diff options
Diffstat (limited to 'kernel')
180 files changed, 22053 insertions, 14129 deletions
diff --git a/kernel/Makefile b/kernel/Makefile index 057472fbc27..353d3fe8ba3 100644 --- a/kernel/Makefile +++ b/kernel/Makefile @@ -10,8 +10,7 @@ obj-y = sched.o fork.o exec_domain.o panic.o printk.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 range.o -obj-$(CONFIG_HAVE_EARLY_RES) += early_res.o + async.o range.o jump_label.o obj-y += groups.o ifdef CONFIG_FUNCTION_TRACER @@ -23,6 +22,7 @@ CFLAGS_REMOVE_rtmutex-debug.o = -pg CFLAGS_REMOVE_cgroup-debug.o = -pg CFLAGS_REMOVE_sched_clock.o = -pg CFLAGS_REMOVE_perf_event.o = -pg +CFLAGS_REMOVE_irq_work.o = -pg endif obj-$(CONFIG_FREEZER) += freezer.o @@ -43,7 +43,7 @@ obj-$(CONFIG_RT_MUTEXES) += rtmutex.o obj-$(CONFIG_DEBUG_RT_MUTEXES) += rtmutex-debug.o obj-$(CONFIG_RT_MUTEX_TESTER) += rtmutex-tester.o obj-$(CONFIG_GENERIC_ISA_DMA) += dma.o -obj-$(CONFIG_USE_GENERIC_SMP_HELPERS) += smp.o +obj-$(CONFIG_SMP) += smp.o ifneq ($(CONFIG_SMP),y) obj-y += up.o endif @@ -70,14 +70,15 @@ obj-$(CONFIG_IKCONFIG) += configs.o obj-$(CONFIG_RESOURCE_COUNTERS) += res_counter.o obj-$(CONFIG_SMP) += stop_machine.o obj-$(CONFIG_KPROBES_SANITY_TEST) += test_kprobes.o -obj-$(CONFIG_AUDIT) += audit.o auditfilter.o audit_watch.o +obj-$(CONFIG_AUDIT) += audit.o auditfilter.o obj-$(CONFIG_AUDITSYSCALL) += auditsc.o -obj-$(CONFIG_GCOV_KERNEL) += gcov/ +obj-$(CONFIG_AUDIT_WATCH) += audit_watch.o obj-$(CONFIG_AUDIT_TREE) += audit_tree.o +obj-$(CONFIG_GCOV_KERNEL) += gcov/ obj-$(CONFIG_KPROBES) += kprobes.o obj-$(CONFIG_KGDB) += debug/ -obj-$(CONFIG_DETECT_SOFTLOCKUP) += softlockup.o obj-$(CONFIG_DETECT_HUNG_TASK) += hung_task.o +obj-$(CONFIG_LOCKUP_DETECTOR) += watchdog.o obj-$(CONFIG_GENERIC_HARDIRQS) += irq/ obj-$(CONFIG_SECCOMP) += seccomp.o obj-$(CONFIG_RCU_TORTURE_TEST) += rcutorture.o @@ -85,6 +86,7 @@ obj-$(CONFIG_TREE_RCU) += rcutree.o obj-$(CONFIG_TREE_PREEMPT_RCU) += rcutree.o obj-$(CONFIG_TREE_RCU_TRACE) += rcutree_trace.o obj-$(CONFIG_TINY_RCU) += rcutiny.o +obj-$(CONFIG_TINY_PREEMPT_RCU) += rcutiny.o obj-$(CONFIG_RELAY) += relay.o obj-$(CONFIG_SYSCTL) += utsname_sysctl.o obj-$(CONFIG_TASK_DELAY_ACCT) += delayacct.o @@ -98,9 +100,9 @@ obj-$(CONFIG_FUNCTION_TRACER) += trace/ obj-$(CONFIG_TRACING) += trace/ obj-$(CONFIG_X86_DS) += trace/ obj-$(CONFIG_RING_BUFFER) += trace/ +obj-$(CONFIG_TRACEPOINTS) += trace/ obj-$(CONFIG_SMP) += sched_cpupri.o -obj-$(CONFIG_SLOW_WORK) += slow-work.o -obj-$(CONFIG_SLOW_WORK_DEBUG) += slow-work-debugfs.o +obj-$(CONFIG_IRQ_WORK) += irq_work.o obj-$(CONFIG_PERF_EVENTS) += perf_event.o obj-$(CONFIG_HAVE_HW_BREAKPOINT) += hw_breakpoint.o obj-$(CONFIG_USER_RETURN_NOTIFIER) += user-return-notifier.o @@ -120,7 +122,7 @@ $(obj)/configs.o: $(obj)/config_data.h # config_data.h contains the same information as ikconfig.h but gzipped. # Info from config_data can be extracted from /proc/config* targets += config_data.gz -$(obj)/config_data.gz: .config FORCE +$(obj)/config_data.gz: $(KCONFIG_CONFIG) FORCE $(call if_changed,gzip) quiet_cmd_ikconfiggz = IKCFG $@ diff --git a/kernel/acct.c b/kernel/acct.c index 385b88461c2..fa7eb3de2dd 100644 --- a/kernel/acct.c +++ b/kernel/acct.c @@ -122,7 +122,7 @@ static int check_free_space(struct bsd_acct_struct *acct, struct file *file) spin_unlock(&acct_lock); /* May block */ - if (vfs_statfs(file->f_path.dentry, &sbuf)) + if (vfs_statfs(&file->f_path, &sbuf)) return res; suspend = sbuf.f_blocks * SUSPEND; resume = sbuf.f_blocks * RESUME; diff --git a/kernel/async.c b/kernel/async.c index 15319d6c18f..cd9dbb913c7 100644 --- a/kernel/async.c +++ b/kernel/async.c @@ -49,40 +49,33 @@ asynchronous and synchronous parts of the kernel. */ #include <linux/async.h> -#include <linux/bug.h> #include <linux/module.h> #include <linux/wait.h> #include <linux/sched.h> -#include <linux/init.h> -#include <linux/kthread.h> -#include <linux/delay.h> #include <linux/slab.h> +#include <linux/workqueue.h> #include <asm/atomic.h> static async_cookie_t next_cookie = 1; -#define MAX_THREADS 256 #define MAX_WORK 32768 static LIST_HEAD(async_pending); static LIST_HEAD(async_running); static DEFINE_SPINLOCK(async_lock); -static int async_enabled = 0; - struct async_entry { - struct list_head list; - async_cookie_t cookie; - async_func_ptr *func; - void *data; - struct list_head *running; + struct list_head list; + struct work_struct work; + async_cookie_t cookie; + async_func_ptr *func; + void *data; + struct list_head *running; }; static DECLARE_WAIT_QUEUE_HEAD(async_done); -static DECLARE_WAIT_QUEUE_HEAD(async_new); static atomic_t entry_count; -static atomic_t thread_count; extern int initcall_debug; @@ -117,27 +110,23 @@ static async_cookie_t lowest_in_progress(struct list_head *running) spin_unlock_irqrestore(&async_lock, flags); return ret; } + /* * pick the first pending entry and run it */ -static void run_one_entry(void) +static void async_run_entry_fn(struct work_struct *work) { + struct async_entry *entry = + container_of(work, struct async_entry, work); unsigned long flags; - struct async_entry *entry; ktime_t calltime, delta, rettime; - /* 1) pick one task from the pending queue */ - + /* 1) move self to the running queue */ spin_lock_irqsave(&async_lock, flags); - if (list_empty(&async_pending)) - goto out; - entry = list_first_entry(&async_pending, struct async_entry, list); - - /* 2) move it to the running queue */ list_move_tail(&entry->list, entry->running); spin_unlock_irqrestore(&async_lock, flags); - /* 3) run it (and print duration)*/ + /* 2) run (and print duration) */ if (initcall_debug && system_state == SYSTEM_BOOTING) { printk("calling %lli_%pF @ %i\n", (long long)entry->cookie, entry->func, task_pid_nr(current)); @@ -153,31 +142,25 @@ static void run_one_entry(void) (long long)ktime_to_ns(delta) >> 10); } - /* 4) remove it from the running queue */ + /* 3) remove self from the running queue */ spin_lock_irqsave(&async_lock, flags); list_del(&entry->list); - /* 5) free the entry */ + /* 4) free the entry */ kfree(entry); atomic_dec(&entry_count); spin_unlock_irqrestore(&async_lock, flags); - /* 6) wake up any waiters. */ + /* 5) wake up any waiters */ wake_up(&async_done); - return; - -out: - spin_unlock_irqrestore(&async_lock, flags); } - static async_cookie_t __async_schedule(async_func_ptr *ptr, void *data, struct list_head *running) { struct async_entry *entry; unsigned long flags; async_cookie_t newcookie; - /* allow irq-off callers */ entry = kzalloc(sizeof(struct async_entry), GFP_ATOMIC); @@ -186,7 +169,7 @@ static async_cookie_t __async_schedule(async_func_ptr *ptr, void *data, struct l * If we're out of memory or if there's too much work * pending already, we execute synchronously. */ - if (!async_enabled || !entry || atomic_read(&entry_count) > MAX_WORK) { + if (!entry || atomic_read(&entry_count) > MAX_WORK) { kfree(entry); spin_lock_irqsave(&async_lock, flags); newcookie = next_cookie++; @@ -196,6 +179,7 @@ static async_cookie_t __async_schedule(async_func_ptr *ptr, void *data, struct l ptr(data, newcookie); return newcookie; } + INIT_WORK(&entry->work, async_run_entry_fn); entry->func = ptr; entry->data = data; entry->running = running; @@ -205,7 +189,10 @@ static async_cookie_t __async_schedule(async_func_ptr *ptr, void *data, struct l list_add_tail(&entry->list, &async_pending); atomic_inc(&entry_count); spin_unlock_irqrestore(&async_lock, flags); - wake_up(&async_new); + + /* schedule for execution */ + queue_work(system_unbound_wq, &entry->work); + return newcookie; } @@ -312,87 +299,3 @@ void async_synchronize_cookie(async_cookie_t cookie) async_synchronize_cookie_domain(cookie, &async_running); } EXPORT_SYMBOL_GPL(async_synchronize_cookie); - - -static int async_thread(void *unused) -{ - DECLARE_WAITQUEUE(wq, current); - add_wait_queue(&async_new, &wq); - - while (!kthread_should_stop()) { - int ret = HZ; - set_current_state(TASK_INTERRUPTIBLE); - /* - * check the list head without lock.. false positives - * are dealt with inside run_one_entry() while holding - * the lock. - */ - rmb(); - if (!list_empty(&async_pending)) - run_one_entry(); - else - ret = schedule_timeout(HZ); - - if (ret == 0) { - /* - * we timed out, this means we as thread are redundant. - * we sign off and die, but we to avoid any races there - * is a last-straw check to see if work snuck in. - */ - atomic_dec(&thread_count); - wmb(); /* manager must see our departure first */ - if (list_empty(&async_pending)) - break; - /* - * woops work came in between us timing out and us - * signing off; we need to stay alive and keep working. - */ - atomic_inc(&thread_count); - } - } - remove_wait_queue(&async_new, &wq); - - return 0; -} - -static int async_manager_thread(void *unused) -{ - DECLARE_WAITQUEUE(wq, current); - add_wait_queue(&async_new, &wq); - - while (!kthread_should_stop()) { - int tc, ec; - - set_current_state(TASK_INTERRUPTIBLE); - - tc = atomic_read(&thread_count); - rmb(); - ec = atomic_read(&entry_count); - - while (tc < ec && tc < MAX_THREADS) { - if (IS_ERR(kthread_run(async_thread, NULL, "async/%i", - tc))) { - msleep(100); - continue; - } - atomic_inc(&thread_count); - tc++; - } - - schedule(); - } - remove_wait_queue(&async_new, &wq); - - return 0; -} - -static int __init async_init(void) -{ - async_enabled = - !IS_ERR(kthread_run(async_manager_thread, NULL, "async/mgr")); - - WARN_ON(!async_enabled); - return 0; -} - -core_initcall(async_init); diff --git a/kernel/audit.c b/kernel/audit.c index c71bd26631a..e4956244ae5 100644 --- a/kernel/audit.c +++ b/kernel/audit.c @@ -56,7 +56,6 @@ #include <net/netlink.h> #include <linux/skbuff.h> #include <linux/netlink.h> -#include <linux/inotify.h> #include <linux/freezer.h> #include <linux/tty.h> @@ -401,13 +400,13 @@ static void kauditd_send_skb(struct sk_buff *skb) if (err < 0) { BUG_ON(err != -ECONNREFUSED); /* Shouldn't happen */ printk(KERN_ERR "audit: *NO* daemon at audit_pid=%d\n", audit_pid); - audit_log_lost("auditd dissapeared\n"); + audit_log_lost("auditd disappeared\n"); audit_pid = 0; /* we might get lucky and get this in the next auditd */ audit_hold_skb(skb); } else /* drop the extra reference if sent ok */ - kfree_skb(skb); + consume_skb(skb); } static int kauditd_thread(void *dummy) @@ -468,23 +467,16 @@ static int audit_prepare_user_tty(pid_t pid, uid_t loginuid, u32 sessionid) struct task_struct *tsk; int err; - read_lock(&tasklist_lock); + rcu_read_lock(); tsk = find_task_by_vpid(pid); - err = -ESRCH; - if (!tsk) - goto out; - err = 0; - - spin_lock_irq(&tsk->sighand->siglock); - if (!tsk->signal->audit_tty) - err = -EPERM; - spin_unlock_irq(&tsk->sighand->siglock); - if (err) - goto out; - - tty_audit_push_task(tsk, loginuid, sessionid); -out: - read_unlock(&tasklist_lock); + if (!tsk) { + rcu_read_unlock(); + return -ESRCH; + } + get_task_struct(tsk); + rcu_read_unlock(); + err = tty_audit_push_task(tsk, loginuid, sessionid); + put_task_struct(tsk); return err; } @@ -507,7 +499,7 @@ int audit_send_list(void *_dest) } struct sk_buff *audit_make_reply(int pid, int seq, int type, int done, - int multi, void *payload, int size) + int multi, const void *payload, int size) { struct sk_buff *skb; struct nlmsghdr *nlh; @@ -556,8 +548,8 @@ static int audit_send_reply_thread(void *arg) * Allocates an skb, builds the netlink message, and sends it to the pid. * No failure notifications. */ -void audit_send_reply(int pid, int seq, int type, int done, int multi, - void *payload, int size) +static void audit_send_reply(int pid, int seq, int type, int done, int multi, + const void *payload, int size) { struct sk_buff *skb; struct task_struct *tsk; @@ -881,40 +873,40 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh) case AUDIT_TTY_GET: { struct audit_tty_status s; struct task_struct *tsk; + unsigned long flags; - read_lock(&tasklist_lock); + rcu_read_lock(); tsk = find_task_by_vpid(pid); - if (!tsk) - err = -ESRCH; - else { - spin_lock_irq(&tsk->sighand->siglock); + if (tsk && lock_task_sighand(tsk, &flags)) { s.enabled = tsk->signal->audit_tty != 0; - spin_unlock_irq(&tsk->sighand->siglock); - } - read_unlock(&tasklist_lock); - audit_send_reply(NETLINK_CB(skb).pid, seq, AUDIT_TTY_GET, 0, 0, - &s, sizeof(s)); + unlock_task_sighand(tsk, &flags); + } else + err = -ESRCH; + rcu_read_unlock(); + + if (!err) + audit_send_reply(NETLINK_CB(skb).pid, seq, + AUDIT_TTY_GET, 0, 0, &s, sizeof(s)); break; } case AUDIT_TTY_SET: { struct audit_tty_status *s; struct task_struct *tsk; + unsigned long flags; if (nlh->nlmsg_len < sizeof(struct audit_tty_status)) return -EINVAL; s = data; if (s->enabled != 0 && s->enabled != 1) return -EINVAL; - read_lock(&tasklist_lock); + rcu_read_lock(); tsk = find_task_by_vpid(pid); - if (!tsk) - err = -ESRCH; - else { - spin_lock_irq(&tsk->sighand->siglock); + if (tsk && lock_task_sighand(tsk, &flags)) { tsk->signal->audit_tty = s->enabled != 0; - spin_unlock_irq(&tsk->sighand->siglock); - } - read_unlock(&tasklist_lock); + unlock_task_sighand(tsk, &flags); + } else + err = -ESRCH; + rcu_read_unlock(); break; } default: diff --git a/kernel/audit.h b/kernel/audit.h index 208687be4f3..91e7071c4d2 100644 --- a/kernel/audit.h +++ b/kernel/audit.h @@ -84,10 +84,7 @@ extern int audit_compare_dname_path(const char *dname, const char *path, int *dirlen); extern struct sk_buff * audit_make_reply(int pid, int seq, int type, int done, int multi, - void *payload, int size); -extern void audit_send_reply(int pid, int seq, int type, - int done, int multi, - void *payload, int size); + const void *payload, int size); extern void audit_panic(const char *message); struct audit_netlink_list { @@ -103,21 +100,27 @@ extern struct mutex audit_filter_mutex; extern void audit_free_rule_rcu(struct rcu_head *); extern struct list_head audit_filter_list[]; +extern struct audit_entry *audit_dupe_rule(struct audit_krule *old); + /* audit watch functions */ -extern unsigned long audit_watch_inode(struct audit_watch *watch); -extern dev_t audit_watch_dev(struct audit_watch *watch); +#ifdef CONFIG_AUDIT_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 int audit_add_watch(struct audit_krule *krule, struct list_head **list); +extern void audit_remove_watch_rule(struct audit_krule *krule); 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); +extern int audit_watch_compare(struct audit_watch *watch, unsigned long ino, dev_t dev); +#else +#define audit_put_watch(w) {} +#define audit_get_watch(w) {} +#define audit_to_watch(k, p, l, o) (-EINVAL) +#define audit_add_watch(k, l) (-EINVAL) +#define audit_remove_watch_rule(k) BUG() +#define audit_watch_path(w) "" +#define audit_watch_compare(w, i, d) 0 + +#endif /* CONFIG_AUDIT_WATCH */ #ifdef CONFIG_AUDIT_TREE extern struct audit_chunk *audit_tree_lookup(const struct inode *); diff --git a/kernel/audit_tree.c b/kernel/audit_tree.c index 46a57b57a33..37b2bea170c 100644 --- a/kernel/audit_tree.c +++ b/kernel/audit_tree.c @@ -1,5 +1,5 @@ #include "audit.h" -#include <linux/inotify.h> +#include <linux/fsnotify_backend.h> #include <linux/namei.h> #include <linux/mount.h> #include <linux/kthread.h> @@ -22,7 +22,7 @@ struct audit_tree { struct audit_chunk { struct list_head hash; - struct inotify_watch watch; + struct fsnotify_mark mark; struct list_head trees; /* with root here */ int dead; int count; @@ -59,7 +59,7 @@ static LIST_HEAD(prune_list); * tree is refcounted; one reference for "some rules on rules_list refer to * it", one for each chunk with pointer to it. * - * chunk is refcounted by embedded inotify_watch + .refs (non-zero refcount + * chunk is refcounted by embedded fsnotify_mark + .refs (non-zero refcount * of watch contributes 1 to .refs). * * node.index allows to get from node.list to containing chunk. @@ -68,7 +68,7 @@ static LIST_HEAD(prune_list); * that makes a difference. Some. */ -static struct inotify_handle *rtree_ih; +static struct fsnotify_group *audit_tree_group; static struct audit_tree *alloc_tree(const char *s) { @@ -111,29 +111,6 @@ const char *audit_tree_path(struct audit_tree *tree) return tree->pathname; } -static struct audit_chunk *alloc_chunk(int count) -{ - struct audit_chunk *chunk; - size_t size; - int i; - - size = offsetof(struct audit_chunk, owners) + count * sizeof(struct node); - chunk = kzalloc(size, GFP_KERNEL); - if (!chunk) - return NULL; - - INIT_LIST_HEAD(&chunk->hash); - INIT_LIST_HEAD(&chunk->trees); - chunk->count = count; - atomic_long_set(&chunk->refs, 1); - for (i = 0; i < count; i++) { - INIT_LIST_HEAD(&chunk->owners[i].list); - chunk->owners[i].index = i; - } - inotify_init_watch(&chunk->watch); - return chunk; -} - static void free_chunk(struct audit_chunk *chunk) { int i; @@ -157,6 +134,35 @@ static void __put_chunk(struct rcu_head *rcu) audit_put_chunk(chunk); } +static void audit_tree_destroy_watch(struct fsnotify_mark *entry) +{ + struct audit_chunk *chunk = container_of(entry, struct audit_chunk, mark); + call_rcu(&chunk->head, __put_chunk); +} + +static struct audit_chunk *alloc_chunk(int count) +{ + struct audit_chunk *chunk; + size_t size; + int i; + + size = offsetof(struct audit_chunk, owners) + count * sizeof(struct node); + chunk = kzalloc(size, GFP_KERNEL); + if (!chunk) + return NULL; + + INIT_LIST_HEAD(&chunk->hash); + INIT_LIST_HEAD(&chunk->trees); + chunk->count = count; + atomic_long_set(&chunk->refs, 1); + for (i = 0; i < count; i++) { + INIT_LIST_HEAD(&chunk->owners[i].list); + chunk->owners[i].index = i; + } + fsnotify_init_mark(&chunk->mark, audit_tree_destroy_watch); + return chunk; +} + enum {HASH_SIZE = 128}; static struct list_head chunk_hash_heads[HASH_SIZE]; static __cacheline_aligned_in_smp DEFINE_SPINLOCK(hash_lock); @@ -167,10 +173,15 @@ static inline struct list_head *chunk_hash(const struct inode *inode) return chunk_hash_heads + n % HASH_SIZE; } -/* hash_lock is held by caller */ +/* hash_lock & entry->lock is held by caller */ static void insert_hash(struct audit_chunk *chunk) { - struct list_head *list = chunk_hash(chunk->watch.inode); + struct fsnotify_mark *entry = &chunk->mark; + struct list_head *list; + + if (!entry->i.inode) + return; + list = chunk_hash(entry->i.inode); list_add_rcu(&chunk->hash, list); } @@ -181,7 +192,8 @@ struct audit_chunk *audit_tree_lookup(const struct inode *inode) struct audit_chunk *p; list_for_each_entry_rcu(p, list, hash) { - if (p->watch.inode == inode) { + /* mark.inode may have gone NULL, but who cares? */ + if (p->mark.i.inode == inode) { atomic_long_inc(&p->refs); return p; } @@ -210,38 +222,24 @@ static struct audit_chunk *find_chunk(struct node *p) static void untag_chunk(struct node *p) { struct audit_chunk *chunk = find_chunk(p); - struct audit_chunk *new; + struct fsnotify_mark *entry = &chunk->mark; + struct audit_chunk *new = NULL; struct audit_tree *owner; int size = chunk->count - 1; int i, j; - if (!pin_inotify_watch(&chunk->watch)) { - /* - * Filesystem is shutting down; all watches are getting - * evicted, just take it off the node list for this - * tree and let the eviction logics take care of the - * rest. - */ - owner = p->owner; - if (owner->root == chunk) { - list_del_init(&owner->same_root); - owner->root = NULL; - } - list_del_init(&p->list); - p->owner = NULL; - put_tree(owner); - return; - } + fsnotify_get_mark(entry); spin_unlock(&hash_lock); - /* - * pin_inotify_watch() succeeded, so the watch won't go away - * from under us. - */ - mutex_lock(&chunk->watch.inode->inotify_mutex); - if (chunk->dead) { - mutex_unlock(&chunk->watch.inode->inotify_mutex); + if (size) + new = alloc_chunk(size); + + spin_lock(&entry->lock); + if (chunk->dead || !entry->i.inode) { + spin_unlock(&entry->lock); + if (new) + free_chunk(new); goto out; } @@ -256,16 +254,17 @@ static void untag_chunk(struct node *p) list_del_init(&p->list); list_del_rcu(&chunk->hash); spin_unlock(&hash_lock); - inotify_evict_watch(&chunk->watch); - mutex_unlock(&chunk->watch.inode->inotify_mutex); - put_inotify_watch(&chunk->watch); + spin_unlock(&entry->lock); + fsnotify_destroy_mark(entry); + fsnotify_put_mark(entry); goto out; } - new = alloc_chunk(size); if (!new) goto Fallback; - if (inotify_clone_watch(&chunk->watch, &new->watch) < 0) { + + fsnotify_duplicate_mark(&new->mark, entry); + if (fsnotify_add_mark(&new->mark, new->mark.group, new->mark.i.inode, NULL, 1)) { free_chunk(new); goto Fallback; } @@ -298,9 +297,9 @@ static void untag_chunk(struct node *p) list_for_each_entry(owner, &new->trees, same_root) owner->root = new; spin_unlock(&hash_lock); - inotify_evict_watch(&chunk->watch); - mutex_unlock(&chunk->watch.inode->inotify_mutex); - put_inotify_watch(&chunk->watch); + spin_unlock(&entry->lock); + fsnotify_destroy_mark(entry); + fsnotify_put_mark(entry); goto out; Fallback: @@ -314,31 +313,33 @@ Fallback: p->owner = NULL; put_tree(owner); spin_unlock(&hash_lock); - mutex_unlock(&chunk->watch.inode->inotify_mutex); + spin_unlock(&entry->lock); out: - unpin_inotify_watch(&chunk->watch); + fsnotify_put_mark(entry); spin_lock(&hash_lock); } static int create_chunk(struct inode *inode, struct audit_tree *tree) { + struct fsnotify_mark *entry; struct audit_chunk *chunk = alloc_chunk(1); if (!chunk) return -ENOMEM; - if (inotify_add_watch(rtree_ih, &chunk->watch, inode, IN_IGNORED | IN_DELETE_SELF) < 0) { + entry = &chunk->mark; + if (fsnotify_add_mark(entry, audit_tree_group, inode, NULL, 0)) { free_chunk(chunk); return -ENOSPC; } - mutex_lock(&inode->inotify_mutex); + spin_lock(&entry->lock); spin_lock(&hash_lock); if (tree->goner) { spin_unlock(&hash_lock); chunk->dead = 1; - inotify_evict_watch(&chunk->watch); - mutex_unlock(&inode->inotify_mutex); - put_inotify_watch(&chunk->watch); + spin_unlock(&entry->lock); + fsnotify_destroy_mark(entry); + fsnotify_put_mark(entry); return 0; } chunk->owners[0].index = (1U << 31); @@ -351,30 +352,31 @@ static int create_chunk(struct inode *inode, struct audit_tree *tree) } insert_hash(chunk); spin_unlock(&hash_lock); - mutex_unlock(&inode->inotify_mutex); + spin_unlock(&entry->lock); return 0; } /* the first tagged inode becomes root of tree */ static int tag_chunk(struct inode *inode, struct audit_tree *tree) { - struct inotify_watch *watch; + struct fsnotify_mark *old_entry, *chunk_entry; struct audit_tree *owner; struct audit_chunk *chunk, *old; struct node *p; int n; - if (inotify_find_watch(rtree_ih, inode, &watch) < 0) + old_entry = fsnotify_find_inode_mark(audit_tree_group, inode); + if (!old_entry) return create_chunk(inode, tree); - old = container_of(watch, struct audit_chunk, watch); + old = container_of(old_entry, struct audit_chunk, mark); /* are we already there? */ spin_lock(&hash_lock); for (n = 0; n < old->count; n++) { if (old->owners[n].owner == tree) { spin_unlock(&hash_lock); - put_inotify_watch(&old->watch); + fsnotify_put_mark(old_entry); return 0; } } @@ -382,25 +384,44 @@ static int tag_chunk(struct inode *inode, struct audit_tree *tree) chunk = alloc_chunk(old->count + 1); if (!chunk) { - put_inotify_watch(&old->watch); + fsnotify_put_mark(old_entry); return -ENOMEM; } - mutex_lock(&inode->inotify_mutex); - if (inotify_clone_watch(&old->watch, &chunk->watch) < 0) { - mutex_unlock(&inode->inotify_mutex); - put_inotify_watch(&old->watch); + chunk_entry = &chunk->mark; + + spin_lock(&old_entry->lock); + if (!old_entry->i.inode) { + /* old_entry is being shot, lets just lie */ + spin_unlock(&old_entry->lock); + fsnotify_put_mark(old_entry); free_chunk(chunk); + return -ENOENT; + } + + fsnotify_duplicate_mark(chunk_entry, old_entry); + if (fsnotify_add_mark(chunk_entry, chunk_entry->group, chunk_entry->i.inode, NULL, 1)) { + spin_unlock(&old_entry->lock); + free_chunk(chunk); + fsnotify_put_mark(old_entry); return -ENOSPC; } + + /* even though we hold old_entry->lock, this is safe since chunk_entry->lock could NEVER have been grabbed before */ + spin_lock(&chunk_entry->lock); spin_lock(&hash_lock); + + /* we now hold old_entry->lock, chunk_entry->lock, and hash_lock */ if (tree->goner) { spin_unlock(&hash_lock); chunk->dead = 1; - inotify_evict_watch(&chunk->watch); - mutex_unlock(&inode->inotify_mutex); - put_inotify_watch(&old->watch); - put_inotify_watch(&chunk->watch); + spin_unlock(&chunk_entry->lock); + spin_unlock(&old_entry->lock); + + fsnotify_destroy_mark(chunk_entry); + + fsnotify_put_mark(chunk_entry); + fsnotify_put_mark(old_entry); return 0; } list_replace_init(&old->trees, &chunk->trees); @@ -426,10 +447,11 @@ static int tag_chunk(struct inode *inode, struct audit_tree *tree) list_add(&tree->same_root, &chunk->trees); } spin_unlock(&hash_lock); - inotify_evict_watch(&old->watch); - mutex_unlock(&inode->inotify_mutex); - put_inotify_watch(&old->watch); /* pair to inotify_find_watch */ - put_inotify_watch(&old->watch); /* and kill it */ + spin_unlock(&chunk_entry->lock); + spin_unlock(&old_entry->lock); + fsnotify_destroy_mark(old_entry); + fsnotify_put_mark(old_entry); /* pair to fsnotify_find mark_entry */ + fsnotify_put_mark(old_entry); /* and kill it */ return 0; } @@ -584,7 +606,9 @@ void audit_trim_trees(void) spin_lock(&hash_lock); list_for_each_entry(node, &tree->chunks, list) { - struct inode *inode = find_chunk(node)->watch.inode; + struct audit_chunk *chunk = find_chunk(node); + /* this could be NULL if the watch is dieing else where... */ + struct inode *inode = chunk->mark.i.inode; node->index |= 1U<<31; if (iterate_mounts(compare_root, inode, root_mnt)) node->index &= ~(1U<<31); @@ -846,7 +870,6 @@ void audit_kill_trees(struct list_head *list) * Here comes the stuff asynchronous to auditctl operations */ -/* inode->inotify_mutex is locked */ static void evict_chunk(struct audit_chunk *chunk) { struct audit_tree *owner; @@ -885,35 +908,46 @@ static void evict_chunk(struct audit_chunk *chunk) mutex_unlock(&audit_filter_mutex); } -static void handle_event(struct inotify_watch *watch, u32 wd, u32 mask, - u32 cookie, const char *dname, struct inode *inode) +static int audit_tree_handle_event(struct fsnotify_group *group, + struct fsnotify_mark *inode_mark, + struct fsnotify_mark *vfsmonut_mark, + struct fsnotify_event *event) +{ + BUG(); + return -EOPNOTSUPP; +} + +static void audit_tree_freeing_mark(struct fsnotify_mark *entry, struct fsnotify_group *group) { - struct audit_chunk *chunk = container_of(watch, struct audit_chunk, watch); + struct audit_chunk *chunk = container_of(entry, struct audit_chunk, mark); - if (mask & IN_IGNORED) { - evict_chunk(chunk); - put_inotify_watch(watch); - } + evict_chunk(chunk); + fsnotify_put_mark(entry); } -static void destroy_watch(struct inotify_watch *watch) +static bool audit_tree_send_event(struct fsnotify_group *group, struct inode *inode, + struct fsnotify_mark *inode_mark, + struct fsnotify_mark *vfsmount_mark, + __u32 mask, void *data, int data_type) { - struct audit_chunk *chunk = container_of(watch, struct audit_chunk, watch); - call_rcu(&chunk->head, __put_chunk); + return false; } -static const struct inotify_operations rtree_inotify_ops = { - .handle_event = handle_event, - .destroy_watch = destroy_watch, +static const struct fsnotify_ops audit_tree_ops = { + .handle_event = audit_tree_handle_event, + .should_send_event = audit_tree_send_event, + .free_group_priv = NULL, + .free_event_priv = NULL, + .freeing_mark = audit_tree_freeing_mark, }; static int __init audit_tree_init(void) { int i; - rtree_ih = inotify_init(&rtree_inotify_ops); - if (IS_ERR(rtree_ih)) - audit_panic("cannot initialize inotify handle for rectree watches"); + audit_tree_group = fsnotify_alloc_group(&audit_tree_ops); + if (IS_ERR(audit_tree_group)) + audit_panic("cannot initialize fsnotify group for rectree watches"); for (i = 0; i < HASH_SIZE; i++) INIT_LIST_HEAD(&chunk_hash_heads[i]); diff --git a/kernel/audit_watch.c b/kernel/audit_watch.c index 8df43696f4b..d2e3c786646 100644 --- a/kernel/audit_watch.c +++ b/kernel/audit_watch.c @@ -24,18 +24,18 @@ #include <linux/kthread.h> #include <linux/mutex.h> #include <linux/fs.h> +#include <linux/fsnotify_backend.h> #include <linux/namei.h> #include <linux/netlink.h> #include <linux/sched.h> #include <linux/slab.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 + * audit_parent: lifetime is from audit_init_parent() to receipt of an FS_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 @@ -51,40 +51,61 @@ struct audit_watch { 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 list_head rules; /* anchor for krule->rlist */ }; 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 */ + struct list_head watches; /* anchor for audit_watch->wlist */ + struct fsnotify_mark mark; /* fsnotify mark on the inode */ }; -/* Inotify handle. */ -struct inotify_handle *audit_ih; +/* fsnotify handle. */ +static struct fsnotify_group *audit_watch_group; -/* - * 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 +/* fsnotify events we care about. */ +#define AUDIT_FS_WATCH (FS_MOVE | FS_CREATE | FS_DELETE | FS_DELETE_SELF |\ + FS_MOVE_SELF | FS_EVENT_ON_CHILD) -/* 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 audit_parent *parent) +{ + WARN_ON(!list_empty(&parent->watches)); + kfree(parent); +} -static void audit_free_parent(struct inotify_watch *i_watch) +static void audit_watch_free_mark(struct fsnotify_mark *entry) { struct audit_parent *parent; - parent = container_of(i_watch, struct audit_parent, wdata); - WARN_ON(!list_empty(&parent->watches)); - kfree(parent); + parent = container_of(entry, struct audit_parent, mark); + audit_free_parent(parent); +} + +static void audit_get_parent(struct audit_parent *parent) +{ + if (likely(parent)) + fsnotify_get_mark(&parent->mark); +} + +static void audit_put_parent(struct audit_parent *parent) +{ + if (likely(parent)) + fsnotify_put_mark(&parent->mark); +} + +/* + * Find and return the audit_parent on the given inode. If found a reference + * is taken on this parent. + */ +static inline struct audit_parent *audit_find_parent(struct inode *inode) +{ + struct audit_parent *parent = NULL; + struct fsnotify_mark *entry; + + entry = fsnotify_find_inode_mark(audit_watch_group, inode); + if (entry) + parent = container_of(entry, struct audit_parent, mark); + + return parent; } void audit_get_watch(struct audit_watch *watch) @@ -102,10 +123,10 @@ void audit_put_watch(struct audit_watch *watch) } } -void audit_remove_watch(struct audit_watch *watch) +static void audit_remove_watch(struct audit_watch *watch) { list_del(&watch->wlist); - put_inotify_watch(&watch->parent->wdata); + audit_put_parent(watch->parent); watch->parent = NULL; audit_put_watch(watch); /* match initial get */ } @@ -115,42 +136,32 @@ 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) +int audit_watch_compare(struct audit_watch *watch, unsigned long ino, dev_t dev) { - return watch->ino; -} - -dev_t audit_watch_dev(struct audit_watch *watch) -{ - return watch->dev; + return (watch->ino != (unsigned long)-1) && + (watch->ino == ino) && + (watch->dev == dev); } /* Initialize a parent watch entry. */ static struct audit_parent *audit_init_parent(struct nameidata *ndp) { + struct inode *inode = ndp->path.dentry->d_inode; struct audit_parent *parent; - s32 wd; + int ret; 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); + + fsnotify_init_mark(&parent->mark, audit_watch_free_mark); + parent->mark.mask = AUDIT_FS_WATCH; + ret = fsnotify_add_mark(&parent->mark, audit_watch_group, inode, NULL, 0); + if (ret < 0) { + audit_free_parent(parent); + return ERR_PTR(ret); } return parent; @@ -179,7 +190,7 @@ int audit_to_watch(struct audit_krule *krule, char *path, int len, u32 op) { struct audit_watch *watch; - if (!audit_ih) + if (!audit_watch_group) return -EOPNOTSUPP; if (path[0] != '/' || path[len-1] == '/' || @@ -217,7 +228,7 @@ static struct audit_watch *audit_dupe_watch(struct audit_watch *old) new->dev = old->dev; new->ino = old->ino; - get_inotify_watch(&old->parent->wdata); + audit_get_parent(old->parent); new->parent = old->parent; out: @@ -251,15 +262,19 @@ static void audit_update_watch(struct audit_parent *parent, struct audit_entry *oentry, *nentry; mutex_lock(&audit_filter_mutex); + /* Run all of the watches on this parent looking for the one that + * matches the given dname */ 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) + if (invalidating && !audit_dummy_context()) audit_filter_inodes(current, current->audit_context); + /* updating ino will likely change which audit_hash_list we + * are on so we need a new watch for the new list */ nwatch = audit_dupe_watch(owatch); if (IS_ERR(nwatch)) { mutex_unlock(&audit_filter_mutex); @@ -275,12 +290,21 @@ static void audit_update_watch(struct audit_parent *parent, list_del(&oentry->rule.rlist); list_del_rcu(&oentry->list); - nentry = audit_dupe_rule(&oentry->rule, nwatch); + nentry = audit_dupe_rule(&oentry->rule); if (IS_ERR(nentry)) { list_del(&oentry->rule.list); audit_panic("error updating watch, removing"); } else { int h = audit_hash_ino((u32)ino); + + /* + * nentry->rule.watch == oentry->rule.watch so + * we must drop that reference and set it to our + * new watch. + */ + audit_put_watch(nentry->rule.watch); + audit_get_watch(nwatch); + nentry->rule.watch = nwatch; list_add(&nentry->rule.rlist, &nwatch->rules); list_add_rcu(&nentry->list, &audit_inode_hash[h]); list_replace(&oentry->rule.list, @@ -312,7 +336,6 @@ static void audit_remove_parent_watches(struct audit_parent *parent) 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); @@ -325,20 +348,8 @@ static void audit_remove_parent_watches(struct audit_parent *parent) 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); - } + fsnotify_destroy_mark(&parent->mark); } /* Get path information necessary for adding watches. */ @@ -389,7 +400,7 @@ static void audit_put_nd(struct nameidata *ndp, struct nameidata *ndw) } } -/* Associate the given rule with an existing parent inotify_watch. +/* Associate the given rule with an existing parent. * Caller must hold audit_filter_mutex. */ static void audit_add_to_parent(struct audit_krule *krule, struct audit_parent *parent) @@ -397,6 +408,8 @@ static void audit_add_to_parent(struct audit_krule *krule, struct audit_watch *w, *watch = krule->watch; int watch_found = 0; + BUG_ON(!mutex_is_locked(&audit_filter_mutex)); + list_for_each_entry(w, &parent->watches, wlist) { if (strcmp(watch->path, w->path)) continue; @@ -413,7 +426,7 @@ static void audit_add_to_parent(struct audit_krule *krule, } if (!watch_found) { - get_inotify_watch(&parent->wdata); + audit_get_parent(parent); watch->parent = parent; list_add(&watch->wlist, &parent->watches); @@ -423,13 +436,12 @@ static void audit_add_to_parent(struct audit_krule *krule, /* Find a matching watch entry, or add this one. * Caller must hold audit_filter_mutex. */ -int audit_add_watch(struct audit_krule *krule) +int audit_add_watch(struct audit_krule *krule, struct list_head **list) { struct audit_watch *watch = krule->watch; - struct inotify_watch *i_watch; struct audit_parent *parent; struct nameidata *ndp = NULL, *ndw = NULL; - int ret = 0; + int h, ret = 0; mutex_unlock(&audit_filter_mutex); @@ -441,47 +453,38 @@ int audit_add_watch(struct audit_krule *krule) goto error; } + mutex_lock(&audit_filter_mutex); + /* 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) { + /* either find an old parent or attach a new one */ + parent = audit_find_parent(ndp->path.dentry->d_inode); + if (!parent) { 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); + audit_add_to_parent(krule, parent); - /* match get in audit_init_parent or inotify_find_watch */ - put_inotify_watch(&parent->wdata); + /* match get in audit_find_parent or audit_init_parent */ + audit_put_parent(parent); + h = audit_hash_ino((u32)watch->ino); + *list = &audit_inode_hash[h]; error: audit_put_nd(ndp, ndw); /* NULL args OK */ return ret; } -void audit_remove_watch_rule(struct audit_krule *krule, struct list_head *list) +void audit_remove_watch_rule(struct audit_krule *krule) { struct audit_watch *watch = krule->watch; struct audit_parent *parent = watch->parent; @@ -492,53 +495,74 @@ void audit_remove_watch_rule(struct audit_krule *krule, struct list_head *list) 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); + audit_get_parent(parent); + fsnotify_destroy_mark(&parent->mark); + audit_put_parent(parent); } } } -/* 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) +static bool audit_watch_should_send_event(struct fsnotify_group *group, struct inode *inode, + struct fsnotify_mark *inode_mark, + struct fsnotify_mark *vfsmount_mark, + __u32 mask, void *data, int data_type) +{ + return true; +} + +/* Update watch data in audit rules based on fsnotify events. */ +static int audit_watch_handle_event(struct fsnotify_group *group, + struct fsnotify_mark *inode_mark, + struct fsnotify_mark *vfsmount_mark, + struct fsnotify_event *event) { + struct inode *inode; + __u32 mask = event->mask; + const char *dname = event->file_name; struct audit_parent *parent; - parent = container_of(i_watch, struct audit_parent, wdata); + parent = container_of(inode_mark, struct audit_parent, mark); - 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)) + BUG_ON(group != audit_watch_group); + + switch (event->data_type) { + case (FSNOTIFY_EVENT_PATH): + inode = event->path.dentry->d_inode; + break; + case (FSNOTIFY_EVENT_INODE): + inode = event->inode; + break; + default: + BUG(); + inode = NULL; + break; + }; + + if (mask & (FS_CREATE|FS_MOVED_TO) && inode) + audit_update_watch(parent, dname, inode->i_sb->s_dev, inode->i_ino, 0); + else if (mask & (FS_DELETE|FS_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) { + else if (mask & (FS_DELETE_SELF|FS_UNMOUNT|FS_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); + + return 0; } -static const struct inotify_operations audit_inotify_ops = { - .handle_event = audit_handle_ievent, - .destroy_watch = audit_free_parent, +static const struct fsnotify_ops audit_watch_fsnotify_ops = { + .should_send_event = audit_watch_should_send_event, + .handle_event = audit_watch_handle_event, + .free_group_priv = NULL, + .freeing_mark = NULL, + .free_event_priv = NULL, }; 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"); + audit_watch_group = fsnotify_alloc_group(&audit_watch_fsnotify_ops); + if (IS_ERR(audit_watch_group)) { + audit_watch_group = NULL; + audit_panic("cannot create audit fsnotify group"); + } return 0; } -subsys_initcall(audit_watch_init); +device_initcall(audit_watch_init); diff --git a/kernel/auditfilter.c b/kernel/auditfilter.c index ce08041f578..add2819af71 100644 --- a/kernel/auditfilter.c +++ b/kernel/auditfilter.c @@ -71,6 +71,7 @@ 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 (erule->watch) audit_put_watch(erule->watch); @@ -746,8 +747,7 @@ 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. */ -struct audit_entry *audit_dupe_rule(struct audit_krule *old, - struct audit_watch *watch) +struct audit_entry *audit_dupe_rule(struct audit_krule *old) { u32 fcount = old->field_count; struct audit_entry *entry; @@ -769,8 +769,8 @@ struct audit_entry *audit_dupe_rule(struct audit_krule *old, new->prio = old->prio; new->buflen = old->buflen; new->inode_f = old->inode_f; - new->watch = NULL; new->field_count = old->field_count; + /* * note that we are OK with not refcounting here; audit_match_tree() * never dereferences tree and we can't get false positives there @@ -811,9 +811,9 @@ struct audit_entry *audit_dupe_rule(struct audit_krule *old, } } - if (watch) { - audit_get_watch(watch); - new->watch = watch; + if (old->watch) { + audit_get_watch(old->watch); + new->watch = old->watch; } return entry; @@ -866,7 +866,7 @@ static inline int audit_add_rule(struct audit_entry *entry) struct audit_watch *watch = entry->rule.watch; struct audit_tree *tree = entry->rule.tree; struct list_head *list; - int h, err; + int err; #ifdef CONFIG_AUDITSYSCALL int dont_count = 0; @@ -889,15 +889,11 @@ static inline int audit_add_rule(struct audit_entry *entry) if (watch) { /* audit_filter_mutex is dropped and re-taken during this call */ - err = audit_add_watch(&entry->rule); + err = audit_add_watch(&entry->rule, &list); if (err) { mutex_unlock(&audit_filter_mutex); goto error; } - /* 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) { err = audit_add_tree_rule(&entry->rule); @@ -949,7 +945,6 @@ static inline int audit_del_rule(struct audit_entry *entry) struct audit_watch *watch = entry->rule.watch; struct audit_tree *tree = entry->rule.tree; struct list_head *list; - LIST_HEAD(inotify_list); int ret = 0; #ifdef CONFIG_AUDITSYSCALL int dont_count = 0; @@ -969,7 +964,7 @@ static inline int audit_del_rule(struct audit_entry *entry) } if (e->rule.watch) - audit_remove_watch_rule(&e->rule, &inotify_list); + audit_remove_watch_rule(&e->rule); if (e->rule.tree) audit_remove_tree_rule(&e->rule); @@ -987,9 +982,6 @@ static inline int audit_del_rule(struct audit_entry *entry) #endif mutex_unlock(&audit_filter_mutex); - if (!list_empty(&inotify_list)) - audit_inotify_unregister(&inotify_list); - out: if (watch) audit_put_watch(watch); /* match initial get */ @@ -1260,6 +1252,18 @@ static int audit_filter_user_rules(struct netlink_skb_parms *cb, case AUDIT_LOGINUID: result = audit_comparator(cb->loginuid, f->op, f->val); break; + case AUDIT_SUBJ_USER: + case AUDIT_SUBJ_ROLE: + case AUDIT_SUBJ_TYPE: + case AUDIT_SUBJ_SEN: + case AUDIT_SUBJ_CLR: + if (f->lsm_rule) + result = security_audit_rule_match(cb->sid, + f->type, + f->op, + f->lsm_rule, + NULL); + break; } if (!result) @@ -1323,30 +1327,23 @@ static int update_lsm_rule(struct audit_krule *r) { struct audit_entry *entry = container_of(r, struct audit_entry, rule); struct audit_entry *nentry; - struct audit_watch *watch; - struct audit_tree *tree; int err = 0; if (!security_audit_rule_known(r)) return 0; - watch = r->watch; - tree = r->tree; - nentry = audit_dupe_rule(r, watch); + nentry = audit_dupe_rule(r); if (IS_ERR(nentry)) { /* save the first error encountered for the * return value */ err = PTR_ERR(nentry); audit_panic("error updating LSM filters"); - if (watch) + if (r->watch) list_del(&r->rlist); list_del_rcu(&entry->list); list_del(&r->list); } else { - if (watch) { - list_add(&nentry->rule.rlist, audit_watch_rules(watch)); - list_del(&r->rlist); - } else if (tree) + if (r->watch || r->tree) list_replace_init(&r->rlist, &nentry->rule.rlist); list_replace_rcu(&entry->list, &nentry->list); list_replace(&r->list, &nentry->rule.list); diff --git a/kernel/auditsc.c b/kernel/auditsc.c index 3828ad5fb8f..f49a0318c2e 100644 --- a/kernel/auditsc.c +++ b/kernel/auditsc.c @@ -65,7 +65,6 @@ #include <linux/binfmts.h> #include <linux/highmem.h> #include <linux/syscalls.h> -#include <linux/inotify.h> #include <linux/capability.h> #include <linux/fs_struct.h> @@ -242,6 +241,10 @@ struct audit_context { pid_t pid; struct audit_cap_data cap; } capset; + struct { + int fd; + int flags; + } mmap; }; int fds[2]; @@ -549,9 +552,8 @@ static int audit_filter_rules(struct task_struct *tsk, } break; case AUDIT_WATCH: - 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)); + if (name) + result = audit_watch_compare(rule->watch, name->ino, name->dev); break; case AUDIT_DIR: if (ctx) @@ -1307,6 +1309,10 @@ static void show_special(struct audit_context *context, int *call_panic) audit_log_cap(ab, "cap_pp", &context->capset.cap.permitted); audit_log_cap(ab, "cap_pe", &context->capset.cap.effective); break; } + case AUDIT_MMAP: { + audit_log_format(ab, "fd=%d flags=0x%x", context->mmap.fd, + context->mmap.flags); + break; } } audit_log_end(ab); } @@ -1726,7 +1732,7 @@ static inline void handle_one(const struct inode *inode) struct audit_tree_refs *p; struct audit_chunk *chunk; int count; - if (likely(list_empty(&inode->inotify_watches))) + if (likely(hlist_empty(&inode->i_fsnotify_marks))) return; context = current->audit_context; p = context->trees; @@ -1769,7 +1775,7 @@ retry: seq = read_seqbegin(&rename_lock); for(;;) { struct inode *inode = d->d_inode; - if (inode && unlikely(!list_empty(&inode->inotify_watches))) { + if (inode && unlikely(!hlist_empty(&inode->i_fsnotify_marks))) { struct audit_chunk *chunk; chunk = audit_tree_lookup(inode); if (chunk) { @@ -1837,13 +1843,8 @@ void __audit_getname(const char *name) context->names[context->name_count].ino = (unsigned long)-1; context->names[context->name_count].osid = 0; ++context->name_count; - if (!context->pwd.dentry) { - read_lock(¤t->fs->lock); - context->pwd = current->fs->pwd; - path_get(¤t->fs->pwd); - read_unlock(¤t->fs->lock); - } - + if (!context->pwd.dentry) + get_fs_pwd(current->fs, &context->pwd); } /* audit_putname - intercept a putname request @@ -2483,6 +2484,14 @@ void __audit_log_capset(pid_t pid, context->type = AUDIT_CAPSET; } +void __audit_mmap_fd(int fd, int flags) +{ + struct audit_context *context = current->audit_context; + context->mmap.fd = fd; + context->mmap.flags = flags; + context->type = AUDIT_MMAP; +} + /** * audit_core_dumps - record information about processes that end abnormally * @signr: signal value diff --git a/kernel/cgroup.c b/kernel/cgroup.c index 291775021b2..b24d7027b83 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -52,7 +52,6 @@ #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 */ @@ -138,7 +137,7 @@ struct css_id { * 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; + struct cgroup_subsys_state __rcu *css; /* * ID of this css. */ @@ -244,6 +243,11 @@ static int notify_on_release(const struct cgroup *cgrp) return test_bit(CGRP_NOTIFY_ON_RELEASE, &cgrp->flags); } +static int clone_children(const struct cgroup *cgrp) +{ + return test_bit(CGRP_CLONE_CHILDREN, &cgrp->flags); +} + /* * for_each_subsys() allows you to iterate on each subsystem attached to * an active hierarchy @@ -760,6 +764,7 @@ EXPORT_SYMBOL_GPL(cgroup_unlock); */ static int cgroup_mkdir(struct inode *dir, struct dentry *dentry, int mode); +static struct dentry *cgroup_lookup(struct inode *, struct dentry *, struct nameidata *); static int cgroup_rmdir(struct inode *unused_dir, struct dentry *dentry); static int cgroup_populate_dir(struct cgroup *cgrp); static const struct inode_operations cgroup_dir_inode_operations; @@ -778,6 +783,7 @@ static struct inode *cgroup_new_inode(mode_t mode, struct super_block *sb) struct inode *inode = new_inode(sb); if (inode) { + inode->i_ino = get_next_ino(); inode->i_mode = mode; inode->i_uid = current_fsuid(); inode->i_gid = current_fsgid(); @@ -855,6 +861,11 @@ static void cgroup_diput(struct dentry *dentry, struct inode *inode) iput(inode); } +static int cgroup_delete(const struct dentry *d) +{ + return 1; +} + static void remove_dir(struct dentry *d) { struct dentry *parent = dget(d->d_parent); @@ -869,25 +880,29 @@ static void cgroup_clear_directory(struct dentry *dentry) struct list_head *node; BUG_ON(!mutex_is_locked(&dentry->d_inode->i_mutex)); - spin_lock(&dcache_lock); + spin_lock(&dentry->d_lock); node = dentry->d_subdirs.next; while (node != &dentry->d_subdirs) { struct dentry *d = list_entry(node, struct dentry, d_u.d_child); + + spin_lock_nested(&d->d_lock, DENTRY_D_LOCK_NESTED); list_del_init(node); if (d->d_inode) { /* This should never be called on a cgroup * directory with child cgroups */ BUG_ON(d->d_inode->i_mode & S_IFDIR); - d = dget_locked(d); - spin_unlock(&dcache_lock); + dget_dlock(d); + spin_unlock(&d->d_lock); + spin_unlock(&dentry->d_lock); d_delete(d); simple_unlink(dentry->d_inode, d); dput(d); - spin_lock(&dcache_lock); - } + spin_lock(&dentry->d_lock); + } else + spin_unlock(&d->d_lock); node = dentry->d_subdirs.next; } - spin_unlock(&dcache_lock); + spin_unlock(&dentry->d_lock); } /* @@ -895,11 +910,16 @@ static void cgroup_clear_directory(struct dentry *dentry) */ static void cgroup_d_remove_dir(struct dentry *dentry) { + struct dentry *parent; + cgroup_clear_directory(dentry); - spin_lock(&dcache_lock); + parent = dentry->d_parent; + spin_lock(&parent->d_lock); + spin_lock_nested(&dentry->d_lock, DENTRY_D_LOCK_NESTED); list_del_init(&dentry->d_u.d_child); - spin_unlock(&dcache_lock); + spin_unlock(&dentry->d_lock); + spin_unlock(&parent->d_lock); remove_dir(dentry); } @@ -1040,6 +1060,8 @@ 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 (clone_children(&root->top_cgroup)) + seq_puts(seq, ",clone_children"); if (strlen(root->name)) seq_printf(seq, ",name=%s", root->name); mutex_unlock(&cgroup_mutex); @@ -1050,6 +1072,7 @@ struct cgroup_sb_opts { unsigned long subsys_bits; unsigned long flags; char *release_agent; + bool clone_children; char *name; /* User explicitly requested empty subsystem */ bool none; @@ -1066,7 +1089,8 @@ struct cgroup_sb_opts { */ static int parse_cgroupfs_options(char *data, struct cgroup_sb_opts *opts) { - char *token, *o = data ?: "all"; + char *token, *o = data; + bool all_ss = false, one_ss = false; unsigned long mask = (unsigned long)-1; int i; bool module_pin_failed = false; @@ -1082,30 +1106,37 @@ static int parse_cgroupfs_options(char *data, struct cgroup_sb_opts *opts) while ((token = strsep(&o, ",")) != NULL) { if (!*token) return -EINVAL; - if (!strcmp(token, "all")) { - /* Add all non-disabled subsystems */ - opts->subsys_bits = 0; - for (i = 0; i < CGROUP_SUBSYS_COUNT; i++) { - struct cgroup_subsys *ss = subsys[i]; - if (ss == NULL) - continue; - if (!ss->disabled) - opts->subsys_bits |= 1ul << i; - } - } else if (!strcmp(token, "none")) { + if (!strcmp(token, "none")) { /* Explicitly have no subsystems */ opts->none = true; - } else if (!strcmp(token, "noprefix")) { + continue; + } + if (!strcmp(token, "all")) { + /* Mutually exclusive option 'all' + subsystem name */ + if (one_ss) + return -EINVAL; + all_ss = true; + continue; + } + if (!strcmp(token, "noprefix")) { set_bit(ROOT_NOPREFIX, &opts->flags); - } else if (!strncmp(token, "release_agent=", 14)) { + continue; + } + if (!strcmp(token, "clone_children")) { + opts->clone_children = true; + continue; + } + if (!strncmp(token, "release_agent=", 14)) { /* Specifying two release agents is forbidden */ if (opts->release_agent) return -EINVAL; opts->release_agent = - kstrndup(token + 14, PATH_MAX, GFP_KERNEL); + kstrndup(token + 14, PATH_MAX - 1, GFP_KERNEL); if (!opts->release_agent) return -ENOMEM; - } else if (!strncmp(token, "name=", 5)) { + continue; + } + if (!strncmp(token, "name=", 5)) { const char *name = token + 5; /* Can't specify an empty name */ if (!strlen(name)) @@ -1123,24 +1154,48 @@ static int parse_cgroupfs_options(char *data, struct cgroup_sb_opts *opts) if (opts->name) return -EINVAL; opts->name = kstrndup(name, - MAX_CGROUP_ROOT_NAMELEN, + MAX_CGROUP_ROOT_NAMELEN - 1, GFP_KERNEL); if (!opts->name) return -ENOMEM; - } else { - struct cgroup_subsys *ss; - for (i = 0; i < CGROUP_SUBSYS_COUNT; i++) { - ss = subsys[i]; - if (ss == NULL) - continue; - if (!strcmp(token, ss->name)) { - if (!ss->disabled) - set_bit(i, &opts->subsys_bits); - break; - } - } - if (i == CGROUP_SUBSYS_COUNT) - return -ENOENT; + + continue; + } + + for (i = 0; i < CGROUP_SUBSYS_COUNT; i++) { + struct cgroup_subsys *ss = subsys[i]; + if (ss == NULL) + continue; + if (strcmp(token, ss->name)) + continue; + if (ss->disabled) + continue; + + /* Mutually exclusive option 'all' + subsystem name */ + if (all_ss) + return -EINVAL; + set_bit(i, &opts->subsys_bits); + one_ss = true; + + break; + } + if (i == CGROUP_SUBSYS_COUNT) + return -ENOENT; + } + + /* + * If the 'all' option was specified select all the subsystems, + * otherwise 'all, 'none' and a subsystem name options were not + * specified, let's default to 'all' + */ + if (all_ss || (!all_ss && !one_ss && !opts->none)) { + for (i = 0; i < CGROUP_SUBSYS_COUNT; i++) { + struct cgroup_subsys *ss = subsys[i]; + if (ss == NULL) + continue; + if (ss->disabled) + continue; + set_bit(i, &opts->subsys_bits); } } @@ -1222,7 +1277,6 @@ 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); @@ -1255,7 +1309,6 @@ static int cgroup_remount(struct super_block *sb, int *flags, char *data) kfree(opts.name); mutex_unlock(&cgroup_mutex); mutex_unlock(&cgrp->dentry->d_inode->i_mutex); - unlock_kernel(); return ret; } @@ -1357,6 +1410,8 @@ static struct cgroupfs_root *cgroup_root_from_opts(struct cgroup_sb_opts *opts) strcpy(root->release_agent_path, opts->release_agent); if (opts->name) strcpy(root->name, opts->name); + if (opts->clone_children) + set_bit(CGRP_CLONE_CHILDREN, &root->top_cgroup.flags); return root; } @@ -1400,6 +1455,11 @@ static int cgroup_set_super(struct super_block *sb, void *data) static int cgroup_get_rootdir(struct super_block *sb) { + static const struct dentry_operations cgroup_dops = { + .d_iput = cgroup_diput, + .d_delete = cgroup_delete, + }; + struct inode *inode = cgroup_new_inode(S_IFDIR | S_IRUGO | S_IXUGO | S_IWUSR, sb); struct dentry *dentry; @@ -1417,12 +1477,14 @@ static int cgroup_get_rootdir(struct super_block *sb) return -ENOMEM; } sb->s_root = dentry; + /* for everything else we want ->d_op set */ + sb->s_d_op = &cgroup_dops; return 0; } -static int cgroup_get_sb(struct file_system_type *fs_type, +static struct dentry *cgroup_mount(struct file_system_type *fs_type, int flags, const char *unused_dev_name, - void *data, struct vfsmount *mnt) + void *data) { struct cgroup_sb_opts opts; struct cgroupfs_root *root; @@ -1556,10 +1618,9 @@ static int cgroup_get_sb(struct file_system_type *fs_type, drop_parsed_module_refcounts(opts.subsys_bits); } - simple_set_mnt(mnt, sb); kfree(opts.release_agent); kfree(opts.name); - return 0; + return dget(sb->s_root); drop_new_super: deactivate_locked_super(sb); @@ -1568,8 +1629,7 @@ static int cgroup_get_sb(struct file_system_type *fs_type, out_err: kfree(opts.release_agent); kfree(opts.name); - - return ret; + return ERR_PTR(ret); } static void cgroup_kill_sb(struct super_block *sb) { @@ -1619,10 +1679,12 @@ static void cgroup_kill_sb(struct super_block *sb) { static struct file_system_type cgroup_fs_type = { .name = "cgroup", - .get_sb = cgroup_get_sb, + .mount = cgroup_mount, .kill_sb = cgroup_kill_sb, }; +static struct kobject *cgroup_kobj; + static inline struct cgroup *__d_cgrp(struct dentry *dentry) { return dentry->d_fsdata; @@ -1788,6 +1850,30 @@ out: return retval; } +/** + * cgroup_attach_task_all - attach task 'tsk' to all cgroups of task 'from' + * @from: attach to all cgroups of a given task + * @tsk: the task to be attached + */ +int cgroup_attach_task_all(struct task_struct *from, struct task_struct *tsk) +{ + struct cgroupfs_root *root; + int retval = 0; + + cgroup_lock(); + for_each_active_root(root) { + struct cgroup *from_cg = task_cgroup_from_root(from, root); + + retval = cgroup_attach_task(from_cg, tsk); + if (retval) + break; + } + cgroup_unlock(); + + return retval; +} +EXPORT_SYMBOL_GPL(cgroup_attach_task_all); + /* * Attach task with pid 'pid' to cgroup 'cgrp'. Call with cgroup_mutex * held. May take task_lock of task @@ -1857,6 +1943,8 @@ static int cgroup_release_agent_write(struct cgroup *cgrp, struct cftype *cft, const char *buffer) { BUILD_BUG_ON(sizeof(cgrp->root->release_agent_path) < PATH_MAX); + if (strlen(buffer) >= PATH_MAX) + return -EINVAL; if (!cgroup_lock_live_group(cgrp)) return -ENODEV; strcpy(cgrp->root->release_agent_path, buffer); @@ -2114,12 +2202,20 @@ static const struct file_operations cgroup_file_operations = { }; static const struct inode_operations cgroup_dir_inode_operations = { - .lookup = simple_lookup, + .lookup = cgroup_lookup, .mkdir = cgroup_mkdir, .rmdir = cgroup_rmdir, .rename = cgroup_rename, }; +static struct dentry *cgroup_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd) +{ + if (dentry->d_name.len > NAME_MAX) + return ERR_PTR(-ENAMETOOLONG); + d_add(dentry, NULL); + return NULL; +} + /* * Check if a file is a control file */ @@ -2133,10 +2229,6 @@ static inline struct cftype *__file_cft(struct file *file) static int cgroup_create_file(struct dentry *dentry, mode_t mode, struct super_block *sb) { - static const struct dentry_operations cgroup_dops = { - .d_iput = cgroup_diput, - }; - struct inode *inode; if (!dentry) @@ -2162,7 +2254,6 @@ static int cgroup_create_file(struct dentry *dentry, mode_t mode, inode->i_size = 0; inode->i_fop = &cgroup_file_operations; } - dentry->d_op = &cgroup_dops; d_instantiate(dentry, inode); dget(dentry); /* Extra count - pin the dentry in core */ return 0; @@ -2994,7 +3085,6 @@ static void cgroup_event_remove(struct work_struct *work) remove); struct cgroup *cgrp = event->cgrp; - /* TODO: check return code */ event->cft->unregister_event(cgrp, event->cft, event->eventfd); eventfd_ctx_put(event->eventfd); @@ -3151,6 +3241,23 @@ fail: return ret; } +static u64 cgroup_clone_children_read(struct cgroup *cgrp, + struct cftype *cft) +{ + return clone_children(cgrp); +} + +static int cgroup_clone_children_write(struct cgroup *cgrp, + struct cftype *cft, + u64 val) +{ + if (val) + set_bit(CGRP_CLONE_CHILDREN, &cgrp->flags); + else + clear_bit(CGRP_CLONE_CHILDREN, &cgrp->flags); + return 0; +} + /* * for the common functions, 'private' gives the type of file */ @@ -3181,6 +3288,11 @@ static struct cftype files[] = { .write_string = cgroup_write_event_control, .mode = S_IWUGO, }, + { + .name = "cgroup.clone_children", + .read_u64 = cgroup_clone_children_read, + .write_u64 = cgroup_clone_children_write, + }, }; static struct cftype cft_release_agent = { @@ -3310,6 +3422,9 @@ static long cgroup_create(struct cgroup *parent, struct dentry *dentry, if (notify_on_release(parent)) set_bit(CGRP_NOTIFY_ON_RELEASE, &cgrp->flags); + if (clone_children(parent)) + set_bit(CGRP_CLONE_CHILDREN, &cgrp->flags); + for_each_subsys(root, ss) { struct cgroup_subsys_state *css = ss->create(ss, cgrp); @@ -3324,6 +3439,8 @@ static long cgroup_create(struct cgroup *parent, struct dentry *dentry, goto err_destroy; } /* At error, ->destroy() callback has to free assigned ID. */ + if (clone_children(parent) && ss->post_clone) + ss->post_clone(ss, cgrp); } cgroup_lock_hierarchy(root); @@ -3546,9 +3663,7 @@ again: list_del(&cgrp->sibling); cgroup_unlock_hierarchy(cgrp->root); - spin_lock(&cgrp->dentry->d_lock); d = dget(cgrp->dentry); - spin_unlock(&d->d_lock); cgroup_d_remove_dir(d); dput(d); @@ -3872,9 +3987,18 @@ int __init cgroup_init(void) hhead = css_set_hash(init_css_set.subsys); hlist_add_head(&init_css_set.hlist, hhead); BUG_ON(!init_root_id(&rootnode)); + + cgroup_kobj = kobject_create_and_add("cgroup", fs_kobj); + if (!cgroup_kobj) { + err = -ENOMEM; + goto out; + } + err = register_filesystem(&cgroup_fs_type); - if (err < 0) + if (err < 0) { + kobject_put(cgroup_kobj); goto out; + } proc_create("cgroups", 0, NULL, &proc_cgroupstats_operations); @@ -4599,7 +4723,7 @@ static int alloc_css_id(struct cgroup_subsys *ss, struct cgroup *parent, parent_css = parent->subsys[subsys_id]; child_css = child->subsys[subsys_id]; parent_id = parent_css->id; - depth = parent_id->depth; + depth = parent_id->depth + 1; child_id = get_new_cssid(ss, depth); if (IS_ERR(child_id)) diff --git a/kernel/cgroup_freezer.c b/kernel/cgroup_freezer.c index ce71ed53e88..e7bebb7c6c3 100644 --- a/kernel/cgroup_freezer.c +++ b/kernel/cgroup_freezer.c @@ -48,20 +48,19 @@ static inline struct freezer *task_freezer(struct task_struct *task) struct freezer, css); } -int cgroup_freezing_or_frozen(struct task_struct *task) +static inline int __cgroup_freezing_or_frozen(struct task_struct *task) { - struct freezer *freezer; - enum freezer_state state; + enum freezer_state state = task_freezer(task)->state; + return (state == CGROUP_FREEZING) || (state == CGROUP_FROZEN); +} +int cgroup_freezing_or_frozen(struct task_struct *task) +{ + int result; task_lock(task); - freezer = task_freezer(task); - if (!freezer->css.cgroup->parent) - state = CGROUP_THAWED; /* root cgroup can't be frozen */ - else - state = freezer->state; + result = __cgroup_freezing_or_frozen(task); task_unlock(task); - - return (state == CGROUP_FREEZING) || (state == CGROUP_FROZEN); + return result; } /* @@ -154,13 +153,6 @@ static void freezer_destroy(struct cgroup_subsys *ss, kfree(cgroup_freezer(cgroup)); } -/* Task is frozen or will freeze immediately when next it gets woken */ -static bool is_task_frozen_enough(struct task_struct *task) -{ - return frozen(task) || - (task_is_stopped_or_traced(task) && freezing(task)); -} - /* * The call to cgroup_lock() in the freezer.state write method prevents * a write to that file racing against an attach, and hence the @@ -174,24 +166,25 @@ static int freezer_can_attach(struct cgroup_subsys *ss, /* * Anything frozen can't move or be moved to/from. - * - * Since orig_freezer->state == FROZEN means that @task has been - * frozen, so it's sufficient to check the latter condition. */ - if (is_task_frozen_enough(task)) + freezer = cgroup_freezer(new_cgroup); + if (freezer->state != CGROUP_THAWED) return -EBUSY; - freezer = cgroup_freezer(new_cgroup); - if (freezer->state == CGROUP_FROZEN) + rcu_read_lock(); + if (__cgroup_freezing_or_frozen(task)) { + rcu_read_unlock(); return -EBUSY; + } + rcu_read_unlock(); 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)) { + if (__cgroup_freezing_or_frozen(c)) { rcu_read_unlock(); return -EBUSY; } @@ -236,31 +229,30 @@ static void freezer_fork(struct cgroup_subsys *ss, struct task_struct *task) /* * caller must hold freezer->lock */ -static void update_freezer_state(struct cgroup *cgroup, +static void update_if_frozen(struct cgroup *cgroup, struct freezer *freezer) { struct cgroup_iter it; struct task_struct *task; unsigned int nfrozen = 0, ntotal = 0; + enum freezer_state old_state = freezer->state; cgroup_iter_start(cgroup, &it); while ((task = cgroup_iter_next(cgroup, &it))) { ntotal++; - if (is_task_frozen_enough(task)) + if (frozen(task)) nfrozen++; } - /* - * Transition to FROZEN when no new tasks can be added ensures - * that we never exist in the FROZEN state while there are unfrozen - * tasks. - */ - if (nfrozen == ntotal) - freezer->state = CGROUP_FROZEN; - else if (nfrozen > 0) - freezer->state = CGROUP_FREEZING; - else - freezer->state = CGROUP_THAWED; + if (old_state == CGROUP_THAWED) { + BUG_ON(nfrozen > 0); + } else if (old_state == CGROUP_FREEZING) { + if (nfrozen == ntotal) + freezer->state = CGROUP_FROZEN; + } else { /* old_state == CGROUP_FROZEN */ + BUG_ON(nfrozen != ntotal); + } + cgroup_iter_end(cgroup, &it); } @@ -279,7 +271,7 @@ static int freezer_read(struct cgroup *cgroup, struct cftype *cft, if (state == CGROUP_FREEZING) { /* We change from FREEZING to FROZEN lazily if the cgroup was * only partially frozen when we exitted write. */ - update_freezer_state(cgroup, freezer); + update_if_frozen(cgroup, freezer); state = freezer->state; } spin_unlock_irq(&freezer->lock); @@ -301,7 +293,7 @@ static int try_to_freeze_cgroup(struct cgroup *cgroup, struct freezer *freezer) while ((task = cgroup_iter_next(cgroup, &it))) { if (!freeze_task(task, true)) continue; - if (is_task_frozen_enough(task)) + if (frozen(task)) continue; if (!freezing(task) && !freezer_should_skip(task)) num_cant_freeze_now++; @@ -335,7 +327,7 @@ static int freezer_change_state(struct cgroup *cgroup, spin_lock_irq(&freezer->lock); - update_freezer_state(cgroup, freezer); + update_if_frozen(cgroup, freezer); if (goal_state == freezer->state) goto out; diff --git a/kernel/compat.c b/kernel/compat.c index 5adab05a317..c9e2ec0b34a 100644 --- a/kernel/compat.c +++ b/kernel/compat.c @@ -279,11 +279,6 @@ asmlinkage long compat_sys_setrlimit(unsigned int resource, struct compat_rlimit __user *rlim) { struct rlimit r; - int ret; - mm_segment_t old_fs = get_fs (); - - if (resource >= RLIM_NLIMITS) - return -EINVAL; if (!access_ok(VERIFY_READ, rlim, sizeof(*rlim)) || __get_user(r.rlim_cur, &rlim->rlim_cur) || @@ -294,10 +289,7 @@ asmlinkage long compat_sys_setrlimit(unsigned int resource, r.rlim_cur = RLIM_INFINITY; if (r.rlim_max == COMPAT_RLIM_INFINITY) r.rlim_max = RLIM_INFINITY; - set_fs(KERNEL_DS); - ret = sys_setrlimit(resource, (struct rlimit __user *) &r); - set_fs(old_fs); - return ret; + return do_prlimit(current, resource, &r, NULL); } #ifdef COMPAT_RLIM_OLD_INFINITY @@ -329,16 +321,13 @@ asmlinkage long compat_sys_old_getrlimit(unsigned int resource, #endif -asmlinkage long compat_sys_getrlimit (unsigned int resource, +asmlinkage long compat_sys_getrlimit(unsigned int resource, struct compat_rlimit __user *rlim) { struct rlimit r; int ret; - mm_segment_t old_fs = get_fs(); - set_fs(KERNEL_DS); - ret = sys_getrlimit(resource, (struct rlimit __user *) &r); - set_fs(old_fs); + ret = do_prlimit(current, resource, NULL, &r); if (!ret) { if (r.rlim_cur > COMPAT_RLIM_INFINITY) r.rlim_cur = COMPAT_RLIM_INFINITY; @@ -1137,3 +1126,24 @@ compat_sys_sysinfo(struct compat_sysinfo __user *info) return 0; } + +/* + * Allocate user-space memory for the duration of a single system call, + * in order to marshall parameters inside a compat thunk. + */ +void __user *compat_alloc_user_space(unsigned long len) +{ + void __user *ptr; + + /* If len would occupy more than half of the entire compat space... */ + if (unlikely(len > (((compat_uptr_t)~0) >> 1))) + return NULL; + + ptr = arch_compat_alloc_user_space(len); + + if (unlikely(!access_ok(VERIFY_WRITE, ptr, len))) + return NULL; + + return ptr; +} +EXPORT_SYMBOL_GPL(compat_alloc_user_space); diff --git a/kernel/configs.c b/kernel/configs.c index abaee684ecb..b4066b44a99 100644 --- a/kernel/configs.c +++ b/kernel/configs.c @@ -66,6 +66,7 @@ ikconfig_read_current(struct file *file, char __user *buf, static const struct file_operations ikconfig_file_ops = { .owner = THIS_MODULE, .read = ikconfig_read_current, + .llseek = default_llseek, }; static int __init ikconfig_init(void) diff --git a/kernel/cpu.c b/kernel/cpu.c index 54577757477..156cc555614 100644 --- a/kernel/cpu.c +++ b/kernel/cpu.c @@ -20,13 +20,29 @@ /* Serializes the updates to cpu_online_mask, cpu_present_mask */ static DEFINE_MUTEX(cpu_add_remove_lock); -static __cpuinitdata RAW_NOTIFIER_HEAD(cpu_chain); +/* + * The following two API's must be used when attempting + * to serialize the updates to cpu_online_mask, cpu_present_mask. + */ +void cpu_maps_update_begin(void) +{ + mutex_lock(&cpu_add_remove_lock); +} + +void cpu_maps_update_done(void) +{ + mutex_unlock(&cpu_add_remove_lock); +} + +static RAW_NOTIFIER_HEAD(cpu_chain); /* If set, cpu_up and cpu_down will return -EBUSY and do nothing. * Should always be manipulated under cpu_add_remove_lock */ static int cpu_hotplug_disabled; +#ifdef CONFIG_HOTPLUG_CPU + static struct { struct task_struct *active_writer; struct mutex lock; /* Synchronizes accesses to refcount, */ @@ -41,8 +57,6 @@ static struct { .refcount = 0, }; -#ifdef CONFIG_HOTPLUG_CPU - void get_online_cpus(void) { might_sleep(); @@ -67,22 +81,6 @@ void put_online_cpus(void) } EXPORT_SYMBOL_GPL(put_online_cpus); -#endif /* CONFIG_HOTPLUG_CPU */ - -/* - * The following two API's must be used when attempting - * to serialize the updates to cpu_online_mask, cpu_present_mask. - */ -void cpu_maps_update_begin(void) -{ - mutex_lock(&cpu_add_remove_lock); -} - -void cpu_maps_update_done(void) -{ - mutex_unlock(&cpu_add_remove_lock); -} - /* * This ensures that the hotplug operation can begin only when the * refcount goes to zero. @@ -124,6 +122,12 @@ static void cpu_hotplug_done(void) cpu_hotplug.active_writer = NULL; mutex_unlock(&cpu_hotplug.lock); } + +#else /* #if CONFIG_HOTPLUG_CPU */ +static void cpu_hotplug_begin(void) {} +static void cpu_hotplug_done(void) {} +#endif /* #esle #if CONFIG_HOTPLUG_CPU */ + /* Need to know about CPUs going up/down? */ int __ref register_cpu_notifier(struct notifier_block *nb) { @@ -134,8 +138,29 @@ int __ref register_cpu_notifier(struct notifier_block *nb) return ret; } +static int __cpu_notify(unsigned long val, void *v, int nr_to_call, + int *nr_calls) +{ + int ret; + + ret = __raw_notifier_call_chain(&cpu_chain, val, v, nr_to_call, + nr_calls); + + return notifier_to_errno(ret); +} + +static int cpu_notify(unsigned long val, void *v) +{ + return __cpu_notify(val, v, -1, NULL); +} + #ifdef CONFIG_HOTPLUG_CPU +static void cpu_notify_nofail(unsigned long val, void *v) +{ + BUG_ON(cpu_notify(val, v)); +} + EXPORT_SYMBOL(register_cpu_notifier); void __ref unregister_cpu_notifier(struct notifier_block *nb) @@ -164,7 +189,6 @@ static inline void check_for_tasks(int cpu) } struct take_cpu_down_param { - struct task_struct *caller; unsigned long mod; void *hcpu; }; @@ -173,7 +197,6 @@ struct take_cpu_down_param { static int __ref take_cpu_down(void *_param) { struct take_cpu_down_param *param = _param; - unsigned int cpu = (unsigned long)param->hcpu; int err; /* Ensure this CPU doesn't handle any more interrupts. */ @@ -181,14 +204,8 @@ static int __ref take_cpu_down(void *_param) if (err < 0) return err; - raw_notifier_call_chain(&cpu_chain, CPU_DYING | param->mod, - param->hcpu); + cpu_notify(CPU_DYING | param->mod, param->hcpu); - if (task_cpu(param->caller) == cpu) - move_task_off_dead_cpu(cpu, param->caller); - /* Force idle task to run as soon as we yield: it should - immediately notice cpu is offline and die quickly. */ - sched_idle_next(); return 0; } @@ -199,7 +216,6 @@ static int __ref _cpu_down(unsigned int cpu, int tasks_frozen) void *hcpu = (void *)(long)cpu; unsigned long mod = tasks_frozen ? CPU_TASKS_FROZEN : 0; struct take_cpu_down_param tcd_param = { - .caller = current, .mod = mod, .hcpu = hcpu, }; @@ -211,54 +227,46 @@ static int __ref _cpu_down(unsigned int cpu, int tasks_frozen) return -EINVAL; cpu_hotplug_begin(); - set_cpu_active(cpu, false); - err = __raw_notifier_call_chain(&cpu_chain, CPU_DOWN_PREPARE | mod, - hcpu, -1, &nr_calls); - if (err == NOTIFY_BAD) { - set_cpu_active(cpu, true); - + err = __cpu_notify(CPU_DOWN_PREPARE | mod, hcpu, -1, &nr_calls); + if (err) { nr_calls--; - __raw_notifier_call_chain(&cpu_chain, CPU_DOWN_FAILED | mod, - hcpu, nr_calls, NULL); + __cpu_notify(CPU_DOWN_FAILED | mod, hcpu, nr_calls, NULL); printk("%s: attempt to take down CPU %u failed\n", __func__, cpu); - err = -EINVAL; goto out_release; } err = __stop_machine(take_cpu_down, &tcd_param, cpumask_of(cpu)); if (err) { - set_cpu_active(cpu, true); /* CPU didn't die: tell everyone. Can't complain. */ - if (raw_notifier_call_chain(&cpu_chain, CPU_DOWN_FAILED | mod, - hcpu) == NOTIFY_BAD) - BUG(); + cpu_notify_nofail(CPU_DOWN_FAILED | mod, hcpu); goto out_release; } BUG_ON(cpu_online(cpu)); - /* Wait for it to sleep (leaving idle task). */ + /* + * The migration_call() CPU_DYING callback will have removed all + * runnable tasks from the cpu, there's only the idle task left now + * that the migration thread is done doing the stop_machine thing. + * + * Wait for the stop thread to go away. + */ while (!idle_cpu(cpu)) - yield(); + cpu_relax(); /* This actually kills the CPU. */ __cpu_die(cpu); /* CPU is completely dead: tell everyone. Too late to complain. */ - if (raw_notifier_call_chain(&cpu_chain, CPU_DEAD | mod, - hcpu) == NOTIFY_BAD) - BUG(); + cpu_notify_nofail(CPU_DEAD | mod, hcpu); check_for_tasks(cpu); out_release: cpu_hotplug_done(); - if (!err) { - if (raw_notifier_call_chain(&cpu_chain, CPU_POST_DEAD | mod, - hcpu) == NOTIFY_BAD) - BUG(); - } + if (!err) + cpu_notify_nofail(CPU_POST_DEAD | mod, hcpu); return err; } @@ -293,13 +301,11 @@ static int __cpuinit _cpu_up(unsigned int cpu, int tasks_frozen) return -EINVAL; cpu_hotplug_begin(); - ret = __raw_notifier_call_chain(&cpu_chain, CPU_UP_PREPARE | mod, hcpu, - -1, &nr_calls); - if (ret == NOTIFY_BAD) { + ret = __cpu_notify(CPU_UP_PREPARE | mod, hcpu, -1, &nr_calls); + if (ret) { nr_calls--; printk("%s: attempt to bring up CPU %u failed\n", __func__, cpu); - ret = -EINVAL; goto out_notify; } @@ -309,15 +315,12 @@ static int __cpuinit _cpu_up(unsigned int cpu, int tasks_frozen) goto out_notify; BUG_ON(!cpu_online(cpu)); - set_cpu_active(cpu, true); - /* Now call notifier in preparation. */ - raw_notifier_call_chain(&cpu_chain, CPU_ONLINE | mod, hcpu); + cpu_notify(CPU_ONLINE | mod, hcpu); out_notify: if (ret != 0) - __raw_notifier_call_chain(&cpu_chain, - CPU_UP_CANCELED | mod, hcpu, nr_calls, NULL); + __cpu_notify(CPU_UP_CANCELED | mod, hcpu, nr_calls, NULL); cpu_hotplug_done(); return ret; @@ -326,6 +329,12 @@ out_notify: int __cpuinit cpu_up(unsigned int cpu) { int err = 0; + +#ifdef CONFIG_MEMORY_HOTPLUG + int nid; + pg_data_t *pgdat; +#endif + if (!cpu_possible(cpu)) { printk(KERN_ERR "can't online cpu %d because it is not " "configured as may-hotadd at boot time\n", cpu); @@ -336,6 +345,28 @@ int __cpuinit cpu_up(unsigned int cpu) return -EINVAL; } +#ifdef CONFIG_MEMORY_HOTPLUG + nid = cpu_to_node(cpu); + if (!node_online(nid)) { + err = mem_online_node(nid); + if (err) + return err; + } + + pgdat = NODE_DATA(nid); + if (!pgdat) { + printk(KERN_ERR + "Can't online cpu %d due to NULL pgdat\n", cpu); + return -ENOMEM; + } + + if (pgdat->node_zonelists->_zonerefs->zone == NULL) { + mutex_lock(&zonelists_mutex); + build_all_zonelists(NULL); + mutex_unlock(&zonelists_mutex); + } +#endif + cpu_maps_update_begin(); if (cpu_hotplug_disabled) { @@ -353,9 +384,17 @@ out: #ifdef CONFIG_PM_SLEEP_SMP static cpumask_var_t frozen_cpus; +void __weak arch_disable_nonboot_cpus_begin(void) +{ +} + +void __weak arch_disable_nonboot_cpus_end(void) +{ +} + int disable_nonboot_cpus(void) { - int cpu, first_cpu, error; + int cpu, first_cpu, error = 0; cpu_maps_update_begin(); first_cpu = cpumask_first(cpu_online_mask); @@ -364,6 +403,7 @@ int disable_nonboot_cpus(void) * with the userspace trying to use the CPU hotplug at the same time */ cpumask_clear(frozen_cpus); + arch_disable_nonboot_cpus_begin(); printk("Disabling non-boot CPUs ...\n"); for_each_online_cpu(cpu) { @@ -379,6 +419,8 @@ int disable_nonboot_cpus(void) } } + arch_disable_nonboot_cpus_end(); + if (!error) { BUG_ON(num_online_cpus() > 1); /* Make sure the CPUs won't be enabled by someone else */ @@ -453,7 +495,7 @@ void __cpuinit notify_cpu_starting(unsigned int cpu) if (frozen_cpus != NULL && cpumask_test_cpu(cpu, frozen_cpus)) val = CPU_STARTING_FROZEN; #endif /* CONFIG_PM_SLEEP_SMP */ - raw_notifier_call_chain(&cpu_chain, val, (void *)(long)cpu); + cpu_notify(val, (void *)(long)cpu); } #endif /* CONFIG_SMP */ diff --git a/kernel/cpuset.c b/kernel/cpuset.c index 9a50c5f6e72..4349935c2ad 100644 --- a/kernel/cpuset.c +++ b/kernel/cpuset.c @@ -105,7 +105,7 @@ struct cpuset { /* for custom sched domain */ int relax_domain_level; - /* used for walking a cpuset heirarchy */ + /* used for walking a cpuset hierarchy */ struct list_head stack_list; }; @@ -231,18 +231,17 @@ static DEFINE_SPINLOCK(cpuset_buffer_lock); * users. If someone tries to mount the "cpuset" filesystem, we * silently switch it to mount "cgroup" instead */ -static int cpuset_get_sb(struct file_system_type *fs_type, - int flags, const char *unused_dev_name, - void *data, struct vfsmount *mnt) +static struct dentry *cpuset_mount(struct file_system_type *fs_type, + int flags, const char *unused_dev_name, void *data) { struct file_system_type *cgroup_fs = get_fs_type("cgroup"); - int ret = -ENODEV; + struct dentry *ret = ERR_PTR(-ENODEV); if (cgroup_fs) { char mountopts[] = "cpuset,noprefix," "release_agent=/sbin/cpuset_release_agent"; - ret = cgroup_fs->get_sb(cgroup_fs, flags, - unused_dev_name, mountopts, mnt); + ret = cgroup_fs->mount(cgroup_fs, flags, + unused_dev_name, mountopts); put_filesystem(cgroup_fs); } return ret; @@ -250,7 +249,7 @@ static int cpuset_get_sb(struct file_system_type *fs_type, static struct file_system_type cpuset_fs_type = { .name = "cpuset", - .get_sb = cpuset_get_sb, + .mount = cpuset_mount, }; /* @@ -946,16 +945,62 @@ static void cpuset_migrate_mm(struct mm_struct *mm, const nodemask_t *from, * 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) { +repeat: + /* + * Allow tasks that have access to memory reserves because they have + * been OOM killed to get memory anywhere. + */ + if (unlikely(test_thread_flag(TIF_MEMDIE))) + return; + if (current->flags & PF_EXITING) /* Let dying task have memory */ + return; + + task_lock(tsk); nodes_or(tsk->mems_allowed, tsk->mems_allowed, *newmems); - mpol_rebind_task(tsk, &tsk->mems_allowed); - mpol_rebind_task(tsk, newmems); + mpol_rebind_task(tsk, newmems, MPOL_REBIND_STEP1); + + + /* + * ensure checking ->mems_allowed_change_disable after setting all new + * allowed nodes. + * + * the read-side task can see an nodemask with new allowed nodes and + * old allowed nodes. and if it allocates page when cpuset clears newly + * disallowed ones continuous, it can see the new allowed bits. + * + * And if setting all new allowed nodes is after the checking, setting + * all new allowed nodes and clearing newly disallowed ones will be done + * continuous, and the read-side task may find no node to alloc page. + */ + smp_mb(); + + /* + * Allocation of memory is very fast, we needn't sleep when waiting + * for the read-side. + */ + while (ACCESS_ONCE(tsk->mems_allowed_change_disable)) { + task_unlock(tsk); + if (!task_curr(tsk)) + yield(); + goto repeat; + } + + /* + * ensure checking ->mems_allowed_change_disable before clearing all new + * disallowed nodes. + * + * if clearing newly disallowed bits before the checking, the read-side + * task may find no node to alloc page. + */ + smp_mb(); + + mpol_rebind_task(tsk, newmems, MPOL_REBIND_STEP2); tsk->mems_allowed = *newmems; + task_unlock(tsk); } /* @@ -978,9 +1023,7 @@ static void cpuset_change_nodemask(struct task_struct *p, cs = cgroup_cs(scan->cg); guarantee_online_mems(cs, newmems); - task_lock(p); cpuset_change_task_nodemask(p, newmems); - task_unlock(p); NODEMASK_FREE(newmems); @@ -1353,7 +1396,7 @@ static int cpuset_can_attach(struct cgroup_subsys *ss, struct cgroup *cont, if (tsk->flags & PF_THREAD_BOUND) return -EINVAL; - ret = security_task_setscheduler(tsk, 0, NULL); + ret = security_task_setscheduler(tsk); if (ret) return ret; if (threadgroup) { @@ -1361,7 +1404,7 @@ static int cpuset_can_attach(struct cgroup_subsys *ss, struct cgroup *cont, rcu_read_lock(); list_for_each_entry_rcu(c, &tsk->thread_group, thread_group) { - ret = security_task_setscheduler(c, 0, NULL); + ret = security_task_setscheduler(c); if (ret) { rcu_read_unlock(); return ret; @@ -1383,9 +1426,7 @@ static void cpuset_attach_task(struct task_struct *tsk, nodemask_t *to, 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); } @@ -2071,31 +2112,17 @@ static void scan_for_empty_cpusets(struct cpuset *root) * but making no active use of cpusets. * * This routine ensures that top_cpuset.cpus_allowed tracks - * cpu_online_map on each CPU hotplug (cpuhp) event. + * cpu_active_mask on each CPU hotplug (cpuhp) event. * * Called within get_online_cpus(). Needs to call cgroup_lock() * before calling generate_sched_domains(). */ -static int cpuset_track_online_cpus(struct notifier_block *unused_nb, - unsigned long phase, void *unused_cpu) +void cpuset_update_active_cpus(void) { struct sched_domain_attr *attr; cpumask_var_t *doms; int ndoms; - switch (phase) { - case CPU_ONLINE: - case CPU_ONLINE_FROZEN: - case CPU_DOWN_PREPARE: - case CPU_DOWN_PREPARE_FROZEN: - case CPU_DOWN_FAILED: - case CPU_DOWN_FAILED_FROZEN: - break; - - default: - return NOTIFY_DONE; - } - cgroup_lock(); mutex_lock(&callback_mutex); cpumask_copy(top_cpuset.cpus_allowed, cpu_active_mask); @@ -2106,8 +2133,6 @@ static int cpuset_track_online_cpus(struct notifier_block *unused_nb, /* Have scheduler rebuild the domains */ partition_sched_domains(ndoms, doms, attr); - - return NOTIFY_OK; } #ifdef CONFIG_MEMORY_HOTPLUG @@ -2161,7 +2186,6 @@ void __init cpuset_init_smp(void) cpumask_copy(top_cpuset.cpus_allowed, cpu_active_mask); top_cpuset.mems_allowed = node_states[N_HIGH_MEMORY]; - hotcpu_notifier(cpuset_track_online_cpus, 0); hotplug_memory_notifier(cpuset_track_online_nodes, 10); cpuset_wq = create_singlethread_workqueue("cpuset"); @@ -2427,7 +2451,8 @@ void cpuset_unlock(void) } /** - * cpuset_mem_spread_node() - On which node to begin search for a page + * cpuset_mem_spread_node() - On which node to begin search for a file page + * cpuset_slab_spread_node() - On which node to begin search for a slab page * * If a task is marked PF_SPREAD_PAGE or PF_SPREAD_SLAB (as for * tasks in a cpuset with is_spread_page or is_spread_slab set), @@ -2452,16 +2477,27 @@ void cpuset_unlock(void) * See kmem_cache_alloc_node(). */ -int cpuset_mem_spread_node(void) +static int cpuset_spread_node(int *rotor) { int node; - node = next_node(current->cpuset_mem_spread_rotor, current->mems_allowed); + node = next_node(*rotor, current->mems_allowed); if (node == MAX_NUMNODES) node = first_node(current->mems_allowed); - current->cpuset_mem_spread_rotor = node; + *rotor = node; return node; } + +int cpuset_mem_spread_node(void) +{ + return cpuset_spread_node(¤t->cpuset_mem_spread_rotor); +} + +int cpuset_slab_spread_node(void) +{ + return cpuset_spread_node(¤t->cpuset_slab_spread_rotor); +} + EXPORT_SYMBOL_GPL(cpuset_mem_spread_node); /** diff --git a/kernel/cred.c b/kernel/cred.c index 2c24870c55d..6a1aa004e37 100644 --- a/kernel/cred.c +++ b/kernel/cred.c @@ -22,10 +22,6 @@ #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 @@ -209,6 +205,31 @@ void exit_creds(struct task_struct *tsk) } } +/** + * get_task_cred - Get another task's objective credentials + * @task: The task to query + * + * Get the objective credentials of a task, pinning them so that they can't go + * away. Accessing a task's credentials directly is not permitted. + * + * The caller must also make sure task doesn't get deleted, either by holding a + * ref on task or by holding tasklist_lock to prevent it from being unlinked. + */ +const struct cred *get_task_cred(struct task_struct *task) +{ + const struct cred *cred; + + rcu_read_lock(); + + do { + cred = __task_cred((task)); + BUG_ON(!cred); + } while (!atomic_inc_not_zero(&((struct cred *)cred)->usage)); + + rcu_read_unlock(); + return cred; +} + /* * Allocate blank credentials, such that the credentials can be filled in at a * later date without risk of ENOMEM. @@ -304,7 +325,7 @@ EXPORT_SYMBOL(prepare_creds); /* * Prepare credentials for current to perform an execve() - * - The caller must hold current->cred_guard_mutex + * - The caller must hold ->cred_guard_mutex */ struct cred *prepare_exec_creds(void) { @@ -347,66 +368,6 @@ struct cred *prepare_exec_creds(void) } /* - * prepare new credentials for the usermode helper dispatcher - */ -struct cred *prepare_usermodehelper_creds(void) -{ -#ifdef CONFIG_KEYS - struct thread_group_cred *tgcred = NULL; -#endif - struct cred *new; - -#ifdef CONFIG_KEYS - tgcred = kzalloc(sizeof(*new->tgcred), GFP_ATOMIC); - if (!tgcred) - return NULL; -#endif - - new = kmem_cache_alloc(cred_jar, GFP_ATOMIC); - if (!new) - goto free_tgcred; - - 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); - -#ifdef CONFIG_KEYS - new->thread_keyring = NULL; - new->request_key_auth = NULL; - new->jit_keyring = KEY_REQKEY_DEFL_DEFAULT; - - atomic_set(&tgcred->usage, 1); - spin_lock_init(&tgcred->lock); - new->tgcred = tgcred; -#endif - -#ifdef CONFIG_SECURITY - new->security = NULL; -#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; - -error: - put_cred(new); - return NULL; - -free_tgcred: -#ifdef CONFIG_KEYS - kfree(tgcred); -#endif - return NULL; -} - -/* * Copy credentials for the new process created by fork() * * We share if we can, but under some circumstances we have to generate a new @@ -423,8 +384,6 @@ int copy_creds(struct task_struct *p, unsigned long clone_flags) struct cred *new; int ret; - mutex_init(&p->cred_guard_mutex); - if ( #ifdef CONFIG_KEYS !p->cred->thread_keyring && diff --git a/kernel/debug/debug_core.c b/kernel/debug/debug_core.c index 5cb7cd1de10..cefd4a11f6d 100644 --- a/kernel/debug/debug_core.c +++ b/kernel/debug/debug_core.c @@ -6,7 +6,7 @@ * Copyright (C) 2000-2001 VERITAS Software Corporation. * Copyright (C) 2002-2004 Timesys Corporation * Copyright (C) 2003-2004 Amit S. Kale <amitkale@linsyssoft.com> - * Copyright (C) 2004 Pavel Machek <pavel@suse.cz> + * Copyright (C) 2004 Pavel Machek <pavel@ucw.cz> * Copyright (C) 2004-2006 Tom Rini <trini@kernel.crashing.org> * Copyright (C) 2004-2006 LinSysSoft Technologies Pvt. Ltd. * Copyright (C) 2005-2009 Wind River Systems, Inc. @@ -47,6 +47,7 @@ #include <linux/pid.h> #include <linux/smp.h> #include <linux/mm.h> +#include <linux/rcupdate.h> #include <asm/cacheflush.h> #include <asm/byteorder.h> @@ -109,13 +110,15 @@ static struct kgdb_bkpt kgdb_break[KGDB_MAX_BREAKPOINTS] = { */ atomic_t kgdb_active = ATOMIC_INIT(-1); EXPORT_SYMBOL_GPL(kgdb_active); +static DEFINE_RAW_SPINLOCK(dbg_master_lock); +static DEFINE_RAW_SPINLOCK(dbg_slave_lock); /* * We use NR_CPUs not PERCPU, in case kgdb is used to debug early * bootup code (which might not have percpu set up yet): */ -static atomic_t passive_cpu_wait[NR_CPUS]; -static atomic_t cpu_in_kgdb[NR_CPUS]; +static atomic_t masters_in_kgdb; +static atomic_t slaves_in_kgdb; static atomic_t kgdb_break_tasklet_var; atomic_t kgdb_setting_breakpoint; @@ -206,18 +209,6 @@ int __weak kgdb_skipexception(int exception, struct pt_regs *regs) return 0; } -/** - * kgdb_disable_hw_debug - Disable hardware debugging while we in kgdb. - * @regs: Current &struct pt_regs. - * - * This function will be called if the particular architecture must - * disable hardware debugging while it is processing gdb packets or - * handling exception. - */ -void __weak kgdb_disable_hw_debug(struct pt_regs *regs) -{ -} - /* * Some architectures need cache flushes when we set/clear a * breakpoint: @@ -457,26 +448,34 @@ static int kgdb_reenter_check(struct kgdb_state *ks) return 1; } -static void dbg_cpu_switch(int cpu, int next_cpu) +static void dbg_touch_watchdogs(void) { - /* Mark the cpu we are switching away from as a slave when it - * holds the kgdb_active token. This must be done so that the - * that all the cpus wait in for the debug core will not enter - * again as the master. */ - if (cpu == atomic_read(&kgdb_active)) { - kgdb_info[cpu].exception_state |= DCPU_IS_SLAVE; - kgdb_info[cpu].exception_state &= ~DCPU_WANT_MASTER; - } - kgdb_info[next_cpu].exception_state |= DCPU_NEXT_MASTER; + touch_softlockup_watchdog_sync(); + clocksource_touch_watchdog(); + rcu_cpu_stall_reset(); } -static int kgdb_cpu_enter(struct kgdb_state *ks, struct pt_regs *regs) +static int kgdb_cpu_enter(struct kgdb_state *ks, struct pt_regs *regs, + int exception_state) { unsigned long flags; int sstep_tries = 100; int error; - int i, cpu; + int cpu; int trace_on = 0; + int online_cpus = num_online_cpus(); + + kgdb_info[ks->cpu].enter_kgdb++; + kgdb_info[ks->cpu].exception_state |= exception_state; + + if (exception_state == DCPU_WANT_MASTER) + atomic_inc(&masters_in_kgdb); + else + atomic_inc(&slaves_in_kgdb); + + if (arch_kgdb_ops.disable_hw_break) + arch_kgdb_ops.disable_hw_break(regs); + acquirelock: /* * Interrupts will be restored by the 'trap return' code, except when @@ -489,14 +488,15 @@ acquirelock: kgdb_info[cpu].task = current; kgdb_info[cpu].ret_state = 0; kgdb_info[cpu].irq_depth = hardirq_count() >> HARDIRQ_SHIFT; - /* - * Make sure the above info reaches the primary CPU before - * our cpu_in_kgdb[] flag setting does: - */ - atomic_inc(&cpu_in_kgdb[cpu]); - if (exception_level == 1) + /* Make sure the above info reaches the primary CPU */ + smp_mb(); + + if (exception_level == 1) { + if (raw_spin_trylock(&dbg_master_lock)) + atomic_xchg(&kgdb_active, cpu); goto cpu_master_loop; + } /* * CPU will loop if it is a slave or request to become a kgdb @@ -508,10 +508,12 @@ cpu_loop: kgdb_info[cpu].exception_state &= ~DCPU_NEXT_MASTER; goto cpu_master_loop; } else if (kgdb_info[cpu].exception_state & DCPU_WANT_MASTER) { - if (atomic_cmpxchg(&kgdb_active, -1, cpu) == cpu) + if (raw_spin_trylock(&dbg_master_lock)) { + atomic_xchg(&kgdb_active, cpu); break; + } } else if (kgdb_info[cpu].exception_state & DCPU_IS_SLAVE) { - if (!atomic_read(&passive_cpu_wait[cpu])) + if (!raw_spin_is_locked(&dbg_slave_lock)) goto return_normal; } else { return_normal: @@ -522,9 +524,12 @@ return_normal: arch_kgdb_ops.correct_hw_break(); if (trace_on) tracing_on(); - atomic_dec(&cpu_in_kgdb[cpu]); - touch_softlockup_watchdog_sync(); - clocksource_touch_watchdog(); + kgdb_info[cpu].exception_state &= + ~(DCPU_WANT_MASTER | DCPU_IS_SLAVE); + kgdb_info[cpu].enter_kgdb--; + smp_mb__before_atomic_dec(); + atomic_dec(&slaves_in_kgdb); + dbg_touch_watchdogs(); local_irq_restore(flags); return 0; } @@ -541,8 +546,8 @@ return_normal: (kgdb_info[cpu].task && kgdb_info[cpu].task->pid != kgdb_sstep_pid) && --sstep_tries) { atomic_set(&kgdb_active, -1); - touch_softlockup_watchdog_sync(); - clocksource_touch_watchdog(); + raw_spin_unlock(&dbg_master_lock); + dbg_touch_watchdogs(); local_irq_restore(flags); goto acquirelock; @@ -563,16 +568,12 @@ return_normal: if (dbg_io_ops->pre_exception) dbg_io_ops->pre_exception(); - kgdb_disable_hw_debug(ks->linux_regs); - /* * Get the passive CPU lock which will hold all the non-primary * CPU in a spin state while the debugger is active */ - if (!kgdb_single_step) { - for (i = 0; i < NR_CPUS; i++) - atomic_inc(&passive_cpu_wait[i]); - } + if (!kgdb_single_step) + raw_spin_lock(&dbg_slave_lock); #ifdef CONFIG_SMP /* Signal the other CPUs to enter kgdb_wait() */ @@ -583,10 +584,9 @@ return_normal: /* * Wait for the other CPUs to be notified and be waiting for us: */ - for_each_online_cpu(i) { - while (kgdb_do_roundup && !atomic_read(&cpu_in_kgdb[i])) - cpu_relax(); - } + while (kgdb_do_roundup && (atomic_read(&masters_in_kgdb) + + atomic_read(&slaves_in_kgdb)) != online_cpus) + cpu_relax(); /* * At this point the primary processor is completely @@ -605,15 +605,18 @@ cpu_master_loop: if (dbg_kdb_mode) { kgdb_connected = 1; error = kdb_stub(ks); + if (error == -1) + continue; + kgdb_connected = 0; } else { error = gdb_serial_stub(ks); } if (error == DBG_PASS_EVENT) { dbg_kdb_mode = !dbg_kdb_mode; - kgdb_connected = 0; } else if (error == DBG_SWITCH_CPU_EVENT) { - dbg_cpu_switch(cpu, dbg_switch_cpu); + kgdb_info[dbg_switch_cpu].exception_state |= + DCPU_NEXT_MASTER; goto cpu_loop; } else { kgdb_info[cpu].ret_state = error; @@ -625,24 +628,11 @@ cpu_master_loop: if (dbg_io_ops->post_exception) dbg_io_ops->post_exception(); - atomic_dec(&cpu_in_kgdb[ks->cpu]); - if (!kgdb_single_step) { - for (i = NR_CPUS-1; i >= 0; i--) - atomic_dec(&passive_cpu_wait[i]); - /* - * Wait till all the CPUs have quit from the debugger, - * but allow a CPU that hit an exception and is - * waiting to become the master to remain in the debug - * core. - */ - for_each_online_cpu(i) { - while (kgdb_do_roundup && - atomic_read(&cpu_in_kgdb[i]) && - !(kgdb_info[i].exception_state & - DCPU_WANT_MASTER)) - cpu_relax(); - } + raw_spin_unlock(&dbg_slave_lock); + /* Wait till all the CPUs have quit from the debugger. */ + while (kgdb_do_roundup && atomic_read(&slaves_in_kgdb)) + cpu_relax(); } kgdb_restore: @@ -653,12 +643,20 @@ kgdb_restore: else kgdb_sstep_pid = 0; } + if (arch_kgdb_ops.correct_hw_break) + arch_kgdb_ops.correct_hw_break(); if (trace_on) tracing_on(); + + kgdb_info[cpu].exception_state &= + ~(DCPU_WANT_MASTER | DCPU_IS_SLAVE); + kgdb_info[cpu].enter_kgdb--; + smp_mb__before_atomic_dec(); + atomic_dec(&masters_in_kgdb); /* Free kgdb_active */ atomic_set(&kgdb_active, -1); - touch_softlockup_watchdog_sync(); - clocksource_touch_watchdog(); + raw_spin_unlock(&dbg_master_lock); + dbg_touch_watchdogs(); local_irq_restore(flags); return kgdb_info[cpu].ret_state; @@ -676,7 +674,6 @@ kgdb_handle_exception(int evector, int signo, int ecode, struct pt_regs *regs) { struct kgdb_state kgdb_var; struct kgdb_state *ks = &kgdb_var; - int ret; ks->cpu = raw_smp_processor_id(); ks->ex_vector = evector; @@ -687,11 +684,10 @@ kgdb_handle_exception(int evector, int signo, int ecode, struct pt_regs *regs) if (kgdb_reenter_check(ks)) return 0; /* Ouch, double exception ! */ - kgdb_info[ks->cpu].exception_state |= DCPU_WANT_MASTER; - ret = kgdb_cpu_enter(ks, regs); - kgdb_info[ks->cpu].exception_state &= ~(DCPU_WANT_MASTER | - DCPU_IS_SLAVE); - return ret; + if (kgdb_info[ks->cpu].enter_kgdb != 0) + return 0; + + return kgdb_cpu_enter(ks, regs, DCPU_WANT_MASTER); } int kgdb_nmicallback(int cpu, void *regs) @@ -704,12 +700,9 @@ int kgdb_nmicallback(int cpu, void *regs) ks->cpu = cpu; ks->linux_regs = regs; - if (!atomic_read(&cpu_in_kgdb[cpu]) && - atomic_read(&kgdb_active) != -1 && - atomic_read(&kgdb_active) != cpu) { - kgdb_info[cpu].exception_state |= DCPU_IS_SLAVE; - kgdb_cpu_enter(ks, regs); - kgdb_info[cpu].exception_state &= ~DCPU_IS_SLAVE; + if (kgdb_info[ks->cpu].enter_kgdb == 0 && + raw_spin_is_locked(&dbg_master_lock)) { + kgdb_cpu_enter(ks, regs, DCPU_IS_SLAVE); return 0; } #endif @@ -739,7 +732,7 @@ static struct console kgdbcons = { }; #ifdef CONFIG_MAGIC_SYSRQ -static void sysrq_handle_dbg(int key, struct tty_struct *tty) +static void sysrq_handle_dbg(int key) { if (!dbg_io_ops) { printk(KERN_CRIT "ERROR: No KGDB I/O module available\n"); diff --git a/kernel/debug/debug_core.h b/kernel/debug/debug_core.h index c5d753d80f6..3494c28a7e7 100644 --- a/kernel/debug/debug_core.h +++ b/kernel/debug/debug_core.h @@ -40,6 +40,7 @@ struct debuggerinfo_struct { int exception_state; int ret_state; int irq_depth; + int enter_kgdb; }; extern struct debuggerinfo_struct kgdb_info[]; diff --git a/kernel/debug/gdbstub.c b/kernel/debug/gdbstub.c index 4b17b326952..481a7bd2dfe 100644 --- a/kernel/debug/gdbstub.c +++ b/kernel/debug/gdbstub.c @@ -6,7 +6,7 @@ * Copyright (C) 2000-2001 VERITAS Software Corporation. * Copyright (C) 2002-2004 Timesys Corporation * Copyright (C) 2003-2004 Amit S. Kale <amitkale@linsyssoft.com> - * Copyright (C) 2004 Pavel Machek <pavel@suse.cz> + * Copyright (C) 2004 Pavel Machek <pavel@ucw.cz> * Copyright (C) 2004-2006 Tom Rini <trini@kernel.crashing.org> * Copyright (C) 2004-2006 LinSysSoft Technologies Pvt. Ltd. * Copyright (C) 2005-2009 Wind River Systems, Inc. @@ -52,17 +52,6 @@ static unsigned long gdb_regs[(NUMREGBYTES + * GDB remote protocol parser: */ -static int hex(char ch) -{ - if ((ch >= 'a') && (ch <= 'f')) - return ch - 'a' + 10; - if ((ch >= '0') && (ch <= '9')) - return ch - '0'; - if ((ch >= 'A') && (ch <= 'F')) - return ch - 'A' + 10; - return -1; -} - #ifdef CONFIG_KGDB_KDB static int gdbstub_read_wait(void) { @@ -123,8 +112,8 @@ static void get_packet(char *buffer) buffer[count] = 0; if (ch == '#') { - xmitcsum = hex(gdbstub_read_wait()) << 4; - xmitcsum += hex(gdbstub_read_wait()); + xmitcsum = hex_to_bin(gdbstub_read_wait()) << 4; + xmitcsum += hex_to_bin(gdbstub_read_wait()); if (checksum != xmitcsum) /* failed checksum */ @@ -236,7 +225,7 @@ void gdbstub_msg_write(const char *s, int len) * buf. Return a pointer to the last char put in buf (null). May * return an error. */ -int kgdb_mem2hex(char *mem, char *buf, int count) +char *kgdb_mem2hex(char *mem, char *buf, int count) { char *tmp; int err; @@ -248,17 +237,16 @@ int kgdb_mem2hex(char *mem, char *buf, int count) tmp = buf + count; err = probe_kernel_read(tmp, mem, count); - if (!err) { - while (count > 0) { - buf = pack_hex_byte(buf, *tmp); - tmp++; - count--; - } - - *buf = 0; + if (err) + return NULL; + while (count > 0) { + buf = pack_hex_byte(buf, *tmp); + tmp++; + count--; } + *buf = 0; - return err; + return buf; } /* @@ -280,8 +268,8 @@ int kgdb_hex2mem(char *buf, char *mem, int count) tmp_hex = tmp_raw - 1; while (tmp_hex >= buf) { tmp_raw--; - *tmp_raw = hex(*tmp_hex--); - *tmp_raw |= hex(*tmp_hex--) << 4; + *tmp_raw = hex_to_bin(*tmp_hex--); + *tmp_raw |= hex_to_bin(*tmp_hex--) << 4; } return probe_kernel_write(mem, tmp_raw, count); @@ -304,7 +292,7 @@ int kgdb_hex2long(char **ptr, unsigned long *long_val) (*ptr)++; } while (**ptr) { - hex_val = hex(**ptr); + hex_val = hex_to_bin(**ptr); if (hex_val < 0) break; @@ -339,6 +327,32 @@ static int kgdb_ebin2mem(char *buf, char *mem, int count) return probe_kernel_write(mem, c, size); } +#if DBG_MAX_REG_NUM > 0 +void pt_regs_to_gdb_regs(unsigned long *gdb_regs, struct pt_regs *regs) +{ + int i; + int idx = 0; + char *ptr = (char *)gdb_regs; + + for (i = 0; i < DBG_MAX_REG_NUM; i++) { + dbg_get_reg(i, ptr + idx, regs); + idx += dbg_reg_def[i].size; + } +} + +void gdb_regs_to_pt_regs(unsigned long *gdb_regs, struct pt_regs *regs) +{ + int i; + int idx = 0; + char *ptr = (char *)gdb_regs; + + for (i = 0; i < DBG_MAX_REG_NUM; i++) { + dbg_set_reg(i, ptr + idx, regs); + idx += dbg_reg_def[i].size; + } +} +#endif /* DBG_MAX_REG_NUM > 0 */ + /* Write memory due to an 'M' or 'X' packet. */ static int write_mem_msg(int binary) { @@ -378,28 +392,31 @@ static void error_packet(char *pkt, int error) * remapped to negative TIDs. */ -#define BUF_THREAD_ID_SIZE 16 +#define BUF_THREAD_ID_SIZE 8 static char *pack_threadid(char *pkt, unsigned char *id) { - char *limit; + unsigned char *limit; + int lzero = 1; + + limit = id + (BUF_THREAD_ID_SIZE / 2); + while (id < limit) { + if (!lzero || *id != 0) { + pkt = pack_hex_byte(pkt, *id); + lzero = 0; + } + id++; + } - limit = pkt + BUF_THREAD_ID_SIZE; - while (pkt < limit) - pkt = pack_hex_byte(pkt, *id++); + if (lzero) + pkt = pack_hex_byte(pkt, 0); return pkt; } static void int_to_threadref(unsigned char *id, int value) { - unsigned char *scan; - int i = 4; - - scan = (unsigned char *)id; - while (i--) - *scan++ = 0; - put_unaligned_be32(value, scan); + put_unaligned_be32(value, id); } static struct task_struct *getthread(struct pt_regs *regs, int tid) @@ -463,8 +480,7 @@ static void gdb_cmd_status(struct kgdb_state *ks) pack_hex_byte(&remcom_out_buffer[1], ks->signo); } -/* Handle the 'g' get registers request */ -static void gdb_cmd_getregs(struct kgdb_state *ks) +static void gdb_get_regs_helper(struct kgdb_state *ks) { struct task_struct *thread; void *local_debuggerinfo; @@ -505,6 +521,12 @@ static void gdb_cmd_getregs(struct kgdb_state *ks) */ sleeping_thread_to_gdb_regs(gdb_regs, thread); } +} + +/* Handle the 'g' get registers request */ +static void gdb_cmd_getregs(struct kgdb_state *ks) +{ + gdb_get_regs_helper(ks); kgdb_mem2hex((char *)gdb_regs, remcom_out_buffer, NUMREGBYTES); } @@ -527,13 +549,13 @@ static void gdb_cmd_memread(struct kgdb_state *ks) char *ptr = &remcom_in_buffer[1]; unsigned long length; unsigned long addr; - int err; + char *err; if (kgdb_hex2long(&ptr, &addr) > 0 && *ptr++ == ',' && kgdb_hex2long(&ptr, &length) > 0) { err = kgdb_mem2hex((char *)addr, remcom_out_buffer, length); - if (err) - error_packet(remcom_out_buffer, err); + if (!err) + error_packet(remcom_out_buffer, -EINVAL); } else { error_packet(remcom_out_buffer, -EINVAL); } @@ -550,6 +572,60 @@ static void gdb_cmd_memwrite(struct kgdb_state *ks) strcpy(remcom_out_buffer, "OK"); } +#if DBG_MAX_REG_NUM > 0 +static char *gdb_hex_reg_helper(int regnum, char *out) +{ + int i; + int offset = 0; + + for (i = 0; i < regnum; i++) + offset += dbg_reg_def[i].size; + return kgdb_mem2hex((char *)gdb_regs + offset, out, + dbg_reg_def[i].size); +} + +/* Handle the 'p' individual regster get */ +static void gdb_cmd_reg_get(struct kgdb_state *ks) +{ + unsigned long regnum; + char *ptr = &remcom_in_buffer[1]; + + kgdb_hex2long(&ptr, ®num); + if (regnum >= DBG_MAX_REG_NUM) { + error_packet(remcom_out_buffer, -EINVAL); + return; + } + gdb_get_regs_helper(ks); + gdb_hex_reg_helper(regnum, remcom_out_buffer); +} + +/* Handle the 'P' individual regster set */ +static void gdb_cmd_reg_set(struct kgdb_state *ks) +{ + unsigned long regnum; + char *ptr = &remcom_in_buffer[1]; + int i = 0; + + kgdb_hex2long(&ptr, ®num); + if (*ptr++ != '=' || + !(!kgdb_usethread || kgdb_usethread == current) || + !dbg_get_reg(regnum, gdb_regs, ks->linux_regs)) { + error_packet(remcom_out_buffer, -EINVAL); + return; + } + memset(gdb_regs, 0, sizeof(gdb_regs)); + while (i < sizeof(gdb_regs) * 2) + if (hex_to_bin(ptr[i]) >= 0) + i++; + else + break; + i = i / 2; + kgdb_hex2mem(ptr, (char *)gdb_regs, i); + dbg_set_reg(regnum, gdb_regs, ks->linux_regs); + strcpy(remcom_out_buffer, "OK"); +} +#endif /* DBG_MAX_REG_NUM > 0 */ + /* Handle the 'X' memory binary write bytes */ static void gdb_cmd_binwrite(struct kgdb_state *ks) { @@ -612,7 +688,7 @@ static void gdb_cmd_query(struct kgdb_state *ks) { struct task_struct *g; struct task_struct *p; - unsigned char thref[8]; + unsigned char thref[BUF_THREAD_ID_SIZE]; char *ptr; int i; int cpu; @@ -621,10 +697,8 @@ static void gdb_cmd_query(struct kgdb_state *ks) switch (remcom_in_buffer[1]) { case 's': case 'f': - if (memcmp(remcom_in_buffer + 2, "ThreadInfo", 10)) { - error_packet(remcom_out_buffer, -EINVAL); + if (memcmp(remcom_in_buffer + 2, "ThreadInfo", 10)) break; - } i = 0; remcom_out_buffer[0] = 'm'; @@ -634,8 +708,7 @@ static void gdb_cmd_query(struct kgdb_state *ks) for_each_online_cpu(cpu) { ks->thr_query = 0; int_to_threadref(thref, -cpu - 2); - pack_threadid(ptr, thref); - ptr += BUF_THREAD_ID_SIZE; + ptr = pack_threadid(ptr, thref); *(ptr++) = ','; i++; } @@ -644,8 +717,7 @@ static void gdb_cmd_query(struct kgdb_state *ks) do_each_thread(g, p) { if (i >= ks->thr_query && !finished) { int_to_threadref(thref, p->pid); - pack_threadid(ptr, thref); - ptr += BUF_THREAD_ID_SIZE; + ptr = pack_threadid(ptr, thref); *(ptr++) = ','; ks->thr_query++; if (ks->thr_query % KGDB_MAX_THREAD_QUERY == 0) @@ -665,10 +737,9 @@ static void gdb_cmd_query(struct kgdb_state *ks) pack_threadid(remcom_out_buffer + 2, thref); break; case 'T': - if (memcmp(remcom_in_buffer + 1, "ThreadExtraInfo,", 16)) { - error_packet(remcom_out_buffer, -EINVAL); + if (memcmp(remcom_in_buffer + 1, "ThreadExtraInfo,", 16)) break; - } + ks->threadid = 0; ptr = remcom_in_buffer + 17; kgdb_hex2long(&ptr, &ks->threadid); @@ -861,11 +932,14 @@ int gdb_serial_stub(struct kgdb_state *ks) int error = 0; int tmp; - /* Clear the out buffer. */ + /* Initialize comm buffer and globals. */ memset(remcom_out_buffer, 0, sizeof(remcom_out_buffer)); + kgdb_usethread = kgdb_info[ks->cpu].task; + ks->kgdb_usethreadid = shadow_pid(kgdb_info[ks->cpu].task->pid); + ks->pass_exception = 0; if (kgdb_connected) { - unsigned char thref[8]; + unsigned char thref[BUF_THREAD_ID_SIZE]; char *ptr; /* Reply to host that an exception has occurred */ @@ -879,10 +953,6 @@ int gdb_serial_stub(struct kgdb_state *ks) put_packet(remcom_out_buffer); } - kgdb_usethread = kgdb_info[ks->cpu].task; - ks->kgdb_usethreadid = shadow_pid(kgdb_info[ks->cpu].task->pid); - ks->pass_exception = 0; - while (1) { error = 0; @@ -907,6 +977,14 @@ int gdb_serial_stub(struct kgdb_state *ks) case 'M': /* MAA..AA,LLLL: Write LLLL bytes at address AA..AA */ gdb_cmd_memwrite(ks); break; +#if DBG_MAX_REG_NUM > 0 + case 'p': /* pXX Return gdb register XX (in hex) */ + gdb_cmd_reg_get(ks); + break; + case 'P': /* PXX=aaaa Set gdb register XX to aaaa (in hex) */ + gdb_cmd_reg_set(ks); + break; +#endif /* DBG_MAX_REG_NUM > 0 */ case 'X': /* XAA..AA,LLLL: Write LLLL bytes at address AA..AA */ gdb_cmd_binwrite(ks); break; diff --git a/kernel/debug/kdb/kdb_bp.c b/kernel/debug/kdb/kdb_bp.c index 75bd9b3ebbb..20059ef4459 100644 --- a/kernel/debug/kdb/kdb_bp.c +++ b/kernel/debug/kdb/kdb_bp.c @@ -274,7 +274,6 @@ static int kdb_bp(int argc, const char **argv) int i, bpno; kdb_bp_t *bp, *bp_check; int diag; - int free; char *symname = NULL; long offset = 0ul; int nextarg; @@ -305,7 +304,6 @@ static int kdb_bp(int argc, const char **argv) /* * Find an empty bp structure to allocate */ - free = KDB_MAXBPT; for (bpno = 0, bp = kdb_breakpoints; bpno < KDB_MAXBPT; bpno++, bp++) { if (bp->bp_free) break; diff --git a/kernel/debug/kdb/kdb_debugger.c b/kernel/debug/kdb/kdb_debugger.c index bf6e8270e95..dd0b1b7dd02 100644 --- a/kernel/debug/kdb/kdb_debugger.c +++ b/kernel/debug/kdb/kdb_debugger.c @@ -86,7 +86,7 @@ int kdb_stub(struct kgdb_state *ks) } /* Set initial kdb state variables */ KDB_STATE_CLEAR(KGDB_TRANS); - kdb_initial_cpu = ks->cpu; + kdb_initial_cpu = atomic_read(&kgdb_active); kdb_current_task = kgdb_info[ks->cpu].task; kdb_current_regs = kgdb_info[ks->cpu].debuggerinfo; /* Remove any breakpoints as needed by kdb and clear single step */ @@ -105,7 +105,6 @@ int kdb_stub(struct kgdb_state *ks) ks->pass_exception = 1; KDB_FLAG_SET(CATASTROPHIC); } - kdb_initial_cpu = ks->cpu; if (KDB_STATE(SSBPT) && reason == KDB_REASON_SSTEP) { KDB_STATE_CLEAR(SSBPT); KDB_STATE_CLEAR(DOING_SS); diff --git a/kernel/debug/kdb/kdb_io.c b/kernel/debug/kdb/kdb_io.c index c9b7f4f90bb..96fdaac46a8 100644 --- a/kernel/debug/kdb/kdb_io.c +++ b/kernel/debug/kdb/kdb_io.c @@ -823,4 +823,4 @@ int kdb_printf(const char *fmt, ...) return r; } - +EXPORT_SYMBOL_GPL(kdb_printf); diff --git a/kernel/debug/kdb/kdb_main.c b/kernel/debug/kdb/kdb_main.c index b724c791b6d..bd3e8e29caa 100644 --- a/kernel/debug/kdb/kdb_main.c +++ b/kernel/debug/kdb/kdb_main.c @@ -82,7 +82,7 @@ static kdbtab_t kdb_base_commands[50]; #define for_each_kdbcmd(cmd, num) \ for ((cmd) = kdb_base_commands, (num) = 0; \ num < kdb_max_commands; \ - num == KDB_BASE_CMD_MAX ? cmd = kdb_commands : cmd++, num++) + num++, num == KDB_BASE_CMD_MAX ? cmd = kdb_commands : cmd++) typedef struct _kdbmsg { int km_diag; /* kdb diagnostic */ @@ -312,7 +312,7 @@ int kdbgetularg(const char *arg, unsigned long *value) if (endp == arg) { /* - * Try base 16, for us folks too lazy to type the + * Also try base 16, for us folks too lazy to type the * leading 0x... */ val = simple_strtoul(arg, &endp, 16); @@ -325,6 +325,25 @@ int kdbgetularg(const char *arg, unsigned long *value) return 0; } +int kdbgetu64arg(const char *arg, u64 *value) +{ + char *endp; + u64 val; + + val = simple_strtoull(arg, &endp, 0); + + if (endp == arg) { + + val = simple_strtoull(arg, &endp, 16); + if (endp == arg) + return KDB_BADINT; + } + + *value = val; + + return 0; +} + /* * kdb_set - This function implements the 'set' command. Alter an * existing environment variable or create a new one. @@ -627,7 +646,7 @@ static int kdb_defcmd2(const char *cmdstr, const char *argv0) } if (!s->usable) return KDB_NOTIMP; - s->command = kmalloc((s->count + 1) * sizeof(*(s->command)), GFP_KDB); + s->command = kzalloc((s->count + 1) * sizeof(*(s->command)), GFP_KDB); if (!s->command) { kdb_printf("Could not allocate new kdb_defcmd table for %s\n", cmdstr); @@ -1108,7 +1127,7 @@ static int kdb_local(kdb_reason_t reason, int error, struct pt_regs *regs, /* special case below */ } else { kdb_printf("\nEntering kdb (current=0x%p, pid %d) ", - kdb_current, kdb_current->pid); + kdb_current, kdb_current ? kdb_current->pid : 0); #if defined(CONFIG_SMP) kdb_printf("on processor %d ", raw_smp_processor_id()); #endif @@ -1730,13 +1749,13 @@ static int kdb_go(int argc, const char **argv) int nextarg; long offset; + if (raw_smp_processor_id() != kdb_initial_cpu) { + kdb_printf("go must execute on the entry cpu, " + "please use \"cpu %d\" and then execute go\n", + kdb_initial_cpu); + return KDB_BADCPUNUM; + } if (argc == 1) { - if (raw_smp_processor_id() != kdb_initial_cpu) { - kdb_printf("go <address> must be issued from the " - "initial cpu, do cpu %d first\n", - kdb_initial_cpu); - return KDB_ARGCOUNT; - } nextarg = 1; diag = kdbgetaddrarg(argc, argv, &nextarg, &addr, &offset, NULL); @@ -1770,11 +1789,65 @@ static int kdb_go(int argc, const char **argv) */ static int kdb_rd(int argc, const char **argv) { - int diag = kdb_check_regs(); - if (diag) - return diag; + int len = kdb_check_regs(); +#if DBG_MAX_REG_NUM > 0 + int i; + char *rname; + int rsize; + u64 reg64; + u32 reg32; + u16 reg16; + u8 reg8; + + if (len) + return len; + + for (i = 0; i < DBG_MAX_REG_NUM; i++) { + rsize = dbg_reg_def[i].size * 2; + if (rsize > 16) + rsize = 2; + if (len + strlen(dbg_reg_def[i].name) + 4 + rsize > 80) { + len = 0; + kdb_printf("\n"); + } + if (len) + len += kdb_printf(" "); + switch(dbg_reg_def[i].size * 8) { + case 8: + rname = dbg_get_reg(i, ®8, kdb_current_regs); + if (!rname) + break; + len += kdb_printf("%s: %02x", rname, reg8); + break; + case 16: + rname = dbg_get_reg(i, ®16, kdb_current_regs); + if (!rname) + break; + len += kdb_printf("%s: %04x", rname, reg16); + break; + case 32: + rname = dbg_get_reg(i, ®32, kdb_current_regs); + if (!rname) + break; + len += kdb_printf("%s: %08x", rname, reg32); + break; + case 64: + rname = dbg_get_reg(i, ®64, kdb_current_regs); + if (!rname) + break; + len += kdb_printf("%s: %016llx", rname, reg64); + break; + default: + len += kdb_printf("%s: ??", dbg_reg_def[i].name); + } + } + kdb_printf("\n"); +#else + if (len) + return len; kdb_dumpregs(kdb_current_regs); +#endif return 0; } @@ -1782,32 +1855,67 @@ static int kdb_rd(int argc, const char **argv) * kdb_rm - This function implements the 'rm' (register modify) command. * rm register-name new-contents * Remarks: - * Currently doesn't allow modification of control or - * debug registers. + * Allows register modification with the same restrictions as gdb */ static int kdb_rm(int argc, const char **argv) { +#if DBG_MAX_REG_NUM > 0 int diag; - int ind = 0; - unsigned long contents; + const char *rname; + int i; + u64 reg64; + u32 reg32; + u16 reg16; + u8 reg8; if (argc != 2) return KDB_ARGCOUNT; /* * Allow presence or absence of leading '%' symbol. */ - if (argv[1][0] == '%') - ind = 1; + rname = argv[1]; + if (*rname == '%') + rname++; - diag = kdbgetularg(argv[2], &contents); + diag = kdbgetu64arg(argv[2], ®64); if (diag) return diag; diag = kdb_check_regs(); if (diag) return diag; + + diag = KDB_BADREG; + for (i = 0; i < DBG_MAX_REG_NUM; i++) { + if (strcmp(rname, dbg_reg_def[i].name) == 0) { + diag = 0; + break; + } + } + if (!diag) { + switch(dbg_reg_def[i].size * 8) { + case 8: + reg8 = reg64; + dbg_set_reg(i, ®8, kdb_current_regs); + break; + case 16: + reg16 = reg64; + dbg_set_reg(i, ®16, kdb_current_regs); + break; + case 32: + reg32 = reg64; + dbg_set_reg(i, ®32, kdb_current_regs); + break; + case 64: + dbg_set_reg(i, ®64, kdb_current_regs); + break; + } + } + return diag; +#else kdb_printf("ERROR: Register set currently not implemented\n"); - return 0; + return 0; +#endif } #if defined(CONFIG_MAGIC_SYSRQ) @@ -1820,9 +1928,8 @@ static int kdb_sr(int argc, const char **argv) { if (argc != 1) return KDB_ARGCOUNT; - sysrq_toggle_support(1); kdb_trap_printk++; - handle_sysrq(*argv[1], NULL); + __handle_sysrq(*argv[1], false); kdb_trap_printk--; return 0; @@ -1857,12 +1964,6 @@ static int kdb_ef(int argc, const char **argv) } #if defined(CONFIG_MODULES) -/* modules using other modules */ -struct module_use { - struct list_head list; - struct module *module_which_uses; -}; - /* * kdb_lsmod - This function implements the 'lsmod' command. Lists * currently loaded kernel modules. @@ -1889,14 +1990,15 @@ static int kdb_lsmod(int argc, const char **argv) kdb_printf(" (Loading)"); else kdb_printf(" (Live)"); + kdb_printf(" 0x%p", mod->module_core); #ifdef CONFIG_MODULE_UNLOAD { struct module_use *use; kdb_printf(" [ "); - list_for_each_entry(use, &mod->modules_which_use_me, - list) - kdb_printf("%s ", use->module_which_uses->name); + list_for_each_entry(use, &mod->source_list, + source_list) + kdb_printf("%s ", use->target->name); kdb_printf("]\n"); } #endif @@ -2259,7 +2361,7 @@ static int kdb_pid(int argc, const char **argv) */ static int kdb_ll(int argc, const char **argv) { - int diag; + int diag = 0; unsigned long addr; long offset = 0; unsigned long va; @@ -2297,18 +2399,22 @@ static int kdb_ll(int argc, const char **argv) while (va) { char buf[80]; + if (KDB_FLAG(CMD_INTERRUPT)) + goto out; + sprintf(buf, "%s " kdb_machreg_fmt "\n", command, va); diag = kdb_parse(buf); if (diag) - return diag; + goto out; addr = va + linkoffset; if (kdb_getword(&va, addr, sizeof(va))) - return 0; + goto out; } - kfree(command); - return 0; +out: + kfree(command); + return diag; } static int kdb_kgdb(int argc, const char **argv) @@ -2443,6 +2549,7 @@ static void kdb_sysinfo(struct sysinfo *val) */ static int kdb_summary(int argc, const char **argv) { + struct timespec now; struct kdb_tm tm; struct sysinfo val; @@ -2457,7 +2564,8 @@ static int kdb_summary(int argc, const char **argv) kdb_printf("domainname %s\n", init_uts_ns.name.domainname); kdb_printf("ccversion %s\n", __stringify(CCVERSION)); - kdb_gmtime(&xtime, &tm); + now = __current_kernel_time(); + kdb_gmtime(&now, &tm); kdb_printf("date %04d-%02d-%02d %02d:%02d:%02d " "tz_minuteswest %d\n", 1900+tm.tm_year, tm.tm_mon+1, tm.tm_mday, @@ -2496,20 +2604,17 @@ static int kdb_summary(int argc, const char **argv) */ static int kdb_per_cpu(int argc, const char **argv) { - char buf[256], fmtstr[64]; - kdb_symtab_t symtab; - cpumask_t suppress = CPU_MASK_NONE; - int cpu, diag; - unsigned long addr, val, bytesperword = 0, whichcpu = ~0UL; + char fmtstr[64]; + int cpu, diag, nextarg = 1; + unsigned long addr, symaddr, val, bytesperword = 0, whichcpu = ~0UL; if (argc < 1 || argc > 3) return KDB_ARGCOUNT; - snprintf(buf, sizeof(buf), "per_cpu__%s", argv[1]); - if (!kdbgetsymval(buf, &symtab)) { - kdb_printf("%s is not a per_cpu variable\n", argv[1]); - return KDB_BADADDR; - } + diag = kdbgetaddrarg(argc, argv, &nextarg, &symaddr, NULL, NULL); + if (diag) + return diag; + if (argc >= 2) { diag = kdbgetularg(argv[2], &bytesperword); if (diag) @@ -2542,46 +2647,25 @@ static int kdb_per_cpu(int argc, const char **argv) #define KDB_PCU(cpu) 0 #endif #endif - for_each_online_cpu(cpu) { + if (KDB_FLAG(CMD_INTERRUPT)) + return 0; + if (whichcpu != ~0UL && whichcpu != cpu) continue; - addr = symtab.sym_start + KDB_PCU(cpu); + addr = symaddr + KDB_PCU(cpu); diag = kdb_getword(&val, addr, bytesperword); if (diag) { kdb_printf("%5d " kdb_bfd_vma_fmt0 " - unable to " "read, diag=%d\n", cpu, addr, diag); continue; } -#ifdef CONFIG_SMP - if (!val) { - cpu_set(cpu, suppress); - continue; - } -#endif /* CONFIG_SMP */ kdb_printf("%5d ", cpu); kdb_md_line(fmtstr, addr, bytesperword == KDB_WORD_SIZE, 1, bytesperword, 1, 1, 0); } - if (cpus_weight(suppress) == 0) - return 0; - kdb_printf("Zero suppressed cpu(s):"); - for (cpu = first_cpu(suppress); cpu < num_possible_cpus(); - cpu = next_cpu(cpu, suppress)) { - kdb_printf(" %d", cpu); - if (cpu == num_possible_cpus() - 1 || - next_cpu(cpu, suppress) != cpu + 1) - continue; - while (cpu < num_possible_cpus() && - next_cpu(cpu, suppress) == cpu + 1) - ++cpu; - kdb_printf("-%d", cpu); - } - kdb_printf("\n"); - #undef KDB_PCU - return 0; } @@ -2656,13 +2740,13 @@ int kdb_register_repeat(char *cmd, } if (kdb_commands) { memcpy(new, kdb_commands, - kdb_max_commands * sizeof(*new)); + (kdb_max_commands - KDB_BASE_CMD_MAX) * sizeof(*new)); kfree(kdb_commands); } memset(new + kdb_max_commands, 0, kdb_command_extend * sizeof(*new)); kdb_commands = new; - kp = kdb_commands + kdb_max_commands; + kp = kdb_commands + kdb_max_commands - KDB_BASE_CMD_MAX; kdb_max_commands += kdb_command_extend; } @@ -2676,6 +2760,8 @@ int kdb_register_repeat(char *cmd, return 0; } +EXPORT_SYMBOL_GPL(kdb_register_repeat); + /* * kdb_register - Compatibility register function for commands that do @@ -2698,6 +2784,7 @@ int kdb_register(char *cmd, return kdb_register_repeat(cmd, func, usage, help, minlen, KDB_REPEAT_NONE); } +EXPORT_SYMBOL_GPL(kdb_register); /* * kdb_unregister - This function is used to unregister a kernel @@ -2716,7 +2803,7 @@ int kdb_unregister(char *cmd) /* * find the command. */ - for (i = 0, kp = kdb_commands; i < kdb_max_commands; i++, kp++) { + for_each_kdbcmd(kp, i) { if (kp->cmd_name && (strcmp(kp->cmd_name, cmd) == 0)) { kp->cmd_name = NULL; return 0; @@ -2726,6 +2813,7 @@ int kdb_unregister(char *cmd) /* Couldn't find it. */ return 1; } +EXPORT_SYMBOL_GPL(kdb_unregister); /* Initialize the kdb command table. */ static void __init kdb_inittab(void) @@ -2826,7 +2914,7 @@ static void __init kdb_cmd_init(void) } } -/* Intialize kdb_printf, breakpoint tables and kdb state */ +/* Initialize kdb_printf, breakpoint tables and kdb state */ void __init kdb_init(int lvl) { static int kdb_init_lvl = KDB_NOT_INITIALIZED; diff --git a/kernel/debug/kdb/kdb_private.h b/kernel/debug/kdb/kdb_private.h index 97d3ba69775..35d69ed1dfb 100644 --- a/kernel/debug/kdb/kdb_private.h +++ b/kernel/debug/kdb/kdb_private.h @@ -15,29 +15,6 @@ #include <linux/kgdb.h> #include "../debug_core.h" -/* Kernel Debugger Error codes. Must not overlap with command codes. */ -#define KDB_NOTFOUND (-1) -#define KDB_ARGCOUNT (-2) -#define KDB_BADWIDTH (-3) -#define KDB_BADRADIX (-4) -#define KDB_NOTENV (-5) -#define KDB_NOENVVALUE (-6) -#define KDB_NOTIMP (-7) -#define KDB_ENVFULL (-8) -#define KDB_ENVBUFFULL (-9) -#define KDB_TOOMANYBPT (-10) -#define KDB_TOOMANYDBREGS (-11) -#define KDB_DUPBPT (-12) -#define KDB_BPTNOTFOUND (-13) -#define KDB_BADMODE (-14) -#define KDB_BADINT (-15) -#define KDB_INVADDRFMT (-16) -#define KDB_BADREG (-17) -#define KDB_BADCPUNUM (-18) -#define KDB_BADLENGTH (-19) -#define KDB_NOBP (-20) -#define KDB_BADADDR (-21) - /* Kernel Debugger Command codes. Must not overlap with error codes. */ #define KDB_CMD_GO (-1001) #define KDB_CMD_CPU (-1002) @@ -93,17 +70,6 @@ */ #define KDB_MAXBPT 16 -/* Maximum number of arguments to a function */ -#define KDB_MAXARGS 16 - -typedef enum { - KDB_REPEAT_NONE = 0, /* Do not repeat this command */ - KDB_REPEAT_NO_ARGS, /* Repeat the command without arguments */ - KDB_REPEAT_WITH_ARGS, /* Repeat the command including its arguments */ -} kdb_repeat_t; - -typedef int (*kdb_func_t)(int, const char **); - /* Symbol table format returned by kallsyms. */ typedef struct __ksymtab { unsigned long value; /* Address of symbol */ @@ -123,11 +89,6 @@ extern int kallsyms_symbol_next(char *prefix_name, int flag); extern int kallsyms_symbol_complete(char *prefix_name, int max_len); /* Exported Symbols for kernel loadable modules to use. */ -extern int kdb_register(char *, kdb_func_t, char *, char *, short); -extern int kdb_register_repeat(char *, kdb_func_t, char *, char *, - short, kdb_repeat_t); -extern int kdb_unregister(char *); - extern int kdb_getarea_size(void *, unsigned long, size_t); extern int kdb_putarea_size(unsigned long, void *, size_t); @@ -144,9 +105,8 @@ extern int kdb_getword(unsigned long *, unsigned long, size_t); extern int kdb_putword(unsigned long, unsigned long, size_t); extern int kdbgetularg(const char *, unsigned long *); -extern int kdb_set(int, const char **); +extern int kdbgetu64arg(const char *, u64 *); extern char *kdbgetenv(const char *); -extern int kdbgetintenv(const char *, int *); extern int kdbgetaddrarg(int, const char **, int*, unsigned long *, long *, char **); extern int kdbgetsymval(const char *, kdb_symtab_t *); @@ -257,7 +217,6 @@ extern void kdb_ps1(const struct task_struct *p); extern void kdb_print_nameval(const char *name, unsigned long val); extern void kdb_send_sig_info(struct task_struct *p, struct siginfo *info); extern void kdb_meminfo_proc_show(void); -extern const char *kdb_walk_kallsyms(loff_t *pos); extern char *kdb_getstr(char *, size_t, char *); /* Defines for kdb_symbol_print */ diff --git a/kernel/debug/kdb/kdb_support.c b/kernel/debug/kdb/kdb_support.c index 45344d5c53d..6b2485dcb05 100644 --- a/kernel/debug/kdb/kdb_support.c +++ b/kernel/debug/kdb/kdb_support.c @@ -82,8 +82,8 @@ static char *kdb_name_table[100]; /* arbitrary size */ int kdbnearsym(unsigned long addr, kdb_symtab_t *symtab) { int ret = 0; - unsigned long symbolsize; - unsigned long offset; + unsigned long symbolsize = 0; + unsigned long offset = 0; #define knt1_size 128 /* must be >= kallsyms table size */ char *knt1 = NULL; diff --git a/kernel/early_res.c b/kernel/early_res.c deleted file mode 100644 index 31aa9332ef3..00000000000 --- a/kernel/early_res.c +++ /dev/null @@ -1,584 +0,0 @@ -/* - * early_res, could be used to replace bootmem - */ -#include <linux/kernel.h> -#include <linux/types.h> -#include <linux/init.h> -#include <linux/bootmem.h> -#include <linux/mm.h> -#include <linux/early_res.h> - -/* - * Early reserved memory areas. - */ -/* - * need to make sure this one is bigger enough before - * find_fw_memmap_area could be used - */ -#define MAX_EARLY_RES_X 32 - -struct early_res { - u64 start, end; - char name[15]; - char overlap_ok; -}; -static struct early_res early_res_x[MAX_EARLY_RES_X] __initdata; - -static int max_early_res __initdata = MAX_EARLY_RES_X; -static struct early_res *early_res __initdata = &early_res_x[0]; -static int early_res_count __initdata; - -static int __init find_overlapped_early(u64 start, u64 end) -{ - int i; - struct early_res *r; - - for (i = 0; i < max_early_res && early_res[i].end; i++) { - r = &early_res[i]; - if (end > r->start && start < r->end) - break; - } - - return i; -} - -/* - * Drop the i-th range from the early reservation map, - * by copying any higher ranges down one over it, and - * clearing what had been the last slot. - */ -static void __init drop_range(int i) -{ - int j; - - for (j = i + 1; j < max_early_res && early_res[j].end; j++) - ; - - memmove(&early_res[i], &early_res[i + 1], - (j - 1 - i) * sizeof(struct early_res)); - - early_res[j - 1].end = 0; - early_res_count--; -} - -static void __init drop_range_partial(int i, u64 start, u64 end) -{ - u64 common_start, common_end; - u64 old_start, old_end; - - old_start = early_res[i].start; - old_end = early_res[i].end; - common_start = max(old_start, start); - common_end = min(old_end, end); - - /* no overlap ? */ - if (common_start >= common_end) - return; - - if (old_start < common_start) { - /* make head segment */ - early_res[i].end = common_start; - if (old_end > common_end) { - char name[15]; - - /* - * Save a local copy of the name, since the - * early_res array could get resized inside - * reserve_early_without_check() -> - * __check_and_double_early_res(), which would - * make the current name pointer invalid. - */ - strncpy(name, early_res[i].name, - sizeof(early_res[i].name) - 1); - /* add another for left over on tail */ - reserve_early_without_check(common_end, old_end, name); - } - return; - } else { - if (old_end > common_end) { - /* reuse the entry for tail left */ - early_res[i].start = common_end; - return; - } - /* all covered */ - drop_range(i); - } -} - -/* - * Split any existing ranges that: - * 1) are marked 'overlap_ok', and - * 2) overlap with the stated range [start, end) - * into whatever portion (if any) of the existing range is entirely - * below or entirely above the stated range. Drop the portion - * of the existing range that overlaps with the stated range, - * which will allow the caller of this routine to then add that - * stated range without conflicting with any existing range. - */ -static void __init drop_overlaps_that_are_ok(u64 start, u64 end) -{ - int i; - struct early_res *r; - u64 lower_start, lower_end; - u64 upper_start, upper_end; - char name[15]; - - for (i = 0; i < max_early_res && early_res[i].end; i++) { - r = &early_res[i]; - - /* Continue past non-overlapping ranges */ - if (end <= r->start || start >= r->end) - continue; - - /* - * Leave non-ok overlaps as is; let caller - * panic "Overlapping early reservations" - * when it hits this overlap. - */ - if (!r->overlap_ok) - return; - - /* - * We have an ok overlap. We will drop it from the early - * reservation map, and add back in any non-overlapping - * portions (lower or upper) as separate, overlap_ok, - * non-overlapping ranges. - */ - - /* 1. Note any non-overlapping (lower or upper) ranges. */ - strncpy(name, r->name, sizeof(name) - 1); - - lower_start = lower_end = 0; - upper_start = upper_end = 0; - if (r->start < start) { - lower_start = r->start; - lower_end = start; - } - if (r->end > end) { - upper_start = end; - upper_end = r->end; - } - - /* 2. Drop the original ok overlapping range */ - drop_range(i); - - i--; /* resume for-loop on copied down entry */ - - /* 3. Add back in any non-overlapping ranges. */ - if (lower_end) - reserve_early_overlap_ok(lower_start, lower_end, name); - if (upper_end) - reserve_early_overlap_ok(upper_start, upper_end, name); - } -} - -static void __init __reserve_early(u64 start, u64 end, char *name, - int overlap_ok) -{ - int i; - struct early_res *r; - - i = find_overlapped_early(start, end); - if (i >= max_early_res) - panic("Too many early reservations"); - r = &early_res[i]; - if (r->end) - panic("Overlapping early reservations " - "%llx-%llx %s to %llx-%llx %s\n", - start, end - 1, name ? name : "", r->start, - r->end - 1, r->name); - r->start = start; - r->end = end; - r->overlap_ok = overlap_ok; - if (name) - strncpy(r->name, name, sizeof(r->name) - 1); - early_res_count++; -} - -/* - * A few early reservtations come here. - * - * The 'overlap_ok' in the name of this routine does -not- mean it - * is ok for these reservations to overlap an earlier reservation. - * Rather it means that it is ok for subsequent reservations to - * overlap this one. - * - * Use this entry point to reserve early ranges when you are doing - * so out of "Paranoia", reserving perhaps more memory than you need, - * just in case, and don't mind a subsequent overlapping reservation - * that is known to be needed. - * - * The drop_overlaps_that_are_ok() call here isn't really needed. - * It would be needed if we had two colliding 'overlap_ok' - * reservations, so that the second such would not panic on the - * overlap with the first. We don't have any such as of this - * writing, but might as well tolerate such if it happens in - * the future. - */ -void __init reserve_early_overlap_ok(u64 start, u64 end, char *name) -{ - drop_overlaps_that_are_ok(start, end); - __reserve_early(start, end, name, 1); -} - -static void __init __check_and_double_early_res(u64 ex_start, u64 ex_end) -{ - u64 start, end, size, mem; - struct early_res *new; - - /* do we have enough slots left ? */ - if ((max_early_res - early_res_count) > max(max_early_res/8, 2)) - return; - - /* double it */ - mem = -1ULL; - size = sizeof(struct early_res) * max_early_res * 2; - if (early_res == early_res_x) - start = 0; - else - start = early_res[0].end; - end = ex_start; - if (start + size < end) - mem = find_fw_memmap_area(start, end, size, - sizeof(struct early_res)); - if (mem == -1ULL) { - start = ex_end; - end = get_max_mapped(); - if (start + size < end) - mem = find_fw_memmap_area(start, end, size, - sizeof(struct early_res)); - } - if (mem == -1ULL) - panic("can not find more space for early_res array"); - - new = __va(mem); - /* save the first one for own */ - new[0].start = mem; - new[0].end = mem + size; - new[0].overlap_ok = 0; - /* copy old to new */ - if (early_res == early_res_x) { - memcpy(&new[1], &early_res[0], - sizeof(struct early_res) * max_early_res); - memset(&new[max_early_res+1], 0, - sizeof(struct early_res) * (max_early_res - 1)); - early_res_count++; - } else { - memcpy(&new[1], &early_res[1], - sizeof(struct early_res) * (max_early_res - 1)); - memset(&new[max_early_res], 0, - sizeof(struct early_res) * max_early_res); - } - memset(&early_res[0], 0, sizeof(struct early_res) * max_early_res); - early_res = new; - max_early_res *= 2; - printk(KERN_DEBUG "early_res array is doubled to %d at [%llx - %llx]\n", - max_early_res, mem, mem + size - 1); -} - -/* - * Most early reservations come here. - * - * We first have drop_overlaps_that_are_ok() drop any pre-existing - * 'overlap_ok' ranges, so that we can then reserve this memory - * range without risk of panic'ing on an overlapping overlap_ok - * early reservation. - */ -void __init reserve_early(u64 start, u64 end, char *name) -{ - if (start >= end) - return; - - __check_and_double_early_res(start, end); - - drop_overlaps_that_are_ok(start, end); - __reserve_early(start, end, name, 0); -} - -void __init reserve_early_without_check(u64 start, u64 end, char *name) -{ - struct early_res *r; - - if (start >= end) - return; - - __check_and_double_early_res(start, end); - - r = &early_res[early_res_count]; - - r->start = start; - r->end = end; - r->overlap_ok = 0; - if (name) - strncpy(r->name, name, sizeof(r->name) - 1); - early_res_count++; -} - -void __init free_early(u64 start, u64 end) -{ - struct early_res *r; - int i; - - i = find_overlapped_early(start, end); - r = &early_res[i]; - if (i >= max_early_res || r->end != end || r->start != start) - panic("free_early on not reserved area: %llx-%llx!", - start, end - 1); - - drop_range(i); -} - -void __init free_early_partial(u64 start, u64 end) -{ - struct early_res *r; - int i; - - if (start == end) - return; - - if (WARN_ONCE(start > end, " wrong range [%#llx, %#llx]\n", start, end)) - return; - -try_next: - i = find_overlapped_early(start, end); - if (i >= max_early_res) - return; - - r = &early_res[i]; - /* hole ? */ - if (r->end >= end && r->start <= start) { - drop_range_partial(i, start, end); - return; - } - - drop_range_partial(i, start, end); - goto try_next; -} - -#ifdef CONFIG_NO_BOOTMEM -static void __init subtract_early_res(struct range *range, int az) -{ - int i, count; - u64 final_start, final_end; - int idx = 0; - - count = 0; - for (i = 0; i < max_early_res && early_res[i].end; i++) - count++; - - /* need to skip first one ?*/ - if (early_res != early_res_x) - idx = 1; - -#define DEBUG_PRINT_EARLY_RES 1 - -#if DEBUG_PRINT_EARLY_RES - printk(KERN_INFO "Subtract (%d early reservations)\n", count); -#endif - for (i = idx; i < count; i++) { - struct early_res *r = &early_res[i]; -#if DEBUG_PRINT_EARLY_RES - printk(KERN_INFO " #%d [%010llx - %010llx] %15s\n", i, - r->start, r->end, r->name); -#endif - final_start = PFN_DOWN(r->start); - final_end = PFN_UP(r->end); - if (final_start >= final_end) - continue; - subtract_range(range, az, final_start, final_end); - } - -} - -int __init get_free_all_memory_range(struct range **rangep, int nodeid) -{ - int i, count; - u64 start = 0, end; - u64 size; - u64 mem; - struct range *range; - int nr_range; - - count = 0; - for (i = 0; i < max_early_res && early_res[i].end; i++) - count++; - - count *= 2; - - size = sizeof(struct range) * count; - end = get_max_mapped(); -#ifdef MAX_DMA32_PFN - if (end > (MAX_DMA32_PFN << PAGE_SHIFT)) - start = MAX_DMA32_PFN << PAGE_SHIFT; -#endif - mem = find_fw_memmap_area(start, end, size, sizeof(struct range)); - if (mem == -1ULL) - panic("can not find more space for range free"); - - range = __va(mem); - /* use early_node_map[] and early_res to get range array at first */ - memset(range, 0, size); - nr_range = 0; - - /* need to go over early_node_map to find out good range for node */ - nr_range = add_from_early_node_map(range, count, nr_range, nodeid); -#ifdef CONFIG_X86_32 - subtract_range(range, count, max_low_pfn, -1ULL); -#endif - subtract_early_res(range, count); - nr_range = clean_sort_range(range, count); - - /* need to clear it ? */ - if (nodeid == MAX_NUMNODES) { - memset(&early_res[0], 0, - sizeof(struct early_res) * max_early_res); - early_res = NULL; - max_early_res = 0; - } - - *rangep = range; - return nr_range; -} -#else -void __init early_res_to_bootmem(u64 start, u64 end) -{ - int i, count; - u64 final_start, final_end; - int idx = 0; - - count = 0; - for (i = 0; i < max_early_res && early_res[i].end; i++) - count++; - - /* need to skip first one ?*/ - if (early_res != early_res_x) - idx = 1; - - printk(KERN_INFO "(%d/%d early reservations) ==> bootmem [%010llx - %010llx]\n", - count - idx, max_early_res, start, end); - for (i = idx; i < count; i++) { - struct early_res *r = &early_res[i]; - printk(KERN_INFO " #%d [%010llx - %010llx] %16s", i, - r->start, r->end, r->name); - final_start = max(start, r->start); - final_end = min(end, r->end); - if (final_start >= final_end) { - printk(KERN_CONT "\n"); - continue; - } - printk(KERN_CONT " ==> [%010llx - %010llx]\n", - final_start, final_end); - reserve_bootmem_generic(final_start, final_end - final_start, - BOOTMEM_DEFAULT); - } - /* clear them */ - memset(&early_res[0], 0, sizeof(struct early_res) * max_early_res); - early_res = NULL; - max_early_res = 0; - early_res_count = 0; -} -#endif - -/* Check for already reserved areas */ -static inline int __init bad_addr(u64 *addrp, u64 size, u64 align) -{ - int i; - u64 addr = *addrp; - int changed = 0; - struct early_res *r; -again: - i = find_overlapped_early(addr, addr + size); - r = &early_res[i]; - if (i < max_early_res && r->end) { - *addrp = addr = round_up(r->end, align); - changed = 1; - goto again; - } - return changed; -} - -/* Check for already reserved areas */ -static inline int __init bad_addr_size(u64 *addrp, u64 *sizep, u64 align) -{ - int i; - u64 addr = *addrp, last; - u64 size = *sizep; - int changed = 0; -again: - last = addr + size; - for (i = 0; i < max_early_res && early_res[i].end; i++) { - struct early_res *r = &early_res[i]; - if (last > r->start && addr < r->start) { - size = r->start - addr; - changed = 1; - goto again; - } - if (last > r->end && addr < r->end) { - addr = round_up(r->end, align); - size = last - addr; - changed = 1; - goto again; - } - if (last <= r->end && addr >= r->start) { - (*sizep)++; - return 0; - } - } - if (changed) { - *addrp = addr; - *sizep = size; - } - return changed; -} - -/* - * Find a free area with specified alignment in a specific range. - * only with the area.between start to end is active range from early_node_map - * so they are good as RAM - */ -u64 __init find_early_area(u64 ei_start, u64 ei_last, u64 start, u64 end, - u64 size, u64 align) -{ - u64 addr, last; - - addr = round_up(ei_start, align); - if (addr < start) - addr = round_up(start, align); - if (addr >= ei_last) - goto out; - while (bad_addr(&addr, size, align) && addr+size <= ei_last) - ; - last = addr + size; - if (last > ei_last) - goto out; - if (last > end) - goto out; - - return addr; - -out: - return -1ULL; -} - -u64 __init find_early_area_size(u64 ei_start, u64 ei_last, u64 start, - u64 *sizep, u64 align) -{ - u64 addr, last; - - addr = round_up(ei_start, align); - if (addr < start) - addr = round_up(start, align); - if (addr >= ei_last) - goto out; - *sizep = ei_last - addr; - while (bad_addr_size(&addr, sizep, align) && addr + *sizep <= ei_last) - ; - last = addr + *sizep; - if (last > ei_last) - goto out; - - return addr; - -out: - return -1ULL; -} diff --git a/kernel/exec_domain.c b/kernel/exec_domain.c index c35452cadde..0dbeae37422 100644 --- a/kernel/exec_domain.c +++ b/kernel/exec_domain.c @@ -27,7 +27,7 @@ static struct exec_domain *exec_domains = &default_exec_domain; static DEFINE_RWLOCK(exec_domains_lock); -static u_long ident_map[32] = { +static unsigned long ident_map[32] = { 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, @@ -56,10 +56,10 @@ default_handler(int segment, struct pt_regs *regp) } static struct exec_domain * -lookup_exec_domain(u_long personality) +lookup_exec_domain(unsigned int personality) { - struct exec_domain * ep; - u_long pers = personality(personality); + unsigned int pers = personality(personality); + struct exec_domain *ep; read_lock(&exec_domains_lock); for (ep = exec_domains; ep; ep = ep->next) { @@ -70,7 +70,7 @@ lookup_exec_domain(u_long personality) #ifdef CONFIG_MODULES read_unlock(&exec_domains_lock); - request_module("personality-%ld", pers); + request_module("personality-%d", pers); read_lock(&exec_domains_lock); for (ep = exec_domains; ep; ep = ep->next) { @@ -134,23 +134,14 @@ unregister: return 0; } -int -__set_personality(u_long personality) +int __set_personality(unsigned int personality) { - struct exec_domain *ep, *oep; - - ep = lookup_exec_domain(personality); - if (ep == current_thread_info()->exec_domain) { - current->personality = personality; - module_put(ep->module); - return 0; - } + struct exec_domain *oep = current_thread_info()->exec_domain; + current_thread_info()->exec_domain = lookup_exec_domain(personality); current->personality = personality; - oep = current_thread_info()->exec_domain; - current_thread_info()->exec_domain = ep; - module_put(oep->module); + return 0; } @@ -188,17 +179,14 @@ static int __init proc_execdomains_init(void) module_init(proc_execdomains_init); #endif -SYSCALL_DEFINE1(personality, u_long, personality) +SYSCALL_DEFINE1(personality, unsigned int, personality) { - u_long old = current->personality; + unsigned int old = current->personality; - if (personality != 0xffffffff) { + if (personality != 0xffffffff) set_personality(personality); - if (current->personality != personality) - return -EINVAL; - } - return (long)old; + return old; } diff --git a/kernel/exit.c b/kernel/exit.c index eabca5a73a8..f9a45ebcc7b 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -50,6 +50,7 @@ #include <linux/perf_event.h> #include <trace/events/sched.h> #include <linux/hw_breakpoint.h> +#include <linux/oom.h> #include <asm/uaccess.h> #include <asm/unistd.h> @@ -58,17 +59,17 @@ static void exit_mm(struct task_struct * tsk); -static void __unhash_process(struct task_struct *p) +static void __unhash_process(struct task_struct *p, bool group_dead) { nr_threads--; detach_pid(p, PIDTYPE_PID); - if (thread_group_leader(p)) { + if (group_dead) { detach_pid(p, PIDTYPE_PGID); detach_pid(p, PIDTYPE_SID); list_del_rcu(&p->tasks); list_del_init(&p->sibling); - __get_cpu_var(process_counts)--; + __this_cpu_dec(process_counts); } list_del_rcu(&p->thread_group); } @@ -79,10 +80,9 @@ static void __unhash_process(struct task_struct *p) static void __exit_signal(struct task_struct *tsk) { struct signal_struct *sig = tsk->signal; + bool group_dead = thread_group_leader(tsk); struct sighand_struct *sighand; - - BUG_ON(!sig); - BUG_ON(!atomic_read(&sig->count)); + struct tty_struct *uninitialized_var(tty); sighand = rcu_dereference_check(tsk->sighand, rcu_read_lock_held() || @@ -90,14 +90,24 @@ static void __exit_signal(struct task_struct *tsk) spin_lock(&sighand->siglock); posix_cpu_timers_exit(tsk); - if (atomic_dec_and_test(&sig->count)) + if (group_dead) { posix_cpu_timers_exit_group(tsk); - else { + tty = sig->tty; + sig->tty = NULL; + } else { + /* + * This can only happen if the caller is de_thread(). + * FIXME: this is the temporary hack, we should teach + * posix-cpu-timers to handle this case correctly. + */ + if (unlikely(has_group_leader_pid(tsk))) + posix_cpu_timers_exit_group(tsk); + /* * If there is any task waiting for the group exit * then notify it: */ - if (sig->group_exit_task && atomic_read(&sig->count) == sig->notify_count) + if (sig->notify_count > 0 && !--sig->notify_count) wake_up_process(sig->group_exit_task); if (tsk == sig->curr_target) @@ -123,32 +133,24 @@ static void __exit_signal(struct task_struct *tsk) sig->oublock += task_io_get_oublock(tsk); task_io_accounting_add(&sig->ioac, &tsk->ioac); sig->sum_sched_runtime += tsk->se.sum_exec_runtime; - sig = NULL; /* Marker for below. */ } - __unhash_process(tsk); + sig->nr_threads--; + __unhash_process(tsk, group_dead); /* * Do this under ->siglock, we can race with another thread * doing sigqueue_free() if we have SIGQUEUE_PREALLOC signals. */ flush_sigqueue(&tsk->pending); - - tsk->signal = NULL; tsk->sighand = NULL; spin_unlock(&sighand->siglock); __cleanup_sighand(sighand); clear_tsk_thread_flag(tsk,TIF_SIGPENDING); - if (sig) { + if (group_dead) { flush_sigqueue(&sig->shared_pending); - taskstats_tgid_free(sig); - /* - * Make sure ->signal can't go away under rq->lock, - * see account_group_exec_runtime(). - */ - task_rq_unlock_wait(tsk); - __cleanup_signal(sig); + tty_kref_put(tty); } } @@ -156,9 +158,7 @@ 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 + perf_event_delayed_put(tsk); trace_sched_process_free(tsk); put_task_struct(tsk); } @@ -696,6 +696,8 @@ static void exit_mm(struct task_struct * tsk) enter_lazy_tlb(mm, current); /* We don't want this task to be frozen prematurely */ clear_freeze_flag(tsk); + if (tsk->signal->oom_score_adj == OOM_SCORE_ADJ_MIN) + atomic_dec(&mm->oom_disable_count); task_unlock(tsk); mm_update_next_owner(mm); mmput(mm); @@ -709,6 +711,8 @@ static void exit_mm(struct task_struct * tsk) * space. */ static struct task_struct *find_new_reaper(struct task_struct *father) + __releases(&tasklist_lock) + __acquires(&tasklist_lock) { struct pid_namespace *pid_ns = task_active_pid_ns(father); struct task_struct *thread; @@ -778,9 +782,12 @@ static void forget_original_parent(struct task_struct *father) struct task_struct *p, *n, *reaper; LIST_HEAD(dead_children); - exit_ptrace(father); - write_lock_irq(&tasklist_lock); + /* + * Note that exit_ptrace() and find_new_reaper() might + * drop tasklist_lock and reacquire it. + */ + exit_ptrace(father); reaper = find_new_reaper(father); list_for_each_entry_safe(p, n, &father->children, sibling) { @@ -856,12 +863,9 @@ static void exit_notify(struct task_struct *tsk, int group_dead) tsk->exit_state = signal == DEATH_REAP ? EXIT_DEAD : EXIT_ZOMBIE; - /* mt-exec, de_thread() is waiting for us */ - if (thread_group_leader(tsk) && - tsk->signal->group_exit_task && - tsk->signal->notify_count < 0) + /* mt-exec, de_thread() is waiting for group leader */ + if (unlikely(tsk->signal->notify_count < 0)) wake_up_process(tsk->signal->group_exit_task); - write_unlock_irq(&tasklist_lock); tracehook_report_death(tsk, signal, cookie, group_dead); @@ -910,6 +914,15 @@ NORET_TYPE void do_exit(long code) if (unlikely(!tsk->pid)) panic("Attempted to kill the idle task!"); + /* + * If do_exit is called because this processes oopsed, it's possible + * that get_fs() was left as KERNEL_DS, so reset it to USER_DS before + * continuing. Amongst other possible reasons, this is to prevent + * mm_release()->clear_child_tid() from writing to a user-controlled + * kernel address. + */ + set_fs(USER_DS); + tracehook_report_exit(&code); validate_creds_for_do_exit(tsk); @@ -981,6 +994,15 @@ NORET_TYPE void do_exit(long code) exit_fs(tsk); check_stack_usage(); exit_thread(); + + /* + * Flush inherited counters to the parent - before the parent + * gets woken up by child-exit notifications. + * + * because of cgroup mode, must be called before cgroup_exit() + */ + perf_event_exit_task(tsk); + cgroup_exit(tsk, 1); if (group_dead) @@ -994,16 +1016,13 @@ NORET_TYPE void do_exit(long code) * 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 + task_lock(tsk); mpol_put(tsk->mempolicy); tsk->mempolicy = NULL; + task_unlock(tsk); #endif #ifdef CONFIG_FUTEX if (unlikely(current->pi_state_cache)) @@ -1391,8 +1410,7 @@ static int wait_task_stopped(struct wait_opts *wo, 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; + uid = task_uid(p); unlock_sig: spin_unlock_irq(&p->sighand->siglock); if (!exit_code) @@ -1465,7 +1483,7 @@ static int wait_task_continued(struct wait_opts *wo, struct task_struct *p) } if (!unlikely(wo->wo_flags & WNOWAIT)) p->signal->flags &= ~SIGNAL_STOP_CONTINUED; - uid = __task_cred(p)->uid; + uid = task_uid(p); spin_unlock_irq(&p->sighand->siglock); pid = task_pid_vnr(p); diff --git a/kernel/fork.c b/kernel/fork.c index 4d57d9e3a6e..25e429152dd 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -65,6 +65,8 @@ #include <linux/perf_event.h> #include <linux/posix-timers.h> #include <linux/user-return-notifier.h> +#include <linux/oom.h> +#include <linux/khugepaged.h> #include <asm/pgtable.h> #include <asm/pgalloc.h> @@ -165,6 +167,19 @@ void free_task(struct task_struct *tsk) } EXPORT_SYMBOL(free_task); +static inline void free_signal_struct(struct signal_struct *sig) +{ + taskstats_tgid_free(sig); + sched_autogroup_exit(sig); + kmem_cache_free(signal_cachep, sig); +} + +static inline void put_signal_struct(struct signal_struct *sig) +{ + if (atomic_dec_and_test(&sig->sigcnt)) + free_signal_struct(sig); +} + void __put_task_struct(struct task_struct *tsk) { WARN_ON(!tsk->exit_state); @@ -173,6 +188,7 @@ void __put_task_struct(struct task_struct *tsk) exit_creds(tsk); delayacct_tsk_free(tsk); + put_signal_struct(tsk->signal); if (!profile_handoff_task(tsk)) free_task(tsk); @@ -259,6 +275,7 @@ static struct task_struct *dup_task_struct(struct task_struct *orig) setup_thread_stack(tsk, orig); clear_user_return_notifier(tsk); + clear_tsk_need_resched(tsk); stackend = end_of_stack(tsk); *stackend = STACK_END_MAGIC; /* for overflow detection */ @@ -287,7 +304,7 @@ out: #ifdef CONFIG_MMU static int dup_mmap(struct mm_struct *mm, struct mm_struct *oldmm) { - struct vm_area_struct *mpnt, *tmp, **pprev; + struct vm_area_struct *mpnt, *tmp, *prev, **pprev; struct rb_node **rb_link, *rb_parent; int retval; unsigned long charge; @@ -314,7 +331,11 @@ static int dup_mmap(struct mm_struct *mm, struct mm_struct *oldmm) retval = ksm_fork(mm, oldmm); if (retval) goto out; + retval = khugepaged_fork(mm, oldmm); + if (retval) + goto out; + prev = NULL; for (mpnt = oldmm->mmap; mpnt; mpnt = mpnt->vm_next) { struct file *file; @@ -342,11 +363,11 @@ static int dup_mmap(struct mm_struct *mm, struct mm_struct *oldmm) if (IS_ERR(pol)) goto fail_nomem_policy; vma_set_policy(tmp, pol); + tmp->vm_mm = mm; if (anon_vma_fork(tmp, mpnt)) goto fail_nomem_anon_vma_fork; tmp->vm_flags &= ~VM_LOCKED; - tmp->vm_mm = mm; - tmp->vm_next = NULL; + tmp->vm_next = tmp->vm_prev = NULL; file = tmp->vm_file; if (file) { struct inode *inode = file->f_path.dentry->d_inode; @@ -379,6 +400,8 @@ static int dup_mmap(struct mm_struct *mm, struct mm_struct *oldmm) */ *pprev = tmp; pprev = &tmp->vm_next; + tmp->vm_prev = prev; + prev = tmp; __vma_link_rb(mm, tmp, rb_link, rb_parent); rb_link = &tmp->vm_rb.rb_right; @@ -472,6 +495,7 @@ static struct mm_struct * mm_init(struct mm_struct * mm, struct task_struct *p) mm->cached_hole_size = ~0UL; mm_init_aio(mm); mm_init_owner(mm, p); + atomic_set(&mm->oom_disable_count, 0); if (likely(!mm_alloc_pgd(mm))) { mm->def_flags = 0; @@ -509,6 +533,9 @@ void __mmdrop(struct mm_struct *mm) mm_free_pgd(mm); destroy_context(mm); mmu_notifier_mm_destroy(mm); +#ifdef CONFIG_TRANSPARENT_HUGEPAGE + VM_BUG_ON(mm->pmd_huge_pte); +#endif free_mm(mm); } EXPORT_SYMBOL_GPL(__mmdrop); @@ -523,6 +550,7 @@ void mmput(struct mm_struct *mm) if (atomic_dec_and_test(&mm->mm_users)) { exit_aio(mm); ksm_exit(mm); + khugepaged_exit(mm); /* must run before exit_mmap */ exit_mmap(mm); set_mm_exe_file(mm, NULL); if (!list_empty(&mm->mmlist)) { @@ -649,6 +677,10 @@ struct mm_struct *dup_mm(struct task_struct *tsk) mm->token_priority = 0; mm->last_interval = 0; +#ifdef CONFIG_TRANSPARENT_HUGEPAGE + mm->pmd_huge_pte = NULL; +#endif + if (!mm_init(mm, tsk)) goto fail_nomem; @@ -725,6 +757,8 @@ good_mm: /* Initializing for Swap token stuff */ mm->token_priority = 0; mm->last_interval = 0; + if (tsk->signal->oom_score_adj == OOM_SCORE_ADJ_MIN) + atomic_inc(&mm->oom_disable_count); tsk->mm = mm; tsk->active_mm = mm; @@ -739,13 +773,13 @@ static int copy_fs(unsigned long clone_flags, struct task_struct *tsk) struct fs_struct *fs = current->fs; if (clone_flags & CLONE_FS) { /* tsk->fs is already what we want */ - write_lock(&fs->lock); + spin_lock(&fs->lock); if (fs->in_exec) { - write_unlock(&fs->lock); + spin_unlock(&fs->lock); return -EAGAIN; } fs->users++; - write_unlock(&fs->lock); + spin_unlock(&fs->lock); return 0; } tsk->fs = copy_fs_struct(fs); @@ -864,8 +898,9 @@ static int copy_signal(unsigned long clone_flags, struct task_struct *tsk) if (!sig) return -ENOMEM; - atomic_set(&sig->count, 1); + sig->nr_threads = 1; atomic_set(&sig->live, 1); + atomic_set(&sig->sigcnt, 1); init_waitqueue_head(&sig->wait_chldexit); if (clone_flags & CLONE_NEWPID) sig->flags |= SIGNAL_UNKILLABLE; @@ -883,24 +918,22 @@ static int copy_signal(unsigned long clone_flags, struct task_struct *tsk) posix_cpu_timers_init_group(sig); tty_audit_fork(sig); + sched_autogroup_fork(sig); sig->oom_adj = current->signal->oom_adj; + sig->oom_score_adj = current->signal->oom_score_adj; + sig->oom_score_adj_min = current->signal->oom_score_adj_min; - return 0; -} + mutex_init(&sig->cred_guard_mutex); -void __cleanup_signal(struct signal_struct *sig) -{ - thread_group_cputime_free(sig); - tty_kref_put(sig->tty); - kmem_cache_free(signal_cachep, sig); + return 0; } static void copy_flags(unsigned long clone_flags, struct task_struct *p) { unsigned long new_flags = p->flags; - new_flags &= ~PF_SUPERPRIV; + new_flags &= ~(PF_SUPERPRIV | PF_WQ_WORKER); new_flags |= PF_FORKNOEXEC; new_flags |= PF_STARTING; p->flags = new_flags; @@ -1245,8 +1278,9 @@ static struct task_struct *copy_process(unsigned long clone_flags, } if (clone_flags & CLONE_THREAD) { - atomic_inc(¤t->signal->count); + current->signal->nr_threads++; atomic_inc(¤t->signal->live); + atomic_inc(¤t->signal->sigcnt); p->group_leader = current->group_leader; list_add_tail_rcu(&p->thread_group, &p->group_leader->thread_group); } @@ -1259,13 +1293,12 @@ static struct task_struct *copy_process(unsigned long clone_flags, p->nsproxy->pid_ns->child_reaper = p; p->signal->leader_pid = pid; - tty_kref_put(p->signal->tty); p->signal->tty = tty_kref_get(current->signal->tty); attach_pid(p, PIDTYPE_PGID, task_pgrp(current)); attach_pid(p, PIDTYPE_SID, task_session(current)); list_add_tail(&p->sibling, &p->real_parent->children); list_add_tail_rcu(&p->tasks, &init_task.tasks); - __get_cpu_var(process_counts)++; + __this_cpu_inc(process_counts); } attach_pid(p, PIDTYPE_PID, pid); nr_threads++; @@ -1288,11 +1321,16 @@ bad_fork_cleanup_io: bad_fork_cleanup_namespaces: exit_task_namespaces(p); bad_fork_cleanup_mm: - if (p->mm) + if (p->mm) { + task_lock(p); + if (p->signal->oom_score_adj == OOM_SCORE_ADJ_MIN) + atomic_dec(&p->mm->oom_disable_count); + task_unlock(p); mmput(p->mm); + } bad_fork_cleanup_signal: if (!(clone_flags & CLONE_THREAD)) - __cleanup_signal(p->signal); + free_signal_struct(p->signal); bad_fork_cleanup_sighand: __cleanup_sighand(p->sighand); bad_fork_cleanup_fs: @@ -1327,6 +1365,16 @@ noinline struct pt_regs * __cpuinit __attribute__((weak)) idle_regs(struct pt_re return regs; } +static inline void init_idle_pids(struct pid_link *links) +{ + enum pid_type type; + + for (type = PIDTYPE_PID; type < PIDTYPE_MAX; ++type) { + INIT_HLIST_NODE(&links[type].node); /* not really needed */ + links[type].pid = &init_struct_pid; + } +} + struct task_struct * __cpuinit fork_idle(int cpu) { struct task_struct *task; @@ -1334,8 +1382,10 @@ struct task_struct * __cpuinit fork_idle(int cpu) task = copy_process(CLONE_VM, 0, idle_regs(®s), 0, NULL, &init_struct_pid, 0); - if (!IS_ERR(task)) + if (!IS_ERR(task)) { + init_idle_pids(task->pids); init_idle(task, cpu); + } return task; } @@ -1373,23 +1423,6 @@ long do_fork(unsigned long clone_flags, } /* - * We hope to recycle these flags after 2.6.26 - */ - if (unlikely(clone_flags & CLONE_STOPPED)) { - static int __read_mostly count = 100; - - if (count > 0 && printk_ratelimit()) { - char comm[TASK_COMM_LEN]; - - count--; - printk(KERN_INFO "fork(): process `%s' used deprecated " - "clone flags 0x%lx\n", - get_task_comm(comm, current), - clone_flags & CLONE_STOPPED); - } - } - - /* * When called from kernel_thread, don't do user tracing stuff. */ if (likely(user_mode(regs))) @@ -1427,16 +1460,7 @@ long do_fork(unsigned long clone_flags, */ p->flags &= ~PF_STARTING; - if (unlikely(clone_flags & CLONE_STOPPED)) { - /* - * We'll start up with an immediate SIGSTOP. - */ - sigaddset(&p->pending.signal, SIGSTOP); - set_tsk_thread_flag(p, TIF_SIGPENDING); - __set_task_state(p, TASK_STOPPED); - } else { - wake_up_new_task(p, clone_flags); - } + wake_up_new_task(p, clone_flags); tracehook_report_clone_complete(trace, regs, clone_flags, nr, p); @@ -1507,14 +1531,6 @@ static void check_unshare_flags(unsigned long *flags_ptr) *flags_ptr |= CLONE_SIGHAND; /* - * If unsharing signal handlers and the task was created - * using CLONE_THREAD, then must unshare the thread - */ - if ((*flags_ptr & CLONE_SIGHAND) && - (atomic_read(¤t->signal->count) > 1)) - *flags_ptr |= CLONE_THREAD; - - /* * If unsharing namespace, must also unshare filesystem information. */ if (*flags_ptr & CLONE_NEWNS) @@ -1664,13 +1680,13 @@ SYSCALL_DEFINE1(unshare, unsigned long, unshare_flags) if (new_fs) { fs = current->fs; - write_lock(&fs->lock); + spin_lock(&fs->lock); current->fs = new_fs; if (--fs->users) new_fs = NULL; else new_fs = fs; - write_unlock(&fs->lock); + spin_unlock(&fs->lock); } if (new_mm) { @@ -1678,6 +1694,10 @@ SYSCALL_DEFINE1(unshare, unsigned long, unshare_flags) active_mm = current->active_mm; current->mm = new_mm; current->active_mm = new_mm; + if (current->signal->oom_score_adj == OOM_SCORE_ADJ_MIN) { + atomic_dec(&mm->oom_disable_count); + atomic_inc(&new_mm->oom_disable_count); + } activate_mm(active_mm, new_mm); new_mm = mm; } diff --git a/kernel/freezer.c b/kernel/freezer.c index bd1d42b17cb..66ecd2ead21 100644 --- a/kernel/freezer.c +++ b/kernel/freezer.c @@ -104,8 +104,13 @@ bool freeze_task(struct task_struct *p, bool sig_only) } if (should_send_signal(p)) { - if (!signal_pending(p)) - fake_signal_wake_up(p); + fake_signal_wake_up(p); + /* + * fake_signal_wake_up() goes through p's scheduler + * lock and guarantees that TASK_STOPPED/TRACED -> + * TASK_RUNNING transition can't race with task state + * testing in try_to_freeze_tasks(). + */ } else if (sig_only) { return false; } else { diff --git a/kernel/futex.c b/kernel/futex.c index e7a35f1039e..b766d28accd 100644 --- a/kernel/futex.c +++ b/kernel/futex.c @@ -69,6 +69,14 @@ int __read_mostly futex_cmpxchg_enabled; #define FUTEX_HASHBITS (CONFIG_BASE_SMALL ? 4 : 8) /* + * Futex flags used to encode options to functions and preserve them across + * restarts. + */ +#define FLAGS_SHARED 0x01 +#define FLAGS_CLOCKRT 0x02 +#define FLAGS_HAS_TIMEOUT 0x04 + +/* * Priority Inheritance state: */ struct futex_pi_state { @@ -91,6 +99,7 @@ struct futex_pi_state { /** * struct futex_q - The hashed futex queue entry, one per waiting task + * @list: priority-sorted list of tasks waiting on this futex * @task: the task waiting on the futex * @lock_ptr: the hash bucket lock * @key: the key the futex is hashed on @@ -104,7 +113,7 @@ struct futex_pi_state { * * 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 + * The order of wakeup is always to make the first condition true, then * the second. * * PI futexes are typically woken before they are removed from the hash list via @@ -122,6 +131,12 @@ struct futex_q { u32 bitset; }; +static const struct futex_q futex_q_init = { + /* list gets initialized in queue_me()*/ + .key = FUTEX_KEY_INIT, + .bitset = FUTEX_BITSET_MATCH_ANY +}; + /* * Hash buckets are shared by all the futex_keys that hash to the same * location. Each key may have multiple futex_q structures, one for each task @@ -168,7 +183,7 @@ static void get_futex_key_refs(union futex_key *key) switch (key->both.offset & (FUT_OFF_INODE|FUT_OFF_MMSHARED)) { case FUT_OFF_INODE: - atomic_inc(&key->shared.inode->i_count); + ihold(key->shared.inode); break; case FUT_OFF_MMSHARED: atomic_inc(&key->private.mm->mm_count); @@ -218,7 +233,7 @@ get_futex_key(u32 __user *uaddr, int fshared, union futex_key *key) { unsigned long address = (unsigned long)uaddr; struct mm_struct *mm = current->mm; - struct page *page; + struct page *page, *page_head; int err; /* @@ -250,11 +265,46 @@ again: if (err < 0) return err; - page = compound_head(page); - lock_page(page); - if (!page->mapping) { - unlock_page(page); +#ifdef CONFIG_TRANSPARENT_HUGEPAGE + page_head = page; + if (unlikely(PageTail(page))) { + put_page(page); + /* serialize against __split_huge_page_splitting() */ + local_irq_disable(); + if (likely(__get_user_pages_fast(address, 1, 1, &page) == 1)) { + page_head = compound_head(page); + /* + * page_head is valid pointer but we must pin + * it before taking the PG_lock and/or + * PG_compound_lock. The moment we re-enable + * irqs __split_huge_page_splitting() can + * return and the head page can be freed from + * under us. We can't take the PG_lock and/or + * PG_compound_lock on a page that could be + * freed from under us. + */ + if (page != page_head) { + get_page(page_head); + put_page(page); + } + local_irq_enable(); + } else { + local_irq_enable(); + goto again; + } + } +#else + page_head = compound_head(page); + if (page != page_head) { + get_page(page_head); put_page(page); + } +#endif + + lock_page(page_head); + if (!page_head->mapping) { + unlock_page(page_head); + put_page(page_head); goto again; } @@ -265,25 +315,24 @@ again: * it's a read-only handle, it's expected that futexes attach to * the object not the particular process. */ - if (PageAnon(page)) { + if (PageAnon(page_head)) { key->both.offset |= FUT_OFF_MMSHARED; /* ref taken on mm */ key->private.mm = mm; key->private.address = address; } else { key->both.offset |= FUT_OFF_INODE; /* inode-based key */ - key->shared.inode = page->mapping->host; - key->shared.pgoff = page->index; + key->shared.inode = page_head->mapping->host; + key->shared.pgoff = page_head->index; } get_futex_key_refs(key); - unlock_page(page); - put_page(page); + unlock_page(page_head); + put_page(page_head); return 0; } -static inline -void put_futex_key(int fshared, union futex_key *key) +static inline void put_futex_key(union futex_key *key) { drop_futex_key_refs(key); } @@ -295,7 +344,7 @@ void put_futex_key(int fshared, union futex_key *key) * 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 + * 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. @@ -429,20 +478,11 @@ static void free_pi_state(struct futex_pi_state *pi_state) static struct task_struct * futex_find_get_task(pid_t pid) { struct task_struct *p; - const struct cred *cred = current_cred(), *pcred; rcu_read_lock(); p = find_task_by_vpid(pid); - if (!p) { - p = ERR_PTR(-ESRCH); - } else { - pcred = __task_cred(p); - if (cred->euid != pcred->euid && - cred->euid != pcred->uid) - p = ERR_PTR(-ESRCH); - else - get_task_struct(p); - } + if (p) + get_task_struct(p); rcu_read_unlock(); @@ -524,7 +564,7 @@ lookup_pi_state(u32 uval, struct futex_hash_bucket *hb, */ pi_state = this->pi_state; /* - * Userspace might have messed up non PI and PI futexes + * Userspace might have messed up non-PI and PI futexes */ if (unlikely(!pi_state)) return -EINVAL; @@ -564,8 +604,8 @@ lookup_pi_state(u32 uval, struct futex_hash_bucket *hb, if (!pid) return -ESRCH; p = futex_find_get_task(pid); - if (IS_ERR(p)) - return PTR_ERR(p); + if (!p) + return -ESRCH; /* * We need to look at the task state flags to figure out, @@ -745,8 +785,8 @@ static void wake_futex(struct futex_q *q) /* * 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 + * 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. */ @@ -786,10 +826,9 @@ static int wake_futex_pi(u32 __user *uaddr, u32 uval, struct futex_q *this) new_owner = rt_mutex_next_owner(&pi_state->pi_mutex); /* - * This happens when we have stolen the lock and the original - * pending owner did not enqueue itself back on the rt_mutex. - * Thats not a tragedy. We know that way, that a lock waiter - * is on the fly. We make the futex_q waiter the pending owner. + * It is possible that the next waiter (the one that brought + * this owner to the kernel) timed out and is no longer + * waiting on the lock. */ if (!new_owner) new_owner = this->task; @@ -878,7 +917,8 @@ double_unlock_hb(struct futex_hash_bucket *hb1, struct futex_hash_bucket *hb2) /* * Wake up waiters matching bitset queued on this futex (uaddr). */ -static int futex_wake(u32 __user *uaddr, int fshared, int nr_wake, u32 bitset) +static int +futex_wake(u32 __user *uaddr, unsigned int flags, int nr_wake, u32 bitset) { struct futex_hash_bucket *hb; struct futex_q *this, *next; @@ -889,7 +929,7 @@ 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, flags & FLAGS_SHARED, &key); if (unlikely(ret != 0)) goto out; @@ -915,7 +955,7 @@ static int futex_wake(u32 __user *uaddr, int fshared, int nr_wake, u32 bitset) } spin_unlock(&hb->lock); - put_futex_key(fshared, &key); + put_futex_key(&key); out: return ret; } @@ -925,7 +965,7 @@ out: * to this virtual address: */ static int -futex_wake_op(u32 __user *uaddr1, int fshared, u32 __user *uaddr2, +futex_wake_op(u32 __user *uaddr1, unsigned int flags, u32 __user *uaddr2, int nr_wake, int nr_wake2, int op) { union futex_key key1 = FUTEX_KEY_INIT, key2 = FUTEX_KEY_INIT; @@ -935,10 +975,10 @@ 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, flags & FLAGS_SHARED, &key1); if (unlikely(ret != 0)) goto out; - ret = get_futex_key(uaddr2, fshared, &key2); + ret = get_futex_key(uaddr2, flags & FLAGS_SHARED, &key2); if (unlikely(ret != 0)) goto out_put_key1; @@ -970,11 +1010,11 @@ retry_private: if (ret) goto out_put_keys; - if (!fshared) + if (!(flags & FLAGS_SHARED)) goto retry_private; - put_futex_key(fshared, &key2); - put_futex_key(fshared, &key1); + put_futex_key(&key2); + put_futex_key(&key1); goto retry; } @@ -1004,9 +1044,9 @@ retry_private: double_unlock_hb(hb1, hb2); out_put_keys: - put_futex_key(fshared, &key2); + put_futex_key(&key2); out_put_key1: - put_futex_key(fshared, &key1); + put_futex_key(&key1); out: return ret; } @@ -1140,12 +1180,14 @@ static int futex_proxy_trylock_atomic(u32 __user *pifutex, /** * 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) + * @uaddr1: source futex user address + * @flags: futex flags (FLAGS_SHARED, etc.) + * @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) + * @cmpval: @uaddr1 expected value (or %NULL) + * @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. @@ -1154,9 +1196,9 @@ static int futex_proxy_trylock_atomic(u32 __user *pifutex, * >=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 requeue_pi) +static int futex_requeue(u32 __user *uaddr1, unsigned int flags, + u32 __user *uaddr2, 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; @@ -1197,10 +1239,10 @@ retry: pi_state = NULL; } - ret = get_futex_key(uaddr1, fshared, &key1); + ret = get_futex_key(uaddr1, flags & FLAGS_SHARED, &key1); if (unlikely(ret != 0)) goto out; - ret = get_futex_key(uaddr2, fshared, &key2); + ret = get_futex_key(uaddr2, flags & FLAGS_SHARED, &key2); if (unlikely(ret != 0)) goto out_put_key1; @@ -1222,11 +1264,11 @@ retry_private: if (ret) goto out_put_keys; - if (!fshared) + if (!(flags & FLAGS_SHARED)) goto retry_private; - put_futex_key(fshared, &key2); - put_futex_key(fshared, &key1); + put_futex_key(&key2); + put_futex_key(&key1); goto retry; } if (curval != *cmpval) { @@ -1266,8 +1308,8 @@ retry_private: break; case -EFAULT: double_unlock_hb(hb1, hb2); - put_futex_key(fshared, &key2); - put_futex_key(fshared, &key1); + put_futex_key(&key2); + put_futex_key(&key1); ret = fault_in_user_writeable(uaddr2); if (!ret) goto retry; @@ -1275,8 +1317,8 @@ retry_private: case -EAGAIN: /* The owner was exiting, try again. */ double_unlock_hb(hb1, hb2); - put_futex_key(fshared, &key2); - put_futex_key(fshared, &key1); + put_futex_key(&key2); + put_futex_key(&key1); cond_resched(); goto retry; default: @@ -1358,9 +1400,9 @@ out_unlock: drop_futex_key_refs(&key1); out_put_keys: - put_futex_key(fshared, &key2); + put_futex_key(&key2); out_put_key1: - put_futex_key(fshared, &key1); + put_futex_key(&key1); out: if (pi_state != NULL) free_pi_state(pi_state); @@ -1369,10 +1411,10 @@ out: /* The key must be already stored in q->key. */ static inline struct futex_hash_bucket *queue_lock(struct futex_q *q) + __acquires(&hb->lock) { struct futex_hash_bucket *hb; - get_futex_key_refs(&q->key); hb = hash_futex(&q->key); q->lock_ptr = &hb->lock; @@ -1382,9 +1424,9 @@ static inline struct futex_hash_bucket *queue_lock(struct futex_q *q) static inline void queue_unlock(struct futex_q *q, struct futex_hash_bucket *hb) + __releases(&hb->lock) { spin_unlock(&hb->lock); - drop_futex_key_refs(&q->key); } /** @@ -1400,6 +1442,7 @@ queue_unlock(struct futex_q *q, struct futex_hash_bucket *hb) * an example). */ static inline void queue_me(struct futex_q *q, struct futex_hash_bucket *hb) + __releases(&hb->lock) { int prio; @@ -1480,6 +1523,7 @@ retry: * and dropped here. */ static void unqueue_me_pi(struct futex_q *q) + __releases(q->lock_ptr) { WARN_ON(plist_node_empty(&q->list)); plist_del(&q->list, &q->list.plist); @@ -1489,8 +1533,6 @@ static void unqueue_me_pi(struct futex_q *q) q->pi_state = NULL; spin_unlock(q->lock_ptr); - - drop_futex_key_refs(&q->key); } /* @@ -1500,7 +1542,7 @@ static void unqueue_me_pi(struct futex_q *q) * private futexes. */ static int fixup_pi_state_owner(u32 __user *uaddr, struct futex_q *q, - struct task_struct *newowner, int fshared) + struct task_struct *newowner) { u32 newtid = task_pid_vnr(newowner) | FUTEX_WAITERS; struct futex_pi_state *pi_state = q->pi_state; @@ -1593,20 +1635,11 @@ handle_fault: goto retry; } -/* - * In case we must use restart_block to restart a futex_wait, - * we encode in the 'flags' shared capability - */ -#define FLAGS_SHARED 0x01 -#define FLAGS_CLOCKRT 0x02 -#define FLAGS_HAS_TIMEOUT 0x04 - static long futex_wait_restart(struct restart_block *restart); /** * 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) * @@ -1619,8 +1652,7 @@ static long futex_wait_restart(struct restart_block *restart); * 0 - success, lock not taken * <0 - on error (-EFAULT) */ -static int fixup_owner(u32 __user *uaddr, int fshared, struct futex_q *q, - int locked) +static int fixup_owner(u32 __user *uaddr, struct futex_q *q, int locked) { struct task_struct *owner; int ret = 0; @@ -1631,7 +1663,7 @@ static int fixup_owner(u32 __user *uaddr, int fshared, struct futex_q *q, * 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); + ret = fixup_pi_state_owner(uaddr, q, current); goto out; } @@ -1658,7 +1690,7 @@ static int fixup_owner(u32 __user *uaddr, int fshared, struct futex_q *q, * lock. Fix the state up. */ owner = rt_mutex_owner(&q->pi_state->pi_mutex); - ret = fixup_pi_state_owner(uaddr, q, owner, fshared); + ret = fixup_pi_state_owner(uaddr, q, owner); goto out; } @@ -1721,7 +1753,7 @@ static void futex_wait_queue_me(struct futex_hash_bucket *hb, struct futex_q *q, * 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) + * @flags: futex flags (FLAGS_SHARED, etc.) * @q: the associated futex_q * @hb: storage for hash_bucket pointer to be returned to caller * @@ -1734,7 +1766,7 @@ static void futex_wait_queue_me(struct futex_hash_bucket *hb, struct futex_q *q, * 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, +static int futex_wait_setup(u32 __user *uaddr, u32 val, unsigned int flags, struct futex_q *q, struct futex_hash_bucket **hb) { u32 uval; @@ -1758,8 +1790,7 @@ static int futex_wait_setup(u32 __user *uaddr, u32 val, int fshared, * rare, but normal. */ retry: - q->key = FUTEX_KEY_INIT; - ret = get_futex_key(uaddr, fshared, &q->key); + ret = get_futex_key(uaddr, flags & FLAGS_SHARED, &q->key); if (unlikely(ret != 0)) return ret; @@ -1775,10 +1806,10 @@ retry_private: if (ret) goto out; - if (!fshared) + if (!(flags & FLAGS_SHARED)) goto retry_private; - put_futex_key(fshared, &q->key); + put_futex_key(&q->key); goto retry; } @@ -1789,40 +1820,40 @@ retry_private: out: if (ret) - put_futex_key(fshared, &q->key); + put_futex_key(&q->key); return ret; } -static int futex_wait(u32 __user *uaddr, int fshared, - u32 val, ktime_t *abs_time, u32 bitset, int clockrt) +static int futex_wait(u32 __user *uaddr, unsigned int flags, u32 val, + ktime_t *abs_time, u32 bitset) { struct hrtimer_sleeper timeout, *to = NULL; struct restart_block *restart; struct futex_hash_bucket *hb; - struct futex_q q; + struct futex_q q = futex_q_init; int ret; if (!bitset) return -EINVAL; - - q.pi_state = NULL; q.bitset = bitset; - q.rt_waiter = NULL; - q.requeue_pi_key = NULL; if (abs_time) { to = &timeout; - hrtimer_init_on_stack(&to->timer, clockrt ? CLOCK_REALTIME : - CLOCK_MONOTONIC, HRTIMER_MODE_ABS); + hrtimer_init_on_stack(&to->timer, (flags & FLAGS_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); } retry: - /* Prepare to wait on uaddr. */ - ret = futex_wait_setup(uaddr, val, fshared, &q, &hb); + /* + * Prepare to wait on uaddr. On success, holds hb lock and increments + * q.key refs. + */ + ret = futex_wait_setup(uaddr, val, flags, &q, &hb); if (ret) goto out; @@ -1831,42 +1862,34 @@ retry: /* If we were woken (and unqueued), we succeeded, whatever. */ ret = 0; + /* unqueue_me() drops q.key ref */ if (!unqueue_me(&q)) - goto out_put_key; + goto out; ret = -ETIMEDOUT; if (to && !to->task) - goto out_put_key; + goto out; /* * 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); + if (!signal_pending(current)) goto retry; - } ret = -ERESTARTSYS; if (!abs_time) - goto out_put_key; + goto out; restart = ¤t_thread_info()->restart_block; restart->fn = futex_wait_restart; - restart->futex.uaddr = (u32 *)uaddr; + restart->futex.uaddr = uaddr; restart->futex.val = val; restart->futex.time = abs_time->tv64; restart->futex.bitset = bitset; - restart->futex.flags = FLAGS_HAS_TIMEOUT; - - if (fshared) - restart->futex.flags |= FLAGS_SHARED; - if (clockrt) - restart->futex.flags |= FLAGS_CLOCKRT; + restart->futex.flags = flags; ret = -ERESTART_RESTARTBLOCK; -out_put_key: - put_futex_key(fshared, &q.key); out: if (to) { hrtimer_cancel(&to->timer); @@ -1878,8 +1901,7 @@ out: static long futex_wait_restart(struct restart_block *restart) { - u32 __user *uaddr = (u32 __user *)restart->futex.uaddr; - int fshared = 0; + u32 __user *uaddr = restart->futex.uaddr; ktime_t t, *tp = NULL; if (restart->futex.flags & FLAGS_HAS_TIMEOUT) { @@ -1887,11 +1909,9 @@ static long futex_wait_restart(struct restart_block *restart) 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, tp, - restart->futex.bitset, - restart->futex.flags & FLAGS_CLOCKRT); + + return (long)futex_wait(uaddr, restart->futex.flags, + restart->futex.val, tp, restart->futex.bitset); } @@ -1901,12 +1921,12 @@ static long futex_wait_restart(struct restart_block *restart) * if there are waiters then it will block, it does PI, etc. (Due to * races the kernel might see a 0 value of the futex too.) */ -static int futex_lock_pi(u32 __user *uaddr, int fshared, - int detect, ktime_t *time, int trylock) +static int futex_lock_pi(u32 __user *uaddr, unsigned int flags, int detect, + ktime_t *time, int trylock) { struct hrtimer_sleeper timeout, *to = NULL; struct futex_hash_bucket *hb; - struct futex_q q; + struct futex_q q = futex_q_init; int res, ret; if (refill_pi_state_cache()) @@ -1920,12 +1940,8 @@ static int futex_lock_pi(u32 __user *uaddr, int fshared, hrtimer_set_expires(&to->timer, *time); } - 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, flags & FLAGS_SHARED, &q.key); if (unlikely(ret != 0)) goto out; @@ -1947,7 +1963,7 @@ retry_private: * exit to complete. */ queue_unlock(&q, hb); - put_futex_key(fshared, &q.key); + put_futex_key(&q.key); cond_resched(); goto retry; default: @@ -1977,7 +1993,7 @@ retry_private: * Fixup the pi_state owner and possibly acquire the lock if we * haven't already. */ - res = fixup_owner(uaddr, fshared, &q, !ret); + res = fixup_owner(uaddr, &q, !ret); /* * If fixup_owner() returned an error, proprogate that. If it acquired * the lock, clear our -ETIMEDOUT or -EINTR. @@ -2001,7 +2017,7 @@ out_unlock_put_key: queue_unlock(&q, hb); out_put_key: - put_futex_key(fshared, &q.key); + put_futex_key(&q.key); out: if (to) destroy_hrtimer_on_stack(&to->timer); @@ -2014,10 +2030,10 @@ uaddr_faulted: if (ret) goto out_put_key; - if (!fshared) + if (!(flags & FLAGS_SHARED)) goto retry_private; - put_futex_key(fshared, &q.key); + put_futex_key(&q.key); goto retry; } @@ -2026,7 +2042,7 @@ uaddr_faulted: * This is the in-kernel slowpath: we look up the PI state (if any), * and do the rt-mutex unlock. */ -static int futex_unlock_pi(u32 __user *uaddr, int fshared) +static int futex_unlock_pi(u32 __user *uaddr, unsigned int flags) { struct futex_hash_bucket *hb; struct futex_q *this, *next; @@ -2044,7 +2060,7 @@ retry: if ((uval & FUTEX_TID_MASK) != task_pid_vnr(current)) return -EPERM; - ret = get_futex_key(uaddr, fshared, &key); + ret = get_futex_key(uaddr, flags & FLAGS_SHARED, &key); if (unlikely(ret != 0)) goto out; @@ -2099,14 +2115,14 @@ retry: out_unlock: spin_unlock(&hb->lock); - put_futex_key(fshared, &key); + put_futex_key(&key); out: return ret; pi_faulted: spin_unlock(&hb->lock); - put_futex_key(fshared, &key); + put_futex_key(&key); ret = fault_in_user_writeable(uaddr); if (!ret) @@ -2166,7 +2182,7 @@ int handle_early_requeue_pi_wakeup(struct futex_hash_bucket *hb, /** * 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 + * @flags: futex flags (FLAGS_SHARED, FLAGS_CLOCKRT, etc.), they must be * the same type, no requeueing from private to shared, etc. * @val: the expected value of uaddr * @abs_time: absolute timeout @@ -2204,16 +2220,16 @@ int handle_early_requeue_pi_wakeup(struct futex_hash_bucket *hb, * 0 - On success * <0 - On error */ -static int futex_wait_requeue_pi(u32 __user *uaddr, int fshared, +static int futex_wait_requeue_pi(u32 __user *uaddr, unsigned int flags, u32 val, ktime_t *abs_time, u32 bitset, - int clockrt, u32 __user *uaddr2) + 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; + union futex_key key2 = FUTEX_KEY_INIT; + struct futex_q q = futex_q_init; int res, ret; if (!bitset) @@ -2221,8 +2237,9 @@ static int futex_wait_requeue_pi(u32 __user *uaddr, int fshared, if (abs_time) { to = &timeout; - hrtimer_init_on_stack(&to->timer, clockrt ? CLOCK_REALTIME : - CLOCK_MONOTONIC, HRTIMER_MODE_ABS); + hrtimer_init_on_stack(&to->timer, (flags & FLAGS_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); @@ -2235,18 +2252,19 @@ static int futex_wait_requeue_pi(u32 __user *uaddr, int fshared, debug_rt_mutex_init_waiter(&rt_waiter); rt_waiter.task = NULL; - key2 = FUTEX_KEY_INIT; - ret = get_futex_key(uaddr2, fshared, &key2); + ret = get_futex_key(uaddr2, flags & FLAGS_SHARED, &key2); 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); + /* + * Prepare to wait on uaddr. On success, increments q.key (key1) ref + * count. + */ + ret = futex_wait_setup(uaddr, val, flags, &q, &hb); if (ret) goto out_key2; @@ -2263,7 +2281,9 @@ static int futex_wait_requeue_pi(u32 __user *uaddr, int fshared, * 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. + * race with the atomic proxy lock acquisition by the requeue code. The + * futex_requeue dropped our key1 reference and incremented our key2 + * reference count. */ /* Check if the requeue code acquired the second futex for us. */ @@ -2274,8 +2294,7 @@ static int futex_wait_requeue_pi(u32 __user *uaddr, int fshared, */ if (q.pi_state && (q.pi_state->owner != current)) { spin_lock(q.lock_ptr); - ret = fixup_pi_state_owner(uaddr2, &q, current, - fshared); + ret = fixup_pi_state_owner(uaddr2, &q, current); spin_unlock(q.lock_ptr); } } else { @@ -2294,7 +2313,7 @@ static int futex_wait_requeue_pi(u32 __user *uaddr, int fshared, * Fixup the pi_state owner and possibly acquire the lock if we * haven't already. */ - res = fixup_owner(uaddr2, fshared, &q, !ret); + res = fixup_owner(uaddr2, &q, !ret); /* * If fixup_owner() returned an error, proprogate that. If it * acquired the lock, clear -ETIMEDOUT or -EINTR. @@ -2325,9 +2344,9 @@ static int futex_wait_requeue_pi(u32 __user *uaddr, int fshared, } out_put_keys: - put_futex_key(fshared, &q.key); + put_futex_key(&q.key); out_key2: - put_futex_key(fshared, &key2); + put_futex_key(&key2); out: if (to) { @@ -2467,7 +2486,7 @@ retry: */ static inline int fetch_robust_entry(struct robust_list __user **entry, struct robust_list __user * __user *head, - int *pi) + unsigned int *pi) { unsigned long uentry; @@ -2490,7 +2509,8 @@ void exit_robust_list(struct task_struct *curr) { struct robust_list_head __user *head = curr->robust_list; struct robust_list __user *entry, *next_entry, *pending; - unsigned int limit = ROBUST_LIST_LIMIT, pi, next_pi, pip; + unsigned int limit = ROBUST_LIST_LIMIT, pi, pip; + unsigned int uninitialized_var(next_pi); unsigned long futex_offset; int rc; @@ -2551,58 +2571,57 @@ void exit_robust_list(struct task_struct *curr) long do_futex(u32 __user *uaddr, int op, u32 val, ktime_t *timeout, u32 __user *uaddr2, u32 val2, u32 val3) { - int clockrt, ret = -ENOSYS; - int cmd = op & FUTEX_CMD_MASK; - int fshared = 0; + int ret = -ENOSYS, cmd = op & FUTEX_CMD_MASK; + unsigned int flags = 0; if (!(op & FUTEX_PRIVATE_FLAG)) - fshared = 1; + flags |= FLAGS_SHARED; - clockrt = op & FUTEX_CLOCK_REALTIME; - if (clockrt && cmd != FUTEX_WAIT_BITSET && cmd != FUTEX_WAIT_REQUEUE_PI) - return -ENOSYS; + if (op & FUTEX_CLOCK_REALTIME) { + flags |= FLAGS_CLOCKRT; + if (cmd != FUTEX_WAIT_BITSET && cmd != FUTEX_WAIT_REQUEUE_PI) + return -ENOSYS; + } switch (cmd) { case FUTEX_WAIT: val3 = FUTEX_BITSET_MATCH_ANY; case FUTEX_WAIT_BITSET: - ret = futex_wait(uaddr, fshared, val, timeout, val3, clockrt); + ret = futex_wait(uaddr, flags, val, timeout, val3); break; case FUTEX_WAKE: val3 = FUTEX_BITSET_MATCH_ANY; case FUTEX_WAKE_BITSET: - ret = futex_wake(uaddr, fshared, val, val3); + ret = futex_wake(uaddr, flags, val, val3); break; case FUTEX_REQUEUE: - ret = futex_requeue(uaddr, fshared, uaddr2, val, val2, NULL, 0); + ret = futex_requeue(uaddr, flags, uaddr2, val, val2, NULL, 0); break; case FUTEX_CMP_REQUEUE: - ret = futex_requeue(uaddr, fshared, uaddr2, val, val2, &val3, - 0); + ret = futex_requeue(uaddr, flags, uaddr2, val, val2, &val3, 0); break; case FUTEX_WAKE_OP: - ret = futex_wake_op(uaddr, fshared, uaddr2, val, val2, val3); + ret = futex_wake_op(uaddr, flags, uaddr2, val, val2, val3); break; case FUTEX_LOCK_PI: if (futex_cmpxchg_enabled) - ret = futex_lock_pi(uaddr, fshared, val, timeout, 0); + ret = futex_lock_pi(uaddr, flags, val, timeout, 0); break; case FUTEX_UNLOCK_PI: if (futex_cmpxchg_enabled) - ret = futex_unlock_pi(uaddr, fshared); + ret = futex_unlock_pi(uaddr, flags); break; case FUTEX_TRYLOCK_PI: if (futex_cmpxchg_enabled) - ret = futex_lock_pi(uaddr, fshared, 0, timeout, 1); + ret = futex_lock_pi(uaddr, flags, 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); + ret = futex_wait_requeue_pi(uaddr, flags, val, timeout, val3, + uaddr2); break; case FUTEX_CMP_REQUEUE_PI: - ret = futex_requeue(uaddr, fshared, uaddr2, val, val2, &val3, - 1); + ret = futex_requeue(uaddr, flags, uaddr2, val, val2, &val3, 1); break; default: ret = -ENOSYS; @@ -2656,7 +2675,7 @@ static int __init futex_init(void) * of the complex code paths. Also we want to prevent * registration of robust lists in that case. NULL is * guaranteed to fault and we get -EFAULT on functional - * implementation, the non functional ones will return + * implementation, the non-functional ones will return * -ENOSYS. */ curval = cmpxchg_futex_value_locked(NULL, 0, 0); diff --git a/kernel/futex_compat.c b/kernel/futex_compat.c index d49afb2395e..a7934ac75e5 100644 --- a/kernel/futex_compat.c +++ b/kernel/futex_compat.c @@ -19,7 +19,7 @@ */ static inline int fetch_robust_entry(compat_uptr_t *uentry, struct robust_list __user **entry, - compat_uptr_t __user *head, int *pi) + compat_uptr_t __user *head, unsigned int *pi) { if (get_user(*uentry, head)) return -EFAULT; @@ -49,7 +49,8 @@ void compat_exit_robust_list(struct task_struct *curr) { struct compat_robust_list_head __user *head = curr->compat_robust_list; struct robust_list __user *entry, *next_entry, *pending; - unsigned int limit = ROBUST_LIST_LIMIT, pi, next_pi, pip; + unsigned int limit = ROBUST_LIST_LIMIT, pi, pip; + unsigned int uninitialized_var(next_pi); compat_uptr_t uentry, next_uentry, upending; compat_long_t futex_offset; int rc; diff --git a/kernel/gcov/fs.c b/kernel/gcov/fs.c index ef3c3f88a7a..9bd0934f6c3 100644 --- a/kernel/gcov/fs.c +++ b/kernel/gcov/fs.c @@ -33,10 +33,11 @@ * @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". + * @loaded_info: array of pointers to profiling data sets for loaded object + * files. + * @num_loaded: number of profiling data sets for loaded object files. + * @unloaded_info: accumulated copy of profiling data sets for unloaded + * object files. Used only when gcov_persist=1. * @dentry: main debugfs entry, either a directory or data file * @links: associated symbolic links * @name: data file basename @@ -51,10 +52,11 @@ struct gcov_node { struct list_head children; struct list_head all; struct gcov_node *parent; - struct gcov_info *info; - struct gcov_info *ghost; + struct gcov_info **loaded_info; + struct gcov_info *unloaded_info; struct dentry *dentry; struct dentry **links; + int num_loaded; char name[0]; }; @@ -136,16 +138,37 @@ static const struct seq_operations gcov_seq_ops = { }; /* - * 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. + * Return a profiling data set associated with the given node. This is + * either a data set for a loaded object file or a data set copy in case + * all associated object files have been unloaded. */ static struct gcov_info *get_node_info(struct gcov_node *node) { - if (node->info) - return node->info; + if (node->num_loaded > 0) + return node->loaded_info[0]; - return node->ghost; + return node->unloaded_info; +} + +/* + * Return a newly allocated profiling data set which contains the sum of + * all profiling data associated with the given node. + */ +static struct gcov_info *get_accumulated_info(struct gcov_node *node) +{ + struct gcov_info *info; + int i = 0; + + if (node->unloaded_info) + info = gcov_info_dup(node->unloaded_info); + else + info = gcov_info_dup(node->loaded_info[i++]); + if (!info) + return NULL; + for (; i < node->num_loaded; i++) + gcov_info_add(info, node->loaded_info[i]); + + return info; } /* @@ -163,9 +186,10 @@ static int gcov_seq_open(struct inode *inode, struct file *file) mutex_lock(&node_lock); /* * Read from a profiling data copy to minimize reference tracking - * complexity and concurrent access. + * complexity and concurrent access and to keep accumulating multiple + * profiling data sets associated with one node simple. */ - info = gcov_info_dup(get_node_info(node)); + info = get_accumulated_info(node); if (!info) goto out_unlock; iter = gcov_iter_new(info); @@ -225,12 +249,25 @@ static struct gcov_node *get_node_by_name(const char *name) return NULL; } +/* + * Reset all profiling data associated with the specified node. + */ +static void reset_node(struct gcov_node *node) +{ + int i; + + if (node->unloaded_info) + gcov_info_reset(node->unloaded_info); + for (i = 0; i < node->num_loaded; i++) + gcov_info_reset(node->loaded_info[i]); +} + 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. + * corresponding file. If all associated object files have been unloaded, + * 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) @@ -245,10 +282,10 @@ static ssize_t gcov_seq_write(struct file *file, const char __user *addr, node = get_node_by_name(info->filename); if (node) { /* Reset counts or remove node for unloaded modules. */ - if (node->ghost) + if (node->num_loaded == 0) remove_node(node); else - gcov_info_reset(node->info); + reset_node(node); } /* Reset counts for open file. */ gcov_info_reset(info); @@ -378,7 +415,10 @@ static void init_node(struct gcov_node *node, struct gcov_info *info, INIT_LIST_HEAD(&node->list); INIT_LIST_HEAD(&node->children); INIT_LIST_HEAD(&node->all); - node->info = info; + if (node->loaded_info) { + node->loaded_info[0] = info; + node->num_loaded = 1; + } node->parent = parent; if (name) strcpy(node->name, name); @@ -394,9 +434,13 @@ static struct gcov_node *new_node(struct gcov_node *parent, 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; + if (!node) + goto err_nomem; + if (info) { + node->loaded_info = kcalloc(1, sizeof(struct gcov_info *), + GFP_KERNEL); + if (!node->loaded_info) + goto err_nomem; } init_node(node, info, name, parent); /* Differentiate between gcov data file nodes and directory nodes. */ @@ -416,6 +460,11 @@ static struct gcov_node *new_node(struct gcov_node *parent, list_add(&node->all, &all_head); return node; + +err_nomem: + kfree(node); + pr_warning("out of memory\n"); + return NULL; } /* Remove symbolic links associated with node. */ @@ -441,8 +490,9 @@ static void release_node(struct gcov_node *node) list_del(&node->all); debugfs_remove(node->dentry); remove_links(node); - if (node->ghost) - gcov_info_free(node->ghost); + kfree(node->loaded_info); + if (node->unloaded_info) + gcov_info_free(node->unloaded_info); kfree(node); } @@ -477,7 +527,7 @@ static struct gcov_node *get_child_by_name(struct gcov_node *parent, /* * write() implementation for reset file. Reset all profiling data to zero - * and remove ghost nodes. + * and remove nodes for which all associated object files are unloaded. */ static ssize_t reset_write(struct file *file, const char __user *addr, size_t len, loff_t *pos) @@ -487,8 +537,8 @@ static ssize_t reset_write(struct file *file, const char __user *addr, mutex_lock(&node_lock); restart: list_for_each_entry(node, &all_head, all) { - if (node->info) - gcov_info_reset(node->info); + if (node->num_loaded > 0) + reset_node(node); else if (list_empty(&node->children)) { remove_node(node); /* Several nodes may have gone - restart loop. */ @@ -511,6 +561,7 @@ static ssize_t reset_read(struct file *file, char __user *addr, size_t len, static const struct file_operations gcov_reset_fops = { .write = reset_write, .read = reset_read, + .llseek = noop_llseek, }; /* @@ -564,37 +615,115 @@ err_remove: } /* - * 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". + * Associate a profiling data set with an existing node. Needs to be called + * with node_lock held. */ -static int ghost_node(struct gcov_node *node) +static void add_info(struct gcov_node *node, struct gcov_info *info) { - 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; + struct gcov_info **loaded_info; + int num = node->num_loaded; + + /* + * Prepare new array. This is done first to simplify cleanup in + * case the new data set is incompatible, the node only contains + * unloaded data sets and there's not enough memory for the array. + */ + loaded_info = kcalloc(num + 1, sizeof(struct gcov_info *), GFP_KERNEL); + if (!loaded_info) { + pr_warning("could not add '%s' (out of memory)\n", + info->filename); + return; + } + memcpy(loaded_info, node->loaded_info, + num * sizeof(struct gcov_info *)); + loaded_info[num] = info; + /* Check if the new data set is compatible. */ + if (num == 0) { + /* + * A module was unloaded, modified and reloaded. The new + * data set replaces the copy of the last one. + */ + if (!gcov_info_is_compatible(node->unloaded_info, info)) { + pr_warning("discarding saved data for %s " + "(incompatible version)\n", info->filename); + gcov_info_free(node->unloaded_info); + node->unloaded_info = NULL; + } + } else { + /* + * Two different versions of the same object file are loaded. + * The initial one takes precedence. + */ + if (!gcov_info_is_compatible(node->loaded_info[0], info)) { + pr_warning("could not add '%s' (incompatible " + "version)\n", info->filename); + kfree(loaded_info); + return; + } } - node->info = NULL; + /* Overwrite previous array. */ + kfree(node->loaded_info); + node->loaded_info = loaded_info; + node->num_loaded = num + 1; +} - return 0; +/* + * Return the index of a profiling data set associated with a node. + */ +static int get_info_index(struct gcov_node *node, struct gcov_info *info) +{ + int i; + + for (i = 0; i < node->num_loaded; i++) { + if (node->loaded_info[i] == info) + return i; + } + return -ENOENT; } /* - * Profiling data for this node has been loaded again. Add profiling data - * from previous instantiation and turn this node into a regular node. + * Save the data of a profiling data set which is being unloaded. */ -static void revive_node(struct gcov_node *node, struct gcov_info *info) +static void save_info(struct gcov_node *node, struct gcov_info *info) { - if (gcov_info_is_compatible(node->ghost, info)) - gcov_info_add(info, node->ghost); + if (node->unloaded_info) + gcov_info_add(node->unloaded_info, info); else { - pr_warning("discarding saved data for '%s' (version changed)\n", + node->unloaded_info = gcov_info_dup(info); + if (!node->unloaded_info) { + pr_warning("could not save data for '%s' " + "(out of memory)\n", info->filename); + } + } +} + +/* + * Disassociate a profiling data set from a node. Needs to be called with + * node_lock held. + */ +static void remove_info(struct gcov_node *node, struct gcov_info *info) +{ + int i; + + i = get_info_index(node, info); + if (i < 0) { + pr_warning("could not remove '%s' (not found)\n", info->filename); + return; } - gcov_info_free(node->ghost); - node->ghost = NULL; - node->info = info; + if (gcov_persist) + save_info(node, info); + /* Shrink array. */ + node->loaded_info[i] = node->loaded_info[node->num_loaded - 1]; + node->num_loaded--; + if (node->num_loaded > 0) + return; + /* Last loaded data set was removed. */ + kfree(node->loaded_info); + node->loaded_info = NULL; + node->num_loaded = 0; + if (!node->unloaded_info) + remove_node(node); } /* @@ -609,30 +738,18 @@ void gcov_event(enum gcov_action action, struct gcov_info *info) node = get_node_by_name(info->filename); switch (action) { case GCOV_ADD: - /* Add new node or revive ghost. */ - if (!node) { + if (node) + add_info(node, info); + else 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) { + if (node) + remove_info(node, info); + else { 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); diff --git a/kernel/groups.c b/kernel/groups.c index 53b1916c949..253dc0f35cf 100644 --- a/kernel/groups.c +++ b/kernel/groups.c @@ -143,10 +143,9 @@ int groups_search(const struct group_info *group_info, gid_t grp) right = group_info->ngroups; while (left < right) { unsigned int mid = (left+right)/2; - int cmp = grp - GROUP_AT(group_info, mid); - if (cmp > 0) + if (grp > GROUP_AT(group_info, mid)) left = mid + 1; - else if (cmp < 0) + else if (grp < GROUP_AT(group_info, mid)) right = mid; else return 1; diff --git a/kernel/hrtimer.c b/kernel/hrtimer.c index b9b134b3508..0c8d7c04861 100644 --- a/kernel/hrtimer.c +++ b/kernel/hrtimer.c @@ -89,8 +89,8 @@ static void hrtimer_get_softirq_time(struct hrtimer_cpu_base *base) do { seq = read_seqbegin(&xtime_lock); - xts = current_kernel_time(); - tom = wall_to_monotonic; + xts = __current_kernel_time(); + tom = __get_wall_to_monotonic(); } while (read_seqretry(&xtime_lock, seq)); xtim = timespec_to_ktime(xts); @@ -144,12 +144,8 @@ struct hrtimer_clock_base *lock_hrtimer_base(const struct hrtimer *timer, 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; - } + if (!pinned && get_sysctl_timer_migration() && idle_cpu(this_cpu)) + return get_nohz_timer_target(); #endif return this_cpu; } @@ -501,7 +497,7 @@ static inline int hrtimer_is_hres_enabled(void) */ static inline int hrtimer_hres_active(void) { - return __get_cpu_var(hrtimer_bases).hres_active; + return __this_cpu_read(hrtimer_bases.hres_active); } /* @@ -520,10 +516,13 @@ hrtimer_force_reprogram(struct hrtimer_cpu_base *cpu_base, int skip_equal) for (i = 0; i < HRTIMER_MAX_CLOCK_BASES; i++, base++) { struct hrtimer *timer; + struct timerqueue_node *next; - if (!base->first) + next = timerqueue_getnext(&base->active); + if (!next) continue; - timer = rb_entry(base->first, struct hrtimer, node); + timer = container_of(next, struct hrtimer, node); + expires = ktime_sub(hrtimer_get_expires(timer), base->offset); /* * clock_was_set() has changed base->offset so the @@ -612,7 +611,7 @@ static int hrtimer_reprogram(struct hrtimer *timer, static void retrigger_next_event(void *arg) { struct hrtimer_cpu_base *base; - struct timespec realtime_offset; + struct timespec realtime_offset, wtm; unsigned long seq; if (!hrtimer_hres_active()) @@ -620,10 +619,9 @@ static void retrigger_next_event(void *arg) do { seq = read_seqbegin(&xtime_lock); - set_normalized_timespec(&realtime_offset, - -wall_to_monotonic.tv_sec, - -wall_to_monotonic.tv_nsec); + wtm = __get_wall_to_monotonic(); } while (read_seqretry(&xtime_lock, seq)); + set_normalized_timespec(&realtime_offset, -wtm.tv_sec, -wtm.tv_nsec); base = &__get_cpu_var(hrtimer_bases); @@ -845,48 +843,17 @@ EXPORT_SYMBOL_GPL(hrtimer_forward); static int enqueue_hrtimer(struct hrtimer *timer, struct hrtimer_clock_base *base) { - struct rb_node **link = &base->active.rb_node; - struct rb_node *parent = NULL; - struct hrtimer *entry; - int leftmost = 1; - debug_activate(timer); - /* - * Find the right place in the rbtree: - */ - while (*link) { - parent = *link; - entry = rb_entry(parent, struct hrtimer, node); - /* - * We dont care about collisions. Nodes with - * the same expiry time stay together. - */ - if (hrtimer_get_expires_tv64(timer) < - hrtimer_get_expires_tv64(entry)) { - link = &(*link)->rb_left; - } else { - link = &(*link)->rb_right; - leftmost = 0; - } - } - - /* - * Insert the timer to the rbtree and check whether it - * replaces the first pending timer - */ - if (leftmost) - base->first = &timer->node; + timerqueue_add(&base->active, &timer->node); - rb_link_node(&timer->node, parent, link); - rb_insert_color(&timer->node, &base->active); /* * HRTIMER_STATE_ENQUEUED is or'ed to the current state to preserve the * state of a possibly running callback. */ timer->state |= HRTIMER_STATE_ENQUEUED; - return leftmost; + return (&timer->node == base->active.next); } /* @@ -906,12 +873,7 @@ static void __remove_hrtimer(struct hrtimer *timer, 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); + if (&timer->node == timerqueue_getnext(&base->active)) { #ifdef CONFIG_HIGH_RES_TIMERS /* Reprogram the clock event device. if enabled */ if (reprogram && hrtimer_hres_active()) { @@ -924,7 +886,7 @@ static void __remove_hrtimer(struct hrtimer *timer, } #endif } - rb_erase(&timer->node, &base->active); + timerqueue_del(&base->active, &timer->node); out: timer->state = newstate; } @@ -936,6 +898,7 @@ static inline int remove_hrtimer(struct hrtimer *timer, struct hrtimer_clock_base *base) { if (hrtimer_is_queued(timer)) { + unsigned long state; int reprogram; /* @@ -949,8 +912,13 @@ remove_hrtimer(struct hrtimer *timer, struct hrtimer_clock_base *base) 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, - reprogram); + /* + * We must preserve the CALLBACK state flag here, + * otherwise we could move the timer base in + * switch_hrtimer_base. + */ + state = timer->state & HRTIMER_STATE_CALLBACK; + __remove_hrtimer(timer, base, state, reprogram); return 1; } return 0; @@ -1096,11 +1064,10 @@ EXPORT_SYMBOL_GPL(hrtimer_cancel); */ ktime_t hrtimer_get_remaining(const struct hrtimer *timer) { - struct hrtimer_clock_base *base; unsigned long flags; ktime_t rem; - base = lock_hrtimer_base(timer, &flags); + lock_hrtimer_base(timer, &flags); rem = hrtimer_expires_remaining(timer); unlock_hrtimer_base(timer, &flags); @@ -1128,11 +1095,13 @@ ktime_t hrtimer_get_next_event(void) if (!hrtimer_hres_active()) { for (i = 0; i < HRTIMER_MAX_CLOCK_BASES; i++, base++) { struct hrtimer *timer; + struct timerqueue_node *next; - if (!base->first) + next = timerqueue_getnext(&base->active); + if (!next) continue; - timer = rb_entry(base->first, struct hrtimer, node); + timer = container_of(next, struct hrtimer, node); delta.tv64 = hrtimer_get_expires_tv64(timer); delta = ktime_sub(delta, base->get_time()); if (delta.tv64 < mindelta.tv64) @@ -1162,6 +1131,7 @@ static void __hrtimer_init(struct hrtimer *timer, clockid_t clock_id, timer->base = &cpu_base->clock_base[clock_id]; hrtimer_init_timer_hres(timer); + timerqueue_init(&timer->node); #ifdef CONFIG_TIMER_STATS timer->start_site = NULL; @@ -1237,6 +1207,9 @@ static void __run_hrtimer(struct hrtimer *timer, ktime_t *now) BUG_ON(timer->state != HRTIMER_STATE_CALLBACK); enqueue_hrtimer(timer, base); } + + WARN_ON_ONCE(!(timer->state & HRTIMER_STATE_CALLBACK)); + timer->state &= ~HRTIMER_STATE_CALLBACK; } @@ -1275,14 +1248,14 @@ retry: for (i = 0; i < HRTIMER_MAX_CLOCK_BASES; i++) { ktime_t basenow; - struct rb_node *node; + struct timerqueue_node *node; basenow = ktime_add(now, base->offset); - while ((node = base->first)) { + while ((node = timerqueue_getnext(&base->active))) { struct hrtimer *timer; - timer = rb_entry(node, struct hrtimer, node); + timer = container_of(node, struct hrtimer, node); /* * The immediate goal for using the softexpires is @@ -1438,7 +1411,7 @@ void hrtimer_run_pending(void) */ void hrtimer_run_queues(void) { - struct rb_node *node; + struct timerqueue_node *node; struct hrtimer_cpu_base *cpu_base = &__get_cpu_var(hrtimer_bases); struct hrtimer_clock_base *base; int index, gettime = 1; @@ -1448,8 +1421,7 @@ void hrtimer_run_queues(void) for (index = 0; index < HRTIMER_MAX_CLOCK_BASES; index++) { base = &cpu_base->clock_base[index]; - - if (!base->first) + if (!timerqueue_getnext(&base->active)) continue; if (gettime) { @@ -1459,10 +1431,10 @@ void hrtimer_run_queues(void) raw_spin_lock(&cpu_base->lock); - while ((node = base->first)) { + while ((node = timerqueue_getnext(&base->active))) { struct hrtimer *timer; - timer = rb_entry(node, struct hrtimer, node); + timer = container_of(node, struct hrtimer, node); if (base->softirq_time.tv64 <= hrtimer_get_expires_tv64(timer)) break; @@ -1627,8 +1599,10 @@ static void __cpuinit init_hrtimers_cpu(int cpu) raw_spin_lock_init(&cpu_base->lock); - for (i = 0; i < HRTIMER_MAX_CLOCK_BASES; i++) + for (i = 0; i < HRTIMER_MAX_CLOCK_BASES; i++) { cpu_base->clock_base[i].cpu_base = cpu_base; + timerqueue_init_head(&cpu_base->clock_base[i].active); + } hrtimer_init_hres(cpu_base); } @@ -1639,10 +1613,10 @@ static void migrate_hrtimer_list(struct hrtimer_clock_base *old_base, struct hrtimer_clock_base *new_base) { struct hrtimer *timer; - struct rb_node *node; + struct timerqueue_node *node; - while ((node = rb_first(&old_base->active))) { - timer = rb_entry(node, struct hrtimer, node); + while ((node = timerqueue_getnext(&old_base->active))) { + timer = container_of(node, struct hrtimer, node); BUG_ON(hrtimer_callback_running(timer)); debug_deactivate(timer); @@ -1771,7 +1745,7 @@ schedule_hrtimeout_range_clock(ktime_t *expires, unsigned long delta, } /* - * A NULL parameter means "inifinte" + * A NULL parameter means "infinite" */ if (!expires) { schedule(); diff --git a/kernel/hung_task.c b/kernel/hung_task.c index 0c642d51aac..53ead174da2 100644 --- a/kernel/hung_task.c +++ b/kernel/hung_task.c @@ -98,7 +98,7 @@ static void check_hung_task(struct task_struct *t, unsigned long 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); + debug_show_held_locks(t); touch_nmi_watchdog(); @@ -111,7 +111,7 @@ static void check_hung_task(struct task_struct *t, unsigned long timeout) * 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. + * to exit the grace period. For classic RCU, a reschedule is required. */ static void rcu_lock_break(struct task_struct *g, struct task_struct *t) { diff --git a/kernel/hw_breakpoint.c b/kernel/hw_breakpoint.c index 7a56b22e060..086adf25a55 100644 --- a/kernel/hw_breakpoint.c +++ b/kernel/hw_breakpoint.c @@ -41,6 +41,7 @@ #include <linux/sched.h> #include <linux/init.h> #include <linux/slab.h> +#include <linux/list.h> #include <linux/cpu.h> #include <linux/smp.h> @@ -62,6 +63,9 @@ static DEFINE_PER_CPU(unsigned int, nr_bp_flexible[TYPE_MAX]); static int nr_slots[TYPE_MAX]; +/* Keep track of the breakpoints attached to tasks */ +static LIST_HEAD(bp_task_head); + static int constraints_initialized; /* Gather the number of total pinned and un-pinned bp in a cpuset */ @@ -103,33 +107,21 @@ static unsigned int max_task_bp_pinned(int cpu, enum bp_type_idx type) return 0; } -static int task_bp_pinned(struct task_struct *tsk, enum bp_type_idx type) +/* + * Count the number of breakpoints of the same type and same task. + * The given event must be not on the list. + */ +static int task_bp_pinned(struct perf_event *bp, enum bp_type_idx type) { - struct perf_event_context *ctx = tsk->perf_event_ctxp; - struct list_head *list; - struct perf_event *bp; - unsigned long flags; + struct task_struct *tsk = bp->hw.bp_target; + struct perf_event *iter; int count = 0; - if (WARN_ONCE(!ctx, "No perf context for this task")) - return 0; - - list = &ctx->event_list; - - raw_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) - if (find_slot_idx(bp) == type) - count += hw_breakpoint_weight(bp); + list_for_each_entry(iter, &bp_task_head, hw.bp_list) { + if (iter->hw.bp_target == tsk && find_slot_idx(iter) == type) + count += hw_breakpoint_weight(iter); } - raw_spin_unlock_irqrestore(&ctx->lock, flags); - return count; } @@ -142,14 +134,14 @@ fetch_bp_busy_slots(struct bp_busy_slots *slots, struct perf_event *bp, enum bp_type_idx type) { int cpu = bp->cpu; - struct task_struct *tsk = bp->ctx->task; + struct task_struct *tsk = bp->hw.bp_target; if (cpu >= 0) { slots->pinned = per_cpu(nr_cpu_bp_pinned[type], cpu); if (!tsk) slots->pinned += max_task_bp_pinned(cpu, type); else - slots->pinned += task_bp_pinned(tsk, type); + slots->pinned += task_bp_pinned(bp, type); slots->flexible = per_cpu(nr_bp_flexible[type], cpu); return; @@ -162,7 +154,7 @@ fetch_bp_busy_slots(struct bp_busy_slots *slots, struct perf_event *bp, if (!tsk) nr += max_task_bp_pinned(cpu, type); else - nr += task_bp_pinned(tsk, type); + nr += task_bp_pinned(bp, type); if (nr > slots->pinned) slots->pinned = nr; @@ -188,7 +180,7 @@ fetch_this_slot(struct bp_busy_slots *slots, int weight) /* * 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, +static void toggle_bp_task_slot(struct perf_event *bp, int cpu, bool enable, enum bp_type_idx type, int weight) { unsigned int *tsk_pinned; @@ -196,10 +188,11 @@ static void toggle_bp_task_slot(struct task_struct *tsk, int cpu, bool enable, int old_idx = 0; int idx = 0; - old_count = task_bp_pinned(tsk, type); + old_count = task_bp_pinned(bp, type); old_idx = old_count - 1; idx = old_idx + weight; + /* tsk_pinned[n] is the number of tasks having n breakpoints */ tsk_pinned = per_cpu(nr_task_bp_pinned[type], cpu); if (enable) { tsk_pinned[idx]++; @@ -220,25 +213,43 @@ toggle_bp_slot(struct perf_event *bp, bool enable, enum bp_type_idx type, int weight) { int cpu = bp->cpu; - struct task_struct *tsk = bp->ctx->task; + struct task_struct *tsk = bp->hw.bp_target; + + /* Pinned counter cpu profiling */ + if (!tsk) { + + if (enable) + per_cpu(nr_cpu_bp_pinned[type], bp->cpu) += weight; + else + per_cpu(nr_cpu_bp_pinned[type], bp->cpu) -= weight; + return; + } /* Pinned counter task profiling */ - if (tsk) { - if (cpu >= 0) { - toggle_bp_task_slot(tsk, cpu, enable, type, weight); - return; - } + if (!enable) + list_del(&bp->hw.bp_list); + + if (cpu >= 0) { + toggle_bp_task_slot(bp, cpu, enable, type, weight); + } else { for_each_online_cpu(cpu) - toggle_bp_task_slot(tsk, cpu, enable, type, weight); - return; + toggle_bp_task_slot(bp, cpu, enable, type, weight); } - /* Pinned counter cpu profiling */ if (enable) - per_cpu(nr_cpu_bp_pinned[type], bp->cpu) += weight; - else - per_cpu(nr_cpu_bp_pinned[type], bp->cpu) -= weight; + list_add_tail(&bp->hw.bp_list, &bp_task_head); +} + +/* + * Function to perform processor-specific cleanup during unregistration + */ +__weak void arch_unregister_hw_breakpoint(struct perf_event *bp) +{ + /* + * A weak stub function here for those archs that don't define + * it inside arch/.../kernel/hw_breakpoint.c + */ } /* @@ -301,6 +312,10 @@ static int __reserve_bp_slot(struct perf_event *bp) weight = hw_breakpoint_weight(bp); fetch_bp_busy_slots(&slots, bp, type); + /* + * Simulate the addition of this breakpoint to the constraints + * and see the result. + */ fetch_this_slot(&slots, weight); /* Flexible counters need to keep at least one slot */ @@ -339,6 +354,7 @@ void release_bp_slot(struct perf_event *bp) { mutex_lock(&nr_bp_mutex); + arch_unregister_hw_breakpoint(bp); __release_bp_slot(bp); mutex_unlock(&nr_bp_mutex); @@ -417,7 +433,7 @@ register_user_hw_breakpoint(struct perf_event_attr *attr, perf_overflow_handler_t triggered, struct task_struct *tsk) { - return perf_event_create_kernel_counter(attr, -1, tsk->pid, triggered); + return perf_event_create_kernel_counter(attr, -1, tsk, triggered); } EXPORT_SYMBOL_GPL(register_user_hw_breakpoint); @@ -499,7 +515,7 @@ register_wide_hw_breakpoint(struct perf_event_attr *attr, get_online_cpus(); for_each_online_cpu(cpu) { pevent = per_cpu_ptr(cpu_events, cpu); - bp = perf_event_create_kernel_counter(attr, cpu, -1, triggered); + bp = perf_event_create_kernel_counter(attr, cpu, NULL, triggered); *pevent = bp; @@ -549,7 +565,62 @@ static struct notifier_block hw_breakpoint_exceptions_nb = { .priority = 0x7fffffff }; -static int __init init_hw_breakpoint(void) +static void bp_perf_event_destroy(struct perf_event *event) +{ + release_bp_slot(event); +} + +static int hw_breakpoint_event_init(struct perf_event *bp) +{ + int err; + + if (bp->attr.type != PERF_TYPE_BREAKPOINT) + return -ENOENT; + + err = register_perf_hw_breakpoint(bp); + if (err) + return err; + + bp->destroy = bp_perf_event_destroy; + + return 0; +} + +static int hw_breakpoint_add(struct perf_event *bp, int flags) +{ + if (!(flags & PERF_EF_START)) + bp->hw.state = PERF_HES_STOPPED; + + return arch_install_hw_breakpoint(bp); +} + +static void hw_breakpoint_del(struct perf_event *bp, int flags) +{ + arch_uninstall_hw_breakpoint(bp); +} + +static void hw_breakpoint_start(struct perf_event *bp, int flags) +{ + bp->hw.state = 0; +} + +static void hw_breakpoint_stop(struct perf_event *bp, int flags) +{ + bp->hw.state = PERF_HES_STOPPED; +} + +static struct pmu perf_breakpoint = { + .task_ctx_nr = perf_sw_context, /* could eventually get its own */ + + .event_init = hw_breakpoint_event_init, + .add = hw_breakpoint_add, + .del = hw_breakpoint_del, + .start = hw_breakpoint_start, + .stop = hw_breakpoint_stop, + .read = hw_breakpoint_pmu_read, +}; + +int __init init_hw_breakpoint(void) { unsigned int **task_bp_pinned; int cpu, err_cpu; @@ -570,6 +641,8 @@ static int __init init_hw_breakpoint(void) constraints_initialized = 1; + perf_pmu_register(&perf_breakpoint, "breakpoint", PERF_TYPE_BREAKPOINT); + return register_die_notifier(&hw_breakpoint_exceptions_nb); err_alloc: @@ -582,11 +655,5 @@ static int __init init_hw_breakpoint(void) return -ENOMEM; } -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, -}; diff --git a/kernel/irq/Kconfig b/kernel/irq/Kconfig new file mode 100644 index 00000000000..8e42fec7686 --- /dev/null +++ b/kernel/irq/Kconfig @@ -0,0 +1,50 @@ +config HAVE_GENERIC_HARDIRQS + def_bool n + +if HAVE_GENERIC_HARDIRQS +menu "IRQ subsystem" +# +# Interrupt subsystem related configuration options +# +config GENERIC_HARDIRQS + def_bool y + +# Select this to disable the deprecated stuff +config GENERIC_HARDIRQS_NO_DEPRECATED + def_bool n + +# Options selectable by the architecture code +config HAVE_SPARSE_IRQ + def_bool n + +config GENERIC_IRQ_PROBE + def_bool n + +config GENERIC_PENDING_IRQ + def_bool n + +config AUTO_IRQ_AFFINITY + def_bool n + +config IRQ_PER_CPU + def_bool n + +config HARDIRQS_SW_RESEND + def_bool n + +config SPARSE_IRQ + bool "Support sparse irq numbering" + depends on HAVE_SPARSE_IRQ + ---help--- + + Sparse irq numbering is useful for distro kernels that want + to define a high CONFIG_NR_CPUS value but still want to have + low kernel memory footprint on smaller machines. + + ( Sparse irqs can also be beneficial on NUMA boxes, as they spread + out the interrupt descriptors in a more NUMA-friendly way. ) + + If you don't know what to do here, say N. + +endmenu +endif diff --git a/kernel/irq/Makefile b/kernel/irq/Makefile index 7d047808419..54329cd7b3e 100644 --- a/kernel/irq/Makefile +++ b/kernel/irq/Makefile @@ -1,7 +1,6 @@ -obj-y := handle.o manage.o spurious.o resend.o chip.o devres.o +obj-y := irqdesc.o handle.o manage.o spurious.o resend.o chip.o dummychip.o devres.o obj-$(CONFIG_GENERIC_IRQ_PROBE) += autoprobe.o obj-$(CONFIG_PROC_FS) += proc.o obj-$(CONFIG_GENERIC_PENDING_IRQ) += migration.o -obj-$(CONFIG_NUMA_IRQ_DESC) += numa_migrate.o obj-$(CONFIG_PM_SLEEP) += pm.o diff --git a/kernel/irq/autoprobe.c b/kernel/irq/autoprobe.c index 2295a31ef11..505798f86c3 100644 --- a/kernel/irq/autoprobe.c +++ b/kernel/irq/autoprobe.c @@ -57,9 +57,10 @@ unsigned long probe_irq_on(void) * Some chips need to know about probing in * progress: */ - if (desc->chip->set_type) - desc->chip->set_type(i, IRQ_TYPE_PROBE); - desc->chip->startup(i); + if (desc->irq_data.chip->irq_set_type) + desc->irq_data.chip->irq_set_type(&desc->irq_data, + IRQ_TYPE_PROBE); + desc->irq_data.chip->irq_startup(&desc->irq_data); } raw_spin_unlock_irq(&desc->lock); } @@ -76,7 +77,7 @@ unsigned long probe_irq_on(void) raw_spin_lock_irq(&desc->lock); if (!desc->action && !(desc->status & IRQ_NOPROBE)) { desc->status |= IRQ_AUTODETECT | IRQ_WAITING; - if (desc->chip->startup(i)) + if (desc->irq_data.chip->irq_startup(&desc->irq_data)) desc->status |= IRQ_PENDING; } raw_spin_unlock_irq(&desc->lock); @@ -98,7 +99,7 @@ unsigned long probe_irq_on(void) /* It triggered already - consider it spurious. */ if (!(status & IRQ_WAITING)) { desc->status = status & ~IRQ_AUTODETECT; - desc->chip->shutdown(i); + desc->irq_data.chip->irq_shutdown(&desc->irq_data); } else if (i < 32) mask |= 1 << i; @@ -137,7 +138,7 @@ unsigned int probe_irq_mask(unsigned long val) mask |= 1 << i; desc->status = status & ~IRQ_AUTODETECT; - desc->chip->shutdown(i); + desc->irq_data.chip->irq_shutdown(&desc->irq_data); } raw_spin_unlock_irq(&desc->lock); } @@ -181,7 +182,7 @@ int probe_irq_off(unsigned long val) nr_of_irqs++; } desc->status = status & ~IRQ_AUTODETECT; - desc->chip->shutdown(i); + desc->irq_data.chip->irq_shutdown(&desc->irq_data); } raw_spin_unlock_irq(&desc->lock); } diff --git a/kernel/irq/chip.c b/kernel/irq/chip.c index b7091d5ca2f..baa5c4acad8 100644 --- a/kernel/irq/chip.c +++ b/kernel/irq/chip.c @@ -18,108 +18,6 @@ #include "internals.h" -static void dynamic_irq_init_x(unsigned int irq, bool keep_chip_data) -{ - struct irq_desc *desc; - unsigned long flags; - - desc = irq_to_desc(irq); - if (!desc) { - WARN(1, KERN_ERR "Trying to initialize invalid IRQ%d\n", irq); - return; - } - - /* Ensure we don't have left over values from a previous use of this irq */ - raw_spin_lock_irqsave(&desc->lock, flags); - desc->status = IRQ_DISABLED; - desc->chip = &no_irq_chip; - desc->handle_irq = handle_bad_irq; - desc->depth = 1; - desc->msi_desc = NULL; - desc->handler_data = NULL; - if (!keep_chip_data) - desc->chip_data = NULL; - desc->action = NULL; - desc->irq_count = 0; - desc->irqs_unhandled = 0; -#ifdef CONFIG_SMP - cpumask_setall(desc->affinity); -#ifdef CONFIG_GENERIC_PENDING_IRQ - cpumask_clear(desc->pending_mask); -#endif -#endif - raw_spin_unlock_irqrestore(&desc->lock, flags); -} - -/** - * dynamic_irq_init - initialize a dynamically allocated irq - * @irq: irq number to initialize - */ -void dynamic_irq_init(unsigned int irq) -{ - dynamic_irq_init_x(irq, false); -} - -/** - * dynamic_irq_init_keep_chip_data - initialize a dynamically allocated irq - * @irq: irq number to initialize - * - * does not set irq_to_desc(irq)->chip_data to NULL - */ -void dynamic_irq_init_keep_chip_data(unsigned int irq) -{ - dynamic_irq_init_x(irq, true); -} - -static void dynamic_irq_cleanup_x(unsigned int irq, bool keep_chip_data) -{ - struct irq_desc *desc = irq_to_desc(irq); - unsigned long flags; - - if (!desc) { - WARN(1, KERN_ERR "Trying to cleanup invalid IRQ%d\n", irq); - return; - } - - raw_spin_lock_irqsave(&desc->lock, flags); - if (desc->action) { - raw_spin_unlock_irqrestore(&desc->lock, flags); - WARN(1, KERN_ERR "Destroying IRQ%d without calling free_irq\n", - irq); - return; - } - desc->msi_desc = NULL; - desc->handler_data = NULL; - if (!keep_chip_data) - desc->chip_data = NULL; - desc->handle_irq = handle_bad_irq; - desc->chip = &no_irq_chip; - desc->name = NULL; - clear_kstat_irqs(desc); - raw_spin_unlock_irqrestore(&desc->lock, flags); -} - -/** - * dynamic_irq_cleanup - cleanup a dynamically allocated irq - * @irq: irq number to initialize - */ -void dynamic_irq_cleanup(unsigned int irq) -{ - dynamic_irq_cleanup_x(irq, false); -} - -/** - * dynamic_irq_cleanup_keep_chip_data - cleanup a dynamically allocated irq - * @irq: irq number to initialize - * - * does not set irq_to_desc(irq)->chip_data to NULL - */ -void dynamic_irq_cleanup_keep_chip_data(unsigned int irq) -{ - dynamic_irq_cleanup_x(irq, true); -} - - /** * set_irq_chip - set the irq chip for an irq * @irq: irq number @@ -140,7 +38,7 @@ int set_irq_chip(unsigned int irq, struct irq_chip *chip) raw_spin_lock_irqsave(&desc->lock, flags); irq_chip_set_defaults(chip); - desc->chip = chip; + desc->irq_data.chip = chip; raw_spin_unlock_irqrestore(&desc->lock, flags); return 0; @@ -193,7 +91,7 @@ int set_irq_data(unsigned int irq, void *data) } raw_spin_lock_irqsave(&desc->lock, flags); - desc->handler_data = data; + desc->irq_data.handler_data = data; raw_spin_unlock_irqrestore(&desc->lock, flags); return 0; } @@ -218,7 +116,7 @@ int set_irq_msi(unsigned int irq, struct msi_desc *entry) } raw_spin_lock_irqsave(&desc->lock, flags); - desc->msi_desc = entry; + desc->irq_data.msi_desc = entry; if (entry) entry->irq = irq; raw_spin_unlock_irqrestore(&desc->lock, flags); @@ -243,19 +141,27 @@ int set_irq_chip_data(unsigned int irq, void *data) return -EINVAL; } - if (!desc->chip) { + if (!desc->irq_data.chip) { printk(KERN_ERR "BUG: bad set_irq_chip_data(IRQ#%d)\n", irq); return -EINVAL; } raw_spin_lock_irqsave(&desc->lock, flags); - desc->chip_data = data; + desc->irq_data.chip_data = data; raw_spin_unlock_irqrestore(&desc->lock, flags); return 0; } EXPORT_SYMBOL(set_irq_chip_data); +struct irq_data *irq_get_irq_data(unsigned int irq) +{ + struct irq_desc *desc = irq_to_desc(irq); + + return desc ? &desc->irq_data : NULL; +} +EXPORT_SYMBOL_GPL(irq_get_irq_data); + /** * set_irq_nested_thread - Set/Reset the IRQ_NESTED_THREAD flag of an irq * @@ -287,93 +193,216 @@ EXPORT_SYMBOL_GPL(set_irq_nested_thread); /* * default enable function */ -static void default_enable(unsigned int irq) +static void default_enable(struct irq_data *data) { - struct irq_desc *desc = irq_to_desc(irq); + struct irq_desc *desc = irq_data_to_desc(data); - desc->chip->unmask(irq); + desc->irq_data.chip->irq_unmask(&desc->irq_data); desc->status &= ~IRQ_MASKED; } /* * default disable function */ -static void default_disable(unsigned int irq) +static void default_disable(struct irq_data *data) { } /* * default startup function */ -static unsigned int default_startup(unsigned int irq) +static unsigned int default_startup(struct irq_data *data) { - struct irq_desc *desc = irq_to_desc(irq); + struct irq_desc *desc = irq_data_to_desc(data); - desc->chip->enable(irq); + desc->irq_data.chip->irq_enable(data); return 0; } /* * default shutdown function */ -static void default_shutdown(unsigned int irq) +static void default_shutdown(struct irq_data *data) { - struct irq_desc *desc = irq_to_desc(irq); + struct irq_desc *desc = irq_data_to_desc(data); - desc->chip->mask(irq); + desc->irq_data.chip->irq_mask(&desc->irq_data); desc->status |= IRQ_MASKED; } +#ifndef CONFIG_GENERIC_HARDIRQS_NO_DEPRECATED +/* Temporary migration helpers */ +static void compat_irq_mask(struct irq_data *data) +{ + data->chip->mask(data->irq); +} + +static void compat_irq_unmask(struct irq_data *data) +{ + data->chip->unmask(data->irq); +} + +static void compat_irq_ack(struct irq_data *data) +{ + data->chip->ack(data->irq); +} + +static void compat_irq_mask_ack(struct irq_data *data) +{ + data->chip->mask_ack(data->irq); +} + +static void compat_irq_eoi(struct irq_data *data) +{ + data->chip->eoi(data->irq); +} + +static void compat_irq_enable(struct irq_data *data) +{ + data->chip->enable(data->irq); +} + +static void compat_irq_disable(struct irq_data *data) +{ + data->chip->disable(data->irq); +} + +static void compat_irq_shutdown(struct irq_data *data) +{ + data->chip->shutdown(data->irq); +} + +static unsigned int compat_irq_startup(struct irq_data *data) +{ + return data->chip->startup(data->irq); +} + +static int compat_irq_set_affinity(struct irq_data *data, + const struct cpumask *dest, bool force) +{ + return data->chip->set_affinity(data->irq, dest); +} + +static int compat_irq_set_type(struct irq_data *data, unsigned int type) +{ + return data->chip->set_type(data->irq, type); +} + +static int compat_irq_set_wake(struct irq_data *data, unsigned int on) +{ + return data->chip->set_wake(data->irq, on); +} + +static int compat_irq_retrigger(struct irq_data *data) +{ + return data->chip->retrigger(data->irq); +} + +static void compat_bus_lock(struct irq_data *data) +{ + data->chip->bus_lock(data->irq); +} + +static void compat_bus_sync_unlock(struct irq_data *data) +{ + data->chip->bus_sync_unlock(data->irq); +} +#endif + /* * Fixup enable/disable function pointers */ void irq_chip_set_defaults(struct irq_chip *chip) { - if (!chip->enable) - chip->enable = default_enable; - if (!chip->disable) - chip->disable = default_disable; - if (!chip->startup) - chip->startup = default_startup; +#ifndef CONFIG_GENERIC_HARDIRQS_NO_DEPRECATED /* - * We use chip->disable, when the user provided its own. When - * we have default_disable set for chip->disable, then we need + * Compat fixup functions need to be before we set the + * defaults for enable/disable/startup/shutdown + */ + if (chip->enable) + chip->irq_enable = compat_irq_enable; + if (chip->disable) + chip->irq_disable = compat_irq_disable; + if (chip->shutdown) + chip->irq_shutdown = compat_irq_shutdown; + if (chip->startup) + chip->irq_startup = compat_irq_startup; +#endif + /* + * The real defaults + */ + if (!chip->irq_enable) + chip->irq_enable = default_enable; + if (!chip->irq_disable) + chip->irq_disable = default_disable; + if (!chip->irq_startup) + chip->irq_startup = default_startup; + /* + * We use chip->irq_disable, when the user provided its own. When + * we have default_disable set for chip->irq_disable, then we need * to use default_shutdown, otherwise the irq line is not * disabled on free_irq(): */ - if (!chip->shutdown) - chip->shutdown = chip->disable != default_disable ? - chip->disable : default_shutdown; - if (!chip->name) - chip->name = chip->typename; + if (!chip->irq_shutdown) + chip->irq_shutdown = chip->irq_disable != default_disable ? + chip->irq_disable : default_shutdown; + +#ifndef CONFIG_GENERIC_HARDIRQS_NO_DEPRECATED if (!chip->end) chip->end = dummy_irq_chip.end; + + /* + * Now fix up the remaining compat handlers + */ + if (chip->bus_lock) + chip->irq_bus_lock = compat_bus_lock; + if (chip->bus_sync_unlock) + chip->irq_bus_sync_unlock = compat_bus_sync_unlock; + if (chip->mask) + chip->irq_mask = compat_irq_mask; + if (chip->unmask) + chip->irq_unmask = compat_irq_unmask; + if (chip->ack) + chip->irq_ack = compat_irq_ack; + if (chip->mask_ack) + chip->irq_mask_ack = compat_irq_mask_ack; + if (chip->eoi) + chip->irq_eoi = compat_irq_eoi; + if (chip->set_affinity) + chip->irq_set_affinity = compat_irq_set_affinity; + if (chip->set_type) + chip->irq_set_type = compat_irq_set_type; + if (chip->set_wake) + chip->irq_set_wake = compat_irq_set_wake; + if (chip->retrigger) + chip->irq_retrigger = compat_irq_retrigger; +#endif } -static inline void mask_ack_irq(struct irq_desc *desc, int irq) +static inline void mask_ack_irq(struct irq_desc *desc) { - if (desc->chip->mask_ack) - desc->chip->mask_ack(irq); + if (desc->irq_data.chip->irq_mask_ack) + desc->irq_data.chip->irq_mask_ack(&desc->irq_data); else { - desc->chip->mask(irq); - if (desc->chip->ack) - desc->chip->ack(irq); + desc->irq_data.chip->irq_mask(&desc->irq_data); + if (desc->irq_data.chip->irq_ack) + desc->irq_data.chip->irq_ack(&desc->irq_data); } desc->status |= IRQ_MASKED; } -static inline void mask_irq(struct irq_desc *desc, int irq) +static inline void mask_irq(struct irq_desc *desc) { - if (desc->chip->mask) { - desc->chip->mask(irq); + if (desc->irq_data.chip->irq_mask) { + desc->irq_data.chip->irq_mask(&desc->irq_data); desc->status |= IRQ_MASKED; } } -static inline void unmask_irq(struct irq_desc *desc, int irq) +static inline void unmask_irq(struct irq_desc *desc) { - if (desc->chip->unmask) { - desc->chip->unmask(irq); + if (desc->irq_data.chip->irq_unmask) { + desc->irq_data.chip->irq_unmask(&desc->irq_data); desc->status &= ~IRQ_MASKED; } } @@ -476,7 +505,7 @@ handle_level_irq(unsigned int irq, struct irq_desc *desc) irqreturn_t action_ret; raw_spin_lock(&desc->lock); - mask_ack_irq(desc, irq); + mask_ack_irq(desc); if (unlikely(desc->status & IRQ_INPROGRESS)) goto out_unlock; @@ -502,7 +531,7 @@ handle_level_irq(unsigned int irq, struct irq_desc *desc) desc->status &= ~IRQ_INPROGRESS; if (!(desc->status & (IRQ_DISABLED | IRQ_ONESHOT))) - unmask_irq(desc, irq); + unmask_irq(desc); out_unlock: raw_spin_unlock(&desc->lock); } @@ -539,7 +568,7 @@ handle_fasteoi_irq(unsigned int irq, struct irq_desc *desc) action = desc->action; if (unlikely(!action || (desc->status & IRQ_DISABLED))) { desc->status |= IRQ_PENDING; - mask_irq(desc, irq); + mask_irq(desc); goto out; } @@ -554,7 +583,7 @@ handle_fasteoi_irq(unsigned int irq, struct irq_desc *desc) raw_spin_lock(&desc->lock); desc->status &= ~IRQ_INPROGRESS; out: - desc->chip->eoi(irq); + desc->irq_data.chip->irq_eoi(&desc->irq_data); raw_spin_unlock(&desc->lock); } @@ -590,14 +619,13 @@ handle_edge_irq(unsigned int irq, struct irq_desc *desc) if (unlikely((desc->status & (IRQ_INPROGRESS | IRQ_DISABLED)) || !desc->action)) { desc->status |= (IRQ_PENDING | IRQ_MASKED); - mask_ack_irq(desc, irq); + mask_ack_irq(desc); goto out_unlock; } kstat_incr_irqs_this_cpu(irq, desc); /* Start handling the irq */ - if (desc->chip->ack) - desc->chip->ack(irq); + desc->irq_data.chip->irq_ack(&desc->irq_data); /* Mark the IRQ currently in progress.*/ desc->status |= IRQ_INPROGRESS; @@ -607,7 +635,7 @@ handle_edge_irq(unsigned int irq, struct irq_desc *desc) irqreturn_t action_ret; if (unlikely(!action)) { - mask_irq(desc, irq); + mask_irq(desc); goto out_unlock; } @@ -619,7 +647,7 @@ handle_edge_irq(unsigned int irq, struct irq_desc *desc) if (unlikely((desc->status & (IRQ_PENDING | IRQ_MASKED | IRQ_DISABLED)) == (IRQ_PENDING | IRQ_MASKED))) { - unmask_irq(desc, irq); + unmask_irq(desc); } desc->status &= ~IRQ_PENDING; @@ -650,15 +678,15 @@ handle_percpu_irq(unsigned int irq, struct irq_desc *desc) kstat_incr_irqs_this_cpu(irq, desc); - if (desc->chip->ack) - desc->chip->ack(irq); + if (desc->irq_data.chip->irq_ack) + desc->irq_data.chip->irq_ack(&desc->irq_data); action_ret = handle_IRQ_event(irq, desc->action); if (!noirqdebug) note_interrupt(irq, desc, action_ret); - if (desc->chip->eoi) - desc->chip->eoi(irq); + if (desc->irq_data.chip->irq_eoi) + desc->irq_data.chip->irq_eoi(&desc->irq_data); } void @@ -676,7 +704,7 @@ __set_irq_handler(unsigned int irq, irq_flow_handler_t handle, int is_chained, if (!handle) handle = handle_bad_irq; - else if (desc->chip == &no_irq_chip) { + else if (desc->irq_data.chip == &no_irq_chip) { printk(KERN_WARNING "Trying to install %sinterrupt handler " "for IRQ%d\n", is_chained ? "chained " : "", irq); /* @@ -686,16 +714,16 @@ __set_irq_handler(unsigned int irq, irq_flow_handler_t handle, int is_chained, * prevent us to setup the interrupt at all. Switch it to * dummy_irq_chip for easy transition. */ - desc->chip = &dummy_irq_chip; + desc->irq_data.chip = &dummy_irq_chip; } - chip_bus_lock(irq, desc); + chip_bus_lock(desc); raw_spin_lock_irqsave(&desc->lock, flags); /* Uninstall? */ if (handle == handle_bad_irq) { - if (desc->chip != &no_irq_chip) - mask_ack_irq(desc, irq); + if (desc->irq_data.chip != &no_irq_chip) + mask_ack_irq(desc); desc->status |= IRQ_DISABLED; desc->depth = 1; } @@ -706,10 +734,10 @@ __set_irq_handler(unsigned int irq, irq_flow_handler_t handle, int is_chained, desc->status &= ~IRQ_DISABLED; desc->status |= IRQ_NOREQUEST | IRQ_NOPROBE; desc->depth = 0; - desc->chip->startup(irq); + desc->irq_data.chip->irq_startup(&desc->irq_data); } raw_spin_unlock_irqrestore(&desc->lock, flags); - chip_bus_sync_unlock(irq, desc); + chip_bus_sync_unlock(desc); } EXPORT_SYMBOL_GPL(__set_irq_handler); @@ -729,32 +757,20 @@ set_irq_chip_and_handler_name(unsigned int irq, struct irq_chip *chip, __set_irq_handler(irq, handle, 0, name); } -void set_irq_noprobe(unsigned int irq) +void irq_modify_status(unsigned int irq, unsigned long clr, unsigned long set) { struct irq_desc *desc = irq_to_desc(irq); unsigned long flags; - if (!desc) { - printk(KERN_ERR "Trying to mark IRQ%d non-probeable\n", irq); + if (!desc) return; - } - - raw_spin_lock_irqsave(&desc->lock, flags); - desc->status |= IRQ_NOPROBE; - raw_spin_unlock_irqrestore(&desc->lock, flags); -} - -void set_irq_probe(unsigned int irq) -{ - struct irq_desc *desc = irq_to_desc(irq); - unsigned long flags; - if (!desc) { - printk(KERN_ERR "Trying to mark IRQ%d probeable\n", irq); - return; - } + /* Sanitize flags */ + set &= IRQF_MODIFY_MASK; + clr &= IRQF_MODIFY_MASK; raw_spin_lock_irqsave(&desc->lock, flags); - desc->status &= ~IRQ_NOPROBE; + desc->status &= ~clr; + desc->status |= set; raw_spin_unlock_irqrestore(&desc->lock, flags); } diff --git a/kernel/irq/dummychip.c b/kernel/irq/dummychip.c new file mode 100644 index 00000000000..20dc5474947 --- /dev/null +++ b/kernel/irq/dummychip.c @@ -0,0 +1,68 @@ +/* + * Copyright (C) 1992, 1998-2006 Linus Torvalds, Ingo Molnar + * Copyright (C) 2005-2006, Thomas Gleixner, Russell King + * + * This file contains the dummy interrupt chip implementation + */ +#include <linux/interrupt.h> +#include <linux/irq.h> + +#include "internals.h" + +/* + * What should we do if we get a hw irq event on an illegal vector? + * Each architecture has to answer this themself. + */ +static void ack_bad(struct irq_data *data) +{ + struct irq_desc *desc = irq_data_to_desc(data); + + print_irq_desc(data->irq, desc); + ack_bad_irq(data->irq); +} + +/* + * NOP functions + */ +static void noop(struct irq_data *data) { } + +static unsigned int noop_ret(struct irq_data *data) +{ + return 0; +} + +#ifndef CONFIG_GENERIC_HARDIRQS_NO_DEPRECATED +static void compat_noop(unsigned int irq) { } +#define END_INIT .end = compat_noop +#else +#define END_INIT +#endif + +/* + * Generic no controller implementation + */ +struct irq_chip no_irq_chip = { + .name = "none", + .irq_startup = noop_ret, + .irq_shutdown = noop, + .irq_enable = noop, + .irq_disable = noop, + .irq_ack = ack_bad, + END_INIT +}; + +/* + * Generic dummy implementation which can be used for + * real dumb interrupt sources + */ +struct irq_chip dummy_irq_chip = { + .name = "dummy", + .irq_startup = noop_ret, + .irq_shutdown = noop, + .irq_enable = noop, + .irq_disable = noop, + .irq_ack = noop, + .irq_mask = noop, + .irq_unmask = noop, + END_INIT +}; diff --git a/kernel/irq/handle.c b/kernel/irq/handle.c index 27e5c691122..3540a719012 100644 --- a/kernel/irq/handle.c +++ b/kernel/irq/handle.c @@ -11,24 +11,15 @@ */ #include <linux/irq.h> -#include <linux/sched.h> -#include <linux/slab.h> -#include <linux/module.h> #include <linux/random.h> +#include <linux/sched.h> #include <linux/interrupt.h> #include <linux/kernel_stat.h> -#include <linux/rculist.h> -#include <linux/hash.h> -#include <linux/radix-tree.h> + #include <trace/events/irq.h> #include "internals.h" -/* - * lockdep: we want to handle all irq_desc locks as a single lock-class: - */ -struct lock_class_key irq_desc_lock_class; - /** * handle_bad_irq - handle spurious and unhandled irqs * @irq: the interrupt number @@ -43,304 +34,6 @@ void handle_bad_irq(unsigned int irq, struct irq_desc *desc) ack_bad_irq(irq); } -#if defined(CONFIG_SMP) && defined(CONFIG_GENERIC_HARDIRQS) -static void __init init_irq_default_affinity(void) -{ - alloc_cpumask_var(&irq_default_affinity, GFP_NOWAIT); - cpumask_setall(irq_default_affinity); -} -#else -static void __init init_irq_default_affinity(void) -{ -} -#endif - -/* - * Linux has a controller-independent interrupt architecture. - * Every controller has a 'controller-template', that is used - * by the main code to do the right thing. Each driver-visible - * interrupt source is transparently wired to the appropriate - * controller. Thus drivers need not be aware of the - * interrupt-controller. - * - * The code is designed to be easily extended with new/different - * interrupt controllers, without having to do assembly magic or - * having to touch the generic code. - * - * Controller mappings for all interrupt sources: - */ -int nr_irqs = NR_IRQS; -EXPORT_SYMBOL_GPL(nr_irqs); - -#ifdef CONFIG_SPARSE_IRQ - -static struct irq_desc irq_desc_init = { - .irq = -1, - .status = IRQ_DISABLED, - .chip = &no_irq_chip, - .handle_irq = handle_bad_irq, - .depth = 1, - .lock = __RAW_SPIN_LOCK_UNLOCKED(irq_desc_init.lock), -}; - -void __ref init_kstat_irqs(struct irq_desc *desc, int node, int nr) -{ - void *ptr; - - ptr = kzalloc_node(nr * sizeof(*desc->kstat_irqs), - GFP_ATOMIC, node); - - /* - * 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 node %d\n", node); - desc->kstat_irqs = ptr; - } -} - -static void init_one_irq_desc(int irq, struct irq_desc *desc, int node) -{ - memcpy(desc, &irq_desc_init, sizeof(struct irq_desc)); - - raw_spin_lock_init(&desc->lock); - desc->irq = irq; -#ifdef CONFIG_SMP - desc->node = node; -#endif - lockdep_set_class(&desc->lock, &irq_desc_lock_class); - 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 (!alloc_desc_masks(desc, node, false)) { - printk(KERN_ERR "can not alloc irq_desc cpumasks\n"); - BUG_ON(1); - } - init_desc_masks(desc); - arch_init_chip_data(desc, node); -} - -/* - * Protect the sparse_irqs: - */ -DEFINE_RAW_SPINLOCK(sparse_irq_lock); - -static RADIX_TREE(irq_desc_tree, GFP_ATOMIC); - -static void set_irq_desc(unsigned int irq, struct irq_desc *desc) -{ - radix_tree_insert(&irq_desc_tree, irq, desc); -} - -struct irq_desc *irq_to_desc(unsigned int irq) -{ - return radix_tree_lookup(&irq_desc_tree, irq); -} - -void replace_irq_desc(unsigned int irq, struct irq_desc *desc) -{ - void **ptr; - - ptr = radix_tree_lookup_slot(&irq_desc_tree, irq); - if (ptr) - radix_tree_replace_slot(ptr, desc); -} - -static struct irq_desc irq_desc_legacy[NR_IRQS_LEGACY] __cacheline_aligned_in_smp = { - [0 ... NR_IRQS_LEGACY-1] = { - .irq = -1, - .status = IRQ_DISABLED, - .chip = &no_irq_chip, - .handle_irq = handle_bad_irq, - .depth = 1, - .lock = __RAW_SPIN_LOCK_UNLOCKED(irq_desc_init.lock), - } -}; - -static unsigned int *kstat_irqs_legacy; - -int __init early_irq_init(void) -{ - struct irq_desc *desc; - int legacy_count; - int node; - int i; - - init_irq_default_affinity(); - - /* initialize nr_irqs based on nr_cpu_ids */ - arch_probe_nr_irqs(); - printk(KERN_INFO "NR_IRQS:%d nr_irqs:%d\n", NR_IRQS, nr_irqs); - - desc = irq_desc_legacy; - legacy_count = ARRAY_SIZE(irq_desc_legacy); - node = first_online_node; - - /* allocate based on nr_cpu_ids */ - 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); - alloc_desc_masks(&desc[i], node, true); - init_desc_masks(&desc[i]); - set_irq_desc(i, &desc[i]); - } - - return arch_early_irq_init(); -} - -struct irq_desc * __ref irq_to_desc_alloc_node(unsigned int irq, int node) -{ - struct irq_desc *desc; - unsigned long flags; - - if (irq >= nr_irqs) { - WARN(1, "irq (%d) >= nr_irqs (%d) in irq_to_desc_alloc\n", - irq, nr_irqs); - return NULL; - } - - desc = irq_to_desc(irq); - if (desc) - return desc; - - raw_spin_lock_irqsave(&sparse_irq_lock, flags); - - /* We have to check it to avoid races with another CPU */ - desc = irq_to_desc(irq); - if (desc) - goto out_unlock; - - desc = kzalloc_node(sizeof(*desc), GFP_ATOMIC, node); - - 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, node); - - set_irq_desc(irq, desc); - -out_unlock: - raw_spin_unlock_irqrestore(&sparse_irq_lock, flags); - - return desc; -} - -#else /* !CONFIG_SPARSE_IRQ */ - -struct irq_desc irq_desc[NR_IRQS] __cacheline_aligned_in_smp = { - [0 ... NR_IRQS-1] = { - .status = IRQ_DISABLED, - .chip = &no_irq_chip, - .handle_irq = handle_bad_irq, - .depth = 1, - .lock = __RAW_SPIN_LOCK_UNLOCKED(irq_desc->lock), - } -}; - -static unsigned int kstat_irqs_all[NR_IRQS][NR_CPUS]; -int __init early_irq_init(void) -{ - struct irq_desc *desc; - int count; - int i; - - init_irq_default_affinity(); - - printk(KERN_INFO "NR_IRQS:%d\n", NR_IRQS); - - desc = irq_desc; - count = ARRAY_SIZE(irq_desc); - - for (i = 0; i < count; i++) { - desc[i].irq = i; - 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(); -} - -struct irq_desc *irq_to_desc(unsigned int irq) -{ - return (irq < NR_IRQS) ? irq_desc + irq : NULL; -} - -struct irq_desc *irq_to_desc_alloc_node(unsigned int irq, int node) -{ - return irq_to_desc(irq); -} -#endif /* !CONFIG_SPARSE_IRQ */ - -void clear_kstat_irqs(struct irq_desc *desc) -{ - memset(desc->kstat_irqs, 0, nr_cpu_ids * sizeof(*(desc->kstat_irqs))); -} - -/* - * What should we do if we get a hw irq event on an illegal vector? - * Each architecture has to answer this themself. - */ -static void ack_bad(unsigned int irq) -{ - struct irq_desc *desc = irq_to_desc(irq); - - print_irq_desc(irq, desc); - ack_bad_irq(irq); -} - -/* - * NOP functions - */ -static void noop(unsigned int irq) -{ -} - -static unsigned int noop_ret(unsigned int irq) -{ - return 0; -} - -/* - * Generic no controller implementation - */ -struct irq_chip no_irq_chip = { - .name = "none", - .startup = noop_ret, - .shutdown = noop, - .enable = noop, - .disable = noop, - .ack = ack_bad, - .end = noop, -}; - -/* - * Generic dummy implementation which can be used for - * real dumb interrupt sources - */ -struct irq_chip dummy_irq_chip = { - .name = "dummy", - .startup = noop_ret, - .shutdown = noop, - .enable = noop, - .disable = noop, - .ack = noop, - .mask = noop, - .unmask = noop, - .end = noop, -}; - /* * Special, empty irq handler: */ @@ -425,132 +118,3 @@ irqreturn_t handle_IRQ_event(unsigned int irq, struct irqaction *action) return retval; } - -#ifndef CONFIG_GENERIC_HARDIRQS_NO__DO_IRQ - -#ifdef CONFIG_ENABLE_WARN_DEPRECATED -# warning __do_IRQ is deprecated. Please convert to proper flow handlers -#endif - -/** - * __do_IRQ - original all in one highlevel IRQ handler - * @irq: the interrupt number - * - * __do_IRQ handles all normal device IRQ's (the special - * SMP cross-CPU interrupts have their own specific - * handlers). - * - * This is the original x86 implementation which is used for every - * interrupt type. - */ -unsigned int __do_IRQ(unsigned int irq) -{ - struct irq_desc *desc = irq_to_desc(irq); - struct irqaction *action; - unsigned int status; - - kstat_incr_irqs_this_cpu(irq, desc); - - if (CHECK_IRQ_PER_CPU(desc->status)) { - irqreturn_t action_ret; - - /* - * No locking required for CPU-local interrupts: - */ - if (desc->chip->ack) - desc->chip->ack(irq); - if (likely(!(desc->status & IRQ_DISABLED))) { - action_ret = handle_IRQ_event(irq, desc->action); - if (!noirqdebug) - note_interrupt(irq, desc, action_ret); - } - desc->chip->end(irq); - return 1; - } - - raw_spin_lock(&desc->lock); - if (desc->chip->ack) - desc->chip->ack(irq); - /* - * REPLAY is when Linux resends an IRQ that was dropped earlier - * WAITING is used by probe to mark irqs that are being tested - */ - status = desc->status & ~(IRQ_REPLAY | IRQ_WAITING); - status |= IRQ_PENDING; /* we _want_ to handle it */ - - /* - * If the IRQ is disabled for whatever reason, we cannot - * use the action we have. - */ - action = NULL; - if (likely(!(status & (IRQ_DISABLED | IRQ_INPROGRESS)))) { - action = desc->action; - status &= ~IRQ_PENDING; /* we commit to handling */ - status |= IRQ_INPROGRESS; /* we are handling it */ - } - desc->status = status; - - /* - * If there is no IRQ handler or it was disabled, exit early. - * Since we set PENDING, if another processor is handling - * a different instance of this same irq, the other processor - * will take care of it. - */ - if (unlikely(!action)) - goto out; - - /* - * Edge triggered interrupts need to remember - * pending events. - * This applies to any hw interrupts that allow a second - * instance of the same irq to arrive while we are in do_IRQ - * or in the handler. But the code here only handles the _second_ - * instance of the irq, not the third or fourth. So it is mostly - * useful for irq hardware that does not mask cleanly in an - * SMP environment. - */ - for (;;) { - irqreturn_t action_ret; - - raw_spin_unlock(&desc->lock); - - action_ret = handle_IRQ_event(irq, action); - if (!noirqdebug) - note_interrupt(irq, desc, action_ret); - - raw_spin_lock(&desc->lock); - if (likely(!(desc->status & IRQ_PENDING))) - break; - desc->status &= ~IRQ_PENDING; - } - desc->status &= ~IRQ_INPROGRESS; - -out: - /* - * The ->end() handler has to deal with interrupts which got - * disabled while the handler was running. - */ - desc->chip->end(irq); - raw_spin_unlock(&desc->lock); - - return 1; -} -#endif - -void early_init_irq_lock_class(void) -{ - struct irq_desc *desc; - int i; - - for_each_irq_desc(i, desc) { - lockdep_set_class(&desc->lock, &irq_desc_lock_class); - } -} - -unsigned int kstat_irqs_cpu(unsigned int irq, int cpu) -{ - struct irq_desc *desc = irq_to_desc(irq); - return desc ? desc->kstat_irqs[cpu] : 0; -} -EXPORT_SYMBOL(kstat_irqs_cpu); - diff --git a/kernel/irq/internals.h b/kernel/irq/internals.h index c63f3bc88f0..4571ae7e085 100644 --- a/kernel/irq/internals.h +++ b/kernel/irq/internals.h @@ -1,9 +1,12 @@ /* * IRQ subsystem internal functions and variables: */ +#include <linux/irqdesc.h> extern int noirqdebug; +#define irq_data_to_desc(data) container_of(data, struct irq_desc, irq_data) + /* Set default functions for irq_chip structures: */ extern void irq_chip_set_defaults(struct irq_chip *chip); @@ -15,21 +18,19 @@ extern int __irq_set_trigger(struct irq_desc *desc, unsigned int irq, 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 node, int nr); -extern void clear_kstat_irqs(struct irq_desc *desc); -extern raw_spinlock_t sparse_irq_lock; -#ifdef CONFIG_SPARSE_IRQ -void replace_irq_desc(unsigned int irq, struct irq_desc *desc); -#endif +/* Resending of interrupts :*/ +void check_irq_resend(struct irq_desc *desc, unsigned int irq); #ifdef CONFIG_PROC_FS extern void register_irq_proc(unsigned int irq, struct irq_desc *desc); +extern void unregister_irq_proc(unsigned int irq, struct irq_desc *desc); extern void register_handler_proc(unsigned int irq, struct irqaction *action); extern void unregister_handler_proc(unsigned int irq, struct irqaction *action); #else static inline void register_irq_proc(unsigned int irq, struct irq_desc *desc) { } +static inline void unregister_irq_proc(unsigned int irq, struct irq_desc *desc) { } static inline void register_handler_proc(unsigned int irq, struct irqaction *action) { } static inline void unregister_handler_proc(unsigned int irq, @@ -40,17 +41,27 @@ extern int irq_select_affinity_usr(unsigned int irq); extern void irq_set_thread_affinity(struct irq_desc *desc); +#ifndef CONFIG_GENERIC_HARDIRQS_NO_DEPRECATED +static inline void irq_end(unsigned int irq, struct irq_desc *desc) +{ + if (desc->irq_data.chip && desc->irq_data.chip->end) + desc->irq_data.chip->end(irq); +} +#else +static inline void irq_end(unsigned int irq, struct irq_desc *desc) { } +#endif + /* Inline functions for support of irq chips on slow busses */ -static inline void chip_bus_lock(unsigned int irq, struct irq_desc *desc) +static inline void chip_bus_lock(struct irq_desc *desc) { - if (unlikely(desc->chip->bus_lock)) - desc->chip->bus_lock(irq); + if (unlikely(desc->irq_data.chip->irq_bus_lock)) + desc->irq_data.chip->irq_bus_lock(&desc->irq_data); } -static inline void chip_bus_sync_unlock(unsigned int irq, struct irq_desc *desc) +static inline void chip_bus_sync_unlock(struct irq_desc *desc) { - if (unlikely(desc->chip->bus_sync_unlock)) - desc->chip->bus_sync_unlock(irq); + if (unlikely(desc->irq_data.chip->irq_bus_sync_unlock)) + desc->irq_data.chip->irq_bus_sync_unlock(&desc->irq_data); } /* @@ -67,8 +78,8 @@ static inline void print_irq_desc(unsigned int irq, struct irq_desc *desc) irq, desc, desc->depth, desc->irq_count, desc->irqs_unhandled); printk("->handle_irq(): %p, ", desc->handle_irq); print_symbol("%s\n", (unsigned long)desc->handle_irq); - printk("->chip(): %p, ", desc->chip); - print_symbol("%s\n", (unsigned long)desc->chip); + printk("->irq_data.chip(): %p, ", desc->irq_data.chip); + print_symbol("%s\n", (unsigned long)desc->irq_data.chip); printk("->action(): %p\n", desc->action); if (desc->action) { printk("->action->handler(): %p, ", desc->action->handler); diff --git a/kernel/irq/irqdesc.c b/kernel/irq/irqdesc.c new file mode 100644 index 00000000000..282f20230e6 --- /dev/null +++ b/kernel/irq/irqdesc.c @@ -0,0 +1,430 @@ +/* + * Copyright (C) 1992, 1998-2006 Linus Torvalds, Ingo Molnar + * Copyright (C) 2005-2006, Thomas Gleixner, Russell King + * + * This file contains the interrupt descriptor management code + * + * Detailed information is available in Documentation/DocBook/genericirq + * + */ +#include <linux/irq.h> +#include <linux/slab.h> +#include <linux/module.h> +#include <linux/interrupt.h> +#include <linux/kernel_stat.h> +#include <linux/radix-tree.h> +#include <linux/bitmap.h> + +#include "internals.h" + +/* + * lockdep: we want to handle all irq_desc locks as a single lock-class: + */ +static struct lock_class_key irq_desc_lock_class; + +#if defined(CONFIG_SMP) && defined(CONFIG_GENERIC_HARDIRQS) +static void __init init_irq_default_affinity(void) +{ + alloc_cpumask_var(&irq_default_affinity, GFP_NOWAIT); + cpumask_setall(irq_default_affinity); +} +#else +static void __init init_irq_default_affinity(void) +{ +} +#endif + +#ifdef CONFIG_SMP +static int alloc_masks(struct irq_desc *desc, gfp_t gfp, int node) +{ + if (!zalloc_cpumask_var_node(&desc->irq_data.affinity, gfp, node)) + return -ENOMEM; + +#ifdef CONFIG_GENERIC_PENDING_IRQ + if (!zalloc_cpumask_var_node(&desc->pending_mask, gfp, node)) { + free_cpumask_var(desc->irq_data.affinity); + return -ENOMEM; + } +#endif + return 0; +} + +static void desc_smp_init(struct irq_desc *desc, int node) +{ + desc->irq_data.node = node; + cpumask_copy(desc->irq_data.affinity, irq_default_affinity); +#ifdef CONFIG_GENERIC_PENDING_IRQ + cpumask_clear(desc->pending_mask); +#endif +} + +static inline int desc_node(struct irq_desc *desc) +{ + return desc->irq_data.node; +} + +#else +static inline int +alloc_masks(struct irq_desc *desc, gfp_t gfp, int node) { return 0; } +static inline void desc_smp_init(struct irq_desc *desc, int node) { } +static inline int desc_node(struct irq_desc *desc) { return 0; } +#endif + +static void desc_set_defaults(unsigned int irq, struct irq_desc *desc, int node) +{ + int cpu; + + desc->irq_data.irq = irq; + desc->irq_data.chip = &no_irq_chip; + desc->irq_data.chip_data = NULL; + desc->irq_data.handler_data = NULL; + desc->irq_data.msi_desc = NULL; + desc->status = IRQ_DEFAULT_INIT_FLAGS; + desc->handle_irq = handle_bad_irq; + desc->depth = 1; + desc->irq_count = 0; + desc->irqs_unhandled = 0; + desc->name = NULL; + for_each_possible_cpu(cpu) + *per_cpu_ptr(desc->kstat_irqs, cpu) = 0; + desc_smp_init(desc, node); +} + +int nr_irqs = NR_IRQS; +EXPORT_SYMBOL_GPL(nr_irqs); + +static DEFINE_MUTEX(sparse_irq_lock); +static DECLARE_BITMAP(allocated_irqs, NR_IRQS); + +#ifdef CONFIG_SPARSE_IRQ + +static RADIX_TREE(irq_desc_tree, GFP_KERNEL); + +static void irq_insert_desc(unsigned int irq, struct irq_desc *desc) +{ + radix_tree_insert(&irq_desc_tree, irq, desc); +} + +struct irq_desc *irq_to_desc(unsigned int irq) +{ + return radix_tree_lookup(&irq_desc_tree, irq); +} + +static void delete_irq_desc(unsigned int irq) +{ + radix_tree_delete(&irq_desc_tree, irq); +} + +#ifdef CONFIG_SMP +static void free_masks(struct irq_desc *desc) +{ +#ifdef CONFIG_GENERIC_PENDING_IRQ + free_cpumask_var(desc->pending_mask); +#endif + free_cpumask_var(desc->irq_data.affinity); +} +#else +static inline void free_masks(struct irq_desc *desc) { } +#endif + +static struct irq_desc *alloc_desc(int irq, int node) +{ + struct irq_desc *desc; + gfp_t gfp = GFP_KERNEL; + + desc = kzalloc_node(sizeof(*desc), gfp, node); + if (!desc) + return NULL; + /* allocate based on nr_cpu_ids */ + desc->kstat_irqs = alloc_percpu(unsigned int); + if (!desc->kstat_irqs) + goto err_desc; + + if (alloc_masks(desc, gfp, node)) + goto err_kstat; + + raw_spin_lock_init(&desc->lock); + lockdep_set_class(&desc->lock, &irq_desc_lock_class); + + desc_set_defaults(irq, desc, node); + + return desc; + +err_kstat: + free_percpu(desc->kstat_irqs); +err_desc: + kfree(desc); + return NULL; +} + +static void free_desc(unsigned int irq) +{ + struct irq_desc *desc = irq_to_desc(irq); + + unregister_irq_proc(irq, desc); + + mutex_lock(&sparse_irq_lock); + delete_irq_desc(irq); + mutex_unlock(&sparse_irq_lock); + + free_masks(desc); + free_percpu(desc->kstat_irqs); + kfree(desc); +} + +static int alloc_descs(unsigned int start, unsigned int cnt, int node) +{ + struct irq_desc *desc; + int i; + + for (i = 0; i < cnt; i++) { + desc = alloc_desc(start + i, node); + if (!desc) + goto err; + mutex_lock(&sparse_irq_lock); + irq_insert_desc(start + i, desc); + mutex_unlock(&sparse_irq_lock); + } + return start; + +err: + for (i--; i >= 0; i--) + free_desc(start + i); + + mutex_lock(&sparse_irq_lock); + bitmap_clear(allocated_irqs, start, cnt); + mutex_unlock(&sparse_irq_lock); + return -ENOMEM; +} + +struct irq_desc * __ref irq_to_desc_alloc_node(unsigned int irq, int node) +{ + int res = irq_alloc_descs(irq, irq, 1, node); + + if (res == -EEXIST || res == irq) + return irq_to_desc(irq); + return NULL; +} + +int __init early_irq_init(void) +{ + int i, initcnt, node = first_online_node; + struct irq_desc *desc; + + init_irq_default_affinity(); + + /* Let arch update nr_irqs and return the nr of preallocated irqs */ + initcnt = arch_probe_nr_irqs(); + printk(KERN_INFO "NR_IRQS:%d nr_irqs:%d %d\n", NR_IRQS, nr_irqs, initcnt); + + for (i = 0; i < initcnt; i++) { + desc = alloc_desc(i, node); + set_bit(i, allocated_irqs); + irq_insert_desc(i, desc); + } + return arch_early_irq_init(); +} + +#else /* !CONFIG_SPARSE_IRQ */ + +struct irq_desc irq_desc[NR_IRQS] __cacheline_aligned_in_smp = { + [0 ... NR_IRQS-1] = { + .status = IRQ_DEFAULT_INIT_FLAGS, + .handle_irq = handle_bad_irq, + .depth = 1, + .lock = __RAW_SPIN_LOCK_UNLOCKED(irq_desc->lock), + } +}; + +int __init early_irq_init(void) +{ + int count, i, node = first_online_node; + struct irq_desc *desc; + + init_irq_default_affinity(); + + printk(KERN_INFO "NR_IRQS:%d\n", NR_IRQS); + + desc = irq_desc; + count = ARRAY_SIZE(irq_desc); + + for (i = 0; i < count; i++) { + desc[i].irq_data.irq = i; + desc[i].irq_data.chip = &no_irq_chip; + /* TODO : do this allocation on-demand ... */ + desc[i].kstat_irqs = alloc_percpu(unsigned int); + alloc_masks(desc + i, GFP_KERNEL, node); + desc_smp_init(desc + i, node); + lockdep_set_class(&desc[i].lock, &irq_desc_lock_class); + } + return arch_early_irq_init(); +} + +struct irq_desc *irq_to_desc(unsigned int irq) +{ + return (irq < NR_IRQS) ? irq_desc + irq : NULL; +} + +struct irq_desc *irq_to_desc_alloc_node(unsigned int irq, int node) +{ + return irq_to_desc(irq); +} + +static void free_desc(unsigned int irq) +{ + dynamic_irq_cleanup(irq); +} + +static inline int alloc_descs(unsigned int start, unsigned int cnt, int node) +{ +#if defined(CONFIG_KSTAT_IRQS_ONDEMAND) + struct irq_desc *desc; + unsigned int i; + + for (i = 0; i < cnt; i++) { + desc = irq_to_desc(start + i); + if (desc && !desc->kstat_irqs) { + unsigned int __percpu *stats = alloc_percpu(unsigned int); + + if (!stats) + return -1; + if (cmpxchg(&desc->kstat_irqs, NULL, stats) != NULL) + free_percpu(stats); + } + } +#endif + return start; +} +#endif /* !CONFIG_SPARSE_IRQ */ + +/* Dynamic interrupt handling */ + +/** + * irq_free_descs - free irq descriptors + * @from: Start of descriptor range + * @cnt: Number of consecutive irqs to free + */ +void irq_free_descs(unsigned int from, unsigned int cnt) +{ + int i; + + if (from >= nr_irqs || (from + cnt) > nr_irqs) + return; + + for (i = 0; i < cnt; i++) + free_desc(from + i); + + mutex_lock(&sparse_irq_lock); + bitmap_clear(allocated_irqs, from, cnt); + mutex_unlock(&sparse_irq_lock); +} + +/** + * irq_alloc_descs - allocate and initialize a range of irq descriptors + * @irq: Allocate for specific irq number if irq >= 0 + * @from: Start the search from this irq number + * @cnt: Number of consecutive irqs to allocate. + * @node: Preferred node on which the irq descriptor should be allocated + * + * Returns the first irq number or error code + */ +int __ref +irq_alloc_descs(int irq, unsigned int from, unsigned int cnt, int node) +{ + int start, ret; + + if (!cnt) + return -EINVAL; + + mutex_lock(&sparse_irq_lock); + + start = bitmap_find_next_zero_area(allocated_irqs, nr_irqs, from, cnt, 0); + ret = -EEXIST; + if (irq >=0 && start != irq) + goto err; + + ret = -ENOMEM; + if (start >= nr_irqs) + goto err; + + bitmap_set(allocated_irqs, start, cnt); + mutex_unlock(&sparse_irq_lock); + return alloc_descs(start, cnt, node); + +err: + mutex_unlock(&sparse_irq_lock); + return ret; +} + +/** + * irq_reserve_irqs - mark irqs allocated + * @from: mark from irq number + * @cnt: number of irqs to mark + * + * Returns 0 on success or an appropriate error code + */ +int irq_reserve_irqs(unsigned int from, unsigned int cnt) +{ + unsigned int start; + int ret = 0; + + if (!cnt || (from + cnt) > nr_irqs) + return -EINVAL; + + mutex_lock(&sparse_irq_lock); + start = bitmap_find_next_zero_area(allocated_irqs, nr_irqs, from, cnt, 0); + if (start == from) + bitmap_set(allocated_irqs, start, cnt); + else + ret = -EEXIST; + mutex_unlock(&sparse_irq_lock); + return ret; +} + +/** + * irq_get_next_irq - get next allocated irq number + * @offset: where to start the search + * + * Returns next irq number after offset or nr_irqs if none is found. + */ +unsigned int irq_get_next_irq(unsigned int offset) +{ + return find_next_bit(allocated_irqs, nr_irqs, offset); +} + +/** + * dynamic_irq_cleanup - cleanup a dynamically allocated irq + * @irq: irq number to initialize + */ +void dynamic_irq_cleanup(unsigned int irq) +{ + struct irq_desc *desc = irq_to_desc(irq); + unsigned long flags; + + raw_spin_lock_irqsave(&desc->lock, flags); + desc_set_defaults(irq, desc, desc_node(desc)); + raw_spin_unlock_irqrestore(&desc->lock, flags); +} + +unsigned int kstat_irqs_cpu(unsigned int irq, int cpu) +{ + struct irq_desc *desc = irq_to_desc(irq); + + return desc && desc->kstat_irqs ? + *per_cpu_ptr(desc->kstat_irqs, cpu) : 0; +} + +#ifdef CONFIG_GENERIC_HARDIRQS +unsigned int kstat_irqs(unsigned int irq) +{ + struct irq_desc *desc = irq_to_desc(irq); + int cpu; + int sum = 0; + + if (!desc || !desc->kstat_irqs) + return 0; + for_each_possible_cpu(cpu) + sum += *per_cpu_ptr(desc->kstat_irqs, cpu); + return sum; +} +#endif /* CONFIG_GENERIC_HARDIRQS */ diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c index 3164ba7ce15..0caa59f747d 100644 --- a/kernel/irq/manage.c +++ b/kernel/irq/manage.c @@ -73,8 +73,8 @@ int irq_can_set_affinity(unsigned int irq) { struct irq_desc *desc = irq_to_desc(irq); - if (CHECK_IRQ_PER_CPU(desc->status) || !desc->chip || - !desc->chip->set_affinity) + if (CHECK_IRQ_PER_CPU(desc->status) || !desc->irq_data.chip || + !desc->irq_data.chip->irq_set_affinity) return 0; return 1; @@ -109,17 +109,18 @@ void irq_set_thread_affinity(struct irq_desc *desc) int irq_set_affinity(unsigned int irq, const struct cpumask *cpumask) { struct irq_desc *desc = irq_to_desc(irq); + struct irq_chip *chip = desc->irq_data.chip; unsigned long flags; - if (!desc->chip->set_affinity) + if (!chip->irq_set_affinity) return -EINVAL; raw_spin_lock_irqsave(&desc->lock, flags); #ifdef CONFIG_GENERIC_PENDING_IRQ if (desc->status & IRQ_MOVE_PCNTXT) { - if (!desc->chip->set_affinity(irq, cpumask)) { - cpumask_copy(desc->affinity, cpumask); + if (!chip->irq_set_affinity(&desc->irq_data, cpumask, false)) { + cpumask_copy(desc->irq_data.affinity, cpumask); irq_set_thread_affinity(desc); } } @@ -128,8 +129,8 @@ int irq_set_affinity(unsigned int irq, const struct cpumask *cpumask) cpumask_copy(desc->pending_mask, cpumask); } #else - if (!desc->chip->set_affinity(irq, cpumask)) { - cpumask_copy(desc->affinity, cpumask); + if (!chip->irq_set_affinity(&desc->irq_data, cpumask, false)) { + cpumask_copy(desc->irq_data.affinity, cpumask); irq_set_thread_affinity(desc); } #endif @@ -168,16 +169,16 @@ static int setup_affinity(unsigned int irq, struct irq_desc *desc) * one of the targets is online. */ if (desc->status & (IRQ_AFFINITY_SET | IRQ_NO_BALANCING)) { - if (cpumask_any_and(desc->affinity, cpu_online_mask) + if (cpumask_any_and(desc->irq_data.affinity, cpu_online_mask) < nr_cpu_ids) goto set_affinity; else desc->status &= ~IRQ_AFFINITY_SET; } - cpumask_and(desc->affinity, cpu_online_mask, irq_default_affinity); + cpumask_and(desc->irq_data.affinity, cpu_online_mask, irq_default_affinity); set_affinity: - desc->chip->set_affinity(irq, desc->affinity); + desc->irq_data.chip->irq_set_affinity(&desc->irq_data, desc->irq_data.affinity, false); return 0; } @@ -216,14 +217,14 @@ static inline int setup_affinity(unsigned int irq, struct irq_desc *desc) void __disable_irq(struct irq_desc *desc, unsigned int irq, bool suspend) { if (suspend) { - if (!desc->action || (desc->action->flags & IRQF_TIMER)) + if (!desc->action || (desc->action->flags & IRQF_NO_SUSPEND)) return; desc->status |= IRQ_SUSPENDED; } if (!desc->depth++) { desc->status |= IRQ_DISABLED; - desc->chip->disable(irq); + desc->irq_data.chip->irq_disable(&desc->irq_data); } } @@ -246,11 +247,11 @@ void disable_irq_nosync(unsigned int irq) if (!desc) return; - chip_bus_lock(irq, desc); + chip_bus_lock(desc); raw_spin_lock_irqsave(&desc->lock, flags); __disable_irq(desc, irq, false); raw_spin_unlock_irqrestore(&desc->lock, flags); - chip_bus_sync_unlock(irq, desc); + chip_bus_sync_unlock(desc); } EXPORT_SYMBOL(disable_irq_nosync); @@ -313,7 +314,7 @@ void __enable_irq(struct irq_desc *desc, unsigned int irq, bool resume) * IRQ line is re-enabled. * * This function may be called from IRQ context only when - * desc->chip->bus_lock and desc->chip->bus_sync_unlock are NULL ! + * desc->irq_data.chip->bus_lock and desc->chip->bus_sync_unlock are NULL ! */ void enable_irq(unsigned int irq) { @@ -323,11 +324,15 @@ void enable_irq(unsigned int irq) if (!desc) return; - chip_bus_lock(irq, desc); + if (WARN(!desc->irq_data.chip || !desc->irq_data.chip->irq_enable, + KERN_ERR "enable_irq before setup/request_irq: irq %u\n", irq)) + return; + + chip_bus_lock(desc); raw_spin_lock_irqsave(&desc->lock, flags); __enable_irq(desc, irq, false); raw_spin_unlock_irqrestore(&desc->lock, flags); - chip_bus_sync_unlock(irq, desc); + chip_bus_sync_unlock(desc); } EXPORT_SYMBOL(enable_irq); @@ -336,8 +341,8 @@ static int set_irq_wake_real(unsigned int irq, unsigned int on) struct irq_desc *desc = irq_to_desc(irq); int ret = -ENXIO; - if (desc->chip->set_wake) - ret = desc->chip->set_wake(irq, on); + if (desc->irq_data.chip->irq_set_wake) + ret = desc->irq_data.chip->irq_set_wake(&desc->irq_data, on); return ret; } @@ -429,12 +434,12 @@ void compat_irq_chip_set_default_handler(struct irq_desc *desc) } int __irq_set_trigger(struct irq_desc *desc, unsigned int irq, - unsigned long flags) + unsigned long flags) { int ret; - struct irq_chip *chip = desc->chip; + struct irq_chip *chip = desc->irq_data.chip; - if (!chip || !chip->set_type) { + if (!chip || !chip->irq_set_type) { /* * IRQF_TRIGGER_* but the PIC does not support multiple * flow-types? @@ -445,17 +450,20 @@ int __irq_set_trigger(struct irq_desc *desc, unsigned int irq, } /* caller masked out all except trigger mode flags */ - ret = chip->set_type(irq, flags); + ret = chip->irq_set_type(&desc->irq_data, flags); if (ret) - pr_err("setting trigger mode %d for irq %u failed (%pF)\n", - (int)flags, irq, chip->set_type); + pr_err("setting trigger mode %lu for irq %u failed (%pF)\n", + flags, irq, chip->irq_set_type); else { if (flags & (IRQ_TYPE_LEVEL_LOW | IRQ_TYPE_LEVEL_HIGH)) flags |= IRQ_LEVEL; /* note that IRQF_TRIGGER_MASK == IRQ_TYPE_SENSE_MASK */ desc->status &= ~(IRQ_LEVEL | IRQ_TYPE_SENSE_MASK); desc->status |= flags; + + if (chip != desc->irq_data.chip) + irq_chip_set_defaults(desc->irq_data.chip); } return ret; @@ -504,7 +512,7 @@ static int irq_wait_for_interrupt(struct irqaction *action) static void irq_finalize_oneshot(unsigned int irq, struct irq_desc *desc) { again: - chip_bus_lock(irq, desc); + chip_bus_lock(desc); raw_spin_lock_irq(&desc->lock); /* @@ -518,17 +526,17 @@ again: */ if (unlikely(desc->status & IRQ_INPROGRESS)) { raw_spin_unlock_irq(&desc->lock); - chip_bus_sync_unlock(irq, desc); + chip_bus_sync_unlock(desc); cpu_relax(); goto again; } if (!(desc->status & IRQ_DISABLED) && (desc->status & IRQ_MASKED)) { desc->status &= ~IRQ_MASKED; - desc->chip->unmask(irq); + desc->irq_data.chip->irq_unmask(&desc->irq_data); } raw_spin_unlock_irq(&desc->lock); - chip_bus_sync_unlock(irq, desc); + chip_bus_sync_unlock(desc); } #ifdef CONFIG_SMP @@ -553,7 +561,7 @@ irq_thread_check_affinity(struct irq_desc *desc, struct irqaction *action) } raw_spin_lock_irq(&desc->lock); - cpumask_copy(mask, desc->affinity); + cpumask_copy(mask, desc->irq_data.affinity); raw_spin_unlock_irq(&desc->lock); set_cpus_allowed_ptr(current, mask); @@ -569,7 +577,9 @@ irq_thread_check_affinity(struct irq_desc *desc, struct irqaction *action) { } */ static int irq_thread(void *data) { - struct sched_param param = { .sched_priority = MAX_USER_RT_PRIO/2, }; + static const 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; @@ -654,7 +664,7 @@ __setup_irq(unsigned int irq, struct irq_desc *desc, struct irqaction *new) if (!desc) return -EINVAL; - if (desc->chip == &no_irq_chip) + if (desc->irq_data.chip == &no_irq_chip) return -ENOSYS; /* * Some drivers like serial.c use request_irq() heavily, @@ -749,7 +759,7 @@ __setup_irq(unsigned int irq, struct irq_desc *desc, struct irqaction *new) } if (!shared) { - irq_chip_set_defaults(desc->chip); + irq_chip_set_defaults(desc->irq_data.chip); init_waitqueue_head(&desc->wait_for_threads); @@ -776,7 +786,7 @@ __setup_irq(unsigned int irq, struct irq_desc *desc, struct irqaction *new) if (!(desc->status & IRQ_NOAUTOEN)) { desc->depth = 0; desc->status &= ~IRQ_DISABLED; - desc->chip->startup(irq); + desc->irq_data.chip->irq_startup(&desc->irq_data); } else /* Undo nested disables: */ desc->depth = 1; @@ -909,17 +919,17 @@ static struct irqaction *__free_irq(unsigned int irq, void *dev_id) /* Currently used only by UML, might disappear one day: */ #ifdef CONFIG_IRQ_RELEASE_METHOD - if (desc->chip->release) - desc->chip->release(irq, dev_id); + if (desc->irq_data.chip->release) + desc->irq_data.chip->release(irq, dev_id); #endif /* If this was the last handler, shut down the IRQ line: */ if (!desc->action) { desc->status |= IRQ_DISABLED; - if (desc->chip->shutdown) - desc->chip->shutdown(irq); + if (desc->irq_data.chip->irq_shutdown) + desc->irq_data.chip->irq_shutdown(&desc->irq_data); else - desc->chip->disable(irq); + desc->irq_data.chip->irq_disable(&desc->irq_data); } #ifdef CONFIG_SMP @@ -994,9 +1004,9 @@ void free_irq(unsigned int irq, void *dev_id) if (!desc) return; - chip_bus_lock(irq, desc); + chip_bus_lock(desc); kfree(__free_irq(irq, dev_id)); - chip_bus_sync_unlock(irq, desc); + chip_bus_sync_unlock(desc); } EXPORT_SYMBOL(free_irq); @@ -1083,9 +1093,9 @@ int request_threaded_irq(unsigned int irq, irq_handler_t handler, action->name = devname; action->dev_id = dev_id; - chip_bus_lock(irq, desc); + chip_bus_lock(desc); retval = __setup_irq(irq, desc, action); - chip_bus_sync_unlock(irq, desc); + chip_bus_sync_unlock(desc); if (retval) kfree(action); diff --git a/kernel/irq/migration.c b/kernel/irq/migration.c index 24196228083..1d254194048 100644 --- a/kernel/irq/migration.c +++ b/kernel/irq/migration.c @@ -7,6 +7,7 @@ void move_masked_irq(int irq) { struct irq_desc *desc = irq_to_desc(irq); + struct irq_chip *chip = desc->irq_data.chip; if (likely(!(desc->status & IRQ_MOVE_PENDING))) return; @@ -24,7 +25,7 @@ void move_masked_irq(int irq) if (unlikely(cpumask_empty(desc->pending_mask))) return; - if (!desc->chip->set_affinity) + if (!chip->irq_set_affinity) return; assert_raw_spin_locked(&desc->lock); @@ -43,8 +44,9 @@ void move_masked_irq(int irq) */ if (likely(cpumask_any_and(desc->pending_mask, cpu_online_mask) < nr_cpu_ids)) - if (!desc->chip->set_affinity(irq, desc->pending_mask)) { - cpumask_copy(desc->affinity, desc->pending_mask); + if (!chip->irq_set_affinity(&desc->irq_data, + desc->pending_mask, false)) { + cpumask_copy(desc->irq_data.affinity, desc->pending_mask); irq_set_thread_affinity(desc); } @@ -61,8 +63,8 @@ void move_native_irq(int irq) if (unlikely(desc->status & IRQ_DISABLED)) return; - desc->chip->mask(irq); + desc->irq_data.chip->irq_mask(&desc->irq_data); move_masked_irq(irq); - desc->chip->unmask(irq); + desc->irq_data.chip->irq_unmask(&desc->irq_data); } diff --git a/kernel/irq/numa_migrate.c b/kernel/irq/numa_migrate.c deleted file mode 100644 index 65d3845665a..00000000000 --- a/kernel/irq/numa_migrate.c +++ /dev/null @@ -1,120 +0,0 @@ -/* - * NUMA irq-desc migration code - * - * Migrate IRQ data structures (irq_desc, chip_data, etc.) over to - * the new "home node" of the IRQ. - */ - -#include <linux/irq.h> -#include <linux/slab.h> -#include <linux/module.h> -#include <linux/random.h> -#include <linux/interrupt.h> -#include <linux/kernel_stat.h> - -#include "internals.h" - -static void init_copy_kstat_irqs(struct irq_desc *old_desc, - struct irq_desc *desc, - int node, int nr) -{ - init_kstat_irqs(desc, node, nr); - - if (desc->kstat_irqs != old_desc->kstat_irqs) - memcpy(desc->kstat_irqs, old_desc->kstat_irqs, - nr * sizeof(*desc->kstat_irqs)); -} - -static void free_kstat_irqs(struct irq_desc *old_desc, struct irq_desc *desc) -{ - if (old_desc->kstat_irqs == desc->kstat_irqs) - return; - - kfree(old_desc->kstat_irqs); - old_desc->kstat_irqs = NULL; -} - -static bool init_copy_one_irq_desc(int irq, struct irq_desc *old_desc, - struct irq_desc *desc, int node) -{ - memcpy(desc, old_desc, sizeof(struct irq_desc)); - 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; - } - raw_spin_lock_init(&desc->lock); - desc->node = node; - lockdep_set_class(&desc->lock, &irq_desc_lock_class); - 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, 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 node) -{ - struct irq_desc *desc; - unsigned int irq; - unsigned long flags; - - irq = old_desc->irq; - - raw_spin_lock_irqsave(&sparse_irq_lock, flags); - - /* We have to check it to avoid races with another CPU */ - desc = irq_to_desc(irq); - - if (desc && old_desc != desc) - goto out_unlock; - - desc = kzalloc_node(sizeof(*desc), GFP_ATOMIC, node); - if (!desc) { - printk(KERN_ERR "irq %d: can not get new irq_desc " - "for migration.\n", irq); - /* still use old one */ - desc = old_desc; - goto out_unlock; - } - if (!init_copy_one_irq_desc(irq, old_desc, desc, node)) { - /* still use old one */ - kfree(desc); - desc = old_desc; - goto out_unlock; - } - - replace_irq_desc(irq, desc); - raw_spin_unlock_irqrestore(&sparse_irq_lock, flags); - - /* free the old one */ - free_one_irq_desc(old_desc, desc); - kfree(old_desc); - - return desc; - -out_unlock: - raw_spin_unlock_irqrestore(&sparse_irq_lock, flags); - - return desc; -} - -struct irq_desc *move_irq_desc(struct irq_desc *desc, int node) -{ - /* those static or target node is -1, do not move them */ - if (desc->irq < NR_IRQS_LEGACY || node == -1) - return desc; - - if (desc->node != node) - desc = __real_move_irq_desc(desc, node); - - return desc; -} - diff --git a/kernel/irq/proc.c b/kernel/irq/proc.c index 09a2ee540bd..6c8a2a9f8a7 100644 --- a/kernel/irq/proc.c +++ b/kernel/irq/proc.c @@ -21,7 +21,7 @@ static struct proc_dir_entry *root_irq_dir; static int irq_affinity_proc_show(struct seq_file *m, void *v) { struct irq_desc *desc = irq_to_desc((long)m->private); - const struct cpumask *mask = desc->affinity; + const struct cpumask *mask = desc->irq_data.affinity; #ifdef CONFIG_GENERIC_PENDING_IRQ if (desc->status & IRQ_MOVE_PENDING) @@ -65,7 +65,7 @@ static ssize_t irq_affinity_proc_write(struct file *file, cpumask_var_t new_value; int err; - if (!irq_to_desc(irq)->chip->set_affinity || no_irq_affinity || + if (!irq_to_desc(irq)->irq_data.chip->irq_set_affinity || no_irq_affinity || irq_balancing_disabled(irq)) return -EIO; @@ -185,7 +185,7 @@ static int irq_node_proc_show(struct seq_file *m, void *v) { struct irq_desc *desc = irq_to_desc((long) m->private); - seq_printf(m, "%d\n", desc->node); + seq_printf(m, "%d\n", desc->irq_data.node); return 0; } @@ -214,7 +214,7 @@ static int irq_spurious_proc_show(struct seq_file *m, void *v) static int irq_spurious_proc_open(struct inode *inode, struct file *file) { - return single_open(file, irq_spurious_proc_show, NULL); + return single_open(file, irq_spurious_proc_show, PDE(inode)->data); } static const struct file_operations irq_spurious_proc_fops = { @@ -269,7 +269,7 @@ void register_irq_proc(unsigned int irq, struct irq_desc *desc) { char name [MAX_NAMELEN]; - if (!root_irq_dir || (desc->chip == &no_irq_chip) || desc->dir) + if (!root_irq_dir || (desc->irq_data.chip == &no_irq_chip) || desc->dir) return; memset(name, 0, MAX_NAMELEN); @@ -297,6 +297,24 @@ void register_irq_proc(unsigned int irq, struct irq_desc *desc) &irq_spurious_proc_fops, (void *)(long)irq); } +void unregister_irq_proc(unsigned int irq, struct irq_desc *desc) +{ + char name [MAX_NAMELEN]; + + if (!root_irq_dir || !desc->dir) + return; +#ifdef CONFIG_SMP + remove_proc_entry("smp_affinity", desc->dir); + remove_proc_entry("affinity_hint", desc->dir); + remove_proc_entry("node", desc->dir); +#endif + remove_proc_entry("spurious", desc->dir); + + memset(name, 0, MAX_NAMELEN); + sprintf(name, "%u", irq); + remove_proc_entry(name, root_irq_dir); +} + #undef MAX_NAMELEN void unregister_handler_proc(unsigned int irq, struct irqaction *action) diff --git a/kernel/irq/resend.c b/kernel/irq/resend.c index 090c3763f3a..891115a929a 100644 --- a/kernel/irq/resend.c +++ b/kernel/irq/resend.c @@ -60,7 +60,7 @@ void check_irq_resend(struct irq_desc *desc, unsigned int irq) /* * Make sure the interrupt is enabled, before resending it: */ - desc->chip->enable(irq); + desc->irq_data.chip->irq_enable(&desc->irq_data); /* * We do not resend level type interrupts. Level type @@ -70,7 +70,8 @@ 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->retrigger || !desc->chip->retrigger(irq)) { + if (!desc->irq_data.chip->irq_retrigger || + !desc->irq_data.chip->irq_retrigger(&desc->irq_data)) { #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 89fb90ae534..3089d3b9d5f 100644 --- a/kernel/irq/spurious.c +++ b/kernel/irq/spurious.c @@ -14,6 +14,8 @@ #include <linux/moduleparam.h> #include <linux/timer.h> +#include "internals.h" + static int irqfixup __read_mostly; #define POLL_SPURIOUS_IRQ_INTERVAL (HZ/10) @@ -78,8 +80,8 @@ static int try_one_irq(int irq, struct irq_desc *desc) * If we did actual work for the real IRQ line we must let the * IRQ controller clean up too */ - if (work && desc->chip && desc->chip->end) - desc->chip->end(irq); + if (work) + irq_end(irq, desc); raw_spin_unlock(&desc->lock); return ok; @@ -254,7 +256,7 @@ void note_interrupt(unsigned int irq, struct irq_desc *desc, printk(KERN_EMERG "Disabling IRQ #%d\n", irq); desc->status |= IRQ_DISABLED | IRQ_SPURIOUS_DISABLED; desc->depth++; - desc->chip->disable(irq); + desc->irq_data.chip->irq_disable(&desc->irq_data); mod_timer(&poll_spurious_irq_timer, jiffies + POLL_SPURIOUS_IRQ_INTERVAL); diff --git a/kernel/irq_work.c b/kernel/irq_work.c new file mode 100644 index 00000000000..c58fa7da8ae --- /dev/null +++ b/kernel/irq_work.c @@ -0,0 +1,166 @@ +/* + * Copyright (C) 2010 Red Hat, Inc., Peter Zijlstra <pzijlstr@redhat.com> + * + * Provides a framework for enqueueing and running callbacks from hardirq + * context. The enqueueing is NMI-safe. + */ + +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/irq_work.h> +#include <linux/hardirq.h> + +/* + * An entry can be in one of four states: + * + * free NULL, 0 -> {claimed} : free to be used + * claimed NULL, 3 -> {pending} : claimed to be enqueued + * pending next, 3 -> {busy} : queued, pending callback + * busy NULL, 2 -> {free, claimed} : callback in progress, can be claimed + * + * We use the lower two bits of the next pointer to keep PENDING and BUSY + * flags. + */ + +#define IRQ_WORK_PENDING 1UL +#define IRQ_WORK_BUSY 2UL +#define IRQ_WORK_FLAGS 3UL + +static inline bool irq_work_is_set(struct irq_work *entry, int flags) +{ + return (unsigned long)entry->next & flags; +} + +static inline struct irq_work *irq_work_next(struct irq_work *entry) +{ + unsigned long next = (unsigned long)entry->next; + next &= ~IRQ_WORK_FLAGS; + return (struct irq_work *)next; +} + +static inline struct irq_work *next_flags(struct irq_work *entry, int flags) +{ + unsigned long next = (unsigned long)entry; + next |= flags; + return (struct irq_work *)next; +} + +static DEFINE_PER_CPU(struct irq_work *, irq_work_list); + +/* + * Claim the entry so that no one else will poke at it. + */ +static bool irq_work_claim(struct irq_work *entry) +{ + struct irq_work *next, *nflags; + + do { + next = entry->next; + if ((unsigned long)next & IRQ_WORK_PENDING) + return false; + nflags = next_flags(next, IRQ_WORK_FLAGS); + } while (cmpxchg(&entry->next, next, nflags) != next); + + return true; +} + + +void __weak arch_irq_work_raise(void) +{ + /* + * Lame architectures will get the timer tick callback + */ +} + +/* + * Queue the entry and raise the IPI if needed. + */ +static void __irq_work_queue(struct irq_work *entry) +{ + struct irq_work *next; + + preempt_disable(); + + do { + next = __this_cpu_read(irq_work_list); + /* Can assign non-atomic because we keep the flags set. */ + entry->next = next_flags(next, IRQ_WORK_FLAGS); + } while (this_cpu_cmpxchg(irq_work_list, next, entry) != next); + + /* The list was empty, raise self-interrupt to start processing. */ + if (!irq_work_next(entry)) + arch_irq_work_raise(); + + preempt_enable(); +} + +/* + * Enqueue the irq_work @entry, returns true on success, failure when the + * @entry was already enqueued by someone else. + * + * Can be re-enqueued while the callback is still in progress. + */ +bool irq_work_queue(struct irq_work *entry) +{ + if (!irq_work_claim(entry)) { + /* + * Already enqueued, can't do! + */ + return false; + } + + __irq_work_queue(entry); + return true; +} +EXPORT_SYMBOL_GPL(irq_work_queue); + +/* + * Run the irq_work entries on this cpu. Requires to be ran from hardirq + * context with local IRQs disabled. + */ +void irq_work_run(void) +{ + struct irq_work *list; + + if (this_cpu_read(irq_work_list) == NULL) + return; + + BUG_ON(!in_irq()); + BUG_ON(!irqs_disabled()); + + list = this_cpu_xchg(irq_work_list, NULL); + + while (list != NULL) { + struct irq_work *entry = list; + + list = irq_work_next(list); + + /* + * Clear the PENDING bit, after this point the @entry + * can be re-used. + */ + entry->next = next_flags(NULL, IRQ_WORK_BUSY); + entry->func(entry); + /* + * Clear the BUSY bit and return to the free state if + * no-one else claimed it meanwhile. + */ + (void)cmpxchg(&entry->next, + next_flags(NULL, IRQ_WORK_BUSY), + NULL); + } +} +EXPORT_SYMBOL_GPL(irq_work_run); + +/* + * Synchronize against the irq_work @entry, ensures the entry is not + * currently in use. + */ +void irq_work_sync(struct irq_work *entry) +{ + WARN_ON_ONCE(irqs_disabled()); + + while (irq_work_is_set(entry, IRQ_WORK_BUSY)) + cpu_relax(); +} +EXPORT_SYMBOL_GPL(irq_work_sync); diff --git a/kernel/jump_label.c b/kernel/jump_label.c new file mode 100644 index 00000000000..3b79bd93833 --- /dev/null +++ b/kernel/jump_label.c @@ -0,0 +1,484 @@ +/* + * jump label support + * + * Copyright (C) 2009 Jason Baron <jbaron@redhat.com> + * + */ +#include <linux/jump_label.h> +#include <linux/memory.h> +#include <linux/uaccess.h> +#include <linux/module.h> +#include <linux/list.h> +#include <linux/jhash.h> +#include <linux/slab.h> +#include <linux/sort.h> +#include <linux/err.h> + +#ifdef HAVE_JUMP_LABEL + +#define JUMP_LABEL_HASH_BITS 6 +#define JUMP_LABEL_TABLE_SIZE (1 << JUMP_LABEL_HASH_BITS) +static struct hlist_head jump_label_table[JUMP_LABEL_TABLE_SIZE]; + +/* mutex to protect coming/going of the the jump_label table */ +static DEFINE_MUTEX(jump_label_mutex); + +struct jump_label_entry { + struct hlist_node hlist; + struct jump_entry *table; + int nr_entries; + /* hang modules off here */ + struct hlist_head modules; + unsigned long key; +}; + +struct jump_label_module_entry { + struct hlist_node hlist; + struct jump_entry *table; + int nr_entries; + struct module *mod; +}; + +void jump_label_lock(void) +{ + mutex_lock(&jump_label_mutex); +} + +void jump_label_unlock(void) +{ + mutex_unlock(&jump_label_mutex); +} + +static int jump_label_cmp(const void *a, const void *b) +{ + const struct jump_entry *jea = a; + const struct jump_entry *jeb = b; + + if (jea->key < jeb->key) + return -1; + + if (jea->key > jeb->key) + return 1; + + return 0; +} + +static void +sort_jump_label_entries(struct jump_entry *start, struct jump_entry *stop) +{ + unsigned long size; + + size = (((unsigned long)stop - (unsigned long)start) + / sizeof(struct jump_entry)); + sort(start, size, sizeof(struct jump_entry), jump_label_cmp, NULL); +} + +static struct jump_label_entry *get_jump_label_entry(jump_label_t key) +{ + struct hlist_head *head; + struct hlist_node *node; + struct jump_label_entry *e; + u32 hash = jhash((void *)&key, sizeof(jump_label_t), 0); + + head = &jump_label_table[hash & (JUMP_LABEL_TABLE_SIZE - 1)]; + hlist_for_each_entry(e, node, head, hlist) { + if (key == e->key) + return e; + } + return NULL; +} + +static struct jump_label_entry * +add_jump_label_entry(jump_label_t key, int nr_entries, struct jump_entry *table) +{ + struct hlist_head *head; + struct jump_label_entry *e; + u32 hash; + + e = get_jump_label_entry(key); + if (e) + return ERR_PTR(-EEXIST); + + e = kmalloc(sizeof(struct jump_label_entry), GFP_KERNEL); + if (!e) + return ERR_PTR(-ENOMEM); + + hash = jhash((void *)&key, sizeof(jump_label_t), 0); + head = &jump_label_table[hash & (JUMP_LABEL_TABLE_SIZE - 1)]; + e->key = key; + e->table = table; + e->nr_entries = nr_entries; + INIT_HLIST_HEAD(&(e->modules)); + hlist_add_head(&e->hlist, head); + return e; +} + +static int +build_jump_label_hashtable(struct jump_entry *start, struct jump_entry *stop) +{ + struct jump_entry *iter, *iter_begin; + struct jump_label_entry *entry; + int count; + + sort_jump_label_entries(start, stop); + iter = start; + while (iter < stop) { + entry = get_jump_label_entry(iter->key); + if (!entry) { + iter_begin = iter; + count = 0; + while ((iter < stop) && + (iter->key == iter_begin->key)) { + iter++; + count++; + } + entry = add_jump_label_entry(iter_begin->key, + count, iter_begin); + if (IS_ERR(entry)) + return PTR_ERR(entry); + } else { + WARN_ONCE(1, KERN_ERR "build_jump_hashtable: unexpected entry!\n"); + return -1; + } + } + return 0; +} + +/*** + * jump_label_update - update jump label text + * @key - key value associated with a a jump label + * @type - enum set to JUMP_LABEL_ENABLE or JUMP_LABEL_DISABLE + * + * Will enable/disable the jump for jump label @key, depending on the + * value of @type. + * + */ + +void jump_label_update(unsigned long key, enum jump_label_type type) +{ + struct jump_entry *iter; + struct jump_label_entry *entry; + struct hlist_node *module_node; + struct jump_label_module_entry *e_module; + int count; + + jump_label_lock(); + entry = get_jump_label_entry((jump_label_t)key); + if (entry) { + count = entry->nr_entries; + iter = entry->table; + while (count--) { + if (kernel_text_address(iter->code)) + arch_jump_label_transform(iter, type); + iter++; + } + /* eanble/disable jump labels in modules */ + hlist_for_each_entry(e_module, module_node, &(entry->modules), + hlist) { + count = e_module->nr_entries; + iter = e_module->table; + while (count--) { + if (iter->key && + kernel_text_address(iter->code)) + arch_jump_label_transform(iter, type); + iter++; + } + } + } + jump_label_unlock(); +} + +static int addr_conflict(struct jump_entry *entry, void *start, void *end) +{ + if (entry->code <= (unsigned long)end && + entry->code + JUMP_LABEL_NOP_SIZE > (unsigned long)start) + return 1; + + return 0; +} + +#ifdef CONFIG_MODULES + +static int module_conflict(void *start, void *end) +{ + struct hlist_head *head; + struct hlist_node *node, *node_next, *module_node, *module_node_next; + struct jump_label_entry *e; + struct jump_label_module_entry *e_module; + struct jump_entry *iter; + int i, count; + int conflict = 0; + + for (i = 0; i < JUMP_LABEL_TABLE_SIZE; i++) { + head = &jump_label_table[i]; + hlist_for_each_entry_safe(e, node, node_next, head, hlist) { + hlist_for_each_entry_safe(e_module, module_node, + module_node_next, + &(e->modules), hlist) { + count = e_module->nr_entries; + iter = e_module->table; + while (count--) { + if (addr_conflict(iter, start, end)) { + conflict = 1; + goto out; + } + iter++; + } + } + } + } +out: + return conflict; +} + +#endif + +/*** + * jump_label_text_reserved - check if addr range is reserved + * @start: start text addr + * @end: end text addr + * + * checks if the text addr located between @start and @end + * overlaps with any of the jump label patch addresses. Code + * that wants to modify kernel text should first verify that + * it does not overlap with any of the jump label addresses. + * Caller must hold jump_label_mutex. + * + * returns 1 if there is an overlap, 0 otherwise + */ +int jump_label_text_reserved(void *start, void *end) +{ + struct jump_entry *iter; + struct jump_entry *iter_start = __start___jump_table; + struct jump_entry *iter_stop = __start___jump_table; + int conflict = 0; + + iter = iter_start; + while (iter < iter_stop) { + if (addr_conflict(iter, start, end)) { + conflict = 1; + goto out; + } + iter++; + } + + /* now check modules */ +#ifdef CONFIG_MODULES + conflict = module_conflict(start, end); +#endif +out: + return conflict; +} + +/* + * Not all archs need this. + */ +void __weak arch_jump_label_text_poke_early(jump_label_t addr) +{ +} + +static __init int init_jump_label(void) +{ + int ret; + struct jump_entry *iter_start = __start___jump_table; + struct jump_entry *iter_stop = __stop___jump_table; + struct jump_entry *iter; + + jump_label_lock(); + ret = build_jump_label_hashtable(__start___jump_table, + __stop___jump_table); + iter = iter_start; + while (iter < iter_stop) { + arch_jump_label_text_poke_early(iter->code); + iter++; + } + jump_label_unlock(); + return ret; +} +early_initcall(init_jump_label); + +#ifdef CONFIG_MODULES + +static struct jump_label_module_entry * +add_jump_label_module_entry(struct jump_label_entry *entry, + struct jump_entry *iter_begin, + int count, struct module *mod) +{ + struct jump_label_module_entry *e; + + e = kmalloc(sizeof(struct jump_label_module_entry), GFP_KERNEL); + if (!e) + return ERR_PTR(-ENOMEM); + e->mod = mod; + e->nr_entries = count; + e->table = iter_begin; + hlist_add_head(&e->hlist, &entry->modules); + return e; +} + +static int add_jump_label_module(struct module *mod) +{ + struct jump_entry *iter, *iter_begin; + struct jump_label_entry *entry; + struct jump_label_module_entry *module_entry; + int count; + + /* if the module doesn't have jump label entries, just return */ + if (!mod->num_jump_entries) + return 0; + + sort_jump_label_entries(mod->jump_entries, + mod->jump_entries + mod->num_jump_entries); + iter = mod->jump_entries; + while (iter < mod->jump_entries + mod->num_jump_entries) { + entry = get_jump_label_entry(iter->key); + iter_begin = iter; + count = 0; + while ((iter < mod->jump_entries + mod->num_jump_entries) && + (iter->key == iter_begin->key)) { + iter++; + count++; + } + if (!entry) { + entry = add_jump_label_entry(iter_begin->key, 0, NULL); + if (IS_ERR(entry)) + return PTR_ERR(entry); + } + module_entry = add_jump_label_module_entry(entry, iter_begin, + count, mod); + if (IS_ERR(module_entry)) + return PTR_ERR(module_entry); + } + return 0; +} + +static void remove_jump_label_module(struct module *mod) +{ + struct hlist_head *head; + struct hlist_node *node, *node_next, *module_node, *module_node_next; + struct jump_label_entry *e; + struct jump_label_module_entry *e_module; + int i; + + /* if the module doesn't have jump label entries, just return */ + if (!mod->num_jump_entries) + return; + + for (i = 0; i < JUMP_LABEL_TABLE_SIZE; i++) { + head = &jump_label_table[i]; + hlist_for_each_entry_safe(e, node, node_next, head, hlist) { + hlist_for_each_entry_safe(e_module, module_node, + module_node_next, + &(e->modules), hlist) { + if (e_module->mod == mod) { + hlist_del(&e_module->hlist); + kfree(e_module); + } + } + if (hlist_empty(&e->modules) && (e->nr_entries == 0)) { + hlist_del(&e->hlist); + kfree(e); + } + } + } +} + +static void remove_jump_label_module_init(struct module *mod) +{ + struct hlist_head *head; + struct hlist_node *node, *node_next, *module_node, *module_node_next; + struct jump_label_entry *e; + struct jump_label_module_entry *e_module; + struct jump_entry *iter; + int i, count; + + /* if the module doesn't have jump label entries, just return */ + if (!mod->num_jump_entries) + return; + + for (i = 0; i < JUMP_LABEL_TABLE_SIZE; i++) { + head = &jump_label_table[i]; + hlist_for_each_entry_safe(e, node, node_next, head, hlist) { + hlist_for_each_entry_safe(e_module, module_node, + module_node_next, + &(e->modules), hlist) { + if (e_module->mod != mod) + continue; + count = e_module->nr_entries; + iter = e_module->table; + while (count--) { + if (within_module_init(iter->code, mod)) + iter->key = 0; + iter++; + } + } + } + } +} + +static int +jump_label_module_notify(struct notifier_block *self, unsigned long val, + void *data) +{ + struct module *mod = data; + int ret = 0; + + switch (val) { + case MODULE_STATE_COMING: + jump_label_lock(); + ret = add_jump_label_module(mod); + if (ret) + remove_jump_label_module(mod); + jump_label_unlock(); + break; + case MODULE_STATE_GOING: + jump_label_lock(); + remove_jump_label_module(mod); + jump_label_unlock(); + break; + case MODULE_STATE_LIVE: + jump_label_lock(); + remove_jump_label_module_init(mod); + jump_label_unlock(); + break; + } + return ret; +} + +/*** + * apply_jump_label_nops - patch module jump labels with arch_get_jump_label_nop() + * @mod: module to patch + * + * Allow for run-time selection of the optimal nops. Before the module + * loads patch these with arch_get_jump_label_nop(), which is specified by + * the arch specific jump label code. + */ +void jump_label_apply_nops(struct module *mod) +{ + struct jump_entry *iter; + + /* if the module doesn't have jump label entries, just return */ + if (!mod->num_jump_entries) + return; + + iter = mod->jump_entries; + while (iter < mod->jump_entries + mod->num_jump_entries) { + arch_jump_label_text_poke_early(iter->code); + iter++; + } +} + +struct notifier_block jump_label_module_nb = { + .notifier_call = jump_label_module_notify, + .priority = 0, +}; + +static __init int init_jump_label_module(void) +{ + return register_module_notifier(&jump_label_module_nb); +} +early_initcall(init_jump_label_module); + +#endif /* CONFIG_MODULES */ + +#endif diff --git a/kernel/kexec.c b/kernel/kexec.c index 474a84715ea..ec19b92c7eb 100644 --- a/kernel/kexec.c +++ b/kernel/kexec.c @@ -151,8 +151,10 @@ static int do_kimage_alloc(struct kimage **rimage, unsigned long entry, image->nr_segments = nr_segments; segment_bytes = nr_segments * sizeof(*segments); result = copy_from_user(image->segment, segments, segment_bytes); - if (result) + if (result) { + result = -EFAULT; goto out; + } /* * Verify we have good destination addresses. The caller is @@ -161,7 +163,7 @@ static int do_kimage_alloc(struct kimage **rimage, unsigned long entry, * just verifies it is an address we can use. * * Since the kernel does everything in page size chunks ensure - * the destination addreses are page aligned. Too many + * the destination addresses are page aligned. Too many * special cases crop of when we don't do this. The most * insidious is getting overlapping destination addresses * simply because addresses are changed to page size @@ -814,7 +816,7 @@ static int kimage_load_normal_segment(struct kimage *image, ptr = kmap(page); /* Start with a clear page */ - memset(ptr, 0, PAGE_SIZE); + clear_page(ptr); ptr += maddr & ~PAGE_MASK; mchunk = PAGE_SIZE - (maddr & ~PAGE_MASK); if (mchunk > mbytes) @@ -827,7 +829,7 @@ static int kimage_load_normal_segment(struct kimage *image, result = copy_from_user(ptr, buf, uchunk); kunmap(page); if (result) { - result = (result < 0) ? result : -EIO; + result = -EFAULT; goto out; } ubytes -= uchunk; @@ -882,7 +884,7 @@ static int kimage_load_crash_segment(struct kimage *image, kexec_flush_icache_page(page); kunmap(page); if (result) { - result = (result < 0) ? result : -EIO; + result = -EFAULT; goto out; } ubytes -= uchunk; @@ -1089,9 +1091,10 @@ void crash_kexec(struct pt_regs *regs) size_t crash_get_memory_size(void) { - size_t size; + size_t size = 0; mutex_lock(&kexec_mutex); - size = crashk_res.end - crashk_res.start + 1; + if (crashk_res.end != crashk_res.start) + size = crashk_res.end - crashk_res.start + 1; mutex_unlock(&kexec_mutex); return size; } @@ -1134,7 +1137,7 @@ int crash_shrink_memory(unsigned long new_size) free_reserved_phys_range(end, crashk_res.end); - if (start == end) + if ((start == end) && (crashk_res.parent != NULL)) release_resource(&crashk_res); crashk_res.end = end - 1; diff --git a/kernel/kfifo.c b/kernel/kfifo.c index 35edbe22e9a..01a0700e873 100644 --- a/kernel/kfifo.c +++ b/kernel/kfifo.c @@ -1,8 +1,7 @@ /* - * A generic kernel FIFO implementation. + * A generic kernel FIFO implementation * - * Copyright (C) 2009 Stefani Seibold <stefani@seibold.net> - * Copyright (C) 2004 Stelian Pop <stelian@popies.net> + * Copyright (C) 2009/2010 Stefani Seibold <stefani@seibold.net> * * 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 @@ -11,7 +10,7 @@ * * 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 + * 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 @@ -24,422 +23,586 @@ #include <linux/module.h> #include <linux/slab.h> #include <linux/err.h> -#include <linux/kfifo.h> #include <linux/log2.h> #include <linux/uaccess.h> +#include <linux/kfifo.h> -static void _kfifo_init(struct kfifo *fifo, void *buffer, - unsigned int size) -{ - fifo->buffer = buffer; - fifo->size = size; - - kfifo_reset(fifo); -} - -/** - * kfifo_init - initialize a FIFO using a preallocated buffer - * @fifo: the fifo to assign the buffer - * @buffer: the preallocated buffer to be used. - * @size: the size of the internal buffer, this has to be a power of 2. - * +/* + * internal helper to calculate the unused elements in a fifo */ -void kfifo_init(struct kfifo *fifo, void *buffer, unsigned int size) +static inline unsigned int kfifo_unused(struct __kfifo *fifo) { - /* size must be a power of 2 */ - BUG_ON(!is_power_of_2(size)); - - _kfifo_init(fifo, buffer, size); + return (fifo->mask + 1) - (fifo->in - fifo->out); } -EXPORT_SYMBOL(kfifo_init); -/** - * kfifo_alloc - allocates a new FIFO internal buffer - * @fifo: the fifo to assign then new buffer - * @size: the size of the buffer to be allocated, this have to be a power of 2. - * @gfp_mask: get_free_pages mask, passed to kmalloc() - * - * This function dynamically allocates a new fifo internal buffer - * - * The size will be rounded-up to a power of 2. - * The buffer will be release with kfifo_free(). - * Return 0 if no error, otherwise the an error code - */ -int kfifo_alloc(struct kfifo *fifo, unsigned int size, gfp_t gfp_mask) +int __kfifo_alloc(struct __kfifo *fifo, unsigned int size, + size_t esize, gfp_t gfp_mask) { - unsigned char *buffer; - /* - * round up to the next power of 2, since our 'let the indices + * round down to the next power of 2, since our 'let the indices * wrap' technique works only in this case. */ - if (!is_power_of_2(size)) { - BUG_ON(size > 0x80000000); - size = roundup_pow_of_two(size); + if (!is_power_of_2(size)) + size = rounddown_pow_of_two(size); + + fifo->in = 0; + fifo->out = 0; + fifo->esize = esize; + + if (size < 2) { + fifo->data = NULL; + fifo->mask = 0; + return -EINVAL; } - buffer = kmalloc(size, gfp_mask); - if (!buffer) { - _kfifo_init(fifo, NULL, 0); + fifo->data = kmalloc(size * esize, gfp_mask); + + if (!fifo->data) { + fifo->mask = 0; return -ENOMEM; } - - _kfifo_init(fifo, buffer, size); + fifo->mask = size - 1; return 0; } -EXPORT_SYMBOL(kfifo_alloc); +EXPORT_SYMBOL(__kfifo_alloc); -/** - * kfifo_free - frees the FIFO internal buffer - * @fifo: the fifo to be freed. - */ -void kfifo_free(struct kfifo *fifo) +void __kfifo_free(struct __kfifo *fifo) { - kfree(fifo->buffer); - _kfifo_init(fifo, NULL, 0); + kfree(fifo->data); + fifo->in = 0; + fifo->out = 0; + fifo->esize = 0; + fifo->data = NULL; + fifo->mask = 0; } -EXPORT_SYMBOL(kfifo_free); +EXPORT_SYMBOL(__kfifo_free); -/** - * kfifo_skip - skip output data - * @fifo: the fifo to be used. - * @len: number of bytes to skip - */ -void kfifo_skip(struct kfifo *fifo, unsigned int len) +int __kfifo_init(struct __kfifo *fifo, void *buffer, + unsigned int size, size_t esize) { - if (len < kfifo_len(fifo)) { - __kfifo_add_out(fifo, len); - return; + size /= esize; + + if (!is_power_of_2(size)) + size = rounddown_pow_of_two(size); + + fifo->in = 0; + fifo->out = 0; + fifo->esize = esize; + fifo->data = buffer; + + if (size < 2) { + fifo->mask = 0; + return -EINVAL; } - kfifo_reset_out(fifo); + fifo->mask = size - 1; + + return 0; } -EXPORT_SYMBOL(kfifo_skip); +EXPORT_SYMBOL(__kfifo_init); -static inline void __kfifo_in_data(struct kfifo *fifo, - const void *from, unsigned int len, unsigned int off) +static void kfifo_copy_in(struct __kfifo *fifo, const void *src, + unsigned int len, unsigned int off) { + unsigned int size = fifo->mask + 1; + unsigned int esize = fifo->esize; unsigned int l; + off &= fifo->mask; + if (esize != 1) { + off *= esize; + size *= esize; + len *= esize; + } + l = min(len, size - off); + + memcpy(fifo->data + off, src, l); + memcpy(fifo->data, src + l, len - l); /* - * Ensure that we sample the fifo->out index -before- we - * start putting bytes into the kfifo. + * make sure that the data in the fifo is up to date before + * incrementing the fifo->in index counter */ + smp_wmb(); +} - smp_mb(); - - off = __kfifo_off(fifo, fifo->in + off); +unsigned int __kfifo_in(struct __kfifo *fifo, + const void *buf, unsigned int len) +{ + unsigned int l; - /* first put the data starting from fifo->in to buffer end */ - l = min(len, fifo->size - off); - memcpy(fifo->buffer + off, from, l); + l = kfifo_unused(fifo); + if (len > l) + len = l; - /* then put the rest (if any) at the beginning of the buffer */ - memcpy(fifo->buffer, from + l, len - l); + kfifo_copy_in(fifo, buf, len, fifo->in); + fifo->in += len; + return len; } +EXPORT_SYMBOL(__kfifo_in); -static inline void __kfifo_out_data(struct kfifo *fifo, - void *to, unsigned int len, unsigned int off) +static void kfifo_copy_out(struct __kfifo *fifo, void *dst, + unsigned int len, unsigned int off) { + unsigned int size = fifo->mask + 1; + unsigned int esize = fifo->esize; unsigned int l; + off &= fifo->mask; + if (esize != 1) { + off *= esize; + size *= esize; + len *= esize; + } + l = min(len, size - off); + + memcpy(dst, fifo->data + off, l); + memcpy(dst + l, fifo->data, len - l); /* - * Ensure that we sample the fifo->in index -before- we - * start removing bytes from the kfifo. + * make sure that the data is copied before + * incrementing the fifo->out index counter */ + smp_wmb(); +} - smp_rmb(); +unsigned int __kfifo_out_peek(struct __kfifo *fifo, + void *buf, unsigned int len) +{ + unsigned int l; - off = __kfifo_off(fifo, fifo->out + off); + l = fifo->in - fifo->out; + if (len > l) + len = l; - /* first get the data from fifo->out until the end of the buffer */ - l = min(len, fifo->size - off); - memcpy(to, fifo->buffer + off, l); + kfifo_copy_out(fifo, buf, len, fifo->out); + return len; +} +EXPORT_SYMBOL(__kfifo_out_peek); - /* then get the rest (if any) from the beginning of the buffer */ - memcpy(to + l, fifo->buffer, len - l); +unsigned int __kfifo_out(struct __kfifo *fifo, + void *buf, unsigned int len) +{ + len = __kfifo_out_peek(fifo, buf, len); + fifo->out += len; + return len; } +EXPORT_SYMBOL(__kfifo_out); -static inline int __kfifo_from_user_data(struct kfifo *fifo, - const void __user *from, unsigned int len, unsigned int off, - unsigned *lenout) +static unsigned long kfifo_copy_from_user(struct __kfifo *fifo, + const void __user *from, unsigned int len, unsigned int off, + unsigned int *copied) { + unsigned int size = fifo->mask + 1; + unsigned int esize = fifo->esize; unsigned int l; - int ret; + unsigned long ret; + off &= fifo->mask; + if (esize != 1) { + off *= esize; + size *= esize; + len *= esize; + } + l = min(len, size - off); + + ret = copy_from_user(fifo->data + off, from, l); + if (unlikely(ret)) + ret = DIV_ROUND_UP(ret + len - l, esize); + else { + ret = copy_from_user(fifo->data, from + l, len - l); + if (unlikely(ret)) + ret = DIV_ROUND_UP(ret, esize); + } /* - * Ensure that we sample the fifo->out index -before- we - * start putting bytes into the kfifo. + * make sure that the data in the fifo is up to date before + * incrementing the fifo->in index counter */ + smp_wmb(); + *copied = len - ret; + /* return the number of elements which are not copied */ + return ret; +} - smp_mb(); +int __kfifo_from_user(struct __kfifo *fifo, const void __user *from, + unsigned long len, unsigned int *copied) +{ + unsigned int l; + unsigned long ret; + unsigned int esize = fifo->esize; + int err; - off = __kfifo_off(fifo, fifo->in + off); + if (esize != 1) + len /= esize; - /* first put the data starting from fifo->in to buffer end */ - l = min(len, fifo->size - off); - ret = copy_from_user(fifo->buffer + off, from, l); - if (unlikely(ret)) { - *lenout = ret; - return -EFAULT; - } - *lenout = l; + l = kfifo_unused(fifo); + if (len > l) + len = l; - /* then put the rest (if any) at the beginning of the buffer */ - ret = copy_from_user(fifo->buffer, from + l, len - l); - *lenout += ret ? ret : len - l; - return ret ? -EFAULT : 0; + ret = kfifo_copy_from_user(fifo, from, len, fifo->in, copied); + if (unlikely(ret)) { + len -= ret; + err = -EFAULT; + } else + err = 0; + fifo->in += len; + return err; } +EXPORT_SYMBOL(__kfifo_from_user); -static inline int __kfifo_to_user_data(struct kfifo *fifo, - void __user *to, unsigned int len, unsigned int off, unsigned *lenout) +static unsigned long kfifo_copy_to_user(struct __kfifo *fifo, void __user *to, + unsigned int len, unsigned int off, unsigned int *copied) { unsigned int l; - int ret; - + unsigned long ret; + unsigned int size = fifo->mask + 1; + unsigned int esize = fifo->esize; + + off &= fifo->mask; + if (esize != 1) { + off *= esize; + size *= esize; + len *= esize; + } + l = min(len, size - off); + + ret = copy_to_user(to, fifo->data + off, l); + if (unlikely(ret)) + ret = DIV_ROUND_UP(ret + len - l, esize); + else { + ret = copy_to_user(to + l, fifo->data, len - l); + if (unlikely(ret)) + ret = DIV_ROUND_UP(ret, esize); + } /* - * Ensure that we sample the fifo->in index -before- we - * start removing bytes from the kfifo. + * make sure that the data is copied before + * incrementing the fifo->out index counter */ + smp_wmb(); + *copied = len - ret; + /* return the number of elements which are not copied */ + return ret; +} - smp_rmb(); +int __kfifo_to_user(struct __kfifo *fifo, void __user *to, + unsigned long len, unsigned int *copied) +{ + unsigned int l; + unsigned long ret; + unsigned int esize = fifo->esize; + int err; - off = __kfifo_off(fifo, fifo->out + off); + if (esize != 1) + len /= esize; - /* first get the data from fifo->out until the end of the buffer */ - l = min(len, fifo->size - off); - ret = copy_to_user(to, fifo->buffer + off, l); - *lenout = l; + l = fifo->in - fifo->out; + if (len > l) + len = l; + ret = kfifo_copy_to_user(fifo, to, len, fifo->out, copied); if (unlikely(ret)) { - *lenout -= ret; - return -EFAULT; - } + len -= ret; + err = -EFAULT; + } else + err = 0; + fifo->out += len; + return err; +} +EXPORT_SYMBOL(__kfifo_to_user); - /* then get the rest (if any) from the beginning of the buffer */ - len -= l; - ret = copy_to_user(to + l, fifo->buffer, len); - if (unlikely(ret)) { - *lenout += len - ret; - return -EFAULT; +static int setup_sgl_buf(struct scatterlist *sgl, void *buf, + int nents, unsigned int len) +{ + int n; + unsigned int l; + unsigned int off; + struct page *page; + + if (!nents) + return 0; + + if (!len) + return 0; + + n = 0; + page = virt_to_page(buf); + off = offset_in_page(buf); + l = 0; + + while (len >= l + PAGE_SIZE - off) { + struct page *npage; + + l += PAGE_SIZE; + buf += PAGE_SIZE; + npage = virt_to_page(buf); + if (page_to_phys(page) != page_to_phys(npage) - l) { + sg_set_page(sgl, page, l - off, off); + sgl = sg_next(sgl); + if (++n == nents || sgl == NULL) + return n; + page = npage; + len -= l - off; + l = off = 0; + } } - *lenout += len; - return 0; + sg_set_page(sgl, page, len, off); + return n + 1; } -unsigned int __kfifo_in_n(struct kfifo *fifo, - const void *from, unsigned int len, unsigned int recsize) +static unsigned int setup_sgl(struct __kfifo *fifo, struct scatterlist *sgl, + int nents, unsigned int len, unsigned int off) { - if (kfifo_avail(fifo) < len + recsize) - return len + 1; + unsigned int size = fifo->mask + 1; + unsigned int esize = fifo->esize; + unsigned int l; + unsigned int n; - __kfifo_in_data(fifo, from, len, recsize); - return 0; + off &= fifo->mask; + if (esize != 1) { + off *= esize; + size *= esize; + len *= esize; + } + l = min(len, size - off); + + n = setup_sgl_buf(sgl, fifo->data + off, nents, l); + n += setup_sgl_buf(sgl + n, fifo->data, nents - n, len - l); + + return n; } -EXPORT_SYMBOL(__kfifo_in_n); -/** - * kfifo_in - puts some data into the FIFO - * @fifo: the fifo to be used. - * @from: the data to be added. - * @len: the length of the data to be added. - * - * This function copies at most @len bytes from the @from buffer into - * the FIFO depending on the free space, and returns the number of - * bytes copied. - * - * Note that with only one concurrent reader and one concurrent - * writer, you don't need extra locking to use these functions. - */ -unsigned int kfifo_in(struct kfifo *fifo, const void *from, - unsigned int len) +unsigned int __kfifo_dma_in_prepare(struct __kfifo *fifo, + struct scatterlist *sgl, int nents, unsigned int len) { - len = min(kfifo_avail(fifo), len); + unsigned int l; - __kfifo_in_data(fifo, from, len, 0); - __kfifo_add_in(fifo, len); - return len; + l = kfifo_unused(fifo); + if (len > l) + len = l; + + return setup_sgl(fifo, sgl, nents, len, fifo->in); } -EXPORT_SYMBOL(kfifo_in); +EXPORT_SYMBOL(__kfifo_dma_in_prepare); -unsigned int __kfifo_in_generic(struct kfifo *fifo, - const void *from, unsigned int len, unsigned int recsize) +unsigned int __kfifo_dma_out_prepare(struct __kfifo *fifo, + struct scatterlist *sgl, int nents, unsigned int len) { - return __kfifo_in_rec(fifo, from, len, recsize); + unsigned int l; + + l = fifo->in - fifo->out; + if (len > l) + len = l; + + return setup_sgl(fifo, sgl, nents, len, fifo->out); } -EXPORT_SYMBOL(__kfifo_in_generic); +EXPORT_SYMBOL(__kfifo_dma_out_prepare); -unsigned int __kfifo_out_n(struct kfifo *fifo, - void *to, unsigned int len, unsigned int recsize) +unsigned int __kfifo_max_r(unsigned int len, size_t recsize) { - if (kfifo_len(fifo) < len + recsize) - return len; + unsigned int max = (1 << (recsize << 3)) - 1; - __kfifo_out_data(fifo, to, len, recsize); - __kfifo_add_out(fifo, len + recsize); - return 0; + if (len > max) + return max; + return len; } -EXPORT_SYMBOL(__kfifo_out_n); -/** - * kfifo_out - gets some data from the FIFO - * @fifo: the fifo to be used. - * @to: where the data must be copied. - * @len: the size of the destination buffer. - * - * This function copies at most @len bytes from the FIFO into the - * @to buffer and returns the number of copied bytes. - * - * Note that with only one concurrent reader and one concurrent - * writer, you don't need extra locking to use these functions. +#define __KFIFO_PEEK(data, out, mask) \ + ((data)[(out) & (mask)]) +/* + * __kfifo_peek_n internal helper function for determinate the length of + * the next record in the fifo */ -unsigned int kfifo_out(struct kfifo *fifo, void *to, unsigned int len) +static unsigned int __kfifo_peek_n(struct __kfifo *fifo, size_t recsize) { - len = min(kfifo_len(fifo), len); + unsigned int l; + unsigned int mask = fifo->mask; + unsigned char *data = fifo->data; - __kfifo_out_data(fifo, to, len, 0); - __kfifo_add_out(fifo, len); + l = __KFIFO_PEEK(data, fifo->out, mask); - return len; + if (--recsize) + l |= __KFIFO_PEEK(data, fifo->out + 1, mask) << 8; + + return l; } -EXPORT_SYMBOL(kfifo_out); -/** - * kfifo_out_peek - copy some data from the FIFO, but do not remove it - * @fifo: the fifo to be used. - * @to: where the data must be copied. - * @len: the size of the destination buffer. - * @offset: offset into the fifo - * - * This function copies at most @len bytes at @offset from the FIFO - * into the @to buffer and returns the number of copied bytes. - * The data is not removed from the FIFO. +#define __KFIFO_POKE(data, in, mask, val) \ + ( \ + (data)[(in) & (mask)] = (unsigned char)(val) \ + ) + +/* + * __kfifo_poke_n internal helper function for storeing the length of + * the record into the fifo */ -unsigned int kfifo_out_peek(struct kfifo *fifo, void *to, unsigned int len, - unsigned offset) +static void __kfifo_poke_n(struct __kfifo *fifo, unsigned int n, size_t recsize) { - len = min(kfifo_len(fifo), len + offset); + unsigned int mask = fifo->mask; + unsigned char *data = fifo->data; - __kfifo_out_data(fifo, to, len, offset); - return len; + __KFIFO_POKE(data, fifo->in, mask, n); + + if (recsize > 1) + __KFIFO_POKE(data, fifo->in + 1, mask, n >> 8); } -EXPORT_SYMBOL(kfifo_out_peek); -unsigned int __kfifo_out_generic(struct kfifo *fifo, - void *to, unsigned int len, unsigned int recsize, - unsigned int *total) +unsigned int __kfifo_len_r(struct __kfifo *fifo, size_t recsize) { - return __kfifo_out_rec(fifo, to, len, recsize, total); + return __kfifo_peek_n(fifo, recsize); } -EXPORT_SYMBOL(__kfifo_out_generic); +EXPORT_SYMBOL(__kfifo_len_r); -unsigned int __kfifo_from_user_n(struct kfifo *fifo, - const void __user *from, unsigned int len, unsigned int recsize) +unsigned int __kfifo_in_r(struct __kfifo *fifo, const void *buf, + unsigned int len, size_t recsize) { - unsigned total; + if (len + recsize > kfifo_unused(fifo)) + return 0; - if (kfifo_avail(fifo) < len + recsize) - return len + 1; + __kfifo_poke_n(fifo, len, recsize); - __kfifo_from_user_data(fifo, from, len, recsize, &total); - return total; + kfifo_copy_in(fifo, buf, len, fifo->in + recsize); + fifo->in += len + recsize; + return len; } -EXPORT_SYMBOL(__kfifo_from_user_n); +EXPORT_SYMBOL(__kfifo_in_r); -/** - * kfifo_from_user - puts some data from user space into the FIFO - * @fifo: the fifo to be used. - * @from: pointer to the data to be added. - * @len: the length of the data to be added. - * @total: the actual returned data length. - * - * This function copies at most @len bytes from the @from into the - * FIFO depending and returns -EFAULT/0. - * - * Note that with only one concurrent reader and one concurrent - * writer, you don't need extra locking to use these functions. - */ -int kfifo_from_user(struct kfifo *fifo, - const void __user *from, unsigned int len, unsigned *total) -{ - int ret; - len = min(kfifo_avail(fifo), len); - ret = __kfifo_from_user_data(fifo, from, len, 0, total); - if (ret) - return ret; - __kfifo_add_in(fifo, len); - return 0; +static unsigned int kfifo_out_copy_r(struct __kfifo *fifo, + void *buf, unsigned int len, size_t recsize, unsigned int *n) +{ + *n = __kfifo_peek_n(fifo, recsize); + + if (len > *n) + len = *n; + + kfifo_copy_out(fifo, buf, len, fifo->out + recsize); + return len; } -EXPORT_SYMBOL(kfifo_from_user); -unsigned int __kfifo_from_user_generic(struct kfifo *fifo, - const void __user *from, unsigned int len, unsigned int recsize) +unsigned int __kfifo_out_peek_r(struct __kfifo *fifo, void *buf, + unsigned int len, size_t recsize) { - return __kfifo_from_user_rec(fifo, from, len, recsize); + unsigned int n; + + if (fifo->in == fifo->out) + return 0; + + return kfifo_out_copy_r(fifo, buf, len, recsize, &n); } -EXPORT_SYMBOL(__kfifo_from_user_generic); +EXPORT_SYMBOL(__kfifo_out_peek_r); -unsigned int __kfifo_to_user_n(struct kfifo *fifo, - void __user *to, unsigned int len, unsigned int reclen, - unsigned int recsize) +unsigned int __kfifo_out_r(struct __kfifo *fifo, void *buf, + unsigned int len, size_t recsize) { - unsigned int ret, total; + unsigned int n; - if (kfifo_len(fifo) < reclen + recsize) - return len; + if (fifo->in == fifo->out) + return 0; - ret = __kfifo_to_user_data(fifo, to, reclen, recsize, &total); + len = kfifo_out_copy_r(fifo, buf, len, recsize, &n); + fifo->out += n + recsize; + return len; +} +EXPORT_SYMBOL(__kfifo_out_r); - if (likely(ret == 0)) - __kfifo_add_out(fifo, reclen + recsize); +void __kfifo_skip_r(struct __kfifo *fifo, size_t recsize) +{ + unsigned int n; - return total; + n = __kfifo_peek_n(fifo, recsize); + fifo->out += n + recsize; } -EXPORT_SYMBOL(__kfifo_to_user_n); +EXPORT_SYMBOL(__kfifo_skip_r); -/** - * kfifo_to_user - gets data from the FIFO and write it to user space - * @fifo: the fifo to be used. - * @to: where the data must be copied. - * @len: the size of the destination buffer. - * @lenout: pointer to output variable with copied data - * - * This function copies at most @len bytes from the FIFO into the - * @to buffer and 0 or -EFAULT. - * - * Note that with only one concurrent reader and one concurrent - * writer, you don't need extra locking to use these functions. - */ -int kfifo_to_user(struct kfifo *fifo, - void __user *to, unsigned int len, unsigned *lenout) +int __kfifo_from_user_r(struct __kfifo *fifo, const void __user *from, + unsigned long len, unsigned int *copied, size_t recsize) { - int ret; - len = min(kfifo_len(fifo), len); - ret = __kfifo_to_user_data(fifo, to, len, 0, lenout); - __kfifo_add_out(fifo, *lenout); - return ret; + unsigned long ret; + + len = __kfifo_max_r(len, recsize); + + if (len + recsize > kfifo_unused(fifo)) { + *copied = 0; + return 0; + } + + __kfifo_poke_n(fifo, len, recsize); + + ret = kfifo_copy_from_user(fifo, from, len, fifo->in + recsize, copied); + if (unlikely(ret)) { + *copied = 0; + return -EFAULT; + } + fifo->in += len + recsize; + return 0; } -EXPORT_SYMBOL(kfifo_to_user); +EXPORT_SYMBOL(__kfifo_from_user_r); -unsigned int __kfifo_to_user_generic(struct kfifo *fifo, - void __user *to, unsigned int len, unsigned int recsize, - unsigned int *total) +int __kfifo_to_user_r(struct __kfifo *fifo, void __user *to, + unsigned long len, unsigned int *copied, size_t recsize) { - return __kfifo_to_user_rec(fifo, to, len, recsize, total); + unsigned long ret; + unsigned int n; + + if (fifo->in == fifo->out) { + *copied = 0; + return 0; + } + + n = __kfifo_peek_n(fifo, recsize); + if (len > n) + len = n; + + ret = kfifo_copy_to_user(fifo, to, len, fifo->out + recsize, copied); + if (unlikely(ret)) { + *copied = 0; + return -EFAULT; + } + fifo->out += n + recsize; + return 0; } -EXPORT_SYMBOL(__kfifo_to_user_generic); +EXPORT_SYMBOL(__kfifo_to_user_r); -unsigned int __kfifo_peek_generic(struct kfifo *fifo, unsigned int recsize) +unsigned int __kfifo_dma_in_prepare_r(struct __kfifo *fifo, + struct scatterlist *sgl, int nents, unsigned int len, size_t recsize) { - if (recsize == 0) - return kfifo_avail(fifo); + if (!nents) + BUG(); - return __kfifo_peek_n(fifo, recsize); + len = __kfifo_max_r(len, recsize); + + if (len + recsize > kfifo_unused(fifo)) + return 0; + + return setup_sgl(fifo, sgl, nents, len, fifo->in + recsize); } -EXPORT_SYMBOL(__kfifo_peek_generic); +EXPORT_SYMBOL(__kfifo_dma_in_prepare_r); -void __kfifo_skip_generic(struct kfifo *fifo, unsigned int recsize) +void __kfifo_dma_in_finish_r(struct __kfifo *fifo, + unsigned int len, size_t recsize) { - __kfifo_skip_rec(fifo, recsize); + len = __kfifo_max_r(len, recsize); + __kfifo_poke_n(fifo, len, recsize); + fifo->in += len + recsize; } -EXPORT_SYMBOL(__kfifo_skip_generic); +EXPORT_SYMBOL(__kfifo_dma_in_finish_r); +unsigned int __kfifo_dma_out_prepare_r(struct __kfifo *fifo, + struct scatterlist *sgl, int nents, unsigned int len, size_t recsize) +{ + if (!nents) + BUG(); + + len = __kfifo_max_r(len, recsize); + + if (len + recsize > fifo->in - fifo->out) + return 0; + + return setup_sgl(fifo, sgl, nents, len, fifo->out + recsize); +} +EXPORT_SYMBOL(__kfifo_dma_out_prepare_r); + +void __kfifo_dma_out_finish_r(struct __kfifo *fifo, size_t recsize) +{ + unsigned int len; + + len = __kfifo_peek_n(fifo, recsize); + fifo->out += len + recsize; +} +EXPORT_SYMBOL(__kfifo_dma_out_finish_r); diff --git a/kernel/kmod.c b/kernel/kmod.c index bf0e231d970..9cd0591c96a 100644 --- a/kernel/kmod.c +++ b/kernel/kmod.c @@ -116,27 +116,16 @@ int __request_module(bool wait, const char *fmt, ...) trace_module_request(module_name, wait, _RET_IP_); - ret = call_usermodehelper(modprobe_path, argv, envp, - wait ? UMH_WAIT_PROC : UMH_WAIT_EXEC); + ret = call_usermodehelper_fns(modprobe_path, argv, envp, + wait ? UMH_WAIT_PROC : UMH_WAIT_EXEC, + NULL, NULL, NULL); + atomic_dec(&kmod_concurrent); return ret; } EXPORT_SYMBOL(__request_module); #endif /* CONFIG_MODULES */ -struct subprocess_info { - struct work_struct work; - struct completion *complete; - struct cred *cred; - char *path; - char **argv; - char **envp; - enum umh_wait wait; - int retval; - struct file *stdin; - void (*cleanup)(char **argv, char **envp); -}; - /* * This is the task which runs the usermode application */ @@ -145,36 +134,10 @@ static int ____call_usermodehelper(void *data) struct subprocess_info *sub_info = data; int retval; - BUG_ON(atomic_read(&sub_info->cred->usage) != 1); - - /* Unblock all signals */ spin_lock_irq(¤t->sighand->siglock); flush_signal_handlers(current, 1); - sigemptyset(¤t->blocked); - recalc_sigpending(); spin_unlock_irq(¤t->sighand->siglock); - /* Install the credentials */ - commit_creds(sub_info->cred); - sub_info->cred = NULL; - - /* Install input pipe when needed */ - if (sub_info->stdin) { - struct files_struct *f = current->files; - struct fdtable *fdt; - /* no races because files should be private here */ - sys_close(0); - fd_install(0, sub_info->stdin); - spin_lock(&f->file_lock); - fdt = files_fdtable(f); - FD_SET(0, fdt->open_fds); - FD_CLR(0, fdt->close_on_exec); - spin_unlock(&f->file_lock); - - /* and disallow core files too */ - current->signal->rlim[RLIMIT_CORE] = (struct rlimit){0, 0}; - } - /* We can run anywhere, unlike our parent keventd(). */ set_cpus_allowed_ptr(current, cpu_all_mask); @@ -184,9 +147,18 @@ static int ____call_usermodehelper(void *data) */ set_user_nice(current, 0); - retval = kernel_execve(sub_info->path, sub_info->argv, sub_info->envp); + if (sub_info->init) { + retval = sub_info->init(sub_info); + if (retval) + goto fail; + } + + retval = kernel_execve(sub_info->path, + (const char *const *)sub_info->argv, + (const char *const *)sub_info->envp); /* Exec failed? */ +fail: sub_info->retval = retval; do_exit(0); } @@ -194,9 +166,7 @@ static int ____call_usermodehelper(void *data) void call_usermodehelper_freeinfo(struct subprocess_info *info) { if (info->cleanup) - (*info->cleanup)(info->argv, info->envp); - if (info->cred) - put_cred(info->cred); + (*info->cleanup)(info); kfree(info); } EXPORT_SYMBOL(call_usermodehelper_freeinfo); @@ -207,16 +177,16 @@ static int wait_for_helper(void *data) struct subprocess_info *sub_info = data; pid_t pid; - /* Install a handler: if SIGCLD isn't handled sys_wait4 won't - * populate the status, but will return -ECHILD. */ - allow_signal(SIGCHLD); + /* If SIGCLD is ignored sys_wait4 won't populate the status. */ + spin_lock_irq(¤t->sighand->siglock); + current->sighand->action[SIGCHLD-1].sa.sa_handler = SIG_DFL; + spin_unlock_irq(¤t->sighand->siglock); pid = kernel_thread(____call_usermodehelper, sub_info, SIGCHLD); if (pid < 0) { sub_info->retval = pid; } else { - int ret; - + int ret = -ECHILD; /* * Normally it is bogus to call wait4() from in-kernel because * wait4() wants to write the exit code to a userspace address. @@ -237,10 +207,7 @@ static int wait_for_helper(void *data) sub_info->retval = ret; } - if (sub_info->wait == UMH_NO_WAIT) - call_usermodehelper_freeinfo(sub_info); - else - complete(sub_info->complete); + complete(sub_info->complete); return 0; } @@ -249,15 +216,13 @@ static void __call_usermodehelper(struct work_struct *work) { struct subprocess_info *sub_info = container_of(work, struct subprocess_info, work); - pid_t pid; enum umh_wait wait = sub_info->wait; - - BUG_ON(atomic_read(&sub_info->cred->usage) != 1); + pid_t pid; /* CLONE_VFORK: wait until the usermode helper has execve'd * successfully We need the data structures to stay around * until that is done. */ - if (wait == UMH_WAIT_PROC || wait == UMH_NO_WAIT) + if (wait == UMH_WAIT_PROC) pid = kernel_thread(wait_for_helper, sub_info, CLONE_FS | CLONE_FILES | SIGCHLD); else @@ -266,15 +231,16 @@ static void __call_usermodehelper(struct work_struct *work) switch (wait) { case UMH_NO_WAIT: + call_usermodehelper_freeinfo(sub_info); break; case UMH_WAIT_PROC: if (pid > 0) break; - sub_info->retval = pid; /* FALLTHROUGH */ - case UMH_WAIT_EXEC: + if (pid < 0) + sub_info->retval = pid; complete(sub_info->complete); } } @@ -376,80 +342,37 @@ struct subprocess_info *call_usermodehelper_setup(char *path, char **argv, sub_info->path = path; sub_info->argv = argv; sub_info->envp = envp; - sub_info->cred = prepare_usermodehelper_creds(); - if (!sub_info->cred) { - kfree(sub_info); - return NULL; - } - out: return sub_info; } EXPORT_SYMBOL(call_usermodehelper_setup); /** - * call_usermodehelper_setkeys - set the session keys for usermode helper - * @info: a subprocess_info returned by call_usermodehelper_setup - * @session_keyring: the session keyring for the process - */ -void call_usermodehelper_setkeys(struct subprocess_info *info, - struct key *session_keyring) -{ -#ifdef CONFIG_KEYS - struct thread_group_cred *tgcred = info->cred->tgcred; - key_put(tgcred->session_keyring); - tgcred->session_keyring = key_get(session_keyring); -#else - BUG(); -#endif -} -EXPORT_SYMBOL(call_usermodehelper_setkeys); - -/** - * call_usermodehelper_setcleanup - set a cleanup function + * call_usermodehelper_setfns - set a cleanup/init function * @info: a subprocess_info returned by call_usermodehelper_setup * @cleanup: a cleanup function + * @init: an init function + * @data: arbitrary context sensitive data * - * The cleanup function is just befor ethe subprocess_info is about to + * The init function is used to customize the helper process prior to + * exec. A non-zero return code causes the process to error out, exit, + * and return the failure to the calling process + * + * The cleanup function is just before ethe subprocess_info is about to * be freed. This can be used for freeing the argv and envp. The * Function must be runnable in either a process context or the * context in which call_usermodehelper_exec is called. */ -void call_usermodehelper_setcleanup(struct subprocess_info *info, - void (*cleanup)(char **argv, char **envp)) +void call_usermodehelper_setfns(struct subprocess_info *info, + int (*init)(struct subprocess_info *info), + void (*cleanup)(struct subprocess_info *info), + void *data) { info->cleanup = cleanup; + info->init = init; + info->data = data; } -EXPORT_SYMBOL(call_usermodehelper_setcleanup); - -/** - * call_usermodehelper_stdinpipe - set up a pipe to be used for stdin - * @sub_info: a subprocess_info returned by call_usermodehelper_setup - * @filp: set to the write-end of a pipe - * - * This constructs a pipe, and sets the read end to be the stdin of the - * subprocess, and returns the write-end in *@filp. - */ -int call_usermodehelper_stdinpipe(struct subprocess_info *sub_info, - struct file **filp) -{ - struct file *f; - - f = create_write_pipe(0); - if (IS_ERR(f)) - return PTR_ERR(f); - *filp = f; - - f = create_read_pipe(f, 0); - if (IS_ERR(f)) { - free_write_pipe(*filp); - return PTR_ERR(f); - } - sub_info->stdin = f; - - return 0; -} -EXPORT_SYMBOL(call_usermodehelper_stdinpipe); +EXPORT_SYMBOL(call_usermodehelper_setfns); /** * call_usermodehelper_exec - start a usermode application @@ -469,9 +392,6 @@ int call_usermodehelper_exec(struct subprocess_info *sub_info, DECLARE_COMPLETION_ONSTACK(done); 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') goto out; @@ -498,41 +418,6 @@ unlock: } EXPORT_SYMBOL(call_usermodehelper_exec); -/** - * call_usermodehelper_pipe - call a usermode helper process with a pipe stdin - * @path: path to usermode executable - * @argv: arg vector for process - * @envp: environment for process - * @filp: set to the write-end of a pipe - * - * This is a simple wrapper which executes a usermode-helper function - * with a pipe as stdin. It is implemented entirely in terms of - * lower-level call_usermodehelper_* functions. - */ -int call_usermodehelper_pipe(char *path, char **argv, char **envp, - struct file **filp) -{ - struct subprocess_info *sub_info; - int ret; - - sub_info = call_usermodehelper_setup(path, argv, envp, GFP_KERNEL); - if (sub_info == NULL) - return -ENOMEM; - - ret = call_usermodehelper_stdinpipe(sub_info, filp); - if (ret < 0) { - call_usermodehelper_freeinfo(sub_info); - return ret; - } - - ret = call_usermodehelper_exec(sub_info, UMH_WAIT_EXEC); - if (ret < 0) /* Failed to execute helper, close pipe */ - filp_close(*filp, NULL); - - return ret; -} -EXPORT_SYMBOL(call_usermodehelper_pipe); - void __init usermodehelper_init(void) { khelper_wq = create_singlethread_workqueue("khelper"); diff --git a/kernel/kprobes.c b/kernel/kprobes.c index 282035f3ae9..77981813a1e 100644 --- a/kernel/kprobes.c +++ b/kernel/kprobes.c @@ -47,6 +47,7 @@ #include <linux/memory.h> #include <linux/ftrace.h> #include <linux/cpu.h> +#include <linux/jump_label.h> #include <asm-generic/sections.h> #include <asm/cacheflush.h> @@ -73,7 +74,8 @@ static struct hlist_head kretprobe_inst_table[KPROBE_TABLE_SIZE]; /* NOTE: change this value only with kprobe_mutex held */ static bool kprobes_all_disarmed; -static DEFINE_MUTEX(kprobe_mutex); /* Protects kprobe_table */ +/* This protects kprobe_table and optimizing_list */ +static DEFINE_MUTEX(kprobe_mutex); static DEFINE_PER_CPU(struct kprobe *, kprobe_instance) = NULL; static struct { spinlock_t lock ____cacheline_aligned_in_smp; @@ -315,12 +317,12 @@ void __kprobes free_optinsn_slot(kprobe_opcode_t * slot, int dirty) /* We have preemption disabled.. so it is safe to use __ versions */ static inline void set_kprobe_instance(struct kprobe *kp) { - __get_cpu_var(kprobe_instance) = kp; + __this_cpu_write(kprobe_instance, kp); } static inline void reset_kprobe_instance(void) { - __get_cpu_var(kprobe_instance) = NULL; + __this_cpu_write(kprobe_instance, NULL); } /* @@ -352,13 +354,20 @@ static inline int kprobe_aggrprobe(struct kprobe *p) return p->pre_handler == aggr_pre_handler; } +/* Return true(!0) if the kprobe is unused */ +static inline int kprobe_unused(struct kprobe *p) +{ + return kprobe_aggrprobe(p) && kprobe_disabled(p) && + list_empty(&p->list); +} + /* * Keep all fields in the kprobe consistent */ -static inline void copy_kprobe(struct kprobe *old_p, struct kprobe *p) +static inline void copy_kprobe(struct kprobe *ap, struct kprobe *p) { - memcpy(&p->opcode, &old_p->opcode, sizeof(kprobe_opcode_t)); - memcpy(&p->ainsn, &old_p->ainsn, sizeof(struct arch_specific_insn)); + memcpy(&p->opcode, &ap->opcode, sizeof(kprobe_opcode_t)); + memcpy(&p->ainsn, &ap->ainsn, sizeof(struct arch_specific_insn)); } #ifdef CONFIG_OPTPROBES @@ -382,6 +391,17 @@ void __kprobes opt_pre_handler(struct kprobe *p, struct pt_regs *regs) } } +/* Free optimized instructions and optimized_kprobe */ +static __kprobes void free_aggr_kprobe(struct kprobe *p) +{ + struct optimized_kprobe *op; + + op = container_of(p, struct optimized_kprobe, kp); + arch_remove_optimized_kprobe(op); + arch_remove_kprobe(p); + kfree(op); +} + /* Return true(!0) if the kprobe is ready for optimization. */ static inline int kprobe_optready(struct kprobe *p) { @@ -395,11 +415,38 @@ static inline int kprobe_optready(struct kprobe *p) return 0; } +/* Return true(!0) if the kprobe is disarmed. Note: p must be on hash list */ +static inline int kprobe_disarmed(struct kprobe *p) +{ + struct optimized_kprobe *op; + + /* If kprobe is not aggr/opt probe, just return kprobe is disabled */ + if (!kprobe_aggrprobe(p)) + return kprobe_disabled(p); + + op = container_of(p, struct optimized_kprobe, kp); + + return kprobe_disabled(p) && list_empty(&op->list); +} + +/* Return true(!0) if the probe is queued on (un)optimizing lists */ +static int __kprobes kprobe_queued(struct kprobe *p) +{ + struct optimized_kprobe *op; + + if (kprobe_aggrprobe(p)) { + op = container_of(p, struct optimized_kprobe, kp); + if (!list_empty(&op->list)) + return 1; + } + return 0; +} + /* * Return an optimized kprobe whose optimizing code replaces * instructions including addr (exclude breakpoint). */ -struct kprobe *__kprobes get_optimized_kprobe(unsigned long addr) +static struct kprobe *__kprobes get_optimized_kprobe(unsigned long addr) { int i; struct kprobe *p = NULL; @@ -420,30 +467,23 @@ struct kprobe *__kprobes get_optimized_kprobe(unsigned long addr) /* Optimization staging list, protected by kprobe_mutex */ static LIST_HEAD(optimizing_list); +static LIST_HEAD(unoptimizing_list); static void kprobe_optimizer(struct work_struct *work); static DECLARE_DELAYED_WORK(optimizing_work, kprobe_optimizer); +static DECLARE_COMPLETION(optimizer_comp); #define OPTIMIZE_DELAY 5 -/* Kprobe jump optimizer */ -static __kprobes void kprobe_optimizer(struct work_struct *work) +/* + * Optimize (replace a breakpoint with a jump) kprobes listed on + * optimizing_list. + */ +static __kprobes void do_optimize_kprobes(void) { - struct optimized_kprobe *op, *tmp; - - /* Lock modules while optimizing kprobes */ - mutex_lock(&module_mutex); - mutex_lock(&kprobe_mutex); - if (kprobes_all_disarmed || !kprobes_allow_optimization) - goto end; - - /* - * Wait for quiesence period to ensure all running interrupts - * are done. Because optprobe may modify multiple instructions - * there is a chance that Nth instruction is interrupted. In that - * case, running interrupt can return to 2nd-Nth byte of jump - * instruction. This wait is for avoiding it. - */ - synchronize_sched(); + /* Optimization never be done when disarmed */ + if (kprobes_all_disarmed || !kprobes_allow_optimization || + list_empty(&optimizing_list)) + return; /* * The optimization/unoptimization refers online_cpus via @@ -457,17 +497,111 @@ static __kprobes void kprobe_optimizer(struct work_struct *work) */ get_online_cpus(); mutex_lock(&text_mutex); - list_for_each_entry_safe(op, tmp, &optimizing_list, list) { - WARN_ON(kprobe_disabled(&op->kp)); - if (arch_optimize_kprobe(op) < 0) - op->kp.flags &= ~KPROBE_FLAG_OPTIMIZED; - list_del_init(&op->list); + arch_optimize_kprobes(&optimizing_list); + mutex_unlock(&text_mutex); + put_online_cpus(); +} + +/* + * Unoptimize (replace a jump with a breakpoint and remove the breakpoint + * if need) kprobes listed on unoptimizing_list. + */ +static __kprobes void do_unoptimize_kprobes(struct list_head *free_list) +{ + struct optimized_kprobe *op, *tmp; + + /* Unoptimization must be done anytime */ + if (list_empty(&unoptimizing_list)) + return; + + /* Ditto to do_optimize_kprobes */ + get_online_cpus(); + mutex_lock(&text_mutex); + arch_unoptimize_kprobes(&unoptimizing_list, free_list); + /* Loop free_list for disarming */ + list_for_each_entry_safe(op, tmp, free_list, list) { + /* Disarm probes if marked disabled */ + if (kprobe_disabled(&op->kp)) + arch_disarm_kprobe(&op->kp); + if (kprobe_unused(&op->kp)) { + /* + * Remove unused probes from hash list. After waiting + * for synchronization, these probes are reclaimed. + * (reclaiming is done by do_free_cleaned_kprobes.) + */ + hlist_del_rcu(&op->kp.hlist); + } else + list_del_init(&op->list); } mutex_unlock(&text_mutex); put_online_cpus(); -end: +} + +/* Reclaim all kprobes on the free_list */ +static __kprobes void do_free_cleaned_kprobes(struct list_head *free_list) +{ + struct optimized_kprobe *op, *tmp; + + list_for_each_entry_safe(op, tmp, free_list, list) { + BUG_ON(!kprobe_unused(&op->kp)); + list_del_init(&op->list); + free_aggr_kprobe(&op->kp); + } +} + +/* Start optimizer after OPTIMIZE_DELAY passed */ +static __kprobes void kick_kprobe_optimizer(void) +{ + if (!delayed_work_pending(&optimizing_work)) + schedule_delayed_work(&optimizing_work, OPTIMIZE_DELAY); +} + +/* Kprobe jump optimizer */ +static __kprobes void kprobe_optimizer(struct work_struct *work) +{ + LIST_HEAD(free_list); + + /* Lock modules while optimizing kprobes */ + mutex_lock(&module_mutex); + mutex_lock(&kprobe_mutex); + + /* + * Step 1: Unoptimize kprobes and collect cleaned (unused and disarmed) + * kprobes before waiting for quiesence period. + */ + do_unoptimize_kprobes(&free_list); + + /* + * Step 2: Wait for quiesence period to ensure all running interrupts + * are done. Because optprobe may modify multiple instructions + * there is a chance that Nth instruction is interrupted. In that + * case, running interrupt can return to 2nd-Nth byte of jump + * instruction. This wait is for avoiding it. + */ + synchronize_sched(); + + /* Step 3: Optimize kprobes after quiesence period */ + do_optimize_kprobes(); + + /* Step 4: Free cleaned kprobes after quiesence period */ + do_free_cleaned_kprobes(&free_list); + mutex_unlock(&kprobe_mutex); mutex_unlock(&module_mutex); + + /* Step 5: Kick optimizer again if needed */ + if (!list_empty(&optimizing_list) || !list_empty(&unoptimizing_list)) + kick_kprobe_optimizer(); + else + /* Wake up all waiters */ + complete_all(&optimizer_comp); +} + +/* Wait for completing optimization and unoptimization */ +static __kprobes void wait_for_kprobe_optimizer(void) +{ + if (delayed_work_pending(&optimizing_work)) + wait_for_completion(&optimizer_comp); } /* Optimize kprobe if p is ready to be optimized */ @@ -493,42 +627,99 @@ static __kprobes void optimize_kprobe(struct kprobe *p) /* Check if it is already optimized. */ if (op->kp.flags & KPROBE_FLAG_OPTIMIZED) return; - op->kp.flags |= KPROBE_FLAG_OPTIMIZED; - list_add(&op->list, &optimizing_list); - if (!delayed_work_pending(&optimizing_work)) - schedule_delayed_work(&optimizing_work, OPTIMIZE_DELAY); + + if (!list_empty(&op->list)) + /* This is under unoptimizing. Just dequeue the probe */ + list_del_init(&op->list); + else { + list_add(&op->list, &optimizing_list); + kick_kprobe_optimizer(); + } +} + +/* Short cut to direct unoptimizing */ +static __kprobes void force_unoptimize_kprobe(struct optimized_kprobe *op) +{ + get_online_cpus(); + arch_unoptimize_kprobe(op); + put_online_cpus(); + if (kprobe_disabled(&op->kp)) + arch_disarm_kprobe(&op->kp); } /* Unoptimize a kprobe if p is optimized */ -static __kprobes void unoptimize_kprobe(struct kprobe *p) +static __kprobes void unoptimize_kprobe(struct kprobe *p, bool force) { struct optimized_kprobe *op; - if ((p->flags & KPROBE_FLAG_OPTIMIZED) && kprobe_aggrprobe(p)) { - op = container_of(p, struct optimized_kprobe, kp); - if (!list_empty(&op->list)) - /* Dequeue from the optimization queue */ + if (!kprobe_aggrprobe(p) || kprobe_disarmed(p)) + return; /* This is not an optprobe nor optimized */ + + op = container_of(p, struct optimized_kprobe, kp); + if (!kprobe_optimized(p)) { + /* Unoptimized or unoptimizing case */ + if (force && !list_empty(&op->list)) { + /* + * Only if this is unoptimizing kprobe and forced, + * forcibly unoptimize it. (No need to unoptimize + * unoptimized kprobe again :) + */ list_del_init(&op->list); - else - /* Replace jump with break */ - arch_unoptimize_kprobe(op); - op->kp.flags &= ~KPROBE_FLAG_OPTIMIZED; + force_unoptimize_kprobe(op); + } + return; + } + + op->kp.flags &= ~KPROBE_FLAG_OPTIMIZED; + if (!list_empty(&op->list)) { + /* Dequeue from the optimization queue */ + list_del_init(&op->list); + return; + } + /* Optimized kprobe case */ + if (force) + /* Forcibly update the code: this is a special case */ + force_unoptimize_kprobe(op); + else { + list_add(&op->list, &unoptimizing_list); + kick_kprobe_optimizer(); } } +/* Cancel unoptimizing for reusing */ +static void reuse_unused_kprobe(struct kprobe *ap) +{ + struct optimized_kprobe *op; + + BUG_ON(!kprobe_unused(ap)); + /* + * Unused kprobe MUST be on the way of delayed unoptimizing (means + * there is still a relative jump) and disabled. + */ + op = container_of(ap, struct optimized_kprobe, kp); + if (unlikely(list_empty(&op->list))) + printk(KERN_WARNING "Warning: found a stray unused " + "aggrprobe@%p\n", ap->addr); + /* Enable the probe again */ + ap->flags &= ~KPROBE_FLAG_DISABLED; + /* Optimize it again (remove from op->list) */ + BUG_ON(!kprobe_optready(ap)); + optimize_kprobe(ap); +} + /* Remove optimized instructions */ static void __kprobes kill_optimized_kprobe(struct kprobe *p) { struct optimized_kprobe *op; op = container_of(p, struct optimized_kprobe, kp); - if (!list_empty(&op->list)) { - /* Dequeue from the optimization queue */ + if (!list_empty(&op->list)) + /* Dequeue from the (un)optimization queue */ list_del_init(&op->list); - op->kp.flags &= ~KPROBE_FLAG_OPTIMIZED; - } - /* Don't unoptimize, because the target code will be freed. */ + + op->kp.flags &= ~KPROBE_FLAG_OPTIMIZED; + /* Don't touch the code, because it is already freed. */ arch_remove_optimized_kprobe(op); } @@ -541,16 +732,6 @@ static __kprobes void prepare_optimized_kprobe(struct kprobe *p) arch_prepare_optimized_kprobe(op); } -/* Free optimized instructions and optimized_kprobe */ -static __kprobes void free_aggr_kprobe(struct kprobe *p) -{ - struct optimized_kprobe *op; - - op = container_of(p, struct optimized_kprobe, kp); - arch_remove_optimized_kprobe(op); - kfree(op); -} - /* Allocate new optimized_kprobe and try to prepare optimized instructions */ static __kprobes struct kprobe *alloc_aggr_kprobe(struct kprobe *p) { @@ -585,7 +766,8 @@ static __kprobes void try_to_optimize_kprobe(struct kprobe *p) op = container_of(ap, struct optimized_kprobe, kp); if (!arch_prepared_optinsn(&op->optinsn)) { /* If failed to setup optimizing, fallback to kprobe */ - free_aggr_kprobe(ap); + arch_remove_optimized_kprobe(op); + kfree(op); return; } @@ -594,6 +776,7 @@ static __kprobes void try_to_optimize_kprobe(struct kprobe *p) } #ifdef CONFIG_SYSCTL +/* This should be called with kprobe_mutex locked */ static void __kprobes optimize_all_kprobes(void) { struct hlist_head *head; @@ -606,17 +789,16 @@ static void __kprobes optimize_all_kprobes(void) return; kprobes_allow_optimization = true; - 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_disabled(p)) optimize_kprobe(p); } - mutex_unlock(&text_mutex); printk(KERN_INFO "Kprobes globally optimized\n"); } +/* This should be called with kprobe_mutex locked */ static void __kprobes unoptimize_all_kprobes(void) { struct hlist_head *head; @@ -629,21 +811,16 @@ static void __kprobes unoptimize_all_kprobes(void) return; kprobes_allow_optimization = false; - printk(KERN_INFO "Kprobes globally unoptimized\n"); - get_online_cpus(); /* For avoiding text_mutex deadlock */ - 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_disabled(p)) - unoptimize_kprobe(p); + unoptimize_kprobe(p, false); } } - - mutex_unlock(&text_mutex); - put_online_cpus(); - /* Allow all currently running kprobes to complete */ - synchronize_sched(); + /* Wait for unoptimizing completion */ + wait_for_kprobe_optimizer(); + printk(KERN_INFO "Kprobes globally unoptimized\n"); } int sysctl_kprobes_optimization; @@ -667,44 +844,60 @@ int proc_kprobes_optimization_handler(struct ctl_table *table, int write, } #endif /* CONFIG_SYSCTL */ +/* Put a breakpoint for a probe. Must be called with text_mutex locked */ static void __kprobes __arm_kprobe(struct kprobe *p) { - struct kprobe *old_p; + struct kprobe *_p; /* Check collision with other optimized kprobes */ - old_p = get_optimized_kprobe((unsigned long)p->addr); - if (unlikely(old_p)) - unoptimize_kprobe(old_p); /* Fallback to unoptimized kprobe */ + _p = get_optimized_kprobe((unsigned long)p->addr); + if (unlikely(_p)) + /* Fallback to unoptimized kprobe */ + unoptimize_kprobe(_p, true); arch_arm_kprobe(p); optimize_kprobe(p); /* Try to optimize (add kprobe to a list) */ } -static void __kprobes __disarm_kprobe(struct kprobe *p) +/* Remove the breakpoint of a probe. Must be called with text_mutex locked */ +static void __kprobes __disarm_kprobe(struct kprobe *p, bool reopt) { - struct kprobe *old_p; + struct kprobe *_p; - unoptimize_kprobe(p); /* Try to unoptimize */ - arch_disarm_kprobe(p); + unoptimize_kprobe(p, false); /* Try to unoptimize */ - /* If another kprobe was blocked, optimize it. */ - old_p = get_optimized_kprobe((unsigned long)p->addr); - if (unlikely(old_p)) - optimize_kprobe(old_p); + if (!kprobe_queued(p)) { + arch_disarm_kprobe(p); + /* If another kprobe was blocked, optimize it. */ + _p = get_optimized_kprobe((unsigned long)p->addr); + if (unlikely(_p) && reopt) + optimize_kprobe(_p); + } + /* TODO: reoptimize others after unoptimized this probe */ } #else /* !CONFIG_OPTPROBES */ #define optimize_kprobe(p) do {} while (0) -#define unoptimize_kprobe(p) do {} while (0) +#define unoptimize_kprobe(p, f) do {} while (0) #define kill_optimized_kprobe(p) do {} while (0) #define prepare_optimized_kprobe(p) do {} while (0) #define try_to_optimize_kprobe(p) do {} while (0) #define __arm_kprobe(p) arch_arm_kprobe(p) -#define __disarm_kprobe(p) arch_disarm_kprobe(p) +#define __disarm_kprobe(p, o) arch_disarm_kprobe(p) +#define kprobe_disarmed(p) kprobe_disabled(p) +#define wait_for_kprobe_optimizer() do {} while (0) + +/* There should be no unused kprobes can be reused without optimization */ +static void reuse_unused_kprobe(struct kprobe *ap) +{ + printk(KERN_ERR "Error: There should be no unused kprobe here.\n"); + BUG_ON(kprobe_unused(ap)); +} static __kprobes void free_aggr_kprobe(struct kprobe *p) { + arch_remove_kprobe(p); kfree(p); } @@ -730,11 +923,10 @@ static void __kprobes arm_kprobe(struct kprobe *kp) /* Disarm a kprobe with text_mutex */ static void __kprobes disarm_kprobe(struct kprobe *kp) { - get_online_cpus(); /* For avoiding text_mutex deadlock */ + /* Ditto */ mutex_lock(&text_mutex); - __disarm_kprobe(kp); + __disarm_kprobe(kp, true); mutex_unlock(&text_mutex); - put_online_cpus(); } /* @@ -773,7 +965,7 @@ static void __kprobes aggr_post_handler(struct kprobe *p, struct pt_regs *regs, static int __kprobes aggr_fault_handler(struct kprobe *p, struct pt_regs *regs, int trapnr) { - struct kprobe *cur = __get_cpu_var(kprobe_instance); + struct kprobe *cur = __this_cpu_read(kprobe_instance); /* * if we faulted "during" the execution of a user specified @@ -788,7 +980,7 @@ static int __kprobes aggr_fault_handler(struct kprobe *p, struct pt_regs *regs, static int __kprobes aggr_break_handler(struct kprobe *p, struct pt_regs *regs) { - struct kprobe *cur = __get_cpu_var(kprobe_instance); + struct kprobe *cur = __this_cpu_read(kprobe_instance); int ret = 0; if (cur && cur->break_handler) { @@ -831,6 +1023,7 @@ void __kprobes recycle_rp_inst(struct kretprobe_instance *ri, void __kprobes kretprobe_hash_lock(struct task_struct *tsk, struct hlist_head **head, unsigned long *flags) +__acquires(hlist_lock) { unsigned long hash = hash_ptr(tsk, KPROBE_HASH_BITS); spinlock_t *hlist_lock; @@ -842,6 +1035,7 @@ void __kprobes kretprobe_hash_lock(struct task_struct *tsk, static void __kprobes kretprobe_table_lock(unsigned long hash, unsigned long *flags) +__acquires(hlist_lock) { spinlock_t *hlist_lock = kretprobe_table_lock_ptr(hash); spin_lock_irqsave(hlist_lock, *flags); @@ -849,6 +1043,7 @@ static void __kprobes kretprobe_table_lock(unsigned long hash, void __kprobes kretprobe_hash_unlock(struct task_struct *tsk, unsigned long *flags) +__releases(hlist_lock) { unsigned long hash = hash_ptr(tsk, KPROBE_HASH_BITS); spinlock_t *hlist_lock; @@ -857,7 +1052,9 @@ void __kprobes kretprobe_hash_unlock(struct task_struct *tsk, spin_unlock_irqrestore(hlist_lock, *flags); } -void __kprobes kretprobe_table_unlock(unsigned long hash, unsigned long *flags) +static void __kprobes kretprobe_table_unlock(unsigned long hash, + unsigned long *flags) +__releases(hlist_lock) { spinlock_t *hlist_lock = kretprobe_table_lock_ptr(hash); spin_unlock_irqrestore(hlist_lock, *flags); @@ -935,7 +1132,7 @@ static int __kprobes add_new_kprobe(struct kprobe *ap, struct kprobe *p) BUG_ON(kprobe_gone(ap) || kprobe_gone(p)); if (p->break_handler || p->post_handler) - unoptimize_kprobe(ap); /* Fall back to normal kprobe */ + unoptimize_kprobe(ap, true); /* Fall back to normal kprobe */ if (p->break_handler) { if (ap->break_handler) @@ -986,19 +1183,21 @@ static void __kprobes init_aggr_kprobe(struct kprobe *ap, struct kprobe *p) * This is the second or subsequent kprobe at the address - handle * the intricacies */ -static int __kprobes register_aggr_kprobe(struct kprobe *old_p, +static int __kprobes register_aggr_kprobe(struct kprobe *orig_p, struct kprobe *p) { int ret = 0; - struct kprobe *ap = old_p; + struct kprobe *ap = orig_p; - if (!kprobe_aggrprobe(old_p)) { - /* If old_p is not an aggr_kprobe, create new aggr_kprobe. */ - ap = alloc_aggr_kprobe(old_p); + if (!kprobe_aggrprobe(orig_p)) { + /* If orig_p is not an aggr_kprobe, create new aggr_kprobe. */ + ap = alloc_aggr_kprobe(orig_p); if (!ap) return -ENOMEM; - init_aggr_kprobe(ap, old_p); - } + init_aggr_kprobe(ap, orig_p); + } else if (kprobe_unused(ap)) + /* This probe is going to die. Rescue it */ + reuse_unused_kprobe(ap); if (kprobe_gone(ap)) { /* @@ -1032,23 +1231,6 @@ static int __kprobes register_aggr_kprobe(struct kprobe *old_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) { struct kprobe_blackpoint *kb; @@ -1091,34 +1273,33 @@ static kprobe_opcode_t __kprobes *kprobe_addr(struct kprobe *p) /* 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; + struct kprobe *ap, *list_p; - old_p = get_kprobe(p->addr); - if (unlikely(!old_p)) + ap = get_kprobe(p->addr); + if (unlikely(!ap)) return NULL; - if (p != old_p) { - list_for_each_entry_rcu(list_p, &old_p->list, list) + if (p != ap) { + list_for_each_entry_rcu(list_p, &ap->list, list) if (list_p == p) /* kprobe p is a valid probe */ goto valid; return NULL; } valid: - return old_p; + return ap; } /* 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) + if (__get_valid_kprobe(p)) ret = -EINVAL; mutex_unlock(&kprobe_mutex); + return ret; } @@ -1138,13 +1319,13 @@ int __kprobes register_kprobe(struct kprobe *p) if (ret) return ret; + jump_label_lock(); preempt_disable(); if (!kernel_text_address((unsigned long) p->addr) || in_kprobes_functions((unsigned long) p->addr) || - ftrace_text_reserved(p->addr, p->addr)) { - preempt_enable(); - return -EINVAL; - } + ftrace_text_reserved(p->addr, p->addr) || + jump_label_text_reserved(p->addr, p->addr)) + goto fail_with_jump_label; /* User can pass only KPROBE_FLAG_DISABLED to register_kprobe */ p->flags &= KPROBE_FLAG_DISABLED; @@ -1158,10 +1339,9 @@ int __kprobes register_kprobe(struct kprobe *p) * We must hold a refcount of the probed module while updating * its code to prohibit unexpected unloading. */ - if (unlikely(!try_module_get(probed_mod))) { - preempt_enable(); - return -EINVAL; - } + if (unlikely(!try_module_get(probed_mod))) + goto fail_with_jump_label; + /* * If the module freed .init.text, we couldn't insert * kprobes in there. @@ -1169,16 +1349,18 @@ int __kprobes register_kprobe(struct kprobe *p) if (within_module_init((unsigned long)p->addr, probed_mod) && probed_mod->state != MODULE_STATE_COMING) { module_put(probed_mod); - preempt_enable(); - return -EINVAL; + goto fail_with_jump_label; } } preempt_enable(); + jump_label_unlock(); p->nmissed = 0; INIT_LIST_HEAD(&p->list); mutex_lock(&kprobe_mutex); + jump_label_lock(); /* needed to call jump_label_text_reserved() */ + get_online_cpus(); /* For avoiding text_mutex deadlock. */ mutex_lock(&text_mutex); @@ -1206,76 +1388,136 @@ int __kprobes register_kprobe(struct kprobe *p) out: mutex_unlock(&text_mutex); put_online_cpus(); + jump_label_unlock(); mutex_unlock(&kprobe_mutex); if (probed_mod) module_put(probed_mod); return ret; + +fail_with_jump_label: + preempt_enable(); + jump_label_unlock(); + return -EINVAL; } EXPORT_SYMBOL_GPL(register_kprobe); +/* Check if all probes on the aggrprobe are disabled */ +static int __kprobes aggr_kprobe_disabled(struct kprobe *ap) +{ + struct kprobe *kp; + + list_for_each_entry_rcu(kp, &ap->list, list) + if (!kprobe_disabled(kp)) + /* + * There is an active probe on the list. + * We can't disable this ap. + */ + return 0; + + return 1; +} + +/* Disable one kprobe: Make sure called under kprobe_mutex is locked */ +static struct kprobe *__kprobes __disable_kprobe(struct kprobe *p) +{ + struct kprobe *orig_p; + + /* Get an original kprobe for return */ + orig_p = __get_valid_kprobe(p); + if (unlikely(orig_p == NULL)) + return NULL; + + if (!kprobe_disabled(p)) { + /* Disable probe if it is a child probe */ + if (p != orig_p) + p->flags |= KPROBE_FLAG_DISABLED; + + /* Try to disarm and disable this/parent probe */ + if (p == orig_p || aggr_kprobe_disabled(orig_p)) { + disarm_kprobe(orig_p); + orig_p->flags |= KPROBE_FLAG_DISABLED; + } + } + + return orig_p; +} + /* * Unregister a kprobe without a scheduler synchronization. */ static int __kprobes __unregister_kprobe_top(struct kprobe *p) { - struct kprobe *old_p, *list_p; + struct kprobe *ap, *list_p; - old_p = __get_valid_kprobe(p); - if (old_p == NULL) + /* Disable kprobe. This will disarm it if needed. */ + ap = __disable_kprobe(p); + if (ap == NULL) return -EINVAL; - if (old_p == p || - (kprobe_aggrprobe(old_p) && - list_is_singular(&old_p->list))) { + if (ap == p) /* - * Only probe on the hash list. Disarm only if kprobes are - * enabled and not gone - otherwise, the breakpoint would - * already have been removed. We save on flushing icache. + * This probe is an independent(and non-optimized) kprobe + * (not an aggrprobe). Remove from the hash list. */ - if (!kprobes_all_disarmed && !kprobe_disabled(old_p)) - disarm_kprobe(old_p); - hlist_del_rcu(&old_p->hlist); - } else { + goto disarmed; + + /* Following process expects this probe is an aggrprobe */ + WARN_ON(!kprobe_aggrprobe(ap)); + + if (list_is_singular(&ap->list) && kprobe_disarmed(ap)) + /* + * !disarmed could be happen if the probe is under delayed + * unoptimizing. + */ + goto disarmed; + else { + /* If disabling probe has special handlers, update aggrprobe */ if (p->break_handler && !kprobe_gone(p)) - old_p->break_handler = NULL; + ap->break_handler = NULL; if (p->post_handler && !kprobe_gone(p)) { - list_for_each_entry_rcu(list_p, &old_p->list, list) { + list_for_each_entry_rcu(list_p, &ap->list, list) { if ((list_p != p) && (list_p->post_handler)) goto noclean; } - old_p->post_handler = NULL; + ap->post_handler = NULL; } noclean: + /* + * Remove from the aggrprobe: this path will do nothing in + * __unregister_kprobe_bottom(). + */ list_del_rcu(&p->list); - if (!kprobe_disabled(old_p)) { - try_to_disable_aggr_kprobe(old_p); - if (!kprobes_all_disarmed) { - if (kprobe_disabled(old_p)) - disarm_kprobe(old_p); - else - /* Try to optimize this probe again */ - optimize_kprobe(old_p); - } - } + if (!kprobe_disabled(ap) && !kprobes_all_disarmed) + /* + * Try to optimize this probe again, because post + * handler may have been changed. + */ + optimize_kprobe(ap); } return 0; + +disarmed: + BUG_ON(!kprobe_disarmed(ap)); + hlist_del_rcu(&ap->hlist); + return 0; } static void __kprobes __unregister_kprobe_bottom(struct kprobe *p) { - struct kprobe *old_p; + struct kprobe *ap; if (list_empty(&p->list)) + /* This is an independent kprobe */ arch_remove_kprobe(p); else if (list_is_singular(&p->list)) { - /* "p" is the last child of an aggr_kprobe */ - old_p = list_entry(p->list.next, struct kprobe, list); + /* This is the last child of an aggrprobe */ + ap = list_entry(p->list.next, struct kprobe, list); list_del(&p->list); - arch_remove_kprobe(old_p); - free_aggr_kprobe(old_p); + free_aggr_kprobe(ap); } + /* Otherwise, do nothing. */ } int __kprobes register_kprobes(struct kprobe **kps, int num) @@ -1339,18 +1581,19 @@ int __kprobes register_jprobes(struct jprobe **jps, int num) if (num <= 0) return -EINVAL; for (i = 0; i < num; i++) { - unsigned long addr; + unsigned long addr, offset; jp = jps[i]; addr = arch_deref_entry_point(jp->entry); - if (!kernel_text_address(addr)) - ret = -EINVAL; - else { - /* Todo: Verify probepoint is a function entry point */ + /* Verify probepoint is a function entry point */ + if (kallsyms_lookup_size_offset(addr, NULL, &offset) && + offset == 0) { jp->kp.pre_handler = setjmp_pre_handler; jp->kp.break_handler = longjmp_break_handler; ret = register_kprobe(&jp->kp); - } + } else + ret = -EINVAL; + if (ret < 0) { if (i > 0) unregister_jprobes(jps, i); @@ -1592,29 +1835,13 @@ static void __kprobes kill_kprobe(struct kprobe *p) 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)) { + /* Disable this kprobe */ + if (__disable_kprobe(kp) == 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; } @@ -1912,36 +2139,27 @@ static void __kprobes disarm_all_kprobes(void) mutex_lock(&kprobe_mutex); /* If kprobes are already disarmed, just return */ - if (kprobes_all_disarmed) - goto already_disabled; + if (kprobes_all_disarmed) { + mutex_unlock(&kprobe_mutex); + return; + } kprobes_all_disarmed = true; printk(KERN_INFO "Kprobes globally disabled\n"); - /* - * Here we call get_online_cpus() for avoiding text_mutex deadlock, - * because disarming may also unoptimize kprobes. - */ - get_online_cpus(); 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_disabled(p)) - __disarm_kprobe(p); + __disarm_kprobe(p, false); } } - mutex_unlock(&text_mutex); - put_online_cpus(); mutex_unlock(&kprobe_mutex); - /* Allow all currently running kprobes to complete */ - synchronize_sched(); - return; -already_disabled: - mutex_unlock(&kprobe_mutex); - return; + /* Wait for disarming all kprobes by optimizer */ + wait_for_kprobe_optimizer(); } /* @@ -1992,6 +2210,7 @@ static ssize_t write_enabled_file_bool(struct file *file, static const struct file_operations fops_kp = { .read = read_enabled_file_bool, .write = write_enabled_file_bool, + .llseek = default_llseek, }; static int __kprobes debugfs_kprobe_init(void) diff --git a/kernel/kthread.c b/kernel/kthread.c index 83911c78017..c55afba990a 100644 --- a/kernel/kthread.c +++ b/kernel/kthread.c @@ -14,6 +14,8 @@ #include <linux/file.h> #include <linux/module.h> #include <linux/mutex.h> +#include <linux/slab.h> +#include <linux/freezer.h> #include <trace/events/sched.h> static DEFINE_SPINLOCK(kthread_create_lock); @@ -35,6 +37,7 @@ struct kthread_create_info struct kthread { int should_stop; + void *data; struct completion exited; }; @@ -54,6 +57,19 @@ int kthread_should_stop(void) } EXPORT_SYMBOL(kthread_should_stop); +/** + * kthread_data - return data value specified on kthread creation + * @task: kthread task in question + * + * Return the data value specified when kthread @task was created. + * The caller is responsible for ensuring the validity of @task when + * calling this function. + */ +void *kthread_data(struct task_struct *task) +{ + return to_kthread(task)->data; +} + static int kthread(void *_create) { /* Copy data: it's on kthread's stack */ @@ -64,6 +80,7 @@ static int kthread(void *_create) int ret; self.should_stop = 0; + self.data = data; init_completion(&self.exited); current->vfork_done = &self.exited; @@ -131,7 +148,7 @@ 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 }; + static const struct sched_param param = { .sched_priority = 0 }; va_list args; va_start(args, namefmt); @@ -247,3 +264,161 @@ int kthreadd(void *unused) return 0; } + +void __init_kthread_worker(struct kthread_worker *worker, + const char *name, + struct lock_class_key *key) +{ + spin_lock_init(&worker->lock); + lockdep_set_class_and_name(&worker->lock, key, name); + INIT_LIST_HEAD(&worker->work_list); + worker->task = NULL; +} +EXPORT_SYMBOL_GPL(__init_kthread_worker); + +/** + * kthread_worker_fn - kthread function to process kthread_worker + * @worker_ptr: pointer to initialized kthread_worker + * + * This function can be used as @threadfn to kthread_create() or + * kthread_run() with @worker_ptr argument pointing to an initialized + * kthread_worker. The started kthread will process work_list until + * the it is stopped with kthread_stop(). A kthread can also call + * this function directly after extra initialization. + * + * Different kthreads can be used for the same kthread_worker as long + * as there's only one kthread attached to it at any given time. A + * kthread_worker without an attached kthread simply collects queued + * kthread_works. + */ +int kthread_worker_fn(void *worker_ptr) +{ + struct kthread_worker *worker = worker_ptr; + struct kthread_work *work; + + WARN_ON(worker->task); + worker->task = current; +repeat: + set_current_state(TASK_INTERRUPTIBLE); /* mb paired w/ kthread_stop */ + + if (kthread_should_stop()) { + __set_current_state(TASK_RUNNING); + spin_lock_irq(&worker->lock); + worker->task = NULL; + spin_unlock_irq(&worker->lock); + return 0; + } + + work = NULL; + spin_lock_irq(&worker->lock); + if (!list_empty(&worker->work_list)) { + work = list_first_entry(&worker->work_list, + struct kthread_work, node); + list_del_init(&work->node); + } + spin_unlock_irq(&worker->lock); + + if (work) { + __set_current_state(TASK_RUNNING); + work->func(work); + smp_wmb(); /* wmb worker-b0 paired with flush-b1 */ + work->done_seq = work->queue_seq; + smp_mb(); /* mb worker-b1 paired with flush-b0 */ + if (atomic_read(&work->flushing)) + wake_up_all(&work->done); + } else if (!freezing(current)) + schedule(); + + try_to_freeze(); + goto repeat; +} +EXPORT_SYMBOL_GPL(kthread_worker_fn); + +/** + * queue_kthread_work - queue a kthread_work + * @worker: target kthread_worker + * @work: kthread_work to queue + * + * Queue @work to work processor @task for async execution. @task + * must have been created with kthread_worker_create(). Returns %true + * if @work was successfully queued, %false if it was already pending. + */ +bool queue_kthread_work(struct kthread_worker *worker, + struct kthread_work *work) +{ + bool ret = false; + unsigned long flags; + + spin_lock_irqsave(&worker->lock, flags); + if (list_empty(&work->node)) { + list_add_tail(&work->node, &worker->work_list); + work->queue_seq++; + if (likely(worker->task)) + wake_up_process(worker->task); + ret = true; + } + spin_unlock_irqrestore(&worker->lock, flags); + return ret; +} +EXPORT_SYMBOL_GPL(queue_kthread_work); + +/** + * flush_kthread_work - flush a kthread_work + * @work: work to flush + * + * If @work is queued or executing, wait for it to finish execution. + */ +void flush_kthread_work(struct kthread_work *work) +{ + int seq = work->queue_seq; + + atomic_inc(&work->flushing); + + /* + * mb flush-b0 paired with worker-b1, to make sure either + * worker sees the above increment or we see done_seq update. + */ + smp_mb__after_atomic_inc(); + + /* A - B <= 0 tests whether B is in front of A regardless of overflow */ + wait_event(work->done, seq - work->done_seq <= 0); + atomic_dec(&work->flushing); + + /* + * rmb flush-b1 paired with worker-b0, to make sure our caller + * sees every change made by work->func(). + */ + smp_mb__after_atomic_dec(); +} +EXPORT_SYMBOL_GPL(flush_kthread_work); + +struct kthread_flush_work { + struct kthread_work work; + struct completion done; +}; + +static void kthread_flush_work_fn(struct kthread_work *work) +{ + struct kthread_flush_work *fwork = + container_of(work, struct kthread_flush_work, work); + complete(&fwork->done); +} + +/** + * flush_kthread_worker - flush all current works on a kthread_worker + * @worker: worker to flush + * + * Wait until all currently executing or pending works on @worker are + * finished. + */ +void flush_kthread_worker(struct kthread_worker *worker) +{ + struct kthread_flush_work fwork = { + KTHREAD_WORK_INIT(fwork.work, kthread_flush_work_fn), + COMPLETION_INITIALIZER_ONSTACK(fwork.done), + }; + + queue_kthread_work(worker, &fwork.work); + wait_for_completion(&fwork.done); +} +EXPORT_SYMBOL_GPL(flush_kthread_worker); diff --git a/kernel/latencytop.c b/kernel/latencytop.c index 877fb306d41..ee74b35e528 100644 --- a/kernel/latencytop.c +++ b/kernel/latencytop.c @@ -194,14 +194,7 @@ __account_scheduler_latency(struct task_struct *tsk, int usecs, int inter) account_global_scheduler_latency(tsk, &lat); - /* - * short term hack; if we're > 32 we stop; future we recycle: - */ - tsk->latency_record_count++; - if (tsk->latency_record_count >= LT_SAVECOUNT) - goto out_unlock; - - for (i = 0; i < LT_SAVECOUNT; i++) { + for (i = 0; i < tsk->latency_record_count; i++) { struct latency_record *mylat; int same = 1; @@ -227,8 +220,14 @@ __account_scheduler_latency(struct task_struct *tsk, int usecs, int inter) } } + /* + * short term hack; if we're > 32 we stop; future we recycle: + */ + if (tsk->latency_record_count >= LT_SAVECOUNT) + goto out_unlock; + /* Allocated a new one: */ - i = tsk->latency_record_count; + i = tsk->latency_record_count++; memcpy(&tsk->latency_record[i], &lat, sizeof(struct latency_record)); out_unlock: @@ -242,24 +241,19 @@ static int lstats_show(struct seq_file *m, void *v) seq_puts(m, "Latency Top version : v0.1\n"); for (i = 0; i < MAXLR; i++) { - if (latency_record[i].backtrace[0]) { + struct latency_record *lr = &latency_record[i]; + + if (lr->backtrace[0]) { int q; - seq_printf(m, "%i %lu %lu ", - latency_record[i].count, - latency_record[i].time, - latency_record[i].max); + seq_printf(m, "%i %lu %lu", + lr->count, lr->time, lr->max); for (q = 0; q < LT_BACKTRACEDEPTH; q++) { - char sym[KSYM_SYMBOL_LEN]; - char *c; - if (!latency_record[i].backtrace[q]) + unsigned long bt = lr->backtrace[q]; + if (!bt) break; - if (latency_record[i].backtrace[q] == ULONG_MAX) + if (bt == ULONG_MAX) break; - sprint_symbol(sym, latency_record[i].backtrace[q]); - c = strchr(sym, '+'); - if (c) - *c = 0; - seq_printf(m, "%s ", sym); + seq_printf(m, " %ps", (void *)bt); } seq_printf(m, "\n"); } diff --git a/kernel/lockdep.c b/kernel/lockdep.c index 54286798c37..0d2058da80f 100644 --- a/kernel/lockdep.c +++ b/kernel/lockdep.c @@ -146,7 +146,7 @@ static DEFINE_PER_CPU(struct lock_class_stats[MAX_LOCKDEP_KEYS], static inline u64 lockstat_clock(void) { - return cpu_clock(smp_processor_id()); + return local_clock(); } static int lock_point(unsigned long points[], unsigned long ip) @@ -639,6 +639,16 @@ look_up_lock_class(struct lockdep_map *lock, unsigned int subclass) } #endif + if (unlikely(subclass >= MAX_LOCKDEP_SUBCLASSES)) { + debug_locks_off(); + printk(KERN_ERR + "BUG: looking up invalid subclass: %u\n", subclass); + printk(KERN_ERR + "turning off the locking correctness validator.\n"); + dump_stack(); + return NULL; + } + /* * Static locks do not have their class-keys yet - for them the key * is the lock object itself: @@ -774,7 +784,9 @@ out_unlock_set: raw_local_irq_restore(flags); if (!subclass || force) - lock->class_cache = class; + lock->class_cache[0] = class; + else if (subclass < NR_LOCKDEP_CACHING_CLASSES) + lock->class_cache[subclass] = class; if (DEBUG_LOCKS_WARN_ON(class->subclass != subclass)) return NULL; @@ -2280,22 +2292,6 @@ mark_held_locks(struct task_struct *curr, enum mark_type mark) } /* - * Debugging helper: via this flag we know that we are in - * 'early bootup code', and will warn about any invalid irqs-on event: - */ -static int early_boot_irqs_enabled; - -void early_boot_irqs_off(void) -{ - early_boot_irqs_enabled = 0; -} - -void early_boot_irqs_on(void) -{ - early_boot_irqs_enabled = 1; -} - -/* * Hardirqs will be enabled: */ void trace_hardirqs_on_caller(unsigned long ip) @@ -2307,7 +2303,7 @@ void trace_hardirqs_on_caller(unsigned long ip) if (unlikely(!debug_locks || current->lockdep_recursion)) return; - if (DEBUG_LOCKS_WARN_ON(unlikely(!early_boot_irqs_enabled))) + if (DEBUG_LOCKS_WARN_ON(unlikely(early_boot_irqs_disabled))) return; if (unlikely(curr->hardirqs_enabled)) { @@ -2679,7 +2675,11 @@ 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) { - lock->class_cache = NULL; + int i; + + for (i = 0; i < NR_LOCKDEP_CACHING_CLASSES; i++) + lock->class_cache[i] = NULL; + #ifdef CONFIG_LOCK_STAT lock->cpu = raw_smp_processor_id(); #endif @@ -2739,21 +2739,13 @@ static int __lock_acquire(struct lockdep_map *lock, unsigned int subclass, if (DEBUG_LOCKS_WARN_ON(!irqs_disabled())) return 0; - if (unlikely(subclass >= MAX_LOCKDEP_SUBCLASSES)) { - debug_locks_off(); - printk("BUG: MAX_LOCKDEP_SUBCLASSES too low!\n"); - printk("turning off the locking correctness validator.\n"); - dump_stack(); - return 0; - } - if (lock->key == &__lockdep_no_validate__) check = 1; - if (!subclass) - class = lock->class_cache; + if (subclass < NR_LOCKDEP_CACHING_CLASSES) + class = lock->class_cache[subclass]; /* - * Not cached yet or subclass? + * Not cached? */ if (unlikely(!class)) { class = register_lock_class(lock, subclass, 0); @@ -2918,7 +2910,7 @@ static int match_held_lock(struct held_lock *hlock, struct lockdep_map *lock) return 1; if (hlock->references) { - struct lock_class *class = lock->class_cache; + struct lock_class *class = lock->class_cache[0]; if (!class) class = look_up_lock_class(lock, 0); @@ -3559,7 +3551,12 @@ void lockdep_reset_lock(struct lockdep_map *lock) if (list_empty(head)) continue; list_for_each_entry_safe(class, next, head, hash_entry) { - if (unlikely(class == lock->class_cache)) { + int match = 0; + + for (j = 0; j < NR_LOCKDEP_CACHING_CLASSES; j++) + match |= class == lock->class_cache[j]; + + if (unlikely(match)) { if (debug_locks_off_graph_unlock()) WARN_ON(1); goto out_restore; @@ -3775,7 +3772,7 @@ EXPORT_SYMBOL_GPL(debug_show_all_locks); * Careful: only use this function if you are sure that * the task cannot run in parallel! */ -void __debug_show_held_locks(struct task_struct *task) +void debug_show_held_locks(struct task_struct *task) { if (unlikely(!debug_locks)) { printk("INFO: lockdep is turned off.\n"); @@ -3783,12 +3780,6 @@ void __debug_show_held_locks(struct task_struct *task) } lockdep_print_held_locks(task); } -EXPORT_SYMBOL_GPL(__debug_show_held_locks); - -void debug_show_held_locks(struct task_struct *task) -{ - __debug_show_held_locks(task); -} EXPORT_SYMBOL_GPL(debug_show_held_locks); void lockdep_sys_exit(void) diff --git a/kernel/lockdep_proc.c b/kernel/lockdep_proc.c index 59b76c8ce9d..1969d2fc4b3 100644 --- a/kernel/lockdep_proc.c +++ b/kernel/lockdep_proc.c @@ -494,7 +494,6 @@ static void seq_stats(struct seq_file *m, struct lock_stat_data *data) namelen += 2; for (i = 0; i < LOCKSTAT_POINTS; i++) { - char sym[KSYM_SYMBOL_LEN]; char ip[32]; if (class->contention_point[i] == 0) @@ -503,15 +502,13 @@ static void seq_stats(struct seq_file *m, struct lock_stat_data *data) if (!i) seq_line(m, '-', 40-namelen, namelen); - sprint_symbol(sym, class->contention_point[i]); snprintf(ip, sizeof(ip), "[<%p>]", (void *)class->contention_point[i]); - seq_printf(m, "%40s %14lu %29s %s\n", name, - stats->contention_point[i], - ip, sym); + seq_printf(m, "%40s %14lu %29s %pS\n", + name, stats->contention_point[i], + ip, (void *)class->contention_point[i]); } for (i = 0; i < LOCKSTAT_POINTS; i++) { - char sym[KSYM_SYMBOL_LEN]; char ip[32]; if (class->contending_point[i] == 0) @@ -520,12 +517,11 @@ static void seq_stats(struct seq_file *m, struct lock_stat_data *data) if (!i) seq_line(m, '-', 40-namelen, namelen); - sprint_symbol(sym, class->contending_point[i]); snprintf(ip, sizeof(ip), "[<%p>]", (void *)class->contending_point[i]); - seq_printf(m, "%40s %14lu %29s %s\n", name, - stats->contending_point[i], - ip, sym); + seq_printf(m, "%40s %14lu %29s %pS\n", + name, stats->contending_point[i], + ip, (void *)class->contending_point[i]); } if (i) { seq_puts(m, "\n"); diff --git a/kernel/module.c b/kernel/module.c index a8014bfb5a4..34e00b708fa 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -1,6 +1,6 @@ /* Copyright (C) 2002 Richard Henderson - Copyright (C) 2001 Rusty Russell, 2002 Rusty Russell IBM. + Copyright (C) 2001 Rusty Russell, 2002, 2010 Rusty Russell IBM. 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 @@ -55,6 +55,8 @@ #include <linux/async.h> #include <linux/percpu.h> #include <linux/kmemleak.h> +#include <linux/jump_label.h> +#include <linux/pfn.h> #define CREATE_TRACE_POINTS #include <trace/events/module.h> @@ -69,10 +71,34 @@ #define ARCH_SHF_SMALL 0 #endif +/* + * Modules' sections will be aligned on page boundaries + * to ensure complete separation of code and data, but + * only when CONFIG_DEBUG_SET_MODULE_RONX=y + */ +#ifdef CONFIG_DEBUG_SET_MODULE_RONX +# define debug_align(X) ALIGN(X, PAGE_SIZE) +#else +# define debug_align(X) (X) +#endif + +/* + * Given BASE and SIZE this macro calculates the number of pages the + * memory regions occupies + */ +#define MOD_NUMBER_OF_PAGES(BASE, SIZE) (((SIZE) > 0) ? \ + (PFN_DOWN((unsigned long)(BASE) + (SIZE) - 1) - \ + PFN_DOWN((unsigned long)BASE) + 1) \ + : (0UL)) + /* If this is set, the section belongs in the init part of the module */ #define INIT_OFFSET_MASK (1UL << (BITS_PER_LONG-1)) -/* List of modules, protected by module_mutex or preempt_disable +/* + * Mutex protects: + * 1) List of modules (also safely readable with preempt_disable), + * 2) module_use links, + * 3) module_addr_min/module_addr_max. * (delete uses stop_machine/add uses RCU list operations). */ DEFINE_MUTEX(module_mutex); EXPORT_SYMBOL_GPL(module_mutex); @@ -90,7 +116,8 @@ static DECLARE_WAIT_QUEUE_HEAD(module_wq); static BLOCKING_NOTIFIER_HEAD(module_notify_list); -/* Bounds of module allocation, for speeding __module_address */ +/* Bounds of module allocation, for speeding __module_address. + * Protected by module_mutex. */ static unsigned long module_addr_min = -1UL, module_addr_max = 0; int register_module_notifier(struct notifier_block * nb) @@ -105,6 +132,20 @@ int unregister_module_notifier(struct notifier_block * nb) } EXPORT_SYMBOL(unregister_module_notifier); +struct load_info { + Elf_Ehdr *hdr; + unsigned long len; + Elf_Shdr *sechdrs; + char *secstrings, *strtab; + unsigned long *strmap; + unsigned long symoffs, stroffs; + struct _ddebug *debug; + unsigned int num_debug; + struct { + unsigned int sym, str, mod, vers, info, pcpu; + } index; +}; + /* We require a truly strong try_module_get(): 0 means failure due to ongoing or failed initialization etc. */ static inline int strong_try_module_get(struct module *mod) @@ -135,42 +176,38 @@ void __module_put_and_exit(struct module *mod, long code) EXPORT_SYMBOL(__module_put_and_exit); /* Find a module section: 0 means not found. */ -static unsigned int find_sec(Elf_Ehdr *hdr, - Elf_Shdr *sechdrs, - const char *secstrings, - const char *name) +static unsigned int find_sec(const struct load_info *info, const char *name) { unsigned int i; - for (i = 1; i < hdr->e_shnum; i++) + for (i = 1; i < info->hdr->e_shnum; i++) { + Elf_Shdr *shdr = &info->sechdrs[i]; /* Alloc bit cleared means "ignore it." */ - if ((sechdrs[i].sh_flags & SHF_ALLOC) - && strcmp(secstrings+sechdrs[i].sh_name, name) == 0) + if ((shdr->sh_flags & SHF_ALLOC) + && strcmp(info->secstrings + shdr->sh_name, name) == 0) return i; + } return 0; } /* Find a module section, or NULL. */ -static void *section_addr(Elf_Ehdr *hdr, Elf_Shdr *shdrs, - const char *secstrings, const char *name) +static void *section_addr(const struct load_info *info, const char *name) { /* Section 0 has sh_addr 0. */ - return (void *)shdrs[find_sec(hdr, shdrs, secstrings, name)].sh_addr; + return (void *)info->sechdrs[find_sec(info, name)].sh_addr; } /* Find a module section, or NULL. Fill in number of "objects" in section. */ -static void *section_objs(Elf_Ehdr *hdr, - Elf_Shdr *sechdrs, - const char *secstrings, +static void *section_objs(const struct load_info *info, const char *name, size_t object_size, unsigned int *num) { - unsigned int sec = find_sec(hdr, sechdrs, secstrings, name); + unsigned int sec = find_sec(info, name); /* Section 0 has sh_addr 0 and sh_size 0. */ - *num = sechdrs[sec].sh_size / object_size; - return (void *)sechdrs[sec].sh_addr; + *num = info->sechdrs[sec].sh_size / object_size; + return (void *)info->sechdrs[sec].sh_addr; } /* Provided by the linker */ @@ -180,8 +217,6 @@ extern const struct kernel_symbol __start___ksymtab_gpl[]; extern const struct kernel_symbol __stop___ksymtab_gpl[]; extern const struct kernel_symbol __start___ksymtab_gpl_future[]; extern const struct kernel_symbol __stop___ksymtab_gpl_future[]; -extern const struct kernel_symbol __start___ksymtab_gpl_future[]; -extern const struct kernel_symbol __stop___ksymtab_gpl_future[]; extern const unsigned long __start___kcrctab[]; extern const unsigned long __start___kcrctab_gpl[]; extern const unsigned long __start___kcrctab_gpl_future[]; @@ -224,7 +259,7 @@ 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[] = { + static const struct symsearch arr[] = { { __start___ksymtab, __stop___ksymtab, __start___kcrctab, NOT_GPL_ONLY, false }, { __start___ksymtab_gpl, __stop___ksymtab_gpl, @@ -331,7 +366,7 @@ static bool find_symbol_in_section(const struct symsearch *syms, } /* Find a symbol and return it, along with, (optional) crc and - * (optional) module which owns it */ + * (optional) module which owns it. Needs preempt disabled or module_mutex. */ const struct kernel_symbol *find_symbol(const char *name, struct module **owner, const unsigned long **crc, @@ -389,7 +424,8 @@ static int percpu_modalloc(struct module *mod, mod->percpu = __alloc_reserved_percpu(size, align); if (!mod->percpu) { printk(KERN_WARNING - "Could not allocate %lu bytes percpu data\n", size); + "%s: Could not allocate %lu bytes percpu data\n", + mod->name, size); return -ENOMEM; } mod->percpu_size = size; @@ -401,11 +437,9 @@ static void percpu_modfree(struct module *mod) free_percpu(mod->percpu); } -static unsigned int find_pcpusec(Elf_Ehdr *hdr, - Elf_Shdr *sechdrs, - const char *secstrings) +static unsigned int find_pcpusec(struct load_info *info) { - return find_sec(hdr, sechdrs, secstrings, ".data.percpu"); + return find_sec(info, ".data..percpu"); } static void percpu_modcopy(struct module *mod, @@ -465,9 +499,7 @@ static inline int percpu_modalloc(struct module *mod, static inline void percpu_modfree(struct module *mod) { } -static inline unsigned int find_pcpusec(Elf_Ehdr *hdr, - Elf_Shdr *sechdrs, - const char *secstrings) +static unsigned int find_pcpusec(struct load_info *info) { return 0; } @@ -521,36 +553,30 @@ static char last_unloaded_module[MODULE_NAME_LEN+1]; EXPORT_TRACEPOINT_SYMBOL(module_get); /* Init the unload section of the module. */ -static void module_unload_init(struct module *mod) +static int module_unload_init(struct module *mod) { - int cpu; + mod->refptr = alloc_percpu(struct module_ref); + if (!mod->refptr) + return -ENOMEM; - INIT_LIST_HEAD(&mod->modules_which_use_me); - for_each_possible_cpu(cpu) { - per_cpu_ptr(mod->refptr, cpu)->incs = 0; - per_cpu_ptr(mod->refptr, cpu)->decs = 0; - } + INIT_LIST_HEAD(&mod->source_list); + INIT_LIST_HEAD(&mod->target_list); /* Hold reference count during initialization. */ __this_cpu_write(mod->refptr->incs, 1); /* Backwards compatibility macros put refcount during init. */ mod->waiter = current; -} -/* modules using other modules */ -struct module_use -{ - struct list_head list; - struct module *module_which_uses; -}; + return 0; +} /* Does a already use b? */ static int already_uses(struct module *a, struct module *b) { struct module_use *use; - list_for_each_entry(use, &b->modules_which_use_me, list) { - if (use->module_which_uses == a) { + list_for_each_entry(use, &b->source_list, source_list) { + if (use->source == a) { DEBUGP("%s uses %s!\n", a->name, b->name); return 1; } @@ -559,55 +585,70 @@ static int already_uses(struct module *a, struct module *b) return 0; } -/* Module a uses b */ -int use_module(struct module *a, struct module *b) +/* + * Module a uses b + * - we add 'a' as a "source", 'b' as a "target" of module use + * - the module_use is added to the list of 'b' sources (so + * 'b' can walk the list to see who sourced them), and of 'a' + * targets (so 'a' can see what modules it targets). + */ +static int add_module_usage(struct module *a, struct module *b) { struct module_use *use; - int no_warn, err; + + DEBUGP("Allocating new usage for %s.\n", a->name); + use = kmalloc(sizeof(*use), GFP_ATOMIC); + if (!use) { + printk(KERN_WARNING "%s: out of memory loading\n", a->name); + return -ENOMEM; + } + + use->source = a; + use->target = b; + list_add(&use->source_list, &b->source_list); + list_add(&use->target_list, &a->target_list); + return 0; +} + +/* Module a uses b: caller needs module_mutex() */ +int ref_module(struct module *a, struct module *b) +{ + int err; if (b == NULL || already_uses(a, b)) return 0; - /* If we're interrupted or time out, we fail. */ + /* If module isn't available, we fail. */ err = strong_try_module_get(b); if (err) return err; - DEBUGP("Allocating new usage for %s.\n", a->name); - use = kmalloc(sizeof(*use), GFP_ATOMIC); - if (!use) { - printk("%s: out of memory loading\n", a->name); + err = add_module_usage(a, b); + if (err) { module_put(b); - return -ENOMEM; + return err; } - - use->module_which_uses = a; - list_add(&use->list, &b->modules_which_use_me); - no_warn = sysfs_create_link(b->holders_dir, &a->mkobj.kobj, a->name); return 0; } -EXPORT_SYMBOL_GPL(use_module); +EXPORT_SYMBOL_GPL(ref_module); /* Clear the unload stuff of the module. */ static void module_unload_free(struct module *mod) { - struct module *i; - - list_for_each_entry(i, &modules, list) { - struct module_use *use; + struct module_use *use, *tmp; - list_for_each_entry(use, &i->modules_which_use_me, list) { - if (use->module_which_uses == mod) { - DEBUGP("%s unusing %s\n", mod->name, i->name); - module_put(i); - list_del(&use->list); - kfree(use); - sysfs_remove_link(i->holders_dir, mod->name); - /* There can be at most one match. */ - break; - } - } + mutex_lock(&module_mutex); + list_for_each_entry_safe(use, tmp, &mod->target_list, target_list) { + struct module *i = use->target; + DEBUGP("%s unusing %s\n", mod->name, i->name); + module_put(i); + list_del(&use->source_list); + list_del(&use->target_list); + kfree(use); } + mutex_unlock(&module_mutex); + + free_percpu(mod->refptr); } #ifdef CONFIG_MODULE_FORCE_UNLOAD @@ -730,7 +771,7 @@ SYSCALL_DEFINE2(delete_module, const char __user *, name_user, goto out; } - if (!list_empty(&mod->modules_which_use_me)) { + if (!list_empty(&mod->source_list)) { /* Other modules depend on us: get rid of them first. */ ret = -EWOULDBLOCK; goto out; @@ -774,13 +815,13 @@ SYSCALL_DEFINE2(delete_module, const char __user *, name_user, blocking_notifier_call_chain(&module_notify_list, MODULE_STATE_GOING, mod); async_synchronize_full(); - mutex_lock(&module_mutex); + /* Store the name of the last unloaded module for diagnostic purposes */ strlcpy(last_unloaded_module, mod->name, sizeof(last_unloaded_module)); - ddebug_remove_module(mod->name); - free_module(mod); - out: + free_module(mod); + return 0; +out: mutex_unlock(&module_mutex); return ret; } @@ -794,9 +835,9 @@ static inline void print_unload_info(struct seq_file *m, struct module *mod) /* Always include a trailing , so userspace can differentiate between this and the old multi-field proc format. */ - list_for_each_entry(use, &mod->modules_which_use_me, list) { + list_for_each_entry(use, &mod->source_list, source_list) { printed_something = 1; - seq_printf(m, "%s,", use->module_which_uses->name); + seq_printf(m, "%s,", use->source->name); } if (mod->init != NULL && mod->exit == NULL) { @@ -875,14 +916,15 @@ static inline void module_unload_free(struct module *mod) { } -int use_module(struct module *a, struct module *b) +int ref_module(struct module *a, struct module *b) { return strong_try_module_get(b); } -EXPORT_SYMBOL_GPL(use_module); +EXPORT_SYMBOL_GPL(ref_module); -static inline void module_unload_init(struct module *mod) +static inline int module_unload_init(struct module *mod) { + return 0; } #endif /* CONFIG_MODULE_UNLOAD */ @@ -996,6 +1038,8 @@ static inline int check_modstruct_version(Elf_Shdr *sechdrs, { const unsigned long *crc; + /* Since this should be found in kernel (which can't be removed), + * no locking is necessary. */ if (!find_symbol(MODULE_SYMBOL_PREFIX "module_layout", NULL, &crc, true, false)) BUG(); @@ -1038,57 +1082,68 @@ static inline int same_magic(const char *amagic, const char *bmagic, } #endif /* CONFIG_MODVERSIONS */ -/* Resolve a symbol for this module. I.e. if we find one, record usage. - Must be holding module_mutex. */ -static const struct kernel_symbol *resolve_symbol(Elf_Shdr *sechdrs, - unsigned int versindex, +/* Resolve a symbol for this module. I.e. if we find one, record usage. */ +static const struct kernel_symbol *resolve_symbol(struct module *mod, + const struct load_info *info, const char *name, - struct module *mod) + char ownername[]) { struct module *owner; const struct kernel_symbol *sym; const unsigned long *crc; - DEFINE_WAIT(wait); int err; - long timeleft = 30 * HZ; -again: + mutex_lock(&module_mutex); sym = find_symbol(name, &owner, &crc, !(mod->taints & (1 << TAINT_PROPRIETARY_MODULE)), true); if (!sym) - return NULL; + goto unlock; - if (!check_version(sechdrs, versindex, name, mod, crc, owner)) - return NULL; + if (!check_version(info->sechdrs, info->index.vers, name, mod, crc, + owner)) { + sym = ERR_PTR(-EINVAL); + goto getname; + } - prepare_to_wait(&module_wq, &wait, TASK_INTERRUPTIBLE); - err = use_module(mod, owner); - if (likely(!err) || err != -EBUSY || signal_pending(current)) { - finish_wait(&module_wq, &wait); - return err ? NULL : sym; + err = ref_module(mod, owner); + if (err) { + sym = ERR_PTR(err); + goto getname; } - /* Module is still loading. Drop lock and wait. */ +getname: + /* We must make copy under the lock if we failed to get ref. */ + strncpy(ownername, module_name(owner), MODULE_NAME_LEN); +unlock: mutex_unlock(&module_mutex); - timeleft = schedule_timeout(timeleft); - mutex_lock(&module_mutex); - finish_wait(&module_wq, &wait); + return sym; +} - /* Module might be gone entirely, or replaced. Re-lookup. */ - if (timeleft) - goto again; +static const struct kernel_symbol * +resolve_symbol_wait(struct module *mod, + const struct load_info *info, + const char *name) +{ + const struct kernel_symbol *ksym; + char owner[MODULE_NAME_LEN]; - printk(KERN_WARNING "%s: gave up waiting for init of module %s.\n", - mod->name, owner->name); - return NULL; + if (wait_event_interruptible_timeout(module_wq, + !IS_ERR(ksym = resolve_symbol(mod, info, name, owner)) + || PTR_ERR(ksym) != -EBUSY, + 30 * HZ) <= 0) { + printk(KERN_WARNING "%s: gave up waiting for init of module %s.\n", + mod->name, owner); + } + return ksym; } /* * /sys/module/foo/sections stuff * J. Corbet <corbet@lwn.net> */ -#if defined(CONFIG_KALLSYMS) && defined(CONFIG_SYSFS) +#ifdef CONFIG_SYSFS +#ifdef CONFIG_KALLSYMS static inline bool sect_empty(const Elf_Shdr *sect) { return !(sect->sh_flags & SHF_ALLOC) || sect->sh_size == 0; @@ -1125,8 +1180,7 @@ static void free_sect_attrs(struct module_sect_attrs *sect_attrs) kfree(sect_attrs); } -static void add_sect_attrs(struct module *mod, unsigned int nsect, - char *secstrings, Elf_Shdr *sechdrs) +static void add_sect_attrs(struct module *mod, const struct load_info *info) { unsigned int nloaded = 0, i, size[2]; struct module_sect_attrs *sect_attrs; @@ -1134,8 +1188,8 @@ static void add_sect_attrs(struct module *mod, unsigned int nsect, struct attribute **gattr; /* Count loaded sections and allocate structures */ - for (i = 0; i < nsect; i++) - if (!sect_empty(&sechdrs[i])) + for (i = 0; i < info->hdr->e_shnum; i++) + if (!sect_empty(&info->sechdrs[i])) nloaded++; size[0] = ALIGN(sizeof(*sect_attrs) + nloaded * sizeof(sect_attrs->attrs[0]), @@ -1152,11 +1206,12 @@ static void add_sect_attrs(struct module *mod, unsigned int nsect, sect_attrs->nsections = 0; sattr = §_attrs->attrs[0]; gattr = §_attrs->grp.attrs[0]; - for (i = 0; i < nsect; i++) { - if (sect_empty(&sechdrs[i])) + for (i = 0; i < info->hdr->e_shnum; i++) { + Elf_Shdr *sec = &info->sechdrs[i]; + if (sect_empty(sec)) continue; - sattr->address = sechdrs[i].sh_addr; - sattr->name = kstrdup(secstrings + sechdrs[i].sh_name, + sattr->address = sec->sh_addr; + sattr->name = kstrdup(info->secstrings + sec->sh_name, GFP_KERNEL); if (sattr->name == NULL) goto out; @@ -1224,8 +1279,7 @@ static void free_notes_attrs(struct module_notes_attrs *notes_attrs, kfree(notes_attrs); } -static void add_notes_attrs(struct module *mod, unsigned int nsect, - char *secstrings, Elf_Shdr *sechdrs) +static void add_notes_attrs(struct module *mod, const struct load_info *info) { unsigned int notes, loaded, i; struct module_notes_attrs *notes_attrs; @@ -1237,9 +1291,9 @@ static void add_notes_attrs(struct module *mod, unsigned int nsect, /* Count notes sections and allocate structures. */ notes = 0; - for (i = 0; i < nsect; i++) - if (!sect_empty(&sechdrs[i]) && - (sechdrs[i].sh_type == SHT_NOTE)) + for (i = 0; i < info->hdr->e_shnum; i++) + if (!sect_empty(&info->sechdrs[i]) && + (info->sechdrs[i].sh_type == SHT_NOTE)) ++notes; if (notes == 0) @@ -1253,15 +1307,15 @@ static void add_notes_attrs(struct module *mod, unsigned int nsect, notes_attrs->notes = notes; nattr = ¬es_attrs->attrs[0]; - for (loaded = i = 0; i < nsect; ++i) { - if (sect_empty(&sechdrs[i])) + for (loaded = i = 0; i < info->hdr->e_shnum; ++i) { + if (sect_empty(&info->sechdrs[i])) continue; - if (sechdrs[i].sh_type == SHT_NOTE) { + if (info->sechdrs[i].sh_type == SHT_NOTE) { sysfs_bin_attr_init(nattr); nattr->attr.name = mod->sect_attrs->attrs[loaded].name; nattr->attr.mode = S_IRUGO; - nattr->size = sechdrs[i].sh_size; - nattr->private = (void *) sechdrs[i].sh_addr; + nattr->size = info->sechdrs[i].sh_size; + nattr->private = (void *) info->sechdrs[i].sh_addr; nattr->read = module_notes_read; ++nattr; } @@ -1292,8 +1346,8 @@ static void remove_notes_attrs(struct module *mod) #else -static inline void add_sect_attrs(struct module *mod, unsigned int nsect, - char *sectstrings, Elf_Shdr *sechdrs) +static inline void add_sect_attrs(struct module *mod, + const struct load_info *info) { } @@ -1301,18 +1355,44 @@ static inline void remove_sect_attrs(struct module *mod) { } -static inline void add_notes_attrs(struct module *mod, unsigned int nsect, - char *sectstrings, Elf_Shdr *sechdrs) +static inline void add_notes_attrs(struct module *mod, + const struct load_info *info) { } static inline void remove_notes_attrs(struct module *mod) { } +#endif /* CONFIG_KALLSYMS */ + +static void add_usage_links(struct module *mod) +{ +#ifdef CONFIG_MODULE_UNLOAD + struct module_use *use; + int nowarn; + + mutex_lock(&module_mutex); + list_for_each_entry(use, &mod->target_list, target_list) { + nowarn = sysfs_create_link(use->target->holders_dir, + &mod->mkobj.kobj, mod->name); + } + mutex_unlock(&module_mutex); #endif +} -#ifdef CONFIG_SYSFS -int module_add_modinfo_attrs(struct module *mod) +static void del_usage_links(struct module *mod) +{ +#ifdef CONFIG_MODULE_UNLOAD + struct module_use *use; + + mutex_lock(&module_mutex); + list_for_each_entry(use, &mod->target_list, target_list) + sysfs_remove_link(use->target->holders_dir, mod->name); + mutex_unlock(&module_mutex); +#endif +} + +static int module_add_modinfo_attrs(struct module *mod) { struct module_attribute *attr; struct module_attribute *temp_attr; @@ -1338,7 +1418,7 @@ int module_add_modinfo_attrs(struct module *mod) return error; } -void module_remove_modinfo_attrs(struct module *mod) +static void module_remove_modinfo_attrs(struct module *mod) { struct module_attribute *attr; int i; @@ -1354,7 +1434,7 @@ void module_remove_modinfo_attrs(struct module *mod) kfree(mod->modinfo_attrs); } -int mod_sysfs_init(struct module *mod) +static int mod_sysfs_init(struct module *mod) { int err; struct kobject *kobj; @@ -1388,12 +1468,17 @@ out: return err; } -int mod_sysfs_setup(struct module *mod, +static int mod_sysfs_setup(struct module *mod, + const struct load_info *info, struct kernel_param *kparam, unsigned int num_params) { int err; + err = mod_sysfs_init(mod); + if (err) + goto out; + mod->holders_dir = kobject_create_and_add("holders", &mod->mkobj.kobj); if (!mod->holders_dir) { err = -ENOMEM; @@ -1408,6 +1493,10 @@ int mod_sysfs_setup(struct module *mod, if (err) goto out_unreg_param; + add_usage_links(mod); + add_sect_attrs(mod, info); + add_notes_attrs(mod, info); + kobject_uevent(&mod->mkobj.kobj, KOBJ_ADD); return 0; @@ -1417,24 +1506,44 @@ out_unreg_holders: kobject_put(mod->holders_dir); out_unreg: kobject_put(&mod->mkobj.kobj); +out: return err; } static void mod_sysfs_fini(struct module *mod) { + remove_notes_attrs(mod); + remove_sect_attrs(mod); kobject_put(&mod->mkobj.kobj); } -#else /* CONFIG_SYSFS */ +#else /* !CONFIG_SYSFS */ + +static int mod_sysfs_setup(struct module *mod, + const struct load_info *info, + struct kernel_param *kparam, + unsigned int num_params) +{ + return 0; +} static void mod_sysfs_fini(struct module *mod) { } +static void module_remove_modinfo_attrs(struct module *mod) +{ +} + +static void del_usage_links(struct module *mod) +{ +} + #endif /* CONFIG_SYSFS */ -static void mod_kobject_remove(struct module *mod) +static void mod_sysfs_teardown(struct module *mod) { + del_usage_links(mod); module_remove_modinfo_attrs(mod); module_param_sysfs_remove(mod); kobject_put(mod->mkobj.drivers_dir); @@ -1450,19 +1559,132 @@ static int __unlink_module(void *_mod) { struct module *mod = _mod; list_del(&mod->list); + module_bug_cleanup(mod); return 0; } -/* Free a module, remove from lists, etc (must hold module_mutex). */ +#ifdef CONFIG_DEBUG_SET_MODULE_RONX +/* + * LKM RO/NX protection: protect module's text/ro-data + * from modification and any data from execution. + */ +void set_page_attributes(void *start, void *end, int (*set)(unsigned long start, int num_pages)) +{ + unsigned long begin_pfn = PFN_DOWN((unsigned long)start); + unsigned long end_pfn = PFN_DOWN((unsigned long)end); + + if (end_pfn > begin_pfn) + set(begin_pfn << PAGE_SHIFT, end_pfn - begin_pfn); +} + +static void set_section_ro_nx(void *base, + unsigned long text_size, + unsigned long ro_size, + unsigned long total_size) +{ + /* begin and end PFNs of the current subsection */ + unsigned long begin_pfn; + unsigned long end_pfn; + + /* + * Set RO for module text and RO-data: + * - Always protect first page. + * - Do not protect last partial page. + */ + if (ro_size > 0) + set_page_attributes(base, base + ro_size, set_memory_ro); + + /* + * Set NX permissions for module data: + * - Do not protect first partial page. + * - Always protect last page. + */ + if (total_size > text_size) { + begin_pfn = PFN_UP((unsigned long)base + text_size); + end_pfn = PFN_UP((unsigned long)base + total_size); + if (end_pfn > begin_pfn) + set_memory_nx(begin_pfn << PAGE_SHIFT, end_pfn - begin_pfn); + } +} + +/* Setting memory back to RW+NX before releasing it */ +void unset_section_ro_nx(struct module *mod, void *module_region) +{ + unsigned long total_pages; + + if (mod->module_core == module_region) { + /* Set core as NX+RW */ + total_pages = MOD_NUMBER_OF_PAGES(mod->module_core, mod->core_size); + set_memory_nx((unsigned long)mod->module_core, total_pages); + set_memory_rw((unsigned long)mod->module_core, total_pages); + + } else if (mod->module_init == module_region) { + /* Set init as NX+RW */ + total_pages = MOD_NUMBER_OF_PAGES(mod->module_init, mod->init_size); + set_memory_nx((unsigned long)mod->module_init, total_pages); + set_memory_rw((unsigned long)mod->module_init, total_pages); + } +} + +/* Iterate through all modules and set each module's text as RW */ +void set_all_modules_text_rw() +{ + struct module *mod; + + mutex_lock(&module_mutex); + list_for_each_entry_rcu(mod, &modules, list) { + if ((mod->module_core) && (mod->core_text_size)) { + set_page_attributes(mod->module_core, + mod->module_core + mod->core_text_size, + set_memory_rw); + } + if ((mod->module_init) && (mod->init_text_size)) { + set_page_attributes(mod->module_init, + mod->module_init + mod->init_text_size, + set_memory_rw); + } + } + mutex_unlock(&module_mutex); +} + +/* Iterate through all modules and set each module's text as RO */ +void set_all_modules_text_ro() +{ + struct module *mod; + + mutex_lock(&module_mutex); + list_for_each_entry_rcu(mod, &modules, list) { + if ((mod->module_core) && (mod->core_text_size)) { + set_page_attributes(mod->module_core, + mod->module_core + mod->core_text_size, + set_memory_ro); + } + if ((mod->module_init) && (mod->init_text_size)) { + set_page_attributes(mod->module_init, + mod->module_init + mod->init_text_size, + set_memory_ro); + } + } + mutex_unlock(&module_mutex); +} +#else +static inline void set_section_ro_nx(void *base, unsigned long text_size, unsigned long ro_size, unsigned long total_size) { } +static inline void unset_section_ro_nx(struct module *mod, void *module_region) { } +#endif + +/* Free a module, remove from lists, etc. */ static void free_module(struct module *mod) { trace_module_free(mod); /* Delete from various lists */ + mutex_lock(&module_mutex); stop_machine(__unlink_module, mod, NULL); - remove_notes_attrs(mod); - remove_sect_attrs(mod); - mod_kobject_remove(mod); + mutex_unlock(&module_mutex); + mod_sysfs_teardown(mod); + + /* Remove dynamic debug info */ + ddebug_remove_module(mod->name); /* Arch-specific cleanup. */ module_arch_cleanup(mod); @@ -1474,17 +1696,16 @@ static void free_module(struct module *mod) destroy_params(mod->kp, mod->num_kp); /* This may be NULL, but that's OK */ + unset_section_ro_nx(mod, mod->module_init); module_free(mod, mod->module_init); kfree(mod->args); percpu_modfree(mod); -#if defined(CONFIG_MODULE_UNLOAD) - if (mod->refptr) - free_percpu(mod->refptr); -#endif + /* Free lock-classes: */ lockdep_free_key_range(mod->module_core, mod->core_size); /* Finally, free the core (containing the module structure) */ + unset_section_ro_nx(mod, mod->module_core); module_free(mod, mod->module_core); #ifdef CONFIG_MPU @@ -1510,6 +1731,8 @@ EXPORT_SYMBOL_GPL(__symbol_get); /* * Ensure that an exported symbol [global namespace] does not already exist * in the kernel or in some other module's exported symbol table. + * + * You must hold the module_mutex. */ static int verify_export_symbols(struct module *mod) { @@ -1544,25 +1767,23 @@ static int verify_export_symbols(struct module *mod) } /* Change all symbols so that st_value encodes the pointer directly. */ -static int simplify_symbols(Elf_Shdr *sechdrs, - unsigned int symindex, - const char *strtab, - unsigned int versindex, - unsigned int pcpuindex, - struct module *mod) -{ - Elf_Sym *sym = (void *)sechdrs[symindex].sh_addr; +static int simplify_symbols(struct module *mod, const struct load_info *info) +{ + Elf_Shdr *symsec = &info->sechdrs[info->index.sym]; + Elf_Sym *sym = (void *)symsec->sh_addr; unsigned long secbase; - unsigned int i, n = sechdrs[symindex].sh_size / sizeof(Elf_Sym); + unsigned int i; int ret = 0; const struct kernel_symbol *ksym; - for (i = 1; i < n; i++) { + for (i = 1; i < symsec->sh_size / sizeof(Elf_Sym); i++) { + const char *name = info->strtab + sym[i].st_name; + switch (sym[i].st_shndx) { case SHN_COMMON: /* We compiled with -fno-common. These are not supposed to happen. */ - DEBUGP("Common symbol: %s\n", strtab + sym[i].st_name); + DEBUGP("Common symbol: %s\n", name); printk("%s: please compile with -fno-common\n", mod->name); ret = -ENOEXEC; @@ -1575,29 +1796,28 @@ static int simplify_symbols(Elf_Shdr *sechdrs, break; case SHN_UNDEF: - ksym = resolve_symbol(sechdrs, versindex, - strtab + sym[i].st_name, mod); + ksym = resolve_symbol_wait(mod, info, name); /* Ok if resolved. */ - if (ksym) { + if (ksym && !IS_ERR(ksym)) { sym[i].st_value = ksym->value; break; } /* Ok if weak. */ - if (ELF_ST_BIND(sym[i].st_info) == STB_WEAK) + if (!ksym && ELF_ST_BIND(sym[i].st_info) == STB_WEAK) break; - printk(KERN_WARNING "%s: Unknown symbol %s\n", - mod->name, strtab + sym[i].st_name); - ret = -ENOENT; + printk(KERN_WARNING "%s: Unknown symbol %s (err %li)\n", + mod->name, name, PTR_ERR(ksym)); + ret = PTR_ERR(ksym) ?: -ENOENT; break; default: /* Divert to percpu allocation if a percpu var. */ - if (sym[i].st_shndx == pcpuindex) + if (sym[i].st_shndx == info->index.pcpu) secbase = (unsigned long)mod_percpu(mod); else - secbase = sechdrs[sym[i].st_shndx].sh_addr; + secbase = info->sechdrs[sym[i].st_shndx].sh_addr; sym[i].st_value += secbase; break; } @@ -1606,6 +1826,35 @@ static int simplify_symbols(Elf_Shdr *sechdrs, return ret; } +static int apply_relocations(struct module *mod, const struct load_info *info) +{ + unsigned int i; + int err = 0; + + /* Now do relocations. */ + for (i = 1; i < info->hdr->e_shnum; i++) { + unsigned int infosec = info->sechdrs[i].sh_info; + + /* Not a valid relocation section? */ + if (infosec >= info->hdr->e_shnum) + continue; + + /* Don't bother with non-allocated sections */ + if (!(info->sechdrs[infosec].sh_flags & SHF_ALLOC)) + continue; + + if (info->sechdrs[i].sh_type == SHT_REL) + err = apply_relocate(info->sechdrs, info->strtab, + info->index.sym, i, mod); + else if (info->sechdrs[i].sh_type == SHT_RELA) + err = apply_relocate_add(info->sechdrs, info->strtab, + info->index.sym, i, mod); + if (err < 0) + break; + } + return err; +} + /* Additional bytes needed by arch in front of individual sections */ unsigned int __weak arch_mod_section_prepend(struct module *mod, unsigned int section) @@ -1630,10 +1879,7 @@ static long get_offset(struct module *mod, unsigned int *size, might -- code, read-only data, read-write data, small data. Tally sizes, and place the offsets into sh_entsize fields: high bit means it belongs in init. */ -static void layout_sections(struct module *mod, - const Elf_Ehdr *hdr, - Elf_Shdr *sechdrs, - const char *secstrings) +static void layout_sections(struct module *mod, struct load_info *info) { static unsigned long const masks[][2] = { /* NOTE: all executable code must be the first section @@ -1646,42 +1892,66 @@ static void layout_sections(struct module *mod, }; unsigned int m, i; - for (i = 0; i < hdr->e_shnum; i++) - sechdrs[i].sh_entsize = ~0UL; + for (i = 0; i < info->hdr->e_shnum; i++) + info->sechdrs[i].sh_entsize = ~0UL; DEBUGP("Core section allocation order:\n"); for (m = 0; m < ARRAY_SIZE(masks); ++m) { - for (i = 0; i < hdr->e_shnum; ++i) { - Elf_Shdr *s = &sechdrs[i]; + for (i = 0; i < info->hdr->e_shnum; ++i) { + Elf_Shdr *s = &info->sechdrs[i]; + const char *sname = info->secstrings + s->sh_name; if ((s->sh_flags & masks[m][0]) != masks[m][0] || (s->sh_flags & masks[m][1]) || s->sh_entsize != ~0UL - || strstarts(secstrings + s->sh_name, ".init")) + || strstarts(sname, ".init")) continue; s->sh_entsize = get_offset(mod, &mod->core_size, s, i); - DEBUGP("\t%s\n", secstrings + s->sh_name); + DEBUGP("\t%s\n", name); } - if (m == 0) + switch (m) { + case 0: /* executable */ + mod->core_size = debug_align(mod->core_size); mod->core_text_size = mod->core_size; + break; + case 1: /* RO: text and ro-data */ + mod->core_size = debug_align(mod->core_size); + mod->core_ro_size = mod->core_size; + break; + case 3: /* whole core */ + mod->core_size = debug_align(mod->core_size); + break; + } } DEBUGP("Init section allocation order:\n"); for (m = 0; m < ARRAY_SIZE(masks); ++m) { - for (i = 0; i < hdr->e_shnum; ++i) { - Elf_Shdr *s = &sechdrs[i]; + for (i = 0; i < info->hdr->e_shnum; ++i) { + Elf_Shdr *s = &info->sechdrs[i]; + const char *sname = info->secstrings + s->sh_name; if ((s->sh_flags & masks[m][0]) != masks[m][0] || (s->sh_flags & masks[m][1]) || s->sh_entsize != ~0UL - || !strstarts(secstrings + s->sh_name, ".init")) + || !strstarts(sname, ".init")) continue; s->sh_entsize = (get_offset(mod, &mod->init_size, s, i) | INIT_OFFSET_MASK); - DEBUGP("\t%s\n", secstrings + s->sh_name); + DEBUGP("\t%s\n", sname); } - if (m == 0) + switch (m) { + case 0: /* executable */ + mod->init_size = debug_align(mod->init_size); mod->init_text_size = mod->init_size; + break; + case 1: /* RO: text and ro-data */ + mod->init_size = debug_align(mod->init_size); + mod->init_ro_size = mod->init_size; + break; + case 3: /* whole init */ + mod->init_size = debug_align(mod->init_size); + break; + } } } @@ -1717,33 +1987,28 @@ static char *next_string(char *string, unsigned long *secsize) return string; } -static char *get_modinfo(Elf_Shdr *sechdrs, - unsigned int info, - const char *tag) +static char *get_modinfo(struct load_info *info, const char *tag) { char *p; unsigned int taglen = strlen(tag); - unsigned long size = sechdrs[info].sh_size; + Elf_Shdr *infosec = &info->sechdrs[info->index.info]; + unsigned long size = infosec->sh_size; - for (p = (char *)sechdrs[info].sh_addr; p; p = next_string(p, &size)) { + for (p = (char *)infosec->sh_addr; p; p = next_string(p, &size)) { if (strncmp(p, tag, taglen) == 0 && p[taglen] == '=') return p + taglen + 1; } return NULL; } -static void setup_modinfo(struct module *mod, Elf_Shdr *sechdrs, - unsigned int infoindex) +static void setup_modinfo(struct module *mod, struct load_info *info) { struct module_attribute *attr; int i; for (i = 0; (attr = modinfo_attrs[i]); i++) { if (attr->setup) - attr->setup(mod, - get_modinfo(sechdrs, - infoindex, - attr->attr.name)); + attr->setup(mod, get_modinfo(info, attr->attr.name)); } } @@ -1784,11 +2049,10 @@ static int is_exported(const char *name, unsigned long value, } /* As per nm */ -static char elf_type(const Elf_Sym *sym, - Elf_Shdr *sechdrs, - const char *secstrings, - struct module *mod) +static char elf_type(const Elf_Sym *sym, const struct load_info *info) { + const Elf_Shdr *sechdrs = info->sechdrs; + if (ELF_ST_BIND(sym->st_info) == STB_WEAK) { if (ELF_ST_TYPE(sym->st_info) == STT_OBJECT) return 'v'; @@ -1818,8 +2082,10 @@ static char elf_type(const Elf_Sym *sym, else return 'b'; } - if (strstarts(secstrings + sechdrs[sym->st_shndx].sh_name, ".debug")) + if (strstarts(info->secstrings + sechdrs[sym->st_shndx].sh_name, + ".debug")) { return 'n'; + } return '?'; } @@ -1844,127 +2110,96 @@ static bool is_core_symbol(const Elf_Sym *src, const Elf_Shdr *sechdrs, 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) +static void layout_symtab(struct module *mod, struct load_info *info) { - unsigned long symoffs; - Elf_Shdr *symsect = sechdrs + symindex; - Elf_Shdr *strsect = sechdrs + strindex; + Elf_Shdr *symsect = info->sechdrs + info->index.sym; + Elf_Shdr *strsect = info->sechdrs + info->index.str; 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); + info->index.sym) | INIT_OFFSET_MASK; + DEBUGP("\t%s\n", info->secstrings + symsect->sh_name); - src = (void *)hdr + symsect->sh_offset; + src = (void *)info->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)) { + if (is_core_symbol(src, info->sechdrs, info->hdr->e_shnum)) { unsigned int j = src->st_name; - while(!__test_and_set_bit(j, strmap) && strtab[j]) + while (!__test_and_set_bit(j, info->strmap) + && info->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); + info->symoffs = ALIGN(mod->core_size, symsect->sh_addralign ?: 1); + mod->core_size = info->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); + info->index.str) | INIT_OFFSET_MASK; + DEBUGP("\t%s\n", info->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; + info->stroffs = mod->core_size; + __set_bit(0, info->strmap); + mod->core_size += bitmap_weight(info->strmap, strsect->sh_size); } -static void add_kallsyms(struct module *mod, - Elf_Shdr *sechdrs, - unsigned int shnum, - unsigned int symindex, - unsigned int strindex, - unsigned long symoffs, - unsigned long stroffs, - const char *secstrings, - unsigned long *strmap) +static void add_kallsyms(struct module *mod, const struct load_info *info) { unsigned int i, ndst; const Elf_Sym *src; Elf_Sym *dst; char *s; + Elf_Shdr *symsec = &info->sechdrs[info->index.sym]; - mod->symtab = (void *)sechdrs[symindex].sh_addr; - mod->num_symtab = sechdrs[symindex].sh_size / sizeof(Elf_Sym); - mod->strtab = (void *)sechdrs[strindex].sh_addr; + mod->symtab = (void *)symsec->sh_addr; + mod->num_symtab = symsec->sh_size / sizeof(Elf_Sym); + /* Make sure we get permanent strtab: don't use info->strtab. */ + mod->strtab = (void *)info->sechdrs[info->index.str].sh_addr; /* Set types up while we still have access to sections. */ for (i = 0; i < mod->num_symtab; i++) - mod->symtab[i].st_info - = elf_type(&mod->symtab[i], sechdrs, secstrings, mod); + mod->symtab[i].st_info = elf_type(&mod->symtab[i], info); - mod->core_symtab = dst = mod->module_core + symoffs; + mod->core_symtab = dst = mod->module_core + info->symoffs; src = mod->symtab; *dst = *src; for (ndst = i = 1; i < mod->num_symtab; ++i, ++src) { - if (!is_core_symbol(src, sechdrs, shnum)) + if (!is_core_symbol(src, info->sechdrs, info->hdr->e_shnum)) continue; dst[ndst] = *src; - dst[ndst].st_name = bitmap_weight(strmap, dst[ndst].st_name); + dst[ndst].st_name = bitmap_weight(info->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)) + mod->core_strtab = s = mod->module_core + info->stroffs; + for (*s = 0, i = 1; i < info->sechdrs[info->index.str].sh_size; ++i) + if (test_bit(i, info->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) +static inline void layout_symtab(struct module *mod, struct load_info *info) { - return 0; } -static inline void add_kallsyms(struct module *mod, - Elf_Shdr *sechdrs, - unsigned int shnum, - unsigned int symindex, - unsigned int strindex, - unsigned long symoffs, - unsigned long stroffs, - const char *secstrings, - const unsigned long *strmap) +static void add_kallsyms(struct module *mod, const struct load_info *info) { } #endif /* CONFIG_KALLSYMS */ static void dynamic_debug_setup(struct _ddebug *debug, unsigned int num) { + if (!debug) + return; #ifdef CONFIG_DYNAMIC_DEBUG if (ddebug_add_module(debug, num, debug->modname)) printk(KERN_ERR "dynamic debug error adding module: %s\n", @@ -1972,77 +2207,70 @@ static void dynamic_debug_setup(struct _ddebug *debug, unsigned int num) #endif } +static void dynamic_debug_remove(struct _ddebug *debug) +{ + if (debug) + ddebug_remove_module(debug->modname); +} + static void *module_alloc_update_bounds(unsigned long size) { void *ret = module_alloc(size); if (ret) { + mutex_lock(&module_mutex); /* Update module bounds. */ if ((unsigned long)ret < module_addr_min) module_addr_min = (unsigned long)ret; if ((unsigned long)ret + size > module_addr_max) module_addr_max = (unsigned long)ret + size; + mutex_unlock(&module_mutex); } return ret; } #ifdef CONFIG_DEBUG_KMEMLEAK -static void kmemleak_load_module(struct module *mod, Elf_Ehdr *hdr, - Elf_Shdr *sechdrs, char *secstrings) +static void kmemleak_load_module(const struct module *mod, + const struct load_info *info) { unsigned int i; /* only scan the sections containing data */ kmemleak_scan_area(mod, sizeof(struct module), GFP_KERNEL); - for (i = 1; i < hdr->e_shnum; i++) { - if (!(sechdrs[i].sh_flags & SHF_ALLOC)) + for (i = 1; i < info->hdr->e_shnum; i++) { + const char *name = info->secstrings + info->sechdrs[i].sh_name; + if (!(info->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) + if (!strstarts(name, ".data") && !strstarts(name, ".bss")) continue; - kmemleak_scan_area((void *)sechdrs[i].sh_addr, - sechdrs[i].sh_size, GFP_KERNEL); + kmemleak_scan_area((void *)info->sechdrs[i].sh_addr, + info->sechdrs[i].sh_size, GFP_KERNEL); } } #else -static inline void kmemleak_load_module(struct module *mod, Elf_Ehdr *hdr, - Elf_Shdr *sechdrs, char *secstrings) +static inline void kmemleak_load_module(const struct module *mod, + const struct load_info *info) { } #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, - unsigned long len, - const char __user *uargs) +/* Sets info->hdr and info->len. */ +static int copy_and_check(struct load_info *info, + const void __user *umod, unsigned long len, + const char __user *uargs) { + int err; Elf_Ehdr *hdr; - Elf_Shdr *sechdrs; - char *secstrings, *args, *modmagic, *strtab = NULL; - char *staging; - unsigned int i; - unsigned int symindex = 0; - unsigned int strindex = 0; - unsigned int modindex, versindex, infoindex, pcpuindex; - struct module *mod; - long err = 0; - void *ptr = NULL; /* Stops spurious gcc warning */ - unsigned long symoffs, stroffs, *strmap; - mm_segment_t old_fs; - - DEBUGP("load_module: umod=%p, len=%lu, uargs=%p\n", - umod, len, uargs); if (len < sizeof(*hdr)) - return ERR_PTR(-ENOEXEC); + return -ENOEXEC; /* Suck in entire file: we'll want most of it. */ /* vmalloc barfs on "unusual" numbers. Check here */ if (len > 64 * 1024 * 1024 || (hdr = vmalloc(len)) == NULL) - return ERR_PTR(-ENOMEM); + return -ENOMEM; if (copy_from_user(hdr, umod, len) != 0) { err = -EFAULT; @@ -2050,138 +2278,242 @@ static noinline struct module *load_module(void __user *umod, } /* Sanity checks against insmoding binaries or wrong arch, - weird elf version */ + weird elf version */ if (memcmp(hdr->e_ident, ELFMAG, SELFMAG) != 0 || hdr->e_type != ET_REL || !elf_check_arch(hdr) - || hdr->e_shentsize != sizeof(*sechdrs)) { + || hdr->e_shentsize != sizeof(Elf_Shdr)) { err = -ENOEXEC; goto free_hdr; } - if (len < hdr->e_shoff + hdr->e_shnum * sizeof(Elf_Shdr)) - goto truncated; + if (len < hdr->e_shoff + hdr->e_shnum * sizeof(Elf_Shdr)) { + err = -ENOEXEC; + goto free_hdr; + } - /* Convenience variables */ - sechdrs = (void *)hdr + hdr->e_shoff; - secstrings = (void *)hdr + sechdrs[hdr->e_shstrndx].sh_offset; - sechdrs[0].sh_addr = 0; + info->hdr = hdr; + info->len = len; + return 0; - for (i = 1; i < hdr->e_shnum; i++) { - if (sechdrs[i].sh_type != SHT_NOBITS - && len < sechdrs[i].sh_offset + sechdrs[i].sh_size) - goto truncated; +free_hdr: + vfree(hdr); + return err; +} + +static void free_copy(struct load_info *info) +{ + vfree(info->hdr); +} + +static int rewrite_section_headers(struct load_info *info) +{ + unsigned int i; + + /* This should always be true, but let's be sure. */ + info->sechdrs[0].sh_addr = 0; + + for (i = 1; i < info->hdr->e_shnum; i++) { + Elf_Shdr *shdr = &info->sechdrs[i]; + if (shdr->sh_type != SHT_NOBITS + && info->len < shdr->sh_offset + shdr->sh_size) { + printk(KERN_ERR "Module len %lu truncated\n", + info->len); + return -ENOEXEC; + } /* Mark all sections sh_addr with their address in the temporary image. */ - sechdrs[i].sh_addr = (size_t)hdr + sechdrs[i].sh_offset; + shdr->sh_addr = (size_t)info->hdr + shdr->sh_offset; - /* Internal symbols and strings. */ - if (sechdrs[i].sh_type == SHT_SYMTAB) { - symindex = i; - strindex = sechdrs[i].sh_link; - strtab = (char *)hdr + sechdrs[strindex].sh_offset; - } #ifndef CONFIG_MODULE_UNLOAD /* Don't load .exit sections */ - if (strstarts(secstrings+sechdrs[i].sh_name, ".exit")) - sechdrs[i].sh_flags &= ~(unsigned long)SHF_ALLOC; + if (strstarts(info->secstrings+shdr->sh_name, ".exit")) + shdr->sh_flags &= ~(unsigned long)SHF_ALLOC; #endif } - modindex = find_sec(hdr, sechdrs, secstrings, - ".gnu.linkonce.this_module"); - if (!modindex) { + /* Track but don't keep modinfo and version sections. */ + info->index.vers = find_sec(info, "__versions"); + info->index.info = find_sec(info, ".modinfo"); + info->sechdrs[info->index.info].sh_flags &= ~(unsigned long)SHF_ALLOC; + info->sechdrs[info->index.vers].sh_flags &= ~(unsigned long)SHF_ALLOC; + return 0; +} + +/* + * Set up our basic convenience variables (pointers to section headers, + * search for module section index etc), and do some basic section + * verification. + * + * Return the temporary module pointer (we'll replace it with the final + * one when we move the module sections around). + */ +static struct module *setup_load_info(struct load_info *info) +{ + unsigned int i; + int err; + struct module *mod; + + /* Set up the convenience variables */ + info->sechdrs = (void *)info->hdr + info->hdr->e_shoff; + info->secstrings = (void *)info->hdr + + info->sechdrs[info->hdr->e_shstrndx].sh_offset; + + err = rewrite_section_headers(info); + if (err) + return ERR_PTR(err); + + /* Find internal symbols and strings. */ + for (i = 1; i < info->hdr->e_shnum; i++) { + if (info->sechdrs[i].sh_type == SHT_SYMTAB) { + info->index.sym = i; + info->index.str = info->sechdrs[i].sh_link; + info->strtab = (char *)info->hdr + + info->sechdrs[info->index.str].sh_offset; + break; + } + } + + info->index.mod = find_sec(info, ".gnu.linkonce.this_module"); + if (!info->index.mod) { printk(KERN_WARNING "No module found in object\n"); - err = -ENOEXEC; - goto free_hdr; + return ERR_PTR(-ENOEXEC); } /* This is temporary: point mod into copy of data. */ - mod = (void *)sechdrs[modindex].sh_addr; + mod = (void *)info->sechdrs[info->index.mod].sh_addr; - if (symindex == 0) { + if (info->index.sym == 0) { printk(KERN_WARNING "%s: module has no symbols (stripped?)\n", mod->name); - err = -ENOEXEC; - goto free_hdr; + return ERR_PTR(-ENOEXEC); } - versindex = find_sec(hdr, sechdrs, secstrings, "__versions"); - infoindex = find_sec(hdr, sechdrs, secstrings, ".modinfo"); - pcpuindex = find_pcpusec(hdr, sechdrs, secstrings); - - /* Don't keep modinfo and version sections. */ - sechdrs[infoindex].sh_flags &= ~(unsigned long)SHF_ALLOC; - sechdrs[versindex].sh_flags &= ~(unsigned long)SHF_ALLOC; + info->index.pcpu = find_pcpusec(info); /* Check module struct version now, before we try to use module. */ - if (!check_modstruct_version(sechdrs, versindex, mod)) { - err = -ENOEXEC; - goto free_hdr; - } + if (!check_modstruct_version(info->sechdrs, info->index.vers, mod)) + return ERR_PTR(-ENOEXEC); + + return mod; +} + +static int check_modinfo(struct module *mod, struct load_info *info) +{ + const char *modmagic = get_modinfo(info, "vermagic"); + int err; - modmagic = get_modinfo(sechdrs, infoindex, "vermagic"); /* This is allowed: modprobe --force will invalidate it. */ if (!modmagic) { err = try_to_force_load(mod, "bad vermagic"); if (err) - goto free_hdr; - } else if (!same_magic(modmagic, vermagic, versindex)) { + return err; + } else if (!same_magic(modmagic, vermagic, info->index.vers)) { printk(KERN_ERR "%s: version magic '%s' should be '%s'\n", mod->name, modmagic, vermagic); - err = -ENOEXEC; - goto free_hdr; + return -ENOEXEC; } - staging = get_modinfo(sechdrs, infoindex, "staging"); - if (staging) { + if (get_modinfo(info, "staging")) { add_taint_module(mod, TAINT_CRAP); printk(KERN_WARNING "%s: module is from the staging directory," " the quality is unknown, you have been warned.\n", mod->name); } - /* Now copy in args */ - args = strndup_user(uargs, ~0UL >> 1); - if (IS_ERR(args)) { - err = PTR_ERR(args); - goto free_hdr; - } + /* Set up license info based on the info section */ + set_license(mod, get_modinfo(info, "license")); - strmap = kzalloc(BITS_TO_LONGS(sechdrs[strindex].sh_size) - * sizeof(long), GFP_KERNEL); - if (!strmap) { - err = -ENOMEM; - goto free_mod; - } + return 0; +} - if (find_module(mod->name)) { - err = -EEXIST; - goto free_mod; - } +static void find_module_sections(struct module *mod, struct load_info *info) +{ + mod->kp = section_objs(info, "__param", + sizeof(*mod->kp), &mod->num_kp); + mod->syms = section_objs(info, "__ksymtab", + sizeof(*mod->syms), &mod->num_syms); + mod->crcs = section_addr(info, "__kcrctab"); + mod->gpl_syms = section_objs(info, "__ksymtab_gpl", + sizeof(*mod->gpl_syms), + &mod->num_gpl_syms); + mod->gpl_crcs = section_addr(info, "__kcrctab_gpl"); + mod->gpl_future_syms = section_objs(info, + "__ksymtab_gpl_future", + sizeof(*mod->gpl_future_syms), + &mod->num_gpl_future_syms); + mod->gpl_future_crcs = section_addr(info, "__kcrctab_gpl_future"); - mod->state = MODULE_STATE_COMING; +#ifdef CONFIG_UNUSED_SYMBOLS + mod->unused_syms = section_objs(info, "__ksymtab_unused", + sizeof(*mod->unused_syms), + &mod->num_unused_syms); + mod->unused_crcs = section_addr(info, "__kcrctab_unused"); + mod->unused_gpl_syms = section_objs(info, "__ksymtab_unused_gpl", + sizeof(*mod->unused_gpl_syms), + &mod->num_unused_gpl_syms); + mod->unused_gpl_crcs = section_addr(info, "__kcrctab_unused_gpl"); +#endif +#ifdef CONFIG_CONSTRUCTORS + mod->ctors = section_objs(info, ".ctors", + sizeof(*mod->ctors), &mod->num_ctors); +#endif - /* Allow arches to frob section contents and sizes. */ - err = module_frob_arch_sections(hdr, sechdrs, secstrings, mod); - if (err < 0) - goto free_mod; +#ifdef CONFIG_TRACEPOINTS + mod->tracepoints = section_objs(info, "__tracepoints", + sizeof(*mod->tracepoints), + &mod->num_tracepoints); +#endif +#ifdef HAVE_JUMP_LABEL + mod->jump_entries = section_objs(info, "__jump_table", + sizeof(*mod->jump_entries), + &mod->num_jump_entries); +#endif +#ifdef CONFIG_EVENT_TRACING + mod->trace_events = section_objs(info, "_ftrace_events", + sizeof(*mod->trace_events), + &mod->num_trace_events); + /* + * This section contains pointers to allocated objects in the trace + * code and not scanning it leads to false positives. + */ + kmemleak_scan_area(mod->trace_events, sizeof(*mod->trace_events) * + mod->num_trace_events, GFP_KERNEL); +#endif +#ifdef CONFIG_TRACING + mod->trace_bprintk_fmt_start = section_objs(info, "__trace_printk_fmt", + sizeof(*mod->trace_bprintk_fmt_start), + &mod->num_trace_bprintk_fmt); + /* + * This section contains pointers to allocated objects in the trace + * code and not scanning it leads to false positives. + */ + kmemleak_scan_area(mod->trace_bprintk_fmt_start, + sizeof(*mod->trace_bprintk_fmt_start) * + mod->num_trace_bprintk_fmt, GFP_KERNEL); +#endif +#ifdef CONFIG_FTRACE_MCOUNT_RECORD + /* sechdrs[0].sh_size is always zero */ + mod->ftrace_callsites = section_objs(info, "__mcount_loc", + sizeof(*mod->ftrace_callsites), + &mod->num_ftrace_callsites); +#endif - if (pcpuindex) { - /* We have a special allocation for this section. */ - err = percpu_modalloc(mod, sechdrs[pcpuindex].sh_size, - sechdrs[pcpuindex].sh_addralign); - if (err) - goto free_mod; - sechdrs[pcpuindex].sh_flags &= ~(unsigned long)SHF_ALLOC; - } + mod->extable = section_objs(info, "__ex_table", + sizeof(*mod->extable), &mod->num_exentries); - /* Determine total sizes, and put offsets in sh_entsize. For now - 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); + if (section_addr(info, "__obsparm")) + printk(KERN_WARNING "%s: Ignoring obsolete parameters\n", + mod->name); + + info->debug = section_objs(info, "__verbose", + sizeof(*info->debug), &info->num_debug); +} + +static int move_module(struct module *mod, struct load_info *info) +{ + int i; + void *ptr; /* Do the allocs. */ ptr = module_alloc_update_bounds(mod->core_size); @@ -2191,10 +2523,9 @@ static noinline struct module *load_module(void __user *umod, * leak. */ kmemleak_not_leak(ptr); - if (!ptr) { - err = -ENOMEM; - goto free_percpu; - } + if (!ptr) + return -ENOMEM; + memset(ptr, 0, mod->core_size); mod->module_core = ptr; @@ -2207,55 +2538,40 @@ static noinline struct module *load_module(void __user *umod, */ kmemleak_ignore(ptr); if (!ptr && mod->init_size) { - err = -ENOMEM; - goto free_core; + module_free(mod, mod->module_core); + return -ENOMEM; } memset(ptr, 0, mod->init_size); mod->module_init = ptr; /* Transfer each section which specifies SHF_ALLOC */ DEBUGP("final section addresses:\n"); - for (i = 0; i < hdr->e_shnum; i++) { + for (i = 0; i < info->hdr->e_shnum; i++) { void *dest; + Elf_Shdr *shdr = &info->sechdrs[i]; - if (!(sechdrs[i].sh_flags & SHF_ALLOC)) + if (!(shdr->sh_flags & SHF_ALLOC)) continue; - if (sechdrs[i].sh_entsize & INIT_OFFSET_MASK) + if (shdr->sh_entsize & INIT_OFFSET_MASK) dest = mod->module_init - + (sechdrs[i].sh_entsize & ~INIT_OFFSET_MASK); + + (shdr->sh_entsize & ~INIT_OFFSET_MASK); else - dest = mod->module_core + sechdrs[i].sh_entsize; + dest = mod->module_core + shdr->sh_entsize; - if (sechdrs[i].sh_type != SHT_NOBITS) - memcpy(dest, (void *)sechdrs[i].sh_addr, - sechdrs[i].sh_size); + if (shdr->sh_type != SHT_NOBITS) + memcpy(dest, (void *)shdr->sh_addr, shdr->sh_size); /* Update sh_addr to point to copy in image. */ - sechdrs[i].sh_addr = (unsigned long)dest; - DEBUGP("\t0x%lx %s\n", sechdrs[i].sh_addr, secstrings + sechdrs[i].sh_name); + shdr->sh_addr = (unsigned long)dest; + DEBUGP("\t0x%lx %s\n", + shdr->sh_addr, info->secstrings + shdr->sh_name); } - /* Module has been moved. */ - mod = (void *)sechdrs[modindex].sh_addr; - kmemleak_load_module(mod, hdr, sechdrs, secstrings); -#if defined(CONFIG_MODULE_UNLOAD) - mod->refptr = alloc_percpu(struct module_ref); - if (!mod->refptr) { - err = -ENOMEM; - goto free_init; - } -#endif - /* Now we've moved module, initialize linked lists, etc. */ - module_unload_init(mod); - - /* add kobject, so we can reference it. */ - err = mod_sysfs_init(mod); - if (err) - goto free_unload; - - /* Set up license info based on the info section */ - set_license(mod, get_modinfo(sechdrs, infoindex, "license")); + return 0; +} +static int check_module_license_and_versions(struct module *mod) +{ /* * ndiswrapper is under GPL by itself, but loads proprietary modules. * Don't use add_taint_module(), as it would prevent ndiswrapper from @@ -2268,77 +2584,6 @@ static noinline struct module *load_module(void __user *umod, if (strcmp(mod->name, "driverloader") == 0) add_taint_module(mod, TAINT_PROPRIETARY_MODULE); - /* Set up MODINFO_ATTR fields */ - setup_modinfo(mod, sechdrs, infoindex); - - /* Fix up syms, so that st_value is a pointer to location. */ - err = simplify_symbols(sechdrs, symindex, strtab, versindex, pcpuindex, - mod); - if (err < 0) - goto cleanup; - - /* Now we've got everything in the final locations, we can - * find optional sections. */ - 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"); - mod->gpl_syms = section_objs(hdr, sechdrs, secstrings, "__ksymtab_gpl", - sizeof(*mod->gpl_syms), - &mod->num_gpl_syms); - mod->gpl_crcs = section_addr(hdr, sechdrs, secstrings, "__kcrctab_gpl"); - mod->gpl_future_syms = section_objs(hdr, sechdrs, secstrings, - "__ksymtab_gpl_future", - sizeof(*mod->gpl_future_syms), - &mod->num_gpl_future_syms); - mod->gpl_future_crcs = section_addr(hdr, sechdrs, secstrings, - "__kcrctab_gpl_future"); - -#ifdef CONFIG_UNUSED_SYMBOLS - mod->unused_syms = section_objs(hdr, sechdrs, secstrings, - "__ksymtab_unused", - sizeof(*mod->unused_syms), - &mod->num_unused_syms); - mod->unused_crcs = section_addr(hdr, sechdrs, secstrings, - "__kcrctab_unused"); - mod->unused_gpl_syms = section_objs(hdr, sechdrs, secstrings, - "__ksymtab_unused_gpl", - sizeof(*mod->unused_gpl_syms), - &mod->num_unused_gpl_syms); - mod->unused_gpl_crcs = section_addr(hdr, sechdrs, secstrings, - "__kcrctab_unused_gpl"); -#endif -#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); - /* - * This section contains pointers to allocated objects in the trace - * code and not scanning it leads to false positives. - */ - kmemleak_scan_area(mod->trace_events, sizeof(*mod->trace_events) * - mod->num_trace_events, GFP_KERNEL); -#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) @@ -2348,67 +2593,16 @@ static noinline struct module *load_module(void __user *umod, || (mod->num_unused_gpl_syms && !mod->unused_gpl_crcs) #endif ) { - err = try_to_force_load(mod, - "no versions for exported symbols"); - if (err) - goto cleanup; + return try_to_force_load(mod, + "no versions for exported symbols"); } #endif + return 0; +} - /* Now do relocations. */ - for (i = 1; i < hdr->e_shnum; i++) { - const char *strtab = (char *)sechdrs[strindex].sh_addr; - unsigned int info = sechdrs[i].sh_info; - - /* Not a valid relocation section? */ - if (info >= hdr->e_shnum) - continue; - - /* Don't bother with non-allocated sections */ - if (!(sechdrs[info].sh_flags & SHF_ALLOC)) - continue; - - if (sechdrs[i].sh_type == SHT_REL) - err = apply_relocate(sechdrs, strtab, symindex, i,mod); - else if (sechdrs[i].sh_type == SHT_RELA) - err = apply_relocate_add(sechdrs, strtab, symindex, i, - mod); - if (err < 0) - goto cleanup; - } - - /* Find duplicate symbols */ - err = verify_export_symbols(mod); - if (err < 0) - goto cleanup; - - /* Set up and sort exception table */ - mod->extable = section_objs(hdr, sechdrs, secstrings, "__ex_table", - sizeof(*mod->extable), &mod->num_exentries); - sort_extable(mod->extable, mod->extable + mod->num_exentries); - - /* Finally, copy percpu area over. */ - percpu_modcopy(mod, (void *)sechdrs[pcpuindex].sh_addr, - sechdrs[pcpuindex].sh_size); - - add_kallsyms(mod, sechdrs, hdr->e_shnum, symindex, strindex, - symoffs, stroffs, secstrings, strmap); - kfree(strmap); - strmap = NULL; - - if (!mod->taints) { - struct _ddebug *debug; - unsigned int num_debug; - - debug = section_objs(hdr, sechdrs, secstrings, "__verbose", - sizeof(*debug), &num_debug); - if (debug) - dynamic_debug_setup(debug, num_debug); - } - - err = module_finalize(hdr, sechdrs, mod); - if (err < 0) - goto cleanup; +static void flush_module_icache(const struct module *mod) +{ + mm_segment_t old_fs; /* flush the icache in correct context */ old_fs = get_fs(); @@ -2427,11 +2621,160 @@ static noinline struct module *load_module(void __user *umod, (unsigned long)mod->module_core + mod->core_size); set_fs(old_fs); +} - mod->args = args; - if (section_addr(hdr, sechdrs, secstrings, "__obsparm")) - printk(KERN_WARNING "%s: Ignoring obsolete parameters\n", - mod->name); +static struct module *layout_and_allocate(struct load_info *info) +{ + /* Module within temporary copy. */ + struct module *mod; + Elf_Shdr *pcpusec; + int err; + + mod = setup_load_info(info); + if (IS_ERR(mod)) + return mod; + + err = check_modinfo(mod, info); + if (err) + return ERR_PTR(err); + + /* Allow arches to frob section contents and sizes. */ + err = module_frob_arch_sections(info->hdr, info->sechdrs, + info->secstrings, mod); + if (err < 0) + goto out; + + pcpusec = &info->sechdrs[info->index.pcpu]; + if (pcpusec->sh_size) { + /* We have a special allocation for this section. */ + err = percpu_modalloc(mod, + pcpusec->sh_size, pcpusec->sh_addralign); + if (err) + goto out; + pcpusec->sh_flags &= ~(unsigned long)SHF_ALLOC; + } + + /* Determine total sizes, and put offsets in sh_entsize. For now + this is done generically; there doesn't appear to be any + special cases for the architectures. */ + layout_sections(mod, info); + + info->strmap = kzalloc(BITS_TO_LONGS(info->sechdrs[info->index.str].sh_size) + * sizeof(long), GFP_KERNEL); + if (!info->strmap) { + err = -ENOMEM; + goto free_percpu; + } + layout_symtab(mod, info); + + /* Allocate and move to the final place */ + err = move_module(mod, info); + if (err) + goto free_strmap; + + /* Module has been copied to its final place now: return it. */ + mod = (void *)info->sechdrs[info->index.mod].sh_addr; + kmemleak_load_module(mod, info); + return mod; + +free_strmap: + kfree(info->strmap); +free_percpu: + percpu_modfree(mod); +out: + return ERR_PTR(err); +} + +/* mod is no longer valid after this! */ +static void module_deallocate(struct module *mod, struct load_info *info) +{ + kfree(info->strmap); + percpu_modfree(mod); + module_free(mod, mod->module_init); + module_free(mod, mod->module_core); +} + +static int post_relocation(struct module *mod, const struct load_info *info) +{ + /* Sort exception table now relocations are done. */ + sort_extable(mod->extable, mod->extable + mod->num_exentries); + + /* Copy relocated percpu area over. */ + percpu_modcopy(mod, (void *)info->sechdrs[info->index.pcpu].sh_addr, + info->sechdrs[info->index.pcpu].sh_size); + + /* Setup kallsyms-specific fields. */ + add_kallsyms(mod, info); + + /* Arch-specific module finalizing. */ + return module_finalize(info->hdr, info->sechdrs, mod); +} + +/* Allocate and load the module: note that size of section 0 is always + zero, and we rely on this for optional sections. */ +static struct module *load_module(void __user *umod, + unsigned long len, + const char __user *uargs) +{ + struct load_info info = { NULL, }; + struct module *mod; + long err; + + DEBUGP("load_module: umod=%p, len=%lu, uargs=%p\n", + umod, len, uargs); + + /* Copy in the blobs from userspace, check they are vaguely sane. */ + err = copy_and_check(&info, umod, len, uargs); + if (err) + return ERR_PTR(err); + + /* Figure out module layout, and allocate all the memory. */ + mod = layout_and_allocate(&info); + if (IS_ERR(mod)) { + err = PTR_ERR(mod); + goto free_copy; + } + + /* Now module is in final location, initialize linked lists, etc. */ + err = module_unload_init(mod); + if (err) + goto free_module; + + /* Now we've got everything in the final locations, we can + * find optional sections. */ + find_module_sections(mod, &info); + + err = check_module_license_and_versions(mod); + if (err) + goto free_unload; + + /* Set up MODINFO_ATTR fields */ + setup_modinfo(mod, &info); + + /* Fix up syms, so that st_value is a pointer to location. */ + err = simplify_symbols(mod, &info); + if (err < 0) + goto free_modinfo; + + err = apply_relocations(mod, &info); + if (err < 0) + goto free_modinfo; + + err = post_relocation(mod, &info); + if (err < 0) + goto free_modinfo; + + flush_module_icache(mod); + + /* Now copy in args */ + mod->args = strndup_user(uargs, ~0UL >> 1); + if (IS_ERR(mod->args)) { + err = PTR_ERR(mod->args); + goto free_arch_cleanup; + } + + /* Mark state as coming so strong_try_module_get() ignores us. */ + mod->state = MODULE_STATE_COMING; /* Now sew it into the lists so we can get lockdep and oops * info during argument parsing. Noone should access us, since @@ -2440,58 +2783,67 @@ static noinline struct module *load_module(void __user *umod, * function to insert in a way safe to concurrent readers. * The mutex protects against concurrent writers. */ + mutex_lock(&module_mutex); + if (find_module(mod->name)) { + err = -EEXIST; + goto unlock; + } + + /* This has to be done once we're sure module name is unique. */ + if (!mod->taints) + dynamic_debug_setup(info.debug, info.num_debug); + + /* Find duplicate symbols */ + err = verify_export_symbols(mod); + if (err < 0) + goto ddebug; + + module_bug_finalize(info.hdr, info.sechdrs, mod); list_add_rcu(&mod->list, &modules); + mutex_unlock(&module_mutex); + /* Module is ready to execute: parsing args may do that. */ err = parse_args(mod->name, mod->args, mod->kp, mod->num_kp, NULL); if (err < 0) goto unlink; - err = mod_sysfs_setup(mod, mod->kp, mod->num_kp); + /* Link in to syfs. */ + err = mod_sysfs_setup(mod, &info, mod->kp, mod->num_kp); if (err < 0) goto unlink; - add_sect_attrs(mod, hdr->e_shnum, secstrings, sechdrs); - add_notes_attrs(mod, hdr->e_shnum, secstrings, sechdrs); - - /* Get rid of temporary copy */ - vfree(hdr); - trace_module_load(mod); + /* Get rid of temporary copy and strmap. */ + kfree(info.strmap); + free_copy(&info); /* Done! */ + trace_module_load(mod); return mod; unlink: + mutex_lock(&module_mutex); /* Unlink carefully: kallsyms could be walking list. */ list_del_rcu(&mod->list); + module_bug_cleanup(mod); + + ddebug: + if (!mod->taints) + dynamic_debug_remove(info.debug); + unlock: + mutex_unlock(&module_mutex); synchronize_sched(); + kfree(mod->args); + free_arch_cleanup: module_arch_cleanup(mod); - cleanup: + free_modinfo: free_modinfo(mod); - kobject_del(&mod->mkobj.kobj); - kobject_put(&mod->mkobj.kobj); free_unload: module_unload_free(mod); -#if defined(CONFIG_MODULE_UNLOAD) - free_percpu(mod->refptr); - free_init: -#endif - module_free(mod, mod->module_init); - free_core: - module_free(mod, mod->module_core); - /* mod will be freed with core. Don't access it beyond this line! */ - free_percpu: - percpu_modfree(mod); - free_mod: - kfree(args); - kfree(strmap); - free_hdr: - vfree(hdr); + free_module: + module_deallocate(mod, &info); + free_copy: + free_copy(&info); return ERR_PTR(err); - - truncated: - printk(KERN_ERR "Module len %lu truncated\n", len); - err = -ENOEXEC; - goto free_hdr; } /* Call module constructors. */ @@ -2516,23 +2868,26 @@ SYSCALL_DEFINE3(init_module, void __user *, umod, if (!capable(CAP_SYS_MODULE) || modules_disabled) return -EPERM; - /* Only one module load at a time, please */ - if (mutex_lock_interruptible(&module_mutex) != 0) - return -EINTR; - /* Do all the hard work */ mod = load_module(umod, len, uargs); - if (IS_ERR(mod)) { - mutex_unlock(&module_mutex); + if (IS_ERR(mod)) return PTR_ERR(mod); - } - - /* Drop lock so they can recurse */ - mutex_unlock(&module_mutex); blocking_notifier_call_chain(&module_notify_list, MODULE_STATE_COMING, mod); + /* Set RO and NX regions for core */ + set_section_ro_nx(mod->module_core, + mod->core_text_size, + mod->core_ro_size, + mod->core_size); + + /* Set RO and NX regions for init */ + set_section_ro_nx(mod->module_init, + mod->init_text_size, + mod->init_ro_size, + mod->init_size); + do_mod_ctors(mod); /* Start the module */ if (mod->init != NULL) @@ -2545,9 +2900,7 @@ SYSCALL_DEFINE3(init_module, void __user *, umod, module_put(mod); blocking_notifier_call_chain(&module_notify_list, MODULE_STATE_GOING, mod); - mutex_lock(&module_mutex); free_module(mod); - mutex_unlock(&module_mutex); wake_up(&module_wq); return ret; } @@ -2578,6 +2931,7 @@ SYSCALL_DEFINE3(init_module, void __user *, umod, mod->symtab = mod->core_symtab; mod->strtab = mod->core_strtab; #endif + unset_section_ro_nx(mod, mod->module_init); module_free(mod, mod->module_init); mod->module_init = NULL; mod->init_size = 0; diff --git a/kernel/mutex.c b/kernel/mutex.c index 632f04c57d8..a5889fb28ec 100644 --- a/kernel/mutex.c +++ b/kernel/mutex.c @@ -36,15 +36,6 @@ # include <asm/mutex.h> #endif -/*** - * mutex_init - initialize the mutex - * @lock: the mutex to be initialized - * @key: the lock_class_key for the class; used by mutex lock debugging - * - * Initialize the mutex to unlocked state. - * - * It is not allowed to initialize an already locked mutex. - */ void __mutex_init(struct mutex *lock, const char *name, struct lock_class_key *key) { @@ -68,7 +59,7 @@ EXPORT_SYMBOL(__mutex_init); static __used noinline void __sched __mutex_lock_slowpath(atomic_t *lock_count); -/*** +/** * mutex_lock - acquire the mutex * @lock: the mutex to be acquired * @@ -105,7 +96,7 @@ EXPORT_SYMBOL(mutex_lock); static __used noinline void __sched __mutex_unlock_slowpath(atomic_t *lock_count); -/*** +/** * mutex_unlock - release the mutex * @lock: the mutex to be released * @@ -172,6 +163,13 @@ __mutex_lock_common(struct mutex *lock, long state, unsigned int subclass, struct thread_info *owner; /* + * If we own the BKL, then don't spin. The owner of + * the mutex might be waiting on us to release the BKL. + */ + if (unlikely(current->lock_depth >= 0)) + break; + + /* * If there's an owner, wait for it to either * release the lock or go to sleep. */ @@ -201,7 +199,7 @@ __mutex_lock_common(struct mutex *lock, long state, unsigned int subclass, * memory barriers as we'll eventually observe the right * values at the cost of a few extra spins. */ - cpu_relax(); + arch_mutex_cpu_relax(); } #endif spin_lock_mutex(&lock->wait_lock, flags); @@ -357,8 +355,8 @@ __mutex_lock_killable_slowpath(atomic_t *lock_count); static noinline int __sched __mutex_lock_interruptible_slowpath(atomic_t *lock_count); -/*** - * mutex_lock_interruptible - acquire the mutex, interruptable +/** + * mutex_lock_interruptible - acquire the mutex, interruptible * @lock: the mutex to be acquired * * Lock the mutex like mutex_lock(), and return 0 if the mutex has @@ -449,15 +447,15 @@ static inline int __mutex_trylock_slowpath(atomic_t *lock_count) return prev == 1; } -/*** - * mutex_trylock - try acquire the mutex, without waiting +/** + * mutex_trylock - try to acquire the mutex, without waiting * @lock: the mutex to be acquired * * Try to acquire the mutex atomically. Returns 1 if the mutex * has been acquired successfully, and 0 on contention. * * NOTE: this function follows the spin_trylock() convention, so - * it is negated to the down_trylock() return values! Be careful + * it is negated from the down_trylock() return values! Be careful * about this when converting semaphore users to mutexes. * * This function must not be used in interrupt context. The diff --git a/kernel/ns_cgroup.c b/kernel/ns_cgroup.c index 2a5dfec8efe..2c98ad94ba0 100644 --- a/kernel/ns_cgroup.c +++ b/kernel/ns_cgroup.c @@ -85,6 +85,14 @@ static struct cgroup_subsys_state *ns_create(struct cgroup_subsys *ss, return ERR_PTR(-EPERM); if (!cgroup_is_descendant(cgroup, current)) return ERR_PTR(-EPERM); + if (test_bit(CGRP_CLONE_CHILDREN, &cgroup->flags)) { + printk("ns_cgroup can't be created with parent " + "'clone_children' set.\n"); + return ERR_PTR(-EINVAL); + } + + printk_once("ns_cgroup deprecated: consider using the " + "'clone_children' flag without the ns_cgroup.\n"); ns_cgroup = kzalloc(sizeof(*ns_cgroup), GFP_KERNEL); if (!ns_cgroup) diff --git a/kernel/padata.c b/kernel/padata.c index b1c9857f840..751019415d2 100644 --- a/kernel/padata.c +++ b/kernel/padata.c @@ -26,18 +26,19 @@ #include <linux/mutex.h> #include <linux/sched.h> #include <linux/slab.h> +#include <linux/sysfs.h> #include <linux/rcupdate.h> -#define MAX_SEQ_NR INT_MAX - NR_CPUS +#define MAX_SEQ_NR (INT_MAX - NR_CPUS) #define MAX_OBJ_NUM 1000 static int padata_index_to_cpu(struct parallel_data *pd, int cpu_index) { int cpu, target_cpu; - target_cpu = cpumask_first(pd->cpumask); + target_cpu = cpumask_first(pd->cpumask.pcpu); for (cpu = 0; cpu < cpu_index; cpu++) - target_cpu = cpumask_next(target_cpu, pd->cpumask); + target_cpu = cpumask_next(target_cpu, pd->cpumask.pcpu); return target_cpu; } @@ -53,26 +54,27 @@ static int padata_cpu_hash(struct padata_priv *padata) * Hash the sequence numbers to the cpus by taking * seq_nr mod. number of cpus in use. */ - cpu_index = padata->seq_nr % cpumask_weight(pd->cpumask); + cpu_index = padata->seq_nr % cpumask_weight(pd->cpumask.pcpu); return padata_index_to_cpu(pd, cpu_index); } -static void padata_parallel_worker(struct work_struct *work) +static void padata_parallel_worker(struct work_struct *parallel_work) { - struct padata_queue *queue; + struct padata_parallel_queue *pqueue; struct parallel_data *pd; struct padata_instance *pinst; LIST_HEAD(local_list); local_bh_disable(); - queue = container_of(work, struct padata_queue, pwork); - pd = queue->pd; + pqueue = container_of(parallel_work, + struct padata_parallel_queue, work); + pd = pqueue->pd; pinst = pd->pinst; - spin_lock(&queue->parallel.lock); - list_replace_init(&queue->parallel.list, &local_list); - spin_unlock(&queue->parallel.lock); + spin_lock(&pqueue->parallel.lock); + list_replace_init(&pqueue->parallel.list, &local_list); + spin_unlock(&pqueue->parallel.lock); while (!list_empty(&local_list)) { struct padata_priv *padata; @@ -94,7 +96,7 @@ static void padata_parallel_worker(struct work_struct *work) * @pinst: padata instance * @padata: object to be parallelized * @cb_cpu: cpu the serialization callback function will run on, - * must be in the cpumask of padata. + * must be in the serial cpumask of padata(i.e. cpumask.cbcpu). * * The parallelization callback function will run with BHs off. * Note: Every object which is parallelized by padata_do_parallel @@ -104,15 +106,18 @@ int padata_do_parallel(struct padata_instance *pinst, struct padata_priv *padata, int cb_cpu) { int target_cpu, err; - struct padata_queue *queue; + struct padata_parallel_queue *queue; struct parallel_data *pd; rcu_read_lock_bh(); pd = rcu_dereference(pinst->pd); - err = 0; - if (!(pinst->flags & PADATA_INIT)) + err = -EINVAL; + if (!(pinst->flags & PADATA_INIT) || pinst->flags & PADATA_INVALID) + goto out; + + if (!cpumask_test_cpu(cb_cpu, pd->cpumask.cbcpu)) goto out; err = -EBUSY; @@ -122,11 +127,7 @@ int padata_do_parallel(struct padata_instance *pinst, if (atomic_read(&pd->refcnt) >= MAX_OBJ_NUM) goto out; - err = -EINVAL; - if (!cpumask_test_cpu(cb_cpu, pd->cpumask)) - goto out; - - err = -EINPROGRESS; + err = 0; atomic_inc(&pd->refcnt); padata->pd = pd; padata->cb_cpu = cb_cpu; @@ -137,13 +138,13 @@ int padata_do_parallel(struct padata_instance *pinst, padata->seq_nr = atomic_inc_return(&pd->seq_nr); target_cpu = padata_cpu_hash(padata); - queue = per_cpu_ptr(pd->queue, target_cpu); + queue = per_cpu_ptr(pd->pqueue, target_cpu); spin_lock(&queue->parallel.lock); list_add_tail(&padata->list, &queue->parallel.list); spin_unlock(&queue->parallel.lock); - queue_work_on(target_cpu, pinst->wq, &queue->pwork); + queue_work_on(target_cpu, pinst->wq, &queue->work); out: rcu_read_unlock_bh(); @@ -171,84 +172,52 @@ EXPORT_SYMBOL(padata_do_parallel); */ static struct padata_priv *padata_get_next(struct parallel_data *pd) { - int cpu, num_cpus, empty, calc_seq_nr; - int seq_nr, next_nr, overrun, next_overrun; - struct padata_queue *queue, *next_queue; + int cpu, num_cpus; + int next_nr, next_index; + struct padata_parallel_queue *queue, *next_queue; struct padata_priv *padata; struct padata_list *reorder; - empty = 0; - next_nr = -1; - next_overrun = 0; - next_queue = NULL; - - num_cpus = cpumask_weight(pd->cpumask); - - for_each_cpu(cpu, pd->cpumask) { - queue = per_cpu_ptr(pd->queue, cpu); - reorder = &queue->reorder; - - /* - * Calculate the seq_nr of the object that should be - * next in this reorder queue. - */ - overrun = 0; - calc_seq_nr = (atomic_read(&queue->num_obj) * num_cpus) - + queue->cpu_index; + num_cpus = cpumask_weight(pd->cpumask.pcpu); - if (unlikely(calc_seq_nr > pd->max_seq_nr)) { - calc_seq_nr = calc_seq_nr - pd->max_seq_nr - 1; - overrun = 1; - } - - if (!list_empty(&reorder->list)) { - padata = list_entry(reorder->list.next, - struct padata_priv, list); - - seq_nr = padata->seq_nr; - BUG_ON(calc_seq_nr != seq_nr); - } else { - seq_nr = calc_seq_nr; - empty++; - } - - if (next_nr < 0 || seq_nr < next_nr - || (next_overrun && !overrun)) { - next_nr = seq_nr; - next_overrun = overrun; - next_queue = queue; - } + /* + * Calculate the percpu reorder queue and the sequence + * number of the next object. + */ + next_nr = pd->processed; + next_index = next_nr % num_cpus; + cpu = padata_index_to_cpu(pd, next_index); + next_queue = per_cpu_ptr(pd->pqueue, cpu); + + if (unlikely(next_nr > pd->max_seq_nr)) { + next_nr = next_nr - pd->max_seq_nr - 1; + next_index = next_nr % num_cpus; + cpu = padata_index_to_cpu(pd, next_index); + next_queue = per_cpu_ptr(pd->pqueue, cpu); + pd->processed = 0; } padata = NULL; - if (empty == num_cpus) - goto out; - reorder = &next_queue->reorder; if (!list_empty(&reorder->list)) { padata = list_entry(reorder->list.next, struct padata_priv, list); - if (unlikely(next_overrun)) { - for_each_cpu(cpu, pd->cpumask) { - queue = per_cpu_ptr(pd->queue, cpu); - atomic_set(&queue->num_obj, 0); - } - } + BUG_ON(next_nr != padata->seq_nr); spin_lock(&reorder->lock); list_del_init(&padata->list); atomic_dec(&pd->reorder_objects); spin_unlock(&reorder->lock); - atomic_inc(&next_queue->num_obj); + pd->processed++; goto out; } - queue = per_cpu_ptr(pd->queue, smp_processor_id()); + queue = per_cpu_ptr(pd->pqueue, smp_processor_id()); if (queue->cpu_index == next_queue->cpu_index) { padata = ERR_PTR(-ENODATA); goto out; @@ -262,7 +231,7 @@ out: static void padata_reorder(struct parallel_data *pd) { struct padata_priv *padata; - struct padata_queue *queue; + struct padata_serial_queue *squeue; struct padata_instance *pinst = pd->pinst; /* @@ -301,13 +270,13 @@ static void padata_reorder(struct parallel_data *pd) return; } - queue = per_cpu_ptr(pd->queue, padata->cb_cpu); + squeue = per_cpu_ptr(pd->squeue, padata->cb_cpu); - spin_lock(&queue->serial.lock); - list_add_tail(&padata->list, &queue->serial.list); - spin_unlock(&queue->serial.lock); + spin_lock(&squeue->serial.lock); + list_add_tail(&padata->list, &squeue->serial.list); + spin_unlock(&squeue->serial.lock); - queue_work_on(padata->cb_cpu, pinst->wq, &queue->swork); + queue_work_on(padata->cb_cpu, pinst->wq, &squeue->work); } spin_unlock_bh(&pd->lock); @@ -333,19 +302,19 @@ static void padata_reorder_timer(unsigned long arg) padata_reorder(pd); } -static void padata_serial_worker(struct work_struct *work) +static void padata_serial_worker(struct work_struct *serial_work) { - struct padata_queue *queue; + struct padata_serial_queue *squeue; struct parallel_data *pd; LIST_HEAD(local_list); local_bh_disable(); - queue = container_of(work, struct padata_queue, swork); - pd = queue->pd; + squeue = container_of(serial_work, struct padata_serial_queue, work); + pd = squeue->pd; - spin_lock(&queue->serial.lock); - list_replace_init(&queue->serial.list, &local_list); - spin_unlock(&queue->serial.lock); + spin_lock(&squeue->serial.lock); + list_replace_init(&squeue->serial.list, &local_list); + spin_unlock(&squeue->serial.lock); while (!list_empty(&local_list)) { struct padata_priv *padata; @@ -372,18 +341,18 @@ static void padata_serial_worker(struct work_struct *work) void padata_do_serial(struct padata_priv *padata) { int cpu; - struct padata_queue *queue; + struct padata_parallel_queue *pqueue; struct parallel_data *pd; pd = padata->pd; cpu = get_cpu(); - queue = per_cpu_ptr(pd->queue, cpu); + pqueue = per_cpu_ptr(pd->pqueue, cpu); - spin_lock(&queue->reorder.lock); + spin_lock(&pqueue->reorder.lock); atomic_inc(&pd->reorder_objects); - list_add_tail(&padata->list, &queue->reorder.list); - spin_unlock(&queue->reorder.lock); + list_add_tail(&padata->list, &pqueue->reorder.list); + spin_unlock(&pqueue->reorder.lock); put_cpu(); @@ -391,52 +360,89 @@ void padata_do_serial(struct padata_priv *padata) } EXPORT_SYMBOL(padata_do_serial); -/* Allocate and initialize the internal cpumask dependend resources. */ -static struct parallel_data *padata_alloc_pd(struct padata_instance *pinst, - const struct cpumask *cpumask) +static int padata_setup_cpumasks(struct parallel_data *pd, + const struct cpumask *pcpumask, + const struct cpumask *cbcpumask) { - int cpu, cpu_index, num_cpus; - struct padata_queue *queue; - struct parallel_data *pd; - - cpu_index = 0; + if (!alloc_cpumask_var(&pd->cpumask.pcpu, GFP_KERNEL)) + return -ENOMEM; - pd = kzalloc(sizeof(struct parallel_data), GFP_KERNEL); - if (!pd) - goto err; + cpumask_and(pd->cpumask.pcpu, pcpumask, cpu_active_mask); + if (!alloc_cpumask_var(&pd->cpumask.cbcpu, GFP_KERNEL)) { + free_cpumask_var(pd->cpumask.cbcpu); + return -ENOMEM; + } - pd->queue = alloc_percpu(struct padata_queue); - if (!pd->queue) - goto err_free_pd; + cpumask_and(pd->cpumask.cbcpu, cbcpumask, cpu_active_mask); + return 0; +} - if (!alloc_cpumask_var(&pd->cpumask, GFP_KERNEL)) - goto err_free_queue; +static void __padata_list_init(struct padata_list *pd_list) +{ + INIT_LIST_HEAD(&pd_list->list); + spin_lock_init(&pd_list->lock); +} - cpumask_and(pd->cpumask, cpumask, cpu_active_mask); +/* Initialize all percpu queues used by serial workers */ +static void padata_init_squeues(struct parallel_data *pd) +{ + int cpu; + struct padata_serial_queue *squeue; - for_each_cpu(cpu, pd->cpumask) { - queue = per_cpu_ptr(pd->queue, cpu); + for_each_cpu(cpu, pd->cpumask.cbcpu) { + squeue = per_cpu_ptr(pd->squeue, cpu); + squeue->pd = pd; + __padata_list_init(&squeue->serial); + INIT_WORK(&squeue->work, padata_serial_worker); + } +} - queue->pd = pd; +/* Initialize all percpu queues used by parallel workers */ +static void padata_init_pqueues(struct parallel_data *pd) +{ + int cpu_index, num_cpus, cpu; + struct padata_parallel_queue *pqueue; - queue->cpu_index = cpu_index; + cpu_index = 0; + for_each_cpu(cpu, pd->cpumask.pcpu) { + pqueue = per_cpu_ptr(pd->pqueue, cpu); + pqueue->pd = pd; + pqueue->cpu_index = cpu_index; cpu_index++; - INIT_LIST_HEAD(&queue->reorder.list); - INIT_LIST_HEAD(&queue->parallel.list); - INIT_LIST_HEAD(&queue->serial.list); - spin_lock_init(&queue->reorder.lock); - spin_lock_init(&queue->parallel.lock); - spin_lock_init(&queue->serial.lock); - - INIT_WORK(&queue->pwork, padata_parallel_worker); - INIT_WORK(&queue->swork, padata_serial_worker); - atomic_set(&queue->num_obj, 0); + __padata_list_init(&pqueue->reorder); + __padata_list_init(&pqueue->parallel); + INIT_WORK(&pqueue->work, padata_parallel_worker); + atomic_set(&pqueue->num_obj, 0); } - num_cpus = cpumask_weight(pd->cpumask); - pd->max_seq_nr = (MAX_SEQ_NR / num_cpus) * num_cpus - 1; + num_cpus = cpumask_weight(pd->cpumask.pcpu); + pd->max_seq_nr = num_cpus ? (MAX_SEQ_NR / num_cpus) * num_cpus - 1 : 0; +} + +/* Allocate and initialize the internal cpumask dependend resources. */ +static struct parallel_data *padata_alloc_pd(struct padata_instance *pinst, + const struct cpumask *pcpumask, + const struct cpumask *cbcpumask) +{ + struct parallel_data *pd; + pd = kzalloc(sizeof(struct parallel_data), GFP_KERNEL); + if (!pd) + goto err; + + pd->pqueue = alloc_percpu(struct padata_parallel_queue); + if (!pd->pqueue) + goto err_free_pd; + + pd->squeue = alloc_percpu(struct padata_serial_queue); + if (!pd->squeue) + goto err_free_pqueue; + if (padata_setup_cpumasks(pd, pcpumask, cbcpumask) < 0) + goto err_free_squeue; + + padata_init_pqueues(pd); + padata_init_squeues(pd); setup_timer(&pd->timer, padata_reorder_timer, (unsigned long)pd); atomic_set(&pd->seq_nr, -1); atomic_set(&pd->reorder_objects, 0); @@ -446,8 +452,10 @@ static struct parallel_data *padata_alloc_pd(struct padata_instance *pinst, return pd; -err_free_queue: - free_percpu(pd->queue); +err_free_squeue: + free_percpu(pd->squeue); +err_free_pqueue: + free_percpu(pd->pqueue); err_free_pd: kfree(pd); err: @@ -456,8 +464,10 @@ err: static void padata_free_pd(struct parallel_data *pd) { - free_cpumask_var(pd->cpumask); - free_percpu(pd->queue); + free_cpumask_var(pd->cpumask.pcpu); + free_cpumask_var(pd->cpumask.cbcpu); + free_percpu(pd->pqueue); + free_percpu(pd->squeue); kfree(pd); } @@ -465,11 +475,12 @@ static void padata_free_pd(struct parallel_data *pd) static void padata_flush_queues(struct parallel_data *pd) { int cpu; - struct padata_queue *queue; + struct padata_parallel_queue *pqueue; + struct padata_serial_queue *squeue; - for_each_cpu(cpu, pd->cpumask) { - queue = per_cpu_ptr(pd->queue, cpu); - flush_work(&queue->pwork); + for_each_cpu(cpu, pd->cpumask.pcpu) { + pqueue = per_cpu_ptr(pd->pqueue, cpu); + flush_work(&pqueue->work); } del_timer_sync(&pd->timer); @@ -477,19 +488,39 @@ static void padata_flush_queues(struct parallel_data *pd) if (atomic_read(&pd->reorder_objects)) padata_reorder(pd); - for_each_cpu(cpu, pd->cpumask) { - queue = per_cpu_ptr(pd->queue, cpu); - flush_work(&queue->swork); + for_each_cpu(cpu, pd->cpumask.cbcpu) { + squeue = per_cpu_ptr(pd->squeue, cpu); + flush_work(&squeue->work); } BUG_ON(atomic_read(&pd->refcnt) != 0); } +static void __padata_start(struct padata_instance *pinst) +{ + pinst->flags |= PADATA_INIT; +} + +static void __padata_stop(struct padata_instance *pinst) +{ + if (!(pinst->flags & PADATA_INIT)) + return; + + pinst->flags &= ~PADATA_INIT; + + synchronize_rcu(); + + get_online_cpus(); + padata_flush_queues(pinst->pd); + put_online_cpus(); +} + /* Replace the internal control stucture with a new one. */ static void padata_replace(struct padata_instance *pinst, struct parallel_data *pd_new) { struct parallel_data *pd_old = pinst->pd; + int notification_mask = 0; pinst->flags |= PADATA_RESET; @@ -497,41 +528,162 @@ static void padata_replace(struct padata_instance *pinst, synchronize_rcu(); + if (!cpumask_equal(pd_old->cpumask.pcpu, pd_new->cpumask.pcpu)) + notification_mask |= PADATA_CPU_PARALLEL; + if (!cpumask_equal(pd_old->cpumask.cbcpu, pd_new->cpumask.cbcpu)) + notification_mask |= PADATA_CPU_SERIAL; + padata_flush_queues(pd_old); padata_free_pd(pd_old); + if (notification_mask) + blocking_notifier_call_chain(&pinst->cpumask_change_notifier, + notification_mask, + &pd_new->cpumask); + pinst->flags &= ~PADATA_RESET; } /** - * padata_set_cpumask - set the cpumask that padata should use + * padata_register_cpumask_notifier - Registers a notifier that will be called + * if either pcpu or cbcpu or both cpumasks change. * - * @pinst: padata instance - * @cpumask: the cpumask to use + * @pinst: A poineter to padata instance + * @nblock: A pointer to notifier block. */ -int padata_set_cpumask(struct padata_instance *pinst, - cpumask_var_t cpumask) +int padata_register_cpumask_notifier(struct padata_instance *pinst, + struct notifier_block *nblock) { + return blocking_notifier_chain_register(&pinst->cpumask_change_notifier, + nblock); +} +EXPORT_SYMBOL(padata_register_cpumask_notifier); + +/** + * padata_unregister_cpumask_notifier - Unregisters cpumask notifier + * registered earlier using padata_register_cpumask_notifier + * + * @pinst: A pointer to data instance. + * @nlock: A pointer to notifier block. + */ +int padata_unregister_cpumask_notifier(struct padata_instance *pinst, + struct notifier_block *nblock) +{ + return blocking_notifier_chain_unregister( + &pinst->cpumask_change_notifier, + nblock); +} +EXPORT_SYMBOL(padata_unregister_cpumask_notifier); + + +/* If cpumask contains no active cpu, we mark the instance as invalid. */ +static bool padata_validate_cpumask(struct padata_instance *pinst, + const struct cpumask *cpumask) +{ + if (!cpumask_intersects(cpumask, cpu_active_mask)) { + pinst->flags |= PADATA_INVALID; + return false; + } + + pinst->flags &= ~PADATA_INVALID; + return true; +} + +static int __padata_set_cpumasks(struct padata_instance *pinst, + cpumask_var_t pcpumask, + cpumask_var_t cbcpumask) +{ + int valid; struct parallel_data *pd; - int err = 0; + + valid = padata_validate_cpumask(pinst, pcpumask); + if (!valid) { + __padata_stop(pinst); + goto out_replace; + } + + valid = padata_validate_cpumask(pinst, cbcpumask); + if (!valid) + __padata_stop(pinst); + +out_replace: + pd = padata_alloc_pd(pinst, pcpumask, cbcpumask); + if (!pd) + return -ENOMEM; + + cpumask_copy(pinst->cpumask.pcpu, pcpumask); + cpumask_copy(pinst->cpumask.cbcpu, cbcpumask); + + padata_replace(pinst, pd); + + if (valid) + __padata_start(pinst); + + return 0; +} + +/** + * padata_set_cpumasks - Set both parallel and serial cpumasks. The first + * one is used by parallel workers and the second one + * by the wokers doing serialization. + * + * @pinst: padata instance + * @pcpumask: the cpumask to use for parallel workers + * @cbcpumask: the cpumsak to use for serial workers + */ +int padata_set_cpumasks(struct padata_instance *pinst, cpumask_var_t pcpumask, + cpumask_var_t cbcpumask) +{ + int err; mutex_lock(&pinst->lock); + get_online_cpus(); + err = __padata_set_cpumasks(pinst, pcpumask, cbcpumask); + + put_online_cpus(); + mutex_unlock(&pinst->lock); + + return err; + +} +EXPORT_SYMBOL(padata_set_cpumasks); + +/** + * padata_set_cpumask: Sets specified by @cpumask_type cpumask to the value + * equivalent to @cpumask. + * + * @pinst: padata instance + * @cpumask_type: PADATA_CPU_SERIAL or PADATA_CPU_PARALLEL corresponding + * to parallel and serial cpumasks respectively. + * @cpumask: the cpumask to use + */ +int padata_set_cpumask(struct padata_instance *pinst, int cpumask_type, + cpumask_var_t cpumask) +{ + struct cpumask *serial_mask, *parallel_mask; + int err = -EINVAL; + + mutex_lock(&pinst->lock); get_online_cpus(); - pd = padata_alloc_pd(pinst, cpumask); - if (!pd) { - err = -ENOMEM; - goto out; + switch (cpumask_type) { + case PADATA_CPU_PARALLEL: + serial_mask = pinst->cpumask.cbcpu; + parallel_mask = cpumask; + break; + case PADATA_CPU_SERIAL: + parallel_mask = pinst->cpumask.pcpu; + serial_mask = cpumask; + break; + default: + goto out; } - cpumask_copy(pinst->cpumask, cpumask); - - padata_replace(pinst, pd); + err = __padata_set_cpumasks(pinst, parallel_mask, serial_mask); out: put_online_cpus(); - mutex_unlock(&pinst->lock); return err; @@ -543,30 +695,48 @@ static int __padata_add_cpu(struct padata_instance *pinst, int cpu) struct parallel_data *pd; if (cpumask_test_cpu(cpu, cpu_active_mask)) { - pd = padata_alloc_pd(pinst, pinst->cpumask); + pd = padata_alloc_pd(pinst, pinst->cpumask.pcpu, + pinst->cpumask.cbcpu); if (!pd) return -ENOMEM; padata_replace(pinst, pd); + + if (padata_validate_cpumask(pinst, pinst->cpumask.pcpu) && + padata_validate_cpumask(pinst, pinst->cpumask.cbcpu)) + __padata_start(pinst); } return 0; } -/** - * padata_add_cpu - add a cpu to the padata cpumask + /** + * padata_add_cpu - add a cpu to one or both(parallel and serial) + * padata cpumasks. * * @pinst: padata instance * @cpu: cpu to add + * @mask: bitmask of flags specifying to which cpumask @cpu shuld be added. + * The @mask may be any combination of the following flags: + * PADATA_CPU_SERIAL - serial cpumask + * PADATA_CPU_PARALLEL - parallel cpumask */ -int padata_add_cpu(struct padata_instance *pinst, int cpu) + +int padata_add_cpu(struct padata_instance *pinst, int cpu, int mask) { int err; + if (!(mask & (PADATA_CPU_SERIAL | PADATA_CPU_PARALLEL))) + return -EINVAL; + mutex_lock(&pinst->lock); get_online_cpus(); - cpumask_set_cpu(cpu, pinst->cpumask); + if (mask & PADATA_CPU_SERIAL) + cpumask_set_cpu(cpu, pinst->cpumask.cbcpu); + if (mask & PADATA_CPU_PARALLEL) + cpumask_set_cpu(cpu, pinst->cpumask.pcpu); + err = __padata_add_cpu(pinst, cpu); put_online_cpus(); @@ -578,10 +748,16 @@ EXPORT_SYMBOL(padata_add_cpu); static int __padata_remove_cpu(struct padata_instance *pinst, int cpu) { - struct parallel_data *pd; + struct parallel_data *pd = NULL; if (cpumask_test_cpu(cpu, cpu_online_mask)) { - pd = padata_alloc_pd(pinst, pinst->cpumask); + + if (!padata_validate_cpumask(pinst, pinst->cpumask.pcpu) || + !padata_validate_cpumask(pinst, pinst->cpumask.cbcpu)) + __padata_stop(pinst); + + pd = padata_alloc_pd(pinst, pinst->cpumask.pcpu, + pinst->cpumask.cbcpu); if (!pd) return -ENOMEM; @@ -591,20 +767,32 @@ static int __padata_remove_cpu(struct padata_instance *pinst, int cpu) return 0; } -/** - * padata_remove_cpu - remove a cpu from the padata cpumask + /** + * padata_remove_cpu - remove a cpu from the one or both(serial and paralell) + * padata cpumasks. * * @pinst: padata instance * @cpu: cpu to remove + * @mask: bitmask specifying from which cpumask @cpu should be removed + * The @mask may be any combination of the following flags: + * PADATA_CPU_SERIAL - serial cpumask + * PADATA_CPU_PARALLEL - parallel cpumask */ -int padata_remove_cpu(struct padata_instance *pinst, int cpu) +int padata_remove_cpu(struct padata_instance *pinst, int cpu, int mask) { int err; + if (!(mask & (PADATA_CPU_SERIAL | PADATA_CPU_PARALLEL))) + return -EINVAL; + mutex_lock(&pinst->lock); get_online_cpus(); - cpumask_clear_cpu(cpu, pinst->cpumask); + if (mask & PADATA_CPU_SERIAL) + cpumask_clear_cpu(cpu, pinst->cpumask.cbcpu); + if (mask & PADATA_CPU_PARALLEL) + cpumask_clear_cpu(cpu, pinst->cpumask.pcpu); + err = __padata_remove_cpu(pinst, cpu); put_online_cpus(); @@ -619,11 +807,20 @@ EXPORT_SYMBOL(padata_remove_cpu); * * @pinst: padata instance to start */ -void padata_start(struct padata_instance *pinst) +int padata_start(struct padata_instance *pinst) { + int err = 0; + mutex_lock(&pinst->lock); - pinst->flags |= PADATA_INIT; + + if (pinst->flags & PADATA_INVALID) + err =-EINVAL; + + __padata_start(pinst); + mutex_unlock(&pinst->lock); + + return err; } EXPORT_SYMBOL(padata_start); @@ -635,12 +832,20 @@ EXPORT_SYMBOL(padata_start); void padata_stop(struct padata_instance *pinst) { mutex_lock(&pinst->lock); - pinst->flags &= ~PADATA_INIT; + __padata_stop(pinst); mutex_unlock(&pinst->lock); } EXPORT_SYMBOL(padata_stop); #ifdef CONFIG_HOTPLUG_CPU + +static inline int pinst_has_cpu(struct padata_instance *pinst, int cpu) +{ + return cpumask_test_cpu(cpu, pinst->cpumask.pcpu) || + cpumask_test_cpu(cpu, pinst->cpumask.cbcpu); +} + + static int padata_cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu) { @@ -653,29 +858,29 @@ static int padata_cpu_callback(struct notifier_block *nfb, switch (action) { case CPU_ONLINE: case CPU_ONLINE_FROZEN: - if (!cpumask_test_cpu(cpu, pinst->cpumask)) + if (!pinst_has_cpu(pinst, cpu)) break; mutex_lock(&pinst->lock); err = __padata_add_cpu(pinst, cpu); mutex_unlock(&pinst->lock); if (err) - return NOTIFY_BAD; + return notifier_from_errno(err); break; case CPU_DOWN_PREPARE: case CPU_DOWN_PREPARE_FROZEN: - if (!cpumask_test_cpu(cpu, pinst->cpumask)) + if (!pinst_has_cpu(pinst, cpu)) break; mutex_lock(&pinst->lock); err = __padata_remove_cpu(pinst, cpu); mutex_unlock(&pinst->lock); if (err) - return NOTIFY_BAD; + return notifier_from_errno(err); break; case CPU_UP_CANCELED: case CPU_UP_CANCELED_FROZEN: - if (!cpumask_test_cpu(cpu, pinst->cpumask)) + if (!pinst_has_cpu(pinst, cpu)) break; mutex_lock(&pinst->lock); __padata_remove_cpu(pinst, cpu); @@ -683,7 +888,7 @@ static int padata_cpu_callback(struct notifier_block *nfb, case CPU_DOWN_FAILED: case CPU_DOWN_FAILED_FROZEN: - if (!cpumask_test_cpu(cpu, pinst->cpumask)) + if (!pinst_has_cpu(pinst, cpu)) break; mutex_lock(&pinst->lock); __padata_add_cpu(pinst, cpu); @@ -694,36 +899,202 @@ static int padata_cpu_callback(struct notifier_block *nfb, } #endif +static void __padata_free(struct padata_instance *pinst) +{ +#ifdef CONFIG_HOTPLUG_CPU + unregister_hotcpu_notifier(&pinst->cpu_notifier); +#endif + + padata_stop(pinst); + padata_free_pd(pinst->pd); + free_cpumask_var(pinst->cpumask.pcpu); + free_cpumask_var(pinst->cpumask.cbcpu); + kfree(pinst); +} + +#define kobj2pinst(_kobj) \ + container_of(_kobj, struct padata_instance, kobj) +#define attr2pentry(_attr) \ + container_of(_attr, struct padata_sysfs_entry, attr) + +static void padata_sysfs_release(struct kobject *kobj) +{ + struct padata_instance *pinst = kobj2pinst(kobj); + __padata_free(pinst); +} + +struct padata_sysfs_entry { + struct attribute attr; + ssize_t (*show)(struct padata_instance *, struct attribute *, char *); + ssize_t (*store)(struct padata_instance *, struct attribute *, + const char *, size_t); +}; + +static ssize_t show_cpumask(struct padata_instance *pinst, + struct attribute *attr, char *buf) +{ + struct cpumask *cpumask; + ssize_t len; + + mutex_lock(&pinst->lock); + if (!strcmp(attr->name, "serial_cpumask")) + cpumask = pinst->cpumask.cbcpu; + else + cpumask = pinst->cpumask.pcpu; + + len = bitmap_scnprintf(buf, PAGE_SIZE, cpumask_bits(cpumask), + nr_cpu_ids); + if (PAGE_SIZE - len < 2) + len = -EINVAL; + else + len += sprintf(buf + len, "\n"); + + mutex_unlock(&pinst->lock); + return len; +} + +static ssize_t store_cpumask(struct padata_instance *pinst, + struct attribute *attr, + const char *buf, size_t count) +{ + cpumask_var_t new_cpumask; + ssize_t ret; + int mask_type; + + if (!alloc_cpumask_var(&new_cpumask, GFP_KERNEL)) + return -ENOMEM; + + ret = bitmap_parse(buf, count, cpumask_bits(new_cpumask), + nr_cpumask_bits); + if (ret < 0) + goto out; + + mask_type = !strcmp(attr->name, "serial_cpumask") ? + PADATA_CPU_SERIAL : PADATA_CPU_PARALLEL; + ret = padata_set_cpumask(pinst, mask_type, new_cpumask); + if (!ret) + ret = count; + +out: + free_cpumask_var(new_cpumask); + return ret; +} + +#define PADATA_ATTR_RW(_name, _show_name, _store_name) \ + static struct padata_sysfs_entry _name##_attr = \ + __ATTR(_name, 0644, _show_name, _store_name) +#define PADATA_ATTR_RO(_name, _show_name) \ + static struct padata_sysfs_entry _name##_attr = \ + __ATTR(_name, 0400, _show_name, NULL) + +PADATA_ATTR_RW(serial_cpumask, show_cpumask, store_cpumask); +PADATA_ATTR_RW(parallel_cpumask, show_cpumask, store_cpumask); + +/* + * Padata sysfs provides the following objects: + * serial_cpumask [RW] - cpumask for serial workers + * parallel_cpumask [RW] - cpumask for parallel workers + */ +static struct attribute *padata_default_attrs[] = { + &serial_cpumask_attr.attr, + ¶llel_cpumask_attr.attr, + NULL, +}; + +static ssize_t padata_sysfs_show(struct kobject *kobj, + struct attribute *attr, char *buf) +{ + struct padata_instance *pinst; + struct padata_sysfs_entry *pentry; + ssize_t ret = -EIO; + + pinst = kobj2pinst(kobj); + pentry = attr2pentry(attr); + if (pentry->show) + ret = pentry->show(pinst, attr, buf); + + return ret; +} + +static ssize_t padata_sysfs_store(struct kobject *kobj, struct attribute *attr, + const char *buf, size_t count) +{ + struct padata_instance *pinst; + struct padata_sysfs_entry *pentry; + ssize_t ret = -EIO; + + pinst = kobj2pinst(kobj); + pentry = attr2pentry(attr); + if (pentry->show) + ret = pentry->store(pinst, attr, buf, count); + + return ret; +} + +static const struct sysfs_ops padata_sysfs_ops = { + .show = padata_sysfs_show, + .store = padata_sysfs_store, +}; + +static struct kobj_type padata_attr_type = { + .sysfs_ops = &padata_sysfs_ops, + .default_attrs = padata_default_attrs, + .release = padata_sysfs_release, +}; + /** - * padata_alloc - allocate and initialize a padata instance + * padata_alloc_possible - Allocate and initialize padata instance. + * Use the cpu_possible_mask for serial and + * parallel workers. * - * @cpumask: cpumask that padata uses for parallelization * @wq: workqueue to use for the allocated padata instance */ -struct padata_instance *padata_alloc(const struct cpumask *cpumask, - struct workqueue_struct *wq) +struct padata_instance *padata_alloc_possible(struct workqueue_struct *wq) +{ + return padata_alloc(wq, cpu_possible_mask, cpu_possible_mask); +} +EXPORT_SYMBOL(padata_alloc_possible); + +/** + * padata_alloc - allocate and initialize a padata instance and specify + * cpumasks for serial and parallel workers. + * + * @wq: workqueue to use for the allocated padata instance + * @pcpumask: cpumask that will be used for padata parallelization + * @cbcpumask: cpumask that will be used for padata serialization + */ +struct padata_instance *padata_alloc(struct workqueue_struct *wq, + const struct cpumask *pcpumask, + const struct cpumask *cbcpumask) { struct padata_instance *pinst; - struct parallel_data *pd; + struct parallel_data *pd = NULL; pinst = kzalloc(sizeof(struct padata_instance), GFP_KERNEL); if (!pinst) goto err; get_online_cpus(); - - pd = padata_alloc_pd(pinst, cpumask); - if (!pd) + if (!alloc_cpumask_var(&pinst->cpumask.pcpu, GFP_KERNEL)) goto err_free_inst; + if (!alloc_cpumask_var(&pinst->cpumask.cbcpu, GFP_KERNEL)) { + free_cpumask_var(pinst->cpumask.pcpu); + goto err_free_inst; + } + if (!padata_validate_cpumask(pinst, pcpumask) || + !padata_validate_cpumask(pinst, cbcpumask)) + goto err_free_masks; - if (!alloc_cpumask_var(&pinst->cpumask, GFP_KERNEL)) - goto err_free_pd; + pd = padata_alloc_pd(pinst, pcpumask, cbcpumask); + if (!pd) + goto err_free_masks; rcu_assign_pointer(pinst->pd, pd); pinst->wq = wq; - cpumask_copy(pinst->cpumask, cpumask); + cpumask_copy(pinst->cpumask.pcpu, pcpumask); + cpumask_copy(pinst->cpumask.cbcpu, cbcpumask); pinst->flags = 0; @@ -735,12 +1106,15 @@ struct padata_instance *padata_alloc(const struct cpumask *cpumask, put_online_cpus(); + BLOCKING_INIT_NOTIFIER_HEAD(&pinst->cpumask_change_notifier); + kobject_init(&pinst->kobj, &padata_attr_type); mutex_init(&pinst->lock); return pinst; -err_free_pd: - padata_free_pd(pd); +err_free_masks: + free_cpumask_var(pinst->cpumask.pcpu); + free_cpumask_var(pinst->cpumask.cbcpu); err_free_inst: kfree(pinst); put_online_cpus(); @@ -756,19 +1130,6 @@ EXPORT_SYMBOL(padata_alloc); */ void padata_free(struct padata_instance *pinst) { - padata_stop(pinst); - - synchronize_rcu(); - -#ifdef CONFIG_HOTPLUG_CPU - unregister_hotcpu_notifier(&pinst->cpu_notifier); -#endif - get_online_cpus(); - padata_flush_queues(pinst->pd); - put_online_cpus(); - - padata_free_pd(pinst->pd); - free_cpumask_var(pinst->cpumask); - kfree(pinst); + kobject_put(&pinst->kobj); } EXPORT_SYMBOL(padata_free); diff --git a/kernel/panic.c b/kernel/panic.c index dbe13dbb057..991bb87a170 100644 --- a/kernel/panic.c +++ b/kernel/panic.c @@ -24,6 +24,9 @@ #include <linux/nmi.h> #include <linux/dmi.h> +#define PANIC_TIMER_STEP 100 +#define PANIC_BLINK_SPD 18 + int panic_on_oops; static unsigned long tainted_mask; static int pause_on_oops; @@ -31,41 +34,21 @@ static int pause_on_oops_flag; static DEFINE_SPINLOCK(pause_on_oops_lock); int panic_timeout; +EXPORT_SYMBOL_GPL(panic_timeout); ATOMIC_NOTIFIER_HEAD(panic_notifier_list); EXPORT_SYMBOL(panic_notifier_list); -/* Returns how long it waited in ms */ -long (*panic_blink)(long time); -EXPORT_SYMBOL(panic_blink); - -static void panic_blink_one_second(void) +static long no_blink(int state) { - static long i = 0, end; - - if (panic_blink) { - end = i + MSEC_PER_SEC; - - while (i < end) { - i += panic_blink(i); - mdelay(1); - i++; - } - } else { - /* - * When running under a hypervisor a small mdelay may get - * rounded up to the hypervisor timeslice. For example, with - * a 1ms in 10ms hypervisor timeslice we might inflate a - * mdelay(1) loop by 10x. - * - * If we have nothing to blink, spin on 1 second calls to - * mdelay to avoid this. - */ - mdelay(MSEC_PER_SEC); - } + return 0; } +/* Returns how long it waited in ms */ +long (*panic_blink)(int state); +EXPORT_SYMBOL(panic_blink); + /** * panic - halt the system * @fmt: The text string to print @@ -78,7 +61,8 @@ NORET_TYPE void panic(const char * fmt, ...) { static char buf[1024]; va_list args; - long i; + long i, i_next = 0; + int state = 0; /* * It's possible to come here directly from a panic-assertion and @@ -87,6 +71,7 @@ NORET_TYPE void panic(const char * fmt, ...) */ preempt_disable(); + console_verbose(); bust_spinlocks(1); va_start(args, fmt); vsnprintf(buf, sizeof(buf), fmt, args); @@ -116,6 +101,9 @@ NORET_TYPE void panic(const char * fmt, ...) bust_spinlocks(0); + if (!panic_blink) + panic_blink = no_blink; + if (panic_timeout > 0) { /* * Delay timeout seconds before rebooting the machine. @@ -123,9 +111,13 @@ NORET_TYPE void panic(const char * fmt, ...) */ printk(KERN_EMERG "Rebooting in %d seconds..", panic_timeout); - for (i = 0; i < panic_timeout; i++) { + for (i = 0; i < panic_timeout * 1000; i += PANIC_TIMER_STEP) { touch_nmi_watchdog(); - panic_blink_one_second(); + if (i >= i_next) { + i += panic_blink(state ^= 1); + i_next = i + 3600 / PANIC_BLINK_SPD; + } + mdelay(PANIC_TIMER_STEP); } /* * This will not be a clean reboot, with everything @@ -151,9 +143,13 @@ NORET_TYPE void panic(const char * fmt, ...) } #endif local_irq_enable(); - while (1) { + for (i = 0; ; i += PANIC_TIMER_STEP) { touch_softlockup_watchdog(); - panic_blink_one_second(); + if (i >= i_next) { + i += panic_blink(state ^= 1); + i_next = i + 3600 / PANIC_BLINK_SPD; + } + mdelay(PANIC_TIMER_STEP); } } @@ -343,7 +339,7 @@ static int init_oops_id(void) } late_initcall(init_oops_id); -static void print_oops_end_marker(void) +void print_oops_end_marker(void) { init_oops_id(); printk(KERN_WARNING "---[ end trace %016llx ]---\n", diff --git a/kernel/params.c b/kernel/params.c index 0b30ecd53a5..0da1411222b 100644 --- a/kernel/params.c +++ b/kernel/params.c @@ -31,6 +31,42 @@ #define DEBUGP(fmt, a...) #endif +/* Protects all parameters, and incidentally kmalloced_param list. */ +static DEFINE_MUTEX(param_lock); + +/* This just allows us to keep track of which parameters are kmalloced. */ +struct kmalloced_param { + struct list_head list; + char val[]; +}; +static LIST_HEAD(kmalloced_params); + +static void *kmalloc_parameter(unsigned int size) +{ + struct kmalloced_param *p; + + p = kmalloc(sizeof(*p) + size, GFP_KERNEL); + if (!p) + return NULL; + + list_add(&p->list, &kmalloced_params); + return p->val; +} + +/* Does nothing if parameter wasn't kmalloced above. */ +static void maybe_kfree_parameter(void *param) +{ + struct kmalloced_param *p; + + list_for_each_entry(p, &kmalloced_params, list) { + if (p->val == param) { + list_del(&p->list); + kfree(p); + break; + } + } +} + static inline char dash2underscore(char c) { if (c == '-') @@ -49,18 +85,25 @@ static inline int parameq(const char *input, const char *paramname) static int parse_one(char *param, char *val, - struct kernel_param *params, + const struct kernel_param *params, unsigned num_params, int (*handle_unknown)(char *param, char *val)) { unsigned int i; + int err; /* Find parameter */ for (i = 0; i < num_params; i++) { if (parameq(param, params[i].name)) { + /* Noone handled NULL, so do it here. */ + if (!val && params[i].ops->set != param_set_bool) + return -EINVAL; DEBUGP("They are equal! Calling %p\n", - params[i].set); - return params[i].set(val, ¶ms[i]); + params[i].ops->set); + mutex_lock(¶m_lock); + err = params[i].ops->set(val, ¶ms[i]); + mutex_unlock(¶m_lock); + return err; } } @@ -128,7 +171,7 @@ static char *next_arg(char *args, char **param, char **val) /* Args looks like "foo=bar,bar2 baz=fuz wiz". */ int parse_args(const char *name, char *args, - struct kernel_param *params, + const struct kernel_param *params, unsigned num, int (*unknown)(char *param, char *val)) { @@ -176,22 +219,29 @@ int parse_args(const char *name, /* Lazy bastard, eh? */ #define STANDARD_PARAM_DEF(name, type, format, tmptype, strtolfn) \ - int param_set_##name(const char *val, struct kernel_param *kp) \ + int param_set_##name(const char *val, const struct kernel_param *kp) \ { \ tmptype l; \ int ret; \ \ - if (!val) return -EINVAL; \ ret = strtolfn(val, 0, &l); \ if (ret == -EINVAL || ((type)l != l)) \ return -EINVAL; \ *((type *)kp->arg) = l; \ return 0; \ } \ - int param_get_##name(char *buffer, struct kernel_param *kp) \ + int param_get_##name(char *buffer, const struct kernel_param *kp) \ { \ return sprintf(buffer, format, *((type *)kp->arg)); \ - } + } \ + struct kernel_param_ops param_ops_##name = { \ + .set = param_set_##name, \ + .get = param_get_##name, \ + }; \ + EXPORT_SYMBOL(param_set_##name); \ + EXPORT_SYMBOL(param_get_##name); \ + EXPORT_SYMBOL(param_ops_##name) + STANDARD_PARAM_DEF(byte, unsigned char, "%c", unsigned long, strict_strtoul); STANDARD_PARAM_DEF(short, short, "%hi", long, strict_strtol); @@ -201,39 +251,50 @@ STANDARD_PARAM_DEF(uint, unsigned int, "%u", unsigned long, strict_strtoul); STANDARD_PARAM_DEF(long, long, "%li", long, strict_strtol); STANDARD_PARAM_DEF(ulong, unsigned long, "%lu", unsigned long, strict_strtoul); -int param_set_charp(const char *val, struct kernel_param *kp) +int param_set_charp(const char *val, const struct kernel_param *kp) { - if (!val) { - printk(KERN_ERR "%s: string parameter expected\n", - kp->name); - return -EINVAL; - } - if (strlen(val) > 1024) { printk(KERN_ERR "%s: string parameter too long\n", kp->name); return -ENOSPC; } - /* This is a hack. We can't need to strdup in early boot, and we + maybe_kfree_parameter(*(char **)kp->arg); + + /* This is a hack. We can't kmalloc 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); + *(char **)kp->arg = kmalloc_parameter(strlen(val)+1); if (!*(char **)kp->arg) return -ENOMEM; + strcpy(*(char **)kp->arg, val); } else *(const char **)kp->arg = val; return 0; } +EXPORT_SYMBOL(param_set_charp); -int param_get_charp(char *buffer, struct kernel_param *kp) +int param_get_charp(char *buffer, const struct kernel_param *kp) { return sprintf(buffer, "%s", *((char **)kp->arg)); } +EXPORT_SYMBOL(param_get_charp); + +static void param_free_charp(void *arg) +{ + maybe_kfree_parameter(*((char **)arg)); +} + +struct kernel_param_ops param_ops_charp = { + .set = param_set_charp, + .get = param_get_charp, + .free = param_free_charp, +}; +EXPORT_SYMBOL(param_ops_charp); /* Actually could be a bool or an int, for historical reasons. */ -int param_set_bool(const char *val, struct kernel_param *kp) +int param_set_bool(const char *val, const struct kernel_param *kp) { bool v; @@ -258,8 +319,9 @@ int param_set_bool(const char *val, struct kernel_param *kp) *(int *)kp->arg = v; return 0; } +EXPORT_SYMBOL(param_set_bool); -int param_get_bool(char *buffer, struct kernel_param *kp) +int param_get_bool(char *buffer, const struct kernel_param *kp) { bool val; if (kp->flags & KPARAM_ISBOOL) @@ -270,9 +332,16 @@ int param_get_bool(char *buffer, struct kernel_param *kp) /* Y and N chosen as being relatively non-coder friendly */ return sprintf(buffer, "%c", val ? 'Y' : 'N'); } +EXPORT_SYMBOL(param_get_bool); + +struct kernel_param_ops param_ops_bool = { + .set = param_set_bool, + .get = param_get_bool, +}; +EXPORT_SYMBOL(param_ops_bool); /* This one must be bool. */ -int param_set_invbool(const char *val, struct kernel_param *kp) +int param_set_invbool(const char *val, const struct kernel_param *kp) { int ret; bool boolval; @@ -285,18 +354,26 @@ int param_set_invbool(const char *val, struct kernel_param *kp) *(bool *)kp->arg = !boolval; return ret; } +EXPORT_SYMBOL(param_set_invbool); -int param_get_invbool(char *buffer, struct kernel_param *kp) +int param_get_invbool(char *buffer, const struct kernel_param *kp) { return sprintf(buffer, "%c", (*(bool *)kp->arg) ? 'N' : 'Y'); } +EXPORT_SYMBOL(param_get_invbool); + +struct kernel_param_ops param_ops_invbool = { + .set = param_set_invbool, + .get = param_get_invbool, +}; +EXPORT_SYMBOL(param_ops_invbool); /* We break the rule and mangle the string. */ static int param_array(const char *name, const char *val, unsigned int min, unsigned int max, void *elem, int elemsize, - int (*set)(const char *, struct kernel_param *kp), + int (*set)(const char *, const struct kernel_param *kp), u16 flags, unsigned int *num) { @@ -309,12 +386,6 @@ static int param_array(const char *name, kp.arg = elem; kp.flags = flags; - /* No equals sign? */ - if (!val) { - printk(KERN_ERR "%s: expects arguments\n", name); - return -EINVAL; - } - *num = 0; /* We expect a comma-separated list of values. */ do { @@ -330,6 +401,7 @@ static int param_array(const char *name, /* nul-terminate and parse */ save = val[len]; ((char *)val)[len] = '\0'; + BUG_ON(!mutex_is_locked(¶m_lock)); ret = set(val, &kp); if (ret != 0) @@ -347,17 +419,17 @@ static int param_array(const char *name, return 0; } -int param_array_set(const char *val, struct kernel_param *kp) +static int param_array_set(const char *val, const struct kernel_param *kp) { const struct kparam_array *arr = kp->arr; unsigned int temp_num; return param_array(kp->name, val, 1, arr->max, arr->elem, - arr->elemsize, arr->set, kp->flags, + arr->elemsize, arr->ops->set, kp->flags, arr->num ?: &temp_num); } -int param_array_get(char *buffer, struct kernel_param *kp) +static int param_array_get(char *buffer, const struct kernel_param *kp) { int i, off, ret; const struct kparam_array *arr = kp->arr; @@ -368,7 +440,8 @@ int param_array_get(char *buffer, struct kernel_param *kp) if (i) buffer[off++] = ','; p.arg = arr->elem + arr->elemsize * i; - ret = arr->get(buffer + off, &p); + BUG_ON(!mutex_is_locked(¶m_lock)); + ret = arr->ops->get(buffer + off, &p); if (ret < 0) return ret; off += ret; @@ -377,14 +450,27 @@ int param_array_get(char *buffer, struct kernel_param *kp) return off; } -int param_set_copystring(const char *val, struct kernel_param *kp) +static void param_array_free(void *arg) +{ + unsigned int i; + const struct kparam_array *arr = arg; + + if (arr->ops->free) + for (i = 0; i < (arr->num ? *arr->num : arr->max); i++) + arr->ops->free(arr->elem + arr->elemsize * i); +} + +struct kernel_param_ops param_array_ops = { + .set = param_array_set, + .get = param_array_get, + .free = param_array_free, +}; +EXPORT_SYMBOL(param_array_ops); + +int param_set_copystring(const char *val, const struct kernel_param *kp) { const struct kparam_string *kps = kp->str; - if (!val) { - printk(KERN_ERR "%s: missing param set value\n", kp->name); - return -EINVAL; - } if (strlen(val)+1 > kps->maxlen) { printk(KERN_ERR "%s: string doesn't fit in %u chars.\n", kp->name, kps->maxlen-1); @@ -393,12 +479,20 @@ int param_set_copystring(const char *val, struct kernel_param *kp) strcpy(kps->string, val); return 0; } +EXPORT_SYMBOL(param_set_copystring); -int param_get_string(char *buffer, struct kernel_param *kp) +int param_get_string(char *buffer, const struct kernel_param *kp) { const struct kparam_string *kps = kp->str; return strlcpy(buffer, kps->string, kps->maxlen); } +EXPORT_SYMBOL(param_get_string); + +struct kernel_param_ops param_ops_string = { + .set = param_set_copystring, + .get = param_get_string, +}; +EXPORT_SYMBOL(param_ops_string); /* sysfs output in /sys/modules/XYZ/parameters/ */ #define to_module_attr(n) container_of(n, struct module_attribute, attr) @@ -409,7 +503,7 @@ extern struct kernel_param __start___param[], __stop___param[]; struct param_attribute { struct module_attribute mattr; - struct kernel_param *param; + const struct kernel_param *param; }; struct module_param_attrs @@ -428,10 +522,12 @@ static ssize_t param_attr_show(struct module_attribute *mattr, int count; struct param_attribute *attribute = to_param_attr(mattr); - if (!attribute->param->get) + if (!attribute->param->ops->get) return -EPERM; - count = attribute->param->get(buf, attribute->param); + mutex_lock(¶m_lock); + count = attribute->param->ops->get(buf, attribute->param); + mutex_unlock(¶m_lock); if (count > 0) { strcat(buf, "\n"); ++count; @@ -447,10 +543,12 @@ static ssize_t param_attr_store(struct module_attribute *mattr, int err; struct param_attribute *attribute = to_param_attr(mattr); - if (!attribute->param->set) + if (!attribute->param->ops->set) return -EPERM; - err = attribute->param->set(buf, attribute->param); + mutex_lock(¶m_lock); + err = attribute->param->ops->set(buf, attribute->param); + mutex_unlock(¶m_lock); if (!err) return len; return err; @@ -464,6 +562,18 @@ static ssize_t param_attr_store(struct module_attribute *mattr, #endif #ifdef CONFIG_SYSFS +void __kernel_param_lock(void) +{ + mutex_lock(¶m_lock); +} +EXPORT_SYMBOL(__kernel_param_lock); + +void __kernel_param_unlock(void) +{ + mutex_unlock(¶m_lock); +} +EXPORT_SYMBOL(__kernel_param_unlock); + /* * add_sysfs_param - add a parameter to sysfs * @mk: struct module_kobject @@ -475,7 +585,7 @@ static ssize_t param_attr_store(struct module_attribute *mattr, * if there's an error. */ static __modinit int add_sysfs_param(struct module_kobject *mk, - struct kernel_param *kp, + const struct kernel_param *kp, const char *name) { struct module_param_attrs *new; @@ -557,7 +667,7 @@ static void free_module_param_attrs(struct module_kobject *mk) * /sys/module/[mod->name]/parameters/ */ int module_param_sysfs_setup(struct module *mod, - struct kernel_param *kparam, + const struct kernel_param *kparam, unsigned int num_params) { int i, err; @@ -602,12 +712,14 @@ void module_param_sysfs_remove(struct module *mod) void destroy_params(const struct kernel_param *params, unsigned num) { - /* FIXME: This should free kmalloced charp parameters. It doesn't. */ + unsigned int i; + + for (i = 0; i < num; i++) + if (params[i].ops->free) + params[i].ops->free(params[i].arg); } -static void __init kernel_add_sysfs_param(const char *name, - struct kernel_param *kparam, - unsigned int name_skip) +static struct module_kobject * __init locate_module_kobject(const char *name) { struct module_kobject *mk; struct kobject *kobj; @@ -615,10 +727,7 @@ static void __init kernel_add_sysfs_param(const char *name, kobj = kset_find_obj(module_kset, name); if (kobj) { - /* We already have one. Remove params so we can add more. */ mk = to_module_kobject(kobj); - /* We need to remove it before adding parameters. */ - sysfs_remove_group(&mk->kobj, &mk->mp->grp); } else { mk = kzalloc(sizeof(struct module_kobject), GFP_KERNEL); BUG_ON(!mk); @@ -629,15 +738,36 @@ static void __init kernel_add_sysfs_param(const char *name, "%s", name); if (err) { kobject_put(&mk->kobj); - printk(KERN_ERR "Module '%s' failed add to sysfs, " - "error number %d\n", name, err); - printk(KERN_ERR "The system will be unstable now.\n"); - return; + printk(KERN_ERR + "Module '%s' failed add to sysfs, error number %d\n", + name, err); + printk(KERN_ERR + "The system will be unstable now.\n"); + return NULL; } - /* So that exit path is even. */ + + /* So that we hold reference in both cases. */ kobject_get(&mk->kobj); } + return mk; +} + +static void __init kernel_add_sysfs_param(const char *name, + struct kernel_param *kparam, + unsigned int name_skip) +{ + struct module_kobject *mk; + int err; + + mk = locate_module_kobject(name); + if (!mk) + return; + + /* We need to remove old parameters before adding more. */ + if (mk->mp) + sysfs_remove_group(&mk->kobj, &mk->mp->grp); + /* These should not fail at boot. */ err = add_sysfs_param(mk, kparam, kparam->name + name_skip); BUG_ON(err); @@ -682,6 +812,32 @@ static void __init param_sysfs_builtin(void) } } +ssize_t __modver_version_show(struct module_attribute *mattr, + struct module *mod, char *buf) +{ + struct module_version_attribute *vattr = + container_of(mattr, struct module_version_attribute, mattr); + + return sprintf(buf, "%s\n", vattr->version); +} + +extern struct module_version_attribute __start___modver[], __stop___modver[]; + +static void __init version_sysfs_builtin(void) +{ + const struct module_version_attribute *vattr; + struct module_kobject *mk; + int err; + + for (vattr = __start___modver; vattr < __stop___modver; vattr++) { + mk = locate_module_kobject(vattr->module_name); + if (mk) { + err = sysfs_create_file(&mk->kobj, &vattr->mattr.attr); + kobject_uevent(&mk->kobj, KOBJ_ADD); + kobject_put(&mk->kobj); + } + } +} /* module-related sysfs stuff */ @@ -761,6 +917,7 @@ static int __init param_sysfs_init(void) } module_sysfs_initialized = 1; + version_sysfs_builtin(); param_sysfs_builtin(); return 0; @@ -768,28 +925,3 @@ static int __init param_sysfs_init(void) subsys_initcall(param_sysfs_init); #endif /* CONFIG_SYSFS */ - -EXPORT_SYMBOL(param_set_byte); -EXPORT_SYMBOL(param_get_byte); -EXPORT_SYMBOL(param_set_short); -EXPORT_SYMBOL(param_get_short); -EXPORT_SYMBOL(param_set_ushort); -EXPORT_SYMBOL(param_get_ushort); -EXPORT_SYMBOL(param_set_int); -EXPORT_SYMBOL(param_get_int); -EXPORT_SYMBOL(param_set_uint); -EXPORT_SYMBOL(param_get_uint); -EXPORT_SYMBOL(param_set_long); -EXPORT_SYMBOL(param_get_long); -EXPORT_SYMBOL(param_set_ulong); -EXPORT_SYMBOL(param_get_ulong); -EXPORT_SYMBOL(param_set_charp); -EXPORT_SYMBOL(param_get_charp); -EXPORT_SYMBOL(param_set_bool); -EXPORT_SYMBOL(param_get_bool); -EXPORT_SYMBOL(param_set_invbool); -EXPORT_SYMBOL(param_get_invbool); -EXPORT_SYMBOL(param_array_set); -EXPORT_SYMBOL(param_array_get); -EXPORT_SYMBOL(param_set_copystring); -EXPORT_SYMBOL(param_get_string); diff --git a/kernel/perf_event.c b/kernel/perf_event.c index a4fa381db3c..126a302c481 100644 --- a/kernel/perf_event.c +++ b/kernel/perf_event.c @@ -13,6 +13,7 @@ #include <linux/mm.h> #include <linux/cpu.h> #include <linux/smp.h> +#include <linux/idr.h> #include <linux/file.h> #include <linux/poll.h> #include <linux/slab.h> @@ -21,7 +22,9 @@ #include <linux/dcache.h> #include <linux/percpu.h> #include <linux/ptrace.h> +#include <linux/reboot.h> #include <linux/vmstat.h> +#include <linux/device.h> #include <linux/vmalloc.h> #include <linux/hardirq.h> #include <linux/rculist.h> @@ -35,20 +38,21 @@ #include <asm/irq_regs.h> -/* - * Each CPU has a list of per CPU events: - */ -static 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; +enum event_type_t { + EVENT_FLEXIBLE = 0x1, + EVENT_PINNED = 0x2, + EVENT_ALL = EVENT_FLEXIBLE | EVENT_PINNED, +}; -static atomic_t nr_events __read_mostly; +atomic_t perf_task_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; +static LIST_HEAD(pmus); +static DEFINE_MUTEX(pmus_lock); +static struct srcu_struct pmus_srcu; + /* * perf event paranoia level: * -1 - not paranoid at all @@ -67,36 +71,54 @@ 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); +static void cpu_ctx_sched_out(struct perf_cpu_context *cpuctx, + enum event_type_t event_type); -/* - * Architecture provided APIs - weak aliases: - */ -extern __weak const struct pmu *hw_perf_event_init(struct perf_event *event) +static void cpu_ctx_sched_in(struct perf_cpu_context *cpuctx, + enum event_type_t event_type); + +void __weak perf_event_print_debug(void) { } + +extern __weak const char *perf_pmu_name(void) { - return NULL; + return "pmu"; } -void __weak hw_perf_disable(void) { barrier(); } -void __weak hw_perf_enable(void) { barrier(); } - -void __weak perf_event_print_debug(void) { } +static inline u64 perf_clock(void) +{ + return local_clock(); +} -static DEFINE_PER_CPU(int, perf_disable_count); +void perf_pmu_disable(struct pmu *pmu) +{ + int *count = this_cpu_ptr(pmu->pmu_disable_count); + if (!(*count)++) + pmu->pmu_disable(pmu); +} -void perf_disable(void) +void perf_pmu_enable(struct pmu *pmu) { - if (!__get_cpu_var(perf_disable_count)++) - hw_perf_disable(); + int *count = this_cpu_ptr(pmu->pmu_disable_count); + if (!--(*count)) + pmu->pmu_enable(pmu); } -void perf_enable(void) +static DEFINE_PER_CPU(struct list_head, rotation_list); + +/* + * perf_pmu_rotate_start() and perf_rotate_context() are fully serialized + * because they're strictly cpu affine and rotate_start is called with IRQs + * disabled, while rotate_context is called from IRQ context. + */ +static void perf_pmu_rotate_start(struct pmu *pmu) { - if (!--__get_cpu_var(perf_disable_count)) - hw_perf_enable(); + struct perf_cpu_context *cpuctx = this_cpu_ptr(pmu->pmu_cpu_context); + struct list_head *head = &__get_cpu_var(rotation_list); + + WARN_ON(!irqs_disabled()); + + if (list_empty(&cpuctx->rotation_list)) + list_add(&cpuctx->rotation_list, head); } static void get_ctx(struct perf_event_context *ctx) @@ -131,6 +153,28 @@ static void unclone_ctx(struct perf_event_context *ctx) } } +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); +} + /* * If we inherit events we want to return the parent event id * to userspace. @@ -151,13 +195,13 @@ static u64 primary_event_id(struct perf_event *event) * the context could get moved to another task. */ static struct perf_event_context * -perf_lock_task_context(struct task_struct *task, unsigned long *flags) +perf_lock_task_context(struct task_struct *task, int ctxn, unsigned long *flags) { struct perf_event_context *ctx; rcu_read_lock(); - retry: - ctx = rcu_dereference(task->perf_event_ctxp); +retry: + ctx = rcu_dereference(task->perf_event_ctxp[ctxn]); if (ctx) { /* * If this context is a clone of another, it might @@ -170,7 +214,7 @@ perf_lock_task_context(struct task_struct *task, unsigned long *flags) * can't get swapped on us any more. */ raw_spin_lock_irqsave(&ctx->lock, *flags); - if (ctx != rcu_dereference(task->perf_event_ctxp)) { + if (ctx != rcu_dereference(task->perf_event_ctxp[ctxn])) { raw_spin_unlock_irqrestore(&ctx->lock, *flags); goto retry; } @@ -189,12 +233,13 @@ perf_lock_task_context(struct task_struct *task, unsigned long *flags) * 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) +static struct perf_event_context * +perf_pin_task_context(struct task_struct *task, int ctxn) { struct perf_event_context *ctx; unsigned long flags; - ctx = perf_lock_task_context(task, &flags); + ctx = perf_lock_task_context(task, ctxn, &flags); if (ctx) { ++ctx->pin_count; raw_spin_unlock_irqrestore(&ctx->lock, flags); @@ -212,11 +257,6 @@ static void perf_unpin_context(struct perf_event_context *ctx) put_ctx(ctx); } -static inline u64 perf_clock(void) -{ - return cpu_clock(raw_smp_processor_id()); -} - /* * Update the record of the current time in a context. */ @@ -228,6 +268,12 @@ static void update_context_time(struct perf_event_context *ctx) ctx->timestamp = now; } +static u64 perf_event_time(struct perf_event *event) +{ + struct perf_event_context *ctx = event->ctx; + return ctx ? ctx->time : 0; +} + /* * Update the total_time_enabled and total_time_running fields for a event. */ @@ -241,7 +287,7 @@ static void update_event_times(struct perf_event *event) return; if (ctx->is_active) - run_end = ctx->time; + run_end = perf_event_time(event); else run_end = event->tstamp_stopped; @@ -250,7 +296,7 @@ static void update_event_times(struct perf_event *event) if (event->state == PERF_EVENT_STATE_INACTIVE) run_end = event->tstamp_stopped; else - run_end = ctx->time; + run_end = perf_event_time(event); event->total_time_running = run_end - event->tstamp_running; } @@ -283,14 +329,15 @@ ctx_group_list(struct perf_event *event, struct perf_event_context *ctx) static void list_add_event(struct perf_event *event, struct perf_event_context *ctx) { - struct perf_event *group_leader = event->group_leader; + WARN_ON_ONCE(event->attach_state & PERF_ATTACH_CONTEXT); + event->attach_state |= PERF_ATTACH_CONTEXT; /* - * 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 we're a stand alone event or group leader, we go to the context + * list, group events are kept attached to the group so that + * perf_group_detach can, at all times, locate all siblings. */ - if (group_leader == event) { + if (event->group_leader == event) { struct list_head *list; if (is_software_event(event)) @@ -298,39 +345,142 @@ list_add_event(struct perf_event *event, struct perf_event_context *ctx) list = ctx_group_list(event, ctx); list_add_tail(&event->group_entry, list); - } else { - if (group_leader->group_flags & PERF_GROUP_SOFTWARE && - !is_software_event(event)) - group_leader->group_flags &= ~PERF_GROUP_SOFTWARE; - - list_add_tail(&event->group_entry, &group_leader->sibling_list); - group_leader->nr_siblings++; } list_add_rcu(&event->event_entry, &ctx->event_list); + if (!ctx->nr_events) + perf_pmu_rotate_start(ctx->pmu); ctx->nr_events++; if (event->attr.inherit_stat) ctx->nr_stat++; } /* + * Called at perf_event creation and when events are attached/detached from a + * group. + */ +static void 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; + event->read_size = size; +} + +static void perf_event__header_size(struct perf_event *event) +{ + struct perf_sample_data *data; + u64 sample_type = event->attr.sample_type; + u16 size = 0; + + perf_event__read_size(event); + + if (sample_type & PERF_SAMPLE_IP) + size += sizeof(data->ip); + + if (sample_type & PERF_SAMPLE_ADDR) + size += sizeof(data->addr); + + if (sample_type & PERF_SAMPLE_PERIOD) + size += sizeof(data->period); + + if (sample_type & PERF_SAMPLE_READ) + size += event->read_size; + + event->header_size = size; +} + +static void perf_event__id_header_size(struct perf_event *event) +{ + struct perf_sample_data *data; + u64 sample_type = event->attr.sample_type; + u16 size = 0; + + if (sample_type & PERF_SAMPLE_TID) + size += sizeof(data->tid_entry); + + if (sample_type & PERF_SAMPLE_TIME) + size += sizeof(data->time); + + if (sample_type & PERF_SAMPLE_ID) + size += sizeof(data->id); + + if (sample_type & PERF_SAMPLE_STREAM_ID) + size += sizeof(data->stream_id); + + if (sample_type & PERF_SAMPLE_CPU) + size += sizeof(data->cpu_entry); + + event->id_header_size = size; +} + +static void perf_group_attach(struct perf_event *event) +{ + struct perf_event *group_leader = event->group_leader, *pos; + + /* + * We can have double attach due to group movement in perf_event_open. + */ + if (event->attach_state & PERF_ATTACH_GROUP) + return; + + event->attach_state |= PERF_ATTACH_GROUP; + + if (group_leader == event) + return; + + if (group_leader->group_flags & PERF_GROUP_SOFTWARE && + !is_software_event(event)) + group_leader->group_flags &= ~PERF_GROUP_SOFTWARE; + + list_add_tail(&event->group_entry, &group_leader->sibling_list); + group_leader->nr_siblings++; + + perf_event__header_size(group_leader); + + list_for_each_entry(pos, &group_leader->sibling_list, group_entry) + perf_event__header_size(pos); +} + +/* * 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) { - if (list_empty(&event->group_entry)) + /* + * We can have double detach due to exit/hot-unplug + close. + */ + if (!(event->attach_state & PERF_ATTACH_CONTEXT)) return; + + event->attach_state &= ~PERF_ATTACH_CONTEXT; + 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--; + if (event->group_leader == event) + list_del_init(&event->group_entry); update_group_times(event); @@ -345,26 +495,56 @@ list_del_event(struct perf_event *event, struct perf_event_context *ctx) event->state = PERF_EVENT_STATE_OFF; } -static void -perf_destroy_group(struct perf_event *event, struct perf_event_context *ctx) +static void perf_group_detach(struct perf_event *event) { struct perf_event *sibling, *tmp; + struct list_head *list = NULL; + + /* + * We can have double detach due to exit/hot-unplug + close. + */ + if (!(event->attach_state & PERF_ATTACH_GROUP)) + return; + + event->attach_state &= ~PERF_ATTACH_GROUP; + + /* + * If this is a sibling, remove it from its group. + */ + if (event->group_leader != event) { + list_del_init(&event->group_entry); + event->group_leader->nr_siblings--; + goto out; + } + + if (!list_empty(&event->group_entry)) + list = &event->group_entry; /* * If this was a group event with sibling events then * upgrade the siblings to singleton events by adding them - * to the context list directly: + * to whatever list we are on. */ list_for_each_entry_safe(sibling, tmp, &event->sibling_list, group_entry) { - struct list_head *list; - - list = ctx_group_list(event, ctx); - list_move_tail(&sibling->group_entry, list); + if (list) + list_move_tail(&sibling->group_entry, list); sibling->group_leader = sibling; /* Inherit group flags from the previous leader */ sibling->group_flags = event->group_flags; } + +out: + perf_event__header_size(event->group_leader); + + list_for_each_entry(tmp, &event->group_leader->sibling_list, group_entry) + perf_event__header_size(tmp); +} + +static inline int +event_filter_match(struct perf_event *event) +{ + return event->cpu == -1 || event->cpu == smp_processor_id(); } static void @@ -372,6 +552,21 @@ event_sched_out(struct perf_event *event, struct perf_cpu_context *cpuctx, struct perf_event_context *ctx) { + u64 tstamp = perf_event_time(event); + u64 delta; + /* + * An event which could not be activated because of + * filter mismatch still needs to have its timings + * maintained, otherwise bogus information is return + * via read() for time_enabled, time_running: + */ + if (event->state == PERF_EVENT_STATE_INACTIVE + && !event_filter_match(event)) { + delta = ctx->time - event->tstamp_stopped; + event->tstamp_running += delta; + event->tstamp_stopped = tstamp; + } + if (event->state != PERF_EVENT_STATE_ACTIVE) return; @@ -380,8 +575,8 @@ event_sched_out(struct perf_event *event, event->pending_disable = 0; event->state = PERF_EVENT_STATE_OFF; } - event->tstamp_stopped = ctx->time; - event->pmu->disable(event); + event->tstamp_stopped = tstamp; + event->pmu->del(event, 0); event->oncpu = -1; if (!is_software_event(event)) @@ -397,9 +592,7 @@ group_sched_out(struct perf_event *group_event, struct perf_event_context *ctx) { struct perf_event *event; - - if (group_event->state != PERF_EVENT_STATE_ACTIVE) - return; + int state = group_event->state; event_sched_out(group_event, cpuctx, ctx); @@ -409,10 +602,16 @@ group_sched_out(struct perf_event *group_event, list_for_each_entry(event, &group_event->sibling_list, group_entry) event_sched_out(event, cpuctx, ctx); - if (group_event->attr.exclusive) + if (state == PERF_EVENT_STATE_ACTIVE && group_event->attr.exclusive) cpuctx->exclusive = 0; } +static inline struct perf_cpu_context * +__get_cpu_context(struct perf_event_context *ctx) +{ + return this_cpu_ptr(ctx->pmu->pmu_cpu_context); +} + /* * Cross CPU call to remove a performance event * @@ -421,9 +620,9 @@ group_sched_out(struct perf_event *group_event, */ 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; + struct perf_cpu_context *cpuctx = __get_cpu_context(ctx); /* * If this is a task context, we need to check whether it is @@ -434,27 +633,11 @@ static void __perf_event_remove_from_context(void *info) return; raw_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(); raw_spin_unlock(&ctx->lock); } @@ -519,8 +702,8 @@ retry: 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; + struct perf_cpu_context *cpuctx = __get_cpu_context(ctx); /* * If this is a per-task event, need to check whether this @@ -575,7 +758,7 @@ void perf_event_disable(struct perf_event *event) return; } - retry: +retry: task_oncpu_function_call(task, __perf_event_disable, event); raw_spin_lock_irq(&ctx->lock); @@ -604,6 +787,8 @@ event_sched_in(struct perf_event *event, struct perf_cpu_context *cpuctx, struct perf_event_context *ctx) { + u64 tstamp = perf_event_time(event); + if (event->state <= PERF_EVENT_STATE_OFF) return 0; @@ -614,13 +799,15 @@ event_sched_in(struct perf_event *event, */ smp_wmb(); - if (event->pmu->enable(event)) { + if (event->pmu->add(event, PERF_EF_START)) { event->state = PERF_EVENT_STATE_INACTIVE; event->oncpu = -1; return -EAGAIN; } - event->tstamp_running += ctx->time - event->tstamp_stopped; + event->tstamp_running += tstamp - event->tstamp_stopped; + + event->shadow_ctx_time = tstamp - ctx->timestamp; if (!is_software_event(event)) cpuctx->active_oncpu++; @@ -638,22 +825,19 @@ group_sched_in(struct perf_event *group_event, struct perf_event_context *ctx) { struct perf_event *event, *partial_group = NULL; - const struct pmu *pmu = group_event->pmu; - bool txn = false; - int ret; + struct pmu *pmu = group_event->pmu; + u64 now = ctx->time; + bool simulate = false; if (group_event->state == PERF_EVENT_STATE_OFF) return 0; - /* Check if group transaction availabe */ - if (pmu->start_txn) - txn = true; + pmu->start_txn(pmu); - if (txn) - pmu->start_txn(pmu); - - if (event_sched_in(group_event, cpuctx, ctx)) + if (event_sched_in(group_event, cpuctx, ctx)) { + pmu->cancel_txn(pmu); return -EAGAIN; + } /* * Schedule in siblings as one group (if any): @@ -665,30 +849,39 @@ group_sched_in(struct perf_event *group_event, } } - if (!txn) - return 0; - - ret = pmu->commit_txn(pmu); - if (!ret) { - pmu->cancel_txn(pmu); + if (!pmu->commit_txn(pmu)) return 0; - } group_error: - if (txn) - pmu->cancel_txn(pmu); - /* * Groups can be scheduled in as one unit only, so undo any * partial group before returning: + * The events up to the failed event are scheduled out normally, + * tstamp_stopped will be updated. + * + * The failed events and the remaining siblings need to have + * their timings updated as if they had gone thru event_sched_in() + * and event_sched_out(). This is required to get consistent timings + * across the group. This also takes care of the case where the group + * could never be scheduled by ensuring tstamp_stopped is set to mark + * the time the event was actually stopped, such that time delta + * calculation in update_event_times() is correct. */ list_for_each_entry(event, &group_event->sibling_list, group_entry) { if (event == partial_group) - break; - event_sched_out(event, cpuctx, ctx); + simulate = true; + + if (simulate) { + event->tstamp_running += now - event->tstamp_stopped; + event->tstamp_stopped = now; + } else { + event_sched_out(event, cpuctx, ctx); + } } event_sched_out(group_event, cpuctx, ctx); + pmu->cancel_txn(pmu); + return -EAGAIN; } @@ -726,10 +919,13 @@ static int group_can_go_on(struct perf_event *event, static void add_event_to_ctx(struct perf_event *event, struct perf_event_context *ctx) { + u64 tstamp = perf_event_time(event); + list_add_event(event, ctx); - event->tstamp_enabled = ctx->time; - event->tstamp_running = ctx->time; - event->tstamp_stopped = ctx->time; + perf_group_attach(event); + event->tstamp_enabled = tstamp; + event->tstamp_running = tstamp; + event->tstamp_stopped = tstamp; } /* @@ -739,10 +935,10 @@ static void add_event_to_ctx(struct perf_event *event, */ 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; + struct perf_cpu_context *cpuctx = __get_cpu_context(ctx); int err; /* @@ -762,15 +958,9 @@ static void __perf_install_in_context(void *info) 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); - if (event->cpu != -1 && event->cpu != smp_processor_id()) + if (!event_filter_match(event)) goto unlock; /* @@ -805,12 +995,7 @@ static void __perf_install_in_context(void *info) } } - if (!err && !ctx->task && cpuctx->max_pertask) - cpuctx->max_pertask--; - - unlock: - perf_enable(); - +unlock: raw_spin_unlock(&ctx->lock); } @@ -833,6 +1018,8 @@ perf_install_in_context(struct perf_event_context *ctx, { struct task_struct *task = ctx->task; + event->ctx = ctx; + if (!task) { /* * Per cpu events are installed via an smp call and @@ -878,13 +1065,14 @@ static void __perf_event_mark_enabled(struct perf_event *event, struct perf_event_context *ctx) { struct perf_event *sub; + u64 tstamp = perf_event_time(event); 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) + event->tstamp_enabled = tstamp - 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; + sub->tstamp_enabled = tstamp - sub->total_time_enabled; + } } /* @@ -893,9 +1081,9 @@ static void __perf_event_mark_enabled(struct perf_event *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; + struct perf_cpu_context *cpuctx = __get_cpu_context(ctx); int err; /* @@ -916,7 +1104,7 @@ static void __perf_event_enable(void *info) goto unlock; __perf_event_mark_enabled(event, ctx); - if (event->cpu != -1 && event->cpu != smp_processor_id()) + if (!event_filter_match(event)) goto unlock; /* @@ -929,12 +1117,10 @@ static void __perf_event_enable(void *info) if (!group_can_go_on(event, cpuctx, 1)) { err = -EEXIST; } else { - perf_disable(); if (event == leader) err = group_sched_in(event, cpuctx, ctx); else err = event_sched_in(event, cpuctx, ctx); - perf_enable(); } if (err) { @@ -950,7 +1136,7 @@ static void __perf_event_enable(void *info) } } - unlock: +unlock: raw_spin_unlock(&ctx->lock); } @@ -991,7 +1177,7 @@ void perf_event_enable(struct perf_event *event) if (event->state == PERF_EVENT_STATE_ERROR) event->state = PERF_EVENT_STATE_OFF; - retry: +retry: raw_spin_unlock_irq(&ctx->lock); task_oncpu_function_call(task, __perf_event_enable, event); @@ -1011,7 +1197,7 @@ void perf_event_enable(struct perf_event *event) if (event->state == PERF_EVENT_STATE_OFF) __perf_event_mark_enabled(event, ctx); - out: +out: raw_spin_unlock_irq(&ctx->lock); } @@ -1020,7 +1206,7 @@ static int perf_event_refresh(struct perf_event *event, int refresh) /* * not supported on inherited events */ - if (event->attr.inherit) + if (event->attr.inherit || !is_sampling_event(event)) return -EINVAL; atomic_add(refresh, &event->event_limit); @@ -1029,12 +1215,6 @@ static int perf_event_refresh(struct perf_event *event, int refresh) return 0; } -enum event_type_t { - EVENT_FLEXIBLE = 0x1, - EVENT_PINNED = 0x2, - EVENT_ALL = EVENT_FLEXIBLE | EVENT_PINNED, -}; - static void ctx_sched_out(struct perf_event_context *ctx, struct perf_cpu_context *cpuctx, enum event_type_t event_type) @@ -1042,26 +1222,26 @@ static void ctx_sched_out(struct perf_event_context *ctx, struct perf_event *event; raw_spin_lock(&ctx->lock); + perf_pmu_disable(ctx->pmu); ctx->is_active = 0; if (likely(!ctx->nr_events)) goto out; update_context_time(ctx); - perf_disable(); if (!ctx->nr_active) - goto out_enable; + goto out; - if (event_type & EVENT_PINNED) + if (event_type & EVENT_PINNED) { list_for_each_entry(event, &ctx->pinned_groups, group_entry) group_sched_out(event, cpuctx, ctx); + } - if (event_type & EVENT_FLEXIBLE) + if (event_type & EVENT_FLEXIBLE) { list_for_each_entry(event, &ctx->flexible_groups, group_entry) group_sched_out(event, cpuctx, ctx); - - out_enable: - perf_enable(); - out: + } +out: + perf_pmu_enable(ctx->pmu); raw_spin_unlock(&ctx->lock); } @@ -1116,9 +1296,9 @@ static void __perf_event_sync_stat(struct perf_event *event, * 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); + value = local64_read(&next_event->count); + value = local64_xchg(&event->count, value); + local64_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); @@ -1159,34 +1339,25 @@ static void perf_event_sync_stat(struct perf_event_context *ctx, } } -/* - * 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) +void perf_event_context_sched_out(struct task_struct *task, int ctxn, + struct task_struct *next) { - struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context); - struct perf_event_context *ctx = task->perf_event_ctxp; + struct perf_event_context *ctx = task->perf_event_ctxp[ctxn]; struct perf_event_context *next_ctx; struct perf_event_context *parent; + struct perf_cpu_context *cpuctx; int do_switch = 1; - perf_sw_event(PERF_COUNT_SW_CONTEXT_SWITCHES, 1, 1, NULL, 0); + if (likely(!ctx)) + return; - if (likely(!ctx || !cpuctx->task_ctx)) + cpuctx = __get_cpu_context(ctx); + if (!cpuctx->task_ctx) return; rcu_read_lock(); parent = rcu_dereference(ctx->parent_ctx); - next_ctx = next->perf_event_ctxp; + next_ctx = next->perf_event_ctxp[ctxn]; if (parent && next_ctx && rcu_dereference(next_ctx->parent_ctx) == parent) { /* @@ -1205,8 +1376,8 @@ void perf_event_task_sched_out(struct task_struct *task, * 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; + task->perf_event_ctxp[ctxn] = next_ctx; + next->perf_event_ctxp[ctxn] = ctx; ctx->task = next; next_ctx->task = task; do_switch = 0; @@ -1224,10 +1395,33 @@ void perf_event_task_sched_out(struct task_struct *task, } } +#define for_each_task_context_nr(ctxn) \ + for ((ctxn) = 0; (ctxn) < perf_nr_task_contexts; (ctxn)++) + +/* + * 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 ctxn; + + for_each_task_context_nr(ctxn) + perf_event_context_sched_out(task, ctxn, next); +} + static void task_ctx_sched_out(struct perf_event_context *ctx, enum event_type_t event_type) { - struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context); + struct perf_cpu_context *cpuctx = __get_cpu_context(ctx); if (!cpuctx->task_ctx) return; @@ -1242,14 +1436,6 @@ static void task_ctx_sched_out(struct perf_event_context *ctx, /* * Called with IRQs disabled */ -static void __perf_event_task_sched_out(struct perf_event_context *ctx) -{ - task_ctx_sched_out(ctx, EVENT_ALL); -} - -/* - * Called with IRQs disabled - */ static void cpu_ctx_sched_out(struct perf_cpu_context *cpuctx, enum event_type_t event_type) { @@ -1265,7 +1451,7 @@ ctx_pinned_sched_in(struct perf_event_context *ctx, list_for_each_entry(event, &ctx->pinned_groups, group_entry) { if (event->state <= PERF_EVENT_STATE_OFF) continue; - if (event->cpu != -1 && event->cpu != smp_processor_id()) + if (!event_filter_match(event)) continue; if (group_can_go_on(event, cpuctx, 1)) @@ -1297,12 +1483,13 @@ ctx_flexible_sched_in(struct perf_event_context *ctx, * Listen to the 'cpu' scheduling filter constraint * of events: */ - if (event->cpu != -1 && event->cpu != smp_processor_id()) + if (!event_filter_match(event)) continue; - if (group_can_go_on(event, cpuctx, can_add_hw)) + if (group_can_go_on(event, cpuctx, can_add_hw)) { if (group_sched_in(event, cpuctx, ctx)) can_add_hw = 0; + } } } @@ -1318,8 +1505,6 @@ ctx_sched_in(struct perf_event_context *ctx, 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. @@ -1331,8 +1516,7 @@ ctx_sched_in(struct perf_event_context *ctx, if (event_type & EVENT_FLEXIBLE) ctx_flexible_sched_in(ctx, cpuctx); - perf_enable(); - out: +out: raw_spin_unlock(&ctx->lock); } @@ -1344,43 +1528,28 @@ static void cpu_ctx_sched_in(struct perf_cpu_context *cpuctx, ctx_sched_in(ctx, cpuctx, event_type); } -static void task_ctx_sched_in(struct task_struct *task, +static void task_ctx_sched_in(struct perf_event_context *ctx, enum event_type_t event_type) { - struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context); - struct perf_event_context *ctx = task->perf_event_ctxp; + struct perf_cpu_context *cpuctx; - if (likely(!ctx)) - return; + cpuctx = __get_cpu_context(ctx); if (cpuctx->task_ctx == ctx) return; + ctx_sched_in(ctx, cpuctx, event_type); cpuctx->task_ctx = ctx; } -/* - * 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) -{ - struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context); - struct perf_event_context *ctx = task->perf_event_ctxp; - if (likely(!ctx)) - return; +void perf_event_context_sched_in(struct perf_event_context *ctx) +{ + struct perf_cpu_context *cpuctx; + cpuctx = __get_cpu_context(ctx); if (cpuctx->task_ctx == ctx) return; - perf_disable(); - + perf_pmu_disable(ctx->pmu); /* * We want to keep the following priority order: * cpu pinned (that don't need to move), task pinned, @@ -1394,7 +1563,37 @@ void perf_event_task_sched_in(struct task_struct *task) cpuctx->task_ctx = ctx; - perf_enable(); + /* + * Since these rotations are per-cpu, we need to ensure the + * cpu-context we got scheduled on is actually rotating. + */ + perf_pmu_rotate_start(ctx->pmu); + perf_pmu_enable(ctx->pmu); +} + +/* + * 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) +{ + struct perf_event_context *ctx; + int ctxn; + + for_each_task_context_nr(ctxn) { + ctx = task->perf_event_ctxp[ctxn]; + if (likely(!ctx)) + continue; + + perf_event_context_sched_in(ctx); + } } #define MAX_INTERRUPTS (~0ULL) @@ -1468,29 +1667,16 @@ do { \ divisor = nsec * frequency; } - return div64_u64(dividend, divisor); -} - -static void perf_event_stop(struct perf_event *event) -{ - if (!event->pmu->stop) - return event->pmu->disable(event); - - return event->pmu->stop(event); -} - -static int perf_event_start(struct perf_event *event) -{ - if (!event->pmu->start) - return event->pmu->enable(event); + if (!divisor) + return dividend; - return event->pmu->start(event); + return div64_u64(dividend, divisor); } static void perf_adjust_period(struct perf_event *event, u64 nsec, u64 count) { struct hw_perf_event *hwc = &event->hw; - u64 period, sample_period; + s64 period, sample_period; s64 delta; period = perf_calculate_period(event, nsec, count); @@ -1505,16 +1691,14 @@ static void perf_adjust_period(struct perf_event *event, u64 nsec, u64 count) hwc->sample_period = sample_period; - if (atomic64_read(&hwc->period_left) > 8*sample_period) { - perf_disable(); - perf_event_stop(event); - atomic64_set(&hwc->period_left, 0); - perf_event_start(event); - perf_enable(); + if (local64_read(&hwc->period_left) > 8*sample_period) { + event->pmu->stop(event, PERF_EF_UPDATE); + local64_set(&hwc->period_left, 0); + event->pmu->start(event, PERF_EF_RELOAD); } } -static void perf_ctx_adjust_freq(struct perf_event_context *ctx) +static void perf_ctx_adjust_freq(struct perf_event_context *ctx, u64 period) { struct perf_event *event; struct hw_perf_event *hwc; @@ -1526,7 +1710,7 @@ static void perf_ctx_adjust_freq(struct perf_event_context *ctx) if (event->state != PERF_EVENT_STATE_ACTIVE) continue; - if (event->cpu != -1 && event->cpu != smp_processor_id()) + if (!event_filter_match(event)) continue; hwc = &event->hw; @@ -1539,23 +1723,19 @@ static void perf_ctx_adjust_freq(struct perf_event_context *ctx) */ if (interrupts == MAX_INTERRUPTS) { perf_log_throttle(event, 1); - perf_disable(); - event->pmu->unthrottle(event); - perf_enable(); + event->pmu->start(event, 0); } if (!event->attr.freq || !event->attr.sample_freq) continue; - perf_disable(); event->pmu->read(event); - now = atomic64_read(&event->count); + now = local64_read(&event->count); delta = now - hwc->freq_count_stamp; hwc->freq_count_stamp = now; if (delta > 0) - perf_adjust_period(event, TICK_NSEC, delta); - perf_enable(); + perf_adjust_period(event, period, delta); } raw_spin_unlock(&ctx->lock); } @@ -1567,38 +1747,48 @@ static void rotate_ctx(struct perf_event_context *ctx) { raw_spin_lock(&ctx->lock); - /* Rotate the first entry last of non-pinned groups */ - list_rotate_left(&ctx->flexible_groups); + /* + * Rotate the first entry last of non-pinned groups. Rotation might be + * disabled by the inheritance code. + */ + if (!ctx->rotate_disable) + list_rotate_left(&ctx->flexible_groups); raw_spin_unlock(&ctx->lock); } -void perf_event_task_tick(struct task_struct *curr) +/* + * perf_pmu_rotate_start() and perf_rotate_context() are fully serialized + * because they're strictly cpu affine and rotate_start is called with IRQs + * disabled, while rotate_context is called from IRQ context. + */ +static void perf_rotate_context(struct perf_cpu_context *cpuctx) { - struct perf_cpu_context *cpuctx; - struct perf_event_context *ctx; - int rotate = 0; + u64 interval = (u64)cpuctx->jiffies_interval * TICK_NSEC; + struct perf_event_context *ctx = NULL; + int rotate = 0, remove = 1; - if (!atomic_read(&nr_events)) - return; - - cpuctx = &__get_cpu_var(perf_cpu_context); - if (cpuctx->ctx.nr_events && - cpuctx->ctx.nr_events != cpuctx->ctx.nr_active) - rotate = 1; + if (cpuctx->ctx.nr_events) { + remove = 0; + if (cpuctx->ctx.nr_events != cpuctx->ctx.nr_active) + rotate = 1; + } - ctx = curr->perf_event_ctxp; - if (ctx && ctx->nr_events && ctx->nr_events != ctx->nr_active) - rotate = 1; + ctx = cpuctx->task_ctx; + if (ctx && ctx->nr_events) { + remove = 0; + if (ctx->nr_events != ctx->nr_active) + rotate = 1; + } - perf_ctx_adjust_freq(&cpuctx->ctx); + perf_pmu_disable(cpuctx->ctx.pmu); + perf_ctx_adjust_freq(&cpuctx->ctx, interval); if (ctx) - perf_ctx_adjust_freq(ctx); + perf_ctx_adjust_freq(ctx, interval); if (!rotate) - return; + goto done; - perf_disable(); cpu_ctx_sched_out(cpuctx, EVENT_FLEXIBLE); if (ctx) task_ctx_sched_out(ctx, EVENT_FLEXIBLE); @@ -1609,8 +1799,27 @@ void perf_event_task_tick(struct task_struct *curr) cpu_ctx_sched_in(cpuctx, EVENT_FLEXIBLE); if (ctx) - task_ctx_sched_in(curr, EVENT_FLEXIBLE); - perf_enable(); + task_ctx_sched_in(ctx, EVENT_FLEXIBLE); + +done: + if (remove) + list_del_init(&cpuctx->rotation_list); + + perf_pmu_enable(cpuctx->ctx.pmu); +} + +void perf_event_task_tick(void) +{ + struct list_head *head = &__get_cpu_var(rotation_list); + struct perf_cpu_context *cpuctx, *tmp; + + WARN_ON(!irqs_disabled()); + + list_for_each_entry_safe(cpuctx, tmp, head, rotation_list) { + if (cpuctx->jiffies_interval == 1 || + !(jiffies % cpuctx->jiffies_interval)) + perf_rotate_context(cpuctx); + } } static int event_enable_on_exec(struct perf_event *event, @@ -1632,20 +1841,18 @@ static int event_enable_on_exec(struct perf_event *event, * 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) +static void perf_event_enable_on_exec(struct perf_event_context *ctx) { - struct perf_event_context *ctx; struct perf_event *event; unsigned long flags; int enabled = 0; int ret; local_irq_save(flags); - ctx = task->perf_event_ctxp; if (!ctx || !ctx->nr_events) goto out; - __perf_event_task_sched_out(ctx); + task_ctx_sched_out(ctx, EVENT_ALL); raw_spin_lock(&ctx->lock); @@ -1669,8 +1876,8 @@ static void perf_event_enable_on_exec(struct task_struct *task) raw_spin_unlock(&ctx->lock); - perf_event_task_sched_in(task); - out: + perf_event_context_sched_in(ctx); +out: local_irq_restore(flags); } @@ -1679,9 +1886,9 @@ static void perf_event_enable_on_exec(struct task_struct *task) */ 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; + struct perf_cpu_context *cpuctx = __get_cpu_context(ctx); /* * If this is a task context, we need to check whether it is @@ -1701,6 +1908,11 @@ static void __perf_event_read(void *info) event->pmu->read(event); } +static inline u64 perf_event_count(struct perf_event *event) +{ + return local64_read(&event->count) + atomic64_read(&event->child_count); +} + static u64 perf_event_read(struct perf_event *event) { /* @@ -1715,20 +1927,234 @@ static u64 perf_event_read(struct perf_event *event) unsigned long flags; raw_spin_lock_irqsave(&ctx->lock, flags); - update_context_time(ctx); + /* + * may read while context is not active + * (e.g., thread is blocked), in that case + * we cannot update context time + */ + if (ctx->is_active) + update_context_time(ctx); update_event_times(event); raw_spin_unlock_irqrestore(&ctx->lock, flags); } - return atomic64_read(&event->count); + return perf_event_count(event); } /* - * Initialize the perf_event context in a task_struct: + * Callchain support */ + +struct callchain_cpus_entries { + struct rcu_head rcu_head; + struct perf_callchain_entry *cpu_entries[0]; +}; + +static DEFINE_PER_CPU(int, callchain_recursion[PERF_NR_CONTEXTS]); +static atomic_t nr_callchain_events; +static DEFINE_MUTEX(callchain_mutex); +struct callchain_cpus_entries *callchain_cpus_entries; + + +__weak void perf_callchain_kernel(struct perf_callchain_entry *entry, + struct pt_regs *regs) +{ +} + +__weak void perf_callchain_user(struct perf_callchain_entry *entry, + struct pt_regs *regs) +{ +} + +static void release_callchain_buffers_rcu(struct rcu_head *head) +{ + struct callchain_cpus_entries *entries; + int cpu; + + entries = container_of(head, struct callchain_cpus_entries, rcu_head); + + for_each_possible_cpu(cpu) + kfree(entries->cpu_entries[cpu]); + + kfree(entries); +} + +static void release_callchain_buffers(void) +{ + struct callchain_cpus_entries *entries; + + entries = callchain_cpus_entries; + rcu_assign_pointer(callchain_cpus_entries, NULL); + call_rcu(&entries->rcu_head, release_callchain_buffers_rcu); +} + +static int alloc_callchain_buffers(void) +{ + int cpu; + int size; + struct callchain_cpus_entries *entries; + + /* + * We can't use the percpu allocation API for data that can be + * accessed from NMI. Use a temporary manual per cpu allocation + * until that gets sorted out. + */ + size = sizeof(*entries) + sizeof(struct perf_callchain_entry *) * + num_possible_cpus(); + + entries = kzalloc(size, GFP_KERNEL); + if (!entries) + return -ENOMEM; + + size = sizeof(struct perf_callchain_entry) * PERF_NR_CONTEXTS; + + for_each_possible_cpu(cpu) { + entries->cpu_entries[cpu] = kmalloc_node(size, GFP_KERNEL, + cpu_to_node(cpu)); + if (!entries->cpu_entries[cpu]) + goto fail; + } + + rcu_assign_pointer(callchain_cpus_entries, entries); + + return 0; + +fail: + for_each_possible_cpu(cpu) + kfree(entries->cpu_entries[cpu]); + kfree(entries); + + return -ENOMEM; +} + +static int get_callchain_buffers(void) +{ + int err = 0; + int count; + + mutex_lock(&callchain_mutex); + + count = atomic_inc_return(&nr_callchain_events); + if (WARN_ON_ONCE(count < 1)) { + err = -EINVAL; + goto exit; + } + + if (count > 1) { + /* If the allocation failed, give up */ + if (!callchain_cpus_entries) + err = -ENOMEM; + goto exit; + } + + err = alloc_callchain_buffers(); + if (err) + release_callchain_buffers(); +exit: + mutex_unlock(&callchain_mutex); + + return err; +} + +static void put_callchain_buffers(void) +{ + if (atomic_dec_and_mutex_lock(&nr_callchain_events, &callchain_mutex)) { + release_callchain_buffers(); + mutex_unlock(&callchain_mutex); + } +} + +static int get_recursion_context(int *recursion) +{ + int rctx; + + if (in_nmi()) + rctx = 3; + else if (in_irq()) + rctx = 2; + else if (in_softirq()) + rctx = 1; + else + rctx = 0; + + if (recursion[rctx]) + return -1; + + recursion[rctx]++; + barrier(); + + return rctx; +} + +static inline void put_recursion_context(int *recursion, int rctx) +{ + barrier(); + recursion[rctx]--; +} + +static struct perf_callchain_entry *get_callchain_entry(int *rctx) +{ + int cpu; + struct callchain_cpus_entries *entries; + + *rctx = get_recursion_context(__get_cpu_var(callchain_recursion)); + if (*rctx == -1) + return NULL; + + entries = rcu_dereference(callchain_cpus_entries); + if (!entries) + return NULL; + + cpu = smp_processor_id(); + + return &entries->cpu_entries[cpu][*rctx]; +} + static void -__perf_event_init_context(struct perf_event_context *ctx, - struct task_struct *task) +put_callchain_entry(int rctx) +{ + put_recursion_context(__get_cpu_var(callchain_recursion), rctx); +} + +static struct perf_callchain_entry *perf_callchain(struct pt_regs *regs) +{ + int rctx; + struct perf_callchain_entry *entry; + + + entry = get_callchain_entry(&rctx); + if (rctx == -1) + return NULL; + + if (!entry) + goto exit_put; + + entry->nr = 0; + + if (!user_mode(regs)) { + perf_callchain_store(entry, PERF_CONTEXT_KERNEL); + perf_callchain_kernel(entry, regs); + if (current->mm) + regs = task_pt_regs(current); + else + regs = NULL; + } + + if (regs) { + perf_callchain_store(entry, PERF_CONTEXT_USER); + perf_callchain_user(entry, regs); + } + +exit_put: + put_callchain_entry(rctx); + + return entry; +} + +/* + * Initialize the perf_event context in a task_struct: + */ +static void __perf_event_init_context(struct perf_event_context *ctx) { raw_spin_lock_init(&ctx->lock); mutex_init(&ctx->mutex); @@ -1736,25 +2162,70 @@ __perf_event_init_context(struct perf_event_context *ctx, INIT_LIST_HEAD(&ctx->flexible_groups); 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) +static struct perf_event_context * +alloc_perf_context(struct pmu *pmu, struct task_struct *task) { struct perf_event_context *ctx; - struct perf_cpu_context *cpuctx; + + ctx = kzalloc(sizeof(struct perf_event_context), GFP_KERNEL); + if (!ctx) + return NULL; + + __perf_event_init_context(ctx); + if (task) { + ctx->task = task; + get_task_struct(task); + } + ctx->pmu = pmu; + + return ctx; +} + +static struct task_struct * +find_lively_task_by_vpid(pid_t vpid) +{ struct task_struct *task; - unsigned long flags; int err; - if (pid == -1 && cpu != -1) { + rcu_read_lock(); + if (!vpid) + task = current; + else + task = find_task_by_vpid(vpid); + if (task) + get_task_struct(task); + rcu_read_unlock(); + + if (!task) + return ERR_PTR(-ESRCH); + + /* Reuse ptrace permission checks for now. */ + err = -EACCES; + if (!ptrace_may_access(task, PTRACE_MODE_READ)) + goto errout; + + return task; +errout: + put_task_struct(task); + return ERR_PTR(err); + +} + +static struct perf_event_context * +find_get_context(struct pmu *pmu, struct task_struct *task, int cpu) +{ + struct perf_event_context *ctx; + struct perf_cpu_context *cpuctx; + unsigned long flags; + int ctxn, err; + + if (!task) { /* 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 >= nr_cpumask_bits) - 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 @@ -1763,67 +2234,60 @@ static struct perf_event_context *find_get_context(pid_t pid, int cpu) if (!cpu_online(cpu)) return ERR_PTR(-ENODEV); - cpuctx = &per_cpu(perf_cpu_context, cpu); + cpuctx = per_cpu_ptr(pmu->pmu_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)) + err = -EINVAL; + ctxn = pmu->task_ctx_nr; + if (ctxn < 0) goto errout; - retry: - ctx = perf_lock_task_context(task, &flags); +retry: + ctx = perf_lock_task_context(task, ctxn, &flags); if (ctx) { unclone_ctx(ctx); raw_spin_unlock_irqrestore(&ctx->lock, flags); } if (!ctx) { - ctx = kzalloc(sizeof(struct perf_event_context), GFP_KERNEL); + ctx = alloc_perf_context(pmu, task); 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. - */ + + err = 0; + mutex_lock(&task->perf_event_mutex); + /* + * If it has already passed perf_event_exit_task(). + * we must see PF_EXITING, it takes this mutex too. + */ + if (task->flags & PF_EXITING) + err = -ESRCH; + else if (task->perf_event_ctxp[ctxn]) + err = -EAGAIN; + else + rcu_assign_pointer(task->perf_event_ctxp[ctxn], ctx); + mutex_unlock(&task->perf_event_mutex); + + if (unlikely(err)) { + put_task_struct(task); kfree(ctx); - goto retry; + + if (err == -EAGAIN) + goto retry; + goto errout; } - get_task_struct(task); } - put_task_struct(task); return ctx; - errout: - put_task_struct(task); +errout: return ERR_PTR(err); } @@ -1840,31 +2304,36 @@ static void free_event_rcu(struct rcu_head *head) kfree(event); } -static void perf_pending_sync(struct perf_event *event); +static void perf_buffer_put(struct perf_buffer *buffer); static void free_event(struct perf_event *event) { - perf_pending_sync(event); + irq_work_sync(&event->pending); if (!event->parent) { - atomic_dec(&nr_events); - if (event->attr.mmap) + if (event->attach_state & PERF_ATTACH_TASK) + jump_label_dec(&perf_task_events); + if (event->attr.mmap || event->attr.mmap_data) 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->attr.sample_type & PERF_SAMPLE_CALLCHAIN) + put_callchain_buffers(); } - if (event->output) { - fput(event->output->filp); - event->output = NULL; + if (event->buffer) { + perf_buffer_put(event->buffer); + event->buffer = NULL; } if (event->destroy) event->destroy(event); - put_ctx(event->ctx); + if (event->ctx) + put_ctx(event->ctx); + call_rcu(&event->rcu_head, free_event_rcu); } @@ -1893,16 +2362,11 @@ int perf_event_release_kernel(struct perf_event *event) */ mutex_lock_nested(&ctx->mutex, SINGLE_DEPTH_NESTING); raw_spin_lock_irq(&ctx->lock); + perf_group_detach(event); list_del_event(event, ctx); - perf_destroy_group(event, ctx); raw_spin_unlock_irq(&ctx->lock); 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; @@ -1915,35 +2379,44 @@ EXPORT_SYMBOL_GPL(perf_event_release_kernel); static int perf_release(struct inode *inode, struct file *file) { struct perf_event *event = file->private_data; + struct task_struct *owner; 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); + rcu_read_lock(); + owner = ACCESS_ONCE(event->owner); + /* + * Matches the smp_wmb() in perf_event_exit_task(). If we observe + * !owner it means the list deletion is complete and we can indeed + * free this event, otherwise we need to serialize on + * owner->perf_event_mutex. + */ + smp_read_barrier_depends(); + if (owner) { + /* + * Since delayed_put_task_struct() also drops the last + * task reference we can safely take a new reference + * while holding the rcu_read_lock(). + */ + get_task_struct(owner); } + rcu_read_unlock(); - size += entry * nr; + if (owner) { + mutex_lock(&owner->perf_event_mutex); + /* + * We have to re-check the event->owner field, if it is cleared + * we raced with perf_event_exit_task(), acquiring the mutex + * ensured they're done, and we can proceed with freeing the + * event. + */ + if (event->owner) + list_del_init(&event->owner_entry); + mutex_unlock(&owner->perf_event_mutex); + put_task_struct(owner); + } - return size; + return perf_event_release_kernel(event); } u64 perf_event_read_value(struct perf_event *event, u64 *enabled, u64 *running) @@ -2060,7 +2533,7 @@ perf_read_hw(struct perf_event *event, char __user *buf, size_t count) if (event->state == PERF_EVENT_STATE_ERROR) return 0; - if (count < perf_event_read_size(event)) + if (count < event->read_size) return -ENOSPC; WARN_ON_ONCE(event->ctx->parent_ctx); @@ -2083,13 +2556,13 @@ perf_read(struct file *file, char __user *buf, size_t count, loff_t *ppos) static unsigned int perf_poll(struct file *file, poll_table *wait) { struct perf_event *event = file->private_data; - struct perf_mmap_data *data; + struct perf_buffer *buffer; unsigned int events = POLL_HUP; rcu_read_lock(); - data = rcu_dereference(event->data); - if (data) - events = atomic_xchg(&data->poll, 0); + buffer = rcu_dereference(event->buffer); + if (buffer) + events = atomic_xchg(&buffer->poll, 0); rcu_read_unlock(); poll_wait(file, &event->waitq, wait); @@ -2100,7 +2573,7 @@ static unsigned int perf_poll(struct file *file, poll_table *wait) static void perf_event_reset(struct perf_event *event) { (void)perf_event_read(event); - atomic64_set(&event->count, 0); + local64_set(&event->count, 0); perf_event_update_userpage(event); } @@ -2143,15 +2616,13 @@ static void perf_event_for_each(struct perf_event *event, 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) + if (!is_sampling_event(event)) return -EINVAL; - size = copy_from_user(&value, arg, sizeof(value)); - if (size != sizeof(value)) + if (copy_from_user(&value, arg, sizeof(value))) return -EFAULT; if (!value) @@ -2175,7 +2646,27 @@ unlock: return ret; } -static int perf_event_set_output(struct perf_event *event, int output_fd); +static const struct file_operations perf_fops; + +static struct perf_event *perf_fget_light(int fd, int *fput_needed) +{ + struct file *file; + + file = fget_light(fd, fput_needed); + if (!file) + return ERR_PTR(-EBADF); + + if (file->f_op != &perf_fops) { + fput_light(file, *fput_needed); + *fput_needed = 0; + return ERR_PTR(-EBADF); + } + + return file->private_data; +} + +static int perf_event_set_output(struct perf_event *event, + struct perf_event *output_event); 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) @@ -2202,7 +2693,23 @@ static long perf_ioctl(struct file *file, unsigned int cmd, unsigned long arg) return perf_event_period(event, (u64 __user *)arg); case PERF_EVENT_IOC_SET_OUTPUT: - return perf_event_set_output(event, arg); + { + struct perf_event *output_event = NULL; + int fput_needed = 0; + int ret; + + if (arg != -1) { + output_event = perf_fget_light(arg, &fput_needed); + if (IS_ERR(output_event)) + return PTR_ERR(output_event); + } + + ret = perf_event_set_output(event, output_event); + if (output_event) + fput_light(output_event->filp, fput_needed); + + return ret; + } case PERF_EVENT_IOC_SET_FILTER: return perf_event_set_filter(event, (void __user *)arg); @@ -2249,6 +2756,9 @@ int perf_event_task_disable(void) static int perf_event_index(struct perf_event *event) { + if (event->hw.state & PERF_HES_STOPPED) + return 0; + if (event->state != PERF_EVENT_STATE_ACTIVE) return 0; @@ -2263,14 +2773,14 @@ static int perf_event_index(struct perf_event *event) void perf_event_update_userpage(struct perf_event *event) { struct perf_event_mmap_page *userpg; - struct perf_mmap_data *data; + struct perf_buffer *buffer; rcu_read_lock(); - data = rcu_dereference(event->data); - if (!data) + buffer = rcu_dereference(event->buffer); + if (!buffer) goto unlock; - userpg = data->user_page; + userpg = buffer->user_page; /* * Disable preemption so as to not let the corresponding user-space @@ -2280,9 +2790,9 @@ void perf_event_update_userpage(struct perf_event *event) ++userpg->lock; barrier(); userpg->index = perf_event_index(event); - userpg->offset = atomic64_read(&event->count); + userpg->offset = perf_event_count(event); if (event->state == PERF_EVENT_STATE_ACTIVE) - userpg->offset -= atomic64_read(&event->hw.prev_count); + userpg->offset -= local64_read(&event->hw.prev_count); userpg->time_enabled = event->total_time_enabled + atomic64_read(&event->child_total_time_enabled); @@ -2297,9 +2807,23 @@ unlock: rcu_read_unlock(); } -static unsigned long perf_data_size(struct perf_mmap_data *data) +static unsigned long perf_data_size(struct perf_buffer *buffer); + +static void +perf_buffer_init(struct perf_buffer *buffer, long watermark, int flags) { - return data->nr_pages << (PAGE_SHIFT + data->data_order); + long max_size = perf_data_size(buffer); + + if (watermark) + buffer->watermark = min(max_size, watermark); + + if (!buffer->watermark) + buffer->watermark = max_size / 2; + + if (flags & PERF_BUFFER_WRITABLE) + buffer->writable = 1; + + atomic_set(&buffer->refcount, 1); } #ifndef CONFIG_PERF_USE_VMALLOC @@ -2309,56 +2833,68 @@ static unsigned long perf_data_size(struct perf_mmap_data *data) */ static struct page * -perf_mmap_to_page(struct perf_mmap_data *data, unsigned long pgoff) +perf_mmap_to_page(struct perf_buffer *buffer, unsigned long pgoff) { - if (pgoff > data->nr_pages) + if (pgoff > buffer->nr_pages) return NULL; if (pgoff == 0) - return virt_to_page(data->user_page); + return virt_to_page(buffer->user_page); - return virt_to_page(data->data_pages[pgoff - 1]); + return virt_to_page(buffer->data_pages[pgoff - 1]); } -static struct perf_mmap_data * -perf_mmap_data_alloc(struct perf_event *event, int nr_pages) +static void *perf_mmap_alloc_page(int cpu) { - struct perf_mmap_data *data; + struct page *page; + int node; + + node = (cpu == -1) ? cpu : cpu_to_node(cpu); + page = alloc_pages_node(node, GFP_KERNEL | __GFP_ZERO, 0); + if (!page) + return NULL; + + return page_address(page); +} + +static struct perf_buffer * +perf_buffer_alloc(int nr_pages, long watermark, int cpu, int flags) +{ + struct perf_buffer *buffer; unsigned long size; int i; - WARN_ON(atomic_read(&event->mmap_count)); - - size = sizeof(struct perf_mmap_data); + size = sizeof(struct perf_buffer); size += nr_pages * sizeof(void *); - data = kzalloc(size, GFP_KERNEL); - if (!data) + buffer = kzalloc(size, GFP_KERNEL); + if (!buffer) goto fail; - data->user_page = (void *)get_zeroed_page(GFP_KERNEL); - if (!data->user_page) + buffer->user_page = perf_mmap_alloc_page(cpu); + if (!buffer->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]) + buffer->data_pages[i] = perf_mmap_alloc_page(cpu); + if (!buffer->data_pages[i]) goto fail_data_pages; } - data->data_order = 0; - data->nr_pages = nr_pages; + buffer->nr_pages = nr_pages; - return data; + perf_buffer_init(buffer, watermark, flags); + + return buffer; fail_data_pages: for (i--; i >= 0; i--) - free_page((unsigned long)data->data_pages[i]); + free_page((unsigned long)buffer->data_pages[i]); - free_page((unsigned long)data->user_page); + free_page((unsigned long)buffer->user_page); fail_user_page: - kfree(data); + kfree(buffer); fail: return NULL; @@ -2372,14 +2908,19 @@ static void perf_mmap_free_page(unsigned long addr) __free_page(page); } -static void perf_mmap_data_free(struct perf_mmap_data *data) +static void perf_buffer_free(struct perf_buffer *buffer) { 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); + perf_mmap_free_page((unsigned long)buffer->user_page); + for (i = 0; i < buffer->nr_pages; i++) + perf_mmap_free_page((unsigned long)buffer->data_pages[i]); + kfree(buffer); +} + +static inline int page_order(struct perf_buffer *buffer) +{ + return 0; } #else @@ -2390,13 +2931,18 @@ static void perf_mmap_data_free(struct perf_mmap_data *data) * Required for architectures that have d-cache aliasing issues. */ +static inline int page_order(struct perf_buffer *buffer) +{ + return buffer->page_order; +} + static struct page * -perf_mmap_to_page(struct perf_mmap_data *data, unsigned long pgoff) +perf_mmap_to_page(struct perf_buffer *buffer, unsigned long pgoff) { - if (pgoff > (1UL << data->data_order)) + if (pgoff > (1UL << page_order(buffer))) return NULL; - return vmalloc_to_page((void *)data->user_page + pgoff * PAGE_SIZE); + return vmalloc_to_page((void *)buffer->user_page + pgoff * PAGE_SIZE); } static void perf_mmap_unmark_page(void *addr) @@ -2406,59 +2952,59 @@ static void perf_mmap_unmark_page(void *addr) page->mapping = NULL; } -static void perf_mmap_data_free_work(struct work_struct *work) +static void perf_buffer_free_work(struct work_struct *work) { - struct perf_mmap_data *data; + struct perf_buffer *buffer; void *base; int i, nr; - data = container_of(work, struct perf_mmap_data, work); - nr = 1 << data->data_order; + buffer = container_of(work, struct perf_buffer, work); + nr = 1 << page_order(buffer); - base = data->user_page; + base = buffer->user_page; for (i = 0; i < nr + 1; i++) perf_mmap_unmark_page(base + (i * PAGE_SIZE)); vfree(base); - kfree(data); + kfree(buffer); } -static void perf_mmap_data_free(struct perf_mmap_data *data) +static void perf_buffer_free(struct perf_buffer *buffer) { - schedule_work(&data->work); + schedule_work(&buffer->work); } -static struct perf_mmap_data * -perf_mmap_data_alloc(struct perf_event *event, int nr_pages) +static struct perf_buffer * +perf_buffer_alloc(int nr_pages, long watermark, int cpu, int flags) { - struct perf_mmap_data *data; + struct perf_buffer *buffer; unsigned long size; void *all_buf; - WARN_ON(atomic_read(&event->mmap_count)); - - size = sizeof(struct perf_mmap_data); + size = sizeof(struct perf_buffer); size += sizeof(void *); - data = kzalloc(size, GFP_KERNEL); - if (!data) + buffer = kzalloc(size, GFP_KERNEL); + if (!buffer) goto fail; - INIT_WORK(&data->work, perf_mmap_data_free_work); + INIT_WORK(&buffer->work, perf_buffer_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; + buffer->user_page = all_buf; + buffer->data_pages[0] = all_buf + PAGE_SIZE; + buffer->page_order = ilog2(nr_pages); + buffer->nr_pages = 1; - return data; + perf_buffer_init(buffer, watermark, flags); + + return buffer; fail_all_buf: - kfree(data); + kfree(buffer); fail: return NULL; @@ -2466,10 +3012,15 @@ fail: #endif +static unsigned long perf_data_size(struct perf_buffer *buffer) +{ + return buffer->nr_pages << (PAGE_SHIFT + page_order(buffer)); +} + 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; + struct perf_buffer *buffer; int ret = VM_FAULT_SIGBUS; if (vmf->flags & FAULT_FLAG_MKWRITE) { @@ -2479,14 +3030,14 @@ static int perf_mmap_fault(struct vm_area_struct *vma, struct vm_fault *vmf) } rcu_read_lock(); - data = rcu_dereference(event->data); - if (!data) + buffer = rcu_dereference(event->buffer); + if (!buffer) goto unlock; if (vmf->pgoff && (vmf->flags & FAULT_FLAG_WRITE)) goto unlock; - vmf->page = perf_mmap_to_page(data, vmf->pgoff); + vmf->page = perf_mmap_to_page(buffer, vmf->pgoff); if (!vmf->page) goto unlock; @@ -2501,41 +3052,35 @@ unlock: return ret; } -static void -perf_mmap_data_init(struct perf_event *event, struct perf_mmap_data *data) +static void perf_buffer_free_rcu(struct rcu_head *rcu_head) { - 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); - } + struct perf_buffer *buffer; - if (!data->watermark) - data->watermark = max_size / 2; - - - rcu_assign_pointer(event->data, data); + buffer = container_of(rcu_head, struct perf_buffer, rcu_head); + perf_buffer_free(buffer); } -static void perf_mmap_data_free_rcu(struct rcu_head *rcu_head) +static struct perf_buffer *perf_buffer_get(struct perf_event *event) { - struct perf_mmap_data *data; + struct perf_buffer *buffer; - data = container_of(rcu_head, struct perf_mmap_data, rcu_head); - perf_mmap_data_free(data); + rcu_read_lock(); + buffer = rcu_dereference(event->buffer); + if (buffer) { + if (!atomic_inc_not_zero(&buffer->refcount)) + buffer = NULL; + } + rcu_read_unlock(); + + return buffer; } -static void perf_mmap_data_release(struct perf_event *event) +static void perf_buffer_put(struct perf_buffer *buffer) { - struct perf_mmap_data *data = event->data; - - WARN_ON(atomic_read(&event->mmap_count)); + if (!atomic_dec_and_test(&buffer->refcount)) + return; - rcu_assign_pointer(event->data, NULL); - call_rcu(&data->rcu_head, perf_mmap_data_free_rcu); + call_rcu(&buffer->rcu_head, perf_buffer_free_rcu); } static void perf_mmap_open(struct vm_area_struct *vma) @@ -2549,15 +3094,18 @@ 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(); + unsigned long size = perf_data_size(event->buffer); + struct user_struct *user = event->mmap_user; + struct perf_buffer *buffer = event->buffer; atomic_long_sub((size >> PAGE_SHIFT) + 1, &user->locked_vm); - vma->vm_mm->locked_vm -= event->data->nr_locked; - perf_mmap_data_release(event); + vma->vm_mm->locked_vm -= event->mmap_locked; + rcu_assign_pointer(event->buffer, NULL); mutex_unlock(&event->mmap_mutex); + + perf_buffer_put(buffer); + free_uid(user); } } @@ -2574,11 +3122,19 @@ static int perf_mmap(struct file *file, struct vm_area_struct *vma) unsigned long user_locked, user_lock_limit; struct user_struct *user = current_user(); unsigned long locked, lock_limit; - struct perf_mmap_data *data; + struct perf_buffer *buffer; unsigned long vma_size; unsigned long nr_pages; long user_extra, extra; - int ret = 0; + int ret = 0, flags = 0; + + /* + * Don't allow mmap() of inherited per-task counters. This would + * create a performance issue due to all children writing to the + * same buffer. + */ + if (event->cpu == -1 && event->attr.inherit) + return -EINVAL; if (!(vma->vm_flags & VM_SHARED)) return -EINVAL; @@ -2587,7 +3143,7 @@ static int perf_mmap(struct file *file, struct vm_area_struct *vma) nr_pages = (vma_size / PAGE_SIZE) - 1; /* - * If we have data pages ensure they're a power-of-two number, so we + * If we have buffer 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)) @@ -2601,13 +3157,10 @@ static int perf_mmap(struct file *file, struct vm_area_struct *vma) 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) + if (event->buffer) { + if (event->buffer->nr_pages == nr_pages) + atomic_inc(&event->buffer->refcount); + else ret = -EINVAL; goto unlock; } @@ -2636,24 +3189,27 @@ static int perf_mmap(struct file *file, struct vm_area_struct *vma) goto unlock; } - WARN_ON(event->data); + WARN_ON(event->buffer); - data = perf_mmap_data_alloc(event, nr_pages); - ret = -ENOMEM; - if (!data) - goto unlock; + if (vma->vm_flags & VM_WRITE) + flags |= PERF_BUFFER_WRITABLE; - ret = 0; - perf_mmap_data_init(event, data); + buffer = perf_buffer_alloc(nr_pages, event->attr.wakeup_watermark, + event->cpu, flags); + if (!buffer) { + ret = -ENOMEM; + goto unlock; + } + rcu_assign_pointer(event->buffer, buffer); - 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; + event->mmap_locked = extra; + event->mmap_user = get_current_user(); + vma->vm_mm->locked_vm += event->mmap_locked; unlock: + if (!ret) + atomic_inc(&event->mmap_count); mutex_unlock(&event->mmap_mutex); vma->vm_flags |= VM_RESERVED; @@ -2706,16 +3262,7 @@ void perf_event_wakeup(struct perf_event *event) } } -/* - * 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) +static void perf_pending_event(struct irq_work *entry) { struct perf_event *event = container_of(entry, struct perf_event, pending); @@ -2731,104 +3278,6 @@ static void perf_pending_event(struct perf_pending_entry *entry) } } -#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; -} - -__weak -void perf_arch_fetch_caller_regs(struct pt_regs *regs, unsigned long ip, int skip) -{ -} - - /* * We assume there is only KVM supporting the callbacks. * Later on, we might change it to a list if there is @@ -2853,15 +3302,15 @@ EXPORT_SYMBOL_GPL(perf_unregister_guest_info_callbacks); /* * Output */ -static bool perf_output_space(struct perf_mmap_data *data, unsigned long tail, +static bool perf_output_space(struct perf_buffer *buffer, unsigned long tail, unsigned long offset, unsigned long head) { unsigned long mask; - if (!data->writable) + if (!buffer->writable) return true; - mask = perf_data_size(data) - 1; + mask = perf_data_size(buffer) - 1; offset = (offset - tail) & mask; head = (head - tail) & mask; @@ -2874,141 +3323,168 @@ static bool perf_output_space(struct perf_mmap_data *data, unsigned long tail, static void perf_output_wakeup(struct perf_output_handle *handle) { - atomic_set(&handle->data->poll, POLL_IN); + atomic_set(&handle->buffer->poll, POLL_IN); if (handle->nmi) { handle->event->pending_wakeup = 1; - perf_pending_queue(&handle->event->pending, - perf_pending_event); + irq_work_queue(&handle->event->pending); } 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 + * event 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. + * event completes. */ -static void perf_output_lock(struct perf_output_handle *handle) +static void perf_output_get_handle(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; + struct perf_buffer *buffer = handle->buffer; - cpu_relax(); - } + preempt_disable(); + local_inc(&buffer->nest); + handle->wakeup = local_read(&buffer->wakeup); } -static void perf_output_unlock(struct perf_output_handle *handle) +static void perf_output_put_handle(struct perf_output_handle *handle) { - struct perf_mmap_data *data = handle->data; + struct perf_buffer *buffer = handle->buffer; 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; + head = local_read(&buffer->head); /* - * NMI can happen here, which means we can miss a done_head update. + * IRQ/NMI can happen here, which means we can miss a head update. */ - cpu = atomic_xchg(&data->lock, -1); - WARN_ON_ONCE(cpu != smp_processor_id()); + if (!local_dec_and_test(&buffer->nest)) + goto out; /* - * Therefore we have to validate we did not indeed do so. + * Publish the known good head. Rely on the full barrier implied + * by atomic_dec_and_test() order the buffer->head read and this + * write. */ - 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(); + buffer->user_page->data_head = head; + /* + * Now check if we missed an update, rely on the (compiler) + * barrier in atomic_dec_and_test() to re-read buffer->head. + */ + if (unlikely(head != local_read(&buffer->head))) { + local_inc(&buffer->nest); goto again; } - if (atomic_xchg(&data->wakeup, 0)) + if (handle->wakeup != local_read(&buffer->wakeup)) perf_output_wakeup(handle); + out: - put_cpu(); + preempt_enable(); } -void perf_output_copy(struct perf_output_handle *handle, +__always_inline 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; + unsigned long size = min_t(unsigned long, handle->size, len); - 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(handle->addr, buf, size); - memcpy(pages[nr] + page_offset, buf, size); + len -= size; + handle->addr += size; + buf += size; + handle->size -= size; + if (!handle->size) { + struct perf_buffer *buffer = handle->buffer; - len -= size; - buf += size; - offset += size; + handle->page++; + handle->page &= buffer->nr_pages - 1; + handle->addr = buffer->data_pages[handle->page]; + handle->size = PAGE_SIZE << page_order(buffer); + } } while (len); +} + +static void __perf_event_header__init_id(struct perf_event_header *header, + struct perf_sample_data *data, + struct perf_event *event) +{ + u64 sample_type = event->attr.sample_type; - handle->offset = offset; + data->type = sample_type; + header->size += event->id_header_size; - /* - * 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); + 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); + } + + if (sample_type & PERF_SAMPLE_TIME) + data->time = perf_clock(); + + if (sample_type & PERF_SAMPLE_ID) + data->id = primary_event_id(event); + + if (sample_type & PERF_SAMPLE_STREAM_ID) + data->stream_id = event->id; + + if (sample_type & PERF_SAMPLE_CPU) { + data->cpu_entry.cpu = raw_smp_processor_id(); + data->cpu_entry.reserved = 0; + } +} + +static void perf_event_header__init_id(struct perf_event_header *header, + struct perf_sample_data *data, + struct perf_event *event) +{ + if (event->attr.sample_id_all) + __perf_event_header__init_id(header, data, event); +} + +static void __perf_event__output_id_sample(struct perf_output_handle *handle, + struct perf_sample_data *data) +{ + u64 sample_type = data->type; + + 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_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); +} + +static void perf_event__output_id_sample(struct perf_event *event, + struct perf_output_handle *handle, + struct perf_sample_data *sample) +{ + if (event->attr.sample_id_all) + __perf_event__output_id_sample(handle, sample); } 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; + struct perf_buffer *buffer; unsigned long tail, offset, head; int have_lost; + struct perf_sample_data sample_data; struct { struct perf_event_header header; u64 id; @@ -3022,27 +3498,27 @@ int perf_output_begin(struct perf_output_handle *handle, 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) + buffer = rcu_dereference(event->buffer); + if (!buffer) goto out; - handle->data = data; + handle->buffer = buffer; handle->event = event; handle->nmi = nmi; handle->sample = sample; - if (!data->nr_pages) - goto fail; + if (!buffer->nr_pages) + goto out; - have_lost = atomic_read(&data->lost); - if (have_lost) - size += sizeof(lost_event); + have_lost = local_read(&buffer->lost); + if (have_lost) { + lost_event.header.size = sizeof(lost_event); + perf_event_header__init_id(&lost_event.header, &sample_data, + event); + size += lost_event.header.size; + } - perf_output_lock(handle); + perf_output_get_handle(handle); do { /* @@ -3050,35 +3526,39 @@ int perf_output_begin(struct perf_output_handle *handle, * tail pointer. So that all reads will be completed before the * write is issued. */ - tail = ACCESS_ONCE(data->user_page->data_tail); + tail = ACCESS_ONCE(buffer->user_page->data_tail); smp_rmb(); - offset = head = atomic_long_read(&data->head); + offset = head = local_read(&buffer->head); head += size; - if (unlikely(!perf_output_space(data, tail, offset, head))) + if (unlikely(!perf_output_space(buffer, tail, offset, head))) goto fail; - } while (atomic_long_cmpxchg(&data->head, offset, head) != offset); + } while (local_cmpxchg(&buffer->head, offset, head) != offset); - handle->offset = offset; - handle->head = head; + if (head - local_read(&buffer->wakeup) > buffer->watermark) + local_add(buffer->watermark, &buffer->wakeup); - if (head - tail > data->watermark) - atomic_set(&data->wakeup, 1); + handle->page = offset >> (PAGE_SHIFT + page_order(buffer)); + handle->page &= buffer->nr_pages - 1; + handle->size = offset & ((PAGE_SIZE << page_order(buffer)) - 1); + handle->addr = buffer->data_pages[handle->page]; + handle->addr += handle->size; + handle->size = (PAGE_SIZE << page_order(buffer)) - handle->size; 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); + lost_event.lost = local_xchg(&buffer->lost, 0); perf_output_put(handle, lost_event); + perf_event__output_id_sample(event, handle, &sample_data); } return 0; fail: - atomic_inc(&data->lost); - perf_output_unlock(handle); + local_inc(&buffer->lost); + perf_output_put_handle(handle); out: rcu_read_unlock(); @@ -3088,58 +3568,37 @@ out: void perf_output_end(struct perf_output_handle *handle) { struct perf_event *event = handle->event; - struct perf_mmap_data *data = handle->data; + struct perf_buffer *buffer = handle->buffer; int wakeup_events = event->attr.wakeup_events; if (handle->sample && wakeup_events) { - int events = atomic_inc_return(&data->events); + int events = local_inc_return(&buffer->events); if (events >= wakeup_events) { - atomic_sub(wakeup_events, &data->events); - atomic_set(&data->wakeup, 1); + local_sub(wakeup_events, &buffer->events); + local_inc(&buffer->wakeup); } } - perf_output_unlock(handle); + perf_output_put_handle(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) + struct perf_event *event, + u64 enabled, u64 running) { u64 read_format = event->attr.read_format; u64 values[4]; int n = 0; - values[n++] = atomic64_read(&event->count); + values[n++] = perf_event_count(event); if (read_format & PERF_FORMAT_TOTAL_TIME_ENABLED) { - values[n++] = event->total_time_enabled + + values[n++] = enabled + atomic64_read(&event->child_total_time_enabled); } if (read_format & PERF_FORMAT_TOTAL_TIME_RUNNING) { - values[n++] = event->total_time_running + + values[n++] = running + atomic64_read(&event->child_total_time_running); } if (read_format & PERF_FORMAT_ID) @@ -3152,7 +3611,8 @@ static void perf_output_read_one(struct perf_output_handle *handle, * 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 *event, + u64 enabled, u64 running) { struct perf_event *leader = event->group_leader, *sub; u64 read_format = event->attr.read_format; @@ -3162,15 +3622,15 @@ static void perf_output_read_group(struct perf_output_handle *handle, values[n++] = 1 + leader->nr_siblings; if (read_format & PERF_FORMAT_TOTAL_TIME_ENABLED) - values[n++] = leader->total_time_enabled; + values[n++] = enabled; if (read_format & PERF_FORMAT_TOTAL_TIME_RUNNING) - values[n++] = leader->total_time_running; + values[n++] = running; if (leader != event) leader->pmu->read(leader); - values[n++] = atomic64_read(&leader->count); + values[n++] = perf_event_count(leader); if (read_format & PERF_FORMAT_ID) values[n++] = primary_event_id(leader); @@ -3182,7 +3642,7 @@ static void perf_output_read_group(struct perf_output_handle *handle, if (sub != event) sub->pmu->read(sub); - values[n++] = atomic64_read(&sub->count); + values[n++] = perf_event_count(sub); if (read_format & PERF_FORMAT_ID) values[n++] = primary_event_id(sub); @@ -3190,13 +3650,35 @@ static void perf_output_read_group(struct perf_output_handle *handle, } } +#define PERF_FORMAT_TOTAL_TIMES (PERF_FORMAT_TOTAL_TIME_ENABLED|\ + PERF_FORMAT_TOTAL_TIME_RUNNING) + static void perf_output_read(struct perf_output_handle *handle, struct perf_event *event) { + u64 enabled = 0, running = 0, now, ctx_time; + u64 read_format = event->attr.read_format; + + /* + * compute total_time_enabled, total_time_running + * based on snapshot values taken when the event + * was last scheduled in. + * + * we cannot simply called update_context_time() + * because of locking issue as we are called in + * NMI context + */ + if (read_format & PERF_FORMAT_TOTAL_TIMES) { + now = perf_clock(); + ctx_time = event->shadow_ctx_time + now; + enabled = ctx_time - event->tstamp_enabled; + running = ctx_time - event->tstamp_running; + } + if (event->attr.read_format & PERF_FORMAT_GROUP) - perf_output_read_group(handle, event); + perf_output_read_group(handle, event, enabled, running); else - perf_output_read_one(handle, event); + perf_output_read_one(handle, event, enabled, running); } void perf_output_sample(struct perf_output_handle *handle, @@ -3276,61 +3758,16 @@ void perf_prepare_sample(struct perf_event_header *header, { u64 sample_type = event->attr.sample_type; - data->type = sample_type; - header->type = PERF_RECORD_SAMPLE; - header->size = sizeof(*header); + header->size = sizeof(*header) + event->header_size; 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); - } + __perf_event_header__init_id(header, data, event); - 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_IP) + data->ip = perf_instruction_pointer(regs); if (sample_type & PERF_SAMPLE_CALLCHAIN) { int size = 1; @@ -3363,14 +3800,20 @@ static void perf_event_output(struct perf_event *event, int nmi, struct perf_output_handle handle; struct perf_event_header header; + /* protect the callchain buffers */ + rcu_read_lock(); + perf_prepare_sample(&header, data, event, regs); if (perf_output_begin(&handle, event, header.size, nmi, 1)) - return; + goto exit; perf_output_sample(&handle, &header, data, event); perf_output_end(&handle); + +exit: + rcu_read_unlock(); } /* @@ -3389,23 +3832,26 @@ perf_event_read_event(struct perf_event *event, struct task_struct *task) { struct perf_output_handle handle; + struct perf_sample_data sample; struct perf_read_event read_event = { .header = { .type = PERF_RECORD_READ, .misc = 0, - .size = sizeof(read_event) + perf_event_read_size(event), + .size = sizeof(read_event) + event->read_size, }, .pid = perf_event_pid(event, task), .tid = perf_event_tid(event, task), }; int ret; + perf_event_header__init_id(&read_event.header, &sample, event); 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_event__output_id_sample(event, &handle, &sample); perf_output_end(&handle); } @@ -3413,7 +3859,7 @@ perf_event_read_event(struct perf_event *event, /* * task tracking -- fork/exit * - * enabled by: attr.comm | attr.mmap | attr.task + * enabled by: attr.comm | attr.mmap | attr.mmap_data | attr.task */ struct perf_task_event { @@ -3435,23 +3881,16 @@ static void perf_event_task_output(struct perf_event *event, struct perf_task_event *task_event) { struct perf_output_handle handle; + struct perf_sample_data sample; struct task_struct *task = task_event->task; - unsigned long flags; - int size, ret; + int ret, size = task_event->event_id.header.size; - /* - * If this CPU attempts to acquire an rq lock held by a CPU spinning - * in perf_output_lock() from interrupt context, it's game over. - */ - local_irq_save(flags); + perf_event_header__init_id(&task_event->event_id.header, &sample, event); - size = task_event->event_id.header.size; - ret = perf_output_begin(&handle, event, size, 0, 0); - - if (ret) { - local_irq_restore(flags); - return; - } + ret = perf_output_begin(&handle, event, + task_event->event_id.header.size, 0, 0); + if (ret) + goto out; task_event->event_id.pid = perf_event_pid(event, task); task_event->event_id.ppid = perf_event_pid(event, current); @@ -3461,8 +3900,11 @@ static void perf_event_task_output(struct perf_event *event, perf_output_put(&handle, task_event->event_id); + perf_event__output_id_sample(event, &handle, &sample); + perf_output_end(&handle); - local_irq_restore(flags); +out: + task_event->event_id.header.size = size; } static int perf_event_task_match(struct perf_event *event) @@ -3470,10 +3912,11 @@ static int perf_event_task_match(struct perf_event *event) if (event->state < PERF_EVENT_STATE_INACTIVE) return 0; - if (event->cpu != -1 && event->cpu != smp_processor_id()) + if (!event_filter_match(event)) return 0; - if (event->attr.comm || event->attr.mmap || event->attr.task) + if (event->attr.comm || event->attr.mmap || + event->attr.mmap_data || event->attr.task) return 1; return 0; @@ -3493,16 +3936,29 @@ static void perf_event_task_ctx(struct perf_event_context *ctx, 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; + struct perf_event_context *ctx; + struct pmu *pmu; + int ctxn; rcu_read_lock(); - cpuctx = &get_cpu_var(perf_cpu_context); - perf_event_task_ctx(&cpuctx->ctx, task_event); - if (!ctx) - ctx = rcu_dereference(current->perf_event_ctxp); - if (ctx) - perf_event_task_ctx(ctx, task_event); - put_cpu_var(perf_cpu_context); + list_for_each_entry_rcu(pmu, &pmus, entry) { + cpuctx = get_cpu_ptr(pmu->pmu_cpu_context); + if (cpuctx->active_pmu != pmu) + goto next; + perf_event_task_ctx(&cpuctx->ctx, task_event); + + ctx = task_event->task_ctx; + if (!ctx) { + ctxn = pmu->task_ctx_nr; + if (ctxn < 0) + goto next; + ctx = rcu_dereference(current->perf_event_ctxp[ctxn]); + } + if (ctx) + perf_event_task_ctx(ctx, task_event); +next: + put_cpu_ptr(pmu->pmu_cpu_context); + } rcu_read_unlock(); } @@ -3563,11 +4019,16 @@ static void perf_event_comm_output(struct perf_event *event, struct perf_comm_event *comm_event) { struct perf_output_handle handle; + struct perf_sample_data sample; int size = comm_event->event_id.header.size; - int ret = perf_output_begin(&handle, event, size, 0, 0); + int ret; + + perf_event_header__init_id(&comm_event->event_id.header, &sample, event); + ret = perf_output_begin(&handle, event, + comm_event->event_id.header.size, 0, 0); if (ret) - return; + goto out; comm_event->event_id.pid = perf_event_pid(event, comm_event->task); comm_event->event_id.tid = perf_event_tid(event, comm_event->task); @@ -3575,7 +4036,12 @@ static void perf_event_comm_output(struct perf_event *event, perf_output_put(&handle, comm_event->event_id); perf_output_copy(&handle, comm_event->comm, comm_event->comm_size); + + perf_event__output_id_sample(event, &handle, &sample); + perf_output_end(&handle); +out: + comm_event->event_id.header.size = size; } static int perf_event_comm_match(struct perf_event *event) @@ -3583,7 +4049,7 @@ static int perf_event_comm_match(struct perf_event *event) if (event->state < PERF_EVENT_STATE_INACTIVE) return 0; - if (event->cpu != -1 && event->cpu != smp_processor_id()) + if (!event_filter_match(event)) return 0; if (event->attr.comm) @@ -3607,8 +4073,10 @@ 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]; + unsigned int size; + struct pmu *pmu; + int ctxn; memset(comm, 0, sizeof(comm)); strlcpy(comm, comm_event->task->comm, sizeof(comm)); @@ -3618,23 +4086,39 @@ static void perf_event_comm_event(struct perf_comm_event *comm_event) 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); - ctx = rcu_dereference(current->perf_event_ctxp); - if (ctx) - perf_event_comm_ctx(ctx, comm_event); - put_cpu_var(perf_cpu_context); + list_for_each_entry_rcu(pmu, &pmus, entry) { + cpuctx = get_cpu_ptr(pmu->pmu_cpu_context); + if (cpuctx->active_pmu != pmu) + goto next; + perf_event_comm_ctx(&cpuctx->ctx, comm_event); + + ctxn = pmu->task_ctx_nr; + if (ctxn < 0) + goto next; + + ctx = rcu_dereference(current->perf_event_ctxp[ctxn]); + if (ctx) + perf_event_comm_ctx(ctx, comm_event); +next: + put_cpu_ptr(pmu->pmu_cpu_context); + } rcu_read_unlock(); } void perf_event_comm(struct task_struct *task) { struct perf_comm_event comm_event; + struct perf_event_context *ctx; + int ctxn; - if (task->perf_event_ctxp) - perf_event_enable_on_exec(task); + for_each_task_context_nr(ctxn) { + ctx = task->perf_event_ctxp[ctxn]; + if (!ctx) + continue; + + perf_event_enable_on_exec(ctx); + } if (!atomic_read(&nr_comm_events)) return; @@ -3682,11 +4166,15 @@ static void perf_event_mmap_output(struct perf_event *event, struct perf_mmap_event *mmap_event) { struct perf_output_handle handle; + struct perf_sample_data sample; int size = mmap_event->event_id.header.size; - int ret = perf_output_begin(&handle, event, size, 0, 0); + int ret; + perf_event_header__init_id(&mmap_event->event_id.header, &sample, event); + ret = perf_output_begin(&handle, event, + mmap_event->event_id.header.size, 0, 0); if (ret) - return; + goto out; mmap_event->event_id.pid = perf_event_pid(event, current); mmap_event->event_id.tid = perf_event_tid(event, current); @@ -3694,31 +4182,39 @@ static void perf_event_mmap_output(struct perf_event *event, perf_output_put(&handle, mmap_event->event_id); perf_output_copy(&handle, mmap_event->file_name, mmap_event->file_size); + + perf_event__output_id_sample(event, &handle, &sample); + perf_output_end(&handle); +out: + mmap_event->event_id.header.size = size; } static int perf_event_mmap_match(struct perf_event *event, - struct perf_mmap_event *mmap_event) + struct perf_mmap_event *mmap_event, + int executable) { if (event->state < PERF_EVENT_STATE_INACTIVE) return 0; - if (event->cpu != -1 && event->cpu != smp_processor_id()) + if (!event_filter_match(event)) return 0; - if (event->attr.mmap) + if ((!executable && event->attr.mmap_data) || + (executable && 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_mmap_event *mmap_event, + int executable) { struct perf_event *event; list_for_each_entry_rcu(event, &ctx->event_list, event_entry) { - if (perf_event_mmap_match(event, mmap_event)) + if (perf_event_mmap_match(event, mmap_event, executable)) perf_event_mmap_output(event, mmap_event); } } @@ -3733,6 +4229,8 @@ static void perf_event_mmap_event(struct perf_mmap_event *mmap_event) char tmp[16]; char *buf = NULL; const char *name; + struct pmu *pmu; + int ctxn; memset(tmp, 0, sizeof(tmp)); @@ -3762,6 +4260,14 @@ static void perf_event_mmap_event(struct perf_mmap_event *mmap_event) if (!vma->vm_mm) { name = strncpy(tmp, "[vdso]", sizeof(tmp)); goto got_name; + } else if (vma->vm_start <= vma->vm_mm->start_brk && + vma->vm_end >= vma->vm_mm->brk) { + name = strncpy(tmp, "[heap]", sizeof(tmp)); + goto got_name; + } else if (vma->vm_start <= vma->vm_mm->start_stack && + vma->vm_end >= vma->vm_mm->start_stack) { + name = strncpy(tmp, "[stack]", sizeof(tmp)); + goto got_name; } name = strncpy(tmp, "//anon", sizeof(tmp)); @@ -3777,18 +4283,31 @@ got_name: 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); - ctx = rcu_dereference(current->perf_event_ctxp); - if (ctx) - perf_event_mmap_ctx(ctx, mmap_event); - put_cpu_var(perf_cpu_context); + list_for_each_entry_rcu(pmu, &pmus, entry) { + cpuctx = get_cpu_ptr(pmu->pmu_cpu_context); + if (cpuctx->active_pmu != pmu) + goto next; + perf_event_mmap_ctx(&cpuctx->ctx, mmap_event, + vma->vm_flags & VM_EXEC); + + ctxn = pmu->task_ctx_nr; + if (ctxn < 0) + goto next; + + ctx = rcu_dereference(current->perf_event_ctxp[ctxn]); + if (ctx) { + perf_event_mmap_ctx(ctx, mmap_event, + vma->vm_flags & VM_EXEC); + } +next: + put_cpu_ptr(pmu->pmu_cpu_context); + } rcu_read_unlock(); kfree(buf); } -void __perf_event_mmap(struct vm_area_struct *vma) +void perf_event_mmap(struct vm_area_struct *vma) { struct perf_mmap_event mmap_event; @@ -3823,6 +4342,7 @@ void __perf_event_mmap(struct vm_area_struct *vma) static void perf_log_throttle(struct perf_event *event, int enable) { struct perf_output_handle handle; + struct perf_sample_data sample; int ret; struct { @@ -3844,11 +4364,15 @@ static void perf_log_throttle(struct perf_event *event, int enable) if (enable) throttle_event.header.type = PERF_RECORD_UNTHROTTLE; - ret = perf_output_begin(&handle, event, sizeof(throttle_event), 1, 0); + perf_event_header__init_id(&throttle_event.header, &sample, event); + + ret = perf_output_begin(&handle, event, + throttle_event.header.size, 1, 0); if (ret) return; perf_output_put(&handle, throttle_event); + perf_event__output_id_sample(event, &handle, &sample); perf_output_end(&handle); } @@ -3864,7 +4388,12 @@ static int __perf_event_overflow(struct perf_event *event, int nmi, struct hw_perf_event *hwc = &event->hw; int ret = 0; - throttle = (throttle && event->pmu->unthrottle != NULL); + /* + * Non-sampling counters might still use the PMI to fold short + * hardware counters, ignore those. + */ + if (unlikely(!is_sampling_event(event))) + return 0; if (!throttle) { hwc->interrupts++; @@ -3908,8 +4437,7 @@ static int __perf_event_overflow(struct perf_event *event, int nmi, event->pending_kill = POLL_HUP; if (nmi) { event->pending_disable = 1; - perf_pending_queue(&event->pending, - perf_pending_event); + irq_work_queue(&event->pending); } else perf_event_disable(event); } @@ -3933,6 +4461,17 @@ int perf_event_overflow(struct perf_event *event, int nmi, * Generic software event infrastructure */ +struct swevent_htable { + struct swevent_hlist *swevent_hlist; + struct mutex hlist_mutex; + int hlist_refcount; + + /* Recursion avoidance in each contexts */ + int recursion[PERF_NR_CONTEXTS]; +}; + +static DEFINE_PER_CPU(struct swevent_htable, swevent_htable); + /* * We directly increment event->count and keep a second value in * event->hw.period_left to count intervals. This period event @@ -3950,14 +4489,14 @@ static u64 perf_swevent_set_period(struct perf_event *event) hwc->last_period = hwc->sample_period; again: - old = val = atomic64_read(&hwc->period_left); + old = val = local64_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) + if (local64_cmpxchg(&hwc->period_left, old, val) != old) goto again; return nr; @@ -3990,42 +4529,35 @@ static void perf_swevent_overflow(struct perf_event *event, u64 overflow, } } -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, +static void perf_swevent_event(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); + local64_add(nr, &event->count); if (!regs) return; - if (!hwc->sample_period) + if (!is_sampling_event(event)) 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)) + if (local64_add_negative(nr, &hwc->period_left)) return; perf_swevent_overflow(event, 0, nmi, data, regs); } -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 (event->hw.state & PERF_HES_STOPPED) + return 0; + if (regs) { if (event->attr.exclude_user && user_mode(regs)) return 1; @@ -4052,10 +4584,6 @@ static int perf_swevent_match(struct perf_event *event, 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; } @@ -4066,19 +4594,46 @@ static inline u64 swevent_hash(u64 type, u32 event_id) return hash_64(val, SWEVENT_HLIST_BITS); } -static struct hlist_head * -find_swevent_head(struct perf_cpu_context *ctx, u64 type, u32 event_id) +static inline struct hlist_head * +__find_swevent_head(struct swevent_hlist *hlist, u64 type, u32 event_id) +{ + u64 hash = swevent_hash(type, event_id); + + return &hlist->heads[hash]; +} + +/* For the read side: events when they trigger */ +static inline struct hlist_head * +find_swevent_head_rcu(struct swevent_htable *swhash, u64 type, u32 event_id) { - u64 hash; struct swevent_hlist *hlist; - hash = swevent_hash(type, event_id); + hlist = rcu_dereference(swhash->swevent_hlist); + if (!hlist) + return NULL; - hlist = rcu_dereference(ctx->swevent_hlist); + return __find_swevent_head(hlist, type, event_id); +} + +/* For the event head insertion and removal in the hlist */ +static inline struct hlist_head * +find_swevent_head(struct swevent_htable *swhash, struct perf_event *event) +{ + struct swevent_hlist *hlist; + u32 event_id = event->attr.config; + u64 type = event->attr.type; + + /* + * Event scheduling is always serialized against hlist allocation + * and release. Which makes the protected version suitable here. + * The context lock guarantees that. + */ + hlist = rcu_dereference_protected(swhash->swevent_hlist, + lockdep_is_held(&event->ctx->lock)); if (!hlist) return NULL; - return &hlist->heads[hash]; + return __find_swevent_head(hlist, type, event_id); } static void do_perf_sw_event(enum perf_type_id type, u32 event_id, @@ -4086,23 +4641,19 @@ static void do_perf_sw_event(enum perf_type_id type, u32 event_id, struct perf_sample_data *data, struct pt_regs *regs) { - struct perf_cpu_context *cpuctx; + struct swevent_htable *swhash = &__get_cpu_var(swevent_htable); struct perf_event *event; struct hlist_node *node; struct hlist_head *head; - cpuctx = &__get_cpu_var(perf_cpu_context); - rcu_read_lock(); - - head = find_swevent_head(cpuctx, type, event_id); - + head = find_swevent_head_rcu(swhash, type, event_id); if (!head) goto end; hlist_for_each_entry_rcu(event, node, head, hlist_entry) { if (perf_swevent_match(event, type, event_id, data, regs)) - perf_swevent_add(event, nr, nmi, data, regs); + perf_swevent_event(event, nr, nmi, data, regs); } end: rcu_read_unlock(); @@ -4110,39 +4661,18 @@ end: 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; + struct swevent_htable *swhash = &__get_cpu_var(swevent_htable); - if (cpuctx->recursion[rctx]) { - put_cpu_var(perf_cpu_context); - return -1; - } - - cpuctx->recursion[rctx]++; - barrier(); - - return rctx; + return get_recursion_context(swhash->recursion); } EXPORT_SYMBOL_GPL(perf_swevent_get_recursion_context); -void perf_swevent_put_recursion_context(int rctx) +inline 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); + struct swevent_htable *swhash = &__get_cpu_var(swevent_htable); + put_recursion_context(swhash->recursion, rctx); +} void __perf_sw_event(u32 event_id, u64 nr, int nmi, struct pt_regs *regs, u64 addr) @@ -4150,6 +4680,7 @@ void __perf_sw_event(u32 event_id, u64 nr, int nmi, struct perf_sample_data data; int rctx; + preempt_disable_notrace(); rctx = perf_swevent_get_recursion_context(); if (rctx < 0) return; @@ -4159,26 +4690,27 @@ void __perf_sw_event(u32 event_id, u64 nr, int nmi, do_perf_sw_event(PERF_TYPE_SOFTWARE, event_id, nr, nmi, &data, regs); perf_swevent_put_recursion_context(rctx); + preempt_enable_notrace(); } static void perf_swevent_read(struct perf_event *event) { } -static int perf_swevent_enable(struct perf_event *event) +static int perf_swevent_add(struct perf_event *event, int flags) { + struct swevent_htable *swhash = &__get_cpu_var(swevent_htable); struct hw_perf_event *hwc = &event->hw; - struct perf_cpu_context *cpuctx; struct hlist_head *head; - cpuctx = &__get_cpu_var(perf_cpu_context); - - if (hwc->sample_period) { + if (is_sampling_event(event)) { hwc->last_period = hwc->sample_period; perf_swevent_set_period(event); } - head = find_swevent_head(cpuctx, event->attr.type, event->attr.config); + hwc->state = !(flags & PERF_EF_START); + + head = find_swevent_head(swhash, event); if (WARN_ON_ONCE(!head)) return -EINVAL; @@ -4187,185 +4719,29 @@ static int perf_swevent_enable(struct perf_event *event) return 0; } -static void perf_swevent_disable(struct perf_event *event) +static void perf_swevent_del(struct perf_event *event, int flags) { hlist_del_rcu(&event->hlist_entry); } -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); - - perf_sample_data_init(&data, 0); - data.period = event->hw.last_period; - regs = get_irq_regs(); - - if (regs && !perf_exclude_event(event, 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_xchg(&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) +static void perf_swevent_start(struct perf_event *event, int flags) { - cpu_clock_perf_event_update(event); + event->hw.state = 0; } -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) +static void perf_swevent_stop(struct perf_event *event, int flags) { - 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; + event->hw.state = PERF_HES_STOPPED; } -static void task_clock_perf_event_disable(struct perf_event *event) +/* Deref the hlist from the update side */ +static inline struct swevent_hlist * +swevent_hlist_deref(struct swevent_htable *swhash) { - perf_swevent_cancel_hrtimer(event); - task_clock_perf_event_update(event, event->ctx->time); - + return rcu_dereference_protected(swhash->swevent_hlist, + lockdep_is_held(&swhash->hlist_mutex)); } -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, -}; - static void swevent_hlist_release_rcu(struct rcu_head *rcu_head) { struct swevent_hlist *hlist; @@ -4374,28 +4750,27 @@ static void swevent_hlist_release_rcu(struct rcu_head *rcu_head) kfree(hlist); } -static void swevent_hlist_release(struct perf_cpu_context *cpuctx) +static void swevent_hlist_release(struct swevent_htable *swhash) { - struct swevent_hlist *hlist; + struct swevent_hlist *hlist = swevent_hlist_deref(swhash); - if (!cpuctx->swevent_hlist) + if (!hlist) return; - hlist = cpuctx->swevent_hlist; - rcu_assign_pointer(cpuctx->swevent_hlist, NULL); + rcu_assign_pointer(swhash->swevent_hlist, NULL); call_rcu(&hlist->rcu_head, swevent_hlist_release_rcu); } static void swevent_hlist_put_cpu(struct perf_event *event, int cpu) { - struct perf_cpu_context *cpuctx = &per_cpu(perf_cpu_context, cpu); + struct swevent_htable *swhash = &per_cpu(swevent_htable, cpu); - mutex_lock(&cpuctx->hlist_mutex); + mutex_lock(&swhash->hlist_mutex); - if (!--cpuctx->hlist_refcount) - swevent_hlist_release(cpuctx); + if (!--swhash->hlist_refcount) + swevent_hlist_release(swhash); - mutex_unlock(&cpuctx->hlist_mutex); + mutex_unlock(&swhash->hlist_mutex); } static void swevent_hlist_put(struct perf_event *event) @@ -4413,12 +4788,12 @@ static void swevent_hlist_put(struct perf_event *event) static int swevent_hlist_get_cpu(struct perf_event *event, int cpu) { - struct perf_cpu_context *cpuctx = &per_cpu(perf_cpu_context, cpu); + struct swevent_htable *swhash = &per_cpu(swevent_htable, cpu); int err = 0; - mutex_lock(&cpuctx->hlist_mutex); + mutex_lock(&swhash->hlist_mutex); - if (!cpuctx->swevent_hlist && cpu_online(cpu)) { + if (!swevent_hlist_deref(swhash) && cpu_online(cpu)) { struct swevent_hlist *hlist; hlist = kzalloc(sizeof(*hlist), GFP_KERNEL); @@ -4426,11 +4801,11 @@ static int swevent_hlist_get_cpu(struct perf_event *event, int cpu) err = -ENOMEM; goto exit; } - rcu_assign_pointer(cpuctx->swevent_hlist, hlist); + rcu_assign_pointer(swhash->swevent_hlist, hlist); } - cpuctx->hlist_refcount++; - exit: - mutex_unlock(&cpuctx->hlist_mutex); + swhash->hlist_refcount++; +exit: + mutex_unlock(&swhash->hlist_mutex); return err; } @@ -4454,7 +4829,7 @@ static int swevent_hlist_get(struct perf_event *event) put_online_cpus(); return 0; - fail: +fail: for_each_possible_cpu(cpu) { if (cpu == failed_cpu) break; @@ -4465,27 +4840,65 @@ static int swevent_hlist_get(struct perf_event *event) return err; } -#ifdef CONFIG_EVENT_TRACING +atomic_t perf_swevent_enabled[PERF_COUNT_SW_MAX]; -void perf_tp_event(int event_id, u64 addr, u64 count, void *record, - int entry_size, struct pt_regs *regs) +static void sw_perf_event_destroy(struct perf_event *event) { - struct perf_sample_data data; - struct perf_raw_record raw = { - .size = entry_size, - .data = record, - }; + u64 event_id = event->attr.config; - perf_sample_data_init(&data, addr); - data.raw = &raw; + WARN_ON(event->parent); - /* Trace events already protected against recursion */ - do_perf_sw_event(PERF_TYPE_TRACEPOINT, event_id, count, 1, - &data, regs); + jump_label_dec(&perf_swevent_enabled[event_id]); + swevent_hlist_put(event); } -EXPORT_SYMBOL_GPL(perf_tp_event); -static int perf_tp_event_match(struct perf_event *event, +static int perf_swevent_init(struct perf_event *event) +{ + int event_id = event->attr.config; + + if (event->attr.type != PERF_TYPE_SOFTWARE) + return -ENOENT; + + switch (event_id) { + case PERF_COUNT_SW_CPU_CLOCK: + case PERF_COUNT_SW_TASK_CLOCK: + return -ENOENT; + + default: + break; + } + + if (event_id >= PERF_COUNT_SW_MAX) + return -ENOENT; + + if (!event->parent) { + int err; + + err = swevent_hlist_get(event); + if (err) + return err; + + jump_label_inc(&perf_swevent_enabled[event_id]); + event->destroy = sw_perf_event_destroy; + } + + return 0; +} + +static struct pmu perf_swevent = { + .task_ctx_nr = perf_sw_context, + + .event_init = perf_swevent_init, + .add = perf_swevent_add, + .del = perf_swevent_del, + .start = perf_swevent_start, + .stop = perf_swevent_stop, + .read = perf_swevent_read, +}; + +#ifdef CONFIG_EVENT_TRACING + +static int perf_tp_filter_match(struct perf_event *event, struct perf_sample_data *data) { void *record = data->raw->data; @@ -4495,36 +4908,81 @@ static int perf_tp_event_match(struct perf_event *event, return 0; } +static int perf_tp_event_match(struct perf_event *event, + struct perf_sample_data *data, + struct pt_regs *regs) +{ + /* + * All tracepoints are from kernel-space. + */ + if (event->attr.exclude_kernel) + return 0; + + if (!perf_tp_filter_match(event, data)) + return 0; + + return 1; +} + +void perf_tp_event(u64 addr, u64 count, void *record, int entry_size, + struct pt_regs *regs, struct hlist_head *head, int rctx) +{ + struct perf_sample_data data; + struct perf_event *event; + struct hlist_node *node; + + struct perf_raw_record raw = { + .size = entry_size, + .data = record, + }; + + perf_sample_data_init(&data, addr); + data.raw = &raw; + + hlist_for_each_entry_rcu(event, node, head, hlist_entry) { + if (perf_tp_event_match(event, &data, regs)) + perf_swevent_event(event, count, 1, &data, regs); + } + + perf_swevent_put_recursion_context(rctx); +} +EXPORT_SYMBOL_GPL(perf_tp_event); + static void tp_perf_event_destroy(struct perf_event *event) { - perf_trace_disable(event->attr.config); - swevent_hlist_put(event); + perf_trace_destroy(event); } -static const struct pmu *tp_perf_event_init(struct perf_event *event) +static int perf_tp_event_init(struct perf_event *event) { int err; - /* - * 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 (event->attr.type != PERF_TYPE_TRACEPOINT) + return -ENOENT; - if (perf_trace_enable(event->attr.config)) - return NULL; + err = perf_trace_init(event); + if (err) + return err; event->destroy = tp_perf_event_destroy; - err = swevent_hlist_get(event); - if (err) { - perf_trace_disable(event->attr.config); - return ERR_PTR(err); - } - return &perf_ops_generic; + return 0; +} + +static struct pmu perf_tracepoint = { + .task_ctx_nr = perf_sw_context, + + .event_init = perf_tp_event_init, + .add = perf_trace_add, + .del = perf_trace_del, + .start = perf_swevent_start, + .stop = perf_swevent_stop, + .read = perf_swevent_read, +}; + +static inline void perf_tp_register(void) +{ + perf_pmu_register(&perf_tracepoint, "tracepoint", PERF_TYPE_TRACEPOINT); } static int perf_event_set_filter(struct perf_event *event, void __user *arg) @@ -4552,15 +5010,8 @@ static void perf_event_free_filter(struct perf_event *event) #else -static int perf_tp_event_match(struct perf_event *event, - struct perf_sample_data *data) +static inline void perf_tp_register(void) { - 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) @@ -4575,105 +5026,508 @@ static void perf_event_free_filter(struct perf_event *event) #endif /* CONFIG_EVENT_TRACING */ #ifdef CONFIG_HAVE_HW_BREAKPOINT -static void bp_perf_event_destroy(struct perf_event *event) +void perf_bp_event(struct perf_event *bp, void *data) { - release_bp_slot(event); + struct perf_sample_data sample; + struct pt_regs *regs = data; + + perf_sample_data_init(&sample, bp->attr.bp_addr); + + if (!bp->hw.state && !perf_exclude_event(bp, regs)) + perf_swevent_event(bp, 1, 1, &sample, regs); } +#endif -static const struct pmu *bp_perf_event_init(struct perf_event *bp) +/* + * hrtimer based swevent callback + */ + +static enum hrtimer_restart perf_swevent_hrtimer(struct hrtimer *hrtimer) { - int err; + enum hrtimer_restart ret = HRTIMER_RESTART; + struct perf_sample_data data; + struct pt_regs *regs; + struct perf_event *event; + u64 period; - err = register_perf_hw_breakpoint(bp); - if (err) - return ERR_PTR(err); + event = container_of(hrtimer, struct perf_event, hw.hrtimer); + event->pmu->read(event); - bp->destroy = bp_perf_event_destroy; + perf_sample_data_init(&data, 0); + data.period = event->hw.last_period; + regs = get_irq_regs(); - return &perf_ops_bp; + if (regs && !perf_exclude_event(event, 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; } -void perf_bp_event(struct perf_event *bp, void *data) +static void perf_swevent_start_hrtimer(struct perf_event *event) { - struct perf_sample_data sample; - struct pt_regs *regs = data; + struct hw_perf_event *hwc = &event->hw; + s64 period; - perf_sample_data_init(&sample, bp->attr.bp_addr); + if (!is_sampling_event(event)) + return; + + hrtimer_init(&hwc->hrtimer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); + hwc->hrtimer.function = perf_swevent_hrtimer; - if (!perf_exclude_event(bp, regs)) - perf_swevent_add(bp, 1, 1, &sample, regs); + period = local64_read(&hwc->period_left); + if (period) { + if (period < 0) + period = 10000; + + local64_set(&hwc->period_left, 0); + } else { + period = max_t(u64, 10000, hwc->sample_period); + } + __hrtimer_start_range_ns(&hwc->hrtimer, + ns_to_ktime(period), 0, + HRTIMER_MODE_REL_PINNED, 0); } -#else -static const struct pmu *bp_perf_event_init(struct perf_event *bp) + +static void perf_swevent_cancel_hrtimer(struct perf_event *event) +{ + struct hw_perf_event *hwc = &event->hw; + + if (is_sampling_event(event)) { + ktime_t remaining = hrtimer_get_remaining(&hwc->hrtimer); + local64_set(&hwc->period_left, ktime_to_ns(remaining)); + + hrtimer_cancel(&hwc->hrtimer); + } +} + +/* + * Software event: cpu wall time clock + */ + +static void cpu_clock_event_update(struct perf_event *event) +{ + s64 prev; + u64 now; + + now = local_clock(); + prev = local64_xchg(&event->hw.prev_count, now); + local64_add(now - prev, &event->count); +} + +static void cpu_clock_event_start(struct perf_event *event, int flags) +{ + local64_set(&event->hw.prev_count, local_clock()); + perf_swevent_start_hrtimer(event); +} + +static void cpu_clock_event_stop(struct perf_event *event, int flags) +{ + perf_swevent_cancel_hrtimer(event); + cpu_clock_event_update(event); +} + +static int cpu_clock_event_add(struct perf_event *event, int flags) +{ + if (flags & PERF_EF_START) + cpu_clock_event_start(event, flags); + + return 0; +} + +static void cpu_clock_event_del(struct perf_event *event, int flags) +{ + cpu_clock_event_stop(event, flags); +} + +static void cpu_clock_event_read(struct perf_event *event) +{ + cpu_clock_event_update(event); +} + +static int cpu_clock_event_init(struct perf_event *event) +{ + if (event->attr.type != PERF_TYPE_SOFTWARE) + return -ENOENT; + + if (event->attr.config != PERF_COUNT_SW_CPU_CLOCK) + return -ENOENT; + + return 0; +} + +static struct pmu perf_cpu_clock = { + .task_ctx_nr = perf_sw_context, + + .event_init = cpu_clock_event_init, + .add = cpu_clock_event_add, + .del = cpu_clock_event_del, + .start = cpu_clock_event_start, + .stop = cpu_clock_event_stop, + .read = cpu_clock_event_read, +}; + +/* + * Software event: task time clock + */ + +static void task_clock_event_update(struct perf_event *event, u64 now) +{ + u64 prev; + s64 delta; + + prev = local64_xchg(&event->hw.prev_count, now); + delta = now - prev; + local64_add(delta, &event->count); +} + +static void task_clock_event_start(struct perf_event *event, int flags) +{ + local64_set(&event->hw.prev_count, event->ctx->time); + perf_swevent_start_hrtimer(event); +} + +static void task_clock_event_stop(struct perf_event *event, int flags) +{ + perf_swevent_cancel_hrtimer(event); + task_clock_event_update(event, event->ctx->time); +} + +static int task_clock_event_add(struct perf_event *event, int flags) +{ + if (flags & PERF_EF_START) + task_clock_event_start(event, flags); + + return 0; +} + +static void task_clock_event_del(struct perf_event *event, int flags) +{ + task_clock_event_stop(event, PERF_EF_UPDATE); +} + +static void task_clock_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_event_update(event, time); +} + +static int task_clock_event_init(struct perf_event *event) +{ + if (event->attr.type != PERF_TYPE_SOFTWARE) + return -ENOENT; + + if (event->attr.config != PERF_COUNT_SW_TASK_CLOCK) + return -ENOENT; + + return 0; +} + +static struct pmu perf_task_clock = { + .task_ctx_nr = perf_sw_context, + + .event_init = task_clock_event_init, + .add = task_clock_event_add, + .del = task_clock_event_del, + .start = task_clock_event_start, + .stop = task_clock_event_stop, + .read = task_clock_event_read, +}; + +static void perf_pmu_nop_void(struct pmu *pmu) +{ +} + +static int perf_pmu_nop_int(struct pmu *pmu) +{ + return 0; +} + +static void perf_pmu_start_txn(struct pmu *pmu) +{ + perf_pmu_disable(pmu); +} + +static int perf_pmu_commit_txn(struct pmu *pmu) +{ + perf_pmu_enable(pmu); + return 0; +} + +static void perf_pmu_cancel_txn(struct pmu *pmu) +{ + perf_pmu_enable(pmu); +} + +/* + * Ensures all contexts with the same task_ctx_nr have the same + * pmu_cpu_context too. + */ +static void *find_pmu_context(int ctxn) +{ + struct pmu *pmu; + + if (ctxn < 0) + return NULL; + + list_for_each_entry(pmu, &pmus, entry) { + if (pmu->task_ctx_nr == ctxn) + return pmu->pmu_cpu_context; + } + return NULL; } -void perf_bp_event(struct perf_event *bp, void *regs) +static void update_pmu_context(struct pmu *pmu, struct pmu *old_pmu) { + int cpu; + + for_each_possible_cpu(cpu) { + struct perf_cpu_context *cpuctx; + + cpuctx = per_cpu_ptr(pmu->pmu_cpu_context, cpu); + + if (cpuctx->active_pmu == old_pmu) + cpuctx->active_pmu = pmu; + } } -#endif -atomic_t perf_swevent_enabled[PERF_COUNT_SW_MAX]; +static void free_pmu_context(struct pmu *pmu) +{ + struct pmu *i; -static void sw_perf_event_destroy(struct perf_event *event) + mutex_lock(&pmus_lock); + /* + * Like a real lame refcount. + */ + list_for_each_entry(i, &pmus, entry) { + if (i->pmu_cpu_context == pmu->pmu_cpu_context) { + update_pmu_context(i, pmu); + goto out; + } + } + + free_percpu(pmu->pmu_cpu_context); +out: + mutex_unlock(&pmus_lock); +} +static struct idr pmu_idr; + +static ssize_t +type_show(struct device *dev, struct device_attribute *attr, char *page) { - u64 event_id = event->attr.config; + struct pmu *pmu = dev_get_drvdata(dev); - WARN_ON(event->parent); + return snprintf(page, PAGE_SIZE-1, "%d\n", pmu->type); +} - atomic_dec(&perf_swevent_enabled[event_id]); - swevent_hlist_put(event); +static struct device_attribute pmu_dev_attrs[] = { + __ATTR_RO(type), + __ATTR_NULL, +}; + +static int pmu_bus_running; +static struct bus_type pmu_bus = { + .name = "event_source", + .dev_attrs = pmu_dev_attrs, +}; + +static void pmu_dev_release(struct device *dev) +{ + kfree(dev); } -static const struct pmu *sw_perf_event_init(struct perf_event *event) +static int pmu_dev_alloc(struct pmu *pmu) { - const struct pmu *pmu = NULL; - u64 event_id = event->attr.config; + int ret = -ENOMEM; + + pmu->dev = kzalloc(sizeof(struct device), GFP_KERNEL); + if (!pmu->dev) + goto out; + + device_initialize(pmu->dev); + ret = dev_set_name(pmu->dev, "%s", pmu->name); + if (ret) + goto free_dev; + + dev_set_drvdata(pmu->dev, pmu); + pmu->dev->bus = &pmu_bus; + pmu->dev->release = pmu_dev_release; + ret = device_add(pmu->dev); + if (ret) + goto free_dev; + +out: + return ret; + +free_dev: + put_device(pmu->dev); + goto out; +} + +static struct lock_class_key cpuctx_mutex; + +int perf_pmu_register(struct pmu *pmu, char *name, int type) +{ + int cpu, ret; + + mutex_lock(&pmus_lock); + ret = -ENOMEM; + pmu->pmu_disable_count = alloc_percpu(int); + if (!pmu->pmu_disable_count) + goto unlock; + + pmu->type = -1; + if (!name) + goto skip_type; + pmu->name = name; + + if (type < 0) { + int err = idr_pre_get(&pmu_idr, GFP_KERNEL); + if (!err) + goto free_pdc; + + err = idr_get_new_above(&pmu_idr, pmu, PERF_TYPE_MAX, &type); + if (err) { + ret = err; + goto free_pdc; + } + } + pmu->type = type; + + if (pmu_bus_running) { + ret = pmu_dev_alloc(pmu); + if (ret) + goto free_idr; + } + +skip_type: + pmu->pmu_cpu_context = find_pmu_context(pmu->task_ctx_nr); + if (pmu->pmu_cpu_context) + goto got_cpu_context; + + pmu->pmu_cpu_context = alloc_percpu(struct perf_cpu_context); + if (!pmu->pmu_cpu_context) + goto free_dev; + + for_each_possible_cpu(cpu) { + struct perf_cpu_context *cpuctx; + + cpuctx = per_cpu_ptr(pmu->pmu_cpu_context, cpu); + __perf_event_init_context(&cpuctx->ctx); + lockdep_set_class(&cpuctx->ctx.mutex, &cpuctx_mutex); + cpuctx->ctx.type = cpu_context; + cpuctx->ctx.pmu = pmu; + cpuctx->jiffies_interval = 1; + INIT_LIST_HEAD(&cpuctx->rotation_list); + cpuctx->active_pmu = pmu; + } + +got_cpu_context: + if (!pmu->start_txn) { + if (pmu->pmu_enable) { + /* + * If we have pmu_enable/pmu_disable calls, install + * transaction stubs that use that to try and batch + * hardware accesses. + */ + pmu->start_txn = perf_pmu_start_txn; + pmu->commit_txn = perf_pmu_commit_txn; + pmu->cancel_txn = perf_pmu_cancel_txn; + } else { + pmu->start_txn = perf_pmu_nop_void; + pmu->commit_txn = perf_pmu_nop_int; + pmu->cancel_txn = perf_pmu_nop_void; + } + } + + if (!pmu->pmu_enable) { + pmu->pmu_enable = perf_pmu_nop_void; + pmu->pmu_disable = perf_pmu_nop_void; + } + + list_add_rcu(&pmu->entry, &pmus); + ret = 0; +unlock: + mutex_unlock(&pmus_lock); + + return ret; + +free_dev: + device_del(pmu->dev); + put_device(pmu->dev); + +free_idr: + if (pmu->type >= PERF_TYPE_MAX) + idr_remove(&pmu_idr, pmu->type); + +free_pdc: + free_percpu(pmu->pmu_disable_count); + goto unlock; +} + +void perf_pmu_unregister(struct pmu *pmu) +{ + mutex_lock(&pmus_lock); + list_del_rcu(&pmu->entry); + mutex_unlock(&pmus_lock); /* - * 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. + * We dereference the pmu list under both SRCU and regular RCU, so + * synchronize against both of those. */ - switch (event_id) { - case PERF_COUNT_SW_CPU_CLOCK: - pmu = &perf_ops_cpu_clock; + synchronize_srcu(&pmus_srcu); + synchronize_rcu(); - 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; + free_percpu(pmu->pmu_disable_count); + if (pmu->type >= PERF_TYPE_MAX) + idr_remove(&pmu_idr, pmu->type); + device_del(pmu->dev); + put_device(pmu->dev); + free_pmu_context(pmu); +} - 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) { - int err; - - err = swevent_hlist_get(event); - if (err) - return ERR_PTR(err); +struct pmu *perf_init_event(struct perf_event *event) +{ + struct pmu *pmu = NULL; + int idx; - atomic_inc(&perf_swevent_enabled[event_id]); - event->destroy = sw_perf_event_destroy; + idx = srcu_read_lock(&pmus_srcu); + + rcu_read_lock(); + pmu = idr_find(&pmu_idr, event->attr.type); + rcu_read_unlock(); + if (pmu) + goto unlock; + + list_for_each_entry_rcu(pmu, &pmus, entry) { + int ret = pmu->event_init(event); + if (!ret) + goto unlock; + + if (ret != -ENOENT) { + pmu = ERR_PTR(ret); + goto unlock; } - pmu = &perf_ops_generic; - break; } + pmu = ERR_PTR(-ENOENT); +unlock: + srcu_read_unlock(&pmus_srcu, idx); return pmu; } @@ -4682,20 +5536,23 @@ static const struct pmu *sw_perf_event_init(struct perf_event *event) * 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_overflow_handler_t overflow_handler, - gfp_t gfpflags) -{ - const struct pmu *pmu; +perf_event_alloc(struct perf_event_attr *attr, int cpu, + struct task_struct *task, + struct perf_event *group_leader, + struct perf_event *parent_event, + perf_overflow_handler_t overflow_handler) +{ + struct pmu *pmu; struct perf_event *event; struct hw_perf_event *hwc; long err; - event = kzalloc(sizeof(*event), gfpflags); + if ((unsigned)cpu >= nr_cpu_ids) { + if (!task || cpu != -1) + return ERR_PTR(-EINVAL); + } + + event = kzalloc(sizeof(*event), GFP_KERNEL); if (!event) return ERR_PTR(-ENOMEM); @@ -4713,6 +5570,7 @@ perf_event_alloc(struct perf_event_attr *attr, INIT_LIST_HEAD(&event->event_entry); INIT_LIST_HEAD(&event->sibling_list); init_waitqueue_head(&event->waitq); + init_irq_work(&event->pending, perf_pending_event); mutex_init(&event->mmap_mutex); @@ -4720,7 +5578,6 @@ perf_event_alloc(struct perf_event_attr *attr, event->attr = *attr; event->group_leader = group_leader; event->pmu = NULL; - event->ctx = ctx; event->oncpu = -1; event->parent = parent_event; @@ -4730,9 +5587,20 @@ perf_event_alloc(struct perf_event_attr *attr, event->state = PERF_EVENT_STATE_INACTIVE; + if (task) { + event->attach_state = PERF_ATTACH_TASK; +#ifdef CONFIG_HAVE_HW_BREAKPOINT + /* + * hw_breakpoint is a bit difficult here.. + */ + if (attr->type == PERF_TYPE_BREAKPOINT) + event->hw.bp_target = task; +#endif + } + if (!overflow_handler && parent_event) overflow_handler = parent_event->overflow_handler; - + event->overflow_handler = overflow_handler; if (attr->disabled) @@ -4746,7 +5614,7 @@ perf_event_alloc(struct perf_event_attr *attr, hwc->sample_period = 1; hwc->last_period = hwc->sample_period; - atomic64_set(&hwc->period_left, hwc->sample_period); + local64_set(&hwc->period_left, hwc->sample_period); /* * we currently do not support PERF_FORMAT_GROUP on inherited events @@ -4754,29 +5622,8 @@ perf_event_alloc(struct perf_event_attr *attr, 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; - + pmu = perf_init_event(event); - default: - break; - } done: err = 0; if (!pmu) @@ -4794,13 +5641,21 @@ done: event->pmu = pmu; if (!event->parent) { - atomic_inc(&nr_events); - if (event->attr.mmap) + if (event->attach_state & PERF_ATTACH_TASK) + jump_label_inc(&perf_task_events); + if (event->attr.mmap || event->attr.mmap_data) atomic_inc(&nr_mmap_events); if (event->attr.comm) atomic_inc(&nr_comm_events); if (event->attr.task) atomic_inc(&nr_task_events); + if (event->attr.sample_type & PERF_SAMPLE_CALLCHAIN) { + err = get_callchain_buffers(); + if (err) { + free_event(event); + return ERR_PTR(err); + } + } } return event; @@ -4886,54 +5741,53 @@ err_size: goto out; } -static int perf_event_set_output(struct perf_event *event, int output_fd) +static int +perf_event_set_output(struct perf_event *event, struct perf_event *output_event) { - struct perf_event *output_event = NULL; - struct file *output_file = NULL; - struct perf_event *old_output; - int fput_needed = 0; + struct perf_buffer *buffer = NULL, *old_buffer = NULL; int ret = -EINVAL; - if (!output_fd) + if (!output_event) goto set; - output_file = fget_light(output_fd, &fput_needed); - if (!output_file) - return -EBADF; - - if (output_file->f_op != &perf_fops) + /* don't allow circular references */ + if (event == output_event) goto out; - output_event = output_file->private_data; - - /* Don't chain output fds */ - if (output_event->output) + /* + * Don't allow cross-cpu buffers + */ + if (output_event->cpu != event->cpu) goto out; - /* Don't set an output fd when we already have an output channel */ - if (event->data) + /* + * If its not a per-cpu buffer, it must be the same task. + */ + if (output_event->cpu == -1 && output_event->ctx != event->ctx) 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); + /* Can't redirect output if we've got an active mmap() */ + if (atomic_read(&event->mmap_count)) + goto unlock; - if (old_output) { - /* - * we need to make sure no existing perf_output_*() - * is still referencing this event. - */ - synchronize_rcu(); - fput(old_output->filp); + if (output_event) { + /* get the buffer we want to redirect to */ + buffer = perf_buffer_get(output_event); + if (!buffer) + goto unlock; } + old_buffer = event->buffer; + rcu_assign_pointer(event->buffer, buffer); ret = 0; +unlock: + mutex_unlock(&event->mmap_mutex); + + if (old_buffer) + perf_buffer_put(old_buffer); out: - fput_light(output_file, fput_needed); return ret; } @@ -4949,13 +5803,17 @@ 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 *group_leader = NULL, *output_event = NULL; + struct perf_event *event, *sibling; struct perf_event_attr attr; struct perf_event_context *ctx; struct file *event_file = NULL; struct file *group_file = NULL; + struct task_struct *task = NULL; + struct pmu *pmu; + int event_fd; + int move_group = 0; int fput_needed = 0; - int fput_needed2 = 0; int err; /* for future expandability... */ @@ -4976,91 +5834,183 @@ SYSCALL_DEFINE5(perf_event_open, return -EINVAL; } + event_fd = get_unused_fd_flags(O_RDWR); + if (event_fd < 0) + return event_fd; + + if (group_fd != -1) { + group_leader = perf_fget_light(group_fd, &fput_needed); + if (IS_ERR(group_leader)) { + err = PTR_ERR(group_leader); + goto err_fd; + } + group_file = group_leader->filp; + if (flags & PERF_FLAG_FD_OUTPUT) + output_event = group_leader; + if (flags & PERF_FLAG_FD_NO_GROUP) + group_leader = NULL; + } + + if (pid != -1) { + task = find_lively_task_by_vpid(pid); + if (IS_ERR(task)) { + err = PTR_ERR(task); + goto err_group_fd; + } + } + + event = perf_event_alloc(&attr, cpu, task, group_leader, NULL, NULL); + if (IS_ERR(event)) { + err = PTR_ERR(event); + goto err_task; + } + + /* + * Special case software events and allow them to be part of + * any hardware group. + */ + pmu = event->pmu; + + if (group_leader && + (is_software_event(event) != is_software_event(group_leader))) { + if (is_software_event(event)) { + /* + * If event and group_leader are not both a software + * event, and event is, then group leader is not. + * + * Allow the addition of software events to !software + * groups, this is safe because software events never + * fail to schedule. + */ + pmu = group_leader->pmu; + } else if (is_software_event(group_leader) && + (group_leader->group_flags & PERF_GROUP_SOFTWARE)) { + /* + * In case the group is a pure software group, and we + * try to add a hardware event, move the whole group to + * the hardware context. + */ + move_group = 1; + } + } + /* * Get the target context (task or percpu): */ - ctx = find_get_context(pid, cpu); - if (IS_ERR(ctx)) - return PTR_ERR(ctx); + ctx = find_get_context(pmu, task, cpu); + if (IS_ERR(ctx)) { + err = PTR_ERR(ctx); + goto err_alloc; + } /* * 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)) { + if (group_leader) { 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; + goto err_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; + if (move_group) { + if (group_leader->ctx->type != ctx->type) + goto err_context; + } else { + if (group_leader->ctx != ctx) + goto err_context; + } + /* * Only a group leader can be exclusive or pinned */ if (attr.exclusive || attr.pinned) - goto err_put_context; + goto err_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; + if (output_event) { + err = perf_event_set_output(event, output_event); + if (err) + goto err_context; + } - err = anon_inode_getfd("[perf_event]", &perf_fops, event, O_RDWR); - if (err < 0) - goto err_free_put_context; + event_file = anon_inode_getfile("[perf_event]", &perf_fops, event, O_RDWR); + if (IS_ERR(event_file)) { + err = PTR_ERR(event_file); + goto err_context; + } - event_file = fget_light(err, &fput_needed2); - if (!event_file) - goto err_free_put_context; + if (move_group) { + struct perf_event_context *gctx = group_leader->ctx; - if (flags & PERF_FLAG_FD_OUTPUT) { - err = perf_event_set_output(event, group_fd); - if (err) - goto err_fput_free_put_context; + mutex_lock(&gctx->mutex); + perf_event_remove_from_context(group_leader); + list_for_each_entry(sibling, &group_leader->sibling_list, + group_entry) { + perf_event_remove_from_context(sibling); + put_ctx(gctx); + } + mutex_unlock(&gctx->mutex); + put_ctx(gctx); } event->filp = event_file; WARN_ON_ONCE(ctx->parent_ctx); mutex_lock(&ctx->mutex); + + if (move_group) { + perf_install_in_context(ctx, group_leader, cpu); + get_ctx(ctx); + list_for_each_entry(sibling, &group_leader->sibling_list, + group_entry) { + perf_install_in_context(ctx, sibling, cpu); + get_ctx(ctx); + } + } + 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) - free_event(event); - -err_put_context: - if (err < 0) - put_ctx(ctx); + /* + * Precalculate sample_data sizes + */ + perf_event__header_size(event); + perf_event__id_header_size(event); + /* + * Drop the reference on the group_event after placing the + * new event on the sibling_list. This ensures destruction + * of the group leader will find the pointer to itself in + * perf_group_detach(). + */ fput_light(group_file, fput_needed); + fd_install(event_fd, event_file); + return event_fd; +err_context: + put_ctx(ctx); +err_alloc: + free_event(event); +err_task: + if (task) + put_task_struct(task); +err_group_fd: + fput_light(group_file, fput_needed); +err_fd: + put_unused_fd(event_fd); return err; } @@ -5069,32 +6019,31 @@ err_put_context: * * @attr: attributes of the counter to create * @cpu: cpu in which the counter is bound - * @pid: task to profile + * @task: task to profile (NULL for percpu) */ struct perf_event * perf_event_create_kernel_counter(struct perf_event_attr *attr, int cpu, - pid_t pid, + struct task_struct *task, perf_overflow_handler_t overflow_handler) { - struct perf_event *event; struct perf_event_context *ctx; + struct perf_event *event; 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, overflow_handler, GFP_KERNEL); + event = perf_event_alloc(attr, cpu, task, NULL, NULL, overflow_handler); if (IS_ERR(event)) { err = PTR_ERR(event); - goto err_put_context; + goto err; + } + + ctx = find_get_context(event->pmu, task, cpu); + if (IS_ERR(ctx)) { + err = PTR_ERR(ctx); + goto err_free; } event->filp = NULL; @@ -5104,120 +6053,15 @@ perf_event_create_kernel_counter(struct perf_event_attr *attr, int 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: +err_free: + free_event(event); +err: 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) { - u64 sample_period = parent_event->hw.sample_period; - struct hw_perf_event *hwc = &child_event->hw; - - hwc->sample_period = sample_period; - hwc->last_period = sample_period; - - atomic64_set(&hwc->period_left, 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) { @@ -5227,12 +6071,12 @@ static void sync_child_event(struct perf_event *child_event, if (child_event->attr.inherit_stat) perf_event_read_event(child_event, child); - child_val = atomic64_read(&child_event->count); + child_val = perf_event_count(child_event); /* * Add back the child's count to the parent's count: */ - atomic64_add(child_val, &parent_event->count); + atomic64_add(child_val, &parent_event->child_count); atomic64_add(child_event->total_time_enabled, &parent_event->child_total_time_enabled); atomic64_add(child_event->total_time_running, @@ -5274,16 +6118,13 @@ __perf_event_exit_task(struct perf_event *child_event, } } -/* - * When a child task exits, feed back event values to parent events. - */ -void perf_event_exit_task(struct task_struct *child) +static void perf_event_exit_task_context(struct task_struct *child, int ctxn) { struct perf_event *child_event, *tmp; struct perf_event_context *child_ctx; unsigned long flags; - if (likely(!child->perf_event_ctxp)) { + if (likely(!child->perf_event_ctxp[ctxn])) { perf_event_task(child, NULL, 0); return; } @@ -5295,8 +6136,8 @@ void perf_event_exit_task(struct task_struct *child) * scheduled, so we are now safe from rescheduling changing * our context. */ - child_ctx = child->perf_event_ctxp; - __perf_event_task_sched_out(child_ctx); + child_ctx = rcu_dereference_raw(child->perf_event_ctxp[ctxn]); + task_ctx_sched_out(child_ctx, EVENT_ALL); /* * Take the context lock here so that if find_get_context is @@ -5304,7 +6145,7 @@ void perf_event_exit_task(struct task_struct *child) * incremented the context's refcount before we do put_ctx below. */ raw_spin_lock(&child_ctx->lock); - child->perf_event_ctxp = NULL; + child->perf_event_ctxp[ctxn] = NULL; /* * If this context is a clone; unclone it so it can't get * swapped to another process while we're removing all @@ -5357,6 +6198,33 @@ again: put_ctx(child_ctx); } +/* + * When a child task exits, feed back event values to parent events. + */ +void perf_event_exit_task(struct task_struct *child) +{ + struct perf_event *event, *tmp; + int ctxn; + + mutex_lock(&child->perf_event_mutex); + list_for_each_entry_safe(event, tmp, &child->perf_event_list, + owner_entry) { + list_del_init(&event->owner_entry); + + /* + * Ensure the list deletion is visible before we clear + * the owner, closes a race against perf_release() where + * we need to serialize on the owner->perf_event_mutex. + */ + smp_wmb(); + event->owner = NULL; + } + mutex_unlock(&child->perf_event_mutex); + + for_each_task_context_nr(ctxn) + perf_event_exit_task_context(child, ctxn); +} + static void perf_free_event(struct perf_event *event, struct perf_event_context *ctx) { @@ -5371,54 +6239,179 @@ static void perf_free_event(struct perf_event *event, fput(parent->filp); + perf_group_detach(event); list_del_event(event, ctx); free_event(event); } /* * free an unexposed, unused context as created by inheritance by - * init_task below, used by fork() in case of fail. + * perf_event_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_context *ctx; struct perf_event *event, *tmp; + int ctxn; - if (!ctx) - return; + for_each_task_context_nr(ctxn) { + ctx = task->perf_event_ctxp[ctxn]; + if (!ctx) + continue; - mutex_lock(&ctx->mutex); + mutex_lock(&ctx->mutex); again: - list_for_each_entry_safe(event, tmp, &ctx->pinned_groups, group_entry) - perf_free_event(event, ctx); + list_for_each_entry_safe(event, tmp, &ctx->pinned_groups, + group_entry) + perf_free_event(event, ctx); - list_for_each_entry_safe(event, tmp, &ctx->flexible_groups, - group_entry) - perf_free_event(event, ctx); + list_for_each_entry_safe(event, tmp, &ctx->flexible_groups, + group_entry) + perf_free_event(event, ctx); - if (!list_empty(&ctx->pinned_groups) || - !list_empty(&ctx->flexible_groups)) - goto again; + if (!list_empty(&ctx->pinned_groups) || + !list_empty(&ctx->flexible_groups)) + goto again; - mutex_unlock(&ctx->mutex); + mutex_unlock(&ctx->mutex); - put_ctx(ctx); + put_ctx(ctx); + } +} + +void perf_event_delayed_put(struct task_struct *task) +{ + int ctxn; + + for_each_task_context_nr(ctxn) + WARN_ON_ONCE(task->perf_event_ctxp[ctxn]); +} + +/* + * 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; + unsigned long flags; + + /* + * 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, + group_leader, parent_event, + NULL); + 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) { + u64 sample_period = parent_event->hw.sample_period; + struct hw_perf_event *hwc = &child_event->hw; + + hwc->sample_period = sample_period; + hwc->last_period = sample_period; + + local64_set(&hwc->period_left, sample_period); + } + + child_event->ctx = child_ctx; + child_event->overflow_handler = parent_event->overflow_handler; + + /* + * Precalculate sample_data sizes + */ + perf_event__header_size(child_event); + perf_event__id_header_size(child_event); + + /* + * Link it up in the child's context: + */ + raw_spin_lock_irqsave(&child_ctx->lock, flags); + add_event_to_ctx(child_event, child_ctx); + raw_spin_unlock_irqrestore(&child_ctx->lock, flags); + + /* + * 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 int inherit_task_group(struct perf_event *event, struct task_struct *parent, struct perf_event_context *parent_ctx, - struct task_struct *child, + struct task_struct *child, int ctxn, int *inherited_all) { int ret; - struct perf_event_context *child_ctx = child->perf_event_ctxp; + struct perf_event_context *child_ctx; if (!event->attr.inherit) { *inherited_all = 0; return 0; } + child_ctx = child->perf_event_ctxp[ctxn]; if (!child_ctx) { /* * This is executed from the parent task context, so @@ -5427,14 +6420,11 @@ inherit_task_group(struct perf_event *event, struct task_struct *parent, * child. */ - child_ctx = kzalloc(sizeof(struct perf_event_context), - GFP_KERNEL); + child_ctx = alloc_perf_context(event->pmu, child); if (!child_ctx) return -ENOMEM; - __perf_event_init_context(child_ctx, child); - child->perf_event_ctxp = child_ctx; - get_task_struct(child); + child->perf_event_ctxp[ctxn] = child_ctx; } ret = inherit_group(event, parent, parent_ctx, @@ -5446,32 +6436,27 @@ inherit_task_group(struct perf_event *event, struct task_struct *parent, return ret; } - /* * Initialize the perf_event context in task_struct */ -int perf_event_init_task(struct task_struct *child) +int perf_event_init_context(struct task_struct *child, int ctxn) { 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; + unsigned long flags; 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)) + if (likely(!parent->perf_event_ctxp[ctxn])) return 0; /* * 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); + parent_ctx = perf_pin_task_context(parent, ctxn); /* * No need to check if parent_ctx != NULL here; since we saw @@ -5491,31 +6476,42 @@ int perf_event_init_task(struct task_struct *child) * the list, not manipulating it: */ list_for_each_entry(event, &parent_ctx->pinned_groups, group_entry) { - ret = inherit_task_group(event, parent, parent_ctx, child, - &inherited_all); + ret = inherit_task_group(event, parent, parent_ctx, + child, ctxn, &inherited_all); if (ret) break; } + /* + * We can't hold ctx->lock when iterating the ->flexible_group list due + * to allocations, but we need to prevent rotation because + * rotate_ctx() will change the list from interrupt context. + */ + raw_spin_lock_irqsave(&parent_ctx->lock, flags); + parent_ctx->rotate_disable = 1; + raw_spin_unlock_irqrestore(&parent_ctx->lock, flags); + list_for_each_entry(event, &parent_ctx->flexible_groups, group_entry) { - ret = inherit_task_group(event, parent, parent_ctx, child, - &inherited_all); + ret = inherit_task_group(event, parent, parent_ctx, + child, ctxn, &inherited_all); if (ret) break; } - child_ctx = child->perf_event_ctxp; + raw_spin_lock_irqsave(&parent_ctx->lock, flags); + parent_ctx->rotate_disable = 0; + + child_ctx = child->perf_event_ctxp[ctxn]; if (child_ctx && 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. + * + * Note that if the parent is a clone, the holding of + * parent_ctx->lock avoids it from being uncloned. */ - cloned_ctx = rcu_dereference(parent_ctx->parent_ctx); + cloned_ctx = parent_ctx->parent_ctx; if (cloned_ctx) { child_ctx->parent_ctx = cloned_ctx; child_ctx->parent_gen = parent_ctx->parent_gen; @@ -5526,6 +6522,7 @@ int perf_event_init_task(struct task_struct *child) get_ctx(child_ctx->parent_ctx); } + raw_spin_unlock_irqrestore(&parent_ctx->lock, flags); mutex_unlock(&parent_ctx->mutex); perf_unpin_context(parent_ctx); @@ -5533,82 +6530,141 @@ int perf_event_init_task(struct task_struct *child) return ret; } +/* + * Initialize the perf_event context in task_struct + */ +int perf_event_init_task(struct task_struct *child) +{ + int ctxn, ret; + + memset(child->perf_event_ctxp, 0, sizeof(child->perf_event_ctxp)); + mutex_init(&child->perf_event_mutex); + INIT_LIST_HEAD(&child->perf_event_list); + + for_each_task_context_nr(ctxn) { + ret = perf_event_init_context(child, ctxn); + if (ret) + return ret; + } + + return 0; +} + static void __init perf_event_init_all_cpus(void) { + struct swevent_htable *swhash; int cpu; - struct perf_cpu_context *cpuctx; for_each_possible_cpu(cpu) { - cpuctx = &per_cpu(perf_cpu_context, cpu); - mutex_init(&cpuctx->hlist_mutex); - __perf_event_init_context(&cpuctx->ctx, NULL); + swhash = &per_cpu(swevent_htable, cpu); + mutex_init(&swhash->hlist_mutex); + INIT_LIST_HEAD(&per_cpu(rotation_list, cpu)); } } static void __cpuinit perf_event_init_cpu(int cpu) { - struct perf_cpu_context *cpuctx; + struct swevent_htable *swhash = &per_cpu(swevent_htable, cpu); - cpuctx = &per_cpu(perf_cpu_context, cpu); - - spin_lock(&perf_resource_lock); - cpuctx->max_pertask = perf_max_events - perf_reserved_percpu; - spin_unlock(&perf_resource_lock); - - mutex_lock(&cpuctx->hlist_mutex); - if (cpuctx->hlist_refcount > 0) { + mutex_lock(&swhash->hlist_mutex); + if (swhash->hlist_refcount > 0) { struct swevent_hlist *hlist; - hlist = kzalloc(sizeof(*hlist), GFP_KERNEL); - WARN_ON_ONCE(!hlist); - rcu_assign_pointer(cpuctx->swevent_hlist, hlist); + hlist = kzalloc_node(sizeof(*hlist), GFP_KERNEL, cpu_to_node(cpu)); + WARN_ON(!hlist); + rcu_assign_pointer(swhash->swevent_hlist, hlist); } - mutex_unlock(&cpuctx->hlist_mutex); + mutex_unlock(&swhash->hlist_mutex); } -#ifdef CONFIG_HOTPLUG_CPU -static void __perf_event_exit_cpu(void *info) +#if defined CONFIG_HOTPLUG_CPU || defined CONFIG_KEXEC +static void perf_pmu_rotate_stop(struct pmu *pmu) { - struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context); - struct perf_event_context *ctx = &cpuctx->ctx; + struct perf_cpu_context *cpuctx = this_cpu_ptr(pmu->pmu_cpu_context); + + WARN_ON(!irqs_disabled()); + + list_del_init(&cpuctx->rotation_list); +} + +static void __perf_event_exit_context(void *__info) +{ + struct perf_event_context *ctx = __info; struct perf_event *event, *tmp; + perf_pmu_rotate_stop(ctx->pmu); + list_for_each_entry_safe(event, tmp, &ctx->pinned_groups, group_entry) __perf_event_remove_from_context(event); list_for_each_entry_safe(event, tmp, &ctx->flexible_groups, group_entry) __perf_event_remove_from_context(event); } + +static void perf_event_exit_cpu_context(int cpu) +{ + struct perf_event_context *ctx; + struct pmu *pmu; + int idx; + + idx = srcu_read_lock(&pmus_srcu); + list_for_each_entry_rcu(pmu, &pmus, entry) { + ctx = &per_cpu_ptr(pmu->pmu_cpu_context, cpu)->ctx; + + mutex_lock(&ctx->mutex); + smp_call_function_single(cpu, __perf_event_exit_context, ctx, 1); + mutex_unlock(&ctx->mutex); + } + srcu_read_unlock(&pmus_srcu, idx); +} + 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; + struct swevent_htable *swhash = &per_cpu(swevent_htable, cpu); - mutex_lock(&cpuctx->hlist_mutex); - swevent_hlist_release(cpuctx); - mutex_unlock(&cpuctx->hlist_mutex); + mutex_lock(&swhash->hlist_mutex); + swevent_hlist_release(swhash); + mutex_unlock(&swhash->hlist_mutex); - mutex_lock(&ctx->mutex); - smp_call_function_single(cpu, __perf_event_exit_cpu, NULL, 1); - mutex_unlock(&ctx->mutex); + perf_event_exit_cpu_context(cpu); } #else static inline void perf_event_exit_cpu(int cpu) { } #endif +static int +perf_reboot(struct notifier_block *notifier, unsigned long val, void *v) +{ + int cpu; + + for_each_online_cpu(cpu) + perf_event_exit_cpu(cpu); + + return NOTIFY_OK; +} + +/* + * Run the perf reboot notifier at the very last possible moment so that + * the generic watchdog code runs as long as possible. + */ +static struct notifier_block perf_reboot_notifier = { + .notifier_call = perf_reboot, + .priority = INT_MIN, +}; + static int __cpuinit perf_cpu_notify(struct notifier_block *self, unsigned long action, void *hcpu) { unsigned int cpu = (long)hcpu; - switch (action) { + switch (action & ~CPU_TASKS_FROZEN) { case CPU_UP_PREPARE: - case CPU_UP_PREPARE_FROZEN: + case CPU_DOWN_FAILED: perf_event_init_cpu(cpu); break; + case CPU_UP_CANCELED: case CPU_DOWN_PREPARE: - case CPU_DOWN_PREPARE_FROZEN: perf_event_exit_cpu(cpu); break; @@ -5619,118 +6675,49 @@ perf_cpu_notify(struct notifier_block *self, unsigned long action, void *hcpu) 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_event_init_all_cpus(); - 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, - struct sysdev_class_attribute *attr, - char *buf) -{ - return sprintf(buf, "%d\n", perf_reserved_percpu); -} - -static ssize_t -perf_set_reserve_percpu(struct sysdev_class *class, - struct sysdev_class_attribute *attr, - 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; + int ret; - spin_lock(&perf_resource_lock); - perf_reserved_percpu = val; - for_each_online_cpu(cpu) { - cpuctx = &per_cpu(perf_cpu_context, cpu); - raw_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; - raw_spin_unlock_irq(&cpuctx->ctx.lock); - } - spin_unlock(&perf_resource_lock); + idr_init(&pmu_idr); - return count; -} + perf_event_init_all_cpus(); + init_srcu_struct(&pmus_srcu); + perf_pmu_register(&perf_swevent, "software", PERF_TYPE_SOFTWARE); + perf_pmu_register(&perf_cpu_clock, NULL, -1); + perf_pmu_register(&perf_task_clock, NULL, -1); + perf_tp_register(); + perf_cpu_notifier(perf_cpu_notify); + register_reboot_notifier(&perf_reboot_notifier); -static ssize_t perf_show_overcommit(struct sysdev_class *class, - struct sysdev_class_attribute *attr, - char *buf) -{ - return sprintf(buf, "%d\n", perf_overcommit); + ret = init_hw_breakpoint(); + WARN(ret, "hw_breakpoint initialization failed with: %d", ret); } -static ssize_t -perf_set_overcommit(struct sysdev_class *class, - struct sysdev_class_attribute *attr, - const char *buf, size_t count) +static int __init perf_event_sysfs_init(void) { - 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); + struct pmu *pmu; + int ret; - return count; -} + mutex_lock(&pmus_lock); -static SYSDEV_CLASS_ATTR( - reserve_percpu, - 0644, - perf_show_reserve_percpu, - perf_set_reserve_percpu - ); + ret = bus_register(&pmu_bus); + if (ret) + goto unlock; -static SYSDEV_CLASS_ATTR( - overcommit, - 0644, - perf_show_overcommit, - perf_set_overcommit - ); + list_for_each_entry(pmu, &pmus, entry) { + if (!pmu->name || pmu->type < 0) + continue; -static struct attribute *perfclass_attrs[] = { - &attr_reserve_percpu.attr, - &attr_overcommit.attr, - NULL -}; + ret = pmu_dev_alloc(pmu); + WARN(ret, "Failed to register pmu: %s, reason %d\n", pmu->name, ret); + } + pmu_bus_running = 1; + ret = 0; -static struct attribute_group perfclass_attr_group = { - .attrs = perfclass_attrs, - .name = "perf_events", -}; +unlock: + mutex_unlock(&pmus_lock); -static int __init perf_event_sysfs_init(void) -{ - return sysfs_create_group(&cpu_sysdev_class.kset.kobj, - &perfclass_attr_group); + return ret; } device_initcall(perf_event_sysfs_init); diff --git a/kernel/pid.c b/kernel/pid.c index aebb30d9c23..39b65b69584 100644 --- a/kernel/pid.c +++ b/kernel/pid.c @@ -122,6 +122,43 @@ static void free_pidmap(struct upid *upid) atomic_inc(&map->nr_free); } +/* + * If we started walking pids at 'base', is 'a' seen before 'b'? + */ +static int pid_before(int base, int a, int b) +{ + /* + * This is the same as saying + * + * (a - base + MAXUINT) % MAXUINT < (b - base + MAXUINT) % MAXUINT + * and that mapping orders 'a' and 'b' with respect to 'base'. + */ + return (unsigned)(a - base) < (unsigned)(b - base); +} + +/* + * We might be racing with someone else trying to set pid_ns->last_pid. + * We want the winner to have the "later" value, because if the + * "earlier" value prevails, then a pid may get reused immediately. + * + * Since pids rollover, it is not sufficient to just pick the bigger + * value. We have to consider where we started counting from. + * + * 'base' is the value of pid_ns->last_pid that we observed when + * we started looking for a pid. + * + * 'pid' is the pid that we eventually found. + */ +static void set_last_pid(struct pid_namespace *pid_ns, int base, int pid) +{ + int prev; + int last_write = base; + do { + prev = last_write; + last_write = cmpxchg(&pid_ns->last_pid, prev, pid); + } while ((prev != last_write) && (pid_before(base, last_write, pid))); +} + static int alloc_pidmap(struct pid_namespace *pid_ns) { int i, offset, max_scan, pid, last = pid_ns->last_pid; @@ -132,7 +169,12 @@ static int alloc_pidmap(struct pid_namespace *pid_ns) pid = RESERVED_PIDS; offset = pid & BITS_PER_PAGE_MASK; map = &pid_ns->pidmap[pid/BITS_PER_PAGE]; - max_scan = (pid_max + BITS_PER_PAGE - 1)/BITS_PER_PAGE - !offset; + /* + * If last_pid points into the middle of the map->page we + * want to scan this bitmap block twice, the second time + * we start with offset == 0 (or RESERVED_PIDS). + */ + max_scan = DIV_ROUND_UP(pid_max, BITS_PER_PAGE) - !offset; for (i = 0; i <= max_scan; ++i) { if (unlikely(!map->page)) { void *page = kzalloc(PAGE_SIZE, GFP_KERNEL); @@ -154,20 +196,12 @@ static int alloc_pidmap(struct pid_namespace *pid_ns) do { if (!test_and_set_bit(offset, map->page)) { atomic_dec(&map->nr_free); - pid_ns->last_pid = pid; + set_last_pid(pid_ns, last, pid); return pid; } offset = find_next_offset(map, offset); pid = mk_pid(pid_ns, map, offset); - /* - * find_next_offset() found a bit, the pid from it - * is in-bounds, and if we fell back to the last - * bitmap block and the final block was the same - * as the starting point, pid is before last_pid. - */ - } while (offset < BITS_PER_PAGE && pid < pid_max && - (i != max_scan || pid < last || - !((last+1) & BITS_PER_PAGE_MASK))); + } while (offset < BITS_PER_PAGE && pid < pid_max); } if (map < &pid_ns->pidmap[(pid_max-1)/BITS_PER_PAGE]) { ++map; @@ -367,7 +401,7 @@ struct task_struct *pid_task(struct pid *pid, enum pid_type type) struct task_struct *result = NULL; if (pid) { struct hlist_node *first; - first = rcu_dereference_check(pid->tasks[type].first, + first = rcu_dereference_check(hlist_first_rcu(&pid->tasks[type]), rcu_read_lock_held() || lockdep_tasklist_lock_is_held()); if (first) @@ -382,6 +416,7 @@ EXPORT_SYMBOL(pid_task); */ struct task_struct *find_task_by_pid_ns(pid_t nr, struct pid_namespace *ns) { + rcu_lockdep_assert(rcu_read_lock_held()); return pid_task(find_pid_ns(nr, ns), PIDTYPE_PID); } @@ -513,6 +548,13 @@ void __init pidhash_init(void) void __init pidmap_init(void) { + /* bump default and minimum pid_max based on number of cpus */ + pid_max = min(pid_max_max, max_t(int, pid_max, + PIDS_PER_CPU_DEFAULT * num_possible_cpus())); + pid_max_min = max_t(int, pid_max_min, + PIDS_PER_CPU_MIN * num_possible_cpus()); + pr_info("pid_max: default: %u minimum: %u\n", pid_max, pid_max_min); + init_pid_ns.pidmap[0].page = kzalloc(PAGE_SIZE, GFP_KERNEL); /* Reserve PID 0. We never call free_pidmap(0) */ set_bit(0, init_pid_ns.pidmap[0].page); diff --git a/kernel/pm_qos_params.c b/kernel/pm_qos_params.c index f42d3f737a3..aeaa7f84682 100644 --- a/kernel/pm_qos_params.c +++ b/kernel/pm_qos_params.c @@ -48,59 +48,49 @@ * or pm_qos_object list and pm_qos_objects need to happen with pm_qos_lock * held, taken with _irqsave. One lock to rule them all */ -struct pm_qos_request_list { - struct list_head list; - union { - s32 value; - s32 usec; - s32 kbps; - }; - int pm_qos_class; +enum pm_qos_type { + PM_QOS_MAX, /* return the largest value */ + PM_QOS_MIN /* return the smallest value */ }; -static s32 max_compare(s32 v1, s32 v2); -static s32 min_compare(s32 v1, s32 v2); - struct pm_qos_object { - struct pm_qos_request_list requests; + struct plist_head requests; struct blocking_notifier_head *notifiers; struct miscdevice pm_qos_power_miscdev; char *name; s32 default_value; - atomic_t target_value; - s32 (*comparitor)(s32, s32); + enum pm_qos_type type; }; +static DEFINE_SPINLOCK(pm_qos_lock); + static struct pm_qos_object null_pm_qos; static BLOCKING_NOTIFIER_HEAD(cpu_dma_lat_notifier); static struct pm_qos_object cpu_dma_pm_qos = { - .requests = {LIST_HEAD_INIT(cpu_dma_pm_qos.requests.list)}, + .requests = PLIST_HEAD_INIT(cpu_dma_pm_qos.requests, pm_qos_lock), .notifiers = &cpu_dma_lat_notifier, .name = "cpu_dma_latency", .default_value = 2000 * USEC_PER_SEC, - .target_value = ATOMIC_INIT(2000 * USEC_PER_SEC), - .comparitor = min_compare + .type = PM_QOS_MIN, }; static BLOCKING_NOTIFIER_HEAD(network_lat_notifier); static struct pm_qos_object network_lat_pm_qos = { - .requests = {LIST_HEAD_INIT(network_lat_pm_qos.requests.list)}, + .requests = PLIST_HEAD_INIT(network_lat_pm_qos.requests, pm_qos_lock), .notifiers = &network_lat_notifier, .name = "network_latency", .default_value = 2000 * USEC_PER_SEC, - .target_value = ATOMIC_INIT(2000 * USEC_PER_SEC), - .comparitor = min_compare + .type = PM_QOS_MIN }; static BLOCKING_NOTIFIER_HEAD(network_throughput_notifier); static struct pm_qos_object network_throughput_pm_qos = { - .requests = {LIST_HEAD_INIT(network_throughput_pm_qos.requests.list)}, + .requests = PLIST_HEAD_INIT(network_throughput_pm_qos.requests, pm_qos_lock), .notifiers = &network_throughput_notifier, .name = "network_throughput", .default_value = 0, - .target_value = ATOMIC_INIT(0), - .comparitor = max_compare + .type = PM_QOS_MAX, }; @@ -111,8 +101,6 @@ static struct pm_qos_object *pm_qos_array[] = { &network_throughput_pm_qos }; -static DEFINE_SPINLOCK(pm_qos_lock); - static ssize_t pm_qos_power_write(struct file *filp, const char __user *buf, size_t count, loff_t *f_pos); static int pm_qos_power_open(struct inode *inode, struct file *filp); @@ -122,48 +110,58 @@ static const struct file_operations pm_qos_power_fops = { .write = pm_qos_power_write, .open = pm_qos_power_open, .release = pm_qos_power_release, + .llseek = noop_llseek, }; -/* static helper functions */ -static s32 max_compare(s32 v1, s32 v2) +/* unlocked internal variant */ +static inline int pm_qos_get_value(struct pm_qos_object *o) { - return max(v1, v2); -} + if (plist_head_empty(&o->requests)) + return o->default_value; -static s32 min_compare(s32 v1, s32 v2) -{ - return min(v1, v2); -} + switch (o->type) { + case PM_QOS_MIN: + return plist_first(&o->requests)->prio; + + case PM_QOS_MAX: + return plist_last(&o->requests)->prio; + default: + /* runtime check for not using enum */ + BUG(); + } +} -static void update_target(int pm_qos_class) +static void update_target(struct pm_qos_object *o, struct plist_node *node, + int del, int value) { - s32 extreme_value; - struct pm_qos_request_list *node; unsigned long flags; - int call_notifier = 0; + int prev_value, curr_value; spin_lock_irqsave(&pm_qos_lock, flags); - extreme_value = pm_qos_array[pm_qos_class]->default_value; - list_for_each_entry(node, - &pm_qos_array[pm_qos_class]->requests.list, list) { - extreme_value = pm_qos_array[pm_qos_class]->comparitor( - extreme_value, node->value); - } - if (atomic_read(&pm_qos_array[pm_qos_class]->target_value) != - extreme_value) { - call_notifier = 1; - atomic_set(&pm_qos_array[pm_qos_class]->target_value, - extreme_value); - pr_debug(KERN_ERR "new target for qos %d is %d\n", pm_qos_class, - atomic_read(&pm_qos_array[pm_qos_class]->target_value)); + prev_value = pm_qos_get_value(o); + /* PM_QOS_DEFAULT_VALUE is a signal that the value is unchanged */ + if (value != PM_QOS_DEFAULT_VALUE) { + /* + * to change the list, we atomically remove, reinit + * with new value and add, then see if the extremal + * changed + */ + plist_del(node, &o->requests); + plist_node_init(node, value); + plist_add(node, &o->requests); + } else if (del) { + plist_del(node, &o->requests); + } else { + plist_add(node, &o->requests); } + curr_value = pm_qos_get_value(o); spin_unlock_irqrestore(&pm_qos_lock, flags); - if (call_notifier) - blocking_notifier_call_chain( - pm_qos_array[pm_qos_class]->notifiers, - (unsigned long) extreme_value, NULL); + if (prev_value != curr_value) + blocking_notifier_call_chain(o->notifiers, + (unsigned long)curr_value, + NULL); } static int register_pm_qos_misc(struct pm_qos_object *qos) @@ -196,42 +194,53 @@ static int find_pm_qos_object_by_minor(int minor) */ int pm_qos_request(int pm_qos_class) { - return atomic_read(&pm_qos_array[pm_qos_class]->target_value); + unsigned long flags; + int value; + + spin_lock_irqsave(&pm_qos_lock, flags); + value = pm_qos_get_value(pm_qos_array[pm_qos_class]); + spin_unlock_irqrestore(&pm_qos_lock, flags); + + return value; } EXPORT_SYMBOL_GPL(pm_qos_request); +int pm_qos_request_active(struct pm_qos_request_list *req) +{ + return req->pm_qos_class != 0; +} +EXPORT_SYMBOL_GPL(pm_qos_request_active); + /** * pm_qos_add_request - inserts new qos request into the list - * @pm_qos_class: identifies which list of qos request to us + * @dep: pointer to a preallocated handle + * @pm_qos_class: identifies which list of qos request to use * @value: defines the qos request * * This function inserts a new entry in the pm_qos_class list of requested qos * performance characteristics. It recomputes the aggregate QoS expectations - * for the pm_qos_class of parameters, and returns the pm_qos_request list - * element as a handle for use in updating and removal. Call needs to save - * this handle for later use. + * for the pm_qos_class of parameters and initializes the pm_qos_request_list + * handle. Caller needs to save this handle for later use in updates and + * removal. */ -struct pm_qos_request_list *pm_qos_add_request(int pm_qos_class, s32 value) + +void pm_qos_add_request(struct pm_qos_request_list *dep, + int pm_qos_class, s32 value) { - struct pm_qos_request_list *dep; - unsigned long flags; + struct pm_qos_object *o = pm_qos_array[pm_qos_class]; + int new_value; - dep = kzalloc(sizeof(struct pm_qos_request_list), GFP_KERNEL); - if (dep) { - if (value == PM_QOS_DEFAULT_VALUE) - dep->value = pm_qos_array[pm_qos_class]->default_value; - else - dep->value = value; - dep->pm_qos_class = pm_qos_class; - - spin_lock_irqsave(&pm_qos_lock, flags); - list_add(&dep->list, - &pm_qos_array[pm_qos_class]->requests.list); - spin_unlock_irqrestore(&pm_qos_lock, flags); - update_target(pm_qos_class); + if (pm_qos_request_active(dep)) { + WARN(1, KERN_ERR "pm_qos_add_request() called for already added request\n"); + return; } - - return dep; + if (value == PM_QOS_DEFAULT_VALUE) + new_value = o->default_value; + else + new_value = value; + plist_node_init(&dep->list, new_value); + dep->pm_qos_class = pm_qos_class; + update_target(o, &dep->list, 0, PM_QOS_DEFAULT_VALUE); } EXPORT_SYMBOL_GPL(pm_qos_add_request); @@ -246,27 +255,28 @@ EXPORT_SYMBOL_GPL(pm_qos_add_request); * Attempts are made to make this code callable on hot code paths. */ void pm_qos_update_request(struct pm_qos_request_list *pm_qos_req, - s32 new_value) + s32 new_value) { - unsigned long flags; - int pending_update = 0; s32 temp; + struct pm_qos_object *o; + + if (!pm_qos_req) /*guard against callers passing in null */ + return; - if (pm_qos_req) { /*guard against callers passing in null */ - spin_lock_irqsave(&pm_qos_lock, flags); - if (new_value == PM_QOS_DEFAULT_VALUE) - temp = pm_qos_array[pm_qos_req->pm_qos_class]->default_value; - else - temp = new_value; - - if (temp != pm_qos_req->value) { - pending_update = 1; - pm_qos_req->value = temp; - } - spin_unlock_irqrestore(&pm_qos_lock, flags); - if (pending_update) - update_target(pm_qos_req->pm_qos_class); + if (!pm_qos_request_active(pm_qos_req)) { + WARN(1, KERN_ERR "pm_qos_update_request() called for unknown object\n"); + return; } + + o = pm_qos_array[pm_qos_req->pm_qos_class]; + + if (new_value == PM_QOS_DEFAULT_VALUE) + temp = o->default_value; + else + temp = new_value; + + if (temp != pm_qos_req->list.prio) + update_target(o, &pm_qos_req->list, 0, temp); } EXPORT_SYMBOL_GPL(pm_qos_update_request); @@ -280,19 +290,20 @@ EXPORT_SYMBOL_GPL(pm_qos_update_request); */ void pm_qos_remove_request(struct pm_qos_request_list *pm_qos_req) { - unsigned long flags; - int qos_class; + struct pm_qos_object *o; if (pm_qos_req == NULL) return; /* silent return to keep pcm code cleaner */ - qos_class = pm_qos_req->pm_qos_class; - spin_lock_irqsave(&pm_qos_lock, flags); - list_del(&pm_qos_req->list); - kfree(pm_qos_req); - spin_unlock_irqrestore(&pm_qos_lock, flags); - update_target(qos_class); + if (!pm_qos_request_active(pm_qos_req)) { + WARN(1, KERN_ERR "pm_qos_remove_request() called for unknown object\n"); + return; + } + + o = pm_qos_array[pm_qos_req->pm_qos_class]; + update_target(o, &pm_qos_req->list, 1, PM_QOS_DEFAULT_VALUE); + memset(pm_qos_req, 0, sizeof(*pm_qos_req)); } EXPORT_SYMBOL_GPL(pm_qos_remove_request); @@ -340,8 +351,12 @@ static int pm_qos_power_open(struct inode *inode, struct file *filp) pm_qos_class = find_pm_qos_object_by_minor(iminor(inode)); if (pm_qos_class >= 0) { - filp->private_data = (void *) pm_qos_add_request(pm_qos_class, - PM_QOS_DEFAULT_VALUE); + struct pm_qos_request_list *req = kzalloc(sizeof(*req), GFP_KERNEL); + if (!req) + return -ENOMEM; + + pm_qos_add_request(req, pm_qos_class, PM_QOS_DEFAULT_VALUE); + filp->private_data = req; if (filp->private_data) return 0; @@ -353,8 +368,9 @@ static int pm_qos_power_release(struct inode *inode, struct file *filp) { struct pm_qos_request_list *req; - req = (struct pm_qos_request_list *)filp->private_data; + req = filp->private_data; pm_qos_remove_request(req); + kfree(req); return 0; } @@ -374,14 +390,16 @@ static ssize_t pm_qos_power_write(struct file *filp, const char __user *buf, } else if (count == 11) { /* len('0x12345678/0') */ if (copy_from_user(ascii_value, buf, 11)) return -EFAULT; + if (strlen(ascii_value) != 10) + return -EINVAL; x = sscanf(ascii_value, "%x", &value); if (x != 1) return -EINVAL; - pr_debug(KERN_ERR "%s, %d, 0x%x\n", ascii_value, x, value); + pr_debug("%s, %d, 0x%x\n", ascii_value, x, value); } else return -EINVAL; - pm_qos_req = (struct pm_qos_request_list *)filp->private_data; + pm_qos_req = filp->private_data; pm_qos_update_request(pm_qos_req, value); return count; diff --git a/kernel/posix-cpu-timers.c b/kernel/posix-cpu-timers.c index 00bb252f29a..05bb7173850 100644 --- a/kernel/posix-cpu-timers.c +++ b/kernel/posix-cpu-timers.c @@ -16,13 +16,13 @@ * siglock protection since other code may update expiration cache as * well. */ -void update_rlimit_cpu(unsigned long rlim_new) +void update_rlimit_cpu(struct task_struct *task, unsigned long rlim_new) { cputime_t cputime = secs_to_cputime(rlim_new); - spin_lock_irq(¤t->sighand->siglock); - set_process_cpu_timer(current, CPUCLOCK_PROF, &cputime, NULL); - spin_unlock_irq(¤t->sighand->siglock); + spin_lock_irq(&task->sighand->siglock); + set_process_cpu_timer(task, CPUCLOCK_PROF, &cputime, NULL); + spin_unlock_irq(&task->sighand->siglock); } static int check_clock(const clockid_t which_clock) @@ -37,13 +37,13 @@ static int check_clock(const clockid_t which_clock) if (pid == 0) return 0; - read_lock(&tasklist_lock); + rcu_read_lock(); p = find_task_by_vpid(pid); if (!p || !(CPUCLOCK_PERTHREAD(which_clock) ? - same_thread_group(p, current) : thread_group_leader(p))) { + same_thread_group(p, current) : has_group_leader_pid(p))) { error = -EINVAL; } - read_unlock(&tasklist_lock); + rcu_read_unlock(); return error; } @@ -232,31 +232,24 @@ static int cpu_clock_sample(const clockid_t which_clock, struct task_struct *p, void thread_group_cputime(struct task_struct *tsk, struct task_cputime *times) { - struct sighand_struct *sighand; - struct signal_struct *sig; + struct signal_struct *sig = tsk->signal; struct task_struct *t; - *times = INIT_CPUTIME; + times->utime = sig->utime; + times->stime = sig->stime; + times->sum_exec_runtime = sig->sum_sched_runtime; rcu_read_lock(); - sighand = rcu_dereference(tsk->sighand); - if (!sighand) + /* make sure we can trust tsk->thread_group list */ + if (!likely(pid_alive(tsk))) goto out; - sig = tsk->signal; - t = tsk; do { times->utime = cputime_add(times->utime, t->utime); times->stime = cputime_add(times->stime, t->stime); times->sum_exec_runtime += t->se.sum_exec_runtime; - - t = next_thread(t); - } while (t != tsk); - - times->utime = cputime_add(times->utime, sig->utime); - times->stime = cputime_add(times->stime, sig->stime); - times->sum_exec_runtime += sig->sum_sched_runtime; + } while_each_thread(tsk, t); out: rcu_read_unlock(); } @@ -363,7 +356,7 @@ int posix_cpu_clock_get(const clockid_t which_clock, struct timespec *tp) } } else { read_lock(&tasklist_lock); - if (thread_group_leader(p) && p->signal) { + if (thread_group_leader(p) && p->sighand) { error = cpu_clock_sample_group(which_clock, p, &rtn); @@ -397,7 +390,7 @@ int posix_cpu_timer_create(struct k_itimer *new_timer) INIT_LIST_HEAD(&new_timer->it.cpu.entry); - read_lock(&tasklist_lock); + rcu_read_lock(); if (CPUCLOCK_PERTHREAD(new_timer->it_clock)) { if (pid == 0) { p = current; @@ -411,7 +404,7 @@ int posix_cpu_timer_create(struct k_itimer *new_timer) p = current->group_leader; } else { p = find_task_by_vpid(pid); - if (p && !thread_group_leader(p)) + if (p && !has_group_leader_pid(p)) p = NULL; } } @@ -421,7 +414,7 @@ int posix_cpu_timer_create(struct k_itimer *new_timer) } else { ret = -EINVAL; } - read_unlock(&tasklist_lock); + rcu_read_unlock(); return ret; } @@ -439,7 +432,7 @@ int posix_cpu_timer_del(struct k_itimer *timer) if (likely(p != NULL)) { read_lock(&tasklist_lock); - if (unlikely(p->signal == NULL)) { + if (unlikely(p->sighand == NULL)) { /* * We raced with the reaping of the task. * The deletion should have cleared us off the list. @@ -691,10 +684,10 @@ int posix_cpu_timer_set(struct k_itimer *timer, int flags, read_lock(&tasklist_lock); /* * We need the tasklist_lock to protect against reaping that - * clears p->signal. If p has just been reaped, we can no + * clears p->sighand. If p has just been reaped, we can no * longer get any information about it at all. */ - if (unlikely(p->signal == NULL)) { + if (unlikely(p->sighand == NULL)) { read_unlock(&tasklist_lock); put_task_struct(p); timer->it.cpu.task = NULL; @@ -863,7 +856,7 @@ void posix_cpu_timer_get(struct k_itimer *timer, struct itimerspec *itp) clear_dead = p->exit_state; } else { read_lock(&tasklist_lock); - if (unlikely(p->signal == NULL)) { + if (unlikely(p->sighand == NULL)) { /* * The process has been reaped. * We can't even collect a sample any more. @@ -1199,7 +1192,7 @@ void posix_cpu_timer_schedule(struct k_itimer *timer) spin_lock(&p->sighand->siglock); } else { read_lock(&tasklist_lock); - if (unlikely(p->signal == NULL)) { + if (unlikely(p->sighand == NULL)) { /* * The process has been reaped. * We can't even collect a sample any more. @@ -1279,10 +1272,6 @@ static inline int fastpath_timer_check(struct task_struct *tsk) { struct signal_struct *sig; - /* tsk == current, ensure it is safe to use ->signal/sighand */ - if (unlikely(tsk->exit_state)) - return 0; - if (!task_cputime_zero(&tsk->cputime_expires)) { struct task_cputime task_sample = { .utime = tsk->utime, @@ -1298,7 +1287,10 @@ static inline int fastpath_timer_check(struct task_struct *tsk) if (sig->cputimer.running) { struct task_cputime group_sample; - thread_group_cputimer(tsk, &group_sample); + spin_lock(&sig->cputimer.lock); + group_sample = sig->cputimer.cputime; + spin_unlock(&sig->cputimer.lock); + if (task_cputime_expired(&group_sample, &sig->cputime_expires)) return 1; } @@ -1315,6 +1307,7 @@ void run_posix_cpu_timers(struct task_struct *tsk) { LIST_HEAD(firing); struct k_itimer *timer, *next; + unsigned long flags; BUG_ON(!irqs_disabled()); @@ -1325,7 +1318,8 @@ void run_posix_cpu_timers(struct task_struct *tsk) if (!fastpath_timer_check(tsk)) return; - spin_lock(&tsk->sighand->siglock); + if (!lock_task_sighand(tsk, &flags)) + return; /* * Here we take off tsk->signal->cpu_timers[N] and * tsk->cpu_timers[N] all the timers that are firing, and @@ -1347,7 +1341,7 @@ void run_posix_cpu_timers(struct task_struct *tsk) * that gets the timer lock before we do will give it up and * spin until we've taken care of that timer below. */ - spin_unlock(&tsk->sighand->siglock); + unlock_task_sighand(tsk, &flags); /* * Now that all the timers on our list have the firing flag, diff --git a/kernel/posix-timers.c b/kernel/posix-timers.c index 00d1fda58ab..93bd2eb2bc5 100644 --- a/kernel/posix-timers.c +++ b/kernel/posix-timers.c @@ -145,7 +145,13 @@ static int common_timer_del(struct k_itimer *timer); static enum hrtimer_restart posix_timer_fn(struct hrtimer *data); -static struct k_itimer *lock_timer(timer_t timer_id, unsigned long *flags); +static struct k_itimer *__lock_timer(timer_t timer_id, unsigned long *flags); + +#define lock_timer(tid, flags) \ +({ struct k_itimer *__timr; \ + __cond_lock(&__timr->it_lock, __timr = __lock_timer(tid, flags)); \ + __timr; \ +}) static inline void unlock_timer(struct k_itimer *timr, unsigned long flags) { @@ -559,19 +565,7 @@ SYSCALL_DEFINE3(timer_create, const clockid_t, which_clock, new_timer->it_id = (timer_t) new_timer_id; new_timer->it_clock = which_clock; new_timer->it_overrun = -1; - error = CLOCK_DISPATCH(which_clock, timer_create, (new_timer)); - if (error) - goto out; - /* - * return the timer_id now. The next step is hard to - * back out if there is an error. - */ - if (copy_to_user(created_timer_id, - &new_timer_id, sizeof (new_timer_id))) { - error = -EFAULT; - goto out; - } if (timer_event_spec) { if (copy_from_user(&event, timer_event_spec, sizeof (event))) { error = -EFAULT; @@ -597,6 +591,16 @@ SYSCALL_DEFINE3(timer_create, const clockid_t, which_clock, new_timer->sigq->info.si_tid = new_timer->it_id; new_timer->sigq->info.si_code = SI_TIMER; + if (copy_to_user(created_timer_id, + &new_timer_id, sizeof (new_timer_id))) { + error = -EFAULT; + goto out; + } + + error = CLOCK_DISPATCH(which_clock, timer_create, (new_timer)); + if (error) + goto out; + spin_lock_irq(¤t->sighand->siglock); new_timer->it_signal = current->signal; list_add(&new_timer->list, ¤t->signal->posix_timers); @@ -621,7 +625,7 @@ out: * the find to the timer lock. To avoid a dead lock, the timer id MUST * be release with out holding the timer lock. */ -static struct k_itimer *lock_timer(timer_t timer_id, unsigned long *flags) +static struct k_itimer *__lock_timer(timer_t timer_id, unsigned long *flags) { struct k_itimer *timr; /* diff --git a/kernel/power/Kconfig b/kernel/power/Kconfig index 5c36ea9d55d..265729966ec 100644 --- a/kernel/power/Kconfig +++ b/kernel/power/Kconfig @@ -86,6 +86,7 @@ config PM_SLEEP_SMP depends on SMP depends on ARCH_SUSPEND_POSSIBLE || ARCH_HIBERNATION_POSSIBLE depends on PM_SLEEP + select HOTPLUG select HOTPLUG_CPU default y @@ -130,13 +131,11 @@ 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 + select LZO_COMPRESS + select LZO_DECOMPRESS ---help--- Enable the suspend to disk (STD) functionality, which is usually called "hibernation" in user interfaces. STD checkpoints the @@ -241,3 +240,21 @@ config PM_OPS bool depends on PM_SLEEP || PM_RUNTIME default y + +config ARCH_HAS_OPP + bool + +config PM_OPP + bool "Operating Performance Point (OPP) Layer library" + depends on PM + depends on ARCH_HAS_OPP + ---help--- + SOCs have a standard set of tuples consisting of frequency and + voltage pairs that the device will support per voltage domain. This + is called Operating Performance Point or OPP. The actual definitions + of OPP varies over silicon within the same family of devices. + + OPP layer organizes the data internally using device pointers + representing individual voltage domains and provides SOC + implementations a ready to use framework to manage OPPs. + For more information, read <file:Documentation/power/opp.txt> diff --git a/kernel/power/Makefile b/kernel/power/Makefile index 524e058dcf0..c350e18b53e 100644 --- a/kernel/power/Makefile +++ b/kernel/power/Makefile @@ -1,7 +1,4 @@ - -ifeq ($(CONFIG_PM_DEBUG),y) -EXTRA_CFLAGS += -DDEBUG -endif +ccflags-$(CONFIG_PM_DEBUG) := -DDEBUG obj-$(CONFIG_PM) += main.o obj-$(CONFIG_PM_SLEEP) += console.o @@ -10,6 +7,5 @@ obj-$(CONFIG_SUSPEND) += suspend.o obj-$(CONFIG_PM_TEST_SUSPEND) += suspend_test.o obj-$(CONFIG_HIBERNATION) += hibernate.o snapshot.o swap.o user.o \ block_io.o -obj-$(CONFIG_HIBERNATION_NVS) += hibernate_nvs.o obj-$(CONFIG_MAGIC_SYSRQ) += poweroff.o diff --git a/kernel/power/block_io.c b/kernel/power/block_io.c index 97024fd40cd..83bbc7c02df 100644 --- a/kernel/power/block_io.c +++ b/kernel/power/block_io.c @@ -28,7 +28,7 @@ static int submit(int rw, struct block_device *bdev, sector_t sector, struct page *page, struct bio **bio_chain) { - const int bio_rw = rw | (1 << BIO_RW_SYNCIO) | (1 << BIO_RW_UNPLUG); + const int bio_rw = rw | REQ_SYNC | REQ_UNPLUG; struct bio *bio; bio = bio_alloc(__GFP_WAIT | __GFP_HIGH, 1); diff --git a/kernel/power/hibernate.c b/kernel/power/hibernate.c index aa9e916da4d..1832bd26421 100644 --- a/kernel/power/hibernate.c +++ b/kernel/power/hibernate.c @@ -3,7 +3,7 @@ * * Copyright (c) 2003 Patrick Mochel * Copyright (c) 2003 Open Source Development Lab - * Copyright (c) 2004 Pavel Machek <pavel@suse.cz> + * Copyright (c) 2004 Pavel Machek <pavel@ucw.cz> * Copyright (c) 2009 Rafael J. Wysocki, Novell Inc. * * This file is released under the GPLv2. @@ -29,6 +29,7 @@ #include "power.h" +static int nocompress = 0; static int noresume = 0; static char resume_file[256] = CONFIG_PM_STD_PARTITION; dev_t swsusp_resume_device; @@ -50,18 +51,18 @@ enum { static int hibernation_mode = HIBERNATION_SHUTDOWN; -static struct platform_hibernation_ops *hibernation_ops; +static const struct platform_hibernation_ops *hibernation_ops; /** * hibernation_set_ops - set the global hibernate operations * @ops: the hibernation operations to use in subsequent hibernation transitions */ -void hibernation_set_ops(struct platform_hibernation_ops *ops) +void hibernation_set_ops(const struct platform_hibernation_ops *ops) { if (ops && !(ops->begin && ops->end && ops->pre_snapshot && ops->prepare && ops->finish && ops->enter && ops->pre_restore - && ops->restore_cleanup)) { + && ops->restore_cleanup && ops->leave)) { WARN_ON(1); return; } @@ -277,7 +278,7 @@ static int create_image(int platform_mode) goto Enable_irqs; } - if (hibernation_test(TEST_CORE)) + if (hibernation_test(TEST_CORE) || pm_wakeup_pending()) goto Power_up; in_suspend = 1; @@ -288,8 +289,10 @@ static int create_image(int platform_mode) error); /* Restore control flow magically appears here */ restore_processor_state(); - if (!in_suspend) + if (!in_suspend) { + events_check_enabled = false; platform_leave(platform_mode); + } Power_up: sysdev_resume(); @@ -324,11 +327,10 @@ static int create_image(int platform_mode) int hibernation_snapshot(int platform_mode) { int error; - gfp_t saved_mask; error = platform_begin(platform_mode); if (error) - return error; + goto Close; /* Preallocate image memory before shutting down devices. */ error = hibernate_preallocate_memory(); @@ -336,7 +338,7 @@ int hibernation_snapshot(int platform_mode) goto Close; suspend_console(); - saved_mask = clear_gfp_allowed_mask(GFP_IOFS); + pm_restrict_gfp_mask(); error = dpm_suspend_start(PMSG_FREEZE); if (error) goto Recover_platform; @@ -345,7 +347,10 @@ int hibernation_snapshot(int platform_mode) goto Recover_platform; error = create_image(platform_mode); - /* Control returns here after successful restore */ + /* + * Control returns here (1) after the image has been created or the + * image creation has failed and (2) after a successful restore. + */ Resume_devices: /* We may need to release the preallocated image pages here. */ @@ -354,7 +359,10 @@ int hibernation_snapshot(int platform_mode) dpm_resume_end(in_suspend ? (error ? PMSG_RECOVER : PMSG_THAW) : PMSG_RESTORE); - set_gfp_allowed_mask(saved_mask); + + if (error || !in_suspend) + pm_restore_gfp_mask(); + resume_console(); Close: platform_end(platform_mode); @@ -449,17 +457,16 @@ static int resume_target_kernel(bool platform_mode) int hibernation_restore(int platform_mode) { int error; - gfp_t saved_mask; pm_prepare_console(); suspend_console(); - saved_mask = clear_gfp_allowed_mask(GFP_IOFS); + pm_restrict_gfp_mask(); error = dpm_suspend_start(PMSG_QUIESCE); if (!error) { error = resume_target_kernel(platform_mode); dpm_resume_end(PMSG_RECOVER); } - set_gfp_allowed_mask(saved_mask); + pm_restore_gfp_mask(); resume_console(); pm_restore_console(); return error; @@ -473,7 +480,6 @@ int hibernation_restore(int platform_mode) int hibernation_platform_enter(void) { int error; - gfp_t saved_mask; if (!hibernation_ops) return -ENOSYS; @@ -489,7 +495,6 @@ int hibernation_platform_enter(void) entering_platform_hibernation = true; suspend_console(); - saved_mask = clear_gfp_allowed_mask(GFP_IOFS); error = dpm_suspend_start(PMSG_HIBERNATE); if (error) { if (hibernation_ops->recover) @@ -511,23 +516,28 @@ int hibernation_platform_enter(void) local_irq_disable(); sysdev_suspend(PMSG_HIBERNATE); + if (pm_wakeup_pending()) { + error = -EAGAIN; + goto Power_up; + } + hibernation_ops->enter(); /* We should never get here */ while (1); - /* - * We don't need to reenable the nonboot CPUs or resume consoles, since - * the system is going to be halted anyway. - */ + Power_up: + sysdev_resume(); + local_irq_enable(); + enable_nonboot_cpus(); + Platform_finish: hibernation_ops->finish(); - dpm_suspend_noirq(PMSG_RESTORE); + dpm_resume_noirq(PMSG_RESTORE); Resume_devices: entering_platform_hibernation = false; dpm_resume_end(PMSG_RESTORE); - set_gfp_allowed_mask(saved_mask); resume_console(); Close: @@ -630,11 +640,15 @@ int hibernate(void) if (hibernation_mode == HIBERNATION_PLATFORM) flags |= SF_PLATFORM_MODE; + if (nocompress) + flags |= SF_NOCOMPRESS_MODE; pr_debug("PM: writing image.\n"); error = swsusp_write(flags); swsusp_free(); if (!error) power_down(); + in_suspend = 0; + pm_restore_gfp_mask(); } else { pr_debug("PM: Image restored successfully.\n"); } @@ -697,7 +711,7 @@ static int software_resume(void) goto Unlock; } - pr_debug("PM: Checking image partition %s\n", resume_file); + pr_debug("PM: Checking hibernation image partition %s\n", resume_file); /* Check if the device is there */ swsusp_resume_device = name_to_dev_t(resume_file); @@ -722,10 +736,10 @@ static int software_resume(void) } Check_image: - pr_debug("PM: Resume from partition %d:%d\n", + pr_debug("PM: Hibernation image partition %d:%d present\n", MAJOR(swsusp_resume_device), MINOR(swsusp_resume_device)); - pr_debug("PM: Checking hibernation image.\n"); + pr_debug("PM: Looking for hibernation image.\n"); error = swsusp_check(); if (error) goto Unlock; @@ -757,14 +771,14 @@ static int software_resume(void) goto Done; } - pr_debug("PM: Reading hibernation image.\n"); + pr_debug("PM: Loading hibernation image.\n"); error = swsusp_read(&flags); swsusp_close(FMODE_READ); if (!error) hibernation_restore(flags & SF_PLATFORM_MODE); - printk(KERN_ERR "PM: Restore failed, recovering.\n"); + printk(KERN_ERR "PM: Failed to load hibernation image, recovering.\n"); swsusp_free(); thaw_processes(); Done: @@ -777,7 +791,7 @@ static int software_resume(void) /* For success case, the suspend path will release the lock */ Unlock: mutex_unlock(&pm_mutex); - pr_debug("PM: Resume from disk failed.\n"); + pr_debug("PM: Hibernation image not present or could not be loaded.\n"); return error; close_finish: swsusp_close(FMODE_READ); @@ -996,6 +1010,15 @@ static int __init resume_offset_setup(char *str) return 1; } +static int __init hibernate_setup(char *str) +{ + if (!strncmp(str, "noresume", 8)) + noresume = 1; + else if (!strncmp(str, "nocompress", 10)) + nocompress = 1; + return 1; +} + static int __init noresume_setup(char *str) { noresume = 1; @@ -1005,3 +1028,4 @@ static int __init noresume_setup(char *str) __setup("noresume", noresume_setup); __setup("resume_offset=", resume_offset_setup); __setup("resume=", resume_setup); +__setup("hibernate=", hibernate_setup); diff --git a/kernel/power/hibernate_nvs.c b/kernel/power/hibernate_nvs.c deleted file mode 100644 index fdcad9ed5a7..00000000000 --- a/kernel/power/hibernate_nvs.c +++ /dev/null @@ -1,136 +0,0 @@ -/* - * 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/slab.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 b58800b21fc..7b5db6a8561 100644 --- a/kernel/power/main.c +++ b/kernel/power/main.c @@ -204,6 +204,60 @@ static ssize_t state_store(struct kobject *kobj, struct kobj_attribute *attr, power_attr(state); +#ifdef CONFIG_PM_SLEEP +/* + * The 'wakeup_count' attribute, along with the functions defined in + * drivers/base/power/wakeup.c, provides a means by which wakeup events can be + * handled in a non-racy way. + * + * If a wakeup event occurs when the system is in a sleep state, it simply is + * woken up. In turn, if an event that would wake the system up from a sleep + * state occurs when it is undergoing a transition to that sleep state, the + * transition should be aborted. Moreover, if such an event occurs when the + * system is in the working state, an attempt to start a transition to the + * given sleep state should fail during certain period after the detection of + * the event. Using the 'state' attribute alone is not sufficient to satisfy + * these requirements, because a wakeup event may occur exactly when 'state' + * is being written to and may be delivered to user space right before it is + * frozen, so the event will remain only partially processed until the system is + * woken up by another event. In particular, it won't cause the transition to + * a sleep state to be aborted. + * + * This difficulty may be overcome if user space uses 'wakeup_count' before + * writing to 'state'. It first should read from 'wakeup_count' and store + * the read value. Then, after carrying out its own preparations for the system + * transition to a sleep state, it should write the stored value to + * 'wakeup_count'. If that fails, at least one wakeup event has occured since + * 'wakeup_count' was read and 'state' should not be written to. Otherwise, it + * is allowed to write to 'state', but the transition will be aborted if there + * are any wakeup events detected after 'wakeup_count' was written to. + */ + +static ssize_t wakeup_count_show(struct kobject *kobj, + struct kobj_attribute *attr, + char *buf) +{ + unsigned int val; + + return pm_get_wakeup_count(&val) ? sprintf(buf, "%u\n", val) : -EINTR; +} + +static ssize_t wakeup_count_store(struct kobject *kobj, + struct kobj_attribute *attr, + const char *buf, size_t n) +{ + unsigned int val; + + if (sscanf(buf, "%u", &val) == 1) { + if (pm_save_wakeup_count(val)) + return n; + } + return -EINVAL; +} + +power_attr(wakeup_count); +#endif /* CONFIG_PM_SLEEP */ + #ifdef CONFIG_PM_TRACE int pm_trace_enabled; @@ -227,15 +281,34 @@ pm_trace_store(struct kobject *kobj, struct kobj_attribute *attr, } power_attr(pm_trace); + +static ssize_t pm_trace_dev_match_show(struct kobject *kobj, + struct kobj_attribute *attr, + char *buf) +{ + return show_trace_dev_match(buf, PAGE_SIZE); +} + +static ssize_t +pm_trace_dev_match_store(struct kobject *kobj, struct kobj_attribute *attr, + const char *buf, size_t n) +{ + return -EINVAL; +} + +power_attr(pm_trace_dev_match); + #endif /* CONFIG_PM_TRACE */ static struct attribute * g[] = { &state_attr.attr, #ifdef CONFIG_PM_TRACE &pm_trace_attr.attr, + &pm_trace_dev_match_attr.attr, #endif #ifdef CONFIG_PM_SLEEP &pm_async_attr.attr, + &wakeup_count_attr.attr, #ifdef CONFIG_PM_DEBUG &pm_test_attr.attr, #endif @@ -253,7 +326,7 @@ EXPORT_SYMBOL_GPL(pm_wq); static int __init pm_start_workqueue(void) { - pm_wq = create_freezeable_workqueue("pm"); + pm_wq = alloc_workqueue("pm", WQ_FREEZEABLE, 0); return pm_wq ? 0 : -ENOMEM; } @@ -266,6 +339,7 @@ static int __init pm_init(void) int error = pm_start_workqueue(); if (error) return error; + hibernate_image_size_init(); power_kobj = kobject_create_and_add("power", NULL); if (!power_kobj) return -ENOMEM; diff --git a/kernel/power/power.h b/kernel/power/power.h index 006270fe382..03634be55f6 100644 --- a/kernel/power/power.h +++ b/kernel/power/power.h @@ -14,6 +14,9 @@ struct swsusp_info { } __attribute__((aligned(PAGE_SIZE))); #ifdef CONFIG_HIBERNATION +/* kernel/power/snapshot.c */ +extern void __init hibernate_image_size_init(void); + #ifdef CONFIG_ARCH_HIBERNATION_HEADER /* Maximum size of architecture specific data in a hibernation header */ #define MAX_ARCH_HEADER_SIZE (sizeof(struct new_utsname) + 4) @@ -49,7 +52,11 @@ static inline char *check_image_kernel(struct swsusp_info *info) extern int hibernation_snapshot(int platform_mode); extern int hibernation_restore(int platform_mode); extern int hibernation_platform_enter(void); -#endif + +#else /* !CONFIG_HIBERNATION */ + +static inline void hibernate_image_size_init(void) {} +#endif /* !CONFIG_HIBERNATION */ extern int pfn_is_nosave(unsigned long); @@ -134,6 +141,7 @@ extern int swsusp_swap_in_use(void); * the image header. */ #define SF_PLATFORM_MODE 1 +#define SF_NOCOMPRESS_MODE 2 /* kernel/power/hibernate.c */ extern int swsusp_check(void); diff --git a/kernel/power/poweroff.c b/kernel/power/poweroff.c index e8b33700627..d52359374e8 100644 --- a/kernel/power/poweroff.c +++ b/kernel/power/poweroff.c @@ -24,7 +24,7 @@ static void do_poweroff(struct work_struct *dummy) static DECLARE_WORK(poweroff_work, do_poweroff); -static void handle_poweroff(int key, struct tty_struct *tty) +static void handle_poweroff(int key) { /* run sysrq poweroff on boot cpu */ schedule_work_on(cpumask_first(cpu_online_mask), &poweroff_work); diff --git a/kernel/power/process.c b/kernel/power/process.c index 71ae29052ab..d6d2a10320e 100644 --- a/kernel/power/process.c +++ b/kernel/power/process.c @@ -15,6 +15,7 @@ #include <linux/syscalls.h> #include <linux/freezer.h> #include <linux/delay.h> +#include <linux/workqueue.h> /* * Timeout for stopping processes @@ -35,13 +36,19 @@ static int try_to_freeze_tasks(bool sig_only) struct task_struct *g, *p; unsigned long end_time; unsigned int todo; + bool wq_busy = false; struct timeval start, end; u64 elapsed_csecs64; unsigned int elapsed_csecs; + bool wakeup = false; do_gettimeofday(&start); end_time = jiffies + TIMEOUT; + + if (!sig_only) + freeze_workqueues_begin(); + while (true) { todo = 0; read_lock(&tasklist_lock); @@ -57,15 +64,32 @@ static int try_to_freeze_tasks(bool sig_only) * perturb a task in TASK_STOPPED or TASK_TRACED. * It is "frozen enough". If the task does wake * up, it will immediately call try_to_freeze. + * + * Because freeze_task() goes through p's + * scheduler lock after setting TIF_FREEZE, it's + * guaranteed that either we see TASK_RUNNING or + * try_to_stop() after schedule() in ptrace/signal + * stop sees TIF_FREEZE. */ if (!task_is_stopped_or_traced(p) && !freezer_should_skip(p)) todo++; } while_each_thread(g, p); read_unlock(&tasklist_lock); + + if (!sig_only) { + wq_busy = freeze_workqueues_busy(); + todo += wq_busy; + } + if (!todo || time_after(jiffies, end_time)) break; + if (pm_wakeup_pending()) { + wakeup = true; + break; + } + /* * We need to retry, but first give the freezing tasks some * time to enter the regrigerator. @@ -85,13 +109,18 @@ static int try_to_freeze_tasks(bool sig_only) * but it cleans up leftover PF_FREEZE requests. */ printk("\n"); - printk(KERN_ERR "Freezing of tasks failed after %d.%02d seconds " - "(%d tasks refusing to freeze):\n", - elapsed_csecs / 100, elapsed_csecs % 100, todo); + printk(KERN_ERR "Freezing of tasks %s after %d.%02d seconds " + "(%d tasks refusing to freeze, wq_busy=%d):\n", + wakeup ? "aborted" : "failed", + elapsed_csecs / 100, elapsed_csecs % 100, + todo - wq_busy, wq_busy); + + thaw_workqueues(); + read_lock(&tasklist_lock); do_each_thread(g, p) { task_lock(p); - if (freezing(p) && !freezer_should_skip(p)) + if (!wakeup && freezing(p) && !freezer_should_skip(p)) sched_show_task(p); cancel_freezing(p); task_unlock(p); @@ -157,6 +186,7 @@ void thaw_processes(void) oom_killer_enable(); printk("Restarting tasks ... "); + thaw_workqueues(); thaw_tasks(true); thaw_tasks(false); schedule(); diff --git a/kernel/power/snapshot.c b/kernel/power/snapshot.c index 25ce010e9f8..0dac75ea445 100644 --- a/kernel/power/snapshot.c +++ b/kernel/power/snapshot.c @@ -3,7 +3,7 @@ * * This file provides system snapshot/restore functionality for swsusp. * - * Copyright (C) 1998-2005 Pavel Machek <pavel@suse.cz> + * Copyright (C) 1998-2005 Pavel Machek <pavel@ucw.cz> * Copyright (C) 2006 Rafael J. Wysocki <rjw@sisk.pl> * * This file is released under the GPLv2. @@ -46,7 +46,12 @@ static void swsusp_unset_page_forbidden(struct page *); * 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; +unsigned long image_size; + +void __init hibernate_image_size_init(void) +{ + image_size = ((totalram_pages * 2) / 5) * PAGE_SIZE; +} /* List of PBEs needed for restoring the pages that were allocated before * the suspend and included in the suspend image, but have also been @@ -979,8 +984,8 @@ static void copy_data_page(unsigned long dst_pfn, unsigned long src_pfn) src = kmap_atomic(s_page, KM_USER0); dst = kmap_atomic(d_page, KM_USER1); do_copy_page(dst, src); - kunmap_atomic(src, KM_USER0); kunmap_atomic(dst, KM_USER1); + kunmap_atomic(src, KM_USER0); } else { if (PageHighMem(d_page)) { /* Page pointed to by src may contain some kernel @@ -988,7 +993,7 @@ static void copy_data_page(unsigned long dst_pfn, unsigned long src_pfn) */ safe_copy_page(buffer, s_page); dst = kmap_atomic(d_page, KM_USER0); - memcpy(dst, buffer, PAGE_SIZE); + copy_page(dst, buffer); kunmap_atomic(dst, KM_USER0); } else { safe_copy_page(page_address(d_page), s_page); @@ -1121,9 +1126,19 @@ static unsigned long preallocate_image_pages(unsigned long nr_pages, gfp_t mask) return nr_alloc; } -static unsigned long preallocate_image_memory(unsigned long nr_pages) +static unsigned long preallocate_image_memory(unsigned long nr_pages, + unsigned long avail_normal) { - return preallocate_image_pages(nr_pages, GFP_IMAGE); + unsigned long alloc; + + if (avail_normal <= alloc_normal) + return 0; + + alloc = avail_normal - alloc_normal; + if (nr_pages < alloc) + alloc = nr_pages; + + return preallocate_image_pages(alloc, GFP_IMAGE); } #ifdef CONFIG_HIGHMEM @@ -1169,15 +1184,22 @@ static inline unsigned long preallocate_highmem_fraction(unsigned long nr_pages, */ static void free_unnecessary_pages(void) { - unsigned long save_highmem, to_free_normal, to_free_highmem; + unsigned long save, 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; + save = count_data_pages(); + if (alloc_normal >= save) { + to_free_normal = alloc_normal - save; + save = 0; + } else { + to_free_normal = 0; + save -= alloc_normal; + } + save += count_highmem_pages(); + if (alloc_highmem >= save) { + to_free_highmem = alloc_highmem - save; } else { to_free_highmem = 0; - to_free_normal -= save_highmem - alloc_highmem; + to_free_normal -= save - alloc_highmem; } memory_bm_position_reset(©_bm); @@ -1258,7 +1280,7 @@ 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; + unsigned long alloc, save_highmem, pages_highmem, avail_normal; struct timeval start, stop; int error; @@ -1295,26 +1317,38 @@ int hibernate_preallocate_memory(void) else count += zone_page_state(zone, NR_FREE_PAGES); } + avail_normal = count; 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; + /* Compute the desired number of image pages specified by image_size. */ 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 the desired number of image pages is at least as large as 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); + pages += preallocate_image_memory(saveable - pages, avail_normal); goto out; } /* Estimate the minimum size of the image. */ pages = minimum_image_size(saveable); + /* + * To avoid excessive pressure on the normal zone, leave room in it to + * accommodate an image of the minimum size (unless it's already too + * small, in which case don't preallocate pages from it at all). + */ + if (avail_normal > pages) + avail_normal -= pages; + else + avail_normal = 0; if (size < pages) size = min_t(unsigned long, pages, max_size); @@ -1335,16 +1369,34 @@ int hibernate_preallocate_memory(void) */ 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; + pages = preallocate_image_memory(alloc, avail_normal); + if (pages < alloc) { + /* We have exhausted non-highmem pages, try highmem. */ + alloc -= pages; + pages += pages_highmem; + pages_highmem = preallocate_image_highmem(alloc); + if (pages_highmem < alloc) + goto err_out; + pages += pages_highmem; + /* + * size is the desired number of saveable pages to leave in + * memory, so try to preallocate (all memory - size) pages. + */ + alloc = (count - pages) - size; + pages += preallocate_image_highmem(alloc); + } else { + /* + * There are approximately max_size saveable pages at this point + * and we want to reduce this number down to size. + */ + alloc = max_size - size; + size = preallocate_highmem_fraction(alloc, highmem, count); + pages_highmem += size; + alloc -= size; + size = preallocate_image_memory(alloc, avail_normal); + pages_highmem += preallocate_image_highmem(alloc - size); + pages += pages_highmem + size; + } /* * We only need as many page frames for the image as there are saveable @@ -1635,7 +1687,7 @@ int snapshot_read_next(struct snapshot_handle *handle) memory_bm_position_reset(&orig_bm); memory_bm_position_reset(©_bm); } else if (handle->cur <= nr_meta_pages) { - memset(buffer, 0, PAGE_SIZE); + clear_page(buffer); pack_pfns(buffer, &orig_bm); } else { struct page *page; @@ -1649,7 +1701,7 @@ int snapshot_read_next(struct snapshot_handle *handle) void *kaddr; kaddr = kmap_atomic(page, KM_USER0); - memcpy(buffer, kaddr, PAGE_SIZE); + copy_page(buffer, kaddr); kunmap_atomic(kaddr, KM_USER0); handle->buffer = buffer; } else { @@ -1932,7 +1984,7 @@ static void copy_last_highmem_page(void) void *dst; dst = kmap_atomic(last_highmem_page, KM_USER0); - memcpy(dst, buffer, PAGE_SIZE); + copy_page(dst, buffer); kunmap_atomic(dst, KM_USER0); last_highmem_page = NULL; } @@ -2218,11 +2270,11 @@ swap_two_pages_data(struct page *p1, struct page *p2, void *buf) kaddr1 = kmap_atomic(p1, KM_USER0); kaddr2 = kmap_atomic(p2, KM_USER1); - memcpy(buf, kaddr1, PAGE_SIZE); - memcpy(kaddr1, kaddr2, PAGE_SIZE); - memcpy(kaddr2, buf, PAGE_SIZE); - kunmap_atomic(kaddr1, KM_USER0); + copy_page(buf, kaddr1); + copy_page(kaddr1, kaddr2); + copy_page(kaddr2, buf); kunmap_atomic(kaddr2, KM_USER1); + kunmap_atomic(kaddr1, KM_USER0); } /** diff --git a/kernel/power/suspend.c b/kernel/power/suspend.c index 56e7dbb8b99..de6f86bfa30 100644 --- a/kernel/power/suspend.c +++ b/kernel/power/suspend.c @@ -16,6 +16,13 @@ #include <linux/cpu.h> #include <linux/syscalls.h> #include <linux/gfp.h> +#include <linux/io.h> +#include <linux/kernel.h> +#include <linux/list.h> +#include <linux/mm.h> +#include <linux/slab.h> +#include <linux/suspend.h> +#include <trace/events/power.h> #include "power.h" @@ -24,13 +31,13 @@ const char *const pm_states[PM_SUSPEND_MAX] = { [PM_SUSPEND_MEM] = "mem", }; -static struct platform_suspend_ops *suspend_ops; +static const 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) +void suspend_set_ops(const struct platform_suspend_ops *ops) { mutex_lock(&pm_mutex); suspend_ops = ops; @@ -130,19 +137,19 @@ static int suspend_enter(suspend_state_t state) if (suspend_ops->prepare) { error = suspend_ops->prepare(); if (error) - return error; + goto Platform_finish; } error = dpm_suspend_noirq(PMSG_SUSPEND); if (error) { printk(KERN_ERR "PM: Some devices failed to power down\n"); - goto Platfrom_finish; + goto Platform_finish; } if (suspend_ops->prepare_late) { error = suspend_ops->prepare_late(); if (error) - goto Power_up_devices; + goto Platform_wake; } if (suspend_test(TEST_PLATFORM)) @@ -157,8 +164,10 @@ static int suspend_enter(suspend_state_t state) error = sysdev_suspend(PMSG_SUSPEND); if (!error) { - if (!suspend_test(TEST_CORE)) + if (!(suspend_test(TEST_CORE) || pm_wakeup_pending())) { error = suspend_ops->enter(state); + events_check_enabled = false; + } sysdev_resume(); } @@ -172,10 +181,9 @@ static int suspend_enter(suspend_state_t state) if (suspend_ops->wake) suspend_ops->wake(); - Power_up_devices: dpm_resume_noirq(PMSG_RESUME); - Platfrom_finish: + Platform_finish: if (suspend_ops->finish) suspend_ops->finish(); @@ -190,18 +198,18 @@ static int suspend_enter(suspend_state_t state) int suspend_devices_and_enter(suspend_state_t state) { int error; - gfp_t saved_mask; if (!suspend_ops) return -ENOSYS; + trace_machine_suspend(state); if (suspend_ops->begin) { error = suspend_ops->begin(state); if (error) goto Close; } suspend_console(); - saved_mask = clear_gfp_allowed_mask(GFP_IOFS); + pm_restrict_gfp_mask(); suspend_test_start(); error = dpm_suspend_start(PMSG_SUSPEND); if (error) { @@ -218,11 +226,12 @@ int suspend_devices_and_enter(suspend_state_t state) suspend_test_start(); dpm_resume_end(PMSG_RESUME); suspend_test_finish("resume devices"); - set_gfp_allowed_mask(saved_mask); + pm_restore_gfp_mask(); resume_console(); Close: if (suspend_ops->end) suspend_ops->end(); + trace_machine_suspend(PWR_EVENT_EXIT); return error; Recover_platform: diff --git a/kernel/power/swap.c b/kernel/power/swap.c index b0bb2177839..7c97c3a0eee 100644 --- a/kernel/power/swap.c +++ b/kernel/power/swap.c @@ -4,8 +4,9 @@ * This file provides functions for reading the suspend image from * and writing it to a swap partition. * - * Copyright (C) 1998,2001-2005 Pavel Machek <pavel@suse.cz> + * Copyright (C) 1998,2001-2005 Pavel Machek <pavel@ucw.cz> * Copyright (C) 2006 Rafael J. Wysocki <rjw@sisk.pl> + * Copyright (C) 2010 Bojan Smojver <bojan@rexursive.com> * * This file is released under the GPLv2. * @@ -24,15 +25,17 @@ #include <linux/swapops.h> #include <linux/pm.h> #include <linux/slab.h> +#include <linux/lzo.h> +#include <linux/vmalloc.h> #include "power.h" -#define SWSUSP_SIG "S1SUSPEND" +#define HIBERNATE_SIG "S1SUSPEND" /* * The swap map is a data structure used for keeping track of each page * written to a swap partition. It consists of many swap_map_page - * structures that contain each an array of MAP_PAGE_SIZE swap entries. + * structures that contain each an array of MAP_PAGE_ENTRIES swap entries. * These structures are stored on the swap and linked together with the * help of the .next_swap member. * @@ -148,7 +151,7 @@ sector_t alloc_swapdev_block(int swap) /** * 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 + * It also frees the extents used to register which swap entries had been * allocated. */ @@ -193,7 +196,7 @@ static int mark_swapfiles(struct swap_map_handle *handle, unsigned int flags) if (!memcmp("SWAP-SPACE",swsusp_header->sig, 10) || !memcmp("SWAPSPACE2",swsusp_header->sig, 10)) { memcpy(swsusp_header->orig_sig,swsusp_header->sig, 10); - memcpy(swsusp_header->sig,SWSUSP_SIG, 10); + memcpy(swsusp_header->sig, HIBERNATE_SIG, 10); swsusp_header->image = handle->first_sector; swsusp_header->flags = flags; error = hib_bio_write_page(swsusp_resume_block, @@ -221,7 +224,7 @@ static int swsusp_swap_check(void) return res; root_swap = res; - res = blkdev_get(hib_resume_bdev, FMODE_WRITE); + res = blkdev_get(hib_resume_bdev, FMODE_WRITE, NULL); if (res) return res; @@ -249,7 +252,7 @@ static int write_page(void *buf, sector_t offset, struct bio **bio_chain) if (bio_chain) { src = (void *)__get_free_page(__GFP_WAIT | __GFP_HIGH); if (src) { - memcpy(src, buf, PAGE_SIZE); + copy_page(src, buf); } else { WARN_ON_ONCE(1); bio_chain = NULL; /* Go synchronous */ @@ -323,7 +326,7 @@ static int swap_write_page(struct swap_map_handle *handle, void *buf, error = write_page(handle->cur, handle->cur_swap, NULL); if (error) goto out; - memset(handle->cur, 0, PAGE_SIZE); + clear_page(handle->cur); handle->cur_swap = offset; handle->k = 0; } @@ -357,6 +360,18 @@ static int swap_writer_finish(struct swap_map_handle *handle, return error; } +/* We need to remember how much compressed data we need to read. */ +#define LZO_HEADER sizeof(size_t) + +/* Number of pages/bytes we'll compress at one time. */ +#define LZO_UNC_PAGES 32 +#define LZO_UNC_SIZE (LZO_UNC_PAGES * PAGE_SIZE) + +/* Number of pages/bytes we need for compressed data (worst case). */ +#define LZO_CMP_PAGES DIV_ROUND_UP(lzo1x_worst_compress(LZO_UNC_SIZE) + \ + LZO_HEADER, PAGE_SIZE) +#define LZO_CMP_SIZE (LZO_CMP_PAGES * PAGE_SIZE) + /** * save_image - save the suspend image data */ @@ -404,6 +419,137 @@ static int save_image(struct swap_map_handle *handle, return ret; } + +/** + * save_image_lzo - Save the suspend image data compressed with LZO. + * @handle: Swap mam handle to use for saving the image. + * @snapshot: Image to read data from. + * @nr_to_write: Number of pages to save. + */ +static int save_image_lzo(struct swap_map_handle *handle, + struct snapshot_handle *snapshot, + unsigned int nr_to_write) +{ + unsigned int m; + int ret = 0; + int nr_pages; + int err2; + struct bio *bio; + struct timeval start; + struct timeval stop; + size_t off, unc_len, cmp_len; + unsigned char *unc, *cmp, *wrk, *page; + + page = (void *)__get_free_page(__GFP_WAIT | __GFP_HIGH); + if (!page) { + printk(KERN_ERR "PM: Failed to allocate LZO page\n"); + return -ENOMEM; + } + + wrk = vmalloc(LZO1X_1_MEM_COMPRESS); + if (!wrk) { + printk(KERN_ERR "PM: Failed to allocate LZO workspace\n"); + free_page((unsigned long)page); + return -ENOMEM; + } + + unc = vmalloc(LZO_UNC_SIZE); + if (!unc) { + printk(KERN_ERR "PM: Failed to allocate LZO uncompressed\n"); + vfree(wrk); + free_page((unsigned long)page); + return -ENOMEM; + } + + cmp = vmalloc(LZO_CMP_SIZE); + if (!cmp) { + printk(KERN_ERR "PM: Failed to allocate LZO compressed\n"); + vfree(unc); + vfree(wrk); + free_page((unsigned long)page); + return -ENOMEM; + } + + printk(KERN_INFO + "PM: Compressing and saving image data (%u pages) ... ", + nr_to_write); + m = nr_to_write / 100; + if (!m) + m = 1; + nr_pages = 0; + bio = NULL; + do_gettimeofday(&start); + for (;;) { + for (off = 0; off < LZO_UNC_SIZE; off += PAGE_SIZE) { + ret = snapshot_read_next(snapshot); + if (ret < 0) + goto out_finish; + + if (!ret) + break; + + memcpy(unc + off, data_of(*snapshot), PAGE_SIZE); + + if (!(nr_pages % m)) + printk(KERN_CONT "\b\b\b\b%3d%%", nr_pages / m); + nr_pages++; + } + + if (!off) + break; + + unc_len = off; + ret = lzo1x_1_compress(unc, unc_len, + cmp + LZO_HEADER, &cmp_len, wrk); + if (ret < 0) { + printk(KERN_ERR "PM: LZO compression failed\n"); + break; + } + + if (unlikely(!cmp_len || + cmp_len > lzo1x_worst_compress(unc_len))) { + printk(KERN_ERR "PM: Invalid LZO compressed length\n"); + ret = -1; + break; + } + + *(size_t *)cmp = cmp_len; + + /* + * Given we are writing one page at a time to disk, we copy + * that much from the buffer, although the last bit will likely + * be smaller than full page. This is OK - we saved the length + * of the compressed data, so any garbage at the end will be + * discarded when we read it. + */ + for (off = 0; off < LZO_HEADER + cmp_len; off += PAGE_SIZE) { + memcpy(page, cmp + off, PAGE_SIZE); + + ret = swap_write_page(handle, page, &bio); + if (ret) + goto out_finish; + } + } + +out_finish: + err2 = hib_wait_on_bio_chain(&bio); + do_gettimeofday(&stop); + 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"); + + vfree(cmp); + vfree(unc); + vfree(wrk); + free_page((unsigned long)page); + + return ret; +} + /** * enough_swap - Make sure we have enough swap to save the image. * @@ -411,12 +557,16 @@ static int save_image(struct swap_map_handle *handle, * space avaiable from the resume partition. */ -static int enough_swap(unsigned int nr_pages) +static int enough_swap(unsigned int nr_pages, unsigned int flags) { unsigned int free_swap = count_swap_pages(root_swap, 1); + unsigned int required; pr_debug("PM: Free swap pages: %u\n", free_swap); - return free_swap > nr_pages + PAGES_FOR_IO; + + required = PAGES_FOR_IO + ((flags & SF_NOCOMPRESS_MODE) ? + nr_pages : (nr_pages * LZO_CMP_PAGES) / LZO_UNC_PAGES + 1); + return free_swap > required; } /** @@ -443,7 +593,7 @@ int swsusp_write(unsigned int flags) printk(KERN_ERR "PM: Cannot get swap writer\n"); return error; } - if (!enough_swap(pages)) { + if (!enough_swap(pages, flags)) { printk(KERN_ERR "PM: Not enough free swap\n"); error = -ENOSPC; goto out_finish; @@ -458,8 +608,11 @@ int swsusp_write(unsigned int flags) } header = (struct swsusp_info *)data_of(snapshot); error = swap_write_page(&handle, header, NULL); - if (!error) - error = save_image(&handle, &snapshot, pages - 1); + if (!error) { + error = (flags & SF_NOCOMPRESS_MODE) ? + save_image(&handle, &snapshot, pages - 1) : + save_image_lzo(&handle, &snapshot, pages - 1); + } out_finish: error = swap_writer_finish(&handle, flags, error); return error; @@ -590,9 +743,152 @@ static int load_image(struct swap_map_handle *handle, } /** + * load_image_lzo - Load compressed image data and decompress them with LZO. + * @handle: Swap map handle to use for loading data. + * @snapshot: Image to copy uncompressed data into. + * @nr_to_read: Number of pages to load. + */ +static int load_image_lzo(struct swap_map_handle *handle, + struct snapshot_handle *snapshot, + unsigned int nr_to_read) +{ + unsigned int m; + int error = 0; + struct bio *bio; + struct timeval start; + struct timeval stop; + unsigned nr_pages; + size_t i, off, unc_len, cmp_len; + unsigned char *unc, *cmp, *page[LZO_CMP_PAGES]; + + for (i = 0; i < LZO_CMP_PAGES; i++) { + page[i] = (void *)__get_free_page(__GFP_WAIT | __GFP_HIGH); + if (!page[i]) { + printk(KERN_ERR "PM: Failed to allocate LZO page\n"); + + while (i) + free_page((unsigned long)page[--i]); + + return -ENOMEM; + } + } + + unc = vmalloc(LZO_UNC_SIZE); + if (!unc) { + printk(KERN_ERR "PM: Failed to allocate LZO uncompressed\n"); + + for (i = 0; i < LZO_CMP_PAGES; i++) + free_page((unsigned long)page[i]); + + return -ENOMEM; + } + + cmp = vmalloc(LZO_CMP_SIZE); + if (!cmp) { + printk(KERN_ERR "PM: Failed to allocate LZO compressed\n"); + + vfree(unc); + for (i = 0; i < LZO_CMP_PAGES; i++) + free_page((unsigned long)page[i]); + + return -ENOMEM; + } + + printk(KERN_INFO + "PM: Loading and decompressing image data (%u pages) ... ", + nr_to_read); + m = nr_to_read / 100; + if (!m) + m = 1; + nr_pages = 0; + bio = NULL; + do_gettimeofday(&start); + + error = snapshot_write_next(snapshot); + if (error <= 0) + goto out_finish; + + for (;;) { + error = swap_read_page(handle, page[0], NULL); /* sync */ + if (error) + break; + + cmp_len = *(size_t *)page[0]; + if (unlikely(!cmp_len || + cmp_len > lzo1x_worst_compress(LZO_UNC_SIZE))) { + printk(KERN_ERR "PM: Invalid LZO compressed length\n"); + error = -1; + break; + } + + for (off = PAGE_SIZE, i = 1; + off < LZO_HEADER + cmp_len; off += PAGE_SIZE, i++) { + error = swap_read_page(handle, page[i], &bio); + if (error) + goto out_finish; + } + + error = hib_wait_on_bio_chain(&bio); /* need all data now */ + if (error) + goto out_finish; + + for (off = 0, i = 0; + off < LZO_HEADER + cmp_len; off += PAGE_SIZE, i++) { + memcpy(cmp + off, page[i], PAGE_SIZE); + } + + unc_len = LZO_UNC_SIZE; + error = lzo1x_decompress_safe(cmp + LZO_HEADER, cmp_len, + unc, &unc_len); + if (error < 0) { + printk(KERN_ERR "PM: LZO decompression failed\n"); + break; + } + + if (unlikely(!unc_len || + unc_len > LZO_UNC_SIZE || + unc_len & (PAGE_SIZE - 1))) { + printk(KERN_ERR "PM: Invalid LZO uncompressed length\n"); + error = -1; + break; + } + + for (off = 0; off < unc_len; off += PAGE_SIZE) { + memcpy(data_of(*snapshot), unc + off, PAGE_SIZE); + + if (!(nr_pages % m)) + printk("\b\b\b\b%3d%%", nr_pages / m); + nr_pages++; + + error = snapshot_write_next(snapshot); + if (error <= 0) + goto out_finish; + } + } + +out_finish: + do_gettimeofday(&stop); + if (!error) { + printk("\b\b\b\bdone\n"); + snapshot_write_finalize(snapshot); + if (!snapshot_image_loaded(snapshot)) + error = -ENODATA; + } else + printk("\n"); + swsusp_show_speed(&start, &stop, nr_to_read, "Read"); + + vfree(cmp); + vfree(unc); + for (i = 0; i < LZO_CMP_PAGES; i++) + free_page((unsigned long)page[i]); + + return error; +} + +/** * swsusp_read - read the hibernation image. * @flags_p: flags passed by the "frozen" kernel in the image header should - * be written into this memeory location + * be written into this memory location */ int swsusp_read(unsigned int *flags_p) @@ -612,8 +908,11 @@ int swsusp_read(unsigned int *flags_p) goto end; if (!error) error = swap_read_page(&handle, header, NULL); - if (!error) - error = load_image(&handle, &snapshot, header->pages - 1); + if (!error) { + error = (*flags_p & SF_NOCOMPRESS_MODE) ? + load_image(&handle, &snapshot, header->pages - 1) : + load_image_lzo(&handle, &snapshot, header->pages - 1); + } swap_reader_finish(&handle); end: if (!error) @@ -631,16 +930,17 @@ int swsusp_check(void) { int error; - hib_resume_bdev = open_by_devnum(swsusp_resume_device, FMODE_READ); + hib_resume_bdev = blkdev_get_by_dev(swsusp_resume_device, + FMODE_READ, NULL); if (!IS_ERR(hib_resume_bdev)) { set_blocksize(hib_resume_bdev, PAGE_SIZE); - memset(swsusp_header, 0, PAGE_SIZE); + clear_page(swsusp_header); error = hib_bio_read_page(swsusp_resume_block, swsusp_header, NULL); if (error) goto put; - if (!memcmp(SWSUSP_SIG, swsusp_header->sig, 10)) { + if (!memcmp(HIBERNATE_SIG, swsusp_header->sig, 10)) { memcpy(swsusp_header->sig, swsusp_header->orig_sig, 10); /* Reset swap signature now */ error = hib_bio_write_page(swsusp_resume_block, @@ -653,13 +953,13 @@ put: if (error) blkdev_put(hib_resume_bdev, FMODE_READ); else - pr_debug("PM: Signature found, resuming\n"); + pr_debug("PM: Image signature found, resuming\n"); } else { error = PTR_ERR(hib_resume_bdev); } if (error) - pr_debug("PM: Error %d checking image file\n", error); + pr_debug("PM: Image not found (code %d)\n", error); return error; } diff --git a/kernel/power/user.c b/kernel/power/user.c index e819e17877c..c36c3b9e8a8 100644 --- a/kernel/power/user.c +++ b/kernel/power/user.c @@ -137,7 +137,7 @@ static int snapshot_release(struct inode *inode, struct file *filp) free_all_swap_pages(data->swap); if (data->frozen) thaw_processes(); - pm_notifier_call_chain(data->mode == O_WRONLY ? + pm_notifier_call_chain(data->mode == O_RDONLY ? PM_POST_HIBERNATION : PM_POST_RESTORE); atomic_inc(&snapshot_device_available); @@ -263,6 +263,7 @@ static long snapshot_ioctl(struct file *filp, unsigned int cmd, case SNAPSHOT_UNFREEZE: if (!data->frozen || data->ready) break; + pm_restore_gfp_mask(); thaw_processes(); usermodehelper_enable(); data->frozen = 0; @@ -275,6 +276,7 @@ static long snapshot_ioctl(struct file *filp, unsigned int cmd, error = -EPERM; break; } + pm_restore_gfp_mask(); error = hibernation_snapshot(data->platform_support); if (!error) error = put_user(in_suspend, (int __user *)arg); diff --git a/kernel/printk.c b/kernel/printk.c index 444b770c959..2ddbdc73aad 100644 --- a/kernel/printk.c +++ b/kernel/printk.c @@ -37,16 +37,13 @@ #include <linux/ratelimit.h> #include <linux/kmsg_dump.h> #include <linux/syslog.h> +#include <linux/cpu.h> +#include <linux/notifier.h> +#include <linux/rculist.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, ...) @@ -83,7 +80,7 @@ EXPORT_SYMBOL(oops_in_progress); * provides serialisation for access to the entire console * driver system. */ -static DECLARE_MUTEX(console_sem); +static DEFINE_SEMAPHORE(console_sem); struct console *console_drivers; EXPORT_SYMBOL_GPL(console_drivers); @@ -100,7 +97,7 @@ static int console_locked, console_suspended; /* * logbuf_lock protects log_buf, log_start, log_end, con_start and logged_chars * It is also used in interesting ways to provide interlocking in - * release_console_sem(). + * console_unlock();. */ static DEFINE_SPINLOCK(logbuf_lock); @@ -208,7 +205,7 @@ __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 int boot_delay; /* msecs delay after each printk during bootup */ static unsigned long long loops_per_msec; /* based on boot_delay */ static int __init boot_delay_setup(char *str) @@ -259,6 +256,12 @@ static inline void boot_delay_msec(void) } #endif +#ifdef CONFIG_SECURITY_DMESG_RESTRICT +int dmesg_restrict = 1; +#else +int dmesg_restrict; +#endif + int do_syslog(int type, char __user *buf, int len, bool from_file) { unsigned i, j, limit, count; @@ -266,7 +269,20 @@ int do_syslog(int type, char __user *buf, int len, bool from_file) char c; int error = 0; - error = security_syslog(type, from_file); + /* + * If this is from /proc/kmsg we only do the capabilities checks + * at open time. + */ + if (type == SYSLOG_ACTION_OPEN || !from_file) { + if (dmesg_restrict && !capable(CAP_SYSLOG)) + goto warn; /* switch to return -EPERM after 2.6.39 */ + if ((type != SYSLOG_ACTION_READ_ALL && + type != SYSLOG_ACTION_SIZE_BUFFER) && + !capable(CAP_SYSLOG)) + goto warn; /* switch to return -EPERM after 2.6.39 */ + } + + error = security_syslog(type); if (error) return error; @@ -407,6 +423,12 @@ int do_syslog(int type, char __user *buf, int len, bool from_file) } out: return error; +warn: + /* remove after 2.6.39 */ + if (capable(CAP_SYS_ADMIN)) + WARN_ONCE(1, "Attempt to access syslog with CAP_SYS_ADMIN " + "but no CAP_SYSLOG (deprecated and denied).\n"); + return -EPERM; } SYSCALL_DEFINE3(syslog, int, type, char __user *, buf, int, len) @@ -479,7 +501,7 @@ static void _call_console_drivers(unsigned start, /* * Call the console drivers, asking them to write out * log_buf[start] to log_buf[end - 1]. - * The console_sem must be held. + * The console_lock must be held. */ static void call_console_drivers(unsigned start, unsigned end) { @@ -554,7 +576,7 @@ static void zap_locks(void) /* If a crash is occurring, make sure we can't deadlock */ spin_lock_init(&logbuf_lock); /* And make sure that we print immediately */ - init_MUTEX(&console_sem); + sema_init(&console_sem, 1); } #if defined(CONFIG_PRINTK_TIME) @@ -582,11 +604,11 @@ static int have_callable_console(void) * * This is printk(). It can be called from any context. We want it to work. * - * We try to grab the console_sem. If we succeed, it's easy - we log the output and + * We try to grab the console_lock. If we succeed, it's easy - we log the output and * call the console drivers. If we fail to get the semaphore we place the output * into the log buffer and return. The current holder of the console_sem will - * notice the new output in release_console_sem() and will send it to the - * consoles before releasing the semaphore. + * notice the new output in console_unlock(); and will send it to the + * consoles before releasing the lock. * * One effect of this deferred printing is that code which calls printk() and * then changes console_loglevel may break. This is because console_loglevel @@ -637,18 +659,19 @@ static inline int can_use_console(unsigned int cpu) /* * Try to get console ownership to actually show the kernel * messages from a 'printk'. Return true (and with the - * console_semaphore held, and 'console_locked' set) if it + * console_lock held, and 'console_locked' set) if it * is successful, false otherwise. * * This gets called with the 'logbuf_lock' spinlock held and * interrupts disabled. It should return with 'lockbuf_lock' * released but interrupts still disabled. */ -static int acquire_console_semaphore_for_printk(unsigned int cpu) +static int console_trylock_for_printk(unsigned int cpu) + __releases(&logbuf_lock) { int retval = 0; - if (!try_acquire_console_sem()) { + if (console_trylock()) { retval = 1; /* @@ -804,12 +827,12 @@ asmlinkage int vprintk(const char *fmt, va_list args) * actual magic (print out buffers, wake up klogd, * etc). * - * The acquire_console_semaphore_for_printk() function + * The console_trylock_for_printk() function * will release 'logbuf_lock' regardless of whether it * actually gets the semaphore or not. */ - if (acquire_console_semaphore_for_printk(this_cpu)) - release_console_sem(); + if (console_trylock_for_printk(this_cpu)) + console_unlock(); lockdep_on(); out_restore_irqs: @@ -970,7 +993,7 @@ void suspend_console(void) if (!console_suspend_enabled) return; printk("Suspending console(s) (use no_console_suspend to debug)\n"); - acquire_console_sem(); + console_lock(); console_suspended = 1; up(&console_sem); } @@ -981,18 +1004,44 @@ void resume_console(void) return; down(&console_sem); console_suspended = 0; - release_console_sem(); + console_unlock(); +} + +/** + * console_cpu_notify - print deferred console messages after CPU hotplug + * @self: notifier struct + * @action: CPU hotplug event + * @hcpu: unused + * + * If printk() is called from a CPU that is not online yet, the messages + * will be spooled but will not show up on the console. This function is + * called when a new CPU comes online (or fails to come up), and ensures + * that any such output gets printed. + */ +static int __cpuinit console_cpu_notify(struct notifier_block *self, + unsigned long action, void *hcpu) +{ + switch (action) { + case CPU_ONLINE: + case CPU_DEAD: + case CPU_DYING: + case CPU_DOWN_FAILED: + case CPU_UP_CANCELED: + console_lock(); + console_unlock(); + } + return NOTIFY_OK; } /** - * acquire_console_sem - lock the console system for exclusive use. + * console_lock - lock the console system for exclusive use. * - * Acquires a semaphore which guarantees that the caller has + * Acquires a lock which guarantees that the caller has * exclusive access to the console system and the console_drivers list. * * Can sleep, returns nothing. */ -void acquire_console_sem(void) +void console_lock(void) { BUG_ON(in_interrupt()); down(&console_sem); @@ -1001,21 +1050,29 @@ void acquire_console_sem(void) console_locked = 1; console_may_schedule = 1; } -EXPORT_SYMBOL(acquire_console_sem); +EXPORT_SYMBOL(console_lock); -int try_acquire_console_sem(void) +/** + * console_trylock - try to lock the console system for exclusive use. + * + * Tried to acquire a lock which guarantees that the caller has + * exclusive access to the console system and the console_drivers list. + * + * returns 1 on success, and 0 on failure to acquire the lock. + */ +int console_trylock(void) { if (down_trylock(&console_sem)) - return -1; + return 0; if (console_suspended) { up(&console_sem); - return -1; + return 0; } console_locked = 1; console_may_schedule = 0; - return 0; + return 1; } -EXPORT_SYMBOL(try_acquire_console_sem); +EXPORT_SYMBOL(console_trylock); int is_console_locked(void) { @@ -1026,38 +1083,40 @@ static DEFINE_PER_CPU(int, printk_pending); void printk_tick(void) { - if (__get_cpu_var(printk_pending)) { - __get_cpu_var(printk_pending) = 0; + if (__this_cpu_read(printk_pending)) { + __this_cpu_write(printk_pending, 0); wake_up_interruptible(&log_wait); } } int printk_needs_cpu(int cpu) { - return per_cpu(printk_pending, cpu); + if (cpu_is_offline(cpu)) + printk_tick(); + return __this_cpu_read(printk_pending); } void wake_up_klogd(void) { if (waitqueue_active(&log_wait)) - __raw_get_cpu_var(printk_pending) = 1; + this_cpu_write(printk_pending, 1); } /** - * release_console_sem - unlock the console system + * console_unlock - unlock the console system * - * Releases the semaphore which the caller holds on the console system + * Releases the console_lock which the caller holds on the console system * and the console driver list. * - * While the semaphore was held, console output may have been buffered - * by printk(). If this is the case, release_console_sem() emits - * the output prior to releasing the semaphore. + * While the console_lock was held, console output may have been buffered + * by printk(). If this is the case, console_unlock(); emits + * the output prior to releasing the lock. * * If there is output waiting for klogd, we wake it up. * - * release_console_sem() may be called from any context. + * console_unlock(); may be called from any context. */ -void release_console_sem(void) +void console_unlock(void) { unsigned long flags; unsigned _con_start, _log_end; @@ -1090,7 +1149,7 @@ void release_console_sem(void) if (wake_klogd) wake_up_klogd(); } -EXPORT_SYMBOL(release_console_sem); +EXPORT_SYMBOL(console_unlock); /** * console_conditional_schedule - yield the CPU if required @@ -1099,7 +1158,7 @@ EXPORT_SYMBOL(release_console_sem); * if this CPU should yield the CPU to another task, do * so here. * - * Must be called within acquire_console_sem(). + * Must be called within console_lock();. */ void __sched console_conditional_schedule(void) { @@ -1120,14 +1179,14 @@ void console_unblank(void) if (down_trylock(&console_sem) != 0) return; } else - acquire_console_sem(); + console_lock(); console_locked = 1; console_may_schedule = 0; for_each_console(c) if ((c->flags & CON_ENABLED) && c->unblank) c->unblank(); - release_console_sem(); + console_unlock(); } /* @@ -1138,7 +1197,7 @@ struct tty_driver *console_device(int *index) struct console *c; struct tty_driver *driver = NULL; - acquire_console_sem(); + console_lock(); for_each_console(c) { if (!c->device) continue; @@ -1146,7 +1205,7 @@ struct tty_driver *console_device(int *index) if (driver) break; } - release_console_sem(); + console_unlock(); return driver; } @@ -1157,17 +1216,17 @@ struct tty_driver *console_device(int *index) */ void console_stop(struct console *console) { - acquire_console_sem(); + console_lock(); console->flags &= ~CON_ENABLED; - release_console_sem(); + console_unlock(); } EXPORT_SYMBOL(console_stop); void console_start(struct console *console) { - acquire_console_sem(); + console_lock(); console->flags |= CON_ENABLED; - release_console_sem(); + console_unlock(); } EXPORT_SYMBOL(console_start); @@ -1289,7 +1348,7 @@ void register_console(struct console *newcon) * Put this console in the list - keep the * preferred driver at the head of the list. */ - acquire_console_sem(); + console_lock(); if ((newcon->flags & CON_CONSDEV) || console_drivers == NULL) { newcon->next = console_drivers; console_drivers = newcon; @@ -1301,14 +1360,15 @@ void register_console(struct console *newcon) } if (newcon->flags & CON_PRINTBUFFER) { /* - * release_console_sem() will print out the buffered messages + * console_unlock(); will print out the buffered messages * for us. */ spin_lock_irqsave(&logbuf_lock, flags); con_start = log_start; spin_unlock_irqrestore(&logbuf_lock, flags); } - release_console_sem(); + console_unlock(); + console_sysfs_notify(); /* * By unregistering the bootconsoles after we enable the real console @@ -1344,7 +1404,7 @@ int unregister_console(struct console *console) return braille_unregister_console(console); #endif - acquire_console_sem(); + console_lock(); if (console_drivers == console) { console_drivers=console->next; res = 0; @@ -1366,12 +1426,13 @@ int unregister_console(struct console *console) if (console_drivers != NULL && console->flags & CON_CONSDEV) console_drivers->flags |= CON_CONSDEV; - release_console_sem(); + console_unlock(); + console_sysfs_notify(); return res; } EXPORT_SYMBOL(unregister_console); -static int __init disable_boot_consoles(void) +static int __init printk_late_init(void) { struct console *con; @@ -1382,9 +1443,10 @@ static int __init disable_boot_consoles(void) unregister_console(con); } } + hotcpu_notifier(console_cpu_notify, 0); return 0; } -late_initcall(disable_boot_consoles); +late_initcall(printk_late_init); #if defined CONFIG_PRINTK @@ -1449,7 +1511,7 @@ int kmsg_dump_register(struct kmsg_dumper *dumper) /* Don't allow registering multiple times */ if (!dumper->registered) { dumper->registered = 1; - list_add_tail(&dumper->list, &dump_list); + list_add_tail_rcu(&dumper->list, &dump_list); err = 0; } spin_unlock_irqrestore(&dump_list_lock, flags); @@ -1473,29 +1535,16 @@ int kmsg_dump_unregister(struct kmsg_dumper *dumper) spin_lock_irqsave(&dump_list_lock, flags); if (dumper->registered) { dumper->registered = 0; - list_del(&dumper->list); + list_del_rcu(&dumper->list); err = 0; } spin_unlock_irqrestore(&dump_list_lock, flags); + synchronize_rcu(); return err; } EXPORT_SYMBOL_GPL(kmsg_dump_unregister); -static const char const *kmsg_reasons[] = { - [KMSG_DUMP_OOPS] = "oops", - [KMSG_DUMP_PANIC] = "panic", - [KMSG_DUMP_KEXEC] = "kexec", -}; - -static const char *kmsg_to_str(enum kmsg_dump_reason reason) -{ - if (reason >= ARRAY_SIZE(kmsg_reasons) || reason < 0) - return "unknown"; - - return kmsg_reasons[reason]; -} - /** * kmsg_dump - dump kernel log to kernel message dumpers. * @reason: the reason (oops, panic etc) for dumping @@ -1520,9 +1569,9 @@ void kmsg_dump(enum kmsg_dump_reason reason) chars = logged_chars; spin_unlock_irqrestore(&logbuf_lock, flags); - if (logged_chars > end) { - s1 = log_buf + log_buf_len - logged_chars + end; - l1 = logged_chars - end; + if (chars > end) { + s1 = log_buf + log_buf_len - chars + end; + l1 = chars - end; s2 = log_buf; l2 = end; @@ -1530,17 +1579,13 @@ void kmsg_dump(enum kmsg_dump_reason reason) s1 = ""; l1 = 0; - s2 = log_buf + end - logged_chars; - l2 = logged_chars; + s2 = log_buf + end - chars; + l2 = chars; } - if (!spin_trylock_irqsave(&dump_list_lock, flags)) { - printk(KERN_ERR "dump_kmsg: dump list lock is held during %s, skipping dump\n", - kmsg_to_str(reason)); - return; - } - list_for_each_entry(dumper, &dump_list, list) + rcu_read_lock(); + list_for_each_entry_rcu(dumper, &dump_list, list) dumper->dump(dumper, reason, s1, l1, s2, l2); - spin_unlock_irqrestore(&dump_list_lock, flags); + rcu_read_unlock(); } #endif diff --git a/kernel/profile.c b/kernel/profile.c index dfadc5b729f..66f841b7fbd 100644 --- a/kernel/profile.c +++ b/kernel/profile.c @@ -365,14 +365,14 @@ static int __cpuinit profile_cpu_callback(struct notifier_block *info, switch (action) { case CPU_UP_PREPARE: case CPU_UP_PREPARE_FROZEN: - node = cpu_to_node(cpu); + node = cpu_to_mem(cpu); per_cpu(cpu_profile_flip, cpu) = 0; if (!per_cpu(cpu_profile_hits, cpu)[1]) { page = alloc_pages_exact_node(node, GFP_KERNEL | __GFP_ZERO, 0); if (!page) - return NOTIFY_BAD; + return notifier_from_errno(-ENOMEM); per_cpu(cpu_profile_hits, cpu)[1] = page_address(page); } if (!per_cpu(cpu_profile_hits, cpu)[0]) { @@ -388,7 +388,7 @@ out_free: page = virt_to_page(per_cpu(cpu_profile_hits, cpu)[1]); per_cpu(cpu_profile_hits, cpu)[1] = NULL; __free_page(page); - return NOTIFY_BAD; + return notifier_from_errno(-ENOMEM); case CPU_ONLINE: case CPU_ONLINE_FROZEN: if (prof_cpu_mask != NULL) @@ -555,6 +555,7 @@ static ssize_t write_profile(struct file *file, const char __user *buf, static const struct file_operations proc_profile_operations = { .read = read_profile, .write = write_profile, + .llseek = default_llseek, }; #ifdef CONFIG_SMP @@ -567,7 +568,7 @@ static int create_hash_tables(void) int cpu; for_each_online_cpu(cpu) { - int node = cpu_to_node(cpu); + int node = cpu_to_mem(cpu); struct page *page; page = alloc_pages_exact_node(node, diff --git a/kernel/ptrace.c b/kernel/ptrace.c index 6af9cdd558b..99bbaa3e5b0 100644 --- a/kernel/ptrace.c +++ b/kernel/ptrace.c @@ -181,7 +181,7 @@ int ptrace_attach(struct task_struct *task) * under ptrace. */ retval = -ERESTARTNOINTR; - if (mutex_lock_interruptible(&task->cred_guard_mutex)) + if (mutex_lock_interruptible(&task->signal->cred_guard_mutex)) goto out; task_lock(task); @@ -208,7 +208,7 @@ int ptrace_attach(struct task_struct *task) unlock_tasklist: write_unlock_irq(&tasklist_lock); unlock_creds: - mutex_unlock(&task->cred_guard_mutex); + mutex_unlock(&task->signal->cred_guard_mutex); out: return retval; } @@ -324,26 +324,34 @@ int ptrace_detach(struct task_struct *child, unsigned int data) } /* - * Detach all tasks we were using ptrace on. + * Detach all tasks we were using ptrace on. Called with tasklist held + * for writing, and returns with it held too. But note it can release + * and reacquire the lock. */ void exit_ptrace(struct task_struct *tracer) + __releases(&tasklist_lock) + __acquires(&tasklist_lock) { struct task_struct *p, *n; LIST_HEAD(ptrace_dead); - write_lock_irq(&tasklist_lock); + if (likely(list_empty(&tracer->ptraced))) + return; + 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); + 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); } + + write_lock_irq(&tasklist_lock); } int ptrace_readdata(struct task_struct *tsk, unsigned long src, char __user *dst, int len) @@ -396,7 +404,7 @@ int ptrace_writedata(struct task_struct *tsk, char __user *src, unsigned long ds return copied; } -static int ptrace_setoptions(struct task_struct *child, long data) +static int ptrace_setoptions(struct task_struct *child, unsigned long data) { child->ptrace &= ~PT_TRACE_MASK; @@ -475,7 +483,8 @@ static int ptrace_setsiginfo(struct task_struct *child, const siginfo_t *info) #define is_sysemu_singlestep(request) 0 #endif -static int ptrace_resume(struct task_struct *child, long request, long data) +static int ptrace_resume(struct task_struct *child, long request, + unsigned long data) { if (!valid_signal(data)) return -EIO; @@ -552,10 +561,12 @@ static int ptrace_regset(struct task_struct *task, int req, unsigned int type, #endif int ptrace_request(struct task_struct *child, long request, - long addr, long data) + unsigned long addr, unsigned long data) { int ret = -EIO; siginfo_t siginfo; + void __user *datavp = (void __user *) data; + unsigned long __user *datalp = datavp; switch (request) { case PTRACE_PEEKTEXT: @@ -572,19 +583,17 @@ int ptrace_request(struct task_struct *child, long request, ret = ptrace_setoptions(child, data); break; case PTRACE_GETEVENTMSG: - ret = put_user(child->ptrace_message, (unsigned long __user *) data); + ret = put_user(child->ptrace_message, datalp); break; case PTRACE_GETSIGINFO: ret = ptrace_getsiginfo(child, &siginfo); if (!ret) - ret = copy_siginfo_to_user((siginfo_t __user *) data, - &siginfo); + ret = copy_siginfo_to_user(datavp, &siginfo); break; case PTRACE_SETSIGINFO: - if (copy_from_user(&siginfo, (siginfo_t __user *) data, - sizeof siginfo)) + if (copy_from_user(&siginfo, datavp, sizeof siginfo)) ret = -EFAULT; else ret = ptrace_setsiginfo(child, &siginfo); @@ -594,6 +603,32 @@ int ptrace_request(struct task_struct *child, long request, ret = ptrace_detach(child, data); break; +#ifdef CONFIG_BINFMT_ELF_FDPIC + case PTRACE_GETFDPIC: { + struct mm_struct *mm = get_task_mm(child); + unsigned long tmp = 0; + + ret = -ESRCH; + if (!mm) + break; + + switch (addr) { + case PTRACE_GETFDPIC_EXEC: + tmp = mm->context.exec_fdpic_loadmap; + break; + case PTRACE_GETFDPIC_INTERP: + tmp = mm->context.interp_fdpic_loadmap; + break; + default: + break; + } + mmput(mm); + + ret = put_user(tmp, datalp); + break; + } +#endif + #ifdef PTRACE_SINGLESTEP case PTRACE_SINGLESTEP: #endif @@ -618,7 +653,7 @@ int ptrace_request(struct task_struct *child, long request, case PTRACE_SETREGSET: { struct iovec kiov; - struct iovec __user *uiov = (struct iovec __user *) data; + struct iovec __user *uiov = datavp; if (!access_ok(VERIFY_WRITE, uiov, sizeof(*uiov))) return -EFAULT; @@ -659,7 +694,8 @@ static struct task_struct *ptrace_get_task_struct(pid_t pid) #define arch_ptrace_attach(child) do { } while (0) #endif -SYSCALL_DEFINE4(ptrace, long, request, long, pid, long, addr, long, data) +SYSCALL_DEFINE4(ptrace, long, request, long, pid, unsigned long, addr, + unsigned long, data) { struct task_struct *child; long ret; @@ -700,7 +736,8 @@ SYSCALL_DEFINE4(ptrace, long, request, long, pid, long, addr, long, data) return ret; } -int generic_ptrace_peekdata(struct task_struct *tsk, long addr, long data) +int generic_ptrace_peekdata(struct task_struct *tsk, unsigned long addr, + unsigned long data) { unsigned long tmp; int copied; @@ -711,7 +748,8 @@ int generic_ptrace_peekdata(struct task_struct *tsk, long addr, long data) return put_user(tmp, (unsigned long __user *)data); } -int generic_ptrace_pokedata(struct task_struct *tsk, long addr, long data) +int generic_ptrace_pokedata(struct task_struct *tsk, unsigned long addr, + unsigned long data) { int copied; diff --git a/kernel/range.c b/kernel/range.c index 74e2e611492..37fa9b99ad5 100644 --- a/kernel/range.c +++ b/kernel/range.c @@ -7,10 +7,6 @@ #include <linux/range.h> -#ifndef ARRAY_SIZE -#define ARRAY_SIZE(x) (sizeof(x) / sizeof((x)[0])) -#endif - int add_range(struct range *range, int az, int nr_range, u64 start, u64 end) { if (start >= end) @@ -123,7 +119,7 @@ static int cmp_range(const void *x1, const void *x2) int clean_sort_range(struct range *range, int az) { - int i, j, k = az - 1, nr_range = 0; + int i, j, k = az - 1, nr_range = az; for (i = 0; i < k; i++) { if (range[i].end) diff --git a/kernel/rcupdate.c b/kernel/rcupdate.c index 72a8dc9567f..a23a57a976d 100644 --- a/kernel/rcupdate.c +++ b/kernel/rcupdate.c @@ -73,12 +73,14 @@ int debug_lockdep_rcu_enabled(void) EXPORT_SYMBOL_GPL(debug_lockdep_rcu_enabled); /** - * rcu_read_lock_bh_held - might we be in RCU-bh read-side critical section? + * rcu_read_lock_bh_held() - might we be in RCU-bh read-side critical section? * * Check for bottom half being disabled, which covers both the * CONFIG_PROVE_RCU and not cases. Note that if someone uses * rcu_read_lock_bh(), but then later enables BH, lockdep (if enabled) - * will show the situation. + * will show the situation. This is useful for debug checks in functions + * that require that they be called within an RCU read-side critical + * section. * * Check debug_lockdep_rcu_enabled() to prevent false positives during boot. */ @@ -86,7 +88,7 @@ int rcu_read_lock_bh_held(void) { if (!debug_lockdep_rcu_enabled()) return 1; - return in_softirq(); + return in_softirq() || irqs_disabled(); } EXPORT_SYMBOL_GPL(rcu_read_lock_bh_held); @@ -114,3 +116,163 @@ int rcu_my_thread_group_empty(void) } EXPORT_SYMBOL_GPL(rcu_my_thread_group_empty); #endif /* #ifdef CONFIG_PROVE_RCU */ + +#ifdef CONFIG_DEBUG_OBJECTS_RCU_HEAD +static inline void debug_init_rcu_head(struct rcu_head *head) +{ + debug_object_init(head, &rcuhead_debug_descr); +} + +static inline void debug_rcu_head_free(struct rcu_head *head) +{ + debug_object_free(head, &rcuhead_debug_descr); +} + +/* + * fixup_init is called when: + * - an active object is initialized + */ +static int rcuhead_fixup_init(void *addr, enum debug_obj_state state) +{ + struct rcu_head *head = addr; + + switch (state) { + case ODEBUG_STATE_ACTIVE: + /* + * Ensure that queued callbacks are all executed. + * If we detect that we are nested in a RCU read-side critical + * section, we should simply fail, otherwise we would deadlock. + */ + if (rcu_preempt_depth() != 0 || preempt_count() != 0 || + irqs_disabled()) { + WARN_ON(1); + return 0; + } + rcu_barrier(); + rcu_barrier_sched(); + rcu_barrier_bh(); + debug_object_init(head, &rcuhead_debug_descr); + return 1; + default: + return 0; + } +} + +/* + * fixup_activate is called when: + * - an active object is activated + * - an unknown object is activated (might be a statically initialized object) + * Activation is performed internally by call_rcu(). + */ +static int rcuhead_fixup_activate(void *addr, enum debug_obj_state state) +{ + struct rcu_head *head = addr; + + switch (state) { + + case ODEBUG_STATE_NOTAVAILABLE: + /* + * This is not really a fixup. We just make sure that it is + * tracked in the object tracker. + */ + debug_object_init(head, &rcuhead_debug_descr); + debug_object_activate(head, &rcuhead_debug_descr); + return 0; + + case ODEBUG_STATE_ACTIVE: + /* + * Ensure that queued callbacks are all executed. + * If we detect that we are nested in a RCU read-side critical + * section, we should simply fail, otherwise we would deadlock. + */ + if (rcu_preempt_depth() != 0 || preempt_count() != 0 || + irqs_disabled()) { + WARN_ON(1); + return 0; + } + rcu_barrier(); + rcu_barrier_sched(); + rcu_barrier_bh(); + debug_object_activate(head, &rcuhead_debug_descr); + return 1; + default: + return 0; + } +} + +/* + * fixup_free is called when: + * - an active object is freed + */ +static int rcuhead_fixup_free(void *addr, enum debug_obj_state state) +{ + struct rcu_head *head = addr; + + switch (state) { + case ODEBUG_STATE_ACTIVE: + /* + * Ensure that queued callbacks are all executed. + * If we detect that we are nested in a RCU read-side critical + * section, we should simply fail, otherwise we would deadlock. + */ +#ifndef CONFIG_PREEMPT + WARN_ON(1); + return 0; +#else + if (rcu_preempt_depth() != 0 || preempt_count() != 0 || + irqs_disabled()) { + WARN_ON(1); + return 0; + } + rcu_barrier(); + rcu_barrier_sched(); + rcu_barrier_bh(); + debug_object_free(head, &rcuhead_debug_descr); + return 1; +#endif + default: + return 0; + } +} + +/** + * init_rcu_head_on_stack() - initialize on-stack rcu_head for debugobjects + * @head: pointer to rcu_head structure to be initialized + * + * This function informs debugobjects of a new rcu_head structure that + * has been allocated as an auto variable on the stack. This function + * is not required for rcu_head structures that are statically defined or + * that are dynamically allocated on the heap. This function has no + * effect for !CONFIG_DEBUG_OBJECTS_RCU_HEAD kernel builds. + */ +void init_rcu_head_on_stack(struct rcu_head *head) +{ + debug_object_init_on_stack(head, &rcuhead_debug_descr); +} +EXPORT_SYMBOL_GPL(init_rcu_head_on_stack); + +/** + * destroy_rcu_head_on_stack() - destroy on-stack rcu_head for debugobjects + * @head: pointer to rcu_head structure to be initialized + * + * This function informs debugobjects that an on-stack rcu_head structure + * is about to go out of scope. As with init_rcu_head_on_stack(), this + * function is not required for rcu_head structures that are statically + * defined or that are dynamically allocated on the heap. Also as with + * init_rcu_head_on_stack(), this function has no effect for + * !CONFIG_DEBUG_OBJECTS_RCU_HEAD kernel builds. + */ +void destroy_rcu_head_on_stack(struct rcu_head *head) +{ + debug_object_free(head, &rcuhead_debug_descr); +} +EXPORT_SYMBOL_GPL(destroy_rcu_head_on_stack); + +struct debug_obj_descr rcuhead_debug_descr = { + .name = "rcu_head", + .fixup_init = rcuhead_fixup_init, + .fixup_activate = rcuhead_fixup_activate, + .fixup_free = rcuhead_fixup_free, +}; +EXPORT_SYMBOL_GPL(rcuhead_debug_descr); +#endif /* #ifdef CONFIG_DEBUG_OBJECTS_RCU_HEAD */ diff --git a/kernel/rcutiny.c b/kernel/rcutiny.c index 38729d3cd23..0c343b9a46d 100644 --- a/kernel/rcutiny.c +++ b/kernel/rcutiny.c @@ -36,28 +36,21 @@ #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_sched_ctrlblk = { - .donetail = &rcu_sched_ctrlblk.rcucblist, - .curtail = &rcu_sched_ctrlblk.rcucblist, -}; - -static struct rcu_ctrlblk rcu_bh_ctrlblk = { - .donetail = &rcu_bh_ctrlblk.rcucblist, - .curtail = &rcu_bh_ctrlblk.rcucblist, -}; - -#ifdef CONFIG_DEBUG_LOCK_ALLOC -int rcu_scheduler_active __read_mostly; -EXPORT_SYMBOL_GPL(rcu_scheduler_active); -#endif /* #ifdef CONFIG_DEBUG_LOCK_ALLOC */ +/* Controls for rcu_kthread() kthread, replacing RCU_SOFTIRQ used previously. */ +static struct task_struct *rcu_kthread_task; +static DECLARE_WAIT_QUEUE_HEAD(rcu_kthread_wq); +static unsigned long have_rcu_kthread_work; +static void invoke_rcu_kthread(void); + +/* Forward declarations for rcutiny_plugin.h. */ +struct rcu_ctrlblk; +static void rcu_process_callbacks(struct rcu_ctrlblk *rcp); +static int rcu_kthread(void *arg); +static void __call_rcu(struct rcu_head *head, + void (*func)(struct rcu_head *rcu), + struct rcu_ctrlblk *rcp); + +#include "rcutiny_plugin.h" #ifdef CONFIG_NO_HZ @@ -115,7 +108,7 @@ void rcu_sched_qs(int cpu) { if (rcu_qsctr_help(&rcu_sched_ctrlblk) + rcu_qsctr_help(&rcu_bh_ctrlblk)) - raise_softirq(RCU_SOFTIRQ); + invoke_rcu_kthread(); } /* @@ -124,7 +117,7 @@ void rcu_sched_qs(int cpu) void rcu_bh_qs(int cpu) { if (rcu_qsctr_help(&rcu_bh_ctrlblk)) - raise_softirq(RCU_SOFTIRQ); + invoke_rcu_kthread(); } /* @@ -140,16 +133,18 @@ void rcu_check_callbacks(int cpu, int user) rcu_sched_qs(cpu); else if (!in_softirq()) rcu_bh_qs(cpu); + rcu_preempt_check_callbacks(); } /* - * Helper function for rcu_process_callbacks() that operates on the - * specified rcu_ctrlkblk structure. + * Invoke the RCU callbacks on the specified rcu_ctrlkblk structure + * whose grace period has elapsed. */ -static void __rcu_process_callbacks(struct rcu_ctrlblk *rcp) +static void rcu_process_callbacks(struct rcu_ctrlblk *rcp) { struct rcu_head *next, *list; unsigned long flags; + RCU_TRACE(int cb_count = 0); /* If no RCU callbacks ready to invoke, just return. */ if (&rcp->rcucblist == rcp->donetail) @@ -162,6 +157,7 @@ static void __rcu_process_callbacks(struct rcu_ctrlblk *rcp) *rcp->donetail = NULL; if (rcp->curtail == rcp->donetail) rcp->curtail = &rcp->rcucblist; + rcu_preempt_remove_callbacks(rcp); rcp->donetail = &rcp->rcucblist; local_irq_restore(flags); @@ -169,18 +165,60 @@ static void __rcu_process_callbacks(struct rcu_ctrlblk *rcp) while (list) { next = list->next; prefetch(next); + debug_rcu_head_unqueue(list); + local_bh_disable(); list->func(list); + local_bh_enable(); list = next; + RCU_TRACE(cb_count++); + } + RCU_TRACE(rcu_trace_sub_qlen(rcp, cb_count)); +} + +/* + * This kthread invokes RCU callbacks whose grace periods have + * elapsed. It is awakened as needed, and takes the place of the + * RCU_SOFTIRQ that was used previously for this purpose. + * This is a kthread, but it is never stopped, at least not until + * the system goes down. + */ +static int rcu_kthread(void *arg) +{ + unsigned long work; + unsigned long morework; + unsigned long flags; + + for (;;) { + wait_event_interruptible(rcu_kthread_wq, + have_rcu_kthread_work != 0); + morework = rcu_boost(); + local_irq_save(flags); + work = have_rcu_kthread_work; + have_rcu_kthread_work = morework; + local_irq_restore(flags); + if (work) { + rcu_process_callbacks(&rcu_sched_ctrlblk); + rcu_process_callbacks(&rcu_bh_ctrlblk); + rcu_preempt_process_callbacks(); + } + schedule_timeout_interruptible(1); /* Leave CPU for others. */ } + + return 0; /* Not reached, but needed to shut gcc up. */ } /* - * Invoke any callbacks whose grace period has completed. + * Wake up rcu_kthread() to process callbacks now eligible for invocation + * or to boost readers. */ -static void rcu_process_callbacks(struct softirq_action *unused) +static void invoke_rcu_kthread(void) { - __rcu_process_callbacks(&rcu_sched_ctrlblk); - __rcu_process_callbacks(&rcu_bh_ctrlblk); + unsigned long flags; + + local_irq_save(flags); + have_rcu_kthread_work = 1; + wake_up(&rcu_kthread_wq); + local_irq_restore(flags); } /* @@ -211,25 +249,27 @@ static void __call_rcu(struct rcu_head *head, { unsigned long flags; + debug_rcu_head_queue(head); head->func = func; head->next = NULL; local_irq_save(flags); *rcp->curtail = head; rcp->curtail = &head->next; + RCU_TRACE(rcp->qlen++); local_irq_restore(flags); } /* - * Post an RCU callback to be invoked after the end of an RCU grace + * Post an RCU callback to be invoked after the end of an RCU-sched 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)) +void call_rcu_sched(struct rcu_head *head, void (*func)(struct rcu_head *rcu)) { __call_rcu(head, func, &rcu_sched_ctrlblk); } -EXPORT_SYMBOL_GPL(call_rcu); +EXPORT_SYMBOL_GPL(call_rcu_sched); /* * Post an RCU bottom-half callback to be invoked after any subsequent @@ -241,20 +281,6 @@ void call_rcu_bh(struct rcu_head *head, void (*func)(struct rcu_head *rcu)) } EXPORT_SYMBOL_GPL(call_rcu_bh); -void rcu_barrier(void) -{ - struct rcu_synchronize rcu; - - init_rcu_head_on_stack(&rcu.head); - 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); - destroy_rcu_head_on_stack(&rcu.head); -} -EXPORT_SYMBOL_GPL(rcu_barrier); - void rcu_barrier_bh(void) { struct rcu_synchronize rcu; @@ -283,9 +309,16 @@ void rcu_barrier_sched(void) } EXPORT_SYMBOL_GPL(rcu_barrier_sched); -void __init rcu_init(void) +/* + * Spawn the kthread that invokes RCU callbacks. + */ +static int __init rcu_spawn_kthreads(void) { - open_softirq(RCU_SOFTIRQ, rcu_process_callbacks); -} + struct sched_param sp; -#include "rcutiny_plugin.h" + rcu_kthread_task = kthread_run(rcu_kthread, NULL, "rcu_kthread"); + sp.sched_priority = RCU_BOOST_PRIO; + sched_setscheduler_nocheck(rcu_kthread_task, SCHED_FIFO, &sp); + return 0; +} +early_initcall(rcu_spawn_kthreads); diff --git a/kernel/rcutiny_plugin.h b/kernel/rcutiny_plugin.h index d223a92bc74..015abaea962 100644 --- a/kernel/rcutiny_plugin.h +++ b/kernel/rcutiny_plugin.h @@ -1,7 +1,7 @@ /* - * Read-Copy Update mechanism for mutual exclusion (tree-based version) + * Read-Copy Update mechanism for mutual exclusion, the Bloatwatch edition * Internal non-public definitions that provide either classic - * or preemptable semantics. + * or preemptible 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 @@ -17,23 +17,1004 @@ * 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, 2009 + * Copyright (c) 2010 Linaro * * Author: Paul E. McKenney <paulmck@linux.vnet.ibm.com> */ +#include <linux/kthread.h> +#include <linux/debugfs.h> +#include <linux/seq_file.h> + +#ifdef CONFIG_RCU_TRACE +#define RCU_TRACE(stmt) stmt +#else /* #ifdef CONFIG_RCU_TRACE */ +#define RCU_TRACE(stmt) +#endif /* #else #ifdef CONFIG_RCU_TRACE */ + +/* 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. */ + RCU_TRACE(long qlen); /* Number of pending CBs. */ +}; + +/* Definition for rcupdate control block. */ +static struct rcu_ctrlblk rcu_sched_ctrlblk = { + .donetail = &rcu_sched_ctrlblk.rcucblist, + .curtail = &rcu_sched_ctrlblk.rcucblist, +}; + +static struct rcu_ctrlblk rcu_bh_ctrlblk = { + .donetail = &rcu_bh_ctrlblk.rcucblist, + .curtail = &rcu_bh_ctrlblk.rcucblist, +}; + +#ifdef CONFIG_DEBUG_LOCK_ALLOC +int rcu_scheduler_active __read_mostly; +EXPORT_SYMBOL_GPL(rcu_scheduler_active); +#endif /* #ifdef CONFIG_DEBUG_LOCK_ALLOC */ + +#ifdef CONFIG_TINY_PREEMPT_RCU + +#include <linux/delay.h> + +/* Global control variables for preemptible RCU. */ +struct rcu_preempt_ctrlblk { + struct rcu_ctrlblk rcb; /* curtail: ->next ptr of last CB for GP. */ + struct rcu_head **nexttail; + /* Tasks blocked in a preemptible RCU */ + /* read-side critical section while an */ + /* preemptible-RCU grace period is in */ + /* progress must wait for a later grace */ + /* period. This pointer points to the */ + /* ->next pointer of the last task that */ + /* must wait for a later grace period, or */ + /* to &->rcb.rcucblist if there is no */ + /* such task. */ + struct list_head blkd_tasks; + /* Tasks blocked in RCU read-side critical */ + /* section. Tasks are placed at the head */ + /* of this list and age towards the tail. */ + struct list_head *gp_tasks; + /* Pointer to the first task blocking the */ + /* current grace period, or NULL if there */ + /* is no such task. */ + struct list_head *exp_tasks; + /* Pointer to first task blocking the */ + /* current expedited grace period, or NULL */ + /* if there is no such task. If there */ + /* is no current expedited grace period, */ + /* then there cannot be any such task. */ +#ifdef CONFIG_RCU_BOOST + struct list_head *boost_tasks; + /* Pointer to first task that needs to be */ + /* priority-boosted, or NULL if no priority */ + /* boosting is needed. If there is no */ + /* current or expedited grace period, there */ + /* can be no such task. */ +#endif /* #ifdef CONFIG_RCU_BOOST */ + u8 gpnum; /* Current grace period. */ + u8 gpcpu; /* Last grace period blocked by the CPU. */ + u8 completed; /* Last grace period completed. */ + /* If all three are equal, RCU is idle. */ +#ifdef CONFIG_RCU_BOOST + s8 boosted_this_gp; /* Has boosting already happened? */ + unsigned long boost_time; /* When to start boosting (jiffies) */ +#endif /* #ifdef CONFIG_RCU_BOOST */ +#ifdef CONFIG_RCU_TRACE + unsigned long n_grace_periods; +#ifdef CONFIG_RCU_BOOST + unsigned long n_tasks_boosted; + unsigned long n_exp_boosts; + unsigned long n_normal_boosts; + unsigned long n_normal_balk_blkd_tasks; + unsigned long n_normal_balk_gp_tasks; + unsigned long n_normal_balk_boost_tasks; + unsigned long n_normal_balk_boosted; + unsigned long n_normal_balk_notyet; + unsigned long n_normal_balk_nos; + unsigned long n_exp_balk_blkd_tasks; + unsigned long n_exp_balk_nos; +#endif /* #ifdef CONFIG_RCU_BOOST */ +#endif /* #ifdef CONFIG_RCU_TRACE */ +}; + +static struct rcu_preempt_ctrlblk rcu_preempt_ctrlblk = { + .rcb.donetail = &rcu_preempt_ctrlblk.rcb.rcucblist, + .rcb.curtail = &rcu_preempt_ctrlblk.rcb.rcucblist, + .nexttail = &rcu_preempt_ctrlblk.rcb.rcucblist, + .blkd_tasks = LIST_HEAD_INIT(rcu_preempt_ctrlblk.blkd_tasks), +}; + +static int rcu_preempted_readers_exp(void); +static void rcu_report_exp_done(void); + +/* + * Return true if the CPU has not yet responded to the current grace period. + */ +static int rcu_cpu_blocking_cur_gp(void) +{ + return rcu_preempt_ctrlblk.gpcpu != rcu_preempt_ctrlblk.gpnum; +} + +/* + * Check for a running RCU reader. Because there is only one CPU, + * there can be but one running RCU reader at a time. ;-) + */ +static int rcu_preempt_running_reader(void) +{ + return current->rcu_read_lock_nesting; +} + +/* + * Check for preempted RCU readers blocking any grace period. + * If the caller needs a reliable answer, it must disable hard irqs. + */ +static int rcu_preempt_blocked_readers_any(void) +{ + return !list_empty(&rcu_preempt_ctrlblk.blkd_tasks); +} + +/* + * Check for preempted RCU readers blocking the current grace period. + * If the caller needs a reliable answer, it must disable hard irqs. + */ +static int rcu_preempt_blocked_readers_cgp(void) +{ + return rcu_preempt_ctrlblk.gp_tasks != NULL; +} + +/* + * Return true if another preemptible-RCU grace period is needed. + */ +static int rcu_preempt_needs_another_gp(void) +{ + return *rcu_preempt_ctrlblk.rcb.curtail != NULL; +} + +/* + * Return true if a preemptible-RCU grace period is in progress. + * The caller must disable hardirqs. + */ +static int rcu_preempt_gp_in_progress(void) +{ + return rcu_preempt_ctrlblk.completed != rcu_preempt_ctrlblk.gpnum; +} + +/* + * Advance a ->blkd_tasks-list pointer to the next entry, instead + * returning NULL if at the end of the list. + */ +static struct list_head *rcu_next_node_entry(struct task_struct *t) +{ + struct list_head *np; + + np = t->rcu_node_entry.next; + if (np == &rcu_preempt_ctrlblk.blkd_tasks) + np = NULL; + return np; +} + +#ifdef CONFIG_RCU_TRACE + +#ifdef CONFIG_RCU_BOOST +static void rcu_initiate_boost_trace(void); +static void rcu_initiate_exp_boost_trace(void); +#endif /* #ifdef CONFIG_RCU_BOOST */ + +/* + * Dump additional statistice for TINY_PREEMPT_RCU. + */ +static void show_tiny_preempt_stats(struct seq_file *m) +{ + seq_printf(m, "rcu_preempt: qlen=%ld gp=%lu g%u/p%u/c%u tasks=%c%c%c\n", + rcu_preempt_ctrlblk.rcb.qlen, + rcu_preempt_ctrlblk.n_grace_periods, + rcu_preempt_ctrlblk.gpnum, + rcu_preempt_ctrlblk.gpcpu, + rcu_preempt_ctrlblk.completed, + "T."[list_empty(&rcu_preempt_ctrlblk.blkd_tasks)], + "N."[!rcu_preempt_ctrlblk.gp_tasks], + "E."[!rcu_preempt_ctrlblk.exp_tasks]); +#ifdef CONFIG_RCU_BOOST + seq_printf(m, " ttb=%c btg=", + "B."[!rcu_preempt_ctrlblk.boost_tasks]); + switch (rcu_preempt_ctrlblk.boosted_this_gp) { + case -1: + seq_puts(m, "exp"); + break; + case 0: + seq_puts(m, "no"); + break; + case 1: + seq_puts(m, "begun"); + break; + case 2: + seq_puts(m, "done"); + break; + default: + seq_printf(m, "?%d?", rcu_preempt_ctrlblk.boosted_this_gp); + } + seq_printf(m, " ntb=%lu neb=%lu nnb=%lu j=%04x bt=%04x\n", + rcu_preempt_ctrlblk.n_tasks_boosted, + rcu_preempt_ctrlblk.n_exp_boosts, + rcu_preempt_ctrlblk.n_normal_boosts, + (int)(jiffies & 0xffff), + (int)(rcu_preempt_ctrlblk.boost_time & 0xffff)); + seq_printf(m, " %s: nt=%lu gt=%lu bt=%lu b=%lu ny=%lu nos=%lu\n", + "normal balk", + rcu_preempt_ctrlblk.n_normal_balk_blkd_tasks, + rcu_preempt_ctrlblk.n_normal_balk_gp_tasks, + rcu_preempt_ctrlblk.n_normal_balk_boost_tasks, + rcu_preempt_ctrlblk.n_normal_balk_boosted, + rcu_preempt_ctrlblk.n_normal_balk_notyet, + rcu_preempt_ctrlblk.n_normal_balk_nos); + seq_printf(m, " exp balk: bt=%lu nos=%lu\n", + rcu_preempt_ctrlblk.n_exp_balk_blkd_tasks, + rcu_preempt_ctrlblk.n_exp_balk_nos); +#endif /* #ifdef CONFIG_RCU_BOOST */ +} + +#endif /* #ifdef CONFIG_RCU_TRACE */ + +#ifdef CONFIG_RCU_BOOST + +#include "rtmutex_common.h" + +/* + * Carry out RCU priority boosting on the task indicated by ->boost_tasks, + * and advance ->boost_tasks to the next task in the ->blkd_tasks list. + */ +static int rcu_boost(void) +{ + unsigned long flags; + struct rt_mutex mtx; + struct list_head *np; + struct task_struct *t; + + if (rcu_preempt_ctrlblk.boost_tasks == NULL) + return 0; /* Nothing to boost. */ + raw_local_irq_save(flags); + rcu_preempt_ctrlblk.boosted_this_gp++; + t = container_of(rcu_preempt_ctrlblk.boost_tasks, struct task_struct, + rcu_node_entry); + np = rcu_next_node_entry(t); + rt_mutex_init_proxy_locked(&mtx, t); + t->rcu_boost_mutex = &mtx; + t->rcu_read_unlock_special |= RCU_READ_UNLOCK_BOOSTED; + raw_local_irq_restore(flags); + rt_mutex_lock(&mtx); + RCU_TRACE(rcu_preempt_ctrlblk.n_tasks_boosted++); + rcu_preempt_ctrlblk.boosted_this_gp++; + rt_mutex_unlock(&mtx); + return rcu_preempt_ctrlblk.boost_tasks != NULL; +} + +/* + * Check to see if it is now time to start boosting RCU readers blocking + * the current grace period, and, if so, tell the rcu_kthread_task to + * start boosting them. If there is an expedited boost in progress, + * we wait for it to complete. + * + * If there are no blocked readers blocking the current grace period, + * return 0 to let the caller know, otherwise return 1. Note that this + * return value is independent of whether or not boosting was done. + */ +static int rcu_initiate_boost(void) +{ + if (!rcu_preempt_blocked_readers_cgp()) { + RCU_TRACE(rcu_preempt_ctrlblk.n_normal_balk_blkd_tasks++); + return 0; + } + if (rcu_preempt_ctrlblk.gp_tasks != NULL && + rcu_preempt_ctrlblk.boost_tasks == NULL && + rcu_preempt_ctrlblk.boosted_this_gp == 0 && + ULONG_CMP_GE(jiffies, rcu_preempt_ctrlblk.boost_time)) { + rcu_preempt_ctrlblk.boost_tasks = rcu_preempt_ctrlblk.gp_tasks; + invoke_rcu_kthread(); + RCU_TRACE(rcu_preempt_ctrlblk.n_normal_boosts++); + } else + RCU_TRACE(rcu_initiate_boost_trace()); + return 1; +} + +/* + * Initiate boosting for an expedited grace period. + */ +static void rcu_initiate_expedited_boost(void) +{ + unsigned long flags; + + raw_local_irq_save(flags); + if (!list_empty(&rcu_preempt_ctrlblk.blkd_tasks)) { + rcu_preempt_ctrlblk.boost_tasks = + rcu_preempt_ctrlblk.blkd_tasks.next; + rcu_preempt_ctrlblk.boosted_this_gp = -1; + invoke_rcu_kthread(); + RCU_TRACE(rcu_preempt_ctrlblk.n_exp_boosts++); + } else + RCU_TRACE(rcu_initiate_exp_boost_trace()); + raw_local_irq_restore(flags); +} + +#define RCU_BOOST_DELAY_JIFFIES DIV_ROUND_UP(CONFIG_RCU_BOOST_DELAY * HZ, 1000); + +/* + * Do priority-boost accounting for the start of a new grace period. + */ +static void rcu_preempt_boost_start_gp(void) +{ + rcu_preempt_ctrlblk.boost_time = jiffies + RCU_BOOST_DELAY_JIFFIES; + if (rcu_preempt_ctrlblk.boosted_this_gp > 0) + rcu_preempt_ctrlblk.boosted_this_gp = 0; +} + +#else /* #ifdef CONFIG_RCU_BOOST */ + +/* + * If there is no RCU priority boosting, we don't boost. + */ +static int rcu_boost(void) +{ + return 0; +} + +/* + * If there is no RCU priority boosting, we don't initiate boosting, + * but we do indicate whether there are blocked readers blocking the + * current grace period. + */ +static int rcu_initiate_boost(void) +{ + return rcu_preempt_blocked_readers_cgp(); +} + +/* + * If there is no RCU priority boosting, we don't initiate expedited boosting. + */ +static void rcu_initiate_expedited_boost(void) +{ +} + +/* + * If there is no RCU priority boosting, nothing to do at grace-period start. + */ +static void rcu_preempt_boost_start_gp(void) +{ +} + +#endif /* else #ifdef CONFIG_RCU_BOOST */ + +/* + * Record a preemptible-RCU quiescent state for the specified CPU. Note + * that this just means that the task currently running on the CPU is + * in a quiescent state. There might be any number of tasks blocked + * while in an RCU read-side critical section. + * + * Unlike the other rcu_*_qs() functions, callers to this function + * must disable irqs in order to protect the assignment to + * ->rcu_read_unlock_special. + * + * Because this is a single-CPU implementation, the only way a grace + * period can end is if the CPU is in a quiescent state. The reason is + * that a blocked preemptible-RCU reader can exit its critical section + * only if the CPU is running it at the time. Therefore, when the + * last task blocking the current grace period exits its RCU read-side + * critical section, neither the CPU nor blocked tasks will be stopping + * the current grace period. (In contrast, SMP implementations + * might have CPUs running in RCU read-side critical sections that + * block later grace periods -- but this is not possible given only + * one CPU.) + */ +static void rcu_preempt_cpu_qs(void) +{ + /* Record both CPU and task as having responded to current GP. */ + rcu_preempt_ctrlblk.gpcpu = rcu_preempt_ctrlblk.gpnum; + current->rcu_read_unlock_special &= ~RCU_READ_UNLOCK_NEED_QS; + + /* If there is no GP then there is nothing more to do. */ + if (!rcu_preempt_gp_in_progress()) + return; + /* + * Check up on boosting. If there are no readers blocking the + * current grace period, leave. + */ + if (rcu_initiate_boost()) + return; + + /* Advance callbacks. */ + rcu_preempt_ctrlblk.completed = rcu_preempt_ctrlblk.gpnum; + rcu_preempt_ctrlblk.rcb.donetail = rcu_preempt_ctrlblk.rcb.curtail; + rcu_preempt_ctrlblk.rcb.curtail = rcu_preempt_ctrlblk.nexttail; + + /* If there are no blocked readers, next GP is done instantly. */ + if (!rcu_preempt_blocked_readers_any()) + rcu_preempt_ctrlblk.rcb.donetail = rcu_preempt_ctrlblk.nexttail; + + /* If there are done callbacks, cause them to be invoked. */ + if (*rcu_preempt_ctrlblk.rcb.donetail != NULL) + invoke_rcu_kthread(); +} + +/* + * Start a new RCU grace period if warranted. Hard irqs must be disabled. + */ +static void rcu_preempt_start_gp(void) +{ + if (!rcu_preempt_gp_in_progress() && rcu_preempt_needs_another_gp()) { + + /* Official start of GP. */ + rcu_preempt_ctrlblk.gpnum++; + RCU_TRACE(rcu_preempt_ctrlblk.n_grace_periods++); + + /* Any blocked RCU readers block new GP. */ + if (rcu_preempt_blocked_readers_any()) + rcu_preempt_ctrlblk.gp_tasks = + rcu_preempt_ctrlblk.blkd_tasks.next; + + /* Set up for RCU priority boosting. */ + rcu_preempt_boost_start_gp(); + + /* If there is no running reader, CPU is done with GP. */ + if (!rcu_preempt_running_reader()) + rcu_preempt_cpu_qs(); + } +} + +/* + * 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 blkd_tasks list. + * If the task started after the current grace period began, as recorded + * by ->gpcpu, we enqueue at the beginning of the list. Otherwise + * before the element referenced by ->gp_tasks (or at the tail if + * ->gp_tasks is NULL) and point ->gp_tasks at the newly added element. + * 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 ->gp_tasks pointer becomes + * NULL. + * + * Caller must disable preemption. + */ +void rcu_preempt_note_context_switch(void) +{ + struct task_struct *t = current; + unsigned long flags; + + local_irq_save(flags); /* must exclude scheduler_tick(). */ + if (rcu_preempt_running_reader() && + (t->rcu_read_unlock_special & RCU_READ_UNLOCK_BLOCKED) == 0) { + + /* Possibly blocking in an RCU read-side critical section. */ + t->rcu_read_unlock_special |= RCU_READ_UNLOCK_BLOCKED; + + /* + * 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. + */ + list_add(&t->rcu_node_entry, &rcu_preempt_ctrlblk.blkd_tasks); + if (rcu_cpu_blocking_cur_gp()) + rcu_preempt_ctrlblk.gp_tasks = &t->rcu_node_entry; + } + + /* + * 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 current grace period continues to be blocked. + */ + rcu_preempt_cpu_qs(); + local_irq_restore(flags); +} + +/* + * Tiny-preemptible 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) +{ + current->rcu_read_lock_nesting++; + barrier(); /* needed if we ever invoke rcu_read_lock in rcutiny.c */ +} +EXPORT_SYMBOL_GPL(__rcu_read_lock); + +/* + * 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 list_head *np; + int special; + + /* + * NMI handlers cannot block and cannot safely manipulate state. + * They therefore cannot possibly be special, so just leave. + */ + 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) + rcu_preempt_cpu_qs(); + + /* 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 ->blkd_tasks list and adjust + * any pointers that might have been referencing it. + */ + empty = !rcu_preempt_blocked_readers_cgp(); + empty_exp = rcu_preempt_ctrlblk.exp_tasks == NULL; + np = rcu_next_node_entry(t); + list_del(&t->rcu_node_entry); + if (&t->rcu_node_entry == rcu_preempt_ctrlblk.gp_tasks) + rcu_preempt_ctrlblk.gp_tasks = np; + if (&t->rcu_node_entry == rcu_preempt_ctrlblk.exp_tasks) + rcu_preempt_ctrlblk.exp_tasks = np; +#ifdef CONFIG_RCU_BOOST + if (&t->rcu_node_entry == rcu_preempt_ctrlblk.boost_tasks) + rcu_preempt_ctrlblk.boost_tasks = np; +#endif /* #ifdef CONFIG_RCU_BOOST */ + INIT_LIST_HEAD(&t->rcu_node_entry); + + /* + * If this was the last task on the current list, and if + * we aren't waiting on the CPU, report the quiescent state + * and start a new grace period if needed. + */ + if (!empty && !rcu_preempt_blocked_readers_cgp()) { + rcu_preempt_cpu_qs(); + rcu_preempt_start_gp(); + } + + /* + * If this was the last task on the expedited lists, + * then we need wake up the waiting task. + */ + if (!empty_exp && rcu_preempt_ctrlblk.exp_tasks == NULL) + rcu_report_exp_done(); + } +#ifdef CONFIG_RCU_BOOST + /* Unboost self if was boosted. */ + if (special & RCU_READ_UNLOCK_BOOSTED) { + t->rcu_read_unlock_special &= ~RCU_READ_UNLOCK_BOOSTED; + rt_mutex_unlock(t->rcu_boost_mutex); + t->rcu_boost_mutex = NULL; + } +#endif /* #ifdef CONFIG_RCU_BOOST */ + local_irq_restore(flags); +} + +/* + * Tiny-preemptible 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 rcutiny.c */ + --t->rcu_read_lock_nesting; + barrier(); /* decrement before load of ->rcu_read_unlock_special */ + if (t->rcu_read_lock_nesting == 0 && + unlikely(ACCESS_ONCE(t->rcu_read_unlock_special))) + rcu_read_unlock_special(t); +#ifdef CONFIG_PROVE_LOCKING + WARN_ON_ONCE(t->rcu_read_lock_nesting < 0); +#endif /* #ifdef CONFIG_PROVE_LOCKING */ +} +EXPORT_SYMBOL_GPL(__rcu_read_unlock); + +/* + * Check for a quiescent state from the current CPU. When a task blocks, + * the task is recorded in the rcu_preempt_ctrlblk structure, which is + * checked elsewhere. This is called from the scheduling-clock interrupt. + * + * Caller must disable hard irqs. + */ +static void rcu_preempt_check_callbacks(void) +{ + struct task_struct *t = current; + + if (rcu_preempt_gp_in_progress() && + (!rcu_preempt_running_reader() || + !rcu_cpu_blocking_cur_gp())) + rcu_preempt_cpu_qs(); + if (&rcu_preempt_ctrlblk.rcb.rcucblist != + rcu_preempt_ctrlblk.rcb.donetail) + invoke_rcu_kthread(); + if (rcu_preempt_gp_in_progress() && + rcu_cpu_blocking_cur_gp() && + rcu_preempt_running_reader()) + t->rcu_read_unlock_special |= RCU_READ_UNLOCK_NEED_QS; +} + +/* + * TINY_PREEMPT_RCU has an extra callback-list tail pointer to + * update, so this is invoked from rcu_process_callbacks() to + * handle that case. Of course, it is invoked for all flavors of + * RCU, but RCU callbacks can appear only on one of the lists, and + * neither ->nexttail nor ->donetail can possibly be NULL, so there + * is no need for an explicit check. + */ +static void rcu_preempt_remove_callbacks(struct rcu_ctrlblk *rcp) +{ + if (rcu_preempt_ctrlblk.nexttail == rcp->donetail) + rcu_preempt_ctrlblk.nexttail = &rcp->rcucblist; +} + +/* + * Process callbacks for preemptible RCU. + */ +static void rcu_preempt_process_callbacks(void) +{ + rcu_process_callbacks(&rcu_preempt_ctrlblk.rcb); +} + +/* + * Queue a preemptible -RCU callback for invocation after a grace period. + */ +void call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu)) +{ + unsigned long flags; + + debug_rcu_head_queue(head); + head->func = func; + head->next = NULL; + + local_irq_save(flags); + *rcu_preempt_ctrlblk.nexttail = head; + rcu_preempt_ctrlblk.nexttail = &head->next; + RCU_TRACE(rcu_preempt_ctrlblk.rcb.qlen++); + rcu_preempt_start_gp(); /* checks to see if GP needed. */ + local_irq_restore(flags); +} +EXPORT_SYMBOL_GPL(call_rcu); + +void rcu_barrier(void) +{ + struct rcu_synchronize rcu; + + init_rcu_head_on_stack(&rcu.head); + 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); + destroy_rcu_head_on_stack(&rcu.head); +} +EXPORT_SYMBOL_GPL(rcu_barrier); + +/* + * 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) +{ #ifdef CONFIG_DEBUG_LOCK_ALLOC + if (!rcu_scheduler_active) + return; +#endif /* #ifdef CONFIG_DEBUG_LOCK_ALLOC */ + WARN_ON_ONCE(rcu_preempt_running_reader()); + if (!rcu_preempt_blocked_readers_any()) + return; + + /* Once we get past the fastpath checks, same code as rcu_barrier(). */ + rcu_barrier(); +} +EXPORT_SYMBOL_GPL(synchronize_rcu); + +static DECLARE_WAIT_QUEUE_HEAD(sync_rcu_preempt_exp_wq); +static unsigned 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(void) +{ + return rcu_preempt_ctrlblk.exp_tasks != NULL; +} + +/* + * 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. + */ +static void rcu_report_exp_done(void) +{ + wake_up(&sync_rcu_preempt_exp_wq); +} + +/* + * Wait for an rcu-preempt grace period, but expedite it. The basic idea + * is to rely in the fact that there is but one CPU, and that it is + * illegal for a task to invoke synchronize_rcu_expedited() while in a + * preemptible-RCU read-side critical section. Therefore, any such + * critical sections must correspond to blocked tasks, which must therefore + * be on the ->blkd_tasks list. So just record the current head of the + * list in the ->exp_tasks pointer, and wait for all tasks including and + * after the task pointed to by ->exp_tasks to drain. + */ +void synchronize_rcu_expedited(void) +{ + unsigned long flags; + struct rcu_preempt_ctrlblk *rpcp = &rcu_preempt_ctrlblk; + unsigned long snap; + + barrier(); /* ensure prior action seen before grace period. */ + + WARN_ON_ONCE(rcu_preempt_running_reader()); + + /* + * Acquire lock so that there is only one preemptible RCU grace + * period in flight. Of course, if someone does the expedited + * grace period for us while we are acquiring the lock, just leave. + */ + snap = sync_rcu_preempt_exp_count + 1; + mutex_lock(&sync_rcu_preempt_exp_mutex); + if (ULONG_CMP_LT(snap, sync_rcu_preempt_exp_count)) + goto unlock_mb_ret; /* Others did our work for us. */ + + local_irq_save(flags); + + /* + * All RCU readers have to already be on blkd_tasks because + * we cannot legally be executing in an RCU read-side critical + * section. + */ + + /* Snapshot current head of ->blkd_tasks list. */ + rpcp->exp_tasks = rpcp->blkd_tasks.next; + if (rpcp->exp_tasks == &rpcp->blkd_tasks) + rpcp->exp_tasks = NULL; + local_irq_restore(flags); + + /* Wait for tail of ->blkd_tasks list to drain. */ + if (rcu_preempted_readers_exp()) + rcu_initiate_expedited_boost(); + wait_event(sync_rcu_preempt_exp_wq, + !rcu_preempted_readers_exp()); + + /* Clean up and exit. */ + barrier(); /* ensure expedited GP seen before counter increment. */ + sync_rcu_preempt_exp_count++; +unlock_mb_ret: + mutex_unlock(&sync_rcu_preempt_exp_mutex); + barrier(); /* ensure subsequent action seen after grace period. */ +} +EXPORT_SYMBOL_GPL(synchronize_rcu_expedited); + +/* + * Does preemptible RCU need the CPU to stay out of dynticks mode? + */ +int rcu_preempt_needs_cpu(void) +{ + if (!rcu_preempt_running_reader()) + rcu_preempt_cpu_qs(); + return rcu_preempt_ctrlblk.rcb.rcucblist != NULL; +} + +/* + * Check for a task exiting while in a preemptible -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_TINY_PREEMPT_RCU */ + +#ifdef CONFIG_RCU_TRACE + +/* + * Because preemptible RCU does not exist, it is not necessary to + * dump out its statistics. + */ +static void show_tiny_preempt_stats(struct seq_file *m) +{ +} + +#endif /* #ifdef CONFIG_RCU_TRACE */ + +/* + * Because preemptible RCU does not exist, it is never necessary to + * boost preempted RCU readers. + */ +static int rcu_boost(void) +{ + return 0; +} + +/* + * Because preemptible RCU does not exist, it never has any callbacks + * to check. + */ +static void rcu_preempt_check_callbacks(void) +{ +} + +/* + * Because preemptible RCU does not exist, it never has any callbacks + * to remove. + */ +static void rcu_preempt_remove_callbacks(struct rcu_ctrlblk *rcp) +{ +} + +/* + * Because preemptible RCU does not exist, it never has any callbacks + * to process. + */ +static void rcu_preempt_process_callbacks(void) +{ +} + +#endif /* #else #ifdef CONFIG_TINY_PREEMPT_RCU */ + +#ifdef CONFIG_DEBUG_LOCK_ALLOC #include <linux/kernel_stat.h> /* * During boot, we forgive RCU lockdep issues. After this function is * invoked, we start taking RCU lockdep issues seriously. */ -void rcu_scheduler_starting(void) +void __init rcu_scheduler_starting(void) { WARN_ON(nr_context_switches() > 0); rcu_scheduler_active = 1; } #endif /* #ifdef CONFIG_DEBUG_LOCK_ALLOC */ + +#ifdef CONFIG_RCU_BOOST +#define RCU_BOOST_PRIO CONFIG_RCU_BOOST_PRIO +#else /* #ifdef CONFIG_RCU_BOOST */ +#define RCU_BOOST_PRIO 1 +#endif /* #else #ifdef CONFIG_RCU_BOOST */ + +#ifdef CONFIG_RCU_TRACE + +#ifdef CONFIG_RCU_BOOST + +static void rcu_initiate_boost_trace(void) +{ + if (rcu_preempt_ctrlblk.gp_tasks == NULL) + rcu_preempt_ctrlblk.n_normal_balk_gp_tasks++; + else if (rcu_preempt_ctrlblk.boost_tasks != NULL) + rcu_preempt_ctrlblk.n_normal_balk_boost_tasks++; + else if (rcu_preempt_ctrlblk.boosted_this_gp != 0) + rcu_preempt_ctrlblk.n_normal_balk_boosted++; + else if (!ULONG_CMP_GE(jiffies, rcu_preempt_ctrlblk.boost_time)) + rcu_preempt_ctrlblk.n_normal_balk_notyet++; + else + rcu_preempt_ctrlblk.n_normal_balk_nos++; +} + +static void rcu_initiate_exp_boost_trace(void) +{ + if (list_empty(&rcu_preempt_ctrlblk.blkd_tasks)) + rcu_preempt_ctrlblk.n_exp_balk_blkd_tasks++; + else + rcu_preempt_ctrlblk.n_exp_balk_nos++; +} + +#endif /* #ifdef CONFIG_RCU_BOOST */ + +static void rcu_trace_sub_qlen(struct rcu_ctrlblk *rcp, int n) +{ + unsigned long flags; + + raw_local_irq_save(flags); + rcp->qlen -= n; + raw_local_irq_restore(flags); +} + +/* + * Dump statistics for TINY_RCU, such as they are. + */ +static int show_tiny_stats(struct seq_file *m, void *unused) +{ + show_tiny_preempt_stats(m); + seq_printf(m, "rcu_sched: qlen: %ld\n", rcu_sched_ctrlblk.qlen); + seq_printf(m, "rcu_bh: qlen: %ld\n", rcu_bh_ctrlblk.qlen); + return 0; +} + +static int show_tiny_stats_open(struct inode *inode, struct file *file) +{ + return single_open(file, show_tiny_stats, NULL); +} + +static const struct file_operations show_tiny_stats_fops = { + .owner = THIS_MODULE, + .open = show_tiny_stats_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; + +static struct dentry *rcudir; + +static int __init rcutiny_trace_init(void) +{ + struct dentry *retval; + + rcudir = debugfs_create_dir("rcu", NULL); + if (!rcudir) + goto free_out; + retval = debugfs_create_file("rcudata", 0444, rcudir, + NULL, &show_tiny_stats_fops); + if (!retval) + goto free_out; + return 0; +free_out: + debugfs_remove_recursive(rcudir); + return 1; +} + +static void __exit rcutiny_trace_cleanup(void) +{ + debugfs_remove_recursive(rcudir); +} + +module_init(rcutiny_trace_init); +module_exit(rcutiny_trace_cleanup); + +MODULE_AUTHOR("Paul E. McKenney"); +MODULE_DESCRIPTION("Read-Copy Update tracing for tiny implementation"); +MODULE_LICENSE("GPL"); + +#endif /* #ifdef CONFIG_RCU_TRACE */ diff --git a/kernel/rcutorture.c b/kernel/rcutorture.c index 6535ac8bc6a..89613f97ff2 100644 --- a/kernel/rcutorture.c +++ b/kernel/rcutorture.c @@ -47,6 +47,7 @@ #include <linux/srcu.h> #include <linux/slab.h> #include <asm/byteorder.h> +#include <linux/sched.h> MODULE_LICENSE("GPL"); MODULE_AUTHOR("Paul E. McKenney <paulmck@us.ibm.com> and " @@ -64,6 +65,9 @@ static int irqreader = 1; /* RCU readers from irq (timers). */ static int fqs_duration = 0; /* Duration of bursts (us), 0 to disable. */ static int fqs_holdoff = 0; /* Hold time within burst (us). */ static int fqs_stutter = 3; /* Wait time between bursts (s). */ +static int test_boost = 1; /* Test RCU prio boost: 0=no, 1=maybe, 2=yes. */ +static int test_boost_interval = 7; /* Interval between boost tests, seconds. */ +static int test_boost_duration = 4; /* Duration of each boost test, seconds. */ static char *torture_type = "rcu"; /* What RCU implementation to torture. */ module_param(nreaders, int, 0444); @@ -88,6 +92,12 @@ module_param(fqs_holdoff, int, 0444); MODULE_PARM_DESC(fqs_holdoff, "Holdoff time within fqs bursts (us)"); module_param(fqs_stutter, int, 0444); MODULE_PARM_DESC(fqs_stutter, "Wait time between fqs bursts (s)"); +module_param(test_boost, int, 0444); +MODULE_PARM_DESC(test_boost, "Test RCU prio boost: 0=no, 1=maybe, 2=yes."); +module_param(test_boost_interval, int, 0444); +MODULE_PARM_DESC(test_boost_interval, "Interval between boost tests, seconds."); +module_param(test_boost_duration, int, 0444); +MODULE_PARM_DESC(test_boost_duration, "Duration of each boost test, seconds."); module_param(torture_type, charp, 0444); MODULE_PARM_DESC(torture_type, "Type of RCU to torture (rcu, rcu_bh, srcu)"); @@ -109,6 +119,7 @@ static struct task_struct *stats_task; static struct task_struct *shuffler_task; static struct task_struct *stutter_task; static struct task_struct *fqs_task; +static struct task_struct *boost_tasks[NR_CPUS]; #define RCU_TORTURE_PIPE_LEN 10 @@ -120,7 +131,7 @@ struct rcu_torture { }; static LIST_HEAD(rcu_torture_freelist); -static struct rcu_torture *rcu_torture_current; +static struct rcu_torture __rcu *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); @@ -134,6 +145,12 @@ 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_boost_ktrerror; +static long n_rcu_torture_boost_rterror; +static long n_rcu_torture_boost_allocerror; +static long n_rcu_torture_boost_afferror; +static long n_rcu_torture_boost_failure; +static long n_rcu_torture_boosts; static long n_rcu_torture_timers; static struct list_head rcu_torture_removed; static cpumask_var_t shuffle_tmp_mask; @@ -147,14 +164,26 @@ static int stutter_pause_test; #endif int rcutorture_runnable = RCUTORTURE_RUNNABLE_INIT; +#ifdef CONFIG_RCU_BOOST +#define rcu_can_boost() 1 +#else /* #ifdef CONFIG_RCU_BOOST */ +#define rcu_can_boost() 0 +#endif /* #else #ifdef CONFIG_RCU_BOOST */ + +static unsigned long boost_starttime; /* jiffies of next boost test start. */ +DEFINE_MUTEX(boost_mutex); /* protect setting boost_starttime */ + /* and boost task create/destroy. */ + /* Mediate rmmod and system shutdown. Concurrent rmmod & shutdown illegal! */ #define FULLSTOP_DONTSTOP 0 /* Normal operation. */ #define FULLSTOP_SHUTDOWN 1 /* System shutdown with rcutorture running. */ #define FULLSTOP_RMMOD 2 /* Normal rmmod of rcutorture. */ static int fullstop = FULLSTOP_RMMOD; -DEFINE_MUTEX(fullstop_mutex); /* Protect fullstop transitions and spawning */ - /* of kthreads. */ +/* + * Protect fullstop transitions and spawning of kthreads. + */ +static DEFINE_MUTEX(fullstop_mutex); /* * Detect and respond to a system shutdown. @@ -239,8 +268,7 @@ static unsigned long rcu_random(struct rcu_random_state *rrsp) { if (--rrsp->rrs_count < 0) { - rrsp->rrs_state += - (unsigned long)cpu_clock(raw_smp_processor_id()); + rrsp->rrs_state += (unsigned long)local_clock(); rrsp->rrs_count = RCU_RANDOM_REFRESH; } rrsp->rrs_state = rrsp->rrs_state * RCU_RANDOM_MULT + RCU_RANDOM_ADD; @@ -276,6 +304,7 @@ struct rcu_torture_ops { void (*fqs)(void); int (*stats)(char *page); int irq_capable; + int can_boost; char *name; }; @@ -304,6 +333,10 @@ static void rcu_read_delay(struct rcu_random_state *rrsp) mdelay(longdelay_ms); if (!(rcu_random(rrsp) % (nrealreaders * 2 * shortdelay_us))) udelay(shortdelay_us); +#ifdef CONFIG_PREEMPT + if (!preempt_count() && !(rcu_random(rrsp) % (nrealreaders * 20000))) + preempt_schedule(); /* No QS if preempt_disable() in effect */ +#endif } static void rcu_torture_read_unlock(int idx) __releases(RCU) @@ -361,6 +394,7 @@ static struct rcu_torture_ops rcu_ops = { .fqs = rcu_force_quiescent_state, .stats = NULL, .irq_capable = 1, + .can_boost = rcu_can_boost(), .name = "rcu" }; @@ -403,6 +437,7 @@ static struct rcu_torture_ops rcu_sync_ops = { .fqs = rcu_force_quiescent_state, .stats = NULL, .irq_capable = 1, + .can_boost = rcu_can_boost(), .name = "rcu_sync" }; @@ -419,6 +454,7 @@ static struct rcu_torture_ops rcu_expedited_ops = { .fqs = rcu_force_quiescent_state, .stats = NULL, .irq_capable = 1, + .can_boost = rcu_can_boost(), .name = "rcu_expedited" }; @@ -537,6 +573,8 @@ static void srcu_read_delay(struct rcu_random_state *rrsp) delay = rcu_random(rrsp) % (nrealreaders * 2 * longdelay * uspertick); if (!delay) schedule_timeout_interruptible(longdelay); + else + rcu_read_delay(rrsp); } static void srcu_torture_read_unlock(int idx) __releases(&srcu_ctl) @@ -677,6 +715,110 @@ static struct rcu_torture_ops sched_expedited_ops = { }; /* + * RCU torture priority-boost testing. Runs one real-time thread per + * CPU for moderate bursts, repeatedly registering RCU callbacks and + * spinning waiting for them to be invoked. If a given callback takes + * too long to be invoked, we assume that priority inversion has occurred. + */ + +struct rcu_boost_inflight { + struct rcu_head rcu; + int inflight; +}; + +static void rcu_torture_boost_cb(struct rcu_head *head) +{ + struct rcu_boost_inflight *rbip = + container_of(head, struct rcu_boost_inflight, rcu); + + smp_mb(); /* Ensure RCU-core accesses precede clearing ->inflight */ + rbip->inflight = 0; +} + +static int rcu_torture_boost(void *arg) +{ + unsigned long call_rcu_time; + unsigned long endtime; + unsigned long oldstarttime; + struct rcu_boost_inflight rbi = { .inflight = 0 }; + struct sched_param sp; + + VERBOSE_PRINTK_STRING("rcu_torture_boost started"); + + /* Set real-time priority. */ + sp.sched_priority = 1; + if (sched_setscheduler(current, SCHED_FIFO, &sp) < 0) { + VERBOSE_PRINTK_STRING("rcu_torture_boost RT prio failed!"); + n_rcu_torture_boost_rterror++; + } + + /* Each pass through the following loop does one boost-test cycle. */ + do { + /* Wait for the next test interval. */ + oldstarttime = boost_starttime; + while (jiffies - oldstarttime > ULONG_MAX / 2) { + schedule_timeout_uninterruptible(1); + rcu_stutter_wait("rcu_torture_boost"); + if (kthread_should_stop() || + fullstop != FULLSTOP_DONTSTOP) + goto checkwait; + } + + /* Do one boost-test interval. */ + endtime = oldstarttime + test_boost_duration * HZ; + call_rcu_time = jiffies; + while (jiffies - endtime > ULONG_MAX / 2) { + /* If we don't have a callback in flight, post one. */ + if (!rbi.inflight) { + smp_mb(); /* RCU core before ->inflight = 1. */ + rbi.inflight = 1; + call_rcu(&rbi.rcu, rcu_torture_boost_cb); + if (jiffies - call_rcu_time > + test_boost_duration * HZ - HZ / 2) { + VERBOSE_PRINTK_STRING("rcu_torture_boost boosting failed"); + n_rcu_torture_boost_failure++; + } + call_rcu_time = jiffies; + } + cond_resched(); + rcu_stutter_wait("rcu_torture_boost"); + if (kthread_should_stop() || + fullstop != FULLSTOP_DONTSTOP) + goto checkwait; + } + + /* + * Set the start time of the next test interval. + * Yes, this is vulnerable to long delays, but such + * delays simply cause a false negative for the next + * interval. Besides, we are running at RT priority, + * so delays should be relatively rare. + */ + while (oldstarttime == boost_starttime) { + if (mutex_trylock(&boost_mutex)) { + boost_starttime = jiffies + + test_boost_interval * HZ; + n_rcu_torture_boosts++; + mutex_unlock(&boost_mutex); + break; + } + schedule_timeout_uninterruptible(1); + } + + /* Go do the stutter. */ +checkwait: rcu_stutter_wait("rcu_torture_boost"); + } while (!kthread_should_stop() && fullstop == FULLSTOP_DONTSTOP); + + /* Clean up and exit. */ + VERBOSE_PRINTK_STRING("rcu_torture_boost task stopping"); + rcutorture_shutdown_absorb("rcu_torture_boost"); + while (!kthread_should_stop() || rbi.inflight) + schedule_timeout_uninterruptible(1); + smp_mb(); /* order accesses to ->inflight before stack-frame death. */ + return 0; +} + +/* * RCU torture force-quiescent-state kthread. Repeatedly induces * bursts of calls to force_quiescent_state(), increasing the probability * of occurrence of some important types of race conditions. @@ -732,7 +874,8 @@ rcu_torture_writer(void *arg) continue; rp->rtort_pipe_count = 0; udelay(rcu_random(&rand) & 0x3ff); - old_rp = rcu_torture_current; + old_rp = rcu_dereference_check(rcu_torture_current, + current == writer_task); rp->rtort_mbtest = 1; rcu_assign_pointer(rcu_torture_current, rp); smp_wmb(); /* Mods to old_rp must follow rcu_assign_pointer() */ @@ -925,7 +1068,8 @@ rcu_torture_printk(char *page) cnt += sprintf(&page[cnt], "%s%s ", torture_type, TORTURE_FLAG); cnt += sprintf(&page[cnt], "rtc: %p ver: %ld tfle: %d rta: %d rtaf: %d rtf: %d " - "rtmbe: %d nt: %ld", + "rtmbe: %d rtbke: %ld rtbre: %ld rtbae: %ld rtbafe: %ld " + "rtbf: %ld rtb: %ld nt: %ld", rcu_torture_current, rcu_torture_current_version, list_empty(&rcu_torture_freelist), @@ -933,8 +1077,19 @@ rcu_torture_printk(char *page) atomic_read(&n_rcu_torture_alloc_fail), atomic_read(&n_rcu_torture_free), atomic_read(&n_rcu_torture_mberror), + n_rcu_torture_boost_ktrerror, + n_rcu_torture_boost_rterror, + n_rcu_torture_boost_allocerror, + n_rcu_torture_boost_afferror, + n_rcu_torture_boost_failure, + n_rcu_torture_boosts, n_rcu_torture_timers); - if (atomic_read(&n_rcu_torture_mberror) != 0) + if (atomic_read(&n_rcu_torture_mberror) != 0 || + n_rcu_torture_boost_ktrerror != 0 || + n_rcu_torture_boost_rterror != 0 || + n_rcu_torture_boost_allocerror != 0 || + n_rcu_torture_boost_afferror != 0 || + n_rcu_torture_boost_failure != 0) cnt += sprintf(&page[cnt], " !!!"); cnt += sprintf(&page[cnt], "\n%s%s ", torture_type, TORTURE_FLAG); if (i > 1) { @@ -1086,22 +1241,91 @@ rcu_torture_stutter(void *arg) } static inline void -rcu_torture_print_module_parms(char *tag) +rcu_torture_print_module_parms(struct rcu_torture_ops *cur_ops, char *tag) { printk(KERN_ALERT "%s" TORTURE_FLAG "--- %s: nreaders=%d nfakewriters=%d " "stat_interval=%d verbose=%d test_no_idle_hz=%d " "shuffle_interval=%d stutter=%d irqreader=%d " - "fqs_duration=%d fqs_holdoff=%d fqs_stutter=%d\n", + "fqs_duration=%d fqs_holdoff=%d fqs_stutter=%d " + "test_boost=%d/%d test_boost_interval=%d " + "test_boost_duration=%d\n", torture_type, tag, nrealreaders, nfakewriters, stat_interval, verbose, test_no_idle_hz, shuffle_interval, - stutter, irqreader, fqs_duration, fqs_holdoff, fqs_stutter); + stutter, irqreader, fqs_duration, fqs_holdoff, fqs_stutter, + test_boost, cur_ops->can_boost, + test_boost_interval, test_boost_duration); } -static struct notifier_block rcutorture_nb = { +static struct notifier_block rcutorture_shutdown_nb = { .notifier_call = rcutorture_shutdown_notify, }; +static void rcutorture_booster_cleanup(int cpu) +{ + struct task_struct *t; + + if (boost_tasks[cpu] == NULL) + return; + mutex_lock(&boost_mutex); + VERBOSE_PRINTK_STRING("Stopping rcu_torture_boost task"); + t = boost_tasks[cpu]; + boost_tasks[cpu] = NULL; + mutex_unlock(&boost_mutex); + + /* This must be outside of the mutex, otherwise deadlock! */ + kthread_stop(t); +} + +static int rcutorture_booster_init(int cpu) +{ + int retval; + + if (boost_tasks[cpu] != NULL) + return 0; /* Already created, nothing more to do. */ + + /* Don't allow time recalculation while creating a new task. */ + mutex_lock(&boost_mutex); + VERBOSE_PRINTK_STRING("Creating rcu_torture_boost task"); + boost_tasks[cpu] = kthread_create(rcu_torture_boost, NULL, + "rcu_torture_boost"); + if (IS_ERR(boost_tasks[cpu])) { + retval = PTR_ERR(boost_tasks[cpu]); + VERBOSE_PRINTK_STRING("rcu_torture_boost task create failed"); + n_rcu_torture_boost_ktrerror++; + boost_tasks[cpu] = NULL; + mutex_unlock(&boost_mutex); + return retval; + } + kthread_bind(boost_tasks[cpu], cpu); + wake_up_process(boost_tasks[cpu]); + mutex_unlock(&boost_mutex); + return 0; +} + +static int rcutorture_cpu_notify(struct notifier_block *self, + unsigned long action, void *hcpu) +{ + long cpu = (long)hcpu; + + switch (action) { + case CPU_ONLINE: + case CPU_DOWN_FAILED: + (void)rcutorture_booster_init(cpu); + break; + case CPU_DOWN_PREPARE: + rcutorture_booster_cleanup(cpu); + break; + default: + break; + } + return NOTIFY_OK; +} + +static struct notifier_block rcutorture_cpu_nb = { + .notifier_call = rcutorture_cpu_notify, +}; + static void rcu_torture_cleanup(void) { @@ -1119,7 +1343,7 @@ rcu_torture_cleanup(void) } fullstop = FULLSTOP_RMMOD; mutex_unlock(&fullstop_mutex); - unregister_reboot_notifier(&rcutorture_nb); + unregister_reboot_notifier(&rcutorture_shutdown_nb); if (stutter_task) { VERBOSE_PRINTK_STRING("Stopping rcu_torture_stutter task"); kthread_stop(stutter_task); @@ -1176,6 +1400,12 @@ rcu_torture_cleanup(void) kthread_stop(fqs_task); } fqs_task = NULL; + if ((test_boost == 1 && cur_ops->can_boost) || + test_boost == 2) { + unregister_cpu_notifier(&rcutorture_cpu_nb); + for_each_possible_cpu(i) + rcutorture_booster_cleanup(i); + } /* Wait for all RCU callbacks to fire. */ @@ -1187,9 +1417,9 @@ rcu_torture_cleanup(void) if (cur_ops->cleanup) cur_ops->cleanup(); if (atomic_read(&n_rcu_torture_error)) - rcu_torture_print_module_parms("End of test: FAILURE"); + rcu_torture_print_module_parms(cur_ops, "End of test: FAILURE"); else - rcu_torture_print_module_parms("End of test: SUCCESS"); + rcu_torture_print_module_parms(cur_ops, "End of test: SUCCESS"); } static int __init @@ -1234,7 +1464,7 @@ rcu_torture_init(void) nrealreaders = nreaders; else nrealreaders = 2 * num_online_cpus(); - rcu_torture_print_module_parms("Start of test"); + rcu_torture_print_module_parms(cur_ops, "Start of test"); fullstop = FULLSTOP_DONTSTOP; /* Set up the freelist. */ @@ -1255,6 +1485,12 @@ rcu_torture_init(void) atomic_set(&n_rcu_torture_free, 0); atomic_set(&n_rcu_torture_mberror, 0); atomic_set(&n_rcu_torture_error, 0); + n_rcu_torture_boost_ktrerror = 0; + n_rcu_torture_boost_rterror = 0; + n_rcu_torture_boost_allocerror = 0; + n_rcu_torture_boost_afferror = 0; + n_rcu_torture_boost_failure = 0; + n_rcu_torture_boosts = 0; for (i = 0; i < RCU_TORTURE_PIPE_LEN + 1; i++) atomic_set(&rcu_torture_wcount[i], 0); for_each_possible_cpu(cpu) { @@ -1368,7 +1604,27 @@ rcu_torture_init(void) goto unwind; } } - register_reboot_notifier(&rcutorture_nb); + if (test_boost_interval < 1) + test_boost_interval = 1; + if (test_boost_duration < 2) + test_boost_duration = 2; + if ((test_boost == 1 && cur_ops->can_boost) || + test_boost == 2) { + int retval; + + boost_starttime = jiffies + test_boost_interval * HZ; + register_cpu_notifier(&rcutorture_cpu_nb); + for_each_possible_cpu(i) { + if (cpu_is_offline(i)) + continue; /* Heuristic: CPU can go offline. */ + retval = rcutorture_booster_init(i); + if (retval < 0) { + firsterr = retval; + goto unwind; + } + } + } + register_reboot_notifier(&rcutorture_shutdown_nb); mutex_unlock(&fullstop_mutex); return 0; diff --git a/kernel/rcutree.c b/kernel/rcutree.c index d4437345706..dd4aea806f8 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -67,9 +67,6 @@ static struct lock_class_key rcu_node_class[NUM_RCU_LVLS]; .gpnum = -300, \ .completed = -300, \ .onofflock = __RAW_SPIN_LOCK_UNLOCKED(&structname.onofflock), \ - .orphan_cbs_list = NULL, \ - .orphan_cbs_tail = &structname.orphan_cbs_list, \ - .orphan_qlen = 0, \ .fqslock = __RAW_SPIN_LOCK_UNLOCKED(&structname.fqslock), \ .n_force_qs = 0, \ .n_force_qs_ngp = 0, \ @@ -143,6 +140,11 @@ module_param(blimit, int, 0); module_param(qhimark, int, 0); module_param(qlowmark, int, 0); +#ifdef CONFIG_RCU_CPU_STALL_DETECTOR +int rcu_cpu_stall_suppress __read_mostly = RCU_CPU_STALL_SUPPRESS_INIT; +module_param(rcu_cpu_stall_suppress, int, 0644); +#endif /* #ifdef CONFIG_RCU_CPU_STALL_DETECTOR */ + static void force_quiescent_state(struct rcu_state *rsp, int relaxed); static int rcu_pending(int cpu); @@ -362,8 +364,8 @@ void rcu_irq_exit(void) WARN_ON_ONCE(rdtp->dynticks & 0x1); /* If the interrupt queued a callback, get out of dyntick mode. */ - if (__get_cpu_var(rcu_sched_data).nxtlist || - __get_cpu_var(rcu_bh_data).nxtlist) + if (__this_cpu_read(rcu_sched_data.nxtlist) || + __this_cpu_read(rcu_bh_data.nxtlist)) set_need_resched(); } @@ -450,7 +452,7 @@ static int rcu_implicit_dynticks_qs(struct rcu_data *rdp) #ifdef CONFIG_RCU_CPU_STALL_DETECTOR -int rcu_cpu_stall_panicking __read_mostly; +int rcu_cpu_stall_suppress __read_mostly; static void record_gp_stall_check_time(struct rcu_state *rsp) { @@ -482,8 +484,11 @@ static void print_other_cpu_stall(struct rcu_state *rsp) rcu_print_task_stall(rnp); raw_spin_unlock_irqrestore(&rnp->lock, flags); - /* OK, time to rat on our buddy... */ - + /* + * OK, time to rat on our buddy... + * See Documentation/RCU/stallwarn.txt for info on how to debug + * RCU CPU stall warnings. + */ printk(KERN_ERR "INFO: %s detected stalls on CPUs/tasks: {", rsp->name); rcu_for_each_leaf_node(rsp, rnp) { @@ -512,6 +517,11 @@ static void print_cpu_stall(struct rcu_state *rsp) unsigned long flags; struct rcu_node *rnp = rcu_get_root(rsp); + /* + * OK, time to rat on ourselves... + * See Documentation/RCU/stallwarn.txt for info on how to debug + * RCU CPU stall warnings. + */ printk(KERN_ERR "INFO: %s detected stall on CPU %d (t=%lu jiffies)\n", rsp->name, smp_processor_id(), jiffies - rsp->gp_start); trigger_all_cpu_backtrace(); @@ -530,11 +540,11 @@ static void check_cpu_stall(struct rcu_state *rsp, struct rcu_data *rdp) long delta; struct rcu_node *rnp; - if (rcu_cpu_stall_panicking) + if (rcu_cpu_stall_suppress) return; - delta = jiffies - rsp->jiffies_stall; + delta = jiffies - ACCESS_ONCE(rsp->jiffies_stall); rnp = rdp->mynode; - if ((rnp->qsmask & rdp->grpmask) && delta >= 0) { + if ((ACCESS_ONCE(rnp->qsmask) & rdp->grpmask) && delta >= 0) { /* We haven't checked in, so go dump stack. */ print_cpu_stall(rsp); @@ -548,10 +558,26 @@ static void check_cpu_stall(struct rcu_state *rsp, struct rcu_data *rdp) static int rcu_panic(struct notifier_block *this, unsigned long ev, void *ptr) { - rcu_cpu_stall_panicking = 1; + rcu_cpu_stall_suppress = 1; return NOTIFY_DONE; } +/** + * rcu_cpu_stall_reset - prevent further stall warnings in current grace period + * + * Set the stall-warning timeout way off into the future, thus preventing + * any RCU CPU stall-warning messages from appearing in the current set of + * RCU grace periods. + * + * The caller must disable hard irqs. + */ +void rcu_cpu_stall_reset(void) +{ + rcu_sched_state.jiffies_stall = jiffies + ULONG_MAX / 2; + rcu_bh_state.jiffies_stall = jiffies + ULONG_MAX / 2; + rcu_preempt_stall_reset(); +} + static struct notifier_block rcu_panic_block = { .notifier_call = rcu_panic, }; @@ -571,6 +597,10 @@ static void check_cpu_stall(struct rcu_state *rsp, struct rcu_data *rdp) { } +void rcu_cpu_stall_reset(void) +{ +} + static void __init check_cpu_stall_init(void) { } @@ -587,9 +617,17 @@ static void __init check_cpu_stall_init(void) 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; + /* + * If the current grace period is waiting for this CPU, + * set up to detect a quiescent state, otherwise don't + * go looking for one. + */ rdp->gpnum = rnp->gpnum; + if (rnp->qsmask & rdp->grpmask) { + rdp->qs_pending = 1; + rdp->passed_quiesc = 0; + } else + rdp->qs_pending = 0; } } @@ -648,6 +686,24 @@ __rcu_process_gp_end(struct rcu_state *rsp, struct rcu_node *rnp, struct rcu_dat /* Remember that we saw this grace-period completion. */ rdp->completed = rnp->completed; + + /* + * If we were in an extended quiescent state, we may have + * missed some grace periods that others CPUs handled on + * our behalf. Catch up with this state to avoid noting + * spurious new grace periods. If another grace period + * has started, then rnp->gpnum will have advanced, so + * we will detect this later on. + */ + if (ULONG_CMP_LT(rdp->gpnum, rdp->completed)) + rdp->gpnum = rdp->completed; + + /* + * If RCU does not need a quiescent state from this CPU, + * then make sure that this CPU doesn't go looking for one. + */ + if ((rnp->qsmask & rdp->grpmask) == 0) + rdp->qs_pending = 0; } } @@ -712,7 +768,7 @@ static void rcu_start_gp(struct rcu_state *rsp, unsigned long flags) __releases(rcu_get_root(rsp)->lock) { - struct rcu_data *rdp = rsp->rda[smp_processor_id()]; + struct rcu_data *rdp = this_cpu_ptr(rsp->rda); struct rcu_node *rnp = rcu_get_root(rsp); if (!cpu_needs_another_gp(rsp, rdp) || rsp->fqs_active) { @@ -951,51 +1007,31 @@ rcu_check_quiescent_state(struct rcu_state *rsp, struct rcu_data *rdp) #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. + * Move a dying CPU's RCU callbacks to online CPU's callback list. + * Synchronization is not required because this function executes + * in stop_machine() context. */ -static void rcu_send_cbs_to_orphanage(struct rcu_state *rsp) +static void rcu_send_cbs_to_online(struct rcu_state *rsp) { int i; - struct rcu_data *rdp = rsp->rda[smp_processor_id()]; + /* current DYING CPU is cleared in the cpu_online_mask */ + int receive_cpu = cpumask_any(cpu_online_mask); + struct rcu_data *rdp = this_cpu_ptr(rsp->rda); + struct rcu_data *receive_rdp = per_cpu_ptr(rsp->rda, receive_cpu); if (rdp->nxtlist == NULL) return; /* irqs disabled, so comparison is stable. */ - raw_spin_lock(&rsp->onofflock); /* irqs already disabled. */ - *rsp->orphan_cbs_tail = rdp->nxtlist; - rsp->orphan_cbs_tail = rdp->nxttail[RCU_NEXT_TAIL]; + + *receive_rdp->nxttail[RCU_NEXT_TAIL] = rdp->nxtlist; + receive_rdp->nxttail[RCU_NEXT_TAIL] = rdp->nxttail[RCU_NEXT_TAIL]; + receive_rdp->qlen += rdp->qlen; + receive_rdp->n_cbs_adopted += rdp->qlen; + rdp->n_cbs_orphaned += rdp->qlen; + rdp->nxtlist = NULL; for (i = 0; i < RCU_NEXT_SIZE; i++) rdp->nxttail[i] = &rdp->nxtlist; - rsp->orphan_qlen += rdp->qlen; rdp->qlen = 0; - raw_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; - - raw_spin_lock_irqsave(&rsp->onofflock, flags); - rdp = rsp->rda[smp_processor_id()]; - if (rsp->orphan_cbs_list == NULL) { - raw_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; - raw_spin_unlock_irqrestore(&rsp->onofflock, flags); } /* @@ -1007,7 +1043,7 @@ static void __rcu_offline_cpu(int cpu, struct rcu_state *rsp) unsigned long flags; unsigned long mask; int need_report = 0; - struct rcu_data *rdp = rsp->rda[cpu]; + struct rcu_data *rdp = per_cpu_ptr(rsp->rda, cpu); struct rcu_node *rnp; /* Exclude any attempts to start a new grace period. */ @@ -1046,8 +1082,6 @@ static void __rcu_offline_cpu(int cpu, struct rcu_state *rsp) raw_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); } /* @@ -1065,11 +1099,7 @@ static void rcu_offline_cpu(int 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_send_cbs_to_online(struct rcu_state *rsp) { } @@ -1112,6 +1142,7 @@ static void rcu_do_batch(struct rcu_state *rsp, struct rcu_data *rdp) while (list) { next = list->next; prefetch(next); + debug_rcu_head_unqueue(list); list->func(list); list = next; if (++count >= rdp->blimit) @@ -1122,6 +1153,7 @@ static void rcu_do_batch(struct rcu_state *rsp, struct rcu_data *rdp) /* Update count, and requeue any remaining callbacks. */ rdp->qlen -= count; + rdp->n_cbs_invoked += count; if (list != NULL) { *tail = rdp->nxtlist; rdp->nxtlist = list; @@ -1225,7 +1257,8 @@ static void force_qs_rnp(struct rcu_state *rsp, int (*f)(struct rcu_data *)) cpu = rnp->grplo; bit = 1; for (; cpu <= rnp->grphi; cpu++, bit <<= 1) { - if ((rnp->qsmask & bit) != 0 && f(rsp->rda[cpu])) + if ((rnp->qsmask & bit) != 0 && + f(per_cpu_ptr(rsp->rda, cpu))) mask |= bit; } if (mask != 0) { @@ -1388,6 +1421,7 @@ __call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu), unsigned long flags; struct rcu_data *rdp; + debug_rcu_head_queue(head); head->func = func; head->next = NULL; @@ -1400,23 +1434,12 @@ __call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu), * a quiescent state betweentimes. */ local_irq_save(flags); - rdp = rsp->rda[smp_processor_id()]; - rcu_process_gp_end(rsp, rdp); - check_for_new_grace_period(rsp, rdp); + rdp = this_cpu_ptr(rsp->rda); /* Add the callback to our list. */ *rdp->nxttail[RCU_NEXT_TAIL] = head; rdp->nxttail[RCU_NEXT_TAIL] = &head->next; - /* Start a new grace period if one not already started. */ - if (!rcu_gp_in_progress(rsp)) { - unsigned long nestflag; - struct rcu_node *rnp_root = rcu_get_root(rsp); - - raw_spin_lock_irqsave(&rnp_root->lock, nestflag); - rcu_start_gp(rsp, nestflag); /* releases rnp_root->lock. */ - } - /* * Force the grace period if too many callbacks or too long waiting. * Enforce hysteresis, and don't invoke force_quiescent_state() @@ -1425,12 +1448,27 @@ __call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu), * 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; - 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; + + /* Are we ignoring a completed grace period? */ + rcu_process_gp_end(rsp, rdp); + check_for_new_grace_period(rsp, rdp); + + /* Start a new grace period if one not already started. */ + if (!rcu_gp_in_progress(rsp)) { + unsigned long nestflag; + struct rcu_node *rnp_root = rcu_get_root(rsp); + + raw_spin_lock_irqsave(&rnp_root->lock, nestflag); + rcu_start_gp(rsp, nestflag); /* rlses rnp_root->lock */ + } else { + /* Give the grace period a kick. */ + rdp->blimit = LONG_MAX; + 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 (ULONG_CMP_LT(ACCESS_ONCE(rsp->jiffies_force_qs), jiffies)) force_quiescent_state(rsp, 1); local_irq_restore(flags); @@ -1660,13 +1698,12 @@ static void _rcu_barrier(struct rcu_state *rsp, * 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. + * early. Note that on_each_cpu() disables irqs, which prevents + * any CPUs from coming online or going offline until each online + * CPU has queued its RCU-barrier callback. */ 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); @@ -1699,7 +1736,7 @@ rcu_boot_init_percpu_data(int cpu, struct rcu_state *rsp) { unsigned long flags; int i; - struct rcu_data *rdp = rsp->rda[cpu]; + struct rcu_data *rdp = per_cpu_ptr(rsp->rda, cpu); struct rcu_node *rnp = rcu_get_root(rsp); /* Set up local state, ensuring consistent view of global state. */ @@ -1727,7 +1764,7 @@ 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_data *rdp = per_cpu_ptr(rsp->rda, cpu); struct rcu_node *rnp = rcu_get_root(rsp); /* Set up local state, ensuring consistent view of global state. */ @@ -1792,18 +1829,13 @@ static int __cpuinit rcu_cpu_notify(struct notifier_block *self, 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. + * The whole machine is "stopped" except this CPU, so we can + * touch any data without introducing corruption. We send the + * dying CPU's callbacks to an arbitrarily chosen online CPU. */ - rcu_send_cbs_to_orphanage(&rcu_bh_state); - rcu_send_cbs_to_orphanage(&rcu_sched_state); - rcu_preempt_send_cbs_to_orphanage(); + rcu_send_cbs_to_online(&rcu_bh_state); + rcu_send_cbs_to_online(&rcu_sched_state); + rcu_preempt_send_cbs_to_online(); break; case CPU_DEAD: case CPU_DEAD_FROZEN: @@ -1841,8 +1873,9 @@ static void __init rcu_init_levelspread(struct rcu_state *rsp) { int i; - for (i = NUM_RCU_LVLS - 1; i >= 0; i--) + for (i = NUM_RCU_LVLS - 1; i > 0; i--) rsp->levelspread[i] = CONFIG_RCU_FANOUT; + rsp->levelspread[0] = RCU_FANOUT_LEAF; } #else /* #ifdef CONFIG_RCU_FANOUT_EXACT */ static void __init rcu_init_levelspread(struct rcu_state *rsp) @@ -1863,7 +1896,8 @@ static void __init rcu_init_levelspread(struct rcu_state *rsp) /* * Helper function for rcu_init() that initializes one rcu_state structure. */ -static void __init rcu_init_one(struct rcu_state *rsp) +static void __init rcu_init_one(struct rcu_state *rsp, + struct rcu_data __percpu *rda) { static char *buf[] = { "rcu_node_level_0", "rcu_node_level_1", @@ -1916,37 +1950,23 @@ static void __init rcu_init_one(struct rcu_state *rsp) } } + rsp->rda = rda; rnp = rsp->level[NUM_RCU_LVLS - 1]; for_each_possible_cpu(i) { while (i > rnp->grphi) rnp++; - rsp->rda[i]->mynode = rnp; + per_cpu_ptr(rsp->rda, i)->mynode = rnp; rcu_boot_init_percpu_data(i, rsp); } } -/* - * 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_INIT_FLAVOR(rsp, rcu_data) \ -do { \ - int i; \ - \ - for_each_possible_cpu(i) { \ - (rsp)->rda[i] = &per_cpu(rcu_data, i); \ - } \ - rcu_init_one(rsp); \ -} while (0) - void __init rcu_init(void) { int cpu; rcu_bootup_announce(); - RCU_INIT_FLAVOR(&rcu_sched_state, rcu_sched_data); - RCU_INIT_FLAVOR(&rcu_bh_state, rcu_bh_data); + rcu_init_one(&rcu_sched_state, &rcu_sched_data); + rcu_init_one(&rcu_bh_state, &rcu_bh_data); __rcu_init_preempt(); open_softirq(RCU_SOFTIRQ, rcu_process_callbacks); diff --git a/kernel/rcutree.h b/kernel/rcutree.h index 14c040b18ed..e8f057e44e3 100644 --- a/kernel/rcutree.h +++ b/kernel/rcutree.h @@ -31,46 +31,51 @@ /* * 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. + * In practice, this did work well going from three levels to four. + * Of course, your mileage may vary. */ #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 +#if CONFIG_RCU_FANOUT > 16 +#define RCU_FANOUT_LEAF 16 +#else /* #if CONFIG_RCU_FANOUT > 16 */ +#define RCU_FANOUT_LEAF (CONFIG_RCU_FANOUT) +#endif /* #else #if CONFIG_RCU_FANOUT > 16 */ +#define RCU_FANOUT_1 (RCU_FANOUT_LEAF) +#define RCU_FANOUT_2 (RCU_FANOUT_1 * CONFIG_RCU_FANOUT) +#define RCU_FANOUT_3 (RCU_FANOUT_2 * CONFIG_RCU_FANOUT) +#define RCU_FANOUT_4 (RCU_FANOUT_3 * CONFIG_RCU_FANOUT) + +#if NR_CPUS <= RCU_FANOUT_1 # 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 +#elif NR_CPUS <= RCU_FANOUT_2 # 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_1 DIV_ROUND_UP(NR_CPUS, RCU_FANOUT_1) # 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 +#elif NR_CPUS <= RCU_FANOUT_3 # 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_1 DIV_ROUND_UP(NR_CPUS, RCU_FANOUT_2) +# define NUM_RCU_LVL_2 DIV_ROUND_UP(NR_CPUS, RCU_FANOUT_1) +# define NUM_RCU_LVL_3 (NR_CPUS) # define NUM_RCU_LVL_4 0 -#elif NR_CPUS <= RCU_FANOUT_FOURTH +#elif NR_CPUS <= RCU_FANOUT_4 # 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 +# define NUM_RCU_LVL_1 DIV_ROUND_UP(NR_CPUS, RCU_FANOUT_3) +# define NUM_RCU_LVL_2 DIV_ROUND_UP(NR_CPUS, RCU_FANOUT_2) +# define NUM_RCU_LVL_3 DIV_ROUND_UP(NR_CPUS, RCU_FANOUT_1) +# define NUM_RCU_LVL_4 (NR_CPUS) #else # error "CONFIG_RCU_FANOUT insufficient for NR_CPUS" -#endif /* #if (NR_CPUS) <= RCU_FANOUT */ +#endif /* #if (NR_CPUS) <= RCU_FANOUT_1 */ #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) @@ -202,6 +207,9 @@ struct rcu_data { long qlen; /* # of queued callbacks */ long qlen_last_fqs_check; /* qlen at last check for QS forcing */ + unsigned long n_cbs_invoked; /* count of RCU cbs invoked. */ + unsigned long n_cbs_orphaned; /* RCU cbs orphaned by dying CPU */ + unsigned long n_cbs_adopted; /* RCU cbs adopted from dying CPU */ unsigned long n_force_qs_snap; /* did other CPU force QS recently? */ long blimit; /* Upper limit on a processed batch */ @@ -254,19 +262,23 @@ struct rcu_data { #define RCU_STALL_DELAY_DELTA 0 #endif -#define RCU_SECONDS_TILL_STALL_CHECK (10 * HZ + RCU_STALL_DELAY_DELTA) +#define RCU_SECONDS_TILL_STALL_CHECK (CONFIG_RCU_CPU_STALL_TIMEOUT * HZ + \ + RCU_STALL_DELAY_DELTA) /* for rsp->jiffies_stall */ -#define RCU_SECONDS_TILL_STALL_RECHECK (30 * HZ + RCU_STALL_DELAY_DELTA) +#define RCU_SECONDS_TILL_STALL_RECHECK (3 * RCU_SECONDS_TILL_STALL_CHECK + 30) /* 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 */ +#ifdef CONFIG_RCU_CPU_STALL_DETECTOR_RUNNABLE +#define RCU_CPU_STALL_SUPPRESS_INIT 0 +#else +#define RCU_CPU_STALL_SUPPRESS_INIT 1 +#endif -#define ULONG_CMP_GE(a, b) (ULONG_MAX / 2 >= (a) - (b)) -#define ULONG_CMP_LT(a, b) (ULONG_MAX / 2 < (a) - (b)) +#endif /* #ifdef CONFIG_RCU_CPU_STALL_DETECTOR */ /* * RCU global state, including node hierarchy. This hierarchy is @@ -283,7 +295,7 @@ struct rcu_state { 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. */ + struct rcu_data __percpu *rda; /* pointer of percu rcu_data. */ /* The following fields are guarded by the root rcu_node's lock. */ @@ -302,15 +314,7 @@ struct rcu_state { /* End of fields guarded by root rcu_node's lock. */ raw_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. */ + /* starting new GP. */ raw_spinlock_t fqslock; /* Only one task forcing */ /* quiescent states. */ unsigned long jiffies_force_qs; /* Time at which to invoke */ @@ -365,6 +369,7 @@ static void rcu_report_unblock_qs_rnp(struct rcu_node *rnp, #ifdef CONFIG_RCU_CPU_STALL_DETECTOR static void rcu_print_detail_task_stall(struct rcu_state *rsp); static void rcu_print_task_stall(struct rcu_node *rnp); +static void rcu_preempt_stall_reset(void); #endif /* #ifdef CONFIG_RCU_CPU_STALL_DETECTOR */ static void rcu_preempt_check_blocked_tasks(struct rcu_node *rnp); #ifdef CONFIG_HOTPLUG_CPU @@ -382,7 +387,7 @@ static void rcu_report_exp_rnp(struct rcu_state *rsp, struct rcu_node *rnp); 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 rcu_preempt_send_cbs_to_online(void); static void __init __rcu_init_preempt(void); static void rcu_needs_cpu_flush(void); diff --git a/kernel/rcutree_plugin.h b/kernel/rcutree_plugin.h index 0e4f420245d..a3638710dc6 100644 --- a/kernel/rcutree_plugin.h +++ b/kernel/rcutree_plugin.h @@ -25,6 +25,7 @@ */ #include <linux/delay.h> +#include <linux/stop_machine.h> /* * Check the RCU kernel configuration parameters and print informative @@ -57,7 +58,7 @@ static void __init rcu_bootup_announce_oddness(void) printk(KERN_INFO "\tRCU-based detection of stalled CPUs is disabled.\n"); #endif -#ifndef CONFIG_RCU_CPU_STALL_VERBOSE +#if defined(CONFIG_TREE_PREEMPT_RCU) && !defined(CONFIG_RCU_CPU_STALL_VERBOSE) printk(KERN_INFO "\tVerbose stalled-CPUs detection is disabled.\n"); #endif #if NUM_RCU_LVL_4 != 0 @@ -154,7 +155,7 @@ static void rcu_preempt_note_context_switch(int cpu) (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]; + rdp = per_cpu_ptr(rcu_preempt_state.rda, cpu); rnp = rdp->mynode; raw_spin_lock_irqsave(&rnp->lock, flags); t->rcu_read_unlock_special |= RCU_READ_UNLOCK_BLOCKED; @@ -201,7 +202,7 @@ static void rcu_preempt_note_context_switch(int cpu) */ void __rcu_read_lock(void) { - ACCESS_ONCE(current->rcu_read_lock_nesting)++; + current->rcu_read_lock_nesting++; barrier(); /* needed if we ever invoke rcu_read_lock in rcutree.c */ } EXPORT_SYMBOL_GPL(__rcu_read_lock); @@ -344,7 +345,9 @@ 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 && + --t->rcu_read_lock_nesting; + barrier(); /* decrement before load of ->rcu_read_unlock_special */ + if (t->rcu_read_lock_nesting == 0 && unlikely(ACCESS_ONCE(t->rcu_read_unlock_special))) rcu_read_unlock_special(t); #ifdef CONFIG_PROVE_LOCKING @@ -417,6 +420,16 @@ static void rcu_print_task_stall(struct rcu_node *rnp) } } +/* + * Suppress preemptible RCU's CPU stall warnings by pushing the + * time of the next stall-warning message comfortably far into the + * future. + */ +static void rcu_preempt_stall_reset(void) +{ + rcu_preempt_state.jiffies_stall = jiffies + ULONG_MAX / 2; +} + #endif /* #ifdef CONFIG_RCU_CPU_STALL_DETECTOR */ /* @@ -546,9 +559,11 @@ EXPORT_SYMBOL_GPL(call_rcu); * * 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. + * read-side critical sections have completed. Note, however, that + * upon return from synchronize_rcu(), the caller might well be executing + * concurrently with new RCU read-side critical sections that began while + * synchronize_rcu() was waiting. RCU read-side critical sections are + * delimited by rcu_read_lock() and rcu_read_unlock(), and may be nested. */ void synchronize_rcu(void) { @@ -759,11 +774,11 @@ static void __cpuinit rcu_preempt_init_percpu_data(int cpu) } /* - * Move preemptable RCU's callbacks to ->orphan_cbs_list. + * Move preemptable RCU's callbacks from dying CPU to other online CPU. */ -static void rcu_preempt_send_cbs_to_orphanage(void) +static void rcu_preempt_send_cbs_to_online(void) { - rcu_send_cbs_to_orphanage(&rcu_preempt_state); + rcu_send_cbs_to_online(&rcu_preempt_state); } /* @@ -771,7 +786,7 @@ static void rcu_preempt_send_cbs_to_orphanage(void) */ static void __init __rcu_init_preempt(void) { - RCU_INIT_FLAVOR(&rcu_preempt_state, rcu_preempt_data); + rcu_init_one(&rcu_preempt_state, &rcu_preempt_data); } /* @@ -865,6 +880,14 @@ static void rcu_print_task_stall(struct rcu_node *rnp) { } +/* + * Because preemptible RCU does not exist, there is no need to suppress + * its CPU stall warnings. + */ +static void rcu_preempt_stall_reset(void) +{ +} + #endif /* #ifdef CONFIG_RCU_CPU_STALL_DETECTOR */ /* @@ -919,15 +942,6 @@ 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. */ @@ -988,7 +1002,7 @@ 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) +static void rcu_preempt_send_cbs_to_online(void) { } @@ -1001,6 +1015,132 @@ static void __init __rcu_init_preempt(void) #endif /* #else #ifdef CONFIG_TREE_PREEMPT_RCU */ +#ifndef CONFIG_SMP + +void synchronize_sched_expedited(void) +{ + cond_resched(); +} +EXPORT_SYMBOL_GPL(synchronize_sched_expedited); + +#else /* #ifndef CONFIG_SMP */ + +static atomic_t sync_sched_expedited_started = ATOMIC_INIT(0); +static atomic_t sync_sched_expedited_done = ATOMIC_INIT(0); + +static int synchronize_sched_expedited_cpu_stop(void *data) +{ + /* + * There must be a full memory barrier on each affected CPU + * between the time that try_stop_cpus() is called and the + * time that it returns. + * + * In the current initial implementation of cpu_stop, the + * above condition is already met when the control reaches + * this point and the following smp_mb() is not strictly + * necessary. Do smp_mb() anyway for documentation and + * robustness against future implementation changes. + */ + smp_mb(); /* See above comment block. */ + return 0; +} + +/* + * 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. + * + * This implementation can be thought of as an application of ticket + * locking to RCU, with sync_sched_expedited_started and + * sync_sched_expedited_done taking on the roles of the halves + * of the ticket-lock word. Each task atomically increments + * sync_sched_expedited_started upon entry, snapshotting the old value, + * then attempts to stop all the CPUs. If this succeeds, then each + * CPU will have executed a context switch, resulting in an RCU-sched + * grace period. We are then done, so we use atomic_cmpxchg() to + * update sync_sched_expedited_done to match our snapshot -- but + * only if someone else has not already advanced past our snapshot. + * + * On the other hand, if try_stop_cpus() fails, we check the value + * of sync_sched_expedited_done. If it has advanced past our + * initial snapshot, then someone else must have forced a grace period + * some time after we took our snapshot. In this case, our work is + * done for us, and we can simply return. Otherwise, we try again, + * but keep our initial snapshot for purposes of checking for someone + * doing our work for us. + * + * If we fail too many times in a row, we fall back to synchronize_sched(). + */ +void synchronize_sched_expedited(void) +{ + int firstsnap, s, snap, trycount = 0; + + /* Note that atomic_inc_return() implies full memory barrier. */ + firstsnap = snap = atomic_inc_return(&sync_sched_expedited_started); + get_online_cpus(); + + /* + * Each pass through the following loop attempts to force a + * context switch on each CPU. + */ + while (try_stop_cpus(cpu_online_mask, + synchronize_sched_expedited_cpu_stop, + NULL) == -EAGAIN) { + put_online_cpus(); + + /* No joy, try again later. Or just synchronize_sched(). */ + if (trycount++ < 10) + udelay(trycount * num_online_cpus()); + else { + synchronize_sched(); + return; + } + + /* Check to see if someone else did our work for us. */ + s = atomic_read(&sync_sched_expedited_done); + if (UINT_CMP_GE((unsigned)s, (unsigned)firstsnap)) { + smp_mb(); /* ensure test happens before caller kfree */ + return; + } + + /* + * Refetching sync_sched_expedited_started allows later + * callers to piggyback on our grace period. We subtract + * 1 to get the same token that the last incrementer got. + * We retry after they started, so our grace period works + * for them, and they started after our first try, so their + * grace period works for us. + */ + get_online_cpus(); + snap = atomic_read(&sync_sched_expedited_started) - 1; + smp_mb(); /* ensure read is before try_stop_cpus(). */ + } + + /* + * Everyone up to our most recent fetch is covered by our grace + * period. Update the counter, but only if our work is still + * relevant -- which it won't be if someone who started later + * than we did beat us to the punch. + */ + do { + s = atomic_read(&sync_sched_expedited_done); + if (UINT_CMP_GE((unsigned)s, (unsigned)snap)) { + smp_mb(); /* ensure test happens before caller kfree */ + break; + } + } while (atomic_cmpxchg(&sync_sched_expedited_done, s, snap) != s); + + put_online_cpus(); +} +EXPORT_SYMBOL_GPL(synchronize_sched_expedited); + +#endif /* #else #ifndef CONFIG_SMP */ + #if !defined(CONFIG_RCU_FAST_NO_HZ) /* diff --git a/kernel/rcutree_trace.c b/kernel/rcutree_trace.c index 36c95b45738..c8e97853b97 100644 --- a/kernel/rcutree_trace.c +++ b/kernel/rcutree_trace.c @@ -64,7 +64,9 @@ static void print_one_rcu_data(struct seq_file *m, struct rcu_data *rdp) rdp->dynticks_fqs); #endif /* #ifdef CONFIG_NO_HZ */ seq_printf(m, " of=%lu ri=%lu", rdp->offline_fqs, rdp->resched_ipi); - seq_printf(m, " ql=%ld b=%ld\n", rdp->qlen, rdp->blimit); + seq_printf(m, " ql=%ld b=%ld", rdp->qlen, rdp->blimit); + seq_printf(m, " ci=%lu co=%lu ca=%lu\n", + rdp->n_cbs_invoked, rdp->n_cbs_orphaned, rdp->n_cbs_adopted); } #define PRINT_RCU_DATA(name, func, m) \ @@ -119,7 +121,9 @@ static void print_one_rcu_data_csv(struct seq_file *m, struct rcu_data *rdp) rdp->dynticks_fqs); #endif /* #ifdef CONFIG_NO_HZ */ seq_printf(m, ",%lu,%lu", rdp->offline_fqs, rdp->resched_ipi); - seq_printf(m, ",%ld,%ld\n", rdp->qlen, rdp->blimit); + seq_printf(m, ",%ld,%ld", rdp->qlen, rdp->blimit); + seq_printf(m, ",%lu,%lu,%lu\n", + rdp->n_cbs_invoked, rdp->n_cbs_orphaned, rdp->n_cbs_adopted); } static int show_rcudata_csv(struct seq_file *m, void *unused) @@ -128,7 +132,7 @@ static int show_rcudata_csv(struct seq_file *m, void *unused) #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, "\"of\",\"ri\",\"ql\",\"b\",\"ci\",\"co\",\"ca\"\n"); #ifdef CONFIG_TREE_PREEMPT_RCU seq_puts(m, "\"rcu_preempt:\"\n"); PRINT_RCU_DATA(rcu_preempt_data, print_one_rcu_data_csv, m); @@ -162,13 +166,13 @@ static void print_one_rcu_state(struct seq_file *m, struct rcu_state *rsp) gpnum = rsp->gpnum; seq_printf(m, "c=%lu g=%lu s=%d jfq=%ld j=%x " - "nfqs=%lu/nfqsng=%lu(%lu) fqlh=%lu oqlen=%ld\n", + "nfqs=%lu/nfqsng=%lu(%lu) fqlh=%lu\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->orphan_qlen); + rsp->n_force_qs_lh); for (rnp = &rsp->node[0]; rnp - &rsp->node[0] < NUM_RCU_NODES; rnp++) { if (rnp->level != level) { seq_puts(m, "\n"); @@ -262,7 +266,7 @@ static void print_rcu_pendings(struct seq_file *m, struct rcu_state *rsp) struct rcu_data *rdp; for_each_possible_cpu(cpu) { - rdp = rsp->rda[cpu]; + rdp = per_cpu_ptr(rsp->rda, cpu); if (rdp->beenonline) print_one_rcu_pending(m, rdp); } @@ -296,7 +300,7 @@ static const struct file_operations rcu_pending_fops = { static struct dentry *rcudir; -static int __init rcuclassic_trace_init(void) +static int __init rcutree_trace_init(void) { struct dentry *retval; @@ -333,14 +337,14 @@ free_out: return 1; } -static void __exit rcuclassic_trace_cleanup(void) +static void __exit rcutree_trace_cleanup(void) { debugfs_remove_recursive(rcudir); } -module_init(rcuclassic_trace_init); -module_exit(rcuclassic_trace_cleanup); +module_init(rcutree_trace_init); +module_exit(rcutree_trace_cleanup); MODULE_AUTHOR("Paul E. McKenney"); MODULE_DESCRIPTION("Read-Copy Update tracing for hierarchical implementation"); diff --git a/kernel/relay.c b/kernel/relay.c index 4268287148c..859ea5a9605 100644 --- a/kernel/relay.c +++ b/kernel/relay.c @@ -70,17 +70,10 @@ static const struct vm_operations_struct relay_file_mmap_ops = { */ static struct page **relay_alloc_page_array(unsigned int n_pages) { - struct page **array; - size_t pa_size = n_pages * sizeof(struct page *); - - if (pa_size > PAGE_SIZE) { - array = vmalloc(pa_size); - if (array) - memset(array, 0, pa_size); - } else { - array = kzalloc(pa_size, GFP_KERNEL); - } - return array; + const size_t pa_size = n_pages * sizeof(struct page *); + if (pa_size > PAGE_SIZE) + return vzalloc(pa_size); + return kzalloc(pa_size, GFP_KERNEL); } /* @@ -539,7 +532,7 @@ static int __cpuinit relay_hotcpu_callback(struct notifier_block *nb, "relay_hotcpu_callback: cpu %d buffer " "creation failed\n", hotcpu); mutex_unlock(&relay_channels_mutex); - return NOTIFY_BAD; + return notifier_from_errno(-ENOMEM); } } mutex_unlock(&relay_channels_mutex); diff --git a/kernel/resource.c b/kernel/resource.c index 7b36976e5de..798e2fae2a0 100644 --- a/kernel/resource.c +++ b/kernel/resource.c @@ -357,6 +357,32 @@ int __weak page_is_ram(unsigned long pfn) return walk_system_ram_range(pfn, 1, NULL, __is_ram) == 1; } +void __weak arch_remove_reservations(struct resource *avail) +{ +} + +static resource_size_t simple_align_resource(void *data, + const struct resource *avail, + resource_size_t size, + resource_size_t align) +{ + return avail->start; +} + +static void resource_clip(struct resource *res, resource_size_t min, + resource_size_t max) +{ + if (res->start < min) + res->start = min; + if (res->end > max) + res->end = max; +} + +static bool resource_contains(struct resource *res1, struct resource *res2) +{ + return res1->start <= res2->start && res1->end >= res2->end; +} + /* * Find empty slot in the resource tree given range and alignment. */ @@ -370,8 +396,9 @@ static int find_resource(struct resource *root, struct resource *new, void *alignf_data) { struct resource *this = root->child; - struct resource tmp = *new; + struct resource tmp = *new, avail, alloc; + tmp.flags = new->flags; tmp.start = root->start; /* * Skip past an allocated resource that starts at 0, since the assignment @@ -386,17 +413,22 @@ static int find_resource(struct resource *root, struct resource *new, tmp.end = this->start - 1; else tmp.end = root->end; - if (tmp.start < min) - tmp.start = min; - if (tmp.end > max) - tmp.end = max; - tmp.start = ALIGN(tmp.start, align); - if (alignf) - tmp.start = alignf(alignf_data, &tmp, size, align); - if (tmp.start < tmp.end && tmp.end - tmp.start >= size - 1) { - new->start = tmp.start; - new->end = tmp.start + size - 1; - return 0; + + resource_clip(&tmp, min, max); + arch_remove_reservations(&tmp); + + /* Check for overflow after ALIGN() */ + avail = *new; + avail.start = ALIGN(tmp.start, align); + avail.end = tmp.end; + if (avail.start >= tmp.start) { + alloc.start = alignf(alignf_data, &avail, size, align); + alloc.end = alloc.start + size - 1; + if (resource_contains(&avail, &alloc)) { + new->start = alloc.start; + new->end = alloc.end; + return 0; + } } if (!this) break; @@ -428,6 +460,9 @@ int allocate_resource(struct resource *root, struct resource *new, { int err; + if (!alignf) + alignf = simple_align_resource; + write_lock(&resource_lock); err = find_resource(root, new, size, min, max, align, alignf, alignf_data); if (err >= 0 && __request_resource(root, new)) @@ -453,6 +488,8 @@ static struct resource * __insert_resource(struct resource *parent, struct resou if (first == parent) return first; + if (WARN_ON(first == new)) /* duplicated insertion */ + return first; if ((first->start > new->start) || (first->end < new->end)) break; diff --git a/kernel/rtmutex-tester.c b/kernel/rtmutex-tester.c index a56f629b057..66cb89bc5ef 100644 --- a/kernel/rtmutex-tester.c +++ b/kernel/rtmutex-tester.c @@ -76,7 +76,9 @@ static int handle_op(struct test_thread_data *td, int lockwakeup) } if (!lockwakeup && td->bkl == 4) { +#ifdef CONFIG_LOCK_KERNEL unlock_kernel(); +#endif td->bkl = 0; } return 0; @@ -133,14 +135,18 @@ static int handle_op(struct test_thread_data *td, int lockwakeup) if (td->bkl) return 0; td->bkl = 1; +#ifdef CONFIG_LOCK_KERNEL lock_kernel(); +#endif td->bkl = 4; return 0; case RTTEST_UNLOCKBKL: if (td->bkl != 4) break; +#ifdef CONFIG_LOCK_KERNEL unlock_kernel(); +#endif td->bkl = 0; return 0; diff --git a/kernel/sched.c b/kernel/sched.c index 054a6012de9..18d38e4ec7b 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -75,8 +75,11 @@ #include <asm/tlb.h> #include <asm/irq_regs.h> +#include <asm/mutex.h> #include "sched_cpupri.h" +#include "workqueue_sched.h" +#include "sched_autogroup.h" #define CREATE_TRACE_POINTS #include <trace/events/sched.h> @@ -252,6 +255,8 @@ struct task_group { /* runqueue "owned" by this group on each cpu */ struct cfs_rq **cfs_rq; unsigned long shares; + + atomic_t load_weight; #endif #ifdef CONFIG_RT_GROUP_SCHED @@ -267,25 +272,18 @@ struct task_group { struct task_group *parent; struct list_head siblings; struct list_head children; -}; -#define root_task_group init_task_group +#ifdef CONFIG_SCHED_AUTOGROUP + struct autogroup *autogroup; +#endif +}; -/* task_group_lock serializes add/remove of task groups and also changes to - * a task group's cpu shares. - */ +/* task_group_lock serializes the addition/removal of task groups */ static DEFINE_SPINLOCK(task_group_lock); #ifdef CONFIG_FAIR_GROUP_SCHED -#ifdef CONFIG_SMP -static int root_task_group_empty(void) -{ - return list_empty(&root_task_group.children); -} -#endif - -# define INIT_TASK_GROUP_LOAD NICE_0_LOAD +# define ROOT_TASK_GROUP_LOAD NICE_0_LOAD /* * A weight of 0 or 1 can cause arithmetics problems. @@ -298,59 +296,13 @@ static int root_task_group_empty(void) #define MIN_SHARES 2 #define MAX_SHARES (1UL << 18) -static int init_task_group_load = INIT_TASK_GROUP_LOAD; +static int root_task_group_load = ROOT_TASK_GROUP_LOAD; #endif /* Default task group. * Every task in system belong to this group at bootup. */ -struct task_group init_task_group; - -/* return group to which a task belongs */ -static inline struct task_group *task_group(struct task_struct *p) -{ - struct task_group *tg; - -#ifdef CONFIG_CGROUP_SCHED - tg = container_of(task_subsys_state(p, cpu_cgroup_subsys_id), - struct task_group, css); -#else - tg = &init_task_group; -#endif - return tg; -} - -/* Change a task's cfs_rq and parent entity if it moves across CPUs/groups */ -static inline void set_task_rq(struct task_struct *p, unsigned int cpu) -{ - /* - * Strictly speaking this rcu_read_lock() is not needed since the - * task_group is tied to the cgroup, which in turn can never go away - * as long as there are tasks attached to it. - * - * However since task_group() uses task_subsys_state() which is an - * rcu_dereference() user, this quiets CONFIG_PROVE_RCU. - */ - rcu_read_lock(); -#ifdef CONFIG_FAIR_GROUP_SCHED - p->se.cfs_rq = task_group(p)->cfs_rq[cpu]; - p->se.parent = task_group(p)->se[cpu]; -#endif - -#ifdef CONFIG_RT_GROUP_SCHED - p->rt.rt_rq = task_group(p)->rt_rq[cpu]; - p->rt.parent = task_group(p)->rt_se[cpu]; -#endif - rcu_read_unlock(); -} - -#else - -static inline void set_task_rq(struct task_struct *p, unsigned int cpu) { } -static inline struct task_group *task_group(struct task_struct *p) -{ - return NULL; -} +struct task_group root_task_group; #endif /* CONFIG_CGROUP_SCHED */ @@ -387,6 +339,7 @@ struct cfs_rq { * leaf_cfs_rq_list ties together list of leaf cfs_rq's in a cpu. This * list is used during load balance. */ + int on_list; struct list_head leaf_cfs_rq_list; struct task_group *tg; /* group that "owns" this runqueue */ @@ -405,14 +358,17 @@ struct cfs_rq { unsigned long h_load; /* - * this cpu's part of tg->shares + * Maintaining per-cpu shares distribution for group scheduling + * + * load_stamp is the last time we updated the load average + * load_last is the last time we updated the load average and saw load + * load_unacc_exec_time is currently unaccounted execution time */ - unsigned long shares; + u64 load_avg; + u64 load_period; + u64 load_stamp, load_last, load_unacc_exec_time; - /* - * load.weight at the time we set shares - */ - unsigned long rq_weight; + unsigned long load_contribution; #endif #endif }; @@ -471,9 +427,7 @@ struct root_domain { */ cpumask_var_t rto_mask; atomic_t rto_count; -#ifdef CONFIG_SMP struct cpupri cpupri; -#endif }; /* @@ -482,7 +436,7 @@ struct root_domain { */ static struct root_domain def_root_domain; -#endif +#endif /* CONFIG_SMP */ /* * This is the main, per-CPU runqueue data structure. @@ -502,9 +456,10 @@ struct rq { unsigned long nr_running; #define CPU_LOAD_IDX_MAX 5 unsigned long cpu_load[CPU_LOAD_IDX_MAX]; + unsigned long last_load_update_tick; #ifdef CONFIG_NO_HZ u64 nohz_stamp; - unsigned char in_nohz_recently; + unsigned char nohz_balance_kick; #endif unsigned int skip_clock_update; @@ -532,11 +487,12 @@ struct rq { */ unsigned long nr_uninterruptible; - struct task_struct *curr, *idle; + struct task_struct *curr, *idle, *stop; unsigned long next_balance; struct mm_struct *prev_mm; u64 clock; + u64 clock_task; atomic_t nr_iowait; @@ -544,6 +500,8 @@ struct rq { struct root_domain *rd; struct sched_domain *sd; + unsigned long cpu_power; + unsigned char idle_at_tick; /* For active balancing */ int post_schedule; @@ -562,6 +520,10 @@ struct rq { u64 avg_idle; #endif +#ifdef CONFIG_IRQ_TIME_ACCOUNTING + u64 prev_irq_time; +#endif + /* calc_load related fields */ unsigned long calc_load_update; long calc_load_active; @@ -591,26 +553,13 @@ struct rq { /* try_to_wake_up() stats */ unsigned int ttwu_count; unsigned int ttwu_local; - - /* BKL stats */ - unsigned int bkl_count; #endif }; static DEFINE_PER_CPU_SHARED_ALIGNED(struct rq, runqueues); -static inline -void check_preempt_curr(struct rq *rq, struct task_struct *p, int flags) -{ - rq->curr->sched_class->check_preempt_curr(rq, p, flags); - /* - * A queue event has occurred, and we're going to schedule. In - * this case, we can save a useless back to back clock update. - */ - if (test_tsk_need_resched(p)) - rq->skip_clock_update = 1; -} +static void check_preempt_curr(struct rq *rq, struct task_struct *p, int flags); static inline int cpu_of(struct rq *rq) { @@ -642,10 +591,67 @@ static inline int cpu_of(struct rq *rq) #define cpu_curr(cpu) (cpu_rq(cpu)->curr) #define raw_rq() (&__raw_get_cpu_var(runqueues)) -inline void update_rq_clock(struct rq *rq) +#ifdef CONFIG_CGROUP_SCHED + +/* + * Return the group to which this tasks belongs. + * + * We use task_subsys_state_check() and extend the RCU verification + * with lockdep_is_held(&task_rq(p)->lock) because cpu_cgroup_attach() + * holds that lock for each task it moves into the cgroup. Therefore + * by holding that lock, we pin the task to the current cgroup. + */ +static inline struct task_group *task_group(struct task_struct *p) +{ + struct task_group *tg; + struct cgroup_subsys_state *css; + + if (p->flags & PF_EXITING) + return &root_task_group; + + css = task_subsys_state_check(p, cpu_cgroup_subsys_id, + lockdep_is_held(&task_rq(p)->lock)); + tg = container_of(css, struct task_group, css); + + return autogroup_task_group(p, tg); +} + +/* Change a task's cfs_rq and parent entity if it moves across CPUs/groups */ +static inline void set_task_rq(struct task_struct *p, unsigned int cpu) +{ +#ifdef CONFIG_FAIR_GROUP_SCHED + p->se.cfs_rq = task_group(p)->cfs_rq[cpu]; + p->se.parent = task_group(p)->se[cpu]; +#endif + +#ifdef CONFIG_RT_GROUP_SCHED + p->rt.rt_rq = task_group(p)->rt_rq[cpu]; + p->rt.parent = task_group(p)->rt_se[cpu]; +#endif +} + +#else /* CONFIG_CGROUP_SCHED */ + +static inline void set_task_rq(struct task_struct *p, unsigned int cpu) { } +static inline struct task_group *task_group(struct task_struct *p) +{ + return NULL; +} + +#endif /* CONFIG_CGROUP_SCHED */ + +static void update_rq_clock_task(struct rq *rq, s64 delta); + +static void update_rq_clock(struct rq *rq) { - if (!rq->skip_clock_update) - rq->clock = sched_clock_cpu(cpu_of(rq)); + s64 delta; + + if (rq->skip_clock_update) + return; + + delta = sched_clock_cpu(cpu_of(rq)) - rq->clock; + rq->clock += delta; + update_rq_clock_task(rq, delta); } /* @@ -722,7 +728,7 @@ sched_feat_write(struct file *filp, const char __user *ubuf, size_t cnt, loff_t *ppos) { char buf[64]; - char *cmp = buf; + char *cmp; int neg = 0; int i; @@ -733,16 +739,15 @@ sched_feat_write(struct file *filp, const char __user *ubuf, return -EFAULT; buf[cnt] = 0; + cmp = strstrip(buf); - if (strncmp(buf, "NO_", 3) == 0) { + if (strncmp(cmp, "NO_", 3) == 0) { neg = 1; cmp += 3; } for (i = 0; sched_feat_names[i]; i++) { - int len = strlen(sched_feat_names[i]); - - if (strncmp(cmp, sched_feat_names[i], len) == 0) { + if (strcmp(cmp, sched_feat_names[i]) == 0) { if (neg) sysctl_sched_features &= ~(1UL << i); else @@ -792,20 +797,6 @@ late_initcall(sched_init_debug); const_debug unsigned int sysctl_sched_nr_migrate = 32; /* - * ratelimit for updating the group shares. - * default: 0.25ms - */ -unsigned int sysctl_sched_shares_ratelimit = 250000; -unsigned int normalized_sysctl_sched_shares_ratelimit = 250000; - -/* - * Inject some fuzzyness into changing the per-cpu group shares - * this avoids remote rq-locks at the expense of fairness. - * default: 4 - */ -unsigned int sysctl_sched_shares_thresh = 4; - -/* * period over which we average the RT time consumption, measured * in ms. * @@ -969,14 +960,6 @@ static struct rq *task_rq_lock(struct task_struct *p, unsigned long *flags) } } -void task_rq_unlock_wait(struct task_struct *p) -{ - struct rq *rq = task_rq(p); - - smp_mb(); /* spin-unlock-wait is not a full memory barrier */ - raw_spin_unlock_wait(&rq->lock); -} - static void __task_rq_unlock(struct rq *rq) __releases(rq->lock) { @@ -1202,6 +1185,27 @@ static void resched_cpu(int cpu) #ifdef CONFIG_NO_HZ /* + * In the semi idle case, use the nearest busy cpu for migrating timers + * from an idle cpu. This is good for power-savings. + * + * We don't do similar optimization for completely idle system, as + * selecting an idle cpu will add more delays to the timers than intended + * (as that cpu's timer base may not be uptodate wrt jiffies etc). + */ +int get_nohz_timer_target(void) +{ + int cpu = smp_processor_id(); + int i; + struct sched_domain *sd; + + for_each_domain(cpu, sd) { + for_each_cpu(i, sched_domain_span(sd)) + if (!idle_cpu(i)) + return i; + } + return cpu; +} +/* * When add_timer_on() enqueues a timer into the timer wheel of an * idle CPU then this timer might expire before the next timer event * which is scheduled to wake up that CPU. In case of a completely @@ -1241,16 +1245,6 @@ void wake_up_idle_cpu(int cpu) smp_send_reschedule(cpu); } -int nohz_ratelimit(int cpu) -{ - struct rq *rq = cpu_rq(cpu); - u64 diff = rq->clock - rq->nohz_stamp; - - rq->nohz_stamp = rq->clock; - - return diff < (NSEC_PER_SEC / HZ) >> 1; -} - #endif /* CONFIG_NO_HZ */ static u64 sched_avg_period(void) @@ -1263,6 +1257,12 @@ static void sched_avg_update(struct rq *rq) s64 period = sched_avg_period(); while ((s64)(rq->clock - rq->age_stamp) > period) { + /* + * Inline assembly required to prevent the compiler + * optimising this loop into a divmod call. + * See __iter_div_u64_rem() for another example of this. + */ + asm("" : "+rm" (rq->age_stamp)); rq->age_stamp += period; rq->rt_avg /= 2; } @@ -1284,6 +1284,10 @@ static void resched_task(struct task_struct *p) static void sched_rt_avg_update(struct rq *rq, u64 rt_delta) { } + +static void sched_avg_update(struct rq *rq) +{ +} #endif /* CONFIG_SMP */ #if BITS_PER_LONG == 32 @@ -1341,6 +1345,12 @@ static inline void update_load_sub(struct load_weight *lw, unsigned long dec) lw->inv_weight = 0; } +static inline void update_load_set(struct load_weight *lw, unsigned long w) +{ + lw->weight = w; + lw->inv_weight = 0; +} + /* * To aid in avoiding the subversion of "niceness" due to uneven distribution * of tasks with abnormal "nice" values across CPUs the contribution that @@ -1507,24 +1517,9 @@ static unsigned long target_load(int cpu, int type) return max(rq->cpu_load[type-1], total); } -static struct sched_group *group_of(int cpu) -{ - struct sched_domain *sd = rcu_dereference_sched(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; + return cpu_rq(cpu)->cpu_power; } static int task_hot(struct task_struct *p, u64 now, struct sched_domain *sd); @@ -1544,101 +1539,6 @@ static unsigned long cpu_avg_load_per_task(int cpu) #ifdef CONFIG_FAIR_GROUP_SCHED -static __read_mostly unsigned long __percpu *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, - unsigned long *usd_rq_weight) -{ - unsigned long shares, rq_weight; - int boost = 0; - - rq_weight = usd_rq_weight[cpu]; - if (!rq_weight) { - boost = 1; - rq_weight = NICE_0_LOAD; - } - - /* - * \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); - - if (abs(shares - tg->se[cpu]->load.weight) > - sysctl_sched_shares_thresh) { - struct rq *rq = cpu_rq(cpu); - unsigned long flags; - - raw_spin_lock_irqsave(&rq->lock, flags); - 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); - raw_spin_unlock_irqrestore(&rq->lock, flags); - } -} - -/* - * Re-compute the task group their per cpu shares over the given domain. - * This needs to be done in a bottom-up fashion because the rq weight of a - * parent group depends on the shares of its child groups. - */ -static int tg_shares_up(struct task_group *tg, void *data) -{ - unsigned long weight, rq_weight = 0, sum_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; - - rq_weight += 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. - */ - if (!weight) - weight = NICE_0_LOAD; - - sum_weight += weight; - shares += tg->cfs_rq[i]->shares; - } - - if (!rq_weight) - rq_weight = sum_weight; - - if ((!shares && rq_weight) || shares > tg->shares) - shares = tg->shares; - - if (!sd->parent || !(sd->parent->flags & SD_LOAD_BALANCE)) - shares = tg->shares; - - for_each_cpu(i, sched_domain_span(sd)) - update_group_shares_cpu(tg, i, shares, rq_weight, usd_rq_weight); - - local_irq_restore(flags); - - return 0; -} - /* * Compute the cpu's hierarchical load factor for each task group. * This needs to be done in a top-down fashion because the load of a child @@ -1653,7 +1553,7 @@ static int tg_load_down(struct task_group *tg, void *data) load = cpu_rq(cpu)->load.weight; } else { load = tg->parent->cfs_rq[cpu]->h_load; - load *= tg->cfs_rq[cpu]->shares; + load *= tg->se[cpu]->load.weight; load /= tg->parent->cfs_rq[cpu]->load.weight + 1; } @@ -1662,37 +1562,11 @@ static int tg_load_down(struct task_group *tg, void *data) return 0; } -static void update_shares(struct sched_domain *sd) -{ - 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; - walk_tg_tree(tg_nop, tg_shares_up, sd); - } -} - static void update_h_load(long cpu) { - if (root_task_group_empty()) - return; - walk_tg_tree(tg_load_down, tg_nop, (void *)cpu); } -#else - -static inline void update_shares(struct sched_domain *sd) -{ -} - #endif #ifdef CONFIG_PREEMPT @@ -1814,18 +1688,10 @@ static void double_rq_unlock(struct rq *rq1, struct rq *rq2) #endif -#ifdef CONFIG_FAIR_GROUP_SCHED -static void cfs_rq_set_shares(struct cfs_rq *cfs_rq, unsigned long shares) -{ -#ifdef CONFIG_SMP - cfs_rq->shares = shares; -#endif -} -#endif - static void calc_load_account_idle(struct rq *this_rq); static void update_sysctl(void); static int get_update_sysctl_factor(void); +static void update_cpu_load(struct rq *this_rq); static inline void __set_task_cpu(struct task_struct *p, unsigned int cpu) { @@ -1843,7 +1709,7 @@ static inline void __set_task_cpu(struct task_struct *p, unsigned int cpu) static const struct sched_class rt_sched_class; -#define sched_class_highest (&rt_sched_class) +#define sched_class_highest (&stop_sched_class) #define for_each_class(class) \ for (class = sched_class_highest; class; class = class->next) @@ -1861,12 +1727,6 @@ static void dec_nr_running(struct rq *rq) static void set_load_weight(struct task_struct *p) { - if (task_has_rt_policy(p)) { - p->se.load.weight = prio_to_weight[0] * 2; - p->se.load.inv_weight = prio_to_wmult[0] >> 1; - return; - } - /* * SCHED_IDLE tasks get minimal weight: */ @@ -1920,13 +1780,194 @@ static void deactivate_task(struct rq *rq, struct task_struct *p, int flags) dec_nr_running(rq); } +#ifdef CONFIG_IRQ_TIME_ACCOUNTING + +/* + * There are no locks covering percpu hardirq/softirq time. + * They are only modified in account_system_vtime, on corresponding CPU + * with interrupts disabled. So, writes are safe. + * They are read and saved off onto struct rq in update_rq_clock(). + * This may result in other CPU reading this CPU's irq time and can + * race with irq/account_system_vtime on this CPU. We would either get old + * or new value with a side effect of accounting a slice of irq time to wrong + * task when irq is in progress while we read rq->clock. That is a worthy + * compromise in place of having locks on each irq in account_system_time. + */ +static DEFINE_PER_CPU(u64, cpu_hardirq_time); +static DEFINE_PER_CPU(u64, cpu_softirq_time); + +static DEFINE_PER_CPU(u64, irq_start_time); +static int sched_clock_irqtime; + +void enable_sched_clock_irqtime(void) +{ + sched_clock_irqtime = 1; +} + +void disable_sched_clock_irqtime(void) +{ + sched_clock_irqtime = 0; +} + +#ifndef CONFIG_64BIT +static DEFINE_PER_CPU(seqcount_t, irq_time_seq); + +static inline void irq_time_write_begin(void) +{ + __this_cpu_inc(irq_time_seq.sequence); + smp_wmb(); +} + +static inline void irq_time_write_end(void) +{ + smp_wmb(); + __this_cpu_inc(irq_time_seq.sequence); +} + +static inline u64 irq_time_read(int cpu) +{ + u64 irq_time; + unsigned seq; + + do { + seq = read_seqcount_begin(&per_cpu(irq_time_seq, cpu)); + irq_time = per_cpu(cpu_softirq_time, cpu) + + per_cpu(cpu_hardirq_time, cpu); + } while (read_seqcount_retry(&per_cpu(irq_time_seq, cpu), seq)); + + return irq_time; +} +#else /* CONFIG_64BIT */ +static inline void irq_time_write_begin(void) +{ +} + +static inline void irq_time_write_end(void) +{ +} + +static inline u64 irq_time_read(int cpu) +{ + return per_cpu(cpu_softirq_time, cpu) + per_cpu(cpu_hardirq_time, cpu); +} +#endif /* CONFIG_64BIT */ + +/* + * Called before incrementing preempt_count on {soft,}irq_enter + * and before decrementing preempt_count on {soft,}irq_exit. + */ +void account_system_vtime(struct task_struct *curr) +{ + unsigned long flags; + s64 delta; + int cpu; + + if (!sched_clock_irqtime) + return; + + local_irq_save(flags); + + cpu = smp_processor_id(); + delta = sched_clock_cpu(cpu) - __this_cpu_read(irq_start_time); + __this_cpu_add(irq_start_time, delta); + + irq_time_write_begin(); + /* + * We do not account for softirq time from ksoftirqd here. + * We want to continue accounting softirq time to ksoftirqd thread + * in that case, so as not to confuse scheduler with a special task + * that do not consume any time, but still wants to run. + */ + if (hardirq_count()) + __this_cpu_add(cpu_hardirq_time, delta); + else if (in_serving_softirq() && !(curr->flags & PF_KSOFTIRQD)) + __this_cpu_add(cpu_softirq_time, delta); + + irq_time_write_end(); + local_irq_restore(flags); +} +EXPORT_SYMBOL_GPL(account_system_vtime); + +static void update_rq_clock_task(struct rq *rq, s64 delta) +{ + s64 irq_delta; + + irq_delta = irq_time_read(cpu_of(rq)) - rq->prev_irq_time; + + /* + * Since irq_time is only updated on {soft,}irq_exit, we might run into + * this case when a previous update_rq_clock() happened inside a + * {soft,}irq region. + * + * When this happens, we stop ->clock_task and only update the + * prev_irq_time stamp to account for the part that fit, so that a next + * update will consume the rest. This ensures ->clock_task is + * monotonic. + * + * It does however cause some slight miss-attribution of {soft,}irq + * time, a more accurate solution would be to update the irq_time using + * the current rq->clock timestamp, except that would require using + * atomic ops. + */ + if (irq_delta > delta) + irq_delta = delta; + + rq->prev_irq_time += irq_delta; + delta -= irq_delta; + rq->clock_task += delta; + + if (irq_delta && sched_feat(NONIRQ_POWER)) + sched_rt_avg_update(rq, irq_delta); +} + +#else /* CONFIG_IRQ_TIME_ACCOUNTING */ + +static void update_rq_clock_task(struct rq *rq, s64 delta) +{ + rq->clock_task += delta; +} + +#endif /* CONFIG_IRQ_TIME_ACCOUNTING */ + #include "sched_idletask.c" #include "sched_fair.c" #include "sched_rt.c" +#include "sched_autogroup.c" +#include "sched_stoptask.c" #ifdef CONFIG_SCHED_DEBUG # include "sched_debug.c" #endif +void sched_set_stop_task(int cpu, struct task_struct *stop) +{ + struct sched_param param = { .sched_priority = MAX_RT_PRIO - 1 }; + struct task_struct *old_stop = cpu_rq(cpu)->stop; + + if (stop) { + /* + * Make it appear like a SCHED_FIFO task, its something + * userspace knows about and won't get confused about. + * + * Also, it will make PI more or less work without too + * much confusion -- but then, stop work should not + * rely on PI working anyway. + */ + sched_setscheduler_nocheck(stop, SCHED_FIFO, ¶m); + + stop->sched_class = &stop_sched_class; + } + + cpu_rq(cpu)->stop = stop; + + if (old_stop) { + /* + * Reset it back to a normal scheduling class so that + * it can die in pieces. + */ + old_stop->sched_class = &rt_sched_class; + } +} + /* * __normal_prio - return the priority that is based on the static prio */ @@ -1994,6 +2035,31 @@ static inline void check_class_changed(struct rq *rq, struct task_struct *p, p->sched_class->prio_changed(rq, p, oldprio, running); } +static void check_preempt_curr(struct rq *rq, struct task_struct *p, int flags) +{ + const struct sched_class *class; + + if (p->sched_class == rq->curr->sched_class) { + rq->curr->sched_class->check_preempt_curr(rq, p, flags); + } else { + for_each_class(class) { + if (class == rq->curr->sched_class) + break; + if (class == p->sched_class) { + resched_task(rq->curr); + break; + } + } + } + + /* + * A queue event has occurred, and we're going to schedule. In + * this case, we can save a useless back to back clock update. + */ + if (rq->curr->se.on_rq && test_tsk_need_resched(rq->curr)) + rq->skip_clock_update = 1; +} + #ifdef CONFIG_SMP /* * Is this task likely cache-hot: @@ -2006,6 +2072,9 @@ task_hot(struct task_struct *p, u64 now, struct sched_domain *sd) if (p->sched_class != &fair_sched_class) return 0; + if (unlikely(p->policy == SCHED_IDLE)) + return 0; + /* * Buddy candidates are cache hot: */ @@ -2056,10 +2125,8 @@ static int migration_cpu_stop(void *data); * The task's runqueue lock must be held. * Returns true if you have to wait for migration thread. */ -static bool migrate_task(struct task_struct *p, int dest_cpu) +static bool migrate_task(struct task_struct *p, struct rq *rq) { - struct rq *rq = task_rq(p); - /* * If the task is not on a runqueue (and not running), then * the next wake-up will properly place the task. @@ -2239,18 +2306,15 @@ static int select_fallback_rq(int cpu, struct task_struct *p) return dest_cpu; /* No more Mr. Nice Guy. */ - if (unlikely(dest_cpu >= nr_cpu_ids)) { - dest_cpu = cpuset_cpus_allowed_fallback(p); - /* - * Don't tell them about moving exiting tasks or - * kernel threads (both mm NULL), since they never - * leave kernel. - */ - if (p->mm && printk_ratelimit()) { - printk(KERN_INFO "process %d (%s) no " - "longer affine to cpu%d\n", - task_pid_nr(p), p->comm, cpu); - } + dest_cpu = cpuset_cpus_allowed_fallback(p); + /* + * Don't tell them about moving exiting tasks or + * kernel threads (both mm NULL), since they never + * leave kernel. + */ + if (p->mm && printk_ratelimit()) { + printk(KERN_INFO "process %d (%s) no longer affine to cpu%d\n", + task_pid_nr(p), p->comm, cpu); } return dest_cpu; @@ -2288,11 +2352,55 @@ static void update_avg(u64 *avg, u64 sample) } #endif -/*** +static inline void ttwu_activate(struct task_struct *p, struct rq *rq, + bool is_sync, bool is_migrate, bool is_local, + unsigned long en_flags) +{ + schedstat_inc(p, se.statistics.nr_wakeups); + if (is_sync) + schedstat_inc(p, se.statistics.nr_wakeups_sync); + if (is_migrate) + schedstat_inc(p, se.statistics.nr_wakeups_migrate); + if (is_local) + schedstat_inc(p, se.statistics.nr_wakeups_local); + else + schedstat_inc(p, se.statistics.nr_wakeups_remote); + + activate_task(rq, p, en_flags); +} + +static inline void ttwu_post_activation(struct task_struct *p, struct rq *rq, + int wake_flags, bool success) +{ + trace_sched_wakeup(p, success); + check_preempt_curr(rq, p, wake_flags); + + p->state = TASK_RUNNING; +#ifdef CONFIG_SMP + if (p->sched_class->task_woken) + p->sched_class->task_woken(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 + /* if a worker is waking up, notify workqueue */ + if ((p->flags & PF_WQ_WORKER) && success) + wq_worker_waking_up(p, cpu_of(rq)); +} + +/** * try_to_wake_up - wake up a thread - * @p: the to-be-woken-up thread + * @p: the thread to be awakened * @state: the mask of task states that can be woken - * @sync: do a synchronous wakeup? + * @wake_flags: wake modifier flags (WF_*) * * Put it on the run-queue if it's not already there. The "current" * thread is always on the run-queue (except when the actual @@ -2300,7 +2408,8 @@ static void update_avg(u64 *avg, u64 sample) * the simpler "current->state = TASK_RUNNING" to mark yourself * runnable without the overhead of this. * - * returns failure only if the task is already active. + * Returns %true if @p was woken up, %false if it was already running + * or @state didn't match @p's state. */ static int try_to_wake_up(struct task_struct *p, unsigned int state, int wake_flags) @@ -2380,38 +2489,11 @@ static int try_to_wake_up(struct task_struct *p, unsigned int state, out_activate: #endif /* CONFIG_SMP */ - schedstat_inc(p, se.statistics.nr_wakeups); - if (wake_flags & WF_SYNC) - schedstat_inc(p, se.statistics.nr_wakeups_sync); - if (orig_cpu != cpu) - schedstat_inc(p, se.statistics.nr_wakeups_migrate); - if (cpu == this_cpu) - schedstat_inc(p, se.statistics.nr_wakeups_local); - else - schedstat_inc(p, se.statistics.nr_wakeups_remote); - activate_task(rq, p, en_flags); + ttwu_activate(p, rq, wake_flags & WF_SYNC, orig_cpu != cpu, + cpu == this_cpu, en_flags); success = 1; - out_running: - trace_sched_wakeup(p, success); - check_preempt_curr(rq, p, wake_flags); - - p->state = TASK_RUNNING; -#ifdef CONFIG_SMP - if (p->sched_class->task_woken) - p->sched_class->task_woken(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 + ttwu_post_activation(p, rq, wake_flags, success); out: task_rq_unlock(rq, &flags); put_cpu(); @@ -2420,6 +2502,37 @@ out: } /** + * try_to_wake_up_local - try to wake up a local task with rq lock held + * @p: the thread to be awakened + * + * Put @p on the run-queue if it's not already there. The caller must + * ensure that this_rq() is locked, @p is bound to this_rq() and not + * the current task. this_rq() stays locked over invocation. + */ +static void try_to_wake_up_local(struct task_struct *p) +{ + struct rq *rq = task_rq(p); + bool success = false; + + BUG_ON(rq != this_rq()); + BUG_ON(p == current); + lockdep_assert_held(&rq->lock); + + if (!(p->state & TASK_NORMAL)) + return; + + if (!p->se.on_rq) { + if (likely(!task_running(rq, p))) { + schedstat_inc(rq, ttwu_count); + schedstat_inc(rq, ttwu_local); + } + ttwu_activate(p, rq, false, false, true, ENQUEUE_WAKEUP); + success = true; + } + ttwu_post_activation(p, rq, 0, success); +} + +/** * wake_up_process - Wake up a specific process * @p: The process to be woken up. * @@ -2515,7 +2628,16 @@ void sched_fork(struct task_struct *p, int clone_flags) if (p->sched_class->task_fork) p->sched_class->task_fork(p); + /* + * The child is not yet in the pid-hash so no cgroup attach races, + * and the cgroup is pinned to this child due to cgroup_fork() + * is ran before sched_fork(). + * + * Silence PROVE_RCU. + */ + rcu_read_lock(); set_task_cpu(p, cpu); + rcu_read_unlock(); #if defined(CONFIG_SCHEDSTATS) || defined(CONFIG_TASK_DELAY_ACCT) if (likely(sched_info_on())) @@ -2528,7 +2650,9 @@ void sched_fork(struct task_struct *p, int clone_flags) /* Want to start with kernel preemption disabled. */ task_thread_info(p)->preempt_count = 1; #endif +#ifdef CONFIG_SMP plist_node_init(&p->pushable_tasks, MAX_PRIO); +#endif put_cpu(); } @@ -2797,14 +2921,14 @@ context_switch(struct rq *rq, struct task_struct *prev, */ arch_start_context_switch(prev); - if (likely(!mm)) { + if (!mm) { next->active_mm = oldmm; atomic_inc(&oldmm->mm_count); enter_lazy_tlb(oldmm, next); } else switch_mm(oldmm, mm, next); - if (likely(!prev->mm)) { + if (!prev->mm) { prev->active_mm = NULL; rq->prev_mm = oldmm; } @@ -2885,9 +3009,9 @@ unsigned long nr_iowait(void) return sum; } -unsigned long nr_iowait_cpu(void) +unsigned long nr_iowait_cpu(int cpu) { - struct rq *this = this_rq(); + struct rq *this = cpu_rq(cpu); return atomic_read(&this->nr_iowait); } @@ -2919,6 +3043,15 @@ static long calc_load_fold_active(struct rq *this_rq) return delta; } +static unsigned long +calc_load(unsigned long load, unsigned long exp, unsigned long active) +{ + load *= exp; + load += active * (FIXED_1 - exp); + load += 1UL << (FSHIFT - 1); + return load >> FSHIFT; +} + #ifdef CONFIG_NO_HZ /* * For NO_HZ we delay the active fold to the next LOAD_FREQ update. @@ -2948,6 +3081,128 @@ static long calc_load_fold_idle(void) return delta; } + +/** + * fixed_power_int - compute: x^n, in O(log n) time + * + * @x: base of the power + * @frac_bits: fractional bits of @x + * @n: power to raise @x to. + * + * By exploiting the relation between the definition of the natural power + * function: x^n := x*x*...*x (x multiplied by itself for n times), and + * the binary encoding of numbers used by computers: n := \Sum n_i * 2^i, + * (where: n_i \elem {0, 1}, the binary vector representing n), + * we find: x^n := x^(\Sum n_i * 2^i) := \Prod x^(n_i * 2^i), which is + * of course trivially computable in O(log_2 n), the length of our binary + * vector. + */ +static unsigned long +fixed_power_int(unsigned long x, unsigned int frac_bits, unsigned int n) +{ + unsigned long result = 1UL << frac_bits; + + if (n) for (;;) { + if (n & 1) { + result *= x; + result += 1UL << (frac_bits - 1); + result >>= frac_bits; + } + n >>= 1; + if (!n) + break; + x *= x; + x += 1UL << (frac_bits - 1); + x >>= frac_bits; + } + + return result; +} + +/* + * a1 = a0 * e + a * (1 - e) + * + * a2 = a1 * e + a * (1 - e) + * = (a0 * e + a * (1 - e)) * e + a * (1 - e) + * = a0 * e^2 + a * (1 - e) * (1 + e) + * + * a3 = a2 * e + a * (1 - e) + * = (a0 * e^2 + a * (1 - e) * (1 + e)) * e + a * (1 - e) + * = a0 * e^3 + a * (1 - e) * (1 + e + e^2) + * + * ... + * + * an = a0 * e^n + a * (1 - e) * (1 + e + ... + e^n-1) [1] + * = a0 * e^n + a * (1 - e) * (1 - e^n)/(1 - e) + * = a0 * e^n + a * (1 - e^n) + * + * [1] application of the geometric series: + * + * n 1 - x^(n+1) + * S_n := \Sum x^i = ------------- + * i=0 1 - x + */ +static unsigned long +calc_load_n(unsigned long load, unsigned long exp, + unsigned long active, unsigned int n) +{ + + return calc_load(load, fixed_power_int(exp, FSHIFT, n), active); +} + +/* + * NO_HZ can leave us missing all per-cpu ticks calling + * calc_load_account_active(), but since an idle CPU folds its delta into + * calc_load_tasks_idle per calc_load_account_idle(), all we need to do is fold + * in the pending idle delta if our idle period crossed a load cycle boundary. + * + * Once we've updated the global active value, we need to apply the exponential + * weights adjusted to the number of cycles missed. + */ +static void calc_global_nohz(unsigned long ticks) +{ + long delta, active, n; + + if (time_before(jiffies, calc_load_update)) + return; + + /* + * If we crossed a calc_load_update boundary, make sure to fold + * any pending idle changes, the respective CPUs might have + * missed the tick driven calc_load_account_active() update + * due to NO_HZ. + */ + delta = calc_load_fold_idle(); + if (delta) + atomic_long_add(delta, &calc_load_tasks); + + /* + * If we were idle for multiple load cycles, apply them. + */ + if (ticks >= LOAD_FREQ) { + n = ticks / LOAD_FREQ; + + active = atomic_long_read(&calc_load_tasks); + active = active > 0 ? active * FIXED_1 : 0; + + avenrun[0] = calc_load_n(avenrun[0], EXP_1, active, n); + avenrun[1] = calc_load_n(avenrun[1], EXP_5, active, n); + avenrun[2] = calc_load_n(avenrun[2], EXP_15, active, n); + + calc_load_update += n * LOAD_FREQ; + } + + /* + * Its possible the remainder of the above division also crosses + * a LOAD_FREQ period, the regular check in calc_global_load() + * which comes after this will take care of that. + * + * Consider us being 11 ticks before a cycle completion, and us + * sleeping for 4*LOAD_FREQ + 22 ticks, then the above code will + * age us 4 cycles, and the test in calc_global_load() will + * pick up the final one. + */ +} #else static void calc_load_account_idle(struct rq *this_rq) { @@ -2957,6 +3212,10 @@ static inline long calc_load_fold_idle(void) { return 0; } + +static void calc_global_nohz(unsigned long ticks) +{ +} #endif /** @@ -2974,24 +3233,17 @@ void get_avenrun(unsigned long *loads, unsigned long offset, int 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) +void calc_global_load(unsigned long ticks) { - unsigned long upd = calc_load_update + 10; long active; - if (time_before(jiffies, upd)) + calc_global_nohz(ticks); + + if (time_before(jiffies, calc_load_update + 10)) return; active = atomic_long_read(&calc_load_tasks); @@ -3024,23 +3276,102 @@ static void calc_load_account_active(struct rq *this_rq) } /* + * The exact cpuload at various idx values, calculated at every tick would be + * load = (2^idx - 1) / 2^idx * load + 1 / 2^idx * cur_load + * + * If a cpu misses updates for n-1 ticks (as it was idle) and update gets called + * on nth tick when cpu may be busy, then we have: + * load = ((2^idx - 1) / 2^idx)^(n-1) * load + * load = (2^idx - 1) / 2^idx) * load + 1 / 2^idx * cur_load + * + * decay_load_missed() below does efficient calculation of + * load = ((2^idx - 1) / 2^idx)^(n-1) * load + * avoiding 0..n-1 loop doing load = ((2^idx - 1) / 2^idx) * load + * + * The calculation is approximated on a 128 point scale. + * degrade_zero_ticks is the number of ticks after which load at any + * particular idx is approximated to be zero. + * degrade_factor is a precomputed table, a row for each load idx. + * Each column corresponds to degradation factor for a power of two ticks, + * based on 128 point scale. + * Example: + * row 2, col 3 (=12) says that the degradation at load idx 2 after + * 8 ticks is 12/128 (which is an approximation of exact factor 3^8/4^8). + * + * With this power of 2 load factors, we can degrade the load n times + * by looking at 1 bits in n and doing as many mult/shift instead of + * n mult/shifts needed by the exact degradation. + */ +#define DEGRADE_SHIFT 7 +static const unsigned char + degrade_zero_ticks[CPU_LOAD_IDX_MAX] = {0, 8, 32, 64, 128}; +static const unsigned char + degrade_factor[CPU_LOAD_IDX_MAX][DEGRADE_SHIFT + 1] = { + {0, 0, 0, 0, 0, 0, 0, 0}, + {64, 32, 8, 0, 0, 0, 0, 0}, + {96, 72, 40, 12, 1, 0, 0}, + {112, 98, 75, 43, 15, 1, 0}, + {120, 112, 98, 76, 45, 16, 2} }; + +/* + * Update cpu_load for any missed ticks, due to tickless idle. The backlog + * would be when CPU is idle and so we just decay the old load without + * adding any new load. + */ +static unsigned long +decay_load_missed(unsigned long load, unsigned long missed_updates, int idx) +{ + int j = 0; + + if (!missed_updates) + return load; + + if (missed_updates >= degrade_zero_ticks[idx]) + return 0; + + if (idx == 1) + return load >> missed_updates; + + while (missed_updates) { + if (missed_updates % 2) + load = (load * degrade_factor[idx][j]) >> DEGRADE_SHIFT; + + missed_updates >>= 1; + j++; + } + return load; +} + +/* * Update rq->cpu_load[] statistics. This function is usually called every - * scheduler tick (TICK_NSEC). + * scheduler tick (TICK_NSEC). With tickless idle this will not be called + * every tick. We fix it up based on jiffies. */ static void update_cpu_load(struct rq *this_rq) { unsigned long this_load = this_rq->load.weight; + unsigned long curr_jiffies = jiffies; + unsigned long pending_updates; int i, scale; this_rq->nr_load_updates++; + /* Avoid repeated calls on same jiffy, when moving in and out of idle */ + if (curr_jiffies == this_rq->last_load_update_tick) + return; + + pending_updates = curr_jiffies - this_rq->last_load_update_tick; + this_rq->last_load_update_tick = curr_jiffies; + /* Update our load: */ - for (i = 0, scale = 1; i < CPU_LOAD_IDX_MAX; i++, scale += scale) { + this_rq->cpu_load[0] = this_load; /* Fasttrack for idx 0 */ + for (i = 1, scale = 2; i < CPU_LOAD_IDX_MAX; i++, scale += scale) { unsigned long old_load, new_load; /* scale is effectively 1 << i now, and >> i divides by scale */ old_load = this_rq->cpu_load[i]; + old_load = decay_load_missed(old_load, pending_updates - 1, i); new_load = this_load; /* * Round up the averaging division if load is increasing. This @@ -3048,10 +3379,18 @@ static void update_cpu_load(struct rq *this_rq) * example. */ if (new_load > old_load) - new_load += scale-1; - this_rq->cpu_load[i] = (old_load*(scale-1) + new_load) >> i; + new_load += scale - 1; + + this_rq->cpu_load[i] = (old_load * (scale - 1) + new_load) >> i; } + sched_avg_update(this_rq); +} + +static void update_cpu_load_active(struct rq *this_rq) +{ + update_cpu_load(this_rq); + calc_load_account_active(this_rq); } @@ -3077,7 +3416,7 @@ void sched_exec(void) * select_task_rq() can race against ->cpus_allowed */ if (cpumask_test_cpu(dest_cpu, &p->cpus_allowed) && - likely(cpu_active(dest_cpu)) && migrate_task(p, dest_cpu)) { + likely(cpu_active(dest_cpu)) && migrate_task(p, rq)) { struct migration_arg arg = { p, dest_cpu }; task_rq_unlock(rq, &flags); @@ -3106,7 +3445,7 @@ static u64 do_task_delta_exec(struct task_struct *p, struct rq *rq) if (task_current(rq, p)) { update_rq_clock(rq); - ns = rq->clock - p->se.exec_start; + ns = rq->clock_task - p->se.exec_start; if ((s64)ns < 0) ns = 0; } @@ -3255,7 +3594,7 @@ void account_system_time(struct task_struct *p, int hardirq_offset, tmp = cputime_to_cputime64(cputime); if (hardirq_count() - hardirq_offset) cpustat->irq = cputime64_add(cpustat->irq, tmp); - else if (softirq_count()) + else if (in_serving_softirq()) cpustat->softirq = cputime64_add(cpustat->softirq, tmp); else cpustat->system = cputime64_add(cpustat->system, tmp); @@ -3371,9 +3710,9 @@ void task_times(struct task_struct *p, cputime_t *ut, cputime_t *st) rtime = nsecs_to_cputime(p->se.sum_exec_runtime); if (total) { - u64 temp; + u64 temp = rtime; - temp = (u64)(rtime * utime); + temp *= utime; do_div(temp, total); utime = (cputime_t)temp; } else @@ -3404,9 +3743,9 @@ void thread_group_times(struct task_struct *p, cputime_t *ut, cputime_t *st) rtime = nsecs_to_cputime(cputime.sum_exec_runtime); if (total) { - u64 temp; + u64 temp = rtime; - temp = (u64)(rtime * cputime.utime); + temp *= cputime.utime; do_div(temp, total); utime = (cputime_t)temp; } else @@ -3438,11 +3777,11 @@ void scheduler_tick(void) raw_spin_lock(&rq->lock); update_rq_clock(rq); - update_cpu_load(rq); + update_cpu_load_active(rq); curr->sched_class->task_tick(rq, curr, 0); raw_spin_unlock(&rq->lock); - perf_event_task_tick(curr); + perf_event_task_tick(); #ifdef CONFIG_SMP rq->idle_at_tick = idle_cpu(cpu); @@ -3548,7 +3887,7 @@ static inline void schedule_debug(struct task_struct *prev) schedstat_inc(this_rq(), sched_count); #ifdef CONFIG_SCHEDSTATS if (unlikely(prev->lock_depth >= 0)) { - schedstat_inc(this_rq(), bkl_count); + schedstat_inc(this_rq(), rq_sched_info.bkl_count); schedstat_inc(prev, sched_info.bkl_count); } #endif @@ -3558,7 +3897,6 @@ static void put_prev_task(struct rq *rq, struct task_struct *prev) { if (prev->se.on_rq) update_rq_clock(rq); - rq->skip_clock_update = 0; prev->sched_class->put_prev_task(rq, prev); } @@ -3581,17 +3919,13 @@ pick_next_task(struct rq *rq) return p; } - class = sched_class_highest; - for ( ; ; ) { + for_each_class(class) { p = class->pick_next_task(rq); if (p) return p; - /* - * Will never be NULL as the idle class always - * returns a non-NULL p: - */ - class = class->next; } + + BUG(); /* the idle class will always have a runnable task */ } /* @@ -3610,7 +3944,6 @@ need_resched: rq = cpu_rq(cpu); rcu_note_context_switch(cpu); prev = rq->curr; - switch_count = &prev->nivcsw; release_kernel_lock(prev); need_resched_nonpreemptible: @@ -3621,13 +3954,27 @@ need_resched_nonpreemptible: hrtick_clear(rq); raw_spin_lock_irq(&rq->lock); - clear_tsk_need_resched(prev); + switch_count = &prev->nivcsw; if (prev->state && !(preempt_count() & PREEMPT_ACTIVE)) { - if (unlikely(signal_pending_state(prev->state, prev))) + if (unlikely(signal_pending_state(prev->state, prev))) { prev->state = TASK_RUNNING; - else + } else { + /* + * If a worker is going to sleep, notify and + * ask workqueue whether it wants to wake up a + * task to maintain concurrency. If so, wake + * up the task. + */ + if (prev->flags & PF_WQ_WORKER) { + struct task_struct *to_wakeup; + + to_wakeup = wq_worker_sleeping(prev, cpu); + if (to_wakeup) + try_to_wake_up_local(to_wakeup); + } deactivate_task(rq, prev, DEQUEUE_SLEEP); + } switch_count = &prev->nvcsw; } @@ -3638,6 +3985,8 @@ need_resched_nonpreemptible: put_prev_task(rq, prev); next = pick_next_task(rq); + clear_tsk_need_resched(prev); + rq->skip_clock_update = 0; if (likely(prev != next)) { sched_info_switch(prev, next); @@ -3649,8 +3998,10 @@ need_resched_nonpreemptible: context_switch(rq, prev, next); /* unlocks the rq */ /* - * the context switch might have flipped the stack from under - * us, hence refresh the local variables. + * The context switch have flipped the stack from under us + * and restored the local variables which were saved when + * this task called schedule() in the past. prev == current + * is still correct, but it can be moved to another cpu/rq. */ cpu = smp_processor_id(); rq = cpu_rq(cpu); @@ -3659,11 +4010,8 @@ need_resched_nonpreemptible: post_schedule(rq); - if (unlikely(reacquire_kernel_lock(current) < 0)) { - prev = rq->curr; - switch_count = &prev->nivcsw; + if (unlikely(reacquire_kernel_lock(prev))) goto need_resched_nonpreemptible; - } preempt_enable_no_resched(); if (need_resched()) @@ -3716,8 +4064,16 @@ int mutex_spin_on_owner(struct mutex *lock, struct thread_info *owner) /* * Owner changed, break to re-assess state. */ - if (lock->owner != owner) + if (lock->owner != owner) { + /* + * If the lock has switched to a different owner, + * we likely have heavy contention. Return 0 to quit + * optimistic spinning and not contend further: + */ + if (lock->owner) + return 0; break; + } /* * Is that owner really running on that cpu? @@ -3725,7 +4081,7 @@ int mutex_spin_on_owner(struct mutex *lock, struct thread_info *owner) if (task_thread_info(rq->curr) != owner || need_resched()) return 0; - cpu_relax(); + arch_mutex_cpu_relax(); } return 1; @@ -3738,7 +4094,7 @@ int mutex_spin_on_owner(struct mutex *lock, struct thread_info *owner) * off of preempt_enable. Kernel preemptions off return from interrupt * occur there and call schedule directly. */ -asmlinkage void __sched preempt_schedule(void) +asmlinkage void __sched notrace preempt_schedule(void) { struct thread_info *ti = current_thread_info(); @@ -3750,9 +4106,9 @@ asmlinkage void __sched preempt_schedule(void) return; do { - add_preempt_count(PREEMPT_ACTIVE); + add_preempt_count_notrace(PREEMPT_ACTIVE); schedule(); - sub_preempt_count(PREEMPT_ACTIVE); + sub_preempt_count_notrace(PREEMPT_ACTIVE); /* * Check again in case we missed a preemption opportunity @@ -4037,7 +4393,7 @@ EXPORT_SYMBOL(wait_for_completion_interruptible); * This waits for either a completion of a specific task to be signaled or for a * specified timeout to expire. It is interruptible. The timeout is in jiffies. */ -unsigned long __sched +long __sched wait_for_completion_interruptible_timeout(struct completion *x, unsigned long timeout) { @@ -4062,6 +4418,23 @@ int __sched wait_for_completion_killable(struct completion *x) EXPORT_SYMBOL(wait_for_completion_killable); /** + * wait_for_completion_killable_timeout: - waits for completion of a task (w/(to,killable)) + * @x: holds the state of this particular completion + * @timeout: timeout value in jiffies + * + * This waits for either a completion of a specific task to be + * signaled or for a specified timeout to expire. It can be + * interrupted by a kill signal. The timeout is in jiffies. + */ +long __sched +wait_for_completion_killable_timeout(struct completion *x, + unsigned long timeout) +{ + return wait_for_common(x, timeout, TASK_KILLABLE); +} +EXPORT_SYMBOL(wait_for_completion_killable_timeout); + +/** * try_wait_for_completion - try to decrement a completion without blocking * @x: completion structure * @@ -4178,6 +4551,7 @@ void rt_mutex_setprio(struct task_struct *p, int prio) rq = task_rq_lock(p, &flags); + trace_sched_pi_setprio(p, prio); oldprio = p->prio; prev_class = p->sched_class; on_rq = p->se.on_rq; @@ -4394,7 +4768,7 @@ static bool check_same_owner(struct task_struct *p) } static int __sched_setscheduler(struct task_struct *p, int policy, - struct sched_param *param, bool user) + const struct sched_param *param, bool user) { int retval, oldprio, oldpolicy = -1, on_rq, running; unsigned long flags; @@ -4436,12 +4810,8 @@ recheck: */ if (user && !capable(CAP_SYS_NICE)) { if (rt_policy(policy)) { - unsigned long rlim_rtprio; - - if (!lock_task_sighand(p, &flags)) - return -ESRCH; - rlim_rtprio = task_rlimit(p, RLIMIT_RTPRIO); - unlock_task_sighand(p, &flags); + unsigned long rlim_rtprio = + task_rlimit(p, RLIMIT_RTPRIO); /* can't set/change the rt policy */ if (policy != p->policy && !rlim_rtprio) @@ -4469,17 +4839,7 @@ recheck: } if (user) { -#ifdef CONFIG_RT_GROUP_SCHED - /* - * Do not allow realtime tasks into groups that have no runtime - * assigned. - */ - if (rt_bandwidth_enabled() && rt_policy(policy) && - task_group(p)->rt_bandwidth.rt_runtime == 0) - return -EPERM; -#endif - - retval = security_task_setscheduler(p, policy, param); + retval = security_task_setscheduler(p); if (retval) return retval; } @@ -4494,6 +4854,32 @@ recheck: * runqueue lock must be held. */ rq = __task_rq_lock(p); + + /* + * Changing the policy of the stop threads its a very bad idea + */ + if (p == rq->stop) { + __task_rq_unlock(rq); + raw_spin_unlock_irqrestore(&p->pi_lock, flags); + return -EINVAL; + } + +#ifdef CONFIG_RT_GROUP_SCHED + if (user) { + /* + * Do not allow realtime tasks into groups that have no runtime + * assigned. + */ + if (rt_bandwidth_enabled() && rt_policy(policy) && + task_group(p)->rt_bandwidth.rt_runtime == 0 && + !task_group_is_autogroup(task_group(p))) { + __task_rq_unlock(rq); + raw_spin_unlock_irqrestore(&p->pi_lock, flags); + return -EPERM; + } + } +#endif + /* recheck policy now with rq lock held */ if (unlikely(oldpolicy != -1 && oldpolicy != p->policy)) { policy = oldpolicy = -1; @@ -4538,7 +4924,7 @@ recheck: * NOTE that the task may be already dead. */ int sched_setscheduler(struct task_struct *p, int policy, - struct sched_param *param) + const struct sched_param *param) { return __sched_setscheduler(p, policy, param, true); } @@ -4556,7 +4942,7 @@ EXPORT_SYMBOL_GPL(sched_setscheduler); * but our caller might not have that capability. */ int sched_setscheduler_nocheck(struct task_struct *p, int policy, - struct sched_param *param) + const struct sched_param *param) { return __sched_setscheduler(p, policy, param, false); } @@ -4705,13 +5091,13 @@ long sched_setaffinity(pid_t pid, const struct cpumask *in_mask) if (!check_same_owner(p) && !capable(CAP_SYS_NICE)) goto out_unlock; - retval = security_task_setscheduler(p, 0, NULL); + retval = security_task_setscheduler(p); if (retval) goto out_unlock; cpuset_cpus_allowed(p, cpus_allowed); cpumask_and(new_mask, in_mask, cpus_allowed); - again: +again: retval = set_cpus_allowed_ptr(p, new_mask); if (!retval) { @@ -5072,7 +5458,7 @@ void sched_show_task(struct task_struct *p) unsigned state; state = p->state ? __ffs(p->state) + 1 : 0; - printk(KERN_INFO "%-13.13s %c", p->comm, + printk(KERN_INFO "%-15.15s %c", p->comm, state < sizeof(stat_nam) - 1 ? stat_nam[state] : '?'); #if BITS_PER_LONG == 32 if (state == TASK_RUNNING) @@ -5155,7 +5541,19 @@ void __cpuinit init_idle(struct task_struct *idle, int cpu) idle->se.exec_start = sched_clock(); cpumask_copy(&idle->cpus_allowed, cpumask_of(cpu)); + /* + * We're having a chicken and egg problem, even though we are + * holding rq->lock, the cpu isn't yet set to this cpu so the + * lockdep check in task_group() will fail. + * + * Similar case to sched_fork(). / Alternatively we could + * use task_rq_lock() here and obtain the other rq->lock. + * + * Silence PROVE_RCU + */ + rcu_read_lock(); __set_task_cpu(idle, cpu); + rcu_read_unlock(); rq->curr = rq->idle = idle; #if defined(CONFIG_SMP) && defined(__ARCH_WANT_UNLOCKED_CTXSW) @@ -5224,7 +5622,6 @@ static void update_sysctl(void) SET_SYSCTL(sched_min_granularity); SET_SYSCTL(sched_latency); SET_SYSCTL(sched_wakeup_granularity); - SET_SYSCTL(sched_shares_ratelimit); #undef SET_SYSCTL } @@ -5300,7 +5697,7 @@ again: goto out; dest_cpu = cpumask_any_and(cpu_active_mask, new_mask); - if (migrate_task(p, dest_cpu)) { + if (migrate_task(p, rq)) { struct migration_arg arg = { p, dest_cpu }; /* Need help from migration thread: drop lock and wait. */ task_rq_unlock(rq, &flags); @@ -5382,29 +5779,20 @@ static int migration_cpu_stop(void *data) } #ifdef CONFIG_HOTPLUG_CPU + /* - * Figure out where task on dead CPU should go, use force if necessary. + * Ensures that the idle task is using init_mm right before its cpu goes + * offline. */ -void move_task_off_dead_cpu(int dead_cpu, struct task_struct *p) +void idle_task_exit(void) { - struct rq *rq = cpu_rq(dead_cpu); - int needs_cpu, uninitialized_var(dest_cpu); - unsigned long flags; + struct mm_struct *mm = current->active_mm; - local_irq_save(flags); + BUG_ON(cpu_online(smp_processor_id())); - raw_spin_lock(&rq->lock); - needs_cpu = (task_cpu(p) == dead_cpu) && (p->state != TASK_WAKING); - if (needs_cpu) - dest_cpu = select_fallback_rq(dead_cpu, p); - raw_spin_unlock(&rq->lock); - /* - * It can only fail if we race with set_cpus_allowed(), - * in the racer should migrate the task anyway. - */ - if (needs_cpu) - __migrate_task(p, dead_cpu, dest_cpu); - local_irq_restore(flags); + if (mm != &init_mm) + switch_mm(mm, &init_mm, current); + mmdrop(mm); } /* @@ -5417,128 +5805,69 @@ void move_task_off_dead_cpu(int dead_cpu, struct task_struct *p) static void migrate_nr_uninterruptible(struct rq *rq_src) { struct rq *rq_dest = cpu_rq(cpumask_any(cpu_active_mask)); - unsigned long flags; - local_irq_save(flags); - double_rq_lock(rq_src, rq_dest); rq_dest->nr_uninterruptible += rq_src->nr_uninterruptible; rq_src->nr_uninterruptible = 0; - double_rq_unlock(rq_src, rq_dest); - local_irq_restore(flags); -} - -/* Run through task list and migrate tasks from the dead cpu. */ -static void migrate_live_tasks(int src_cpu) -{ - struct task_struct *p, *t; - - read_lock(&tasklist_lock); - - do_each_thread(t, p) { - if (p == current) - continue; - - if (task_cpu(p) == src_cpu) - move_task_off_dead_cpu(src_cpu, p); - } while_each_thread(t, p); - - read_unlock(&tasklist_lock); } /* - * Schedules idle task to be the next runnable task on current CPU. - * It does so by boosting its priority to highest possible. - * Used by CPU offline code. + * remove the tasks which were accounted by rq from calc_load_tasks. */ -void sched_idle_next(void) +static void calc_global_load_remove(struct rq *rq) { - int this_cpu = smp_processor_id(); - struct rq *rq = cpu_rq(this_cpu); - struct task_struct *p = rq->idle; - unsigned long flags; - - /* cpu has to be offline */ - BUG_ON(cpu_online(this_cpu)); - - /* - * Strictly not necessary since rest of the CPUs are stopped by now - * and interrupts disabled on the current cpu. - */ - raw_spin_lock_irqsave(&rq->lock, flags); - - __setscheduler(rq, p, SCHED_FIFO, MAX_RT_PRIO-1); - - activate_task(rq, p, 0); - - raw_spin_unlock_irqrestore(&rq->lock, flags); + atomic_long_sub(rq->calc_load_active, &calc_load_tasks); + rq->calc_load_active = 0; } /* - * Ensures that the idle task is using init_mm right before its cpu goes - * offline. + * Migrate all tasks from the rq, sleeping tasks will be migrated by + * try_to_wake_up()->select_task_rq(). + * + * Called with rq->lock held even though we'er in stop_machine() and + * there's no concurrency possible, we hold the required locks anyway + * because of lock validation efforts. */ -void idle_task_exit(void) -{ - struct mm_struct *mm = current->active_mm; - - BUG_ON(cpu_online(smp_processor_id())); - - if (mm != &init_mm) - switch_mm(mm, &init_mm, current); - mmdrop(mm); -} - -/* called under rq->lock with disabled interrupts */ -static void migrate_dead(unsigned int dead_cpu, struct task_struct *p) +static void migrate_tasks(unsigned int dead_cpu) { struct rq *rq = cpu_rq(dead_cpu); - - /* Must be exiting, otherwise would be on tasklist. */ - BUG_ON(!p->exit_state); - - /* Cannot have done final schedule yet: would have vanished. */ - BUG_ON(p->state == TASK_DEAD); - - get_task_struct(p); + struct task_struct *next, *stop = rq->stop; + int dest_cpu; /* - * Drop lock around migration; if someone else moves it, - * that's OK. No task can be added to this CPU, so iteration is - * fine. + * Fudge the rq selection such that the below task selection loop + * doesn't get stuck on the currently eligible stop task. + * + * We're currently inside stop_machine() and the rq is either stuck + * in the stop_machine_cpu_stop() loop, or we're executing this code, + * either way we should never end up calling schedule() until we're + * done here. */ - raw_spin_unlock_irq(&rq->lock); - move_task_off_dead_cpu(dead_cpu, p); - raw_spin_lock_irq(&rq->lock); - - put_task_struct(p); -} - -/* release_task() removes task from tasklist, so we won't find dead tasks. */ -static void migrate_dead_tasks(unsigned int dead_cpu) -{ - struct rq *rq = cpu_rq(dead_cpu); - struct task_struct *next; + rq->stop = NULL; for ( ; ; ) { - if (!rq->nr_running) + /* + * There's this thread running, bail when that's the only + * remaining thread. + */ + if (rq->nr_running == 1) break; + next = pick_next_task(rq); - if (!next) - break; + BUG_ON(!next); next->sched_class->put_prev_task(rq, next); - migrate_dead(dead_cpu, next); + /* Find suitable destination for @next, with force if needed. */ + dest_cpu = select_fallback_rq(dead_cpu, next); + raw_spin_unlock(&rq->lock); + + __migrate_task(next, dead_cpu, dest_cpu); + + raw_spin_lock(&rq->lock); } -} -/* - * 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; + rq->stop = stop; } + #endif /* CONFIG_HOTPLUG_CPU */ #if defined(CONFIG_SCHED_DEBUG) && defined(CONFIG_SYSCTL) @@ -5748,15 +6077,13 @@ migration_call(struct notifier_block *nfb, unsigned long action, void *hcpu) unsigned long flags; struct rq *rq = cpu_rq(cpu); - switch (action) { + switch (action & ~CPU_TASKS_FROZEN) { case CPU_UP_PREPARE: - case CPU_UP_PREPARE_FROZEN: rq->calc_load_update = calc_load_update; break; case CPU_ONLINE: - case CPU_ONLINE_FROZEN: /* Update our root-domain */ raw_spin_lock_irqsave(&rq->lock, flags); if (rq->rd) { @@ -5768,30 +6095,19 @@ migration_call(struct notifier_block *nfb, unsigned long action, void *hcpu) break; #ifdef CONFIG_HOTPLUG_CPU - case CPU_DEAD: - case CPU_DEAD_FROZEN: - migrate_live_tasks(cpu); - /* Idle task back to normal (off runqueue, low prio) */ - raw_spin_lock_irq(&rq->lock); - deactivate_task(rq, rq->idle, 0); - __setscheduler(rq, rq->idle, SCHED_NORMAL, 0); - rq->idle->sched_class = &idle_sched_class; - migrate_dead_tasks(cpu); - raw_spin_unlock_irq(&rq->lock); - migrate_nr_uninterruptible(rq); - BUG_ON(rq->nr_running != 0); - calc_global_load_remove(rq); - break; - case CPU_DYING: - case CPU_DYING_FROZEN: /* Update our root-domain */ raw_spin_lock_irqsave(&rq->lock, flags); if (rq->rd) { BUG_ON(!cpumask_test_cpu(cpu, rq->rd->span)); set_rq_offline(rq); } + migrate_tasks(cpu); + BUG_ON(rq->nr_running != 1); /* the migration thread */ raw_spin_unlock_irqrestore(&rq->lock, flags); + + migrate_nr_uninterruptible(rq); + calc_global_load_remove(rq); break; #endif } @@ -5805,20 +6121,49 @@ migration_call(struct notifier_block *nfb, unsigned long action, void *hcpu) */ static struct notifier_block __cpuinitdata migration_notifier = { .notifier_call = migration_call, - .priority = 10 + .priority = CPU_PRI_MIGRATION, }; +static int __cpuinit sched_cpu_active(struct notifier_block *nfb, + unsigned long action, void *hcpu) +{ + switch (action & ~CPU_TASKS_FROZEN) { + case CPU_ONLINE: + case CPU_DOWN_FAILED: + set_cpu_active((long)hcpu, true); + return NOTIFY_OK; + default: + return NOTIFY_DONE; + } +} + +static int __cpuinit sched_cpu_inactive(struct notifier_block *nfb, + unsigned long action, void *hcpu) +{ + switch (action & ~CPU_TASKS_FROZEN) { + case CPU_DOWN_PREPARE: + set_cpu_active((long)hcpu, false); + return NOTIFY_OK; + default: + return NOTIFY_DONE; + } +} + static int __init migration_init(void) { void *cpu = (void *)(long)smp_processor_id(); int err; - /* Start one for the boot CPU: */ + /* Initialize migration for the boot CPU */ err = migration_call(&migration_notifier, CPU_UP_PREPARE, cpu); BUG_ON(err == NOTIFY_BAD); migration_call(&migration_notifier, CPU_ONLINE, cpu); register_cpu_notifier(&migration_notifier); + /* Register cpu active notifiers */ + cpu_notifier(sched_cpu_active, CPU_PRI_SCHED_ACTIVE); + cpu_notifier(sched_cpu_inactive, CPU_PRI_SCHED_INACTIVE); + return 0; } early_initcall(migration_init); @@ -6053,23 +6398,18 @@ static void rq_attach_root(struct rq *rq, struct root_domain *rd) free_rootdomain(old_rd); } -static int init_rootdomain(struct root_domain *rd, bool bootmem) +static int init_rootdomain(struct root_domain *rd) { - gfp_t gfp = GFP_KERNEL; - memset(rd, 0, sizeof(*rd)); - if (bootmem) - gfp = GFP_NOWAIT; - - if (!alloc_cpumask_var(&rd->span, gfp)) + if (!alloc_cpumask_var(&rd->span, GFP_KERNEL)) goto out; - if (!alloc_cpumask_var(&rd->online, gfp)) + if (!alloc_cpumask_var(&rd->online, GFP_KERNEL)) goto free_span; - if (!alloc_cpumask_var(&rd->rto_mask, gfp)) + if (!alloc_cpumask_var(&rd->rto_mask, GFP_KERNEL)) goto free_online; - if (cpupri_init(&rd->cpupri, bootmem) != 0) + if (cpupri_init(&rd->cpupri) != 0) goto free_rto_mask; return 0; @@ -6085,7 +6425,7 @@ out: static void init_defrootdomain(void) { - init_rootdomain(&def_root_domain, true); + init_rootdomain(&def_root_domain); atomic_set(&def_root_domain.refcount, 1); } @@ -6098,7 +6438,7 @@ static struct root_domain *alloc_rootdomain(void) if (!rd) return NULL; - if (init_rootdomain(rd, false) != 0) { + if (init_rootdomain(rd) != 0) { kfree(rd); return NULL; } @@ -6308,6 +6648,7 @@ struct s_data { cpumask_var_t nodemask; cpumask_var_t this_sibling_map; cpumask_var_t this_core_map; + cpumask_var_t this_book_map; cpumask_var_t send_covered; cpumask_var_t tmpmask; struct sched_group **sched_group_nodes; @@ -6319,6 +6660,7 @@ enum s_alloc { sa_rootdomain, sa_tmpmask, sa_send_covered, + sa_this_book_map, sa_this_core_map, sa_this_sibling_map, sa_nodemask, @@ -6354,31 +6696,48 @@ cpu_to_cpu_group(int cpu, const struct cpumask *cpu_map, #ifdef CONFIG_SCHED_MC static DEFINE_PER_CPU(struct static_sched_domain, core_domains); static DEFINE_PER_CPU(struct static_sched_group, sched_group_core); -#endif /* CONFIG_SCHED_MC */ -#if defined(CONFIG_SCHED_MC) && defined(CONFIG_SCHED_SMT) static int cpu_to_core_group(int cpu, const struct cpumask *cpu_map, struct sched_group **sg, struct cpumask *mask) { int group; - +#ifdef CONFIG_SCHED_SMT cpumask_and(mask, topology_thread_cpumask(cpu), cpu_map); group = cpumask_first(mask); +#else + group = cpu; +#endif if (sg) *sg = &per_cpu(sched_group_core, group).sg; return group; } -#elif defined(CONFIG_SCHED_MC) +#endif /* CONFIG_SCHED_MC */ + +/* + * book sched-domains: + */ +#ifdef CONFIG_SCHED_BOOK +static DEFINE_PER_CPU(struct static_sched_domain, book_domains); +static DEFINE_PER_CPU(struct static_sched_group, sched_group_book); + static int -cpu_to_core_group(int cpu, const struct cpumask *cpu_map, - struct sched_group **sg, struct cpumask *unused) +cpu_to_book_group(int cpu, const struct cpumask *cpu_map, + struct sched_group **sg, struct cpumask *mask) { + int group = cpu; +#ifdef CONFIG_SCHED_MC + cpumask_and(mask, cpu_coregroup_mask(cpu), cpu_map); + group = cpumask_first(mask); +#elif defined(CONFIG_SCHED_SMT) + cpumask_and(mask, topology_thread_cpumask(cpu), cpu_map); + group = cpumask_first(mask); +#endif if (sg) - *sg = &per_cpu(sched_group_core, cpu).sg; - return cpu; + *sg = &per_cpu(sched_group_book, group).sg; + return group; } -#endif +#endif /* CONFIG_SCHED_BOOK */ static DEFINE_PER_CPU(struct static_sched_domain, phys_domains); static DEFINE_PER_CPU(struct static_sched_group, sched_group_phys); @@ -6388,7 +6747,10 @@ cpu_to_phys_group(int cpu, const struct cpumask *cpu_map, struct sched_group **sg, struct cpumask *mask) { int group; -#ifdef CONFIG_SCHED_MC +#ifdef CONFIG_SCHED_BOOK + cpumask_and(mask, cpu_book_mask(cpu), cpu_map); + group = cpumask_first(mask); +#elif defined(CONFIG_SCHED_MC) cpumask_and(mask, cpu_coregroup_mask(cpu), cpu_map); group = cpumask_first(mask); #elif defined(CONFIG_SCHED_SMT) @@ -6584,6 +6946,8 @@ static void init_sched_groups_power(int cpu, struct sched_domain *sd) if (cpu != group_first_cpu(sd->groups)) return; + sd->groups->group_weight = cpumask_weight(sched_group_cpus(sd->groups)); + child = sd->child; sd->groups->cpu_power = 0; @@ -6649,6 +7013,9 @@ SD_INIT_FUNC(CPU) #ifdef CONFIG_SCHED_MC SD_INIT_FUNC(MC) #endif +#ifdef CONFIG_SCHED_BOOK + SD_INIT_FUNC(BOOK) +#endif static int default_relax_domain_level = -1; @@ -6698,6 +7065,8 @@ static void __free_domain_allocs(struct s_data *d, enum s_alloc what, free_cpumask_var(d->tmpmask); /* fall through */ case sa_send_covered: free_cpumask_var(d->send_covered); /* fall through */ + case sa_this_book_map: + free_cpumask_var(d->this_book_map); /* fall through */ case sa_this_core_map: free_cpumask_var(d->this_core_map); /* fall through */ case sa_this_sibling_map: @@ -6744,8 +7113,10 @@ static enum s_alloc __visit_domain_allocation_hell(struct s_data *d, 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)) + if (!alloc_cpumask_var(&d->this_book_map, GFP_KERNEL)) return sa_this_core_map; + if (!alloc_cpumask_var(&d->send_covered, GFP_KERNEL)) + return sa_this_book_map; if (!alloc_cpumask_var(&d->tmpmask, GFP_KERNEL)) return sa_send_covered; d->rd = alloc_rootdomain(); @@ -6803,6 +7174,23 @@ static struct sched_domain *__build_cpu_sched_domain(struct s_data *d, return sd; } +static struct sched_domain *__build_book_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_BOOK + sd = &per_cpu(book_domains, i).sd; + SD_INIT(sd, BOOK); + set_domain_attribute(sd, attr); + cpumask_and(sched_domain_span(sd), cpu_map, cpu_book_mask(i)); + sd->parent = parent; + parent->child = sd; + cpu_to_book_group(i, cpu_map, &sd->groups, d->tmpmask); +#endif + 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) @@ -6860,6 +7248,15 @@ static void build_sched_groups(struct s_data *d, enum sched_domain_level l, d->send_covered, d->tmpmask); break; #endif +#ifdef CONFIG_SCHED_BOOK + case SD_LV_BOOK: /* set up book groups */ + cpumask_and(d->this_book_map, cpu_map, cpu_book_mask(cpu)); + if (cpu == cpumask_first(d->this_book_map)) + init_sched_build_groups(d->this_book_map, cpu_map, + &cpu_to_book_group, + d->send_covered, d->tmpmask); + break; +#endif case SD_LV_CPU: /* set up physical groups */ cpumask_and(d->nodemask, cpumask_of_node(cpu), cpu_map); if (!cpumask_empty(d->nodemask)) @@ -6907,12 +7304,14 @@ static int __build_sched_domains(const struct cpumask *cpu_map, sd = __build_numa_sched_domains(&d, cpu_map, attr, i); sd = __build_cpu_sched_domain(&d, cpu_map, attr, sd, i); + sd = __build_book_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_each_cpu(i, cpu_map) { build_sched_groups(&d, SD_LV_SIBLING, cpu_map, i); + build_sched_groups(&d, SD_LV_BOOK, cpu_map, i); build_sched_groups(&d, SD_LV_MC, cpu_map, i); } @@ -6943,6 +7342,12 @@ static int __build_sched_domains(const struct cpumask *cpu_map, init_sched_groups_power(i, sd); } #endif +#ifdef CONFIG_SCHED_BOOK + for_each_cpu(i, cpu_map) { + sd = &per_cpu(book_domains, i).sd; + init_sched_groups_power(i, sd); + } +#endif for_each_cpu(i, cpu_map) { sd = &per_cpu(phys_domains, i).sd; @@ -6968,6 +7373,8 @@ static int __build_sched_domains(const struct cpumask *cpu_map, sd = &per_cpu(cpu_domains, i).sd; #elif defined(CONFIG_SCHED_MC) sd = &per_cpu(core_domains, i).sd; +#elif defined(CONFIG_SCHED_BOOK) + sd = &per_cpu(book_domains, i).sd; #else sd = &per_cpu(phys_domains, i).sd; #endif @@ -7277,29 +7684,35 @@ int __init sched_create_sysfs_power_savings_entries(struct sysdev_class *cls) } #endif /* CONFIG_SCHED_MC || CONFIG_SCHED_SMT */ -#ifndef CONFIG_CPUSETS /* - * Add online and remove offline CPUs from the scheduler domains. - * When cpusets are enabled they take over this function. + * Update cpusets according to cpu_active mask. If cpusets are + * disabled, cpuset_update_active_cpus() becomes a simple wrapper + * around partition_sched_domains(). */ -static int update_sched_domains(struct notifier_block *nfb, - unsigned long action, void *hcpu) +static int cpuset_cpu_active(struct notifier_block *nfb, unsigned long action, + void *hcpu) { - switch (action) { + switch (action & ~CPU_TASKS_FROZEN) { case CPU_ONLINE: - case CPU_ONLINE_FROZEN: - case CPU_DOWN_PREPARE: - case CPU_DOWN_PREPARE_FROZEN: case CPU_DOWN_FAILED: - case CPU_DOWN_FAILED_FROZEN: - partition_sched_domains(1, NULL, NULL); + cpuset_update_active_cpus(); return NOTIFY_OK; + default: + return NOTIFY_DONE; + } +} +static int cpuset_cpu_inactive(struct notifier_block *nfb, unsigned long action, + void *hcpu) +{ + switch (action & ~CPU_TASKS_FROZEN) { + case CPU_DOWN_PREPARE: + cpuset_update_active_cpus(); + return NOTIFY_OK; default: return NOTIFY_DONE; } } -#endif static int update_runtime(struct notifier_block *nfb, unsigned long action, void *hcpu) @@ -7345,10 +7758,8 @@ void __init sched_init_smp(void) mutex_unlock(&sched_domains_mutex); put_online_cpus(); -#ifndef CONFIG_CPUSETS - /* XXX: Theoretical race here - CPU may be hotplugged now */ - hotcpu_notifier(update_sched_domains, 0); -#endif + hotcpu_notifier(cpuset_cpu_active, CPU_PRI_CPUSET_ACTIVE); + hotcpu_notifier(cpuset_cpu_inactive, CPU_PRI_CPUSET_INACTIVE); /* RT runtime code needs to handle some hotplug events */ hotcpu_notifier(update_runtime, 0); @@ -7427,18 +7838,16 @@ static void init_rt_rq(struct rt_rq *rt_rq, struct rq *rq) #ifdef CONFIG_FAIR_GROUP_SCHED static void init_tg_cfs_entry(struct task_group *tg, struct cfs_rq *cfs_rq, - struct sched_entity *se, int cpu, int add, + struct sched_entity *se, int cpu, struct sched_entity *parent) { struct rq *rq = cpu_rq(cpu); tg->cfs_rq[cpu] = cfs_rq; init_cfs_rq(cfs_rq, rq); cfs_rq->tg = tg; - if (add) - list_add(&cfs_rq->leaf_cfs_rq_list, &rq->leaf_cfs_rq_list); tg->se[cpu] = se; - /* se could be NULL for init_task_group */ + /* se could be NULL for root_task_group */ if (!se) return; @@ -7448,15 +7857,14 @@ static void init_tg_cfs_entry(struct task_group *tg, struct cfs_rq *cfs_rq, se->cfs_rq = parent->my_q; se->my_q = cfs_rq; - se->load.weight = tg->shares; - se->load.inv_weight = 0; + update_load_set(&se->load, 0); se->parent = parent; } #endif #ifdef CONFIG_RT_GROUP_SCHED static void init_tg_rt_entry(struct task_group *tg, struct rt_rq *rt_rq, - struct sched_rt_entity *rt_se, int cpu, int add, + struct sched_rt_entity *rt_se, int cpu, struct sched_rt_entity *parent) { struct rq *rq = cpu_rq(cpu); @@ -7465,8 +7873,6 @@ static void init_tg_rt_entry(struct task_group *tg, struct rt_rq *rt_rq, init_rt_rq(rt_rq, rq); rt_rq->tg = tg; rt_rq->rt_runtime = tg->rt_bandwidth.rt_runtime; - if (add) - list_add(&rt_rq->leaf_rt_rq_list, &rq->leaf_rt_rq_list); tg->rt_se[cpu] = rt_se; if (!rt_se) @@ -7501,18 +7907,18 @@ void __init sched_init(void) ptr = (unsigned long)kzalloc(alloc_size, GFP_NOWAIT); #ifdef CONFIG_FAIR_GROUP_SCHED - init_task_group.se = (struct sched_entity **)ptr; + root_task_group.se = (struct sched_entity **)ptr; ptr += nr_cpu_ids * sizeof(void **); - init_task_group.cfs_rq = (struct cfs_rq **)ptr; + root_task_group.cfs_rq = (struct cfs_rq **)ptr; ptr += nr_cpu_ids * sizeof(void **); #endif /* CONFIG_FAIR_GROUP_SCHED */ #ifdef CONFIG_RT_GROUP_SCHED - init_task_group.rt_se = (struct sched_rt_entity **)ptr; + root_task_group.rt_se = (struct sched_rt_entity **)ptr; ptr += nr_cpu_ids * sizeof(void **); - init_task_group.rt_rq = (struct rt_rq **)ptr; + root_task_group.rt_rq = (struct rt_rq **)ptr; ptr += nr_cpu_ids * sizeof(void **); #endif /* CONFIG_RT_GROUP_SCHED */ @@ -7532,20 +7938,16 @@ void __init sched_init(void) global_rt_period(), global_rt_runtime()); #ifdef CONFIG_RT_GROUP_SCHED - init_rt_bandwidth(&init_task_group.rt_bandwidth, + init_rt_bandwidth(&root_task_group.rt_bandwidth, global_rt_period(), global_rt_runtime()); #endif /* CONFIG_RT_GROUP_SCHED */ #ifdef CONFIG_CGROUP_SCHED - list_add(&init_task_group.list, &task_groups); - INIT_LIST_HEAD(&init_task_group.children); - + list_add(&root_task_group.list, &task_groups); + INIT_LIST_HEAD(&root_task_group.children); + autogroup_init(&init_task); #endif /* CONFIG_CGROUP_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; @@ -7557,45 +7959,45 @@ void __init sched_init(void) init_cfs_rq(&rq->cfs, rq); init_rt_rq(&rq->rt, rq); #ifdef CONFIG_FAIR_GROUP_SCHED - init_task_group.shares = init_task_group_load; + root_task_group.shares = root_task_group_load; INIT_LIST_HEAD(&rq->leaf_cfs_rq_list); -#ifdef CONFIG_CGROUP_SCHED /* - * How much cpu bandwidth does init_task_group get? + * How much cpu bandwidth does root_task_group get? * * In case of task-groups formed thr' the cgroup filesystem, it * gets 100% of the cpu resources in the system. This overall * system cpu resource is divided among the tasks of - * init_task_group and its child task-groups in a fair manner, + * root_task_group and its child task-groups in a fair manner, * based on each entity's (task or task-group's) weight * (se->load.weight). * - * In other words, if init_task_group has 10 tasks of weight + * In other words, if root_task_group has 10 tasks of weight * 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% * - * We achieve this by letting init_task_group's tasks sit - * directly in rq->cfs (i.e init_task_group->se[] = NULL). + * We achieve this by letting root_task_group's tasks sit + * directly in rq->cfs (i.e root_task_group->se[] = NULL). */ - init_tg_cfs_entry(&init_task_group, &rq->cfs, NULL, i, 1, NULL); -#endif + init_tg_cfs_entry(&root_task_group, &rq->cfs, NULL, i, NULL); #endif /* CONFIG_FAIR_GROUP_SCHED */ rq->rt.rt_runtime = def_rt_bandwidth.rt_runtime; #ifdef CONFIG_RT_GROUP_SCHED INIT_LIST_HEAD(&rq->leaf_rt_rq_list); -#ifdef CONFIG_CGROUP_SCHED - init_tg_rt_entry(&init_task_group, &rq->rt, NULL, i, 1, NULL); -#endif + init_tg_rt_entry(&root_task_group, &rq->rt, NULL, i, NULL); #endif for (j = 0; j < CPU_LOAD_IDX_MAX; j++) rq->cpu_load[j] = 0; + + rq->last_load_update_tick = jiffies; + #ifdef CONFIG_SMP rq->sd = NULL; rq->rd = NULL; + rq->cpu_power = SCHED_LOAD_SCALE; rq->post_schedule = 0; rq->active_balance = 0; rq->next_balance = jiffies; @@ -7605,6 +8007,10 @@ void __init sched_init(void) rq->idle_stamp = 0; rq->avg_idle = 2*sysctl_sched_migration_cost; rq_attach_root(rq, &def_root_domain); +#ifdef CONFIG_NO_HZ + rq->nohz_balance_kick = 0; + init_sched_softirq_csd(&per_cpu(remote_sched_softirq_cb, i)); +#endif #endif init_rq_hrtick(rq); atomic_set(&rq->nr_iowait, 0); @@ -7649,16 +8055,17 @@ void __init sched_init(void) zalloc_cpumask_var(&nohz_cpu_mask, GFP_NOWAIT); #ifdef CONFIG_SMP #ifdef CONFIG_NO_HZ - zalloc_cpumask_var(&nohz.cpu_mask, GFP_NOWAIT); - alloc_cpumask_var(&nohz.ilb_grp_nohz_mask, GFP_NOWAIT); + zalloc_cpumask_var(&nohz.idle_cpus_mask, GFP_NOWAIT); + alloc_cpumask_var(&nohz.grp_idle_mask, GFP_NOWAIT); + atomic_set(&nohz.load_balancer, nr_cpu_ids); + atomic_set(&nohz.first_pick_cpu, nr_cpu_ids); + atomic_set(&nohz.second_pick_cpu, nr_cpu_ids); #endif /* 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; } @@ -7852,26 +8259,32 @@ int alloc_fair_sched_group(struct task_group *tg, struct task_group *parent) if (!se) goto err_free_rq; - init_tg_cfs_entry(tg, cfs_rq, se, i, 0, parent->se[i]); + init_tg_cfs_entry(tg, cfs_rq, se, i, parent->se[i]); } return 1; - err_free_rq: +err_free_rq: kfree(cfs_rq); - err: +err: return 0; } -static inline void register_fair_sched_group(struct task_group *tg, int cpu) -{ - list_add_rcu(&tg->cfs_rq[cpu]->leaf_cfs_rq_list, - &cpu_rq(cpu)->leaf_cfs_rq_list); -} - static inline void unregister_fair_sched_group(struct task_group *tg, int cpu) { - list_del_rcu(&tg->cfs_rq[cpu]->leaf_cfs_rq_list); + struct rq *rq = cpu_rq(cpu); + unsigned long flags; + + /* + * Only empty task groups can be destroyed; so we can speculatively + * check on_list without danger of it being re-added. + */ + if (!tg->cfs_rq[cpu]->on_list) + return; + + raw_spin_lock_irqsave(&rq->lock, flags); + list_del_leaf_cfs_rq(tg->cfs_rq[cpu]); + raw_spin_unlock_irqrestore(&rq->lock, flags); } #else /* !CONFG_FAIR_GROUP_SCHED */ static inline void free_fair_sched_group(struct task_group *tg) @@ -7884,10 +8297,6 @@ int alloc_fair_sched_group(struct task_group *tg, struct task_group *parent) return 1; } -static inline void register_fair_sched_group(struct task_group *tg, int cpu) -{ -} - static inline void unregister_fair_sched_group(struct task_group *tg, int cpu) { } @@ -7942,27 +8351,16 @@ int alloc_rt_sched_group(struct task_group *tg, struct task_group *parent) if (!rt_se) goto err_free_rq; - init_tg_rt_entry(tg, rt_rq, rt_se, i, 0, parent->rt_se[i]); + init_tg_rt_entry(tg, rt_rq, rt_se, i, parent->rt_se[i]); } return 1; - err_free_rq: +err_free_rq: kfree(rt_rq); - err: +err: return 0; } - -static inline void register_rt_sched_group(struct task_group *tg, int cpu) -{ - list_add_rcu(&tg->rt_rq[cpu]->leaf_rt_rq_list, - &cpu_rq(cpu)->leaf_rt_rq_list); -} - -static inline void unregister_rt_sched_group(struct task_group *tg, int cpu) -{ - list_del_rcu(&tg->rt_rq[cpu]->leaf_rt_rq_list); -} #else /* !CONFIG_RT_GROUP_SCHED */ static inline void free_rt_sched_group(struct task_group *tg) { @@ -7973,14 +8371,6 @@ int alloc_rt_sched_group(struct task_group *tg, struct task_group *parent) { return 1; } - -static inline void register_rt_sched_group(struct task_group *tg, int cpu) -{ -} - -static inline void unregister_rt_sched_group(struct task_group *tg, int cpu) -{ -} #endif /* CONFIG_RT_GROUP_SCHED */ #ifdef CONFIG_CGROUP_SCHED @@ -7988,6 +8378,7 @@ static void free_sched_group(struct task_group *tg) { free_fair_sched_group(tg); free_rt_sched_group(tg); + autogroup_free(tg); kfree(tg); } @@ -7996,7 +8387,6 @@ struct task_group *sched_create_group(struct task_group *parent) { struct task_group *tg; unsigned long flags; - int i; tg = kzalloc(sizeof(*tg), GFP_KERNEL); if (!tg) @@ -8009,10 +8399,6 @@ struct task_group *sched_create_group(struct task_group *parent) goto err; spin_lock_irqsave(&task_group_lock, flags); - for_each_possible_cpu(i) { - register_fair_sched_group(tg, i); - register_rt_sched_group(tg, i); - } list_add_rcu(&tg->list, &task_groups); WARN_ON(!parent); /* root should already exist */ @@ -8042,11 +8428,11 @@ void sched_destroy_group(struct task_group *tg) unsigned long flags; int i; - spin_lock_irqsave(&task_group_lock, flags); - for_each_possible_cpu(i) { + /* end participation in shares distribution */ + for_each_possible_cpu(i) unregister_fair_sched_group(tg, i); - unregister_rt_sched_group(tg, i); - } + + spin_lock_irqsave(&task_group_lock, flags); list_del_rcu(&tg->list); list_del_rcu(&tg->siblings); spin_unlock_irqrestore(&task_group_lock, flags); @@ -8076,12 +8462,12 @@ void sched_move_task(struct task_struct *tsk) if (unlikely(running)) tsk->sched_class->put_prev_task(rq, tsk); - set_task_rq(tsk, task_cpu(tsk)); - #ifdef CONFIG_FAIR_GROUP_SCHED - if (tsk->sched_class->moved_group) - tsk->sched_class->moved_group(tsk, on_rq); + if (tsk->sched_class->task_move_group) + tsk->sched_class->task_move_group(tsk, on_rq); + else #endif + set_task_rq(tsk, task_cpu(tsk)); if (unlikely(running)) tsk->sched_class->set_curr_task(rq); @@ -8093,33 +8479,6 @@ void sched_move_task(struct task_struct *tsk) #endif /* CONFIG_CGROUP_SCHED */ #ifdef CONFIG_FAIR_GROUP_SCHED -static void __set_se_shares(struct sched_entity *se, unsigned long shares) -{ - struct cfs_rq *cfs_rq = se->cfs_rq; - int on_rq; - - on_rq = se->on_rq; - if (on_rq) - dequeue_entity(cfs_rq, se, 0); - - se->load.weight = shares; - se->load.inv_weight = 0; - - if (on_rq) - enqueue_entity(cfs_rq, se, 0); -} - -static void set_se_shares(struct sched_entity *se, unsigned long shares) -{ - struct cfs_rq *cfs_rq = se->cfs_rq; - struct rq *rq = cfs_rq->rq; - unsigned long flags; - - raw_spin_lock_irqsave(&rq->lock, flags); - __set_se_shares(se, shares); - raw_spin_unlock_irqrestore(&rq->lock, flags); -} - static DEFINE_MUTEX(shares_mutex); int sched_group_set_shares(struct task_group *tg, unsigned long shares) @@ -8142,37 +8501,19 @@ int sched_group_set_shares(struct task_group *tg, unsigned long shares) if (tg->shares == shares) goto done; - spin_lock_irqsave(&task_group_lock, flags); - for_each_possible_cpu(i) - unregister_fair_sched_group(tg, i); - list_del_rcu(&tg->siblings); - spin_unlock_irqrestore(&task_group_lock, flags); - - /* wait for any ongoing reference to this group to finish */ - synchronize_sched(); - - /* - * Now we are free to modify the group's share on each cpu - * w/o tripping rebalance_share or load_balance_fair. - */ tg->shares = shares; for_each_possible_cpu(i) { - /* - * force a rebalance - */ - cfs_rq_set_shares(tg->cfs_rq[i], 0); - set_se_shares(tg->se[i], shares); + struct rq *rq = cpu_rq(i); + struct sched_entity *se; + + se = tg->se[i]; + /* Propagate contribution to hierarchy */ + raw_spin_lock_irqsave(&rq->lock, flags); + for_each_sched_entity(se) + update_cfs_shares(group_cfs_rq(se), 0); + raw_spin_unlock_irqrestore(&rq->lock, flags); } - /* - * Enable load balance activity on this group, by inserting it back on - * each cpu's rq->leaf_cfs_rq_list. - */ - spin_lock_irqsave(&task_group_lock, flags); - for_each_possible_cpu(i) - register_fair_sched_group(tg, i); - list_add_rcu(&tg->siblings, &tg->parent->children); - spin_unlock_irqrestore(&task_group_lock, flags); done: mutex_unlock(&shares_mutex); return 0; @@ -8307,7 +8648,7 @@ static int tg_set_bandwidth(struct task_group *tg, raw_spin_unlock(&rt_rq->rt_runtime_lock); } raw_spin_unlock_irq(&tg->rt_bandwidth.rt_runtime_lock); - unlock: +unlock: read_unlock(&tasklist_lock); mutex_unlock(&rt_constraints_mutex); @@ -8471,7 +8812,7 @@ cpu_cgroup_create(struct cgroup_subsys *ss, struct cgroup *cgrp) if (!cgrp->parent) { /* This is early initialization for the top cgroup */ - return &init_task_group.css; + return &root_task_group.css; } parent = cgroup_tg(cgrp->parent); @@ -8542,6 +8883,20 @@ cpu_cgroup_attach(struct cgroup_subsys *ss, struct cgroup *cgrp, } } +static void +cpu_cgroup_exit(struct cgroup_subsys *ss, struct task_struct *task) +{ + /* + * cgroup_exit() is called in the copy_process() failure path. + * Ignore this case since the task hasn't ran yet, this avoids + * trying to poke a half freed task state from generic code. + */ + if (!(task->flags & PF_EXITING)) + return; + + sched_move_task(task); +} + #ifdef CONFIG_FAIR_GROUP_SCHED static int cpu_shares_write_u64(struct cgroup *cgrp, struct cftype *cftype, u64 shareval) @@ -8614,6 +8969,7 @@ struct cgroup_subsys cpu_cgroup_subsys = { .destroy = cpu_cgroup_destroy, .can_attach = cpu_cgroup_can_attach, .attach = cpu_cgroup_attach, + .exit = cpu_cgroup_exit, .populate = cpu_cgroup_populate, .subsys_id = cpu_cgroup_subsys_id, .early_init = 1, @@ -8898,72 +9254,3 @@ struct cgroup_subsys cpuacct_subsys = { }; #endif /* CONFIG_CGROUP_CPUACCT */ -#ifndef CONFIG_SMP - -void synchronize_sched_expedited(void) -{ - barrier(); -} -EXPORT_SYMBOL_GPL(synchronize_sched_expedited); - -#else /* #ifndef CONFIG_SMP */ - -static atomic_t synchronize_sched_expedited_count = ATOMIC_INIT(0); - -static int synchronize_sched_expedited_cpu_stop(void *data) -{ - /* - * There must be a full memory barrier on each affected CPU - * between the time that try_stop_cpus() is called and the - * time that it returns. - * - * In the current initial implementation of cpu_stop, the - * above condition is already met when the control reaches - * this point and the following smp_mb() is not strictly - * necessary. Do smp_mb() anyway for documentation and - * robustness against future implementation changes. - */ - smp_mb(); /* See above comment block. */ - return 0; -} - -/* - * 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 snap, trycount = 0; - - smp_mb(); /* ensure prior mod happens before capturing snap. */ - snap = atomic_read(&synchronize_sched_expedited_count) + 1; - get_online_cpus(); - while (try_stop_cpus(cpu_online_mask, - synchronize_sched_expedited_cpu_stop, - NULL) == -EAGAIN) { - put_online_cpus(); - if (trycount++ < 10) - udelay(trycount * num_online_cpus()); - else { - synchronize_sched(); - return; - } - if (atomic_read(&synchronize_sched_expedited_count) - snap > 0) { - smp_mb(); /* ensure test happens before caller kfree */ - return; - } - get_online_cpus(); - } - atomic_inc(&synchronize_sched_expedited_count); - smp_mb__after_atomic_inc(); /* ensure post-GP actions seen after GP. */ - put_online_cpus(); -} -EXPORT_SYMBOL_GPL(synchronize_sched_expedited); - -#endif /* #else #ifndef CONFIG_SMP */ diff --git a/kernel/sched_autogroup.c b/kernel/sched_autogroup.c new file mode 100644 index 00000000000..9fb65628315 --- /dev/null +++ b/kernel/sched_autogroup.c @@ -0,0 +1,270 @@ +#ifdef CONFIG_SCHED_AUTOGROUP + +#include <linux/proc_fs.h> +#include <linux/seq_file.h> +#include <linux/kallsyms.h> +#include <linux/utsname.h> + +unsigned int __read_mostly sysctl_sched_autogroup_enabled = 1; +static struct autogroup autogroup_default; +static atomic_t autogroup_seq_nr; + +static void __init autogroup_init(struct task_struct *init_task) +{ + autogroup_default.tg = &root_task_group; + root_task_group.autogroup = &autogroup_default; + kref_init(&autogroup_default.kref); + init_rwsem(&autogroup_default.lock); + init_task->signal->autogroup = &autogroup_default; +} + +static inline void autogroup_free(struct task_group *tg) +{ + kfree(tg->autogroup); +} + +static inline void autogroup_destroy(struct kref *kref) +{ + struct autogroup *ag = container_of(kref, struct autogroup, kref); + +#ifdef CONFIG_RT_GROUP_SCHED + /* We've redirected RT tasks to the root task group... */ + ag->tg->rt_se = NULL; + ag->tg->rt_rq = NULL; +#endif + sched_destroy_group(ag->tg); +} + +static inline void autogroup_kref_put(struct autogroup *ag) +{ + kref_put(&ag->kref, autogroup_destroy); +} + +static inline struct autogroup *autogroup_kref_get(struct autogroup *ag) +{ + kref_get(&ag->kref); + return ag; +} + +static inline struct autogroup *autogroup_task_get(struct task_struct *p) +{ + struct autogroup *ag; + unsigned long flags; + + if (!lock_task_sighand(p, &flags)) + return autogroup_kref_get(&autogroup_default); + + ag = autogroup_kref_get(p->signal->autogroup); + unlock_task_sighand(p, &flags); + + return ag; +} + +#ifdef CONFIG_RT_GROUP_SCHED +static void free_rt_sched_group(struct task_group *tg); +#endif + +static inline struct autogroup *autogroup_create(void) +{ + struct autogroup *ag = kzalloc(sizeof(*ag), GFP_KERNEL); + struct task_group *tg; + + if (!ag) + goto out_fail; + + tg = sched_create_group(&root_task_group); + + if (IS_ERR(tg)) + goto out_free; + + kref_init(&ag->kref); + init_rwsem(&ag->lock); + ag->id = atomic_inc_return(&autogroup_seq_nr); + ag->tg = tg; +#ifdef CONFIG_RT_GROUP_SCHED + /* + * Autogroup RT tasks are redirected to the root task group + * so we don't have to move tasks around upon policy change, + * or flail around trying to allocate bandwidth on the fly. + * A bandwidth exception in __sched_setscheduler() allows + * the policy change to proceed. Thereafter, task_group() + * returns &root_task_group, so zero bandwidth is required. + */ + free_rt_sched_group(tg); + tg->rt_se = root_task_group.rt_se; + tg->rt_rq = root_task_group.rt_rq; +#endif + tg->autogroup = ag; + + return ag; + +out_free: + kfree(ag); +out_fail: + if (printk_ratelimit()) { + printk(KERN_WARNING "autogroup_create: %s failure.\n", + ag ? "sched_create_group()" : "kmalloc()"); + } + + return autogroup_kref_get(&autogroup_default); +} + +static inline bool +task_wants_autogroup(struct task_struct *p, struct task_group *tg) +{ + if (tg != &root_task_group) + return false; + + if (p->sched_class != &fair_sched_class) + return false; + + /* + * We can only assume the task group can't go away on us if + * autogroup_move_group() can see us on ->thread_group list. + */ + if (p->flags & PF_EXITING) + return false; + + return true; +} + +static inline bool task_group_is_autogroup(struct task_group *tg) +{ + return tg != &root_task_group && tg->autogroup; +} + +static inline struct task_group * +autogroup_task_group(struct task_struct *p, struct task_group *tg) +{ + int enabled = ACCESS_ONCE(sysctl_sched_autogroup_enabled); + + if (enabled && task_wants_autogroup(p, tg)) + return p->signal->autogroup->tg; + + return tg; +} + +static void +autogroup_move_group(struct task_struct *p, struct autogroup *ag) +{ + struct autogroup *prev; + struct task_struct *t; + unsigned long flags; + + BUG_ON(!lock_task_sighand(p, &flags)); + + prev = p->signal->autogroup; + if (prev == ag) { + unlock_task_sighand(p, &flags); + return; + } + + p->signal->autogroup = autogroup_kref_get(ag); + + t = p; + do { + sched_move_task(t); + } while_each_thread(p, t); + + unlock_task_sighand(p, &flags); + autogroup_kref_put(prev); +} + +/* Allocates GFP_KERNEL, cannot be called under any spinlock */ +void sched_autogroup_create_attach(struct task_struct *p) +{ + struct autogroup *ag = autogroup_create(); + + autogroup_move_group(p, ag); + /* drop extra refrence added by autogroup_create() */ + autogroup_kref_put(ag); +} +EXPORT_SYMBOL(sched_autogroup_create_attach); + +/* Cannot be called under siglock. Currently has no users */ +void sched_autogroup_detach(struct task_struct *p) +{ + autogroup_move_group(p, &autogroup_default); +} +EXPORT_SYMBOL(sched_autogroup_detach); + +void sched_autogroup_fork(struct signal_struct *sig) +{ + sig->autogroup = autogroup_task_get(current); +} + +void sched_autogroup_exit(struct signal_struct *sig) +{ + autogroup_kref_put(sig->autogroup); +} + +static int __init setup_autogroup(char *str) +{ + sysctl_sched_autogroup_enabled = 0; + + return 1; +} + +__setup("noautogroup", setup_autogroup); + +#ifdef CONFIG_PROC_FS + +int proc_sched_autogroup_set_nice(struct task_struct *p, int *nice) +{ + static unsigned long next = INITIAL_JIFFIES; + struct autogroup *ag; + int err; + + if (*nice < -20 || *nice > 19) + return -EINVAL; + + err = security_task_setnice(current, *nice); + if (err) + return err; + + if (*nice < 0 && !can_nice(current, *nice)) + return -EPERM; + + /* this is a heavy operation taking global locks.. */ + if (!capable(CAP_SYS_ADMIN) && time_before(jiffies, next)) + return -EAGAIN; + + next = HZ / 10 + jiffies; + ag = autogroup_task_get(p); + + down_write(&ag->lock); + err = sched_group_set_shares(ag->tg, prio_to_weight[*nice + 20]); + if (!err) + ag->nice = *nice; + up_write(&ag->lock); + + autogroup_kref_put(ag); + + return err; +} + +void proc_sched_autogroup_show_task(struct task_struct *p, struct seq_file *m) +{ + struct autogroup *ag = autogroup_task_get(p); + + down_read(&ag->lock); + seq_printf(m, "/autogroup-%ld nice %d\n", ag->id, ag->nice); + up_read(&ag->lock); + + autogroup_kref_put(ag); +} +#endif /* CONFIG_PROC_FS */ + +#ifdef CONFIG_SCHED_DEBUG +static inline int autogroup_path(struct task_group *tg, char *buf, int buflen) +{ + int enabled = ACCESS_ONCE(sysctl_sched_autogroup_enabled); + + if (!enabled || !tg->autogroup) + return 0; + + return snprintf(buf, buflen, "%s-%ld", "/autogroup", tg->autogroup->id); +} +#endif /* CONFIG_SCHED_DEBUG */ + +#endif /* CONFIG_SCHED_AUTOGROUP */ diff --git a/kernel/sched_autogroup.h b/kernel/sched_autogroup.h new file mode 100644 index 00000000000..7b859ffe5da --- /dev/null +++ b/kernel/sched_autogroup.h @@ -0,0 +1,36 @@ +#ifdef CONFIG_SCHED_AUTOGROUP + +struct autogroup { + struct kref kref; + struct task_group *tg; + struct rw_semaphore lock; + unsigned long id; + int nice; +}; + +static inline struct task_group * +autogroup_task_group(struct task_struct *p, struct task_group *tg); + +#else /* !CONFIG_SCHED_AUTOGROUP */ + +static inline void autogroup_init(struct task_struct *init_task) { } +static inline void autogroup_free(struct task_group *tg) { } +static inline bool task_group_is_autogroup(struct task_group *tg) +{ + return 0; +} + +static inline struct task_group * +autogroup_task_group(struct task_struct *p, struct task_group *tg) +{ + return tg; +} + +#ifdef CONFIG_SCHED_DEBUG +static inline int autogroup_path(struct task_group *tg, char *buf, int buflen) +{ + return 0; +} +#endif + +#endif /* CONFIG_SCHED_AUTOGROUP */ diff --git a/kernel/sched_clock.c b/kernel/sched_clock.c index 906a0f718cb..9d8af0b3fb6 100644 --- a/kernel/sched_clock.c +++ b/kernel/sched_clock.c @@ -10,19 +10,55 @@ * Ingo Molnar <mingo@redhat.com> * Guillaume Chazarain <guichaz@gmail.com> * - * Create a semi stable clock from a mixture of other events, including: - * - gtod + * + * What: + * + * cpu_clock(i) provides a fast (execution time) high resolution + * clock with bounded drift between CPUs. The value of cpu_clock(i) + * is monotonic for constant i. The timestamp returned is in nanoseconds. + * + * ######################### BIG FAT WARNING ########################## + * # when comparing cpu_clock(i) to cpu_clock(j) for i != j, time can # + * # go backwards !! # + * #################################################################### + * + * There is no strict promise about the base, although it tends to start + * at 0 on boot (but people really shouldn't rely on that). + * + * cpu_clock(i) -- can be used from any context, including NMI. + * sched_clock_cpu(i) -- must be used with local IRQs disabled (implied by NMI) + * local_clock() -- is cpu_clock() on the current cpu. + * + * How: + * + * The implementation either uses sched_clock() when + * !CONFIG_HAVE_UNSTABLE_SCHED_CLOCK, which means in that case the + * sched_clock() is assumed to provide these properties (mostly it means + * the architecture provides a globally synchronized highres time source). + * + * Otherwise it tries to create a semi stable clock from a mixture of other + * clocks, including: + * + * - GTOD (clock monotomic) * - sched_clock() * - explicit idle events * - * We use gtod as base and the unstable clock deltas. The deltas are filtered, - * making it monotonic and keeping it within an expected window. + * We use GTOD as base and use sched_clock() deltas to improve resolution. The + * deltas are filtered to provide monotonicity and keeping it within an + * expected window. * * Furthermore, explicit sleep and wakeup hooks allow us to account for time * that is otherwise invisible (TSC gets stopped). * - * The clock: sched_clock_cpu() is monotonic per cpu, and should be somewhat - * consistent between cpus (never more than 2 jiffies difference). + * + * Notes: + * + * The !IRQ-safetly of sched_clock() and sched_clock_cpu() comes from things + * like cpufreq interrupts that can change the base clock (TSC) multiplier + * and cause funny jumps in time -- although the filtering provided by + * sched_clock_cpu() should mitigate serious artifacts we cannot rely on it + * in general since for !CONFIG_HAVE_UNSTABLE_SCHED_CLOCK we fully rely on + * sched_clock(). */ #include <linux/spinlock.h> #include <linux/hardirq.h> @@ -43,7 +79,7 @@ unsigned long long __attribute__((weak)) sched_clock(void) } EXPORT_SYMBOL_GPL(sched_clock); -static __read_mostly int sched_clock_running; +__read_mostly int sched_clock_running; #ifdef CONFIG_HAVE_UNSTABLE_SCHED_CLOCK __read_mostly int sched_clock_stable; @@ -170,6 +206,11 @@ again: return val; } +/* + * Similar to cpu_clock(), but requires local IRQs to be disabled. + * + * See cpu_clock(). + */ u64 sched_clock_cpu(int cpu) { struct sched_clock_data *scd; @@ -237,9 +278,19 @@ void sched_clock_idle_wakeup_event(u64 delta_ns) } EXPORT_SYMBOL_GPL(sched_clock_idle_wakeup_event); -unsigned long long cpu_clock(int cpu) +/* + * As outlined at the top, provides a fast, high resolution, nanosecond + * time source that is monotonic per cpu argument and has bounded drift + * between cpus. + * + * ######################### BIG FAT WARNING ########################## + * # when comparing cpu_clock(i) to cpu_clock(j) for i != j, time can # + * # go backwards !! # + * #################################################################### + */ +u64 cpu_clock(int cpu) { - unsigned long long clock; + u64 clock; unsigned long flags; local_irq_save(flags); @@ -249,6 +300,25 @@ unsigned long long cpu_clock(int cpu) return clock; } +/* + * Similar to cpu_clock() for the current cpu. Time will only be observed + * to be monotonic if care is taken to only compare timestampt taken on the + * same CPU. + * + * See cpu_clock(). + */ +u64 local_clock(void) +{ + u64 clock; + unsigned long flags; + + local_irq_save(flags); + clock = sched_clock_cpu(smp_processor_id()); + local_irq_restore(flags); + + return clock; +} + #else /* CONFIG_HAVE_UNSTABLE_SCHED_CLOCK */ void sched_clock_init(void) @@ -264,12 +334,17 @@ u64 sched_clock_cpu(int cpu) return sched_clock(); } - -unsigned long long cpu_clock(int cpu) +u64 cpu_clock(int cpu) { return sched_clock_cpu(cpu); } +u64 local_clock(void) +{ + return sched_clock_cpu(0); +} + #endif /* CONFIG_HAVE_UNSTABLE_SCHED_CLOCK */ EXPORT_SYMBOL_GPL(cpu_clock); +EXPORT_SYMBOL_GPL(local_clock); diff --git a/kernel/sched_cpupri.c b/kernel/sched_cpupri.c index e6871cb3fc8..2722dc1b413 100644 --- a/kernel/sched_cpupri.c +++ b/kernel/sched_cpupri.c @@ -166,14 +166,10 @@ void cpupri_set(struct cpupri *cp, int cpu, int newpri) * * Returns: -ENOMEM if memory fails. */ -int cpupri_init(struct cpupri *cp, bool bootmem) +int cpupri_init(struct cpupri *cp) { - 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++) { @@ -181,7 +177,7 @@ int cpupri_init(struct cpupri *cp, bool bootmem) raw_spin_lock_init(&vec->lock); vec->count = 0; - if (!zalloc_cpumask_var(&vec->mask, gfp)) + if (!zalloc_cpumask_var(&vec->mask, GFP_KERNEL)) goto cleanup; } diff --git a/kernel/sched_cpupri.h b/kernel/sched_cpupri.h index 7cb5bb6b95b..9fc7d386fea 100644 --- a/kernel/sched_cpupri.h +++ b/kernel/sched_cpupri.h @@ -27,7 +27,7 @@ struct cpupri { int cpupri_find(struct cpupri *cp, struct task_struct *p, struct cpumask *lowest_mask); void cpupri_set(struct cpupri *cp, int cpu, int pri); -int cpupri_init(struct cpupri *cp, bool bootmem); +int cpupri_init(struct cpupri *cp); void cpupri_cleanup(struct cpupri *cp); #else #define cpupri_set(cp, cpu, pri) do { } while (0) diff --git a/kernel/sched_debug.c b/kernel/sched_debug.c index 87a330a7185..eb6cb8edd07 100644 --- a/kernel/sched_debug.c +++ b/kernel/sched_debug.c @@ -16,6 +16,8 @@ #include <linux/kallsyms.h> #include <linux/utsname.h> +static DEFINE_SPINLOCK(sched_debug_lock); + /* * This allows printing both to /proc/sched_debug and * to the console @@ -54,8 +56,7 @@ static unsigned long nsec_low(unsigned long long nsec) #define SPLIT_NS(x) nsec_high(x), nsec_low(x) #ifdef CONFIG_FAIR_GROUP_SCHED -static void print_cfs_group_stats(struct seq_file *m, int cpu, - struct task_group *tg) +static void print_cfs_group_stats(struct seq_file *m, int cpu, struct task_group *tg) { struct sched_entity *se = tg->se[cpu]; if (!se) @@ -87,6 +88,26 @@ static void print_cfs_group_stats(struct seq_file *m, int cpu, } #endif +#ifdef CONFIG_CGROUP_SCHED +static char group_path[PATH_MAX]; + +static char *task_group_path(struct task_group *tg) +{ + if (autogroup_path(tg, group_path, PATH_MAX)) + return group_path; + + /* + * May be NULL if the underlying cgroup isn't fully-created yet + */ + if (!tg->css.cgroup) { + group_path[0] = '\0'; + return group_path; + } + cgroup_path(tg->css.cgroup, group_path, PATH_MAX); + return group_path; +} +#endif + static void print_task(struct seq_file *m, struct rq *rq, struct task_struct *p) { @@ -109,17 +130,10 @@ print_task(struct seq_file *m, struct rq *rq, struct task_struct *p) SEQ_printf(m, "%15Ld %15Ld %15Ld.%06ld %15Ld.%06ld %15Ld.%06ld", 0LL, 0LL, 0LL, 0L, 0LL, 0L, 0LL, 0L); #endif - #ifdef CONFIG_CGROUP_SCHED - { - char path[64]; - - rcu_read_lock(); - cgroup_path(task_group(p)->css.cgroup, path, sizeof(path)); - rcu_read_unlock(); - SEQ_printf(m, " %s", path); - } + SEQ_printf(m, " %s", task_group_path(task_group(p))); #endif + SEQ_printf(m, "\n"); } @@ -147,19 +161,6 @@ static void print_rq(struct seq_file *m, struct rq *rq, int rq_cpu) read_unlock_irqrestore(&tasklist_lock, flags); } -#if defined(CONFIG_CGROUP_SCHED) && \ - (defined(CONFIG_FAIR_GROUP_SCHED) || defined(CONFIG_RT_GROUP_SCHED)) -static void task_group_path(struct task_group *tg, char *buf, int buflen) -{ - /* may be NULL if the underlying cgroup isn't fully-created yet */ - if (!tg->css.cgroup) { - buf[0] = '\0'; - return; - } - cgroup_path(tg->css.cgroup, buf, buflen); -} -#endif - void print_cfs_rq(struct seq_file *m, int cpu, struct cfs_rq *cfs_rq) { s64 MIN_vruntime = -1, min_vruntime, max_vruntime = -1, @@ -168,13 +169,8 @@ void print_cfs_rq(struct seq_file *m, int cpu, struct cfs_rq *cfs_rq) struct sched_entity *last; unsigned long flags; -#if defined(CONFIG_CGROUP_SCHED) && defined(CONFIG_FAIR_GROUP_SCHED) - char path[128]; - struct task_group *tg = cfs_rq->tg; - - task_group_path(tg, path, sizeof(path)); - - SEQ_printf(m, "\ncfs_rq[%d]:%s\n", cpu, path); +#ifdef CONFIG_FAIR_GROUP_SCHED + SEQ_printf(m, "\ncfs_rq[%d]:%s\n", cpu, task_group_path(cfs_rq->tg)); #else SEQ_printf(m, "\ncfs_rq[%d]:\n", cpu); #endif @@ -202,33 +198,34 @@ void print_cfs_rq(struct seq_file *m, int cpu, struct cfs_rq *cfs_rq) spread0 = min_vruntime - rq0_min_vruntime; SEQ_printf(m, " .%-30s: %Ld.%06ld\n", "spread0", SPLIT_NS(spread0)); - SEQ_printf(m, " .%-30s: %ld\n", "nr_running", cfs_rq->nr_running); - SEQ_printf(m, " .%-30s: %ld\n", "load", cfs_rq->load.weight); - SEQ_printf(m, " .%-30s: %d\n", "nr_spread_over", cfs_rq->nr_spread_over); + SEQ_printf(m, " .%-30s: %ld\n", "nr_running", cfs_rq->nr_running); + SEQ_printf(m, " .%-30s: %ld\n", "load", cfs_rq->load.weight); #ifdef CONFIG_FAIR_GROUP_SCHED #ifdef CONFIG_SMP - SEQ_printf(m, " .%-30s: %lu\n", "shares", cfs_rq->shares); + SEQ_printf(m, " .%-30s: %Ld.%06ld\n", "load_avg", + SPLIT_NS(cfs_rq->load_avg)); + SEQ_printf(m, " .%-30s: %Ld.%06ld\n", "load_period", + SPLIT_NS(cfs_rq->load_period)); + SEQ_printf(m, " .%-30s: %ld\n", "load_contrib", + cfs_rq->load_contribution); + SEQ_printf(m, " .%-30s: %d\n", "load_tg", + atomic_read(&cfs_rq->tg->load_weight)); #endif + print_cfs_group_stats(m, cpu, cfs_rq->tg); #endif } void print_rt_rq(struct seq_file *m, int cpu, struct rt_rq *rt_rq) { -#if defined(CONFIG_CGROUP_SCHED) && defined(CONFIG_RT_GROUP_SCHED) - char path[128]; - struct task_group *tg = rt_rq->tg; - - task_group_path(tg, path, sizeof(path)); - - SEQ_printf(m, "\nrt_rq[%d]:%s\n", cpu, path); +#ifdef CONFIG_RT_GROUP_SCHED + SEQ_printf(m, "\nrt_rq[%d]:%s\n", cpu, task_group_path(rt_rq->tg)); #else SEQ_printf(m, "\nrt_rq[%d]:\n", cpu); #endif - #define P(x) \ SEQ_printf(m, " .%-30s: %Ld\n", #x, (long long)(rt_rq->x)) #define PN(x) \ @@ -243,9 +240,12 @@ void print_rt_rq(struct seq_file *m, int cpu, struct rt_rq *rt_rq) #undef P } +extern __read_mostly int sched_clock_running; + static void print_cpu(struct seq_file *m, int cpu) { struct rq *rq = cpu_rq(cpu); + unsigned long flags; #ifdef CONFIG_X86 { @@ -296,14 +296,20 @@ static void print_cpu(struct seq_file *m, int cpu) P(ttwu_count); P(ttwu_local); - P(bkl_count); + SEQ_printf(m, " .%-30s: %d\n", "bkl_count", + rq->rq_sched_info.bkl_count); #undef P +#undef P64 #endif + spin_lock_irqsave(&sched_debug_lock, flags); print_cfs_stats(m, cpu); print_rt_stats(m, cpu); + rcu_read_lock(); print_rq(m, rq, cpu); + rcu_read_unlock(); + spin_unlock_irqrestore(&sched_debug_lock, flags); } static const char *sched_tunable_scaling_names[] = { @@ -314,25 +320,46 @@ static const char *sched_tunable_scaling_names[] = { static int sched_debug_show(struct seq_file *m, void *v) { - u64 now = ktime_to_ns(ktime_get()); + u64 ktime, sched_clk, cpu_clk; + unsigned long flags; int cpu; - SEQ_printf(m, "Sched Debug Version: v0.09, %s %.*s\n", + local_irq_save(flags); + ktime = ktime_to_ns(ktime_get()); + sched_clk = sched_clock(); + cpu_clk = local_clock(); + local_irq_restore(flags); + + SEQ_printf(m, "Sched Debug Version: v0.10, %s %.*s\n", init_utsname()->release, (int)strcspn(init_utsname()->version, " "), init_utsname()->version); - SEQ_printf(m, "now at %Lu.%06ld msecs\n", SPLIT_NS(now)); +#define P(x) \ + SEQ_printf(m, "%-40s: %Ld\n", #x, (long long)(x)) +#define PN(x) \ + SEQ_printf(m, "%-40s: %Ld.%06ld\n", #x, SPLIT_NS(x)) + PN(ktime); + PN(sched_clk); + PN(cpu_clk); + P(jiffies); +#ifdef CONFIG_HAVE_UNSTABLE_SCHED_CLOCK + P(sched_clock_stable); +#endif +#undef PN +#undef P + + SEQ_printf(m, "\n"); + SEQ_printf(m, "sysctl_sched\n"); #define P(x) \ SEQ_printf(m, " .%-40s: %Ld\n", #x, (long long)(x)) #define PN(x) \ SEQ_printf(m, " .%-40s: %Ld.%06ld\n", #x, SPLIT_NS(x)) - P(jiffies); PN(sysctl_sched_latency); PN(sysctl_sched_min_granularity); PN(sysctl_sched_wakeup_granularity); - PN(sysctl_sched_child_runs_first); + P(sysctl_sched_child_runs_first); P(sysctl_sched_features); #undef PN #undef P @@ -381,15 +408,9 @@ __initcall(init_sched_debug_procfs); void proc_sched_show_task(struct task_struct *p, struct seq_file *m) { unsigned long nr_switches; - unsigned long flags; - int num_threads = 1; - - if (lock_task_sighand(p, &flags)) { - num_threads = atomic_read(&p->signal->count); - unlock_task_sighand(p, &flags); - } - SEQ_printf(m, "%s (%d, #threads: %d)\n", p->comm, p->pid, num_threads); + SEQ_printf(m, "%s (%d, #threads: %d)\n", p->comm, p->pid, + get_nr_threads(p)); SEQ_printf(m, "---------------------------------------------------------\n"); #define __P(F) \ diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c index 217e4a9393e..354769979c0 100644 --- a/kernel/sched_fair.c +++ b/kernel/sched_fair.c @@ -25,7 +25,7 @@ /* * Targeted preemption latency for CPU-bound tasks: - * (default: 5ms * (1 + ilog(ncpus)), units: nanoseconds) + * (default: 6ms * (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 @@ -52,15 +52,15 @@ enum sched_tunable_scaling sysctl_sched_tunable_scaling /* * Minimal preemption granularity for CPU-bound tasks: - * (default: 2 msec * (1 + ilog(ncpus)), units: nanoseconds) + * (default: 0.75 msec * (1 + ilog(ncpus)), units: nanoseconds) */ -unsigned int sysctl_sched_min_granularity = 2000000ULL; -unsigned int normalized_sysctl_sched_min_granularity = 2000000ULL; +unsigned int sysctl_sched_min_granularity = 750000ULL; +unsigned int normalized_sysctl_sched_min_granularity = 750000ULL; /* * is kept at sysctl_sched_latency / sysctl_sched_min_granularity */ -static unsigned int sched_nr_latency = 3; +static unsigned int sched_nr_latency = 8; /* * After fork, child runs first. If set to 0 (default) then @@ -89,6 +89,13 @@ unsigned int normalized_sysctl_sched_wakeup_granularity = 1000000UL; const_debug unsigned int sysctl_sched_migration_cost = 500000UL; +/* + * The exponential sliding window over which load is averaged for shares + * distribution. + * (default: 10msec) + */ +unsigned int __read_mostly sysctl_sched_shares_window = 10000000UL; + static const struct sched_class fair_sched_class; /************************************************************** @@ -143,6 +150,36 @@ static inline struct cfs_rq *cpu_cfs_rq(struct cfs_rq *cfs_rq, int this_cpu) return cfs_rq->tg->cfs_rq[this_cpu]; } +static inline void list_add_leaf_cfs_rq(struct cfs_rq *cfs_rq) +{ + if (!cfs_rq->on_list) { + /* + * Ensure we either appear before our parent (if already + * enqueued) or force our parent to appear after us when it is + * enqueued. The fact that we always enqueue bottom-up + * reduces this to two cases. + */ + if (cfs_rq->tg->parent && + cfs_rq->tg->parent->cfs_rq[cpu_of(rq_of(cfs_rq))]->on_list) { + list_add_rcu(&cfs_rq->leaf_cfs_rq_list, + &rq_of(cfs_rq)->leaf_cfs_rq_list); + } else { + list_add_tail_rcu(&cfs_rq->leaf_cfs_rq_list, + &rq_of(cfs_rq)->leaf_cfs_rq_list); + } + + cfs_rq->on_list = 1; + } +} + +static inline void list_del_leaf_cfs_rq(struct cfs_rq *cfs_rq) +{ + if (cfs_rq->on_list) { + list_del_rcu(&cfs_rq->leaf_cfs_rq_list); + cfs_rq->on_list = 0; + } +} + /* Iterate thr' all leaf cfs_rq's on a runqueue */ #define for_each_leaf_cfs_rq(rq, cfs_rq) \ list_for_each_entry_rcu(cfs_rq, &rq->leaf_cfs_rq_list, leaf_cfs_rq_list) @@ -246,6 +283,14 @@ static inline struct cfs_rq *cpu_cfs_rq(struct cfs_rq *cfs_rq, int this_cpu) return &cpu_rq(this_cpu)->cfs; } +static inline void list_add_leaf_cfs_rq(struct cfs_rq *cfs_rq) +{ +} + +static inline void list_del_leaf_cfs_rq(struct cfs_rq *cfs_rq) +{ +} + #define for_each_leaf_cfs_rq(rq, cfs_rq) \ for (cfs_rq = &rq->cfs; cfs_rq; cfs_rq = NULL) @@ -417,7 +462,6 @@ int sched_proc_update_handler(struct ctl_table *table, int write, WRT_SYSCTL(sched_min_granularity); WRT_SYSCTL(sched_latency); WRT_SYSCTL(sched_wakeup_granularity); - WRT_SYSCTL(sched_shares_ratelimit); #undef WRT_SYSCTL return 0; @@ -495,6 +539,9 @@ static u64 sched_vslice(struct cfs_rq *cfs_rq, struct sched_entity *se) return calc_delta_fair(sched_slice(cfs_rq, se), se); } +static void update_cfs_load(struct cfs_rq *cfs_rq, int global_update); +static void update_cfs_shares(struct cfs_rq *cfs_rq, long weight_delta); + /* * Update the current task's runtime statistics. Skip current tasks that * are not in our scheduling class. @@ -514,12 +561,16 @@ __update_curr(struct cfs_rq *cfs_rq, struct sched_entity *curr, curr->vruntime += delta_exec_weighted; update_min_vruntime(cfs_rq); + +#if defined CONFIG_SMP && defined CONFIG_FAIR_GROUP_SCHED + cfs_rq->load_unacc_exec_time += delta_exec; +#endif } static void update_curr(struct cfs_rq *cfs_rq) { struct sched_entity *curr = cfs_rq->curr; - u64 now = rq_of(cfs_rq)->clock; + u64 now = rq_of(cfs_rq)->clock_task; unsigned long delta_exec; if (unlikely(!curr)) @@ -602,7 +653,7 @@ update_stats_curr_start(struct cfs_rq *cfs_rq, struct sched_entity *se) /* * We are starting a new run period: */ - se->exec_start = rq_of(cfs_rq)->clock; + se->exec_start = rq_of(cfs_rq)->clock_task; } /************************************************** @@ -633,7 +684,6 @@ account_entity_enqueue(struct cfs_rq *cfs_rq, struct sched_entity *se) list_add(&se->group_node, &cfs_rq->tasks); } cfs_rq->nr_running++; - se->on_rq = 1; } static void @@ -647,9 +697,168 @@ account_entity_dequeue(struct cfs_rq *cfs_rq, struct sched_entity *se) list_del_init(&se->group_node); } cfs_rq->nr_running--; - se->on_rq = 0; } +#ifdef CONFIG_FAIR_GROUP_SCHED +# ifdef CONFIG_SMP +static void update_cfs_rq_load_contribution(struct cfs_rq *cfs_rq, + int global_update) +{ + struct task_group *tg = cfs_rq->tg; + long load_avg; + + load_avg = div64_u64(cfs_rq->load_avg, cfs_rq->load_period+1); + load_avg -= cfs_rq->load_contribution; + + if (global_update || abs(load_avg) > cfs_rq->load_contribution / 8) { + atomic_add(load_avg, &tg->load_weight); + cfs_rq->load_contribution += load_avg; + } +} + +static void update_cfs_load(struct cfs_rq *cfs_rq, int global_update) +{ + u64 period = sysctl_sched_shares_window; + u64 now, delta; + unsigned long load = cfs_rq->load.weight; + + if (!cfs_rq) + return; + + now = rq_of(cfs_rq)->clock; + delta = now - cfs_rq->load_stamp; + + /* truncate load history at 4 idle periods */ + if (cfs_rq->load_stamp > cfs_rq->load_last && + now - cfs_rq->load_last > 4 * period) { + cfs_rq->load_period = 0; + cfs_rq->load_avg = 0; + } + + cfs_rq->load_stamp = now; + cfs_rq->load_unacc_exec_time = 0; + cfs_rq->load_period += delta; + if (load) { + cfs_rq->load_last = now; + cfs_rq->load_avg += delta * load; + } + + /* consider updating load contribution on each fold or truncate */ + if (global_update || cfs_rq->load_period > period + || !cfs_rq->load_period) + update_cfs_rq_load_contribution(cfs_rq, global_update); + + while (cfs_rq->load_period > period) { + /* + * Inline assembly required to prevent the compiler + * optimising this loop into a divmod call. + * See __iter_div_u64_rem() for another example of this. + */ + asm("" : "+rm" (cfs_rq->load_period)); + cfs_rq->load_period /= 2; + cfs_rq->load_avg /= 2; + } + + if (!cfs_rq->curr && !cfs_rq->nr_running && !cfs_rq->load_avg) + list_del_leaf_cfs_rq(cfs_rq); +} + +static long calc_cfs_shares(struct cfs_rq *cfs_rq, struct task_group *tg, + long weight_delta) +{ + long load_weight, load, shares; + + load = cfs_rq->load.weight + weight_delta; + + load_weight = atomic_read(&tg->load_weight); + load_weight -= cfs_rq->load_contribution; + load_weight += load; + + shares = (tg->shares * load); + if (load_weight) + shares /= load_weight; + + if (shares < MIN_SHARES) + shares = MIN_SHARES; + if (shares > tg->shares) + shares = tg->shares; + + return shares; +} + +static void update_entity_shares_tick(struct cfs_rq *cfs_rq) +{ + if (cfs_rq->load_unacc_exec_time > sysctl_sched_shares_window) { + update_cfs_load(cfs_rq, 0); + update_cfs_shares(cfs_rq, 0); + } +} +# else /* CONFIG_SMP */ +static void update_cfs_load(struct cfs_rq *cfs_rq, int global_update) +{ +} + +static inline long calc_cfs_shares(struct cfs_rq *cfs_rq, struct task_group *tg, + long weight_delta) +{ + return tg->shares; +} + +static inline void update_entity_shares_tick(struct cfs_rq *cfs_rq) +{ +} +# endif /* CONFIG_SMP */ +static void reweight_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, + unsigned long weight) +{ + if (se->on_rq) { + /* commit outstanding execution time */ + if (cfs_rq->curr == se) + update_curr(cfs_rq); + account_entity_dequeue(cfs_rq, se); + } + + update_load_set(&se->load, weight); + + if (se->on_rq) + account_entity_enqueue(cfs_rq, se); +} + +static void update_cfs_shares(struct cfs_rq *cfs_rq, long weight_delta) +{ + struct task_group *tg; + struct sched_entity *se; + long shares; + + if (!cfs_rq) + return; + + tg = cfs_rq->tg; + se = tg->se[cpu_of(rq_of(cfs_rq))]; + if (!se) + return; +#ifndef CONFIG_SMP + if (likely(se->load.weight == tg->shares)) + return; +#endif + shares = calc_cfs_shares(cfs_rq, tg, weight_delta); + + reweight_entity(cfs_rq_of(se), se, shares); +} +#else /* CONFIG_FAIR_GROUP_SCHED */ +static void update_cfs_load(struct cfs_rq *cfs_rq, int global_update) +{ +} + +static inline void update_cfs_shares(struct cfs_rq *cfs_rq, long weight_delta) +{ +} + +static inline void update_entity_shares_tick(struct cfs_rq *cfs_rq) +{ +} +#endif /* CONFIG_FAIR_GROUP_SCHED */ + static void enqueue_sleeper(struct cfs_rq *cfs_rq, struct sched_entity *se) { #ifdef CONFIG_SCHEDSTATS @@ -771,6 +980,8 @@ enqueue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) * Update run-time statistics of the 'current'. */ update_curr(cfs_rq); + update_cfs_load(cfs_rq, 0); + update_cfs_shares(cfs_rq, se->load.weight); account_entity_enqueue(cfs_rq, se); if (flags & ENQUEUE_WAKEUP) { @@ -782,6 +993,10 @@ enqueue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) check_spread(cfs_rq, se); if (se != cfs_rq->curr) __enqueue_entity(cfs_rq, se); + se->on_rq = 1; + + if (cfs_rq->nr_running == 1) + list_add_leaf_cfs_rq(cfs_rq); } static void __clear_buddies(struct cfs_rq *cfs_rq, struct sched_entity *se) @@ -825,8 +1040,11 @@ dequeue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) if (se != cfs_rq->curr) __dequeue_entity(cfs_rq, se); + se->on_rq = 0; + update_cfs_load(cfs_rq, 0); account_entity_dequeue(cfs_rq, se); update_min_vruntime(cfs_rq); + update_cfs_shares(cfs_rq, 0); /* * Normalize the entity after updating the min_vruntime because the @@ -872,6 +1090,9 @@ check_preempt_tick(struct cfs_rq *cfs_rq, struct sched_entity *curr) struct sched_entity *se = __pick_next_entity(cfs_rq); s64 delta = curr->vruntime - se->vruntime; + if (delta < 0) + return; + if (delta > ideal_runtime) resched_task(rq_of(cfs_rq)->curr); } @@ -955,6 +1176,11 @@ entity_tick(struct cfs_rq *cfs_rq, struct sched_entity *curr, int queued) */ update_curr(cfs_rq); + /* + * Update share accounting for long-running entities. + */ + update_entity_shares_tick(cfs_rq); + #ifdef CONFIG_SCHED_HRTICK /* * queued ticks are scheduled to match the slice, so don't bother @@ -1055,6 +1281,13 @@ enqueue_task_fair(struct rq *rq, struct task_struct *p, int flags) flags = ENQUEUE_WAKEUP; } + for_each_sched_entity(se) { + struct cfs_rq *cfs_rq = cfs_rq_of(se); + + update_cfs_load(cfs_rq, 0); + update_cfs_shares(cfs_rq, 0); + } + hrtick_update(rq); } @@ -1071,12 +1304,20 @@ static void dequeue_task_fair(struct rq *rq, struct task_struct *p, int flags) for_each_sched_entity(se) { cfs_rq = cfs_rq_of(se); dequeue_entity(cfs_rq, se, flags); + /* Don't dequeue parent if it has other entities besides us */ if (cfs_rq->load.weight) break; flags |= DEQUEUE_SLEEP; } + for_each_sched_entity(se) { + struct cfs_rq *cfs_rq = cfs_rq_of(se); + + update_cfs_load(cfs_rq, 0); + update_cfs_shares(cfs_rq, 0); + } + hrtick_update(rq); } @@ -1143,67 +1384,36 @@ static void task_waking_fair(struct rq *rq, struct task_struct *p) * Adding load to a group doesn't make a group heavier, but can cause movement * of group shares between cpus. Assuming the shares were perfectly aligned one * can calculate the shift in shares. - * - * The problem is that perfectly aligning the shares is rather expensive, hence - * we try to avoid doing that too often - see update_shares(), which ratelimits - * this change. - * - * We compensate this by not only taking the current delta into account, but - * also considering the delta between when the shares were last adjusted and - * now. - * - * We still saw a performance dip, some tracing learned us that between - * cgroup:/ and cgroup:/foo balancing the number of affine wakeups increased - * significantly. Therefore try to bias the error in direction of failing - * the affine wakeup. - * */ -static long effective_load(struct task_group *tg, int cpu, - long wl, long wg) +static long effective_load(struct task_group *tg, int cpu, long wl, long wg) { struct sched_entity *se = tg->se[cpu]; if (!tg->parent) return wl; - /* - * By not taking the decrease of shares on the other cpu into - * account our error leans towards reducing the affine wakeups. - */ - if (!wl && sched_feat(ASYM_EFF_LOAD)) - return wl; - for_each_sched_entity(se) { - long S, rw, s, a, b; - long more_w; + long lw, w; - /* - * Instead of using this increment, also add the difference - * between when the shares were last updated and now. - */ - more_w = se->my_q->load.weight - se->my_q->rq_weight; - wl += more_w; - wg += more_w; - - S = se->my_q->tg->shares; - s = se->my_q->shares; - rw = se->my_q->rq_weight; + tg = se->my_q->tg; + w = se->my_q->load.weight; - a = S*(rw + wl); - b = S*rw + s*wg; + /* use this cpu's instantaneous contribution */ + lw = atomic_read(&tg->load_weight); + lw -= se->my_q->load_contribution; + lw += w + wg; - wl = s*(a-b); + wl += w; - if (likely(b)) - wl /= b; + if (lw > 0 && wl < lw) + wl = (wl * tg->shares) / lw; + else + wl = tg->shares; - /* - * Assume the group is already running and will - * thus already be accounted for in the weight. - * - * That is, moving shares between CPUs, does not - * alter the group weight. - */ + /* zero point is MIN_SHARES */ + if (wl < MIN_SHARES) + wl = MIN_SHARES; + wl -= se->load.weight; wg = 0; } @@ -1225,7 +1435,6 @@ static int wake_affine(struct sched_domain *sd, struct task_struct *p, int sync) 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; @@ -1241,6 +1450,7 @@ static int wake_affine(struct sched_domain *sd, struct task_struct *p, int sync) * effect of the currently running task from the load * of the current CPU: */ + rcu_read_lock(); if (sync) { tg = task_group(current); weight = current->se.load.weight; @@ -1252,8 +1462,6 @@ static int wake_affine(struct sched_domain *sd, struct task_struct *p, int sync) tg = task_group(p); weight = p->se.load.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 @@ -1263,9 +1471,22 @@ static int wake_affine(struct sched_domain *sd, struct task_struct *p, int sync) * 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)); + if (this_load) { + unsigned long this_eff_load, prev_eff_load; + + this_eff_load = 100; + this_eff_load *= power_of(prev_cpu); + this_eff_load *= this_load + + effective_load(tg, this_cpu, weight, weight); + + prev_eff_load = 100 + (sd->imbalance_pct - 100) / 2; + prev_eff_load *= power_of(this_cpu); + prev_eff_load *= load + effective_load(tg, prev_cpu, 0, weight); + + balanced = this_eff_load <= prev_eff_load; + } else + balanced = true; + rcu_read_unlock(); /* * If the currently running task will sleep within @@ -1302,7 +1523,7 @@ static struct sched_group * find_idlest_group(struct sched_domain *sd, struct task_struct *p, int this_cpu, int load_idx) { - struct sched_group *idlest = NULL, *this = NULL, *group = sd->groups; + struct sched_group *idlest = NULL, *group = sd->groups; unsigned long min_load = ULONG_MAX, this_load = 0; int imbalance = 100 + (sd->imbalance_pct-100)/2; @@ -1337,7 +1558,6 @@ find_idlest_group(struct sched_domain *sd, struct task_struct *p, if (local_group) { this_load = avg_load; - this = group; } else if (avg_load < min_load) { min_load = avg_load; idlest = group; @@ -1498,23 +1718,6 @@ select_task_rq_fair(struct rq *rq, struct task_struct *p, int sd_flag, int wake_ sd = tmp; } -#ifdef CONFIG_FAIR_GROUP_SCHED - if (sched_feat(LB_SHARES_UPDATE)) { - /* - * Pick the largest domain to update shares over - */ - tmp = sd; - if (affine_sd && (!tmp || affine_sd->span_weight > sd->span_weight)) - tmp = affine_sd; - - if (tmp) { - raw_spin_unlock(&rq->lock); - update_shares(tmp); - raw_spin_lock(&rq->lock); - } - } -#endif - if (affine_sd) { if (cpu == prev_cpu || wake_affine(affine_sd, p, sync)) return select_idle_sibling(p, cpu); @@ -1644,12 +1847,6 @@ static void check_preempt_wakeup(struct rq *rq, struct task_struct *p, int wake_ struct cfs_rq *cfs_rq = task_cfs_rq(curr); int scale = cfs_rq->nr_running >= sched_nr_latency; - if (unlikely(rt_prio(p->prio))) - goto preempt; - - if (unlikely(p->sched_class != &fair_sched_class)) - return; - if (unlikely(se == pse)) return; @@ -1788,7 +1985,7 @@ int can_migrate_task(struct task_struct *p, struct rq *rq, int this_cpu, * 2) too many balance attempts have failed. */ - tsk_cache_hot = task_hot(p, rq->clock, sd); + tsk_cache_hot = task_hot(p, rq->clock_task, sd); if (!tsk_cache_hot || sd->nr_balance_failed > sd->cache_nice_tries) { #ifdef CONFIG_SCHEDSTATS @@ -1905,6 +2102,48 @@ out: } #ifdef CONFIG_FAIR_GROUP_SCHED +/* + * update tg->load_weight by folding this cpu's load_avg + */ +static int update_shares_cpu(struct task_group *tg, int cpu) +{ + struct cfs_rq *cfs_rq; + unsigned long flags; + struct rq *rq; + + if (!tg->se[cpu]) + return 0; + + rq = cpu_rq(cpu); + cfs_rq = tg->cfs_rq[cpu]; + + raw_spin_lock_irqsave(&rq->lock, flags); + + update_rq_clock(rq); + update_cfs_load(cfs_rq, 1); + + /* + * We need to update shares after updating tg->load_weight in + * order to adjust the weight of groups with long running tasks. + */ + update_cfs_shares(cfs_rq, 0); + + raw_spin_unlock_irqrestore(&rq->lock, flags); + + return 0; +} + +static void update_shares(int cpu) +{ + struct cfs_rq *cfs_rq; + struct rq *rq = cpu_rq(cpu); + + rcu_read_lock(); + for_each_leaf_cfs_rq(rq, cfs_rq) + update_shares_cpu(cfs_rq->tg, cpu); + rcu_read_unlock(); +} + static unsigned long load_balance_fair(struct rq *this_rq, int this_cpu, struct rq *busiest, unsigned long max_load_move, @@ -1952,6 +2191,10 @@ load_balance_fair(struct rq *this_rq, int this_cpu, struct rq *busiest, return max_load_move - rem_load_move; } #else +static inline void update_shares(int cpu) +{ +} + static unsigned long load_balance_fair(struct rq *this_rq, int this_cpu, struct rq *busiest, unsigned long max_load_move, @@ -2020,12 +2263,17 @@ struct sd_lb_stats { unsigned long this_load; unsigned long this_load_per_task; unsigned long this_nr_running; + unsigned long this_has_capacity; + unsigned int this_idle_cpus; /* Statistics of the busiest group */ + unsigned int busiest_idle_cpus; unsigned long max_load; unsigned long busiest_load_per_task; unsigned long busiest_nr_running; unsigned long busiest_group_capacity; + unsigned long busiest_has_capacity; + unsigned int busiest_group_weight; int group_imb; /* Is there imbalance in this sd */ #if defined(CONFIG_SCHED_MC) || defined(CONFIG_SCHED_SMT) @@ -2047,7 +2295,10 @@ struct sg_lb_stats { unsigned long sum_nr_running; /* Nr tasks running in the group */ unsigned long sum_weighted_load; /* Weighted load of group's tasks */ unsigned long group_capacity; + unsigned long idle_cpus; + unsigned long group_weight; int group_imb; /* Is there an imbalance in the group ? */ + int group_has_capacity; /* Is there extra capacity in the group? */ }; /** @@ -2257,10 +2508,14 @@ 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(total < rq->rt_avg)) { + /* Ensures that power won't end up being negative */ + available = 0; + } else { + available = total - rq->rt_avg; + } if (unlikely((s64)total < SCHED_LOAD_SCALE)) total = SCHED_LOAD_SCALE; @@ -2276,13 +2531,6 @@ static void update_cpu_power(struct sched_domain *sd, int cpu) 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); @@ -2292,12 +2540,22 @@ static void update_cpu_power(struct sched_domain *sd, int cpu) power >>= SCHED_LOAD_SHIFT; } + sdg->cpu_power_orig = power; + + if (sched_feat(ARCH_POWER)) + power *= arch_scale_freq_power(sd, cpu); + else + power *= default_scale_freq_power(sd, cpu); + + power >>= SCHED_LOAD_SHIFT; + power *= scale_rt_power(cpu); power >>= SCHED_LOAD_SHIFT; if (!power) power = 1; + cpu_rq(cpu)->cpu_power = power; sdg->cpu_power = power; } @@ -2323,6 +2581,31 @@ static void update_group_power(struct sched_domain *sd, int cpu) sdg->cpu_power = power; } +/* + * Try and fix up capacity for tiny siblings, this is needed when + * things like SD_ASYM_PACKING need f_b_g to select another sibling + * which on its own isn't powerful enough. + * + * See update_sd_pick_busiest() and check_asym_packing(). + */ +static inline int +fix_small_capacity(struct sched_domain *sd, struct sched_group *group) +{ + /* + * Only siblings can have significantly less than SCHED_LOAD_SCALE + */ + if (sd->level != SD_LV_SIBLING) + return 0; + + /* + * If ~90% of the cpu_power is still there, we're good. + */ + if (group->cpu_power * 32 > group->cpu_power_orig * 29) + return 1; + + return 0; +} + /** * update_sg_lb_stats - Update sched_group's statistics for load balancing. * @sd: The sched_domain whose statistics are to be updated. @@ -2342,7 +2625,7 @@ static inline void update_sg_lb_stats(struct sched_domain *sd, int local_group, const struct cpumask *cpus, int *balance, struct sg_lb_stats *sgs) { - unsigned long load, max_cpu_load, min_cpu_load; + unsigned long load, max_cpu_load, min_cpu_load, max_nr_running; int i; unsigned int balance_cpu = -1, first_idle_cpu = 0; unsigned long avg_load_per_task = 0; @@ -2353,6 +2636,7 @@ static inline void update_sg_lb_stats(struct sched_domain *sd, /* Tally up the load of all CPUs in the group */ max_cpu_load = 0; min_cpu_load = ~0UL; + max_nr_running = 0; for_each_cpu_and(i, sched_group_cpus(group), cpus) { struct rq *rq = cpu_rq(i); @@ -2370,8 +2654,10 @@ static inline void update_sg_lb_stats(struct sched_domain *sd, load = target_load(i, load_idx); } else { load = source_load(i, load_idx); - if (load > max_cpu_load) + if (load > max_cpu_load) { max_cpu_load = load; + max_nr_running = rq->nr_running; + } if (min_cpu_load > load) min_cpu_load = load; } @@ -2379,7 +2665,8 @@ static inline void update_sg_lb_stats(struct sched_domain *sd, sgs->group_load += load; sgs->sum_nr_running += rq->nr_running; sgs->sum_weighted_load += weighted_cpuload(i); - + if (idle_cpu(i)) + sgs->idle_cpus++; } /* @@ -2388,14 +2675,14 @@ static inline void update_sg_lb_stats(struct sched_domain *sd, * domains. In the newly idle case, we will allow all the cpu's * to do the newly idle load balance. */ - if (idle != CPU_NEWLY_IDLE && local_group && - balance_cpu != this_cpu) { - *balance = 0; - return; + if (idle != CPU_NEWLY_IDLE && local_group) { + if (balance_cpu != this_cpu) { + *balance = 0; + return; + } + update_group_power(sd, this_cpu); } - update_group_power(sd, this_cpu); - /* Adjust by relative CPU power of the group */ sgs->avg_load = (sgs->group_load * SCHED_LOAD_SCALE) / group->cpu_power; @@ -2411,11 +2698,59 @@ static inline void update_sg_lb_stats(struct sched_domain *sd, if (sgs->sum_nr_running) avg_load_per_task = sgs->sum_weighted_load / sgs->sum_nr_running; - if ((max_cpu_load - min_cpu_load) > 2*avg_load_per_task) + if ((max_cpu_load - min_cpu_load) > 2*avg_load_per_task && max_nr_running > 1) sgs->group_imb = 1; - sgs->group_capacity = - DIV_ROUND_CLOSEST(group->cpu_power, SCHED_LOAD_SCALE); + sgs->group_capacity = DIV_ROUND_CLOSEST(group->cpu_power, SCHED_LOAD_SCALE); + if (!sgs->group_capacity) + sgs->group_capacity = fix_small_capacity(sd, group); + sgs->group_weight = group->group_weight; + + if (sgs->group_capacity > sgs->sum_nr_running) + sgs->group_has_capacity = 1; +} + +/** + * update_sd_pick_busiest - return 1 on busiest group + * @sd: sched_domain whose statistics are to be checked + * @sds: sched_domain statistics + * @sg: sched_group candidate to be checked for being the busiest + * @sgs: sched_group statistics + * @this_cpu: the current cpu + * + * Determine if @sg is a busier group than the previously selected + * busiest group. + */ +static bool update_sd_pick_busiest(struct sched_domain *sd, + struct sd_lb_stats *sds, + struct sched_group *sg, + struct sg_lb_stats *sgs, + int this_cpu) +{ + if (sgs->avg_load <= sds->max_load) + return false; + + if (sgs->sum_nr_running > sgs->group_capacity) + return true; + + if (sgs->group_imb) + return true; + + /* + * ASYM_PACKING needs to move all the work to the lowest + * numbered CPUs in the group, therefore mark all groups + * higher than ourself as busy. + */ + if ((sd->flags & SD_ASYM_PACKING) && sgs->sum_nr_running && + this_cpu < group_first_cpu(sg)) { + if (!sds->busiest) + return true; + + if (group_first_cpu(sds->busiest) > group_first_cpu(sg)) + return true; + } + + return false; } /** @@ -2423,7 +2758,7 @@ static inline void update_sg_lb_stats(struct sched_domain *sd, * @sd: sched_domain whose statistics are to be updated. * @this_cpu: Cpu for which load balance is currently performed. * @idle: Idle status of this_cpu - * @sd_idle: Idle status of the sched_domain containing group. + * @sd_idle: Idle status of the sched_domain containing sg. * @cpus: Set of cpus considered for load balancing. * @balance: Should we balance. * @sds: variable to hold the statistics for this sched_domain. @@ -2434,7 +2769,7 @@ static inline void update_sd_lb_stats(struct sched_domain *sd, int this_cpu, struct sd_lb_stats *sds) { struct sched_domain *child = sd->child; - struct sched_group *group = sd->groups; + struct sched_group *sg = sd->groups; struct sg_lb_stats sgs; int load_idx, prefer_sibling = 0; @@ -2447,45 +2782,103 @@ static inline void update_sd_lb_stats(struct sched_domain *sd, int this_cpu, do { int local_group; - local_group = cpumask_test_cpu(this_cpu, - sched_group_cpus(group)); + local_group = cpumask_test_cpu(this_cpu, sched_group_cpus(sg)); memset(&sgs, 0, sizeof(sgs)); - update_sg_lb_stats(sd, group, this_cpu, idle, load_idx, sd_idle, + update_sg_lb_stats(sd, sg, this_cpu, idle, load_idx, sd_idle, local_group, cpus, balance, &sgs); if (local_group && !(*balance)) return; sds->total_load += sgs.group_load; - sds->total_pwr += group->cpu_power; + sds->total_pwr += sg->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. + * first, lower the sg capacity to one so that we'll try + * and move all the excess tasks away. We lower the capacity + * of a group only if the local group has the capacity to fit + * these excess tasks, i.e. nr_running < group_capacity. The + * extra check prevents the case where you always pull from the + * heaviest group when it is already under-utilized (possible + * with a large weight task outweighs the tasks on the system). */ - if (prefer_sibling) + if (prefer_sibling && !local_group && sds->this_has_capacity) sgs.group_capacity = min(sgs.group_capacity, 1UL); if (local_group) { sds->this_load = sgs.avg_load; - sds->this = group; + sds->this = sg; sds->this_nr_running = sgs.sum_nr_running; sds->this_load_per_task = sgs.sum_weighted_load; - } else if (sgs.avg_load > sds->max_load && - (sgs.sum_nr_running > sgs.group_capacity || - sgs.group_imb)) { + sds->this_has_capacity = sgs.group_has_capacity; + sds->this_idle_cpus = sgs.idle_cpus; + } else if (update_sd_pick_busiest(sd, sds, sg, &sgs, this_cpu)) { sds->max_load = sgs.avg_load; - sds->busiest = group; + sds->busiest = sg; sds->busiest_nr_running = sgs.sum_nr_running; + sds->busiest_idle_cpus = sgs.idle_cpus; sds->busiest_group_capacity = sgs.group_capacity; sds->busiest_load_per_task = sgs.sum_weighted_load; + sds->busiest_has_capacity = sgs.group_has_capacity; + sds->busiest_group_weight = sgs.group_weight; sds->group_imb = sgs.group_imb; } - update_sd_power_savings_stats(group, sds, local_group, &sgs); - group = group->next; - } while (group != sd->groups); + update_sd_power_savings_stats(sg, sds, local_group, &sgs); + sg = sg->next; + } while (sg != sd->groups); +} + +int __weak arch_sd_sibling_asym_packing(void) +{ + return 0*SD_ASYM_PACKING; +} + +/** + * check_asym_packing - Check to see if the group is packed into the + * sched doman. + * + * This is primarily intended to used at the sibling level. Some + * cores like POWER7 prefer to use lower numbered SMT threads. In the + * case of POWER7, it can move to lower SMT modes only when higher + * threads are idle. When in lower SMT modes, the threads will + * perform better since they share less core resources. Hence when we + * have idle threads, we want them to be the higher ones. + * + * This packing function is run on idle threads. It checks to see if + * the busiest CPU in this domain (core in the P7 case) has a higher + * CPU number than the packing function is being run on. Here we are + * assuming lower CPU number will be equivalent to lower a SMT thread + * number. + * + * Returns 1 when packing is required and a task should be moved to + * this CPU. The amount of the imbalance is returned in *imbalance. + * + * @sd: The sched_domain whose packing is to be checked. + * @sds: Statistics of the sched_domain which is to be packed + * @this_cpu: The cpu at whose sched_domain we're performing load-balance. + * @imbalance: returns amount of imbalanced due to packing. + */ +static int check_asym_packing(struct sched_domain *sd, + struct sd_lb_stats *sds, + int this_cpu, unsigned long *imbalance) +{ + int busiest_cpu; + + if (!(sd->flags & SD_ASYM_PACKING)) + return 0; + + if (!sds->busiest) + return 0; + + busiest_cpu = group_first_cpu(sds->busiest); + if (this_cpu > busiest_cpu) + return 0; + + *imbalance = DIV_ROUND_CLOSEST(sds->max_load * sds->busiest->cpu_power, + SCHED_LOAD_SCALE); + return 1; } /** @@ -2625,6 +3018,7 @@ static inline void calculate_imbalance(struct sd_lb_stats *sds, int this_cpu, return fix_small_imbalance(sds, this_cpu, imbalance); } + /******* find_busiest_group() helpers end here *********************/ /** @@ -2676,13 +3070,27 @@ find_busiest_group(struct sched_domain *sd, int this_cpu, * 4) This group is more busy than the avg busieness at this * sched_domain. * 5) The imbalance is within the specified limit. + * + * Note: when doing newidle balance, if the local group has excess + * capacity (i.e. nr_running < group_capacity) and the busiest group + * does not have any capacity, we force a load balance to pull tasks + * to the local group. In this case, we skip past checks 3, 4 and 5. */ if (!(*balance)) goto ret; + if ((idle == CPU_IDLE || idle == CPU_NEWLY_IDLE) && + check_asym_packing(sd, &sds, this_cpu, imbalance)) + return sds.busiest; + if (!sds.busiest || sds.busiest_nr_running == 0) goto out_balanced; + /* SD_BALANCE_NEWIDLE trumps SMP nice when underutilized */ + if (idle == CPU_NEWLY_IDLE && sds.this_has_capacity && + !sds.busiest_has_capacity) + goto force_balance; + if (sds.this_load >= sds.max_load) goto out_balanced; @@ -2691,9 +3099,28 @@ find_busiest_group(struct sched_domain *sd, int this_cpu, if (sds.this_load >= sds.avg_load) goto out_balanced; - if (100 * sds.max_load <= sd->imbalance_pct * sds.this_load) - goto out_balanced; + /* + * In the CPU_NEWLY_IDLE, use imbalance_pct to be conservative. + * And to check for busy balance use !idle_cpu instead of + * CPU_NOT_IDLE. This is because HT siblings will use CPU_NOT_IDLE + * even when they are idle. + */ + if (idle == CPU_NEWLY_IDLE || !idle_cpu(this_cpu)) { + if (100 * sds.max_load <= sd->imbalance_pct * sds.this_load) + goto out_balanced; + } else { + /* + * This cpu is idle. If the busiest group load doesn't + * have more tasks than the number of available cpu's and + * there is no imbalance between this and busiest group + * wrt to idle cpu's, it is balanced. + */ + if ((sds.this_idle_cpus <= sds.busiest_idle_cpus + 1) && + sds.busiest_nr_running <= sds.busiest_group_weight) + goto out_balanced; + } +force_balance: /* Looks like there is an imbalance. Compute it */ calculate_imbalance(&sds, this_cpu, imbalance); return sds.busiest; @@ -2714,8 +3141,9 @@ ret: * find_busiest_queue - find the busiest runqueue among the cpus in group. */ static struct rq * -find_busiest_queue(struct sched_group *group, enum cpu_idle_type idle, - unsigned long imbalance, const struct cpumask *cpus) +find_busiest_queue(struct sched_domain *sd, struct sched_group *group, + enum cpu_idle_type idle, unsigned long imbalance, + const struct cpumask *cpus) { struct rq *busiest = NULL, *rq; unsigned long max_load = 0; @@ -2726,6 +3154,9 @@ find_busiest_queue(struct sched_group *group, enum cpu_idle_type idle, unsigned long capacity = DIV_ROUND_CLOSEST(power, SCHED_LOAD_SCALE); unsigned long wl; + if (!capacity) + capacity = fix_small_capacity(sd, group); + if (!cpumask_test_cpu(i, cpus)) continue; @@ -2765,9 +3196,19 @@ find_busiest_queue(struct sched_group *group, enum cpu_idle_type idle, /* Working cpumask for load_balance and load_balance_newidle. */ static DEFINE_PER_CPU(cpumask_var_t, load_balance_tmpmask); -static int need_active_balance(struct sched_domain *sd, int sd_idle, int idle) +static int need_active_balance(struct sched_domain *sd, int sd_idle, int idle, + int busiest_cpu, int this_cpu) { if (idle == CPU_NEWLY_IDLE) { + + /* + * ASYM_PACKING needs to force migrate tasks from busy but + * higher numbered CPUs in order to pack all tasks in the + * lowest numbered CPUs. + */ + if ((sd->flags & SD_ASYM_PACKING) && busiest_cpu > this_cpu) + return 1; + /* * The only task running in a non-idle cpu can be moved to this * cpu in an attempt to completely freeup the other CPU @@ -2830,7 +3271,6 @@ static int load_balance(int this_cpu, struct rq *this_rq, schedstat_inc(sd, lb_count[idle]); redo: - update_shares(sd); group = find_busiest_group(sd, this_cpu, &imbalance, idle, &sd_idle, cpus, balance); @@ -2842,7 +3282,7 @@ redo: goto out_balanced; } - busiest = find_busiest_queue(group, idle, imbalance, cpus); + busiest = find_busiest_queue(sd, group, idle, imbalance, cpus); if (!busiest) { schedstat_inc(sd, lb_nobusyq[idle]); goto out_balanced; @@ -2884,9 +3324,17 @@ redo: if (!ld_moved) { schedstat_inc(sd, lb_failed[idle]); - sd->nr_balance_failed++; + /* + * Increment the failure counter only on periodic balance. + * We do not want newidle balance, which can be very + * frequent, pollute the failure counter causing + * excessive cache_hot migrations and active balances. + */ + if (idle != CPU_NEWLY_IDLE) + sd->nr_balance_failed++; - if (need_active_balance(sd, sd_idle, idle)) { + if (need_active_balance(sd, sd_idle, idle, cpu_of(busiest), + this_cpu)) { raw_spin_lock_irqsave(&busiest->lock, flags); /* don't kick the active_load_balance_cpu_stop, @@ -2964,8 +3412,6 @@ out_one_pinned: else ld_moved = 0; out: - if (ld_moved) - update_shares(sd); return ld_moved; } @@ -2989,6 +3435,7 @@ static void idle_balance(int this_cpu, struct rq *this_rq) */ raw_spin_unlock(&this_rq->lock); + update_shares(this_cpu); for_each_domain(this_cpu, sd) { unsigned long interval; int balance = 1; @@ -3081,13 +3528,40 @@ out_unlock: } #ifdef CONFIG_NO_HZ + +static DEFINE_PER_CPU(struct call_single_data, remote_sched_softirq_cb); + +static void trigger_sched_softirq(void *data) +{ + raise_softirq_irqoff(SCHED_SOFTIRQ); +} + +static inline void init_sched_softirq_csd(struct call_single_data *csd) +{ + csd->func = trigger_sched_softirq; + csd->info = NULL; + csd->flags = 0; + csd->priv = 0; +} + +/* + * idle load balancing details + * - One of the idle CPUs nominates itself as idle load_balancer, while + * entering idle. + * - This idle load balancer CPU will also go into tickless mode when + * it is idle, just like all other idle CPUs + * - When one of the busy CPUs notice that there may be an idle rebalancing + * needed, they will kick the idle load balancer, which then does idle + * load balancing for all the idle CPUs. + */ 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), -}; + atomic_t first_pick_cpu; + atomic_t second_pick_cpu; + cpumask_var_t idle_cpus_mask; + cpumask_var_t grp_idle_mask; + unsigned long next_balance; /* in jiffy units */ +} nohz ____cacheline_aligned; int get_nohz_load_balancer(void) { @@ -3141,17 +3615,17 @@ static inline struct sched_domain *lowest_flag_domain(int cpu, int flag) */ static inline int is_semi_idle_group(struct sched_group *ilb_group) { - cpumask_and(nohz.ilb_grp_nohz_mask, nohz.cpu_mask, + cpumask_and(nohz.grp_idle_mask, nohz.idle_cpus_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)) + if (cpumask_empty(nohz.grp_idle_mask)) return 0; - if (cpumask_equal(nohz.ilb_grp_nohz_mask, sched_group_cpus(ilb_group))) + if (cpumask_equal(nohz.grp_idle_mask, sched_group_cpus(ilb_group))) return 0; return 1; @@ -3184,7 +3658,7 @@ static int find_new_ilb(int cpu) * 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) + if (cpumask_weight(nohz.idle_cpus_mask) < 2) goto out_done; for_each_flag_domain(cpu, sd, SD_POWERSAVINGS_BALANCE) { @@ -3192,7 +3666,7 @@ static int find_new_ilb(int cpu) do { if (is_semi_idle_group(ilb_group)) - return cpumask_first(nohz.ilb_grp_nohz_mask); + return cpumask_first(nohz.grp_idle_mask); ilb_group = ilb_group->next; @@ -3200,98 +3674,116 @@ static int find_new_ilb(int cpu) } out_done: - return cpumask_first(nohz.cpu_mask); + return nr_cpu_ids; } #else /* (CONFIG_SCHED_MC || CONFIG_SCHED_SMT) */ static inline int find_new_ilb(int call_cpu) { - return cpumask_first(nohz.cpu_mask); + return nr_cpu_ids; } #endif /* + * Kick a CPU to do the nohz balancing, if it is time for it. We pick the + * nohz_load_balancer CPU (if there is one) otherwise fallback to any idle + * CPU (if there is one). + */ +static void nohz_balancer_kick(int cpu) +{ + int ilb_cpu; + + nohz.next_balance++; + + ilb_cpu = get_nohz_load_balancer(); + + if (ilb_cpu >= nr_cpu_ids) { + ilb_cpu = cpumask_first(nohz.idle_cpus_mask); + if (ilb_cpu >= nr_cpu_ids) + return; + } + + if (!cpu_rq(ilb_cpu)->nohz_balance_kick) { + struct call_single_data *cp; + + cpu_rq(ilb_cpu)->nohz_balance_kick = 1; + cp = &per_cpu(remote_sched_softirq_cb, cpu); + __smp_call_function_single(ilb_cpu, cp, 0); + } + return; +} + +/* * 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 - * load balancing on behalf of all those cpus. If all the cpus in the system - * go into this tickless mode, then there will be no ilb owner (as there is - * no need for one) and all the cpus will sleep till the next wakeup event - * arrives... - * - * For the ilb owner, tick is not stopped. And this tick will be used - * for idle load balancing. ilb owner will still be part of - * nohz.cpu_mask.. + * load balancing on behalf of all those cpus. * - * While stopping the tick, this cpu will become the ilb owner if there - * is no other owner. And will be the owner till that cpu becomes busy - * or if all cpus in the system stop their ticks at which point - * there is no need for ilb owner. + * When the ilb owner becomes busy, we will not have new ilb owner until some + * idle CPU wakes up and goes back to idle or some busy CPU tries to kick + * idle load balancing by kicking one of the idle CPUs. * - * When the ilb owner becomes busy, it nominates another owner, during the - * next busy scheduler_tick() + * Ticks are stopped for the ilb owner as well, with busy CPU kicking this + * ilb owner CPU in future (when there is a need for idle load balancing on + * behalf of all idle CPUs). */ -int select_nohz_load_balancer(int stop_tick) +void select_nohz_load_balancer(int stop_tick) { int cpu = smp_processor_id(); if (stop_tick) { - cpu_rq(cpu)->in_nohz_recently = 1; - if (!cpu_active(cpu)) { if (atomic_read(&nohz.load_balancer) != cpu) - return 0; + return; /* * If we are going offline and still the leader, * give up! */ - if (atomic_cmpxchg(&nohz.load_balancer, cpu, -1) != cpu) + if (atomic_cmpxchg(&nohz.load_balancer, cpu, + nr_cpu_ids) != cpu) BUG(); - return 0; + return; } - cpumask_set_cpu(cpu, nohz.cpu_mask); + cpumask_set_cpu(cpu, nohz.idle_cpus_mask); - /* time for ilb owner also to sleep */ - if (cpumask_weight(nohz.cpu_mask) == num_active_cpus()) { - if (atomic_read(&nohz.load_balancer) == cpu) - atomic_set(&nohz.load_balancer, -1); - return 0; - } + if (atomic_read(&nohz.first_pick_cpu) == cpu) + atomic_cmpxchg(&nohz.first_pick_cpu, cpu, nr_cpu_ids); + if (atomic_read(&nohz.second_pick_cpu) == cpu) + atomic_cmpxchg(&nohz.second_pick_cpu, cpu, nr_cpu_ids); - if (atomic_read(&nohz.load_balancer) == -1) { - /* make me the ilb owner */ - if (atomic_cmpxchg(&nohz.load_balancer, -1, cpu) == -1) - return 1; - } else if (atomic_read(&nohz.load_balancer) == cpu) { + if (atomic_read(&nohz.load_balancer) >= nr_cpu_ids) { int new_ilb; - if (!(sched_smt_power_savings || - sched_mc_power_savings)) - return 1; + /* make me the ilb owner */ + if (atomic_cmpxchg(&nohz.load_balancer, nr_cpu_ids, + cpu) != nr_cpu_ids) + return; + /* * 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); + atomic_set(&nohz.load_balancer, nr_cpu_ids); resched_cpu(new_ilb); - return 0; + return; } - return 1; + return; } } else { - if (!cpumask_test_cpu(cpu, nohz.cpu_mask)) - return 0; + if (!cpumask_test_cpu(cpu, nohz.idle_cpus_mask)) + return; - cpumask_clear_cpu(cpu, nohz.cpu_mask); + cpumask_clear_cpu(cpu, nohz.idle_cpus_mask); if (atomic_read(&nohz.load_balancer) == cpu) - if (atomic_cmpxchg(&nohz.load_balancer, cpu, -1) != cpu) + if (atomic_cmpxchg(&nohz.load_balancer, cpu, + nr_cpu_ids) != cpu) BUG(); } - return 0; + return; } #endif @@ -3314,6 +3806,8 @@ static void rebalance_domains(int cpu, enum cpu_idle_type idle) int update_next_balance = 0; int need_serialize; + update_shares(cpu); + for_each_domain(cpu, sd) { if (!(sd->flags & SD_LOAD_BALANCE)) continue; @@ -3373,11 +3867,102 @@ out: rq->next_balance = next_balance; } +#ifdef CONFIG_NO_HZ /* - * run_rebalance_domains is triggered when needed from the scheduler tick. - * In CONFIG_NO_HZ case, the idle load balance owner will do the + * In CONFIG_NO_HZ case, the idle balance kickee will do the * rebalancing for all the cpus for whom scheduler ticks are stopped. */ +static void nohz_idle_balance(int this_cpu, enum cpu_idle_type idle) +{ + struct rq *this_rq = cpu_rq(this_cpu); + struct rq *rq; + int balance_cpu; + + if (idle != CPU_IDLE || !this_rq->nohz_balance_kick) + return; + + for_each_cpu(balance_cpu, nohz.idle_cpus_mask) { + if (balance_cpu == this_cpu) + continue; + + /* + * If this cpu gets work to do, stop the load balancing + * work being done for other cpus. Next load + * balancing owner will pick it up. + */ + if (need_resched()) { + this_rq->nohz_balance_kick = 0; + break; + } + + raw_spin_lock_irq(&this_rq->lock); + update_rq_clock(this_rq); + update_cpu_load(this_rq); + raw_spin_unlock_irq(&this_rq->lock); + + rebalance_domains(balance_cpu, CPU_IDLE); + + rq = cpu_rq(balance_cpu); + if (time_after(this_rq->next_balance, rq->next_balance)) + this_rq->next_balance = rq->next_balance; + } + nohz.next_balance = this_rq->next_balance; + this_rq->nohz_balance_kick = 0; +} + +/* + * Current heuristic for kicking the idle load balancer + * - first_pick_cpu is the one of the busy CPUs. It will kick + * idle load balancer when it has more than one process active. This + * eliminates the need for idle load balancing altogether when we have + * only one running process in the system (common case). + * - If there are more than one busy CPU, idle load balancer may have + * to run for active_load_balance to happen (i.e., two busy CPUs are + * SMT or core siblings and can run better if they move to different + * physical CPUs). So, second_pick_cpu is the second of the busy CPUs + * which will kick idle load balancer as soon as it has any load. + */ +static inline int nohz_kick_needed(struct rq *rq, int cpu) +{ + unsigned long now = jiffies; + int ret; + int first_pick_cpu, second_pick_cpu; + + if (time_before(now, nohz.next_balance)) + return 0; + + if (rq->idle_at_tick) + return 0; + + first_pick_cpu = atomic_read(&nohz.first_pick_cpu); + second_pick_cpu = atomic_read(&nohz.second_pick_cpu); + + if (first_pick_cpu < nr_cpu_ids && first_pick_cpu != cpu && + second_pick_cpu < nr_cpu_ids && second_pick_cpu != cpu) + return 0; + + ret = atomic_cmpxchg(&nohz.first_pick_cpu, nr_cpu_ids, cpu); + if (ret == nr_cpu_ids || ret == cpu) { + atomic_cmpxchg(&nohz.second_pick_cpu, cpu, nr_cpu_ids); + if (rq->nr_running > 1) + return 1; + } else { + ret = atomic_cmpxchg(&nohz.second_pick_cpu, nr_cpu_ids, cpu); + if (ret == nr_cpu_ids || ret == cpu) { + if (rq->nr_running) + return 1; + } + } + return 0; +} +#else +static void nohz_idle_balance(int this_cpu, enum cpu_idle_type idle) { } +#endif + +/* + * run_rebalance_domains is triggered when needed from the scheduler tick. + * Also triggered for nohz idle balancing (with nohz_balancing_kick set). + */ static void run_rebalance_domains(struct softirq_action *h) { int this_cpu = smp_processor_id(); @@ -3387,37 +3972,12 @@ static void run_rebalance_domains(struct softirq_action *h) rebalance_domains(this_cpu, idle); -#ifdef CONFIG_NO_HZ /* - * If this cpu is the owner for idle load balancing, then do the + * If this cpu has a pending nohz_balance_kick, then do the * balancing on behalf of the other idle cpus whose ticks are * stopped. */ - if (this_rq->idle_at_tick && - atomic_read(&nohz.load_balancer) == this_cpu) { - struct rq *rq; - int balance_cpu; - - for_each_cpu(balance_cpu, nohz.cpu_mask) { - if (balance_cpu == this_cpu) - continue; - - /* - * If this cpu gets work to do, stop the load balancing - * work being done for other cpus. Next load - * balancing owner will pick it up. - */ - if (need_resched()) - break; - - rebalance_domains(balance_cpu, CPU_IDLE); - - rq = cpu_rq(balance_cpu); - if (time_after(this_rq->next_balance, rq->next_balance)) - this_rq->next_balance = rq->next_balance; - } - } -#endif + nohz_idle_balance(this_cpu, idle); } static inline int on_null_domain(int cpu) @@ -3427,57 +3987,17 @@ static inline int on_null_domain(int cpu) /* * Trigger the SCHED_SOFTIRQ if it is time to do periodic load balancing. - * - * In case of CONFIG_NO_HZ, this is the place where we nominate a new - * idle load balancing owner or decide to stop the periodic load balancing, - * if the whole system is idle. */ static inline void trigger_load_balance(struct rq *rq, int cpu) { -#ifdef CONFIG_NO_HZ - /* - * If we were in the nohz mode recently and busy at the current - * scheduler tick, then check if we need to nominate new idle - * load balancer. - */ - if (rq->in_nohz_recently && !rq->idle_at_tick) { - rq->in_nohz_recently = 0; - - if (atomic_read(&nohz.load_balancer) == cpu) { - cpumask_clear_cpu(cpu, nohz.cpu_mask); - atomic_set(&nohz.load_balancer, -1); - } - - if (atomic_read(&nohz.load_balancer) == -1) { - int ilb = find_new_ilb(cpu); - - if (ilb < nr_cpu_ids) - resched_cpu(ilb); - } - } - - /* - * If this cpu is idle and doing idle load balancing for all the - * cpus with ticks stopped, is it time for that to stop? - */ - if (rq->idle_at_tick && atomic_read(&nohz.load_balancer) == cpu && - cpumask_weight(nohz.cpu_mask) == num_online_cpus()) { - resched_cpu(cpu); - return; - } - - /* - * If this cpu is idle and the idle load balancing is done by - * someone else, then no need raise the SCHED_SOFTIRQ - */ - if (rq->idle_at_tick && atomic_read(&nohz.load_balancer) != cpu && - cpumask_test_cpu(cpu, nohz.cpu_mask)) - return; -#endif /* Don't need to rebalance while attached to NULL domain */ if (time_after_eq(jiffies, rq->next_balance) && likely(!on_null_domain(cpu))) raise_softirq(SCHED_SOFTIRQ); +#ifdef CONFIG_NO_HZ + else if (nohz_kick_needed(rq, cpu) && likely(!on_null_domain(cpu))) + nohz_balancer_kick(cpu); +#endif } static void rq_online_fair(struct rq *rq) @@ -3530,8 +4050,13 @@ static void task_fork_fair(struct task_struct *p) raw_spin_lock_irqsave(&rq->lock, flags); - if (unlikely(task_cpu(p) != this_cpu)) + update_rq_clock(rq); + + if (unlikely(task_cpu(p) != this_cpu)) { + rcu_read_lock(); __set_task_cpu(p, this_cpu); + rcu_read_unlock(); + } update_curr(cfs_rq); @@ -3603,13 +4128,26 @@ static void set_curr_task_fair(struct rq *rq) } #ifdef CONFIG_FAIR_GROUP_SCHED -static void moved_group_fair(struct task_struct *p, int on_rq) +static void task_move_group_fair(struct task_struct *p, int on_rq) { - struct cfs_rq *cfs_rq = task_cfs_rq(p); - - update_curr(cfs_rq); + /* + * If the task was not on the rq at the time of this cgroup movement + * it must have been asleep, sleeping tasks keep their ->vruntime + * absolute on their old rq until wakeup (needed for the fair sleeper + * bonus in place_entity()). + * + * If it was on the rq, we've just 'preempted' it, which does convert + * ->vruntime to a relative base. + * + * Make sure both cases convert their relative position when migrating + * to another cgroup's rq. This does somewhat interfere with the + * fair sleeper stuff for the first placement, but who cares. + */ + if (!on_rq) + p->se.vruntime -= cfs_rq_of(&p->se)->min_vruntime; + set_task_rq(p, task_cpu(p)); if (!on_rq) - place_entity(cfs_rq, &p->se, 1); + p->se.vruntime += cfs_rq_of(&p->se)->min_vruntime; } #endif @@ -3661,7 +4199,7 @@ static const struct sched_class fair_sched_class = { .get_rr_interval = get_rr_interval_fair, #ifdef CONFIG_FAIR_GROUP_SCHED - .moved_group = moved_group_fair, + .task_move_group = task_move_group_fair, #endif }; diff --git a/kernel/sched_features.h b/kernel/sched_features.h index 83c66e8ad3e..68e69acc29b 100644 --- a/kernel/sched_features.h +++ b/kernel/sched_features.h @@ -52,8 +52,6 @@ SCHED_FEAT(ARCH_POWER, 0) SCHED_FEAT(HRTICK, 0) SCHED_FEAT(DOUBLE_TICK, 0) SCHED_FEAT(LB_BIAS, 1) -SCHED_FEAT(LB_SHARES_UPDATE, 1) -SCHED_FEAT(ASYM_EFF_LOAD, 1) /* * Spin-wait on mutex acquisition when the mutex owner is running on @@ -61,3 +59,8 @@ SCHED_FEAT(ASYM_EFF_LOAD, 1) * release the lock. Decreases scheduling overhead. */ SCHED_FEAT(OWNER_SPIN, 1) + +/* + * Decrement CPU power based on irq activity + */ +SCHED_FEAT(NONIRQ_POWER, 1) diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c index 8afb953e31c..c914ec747ca 100644 --- a/kernel/sched_rt.c +++ b/kernel/sched_rt.c @@ -183,6 +183,17 @@ static inline u64 sched_rt_period(struct rt_rq *rt_rq) return ktime_to_ns(rt_rq->tg->rt_bandwidth.rt_period); } +static inline void list_add_leaf_rt_rq(struct rt_rq *rt_rq) +{ + list_add_rcu(&rt_rq->leaf_rt_rq_list, + &rq_of_rt_rq(rt_rq)->leaf_rt_rq_list); +} + +static inline void list_del_leaf_rt_rq(struct rt_rq *rt_rq) +{ + list_del_rcu(&rt_rq->leaf_rt_rq_list); +} + #define for_each_leaf_rt_rq(rt_rq, rq) \ list_for_each_entry_rcu(rt_rq, &rq->leaf_rt_rq_list, leaf_rt_rq_list) @@ -276,6 +287,14 @@ static inline u64 sched_rt_period(struct rt_rq *rt_rq) return ktime_to_ns(def_rt_bandwidth.rt_period); } +static inline void list_add_leaf_rt_rq(struct rt_rq *rt_rq) +{ +} + +static inline void list_del_leaf_rt_rq(struct rt_rq *rt_rq) +{ +} + #define for_each_leaf_rt_rq(rt_rq, rq) \ for (rt_rq = &rq->rt; rt_rq; rt_rq = NULL) @@ -609,7 +628,7 @@ static void update_curr_rt(struct rq *rq) if (!task_has_rt_policy(curr)) return; - delta_exec = rq->clock - curr->se.exec_start; + delta_exec = rq->clock_task - curr->se.exec_start; if (unlikely((s64)delta_exec < 0)) delta_exec = 0; @@ -618,7 +637,7 @@ static void update_curr_rt(struct rq *rq) curr->se.sum_exec_runtime += delta_exec; account_group_exec_runtime(curr, delta_exec); - curr->se.exec_start = rq->clock; + curr->se.exec_start = rq->clock_task; cpuacct_charge(curr, delta_exec); sched_rt_avg_update(rq, delta_exec); @@ -825,6 +844,9 @@ static void __enqueue_rt_entity(struct sched_rt_entity *rt_se, bool head) if (group_rq && (rt_rq_throttled(group_rq) || !group_rq->rt_nr_running)) return; + if (!rt_rq->rt_nr_running) + list_add_leaf_rt_rq(rt_rq); + if (head) list_add(&rt_se->run_list, queue); else @@ -844,6 +866,8 @@ static void __dequeue_rt_entity(struct sched_rt_entity *rt_se) __clear_bit(rt_se_prio(rt_se), array->bitmap); dec_rt_tasks(rt_se, rt_rq); + if (!rt_rq->rt_nr_running) + list_del_leaf_rt_rq(rt_rq); } /* @@ -960,18 +984,19 @@ select_task_rq_rt(struct rq *rq, struct task_struct *p, int sd_flag, int flags) * runqueue. Otherwise simply start this RT task * on its current runqueue. * - * We want to avoid overloading runqueues. Even if - * the RT task is of higher priority than the current RT task. - * RT tasks behave differently than other tasks. If - * one gets preempted, we try to push it off to another queue. - * So trying to keep a preempting RT task on the same - * cache hot CPU will force the running RT task to - * a cold CPU. So we waste all the cache for the lower - * RT task in hopes of saving some of a RT task - * that is just being woken and probably will have - * cold cache anyway. + * We want to avoid overloading runqueues. If the woken + * task is a higher priority, then it will stay on this CPU + * and the lower prio task should be moved to another CPU. + * Even though this will probably make the lower prio task + * lose its cache, we do not want to bounce a higher task + * around just because it gave up its CPU, perhaps for a + * lock? + * + * For equal prio tasks, we just let the scheduler sort it out. */ if (unlikely(rt_task(rq->curr)) && + (rq->curr->rt.nr_cpus_allowed < 2 || + rq->curr->prio < p->prio) && (p->rt.nr_cpus_allowed > 1)) { int cpu = find_lowest_rq(p); @@ -1074,7 +1099,7 @@ static struct task_struct *_pick_next_task_rt(struct rq *rq) } while (rt_rq); p = rt_task_of(rt_se); - p->se.exec_start = rq->clock; + p->se.exec_start = rq->clock_task; return p; } @@ -1139,7 +1164,7 @@ static struct task_struct *pick_next_highest_task_rt(struct rq *rq, int cpu) for_each_leaf_rt_rq(rt_rq, rq) { array = &rt_rq->active; idx = sched_find_first_bit(array->bitmap); - next_idx: +next_idx: if (idx >= MAX_RT_PRIO) continue; if (next && next->prio < idx) @@ -1315,7 +1340,7 @@ static int push_rt_task(struct rq *rq) if (!next_task) return 0; - retry: +retry: if (unlikely(next_task == rq->curr)) { WARN_ON(1); return 0; @@ -1463,7 +1488,7 @@ static int pull_rt_task(struct rq *this_rq) * but possible) */ } - skip: +skip: double_unlock_balance(this_rq, src_rq); } @@ -1491,7 +1516,10 @@ static void task_woken_rt(struct rq *rq, struct task_struct *p) if (!task_running(rq, p) && !test_tsk_need_resched(rq->curr) && has_pushable_tasks(rq) && - p->rt.nr_cpus_allowed > 1) + p->rt.nr_cpus_allowed > 1 && + rt_task(rq->curr) && + (rq->curr->rt.nr_cpus_allowed < 2 || + rq->curr->prio < p->prio)) push_rt_tasks(rq); } @@ -1663,9 +1691,6 @@ static void watchdog(struct rq *rq, struct task_struct *p) { unsigned long soft, hard; - if (!p->signal) - return; - /* max may change after cur was read, this will be fixed next tick */ soft = task_rlimit(p, RLIMIT_RTTIME); hard = task_rlimit_max(p, RLIMIT_RTTIME); @@ -1712,7 +1737,7 @@ static void set_curr_task_rt(struct rq *rq) { struct task_struct *p = rq->curr; - p->se.exec_start = rq->clock; + p->se.exec_start = rq->clock_task; /* The running task is never eligible for pushing */ dequeue_pushable_task(rq, p); diff --git a/kernel/sched_stats.h b/kernel/sched_stats.h index 32d2bd4061b..48ddf431db0 100644 --- a/kernel/sched_stats.h +++ b/kernel/sched_stats.h @@ -157,15 +157,7 @@ static inline void sched_info_reset_dequeued(struct task_struct *t) } /* - * Called when a process is dequeued from the active array and given - * the cpu. We should note that with the exception of interactive - * tasks, the expired queue will become the active queue after the active - * queue is empty, without explicitly dequeuing and requeuing tasks in the - * expired queue. (Interactive tasks may be requeued directly to the - * active queue, thus delaying tasks in the expired queue from running; - * see scheduler_tick()). - * - * Though we are interested in knowing how long it was from the *first* time a + * We are interested in knowing how long it was from the *first* time a * task was queued to the time that it finally hit a cpu, we call this routine * from dequeue_task() to account for possible rq->clock skew across cpus. The * delta taken on each cpu would annul the skew. @@ -203,16 +195,6 @@ static void sched_info_arrive(struct task_struct *t) } /* - * Called when a process is queued into either the active or expired - * array. The time is noted and later used to determine how long we - * had to wait for us to reach the cpu. Since the expired queue will - * become the active queue after active queue is empty, without dequeuing - * and requeuing any tasks, we are interested in queuing to either. It - * is unusual but not impossible for tasks to be dequeued and immediately - * requeued in the same or another array: this can happen in sched_yield(), - * set_user_nice(), and even load_balance() as it moves tasks from runqueue - * to runqueue. - * * This function is only called from enqueue_task(), but also only updates * the timestamp if it is already not set. It's assumed that * sched_info_dequeued() will clear that stamp when appropriate. @@ -295,13 +277,7 @@ sched_info_switch(struct task_struct *prev, struct task_struct *next) static inline void account_group_user_time(struct task_struct *tsk, cputime_t cputime) { - struct thread_group_cputimer *cputimer; - - /* tsk == current, ensure it is safe to use ->signal */ - if (unlikely(tsk->exit_state)) - return; - - cputimer = &tsk->signal->cputimer; + struct thread_group_cputimer *cputimer = &tsk->signal->cputimer; if (!cputimer->running) return; @@ -325,13 +301,7 @@ static inline void account_group_user_time(struct task_struct *tsk, static inline void account_group_system_time(struct task_struct *tsk, cputime_t cputime) { - struct thread_group_cputimer *cputimer; - - /* tsk == current, ensure it is safe to use ->signal */ - if (unlikely(tsk->exit_state)) - return; - - cputimer = &tsk->signal->cputimer; + struct thread_group_cputimer *cputimer = &tsk->signal->cputimer; if (!cputimer->running) return; @@ -355,16 +325,7 @@ static inline void account_group_system_time(struct task_struct *tsk, static inline void account_group_exec_runtime(struct task_struct *tsk, unsigned long long ns) { - struct thread_group_cputimer *cputimer; - struct signal_struct *sig; - - sig = tsk->signal; - /* see __exit_signal()->task_rq_unlock_wait() */ - barrier(); - if (unlikely(!sig)) - return; - - cputimer = &sig->cputimer; + struct thread_group_cputimer *cputimer = &tsk->signal->cputimer; if (!cputimer->running) return; diff --git a/kernel/sched_stoptask.c b/kernel/sched_stoptask.c new file mode 100644 index 00000000000..2bf6b47058c --- /dev/null +++ b/kernel/sched_stoptask.c @@ -0,0 +1,108 @@ +/* + * stop-task scheduling class. + * + * The stop task is the highest priority task in the system, it preempts + * everything and will be preempted by nothing. + * + * See kernel/stop_machine.c + */ + +#ifdef CONFIG_SMP +static int +select_task_rq_stop(struct rq *rq, struct task_struct *p, + int sd_flag, int flags) +{ + return task_cpu(p); /* stop tasks as never migrate */ +} +#endif /* CONFIG_SMP */ + +static void +check_preempt_curr_stop(struct rq *rq, struct task_struct *p, int flags) +{ + /* we're never preempted */ +} + +static struct task_struct *pick_next_task_stop(struct rq *rq) +{ + struct task_struct *stop = rq->stop; + + if (stop && stop->se.on_rq) + return stop; + + return NULL; +} + +static void +enqueue_task_stop(struct rq *rq, struct task_struct *p, int flags) +{ +} + +static void +dequeue_task_stop(struct rq *rq, struct task_struct *p, int flags) +{ +} + +static void yield_task_stop(struct rq *rq) +{ + BUG(); /* the stop task should never yield, its pointless. */ +} + +static void put_prev_task_stop(struct rq *rq, struct task_struct *prev) +{ +} + +static void task_tick_stop(struct rq *rq, struct task_struct *curr, int queued) +{ +} + +static void set_curr_task_stop(struct rq *rq) +{ +} + +static void switched_to_stop(struct rq *rq, struct task_struct *p, + int running) +{ + BUG(); /* its impossible to change to this class */ +} + +static void prio_changed_stop(struct rq *rq, struct task_struct *p, + int oldprio, int running) +{ + BUG(); /* how!?, what priority? */ +} + +static unsigned int +get_rr_interval_stop(struct rq *rq, struct task_struct *task) +{ + return 0; +} + +/* + * Simple, special scheduling class for the per-CPU stop tasks: + */ +static const struct sched_class stop_sched_class = { + .next = &rt_sched_class, + + .enqueue_task = enqueue_task_stop, + .dequeue_task = dequeue_task_stop, + .yield_task = yield_task_stop, + + .check_preempt_curr = check_preempt_curr_stop, + + .pick_next_task = pick_next_task_stop, + .put_prev_task = put_prev_task_stop, + +#ifdef CONFIG_SMP + .select_task_rq = select_task_rq_stop, +#endif + + .set_curr_task = set_curr_task_stop, + .task_tick = task_tick_stop, + + .get_rr_interval = get_rr_interval_stop, + + .prio_changed = prio_changed_stop, + .switched_to = switched_to_stop, + + /* no .task_new for stop tasks */ +}; diff --git a/kernel/signal.c b/kernel/signal.c index 825a3f24ad7..4e3cff10fdc 100644 --- a/kernel/signal.c +++ b/kernel/signal.c @@ -637,12 +637,12 @@ static inline bool si_fromuser(const struct siginfo *info) /* * Bad permissions for sending the signal - * - the caller must hold at least the RCU read lock + * - the caller must hold the RCU read lock */ static int check_kill_permission(int sig, struct siginfo *info, struct task_struct *t) { - const struct cred *cred = current_cred(), *tcred; + const struct cred *cred, *tcred; struct pid *sid; int error; @@ -656,8 +656,10 @@ static int check_kill_permission(int sig, struct siginfo *info, if (error) return error; + cred = current_cred(); tcred = __task_cred(t); - if ((cred->euid ^ tcred->suid) && + if (!same_thread_group(current, t) && + (cred->euid ^ tcred->suid) && (cred->euid ^ tcred->uid) && (cred->uid ^ tcred->suid) && (cred->uid ^ tcred->uid) && @@ -1083,26 +1085,28 @@ force_sig_info(int sig, struct siginfo *info, struct task_struct *t) /* * Nuke all other threads in the group. */ -void zap_other_threads(struct task_struct *p) +int zap_other_threads(struct task_struct *p) { - struct task_struct *t; + struct task_struct *t = p; + int count = 0; p->signal->group_stop_count = 0; - for (t = next_thread(p); t != p; t = next_thread(t)) { - /* - * Don't bother with already dead threads - */ + while_each_thread(p, t) { + count++; + + /* Don't bother with already dead threads */ if (t->exit_state) continue; - - /* SIGKILL will be handled before any pending SIGSTOP */ sigaddset(&t->pending.signal, SIGKILL); signal_wake_up(t, 1); } + + return count; } -struct sighand_struct *lock_task_sighand(struct task_struct *tsk, unsigned long *flags) +struct sighand_struct *__lock_task_sighand(struct task_struct *tsk, + unsigned long *flags) { struct sighand_struct *sighand; @@ -1124,11 +1128,14 @@ struct sighand_struct *lock_task_sighand(struct task_struct *tsk, unsigned long /* * send signal info to all the members of a group - * - the caller must hold the RCU read lock at least */ int group_send_sig_info(int sig, struct siginfo *info, struct task_struct *p) { - int ret = check_kill_permission(sig, info, p); + int ret; + + rcu_read_lock(); + ret = check_kill_permission(sig, info, p); + rcu_read_unlock(); if (!ret && sig) ret = do_send_sig_info(sig, info, p, true); @@ -1611,6 +1618,8 @@ static int sigkill_pending(struct task_struct *tsk) * is gone, we keep current->exit_code unless clear_code. */ static void ptrace_stop(int exit_code, int clear_code, siginfo_t *info) + __releases(¤t->sighand->siglock) + __acquires(¤t->sighand->siglock) { if (arch_ptrace_stop_needed(exit_code, info)) { /* @@ -2209,6 +2218,14 @@ int copy_siginfo_to_user(siginfo_t __user *to, siginfo_t *from) #ifdef __ARCH_SI_TRAPNO err |= __put_user(from->si_trapno, &to->si_trapno); #endif +#ifdef BUS_MCEERR_AO + /* + * Other callers might not initialize the si_lsb field, + * so check explicitely for the right codes here. + */ + if (from->si_code == BUS_MCEERR_AR || from->si_code == BUS_MCEERR_AO) + err |= __put_user(from->si_addr_lsb, &to->si_addr_lsb); +#endif break; case __SI_CHLD: err |= __put_user(from->si_pid, &to->si_pid); diff --git a/kernel/slow-work-debugfs.c b/kernel/slow-work-debugfs.c deleted file mode 100644 index e45c4364529..00000000000 --- a/kernel/slow-work-debugfs.c +++ /dev/null @@ -1,227 +0,0 @@ -/* 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 deleted file mode 100644 index 7d3f4fa9ef4..00000000000 --- a/kernel/slow-work.c +++ /dev/null @@ -1,1068 +0,0 @@ -/* 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 = slow_work_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 deleted file mode 100644 index a29ebd1ef41..00000000000 --- a/kernel/slow-work.h +++ /dev/null @@ -1,72 +0,0 @@ -/* 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_DEBUG - slow_work_pids[id] = pid; -#endif -} - -static inline void slow_work_mark_time(struct slow_work *work) -{ -#ifdef CONFIG_SLOW_WORK_DEBUG - work->mark = CURRENT_TIME; -#endif -} - -static inline void slow_work_begin_exec(int id, struct slow_work *work) -{ -#ifdef CONFIG_SLOW_WORK_DEBUG - slow_work_execs[id] = work; -#endif -} - -static inline void slow_work_end_exec(int id, struct slow_work *work) -{ -#ifdef CONFIG_SLOW_WORK_DEBUG - 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 3fc69733618..9910744f085 100644 --- a/kernel/smp.c +++ b/kernel/smp.c @@ -13,6 +13,7 @@ #include <linux/smp.h> #include <linux/cpu.h> +#ifdef CONFIG_USE_GENERIC_SMP_HELPERS static struct { struct list_head queue; raw_spinlock_t lock; @@ -52,7 +53,7 @@ hotplug_cfd(struct notifier_block *nfb, unsigned long action, void *hcpu) case CPU_UP_PREPARE_FROZEN: if (!zalloc_cpumask_var_node(&cfd->cpumask, GFP_KERNEL, cpu_to_node(cpu))) - return NOTIFY_BAD; + return notifier_from_errno(-ENOMEM); break; #ifdef CONFIG_HOTPLUG_CPU @@ -193,23 +194,52 @@ void generic_smp_call_function_interrupt(void) */ list_for_each_entry_rcu(data, &call_function.queue, csd.list) { int refs; + void (*func) (void *info); - if (!cpumask_test_and_clear_cpu(cpu, data->cpumask)) + /* + * Since we walk the list without any locks, we might + * see an entry that was completed, removed from the + * list and is in the process of being reused. + * + * We must check that the cpu is in the cpumask before + * checking the refs, and both must be set before + * executing the callback on this cpu. + */ + + if (!cpumask_test_cpu(cpu, data->cpumask)) + continue; + + smp_rmb(); + + if (atomic_read(&data->refs) == 0) continue; + func = data->csd.func; /* for later warn */ data->csd.func(data->csd.info); + /* + * If the cpu mask is not still set then it enabled interrupts, + * we took another smp interrupt, and executed the function + * twice on this cpu. In theory that copy decremented refs. + */ + if (!cpumask_test_and_clear_cpu(cpu, data->cpumask)) { + WARN(1, "%pS enabled interrupts and double executed\n", + func); + continue; + } + refs = atomic_dec_return(&data->refs); WARN_ON(refs < 0); - if (!refs) { - raw_spin_lock(&call_function.lock); - list_del_rcu(&data->csd.list); - raw_spin_unlock(&call_function.lock); - } if (refs) continue; + WARN_ON(!cpumask_empty(data->cpumask)); + + raw_spin_lock(&call_function.lock); + list_del_rcu(&data->csd.list); + raw_spin_unlock(&call_function.lock); + csd_unlock(&data->csd); } @@ -267,7 +297,7 @@ static DEFINE_PER_CPU_SHARED_ALIGNED(struct call_single_data, csd_data); * * Returns 0 on success, else a negative status code. */ -int smp_call_function_single(int cpu, void (*func) (void *info), void *info, +int smp_call_function_single(int cpu, smp_call_func_t func, void *info, int wait) { struct call_single_data d = { @@ -336,7 +366,7 @@ EXPORT_SYMBOL(smp_call_function_single); * 3) any other online cpu in @mask */ int smp_call_function_any(const struct cpumask *mask, - void (*func)(void *info), void *info, int wait) + smp_call_func_t func, void *info, int wait) { unsigned int cpu; const struct cpumask *nodemask; @@ -365,9 +395,10 @@ call: EXPORT_SYMBOL_GPL(smp_call_function_any); /** - * __smp_call_function_single(): Run a function on another CPU + * __smp_call_function_single(): Run a function on a specific CPU * @cpu: The CPU to run on. * @data: Pre-allocated and setup data structure + * @wait: If true, wait until function has completed on specified CPU. * * Like smp_call_function_single(), but allow caller to pass in a * pre-allocated data structure. Useful for embedding @data inside @@ -376,8 +407,10 @@ EXPORT_SYMBOL_GPL(smp_call_function_any); void __smp_call_function_single(int cpu, struct call_single_data *data, int wait) { - csd_lock(data); + unsigned int this_cpu; + unsigned long flags; + 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 @@ -387,7 +420,15 @@ void __smp_call_function_single(int cpu, struct call_single_data *data, WARN_ON_ONCE(cpu_online(smp_processor_id()) && wait && irqs_disabled() && !oops_in_progress); - generic_exec_single(cpu, data, wait); + if (cpu == this_cpu) { + local_irq_save(flags); + data->func(data->info); + local_irq_restore(flags); + } else { + csd_lock(data); + generic_exec_single(cpu, data, wait); + } + put_cpu(); } /** @@ -405,7 +446,7 @@ void __smp_call_function_single(int cpu, struct call_single_data *data, * must be disabled when calling this function. */ void smp_call_function_many(const struct cpumask *mask, - void (*func)(void *), void *info, bool wait) + smp_call_func_t func, void *info, bool wait) { struct call_function_data *data; unsigned long flags; @@ -418,7 +459,7 @@ void smp_call_function_many(const struct cpumask *mask, * can't happen. */ WARN_ON_ONCE(cpu_online(this_cpu) && irqs_disabled() - && !oops_in_progress); + && !oops_in_progress && !early_boot_irqs_disabled); /* So, what's a CPU they want? Ignoring this one. */ cpu = cpumask_first_and(mask, cpu_online_mask); @@ -442,11 +483,21 @@ void smp_call_function_many(const struct cpumask *mask, data = &__get_cpu_var(cfd_data); csd_lock(&data->csd); + BUG_ON(atomic_read(&data->refs) || !cpumask_empty(data->cpumask)); data->csd.func = func; data->csd.info = info; cpumask_and(data->cpumask, mask, cpu_online_mask); cpumask_clear_cpu(this_cpu, data->cpumask); + + /* + * To ensure the interrupt handler gets an complete view + * we order the cpumask and refs writes and order the read + * of them in the interrupt handler. In addition we may + * only clear our own cpu bit from the mask. + */ + smp_wmb(); + atomic_set(&data->refs, cpumask_weight(data->cpumask)); raw_spin_lock_irqsave(&call_function.lock, flags); @@ -489,7 +540,7 @@ EXPORT_SYMBOL(smp_call_function_many); * You must not call this function with disabled interrupts or from a * hardware interrupt handler or from a bottom half handler. */ -int smp_call_function(void (*func)(void *), void *info, int wait) +int smp_call_function(smp_call_func_t func, void *info, int wait) { preempt_disable(); smp_call_function_many(cpu_online_mask, func, info, wait); @@ -518,3 +569,24 @@ void ipi_call_unlock_irq(void) { raw_spin_unlock_irq(&call_function.lock); } +#endif /* USE_GENERIC_SMP_HELPERS */ + +/* + * Call a function on all processors. May be used during early boot while + * early_boot_irqs_disabled is set. Use local_irq_save/restore() instead + * of local_irq_disable/enable(). + */ +int on_each_cpu(void (*func) (void *info), void *info, int wait) +{ + unsigned long flags; + int ret = 0; + + preempt_disable(); + ret = smp_call_function(func, info, wait); + local_irq_save(flags); + func(info); + local_irq_restore(flags); + preempt_enable(); + return ret; +} +EXPORT_SYMBOL(on_each_cpu); diff --git a/kernel/softirq.c b/kernel/softirq.c index 0db913a5c60..68eb5efec38 100644 --- a/kernel/softirq.c +++ b/kernel/softirq.c @@ -67,21 +67,31 @@ char *softirq_to_name[NR_SOFTIRQS] = { * to the pending events, so lets the scheduler to balance * the softirq load for us. */ -void wakeup_softirqd(void) +static void wakeup_softirqd(void) { /* Interrupts are disabled: no need to stop preemption */ - struct task_struct *tsk = __get_cpu_var(ksoftirqd); + struct task_struct *tsk = __this_cpu_read(ksoftirqd); if (tsk && tsk->state != TASK_RUNNING) wake_up_process(tsk); } /* + * preempt_count and SOFTIRQ_OFFSET usage: + * - preempt_count is changed by SOFTIRQ_OFFSET on entering or leaving + * softirq processing. + * - preempt_count is changed by SOFTIRQ_DISABLE_OFFSET (= 2 * SOFTIRQ_OFFSET) + * on local_bh_disable or local_bh_enable. + * This lets us distinguish between whether we are currently processing + * softirq and whether we just have bh disabled. + */ + +/* * This one is for softirq.c-internal use, * where hardirqs are disabled legitimately: */ #ifdef CONFIG_TRACE_IRQFLAGS -static void __local_bh_disable(unsigned long ip) +static void __local_bh_disable(unsigned long ip, unsigned int cnt) { unsigned long flags; @@ -95,32 +105,43 @@ static void __local_bh_disable(unsigned long ip) * We must manually increment preempt_count here and manually * call the trace_preempt_off later. */ - preempt_count() += SOFTIRQ_OFFSET; + preempt_count() += cnt; /* * Were softirqs turned off above: */ - if (softirq_count() == SOFTIRQ_OFFSET) + if (softirq_count() == cnt) trace_softirqs_off(ip); raw_local_irq_restore(flags); - if (preempt_count() == SOFTIRQ_OFFSET) + if (preempt_count() == cnt) trace_preempt_off(CALLER_ADDR0, get_parent_ip(CALLER_ADDR1)); } #else /* !CONFIG_TRACE_IRQFLAGS */ -static inline void __local_bh_disable(unsigned long ip) +static inline void __local_bh_disable(unsigned long ip, unsigned int cnt) { - add_preempt_count(SOFTIRQ_OFFSET); + add_preempt_count(cnt); barrier(); } #endif /* CONFIG_TRACE_IRQFLAGS */ void local_bh_disable(void) { - __local_bh_disable((unsigned long)__builtin_return_address(0)); + __local_bh_disable((unsigned long)__builtin_return_address(0), + SOFTIRQ_DISABLE_OFFSET); } EXPORT_SYMBOL(local_bh_disable); +static void __local_bh_enable(unsigned int cnt) +{ + WARN_ON_ONCE(in_irq()); + WARN_ON_ONCE(!irqs_disabled()); + + if (softirq_count() == cnt) + trace_softirqs_on((unsigned long)__builtin_return_address(0)); + sub_preempt_count(cnt); +} + /* * Special-case - softirqs can safely be enabled in * cond_resched_softirq(), or by __do_softirq(), @@ -128,12 +149,7 @@ EXPORT_SYMBOL(local_bh_disable); */ void _local_bh_enable(void) { - WARN_ON_ONCE(in_irq()); - WARN_ON_ONCE(!irqs_disabled()); - - if (softirq_count() == SOFTIRQ_OFFSET) - trace_softirqs_on((unsigned long)__builtin_return_address(0)); - sub_preempt_count(SOFTIRQ_OFFSET); + __local_bh_enable(SOFTIRQ_DISABLE_OFFSET); } EXPORT_SYMBOL(_local_bh_enable); @@ -147,13 +163,13 @@ static inline void _local_bh_enable_ip(unsigned long ip) /* * Are softirqs going to be turned on now: */ - if (softirq_count() == SOFTIRQ_OFFSET) + if (softirq_count() == SOFTIRQ_DISABLE_OFFSET) trace_softirqs_on(ip); /* * Keep preemption disabled until we are done with * softirq processing: */ - sub_preempt_count(SOFTIRQ_OFFSET - 1); + sub_preempt_count(SOFTIRQ_DISABLE_OFFSET - 1); if (unlikely(!in_interrupt() && local_softirq_pending())) do_softirq(); @@ -198,7 +214,8 @@ asmlinkage void __do_softirq(void) pending = local_softirq_pending(); account_system_vtime(current); - __local_bh_disable((unsigned long)__builtin_return_address(0)); + __local_bh_disable((unsigned long)__builtin_return_address(0), + SOFTIRQ_OFFSET); lockdep_softirq_enter(); cpu = smp_processor_id(); @@ -212,18 +229,20 @@ restart: do { if (pending & 1) { + unsigned int vec_nr = h - softirq_vec; int prev_count = preempt_count(); - kstat_incr_softirqs_this_cpu(h - softirq_vec); - trace_softirq_entry(h, softirq_vec); + kstat_incr_softirqs_this_cpu(vec_nr); + + trace_softirq_entry(vec_nr); h->action(h); - trace_softirq_exit(h, softirq_vec); + trace_softirq_exit(vec_nr); if (unlikely(prev_count != preempt_count())) { - printk(KERN_ERR "huh, entered softirq %td %s %p" + printk(KERN_ERR "huh, entered softirq %u %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()); + " exited with %08x?\n", vec_nr, + softirq_to_name[vec_nr], h->action, + prev_count, preempt_count()); preempt_count() = prev_count; } @@ -245,7 +264,7 @@ restart: lockdep_softirq_exit(); account_system_vtime(current); - _local_bh_enable(); + __local_bh_enable(SOFTIRQ_OFFSET); } #ifndef __ARCH_HAS_DO_SOFTIRQ @@ -279,10 +298,16 @@ void irq_enter(void) rcu_irq_enter(); if (idle_cpu(cpu) && !in_interrupt()) { - __irq_enter(); + /* + * Prevent raise_softirq from needlessly waking up ksoftirqd + * here, as softirq will be serviced on return from interrupt. + */ + local_bh_disable(); tick_check_idle(cpu); - } else - __irq_enter(); + _local_bh_enable(); + } + + __irq_enter(); } #ifdef __ARCH_IRQ_EXIT_IRQS_DISABLED @@ -363,8 +388,8 @@ void __tasklet_schedule(struct tasklet_struct *t) local_irq_save(flags); t->next = NULL; - *__get_cpu_var(tasklet_vec).tail = t; - __get_cpu_var(tasklet_vec).tail = &(t->next); + *__this_cpu_read(tasklet_vec.tail) = t; + __this_cpu_write(tasklet_vec.tail, &(t->next)); raise_softirq_irqoff(TASKLET_SOFTIRQ); local_irq_restore(flags); } @@ -377,8 +402,8 @@ void __tasklet_hi_schedule(struct tasklet_struct *t) local_irq_save(flags); t->next = NULL; - *__get_cpu_var(tasklet_hi_vec).tail = t; - __get_cpu_var(tasklet_hi_vec).tail = &(t->next); + *__this_cpu_read(tasklet_hi_vec.tail) = t; + __this_cpu_write(tasklet_hi_vec.tail, &(t->next)); raise_softirq_irqoff(HI_SOFTIRQ); local_irq_restore(flags); } @@ -389,8 +414,8 @@ 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; + t->next = __this_cpu_read(tasklet_hi_vec.head); + __this_cpu_write(tasklet_hi_vec.head, t); __raise_softirq_irqoff(HI_SOFTIRQ); } @@ -401,9 +426,9 @@ static void tasklet_action(struct softirq_action *a) struct tasklet_struct *list; local_irq_disable(); - list = __get_cpu_var(tasklet_vec).head; - __get_cpu_var(tasklet_vec).head = NULL; - __get_cpu_var(tasklet_vec).tail = &__get_cpu_var(tasklet_vec).head; + list = __this_cpu_read(tasklet_vec.head); + __this_cpu_write(tasklet_vec.head, NULL); + __this_cpu_write(tasklet_vec.tail, &__get_cpu_var(tasklet_vec).head); local_irq_enable(); while (list) { @@ -424,8 +449,8 @@ static void tasklet_action(struct softirq_action *a) local_irq_disable(); t->next = NULL; - *__get_cpu_var(tasklet_vec).tail = t; - __get_cpu_var(tasklet_vec).tail = &(t->next); + *__this_cpu_read(tasklet_vec.tail) = t; + __this_cpu_write(tasklet_vec.tail, &(t->next)); __raise_softirq_irqoff(TASKLET_SOFTIRQ); local_irq_enable(); } @@ -436,9 +461,9 @@ static void tasklet_hi_action(struct softirq_action *a) struct tasklet_struct *list; local_irq_disable(); - list = __get_cpu_var(tasklet_hi_vec).head; - __get_cpu_var(tasklet_hi_vec).head = NULL; - __get_cpu_var(tasklet_hi_vec).tail = &__get_cpu_var(tasklet_hi_vec).head; + list = __this_cpu_read(tasklet_hi_vec.head); + __this_cpu_write(tasklet_hi_vec.head, NULL); + __this_cpu_write(tasklet_hi_vec.tail, &__get_cpu_var(tasklet_hi_vec).head); local_irq_enable(); while (list) { @@ -459,8 +484,8 @@ static void tasklet_hi_action(struct softirq_action *a) local_irq_disable(); t->next = NULL; - *__get_cpu_var(tasklet_hi_vec).tail = t; - __get_cpu_var(tasklet_hi_vec).tail = &(t->next); + *__this_cpu_read(tasklet_hi_vec.tail) = t; + __this_cpu_write(tasklet_hi_vec.tail, &(t->next)); __raise_softirq_irqoff(HI_SOFTIRQ); local_irq_enable(); } @@ -696,6 +721,7 @@ static int run_ksoftirqd(void * __bind_cpu) { set_current_state(TASK_INTERRUPTIBLE); + current->flags |= PF_KSOFTIRQD; while (!kthread_should_stop()) { preempt_disable(); if (!local_softirq_pending()) { @@ -776,16 +802,16 @@ static void takeover_tasklets(unsigned int cpu) /* Find end, append list for that CPU. */ if (&per_cpu(tasklet_vec, cpu).head != per_cpu(tasklet_vec, cpu).tail) { - *(__get_cpu_var(tasklet_vec).tail) = per_cpu(tasklet_vec, cpu).head; - __get_cpu_var(tasklet_vec).tail = per_cpu(tasklet_vec, cpu).tail; + *__this_cpu_read(tasklet_vec.tail) = per_cpu(tasklet_vec, cpu).head; + this_cpu_write(tasklet_vec.tail, per_cpu(tasklet_vec, cpu).tail); per_cpu(tasklet_vec, cpu).head = NULL; per_cpu(tasklet_vec, cpu).tail = &per_cpu(tasklet_vec, cpu).head; } raise_softirq_irqoff(TASKLET_SOFTIRQ); if (&per_cpu(tasklet_hi_vec, cpu).head != per_cpu(tasklet_hi_vec, cpu).tail) { - *__get_cpu_var(tasklet_hi_vec).tail = per_cpu(tasklet_hi_vec, cpu).head; - __get_cpu_var(tasklet_hi_vec).tail = per_cpu(tasklet_hi_vec, cpu).tail; + *__this_cpu_read(tasklet_hi_vec.tail) = per_cpu(tasklet_hi_vec, cpu).head; + __this_cpu_write(tasklet_hi_vec.tail, per_cpu(tasklet_hi_vec, cpu).tail); per_cpu(tasklet_hi_vec, cpu).head = NULL; per_cpu(tasklet_hi_vec, cpu).tail = &per_cpu(tasklet_hi_vec, cpu).head; } @@ -808,7 +834,7 @@ static int __cpuinit cpu_callback(struct notifier_block *nfb, p = kthread_create(run_ksoftirqd, hcpu, "ksoftirqd/%d", hotcpu); if (IS_ERR(p)) { printk("ksoftirqd for %i failed\n", hotcpu); - return NOTIFY_BAD; + return notifier_from_errno(PTR_ERR(p)); } kthread_bind(p, hotcpu); per_cpu(ksoftirqd, hotcpu) = p; @@ -827,7 +853,9 @@ static int __cpuinit cpu_callback(struct notifier_block *nfb, cpumask_any(cpu_online_mask)); case CPU_DEAD: case CPU_DEAD_FROZEN: { - struct sched_param param = { .sched_priority = MAX_RT_PRIO-1 }; + static const struct sched_param param = { + .sched_priority = MAX_RT_PRIO-1 + }; p = per_cpu(ksoftirqd, hotcpu); per_cpu(ksoftirqd, hotcpu) = NULL; @@ -850,32 +878,13 @@ static __init int spawn_ksoftirqd(void) void *cpu = (void *)(long)smp_processor_id(); int err = cpu_callback(&cpu_nfb, CPU_UP_PREPARE, cpu); - BUG_ON(err == NOTIFY_BAD); + BUG_ON(err != NOTIFY_OK); cpu_callback(&cpu_nfb, CPU_ONLINE, cpu); register_cpu_notifier(&cpu_nfb); return 0; } early_initcall(spawn_ksoftirqd); -#ifdef CONFIG_SMP -/* - * Call a function on all processors - */ -int on_each_cpu(void (*func) (void *info), void *info, int wait) -{ - int ret = 0; - - preempt_disable(); - ret = smp_call_function(func, info, wait); - local_irq_disable(); - func(info); - local_irq_enable(); - preempt_enable(); - return ret; -} -EXPORT_SYMBOL(on_each_cpu); -#endif - /* * [ These __weak aliases are kept in a separate compilation unit, so that * GCC does not inline them incorrectly. ] @@ -886,17 +895,14 @@ int __init __weak early_irq_init(void) return 0; } +#ifdef CONFIG_GENERIC_HARDIRQS int __init __weak arch_probe_nr_irqs(void) { - return 0; + return NR_IRQS_LEGACY; } int __init __weak arch_early_irq_init(void) { return 0; } - -int __weak arch_init_chip_data(struct irq_desc *desc, int node) -{ - return 0; -} +#endif diff --git a/kernel/softlockup.c b/kernel/softlockup.c deleted file mode 100644 index 4b493f67dcb..00000000000 --- a/kernel/softlockup.c +++ /dev/null @@ -1,293 +0,0 @@ -/* - * Detect Soft Lockups - * - * started by Ingo Molnar, Copyright (C) 2005, 2006 Red Hat, Inc. - * - * this code detects soft lockups: incidents in where on a CPU - * the kernel does not reschedule for 10 seconds or more. - */ -#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/notifier.h> -#include <linux/module.h> -#include <linux/sysctl.h> - -#include <asm/irq_regs.h> - -static DEFINE_SPINLOCK(print_lock); - -static DEFINE_PER_CPU(unsigned long, softlockup_touch_ts); /* touch timestamp */ -static DEFINE_PER_CPU(unsigned long, softlockup_print_ts); /* print timestamp */ -static DEFINE_PER_CPU(struct task_struct *, softlockup_watchdog); -static DEFINE_PER_CPU(bool, softlock_touch_sync); - -static int __read_mostly did_panic; -int __read_mostly softlockup_thresh = 60; - -/* - * Should we panic (and reboot, if panic_timeout= is set) when a - * soft-lockup occurs: - */ -unsigned int __read_mostly softlockup_panic = - CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC_VALUE; - -static int __init softlockup_panic_setup(char *str) -{ - softlockup_panic = simple_strtoul(str, NULL, 0); - - return 1; -} -__setup("softlockup_panic=", softlockup_panic_setup); - -static int -softlock_panic(struct notifier_block *this, unsigned long event, void *ptr) -{ - did_panic = 1; - - return NOTIFY_DONE; -} - -static struct notifier_block panic_block = { - .notifier_call = softlock_panic, -}; - -/* - * Returns seconds, approximately. We don't need nanosecond - * resolution, and we don't need to waste time with a big divide when - * 2^30ns == 1.074s. - */ -static unsigned long get_timestamp(int this_cpu) -{ - return cpu_clock(this_cpu) >> 30LL; /* 2^30 ~= 10^9 */ -} - -static void __touch_softlockup_watchdog(void) -{ - int this_cpu = raw_smp_processor_id(); - - __raw_get_cpu_var(softlockup_touch_ts) = get_timestamp(this_cpu); -} - -void touch_softlockup_watchdog(void) -{ - __raw_get_cpu_var(softlockup_touch_ts) = 0; -} -EXPORT_SYMBOL(touch_softlockup_watchdog); - -void touch_softlockup_watchdog_sync(void) -{ - __raw_get_cpu_var(softlock_touch_sync) = true; - __raw_get_cpu_var(softlockup_touch_ts) = 0; -} - -void touch_all_softlockup_watchdogs(void) -{ - int cpu; - - /* Cause each CPU to re-update its timestamp rather than complain */ - for_each_online_cpu(cpu) - per_cpu(softlockup_touch_ts, cpu) = 0; -} -EXPORT_SYMBOL(touch_all_softlockup_watchdogs); - -int proc_dosoftlockup_thresh(struct ctl_table *table, int write, - void __user *buffer, - size_t *lenp, loff_t *ppos) -{ - touch_all_softlockup_watchdogs(); - return proc_dointvec_minmax(table, write, buffer, lenp, ppos); -} - -/* - * This callback runs from the timer interrupt, and checks - * whether the watchdog thread has hung or not: - */ -void softlockup_tick(void) -{ - int this_cpu = smp_processor_id(); - unsigned long touch_ts = per_cpu(softlockup_touch_ts, this_cpu); - unsigned long print_ts; - struct pt_regs *regs = get_irq_regs(); - unsigned long now; - - /* Is detection switched off? */ - if (!per_cpu(softlockup_watchdog, this_cpu) || softlockup_thresh <= 0) { - /* Be sure we don't false trigger if switched back on */ - if (touch_ts) - per_cpu(softlockup_touch_ts, this_cpu) = 0; - return; - } - - if (touch_ts == 0) { - if (unlikely(per_cpu(softlock_touch_sync, this_cpu))) { - /* - * If the time stamp was touched atomically - * make sure the scheduler tick is up to date. - */ - per_cpu(softlock_touch_sync, this_cpu) = false; - sched_clock_tick(); - } - __touch_softlockup_watchdog(); - return; - } - - print_ts = per_cpu(softlockup_print_ts, this_cpu); - - /* report at most once a second */ - if (print_ts == touch_ts || did_panic) - return; - - /* do not print during early bootup: */ - if (unlikely(system_state != SYSTEM_RUNNING)) { - __touch_softlockup_watchdog(); - return; - } - - now = get_timestamp(this_cpu); - - /* - * Wake up the high-prio watchdog task twice per - * threshold timespan. - */ - if (time_after(now - softlockup_thresh/2, touch_ts)) - wake_up_process(per_cpu(softlockup_watchdog, this_cpu)); - - /* Warn about unreasonable delays: */ - if (time_before_eq(now - softlockup_thresh, touch_ts)) - return; - - per_cpu(softlockup_print_ts, this_cpu) = touch_ts; - - spin_lock(&print_lock); - printk(KERN_ERR "BUG: soft lockup - CPU#%d stuck for %lus! [%s:%d]\n", - this_cpu, now - touch_ts, - current->comm, task_pid_nr(current)); - print_modules(); - print_irqtrace_events(current); - if (regs) - show_regs(regs); - else - dump_stack(); - spin_unlock(&print_lock); - - if (softlockup_panic) - panic("softlockup: hung tasks"); -} - -/* - * 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 }; - - sched_setscheduler(current, SCHED_FIFO, ¶m); - - /* initialize timestamp */ - __touch_softlockup_watchdog(); - - set_current_state(TASK_INTERRUPTIBLE); - /* - * Run briefly once per second to reset the softlockup timestamp. - * If this gets delayed for more than 60 seconds then the - * debug-printout triggers in softlockup_tick(). - */ - while (!kthread_should_stop()) { - __touch_softlockup_watchdog(); - schedule(); - - if (kthread_should_stop()) - break; - - set_current_state(TASK_INTERRUPTIBLE); - } - __set_current_state(TASK_RUNNING); - - return 0; -} - -/* - * Create/destroy watchdog threads as CPUs come and go: - */ -static int __cpuinit -cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu) -{ - int hotcpu = (unsigned long)hcpu; - struct task_struct *p; - - switch (action) { - case CPU_UP_PREPARE: - case CPU_UP_PREPARE_FROZEN: - BUG_ON(per_cpu(softlockup_watchdog, hotcpu)); - p = kthread_create(watchdog, hcpu, "watchdog/%d", hotcpu); - if (IS_ERR(p)) { - printk(KERN_ERR "watchdog for %i failed\n", hotcpu); - return NOTIFY_BAD; - } - per_cpu(softlockup_touch_ts, hotcpu) = 0; - per_cpu(softlockup_watchdog, hotcpu) = p; - kthread_bind(p, hotcpu); - break; - case CPU_ONLINE: - case CPU_ONLINE_FROZEN: - wake_up_process(per_cpu(softlockup_watchdog, hotcpu)); - break; -#ifdef CONFIG_HOTPLUG_CPU - case CPU_UP_CANCELED: - case CPU_UP_CANCELED_FROZEN: - if (!per_cpu(softlockup_watchdog, hotcpu)) - break; - /* Unbind so it can run. Fall thru. */ - kthread_bind(per_cpu(softlockup_watchdog, hotcpu), - cpumask_any(cpu_online_mask)); - case CPU_DEAD: - case CPU_DEAD_FROZEN: - p = per_cpu(softlockup_watchdog, hotcpu); - per_cpu(softlockup_watchdog, hotcpu) = NULL; - kthread_stop(p); - break; -#endif /* CONFIG_HOTPLUG_CPU */ - } - return NOTIFY_OK; -} - -static struct notifier_block __cpuinitdata cpu_nfb = { - .notifier_call = cpu_callback -}; - -static int __initdata nosoftlockup; - -static int __init nosoftlockup_setup(char *str) -{ - nosoftlockup = 1; - return 1; -} -__setup("nosoftlockup", nosoftlockup_setup); - -static int __init spawn_softlockup_task(void) -{ - void *cpu = (void *)(long)smp_processor_id(); - int err; - - if (nosoftlockup) - return 0; - - err = cpu_callback(&cpu_nfb, CPU_UP_PREPARE, cpu); - if (err == NOTIFY_BAD) { - BUG(); - return 1; - } - cpu_callback(&cpu_nfb, CPU_ONLINE, cpu); - register_cpu_notifier(&cpu_nfb); - - atomic_notifier_chain_register(&panic_notifier_list, &panic_block); - - return 0; -} -early_initcall(spawn_softlockup_task); diff --git a/kernel/srcu.c b/kernel/srcu.c index 2980da3fd50..73ce23feaea 100644 --- a/kernel/srcu.c +++ b/kernel/srcu.c @@ -31,6 +31,7 @@ #include <linux/rcupdate.h> #include <linux/sched.h> #include <linux/smp.h> +#include <linux/delay.h> #include <linux/srcu.h> static int init_srcu_struct_fields(struct srcu_struct *sp) @@ -46,11 +47,9 @@ static int init_srcu_struct_fields(struct srcu_struct *sp) int __init_srcu_struct(struct srcu_struct *sp, const char *name, struct lock_class_key *key) { -#ifdef CONFIG_DEBUG_LOCK_ALLOC /* Don't re-initialize a lock while it is held. */ debug_check_no_locks_freed((void *)sp, sizeof(*sp)); lockdep_init_map(&sp->dep_map, name, key, 0); -#endif /* #ifdef CONFIG_DEBUG_LOCK_ALLOC */ return init_srcu_struct_fields(sp); } EXPORT_SYMBOL_GPL(__init_srcu_struct); @@ -157,6 +156,16 @@ void __srcu_read_unlock(struct srcu_struct *sp, int idx) EXPORT_SYMBOL_GPL(__srcu_read_unlock); /* + * We use an adaptive strategy for synchronize_srcu() and especially for + * synchronize_srcu_expedited(). We spin for a fixed time period + * (defined below) to allow SRCU readers to exit their read-side critical + * sections. If there are still some readers after 10 microseconds, + * we repeatedly block for 1-millisecond time periods. This approach + * has done well in testing, so there is no need for a config parameter. + */ +#define SYNCHRONIZE_SRCU_READER_DELAY 10 + +/* * Helper function for synchronize_srcu() and synchronize_srcu_expedited(). */ static void __synchronize_srcu(struct srcu_struct *sp, void (*sync_func)(void)) @@ -205,9 +214,15 @@ static void __synchronize_srcu(struct srcu_struct *sp, void (*sync_func)(void)) * all srcu_read_lock() calls using the old counters have completed. * Their corresponding critical sections might well be still * executing, but the srcu_read_lock() primitives themselves - * will have finished executing. + * will have finished executing. We initially give readers + * an arbitrarily chosen 10 microseconds to get out of their + * SRCU read-side critical sections, then loop waiting 1/HZ + * seconds per iteration. The 10-microsecond value has done + * very well in testing. */ + if (srcu_readers_active_idx(sp, idx)) + udelay(SYNCHRONIZE_SRCU_READER_DELAY); while (srcu_readers_active_idx(sp, idx)) schedule_timeout_interruptible(1); diff --git a/kernel/stop_machine.c b/kernel/stop_machine.c index b4e7431e7c7..2df820b03be 100644 --- a/kernel/stop_machine.c +++ b/kernel/stop_machine.c @@ -35,9 +35,9 @@ struct cpu_stop_done { /* the actual stopper, one per every possible cpu, enabled on online cpus */ struct cpu_stopper { spinlock_t lock; + bool enabled; /* is this stopper enabled? */ struct list_head works; /* list of pending works */ struct task_struct *thread; /* stopper thread */ - bool enabled; /* is this stopper enabled? */ }; static DEFINE_PER_CPU(struct cpu_stopper, cpu_stopper); @@ -262,7 +262,7 @@ repeat: cpu_stop_fn_t fn = work->fn; void *arg = work->arg; struct cpu_stop_done *done = work->done; - char ksym_buf[KSYM_NAME_LEN]; + char ksym_buf[KSYM_NAME_LEN] __maybe_unused; __set_current_state(TASK_RUNNING); @@ -287,11 +287,12 @@ repeat: goto repeat; } +extern void sched_set_stop_task(int cpu, struct task_struct *stop); + /* manage stopper for a cpu, mostly lifted from sched migration thread mgmt */ static int __cpuinit cpu_stop_cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu) { - struct sched_param param = { .sched_priority = MAX_RT_PRIO - 1 }; unsigned int cpu = (unsigned long)hcpu; struct cpu_stopper *stopper = &per_cpu(cpu_stopper, cpu); struct task_struct *p; @@ -303,14 +304,14 @@ static int __cpuinit cpu_stop_cpu_callback(struct notifier_block *nfb, p = kthread_create(cpu_stopper_thread, stopper, "migration/%d", cpu); if (IS_ERR(p)) - return NOTIFY_BAD; - sched_setscheduler_nocheck(p, SCHED_FIFO, ¶m); + return notifier_from_errno(PTR_ERR(p)); get_task_struct(p); + kthread_bind(p, cpu); + sched_set_stop_task(cpu, p); stopper->thread = p; break; case CPU_ONLINE: - kthread_bind(stopper->thread, cpu); /* strictly unnecessary, as first user will wake it */ wake_up_process(stopper->thread); /* mark enabled */ @@ -321,10 +322,11 @@ static int __cpuinit cpu_stop_cpu_callback(struct notifier_block *nfb, #ifdef CONFIG_HOTPLUG_CPU case CPU_UP_CANCELED: - case CPU_DEAD: + case CPU_POST_DEAD: { struct cpu_stop_work *work; + sched_set_stop_task(cpu, NULL); /* kill the stopper */ kthread_stop(stopper->thread); /* drain remaining works */ @@ -370,7 +372,7 @@ static int __init cpu_stop_init(void) /* start one for the boot cpu */ err = cpu_stop_cpu_callback(&cpu_stop_cpu_notifier, CPU_UP_PREPARE, bcpu); - BUG_ON(err == NOTIFY_BAD); + BUG_ON(err != NOTIFY_OK); cpu_stop_cpu_callback(&cpu_stop_cpu_notifier, CPU_ONLINE, bcpu); register_cpu_notifier(&cpu_stop_cpu_notifier); diff --git a/kernel/sys.c b/kernel/sys.c index 0d36d889c74..31b71a276b4 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -43,6 +43,8 @@ #include <linux/kprobes.h> #include <linux/user_namespace.h> +#include <linux/kmsg_dump.h> + #include <asm/uaccess.h> #include <asm/io.h> #include <asm/unistd.h> @@ -285,6 +287,7 @@ out_unlock: */ void emergency_restart(void) { + kmsg_dump(KMSG_DUMP_EMERG); machine_emergency_restart(); } EXPORT_SYMBOL_GPL(emergency_restart); @@ -312,6 +315,7 @@ void kernel_restart(char *cmd) printk(KERN_EMERG "Restarting system.\n"); else printk(KERN_EMERG "Restarting system with command '%s'.\n", cmd); + kmsg_dump(KMSG_DUMP_RESTART); machine_restart(cmd); } EXPORT_SYMBOL_GPL(kernel_restart); @@ -333,6 +337,7 @@ void kernel_halt(void) kernel_shutdown_prepare(SYSTEM_HALT); sysdev_shutdown(); printk(KERN_EMERG "System halted.\n"); + kmsg_dump(KMSG_DUMP_HALT); machine_halt(); } @@ -351,6 +356,7 @@ void kernel_power_off(void) disable_nonboot_cpus(); sysdev_shutdown(); printk(KERN_EMERG "Power down.\n"); + kmsg_dump(KMSG_DUMP_POWEROFF); machine_power_off(); } EXPORT_SYMBOL_GPL(kernel_power_off); @@ -931,6 +937,7 @@ SYSCALL_DEFINE2(setpgid, pid_t, pid, pid_t, pgid) pgid = pid; if (pgid < 0) return -EINVAL; + rcu_read_lock(); /* From this point forward we keep holding onto the tasklist lock * so that our parent does not change from under us. -DaveM @@ -984,6 +991,7 @@ SYSCALL_DEFINE2(setpgid, pid_t, pid, pid_t, pgid) out: /* All paths lead to here, thus we are safe. -DaveM */ write_unlock_irq(&tasklist_lock); + rcu_read_unlock(); return err; } @@ -1078,8 +1086,10 @@ SYSCALL_DEFINE0(setsid) err = session; out: write_unlock_irq(&tasklist_lock); - if (err > 0) + if (err > 0) { proc_sid_connector(group_leader); + sched_autogroup_create_attach(group_leader); + } return err; } @@ -1236,15 +1246,14 @@ SYSCALL_DEFINE2(setdomainname, char __user *, name, int, len) SYSCALL_DEFINE2(getrlimit, unsigned int, resource, struct rlimit __user *, rlim) { - if (resource >= RLIM_NLIMITS) - return -EINVAL; - else { - struct rlimit value; - task_lock(current->group_leader); - value = current->signal->rlim[resource]; - task_unlock(current->group_leader); - return copy_to_user(rlim, &value, sizeof(*rlim)) ? -EFAULT : 0; - } + struct rlimit value; + int ret; + + ret = do_prlimit(current, resource, NULL, &value); + if (!ret) + ret = copy_to_user(rlim, &value, sizeof(*rlim)) ? -EFAULT : 0; + + return ret; } #ifdef __ARCH_WANT_SYS_OLD_GETRLIMIT @@ -1272,44 +1281,89 @@ SYSCALL_DEFINE2(old_getrlimit, unsigned int, resource, #endif -SYSCALL_DEFINE2(setrlimit, unsigned int, resource, struct rlimit __user *, rlim) +static inline bool rlim64_is_infinity(__u64 rlim64) { - struct rlimit new_rlim, *old_rlim; - int retval; +#if BITS_PER_LONG < 64 + return rlim64 >= ULONG_MAX; +#else + return rlim64 == RLIM64_INFINITY; +#endif +} + +static void rlim_to_rlim64(const struct rlimit *rlim, struct rlimit64 *rlim64) +{ + if (rlim->rlim_cur == RLIM_INFINITY) + rlim64->rlim_cur = RLIM64_INFINITY; + else + rlim64->rlim_cur = rlim->rlim_cur; + if (rlim->rlim_max == RLIM_INFINITY) + rlim64->rlim_max = RLIM64_INFINITY; + else + rlim64->rlim_max = rlim->rlim_max; +} + +static void rlim64_to_rlim(const struct rlimit64 *rlim64, struct rlimit *rlim) +{ + if (rlim64_is_infinity(rlim64->rlim_cur)) + rlim->rlim_cur = RLIM_INFINITY; + else + rlim->rlim_cur = (unsigned long)rlim64->rlim_cur; + if (rlim64_is_infinity(rlim64->rlim_max)) + rlim->rlim_max = RLIM_INFINITY; + else + rlim->rlim_max = (unsigned long)rlim64->rlim_max; +} + +/* make sure you are allowed to change @tsk limits before calling this */ +int do_prlimit(struct task_struct *tsk, unsigned int resource, + struct rlimit *new_rlim, struct rlimit *old_rlim) +{ + struct rlimit *rlim; + int retval = 0; if (resource >= RLIM_NLIMITS) return -EINVAL; - if (copy_from_user(&new_rlim, rlim, sizeof(*rlim))) - return -EFAULT; - if (new_rlim.rlim_cur > new_rlim.rlim_max) - return -EINVAL; - old_rlim = current->signal->rlim + resource; - if ((new_rlim.rlim_max > old_rlim->rlim_max) && - !capable(CAP_SYS_RESOURCE)) - return -EPERM; - if (resource == RLIMIT_NOFILE && new_rlim.rlim_max > sysctl_nr_open) - return -EPERM; - - retval = security_task_setrlimit(resource, &new_rlim); - if (retval) - return retval; - - if (resource == RLIMIT_CPU && new_rlim.rlim_cur == 0) { - /* - * The caller is asking for an immediate RLIMIT_CPU - * expiry. But we use the zero value to mean "it was - * never set". So let's cheat and make it one second - * instead - */ - new_rlim.rlim_cur = 1; + if (new_rlim) { + if (new_rlim->rlim_cur > new_rlim->rlim_max) + return -EINVAL; + if (resource == RLIMIT_NOFILE && + new_rlim->rlim_max > sysctl_nr_open) + return -EPERM; } - task_lock(current->group_leader); - *old_rlim = new_rlim; - task_unlock(current->group_leader); - - if (resource != RLIMIT_CPU) + /* protect tsk->signal and tsk->sighand from disappearing */ + read_lock(&tasklist_lock); + if (!tsk->sighand) { + retval = -ESRCH; goto out; + } + + rlim = tsk->signal->rlim + resource; + task_lock(tsk->group_leader); + if (new_rlim) { + if (new_rlim->rlim_max > rlim->rlim_max && + !capable(CAP_SYS_RESOURCE)) + retval = -EPERM; + if (!retval) + retval = security_task_setrlimit(tsk->group_leader, + resource, new_rlim); + if (resource == RLIMIT_CPU && new_rlim->rlim_cur == 0) { + /* + * The caller is asking for an immediate RLIMIT_CPU + * expiry. But we use the zero value to mean "it was + * never set". So let's cheat and make it one second + * instead + */ + new_rlim->rlim_cur = 1; + } + } + if (!retval) { + if (old_rlim) + *old_rlim = *rlim; + if (new_rlim) + *rlim = *new_rlim; + } + task_unlock(tsk->group_leader); /* * RLIMIT_CPU handling. Note that the kernel fails to return an error @@ -1317,14 +1371,84 @@ SYSCALL_DEFINE2(setrlimit, unsigned int, resource, struct rlimit __user *, rlim) * very long-standing error, and fixing it now risks breakage of * applications, so we live with it */ - if (new_rlim.rlim_cur == RLIM_INFINITY) - goto out; - - update_rlimit_cpu(new_rlim.rlim_cur); + if (!retval && new_rlim && resource == RLIMIT_CPU && + new_rlim->rlim_cur != RLIM_INFINITY) + update_rlimit_cpu(tsk, new_rlim->rlim_cur); out: + read_unlock(&tasklist_lock); + return retval; +} + +/* rcu lock must be held */ +static int check_prlimit_permission(struct task_struct *task) +{ + const struct cred *cred = current_cred(), *tcred; + + tcred = __task_cred(task); + if ((cred->uid != tcred->euid || + cred->uid != tcred->suid || + cred->uid != tcred->uid || + cred->gid != tcred->egid || + cred->gid != tcred->sgid || + cred->gid != tcred->gid) && + !capable(CAP_SYS_RESOURCE)) { + return -EPERM; + } + return 0; } +SYSCALL_DEFINE4(prlimit64, pid_t, pid, unsigned int, resource, + const struct rlimit64 __user *, new_rlim, + struct rlimit64 __user *, old_rlim) +{ + struct rlimit64 old64, new64; + struct rlimit old, new; + struct task_struct *tsk; + int ret; + + if (new_rlim) { + if (copy_from_user(&new64, new_rlim, sizeof(new64))) + return -EFAULT; + rlim64_to_rlim(&new64, &new); + } + + rcu_read_lock(); + tsk = pid ? find_task_by_vpid(pid) : current; + if (!tsk) { + rcu_read_unlock(); + return -ESRCH; + } + ret = check_prlimit_permission(tsk); + if (ret) { + rcu_read_unlock(); + return ret; + } + get_task_struct(tsk); + rcu_read_unlock(); + + ret = do_prlimit(tsk, resource, new_rlim ? &new : NULL, + old_rlim ? &old : NULL); + + if (!ret && old_rlim) { + rlim_to_rlim64(&old, &old64); + if (copy_to_user(old_rlim, &old64, sizeof(old64))) + ret = -EFAULT; + } + + put_task_struct(tsk); + return ret; +} + +SYSCALL_DEFINE2(setrlimit, unsigned int, resource, struct rlimit __user *, rlim) +{ + struct rlimit new_rlim; + + if (copy_from_user(&new_rlim, rlim, sizeof(*rlim))) + return -EFAULT; + return do_prlimit(current, resource, &new_rlim, NULL); +} + /* * It would make sense to put struct rusage in the task_struct, * except that would make the task_struct be *really big*. After @@ -1632,9 +1756,9 @@ SYSCALL_DEFINE3(getcpu, unsigned __user *, cpup, unsigned __user *, nodep, char poweroff_cmd[POWEROFF_CMD_PATH_LEN] = "/sbin/poweroff"; -static void argv_cleanup(char **argv, char **envp) +static void argv_cleanup(struct subprocess_info *info) { - argv_free(argv); + argv_free(info->argv); } /** @@ -1668,7 +1792,7 @@ int orderly_poweroff(bool force) goto out; } - call_usermodehelper_setcleanup(info, argv_cleanup); + call_usermodehelper_setfns(info, NULL, argv_cleanup, NULL); ret = call_usermodehelper_exec(info, UMH_NO_WAIT); diff --git a/kernel/sys_ni.c b/kernel/sys_ni.c index 70f2ea758ff..c782fe9924c 100644 --- a/kernel/sys_ni.c +++ b/kernel/sys_ni.c @@ -50,6 +50,7 @@ cond_syscall(compat_sys_sendmsg); cond_syscall(sys_recvmsg); cond_syscall(sys_recvmmsg); cond_syscall(compat_sys_recvmsg); +cond_syscall(compat_sys_recv); cond_syscall(compat_sys_recvfrom); cond_syscall(compat_sys_recvmmsg); cond_syscall(sys_socketcall); @@ -181,3 +182,7 @@ cond_syscall(sys_eventfd2); /* performance counters: */ cond_syscall(sys_perf_event_open); + +/* fanotify! */ +cond_syscall(sys_fanotify_init); +cond_syscall(sys_fanotify_mark); diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 4c93486b45d..0f1bd83db98 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -24,6 +24,7 @@ #include <linux/slab.h> #include <linux/sysctl.h> #include <linux/signal.h> +#include <linux/printk.h> #include <linux/proc_fs.h> #include <linux/security.h> #include <linux/ctype.h> @@ -37,22 +38,24 @@ #include <linux/highuid.h> #include <linux/writeback.h> #include <linux/ratelimit.h> +#include <linux/compaction.h> #include <linux/hugetlb.h> #include <linux/initrd.h> #include <linux/key.h> #include <linux/times.h> #include <linux/limits.h> #include <linux/dcache.h> +#include <linux/dnotify.h> #include <linux/syscalls.h> #include <linux/vmstat.h> #include <linux/nfs_fs.h> #include <linux/acpi.h> #include <linux/reboot.h> #include <linux/ftrace.h> -#include <linux/slow-work.h> #include <linux/perf_event.h> #include <linux/kprobes.h> #include <linux/pipe_fs_i.h> +#include <linux/oom.h> #include <asm/uaccess.h> #include <asm/processor.h> @@ -75,15 +78,16 @@ #include <scsi/sg.h> #endif +#ifdef CONFIG_LOCKUP_DETECTOR +#include <linux/nmi.h> +#endif + #if defined(CONFIG_SYSCTL) /* External variables not in a header file. */ extern int sysctl_overcommit_memory; extern int sysctl_overcommit_ratio; -extern int sysctl_panic_on_oom; -extern int sysctl_oom_kill_allocating_task; -extern int sysctl_oom_dump_tasks; extern int max_threads; extern int core_uses_pid; extern int suid_dumpable; @@ -105,7 +109,7 @@ extern int blk_iopoll_enabled; #endif /* Constants used for minimum and maximum */ -#ifdef CONFIG_DETECT_SOFTLOCKUP +#ifdef CONFIG_LOCKUP_DETECTOR static int sixty = 60; static int neg_one = -1; #endif @@ -129,6 +133,9 @@ static int min_percpu_pagelist_fract = 8; static int ngroups_max = NGROUPS_MAX; +#ifdef CONFIG_INOTIFY_USER +#include <linux/inotify.h> +#endif #ifdef CONFIG_SPARC #include <asm/system.h> #endif @@ -155,8 +162,6 @@ extern int no_unaligned_warning; extern int unaligned_dump_stack; #endif -extern struct ratelimit_state printk_ratelimit_state; - #ifdef CONFIG_PROC_SYSCTL static int proc_do_cad_pid(struct ctl_table *table, int write, void __user *buffer, size_t *lenp, loff_t *ppos); @@ -165,7 +170,8 @@ static int proc_taint(struct ctl_table *table, int write, #endif #ifdef CONFIG_MAGIC_SYSRQ -static int __sysrq_enabled; /* Note: sysrq code ises it's own private copy */ +/* Note: sysrq code uses it's own private copy */ +static int __sysrq_enabled = SYSRQ_DEFAULT_ENABLE; static int sysrq_sysctl_handler(ctl_table *table, int write, void __user *buffer, size_t *lenp, @@ -205,9 +211,6 @@ static struct ctl_table fs_table[]; static struct ctl_table debug_table[]; static struct ctl_table dev_table[]; extern struct ctl_table random_table[]; -#ifdef CONFIG_INOTIFY_USER -extern struct ctl_table inotify_table[]; -#endif #ifdef CONFIG_EPOLL extern struct ctl_table epoll_table[]; #endif @@ -244,10 +247,6 @@ static struct ctl_table root_table[] = { .mode = 0555, .child = dev_table, }, -/* - * NOTE: do not add new entries to this table unless you have read - * Documentation/sysctl/ctl_unnumbered.txt - */ { } }; @@ -258,8 +257,11 @@ static int min_wakeup_granularity_ns; /* 0 usecs */ static int max_wakeup_granularity_ns = NSEC_PER_SEC; /* 1 second */ static int min_sched_tunable_scaling = SCHED_TUNABLESCALING_NONE; static int max_sched_tunable_scaling = SCHED_TUNABLESCALING_END-1; -static int min_sched_shares_ratelimit = 100000; /* 100 usec */ -static int max_sched_shares_ratelimit = NSEC_PER_SEC; /* 1 second */ +#endif + +#ifdef CONFIG_COMPACTION +static int min_extfrag_threshold; +static int max_extfrag_threshold = 1000; #endif static struct ctl_table kern_table[] = { @@ -299,15 +301,6 @@ static struct ctl_table kern_table[] = { .extra2 = &max_wakeup_granularity_ns, }, { - .procname = "sched_shares_ratelimit", - .data = &sysctl_sched_shares_ratelimit, - .maxlen = sizeof(unsigned int), - .mode = 0644, - .proc_handler = sched_proc_update_handler, - .extra1 = &min_sched_shares_ratelimit, - .extra2 = &max_sched_shares_ratelimit, - }, - { .procname = "sched_tunable_scaling", .data = &sysctl_sched_tunable_scaling, .maxlen = sizeof(enum sched_tunable_scaling), @@ -317,14 +310,6 @@ static struct ctl_table kern_table[] = { .extra2 = &max_sched_tunable_scaling, }, { - .procname = "sched_shares_thresh", - .data = &sysctl_sched_shares_thresh, - .maxlen = sizeof(unsigned int), - .mode = 0644, - .proc_handler = proc_dointvec_minmax, - .extra1 = &zero, - }, - { .procname = "sched_migration_cost", .data = &sysctl_sched_migration_cost, .maxlen = sizeof(unsigned int), @@ -346,6 +331,13 @@ static struct ctl_table kern_table[] = { .proc_handler = proc_dointvec, }, { + .procname = "sched_shares_window", + .data = &sysctl_sched_shares_window, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = proc_dointvec, + }, + { .procname = "timer_migration", .data = &sysctl_timer_migration, .maxlen = sizeof(unsigned int), @@ -376,6 +368,17 @@ static struct ctl_table kern_table[] = { .mode = 0644, .proc_handler = proc_dointvec, }, +#ifdef CONFIG_SCHED_AUTOGROUP + { + .procname = "sched_autogroup_enabled", + .data = &sysctl_sched_autogroup_enabled, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = proc_dointvec, + .extra1 = &zero, + .extra2 = &one, + }, +#endif #ifdef CONFIG_PROVE_LOCKING { .procname = "prove_locking", @@ -556,7 +559,7 @@ static struct ctl_table kern_table[] = { .extra2 = &one, }, #endif -#if defined(CONFIG_HOTPLUG) && defined(CONFIG_NET) +#ifdef CONFIG_HOTPLUG { .procname = "hotplug", .data = &uevent_helper, @@ -696,6 +699,24 @@ static struct ctl_table kern_table[] = { .extra1 = &zero, .extra2 = &ten_thousand, }, + { + .procname = "dmesg_restrict", + .data = &dmesg_restrict, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec_minmax, + .extra1 = &zero, + .extra2 = &one, + }, + { + .procname = "kptr_restrict", + .data = &kptr_restrict, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec_minmax, + .extra1 = &zero, + .extra2 = &two, + }, #endif { .procname = "ngroups_max", @@ -704,20 +725,47 @@ static struct ctl_table kern_table[] = { .mode = 0444, .proc_handler = proc_dointvec, }, -#if defined(CONFIG_X86_LOCAL_APIC) && defined(CONFIG_X86) +#if defined(CONFIG_LOCKUP_DETECTOR) { - .procname = "unknown_nmi_panic", - .data = &unknown_nmi_panic, + .procname = "watchdog", + .data = &watchdog_enabled, .maxlen = sizeof (int), .mode = 0644, - .proc_handler = proc_dointvec, + .proc_handler = proc_dowatchdog_enabled, + }, + { + .procname = "watchdog_thresh", + .data = &softlockup_thresh, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dowatchdog_thresh, + .extra1 = &neg_one, + .extra2 = &sixty, + }, + { + .procname = "softlockup_panic", + .data = &softlockup_panic, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec_minmax, + .extra1 = &zero, + .extra2 = &one, }, { .procname = "nmi_watchdog", - .data = &nmi_watchdog_enabled, + .data = &watchdog_enabled, + .maxlen = sizeof (int), + .mode = 0644, + .proc_handler = proc_dowatchdog_enabled, + }, +#endif +#if defined(CONFIG_X86_LOCAL_APIC) && defined(CONFIG_X86) + { + .procname = "unknown_nmi_panic", + .data = &unknown_nmi_panic, .maxlen = sizeof (int), .mode = 0644, - .proc_handler = proc_nmi_enabled, + .proc_handler = proc_dointvec, }, #endif #if defined(CONFIG_X86) @@ -807,26 +855,6 @@ static struct ctl_table kern_table[] = { .proc_handler = proc_dointvec, }, #endif -#ifdef CONFIG_DETECT_SOFTLOCKUP - { - .procname = "softlockup_panic", - .data = &softlockup_panic, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec_minmax, - .extra1 = &zero, - .extra2 = &one, - }, - { - .procname = "softlockup_thresh", - .data = &softlockup_thresh, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dosoftlockup_thresh, - .extra1 = &neg_one, - .extra2 = &sixty, - }, -#endif #ifdef CONFIG_DETECT_HUNG_TASK { .procname = "hung_task_panic", @@ -900,13 +928,6 @@ static struct ctl_table kern_table[] = { .proc_handler = proc_dointvec, }, #endif -#ifdef CONFIG_SLOW_WORK - { - .procname = "slow-work", - .mode = 0555, - .child = slow_work_sysctls, - }, -#endif #ifdef CONFIG_PERF_EVENTS { .procname = "perf_event_paranoid", @@ -948,10 +969,6 @@ static struct ctl_table kern_table[] = { .proc_handler = proc_dointvec, }, #endif -/* - * NOTE: do not add new entries to this table unless you have read - * Documentation/sysctl/ctl_unnumbered.txt - */ { } }; @@ -1121,6 +1138,25 @@ static struct ctl_table vm_table[] = { .mode = 0644, .proc_handler = drop_caches_sysctl_handler, }, +#ifdef CONFIG_COMPACTION + { + .procname = "compact_memory", + .data = &sysctl_compact_memory, + .maxlen = sizeof(int), + .mode = 0200, + .proc_handler = sysctl_compaction_handler, + }, + { + .procname = "extfrag_threshold", + .data = &sysctl_extfrag_threshold, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = sysctl_extfrag_handler, + .extra1 = &min_extfrag_threshold, + .extra2 = &max_extfrag_threshold, + }, + +#endif /* CONFIG_COMPACTION */ { .procname = "min_free_kbytes", .data = &min_free_kbytes, @@ -1293,11 +1329,6 @@ static struct ctl_table vm_table[] = { .extra2 = &one, }, #endif - -/* - * NOTE: do not add new entries to this table unless you have read - * Documentation/sysctl/ctl_unnumbered.txt - */ { } }; @@ -1313,28 +1344,28 @@ static struct ctl_table fs_table[] = { .data = &inodes_stat, .maxlen = 2*sizeof(int), .mode = 0444, - .proc_handler = proc_dointvec, + .proc_handler = proc_nr_inodes, }, { .procname = "inode-state", .data = &inodes_stat, .maxlen = 7*sizeof(int), .mode = 0444, - .proc_handler = proc_dointvec, + .proc_handler = proc_nr_inodes, }, { .procname = "file-nr", .data = &files_stat, - .maxlen = 3*sizeof(int), + .maxlen = sizeof(files_stat), .mode = 0444, .proc_handler = proc_nr_files, }, { .procname = "file-max", .data = &files_stat.max_files, - .maxlen = sizeof(int), + .maxlen = sizeof(files_stat.max_files), .mode = 0644, - .proc_handler = proc_dointvec, + .proc_handler = proc_doulongvec_minmax, }, { .procname = "nr_open", @@ -1350,7 +1381,7 @@ static struct ctl_table fs_table[] = { .data = &dentry_stat, .maxlen = 6*sizeof(int), .mode = 0444, - .proc_handler = proc_dointvec, + .proc_handler = proc_nr_dentry, }, { .procname = "overflowuid", @@ -1446,17 +1477,13 @@ static struct ctl_table fs_table[] = { }, #endif { - .procname = "pipe-max-pages", - .data = &pipe_max_pages, + .procname = "pipe-max-size", + .data = &pipe_max_size, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = &proc_dointvec_minmax, - .extra1 = &two, + .proc_handler = &pipe_proc_fn, + .extra1 = &pipe_min_size, }, -/* - * NOTE: do not add new entries to this table unless you have read - * Documentation/sysctl/ctl_unnumbered.txt - */ { } }; @@ -1686,10 +1713,7 @@ static __init int sysctl_init(void) { sysctl_set_parent(NULL, root_table); #ifdef CONFIG_SYSCTL_SYSCALL_CHECK - { - int err; - err = sysctl_check_table(current->nsproxy, root_table); - } + sysctl_check_table(current->nsproxy, root_table); #endif return 0; } @@ -2262,6 +2286,8 @@ static int __do_proc_dointvec(void *tbl_data, struct ctl_table *table, if (write) { left -= proc_skip_spaces(&kbuf); + if (!left) + break; err = proc_get_long(&kbuf, &left, &lval, &neg, proc_wspace_sep, sizeof(proc_wspace_sep), NULL); @@ -2288,7 +2314,7 @@ static int __do_proc_dointvec(void *tbl_data, struct ctl_table *table, if (!write && !first && left && !err) err = proc_put_char(&buffer, &left, '\n'); - if (write && !err) + if (write && !err && left) left -= proc_skip_spaces(&kbuf); free: if (write) { @@ -2459,7 +2485,7 @@ static int __do_proc_doulongvec_minmax(void *data, struct ctl_table *table, int kbuf[left] = 0; } - for (; left && vleft--; i++, min++, max++, first=0) { + for (; left && vleft--; i++, first = 0) { unsigned long val; if (write) { @@ -2867,7 +2893,7 @@ int proc_do_large_bitmap(struct ctl_table *table, int write, } } -#else /* CONFIG_PROC_FS */ +#else /* CONFIG_PROC_SYSCTL */ int proc_dostring(struct ctl_table *table, int write, void __user *buffer, size_t *lenp, loff_t *ppos) @@ -2919,7 +2945,7 @@ int proc_doulongvec_ms_jiffies_minmax(struct ctl_table *table, int write, } -#endif /* CONFIG_PROC_FS */ +#endif /* CONFIG_PROC_SYSCTL */ /* * No sense putting this after each symbol definition, twice, diff --git a/kernel/sysctl_binary.c b/kernel/sysctl_binary.c index 937d31dc856..b875bedf7c9 100644 --- a/kernel/sysctl_binary.c +++ b/kernel/sysctl_binary.c @@ -13,6 +13,7 @@ #include <linux/file.h> #include <linux/ctype.h> #include <linux/netdevice.h> +#include <linux/kernel.h> #include <linux/slab.h> #ifdef CONFIG_SYSCTL_SYSCALL @@ -135,7 +136,6 @@ static const struct bin_table bin_kern_table[] = { { 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" }, {} }; @@ -1124,11 +1124,6 @@ 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) { @@ -1156,7 +1151,8 @@ static ssize_t bin_uuid(struct file *file, if (!isxdigit(str[0]) || !isxdigit(str[1])) goto out; - uuid[i] = (hex_value(str[0]) << 4) | hex_value(str[1]); + uuid[i] = (hex_to_bin(str[0]) << 4) | + hex_to_bin(str[1]); str += 2; if (*str == '-') str++; @@ -1196,7 +1192,7 @@ static ssize_t bin_dn_node_address(struct file *file, buf[result] = '\0'; - /* Convert the decnet addresss to binary */ + /* Convert the decnet address to binary */ result = -EIO; nodep = strchr(buf, '.') + 1; if (!nodep) diff --git a/kernel/sysctl_check.c b/kernel/sysctl_check.c index 04cdcf72c82..10b90d8a03c 100644 --- a/kernel/sysctl_check.c +++ b/kernel/sysctl_check.c @@ -143,15 +143,6 @@ int sysctl_check_table(struct nsproxy *namespaces, struct ctl_table *table) if (!table->maxlen) set_fail(&fail, table, "No maxlen"); } - if ((table->proc_handler == proc_doulongvec_minmax) || - (table->proc_handler == proc_doulongvec_ms_jiffies_minmax)) { - if (table->maxlen > sizeof (unsigned long)) { - if (!table->extra1) - set_fail(&fail, table, "No min"); - if (!table->extra2) - set_fail(&fail, table, "No max"); - } - } #ifdef CONFIG_PROC_SYSCTL if (table->procname && !table->proc_handler) set_fail(&fail, table, "No proc_handler"); diff --git a/kernel/taskstats.c b/kernel/taskstats.c index 11281d5792b..3971c6b9d58 100644 --- a/kernel/taskstats.c +++ b/kernel/taskstats.c @@ -89,8 +89,7 @@ static int prepare_reply(struct genl_info *info, u8 cmd, struct sk_buff **skbp, return -ENOMEM; if (!info) { - int seq = get_cpu_var(taskstats_seqnum)++; - put_cpu_var(taskstats_seqnum); + int seq = this_cpu_inc_return(taskstats_seqnum) - 1; reply = genlmsg_put(skb, 0, seq, &family, 0, cmd); } else @@ -175,22 +174,8 @@ static void send_cpu_listeners(struct sk_buff *skb, up_write(&listeners->sem); } -static int fill_pid(pid_t pid, struct task_struct *tsk, - struct taskstats *stats) +static void fill_stats(struct task_struct *tsk, struct taskstats *stats) { - int rc = 0; - - if (!tsk) { - rcu_read_lock(); - tsk = find_task_by_vpid(pid); - if (tsk) - get_task_struct(tsk); - rcu_read_unlock(); - if (!tsk) - return -ESRCH; - } else - get_task_struct(tsk); - memset(stats, 0, sizeof(*stats)); /* * Each accounting subsystem adds calls to its functions to @@ -209,17 +194,27 @@ static int fill_pid(pid_t pid, struct task_struct *tsk, /* fill in extended acct fields */ xacct_add_tsk(stats, tsk); +} - /* Define err: label here if needed */ - put_task_struct(tsk); - return rc; +static int fill_stats_for_pid(pid_t pid, struct taskstats *stats) +{ + struct task_struct *tsk; + rcu_read_lock(); + tsk = find_task_by_vpid(pid); + if (tsk) + get_task_struct(tsk); + rcu_read_unlock(); + if (!tsk) + return -ESRCH; + fill_stats(tsk, stats); + put_task_struct(tsk); + return 0; } -static int fill_tgid(pid_t tgid, struct task_struct *first, - struct taskstats *stats) +static int fill_stats_for_tgid(pid_t tgid, struct taskstats *stats) { - struct task_struct *tsk; + struct task_struct *tsk, *first; unsigned long flags; int rc = -ESRCH; @@ -228,8 +223,7 @@ static int fill_tgid(pid_t tgid, struct task_struct *first, * leaders who are already counted with the dead tasks */ rcu_read_lock(); - if (!first) - first = find_task_by_vpid(tgid); + first = find_task_by_vpid(tgid); if (!first || !lock_task_sighand(first, &flags)) goto out; @@ -268,7 +262,6 @@ out: return rc; } - static void fill_tgid_exit(struct task_struct *tsk) { unsigned long flags; @@ -355,6 +348,10 @@ static int parse(struct nlattr *na, struct cpumask *mask) return ret; } +#if defined(CONFIG_64BIT) && !defined(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS) +#define TASKSTATS_NEEDS_PADDING 1 +#endif + static struct taskstats *mk_reply(struct sk_buff *skb, int type, u32 pid) { struct nlattr *na, *ret; @@ -364,9 +361,33 @@ static struct taskstats *mk_reply(struct sk_buff *skb, int type, u32 pid) ? TASKSTATS_TYPE_AGGR_PID : TASKSTATS_TYPE_AGGR_TGID; + /* + * The taskstats structure is internally aligned on 8 byte + * boundaries but the layout of the aggregrate reply, with + * two NLA headers and the pid (each 4 bytes), actually + * force the entire structure to be unaligned. This causes + * the kernel to issue unaligned access warnings on some + * architectures like ia64. Unfortunately, some software out there + * doesn't properly unroll the NLA packet and assumes that the start + * of the taskstats structure will always be 20 bytes from the start + * of the netlink payload. Aligning the start of the taskstats + * structure breaks this software, which we don't want. So, for now + * the alignment only happens on architectures that require it + * and those users will have to update to fixed versions of those + * packages. Space is reserved in the packet only when needed. + * This ifdef should be removed in several years e.g. 2012 once + * we can be confident that fixed versions are installed on most + * systems. We add the padding before the aggregate since the + * aggregate is already a defined type. + */ +#ifdef TASKSTATS_NEEDS_PADDING + if (nla_put(skb, TASKSTATS_TYPE_NULL, 0, NULL) < 0) + goto err; +#endif na = nla_nest_start(skb, aggr); if (!na) goto err; + if (nla_put(skb, type, sizeof(pid), &pid) < 0) goto err; ret = nla_reserve(skb, TASKSTATS_TYPE_STATS, sizeof(struct taskstats)); @@ -424,74 +445,122 @@ err: return rc; } -static int taskstats_user_cmd(struct sk_buff *skb, struct genl_info *info) +static int cmd_attr_register_cpumask(struct genl_info *info) { - int rc; - struct sk_buff *rep_skb; - struct taskstats *stats; - size_t size; cpumask_var_t mask; + int rc; if (!alloc_cpumask_var(&mask, GFP_KERNEL)) return -ENOMEM; - rc = parse(info->attrs[TASKSTATS_CMD_ATTR_REGISTER_CPUMASK], mask); if (rc < 0) - goto free_return_rc; - if (rc == 0) { - rc = add_del_listener(info->snd_pid, mask, REGISTER); - goto free_return_rc; - } + goto out; + rc = add_del_listener(info->snd_pid, mask, REGISTER); +out: + free_cpumask_var(mask); + return rc; +} + +static int cmd_attr_deregister_cpumask(struct genl_info *info) +{ + cpumask_var_t mask; + int rc; + if (!alloc_cpumask_var(&mask, GFP_KERNEL)) + return -ENOMEM; rc = parse(info->attrs[TASKSTATS_CMD_ATTR_DEREGISTER_CPUMASK], mask); if (rc < 0) - goto free_return_rc; - if (rc == 0) { - rc = add_del_listener(info->snd_pid, mask, DEREGISTER); -free_return_rc: - free_cpumask_var(mask); - return rc; - } + goto out; + rc = add_del_listener(info->snd_pid, mask, DEREGISTER); +out: free_cpumask_var(mask); + return rc; +} + +static size_t taskstats_packet_size(void) +{ + size_t size; - /* - * Size includes space for nested attributes - */ size = nla_total_size(sizeof(u32)) + nla_total_size(sizeof(struct taskstats)) + nla_total_size(0); +#ifdef TASKSTATS_NEEDS_PADDING + size += nla_total_size(0); /* Padding for alignment */ +#endif + return size; +} + +static int cmd_attr_pid(struct genl_info *info) +{ + struct taskstats *stats; + struct sk_buff *rep_skb; + size_t size; + u32 pid; + int rc; + + size = taskstats_packet_size(); rc = prepare_reply(info, TASKSTATS_CMD_NEW, &rep_skb, size); if (rc < 0) return rc; rc = -EINVAL; - if (info->attrs[TASKSTATS_CMD_ATTR_PID]) { - u32 pid = nla_get_u32(info->attrs[TASKSTATS_CMD_ATTR_PID]); - stats = mk_reply(rep_skb, TASKSTATS_TYPE_PID, pid); - if (!stats) - goto err; - - rc = fill_pid(pid, NULL, stats); - if (rc < 0) - goto err; - } else if (info->attrs[TASKSTATS_CMD_ATTR_TGID]) { - u32 tgid = nla_get_u32(info->attrs[TASKSTATS_CMD_ATTR_TGID]); - stats = mk_reply(rep_skb, TASKSTATS_TYPE_TGID, tgid); - if (!stats) - goto err; - - rc = fill_tgid(tgid, NULL, stats); - if (rc < 0) - goto err; - } else + pid = nla_get_u32(info->attrs[TASKSTATS_CMD_ATTR_PID]); + stats = mk_reply(rep_skb, TASKSTATS_TYPE_PID, pid); + if (!stats) goto err; + rc = fill_stats_for_pid(pid, stats); + if (rc < 0) + goto err; return send_reply(rep_skb, info); err: nlmsg_free(rep_skb); return rc; } +static int cmd_attr_tgid(struct genl_info *info) +{ + struct taskstats *stats; + struct sk_buff *rep_skb; + size_t size; + u32 tgid; + int rc; + + size = taskstats_packet_size(); + + rc = prepare_reply(info, TASKSTATS_CMD_NEW, &rep_skb, size); + if (rc < 0) + return rc; + + rc = -EINVAL; + tgid = nla_get_u32(info->attrs[TASKSTATS_CMD_ATTR_TGID]); + stats = mk_reply(rep_skb, TASKSTATS_TYPE_TGID, tgid); + if (!stats) + goto err; + + rc = fill_stats_for_tgid(tgid, stats); + if (rc < 0) + goto err; + return send_reply(rep_skb, info); +err: + nlmsg_free(rep_skb); + return rc; +} + +static int taskstats_user_cmd(struct sk_buff *skb, struct genl_info *info) +{ + if (info->attrs[TASKSTATS_CMD_ATTR_REGISTER_CPUMASK]) + return cmd_attr_register_cpumask(info); + else if (info->attrs[TASKSTATS_CMD_ATTR_DEREGISTER_CPUMASK]) + return cmd_attr_deregister_cpumask(info); + else if (info->attrs[TASKSTATS_CMD_ATTR_PID]) + return cmd_attr_pid(info); + else if (info->attrs[TASKSTATS_CMD_ATTR_TGID]) + return cmd_attr_tgid(info); + else + return -EINVAL; +} + static struct taskstats *taskstats_tgid_alloc(struct task_struct *tsk) { struct signal_struct *sig = tsk->signal; @@ -532,8 +601,7 @@ void taskstats_exit(struct task_struct *tsk, int group_dead) /* * Size includes space for nested attributes */ - size = nla_total_size(sizeof(u32)) + - nla_total_size(sizeof(struct taskstats)) + nla_total_size(0); + size = taskstats_packet_size(); is_thread_group = !!taskstats_tgid_alloc(tsk); if (is_thread_group) { @@ -543,7 +611,7 @@ void taskstats_exit(struct task_struct *tsk, int group_dead) fill_tgid_exit(tsk); } - listeners = &__raw_get_cpu_var(listener_array); + listeners = __this_cpu_ptr(&listener_array); if (list_empty(&listeners->list)) return; @@ -555,9 +623,7 @@ void taskstats_exit(struct task_struct *tsk, int group_dead) if (!stats) goto err; - rc = fill_pid(-1, tsk, stats); - if (rc < 0) - goto err; + fill_stats(tsk, stats); /* * Doesn't matter if tsk is the leader or the last group member leaving diff --git a/kernel/test_kprobes.c b/kernel/test_kprobes.c index 4f104515a19..f8b11a28317 100644 --- a/kernel/test_kprobes.c +++ b/kernel/test_kprobes.c @@ -115,7 +115,9 @@ static int test_kprobes(void) int ret; struct kprobe *kps[2] = {&kp, &kp2}; - kp.addr = 0; /* addr should be cleard for reusing kprobe. */ + /* addr and flags should be cleard for reusing kprobe. */ + kp.addr = NULL; + kp.flags = 0; ret = register_kprobes(kps, 2); if (ret < 0) { printk(KERN_ERR "Kprobe smoke test failed: " @@ -210,7 +212,9 @@ static int test_jprobes(void) int ret; struct jprobe *jps[2] = {&jp, &jp2}; - jp.kp.addr = 0; /* addr should be cleard for reusing kprobe. */ + /* addr and flags should be cleard for reusing kprobe. */ + jp.kp.addr = NULL; + jp.kp.flags = 0; ret = register_jprobes(jps, 2); if (ret < 0) { printk(KERN_ERR "Kprobe smoke test failed: " @@ -323,7 +327,9 @@ static int test_kretprobes(void) int ret; struct kretprobe *rps[2] = {&rp, &rp2}; - rp.kp.addr = 0; /* addr should be cleard for reusing kprobe. */ + /* addr and flags should be cleard for reusing kprobe. */ + rp.kp.addr = NULL; + rp.kp.flags = 0; ret = register_kretprobes(rps, 2); if (ret < 0) { printk(KERN_ERR "Kprobe smoke test failed: " diff --git a/kernel/time.c b/kernel/time.c index 848b1c2ab09..32174359576 100644 --- a/kernel/time.c +++ b/kernel/time.c @@ -238,7 +238,7 @@ EXPORT_SYMBOL(current_fs_time); * Avoid unnecessary multiplications/divisions in the * two most common HZ cases: */ -unsigned int inline jiffies_to_msecs(const unsigned long j) +inline unsigned int jiffies_to_msecs(const unsigned long j) { #if HZ <= MSEC_PER_SEC && !(MSEC_PER_SEC % HZ) return (MSEC_PER_SEC / HZ) * j; @@ -254,7 +254,7 @@ unsigned int inline jiffies_to_msecs(const unsigned long j) } EXPORT_SYMBOL(jiffies_to_msecs); -unsigned int inline jiffies_to_usecs(const unsigned long j) +inline unsigned int jiffies_to_usecs(const unsigned long j) { #if HZ <= USEC_PER_SEC && !(USEC_PER_SEC % HZ) return (USEC_PER_SEC / HZ) * j; @@ -300,22 +300,6 @@ struct timespec timespec_trunc(struct timespec t, unsigned gran) } EXPORT_SYMBOL(timespec_trunc); -#ifndef CONFIG_GENERIC_TIME -/* - * Simulate gettimeofday using do_gettimeofday which only allows a timeval - * and therefore only yields usec accuracy - */ -void getnstimeofday(struct timespec *tv) -{ - struct timeval x; - - do_gettimeofday(&x); - tv->tv_sec = x.tv_sec; - tv->tv_nsec = x.tv_usec * NSEC_PER_USEC; -} -EXPORT_SYMBOL_GPL(getnstimeofday); -#endif - /* Converts Gregorian date to seconds since 1970-01-01 00:00:00. * Assumes input in normal date format, i.e. 1980-12-31 23:59:59 * => year=1980, mon=12, day=31, hour=23, min=59, sec=59. diff --git a/kernel/time/Kconfig b/kernel/time/Kconfig index 95ed42951e0..f06a8a36564 100644 --- a/kernel/time/Kconfig +++ b/kernel/time/Kconfig @@ -6,7 +6,7 @@ config TICK_ONESHOT config NO_HZ bool "Tickless System (Dynamic Ticks)" - depends on GENERIC_TIME && GENERIC_CLOCKEVENTS + depends on !ARCH_USES_GETTIMEOFFSET && GENERIC_CLOCKEVENTS select TICK_ONESHOT help This option enables a tickless system: timer interrupts will @@ -15,7 +15,7 @@ config NO_HZ config HIGH_RES_TIMERS bool "High Resolution Timer Support" - depends on GENERIC_TIME && GENERIC_CLOCKEVENTS + depends on !ARCH_USES_GETTIMEOFFSET && GENERIC_CLOCKEVENTS select TICK_ONESHOT help This option enables high resolution timer support. If your diff --git a/kernel/time/clocksource.c b/kernel/time/clocksource.c index f08e99c1d56..6519cf62d9c 100644 --- a/kernel/time/clocksource.c +++ b/kernel/time/clocksource.c @@ -113,7 +113,7 @@ EXPORT_SYMBOL_GPL(timecounter_cyc2time); * @shift: pointer to shift variable * @from: frequency to convert from * @to: frequency to convert to - * @minsec: guaranteed runtime conversion range in seconds + * @maxsec: guaranteed runtime conversion range in seconds * * The function evaluates the shift/mult pair for the scaled math * operations of clocksources and clockevents. @@ -122,7 +122,7 @@ EXPORT_SYMBOL_GPL(timecounter_cyc2time); * 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 + * The @maxsec 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 @@ -131,7 +131,7 @@ EXPORT_SYMBOL_GPL(timecounter_cyc2time); * factors. */ void -clocks_calc_mult_shift(u32 *mult, u32 *shift, u32 from, u32 to, u32 minsec) +clocks_calc_mult_shift(u32 *mult, u32 *shift, u32 from, u32 to, u32 maxsec) { u64 tmp; u32 sft, sftacc= 32; @@ -140,7 +140,7 @@ clocks_calc_mult_shift(u32 *mult, u32 *shift, u32 from, u32 to, u32 minsec) * Calculate the shift factor which is limiting the conversion * range: */ - tmp = ((u64)minsec * from) >> 32; + tmp = ((u64)maxsec * from) >> 32; while (tmp) { tmp >>=1; sftacc--; @@ -152,6 +152,7 @@ clocks_calc_mult_shift(u32 *mult, u32 *shift, u32 from, u32 to, u32 minsec) */ for (sft = 32; sft > 0; sft--) { tmp = (u64) to << sft; + tmp += from / 2; do_div(tmp, from); if ((tmp >> sftacc) == 0) break; @@ -531,7 +532,7 @@ static u64 clocksource_max_deferment(struct clocksource *cs) return max_nsecs - (max_nsecs >> 5); } -#ifdef CONFIG_GENERIC_TIME +#ifndef CONFIG_ARCH_USES_GETTIMEOFFSET /** * clocksource_select - Select the best clocksource available @@ -577,7 +578,7 @@ static void clocksource_select(void) } } -#else /* CONFIG_GENERIC_TIME */ +#else /* !CONFIG_ARCH_USES_GETTIMEOFFSET */ static inline void clocksource_select(void) { } @@ -639,19 +640,18 @@ static void clocksource_enqueue(struct clocksource *cs) #define MAX_UPDATE_LENGTH 5 /* Seconds */ /** - * __clocksource_register_scale - Used to install new clocksources + * __clocksource_updatefreq_scale - Used update clocksource with new freq * @t: clocksource to be registered * @scale: Scale factor multiplied against freq to get clocksource hz * @freq: clocksource frequency (cycles per second) divided by scale * - * Returns -EBUSY if registration fails, zero otherwise. + * This should only be called from the clocksource->enable() method. * * This *SHOULD NOT* be called directly! Please use the - * clocksource_register_hz() or clocksource_register_khz helper functions. + * clocksource_updatefreq_hz() or clocksource_updatefreq_khz helper functions. */ -int __clocksource_register_scale(struct clocksource *cs, u32 scale, u32 freq) +void __clocksource_updatefreq_scale(struct clocksource *cs, u32 scale, u32 freq) { - /* * Ideally we want to use some of the limits used in * clocksource_max_deferment, to provide a more informed @@ -662,7 +662,27 @@ int __clocksource_register_scale(struct clocksource *cs, u32 scale, u32 freq) NSEC_PER_SEC/scale, MAX_UPDATE_LENGTH*scale); cs->max_idle_ns = clocksource_max_deferment(cs); +} +EXPORT_SYMBOL_GPL(__clocksource_updatefreq_scale); + +/** + * __clocksource_register_scale - Used to install new clocksources + * @t: clocksource to be registered + * @scale: Scale factor multiplied against freq to get clocksource hz + * @freq: clocksource frequency (cycles per second) divided by scale + * + * Returns -EBUSY if registration fails, zero otherwise. + * + * This *SHOULD NOT* be called directly! Please use the + * clocksource_register_hz() or clocksource_register_khz helper functions. + */ +int __clocksource_register_scale(struct clocksource *cs, u32 scale, u32 freq) +{ + + /* Initialize mult/shift and max_idle_ns */ + __clocksource_updatefreq_scale(cs, scale, freq); + /* Add clocksource to the clcoksource list */ mutex_lock(&clocksource_mutex); clocksource_enqueue(cs); clocksource_select(); diff --git a/kernel/time/ntp.c b/kernel/time/ntp.c index c63116863a8..5c00242fa92 100644 --- a/kernel/time/ntp.c +++ b/kernel/time/ntp.c @@ -14,6 +14,7 @@ #include <linux/timex.h> #include <linux/time.h> #include <linux/mm.h> +#include <linux/module.h> /* * NTP timekeeping variables: @@ -74,6 +75,162 @@ static long time_adjust; /* constant (boot-param configurable) NTP tick adjustment (upscaled) */ static s64 ntp_tick_adj; +#ifdef CONFIG_NTP_PPS + +/* + * The following variables are used when a pulse-per-second (PPS) signal + * is available. They establish the engineering parameters of the clock + * discipline loop when controlled by the PPS signal. + */ +#define PPS_VALID 10 /* PPS signal watchdog max (s) */ +#define PPS_POPCORN 4 /* popcorn spike threshold (shift) */ +#define PPS_INTMIN 2 /* min freq interval (s) (shift) */ +#define PPS_INTMAX 8 /* max freq interval (s) (shift) */ +#define PPS_INTCOUNT 4 /* number of consecutive good intervals to + increase pps_shift or consecutive bad + intervals to decrease it */ +#define PPS_MAXWANDER 100000 /* max PPS freq wander (ns/s) */ + +static int pps_valid; /* signal watchdog counter */ +static long pps_tf[3]; /* phase median filter */ +static long pps_jitter; /* current jitter (ns) */ +static struct timespec pps_fbase; /* beginning of the last freq interval */ +static int pps_shift; /* current interval duration (s) (shift) */ +static int pps_intcnt; /* interval counter */ +static s64 pps_freq; /* frequency offset (scaled ns/s) */ +static long pps_stabil; /* current stability (scaled ns/s) */ + +/* + * PPS signal quality monitors + */ +static long pps_calcnt; /* calibration intervals */ +static long pps_jitcnt; /* jitter limit exceeded */ +static long pps_stbcnt; /* stability limit exceeded */ +static long pps_errcnt; /* calibration errors */ + + +/* PPS kernel consumer compensates the whole phase error immediately. + * Otherwise, reduce the offset by a fixed factor times the time constant. + */ +static inline s64 ntp_offset_chunk(s64 offset) +{ + if (time_status & STA_PPSTIME && time_status & STA_PPSSIGNAL) + return offset; + else + return shift_right(offset, SHIFT_PLL + time_constant); +} + +static inline void pps_reset_freq_interval(void) +{ + /* the PPS calibration interval may end + surprisingly early */ + pps_shift = PPS_INTMIN; + pps_intcnt = 0; +} + +/** + * pps_clear - Clears the PPS state variables + * + * Must be called while holding a write on the xtime_lock + */ +static inline void pps_clear(void) +{ + pps_reset_freq_interval(); + pps_tf[0] = 0; + pps_tf[1] = 0; + pps_tf[2] = 0; + pps_fbase.tv_sec = pps_fbase.tv_nsec = 0; + pps_freq = 0; +} + +/* Decrease pps_valid to indicate that another second has passed since + * the last PPS signal. When it reaches 0, indicate that PPS signal is + * missing. + * + * Must be called while holding a write on the xtime_lock + */ +static inline void pps_dec_valid(void) +{ + if (pps_valid > 0) + pps_valid--; + else { + time_status &= ~(STA_PPSSIGNAL | STA_PPSJITTER | + STA_PPSWANDER | STA_PPSERROR); + pps_clear(); + } +} + +static inline void pps_set_freq(s64 freq) +{ + pps_freq = freq; +} + +static inline int is_error_status(int status) +{ + return (time_status & (STA_UNSYNC|STA_CLOCKERR)) + /* PPS signal lost when either PPS time or + * PPS frequency synchronization requested + */ + || ((time_status & (STA_PPSFREQ|STA_PPSTIME)) + && !(time_status & STA_PPSSIGNAL)) + /* PPS jitter exceeded when + * PPS time synchronization requested */ + || ((time_status & (STA_PPSTIME|STA_PPSJITTER)) + == (STA_PPSTIME|STA_PPSJITTER)) + /* PPS wander exceeded or calibration error when + * PPS frequency synchronization requested + */ + || ((time_status & STA_PPSFREQ) + && (time_status & (STA_PPSWANDER|STA_PPSERROR))); +} + +static inline void pps_fill_timex(struct timex *txc) +{ + txc->ppsfreq = shift_right((pps_freq >> PPM_SCALE_INV_SHIFT) * + PPM_SCALE_INV, NTP_SCALE_SHIFT); + txc->jitter = pps_jitter; + if (!(time_status & STA_NANO)) + txc->jitter /= NSEC_PER_USEC; + txc->shift = pps_shift; + txc->stabil = pps_stabil; + txc->jitcnt = pps_jitcnt; + txc->calcnt = pps_calcnt; + txc->errcnt = pps_errcnt; + txc->stbcnt = pps_stbcnt; +} + +#else /* !CONFIG_NTP_PPS */ + +static inline s64 ntp_offset_chunk(s64 offset) +{ + return shift_right(offset, SHIFT_PLL + time_constant); +} + +static inline void pps_reset_freq_interval(void) {} +static inline void pps_clear(void) {} +static inline void pps_dec_valid(void) {} +static inline void pps_set_freq(s64 freq) {} + +static inline int is_error_status(int status) +{ + return status & (STA_UNSYNC|STA_CLOCKERR); +} + +static inline void pps_fill_timex(struct timex *txc) +{ + /* PPS is not implemented, so these are zero */ + txc->ppsfreq = 0; + txc->jitter = 0; + txc->shift = 0; + txc->stabil = 0; + txc->jitcnt = 0; + txc->calcnt = 0; + txc->errcnt = 0; + txc->stbcnt = 0; +} + +#endif /* CONFIG_NTP_PPS */ + /* * NTP methods: */ @@ -149,10 +306,18 @@ static void ntp_update_offset(long offset) time_reftime = get_seconds(); offset64 = offset; - freq_adj = (offset64 * secs) << - (NTP_SCALE_SHIFT - 2 * (SHIFT_PLL + 2 + time_constant)); + freq_adj = ntp_update_offset_fll(offset64, secs); - freq_adj += ntp_update_offset_fll(offset64, secs); + /* + * Clamp update interval to reduce PLL gain with low + * sampling rate (e.g. intermittent network connection) + * to avoid instability. + */ + if (unlikely(secs > 1 << (SHIFT_PLL + 1 + time_constant))) + secs = 1 << (SHIFT_PLL + 1 + time_constant); + + freq_adj += (offset64 * secs) << + (NTP_SCALE_SHIFT - 2 * (SHIFT_PLL + 2 + time_constant)); freq_adj = min(freq_adj + time_freq, MAXFREQ_SCALED); @@ -177,6 +342,9 @@ void ntp_clear(void) tick_length = tick_length_base; time_offset = 0; + + /* Clear PPS state variables */ + pps_clear(); } /* @@ -242,16 +410,16 @@ void second_overflow(void) time_status |= STA_UNSYNC; } - /* - * Compute the phase adjustment for the next second. The offset is - * reduced by a fixed factor times the time constant. - */ + /* Compute the phase adjustment for the next second */ tick_length = tick_length_base; - delta = shift_right(time_offset, SHIFT_PLL + time_constant); + delta = ntp_offset_chunk(time_offset); time_offset -= delta; tick_length += delta; + /* Check PPS signal */ + pps_dec_valid(); + if (!time_adjust) return; @@ -361,6 +529,8 @@ static inline void process_adj_status(struct timex *txc, struct timespec *ts) if ((time_status & STA_PLL) && !(txc->status & STA_PLL)) { time_state = TIME_OK; time_status = STA_UNSYNC; + /* restart PPS frequency calibration */ + pps_reset_freq_interval(); } /* @@ -410,6 +580,8 @@ static inline void process_adjtimex_modes(struct timex *txc, struct timespec *ts time_freq = txc->freq * PPM_SCALE; time_freq = min(time_freq, MAXFREQ_SCALED); time_freq = max(time_freq, -MAXFREQ_SCALED); + /* update pps_freq */ + pps_set_freq(time_freq); } if (txc->modes & ADJ_MAXERROR) @@ -500,7 +672,8 @@ int do_adjtimex(struct timex *txc) } result = time_state; /* mostly `TIME_OK' */ - if (time_status & (STA_UNSYNC|STA_CLOCKERR)) + /* check for errors */ + if (is_error_status(time_status)) result = TIME_ERROR; txc->freq = shift_right((time_freq >> PPM_SCALE_INV_SHIFT) * @@ -514,15 +687,8 @@ int do_adjtimex(struct timex *txc) txc->tick = tick_usec; txc->tai = time_tai; - /* PPS is not implemented, so these are zero */ - txc->ppsfreq = 0; - txc->jitter = 0; - txc->shift = 0; - txc->stabil = 0; - txc->jitcnt = 0; - txc->calcnt = 0; - txc->errcnt = 0; - txc->stbcnt = 0; + /* fill PPS status fields */ + pps_fill_timex(txc); write_sequnlock_irq(&xtime_lock); @@ -536,6 +702,243 @@ int do_adjtimex(struct timex *txc) return result; } +#ifdef CONFIG_NTP_PPS + +/* actually struct pps_normtime is good old struct timespec, but it is + * semantically different (and it is the reason why it was invented): + * pps_normtime.nsec has a range of ( -NSEC_PER_SEC / 2, NSEC_PER_SEC / 2 ] + * while timespec.tv_nsec has a range of [0, NSEC_PER_SEC) */ +struct pps_normtime { + __kernel_time_t sec; /* seconds */ + long nsec; /* nanoseconds */ +}; + +/* normalize the timestamp so that nsec is in the + ( -NSEC_PER_SEC / 2, NSEC_PER_SEC / 2 ] interval */ +static inline struct pps_normtime pps_normalize_ts(struct timespec ts) +{ + struct pps_normtime norm = { + .sec = ts.tv_sec, + .nsec = ts.tv_nsec + }; + + if (norm.nsec > (NSEC_PER_SEC >> 1)) { + norm.nsec -= NSEC_PER_SEC; + norm.sec++; + } + + return norm; +} + +/* get current phase correction and jitter */ +static inline long pps_phase_filter_get(long *jitter) +{ + *jitter = pps_tf[0] - pps_tf[1]; + if (*jitter < 0) + *jitter = -*jitter; + + /* TODO: test various filters */ + return pps_tf[0]; +} + +/* add the sample to the phase filter */ +static inline void pps_phase_filter_add(long err) +{ + pps_tf[2] = pps_tf[1]; + pps_tf[1] = pps_tf[0]; + pps_tf[0] = err; +} + +/* decrease frequency calibration interval length. + * It is halved after four consecutive unstable intervals. + */ +static inline void pps_dec_freq_interval(void) +{ + if (--pps_intcnt <= -PPS_INTCOUNT) { + pps_intcnt = -PPS_INTCOUNT; + if (pps_shift > PPS_INTMIN) { + pps_shift--; + pps_intcnt = 0; + } + } +} + +/* increase frequency calibration interval length. + * It is doubled after four consecutive stable intervals. + */ +static inline void pps_inc_freq_interval(void) +{ + if (++pps_intcnt >= PPS_INTCOUNT) { + pps_intcnt = PPS_INTCOUNT; + if (pps_shift < PPS_INTMAX) { + pps_shift++; + pps_intcnt = 0; + } + } +} + +/* update clock frequency based on MONOTONIC_RAW clock PPS signal + * timestamps + * + * At the end of the calibration interval the difference between the + * first and last MONOTONIC_RAW clock timestamps divided by the length + * of the interval becomes the frequency update. If the interval was + * too long, the data are discarded. + * Returns the difference between old and new frequency values. + */ +static long hardpps_update_freq(struct pps_normtime freq_norm) +{ + long delta, delta_mod; + s64 ftemp; + + /* check if the frequency interval was too long */ + if (freq_norm.sec > (2 << pps_shift)) { + time_status |= STA_PPSERROR; + pps_errcnt++; + pps_dec_freq_interval(); + pr_err("hardpps: PPSERROR: interval too long - %ld s\n", + freq_norm.sec); + return 0; + } + + /* here the raw frequency offset and wander (stability) is + * calculated. If the wander is less than the wander threshold + * the interval is increased; otherwise it is decreased. + */ + ftemp = div_s64(((s64)(-freq_norm.nsec)) << NTP_SCALE_SHIFT, + freq_norm.sec); + delta = shift_right(ftemp - pps_freq, NTP_SCALE_SHIFT); + pps_freq = ftemp; + if (delta > PPS_MAXWANDER || delta < -PPS_MAXWANDER) { + pr_warning("hardpps: PPSWANDER: change=%ld\n", delta); + time_status |= STA_PPSWANDER; + pps_stbcnt++; + pps_dec_freq_interval(); + } else { /* good sample */ + pps_inc_freq_interval(); + } + + /* the stability metric is calculated as the average of recent + * frequency changes, but is used only for performance + * monitoring + */ + delta_mod = delta; + if (delta_mod < 0) + delta_mod = -delta_mod; + pps_stabil += (div_s64(((s64)delta_mod) << + (NTP_SCALE_SHIFT - SHIFT_USEC), + NSEC_PER_USEC) - pps_stabil) >> PPS_INTMIN; + + /* if enabled, the system clock frequency is updated */ + if ((time_status & STA_PPSFREQ) != 0 && + (time_status & STA_FREQHOLD) == 0) { + time_freq = pps_freq; + ntp_update_frequency(); + } + + return delta; +} + +/* correct REALTIME clock phase error against PPS signal */ +static void hardpps_update_phase(long error) +{ + long correction = -error; + long jitter; + + /* add the sample to the median filter */ + pps_phase_filter_add(correction); + correction = pps_phase_filter_get(&jitter); + + /* Nominal jitter is due to PPS signal noise. If it exceeds the + * threshold, the sample is discarded; otherwise, if so enabled, + * the time offset is updated. + */ + if (jitter > (pps_jitter << PPS_POPCORN)) { + pr_warning("hardpps: PPSJITTER: jitter=%ld, limit=%ld\n", + jitter, (pps_jitter << PPS_POPCORN)); + time_status |= STA_PPSJITTER; + pps_jitcnt++; + } else if (time_status & STA_PPSTIME) { + /* correct the time using the phase offset */ + time_offset = div_s64(((s64)correction) << NTP_SCALE_SHIFT, + NTP_INTERVAL_FREQ); + /* cancel running adjtime() */ + time_adjust = 0; + } + /* update jitter */ + pps_jitter += (jitter - pps_jitter) >> PPS_INTMIN; +} + +/* + * hardpps() - discipline CPU clock oscillator to external PPS signal + * + * This routine is called at each PPS signal arrival in order to + * discipline the CPU clock oscillator to the PPS signal. It takes two + * parameters: REALTIME and MONOTONIC_RAW clock timestamps. The former + * is used to correct clock phase error and the latter is used to + * correct the frequency. + * + * This code is based on David Mills's reference nanokernel + * implementation. It was mostly rewritten but keeps the same idea. + */ +void hardpps(const struct timespec *phase_ts, const struct timespec *raw_ts) +{ + struct pps_normtime pts_norm, freq_norm; + unsigned long flags; + + pts_norm = pps_normalize_ts(*phase_ts); + + write_seqlock_irqsave(&xtime_lock, flags); + + /* clear the error bits, they will be set again if needed */ + time_status &= ~(STA_PPSJITTER | STA_PPSWANDER | STA_PPSERROR); + + /* indicate signal presence */ + time_status |= STA_PPSSIGNAL; + pps_valid = PPS_VALID; + + /* when called for the first time, + * just start the frequency interval */ + if (unlikely(pps_fbase.tv_sec == 0)) { + pps_fbase = *raw_ts; + write_sequnlock_irqrestore(&xtime_lock, flags); + return; + } + + /* ok, now we have a base for frequency calculation */ + freq_norm = pps_normalize_ts(timespec_sub(*raw_ts, pps_fbase)); + + /* check that the signal is in the range + * [1s - MAXFREQ us, 1s + MAXFREQ us], otherwise reject it */ + if ((freq_norm.sec == 0) || + (freq_norm.nsec > MAXFREQ * freq_norm.sec) || + (freq_norm.nsec < -MAXFREQ * freq_norm.sec)) { + time_status |= STA_PPSJITTER; + /* restart the frequency calibration interval */ + pps_fbase = *raw_ts; + write_sequnlock_irqrestore(&xtime_lock, flags); + pr_err("hardpps: PPSJITTER: bad pulse\n"); + return; + } + + /* signal is ok */ + + /* check if the current frequency interval is finished */ + if (freq_norm.sec >= (1 << pps_shift)) { + pps_calcnt++; + /* restart the frequency calibration interval */ + pps_fbase = *raw_ts; + hardpps_update_freq(freq_norm); + } + + hardpps_update_phase(pts_norm.nsec); + + write_sequnlock_irqrestore(&xtime_lock, flags); +} +EXPORT_SYMBOL(hardpps); + +#endif /* CONFIG_NTP_PPS */ + static int __init ntp_tick_adj_setup(char *str) { ntp_tick_adj = simple_strtol(str, NULL, 0); diff --git a/kernel/time/tick-broadcast.c b/kernel/time/tick-broadcast.c index b3bafd5fc66..48b2761b566 100644 --- a/kernel/time/tick-broadcast.c +++ b/kernel/time/tick-broadcast.c @@ -188,7 +188,7 @@ static void tick_handle_periodic_broadcast(struct clock_event_device *dev) /* * Setup the next period for devices, which do not have * periodic mode. We read dev->next_event first and add to it - * when the event alrady expired. clockevents_program_event() + * when the event already expired. clockevents_program_event() * sets dev->next_event only when the event is really * programmed to the device. */ diff --git a/kernel/time/tick-common.c b/kernel/time/tick-common.c index b6b898d2eee..051bc80a0c4 100644 --- a/kernel/time/tick-common.c +++ b/kernel/time/tick-common.c @@ -49,7 +49,7 @@ struct tick_device *tick_get_device(int cpu) */ int tick_is_oneshot_available(void) { - struct clock_event_device *dev = __get_cpu_var(tick_cpu_device).evtdev; + struct clock_event_device *dev = __this_cpu_read(tick_cpu_device.evtdev); return dev && (dev->features & CLOCK_EVT_FEAT_ONESHOT); } diff --git a/kernel/time/tick-oneshot.c b/kernel/time/tick-oneshot.c index aada0e52680..5cbc101f908 100644 --- a/kernel/time/tick-oneshot.c +++ b/kernel/time/tick-oneshot.c @@ -95,7 +95,7 @@ int tick_dev_program_event(struct clock_event_device *dev, ktime_t expires, */ int tick_program_event(ktime_t expires, int force) { - struct clock_event_device *dev = __get_cpu_var(tick_cpu_device).evtdev; + struct clock_event_device *dev = __this_cpu_read(tick_cpu_device.evtdev); return tick_dev_program_event(dev, expires, force); } @@ -167,7 +167,7 @@ int tick_oneshot_mode_active(void) int ret; local_irq_save(flags); - ret = __get_cpu_var(tick_cpu_device).mode == TICKDEV_MODE_ONESHOT; + ret = __this_cpu_read(tick_cpu_device.mode) == TICKDEV_MODE_ONESHOT; local_irq_restore(flags); return ret; diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c index 1d7b9bc1c03..c55ea243347 100644 --- a/kernel/time/tick-sched.c +++ b/kernel/time/tick-sched.c @@ -154,14 +154,14 @@ static void tick_nohz_update_jiffies(ktime_t now) * Updates the per cpu time idle statistics counters */ static void -update_ts_time_stats(struct tick_sched *ts, ktime_t now, u64 *last_update_time) +update_ts_time_stats(int cpu, struct tick_sched *ts, ktime_t now, u64 *last_update_time) { ktime_t delta; if (ts->idle_active) { delta = ktime_sub(now, ts->idle_entrytime); ts->idle_sleeptime = ktime_add(ts->idle_sleeptime, delta); - if (nr_iowait_cpu() > 0) + if (nr_iowait_cpu(cpu) > 0) ts->iowait_sleeptime = ktime_add(ts->iowait_sleeptime, delta); ts->idle_entrytime = now; } @@ -175,19 +175,19 @@ static void tick_nohz_stop_idle(int cpu, ktime_t now) { struct tick_sched *ts = &per_cpu(tick_cpu_sched, cpu); - update_ts_time_stats(ts, now, NULL); + update_ts_time_stats(cpu, ts, now, NULL); ts->idle_active = 0; sched_clock_idle_wakeup_event(0); } -static ktime_t tick_nohz_start_idle(struct tick_sched *ts) +static ktime_t tick_nohz_start_idle(int cpu, struct tick_sched *ts) { ktime_t now; now = ktime_get(); - update_ts_time_stats(ts, now, NULL); + update_ts_time_stats(cpu, ts, now, NULL); ts->idle_entrytime = now; ts->idle_active = 1; @@ -216,7 +216,7 @@ u64 get_cpu_idle_time_us(int cpu, u64 *last_update_time) if (!tick_nohz_enabled) return -1; - update_ts_time_stats(ts, ktime_get(), last_update_time); + update_ts_time_stats(cpu, ts, ktime_get(), last_update_time); return ktime_to_us(ts->idle_sleeptime); } @@ -242,7 +242,7 @@ u64 get_cpu_iowait_time_us(int cpu, u64 *last_update_time) if (!tick_nohz_enabled) return -1; - update_ts_time_stats(ts, ktime_get(), last_update_time); + update_ts_time_stats(cpu, ts, ktime_get(), last_update_time); return ktime_to_us(ts->iowait_sleeptime); } @@ -284,7 +284,7 @@ void tick_nohz_stop_sched_tick(int inidle) */ ts->inidle = 1; - now = tick_nohz_start_idle(ts); + now = tick_nohz_start_idle(cpu, ts); /* * If this cpu is offline and it is the one which updates @@ -315,9 +315,6 @@ void tick_nohz_stop_sched_tick(int inidle) goto end; } - if (nohz_ratelimit(cpu)) - goto end; - ts->idle_calls++; /* Read jiffies and the time when jiffies were updated last */ do { @@ -408,13 +405,7 @@ void tick_nohz_stop_sched_tick(int inidle) * the scheduler tick in nohz_restart_sched_tick. */ if (!ts->tick_stopped) { - if (select_nohz_load_balancer(1)) { - /* - * sched tick not stopped! - */ - cpumask_clear_cpu(cpu, nohz_cpu_mask); - goto out; - } + select_nohz_load_balancer(1); ts->idle_tick = hrtimer_get_expires(&ts->sched_timer); ts->tick_stopped = 1; @@ -651,8 +642,7 @@ static void tick_nohz_switch_to_nohz(void) } local_irq_enable(); - printk(KERN_INFO "Switched to NOHz mode on CPU #%d\n", - smp_processor_id()); + printk(KERN_INFO "Switched to NOHz mode on CPU #%d\n", smp_processor_id()); } /* @@ -783,7 +773,6 @@ void tick_setup_sched_timer(void) { struct tick_sched *ts = &__get_cpu_var(tick_cpu_sched); ktime_t now = ktime_get(); - u64 offset; /* * Emulate tick processing via per-CPU hrtimers: @@ -793,10 +782,6 @@ void tick_setup_sched_timer(void) /* Get the next period (per cpu) */ hrtimer_set_expires(&ts->sched_timer, tick_init_jiffy_update()); - offset = ktime_to_ns(tick_period) >> 1; - do_div(offset, num_possible_cpus()); - offset *= smp_processor_id(); - hrtimer_add_expires_ns(&ts->sched_timer, offset); for (;;) { hrtimer_forward(&ts->sched_timer, now, tick_period); @@ -809,8 +794,10 @@ void tick_setup_sched_timer(void) } #ifdef CONFIG_NO_HZ - if (tick_nohz_enabled) + if (tick_nohz_enabled) { ts->nohz_mode = NOHZ_MODE_HIGHRES; + printk(KERN_INFO "Switched to NOHz mode on CPU #%d\n", smp_processor_id()); + } #endif } #endif /* HIGH_RES_TIMERS */ diff --git a/kernel/time/timecompare.c b/kernel/time/timecompare.c index ac38fbb176c..a9ae369925c 100644 --- a/kernel/time/timecompare.c +++ b/kernel/time/timecompare.c @@ -21,6 +21,7 @@ #include <linux/module.h> #include <linux/slab.h> #include <linux/math64.h> +#include <linux/kernel.h> /* * fixed point arithmetic scale factor for skew @@ -57,11 +58,11 @@ int timecompare_offset(struct timecompare *sync, int index; int num_samples = sync->num_samples; - if (num_samples > sizeof(buffer)/sizeof(buffer[0])) { + if (num_samples > ARRAY_SIZE(buffer)) { samples = kmalloc(sizeof(*samples) * num_samples, GFP_ATOMIC); if (!samples) { samples = buffer; - num_samples = sizeof(buffer)/sizeof(buffer[0]); + num_samples = ARRAY_SIZE(buffer); } } else { samples = buffer; diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c index caf8d4d4f5c..d27c7562902 100644 --- a/kernel/time/timekeeping.c +++ b/kernel/time/timekeeping.c @@ -32,6 +32,8 @@ struct timekeeper { cycle_t cycle_interval; /* Number of clock shifted nano seconds in one NTP interval. */ u64 xtime_interval; + /* shifted nano seconds left over when rounding cycle_interval */ + s64 xtime_remainder; /* Raw nano seconds accumulated per NTP interval. */ u32 raw_interval; @@ -47,7 +49,7 @@ struct timekeeper { u32 mult; }; -struct timekeeper timekeeper; +static struct timekeeper timekeeper; /** * timekeeper_setup_internals - Set up internals to use clocksource clock. @@ -62,7 +64,7 @@ struct timekeeper timekeeper; static void timekeeper_setup_internals(struct clocksource *clock) { cycle_t interval; - u64 tmp; + u64 tmp, ntpinterval; timekeeper.clock = clock; clock->cycle_last = clock->read(clock); @@ -70,6 +72,7 @@ static void timekeeper_setup_internals(struct clocksource *clock) /* Do the ns -> cycle conversion first, using original mult */ tmp = NTP_INTERVAL_LENGTH; tmp <<= clock->shift; + ntpinterval = tmp; tmp += clock->mult/2; do_div(tmp, clock->mult); if (tmp == 0) @@ -80,6 +83,7 @@ static void timekeeper_setup_internals(struct clocksource *clock) /* Go back from cycles -> shifted ns */ timekeeper.xtime_interval = (u64) interval * clock->mult; + timekeeper.xtime_remainder = ntpinterval - timekeeper.xtime_interval; timekeeper.raw_interval = ((u64) interval * clock->mult) >> clock->shift; @@ -153,14 +157,14 @@ __cacheline_aligned_in_smp DEFINE_SEQLOCK(xtime_lock); * - wall_to_monotonic is no longer the boot time, getboottime must be * used instead. */ -struct timespec xtime __attribute__ ((aligned (16))); -struct timespec wall_to_monotonic __attribute__ ((aligned (16))); +static struct timespec xtime __attribute__ ((aligned (16))); +static struct timespec wall_to_monotonic __attribute__ ((aligned (16))); static struct timespec total_sleep_time; /* * The raw monotonic time for the CLOCK_MONOTONIC_RAW posix clock. */ -struct timespec raw_time; +static struct timespec raw_time; /* flag for if timekeeping is suspended */ int __read_mostly timekeeping_suspended; @@ -170,11 +174,10 @@ void timekeeping_leap_insert(int leapsecond) { xtime.tv_sec += leapsecond; wall_to_monotonic.tv_sec -= leapsecond; - update_vsyscall(&xtime, timekeeper.clock, timekeeper.mult); + update_vsyscall(&xtime, &wall_to_monotonic, timekeeper.clock, + timekeeper.mult); } -#ifdef CONFIG_GENERIC_TIME - /** * timekeeping_forward_now - update clock to the current time * @@ -285,6 +288,49 @@ void ktime_get_ts(struct timespec *ts) } EXPORT_SYMBOL_GPL(ktime_get_ts); +#ifdef CONFIG_NTP_PPS + +/** + * getnstime_raw_and_real - get day and raw monotonic time in timespec format + * @ts_raw: pointer to the timespec to be set to raw monotonic time + * @ts_real: pointer to the timespec to be set to the time of day + * + * This function reads both the time of day and raw monotonic time at the + * same time atomically and stores the resulting timestamps in timespec + * format. + */ +void getnstime_raw_and_real(struct timespec *ts_raw, struct timespec *ts_real) +{ + unsigned long seq; + s64 nsecs_raw, nsecs_real; + + WARN_ON_ONCE(timekeeping_suspended); + + do { + u32 arch_offset; + + seq = read_seqbegin(&xtime_lock); + + *ts_raw = raw_time; + *ts_real = xtime; + + nsecs_raw = timekeeping_get_ns_raw(); + nsecs_real = timekeeping_get_ns(); + + /* If arch requires, add in gettimeoffset() */ + arch_offset = arch_gettimeoffset(); + nsecs_raw += arch_offset; + nsecs_real += arch_offset; + + } while (read_seqretry(&xtime_lock, seq)); + + timespec_add_ns(ts_raw, nsecs_raw); + timespec_add_ns(ts_real, nsecs_real); +} +EXPORT_SYMBOL(getnstime_raw_and_real); + +#endif /* CONFIG_NTP_PPS */ + /** * do_gettimeofday - Returns the time of day in a timeval * @tv: pointer to the timeval to be set @@ -328,7 +374,8 @@ int do_settimeofday(struct timespec *tv) timekeeper.ntp_error = 0; ntp_clear(); - update_vsyscall(&xtime, timekeeper.clock, timekeeper.mult); + update_vsyscall(&xtime, &wall_to_monotonic, timekeeper.clock, + timekeeper.mult); write_sequnlock_irqrestore(&xtime_lock, flags); @@ -376,52 +423,6 @@ void timekeeping_notify(struct clocksource *clock) tick_clock_notify(); } -#else /* GENERIC_TIME */ - -static inline void timekeeping_forward_now(void) { } - -/** - * 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_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 * @@ -579,9 +580,9 @@ static int timekeeping_resume(struct sys_device *dev) if (timespec_compare(&ts, &timekeeping_suspend_time) > 0) { ts = timespec_sub(ts, timekeeping_suspend_time); - xtime = timespec_add_safe(xtime, ts); + xtime = timespec_add(xtime, ts); wall_to_monotonic = timespec_sub(wall_to_monotonic, ts); - total_sleep_time = timespec_add_safe(total_sleep_time, ts); + total_sleep_time = timespec_add(total_sleep_time, ts); } /* re-base the last cycle value */ timekeeper.clock->cycle_last = timekeeper.clock->read(timekeeper.clock); @@ -736,6 +737,7 @@ static void timekeeping_adjust(s64 offset) static cycle_t logarithmic_accumulation(cycle_t offset, int shift) { u64 nsecps = (u64)NSEC_PER_SEC << timekeeper.shift; + u64 raw_nsecs; /* If the offset is smaller then a shifted interval, do nothing */ if (offset < timekeeper.cycle_interval<<shift) @@ -752,16 +754,20 @@ static cycle_t logarithmic_accumulation(cycle_t offset, int shift) 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 raw time */ + raw_nsecs = timekeeper.raw_interval << shift; + raw_nsecs += raw_time.tv_nsec; + if (raw_nsecs >= NSEC_PER_SEC) { + u64 raw_secs = raw_nsecs; + raw_nsecs = do_div(raw_secs, NSEC_PER_SEC); + raw_time.tv_sec += raw_secs; } + raw_time.tv_nsec = raw_nsecs; /* Accumulate error between NTP and clock interval */ timekeeper.ntp_error += tick_length << shift; - timekeeper.ntp_error -= timekeeper.xtime_interval << + timekeeper.ntp_error -= + (timekeeper.xtime_interval + timekeeper.xtime_remainder) << (timekeeper.ntp_error_shift + shift); return offset; @@ -784,10 +790,11 @@ void update_wall_time(void) return; clock = timekeeper.clock; -#ifdef CONFIG_GENERIC_TIME - offset = (clock->read(clock) - clock->cycle_last) & clock->mask; -#else + +#ifdef CONFIG_ARCH_USES_GETTIMEOFFSET offset = timekeeper.cycle_interval; +#else + offset = (clock->read(clock) - clock->cycle_last) & clock->mask; #endif timekeeper.xtime_nsec = (s64)xtime.tv_nsec << timekeeper.shift; @@ -856,7 +863,8 @@ void update_wall_time(void) } /* check to see if there is a new clocksource to use */ - update_vsyscall(&xtime, timekeeper.clock, timekeeper.mult); + update_vsyscall(&xtime, &wall_to_monotonic, timekeeper.clock, + timekeeper.mult); } /** @@ -887,7 +895,7 @@ EXPORT_SYMBOL_GPL(getboottime); */ void monotonic_to_bootbased(struct timespec *ts) { - *ts = timespec_add_safe(*ts, total_sleep_time); + *ts = timespec_add(*ts, total_sleep_time); } EXPORT_SYMBOL_GPL(monotonic_to_bootbased); @@ -902,6 +910,11 @@ struct timespec __current_kernel_time(void) return xtime; } +struct timespec __get_wall_to_monotonic(void) +{ + return wall_to_monotonic; +} + struct timespec current_kernel_time(void) { struct timespec now; diff --git a/kernel/time/timer_list.c b/kernel/time/timer_list.c index ab8f5e33fa9..32a19f9397f 100644 --- a/kernel/time/timer_list.c +++ b/kernel/time/timer_list.c @@ -79,26 +79,26 @@ print_active_timers(struct seq_file *m, struct hrtimer_clock_base *base, { struct hrtimer *timer, tmp; unsigned long next = 0, i; - struct rb_node *curr; + struct timerqueue_node *curr; unsigned long flags; next_one: i = 0; raw_spin_lock_irqsave(&base->cpu_base->lock, flags); - curr = base->first; + curr = timerqueue_getnext(&base->active); /* * Crude but we have to do this O(N*N) thing, because * we have to unlock the base when printing: */ while (curr && i < next) { - curr = rb_next(curr); + curr = timerqueue_iterate_next(curr); i++; } if (curr) { - timer = rb_entry(curr, struct hrtimer, node); + timer = container_of(curr, struct hrtimer, node); tmp = *timer; raw_spin_unlock_irqrestore(&base->cpu_base->lock, flags); diff --git a/kernel/timer.c b/kernel/timer.c index be394af5bc2..43ca9936f2d 100644 --- a/kernel/timer.c +++ b/kernel/timer.c @@ -37,7 +37,7 @@ #include <linux/delay.h> #include <linux/tick.h> #include <linux/kallsyms.h> -#include <linux/perf_event.h> +#include <linux/irq_work.h> #include <linux/sched.h> #include <linux/slab.h> @@ -88,13 +88,6 @@ struct tvec_base boot_tvec_bases; EXPORT_SYMBOL(boot_tvec_bases); static DEFINE_PER_CPU(struct tvec_base *, tvec_bases) = &boot_tvec_bases; -/* - * Note that all tvec_bases are 2 byte aligned and lower bit of - * base in timer_list is guaranteed to be zero. Use the LSB for - * the new flag to indicate whether the timer is deferrable - */ -#define TBASE_DEFERRABLE_FLAG (0x1) - /* Functions below help us manage 'deferrable' flag */ static inline unsigned int tbase_get_deferrable(struct tvec_base *base) { @@ -108,8 +101,7 @@ static inline struct tvec_base *tbase_get_base(struct tvec_base *base) static inline void timer_set_deferrable(struct timer_list *timer) { - timer->base = ((struct tvec_base *)((unsigned long)(timer->base) | - TBASE_DEFERRABLE_FLAG)); + timer->base = TBASE_MAKE_DEFERRED(timer->base); } static inline void @@ -321,6 +313,7 @@ EXPORT_SYMBOL_GPL(round_jiffies_up_relative); /** * set_timer_slack - set the allowed slack for a timer + * @timer: the timer to be modified * @slack_hz: the amount of time (in jiffies) allowed for rounding * * Set the amount of time, in jiffies, that a certain timer has @@ -337,15 +330,6 @@ void set_timer_slack(struct timer_list *timer, int slack_hz) } EXPORT_SYMBOL_GPL(set_timer_slack); - -static inline void set_running_timer(struct tvec_base *base, - struct timer_list *timer) -{ -#ifdef CONFIG_SMP - base->running_timer = timer; -#endif -} - static void internal_add_timer(struct tvec_base *base, struct timer_list *timer) { unsigned long expires = timer->expires; @@ -577,6 +561,19 @@ static void __init_timer(struct timer_list *timer, lockdep_init_map(&timer->lockdep_map, name, key, 0); } +void setup_deferrable_timer_on_stack_key(struct timer_list *timer, + const char *name, + struct lock_class_key *key, + void (*function)(unsigned long), + unsigned long data) +{ + timer->function = function; + timer->data = data; + init_timer_on_stack_key(timer, name, key); + timer_set_deferrable(timer); +} +EXPORT_SYMBOL_GPL(setup_deferrable_timer_on_stack_key); + /** * init_timer_key - initialize a timer * @timer: the timer to be initialized @@ -679,12 +676,8 @@ __mod_timer(struct timer_list *timer, unsigned long expires, 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; - } + if (!pinned && get_sysctl_timer_migration() && idle_cpu(cpu)) + cpu = get_nohz_timer_target(); #endif new_base = per_cpu(tvec_bases, cpu); @@ -752,11 +745,15 @@ unsigned long apply_slack(struct timer_list *timer, unsigned long expires) expires_limit = expires; - if (timer->slack > -1) + if (timer->slack >= 0) { expires_limit = expires + timer->slack; - else if (time_after(expires, jiffies)) /* auto slack: use 0.4% */ - expires_limit = expires + (expires - jiffies)/256; + } else { + unsigned long now = jiffies; + /* No slack, if already expired else auto slack 0.4% */ + if (time_after(expires, now)) + expires_limit = expires + (expires - now)/256; + } mask = expires ^ expires_limit; if (mask == 0) return expires; @@ -917,15 +914,12 @@ int del_timer(struct timer_list *timer) } EXPORT_SYMBOL(del_timer); -#ifdef CONFIG_SMP /** * try_to_del_timer_sync - Try to deactivate a timer * @timer: timer do del * * This function tries to deactivate a timer. Upon successful (ret >= 0) * exit the timer is not queued and the handler is not running on any CPU. - * - * It must not be called from interrupt contexts. */ int try_to_del_timer_sync(struct timer_list *timer) { @@ -954,6 +948,7 @@ out: } EXPORT_SYMBOL(try_to_del_timer_sync); +#ifdef CONFIG_SMP /** * del_timer_sync - deactivate a timer and wait for the handler to finish. * @timer: the timer to be deactivated @@ -964,7 +959,7 @@ EXPORT_SYMBOL(try_to_del_timer_sync); * * Synchronization rules: Callers must prevent restarting of the timer, * otherwise this function is meaningless. It must not be called from - * interrupt contexts. The caller must not hold locks which would prevent + * hardirq contexts. The caller must not hold locks which would prevent * completion of the timer's handler. The timer's handler must not call * add_timer_on(). Upon exit the timer is not queued and the handler is * not running on any CPU. @@ -974,14 +969,16 @@ EXPORT_SYMBOL(try_to_del_timer_sync); int del_timer_sync(struct timer_list *timer) { #ifdef CONFIG_LOCKDEP - unsigned long flags; - - local_irq_save(flags); + local_bh_disable(); lock_map_acquire(&timer->lockdep_map); lock_map_release(&timer->lockdep_map); - local_irq_restore(flags); + local_bh_enable(); #endif - + /* + * don't use it in hardirq context, because it + * could lead to deadlock. + */ + WARN_ON(in_irq()); for (;;) { int ret = try_to_del_timer_sync(timer); if (ret >= 0) @@ -1092,7 +1089,7 @@ static inline void __run_timers(struct tvec_base *base) timer_stats_account_timer(timer); - set_running_timer(base, timer); + base->running_timer = timer; detach_timer(timer, 1); spin_unlock_irq(&base->lock); @@ -1100,7 +1097,7 @@ static inline void __run_timers(struct tvec_base *base) spin_lock_irq(&base->lock); } } - set_running_timer(base, NULL); + base->running_timer = NULL; spin_unlock_irq(&base->lock); } @@ -1230,9 +1227,15 @@ static unsigned long cmp_next_hrtimer_event(unsigned long now, */ unsigned long get_next_timer_interrupt(unsigned long now) { - struct tvec_base *base = __get_cpu_var(tvec_bases); + struct tvec_base *base = __this_cpu_read(tvec_bases); unsigned long expires; + /* + * Pretend that there is no timer pending if the cpu is offline. + * Possible pending timers will be migrated later to an active cpu. + */ + if (cpu_is_offline(smp_processor_id())) + return now + NEXT_TIMER_MAX_DELTA; spin_lock(&base->lock); if (time_before_eq(base->next_timer, base->timer_jiffies)) base->next_timer = __next_timer_interrupt(base); @@ -1260,7 +1263,10 @@ void update_process_times(int user_tick) run_local_timers(); rcu_check_callbacks(cpu, user_tick); printk_tick(); - perf_event_do_pending(); +#ifdef CONFIG_IRQ_WORK + if (in_irq()) + irq_work_run(); +#endif scheduler_tick(); run_posix_cpu_timers(p); } @@ -1270,7 +1276,7 @@ void update_process_times(int user_tick) */ static void run_timer_softirq(struct softirq_action *h) { - struct tvec_base *base = __get_cpu_var(tvec_bases); + struct tvec_base *base = __this_cpu_read(tvec_bases); hrtimer_run_pending(); @@ -1285,7 +1291,6 @@ void run_local_timers(void) { hrtimer_run_queues(); raise_softirq(TIMER_SOFTIRQ); - softlockup_tick(); } /* @@ -1298,7 +1303,7 @@ void do_timer(unsigned long ticks) { jiffies_64 += ticks; update_wall_time(); - calc_global_load(); + calc_global_load(ticks); } #ifdef __ARCH_WANT_SYS_ALARM @@ -1680,11 +1685,14 @@ static int __cpuinit timer_cpu_notify(struct notifier_block *self, unsigned long action, void *hcpu) { long cpu = (long)hcpu; + int err; + switch(action) { case CPU_UP_PREPARE: case CPU_UP_PREPARE_FROZEN: - if (init_timers_cpu(cpu) < 0) - return NOTIFY_BAD; + err = init_timers_cpu(cpu); + if (err < 0) + return notifier_from_errno(err); break; #ifdef CONFIG_HOTPLUG_CPU case CPU_DEAD: @@ -1710,7 +1718,7 @@ void __init init_timers(void) init_timer_stats(); - BUG_ON(err == NOTIFY_BAD); + BUG_ON(err != NOTIFY_OK); register_cpu_notifier(&timers_nb); open_softirq(TIMER_SOFTIRQ, run_timer_softirq); } @@ -1743,3 +1751,25 @@ unsigned long msleep_interruptible(unsigned int msecs) } EXPORT_SYMBOL(msleep_interruptible); + +static int __sched do_usleep_range(unsigned long min, unsigned long max) +{ + ktime_t kmin; + unsigned long delta; + + kmin = ktime_set(0, min * NSEC_PER_USEC); + delta = (max - min) * NSEC_PER_USEC; + return schedule_hrtimeout_range(&kmin, delta, HRTIMER_MODE_REL); +} + +/** + * usleep_range - Drop in replacement for udelay where wakeup is flexible + * @min: Minimum time in usecs to sleep + * @max: Maximum time in usecs to sleep + */ +void usleep_range(unsigned long min, unsigned long max) +{ + __set_current_state(TASK_UNINTERRUPTIBLE); + do_usleep_range(min, max); +} +EXPORT_SYMBOL(usleep_range); diff --git a/kernel/trace/Kconfig b/kernel/trace/Kconfig index 8b1797c4545..14674dce77a 100644 --- a/kernel/trace/Kconfig +++ b/kernel/trace/Kconfig @@ -49,6 +49,11 @@ config HAVE_SYSCALL_TRACEPOINTS help See Documentation/trace/ftrace-design.txt +config HAVE_C_RECORDMCOUNT + bool + help + C version of recordmcount available? + config TRACER_MAX_TRACE bool @@ -64,6 +69,21 @@ config EVENT_TRACING select CONTEXT_SWITCH_TRACER bool +config EVENT_POWER_TRACING_DEPRECATED + depends on EVENT_TRACING + bool "Deprecated power event trace API, to be removed" + default y + help + Provides old power event types: + C-state/idle accounting events: + power:power_start + power:power_end + and old cpufreq accounting event: + power:power_frequency + This is for userspace compatibility + and will vanish after 5 kernel iterations, + namely 2.6.41. + config CONTEXT_SWITCH_TRACER bool @@ -121,7 +141,7 @@ if FTRACE config FUNCTION_TRACER bool "Kernel Function Tracer" depends on HAVE_FUNCTION_TRACER - select FRAME_POINTER + select FRAME_POINTER if !ARM_UNWIND && !S390 select KALLSYMS select GENERIC_TRACER select CONTEXT_SWITCH_TRACER @@ -153,7 +173,7 @@ config IRQSOFF_TRACER bool "Interrupts-off Latency Tracer" default n depends on TRACE_IRQFLAGS_SUPPORT - depends on GENERIC_TIME + depends on !ARCH_USES_GETTIMEOFFSET select TRACE_IRQFLAGS select GENERIC_TRACER select TRACER_MAX_TRACE @@ -175,7 +195,7 @@ config IRQSOFF_TRACER config PREEMPT_TRACER bool "Preemption-off Latency Tracer" default n - depends on GENERIC_TIME + depends on !ARCH_USES_GETTIMEOFFSET depends on PREEMPT select GENERIC_TRACER select TRACER_MAX_TRACE @@ -194,15 +214,6 @@ config PREEMPT_TRACER enabled. This option and the irqs-off timing option can be used together or separately.) -config SYSPROF_TRACER - bool "Sysprof Tracer" - depends on X86 - 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" select GENERIC_TRACER @@ -229,23 +240,6 @@ config FTRACE_SYSCALLS help Basic tracer to catch the syscall entry and exit events. -config BOOT_TRACER - bool "Trace boot initcalls" - 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 - 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. - - 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 @@ -325,28 +319,6 @@ config BRANCH_TRACER Say N if unsure. -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 @@ -371,37 +343,6 @@ config STACK_TRACER Say N if unsure. -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 information - 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 has 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 diff --git a/kernel/trace/Makefile b/kernel/trace/Makefile index ffb1a5b0550..761c510a06c 100644 --- a/kernel/trace/Makefile +++ b/kernel/trace/Makefile @@ -30,7 +30,6 @@ 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-$(CONFIG_FUNCTION_TRACER) += trace_functions.o obj-$(CONFIG_IRQSOFF_TRACER) += trace_irqsoff.o obj-$(CONFIG_PREEMPT_TRACER) += trace_irqsoff.o @@ -38,10 +37,8 @@ obj-$(CONFIG_SCHED_TRACER) += trace_sched_wakeup.o obj-$(CONFIG_NOP_TRACER) += trace_nop.o obj-$(CONFIG_STACK_TRACER) += trace_stack.o obj-$(CONFIG_MMIOTRACE) += trace_mmiotrace.o -obj-$(CONFIG_BOOT_TRACER) += trace_boot.o obj-$(CONFIG_FUNCTION_GRAPH_TRACER) += trace_functions_graph.o obj-$(CONFIG_TRACE_BRANCH_PROFILING) += trace_branch.o -obj-$(CONFIG_KMEMTRACE) += kmemtrace.o obj-$(CONFIG_WORKQUEUE_TRACER) += trace_workqueue.o obj-$(CONFIG_BLK_DEV_IO_TRACE) += blktrace.o ifeq ($(CONFIG_BLOCK),y) @@ -55,7 +52,9 @@ obj-$(CONFIG_EVENT_TRACING) += trace_event_perf.o endif obj-$(CONFIG_EVENT_TRACING) += trace_events_filter.o obj-$(CONFIG_KPROBE_EVENT) += trace_kprobe.o -obj-$(CONFIG_KSYM_TRACER) += trace_ksym.o -obj-$(CONFIG_EVENT_TRACING) += power-traces.o +obj-$(CONFIG_TRACEPOINTS) += power-traces.o +ifeq ($(CONFIG_TRACING),y) +obj-$(CONFIG_KGDB_KDB) += trace_kdb.o +endif libftrace-y := ftrace.o diff --git a/kernel/trace/blktrace.c b/kernel/trace/blktrace.c index b3bc91a3f51..153562d0b93 100644 --- a/kernel/trace/blktrace.c +++ b/kernel/trace/blktrace.c @@ -23,7 +23,6 @@ #include <linux/mutex.h> #include <linux/slab.h> #include <linux/debugfs.h> -#include <linux/smp_lock.h> #include <linux/time.h> #include <linux/uaccess.h> @@ -169,9 +168,11 @@ static int act_log_check(struct blk_trace *bt, u32 what, sector_t sector, static const u32 ddir_act[2] = { BLK_TC_ACT(BLK_TC_READ), BLK_TC_ACT(BLK_TC_WRITE) }; +#define BLK_TC_RAHEAD BLK_TC_AHEAD + /* 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)) +#define MASK_TC_BIT(rw, __name) ((rw & REQ_ ## __name) << \ + (ilog2(BLK_TC_ ## __name) + BLK_TC_SHIFT - __REQ_ ## __name)) /* * The worker for the various blk_add_trace*() types. Fills out a @@ -194,9 +195,8 @@ static void __blk_add_trace(struct blk_trace *bt, sector_t sector, int bytes, 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, SYNC); + what |= MASK_TC_BIT(rw, RAHEAD); what |= MASK_TC_BIT(rw, META); what |= MASK_TC_BIT(rw, DISCARD); @@ -323,6 +323,7 @@ static const struct file_operations blk_dropped_fops = { .owner = THIS_MODULE, .open = blk_dropped_open, .read = blk_dropped_read, + .llseek = default_llseek, }; static int blk_msg_open(struct inode *inode, struct file *filp) @@ -362,6 +363,7 @@ static const struct file_operations blk_msg_fops = { .owner = THIS_MODULE, .open = blk_msg_open, .write = blk_msg_write, + .llseek = noop_llseek, }; /* @@ -549,6 +551,41 @@ int blk_trace_setup(struct request_queue *q, char *name, dev_t dev, } EXPORT_SYMBOL_GPL(blk_trace_setup); +#if defined(CONFIG_COMPAT) && defined(CONFIG_X86_64) +static int compat_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; + struct compat_blk_user_trace_setup cbuts; + int ret; + + if (copy_from_user(&cbuts, arg, sizeof(cbuts))) + return -EFAULT; + + buts = (struct blk_user_trace_setup) { + .act_mask = cbuts.act_mask, + .buf_size = cbuts.buf_size, + .buf_nr = cbuts.buf_nr, + .start_lba = cbuts.start_lba, + .end_lba = cbuts.end_lba, + .pid = cbuts.pid, + }; + memcpy(&buts.name, &cbuts.name, 32); + + ret = do_blk_trace_setup(q, name, dev, bdev, &buts); + if (ret) + return ret; + + if (copy_to_user(arg, &buts.name, 32)) { + blk_trace_remove(q); + return -EFAULT; + } + + return 0; +} +#endif + int blk_trace_startstop(struct request_queue *q, int start) { int ret; @@ -608,6 +645,12 @@ int blk_trace_ioctl(struct block_device *bdev, unsigned cmd, char __user *arg) bdevname(bdev, b); ret = blk_trace_setup(q, b, bdev->bd_dev, bdev, arg); break; +#if defined(CONFIG_COMPAT) && defined(CONFIG_X86_64) + case BLKTRACESETUP32: + bdevname(bdev, b); + ret = compat_blk_trace_setup(q, b, bdev->bd_dev, bdev, arg); + break; +#endif case BLKTRACESTART: start = 1; case BLKTRACESTOP: @@ -661,10 +704,13 @@ static void blk_add_trace_rq(struct request_queue *q, struct request *rq, if (likely(!bt)) return; - if (blk_discard_rq(rq)) - rw |= (1 << BIO_RW_DISCARD); + if (rq->cmd_flags & REQ_DISCARD) + rw |= REQ_DISCARD; + + if (rq->cmd_flags & REQ_SECURE) + rw |= REQ_SECURE; - if (blk_pc_request(rq)) { + if (rq->cmd_type == REQ_TYPE_BLOCK_PC) { 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); @@ -675,28 +721,33 @@ static void blk_add_trace_rq(struct request_queue *q, struct request *rq, } } -static void blk_add_trace_rq_abort(struct request_queue *q, struct request *rq) +static void blk_add_trace_rq_abort(void *ignore, + 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) +static void blk_add_trace_rq_insert(void *ignore, + 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) +static void blk_add_trace_rq_issue(void *ignore, + 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, +static void blk_add_trace_rq_requeue(void *ignore, + 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, +static void blk_add_trace_rq_complete(void *ignore, + struct request_queue *q, struct request *rq) { blk_add_trace_rq(q, rq, BLK_TA_COMPLETE); @@ -707,55 +758,66 @@ static void blk_add_trace_rq_complete(struct request_queue *q, * @q: queue the io is for * @bio: the source bio * @what: the action + * @error: error, if any * * 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) + u32 what, int error) { struct blk_trace *bt = q->blk_trace; if (likely(!bt)) return; + if (!error && !bio_flagged(bio, BIO_UPTODATE)) + error = EIO; + __blk_add_trace(bt, bio->bi_sector, bio->bi_size, bio->bi_rw, what, - !bio_flagged(bio, BIO_UPTODATE), 0, NULL); + error, 0, NULL); } -static void blk_add_trace_bio_bounce(struct request_queue *q, struct bio *bio) +static void blk_add_trace_bio_bounce(void *ignore, + struct request_queue *q, struct bio *bio) { - blk_add_trace_bio(q, bio, BLK_TA_BOUNCE); + blk_add_trace_bio(q, bio, BLK_TA_BOUNCE, 0); } -static void blk_add_trace_bio_complete(struct request_queue *q, struct bio *bio) +static void blk_add_trace_bio_complete(void *ignore, + struct request_queue *q, struct bio *bio, + int error) { - blk_add_trace_bio(q, bio, BLK_TA_COMPLETE); + blk_add_trace_bio(q, bio, BLK_TA_COMPLETE, error); } -static void blk_add_trace_bio_backmerge(struct request_queue *q, +static void blk_add_trace_bio_backmerge(void *ignore, + struct request_queue *q, struct bio *bio) { - blk_add_trace_bio(q, bio, BLK_TA_BACKMERGE); + blk_add_trace_bio(q, bio, BLK_TA_BACKMERGE, 0); } -static void blk_add_trace_bio_frontmerge(struct request_queue *q, +static void blk_add_trace_bio_frontmerge(void *ignore, + struct request_queue *q, struct bio *bio) { - blk_add_trace_bio(q, bio, BLK_TA_FRONTMERGE); + blk_add_trace_bio(q, bio, BLK_TA_FRONTMERGE, 0); } -static void blk_add_trace_bio_queue(struct request_queue *q, struct bio *bio) +static void blk_add_trace_bio_queue(void *ignore, + struct request_queue *q, struct bio *bio) { - blk_add_trace_bio(q, bio, BLK_TA_QUEUE); + blk_add_trace_bio(q, bio, BLK_TA_QUEUE, 0); } -static void blk_add_trace_getrq(struct request_queue *q, +static void blk_add_trace_getrq(void *ignore, + struct request_queue *q, struct bio *bio, int rw) { if (bio) - blk_add_trace_bio(q, bio, BLK_TA_GETRQ); + blk_add_trace_bio(q, bio, BLK_TA_GETRQ, 0); else { struct blk_trace *bt = q->blk_trace; @@ -765,11 +827,12 @@ static void blk_add_trace_getrq(struct request_queue *q, } -static void blk_add_trace_sleeprq(struct request_queue *q, +static void blk_add_trace_sleeprq(void *ignore, + struct request_queue *q, struct bio *bio, int rw) { if (bio) - blk_add_trace_bio(q, bio, BLK_TA_SLEEPRQ); + blk_add_trace_bio(q, bio, BLK_TA_SLEEPRQ, 0); else { struct blk_trace *bt = q->blk_trace; @@ -779,7 +842,7 @@ static void blk_add_trace_sleeprq(struct request_queue *q, } } -static void blk_add_trace_plug(struct request_queue *q) +static void blk_add_trace_plug(void *ignore, struct request_queue *q) { struct blk_trace *bt = q->blk_trace; @@ -787,7 +850,7 @@ static void blk_add_trace_plug(struct request_queue *q) __blk_add_trace(bt, 0, 0, 0, BLK_TA_PLUG, 0, 0, NULL); } -static void blk_add_trace_unplug_io(struct request_queue *q) +static void blk_add_trace_unplug_io(void *ignore, struct request_queue *q) { struct blk_trace *bt = q->blk_trace; @@ -800,7 +863,7 @@ static void blk_add_trace_unplug_io(struct request_queue *q) } } -static void blk_add_trace_unplug_timer(struct request_queue *q) +static void blk_add_trace_unplug_timer(void *ignore, struct request_queue *q) { struct blk_trace *bt = q->blk_trace; @@ -813,7 +876,8 @@ static void blk_add_trace_unplug_timer(struct request_queue *q) } } -static void blk_add_trace_split(struct request_queue *q, struct bio *bio, +static void blk_add_trace_split(void *ignore, + struct request_queue *q, struct bio *bio, unsigned int pdu) { struct blk_trace *bt = q->blk_trace; @@ -828,7 +892,8 @@ static void blk_add_trace_split(struct request_queue *q, struct bio *bio, } /** - * blk_add_trace_remap - Add a trace for a remap operation + * blk_add_trace_bio_remap - Add a trace for a bio-remap operation + * @ignore: trace callback data parameter (not used) * @q: queue the io is for * @bio: the source bio * @dev: target device @@ -839,8 +904,9 @@ static void blk_add_trace_split(struct request_queue *q, struct bio *bio, * 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) +static void blk_add_trace_bio_remap(void *ignore, + 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; @@ -859,6 +925,7 @@ static void blk_add_trace_remap(struct request_queue *q, struct bio *bio, /** * blk_add_trace_rq_remap - Add a trace for a request-remap operation + * @ignore: trace callback data parameter (not used) * @q: queue the io is for * @rq: the source request * @dev: target device @@ -869,7 +936,8 @@ static void blk_add_trace_remap(struct request_queue *q, struct bio *bio, * Add a trace for that action. * **/ -static void blk_add_trace_rq_remap(struct request_queue *q, +static void blk_add_trace_rq_remap(void *ignore, + struct request_queue *q, struct request *rq, dev_t dev, sector_t from) { @@ -908,7 +976,7 @@ void blk_add_driver_data(struct request_queue *q, if (likely(!bt)) return; - if (blk_pc_request(rq)) + if (rq->cmd_type == REQ_TYPE_BLOCK_PC) __blk_add_trace(bt, 0, blk_rq_bytes(rq), 0, BLK_TA_DRV_DATA, rq->errors, len, data); else @@ -921,64 +989,64 @@ static void blk_register_tracepoints(void) { int ret; - ret = register_trace_block_rq_abort(blk_add_trace_rq_abort); + ret = register_trace_block_rq_abort(blk_add_trace_rq_abort, NULL); WARN_ON(ret); - ret = register_trace_block_rq_insert(blk_add_trace_rq_insert); + ret = register_trace_block_rq_insert(blk_add_trace_rq_insert, NULL); WARN_ON(ret); - ret = register_trace_block_rq_issue(blk_add_trace_rq_issue); + ret = register_trace_block_rq_issue(blk_add_trace_rq_issue, NULL); WARN_ON(ret); - ret = register_trace_block_rq_requeue(blk_add_trace_rq_requeue); + ret = register_trace_block_rq_requeue(blk_add_trace_rq_requeue, NULL); WARN_ON(ret); - ret = register_trace_block_rq_complete(blk_add_trace_rq_complete); + ret = register_trace_block_rq_complete(blk_add_trace_rq_complete, NULL); WARN_ON(ret); - ret = register_trace_block_bio_bounce(blk_add_trace_bio_bounce); + ret = register_trace_block_bio_bounce(blk_add_trace_bio_bounce, NULL); WARN_ON(ret); - ret = register_trace_block_bio_complete(blk_add_trace_bio_complete); + ret = register_trace_block_bio_complete(blk_add_trace_bio_complete, NULL); WARN_ON(ret); - ret = register_trace_block_bio_backmerge(blk_add_trace_bio_backmerge); + ret = register_trace_block_bio_backmerge(blk_add_trace_bio_backmerge, NULL); WARN_ON(ret); - ret = register_trace_block_bio_frontmerge(blk_add_trace_bio_frontmerge); + ret = register_trace_block_bio_frontmerge(blk_add_trace_bio_frontmerge, NULL); WARN_ON(ret); - ret = register_trace_block_bio_queue(blk_add_trace_bio_queue); + ret = register_trace_block_bio_queue(blk_add_trace_bio_queue, NULL); WARN_ON(ret); - ret = register_trace_block_getrq(blk_add_trace_getrq); + ret = register_trace_block_getrq(blk_add_trace_getrq, NULL); WARN_ON(ret); - ret = register_trace_block_sleeprq(blk_add_trace_sleeprq); + ret = register_trace_block_sleeprq(blk_add_trace_sleeprq, NULL); WARN_ON(ret); - ret = register_trace_block_plug(blk_add_trace_plug); + ret = register_trace_block_plug(blk_add_trace_plug, NULL); WARN_ON(ret); - ret = register_trace_block_unplug_timer(blk_add_trace_unplug_timer); + ret = register_trace_block_unplug_timer(blk_add_trace_unplug_timer, NULL); WARN_ON(ret); - ret = register_trace_block_unplug_io(blk_add_trace_unplug_io); + ret = register_trace_block_unplug_io(blk_add_trace_unplug_io, NULL); WARN_ON(ret); - ret = register_trace_block_split(blk_add_trace_split); + ret = register_trace_block_split(blk_add_trace_split, NULL); WARN_ON(ret); - ret = register_trace_block_remap(blk_add_trace_remap); + ret = register_trace_block_bio_remap(blk_add_trace_bio_remap, NULL); WARN_ON(ret); - ret = register_trace_block_rq_remap(blk_add_trace_rq_remap); + ret = register_trace_block_rq_remap(blk_add_trace_rq_remap, NULL); 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); + unregister_trace_block_rq_remap(blk_add_trace_rq_remap, NULL); + unregister_trace_block_bio_remap(blk_add_trace_bio_remap, NULL); + unregister_trace_block_split(blk_add_trace_split, NULL); + unregister_trace_block_unplug_io(blk_add_trace_unplug_io, NULL); + unregister_trace_block_unplug_timer(blk_add_trace_unplug_timer, NULL); + unregister_trace_block_plug(blk_add_trace_plug, NULL); + unregister_trace_block_sleeprq(blk_add_trace_sleeprq, NULL); + unregister_trace_block_getrq(blk_add_trace_getrq, NULL); + unregister_trace_block_bio_queue(blk_add_trace_bio_queue, NULL); + unregister_trace_block_bio_frontmerge(blk_add_trace_bio_frontmerge, NULL); + unregister_trace_block_bio_backmerge(blk_add_trace_bio_backmerge, NULL); + unregister_trace_block_bio_complete(blk_add_trace_bio_complete, NULL); + unregister_trace_block_bio_bounce(blk_add_trace_bio_bounce, NULL); + unregister_trace_block_rq_complete(blk_add_trace_rq_complete, NULL); + unregister_trace_block_rq_requeue(blk_add_trace_rq_requeue, NULL); + unregister_trace_block_rq_issue(blk_add_trace_rq_issue, NULL); + unregister_trace_block_rq_insert(blk_add_trace_rq_insert, NULL); + unregister_trace_block_rq_abort(blk_add_trace_rq_abort, NULL); tracepoint_synchronize_unregister(); } @@ -1321,7 +1389,7 @@ out: } static enum print_line_t blk_trace_event_print(struct trace_iterator *iter, - int flags) + int flags, struct trace_event *event) { return print_one_line(iter, false); } @@ -1343,7 +1411,8 @@ static int blk_trace_synthesize_old_trace(struct trace_iterator *iter) } static enum print_line_t -blk_trace_event_print_binary(struct trace_iterator *iter, int flags) +blk_trace_event_print_binary(struct trace_iterator *iter, int flags, + struct trace_event *event) { return blk_trace_synthesize_old_trace(iter) ? TRACE_TYPE_HANDLED : TRACE_TYPE_PARTIAL_LINE; @@ -1381,12 +1450,16 @@ static struct tracer blk_tracer __read_mostly = { .set_flag = blk_tracer_set_flag, }; -static struct trace_event trace_blk_event = { - .type = TRACE_BLK, +static struct trace_event_functions trace_blk_event_funcs = { .trace = blk_trace_event_print, .binary = blk_trace_event_print_binary, }; +static struct trace_event trace_blk_event = { + .type = TRACE_BLK, + .funcs = &trace_blk_event_funcs, +}; + static int __init init_blk_tracer(void) { if (!register_ftrace_event(&trace_blk_event)) { @@ -1581,10 +1654,9 @@ static ssize_t sysfs_blk_trace_attr_show(struct device *dev, struct block_device *bdev; ssize_t ret = -ENXIO; - lock_kernel(); bdev = bdget(part_devt(p)); if (bdev == NULL) - goto out_unlock_kernel; + goto out; q = blk_trace_get_queue(bdev); if (q == NULL) @@ -1612,8 +1684,7 @@ out_unlock_bdev: mutex_unlock(&bdev->bd_mutex); out_bdput: bdput(bdev); -out_unlock_kernel: - unlock_kernel(); +out: return ret; } @@ -1643,11 +1714,10 @@ static ssize_t sysfs_blk_trace_attr_store(struct device *dev, ret = -ENXIO; - lock_kernel(); p = dev_to_part(dev); bdev = bdget(part_devt(p)); if (bdev == NULL) - goto out_unlock_kernel; + goto out; q = blk_trace_get_queue(bdev); if (q == NULL) @@ -1682,8 +1752,6 @@ out_unlock_bdev: mutex_unlock(&bdev->bd_mutex); out_bdput: bdput(bdev); -out_unlock_kernel: - unlock_kernel(); out: return ret ? ret : count; } @@ -1708,7 +1776,7 @@ void blk_dump_cmd(char *buf, struct request *rq) int len = rq->cmd_len; unsigned char *cmd = rq->cmd; - if (!blk_pc_request(rq)) { + if (rq->cmd_type != REQ_TYPE_BLOCK_PC) { buf[0] = '\0'; return; } @@ -1733,21 +1801,21 @@ void blk_fill_rwbs(char *rwbs, u32 rw, int bytes) if (rw & WRITE) rwbs[i++] = 'W'; - else if (rw & 1 << BIO_RW_DISCARD) + else if (rw & REQ_DISCARD) rwbs[i++] = 'D'; else if (bytes) rwbs[i++] = 'R'; else rwbs[i++] = 'N'; - if (rw & 1 << BIO_RW_AHEAD) + if (rw & REQ_RAHEAD) rwbs[i++] = 'A'; - if (rw & 1 << BIO_RW_BARRIER) - rwbs[i++] = 'B'; - if (rw & 1 << BIO_RW_SYNCIO) + if (rw & REQ_SYNC) rwbs[i++] = 'S'; - if (rw & 1 << BIO_RW_META) + if (rw & REQ_META) rwbs[i++] = 'M'; + if (rw & REQ_SECURE) + rwbs[i++] = 'E'; rwbs[i] = '\0'; } @@ -1757,8 +1825,11 @@ 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); + if (rq->cmd_flags & REQ_DISCARD) + rw |= REQ_DISCARD; + + if (rq->cmd_flags & REQ_SECURE) + rw |= REQ_SECURE; bytes = blk_rq_bytes(rq); diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index 32837e19e3b..f3dadae8388 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -381,12 +381,19 @@ static int function_stat_show(struct seq_file *m, void *v) { struct ftrace_profile *rec = v; char str[KSYM_SYMBOL_LEN]; + int ret = 0; #ifdef CONFIG_FUNCTION_GRAPH_TRACER - static DEFINE_MUTEX(mutex); static struct trace_seq s; unsigned long long avg; unsigned long long stddev; #endif + mutex_lock(&ftrace_profile_lock); + + /* we raced with function_profile_reset() */ + if (unlikely(rec->counter == 0)) { + ret = -EBUSY; + goto out; + } kallsyms_lookup(rec->ip, NULL, NULL, NULL, str); seq_printf(m, " %-30.30s %10lu", str, rec->counter); @@ -408,7 +415,6 @@ static int function_stat_show(struct seq_file *m, void *v) do_div(stddev, (rec->counter - 1) * 1000); } - mutex_lock(&mutex); trace_seq_init(&s); trace_print_graph_duration(rec->time, &s); trace_seq_puts(&s, " "); @@ -416,11 +422,12 @@ static int function_stat_show(struct seq_file *m, void *v) trace_seq_puts(&s, " "); trace_print_graph_duration(stddev, &s); trace_print_seq(m, &s); - mutex_unlock(&mutex); #endif seq_putc(m, '\n'); +out: + mutex_unlock(&ftrace_profile_lock); - return 0; + return ret; } static void ftrace_profile_reset(struct ftrace_profile_stat *stat) @@ -793,6 +800,7 @@ static const struct file_operations ftrace_profile_fops = { .open = tracing_open_generic, .read = ftrace_profile_read, .write = ftrace_profile_write, + .llseek = default_llseek, }; /* used to initialize the real stat files */ @@ -877,10 +885,8 @@ enum { FTRACE_ENABLE_CALLS = (1 << 0), FTRACE_DISABLE_CALLS = (1 << 1), FTRACE_UPDATE_TRACE_FUNC = (1 << 2), - FTRACE_ENABLE_MCOUNT = (1 << 3), - FTRACE_DISABLE_MCOUNT = (1 << 4), - FTRACE_START_FUNC_RET = (1 << 5), - FTRACE_STOP_FUNC_RET = (1 << 6), + FTRACE_START_FUNC_RET = (1 << 3), + FTRACE_STOP_FUNC_RET = (1 << 4), }; static int ftrace_filtered; @@ -1219,8 +1225,6 @@ static void ftrace_shutdown(int command) static void ftrace_startup_sysctl(void) { - int command = FTRACE_ENABLE_MCOUNT; - if (unlikely(ftrace_disabled)) return; @@ -1228,23 +1232,17 @@ static void ftrace_startup_sysctl(void) saved_ftrace_func = NULL; /* ftrace_start_up is true if we want ftrace running */ if (ftrace_start_up) - command |= FTRACE_ENABLE_CALLS; - - ftrace_run_update_code(command); + ftrace_run_update_code(FTRACE_ENABLE_CALLS); } static void ftrace_shutdown_sysctl(void) { - int command = FTRACE_DISABLE_MCOUNT; - if (unlikely(ftrace_disabled)) return; /* ftrace_start_up is true if ftrace is running */ if (ftrace_start_up) - command |= FTRACE_DISABLE_CALLS; - - ftrace_run_update_code(command); + ftrace_run_update_code(FTRACE_DISABLE_CALLS); } static cycle_t ftrace_update_time; @@ -1361,24 +1359,29 @@ enum { #define FTRACE_BUFF_MAX (KSYM_SYMBOL_LEN+4) /* room for wildcards */ struct ftrace_iterator { - struct ftrace_page *pg; - int hidx; - int idx; - unsigned flags; - struct trace_parser parser; + loff_t pos; + loff_t func_pos; + struct ftrace_page *pg; + struct dyn_ftrace *func; + struct ftrace_func_probe *probe; + struct trace_parser parser; + int hidx; + int idx; + unsigned flags; }; static void * -t_hash_next(struct seq_file *m, void *v, loff_t *pos) +t_hash_next(struct seq_file *m, loff_t *pos) { struct ftrace_iterator *iter = m->private; - struct hlist_node *hnd = v; + struct hlist_node *hnd = NULL; struct hlist_head *hhd; - WARN_ON(!(iter->flags & FTRACE_ITER_HASH)); - (*pos)++; + iter->pos = *pos; + if (iter->probe) + hnd = &iter->probe->node; retry: if (iter->hidx >= FTRACE_FUNC_HASHSIZE) return NULL; @@ -1401,7 +1404,12 @@ t_hash_next(struct seq_file *m, void *v, loff_t *pos) } } - return hnd; + if (WARN_ON_ONCE(!hnd)) + return NULL; + + iter->probe = hlist_entry(hnd, struct ftrace_func_probe, node); + + return iter; } static void *t_hash_start(struct seq_file *m, loff_t *pos) @@ -1410,26 +1418,32 @@ static void *t_hash_start(struct seq_file *m, loff_t *pos) void *p = NULL; loff_t l; - if (!(iter->flags & FTRACE_ITER_HASH)) - *pos = 0; - - iter->flags |= FTRACE_ITER_HASH; + if (iter->func_pos > *pos) + return NULL; iter->hidx = 0; - for (l = 0; l <= *pos; ) { - p = t_hash_next(m, p, &l); + for (l = 0; l <= (*pos - iter->func_pos); ) { + p = t_hash_next(m, &l); if (!p) break; } - return p; + if (!p) + return NULL; + + /* Only set this if we have an item */ + iter->flags |= FTRACE_ITER_HASH; + + return iter; } -static int t_hash_show(struct seq_file *m, void *v) +static int +t_hash_show(struct seq_file *m, struct ftrace_iterator *iter) { struct ftrace_func_probe *rec; - struct hlist_node *hnd = v; - rec = hlist_entry(hnd, struct ftrace_func_probe, node); + rec = iter->probe; + if (WARN_ON_ONCE(!rec)) + return -EIO; if (rec->ops->print) return rec->ops->print(m, rec->ip, rec->ops, rec->data); @@ -1450,12 +1464,13 @@ t_next(struct seq_file *m, void *v, loff_t *pos) struct dyn_ftrace *rec = NULL; if (iter->flags & FTRACE_ITER_HASH) - return t_hash_next(m, v, pos); + return t_hash_next(m, pos); (*pos)++; + iter->pos = *pos; if (iter->flags & FTRACE_ITER_PRINTALL) - return NULL; + return t_hash_start(m, pos); retry: if (iter->idx >= iter->pg->index) { @@ -1484,7 +1499,20 @@ t_next(struct seq_file *m, void *v, loff_t *pos) } } - return rec; + if (!rec) + return t_hash_start(m, pos); + + iter->func_pos = *pos; + iter->func = rec; + + return iter; +} + +static void reset_iter_read(struct ftrace_iterator *iter) +{ + iter->pos = 0; + iter->func_pos = 0; + iter->flags &= ~(FTRACE_ITER_PRINTALL & FTRACE_ITER_HASH); } static void *t_start(struct seq_file *m, loff_t *pos) @@ -1495,6 +1523,12 @@ static void *t_start(struct seq_file *m, loff_t *pos) mutex_lock(&ftrace_lock); /* + * If an lseek was done, then reset and start from beginning. + */ + if (*pos < iter->pos) + reset_iter_read(iter); + + /* * For set_ftrace_filter reading, if we have the filter * off, we can short cut and just print out that all * functions are enabled. @@ -1503,12 +1537,19 @@ static void *t_start(struct seq_file *m, loff_t *pos) if (*pos > 0) return t_hash_start(m, pos); iter->flags |= FTRACE_ITER_PRINTALL; + /* reset in case of seek/pread */ + iter->flags &= ~FTRACE_ITER_HASH; return iter; } if (iter->flags & FTRACE_ITER_HASH) return t_hash_start(m, pos); + /* + * Unfortunately, we need to restart at ftrace_pages_start + * every time we let go of the ftrace_mutex. This is because + * those pointers can change without the lock. + */ iter->pg = ftrace_pages_start; iter->idx = 0; for (l = 0; l <= *pos; ) { @@ -1517,10 +1558,14 @@ static void *t_start(struct seq_file *m, loff_t *pos) break; } - if (!p && iter->flags & FTRACE_ITER_FILTER) - return t_hash_start(m, pos); + if (!p) { + if (iter->flags & FTRACE_ITER_FILTER) + return t_hash_start(m, pos); - return p; + return NULL; + } + + return iter; } static void t_stop(struct seq_file *m, void *p) @@ -1531,16 +1576,18 @@ static void t_stop(struct seq_file *m, void *p) static int t_show(struct seq_file *m, void *v) { struct ftrace_iterator *iter = m->private; - struct dyn_ftrace *rec = v; + struct dyn_ftrace *rec; if (iter->flags & FTRACE_ITER_HASH) - return t_hash_show(m, v); + return t_hash_show(m, iter); if (iter->flags & FTRACE_ITER_PRINTALL) { seq_printf(m, "#### all functions enabled ####\n"); return 0; } + rec = iter->func; + if (!rec) return 0; @@ -1592,8 +1639,8 @@ ftrace_failures_open(struct inode *inode, struct file *file) ret = ftrace_avail_open(inode, file); if (!ret) { - m = (struct seq_file *)file->private_data; - iter = (struct ftrace_iterator *)m->private; + m = file->private_data; + iter = m->private; iter->flags = FTRACE_ITER_FAILURES; } @@ -1883,7 +1930,6 @@ function_trace_probe_call(unsigned long ip, unsigned long parent_ip) struct hlist_head *hhd; struct hlist_node *n; unsigned long key; - int resched; key = hash_long(ip, FTRACE_HASH_BITS); @@ -1897,12 +1943,12 @@ function_trace_probe_call(unsigned long ip, unsigned long parent_ip) * period. This syncs the hash iteration and freeing of items * on the hash. rcu_read_lock is too dangerous here. */ - resched = ftrace_preempt_disable(); + preempt_disable_notrace(); 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); + preempt_enable_notrace(); } static struct ftrace_ops trace_probe_ops __read_mostly = @@ -2624,6 +2670,7 @@ static const struct file_operations ftrace_graph_fops = { .read = seq_read, .write = ftrace_graph_write, .release = ftrace_graph_release, + .llseek = seq_lseek, }; #endif /* CONFIG_FUNCTION_GRAPH_TRACER */ @@ -3234,7 +3281,8 @@ free: } static void -ftrace_graph_probe_sched_switch(struct task_struct *prev, struct task_struct *next) +ftrace_graph_probe_sched_switch(void *ignore, + struct task_struct *prev, struct task_struct *next) { unsigned long long timestamp; int index; @@ -3288,7 +3336,7 @@ static int start_graph_tracing(void) } while (ret == -EAGAIN); if (!ret) { - ret = register_trace_sched_switch(ftrace_graph_probe_sched_switch); + ret = register_trace_sched_switch(ftrace_graph_probe_sched_switch, NULL); if (ret) pr_info("ftrace_graph: Couldn't activate tracepoint" " probe to kernel_sched_switch\n"); @@ -3364,7 +3412,7 @@ void unregister_ftrace_graph(void) ftrace_graph_entry = ftrace_graph_entry_stub; ftrace_shutdown(FTRACE_STOP_FUNC_RET); unregister_pm_notifier(&ftrace_suspend_notifier); - unregister_trace_sched_switch(ftrace_graph_probe_sched_switch); + unregister_trace_sched_switch(ftrace_graph_probe_sched_switch, NULL); out: mutex_unlock(&ftrace_lock); diff --git a/kernel/trace/kmemtrace.c b/kernel/trace/kmemtrace.c deleted file mode 100644 index a91da69f153..00000000000 --- a/kernel/trace/kmemtrace.c +++ /dev/null @@ -1,511 +0,0 @@ -/* - * 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 index a22582a0616..f55fcf61b22 100644 --- a/kernel/trace/power-traces.c +++ b/kernel/trace/power-traces.c @@ -13,5 +13,8 @@ #define CREATE_TRACE_POINTS #include <trace/events/power.h> -EXPORT_TRACEPOINT_SYMBOL_GPL(power_frequency); +#ifdef EVENT_POWER_TRACING_DEPRECATED +EXPORT_TRACEPOINT_SYMBOL_GPL(power_start); +#endif +EXPORT_TRACEPOINT_SYMBOL_GPL(cpu_idle); diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c index 7f6059c5aa9..bd1c35a4fbc 100644 --- a/kernel/trace/ring_buffer.c +++ b/kernel/trace/ring_buffer.c @@ -224,6 +224,9 @@ enum { RB_LEN_TIME_STAMP = 16, }; +#define skip_time_extend(event) \ + ((struct ring_buffer_event *)((char *)event + RB_LEN_TIME_EXTEND)) + static inline int rb_null_event(struct ring_buffer_event *event) { return event->type_len == RINGBUF_TYPE_PADDING && !event->time_delta; @@ -248,8 +251,12 @@ rb_event_data_length(struct ring_buffer_event *event) return length + RB_EVNT_HDR_SIZE; } -/* inline for ring buffer fast paths */ -static unsigned +/* + * Return the length of the given event. Will return + * the length of the time extend if the event is a + * time extend. + */ +static inline unsigned rb_event_length(struct ring_buffer_event *event) { switch (event->type_len) { @@ -274,13 +281,41 @@ rb_event_length(struct ring_buffer_event *event) return 0; } +/* + * Return total length of time extend and data, + * or just the event length for all other events. + */ +static inline unsigned +rb_event_ts_length(struct ring_buffer_event *event) +{ + unsigned len = 0; + + if (event->type_len == RINGBUF_TYPE_TIME_EXTEND) { + /* time extends include the data event after it */ + len = RB_LEN_TIME_EXTEND; + event = skip_time_extend(event); + } + return len + rb_event_length(event); +} + /** * ring_buffer_event_length - return the length of the event * @event: the event to get the length of + * + * Returns the size of the data load of a data event. + * If the event is something other than a data event, it + * returns the size of the event itself. With the exception + * of a TIME EXTEND, where it still returns the size of the + * data load of the data event after it. */ unsigned ring_buffer_event_length(struct ring_buffer_event *event) { - unsigned length = rb_event_length(event); + unsigned length; + + if (event->type_len == RINGBUF_TYPE_TIME_EXTEND) + event = skip_time_extend(event); + + length = rb_event_length(event); if (event->type_len > RINGBUF_TYPE_DATA_TYPE_LEN_MAX) return length; length -= RB_EVNT_HDR_SIZE; @@ -294,6 +329,8 @@ EXPORT_SYMBOL_GPL(ring_buffer_event_length); static void * rb_event_data(struct ring_buffer_event *event) { + if (event->type_len == RINGBUF_TYPE_TIME_EXTEND) + event = skip_time_extend(event); 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->type_len) @@ -404,9 +441,6 @@ static inline int test_time_stamp(u64 delta) /* 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; @@ -443,6 +477,7 @@ int ring_buffer_print_page_header(struct trace_seq *s) */ struct ring_buffer_per_cpu { int cpu; + atomic_t record_disabled; struct ring_buffer *buffer; spinlock_t reader_lock; /* serialize readers */ arch_spinlock_t lock; @@ -462,7 +497,6 @@ struct ring_buffer_per_cpu { unsigned long read; u64 write_stamp; u64 read_stamp; - atomic_t record_disabled; }; struct ring_buffer { @@ -1546,6 +1580,25 @@ static void rb_inc_iter(struct ring_buffer_iter *iter) iter->head = 0; } +/* Slow path, do not inline */ +static noinline struct ring_buffer_event * +rb_add_time_stamp(struct ring_buffer_event *event, u64 delta) +{ + event->type_len = RINGBUF_TYPE_TIME_EXTEND; + + /* Not the first event on the page? */ + if (rb_event_index(event)) { + event->time_delta = delta & TS_MASK; + event->array[0] = delta >> TS_SHIFT; + } else { + /* nope, just zero it */ + event->time_delta = 0; + event->array[0] = 0; + } + + return skip_time_extend(event); +} + /** * ring_buffer_update_event - update event type and data * @event: the even to update @@ -1558,28 +1611,31 @@ static void rb_inc_iter(struct ring_buffer_iter *iter) * data field. */ static void -rb_update_event(struct ring_buffer_event *event, - unsigned type, unsigned length) +rb_update_event(struct ring_buffer_per_cpu *cpu_buffer, + struct ring_buffer_event *event, unsigned length, + int add_timestamp, u64 delta) { - event->type_len = type; - - switch (type) { - - case RINGBUF_TYPE_PADDING: - case RINGBUF_TYPE_TIME_EXTEND: - case RINGBUF_TYPE_TIME_STAMP: - break; + /* Only a commit updates the timestamp */ + if (unlikely(!rb_event_is_commit(cpu_buffer, event))) + delta = 0; - case 0: - length -= RB_EVNT_HDR_SIZE; - if (length > RB_MAX_SMALL_DATA || RB_FORCE_8BYTE_ALIGNMENT) - event->array[0] = length; - else - event->type_len = DIV_ROUND_UP(length, RB_ALIGNMENT); - break; - default: - BUG(); + /* + * If we need to add a timestamp, then we + * add it to the start of the resevered space. + */ + if (unlikely(add_timestamp)) { + event = rb_add_time_stamp(event, delta); + length -= RB_LEN_TIME_EXTEND; + delta = 0; } + + event->time_delta = delta; + length -= RB_EVNT_HDR_SIZE; + if (length > RB_MAX_SMALL_DATA || RB_FORCE_8BYTE_ALIGNMENT) { + event->type_len = 0; + event->array[0] = length; + } else + event->type_len = DIV_ROUND_UP(length, RB_ALIGNMENT); } /* @@ -1768,6 +1824,14 @@ rb_reset_tail(struct ring_buffer_per_cpu *cpu_buffer, * must fill the old tail_page with padding. */ if (tail >= BUF_PAGE_SIZE) { + /* + * If the page was filled, then we still need + * to update the real_end. Reset it to zero + * and the reader will ignore it. + */ + if (tail == BUF_PAGE_SIZE) + tail_page->real_end = 0; + local_sub(length, &tail_page->write); return; } @@ -1815,10 +1879,13 @@ rb_reset_tail(struct ring_buffer_per_cpu *cpu_buffer, local_sub(length, &tail_page->write); } -static struct ring_buffer_event * +/* + * This is the slow path, force gcc not to inline it. + */ +static noinline 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 *tail_page, u64 ts) { struct buffer_page *commit_page = cpu_buffer->commit_page; struct ring_buffer *buffer = cpu_buffer->buffer; @@ -1901,8 +1968,8 @@ rb_move_tail(struct ring_buffer_per_cpu *cpu_buffer, * Nested commits always have zero deltas, so * just reread the time stamp */ - *ts = rb_time_stamp(buffer); - next_page->page->time_stamp = *ts; + ts = rb_time_stamp(buffer); + next_page->page->time_stamp = ts; } out_again: @@ -1921,12 +1988,21 @@ rb_move_tail(struct ring_buffer_per_cpu *cpu_buffer, static struct ring_buffer_event * __rb_reserve_next(struct ring_buffer_per_cpu *cpu_buffer, - unsigned type, unsigned long length, u64 *ts) + unsigned long length, u64 ts, + u64 delta, int add_timestamp) { struct buffer_page *tail_page; struct ring_buffer_event *event; unsigned long tail, write; + /* + * If the time delta since the last event is too big to + * hold in the time field of the event, then we append a + * TIME EXTEND event ahead of the data event. + */ + if (unlikely(add_timestamp)) + length += RB_LEN_TIME_EXTEND; + tail_page = cpu_buffer->tail_page; write = local_add_return(length, &tail_page->write); @@ -1935,7 +2011,7 @@ __rb_reserve_next(struct ring_buffer_per_cpu *cpu_buffer, tail = write - length; /* See if we shot pass the end of this buffer page */ - if (write > BUF_PAGE_SIZE) + if (unlikely(write > BUF_PAGE_SIZE)) return rb_move_tail(cpu_buffer, length, tail, tail_page, ts); @@ -1943,18 +2019,16 @@ __rb_reserve_next(struct ring_buffer_per_cpu *cpu_buffer, event = __rb_page_index(tail_page, tail); kmemcheck_annotate_bitfield(event, bitfield); - rb_update_event(event, type, length); + rb_update_event(cpu_buffer, event, length, add_timestamp, delta); - /* The passed in type is zero for DATA */ - if (likely(!type)) - local_inc(&tail_page->entries); + local_inc(&tail_page->entries); /* * If this is the first commit on the page, then update * its timestamp. */ if (!tail) - tail_page->page->time_stamp = *ts; + tail_page->page->time_stamp = ts; return event; } @@ -1969,7 +2043,7 @@ rb_try_to_discard(struct ring_buffer_per_cpu *cpu_buffer, unsigned long addr; new_index = rb_event_index(event); - old_index = new_index + rb_event_length(event); + old_index = new_index + rb_event_ts_length(event); addr = (unsigned long)event; addr &= PAGE_MASK; @@ -1995,76 +2069,13 @@ rb_try_to_discard(struct ring_buffer_per_cpu *cpu_buffer, return 0; } -static int -rb_add_time_stamp(struct ring_buffer_per_cpu *cpu_buffer, - u64 *ts, u64 *delta) -{ - struct ring_buffer_event *event; - int ret; - - WARN_ONCE(*delta > (1ULL << 59), - KERN_WARNING "Delta way too big! %llu ts=%llu write stamp = %llu\n", - (unsigned long long)*delta, - (unsigned long long)*ts, - (unsigned long long)cpu_buffer->write_stamp); - - /* - * The delta is too big, we to add a - * new timestamp. - */ - event = __rb_reserve_next(cpu_buffer, - RINGBUF_TYPE_TIME_EXTEND, - RB_LEN_TIME_EXTEND, - ts); - if (!event) - return -EBUSY; - - if (PTR_ERR(event) == -EAGAIN) - return -EAGAIN; - - /* Only a commited time event can update the write stamp */ - if (rb_event_is_commit(cpu_buffer, event)) { - /* - * 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 { - /* 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 { - /* 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; - } - - *delta = 0; - - 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) +static inline void rb_end_commit(struct ring_buffer_per_cpu *cpu_buffer) { unsigned long commits; @@ -2102,9 +2113,10 @@ rb_reserve_next_event(struct ring_buffer *buffer, unsigned long length) { struct ring_buffer_event *event; - u64 ts, delta = 0; - int commit = 0; + u64 ts, delta; int nr_loops = 0; + int add_timestamp; + u64 diff; rb_start_commit(cpu_buffer); @@ -2125,6 +2137,9 @@ rb_reserve_next_event(struct ring_buffer *buffer, length = rb_calculate_event_length(length); again: + add_timestamp = 0; + delta = 0; + /* * We allow for interrupts to reenter here and do a trace. * If one does, it will cause this original code to loop @@ -2138,56 +2153,32 @@ rb_reserve_next_event(struct ring_buffer *buffer, goto out_fail; ts = rb_time_stamp(cpu_buffer->buffer); + diff = ts - cpu_buffer->write_stamp; - /* - * Only the first commit can update the timestamp. - * Yes there is a race here. If an interrupt comes in - * just after the conditional and it traces too, then it - * will also check the deltas. More than one timestamp may - * also be made. But only the entry that did the actual - * commit will be something other than zero. - */ - if (likely(cpu_buffer->tail_page == cpu_buffer->commit_page && - rb_page_write(cpu_buffer->tail_page) == - rb_commit_index(cpu_buffer))) { - u64 diff; - - diff = ts - cpu_buffer->write_stamp; - - /* make sure this diff is calculated here */ - barrier(); - - /* Did the write stamp get updated already? */ - if (unlikely(ts < cpu_buffer->write_stamp)) - goto get_event; + /* make sure this diff is calculated here */ + barrier(); + /* Did the write stamp get updated already? */ + if (likely(ts >= cpu_buffer->write_stamp)) { delta = diff; if (unlikely(test_time_stamp(delta))) { - - commit = rb_add_time_stamp(cpu_buffer, &ts, &delta); - if (commit == -EBUSY) - goto out_fail; - - if (commit == -EAGAIN) - goto again; - - RB_WARN_ON(cpu_buffer, commit < 0); + WARN_ONCE(delta > (1ULL << 59), + KERN_WARNING "Delta way too big! %llu ts=%llu write stamp = %llu\n", + (unsigned long long)delta, + (unsigned long long)ts, + (unsigned long long)cpu_buffer->write_stamp); + add_timestamp = 1; } } - get_event: - event = __rb_reserve_next(cpu_buffer, 0, length, &ts); + event = __rb_reserve_next(cpu_buffer, length, ts, + delta, add_timestamp); if (unlikely(PTR_ERR(event) == -EAGAIN)) goto again; if (!event) goto out_fail; - if (!rb_event_is_commit(cpu_buffer, event)) - delta = 0; - - event->time_delta = delta; - return event; out_fail: @@ -2199,13 +2190,9 @@ rb_reserve_next_event(struct ring_buffer *buffer, #define TRACE_RECURSIVE_DEPTH 16 -static int trace_recursive_lock(void) +/* Keep this code out of the fast path cache */ +static noinline void trace_recursive_fail(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(); @@ -2217,10 +2204,21 @@ static int trace_recursive_lock(void) in_nmi()); WARN_ON_ONCE(1); +} + +static inline int trace_recursive_lock(void) +{ + current->trace_recursion++; + + if (likely(current->trace_recursion < TRACE_RECURSIVE_DEPTH)) + return 0; + + trace_recursive_fail(); + return -1; } -static void trace_recursive_unlock(void) +static inline void trace_recursive_unlock(void) { WARN_ON_ONCE(!current->trace_recursion); @@ -2234,8 +2232,6 @@ static void trace_recursive_unlock(void) #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 @@ -2256,13 +2252,13 @@ ring_buffer_lock_reserve(struct ring_buffer *buffer, unsigned long length) { struct ring_buffer_per_cpu *cpu_buffer; struct ring_buffer_event *event; - int cpu, resched; + int cpu; if (ring_buffer_flags != RB_BUFFERS_ON) return NULL; /* If we are tracing schedule, we don't want to recurse */ - resched = ftrace_preempt_disable(); + preempt_disable_notrace(); if (atomic_read(&buffer->record_disabled)) goto out_nocheck; @@ -2287,21 +2283,13 @@ ring_buffer_lock_reserve(struct ring_buffer *buffer, unsigned long length) if (!event) goto out; - /* - * Need to store resched state on this cpu. - * Only the first needs to. - */ - - if (preempt_count() == 1) - per_cpu(rb_need_resched, cpu) = resched; - return event; out: trace_recursive_unlock(); out_nocheck: - ftrace_preempt_enable(resched); + preempt_enable_notrace(); return NULL; } EXPORT_SYMBOL_GPL(ring_buffer_lock_reserve); @@ -2310,12 +2298,28 @@ static void rb_update_write_stamp(struct ring_buffer_per_cpu *cpu_buffer, struct ring_buffer_event *event) { + u64 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; + if (rb_event_is_commit(cpu_buffer, event)) { + /* + * A commit event that is first on a page + * updates the write timestamp with the page stamp + */ + if (!rb_event_index(event)) + cpu_buffer->write_stamp = + cpu_buffer->commit_page->page->time_stamp; + else if (event->type_len == RINGBUF_TYPE_TIME_EXTEND) { + delta = event->array[0]; + delta <<= TS_SHIFT; + delta += event->time_delta; + cpu_buffer->write_stamp += delta; + } else + cpu_buffer->write_stamp += event->time_delta; + } } static void rb_commit(struct ring_buffer_per_cpu *cpu_buffer, @@ -2347,13 +2351,7 @@ int ring_buffer_unlock_commit(struct ring_buffer *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(); + preempt_enable_notrace(); return 0; } @@ -2361,6 +2359,9 @@ EXPORT_SYMBOL_GPL(ring_buffer_unlock_commit); static inline void rb_event_discard(struct ring_buffer_event *event) { + if (event->type_len == RINGBUF_TYPE_TIME_EXTEND) + event = skip_time_extend(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; @@ -2461,13 +2462,7 @@ void ring_buffer_discard_commit(struct ring_buffer *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(); + preempt_enable_notrace(); } EXPORT_SYMBOL_GPL(ring_buffer_discard_commit); @@ -2493,12 +2488,12 @@ int ring_buffer_write(struct ring_buffer *buffer, struct ring_buffer_event *event; void *body; int ret = -EBUSY; - int cpu, resched; + int cpu; if (ring_buffer_flags != RB_BUFFERS_ON) return -EBUSY; - resched = ftrace_preempt_disable(); + preempt_disable_notrace(); if (atomic_read(&buffer->record_disabled)) goto out; @@ -2528,7 +2523,7 @@ int ring_buffer_write(struct ring_buffer *buffer, ret = 0; out: - ftrace_preempt_enable(resched); + preempt_enable_notrace(); return ret; } @@ -2620,6 +2615,19 @@ void ring_buffer_record_enable_cpu(struct ring_buffer *buffer, int cpu) } EXPORT_SYMBOL_GPL(ring_buffer_record_enable_cpu); +/* + * The total entries in the ring buffer is the running counter + * of entries entered into the ring buffer, minus the sum of + * the entries read from the ring buffer and the number of + * entries that were overwritten. + */ +static inline unsigned long +rb_num_of_entries(struct ring_buffer_per_cpu *cpu_buffer) +{ + return local_read(&cpu_buffer->entries) - + (local_read(&cpu_buffer->overrun) + cpu_buffer->read); +} + /** * ring_buffer_entries_cpu - get the number of entries in a cpu buffer * @buffer: The ring buffer @@ -2628,16 +2636,13 @@ 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]; - ret = (local_read(&cpu_buffer->entries) - local_read(&cpu_buffer->overrun)) - - cpu_buffer->read; - return ret; + return rb_num_of_entries(cpu_buffer); } EXPORT_SYMBOL_GPL(ring_buffer_entries_cpu); @@ -2698,8 +2703,7 @@ 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 += (local_read(&cpu_buffer->entries) - - local_read(&cpu_buffer->overrun)) - cpu_buffer->read; + entries += rb_num_of_entries(cpu_buffer); } return entries; @@ -2999,13 +3003,11 @@ static void rb_advance_reader(struct ring_buffer_per_cpu *cpu_buffer) static void rb_advance_iter(struct ring_buffer_iter *iter) { - struct ring_buffer *buffer; struct ring_buffer_per_cpu *cpu_buffer; struct ring_buffer_event *event; unsigned length; cpu_buffer = iter->cpu_buffer; - buffer = cpu_buffer->buffer; /* * Check if we are at the end of the buffer. @@ -3056,12 +3058,12 @@ rb_buffer_peek(struct ring_buffer_per_cpu *cpu_buffer, 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, or from discarded - * commits. The most that we can have is the number on a single page. + * We repeat when a time extend is encountered. + * Since the time extend is always attached to a data event, + * we should never loop more than once. + * (We never hit the following condition more than twice). */ - if (RB_WARN_ON(cpu_buffer, ++nr_loops > RB_TIMESTAMPS_PER_PAGE)) + if (RB_WARN_ON(cpu_buffer, ++nr_loops > 2)) return NULL; reader = rb_get_reader_page(cpu_buffer); @@ -3137,14 +3139,12 @@ rb_iter_peek(struct ring_buffer_iter *iter, u64 *ts) return NULL; /* - * 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. + * We repeat when a time extend is encountered. + * Since the time extend is always attached to a data event, + * we should never loop more than once. + * (We never hit the following condition more than twice). */ - if (RB_WARN_ON(cpu_buffer, ++nr_loops > RB_TIMESTAMPS_PER_PAGE)) + if (RB_WARN_ON(cpu_buffer, ++nr_loops > 2)) return NULL; if (rb_per_cpu_empty(cpu_buffer)) @@ -3842,7 +3842,8 @@ int ring_buffer_read_page(struct ring_buffer *buffer, if (len > (commit - read)) len = (commit - read); - size = rb_event_length(event); + /* Always keep the time extend and data together */ + size = rb_event_ts_length(event); if (len < size) goto out_unlock; @@ -3852,6 +3853,13 @@ int ring_buffer_read_page(struct ring_buffer *buffer, /* Need to copy one event at a time */ do { + /* We need the size of one event, because + * rb_advance_reader only advances by one event, + * whereas rb_event_ts_length may include the size of + * one or two events. + * We have already ensured there's enough space if this + * is a time extend. */ + size = rb_event_length(event); memcpy(bpage->data + pos, rpage->data + rpos, size); len -= size; @@ -3860,9 +3868,13 @@ int ring_buffer_read_page(struct ring_buffer *buffer, rpos = reader->read; pos += size; + if (rpos >= commit) + break; + event = rb_reader_event(cpu_buffer); - size = rb_event_length(event); - } while (len > size); + /* Always keep the time extend and data together */ + size = rb_event_ts_length(event); + } while (len >= size); /* update bpage */ local_set(&bpage->commit, pos); @@ -3894,12 +3906,12 @@ int ring_buffer_read_page(struct ring_buffer *buffer, ret = read; cpu_buffer->lost_events = 0; + + commit = local_read(&bpage->commit); /* * Set a flag in the commit field if we lost events */ if (missed_events) { - commit = local_read(&bpage->commit); - /* If there is room at the end of the page to save the * missed events, then record it there. */ @@ -3907,10 +3919,17 @@ int ring_buffer_read_page(struct ring_buffer *buffer, memcpy(&bpage->data[commit], &missed_events, sizeof(missed_events)); local_add(RB_MISSED_STORED, &bpage->commit); + commit += sizeof(missed_events); } local_add(RB_MISSED_EVENTS, &bpage->commit); } + /* + * This page may be off to user land. Zero it out here. + */ + if (commit < BUF_PAGE_SIZE) + memset(&bpage->data[commit], 0, BUF_PAGE_SIZE - commit); + out_unlock: spin_unlock_irqrestore(&cpu_buffer->reader_lock, flags); @@ -3971,6 +3990,7 @@ static const struct file_operations rb_simple_fops = { .open = tracing_open_generic, .read = rb_simple_read, .write = rb_simple_write, + .llseek = default_llseek, }; diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 8a76339a9e6..dc53ecb8058 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -17,7 +17,6 @@ #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> @@ -101,10 +100,7 @@ static inline void ftrace_enable_cpu(void) preempt_enable(); } -static cpumask_var_t __read_mostly tracing_buffer_mask; - -#define for_each_tracing_cpu(cpu) \ - for_each_cpu(cpu, tracing_buffer_mask) +cpumask_var_t __read_mostly tracing_buffer_mask; /* * ftrace_dump_on_oops - variable to dump ftrace buffer on oops @@ -344,7 +340,7 @@ 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_CONTEXT_INFO | TRACE_ITER_SLEEP_TIME | - TRACE_ITER_GRAPH_TIME; + TRACE_ITER_GRAPH_TIME | TRACE_ITER_RECORD_CMD; static int trace_stop_count; static DEFINE_SPINLOCK(tracing_start_lock); @@ -428,6 +424,7 @@ static const char *trace_options[] = { "latency-format", "sleep-time", "graph-time", + "record-cmd", NULL }; @@ -659,6 +656,10 @@ update_max_tr(struct trace_array *tr, struct task_struct *tsk, int cpu) return; WARN_ON_ONCE(!irqs_disabled()); + if (!current_trace->use_max_tr) { + WARN_ON_ONCE(1); + return; + } arch_spin_lock(&ftrace_max_lock); tr->buffer = max_tr.buffer; @@ -685,6 +686,11 @@ update_max_tr_single(struct trace_array *tr, struct task_struct *tsk, int cpu) return; WARN_ON_ONCE(!irqs_disabled()); + if (!current_trace->use_max_tr) { + WARN_ON_ONCE(1); + return; + } + arch_spin_lock(&ftrace_max_lock); ftrace_disable_cpu(); @@ -729,18 +735,11 @@ __acquires(kernel_lock) return -1; } - if (strlen(type->name) > MAX_TRACER_SIZE) { + 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 - * need to disable and enable preemption for successful tests. - * So we drop the BKL here and grab it after the tests again. - */ - unlock_kernel(); mutex_lock(&trace_types_lock); tracing_selftest_running = true; @@ -822,7 +821,6 @@ __acquires(kernel_lock) #endif out_unlock: - lock_kernel(); return ret; } @@ -1285,6 +1283,8 @@ void trace_dump_stack(void) __ftrace_trace_stack(global_trace.buffer, flags, 3, preempt_count()); } +static DEFINE_PER_CPU(int, user_stack_count); + void ftrace_trace_userstack(struct ring_buffer *buffer, unsigned long flags, int pc) { @@ -1303,10 +1303,20 @@ ftrace_trace_userstack(struct ring_buffer *buffer, unsigned long flags, int pc) if (unlikely(in_nmi())) return; + /* + * prevent recursion, since the user stack tracing may + * trigger other kernel events. + */ + preempt_disable(); + if (__this_cpu_read(user_stack_count)) + goto out; + + __this_cpu_inc(user_stack_count); + event = trace_buffer_lock_reserve(buffer, TRACE_USER_STACK, sizeof(*entry), flags, pc); if (!event) - return; + goto out_drop_count; entry = ring_buffer_event_data(event); entry->tgid = current->tgid; @@ -1320,6 +1330,11 @@ ftrace_trace_userstack(struct ring_buffer *buffer, unsigned long flags, int pc) save_stack_trace_user(&trace); if (!filter_check_discard(call, entry, buffer, event)) ring_buffer_unlock_commit(buffer, event); + + out_drop_count: + __this_cpu_dec(user_stack_count); + out: + preempt_enable(); } #ifdef UNUSED @@ -1331,61 +1346,6 @@ static void __trace_userstack(struct trace_array *tr, unsigned long flags) #endif /* CONFIG_STACKTRACE */ -static void -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 *tr = __tr; - struct ring_buffer *buffer = tr->buffer; - struct special_entry *entry; - - event = trace_buffer_lock_reserve(buffer, TRACE_SPECIAL, - sizeof(*entry), 0, pc); - if (!event) - return; - entry = ring_buffer_event_data(event); - entry->arg1 = arg1; - entry->arg2 = arg2; - entry->arg3 = arg3; - - 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, arg1, arg2, arg3, preempt_count()); -} - -void -ftrace_special(unsigned long arg1, unsigned long arg2, unsigned long arg3) -{ - struct trace_array *tr = &global_trace; - struct trace_array_cpu *data; - unsigned long flags; - int cpu; - int pc; - - if (tracing_disabled) - return; - - pc = preempt_count(); - local_irq_save(flags); - cpu = raw_smp_processor_id(); - data = tr->data[cpu]; - - if (likely(atomic_inc_return(&data->disabled) == 1)) - ftrace_trace_special(tr, arg1, arg2, arg3, pc); - - atomic_dec(&data->disabled); - local_irq_restore(flags); -} - /** * trace_vbprintk - write binary msg to tracing buffer * @@ -1404,7 +1364,6 @@ int trace_vbprintk(unsigned long ip, const char *fmt, va_list args) struct bprint_entry *entry; unsigned long flags; int disable; - int resched; int cpu, len = 0, size, pc; if (unlikely(tracing_selftest_running || tracing_disabled)) @@ -1414,7 +1373,7 @@ int trace_vbprintk(unsigned long ip, const char *fmt, va_list args) pause_graph_tracing(); pc = preempt_count(); - resched = ftrace_preempt_disable(); + preempt_disable_notrace(); cpu = raw_smp_processor_id(); data = tr->data[cpu]; @@ -1452,7 +1411,7 @@ out_unlock: out: atomic_dec_return(&data->disabled); - ftrace_preempt_enable(resched); + preempt_enable_notrace(); unpause_graph_tracing(); return len; @@ -1539,11 +1498,6 @@ int trace_vprintk(unsigned long ip, const char *fmt, va_list args) } EXPORT_SYMBOL_GPL(trace_vprintk); -enum trace_file_type { - TRACE_FILE_LAT_FMT = 1, - TRACE_FILE_ANNOTATE = 2, -}; - static void trace_iterator_increment(struct trace_iterator *iter) { /* Don't allow ftrace to trace into the ring buffers */ @@ -1641,7 +1595,7 @@ struct trace_entry *trace_find_next_entry(struct trace_iterator *iter, } /* Find the next real entry, and increment the iterator to the next entry */ -static void *find_next_entry_inc(struct trace_iterator *iter) +void *trace_find_next_entry_inc(struct trace_iterator *iter) { iter->ent = __find_next_entry(iter, &iter->cpu, &iter->lost_events, &iter->ts); @@ -1676,19 +1630,19 @@ static void *s_next(struct seq_file *m, void *v, loff_t *pos) return NULL; if (iter->idx < 0) - ent = find_next_entry_inc(iter); + ent = trace_find_next_entry_inc(iter); else ent = iter; while (ent && iter->idx < i) - ent = find_next_entry_inc(iter); + ent = trace_find_next_entry_inc(iter); iter->pos = *pos; return ent; } -static void tracing_iter_reset(struct trace_iterator *iter, int cpu) +void tracing_iter_reset(struct trace_iterator *iter, int cpu) { struct trace_array *tr = iter->tr; struct ring_buffer_event *event; @@ -1936,7 +1890,7 @@ static enum print_line_t print_trace_fmt(struct trace_iterator *iter) } if (event) - return event->trace(iter, sym_flags); + return event->funcs->trace(iter, sym_flags, event); if (!trace_seq_printf(s, "Unknown type %d\n", entry->type)) goto partial; @@ -1962,7 +1916,7 @@ static enum print_line_t print_raw_fmt(struct trace_iterator *iter) event = ftrace_find_event(entry->type); if (event) - return event->raw(iter, 0); + return event->funcs->raw(iter, 0, event); if (!trace_seq_printf(s, "%d ?\n", entry->type)) goto partial; @@ -1989,7 +1943,7 @@ static enum print_line_t print_hex_fmt(struct trace_iterator *iter) event = ftrace_find_event(entry->type); if (event) { - enum print_line_t ret = event->hex(iter, 0); + enum print_line_t ret = event->funcs->hex(iter, 0, event); if (ret != TRACE_TYPE_HANDLED) return ret; } @@ -2014,7 +1968,8 @@ static enum print_line_t print_bin_fmt(struct trace_iterator *iter) } event = ftrace_find_event(entry->type); - return event ? event->binary(iter, 0) : TRACE_TYPE_HANDLED; + return event ? event->funcs->binary(iter, 0, event) : + TRACE_TYPE_HANDLED; } int trace_empty(struct trace_iterator *iter) @@ -2048,7 +2003,7 @@ int trace_empty(struct trace_iterator *iter) } /* Called with trace_event_read_lock() held. */ -static enum print_line_t print_trace_line(struct trace_iterator *iter) +enum print_line_t print_trace_line(struct trace_iterator *iter) { enum print_line_t ret; @@ -2257,7 +2212,7 @@ int tracing_open_generic(struct inode *inode, struct file *filp) static int tracing_release(struct inode *inode, struct file *file) { - struct seq_file *m = (struct seq_file *)file->private_data; + struct seq_file *m = file->private_data; struct trace_iterator *iter; int cpu; @@ -2381,11 +2336,19 @@ tracing_write_stub(struct file *filp, const char __user *ubuf, return count; } +static loff_t tracing_seek(struct file *file, loff_t offset, int origin) +{ + if (file->f_mode & FMODE_READ) + return seq_lseek(file, offset, origin); + else + return 0; +} + static const struct file_operations tracing_fops = { .open = tracing_open, .read = seq_read, .write = tracing_write_stub, - .llseek = seq_lseek, + .llseek = tracing_seek, .release = tracing_release, }; @@ -2393,6 +2356,7 @@ static const struct file_operations show_traces_fops = { .open = show_traces_open, .read = seq_read, .release = seq_release, + .llseek = seq_lseek, }; /* @@ -2486,6 +2450,7 @@ static const struct file_operations tracing_cpumask_fops = { .open = tracing_open_generic, .read = tracing_cpumask_read, .write = tracing_cpumask_write, + .llseek = generic_file_llseek, }; static int tracing_trace_options_show(struct seq_file *m, void *v) @@ -2561,6 +2526,9 @@ static void set_tracer_flags(unsigned int mask, int enabled) trace_flags |= mask; else trace_flags &= ~mask; + + if (mask == TRACE_ITER_RECORD_CMD) + trace_event_enable_cmd_record(enabled); } static ssize_t @@ -2652,6 +2620,7 @@ tracing_readme_read(struct file *filp, char __user *ubuf, static const struct file_operations tracing_readme_fops = { .open = tracing_open_generic, .read = tracing_readme_read, + .llseek = generic_file_llseek, }; static ssize_t @@ -2702,6 +2671,7 @@ tracing_saved_cmdlines_read(struct file *file, char __user *ubuf, static const struct file_operations tracing_saved_cmdlines_fops = { .open = tracing_open_generic, .read = tracing_saved_cmdlines_read, + .llseek = generic_file_llseek, }; static ssize_t @@ -2797,6 +2767,9 @@ static int tracing_resize_ring_buffer(unsigned long size) if (ret < 0) return ret; + if (!current_trace->use_max_tr) + goto out; + ret = ring_buffer_resize(max_tr.buffer, size); if (ret < 0) { int r; @@ -2824,11 +2797,14 @@ static int tracing_resize_ring_buffer(unsigned long size) return ret; } + max_tr.entries = size; + out: global_trace.entries = size; return ret; } + /** * tracing_update_buffers - used by tracing facility to expand ring buffers * @@ -2889,12 +2865,26 @@ static int tracing_set_tracer(const char *buf) trace_branch_disable(); if (current_trace && current_trace->reset) current_trace->reset(tr); - + if (current_trace && current_trace->use_max_tr) { + /* + * We don't free the ring buffer. instead, resize it because + * The max_tr ring buffer has some state (e.g. ring->clock) and + * we want preserve it. + */ + ring_buffer_resize(max_tr.buffer, 1); + max_tr.entries = 1; + } destroy_trace_option_files(topts); current_trace = t; topts = create_trace_option_files(current_trace); + if (current_trace->use_max_tr) { + ret = ring_buffer_resize(max_tr.buffer, global_trace.entries); + if (ret < 0) + goto out; + max_tr.entries = global_trace.entries; + } if (t->init) { ret = tracer_init(t, tr); @@ -3031,6 +3021,7 @@ static int tracing_open_pipe(struct inode *inode, struct file *filp) if (iter->trace->pipe_open) iter->trace->pipe_open(iter); + nonseekable_open(inode, filp); out: mutex_unlock(&trace_types_lock); return ret; @@ -3210,7 +3201,7 @@ waitagain: trace_event_read_lock(); trace_access_lock(iter->cpu_file); - while (find_next_entry_inc(iter) != NULL) { + while (trace_find_next_entry_inc(iter) != NULL) { enum print_line_t ret; int len = iter->seq.len; @@ -3293,7 +3284,7 @@ tracing_fill_pipe_page(size_t rem, struct trace_iterator *iter) if (ret != TRACE_TYPE_NO_CONSUME) trace_consume(iter); rem -= count; - if (!find_next_entry_inc(iter)) { + if (!trace_find_next_entry_inc(iter)) { rem = 0; iter->ent = NULL; break; @@ -3349,7 +3340,7 @@ static ssize_t tracing_splice_read_pipe(struct file *filp, if (ret <= 0) goto out_err; - if (!iter->ent && !find_next_entry_inc(iter)) { + if (!iter->ent && !trace_find_next_entry_inc(iter)) { ret = -EFAULT; goto out_err; } @@ -3476,7 +3467,6 @@ tracing_entries_write(struct file *filp, const char __user *ubuf, } tracing_start(); - max_tr.entries = global_trace.entries; mutex_unlock(&trace_types_lock); return cnt; @@ -3497,6 +3487,7 @@ tracing_mark_write(struct file *filp, const char __user *ubuf, size_t cnt, loff_t *fpos) { char *buf; + size_t written; if (tracing_disabled) return -EINVAL; @@ -3518,11 +3509,15 @@ tracing_mark_write(struct file *filp, const char __user *ubuf, } else buf[cnt] = '\0'; - cnt = mark_printk("%s", buf); + written = mark_printk("%s", buf); kfree(buf); - *fpos += cnt; + *fpos += written; - return cnt; + /* don't tell userspace we wrote more - it might confuse them */ + if (written > cnt) + written = cnt; + + return written; } static int tracing_clock_show(struct seq_file *m, void *v) @@ -3589,18 +3584,21 @@ static const struct file_operations tracing_max_lat_fops = { .open = tracing_open_generic, .read = tracing_max_lat_read, .write = tracing_max_lat_write, + .llseek = generic_file_llseek, }; static const struct file_operations tracing_ctrl_fops = { .open = tracing_open_generic, .read = tracing_ctrl_read, .write = tracing_ctrl_write, + .llseek = generic_file_llseek, }; static const struct file_operations set_tracer_fops = { .open = tracing_open_generic, .read = tracing_set_trace_read, .write = tracing_set_trace_write, + .llseek = generic_file_llseek, }; static const struct file_operations tracing_pipe_fops = { @@ -3609,17 +3607,20 @@ static const struct file_operations tracing_pipe_fops = { .read = tracing_read_pipe, .splice_read = tracing_splice_read_pipe, .release = tracing_release_pipe, + .llseek = no_llseek, }; static const struct file_operations tracing_entries_fops = { .open = tracing_open_generic, .read = tracing_entries_read, .write = tracing_entries_write, + .llseek = generic_file_llseek, }; static const struct file_operations tracing_mark_fops = { .open = tracing_open_generic, .write = tracing_mark_write, + .llseek = generic_file_llseek, }; static const struct file_operations trace_clock_fops = { @@ -3665,7 +3666,6 @@ 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; @@ -3692,11 +3692,6 @@ tracing_buffers_read(struct file *filp, char __user *ubuf, 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) @@ -3931,6 +3926,7 @@ tracing_stats_read(struct file *filp, char __user *ubuf, static const struct file_operations tracing_stats_fops = { .open = tracing_open_generic, .read = tracing_stats_read, + .llseek = generic_file_llseek, }; #ifdef CONFIG_DYNAMIC_FTRACE @@ -3967,6 +3963,7 @@ tracing_read_dyn_info(struct file *filp, char __user *ubuf, static const struct file_operations tracing_dyn_info_fops = { .open = tracing_open_generic, .read = tracing_read_dyn_info, + .llseek = generic_file_llseek, }; #endif @@ -4023,13 +4020,9 @@ 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]; + char cpu_dir[30]; /* 30 characters should be more than enough */ - if (cpu > 999 || cpu < 0) - return; - - sprintf(cpu_dir, "cpu%ld", cpu); + snprintf(cpu_dir, 30, "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); @@ -4120,6 +4113,7 @@ static const struct file_operations trace_options_fops = { .open = tracing_open_generic, .read = trace_options_read, .write = trace_options_write, + .llseek = generic_file_llseek, }; static ssize_t @@ -4171,6 +4165,7 @@ static const struct file_operations trace_options_core_fops = { .open = tracing_open_generic, .read = trace_options_core_read, .write = trace_options_core_write, + .llseek = generic_file_llseek, }; struct dentry *trace_create_file(const char *name, @@ -4360,9 +4355,6 @@ static __init int tracer_init_debugfs(void) 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(); @@ -4419,7 +4411,7 @@ static struct notifier_block trace_die_notifier = { */ #define KERN_TRACE KERN_EMERG -static void +void trace_printk_seq(struct trace_seq *s) { /* Probably should print a warning here. */ @@ -4434,6 +4426,13 @@ trace_printk_seq(struct trace_seq *s) trace_seq_init(s); } +void trace_init_global_iter(struct trace_iterator *iter) +{ + iter->tr = &global_trace; + iter->trace = current_trace; + iter->cpu_file = TRACE_PIPE_ALL_CPU; +} + static void __ftrace_dump(bool disable_tracing, enum ftrace_dump_mode oops_dump_mode) { @@ -4459,8 +4458,10 @@ __ftrace_dump(bool disable_tracing, enum ftrace_dump_mode oops_dump_mode) if (disable_tracing) ftrace_kill(); + trace_init_global_iter(&iter); + for_each_tracing_cpu(cpu) { - atomic_inc(&global_trace.data[cpu]->disabled); + atomic_inc(&iter.tr->data[cpu]->disabled); } old_userobj = trace_flags & TRACE_ITER_SYM_USEROBJ; @@ -4509,7 +4510,7 @@ __ftrace_dump(bool disable_tracing, enum ftrace_dump_mode oops_dump_mode) iter.iter_flags |= TRACE_FILE_LAT_FMT; iter.pos = -1; - if (find_next_entry_inc(&iter) != NULL) { + if (trace_find_next_entry_inc(&iter) != NULL) { int ret; ret = print_trace_line(&iter); @@ -4531,7 +4532,7 @@ __ftrace_dump(bool disable_tracing, enum ftrace_dump_mode oops_dump_mode) trace_flags |= old_userobj; for_each_tracing_cpu(cpu) { - atomic_dec(&global_trace.data[cpu]->disabled); + atomic_dec(&iter.tr->data[cpu]->disabled); } tracing_on(); } @@ -4580,16 +4581,14 @@ __init static int tracer_alloc_buffers(void) #ifdef CONFIG_TRACER_MAX_TRACE - max_tr.buffer = ring_buffer_alloc(ring_buf_size, - TRACE_BUFFER_FLAGS); + max_tr.buffer = ring_buffer_alloc(1, TRACE_BUFFER_FLAGS); if (!max_tr.buffer) { printk(KERN_ERR "tracer: failed to allocate max ring buffer!\n"); WARN_ON(1); ring_buffer_free(global_trace.buffer); goto out_free_cpumask; } - max_tr.entries = ring_buffer_size(max_tr.buffer); - WARN_ON(max_tr.entries != global_trace.entries); + max_tr.entries = 1; #endif /* Allocate the first page for all buffers */ @@ -4602,9 +4601,6 @@ __init static int tracer_alloc_buffers(void) register_tracer(&nop_trace); current_trace = &nop_trace; -#ifdef CONFIG_BOOT_TRACER - register_tracer(&boot_tracer); -#endif /* All seems OK, enable tracing */ tracing_disabled = 0; diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h index d1ce0bec1b3..9021f8c0c0c 100644 --- a/kernel/trace/trace.h +++ b/kernel/trace/trace.h @@ -9,10 +9,7 @@ #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> @@ -25,30 +22,17 @@ enum trace_type { TRACE_STACK, TRACE_PRINT, TRACE_BPRINT, - TRACE_SPECIAL, TRACE_MMIO_RW, TRACE_MMIO_MAP, TRACE_BRANCH, - TRACE_BOOT_CALL, - TRACE_BOOT_RET, TRACE_GRAPH_RET, TRACE_GRAPH_ENT, TRACE_USER_STACK, - TRACE_KMEM_ALLOC, - TRACE_KMEM_FREE, TRACE_BLK, - TRACE_KSYM, __TRACE_LAST_TYPE, }; -enum kmemtrace_type_id { - KMEMTRACE_TYPE_KMALLOC = 0, /* kmalloc() or kfree(). */ - KMEMTRACE_TYPE_CACHE, /* kmem_cache_*(). */ - KMEMTRACE_TYPE_PAGES, /* __get_free_pages() and friends. */ -}; - -extern struct tracer boot_tracer; #undef __field #define __field(type, item) type item; @@ -204,23 +188,15 @@ extern void __ftrace_bad_type(void); 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); \ IF_ASSIGN(var, ent, struct trace_mmiotrace_map, \ TRACE_MMIO_MAP); \ - IF_ASSIGN(var, ent, struct trace_boot_call, TRACE_BOOT_CALL);\ - IF_ASSIGN(var, ent, struct trace_boot_ret, TRACE_BOOT_RET);\ IF_ASSIGN(var, ent, struct trace_branch, TRACE_BRANCH); \ IF_ASSIGN(var, ent, struct ftrace_graph_ent_entry, \ TRACE_GRAPH_ENT); \ IF_ASSIGN(var, ent, struct ftrace_graph_ret_entry, \ TRACE_GRAPH_RET); \ - 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) @@ -298,6 +274,7 @@ struct tracer { struct tracer *next; int print_max; struct tracer_flags *flags; + int use_max_tr; }; @@ -318,7 +295,6 @@ struct dentry *trace_create_file(const char *name, const struct file_operations *fops); struct dentry *tracing_init_dentry(void); -void init_tracer_sysprof_debugfs(struct dentry *d_tracer); struct ring_buffer_event; @@ -338,6 +314,14 @@ struct trace_entry *tracing_get_trace_entry(struct trace_array *tr, struct trace_entry *trace_find_next_entry(struct trace_iterator *iter, int *ent_cpu, u64 *ent_ts); +int trace_empty(struct trace_iterator *iter); + +void *trace_find_next_entry_inc(struct trace_iterator *iter); + +void trace_init_global_iter(struct trace_iterator *iter); + +void tracing_iter_reset(struct trace_iterator *iter, int cpu); + void default_wait_pipe(struct trace_iterator *iter); void poll_wait_pipe(struct trace_iterator *iter); @@ -355,15 +339,14 @@ void tracing_sched_wakeup_trace(struct trace_array *tr, struct task_struct *wakee, struct task_struct *cur, unsigned long flags, int pc); -void trace_special(struct trace_array *tr, - struct trace_array_cpu *data, - unsigned long arg1, - unsigned long arg2, - unsigned long arg3, int pc); void trace_function(struct trace_array *tr, unsigned long ip, unsigned long parent_ip, unsigned long flags, int pc); +void trace_graph_function(struct trace_array *tr, + unsigned long ip, + unsigned long parent_ip, + unsigned long flags, int pc); void trace_default_header(struct seq_file *m); void print_trace_header(struct seq_file *m, struct trace_iterator *iter); int trace_empty(struct trace_iterator *iter); @@ -380,8 +363,15 @@ void tracing_start_sched_switch_record(void); int register_tracer(struct tracer *type); void unregister_tracer(struct tracer *type); int is_tracing_stopped(void); +enum trace_file_type { + TRACE_FILE_LAT_FMT = 1, + TRACE_FILE_ANNOTATE = 2, +}; + +extern cpumask_var_t __read_mostly tracing_buffer_mask; -extern int process_new_ksym_entry(char *ksymname, int op, unsigned long addr); +#define for_each_tracing_cpu(cpu) \ + for_each_cpu(cpu, tracing_buffer_mask) extern unsigned long nsecs_to_usecs(unsigned long nsecs); @@ -405,12 +395,12 @@ void ftrace_trace_userstack(struct ring_buffer *buffer, unsigned long flags, void __trace_stack(struct trace_array *tr, unsigned long flags, int skip, int pc); #else -static inline void ftrace_trace_stack(struct trace_array *tr, +static inline void ftrace_trace_stack(struct ring_buffer *buffer, unsigned long flags, int skip, int pc) { } -static inline void ftrace_trace_userstack(struct trace_array *tr, +static inline void ftrace_trace_userstack(struct ring_buffer *buffer, unsigned long flags, int pc) { } @@ -452,12 +442,8 @@ extern int trace_selftest_startup_nop(struct tracer *trace, struct trace_array *tr); extern int trace_selftest_startup_sched_switch(struct tracer *trace, struct trace_array *tr); -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_ksym(struct tracer *trace, - struct trace_array *tr); #endif /* CONFIG_FTRACE_STARTUP_TEST */ extern void *head_page(struct trace_array_cpu *data); @@ -471,6 +457,8 @@ 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, ...); +void trace_printk_seq(struct trace_seq *s); +enum print_line_t print_trace_line(struct trace_iterator *iter); extern unsigned long trace_flags; @@ -617,6 +605,7 @@ enum trace_iterator_flags { TRACE_ITER_LATENCY_FMT = 0x20000, TRACE_ITER_SLEEP_TIME = 0x40000, TRACE_ITER_GRAPH_TIME = 0x80000, + TRACE_ITER_RECORD_CMD = 0x100000, }; /* @@ -628,54 +617,6 @@ enum trace_iterator_flags { extern struct tracer nop_trace; -/** - * ftrace_preempt_disable - disable preemption scheduler safe - * - * When tracing can happen inside the scheduler, there exists - * cases that the tracing might happen before the need_resched - * flag is checked. If this happens and the tracer calls - * preempt_enable (after a disable), a schedule might take place - * causing an infinite recursion. - * - * 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 - * 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. - */ -static inline int ftrace_preempt_disable(void) -{ - int resched; - - resched = need_resched(); - preempt_disable_notrace(); - - return resched; -} - -/** - * ftrace_preempt_enable - enable preemption scheduler safe - * @resched: the return value from ftrace_preempt_disable - * - * 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 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. - */ -static inline void ftrace_preempt_enable(int resched) -{ - if (resched) - preempt_enable_no_resched_notrace(); - else - preempt_enable_notrace(); -} - #ifdef CONFIG_BRANCH_TRACER extern int enable_branch_tracing(struct trace_array *tr); extern void disable_branch_tracing(void); @@ -766,6 +707,8 @@ struct filter_pred { int pop_n; }; +extern struct list_head ftrace_common_fields; + 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, @@ -778,12 +721,15 @@ extern void print_subsystem_event_filter(struct event_subsystem *system, struct trace_seq *s); extern int filter_assign_type(const char *type); +struct list_head * +trace_get_fields(struct ftrace_event_call *event_call); + 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) && + if (unlikely(call->flags & TRACE_EVENT_FL_FILTERED) && !filter_match_preds(call->filter, rec)) { ring_buffer_discard_commit(buffer, event); return 1; @@ -792,6 +738,8 @@ filter_check_discard(struct ftrace_event_call *call, void *rec, return 0; } +extern void trace_event_enable_cmd_record(bool enable); + extern struct mutex event_mutex; extern struct list_head ftrace_events; diff --git a/kernel/trace/trace_boot.c b/kernel/trace/trace_boot.c deleted file mode 100644 index c21d5f3956a..00000000000 --- a/kernel/trace/trace_boot.c +++ /dev/null @@ -1,185 +0,0 @@ -/* - * ring buffer based initcalls tracer - * - * 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/time.h> - -#include "trace.h" -#include "trace_output.h" - -static struct trace_array *boot_trace; -static bool pre_initcalls_finished; - -/* Tells the boot tracer that the pre_smp_initcalls are finished. - * So we are ready . - * It doesn't enable sched events tracing however. - * You have to call enable_boot_trace to do so. - */ -void start_boot_trace(void) -{ - pre_initcalls_finished = true; -} - -void enable_boot_trace(void) -{ - if (boot_trace && pre_initcalls_finished) - tracing_start_sched_switch_record(); -} - -void disable_boot_trace(void) -{ - if (boot_trace && pre_initcalls_finished) - tracing_stop_sched_switch_record(); -} - -static int boot_trace_init(struct trace_array *tr) -{ - boot_trace = tr; - - if (!tr) - return 0; - - tracing_reset_online_cpus(tr); - - tracing_sched_switch_assign_trace(tr); - return 0; -} - -static enum print_line_t -initcall_call_print_line(struct trace_iterator *iter) -{ - struct trace_entry *entry = iter->ent; - struct trace_seq *s = &iter->seq; - struct trace_boot_call *field; - struct boot_trace_call *call; - u64 ts; - unsigned long nsec_rem; - int ret; - - trace_assign_type(field, entry); - call = &field->boot_call; - ts = iter->ts; - 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); - - if (!ret) - return TRACE_TYPE_PARTIAL_LINE; - else - return TRACE_TYPE_HANDLED; -} - -static enum print_line_t -initcall_ret_print_line(struct trace_iterator *iter) -{ - struct trace_entry *entry = iter->ent; - struct trace_seq *s = &iter->seq; - struct trace_boot_ret *field; - struct boot_trace_ret *init_ret; - u64 ts; - unsigned long nsec_rem; - int ret; - - trace_assign_type(field, entry); - init_ret = &field->boot_ret; - ts = iter->ts; - nsec_rem = do_div(ts, NSEC_PER_SEC); - - ret = trace_seq_printf(s, "[%5ld.%09ld] initcall %s " - "returned %d after %llu msecs\n", - (unsigned long) ts, - nsec_rem, - init_ret->func, init_ret->result, init_ret->duration); - - if (!ret) - return TRACE_TYPE_PARTIAL_LINE; - else - return TRACE_TYPE_HANDLED; -} - -static enum print_line_t initcall_print_line(struct trace_iterator *iter) -{ - struct trace_entry *entry = iter->ent; - - switch (entry->type) { - case TRACE_BOOT_CALL: - return initcall_call_print_line(iter); - case TRACE_BOOT_RET: - return initcall_ret_print_line(iter); - default: - return TRACE_TYPE_UNHANDLED; - } -} - -struct tracer boot_tracer __read_mostly = -{ - .name = "initcall", - .init = boot_trace_init, - .reset = tracing_reset_online_cpus, - .print_line = initcall_print_line, -}; - -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; - struct trace_array *tr = boot_trace; - - if (!tr || !pre_initcalls_finished) - return; - - /* Get its name now since this function could - * disappear because it is in the .init section. - */ - sprint_symbol(bt->func, (unsigned long)fn); - preempt_disable(); - - 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); - entry->boot_call = *bt; - 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; - struct trace_array *tr = boot_trace; - - if (!tr || !pre_initcalls_finished) - return; - - sprint_symbol(bt->func, (unsigned long)fn); - preempt_disable(); - - 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); - entry->boot_ret = *bt; - 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 b9bc4d47017..8d3538b4ea5 100644 --- a/kernel/trace/trace_branch.c +++ b/kernel/trace/trace_branch.c @@ -143,7 +143,7 @@ static void branch_trace_reset(struct trace_array *tr) } static enum print_line_t trace_branch_print(struct trace_iterator *iter, - int flags) + int flags, struct trace_event *event) { struct trace_branch *field; @@ -167,9 +167,13 @@ static void branch_print_header(struct seq_file *s) " |\n"); } +static struct trace_event_functions trace_branch_funcs = { + .trace = trace_branch_print, +}; + static struct trace_event trace_branch_event = { .type = TRACE_BRANCH, - .trace = trace_branch_print, + .funcs = &trace_branch_funcs, }; static struct tracer branch_trace __read_mostly = diff --git a/kernel/trace/trace_clock.c b/kernel/trace/trace_clock.c index 9d589d8dcd1..685a67d55db 100644 --- a/kernel/trace/trace_clock.c +++ b/kernel/trace/trace_clock.c @@ -32,16 +32,15 @@ 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(); + preempt_disable_notrace(); clock = sched_clock(); - ftrace_preempt_enable(resched); + preempt_enable_notrace(); return clock; } @@ -56,7 +55,7 @@ u64 notrace trace_clock_local(void) */ u64 notrace trace_clock(void) { - return cpu_clock(raw_smp_processor_id()); + return local_clock(); } diff --git a/kernel/trace/trace_entries.h b/kernel/trace/trace_entries.h index dc008c1240d..6cf223764be 100644 --- a/kernel/trace/trace_entries.h +++ b/kernel/trace/trace_entries.h @@ -53,7 +53,7 @@ */ /* - * Function trace entry - function address and parent function addres: + * Function trace entry - function address and parent function address: */ FTRACE_ENTRY(function, ftrace_entry, @@ -151,23 +151,6 @@ FTRACE_ENTRY_DUP(wakeup, ctx_switch_entry, ); /* - * 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: */ @@ -271,33 +254,6 @@ FTRACE_ENTRY(mmiotrace_map, trace_mmiotrace_map, __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 @@ -318,53 +274,3 @@ FTRACE_ENTRY(branch, trace_branch, __entry->func, __entry->file, __entry->correct) ); -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_perf.c b/kernel/trace/trace_event_perf.c index 0565bb42566..19a359d5e6d 100644 --- a/kernel/trace/trace_event_perf.c +++ b/kernel/trace/trace_event_perf.c @@ -9,13 +9,7 @@ #include <linux/kprobes.h> #include "trace.h" -DEFINE_PER_CPU(struct pt_regs, perf_trace_regs); -EXPORT_PER_CPU_SYMBOL_GPL(perf_trace_regs); - -EXPORT_SYMBOL_GPL(perf_arch_fetch_caller_regs); - -static char *perf_trace_buf; -static char *perf_trace_buf_nmi; +static char __percpu *perf_trace_buf[PERF_NR_CONTEXTS]; /* * Force it to be aligned to unsigned long to avoid misaligned accesses @@ -27,57 +21,107 @@ typedef typeof(unsigned long [PERF_MAX_TRACE_SIZE / sizeof(unsigned long)]) /* Count the events in use (per event id, not per instance) */ static int total_ref_count; -static int perf_trace_event_enable(struct ftrace_event_call *event) +static int perf_trace_event_perm(struct ftrace_event_call *tp_event, + struct perf_event *p_event) { - char *buf; - int ret = -ENOMEM; - - if (event->perf_refcount++ > 0) + /* No tracing, just counting, so no obvious leak */ + if (!(p_event->attr.sample_type & PERF_SAMPLE_RAW)) return 0; - if (!total_ref_count) { - buf = (char *)alloc_percpu(perf_trace_t); - if (!buf) - goto fail_buf; + /* Some events are ok to be traced by non-root users... */ + if (p_event->attach_state == PERF_ATTACH_TASK) { + if (tp_event->flags & TRACE_EVENT_FL_CAP_ANY) + return 0; + } - rcu_assign_pointer(perf_trace_buf, buf); + /* + * ...otherwise raw tracepoint data can be a severe data leak, + * only allow root to have these. + */ + if (perf_paranoid_tracepoint_raw() && !capable(CAP_SYS_ADMIN)) + return -EPERM; - buf = (char *)alloc_percpu(perf_trace_t); - if (!buf) - goto fail_buf_nmi; + return 0; +} - rcu_assign_pointer(perf_trace_buf_nmi, buf); - } +static int perf_trace_event_init(struct ftrace_event_call *tp_event, + struct perf_event *p_event) +{ + struct hlist_head __percpu *list; + int ret; + int cpu; - ret = event->perf_event_enable(event); - if (!ret) { - total_ref_count++; + ret = perf_trace_event_perm(tp_event, p_event); + if (ret) + return ret; + + p_event->tp_event = tp_event; + if (tp_event->perf_refcount++ > 0) return 0; + + ret = -ENOMEM; + + list = alloc_percpu(struct hlist_head); + if (!list) + goto fail; + + for_each_possible_cpu(cpu) + INIT_HLIST_HEAD(per_cpu_ptr(list, cpu)); + + tp_event->perf_events = list; + + if (!total_ref_count) { + char __percpu *buf; + int i; + + for (i = 0; i < PERF_NR_CONTEXTS; i++) { + buf = (char __percpu *)alloc_percpu(perf_trace_t); + if (!buf) + goto fail; + + perf_trace_buf[i] = buf; + } } -fail_buf_nmi: + ret = tp_event->class->reg(tp_event, TRACE_REG_PERF_REGISTER); + if (ret) + goto fail; + + total_ref_count++; + return 0; + +fail: if (!total_ref_count) { - free_percpu(perf_trace_buf_nmi); - free_percpu(perf_trace_buf); - perf_trace_buf_nmi = NULL; - perf_trace_buf = NULL; + int i; + + for (i = 0; i < PERF_NR_CONTEXTS; i++) { + free_percpu(perf_trace_buf[i]); + perf_trace_buf[i] = NULL; + } + } + + if (!--tp_event->perf_refcount) { + free_percpu(tp_event->perf_events); + tp_event->perf_events = NULL; } -fail_buf: - event->perf_refcount--; return ret; } -int perf_trace_enable(int event_id) +int perf_trace_init(struct perf_event *p_event) { - struct ftrace_event_call *event; + struct ftrace_event_call *tp_event; + int event_id = p_event->attr.config; int ret = -EINVAL; mutex_lock(&event_mutex); - list_for_each_entry(event, &ftrace_events, list) { - if (event->id == event_id && event->perf_event_enable && - try_module_get(event->mod)) { - ret = perf_trace_event_enable(event); + list_for_each_entry(tp_event, &ftrace_events, list) { + if (tp_event->event.type == event_id && + tp_event->class && tp_event->class->reg && + try_module_get(tp_event->mod)) { + ret = perf_trace_event_init(tp_event, p_event); + if (ret) + module_put(tp_event->mod); break; } } @@ -86,90 +130,87 @@ int perf_trace_enable(int event_id) return ret; } -static void perf_trace_event_disable(struct ftrace_event_call *event) +int perf_trace_add(struct perf_event *p_event, int flags) { - char *buf, *nmi_buf; - - if (--event->perf_refcount > 0) - return; + struct ftrace_event_call *tp_event = p_event->tp_event; + struct hlist_head __percpu *pcpu_list; + struct hlist_head *list; - event->perf_event_disable(event); + pcpu_list = tp_event->perf_events; + if (WARN_ON_ONCE(!pcpu_list)) + return -EINVAL; - if (!--total_ref_count) { - buf = perf_trace_buf; - rcu_assign_pointer(perf_trace_buf, NULL); + if (!(flags & PERF_EF_START)) + p_event->hw.state = PERF_HES_STOPPED; - nmi_buf = perf_trace_buf_nmi; - rcu_assign_pointer(perf_trace_buf_nmi, NULL); + list = this_cpu_ptr(pcpu_list); + hlist_add_head_rcu(&p_event->hlist_entry, list); - /* - * Ensure every events in profiling have finished before - * releasing the buffers - */ - synchronize_sched(); + return 0; +} - free_percpu(buf); - free_percpu(nmi_buf); - } +void perf_trace_del(struct perf_event *p_event, int flags) +{ + hlist_del_rcu(&p_event->hlist_entry); } -void perf_trace_disable(int event_id) +void perf_trace_destroy(struct perf_event *p_event) { - struct ftrace_event_call *event; + struct ftrace_event_call *tp_event = p_event->tp_event; + int i; mutex_lock(&event_mutex); - list_for_each_entry(event, &ftrace_events, list) { - if (event->id == event_id) { - perf_trace_event_disable(event); - module_put(event->mod); - break; + if (--tp_event->perf_refcount > 0) + goto out; + + tp_event->class->reg(tp_event, TRACE_REG_PERF_UNREGISTER); + + /* + * Ensure our callback won't be called anymore. The buffers + * will be freed after that. + */ + tracepoint_synchronize_unregister(); + + free_percpu(tp_event->perf_events); + tp_event->perf_events = NULL; + + if (!--total_ref_count) { + for (i = 0; i < PERF_NR_CONTEXTS; i++) { + free_percpu(perf_trace_buf[i]); + perf_trace_buf[i] = NULL; } } +out: + module_put(tp_event->mod); mutex_unlock(&event_mutex); } __kprobes void *perf_trace_buf_prepare(int size, unsigned short type, - int *rctxp, unsigned long *irq_flags) + struct pt_regs *regs, int *rctxp) { struct trace_entry *entry; - char *trace_buf, *raw_data; - int pc, cpu; + unsigned long flags; + char *raw_data; + int pc; BUILD_BUG_ON(PERF_MAX_TRACE_SIZE % sizeof(unsigned long)); pc = preempt_count(); - /* Protect the per cpu buffer, begin the rcu read side */ - local_irq_save(*irq_flags); - *rctxp = perf_swevent_get_recursion_context(); if (*rctxp < 0) - goto err_recursion; - - cpu = smp_processor_id(); - - if (in_nmi()) - trace_buf = rcu_dereference_sched(perf_trace_buf_nmi); - else - trace_buf = rcu_dereference_sched(perf_trace_buf); - - if (!trace_buf) - goto err; + return NULL; - raw_data = per_cpu_ptr(trace_buf, cpu); + raw_data = this_cpu_ptr(perf_trace_buf[*rctxp]); /* zero the dead bytes from align to not leak stack to user */ memset(&raw_data[size - sizeof(u64)], 0, sizeof(u64)); entry = (struct trace_entry *)raw_data; - tracing_generic_entry_update(entry, *irq_flags, pc); + local_save_flags(flags); + tracing_generic_entry_update(entry, flags, pc); entry->type = type; return raw_data; -err: - perf_swevent_put_recursion_context(*rctxp); -err_recursion: - local_irq_restore(*irq_flags); - return NULL; } EXPORT_SYMBOL_GPL(perf_trace_buf_prepare); diff --git a/kernel/trace/trace_events.c b/kernel/trace/trace_events.c index c697c704334..35fde09b81d 100644 --- a/kernel/trace/trace_events.c +++ b/kernel/trace/trace_events.c @@ -27,11 +27,26 @@ DEFINE_MUTEX(event_mutex); +DEFINE_MUTEX(event_storage_mutex); +EXPORT_SYMBOL_GPL(event_storage_mutex); + +char event_storage[EVENT_STORAGE_SIZE]; +EXPORT_SYMBOL_GPL(event_storage); + LIST_HEAD(ftrace_events); +LIST_HEAD(ftrace_common_fields); -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 list_head * +trace_get_fields(struct ftrace_event_call *event_call) +{ + if (!event_call->class->get_fields) + return &event_call->class->fields; + return event_call->class->get_fields(event_call); +} + +static int __trace_define_field(struct list_head *head, const char *type, + const char *name, int offset, int size, + int is_signed, int filter_type) { struct ftrace_event_field *field; @@ -56,7 +71,7 @@ int trace_define_field(struct ftrace_event_call *call, const char *type, field->size = size; field->is_signed = is_signed; - list_add(&field->link, &call->fields); + list_add(&field->link, head); return 0; @@ -67,17 +82,32 @@ err: return -ENOMEM; } + +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 list_head *head; + + if (WARN_ON(!call->class)) + return 0; + + head = trace_get_fields(call); + return __trace_define_field(head, type, name, offset, size, + is_signed, filter_type); +} 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); \ + ret = __trace_define_field(&ftrace_common_fields, #type, \ + "common_" #item, \ + offsetof(typeof(ent), item), \ + sizeof(ent.item), \ + is_signed_type(type), FILTER_OTHER); \ if (ret) \ return ret; -static int trace_define_common_fields(struct ftrace_event_call *call) +static int trace_define_common_fields(void) { int ret; struct trace_entry ent; @@ -94,8 +124,10 @@ static int trace_define_common_fields(struct ftrace_event_call *call) void trace_destroy_fields(struct ftrace_event_call *call) { struct ftrace_event_field *field, *next; + struct list_head *head; - list_for_each_entry_safe(field, next, &call->fields, link) { + head = trace_get_fields(call); + list_for_each_entry_safe(field, next, head, link) { list_del(&field->link); kfree(field->type); kfree(field->name); @@ -107,16 +139,63 @@ int trace_event_raw_init(struct ftrace_event_call *call) { int id; - id = register_ftrace_event(call->event); + id = register_ftrace_event(&call->event); if (!id) return -ENODEV; - call->id = id; - INIT_LIST_HEAD(&call->fields); return 0; } EXPORT_SYMBOL_GPL(trace_event_raw_init); +int ftrace_event_reg(struct ftrace_event_call *call, enum trace_reg type) +{ + switch (type) { + case TRACE_REG_REGISTER: + return tracepoint_probe_register(call->name, + call->class->probe, + call); + case TRACE_REG_UNREGISTER: + tracepoint_probe_unregister(call->name, + call->class->probe, + call); + return 0; + +#ifdef CONFIG_PERF_EVENTS + case TRACE_REG_PERF_REGISTER: + return tracepoint_probe_register(call->name, + call->class->perf_probe, + call); + case TRACE_REG_PERF_UNREGISTER: + tracepoint_probe_unregister(call->name, + call->class->perf_probe, + call); + return 0; +#endif + } + return 0; +} +EXPORT_SYMBOL_GPL(ftrace_event_reg); + +void trace_event_enable_cmd_record(bool enable) +{ + struct ftrace_event_call *call; + + mutex_lock(&event_mutex); + list_for_each_entry(call, &ftrace_events, list) { + if (!(call->flags & TRACE_EVENT_FL_ENABLED)) + continue; + + if (enable) { + tracing_start_cmdline_record(); + call->flags |= TRACE_EVENT_FL_RECORDED_CMD; + } else { + tracing_stop_cmdline_record(); + call->flags &= ~TRACE_EVENT_FL_RECORDED_CMD; + } + } + mutex_unlock(&event_mutex); +} + static int ftrace_event_enable_disable(struct ftrace_event_call *call, int enable) { @@ -124,23 +203,29 @@ static int ftrace_event_enable_disable(struct ftrace_event_call *call, switch (enable) { case 0: - if (call->enabled) { - call->enabled = 0; - tracing_stop_cmdline_record(); - call->unregfunc(call); + if (call->flags & TRACE_EVENT_FL_ENABLED) { + call->flags &= ~TRACE_EVENT_FL_ENABLED; + if (call->flags & TRACE_EVENT_FL_RECORDED_CMD) { + tracing_stop_cmdline_record(); + call->flags &= ~TRACE_EVENT_FL_RECORDED_CMD; + } + call->class->reg(call, TRACE_REG_UNREGISTER); } break; case 1: - if (!call->enabled) { - tracing_start_cmdline_record(); - ret = call->regfunc(call); + if (!(call->flags & TRACE_EVENT_FL_ENABLED)) { + if (trace_flags & TRACE_ITER_RECORD_CMD) { + tracing_start_cmdline_record(); + call->flags |= TRACE_EVENT_FL_RECORDED_CMD; + } + ret = call->class->reg(call, TRACE_REG_REGISTER); if (ret) { tracing_stop_cmdline_record(); pr_info("event trace: Could not enable event " "%s\n", call->name); break; } - call->enabled = 1; + call->flags |= TRACE_EVENT_FL_ENABLED; } break; } @@ -171,15 +256,15 @@ static int __ftrace_set_clr_event(const char *match, const char *sub, mutex_lock(&event_mutex); list_for_each_entry(call, &ftrace_events, list) { - if (!call->name || !call->regfunc) + if (!call->name || !call->class || !call->class->reg) continue; if (match && strcmp(match, call->name) != 0 && - strcmp(match, call->system) != 0) + strcmp(match, call->class->system) != 0) continue; - if (sub && strcmp(sub, call->system) != 0) + if (sub && strcmp(sub, call->class->system) != 0) continue; if (event && strcmp(event, call->name) != 0) @@ -297,7 +382,7 @@ t_next(struct seq_file *m, void *v, loff_t *pos) * The ftrace subsystem is for showing formats only. * They can not be enabled or disabled via the event files. */ - if (call->regfunc) + if (call->class && call->class->reg) return call; } @@ -328,7 +413,7 @@ s_next(struct seq_file *m, void *v, loff_t *pos) (*pos)++; list_for_each_entry_continue(call, &ftrace_events, list) { - if (call->enabled) + if (call->flags & TRACE_EVENT_FL_ENABLED) return call; } @@ -355,8 +440,8 @@ 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); + if (strcmp(call->class->system, TRACE_SYSTEM) != 0) + seq_printf(m, "%s:", call->class->system); seq_printf(m, "%s\n", call->name); return 0; @@ -387,7 +472,7 @@ event_enable_read(struct file *filp, char __user *ubuf, size_t cnt, struct ftrace_event_call *call = filp->private_data; char *buf; - if (call->enabled) + if (call->flags & TRACE_EVENT_FL_ENABLED) buf = "1\n"; else buf = "0\n"; @@ -450,10 +535,10 @@ system_enable_read(struct file *filp, char __user *ubuf, size_t cnt, mutex_lock(&event_mutex); list_for_each_entry(call, &ftrace_events, list) { - if (!call->name || !call->regfunc) + if (!call->name || !call->class || !call->class->reg) continue; - if (system && strcmp(call->system, system) != 0) + if (system && strcmp(call->class->system, system) != 0) continue; /* @@ -461,7 +546,7 @@ system_enable_read(struct file *filp, char __user *ubuf, size_t cnt, * or if all events or cleared, or if we have * a mixture. */ - set |= (1 << !!call->enabled); + set |= (1 << !!(call->flags & TRACE_EVENT_FL_ENABLED)); /* * If we have a mixture, no need to look further. @@ -519,83 +604,146 @@ out: return ret; } -static ssize_t -event_format_read(struct file *filp, char __user *ubuf, size_t cnt, - loff_t *ppos) +enum { + FORMAT_HEADER = 1, + FORMAT_FIELD_SEPERATOR = 2, + FORMAT_PRINTFMT = 3, +}; + +static void *f_next(struct seq_file *m, void *v, loff_t *pos) { - struct ftrace_event_call *call = filp->private_data; + struct ftrace_event_call *call = m->private; struct ftrace_event_field *field; - struct trace_seq *s; - int common_field_count = 5; - char *buf; - int r = 0; + struct list_head *common_head = &ftrace_common_fields; + struct list_head *head = trace_get_fields(call); - if (*ppos) - return 0; + (*pos)++; - s = kmalloc(sizeof(*s), GFP_KERNEL); - if (!s) - return -ENOMEM; + switch ((unsigned long)v) { + case FORMAT_HEADER: + if (unlikely(list_empty(common_head))) + return NULL; - trace_seq_init(s); + field = list_entry(common_head->prev, + struct ftrace_event_field, link); + return field; - trace_seq_printf(s, "name: %s\n", call->name); - trace_seq_printf(s, "ID: %d\n", call->id); - trace_seq_printf(s, "format:\n"); + case FORMAT_FIELD_SEPERATOR: + if (unlikely(list_empty(head))) + return NULL; - list_for_each_entry_reverse(field, &call->fields, link) { - /* - * Smartly shows the array type(except dynamic array). - * Normal: - * field:TYPE VAR - * If TYPE := TYPE[LEN], it is shown: - * field:TYPE VAR[LEN] - */ - const char *array_descriptor = strchr(field->type, '['); + field = list_entry(head->prev, struct ftrace_event_field, link); + return field; - if (!strncmp(field->type, "__data_loc", 10)) - array_descriptor = NULL; + case FORMAT_PRINTFMT: + /* all done */ + return NULL; + } - if (!array_descriptor) { - r = trace_seq_printf(s, "\tfield:%s %s;\toffset:%u;" - "\tsize:%u;\tsigned:%d;\n", - field->type, field->name, field->offset, - field->size, !!field->is_signed); - } else { - r = trace_seq_printf(s, "\tfield:%.*s %s%s;\toffset:%u;" - "\tsize:%u;\tsigned:%d;\n", - (int)(array_descriptor - field->type), - field->type, field->name, - array_descriptor, field->offset, - field->size, !!field->is_signed); - } + field = v; + if (field->link.prev == common_head) + return (void *)FORMAT_FIELD_SEPERATOR; + else if (field->link.prev == head) + return (void *)FORMAT_PRINTFMT; - if (--common_field_count == 0) - r = trace_seq_printf(s, "\n"); + field = list_entry(field->link.prev, struct ftrace_event_field, link); - if (!r) - break; - } + return field; +} - if (r) - r = trace_seq_printf(s, "\nprint fmt: %s\n", - call->print_fmt); +static void *f_start(struct seq_file *m, loff_t *pos) +{ + loff_t l = 0; + void *p; - 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; + /* Start by showing the header */ + if (!*pos) + return (void *)FORMAT_HEADER; + + p = (void *)FORMAT_HEADER; + do { + p = f_next(m, p, &l); + } while (p && l < *pos); + + return p; +} + +static int f_show(struct seq_file *m, void *v) +{ + struct ftrace_event_call *call = m->private; + struct ftrace_event_field *field; + const char *array_descriptor; + + switch ((unsigned long)v) { + case FORMAT_HEADER: + seq_printf(m, "name: %s\n", call->name); + seq_printf(m, "ID: %d\n", call->event.type); + seq_printf(m, "format:\n"); + return 0; + + case FORMAT_FIELD_SEPERATOR: + seq_putc(m, '\n'); + return 0; + + case FORMAT_PRINTFMT: + seq_printf(m, "\nprint fmt: %s\n", + call->print_fmt); + return 0; } - r = simple_read_from_buffer(ubuf, cnt, ppos, - s->buffer, s->len); - out: - kfree(s); - return r; + field = v; + + /* + * Smartly shows the array type(except dynamic array). + * Normal: + * field:TYPE VAR + * If TYPE := TYPE[LEN], it is shown: + * field:TYPE VAR[LEN] + */ + array_descriptor = strchr(field->type, '['); + + if (!strncmp(field->type, "__data_loc", 10)) + array_descriptor = NULL; + + if (!array_descriptor) + seq_printf(m, "\tfield:%s %s;\toffset:%u;\tsize:%u;\tsigned:%d;\n", + field->type, field->name, field->offset, + field->size, !!field->is_signed); + else + seq_printf(m, "\tfield:%.*s %s%s;\toffset:%u;\tsize:%u;\tsigned:%d;\n", + (int)(array_descriptor - field->type), + field->type, field->name, + array_descriptor, field->offset, + field->size, !!field->is_signed); + + return 0; +} + +static void f_stop(struct seq_file *m, void *p) +{ +} + +static const struct seq_operations trace_format_seq_ops = { + .start = f_start, + .next = f_next, + .stop = f_stop, + .show = f_show, +}; + +static int trace_format_open(struct inode *inode, struct file *file) +{ + struct ftrace_event_call *call = inode->i_private; + struct seq_file *m; + int ret; + + ret = seq_open(file, &trace_format_seq_ops); + if (ret < 0) + return ret; + + m = file->private_data; + m->private = call; + + return 0; } static ssize_t @@ -613,7 +761,7 @@ event_id_read(struct file *filp, char __user *ubuf, size_t cnt, loff_t *ppos) return -ENOMEM; trace_seq_init(s); - trace_seq_printf(s, "%d\n", call->id); + trace_seq_printf(s, "%d\n", call->event.type); r = simple_read_from_buffer(ubuf, cnt, ppos, s->buffer, s->len); @@ -790,39 +938,47 @@ static const struct file_operations ftrace_enable_fops = { .open = tracing_open_generic, .read = event_enable_read, .write = event_enable_write, + .llseek = default_llseek, }; static const struct file_operations ftrace_event_format_fops = { - .open = tracing_open_generic, - .read = event_format_read, + .open = trace_format_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release, }; static const struct file_operations ftrace_event_id_fops = { .open = tracing_open_generic, .read = event_id_read, + .llseek = default_llseek, }; static const struct file_operations ftrace_event_filter_fops = { .open = tracing_open_generic, .read = event_filter_read, .write = event_filter_write, + .llseek = default_llseek, }; static const struct file_operations ftrace_subsystem_filter_fops = { .open = tracing_open_generic, .read = subsystem_filter_read, .write = subsystem_filter_write, + .llseek = default_llseek, }; static const struct file_operations ftrace_system_enable_fops = { .open = tracing_open_generic, .read = system_enable_read, .write = system_enable_write, + .llseek = default_llseek, }; static const struct file_operations ftrace_show_header_fops = { .open = tracing_open_generic, .read = show_header, + .llseek = default_llseek, }; static struct dentry *event_trace_events_dir(void) @@ -919,14 +1075,15 @@ event_create_dir(struct ftrace_event_call *call, struct dentry *d_events, const struct file_operations *filter, const struct file_operations *format) { + struct list_head *head; 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); + if (strcmp(call->class->system, TRACE_SYSTEM) != 0) + d_events = event_subsystem_dir(call->class->system, d_events); call->dir = debugfs_create_dir(call->name, d_events); if (!call->dir) { @@ -935,26 +1092,31 @@ event_create_dir(struct ftrace_event_call *call, struct dentry *d_events, return -1; } - if (call->regfunc) + if (call->class->reg) trace_create_file("enable", 0644, call->dir, call, enable); - if (call->id && call->perf_event_enable) +#ifdef CONFIG_PERF_EVENTS + if (call->event.type && call->class->reg) trace_create_file("id", 0444, call->dir, call, id); +#endif - if (call->define_fields) { - ret = trace_define_common_fields(call); - if (!ret) - ret = call->define_fields(call); + /* + * Other events may have the same class. Only update + * the fields if they are not already defined. + */ + head = trace_get_fields(call); + if (list_empty(head)) { + ret = call->class->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); } + trace_create_file("filter", 0644, call->dir, call, + filter); trace_create_file("format", 0444, call->dir, call, format); @@ -962,20 +1124,26 @@ event_create_dir(struct ftrace_event_call *call, struct dentry *d_events, return 0; } -static int __trace_add_event_call(struct ftrace_event_call *call) +static int +__trace_add_event_call(struct ftrace_event_call *call, struct module *mod, + const struct file_operations *id, + const struct file_operations *enable, + const struct file_operations *filter, + const struct file_operations *format) { struct dentry *d_events; int ret; + /* The linker may leave blanks */ if (!call->name) return -EINVAL; - if (call->raw_init) { - ret = call->raw_init(call); + if (call->class->raw_init) { + ret = call->class->raw_init(call); if (ret < 0) { if (ret != -ENOSYS) - pr_warning("Could not initialize trace " - "events/%s\n", call->name); + pr_warning("Could not initialize trace events/%s\n", + call->name); return ret; } } @@ -984,11 +1152,10 @@ static int __trace_add_event_call(struct ftrace_event_call *call) 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); + ret = event_create_dir(call, d_events, id, enable, filter, format); if (!ret) list_add(&call->list, &ftrace_events); + call->mod = mod; return ret; } @@ -998,7 +1165,10 @@ int trace_add_event_call(struct ftrace_event_call *call) { int ret; mutex_lock(&event_mutex); - ret = __trace_add_event_call(call); + ret = __trace_add_event_call(call, NULL, &ftrace_event_id_fops, + &ftrace_enable_fops, + &ftrace_event_filter_fops, + &ftrace_event_format_fops); mutex_unlock(&event_mutex); return ret; } @@ -1035,13 +1205,13 @@ static void remove_subsystem_dir(const char *name) 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); + if (call->event.funcs) + __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_subsystem_dir(call->class->system); } /* Remove an event_call */ @@ -1115,8 +1285,6 @@ 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; @@ -1124,38 +1292,14 @@ static void trace_module_add_events(struct module *mod) if (start == end) return; - d_events = event_trace_events_dir(); - if (!d_events) + file_ops = trace_create_file_ops(mod); + if (!file_ops) 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, + __trace_add_event_call(call, mod, &file_ops->id, &file_ops->enable, &file_ops->filter, &file_ops->format); - if (!ret) - list_add(&call->list, &ftrace_events); } } @@ -1282,25 +1426,14 @@ static __init int event_trace_init(void) trace_create_file("enable", 0644, d_events, NULL, &ftrace_system_enable_fops); + if (trace_define_common_fields()) + pr_warning("tracing: Failed to allocate common fields"); + 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, + __trace_add_event_call(call, NULL, &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) { @@ -1388,8 +1521,8 @@ static __init void event_trace_self_tests(void) list_for_each_entry(call, &ftrace_events, list) { - /* Only test those that have a regfunc */ - if (!call->regfunc) + /* Only test those that have a probe */ + if (!call->class || !call->class->probe) continue; /* @@ -1399,8 +1532,8 @@ static __init void event_trace_self_tests(void) * syscalls as we test. */ #ifndef CONFIG_EVENT_TRACE_TEST_SYSCALLS - if (call->system && - strcmp(call->system, "syscalls") == 0) + if (call->class->system && + strcmp(call->class->system, "syscalls") == 0) continue; #endif @@ -1410,7 +1543,7 @@ static __init void event_trace_self_tests(void) * If an event is already enabled, someone is using * it and the self test should not be on. */ - if (call->enabled) { + if (call->flags & TRACE_EVENT_FL_ENABLED) { pr_warning("Enabled event during self test!\n"); WARN_ON_ONCE(1); continue; @@ -1487,12 +1620,11 @@ function_test_events_call(unsigned long ip, unsigned long parent_ip) struct ftrace_entry *entry; unsigned long flags; long disabled; - int resched; int cpu; int pc; pc = preempt_count(); - resched = ftrace_preempt_disable(); + preempt_disable_notrace(); cpu = raw_smp_processor_id(); disabled = atomic_inc_return(&per_cpu(ftrace_test_event_disable, cpu)); @@ -1514,7 +1646,7 @@ function_test_events_call(unsigned long ip, unsigned long parent_ip) out: atomic_dec(&per_cpu(ftrace_test_event_disable, cpu)); - ftrace_preempt_enable(resched); + preempt_enable_notrace(); } static struct ftrace_ops trace_ops __initdata = diff --git a/kernel/trace/trace_events_filter.c b/kernel/trace/trace_events_filter.c index 58092d844a1..36d40104b17 100644 --- a/kernel/trace/trace_events_filter.c +++ b/kernel/trace/trace_events_filter.c @@ -497,11 +497,11 @@ void print_subsystem_event_filter(struct event_subsystem *system, } static struct ftrace_event_field * -find_event_field(struct ftrace_event_call *call, char *name) +__find_event_field(struct list_head *head, char *name) { struct ftrace_event_field *field; - list_for_each_entry(field, &call->fields, link) { + list_for_each_entry(field, head, link) { if (!strcmp(field->name, name)) return field; } @@ -509,6 +509,20 @@ find_event_field(struct ftrace_event_call *call, char *name) return NULL; } +static struct ftrace_event_field * +find_event_field(struct ftrace_event_call *call, char *name) +{ + struct ftrace_event_field *field; + struct list_head *head; + + field = __find_event_field(&ftrace_common_fields, name); + if (field) + return field; + + head = trace_get_fields(call); + return __find_event_field(head, name); +} + static void filter_free_pred(struct filter_pred *pred) { if (!pred) @@ -545,7 +559,7 @@ static void filter_disable_preds(struct ftrace_event_call *call) struct event_filter *filter = call->filter; int i; - call->filter_active = 0; + call->flags &= ~TRACE_EVENT_FL_FILTERED; filter->n_preds = 0; for (i = 0; i < MAX_FILTER_PRED; i++) @@ -572,7 +586,7 @@ void destroy_preds(struct ftrace_event_call *call) { __free_preds(call->filter); call->filter = NULL; - call->filter_active = 0; + call->flags &= ~TRACE_EVENT_FL_FILTERED; } static struct event_filter *__alloc_preds(void) @@ -611,7 +625,7 @@ static int init_preds(struct ftrace_event_call *call) if (call->filter) return 0; - call->filter_active = 0; + call->flags &= ~TRACE_EVENT_FL_FILTERED; call->filter = __alloc_preds(); if (IS_ERR(call->filter)) return PTR_ERR(call->filter); @@ -625,10 +639,7 @@ static int init_subsystem_preds(struct event_subsystem *system) int err; list_for_each_entry(call, &ftrace_events, list) { - if (!call->define_fields) - continue; - - if (strcmp(call->system, system->name) != 0) + if (strcmp(call->class->system, system->name) != 0) continue; err = init_preds(call); @@ -644,10 +655,7 @@ 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) + if (strcmp(call->class->system, system->name) != 0) continue; filter_disable_preds(call); @@ -1249,10 +1257,7 @@ static int replace_system_preds(struct event_subsystem *system, 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) + if (strcmp(call->class->system, system->name) != 0) continue; /* try to see if the filter can be applied */ @@ -1266,7 +1271,7 @@ static int replace_system_preds(struct event_subsystem *system, if (err) filter_disable_preds(call); else { - call->filter_active = 1; + call->flags |= TRACE_EVENT_FL_FILTERED; replace_filter_string(filter, filter_string); } fail = false; @@ -1315,7 +1320,7 @@ int apply_event_filter(struct ftrace_event_call *call, char *filter_string) if (err) append_filter_err(ps, call->filter); else - call->filter_active = 1; + call->flags |= TRACE_EVENT_FL_FILTERED; out: filter_opstack_clear(ps); postfix_clear(ps); @@ -1393,7 +1398,7 @@ int ftrace_profile_set_filter(struct perf_event *event, int event_id, mutex_lock(&event_mutex); list_for_each_entry(call, &ftrace_events, list) { - if (call->id == event_id) + if (call->event.type == event_id) break; } diff --git a/kernel/trace/trace_export.c b/kernel/trace/trace_export.c index e091f64ba6c..4b74d71705c 100644 --- a/kernel/trace/trace_export.c +++ b/kernel/trace/trace_export.c @@ -83,13 +83,19 @@ static void __always_unused ____ftrace_check_##name(void) \ #undef __array #define __array(type, item, len) \ - BUILD_BUG_ON(len > MAX_FILTER_STR_VAL); \ - ret = trace_define_field(event_call, #type "[" #len "]", #item, \ + do { \ + BUILD_BUG_ON(len > MAX_FILTER_STR_VAL); \ + mutex_lock(&event_storage_mutex); \ + snprintf(event_storage, sizeof(event_storage), \ + "%s[%d]", #type, len); \ + ret = trace_define_field(event_call, event_storage, #item, \ offsetof(typeof(field), item), \ sizeof(field.item), \ is_signed_type(type), FILTER_OTHER); \ - if (ret) \ - return ret; + mutex_unlock(&event_storage_mutex); \ + if (ret) \ + return ret; \ + } while (0); #undef __array_desc #define __array_desc(type, container, item, len) \ @@ -125,12 +131,6 @@ ftrace_define_fields_##name(struct ftrace_event_call *event_call) \ #include "trace_entries.h" -static int ftrace_raw_init_event(struct ftrace_event_call *call) -{ - INIT_LIST_HEAD(&call->fields); - return 0; -} - #undef __entry #define __entry REC @@ -153,17 +153,21 @@ static int ftrace_raw_init_event(struct ftrace_event_call *call) #define F_printk(fmt, args...) #fmt ", " __stringify(args) #undef FTRACE_ENTRY -#define FTRACE_ENTRY(call, struct_name, type, tstruct, print) \ +#define FTRACE_ENTRY(call, struct_name, etype, tstruct, print) \ + \ +struct ftrace_event_class event_class_ftrace_##call = { \ + .system = __stringify(TRACE_SYSTEM), \ + .define_fields = ftrace_define_fields_##call, \ + .fields = LIST_HEAD_INIT(event_class_ftrace_##call.fields),\ +}; \ \ 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, \ + .event.type = etype, \ + .class = &event_class_ftrace_##call, \ .print_fmt = print, \ - .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 b3f3776b0cd..16aee4d44e8 100644 --- a/kernel/trace/trace_functions.c +++ b/kernel/trace/trace_functions.c @@ -54,14 +54,14 @@ function_trace_call_preempt_only(unsigned long ip, unsigned long parent_ip) struct trace_array_cpu *data; unsigned long flags; long disabled; - int cpu, resched; + int cpu; int pc; if (unlikely(!ftrace_function_enabled)) return; pc = preempt_count(); - resched = ftrace_preempt_disable(); + preempt_disable_notrace(); local_save_flags(flags); cpu = raw_smp_processor_id(); data = tr->data[cpu]; @@ -71,7 +71,7 @@ function_trace_call_preempt_only(unsigned long ip, unsigned long parent_ip) trace_function(tr, ip, parent_ip, flags, pc); atomic_dec(&data->disabled); - ftrace_preempt_enable(resched); + preempt_enable_notrace(); } static void diff --git a/kernel/trace/trace_functions_graph.c b/kernel/trace/trace_functions_graph.c index dd11c830eb8..76b05980225 100644 --- a/kernel/trace/trace_functions_graph.c +++ b/kernel/trace/trace_functions_graph.c @@ -15,15 +15,19 @@ #include "trace.h" #include "trace_output.h" +/* When set, irq functions will be ignored */ +static int ftrace_graph_skip_irqs; + struct fgraph_cpu_data { pid_t last_pid; int depth; + int depth_irq; int ignore; unsigned long enter_funcs[FTRACE_RETFUNC_DEPTH]; }; struct fgraph_data { - struct fgraph_cpu_data *cpu_data; + struct fgraph_cpu_data __percpu *cpu_data; /* Place to preserve last processed entry. */ struct ftrace_graph_ent_entry ent; @@ -41,6 +45,7 @@ struct fgraph_data { #define TRACE_GRAPH_PRINT_PROC 0x8 #define TRACE_GRAPH_PRINT_DURATION 0x10 #define TRACE_GRAPH_PRINT_ABS_TIME 0x20 +#define TRACE_GRAPH_PRINT_IRQS 0x40 static struct tracer_opt trace_opts[] = { /* Display overruns? (for self-debug purpose) */ @@ -55,13 +60,15 @@ static struct tracer_opt trace_opts[] = { { TRACER_OPT(funcgraph-duration, TRACE_GRAPH_PRINT_DURATION) }, /* Display absolute time of an entry */ { TRACER_OPT(funcgraph-abstime, TRACE_GRAPH_PRINT_ABS_TIME) }, + /* Display interrupts */ + { TRACER_OPT(funcgraph-irqs, TRACE_GRAPH_PRINT_IRQS) }, { } /* 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 | - TRACE_GRAPH_PRINT_DURATION, + TRACE_GRAPH_PRINT_DURATION | TRACE_GRAPH_PRINT_IRQS, .opts = trace_opts }; @@ -204,6 +211,14 @@ int __trace_graph_entry(struct trace_array *tr, return 1; } +static inline int ftrace_graph_ignore_irqs(void) +{ + if (!ftrace_graph_skip_irqs) + return 0; + + return in_irq(); +} + int trace_graph_entry(struct ftrace_graph_ent *trace) { struct trace_array *tr = graph_array; @@ -218,7 +233,8 @@ int trace_graph_entry(struct ftrace_graph_ent *trace) return 0; /* trace it when it is-nested-in or is a function enabled. */ - if (!(trace->depth || ftrace_graph_addr(trace->func))) + if (!(trace->depth || ftrace_graph_addr(trace->func)) || + ftrace_graph_ignore_irqs()) return 0; local_irq_save(flags); @@ -246,6 +262,34 @@ int trace_graph_thresh_entry(struct ftrace_graph_ent *trace) return trace_graph_entry(trace); } +static void +__trace_graph_function(struct trace_array *tr, + unsigned long ip, unsigned long flags, int pc) +{ + u64 time = trace_clock_local(); + struct ftrace_graph_ent ent = { + .func = ip, + .depth = 0, + }; + struct ftrace_graph_ret ret = { + .func = ip, + .depth = 0, + .calltime = time, + .rettime = time, + }; + + __trace_graph_entry(tr, &ent, flags, pc); + __trace_graph_return(tr, &ret, flags, pc); +} + +void +trace_graph_function(struct trace_array *tr, + unsigned long ip, unsigned long parent_ip, + unsigned long flags, int pc) +{ + __trace_graph_function(tr, ip, flags, pc); +} + void __trace_graph_return(struct trace_array *tr, struct ftrace_graph_ret *trace, unsigned long flags, @@ -507,7 +551,15 @@ get_return_for_leaf(struct trace_iterator *iter, * if the output fails. */ data->ent = *curr; - data->ret = *next; + /* + * If the next event is not a return type, then + * we only care about what type it is. Otherwise we can + * safely copy the entire event. + */ + if (next->ent.type == TRACE_GRAPH_RET) + data->ret = *next; + else + data->ret.ent.type = next->ent.type; } } @@ -641,7 +693,9 @@ trace_print_graph_duration(unsigned long long duration, struct trace_seq *s) /* Print nsecs (we don't want to exceed 7 numbers) */ if (len < 7) { - snprintf(nsecs_str, 8 - len, "%03lu", nsecs_rem); + size_t slen = min_t(size_t, sizeof(nsecs_str), 8UL - len); + + snprintf(nsecs_str, slen, "%03lu", nsecs_rem); ret = trace_seq_printf(s, ".%s", nsecs_str); if (!ret) return TRACE_TYPE_PARTIAL_LINE; @@ -846,6 +900,108 @@ print_graph_prologue(struct trace_iterator *iter, struct trace_seq *s, return 0; } +/* + * Entry check for irq code + * + * returns 1 if + * - we are inside irq code + * - we just extered irq code + * + * retunns 0 if + * - funcgraph-interrupts option is set + * - we are not inside irq code + */ +static int +check_irq_entry(struct trace_iterator *iter, u32 flags, + unsigned long addr, int depth) +{ + int cpu = iter->cpu; + int *depth_irq; + struct fgraph_data *data = iter->private; + + /* + * If we are either displaying irqs, or we got called as + * a graph event and private data does not exist, + * then we bypass the irq check. + */ + if ((flags & TRACE_GRAPH_PRINT_IRQS) || + (!data)) + return 0; + + depth_irq = &(per_cpu_ptr(data->cpu_data, cpu)->depth_irq); + + /* + * We are inside the irq code + */ + if (*depth_irq >= 0) + return 1; + + if ((addr < (unsigned long)__irqentry_text_start) || + (addr >= (unsigned long)__irqentry_text_end)) + return 0; + + /* + * We are entering irq code. + */ + *depth_irq = depth; + return 1; +} + +/* + * Return check for irq code + * + * returns 1 if + * - we are inside irq code + * - we just left irq code + * + * returns 0 if + * - funcgraph-interrupts option is set + * - we are not inside irq code + */ +static int +check_irq_return(struct trace_iterator *iter, u32 flags, int depth) +{ + int cpu = iter->cpu; + int *depth_irq; + struct fgraph_data *data = iter->private; + + /* + * If we are either displaying irqs, or we got called as + * a graph event and private data does not exist, + * then we bypass the irq check. + */ + if ((flags & TRACE_GRAPH_PRINT_IRQS) || + (!data)) + return 0; + + depth_irq = &(per_cpu_ptr(data->cpu_data, cpu)->depth_irq); + + /* + * We are not inside the irq code. + */ + if (*depth_irq == -1) + return 0; + + /* + * We are inside the irq code, and this is returning entry. + * Let's not trace it and clear the entry depth, since + * we are out of irq code. + * + * This condition ensures that we 'leave the irq code' once + * we are out of the entry depth. Thus protecting us from + * the RETURN entry loss. + */ + if (*depth_irq >= depth) { + *depth_irq = -1; + return 1; + } + + /* + * We are inside the irq code, and this is not the entry. + */ + return 1; +} + static enum print_line_t print_graph_entry(struct ftrace_graph_ent_entry *field, struct trace_seq *s, struct trace_iterator *iter, u32 flags) @@ -856,6 +1012,9 @@ print_graph_entry(struct ftrace_graph_ent_entry *field, struct trace_seq *s, static enum print_line_t ret; int cpu = iter->cpu; + if (check_irq_entry(iter, flags, call->func, call->depth)) + return TRACE_TYPE_HANDLED; + if (print_graph_prologue(iter, s, TRACE_GRAPH_ENT, call->func, flags)) return TRACE_TYPE_PARTIAL_LINE; @@ -893,6 +1052,9 @@ print_graph_return(struct ftrace_graph_ret *trace, struct trace_seq *s, int ret; int i; + if (check_irq_return(iter, flags, trace->depth)) + return TRACE_TYPE_HANDLED; + if (data) { struct fgraph_cpu_data *cpu_data; int cpu = iter->cpu; @@ -1025,7 +1187,7 @@ print_graph_comment(struct trace_seq *s, struct trace_entry *ent, if (!event) return TRACE_TYPE_UNHANDLED; - ret = event->trace(iter, sym_flags); + ret = event->funcs->trace(iter, sym_flags, event); if (ret != TRACE_TYPE_HANDLED) return ret; } @@ -1045,7 +1207,7 @@ print_graph_comment(struct trace_seq *s, struct trace_entry *ent, enum print_line_t -print_graph_function_flags(struct trace_iterator *iter, u32 flags) +__print_graph_function_flags(struct trace_iterator *iter, u32 flags) { struct ftrace_graph_ent_entry *field; struct fgraph_data *data = iter->private; @@ -1108,11 +1270,23 @@ print_graph_function_flags(struct trace_iterator *iter, u32 flags) static enum print_line_t print_graph_function(struct trace_iterator *iter) { - return print_graph_function_flags(iter, tracer_flags.val); + return __print_graph_function_flags(iter, tracer_flags.val); +} + +enum print_line_t print_graph_function_flags(struct trace_iterator *iter, + u32 flags) +{ + if (trace_flags & TRACE_ITER_LATENCY_FMT) + flags |= TRACE_GRAPH_PRINT_DURATION; + else + flags |= TRACE_GRAPH_PRINT_ABS_TIME; + + return __print_graph_function_flags(iter, flags); } static enum print_line_t -print_graph_function_event(struct trace_iterator *iter, int flags) +print_graph_function_event(struct trace_iterator *iter, int flags, + struct trace_event *event) { return print_graph_function(iter); } @@ -1139,7 +1313,7 @@ static void print_lat_header(struct seq_file *s, u32 flags) seq_printf(s, "#%.*s|||| / \n", size, spaces); } -void print_graph_headers_flags(struct seq_file *s, u32 flags) +static void __print_graph_headers_flags(struct seq_file *s, u32 flags) { int lat = trace_flags & TRACE_ITER_LATENCY_FMT; @@ -1180,6 +1354,23 @@ void print_graph_headers(struct seq_file *s) print_graph_headers_flags(s, tracer_flags.val); } +void print_graph_headers_flags(struct seq_file *s, u32 flags) +{ + struct trace_iterator *iter = s->private; + + if (trace_flags & TRACE_ITER_LATENCY_FMT) { + /* print nothing if the buffers are empty */ + if (trace_empty(iter)) + return; + + print_trace_header(s, iter); + flags |= TRACE_GRAPH_PRINT_DURATION; + } else + flags |= TRACE_GRAPH_PRINT_ABS_TIME; + + __print_graph_headers_flags(s, flags); +} + void graph_trace_open(struct trace_iterator *iter) { /* pid and depth on the last trace processed */ @@ -1200,9 +1391,12 @@ void graph_trace_open(struct trace_iterator *iter) pid_t *pid = &(per_cpu_ptr(data->cpu_data, cpu)->last_pid); int *depth = &(per_cpu_ptr(data->cpu_data, cpu)->depth); int *ignore = &(per_cpu_ptr(data->cpu_data, cpu)->ignore); + int *depth_irq = &(per_cpu_ptr(data->cpu_data, cpu)->depth_irq); + *pid = -1; *depth = 0; *ignore = 0; + *depth_irq = -1; } iter->private = data; @@ -1225,14 +1419,26 @@ void graph_trace_close(struct trace_iterator *iter) } } +static int func_graph_set_flag(u32 old_flags, u32 bit, int set) +{ + if (bit == TRACE_GRAPH_PRINT_IRQS) + ftrace_graph_skip_irqs = !set; + + return 0; +} + +static struct trace_event_functions graph_functions = { + .trace = print_graph_function_event, +}; + static struct trace_event graph_trace_entry_event = { .type = TRACE_GRAPH_ENT, - .trace = print_graph_function_event, + .funcs = &graph_functions, }; static struct trace_event graph_trace_ret_event = { .type = TRACE_GRAPH_RET, - .trace = print_graph_function_event, + .funcs = &graph_functions }; static struct tracer graph_trace __read_mostly = { @@ -1247,6 +1453,7 @@ static struct tracer graph_trace __read_mostly = { .print_line = print_graph_function, .print_header = print_graph_headers, .flags = &tracer_flags, + .set_flag = func_graph_set_flag, #ifdef CONFIG_FTRACE_SELFTEST .selftest = trace_selftest_startup_function_graph, #endif diff --git a/kernel/trace/trace_irqsoff.c b/kernel/trace/trace_irqsoff.c index 6fd486e0cef..92b6e1e12d9 100644 --- a/kernel/trace/trace_irqsoff.c +++ b/kernel/trace/trace_irqsoff.c @@ -87,14 +87,22 @@ static __cacheline_aligned_in_smp unsigned long max_sequence; #ifdef CONFIG_FUNCTION_TRACER /* - * irqsoff uses its own tracer function to keep the overhead down: + * Prologue for the preempt and irqs off function tracers. + * + * Returns 1 if it is OK to continue, and data->disabled is + * incremented. + * 0 if the trace is to be ignored, and data->disabled + * is kept the same. + * + * Note, this function is also used outside this ifdef but + * inside the #ifdef of the function graph tracer below. + * This is OK, since the function graph tracer is + * dependent on the function tracer. */ -static void -irqsoff_tracer_call(unsigned long ip, unsigned long parent_ip) +static int func_prolog_dec(struct trace_array *tr, + struct trace_array_cpu **data, + unsigned long *flags) { - struct trace_array *tr = irqsoff_trace; - struct trace_array_cpu *data; - unsigned long flags; long disabled; int cpu; @@ -106,18 +114,38 @@ irqsoff_tracer_call(unsigned long ip, unsigned long parent_ip) */ cpu = raw_smp_processor_id(); if (likely(!per_cpu(tracing_cpu, cpu))) - return; + return 0; - local_save_flags(flags); + local_save_flags(*flags); /* slight chance to get a false positive on tracing_cpu */ - if (!irqs_disabled_flags(flags)) - return; + if (!irqs_disabled_flags(*flags)) + return 0; - data = tr->data[cpu]; - disabled = atomic_inc_return(&data->disabled); + *data = tr->data[cpu]; + disabled = atomic_inc_return(&(*data)->disabled); if (likely(disabled == 1)) - trace_function(tr, ip, parent_ip, flags, preempt_count()); + return 1; + + atomic_dec(&(*data)->disabled); + + return 0; +} + +/* + * irqsoff uses its own tracer function to keep the overhead down: + */ +static void +irqsoff_tracer_call(unsigned long ip, unsigned long parent_ip) +{ + struct trace_array *tr = irqsoff_trace; + struct trace_array_cpu *data; + unsigned long flags; + + if (!func_prolog_dec(tr, &data, &flags)) + return; + + trace_function(tr, ip, parent_ip, flags, preempt_count()); atomic_dec(&data->disabled); } @@ -155,30 +183,16 @@ static int irqsoff_graph_entry(struct ftrace_graph_ent *trace) struct trace_array *tr = irqsoff_trace; struct trace_array_cpu *data; unsigned long flags; - long disabled; int ret; - int cpu; int pc; - cpu = raw_smp_processor_id(); - if (likely(!per_cpu(tracing_cpu, cpu))) - return 0; - - local_save_flags(flags); - /* slight chance to get a false positive on tracing_cpu */ - if (!irqs_disabled_flags(flags)) + if (!func_prolog_dec(tr, &data, &flags)) return 0; - 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; - + pc = preempt_count(); + ret = __trace_graph_entry(tr, trace, flags, pc); atomic_dec(&data->disabled); + return ret; } @@ -187,27 +201,13 @@ static void irqsoff_graph_return(struct ftrace_graph_ret *trace) struct trace_array *tr = irqsoff_trace; struct trace_array_cpu *data; unsigned long flags; - long disabled; - int cpu; int pc; - cpu = raw_smp_processor_id(); - if (likely(!per_cpu(tracing_cpu, cpu))) - return; - - local_save_flags(flags); - /* slight chance to get a false positive on tracing_cpu */ - if (!irqs_disabled_flags(flags)) + if (!func_prolog_dec(tr, &data, &flags)) return; - data = tr->data[cpu]; - disabled = atomic_inc_return(&data->disabled); - - if (likely(disabled == 1)) { - pc = preempt_count(); - __trace_graph_return(tr, trace, flags, pc); - } - + pc = preempt_count(); + __trace_graph_return(tr, trace, flags, pc); atomic_dec(&data->disabled); } @@ -229,75 +229,33 @@ static void irqsoff_trace_close(struct trace_iterator *iter) static enum print_line_t irqsoff_print_line(struct trace_iterator *iter) { - u32 flags = GRAPH_TRACER_FLAGS; - - if (trace_flags & TRACE_ITER_LATENCY_FMT) - flags |= TRACE_GRAPH_PRINT_DURATION; - else - flags |= TRACE_GRAPH_PRINT_ABS_TIME; - /* * In graph mode call the graph tracer output function, * otherwise go with the TRACE_FN event handler */ if (is_graph()) - return print_graph_function_flags(iter, flags); + return print_graph_function_flags(iter, GRAPH_TRACER_FLAGS); return TRACE_TYPE_UNHANDLED; } static void irqsoff_print_header(struct seq_file *s) { - if (is_graph()) { - struct trace_iterator *iter = s->private; - u32 flags = GRAPH_TRACER_FLAGS; - - if (trace_flags & TRACE_ITER_LATENCY_FMT) { - /* print nothing if the buffers are empty */ - if (trace_empty(iter)) - return; - - print_trace_header(s, iter); - flags |= TRACE_GRAPH_PRINT_DURATION; - } else - flags |= TRACE_GRAPH_PRINT_ABS_TIME; - - print_graph_headers_flags(s, flags); - } else + if (is_graph()) + print_graph_headers_flags(s, GRAPH_TRACER_FLAGS); + else trace_default_header(s); } static void -trace_graph_function(struct trace_array *tr, - unsigned long ip, unsigned long flags, int pc) -{ - u64 time = trace_clock_local(); - struct ftrace_graph_ent ent = { - .func = ip, - .depth = 0, - }; - struct ftrace_graph_ret ret = { - .func = ip, - .depth = 0, - .calltime = time, - .rettime = time, - }; - - __trace_graph_entry(tr, &ent, flags, pc); - __trace_graph_return(tr, &ret, flags, pc); -} - -static void __trace_function(struct trace_array *tr, unsigned long ip, unsigned long parent_ip, unsigned long flags, int pc) { - if (!is_graph()) + if (is_graph()) + trace_graph_function(tr, ip, parent_ip, flags, pc); + else trace_function(tr, ip, parent_ip, flags, pc); - else { - trace_graph_function(tr, parent_ip, flags, pc); - trace_graph_function(tr, ip, flags, pc); - } } #else @@ -495,14 +453,6 @@ void time_hardirqs_off(unsigned long a0, unsigned long a1) * Stubs: */ -void early_boot_irqs_off(void) -{ -} - -void early_boot_irqs_on(void) -{ -} - void trace_softirqs_on(unsigned long ip) { } @@ -649,6 +599,7 @@ static struct tracer irqsoff_tracer __read_mostly = #endif .open = irqsoff_trace_open, .close = irqsoff_trace_close, + .use_max_tr = 1, }; # define register_irqsoff(trace) register_tracer(&trace) #else @@ -681,6 +632,7 @@ static struct tracer preemptoff_tracer __read_mostly = #endif .open = irqsoff_trace_open, .close = irqsoff_trace_close, + .use_max_tr = 1, }; # define register_preemptoff(trace) register_tracer(&trace) #else @@ -715,6 +667,7 @@ static struct tracer preemptirqsoff_tracer __read_mostly = #endif .open = irqsoff_trace_open, .close = irqsoff_trace_close, + .use_max_tr = 1, }; # define register_preemptirqsoff(trace) register_tracer(&trace) diff --git a/kernel/trace/trace_kdb.c b/kernel/trace/trace_kdb.c new file mode 100644 index 00000000000..3c5c5dfea0b --- /dev/null +++ b/kernel/trace/trace_kdb.c @@ -0,0 +1,135 @@ +/* + * kdb helper for dumping the ftrace buffer + * + * Copyright (C) 2010 Jason Wessel <jason.wessel@windriver.com> + * + * ftrace_dump_buf based on ftrace_dump: + * Copyright (C) 2007-2008 Steven Rostedt <srostedt@redhat.com> + * Copyright (C) 2008 Ingo Molnar <mingo@redhat.com> + * + */ +#include <linux/init.h> +#include <linux/kgdb.h> +#include <linux/kdb.h> +#include <linux/ftrace.h> + +#include "trace.h" +#include "trace_output.h" + +static void ftrace_dump_buf(int skip_lines, long cpu_file) +{ + /* use static because iter can be a bit big for the stack */ + static struct trace_iterator iter; + unsigned int old_userobj; + int cnt = 0, cpu; + + trace_init_global_iter(&iter); + + for_each_tracing_cpu(cpu) { + atomic_inc(&iter.tr->data[cpu]->disabled); + } + + old_userobj = trace_flags; + + /* don't look at user memory in panic mode */ + trace_flags &= ~TRACE_ITER_SYM_USEROBJ; + + kdb_printf("Dumping ftrace buffer:\n"); + + /* reset all but tr, trace, and overruns */ + memset(&iter.seq, 0, + sizeof(struct trace_iterator) - + offsetof(struct trace_iterator, seq)); + iter.iter_flags |= TRACE_FILE_LAT_FMT; + iter.pos = -1; + + if (cpu_file == TRACE_PIPE_ALL_CPU) { + for_each_tracing_cpu(cpu) { + iter.buffer_iter[cpu] = + ring_buffer_read_prepare(iter.tr->buffer, cpu); + ring_buffer_read_start(iter.buffer_iter[cpu]); + tracing_iter_reset(&iter, cpu); + } + } else { + iter.cpu_file = cpu_file; + iter.buffer_iter[cpu_file] = + ring_buffer_read_prepare(iter.tr->buffer, cpu_file); + ring_buffer_read_start(iter.buffer_iter[cpu_file]); + tracing_iter_reset(&iter, cpu_file); + } + if (!trace_empty(&iter)) + trace_find_next_entry_inc(&iter); + while (!trace_empty(&iter)) { + if (!cnt) + kdb_printf("---------------------------------\n"); + cnt++; + + if (trace_find_next_entry_inc(&iter) != NULL && !skip_lines) + print_trace_line(&iter); + if (!skip_lines) + trace_printk_seq(&iter.seq); + else + skip_lines--; + if (KDB_FLAG(CMD_INTERRUPT)) + goto out; + } + + if (!cnt) + kdb_printf(" (ftrace buffer empty)\n"); + else + kdb_printf("---------------------------------\n"); + +out: + trace_flags = old_userobj; + + for_each_tracing_cpu(cpu) { + atomic_dec(&iter.tr->data[cpu]->disabled); + } + + for_each_tracing_cpu(cpu) + if (iter.buffer_iter[cpu]) + ring_buffer_read_finish(iter.buffer_iter[cpu]); +} + +/* + * kdb_ftdump - Dump the ftrace log buffer + */ +static int kdb_ftdump(int argc, const char **argv) +{ + int skip_lines = 0; + long cpu_file; + char *cp; + + if (argc > 2) + return KDB_ARGCOUNT; + + if (argc) { + skip_lines = simple_strtol(argv[1], &cp, 0); + if (*cp) + skip_lines = 0; + } + + if (argc == 2) { + cpu_file = simple_strtol(argv[2], &cp, 0); + if (*cp || cpu_file >= NR_CPUS || cpu_file < 0 || + !cpu_online(cpu_file)) + return KDB_BADINT; + } else { + cpu_file = TRACE_PIPE_ALL_CPU; + } + + kdb_trap_printk++; + ftrace_dump_buf(skip_lines, cpu_file); + kdb_trap_printk--; + + return 0; +} + +static __init int kdb_ftrace_register(void) +{ + kdb_register_repeat("ftdump", kdb_ftdump, "[skip_#lines] [cpu]", + "Dump ftrace log", 0, KDB_REPEAT_NONE); + return 0; +} + +late_initcall(kdb_ftrace_register); diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c index a7514326052..2dec9bcde8b 100644 --- a/kernel/trace/trace_kprobe.c +++ b/kernel/trace/trace_kprobe.c @@ -30,6 +30,7 @@ #include <linux/ptrace.h> #include <linux/perf_event.h> #include <linux/stringify.h> +#include <linux/limits.h> #include <asm/bitsperlong.h> #include "trace.h" @@ -38,6 +39,7 @@ #define MAX_TRACE_ARGS 128 #define MAX_ARGSTR_LEN 63 #define MAX_EVENT_NAME_LEN 64 +#define MAX_STRING_SIZE PATH_MAX #define KPROBE_EVENT_SYSTEM "kprobes" /* Reserved field names */ @@ -58,14 +60,16 @@ const char *reserved_field_names[] = { }; /* Printing function type */ -typedef int (*print_type_func_t)(struct trace_seq *, const char *, void *); +typedef int (*print_type_func_t)(struct trace_seq *, const char *, void *, + void *); #define PRINT_TYPE_FUNC_NAME(type) print_type_##type #define PRINT_TYPE_FMT_NAME(type) print_type_format_##type /* Printing in basic type function template */ #define DEFINE_BASIC_PRINT_TYPE_FUNC(type, fmt, cast) \ static __kprobes int PRINT_TYPE_FUNC_NAME(type)(struct trace_seq *s, \ - const char *name, void *data)\ + const char *name, \ + void *data, void *ent)\ { \ return trace_seq_printf(s, " %s=" fmt, name, (cast)*(type *)data);\ } \ @@ -80,6 +84,49 @@ DEFINE_BASIC_PRINT_TYPE_FUNC(s16, "%d", int) DEFINE_BASIC_PRINT_TYPE_FUNC(s32, "%ld", long) DEFINE_BASIC_PRINT_TYPE_FUNC(s64, "%lld", long long) +/* data_rloc: data relative location, compatible with u32 */ +#define make_data_rloc(len, roffs) \ + (((u32)(len) << 16) | ((u32)(roffs) & 0xffff)) +#define get_rloc_len(dl) ((u32)(dl) >> 16) +#define get_rloc_offs(dl) ((u32)(dl) & 0xffff) + +static inline void *get_rloc_data(u32 *dl) +{ + return (u8 *)dl + get_rloc_offs(*dl); +} + +/* For data_loc conversion */ +static inline void *get_loc_data(u32 *dl, void *ent) +{ + return (u8 *)ent + get_rloc_offs(*dl); +} + +/* + * Convert data_rloc to data_loc: + * data_rloc stores the offset from data_rloc itself, but data_loc + * stores the offset from event entry. + */ +#define convert_rloc_to_loc(dl, offs) ((u32)(dl) + (offs)) + +/* For defining macros, define string/string_size types */ +typedef u32 string; +typedef u32 string_size; + +/* Print type function for string type */ +static __kprobes int PRINT_TYPE_FUNC_NAME(string)(struct trace_seq *s, + const char *name, + void *data, void *ent) +{ + int len = *(u32 *)data >> 16; + + if (!len) + return trace_seq_printf(s, " %s=(fault)", name); + else + return trace_seq_printf(s, " %s=\"%s\"", name, + (const char *)get_loc_data(data, ent)); +} +static const char PRINT_TYPE_FMT_NAME(string)[] = "\\\"%s\\\""; + /* Data fetch function type */ typedef void (*fetch_func_t)(struct pt_regs *, void *, void *); @@ -94,32 +141,38 @@ static __kprobes void call_fetch(struct fetch_param *fprm, return fprm->fn(regs, fprm->data, dest); } -#define FETCH_FUNC_NAME(kind, type) fetch_##kind##_##type +#define FETCH_FUNC_NAME(method, type) fetch_##method##_##type /* * Define macro for basic types - we don't need to define s* types, because * we have to care only about bitwidth at recording time. */ -#define DEFINE_BASIC_FETCH_FUNCS(kind) \ -DEFINE_FETCH_##kind(u8) \ -DEFINE_FETCH_##kind(u16) \ -DEFINE_FETCH_##kind(u32) \ -DEFINE_FETCH_##kind(u64) - -#define CHECK_BASIC_FETCH_FUNCS(kind, fn) \ - ((FETCH_FUNC_NAME(kind, u8) == fn) || \ - (FETCH_FUNC_NAME(kind, u16) == fn) || \ - (FETCH_FUNC_NAME(kind, u32) == fn) || \ - (FETCH_FUNC_NAME(kind, u64) == fn)) +#define DEFINE_BASIC_FETCH_FUNCS(method) \ +DEFINE_FETCH_##method(u8) \ +DEFINE_FETCH_##method(u16) \ +DEFINE_FETCH_##method(u32) \ +DEFINE_FETCH_##method(u64) + +#define CHECK_FETCH_FUNCS(method, fn) \ + (((FETCH_FUNC_NAME(method, u8) == fn) || \ + (FETCH_FUNC_NAME(method, u16) == fn) || \ + (FETCH_FUNC_NAME(method, u32) == fn) || \ + (FETCH_FUNC_NAME(method, u64) == fn) || \ + (FETCH_FUNC_NAME(method, string) == fn) || \ + (FETCH_FUNC_NAME(method, string_size) == fn)) \ + && (fn != NULL)) /* Data fetch function templates */ #define DEFINE_FETCH_reg(type) \ static __kprobes void FETCH_FUNC_NAME(reg, type)(struct pt_regs *regs, \ - void *offset, void *dest) \ + void *offset, void *dest) \ { \ *(type *)dest = (type)regs_get_register(regs, \ (unsigned int)((unsigned long)offset)); \ } DEFINE_BASIC_FETCH_FUNCS(reg) +/* No string on the register */ +#define fetch_reg_string NULL +#define fetch_reg_string_size NULL #define DEFINE_FETCH_stack(type) \ static __kprobes void FETCH_FUNC_NAME(stack, type)(struct pt_regs *regs,\ @@ -129,6 +182,9 @@ static __kprobes void FETCH_FUNC_NAME(stack, type)(struct pt_regs *regs,\ (unsigned int)((unsigned long)offset)); \ } DEFINE_BASIC_FETCH_FUNCS(stack) +/* No string on the stack entry */ +#define fetch_stack_string NULL +#define fetch_stack_string_size NULL #define DEFINE_FETCH_retval(type) \ static __kprobes void FETCH_FUNC_NAME(retval, type)(struct pt_regs *regs,\ @@ -137,6 +193,9 @@ static __kprobes void FETCH_FUNC_NAME(retval, type)(struct pt_regs *regs,\ *(type *)dest = (type)regs_return_value(regs); \ } DEFINE_BASIC_FETCH_FUNCS(retval) +/* No string on the retval */ +#define fetch_retval_string NULL +#define fetch_retval_string_size NULL #define DEFINE_FETCH_memory(type) \ static __kprobes void FETCH_FUNC_NAME(memory, type)(struct pt_regs *regs,\ @@ -149,6 +208,62 @@ static __kprobes void FETCH_FUNC_NAME(memory, type)(struct pt_regs *regs,\ *(type *)dest = retval; \ } DEFINE_BASIC_FETCH_FUNCS(memory) +/* + * Fetch a null-terminated string. Caller MUST set *(u32 *)dest with max + * length and relative data location. + */ +static __kprobes void FETCH_FUNC_NAME(memory, string)(struct pt_regs *regs, + void *addr, void *dest) +{ + long ret; + int maxlen = get_rloc_len(*(u32 *)dest); + u8 *dst = get_rloc_data(dest); + u8 *src = addr; + mm_segment_t old_fs = get_fs(); + if (!maxlen) + return; + /* + * Try to get string again, since the string can be changed while + * probing. + */ + set_fs(KERNEL_DS); + pagefault_disable(); + do + ret = __copy_from_user_inatomic(dst++, src++, 1); + while (dst[-1] && ret == 0 && src - (u8 *)addr < maxlen); + dst[-1] = '\0'; + pagefault_enable(); + set_fs(old_fs); + + if (ret < 0) { /* Failed to fetch string */ + ((u8 *)get_rloc_data(dest))[0] = '\0'; + *(u32 *)dest = make_data_rloc(0, get_rloc_offs(*(u32 *)dest)); + } else + *(u32 *)dest = make_data_rloc(src - (u8 *)addr, + get_rloc_offs(*(u32 *)dest)); +} +/* Return the length of string -- including null terminal byte */ +static __kprobes void FETCH_FUNC_NAME(memory, string_size)(struct pt_regs *regs, + void *addr, void *dest) +{ + int ret, len = 0; + u8 c; + mm_segment_t old_fs = get_fs(); + + set_fs(KERNEL_DS); + pagefault_disable(); + do { + ret = __copy_from_user_inatomic(&c, (u8 *)addr + len, 1); + len++; + } while (c && ret == 0 && len < MAX_STRING_SIZE); + pagefault_enable(); + set_fs(old_fs); + + if (ret < 0) /* Failed to check the length */ + *(u32 *)dest = 0; + else + *(u32 *)dest = len; +} /* Memory fetching by symbol */ struct symbol_cache { @@ -203,6 +318,8 @@ static __kprobes void FETCH_FUNC_NAME(symbol, type)(struct pt_regs *regs,\ *(type *)dest = 0; \ } DEFINE_BASIC_FETCH_FUNCS(symbol) +DEFINE_FETCH_symbol(string) +DEFINE_FETCH_symbol(string_size) /* Dereference memory access function */ struct deref_fetch_param { @@ -224,12 +341,14 @@ static __kprobes void FETCH_FUNC_NAME(deref, type)(struct pt_regs *regs,\ *(type *)dest = 0; \ } DEFINE_BASIC_FETCH_FUNCS(deref) +DEFINE_FETCH_deref(string) +DEFINE_FETCH_deref(string_size) static __kprobes void free_deref_fetch_param(struct deref_fetch_param *data) { - if (CHECK_BASIC_FETCH_FUNCS(deref, data->orig.fn)) + if (CHECK_FETCH_FUNCS(deref, data->orig.fn)) free_deref_fetch_param(data->orig.data); - else if (CHECK_BASIC_FETCH_FUNCS(symbol, data->orig.fn)) + else if (CHECK_FETCH_FUNCS(symbol, data->orig.fn)) free_symbol_cache(data->orig.data); kfree(data); } @@ -240,23 +359,43 @@ static __kprobes void free_deref_fetch_param(struct deref_fetch_param *data) #define DEFAULT_FETCH_TYPE _DEFAULT_FETCH_TYPE(BITS_PER_LONG) #define DEFAULT_FETCH_TYPE_STR __stringify(DEFAULT_FETCH_TYPE) -#define ASSIGN_FETCH_FUNC(kind, type) \ - .kind = FETCH_FUNC_NAME(kind, type) - -#define ASSIGN_FETCH_TYPE(ptype, ftype, sign) \ - {.name = #ptype, \ - .size = sizeof(ftype), \ - .is_signed = sign, \ - .print = PRINT_TYPE_FUNC_NAME(ptype), \ - .fmt = PRINT_TYPE_FMT_NAME(ptype), \ -ASSIGN_FETCH_FUNC(reg, ftype), \ -ASSIGN_FETCH_FUNC(stack, ftype), \ -ASSIGN_FETCH_FUNC(retval, ftype), \ -ASSIGN_FETCH_FUNC(memory, ftype), \ -ASSIGN_FETCH_FUNC(symbol, ftype), \ -ASSIGN_FETCH_FUNC(deref, ftype), \ +/* Fetch types */ +enum { + FETCH_MTD_reg = 0, + FETCH_MTD_stack, + FETCH_MTD_retval, + FETCH_MTD_memory, + FETCH_MTD_symbol, + FETCH_MTD_deref, + FETCH_MTD_END, +}; + +#define ASSIGN_FETCH_FUNC(method, type) \ + [FETCH_MTD_##method] = FETCH_FUNC_NAME(method, type) + +#define __ASSIGN_FETCH_TYPE(_name, ptype, ftype, _size, sign, _fmttype) \ + {.name = _name, \ + .size = _size, \ + .is_signed = sign, \ + .print = PRINT_TYPE_FUNC_NAME(ptype), \ + .fmt = PRINT_TYPE_FMT_NAME(ptype), \ + .fmttype = _fmttype, \ + .fetch = { \ +ASSIGN_FETCH_FUNC(reg, ftype), \ +ASSIGN_FETCH_FUNC(stack, ftype), \ +ASSIGN_FETCH_FUNC(retval, ftype), \ +ASSIGN_FETCH_FUNC(memory, ftype), \ +ASSIGN_FETCH_FUNC(symbol, ftype), \ +ASSIGN_FETCH_FUNC(deref, ftype), \ + } \ } +#define ASSIGN_FETCH_TYPE(ptype, ftype, sign) \ + __ASSIGN_FETCH_TYPE(#ptype, ptype, ftype, sizeof(ftype), sign, #ptype) + +#define FETCH_TYPE_STRING 0 +#define FETCH_TYPE_STRSIZE 1 + /* Fetch type information table */ static const struct fetch_type { const char *name; /* Name of type */ @@ -264,14 +403,16 @@ static const struct fetch_type { int is_signed; /* Signed flag */ print_type_func_t print; /* Print functions */ const char *fmt; /* Fromat string */ + const char *fmttype; /* Name in format file */ /* Fetch functions */ - fetch_func_t reg; - fetch_func_t stack; - fetch_func_t retval; - fetch_func_t memory; - fetch_func_t symbol; - fetch_func_t deref; + fetch_func_t fetch[FETCH_MTD_END]; } fetch_type_table[] = { + /* Special types */ + [FETCH_TYPE_STRING] = __ASSIGN_FETCH_TYPE("string", string, string, + sizeof(u32), 1, "__data_loc char[]"), + [FETCH_TYPE_STRSIZE] = __ASSIGN_FETCH_TYPE("string_size", u32, + string_size, sizeof(u32), 0, "u32"), + /* Basic types */ ASSIGN_FETCH_TYPE(u8, u8, 0), ASSIGN_FETCH_TYPE(u16, u16, 0), ASSIGN_FETCH_TYPE(u32, u32, 0), @@ -302,12 +443,28 @@ static __kprobes void fetch_stack_address(struct pt_regs *regs, *(unsigned long *)dest = kernel_stack_pointer(regs); } +static fetch_func_t get_fetch_size_function(const struct fetch_type *type, + fetch_func_t orig_fn) +{ + int i; + + if (type != &fetch_type_table[FETCH_TYPE_STRING]) + return NULL; /* Only string type needs size function */ + for (i = 0; i < FETCH_MTD_END; i++) + if (type->fetch[i] == orig_fn) + return fetch_type_table[FETCH_TYPE_STRSIZE].fetch[i]; + + WARN_ON(1); /* This should not happen */ + return NULL; +} + /** * Kprobe event core functions */ struct probe_arg { struct fetch_param fetch; + struct fetch_param fetch_size; unsigned int offset; /* Offset from argument entry */ const char *name; /* Name of this argument */ const char *comm; /* Command of this argument */ @@ -324,8 +481,8 @@ struct trace_probe { unsigned long nhit; unsigned int flags; /* For TP_FLAG_* */ const char *symbol; /* symbol name */ + struct ftrace_event_class class; struct ftrace_event_call call; - struct trace_event event; ssize_t size; /* trace entry size */ unsigned int nr_args; struct probe_arg args[]; @@ -356,8 +513,8 @@ static int kprobe_dispatcher(struct kprobe *kp, struct pt_regs *regs); static int kretprobe_dispatcher(struct kretprobe_instance *ri, struct pt_regs *regs); -/* Check the name is good for event/group */ -static int check_event_name(const char *name) +/* Check the name is good for event/group/fields */ +static int is_good_name(const char *name) { if (!isalpha(*name) && *name != '_') return 0; @@ -399,22 +556,23 @@ static struct trace_probe *alloc_trace_probe(const char *group, else tp->rp.kp.pre_handler = kprobe_dispatcher; - if (!event || !check_event_name(event)) { + if (!event || !is_good_name(event)) { ret = -EINVAL; goto error; } + tp->call.class = &tp->class; tp->call.name = kstrdup(event, GFP_KERNEL); if (!tp->call.name) goto error; - if (!group || !check_event_name(group)) { + if (!group || !is_good_name(group)) { ret = -EINVAL; goto error; } - tp->call.system = kstrdup(group, GFP_KERNEL); - if (!tp->call.system) + tp->class.system = kstrdup(group, GFP_KERNEL); + if (!tp->class.system) goto error; INIT_LIST_HEAD(&tp->list); @@ -428,9 +586,9 @@ error: static void free_probe_arg(struct probe_arg *arg) { - if (CHECK_BASIC_FETCH_FUNCS(deref, arg->fetch.fn)) + if (CHECK_FETCH_FUNCS(deref, arg->fetch.fn)) free_deref_fetch_param(arg->fetch.data); - else if (CHECK_BASIC_FETCH_FUNCS(symbol, arg->fetch.fn)) + else if (CHECK_FETCH_FUNCS(symbol, arg->fetch.fn)) free_symbol_cache(arg->fetch.data); kfree(arg->name); kfree(arg->comm); @@ -443,7 +601,7 @@ static void free_trace_probe(struct trace_probe *tp) for (i = 0; i < tp->nr_args; i++) free_probe_arg(&tp->args[i]); - kfree(tp->call.system); + kfree(tp->call.class->system); kfree(tp->call.name); kfree(tp->symbol); kfree(tp); @@ -456,7 +614,7 @@ static struct trace_probe *find_probe_event(const char *event, list_for_each_entry(tp, &probe_list, list) if (strcmp(tp->call.name, event) == 0 && - strcmp(tp->call.system, group) == 0) + strcmp(tp->call.class->system, group) == 0) return tp; return NULL; } @@ -481,7 +639,7 @@ static int register_trace_probe(struct trace_probe *tp) mutex_lock(&probe_lock); /* register as an event */ - old_tp = find_probe_event(tp->call.name, tp->call.system); + old_tp = find_probe_event(tp->call.name, tp->call.class->system); if (old_tp) { /* delete old event */ unregister_trace_probe(old_tp); @@ -489,7 +647,7 @@ static int register_trace_probe(struct trace_probe *tp) } ret = register_probe_event(tp); if (ret) { - pr_warning("Faild to register probe event(%d)\n", ret); + pr_warning("Failed to register probe event(%d)\n", ret); goto end; } @@ -547,7 +705,7 @@ static int parse_probe_vars(char *arg, const struct fetch_type *t, if (strcmp(arg, "retval") == 0) { if (is_return) - f->fn = t->retval; + f->fn = t->fetch[FETCH_MTD_retval]; else ret = -EINVAL; } else if (strncmp(arg, "stack", 5) == 0) { @@ -561,7 +719,7 @@ static int parse_probe_vars(char *arg, const struct fetch_type *t, if (ret || param > PARAM_MAX_STACK) ret = -EINVAL; else { - f->fn = t->stack; + f->fn = t->fetch[FETCH_MTD_stack]; f->data = (void *)param; } } else @@ -587,7 +745,7 @@ static int __parse_probe_arg(char *arg, const struct fetch_type *t, case '%': /* named register */ ret = regs_query_register_offset(arg + 1); if (ret >= 0) { - f->fn = t->reg; + f->fn = t->fetch[FETCH_MTD_reg]; f->data = (void *)(unsigned long)ret; ret = 0; } @@ -597,7 +755,7 @@ static int __parse_probe_arg(char *arg, const struct fetch_type *t, ret = strict_strtoul(arg + 1, 0, ¶m); if (ret) break; - f->fn = t->memory; + f->fn = t->fetch[FETCH_MTD_memory]; f->data = (void *)param; } else { ret = split_symbol_offset(arg + 1, &offset); @@ -605,7 +763,7 @@ static int __parse_probe_arg(char *arg, const struct fetch_type *t, break; f->data = alloc_symbol_cache(arg + 1, offset); if (f->data) - f->fn = t->symbol; + f->fn = t->fetch[FETCH_MTD_symbol]; } break; case '+': /* deref memory */ @@ -635,14 +793,17 @@ static int __parse_probe_arg(char *arg, const struct fetch_type *t, if (ret) kfree(dprm); else { - f->fn = t->deref; + f->fn = t->fetch[FETCH_MTD_deref]; f->data = (void *)dprm; } } break; } - if (!ret && !f->fn) + if (!ret && !f->fn) { /* Parsed, but do not find fetch method */ + pr_info("%s type has no corresponding fetch method.\n", + t->name); ret = -EINVAL; + } return ret; } @@ -651,6 +812,7 @@ static int parse_probe_arg(char *arg, struct trace_probe *tp, struct probe_arg *parg, int is_return) { const char *t; + int ret; if (strlen(arg) > MAX_ARGSTR_LEN) { pr_info("Argument is too long.: %s\n", arg); @@ -673,7 +835,13 @@ static int parse_probe_arg(char *arg, struct trace_probe *tp, } parg->offset = tp->size; tp->size += parg->type->size; - return __parse_probe_arg(arg, parg->type, &parg->fetch, is_return); + ret = __parse_probe_arg(arg, parg->type, &parg->fetch, is_return); + if (ret >= 0) { + parg->fetch_size.fn = get_fetch_size_function(parg->type, + parg->fetch.fn); + parg->fetch_size.data = parg->fetch.data; + } + return ret; } /* Return 1 if name is reserved or already used by another argument */ @@ -714,7 +882,7 @@ static int create_trace_probe(int argc, char **argv) int i, ret = 0; int is_return = 0, is_delete = 0; char *symbol = NULL, *event = NULL, *group = NULL; - char *arg, *tmp; + char *arg; unsigned long offset = 0; void *addr = NULL; char buf[MAX_EVENT_NAME_LEN]; @@ -756,14 +924,17 @@ static int create_trace_probe(int argc, char **argv) pr_info("Delete command needs an event name.\n"); return -EINVAL; } + mutex_lock(&probe_lock); tp = find_probe_event(event, group); if (!tp) { + mutex_unlock(&probe_lock); pr_info("Event %s/%s doesn't exist.\n", group, event); return -ENOENT; } /* delete an event */ unregister_trace_probe(tp); free_trace_probe(tp); + mutex_unlock(&probe_lock); return 0; } @@ -820,26 +991,36 @@ static int create_trace_probe(int argc, char **argv) /* parse arguments */ ret = 0; for (i = 0; i < argc && i < MAX_TRACE_ARGS; i++) { + /* Increment count for freeing args in error case */ + tp->nr_args++; + /* Parse argument name */ arg = strchr(argv[i], '='); - if (arg) + if (arg) { *arg++ = '\0'; - else + tp->args[i].name = kstrdup(argv[i], GFP_KERNEL); + } else { arg = argv[i]; + /* If argument name is omitted, set "argN" */ + snprintf(buf, MAX_EVENT_NAME_LEN, "arg%d", i + 1); + tp->args[i].name = kstrdup(buf, GFP_KERNEL); + } - 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]); + pr_info("Failed to allocate argument[%d] name.\n", i); ret = -ENOMEM; goto error; } - tmp = strchr(tp->args[i].name, ':'); - if (tmp) - *tmp = '_'; /* convert : to _ */ + + if (!is_good_name(tp->args[i].name)) { + pr_info("Invalid argument[%d] name: %s\n", + i, tp->args[i].name); + ret = -EINVAL; + goto error; + } if (conflict_field_name(tp->args[i].name, tp->args, i)) { - pr_info("Argument%d name '%s' conflicts with " + pr_info("Argument[%d] name '%s' conflicts with " "another field.\n", i, argv[i]); ret = -EINVAL; goto error; @@ -848,12 +1029,9 @@ static int create_trace_probe(int argc, char **argv) /* Parse fetch argument */ ret = parse_probe_arg(arg, tp, &tp->args[i], is_return); if (ret) { - pr_info("Parse error at argument%d. (%d)\n", i, ret); - kfree(tp->args[i].name); + pr_info("Parse error at argument[%d]. (%d)\n", i, ret); goto error; } - - tp->nr_args++; } ret = register_trace_probe(tp); @@ -904,7 +1082,7 @@ static int probes_seq_show(struct seq_file *m, void *v) int i; seq_printf(m, "%c", probe_is_return(tp) ? 'r' : 'p'); - seq_printf(m, ":%s/%s", tp->call.system, tp->call.name); + seq_printf(m, ":%s/%s", tp->call.class->system, tp->call.name); if (!tp->symbol) seq_printf(m, " 0x%p", tp->rp.kp.addr); @@ -1042,6 +1220,54 @@ static const struct file_operations kprobe_profile_ops = { .release = seq_release, }; +/* Sum up total data length for dynamic arraies (strings) */ +static __kprobes int __get_data_size(struct trace_probe *tp, + struct pt_regs *regs) +{ + int i, ret = 0; + u32 len; + + for (i = 0; i < tp->nr_args; i++) + if (unlikely(tp->args[i].fetch_size.fn)) { + call_fetch(&tp->args[i].fetch_size, regs, &len); + ret += len; + } + + return ret; +} + +/* Store the value of each argument */ +static __kprobes void store_trace_args(int ent_size, struct trace_probe *tp, + struct pt_regs *regs, + u8 *data, int maxlen) +{ + int i; + u32 end = tp->size; + u32 *dl; /* Data (relative) location */ + + for (i = 0; i < tp->nr_args; i++) { + if (unlikely(tp->args[i].fetch_size.fn)) { + /* + * First, we set the relative location and + * maximum data length to *dl + */ + dl = (u32 *)(data + tp->args[i].offset); + *dl = make_data_rloc(maxlen, end - tp->args[i].offset); + /* Then try to fetch string or dynamic array data */ + call_fetch(&tp->args[i].fetch, regs, dl); + /* Reduce maximum length */ + end += get_rloc_len(*dl); + maxlen -= get_rloc_len(*dl); + /* Trick here, convert data_rloc to data_loc */ + *dl = convert_rloc_to_loc(*dl, + ent_size + tp->args[i].offset); + } else + /* Just fetching data normally */ + call_fetch(&tp->args[i].fetch, regs, + data + tp->args[i].offset); + } +} + /* Kprobe handler */ static __kprobes void kprobe_trace_func(struct kprobe *kp, struct pt_regs *regs) { @@ -1049,8 +1275,7 @@ static __kprobes void kprobe_trace_func(struct kprobe *kp, struct pt_regs *regs) struct kprobe_trace_entry_head *entry; struct ring_buffer_event *event; struct ring_buffer *buffer; - u8 *data; - int size, i, pc; + int size, dsize, pc; unsigned long irq_flags; struct ftrace_event_call *call = &tp->call; @@ -1059,18 +1284,17 @@ static __kprobes void kprobe_trace_func(struct kprobe *kp, struct pt_regs *regs) local_save_flags(irq_flags); pc = preempt_count(); - size = sizeof(*entry) + tp->size; + dsize = __get_data_size(tp, regs); + size = sizeof(*entry) + tp->size + dsize; - event = trace_current_buffer_lock_reserve(&buffer, call->id, size, - irq_flags, pc); + event = trace_current_buffer_lock_reserve(&buffer, call->event.type, + size, irq_flags, pc); if (!event) return; entry = ring_buffer_event_data(event); entry->ip = (unsigned long)kp->addr; - data = (u8 *)&entry[1]; - for (i = 0; i < tp->nr_args; i++) - call_fetch(&tp->args[i].fetch, regs, data + tp->args[i].offset); + store_trace_args(sizeof(*entry), tp, regs, (u8 *)&entry[1], dsize); if (!filter_current_check_discard(buffer, call, entry, event)) trace_nowake_buffer_unlock_commit(buffer, event, irq_flags, pc); @@ -1084,27 +1308,25 @@ static __kprobes void kretprobe_trace_func(struct kretprobe_instance *ri, struct kretprobe_trace_entry_head *entry; struct ring_buffer_event *event; struct ring_buffer *buffer; - u8 *data; - int size, i, pc; + int size, pc, dsize; unsigned long irq_flags; struct ftrace_event_call *call = &tp->call; local_save_flags(irq_flags); pc = preempt_count(); - size = sizeof(*entry) + tp->size; + dsize = __get_data_size(tp, regs); + size = sizeof(*entry) + tp->size + dsize; - event = trace_current_buffer_lock_reserve(&buffer, call->id, size, - irq_flags, pc); + event = trace_current_buffer_lock_reserve(&buffer, call->event.type, + size, irq_flags, pc); if (!event) return; entry = ring_buffer_event_data(event); entry->func = (unsigned long)tp->rp.kp.addr; entry->ret_ip = (unsigned long)ri->ret_addr; - data = (u8 *)&entry[1]; - for (i = 0; i < tp->nr_args; i++) - call_fetch(&tp->args[i].fetch, regs, data + tp->args[i].offset); + store_trace_args(sizeof(*entry), tp, regs, (u8 *)&entry[1], dsize); if (!filter_current_check_discard(buffer, call, entry, event)) trace_nowake_buffer_unlock_commit(buffer, event, irq_flags, pc); @@ -1112,18 +1334,17 @@ static __kprobes void kretprobe_trace_func(struct kretprobe_instance *ri, /* Event entry printers */ enum print_line_t -print_kprobe_event(struct trace_iterator *iter, int flags) +print_kprobe_event(struct trace_iterator *iter, int flags, + struct trace_event *event) { struct kprobe_trace_entry_head *field; struct trace_seq *s = &iter->seq; - struct trace_event *event; struct trace_probe *tp; u8 *data; int i; field = (struct kprobe_trace_entry_head *)iter->ent; - event = ftrace_find_event(field->ent.type); - tp = container_of(event, struct trace_probe, event); + tp = container_of(event, struct trace_probe, call.event); if (!trace_seq_printf(s, "%s: (", tp->call.name)) goto partial; @@ -1137,7 +1358,7 @@ print_kprobe_event(struct trace_iterator *iter, int flags) data = (u8 *)&field[1]; for (i = 0; i < tp->nr_args; i++) if (!tp->args[i].type->print(s, tp->args[i].name, - data + tp->args[i].offset)) + data + tp->args[i].offset, field)) goto partial; if (!trace_seq_puts(s, "\n")) @@ -1149,18 +1370,17 @@ partial: } enum print_line_t -print_kretprobe_event(struct trace_iterator *iter, int flags) +print_kretprobe_event(struct trace_iterator *iter, int flags, + struct trace_event *event) { struct kretprobe_trace_entry_head *field; struct trace_seq *s = &iter->seq; - struct trace_event *event; struct trace_probe *tp; u8 *data; int i; field = (struct kretprobe_trace_entry_head *)iter->ent; - event = ftrace_find_event(field->ent.type); - tp = container_of(event, struct trace_probe, event); + tp = container_of(event, struct trace_probe, call.event); if (!trace_seq_printf(s, "%s: (", tp->call.name)) goto partial; @@ -1180,7 +1400,7 @@ print_kretprobe_event(struct trace_iterator *iter, int flags) data = (u8 *)&field[1]; for (i = 0; i < tp->nr_args; i++) if (!tp->args[i].type->print(s, tp->args[i].name, - data + tp->args[i].offset)) + data + tp->args[i].offset, field)) goto partial; if (!trace_seq_puts(s, "\n")) @@ -1215,13 +1435,6 @@ static void probe_event_disable(struct ftrace_event_call *call) } } -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 { \ @@ -1242,7 +1455,7 @@ static int kprobe_event_define_fields(struct ftrace_event_call *event_call) DEFINE_FIELD(unsigned long, ip, FIELD_STRING_IP, 0); /* Set argument names as fields */ for (i = 0; i < tp->nr_args; i++) { - ret = trace_define_field(event_call, tp->args[i].type->name, + ret = trace_define_field(event_call, tp->args[i].type->fmttype, tp->args[i].name, sizeof(field) + tp->args[i].offset, tp->args[i].type->size, @@ -1264,7 +1477,7 @@ static int kretprobe_event_define_fields(struct ftrace_event_call *event_call) DEFINE_FIELD(unsigned long, ret_ip, FIELD_STRING_RETIP, 0); /* Set argument names as fields */ for (i = 0; i < tp->nr_args; i++) { - ret = trace_define_field(event_call, tp->args[i].type->name, + ret = trace_define_field(event_call, tp->args[i].type->fmttype, tp->args[i].name, sizeof(field) + tp->args[i].offset, tp->args[i].type->size, @@ -1304,8 +1517,13 @@ static int __set_print_fmt(struct trace_probe *tp, char *buf, int len) pos += snprintf(buf + pos, LEN_OR_ZERO, "\", %s", arg); for (i = 0; i < tp->nr_args; i++) { - pos += snprintf(buf + pos, LEN_OR_ZERO, ", REC->%s", - tp->args[i].name); + if (strcmp(tp->args[i].type->name, "string") == 0) + pos += snprintf(buf + pos, LEN_OR_ZERO, + ", __get_str(%s)", + tp->args[i].name); + else + pos += snprintf(buf + pos, LEN_OR_ZERO, ", REC->%s", + tp->args[i].name); } #undef LEN_OR_ZERO @@ -1341,28 +1559,28 @@ static __kprobes void kprobe_perf_func(struct kprobe *kp, struct trace_probe *tp = container_of(kp, struct trace_probe, rp.kp); struct ftrace_event_call *call = &tp->call; struct kprobe_trace_entry_head *entry; - u8 *data; - int size, __size, i; - unsigned long irq_flags; + struct hlist_head *head; + int size, __size, dsize; int rctx; - __size = sizeof(*entry) + tp->size; + dsize = __get_data_size(tp, regs); + __size = sizeof(*entry) + tp->size + dsize; size = ALIGN(__size + sizeof(u32), sizeof(u64)); size -= sizeof(u32); if (WARN_ONCE(size > PERF_MAX_TRACE_SIZE, "profile buffer not large enough")) return; - entry = perf_trace_buf_prepare(size, call->id, &rctx, &irq_flags); + entry = perf_trace_buf_prepare(size, call->event.type, regs, &rctx); if (!entry) return; entry->ip = (unsigned long)kp->addr; - data = (u8 *)&entry[1]; - for (i = 0; i < tp->nr_args; i++) - call_fetch(&tp->args[i].fetch, regs, data + tp->args[i].offset); + memset(&entry[1], 0, dsize); + store_trace_args(sizeof(*entry), tp, regs, (u8 *)&entry[1], dsize); - perf_trace_buf_submit(entry, size, rctx, entry->ip, 1, irq_flags, regs); + head = this_cpu_ptr(call->perf_events); + perf_trace_buf_submit(entry, size, rctx, entry->ip, 1, regs, head); } /* Kretprobe profile handler */ @@ -1372,30 +1590,28 @@ static __kprobes void kretprobe_perf_func(struct kretprobe_instance *ri, struct trace_probe *tp = container_of(ri->rp, struct trace_probe, rp); struct ftrace_event_call *call = &tp->call; struct kretprobe_trace_entry_head *entry; - u8 *data; - int size, __size, i; - unsigned long irq_flags; + struct hlist_head *head; + int size, __size, dsize; int rctx; - __size = sizeof(*entry) + tp->size; + dsize = __get_data_size(tp, regs); + __size = sizeof(*entry) + tp->size + dsize; size = ALIGN(__size + sizeof(u32), sizeof(u64)); size -= sizeof(u32); if (WARN_ONCE(size > PERF_MAX_TRACE_SIZE, "profile buffer not large enough")) return; - entry = perf_trace_buf_prepare(size, call->id, &rctx, &irq_flags); + entry = perf_trace_buf_prepare(size, call->event.type, regs, &rctx); if (!entry) return; entry->func = (unsigned long)tp->rp.kp.addr; entry->ret_ip = (unsigned long)ri->ret_addr; - data = (u8 *)&entry[1]; - for (i = 0; i < tp->nr_args; i++) - call_fetch(&tp->args[i].fetch, regs, data + tp->args[i].offset); + store_trace_args(sizeof(*entry), tp, regs, (u8 *)&entry[1], dsize); - perf_trace_buf_submit(entry, size, rctx, entry->ret_ip, 1, - irq_flags, regs); + head = this_cpu_ptr(call->perf_events); + perf_trace_buf_submit(entry, size, rctx, entry->ret_ip, 1, regs, head); } static int probe_perf_enable(struct ftrace_event_call *call) @@ -1425,6 +1641,26 @@ static void probe_perf_disable(struct ftrace_event_call *call) } #endif /* CONFIG_PERF_EVENTS */ +static __kprobes +int kprobe_register(struct ftrace_event_call *event, enum trace_reg type) +{ + switch (type) { + case TRACE_REG_REGISTER: + return probe_event_enable(event); + case TRACE_REG_UNREGISTER: + probe_event_disable(event); + return 0; + +#ifdef CONFIG_PERF_EVENTS + case TRACE_REG_PERF_REGISTER: + return probe_perf_enable(event); + case TRACE_REG_PERF_UNREGISTER: + probe_perf_disable(event); + return 0; +#endif + } + return 0; +} static __kprobes int kprobe_dispatcher(struct kprobe *kp, struct pt_regs *regs) @@ -1454,43 +1690,43 @@ int kretprobe_dispatcher(struct kretprobe_instance *ri, struct pt_regs *regs) return 0; /* We don't tweek kernel, so just return 0 */ } +static struct trace_event_functions kretprobe_funcs = { + .trace = print_kretprobe_event +}; + +static struct trace_event_functions kprobe_funcs = { + .trace = print_kprobe_event +}; + static int register_probe_event(struct trace_probe *tp) { struct ftrace_event_call *call = &tp->call; int ret; /* Initialize ftrace_event_call */ + INIT_LIST_HEAD(&call->class->fields); if (probe_is_return(tp)) { - tp->event.trace = print_kretprobe_event; - call->raw_init = probe_event_raw_init; - call->define_fields = kretprobe_event_define_fields; + call->event.funcs = &kretprobe_funcs; + call->class->define_fields = kretprobe_event_define_fields; } else { - tp->event.trace = print_kprobe_event; - call->raw_init = probe_event_raw_init; - call->define_fields = kprobe_event_define_fields; + call->event.funcs = &kprobe_funcs; + call->class->define_fields = kprobe_event_define_fields; } if (set_print_fmt(tp) < 0) return -ENOMEM; - call->event = &tp->event; - call->id = register_ftrace_event(&tp->event); - if (!call->id) { + ret = register_ftrace_event(&call->event); + if (!ret) { kfree(call->print_fmt); return -ENODEV; } - call->enabled = 0; - call->regfunc = probe_event_enable; - call->unregfunc = probe_event_disable; - -#ifdef CONFIG_PERF_EVENTS - call->perf_event_enable = probe_perf_enable; - call->perf_event_disable = probe_perf_disable; -#endif + call->flags = 0; + call->class->reg = kprobe_register; call->data = tp; ret = trace_add_event_call(call); if (ret) { pr_info("Failed to register kprobe event: %s\n", call->name); kfree(call->print_fmt); - unregister_ftrace_event(&tp->event); + unregister_ftrace_event(&call->event); } return ret; } diff --git a/kernel/trace/trace_ksym.c b/kernel/trace/trace_ksym.c deleted file mode 100644 index 8eaf00749b6..00000000000 --- a/kernel/trace/trace_ksym.c +++ /dev/null @@ -1,508 +0,0 @@ -/* - * 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/slab.h> -#include <linux/fs.h> - -#include "trace_output.h" -#include "trace.h" - -#include <linux/hw_breakpoint.h> -#include <asm/hw_breakpoint.h> - -#include <asm/atomic.h> - -#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 - atomic64_t counter; -#endif - struct hlist_node ksym_hlist; -}; - -static struct trace_array *ksym_trace_array; - -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) { - atomic64_inc(&entry->counter); - break; - } - } - rcu_read_unlock(); -} -#endif /* CONFIG_PROFILE_KSYM_TRACER */ - -void ksym_hbp_handler(struct perf_event *hbp, int nmi, - struct perf_sample_data *data, - struct pt_regs *regs) -{ - struct ring_buffer_event *event; - struct ksym_trace_entry *entry; - 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; - - 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; - - entry->ksym_hbp = register_wide_hw_breakpoint(&entry->attr, - ksym_hbp_handler); - - if (IS_ERR(entry->ksym_hbp)) { - ret = PTR_ERR(entry->ksym_hbp); - if (ret == -ENOSPC) { - printk(KERN_ERR "ksym_tracer: Maximum limit reached." - " No new requests for tracing can be accepted now.\n"); - } else { - printk(KERN_INFO "ksym_tracer request failed. Try again" - " later!!\n"); - } - goto err; - } - - hlist_add_head_rcu(&(entry->ksym_hlist), &ksym_filter_head); - - 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 *)(unsigned long)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); - 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 *buf, *input_string, *ksymname = NULL; - unsigned long ksym_addr = 0; - int ret, op, changed = 0; - - buf = kzalloc(count + 1, GFP_KERNEL); - if (!buf) - return -ENOMEM; - - ret = -EFAULT; - if (copy_from_user(buf, buffer, count)) - goto out; - - buf[count] = '\0'; - input_string = strstrip(buf); - - /* - * 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(); - ret = 0; - goto out; - } - - ret = op = parse_ksym_trace_str(input_string, &ksymname, &ksym_addr); - if (ret < 0) - goto out; - - 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_unlock; - 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_unlock; - } - /* Error or "symbol:---" case: drop it */ - hlist_del_rcu(&(entry->ksym_hlist)); - synchronize_rcu(); - kfree(entry); - goto out_unlock; - } else { - /* Check for malformed request: (4) */ - if (op) - ret = process_new_ksym_entry(ksymname, op, ksym_addr); - } -out_unlock: - mutex_unlock(&ksym_tracer_mutex); -out: - kfree(buf); - return !ret ? count : 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 -}; - -#ifdef CONFIG_PROFILE_KSYM_TRACER -static int ksym_profile_show(struct seq_file *m, void *v) -{ - struct hlist_node *node; - struct trace_ksym *entry; - int access_type = 0; - char fn_name[KSYM_NAME_LEN]; - - seq_puts(m, " Access Type "); - seq_puts(m, " Symbol Counter\n"); - seq_puts(m, " ----------- "); - seq_puts(m, " ------ -------\n"); - - rcu_read_lock(); - hlist_for_each_entry_rcu(entry, node, &ksym_filter_head, 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, " %15llu\n", - (unsigned long long)atomic64_read(&entry->counter)); - } - rcu_read_unlock(); - - return 0; -} - -static int ksym_profile_open(struct inode *node, struct file *file) -{ - return single_open(file, ksym_profile_show, NULL); -} - -static const struct file_operations ksym_profile_fops = { - .open = ksym_profile_open, - .read = seq_read, - .llseek = seq_lseek, - .release = single_release, -}; -#endif /* CONFIG_PROFILE_KSYM_TRACER */ - -__init static int init_ksym_trace(void) -{ - struct dentry *d_tracer; - - d_tracer = tracing_init_dentry(); - - trace_create_file("ksym_trace_filter", 0644, d_tracer, - NULL, &ksym_tracing_fops); - -#ifdef CONFIG_PROFILE_KSYM_TRACER - trace_create_file("ksym_profile", 0444, d_tracer, - NULL, &ksym_profile_fops); -#endif - - return register_tracer(&ksym_tracer); -} -device_initcall(init_ksym_trace); diff --git a/kernel/trace/trace_output.c b/kernel/trace/trace_output.c index ab13d700806..02272baa220 100644 --- a/kernel/trace/trace_output.c +++ b/kernel/trace/trace_output.c @@ -16,9 +16,6 @@ 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; @@ -742,6 +739,9 @@ int register_ftrace_event(struct trace_event *event) if (WARN_ON(!event)) goto out; + if (WARN_ON(!event->funcs)) + goto out; + INIT_LIST_HEAD(&event->list); if (!event->type) { @@ -774,14 +774,14 @@ int register_ftrace_event(struct trace_event *event) 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; + if (event->funcs->trace == NULL) + event->funcs->trace = trace_nop_print; + if (event->funcs->raw == NULL) + event->funcs->raw = trace_nop_print; + if (event->funcs->hex == NULL) + event->funcs->hex = trace_nop_print; + if (event->funcs->binary == NULL) + event->funcs->binary = trace_nop_print; key = event->type & (EVENT_HASHSIZE - 1); @@ -823,13 +823,15 @@ EXPORT_SYMBOL_GPL(unregister_ftrace_event); * Standard events */ -enum print_line_t trace_nop_print(struct trace_iterator *iter, int flags) +enum print_line_t trace_nop_print(struct trace_iterator *iter, int flags, + struct trace_event *event) { return TRACE_TYPE_HANDLED; } /* TRACE_FN */ -static enum print_line_t trace_fn_trace(struct trace_iterator *iter, int flags) +static enum print_line_t trace_fn_trace(struct trace_iterator *iter, int flags, + struct trace_event *event) { struct ftrace_entry *field; struct trace_seq *s = &iter->seq; @@ -856,7 +858,8 @@ static enum print_line_t trace_fn_trace(struct trace_iterator *iter, int flags) return TRACE_TYPE_PARTIAL_LINE; } -static enum print_line_t trace_fn_raw(struct trace_iterator *iter, int flags) +static enum print_line_t trace_fn_raw(struct trace_iterator *iter, int flags, + struct trace_event *event) { struct ftrace_entry *field; @@ -870,7 +873,8 @@ static enum print_line_t trace_fn_raw(struct trace_iterator *iter, int flags) return TRACE_TYPE_HANDLED; } -static enum print_line_t trace_fn_hex(struct trace_iterator *iter, int flags) +static enum print_line_t trace_fn_hex(struct trace_iterator *iter, int flags, + struct trace_event *event) { struct ftrace_entry *field; struct trace_seq *s = &iter->seq; @@ -883,7 +887,8 @@ static enum print_line_t trace_fn_hex(struct trace_iterator *iter, int flags) return TRACE_TYPE_HANDLED; } -static enum print_line_t trace_fn_bin(struct trace_iterator *iter, int flags) +static enum print_line_t trace_fn_bin(struct trace_iterator *iter, int flags, + struct trace_event *event) { struct ftrace_entry *field; struct trace_seq *s = &iter->seq; @@ -896,14 +901,18 @@ static enum print_line_t trace_fn_bin(struct trace_iterator *iter, int flags) return TRACE_TYPE_HANDLED; } -static struct trace_event trace_fn_event = { - .type = TRACE_FN, +static struct trace_event_functions trace_fn_funcs = { .trace = trace_fn_trace, .raw = trace_fn_raw, .hex = trace_fn_hex, .binary = trace_fn_bin, }; +static struct trace_event trace_fn_event = { + .type = TRACE_FN, + .funcs = &trace_fn_funcs, +}; + /* TRACE_CTX an TRACE_WAKE */ static enum print_line_t trace_ctxwake_print(struct trace_iterator *iter, char *delim) @@ -932,13 +941,14 @@ static enum print_line_t trace_ctxwake_print(struct trace_iterator *iter, return TRACE_TYPE_HANDLED; } -static enum print_line_t trace_ctx_print(struct trace_iterator *iter, int flags) +static enum print_line_t trace_ctx_print(struct trace_iterator *iter, int flags, + struct trace_event *event) { return trace_ctxwake_print(iter, "==>"); } static enum print_line_t trace_wake_print(struct trace_iterator *iter, - int flags) + int flags, struct trace_event *event) { return trace_ctxwake_print(iter, " +"); } @@ -966,12 +976,14 @@ static int trace_ctxwake_raw(struct trace_iterator *iter, char S) return TRACE_TYPE_HANDLED; } -static enum print_line_t trace_ctx_raw(struct trace_iterator *iter, int flags) +static enum print_line_t trace_ctx_raw(struct trace_iterator *iter, int flags, + struct trace_event *event) { return trace_ctxwake_raw(iter, 0); } -static enum print_line_t trace_wake_raw(struct trace_iterator *iter, int flags) +static enum print_line_t trace_wake_raw(struct trace_iterator *iter, int flags, + struct trace_event *event) { return trace_ctxwake_raw(iter, '+'); } @@ -1000,18 +1012,20 @@ static int trace_ctxwake_hex(struct trace_iterator *iter, char S) return TRACE_TYPE_HANDLED; } -static enum print_line_t trace_ctx_hex(struct trace_iterator *iter, int flags) +static enum print_line_t trace_ctx_hex(struct trace_iterator *iter, int flags, + struct trace_event *event) { return trace_ctxwake_hex(iter, 0); } -static enum print_line_t trace_wake_hex(struct trace_iterator *iter, int flags) +static enum print_line_t trace_wake_hex(struct trace_iterator *iter, int flags, + struct trace_event *event) { return trace_ctxwake_hex(iter, '+'); } static enum print_line_t trace_ctxwake_bin(struct trace_iterator *iter, - int flags) + int flags, struct trace_event *event) { struct ctx_switch_entry *field; struct trace_seq *s = &iter->seq; @@ -1028,81 +1042,34 @@ static enum print_line_t trace_ctxwake_bin(struct trace_iterator *iter, return TRACE_TYPE_HANDLED; } -static struct trace_event trace_ctx_event = { - .type = TRACE_CTX, +static struct trace_event_functions trace_ctx_funcs = { .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, +static struct trace_event trace_ctx_event = { + .type = TRACE_CTX, + .funcs = &trace_ctx_funcs, +}; + +static struct trace_event_functions trace_wake_funcs = { .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, +static struct trace_event trace_wake_event = { + .type = TRACE_WAKE, + .funcs = &trace_wake_funcs, }; /* TRACE_STACK */ static enum print_line_t trace_stack_print(struct trace_iterator *iter, - int flags) + int flags, struct trace_event *event) { struct stack_entry *field; struct trace_seq *s = &iter->seq; @@ -1130,17 +1097,18 @@ static enum print_line_t trace_stack_print(struct trace_iterator *iter, return TRACE_TYPE_PARTIAL_LINE; } +static struct trace_event_functions trace_stack_funcs = { + .trace = trace_stack_print, +}; + 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, + .funcs = &trace_stack_funcs, }; /* TRACE_USER_STACK */ static enum print_line_t trace_user_stack_print(struct trace_iterator *iter, - int flags) + int flags, struct trace_event *event) { struct userstack_entry *field; struct trace_seq *s = &iter->seq; @@ -1159,17 +1127,19 @@ static enum print_line_t trace_user_stack_print(struct trace_iterator *iter, return TRACE_TYPE_PARTIAL_LINE; } +static struct trace_event_functions trace_user_stack_funcs = { + .trace = trace_user_stack_print, +}; + 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, + .funcs = &trace_user_stack_funcs, }; /* TRACE_BPRINT */ static enum print_line_t -trace_bprint_print(struct trace_iterator *iter, int flags) +trace_bprint_print(struct trace_iterator *iter, int flags, + struct trace_event *event) { struct trace_entry *entry = iter->ent; struct trace_seq *s = &iter->seq; @@ -1194,7 +1164,8 @@ trace_bprint_print(struct trace_iterator *iter, int flags) static enum print_line_t -trace_bprint_raw(struct trace_iterator *iter, int flags) +trace_bprint_raw(struct trace_iterator *iter, int flags, + struct trace_event *event) { struct bprint_entry *field; struct trace_seq *s = &iter->seq; @@ -1213,16 +1184,19 @@ trace_bprint_raw(struct trace_iterator *iter, int flags) return TRACE_TYPE_PARTIAL_LINE; } +static struct trace_event_functions trace_bprint_funcs = { + .trace = trace_bprint_print, + .raw = trace_bprint_raw, +}; static struct trace_event trace_bprint_event = { .type = TRACE_BPRINT, - .trace = trace_bprint_print, - .raw = trace_bprint_raw, + .funcs = &trace_bprint_funcs, }; /* TRACE_PRINT */ static enum print_line_t trace_print_print(struct trace_iterator *iter, - int flags) + int flags, struct trace_event *event) { struct print_entry *field; struct trace_seq *s = &iter->seq; @@ -1241,7 +1215,8 @@ static enum print_line_t trace_print_print(struct trace_iterator *iter, return TRACE_TYPE_PARTIAL_LINE; } -static enum print_line_t trace_print_raw(struct trace_iterator *iter, int flags) +static enum print_line_t trace_print_raw(struct trace_iterator *iter, int flags, + struct trace_event *event) { struct print_entry *field; @@ -1256,18 +1231,21 @@ static enum print_line_t trace_print_raw(struct trace_iterator *iter, int flags) return TRACE_TYPE_PARTIAL_LINE; } -static struct trace_event trace_print_event = { - .type = TRACE_PRINT, +static struct trace_event_functions trace_print_funcs = { .trace = trace_print_print, .raw = trace_print_raw, }; +static struct trace_event trace_print_event = { + .type = TRACE_PRINT, + .funcs = &trace_print_funcs, +}; + 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, diff --git a/kernel/trace/trace_output.h b/kernel/trace/trace_output.h index 9d91c72ba38..c038eba0492 100644 --- a/kernel/trace/trace_output.h +++ b/kernel/trace/trace_output.h @@ -25,7 +25,7 @@ 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); + int flags, struct trace_event *event); extern int trace_print_lat_fmt(struct trace_seq *s, struct trace_entry *entry); diff --git a/kernel/trace/trace_sched_switch.c b/kernel/trace/trace_sched_switch.c index a55fccfede5..8f758d070c4 100644 --- a/kernel/trace/trace_sched_switch.c +++ b/kernel/trace/trace_sched_switch.c @@ -50,7 +50,7 @@ tracing_sched_switch_trace(struct trace_array *tr, } static void -probe_sched_switch(struct task_struct *prev, struct task_struct *next) +probe_sched_switch(void *ignore, struct task_struct *prev, struct task_struct *next) { struct trace_array_cpu *data; unsigned long flags; @@ -108,7 +108,7 @@ tracing_sched_wakeup_trace(struct trace_array *tr, } static void -probe_sched_wakeup(struct task_struct *wakee, int success) +probe_sched_wakeup(void *ignore, struct task_struct *wakee, int success) { struct trace_array_cpu *data; unsigned long flags; @@ -138,21 +138,21 @@ static int tracing_sched_register(void) { int ret; - ret = register_trace_sched_wakeup(probe_sched_wakeup); + ret = register_trace_sched_wakeup(probe_sched_wakeup, NULL); if (ret) { pr_info("wakeup trace: Couldn't activate tracepoint" " probe to kernel_sched_wakeup\n"); return ret; } - ret = register_trace_sched_wakeup_new(probe_sched_wakeup); + ret = register_trace_sched_wakeup_new(probe_sched_wakeup, NULL); if (ret) { pr_info("wakeup trace: Couldn't activate tracepoint" " probe to kernel_sched_wakeup_new\n"); goto fail_deprobe; } - ret = register_trace_sched_switch(probe_sched_switch); + ret = register_trace_sched_switch(probe_sched_switch, NULL); if (ret) { pr_info("sched trace: Couldn't activate tracepoint" " probe to kernel_sched_switch\n"); @@ -161,17 +161,17 @@ static int tracing_sched_register(void) return ret; fail_deprobe_wake_new: - unregister_trace_sched_wakeup_new(probe_sched_wakeup); + unregister_trace_sched_wakeup_new(probe_sched_wakeup, NULL); fail_deprobe: - unregister_trace_sched_wakeup(probe_sched_wakeup); + unregister_trace_sched_wakeup(probe_sched_wakeup, NULL); return ret; } static void tracing_sched_unregister(void) { - unregister_trace_sched_switch(probe_sched_switch); - unregister_trace_sched_wakeup_new(probe_sched_wakeup); - unregister_trace_sched_wakeup(probe_sched_wakeup); + unregister_trace_sched_switch(probe_sched_switch, NULL); + unregister_trace_sched_wakeup_new(probe_sched_wakeup, NULL); + unregister_trace_sched_wakeup(probe_sched_wakeup, NULL); } static void tracing_start_sched_switch(void) diff --git a/kernel/trace/trace_sched_wakeup.c b/kernel/trace/trace_sched_wakeup.c index 8052446ceea..7319559ed59 100644 --- a/kernel/trace/trace_sched_wakeup.c +++ b/kernel/trace/trace_sched_wakeup.c @@ -31,50 +31,99 @@ static int wakeup_rt; static arch_spinlock_t wakeup_lock = (arch_spinlock_t)__ARCH_SPIN_LOCK_UNLOCKED; +static void wakeup_reset(struct trace_array *tr); static void __wakeup_reset(struct trace_array *tr); +static int wakeup_graph_entry(struct ftrace_graph_ent *trace); +static void wakeup_graph_return(struct ftrace_graph_ret *trace); static int save_lat_flag; +#define TRACE_DISPLAY_GRAPH 1 + +static struct tracer_opt trace_opts[] = { +#ifdef CONFIG_FUNCTION_GRAPH_TRACER + /* display latency trace as call graph */ + { TRACER_OPT(display-graph, TRACE_DISPLAY_GRAPH) }, +#endif + { } /* Empty entry */ +}; + +static struct tracer_flags tracer_flags = { + .val = 0, + .opts = trace_opts, +}; + +#define is_graph() (tracer_flags.val & TRACE_DISPLAY_GRAPH) + #ifdef CONFIG_FUNCTION_TRACER + /* - * irqsoff uses its own tracer function to keep the overhead down: + * Prologue for the wakeup function tracers. + * + * Returns 1 if it is OK to continue, and preemption + * is disabled and data->disabled is incremented. + * 0 if the trace is to be ignored, and preemption + * is not disabled and data->disabled is + * kept the same. + * + * Note, this function is also used outside this ifdef but + * inside the #ifdef of the function graph tracer below. + * This is OK, since the function graph tracer is + * dependent on the function tracer. */ -static void -wakeup_tracer_call(unsigned long ip, unsigned long parent_ip) +static int +func_prolog_preempt_disable(struct trace_array *tr, + struct trace_array_cpu **data, + int *pc) { - struct trace_array *tr = wakeup_trace; - struct trace_array_cpu *data; - unsigned long flags; long disabled; - int resched; int cpu; - int pc; if (likely(!wakeup_task)) - return; + return 0; - pc = preempt_count(); - resched = ftrace_preempt_disable(); + *pc = preempt_count(); + preempt_disable_notrace(); cpu = raw_smp_processor_id(); if (cpu != wakeup_current_cpu) goto out_enable; - data = tr->data[cpu]; - disabled = atomic_inc_return(&data->disabled); + *data = tr->data[cpu]; + disabled = atomic_inc_return(&(*data)->disabled); if (unlikely(disabled != 1)) goto out; - local_irq_save(flags); + return 1; - trace_function(tr, ip, parent_ip, flags, pc); +out: + atomic_dec(&(*data)->disabled); + +out_enable: + preempt_enable_notrace(); + return 0; +} + +/* + * wakeup uses its own tracer function to keep the overhead down: + */ +static void +wakeup_tracer_call(unsigned long ip, unsigned long parent_ip) +{ + struct trace_array *tr = wakeup_trace; + struct trace_array_cpu *data; + unsigned long flags; + int pc; + + if (!func_prolog_preempt_disable(tr, &data, &pc)) + return; + local_irq_save(flags); + trace_function(tr, ip, parent_ip, flags, pc); local_irq_restore(flags); - out: atomic_dec(&data->disabled); - out_enable: - ftrace_preempt_enable(resched); + preempt_enable_notrace(); } static struct ftrace_ops trace_ops __read_mostly = @@ -83,6 +132,156 @@ static struct ftrace_ops trace_ops __read_mostly = }; #endif /* CONFIG_FUNCTION_TRACER */ +static int start_func_tracer(int graph) +{ + int ret; + + if (!graph) + ret = register_ftrace_function(&trace_ops); + else + ret = register_ftrace_graph(&wakeup_graph_return, + &wakeup_graph_entry); + + if (!ret && tracing_is_enabled()) + tracer_enabled = 1; + else + tracer_enabled = 0; + + return ret; +} + +static void stop_func_tracer(int graph) +{ + tracer_enabled = 0; + + if (!graph) + unregister_ftrace_function(&trace_ops); + else + unregister_ftrace_graph(); +} + +#ifdef CONFIG_FUNCTION_GRAPH_TRACER +static int wakeup_set_flag(u32 old_flags, u32 bit, int set) +{ + + if (!(bit & TRACE_DISPLAY_GRAPH)) + return -EINVAL; + + if (!(is_graph() ^ set)) + return 0; + + stop_func_tracer(!set); + + wakeup_reset(wakeup_trace); + tracing_max_latency = 0; + + return start_func_tracer(set); +} + +static int wakeup_graph_entry(struct ftrace_graph_ent *trace) +{ + struct trace_array *tr = wakeup_trace; + struct trace_array_cpu *data; + unsigned long flags; + int pc, ret = 0; + + if (!func_prolog_preempt_disable(tr, &data, &pc)) + return 0; + + local_save_flags(flags); + ret = __trace_graph_entry(tr, trace, flags, pc); + atomic_dec(&data->disabled); + preempt_enable_notrace(); + + return ret; +} + +static void wakeup_graph_return(struct ftrace_graph_ret *trace) +{ + struct trace_array *tr = wakeup_trace; + struct trace_array_cpu *data; + unsigned long flags; + int pc; + + if (!func_prolog_preempt_disable(tr, &data, &pc)) + return; + + local_save_flags(flags); + __trace_graph_return(tr, trace, flags, pc); + atomic_dec(&data->disabled); + + preempt_enable_notrace(); + return; +} + +static void wakeup_trace_open(struct trace_iterator *iter) +{ + if (is_graph()) + graph_trace_open(iter); +} + +static void wakeup_trace_close(struct trace_iterator *iter) +{ + if (iter->private) + graph_trace_close(iter); +} + +#define GRAPH_TRACER_FLAGS (TRACE_GRAPH_PRINT_PROC) + +static enum print_line_t wakeup_print_line(struct trace_iterator *iter) +{ + /* + * In graph mode call the graph tracer output function, + * otherwise go with the TRACE_FN event handler + */ + if (is_graph()) + return print_graph_function_flags(iter, GRAPH_TRACER_FLAGS); + + return TRACE_TYPE_UNHANDLED; +} + +static void wakeup_print_header(struct seq_file *s) +{ + if (is_graph()) + print_graph_headers_flags(s, GRAPH_TRACER_FLAGS); + else + trace_default_header(s); +} + +static void +__trace_function(struct trace_array *tr, + unsigned long ip, unsigned long parent_ip, + unsigned long flags, int pc) +{ + if (is_graph()) + trace_graph_function(tr, ip, parent_ip, flags, pc); + else + trace_function(tr, ip, parent_ip, flags, pc); +} +#else +#define __trace_function trace_function + +static int wakeup_set_flag(u32 old_flags, u32 bit, int set) +{ + return -EINVAL; +} + +static int wakeup_graph_entry(struct ftrace_graph_ent *trace) +{ + return -1; +} + +static enum print_line_t wakeup_print_line(struct trace_iterator *iter) +{ + return TRACE_TYPE_UNHANDLED; +} + +static void wakeup_graph_return(struct ftrace_graph_ret *trace) { } +static void wakeup_print_header(struct seq_file *s) { } +static void wakeup_trace_open(struct trace_iterator *iter) { } +static void wakeup_trace_close(struct trace_iterator *iter) { } +#endif /* CONFIG_FUNCTION_GRAPH_TRACER */ + /* * Should this new latency be reported/recorded? */ @@ -98,7 +297,8 @@ static int report_latency(cycle_t delta) return 1; } -static void probe_wakeup_migrate_task(struct task_struct *task, int cpu) +static void +probe_wakeup_migrate_task(void *ignore, struct task_struct *task, int cpu) { if (task != wakeup_task) return; @@ -107,7 +307,8 @@ static void probe_wakeup_migrate_task(struct task_struct *task, int cpu) } static void notrace -probe_wakeup_sched_switch(struct task_struct *prev, struct task_struct *next) +probe_wakeup_sched_switch(void *ignore, + struct task_struct *prev, struct task_struct *next) { struct trace_array_cpu *data; cycle_t T0, T1, delta; @@ -151,7 +352,7 @@ probe_wakeup_sched_switch(struct task_struct *prev, struct task_struct *next) /* 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); + __trace_function(wakeup_trace, CALLER_ADDR0, CALLER_ADDR1, flags, pc); tracing_sched_switch_trace(wakeup_trace, prev, next, flags, pc); T0 = data->preempt_timestamp; @@ -199,7 +400,7 @@ static void wakeup_reset(struct trace_array *tr) } static void -probe_wakeup(struct task_struct *p, int success) +probe_wakeup(void *ignore, struct task_struct *p, int success) { struct trace_array_cpu *data; int cpu = smp_processor_id(); @@ -251,7 +452,7 @@ probe_wakeup(struct task_struct *p, int success) * 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); + __trace_function(wakeup_trace, CALLER_ADDR1, CALLER_ADDR2, flags, pc); out_locked: arch_spin_unlock(&wakeup_lock); @@ -263,28 +464,28 @@ static void start_wakeup_tracer(struct trace_array *tr) { int ret; - ret = register_trace_sched_wakeup(probe_wakeup); + ret = register_trace_sched_wakeup(probe_wakeup, NULL); if (ret) { pr_info("wakeup trace: Couldn't activate tracepoint" " probe to kernel_sched_wakeup\n"); return; } - ret = register_trace_sched_wakeup_new(probe_wakeup); + ret = register_trace_sched_wakeup_new(probe_wakeup, NULL); if (ret) { pr_info("wakeup trace: Couldn't activate tracepoint" " probe to kernel_sched_wakeup_new\n"); goto fail_deprobe; } - ret = register_trace_sched_switch(probe_wakeup_sched_switch); + ret = register_trace_sched_switch(probe_wakeup_sched_switch, NULL); if (ret) { pr_info("sched trace: Couldn't activate tracepoint" " probe to kernel_sched_switch\n"); goto fail_deprobe_wake_new; } - ret = register_trace_sched_migrate_task(probe_wakeup_migrate_task); + ret = register_trace_sched_migrate_task(probe_wakeup_migrate_task, NULL); if (ret) { pr_info("wakeup trace: Couldn't activate tracepoint" " probe to kernel_sched_migrate_task\n"); @@ -302,28 +503,24 @@ static void start_wakeup_tracer(struct trace_array *tr) */ smp_wmb(); - register_ftrace_function(&trace_ops); - - if (tracing_is_enabled()) - tracer_enabled = 1; - else - tracer_enabled = 0; + if (start_func_tracer(is_graph())) + printk(KERN_ERR "failed to start wakeup tracer\n"); return; fail_deprobe_wake_new: - unregister_trace_sched_wakeup_new(probe_wakeup); + unregister_trace_sched_wakeup_new(probe_wakeup, NULL); fail_deprobe: - unregister_trace_sched_wakeup(probe_wakeup); + unregister_trace_sched_wakeup(probe_wakeup, NULL); } static void stop_wakeup_tracer(struct trace_array *tr) { 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); + stop_func_tracer(is_graph()); + unregister_trace_sched_switch(probe_wakeup_sched_switch, NULL); + unregister_trace_sched_wakeup_new(probe_wakeup, NULL); + unregister_trace_sched_wakeup(probe_wakeup, NULL); + unregister_trace_sched_migrate_task(probe_wakeup_migrate_task, NULL); } static int __wakeup_tracer_init(struct trace_array *tr) @@ -378,9 +575,16 @@ static struct tracer wakeup_tracer __read_mostly = .start = wakeup_tracer_start, .stop = wakeup_tracer_stop, .print_max = 1, + .print_header = wakeup_print_header, + .print_line = wakeup_print_line, + .flags = &tracer_flags, + .set_flag = wakeup_set_flag, #ifdef CONFIG_FTRACE_SELFTEST .selftest = trace_selftest_startup_wakeup, #endif + .open = wakeup_trace_open, + .close = wakeup_trace_close, + .use_max_tr = 1, }; static struct tracer wakeup_rt_tracer __read_mostly = @@ -392,9 +596,16 @@ static struct tracer wakeup_rt_tracer __read_mostly = .stop = wakeup_tracer_stop, .wait_pipe = poll_wait_pipe, .print_max = 1, + .print_header = wakeup_print_header, + .print_line = wakeup_print_line, + .flags = &tracer_flags, + .set_flag = wakeup_set_flag, #ifdef CONFIG_FTRACE_SELFTEST .selftest = trace_selftest_startup_wakeup, #endif + .open = wakeup_trace_open, + .close = wakeup_trace_close, + .use_max_tr = 1, }; __init static int init_wakeup_tracer(void) diff --git a/kernel/trace/trace_selftest.c b/kernel/trace/trace_selftest.c index 250e7f9bd2f..659732eba07 100644 --- a/kernel/trace/trace_selftest.c +++ b/kernel/trace/trace_selftest.c @@ -13,11 +13,9 @@ static inline int trace_valid_entry(struct trace_entry *entry) case TRACE_WAKE: case TRACE_STACK: case TRACE_PRINT: - case TRACE_SPECIAL: case TRACE_BRANCH: case TRACE_GRAPH_ENT: case TRACE_GRAPH_RET: - case TRACE_KSYM: return 1; } return 0; @@ -560,7 +558,7 @@ trace_selftest_startup_nop(struct tracer *trace, struct trace_array *tr) static int trace_wakeup_test_thread(void *data) { /* Make this a RT thread, doesn't need to be too high */ - struct sched_param param = { .sched_priority = 5 }; + static const struct sched_param param = { .sched_priority = 5 }; struct completion *x = data; sched_setscheduler(current, SCHED_FIFO, ¶m); @@ -691,38 +689,6 @@ trace_selftest_startup_sched_switch(struct tracer *trace, struct trace_array *tr } #endif /* CONFIG_CONTEXT_SWITCH_TRACER */ -#ifdef CONFIG_SYSPROF_TRACER -int -trace_selftest_startup_sysprof(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; - } - - /* Sleep for a 1/10 of a second */ - msleep(100); - /* stop the tracing. */ - 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; - } - - return ret; -} -#endif /* CONFIG_SYSPROF_TRACER */ - #ifdef CONFIG_BRANCH_TRACER int trace_selftest_startup_branch(struct tracer *trace, struct trace_array *tr) @@ -755,56 +721,3 @@ trace_selftest_startup_branch(struct tracer *trace, struct trace_array *tr) } #endif /* CONFIG_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 f4bc9b27de5..4c5dead0c23 100644 --- a/kernel/trace/trace_stack.c +++ b/kernel/trace/trace_stack.c @@ -110,12 +110,12 @@ static inline void check_stack(void) static void stack_trace_call(unsigned long ip, unsigned long parent_ip) { - int cpu, resched; + int cpu; if (unlikely(!ftrace_enabled || stack_trace_disabled)) return; - resched = ftrace_preempt_disable(); + preempt_disable_notrace(); cpu = raw_smp_processor_id(); /* no atomic needed, we only modify this variable by this cpu */ @@ -127,7 +127,7 @@ stack_trace_call(unsigned long ip, unsigned long parent_ip) out: per_cpu(trace_active, cpu)--; /* prevent recursion in schedule */ - ftrace_preempt_enable(resched); + preempt_enable_notrace(); } static struct ftrace_ops trace_ops __read_mostly = @@ -195,6 +195,7 @@ static const struct file_operations stack_max_size_fops = { .open = tracing_open_generic, .read = stack_max_size_read, .write = stack_max_size_write, + .llseek = default_llseek, }; static void * @@ -249,7 +250,7 @@ static int trace_lookup_stack(struct seq_file *m, long i) { unsigned long addr = stack_dump_trace[i]; - return seq_printf(m, "%pF\n", (void *)addr); + return seq_printf(m, "%pS\n", (void *)addr); } static void print_disabled(struct seq_file *m) diff --git a/kernel/trace/trace_syscalls.c b/kernel/trace/trace_syscalls.c index 4d6d711717f..b706529b4fc 100644 --- a/kernel/trace/trace_syscalls.c +++ b/kernel/trace/trace_syscalls.c @@ -15,6 +15,46 @@ static int sys_refcount_exit; static DECLARE_BITMAP(enabled_enter_syscalls, NR_syscalls); static DECLARE_BITMAP(enabled_exit_syscalls, NR_syscalls); +static int syscall_enter_register(struct ftrace_event_call *event, + enum trace_reg type); +static int syscall_exit_register(struct ftrace_event_call *event, + enum trace_reg type); + +static int syscall_enter_define_fields(struct ftrace_event_call *call); +static int syscall_exit_define_fields(struct ftrace_event_call *call); + +static struct list_head * +syscall_get_enter_fields(struct ftrace_event_call *call) +{ + struct syscall_metadata *entry = call->data; + + return &entry->enter_fields; +} + +struct trace_event_functions enter_syscall_print_funcs = { + .trace = print_syscall_enter, +}; + +struct trace_event_functions exit_syscall_print_funcs = { + .trace = print_syscall_exit, +}; + +struct ftrace_event_class event_class_syscall_enter = { + .system = "syscalls", + .reg = syscall_enter_register, + .define_fields = syscall_enter_define_fields, + .get_fields = syscall_get_enter_fields, + .raw_init = init_syscall_trace, +}; + +struct ftrace_event_class event_class_syscall_exit = { + .system = "syscalls", + .reg = syscall_exit_register, + .define_fields = syscall_exit_define_fields, + .fields = LIST_HEAD_INIT(event_class_syscall_exit.fields), + .raw_init = init_syscall_trace, +}; + extern unsigned long __start_syscalls_metadata[]; extern unsigned long __stop_syscalls_metadata[]; @@ -53,7 +93,8 @@ static struct syscall_metadata *syscall_nr_to_meta(int nr) } enum print_line_t -print_syscall_enter(struct trace_iterator *iter, int flags) +print_syscall_enter(struct trace_iterator *iter, int flags, + struct trace_event *event) { struct trace_seq *s = &iter->seq; struct trace_entry *ent = iter->ent; @@ -68,7 +109,7 @@ print_syscall_enter(struct trace_iterator *iter, int flags) if (!entry) goto end; - if (entry->enter_event->id != ent->type) { + if (entry->enter_event->event.type != ent->type) { WARN_ON_ONCE(1); goto end; } @@ -105,7 +146,8 @@ end: } enum print_line_t -print_syscall_exit(struct trace_iterator *iter, int flags) +print_syscall_exit(struct trace_iterator *iter, int flags, + struct trace_event *event) { struct trace_seq *s = &iter->seq; struct trace_entry *ent = iter->ent; @@ -123,7 +165,7 @@ print_syscall_exit(struct trace_iterator *iter, int flags) return TRACE_TYPE_HANDLED; } - if (entry->exit_event->id != ent->type) { + if (entry->exit_event->event.type != ent->type) { WARN_ON_ONCE(1); return TRACE_TYPE_UNHANDLED; } @@ -205,7 +247,7 @@ static void free_syscall_print_fmt(struct ftrace_event_call *call) kfree(call->print_fmt); } -int syscall_enter_define_fields(struct ftrace_event_call *call) +static int syscall_enter_define_fields(struct ftrace_event_call *call) { struct syscall_trace_enter trace; struct syscall_metadata *meta = call->data; @@ -228,7 +270,7 @@ int syscall_enter_define_fields(struct ftrace_event_call *call) return ret; } -int syscall_exit_define_fields(struct ftrace_event_call *call) +static int syscall_exit_define_fields(struct ftrace_event_call *call) { struct syscall_trace_exit trace; int ret; @@ -243,7 +285,7 @@ int syscall_exit_define_fields(struct ftrace_event_call *call) return ret; } -void ftrace_syscall_enter(struct pt_regs *regs, long id) +void ftrace_syscall_enter(void *ignore, struct pt_regs *regs, long id) { struct syscall_trace_enter *entry; struct syscall_metadata *sys_data; @@ -265,7 +307,7 @@ void ftrace_syscall_enter(struct pt_regs *regs, long id) 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); + sys_data->enter_event->event.type, size, 0, 0); if (!event) return; @@ -278,7 +320,7 @@ void ftrace_syscall_enter(struct pt_regs *regs, long id) trace_current_buffer_unlock_commit(buffer, event, 0, 0); } -void ftrace_syscall_exit(struct pt_regs *regs, long ret) +void ftrace_syscall_exit(void *ignore, struct pt_regs *regs, long ret) { struct syscall_trace_exit *entry; struct syscall_metadata *sys_data; @@ -297,7 +339,7 @@ void ftrace_syscall_exit(struct pt_regs *regs, long ret) return; event = trace_current_buffer_lock_reserve(&buffer, - sys_data->exit_event->id, sizeof(*entry), 0, 0); + sys_data->exit_event->event.type, sizeof(*entry), 0, 0); if (!event) return; @@ -320,7 +362,7 @@ int reg_event_syscall_enter(struct ftrace_event_call *call) return -ENOSYS; mutex_lock(&syscall_trace_lock); if (!sys_refcount_enter) - ret = register_trace_sys_enter(ftrace_syscall_enter); + ret = register_trace_sys_enter(ftrace_syscall_enter, NULL); if (!ret) { set_bit(num, enabled_enter_syscalls); sys_refcount_enter++; @@ -340,7 +382,7 @@ void unreg_event_syscall_enter(struct ftrace_event_call *call) sys_refcount_enter--; clear_bit(num, enabled_enter_syscalls); if (!sys_refcount_enter) - unregister_trace_sys_enter(ftrace_syscall_enter); + unregister_trace_sys_enter(ftrace_syscall_enter, NULL); mutex_unlock(&syscall_trace_lock); } @@ -354,7 +396,7 @@ int reg_event_syscall_exit(struct ftrace_event_call *call) return -ENOSYS; mutex_lock(&syscall_trace_lock); if (!sys_refcount_exit) - ret = register_trace_sys_exit(ftrace_syscall_exit); + ret = register_trace_sys_exit(ftrace_syscall_exit, NULL); if (!ret) { set_bit(num, enabled_exit_syscalls); sys_refcount_exit++; @@ -374,7 +416,7 @@ void unreg_event_syscall_exit(struct ftrace_event_call *call) sys_refcount_exit--; clear_bit(num, enabled_exit_syscalls); if (!sys_refcount_exit) - unregister_trace_sys_exit(ftrace_syscall_exit); + unregister_trace_sys_exit(ftrace_syscall_exit, NULL); mutex_unlock(&syscall_trace_lock); } @@ -434,11 +476,11 @@ static DECLARE_BITMAP(enabled_perf_exit_syscalls, NR_syscalls); static int sys_perf_refcount_enter; static int sys_perf_refcount_exit; -static void perf_syscall_enter(struct pt_regs *regs, long id) +static void perf_syscall_enter(void *ignore, struct pt_regs *regs, long id) { struct syscall_metadata *sys_data; struct syscall_trace_enter *rec; - unsigned long flags; + struct hlist_head *head; int syscall_nr; int rctx; int size; @@ -461,14 +503,16 @@ static void perf_syscall_enter(struct pt_regs *regs, long id) return; rec = (struct syscall_trace_enter *)perf_trace_buf_prepare(size, - sys_data->enter_event->id, &rctx, &flags); + sys_data->enter_event->event.type, regs, &rctx); if (!rec) return; rec->nr = syscall_nr; syscall_get_arguments(current, regs, 0, sys_data->nb_args, (unsigned long *)&rec->args); - perf_trace_buf_submit(rec, size, rctx, 0, 1, flags, regs); + + head = this_cpu_ptr(sys_data->enter_event->perf_events); + perf_trace_buf_submit(rec, size, rctx, 0, 1, regs, head); } int perf_sysenter_enable(struct ftrace_event_call *call) @@ -480,7 +524,7 @@ int perf_sysenter_enable(struct ftrace_event_call *call) mutex_lock(&syscall_trace_lock); if (!sys_perf_refcount_enter) - ret = register_trace_sys_enter(perf_syscall_enter); + ret = register_trace_sys_enter(perf_syscall_enter, NULL); if (ret) { pr_info("event trace: Could not activate" "syscall entry trace point"); @@ -502,15 +546,15 @@ void perf_sysenter_disable(struct ftrace_event_call *call) sys_perf_refcount_enter--; clear_bit(num, enabled_perf_enter_syscalls); if (!sys_perf_refcount_enter) - unregister_trace_sys_enter(perf_syscall_enter); + unregister_trace_sys_enter(perf_syscall_enter, NULL); mutex_unlock(&syscall_trace_lock); } -static void perf_syscall_exit(struct pt_regs *regs, long ret) +static void perf_syscall_exit(void *ignore, struct pt_regs *regs, long ret) { struct syscall_metadata *sys_data; struct syscall_trace_exit *rec; - unsigned long flags; + struct hlist_head *head; int syscall_nr; int rctx; int size; @@ -536,14 +580,15 @@ static void perf_syscall_exit(struct pt_regs *regs, long ret) return; rec = (struct syscall_trace_exit *)perf_trace_buf_prepare(size, - sys_data->exit_event->id, &rctx, &flags); + sys_data->exit_event->event.type, regs, &rctx); if (!rec) return; rec->nr = syscall_nr; rec->ret = syscall_get_return_value(current, regs); - perf_trace_buf_submit(rec, size, rctx, 0, 1, flags, regs); + head = this_cpu_ptr(sys_data->exit_event->perf_events); + perf_trace_buf_submit(rec, size, rctx, 0, 1, regs, head); } int perf_sysexit_enable(struct ftrace_event_call *call) @@ -555,7 +600,7 @@ int perf_sysexit_enable(struct ftrace_event_call *call) mutex_lock(&syscall_trace_lock); if (!sys_perf_refcount_exit) - ret = register_trace_sys_exit(perf_syscall_exit); + ret = register_trace_sys_exit(perf_syscall_exit, NULL); if (ret) { pr_info("event trace: Could not activate" "syscall exit trace point"); @@ -577,9 +622,50 @@ void perf_sysexit_disable(struct ftrace_event_call *call) sys_perf_refcount_exit--; clear_bit(num, enabled_perf_exit_syscalls); if (!sys_perf_refcount_exit) - unregister_trace_sys_exit(perf_syscall_exit); + unregister_trace_sys_exit(perf_syscall_exit, NULL); mutex_unlock(&syscall_trace_lock); } #endif /* CONFIG_PERF_EVENTS */ +static int syscall_enter_register(struct ftrace_event_call *event, + enum trace_reg type) +{ + switch (type) { + case TRACE_REG_REGISTER: + return reg_event_syscall_enter(event); + case TRACE_REG_UNREGISTER: + unreg_event_syscall_enter(event); + return 0; + +#ifdef CONFIG_PERF_EVENTS + case TRACE_REG_PERF_REGISTER: + return perf_sysenter_enable(event); + case TRACE_REG_PERF_UNREGISTER: + perf_sysenter_disable(event); + return 0; +#endif + } + return 0; +} + +static int syscall_exit_register(struct ftrace_event_call *event, + enum trace_reg type) +{ + switch (type) { + case TRACE_REG_REGISTER: + return reg_event_syscall_exit(event); + case TRACE_REG_UNREGISTER: + unreg_event_syscall_exit(event); + return 0; + +#ifdef CONFIG_PERF_EVENTS + case TRACE_REG_PERF_REGISTER: + return perf_sysexit_enable(event); + case TRACE_REG_PERF_UNREGISTER: + perf_sysexit_disable(event); + return 0; +#endif + } + return 0; +} diff --git a/kernel/trace/trace_sysprof.c b/kernel/trace/trace_sysprof.c deleted file mode 100644 index a7974a552ca..00000000000 --- a/kernel/trace/trace_sysprof.c +++ /dev/null @@ -1,329 +0,0 @@ -/* - * trace stack traces - * - * Copyright (C) 2004-2008, Soeren Sandmann - * Copyright (C) 2007 Steven Rostedt <srostedt@redhat.com> - * Copyright (C) 2008 Ingo Molnar <mingo@redhat.com> - */ -#include <linux/kallsyms.h> -#include <linux/debugfs.h> -#include <linux/hrtimer.h> -#include <linux/uaccess.h> -#include <linux/ftrace.h> -#include <linux/module.h> -#include <linux/irq.h> -#include <linux/fs.h> - -#include <asm/stacktrace.h> - -#include "trace.h" - -static struct trace_array *sysprof_trace; -static int __read_mostly tracer_enabled; - -/* - * 1 msec sample interval by default: - */ -static unsigned long sample_period = 1000000; -static const unsigned int sample_max_depth = 512; - -static DEFINE_MUTEX(sample_timer_lock); -/* - * Per CPU hrtimers that do the profiling: - */ -static DEFINE_PER_CPU(struct hrtimer, stack_trace_hrtimer); - -struct stack_frame { - const void __user *next_fp; - unsigned long return_address; -}; - -static int copy_stack_frame(const void __user *fp, struct stack_frame *frame) -{ - int ret; - - if (!access_ok(VERIFY_READ, fp, sizeof(*frame))) - return 0; - - ret = 1; - pagefault_disable(); - if (__copy_from_user_inatomic(frame, fp, sizeof(*frame))) - ret = 0; - pagefault_enable(); - - return ret; -} - -struct backtrace_info { - struct trace_array_cpu *data; - struct trace_array *tr; - int pos; -}; - -static void -backtrace_warning_symbol(void *data, char *msg, unsigned long symbol) -{ - /* Ignore warnings */ -} - -static void backtrace_warning(void *data, char *msg) -{ - /* Ignore warnings */ -} - -static int backtrace_stack(void *data, char *name) -{ - /* Don't bother with IRQ stacks for now */ - return -1; -} - -static void backtrace_address(void *data, unsigned long addr, int reliable) -{ - struct backtrace_info *info = data; - - if (info->pos < sample_max_depth && reliable) { - __trace_special(info->tr, info->data, 1, addr, 0); - - info->pos++; - } -} - -static const struct stacktrace_ops backtrace_ops = { - .warning = backtrace_warning, - .warning_symbol = backtrace_warning_symbol, - .stack = backtrace_stack, - .address = backtrace_address, - .walk_stack = print_context_stack, -}; - -static int -trace_kernel(struct pt_regs *regs, struct trace_array *tr, - struct trace_array_cpu *data) -{ - struct backtrace_info info; - unsigned long bp; - char *stack; - - info.tr = tr; - info.data = data; - info.pos = 1; - - __trace_special(info.tr, info.data, 1, regs->ip, 0); - - stack = ((char *)regs + sizeof(struct pt_regs)); -#ifdef CONFIG_FRAME_POINTER - bp = regs->bp; -#else - bp = 0; -#endif - - dump_trace(NULL, regs, (void *)stack, bp, &backtrace_ops, &info); - - return info.pos; -} - -static void timer_notify(struct pt_regs *regs, int cpu) -{ - struct trace_array_cpu *data; - struct stack_frame frame; - struct trace_array *tr; - const void __user *fp; - int is_user; - int i; - - if (!regs) - return; - - tr = sysprof_trace; - data = tr->data[cpu]; - is_user = user_mode(regs); - - if (!current || current->pid == 0) - return; - - if (is_user && current->state != TASK_RUNNING) - return; - - __trace_special(tr, data, 0, 0, current->pid); - - if (!is_user) - i = trace_kernel(regs, tr, data); - else - i = 0; - - /* - * Trace user stack if we are not a kernel thread - */ - if (current->mm && i < sample_max_depth) { - regs = (struct pt_regs *)current->thread.sp0 - 1; - - fp = (void __user *)regs->bp; - - __trace_special(tr, data, 2, regs->ip, 0); - - while (i < sample_max_depth) { - frame.next_fp = NULL; - frame.return_address = 0; - if (!copy_stack_frame(fp, &frame)) - break; - if ((unsigned long)fp < regs->sp) - break; - - __trace_special(tr, data, 2, frame.return_address, - (unsigned long)fp); - fp = frame.next_fp; - - i++; - } - - } - - /* - * Special trace entry if we overflow the max depth: - */ - if (i == sample_max_depth) - __trace_special(tr, data, -1, -1, -1); - - __trace_special(tr, data, 3, current->pid, i); -} - -static enum hrtimer_restart stack_trace_timer_fn(struct hrtimer *hrtimer) -{ - /* trace here */ - timer_notify(get_irq_regs(), smp_processor_id()); - - hrtimer_forward_now(hrtimer, ns_to_ktime(sample_period)); - - return HRTIMER_RESTART; -} - -static void start_stack_timer(void *unused) -{ - struct hrtimer *hrtimer = &__get_cpu_var(stack_trace_hrtimer); - - 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_PINNED); -} - -static void start_stack_timers(void) -{ - on_each_cpu(start_stack_timer, NULL, 1); -} - -static void stop_stack_timer(int cpu) -{ - struct hrtimer *hrtimer = &per_cpu(stack_trace_hrtimer, cpu); - - hrtimer_cancel(hrtimer); -} - -static void stop_stack_timers(void) -{ - int cpu; - - for_each_online_cpu(cpu) - stop_stack_timer(cpu); -} - -static void stop_stack_trace(struct trace_array *tr) -{ - mutex_lock(&sample_timer_lock); - stop_stack_timers(); - tracer_enabled = 0; - mutex_unlock(&sample_timer_lock); -} - -static int stack_trace_init(struct trace_array *tr) -{ - sysprof_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); -} - -static struct tracer stack_trace __read_mostly = -{ - .name = "sysprof", - .init = stack_trace_init, - .reset = stack_trace_reset, -#ifdef CONFIG_FTRACE_SELFTEST - .selftest = trace_selftest_startup_sysprof, -#endif -}; - -__init static int init_stack_trace(void) -{ - return register_tracer(&stack_trace); -} -device_initcall(init_stack_trace); - -#define MAX_LONG_DIGITS 22 - -static ssize_t -sysprof_sample_read(struct file *filp, char __user *ubuf, - size_t cnt, loff_t *ppos) -{ - char buf[MAX_LONG_DIGITS]; - int r; - - r = sprintf(buf, "%ld\n", nsecs_to_usecs(sample_period)); - - return simple_read_from_buffer(ubuf, cnt, ppos, buf, r); -} - -static ssize_t -sysprof_sample_write(struct file *filp, const char __user *ubuf, - size_t cnt, loff_t *ppos) -{ - char buf[MAX_LONG_DIGITS]; - unsigned long val; - - if (cnt > MAX_LONG_DIGITS-1) - cnt = MAX_LONG_DIGITS-1; - - if (copy_from_user(&buf, ubuf, cnt)) - return -EFAULT; - - buf[cnt] = 0; - - val = simple_strtoul(buf, NULL, 10); - /* - * Enforce a minimum sample period of 100 usecs: - */ - if (val < 100) - val = 100; - - mutex_lock(&sample_timer_lock); - stop_stack_timers(); - sample_period = val * 1000; - start_stack_timers(); - mutex_unlock(&sample_timer_lock); - - return cnt; -} - -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) -{ - - trace_create_file("sysprof_sample_period", 0644, - d_tracer, NULL, &sysprof_sample_fops); -} diff --git a/kernel/trace/trace_workqueue.c b/kernel/trace/trace_workqueue.c index cc2d2faa7d9..209b379a472 100644 --- a/kernel/trace/trace_workqueue.c +++ b/kernel/trace/trace_workqueue.c @@ -49,7 +49,8 @@ static void cpu_workqueue_stat_free(struct kref *kref) /* Insertion of a work */ static void -probe_workqueue_insertion(struct task_struct *wq_thread, +probe_workqueue_insertion(void *ignore, + struct task_struct *wq_thread, struct work_struct *work) { int cpu = cpumask_first(&wq_thread->cpus_allowed); @@ -70,7 +71,8 @@ found: /* Execution of a work */ static void -probe_workqueue_execution(struct task_struct *wq_thread, +probe_workqueue_execution(void *ignore, + struct task_struct *wq_thread, struct work_struct *work) { int cpu = cpumask_first(&wq_thread->cpus_allowed); @@ -90,7 +92,8 @@ found: } /* Creation of a cpu workqueue thread */ -static void probe_workqueue_creation(struct task_struct *wq_thread, int cpu) +static void probe_workqueue_creation(void *ignore, + struct task_struct *wq_thread, int cpu) { struct cpu_workqueue_stats *cws; unsigned long flags; @@ -114,7 +117,8 @@ static void probe_workqueue_creation(struct task_struct *wq_thread, int cpu) } /* Destruction of a cpu workqueue thread */ -static void probe_workqueue_destruction(struct task_struct *wq_thread) +static void +probe_workqueue_destruction(void *ignore, struct task_struct *wq_thread) { /* Workqueue only execute on one cpu */ int cpu = cpumask_first(&wq_thread->cpus_allowed); @@ -259,35 +263,35 @@ int __init trace_workqueue_early_init(void) { int ret, cpu; - ret = register_trace_workqueue_insertion(probe_workqueue_insertion); + for_each_possible_cpu(cpu) { + spin_lock_init(&workqueue_cpu_stat(cpu)->lock); + INIT_LIST_HEAD(&workqueue_cpu_stat(cpu)->list); + } + + ret = register_trace_workqueue_insertion(probe_workqueue_insertion, NULL); if (ret) goto out; - ret = register_trace_workqueue_execution(probe_workqueue_execution); + ret = register_trace_workqueue_execution(probe_workqueue_execution, NULL); if (ret) goto no_insertion; - ret = register_trace_workqueue_creation(probe_workqueue_creation); + ret = register_trace_workqueue_creation(probe_workqueue_creation, NULL); if (ret) goto no_execution; - ret = register_trace_workqueue_destruction(probe_workqueue_destruction); + ret = register_trace_workqueue_destruction(probe_workqueue_destruction, NULL); 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); + unregister_trace_workqueue_creation(probe_workqueue_creation, NULL); no_execution: - unregister_trace_workqueue_execution(probe_workqueue_execution); + unregister_trace_workqueue_execution(probe_workqueue_execution, NULL); no_insertion: - unregister_trace_workqueue_insertion(probe_workqueue_insertion); + unregister_trace_workqueue_insertion(probe_workqueue_insertion, NULL); out: pr_warning("trace_workqueue: unable to trace workqueues\n"); diff --git a/kernel/tracepoint.c b/kernel/tracepoint.c index cc89be5bc0f..e95ee7f31d4 100644 --- a/kernel/tracepoint.c +++ b/kernel/tracepoint.c @@ -25,6 +25,7 @@ #include <linux/err.h> #include <linux/slab.h> #include <linux/sched.h> +#include <linux/jump_label.h> extern struct tracepoint __start___tracepoints[]; extern struct tracepoint __stop___tracepoints[]; @@ -54,7 +55,7 @@ static struct hlist_head tracepoint_table[TRACEPOINT_TABLE_SIZE]; */ struct tracepoint_entry { struct hlist_node hlist; - void **funcs; + struct tracepoint_func *funcs; int refcount; /* Number of times armed. 0 if disarmed. */ char name[0]; }; @@ -64,12 +65,12 @@ struct tp_probes { struct rcu_head rcu; struct list_head list; } u; - void *probes[0]; + struct tracepoint_func probes[0]; }; static inline void *allocate_probes(int count) { - struct tp_probes *p = kmalloc(count * sizeof(void *) + struct tp_probes *p = kmalloc(count * sizeof(struct tracepoint_func) + sizeof(struct tp_probes), GFP_KERNEL); return p == NULL ? NULL : p->probes; } @@ -79,7 +80,7 @@ static void rcu_free_old_probes(struct rcu_head *head) kfree(container_of(head, struct tp_probes, u.rcu)); } -static inline void release_probes(void *old) +static inline void release_probes(struct tracepoint_func *old) { if (old) { struct tp_probes *tp_probes = container_of(old, @@ -95,15 +96,16 @@ static void debug_print_probes(struct tracepoint_entry *entry) if (!tracepoint_debug || !entry->funcs) return; - for (i = 0; entry->funcs[i]; i++) - printk(KERN_DEBUG "Probe %d : %p\n", i, entry->funcs[i]); + for (i = 0; entry->funcs[i].func; i++) + printk(KERN_DEBUG "Probe %d : %p\n", i, entry->funcs[i].func); } -static void * -tracepoint_entry_add_probe(struct tracepoint_entry *entry, void *probe) +static struct tracepoint_func * +tracepoint_entry_add_probe(struct tracepoint_entry *entry, + void *probe, void *data) { int nr_probes = 0; - void **old, **new; + struct tracepoint_func *old, *new; WARN_ON(!probe); @@ -111,8 +113,9 @@ tracepoint_entry_add_probe(struct tracepoint_entry *entry, void *probe) old = entry->funcs; if (old) { /* (N -> N+1), (N != 0, 1) probes */ - for (nr_probes = 0; old[nr_probes]; nr_probes++) - if (old[nr_probes] == probe) + for (nr_probes = 0; old[nr_probes].func; nr_probes++) + if (old[nr_probes].func == probe && + old[nr_probes].data == data) return ERR_PTR(-EEXIST); } /* + 2 : one for new probe, one for NULL func */ @@ -120,9 +123,10 @@ tracepoint_entry_add_probe(struct tracepoint_entry *entry, void *probe) if (new == NULL) return ERR_PTR(-ENOMEM); if (old) - memcpy(new, old, nr_probes * sizeof(void *)); - new[nr_probes] = probe; - new[nr_probes + 1] = NULL; + memcpy(new, old, nr_probes * sizeof(struct tracepoint_func)); + new[nr_probes].func = probe; + new[nr_probes].data = data; + new[nr_probes + 1].func = NULL; entry->refcount = nr_probes + 1; entry->funcs = new; debug_print_probes(entry); @@ -130,10 +134,11 @@ tracepoint_entry_add_probe(struct tracepoint_entry *entry, void *probe) } static void * -tracepoint_entry_remove_probe(struct tracepoint_entry *entry, void *probe) +tracepoint_entry_remove_probe(struct tracepoint_entry *entry, + void *probe, void *data) { int nr_probes = 0, nr_del = 0, i; - void **old, **new; + struct tracepoint_func *old, *new; old = entry->funcs; @@ -142,8 +147,10 @@ tracepoint_entry_remove_probe(struct tracepoint_entry *entry, void *probe) debug_print_probes(entry); /* (N -> M), (N > 1, M >= 0) probes */ - for (nr_probes = 0; old[nr_probes]; nr_probes++) { - if ((!probe || old[nr_probes] == probe)) + for (nr_probes = 0; old[nr_probes].func; nr_probes++) { + if (!probe || + (old[nr_probes].func == probe && + old[nr_probes].data == data)) nr_del++; } @@ -160,10 +167,11 @@ tracepoint_entry_remove_probe(struct tracepoint_entry *entry, void *probe) new = allocate_probes(nr_probes - nr_del + 1); if (new == NULL) return ERR_PTR(-ENOMEM); - for (i = 0; old[i]; i++) - if ((probe && old[i] != probe)) + for (i = 0; old[i].func; i++) + if (probe && + (old[i].func != probe || old[i].data != data)) new[j++] = old[i]; - new[nr_probes - nr_del] = NULL; + new[nr_probes - nr_del].func = NULL; entry->refcount = nr_probes - nr_del; entry->funcs = new; } @@ -256,7 +264,13 @@ static void set_tracepoint(struct tracepoint_entry **entry, * is used. */ rcu_assign_pointer(elem->funcs, (*entry)->funcs); - elem->state = active; + if (!elem->state && active) { + jump_label_enable(&elem->state); + elem->state = active; + } else if (elem->state && !active) { + jump_label_disable(&elem->state); + elem->state = active; + } } /* @@ -270,7 +284,10 @@ static void disable_tracepoint(struct tracepoint *elem) if (elem->unregfunc && elem->state) elem->unregfunc(); - elem->state = 0; + if (elem->state) { + jump_label_disable(&elem->state); + elem->state = 0; + } rcu_assign_pointer(elem->funcs, NULL); } @@ -315,18 +332,19 @@ static void tracepoint_update_probes(void) module_update_tracepoints(); } -static void *tracepoint_add_probe(const char *name, void *probe) +static struct tracepoint_func * +tracepoint_add_probe(const char *name, void *probe, void *data) { struct tracepoint_entry *entry; - void *old; + struct tracepoint_func *old; entry = get_tracepoint(name); if (!entry) { entry = add_tracepoint(name); if (IS_ERR(entry)) - return entry; + return (struct tracepoint_func *)entry; } - old = tracepoint_entry_add_probe(entry, probe); + old = tracepoint_entry_add_probe(entry, probe, data); if (IS_ERR(old) && !entry->refcount) remove_tracepoint(entry); return old; @@ -340,12 +358,12 @@ static void *tracepoint_add_probe(const char *name, void *probe) * Returns 0 if ok, error value on error. * The probe address must at least be aligned on the architecture pointer size. */ -int tracepoint_probe_register(const char *name, void *probe) +int tracepoint_probe_register(const char *name, void *probe, void *data) { - void *old; + struct tracepoint_func *old; mutex_lock(&tracepoints_mutex); - old = tracepoint_add_probe(name, probe); + old = tracepoint_add_probe(name, probe, data); mutex_unlock(&tracepoints_mutex); if (IS_ERR(old)) return PTR_ERR(old); @@ -356,15 +374,16 @@ int tracepoint_probe_register(const char *name, void *probe) } EXPORT_SYMBOL_GPL(tracepoint_probe_register); -static void *tracepoint_remove_probe(const char *name, void *probe) +static struct tracepoint_func * +tracepoint_remove_probe(const char *name, void *probe, void *data) { struct tracepoint_entry *entry; - void *old; + struct tracepoint_func *old; entry = get_tracepoint(name); if (!entry) return ERR_PTR(-ENOENT); - old = tracepoint_entry_remove_probe(entry, probe); + old = tracepoint_entry_remove_probe(entry, probe, data); if (IS_ERR(old)) return old; if (!entry->refcount) @@ -382,12 +401,12 @@ static void *tracepoint_remove_probe(const char *name, void *probe) * itself uses stop_machine(), which insures that every preempt disabled section * have finished. */ -int tracepoint_probe_unregister(const char *name, void *probe) +int tracepoint_probe_unregister(const char *name, void *probe, void *data) { - void *old; + struct tracepoint_func *old; mutex_lock(&tracepoints_mutex); - old = tracepoint_remove_probe(name, probe); + old = tracepoint_remove_probe(name, probe, data); mutex_unlock(&tracepoints_mutex); if (IS_ERR(old)) return PTR_ERR(old); @@ -418,12 +437,13 @@ static void tracepoint_add_old_probes(void *old) * * caller must call tracepoint_probe_update_all() */ -int tracepoint_probe_register_noupdate(const char *name, void *probe) +int tracepoint_probe_register_noupdate(const char *name, void *probe, + void *data) { - void *old; + struct tracepoint_func *old; mutex_lock(&tracepoints_mutex); - old = tracepoint_add_probe(name, probe); + old = tracepoint_add_probe(name, probe, data); if (IS_ERR(old)) { mutex_unlock(&tracepoints_mutex); return PTR_ERR(old); @@ -441,12 +461,13 @@ EXPORT_SYMBOL_GPL(tracepoint_probe_register_noupdate); * * caller must call tracepoint_probe_update_all() */ -int tracepoint_probe_unregister_noupdate(const char *name, void *probe) +int tracepoint_probe_unregister_noupdate(const char *name, void *probe, + void *data) { - void *old; + struct tracepoint_func *old; mutex_lock(&tracepoints_mutex); - old = tracepoint_remove_probe(name, probe); + old = tracepoint_remove_probe(name, probe, data); if (IS_ERR(old)) { mutex_unlock(&tracepoints_mutex); return PTR_ERR(old); diff --git a/kernel/tsacct.c b/kernel/tsacct.c index 0a67e041edf..24dc60d9fa1 100644 --- a/kernel/tsacct.c +++ b/kernel/tsacct.c @@ -63,12 +63,10 @@ void bacct_add_tsk(struct taskstats *stats, struct task_struct *tsk) stats->ac_ppid = pid_alive(tsk) ? rcu_dereference(tsk->real_parent)->tgid : 0; rcu_read_unlock(); - stats->ac_utime = cputime_to_msecs(tsk->utime) * USEC_PER_MSEC; - stats->ac_stime = cputime_to_msecs(tsk->stime) * USEC_PER_MSEC; - stats->ac_utimescaled = - cputime_to_msecs(tsk->utimescaled) * USEC_PER_MSEC; - stats->ac_stimescaled = - cputime_to_msecs(tsk->stimescaled) * USEC_PER_MSEC; + stats->ac_utime = cputime_to_usecs(tsk->utime); + stats->ac_stime = cputime_to_usecs(tsk->stime); + stats->ac_utimescaled = cputime_to_usecs(tsk->utimescaled); + stats->ac_stimescaled = cputime_to_usecs(tsk->stimescaled); stats->ac_minflt = tsk->min_flt; stats->ac_majflt = tsk->maj_flt; diff --git a/kernel/user.c b/kernel/user.c index 7e72614b736..5c598ca781d 100644 --- a/kernel/user.c +++ b/kernel/user.c @@ -91,6 +91,7 @@ static struct user_struct *uid_hash_find(uid_t uid, struct hlist_head *hashent) * upon function exit. */ static void free_user(struct user_struct *up, unsigned long flags) + __releases(&uidhash_lock) { uid_hash_remove(up); spin_unlock_irqrestore(&uidhash_lock, flags); @@ -157,6 +158,7 @@ struct user_struct *alloc_uid(struct user_namespace *ns, uid_t uid) spin_lock_irq(&uidhash_lock); up = uid_hash_find(uid, hashent); if (up) { + put_user_ns(ns); key_put(new->uid_keyring); key_put(new->session_keyring); kmem_cache_free(uid_cachep, new); diff --git a/kernel/user_namespace.c b/kernel/user_namespace.c index b2d70d38dff..9da289c34f2 100644 --- a/kernel/user_namespace.c +++ b/kernel/user_namespace.c @@ -9,8 +9,11 @@ #include <linux/nsproxy.h> #include <linux/slab.h> #include <linux/user_namespace.h> +#include <linux/highuid.h> #include <linux/cred.h> +static struct kmem_cache *user_ns_cachep __read_mostly; + /* * Create a new user namespace, deriving the creator from the user in the * passed credentials, and replacing that user with the new root user for the @@ -25,7 +28,7 @@ int create_user_ns(struct cred *new) struct user_struct *root_user; int n; - ns = kmalloc(sizeof(struct user_namespace), GFP_KERNEL); + ns = kmem_cache_alloc(user_ns_cachep, GFP_KERNEL); if (!ns) return -ENOMEM; @@ -37,7 +40,7 @@ int create_user_ns(struct cred *new) /* Alloc new root user. */ root_user = alloc_uid(ns, 0); if (!root_user) { - kfree(ns); + kmem_cache_free(user_ns_cachep, ns); return -ENOMEM; } @@ -70,7 +73,7 @@ static void free_user_ns_work(struct work_struct *work) struct user_namespace *ns = container_of(work, struct user_namespace, destroyer); free_uid(ns->creator); - kfree(ns); + kmem_cache_free(user_ns_cachep, ns); } void free_user_ns(struct kref *kref) @@ -82,3 +85,53 @@ void free_user_ns(struct kref *kref) schedule_work(&ns->destroyer); } EXPORT_SYMBOL(free_user_ns); + +uid_t user_ns_map_uid(struct user_namespace *to, const struct cred *cred, uid_t uid) +{ + struct user_namespace *tmp; + + if (likely(to == cred->user->user_ns)) + return uid; + + + /* Is cred->user the creator of the target user_ns + * or the creator of one of it's parents? + */ + for ( tmp = to; tmp != &init_user_ns; + tmp = tmp->creator->user_ns ) { + if (cred->user == tmp->creator) { + return (uid_t)0; + } + } + + /* No useful relationship so no mapping */ + return overflowuid; +} + +gid_t user_ns_map_gid(struct user_namespace *to, const struct cred *cred, gid_t gid) +{ + struct user_namespace *tmp; + + if (likely(to == cred->user->user_ns)) + return gid; + + /* Is cred->user the creator of the target user_ns + * or the creator of one of it's parents? + */ + for ( tmp = to; tmp != &init_user_ns; + tmp = tmp->creator->user_ns ) { + if (cred->user == tmp->creator) { + return (gid_t)0; + } + } + + /* No useful relationship so no mapping */ + return overflowgid; +} + +static __init int user_namespaces_init(void) +{ + user_ns_cachep = KMEM_CACHE(user_namespace, SLAB_PANIC); + return 0; +} +module_init(user_namespaces_init); diff --git a/kernel/wait.c b/kernel/wait.c index c4bd3d825f3..b0310eb6cc1 100644 --- a/kernel/wait.c +++ b/kernel/wait.c @@ -92,7 +92,7 @@ prepare_to_wait_exclusive(wait_queue_head_t *q, wait_queue_t *wait, int state) } EXPORT_SYMBOL(prepare_to_wait_exclusive); -/* +/** * finish_wait - clean up after waiting in a queue * @q: waitqueue waited on * @wait: wait descriptor @@ -127,11 +127,11 @@ void finish_wait(wait_queue_head_t *q, wait_queue_t *wait) } EXPORT_SYMBOL(finish_wait); -/* +/** * abort_exclusive_wait - abort exclusive waiting in a queue * @q: waitqueue waited on * @wait: wait descriptor - * @state: runstate of the waiter to be woken + * @mode: runstate of the waiter to be woken * @key: key to identify a wait bit queue or %NULL * * Sets current thread back to running state and removes diff --git a/kernel/watchdog.c b/kernel/watchdog.c new file mode 100644 index 00000000000..d7ebdf4cea9 --- /dev/null +++ b/kernel/watchdog.c @@ -0,0 +1,568 @@ +/* + * Detect hard and soft lockups on a system + * + * started by Don Zickus, Copyright (C) 2010 Red Hat, Inc. + * + * this code detects hard lockups: incidents in where on a CPU + * the kernel does not respond to anything except NMI. + * + * Note: Most of this code is borrowed heavily from softlockup.c, + * so thanks to Ingo for the initial implementation. + * Some chunks also taken from arch/x86/kernel/apic/nmi.c, thanks + * to those contributors as well. + */ + +#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/notifier.h> +#include <linux/module.h> +#include <linux/sysctl.h> + +#include <asm/irq_regs.h> +#include <linux/perf_event.h> + +int watchdog_enabled; +int __read_mostly softlockup_thresh = 60; + +static DEFINE_PER_CPU(unsigned long, watchdog_touch_ts); +static DEFINE_PER_CPU(struct task_struct *, softlockup_watchdog); +static DEFINE_PER_CPU(struct hrtimer, watchdog_hrtimer); +static DEFINE_PER_CPU(bool, softlockup_touch_sync); +static DEFINE_PER_CPU(bool, soft_watchdog_warn); +#ifdef CONFIG_HARDLOCKUP_DETECTOR +static DEFINE_PER_CPU(bool, hard_watchdog_warn); +static DEFINE_PER_CPU(bool, watchdog_nmi_touch); +static DEFINE_PER_CPU(unsigned long, hrtimer_interrupts); +static DEFINE_PER_CPU(unsigned long, hrtimer_interrupts_saved); +static DEFINE_PER_CPU(struct perf_event *, watchdog_ev); +#endif + +static int no_watchdog; + + +/* boot commands */ +/* + * Should we panic when a soft-lockup or hard-lockup occurs: + */ +#ifdef CONFIG_HARDLOCKUP_DETECTOR +static int hardlockup_panic; + +static int __init hardlockup_panic_setup(char *str) +{ + if (!strncmp(str, "panic", 5)) + hardlockup_panic = 1; + else if (!strncmp(str, "0", 1)) + no_watchdog = 1; + return 1; +} +__setup("nmi_watchdog=", hardlockup_panic_setup); +#endif + +unsigned int __read_mostly softlockup_panic = + CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC_VALUE; + +static int __init softlockup_panic_setup(char *str) +{ + softlockup_panic = simple_strtoul(str, NULL, 0); + + return 1; +} +__setup("softlockup_panic=", softlockup_panic_setup); + +static int __init nowatchdog_setup(char *str) +{ + no_watchdog = 1; + return 1; +} +__setup("nowatchdog", nowatchdog_setup); + +/* deprecated */ +static int __init nosoftlockup_setup(char *str) +{ + no_watchdog = 1; + return 1; +} +__setup("nosoftlockup", nosoftlockup_setup); +/* */ + + +/* + * Returns seconds, approximately. We don't need nanosecond + * resolution, and we don't need to waste time with a big divide when + * 2^30ns == 1.074s. + */ +static unsigned long get_timestamp(int this_cpu) +{ + return cpu_clock(this_cpu) >> 30LL; /* 2^30 ~= 10^9 */ +} + +static unsigned long get_sample_period(void) +{ + /* + * convert softlockup_thresh from seconds to ns + * the divide by 5 is to give hrtimer 5 chances to + * increment before the hardlockup detector generates + * a warning + */ + return softlockup_thresh / 5 * NSEC_PER_SEC; +} + +/* Commands for resetting the watchdog */ +static void __touch_watchdog(void) +{ + int this_cpu = smp_processor_id(); + + __this_cpu_write(watchdog_touch_ts, get_timestamp(this_cpu)); +} + +void touch_softlockup_watchdog(void) +{ + __this_cpu_write(watchdog_touch_ts, 0); +} +EXPORT_SYMBOL(touch_softlockup_watchdog); + +void touch_all_softlockup_watchdogs(void) +{ + int cpu; + + /* + * this is done lockless + * do we care if a 0 races with a timestamp? + * all it means is the softlock check starts one cycle later + */ + for_each_online_cpu(cpu) + per_cpu(watchdog_touch_ts, cpu) = 0; +} + +#ifdef CONFIG_HARDLOCKUP_DETECTOR +void touch_nmi_watchdog(void) +{ + if (watchdog_enabled) { + unsigned cpu; + + for_each_present_cpu(cpu) { + if (per_cpu(watchdog_nmi_touch, cpu) != true) + per_cpu(watchdog_nmi_touch, cpu) = true; + } + } + touch_softlockup_watchdog(); +} +EXPORT_SYMBOL(touch_nmi_watchdog); + +#endif + +void touch_softlockup_watchdog_sync(void) +{ + __raw_get_cpu_var(softlockup_touch_sync) = true; + __raw_get_cpu_var(watchdog_touch_ts) = 0; +} + +#ifdef CONFIG_HARDLOCKUP_DETECTOR +/* watchdog detector functions */ +static int is_hardlockup(void) +{ + unsigned long hrint = __this_cpu_read(hrtimer_interrupts); + + if (__this_cpu_read(hrtimer_interrupts_saved) == hrint) + return 1; + + __this_cpu_write(hrtimer_interrupts_saved, hrint); + return 0; +} +#endif + +static int is_softlockup(unsigned long touch_ts) +{ + unsigned long now = get_timestamp(smp_processor_id()); + + /* Warn about unreasonable delays: */ + if (time_after(now, touch_ts + softlockup_thresh)) + return now - touch_ts; + + return 0; +} + +#ifdef CONFIG_HARDLOCKUP_DETECTOR +static struct perf_event_attr wd_hw_attr = { + .type = PERF_TYPE_HARDWARE, + .config = PERF_COUNT_HW_CPU_CYCLES, + .size = sizeof(struct perf_event_attr), + .pinned = 1, + .disabled = 1, +}; + +/* Callback function for perf event subsystem */ +static void watchdog_overflow_callback(struct perf_event *event, int nmi, + struct perf_sample_data *data, + struct pt_regs *regs) +{ + /* Ensure the watchdog never gets throttled */ + event->hw.interrupts = 0; + + if (__this_cpu_read(watchdog_nmi_touch) == true) { + __this_cpu_write(watchdog_nmi_touch, false); + return; + } + + /* check for a hardlockup + * This is done by making sure our timer interrupt + * is incrementing. The timer interrupt should have + * fired multiple times before we overflow'd. If it hasn't + * then this is a good indication the cpu is stuck + */ + if (is_hardlockup()) { + int this_cpu = smp_processor_id(); + + /* only print hardlockups once */ + if (__this_cpu_read(hard_watchdog_warn) == true) + return; + + if (hardlockup_panic) + panic("Watchdog detected hard LOCKUP on cpu %d", this_cpu); + else + WARN(1, "Watchdog detected hard LOCKUP on cpu %d", this_cpu); + + __this_cpu_write(hard_watchdog_warn, true); + return; + } + + __this_cpu_write(hard_watchdog_warn, false); + return; +} +static void watchdog_interrupt_count(void) +{ + __this_cpu_inc(hrtimer_interrupts); +} +#else +static inline void watchdog_interrupt_count(void) { return; } +#endif /* CONFIG_HARDLOCKUP_DETECTOR */ + +/* watchdog kicker functions */ +static enum hrtimer_restart watchdog_timer_fn(struct hrtimer *hrtimer) +{ + unsigned long touch_ts = __this_cpu_read(watchdog_touch_ts); + struct pt_regs *regs = get_irq_regs(); + int duration; + + /* kick the hardlockup detector */ + watchdog_interrupt_count(); + + /* kick the softlockup detector */ + wake_up_process(__this_cpu_read(softlockup_watchdog)); + + /* .. and repeat */ + hrtimer_forward_now(hrtimer, ns_to_ktime(get_sample_period())); + + if (touch_ts == 0) { + if (unlikely(__this_cpu_read(softlockup_touch_sync))) { + /* + * If the time stamp was touched atomically + * make sure the scheduler tick is up to date. + */ + __this_cpu_write(softlockup_touch_sync, false); + sched_clock_tick(); + } + __touch_watchdog(); + return HRTIMER_RESTART; + } + + /* check for a softlockup + * This is done by making sure a high priority task is + * being scheduled. The task touches the watchdog to + * indicate it is getting cpu time. If it hasn't then + * this is a good indication some task is hogging the cpu + */ + duration = is_softlockup(touch_ts); + if (unlikely(duration)) { + /* only warn once */ + if (__this_cpu_read(soft_watchdog_warn) == true) + return HRTIMER_RESTART; + + printk(KERN_ERR "BUG: soft lockup - CPU#%d stuck for %us! [%s:%d]\n", + smp_processor_id(), duration, + current->comm, task_pid_nr(current)); + print_modules(); + print_irqtrace_events(current); + if (regs) + show_regs(regs); + else + dump_stack(); + + if (softlockup_panic) + panic("softlockup: hung tasks"); + __this_cpu_write(soft_watchdog_warn, true); + } else + __this_cpu_write(soft_watchdog_warn, false); + + return HRTIMER_RESTART; +} + + +/* + * The watchdog thread - touches the timestamp. + */ +static int watchdog(void *unused) +{ + static struct sched_param param = { .sched_priority = MAX_RT_PRIO-1 }; + struct hrtimer *hrtimer = &__raw_get_cpu_var(watchdog_hrtimer); + + sched_setscheduler(current, SCHED_FIFO, ¶m); + + /* initialize timestamp */ + __touch_watchdog(); + + /* kick off the timer for the hardlockup detector */ + /* done here because hrtimer_start can only pin to smp_processor_id() */ + hrtimer_start(hrtimer, ns_to_ktime(get_sample_period()), + HRTIMER_MODE_REL_PINNED); + + set_current_state(TASK_INTERRUPTIBLE); + /* + * Run briefly once per second to reset the softlockup timestamp. + * If this gets delayed for more than 60 seconds then the + * debug-printout triggers in watchdog_timer_fn(). + */ + while (!kthread_should_stop()) { + __touch_watchdog(); + schedule(); + + if (kthread_should_stop()) + break; + + set_current_state(TASK_INTERRUPTIBLE); + } + __set_current_state(TASK_RUNNING); + + return 0; +} + + +#ifdef CONFIG_HARDLOCKUP_DETECTOR +static int watchdog_nmi_enable(int cpu) +{ + struct perf_event_attr *wd_attr; + struct perf_event *event = per_cpu(watchdog_ev, cpu); + + /* is it already setup and enabled? */ + if (event && event->state > PERF_EVENT_STATE_OFF) + goto out; + + /* it is setup but not enabled */ + if (event != NULL) + goto out_enable; + + /* Try to register using hardware perf events */ + wd_attr = &wd_hw_attr; + wd_attr->sample_period = hw_nmi_get_sample_period(); + event = perf_event_create_kernel_counter(wd_attr, cpu, NULL, watchdog_overflow_callback); + if (!IS_ERR(event)) { + printk(KERN_INFO "NMI watchdog enabled, takes one hw-pmu counter.\n"); + goto out_save; + } + + printk(KERN_ERR "NMI watchdog disabled for cpu%i: unable to create perf event: %ld\n", + cpu, PTR_ERR(event)); + return PTR_ERR(event); + + /* success path */ +out_save: + per_cpu(watchdog_ev, cpu) = event; +out_enable: + perf_event_enable(per_cpu(watchdog_ev, cpu)); +out: + return 0; +} + +static void watchdog_nmi_disable(int cpu) +{ + struct perf_event *event = per_cpu(watchdog_ev, cpu); + + if (event) { + perf_event_disable(event); + per_cpu(watchdog_ev, cpu) = NULL; + + /* should be in cleanup, but blocks oprofile */ + perf_event_release_kernel(event); + } + return; +} +#else +static int watchdog_nmi_enable(int cpu) { return 0; } +static void watchdog_nmi_disable(int cpu) { return; } +#endif /* CONFIG_HARDLOCKUP_DETECTOR */ + +/* prepare/enable/disable routines */ +static int watchdog_prepare_cpu(int cpu) +{ + struct hrtimer *hrtimer = &per_cpu(watchdog_hrtimer, cpu); + + WARN_ON(per_cpu(softlockup_watchdog, cpu)); + hrtimer_init(hrtimer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); + hrtimer->function = watchdog_timer_fn; + + return 0; +} + +static int watchdog_enable(int cpu) +{ + struct task_struct *p = per_cpu(softlockup_watchdog, cpu); + int err; + + /* enable the perf event */ + err = watchdog_nmi_enable(cpu); + if (err) + return err; + + /* create the watchdog thread */ + if (!p) { + p = kthread_create(watchdog, (void *)(unsigned long)cpu, "watchdog/%d", cpu); + if (IS_ERR(p)) { + printk(KERN_ERR "softlockup watchdog for %i failed\n", cpu); + return PTR_ERR(p); + } + kthread_bind(p, cpu); + per_cpu(watchdog_touch_ts, cpu) = 0; + per_cpu(softlockup_watchdog, cpu) = p; + wake_up_process(p); + } + + /* if any cpu succeeds, watchdog is considered enabled for the system */ + watchdog_enabled = 1; + + return 0; +} + +static void watchdog_disable(int cpu) +{ + struct task_struct *p = per_cpu(softlockup_watchdog, cpu); + struct hrtimer *hrtimer = &per_cpu(watchdog_hrtimer, cpu); + + /* + * cancel the timer first to stop incrementing the stats + * and waking up the kthread + */ + hrtimer_cancel(hrtimer); + + /* disable the perf event */ + watchdog_nmi_disable(cpu); + + /* stop the watchdog thread */ + if (p) { + per_cpu(softlockup_watchdog, cpu) = NULL; + kthread_stop(p); + } +} + +static void watchdog_enable_all_cpus(void) +{ + int cpu; + int result = 0; + + for_each_online_cpu(cpu) + result += watchdog_enable(cpu); + + if (result) + printk(KERN_ERR "watchdog: failed to be enabled on some cpus\n"); + +} + +static void watchdog_disable_all_cpus(void) +{ + int cpu; + + if (no_watchdog) + return; + + for_each_online_cpu(cpu) + watchdog_disable(cpu); + + /* if all watchdogs are disabled, then they are disabled for the system */ + watchdog_enabled = 0; +} + + +/* sysctl functions */ +#ifdef CONFIG_SYSCTL +/* + * proc handler for /proc/sys/kernel/nmi_watchdog + */ + +int proc_dowatchdog_enabled(struct ctl_table *table, int write, + void __user *buffer, size_t *length, loff_t *ppos) +{ + proc_dointvec(table, write, buffer, length, ppos); + + if (watchdog_enabled) + watchdog_enable_all_cpus(); + else + watchdog_disable_all_cpus(); + return 0; +} + +int proc_dowatchdog_thresh(struct ctl_table *table, int write, + void __user *buffer, + size_t *lenp, loff_t *ppos) +{ + return proc_dointvec_minmax(table, write, buffer, lenp, ppos); +} +#endif /* CONFIG_SYSCTL */ + + +/* + * Create/destroy watchdog threads as CPUs come and go: + */ +static int __cpuinit +cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu) +{ + int hotcpu = (unsigned long)hcpu; + int err = 0; + + switch (action) { + case CPU_UP_PREPARE: + case CPU_UP_PREPARE_FROZEN: + err = watchdog_prepare_cpu(hotcpu); + break; + case CPU_ONLINE: + case CPU_ONLINE_FROZEN: + err = watchdog_enable(hotcpu); + break; +#ifdef CONFIG_HOTPLUG_CPU + case CPU_UP_CANCELED: + case CPU_UP_CANCELED_FROZEN: + watchdog_disable(hotcpu); + break; + case CPU_DEAD: + case CPU_DEAD_FROZEN: + watchdog_disable(hotcpu); + break; +#endif /* CONFIG_HOTPLUG_CPU */ + } + return notifier_from_errno(err); +} + +static struct notifier_block __cpuinitdata cpu_nfb = { + .notifier_call = cpu_callback +}; + +void __init lockup_detector_init(void) +{ + void *cpu = (void *)(long)smp_processor_id(); + int err; + + if (no_watchdog) + return; + + err = cpu_callback(&cpu_nfb, CPU_UP_PREPARE, cpu); + WARN_ON(notifier_to_errno(err)); + + cpu_callback(&cpu_nfb, CPU_ONLINE, cpu); + register_cpu_notifier(&cpu_nfb); + + return; +} diff --git a/kernel/workqueue.c b/kernel/workqueue.c index 77dabbf64b8..11869faa681 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -1,19 +1,26 @@ /* - * linux/kernel/workqueue.c + * kernel/workqueue.c - generic async execution with shared worker pool * - * Generic mechanism for defining kernel helper threads for running - * arbitrary tasks in process context. + * Copyright (C) 2002 Ingo Molnar * - * Started by Ingo Molnar, Copyright (C) 2002 + * Derived from the taskqueue/keventd code by: + * David Woodhouse <dwmw2@infradead.org> + * Andrew Morton + * Kai Petzke <wpp@marie.physik.tu-berlin.de> + * Theodore Ts'o <tytso@mit.edu> * - * Derived from the taskqueue/keventd code by: + * Made to use alloc_percpu by Christoph Lameter. * - * David Woodhouse <dwmw2@infradead.org> - * Andrew Morton - * Kai Petzke <wpp@marie.physik.tu-berlin.de> - * Theodore Ts'o <tytso@mit.edu> + * Copyright (C) 2010 SUSE Linux Products GmbH + * Copyright (C) 2010 Tejun Heo <tj@kernel.org> * - * Made to use alloc_percpu by Christoph Lameter. + * This is the generic async execution mechanism. Work items as are + * executed in process context. The worker pool is shared and + * automatically managed. There is one worker pool for each CPU and + * one extra for works which are better served by workers which are + * not bound to any specific CPU. + * + * Please read Documentation/workqueue.txt for details. */ #include <linux/module.h> @@ -33,41 +40,276 @@ #include <linux/kallsyms.h> #include <linux/debug_locks.h> #include <linux/lockdep.h> -#define CREATE_TRACE_POINTS -#include <trace/events/workqueue.h> +#include <linux/idr.h> + +#include "workqueue_sched.h" + +enum { + /* global_cwq flags */ + GCWQ_MANAGE_WORKERS = 1 << 0, /* need to manage workers */ + GCWQ_MANAGING_WORKERS = 1 << 1, /* managing workers */ + GCWQ_DISASSOCIATED = 1 << 2, /* cpu can't serve workers */ + GCWQ_FREEZING = 1 << 3, /* freeze in progress */ + GCWQ_HIGHPRI_PENDING = 1 << 4, /* highpri works on queue */ + + /* worker flags */ + WORKER_STARTED = 1 << 0, /* started */ + WORKER_DIE = 1 << 1, /* die die die */ + WORKER_IDLE = 1 << 2, /* is idle */ + WORKER_PREP = 1 << 3, /* preparing to run works */ + WORKER_ROGUE = 1 << 4, /* not bound to any cpu */ + WORKER_REBIND = 1 << 5, /* mom is home, come back */ + WORKER_CPU_INTENSIVE = 1 << 6, /* cpu intensive */ + WORKER_UNBOUND = 1 << 7, /* worker is unbound */ + + WORKER_NOT_RUNNING = WORKER_PREP | WORKER_ROGUE | WORKER_REBIND | + WORKER_CPU_INTENSIVE | WORKER_UNBOUND, + + /* gcwq->trustee_state */ + TRUSTEE_START = 0, /* start */ + TRUSTEE_IN_CHARGE = 1, /* trustee in charge of gcwq */ + TRUSTEE_BUTCHER = 2, /* butcher workers */ + TRUSTEE_RELEASE = 3, /* release workers */ + TRUSTEE_DONE = 4, /* trustee is done */ + + BUSY_WORKER_HASH_ORDER = 6, /* 64 pointers */ + BUSY_WORKER_HASH_SIZE = 1 << BUSY_WORKER_HASH_ORDER, + BUSY_WORKER_HASH_MASK = BUSY_WORKER_HASH_SIZE - 1, + + MAX_IDLE_WORKERS_RATIO = 4, /* 1/4 of busy can be idle */ + IDLE_WORKER_TIMEOUT = 300 * HZ, /* keep idle ones for 5 mins */ + + MAYDAY_INITIAL_TIMEOUT = HZ / 100, /* call for help after 10ms */ + MAYDAY_INTERVAL = HZ / 10, /* and then every 100ms */ + CREATE_COOLDOWN = HZ, /* time to breath after fail */ + TRUSTEE_COOLDOWN = HZ / 10, /* for trustee draining */ + + /* + * Rescue workers are used only on emergencies and shared by + * all cpus. Give -20. + */ + RESCUER_NICE_LEVEL = -20, +}; /* - * The per-CPU workqueue (if single thread, we always use the first - * possible cpu). + * Structure fields follow one of the following exclusion rules. + * + * I: Modifiable by initialization/destruction paths and read-only for + * everyone else. + * + * P: Preemption protected. Disabling preemption is enough and should + * only be modified and accessed from the local cpu. + * + * L: gcwq->lock protected. Access with gcwq->lock held. + * + * X: During normal operation, modification requires gcwq->lock and + * should be done only from local cpu. Either disabling preemption + * on local cpu or grabbing gcwq->lock is enough for read access. + * If GCWQ_DISASSOCIATED is set, it's identical to L. + * + * F: wq->flush_mutex protected. + * + * W: workqueue_lock protected. */ -struct cpu_workqueue_struct { - spinlock_t lock; +struct global_cwq; + +/* + * The poor guys doing the actual heavy lifting. All on-duty workers + * are either serving the manager role, on idle list or on busy hash. + */ +struct worker { + /* on idle list while idle, on busy hash table while busy */ + union { + struct list_head entry; /* L: while idle */ + struct hlist_node hentry; /* L: while busy */ + }; - struct list_head worklist; - wait_queue_head_t more_work; - struct work_struct *current_work; + struct work_struct *current_work; /* L: work being processed */ + struct cpu_workqueue_struct *current_cwq; /* L: current_work's cwq */ + struct list_head scheduled; /* L: scheduled works */ + struct task_struct *task; /* I: worker task */ + struct global_cwq *gcwq; /* I: the associated gcwq */ + /* 64 bytes boundary on 64bit, 32 on 32bit */ + unsigned long last_active; /* L: last active timestamp */ + unsigned int flags; /* X: flags */ + int id; /* I: worker id */ + struct work_struct rebind_work; /* L: rebind worker to cpu */ +}; - struct workqueue_struct *wq; - struct task_struct *thread; -} ____cacheline_aligned; +/* + * Global per-cpu workqueue. There's one and only one for each cpu + * and all works are queued and processed here regardless of their + * target workqueues. + */ +struct global_cwq { + spinlock_t lock; /* the gcwq lock */ + struct list_head worklist; /* L: list of pending works */ + unsigned int cpu; /* I: the associated cpu */ + unsigned int flags; /* L: GCWQ_* flags */ + + int nr_workers; /* L: total number of workers */ + int nr_idle; /* L: currently idle ones */ + + /* workers are chained either in the idle_list or busy_hash */ + struct list_head idle_list; /* X: list of idle workers */ + struct hlist_head busy_hash[BUSY_WORKER_HASH_SIZE]; + /* L: hash of busy workers */ + + struct timer_list idle_timer; /* L: worker idle timeout */ + struct timer_list mayday_timer; /* L: SOS timer for dworkers */ + + struct ida worker_ida; /* L: for worker IDs */ + + struct task_struct *trustee; /* L: for gcwq shutdown */ + unsigned int trustee_state; /* L: trustee state */ + wait_queue_head_t trustee_wait; /* trustee wait */ + struct worker *first_idle; /* L: first idle worker */ +} ____cacheline_aligned_in_smp; + +/* + * The per-CPU workqueue. The lower WORK_STRUCT_FLAG_BITS of + * work_struct->data are used for flags and thus cwqs need to be + * aligned at two's power of the number of flag bits. + */ +struct cpu_workqueue_struct { + struct global_cwq *gcwq; /* I: the associated gcwq */ + struct workqueue_struct *wq; /* I: the owning workqueue */ + int work_color; /* L: current color */ + int flush_color; /* L: flushing color */ + int nr_in_flight[WORK_NR_COLORS]; + /* L: nr of in_flight works */ + int nr_active; /* L: nr of active works */ + int max_active; /* L: max active works */ + struct list_head delayed_works; /* L: delayed works */ +}; + +/* + * Structure used to wait for workqueue flush. + */ +struct wq_flusher { + struct list_head list; /* F: list of flushers */ + int flush_color; /* F: flush color waiting for */ + struct completion done; /* flush completion */ +}; + +/* + * All cpumasks are assumed to be always set on UP and thus can't be + * used to determine whether there's something to be done. + */ +#ifdef CONFIG_SMP +typedef cpumask_var_t mayday_mask_t; +#define mayday_test_and_set_cpu(cpu, mask) \ + cpumask_test_and_set_cpu((cpu), (mask)) +#define mayday_clear_cpu(cpu, mask) cpumask_clear_cpu((cpu), (mask)) +#define for_each_mayday_cpu(cpu, mask) for_each_cpu((cpu), (mask)) +#define alloc_mayday_mask(maskp, gfp) zalloc_cpumask_var((maskp), (gfp)) +#define free_mayday_mask(mask) free_cpumask_var((mask)) +#else +typedef unsigned long mayday_mask_t; +#define mayday_test_and_set_cpu(cpu, mask) test_and_set_bit(0, &(mask)) +#define mayday_clear_cpu(cpu, mask) clear_bit(0, &(mask)) +#define for_each_mayday_cpu(cpu, mask) if ((cpu) = 0, (mask)) +#define alloc_mayday_mask(maskp, gfp) true +#define free_mayday_mask(mask) do { } while (0) +#endif /* * The externally visible workqueue abstraction is an array of * per-CPU workqueues: */ struct workqueue_struct { - struct cpu_workqueue_struct *cpu_wq; - struct list_head list; - const char *name; - int singlethread; - int freezeable; /* Freeze threads during suspend */ - int rt; + unsigned int flags; /* I: WQ_* flags */ + union { + struct cpu_workqueue_struct __percpu *pcpu; + struct cpu_workqueue_struct *single; + unsigned long v; + } cpu_wq; /* I: cwq's */ + struct list_head list; /* W: list of all workqueues */ + + struct mutex flush_mutex; /* protects wq flushing */ + int work_color; /* F: current work color */ + int flush_color; /* F: current flush color */ + atomic_t nr_cwqs_to_flush; /* flush in progress */ + struct wq_flusher *first_flusher; /* F: first flusher */ + struct list_head flusher_queue; /* F: flush waiters */ + struct list_head flusher_overflow; /* F: flush overflow list */ + + mayday_mask_t mayday_mask; /* cpus requesting rescue */ + struct worker *rescuer; /* I: rescue worker */ + + int saved_max_active; /* W: saved cwq max_active */ + const char *name; /* I: workqueue name */ #ifdef CONFIG_LOCKDEP - struct lockdep_map lockdep_map; + struct lockdep_map lockdep_map; #endif }; +struct workqueue_struct *system_wq __read_mostly; +struct workqueue_struct *system_long_wq __read_mostly; +struct workqueue_struct *system_nrt_wq __read_mostly; +struct workqueue_struct *system_unbound_wq __read_mostly; +EXPORT_SYMBOL_GPL(system_wq); +EXPORT_SYMBOL_GPL(system_long_wq); +EXPORT_SYMBOL_GPL(system_nrt_wq); +EXPORT_SYMBOL_GPL(system_unbound_wq); + +#define CREATE_TRACE_POINTS +#include <trace/events/workqueue.h> + +#define for_each_busy_worker(worker, i, pos, gcwq) \ + for (i = 0; i < BUSY_WORKER_HASH_SIZE; i++) \ + hlist_for_each_entry(worker, pos, &gcwq->busy_hash[i], hentry) + +static inline int __next_gcwq_cpu(int cpu, const struct cpumask *mask, + unsigned int sw) +{ + if (cpu < nr_cpu_ids) { + if (sw & 1) { + cpu = cpumask_next(cpu, mask); + if (cpu < nr_cpu_ids) + return cpu; + } + if (sw & 2) + return WORK_CPU_UNBOUND; + } + return WORK_CPU_NONE; +} + +static inline int __next_wq_cpu(int cpu, const struct cpumask *mask, + struct workqueue_struct *wq) +{ + return __next_gcwq_cpu(cpu, mask, !(wq->flags & WQ_UNBOUND) ? 1 : 2); +} + +/* + * CPU iterators + * + * An extra gcwq is defined for an invalid cpu number + * (WORK_CPU_UNBOUND) to host workqueues which are not bound to any + * specific CPU. The following iterators are similar to + * for_each_*_cpu() iterators but also considers the unbound gcwq. + * + * for_each_gcwq_cpu() : possible CPUs + WORK_CPU_UNBOUND + * for_each_online_gcwq_cpu() : online CPUs + WORK_CPU_UNBOUND + * for_each_cwq_cpu() : possible CPUs for bound workqueues, + * WORK_CPU_UNBOUND for unbound workqueues + */ +#define for_each_gcwq_cpu(cpu) \ + for ((cpu) = __next_gcwq_cpu(-1, cpu_possible_mask, 3); \ + (cpu) < WORK_CPU_NONE; \ + (cpu) = __next_gcwq_cpu((cpu), cpu_possible_mask, 3)) + +#define for_each_online_gcwq_cpu(cpu) \ + for ((cpu) = __next_gcwq_cpu(-1, cpu_online_mask, 3); \ + (cpu) < WORK_CPU_NONE; \ + (cpu) = __next_gcwq_cpu((cpu), cpu_online_mask, 3)) + +#define for_each_cwq_cpu(cpu, wq) \ + for ((cpu) = __next_wq_cpu(-1, cpu_possible_mask, (wq)); \ + (cpu) < WORK_CPU_NONE; \ + (cpu) = __next_wq_cpu((cpu), cpu_possible_mask, (wq))) + #ifdef CONFIG_DEBUG_OBJECTS_WORK static struct debug_obj_descr work_debug_descr; @@ -107,7 +349,7 @@ static int work_fixup_activate(void *addr, enum debug_obj_state state) * statically initialized. We just make sure that it * is tracked in the object tracker. */ - if (test_bit(WORK_STRUCT_STATIC, work_data_bits(work))) { + if (test_bit(WORK_STRUCT_STATIC_BIT, work_data_bits(work))) { debug_object_init(work, &work_debug_descr); debug_object_activate(work, &work_debug_descr); return 0; @@ -181,94 +423,624 @@ static inline void debug_work_deactivate(struct work_struct *work) { } /* Serializes the accesses to the list of workqueues. */ static DEFINE_SPINLOCK(workqueue_lock); static LIST_HEAD(workqueues); +static bool workqueue_freezing; /* W: have wqs started freezing? */ + +/* + * The almighty global cpu workqueues. nr_running is the only field + * which is expected to be used frequently by other cpus via + * try_to_wake_up(). Put it in a separate cacheline. + */ +static DEFINE_PER_CPU(struct global_cwq, global_cwq); +static DEFINE_PER_CPU_SHARED_ALIGNED(atomic_t, gcwq_nr_running); -static int singlethread_cpu __read_mostly; -static const struct cpumask *cpu_singlethread_map __read_mostly; /* - * _cpu_down() first removes CPU from cpu_online_map, then CPU_DEAD - * flushes cwq->worklist. This means that flush_workqueue/wait_on_work - * which comes in between can't use for_each_online_cpu(). We could - * use cpu_possible_map, the cpumask below is more a documentation - * than optimization. + * Global cpu workqueue and nr_running counter for unbound gcwq. The + * gcwq is always online, has GCWQ_DISASSOCIATED set, and all its + * workers have WORKER_UNBOUND set. */ -static cpumask_var_t cpu_populated_map __read_mostly; +static struct global_cwq unbound_global_cwq; +static atomic_t unbound_gcwq_nr_running = ATOMIC_INIT(0); /* always 0 */ -/* If it's single threaded, it isn't in the list of workqueues. */ -static inline int is_wq_single_threaded(struct workqueue_struct *wq) +static int worker_thread(void *__worker); + +static struct global_cwq *get_gcwq(unsigned int cpu) { - return wq->singlethread; + if (cpu != WORK_CPU_UNBOUND) + return &per_cpu(global_cwq, cpu); + else + return &unbound_global_cwq; } -static const struct cpumask *wq_cpu_map(struct workqueue_struct *wq) +static atomic_t *get_gcwq_nr_running(unsigned int cpu) { - return is_wq_single_threaded(wq) - ? cpu_singlethread_map : cpu_populated_map; + if (cpu != WORK_CPU_UNBOUND) + return &per_cpu(gcwq_nr_running, cpu); + else + return &unbound_gcwq_nr_running; } -static -struct cpu_workqueue_struct *wq_per_cpu(struct workqueue_struct *wq, int cpu) +static struct cpu_workqueue_struct *get_cwq(unsigned int cpu, + struct workqueue_struct *wq) { - if (unlikely(is_wq_single_threaded(wq))) - cpu = singlethread_cpu; - return per_cpu_ptr(wq->cpu_wq, cpu); + if (!(wq->flags & WQ_UNBOUND)) { + if (likely(cpu < nr_cpu_ids)) { +#ifdef CONFIG_SMP + return per_cpu_ptr(wq->cpu_wq.pcpu, cpu); +#else + return wq->cpu_wq.single; +#endif + } + } else if (likely(cpu == WORK_CPU_UNBOUND)) + return wq->cpu_wq.single; + return NULL; +} + +static unsigned int work_color_to_flags(int color) +{ + return color << WORK_STRUCT_COLOR_SHIFT; +} + +static int get_work_color(struct work_struct *work) +{ + return (*work_data_bits(work) >> WORK_STRUCT_COLOR_SHIFT) & + ((1 << WORK_STRUCT_COLOR_BITS) - 1); +} + +static int work_next_color(int color) +{ + return (color + 1) % WORK_NR_COLORS; } /* - * Set the workqueue on which a work item is to be run - * - Must *only* be called if the pending flag is set + * A work's data points to the cwq with WORK_STRUCT_CWQ set while the + * work is on queue. Once execution starts, WORK_STRUCT_CWQ is + * cleared and the work data contains the cpu number it was last on. + * + * set_work_{cwq|cpu}() and clear_work_data() can be used to set the + * cwq, cpu or clear work->data. These functions should only be + * called while the work is owned - ie. while the PENDING bit is set. + * + * get_work_[g]cwq() can be used to obtain the gcwq or cwq + * corresponding to a work. gcwq is available once the work has been + * queued anywhere after initialization. cwq is available only from + * queueing until execution starts. */ -static inline void set_wq_data(struct work_struct *work, - struct cpu_workqueue_struct *cwq) +static inline void set_work_data(struct work_struct *work, unsigned long data, + unsigned long flags) { - unsigned long new; - BUG_ON(!work_pending(work)); + atomic_long_set(&work->data, data | flags | work_static(work)); +} - new = (unsigned long) cwq | (1UL << WORK_STRUCT_PENDING); - new |= WORK_STRUCT_FLAG_MASK & *work_data_bits(work); - atomic_long_set(&work->data, new); +static void set_work_cwq(struct work_struct *work, + struct cpu_workqueue_struct *cwq, + unsigned long extra_flags) +{ + set_work_data(work, (unsigned long)cwq, + WORK_STRUCT_PENDING | WORK_STRUCT_CWQ | extra_flags); +} + +static void set_work_cpu(struct work_struct *work, unsigned int cpu) +{ + set_work_data(work, cpu << WORK_STRUCT_FLAG_BITS, WORK_STRUCT_PENDING); +} + +static void clear_work_data(struct work_struct *work) +{ + set_work_data(work, WORK_STRUCT_NO_CPU, 0); +} + +static struct cpu_workqueue_struct *get_work_cwq(struct work_struct *work) +{ + unsigned long data = atomic_long_read(&work->data); + + if (data & WORK_STRUCT_CWQ) + return (void *)(data & WORK_STRUCT_WQ_DATA_MASK); + else + return NULL; +} + +static struct global_cwq *get_work_gcwq(struct work_struct *work) +{ + unsigned long data = atomic_long_read(&work->data); + unsigned int cpu; + + if (data & WORK_STRUCT_CWQ) + return ((struct cpu_workqueue_struct *) + (data & WORK_STRUCT_WQ_DATA_MASK))->gcwq; + + cpu = data >> WORK_STRUCT_FLAG_BITS; + if (cpu == WORK_CPU_NONE) + return NULL; + + BUG_ON(cpu >= nr_cpu_ids && cpu != WORK_CPU_UNBOUND); + return get_gcwq(cpu); } /* - * Clear WORK_STRUCT_PENDING and the workqueue on which it was queued. + * Policy functions. These define the policies on how the global + * worker pool is managed. Unless noted otherwise, these functions + * assume that they're being called with gcwq->lock held. */ -static inline void clear_wq_data(struct work_struct *work) + +static bool __need_more_worker(struct global_cwq *gcwq) { - unsigned long flags = *work_data_bits(work) & - (1UL << WORK_STRUCT_STATIC); - atomic_long_set(&work->data, flags); + return !atomic_read(get_gcwq_nr_running(gcwq->cpu)) || + gcwq->flags & GCWQ_HIGHPRI_PENDING; } -static inline -struct cpu_workqueue_struct *get_wq_data(struct work_struct *work) +/* + * Need to wake up a worker? Called from anything but currently + * running workers. + */ +static bool need_more_worker(struct global_cwq *gcwq) { - return (void *) (atomic_long_read(&work->data) & WORK_STRUCT_WQ_DATA_MASK); + return !list_empty(&gcwq->worklist) && __need_more_worker(gcwq); } +/* Can I start working? Called from busy but !running workers. */ +static bool may_start_working(struct global_cwq *gcwq) +{ + return gcwq->nr_idle; +} + +/* Do I need to keep working? Called from currently running workers. */ +static bool keep_working(struct global_cwq *gcwq) +{ + atomic_t *nr_running = get_gcwq_nr_running(gcwq->cpu); + + return !list_empty(&gcwq->worklist) && + (atomic_read(nr_running) <= 1 || + gcwq->flags & GCWQ_HIGHPRI_PENDING); +} + +/* Do we need a new worker? Called from manager. */ +static bool need_to_create_worker(struct global_cwq *gcwq) +{ + return need_more_worker(gcwq) && !may_start_working(gcwq); +} + +/* Do I need to be the manager? */ +static bool need_to_manage_workers(struct global_cwq *gcwq) +{ + return need_to_create_worker(gcwq) || gcwq->flags & GCWQ_MANAGE_WORKERS; +} + +/* Do we have too many workers and should some go away? */ +static bool too_many_workers(struct global_cwq *gcwq) +{ + bool managing = gcwq->flags & GCWQ_MANAGING_WORKERS; + int nr_idle = gcwq->nr_idle + managing; /* manager is considered idle */ + int nr_busy = gcwq->nr_workers - nr_idle; + + return nr_idle > 2 && (nr_idle - 2) * MAX_IDLE_WORKERS_RATIO >= nr_busy; +} + +/* + * Wake up functions. + */ + +/* Return the first worker. Safe with preemption disabled */ +static struct worker *first_worker(struct global_cwq *gcwq) +{ + if (unlikely(list_empty(&gcwq->idle_list))) + return NULL; + + return list_first_entry(&gcwq->idle_list, struct worker, entry); +} + +/** + * wake_up_worker - wake up an idle worker + * @gcwq: gcwq to wake worker for + * + * Wake up the first idle worker of @gcwq. + * + * CONTEXT: + * spin_lock_irq(gcwq->lock). + */ +static void wake_up_worker(struct global_cwq *gcwq) +{ + struct worker *worker = first_worker(gcwq); + + if (likely(worker)) + wake_up_process(worker->task); +} + +/** + * wq_worker_waking_up - a worker is waking up + * @task: task waking up + * @cpu: CPU @task is waking up to + * + * This function is called during try_to_wake_up() when a worker is + * being awoken. + * + * CONTEXT: + * spin_lock_irq(rq->lock) + */ +void wq_worker_waking_up(struct task_struct *task, unsigned int cpu) +{ + struct worker *worker = kthread_data(task); + + if (!(worker->flags & WORKER_NOT_RUNNING)) + atomic_inc(get_gcwq_nr_running(cpu)); +} + +/** + * wq_worker_sleeping - a worker is going to sleep + * @task: task going to sleep + * @cpu: CPU in question, must be the current CPU number + * + * This function is called during schedule() when a busy worker is + * going to sleep. Worker on the same cpu can be woken up by + * returning pointer to its task. + * + * CONTEXT: + * spin_lock_irq(rq->lock) + * + * RETURNS: + * Worker task on @cpu to wake up, %NULL if none. + */ +struct task_struct *wq_worker_sleeping(struct task_struct *task, + unsigned int cpu) +{ + struct worker *worker = kthread_data(task), *to_wakeup = NULL; + struct global_cwq *gcwq = get_gcwq(cpu); + atomic_t *nr_running = get_gcwq_nr_running(cpu); + + if (worker->flags & WORKER_NOT_RUNNING) + return NULL; + + /* this can only happen on the local cpu */ + BUG_ON(cpu != raw_smp_processor_id()); + + /* + * The counterpart of the following dec_and_test, implied mb, + * worklist not empty test sequence is in insert_work(). + * Please read comment there. + * + * NOT_RUNNING is clear. This means that trustee is not in + * charge and we're running on the local cpu w/ rq lock held + * and preemption disabled, which in turn means that none else + * could be manipulating idle_list, so dereferencing idle_list + * without gcwq lock is safe. + */ + if (atomic_dec_and_test(nr_running) && !list_empty(&gcwq->worklist)) + to_wakeup = first_worker(gcwq); + return to_wakeup ? to_wakeup->task : NULL; +} + +/** + * worker_set_flags - set worker flags and adjust nr_running accordingly + * @worker: self + * @flags: flags to set + * @wakeup: wakeup an idle worker if necessary + * + * Set @flags in @worker->flags and adjust nr_running accordingly. If + * nr_running becomes zero and @wakeup is %true, an idle worker is + * woken up. + * + * CONTEXT: + * spin_lock_irq(gcwq->lock) + */ +static inline void worker_set_flags(struct worker *worker, unsigned int flags, + bool wakeup) +{ + struct global_cwq *gcwq = worker->gcwq; + + WARN_ON_ONCE(worker->task != current); + + /* + * If transitioning into NOT_RUNNING, adjust nr_running and + * wake up an idle worker as necessary if requested by + * @wakeup. + */ + if ((flags & WORKER_NOT_RUNNING) && + !(worker->flags & WORKER_NOT_RUNNING)) { + atomic_t *nr_running = get_gcwq_nr_running(gcwq->cpu); + + if (wakeup) { + if (atomic_dec_and_test(nr_running) && + !list_empty(&gcwq->worklist)) + wake_up_worker(gcwq); + } else + atomic_dec(nr_running); + } + + worker->flags |= flags; +} + +/** + * worker_clr_flags - clear worker flags and adjust nr_running accordingly + * @worker: self + * @flags: flags to clear + * + * Clear @flags in @worker->flags and adjust nr_running accordingly. + * + * CONTEXT: + * spin_lock_irq(gcwq->lock) + */ +static inline void worker_clr_flags(struct worker *worker, unsigned int flags) +{ + struct global_cwq *gcwq = worker->gcwq; + unsigned int oflags = worker->flags; + + WARN_ON_ONCE(worker->task != current); + + worker->flags &= ~flags; + + /* + * If transitioning out of NOT_RUNNING, increment nr_running. Note + * that the nested NOT_RUNNING is not a noop. NOT_RUNNING is mask + * of multiple flags, not a single flag. + */ + if ((flags & WORKER_NOT_RUNNING) && (oflags & WORKER_NOT_RUNNING)) + if (!(worker->flags & WORKER_NOT_RUNNING)) + atomic_inc(get_gcwq_nr_running(gcwq->cpu)); +} + +/** + * busy_worker_head - return the busy hash head for a work + * @gcwq: gcwq of interest + * @work: work to be hashed + * + * Return hash head of @gcwq for @work. + * + * CONTEXT: + * spin_lock_irq(gcwq->lock). + * + * RETURNS: + * Pointer to the hash head. + */ +static struct hlist_head *busy_worker_head(struct global_cwq *gcwq, + struct work_struct *work) +{ + const int base_shift = ilog2(sizeof(struct work_struct)); + unsigned long v = (unsigned long)work; + + /* simple shift and fold hash, do we need something better? */ + v >>= base_shift; + v += v >> BUSY_WORKER_HASH_ORDER; + v &= BUSY_WORKER_HASH_MASK; + + return &gcwq->busy_hash[v]; +} + +/** + * __find_worker_executing_work - find worker which is executing a work + * @gcwq: gcwq of interest + * @bwh: hash head as returned by busy_worker_head() + * @work: work to find worker for + * + * Find a worker which is executing @work on @gcwq. @bwh should be + * the hash head obtained by calling busy_worker_head() with the same + * work. + * + * CONTEXT: + * spin_lock_irq(gcwq->lock). + * + * RETURNS: + * Pointer to worker which is executing @work if found, NULL + * otherwise. + */ +static struct worker *__find_worker_executing_work(struct global_cwq *gcwq, + struct hlist_head *bwh, + struct work_struct *work) +{ + struct worker *worker; + struct hlist_node *tmp; + + hlist_for_each_entry(worker, tmp, bwh, hentry) + if (worker->current_work == work) + return worker; + return NULL; +} + +/** + * find_worker_executing_work - find worker which is executing a work + * @gcwq: gcwq of interest + * @work: work to find worker for + * + * Find a worker which is executing @work on @gcwq. This function is + * identical to __find_worker_executing_work() except that this + * function calculates @bwh itself. + * + * CONTEXT: + * spin_lock_irq(gcwq->lock). + * + * RETURNS: + * Pointer to worker which is executing @work if found, NULL + * otherwise. + */ +static struct worker *find_worker_executing_work(struct global_cwq *gcwq, + struct work_struct *work) +{ + return __find_worker_executing_work(gcwq, busy_worker_head(gcwq, work), + work); +} + +/** + * gcwq_determine_ins_pos - find insertion position + * @gcwq: gcwq of interest + * @cwq: cwq a work is being queued for + * + * A work for @cwq is about to be queued on @gcwq, determine insertion + * position for the work. If @cwq is for HIGHPRI wq, the work is + * queued at the head of the queue but in FIFO order with respect to + * other HIGHPRI works; otherwise, at the end of the queue. This + * function also sets GCWQ_HIGHPRI_PENDING flag to hint @gcwq that + * there are HIGHPRI works pending. + * + * CONTEXT: + * spin_lock_irq(gcwq->lock). + * + * RETURNS: + * Pointer to inserstion position. + */ +static inline struct list_head *gcwq_determine_ins_pos(struct global_cwq *gcwq, + struct cpu_workqueue_struct *cwq) +{ + struct work_struct *twork; + + if (likely(!(cwq->wq->flags & WQ_HIGHPRI))) + return &gcwq->worklist; + + list_for_each_entry(twork, &gcwq->worklist, entry) { + struct cpu_workqueue_struct *tcwq = get_work_cwq(twork); + + if (!(tcwq->wq->flags & WQ_HIGHPRI)) + break; + } + + gcwq->flags |= GCWQ_HIGHPRI_PENDING; + return &twork->entry; +} + +/** + * insert_work - insert a work into gcwq + * @cwq: cwq @work belongs to + * @work: work to insert + * @head: insertion point + * @extra_flags: extra WORK_STRUCT_* flags to set + * + * Insert @work which belongs to @cwq into @gcwq after @head. + * @extra_flags is or'd to work_struct flags. + * + * CONTEXT: + * spin_lock_irq(gcwq->lock). + */ static void insert_work(struct cpu_workqueue_struct *cwq, - struct work_struct *work, struct list_head *head) + struct work_struct *work, struct list_head *head, + unsigned int extra_flags) { - trace_workqueue_insertion(cwq->thread, work); + struct global_cwq *gcwq = cwq->gcwq; + + /* we own @work, set data and link */ + set_work_cwq(work, cwq, extra_flags); - set_wq_data(work, cwq); /* * Ensure that we get the right work->data if we see the * result of list_add() below, see try_to_grab_pending(). */ smp_wmb(); + list_add_tail(&work->entry, head); - wake_up(&cwq->more_work); + + /* + * Ensure either worker_sched_deactivated() sees the above + * list_add_tail() or we see zero nr_running to avoid workers + * lying around lazily while there are works to be processed. + */ + smp_mb(); + + if (__need_more_worker(gcwq)) + wake_up_worker(gcwq); } -static void __queue_work(struct cpu_workqueue_struct *cwq, +/* + * Test whether @work is being queued from another work executing on the + * same workqueue. This is rather expensive and should only be used from + * cold paths. + */ +static bool is_chained_work(struct workqueue_struct *wq) +{ + unsigned long flags; + unsigned int cpu; + + for_each_gcwq_cpu(cpu) { + struct global_cwq *gcwq = get_gcwq(cpu); + struct worker *worker; + struct hlist_node *pos; + int i; + + spin_lock_irqsave(&gcwq->lock, flags); + for_each_busy_worker(worker, i, pos, gcwq) { + if (worker->task != current) + continue; + spin_unlock_irqrestore(&gcwq->lock, flags); + /* + * I'm @worker, no locking necessary. See if @work + * is headed to the same workqueue. + */ + return worker->current_cwq->wq == wq; + } + spin_unlock_irqrestore(&gcwq->lock, flags); + } + return false; +} + +static void __queue_work(unsigned int cpu, struct workqueue_struct *wq, struct work_struct *work) { + struct global_cwq *gcwq; + struct cpu_workqueue_struct *cwq; + struct list_head *worklist; + unsigned int work_flags; unsigned long flags; debug_work_activate(work); - spin_lock_irqsave(&cwq->lock, flags); - insert_work(cwq, work, &cwq->worklist); - spin_unlock_irqrestore(&cwq->lock, flags); + + /* if dying, only works from the same workqueue are allowed */ + if (unlikely(wq->flags & WQ_DYING) && + WARN_ON_ONCE(!is_chained_work(wq))) + return; + + /* determine gcwq to use */ + if (!(wq->flags & WQ_UNBOUND)) { + struct global_cwq *last_gcwq; + + if (unlikely(cpu == WORK_CPU_UNBOUND)) + cpu = raw_smp_processor_id(); + + /* + * It's multi cpu. If @wq is non-reentrant and @work + * was previously on a different cpu, it might still + * be running there, in which case the work needs to + * be queued on that cpu to guarantee non-reentrance. + */ + gcwq = get_gcwq(cpu); + if (wq->flags & WQ_NON_REENTRANT && + (last_gcwq = get_work_gcwq(work)) && last_gcwq != gcwq) { + struct worker *worker; + + spin_lock_irqsave(&last_gcwq->lock, flags); + + worker = find_worker_executing_work(last_gcwq, work); + + if (worker && worker->current_cwq->wq == wq) + gcwq = last_gcwq; + else { + /* meh... not running there, queue here */ + spin_unlock_irqrestore(&last_gcwq->lock, flags); + spin_lock_irqsave(&gcwq->lock, flags); + } + } else + spin_lock_irqsave(&gcwq->lock, flags); + } else { + gcwq = get_gcwq(WORK_CPU_UNBOUND); + spin_lock_irqsave(&gcwq->lock, flags); + } + + /* gcwq determined, get cwq and queue */ + cwq = get_cwq(gcwq->cpu, wq); + trace_workqueue_queue_work(cpu, cwq, work); + + BUG_ON(!list_empty(&work->entry)); + + cwq->nr_in_flight[cwq->work_color]++; + work_flags = work_color_to_flags(cwq->work_color); + + if (likely(cwq->nr_active < cwq->max_active)) { + trace_workqueue_activate_work(work); + cwq->nr_active++; + worklist = gcwq_determine_ins_pos(gcwq, cwq); + } else { + work_flags |= WORK_STRUCT_DELAYED; + worklist = &cwq->delayed_works; + } + + insert_work(cwq, work, worklist, work_flags); + + spin_unlock_irqrestore(&gcwq->lock, flags); } /** @@ -308,9 +1080,8 @@ queue_work_on(int cpu, struct workqueue_struct *wq, struct work_struct *work) { int ret = 0; - if (!test_and_set_bit(WORK_STRUCT_PENDING, work_data_bits(work))) { - BUG_ON(!list_empty(&work->entry)); - __queue_work(wq_per_cpu(wq, cpu), work); + if (!test_and_set_bit(WORK_STRUCT_PENDING_BIT, work_data_bits(work))) { + __queue_work(cpu, wq, work); ret = 1; } return ret; @@ -320,10 +1091,9 @@ EXPORT_SYMBOL_GPL(queue_work_on); static void delayed_work_timer_fn(unsigned long __data) { struct delayed_work *dwork = (struct delayed_work *)__data; - struct cpu_workqueue_struct *cwq = get_wq_data(&dwork->work); - struct workqueue_struct *wq = cwq->wq; + struct cpu_workqueue_struct *cwq = get_work_cwq(&dwork->work); - __queue_work(wq_per_cpu(wq, smp_processor_id()), &dwork->work); + __queue_work(smp_processor_id(), cwq->wq, &dwork->work); } /** @@ -360,14 +1130,31 @@ int queue_delayed_work_on(int cpu, struct workqueue_struct *wq, struct timer_list *timer = &dwork->timer; struct work_struct *work = &dwork->work; - if (!test_and_set_bit(WORK_STRUCT_PENDING, work_data_bits(work))) { + if (!test_and_set_bit(WORK_STRUCT_PENDING_BIT, work_data_bits(work))) { + unsigned int lcpu; + BUG_ON(timer_pending(timer)); BUG_ON(!list_empty(&work->entry)); timer_stats_timer_set_start_info(&dwork->timer); - /* This stores cwq for the moment, for the timer_fn */ - set_wq_data(work, wq_per_cpu(wq, raw_smp_processor_id())); + /* + * This stores cwq for the moment, for the timer_fn. + * Note that the work's gcwq is preserved to allow + * reentrance detection for delayed works. + */ + if (!(wq->flags & WQ_UNBOUND)) { + struct global_cwq *gcwq = get_work_gcwq(work); + + if (gcwq && gcwq->cpu != WORK_CPU_UNBOUND) + lcpu = gcwq->cpu; + else + lcpu = raw_smp_processor_id(); + } else + lcpu = WORK_CPU_UNBOUND; + + set_work_cwq(work, get_cwq(lcpu, wq), 0); + timer->expires = jiffies + delay; timer->data = (unsigned long)dwork; timer->function = delayed_work_timer_fn; @@ -382,80 +1169,889 @@ int queue_delayed_work_on(int cpu, struct workqueue_struct *wq, } EXPORT_SYMBOL_GPL(queue_delayed_work_on); -static void run_workqueue(struct cpu_workqueue_struct *cwq) +/** + * worker_enter_idle - enter idle state + * @worker: worker which is entering idle state + * + * @worker is entering idle state. Update stats and idle timer if + * necessary. + * + * LOCKING: + * spin_lock_irq(gcwq->lock). + */ +static void worker_enter_idle(struct worker *worker) { - spin_lock_irq(&cwq->lock); - while (!list_empty(&cwq->worklist)) { - struct work_struct *work = list_entry(cwq->worklist.next, - struct work_struct, entry); - work_func_t f = work->func; -#ifdef CONFIG_LOCKDEP + struct global_cwq *gcwq = worker->gcwq; + + BUG_ON(worker->flags & WORKER_IDLE); + BUG_ON(!list_empty(&worker->entry) && + (worker->hentry.next || worker->hentry.pprev)); + + /* can't use worker_set_flags(), also called from start_worker() */ + worker->flags |= WORKER_IDLE; + gcwq->nr_idle++; + worker->last_active = jiffies; + + /* idle_list is LIFO */ + list_add(&worker->entry, &gcwq->idle_list); + + if (likely(!(worker->flags & WORKER_ROGUE))) { + if (too_many_workers(gcwq) && !timer_pending(&gcwq->idle_timer)) + mod_timer(&gcwq->idle_timer, + jiffies + IDLE_WORKER_TIMEOUT); + } else + wake_up_all(&gcwq->trustee_wait); + + /* sanity check nr_running */ + WARN_ON_ONCE(gcwq->nr_workers == gcwq->nr_idle && + atomic_read(get_gcwq_nr_running(gcwq->cpu))); +} + +/** + * worker_leave_idle - leave idle state + * @worker: worker which is leaving idle state + * + * @worker is leaving idle state. Update stats. + * + * LOCKING: + * spin_lock_irq(gcwq->lock). + */ +static void worker_leave_idle(struct worker *worker) +{ + struct global_cwq *gcwq = worker->gcwq; + + BUG_ON(!(worker->flags & WORKER_IDLE)); + worker_clr_flags(worker, WORKER_IDLE); + gcwq->nr_idle--; + list_del_init(&worker->entry); +} + +/** + * worker_maybe_bind_and_lock - bind worker to its cpu if possible and lock gcwq + * @worker: self + * + * Works which are scheduled while the cpu is online must at least be + * scheduled to a worker which is bound to the cpu so that if they are + * flushed from cpu callbacks while cpu is going down, they are + * guaranteed to execute on the cpu. + * + * This function is to be used by rogue workers and rescuers to bind + * themselves to the target cpu and may race with cpu going down or + * coming online. kthread_bind() can't be used because it may put the + * worker to already dead cpu and set_cpus_allowed_ptr() can't be used + * verbatim as it's best effort and blocking and gcwq may be + * [dis]associated in the meantime. + * + * This function tries set_cpus_allowed() and locks gcwq and verifies + * the binding against GCWQ_DISASSOCIATED which is set during + * CPU_DYING and cleared during CPU_ONLINE, so if the worker enters + * idle state or fetches works without dropping lock, it can guarantee + * the scheduling requirement described in the first paragraph. + * + * CONTEXT: + * Might sleep. Called without any lock but returns with gcwq->lock + * held. + * + * RETURNS: + * %true if the associated gcwq is online (@worker is successfully + * bound), %false if offline. + */ +static bool worker_maybe_bind_and_lock(struct worker *worker) +__acquires(&gcwq->lock) +{ + struct global_cwq *gcwq = worker->gcwq; + struct task_struct *task = worker->task; + + while (true) { /* - * It is permissible to free the struct work_struct - * from inside the function that is called from it, - * this we need to take into account for lockdep too. - * To avoid bogus "held lock freed" warnings as well - * as problems when looking into work->lockdep_map, - * make a copy and use that here. + * The following call may fail, succeed or succeed + * without actually migrating the task to the cpu if + * it races with cpu hotunplug operation. Verify + * against GCWQ_DISASSOCIATED. */ - struct lockdep_map lockdep_map = work->lockdep_map; -#endif - trace_workqueue_execution(cwq->thread, work); - debug_work_deactivate(work); - cwq->current_work = work; - list_del_init(cwq->worklist.next); - spin_unlock_irq(&cwq->lock); - - BUG_ON(get_wq_data(work) != cwq); - work_clear_pending(work); - lock_map_acquire(&cwq->wq->lockdep_map); - lock_map_acquire(&lockdep_map); - f(work); - lock_map_release(&lockdep_map); - lock_map_release(&cwq->wq->lockdep_map); - - if (unlikely(in_atomic() || lockdep_depth(current) > 0)) { - printk(KERN_ERR "BUG: workqueue leaked lock or atomic: " - "%s/0x%08x/%d\n", - current->comm, preempt_count(), - task_pid_nr(current)); - printk(KERN_ERR " last function: "); - print_symbol("%s\n", (unsigned long)f); - debug_show_held_locks(current); - dump_stack(); + if (!(gcwq->flags & GCWQ_DISASSOCIATED)) + set_cpus_allowed_ptr(task, get_cpu_mask(gcwq->cpu)); + + spin_lock_irq(&gcwq->lock); + if (gcwq->flags & GCWQ_DISASSOCIATED) + return false; + if (task_cpu(task) == gcwq->cpu && + cpumask_equal(¤t->cpus_allowed, + get_cpu_mask(gcwq->cpu))) + return true; + spin_unlock_irq(&gcwq->lock); + + /* CPU has come up inbetween, retry migration */ + cpu_relax(); + } +} + +/* + * Function for worker->rebind_work used to rebind rogue busy workers + * to the associated cpu which is coming back online. This is + * scheduled by cpu up but can race with other cpu hotplug operations + * and may be executed twice without intervening cpu down. + */ +static void worker_rebind_fn(struct work_struct *work) +{ + struct worker *worker = container_of(work, struct worker, rebind_work); + struct global_cwq *gcwq = worker->gcwq; + + if (worker_maybe_bind_and_lock(worker)) + worker_clr_flags(worker, WORKER_REBIND); + + spin_unlock_irq(&gcwq->lock); +} + +static struct worker *alloc_worker(void) +{ + struct worker *worker; + + worker = kzalloc(sizeof(*worker), GFP_KERNEL); + if (worker) { + INIT_LIST_HEAD(&worker->entry); + INIT_LIST_HEAD(&worker->scheduled); + INIT_WORK(&worker->rebind_work, worker_rebind_fn); + /* on creation a worker is in !idle && prep state */ + worker->flags = WORKER_PREP; + } + return worker; +} + +/** + * create_worker - create a new workqueue worker + * @gcwq: gcwq the new worker will belong to + * @bind: whether to set affinity to @cpu or not + * + * Create a new worker which is bound to @gcwq. The returned worker + * can be started by calling start_worker() or destroyed using + * destroy_worker(). + * + * CONTEXT: + * Might sleep. Does GFP_KERNEL allocations. + * + * RETURNS: + * Pointer to the newly created worker. + */ +static struct worker *create_worker(struct global_cwq *gcwq, bool bind) +{ + bool on_unbound_cpu = gcwq->cpu == WORK_CPU_UNBOUND; + struct worker *worker = NULL; + int id = -1; + + spin_lock_irq(&gcwq->lock); + while (ida_get_new(&gcwq->worker_ida, &id)) { + spin_unlock_irq(&gcwq->lock); + if (!ida_pre_get(&gcwq->worker_ida, GFP_KERNEL)) + goto fail; + spin_lock_irq(&gcwq->lock); + } + spin_unlock_irq(&gcwq->lock); + + worker = alloc_worker(); + if (!worker) + goto fail; + + worker->gcwq = gcwq; + worker->id = id; + + if (!on_unbound_cpu) + worker->task = kthread_create(worker_thread, worker, + "kworker/%u:%d", gcwq->cpu, id); + else + worker->task = kthread_create(worker_thread, worker, + "kworker/u:%d", id); + if (IS_ERR(worker->task)) + goto fail; + + /* + * A rogue worker will become a regular one if CPU comes + * online later on. Make sure every worker has + * PF_THREAD_BOUND set. + */ + if (bind && !on_unbound_cpu) + kthread_bind(worker->task, gcwq->cpu); + else { + worker->task->flags |= PF_THREAD_BOUND; + if (on_unbound_cpu) + worker->flags |= WORKER_UNBOUND; + } + + return worker; +fail: + if (id >= 0) { + spin_lock_irq(&gcwq->lock); + ida_remove(&gcwq->worker_ida, id); + spin_unlock_irq(&gcwq->lock); + } + kfree(worker); + return NULL; +} + +/** + * start_worker - start a newly created worker + * @worker: worker to start + * + * Make the gcwq aware of @worker and start it. + * + * CONTEXT: + * spin_lock_irq(gcwq->lock). + */ +static void start_worker(struct worker *worker) +{ + worker->flags |= WORKER_STARTED; + worker->gcwq->nr_workers++; + worker_enter_idle(worker); + wake_up_process(worker->task); +} + +/** + * destroy_worker - destroy a workqueue worker + * @worker: worker to be destroyed + * + * Destroy @worker and adjust @gcwq stats accordingly. + * + * CONTEXT: + * spin_lock_irq(gcwq->lock) which is released and regrabbed. + */ +static void destroy_worker(struct worker *worker) +{ + struct global_cwq *gcwq = worker->gcwq; + int id = worker->id; + + /* sanity check frenzy */ + BUG_ON(worker->current_work); + BUG_ON(!list_empty(&worker->scheduled)); + + if (worker->flags & WORKER_STARTED) + gcwq->nr_workers--; + if (worker->flags & WORKER_IDLE) + gcwq->nr_idle--; + + list_del_init(&worker->entry); + worker->flags |= WORKER_DIE; + + spin_unlock_irq(&gcwq->lock); + + kthread_stop(worker->task); + kfree(worker); + + spin_lock_irq(&gcwq->lock); + ida_remove(&gcwq->worker_ida, id); +} + +static void idle_worker_timeout(unsigned long __gcwq) +{ + struct global_cwq *gcwq = (void *)__gcwq; + + spin_lock_irq(&gcwq->lock); + + if (too_many_workers(gcwq)) { + struct worker *worker; + unsigned long expires; + + /* idle_list is kept in LIFO order, check the last one */ + worker = list_entry(gcwq->idle_list.prev, struct worker, entry); + expires = worker->last_active + IDLE_WORKER_TIMEOUT; + + if (time_before(jiffies, expires)) + mod_timer(&gcwq->idle_timer, expires); + else { + /* it's been idle for too long, wake up manager */ + gcwq->flags |= GCWQ_MANAGE_WORKERS; + wake_up_worker(gcwq); } + } - spin_lock_irq(&cwq->lock); - cwq->current_work = NULL; + spin_unlock_irq(&gcwq->lock); +} + +static bool send_mayday(struct work_struct *work) +{ + struct cpu_workqueue_struct *cwq = get_work_cwq(work); + struct workqueue_struct *wq = cwq->wq; + unsigned int cpu; + + if (!(wq->flags & WQ_RESCUER)) + return false; + + /* mayday mayday mayday */ + cpu = cwq->gcwq->cpu; + /* WORK_CPU_UNBOUND can't be set in cpumask, use cpu 0 instead */ + if (cpu == WORK_CPU_UNBOUND) + cpu = 0; + if (!mayday_test_and_set_cpu(cpu, wq->mayday_mask)) + wake_up_process(wq->rescuer->task); + return true; +} + +static void gcwq_mayday_timeout(unsigned long __gcwq) +{ + struct global_cwq *gcwq = (void *)__gcwq; + struct work_struct *work; + + spin_lock_irq(&gcwq->lock); + + if (need_to_create_worker(gcwq)) { + /* + * We've been trying to create a new worker but + * haven't been successful. We might be hitting an + * allocation deadlock. Send distress signals to + * rescuers. + */ + list_for_each_entry(work, &gcwq->worklist, entry) + send_mayday(work); } - spin_unlock_irq(&cwq->lock); + + spin_unlock_irq(&gcwq->lock); + + mod_timer(&gcwq->mayday_timer, jiffies + MAYDAY_INTERVAL); } -static int worker_thread(void *__cwq) +/** + * maybe_create_worker - create a new worker if necessary + * @gcwq: gcwq to create a new worker for + * + * Create a new worker for @gcwq if necessary. @gcwq is guaranteed to + * have at least one idle worker on return from this function. If + * creating a new worker takes longer than MAYDAY_INTERVAL, mayday is + * sent to all rescuers with works scheduled on @gcwq to resolve + * possible allocation deadlock. + * + * On return, need_to_create_worker() is guaranteed to be false and + * may_start_working() true. + * + * LOCKING: + * spin_lock_irq(gcwq->lock) which may be released and regrabbed + * multiple times. Does GFP_KERNEL allocations. Called only from + * manager. + * + * RETURNS: + * false if no action was taken and gcwq->lock stayed locked, true + * otherwise. + */ +static bool maybe_create_worker(struct global_cwq *gcwq) +__releases(&gcwq->lock) +__acquires(&gcwq->lock) { - struct cpu_workqueue_struct *cwq = __cwq; - DEFINE_WAIT(wait); + if (!need_to_create_worker(gcwq)) + return false; +restart: + spin_unlock_irq(&gcwq->lock); + + /* if we don't make progress in MAYDAY_INITIAL_TIMEOUT, call for help */ + mod_timer(&gcwq->mayday_timer, jiffies + MAYDAY_INITIAL_TIMEOUT); + + while (true) { + struct worker *worker; + + worker = create_worker(gcwq, true); + if (worker) { + del_timer_sync(&gcwq->mayday_timer); + spin_lock_irq(&gcwq->lock); + start_worker(worker); + BUG_ON(need_to_create_worker(gcwq)); + return true; + } - if (cwq->wq->freezeable) - set_freezable(); + if (!need_to_create_worker(gcwq)) + break; - for (;;) { - prepare_to_wait(&cwq->more_work, &wait, TASK_INTERRUPTIBLE); - if (!freezing(current) && - !kthread_should_stop() && - list_empty(&cwq->worklist)) - schedule(); - finish_wait(&cwq->more_work, &wait); + __set_current_state(TASK_INTERRUPTIBLE); + schedule_timeout(CREATE_COOLDOWN); - try_to_freeze(); + if (!need_to_create_worker(gcwq)) + break; + } + + del_timer_sync(&gcwq->mayday_timer); + spin_lock_irq(&gcwq->lock); + if (need_to_create_worker(gcwq)) + goto restart; + return true; +} + +/** + * maybe_destroy_worker - destroy workers which have been idle for a while + * @gcwq: gcwq to destroy workers for + * + * Destroy @gcwq workers which have been idle for longer than + * IDLE_WORKER_TIMEOUT. + * + * LOCKING: + * spin_lock_irq(gcwq->lock) which may be released and regrabbed + * multiple times. Called only from manager. + * + * RETURNS: + * false if no action was taken and gcwq->lock stayed locked, true + * otherwise. + */ +static bool maybe_destroy_workers(struct global_cwq *gcwq) +{ + bool ret = false; + + while (too_many_workers(gcwq)) { + struct worker *worker; + unsigned long expires; - if (kthread_should_stop()) + worker = list_entry(gcwq->idle_list.prev, struct worker, entry); + expires = worker->last_active + IDLE_WORKER_TIMEOUT; + + if (time_before(jiffies, expires)) { + mod_timer(&gcwq->idle_timer, expires); break; + } - run_workqueue(cwq); + destroy_worker(worker); + ret = true; } - return 0; + return ret; +} + +/** + * manage_workers - manage worker pool + * @worker: self + * + * Assume the manager role and manage gcwq worker pool @worker belongs + * to. At any given time, there can be only zero or one manager per + * gcwq. The exclusion is handled automatically by this function. + * + * The caller can safely start processing works on false return. On + * true return, it's guaranteed that need_to_create_worker() is false + * and may_start_working() is true. + * + * CONTEXT: + * spin_lock_irq(gcwq->lock) which may be released and regrabbed + * multiple times. Does GFP_KERNEL allocations. + * + * RETURNS: + * false if no action was taken and gcwq->lock stayed locked, true if + * some action was taken. + */ +static bool manage_workers(struct worker *worker) +{ + struct global_cwq *gcwq = worker->gcwq; + bool ret = false; + + if (gcwq->flags & GCWQ_MANAGING_WORKERS) + return ret; + + gcwq->flags &= ~GCWQ_MANAGE_WORKERS; + gcwq->flags |= GCWQ_MANAGING_WORKERS; + + /* + * Destroy and then create so that may_start_working() is true + * on return. + */ + ret |= maybe_destroy_workers(gcwq); + ret |= maybe_create_worker(gcwq); + + gcwq->flags &= ~GCWQ_MANAGING_WORKERS; + + /* + * The trustee might be waiting to take over the manager + * position, tell it we're done. + */ + if (unlikely(gcwq->trustee)) + wake_up_all(&gcwq->trustee_wait); + + return ret; +} + +/** + * move_linked_works - move linked works to a list + * @work: start of series of works to be scheduled + * @head: target list to append @work to + * @nextp: out paramter for nested worklist walking + * + * Schedule linked works starting from @work to @head. Work series to + * be scheduled starts at @work and includes any consecutive work with + * WORK_STRUCT_LINKED set in its predecessor. + * + * If @nextp is not NULL, it's updated to point to the next work of + * the last scheduled work. This allows move_linked_works() to be + * nested inside outer list_for_each_entry_safe(). + * + * CONTEXT: + * spin_lock_irq(gcwq->lock). + */ +static void move_linked_works(struct work_struct *work, struct list_head *head, + struct work_struct **nextp) +{ + struct work_struct *n; + + /* + * Linked worklist will always end before the end of the list, + * use NULL for list head. + */ + list_for_each_entry_safe_from(work, n, NULL, entry) { + list_move_tail(&work->entry, head); + if (!(*work_data_bits(work) & WORK_STRUCT_LINKED)) + break; + } + + /* + * If we're already inside safe list traversal and have moved + * multiple works to the scheduled queue, the next position + * needs to be updated. + */ + if (nextp) + *nextp = n; +} + +static void cwq_activate_first_delayed(struct cpu_workqueue_struct *cwq) +{ + struct work_struct *work = list_first_entry(&cwq->delayed_works, + struct work_struct, entry); + struct list_head *pos = gcwq_determine_ins_pos(cwq->gcwq, cwq); + + trace_workqueue_activate_work(work); + move_linked_works(work, pos, NULL); + __clear_bit(WORK_STRUCT_DELAYED_BIT, work_data_bits(work)); + cwq->nr_active++; +} + +/** + * cwq_dec_nr_in_flight - decrement cwq's nr_in_flight + * @cwq: cwq of interest + * @color: color of work which left the queue + * @delayed: for a delayed work + * + * A work either has completed or is removed from pending queue, + * decrement nr_in_flight of its cwq and handle workqueue flushing. + * + * CONTEXT: + * spin_lock_irq(gcwq->lock). + */ +static void cwq_dec_nr_in_flight(struct cpu_workqueue_struct *cwq, int color, + bool delayed) +{ + /* ignore uncolored works */ + if (color == WORK_NO_COLOR) + return; + + cwq->nr_in_flight[color]--; + + if (!delayed) { + cwq->nr_active--; + if (!list_empty(&cwq->delayed_works)) { + /* one down, submit a delayed one */ + if (cwq->nr_active < cwq->max_active) + cwq_activate_first_delayed(cwq); + } + } + + /* is flush in progress and are we at the flushing tip? */ + if (likely(cwq->flush_color != color)) + return; + + /* are there still in-flight works? */ + if (cwq->nr_in_flight[color]) + return; + + /* this cwq is done, clear flush_color */ + cwq->flush_color = -1; + + /* + * If this was the last cwq, wake up the first flusher. It + * will handle the rest. + */ + if (atomic_dec_and_test(&cwq->wq->nr_cwqs_to_flush)) + complete(&cwq->wq->first_flusher->done); +} + +/** + * process_one_work - process single work + * @worker: self + * @work: work to process + * + * Process @work. This function contains all the logics necessary to + * process a single work including synchronization against and + * interaction with other workers on the same cpu, queueing and + * flushing. As long as context requirement is met, any worker can + * call this function to process a work. + * + * CONTEXT: + * spin_lock_irq(gcwq->lock) which is released and regrabbed. + */ +static void process_one_work(struct worker *worker, struct work_struct *work) +__releases(&gcwq->lock) +__acquires(&gcwq->lock) +{ + struct cpu_workqueue_struct *cwq = get_work_cwq(work); + struct global_cwq *gcwq = cwq->gcwq; + struct hlist_head *bwh = busy_worker_head(gcwq, work); + bool cpu_intensive = cwq->wq->flags & WQ_CPU_INTENSIVE; + work_func_t f = work->func; + int work_color; + struct worker *collision; +#ifdef CONFIG_LOCKDEP + /* + * It is permissible to free the struct work_struct from + * inside the function that is called from it, this we need to + * take into account for lockdep too. To avoid bogus "held + * lock freed" warnings as well as problems when looking into + * work->lockdep_map, make a copy and use that here. + */ + struct lockdep_map lockdep_map = work->lockdep_map; +#endif + /* + * A single work shouldn't be executed concurrently by + * multiple workers on a single cpu. Check whether anyone is + * already processing the work. If so, defer the work to the + * currently executing one. + */ + collision = __find_worker_executing_work(gcwq, bwh, work); + if (unlikely(collision)) { + move_linked_works(work, &collision->scheduled, NULL); + return; + } + + /* claim and process */ + debug_work_deactivate(work); + hlist_add_head(&worker->hentry, bwh); + worker->current_work = work; + worker->current_cwq = cwq; + work_color = get_work_color(work); + + /* record the current cpu number in the work data and dequeue */ + set_work_cpu(work, gcwq->cpu); + list_del_init(&work->entry); + + /* + * If HIGHPRI_PENDING, check the next work, and, if HIGHPRI, + * wake up another worker; otherwise, clear HIGHPRI_PENDING. + */ + if (unlikely(gcwq->flags & GCWQ_HIGHPRI_PENDING)) { + struct work_struct *nwork = list_first_entry(&gcwq->worklist, + struct work_struct, entry); + + if (!list_empty(&gcwq->worklist) && + get_work_cwq(nwork)->wq->flags & WQ_HIGHPRI) + wake_up_worker(gcwq); + else + gcwq->flags &= ~GCWQ_HIGHPRI_PENDING; + } + + /* + * CPU intensive works don't participate in concurrency + * management. They're the scheduler's responsibility. + */ + if (unlikely(cpu_intensive)) + worker_set_flags(worker, WORKER_CPU_INTENSIVE, true); + + spin_unlock_irq(&gcwq->lock); + + work_clear_pending(work); + lock_map_acquire_read(&cwq->wq->lockdep_map); + lock_map_acquire(&lockdep_map); + trace_workqueue_execute_start(work); + f(work); + /* + * While we must be careful to not use "work" after this, the trace + * point will only record its address. + */ + trace_workqueue_execute_end(work); + lock_map_release(&lockdep_map); + lock_map_release(&cwq->wq->lockdep_map); + + if (unlikely(in_atomic() || lockdep_depth(current) > 0)) { + printk(KERN_ERR "BUG: workqueue leaked lock or atomic: " + "%s/0x%08x/%d\n", + current->comm, preempt_count(), task_pid_nr(current)); + printk(KERN_ERR " last function: "); + print_symbol("%s\n", (unsigned long)f); + debug_show_held_locks(current); + dump_stack(); + } + + spin_lock_irq(&gcwq->lock); + + /* clear cpu intensive status */ + if (unlikely(cpu_intensive)) + worker_clr_flags(worker, WORKER_CPU_INTENSIVE); + + /* we're done with it, release */ + hlist_del_init(&worker->hentry); + worker->current_work = NULL; + worker->current_cwq = NULL; + cwq_dec_nr_in_flight(cwq, work_color, false); +} + +/** + * process_scheduled_works - process scheduled works + * @worker: self + * + * Process all scheduled works. Please note that the scheduled list + * may change while processing a work, so this function repeatedly + * fetches a work from the top and executes it. + * + * CONTEXT: + * spin_lock_irq(gcwq->lock) which may be released and regrabbed + * multiple times. + */ +static void process_scheduled_works(struct worker *worker) +{ + while (!list_empty(&worker->scheduled)) { + struct work_struct *work = list_first_entry(&worker->scheduled, + struct work_struct, entry); + process_one_work(worker, work); + } +} + +/** + * worker_thread - the worker thread function + * @__worker: self + * + * The gcwq worker thread function. There's a single dynamic pool of + * these per each cpu. These workers process all works regardless of + * their specific target workqueue. The only exception is works which + * belong to workqueues with a rescuer which will be explained in + * rescuer_thread(). + */ +static int worker_thread(void *__worker) +{ + struct worker *worker = __worker; + struct global_cwq *gcwq = worker->gcwq; + + /* tell the scheduler that this is a workqueue worker */ + worker->task->flags |= PF_WQ_WORKER; +woke_up: + spin_lock_irq(&gcwq->lock); + + /* DIE can be set only while we're idle, checking here is enough */ + if (worker->flags & WORKER_DIE) { + spin_unlock_irq(&gcwq->lock); + worker->task->flags &= ~PF_WQ_WORKER; + return 0; + } + + worker_leave_idle(worker); +recheck: + /* no more worker necessary? */ + if (!need_more_worker(gcwq)) + goto sleep; + + /* do we need to manage? */ + if (unlikely(!may_start_working(gcwq)) && manage_workers(worker)) + goto recheck; + + /* + * ->scheduled list can only be filled while a worker is + * preparing to process a work or actually processing it. + * Make sure nobody diddled with it while I was sleeping. + */ + BUG_ON(!list_empty(&worker->scheduled)); + + /* + * When control reaches this point, we're guaranteed to have + * at least one idle worker or that someone else has already + * assumed the manager role. + */ + worker_clr_flags(worker, WORKER_PREP); + + do { + struct work_struct *work = + list_first_entry(&gcwq->worklist, + struct work_struct, entry); + + if (likely(!(*work_data_bits(work) & WORK_STRUCT_LINKED))) { + /* optimization path, not strictly necessary */ + process_one_work(worker, work); + if (unlikely(!list_empty(&worker->scheduled))) + process_scheduled_works(worker); + } else { + move_linked_works(work, &worker->scheduled, NULL); + process_scheduled_works(worker); + } + } while (keep_working(gcwq)); + + worker_set_flags(worker, WORKER_PREP, false); +sleep: + if (unlikely(need_to_manage_workers(gcwq)) && manage_workers(worker)) + goto recheck; + + /* + * gcwq->lock is held and there's no work to process and no + * need to manage, sleep. Workers are woken up only while + * holding gcwq->lock or from local cpu, so setting the + * current state before releasing gcwq->lock is enough to + * prevent losing any event. + */ + worker_enter_idle(worker); + __set_current_state(TASK_INTERRUPTIBLE); + spin_unlock_irq(&gcwq->lock); + schedule(); + goto woke_up; +} + +/** + * rescuer_thread - the rescuer thread function + * @__wq: the associated workqueue + * + * Workqueue rescuer thread function. There's one rescuer for each + * workqueue which has WQ_RESCUER set. + * + * Regular work processing on a gcwq may block trying to create a new + * worker which uses GFP_KERNEL allocation which has slight chance of + * developing into deadlock if some works currently on the same queue + * need to be processed to satisfy the GFP_KERNEL allocation. This is + * the problem rescuer solves. + * + * When such condition is possible, the gcwq summons rescuers of all + * workqueues which have works queued on the gcwq and let them process + * those works so that forward progress can be guaranteed. + * + * This should happen rarely. + */ +static int rescuer_thread(void *__wq) +{ + struct workqueue_struct *wq = __wq; + struct worker *rescuer = wq->rescuer; + struct list_head *scheduled = &rescuer->scheduled; + bool is_unbound = wq->flags & WQ_UNBOUND; + unsigned int cpu; + + set_user_nice(current, RESCUER_NICE_LEVEL); +repeat: + set_current_state(TASK_INTERRUPTIBLE); + + if (kthread_should_stop()) + return 0; + + /* + * See whether any cpu is asking for help. Unbounded + * workqueues use cpu 0 in mayday_mask for CPU_UNBOUND. + */ + for_each_mayday_cpu(cpu, wq->mayday_mask) { + unsigned int tcpu = is_unbound ? WORK_CPU_UNBOUND : cpu; + struct cpu_workqueue_struct *cwq = get_cwq(tcpu, wq); + struct global_cwq *gcwq = cwq->gcwq; + struct work_struct *work, *n; + + __set_current_state(TASK_RUNNING); + mayday_clear_cpu(cpu, wq->mayday_mask); + + /* migrate to the target cpu if possible */ + rescuer->gcwq = gcwq; + worker_maybe_bind_and_lock(rescuer); + + /* + * Slurp in all works issued via this workqueue and + * process'em. + */ + BUG_ON(!list_empty(&rescuer->scheduled)); + list_for_each_entry_safe(work, n, &gcwq->worklist, entry) + if (get_work_cwq(work) == cwq) + move_linked_works(work, scheduled, &n); + + process_scheduled_works(rescuer); + spin_unlock_irq(&gcwq->lock); + } + + schedule(); + goto repeat; } struct wq_barrier { @@ -469,44 +2065,137 @@ static void wq_barrier_func(struct work_struct *work) complete(&barr->done); } +/** + * insert_wq_barrier - insert a barrier work + * @cwq: cwq to insert barrier into + * @barr: wq_barrier to insert + * @target: target work to attach @barr to + * @worker: worker currently executing @target, NULL if @target is not executing + * + * @barr is linked to @target such that @barr is completed only after + * @target finishes execution. Please note that the ordering + * guarantee is observed only with respect to @target and on the local + * cpu. + * + * Currently, a queued barrier can't be canceled. This is because + * try_to_grab_pending() can't determine whether the work to be + * grabbed is at the head of the queue and thus can't clear LINKED + * flag of the previous work while there must be a valid next work + * after a work with LINKED flag set. + * + * Note that when @worker is non-NULL, @target may be modified + * underneath us, so we can't reliably determine cwq from @target. + * + * CONTEXT: + * spin_lock_irq(gcwq->lock). + */ static void insert_wq_barrier(struct cpu_workqueue_struct *cwq, - struct wq_barrier *barr, struct list_head *head) + struct wq_barrier *barr, + struct work_struct *target, struct worker *worker) { + struct list_head *head; + unsigned int linked = 0; + /* - * debugobject calls are safe here even with cwq->lock locked + * debugobject calls are safe here even with gcwq->lock locked * as we know for sure that this will not trigger any of the * checks and call back into the fixup functions where we * might deadlock. */ - INIT_WORK_ON_STACK(&barr->work, wq_barrier_func); - __set_bit(WORK_STRUCT_PENDING, work_data_bits(&barr->work)); - + INIT_WORK_ONSTACK(&barr->work, wq_barrier_func); + __set_bit(WORK_STRUCT_PENDING_BIT, work_data_bits(&barr->work)); init_completion(&barr->done); + /* + * If @target is currently being executed, schedule the + * barrier to the worker; otherwise, put it after @target. + */ + if (worker) + head = worker->scheduled.next; + else { + unsigned long *bits = work_data_bits(target); + + head = target->entry.next; + /* there can already be other linked works, inherit and set */ + linked = *bits & WORK_STRUCT_LINKED; + __set_bit(WORK_STRUCT_LINKED_BIT, bits); + } + debug_work_activate(&barr->work); - insert_work(cwq, &barr->work, head); + insert_work(cwq, &barr->work, head, + work_color_to_flags(WORK_NO_COLOR) | linked); } -static int flush_cpu_workqueue(struct cpu_workqueue_struct *cwq) +/** + * flush_workqueue_prep_cwqs - prepare cwqs for workqueue flushing + * @wq: workqueue being flushed + * @flush_color: new flush color, < 0 for no-op + * @work_color: new work color, < 0 for no-op + * + * Prepare cwqs for workqueue flushing. + * + * If @flush_color is non-negative, flush_color on all cwqs should be + * -1. If no cwq has in-flight commands at the specified color, all + * cwq->flush_color's stay at -1 and %false is returned. If any cwq + * has in flight commands, its cwq->flush_color is set to + * @flush_color, @wq->nr_cwqs_to_flush is updated accordingly, cwq + * wakeup logic is armed and %true is returned. + * + * The caller should have initialized @wq->first_flusher prior to + * calling this function with non-negative @flush_color. If + * @flush_color is negative, no flush color update is done and %false + * is returned. + * + * If @work_color is non-negative, all cwqs should have the same + * work_color which is previous to @work_color and all will be + * advanced to @work_color. + * + * CONTEXT: + * mutex_lock(wq->flush_mutex). + * + * RETURNS: + * %true if @flush_color >= 0 and there's something to flush. %false + * otherwise. + */ +static bool flush_workqueue_prep_cwqs(struct workqueue_struct *wq, + int flush_color, int work_color) { - int active = 0; - struct wq_barrier barr; - - WARN_ON(cwq->thread == current); + bool wait = false; + unsigned int cpu; - spin_lock_irq(&cwq->lock); - if (!list_empty(&cwq->worklist) || cwq->current_work != NULL) { - insert_wq_barrier(cwq, &barr, &cwq->worklist); - active = 1; + if (flush_color >= 0) { + BUG_ON(atomic_read(&wq->nr_cwqs_to_flush)); + atomic_set(&wq->nr_cwqs_to_flush, 1); } - spin_unlock_irq(&cwq->lock); - if (active) { - wait_for_completion(&barr.done); - destroy_work_on_stack(&barr.work); + for_each_cwq_cpu(cpu, wq) { + struct cpu_workqueue_struct *cwq = get_cwq(cpu, wq); + struct global_cwq *gcwq = cwq->gcwq; + + spin_lock_irq(&gcwq->lock); + + if (flush_color >= 0) { + BUG_ON(cwq->flush_color != -1); + + if (cwq->nr_in_flight[flush_color]) { + cwq->flush_color = flush_color; + atomic_inc(&wq->nr_cwqs_to_flush); + wait = true; + } + } + + if (work_color >= 0) { + BUG_ON(work_color != work_next_color(cwq->work_color)); + cwq->work_color = work_color; + } + + spin_unlock_irq(&gcwq->lock); } - return active; + if (flush_color >= 0 && atomic_dec_and_test(&wq->nr_cwqs_to_flush)) + complete(&wq->first_flusher->done); + + return wait; } /** @@ -518,158 +2207,350 @@ static int flush_cpu_workqueue(struct cpu_workqueue_struct *cwq) * * We sleep until all works which were queued on entry have been handled, * but we are not livelocked by new incoming ones. - * - * This function used to run the workqueues itself. Now we just wait for the - * helper threads to do it. */ void flush_workqueue(struct workqueue_struct *wq) { - const struct cpumask *cpu_map = wq_cpu_map(wq); - int cpu; + struct wq_flusher this_flusher = { + .list = LIST_HEAD_INIT(this_flusher.list), + .flush_color = -1, + .done = COMPLETION_INITIALIZER_ONSTACK(this_flusher.done), + }; + int next_color; - might_sleep(); lock_map_acquire(&wq->lockdep_map); lock_map_release(&wq->lockdep_map); - for_each_cpu(cpu, cpu_map) - flush_cpu_workqueue(per_cpu_ptr(wq->cpu_wq, cpu)); + + mutex_lock(&wq->flush_mutex); + + /* + * Start-to-wait phase + */ + next_color = work_next_color(wq->work_color); + + if (next_color != wq->flush_color) { + /* + * Color space is not full. The current work_color + * becomes our flush_color and work_color is advanced + * by one. + */ + BUG_ON(!list_empty(&wq->flusher_overflow)); + this_flusher.flush_color = wq->work_color; + wq->work_color = next_color; + + if (!wq->first_flusher) { + /* no flush in progress, become the first flusher */ + BUG_ON(wq->flush_color != this_flusher.flush_color); + + wq->first_flusher = &this_flusher; + + if (!flush_workqueue_prep_cwqs(wq, wq->flush_color, + wq->work_color)) { + /* nothing to flush, done */ + wq->flush_color = next_color; + wq->first_flusher = NULL; + goto out_unlock; + } + } else { + /* wait in queue */ + BUG_ON(wq->flush_color == this_flusher.flush_color); + list_add_tail(&this_flusher.list, &wq->flusher_queue); + flush_workqueue_prep_cwqs(wq, -1, wq->work_color); + } + } else { + /* + * Oops, color space is full, wait on overflow queue. + * The next flush completion will assign us + * flush_color and transfer to flusher_queue. + */ + list_add_tail(&this_flusher.list, &wq->flusher_overflow); + } + + mutex_unlock(&wq->flush_mutex); + + wait_for_completion(&this_flusher.done); + + /* + * Wake-up-and-cascade phase + * + * First flushers are responsible for cascading flushes and + * handling overflow. Non-first flushers can simply return. + */ + if (wq->first_flusher != &this_flusher) + return; + + mutex_lock(&wq->flush_mutex); + + /* we might have raced, check again with mutex held */ + if (wq->first_flusher != &this_flusher) + goto out_unlock; + + wq->first_flusher = NULL; + + BUG_ON(!list_empty(&this_flusher.list)); + BUG_ON(wq->flush_color != this_flusher.flush_color); + + while (true) { + struct wq_flusher *next, *tmp; + + /* complete all the flushers sharing the current flush color */ + list_for_each_entry_safe(next, tmp, &wq->flusher_queue, list) { + if (next->flush_color != wq->flush_color) + break; + list_del_init(&next->list); + complete(&next->done); + } + + BUG_ON(!list_empty(&wq->flusher_overflow) && + wq->flush_color != work_next_color(wq->work_color)); + + /* this flush_color is finished, advance by one */ + wq->flush_color = work_next_color(wq->flush_color); + + /* one color has been freed, handle overflow queue */ + if (!list_empty(&wq->flusher_overflow)) { + /* + * Assign the same color to all overflowed + * flushers, advance work_color and append to + * flusher_queue. This is the start-to-wait + * phase for these overflowed flushers. + */ + list_for_each_entry(tmp, &wq->flusher_overflow, list) + tmp->flush_color = wq->work_color; + + wq->work_color = work_next_color(wq->work_color); + + list_splice_tail_init(&wq->flusher_overflow, + &wq->flusher_queue); + flush_workqueue_prep_cwqs(wq, -1, wq->work_color); + } + + if (list_empty(&wq->flusher_queue)) { + BUG_ON(wq->flush_color != wq->work_color); + break; + } + + /* + * Need to flush more colors. Make the next flusher + * the new first flusher and arm cwqs. + */ + BUG_ON(wq->flush_color == wq->work_color); + BUG_ON(wq->flush_color != next->flush_color); + + list_del_init(&next->list); + wq->first_flusher = next; + + if (flush_workqueue_prep_cwqs(wq, wq->flush_color, -1)) + break; + + /* + * Meh... this color is already done, clear first + * flusher and repeat cascading. + */ + wq->first_flusher = NULL; + } + +out_unlock: + mutex_unlock(&wq->flush_mutex); } EXPORT_SYMBOL_GPL(flush_workqueue); -/** - * flush_work - block until a work_struct's callback has terminated - * @work: the work which is to be flushed - * - * Returns false if @work has already terminated. - * - * It is expected that, prior to calling flush_work(), the caller has - * arranged for the work to not be requeued, otherwise it doesn't make - * sense to use this function. - */ -int flush_work(struct work_struct *work) +static bool start_flush_work(struct work_struct *work, struct wq_barrier *barr, + bool wait_executing) { + struct worker *worker = NULL; + struct global_cwq *gcwq; struct cpu_workqueue_struct *cwq; - struct list_head *prev; - struct wq_barrier barr; might_sleep(); - cwq = get_wq_data(work); - if (!cwq) - return 0; - - lock_map_acquire(&cwq->wq->lockdep_map); - lock_map_release(&cwq->wq->lockdep_map); + gcwq = get_work_gcwq(work); + if (!gcwq) + return false; - prev = NULL; - spin_lock_irq(&cwq->lock); + spin_lock_irq(&gcwq->lock); if (!list_empty(&work->entry)) { /* * See the comment near try_to_grab_pending()->smp_rmb(). - * If it was re-queued under us we are not going to wait. + * If it was re-queued to a different gcwq under us, we + * are not going to wait. */ smp_rmb(); - if (unlikely(cwq != get_wq_data(work))) - goto out; - prev = &work->entry; - } else { - if (cwq->current_work != work) - goto out; - prev = &cwq->worklist; - } - insert_wq_barrier(cwq, &barr, prev->next); -out: - spin_unlock_irq(&cwq->lock); - if (!prev) - return 0; + cwq = get_work_cwq(work); + if (unlikely(!cwq || gcwq != cwq->gcwq)) + goto already_gone; + } else if (wait_executing) { + worker = find_worker_executing_work(gcwq, work); + if (!worker) + goto already_gone; + cwq = worker->current_cwq; + } else + goto already_gone; + + insert_wq_barrier(cwq, barr, work, worker); + spin_unlock_irq(&gcwq->lock); - wait_for_completion(&barr.done); - destroy_work_on_stack(&barr.work); - return 1; + /* + * If @max_active is 1 or rescuer is in use, flushing another work + * item on the same workqueue may lead to deadlock. Make sure the + * flusher is not running on the same workqueue by verifying write + * access. + */ + if (cwq->wq->saved_max_active == 1 || cwq->wq->flags & WQ_RESCUER) + lock_map_acquire(&cwq->wq->lockdep_map); + else + lock_map_acquire_read(&cwq->wq->lockdep_map); + lock_map_release(&cwq->wq->lockdep_map); + + return true; +already_gone: + spin_unlock_irq(&gcwq->lock); + return false; +} + +/** + * flush_work - wait for a work to finish executing the last queueing instance + * @work: the work to flush + * + * Wait until @work has finished execution. This function considers + * only the last queueing instance of @work. If @work has been + * enqueued across different CPUs on a non-reentrant workqueue or on + * multiple workqueues, @work might still be executing on return on + * some of the CPUs from earlier queueing. + * + * If @work was queued only on a non-reentrant, ordered or unbound + * workqueue, @work is guaranteed to be idle on return if it hasn't + * been requeued since flush started. + * + * RETURNS: + * %true if flush_work() waited for the work to finish execution, + * %false if it was already idle. + */ +bool flush_work(struct work_struct *work) +{ + struct wq_barrier barr; + + if (start_flush_work(work, &barr, true)) { + wait_for_completion(&barr.done); + destroy_work_on_stack(&barr.work); + return true; + } else + return false; } EXPORT_SYMBOL_GPL(flush_work); +static bool wait_on_cpu_work(struct global_cwq *gcwq, struct work_struct *work) +{ + struct wq_barrier barr; + struct worker *worker; + + spin_lock_irq(&gcwq->lock); + + worker = find_worker_executing_work(gcwq, work); + if (unlikely(worker)) + insert_wq_barrier(worker->current_cwq, &barr, work, worker); + + spin_unlock_irq(&gcwq->lock); + + if (unlikely(worker)) { + wait_for_completion(&barr.done); + destroy_work_on_stack(&barr.work); + return true; + } else + return false; +} + +static bool wait_on_work(struct work_struct *work) +{ + bool ret = false; + int cpu; + + might_sleep(); + + lock_map_acquire(&work->lockdep_map); + lock_map_release(&work->lockdep_map); + + for_each_gcwq_cpu(cpu) + ret |= wait_on_cpu_work(get_gcwq(cpu), work); + return ret; +} + +/** + * flush_work_sync - wait until a work has finished execution + * @work: the work to flush + * + * Wait until @work has finished execution. On return, it's + * guaranteed that all queueing instances of @work which happened + * before this function is called are finished. In other words, if + * @work hasn't been requeued since this function was called, @work is + * guaranteed to be idle on return. + * + * RETURNS: + * %true if flush_work_sync() waited for the work to finish execution, + * %false if it was already idle. + */ +bool flush_work_sync(struct work_struct *work) +{ + struct wq_barrier barr; + bool pending, waited; + + /* we'll wait for executions separately, queue barr only if pending */ + pending = start_flush_work(work, &barr, false); + + /* wait for executions to finish */ + waited = wait_on_work(work); + + /* wait for the pending one */ + if (pending) { + wait_for_completion(&barr.done); + destroy_work_on_stack(&barr.work); + } + + return pending || waited; +} +EXPORT_SYMBOL_GPL(flush_work_sync); + /* * Upon a successful return (>= 0), the caller "owns" WORK_STRUCT_PENDING bit, * so this work can't be re-armed in any way. */ static int try_to_grab_pending(struct work_struct *work) { - struct cpu_workqueue_struct *cwq; + struct global_cwq *gcwq; int ret = -1; - if (!test_and_set_bit(WORK_STRUCT_PENDING, work_data_bits(work))) + if (!test_and_set_bit(WORK_STRUCT_PENDING_BIT, work_data_bits(work))) return 0; /* * The queueing is in progress, or it is already queued. Try to * steal it from ->worklist without clearing WORK_STRUCT_PENDING. */ - - cwq = get_wq_data(work); - if (!cwq) + gcwq = get_work_gcwq(work); + if (!gcwq) return ret; - spin_lock_irq(&cwq->lock); + spin_lock_irq(&gcwq->lock); if (!list_empty(&work->entry)) { /* - * This work is queued, but perhaps we locked the wrong cwq. + * This work is queued, but perhaps we locked the wrong gcwq. * In that case we must see the new value after rmb(), see * insert_work()->wmb(). */ smp_rmb(); - if (cwq == get_wq_data(work)) { + if (gcwq == get_work_gcwq(work)) { debug_work_deactivate(work); list_del_init(&work->entry); + cwq_dec_nr_in_flight(get_work_cwq(work), + get_work_color(work), + *work_data_bits(work) & WORK_STRUCT_DELAYED); ret = 1; } } - spin_unlock_irq(&cwq->lock); + spin_unlock_irq(&gcwq->lock); return ret; } -static void wait_on_cpu_work(struct cpu_workqueue_struct *cwq, - struct work_struct *work) -{ - struct wq_barrier barr; - int running = 0; - - spin_lock_irq(&cwq->lock); - if (unlikely(cwq->current_work == work)) { - insert_wq_barrier(cwq, &barr, cwq->worklist.next); - running = 1; - } - spin_unlock_irq(&cwq->lock); - - if (unlikely(running)) { - wait_for_completion(&barr.done); - destroy_work_on_stack(&barr.work); - } -} - -static void wait_on_work(struct work_struct *work) -{ - struct cpu_workqueue_struct *cwq; - struct workqueue_struct *wq; - const struct cpumask *cpu_map; - int cpu; - - might_sleep(); - - lock_map_acquire(&work->lockdep_map); - lock_map_release(&work->lockdep_map); - - cwq = get_wq_data(work); - if (!cwq) - return; - - wq = cwq->wq; - cpu_map = wq_cpu_map(wq); - - for_each_cpu(cpu, cpu_map) - wait_on_cpu_work(per_cpu_ptr(wq->cpu_wq, cpu), work); -} - -static int __cancel_work_timer(struct work_struct *work, +static bool __cancel_work_timer(struct work_struct *work, struct timer_list* timer) { int ret; @@ -681,54 +2562,91 @@ static int __cancel_work_timer(struct work_struct *work, wait_on_work(work); } while (unlikely(ret < 0)); - clear_wq_data(work); + clear_work_data(work); return ret; } /** - * cancel_work_sync - block until a work_struct's callback has terminated - * @work: the work which is to be flushed - * - * Returns true if @work was pending. + * cancel_work_sync - cancel a work and wait for it to finish + * @work: the work to cancel * - * cancel_work_sync() will cancel the work if it is queued. If the work's - * callback appears to be running, cancel_work_sync() will block until it - * has completed. - * - * It is possible to use this function if the work re-queues itself. It can - * cancel the work even if it migrates to another workqueue, however in that - * case it only guarantees that work->func() has completed on the last queued - * workqueue. + * Cancel @work and wait for its execution to finish. This function + * can be used even if the work re-queues itself or migrates to + * another workqueue. On return from this function, @work is + * guaranteed to be not pending or executing on any CPU. * - * cancel_work_sync(&delayed_work->work) should be used only if ->timer is not - * pending, otherwise it goes into a busy-wait loop until the timer expires. + * cancel_work_sync(&delayed_work->work) must not be used for + * delayed_work's. Use cancel_delayed_work_sync() instead. * - * The caller must ensure that workqueue_struct on which this work was last + * The caller must ensure that the workqueue on which @work was last * queued can't be destroyed before this function returns. + * + * RETURNS: + * %true if @work was pending, %false otherwise. */ -int cancel_work_sync(struct work_struct *work) +bool cancel_work_sync(struct work_struct *work) { return __cancel_work_timer(work, NULL); } EXPORT_SYMBOL_GPL(cancel_work_sync); /** - * cancel_delayed_work_sync - reliably kill off a delayed work. - * @dwork: the delayed work struct + * flush_delayed_work - wait for a dwork to finish executing the last queueing + * @dwork: the delayed work to flush * - * Returns true if @dwork was pending. + * Delayed timer is cancelled and the pending work is queued for + * immediate execution. Like flush_work(), this function only + * considers the last queueing instance of @dwork. * - * It is possible to use this function if @dwork rearms itself via queue_work() - * or queue_delayed_work(). See also the comment for cancel_work_sync(). + * RETURNS: + * %true if flush_work() waited for the work to finish execution, + * %false if it was already idle. */ -int cancel_delayed_work_sync(struct delayed_work *dwork) +bool flush_delayed_work(struct delayed_work *dwork) +{ + if (del_timer_sync(&dwork->timer)) + __queue_work(raw_smp_processor_id(), + get_work_cwq(&dwork->work)->wq, &dwork->work); + return flush_work(&dwork->work); +} +EXPORT_SYMBOL(flush_delayed_work); + +/** + * flush_delayed_work_sync - wait for a dwork to finish + * @dwork: the delayed work to flush + * + * Delayed timer is cancelled and the pending work is queued for + * execution immediately. Other than timer handling, its behavior + * is identical to flush_work_sync(). + * + * RETURNS: + * %true if flush_work_sync() waited for the work to finish execution, + * %false if it was already idle. + */ +bool flush_delayed_work_sync(struct delayed_work *dwork) +{ + if (del_timer_sync(&dwork->timer)) + __queue_work(raw_smp_processor_id(), + get_work_cwq(&dwork->work)->wq, &dwork->work); + return flush_work_sync(&dwork->work); +} +EXPORT_SYMBOL(flush_delayed_work_sync); + +/** + * cancel_delayed_work_sync - cancel a delayed work and wait for it to finish + * @dwork: the delayed work cancel + * + * This is cancel_work_sync() for delayed works. + * + * RETURNS: + * %true if @dwork was pending, %false otherwise. + */ +bool cancel_delayed_work_sync(struct delayed_work *dwork) { return __cancel_work_timer(&dwork->work, &dwork->timer); } EXPORT_SYMBOL(cancel_delayed_work_sync); -static struct workqueue_struct *keventd_wq __read_mostly; - /** * schedule_work - put work task in global workqueue * @work: job to be done @@ -742,7 +2660,7 @@ static struct workqueue_struct *keventd_wq __read_mostly; */ int schedule_work(struct work_struct *work) { - return queue_work(keventd_wq, work); + return queue_work(system_wq, work); } EXPORT_SYMBOL(schedule_work); @@ -755,7 +2673,7 @@ EXPORT_SYMBOL(schedule_work); */ int schedule_work_on(int cpu, struct work_struct *work) { - return queue_work_on(cpu, keventd_wq, work); + return queue_work_on(cpu, system_wq, work); } EXPORT_SYMBOL(schedule_work_on); @@ -770,29 +2688,11 @@ EXPORT_SYMBOL(schedule_work_on); int schedule_delayed_work(struct delayed_work *dwork, unsigned long delay) { - return queue_delayed_work(keventd_wq, dwork, delay); + return queue_delayed_work(system_wq, dwork, delay); } 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(get_wq_data(&dwork->work)->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 @@ -804,24 +2704,25 @@ EXPORT_SYMBOL(flush_delayed_work); int schedule_delayed_work_on(int cpu, struct delayed_work *dwork, unsigned long delay) { - return queue_delayed_work_on(cpu, keventd_wq, dwork, delay); + return queue_delayed_work_on(cpu, system_wq, dwork, delay); } EXPORT_SYMBOL(schedule_delayed_work_on); /** - * schedule_on_each_cpu - call a function on each online CPU from keventd + * schedule_on_each_cpu - execute a function synchronously on each online CPU * @func: the function to call * - * Returns zero on success. - * Returns -ve errno on failure. - * + * schedule_on_each_cpu() executes @func on each online CPU using the + * system workqueue and blocks until all CPUs have completed. * schedule_on_each_cpu() is very slow. + * + * RETURNS: + * 0 on success, -errno on failure. */ int schedule_on_each_cpu(work_func_t func) { int cpu; - int orig = -1; - struct work_struct *works; + struct work_struct __percpu *works; works = alloc_percpu(struct work_struct); if (!works) @@ -829,23 +2730,12 @@ int schedule_on_each_cpu(work_func_t func) 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); - if (cpu != orig) - schedule_work_on(cpu, work); + 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)); @@ -881,7 +2771,7 @@ int schedule_on_each_cpu(work_func_t func) */ void flush_scheduled_work(void) { - flush_workqueue(keventd_wq); + flush_workqueue(system_wq); } EXPORT_SYMBOL(flush_scheduled_work); @@ -913,170 +2803,178 @@ EXPORT_SYMBOL_GPL(execute_in_process_context); int keventd_up(void) { - return keventd_wq != NULL; + return system_wq != NULL; } -int current_is_keventd(void) +static int alloc_cwqs(struct workqueue_struct *wq) { - struct cpu_workqueue_struct *cwq; - int cpu = raw_smp_processor_id(); /* preempt-safe: keventd is per-cpu */ - int ret = 0; - - BUG_ON(!keventd_wq); + /* + * cwqs are forced aligned according to WORK_STRUCT_FLAG_BITS. + * Make sure that the alignment isn't lower than that of + * unsigned long long. + */ + const size_t size = sizeof(struct cpu_workqueue_struct); + const size_t align = max_t(size_t, 1 << WORK_STRUCT_FLAG_BITS, + __alignof__(unsigned long long)); +#ifdef CONFIG_SMP + bool percpu = !(wq->flags & WQ_UNBOUND); +#else + bool percpu = false; +#endif - cwq = per_cpu_ptr(keventd_wq->cpu_wq, cpu); - if (current == cwq->thread) - ret = 1; + if (percpu) + wq->cpu_wq.pcpu = __alloc_percpu(size, align); + else { + void *ptr; - return ret; + /* + * Allocate enough room to align cwq and put an extra + * pointer at the end pointing back to the originally + * allocated pointer which will be used for free. + */ + ptr = kzalloc(size + align + sizeof(void *), GFP_KERNEL); + if (ptr) { + wq->cpu_wq.single = PTR_ALIGN(ptr, align); + *(void **)(wq->cpu_wq.single + 1) = ptr; + } + } + /* just in case, make sure it's actually aligned + * - this is affected by PERCPU() alignment in vmlinux.lds.S + */ + BUG_ON(!IS_ALIGNED(wq->cpu_wq.v, align)); + return wq->cpu_wq.v ? 0 : -ENOMEM; } -static struct cpu_workqueue_struct * -init_cpu_workqueue(struct workqueue_struct *wq, int cpu) +static void free_cwqs(struct workqueue_struct *wq) { - struct cpu_workqueue_struct *cwq = per_cpu_ptr(wq->cpu_wq, cpu); - - cwq->wq = wq; - spin_lock_init(&cwq->lock); - INIT_LIST_HEAD(&cwq->worklist); - init_waitqueue_head(&cwq->more_work); +#ifdef CONFIG_SMP + bool percpu = !(wq->flags & WQ_UNBOUND); +#else + bool percpu = false; +#endif - return cwq; + if (percpu) + free_percpu(wq->cpu_wq.pcpu); + else if (wq->cpu_wq.single) { + /* the pointer to free is stored right after the cwq */ + kfree(*(void **)(wq->cpu_wq.single + 1)); + } } -static int create_workqueue_thread(struct cpu_workqueue_struct *cwq, int cpu) +static int wq_clamp_max_active(int max_active, unsigned int flags, + const char *name) { - struct sched_param param = { .sched_priority = MAX_RT_PRIO-1 }; - struct workqueue_struct *wq = cwq->wq; - const char *fmt = is_wq_single_threaded(wq) ? "%s" : "%s/%d"; - struct task_struct *p; - - p = kthread_create(worker_thread, cwq, fmt, wq->name, cpu); - /* - * Nobody can add the work_struct to this cwq, - * if (caller is __create_workqueue) - * nobody should see this wq - * else // caller is CPU_UP_PREPARE - * cpu is not on cpu_online_map - * so we can abort safely. - */ - if (IS_ERR(p)) - return PTR_ERR(p); - if (cwq->wq->rt) - sched_setscheduler_nocheck(p, SCHED_FIFO, ¶m); - cwq->thread = p; + int lim = flags & WQ_UNBOUND ? WQ_UNBOUND_MAX_ACTIVE : WQ_MAX_ACTIVE; - trace_workqueue_creation(cwq->thread, cpu); + if (max_active < 1 || max_active > lim) + printk(KERN_WARNING "workqueue: max_active %d requested for %s " + "is out of range, clamping between %d and %d\n", + max_active, name, 1, lim); - return 0; + return clamp_val(max_active, 1, lim); } -static void start_workqueue_thread(struct cpu_workqueue_struct *cwq, int cpu) +struct workqueue_struct *__alloc_workqueue_key(const char *name, + unsigned int flags, + int max_active, + struct lock_class_key *key, + const char *lock_name) { - struct task_struct *p = cwq->thread; + struct workqueue_struct *wq; + unsigned int cpu; - if (p != NULL) { - if (cpu >= 0) - kthread_bind(p, cpu); - wake_up_process(p); - } -} + /* + * Workqueues which may be used during memory reclaim should + * have a rescuer to guarantee forward progress. + */ + if (flags & WQ_MEM_RECLAIM) + flags |= WQ_RESCUER; -struct workqueue_struct *__create_workqueue_key(const char *name, - int singlethread, - int freezeable, - int rt, - struct lock_class_key *key, - const char *lock_name) -{ - struct workqueue_struct *wq; - struct cpu_workqueue_struct *cwq; - int err = 0, cpu; + /* + * Unbound workqueues aren't concurrency managed and should be + * dispatched to workers immediately. + */ + if (flags & WQ_UNBOUND) + flags |= WQ_HIGHPRI; + + max_active = max_active ?: WQ_DFL_ACTIVE; + max_active = wq_clamp_max_active(max_active, flags, name); wq = kzalloc(sizeof(*wq), GFP_KERNEL); if (!wq) - return NULL; + goto err; - wq->cpu_wq = alloc_percpu(struct cpu_workqueue_struct); - if (!wq->cpu_wq) { - kfree(wq); - return NULL; - } + wq->flags = flags; + wq->saved_max_active = max_active; + mutex_init(&wq->flush_mutex); + atomic_set(&wq->nr_cwqs_to_flush, 0); + INIT_LIST_HEAD(&wq->flusher_queue); + INIT_LIST_HEAD(&wq->flusher_overflow); wq->name = name; lockdep_init_map(&wq->lockdep_map, lock_name, key, 0); - wq->singlethread = singlethread; - wq->freezeable = freezeable; - wq->rt = rt; INIT_LIST_HEAD(&wq->list); - if (singlethread) { - cwq = init_cpu_workqueue(wq, singlethread_cpu); - err = create_workqueue_thread(cwq, singlethread_cpu); - start_workqueue_thread(cwq, -1); - } else { - cpu_maps_update_begin(); - /* - * We must place this wq on list even if the code below fails. - * cpu_down(cpu) can remove cpu from cpu_populated_map before - * destroy_workqueue() takes the lock, in that case we leak - * cwq[cpu]->thread. - */ - spin_lock(&workqueue_lock); - list_add(&wq->list, &workqueues); - spin_unlock(&workqueue_lock); - /* - * We must initialize cwqs for each possible cpu even if we - * are going to call destroy_workqueue() finally. Otherwise - * cpu_up() can hit the uninitialized cwq once we drop the - * lock. - */ - for_each_possible_cpu(cpu) { - cwq = init_cpu_workqueue(wq, cpu); - if (err || !cpu_online(cpu)) - continue; - err = create_workqueue_thread(cwq, cpu); - start_workqueue_thread(cwq, cpu); - } - cpu_maps_update_done(); + if (alloc_cwqs(wq) < 0) + goto err; + + for_each_cwq_cpu(cpu, wq) { + struct cpu_workqueue_struct *cwq = get_cwq(cpu, wq); + struct global_cwq *gcwq = get_gcwq(cpu); + + BUG_ON((unsigned long)cwq & WORK_STRUCT_FLAG_MASK); + cwq->gcwq = gcwq; + cwq->wq = wq; + cwq->flush_color = -1; + cwq->max_active = max_active; + INIT_LIST_HEAD(&cwq->delayed_works); } - if (err) { - destroy_workqueue(wq); - wq = NULL; + if (flags & WQ_RESCUER) { + struct worker *rescuer; + + if (!alloc_mayday_mask(&wq->mayday_mask, GFP_KERNEL)) + goto err; + + wq->rescuer = rescuer = alloc_worker(); + if (!rescuer) + goto err; + + rescuer->task = kthread_create(rescuer_thread, wq, "%s", name); + if (IS_ERR(rescuer->task)) + goto err; + + rescuer->task->flags |= PF_THREAD_BOUND; + wake_up_process(rescuer->task); } - return wq; -} -EXPORT_SYMBOL_GPL(__create_workqueue_key); -static void cleanup_workqueue_thread(struct cpu_workqueue_struct *cwq) -{ /* - * Our caller is either destroy_workqueue() or CPU_POST_DEAD, - * cpu_add_remove_lock protects cwq->thread. + * workqueue_lock protects global freeze state and workqueues + * list. Grab it, set max_active accordingly and add the new + * workqueue to workqueues list. */ - if (cwq->thread == NULL) - return; + spin_lock(&workqueue_lock); - lock_map_acquire(&cwq->wq->lockdep_map); - lock_map_release(&cwq->wq->lockdep_map); + if (workqueue_freezing && wq->flags & WQ_FREEZEABLE) + for_each_cwq_cpu(cpu, wq) + get_cwq(cpu, wq)->max_active = 0; - flush_cpu_workqueue(cwq); - /* - * If the caller is CPU_POST_DEAD and cwq->worklist was not empty, - * a concurrent flush_workqueue() can insert a barrier after us. - * However, in that case run_workqueue() won't return and check - * kthread_should_stop() until it flushes all work_struct's. - * When ->worklist becomes empty it is safe to exit because no - * more work_structs can be queued on this cwq: flush_workqueue - * 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; + list_add(&wq->list, &workqueues); + + spin_unlock(&workqueue_lock); + + return wq; +err: + if (wq) { + free_cwqs(wq); + free_mayday_mask(wq->mayday_mask); + kfree(wq->rescuer); + kfree(wq); + } + return NULL; } +EXPORT_SYMBOL_GPL(__alloc_workqueue_key); /** * destroy_workqueue - safely terminate a workqueue @@ -1086,71 +2984,544 @@ static void cleanup_workqueue_thread(struct cpu_workqueue_struct *cwq) */ void destroy_workqueue(struct workqueue_struct *wq) { - const struct cpumask *cpu_map = wq_cpu_map(wq); - int cpu; + unsigned int flush_cnt = 0; + unsigned int cpu; + + /* + * Mark @wq dying and drain all pending works. Once WQ_DYING is + * set, only chain queueing is allowed. IOW, only currently + * pending or running work items on @wq can queue further work + * items on it. @wq is flushed repeatedly until it becomes empty. + * The number of flushing is detemined by the depth of chaining and + * should be relatively short. Whine if it takes too long. + */ + wq->flags |= WQ_DYING; +reflush: + flush_workqueue(wq); + + for_each_cwq_cpu(cpu, wq) { + struct cpu_workqueue_struct *cwq = get_cwq(cpu, wq); + + if (!cwq->nr_active && list_empty(&cwq->delayed_works)) + continue; + + if (++flush_cnt == 10 || + (flush_cnt % 100 == 0 && flush_cnt <= 1000)) + printk(KERN_WARNING "workqueue %s: flush on " + "destruction isn't complete after %u tries\n", + wq->name, flush_cnt); + goto reflush; + } - cpu_maps_update_begin(); + /* + * wq list is used to freeze wq, remove from list after + * flushing is complete in case freeze races us. + */ spin_lock(&workqueue_lock); list_del(&wq->list); spin_unlock(&workqueue_lock); - for_each_cpu(cpu, cpu_map) - cleanup_workqueue_thread(per_cpu_ptr(wq->cpu_wq, cpu)); - cpu_maps_update_done(); + /* sanity check */ + for_each_cwq_cpu(cpu, wq) { + struct cpu_workqueue_struct *cwq = get_cwq(cpu, wq); + int i; - free_percpu(wq->cpu_wq); + for (i = 0; i < WORK_NR_COLORS; i++) + BUG_ON(cwq->nr_in_flight[i]); + BUG_ON(cwq->nr_active); + BUG_ON(!list_empty(&cwq->delayed_works)); + } + + if (wq->flags & WQ_RESCUER) { + kthread_stop(wq->rescuer->task); + free_mayday_mask(wq->mayday_mask); + kfree(wq->rescuer); + } + + free_cwqs(wq); kfree(wq); } EXPORT_SYMBOL_GPL(destroy_workqueue); +/** + * workqueue_set_max_active - adjust max_active of a workqueue + * @wq: target workqueue + * @max_active: new max_active value. + * + * Set max_active of @wq to @max_active. + * + * CONTEXT: + * Don't call from IRQ context. + */ +void workqueue_set_max_active(struct workqueue_struct *wq, int max_active) +{ + unsigned int cpu; + + max_active = wq_clamp_max_active(max_active, wq->flags, wq->name); + + spin_lock(&workqueue_lock); + + wq->saved_max_active = max_active; + + for_each_cwq_cpu(cpu, wq) { + struct global_cwq *gcwq = get_gcwq(cpu); + + spin_lock_irq(&gcwq->lock); + + if (!(wq->flags & WQ_FREEZEABLE) || + !(gcwq->flags & GCWQ_FREEZING)) + get_cwq(gcwq->cpu, wq)->max_active = max_active; + + spin_unlock_irq(&gcwq->lock); + } + + spin_unlock(&workqueue_lock); +} +EXPORT_SYMBOL_GPL(workqueue_set_max_active); + +/** + * workqueue_congested - test whether a workqueue is congested + * @cpu: CPU in question + * @wq: target workqueue + * + * Test whether @wq's cpu workqueue for @cpu is congested. There is + * no synchronization around this function and the test result is + * unreliable and only useful as advisory hints or for debugging. + * + * RETURNS: + * %true if congested, %false otherwise. + */ +bool workqueue_congested(unsigned int cpu, struct workqueue_struct *wq) +{ + struct cpu_workqueue_struct *cwq = get_cwq(cpu, wq); + + return !list_empty(&cwq->delayed_works); +} +EXPORT_SYMBOL_GPL(workqueue_congested); + +/** + * work_cpu - return the last known associated cpu for @work + * @work: the work of interest + * + * RETURNS: + * CPU number if @work was ever queued. WORK_CPU_NONE otherwise. + */ +unsigned int work_cpu(struct work_struct *work) +{ + struct global_cwq *gcwq = get_work_gcwq(work); + + return gcwq ? gcwq->cpu : WORK_CPU_NONE; +} +EXPORT_SYMBOL_GPL(work_cpu); + +/** + * work_busy - test whether a work is currently pending or running + * @work: the work to be tested + * + * Test whether @work is currently pending or running. There is no + * synchronization around this function and the test result is + * unreliable and only useful as advisory hints or for debugging. + * Especially for reentrant wqs, the pending state might hide the + * running state. + * + * RETURNS: + * OR'd bitmask of WORK_BUSY_* bits. + */ +unsigned int work_busy(struct work_struct *work) +{ + struct global_cwq *gcwq = get_work_gcwq(work); + unsigned long flags; + unsigned int ret = 0; + + if (!gcwq) + return false; + + spin_lock_irqsave(&gcwq->lock, flags); + + if (work_pending(work)) + ret |= WORK_BUSY_PENDING; + if (find_worker_executing_work(gcwq, work)) + ret |= WORK_BUSY_RUNNING; + + spin_unlock_irqrestore(&gcwq->lock, flags); + + return ret; +} +EXPORT_SYMBOL_GPL(work_busy); + +/* + * CPU hotplug. + * + * There are two challenges in supporting CPU hotplug. Firstly, there + * are a lot of assumptions on strong associations among work, cwq and + * gcwq which make migrating pending and scheduled works very + * difficult to implement without impacting hot paths. Secondly, + * gcwqs serve mix of short, long and very long running works making + * blocked draining impractical. + * + * This is solved by allowing a gcwq to be detached from CPU, running + * it with unbound (rogue) workers and allowing it to be reattached + * later if the cpu comes back online. A separate thread is created + * to govern a gcwq in such state and is called the trustee of the + * gcwq. + * + * Trustee states and their descriptions. + * + * START Command state used on startup. On CPU_DOWN_PREPARE, a + * new trustee is started with this state. + * + * IN_CHARGE Once started, trustee will enter this state after + * assuming the manager role and making all existing + * workers rogue. DOWN_PREPARE waits for trustee to + * enter this state. After reaching IN_CHARGE, trustee + * tries to execute the pending worklist until it's empty + * and the state is set to BUTCHER, or the state is set + * to RELEASE. + * + * BUTCHER Command state which is set by the cpu callback after + * the cpu has went down. Once this state is set trustee + * knows that there will be no new works on the worklist + * and once the worklist is empty it can proceed to + * killing idle workers. + * + * RELEASE Command state which is set by the cpu callback if the + * cpu down has been canceled or it has come online + * again. After recognizing this state, trustee stops + * trying to drain or butcher and clears ROGUE, rebinds + * all remaining workers back to the cpu and releases + * manager role. + * + * DONE Trustee will enter this state after BUTCHER or RELEASE + * is complete. + * + * trustee CPU draining + * took over down complete + * START -----------> IN_CHARGE -----------> BUTCHER -----------> DONE + * | | ^ + * | CPU is back online v return workers | + * ----------------> RELEASE -------------- + */ + +/** + * trustee_wait_event_timeout - timed event wait for trustee + * @cond: condition to wait for + * @timeout: timeout in jiffies + * + * wait_event_timeout() for trustee to use. Handles locking and + * checks for RELEASE request. + * + * CONTEXT: + * spin_lock_irq(gcwq->lock) which may be released and regrabbed + * multiple times. To be used by trustee. + * + * RETURNS: + * Positive indicating left time if @cond is satisfied, 0 if timed + * out, -1 if canceled. + */ +#define trustee_wait_event_timeout(cond, timeout) ({ \ + long __ret = (timeout); \ + while (!((cond) || (gcwq->trustee_state == TRUSTEE_RELEASE)) && \ + __ret) { \ + spin_unlock_irq(&gcwq->lock); \ + __wait_event_timeout(gcwq->trustee_wait, (cond) || \ + (gcwq->trustee_state == TRUSTEE_RELEASE), \ + __ret); \ + spin_lock_irq(&gcwq->lock); \ + } \ + gcwq->trustee_state == TRUSTEE_RELEASE ? -1 : (__ret); \ +}) + +/** + * trustee_wait_event - event wait for trustee + * @cond: condition to wait for + * + * wait_event() for trustee to use. Automatically handles locking and + * checks for CANCEL request. + * + * CONTEXT: + * spin_lock_irq(gcwq->lock) which may be released and regrabbed + * multiple times. To be used by trustee. + * + * RETURNS: + * 0 if @cond is satisfied, -1 if canceled. + */ +#define trustee_wait_event(cond) ({ \ + long __ret1; \ + __ret1 = trustee_wait_event_timeout(cond, MAX_SCHEDULE_TIMEOUT);\ + __ret1 < 0 ? -1 : 0; \ +}) + +static int __cpuinit trustee_thread(void *__gcwq) +{ + struct global_cwq *gcwq = __gcwq; + struct worker *worker; + struct work_struct *work; + struct hlist_node *pos; + long rc; + int i; + + BUG_ON(gcwq->cpu != smp_processor_id()); + + spin_lock_irq(&gcwq->lock); + /* + * Claim the manager position and make all workers rogue. + * Trustee must be bound to the target cpu and can't be + * cancelled. + */ + BUG_ON(gcwq->cpu != smp_processor_id()); + rc = trustee_wait_event(!(gcwq->flags & GCWQ_MANAGING_WORKERS)); + BUG_ON(rc < 0); + + gcwq->flags |= GCWQ_MANAGING_WORKERS; + + list_for_each_entry(worker, &gcwq->idle_list, entry) + worker->flags |= WORKER_ROGUE; + + for_each_busy_worker(worker, i, pos, gcwq) + worker->flags |= WORKER_ROGUE; + + /* + * Call schedule() so that we cross rq->lock and thus can + * guarantee sched callbacks see the rogue flag. This is + * necessary as scheduler callbacks may be invoked from other + * cpus. + */ + spin_unlock_irq(&gcwq->lock); + schedule(); + spin_lock_irq(&gcwq->lock); + + /* + * Sched callbacks are disabled now. Zap nr_running. After + * this, nr_running stays zero and need_more_worker() and + * keep_working() are always true as long as the worklist is + * not empty. + */ + atomic_set(get_gcwq_nr_running(gcwq->cpu), 0); + + spin_unlock_irq(&gcwq->lock); + del_timer_sync(&gcwq->idle_timer); + spin_lock_irq(&gcwq->lock); + + /* + * We're now in charge. Notify and proceed to drain. We need + * to keep the gcwq running during the whole CPU down + * procedure as other cpu hotunplug callbacks may need to + * flush currently running tasks. + */ + gcwq->trustee_state = TRUSTEE_IN_CHARGE; + wake_up_all(&gcwq->trustee_wait); + + /* + * The original cpu is in the process of dying and may go away + * anytime now. When that happens, we and all workers would + * be migrated to other cpus. Try draining any left work. We + * want to get it over with ASAP - spam rescuers, wake up as + * many idlers as necessary and create new ones till the + * worklist is empty. Note that if the gcwq is frozen, there + * may be frozen works in freezeable cwqs. Don't declare + * completion while frozen. + */ + while (gcwq->nr_workers != gcwq->nr_idle || + gcwq->flags & GCWQ_FREEZING || + gcwq->trustee_state == TRUSTEE_IN_CHARGE) { + int nr_works = 0; + + list_for_each_entry(work, &gcwq->worklist, entry) { + send_mayday(work); + nr_works++; + } + + list_for_each_entry(worker, &gcwq->idle_list, entry) { + if (!nr_works--) + break; + wake_up_process(worker->task); + } + + if (need_to_create_worker(gcwq)) { + spin_unlock_irq(&gcwq->lock); + worker = create_worker(gcwq, false); + spin_lock_irq(&gcwq->lock); + if (worker) { + worker->flags |= WORKER_ROGUE; + start_worker(worker); + } + } + + /* give a breather */ + if (trustee_wait_event_timeout(false, TRUSTEE_COOLDOWN) < 0) + break; + } + + /* + * Either all works have been scheduled and cpu is down, or + * cpu down has already been canceled. Wait for and butcher + * all workers till we're canceled. + */ + do { + rc = trustee_wait_event(!list_empty(&gcwq->idle_list)); + while (!list_empty(&gcwq->idle_list)) + destroy_worker(list_first_entry(&gcwq->idle_list, + struct worker, entry)); + } while (gcwq->nr_workers && rc >= 0); + + /* + * At this point, either draining has completed and no worker + * is left, or cpu down has been canceled or the cpu is being + * brought back up. There shouldn't be any idle one left. + * Tell the remaining busy ones to rebind once it finishes the + * currently scheduled works by scheduling the rebind_work. + */ + WARN_ON(!list_empty(&gcwq->idle_list)); + + for_each_busy_worker(worker, i, pos, gcwq) { + struct work_struct *rebind_work = &worker->rebind_work; + + /* + * Rebind_work may race with future cpu hotplug + * operations. Use a separate flag to mark that + * rebinding is scheduled. + */ + worker->flags |= WORKER_REBIND; + worker->flags &= ~WORKER_ROGUE; + + /* queue rebind_work, wq doesn't matter, use the default one */ + if (test_and_set_bit(WORK_STRUCT_PENDING_BIT, + work_data_bits(rebind_work))) + continue; + + debug_work_activate(rebind_work); + insert_work(get_cwq(gcwq->cpu, system_wq), rebind_work, + worker->scheduled.next, + work_color_to_flags(WORK_NO_COLOR)); + } + + /* relinquish manager role */ + gcwq->flags &= ~GCWQ_MANAGING_WORKERS; + + /* notify completion */ + gcwq->trustee = NULL; + gcwq->trustee_state = TRUSTEE_DONE; + wake_up_all(&gcwq->trustee_wait); + spin_unlock_irq(&gcwq->lock); + return 0; +} + +/** + * wait_trustee_state - wait for trustee to enter the specified state + * @gcwq: gcwq the trustee of interest belongs to + * @state: target state to wait for + * + * Wait for the trustee to reach @state. DONE is already matched. + * + * CONTEXT: + * spin_lock_irq(gcwq->lock) which may be released and regrabbed + * multiple times. To be used by cpu_callback. + */ +static void __cpuinit wait_trustee_state(struct global_cwq *gcwq, int state) +__releases(&gcwq->lock) +__acquires(&gcwq->lock) +{ + if (!(gcwq->trustee_state == state || + gcwq->trustee_state == TRUSTEE_DONE)) { + spin_unlock_irq(&gcwq->lock); + __wait_event(gcwq->trustee_wait, + gcwq->trustee_state == state || + gcwq->trustee_state == TRUSTEE_DONE); + spin_lock_irq(&gcwq->lock); + } +} + static int __devinit workqueue_cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu) { unsigned int cpu = (unsigned long)hcpu; - struct cpu_workqueue_struct *cwq; - struct workqueue_struct *wq; - int ret = NOTIFY_OK; + struct global_cwq *gcwq = get_gcwq(cpu); + struct task_struct *new_trustee = NULL; + struct worker *uninitialized_var(new_worker); + unsigned long flags; action &= ~CPU_TASKS_FROZEN; switch (action) { + case CPU_DOWN_PREPARE: + new_trustee = kthread_create(trustee_thread, gcwq, + "workqueue_trustee/%d\n", cpu); + if (IS_ERR(new_trustee)) + return notifier_from_errno(PTR_ERR(new_trustee)); + kthread_bind(new_trustee, cpu); + /* fall through */ case CPU_UP_PREPARE: - cpumask_set_cpu(cpu, cpu_populated_map); - } -undo: - list_for_each_entry(wq, &workqueues, list) { - cwq = per_cpu_ptr(wq->cpu_wq, cpu); - - switch (action) { - case CPU_UP_PREPARE: - if (!create_workqueue_thread(cwq, cpu)) - break; - printk(KERN_ERR "workqueue [%s] for %i failed\n", - wq->name, cpu); - action = CPU_UP_CANCELED; - ret = NOTIFY_BAD; - goto undo; - - case CPU_ONLINE: - start_workqueue_thread(cwq, cpu); - break; - - case CPU_UP_CANCELED: - start_workqueue_thread(cwq, -1); - case CPU_POST_DEAD: - cleanup_workqueue_thread(cwq); - break; + BUG_ON(gcwq->first_idle); + new_worker = create_worker(gcwq, false); + if (!new_worker) { + if (new_trustee) + kthread_stop(new_trustee); + return NOTIFY_BAD; } } + /* some are called w/ irq disabled, don't disturb irq status */ + spin_lock_irqsave(&gcwq->lock, flags); + switch (action) { - case CPU_UP_CANCELED: + case CPU_DOWN_PREPARE: + /* initialize trustee and tell it to acquire the gcwq */ + BUG_ON(gcwq->trustee || gcwq->trustee_state != TRUSTEE_DONE); + gcwq->trustee = new_trustee; + gcwq->trustee_state = TRUSTEE_START; + wake_up_process(gcwq->trustee); + wait_trustee_state(gcwq, TRUSTEE_IN_CHARGE); + /* fall through */ + case CPU_UP_PREPARE: + BUG_ON(gcwq->first_idle); + gcwq->first_idle = new_worker; + break; + + case CPU_DYING: + /* + * Before this, the trustee and all workers except for + * the ones which are still executing works from + * before the last CPU down must be on the cpu. After + * this, they'll all be diasporas. + */ + gcwq->flags |= GCWQ_DISASSOCIATED; + break; + case CPU_POST_DEAD: - cpumask_clear_cpu(cpu, cpu_populated_map); + gcwq->trustee_state = TRUSTEE_BUTCHER; + /* fall through */ + case CPU_UP_CANCELED: + destroy_worker(gcwq->first_idle); + gcwq->first_idle = NULL; + break; + + case CPU_DOWN_FAILED: + case CPU_ONLINE: + gcwq->flags &= ~GCWQ_DISASSOCIATED; + if (gcwq->trustee_state != TRUSTEE_DONE) { + gcwq->trustee_state = TRUSTEE_RELEASE; + wake_up_process(gcwq->trustee); + wait_trustee_state(gcwq, TRUSTEE_DONE); + } + + /* + * Trustee is done and there might be no worker left. + * Put the first_idle in and request a real manager to + * take a look. + */ + spin_unlock_irq(&gcwq->lock); + kthread_bind(gcwq->first_idle->task, cpu); + spin_lock_irq(&gcwq->lock); + gcwq->flags |= GCWQ_MANAGE_WORKERS; + start_worker(gcwq->first_idle); + gcwq->first_idle = NULL; + break; } - return ret; + spin_unlock_irqrestore(&gcwq->lock, flags); + + return notifier_from_errno(0); } #ifdef CONFIG_SMP @@ -1200,14 +3571,201 @@ long work_on_cpu(unsigned int cpu, long (*fn)(void *), void *arg) EXPORT_SYMBOL_GPL(work_on_cpu); #endif /* CONFIG_SMP */ -void __init init_workqueues(void) +#ifdef CONFIG_FREEZER + +/** + * freeze_workqueues_begin - begin freezing workqueues + * + * Start freezing workqueues. After this function returns, all + * freezeable workqueues will queue new works to their frozen_works + * list instead of gcwq->worklist. + * + * CONTEXT: + * Grabs and releases workqueue_lock and gcwq->lock's. + */ +void freeze_workqueues_begin(void) { - alloc_cpumask_var(&cpu_populated_map, GFP_KERNEL); + unsigned int cpu; + + spin_lock(&workqueue_lock); + + BUG_ON(workqueue_freezing); + workqueue_freezing = true; + + for_each_gcwq_cpu(cpu) { + struct global_cwq *gcwq = get_gcwq(cpu); + struct workqueue_struct *wq; + + spin_lock_irq(&gcwq->lock); + + BUG_ON(gcwq->flags & GCWQ_FREEZING); + gcwq->flags |= GCWQ_FREEZING; - cpumask_copy(cpu_populated_map, cpu_online_mask); - singlethread_cpu = cpumask_first(cpu_possible_mask); - cpu_singlethread_map = cpumask_of(singlethread_cpu); - hotcpu_notifier(workqueue_cpu_callback, 0); - keventd_wq = create_workqueue("events"); - BUG_ON(!keventd_wq); + list_for_each_entry(wq, &workqueues, list) { + struct cpu_workqueue_struct *cwq = get_cwq(cpu, wq); + + if (cwq && wq->flags & WQ_FREEZEABLE) + cwq->max_active = 0; + } + + spin_unlock_irq(&gcwq->lock); + } + + spin_unlock(&workqueue_lock); +} + +/** + * freeze_workqueues_busy - are freezeable workqueues still busy? + * + * Check whether freezing is complete. This function must be called + * between freeze_workqueues_begin() and thaw_workqueues(). + * + * CONTEXT: + * Grabs and releases workqueue_lock. + * + * RETURNS: + * %true if some freezeable workqueues are still busy. %false if + * freezing is complete. + */ +bool freeze_workqueues_busy(void) +{ + unsigned int cpu; + bool busy = false; + + spin_lock(&workqueue_lock); + + BUG_ON(!workqueue_freezing); + + for_each_gcwq_cpu(cpu) { + struct workqueue_struct *wq; + /* + * nr_active is monotonically decreasing. It's safe + * to peek without lock. + */ + list_for_each_entry(wq, &workqueues, list) { + struct cpu_workqueue_struct *cwq = get_cwq(cpu, wq); + + if (!cwq || !(wq->flags & WQ_FREEZEABLE)) + continue; + + BUG_ON(cwq->nr_active < 0); + if (cwq->nr_active) { + busy = true; + goto out_unlock; + } + } + } +out_unlock: + spin_unlock(&workqueue_lock); + return busy; +} + +/** + * thaw_workqueues - thaw workqueues + * + * Thaw workqueues. Normal queueing is restored and all collected + * frozen works are transferred to their respective gcwq worklists. + * + * CONTEXT: + * Grabs and releases workqueue_lock and gcwq->lock's. + */ +void thaw_workqueues(void) +{ + unsigned int cpu; + + spin_lock(&workqueue_lock); + + if (!workqueue_freezing) + goto out_unlock; + + for_each_gcwq_cpu(cpu) { + struct global_cwq *gcwq = get_gcwq(cpu); + struct workqueue_struct *wq; + + spin_lock_irq(&gcwq->lock); + + BUG_ON(!(gcwq->flags & GCWQ_FREEZING)); + gcwq->flags &= ~GCWQ_FREEZING; + + list_for_each_entry(wq, &workqueues, list) { + struct cpu_workqueue_struct *cwq = get_cwq(cpu, wq); + + if (!cwq || !(wq->flags & WQ_FREEZEABLE)) + continue; + + /* restore max_active and repopulate worklist */ + cwq->max_active = wq->saved_max_active; + + while (!list_empty(&cwq->delayed_works) && + cwq->nr_active < cwq->max_active) + cwq_activate_first_delayed(cwq); + } + + wake_up_worker(gcwq); + + spin_unlock_irq(&gcwq->lock); + } + + workqueue_freezing = false; +out_unlock: + spin_unlock(&workqueue_lock); +} +#endif /* CONFIG_FREEZER */ + +static int __init init_workqueues(void) +{ + unsigned int cpu; + int i; + + cpu_notifier(workqueue_cpu_callback, CPU_PRI_WORKQUEUE); + + /* initialize gcwqs */ + for_each_gcwq_cpu(cpu) { + struct global_cwq *gcwq = get_gcwq(cpu); + + spin_lock_init(&gcwq->lock); + INIT_LIST_HEAD(&gcwq->worklist); + gcwq->cpu = cpu; + gcwq->flags |= GCWQ_DISASSOCIATED; + + INIT_LIST_HEAD(&gcwq->idle_list); + for (i = 0; i < BUSY_WORKER_HASH_SIZE; i++) + INIT_HLIST_HEAD(&gcwq->busy_hash[i]); + + init_timer_deferrable(&gcwq->idle_timer); + gcwq->idle_timer.function = idle_worker_timeout; + gcwq->idle_timer.data = (unsigned long)gcwq; + + setup_timer(&gcwq->mayday_timer, gcwq_mayday_timeout, + (unsigned long)gcwq); + + ida_init(&gcwq->worker_ida); + + gcwq->trustee_state = TRUSTEE_DONE; + init_waitqueue_head(&gcwq->trustee_wait); + } + + /* create the initial worker */ + for_each_online_gcwq_cpu(cpu) { + struct global_cwq *gcwq = get_gcwq(cpu); + struct worker *worker; + + if (cpu != WORK_CPU_UNBOUND) + gcwq->flags &= ~GCWQ_DISASSOCIATED; + worker = create_worker(gcwq, true); + BUG_ON(!worker); + spin_lock_irq(&gcwq->lock); + start_worker(worker); + spin_unlock_irq(&gcwq->lock); + } + + system_wq = alloc_workqueue("events", 0, 0); + system_long_wq = alloc_workqueue("events_long", 0, 0); + system_nrt_wq = alloc_workqueue("events_nrt", WQ_NON_REENTRANT, 0); + system_unbound_wq = alloc_workqueue("events_unbound", WQ_UNBOUND, + WQ_UNBOUND_MAX_ACTIVE); + BUG_ON(!system_wq || !system_long_wq || !system_nrt_wq || + !system_unbound_wq); + return 0; } +early_initcall(init_workqueues); diff --git a/kernel/workqueue_sched.h b/kernel/workqueue_sched.h new file mode 100644 index 00000000000..2d10fc98dc7 --- /dev/null +++ b/kernel/workqueue_sched.h @@ -0,0 +1,9 @@ +/* + * kernel/workqueue_sched.h + * + * Scheduler hooks for concurrency managed workqueue. Only to be + * included from sched.c and workqueue.c. + */ +void wq_worker_waking_up(struct task_struct *task, unsigned int cpu); +struct task_struct *wq_worker_sleeping(struct task_struct *task, + unsigned int cpu); |