ftrace_events, list) { + if (call->enabled) + return call; + } + + return NULL; +} + +static void *s_start(struct seq_file *m, loff_t *pos) +{ + struct ftrace_event_call *call; + loff_t l; + + mutex_lock(&event_mutex); + + call = list_entry(&ftrace_events, struct ftrace_event_call, list); + for (l = 0; l <= *pos; ) { + call = s_next(m, call, &l); + if (!call) + break; + } + return call; +} + +static int t_show(struct seq_file *m, void *v) +{ + struct ftrace_event_call *call = v; + + if (strcmp(call->system, TRACE_SYSTEM) != 0) + seq_printf(m, "%s:", call->system); + seq_printf(m, "%s\n", call->name); + + return 0; +} + +static void t_stop(struct seq_file *m, void *p) +{ + mutex_unlock(&event_mutex); +} + +static int +ftrace_event_seq_open(struct inode *inode, struct file *file) +{ + const struct seq_operations *seq_ops; + + if ((file->f_mode & FMODE_WRITE) && + (file->f_flags & O_TRUNC)) + ftrace_clear_events(); + + seq_ops = inode->i_private; + return seq_open(file, seq_ops); +} + +static ssize_t +event_enable_read(struct file *filp, char __user *ubuf, size_t cnt, + loff_t *ppos) +{ + struct ftrace_event_call *call = filp->private_data; + char *buf; + + if (call->enabled) + buf = "1\n"; + else + buf = "0\n"; + + return simple_read_from_buffer(ubuf, cnt, ppos, buf, 2); +} + +static ssize_t +event_enable_write(struct file *filp, const char __user *ubuf, size_t cnt, + loff_t *ppos) +{ + struct ftrace_event_call *call = filp->private_data; + char buf[64]; + unsigned long val; + int ret; + + if (cnt >= sizeof(buf)) + return -EINVAL; + + if (copy_from_user(&buf, ubuf, cnt)) + return -EFAULT; + + buf[cnt] = 0; + + ret = strict_strtoul(buf, 10, &val); + if (ret < 0) + return ret; + + ret = tracing_update_buffers(); + if (ret < 0) + return ret; + + switch (val) { + case 0: + case 1: + mutex_lock(&event_mutex); + ftrace_event_enable_disable(call, val); + mutex_unlock(&event_mutex); + break; + + default: + return -EINVAL; + } + + *ppos += cnt; + + return cnt; +} + +static ssize_t +system_enable_read(struct file *filp, char __user *ubuf, size_t cnt, + loff_t *ppos) +{ + const char set_to_char[4] = { '?', '0', '1', 'X' }; + const char *system = filp->private_data; + struct ftrace_event_call *call; + char buf[2]; + int set = 0; + int ret; + + mutex_lock(&event_mutex); + list_for_each_entry(call, &ftrace_events, list) { + if (!call->name || !call->regfunc) + continue; + + if (system && strcmp(call->system, system) != 0) + continue; + + /* + * We need to find out if all the events are set + * or if all events or cleared, or if we have + * a mixture. + */ + set |= (1 << !!call->enabled); + + /* + * If we have a mixture, no need to look further. + */ + if (set == 3) + break; + } + mutex_unlock(&event_mutex); + + buf[0] = set_to_char[set]; + buf[1] = '\n'; + + ret = simple_read_from_buffer(ubuf, cnt, ppos, buf, 2); + + return ret; +} + +static ssize_t +system_enable_write(struct file *filp, const char __user *ubuf, size_t cnt, + loff_t *ppos) +{ + const char *system = filp->private_data; + unsigned long val; + char buf[64]; + ssize_t ret; + + if (cnt >= sizeof(buf)) + return -EINVAL; + + if (copy_from_user(&buf, ubuf, cnt)) + return -EFAULT; + + buf[cnt] = 0; + + ret = strict_strtoul(buf, 10, &val); + if (ret < 0) + return ret; + + ret = tracing_update_buffers(); + if (ret < 0) + return ret; + + if (val != 0 && val != 1) + return -EINVAL; + + ret = __ftrace_set_clr_event(NULL, system, NULL, val); + if (ret) + goto out; + + ret = cnt; + +out: + *ppos += cnt; + + return ret; +} + +extern char *__bad_type_size(void); + +#undef FIELD +#define FIELD(type, name) \ + sizeof(type) != sizeof(field.name) ? __bad_type_size() : \ + #type, "common_" #name, offsetof(typeof(field), name), \ + sizeof(field.name), is_signed_type(type) + +static int trace_write_header(struct trace_seq *s) +{ + struct trace_entry field; + + /* struct trace_entry */ + return trace_seq_printf(s, + "\tfield:%s %s;\toffset:%zu;\tsize:%zu;\tsigned:%u;\n" + "\tfield:%s %s;\toffset:%zu;\tsize:%zu;\tsigned:%u;\n" + "\tfield:%s %s;\toffset:%zu;\tsize:%zu;\tsigned:%u;\n" + "\tfield:%s %s;\toffset:%zu;\tsize:%zu;\tsigned:%u;\n" + "\tfield:%s %s;\toffset:%zu;\tsize:%zu;\tsigned:%u;\n" + "\n", + FIELD(unsigned short, type), + FIELD(unsigned char, flags), + FIELD(unsigned char, preempt_count), + FIELD(int, pid), + FIELD(int, lock_depth)); +} + +static ssize_t +event_format_read(struct file *filp, char __user *ubuf, size_t cnt, + loff_t *ppos) +{ + struct ftrace_event_call *call = filp->private_data; + struct trace_seq *s; + char *buf; + int r; + + if (*ppos) + return 0; + + s = kmalloc(sizeof(*s), GFP_KERNEL); + if (!s) + return -ENOMEM; + + trace_seq_init(s); + + /* If any of the first writes fail, so will the show_format. */ + + trace_seq_printf(s, "name: %s\n", call->name); + trace_seq_printf(s, "ID: %d\n", call->id); + trace_seq_printf(s, "format:\n"); + trace_write_header(s); + + r = call->show_format(call, s); + if (!r) { + /* + * ug! The format output is bigger than a PAGE!! + */ + buf = "FORMAT TOO BIG\n"; + r = simple_read_from_buffer(ubuf, cnt, ppos, + buf, strlen(buf)); + goto out; + } + + r = simple_read_from_buffer(ubuf, cnt, ppos, + s->buffer, s->len); + out: + kfree(s); + return r; +} + +static ssize_t +event_id_read(struct file *filp, char __user *ubuf, size_t cnt, loff_t *ppos) +{ + struct ftrace_event_call *call = filp->private_data; + struct trace_seq *s; + int r; + + if (*ppos) + return 0; + + s = kmalloc(sizeof(*s), GFP_KERNEL); + if (!s) + return -ENOMEM; + + trace_seq_init(s); + trace_seq_printf(s, "%d\n", call->id); + + r = simple_read_from_buffer(ubuf, cnt, ppos, + s->buffer, s->len); + kfree(s); + return r; +} + +static ssize_t +event_filter_read(struct file *filp, char __user *ubuf, size_t cnt, + loff_t *ppos) +{ + struct ftrace_event_call *call = filp->private_data; + struct trace_seq *s; + int r; + + if (*ppos) + return 0; + + s = kmalloc(sizeof(*s), GFP_KERNEL); + if (!s) + return -ENOMEM; + + trace_seq_init(s); + + print_event_filter(call, s); + r = simple_read_from_buffer(ubuf, cnt, ppos, s->buffer, s->len); + + kfree(s); + + return r; +} + +static ssize_t +event_filter_write(struct file *filp, const char __user *ubuf, size_t cnt, + loff_t *ppos) +{ + struct ftrace_event_call *call = filp->private_data; + char *buf; + int err; + + if (cnt >= PAGE_SIZE) + return -EINVAL; + + buf = (char *)__get_free_page(GFP_TEMPORARY); + if (!buf) + return -ENOMEM; + + if (copy_from_user(buf, ubuf, cnt)) { + free_page((unsigned long) buf); + return -EFAULT; + } + buf[cnt] = '\0'; + + err = apply_event_filter(call, buf); + free_page((unsigned long) buf); + if (err < 0) + return err; + + *ppos += cnt; + + return cnt; +} + +static ssize_t +subsystem_filter_read(struct file *filp, char __user *ubuf, size_t cnt, + loff_t *ppos) +{ + struct event_subsystem *system = filp->private_data; + struct trace_seq *s; + int r; + + if (*ppos) + return 0; + + s = kmalloc(sizeof(*s), GFP_KERNEL); + if (!s) + return -ENOMEM; + + trace_seq_init(s); + + print_subsystem_event_filter(system, s); + r = simple_read_from_buffer(ubuf, cnt, ppos, s->buffer, s->len); + + kfree(s); + + return r; +} + +static ssize_t +subsystem_filter_write(struct file *filp, const char __user *ubuf, size_t cnt, + loff_t *ppos) +{ + struct event_subsystem *system = filp->private_data; + char *buf; + int err; + + if (cnt >= PAGE_SIZE) + return -EINVAL; + + buf = (char *)__get_free_page(GFP_TEMPORARY); + if (!buf) + return -ENOMEM; + + if (copy_from_user(buf, ubuf, cnt)) { + free_page((unsigned long) buf); + return -EFAULT; + } + buf[cnt] = '\0'; + + err = apply_subsystem_event_filter(system, buf); + free_page((unsigned long) buf); + if (err < 0) + return err; + + *ppos += cnt; + + return cnt; +} + +static ssize_t +show_header(struct file *filp, char __user *ubuf, size_t cnt, loff_t *ppos) +{ + int (*func)(struct trace_seq *s) = filp->private_data; + struct trace_seq *s; + int r; + + if (*ppos) + return 0; + + s = kmalloc(sizeof(*s), GFP_KERNEL); + if (!s) + return -ENOMEM; + + trace_seq_init(s); + + func(s); + r = simple_read_from_buffer(ubuf, cnt, ppos, s->buffer, s->len); + + kfree(s); + + return r; +} + +static const struct seq_operations show_event_seq_ops = { + .start = t_start, + .next = t_next, + .show = t_show, + .stop = t_stop, +}; + +static const struct seq_operations show_set_event_seq_ops = { + .start = s_start, + .next = s_next, + .show = t_show, + .stop = t_stop, +}; + +static const struct file_operations ftrace_avail_fops = { + .open = ftrace_event_seq_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release, +}; + +static const struct file_operations ftrace_set_event_fops = { + .open = ftrace_event_seq_open, + .read = seq_read, + .write = ftrace_event_write, + .llseek = seq_lseek, + .release = seq_release, +}; + +static const struct file_operations ftrace_enable_fops = { + .open = tracing_open_generic, + .read = event_enable_read, + .write = event_enable_write, +}; + +static const struct file_operations ftrace_event_format_fops = { + .open = tracing_open_generic, + .read = event_format_read, +}; + +static const struct file_operations ftrace_event_id_fops = { + .open = tracing_open_generic, + .read = event_id_read, +}; + +static const struct file_operations ftrace_event_filter_fops = { + .open = tracing_open_generic, + .read = event_filter_read, + .write = event_filter_write, +}; + +static const struct file_operations ftrace_subsystem_filter_fops = { + .open = tracing_open_generic, + .read = subsystem_filter_read, + .write = subsystem_filter_write, +}; + +static const struct file_operations ftrace_system_enable_fops = { + .open = tracing_open_generic, + .read = system_enable_read, + .write = system_enable_write, +}; + +static const struct file_operations ftrace_show_header_fops = { + .open = tracing_open_generic, + .read = show_header, +}; + +static struct dentry *event_trace_events_dir(void) +{ + static struct dentry *d_tracer; + static struct dentry *d_events; + + if (d_events) + return d_events; + + d_tracer = tracing_init_dentry(); + if (!d_tracer) + return NULL; + + d_events = debugfs_create_dir("events", d_tracer); + if (!d_events) + pr_warning("Could not create debugfs " + "'events' directory\n"); + + return d_events; +} + +static LIST_HEAD(event_subsystems); + +static struct dentry * +event_subsystem_dir(const char *name, struct dentry *d_events) +{ + struct event_subsystem *system; + struct dentry *entry; + + /* First see if we did not already create this dir */ + list_for_each_entry(system, &event_subsystems, list) { + if (strcmp(system->name, name) == 0) { + system->nr_events++; + return system->entry; + } + } + + /* need to create new entry */ + system = kmalloc(sizeof(*system), GFP_KERNEL); + if (!system) { + pr_warning("No memory to create event subsystem %s\n", + name); + return d_events; + } + + system->entry = debugfs_create_dir(name, d_events); + if (!system->entry) { + pr_warning("Could not create event subsystem %s\n", + name); + kfree(system); + return d_events; + } + + system->nr_events = 1; + system->name = kstrdup(name, GFP_KERNEL); + if (!system->name) { + debugfs_remove(system->entry); + kfree(system); + return d_events; + } + + list_add(&system->list, &event_subsystems); + + system->filter = NULL; + + system->filter = kzalloc(sizeof(struct event_filter), GFP_KERNEL); + if (!system->filter) { + pr_warning("Could not allocate filter for subsystem " + "'%s'\n", name); + return system->entry; + } + + entry = debugfs_create_file("filter", 0644, system->entry, system, + &ftrace_subsystem_filter_fops); + if (!entry) { + kfree(system->filter); + system->filter = NULL; + pr_warning("Could not create debugfs " + "'%s/filter' entry\n", name); + } + + trace_create_file("enable", 0644, system->entry, + (void *)system->name, + &ftrace_system_enable_fops); + + return system->entry; +} + +static int +event_create_dir(struct ftrace_event_call *call, struct dentry *d_events, + const struct file_operations *id, + const struct file_operations *enable, + const struct file_operations *filter, + const struct file_operations *format) +{ + int ret; + + /* + * If the trace point header did not define TRACE_SYSTEM + * then the system would be called "TRACE_SYSTEM". + */ + if (strcmp(call->system, TRACE_SYSTEM) != 0) + d_events = event_subsystem_dir(call->system, d_events); + + call->dir = debugfs_create_dir(call->name, d_events); + if (!call->dir) { + pr_warning("Could not create debugfs " + "'%s' directory\n", call->name); + return -1; + } + + if (call->regfunc) + trace_create_file("enable", 0644, call->dir, call, + enable); + + if (call->id && call->profile_enable) + trace_create_file("id", 0444, call->dir, call, + id); + + if (call->define_fields) { + ret = call->define_fields(call); + if (ret < 0) { + pr_warning("Could not initialize trace point" + " events/%s\n", call->name); + return ret; + } + trace_create_file("filter", 0644, call->dir, call, + filter); + } + + /* A trace may not want to export its format */ + if (!call->show_format) + return 0; + + trace_create_file("format", 0444, call->dir, call, + format); + + return 0; +} + +static int __trace_add_event_call(struct ftrace_event_call *call) +{ + struct dentry *d_events; + int ret; + + if (!call->name) + return -EINVAL; + + if (call->raw_init) { + ret = call->raw_init(call); + if (ret < 0) { + if (ret != -ENOSYS) + pr_warning("Could not initialize trace " + "events/%s\n", call->name); + return ret; + } + } + + d_events = event_trace_events_dir(); + if (!d_events) + return -ENOENT; + + ret = event_create_dir(call, d_events, &ftrace_event_id_fops, + &ftrace_enable_fops, &ftrace_event_filter_fops, + &ftrace_event_format_fops); + if (!ret) + list_add(&call->list, &ftrace_events); + + return ret; +} + +/* Add an additional event_call dynamically */ +int trace_add_event_call(struct ftrace_event_call *call) +{ + int ret; + mutex_lock(&event_mutex); + ret = __trace_add_event_call(call); + mutex_unlock(&event_mutex); + return ret; +} + +static void remove_subsystem_dir(const char *name) +{ + struct event_subsystem *system; + + if (strcmp(name, TRACE_SYSTEM) == 0) + return; + + list_for_each_entry(system, &event_subsystems, list) { + if (strcmp(system->name, name) == 0) { + if (!--system->nr_events) { + struct event_filter *filter = system->filter; + + debugfs_remove_recursive(system->entry); + list_del(&system->list); + if (filter) { + kfree(filter->filter_string); + kfree(filter); + } + kfree(system->name); + kfree(system); + } + break; + } + } +} + +/* + * Must be called under locking both of event_mutex and trace_event_mutex. + */ +static void __trace_remove_event_call(struct ftrace_event_call *call) +{ + ftrace_event_enable_disable(call, 0); + if (call->event) + __unregister_ftrace_event(call->event); + debugfs_remove_recursive(call->dir); + list_del(&call->list); + trace_destroy_fields(call); + destroy_preds(call); + remove_subsystem_dir(call->system); +} + +/* Remove an event_call */ +void trace_remove_event_call(struct ftrace_event_call *call) +{ + mutex_lock(&event_mutex); + down_write(&trace_event_mutex); + __trace_remove_event_call(call); + up_write(&trace_event_mutex); + mutex_unlock(&event_mutex); +} + +#define for_each_event(event, start, end) \ + for (event = start; \ + (unsigned long)event < (unsigned long)end; \ + event++) + +#ifdef CONFIG_MODULES + +static LIST_HEAD(ftrace_module_file_list); + +/* + * Modules must own their file_operations to keep up with + * reference counting. + */ +struct ftrace_module_file_ops { + struct list_head list; + struct module *mod; + struct file_operations id; + struct file_operations enable; + struct file_operations format; + struct file_operations filter; +}; + +static struct ftrace_module_file_ops * +trace_create_file_ops(struct module *mod) +{ + struct ftrace_module_file_ops *file_ops; + + /* + * This is a bit of a PITA. To allow for correct reference + * counting, modules must "own" their file_operations. + * To do this, we allocate the file operations that will be + * used in the event directory. + */ + + file_ops = kmalloc(sizeof(*file_ops), GFP_KERNEL); + if (!file_ops) + return NULL; + + file_ops->mod = mod; + + file_ops->id = ftrace_event_id_fops; + file_ops->id.owner = mod; + + file_ops->enable = ftrace_enable_fops; + file_ops->enable.owner = mod; + + file_ops->filter = ftrace_event_filter_fops; + file_ops->filter.owner = mod; + + file_ops->format = ftrace_event_format_fops; + file_ops->format.owner = mod; + + list_add(&file_ops->list, &ftrace_module_file_list); + + return file_ops; +} + +static void trace_module_add_events(struct module *mod) +{ + struct ftrace_module_file_ops *file_ops = NULL; + struct ftrace_event_call *call, *start, *end; + struct dentry *d_events; + int ret; + + start = mod->trace_events; + end = mod->trace_events + mod->num_trace_events; + + if (start == end) + return; + + d_events = event_trace_events_dir(); + if (!d_events) + return; + + for_each_event(call, start, end) { + /* The linker may leave blanks */ + if (!call->name) + continue; + if (call->raw_init) { + ret = call->raw_init(call); + if (ret < 0) { + if (ret != -ENOSYS) + pr_warning("Could not initialize trace " + "point events/%s\n", call->name); + continue; + } + } + /* + * This module has events, create file ops for this module + * if not already done. + */ + if (!file_ops) { + file_ops = trace_create_file_ops(mod); + if (!file_ops) + return; + } + call->mod = mod; + ret = event_create_dir(call, d_events, + &file_ops->id, &file_ops->enable, + &file_ops->filter, &file_ops->format); + if (!ret) + list_add(&call->list, &ftrace_events); + } +} + +static void trace_module_remove_events(struct module *mod) +{ + struct ftrace_module_file_ops *file_ops; + struct ftrace_event_call *call, *p; + bool found = false; + + down_write(&trace_event_mutex); + list_for_each_entry_safe(call, p, &ftrace_events, list) { + if (call->mod == mod) { + found = true; + __trace_remove_event_call(call); + } + } + + /* Now free the file_operations */ + list_for_each_entry(file_ops, &ftrace_module_file_list, list) { + if (file_ops->mod == mod) + break; + } + if (&file_ops->list != &ftrace_module_file_list) { + list_del(&file_ops->list); + kfree(file_ops); + } + + /* + * It is safest to reset the ring buffer if the module being unloaded + * registered any events. + */ + if (found) + tracing_reset_current_online_cpus(); + up_write(&trace_event_mutex); +} + +static int trace_module_notify(struct notifier_block *self, + unsigned long val, void *data) +{ + struct module *mod = data; + + mutex_lock(&event_mutex); + switch (val) { + case MODULE_STATE_COMING: + trace_module_add_events(mod); + break; + case MODULE_STATE_GOING: + trace_module_remove_events(mod); + break; + } + mutex_unlock(&event_mutex); + + return 0; +} +#else +static int trace_module_notify(struct notifier_block *self, + unsigned long val, void *data) +{ + return 0; +} +#endif /* CONFIG_MODULES */ + +static struct notifier_block trace_module_nb = { + .notifier_call = trace_module_notify, + .priority = 0, +}; + +extern struct ftrace_event_call __start_ftrace_events[]; +extern struct ftrace_event_call __stop_ftrace_events[]; + +static char bootup_event_buf[COMMAND_LINE_SIZE] __initdata; + +static __init int setup_trace_event(char *str) +{ + strlcpy(bootup_event_buf, str, COMMAND_LINE_SIZE); + ring_buffer_expanded = 1; + tracing_selftest_disabled = 1; + + return 1; +} +__setup("trace_event=", setup_trace_event); + +static __init int event_trace_init(void) +{ + struct ftrace_event_call *call; + struct dentry *d_tracer; + struct dentry *entry; + struct dentry *d_events; + int ret; + char *buf = bootup_event_buf; + char *token; + + d_tracer = tracing_init_dentry(); + if (!d_tracer) + return 0; + + entry = debugfs_create_file("available_events", 0444, d_tracer, + (void *)&show_event_seq_ops, + &ftrace_avail_fops); + if (!entry) + pr_warning("Could not create debugfs " + "'available_events' entry\n"); + + entry = debugfs_create_file("set_event", 0644, d_tracer, + (void *)&show_set_event_seq_ops, + &ftrace_set_event_fops); + if (!entry) + pr_warning("Could not create debugfs " + "'set_event' entry\n"); + + d_events = event_trace_events_dir(); + if (!d_events) + return 0; + + /* ring buffer internal formats */ + trace_create_file("header_page", 0444, d_events, + ring_buffer_print_page_header, + &ftrace_show_header_fops); + + trace_create_file("header_event", 0444, d_events, + ring_buffer_print_entry_header, + &ftrace_show_header_fops); + + trace_create_file("enable", 0644, d_events, + NULL, &ftrace_system_enable_fops); + + for_each_event(call, __start_ftrace_events, __stop_ftrace_events) { + /* The linker may leave blanks */ + if (!call->name) + continue; + if (call->raw_init) { + ret = call->raw_init(call); + if (ret < 0) { + if (ret != -ENOSYS) + pr_warning("Could not initialize trace " + "point events/%s\n", call->name); + continue; + } + } + ret = event_create_dir(call, d_events, &ftrace_event_id_fops, + &ftrace_enable_fops, + &ftrace_event_filter_fops, + &ftrace_event_format_fops); + if (!ret) + list_add(&call->list, &ftrace_events); + } + + while (true) { + token = strsep(&buf, ","); + + if (!token) + break; + if (!*token) + continue; + + ret = ftrace_set_clr_event(token, 1); + if (ret) + pr_warning("Failed to enable trace event: %s\n", token); + } + + ret = register_module_notifier(&trace_module_nb); + if (ret) + pr_warning("Failed to register trace events module notifier\n"); + + return 0; +} +fs_initcall(event_trace_init); + +#ifdef CONFIG_FTRACE_STARTUP_TEST + +static DEFINE_SPINLOCK(test_spinlock); +static DEFINE_SPINLOCK(test_spinlock_irq); +static DEFINE_MUTEX(test_mutex); + +static __init void test_work(struct work_struct *dummy) +{ + spin_lock(&test_spinlock); + spin_lock_irq(&test_spinlock_irq); + udelay(1); + spin_unlock_irq(&test_spinlock_irq); + spin_unlock(&test_spinlock); + + mutex_lock(&test_mutex); + msleep(1); + mutex_unlock(&test_mutex); +} + +static __init int event_test_thread(void *unused) +{ + void *test_malloc; + + test_malloc = kmalloc(1234, GFP_KERNEL); + if (!test_malloc) + pr_info("failed to kmalloc\n"); + + schedule_on_each_cpu(test_work); + + kfree(test_malloc); + + set_current_state(TASK_INTERRUPTIBLE); + while (!kthread_should_stop()) + schedule(); + + return 0; +} + +/* + * Do various things that may trigger events. + */ +static __init void event_test_stuff(void) +{ + struct task_struct *test_thread; + + test_thread = kthread_run(event_test_thread, NULL, "test-events"); + msleep(1); + kthread_stop(test_thread); +} + +/* + * For every trace event defined, we will test each trace point separately, + * and then by groups, and finally all trace points. + */ +static __init void event_trace_self_tests(void) +{ + struct ftrace_event_call *call; + struct event_subsystem *system; + int ret; + + pr_info("Running tests on trace events:\n"); + + list_for_each_entry(call, &ftrace_events, list) { + + /* Only test those that have a regfunc */ + if (!call->regfunc) + continue; + +/* + * Testing syscall events here is pretty useless, but + * we still do it if configured. But this is time consuming. + * What we really need is a user thread to perform the + * syscalls as we test. + */ +#ifndef CONFIG_EVENT_TRACE_TEST_SYSCALLS + if (call->system && + strcmp(call->system, "syscalls") == 0) + continue; +#endif + + pr_info("Testing event %s: ", call->name); + + /* + * If an event is already enabled, someone is using + * it and the self test should not be on. + */ + if (call->enabled) { + pr_warning("Enabled event during self test!\n"); + WARN_ON_ONCE(1); + continue; + } + + ftrace_event_enable_disable(call, 1); + event_test_stuff(); + ftrace_event_enable_disable(call, 0); + + pr_cont("OK\n"); + } + + /* Now test at the sub system level */ + + pr_info("Running tests on trace event systems:\n"); + + list_for_each_entry(system, &event_subsystems, list) { + + /* the ftrace system is special, skip it */ + if (strcmp(system->name, "ftrace") == 0) + continue; + + pr_info("Testing event system %s: ", system->name); + + ret = __ftrace_set_clr_event(NULL, system->name, NULL, 1); + if (WARN_ON_ONCE(ret)) { + pr_warning("error enabling system %s\n", + system->name); + continue; + } + + event_test_stuff(); + + ret = __ftrace_set_clr_event(NULL, system->name, NULL, 0); + if (WARN_ON_ONCE(ret)) + pr_warning("error disabling system %s\n", + system->name); + + pr_cont("OK\n"); + } + + /* Test with all events enabled */ + + pr_info("Running tests on all trace events:\n"); + pr_info("Testing all events: "); + + ret = __ftrace_set_clr_event(NULL, NULL, NULL, 1); + if (WARN_ON_ONCE(ret)) { + pr_warning("error enabling all events\n"); + return; + } + + event_test_stuff(); + + /* reset sysname */ + ret = __ftrace_set_clr_event(NULL, NULL, NULL, 0); + if (WARN_ON_ONCE(ret)) { + pr_warning("error disabling all events\n"); + return; + } + + pr_cont("OK\n"); +} + +#ifdef CONFIG_FUNCTION_TRACER + +static DEFINE_PER_CPU(atomic_t, ftrace_test_event_disable); + +static void +function_test_events_call(unsigned long ip, unsigned long parent_ip) +{ + struct ring_buffer_event *event; + struct ring_buffer *buffer; + struct ftrace_entry *entry; + unsigned long flags; + long disabled; + int resched; + int cpu; + int pc; + + pc = preempt_count(); + resched = ftrace_preempt_disable(); + cpu = raw_smp_processor_id(); + disabled = atomic_inc_return(&per_cpu(ftrace_test_event_disable, cpu)); + + if (disabled != 1) + goto out; + + local_save_flags(flags); + + event = trace_current_buffer_lock_reserve(&buffer, + TRACE_FN, sizeof(*entry), + flags, pc); + if (!event) + goto out; + entry = ring_buffer_event_data(event); + entry->ip = ip; + entry->parent_ip = parent_ip; + + trace_nowake_buffer_unlock_commit(buffer, event, flags, pc); + + out: + atomic_dec(&per_cpu(ftrace_test_event_disable, cpu)); + ftrace_preempt_enable(resched); +} + +static struct ftrace_ops trace_ops __initdata = +{ + .func = function_test_events_call, +}; + +static __init void event_trace_self_test_with_function(void) +{ + register_ftrace_function(&trace_ops); + pr_info("Running tests again, along with the function tracer\n"); + event_trace_self_tests(); + unregister_ftrace_function(&trace_ops); +} +#else +static __init void event_trace_self_test_with_function(void) +{ +} +#endif + +static __init int event_trace_self_tests_init(void) +{ + if (!tracing_selftest_disabled) { + event_trace_self_tests(); + event_trace_self_test_with_function(); + } + + return 0; +} + +late_initcall(event_trace_self_tests_init); + +#endif diff --git a/kernel/trace/trace_events_filter.c b/kernel/trace/trace_events_filter.c new file mode 100644 index 00000000000..50504cb228d --- /dev/null +++ b/kernel/trace/trace_events_filter.c @@ -0,0 +1,1432 @@ +/* + * trace_events_filter - generic event filtering + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + * + * Copyright (C) 2009 Tom Zanussi <tzanussi@gmail.com> + */ + +#include <linux/module.h> +#include <linux/ctype.h> +#include <linux/mutex.h> +#include <linux/perf_event.h> + +#include "trace.h" +#include "trace_output.h" + +enum filter_op_ids +{ + OP_OR, + OP_AND, + OP_GLOB, + OP_NE, + OP_EQ, + OP_LT, + OP_LE, + OP_GT, + OP_GE, + OP_NONE, + OP_OPEN_PAREN, +}; + +struct filter_op { + int id; + char *string; + int precedence; +}; + +static struct filter_op filter_ops[] = { + { OP_OR, "||", 1 }, + { OP_AND, "&&", 2 }, + { OP_GLOB, "~", 4 }, + { OP_NE, "!=", 4 }, + { OP_EQ, "==", 4 }, + { OP_LT, "<", 5 }, + { OP_LE, "<=", 5 }, + { OP_GT, ">", 5 }, + { OP_GE, ">=", 5 }, + { OP_NONE, "OP_NONE", 0 }, + { OP_OPEN_PAREN, "(", 0 }, +}; + +enum { + FILT_ERR_NONE, + FILT_ERR_INVALID_OP, + FILT_ERR_UNBALANCED_PAREN, + FILT_ERR_TOO_MANY_OPERANDS, + FILT_ERR_OPERAND_TOO_LONG, + FILT_ERR_FIELD_NOT_FOUND, + FILT_ERR_ILLEGAL_FIELD_OP, + FILT_ERR_ILLEGAL_INTVAL, + FILT_ERR_BAD_SUBSYS_FILTER, + FILT_ERR_TOO_MANY_PREDS, + FILT_ERR_MISSING_FIELD, + FILT_ERR_INVALID_FILTER, +}; + +static char *err_text[] = { + "No error", + "Invalid operator", + "Unbalanced parens", + "Too many operands", + "Operand too long", + "Field not found", + "Illegal operation for field type", + "Illegal integer value", + "Couldn't find or set field in one of a subsystem's events", + "Too many terms in predicate expression", + "Missing field name and/or value", + "Meaningless filter expression", +}; + +struct opstack_op { + int op; + struct list_head list; +}; + +struct postfix_elt { + int op; + char *operand; + struct list_head list; +}; + +struct filter_parse_state { + struct filter_op *ops; + struct list_head opstack; + struct list_head postfix; + int lasterr; + int lasterr_pos; + + struct { + char *string; + unsigned int cnt; + unsigned int tail; + } infix; + + struct { + char string[MAX_FILTER_STR_VAL]; + int pos; + unsigned int tail; + } operand; +}; + +#define DEFINE_COMPARISON_PRED(type) \ +static int filter_pred_##type(struct filter_pred *pred, void *event, \ + int val1, int val2) \ +{ \ + type *addr = (type *)(event + pred->offset); \ + type val = (type)pred->val; \ + int match = 0; \ + \ + switch (pred->op) { \ + case OP_LT: \ + match = (*addr < val); \ + break; \ + case OP_LE: \ + match = (*addr <= val); \ + break; \ + case OP_GT: \ + match = (*addr > val); \ + break; \ + case OP_GE: \ + match = (*addr >= val); \ + break; \ + default: \ + break; \ + } \ + \ + return match; \ +} + +#define DEFINE_EQUALITY_PRED(size) \ +static int filter_pred_##size(struct filter_pred *pred, void *event, \ + int val1, int val2) \ +{ \ + u##size *addr = (u##size *)(event + pred->offset); \ + u##size val = (u##size)pred->val; \ + int match; \ + \ + match = (val == *addr) ^ pred->not; \ + \ + return match; \ +} + +DEFINE_COMPARISON_PRED(s64); +DEFINE_COMPARISON_PRED(u64); +DEFINE_COMPARISON_PRED(s32); +DEFINE_COMPARISON_PRED(u32); +DEFINE_COMPARISON_PRED(s16); +DEFINE_COMPARISON_PRED(u16); +DEFINE_COMPARISON_PRED(s8); +DEFINE_COMPARISON_PRED(u8); + +DEFINE_EQUALITY_PRED(64); +DEFINE_EQUALITY_PRED(32); +DEFINE_EQUALITY_PRED(16); +DEFINE_EQUALITY_PRED(8); + +static int filter_pred_and(struct filter_pred *pred __attribute((unused)), + void *event __attribute((unused)), + int val1, int val2) +{ + return val1 && val2; +} + +static int filter_pred_or(struct filter_pred *pred __attribute((unused)), + void *event __attribute((unused)), + int val1, int val2) +{ + return val1 || val2; +} + +/* Filter predicate for fixed sized arrays of characters */ +static int filter_pred_string(struct filter_pred *pred, void *event, + int val1, int val2) +{ + char *addr = (char *)(event + pred->offset); + int cmp, match; + + cmp = pred->regex.match(addr, &pred->regex, pred->regex.field_len); + + match = cmp ^ pred->not; + + return match; +} + +/* Filter predicate for char * pointers */ +static int filter_pred_pchar(struct filter_pred *pred, void *event, + int val1, int val2) +{ + char **addr = (char **)(event + pred->offset); + int cmp, match; + + cmp = pred->regex.match(*addr, &pred->regex, pred->regex.field_len); + + match = cmp ^ pred->not; + + return match; +} + +/* + * Filter predicate for dynamic sized arrays of characters. + * These are implemented through a list of strings at the end + * of the entry. + * Also each of these strings have a field in the entry which + * contains its offset from the beginning of the entry. + * We have then first to get this field, dereference it + * and add it to the address of the entry, and at last we have + * the address of the string. + */ +static int filter_pred_strloc(struct filter_pred *pred, void *event, + int val1, int val2) +{ + u32 str_item = *(u32 *)(event + pred->offset); + int str_loc = str_item & 0xffff; + int str_len = str_item >> 16; + char *addr = (char *)(event + str_loc); + int cmp, match; + + cmp = pred->regex.match(addr, &pred->regex, str_len); + + match = cmp ^ pred->not; + + return match; +} + +static int filter_pred_none(struct filter_pred *pred, void *event, + int val1, int val2) +{ + return 0; +} + +/* Basic regex callbacks */ +static int regex_match_full(char *str, struct regex *r, int len) +{ + if (strncmp(str, r->pattern, len) == 0) + return 1; + return 0; +} + +static int regex_match_front(char *str, struct regex *r, int len) +{ + if (strncmp(str, r->pattern, len) == 0) + return 1; + return 0; +} + +static int regex_match_middle(char *str, struct regex *r, int len) +{ + if (strstr(str, r->pattern)) + return 1; + return 0; +} + +static int regex_match_end(char *str, struct regex *r, int len) +{ + char *ptr = strstr(str, r->pattern); + + if (ptr && (ptr[r->len] == 0)) + return 1; + return 0; +} + +/** + * filter_parse_regex - parse a basic regex + * @buff: the raw regex + * @len: length of the regex + * @search: will point to the beginning of the string to compare + * @not: tell whether the match will have to be inverted + * + * This passes in a buffer containing a regex and this function will + * set search to point to the search part of the buffer and + * return the type of search it is (see enum above). + * This does modify buff. + * + * Returns enum type. + * search returns the pointer to use for comparison. + * not returns 1 if buff started with a '!' + * 0 otherwise. + */ +enum regex_type filter_parse_regex(char *buff, int len, char **search, int *not) +{ + int type = MATCH_FULL; + int i; + + if (buff[0] == '!') { + *not = 1; + buff++; + len--; + } else + *not = 0; + + *search = buff; + + for (i = 0; i < len; i++) { + if (buff[i] == '*') { + if (!i) { + *search = buff + 1; + type = MATCH_END_ONLY; + } else { + if (type == MATCH_END_ONLY) + type = MATCH_MIDDLE_ONLY; + else + type = MATCH_FRONT_ONLY; + buff[i] = 0; + break; + } + } + } + + return type; +} + +static void filter_build_regex(struct filter_pred *pred) +{ + struct regex *r = &pred->regex; + char *search; + enum regex_type type = MATCH_FULL; + int not = 0; + + if (pred->op == OP_GLOB) { + type = filter_parse_regex(r->pattern, r->len, &search, ¬); + r->len = strlen(search); + memmove(r->pattern, search, r->len+1); + } + + switch (type) { + case MATCH_FULL: + r->match = regex_match_full; + break; + case MATCH_FRONT_ONLY: + r->match = regex_match_front; + break; + case MATCH_MIDDLE_ONLY: + r->match = regex_match_middle; + break; + case MATCH_END_ONLY: + r->match = regex_match_end; + break; + } + + pred->not ^= not; +} + +/* return 1 if event matches, 0 otherwise (discard) */ +int filter_match_preds(struct event_filter *filter, void *rec) +{ + int match, top = 0, val1 = 0, val2 = 0; + int stack[MAX_FILTER_PRED]; + struct filter_pred *pred; + int i; + + for (i = 0; i < filter->n_preds; i++) { + pred = filter->preds[i]; + if (!pred->pop_n) { + match = pred->fn(pred, rec, val1, val2); + stack[top++] = match; + continue; + } + if (pred->pop_n > top) { + WARN_ON_ONCE(1); + return 0; + } + val1 = stack[--top]; + val2 = stack[--top]; + match = pred->fn(pred, rec, val1, val2); + stack[top++] = match; + } + + return stack[--top]; +} +EXPORT_SYMBOL_GPL(filter_match_preds); + +static void parse_error(struct filter_parse_state *ps, int err, int pos) +{ + ps->lasterr = err; + ps->lasterr_pos = pos; +} + +static void remove_filter_string(struct event_filter *filter) +{ + kfree(filter->filter_string); + filter->filter_string = NULL; +} + +static int replace_filter_string(struct event_filter *filter, + char *filter_string) +{ + kfree(filter->filter_string); + filter->filter_string = kstrdup(filter_string, GFP_KERNEL); + if (!filter->filter_string) + return -ENOMEM; + + return 0; +} + +static int append_filter_string(struct event_filter *filter, + char *string) +{ + int newlen; + char *new_filter_string; + + BUG_ON(!filter->filter_string); + newlen = strlen(filter->filter_string) + strlen(string) + 1; + new_filter_string = kmalloc(newlen, GFP_KERNEL); + if (!new_filter_string) + return -ENOMEM; + + strcpy(new_filter_string, filter->filter_string); + strcat(new_filter_string, string); + kfree(filter->filter_string); + filter->filter_string = new_filter_string; + + return 0; +} + +static void append_filter_err(struct filter_parse_state *ps, + struct event_filter *filter) +{ + int pos = ps->lasterr_pos; + char *buf, *pbuf; + + buf = (char *)__get_free_page(GFP_TEMPORARY); + if (!buf) + return; + + append_filter_string(filter, "\n"); + memset(buf, ' ', PAGE_SIZE); + if (pos > PAGE_SIZE - 128) + pos = 0; + buf[pos] = '^'; + pbuf = &buf[pos] + 1; + + sprintf(pbuf, "\nparse_error: %s\n", err_text[ps->lasterr]); + append_filter_string(filter, buf); + free_page((unsigned long) buf); +} + +void print_event_filter(struct ftrace_event_call *call, struct trace_seq *s) +{ + struct event_filter *filter = call->filter; + + mutex_lock(&event_mutex); + if (filter && filter->filter_string) + trace_seq_printf(s, "%s\n", filter->filter_string); + else + trace_seq_printf(s, "none\n"); + mutex_unlock(&event_mutex); +} + +void print_subsystem_event_filter(struct event_subsystem *system, + struct trace_seq *s) +{ + struct event_filter *filter = system->filter; + + mutex_lock(&event_mutex); + if (filter && filter->filter_string) + trace_seq_printf(s, "%s\n", filter->filter_string); + else + trace_seq_printf(s, "none\n"); + mutex_unlock(&event_mutex); +} + +static struct ftrace_event_field * +find_event_field(struct ftrace_event_call *call, char *name) +{ + struct ftrace_event_field *field; + + list_for_each_entry(field, &call->fields, link) { + if (!strcmp(field->name, name)) + return field; + } + + return NULL; +} + +static void filter_free_pred(struct filter_pred *pred) +{ + if (!pred) + return; + + kfree(pred->field_name); + kfree(pred); +} + +static void filter_clear_pred(struct filter_pred *pred) +{ + kfree(pred->field_name); + pred->field_name = NULL; + pred->regex.len = 0; +} + +static int filter_set_pred(struct filter_pred *dest, + struct filter_pred *src, + filter_pred_fn_t fn) +{ + *dest = *src; + if (src->field_name) { + dest->field_name = kstrdup(src->field_name, GFP_KERNEL); + if (!dest->field_name) + return -ENOMEM; + } + dest->fn = fn; + + return 0; +} + +static void filter_disable_preds(struct ftrace_event_call *call) +{ + struct event_filter *filter = call->filter; + int i; + + call->filter_active = 0; + filter->n_preds = 0; + + for (i = 0; i < MAX_FILTER_PRED; i++) + filter->preds[i]->fn = filter_pred_none; +} + +static void __free_preds(struct event_filter *filter) +{ + int i; + + if (!filter) + return; + + for (i = 0; i < MAX_FILTER_PRED; i++) { + if (filter->preds[i]) + filter_free_pred(filter->preds[i]); + } + kfree(filter->preds); + kfree(filter->filter_string); + kfree(filter); +} + +void destroy_preds(struct ftrace_event_call *call) +{ + __free_preds(call->filter); + call->filter = NULL; + call->filter_active = 0; +} + +static struct event_filter *__alloc_preds(void) +{ + struct event_filter *filter; + struct filter_pred *pred; + int i; + + filter = kzalloc(sizeof(*filter), GFP_KERNEL); + if (!filter) + return ERR_PTR(-ENOMEM); + + filter->n_preds = 0; + + filter->preds = kzalloc(MAX_FILTER_PRED * sizeof(pred), GFP_KERNEL); + if (!filter->preds) + goto oom; + + for (i = 0; i < MAX_FILTER_PRED; i++) { + pred = kzalloc(sizeof(*pred), GFP_KERNEL); + if (!pred) + goto oom; + pred->fn = filter_pred_none; + filter->preds[i] = pred; + } + + return filter; + +oom: + __free_preds(filter); + return ERR_PTR(-ENOMEM); +} + +static int init_preds(struct ftrace_event_call *call) +{ + if (call->filter) + return 0; + + call->filter_active = 0; + call->filter = __alloc_preds(); + if (IS_ERR(call->filter)) + return PTR_ERR(call->filter); + + return 0; +} + +static int init_subsystem_preds(struct event_subsystem *system) +{ + struct ftrace_event_call *call; + int err; + + list_for_each_entry(call, &ftrace_events, list) { + if (!call->define_fields) + continue; + + if (strcmp(call->system, system->name) != 0) + continue; + + err = init_preds(call); + if (err) + return err; + } + + return 0; +} + +static void filter_free_subsystem_preds(struct event_subsystem *system) +{ + struct ftrace_event_call *call; + + list_for_each_entry(call, &ftrace_events, list) { + if (!call->define_fields) + continue; + + if (strcmp(call->system, system->name) != 0) + continue; + + filter_disable_preds(call); + remove_filter_string(call->filter); + } +} + +static int filter_add_pred_fn(struct filter_parse_state *ps, + struct ftrace_event_call *call, + struct event_filter *filter, + struct filter_pred *pred, + filter_pred_fn_t fn) +{ + int idx, err; + + if (filter->n_preds == MAX_FILTER_PRED) { + parse_error(ps, FILT_ERR_TOO_MANY_PREDS, 0); + return -ENOSPC; + } + + idx = filter->n_preds; + filter_clear_pred(filter->preds[idx]); + err = filter_set_pred(filter->preds[idx], pred, fn); + if (err) + return err; + + filter->n_preds++; + + return 0; +} + +int filter_assign_type(const char *type) +{ + if (strstr(type, "__data_loc") && strstr(type, "char")) + return FILTER_DYN_STRING; + + if (strchr(type, '[') && strstr(type, "char")) + return FILTER_STATIC_STRING; + + return FILTER_OTHER; +} + +static bool is_string_field(struct ftrace_event_field *field) +{ + return field->filter_type == FILTER_DYN_STRING || + field->filter_type == FILTER_STATIC_STRING || + field->filter_type == FILTER_PTR_STRING; +} + +static int is_legal_op(struct ftrace_event_field *field, int op) +{ + if (is_string_field(field) && + (op != OP_EQ && op != OP_NE && op != OP_GLOB)) + return 0; + if (!is_string_field(field) && op == OP_GLOB) + return 0; + + return 1; +} + +static filter_pred_fn_t select_comparison_fn(int op, int field_size, + int field_is_signed) +{ + filter_pred_fn_t fn = NULL; + + switch (field_size) { + case 8: + if (op == OP_EQ || op == OP_NE) + fn = filter_pred_64; + else if (field_is_signed) + fn = filter_pred_s64; + else + fn = filter_pred_u64; + break; + case 4: + if (op == OP_EQ || op == OP_NE) + fn = filter_pred_32; + else if (field_is_signed) + fn = filter_pred_s32; + else + fn = filter_pred_u32; + break; + case 2: + if (op == OP_EQ || op == OP_NE) + fn = filter_pred_16; + else if (field_is_signed) + fn = filter_pred_s16; + else + fn = filter_pred_u16; + break; + case 1: + if (op == OP_EQ || op == OP_NE) + fn = filter_pred_8; + else if (field_is_signed) + fn = filter_pred_s8; + else + fn = filter_pred_u8; + break; + } + + return fn; +} + +static int filter_add_pred(struct filter_parse_state *ps, + struct ftrace_event_call *call, + struct event_filter *filter, + struct filter_pred *pred, + bool dry_run) +{ + struct ftrace_event_field *field; + filter_pred_fn_t fn; + unsigned long long val; + int ret; + + pred->fn = filter_pred_none; + + if (pred->op == OP_AND) { + pred->pop_n = 2; + fn = filter_pred_and; + goto add_pred_fn; + } else if (pred->op == OP_OR) { + pred->pop_n = 2; + fn = filter_pred_or; + goto add_pred_fn; + } + + field = find_event_field(call, pred->field_name); + if (!field) { + parse_error(ps, FILT_ERR_FIELD_NOT_FOUND, 0); + return -EINVAL; + } + + pred->offset = field->offset; + + if (!is_legal_op(field, pred->op)) { + parse_error(ps, FILT_ERR_ILLEGAL_FIELD_OP, 0); + return -EINVAL; + } + + if (is_string_field(field)) { + filter_build_regex(pred); + + if (field->filter_type == FILTER_STATIC_STRING) { + fn = filter_pred_string; + pred->regex.field_len = field->size; + } else if (field->filter_type == FILTER_DYN_STRING) + fn = filter_pred_strloc; + else { + fn = filter_pred_pchar; + pred->regex.field_len = strlen(pred->regex.pattern); + } + } else { + if (field->is_signed) + ret = strict_strtoll(pred->regex.pattern, 0, &val); + else + ret = strict_strtoull(pred->regex.pattern, 0, &val); + if (ret) { + parse_error(ps, FILT_ERR_ILLEGAL_INTVAL, 0); + return -EINVAL; + } + pred->val = val; + + fn = select_comparison_fn(pred->op, field->size, + field->is_signed); + if (!fn) { + parse_error(ps, FILT_ERR_INVALID_OP, 0); + return -EINVAL; + } + } + + if (pred->op == OP_NE) + pred->not = 1; + +add_pred_fn: + if (!dry_run) + return filter_add_pred_fn(ps, call, filter, pred, fn); + return 0; +} + +static void parse_init(struct filter_parse_state *ps, + struct filter_op *ops, + char *infix_string) +{ + memset(ps, '\0', sizeof(*ps)); + + ps->infix.string = infix_string; + ps->infix.cnt = strlen(infix_string); + ps->ops = ops; + + INIT_LIST_HEAD(&ps->opstack); + INIT_LIST_HEAD(&ps->postfix); +} + +static char infix_next(struct filter_parse_state *ps) +{ + ps->infix.cnt--; + + return ps->infix.string[ps->infix.tail++]; +} + +static char infix_peek(struct filter_parse_state *ps) +{ + if (ps->infix.tail == strlen(ps->infix.string)) + return 0; + + return ps->infix.string[ps->infix.tail]; +} + +static void infix_advance(struct filter_parse_state *ps) +{ + ps->infix.cnt--; + ps->infix.tail++; +} + +static inline int is_precedence_lower(struct filter_parse_state *ps, + int a, int b) +{ + return ps->ops[a].precedence < ps->ops[b].precedence; +} + +static inline int is_op_char(struct filter_parse_state *ps, char c) +{ + int i; + + for (i = 0; strcmp(ps->ops[i].string, "OP_NONE"); i++) { + if (ps->ops[i].string[0] == c) + return 1; + } + + return 0; +} + +static int infix_get_op(struct filter_parse_state *ps, char firstc) +{ + char nextc = infix_peek(ps); + char opstr[3]; + int i; + + opstr[0] = firstc; + opstr[1] = nextc; + opstr[2] = '\0'; + + for (i = 0; strcmp(ps->ops[i].string, "OP_NONE"); i++) { + if (!strcmp(opstr, ps->ops[i].string)) { + infix_advance(ps); + return ps->ops[i].id; + } + } + + opstr[1] = '\0'; + + for (i = 0; strcmp(ps->ops[i].string, "OP_NONE"); i++) { + if (!strcmp(opstr, ps->ops[i].string)) + return ps->ops[i].id; + } + + return OP_NONE; +} + +static inline void clear_operand_string(struct filter_parse_state *ps) +{ + memset(ps->operand.string, '\0', MAX_FILTER_STR_VAL); + ps->operand.tail = 0; +} + +static inline int append_operand_char(struct filter_parse_state *ps, char c) +{ + if (ps->operand.tail == MAX_FILTER_STR_VAL - 1) + return -EINVAL; + + ps->operand.string[ps->operand.tail++] = c; + + return 0; +} + +static int filter_opstack_push(struct filter_parse_state *ps, int op) +{ + struct opstack_op *opstack_op; + + opstack_op = kmalloc(sizeof(*opstack_op), GFP_KERNEL); + if (!opstack_op) + return -ENOMEM; + + opstack_op->op = op; + list_add(&opstack_op->list, &ps->opstack); + + return 0; +} + +static int filter_opstack_empty(struct filter_parse_state *ps) +{ + return list_empty(&ps->opstack); +} + +static int filter_opstack_top(struct filter_parse_state *ps) +{ + struct opstack_op *opstack_op; + + if (filter_opstack_empty(ps)) + return OP_NONE; + + opstack_op = list_first_entry(&ps->opstack, struct opstack_op, list); + + return opstack_op->op; +} + +static int filter_opstack_pop(struct filter_parse_state *ps) +{ + struct opstack_op *opstack_op; + int op; + + if (filter_opstack_empty(ps)) + return OP_NONE; + + opstack_op = list_first_entry(&ps->opstack, struct opstack_op, list); + op = opstack_op->op; + list_del(&opstack_op->list); + + kfree(opstack_op); + + return op; +} + +static void filter_opstack_clear(struct filter_parse_state *ps) +{ + while (!filter_opstack_empty(ps)) + filter_opstack_pop(ps); +} + +static char *curr_operand(struct filter_parse_state *ps) +{ + return ps->operand.string; +} + +static int postfix_append_operand(struct filter_parse_state *ps, char *operand) +{ + struct postfix_elt *elt; + + elt = kmalloc(sizeof(*elt), GFP_KERNEL); + if (!elt) + return -ENOMEM; + + elt->op = OP_NONE; + elt->operand = kstrdup(operand, GFP_KERNEL); + if (!elt->operand) { + kfree(elt); + return -ENOMEM; + } + + list_add_tail(&elt->list, &ps->postfix); + + return 0; +} + +static int postfix_append_op(struct filter_parse_state *ps, int op) +{ + struct postfix_elt *elt; + + elt = kmalloc(sizeof(*elt), GFP_KERNEL); + if (!elt) + return -ENOMEM; + + elt->op = op; + elt->operand = NULL; + + list_add_tail(&elt->list, &ps->postfix); + + return 0; +} + +static void postfix_clear(struct filter_parse_state *ps) +{ + struct postfix_elt *elt; + + while (!list_empty(&ps->postfix)) { + elt = list_first_entry(&ps->postfix, struct postfix_elt, list); + list_del(&elt->list); + kfree(elt->operand); + kfree(elt); + } +} + +static int filter_parse(struct filter_parse_state *ps) +{ + int in_string = 0; + int op, top_op; + char ch; + + while ((ch = infix_next(ps))) { + if (ch == '"') { + in_string ^= 1; + continue; + } + + if (in_string) + goto parse_operand; + + if (isspace(ch)) + continue; + + if (is_op_char(ps, ch)) { + op = infix_get_op(ps, ch); + if (op == OP_NONE) { + parse_error(ps, FILT_ERR_INVALID_OP, 0); + return -EINVAL; + } + + if (strlen(curr_operand(ps))) { + postfix_append_operand(ps, curr_operand(ps)); + clear_operand_string(ps); + } + + while (!filter_opstack_empty(ps)) { + top_op = filter_opstack_top(ps); + if (!is_precedence_lower(ps, top_op, op)) { + top_op = filter_opstack_pop(ps); + postfix_append_op(ps, top_op); + continue; + } + break; + } + + filter_opstack_push(ps, op); + continue; + } + + if (ch == '(') { + filter_opstack_push(ps, OP_OPEN_PAREN); + continue; + } + + if (ch == ')') { + if (strlen(curr_operand(ps))) { + postfix_append_operand(ps, curr_operand(ps)); + clear_operand_string(ps); + } + + top_op = filter_opstack_pop(ps); + while (top_op != OP_NONE) { + if (top_op == OP_OPEN_PAREN) + break; + postfix_append_op(ps, top_op); + top_op = filter_opstack_pop(ps); + } + if (top_op == OP_NONE) { + parse_error(ps, FILT_ERR_UNBALANCED_PAREN, 0); + return -EINVAL; + } + continue; + } +parse_operand: + if (append_operand_char(ps, ch)) { + parse_error(ps, FILT_ERR_OPERAND_TOO_LONG, 0); + return -EINVAL; + } + } + + if (strlen(curr_operand(ps))) + postfix_append_operand(ps, curr_operand(ps)); + + while (!filter_opstack_empty(ps)) { + top_op = filter_opstack_pop(ps); + if (top_op == OP_NONE) + break; + if (top_op == OP_OPEN_PAREN) { + parse_error(ps, FILT_ERR_UNBALANCED_PAREN, 0); + return -EINVAL; + } + postfix_append_op(ps, top_op); + } + + return 0; +} + +static struct filter_pred *create_pred(int op, char *operand1, char *operand2) +{ + struct filter_pred *pred; + + pred = kzalloc(sizeof(*pred), GFP_KERNEL); + if (!pred) + return NULL; + + pred->field_name = kstrdup(operand1, GFP_KERNEL); + if (!pred->field_name) { + kfree(pred); + return NULL; + } + + strcpy(pred->regex.pattern, operand2); + pred->regex.len = strlen(pred->regex.pattern); + + pred->op = op; + + return pred; +} + +static struct filter_pred *create_logical_pred(int op) +{ + struct filter_pred *pred; + + pred = kzalloc(sizeof(*pred), GFP_KERNEL); + if (!pred) + return NULL; + + pred->op = op; + + return pred; +} + +static int check_preds(struct filter_parse_state *ps) +{ + int n_normal_preds = 0, n_logical_preds = 0; + struct postfix_elt *elt; + + list_for_each_entry(elt, &ps->postfix, list) { + if (elt->op == OP_NONE) + continue; + + if (elt->op == OP_AND || elt->op == OP_OR) { + n_logical_preds++; + continue; + } + n_normal_preds++; + } + + if (!n_normal_preds || n_logical_preds >= n_normal_preds) { + parse_error(ps, FILT_ERR_INVALID_FILTER, 0); + return -EINVAL; + } + + return 0; +} + +static int replace_preds(struct ftrace_event_call *call, + struct event_filter *filter, + struct filter_parse_state *ps, + char *filter_string, + bool dry_run) +{ + char *operand1 = NULL, *operand2 = NULL; + struct filter_pred *pred; + struct postfix_elt *elt; + int err; + int n_preds = 0; + + err = check_preds(ps); + if (err) + return err; + + list_for_each_entry(elt, &ps->postfix, list) { + if (elt->op == OP_NONE) { + if (!operand1) + operand1 = elt->operand; + else if (!operand2) + operand2 = elt->operand; + else { + parse_error(ps, FILT_ERR_TOO_MANY_OPERANDS, 0); + return -EINVAL; + } + continue; + } + + if (n_preds++ == MAX_FILTER_PRED) { + parse_error(ps, FILT_ERR_TOO_MANY_PREDS, 0); + return -ENOSPC; + } + + if (elt->op == OP_AND || elt->op == OP_OR) { + pred = create_logical_pred(elt->op); + goto add_pred; + } + + if (!operand1 || !operand2) { + parse_error(ps, FILT_ERR_MISSING_FIELD, 0); + return -EINVAL; + } + + pred = create_pred(elt->op, operand1, operand2); +add_pred: + if (!pred) + return -ENOMEM; + err = filter_add_pred(ps, call, filter, pred, dry_run); + filter_free_pred(pred); + if (err) + return err; + + operand1 = operand2 = NULL; + } + + return 0; +} + +static int replace_system_preds(struct event_subsystem *system, + struct filter_parse_state *ps, + char *filter_string) +{ + struct ftrace_event_call *call; + bool fail = true; + int err; + + list_for_each_entry(call, &ftrace_events, list) { + struct event_filter *filter = call->filter; + + if (!call->define_fields) + continue; + + if (strcmp(call->system, system->name) != 0) + continue; + + /* try to see if the filter can be applied */ + err = replace_preds(call, filter, ps, filter_string, true); + if (err) + continue; + + /* really apply the filter */ + filter_disable_preds(call); + err = replace_preds(call, filter, ps, filter_string, false); + if (err) + filter_disable_preds(call); + else { + call->filter_active = 1; + replace_filter_string(filter, filter_string); + } + fail = false; + } + + if (fail) { + parse_error(ps, FILT_ERR_BAD_SUBSYS_FILTER, 0); + return -EINVAL; + } + return 0; +} + +int apply_event_filter(struct ftrace_event_call *call, char *filter_string) +{ + int err; + struct filter_parse_state *ps; + + mutex_lock(&event_mutex); + + err = init_preds(call); + if (err) + goto out_unlock; + + if (!strcmp(strstrip(filter_string), "0")) { + filter_disable_preds(call); + remove_filter_string(call->filter); + goto out_unlock; + } + + err = -ENOMEM; + ps = kzalloc(sizeof(*ps), GFP_KERNEL); + if (!ps) + goto out_unlock; + + filter_disable_preds(call); + replace_filter_string(call->filter, filter_string); + + parse_init(ps, filter_ops, filter_string); + err = filter_parse(ps); + if (err) { + append_filter_err(ps, call->filter); + goto out; + } + + err = replace_preds(call, call->filter, ps, filter_string, false); + if (err) + append_filter_err(ps, call->filter); + else + call->filter_active = 1; +out: + filter_opstack_clear(ps); + postfix_clear(ps); + kfree(ps); +out_unlock: + mutex_unlock(&event_mutex); + + return err; +} + +int apply_subsystem_event_filter(struct event_subsystem *system, + char *filter_string) +{ + int err; + struct filter_parse_state *ps; + + mutex_lock(&event_mutex); + + err = init_subsystem_preds(system); + if (err) + goto out_unlock; + + if (!strcmp(strstrip(filter_string), "0")) { + filter_free_subsystem_preds(system); + remove_filter_string(system->filter); + goto out_unlock; + } + + err = -ENOMEM; + ps = kzalloc(sizeof(*ps), GFP_KERNEL); + if (!ps) + goto out_unlock; + + replace_filter_string(system->filter, filter_string); + + parse_init(ps, filter_ops, filter_string); + err = filter_parse(ps); + if (err) { + append_filter_err(ps, system->filter); + goto out; + } + + err = replace_system_preds(system, ps, filter_string); + if (err) + append_filter_err(ps, system->filter); + +out: + filter_opstack_clear(ps); + postfix_clear(ps); + kfree(ps); +out_unlock: + mutex_unlock(&event_mutex); + + return err; +} + +#ifdef CONFIG_EVENT_PROFILE + +void ftrace_profile_free_filter(struct perf_event *event) +{ + struct event_filter *filter = event->filter; + + event->filter = NULL; + __free_preds(filter); +} + +int ftrace_profile_set_filter(struct perf_event *event, int event_id, + char *filter_str) +{ + int err; + struct event_filter *filter; + struct filter_parse_state *ps; + struct ftrace_event_call *call = NULL; + + mutex_lock(&event_mutex); + + list_for_each_entry(call, &ftrace_events, list) { + if (call->id == event_id) + break; + } + + err = -EINVAL; + if (!call) + goto out_unlock; + + err = -EEXIST; + if (event->filter) + goto out_unlock; + + filter = __alloc_preds(); + if (IS_ERR(filter)) { + err = PTR_ERR(filter); + goto out_unlock; + } + + err = -ENOMEM; + ps = kzalloc(sizeof(*ps), GFP_KERNEL); + if (!ps) + goto free_preds; + + parse_init(ps, filter_ops, filter_str); + err = filter_parse(ps); + if (err) + goto free_ps; + + err = replace_preds(call, filter, ps, filter_str, false); + if (!err) + event->filter = filter; + +free_ps: + filter_opstack_clear(ps); + postfix_clear(ps); + kfree(ps); + +free_preds: + if (err) + __free_preds(filter); + +out_unlock: + mutex_unlock(&event_mutex); + + return err; +} + +#endif /* CONFIG_EVENT_PROFILE */ + diff --git a/kernel/trace/trace_export.c b/kernel/trace/trace_export.c new file mode 100644 index 00000000000..dff8c84ddf1 --- /dev/null +++ b/kernel/trace/trace_export.c @@ -0,0 +1,233 @@ +/* + * trace_export.c - export basic ftrace utilities to user space + * + * Copyright (C) 2009 Steven Rostedt <srostedt@redhat.com> + */ +#include <linux/stringify.h> +#include <linux/kallsyms.h> +#include <linux/seq_file.h> +#include <linux/debugfs.h> +#include <linux/uaccess.h> +#include <linux/ftrace.h> +#include <linux/module.h> +#include <linux/init.h> +#include <linux/fs.h> + +#include "trace_output.h" + +#undef TRACE_SYSTEM +#define TRACE_SYSTEM ftrace + +/* not needed for this file */ +#undef __field_struct +#define __field_struct(type, item) + +#undef __field +#define __field(type, item) type item; + +#undef __field_desc +#define __field_desc(type, container, item) type item; + +#undef __array +#define __array(type, item, size) type item[size]; + +#undef __array_desc +#define __array_desc(type, container, item, size) type item[size]; + +#undef __dynamic_array +#define __dynamic_array(type, item) type item[]; + +#undef F_STRUCT +#define F_STRUCT(args...) args + +#undef F_printk +#define F_printk(fmt, args...) fmt, args + +#undef FTRACE_ENTRY +#define FTRACE_ENTRY(name, struct_name, id, tstruct, print) \ +struct ____ftrace_##name { \ + tstruct \ +}; \ +static void __always_unused ____ftrace_check_##name(void) \ +{ \ + struct ____ftrace_##name *__entry = NULL; \ + \ + /* force compile-time check on F_printk() */ \ + printk(print); \ +} + +#undef FTRACE_ENTRY_DUP +#define FTRACE_ENTRY_DUP(name, struct_name, id, tstruct, print) \ + FTRACE_ENTRY(name, struct_name, id, PARAMS(tstruct), PARAMS(print)) + +#include "trace_entries.h" + + +#undef __field +#define __field(type, item) \ + ret = trace_seq_printf(s, "\tfield:" #type " " #item ";\t" \ + "offset:%zu;\tsize:%zu;\tsigned:%u;\n", \ + offsetof(typeof(field), item), \ + sizeof(field.item), is_signed_type(type)); \ + if (!ret) \ + return 0; + +#undef __field_desc +#define __field_desc(type, container, item) \ + ret = trace_seq_printf(s, "\tfield:" #type " " #item ";\t" \ + "offset:%zu;\tsize:%zu;\tsigned:%u;\n", \ + offsetof(typeof(field), container.item), \ + sizeof(field.container.item), \ + is_signed_type(type)); \ + if (!ret) \ + return 0; + +#undef __array +#define __array(type, item, len) \ + ret = trace_seq_printf(s, "\tfield:" #type " " #item "[" #len "];\t" \ + "offset:%zu;\tsize:%zu;\tsigned:%u;\n", \ + offsetof(typeof(field), item), \ + sizeof(field.item), is_signed_type(type)); \ + if (!ret) \ + return 0; + +#undef __array_desc +#define __array_desc(type, container, item, len) \ + ret = trace_seq_printf(s, "\tfield:" #type " " #item "[" #len "];\t" \ + "offset:%zu;\tsize:%zu;\tsigned:%u;\n", \ + offsetof(typeof(field), container.item), \ + sizeof(field.container.item), \ + is_signed_type(type)); \ + if (!ret) \ + return 0; + +#undef __dynamic_array +#define __dynamic_array(type, item) \ + ret = trace_seq_printf(s, "\tfield:" #type " " #item ";\t" \ + "offset:%zu;\tsize:0;\tsigned:%u;\n", \ + offsetof(typeof(field), item), \ + is_signed_type(type)); \ + if (!ret) \ + return 0; + +#undef F_printk +#define F_printk(fmt, args...) "%s, %s\n", #fmt, __stringify(args) + +#undef __entry +#define __entry REC + +#undef FTRACE_ENTRY +#define FTRACE_ENTRY(name, struct_name, id, tstruct, print) \ +static int \ +ftrace_format_##name(struct ftrace_event_call *unused, \ + struct trace_seq *s) \ +{ \ + struct struct_name field __attribute__((unused)); \ + int ret = 0; \ + \ + tstruct; \ + \ + trace_seq_printf(s, "\nprint fmt: " print); \ + \ + return ret; \ +} + +#include "trace_entries.h" + +#undef __field +#define __field(type, item) \ + ret = trace_define_field(event_call, #type, #item, \ + offsetof(typeof(field), item), \ + sizeof(field.item), \ + is_signed_type(type), FILTER_OTHER); \ + if (ret) \ + return ret; + +#undef __field_desc +#define __field_desc(type, container, item) \ + ret = trace_define_field(event_call, #type, #item, \ + offsetof(typeof(field), \ + container.item), \ + sizeof(field.container.item), \ + is_signed_type(type), FILTER_OTHER); \ + if (ret) \ + return ret; + +#undef __array +#define __array(type, item, len) \ + BUILD_BUG_ON(len > MAX_FILTER_STR_VAL); \ + ret = trace_define_field(event_call, #type "[" #len "]", #item, \ + offsetof(typeof(field), item), \ + sizeof(field.item), 0, FILTER_OTHER); \ + if (ret) \ + return ret; + +#undef __array_desc +#define __array_desc(type, container, item, len) \ + BUILD_BUG_ON(len > MAX_FILTER_STR_VAL); \ + ret = trace_define_field(event_call, #type "[" #len "]", #item, \ + offsetof(typeof(field), \ + container.item), \ + sizeof(field.container.item), 0, \ + FILTER_OTHER); \ + if (ret) \ + return ret; + +#undef __dynamic_array +#define __dynamic_array(type, item) + +#undef FTRACE_ENTRY +#define FTRACE_ENTRY(name, struct_name, id, tstruct, print) \ +int \ +ftrace_define_fields_##name(struct ftrace_event_call *event_call) \ +{ \ + struct struct_name field; \ + int ret; \ + \ + ret = trace_define_common_fields(event_call); \ + if (ret) \ + return ret; \ + \ + tstruct; \ + \ + return ret; \ +} + +#include "trace_entries.h" + +static int ftrace_raw_init_event(struct ftrace_event_call *call) +{ + INIT_LIST_HEAD(&call->fields); + return 0; +} + +#undef __field +#define __field(type, item) + +#undef __field_desc +#define __field_desc(type, container, item) + +#undef __array +#define __array(type, item, len) + +#undef __array_desc +#define __array_desc(type, container, item, len) + +#undef __dynamic_array +#define __dynamic_array(type, item) + +#undef FTRACE_ENTRY +#define FTRACE_ENTRY(call, struct_name, type, tstruct, print) \ + \ +struct ftrace_event_call __used \ +__attribute__((__aligned__(4))) \ +__attribute__((section("_ftrace_events"))) event_##call = { \ + .name = #call, \ + .id = type, \ + .system = __stringify(TRACE_SYSTEM), \ + .raw_init = ftrace_raw_init_event, \ + .show_format = ftrace_format_##call, \ + .define_fields = ftrace_define_fields_##call, \ +}; \ + +#include "trace_entries.h" diff --git a/kernel/trace/trace_functions.c b/kernel/trace/trace_functions.c index 9236d7e25a1..b3f3776b0cd 100644 --- a/kernel/trace/trace_functions.c +++ b/kernel/trace/trace_functions.c @@ -9,6 +9,7 @@ * Copyright (C) 2004-2006 Ingo Molnar * Copyright (C) 2004 William Lee Irwin III */ +#include <linux/ring_buffer.h> #include <linux/debugfs.h> #include <linux/uaccess.h> #include <linux/ftrace.h> @@ -16,52 +17,387 @@ #include "trace.h" -static void start_function_trace(struct trace_array *tr) +/* function tracing enabled */ +static int ftrace_function_enabled; + +static struct trace_array *func_trace; + +static void tracing_start_function_trace(void); +static void tracing_stop_function_trace(void); + +static int function_trace_init(struct trace_array *tr) { + func_trace = tr; tr->cpu = get_cpu(); - tracing_reset_online_cpus(tr); put_cpu(); tracing_start_cmdline_record(); tracing_start_function_trace(); + return 0; } -static void stop_function_trace(struct trace_array *tr) +static void function_trace_reset(struct trace_array *tr) { tracing_stop_function_trace(); tracing_stop_cmdline_record(); } -static int function_trace_init(struct trace_array *tr) +static void function_trace_start(struct trace_array *tr) { - start_function_trace(tr); - return 0; + tracing_reset_online_cpus(tr); } -static void function_trace_reset(struct trace_array *tr) +static void +function_trace_call_preempt_only(unsigned long ip, unsigned long parent_ip) +{ + struct trace_array *tr = func_trace; + struct trace_array_cpu *data; + unsigned long flags; + long disabled; + int cpu, resched; + int pc; + + if (unlikely(!ftrace_function_enabled)) + return; + + pc = preempt_count(); + resched = ftrace_preempt_disable(); + local_save_flags(flags); + cpu = raw_smp_processor_id(); + data = tr->data[cpu]; + disabled = atomic_inc_return(&data->disabled); + + if (likely(disabled == 1)) + trace_function(tr, ip, parent_ip, flags, pc); + + atomic_dec(&data->disabled); + ftrace_preempt_enable(resched); +} + +static void +function_trace_call(unsigned long ip, unsigned long parent_ip) { - stop_function_trace(tr); + struct trace_array *tr = func_trace; + struct trace_array_cpu *data; + unsigned long flags; + long disabled; + int cpu; + int pc; + + if (unlikely(!ftrace_function_enabled)) + return; + + /* + * Need to use raw, since this must be called before the + * recursive protection is performed. + */ + local_irq_save(flags); + cpu = raw_smp_processor_id(); + data = tr->data[cpu]; + disabled = atomic_inc_return(&data->disabled); + + if (likely(disabled == 1)) { + pc = preempt_count(); + trace_function(tr, ip, parent_ip, flags, pc); + } + + atomic_dec(&data->disabled); + local_irq_restore(flags); } -static void function_trace_start(struct trace_array *tr) +static void +function_stack_trace_call(unsigned long ip, unsigned long parent_ip) { - tracing_reset_online_cpus(tr); + struct trace_array *tr = func_trace; + struct trace_array_cpu *data; + unsigned long flags; + long disabled; + int cpu; + int pc; + + if (unlikely(!ftrace_function_enabled)) + return; + + /* + * Need to use raw, since this must be called before the + * recursive protection is performed. + */ + local_irq_save(flags); + cpu = raw_smp_processor_id(); + data = tr->data[cpu]; + disabled = atomic_inc_return(&data->disabled); + + if (likely(disabled == 1)) { + pc = preempt_count(); + trace_function(tr, ip, parent_ip, flags, pc); + /* + * skip over 5 funcs: + * __ftrace_trace_stack, + * __trace_stack, + * function_stack_trace_call + * ftrace_list_func + * ftrace_call + */ + __trace_stack(tr, flags, 5, pc); + } + + atomic_dec(&data->disabled); + local_irq_restore(flags); +} + + +static struct ftrace_ops trace_ops __read_mostly = +{ + .func = function_trace_call, +}; + +static struct ftrace_ops trace_stack_ops __read_mostly = +{ + .func = function_stack_trace_call, +}; + +/* Our two options */ +enum { + TRACE_FUNC_OPT_STACK = 0x1, +}; + +static struct tracer_opt func_opts[] = { +#ifdef CONFIG_STACKTRACE + { TRACER_OPT(func_stack_trace, TRACE_FUNC_OPT_STACK) }, +#endif + { } /* Always set a last empty entry */ +}; + +static struct tracer_flags func_flags = { + .val = 0, /* By default: all flags disabled */ + .opts = func_opts +}; + +static void tracing_start_function_trace(void) +{ + ftrace_function_enabled = 0; + + if (trace_flags & TRACE_ITER_PREEMPTONLY) + trace_ops.func = function_trace_call_preempt_only; + else + trace_ops.func = function_trace_call; + + if (func_flags.val & TRACE_FUNC_OPT_STACK) + register_ftrace_function(&trace_stack_ops); + else + register_ftrace_function(&trace_ops); + + ftrace_function_enabled = 1; +} + +static void tracing_stop_function_trace(void) +{ + ftrace_function_enabled = 0; + + if (func_flags.val & TRACE_FUNC_OPT_STACK) + unregister_ftrace_function(&trace_stack_ops); + else + unregister_ftrace_function(&trace_ops); +} + +static int func_set_flag(u32 old_flags, u32 bit, int set) +{ + if (bit == TRACE_FUNC_OPT_STACK) { + /* do nothing if already set */ + if (!!set == !!(func_flags.val & TRACE_FUNC_OPT_STACK)) + return 0; + + if (set) { + unregister_ftrace_function(&trace_ops); + register_ftrace_function(&trace_stack_ops); + } else { + unregister_ftrace_function(&trace_stack_ops); + register_ftrace_function(&trace_ops); + } + + return 0; + } + + return -EINVAL; } static struct tracer function_trace __read_mostly = { - .name = "function", - .init = function_trace_init, - .reset = function_trace_reset, - .start = function_trace_start, + .name = "function", + .init = function_trace_init, + .reset = function_trace_reset, + .start = function_trace_start, + .wait_pipe = poll_wait_pipe, + .flags = &func_flags, + .set_flag = func_set_flag, #ifdef CONFIG_FTRACE_SELFTEST - .selftest = trace_selftest_startup_function, + .selftest = trace_selftest_startup_function, #endif }; +#ifdef CONFIG_DYNAMIC_FTRACE +static void +ftrace_traceon(unsigned long ip, unsigned long parent_ip, void **data) +{ + long *count = (long *)data; + + if (tracing_is_on()) + return; + + if (!*count) + return; + + if (*count != -1) + (*count)--; + + tracing_on(); +} + +static void +ftrace_traceoff(unsigned long ip, unsigned long parent_ip, void **data) +{ + long *count = (long *)data; + + if (!tracing_is_on()) + return; + + if (!*count) + return; + + if (*count != -1) + (*count)--; + + tracing_off(); +} + +static int +ftrace_trace_onoff_print(struct seq_file *m, unsigned long ip, + struct ftrace_probe_ops *ops, void *data); + +static struct ftrace_probe_ops traceon_probe_ops = { + .func = ftrace_traceon, + .print = ftrace_trace_onoff_print, +}; + +static struct ftrace_probe_ops traceoff_probe_ops = { + .func = ftrace_traceoff, + .print = ftrace_trace_onoff_print, +}; + +static int +ftrace_trace_onoff_print(struct seq_file *m, unsigned long ip, + struct ftrace_probe_ops *ops, void *data) +{ + long count = (long)data; + + seq_printf(m, "%ps:", (void *)ip); + + if (ops == &traceon_probe_ops) + seq_printf(m, "traceon"); + else + seq_printf(m, "traceoff"); + + if (count == -1) + seq_printf(m, ":unlimited\n"); + else + seq_printf(m, ":count=%ld\n", count); + + return 0; +} + +static int +ftrace_trace_onoff_unreg(char *glob, char *cmd, char *param) +{ + struct ftrace_probe_ops *ops; + + /* we register both traceon and traceoff to this callback */ + if (strcmp(cmd, "traceon") == 0) + ops = &traceon_probe_ops; + else + ops = &traceoff_probe_ops; + + unregister_ftrace_function_probe_func(glob, ops); + + return 0; +} + +static int +ftrace_trace_onoff_callback(char *glob, char *cmd, char *param, int enable) +{ + struct ftrace_probe_ops *ops; + void *count = (void *)-1; + char *number; + int ret; + + /* hash funcs only work with set_ftrace_filter */ + if (!enable) + return -EINVAL; + + if (glob[0] == '!') + return ftrace_trace_onoff_unreg(glob+1, cmd, param); + + /* we register both traceon and traceoff to this callback */ + if (strcmp(cmd, "traceon") == 0) + ops = &traceon_probe_ops; + else + ops = &traceoff_probe_ops; + + if (!param) + goto out_reg; + + number = strsep(¶m, ":"); + + if (!strlen(number)) + goto out_reg; + + /* + * We use the callback data field (which is a pointer) + * as our counter. + */ + ret = strict_strtoul(number, 0, (unsigned long *)&count); + if (ret) + return ret; + + out_reg: + ret = register_ftrace_function_probe(glob, ops, count); + + return ret < 0 ? ret : 0; +} + +static struct ftrace_func_command ftrace_traceon_cmd = { + .name = "traceon", + .func = ftrace_trace_onoff_callback, +}; + +static struct ftrace_func_command ftrace_traceoff_cmd = { + .name = "traceoff", + .func = ftrace_trace_onoff_callback, +}; + +static int __init init_func_cmd_traceon(void) +{ + int ret; + + ret = register_ftrace_command(&ftrace_traceoff_cmd); + if (ret) + return ret; + + ret = register_ftrace_command(&ftrace_traceon_cmd); + if (ret) + unregister_ftrace_command(&ftrace_traceoff_cmd); + return ret; +} +#else +static inline int init_func_cmd_traceon(void) +{ + return 0; +} +#endif /* CONFIG_DYNAMIC_FTRACE */ + static __init int init_function_trace(void) { + init_func_cmd_traceon(); return register_tracer(&function_trace); } - device_initcall(init_function_trace); + diff --git a/kernel/trace/trace_functions_graph.c b/kernel/trace/trace_functions_graph.c index dce71a5b51b..45e6c01b2e4 100644 --- a/kernel/trace/trace_functions_graph.c +++ b/kernel/trace/trace_functions_graph.c @@ -1,7 +1,7 @@ /* * * Function graph tracer. - * Copyright (c) 2008 Frederic Weisbecker <fweisbec@gmail.com> + * Copyright (c) 2008-2009 Frederic Weisbecker <fweisbec@gmail.com> * Mostly borrowed from function tracer which * is Copyright (c) Steven Rostedt <srostedt@redhat.com> * @@ -12,6 +12,12 @@ #include <linux/fs.h> #include "trace.h" +#include "trace_output.h" + +struct fgraph_data { + pid_t last_pid; + int depth; +}; #define TRACE_GRAPH_INDENT 2 @@ -20,9 +26,11 @@ #define TRACE_GRAPH_PRINT_CPU 0x2 #define TRACE_GRAPH_PRINT_OVERHEAD 0x4 #define TRACE_GRAPH_PRINT_PROC 0x8 +#define TRACE_GRAPH_PRINT_DURATION 0x10 +#define TRACE_GRAPH_PRINT_ABS_TIME 0X20 static struct tracer_opt trace_opts[] = { - /* Display overruns ? */ + /* Display overruns? (for self-debug purpose) */ { TRACER_OPT(funcgraph-overrun, TRACE_GRAPH_PRINT_OVERRUN) }, /* Display CPU ? */ { TRACER_OPT(funcgraph-cpu, TRACE_GRAPH_PRINT_CPU) }, @@ -30,47 +38,64 @@ static struct tracer_opt trace_opts[] = { { TRACER_OPT(funcgraph-overhead, TRACE_GRAPH_PRINT_OVERHEAD) }, /* Display proc name/pid */ { TRACER_OPT(funcgraph-proc, TRACE_GRAPH_PRINT_PROC) }, + /* Display duration of execution */ + { TRACER_OPT(funcgraph-duration, TRACE_GRAPH_PRINT_DURATION) }, + /* Display absolute time of an entry */ + { TRACER_OPT(funcgraph-abstime, TRACE_GRAPH_PRINT_ABS_TIME) }, { } /* Empty entry */ }; static struct tracer_flags tracer_flags = { /* Don't display overruns and proc by default */ - .val = TRACE_GRAPH_PRINT_CPU | TRACE_GRAPH_PRINT_OVERHEAD, + .val = TRACE_GRAPH_PRINT_CPU | TRACE_GRAPH_PRINT_OVERHEAD | + TRACE_GRAPH_PRINT_DURATION, .opts = trace_opts }; -/* pid on the last trace processed */ -static pid_t last_pid[NR_CPUS] = { [0 ... NR_CPUS-1] = -1 }; +static struct trace_array *graph_array; + /* Add a function return address to the trace stack on thread info.*/ int -ftrace_push_return_trace(unsigned long ret, unsigned long long time, - unsigned long func, int *depth) +ftrace_push_return_trace(unsigned long ret, unsigned long func, int *depth, + unsigned long frame_pointer) { + unsigned long long calltime; int index; if (!current->ret_stack) return -EBUSY; + /* + * We must make sure the ret_stack is tested before we read + * anything else. + */ + smp_rmb(); + /* The return trace stack is full */ if (current->curr_ret_stack == FTRACE_RETFUNC_DEPTH - 1) { atomic_inc(¤t->trace_overrun); return -EBUSY; } + calltime = trace_clock_local(); + index = ++current->curr_ret_stack; barrier(); current->ret_stack[index].ret = ret; current->ret_stack[index].func = func; - current->ret_stack[index].calltime = time; + current->ret_stack[index].calltime = calltime; + current->ret_stack[index].subtime = 0; + current->ret_stack[index].fp = frame_pointer; *depth = index; return 0; } /* Retrieve a function return address to the trace stack on thread info.*/ -void -ftrace_pop_return_trace(struct ftrace_graph_ret *trace, unsigned long *ret) +static void +ftrace_pop_return_trace(struct ftrace_graph_ret *trace, unsigned long *ret, + unsigned long frame_pointer) { int index; @@ -84,28 +109,52 @@ ftrace_pop_return_trace(struct ftrace_graph_ret *trace, unsigned long *ret) return; } +#ifdef CONFIG_HAVE_FUNCTION_GRAPH_FP_TEST + /* + * The arch may choose to record the frame pointer used + * and check it here to make sure that it is what we expect it + * to be. If gcc does not set the place holder of the return + * address in the frame pointer, and does a copy instead, then + * the function graph trace will fail. This test detects this + * case. + * + * Currently, x86_32 with optimize for size (-Os) makes the latest + * gcc do the above. + */ + if (unlikely(current->ret_stack[index].fp != frame_pointer)) { + ftrace_graph_stop(); + WARN(1, "Bad frame pointer: expected %lx, received %lx\n" + " from func %ps return to %lx\n", + current->ret_stack[index].fp, + frame_pointer, + (void *)current->ret_stack[index].func, + current->ret_stack[index].ret); + *ret = (unsigned long)panic; + return; + } +#endif + *ret = current->ret_stack[index].ret; trace->func = current->ret_stack[index].func; trace->calltime = current->ret_stack[index].calltime; trace->overrun = atomic_read(¤t->trace_overrun); trace->depth = index; - barrier(); - current->curr_ret_stack--; - } /* * Send the trace to the ring-buffer. * @return the original return address. */ -unsigned long ftrace_return_to_handler(void) +unsigned long ftrace_return_to_handler(unsigned long frame_pointer) { struct ftrace_graph_ret trace; unsigned long ret; - ftrace_pop_return_trace(&trace, &ret); - trace.rettime = cpu_clock(raw_smp_processor_id()); + ftrace_pop_return_trace(&trace, &ret, frame_pointer); + trace.rettime = trace_clock_local(); ftrace_graph_return(&trace); + barrier(); + current->curr_ret_stack--; if (unlikely(!ret)) { ftrace_graph_stop(); @@ -117,15 +166,123 @@ unsigned long ftrace_return_to_handler(void) return ret; } -static int graph_trace_init(struct trace_array *tr) +static int __trace_graph_entry(struct trace_array *tr, + struct ftrace_graph_ent *trace, + unsigned long flags, + int pc) +{ + struct ftrace_event_call *call = &event_funcgraph_entry; + struct ring_buffer_event *event; + struct ring_buffer *buffer = tr->buffer; + struct ftrace_graph_ent_entry *entry; + + if (unlikely(local_read(&__get_cpu_var(ftrace_cpu_disabled)))) + return 0; + + event = trace_buffer_lock_reserve(buffer, TRACE_GRAPH_ENT, + sizeof(*entry), flags, pc); + if (!event) + return 0; + entry = ring_buffer_event_data(event); + entry->graph_ent = *trace; + if (!filter_current_check_discard(buffer, call, entry, event)) + ring_buffer_unlock_commit(buffer, event); + + return 1; +} + +int trace_graph_entry(struct ftrace_graph_ent *trace) +{ + struct trace_array *tr = graph_array; + struct trace_array_cpu *data; + unsigned long flags; + long disabled; + int ret; + int cpu; + int pc; + + if (unlikely(!tr)) + return 0; + + if (!ftrace_trace_task(current)) + return 0; + + if (!ftrace_graph_addr(trace->func)) + return 0; + + local_irq_save(flags); + cpu = raw_smp_processor_id(); + data = tr->data[cpu]; + disabled = atomic_inc_return(&data->disabled); + if (likely(disabled == 1)) { + pc = preempt_count(); + ret = __trace_graph_entry(tr, trace, flags, pc); + } else { + ret = 0; + } + /* Only do the atomic if it is not already set */ + if (!test_tsk_trace_graph(current)) + set_tsk_trace_graph(current); + + atomic_dec(&data->disabled); + local_irq_restore(flags); + + return ret; +} + +static void __trace_graph_return(struct trace_array *tr, + struct ftrace_graph_ret *trace, + unsigned long flags, + int pc) +{ + struct ftrace_event_call *call = &event_funcgraph_exit; + struct ring_buffer_event *event; + struct ring_buffer *buffer = tr->buffer; + struct ftrace_graph_ret_entry *entry; + + if (unlikely(local_read(&__get_cpu_var(ftrace_cpu_disabled)))) + return; + + event = trace_buffer_lock_reserve(buffer, TRACE_GRAPH_RET, + sizeof(*entry), flags, pc); + if (!event) + return; + entry = ring_buffer_event_data(event); + entry->ret = *trace; + if (!filter_current_check_discard(buffer, call, entry, event)) + ring_buffer_unlock_commit(buffer, event); +} + +void trace_graph_return(struct ftrace_graph_ret *trace) { - int cpu, ret; + struct trace_array *tr = graph_array; + struct trace_array_cpu *data; + unsigned long flags; + long disabled; + int cpu; + int pc; + + local_irq_save(flags); + cpu = raw_smp_processor_id(); + data = tr->data[cpu]; + disabled = atomic_inc_return(&data->disabled); + if (likely(disabled == 1)) { + pc = preempt_count(); + __trace_graph_return(tr, trace, flags, pc); + } + if (!trace->depth) + clear_tsk_trace_graph(current); + atomic_dec(&data->disabled); + local_irq_restore(flags); +} - for_each_online_cpu(cpu) - tracing_reset(tr, cpu); +static int graph_trace_init(struct trace_array *tr) +{ + int ret; + graph_array = tr; ret = register_ftrace_graph(&trace_graph_return, - &trace_graph_entry); + &trace_graph_entry); if (ret) return ret; tracing_start_cmdline_record(); @@ -133,49 +290,30 @@ static int graph_trace_init(struct trace_array *tr) return 0; } +void set_graph_array(struct trace_array *tr) +{ + graph_array = tr; +} + static void graph_trace_reset(struct trace_array *tr) { tracing_stop_cmdline_record(); unregister_ftrace_graph(); } -static inline int log10_cpu(int nb) -{ - if (nb / 100) - return 3; - if (nb / 10) - return 2; - return 1; -} +static int max_bytes_for_cpu; static enum print_line_t print_graph_cpu(struct trace_seq *s, int cpu) { - int i; int ret; - int log10_this = log10_cpu(cpu); - int log10_all = log10_cpu(cpumask_weight(cpu_online_mask)); - /* * Start with a space character - to make it stand out * to the right a bit when trace output is pasted into * email: */ - ret = trace_seq_printf(s, " "); - - /* - * Tricky - we space the CPU field according to the max - * number of online CPUs. On a 2-cpu system it would take - * a maximum of 1 digit - on a 128 cpu system it would - * take up to 3 digits: - */ - for (i = 0; i < log10_all - log10_this; i++) { - ret = trace_seq_printf(s, " "); - if (!ret) - return TRACE_TYPE_PARTIAL_LINE; - } - ret = trace_seq_printf(s, "%d) ", cpu); + ret = trace_seq_printf(s, " %*d) ", max_bytes_for_cpu, cpu); if (!ret) return TRACE_TYPE_PARTIAL_LINE; @@ -187,15 +325,15 @@ print_graph_cpu(struct trace_seq *s, int cpu) static enum print_line_t print_graph_proc(struct trace_seq *s, pid_t pid) { - int i; - int ret; - int len; - char comm[8]; - int spaces = 0; + char comm[TASK_COMM_LEN]; /* sign + log10(MAX_INT) + '\0' */ char pid_str[11]; + int spaces = 0; + int ret; + int len; + int i; - strncpy(comm, trace_find_cmdline(pid), 7); + trace_find_cmdline(pid, comm); comm[7] = '\0'; sprintf(pid_str, "%d", pid); @@ -226,19 +364,36 @@ print_graph_proc(struct trace_seq *s, pid_t pid) } +static enum print_line_t +print_graph_lat_fmt(struct trace_seq *s, struct trace_entry *entry) +{ + if (!trace_seq_putc(s, ' ')) + return 0; + + return trace_print_lat_fmt(s, entry); +} + /* If the pid changed since the last trace, output this event */ static enum print_line_t -verif_pid(struct trace_seq *s, pid_t pid, int cpu) +verif_pid(struct trace_seq *s, pid_t pid, int cpu, struct fgraph_data *data) { pid_t prev_pid; + pid_t *last_pid; int ret; - if (last_pid[cpu] != -1 && last_pid[cpu] == pid) + if (!data) return TRACE_TYPE_HANDLED; - prev_pid = last_pid[cpu]; - last_pid[cpu] = pid; + last_pid = &(per_cpu_ptr(data, cpu)->last_pid); + + if (*last_pid == pid) + return TRACE_TYPE_HANDLED; + prev_pid = *last_pid; + *last_pid = pid; + + if (prev_pid == -1) + return TRACE_TYPE_HANDLED; /* * Context-switch trace line: @@ -250,34 +405,34 @@ verif_pid(struct trace_seq *s, pid_t pid, int cpu) ret = trace_seq_printf(s, " ------------------------------------------\n"); if (!ret) - TRACE_TYPE_PARTIAL_LINE; + return TRACE_TYPE_PARTIAL_LINE; ret = print_graph_cpu(s, cpu); if (ret == TRACE_TYPE_PARTIAL_LINE) - TRACE_TYPE_PARTIAL_LINE; + return TRACE_TYPE_PARTIAL_LINE; ret = print_graph_proc(s, prev_pid); if (ret == TRACE_TYPE_PARTIAL_LINE) - TRACE_TYPE_PARTIAL_LINE; + return TRACE_TYPE_PARTIAL_LINE; ret = trace_seq_printf(s, " => "); if (!ret) - TRACE_TYPE_PARTIAL_LINE; + return TRACE_TYPE_PARTIAL_LINE; ret = print_graph_proc(s, pid); if (ret == TRACE_TYPE_PARTIAL_LINE) - TRACE_TYPE_PARTIAL_LINE; + return TRACE_TYPE_PARTIAL_LINE; ret = trace_seq_printf(s, "\n ------------------------------------------\n\n"); if (!ret) - TRACE_TYPE_PARTIAL_LINE; + return TRACE_TYPE_PARTIAL_LINE; - return ret; + return TRACE_TYPE_HANDLED; } -static bool -trace_branch_is_leaf(struct trace_iterator *iter, +static struct ftrace_graph_ret_entry * +get_return_for_leaf(struct trace_iterator *iter, struct ftrace_graph_ent_entry *curr) { struct ring_buffer_iter *ring_iter; @@ -286,72 +441,131 @@ trace_branch_is_leaf(struct trace_iterator *iter, ring_iter = iter->buffer_iter[iter->cpu]; - if (!ring_iter) - return false; - - event = ring_buffer_iter_peek(ring_iter, NULL); + /* First peek to compare current entry and the next one */ + if (ring_iter) + event = ring_buffer_iter_peek(ring_iter, NULL); + else { + /* We need to consume the current entry to see the next one */ + ring_buffer_consume(iter->tr->buffer, iter->cpu, NULL); + event = ring_buffer_peek(iter->tr->buffer, iter->cpu, + NULL); + } if (!event) - return false; + return NULL; next = ring_buffer_event_data(event); if (next->ent.type != TRACE_GRAPH_RET) - return false; + return NULL; if (curr->ent.pid != next->ent.pid || curr->graph_ent.func != next->ret.func) - return false; + return NULL; + + /* this is a leaf, now advance the iterator */ + if (ring_iter) + ring_buffer_read(ring_iter, NULL); - return true; + return next; +} + +/* Signal a overhead of time execution to the output */ +static int +print_graph_overhead(unsigned long long duration, struct trace_seq *s) +{ + /* If duration disappear, we don't need anything */ + if (!(tracer_flags.val & TRACE_GRAPH_PRINT_DURATION)) + return 1; + + /* Non nested entry or return */ + if (duration == -1) + return trace_seq_printf(s, " "); + + if (tracer_flags.val & TRACE_GRAPH_PRINT_OVERHEAD) { + /* Duration exceeded 100 msecs */ + if (duration > 100000ULL) + return trace_seq_printf(s, "! "); + + /* Duration exceeded 10 msecs */ + if (duration > 10000ULL) + return trace_seq_printf(s, "+ "); + } + + return trace_seq_printf(s, " "); +} + +static int print_graph_abs_time(u64 t, struct trace_seq *s) +{ + unsigned long usecs_rem; + + usecs_rem = do_div(t, NSEC_PER_SEC); + usecs_rem /= 1000; + + return trace_seq_printf(s, "%5lu.%06lu | ", + (unsigned long)t, usecs_rem); } static enum print_line_t -print_graph_irq(struct trace_seq *s, unsigned long addr, - enum trace_type type, int cpu, pid_t pid) +print_graph_irq(struct trace_iterator *iter, unsigned long addr, + enum trace_type type, int cpu, pid_t pid) { int ret; + struct trace_seq *s = &iter->seq; if (addr < (unsigned long)__irqentry_text_start || addr >= (unsigned long)__irqentry_text_end) return TRACE_TYPE_UNHANDLED; - if (type == TRACE_GRAPH_ENT) { - ret = trace_seq_printf(s, "==========> | "); - } else { - /* Cpu */ - if (tracer_flags.val & TRACE_GRAPH_PRINT_CPU) { - ret = print_graph_cpu(s, cpu); - if (ret == TRACE_TYPE_PARTIAL_LINE) - return TRACE_TYPE_PARTIAL_LINE; - } - /* Proc */ - if (tracer_flags.val & TRACE_GRAPH_PRINT_PROC) { - ret = print_graph_proc(s, pid); - if (ret == TRACE_TYPE_PARTIAL_LINE) - return TRACE_TYPE_PARTIAL_LINE; - - ret = trace_seq_printf(s, " | "); - if (!ret) - return TRACE_TYPE_PARTIAL_LINE; - } + /* Absolute time */ + if (tracer_flags.val & TRACE_GRAPH_PRINT_ABS_TIME) { + ret = print_graph_abs_time(iter->ts, s); + if (!ret) + return TRACE_TYPE_PARTIAL_LINE; + } - /* No overhead */ - if (tracer_flags.val & TRACE_GRAPH_PRINT_OVERHEAD) { - ret = trace_seq_printf(s, " "); - if (!ret) - return TRACE_TYPE_PARTIAL_LINE; - } + /* Cpu */ + if (tracer_flags.val & TRACE_GRAPH_PRINT_CPU) { + ret = print_graph_cpu(s, cpu); + if (ret == TRACE_TYPE_PARTIAL_LINE) + return TRACE_TYPE_PARTIAL_LINE; + } - ret = trace_seq_printf(s, "<========== |\n"); + /* Proc */ + if (tracer_flags.val & TRACE_GRAPH_PRINT_PROC) { + ret = print_graph_proc(s, pid); + if (ret == TRACE_TYPE_PARTIAL_LINE) + return TRACE_TYPE_PARTIAL_LINE; + ret = trace_seq_printf(s, " | "); + if (!ret) + return TRACE_TYPE_PARTIAL_LINE; } + + /* No overhead */ + ret = print_graph_overhead(-1, s); + if (!ret) + return TRACE_TYPE_PARTIAL_LINE; + + if (type == TRACE_GRAPH_ENT) + ret = trace_seq_printf(s, "==========>"); + else + ret = trace_seq_printf(s, "<=========="); + + if (!ret) + return TRACE_TYPE_PARTIAL_LINE; + + /* Don't close the duration column if haven't one */ + if (tracer_flags.val & TRACE_GRAPH_PRINT_DURATION) + trace_seq_printf(s, " |"); + ret = trace_seq_printf(s, "\n"); + if (!ret) return TRACE_TYPE_PARTIAL_LINE; return TRACE_TYPE_HANDLED; } -static enum print_line_t -print_graph_duration(unsigned long long duration, struct trace_seq *s) +enum print_line_t +trace_print_graph_duration(unsigned long long duration, struct trace_seq *s) { unsigned long nsecs_rem = do_div(duration, 1000); /* log10(ULONG_MAX) + '\0' */ @@ -363,7 +577,7 @@ print_graph_duration(unsigned long long duration, struct trace_seq *s) sprintf(msecs_str, "%lu", (unsigned long) duration); /* Print msecs */ - ret = trace_seq_printf(s, msecs_str); + ret = trace_seq_printf(s, "%s", msecs_str); if (!ret) return TRACE_TYPE_PARTIAL_LINE; @@ -388,60 +602,66 @@ print_graph_duration(unsigned long long duration, struct trace_seq *s) if (!ret) return TRACE_TYPE_PARTIAL_LINE; } - - ret = trace_seq_printf(s, "| "); - if (!ret) - return TRACE_TYPE_PARTIAL_LINE; return TRACE_TYPE_HANDLED; - } -/* Signal a overhead of time execution to the output */ -static int -print_graph_overhead(unsigned long long duration, struct trace_seq *s) +static enum print_line_t +print_graph_duration(unsigned long long duration, struct trace_seq *s) { - /* Duration exceeded 100 msecs */ - if (duration > 100000ULL) - return trace_seq_printf(s, "! "); + int ret; + + ret = trace_print_graph_duration(duration, s); + if (ret != TRACE_TYPE_HANDLED) + return ret; - /* Duration exceeded 10 msecs */ - if (duration > 10000ULL) - return trace_seq_printf(s, "+ "); + ret = trace_seq_printf(s, "| "); + if (!ret) + return TRACE_TYPE_PARTIAL_LINE; - return trace_seq_printf(s, " "); + return TRACE_TYPE_HANDLED; } /* Case of a leaf function on its call entry */ static enum print_line_t print_graph_entry_leaf(struct trace_iterator *iter, - struct ftrace_graph_ent_entry *entry, struct trace_seq *s) + struct ftrace_graph_ent_entry *entry, + struct ftrace_graph_ret_entry *ret_entry, struct trace_seq *s) { - struct ftrace_graph_ret_entry *ret_entry; + struct fgraph_data *data = iter->private; struct ftrace_graph_ret *graph_ret; - struct ring_buffer_event *event; struct ftrace_graph_ent *call; unsigned long long duration; int ret; int i; - event = ring_buffer_read(iter->buffer_iter[iter->cpu], NULL); - ret_entry = ring_buffer_event_data(event); graph_ret = &ret_entry->ret; call = &entry->graph_ent; duration = graph_ret->rettime - graph_ret->calltime; - /* Overhead */ - if (tracer_flags.val & TRACE_GRAPH_PRINT_OVERHEAD) { - ret = print_graph_overhead(duration, s); - if (!ret) - return TRACE_TYPE_PARTIAL_LINE; + if (data) { + int cpu = iter->cpu; + int *depth = &(per_cpu_ptr(data, cpu)->depth); + + /* + * Comments display at + 1 to depth. Since + * this is a leaf function, keep the comments + * equal to this depth. + */ + *depth = call->depth - 1; } - /* Duration */ - ret = print_graph_duration(duration, s); - if (ret == TRACE_TYPE_PARTIAL_LINE) + /* Overhead */ + ret = print_graph_overhead(duration, s); + if (!ret) return TRACE_TYPE_PARTIAL_LINE; + /* Duration */ + if (tracer_flags.val & TRACE_GRAPH_PRINT_DURATION) { + ret = print_graph_duration(duration, s); + if (ret == TRACE_TYPE_PARTIAL_LINE) + return TRACE_TYPE_PARTIAL_LINE; + } + /* Function */ for (i = 0; i < call->depth * TRACE_GRAPH_INDENT; i++) { ret = trace_seq_printf(s, " "); @@ -449,11 +669,7 @@ print_graph_entry_leaf(struct trace_iterator *iter, return TRACE_TYPE_PARTIAL_LINE; } - ret = seq_print_ip_sym(s, call->func, 0); - if (!ret) - return TRACE_TYPE_PARTIAL_LINE; - - ret = trace_seq_printf(s, "();\n"); + ret = trace_seq_printf(s, "%ps();\n", (void *)call->func); if (!ret) return TRACE_TYPE_PARTIAL_LINE; @@ -461,33 +677,34 @@ print_graph_entry_leaf(struct trace_iterator *iter, } static enum print_line_t -print_graph_entry_nested(struct ftrace_graph_ent_entry *entry, - struct trace_seq *s, pid_t pid, int cpu) +print_graph_entry_nested(struct trace_iterator *iter, + struct ftrace_graph_ent_entry *entry, + struct trace_seq *s, int cpu) { - int i; - int ret; struct ftrace_graph_ent *call = &entry->graph_ent; + struct fgraph_data *data = iter->private; + int ret; + int i; - /* No overhead */ - if (tracer_flags.val & TRACE_GRAPH_PRINT_OVERHEAD) { - ret = trace_seq_printf(s, " "); - if (!ret) - return TRACE_TYPE_PARTIAL_LINE; + if (data) { + int cpu = iter->cpu; + int *depth = &(per_cpu_ptr(data, cpu)->depth); + + *depth = call->depth; } - /* Interrupt */ - ret = print_graph_irq(s, call->func, TRACE_GRAPH_ENT, cpu, pid); - if (ret == TRACE_TYPE_UNHANDLED) { - /* No time */ + /* No overhead */ + ret = print_graph_overhead(-1, s); + if (!ret) + return TRACE_TYPE_PARTIAL_LINE; + + /* No time */ + if (tracer_flags.val & TRACE_GRAPH_PRINT_DURATION) { ret = trace_seq_printf(s, " | "); if (!ret) return TRACE_TYPE_PARTIAL_LINE; - } else { - if (ret == TRACE_TYPE_PARTIAL_LINE) - return TRACE_TYPE_PARTIAL_LINE; } - /* Function */ for (i = 0; i < call->depth * TRACE_GRAPH_INDENT; i++) { ret = trace_seq_printf(s, " "); @@ -495,28 +712,44 @@ print_graph_entry_nested(struct ftrace_graph_ent_entry *entry, return TRACE_TYPE_PARTIAL_LINE; } - ret = seq_print_ip_sym(s, call->func, 0); - if (!ret) - return TRACE_TYPE_PARTIAL_LINE; - - ret = trace_seq_printf(s, "() {\n"); + ret = trace_seq_printf(s, "%ps() {\n", (void *)call->func); if (!ret) return TRACE_TYPE_PARTIAL_LINE; - return TRACE_TYPE_HANDLED; + /* + * we already consumed the current entry to check the next one + * and see if this is a leaf. + */ + return TRACE_TYPE_NO_CONSUME; } static enum print_line_t -print_graph_entry(struct ftrace_graph_ent_entry *field, struct trace_seq *s, - struct trace_iterator *iter, int cpu) +print_graph_prologue(struct trace_iterator *iter, struct trace_seq *s, + int type, unsigned long addr) { - int ret; + struct fgraph_data *data = iter->private; struct trace_entry *ent = iter->ent; + int cpu = iter->cpu; + int ret; /* Pid */ - if (verif_pid(s, ent->pid, cpu) == TRACE_TYPE_PARTIAL_LINE) + if (verif_pid(s, ent->pid, cpu, data) == TRACE_TYPE_PARTIAL_LINE) return TRACE_TYPE_PARTIAL_LINE; + if (type) { + /* Interrupt */ + ret = print_graph_irq(iter, addr, type, cpu, ent->pid); + if (ret == TRACE_TYPE_PARTIAL_LINE) + return TRACE_TYPE_PARTIAL_LINE; + } + + /* Absolute time */ + if (tracer_flags.val & TRACE_GRAPH_PRINT_ABS_TIME) { + ret = print_graph_abs_time(iter->ts, s); + if (!ret) + return TRACE_TYPE_PARTIAL_LINE; + } + /* Cpu */ if (tracer_flags.val & TRACE_GRAPH_PRINT_CPU) { ret = print_graph_cpu(s, cpu); @@ -535,54 +768,72 @@ print_graph_entry(struct ftrace_graph_ent_entry *field, struct trace_seq *s, return TRACE_TYPE_PARTIAL_LINE; } - if (trace_branch_is_leaf(iter, field)) - return print_graph_entry_leaf(iter, field, s); + /* Latency format */ + if (trace_flags & TRACE_ITER_LATENCY_FMT) { + ret = print_graph_lat_fmt(s, ent); + if (ret == TRACE_TYPE_PARTIAL_LINE) + return TRACE_TYPE_PARTIAL_LINE; + } + + return 0; +} + +static enum print_line_t +print_graph_entry(struct ftrace_graph_ent_entry *field, struct trace_seq *s, + struct trace_iterator *iter) +{ + int cpu = iter->cpu; + struct ftrace_graph_ent *call = &field->graph_ent; + struct ftrace_graph_ret_entry *leaf_ret; + + if (print_graph_prologue(iter, s, TRACE_GRAPH_ENT, call->func)) + return TRACE_TYPE_PARTIAL_LINE; + + leaf_ret = get_return_for_leaf(iter, field); + if (leaf_ret) + return print_graph_entry_leaf(iter, field, leaf_ret, s); else - return print_graph_entry_nested(field, s, iter->ent->pid, cpu); + return print_graph_entry_nested(iter, field, s, cpu); } static enum print_line_t print_graph_return(struct ftrace_graph_ret *trace, struct trace_seq *s, - struct trace_entry *ent, int cpu) + struct trace_entry *ent, struct trace_iterator *iter) { - int i; - int ret; unsigned long long duration = trace->rettime - trace->calltime; + struct fgraph_data *data = iter->private; + pid_t pid = ent->pid; + int cpu = iter->cpu; + int ret; + int i; - /* Pid */ - if (verif_pid(s, ent->pid, cpu) == TRACE_TYPE_PARTIAL_LINE) - return TRACE_TYPE_PARTIAL_LINE; + if (data) { + int cpu = iter->cpu; + int *depth = &(per_cpu_ptr(data, cpu)->depth); - /* Cpu */ - if (tracer_flags.val & TRACE_GRAPH_PRINT_CPU) { - ret = print_graph_cpu(s, cpu); - if (ret == TRACE_TYPE_PARTIAL_LINE) - return TRACE_TYPE_PARTIAL_LINE; + /* + * Comments display at + 1 to depth. This is the + * return from a function, we now want the comments + * to display at the same level of the bracket. + */ + *depth = trace->depth - 1; } - /* Proc */ - if (tracer_flags.val & TRACE_GRAPH_PRINT_PROC) { - ret = print_graph_proc(s, ent->pid); - if (ret == TRACE_TYPE_PARTIAL_LINE) - return TRACE_TYPE_PARTIAL_LINE; - - ret = trace_seq_printf(s, " | "); - if (!ret) - return TRACE_TYPE_PARTIAL_LINE; - } + if (print_graph_prologue(iter, s, 0, 0)) + return TRACE_TYPE_PARTIAL_LINE; /* Overhead */ - if (tracer_flags.val & TRACE_GRAPH_PRINT_OVERHEAD) { - ret = print_graph_overhead(duration, s); - if (!ret) - return TRACE_TYPE_PARTIAL_LINE; - } + ret = print_graph_overhead(duration, s); + if (!ret) + return TRACE_TYPE_PARTIAL_LINE; /* Duration */ - ret = print_graph_duration(duration, s); - if (ret == TRACE_TYPE_PARTIAL_LINE) - return TRACE_TYPE_PARTIAL_LINE; + if (tracer_flags.val & TRACE_GRAPH_PRINT_DURATION) { + ret = print_graph_duration(duration, s); + if (ret == TRACE_TYPE_PARTIAL_LINE) + return TRACE_TYPE_PARTIAL_LINE; + } /* Closing brace */ for (i = 0; i < trace->depth * TRACE_GRAPH_INDENT; i++) { @@ -603,7 +854,7 @@ print_graph_return(struct ftrace_graph_ret *trace, struct trace_seq *s, return TRACE_TYPE_PARTIAL_LINE; } - ret = print_graph_irq(s, trace->func, TRACE_GRAPH_RET, cpu, ent->pid); + ret = print_graph_irq(iter, trace->func, TRACE_GRAPH_RET, cpu, pid); if (ret == TRACE_TYPE_PARTIAL_LINE) return TRACE_TYPE_PARTIAL_LINE; @@ -611,61 +862,73 @@ print_graph_return(struct ftrace_graph_ret *trace, struct trace_seq *s, } static enum print_line_t -print_graph_comment(struct print_entry *trace, struct trace_seq *s, - struct trace_entry *ent, struct trace_iterator *iter) +print_graph_comment(struct trace_seq *s, struct trace_entry *ent, + struct trace_iterator *iter) { - int i; + unsigned long sym_flags = (trace_flags & TRACE_ITER_SYM_MASK); + struct fgraph_data *data = iter->private; + struct trace_event *event; + int depth = 0; int ret; + int i; - /* Pid */ - if (verif_pid(s, ent->pid, iter->cpu) == TRACE_TYPE_PARTIAL_LINE) - return TRACE_TYPE_PARTIAL_LINE; + if (data) + depth = per_cpu_ptr(data, iter->cpu)->depth; - /* Cpu */ - if (tracer_flags.val & TRACE_GRAPH_PRINT_CPU) { - ret = print_graph_cpu(s, iter->cpu); - if (ret == TRACE_TYPE_PARTIAL_LINE) - return TRACE_TYPE_PARTIAL_LINE; - } - - /* Proc */ - if (tracer_flags.val & TRACE_GRAPH_PRINT_PROC) { - ret = print_graph_proc(s, ent->pid); - if (ret == TRACE_TYPE_PARTIAL_LINE) - return TRACE_TYPE_PARTIAL_LINE; - - ret = trace_seq_printf(s, " | "); - if (!ret) - return TRACE_TYPE_PARTIAL_LINE; - } + if (print_graph_prologue(iter, s, 0, 0)) + return TRACE_TYPE_PARTIAL_LINE; /* No overhead */ - if (tracer_flags.val & TRACE_GRAPH_PRINT_OVERHEAD) { - ret = trace_seq_printf(s, " "); + ret = print_graph_overhead(-1, s); + if (!ret) + return TRACE_TYPE_PARTIAL_LINE; + + /* No time */ + if (tracer_flags.val & TRACE_GRAPH_PRINT_DURATION) { + ret = trace_seq_printf(s, " | "); if (!ret) return TRACE_TYPE_PARTIAL_LINE; } - /* No time */ - ret = trace_seq_printf(s, " | "); - if (!ret) - return TRACE_TYPE_PARTIAL_LINE; - /* Indentation */ - if (trace->depth > 0) - for (i = 0; i < (trace->depth + 1) * TRACE_GRAPH_INDENT; i++) { + if (depth > 0) + for (i = 0; i < (depth + 1) * TRACE_GRAPH_INDENT; i++) { ret = trace_seq_printf(s, " "); if (!ret) return TRACE_TYPE_PARTIAL_LINE; } /* The comment */ - ret = trace_seq_printf(s, "/* %s", trace->buf); + ret = trace_seq_printf(s, "/* "); if (!ret) return TRACE_TYPE_PARTIAL_LINE; - if (ent->flags & TRACE_FLAG_CONT) - trace_seq_print_cont(s, iter); + switch (iter->ent->type) { + case TRACE_BPRINT: + ret = trace_print_bprintk_msg_only(iter); + if (ret != TRACE_TYPE_HANDLED) + return ret; + break; + case TRACE_PRINT: + ret = trace_print_printk_msg_only(iter); + if (ret != TRACE_TYPE_HANDLED) + return ret; + break; + default: + event = ftrace_find_event(ent->type); + if (!event) + return TRACE_TYPE_UNHANDLED; + + ret = event->trace(iter, sym_flags); + if (ret != TRACE_TYPE_HANDLED) + return ret; + } + + /* Strip ending newline */ + if (s->buffer[s->len - 1] == '\n') { + s->buffer[s->len - 1] = '\0'; + s->len--; + } ret = trace_seq_printf(s, " */\n"); if (!ret) @@ -678,66 +941,135 @@ print_graph_comment(struct print_entry *trace, struct trace_seq *s, enum print_line_t print_graph_function(struct trace_iterator *iter) { - struct trace_seq *s = &iter->seq; struct trace_entry *entry = iter->ent; + struct trace_seq *s = &iter->seq; switch (entry->type) { case TRACE_GRAPH_ENT: { - struct ftrace_graph_ent_entry *field; + /* + * print_graph_entry() may consume the current event, + * thus @field may become invalid, so we need to save it. + * sizeof(struct ftrace_graph_ent_entry) is very small, + * it can be safely saved at the stack. + */ + struct ftrace_graph_ent_entry *field, saved; trace_assign_type(field, entry); - return print_graph_entry(field, s, iter, - iter->cpu); + saved = *field; + return print_graph_entry(&saved, s, iter); } case TRACE_GRAPH_RET: { struct ftrace_graph_ret_entry *field; trace_assign_type(field, entry); - return print_graph_return(&field->ret, s, entry, iter->cpu); - } - case TRACE_PRINT: { - struct print_entry *field; - trace_assign_type(field, entry); - return print_graph_comment(field, s, entry, iter); + return print_graph_return(&field->ret, s, entry, iter); } default: - return TRACE_TYPE_UNHANDLED; + return print_graph_comment(s, entry, iter); } + + return TRACE_TYPE_HANDLED; +} + +static void print_lat_header(struct seq_file *s) +{ + static const char spaces[] = " " /* 16 spaces */ + " " /* 4 spaces */ + " "; /* 17 spaces */ + int size = 0; + + if (tracer_flags.val & TRACE_GRAPH_PRINT_ABS_TIME) + size += 16; + if (tracer_flags.val & TRACE_GRAPH_PRINT_CPU) + size += 4; + if (tracer_flags.val & TRACE_GRAPH_PRINT_PROC) + size += 17; + + seq_printf(s, "#%.*s _-----=> irqs-off \n", size, spaces); + seq_printf(s, "#%.*s / _----=> need-resched \n", size, spaces); + seq_printf(s, "#%.*s| / _---=> hardirq/softirq \n", size, spaces); + seq_printf(s, "#%.*s|| / _--=> preempt-depth \n", size, spaces); + seq_printf(s, "#%.*s||| / _-=> lock-depth \n", size, spaces); + seq_printf(s, "#%.*s|||| / \n", size, spaces); } static void print_graph_headers(struct seq_file *s) { + int lat = trace_flags & TRACE_ITER_LATENCY_FMT; + + if (lat) + print_lat_header(s); + /* 1st line */ - seq_printf(s, "# "); + seq_printf(s, "#"); + if (tracer_flags.val & TRACE_GRAPH_PRINT_ABS_TIME) + seq_printf(s, " TIME "); if (tracer_flags.val & TRACE_GRAPH_PRINT_CPU) - seq_printf(s, "CPU "); + seq_printf(s, " CPU"); if (tracer_flags.val & TRACE_GRAPH_PRINT_PROC) - seq_printf(s, "TASK/PID "); - if (tracer_flags.val & TRACE_GRAPH_PRINT_OVERHEAD) - seq_printf(s, "OVERHEAD/"); - seq_printf(s, "DURATION FUNCTION CALLS\n"); + seq_printf(s, " TASK/PID "); + if (lat) + seq_printf(s, "|||||"); + if (tracer_flags.val & TRACE_GRAPH_PRINT_DURATION) + seq_printf(s, " DURATION "); + seq_printf(s, " FUNCTION CALLS\n"); /* 2nd line */ - seq_printf(s, "# "); + seq_printf(s, "#"); + if (tracer_flags.val & TRACE_GRAPH_PRINT_ABS_TIME) + seq_printf(s, " | "); if (tracer_flags.val & TRACE_GRAPH_PRINT_CPU) - seq_printf(s, "| "); + seq_printf(s, " | "); if (tracer_flags.val & TRACE_GRAPH_PRINT_PROC) - seq_printf(s, "| | "); - if (tracer_flags.val & TRACE_GRAPH_PRINT_OVERHEAD) { - seq_printf(s, "| "); - seq_printf(s, "| | | | |\n"); - } else - seq_printf(s, " | | | | |\n"); + seq_printf(s, " | | "); + if (lat) + seq_printf(s, "|||||"); + if (tracer_flags.val & TRACE_GRAPH_PRINT_DURATION) + seq_printf(s, " | | "); + seq_printf(s, " | | | |\n"); } + +static void graph_trace_open(struct trace_iterator *iter) +{ + /* pid and depth on the last trace processed */ + struct fgraph_data *data = alloc_percpu(struct fgraph_data); + int cpu; + + if (!data) + pr_warning("function graph tracer: not enough memory\n"); + else + for_each_possible_cpu(cpu) { + pid_t *pid = &(per_cpu_ptr(data, cpu)->last_pid); + int *depth = &(per_cpu_ptr(data, cpu)->depth); + *pid = -1; + *depth = 0; + } + + iter->private = data; +} + +static void graph_trace_close(struct trace_iterator *iter) +{ + free_percpu(iter->private); +} + static struct tracer graph_trace __read_mostly = { - .name = "function_graph", - .init = graph_trace_init, - .reset = graph_trace_reset, + .name = "function_graph", + .open = graph_trace_open, + .close = graph_trace_close, + .wait_pipe = poll_wait_pipe, + .init = graph_trace_init, + .reset = graph_trace_reset, .print_line = print_graph_function, .print_header = print_graph_headers, .flags = &tracer_flags, +#ifdef CONFIG_FTRACE_SELFTEST + .selftest = trace_selftest_startup_function_graph, +#endif }; static __init int init_graph_trace(void) { + max_bytes_for_cpu = snprintf(NULL, 0, "%d", nr_cpu_ids - 1); + return register_tracer(&graph_trace); } diff --git a/kernel/trace/trace_hw_branches.c b/kernel/trace/trace_hw_branches.c index 649df22d435..69543a905cd 100644 --- a/kernel/trace/trace_hw_branches.c +++ b/kernel/trace/trace_hw_branches.c @@ -1,89 +1,147 @@ /* - * h/w branch tracer for x86 based on bts - * - * Copyright (C) 2008 Markus Metzger <markus.t.metzger@gmail.com> + * h/w branch tracer for x86 based on BTS * + * Copyright (C) 2008-2009 Intel Corporation. + * Markus Metzger <markus.t.metzger@gmail.com>, 2008-2009 */ - -#include <linux/module.h> -#include <linux/fs.h> +#include <linux/kallsyms.h> #include <linux/debugfs.h> #include <linux/ftrace.h> -#include <linux/kallsyms.h> +#include <linux/module.h> +#include <linux/cpu.h> +#include <linux/smp.h> +#include <linux/fs.h> #include <asm/ds.h> +#include "trace_output.h" #include "trace.h" -#define SIZEOF_BTS (1 << 13) +#define BTS_BUFFER_SIZE (1 << 13) static DEFINE_PER_CPU(struct bts_tracer *, tracer); -static DEFINE_PER_CPU(unsigned char[SIZEOF_BTS], buffer); +static DEFINE_PER_CPU(unsigned char[BTS_BUFFER_SIZE], buffer); #define this_tracer per_cpu(tracer, smp_processor_id()) -#define this_buffer per_cpu(buffer, smp_processor_id()) +static int trace_hw_branches_enabled __read_mostly; +static int trace_hw_branches_suspended __read_mostly; +static struct trace_array *hw_branch_trace __read_mostly; -static void bts_trace_start_cpu(void *arg) + +static void bts_trace_init_cpu(int cpu) { - if (this_tracer) - ds_release_bts(this_tracer); + per_cpu(tracer, cpu) = + ds_request_bts_cpu(cpu, per_cpu(buffer, cpu), BTS_BUFFER_SIZE, + NULL, (size_t)-1, BTS_KERNEL); - this_tracer = - ds_request_bts(/* task = */ NULL, this_buffer, SIZEOF_BTS, - /* ovfl = */ NULL, /* th = */ (size_t)-1, - BTS_KERNEL); - if (IS_ERR(this_tracer)) { - this_tracer = NULL; - return; - } + if (IS_ERR(per_cpu(tracer, cpu))) + per_cpu(tracer, cpu) = NULL; } -static void bts_trace_start(struct trace_array *tr) +static int bts_trace_init(struct trace_array *tr) { int cpu; - tracing_reset_online_cpus(tr); + hw_branch_trace = tr; + trace_hw_branches_enabled = 0; + + get_online_cpus(); + for_each_online_cpu(cpu) { + bts_trace_init_cpu(cpu); + + if (likely(per_cpu(tracer, cpu))) + trace_hw_branches_enabled = 1; + } + trace_hw_branches_suspended = 0; + put_online_cpus(); - for_each_cpu(cpu, cpu_possible_mask) - smp_call_function_single(cpu, bts_trace_start_cpu, NULL, 1); + /* If we could not enable tracing on a single cpu, we fail. */ + return trace_hw_branches_enabled ? 0 : -EOPNOTSUPP; } -static void bts_trace_stop_cpu(void *arg) +static void bts_trace_reset(struct trace_array *tr) { - if (this_tracer) { - ds_release_bts(this_tracer); - this_tracer = NULL; + int cpu; + + get_online_cpus(); + for_each_online_cpu(cpu) { + if (likely(per_cpu(tracer, cpu))) { + ds_release_bts(per_cpu(tracer, cpu)); + per_cpu(tracer, cpu) = NULL; + } } + trace_hw_branches_enabled = 0; + trace_hw_branches_suspended = 0; + put_online_cpus(); +} + +static void bts_trace_start(struct trace_array *tr) +{ + int cpu; + + get_online_cpus(); + for_each_online_cpu(cpu) + if (likely(per_cpu(tracer, cpu))) + ds_resume_bts(per_cpu(tracer, cpu)); + trace_hw_branches_suspended = 0; + put_online_cpus(); } static void bts_trace_stop(struct trace_array *tr) { int cpu; - for_each_cpu(cpu, cpu_possible_mask) - smp_call_function_single(cpu, bts_trace_stop_cpu, NULL, 1); + get_online_cpus(); + for_each_online_cpu(cpu) + if (likely(per_cpu(tracer, cpu))) + ds_suspend_bts(per_cpu(tracer, cpu)); + trace_hw_branches_suspended = 1; + put_online_cpus(); } -static int bts_trace_init(struct trace_array *tr) +static int __cpuinit bts_hotcpu_handler(struct notifier_block *nfb, + unsigned long action, void *hcpu) { - tracing_reset_online_cpus(tr); - bts_trace_start(tr); + int cpu = (long)hcpu; + + switch (action) { + case CPU_ONLINE: + case CPU_DOWN_FAILED: + /* The notification is sent with interrupts enabled. */ + if (trace_hw_branches_enabled) { + bts_trace_init_cpu(cpu); + + if (trace_hw_branches_suspended && + likely(per_cpu(tracer, cpu))) + ds_suspend_bts(per_cpu(tracer, cpu)); + } + break; - return 0; + case CPU_DOWN_PREPARE: + /* The notification is sent with interrupts enabled. */ + if (likely(per_cpu(tracer, cpu))) { + ds_release_bts(per_cpu(tracer, cpu)); + per_cpu(tracer, cpu) = NULL; + } + } + + return NOTIFY_DONE; } +static struct notifier_block bts_hotcpu_notifier __cpuinitdata = { + .notifier_call = bts_hotcpu_handler +}; + static void bts_trace_print_header(struct seq_file *m) { - seq_puts(m, - "# CPU# FROM TO FUNCTION\n"); - seq_puts(m, - "# | | | |\n"); + seq_puts(m, "# CPU# TO <- FROM\n"); } static enum print_line_t bts_trace_print_line(struct trace_iterator *iter) { + unsigned long symflags = TRACE_ITER_SYM_OFFSET; struct trace_entry *entry = iter->ent; struct trace_seq *seq = &iter->seq; struct hw_branch_entry *it; @@ -91,38 +149,57 @@ static enum print_line_t bts_trace_print_line(struct trace_iterator *iter) trace_assign_type(it, entry); if (entry->type == TRACE_HW_BRANCHES) { - if (trace_seq_printf(seq, "%4d ", entry->cpu) && - trace_seq_printf(seq, "0x%016llx -> 0x%016llx ", - it->from, it->to) && - (!it->from || - seq_print_ip_sym(seq, it->from, /* sym_flags = */ 0)) && + if (trace_seq_printf(seq, "%4d ", iter->cpu) && + seq_print_ip_sym(seq, it->to, symflags) && + trace_seq_printf(seq, "\t <- ") && + seq_print_ip_sym(seq, it->from, symflags) && trace_seq_printf(seq, "\n")) return TRACE_TYPE_HANDLED; - return TRACE_TYPE_PARTIAL_LINE;; + return TRACE_TYPE_PARTIAL_LINE; } return TRACE_TYPE_UNHANDLED; } -void trace_hw_branch(struct trace_array *tr, u64 from, u64 to) +void trace_hw_branch(u64 from, u64 to) { + struct ftrace_event_call *call = &event_hw_branch; + struct trace_array *tr = hw_branch_trace; struct ring_buffer_event *event; + struct ring_buffer *buf; struct hw_branch_entry *entry; - unsigned long irq; + unsigned long irq1; + int cpu; - event = ring_buffer_lock_reserve(tr->buffer, sizeof(*entry), &irq); - if (!event) + if (unlikely(!tr)) + return; + + if (unlikely(!trace_hw_branches_enabled)) return; + + local_irq_save(irq1); + cpu = raw_smp_processor_id(); + if (atomic_inc_return(&tr->data[cpu]->disabled) != 1) + goto out; + + buf = tr->buffer; + event = trace_buffer_lock_reserve(buf, TRACE_HW_BRANCHES, + sizeof(*entry), 0, 0); + if (!event) + goto out; entry = ring_buffer_event_data(event); tracing_generic_entry_update(&entry->ent, 0, from); entry->ent.type = TRACE_HW_BRANCHES; - entry->ent.cpu = smp_processor_id(); entry->from = from; entry->to = to; - ring_buffer_unlock_commit(tr->buffer, event, irq); + if (!filter_check_discard(call, entry, buf, event)) + trace_buffer_unlock_commit(buf, event, 0, 0); + + out: + atomic_dec(&tr->data[cpu]->disabled); + local_irq_restore(irq1); } -static void trace_bts_at(struct trace_array *tr, - const struct bts_trace *trace, void *at) +static void trace_bts_at(const struct bts_trace *trace, void *at) { struct bts_struct bts; int err = 0; @@ -137,59 +214,98 @@ static void trace_bts_at(struct trace_array *tr, switch (bts.qualifier) { case BTS_BRANCH: - trace_hw_branch(tr, bts.variant.lbr.from, bts.variant.lbr.to); + trace_hw_branch(bts.variant.lbr.from, bts.variant.lbr.to); break; } } +/* + * Collect the trace on the current cpu and write it into the ftrace buffer. + * + * pre: tracing must be suspended on the current cpu + */ static void trace_bts_cpu(void *arg) { - struct trace_array *tr = (struct trace_array *) arg; + struct trace_array *tr = (struct trace_array *)arg; const struct bts_trace *trace; unsigned char *at; - if (!this_tracer) + if (unlikely(!tr)) + return; + + if (unlikely(atomic_read(&tr->data[raw_smp_processor_id()]->disabled))) + return; + + if (unlikely(!this_tracer)) return; - ds_suspend_bts(this_tracer); trace = ds_read_bts(this_tracer); if (!trace) - goto out; + return; for (at = trace->ds.top; (void *)at < trace->ds.end; at += trace->ds.size) - trace_bts_at(tr, trace, at); + trace_bts_at(trace, at); for (at = trace->ds.begin; (void *)at < trace->ds.top; at += trace->ds.size) - trace_bts_at(tr, trace, at); - -out: - ds_resume_bts(this_tracer); + trace_bts_at(trace, at); } static void trace_bts_prepare(struct trace_iterator *iter) { int cpu; - for_each_cpu(cpu, cpu_possible_mask) - smp_call_function_single(cpu, trace_bts_cpu, iter->tr, 1); + get_online_cpus(); + for_each_online_cpu(cpu) + if (likely(per_cpu(tracer, cpu))) + ds_suspend_bts(per_cpu(tracer, cpu)); + /* + * We need to collect the trace on the respective cpu since ftrace + * implicitly adds the record for the current cpu. + * Once that is more flexible, we could collect the data from any cpu. + */ + on_each_cpu(trace_bts_cpu, iter->tr, 1); + + for_each_online_cpu(cpu) + if (likely(per_cpu(tracer, cpu))) + ds_resume_bts(per_cpu(tracer, cpu)); + put_online_cpus(); +} + +static void trace_bts_close(struct trace_iterator *iter) +{ + tracing_reset_online_cpus(iter->tr); +} + +void trace_hw_branch_oops(void) +{ + if (this_tracer) { + ds_suspend_bts_noirq(this_tracer); + trace_bts_cpu(hw_branch_trace); + ds_resume_bts_noirq(this_tracer); + } } struct tracer bts_tracer __read_mostly = { .name = "hw-branch-tracer", .init = bts_trace_init, - .reset = bts_trace_stop, + .reset = bts_trace_reset, .print_header = bts_trace_print_header, .print_line = bts_trace_print_line, .start = bts_trace_start, .stop = bts_trace_stop, - .open = trace_bts_prepare + .open = trace_bts_prepare, + .close = trace_bts_close, +#ifdef CONFIG_FTRACE_SELFTEST + .selftest = trace_selftest_startup_hw_branches, +#endif /* CONFIG_FTRACE_SELFTEST */ }; __init static int init_bts_trace(void) { + register_hotcpu_notifier(&bts_hotcpu_notifier); return register_tracer(&bts_tracer); } device_initcall(init_bts_trace); diff --git a/kernel/trace/trace_irqsoff.c b/kernel/trace/trace_irqsoff.c index 62a78d94353..3aa7eaa2114 100644 --- a/kernel/trace/trace_irqsoff.c +++ b/kernel/trace/trace_irqsoff.c @@ -1,5 +1,5 @@ /* - * trace irqs off criticall timings + * trace irqs off critical timings * * Copyright (C) 2007-2008 Steven Rostedt <srostedt@redhat.com> * Copyright (C) 2008 Ingo Molnar <mingo@redhat.com> @@ -32,6 +32,8 @@ enum { static int trace_type __read_mostly; +static int save_lat_flag; + #ifdef CONFIG_PREEMPT_TRACER static inline int preempt_trace(void) @@ -95,7 +97,7 @@ irqsoff_tracer_call(unsigned long ip, unsigned long parent_ip) disabled = atomic_inc_return(&data->disabled); if (likely(disabled == 1)) - trace_function(tr, data, ip, parent_ip, flags, preempt_count()); + trace_function(tr, ip, parent_ip, flags, preempt_count()); atomic_dec(&data->disabled); } @@ -127,15 +129,10 @@ check_critical_timing(struct trace_array *tr, unsigned long parent_ip, int cpu) { - unsigned long latency, t0, t1; cycle_t T0, T1, delta; unsigned long flags; int pc; - /* - * usecs conversion is slow so we try to delay the conversion - * as long as possible: - */ T0 = data->preempt_timestamp; T1 = ftrace_now(cpu); delta = T1-T0; @@ -153,20 +150,17 @@ check_critical_timing(struct trace_array *tr, if (!report_latency(delta)) goto out_unlock; - trace_function(tr, data, CALLER_ADDR0, parent_ip, flags, pc); - - latency = nsecs_to_usecs(delta); + trace_function(tr, CALLER_ADDR0, parent_ip, flags, pc); if (data->critical_sequence != max_sequence) goto out_unlock; - tracing_max_latency = delta; - t0 = nsecs_to_usecs(T0); - t1 = nsecs_to_usecs(T1); - data->critical_end = parent_ip; - update_max_tr_single(tr, current, cpu); + if (likely(!is_tracing_stopped())) { + tracing_max_latency = delta; + update_max_tr_single(tr, current, cpu); + } max_sequence++; @@ -176,8 +170,7 @@ out_unlock: out: data->critical_sequence = max_sequence; data->preempt_timestamp = ftrace_now(cpu); - tracing_reset(tr, cpu); - trace_function(tr, data, CALLER_ADDR0, parent_ip, flags, pc); + trace_function(tr, CALLER_ADDR0, parent_ip, flags, pc); } static inline void @@ -206,11 +199,10 @@ start_critical_timing(unsigned long ip, unsigned long parent_ip) data->critical_sequence = max_sequence; data->preempt_timestamp = ftrace_now(cpu); data->critical_start = parent_ip ? : ip; - tracing_reset(tr, cpu); local_save_flags(flags); - trace_function(tr, data, ip, parent_ip, flags, preempt_count()); + trace_function(tr, ip, parent_ip, flags, preempt_count()); per_cpu(tracing_cpu, cpu) = 1; @@ -244,7 +236,7 @@ stop_critical_timing(unsigned long ip, unsigned long parent_ip) atomic_inc(&data->disabled); local_save_flags(flags); - trace_function(tr, data, ip, parent_ip, flags, preempt_count()); + trace_function(tr, ip, parent_ip, flags, preempt_count()); check_critical_timing(tr, data, parent_ip ? : ip, cpu); data->critical_start = 0; atomic_dec(&data->disabled); @@ -353,67 +345,50 @@ void trace_preempt_off(unsigned long a0, unsigned long a1) } #endif /* CONFIG_PREEMPT_TRACER */ -/* - * save_tracer_enabled is used to save the state of the tracer_enabled - * variable when we disable it when we open a trace output file. - */ -static int save_tracer_enabled; - static void start_irqsoff_tracer(struct trace_array *tr) { register_ftrace_function(&trace_ops); - if (tracing_is_enabled()) { + if (tracing_is_enabled()) tracer_enabled = 1; - save_tracer_enabled = 1; - } else { + else tracer_enabled = 0; - save_tracer_enabled = 0; - } } static void stop_irqsoff_tracer(struct trace_array *tr) { tracer_enabled = 0; - save_tracer_enabled = 0; unregister_ftrace_function(&trace_ops); } static void __irqsoff_tracer_init(struct trace_array *tr) { + save_lat_flag = trace_flags & TRACE_ITER_LATENCY_FMT; + trace_flags |= TRACE_ITER_LATENCY_FMT; + tracing_max_latency = 0; irqsoff_trace = tr; /* make sure that the tracer is visible */ smp_wmb(); + tracing_reset_online_cpus(tr); start_irqsoff_tracer(tr); } static void irqsoff_tracer_reset(struct trace_array *tr) { stop_irqsoff_tracer(tr); + + if (!save_lat_flag) + trace_flags &= ~TRACE_ITER_LATENCY_FMT; } static void irqsoff_tracer_start(struct trace_array *tr) { tracer_enabled = 1; - save_tracer_enabled = 1; } static void irqsoff_tracer_stop(struct trace_array *tr) { tracer_enabled = 0; - save_tracer_enabled = 0; -} - -static void irqsoff_tracer_open(struct trace_iterator *iter) -{ - /* stop the trace while dumping */ - tracer_enabled = 0; -} - -static void irqsoff_tracer_close(struct trace_iterator *iter) -{ - /* restart tracing */ - tracer_enabled = save_tracer_enabled; } #ifdef CONFIG_IRQSOFF_TRACER @@ -431,8 +406,6 @@ static struct tracer irqsoff_tracer __read_mostly = .reset = irqsoff_tracer_reset, .start = irqsoff_tracer_start, .stop = irqsoff_tracer_stop, - .open = irqsoff_tracer_open, - .close = irqsoff_tracer_close, .print_max = 1, #ifdef CONFIG_FTRACE_SELFTEST .selftest = trace_selftest_startup_irqsoff, @@ -459,8 +432,6 @@ static struct tracer preemptoff_tracer __read_mostly = .reset = irqsoff_tracer_reset, .start = irqsoff_tracer_start, .stop = irqsoff_tracer_stop, - .open = irqsoff_tracer_open, - .close = irqsoff_tracer_close, .print_max = 1, #ifdef CONFIG_FTRACE_SELFTEST .selftest = trace_selftest_startup_preemptoff, @@ -489,8 +460,6 @@ static struct tracer preemptirqsoff_tracer __read_mostly = .reset = irqsoff_tracer_reset, .start = irqsoff_tracer_start, .stop = irqsoff_tracer_stop, - .open = irqsoff_tracer_open, - .close = irqsoff_tracer_close, .print_max = 1, #ifdef CONFIG_FTRACE_SELFTEST .selftest = trace_selftest_startup_preemptirqsoff, diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c new file mode 100644 index 00000000000..aff5f80b59b --- /dev/null +++ b/kernel/trace/trace_kprobe.c @@ -0,0 +1,1523 @@ +/* + * Kprobes-based tracing events + * + * Created by Masami Hiramatsu <mhiramat@redhat.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + */ + +#include <linux/module.h> +#include <linux/uaccess.h> +#include <linux/kprobes.h> +#include <linux/seq_file.h> +#include <linux/slab.h> +#include <linux/smp.h> +#include <linux/debugfs.h> +#include <linux/types.h> +#include <linux/string.h> +#include <linux/ctype.h> +#include <linux/ptrace.h> +#include <linux/perf_event.h> + +#include "trace.h" +#include "trace_output.h" + +#define MAX_TRACE_ARGS 128 +#define MAX_ARGSTR_LEN 63 +#define MAX_EVENT_NAME_LEN 64 +#define KPROBE_EVENT_SYSTEM "kprobes" + +/* Reserved field names */ +#define FIELD_STRING_IP "__probe_ip" +#define FIELD_STRING_NARGS "__probe_nargs" +#define FIELD_STRING_RETIP "__probe_ret_ip" +#define FIELD_STRING_FUNC "__probe_func" + +const char *reserved_field_names[] = { + "common_type", + "common_flags", + "common_preempt_count", + "common_pid", + "common_tgid", + "common_lock_depth", + FIELD_STRING_IP, + FIELD_STRING_NARGS, + FIELD_STRING_RETIP, + FIELD_STRING_FUNC, +}; + +struct fetch_func { + unsigned long (*func)(struct pt_regs *, void *); + void *data; +}; + +static __kprobes unsigned long call_fetch(struct fetch_func *f, + struct pt_regs *regs) +{ + return f->func(regs, f->data); +} + +/* fetch handlers */ +static __kprobes unsigned long fetch_register(struct pt_regs *regs, + void *offset) +{ + return regs_get_register(regs, (unsigned int)((unsigned long)offset)); +} + +static __kprobes unsigned long fetch_stack(struct pt_regs *regs, + void *num) +{ + return regs_get_kernel_stack_nth(regs, + (unsigned int)((unsigned long)num)); +} + +static __kprobes unsigned long fetch_memory(struct pt_regs *regs, void *addr) +{ + unsigned long retval; + + if (probe_kernel_address(addr, retval)) + return 0; + return retval; +} + +static __kprobes unsigned long fetch_argument(struct pt_regs *regs, void *num) +{ + return regs_get_argument_nth(regs, (unsigned int)((unsigned long)num)); +} + +static __kprobes unsigned long fetch_retvalue(struct pt_regs *regs, + void *dummy) +{ + return regs_return_value(regs); +} + +static __kprobes unsigned long fetch_stack_address(struct pt_regs *regs, + void *dummy) +{ + return kernel_stack_pointer(regs); +} + +/* Memory fetching by symbol */ +struct symbol_cache { + char *symbol; + long offset; + unsigned long addr; +}; + +static unsigned long update_symbol_cache(struct symbol_cache *sc) +{ + sc->addr = (unsigned long)kallsyms_lookup_name(sc->symbol); + if (sc->addr) + sc->addr += sc->offset; + return sc->addr; +} + +static void free_symbol_cache(struct symbol_cache *sc) +{ + kfree(sc->symbol); + kfree(sc); +} + +static struct symbol_cache *alloc_symbol_cache(const char *sym, long offset) +{ + struct symbol_cache *sc; + + if (!sym || strlen(sym) == 0) + return NULL; + sc = kzalloc(sizeof(struct symbol_cache), GFP_KERNEL); + if (!sc) + return NULL; + + sc->symbol = kstrdup(sym, GFP_KERNEL); + if (!sc->symbol) { + kfree(sc); + return NULL; + } + sc->offset = offset; + + update_symbol_cache(sc); + return sc; +} + +static __kprobes unsigned long fetch_symbol(struct pt_regs *regs, void *data) +{ + struct symbol_cache *sc = data; + + if (sc->addr) + return fetch_memory(regs, (void *)sc->addr); + else + return 0; +} + +/* Special indirect memory access interface */ +struct indirect_fetch_data { + struct fetch_func orig; + long offset; +}; + +static __kprobes unsigned long fetch_indirect(struct pt_regs *regs, void *data) +{ + struct indirect_fetch_data *ind = data; + unsigned long addr; + + addr = call_fetch(&ind->orig, regs); + if (addr) { + addr += ind->offset; + return fetch_memory(regs, (void *)addr); + } else + return 0; +} + +static __kprobes void free_indirect_fetch_data(struct indirect_fetch_data *data) +{ + if (data->orig.func == fetch_indirect) + free_indirect_fetch_data(data->orig.data); + else if (data->orig.func == fetch_symbol) + free_symbol_cache(data->orig.data); + kfree(data); +} + +/** + * Kprobe event core functions + */ + +struct probe_arg { + struct fetch_func fetch; + const char *name; +}; + +/* Flags for trace_probe */ +#define TP_FLAG_TRACE 1 +#define TP_FLAG_PROFILE 2 + +struct trace_probe { + struct list_head list; + struct kretprobe rp; /* Use rp.kp for kprobe use */ + unsigned long nhit; + unsigned int flags; /* For TP_FLAG_* */ + const char *symbol; /* symbol name */ + struct ftrace_event_call call; + struct trace_event event; + unsigned int nr_args; + struct probe_arg args[]; +}; + +#define SIZEOF_TRACE_PROBE(n) \ + (offsetof(struct trace_probe, args) + \ + (sizeof(struct probe_arg) * (n))) + +static __kprobes int probe_is_return(struct trace_probe *tp) +{ + return tp->rp.handler != NULL; +} + +static __kprobes const char *probe_symbol(struct trace_probe *tp) +{ + return tp->symbol ? tp->symbol : "unknown"; +} + +static int probe_arg_string(char *buf, size_t n, struct fetch_func *ff) +{ + int ret = -EINVAL; + + if (ff->func == fetch_argument) + ret = snprintf(buf, n, "$arg%lu", (unsigned long)ff->data); + else if (ff->func == fetch_register) { + const char *name; + name = regs_query_register_name((unsigned int)((long)ff->data)); + ret = snprintf(buf, n, "%%%s", name); + } else if (ff->func == fetch_stack) + ret = snprintf(buf, n, "$stack%lu", (unsigned long)ff->data); + else if (ff->func == fetch_memory) + ret = snprintf(buf, n, "@0x%p", ff->data); + else if (ff->func == fetch_symbol) { + struct symbol_cache *sc = ff->data; + if (sc->offset) + ret = snprintf(buf, n, "@%s%+ld", sc->symbol, + sc->offset); + else + ret = snprintf(buf, n, "@%s", sc->symbol); + } else if (ff->func == fetch_retvalue) + ret = snprintf(buf, n, "$retval"); + else if (ff->func == fetch_stack_address) + ret = snprintf(buf, n, "$stack"); + else if (ff->func == fetch_indirect) { + struct indirect_fetch_data *id = ff->data; + size_t l = 0; + ret = snprintf(buf, n, "%+ld(", id->offset); + if (ret >= n) + goto end; + l += ret; + ret = probe_arg_string(buf + l, n - l, &id->orig); + if (ret < 0) + goto end; + l += ret; + ret = snprintf(buf + l, n - l, ")"); + ret += l; + } +end: + if (ret >= n) + return -ENOSPC; + return ret; +} + +static int register_probe_event(struct trace_probe *tp); +static void unregister_probe_event(struct trace_probe *tp); + +static DEFINE_MUTEX(probe_lock); +static LIST_HEAD(probe_list); + +static int kprobe_dispatcher(struct kprobe *kp, struct pt_regs *regs); +static int kretprobe_dispatcher(struct kretprobe_instance *ri, + struct pt_regs *regs); + +/* + * Allocate new trace_probe and initialize it (including kprobes). + */ +static struct trace_probe *alloc_trace_probe(const char *group, + const char *event, + void *addr, + const char *symbol, + unsigned long offs, + int nargs, int is_return) +{ + struct trace_probe *tp; + + tp = kzalloc(SIZEOF_TRACE_PROBE(nargs), GFP_KERNEL); + if (!tp) + return ERR_PTR(-ENOMEM); + + if (symbol) { + tp->symbol = kstrdup(symbol, GFP_KERNEL); + if (!tp->symbol) + goto error; + tp->rp.kp.symbol_name = tp->symbol; + tp->rp.kp.offset = offs; + } else + tp->rp.kp.addr = addr; + + if (is_return) + tp->rp.handler = kretprobe_dispatcher; + else + tp->rp.kp.pre_handler = kprobe_dispatcher; + + if (!event) + goto error; + tp->call.name = kstrdup(event, GFP_KERNEL); + if (!tp->call.name) + goto error; + + if (!group) + goto error; + tp->call.system = kstrdup(group, GFP_KERNEL); + if (!tp->call.system) + goto error; + + INIT_LIST_HEAD(&tp->list); + return tp; +error: + kfree(tp->call.name); + kfree(tp->symbol); + kfree(tp); + return ERR_PTR(-ENOMEM); +} + +static void free_probe_arg(struct probe_arg *arg) +{ + if (arg->fetch.func == fetch_symbol) + free_symbol_cache(arg->fetch.data); + else if (arg->fetch.func == fetch_indirect) + free_indirect_fetch_data(arg->fetch.data); + kfree(arg->name); +} + +static void free_trace_probe(struct trace_probe *tp) +{ + int i; + + for (i = 0; i < tp->nr_args; i++) + free_probe_arg(&tp->args[i]); + + kfree(tp->call.system); + kfree(tp->call.name); + kfree(tp->symbol); + kfree(tp); +} + +static struct trace_probe *find_probe_event(const char *event, + const char *group) +{ + struct trace_probe *tp; + + list_for_each_entry(tp, &probe_list, list) + if (strcmp(tp->call.name, event) == 0 && + strcmp(tp->call.system, group) == 0) + return tp; + return NULL; +} + +/* Unregister a trace_probe and probe_event: call with locking probe_lock */ +static void unregister_trace_probe(struct trace_probe *tp) +{ + if (probe_is_return(tp)) + unregister_kretprobe(&tp->rp); + else + unregister_kprobe(&tp->rp.kp); + list_del(&tp->list); + unregister_probe_event(tp); +} + +/* Register a trace_probe and probe_event */ +static int register_trace_probe(struct trace_probe *tp) +{ + struct trace_probe *old_tp; + int ret; + + mutex_lock(&probe_lock); + + /* register as an event */ + old_tp = find_probe_event(tp->call.name, tp->call.system); + if (old_tp) { + /* delete old event */ + unregister_trace_probe(old_tp); + free_trace_probe(old_tp); + } + ret = register_probe_event(tp); + if (ret) { + pr_warning("Faild to register probe event(%d)\n", ret); + goto end; + } + + tp->rp.kp.flags |= KPROBE_FLAG_DISABLED; + if (probe_is_return(tp)) + ret = register_kretprobe(&tp->rp); + else + ret = register_kprobe(&tp->rp.kp); + + if (ret) { + pr_warning("Could not insert probe(%d)\n", ret); + if (ret == -EILSEQ) { + pr_warning("Probing address(0x%p) is not an " + "instruction boundary.\n", + tp->rp.kp.addr); + ret = -EINVAL; + } + unregister_probe_event(tp); + } else + list_add_tail(&tp->list, &probe_list); +end: + mutex_unlock(&probe_lock); + return ret; +} + +/* Split symbol and offset. */ +static int split_symbol_offset(char *symbol, unsigned long *offset) +{ + char *tmp; + int ret; + + if (!offset) + return -EINVAL; + + tmp = strchr(symbol, '+'); + if (tmp) { + /* skip sign because strict_strtol doesn't accept '+' */ + ret = strict_strtoul(tmp + 1, 0, offset); + if (ret) + return ret; + *tmp = '\0'; + } else + *offset = 0; + return 0; +} + +#define PARAM_MAX_ARGS 16 +#define PARAM_MAX_STACK (THREAD_SIZE / sizeof(unsigned long)) + +static int parse_probe_vars(char *arg, struct fetch_func *ff, int is_return) +{ + int ret = 0; + unsigned long param; + + if (strcmp(arg, "retval") == 0) { + if (is_return) { + ff->func = fetch_retvalue; + ff->data = NULL; + } else + ret = -EINVAL; + } else if (strncmp(arg, "stack", 5) == 0) { + if (arg[5] == '\0') { + ff->func = fetch_stack_address; + ff->data = NULL; + } else if (isdigit(arg[5])) { + ret = strict_strtoul(arg + 5, 10, ¶m); + if (ret || param > PARAM_MAX_STACK) + ret = -EINVAL; + else { + ff->func = fetch_stack; + ff->data = (void *)param; + } + } else + ret = -EINVAL; + } else if (strncmp(arg, "arg", 3) == 0 && isdigit(arg[3])) { + ret = strict_strtoul(arg + 3, 10, ¶m); + if (ret || param > PARAM_MAX_ARGS) + ret = -EINVAL; + else { + ff->func = fetch_argument; + ff->data = (void *)param; + } + } else + ret = -EINVAL; + return ret; +} + +/* Recursive argument parser */ +static int __parse_probe_arg(char *arg, struct fetch_func *ff, int is_return) +{ + int ret = 0; + unsigned long param; + long offset; + char *tmp; + + switch (arg[0]) { + case '$': + ret = parse_probe_vars(arg + 1, ff, is_return); + break; + case '%': /* named register */ + ret = regs_query_register_offset(arg + 1); + if (ret >= 0) { + ff->func = fetch_register; + ff->data = (void *)(unsigned long)ret; + ret = 0; + } + break; + case '@': /* memory or symbol */ + if (isdigit(arg[1])) { + ret = strict_strtoul(arg + 1, 0, ¶m); + if (ret) + break; + ff->func = fetch_memory; + ff->data = (void *)param; + } else { + ret = split_symbol_offset(arg + 1, &offset); + if (ret) + break; + ff->data = alloc_symbol_cache(arg + 1, offset); + if (ff->data) + ff->func = fetch_symbol; + else + ret = -EINVAL; + } + break; + case '+': /* indirect memory */ + case '-': + tmp = strchr(arg, '('); + if (!tmp) { + ret = -EINVAL; + break; + } + *tmp = '\0'; + ret = strict_strtol(arg + 1, 0, &offset); + if (ret) + break; + if (arg[0] == '-') + offset = -offset; + arg = tmp + 1; + tmp = strrchr(arg, ')'); + if (tmp) { + struct indirect_fetch_data *id; + *tmp = '\0'; + id = kzalloc(sizeof(struct indirect_fetch_data), + GFP_KERNEL); + if (!id) + return -ENOMEM; + id->offset = offset; + ret = __parse_probe_arg(arg, &id->orig, is_return); + if (ret) + kfree(id); + else { + ff->func = fetch_indirect; + ff->data = (void *)id; + } + } else + ret = -EINVAL; + break; + default: + /* TODO: support custom handler */ + ret = -EINVAL; + } + return ret; +} + +/* String length checking wrapper */ +static int parse_probe_arg(char *arg, struct fetch_func *ff, int is_return) +{ + if (strlen(arg) > MAX_ARGSTR_LEN) { + pr_info("Argument is too long.: %s\n", arg); + return -ENOSPC; + } + return __parse_probe_arg(arg, ff, is_return); +} + +/* Return 1 if name is reserved or already used by another argument */ +static int conflict_field_name(const char *name, + struct probe_arg *args, int narg) +{ + int i; + for (i = 0; i < ARRAY_SIZE(reserved_field_names); i++) + if (strcmp(reserved_field_names[i], name) == 0) + return 1; + for (i = 0; i < narg; i++) + if (strcmp(args[i].name, name) == 0) + return 1; + return 0; +} + +static int create_trace_probe(int argc, char **argv) +{ + /* + * Argument syntax: + * - Add kprobe: p[:[GRP/]EVENT] KSYM[+OFFS]|KADDR [FETCHARGS] + * - Add kretprobe: r[:[GRP/]EVENT] KSYM[+0] [FETCHARGS] + * Fetch args: + * $argN : fetch Nth of function argument. (N:0-) + * $retval : fetch return value + * $stack : fetch stack address + * $stackN : fetch Nth of stack (N:0-) + * @ADDR : fetch memory at ADDR (ADDR should be in kernel) + * @SYM[+|-offs] : fetch memory at SYM +|- offs (SYM is a data symbol) + * %REG : fetch register REG + * Indirect memory fetch: + * +|-offs(ARG) : fetch memory at ARG +|- offs address. + * Alias name of args: + * NAME=FETCHARG : set NAME as alias of FETCHARG. + */ + struct trace_probe *tp; + int i, ret = 0; + int is_return = 0; + char *symbol = NULL, *event = NULL, *arg = NULL, *group = NULL; + unsigned long offset = 0; + void *addr = NULL; + char buf[MAX_EVENT_NAME_LEN]; + + if (argc < 2) { + pr_info("Probe point is not specified.\n"); + return -EINVAL; + } + + if (argv[0][0] == 'p') + is_return = 0; + else if (argv[0][0] == 'r') + is_return = 1; + else { + pr_info("Probe definition must be started with 'p' or 'r'.\n"); + return -EINVAL; + } + + if (argv[0][1] == ':') { + event = &argv[0][2]; + if (strchr(event, '/')) { + group = event; + event = strchr(group, '/') + 1; + event[-1] = '\0'; + if (strlen(group) == 0) { + pr_info("Group name is not specifiled\n"); + return -EINVAL; + } + } + if (strlen(event) == 0) { + pr_info("Event name is not specifiled\n"); + return -EINVAL; + } + } + + if (isdigit(argv[1][0])) { + if (is_return) { + pr_info("Return probe point must be a symbol.\n"); + return -EINVAL; + } + /* an address specified */ + ret = strict_strtoul(&argv[0][2], 0, (unsigned long *)&addr); + if (ret) { + pr_info("Failed to parse address.\n"); + return ret; + } + } else { + /* a symbol specified */ + symbol = argv[1]; + /* TODO: support .init module functions */ + ret = split_symbol_offset(symbol, &offset); + if (ret) { + pr_info("Failed to parse symbol.\n"); + return ret; + } + if (offset && is_return) { + pr_info("Return probe must be used without offset.\n"); + return -EINVAL; + } + } + argc -= 2; argv += 2; + + /* setup a probe */ + if (!group) + group = KPROBE_EVENT_SYSTEM; + if (!event) { + /* Make a new event name */ + if (symbol) + snprintf(buf, MAX_EVENT_NAME_LEN, "%c@%s%+ld", + is_return ? 'r' : 'p', symbol, offset); + else + snprintf(buf, MAX_EVENT_NAME_LEN, "%c@0x%p", + is_return ? 'r' : 'p', addr); + event = buf; + } + tp = alloc_trace_probe(group, event, addr, symbol, offset, argc, + is_return); + if (IS_ERR(tp)) { + pr_info("Failed to allocate trace_probe.(%d)\n", + (int)PTR_ERR(tp)); + return PTR_ERR(tp); + } + + /* parse arguments */ + ret = 0; + for (i = 0; i < argc && i < MAX_TRACE_ARGS; i++) { + /* Parse argument name */ + arg = strchr(argv[i], '='); + if (arg) + *arg++ = '\0'; + else + arg = argv[i]; + + if (conflict_field_name(argv[i], tp->args, i)) { + pr_info("Argument%d name '%s' conflicts with " + "another field.\n", i, argv[i]); + ret = -EINVAL; + goto error; + } + + tp->args[i].name = kstrdup(argv[i], GFP_KERNEL); + if (!tp->args[i].name) { + pr_info("Failed to allocate argument%d name '%s'.\n", + i, argv[i]); + ret = -ENOMEM; + goto error; + } + + /* Parse fetch argument */ + ret = parse_probe_arg(arg, &tp->args[i].fetch, is_return); + if (ret) { + pr_info("Parse error at argument%d. (%d)\n", i, ret); + kfree(tp->args[i].name); + goto error; + } + + tp->nr_args++; + } + + ret = register_trace_probe(tp); + if (ret) + goto error; + return 0; + +error: + free_trace_probe(tp); + return ret; +} + +static void cleanup_all_probes(void) +{ + struct trace_probe *tp; + + mutex_lock(&probe_lock); + /* TODO: Use batch unregistration */ + while (!list_empty(&probe_list)) { + tp = list_entry(probe_list.next, struct trace_probe, list); + unregister_trace_probe(tp); + free_trace_probe(tp); + } + mutex_unlock(&probe_lock); +} + + +/* Probes listing interfaces */ +static void *probes_seq_start(struct seq_file *m, loff_t *pos) +{ + mutex_lock(&probe_lock); + return seq_list_start(&probe_list, *pos); +} + +static void *probes_seq_next(struct seq_file *m, void *v, loff_t *pos) +{ + return seq_list_next(v, &probe_list, pos); +} + +static void probes_seq_stop(struct seq_file *m, void *v) +{ + mutex_unlock(&probe_lock); +} + +static int probes_seq_show(struct seq_file *m, void *v) +{ + struct trace_probe *tp = v; + int i, ret; + char buf[MAX_ARGSTR_LEN + 1]; + + seq_printf(m, "%c", probe_is_return(tp) ? 'r' : 'p'); + seq_printf(m, ":%s/%s", tp->call.system, tp->call.name); + + if (!tp->symbol) + seq_printf(m, " 0x%p", tp->rp.kp.addr); + else if (tp->rp.kp.offset) + seq_printf(m, " %s+%u", probe_symbol(tp), tp->rp.kp.offset); + else + seq_printf(m, " %s", probe_symbol(tp)); + + for (i = 0; i < tp->nr_args; i++) { + ret = probe_arg_string(buf, MAX_ARGSTR_LEN, &tp->args[i].fetch); + if (ret < 0) { + pr_warning("Argument%d decoding error(%d).\n", i, ret); + return ret; + } + seq_printf(m, " %s=%s", tp->args[i].name, buf); + } + seq_printf(m, "\n"); + return 0; +} + +static const struct seq_operations probes_seq_op = { + .start = probes_seq_start, + .next = probes_seq_next, + .stop = probes_seq_stop, + .show = probes_seq_show +}; + +static int probes_open(struct inode *inode, struct file *file) +{ + if ((file->f_mode & FMODE_WRITE) && + (file->f_flags & O_TRUNC)) + cleanup_all_probes(); + + return seq_open(file, &probes_seq_op); +} + +static int command_trace_probe(const char *buf) +{ + char **argv; + int argc = 0, ret = 0; + + argv = argv_split(GFP_KERNEL, buf, &argc); + if (!argv) + return -ENOMEM; + + if (argc) + ret = create_trace_probe(argc, argv); + + argv_free(argv); + return ret; +} + +#define WRITE_BUFSIZE 128 + +static ssize_t probes_write(struct file *file, const char __user *buffer, + size_t count, loff_t *ppos) +{ + char *kbuf, *tmp; + int ret; + size_t done; + size_t size; + + kbuf = kmalloc(WRITE_BUFSIZE, GFP_KERNEL); + if (!kbuf) + return -ENOMEM; + + ret = done = 0; + while (done < count) { + size = count - done; + if (size >= WRITE_BUFSIZE) + size = WRITE_BUFSIZE - 1; + if (copy_from_user(kbuf, buffer + done, size)) { + ret = -EFAULT; + goto out; + } + kbuf[size] = '\0'; + tmp = strchr(kbuf, '\n'); + if (tmp) { + *tmp = '\0'; + size = tmp - kbuf + 1; + } else if (done + size < count) { + pr_warning("Line length is too long: " + "Should be less than %d.", WRITE_BUFSIZE); + ret = -EINVAL; + goto out; + } + done += size; + /* Remove comments */ + tmp = strchr(kbuf, '#'); + if (tmp) + *tmp = '\0'; + + ret = command_trace_probe(kbuf); + if (ret) + goto out; + } + ret = done; +out: + kfree(kbuf); + return ret; +} + +static const struct file_operations kprobe_events_ops = { + .owner = THIS_MODULE, + .open = probes_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release, + .write = probes_write, +}; + +/* Probes profiling interfaces */ +static int probes_profile_seq_show(struct seq_file *m, void *v) +{ + struct trace_probe *tp = v; + + seq_printf(m, " %-44s %15lu %15lu\n", tp->call.name, tp->nhit, + tp->rp.kp.nmissed); + + return 0; +} + +static const struct seq_operations profile_seq_op = { + .start = probes_seq_start, + .next = probes_seq_next, + .stop = probes_seq_stop, + .show = probes_profile_seq_show +}; + +static int profile_open(struct inode *inode, struct file *file) +{ + return seq_open(file, &profile_seq_op); +} + +static const struct file_operations kprobe_profile_ops = { + .owner = THIS_MODULE, + .open = profile_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release, +}; + +/* Kprobe handler */ +static __kprobes int kprobe_trace_func(struct kprobe *kp, struct pt_regs *regs) +{ + struct trace_probe *tp = container_of(kp, struct trace_probe, rp.kp); + struct kprobe_trace_entry *entry; + struct ring_buffer_event *event; + struct ring_buffer *buffer; + int size, i, pc; + unsigned long irq_flags; + struct ftrace_event_call *call = &tp->call; + + tp->nhit++; + + local_save_flags(irq_flags); + pc = preempt_count(); + + size = SIZEOF_KPROBE_TRACE_ENTRY(tp->nr_args); + + event = trace_current_buffer_lock_reserve(&buffer, call->id, size, + irq_flags, pc); + if (!event) + return 0; + + entry = ring_buffer_event_data(event); + entry->nargs = tp->nr_args; + entry->ip = (unsigned long)kp->addr; + for (i = 0; i < tp->nr_args; i++) + entry->args[i] = call_fetch(&tp->args[i].fetch, regs); + + if (!filter_current_check_discard(buffer, call, entry, event)) + trace_nowake_buffer_unlock_commit(buffer, event, irq_flags, pc); + return 0; +} + +/* Kretprobe handler */ +static __kprobes int kretprobe_trace_func(struct kretprobe_instance *ri, + struct pt_regs *regs) +{ + struct trace_probe *tp = container_of(ri->rp, struct trace_probe, rp); + struct kretprobe_trace_entry *entry; + struct ring_buffer_event *event; + struct ring_buffer *buffer; + int size, i, pc; + unsigned long irq_flags; + struct ftrace_event_call *call = &tp->call; + + local_save_flags(irq_flags); + pc = preempt_count(); + + size = SIZEOF_KRETPROBE_TRACE_ENTRY(tp->nr_args); + + event = trace_current_buffer_lock_reserve(&buffer, call->id, size, + irq_flags, pc); + if (!event) + return 0; + + entry = ring_buffer_event_data(event); + entry->nargs = tp->nr_args; + entry->func = (unsigned long)tp->rp.kp.addr; + entry->ret_ip = (unsigned long)ri->ret_addr; + for (i = 0; i < tp->nr_args; i++) + entry->args[i] = call_fetch(&tp->args[i].fetch, regs); + + if (!filter_current_check_discard(buffer, call, entry, event)) + trace_nowake_buffer_unlock_commit(buffer, event, irq_flags, pc); + + return 0; +} + +/* Event entry printers */ +enum print_line_t +print_kprobe_event(struct trace_iterator *iter, int flags) +{ + struct kprobe_trace_entry *field; + struct trace_seq *s = &iter->seq; + struct trace_event *event; + struct trace_probe *tp; + int i; + + field = (struct kprobe_trace_entry *)iter->ent; + event = ftrace_find_event(field->ent.type); + tp = container_of(event, struct trace_probe, event); + + if (!trace_seq_printf(s, "%s: (", tp->call.name)) + goto partial; + + if (!seq_print_ip_sym(s, field->ip, flags | TRACE_ITER_SYM_OFFSET)) + goto partial; + + if (!trace_seq_puts(s, ")")) + goto partial; + + for (i = 0; i < field->nargs; i++) + if (!trace_seq_printf(s, " %s=%lx", + tp->args[i].name, field->args[i])) + goto partial; + + if (!trace_seq_puts(s, "\n")) + goto partial; + + return TRACE_TYPE_HANDLED; +partial: + return TRACE_TYPE_PARTIAL_LINE; +} + +enum print_line_t +print_kretprobe_event(struct trace_iterator *iter, int flags) +{ + struct kretprobe_trace_entry *field; + struct trace_seq *s = &iter->seq; + struct trace_event *event; + struct trace_probe *tp; + int i; + + field = (struct kretprobe_trace_entry *)iter->ent; + event = ftrace_find_event(field->ent.type); + tp = container_of(event, struct trace_probe, event); + + if (!trace_seq_printf(s, "%s: (", tp->call.name)) + goto partial; + + if (!seq_print_ip_sym(s, field->ret_ip, flags | TRACE_ITER_SYM_OFFSET)) + goto partial; + + if (!trace_seq_puts(s, " <- ")) + goto partial; + + if (!seq_print_ip_sym(s, field->func, flags & ~TRACE_ITER_SYM_OFFSET)) + goto partial; + + if (!trace_seq_puts(s, ")")) + goto partial; + + for (i = 0; i < field->nargs; i++) + if (!trace_seq_printf(s, " %s=%lx", + tp->args[i].name, field->args[i])) + goto partial; + + if (!trace_seq_puts(s, "\n")) + goto partial; + + return TRACE_TYPE_HANDLED; +partial: + return TRACE_TYPE_PARTIAL_LINE; +} + +static int probe_event_enable(struct ftrace_event_call *call) +{ + struct trace_probe *tp = (struct trace_probe *)call->data; + + tp->flags |= TP_FLAG_TRACE; + if (probe_is_return(tp)) + return enable_kretprobe(&tp->rp); + else + return enable_kprobe(&tp->rp.kp); +} + +static void probe_event_disable(struct ftrace_event_call *call) +{ + struct trace_probe *tp = (struct trace_probe *)call->data; + + tp->flags &= ~TP_FLAG_TRACE; + if (!(tp->flags & (TP_FLAG_TRACE | TP_FLAG_PROFILE))) { + if (probe_is_return(tp)) + disable_kretprobe(&tp->rp); + else + disable_kprobe(&tp->rp.kp); + } +} + +static int probe_event_raw_init(struct ftrace_event_call *event_call) +{ + INIT_LIST_HEAD(&event_call->fields); + + return 0; +} + +#undef DEFINE_FIELD +#define DEFINE_FIELD(type, item, name, is_signed) \ + do { \ + ret = trace_define_field(event_call, #type, name, \ + offsetof(typeof(field), item), \ + sizeof(field.item), is_signed, \ + FILTER_OTHER); \ + if (ret) \ + return ret; \ + } while (0) + +static int kprobe_event_define_fields(struct ftrace_event_call *event_call) +{ + int ret, i; + struct kprobe_trace_entry field; + struct trace_probe *tp = (struct trace_probe *)event_call->data; + + ret = trace_define_common_fields(event_call); + if (!ret) + return ret; + + DEFINE_FIELD(unsigned long, ip, FIELD_STRING_IP, 0); + DEFINE_FIELD(int, nargs, FIELD_STRING_NARGS, 1); + /* Set argument names as fields */ + for (i = 0; i < tp->nr_args; i++) + DEFINE_FIELD(unsigned long, args[i], tp->args[i].name, 0); + return 0; +} + +static int kretprobe_event_define_fields(struct ftrace_event_call *event_call) +{ + int ret, i; + struct kretprobe_trace_entry field; + struct trace_probe *tp = (struct trace_probe *)event_call->data; + + ret = trace_define_common_fields(event_call); + if (!ret) + return ret; + + DEFINE_FIELD(unsigned long, func, FIELD_STRING_FUNC, 0); + DEFINE_FIELD(unsigned long, ret_ip, FIELD_STRING_RETIP, 0); + DEFINE_FIELD(int, nargs, FIELD_STRING_NARGS, 1); + /* Set argument names as fields */ + for (i = 0; i < tp->nr_args; i++) + DEFINE_FIELD(unsigned long, args[i], tp->args[i].name, 0); + return 0; +} + +static int __probe_event_show_format(struct trace_seq *s, + struct trace_probe *tp, const char *fmt, + const char *arg) +{ + int i; + + /* Show format */ + if (!trace_seq_printf(s, "\nprint fmt: \"%s", fmt)) + return 0; + + for (i = 0; i < tp->nr_args; i++) + if (!trace_seq_printf(s, " %s=%%lx", tp->args[i].name)) + return 0; + + if (!trace_seq_printf(s, "\", %s", arg)) + return 0; + + for (i = 0; i < tp->nr_args; i++) + if (!trace_seq_printf(s, ", REC->%s", tp->args[i].name)) + return 0; + + return trace_seq_puts(s, "\n"); +} + +#undef SHOW_FIELD +#define SHOW_FIELD(type, item, name) \ + do { \ + ret = trace_seq_printf(s, "\tfield: " #type " %s;\t" \ + "offset:%u;\tsize:%u;\n", name, \ + (unsigned int)offsetof(typeof(field), item),\ + (unsigned int)sizeof(type)); \ + if (!ret) \ + return 0; \ + } while (0) + +static int kprobe_event_show_format(struct ftrace_event_call *call, + struct trace_seq *s) +{ + struct kprobe_trace_entry field __attribute__((unused)); + int ret, i; + struct trace_probe *tp = (struct trace_probe *)call->data; + + SHOW_FIELD(unsigned long, ip, FIELD_STRING_IP); + SHOW_FIELD(int, nargs, FIELD_STRING_NARGS); + + /* Show fields */ + for (i = 0; i < tp->nr_args; i++) + SHOW_FIELD(unsigned long, args[i], tp->args[i].name); + trace_seq_puts(s, "\n"); + + return __probe_event_show_format(s, tp, "(%lx)", + "REC->" FIELD_STRING_IP); +} + +static int kretprobe_event_show_format(struct ftrace_event_call *call, + struct trace_seq *s) +{ + struct kretprobe_trace_entry field __attribute__((unused)); + int ret, i; + struct trace_probe *tp = (struct trace_probe *)call->data; + + SHOW_FIELD(unsigned long, func, FIELD_STRING_FUNC); + SHOW_FIELD(unsigned long, ret_ip, FIELD_STRING_RETIP); + SHOW_FIELD(int, nargs, FIELD_STRING_NARGS); + + /* Show fields */ + for (i = 0; i < tp->nr_args; i++) + SHOW_FIELD(unsigned long, args[i], tp->args[i].name); + trace_seq_puts(s, "\n"); + + return __probe_event_show_format(s, tp, "(%lx <- %lx)", + "REC->" FIELD_STRING_FUNC + ", REC->" FIELD_STRING_RETIP); +} + +#ifdef CONFIG_EVENT_PROFILE + +/* Kprobe profile handler */ +static __kprobes int kprobe_profile_func(struct kprobe *kp, + struct pt_regs *regs) +{ + struct trace_probe *tp = container_of(kp, struct trace_probe, rp.kp); + struct ftrace_event_call *call = &tp->call; + struct kprobe_trace_entry *entry; + struct trace_entry *ent; + int size, __size, i, pc, __cpu; + unsigned long irq_flags; + char *trace_buf; + char *raw_data; + int rctx; + + pc = preempt_count(); + __size = SIZEOF_KPROBE_TRACE_ENTRY(tp->nr_args); + size = ALIGN(__size + sizeof(u32), sizeof(u64)); + size -= sizeof(u32); + if (WARN_ONCE(size > FTRACE_MAX_PROFILE_SIZE, + "profile buffer not large enough")) + return 0; + + /* + * Protect the non nmi buffer + * This also protects the rcu read side + */ + local_irq_save(irq_flags); + + rctx = perf_swevent_get_recursion_context(); + if (rctx < 0) + goto end_recursion; + + __cpu = smp_processor_id(); + + if (in_nmi()) + trace_buf = rcu_dereference(perf_trace_buf_nmi); + else + trace_buf = rcu_dereference(perf_trace_buf); + + if (!trace_buf) + goto end; + + raw_data = per_cpu_ptr(trace_buf, __cpu); + + /* Zero dead bytes from alignment to avoid buffer leak to userspace */ + *(u64 *)(&raw_data[size - sizeof(u64)]) = 0ULL; + entry = (struct kprobe_trace_entry *)raw_data; + ent = &entry->ent; + + tracing_generic_entry_update(ent, irq_flags, pc); + ent->type = call->id; + entry->nargs = tp->nr_args; + entry->ip = (unsigned long)kp->addr; + for (i = 0; i < tp->nr_args; i++) + entry->args[i] = call_fetch(&tp->args[i].fetch, regs); + perf_tp_event(call->id, entry->ip, 1, entry, size); + +end: + perf_swevent_put_recursion_context(rctx); +end_recursion: + local_irq_restore(irq_flags); + + return 0; +} + +/* Kretprobe profile handler */ +static __kprobes int kretprobe_profile_func(struct kretprobe_instance *ri, + struct pt_regs *regs) +{ + struct trace_probe *tp = container_of(ri->rp, struct trace_probe, rp); + struct ftrace_event_call *call = &tp->call; + struct kretprobe_trace_entry *entry; + struct trace_entry *ent; + int size, __size, i, pc, __cpu; + unsigned long irq_flags; + char *trace_buf; + char *raw_data; + int rctx; + + pc = preempt_count(); + __size = SIZEOF_KRETPROBE_TRACE_ENTRY(tp->nr_args); + size = ALIGN(__size + sizeof(u32), sizeof(u64)); + size -= sizeof(u32); + if (WARN_ONCE(size > FTRACE_MAX_PROFILE_SIZE, + "profile buffer not large enough")) + return 0; + + /* + * Protect the non nmi buffer + * This also protects the rcu read side + */ + local_irq_save(irq_flags); + + rctx = perf_swevent_get_recursion_context(); + if (rctx < 0) + goto end_recursion; + + __cpu = smp_processor_id(); + + if (in_nmi()) + trace_buf = rcu_dereference(perf_trace_buf_nmi); + else + trace_buf = rcu_dereference(perf_trace_buf); + + if (!trace_buf) + goto end; + + raw_data = per_cpu_ptr(trace_buf, __cpu); + + /* Zero dead bytes from alignment to avoid buffer leak to userspace */ + *(u64 *)(&raw_data[size - sizeof(u64)]) = 0ULL; + entry = (struct kretprobe_trace_entry *)raw_data; + ent = &entry->ent; + + tracing_generic_entry_update(ent, irq_flags, pc); + ent->type = call->id; + entry->nargs = tp->nr_args; + entry->func = (unsigned long)tp->rp.kp.addr; + entry->ret_ip = (unsigned long)ri->ret_addr; + for (i = 0; i < tp->nr_args; i++) + entry->args[i] = call_fetch(&tp->args[i].fetch, regs); + perf_tp_event(call->id, entry->ret_ip, 1, entry, size); + +end: + perf_swevent_put_recursion_context(rctx); +end_recursion: + local_irq_restore(irq_flags); + + return 0; +} + +static int probe_profile_enable(struct ftrace_event_call *call) +{ + struct trace_probe *tp = (struct trace_probe *)call->data; + + tp->flags |= TP_FLAG_PROFILE; + + if (probe_is_return(tp)) + return enable_kretprobe(&tp->rp); + else + return enable_kprobe(&tp->rp.kp); +} + +static void probe_profile_disable(struct ftrace_event_call *call) +{ + struct trace_probe *tp = (struct trace_probe *)call->data; + + tp->flags &= ~TP_FLAG_PROFILE; + + if (!(tp->flags & TP_FLAG_TRACE)) { + if (probe_is_return(tp)) + disable_kretprobe(&tp->rp); + else + disable_kprobe(&tp->rp.kp); + } +} +#endif /* CONFIG_EVENT_PROFILE */ + + +static __kprobes +int kprobe_dispatcher(struct kprobe *kp, struct pt_regs *regs) +{ + struct trace_probe *tp = container_of(kp, struct trace_probe, rp.kp); + + if (tp->flags & TP_FLAG_TRACE) + kprobe_trace_func(kp, regs); +#ifdef CONFIG_EVENT_PROFILE + if (tp->flags & TP_FLAG_PROFILE) + kprobe_profile_func(kp, regs); +#endif /* CONFIG_EVENT_PROFILE */ + return 0; /* We don't tweek kernel, so just return 0 */ +} + +static __kprobes +int kretprobe_dispatcher(struct kretprobe_instance *ri, struct pt_regs *regs) +{ + struct trace_probe *tp = container_of(ri->rp, struct trace_probe, rp); + + if (tp->flags & TP_FLAG_TRACE) + kretprobe_trace_func(ri, regs); +#ifdef CONFIG_EVENT_PROFILE + if (tp->flags & TP_FLAG_PROFILE) + kretprobe_profile_func(ri, regs); +#endif /* CONFIG_EVENT_PROFILE */ + return 0; /* We don't tweek kernel, so just return 0 */ +} + +static int register_probe_event(struct trace_probe *tp) +{ + struct ftrace_event_call *call = &tp->call; + int ret; + + /* Initialize ftrace_event_call */ + if (probe_is_return(tp)) { + tp->event.trace = print_kretprobe_event; + call->raw_init = probe_event_raw_init; + call->show_format = kretprobe_event_show_format; + call->define_fields = kretprobe_event_define_fields; + } else { + tp->event.trace = print_kprobe_event; + call->raw_init = probe_event_raw_init; + call->show_format = kprobe_event_show_format; + call->define_fields = kprobe_event_define_fields; + } + call->event = &tp->event; + call->id = register_ftrace_event(&tp->event); + if (!call->id) + return -ENODEV; + call->enabled = 0; + call->regfunc = probe_event_enable; + call->unregfunc = probe_event_disable; + +#ifdef CONFIG_EVENT_PROFILE + atomic_set(&call->profile_count, -1); + call->profile_enable = probe_profile_enable; + call->profile_disable = probe_profile_disable; +#endif + call->data = tp; + ret = trace_add_event_call(call); + if (ret) { + pr_info("Failed to register kprobe event: %s\n", call->name); + unregister_ftrace_event(&tp->event); + } + return ret; +} + +static void unregister_probe_event(struct trace_probe *tp) +{ + /* tp->event is unregistered in trace_remove_event_call() */ + trace_remove_event_call(&tp->call); +} + +/* Make a debugfs interface for controling probe points */ +static __init int init_kprobe_trace(void) +{ + struct dentry *d_tracer; + struct dentry *entry; + + d_tracer = tracing_init_dentry(); + if (!d_tracer) + return 0; + + entry = debugfs_create_file("kprobe_events", 0644, d_tracer, + NULL, &kprobe_events_ops); + + /* Event list interface */ + if (!entry) + pr_warning("Could not create debugfs " + "'kprobe_events' entry\n"); + + /* Profile interface */ + entry = debugfs_create_file("kprobe_profile", 0444, d_tracer, + NULL, &kprobe_profile_ops); + + if (!entry) + pr_warning("Could not create debugfs " + "'kprobe_profile' entry\n"); + return 0; +} +fs_initcall(init_kprobe_trace); + + +#ifdef CONFIG_FTRACE_STARTUP_TEST + +static int kprobe_trace_selftest_target(int a1, int a2, int a3, + int a4, int a5, int a6) +{ + return a1 + a2 + a3 + a4 + a5 + a6; +} + +static __init int kprobe_trace_self_tests_init(void) +{ + int ret; + int (*target)(int, int, int, int, int, int); + + target = kprobe_trace_selftest_target; + + pr_info("Testing kprobe tracing: "); + + ret = command_trace_probe("p:testprobe kprobe_trace_selftest_target " + "$arg1 $arg2 $arg3 $arg4 $stack $stack0"); + if (WARN_ON_ONCE(ret)) + pr_warning("error enabling function entry\n"); + + ret = command_trace_probe("r:testprobe2 kprobe_trace_selftest_target " + "$retval"); + if (WARN_ON_ONCE(ret)) + pr_warning("error enabling function return\n"); + + ret = target(1, 2, 3, 4, 5, 6); + + cleanup_all_probes(); + + pr_cont("OK\n"); + return 0; +} + +late_initcall(kprobe_trace_self_tests_init); + +#endif diff --git a/kernel/trace/trace_ksym.c b/kernel/trace/trace_ksym.c new file mode 100644 index 00000000000..ddfa0fd43bc --- /dev/null +++ b/kernel/trace/trace_ksym.c @@ -0,0 +1,550 @@ +/* + * trace_ksym.c - Kernel Symbol Tracer + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + * + * Copyright (C) IBM Corporation, 2009 + */ + +#include <linux/kallsyms.h> +#include <linux/uaccess.h> +#include <linux/debugfs.h> +#include <linux/ftrace.h> +#include <linux/module.h> +#include <linux/fs.h> + +#include "trace_output.h" +#include "trace_stat.h" +#include "trace.h" + +#include <linux/hw_breakpoint.h> +#include <asm/hw_breakpoint.h> + +/* + * For now, let us restrict the no. of symbols traced simultaneously to number + * of available hardware breakpoint registers. + */ +#define KSYM_TRACER_MAX HBP_NUM + +#define KSYM_TRACER_OP_LEN 3 /* rw- */ + +struct trace_ksym { + struct perf_event **ksym_hbp; + struct perf_event_attr attr; +#ifdef CONFIG_PROFILE_KSYM_TRACER + unsigned long counter; +#endif + struct hlist_node ksym_hlist; +}; + +static struct trace_array *ksym_trace_array; + +static unsigned int ksym_filter_entry_count; +static unsigned int ksym_tracing_enabled; + +static HLIST_HEAD(ksym_filter_head); + +static DEFINE_MUTEX(ksym_tracer_mutex); + +#ifdef CONFIG_PROFILE_KSYM_TRACER + +#define MAX_UL_INT 0xffffffff + +void ksym_collect_stats(unsigned long hbp_hit_addr) +{ + struct hlist_node *node; + struct trace_ksym *entry; + + rcu_read_lock(); + hlist_for_each_entry_rcu(entry, node, &ksym_filter_head, ksym_hlist) { + if ((entry->attr.bp_addr == hbp_hit_addr) && + (entry->counter <= MAX_UL_INT)) { + entry->counter++; + break; + } + } + rcu_read_unlock(); +} +#endif /* CONFIG_PROFILE_KSYM_TRACER */ + +void ksym_hbp_handler(struct perf_event *hbp, void *data) +{ + struct ring_buffer_event *event; + struct ksym_trace_entry *entry; + struct pt_regs *regs = data; + struct ring_buffer *buffer; + int pc; + + if (!ksym_tracing_enabled) + return; + + buffer = ksym_trace_array->buffer; + + pc = preempt_count(); + + event = trace_buffer_lock_reserve(buffer, TRACE_KSYM, + sizeof(*entry), 0, pc); + if (!event) + return; + + entry = ring_buffer_event_data(event); + entry->ip = instruction_pointer(regs); + entry->type = hw_breakpoint_type(hbp); + entry->addr = hw_breakpoint_addr(hbp); + strlcpy(entry->cmd, current->comm, TASK_COMM_LEN); + +#ifdef CONFIG_PROFILE_KSYM_TRACER + ksym_collect_stats(hw_breakpoint_addr(hbp)); +#endif /* CONFIG_PROFILE_KSYM_TRACER */ + + trace_buffer_unlock_commit(buffer, event, 0, pc); +} + +/* Valid access types are represented as + * + * rw- : Set Read/Write Access Breakpoint + * -w- : Set Write Access Breakpoint + * --- : Clear Breakpoints + * --x : Set Execution Break points (Not available yet) + * + */ +static int ksym_trace_get_access_type(char *str) +{ + int access = 0; + + if (str[0] == 'r') + access |= HW_BREAKPOINT_R; + + if (str[1] == 'w') + access |= HW_BREAKPOINT_W; + + if (str[2] == 'x') + access |= HW_BREAKPOINT_X; + + switch (access) { + case HW_BREAKPOINT_R: + case HW_BREAKPOINT_W: + case HW_BREAKPOINT_W | HW_BREAKPOINT_R: + return access; + default: + return -EINVAL; + } +} + +/* + * There can be several possible malformed requests and we attempt to capture + * all of them. We enumerate some of the rules + * 1. We will not allow kernel symbols with ':' since it is used as a delimiter. + * i.e. multiple ':' symbols disallowed. Possible uses are of the form + * <module>:<ksym_name>:<op>. + * 2. No delimiter symbol ':' in the input string + * 3. Spurious operator symbols or symbols not in their respective positions + * 4. <ksym_name>:--- i.e. clear breakpoint request when ksym_name not in file + * 5. Kernel symbol not a part of /proc/kallsyms + * 6. Duplicate requests + */ +static int parse_ksym_trace_str(char *input_string, char **ksymname, + unsigned long *addr) +{ + int ret; + + *ksymname = strsep(&input_string, ":"); + *addr = kallsyms_lookup_name(*ksymname); + + /* Check for malformed request: (2), (1) and (5) */ + if ((!input_string) || + (strlen(input_string) != KSYM_TRACER_OP_LEN) || + (*addr == 0)) + return -EINVAL;; + + ret = ksym_trace_get_access_type(input_string); + + return ret; +} + +int process_new_ksym_entry(char *ksymname, int op, unsigned long addr) +{ + struct trace_ksym *entry; + int ret = -ENOMEM; + + if (ksym_filter_entry_count >= KSYM_TRACER_MAX) { + printk(KERN_ERR "ksym_tracer: Maximum limit:(%d) reached. No" + " new requests for tracing can be accepted now.\n", + KSYM_TRACER_MAX); + return -ENOSPC; + } + + entry = kzalloc(sizeof(struct trace_ksym), GFP_KERNEL); + if (!entry) + return -ENOMEM; + + hw_breakpoint_init(&entry->attr); + + entry->attr.bp_type = op; + entry->attr.bp_addr = addr; + entry->attr.bp_len = HW_BREAKPOINT_LEN_4; + + ret = -EAGAIN; + entry->ksym_hbp = register_wide_hw_breakpoint(&entry->attr, + ksym_hbp_handler); + + if (IS_ERR(entry->ksym_hbp)) { + ret = PTR_ERR(entry->ksym_hbp); + printk(KERN_INFO "ksym_tracer request failed. Try again" + " later!!\n"); + goto err; + } + + hlist_add_head_rcu(&(entry->ksym_hlist), &ksym_filter_head); + ksym_filter_entry_count++; + + return 0; + +err: + kfree(entry); + + return ret; +} + +static ssize_t ksym_trace_filter_read(struct file *filp, char __user *ubuf, + size_t count, loff_t *ppos) +{ + struct trace_ksym *entry; + struct hlist_node *node; + struct trace_seq *s; + ssize_t cnt = 0; + int ret; + + s = kmalloc(sizeof(*s), GFP_KERNEL); + if (!s) + return -ENOMEM; + trace_seq_init(s); + + mutex_lock(&ksym_tracer_mutex); + + hlist_for_each_entry(entry, node, &ksym_filter_head, ksym_hlist) { + ret = trace_seq_printf(s, "%pS:", (void *)entry->attr.bp_addr); + if (entry->attr.bp_type == HW_BREAKPOINT_R) + ret = trace_seq_puts(s, "r--\n"); + else if (entry->attr.bp_type == HW_BREAKPOINT_W) + ret = trace_seq_puts(s, "-w-\n"); + else if (entry->attr.bp_type == (HW_BREAKPOINT_W | HW_BREAKPOINT_R)) + ret = trace_seq_puts(s, "rw-\n"); + WARN_ON_ONCE(!ret); + } + + cnt = simple_read_from_buffer(ubuf, count, ppos, s->buffer, s->len); + + mutex_unlock(&ksym_tracer_mutex); + + kfree(s); + + return cnt; +} + +static void __ksym_trace_reset(void) +{ + struct trace_ksym *entry; + struct hlist_node *node, *node1; + + mutex_lock(&ksym_tracer_mutex); + hlist_for_each_entry_safe(entry, node, node1, &ksym_filter_head, + ksym_hlist) { + unregister_wide_hw_breakpoint(entry->ksym_hbp); + ksym_filter_entry_count--; + hlist_del_rcu(&(entry->ksym_hlist)); + synchronize_rcu(); + kfree(entry); + } + mutex_unlock(&ksym_tracer_mutex); +} + +static ssize_t ksym_trace_filter_write(struct file *file, + const char __user *buffer, + size_t count, loff_t *ppos) +{ + struct trace_ksym *entry; + struct hlist_node *node; + char *input_string, *ksymname = NULL; + unsigned long ksym_addr = 0; + int ret, op, changed = 0; + + input_string = kzalloc(count + 1, GFP_KERNEL); + if (!input_string) + return -ENOMEM; + + if (copy_from_user(input_string, buffer, count)) { + kfree(input_string); + return -EFAULT; + } + input_string[count] = '\0'; + + strstrip(input_string); + + /* + * Clear all breakpoints if: + * 1: echo > ksym_trace_filter + * 2: echo 0 > ksym_trace_filter + * 3: echo "*:---" > ksym_trace_filter + */ + if (!input_string[0] || !strcmp(input_string, "0") || + !strcmp(input_string, "*:---")) { + __ksym_trace_reset(); + kfree(input_string); + return count; + } + + ret = op = parse_ksym_trace_str(input_string, &ksymname, &ksym_addr); + if (ret < 0) { + kfree(input_string); + return ret; + } + + mutex_lock(&ksym_tracer_mutex); + + ret = -EINVAL; + hlist_for_each_entry(entry, node, &ksym_filter_head, ksym_hlist) { + if (entry->attr.bp_addr == ksym_addr) { + /* Check for malformed request: (6) */ + if (entry->attr.bp_type != op) + changed = 1; + else + goto out; + break; + } + } + if (changed) { + unregister_wide_hw_breakpoint(entry->ksym_hbp); + entry->attr.bp_type = op; + ret = 0; + if (op > 0) { + entry->ksym_hbp = + register_wide_hw_breakpoint(&entry->attr, + ksym_hbp_handler); + if (IS_ERR(entry->ksym_hbp)) + ret = PTR_ERR(entry->ksym_hbp); + else + goto out; + } + /* Error or "symbol:---" case: drop it */ + ksym_filter_entry_count--; + hlist_del_rcu(&(entry->ksym_hlist)); + synchronize_rcu(); + kfree(entry); + goto out; + } else { + /* Check for malformed request: (4) */ + if (op == 0) + goto out; + ret = process_new_ksym_entry(ksymname, op, ksym_addr); + } +out: + mutex_unlock(&ksym_tracer_mutex); + + kfree(input_string); + + if (!ret) + ret = count; + return ret; +} + +static const struct file_operations ksym_tracing_fops = { + .open = tracing_open_generic, + .read = ksym_trace_filter_read, + .write = ksym_trace_filter_write, +}; + +static void ksym_trace_reset(struct trace_array *tr) +{ + ksym_tracing_enabled = 0; + __ksym_trace_reset(); +} + +static int ksym_trace_init(struct trace_array *tr) +{ + int cpu, ret = 0; + + for_each_online_cpu(cpu) + tracing_reset(tr, cpu); + ksym_tracing_enabled = 1; + ksym_trace_array = tr; + + return ret; +} + +static void ksym_trace_print_header(struct seq_file *m) +{ + seq_puts(m, + "# TASK-PID CPU# Symbol " + "Type Function\n"); + seq_puts(m, + "# | | | " + " | |\n"); +} + +static enum print_line_t ksym_trace_output(struct trace_iterator *iter) +{ + struct trace_entry *entry = iter->ent; + struct trace_seq *s = &iter->seq; + struct ksym_trace_entry *field; + char str[KSYM_SYMBOL_LEN]; + int ret; + + if (entry->type != TRACE_KSYM) + return TRACE_TYPE_UNHANDLED; + + trace_assign_type(field, entry); + + ret = trace_seq_printf(s, "%11s-%-5d [%03d] %pS", field->cmd, + entry->pid, iter->cpu, (char *)field->addr); + if (!ret) + return TRACE_TYPE_PARTIAL_LINE; + + switch (field->type) { + case HW_BREAKPOINT_R: + ret = trace_seq_printf(s, " R "); + break; + case HW_BREAKPOINT_W: + ret = trace_seq_printf(s, " W "); + break; + case HW_BREAKPOINT_R | HW_BREAKPOINT_W: + ret = trace_seq_printf(s, " RW "); + break; + default: + return TRACE_TYPE_PARTIAL_LINE; + } + + if (!ret) + return TRACE_TYPE_PARTIAL_LINE; + + sprint_symbol(str, field->ip); + ret = trace_seq_printf(s, "%s\n", str); + if (!ret) + return TRACE_TYPE_PARTIAL_LINE; + + return TRACE_TYPE_HANDLED; +} + +struct tracer ksym_tracer __read_mostly = +{ + .name = "ksym_tracer", + .init = ksym_trace_init, + .reset = ksym_trace_reset, +#ifdef CONFIG_FTRACE_SELFTEST + .selftest = trace_selftest_startup_ksym, +#endif + .print_header = ksym_trace_print_header, + .print_line = ksym_trace_output +}; + +__init static int init_ksym_trace(void) +{ + struct dentry *d_tracer; + struct dentry *entry; + + d_tracer = tracing_init_dentry(); + ksym_filter_entry_count = 0; + + entry = debugfs_create_file("ksym_trace_filter", 0644, d_tracer, + NULL, &ksym_tracing_fops); + if (!entry) + pr_warning("Could not create debugfs " + "'ksym_trace_filter' file\n"); + + return register_tracer(&ksym_tracer); +} +device_initcall(init_ksym_trace); + + +#ifdef CONFIG_PROFILE_KSYM_TRACER +static int ksym_tracer_stat_headers(struct seq_file *m) +{ + seq_puts(m, " Access Type "); + seq_puts(m, " Symbol Counter\n"); + seq_puts(m, " ----------- "); + seq_puts(m, " ------ -------\n"); + return 0; +} + +static int ksym_tracer_stat_show(struct seq_file *m, void *v) +{ + struct hlist_node *stat = v; + struct trace_ksym *entry; + int access_type = 0; + char fn_name[KSYM_NAME_LEN]; + + entry = hlist_entry(stat, struct trace_ksym, ksym_hlist); + + access_type = entry->attr.bp_type; + + switch (access_type) { + case HW_BREAKPOINT_R: + seq_puts(m, " R "); + break; + case HW_BREAKPOINT_W: + seq_puts(m, " W "); + break; + case HW_BREAKPOINT_R | HW_BREAKPOINT_W: + seq_puts(m, " RW "); + break; + default: + seq_puts(m, " NA "); + } + + if (lookup_symbol_name(entry->attr.bp_addr, fn_name) >= 0) + seq_printf(m, " %-36s", fn_name); + else + seq_printf(m, " %-36s", "<NA>"); + seq_printf(m, " %15lu\n", entry->counter); + + return 0; +} + +static void *ksym_tracer_stat_start(struct tracer_stat *trace) +{ + return ksym_filter_head.first; +} + +static void * +ksym_tracer_stat_next(void *v, int idx) +{ + struct hlist_node *stat = v; + + return stat->next; +} + +static struct tracer_stat ksym_tracer_stats = { + .name = "ksym_tracer", + .stat_start = ksym_tracer_stat_start, + .stat_next = ksym_tracer_stat_next, + .stat_headers = ksym_tracer_stat_headers, + .stat_show = ksym_tracer_stat_show +}; + +__init static int ksym_tracer_stat_init(void) +{ + int ret; + + ret = register_stat_tracer(&ksym_tracer_stats); + if (ret) { + printk(KERN_WARNING "Warning: could not register " + "ksym tracer stats\n"); + return 1; + } + + return 0; +} +fs_initcall(ksym_tracer_stat_init); +#endif /* CONFIG_PROFILE_KSYM_TRACER */ diff --git a/kernel/trace/trace_mmiotrace.c b/kernel/trace/trace_mmiotrace.c index 80e503ef613..0acd834659e 100644 --- a/kernel/trace/trace_mmiotrace.c +++ b/kernel/trace/trace_mmiotrace.c @@ -9,9 +9,12 @@ #include <linux/kernel.h> #include <linux/mmiotrace.h> #include <linux/pci.h> +#include <linux/time.h> + #include <asm/atomic.h> #include "trace.h" +#include "trace_output.h" struct header_iter { struct pci_dev *dev; @@ -173,7 +176,7 @@ static enum print_line_t mmio_print_rw(struct trace_iterator *iter) struct mmiotrace_rw *rw; struct trace_seq *s = &iter->seq; unsigned long long t = ns2usecs(iter->ts); - unsigned long usec_rem = do_div(t, 1000000ULL); + unsigned long usec_rem = do_div(t, USEC_PER_SEC); unsigned secs = (unsigned long)t; int ret = 1; @@ -183,21 +186,22 @@ static enum print_line_t mmio_print_rw(struct trace_iterator *iter) switch (rw->opcode) { case MMIO_READ: ret = trace_seq_printf(s, - "R %d %lu.%06lu %d 0x%llx 0x%lx 0x%lx %d\n", + "R %d %u.%06lu %d 0x%llx 0x%lx 0x%lx %d\n", rw->width, secs, usec_rem, rw->map_id, (unsigned long long)rw->phys, rw->value, rw->pc, 0); break; case MMIO_WRITE: ret = trace_seq_printf(s, - "W %d %lu.%06lu %d 0x%llx 0x%lx 0x%lx %d\n", + "W %d %u.%06lu %d 0x%llx 0x%lx 0x%lx %d\n", rw->width, secs, usec_rem, rw->map_id, (unsigned long long)rw->phys, rw->value, rw->pc, 0); break; case MMIO_UNKNOWN_OP: ret = trace_seq_printf(s, - "UNKNOWN %lu.%06lu %d 0x%llx %02x,%02x,%02x 0x%lx %d\n", + "UNKNOWN %u.%06lu %d 0x%llx %02lx,%02lx," + "%02lx 0x%lx %d\n", secs, usec_rem, rw->map_id, (unsigned long long)rw->phys, (rw->value >> 16) & 0xff, (rw->value >> 8) & 0xff, @@ -219,7 +223,7 @@ static enum print_line_t mmio_print_map(struct trace_iterator *iter) struct mmiotrace_map *m; struct trace_seq *s = &iter->seq; unsigned long long t = ns2usecs(iter->ts); - unsigned long usec_rem = do_div(t, 1000000ULL); + unsigned long usec_rem = do_div(t, USEC_PER_SEC); unsigned secs = (unsigned long)t; int ret; @@ -229,14 +233,14 @@ static enum print_line_t mmio_print_map(struct trace_iterator *iter) switch (m->opcode) { case MMIO_PROBE: ret = trace_seq_printf(s, - "MAP %lu.%06lu %d 0x%llx 0x%lx 0x%lx 0x%lx %d\n", + "MAP %u.%06lu %d 0x%llx 0x%lx 0x%lx 0x%lx %d\n", secs, usec_rem, m->map_id, (unsigned long long)m->phys, m->virt, m->len, 0UL, 0); break; case MMIO_UNPROBE: ret = trace_seq_printf(s, - "UNMAP %lu.%06lu %d 0x%lx %d\n", + "UNMAP %u.%06lu %d 0x%lx %d\n", secs, usec_rem, m->map_id, 0UL, 0); break; default: @@ -255,18 +259,15 @@ static enum print_line_t mmio_print_mark(struct trace_iterator *iter) const char *msg = print->buf; struct trace_seq *s = &iter->seq; unsigned long long t = ns2usecs(iter->ts); - unsigned long usec_rem = do_div(t, 1000000ULL); + unsigned long usec_rem = do_div(t, USEC_PER_SEC); unsigned secs = (unsigned long)t; int ret; /* The trailing newline must be in the message. */ - ret = trace_seq_printf(s, "MARK %lu.%06lu %s", secs, usec_rem, msg); + ret = trace_seq_printf(s, "MARK %u.%06lu %s", secs, usec_rem, msg); if (!ret) return TRACE_TYPE_PARTIAL_LINE; - if (entry->flags & TRACE_FLAG_CONT) - trace_seq_print_cont(s, iter); - return TRACE_TYPE_HANDLED; } @@ -306,23 +307,23 @@ static void __trace_mmiotrace_rw(struct trace_array *tr, struct trace_array_cpu *data, struct mmiotrace_rw *rw) { + struct ftrace_event_call *call = &event_mmiotrace_rw; + struct ring_buffer *buffer = tr->buffer; struct ring_buffer_event *event; struct trace_mmiotrace_rw *entry; - unsigned long irq_flags; + int pc = preempt_count(); - event = ring_buffer_lock_reserve(tr->buffer, sizeof(*entry), - &irq_flags); + event = trace_buffer_lock_reserve(buffer, TRACE_MMIO_RW, + sizeof(*entry), 0, pc); if (!event) { atomic_inc(&dropped_count); return; } entry = ring_buffer_event_data(event); - tracing_generic_entry_update(&entry->ent, 0, preempt_count()); - entry->ent.type = TRACE_MMIO_RW; entry->rw = *rw; - ring_buffer_unlock_commit(tr->buffer, event, irq_flags); - trace_wake_up(); + if (!filter_check_discard(call, entry, buffer, event)) + trace_buffer_unlock_commit(buffer, event, 0, pc); } void mmio_trace_rw(struct mmiotrace_rw *rw) @@ -336,23 +337,23 @@ static void __trace_mmiotrace_map(struct trace_array *tr, struct trace_array_cpu *data, struct mmiotrace_map *map) { + struct ftrace_event_call *call = &event_mmiotrace_map; + struct ring_buffer *buffer = tr->buffer; struct ring_buffer_event *event; struct trace_mmiotrace_map *entry; - unsigned long irq_flags; + int pc = preempt_count(); - event = ring_buffer_lock_reserve(tr->buffer, sizeof(*entry), - &irq_flags); + event = trace_buffer_lock_reserve(buffer, TRACE_MMIO_MAP, + sizeof(*entry), 0, pc); if (!event) { atomic_inc(&dropped_count); return; } entry = ring_buffer_event_data(event); - tracing_generic_entry_update(&entry->ent, 0, preempt_count()); - entry->ent.type = TRACE_MMIO_MAP; entry->map = *map; - ring_buffer_unlock_commit(tr->buffer, event, irq_flags); - trace_wake_up(); + if (!filter_check_discard(call, entry, buffer, event)) + trace_buffer_unlock_commit(buffer, event, 0, pc); } void mmio_trace_mapping(struct mmiotrace_map *map) @@ -368,5 +369,5 @@ void mmio_trace_mapping(struct mmiotrace_map *map) int mmio_trace_printk(const char *fmt, va_list args) { - return trace_vprintk(0, -1, fmt, args); + return trace_vprintk(0, fmt, args); } diff --git a/kernel/trace/trace_nop.c b/kernel/trace/trace_nop.c index b9767acd30a..394f94417e2 100644 --- a/kernel/trace/trace_nop.c +++ b/kernel/trace/trace_nop.c @@ -47,12 +47,7 @@ static void stop_nop_trace(struct trace_array *tr) static int nop_trace_init(struct trace_array *tr) { - int cpu; ctx_trace = tr; - - for_each_online_cpu(cpu) - tracing_reset(tr, cpu); - start_nop_trace(tr); return 0; } @@ -96,6 +91,7 @@ struct tracer nop_trace __read_mostly = .name = "nop", .init = nop_trace_init, .reset = nop_trace_reset, + .wait_pipe = poll_wait_pipe, #ifdef CONFIG_FTRACE_SELFTEST .selftest = trace_selftest_startup_nop, #endif diff --git a/kernel/trace/trace_output.c b/kernel/trace/trace_output.c new file mode 100644 index 00000000000..b6c12c6a1bc --- /dev/null +++ b/kernel/trace/trace_output.c @@ -0,0 +1,1233 @@ +/* + * trace_output.c + * + * Copyright (C) 2008 Red Hat Inc, Steven Rostedt <srostedt@redhat.com> + * + */ + +#include <linux/module.h> +#include <linux/mutex.h> +#include <linux/ftrace.h> + +#include "trace_output.h" + +/* must be a power of 2 */ +#define EVENT_HASHSIZE 128 + +DECLARE_RWSEM(trace_event_mutex); + +DEFINE_PER_CPU(struct trace_seq, ftrace_event_seq); +EXPORT_PER_CPU_SYMBOL(ftrace_event_seq); + +static struct hlist_head event_hash[EVENT_HASHSIZE] __read_mostly; + +static int next_event_type = __TRACE_LAST_TYPE + 1; + +void trace_print_seq(struct seq_file *m, struct trace_seq *s) +{ + int len = s->len >= PAGE_SIZE ? PAGE_SIZE - 1 : s->len; + + seq_write(m, s->buffer, len); + + trace_seq_init(s); +} + +enum print_line_t trace_print_bprintk_msg_only(struct trace_iterator *iter) +{ + struct trace_seq *s = &iter->seq; + struct trace_entry *entry = iter->ent; + struct bprint_entry *field; + int ret; + + trace_assign_type(field, entry); + + ret = trace_seq_bprintf(s, field->fmt, field->buf); + if (!ret) + return TRACE_TYPE_PARTIAL_LINE; + + return TRACE_TYPE_HANDLED; +} + +enum print_line_t trace_print_printk_msg_only(struct trace_iterator *iter) +{ + struct trace_seq *s = &iter->seq; + struct trace_entry *entry = iter->ent; + struct print_entry *field; + int ret; + + trace_assign_type(field, entry); + + ret = trace_seq_printf(s, "%s", field->buf); + if (!ret) + return TRACE_TYPE_PARTIAL_LINE; + + return TRACE_TYPE_HANDLED; +} + +/** + * trace_seq_printf - sequence printing of trace information + * @s: trace sequence descriptor + * @fmt: printf format string + * + * It returns 0 if the trace oversizes the buffer's free + * space, 1 otherwise. + * + * The tracer may use either sequence operations or its own + * copy to user routines. To simplify formating of a trace + * trace_seq_printf is used to store strings into a special + * buffer (@s). Then the output may be either used by + * the sequencer or pulled into another buffer. + */ +int +trace_seq_printf(struct trace_seq *s, const char *fmt, ...) +{ + int len = (PAGE_SIZE - 1) - s->len; + va_list ap; + int ret; + + if (!len) + return 0; + + va_start(ap, fmt); + ret = vsnprintf(s->buffer + s->len, len, fmt, ap); + va_end(ap); + + /* If we can't write it all, don't bother writing anything */ + if (ret >= len) + return 0; + + s->len += ret; + + return 1; +} +EXPORT_SYMBOL_GPL(trace_seq_printf); + +/** + * trace_seq_vprintf - sequence printing of trace information + * @s: trace sequence descriptor + * @fmt: printf format string + * + * The tracer may use either sequence operations or its own + * copy to user routines. To simplify formating of a trace + * trace_seq_printf is used to store strings into a special + * buffer (@s). Then the output may be either used by + * the sequencer or pulled into another buffer. + */ +int +trace_seq_vprintf(struct trace_seq *s, const char *fmt, va_list args) +{ + int len = (PAGE_SIZE - 1) - s->len; + int ret; + + if (!len) + return 0; + + ret = vsnprintf(s->buffer + s->len, len, fmt, args); + + /* If we can't write it all, don't bother writing anything */ + if (ret >= len) + return 0; + + s->len += ret; + + return len; +} +EXPORT_SYMBOL_GPL(trace_seq_vprintf); + +int trace_seq_bprintf(struct trace_seq *s, const char *fmt, const u32 *binary) +{ + int len = (PAGE_SIZE - 1) - s->len; + int ret; + + if (!len) + return 0; + + ret = bstr_printf(s->buffer + s->len, len, fmt, binary); + + /* If we can't write it all, don't bother writing anything */ + if (ret >= len) + return 0; + + s->len += ret; + + return len; +} + +/** + * trace_seq_puts - trace sequence printing of simple string + * @s: trace sequence descriptor + * @str: simple string to record + * + * The tracer may use either the sequence operations or its own + * copy to user routines. This function records a simple string + * into a special buffer (@s) for later retrieval by a sequencer + * or other mechanism. + */ +int trace_seq_puts(struct trace_seq *s, const char *str) +{ + int len = strlen(str); + + if (len > ((PAGE_SIZE - 1) - s->len)) + return 0; + + memcpy(s->buffer + s->len, str, len); + s->len += len; + + return len; +} + +int trace_seq_putc(struct trace_seq *s, unsigned char c) +{ + if (s->len >= (PAGE_SIZE - 1)) + return 0; + + s->buffer[s->len++] = c; + + return 1; +} + +int trace_seq_putmem(struct trace_seq *s, const void *mem, size_t len) +{ + if (len > ((PAGE_SIZE - 1) - s->len)) + return 0; + + memcpy(s->buffer + s->len, mem, len); + s->len += len; + + return len; +} + +int trace_seq_putmem_hex(struct trace_seq *s, const void *mem, size_t len) +{ + unsigned char hex[HEX_CHARS]; + const unsigned char *data = mem; + int i, j; + +#ifdef __BIG_ENDIAN + for (i = 0, j = 0; i < len; i++) { +#else + for (i = len-1, j = 0; i >= 0; i--) { +#endif + hex[j++] = hex_asc_hi(data[i]); + hex[j++] = hex_asc_lo(data[i]); + } + hex[j++] = ' '; + + return trace_seq_putmem(s, hex, j); +} + +void *trace_seq_reserve(struct trace_seq *s, size_t len) +{ + void *ret; + + if (len > ((PAGE_SIZE - 1) - s->len)) + return NULL; + + ret = s->buffer + s->len; + s->len += len; + + return ret; +} + +int trace_seq_path(struct trace_seq *s, struct path *path) +{ + unsigned char *p; + + if (s->len >= (PAGE_SIZE - 1)) + return 0; + p = d_path(path, s->buffer + s->len, PAGE_SIZE - s->len); + if (!IS_ERR(p)) { + p = mangle_path(s->buffer + s->len, p, "\n"); + if (p) { + s->len = p - s->buffer; + return 1; + } + } else { + s->buffer[s->len++] = '?'; + return 1; + } + + return 0; +} + +const char * +ftrace_print_flags_seq(struct trace_seq *p, const char *delim, + unsigned long flags, + const struct trace_print_flags *flag_array) +{ + unsigned long mask; + const char *str; + const char *ret = p->buffer + p->len; + int i; + + for (i = 0; flag_array[i].name && flags; i++) { + + mask = flag_array[i].mask; + if ((flags & mask) != mask) + continue; + + str = flag_array[i].name; + flags &= ~mask; + if (p->len && delim) + trace_seq_puts(p, delim); + trace_seq_puts(p, str); + } + + /* check for left over flags */ + if (flags) { + if (p->len && delim) + trace_seq_puts(p, delim); + trace_seq_printf(p, "0x%lx", flags); + } + + trace_seq_putc(p, 0); + + return ret; +} +EXPORT_SYMBOL(ftrace_print_flags_seq); + +const char * +ftrace_print_symbols_seq(struct trace_seq *p, unsigned long val, + const struct trace_print_flags *symbol_array) +{ + int i; + const char *ret = p->buffer + p->len; + + for (i = 0; symbol_array[i].name; i++) { + + if (val != symbol_array[i].mask) + continue; + + trace_seq_puts(p, symbol_array[i].name); + break; + } + + if (!p->len) + trace_seq_printf(p, "0x%lx", val); + + trace_seq_putc(p, 0); + + return ret; +} +EXPORT_SYMBOL(ftrace_print_symbols_seq); + +#ifdef CONFIG_KRETPROBES +static inline const char *kretprobed(const char *name) +{ + static const char tramp_name[] = "kretprobe_trampoline"; + int size = sizeof(tramp_name); + + if (strncmp(tramp_name, name, size) == 0) + return "[unknown/kretprobe'd]"; + return name; +} +#else +static inline const char *kretprobed(const char *name) +{ + return name; +} +#endif /* CONFIG_KRETPROBES */ + +static int +seq_print_sym_short(struct trace_seq *s, const char *fmt, unsigned long address) +{ +#ifdef CONFIG_KALLSYMS + char str[KSYM_SYMBOL_LEN]; + const char *name; + + kallsyms_lookup(address, NULL, NULL, NULL, str); + + name = kretprobed(str); + + return trace_seq_printf(s, fmt, name); +#endif + return 1; +} + +static int +seq_print_sym_offset(struct trace_seq *s, const char *fmt, + unsigned long address) +{ +#ifdef CONFIG_KALLSYMS + char str[KSYM_SYMBOL_LEN]; + const char *name; + + sprint_symbol(str, address); + name = kretprobed(str); + + return trace_seq_printf(s, fmt, name); +#endif + return 1; +} + +#ifndef CONFIG_64BIT +# define IP_FMT "%08lx" +#else +# define IP_FMT "%016lx" +#endif + +int seq_print_user_ip(struct trace_seq *s, struct mm_struct *mm, + unsigned long ip, unsigned long sym_flags) +{ + struct file *file = NULL; + unsigned long vmstart = 0; + int ret = 1; + + if (mm) { + const struct vm_area_struct *vma; + + down_read(&mm->mmap_sem); + vma = find_vma(mm, ip); + if (vma) { + file = vma->vm_file; + vmstart = vma->vm_start; + } + if (file) { + ret = trace_seq_path(s, &file->f_path); + if (ret) + ret = trace_seq_printf(s, "[+0x%lx]", + ip - vmstart); + } + up_read(&mm->mmap_sem); + } + if (ret && ((sym_flags & TRACE_ITER_SYM_ADDR) || !file)) + ret = trace_seq_printf(s, " <" IP_FMT ">", ip); + return ret; +} + +int +seq_print_userip_objs(const struct userstack_entry *entry, struct trace_seq *s, + unsigned long sym_flags) +{ + struct mm_struct *mm = NULL; + int ret = 1; + unsigned int i; + + if (trace_flags & TRACE_ITER_SYM_USEROBJ) { + struct task_struct *task; + /* + * we do the lookup on the thread group leader, + * since individual threads might have already quit! + */ + rcu_read_lock(); + task = find_task_by_vpid(entry->tgid); + if (task) + mm = get_task_mm(task); + rcu_read_unlock(); + } + + for (i = 0; i < FTRACE_STACK_ENTRIES; i++) { + unsigned long ip = entry->caller[i]; + + if (ip == ULONG_MAX || !ret) + break; + if (ret) + ret = trace_seq_puts(s, " => "); + if (!ip) { + if (ret) + ret = trace_seq_puts(s, "??"); + if (ret) + ret = trace_seq_puts(s, "\n"); + continue; + } + if (!ret) + break; + if (ret) + ret = seq_print_user_ip(s, mm, ip, sym_flags); + ret = trace_seq_puts(s, "\n"); + } + + if (mm) + mmput(mm); + return ret; +} + +int +seq_print_ip_sym(struct trace_seq *s, unsigned long ip, unsigned long sym_flags) +{ + int ret; + + if (!ip) + return trace_seq_printf(s, "0"); + + if (sym_flags & TRACE_ITER_SYM_OFFSET) + ret = seq_print_sym_offset(s, "%s", ip); + else + ret = seq_print_sym_short(s, "%s", ip); + + if (!ret) + return 0; + + if (sym_flags & TRACE_ITER_SYM_ADDR) + ret = trace_seq_printf(s, " <" IP_FMT ">", ip); + return ret; +} + +/** + * trace_print_lat_fmt - print the irq, preempt and lockdep fields + * @s: trace seq struct to write to + * @entry: The trace entry field from the ring buffer + * + * Prints the generic fields of irqs off, in hard or softirq, preempt + * count and lock depth. + */ +int trace_print_lat_fmt(struct trace_seq *s, struct trace_entry *entry) +{ + int hardirq, softirq; + int ret; + + hardirq = entry->flags & TRACE_FLAG_HARDIRQ; + softirq = entry->flags & TRACE_FLAG_SOFTIRQ; + + if (!trace_seq_printf(s, "%c%c%c", + (entry->flags & TRACE_FLAG_IRQS_OFF) ? 'd' : + (entry->flags & TRACE_FLAG_IRQS_NOSUPPORT) ? + 'X' : '.', + (entry->flags & TRACE_FLAG_NEED_RESCHED) ? + 'N' : '.', + (hardirq && softirq) ? 'H' : + hardirq ? 'h' : softirq ? 's' : '.')) + return 0; + + if (entry->preempt_count) + ret = trace_seq_printf(s, "%x", entry->preempt_count); + else + ret = trace_seq_putc(s, '.'); + + if (!ret) + return 0; + + if (entry->lock_depth < 0) + return trace_seq_putc(s, '.'); + + return trace_seq_printf(s, "%d", entry->lock_depth); +} + +static int +lat_print_generic(struct trace_seq *s, struct trace_entry *entry, int cpu) +{ + char comm[TASK_COMM_LEN]; + + trace_find_cmdline(entry->pid, comm); + + if (!trace_seq_printf(s, "%8.8s-%-5d %3d", + comm, entry->pid, cpu)) + return 0; + + return trace_print_lat_fmt(s, entry); +} + +static unsigned long preempt_mark_thresh = 100; + +static int +lat_print_timestamp(struct trace_seq *s, u64 abs_usecs, + unsigned long rel_usecs) +{ + return trace_seq_printf(s, " %4lldus%c: ", abs_usecs, + rel_usecs > preempt_mark_thresh ? '!' : + rel_usecs > 1 ? '+' : ' '); +} + +int trace_print_context(struct trace_iterator *iter) +{ + struct trace_seq *s = &iter->seq; + struct trace_entry *entry = iter->ent; + unsigned long long t = ns2usecs(iter->ts); + unsigned long usec_rem = do_div(t, USEC_PER_SEC); + unsigned long secs = (unsigned long)t; + char comm[TASK_COMM_LEN]; + + trace_find_cmdline(entry->pid, comm); + + return trace_seq_printf(s, "%16s-%-5d [%03d] %5lu.%06lu: ", + comm, entry->pid, iter->cpu, secs, usec_rem); +} + +int trace_print_lat_context(struct trace_iterator *iter) +{ + u64 next_ts; + int ret; + struct trace_seq *s = &iter->seq; + struct trace_entry *entry = iter->ent, + *next_entry = trace_find_next_entry(iter, NULL, + &next_ts); + unsigned long verbose = (trace_flags & TRACE_ITER_VERBOSE); + unsigned long abs_usecs = ns2usecs(iter->ts - iter->tr->time_start); + unsigned long rel_usecs; + + if (!next_entry) + next_ts = iter->ts; + rel_usecs = ns2usecs(next_ts - iter->ts); + + if (verbose) { + char comm[TASK_COMM_LEN]; + + trace_find_cmdline(entry->pid, comm); + + ret = trace_seq_printf(s, "%16s %5d %3d %d %08x %08lx [%08llx]" + " %ld.%03ldms (+%ld.%03ldms): ", comm, + entry->pid, iter->cpu, entry->flags, + entry->preempt_count, iter->idx, + ns2usecs(iter->ts), + abs_usecs / USEC_PER_MSEC, + abs_usecs % USEC_PER_MSEC, + rel_usecs / USEC_PER_MSEC, + rel_usecs % USEC_PER_MSEC); + } else { + ret = lat_print_generic(s, entry, iter->cpu); + if (ret) + ret = lat_print_timestamp(s, abs_usecs, rel_usecs); + } + + return ret; +} + +static const char state_to_char[] = TASK_STATE_TO_CHAR_STR; + +static int task_state_char(unsigned long state) +{ + int bit = state ? __ffs(state) + 1 : 0; + + return bit < sizeof(state_to_char) - 1 ? state_to_char[bit] : '?'; +} + +/** + * ftrace_find_event - find a registered event + * @type: the type of event to look for + * + * Returns an event of type @type otherwise NULL + * Called with trace_event_read_lock() held. + */ +struct trace_event *ftrace_find_event(int type) +{ + struct trace_event *event; + struct hlist_node *n; + unsigned key; + + key = type & (EVENT_HASHSIZE - 1); + + hlist_for_each_entry(event, n, &event_hash[key], node) { + if (event->type == type) + return event; + } + + return NULL; +} + +static LIST_HEAD(ftrace_event_list); + +static int trace_search_list(struct list_head **list) +{ + struct trace_event *e; + int last = __TRACE_LAST_TYPE; + + if (list_empty(&ftrace_event_list)) { + *list = &ftrace_event_list; + return last + 1; + } + + /* + * We used up all possible max events, + * lets see if somebody freed one. + */ + list_for_each_entry(e, &ftrace_event_list, list) { + if (e->type != last + 1) + break; + last++; + } + + /* Did we used up all 65 thousand events??? */ + if ((last + 1) > FTRACE_MAX_EVENT) + return 0; + + *list = &e->list; + return last + 1; +} + +void trace_event_read_lock(void) +{ + down_read(&trace_event_mutex); +} + +void trace_event_read_unlock(void) +{ + up_read(&trace_event_mutex); +} + +/** + * register_ftrace_event - register output for an event type + * @event: the event type to register + * + * Event types are stored in a hash and this hash is used to + * find a way to print an event. If the @event->type is set + * then it will use that type, otherwise it will assign a + * type to use. + * + * If you assign your own type, please make sure it is added + * to the trace_type enum in trace.h, to avoid collisions + * with the dynamic types. + * + * Returns the event type number or zero on error. + */ +int register_ftrace_event(struct trace_event *event) +{ + unsigned key; + int ret = 0; + + down_write(&trace_event_mutex); + + if (WARN_ON(!event)) + goto out; + + INIT_LIST_HEAD(&event->list); + + if (!event->type) { + struct list_head *list = NULL; + + if (next_event_type > FTRACE_MAX_EVENT) { + + event->type = trace_search_list(&list); + if (!event->type) + goto out; + + } else { + + event->type = next_event_type++; + list = &ftrace_event_list; + } + + if (WARN_ON(ftrace_find_event(event->type))) + goto out; + + list_add_tail(&event->list, list); + + } else if (event->type > __TRACE_LAST_TYPE) { + printk(KERN_WARNING "Need to add type to trace.h\n"); + WARN_ON(1); + goto out; + } else { + /* Is this event already used */ + if (ftrace_find_event(event->type)) + goto out; + } + + if (event->trace == NULL) + event->trace = trace_nop_print; + if (event->raw == NULL) + event->raw = trace_nop_print; + if (event->hex == NULL) + event->hex = trace_nop_print; + if (event->binary == NULL) + event->binary = trace_nop_print; + + key = event->type & (EVENT_HASHSIZE - 1); + + hlist_add_head(&event->node, &event_hash[key]); + + ret = event->type; + out: + up_write(&trace_event_mutex); + + return ret; +} +EXPORT_SYMBOL_GPL(register_ftrace_event); + +/* + * Used by module code with the trace_event_mutex held for write. + */ +int __unregister_ftrace_event(struct trace_event *event) +{ + hlist_del(&event->node); + list_del(&event->list); + return 0; +} + +/** + * unregister_ftrace_event - remove a no longer used event + * @event: the event to remove + */ +int unregister_ftrace_event(struct trace_event *event) +{ + down_write(&trace_event_mutex); + __unregister_ftrace_event(event); + up_write(&trace_event_mutex); + + return 0; +} +EXPORT_SYMBOL_GPL(unregister_ftrace_event); + +/* + * Standard events + */ + +enum print_line_t trace_nop_print(struct trace_iterator *iter, int flags) +{ + return TRACE_TYPE_HANDLED; +} + +/* TRACE_FN */ +static enum print_line_t trace_fn_trace(struct trace_iterator *iter, int flags) +{ + struct ftrace_entry *field; + struct trace_seq *s = &iter->seq; + + trace_assign_type(field, iter->ent); + + if (!seq_print_ip_sym(s, field->ip, flags)) + goto partial; + + if ((flags & TRACE_ITER_PRINT_PARENT) && field->parent_ip) { + if (!trace_seq_printf(s, " <-")) + goto partial; + if (!seq_print_ip_sym(s, + field->parent_ip, + flags)) + goto partial; + } + if (!trace_seq_printf(s, "\n")) + goto partial; + + return TRACE_TYPE_HANDLED; + + partial: + return TRACE_TYPE_PARTIAL_LINE; +} + +static enum print_line_t trace_fn_raw(struct trace_iterator *iter, int flags) +{ + struct ftrace_entry *field; + + trace_assign_type(field, iter->ent); + + if (!trace_seq_printf(&iter->seq, "%lx %lx\n", + field->ip, + field->parent_ip)) + return TRACE_TYPE_PARTIAL_LINE; + + return TRACE_TYPE_HANDLED; +} + +static enum print_line_t trace_fn_hex(struct trace_iterator *iter, int flags) +{ + struct ftrace_entry *field; + struct trace_seq *s = &iter->seq; + + trace_assign_type(field, iter->ent); + + SEQ_PUT_HEX_FIELD_RET(s, field->ip); + SEQ_PUT_HEX_FIELD_RET(s, field->parent_ip); + + return TRACE_TYPE_HANDLED; +} + +static enum print_line_t trace_fn_bin(struct trace_iterator *iter, int flags) +{ + struct ftrace_entry *field; + struct trace_seq *s = &iter->seq; + + trace_assign_type(field, iter->ent); + + SEQ_PUT_FIELD_RET(s, field->ip); + SEQ_PUT_FIELD_RET(s, field->parent_ip); + + return TRACE_TYPE_HANDLED; +} + +static struct trace_event trace_fn_event = { + .type = TRACE_FN, + .trace = trace_fn_trace, + .raw = trace_fn_raw, + .hex = trace_fn_hex, + .binary = trace_fn_bin, +}; + +/* TRACE_CTX an TRACE_WAKE */ +static enum print_line_t trace_ctxwake_print(struct trace_iterator *iter, + char *delim) +{ + struct ctx_switch_entry *field; + char comm[TASK_COMM_LEN]; + int S, T; + + + trace_assign_type(field, iter->ent); + + T = task_state_char(field->next_state); + S = task_state_char(field->prev_state); + trace_find_cmdline(field->next_pid, comm); + if (!trace_seq_printf(&iter->seq, + " %5d:%3d:%c %s [%03d] %5d:%3d:%c %s\n", + field->prev_pid, + field->prev_prio, + S, delim, + field->next_cpu, + field->next_pid, + field->next_prio, + T, comm)) + return TRACE_TYPE_PARTIAL_LINE; + + return TRACE_TYPE_HANDLED; +} + +static enum print_line_t trace_ctx_print(struct trace_iterator *iter, int flags) +{ + return trace_ctxwake_print(iter, "==>"); +} + +static enum print_line_t trace_wake_print(struct trace_iterator *iter, + int flags) +{ + return trace_ctxwake_print(iter, " +"); +} + +static int trace_ctxwake_raw(struct trace_iterator *iter, char S) +{ + struct ctx_switch_entry *field; + int T; + + trace_assign_type(field, iter->ent); + + if (!S) + S = task_state_char(field->prev_state); + T = task_state_char(field->next_state); + if (!trace_seq_printf(&iter->seq, "%d %d %c %d %d %d %c\n", + field->prev_pid, + field->prev_prio, + S, + field->next_cpu, + field->next_pid, + field->next_prio, + T)) + return TRACE_TYPE_PARTIAL_LINE; + + return TRACE_TYPE_HANDLED; +} + +static enum print_line_t trace_ctx_raw(struct trace_iterator *iter, int flags) +{ + return trace_ctxwake_raw(iter, 0); +} + +static enum print_line_t trace_wake_raw(struct trace_iterator *iter, int flags) +{ + return trace_ctxwake_raw(iter, '+'); +} + + +static int trace_ctxwake_hex(struct trace_iterator *iter, char S) +{ + struct ctx_switch_entry *field; + struct trace_seq *s = &iter->seq; + int T; + + trace_assign_type(field, iter->ent); + + if (!S) + S = task_state_char(field->prev_state); + T = task_state_char(field->next_state); + + SEQ_PUT_HEX_FIELD_RET(s, field->prev_pid); + SEQ_PUT_HEX_FIELD_RET(s, field->prev_prio); + SEQ_PUT_HEX_FIELD_RET(s, S); + SEQ_PUT_HEX_FIELD_RET(s, field->next_cpu); + SEQ_PUT_HEX_FIELD_RET(s, field->next_pid); + SEQ_PUT_HEX_FIELD_RET(s, field->next_prio); + SEQ_PUT_HEX_FIELD_RET(s, T); + + return TRACE_TYPE_HANDLED; +} + +static enum print_line_t trace_ctx_hex(struct trace_iterator *iter, int flags) +{ + return trace_ctxwake_hex(iter, 0); +} + +static enum print_line_t trace_wake_hex(struct trace_iterator *iter, int flags) +{ + return trace_ctxwake_hex(iter, '+'); +} + +static enum print_line_t trace_ctxwake_bin(struct trace_iterator *iter, + int flags) +{ + struct ctx_switch_entry *field; + struct trace_seq *s = &iter->seq; + + trace_assign_type(field, iter->ent); + + SEQ_PUT_FIELD_RET(s, field->prev_pid); + SEQ_PUT_FIELD_RET(s, field->prev_prio); + SEQ_PUT_FIELD_RET(s, field->prev_state); + SEQ_PUT_FIELD_RET(s, field->next_pid); + SEQ_PUT_FIELD_RET(s, field->next_prio); + SEQ_PUT_FIELD_RET(s, field->next_state); + + return TRACE_TYPE_HANDLED; +} + +static struct trace_event trace_ctx_event = { + .type = TRACE_CTX, + .trace = trace_ctx_print, + .raw = trace_ctx_raw, + .hex = trace_ctx_hex, + .binary = trace_ctxwake_bin, +}; + +static struct trace_event trace_wake_event = { + .type = TRACE_WAKE, + .trace = trace_wake_print, + .raw = trace_wake_raw, + .hex = trace_wake_hex, + .binary = trace_ctxwake_bin, +}; + +/* TRACE_SPECIAL */ +static enum print_line_t trace_special_print(struct trace_iterator *iter, + int flags) +{ + struct special_entry *field; + + trace_assign_type(field, iter->ent); + + if (!trace_seq_printf(&iter->seq, "# %ld %ld %ld\n", + field->arg1, + field->arg2, + field->arg3)) + return TRACE_TYPE_PARTIAL_LINE; + + return TRACE_TYPE_HANDLED; +} + +static enum print_line_t trace_special_hex(struct trace_iterator *iter, + int flags) +{ + struct special_entry *field; + struct trace_seq *s = &iter->seq; + + trace_assign_type(field, iter->ent); + + SEQ_PUT_HEX_FIELD_RET(s, field->arg1); + SEQ_PUT_HEX_FIELD_RET(s, field->arg2); + SEQ_PUT_HEX_FIELD_RET(s, field->arg3); + + return TRACE_TYPE_HANDLED; +} + +static enum print_line_t trace_special_bin(struct trace_iterator *iter, + int flags) +{ + struct special_entry *field; + struct trace_seq *s = &iter->seq; + + trace_assign_type(field, iter->ent); + + SEQ_PUT_FIELD_RET(s, field->arg1); + SEQ_PUT_FIELD_RET(s, field->arg2); + SEQ_PUT_FIELD_RET(s, field->arg3); + + return TRACE_TYPE_HANDLED; +} + +static struct trace_event trace_special_event = { + .type = TRACE_SPECIAL, + .trace = trace_special_print, + .raw = trace_special_print, + .hex = trace_special_hex, + .binary = trace_special_bin, +}; + +/* TRACE_STACK */ + +static enum print_line_t trace_stack_print(struct trace_iterator *iter, + int flags) +{ + struct stack_entry *field; + struct trace_seq *s = &iter->seq; + int i; + + trace_assign_type(field, iter->ent); + + if (!trace_seq_puts(s, "<stack trace>\n")) + goto partial; + for (i = 0; i < FTRACE_STACK_ENTRIES; i++) { + if (!field->caller[i] || (field->caller[i] == ULONG_MAX)) + break; + if (!trace_seq_puts(s, " => ")) + goto partial; + + if (!seq_print_ip_sym(s, field->caller[i], flags)) + goto partial; + if (!trace_seq_puts(s, "\n")) + goto partial; + } + + return TRACE_TYPE_HANDLED; + + partial: + return TRACE_TYPE_PARTIAL_LINE; +} + +static struct trace_event trace_stack_event = { + .type = TRACE_STACK, + .trace = trace_stack_print, + .raw = trace_special_print, + .hex = trace_special_hex, + .binary = trace_special_bin, +}; + +/* TRACE_USER_STACK */ +static enum print_line_t trace_user_stack_print(struct trace_iterator *iter, + int flags) +{ + struct userstack_entry *field; + struct trace_seq *s = &iter->seq; + + trace_assign_type(field, iter->ent); + + if (!trace_seq_puts(s, "<user stack trace>\n")) + goto partial; + + if (!seq_print_userip_objs(field, s, flags)) + goto partial; + + return TRACE_TYPE_HANDLED; + + partial: + return TRACE_TYPE_PARTIAL_LINE; +} + +static struct trace_event trace_user_stack_event = { + .type = TRACE_USER_STACK, + .trace = trace_user_stack_print, + .raw = trace_special_print, + .hex = trace_special_hex, + .binary = trace_special_bin, +}; + +/* TRACE_BPRINT */ +static enum print_line_t +trace_bprint_print(struct trace_iterator *iter, int flags) +{ + struct trace_entry *entry = iter->ent; + struct trace_seq *s = &iter->seq; + struct bprint_entry *field; + + trace_assign_type(field, entry); + + if (!seq_print_ip_sym(s, field->ip, flags)) + goto partial; + + if (!trace_seq_puts(s, ": ")) + goto partial; + + if (!trace_seq_bprintf(s, field->fmt, field->buf)) + goto partial; + + return TRACE_TYPE_HANDLED; + + partial: + return TRACE_TYPE_PARTIAL_LINE; +} + + +static enum print_line_t +trace_bprint_raw(struct trace_iterator *iter, int flags) +{ + struct bprint_entry *field; + struct trace_seq *s = &iter->seq; + + trace_assign_type(field, iter->ent); + + if (!trace_seq_printf(s, ": %lx : ", field->ip)) + goto partial; + + if (!trace_seq_bprintf(s, field->fmt, field->buf)) + goto partial; + + return TRACE_TYPE_HANDLED; + + partial: + return TRACE_TYPE_PARTIAL_LINE; +} + + +static struct trace_event trace_bprint_event = { + .type = TRACE_BPRINT, + .trace = trace_bprint_print, + .raw = trace_bprint_raw, +}; + +/* TRACE_PRINT */ +static enum print_line_t trace_print_print(struct trace_iterator *iter, + int flags) +{ + struct print_entry *field; + struct trace_seq *s = &iter->seq; + + trace_assign_type(field, iter->ent); + + if (!seq_print_ip_sym(s, field->ip, flags)) + goto partial; + + if (!trace_seq_printf(s, ": %s", field->buf)) + goto partial; + + return TRACE_TYPE_HANDLED; + + partial: + return TRACE_TYPE_PARTIAL_LINE; +} + +static enum print_line_t trace_print_raw(struct trace_iterator *iter, int flags) +{ + struct print_entry *field; + + trace_assign_type(field, iter->ent); + + if (!trace_seq_printf(&iter->seq, "# %lx %s", field->ip, field->buf)) + goto partial; + + return TRACE_TYPE_HANDLED; + + partial: + return TRACE_TYPE_PARTIAL_LINE; +} + +static struct trace_event trace_print_event = { + .type = TRACE_PRINT, + .trace = trace_print_print, + .raw = trace_print_raw, +}; + + +static struct trace_event *events[] __initdata = { + &trace_fn_event, + &trace_ctx_event, + &trace_wake_event, + &trace_special_event, + &trace_stack_event, + &trace_user_stack_event, + &trace_bprint_event, + &trace_print_event, + NULL +}; + +__init static int init_events(void) +{ + struct trace_event *event; + int i, ret; + + for (i = 0; events[i]; i++) { + event = events[i]; + + ret = register_ftrace_event(event); + if (!ret) { + printk(KERN_WARNING "event %d failed to register\n", + event->type); + WARN_ON_ONCE(1); + } + } + + return 0; +} +device_initcall(init_events); diff --git a/kernel/trace/trace_output.h b/kernel/trace/trace_output.h new file mode 100644 index 00000000000..9d91c72ba38 --- /dev/null +++ b/kernel/trace/trace_output.h @@ -0,0 +1,53 @@ +#ifndef __TRACE_EVENTS_H +#define __TRACE_EVENTS_H + +#include <linux/trace_seq.h> +#include "trace.h" + +extern enum print_line_t +trace_print_bprintk_msg_only(struct trace_iterator *iter); +extern enum print_line_t +trace_print_printk_msg_only(struct trace_iterator *iter); + +extern int +seq_print_ip_sym(struct trace_seq *s, unsigned long ip, + unsigned long sym_flags); +extern int seq_print_userip_objs(const struct userstack_entry *entry, + struct trace_seq *s, unsigned long sym_flags); +extern int seq_print_user_ip(struct trace_seq *s, struct mm_struct *mm, + unsigned long ip, unsigned long sym_flags); + +extern int trace_print_context(struct trace_iterator *iter); +extern int trace_print_lat_context(struct trace_iterator *iter); + +extern void trace_event_read_lock(void); +extern void trace_event_read_unlock(void); +extern struct trace_event *ftrace_find_event(int type); + +extern enum print_line_t trace_nop_print(struct trace_iterator *iter, + int flags); +extern int +trace_print_lat_fmt(struct trace_seq *s, struct trace_entry *entry); + +/* used by module unregistering */ +extern int __unregister_ftrace_event(struct trace_event *event); +extern struct rw_semaphore trace_event_mutex; + +#define MAX_MEMHEX_BYTES 8 +#define HEX_CHARS (MAX_MEMHEX_BYTES*2 + 1) + +#define SEQ_PUT_FIELD_RET(s, x) \ +do { \ + if (!trace_seq_putmem(s, &(x), sizeof(x))) \ + return TRACE_TYPE_PARTIAL_LINE; \ +} while (0) + +#define SEQ_PUT_HEX_FIELD_RET(s, x) \ +do { \ + BUILD_BUG_ON(sizeof(x) > MAX_MEMHEX_BYTES); \ + if (!trace_seq_putmem_hex(s, &(x), sizeof(x))) \ + return TRACE_TYPE_PARTIAL_LINE; \ +} while (0) + +#endif + diff --git a/kernel/trace/trace_power.c b/kernel/trace/trace_power.c deleted file mode 100644 index 7bda248daf5..00000000000 --- a/kernel/trace/trace_power.c +++ /dev/null @@ -1,179 +0,0 @@ -/* - * ring buffer based C-state tracer - * - * Arjan van de Ven <arjan@linux.intel.com> - * Copyright (C) 2008 Intel Corporation - * - * Much is borrowed from trace_boot.c which is - * Copyright (C) 2008 Frederic Weisbecker <fweisbec@gmail.com> - * - */ - -#include <linux/init.h> -#include <linux/debugfs.h> -#include <linux/ftrace.h> -#include <linux/kallsyms.h> -#include <linux/module.h> - -#include "trace.h" - -static struct trace_array *power_trace; -static int __read_mostly trace_power_enabled; - - -static void start_power_trace(struct trace_array *tr) -{ - trace_power_enabled = 1; -} - -static void stop_power_trace(struct trace_array *tr) -{ - trace_power_enabled = 0; -} - - -static int power_trace_init(struct trace_array *tr) -{ - int cpu; - power_trace = tr; - - trace_power_enabled = 1; - - for_each_cpu(cpu, cpu_possible_mask) - tracing_reset(tr, cpu); - return 0; -} - -static enum print_line_t power_print_line(struct trace_iterator *iter) -{ - int ret = 0; - struct trace_entry *entry = iter->ent; - struct trace_power *field ; - struct power_trace *it; - struct trace_seq *s = &iter->seq; - struct timespec stamp; - struct timespec duration; - - trace_assign_type(field, entry); - it = &field->state_data; - stamp = ktime_to_timespec(it->stamp); - duration = ktime_to_timespec(ktime_sub(it->end, it->stamp)); - - if (entry->type == TRACE_POWER) { - if (it->type == POWER_CSTATE) - ret = trace_seq_printf(s, "[%5ld.%09ld] CSTATE: Going to C%i on cpu %i for %ld.%09ld\n", - stamp.tv_sec, - stamp.tv_nsec, - it->state, iter->cpu, - duration.tv_sec, - duration.tv_nsec); - if (it->type == POWER_PSTATE) - ret = trace_seq_printf(s, "[%5ld.%09ld] PSTATE: Going to P%i on cpu %i\n", - stamp.tv_sec, - stamp.tv_nsec, - it->state, iter->cpu); - if (!ret) - return TRACE_TYPE_PARTIAL_LINE; - return TRACE_TYPE_HANDLED; - } - return TRACE_TYPE_UNHANDLED; -} - -static struct tracer power_tracer __read_mostly = -{ - .name = "power", - .init = power_trace_init, - .start = start_power_trace, - .stop = stop_power_trace, - .reset = stop_power_trace, - .print_line = power_print_line, -}; - -static int init_power_trace(void) -{ - return register_tracer(&power_tracer); -} -device_initcall(init_power_trace); - -void trace_power_start(struct power_trace *it, unsigned int type, - unsigned int level) -{ - if (!trace_power_enabled) - return; - - memset(it, 0, sizeof(struct power_trace)); - it->state = level; - it->type = type; - it->stamp = ktime_get(); -} -EXPORT_SYMBOL_GPL(trace_power_start); - - -void trace_power_end(struct power_trace *it) -{ - struct ring_buffer_event *event; - struct trace_power *entry; - struct trace_array_cpu *data; - unsigned long irq_flags; - struct trace_array *tr = power_trace; - - if (!trace_power_enabled) - return; - - preempt_disable(); - it->end = ktime_get(); - data = tr->data[smp_processor_id()]; - - event = ring_buffer_lock_reserve(tr->buffer, sizeof(*entry), - &irq_flags); - if (!event) - goto out; - entry = ring_buffer_event_data(event); - tracing_generic_entry_update(&entry->ent, 0, 0); - entry->ent.type = TRACE_POWER; - entry->state_data = *it; - ring_buffer_unlock_commit(tr->buffer, event, irq_flags); - - trace_wake_up(); - - out: - preempt_enable(); -} -EXPORT_SYMBOL_GPL(trace_power_end); - -void trace_power_mark(struct power_trace *it, unsigned int type, - unsigned int level) -{ - struct ring_buffer_event *event; - struct trace_power *entry; - struct trace_array_cpu *data; - unsigned long irq_flags; - struct trace_array *tr = power_trace; - - if (!trace_power_enabled) - return; - - memset(it, 0, sizeof(struct power_trace)); - it->state = level; - it->type = type; - it->stamp = ktime_get(); - preempt_disable(); - it->end = it->stamp; - data = tr->data[smp_processor_id()]; - - event = ring_buffer_lock_reserve(tr->buffer, sizeof(*entry), - &irq_flags); - if (!event) - goto out; - entry = ring_buffer_event_data(event); - tracing_generic_entry_update(&entry->ent, 0, 0); - entry->ent.type = TRACE_POWER; - entry->state_data = *it; - ring_buffer_unlock_commit(tr->buffer, event, irq_flags); - - trace_wake_up(); - - out: - preempt_enable(); -} -EXPORT_SYMBOL_GPL(trace_power_mark); diff --git a/kernel/trace/trace_printk.c b/kernel/trace/trace_printk.c new file mode 100644 index 00000000000..2547d8813cf --- /dev/null +++ b/kernel/trace/trace_printk.c @@ -0,0 +1,251 @@ +/* + * trace binary printk + * + * Copyright (C) 2008 Lai Jiangshan <laijs@cn.fujitsu.com> + * + */ +#include <linux/seq_file.h> +#include <linux/debugfs.h> +#include <linux/uaccess.h> +#include <linux/kernel.h> +#include <linux/ftrace.h> +#include <linux/string.h> +#include <linux/module.h> +#include <linux/mutex.h> +#include <linux/ctype.h> +#include <linux/list.h> +#include <linux/slab.h> +#include <linux/fs.h> + +#include "trace.h" + +#ifdef CONFIG_MODULES + +/* + * modules trace_printk()'s formats are autosaved in struct trace_bprintk_fmt + * which are queued on trace_bprintk_fmt_list. + */ +static LIST_HEAD(trace_bprintk_fmt_list); + +/* serialize accesses to trace_bprintk_fmt_list */ +static DEFINE_MUTEX(btrace_mutex); + +struct trace_bprintk_fmt { + struct list_head list; + char fmt[0]; +}; + +static inline struct trace_bprintk_fmt *lookup_format(const char *fmt) +{ + struct trace_bprintk_fmt *pos; + list_for_each_entry(pos, &trace_bprintk_fmt_list, list) { + if (!strcmp(pos->fmt, fmt)) + return pos; + } + return NULL; +} + +static +void hold_module_trace_bprintk_format(const char **start, const char **end) +{ + const char **iter; + + mutex_lock(&btrace_mutex); + for (iter = start; iter < end; iter++) { + struct trace_bprintk_fmt *tb_fmt = lookup_format(*iter); + if (tb_fmt) { + *iter = tb_fmt->fmt; + continue; + } + + tb_fmt = kmalloc(offsetof(struct trace_bprintk_fmt, fmt) + + strlen(*iter) + 1, GFP_KERNEL); + if (tb_fmt) { + list_add_tail(&tb_fmt->list, &trace_bprintk_fmt_list); + strcpy(tb_fmt->fmt, *iter); + *iter = tb_fmt->fmt; + } else + *iter = NULL; + } + mutex_unlock(&btrace_mutex); +} + +static int module_trace_bprintk_format_notify(struct notifier_block *self, + unsigned long val, void *data) +{ + struct module *mod = data; + if (mod->num_trace_bprintk_fmt) { + const char **start = mod->trace_bprintk_fmt_start; + const char **end = start + mod->num_trace_bprintk_fmt; + + if (val == MODULE_STATE_COMING) + hold_module_trace_bprintk_format(start, end); + } + return 0; +} + +#else /* !CONFIG_MODULES */ +__init static int +module_trace_bprintk_format_notify(struct notifier_block *self, + unsigned long val, void *data) +{ + return 0; +} +#endif /* CONFIG_MODULES */ + + +__initdata_or_module static +struct notifier_block module_trace_bprintk_format_nb = { + .notifier_call = module_trace_bprintk_format_notify, +}; + +int __trace_bprintk(unsigned long ip, const char *fmt, ...) + { + int ret; + va_list ap; + + if (unlikely(!fmt)) + return 0; + + if (!(trace_flags & TRACE_ITER_PRINTK)) + return 0; + + va_start(ap, fmt); + ret = trace_vbprintk(ip, fmt, ap); + va_end(ap); + return ret; +} +EXPORT_SYMBOL_GPL(__trace_bprintk); + +int __ftrace_vbprintk(unsigned long ip, const char *fmt, va_list ap) + { + if (unlikely(!fmt)) + return 0; + + if (!(trace_flags & TRACE_ITER_PRINTK)) + return 0; + + return trace_vbprintk(ip, fmt, ap); +} +EXPORT_SYMBOL_GPL(__ftrace_vbprintk); + +int __trace_printk(unsigned long ip, const char *fmt, ...) +{ + int ret; + va_list ap; + + if (!(trace_flags & TRACE_ITER_PRINTK)) + return 0; + + va_start(ap, fmt); + ret = trace_vprintk(ip, fmt, ap); + va_end(ap); + return ret; +} +EXPORT_SYMBOL_GPL(__trace_printk); + +int __ftrace_vprintk(unsigned long ip, const char *fmt, va_list ap) +{ + if (!(trace_flags & TRACE_ITER_PRINTK)) + return 0; + + return trace_vprintk(ip, fmt, ap); +} +EXPORT_SYMBOL_GPL(__ftrace_vprintk); + +static void * +t_start(struct seq_file *m, loff_t *pos) +{ + const char **fmt = __start___trace_bprintk_fmt + *pos; + + if ((unsigned long)fmt >= (unsigned long)__stop___trace_bprintk_fmt) + return NULL; + return fmt; +} + +static void *t_next(struct seq_file *m, void * v, loff_t *pos) +{ + (*pos)++; + return t_start(m, pos); +} + +static int t_show(struct seq_file *m, void *v) +{ + const char **fmt = v; + const char *str = *fmt; + int i; + + seq_printf(m, "0x%lx : \"", *(unsigned long *)fmt); + + /* + * Tabs and new lines need to be converted. + */ + for (i = 0; str[i]; i++) { + switch (str[i]) { + case '\n': + seq_puts(m, "\\n"); + break; + case '\t': + seq_puts(m, "\\t"); + break; + case '\\': + seq_puts(m, "\\"); + break; + case '"': + seq_puts(m, "\\\""); + break; + default: + seq_putc(m, str[i]); + } + } + seq_puts(m, "\"\n"); + + return 0; +} + +static void t_stop(struct seq_file *m, void *p) +{ +} + +static const struct seq_operations show_format_seq_ops = { + .start = t_start, + .next = t_next, + .show = t_show, + .stop = t_stop, +}; + +static int +ftrace_formats_open(struct inode *inode, struct file *file) +{ + return seq_open(file, &show_format_seq_ops); +} + +static const struct file_operations ftrace_formats_fops = { + .open = ftrace_formats_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release, +}; + +static __init int init_trace_printk_function_export(void) +{ + struct dentry *d_tracer; + + d_tracer = tracing_init_dentry(); + if (!d_tracer) + return 0; + + trace_create_file("printk_formats", 0444, d_tracer, + NULL, &ftrace_formats_fops); + + return 0; +} + +fs_initcall(init_trace_printk_function_export); + +static __init int init_trace_printk(void) +{ + return register_module_notifier(&module_trace_bprintk_format_nb); +} + +early_initcall(init_trace_printk); diff --git a/kernel/trace/trace_sched_switch.c b/kernel/trace/trace_sched_switch.c index df175cb4564..5fca0f51fde 100644 --- a/kernel/trace/trace_sched_switch.c +++ b/kernel/trace/trace_sched_switch.c @@ -10,7 +10,7 @@ #include <linux/kallsyms.h> #include <linux/uaccess.h> #include <linux/ftrace.h> -#include <trace/sched.h> +#include <trace/events/sched.h> #include "trace.h" @@ -18,6 +18,36 @@ static struct trace_array *ctx_trace; static int __read_mostly tracer_enabled; static int sched_ref; static DEFINE_MUTEX(sched_register_mutex); +static int sched_stopped; + + +void +tracing_sched_switch_trace(struct trace_array *tr, + struct task_struct *prev, + struct task_struct *next, + unsigned long flags, int pc) +{ + struct ftrace_event_call *call = &event_context_switch; + struct ring_buffer *buffer = tr->buffer; + struct ring_buffer_event *event; + struct ctx_switch_entry *entry; + + event = trace_buffer_lock_reserve(buffer, TRACE_CTX, + sizeof(*entry), flags, pc); + if (!event) + return; + entry = ring_buffer_event_data(event); + entry->prev_pid = prev->pid; + entry->prev_prio = prev->prio; + entry->prev_state = prev->state; + entry->next_pid = next->pid; + entry->next_prio = next->prio; + entry->next_state = next->state; + entry->next_cpu = task_cpu(next); + + if (!filter_check_discard(call, entry, buffer, event)) + trace_buffer_unlock_commit(buffer, event, flags, pc); +} static void probe_sched_switch(struct rq *__rq, struct task_struct *prev, @@ -28,13 +58,13 @@ probe_sched_switch(struct rq *__rq, struct task_struct *prev, int cpu; int pc; - if (!sched_ref) + if (unlikely(!sched_ref)) return; tracing_record_cmdline(prev); tracing_record_cmdline(next); - if (!tracer_enabled) + if (!tracer_enabled || sched_stopped) return; pc = preempt_count(); @@ -43,11 +73,41 @@ probe_sched_switch(struct rq *__rq, struct task_struct *prev, data = ctx_trace->data[cpu]; if (likely(!atomic_read(&data->disabled))) - tracing_sched_switch_trace(ctx_trace, data, prev, next, flags, pc); + tracing_sched_switch_trace(ctx_trace, prev, next, flags, pc); local_irq_restore(flags); } +void +tracing_sched_wakeup_trace(struct trace_array *tr, + struct task_struct *wakee, + struct task_struct *curr, + unsigned long flags, int pc) +{ + struct ftrace_event_call *call = &event_wakeup; + struct ring_buffer_event *event; + struct ctx_switch_entry *entry; + struct ring_buffer *buffer = tr->buffer; + + event = trace_buffer_lock_reserve(buffer, TRACE_WAKE, + sizeof(*entry), flags, pc); + if (!event) + return; + entry = ring_buffer_event_data(event); + entry->prev_pid = curr->pid; + entry->prev_prio = curr->prio; + entry->prev_state = curr->state; + entry->next_pid = wakee->pid; + entry->next_prio = wakee->prio; + entry->next_state = wakee->state; + entry->next_cpu = task_cpu(wakee); + + if (!filter_check_discard(call, entry, buffer, event)) + ring_buffer_unlock_commit(buffer, event); + ftrace_trace_stack(tr->buffer, flags, 6, pc); + ftrace_trace_userstack(tr->buffer, flags, pc); +} + static void probe_sched_wakeup(struct rq *__rq, struct task_struct *wakee, int success) { @@ -55,18 +115,21 @@ probe_sched_wakeup(struct rq *__rq, struct task_struct *wakee, int success) unsigned long flags; int cpu, pc; - if (!likely(tracer_enabled)) + if (unlikely(!sched_ref)) return; - pc = preempt_count(); tracing_record_cmdline(current); + if (!tracer_enabled || sched_stopped) + return; + + pc = preempt_count(); local_irq_save(flags); cpu = raw_smp_processor_id(); data = ctx_trace->data[cpu]; if (likely(!atomic_read(&data->disabled))) - tracing_sched_wakeup_trace(ctx_trace, data, wakee, current, + tracing_sched_wakeup_trace(ctx_trace, wakee, current, flags, pc); local_irq_restore(flags); @@ -93,7 +156,7 @@ static int tracing_sched_register(void) ret = register_trace_sched_switch(probe_sched_switch); if (ret) { pr_info("sched trace: Couldn't activate tracepoint" - " probe to kernel_sched_schedule\n"); + " probe to kernel_sched_switch\n"); goto fail_deprobe_wake_new; } @@ -185,12 +248,6 @@ void tracing_sched_switch_assign_trace(struct trace_array *tr) ctx_trace = tr; } -static void start_sched_trace(struct trace_array *tr) -{ - tracing_reset_online_cpus(tr); - tracing_start_sched_switch_record(); -} - static void stop_sched_trace(struct trace_array *tr) { tracing_stop_sched_switch_record(); @@ -199,7 +256,8 @@ static void stop_sched_trace(struct trace_array *tr) static int sched_switch_trace_init(struct trace_array *tr) { ctx_trace = tr; - start_sched_trace(tr); + tracing_reset_online_cpus(tr); + tracing_start_sched_switch_record(); return 0; } @@ -211,13 +269,12 @@ static void sched_switch_trace_reset(struct trace_array *tr) static void sched_switch_trace_start(struct trace_array *tr) { - tracing_reset_online_cpus(tr); - tracing_start_sched_switch(); + sched_stopped = 0; } static void sched_switch_trace_stop(struct trace_array *tr) { - tracing_stop_sched_switch(); + sched_stopped = 1; } static struct tracer sched_switch_trace __read_mostly = @@ -227,6 +284,7 @@ static struct tracer sched_switch_trace __read_mostly = .reset = sched_switch_trace_reset, .start = sched_switch_trace_start, .stop = sched_switch_trace_stop, + .wait_pipe = poll_wait_pipe, #ifdef CONFIG_FTRACE_SELFTEST .selftest = trace_selftest_startup_sched_switch, #endif diff --git a/kernel/trace/trace_sched_wakeup.c b/kernel/trace/trace_sched_wakeup.c index 42ae1e77b6b..26185d72767 100644 --- a/kernel/trace/trace_sched_wakeup.c +++ b/kernel/trace/trace_sched_wakeup.c @@ -15,7 +15,7 @@ #include <linux/kallsyms.h> #include <linux/uaccess.h> #include <linux/ftrace.h> -#include <trace/sched.h> +#include <trace/events/sched.h> #include "trace.h" @@ -24,13 +24,17 @@ static int __read_mostly tracer_enabled; static struct task_struct *wakeup_task; static int wakeup_cpu; +static int wakeup_current_cpu; static unsigned wakeup_prio = -1; +static int wakeup_rt; static raw_spinlock_t wakeup_lock = (raw_spinlock_t)__RAW_SPIN_LOCK_UNLOCKED; static void __wakeup_reset(struct trace_array *tr); +static int save_lat_flag; + #ifdef CONFIG_FUNCTION_TRACER /* * irqsoff uses its own tracer function to keep the overhead down: @@ -53,33 +57,23 @@ wakeup_tracer_call(unsigned long ip, unsigned long parent_ip) resched = ftrace_preempt_disable(); cpu = raw_smp_processor_id(); + if (cpu != wakeup_current_cpu) + goto out_enable; + data = tr->data[cpu]; disabled = atomic_inc_return(&data->disabled); if (unlikely(disabled != 1)) goto out; local_irq_save(flags); - __raw_spin_lock(&wakeup_lock); - - if (unlikely(!wakeup_task)) - goto unlock; - - /* - * The task can't disappear because it needs to - * wake up first, and we have the wakeup_lock. - */ - if (task_cpu(wakeup_task) != cpu) - goto unlock; - trace_function(tr, data, ip, parent_ip, flags, pc); + trace_function(tr, ip, parent_ip, flags, pc); - unlock: - __raw_spin_unlock(&wakeup_lock); local_irq_restore(flags); out: atomic_dec(&data->disabled); - + out_enable: ftrace_preempt_enable(resched); } @@ -104,11 +98,18 @@ static int report_latency(cycle_t delta) return 1; } +static void probe_wakeup_migrate_task(struct task_struct *task, int cpu) +{ + if (task != wakeup_task) + return; + + wakeup_current_cpu = cpu; +} + static void notrace probe_wakeup_sched_switch(struct rq *rq, struct task_struct *prev, struct task_struct *next) { - unsigned long latency = 0, t0 = 0, t1 = 0; struct trace_array_cpu *data; cycle_t T0, T1, delta; unsigned long flags; @@ -135,9 +136,6 @@ probe_wakeup_sched_switch(struct rq *rq, struct task_struct *prev, pc = preempt_count(); - /* The task we are waiting for is waking up */ - data = wakeup_trace->data[wakeup_cpu]; - /* disable local data, not wakeup_cpu data */ cpu = raw_smp_processor_id(); disabled = atomic_inc_return(&wakeup_trace->data[cpu]->disabled); @@ -151,12 +149,12 @@ probe_wakeup_sched_switch(struct rq *rq, struct task_struct *prev, if (unlikely(!tracer_enabled || next != wakeup_task)) goto out_unlock; - trace_function(wakeup_trace, data, CALLER_ADDR1, CALLER_ADDR2, flags, pc); + /* The task we are waiting for is waking up */ + data = wakeup_trace->data[wakeup_cpu]; + + trace_function(wakeup_trace, CALLER_ADDR0, CALLER_ADDR1, flags, pc); + tracing_sched_switch_trace(wakeup_trace, prev, next, flags, pc); - /* - * usecs conversion is slow so we try to delay the conversion - * as long as possible: - */ T0 = data->preempt_timestamp; T1 = ftrace_now(cpu); delta = T1-T0; @@ -164,13 +162,10 @@ probe_wakeup_sched_switch(struct rq *rq, struct task_struct *prev, if (!report_latency(delta)) goto out_unlock; - latency = nsecs_to_usecs(delta); - - tracing_max_latency = delta; - t0 = nsecs_to_usecs(T0); - t1 = nsecs_to_usecs(T1); - - update_max_tr(wakeup_trace, wakeup_task, wakeup_cpu); + if (likely(!is_tracing_stopped())) { + tracing_max_latency = delta; + update_max_tr(wakeup_trace, wakeup_task, wakeup_cpu); + } out_unlock: __wakeup_reset(wakeup_trace); @@ -182,14 +177,6 @@ out: static void __wakeup_reset(struct trace_array *tr) { - struct trace_array_cpu *data; - int cpu; - - for_each_possible_cpu(cpu) { - data = tr->data[cpu]; - tracing_reset(tr, cpu); - } - wakeup_cpu = -1; wakeup_prio = -1; @@ -203,6 +190,8 @@ static void wakeup_reset(struct trace_array *tr) { unsigned long flags; + tracing_reset_online_cpus(tr); + local_irq_save(flags); __raw_spin_lock(&wakeup_lock); __wakeup_reset(tr); @@ -213,6 +202,7 @@ static void wakeup_reset(struct trace_array *tr) static void probe_wakeup(struct rq *rq, struct task_struct *p, int success) { + struct trace_array_cpu *data; int cpu = smp_processor_id(); unsigned long flags; long disabled; @@ -224,7 +214,7 @@ probe_wakeup(struct rq *rq, struct task_struct *p, int success) tracing_record_cmdline(p); tracing_record_cmdline(current); - if (likely(!rt_task(p)) || + if ((wakeup_rt && !rt_task(p)) || p->prio >= wakeup_prio || p->prio >= current->prio) return; @@ -245,6 +235,7 @@ probe_wakeup(struct rq *rq, struct task_struct *p, int success) __wakeup_reset(wakeup_trace); wakeup_cpu = task_cpu(p); + wakeup_current_cpu = wakeup_cpu; wakeup_prio = p->prio; wakeup_task = p; @@ -252,9 +243,16 @@ probe_wakeup(struct rq *rq, struct task_struct *p, int success) local_save_flags(flags); - wakeup_trace->data[wakeup_cpu]->preempt_timestamp = ftrace_now(cpu); - trace_function(wakeup_trace, wakeup_trace->data[wakeup_cpu], - CALLER_ADDR1, CALLER_ADDR2, flags, pc); + data = wakeup_trace->data[wakeup_cpu]; + data->preempt_timestamp = ftrace_now(cpu); + tracing_sched_wakeup_trace(wakeup_trace, p, current, flags, pc); + + /* + * We must be careful in using CALLER_ADDR2. But since wake_up + * is not called by an assembly function (where as schedule is) + * it should be safe to use it here. + */ + trace_function(wakeup_trace, CALLER_ADDR1, CALLER_ADDR2, flags, pc); out_locked: __raw_spin_unlock(&wakeup_lock); @@ -262,12 +260,6 @@ out: atomic_dec(&wakeup_trace->data[cpu]->disabled); } -/* - * save_tracer_enabled is used to save the state of the tracer_enabled - * variable when we disable it when we open a trace output file. - */ -static int save_tracer_enabled; - static void start_wakeup_tracer(struct trace_array *tr) { int ret; @@ -289,10 +281,17 @@ static void start_wakeup_tracer(struct trace_array *tr) ret = register_trace_sched_switch(probe_wakeup_sched_switch); if (ret) { pr_info("sched trace: Couldn't activate tracepoint" - " probe to kernel_sched_schedule\n"); + " probe to kernel_sched_switch\n"); goto fail_deprobe_wake_new; } + ret = register_trace_sched_migrate_task(probe_wakeup_migrate_task); + if (ret) { + pr_info("wakeup trace: Couldn't activate tracepoint" + " probe to kernel_sched_migrate_task\n"); + return; + } + wakeup_reset(tr); /* @@ -306,13 +305,10 @@ static void start_wakeup_tracer(struct trace_array *tr) register_ftrace_function(&trace_ops); - if (tracing_is_enabled()) { + if (tracing_is_enabled()) tracer_enabled = 1; - save_tracer_enabled = 1; - } else { + else tracer_enabled = 0; - save_tracer_enabled = 0; - } return; fail_deprobe_wake_new: @@ -324,54 +320,55 @@ fail_deprobe: static void stop_wakeup_tracer(struct trace_array *tr) { tracer_enabled = 0; - save_tracer_enabled = 0; unregister_ftrace_function(&trace_ops); unregister_trace_sched_switch(probe_wakeup_sched_switch); unregister_trace_sched_wakeup_new(probe_wakeup); unregister_trace_sched_wakeup(probe_wakeup); + unregister_trace_sched_migrate_task(probe_wakeup_migrate_task); } -static int wakeup_tracer_init(struct trace_array *tr) +static int __wakeup_tracer_init(struct trace_array *tr) { + save_lat_flag = trace_flags & TRACE_ITER_LATENCY_FMT; + trace_flags |= TRACE_ITER_LATENCY_FMT; + tracing_max_latency = 0; wakeup_trace = tr; start_wakeup_tracer(tr); return 0; } +static int wakeup_tracer_init(struct trace_array *tr) +{ + wakeup_rt = 0; + return __wakeup_tracer_init(tr); +} + +static int wakeup_rt_tracer_init(struct trace_array *tr) +{ + wakeup_rt = 1; + return __wakeup_tracer_init(tr); +} + static void wakeup_tracer_reset(struct trace_array *tr) { stop_wakeup_tracer(tr); /* make sure we put back any tasks we are tracing */ wakeup_reset(tr); + + if (!save_lat_flag) + trace_flags &= ~TRACE_ITER_LATENCY_FMT; } static void wakeup_tracer_start(struct trace_array *tr) { wakeup_reset(tr); tracer_enabled = 1; - save_tracer_enabled = 1; } static void wakeup_tracer_stop(struct trace_array *tr) { tracer_enabled = 0; - save_tracer_enabled = 0; -} - -static void wakeup_tracer_open(struct trace_iterator *iter) -{ - /* stop the trace while dumping */ - tracer_enabled = 0; -} - -static void wakeup_tracer_close(struct trace_iterator *iter) -{ - /* forget about any processes we were recording */ - if (save_tracer_enabled) { - wakeup_reset(iter->tr); - tracer_enabled = 1; - } } static struct tracer wakeup_tracer __read_mostly = @@ -381,8 +378,20 @@ static struct tracer wakeup_tracer __read_mostly = .reset = wakeup_tracer_reset, .start = wakeup_tracer_start, .stop = wakeup_tracer_stop, - .open = wakeup_tracer_open, - .close = wakeup_tracer_close, + .print_max = 1, +#ifdef CONFIG_FTRACE_SELFTEST + .selftest = trace_selftest_startup_wakeup, +#endif +}; + +static struct tracer wakeup_rt_tracer __read_mostly = +{ + .name = "wakeup_rt", + .init = wakeup_rt_tracer_init, + .reset = wakeup_tracer_reset, + .start = wakeup_tracer_start, + .stop = wakeup_tracer_stop, + .wait_pipe = poll_wait_pipe, .print_max = 1, #ifdef CONFIG_FTRACE_SELFTEST .selftest = trace_selftest_startup_wakeup, @@ -397,6 +406,10 @@ __init static int init_wakeup_tracer(void) if (ret) return ret; + ret = register_tracer(&wakeup_rt_tracer); + if (ret) + return ret; + return 0; } device_initcall(init_wakeup_tracer); diff --git a/kernel/trace/trace_selftest.c b/kernel/trace/trace_selftest.c index bc8e80a86bc..dc98309e839 100644 --- a/kernel/trace/trace_selftest.c +++ b/kernel/trace/trace_selftest.c @@ -1,5 +1,6 @@ /* Include in trace.c */ +#include <linux/stringify.h> #include <linux/kthread.h> #include <linux/delay.h> @@ -9,11 +10,14 @@ static inline int trace_valid_entry(struct trace_entry *entry) case TRACE_FN: case TRACE_CTX: case TRACE_WAKE: - case TRACE_CONT: case TRACE_STACK: case TRACE_PRINT: case TRACE_SPECIAL: case TRACE_BRANCH: + case TRACE_GRAPH_ENT: + case TRACE_GRAPH_RET: + case TRACE_HW_BRANCHES: + case TRACE_KSYM: return 1; } return 0; @@ -99,9 +103,6 @@ static inline void warn_failed_init_tracer(struct tracer *trace, int init_ret) #ifdef CONFIG_DYNAMIC_FTRACE -#define __STR(x) #x -#define STR(x) __STR(x) - /* Test dynamic code modification and ftrace filters */ int trace_selftest_startup_dynamic_tracing(struct tracer *trace, struct trace_array *tr, @@ -125,17 +126,17 @@ int trace_selftest_startup_dynamic_tracing(struct tracer *trace, func(); /* - * Some archs *cough*PowerPC*cough* add charachters to the + * Some archs *cough*PowerPC*cough* add characters to the * start of the function names. We simply put a '*' to - * accomodate them. + * accommodate them. */ - func_name = "*" STR(DYN_FTRACE_TEST_NAME); + func_name = "*" __stringify(DYN_FTRACE_TEST_NAME); /* filter only on our function */ ftrace_set_filter(func_name, strlen(func_name), 1); /* enable tracing */ - ret = trace->init(tr); + ret = tracer_init(trace, tr); if (ret) { warn_failed_init_tracer(trace, ret); goto out; @@ -189,6 +190,7 @@ int trace_selftest_startup_dynamic_tracing(struct tracer *trace, #else # define trace_selftest_startup_dynamic_tracing(trace, tr, func) ({ 0; }) #endif /* CONFIG_DYNAMIC_FTRACE */ + /* * Simple verification test of ftrace function tracer. * Enable ftrace, sleep 1/10 second, and then read the trace @@ -209,7 +211,7 @@ trace_selftest_startup_function(struct tracer *trace, struct trace_array *tr) ftrace_enabled = 1; tracer_enabled = 1; - ret = trace->init(tr); + ret = tracer_init(trace, tr); if (ret) { warn_failed_init_tracer(trace, ret); goto out; @@ -247,6 +249,91 @@ trace_selftest_startup_function(struct tracer *trace, struct trace_array *tr) } #endif /* CONFIG_FUNCTION_TRACER */ + +#ifdef CONFIG_FUNCTION_GRAPH_TRACER + +/* Maximum number of functions to trace before diagnosing a hang */ +#define GRAPH_MAX_FUNC_TEST 100000000 + +static void __ftrace_dump(bool disable_tracing); +static unsigned int graph_hang_thresh; + +/* Wrap the real function entry probe to avoid possible hanging */ +static int trace_graph_entry_watchdog(struct ftrace_graph_ent *trace) +{ + /* This is harmlessly racy, we want to approximately detect a hang */ + if (unlikely(++graph_hang_thresh > GRAPH_MAX_FUNC_TEST)) { + ftrace_graph_stop(); + printk(KERN_WARNING "BUG: Function graph tracer hang!\n"); + if (ftrace_dump_on_oops) + __ftrace_dump(false); + return 0; + } + + return trace_graph_entry(trace); +} + +/* + * Pretty much the same than for the function tracer from which the selftest + * has been borrowed. + */ +int +trace_selftest_startup_function_graph(struct tracer *trace, + struct trace_array *tr) +{ + int ret; + unsigned long count; + + /* + * Simulate the init() callback but we attach a watchdog callback + * to detect and recover from possible hangs + */ + tracing_reset_online_cpus(tr); + set_graph_array(tr); + ret = register_ftrace_graph(&trace_graph_return, + &trace_graph_entry_watchdog); + if (ret) { + warn_failed_init_tracer(trace, ret); + goto out; + } + tracing_start_cmdline_record(); + + /* Sleep for a 1/10 of a second */ + msleep(100); + + /* Have we just recovered from a hang? */ + if (graph_hang_thresh > GRAPH_MAX_FUNC_TEST) { + tracing_selftest_disabled = true; + ret = -1; + goto out; + } + + tracing_stop(); + + /* check the trace buffer */ + ret = trace_test_buffer(tr, &count); + + trace->reset(tr); + tracing_start(); + + if (!ret && !count) { + printk(KERN_CONT ".. no entries found .."); + ret = -1; + goto out; + } + + /* Don't test dynamic tracing, the function tracer already did */ + +out: + /* Stop it if we failed */ + if (ret) + ftrace_graph_stop(); + + return ret; +} +#endif /* CONFIG_FUNCTION_GRAPH_TRACER */ + + #ifdef CONFIG_IRQSOFF_TRACER int trace_selftest_startup_irqsoff(struct tracer *trace, struct trace_array *tr) @@ -256,7 +343,7 @@ trace_selftest_startup_irqsoff(struct tracer *trace, struct trace_array *tr) int ret; /* start the tracing */ - ret = trace->init(tr); + ret = tracer_init(trace, tr); if (ret) { warn_failed_init_tracer(trace, ret); return ret; @@ -268,6 +355,14 @@ trace_selftest_startup_irqsoff(struct tracer *trace, struct trace_array *tr) local_irq_disable(); udelay(100); local_irq_enable(); + + /* + * Stop the tracer to avoid a warning subsequent + * to buffer flipping failure because tracing_stop() + * disables the tr and max buffers, making flipping impossible + * in case of parallels max irqs off latencies. + */ + trace->stop(tr); /* stop the tracing. */ tracing_stop(); /* check both trace buffers */ @@ -310,7 +405,7 @@ trace_selftest_startup_preemptoff(struct tracer *trace, struct trace_array *tr) } /* start the tracing */ - ret = trace->init(tr); + ret = tracer_init(trace, tr); if (ret) { warn_failed_init_tracer(trace, ret); return ret; @@ -322,6 +417,14 @@ trace_selftest_startup_preemptoff(struct tracer *trace, struct trace_array *tr) preempt_disable(); udelay(100); preempt_enable(); + + /* + * Stop the tracer to avoid a warning subsequent + * to buffer flipping failure because tracing_stop() + * disables the tr and max buffers, making flipping impossible + * in case of parallels max preempt off latencies. + */ + trace->stop(tr); /* stop the tracing. */ tracing_stop(); /* check both trace buffers */ @@ -364,10 +467,10 @@ trace_selftest_startup_preemptirqsoff(struct tracer *trace, struct trace_array * } /* start the tracing */ - ret = trace->init(tr); + ret = tracer_init(trace, tr); if (ret) { warn_failed_init_tracer(trace, ret); - goto out; + goto out_no_start; } /* reset the max latency */ @@ -381,31 +484,35 @@ trace_selftest_startup_preemptirqsoff(struct tracer *trace, struct trace_array * /* reverse the order of preempt vs irqs */ local_irq_enable(); + /* + * Stop the tracer to avoid a warning subsequent + * to buffer flipping failure because tracing_stop() + * disables the tr and max buffers, making flipping impossible + * in case of parallels max irqs/preempt off latencies. + */ + trace->stop(tr); /* stop the tracing. */ tracing_stop(); /* check both trace buffers */ ret = trace_test_buffer(tr, NULL); - if (ret) { - tracing_start(); + if (ret) goto out; - } ret = trace_test_buffer(&max_tr, &count); - if (ret) { - tracing_start(); + if (ret) goto out; - } if (!ret && !count) { printk(KERN_CONT ".. no entries found .."); ret = -1; - tracing_start(); goto out; } /* do the test by disabling interrupts first this time */ tracing_max_latency = 0; tracing_start(); + trace->start(tr); + preempt_disable(); local_irq_disable(); udelay(100); @@ -413,6 +520,7 @@ trace_selftest_startup_preemptirqsoff(struct tracer *trace, struct trace_array * /* reverse the order of preempt vs irqs */ local_irq_enable(); + trace->stop(tr); /* stop the tracing. */ tracing_stop(); /* check both trace buffers */ @@ -428,9 +536,10 @@ trace_selftest_startup_preemptirqsoff(struct tracer *trace, struct trace_array * goto out; } - out: - trace->reset(tr); +out: tracing_start(); +out_no_start: + trace->reset(tr); tracing_max_latency = save_max; return ret; @@ -496,7 +605,7 @@ trace_selftest_startup_wakeup(struct tracer *trace, struct trace_array *tr) wait_for_completion(&isrt); /* start the tracing */ - ret = trace->init(tr); + ret = tracer_init(trace, tr); if (ret) { warn_failed_init_tracer(trace, ret); return ret; @@ -557,7 +666,7 @@ trace_selftest_startup_sched_switch(struct tracer *trace, struct trace_array *tr int ret; /* start the tracing */ - ret = trace->init(tr); + ret = tracer_init(trace, tr); if (ret) { warn_failed_init_tracer(trace, ret); return ret; @@ -589,10 +698,10 @@ trace_selftest_startup_sysprof(struct tracer *trace, struct trace_array *tr) int ret; /* start the tracing */ - ret = trace->init(tr); + ret = tracer_init(trace, tr); if (ret) { warn_failed_init_tracer(trace, ret); - return 0; + return ret; } /* Sleep for a 1/10 of a second */ @@ -604,6 +713,11 @@ trace_selftest_startup_sysprof(struct tracer *trace, struct trace_array *tr) trace->reset(tr); tracing_start(); + if (!ret && !count) { + printk(KERN_CONT ".. no entries found .."); + ret = -1; + } + return ret; } #endif /* CONFIG_SYSPROF_TRACER */ @@ -616,7 +730,7 @@ trace_selftest_startup_branch(struct tracer *trace, struct trace_array *tr) int ret; /* start the tracing */ - ret = trace->init(tr); + ret = tracer_init(trace, tr); if (ret) { warn_failed_init_tracer(trace, ret); return ret; @@ -631,6 +745,121 @@ trace_selftest_startup_branch(struct tracer *trace, struct trace_array *tr) trace->reset(tr); tracing_start(); + if (!ret && !count) { + printk(KERN_CONT ".. no entries found .."); + ret = -1; + } + return ret; } #endif /* CONFIG_BRANCH_TRACER */ + +#ifdef CONFIG_HW_BRANCH_TRACER +int +trace_selftest_startup_hw_branches(struct tracer *trace, + struct trace_array *tr) +{ + struct trace_iterator *iter; + struct tracer tracer; + unsigned long count; + int ret; + + if (!trace->open) { + printk(KERN_CONT "missing open function..."); + return -1; + } + + ret = tracer_init(trace, tr); + if (ret) { + warn_failed_init_tracer(trace, ret); + return ret; + } + + /* + * The hw-branch tracer needs to collect the trace from the various + * cpu trace buffers - before tracing is stopped. + */ + iter = kzalloc(sizeof(*iter), GFP_KERNEL); + if (!iter) + return -ENOMEM; + + memcpy(&tracer, trace, sizeof(tracer)); + + iter->trace = &tracer; + iter->tr = tr; + iter->pos = -1; + mutex_init(&iter->mutex); + + trace->open(iter); + + mutex_destroy(&iter->mutex); + kfree(iter); + + tracing_stop(); + + ret = trace_test_buffer(tr, &count); + trace->reset(tr); + tracing_start(); + + if (!ret && !count) { + printk(KERN_CONT "no entries found.."); + ret = -1; + } + + return ret; +} +#endif /* CONFIG_HW_BRANCH_TRACER */ + +#ifdef CONFIG_KSYM_TRACER +static int ksym_selftest_dummy; + +int +trace_selftest_startup_ksym(struct tracer *trace, struct trace_array *tr) +{ + unsigned long count; + int ret; + + /* start the tracing */ + ret = tracer_init(trace, tr); + if (ret) { + warn_failed_init_tracer(trace, ret); + return ret; + } + + ksym_selftest_dummy = 0; + /* Register the read-write tracing request */ + + ret = process_new_ksym_entry("ksym_selftest_dummy", + HW_BREAKPOINT_R | HW_BREAKPOINT_W, + (unsigned long)(&ksym_selftest_dummy)); + + if (ret < 0) { + printk(KERN_CONT "ksym_trace read-write startup test failed\n"); + goto ret_path; + } + /* Perform a read and a write operation over the dummy variable to + * trigger the tracer + */ + if (ksym_selftest_dummy == 0) + ksym_selftest_dummy++; + + /* stop the tracing. */ + tracing_stop(); + /* check the trace buffer */ + ret = trace_test_buffer(tr, &count); + trace->reset(tr); + tracing_start(); + + /* read & write operations - one each is performed on the dummy variable + * triggering two entries in the trace buffer + */ + if (!ret && count != 2) { + printk(KERN_CONT "Ksym tracer startup test failed"); + ret = -1; + } + +ret_path: + return ret; +} +#endif /* CONFIG_KSYM_TRACER */ + diff --git a/kernel/trace/trace_stack.c b/kernel/trace/trace_stack.c index d0871bc0aca..8504ac71e4e 100644 --- a/kernel/trace/trace_stack.c +++ b/kernel/trace/trace_stack.c @@ -186,43 +186,33 @@ static const struct file_operations stack_max_size_fops = { }; static void * -t_next(struct seq_file *m, void *v, loff_t *pos) +__next(struct seq_file *m, loff_t *pos) { - long i; - - (*pos)++; + long n = *pos - 1; - if (v == SEQ_START_TOKEN) - i = 0; - else { - i = *(long *)v; - i++; - } - - if (i >= max_stack_trace.nr_entries || - stack_dump_trace[i] == ULONG_MAX) + if (n >= max_stack_trace.nr_entries || stack_dump_trace[n] == ULONG_MAX) return NULL; - m->private = (void *)i; - + m->private = (void *)n; return &m->private; } -static void *t_start(struct seq_file *m, loff_t *pos) +static void * +t_next(struct seq_file *m, void *v, loff_t *pos) { - void *t = SEQ_START_TOKEN; - loff_t l = 0; + (*pos)++; + return __next(m, pos); +} +static void *t_start(struct seq_file *m, loff_t *pos) +{ local_irq_disable(); __raw_spin_lock(&max_stack_lock); if (*pos == 0) return SEQ_START_TOKEN; - for (; t && l < *pos; t = t_next(m, t, &l)) - ; - - return t; + return __next(m, pos); } static void t_stop(struct seq_file *m, void *p) @@ -234,15 +224,19 @@ static void t_stop(struct seq_file *m, void *p) static int trace_lookup_stack(struct seq_file *m, long i) { unsigned long addr = stack_dump_trace[i]; -#ifdef CONFIG_KALLSYMS - char str[KSYM_SYMBOL_LEN]; - sprint_symbol(str, addr); + return seq_printf(m, "%pF\n", (void *)addr); +} - return seq_printf(m, "%s\n", str); -#else - return seq_printf(m, "%p\n", (void*)addr); -#endif +static void print_disabled(struct seq_file *m) +{ + seq_puts(m, "#\n" + "# Stack tracer disabled\n" + "#\n" + "# To enable the stack tracer, either add 'stacktrace' to the\n" + "# kernel command line\n" + "# or 'echo 1 > /proc/sys/kernel/stack_tracer_enabled'\n" + "#\n"); } static int t_show(struct seq_file *m, void *v) @@ -251,10 +245,14 @@ static int t_show(struct seq_file *m, void *v) int size; if (v == SEQ_START_TOKEN) { - seq_printf(m, " Depth Size Location" + seq_printf(m, " Depth Size Location" " (%d entries)\n" - " ----- ---- --------\n", - max_stack_trace.nr_entries); + " ----- ---- --------\n", + max_stack_trace.nr_entries - 1); + + if (!stack_tracer_enabled && !max_stack_size) + print_disabled(m); + return 0; } @@ -286,35 +284,32 @@ static const struct seq_operations stack_trace_seq_ops = { static int stack_trace_open(struct inode *inode, struct file *file) { - int ret; - - ret = seq_open(file, &stack_trace_seq_ops); - - return ret; + return seq_open(file, &stack_trace_seq_ops); } static const struct file_operations stack_trace_fops = { .open = stack_trace_open, .read = seq_read, .llseek = seq_lseek, + .release = seq_release, }; int stack_trace_sysctl(struct ctl_table *table, int write, - struct file *file, void __user *buffer, size_t *lenp, + void __user *buffer, size_t *lenp, loff_t *ppos) { int ret; mutex_lock(&stack_sysctl_mutex); - ret = proc_dointvec(table, write, file, buffer, lenp, ppos); + ret = proc_dointvec(table, write, buffer, lenp, ppos); if (ret || !write || - (last_stack_tracer_enabled == stack_tracer_enabled)) + (last_stack_tracer_enabled == !!stack_tracer_enabled)) goto out; - last_stack_tracer_enabled = stack_tracer_enabled; + last_stack_tracer_enabled = !!stack_tracer_enabled; if (stack_tracer_enabled) register_ftrace_function(&trace_ops); @@ -337,19 +332,14 @@ __setup("stacktrace", enable_stacktrace); static __init int stack_trace_init(void) { struct dentry *d_tracer; - struct dentry *entry; d_tracer = tracing_init_dentry(); - entry = debugfs_create_file("stack_max_size", 0644, d_tracer, - &max_stack_size, &stack_max_size_fops); - if (!entry) - pr_warning("Could not create debugfs 'stack_max_size' entry\n"); + trace_create_file("stack_max_size", 0644, d_tracer, + &max_stack_size, &stack_max_size_fops); - entry = debugfs_create_file("stack_trace", 0444, d_tracer, - NULL, &stack_trace_fops); - if (!entry) - pr_warning("Could not create debugfs 'stack_trace' entry\n"); + trace_create_file("stack_trace", 0444, d_tracer, + NULL, &stack_trace_fops); if (stack_tracer_enabled) register_ftrace_function(&trace_ops); diff --git a/kernel/trace/trace_stat.c b/kernel/trace/trace_stat.c new file mode 100644 index 00000000000..a4bb239eb98 --- /dev/null +++ b/kernel/trace/trace_stat.c @@ -0,0 +1,387 @@ +/* + * Infrastructure for statistic tracing (histogram output). + * + * Copyright (C) 2008-2009 Frederic Weisbecker <fweisbec@gmail.com> + * + * Based on the code from trace_branch.c which is + * Copyright (C) 2008 Steven Rostedt <srostedt@redhat.com> + * + */ + + +#include <linux/list.h> +#include <linux/rbtree.h> +#include <linux/debugfs.h> +#include "trace_stat.h" +#include "trace.h" + + +/* + * List of stat red-black nodes from a tracer + * We use a such tree to sort quickly the stat + * entries from the tracer. + */ +struct stat_node { + struct rb_node node; + void *stat; +}; + +/* A stat session is the stats output in one file */ +struct stat_session { + struct list_head session_list; + struct tracer_stat *ts; + struct rb_root stat_root; + struct mutex stat_mutex; + struct dentry *file; +}; + +/* All of the sessions currently in use. Each stat file embed one session */ +static LIST_HEAD(all_stat_sessions); +static DEFINE_MUTEX(all_stat_sessions_mutex); + +/* The root directory for all stat files */ +static struct dentry *stat_dir; + +/* + * Iterate through the rbtree using a post order traversal path + * to release the next node. + * It won't necessary release one at each iteration + * but it will at least advance closer to the next one + * to be released. + */ +static struct rb_node *release_next(struct tracer_stat *ts, + struct rb_node *node) +{ + struct stat_node *snode; + struct rb_node *parent = rb_parent(node); + + if (node->rb_left) + return node->rb_left; + else if (node->rb_right) + return node->rb_right; + else { + if (!parent) + ; + else if (parent->rb_left == node) + parent->rb_left = NULL; + else + parent->rb_right = NULL; + + snode = container_of(node, struct stat_node, node); + if (ts->stat_release) + ts->stat_release(snode->stat); + kfree(snode); + + return parent; + } +} + +static void __reset_stat_session(struct stat_session *session) +{ + struct rb_node *node = session->stat_root.rb_node; + + while (node) + node = release_next(session->ts, node); + + session->stat_root = RB_ROOT; +} + +static void reset_stat_session(struct stat_session *session) +{ + mutex_lock(&session->stat_mutex); + __reset_stat_session(session); + mutex_unlock(&session->stat_mutex); +} + +static void destroy_session(struct stat_session *session) +{ + debugfs_remove(session->file); + __reset_stat_session(session); + mutex_destroy(&session->stat_mutex); + kfree(session); +} + +typedef int (*cmp_stat_t)(void *, void *); + +static int insert_stat(struct rb_root *root, void *stat, cmp_stat_t cmp) +{ + struct rb_node **new = &(root->rb_node), *parent = NULL; + struct stat_node *data; + + data = kzalloc(sizeof(*data), GFP_KERNEL); + if (!data) + return -ENOMEM; + data->stat = stat; + + /* + * Figure out where to put new node + * This is a descendent sorting + */ + while (*new) { + struct stat_node *this; + int result; + + this = container_of(*new, struct stat_node, node); + result = cmp(data->stat, this->stat); + + parent = *new; + if (result >= 0) + new = &((*new)->rb_left); + else + new = &((*new)->rb_right); + } + + rb_link_node(&data->node, parent, new); + rb_insert_color(&data->node, root); + return 0; +} + +/* + * For tracers that don't provide a stat_cmp callback. + * This one will force an insertion as right-most node + * in the rbtree. + */ +static int dummy_cmp(void *p1, void *p2) +{ + return -1; +} + +/* + * Initialize the stat rbtree at each trace_stat file opening. + * All of these copies and sorting are required on all opening + * since the stats could have changed between two file sessions. + */ +static int stat_seq_init(struct stat_session *session) +{ + struct tracer_stat *ts = session->ts; + struct rb_root *root = &session->stat_root; + void *stat; + int ret = 0; + int i; + + mutex_lock(&session->stat_mutex); + __reset_stat_session(session); + + if (!ts->stat_cmp) + ts->stat_cmp = dummy_cmp; + + stat = ts->stat_start(ts); + if (!stat) + goto exit; + + ret = insert_stat(root, stat, ts->stat_cmp); + if (ret) + goto exit; + + /* + * Iterate over the tracer stat entries and store them in an rbtree. + */ + for (i = 1; ; i++) { + stat = ts->stat_next(stat, i); + + /* End of insertion */ + if (!stat) + break; + + ret = insert_stat(root, stat, ts->stat_cmp); + if (ret) + goto exit_free_rbtree; + } + +exit: + mutex_unlock(&session->stat_mutex); + return ret; + +exit_free_rbtree: + __reset_stat_session(session); + mutex_unlock(&session->stat_mutex); + return ret; +} + + +static void *stat_seq_start(struct seq_file *s, loff_t *pos) +{ + struct stat_session *session = s->private; + struct rb_node *node; + int n = *pos; + int i; + + /* Prevent from tracer switch or rbtree modification */ + mutex_lock(&session->stat_mutex); + + /* If we are in the beginning of the file, print the headers */ + if (session->ts->stat_headers) { + if (n == 0) + return SEQ_START_TOKEN; + n--; + } + + node = rb_first(&session->stat_root); + for (i = 0; node && i < n; i++) + node = rb_next(node); + + return node; +} + +static void *stat_seq_next(struct seq_file *s, void *p, loff_t *pos) +{ + struct stat_session *session = s->private; + struct rb_node *node = p; + + (*pos)++; + + if (p == SEQ_START_TOKEN) + return rb_first(&session->stat_root); + + return rb_next(node); +} + +static void stat_seq_stop(struct seq_file *s, void *p) +{ + struct stat_session *session = s->private; + mutex_unlock(&session->stat_mutex); +} + +static int stat_seq_show(struct seq_file *s, void *v) +{ + struct stat_session *session = s->private; + struct stat_node *l = container_of(v, struct stat_node, node); + + if (v == SEQ_START_TOKEN) + return session->ts->stat_headers(s); + + return session->ts->stat_show(s, l->stat); +} + +static const struct seq_operations trace_stat_seq_ops = { + .start = stat_seq_start, + .next = stat_seq_next, + .stop = stat_seq_stop, + .show = stat_seq_show +}; + +/* The session stat is refilled and resorted at each stat file opening */ +static int tracing_stat_open(struct inode *inode, struct file *file) +{ + int ret; + struct seq_file *m; + struct stat_session *session = inode->i_private; + + ret = stat_seq_init(session); + if (ret) + return ret; + + ret = seq_open(file, &trace_stat_seq_ops); + if (ret) { + reset_stat_session(session); + return ret; + } + + m = file->private_data; + m->private = session; + return ret; +} + +/* + * Avoid consuming memory with our now useless rbtree. + */ +static int tracing_stat_release(struct inode *i, struct file *f) +{ + struct stat_session *session = i->i_private; + + reset_stat_session(session); + + return seq_release(i, f); +} + +static const struct file_operations tracing_stat_fops = { + .open = tracing_stat_open, + .read = seq_read, + .llseek = seq_lseek, + .release = tracing_stat_release +}; + +static int tracing_stat_init(void) +{ + struct dentry *d_tracing; + + d_tracing = tracing_init_dentry(); + + stat_dir = debugfs_create_dir("trace_stat", d_tracing); + if (!stat_dir) + pr_warning("Could not create debugfs " + "'trace_stat' entry\n"); + return 0; +} + +static int init_stat_file(struct stat_session *session) +{ + if (!stat_dir && tracing_stat_init()) + return -ENODEV; + + session->file = debugfs_create_file(session->ts->name, 0644, + stat_dir, + session, &tracing_stat_fops); + if (!session->file) + return -ENOMEM; + return 0; +} + +int register_stat_tracer(struct tracer_stat *trace) +{ + struct stat_session *session, *node; + int ret; + + if (!trace) + return -EINVAL; + + if (!trace->stat_start || !trace->stat_next || !trace->stat_show) + return -EINVAL; + + /* Already registered? */ + mutex_lock(&all_stat_sessions_mutex); + list_for_each_entry(node, &all_stat_sessions, session_list) { + if (node->ts == trace) { + mutex_unlock(&all_stat_sessions_mutex); + return -EINVAL; + } + } + mutex_unlock(&all_stat_sessions_mutex); + + /* Init the session */ + session = kzalloc(sizeof(*session), GFP_KERNEL); + if (!session) + return -ENOMEM; + + session->ts = trace; + INIT_LIST_HEAD(&session->session_list); + mutex_init(&session->stat_mutex); + + ret = init_stat_file(session); + if (ret) { + destroy_session(session); + return ret; + } + + /* Register */ + mutex_lock(&all_stat_sessions_mutex); + list_add_tail(&session->session_list, &all_stat_sessions); + mutex_unlock(&all_stat_sessions_mutex); + + return 0; +} + +void unregister_stat_tracer(struct tracer_stat *trace) +{ + struct stat_session *node, *tmp; + + mutex_lock(&all_stat_sessions_mutex); + list_for_each_entry_safe(node, tmp, &all_stat_sessions, session_list) { + if (node->ts == trace) { + list_del(&node->session_list); + destroy_session(node); + break; + } + } + mutex_unlock(&all_stat_sessions_mutex); +} diff --git a/kernel/trace/trace_stat.h b/kernel/trace/trace_stat.h new file mode 100644 index 00000000000..8f03914b9a6 --- /dev/null +++ b/kernel/trace/trace_stat.h @@ -0,0 +1,33 @@ +#ifndef __TRACE_STAT_H +#define __TRACE_STAT_H + +#include <linux/seq_file.h> + +/* + * If you want to provide a stat file (one-shot statistics), fill + * an iterator with stat_start/stat_next and a stat_show callbacks. + * The others callbacks are optional. + */ +struct tracer_stat { + /* The name of your stat file */ + const char *name; + /* Iteration over statistic entries */ + void *(*stat_start)(struct tracer_stat *trace); + void *(*stat_next)(void *prev, int idx); + /* Compare two entries for stats sorting */ + int (*stat_cmp)(void *p1, void *p2); + /* Print a stat entry */ + int (*stat_show)(struct seq_file *s, void *p); + /* Release an entry */ + void (*stat_release)(void *stat); + /* Print the headers of your stat entries */ + int (*stat_headers)(struct seq_file *s); +}; + +/* + * Destroy or create a stat file + */ +extern int register_stat_tracer(struct tracer_stat *trace); +extern void unregister_stat_tracer(struct tracer_stat *trace); + +#endif /* __TRACE_STAT_H */ diff --git a/kernel/trace/trace_syscalls.c b/kernel/trace/trace_syscalls.c new file mode 100644 index 00000000000..57501d90096 --- /dev/null +++ b/kernel/trace/trace_syscalls.c @@ -0,0 +1,645 @@ +#include <trace/syscall.h> +#include <trace/events/syscalls.h> +#include <linux/kernel.h> +#include <linux/ftrace.h> +#include <linux/perf_event.h> +#include <asm/syscall.h> + +#include "trace_output.h" +#include "trace.h" + +static DEFINE_MUTEX(syscall_trace_lock); +static int sys_refcount_enter; +static int sys_refcount_exit; +static DECLARE_BITMAP(enabled_enter_syscalls, NR_syscalls); +static DECLARE_BITMAP(enabled_exit_syscalls, NR_syscalls); + +extern unsigned long __start_syscalls_metadata[]; +extern unsigned long __stop_syscalls_metadata[]; + +static struct syscall_metadata **syscalls_metadata; + +static struct syscall_metadata *find_syscall_meta(unsigned long syscall) +{ + struct syscall_metadata *start; + struct syscall_metadata *stop; + char str[KSYM_SYMBOL_LEN]; + + + start = (struct syscall_metadata *)__start_syscalls_metadata; + stop = (struct syscall_metadata *)__stop_syscalls_metadata; + kallsyms_lookup(syscall, NULL, NULL, NULL, str); + + for ( ; start < stop; start++) { + /* + * Only compare after the "sys" prefix. Archs that use + * syscall wrappers may have syscalls symbols aliases prefixed + * with "SyS" instead of "sys", leading to an unwanted + * mismatch. + */ + if (start->name && !strcmp(start->name + 3, str + 3)) + return start; + } + return NULL; +} + +static struct syscall_metadata *syscall_nr_to_meta(int nr) +{ + if (!syscalls_metadata || nr >= NR_syscalls || nr < 0) + return NULL; + + return syscalls_metadata[nr]; +} + +enum print_line_t +print_syscall_enter(struct trace_iterator *iter, int flags) +{ + struct trace_seq *s = &iter->seq; + struct trace_entry *ent = iter->ent; + struct syscall_trace_enter *trace; + struct syscall_metadata *entry; + int i, ret, syscall; + + trace = (typeof(trace))ent; + syscall = trace->nr; + entry = syscall_nr_to_meta(syscall); + + if (!entry) + goto end; + + if (entry->enter_event->id != ent->type) { + WARN_ON_ONCE(1); + goto end; + } + + ret = trace_seq_printf(s, "%s(", entry->name); + if (!ret) + return TRACE_TYPE_PARTIAL_LINE; + + for (i = 0; i < entry->nb_args; i++) { + /* parameter types */ + if (trace_flags & TRACE_ITER_VERBOSE) { + ret = trace_seq_printf(s, "%s ", entry->types[i]); + if (!ret) + return TRACE_TYPE_PARTIAL_LINE; + } + /* parameter values */ + ret = trace_seq_printf(s, "%s: %lx%s", entry->args[i], + trace->args[i], + i == entry->nb_args - 1 ? "" : ", "); + if (!ret) + return TRACE_TYPE_PARTIAL_LINE; + } + + ret = trace_seq_putc(s, ')'); + if (!ret) + return TRACE_TYPE_PARTIAL_LINE; + +end: + ret = trace_seq_putc(s, '\n'); + if (!ret) + return TRACE_TYPE_PARTIAL_LINE; + + return TRACE_TYPE_HANDLED; +} + +enum print_line_t +print_syscall_exit(struct trace_iterator *iter, int flags) +{ + struct trace_seq *s = &iter->seq; + struct trace_entry *ent = iter->ent; + struct syscall_trace_exit *trace; + int syscall; + struct syscall_metadata *entry; + int ret; + + trace = (typeof(trace))ent; + syscall = trace->nr; + entry = syscall_nr_to_meta(syscall); + + if (!entry) { + trace_seq_printf(s, "\n"); + return TRACE_TYPE_HANDLED; + } + + if (entry->exit_event->id != ent->type) { + WARN_ON_ONCE(1); + return TRACE_TYPE_UNHANDLED; + } + + ret = trace_seq_printf(s, "%s -> 0x%lx\n", entry->name, + trace->ret); + if (!ret) + return TRACE_TYPE_PARTIAL_LINE; + + return TRACE_TYPE_HANDLED; +} + +extern char *__bad_type_size(void); + +#define SYSCALL_FIELD(type, name) \ + sizeof(type) != sizeof(trace.name) ? \ + __bad_type_size() : \ + #type, #name, offsetof(typeof(trace), name), \ + sizeof(trace.name), is_signed_type(type) + +int syscall_enter_format(struct ftrace_event_call *call, struct trace_seq *s) +{ + int i; + int ret; + struct syscall_metadata *entry = call->data; + struct syscall_trace_enter trace; + int offset = offsetof(struct syscall_trace_enter, args); + + ret = trace_seq_printf(s, "\tfield:%s %s;\toffset:%zu;\tsize:%zu;" + "\tsigned:%u;\n", + SYSCALL_FIELD(int, nr)); + if (!ret) + return 0; + + for (i = 0; i < entry->nb_args; i++) { + ret = trace_seq_printf(s, "\tfield:%s %s;", entry->types[i], + entry->args[i]); + if (!ret) + return 0; + ret = trace_seq_printf(s, "\toffset:%d;\tsize:%zu;" + "\tsigned:%u;\n", offset, + sizeof(unsigned long), + is_signed_type(unsigned long)); + if (!ret) + return 0; + offset += sizeof(unsigned long); + } + + trace_seq_puts(s, "\nprint fmt: \""); + for (i = 0; i < entry->nb_args; i++) { + ret = trace_seq_printf(s, "%s: 0x%%0%zulx%s", entry->args[i], + sizeof(unsigned long), + i == entry->nb_args - 1 ? "" : ", "); + if (!ret) + return 0; + } + trace_seq_putc(s, '"'); + + for (i = 0; i < entry->nb_args; i++) { + ret = trace_seq_printf(s, ", ((unsigned long)(REC->%s))", + entry->args[i]); + if (!ret) + return 0; + } + + return trace_seq_putc(s, '\n'); +} + +int syscall_exit_format(struct ftrace_event_call *call, struct trace_seq *s) +{ + int ret; + struct syscall_trace_exit trace; + + ret = trace_seq_printf(s, + "\tfield:%s %s;\toffset:%zu;\tsize:%zu;" + "\tsigned:%u;\n" + "\tfield:%s %s;\toffset:%zu;\tsize:%zu;" + "\tsigned:%u;\n", + SYSCALL_FIELD(int, nr), + SYSCALL_FIELD(long, ret)); + if (!ret) + return 0; + + return trace_seq_printf(s, "\nprint fmt: \"0x%%lx\", REC->ret\n"); +} + +int syscall_enter_define_fields(struct ftrace_event_call *call) +{ + struct syscall_trace_enter trace; + struct syscall_metadata *meta = call->data; + int ret; + int i; + int offset = offsetof(typeof(trace), args); + + ret = trace_define_common_fields(call); + if (ret) + return ret; + + ret = trace_define_field(call, SYSCALL_FIELD(int, nr), FILTER_OTHER); + if (ret) + return ret; + + for (i = 0; i < meta->nb_args; i++) { + ret = trace_define_field(call, meta->types[i], + meta->args[i], offset, + sizeof(unsigned long), 0, + FILTER_OTHER); + offset += sizeof(unsigned long); + } + + return ret; +} + +int syscall_exit_define_fields(struct ftrace_event_call *call) +{ + struct syscall_trace_exit trace; + int ret; + + ret = trace_define_common_fields(call); + if (ret) + return ret; + + ret = trace_define_field(call, SYSCALL_FIELD(int, nr), FILTER_OTHER); + if (ret) + return ret; + + ret = trace_define_field(call, SYSCALL_FIELD(long, ret), + FILTER_OTHER); + + return ret; +} + +void ftrace_syscall_enter(struct pt_regs *regs, long id) +{ + struct syscall_trace_enter *entry; + struct syscall_metadata *sys_data; + struct ring_buffer_event *event; + struct ring_buffer *buffer; + int size; + int syscall_nr; + + syscall_nr = syscall_get_nr(current, regs); + if (syscall_nr < 0) + return; + if (!test_bit(syscall_nr, enabled_enter_syscalls)) + return; + + sys_data = syscall_nr_to_meta(syscall_nr); + if (!sys_data) + return; + + size = sizeof(*entry) + sizeof(unsigned long) * sys_data->nb_args; + + event = trace_current_buffer_lock_reserve(&buffer, + sys_data->enter_event->id, size, 0, 0); + if (!event) + return; + + entry = ring_buffer_event_data(event); + entry->nr = syscall_nr; + syscall_get_arguments(current, regs, 0, sys_data->nb_args, entry->args); + + if (!filter_current_check_discard(buffer, sys_data->enter_event, + entry, event)) + trace_current_buffer_unlock_commit(buffer, event, 0, 0); +} + +void ftrace_syscall_exit(struct pt_regs *regs, long ret) +{ + struct syscall_trace_exit *entry; + struct syscall_metadata *sys_data; + struct ring_buffer_event *event; + struct ring_buffer *buffer; + int syscall_nr; + + syscall_nr = syscall_get_nr(current, regs); + if (syscall_nr < 0) + return; + if (!test_bit(syscall_nr, enabled_exit_syscalls)) + return; + + sys_data = syscall_nr_to_meta(syscall_nr); + if (!sys_data) + return; + + event = trace_current_buffer_lock_reserve(&buffer, + sys_data->exit_event->id, sizeof(*entry), 0, 0); + if (!event) + return; + + entry = ring_buffer_event_data(event); + entry->nr = syscall_nr; + entry->ret = syscall_get_return_value(current, regs); + + if (!filter_current_check_discard(buffer, sys_data->exit_event, + entry, event)) + trace_current_buffer_unlock_commit(buffer, event, 0, 0); +} + +int reg_event_syscall_enter(struct ftrace_event_call *call) +{ + int ret = 0; + int num; + + num = ((struct syscall_metadata *)call->data)->syscall_nr; + if (num < 0 || num >= NR_syscalls) + return -ENOSYS; + mutex_lock(&syscall_trace_lock); + if (!sys_refcount_enter) + ret = register_trace_sys_enter(ftrace_syscall_enter); + if (ret) { + pr_info("event trace: Could not activate" + "syscall entry trace point"); + } else { + set_bit(num, enabled_enter_syscalls); + sys_refcount_enter++; + } + mutex_unlock(&syscall_trace_lock); + return ret; +} + +void unreg_event_syscall_enter(struct ftrace_event_call *call) +{ + int num; + + num = ((struct syscall_metadata *)call->data)->syscall_nr; + if (num < 0 || num >= NR_syscalls) + return; + mutex_lock(&syscall_trace_lock); + sys_refcount_enter--; + clear_bit(num, enabled_enter_syscalls); + if (!sys_refcount_enter) + unregister_trace_sys_enter(ftrace_syscall_enter); + mutex_unlock(&syscall_trace_lock); +} + +int reg_event_syscall_exit(struct ftrace_event_call *call) +{ + int ret = 0; + int num; + + num = ((struct syscall_metadata *)call->data)->syscall_nr; + if (num < 0 || num >= NR_syscalls) + return -ENOSYS; + mutex_lock(&syscall_trace_lock); + if (!sys_refcount_exit) + ret = register_trace_sys_exit(ftrace_syscall_exit); + if (ret) { + pr_info("event trace: Could not activate" + "syscall exit trace point"); + } else { + set_bit(num, enabled_exit_syscalls); + sys_refcount_exit++; + } + mutex_unlock(&syscall_trace_lock); + return ret; +} + +void unreg_event_syscall_exit(struct ftrace_event_call *call) +{ + int num; + + num = ((struct syscall_metadata *)call->data)->syscall_nr; + if (num < 0 || num >= NR_syscalls) + return; + mutex_lock(&syscall_trace_lock); + sys_refcount_exit--; + clear_bit(num, enabled_exit_syscalls); + if (!sys_refcount_exit) + unregister_trace_sys_exit(ftrace_syscall_exit); + mutex_unlock(&syscall_trace_lock); +} + +int init_syscall_trace(struct ftrace_event_call *call) +{ + int id; + + id = register_ftrace_event(call->event); + if (!id) + return -ENODEV; + call->id = id; + INIT_LIST_HEAD(&call->fields); + return 0; +} + +int __init init_ftrace_syscalls(void) +{ + struct syscall_metadata *meta; + unsigned long addr; + int i; + + syscalls_metadata = kzalloc(sizeof(*syscalls_metadata) * + NR_syscalls, GFP_KERNEL); + if (!syscalls_metadata) { + WARN_ON(1); + return -ENOMEM; + } + + for (i = 0; i < NR_syscalls; i++) { + addr = arch_syscall_addr(i); + meta = find_syscall_meta(addr); + if (!meta) + continue; + + meta->syscall_nr = i; + syscalls_metadata[i] = meta; + } + + return 0; +} +core_initcall(init_ftrace_syscalls); + +#ifdef CONFIG_EVENT_PROFILE + +static DECLARE_BITMAP(enabled_prof_enter_syscalls, NR_syscalls); +static DECLARE_BITMAP(enabled_prof_exit_syscalls, NR_syscalls); +static int sys_prof_refcount_enter; +static int sys_prof_refcount_exit; + +static void prof_syscall_enter(struct pt_regs *regs, long id) +{ + struct syscall_metadata *sys_data; + struct syscall_trace_enter *rec; + unsigned long flags; + char *trace_buf; + char *raw_data; + int syscall_nr; + int rctx; + int size; + int cpu; + + syscall_nr = syscall_get_nr(current, regs); + if (!test_bit(syscall_nr, enabled_prof_enter_syscalls)) + return; + + sys_data = syscall_nr_to_meta(syscall_nr); + if (!sys_data) + return; + + /* get the size after alignment with the u32 buffer size field */ + size = sizeof(unsigned long) * sys_data->nb_args + sizeof(*rec); + size = ALIGN(size + sizeof(u32), sizeof(u64)); + size -= sizeof(u32); + + if (WARN_ONCE(size > FTRACE_MAX_PROFILE_SIZE, + "profile buffer not large enough")) + return; + + /* Protect the per cpu buffer, begin the rcu read side */ + local_irq_save(flags); + + rctx = perf_swevent_get_recursion_context(); + if (rctx < 0) + goto end_recursion; + + cpu = smp_processor_id(); + + trace_buf = rcu_dereference(perf_trace_buf); + + if (!trace_buf) + goto end; + + raw_data = per_cpu_ptr(trace_buf, cpu); + + /* zero the dead bytes from align to not leak stack to user */ + *(u64 *)(&raw_data[size - sizeof(u64)]) = 0ULL; + + rec = (struct syscall_trace_enter *) raw_data; + tracing_generic_entry_update(&rec->ent, 0, 0); + rec->ent.type = sys_data->enter_event->id; + rec->nr = syscall_nr; + syscall_get_arguments(current, regs, 0, sys_data->nb_args, + (unsigned long *)&rec->args); + perf_tp_event(sys_data->enter_event->id, 0, 1, rec, size); + +end: + perf_swevent_put_recursion_context(rctx); +end_recursion: + local_irq_restore(flags); +} + +int prof_sysenter_enable(struct ftrace_event_call *call) +{ + int ret = 0; + int num; + + num = ((struct syscall_metadata *)call->data)->syscall_nr; + + mutex_lock(&syscall_trace_lock); + if (!sys_prof_refcount_enter) + ret = register_trace_sys_enter(prof_syscall_enter); + if (ret) { + pr_info("event trace: Could not activate" + "syscall entry trace point"); + } else { + set_bit(num, enabled_prof_enter_syscalls); + sys_prof_refcount_enter++; + } + mutex_unlock(&syscall_trace_lock); + return ret; +} + +void prof_sysenter_disable(struct ftrace_event_call *call) +{ + int num; + + num = ((struct syscall_metadata *)call->data)->syscall_nr; + + mutex_lock(&syscall_trace_lock); + sys_prof_refcount_enter--; + clear_bit(num, enabled_prof_enter_syscalls); + if (!sys_prof_refcount_enter) + unregister_trace_sys_enter(prof_syscall_enter); + mutex_unlock(&syscall_trace_lock); +} + +static void prof_syscall_exit(struct pt_regs *regs, long ret) +{ + struct syscall_metadata *sys_data; + struct syscall_trace_exit *rec; + unsigned long flags; + int syscall_nr; + char *trace_buf; + char *raw_data; + int rctx; + int size; + int cpu; + + syscall_nr = syscall_get_nr(current, regs); + if (!test_bit(syscall_nr, enabled_prof_exit_syscalls)) + return; + + sys_data = syscall_nr_to_meta(syscall_nr); + if (!sys_data) + return; + + /* We can probably do that at build time */ + size = ALIGN(sizeof(*rec) + sizeof(u32), sizeof(u64)); + size -= sizeof(u32); + + /* + * Impossible, but be paranoid with the future + * How to put this check outside runtime? + */ + if (WARN_ONCE(size > FTRACE_MAX_PROFILE_SIZE, + "exit event has grown above profile buffer size")) + return; + + /* Protect the per cpu buffer, begin the rcu read side */ + local_irq_save(flags); + + rctx = perf_swevent_get_recursion_context(); + if (rctx < 0) + goto end_recursion; + + cpu = smp_processor_id(); + + trace_buf = rcu_dereference(perf_trace_buf); + + if (!trace_buf) + goto end; + + raw_data = per_cpu_ptr(trace_buf, cpu); + + /* zero the dead bytes from align to not leak stack to user */ + *(u64 *)(&raw_data[size - sizeof(u64)]) = 0ULL; + + rec = (struct syscall_trace_exit *)raw_data; + + tracing_generic_entry_update(&rec->ent, 0, 0); + rec->ent.type = sys_data->exit_event->id; + rec->nr = syscall_nr; + rec->ret = syscall_get_return_value(current, regs); + + perf_tp_event(sys_data->exit_event->id, 0, 1, rec, size); + +end: + perf_swevent_put_recursion_context(rctx); +end_recursion: + local_irq_restore(flags); +} + +int prof_sysexit_enable(struct ftrace_event_call *call) +{ + int ret = 0; + int num; + + num = ((struct syscall_metadata *)call->data)->syscall_nr; + + mutex_lock(&syscall_trace_lock); + if (!sys_prof_refcount_exit) + ret = register_trace_sys_exit(prof_syscall_exit); + if (ret) { + pr_info("event trace: Could not activate" + "syscall entry trace point"); + } else { + set_bit(num, enabled_prof_exit_syscalls); + sys_prof_refcount_exit++; + } + mutex_unlock(&syscall_trace_lock); + return ret; +} + +void prof_sysexit_disable(struct ftrace_event_call *call) +{ + int num; + + num = ((struct syscall_metadata *)call->data)->syscall_nr; + + mutex_lock(&syscall_trace_lock); + sys_prof_refcount_exit--; + clear_bit(num, enabled_prof_exit_syscalls); + if (!sys_prof_refcount_exit) + unregister_trace_sys_exit(prof_syscall_exit); + mutex_unlock(&syscall_trace_lock); +} + +#endif + + diff --git a/kernel/trace/trace_sysprof.c b/kernel/trace/trace_sysprof.c index eaca5ad803f..f6693969287 100644 --- a/kernel/trace/trace_sysprof.c +++ b/kernel/trace/trace_sysprof.c @@ -88,7 +88,7 @@ static void backtrace_address(void *data, unsigned long addr, int reliable) } } -const static struct stacktrace_ops backtrace_ops = { +static const struct stacktrace_ops backtrace_ops = { .warning = backtrace_warning, .warning_symbol = backtrace_warning_symbol, .stack = backtrace_stack, @@ -203,7 +203,8 @@ static void start_stack_timer(void *unused) hrtimer_init(hrtimer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); hrtimer->function = stack_trace_timer_fn; - hrtimer_start(hrtimer, ns_to_ktime(sample_period), HRTIMER_MODE_REL); + hrtimer_start(hrtimer, ns_to_ktime(sample_period), + HRTIMER_MODE_REL_PINNED); } static void start_stack_timers(void) @@ -226,15 +227,6 @@ static void stop_stack_timers(void) stop_stack_timer(cpu); } -static void start_stack_trace(struct trace_array *tr) -{ - mutex_lock(&sample_timer_lock); - tracing_reset_online_cpus(tr); - start_stack_timers(); - tracer_enabled = 1; - mutex_unlock(&sample_timer_lock); -} - static void stop_stack_trace(struct trace_array *tr) { mutex_lock(&sample_timer_lock); @@ -247,12 +239,18 @@ static int stack_trace_init(struct trace_array *tr) { sysprof_trace = tr; - start_stack_trace(tr); + tracing_start_cmdline_record(); + + mutex_lock(&sample_timer_lock); + start_stack_timers(); + tracer_enabled = 1; + mutex_unlock(&sample_timer_lock); return 0; } static void stack_trace_reset(struct trace_array *tr) { + tracing_stop_cmdline_record(); stop_stack_trace(tr); } @@ -317,18 +315,14 @@ sysprof_sample_write(struct file *filp, const char __user *ubuf, return cnt; } -static struct file_operations sysprof_sample_fops = { +static const struct file_operations sysprof_sample_fops = { .read = sysprof_sample_read, .write = sysprof_sample_write, }; void init_tracer_sysprof_debugfs(struct dentry *d_tracer) { - struct dentry *entry; - entry = debugfs_create_file("sysprof_sample_period", 0644, + trace_create_file("sysprof_sample_period", 0644, d_tracer, NULL, &sysprof_sample_fops); - if (entry) - return; - pr_warning("Could not create debugfs 'dyn_ftrace_total_info' entry\n"); } diff --git a/kernel/trace/trace_workqueue.c b/kernel/trace/trace_workqueue.c new file mode 100644 index 00000000000..40cafb07dff --- /dev/null +++ b/kernel/trace/trace_workqueue.c @@ -0,0 +1,295 @@ +/* + * Workqueue statistical tracer. + * + * Copyright (C) 2008 Frederic Weisbecker <fweisbec@gmail.com> + * + */ + + +#include <trace/events/workqueue.h> +#include <linux/list.h> +#include <linux/percpu.h> +#include <linux/kref.h> +#include "trace_stat.h" +#include "trace.h" + + +/* A cpu workqueue thread */ +struct cpu_workqueue_stats { + struct list_head list; + struct kref kref; + int cpu; + pid_t pid; +/* Can be inserted from interrupt or user context, need to be atomic */ + atomic_t inserted; +/* + * Don't need to be atomic, works are serialized in a single workqueue thread + * on a single CPU. + */ + unsigned int executed; +}; + +/* List of workqueue threads on one cpu */ +struct workqueue_global_stats { + struct list_head list; + spinlock_t lock; +}; + +/* Don't need a global lock because allocated before the workqueues, and + * never freed. + */ +static DEFINE_PER_CPU(struct workqueue_global_stats, all_workqueue_stat); +#define workqueue_cpu_stat(cpu) (&per_cpu(all_workqueue_stat, cpu)) + +static void cpu_workqueue_stat_free(struct kref *kref) +{ + kfree(container_of(kref, struct cpu_workqueue_stats, kref)); +} + +/* Insertion of a work */ +static void +probe_workqueue_insertion(struct task_struct *wq_thread, + struct work_struct *work) +{ + int cpu = cpumask_first(&wq_thread->cpus_allowed); + struct cpu_workqueue_stats *node; + unsigned long flags; + + spin_lock_irqsave(&workqueue_cpu_stat(cpu)->lock, flags); + list_for_each_entry(node, &workqueue_cpu_stat(cpu)->list, list) { + if (node->pid == wq_thread->pid) { + atomic_inc(&node->inserted); + goto found; + } + } + pr_debug("trace_workqueue: entry not found\n"); +found: + spin_unlock_irqrestore(&workqueue_cpu_stat(cpu)->lock, flags); +} + +/* Execution of a work */ +static void +probe_workqueue_execution(struct task_struct *wq_thread, + struct work_struct *work) +{ + int cpu = cpumask_first(&wq_thread->cpus_allowed); + struct cpu_workqueue_stats *node; + unsigned long flags; + + spin_lock_irqsave(&workqueue_cpu_stat(cpu)->lock, flags); + list_for_each_entry(node, &workqueue_cpu_stat(cpu)->list, list) { + if (node->pid == wq_thread->pid) { + node->executed++; + goto found; + } + } + pr_debug("trace_workqueue: entry not found\n"); +found: + spin_unlock_irqrestore(&workqueue_cpu_stat(cpu)->lock, flags); +} + +/* Creation of a cpu workqueue thread */ +static void probe_workqueue_creation(struct task_struct *wq_thread, int cpu) +{ + struct cpu_workqueue_stats *cws; + unsigned long flags; + + WARN_ON(cpu < 0); + + /* Workqueues are sometimes created in atomic context */ + cws = kzalloc(sizeof(struct cpu_workqueue_stats), GFP_ATOMIC); + if (!cws) { + pr_warning("trace_workqueue: not enough memory\n"); + return; + } + INIT_LIST_HEAD(&cws->list); + kref_init(&cws->kref); + cws->cpu = cpu; + cws->pid = wq_thread->pid; + + spin_lock_irqsave(&workqueue_cpu_stat(cpu)->lock, flags); + list_add_tail(&cws->list, &workqueue_cpu_stat(cpu)->list); + spin_unlock_irqrestore(&workqueue_cpu_stat(cpu)->lock, flags); +} + +/* Destruction of a cpu workqueue thread */ +static void probe_workqueue_destruction(struct task_struct *wq_thread) +{ + /* Workqueue only execute on one cpu */ + int cpu = cpumask_first(&wq_thread->cpus_allowed); + struct cpu_workqueue_stats *node, *next; + unsigned long flags; + + spin_lock_irqsave(&workqueue_cpu_stat(cpu)->lock, flags); + list_for_each_entry_safe(node, next, &workqueue_cpu_stat(cpu)->list, + list) { + if (node->pid == wq_thread->pid) { + list_del(&node->list); + kref_put(&node->kref, cpu_workqueue_stat_free); + goto found; + } + } + + pr_debug("trace_workqueue: don't find workqueue to destroy\n"); +found: + spin_unlock_irqrestore(&workqueue_cpu_stat(cpu)->lock, flags); + +} + +static struct cpu_workqueue_stats *workqueue_stat_start_cpu(int cpu) +{ + unsigned long flags; + struct cpu_workqueue_stats *ret = NULL; + + + spin_lock_irqsave(&workqueue_cpu_stat(cpu)->lock, flags); + + if (!list_empty(&workqueue_cpu_stat(cpu)->list)) { + ret = list_entry(workqueue_cpu_stat(cpu)->list.next, + struct cpu_workqueue_stats, list); + kref_get(&ret->kref); + } + + spin_unlock_irqrestore(&workqueue_cpu_stat(cpu)->lock, flags); + + return ret; +} + +static void *workqueue_stat_start(struct tracer_stat *trace) +{ + int cpu; + void *ret = NULL; + + for_each_possible_cpu(cpu) { + ret = workqueue_stat_start_cpu(cpu); + if (ret) + return ret; + } + return NULL; +} + +static void *workqueue_stat_next(void *prev, int idx) +{ + struct cpu_workqueue_stats *prev_cws = prev; + struct cpu_workqueue_stats *ret; + int cpu = prev_cws->cpu; + unsigned long flags; + + spin_lock_irqsave(&workqueue_cpu_stat(cpu)->lock, flags); + if (list_is_last(&prev_cws->list, &workqueue_cpu_stat(cpu)->list)) { + spin_unlock_irqrestore(&workqueue_cpu_stat(cpu)->lock, flags); + do { + cpu = cpumask_next(cpu, cpu_possible_mask); + if (cpu >= nr_cpu_ids) + return NULL; + } while (!(ret = workqueue_stat_start_cpu(cpu))); + return ret; + } else { + ret = list_entry(prev_cws->list.next, + struct cpu_workqueue_stats, list); + kref_get(&ret->kref); + } + spin_unlock_irqrestore(&workqueue_cpu_stat(cpu)->lock, flags); + + return ret; +} + +static int workqueue_stat_show(struct seq_file *s, void *p) +{ + struct cpu_workqueue_stats *cws = p; + struct pid *pid; + struct task_struct *tsk; + + pid = find_get_pid(cws->pid); + if (pid) { + tsk = get_pid_task(pid, PIDTYPE_PID); + if (tsk) { + seq_printf(s, "%3d %6d %6u %s\n", cws->cpu, + atomic_read(&cws->inserted), cws->executed, + tsk->comm); + put_task_struct(tsk); + } + put_pid(pid); + } + + return 0; +} + +static void workqueue_stat_release(void *stat) +{ + struct cpu_workqueue_stats *node = stat; + + kref_put(&node->kref, cpu_workqueue_stat_free); +} + +static int workqueue_stat_headers(struct seq_file *s) +{ + seq_printf(s, "# CPU INSERTED EXECUTED NAME\n"); + seq_printf(s, "# | | | |\n"); + return 0; +} + +struct tracer_stat workqueue_stats __read_mostly = { + .name = "workqueues", + .stat_start = workqueue_stat_start, + .stat_next = workqueue_stat_next, + .stat_show = workqueue_stat_show, + .stat_release = workqueue_stat_release, + .stat_headers = workqueue_stat_headers +}; + + +int __init stat_workqueue_init(void) +{ + if (register_stat_tracer(&workqueue_stats)) { + pr_warning("Unable to register workqueue stat tracer\n"); + return 1; + } + + return 0; +} +fs_initcall(stat_workqueue_init); + +/* + * Workqueues are created very early, just after pre-smp initcalls. + * So we must register our tracepoints at this stage. + */ +int __init trace_workqueue_early_init(void) +{ + int ret, cpu; + + ret = register_trace_workqueue_insertion(probe_workqueue_insertion); + if (ret) + goto out; + + ret = register_trace_workqueue_execution(probe_workqueue_execution); + if (ret) + goto no_insertion; + + ret = register_trace_workqueue_creation(probe_workqueue_creation); + if (ret) + goto no_execution; + + ret = register_trace_workqueue_destruction(probe_workqueue_destruction); + if (ret) + goto no_creation; + + for_each_possible_cpu(cpu) { + spin_lock_init(&workqueue_cpu_stat(cpu)->lock); + INIT_LIST_HEAD(&workqueue_cpu_stat(cpu)->list); + } + + return 0; + +no_creation: + unregister_trace_workqueue_creation(probe_workqueue_creation); +no_execution: + unregister_trace_workqueue_execution(probe_workqueue_execution); +no_insertion: + unregister_trace_workqueue_insertion(probe_workqueue_insertion); +out: + pr_warning("trace_workqueue: unable to trace workqueues\n"); + + return 1; +} +early_initcall(trace_workqueue_early_init); diff --git a/kernel/tracepoint.c b/kernel/tracepoint.c index 79602740bbb..cc89be5bc0f 100644 --- a/kernel/tracepoint.c +++ b/kernel/tracepoint.c @@ -24,6 +24,7 @@ #include <linux/tracepoint.h> #include <linux/err.h> #include <linux/slab.h> +#include <linux/sched.h> extern struct tracepoint __start___tracepoints[]; extern struct tracepoint __stop___tracepoints[]; @@ -47,7 +48,7 @@ static struct hlist_head tracepoint_table[TRACEPOINT_TABLE_SIZE]; /* * Note about RCU : - * It is used to to delay the free of multiple probes array until a quiescent + * It is used to delay the free of multiple probes array until a quiescent * state is reached. * Tracepoint entries modifications are protected by the tracepoints_mutex. */ @@ -242,6 +243,11 @@ static void set_tracepoint(struct tracepoint_entry **entry, { WARN_ON(strcmp((*entry)->name, elem->name) != 0); + if (elem->regfunc && !elem->state && active) + elem->regfunc(); + else if (elem->unregfunc && elem->state && !active) + elem->unregfunc(); + /* * rcu_assign_pointer has a smp_wmb() which makes sure that the new * probe callbacks array is consistent before setting a pointer to it. @@ -261,6 +267,9 @@ static void set_tracepoint(struct tracepoint_entry **entry, */ static void disable_tracepoint(struct tracepoint *elem) { + if (elem->unregfunc && elem->state) + elem->unregfunc(); + elem->state = 0; rcu_assign_pointer(elem->funcs, NULL); } @@ -272,12 +281,15 @@ static void disable_tracepoint(struct tracepoint *elem) * * Updates the probe callback corresponding to a range of tracepoints. */ -void tracepoint_update_probe_range(struct tracepoint *begin, - struct tracepoint *end) +void +tracepoint_update_probe_range(struct tracepoint *begin, struct tracepoint *end) { struct tracepoint *iter; struct tracepoint_entry *mark_entry; + if (!begin) + return; + mutex_lock(&tracepoints_mutex); for (iter = begin; iter < end; iter++) { mark_entry = get_tracepoint(iter->name); @@ -551,9 +563,6 @@ int tracepoint_module_notify(struct notifier_block *self, switch (val) { case MODULE_STATE_COMING: - tracepoint_update_probe_range(mod->tracepoints, - mod->tracepoints + mod->num_tracepoints); - break; case MODULE_STATE_GOING: tracepoint_update_probe_range(mod->tracepoints, mod->tracepoints + mod->num_tracepoints); @@ -574,3 +583,41 @@ static int init_tracepoints(void) __initcall(init_tracepoints); #endif /* CONFIG_MODULES */ + +#ifdef CONFIG_HAVE_SYSCALL_TRACEPOINTS + +/* NB: reg/unreg are called while guarded with the tracepoints_mutex */ +static int sys_tracepoint_refcount; + +void syscall_regfunc(void) +{ + unsigned long flags; + struct task_struct *g, *t; + + if (!sys_tracepoint_refcount) { + read_lock_irqsave(&tasklist_lock, flags); + do_each_thread(g, t) { + /* Skip kernel threads. */ + if (t->mm) + set_tsk_thread_flag(t, TIF_SYSCALL_TRACEPOINT); + } while_each_thread(g, t); + read_unlock_irqrestore(&tasklist_lock, flags); + } + sys_tracepoint_refcount++; +} + +void syscall_unregfunc(void) +{ + unsigned long flags; + struct task_struct *g, *t; + + sys_tracepoint_refcount--; + if (!sys_tracepoint_refcount) { + read_lock_irqsave(&tasklist_lock, flags); + do_each_thread(g, t) { + clear_tsk_thread_flag(t, TIF_SYSCALL_TRACEPOINT); + } while_each_thread(g, t); + read_unlock_irqrestore(&tasklist_lock, flags); + } +} +#endif diff --git a/kernel/uid16.c b/kernel/uid16.c index 0314501688b..419209893d8 100644 --- a/kernel/uid16.c +++ b/kernel/uid16.c @@ -4,7 +4,6 @@ */ #include <linux/mm.h> -#include <linux/utsname.h> #include <linux/mman.h> #include <linux/notifier.h> #include <linux/reboot.h> diff --git a/kernel/user-return-notifier.c b/kernel/user-return-notifier.c new file mode 100644 index 00000000000..eb27fd3430a --- /dev/null +++ b/kernel/user-return-notifier.c @@ -0,0 +1,44 @@ + +#include <linux/user-return-notifier.h> +#include <linux/percpu.h> +#include <linux/sched.h> +#include <linux/module.h> + +static DEFINE_PER_CPU(struct hlist_head, return_notifier_list); + +/* + * Request a notification when the current cpu returns to userspace. Must be + * called in atomic context. The notifier will also be called in atomic + * context. + */ +void user_return_notifier_register(struct user_return_notifier *urn) +{ + set_tsk_thread_flag(current, TIF_USER_RETURN_NOTIFY); + hlist_add_head(&urn->link, &__get_cpu_var(return_notifier_list)); +} +EXPORT_SYMBOL_GPL(user_return_notifier_register); + +/* + * Removes a registered user return notifier. Must be called from atomic + * context, and from the same cpu registration occured in. + */ +void user_return_notifier_unregister(struct user_return_notifier *urn) +{ + hlist_del(&urn->link); + if (hlist_empty(&__get_cpu_var(return_notifier_list))) + clear_tsk_thread_flag(current, TIF_USER_RETURN_NOTIFY); +} +EXPORT_SYMBOL_GPL(user_return_notifier_unregister); + +/* Calls registered user return notifiers */ +void fire_user_return_notifiers(void) +{ + struct user_return_notifier *urn; + struct hlist_node *tmp1, *tmp2; + struct hlist_head *head; + + head = &get_cpu_var(return_notifier_list); + hlist_for_each_entry_safe(urn, tmp1, tmp2, head, link) + urn->on_user_return(urn); + put_cpu_var(return_notifier_list); +} diff --git a/kernel/user.c b/kernel/user.c index 850e0ba41c1..46d0165ca70 100644 --- a/kernel/user.c +++ b/kernel/user.c @@ -75,21 +75,6 @@ static void uid_hash_remove(struct user_struct *up) put_user_ns(up->user_ns); } -static struct user_struct *uid_hash_find(uid_t uid, struct hlist_head *hashent) -{ - struct user_struct *user; - struct hlist_node *h; - - hlist_for_each_entry(user, h, hashent, uidhash_node) { - if (user->uid == uid) { - atomic_inc(&user->__count); - return user; - } - } - - return NULL; -} - #ifdef CONFIG_USER_SCHED static void sched_destroy_user(struct user_struct *up) @@ -119,6 +104,23 @@ static int sched_create_user(struct user_struct *up) { return 0; } #if defined(CONFIG_USER_SCHED) && defined(CONFIG_SYSFS) +static struct user_struct *uid_hash_find(uid_t uid, struct hlist_head *hashent) +{ + struct user_struct *user; + struct hlist_node *h; + + hlist_for_each_entry(user, h, hashent, uidhash_node) { + if (user->uid == uid) { + /* possibly resurrect an "almost deleted" object */ + if (atomic_inc_return(&user->__count) == 1) + cancel_delayed_work(&user->work); + return user; + } + } + + return NULL; +} + static struct kset *uids_kset; /* represents the /sys/kernel/uids/ directory */ static DEFINE_MUTEX(uids_mutex); @@ -283,12 +285,12 @@ int __init uids_sysfs_init(void) return uids_user_create(&root_user); } -/* work function to remove sysfs directory for a user and free up +/* delayed work function to remove sysfs directory for a user and free up * corresponding structures. */ static void cleanup_user_struct(struct work_struct *w) { - struct user_struct *up = container_of(w, struct user_struct, work); + struct user_struct *up = container_of(w, struct user_struct, work.work); unsigned long flags; int remove_user = 0; @@ -297,15 +299,12 @@ static void cleanup_user_struct(struct work_struct *w) */ uids_mutex_lock(); - local_irq_save(flags); - - if (atomic_dec_and_lock(&up->__count, &uidhash_lock)) { + spin_lock_irqsave(&uidhash_lock, flags); + if (atomic_read(&up->__count) == 0) { uid_hash_remove(up); remove_user = 1; - spin_unlock_irqrestore(&uidhash_lock, flags); - } else { - local_irq_restore(flags); } + spin_unlock_irqrestore(&uidhash_lock, flags); if (!remove_user) goto done; @@ -331,16 +330,28 @@ done: */ static void free_user(struct user_struct *up, unsigned long flags) { - /* restore back the count */ - atomic_inc(&up->__count); + INIT_DELAYED_WORK(&up->work, cleanup_user_struct); + schedule_delayed_work(&up->work, msecs_to_jiffies(1000)); spin_unlock_irqrestore(&uidhash_lock, flags); - - INIT_WORK(&up->work, cleanup_user_struct); - schedule_work(&up->work); } #else /* CONFIG_USER_SCHED && CONFIG_SYSFS */ +static struct user_struct *uid_hash_find(uid_t uid, struct hlist_head *hashent) +{ + struct user_struct *user; + struct hlist_node *h; + + hlist_for_each_entry(user, h, hashent, uidhash_node) { + if (user->uid == uid) { + atomic_inc(&user->__count); + return user; + } + } + + return NULL; +} + int uids_sysfs_init(void) { return 0; } static inline int uids_user_create(struct user_struct *up) { return 0; } static inline void uids_mutex_lock(void) { } diff --git a/kernel/utsname.c b/kernel/utsname.c index 815237a55af..8a82b4b8ea5 100644 --- a/kernel/utsname.c +++ b/kernel/utsname.c @@ -15,6 +15,16 @@ #include <linux/err.h> #include <linux/slab.h> +static struct uts_namespace *create_uts_ns(void) +{ + struct uts_namespace *uts_ns; + + uts_ns = kmalloc(sizeof(struct uts_namespace), GFP_KERNEL); + if (uts_ns) + kref_init(&uts_ns->kref); + return uts_ns; +} + /* * Clone a new ns copying an original utsname, setting refcount to 1 * @old_ns: namespace to clone @@ -24,14 +34,13 @@ static struct uts_namespace *clone_uts_ns(struct uts_namespace *old_ns) { struct uts_namespace *ns; - ns = kmalloc(sizeof(struct uts_namespace), GFP_KERNEL); + ns = create_uts_ns(); if (!ns) return ERR_PTR(-ENOMEM); down_read(&uts_sem); memcpy(&ns->name, &old_ns->name, sizeof(ns->name)); up_read(&uts_sem); - kref_init(&ns->kref); return ns; } diff --git a/kernel/utsname_sysctl.c b/kernel/utsname_sysctl.c index 3b34b354593..a2cd77e70d4 100644 --- a/kernel/utsname_sysctl.c +++ b/kernel/utsname_sysctl.c @@ -37,19 +37,19 @@ static void put_uts(ctl_table *table, int write, void *which) up_write(&uts_sem); } -#ifdef CONFIG_PROC_FS +#ifdef CONFIG_PROC_SYSCTL /* * Special case of dostring for the UTS structure. This has locks * to observe. Should this be in kernel/sys.c ???? */ -static int proc_do_uts_string(ctl_table *table, int write, struct file *filp, +static int proc_do_uts_string(ctl_table *table, int write, void __user *buffer, size_t *lenp, loff_t *ppos) { struct ctl_table uts_table; int r; memcpy(&uts_table, table, sizeof(uts_table)); uts_table.data = get_uts(table, write); - r = proc_dostring(&uts_table,write,filp,buffer,lenp, ppos); + r = proc_dostring(&uts_table,write,buffer,lenp, ppos); put_uts(table, write, uts_table.data); return r; } @@ -57,78 +57,47 @@ static int proc_do_uts_string(ctl_table *table, int write, struct file *filp, #define proc_do_uts_string NULL #endif - -#ifdef CONFIG_SYSCTL_SYSCALL -/* The generic string strategy routine: */ -static int sysctl_uts_string(ctl_table *table, - void __user *oldval, size_t __user *oldlenp, - void __user *newval, size_t newlen) -{ - struct ctl_table uts_table; - int r, write; - write = newval && newlen; - memcpy(&uts_table, table, sizeof(uts_table)); - uts_table.data = get_uts(table, write); - r = sysctl_string(&uts_table, oldval, oldlenp, newval, newlen); - put_uts(table, write, uts_table.data); - return r; -} -#else -#define sysctl_uts_string NULL -#endif - static struct ctl_table uts_kern_table[] = { { - .ctl_name = KERN_OSTYPE, .procname = "ostype", .data = init_uts_ns.name.sysname, .maxlen = sizeof(init_uts_ns.name.sysname), .mode = 0444, .proc_handler = proc_do_uts_string, - .strategy = sysctl_uts_string, }, { - .ctl_name = KERN_OSRELEASE, .procname = "osrelease", .data = init_uts_ns.name.release, .maxlen = sizeof(init_uts_ns.name.release), .mode = 0444, .proc_handler = proc_do_uts_string, - .strategy = sysctl_uts_string, }, { - .ctl_name = KERN_VERSION, .procname = "version", .data = init_uts_ns.name.version, .maxlen = sizeof(init_uts_ns.name.version), .mode = 0444, .proc_handler = proc_do_uts_string, - .strategy = sysctl_uts_string, }, { - .ctl_name = KERN_NODENAME, .procname = "hostname", .data = init_uts_ns.name.nodename, .maxlen = sizeof(init_uts_ns.name.nodename), .mode = 0644, .proc_handler = proc_do_uts_string, - .strategy = sysctl_uts_string, }, { - .ctl_name = KERN_DOMAINNAME, .procname = "domainname", .data = init_uts_ns.name.domainname, .maxlen = sizeof(init_uts_ns.name.domainname), .mode = 0644, .proc_handler = proc_do_uts_string, - .strategy = sysctl_uts_string, }, {} }; static struct ctl_table uts_root_table[] = { { - .ctl_name = CTL_KERN, .procname = "kernel", .mode = 0555, .child = uts_kern_table, diff --git a/kernel/wait.c b/kernel/wait.c index 42a2dbc181c..c4bd3d825f3 100644 --- a/kernel/wait.c +++ b/kernel/wait.c @@ -10,13 +10,14 @@ #include <linux/wait.h> #include <linux/hash.h> -void init_waitqueue_head(wait_queue_head_t *q) +void __init_waitqueue_head(wait_queue_head_t *q, struct lock_class_key *key) { spin_lock_init(&q->lock); + lockdep_set_class(&q->lock, key); INIT_LIST_HEAD(&q->task_list); } -EXPORT_SYMBOL(init_waitqueue_head); +EXPORT_SYMBOL(__init_waitqueue_head); void add_wait_queue(wait_queue_head_t *q, wait_queue_t *wait) { @@ -154,7 +155,7 @@ void abort_exclusive_wait(wait_queue_head_t *q, wait_queue_t *wait, if (!list_empty(&wait->task_list)) list_del_init(&wait->task_list); else if (waitqueue_active(q)) - __wake_up_common(q, mode, 1, 0, key); + __wake_up_locked_key(q, mode, key); spin_unlock_irqrestore(&q->lock, flags); } EXPORT_SYMBOL(abort_exclusive_wait); diff --git a/kernel/workqueue.c b/kernel/workqueue.c index 9aedd9fd825..67e526b6ae8 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -33,6 +33,8 @@ #include <linux/kallsyms.h> #include <linux/debug_locks.h> #include <linux/lockdep.h> +#define CREATE_TRACE_POINTS +#include <trace/events/workqueue.h> /* * The per-CPU workqueue (if single thread, we always use the first @@ -48,8 +50,6 @@ struct cpu_workqueue_struct { struct workqueue_struct *wq; struct task_struct *thread; - - int run_depth; /* Detect run_workqueue() recursion depth */ } ____cacheline_aligned; /* @@ -128,6 +128,8 @@ struct cpu_workqueue_struct *get_wq_data(struct work_struct *work) static void insert_work(struct cpu_workqueue_struct *cwq, struct work_struct *work, struct list_head *head) { + trace_workqueue_insertion(cwq->thread, work); + set_wq_data(work, cwq); /* * Ensure that we get the right work->data if we see the @@ -262,13 +264,6 @@ EXPORT_SYMBOL_GPL(queue_delayed_work_on); static void run_workqueue(struct cpu_workqueue_struct *cwq) { spin_lock_irq(&cwq->lock); - cwq->run_depth++; - if (cwq->run_depth > 3) { - /* morton gets to eat his hat */ - printk("%s: recursion depth exceeded: %d\n", - __func__, cwq->run_depth); - dump_stack(); - } while (!list_empty(&cwq->worklist)) { struct work_struct *work = list_entry(cwq->worklist.next, struct work_struct, entry); @@ -284,7 +279,7 @@ static void run_workqueue(struct cpu_workqueue_struct *cwq) */ struct lockdep_map lockdep_map = work->lockdep_map; #endif - + trace_workqueue_execution(cwq->thread, work); cwq->current_work = work; list_del_init(cwq->worklist.next); spin_unlock_irq(&cwq->lock); @@ -311,7 +306,6 @@ static void run_workqueue(struct cpu_workqueue_struct *cwq) spin_lock_irq(&cwq->lock); cwq->current_work = NULL; } - cwq->run_depth--; spin_unlock_irq(&cwq->lock); } @@ -323,8 +317,6 @@ static int worker_thread(void *__cwq) if (cwq->wq->freezeable) set_freezable(); - set_user_nice(current, -5); - for (;;) { prepare_to_wait(&cwq->more_work, &wait, TASK_INTERRUPTIBLE); if (!freezing(current) && @@ -368,29 +360,20 @@ static void insert_wq_barrier(struct cpu_workqueue_struct *cwq, static int flush_cpu_workqueue(struct cpu_workqueue_struct *cwq) { - int active; - - if (cwq->thread == current) { - /* - * Probably keventd trying to flush its own queue. So simply run - * it by hand rather than deadlocking. - */ - run_workqueue(cwq); - active = 1; - } else { - struct wq_barrier barr; + int active = 0; + struct wq_barrier barr; - active = 0; - spin_lock_irq(&cwq->lock); - if (!list_empty(&cwq->worklist) || cwq->current_work != NULL) { - insert_wq_barrier(cwq, &barr, &cwq->worklist); - active = 1; - } - spin_unlock_irq(&cwq->lock); + WARN_ON(cwq->thread == current); - if (active) - wait_for_completion(&barr.done); + spin_lock_irq(&cwq->lock); + if (!list_empty(&cwq->worklist) || cwq->current_work != NULL) { + insert_wq_barrier(cwq, &barr, &cwq->worklist); + active = 1; } + spin_unlock_irq(&cwq->lock); + + if (active) + wait_for_completion(&barr.done); return active; } @@ -615,7 +598,12 @@ static struct workqueue_struct *keventd_wq __read_mostly; * schedule_work - put work task in global workqueue * @work: job to be done * - * This puts a job in the kernel-global workqueue. + * Returns zero if @work was already on the kernel-global workqueue and + * non-zero otherwise. + * + * This puts a job in the kernel-global workqueue if it was not already + * queued and leaves it in the same position on the kernel-global + * workqueue otherwise. */ int schedule_work(struct work_struct *work) { @@ -652,6 +640,24 @@ int schedule_delayed_work(struct delayed_work *dwork, EXPORT_SYMBOL(schedule_delayed_work); /** + * flush_delayed_work - block until a dwork_struct's callback has terminated + * @dwork: the delayed work which is to be flushed + * + * Any timeout is cancelled, and any pending work is run immediately. + */ +void flush_delayed_work(struct delayed_work *dwork) +{ + if (del_timer_sync(&dwork->timer)) { + struct cpu_workqueue_struct *cwq; + cwq = wq_per_cpu(keventd_wq, get_cpu()); + __queue_work(cwq, &dwork->work); + put_cpu(); + } + flush_work(&dwork->work); +} +EXPORT_SYMBOL(flush_delayed_work); + +/** * schedule_delayed_work_on - queue work in global workqueue on CPU after delay * @cpu: cpu to use * @dwork: job to be done @@ -679,6 +685,7 @@ EXPORT_SYMBOL(schedule_delayed_work_on); int schedule_on_each_cpu(work_func_t func) { int cpu; + int orig = -1; struct work_struct *works; works = alloc_percpu(struct work_struct); @@ -686,14 +693,28 @@ int schedule_on_each_cpu(work_func_t func) return -ENOMEM; get_online_cpus(); + + /* + * When running in keventd don't schedule a work item on + * itself. Can just call directly because the work queue is + * already bound. This also is faster. + */ + if (current_is_keventd()) + orig = raw_smp_processor_id(); + for_each_online_cpu(cpu) { struct work_struct *work = per_cpu_ptr(works, cpu); INIT_WORK(work, func); - schedule_work_on(cpu, work); + if (cpu != orig) + schedule_work_on(cpu, work); } + if (orig >= 0) + func(per_cpu_ptr(works, orig)); + for_each_online_cpu(cpu) flush_work(per_cpu_ptr(works, cpu)); + put_online_cpus(); free_percpu(works); return 0; @@ -787,6 +808,8 @@ static int create_workqueue_thread(struct cpu_workqueue_struct *cwq, int cpu) sched_setscheduler_nocheck(p, SCHED_FIFO, ¶m); cwq->thread = p; + trace_workqueue_creation(cwq->thread, cpu); + return 0; } @@ -891,6 +914,7 @@ static void cleanup_workqueue_thread(struct cpu_workqueue_struct *cwq) * checks list_empty(), and a "normal" queue_work() can't use * a dead CPU. */ + trace_workqueue_destruction(cwq->thread); kthread_stop(cwq->thread); cwq->thread = NULL; } @@ -971,20 +995,20 @@ undo: } #ifdef CONFIG_SMP -static struct workqueue_struct *work_on_cpu_wq __read_mostly; struct work_for_cpu { - struct work_struct work; + struct completion completion; long (*fn)(void *); void *arg; long ret; }; -static void do_work_for_cpu(struct work_struct *w) +static int do_work_for_cpu(void *_wfc) { - struct work_for_cpu *wfc = container_of(w, struct work_for_cpu, work); - + struct work_for_cpu *wfc = _wfc; wfc->ret = wfc->fn(wfc->arg); + complete(&wfc->completion); + return 0; } /** @@ -995,17 +1019,23 @@ static void do_work_for_cpu(struct work_struct *w) * * This will return the value @fn returns. * It is up to the caller to ensure that the cpu doesn't go offline. + * The caller must not hold any locks which would prevent @fn from completing. */ long work_on_cpu(unsigned int cpu, long (*fn)(void *), void *arg) { - struct work_for_cpu wfc; - - INIT_WORK(&wfc.work, do_work_for_cpu); - wfc.fn = fn; - wfc.arg = arg; - queue_work_on(cpu, work_on_cpu_wq, &wfc.work); - flush_work(&wfc.work); - + struct task_struct *sub_thread; + struct work_for_cpu wfc = { + .completion = COMPLETION_INITIALIZER_ONSTACK(wfc.completion), + .fn = fn, + .arg = arg, + }; + + sub_thread = kthread_create(do_work_for_cpu, &wfc, "work_for_cpu"); + if (IS_ERR(sub_thread)) + return PTR_ERR(sub_thread); + kthread_bind(sub_thread, cpu); + wake_up_process(sub_thread); + wait_for_completion(&wfc.completion); return wfc.ret; } EXPORT_SYMBOL_GPL(work_on_cpu); @@ -1021,8 +1051,4 @@ void __init init_workqueues(void) hotcpu_notifier(workqueue_cpu_callback, 0); keventd_wq = create_workqueue("events"); BUG_ON(!keventd_wq); -#ifdef CONFIG_SMP - work_on_cpu_wq = create_workqueue("work_on_cpu"); - BUG_ON(!work_on_cpu_wq); -#endif } |