diff options
232 files changed, 4587 insertions, 1831 deletions
diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt index 5c8695a3d13..2d7ea85075b 100644 --- a/Documentation/feature-removal-schedule.txt +++ b/Documentation/feature-removal-schedule.txt @@ -62,7 +62,7 @@ Who: Dan Dennedy <dan@dennedy.org>, Stefan Richter <stefanr@s5r6.in-berlin.de> What: old NCR53C9x driver When: October 2007 Why: Replaced by the much better esp_scsi driver. Actual low-level - driver can ported over almost trivially. + driver can be ported over almost trivially. Who: David Miller <davem@davemloft.net> Christoph Hellwig <hch@lst.de> diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index aae2282600c..ce91560229f 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -1132,9 +1132,9 @@ and is between 256 and 4096 characters. It is defined in the file when set. Format: <int> - noaliencache [MM, NUMA] Disables the allcoation of alien caches in - the slab allocator. Saves per-node memory, but will - impact performance on real NUMA hardware. + noaliencache [MM, NUMA, SLAB] Disables the allocation of alien + caches in the slab allocator. Saves per-node memory, + but will impact performance. noalign [KNL,ARM] @@ -1613,6 +1613,37 @@ and is between 256 and 4096 characters. It is defined in the file slram= [HW,MTD] + slub_debug [MM, SLUB] + Enabling slub_debug allows one to determine the culprit + if slab objects become corrupted. Enabling slub_debug + creates guard zones around objects and poisons objects + when not in use. Also tracks the last alloc / free. + For more information see Documentation/vm/slub.txt. + + slub_max_order= [MM, SLUB] + Determines the maximum allowed order for slabs. Setting + this too high may cause fragmentation. + For more information see Documentation/vm/slub.txt. + + slub_min_objects= [MM, SLUB] + The minimum objects per slab. SLUB will increase the + slab order up to slub_max_order to generate a + sufficiently big slab to satisfy the number of objects. + The higher the number of objects the smaller the overhead + of tracking slabs. + For more information see Documentation/vm/slub.txt. + + slub_min_order= [MM, SLUB] + Determines the mininum page order for slabs. Must be + lower than slub_max_order + For more information see Documentation/vm/slub.txt. + + slub_nomerge [MM, SLUB] + Disable merging of slabs of similar size. May be + necessary if there is some reason to distinguish + allocs to different slabs. + For more information see Documentation/vm/slub.txt. + smart2= [HW] Format: <io1>[,<io2>[,...,<io8>]] diff --git a/Documentation/vm/slub.txt b/Documentation/vm/slub.txt index 727c8d81aea..1523320abd8 100644 --- a/Documentation/vm/slub.txt +++ b/Documentation/vm/slub.txt @@ -1,13 +1,9 @@ Short users guide for SLUB -------------------------- -First of all slub should transparently replace SLAB. If you enable -SLUB then everything should work the same (Note the word "should". -There is likely not much value in that word at this point). - The basic philosophy of SLUB is very different from SLAB. SLAB requires rebuilding the kernel to activate debug options for all -SLABS. SLUB always includes full debugging but its off by default. +slab caches. SLUB always includes full debugging but it is off by default. SLUB can enable debugging only for selected slabs in order to avoid an impact on overall system performance which may make a bug more difficult to find. @@ -76,13 +72,28 @@ of objects. Careful with tracing: It may spew out lots of information and never stop if used on the wrong slab. -SLAB Merging +Slab merging ------------ -If no debugging is specified then SLUB may merge similar slabs together +If no debug options are specified then SLUB may merge similar slabs together in order to reduce overhead and increase cache hotness of objects. slabinfo -a displays which slabs were merged together. +Slab validation +--------------- + +SLUB can validate all object if the kernel was booted with slub_debug. In +order to do so you must have the slabinfo tool. Then you can do + +slabinfo -v + +which will test all objects. Output will be generated to the syslog. + +This also works in a more limited way if boot was without slab debug. +In that case slabinfo -v simply tests all reachable objects. Usually +these are in the cpu slabs and the partial slabs. Full slabs are not +tracked by SLUB in a non debug situation. + Getting more performance ------------------------ @@ -91,9 +102,9 @@ list_lock once in a while to deal with partial slabs. That overhead is governed by the order of the allocation for each slab. The allocations can be influenced by kernel parameters: -slub_min_objects=x (default 8) +slub_min_objects=x (default 4) slub_min_order=x (default 0) -slub_max_order=x (default 4) +slub_max_order=x (default 1) slub_min_objects allows to specify how many objects must at least fit into one slab in order for the allocation order to be acceptable. @@ -109,5 +120,107 @@ longer be checked. This is useful to avoid SLUB trying to generate super large order pages to fit slub_min_objects of a slab cache with large object sizes into one high order page. - -Christoph Lameter, <clameter@sgi.com>, April 10, 2007 +SLUB Debug output +----------------- + +Here is a sample of slub debug output: + +*** SLUB kmalloc-8: Redzone Active@0xc90f6d20 slab 0xc528c530 offset=3360 flags=0x400000c3 inuse=61 freelist=0xc90f6d58 + Bytes b4 0xc90f6d10: 00 00 00 00 00 00 00 00 5a 5a 5a 5a 5a 5a 5a 5a ........ZZZZZZZZ + Object 0xc90f6d20: 31 30 31 39 2e 30 30 35 1019.005 + Redzone 0xc90f6d28: 00 cc cc cc . +FreePointer 0xc90f6d2c -> 0xc90f6d58 +Last alloc: get_modalias+0x61/0xf5 jiffies_ago=53 cpu=1 pid=554 +Filler 0xc90f6d50: 5a 5a 5a 5a 5a 5a 5a 5a ZZZZZZZZ + [<c010523d>] dump_trace+0x63/0x1eb + [<c01053df>] show_trace_log_lvl+0x1a/0x2f + [<c010601d>] show_trace+0x12/0x14 + [<c0106035>] dump_stack+0x16/0x18 + [<c017e0fa>] object_err+0x143/0x14b + [<c017e2cc>] check_object+0x66/0x234 + [<c017eb43>] __slab_free+0x239/0x384 + [<c017f446>] kfree+0xa6/0xc6 + [<c02e2335>] get_modalias+0xb9/0xf5 + [<c02e23b7>] dmi_dev_uevent+0x27/0x3c + [<c027866a>] dev_uevent+0x1ad/0x1da + [<c0205024>] kobject_uevent_env+0x20a/0x45b + [<c020527f>] kobject_uevent+0xa/0xf + [<c02779f1>] store_uevent+0x4f/0x58 + [<c027758e>] dev_attr_store+0x29/0x2f + [<c01bec4f>] sysfs_write_file+0x16e/0x19c + [<c0183ba7>] vfs_write+0xd1/0x15a + [<c01841d7>] sys_write+0x3d/0x72 + [<c0104112>] sysenter_past_esp+0x5f/0x99 + [<b7f7b410>] 0xb7f7b410 + ======================= +@@@ SLUB kmalloc-8: Restoring redzone (0xcc) from 0xc90f6d28-0xc90f6d2b + + + +If SLUB encounters a corrupted object then it will perform the following +actions: + +1. Isolation and report of the issue + +This will be a message in the system log starting with + +*** SLUB <slab cache affected>: <What went wrong>@<object address> +offset=<offset of object into slab> flags=<slabflags> +inuse=<objects in use in this slab> freelist=<first free object in slab> + +2. Report on how the problem was dealt with in order to ensure the continued +operation of the system. + +These are messages in the system log beginning with + +@@@ SLUB <slab cache affected>: <corrective action taken> + + +In the above sample SLUB found that the Redzone of an active object has +been overwritten. Here a string of 8 characters was written into a slab that +has the length of 8 characters. However, a 8 character string needs a +terminating 0. That zero has overwritten the first byte of the Redzone field. +After reporting the details of the issue encountered the @@@ SLUB message +tell us that SLUB has restored the redzone to its proper value and then +system operations continue. + +Various types of lines can follow the @@@ SLUB line: + +Bytes b4 <address> : <bytes> + Show a few bytes before the object where the problem was detected. + Can be useful if the corruption does not stop with the start of the + object. + +Object <address> : <bytes> + The bytes of the object. If the object is inactive then the bytes + typically contain poisoning values. Any non-poison value shows a + corruption by a write after free. + +Redzone <address> : <bytes> + The redzone following the object. The redzone is used to detect + writes after the object. All bytes should always have the same + value. If there is any deviation then it is due to a write after + the object boundary. + +Freepointer + The pointer to the next free object in the slab. May become + corrupted if overwriting continues after the red zone. + +Last alloc: +Last free: + Shows the address from which the object was allocated/freed last. + We note the pid, the time and the CPU that did so. This is usually + the most useful information to figure out where things went wrong. + Here get_modalias() did an kmalloc(8) instead of a kmalloc(9). + +Filler <address> : <bytes> + Unused data to fill up the space in order to get the next object + properly aligned. In the debug case we make sure that there are + at least 4 bytes of filler. This allow for the detection of writes + before the object. + +Following the filler will be a stackdump. That stackdump describes the +location where the error was detected. The cause of the corruption is more +likely to be found by looking at the information about the last alloc / free. + +Christoph Lameter, <clameter@sgi.com>, May 23, 2007 diff --git a/MAINTAINERS b/MAINTAINERS index 953291d08c7..4cc17b993b6 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -2334,7 +2334,7 @@ S: Maintained MEGARAID SCSI DRIVERS P: Neela Syam Kolli -M: Neela.Kolli@engenio.com +M: megaraidlinux@lsi.com S: linux-scsi@vger.kernel.org W: http://megaraid.lsilogic.com S: Maintained @@ -1,8 +1,8 @@ VERSION = 2 PATCHLEVEL = 6 SUBLEVEL = 22 -EXTRAVERSION = -rc2 -NAME = Nocturnal Monster Puppy +EXTRAVERSION = -rc3 +NAME = Jeff Thinks I Should Change This, But To What? # *DOCUMENTATION* # To see a list of typical targets execute "make help" diff --git a/arch/alpha/kernel/entry.S b/arch/alpha/kernel/entry.S index c95e95e1ab0..debc8f03886 100644 --- a/arch/alpha/kernel/entry.S +++ b/arch/alpha/kernel/entry.S @@ -391,11 +391,10 @@ $work_resched: bne $2, $work_resched $work_notifysig: - mov $sp, $17 + mov $sp, $16 br $1, do_switch_stack - mov $5, $21 - mov $sp, $18 - mov $31, $16 + mov $sp, $17 + mov $5, $18 jsr $26, do_notify_resume bsr $1, undo_switch_stack br restore_all diff --git a/arch/alpha/kernel/pci_iommu.c b/arch/alpha/kernel/pci_iommu.c index 6e7d1fe6e93..28c84e55feb 100644 --- a/arch/alpha/kernel/pci_iommu.c +++ b/arch/alpha/kernel/pci_iommu.c @@ -7,6 +7,7 @@ #include <linux/pci.h> #include <linux/slab.h> #include <linux/bootmem.h> +#include <linux/log2.h> #include <asm/io.h> #include <asm/hwrpb.h> @@ -53,7 +54,7 @@ size_for_memory(unsigned long max) { unsigned long mem = max_low_pfn << PAGE_SHIFT; if (mem < max) - max = 1UL << ceil_log2(mem); + max = roundup_pow_of_two(mem); return max; } diff --git a/arch/alpha/kernel/setup.c b/arch/alpha/kernel/setup.c index 915f26345c4..bd5e68cd61e 100644 --- a/arch/alpha/kernel/setup.c +++ b/arch/alpha/kernel/setup.c @@ -43,6 +43,7 @@ #include <linux/notifier.h> #include <asm/setup.h> #include <asm/io.h> +#include <linux/log2.h> extern struct atomic_notifier_head panic_notifier_list; static int alpha_panic_event(struct notifier_block *, unsigned long, void *); @@ -1303,7 +1304,7 @@ external_cache_probe(int minsize, int width) long size = minsize, maxsize = MAX_BCACHE_SIZE * 2; if (maxsize > (max_low_pfn + 1) << PAGE_SHIFT) - maxsize = 1 << (floor_log2(max_low_pfn + 1) + PAGE_SHIFT); + maxsize = 1 << (ilog2(max_low_pfn + 1) + PAGE_SHIFT); /* Get the first block cached. */ read_mem_block(__va(0), stride, size); diff --git a/arch/alpha/kernel/signal.c b/arch/alpha/kernel/signal.c index 7f64aa767d5..410af4f3140 100644 --- a/arch/alpha/kernel/signal.c +++ b/arch/alpha/kernel/signal.c @@ -32,8 +32,8 @@ #define _BLOCKABLE (~(sigmask(SIGKILL) | sigmask(SIGSTOP))) asmlinkage void ret_from_sys_call(void); -static int do_signal(sigset_t *, struct pt_regs *, struct switch_stack *, - unsigned long, unsigned long); +static void do_signal(struct pt_regs *, struct switch_stack *, + unsigned long, unsigned long); /* @@ -146,11 +146,9 @@ sys_rt_sigaction(int sig, const struct sigaction __user *act, asmlinkage int do_sigsuspend(old_sigset_t mask, struct pt_regs *regs, struct switch_stack *sw) { - sigset_t oldset; - mask &= _BLOCKABLE; spin_lock_irq(¤t->sighand->siglock); - oldset = current->blocked; + current->saved_sigmask = current->blocked; siginitset(¤t->blocked, mask); recalc_sigpending(); spin_unlock_irq(¤t->sighand->siglock); @@ -160,19 +158,17 @@ do_sigsuspend(old_sigset_t mask, struct pt_regs *regs, struct switch_stack *sw) regs->r0 = EINTR; regs->r19 = 1; - while (1) { - current->state = TASK_INTERRUPTIBLE; - schedule(); - if (do_signal(&oldset, regs, sw, 0, 0)) - return -EINTR; - } + current->state = TASK_INTERRUPTIBLE; + schedule(); + set_thread_flag(TIF_RESTORE_SIGMASK); + return -ERESTARTNOHAND; } asmlinkage int do_rt_sigsuspend(sigset_t __user *uset, size_t sigsetsize, struct pt_regs *regs, struct switch_stack *sw) { - sigset_t oldset, set; + sigset_t set; /* XXX: Don't preclude handling different sized sigset_t's. */ if (sigsetsize != sizeof(sigset_t)) @@ -182,7 +178,7 @@ do_rt_sigsuspend(sigset_t __user *uset, size_t sigsetsize, sigdelsetmask(&set, ~_BLOCKABLE); spin_lock_irq(¤t->sighand->siglock); - oldset = current->blocked; + current->saved_sigmask = current->blocked; current->blocked = set; recalc_sigpending(); spin_unlock_irq(¤t->sighand->siglock); @@ -192,12 +188,10 @@ do_rt_sigsuspend(sigset_t __user *uset, size_t sigsetsize, regs->r0 = EINTR; regs->r19 = 1; - while (1) { - current->state = TASK_INTERRUPTIBLE; - schedule(); - if (do_signal(&oldset, regs, sw, 0, 0)) - return -EINTR; - } + current->state = TASK_INTERRUPTIBLE; + schedule(); + set_thread_flag(TIF_RESTORE_SIGMASK); + return -ERESTARTNOHAND; } asmlinkage int @@ -436,7 +430,7 @@ setup_sigcontext(struct sigcontext __user *sc, struct pt_regs *regs, return err; } -static void +static int setup_frame(int sig, struct k_sigaction *ka, sigset_t *set, struct pt_regs *regs, struct switch_stack * sw) { @@ -481,13 +475,14 @@ setup_frame(int sig, struct k_sigaction *ka, sigset_t *set, current->comm, current->pid, frame, regs->pc, regs->r26); #endif - return; + return 0; give_sigsegv: force_sigsegv(sig, current); + return -EFAULT; } -static void +static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, sigset_t *set, struct pt_regs *regs, struct switch_stack * sw) { @@ -543,34 +538,38 @@ setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, current->comm, current->pid, frame, regs->pc, regs->r26); #endif - return; + return 0; give_sigsegv: force_sigsegv(sig, current); + return -EFAULT; } /* * OK, we're invoking a handler. */ -static inline void +static inline int handle_signal(int sig, struct k_sigaction *ka, siginfo_t *info, sigset_t *oldset, struct pt_regs * regs, struct switch_stack *sw) { + int ret; + if (ka->sa.sa_flags & SA_SIGINFO) - setup_rt_frame(sig, ka, info, oldset, regs, sw); + ret = setup_rt_frame(sig, ka, info, oldset, regs, sw); else - setup_frame(sig, ka, oldset, regs, sw); + ret = setup_frame(sig, ka, oldset, regs, sw); - if (ka->sa.sa_flags & SA_RESETHAND) - ka->sa.sa_handler = SIG_DFL; + if (ret == 0) { + spin_lock_irq(¤t->sighand->siglock); + sigorsets(¤t->blocked,¤t->blocked,&ka->sa.sa_mask); + if (!(ka->sa.sa_flags & SA_NODEFER)) + sigaddset(¤t->blocked,sig); + recalc_sigpending(); + spin_unlock_irq(¤t->sighand->siglock); + } - spin_lock_irq(¤t->sighand->siglock); - sigorsets(¤t->blocked,¤t->blocked,&ka->sa.sa_mask); - if (!(ka->sa.sa_flags & SA_NODEFER)) - sigaddset(¤t->blocked,sig); - recalc_sigpending(); - spin_unlock_irq(¤t->sighand->siglock); + return ret; } static inline void @@ -611,30 +610,42 @@ syscall_restart(unsigned long r0, unsigned long r19, * restart. "r0" is also used as an indicator whether we can restart at * all (if we get here from anything but a syscall return, it will be 0) */ -static int -do_signal(sigset_t *oldset, struct pt_regs * regs, struct switch_stack * sw, +static void +do_signal(struct pt_regs * regs, struct switch_stack * sw, unsigned long r0, unsigned long r19) { siginfo_t info; int signr; unsigned long single_stepping = ptrace_cancel_bpt(current); struct k_sigaction ka; + sigset_t *oldset; - if (!oldset) + if (test_thread_flag(TIF_RESTORE_SIGMASK)) + oldset = ¤t->saved_sigmask; + else oldset = ¤t->blocked; /* This lets the debugger run, ... */ signr = get_signal_to_deliver(&info, &ka, regs, NULL); + /* ... so re-check the single stepping. */ single_stepping |= ptrace_cancel_bpt(current); if (signr > 0) { /* Whee! Actually deliver the signal. */ - if (r0) syscall_restart(r0, r19, regs, &ka); - handle_signal(signr, &ka, &info, oldset, regs, sw); + if (r0) + syscall_restart(r0, r19, regs, &ka); + if (handle_signal(signr, &ka, &info, oldset, regs, sw) == 0) { + /* A signal was successfully delivered, and the + saved sigmask was stored on the signal frame, + and will be restored by sigreturn. So we can + simply clear the restore sigmask flag. */ + if (test_thread_flag(TIF_RESTORE_SIGMASK)) + clear_thread_flag(TIF_RESTORE_SIGMASK); + } if (single_stepping) ptrace_set_bpt(current); /* re-set bpt */ - return 1; + return; } if (r0) { @@ -654,17 +665,22 @@ do_signal(sigset_t *oldset, struct pt_regs * regs, struct switch_stack * sw, break; } } + + /* If there's no signal to deliver, we just restore the saved mask. */ + if (test_thread_flag(TIF_RESTORE_SIGMASK)) { + clear_thread_flag(TIF_RESTORE_SIGMASK); + sigprocmask(SIG_SETMASK, ¤t->saved_sigmask, NULL); + } + if (single_stepping) ptrace_set_bpt(current); /* re-set breakpoint */ - - return 0; } void -do_notify_resume(sigset_t *oldset, struct pt_regs *regs, - struct switch_stack *sw, unsigned long r0, - unsigned long r19, unsigned long thread_info_flags) +do_notify_resume(struct pt_regs *regs, struct switch_stack *sw, + unsigned long thread_info_flags, + unsigned long r0, unsigned long r19) { - if (thread_info_flags & _TIF_SIGPENDING) - do_signal(oldset, regs, sw, r0, r19); + if (thread_info_flags & (_TIF_SIGPENDING | _TIF_RESTORE_SIGMASK)) + do_signal(regs, sw, r0, r19); } diff --git a/arch/alpha/kernel/systbls.S b/arch/alpha/kernel/systbls.S index f6cfe8ce3f9..79de99e32c3 100644 --- a/arch/alpha/kernel/systbls.S +++ b/arch/alpha/kernel/systbls.S @@ -465,6 +465,38 @@ sys_call_table: .quad sys_inotify_init .quad sys_inotify_add_watch /* 445 */ .quad sys_inotify_rm_watch + .quad sys_fdatasync + .quad sys_kexec_load + .quad sys_migrate_pages + .quad sys_openat /* 450 */ + .quad sys_mkdirat + .quad sys_mknodat + .quad sys_fchownat + .quad sys_futimesat + .quad sys_fstatat64 /* 455 */ + .quad sys_unlinkat + .quad sys_renameat + .quad sys_linkat + .quad sys_symlinkat + .quad sys_readlinkat /* 460 */ + .quad sys_fchmodat + .quad sys_faccessat + .quad sys_pselect6 + .quad sys_ppoll + .quad sys_unshare /* 465 */ + .quad sys_set_robust_list + .quad sys_get_robust_list + .quad sys_splice + .quad sys_sync_file_range + .quad sys_tee /* 470 */ + .quad sys_vmsplice + .quad sys_move_pages + .quad sys_getcpu + .quad sys_epoll_pwait + .quad sys_utimensat /* 475 */ + .quad sys_signalfd + .quad sys_timerfd + .quad sys_eventfd .size sys_call_table, . - sys_call_table .type sys_call_table, @object diff --git a/arch/alpha/lib/Makefile b/arch/alpha/lib/Makefile index ea098f3b629..266f78e1307 100644 --- a/arch/alpha/lib/Makefile +++ b/arch/alpha/lib/Makefile @@ -37,7 +37,8 @@ lib-y = __divqu.o __remqu.o __divlu.o __remlu.o \ $(ev6-y)clear_page.o \ $(ev6-y)copy_page.o \ fpreg.o \ - callback_srm.o srm_puts.o srm_printk.o + callback_srm.o srm_puts.o srm_printk.o \ + fls.o lib-$(CONFIG_SMP) += dec_and_lock.o diff --git a/arch/alpha/lib/fls.c b/arch/alpha/lib/fls.c new file mode 100644 index 00000000000..7ad84ea0acf --- /dev/null +++ b/arch/alpha/lib/fls.c @@ -0,0 +1,38 @@ +/* + * arch/alpha/lib/fls.c + */ + +#include <linux/module.h> +#include <asm/bitops.h> + +/* This is fls(x)-1, except zero is held to zero. This allows most + efficent input into extbl, plus it allows easy handling of fls(0)=0. */ + +const unsigned char __flsm1_tab[256] = +{ + 0, + 0, + 1, 1, + 2, 2, 2, 2, + 3, 3, 3, 3, 3, 3, 3, 3, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + + 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, + 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, + + 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, + 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, + 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, + 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, + + 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, + 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, + 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, + 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, + 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, + 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, + 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, + 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, +}; + +EXPORT_SYMBOL(__flsm1_tab); diff --git a/arch/arm/kernel/armksyms.c b/arch/arm/kernel/armksyms.c index 9179e822031..f73d62e8ab6 100644 --- a/arch/arm/kernel/armksyms.c +++ b/arch/arm/kernel/armksyms.c @@ -57,7 +57,7 @@ extern void fp_enter(void); #define EXPORT_SYMBOL_ALIAS(sym,orig) \ EXPORT_CRC_ALIAS(sym) \ static const struct kernel_symbol __ksymtab_##sym \ - __attribute_used__ __attribute__((section("__ksymtab"))) = \ + __used __attribute__((section("__ksymtab"))) = \ { (unsigned long)&orig, #sym }; /* diff --git a/arch/arm/kernel/stacktrace.c b/arch/arm/kernel/stacktrace.c index 8b63ad89d0a..ae31deb2d06 100644 --- a/arch/arm/kernel/stacktrace.c +++ b/arch/arm/kernel/stacktrace.c @@ -13,7 +13,7 @@ int walk_stackframe(unsigned long fp, unsigned long low, unsigned long high, /* * Check current frame pointer is within bounds */ - if ((fp - 12) < low || fp + 4 >= high) + if (fp < (low + 12) || fp + 4 >= high) break; frame = (struct stackframe *)(fp - 12); diff --git a/arch/arm/mach-at91/board-dk.c b/arch/arm/mach-at91/board-dk.c index 6043c38c0a9..af497896a96 100644 --- a/arch/arm/mach-at91/board-dk.c +++ b/arch/arm/mach-at91/board-dk.c @@ -132,7 +132,7 @@ static struct mtd_partition __initdata dk_nand_partition[] = { }, }; -static struct mtd_partition *nand_partitions(int size, int *num_partitions) +static struct mtd_partition * __init nand_partitions(int size, int *num_partitions) { *num_partitions = ARRAY_SIZE(dk_nand_partition); return dk_nand_partition; diff --git a/arch/arm/mach-at91/board-kb9202.c b/arch/arm/mach-at91/board-kb9202.c index 76f6e1e553e..7d9b1a278fd 100644 --- a/arch/arm/mach-at91/board-kb9202.c +++ b/arch/arm/mach-at91/board-kb9202.c @@ -96,7 +96,7 @@ static struct mtd_partition __initdata kb9202_nand_partition[] = { }, }; -static struct mtd_partition *nand_partitions(int size, int *num_partitions) +static struct mtd_partition * __init nand_partitions(int size, int *num_partitions) { *num_partitions = ARRAY_SIZE(kb9202_nand_partition); return kb9202_nand_partition; diff --git a/arch/arm/mach-at91/board-sam9261ek.c b/arch/arm/mach-at91/board-sam9261ek.c index 1f0c8a400b3..26ca8ab3f62 100644 --- a/arch/arm/mach-at91/board-sam9261ek.c +++ b/arch/arm/mach-at91/board-sam9261ek.c @@ -178,7 +178,7 @@ static struct mtd_partition __initdata ek_nand_partition[] = { }, }; -static struct mtd_partition *nand_partitions(int size, int *num_partitions) +static struct mtd_partition * __init nand_partitions(int size, int *num_partitions) { *num_partitions = ARRAY_SIZE(ek_nand_partition); return ek_nand_partition; diff --git a/arch/arm/mach-at91/board-sam9263ek.c b/arch/arm/mach-at91/board-sam9263ek.c index f57458559fb..c164c8e58ae 100644 --- a/arch/arm/mach-at91/board-sam9263ek.c +++ b/arch/arm/mach-at91/board-sam9263ek.c @@ -180,7 +180,7 @@ static struct mtd_partition __initdata ek_nand_partition[] = { }, }; -static struct mtd_partition *nand_partitions(int size, int *num_partitions) +static struct mtd_partition * __init nand_partitions(int size, int *num_partitions) { *num_partitions = ARRAY_SIZE(ek_nand_partition); return ek_nand_partition; diff --git a/arch/arm/mach-at91/board-sam9rlek.c b/arch/arm/mach-at91/board-sam9rlek.c index 30c79aca84d..9b61320f295 100644 --- a/arch/arm/mach-at91/board-sam9rlek.c +++ b/arch/arm/mach-at91/board-sam9rlek.c @@ -87,7 +87,7 @@ static struct mtd_partition __initdata ek_nand_partition[] = { }, }; -static struct mtd_partition *nand_partitions(int size, int *num_partitions) +static struct mtd_partition * __init nand_partitions(int size, int *num_partitions) { *num_partitions = ARRAY_SIZE(ek_nand_partition); return ek_nand_partition; diff --git a/arch/arm/mach-footbridge/cats-pci.c b/arch/arm/mach-footbridge/cats-pci.c index 4f984fde737..35eb232a649 100644 --- a/arch/arm/mach-footbridge/cats-pci.c +++ b/arch/arm/mach-footbridge/cats-pci.c @@ -45,7 +45,7 @@ static struct hw_pci cats_pci __initdata = { .postinit = dc21285_postinit, }; -static int cats_pci_init(void) +static int __init cats_pci_init(void) { if (machine_is_cats()) pci_common_init(&cats_pci); diff --git a/arch/arm/mach-imx/generic.c b/arch/arm/mach-imx/generic.c index 7a7fa51ec62..1c474cf709c 100644 --- a/arch/arm/mach-imx/generic.c +++ b/arch/arm/mach-imx/generic.c @@ -201,7 +201,6 @@ void __init imx_set_mmc_info(struct imxmmc_platform_data *info) { imx_mmc_device.dev.platform_data = info; } -EXPORT_SYMBOL(imx_set_mmc_info); static struct imxfb_mach_info imx_fb_info; diff --git a/arch/arm/mach-ixp2000/ixdp2400.c b/arch/arm/mach-ixp2000/ixdp2400.c index 0fdd03ab36e..ce7c15c7300 100644 --- a/arch/arm/mach-ixp2000/ixdp2400.c +++ b/arch/arm/mach-ixp2000/ixdp2400.c @@ -164,7 +164,7 @@ int __init ixdp2400_pci_init(void) subsys_initcall(ixdp2400_pci_init); -void ixdp2400_init_irq(void) +void __init ixdp2400_init_irq(void) { ixdp2x00_init_irq(IXDP2400_CPLD_INT_STAT, IXDP2400_CPLD_INT_MASK, IXDP2400_NR_IRQS); } diff --git a/arch/arm/mach-ixp2000/ixdp2800.c b/arch/arm/mach-ixp2000/ixdp2800.c index 70d247f09a7..14f09b80ab7 100644 --- a/arch/arm/mach-ixp2000/ixdp2800.c +++ b/arch/arm/mach-ixp2000/ixdp2800.c @@ -279,7 +279,7 @@ int __init ixdp2800_pci_init(void) subsys_initcall(ixdp2800_pci_init); -void ixdp2800_init_irq(void) +void __init ixdp2800_init_irq(void) { ixdp2x00_init_irq(IXDP2800_CPLD_INT_STAT, IXDP2800_CPLD_INT_MASK, IXDP2800_NR_IRQS); } diff --git a/arch/arm/mach-ixp2000/ixdp2x00.c b/arch/arm/mach-ixp2000/ixdp2x00.c index 011065b967b..73c651e83d9 100644 --- a/arch/arm/mach-ixp2000/ixdp2x00.c +++ b/arch/arm/mach-ixp2000/ixdp2x00.c @@ -145,7 +145,7 @@ static struct irq_chip ixdp2x00_cpld_irq_chip = { .unmask = ixdp2x00_irq_unmask }; -void ixdp2x00_init_irq(volatile unsigned long *stat_reg, volatile unsigned long *mask_reg, unsigned long nr_irqs) +void __init ixdp2x00_init_irq(volatile unsigned long *stat_reg, volatile unsigned long *mask_reg, unsigned long nr_irqs) { unsigned int irq; diff --git a/arch/arm/mach-ixp23xx/ixdp2351.c b/arch/arm/mach-ixp23xx/ixdp2351.c index 7a86a2516ea..c41a6b5a0ac 100644 --- a/arch/arm/mach-ixp23xx/ixdp2351.c +++ b/arch/arm/mach-ixp23xx/ixdp2351.c @@ -124,7 +124,7 @@ static struct irq_chip ixdp2351_intb_chip = { .unmask = ixdp2351_intb_unmask }; -void ixdp2351_init_irq(void) +void __init ixdp2351_init_irq(void) { int irq; diff --git a/arch/arm/mach-ixp23xx/pci.c b/arch/arm/mach-ixp23xx/pci.c index ac7d43d23c2..227f808dc0e 100644 --- a/arch/arm/mach-ixp23xx/pci.c +++ b/arch/arm/mach-ixp23xx/pci.c @@ -284,7 +284,7 @@ int ixp23xx_pci_setup(int nr, struct pci_sys_data *sys) return 1; } -void ixp23xx_pci_slave_init(void) +void __init ixp23xx_pci_slave_init(void) { ixp23xx_pci_common_init(); } diff --git a/arch/arm/mach-ixp23xx/roadrunner.c b/arch/arm/mach-ixp23xx/roadrunner.c index d06e21b70de..e35644961aa 100644 --- a/arch/arm/mach-ixp23xx/roadrunner.c +++ b/arch/arm/mach-ixp23xx/roadrunner.c @@ -110,7 +110,7 @@ static int __init roadrunner_map_irq(struct pci_dev *dev, u8 idsel, u8 pin) return NO_IRQ; } -static void roadrunner_pci_preinit(void) +static void __init roadrunner_pci_preinit(void) { set_irq_type(IRQ_ROADRUNNER_PCI_INTC, IRQT_LOW); set_irq_type(IRQ_ROADRUNNER_PCI_INTD, IRQT_LOW); diff --git a/arch/arm/mach-ixp4xx/Kconfig b/arch/arm/mach-ixp4xx/Kconfig index 9715ef506c2..060909870b5 100644 --- a/arch/arm/mach-ixp4xx/Kconfig +++ b/arch/arm/mach-ixp4xx/Kconfig @@ -104,9 +104,6 @@ config MACH_DSMG600 DSM-G600 RevA device. For more information on this platform, see http://www.nslu2-linux.org/wiki/DSMG600/HomePage -# -# Avila and IXDP share the same source for now. Will change in future -# config ARCH_IXDP4XX bool depends on ARCH_IXDP425 || MACH_IXDP465 || MACH_KIXRP435 diff --git a/arch/arm/mach-ixp4xx/common.c b/arch/arm/mach-ixp4xx/common.c index 64685da1462..8112f726ffa 100644 --- a/arch/arm/mach-ixp4xx/common.c +++ b/arch/arm/mach-ixp4xx/common.c @@ -283,7 +283,7 @@ static struct irqaction ixp4xx_timer_irq = { .handler = ixp4xx_timer_interrupt, }; -static void __init ixp4xx_timer_init(void) +void __init ixp4xx_timer_init(void) { /* Reset/disable counter */ *IXP4XX_OSRT1 = 0; diff --git a/arch/arm/mach-ixp4xx/coyote-pci.c b/arch/arm/mach-ixp4xx/coyote-pci.c index 7bc94f3def1..ad2e5b97966 100644 --- a/arch/arm/mach-ixp4xx/coyote-pci.c +++ b/arch/arm/mach-ixp4xx/coyote-pci.c @@ -25,10 +25,6 @@ #include <asm/mach/pci.h> -extern void ixp4xx_pci_preinit(void); -extern int ixp4xx_setup(int nr, struct pci_sys_data *sys); -extern struct pci_bus *ixp4xx_scan_bus(int nr, struct pci_sys_data *sys); - void __init coyote_pci_preinit(void) { set_irq_type(IRQ_COYOTE_PCI_SLOT0, IRQT_LOW); diff --git a/arch/arm/mach-ixp4xx/dsmg600-setup.c b/arch/arm/mach-ixp4xx/dsmg600-setup.c index 1caff65e22c..1e75e105c4f 100644 --- a/arch/arm/mach-ixp4xx/dsmg600-setup.c +++ b/arch/arm/mach-ixp4xx/dsmg600-setup.c @@ -18,6 +18,7 @@ #include <asm/mach-types.h> #include <asm/mach/arch.h> #include <asm/mach/flash.h> +#include <asm/mach/time.h> static struct flash_platform_data dsmg600_flash_data = { .map_name = "cfi_probe", @@ -128,6 +129,19 @@ static void dsmg600_power_off(void) gpio_line_set(DSMG600_PO_GPIO, IXP4XX_GPIO_HIGH); } +static void __init dsmg600_timer_init(void) +{ + /* The xtal on this machine is non-standard. */ + ixp4xx_timer_freq = DSMG600_FREQ; + + /* Call standard timer_init function. */ + ixp4xx_timer_init(); +} + +static struct sys_timer dsmg600_timer = { + .init = dsmg600_timer_init, +}; + static void __init dsmg600_init(void) { ixp4xx_sys_init(); @@ -155,21 +169,13 @@ static void __init dsmg600_init(void) #endif } -static void __init dsmg600_fixup(struct machine_desc *desc, - struct tag *tags, char **cmdline, struct meminfo *mi) -{ - /* The xtal on this machine is non-standard. */ - ixp4xx_timer_freq = DSMG600_FREQ; -} - MACHINE_START(DSMG600, "D-Link DSM-G600 RevA") /* Maintainer: www.nslu2-linux.org */ .phys_io = IXP4XX_PERIPHERAL_BASE_PHYS, .io_pg_offst = ((IXP4XX_PERIPHERAL_BASE_VIRT) >> 18) & 0xFFFC, .boot_params = 0x00000100, - .fixup = dsmg600_fixup, .map_io = ixp4xx_map_io, .init_irq = ixp4xx_init_irq, - .timer = &ixp4xx_timer, + .timer = &dsmg600_timer, .init_machine = dsmg600_init, MACHINE_END diff --git a/arch/arm/mach-ixp4xx/ixdpg425-pci.c b/arch/arm/mach-ixp4xx/ixdpg425-pci.c index 509a95a692a..d1e75b7dc3b 100644 --- a/arch/arm/mach-ixp4xx/ixdpg425-pci.c +++ b/arch/arm/mach-ixp4xx/ixdpg425-pci.c @@ -23,10 +23,6 @@ #include <asm/mach/pci.h> -extern void ixp4xx_pci_preinit(void); -extern int ixp4xx_setup(int nr, struct pci_sys_data *sys); -extern struct pci_bus *ixp4xx_scan_bus(int nr, struct pci_sys_data *sys); - void __init ixdpg425_pci_preinit(void) { set_irq_type(IRQ_IXP4XX_GPIO6, IRQT_LOW); diff --git a/arch/arm/mach-ixp4xx/nas100d-setup.c b/arch/arm/mach-ixp4xx/nas100d-setup.c index 9a31444d921..78a17413cec 100644 --- a/arch/arm/mach-ixp4xx/nas100d-setup.c +++ b/arch/arm/mach-ixp4xx/nas100d-setup.c @@ -155,7 +155,8 @@ static void __init nas100d_init(void) pm_power_off = nas100d_power_off; - /* This is only useful on a modified machine, but it is valuable + /* + * This is only useful on a modified machine, but it is valuable * to have it first in order to see debug messages, and so that * it does *not* get removed if platform_add_devices fails! */ diff --git a/arch/arm/mach-ixp4xx/nslu2-setup.c b/arch/arm/mach-ixp4xx/nslu2-setup.c index 162c266e5f8..9bf8ccbcacc 100644 --- a/arch/arm/mach-ixp4xx/nslu2-setup.c +++ b/arch/arm/mach-ixp4xx/nslu2-setup.c @@ -22,6 +22,7 @@ #include <asm/mach-types.h> #include <asm/mach/arch.h> #include <asm/mach/flash.h> +#include <asm/mach/time.h> static struct flash_platform_data nslu2_flash_data = { .map_name = "cfi_probe", @@ -49,26 +50,26 @@ static struct ixp4xx_i2c_pins nslu2_i2c_gpio_pins = { static struct resource nslu2_led_resources[] = { { .name = "ready", /* green led */ - .start = NSLU2_LED_GRN, - .end = NSLU2_LED_GRN, + .start = NSLU2_LED_GRN_GPIO, + .end = NSLU2_LED_GRN_GPIO, .flags = IXP4XX_GPIO_HIGH, }, { .name = "status", /* red led */ - .start = NSLU2_LED_RED, - .end = NSLU2_LED_RED, + .start = NSLU2_LED_RED_GPIO, + .end = NSLU2_LED_RED_GPIO, .flags = IXP4XX_GPIO_HIGH, }, { .name = "disk-1", - .start = NSLU2_LED_DISK1, - .end = NSLU2_LED_DISK1, + .start = NSLU2_LED_DISK1_GPIO, + .end = NSLU2_LED_DISK1_GPIO, .flags = IXP4XX_GPIO_LOW, }, { .name = "disk-2", - .start = NSLU2_LED_DISK2, - .end = NSLU2_LED_DISK2, + .start = NSLU2_LED_DISK2_GPIO, + .end = NSLU2_LED_DISK2_GPIO, .flags = IXP4XX_GPIO_LOW, }, }; @@ -157,10 +158,21 @@ static void nslu2_power_off(void) gpio_line_set(NSLU2_PO_GPIO, IXP4XX_GPIO_HIGH); } -static void __init nslu2_init(void) +static void __init nslu2_timer_init(void) { - ixp4xx_timer_freq = NSLU2_FREQ; + /* The xtal on this machine is non-standard. */ + ixp4xx_timer_freq = NSLU2_FREQ; + + /* Call standard timer_init function. */ + ixp4xx_timer_init(); +} +static struct sys_timer nslu2_timer = { + .init = nslu2_timer_init, +}; + +static void __init nslu2_init(void) +{ ixp4xx_sys_init(); nslu2_flash_resource.start = IXP4XX_EXP_BUS_BASE(0); @@ -169,7 +181,8 @@ static void __init nslu2_init(void) pm_power_off = nslu2_power_off; - /* This is only useful on a modified machine, but it is valuable + /* + * This is only useful on a modified machine, but it is valuable * to have it first in order to see debug messages, and so that * it does *not* get removed if platform_add_devices fails! */ @@ -185,6 +198,6 @@ MACHINE_START(NSLU2, "Linksys NSLU2") .boot_params = 0x00000100, .map_io = ixp4xx_map_io, .init_irq = ixp4xx_init_irq, - .timer = &ixp4xx_timer, + .timer = &nslu2_timer, .init_machine = nslu2_init, MACHINE_END diff --git a/arch/arm/mach-s3c2410/bast.h b/arch/arm/mach-s3c2410/bast.h deleted file mode 100644 index e98543742eb..00000000000 --- a/arch/arm/mach-s3c2410/bast.h +++ /dev/null @@ -1,2 +0,0 @@ -/* linux/arch/arm/mach-s3c2410/bast.h -extern void bast_init_irq(void); diff --git a/arch/arm/mach-s3c2410/mach-amlm5900.c b/arch/arm/mach-s3c2410/mach-amlm5900.c index bc308ceb91c..435adcce648 100644 --- a/arch/arm/mach-s3c2410/mach-amlm5900.c +++ b/arch/arm/mach-s3c2410/mach-amlm5900.c @@ -160,7 +160,7 @@ static struct platform_device *amlm5900_devices[] __initdata = { #endif }; -void __init amlm5900_map_io(void) +static void __init amlm5900_map_io(void) { s3c24xx_init_io(amlm5900_iodesc, ARRAY_SIZE(amlm5900_iodesc)); s3c24xx_init_clocks(0); diff --git a/arch/arm/mach-s3c2412/s3c2412.c b/arch/arm/mach-s3c2412/s3c2412.c index c602aa39f9c..782b5814ced 100644 --- a/arch/arm/mach-s3c2412/s3c2412.c +++ b/arch/arm/mach-s3c2412/s3c2412.c @@ -16,6 +16,7 @@ #include <linux/list.h> #include <linux/timer.h> #include <linux/init.h> +#include <linux/delay.h> #include <linux/sysdev.h> #include <linux/serial_core.h> #include <linux/platform_device.h> @@ -29,6 +30,7 @@ #include <asm/io.h> #include <asm/irq.h> +#include <asm/arch/reset.h> #include <asm/arch/idle.h> #include <asm/arch/regs-clock.h> @@ -38,6 +40,7 @@ #include <asm/arch/regs-gpioj.h> #include <asm/arch/regs-dsc.h> #include <asm/arch/regs-spi.h> +#include <asm/arch/regs-s3c2412.h> #include <asm/plat-s3c24xx/s3c2412.h> #include <asm/plat-s3c24xx/cpu.h> @@ -106,6 +109,23 @@ static void s3c2412_idle(void) cpu_do_idle(); } +static void s3c2412_hard_reset(void) +{ + /* errata "Watch-dog/Software Reset Problem" specifies that + * this reset must be done with the SYSCLK sourced from + * EXTCLK instead of FOUT to avoid a glitch in the reset + * mechanism. + * + * See the watchdog section of the S3C2412 manual for more + * information on this fix. + */ + + __raw_writel(0x00, S3C2412_CLKSRC); + __raw_writel(S3C2412_SWRST_RESET, S3C2412_SWRST); + + mdelay(1); +} + /* s3c2412_map_io * * register the standard cpu IO areas, and any passed in from the @@ -122,6 +142,10 @@ void __init s3c2412_map_io(struct map_desc *mach_desc, int mach_size) s3c24xx_idle = s3c2412_idle; + /* set custom reset hook */ + + s3c24xx_reset_hook = s3c2412_hard_reset; + /* register our io-tables */ iotable_init(s3c2412_iodesc, ARRAY_SIZE(s3c2412_iodesc)); diff --git a/arch/arm/mach-s3c2443/clock.c b/arch/arm/mach-s3c2443/clock.c index 5955efb5de8..58402948c47 100644 --- a/arch/arm/mach-s3c2443/clock.c +++ b/arch/arm/mach-s3c2443/clock.c @@ -394,7 +394,7 @@ static int s3c2443_setrate_usbhost(struct clk *clk, unsigned long rate) return 0; } -struct clk clk_usb_bus_host = { +static struct clk clk_usb_bus_host = { .name = "usb-bus-host-parent", .id = -1, .parent = &clk_esysclk, @@ -758,7 +758,6 @@ static struct clk init_clocks[] = { .parent = &clk_h, .enable = s3c2443_clkcon_enable_h, .ctrlbit = S3C2443_HCLKCON_CFC, - .ctrlbit = S3C2443_HCLKCON_HSMMC, }, { .name = "ssmc", .id = -1, diff --git a/arch/arm/mach-sa1100/neponset.c b/arch/arm/mach-sa1100/neponset.c index d7c038a0256..4cbf9468f65 100644 --- a/arch/arm/mach-sa1100/neponset.c +++ b/arch/arm/mach-sa1100/neponset.c @@ -139,12 +139,12 @@ static u_int neponset_get_mctrl(struct uart_port *port) return ret; } -static struct sa1100_port_fns neponset_port_fns __initdata = { +static struct sa1100_port_fns neponset_port_fns __devinitdata = { .set_mctrl = neponset_set_mctrl, .get_mctrl = neponset_get_mctrl, }; -static int neponset_probe(struct platform_device *dev) +static int __devinit neponset_probe(struct platform_device *dev) { sa1100_register_uart_fns(&neponset_port_fns); diff --git a/arch/arm/mm/Kconfig b/arch/arm/mm/Kconfig index 5f472a8b406..e7904bc92c7 100644 --- a/arch/arm/mm/Kconfig +++ b/arch/arm/mm/Kconfig @@ -379,7 +379,7 @@ config CPU_V7 select CPU_CP15_MMU select CPU_HAS_ASID select CPU_COPY_V6 if MMU - select CPU_TLB_V6 if MMU + select CPU_TLB_V7 if MMU # Figure out what processor architecture version we should be using. # This defines the compiler instruction set which depends on the machine type. @@ -498,6 +498,9 @@ config CPU_TLB_V4WBI config CPU_TLB_V6 bool +config CPU_TLB_V7 + bool + endif config CPU_HAS_ASID diff --git a/arch/arm/mm/Makefile b/arch/arm/mm/Makefile index b5bd335ff14..762702765fc 100644 --- a/arch/arm/mm/Makefile +++ b/arch/arm/mm/Makefile @@ -46,6 +46,7 @@ obj-$(CONFIG_CPU_TLB_V4WT) += tlb-v4.o obj-$(CONFIG_CPU_TLB_V4WB) += tlb-v4wb.o obj-$(CONFIG_CPU_TLB_V4WBI) += tlb-v4wbi.o obj-$(CONFIG_CPU_TLB_V6) += tlb-v6.o +obj-$(CONFIG_CPU_TLB_V7) += tlb-v7.o obj-$(CONFIG_CPU_ARM610) += proc-arm6_7.o obj-$(CONFIG_CPU_ARM710) += proc-arm6_7.o diff --git a/arch/arm/mm/proc-v7.S b/arch/arm/mm/proc-v7.S index dd823dd4a37..718f4782ee8 100644 --- a/arch/arm/mm/proc-v7.S +++ b/arch/arm/mm/proc-v7.S @@ -256,7 +256,7 @@ __v7_proc_info: .long HWCAP_SWP|HWCAP_HALF|HWCAP_THUMB|HWCAP_FAST_MULT|HWCAP_EDSP .long cpu_v7_name .long v7_processor_functions - .long v6wbi_tlb_fns + .long v7wbi_tlb_fns .long v6_user_fns .long v7_cache_fns .size __v7_proc_info, . - __v7_proc_info diff --git a/arch/arm/mm/tlb-v7.S b/arch/arm/mm/tlb-v7.S new file mode 100644 index 00000000000..b56dda8052f --- /dev/null +++ b/arch/arm/mm/tlb-v7.S @@ -0,0 +1,88 @@ +/* + * linux/arch/arm/mm/tlb-v7.S + * + * Copyright (C) 1997-2002 Russell King + * Modified for ARMv7 by Catalin Marinas + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * ARM architecture version 6 TLB handling functions. + * These assume a split I/D TLB. + */ +#include <linux/linkage.h> +#include <asm/asm-offsets.h> +#include <asm/page.h> +#include <asm/tlbflush.h> +#include "proc-macros.S" + +/* + * v7wbi_flush_user_tlb_range(start, end, vma) + * + * Invalidate a range of TLB entries in the specified address space. + * + * - start - start address (may not be aligned) + * - end - end address (exclusive, may not be aligned) + * - vma - vma_struct describing address range + * + * It is assumed that: + * - the "Invalidate single entry" instruction will invalidate + * both the I and the D TLBs on Harvard-style TLBs + */ +ENTRY(v7wbi_flush_user_tlb_range) + vma_vm_mm r3, r2 @ get vma->vm_mm + mmid r3, r3 @ get vm_mm->context.id + dsb + mov r0, r0, lsr #PAGE_SHIFT @ align address + mov r1, r1, lsr #PAGE_SHIFT + asid r3, r3 @ mask ASID + orr r0, r3, r0, lsl #PAGE_SHIFT @ Create initial MVA + mov r1, r1, lsl #PAGE_SHIFT + vma_vm_flags r2, r2 @ get vma->vm_flags +1: + mcr p15, 0, r0, c8, c6, 1 @ TLB invalidate D MVA (was 1) + tst r2, #VM_EXEC @ Executable area ? + mcrne p15, 0, r0, c8, c5, 1 @ TLB invalidate I MVA (was 1) + add r0, r0, #PAGE_SZ + cmp r0, r1 + blo 1b + mov ip, #0 + mcr p15, 0, ip, c7, c5, 6 @ flush BTAC/BTB + dsb + mov pc, lr + +/* + * v7wbi_flush_kern_tlb_range(start,end) + * + * Invalidate a range of kernel TLB entries + * + * - start - start address (may not be aligned) + * - end - end address (exclusive, may not be aligned) + */ +ENTRY(v7wbi_flush_kern_tlb_range) + dsb + mov r0, r0, lsr #PAGE_SHIFT @ align address + mov r1, r1, lsr #PAGE_SHIFT + mov r0, r0, lsl #PAGE_SHIFT + mov r1, r1, lsl #PAGE_SHIFT +1: + mcr p15, 0, r0, c8, c6, 1 @ TLB invalidate D MVA + mcr p15, 0, r0, c8, c5, 1 @ TLB invalidate I MVA + add r0, r0, #PAGE_SZ + cmp r0, r1 + blo 1b + mov r2, #0 + mcr p15, 0, r2, c7, c5, 6 @ flush BTAC/BTB + dsb + isb + mov pc, lr + + .section ".text.init", #alloc, #execinstr + + .type v7wbi_tlb_fns, #object +ENTRY(v7wbi_tlb_fns) + .long v7wbi_flush_user_tlb_range + .long v7wbi_flush_kern_tlb_range + .long v6wbi_tlb_flags + .size v7wbi_tlb_fns, . - v7wbi_tlb_fns diff --git a/arch/arm/nwfpe/softfloat.h b/arch/arm/nwfpe/softfloat.h index 0a3067452cd..260fe29d73f 100644 --- a/arch/arm/nwfpe/softfloat.h +++ b/arch/arm/nwfpe/softfloat.h @@ -273,4 +273,7 @@ static inline flag float64_lt_nocheck(float64 a, float64 b) extern flag float32_is_nan( float32 a ); extern flag float64_is_nan( float64 a ); +extern int32 float64_to_uint32( struct roundingData *roundData, float64 a ); +extern int32 float64_to_uint32_round_to_zero( float64 a ); + #endif diff --git a/arch/arm/oprofile/op_model_mpcore.c b/arch/arm/oprofile/op_model_mpcore.c index 89850071824..7791da791f5 100644 --- a/arch/arm/oprofile/op_model_mpcore.c +++ b/arch/arm/oprofile/op_model_mpcore.c @@ -257,8 +257,13 @@ static void em_stop(void) */ static void em_route_irq(int irq, unsigned int cpu) { - irq_desc[irq].affinity = cpumask_of_cpu(cpu); - irq_desc[irq].chip->set_affinity(irq, cpumask_of_cpu(cpu)); + struct irq_desc *desc = irq_desc + irq; + cpumask_t mask = cpumask_of_cpu(cpu); + + spin_lock_irq(&desc->lock); + desc->affinity = mask; + desc->chip->set_affinity(irq, mask); + spin_unlock_irq(&desc->lock); } static int em_setup(void) diff --git a/arch/i386/kernel/smpboot.c b/arch/i386/kernel/smpboot.c index 08f07a74a9d..88baed1e7e8 100644 --- a/arch/i386/kernel/smpboot.c +++ b/arch/i386/kernel/smpboot.c @@ -943,10 +943,9 @@ exit: static void smp_tune_scheduling(void) { - unsigned long cachesize; /* kB */ - if (cpu_khz) { - cachesize = boot_cpu_data.x86_cache_size; + /* cache size in kB */ + long cachesize = boot_cpu_data.x86_cache_size; if (cachesize > 0) max_cache_size = cachesize * 1024; diff --git a/arch/m68k/Kconfig b/arch/m68k/Kconfig index b8536c7c087..85cdd23b044 100644 --- a/arch/m68k/Kconfig +++ b/arch/m68k/Kconfig @@ -355,8 +355,9 @@ config RMW_INSNS adventurous. config SINGLE_MEMORY_CHUNK - bool "Use one physical chunk of memory only" - depends on ADVANCED && !SUN3 + bool "Use one physical chunk of memory only" if ADVANCED && !SUN3 + default y if SUN3 + select NEED_MULTIPLE_NODES help Ignore all but the first contiguous chunk of physical memory for VM purposes. This will save a few bytes kernel size and may speed up @@ -377,6 +378,14 @@ config 060_WRITETHROUGH is hardwired on. The 53c710 SCSI driver is known to suffer from this problem. +config ARCH_DISCONTIGMEM_ENABLE + def_bool !SINGLE_MEMORY_CHUNK + +config NODES_SHIFT + int + default "3" + depends on !SINGLE_MEMORY_CHUNK + source "mm/Kconfig" endmenu diff --git a/arch/m68k/Makefile b/arch/m68k/Makefile index c20831a7e1a..aa383a5ea7a 100644 --- a/arch/m68k/Makefile +++ b/arch/m68k/Makefile @@ -19,6 +19,7 @@ COMPILE_ARCH = $(shell uname -m) # override top level makefile AS += -m68020 LDFLAGS := -m m68kelf +LDFLAGS_MODULE += -T $(srctree)/arch/m68k/kernel/module.lds ifneq ($(COMPILE_ARCH),$(ARCH)) # prefix for cross-compiling binaries CROSS_COMPILE = m68k-linux-gnu- diff --git a/arch/m68k/kernel/Makefile b/arch/m68k/kernel/Makefile index 0b68ab8d63d..a806208c7fb 100644 --- a/arch/m68k/kernel/Makefile +++ b/arch/m68k/kernel/Makefile @@ -9,13 +9,12 @@ else endif extra-y += vmlinux.lds -obj-y := entry.o process.o traps.o ints.o signal.o ptrace.o \ +obj-y := entry.o process.o traps.o ints.o signal.o ptrace.o module.o \ sys_m68k.o time.o semaphore.o setup.o m68k_ksyms.o devres.o devres-y = ../../../kernel/irq/devres.o obj-$(CONFIG_PCI) += bios32.o -obj-$(CONFIG_MODULES) += module.o obj-y$(CONFIG_MMU_SUN3) += dma.o # no, it's not a typo EXTRA_AFLAGS := -traditional diff --git a/arch/m68k/kernel/module.c b/arch/m68k/kernel/module.c index 3b1a2ff61dd..774862bc697 100644 --- a/arch/m68k/kernel/module.c +++ b/arch/m68k/kernel/module.c @@ -1,3 +1,9 @@ +/* + * This file is subject to the terms and conditions of the GNU General Public + * License. See the file COPYING in the main directory of this archive + * for more details. + */ + #include <linux/moduleloader.h> #include <linux/elf.h> #include <linux/vmalloc.h> @@ -11,6 +17,8 @@ #define DEBUGP(fmt...) #endif +#ifdef CONFIG_MODULES + void *module_alloc(unsigned long size) { if (size == 0) @@ -118,11 +126,32 @@ int apply_relocate_add(Elf32_Shdr *sechdrs, int module_finalize(const Elf_Ehdr *hdr, const Elf_Shdr *sechdrs, - struct module *me) + struct module *mod) { + module_fixup(mod, mod->arch.fixup_start, mod->arch.fixup_end); + return 0; } void module_arch_cleanup(struct module *mod) { } + +#endif /* CONFIG_MODULES */ + +void module_fixup(struct module *mod, struct m68k_fixup_info *start, + struct m68k_fixup_info *end) +{ + struct m68k_fixup_info *fixup; + + for (fixup = start; fixup < end; fixup++) { + switch (fixup->type) { + case m68k_fixup_memoffset: + *(u32 *)fixup->addr = m68k_memoffset; + break; + case m68k_fixup_vnode_shift: + *(u16 *)fixup->addr += m68k_virt_to_node_shift; + break; + } + } +} diff --git a/arch/m68k/kernel/module.lds b/arch/m68k/kernel/module.lds new file mode 100644 index 00000000000..fda94fa3824 --- /dev/null +++ b/arch/m68k/kernel/module.lds @@ -0,0 +1,7 @@ +SECTIONS { + .m68k_fixup : { + __start_fixup = .; + *(.m68k_fixup) + __stop_fixup = .; + } +} diff --git a/arch/m68k/kernel/setup.c b/arch/m68k/kernel/setup.c index 61031935669..215c7bd4392 100644 --- a/arch/m68k/kernel/setup.c +++ b/arch/m68k/kernel/setup.c @@ -60,14 +60,12 @@ extern unsigned long availmem; int m68k_num_memory; int m68k_realnum_memory; EXPORT_SYMBOL(m68k_realnum_memory); -#ifdef CONFIG_SINGLE_MEMORY_CHUNK unsigned long m68k_memoffset; EXPORT_SYMBOL(m68k_memoffset); -#endif struct mem_info m68k_memory[NUM_MEMINFO]; EXPORT_SYMBOL(m68k_memory); -static struct mem_info m68k_ramdisk; +struct mem_info m68k_ramdisk; static char m68k_command_line[CL_SIZE]; @@ -208,9 +206,6 @@ static void __init m68k_parse_bootinfo(const struct bi_record *record) void __init setup_arch(char **cmdline_p) { extern int _etext, _edata, _end; -#ifndef CONFIG_SUN3 - unsigned long endmem, startmem; -#endif int i; /* The bootinfo is located right after the kernel bss */ @@ -320,30 +315,16 @@ void __init setup_arch(char **cmdline_p) panic("No configuration setup"); } -#ifndef CONFIG_SUN3 - startmem= m68k_memory[0].addr; - endmem = startmem + m68k_memory[0].size; - high_memory = (void *)PAGE_OFFSET; - for (i = 0; i < m68k_num_memory; i++) { - m68k_memory[i].size &= MASK_256K; - if (m68k_memory[i].addr < startmem) - startmem = m68k_memory[i].addr; - if (m68k_memory[i].addr+m68k_memory[i].size > endmem) - endmem = m68k_memory[i].addr+m68k_memory[i].size; - high_memory += m68k_memory[i].size; - } - - availmem += init_bootmem_node(NODE_DATA(0), availmem >> PAGE_SHIFT, - startmem >> PAGE_SHIFT, endmem >> PAGE_SHIFT); - - for (i = 0; i < m68k_num_memory; i++) - free_bootmem(m68k_memory[i].addr, m68k_memory[i].size); - - reserve_bootmem(m68k_memory[0].addr, availmem - m68k_memory[0].addr); + paging_init(); +#ifndef CONFIG_SUN3 + for (i = 1; i < m68k_num_memory; i++) + free_bootmem_node(NODE_DATA(i), m68k_memory[i].addr, + m68k_memory[i].size); #ifdef CONFIG_BLK_DEV_INITRD if (m68k_ramdisk.size) { - reserve_bootmem(m68k_ramdisk.addr, m68k_ramdisk.size); + reserve_bootmem_node(__virt_to_node(phys_to_virt(m68k_ramdisk.addr)), + m68k_ramdisk.addr, m68k_ramdisk.size); initrd_start = (unsigned long)phys_to_virt(m68k_ramdisk.addr); initrd_end = initrd_start + m68k_ramdisk.size; printk("initrd: %08lx - %08lx\n", initrd_start, initrd_end); @@ -362,8 +343,6 @@ void __init setup_arch(char **cmdline_p) #endif /* !CONFIG_SUN3 */ - paging_init(); - /* set ISA defs early as possible */ #if defined(CONFIG_ISA) && defined(MULTI_ISA) #if defined(CONFIG_Q40) diff --git a/arch/m68k/kernel/vmlinux-std.lds b/arch/m68k/kernel/vmlinux-std.lds index 78f139226a1..40f02b128f2 100644 --- a/arch/m68k/kernel/vmlinux-std.lds +++ b/arch/m68k/kernel/vmlinux-std.lds @@ -60,6 +60,11 @@ SECTIONS __con_initcall_start = .; .con_initcall.init : { *(.con_initcall.init) } __con_initcall_end = .; + .m68k_fixup : { + __start_fixup = .; + *(.m68k_fixup) + __stop_fixup = .; + } SECURITY_INIT #ifdef CONFIG_BLK_DEV_INITRD . = ALIGN(8192); diff --git a/arch/m68k/kernel/vmlinux-sun3.lds b/arch/m68k/kernel/vmlinux-sun3.lds index c8999b2db23..f06425b6d20 100644 --- a/arch/m68k/kernel/vmlinux-sun3.lds +++ b/arch/m68k/kernel/vmlinux-sun3.lds @@ -54,6 +54,11 @@ __init_begin = .; __con_initcall_start = .; .con_initcall.init : { *(.con_initcall.init) } __con_initcall_end = .; + .m68k_fixup : { + __start_fixup = .; + *(.m68k_fixup) + __stop_fixup = .; + } SECURITY_INIT #ifdef CONFIG_BLK_DEV_INITRD . = ALIGN(8192); diff --git a/arch/m68k/mm/init.c b/arch/m68k/mm/init.c index ab90213e5c5..f1de19e1dde 100644 --- a/arch/m68k/mm/init.c +++ b/arch/m68k/mm/init.c @@ -7,6 +7,7 @@ * to motorola.c and sun3mmu.c */ +#include <linux/module.h> #include <linux/signal.h> #include <linux/sched.h> #include <linux/mm.h> @@ -31,6 +32,37 @@ DEFINE_PER_CPU(struct mmu_gather, mmu_gathers); +static bootmem_data_t __initdata bootmem_data[MAX_NUMNODES]; + +pg_data_t pg_data_map[MAX_NUMNODES]; +EXPORT_SYMBOL(pg_data_map); + +int m68k_virt_to_node_shift; + +#ifndef CONFIG_SINGLE_MEMORY_CHUNK +pg_data_t *pg_data_table[65]; +EXPORT_SYMBOL(pg_data_table); +#endif + +void m68k_setup_node(int node) +{ +#ifndef CONFIG_SINGLE_MEMORY_CHUNK + struct mem_info *info = m68k_memory + node; + int i, end; + + i = (unsigned long)phys_to_virt(info->addr) >> __virt_to_node_shift(); + end = (unsigned long)phys_to_virt(info->addr + info->size - 1) >> __virt_to_node_shift(); + for (; i <= end; i++) { + if (pg_data_table[i]) + printk("overlap at %u for chunk %u\n", i, node); + pg_data_table[i] = pg_data_map + node; + } +#endif + pg_data_map[node].bdata = bootmem_data + node; + node_set_online(node); +} + + /* * ZERO_PAGE is a special page that is used for zero-initialized * data and COW. @@ -40,52 +72,51 @@ void *empty_zero_page; void show_mem(void) { - unsigned long i; - int free = 0, total = 0, reserved = 0, shared = 0; - int cached = 0; - - printk("\nMem-info:\n"); - show_free_areas(); - printk("Free swap: %6ldkB\n", nr_swap_pages<<(PAGE_SHIFT-10)); - i = max_mapnr; - while (i-- > 0) { - total++; - if (PageReserved(mem_map+i)) - reserved++; - else if (PageSwapCache(mem_map+i)) - cached++; - else if (!page_count(mem_map+i)) - free++; - else - shared += page_count(mem_map+i) - 1; - } - printk("%d pages of RAM\n",total); - printk("%d free pages\n",free); - printk("%d reserved pages\n",reserved); - printk("%d pages shared\n",shared); - printk("%d pages swap cached\n",cached); + pg_data_t *pgdat; + int free = 0, total = 0, reserved = 0, shared = 0; + int cached = 0; + int i; + + printk("\nMem-info:\n"); + show_free_areas(); + printk("Free swap: %6ldkB\n", nr_swap_pages<<(PAGE_SHIFT-10)); + for_each_online_pgdat(pgdat) { + for (i = 0; i < pgdat->node_spanned_pages; i++) { + struct page *page = pgdat->node_mem_map + i; + total++; + if (PageReserved(page)) + reserved++; + else if (PageSwapCache(page)) + cached++; + else if (!page_count(page)) + free++; + else + shared += page_count(page) - 1; + } + } + printk("%d pages of RAM\n",total); + printk("%d free pages\n",free); + printk("%d reserved pages\n",reserved); + printk("%d pages shared\n",shared); + printk("%d pages swap cached\n",cached); } extern void init_pointer_table(unsigned long ptable); /* References to section boundaries */ -extern char _text, _etext, _edata, __bss_start, _end; -extern char __init_begin, __init_end; +extern char _text[], _etext[]; +extern char __init_begin[], __init_end[]; extern pmd_t *zero_pgtable; void __init mem_init(void) { + pg_data_t *pgdat; int codepages = 0; int datapages = 0; int initpages = 0; - unsigned long tmp; -#ifndef CONFIG_SUN3 int i; -#endif - - max_mapnr = num_physpages = (((unsigned long)high_memory - PAGE_OFFSET) >> PAGE_SHIFT); #ifdef CONFIG_ATARI if (MACH_IS_ATARI) @@ -93,19 +124,25 @@ void __init mem_init(void) #endif /* this will put all memory onto the freelists */ - totalram_pages = free_all_bootmem(); - - for (tmp = PAGE_OFFSET ; tmp < (unsigned long)high_memory; tmp += PAGE_SIZE) { - if (PageReserved(virt_to_page(tmp))) { - if (tmp >= (unsigned long)&_text - && tmp < (unsigned long)&_etext) + totalram_pages = num_physpages = 0; + for_each_online_pgdat(pgdat) { + num_physpages += pgdat->node_present_pages; + + totalram_pages += free_all_bootmem_node(pgdat); + for (i = 0; i < pgdat->node_spanned_pages; i++) { + struct page *page = pgdat->node_mem_map + i; + char *addr = page_to_virt(page); + + if (!PageReserved(page)) + continue; + if (addr >= _text && + addr < _etext) codepages++; - else if (tmp >= (unsigned long) &__init_begin - && tmp < (unsigned long) &__init_end) + else if (addr >= __init_begin && + addr < __init_end) initpages++; else datapages++; - continue; } } @@ -124,7 +161,7 @@ void __init mem_init(void) printk("Memory: %luk/%luk available (%dk kernel code, %dk data, %dk init)\n", (unsigned long)nr_free_pages() << (PAGE_SHIFT-10), - max_mapnr << (PAGE_SHIFT-10), + totalram_pages << (PAGE_SHIFT-10), codepages << (PAGE_SHIFT-10), datapages << (PAGE_SHIFT-10), initpages << (PAGE_SHIFT-10)); diff --git a/arch/m68k/mm/memory.c b/arch/m68k/mm/memory.c index 13c0b4ad01e..b7473525b43 100644 --- a/arch/m68k/mm/memory.c +++ b/arch/m68k/mm/memory.c @@ -127,67 +127,6 @@ int free_pointer_table (pmd_t *ptable) return 0; } -#ifdef DEBUG_INVALID_PTOV -int mm_inv_cnt = 5; -#endif - -#ifndef CONFIG_SINGLE_MEMORY_CHUNK -/* - * The following two routines map from a physical address to a kernel - * virtual address and vice versa. - */ -unsigned long mm_vtop(unsigned long vaddr) -{ - int i=0; - unsigned long voff = (unsigned long)vaddr - PAGE_OFFSET; - - do { - if (voff < m68k_memory[i].size) { -#ifdef DEBUGPV - printk ("VTOP(%p)=%lx\n", vaddr, - m68k_memory[i].addr + voff); -#endif - return m68k_memory[i].addr + voff; - } - voff -= m68k_memory[i].size; - } while (++i < m68k_num_memory); - - /* As a special case allow `__pa(high_memory)'. */ - if (voff == 0) - return m68k_memory[i-1].addr + m68k_memory[i-1].size; - - return -1; -} -EXPORT_SYMBOL(mm_vtop); - -unsigned long mm_ptov (unsigned long paddr) -{ - int i = 0; - unsigned long poff, voff = PAGE_OFFSET; - - do { - poff = paddr - m68k_memory[i].addr; - if (poff < m68k_memory[i].size) { -#ifdef DEBUGPV - printk ("PTOV(%lx)=%lx\n", paddr, poff + voff); -#endif - return poff + voff; - } - voff += m68k_memory[i].size; - } while (++i < m68k_num_memory); - -#ifdef DEBUG_INVALID_PTOV - if (mm_inv_cnt > 0) { - mm_inv_cnt--; - printk("Invalid use of phys_to_virt(0x%lx) at 0x%p!\n", - paddr, __builtin_return_address(0)); - } -#endif - return -1; -} -EXPORT_SYMBOL(mm_ptov); -#endif - /* invalidate page in both caches */ static inline void clear040(unsigned long paddr) { @@ -354,15 +293,3 @@ void cache_push (unsigned long paddr, int len) } EXPORT_SYMBOL(cache_push); -#ifndef CONFIG_SINGLE_MEMORY_CHUNK -int mm_end_of_chunk (unsigned long addr, int len) -{ - int i; - - for (i = 0; i < m68k_num_memory; i++) - if (m68k_memory[i].addr + m68k_memory[i].size == addr + len) - return 1; - return 0; -} -EXPORT_SYMBOL(mm_end_of_chunk); -#endif diff --git a/arch/m68k/mm/motorola.c b/arch/m68k/mm/motorola.c index afcccdc6ad4..7d571a2b44d 100644 --- a/arch/m68k/mm/motorola.c +++ b/arch/m68k/mm/motorola.c @@ -43,6 +43,11 @@ unsigned long mm_cachebits; EXPORT_SYMBOL(mm_cachebits); #endif +/* size of memory already mapped in head.S */ +#define INIT_MAPPED_SIZE (4UL<<20) + +extern unsigned long availmem; + static pte_t * __init kernel_page_table(void) { pte_t *ptablep; @@ -98,19 +103,20 @@ static pmd_t * __init kernel_ptr_table(void) return last_pgtable; } -static unsigned long __init -map_chunk (unsigned long addr, long size) +static void __init map_node(int node) { #define PTRTREESIZE (256*1024) #define ROOTTREESIZE (32*1024*1024) - static unsigned long virtaddr = PAGE_OFFSET; - unsigned long physaddr; + unsigned long physaddr, virtaddr, size; pgd_t *pgd_dir; pmd_t *pmd_dir; pte_t *pte_dir; - physaddr = (addr | m68k_supervisor_cachemode | - _PAGE_PRESENT | _PAGE_ACCESSED | _PAGE_DIRTY); + size = m68k_memory[node].size; + physaddr = m68k_memory[node].addr; + virtaddr = (unsigned long)phys_to_virt(physaddr); + physaddr |= m68k_supervisor_cachemode | + _PAGE_PRESENT | _PAGE_ACCESSED | _PAGE_DIRTY; if (CPU_IS_040_OR_060) physaddr |= _PAGE_GLOBAL040; @@ -190,8 +196,6 @@ map_chunk (unsigned long addr, long size) #ifdef DEBUG printk("\n"); #endif - - return virtaddr; } /* @@ -200,15 +204,16 @@ map_chunk (unsigned long addr, long size) */ void __init paging_init(void) { - int chunk; - unsigned long mem_avail = 0; unsigned long zones_size[MAX_NR_ZONES] = { 0, }; + unsigned long min_addr, max_addr; + unsigned long addr, size, end; + int i; #ifdef DEBUG { extern unsigned long availmem; - printk ("start of paging_init (%p, %lx, %lx, %lx)\n", - kernel_pg_dir, availmem, start_mem, end_mem); + printk ("start of paging_init (%p, %lx)\n", + kernel_pg_dir, availmem); } #endif @@ -222,24 +227,62 @@ void __init paging_init(void) pgprot_val(protection_map[i]) |= _PAGE_CACHE040; } + min_addr = m68k_memory[0].addr; + max_addr = min_addr + m68k_memory[0].size; + for (i = 1; i < m68k_num_memory;) { + if (m68k_memory[i].addr < min_addr) { + printk("Ignoring memory chunk at 0x%lx:0x%lx before the first chunk\n", + m68k_memory[i].addr, m68k_memory[i].size); + printk("Fix your bootloader or use a memfile to make use of this area!\n"); + m68k_num_memory--; + memmove(m68k_memory + i, m68k_memory + i + 1, + (m68k_num_memory - i) * sizeof(struct mem_info)); + continue; + } + addr = m68k_memory[i].addr + m68k_memory[i].size; + if (addr > max_addr) + max_addr = addr; + i++; + } + m68k_memoffset = min_addr - PAGE_OFFSET; + m68k_virt_to_node_shift = fls(max_addr - min_addr - 1) - 6; + + module_fixup(NULL, __start_fixup, __stop_fixup); + flush_icache(); + + high_memory = phys_to_virt(max_addr); + + min_low_pfn = availmem >> PAGE_SHIFT; + max_low_pfn = max_addr >> PAGE_SHIFT; + + for (i = 0; i < m68k_num_memory; i++) { + addr = m68k_memory[i].addr; + end = addr + m68k_memory[i].size; + m68k_setup_node(i); + availmem = PAGE_ALIGN(availmem); + availmem += init_bootmem_node(NODE_DATA(i), + availmem >> PAGE_SHIFT, + addr >> PAGE_SHIFT, + end >> PAGE_SHIFT); + } + /* * Map the physical memory available into the kernel virtual - * address space. It may allocate some memory for page - * tables and thus modify availmem. + * address space. First initialize the bootmem allocator with + * the memory we already mapped, so map_node() has something + * to allocate. */ + addr = m68k_memory[0].addr; + size = m68k_memory[0].size; + free_bootmem_node(NODE_DATA(0), availmem, min(INIT_MAPPED_SIZE, size) - (availmem - addr)); + map_node(0); + if (size > INIT_MAPPED_SIZE) + free_bootmem_node(NODE_DATA(0), addr + INIT_MAPPED_SIZE, size - INIT_MAPPED_SIZE); - for (chunk = 0; chunk < m68k_num_memory; chunk++) { - mem_avail = map_chunk (m68k_memory[chunk].addr, - m68k_memory[chunk].size); - - } + for (i = 1; i < m68k_num_memory; i++) + map_node(i); flush_tlb_all(); -#ifdef DEBUG - printk ("memory available is %ldKB\n", mem_avail >> 10); - printk ("start_mem is %#lx\nvirtual_end is %#lx\n", - start_mem, end_mem); -#endif /* * initialize the bad page table and bad page to point @@ -256,14 +299,11 @@ void __init paging_init(void) #ifdef DEBUG printk ("before free_area_init\n"); #endif - zones_size[ZONE_DMA] = (mach_max_dma_address < (unsigned long)high_memory ? - (mach_max_dma_address+1) : (unsigned long)high_memory); - zones_size[ZONE_NORMAL] = (unsigned long)high_memory - zones_size[0]; - - zones_size[ZONE_DMA] = (zones_size[ZONE_DMA] - PAGE_OFFSET) >> PAGE_SHIFT; - zones_size[ZONE_NORMAL] >>= PAGE_SHIFT; - - free_area_init(zones_size); + for (i = 0; i < m68k_num_memory; i++) { + zones_size[ZONE_DMA] = m68k_memory[i].size >> PAGE_SHIFT; + free_area_init_node(i, pg_data_map + i, zones_size, + m68k_memory[i].addr >> PAGE_SHIFT, NULL); + } } extern char __init_begin, __init_end; diff --git a/arch/m68k/sun3/config.c b/arch/m68k/sun3/config.c index 4851b8437a8..c0fbd278fbb 100644 --- a/arch/m68k/sun3/config.c +++ b/arch/m68k/sun3/config.c @@ -21,6 +21,7 @@ #include <asm/contregs.h> #include <asm/movs.h> #include <asm/pgtable.h> +#include <asm/pgalloc.h> #include <asm/sun3-head.h> #include <asm/sun3mmu.h> #include <asm/rtc.h> @@ -127,6 +128,7 @@ void __init sun3_bootmem_alloc(unsigned long memory_start, unsigned long memory_ high_memory = (void *)memory_end; availmem = memory_start; + m68k_setup_node(0); availmem += init_bootmem_node(NODE_DATA(0), start_page, 0, num_pages); availmem = (availmem + (PAGE_SIZE-1)) & PAGE_MASK; diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig index 56d3c0dcd2b..5eaeafd30bd 100644 --- a/arch/powerpc/Kconfig +++ b/arch/powerpc/Kconfig @@ -118,6 +118,7 @@ config GENERIC_BUG depends on BUG config SYS_SUPPORTS_APM_EMULATION + default y if PMAC_APM_EMU bool config DEFAULT_UIMAGE diff --git a/arch/powerpc/boot/dts/lite5200.dts b/arch/powerpc/boot/dts/lite5200.dts index eae68ab1177..d29308fe4c2 100644 --- a/arch/powerpc/boot/dts/lite5200.dts +++ b/arch/powerpc/boot/dts/lite5200.dts @@ -67,7 +67,7 @@ interrupt-controller; #interrupt-cells = <3>; device_type = "interrupt-controller"; - compatible = "mpc5200_pic"; + compatible = "mpc5200-pic"; reg = <500 80>; built-in; }; diff --git a/arch/powerpc/boot/dts/lite5200b.dts b/arch/powerpc/boot/dts/lite5200b.dts index 5185625a941..f242531f045 100644 --- a/arch/powerpc/boot/dts/lite5200b.dts +++ b/arch/powerpc/boot/dts/lite5200b.dts @@ -67,7 +67,7 @@ interrupt-controller; #interrupt-cells = <3>; device_type = "interrupt-controller"; - compatible = "mpc5200b-pic\0mpc5200_pic"; + compatible = "mpc5200b-pic\0mpc5200-pic"; reg = <500 80>; built-in; }; diff --git a/arch/powerpc/mm/pgtable_32.c b/arch/powerpc/mm/pgtable_32.c index d8232b7a08f..f6ae1a57d65 100644 --- a/arch/powerpc/mm/pgtable_32.c +++ b/arch/powerpc/mm/pgtable_32.c @@ -93,7 +93,7 @@ void pgd_free(pgd_t *pgd) free_pages((unsigned long)pgd, PGDIR_ORDER); } -pte_t *pte_alloc_one_kernel(struct mm_struct *mm, unsigned long address) +__init_refok pte_t *pte_alloc_one_kernel(struct mm_struct *mm, unsigned long address) { pte_t *pte; extern int mem_init_done; diff --git a/arch/powerpc/platforms/chrp/pegasos_eth.c b/arch/powerpc/platforms/chrp/pegasos_eth.c index 71045677559..5bcc58d9a4d 100644 --- a/arch/powerpc/platforms/chrp/pegasos_eth.c +++ b/arch/powerpc/platforms/chrp/pegasos_eth.c @@ -169,7 +169,7 @@ static int Enable_SRAM(void) /***********/ /***********/ -int mv643xx_eth_add_pds(void) +static int __init mv643xx_eth_add_pds(void) { int ret = 0; static struct pci_device_id pci_marvell_mv64360[] = { diff --git a/arch/powerpc/platforms/powermac/setup.c b/arch/powerpc/platforms/powermac/setup.c index a410bc76a8a..07b1c4ec428 100644 --- a/arch/powerpc/platforms/powermac/setup.c +++ b/arch/powerpc/platforms/powermac/setup.c @@ -384,7 +384,7 @@ int boot_part; static dev_t boot_dev; #ifdef CONFIG_SCSI -void __init note_scsi_host(struct device_node *node, void *host) +void note_scsi_host(struct device_node *node, void *host) { int l; char *p; diff --git a/arch/powerpc/sysdev/qe_lib/Kconfig b/arch/powerpc/sysdev/qe_lib/Kconfig index 887739f3bad..f611d344a12 100644 --- a/arch/powerpc/sysdev/qe_lib/Kconfig +++ b/arch/powerpc/sysdev/qe_lib/Kconfig @@ -5,15 +5,13 @@ config UCC_SLOW bool default n - select UCC help This option provides qe_lib support to UCC slow protocols: UART, BISYNC, QMC config UCC_FAST bool - default n - select UCC + default y if UCC_GETH help This option provides qe_lib support to UCC fast protocols: HDLC, Ethernet, ATM, transparent diff --git a/arch/ppc/kernel/entry.S b/arch/ppc/kernel/entry.S index ab64256110b..fba7ca17a67 100644 --- a/arch/ppc/kernel/entry.S +++ b/arch/ppc/kernel/entry.S @@ -596,7 +596,11 @@ fast_exception_return: mr r12,r4 /* restart at exc_exit_restart */ b 2b - .comm fee_restarts,4 + .section .bss + .align 2 +fee_restarts: + .space 4 + .previous /* aargh, a nonrecoverable interrupt, panic */ /* aargh, we don't know which trap this is */ @@ -851,7 +855,11 @@ load_dbcr0: mtspr SPRN_DBSR,r11 /* clear all pending debug events */ blr - .comm global_dbcr0,8 + .section .bss + .align 4 +global_dbcr0: + .space 8 + .previous #endif /* !(CONFIG_4xx || CONFIG_BOOKE) */ do_work: /* r10 contains MSR_KERNEL here */ @@ -926,4 +934,8 @@ END_FTR_SECTION_IFSET(CPU_FTR_601) /* shouldn't return */ b 4b - .comm ee_restarts,4 + .section .bss + .align 2 +ee_restarts: + .space 4 + .previous diff --git a/arch/ppc/kernel/ppc_ksyms.c b/arch/ppc/kernel/ppc_ksyms.c index 4ad499605d0..a4165209ac7 100644 --- a/arch/ppc/kernel/ppc_ksyms.c +++ b/arch/ppc/kernel/ppc_ksyms.c @@ -40,7 +40,6 @@ #include <asm/time.h> #include <asm/cputable.h> #include <asm/btext.h> -#include <asm/div64.h> #include <asm/xmon.h> #include <asm/signal.h> #include <asm/dcr.h> @@ -93,7 +92,6 @@ EXPORT_SYMBOL(strncpy); EXPORT_SYMBOL(strcat); EXPORT_SYMBOL(strlen); EXPORT_SYMBOL(strcmp); -EXPORT_SYMBOL(__div64_32); EXPORT_SYMBOL(csum_partial); EXPORT_SYMBOL(csum_partial_copy_generic); diff --git a/arch/ppc/mm/hashtable.S b/arch/ppc/mm/hashtable.S index e756942e65c..5f364dc5015 100644 --- a/arch/ppc/mm/hashtable.S +++ b/arch/ppc/mm/hashtable.S @@ -30,7 +30,11 @@ #include <asm/asm-offsets.h> #ifdef CONFIG_SMP - .comm mmu_hash_lock,4 + .section .bss + .align 2 + .globl mmu_hash_lock +mmu_hash_lock: + .space 4 #endif /* CONFIG_SMP */ /* @@ -461,9 +465,17 @@ found_slot: sync /* make sure pte updates get to memory */ blr - .comm next_slot,4 - .comm primary_pteg_full,4 - .comm htab_hash_searches,4 + .section .bss + .align 2 +next_slot: + .space 4 + .globl primary_pteg_full +primary_pteg_full: + .space 4 + .globl htab_hash_searches +htab_hash_searches: + .space 4 + .previous /* * Flush the entry for a particular page from the hash table. diff --git a/arch/ppc/mm/pgtable.c b/arch/ppc/mm/pgtable.c index c023b729880..35ebb6395ae 100644 --- a/arch/ppc/mm/pgtable.c +++ b/arch/ppc/mm/pgtable.c @@ -92,7 +92,7 @@ void pgd_free(pgd_t *pgd) free_pages((unsigned long)pgd, PGDIR_ORDER); } -pte_t *pte_alloc_one_kernel(struct mm_struct *mm, unsigned long address) +__init_refok pte_t *pte_alloc_one_kernel(struct mm_struct *mm, unsigned long address) { pte_t *pte; extern int mem_init_done; diff --git a/arch/sparc/Kconfig b/arch/sparc/Kconfig index bd992c0048f..fbcc00c6c06 100644 --- a/arch/sparc/Kconfig +++ b/arch/sparc/Kconfig @@ -178,6 +178,13 @@ config ARCH_HAS_ILOG2_U64 bool default n +config EMULATED_CMPXCHG + bool + default y + help + Sparc32 does not have a CAS instruction like sparc64. cmpxchg() + is emulated, and therefore it is not completely atomic. + config SUN_PM bool default y diff --git a/arch/sparc/kernel/time.c b/arch/sparc/kernel/time.c index f1401b57ccc..7b4612da74a 100644 --- a/arch/sparc/kernel/time.c +++ b/arch/sparc/kernel/time.c @@ -148,7 +148,7 @@ irqreturn_t timer_interrupt(int irq, void *dev_id) } /* Kick start a stopped clock (procedure from the Sun NVRAM/hostid FAQ). */ -static void __init kick_start_clock(void) +static void __devinit kick_start_clock(void) { struct mostek48t02 *regs = (struct mostek48t02 *)mstk48t02_regs; unsigned char sec; @@ -223,7 +223,7 @@ static __inline__ int has_low_battery(void) return (data1 == data2); /* Was the write blocked? */ } -static void __init mostek_set_system_time(void) +static void __devinit mostek_set_system_time(void) { unsigned int year, mon, day, hour, min, sec; struct mostek48t02 *mregs; diff --git a/arch/sparc/lib/atomic32.c b/arch/sparc/lib/atomic32.c index 559335f4917..617d29832e1 100644 --- a/arch/sparc/lib/atomic32.c +++ b/arch/sparc/lib/atomic32.c @@ -2,6 +2,7 @@ * atomic32.c: 32-bit atomic_t implementation * * Copyright (C) 2004 Keith M Wesolowski + * Copyright (C) 2007 Kyle McMartin * * Based on asm-parisc/atomic.h Copyright (C) 2000 Philipp Rumpf */ @@ -117,3 +118,17 @@ unsigned long ___change_bit(unsigned long *addr, unsigned long mask) return old & mask; } EXPORT_SYMBOL(___change_bit); + +unsigned long __cmpxchg_u32(volatile u32 *ptr, u32 old, u32 new) +{ + unsigned long flags; + u32 prev; + + spin_lock_irqsave(ATOMIC_HASH(addr), flags); + if ((prev = *ptr) == old) + *ptr = new; + spin_unlock_irqrestore(ATOMIC_HASH(addr), flags); + + return (unsigned long)prev; +} +EXPORT_SYMBOL(__cmpxchg_u32); diff --git a/arch/sparc64/Kconfig b/arch/sparc64/Kconfig index 831781cab27..bd00f89eed1 100644 --- a/arch/sparc64/Kconfig +++ b/arch/sparc64/Kconfig @@ -147,10 +147,10 @@ config SMP If you don't know what to do here, say N. config NR_CPUS - int "Maximum number of CPUs (2-64)" - range 2 64 + int "Maximum number of CPUs (2-1024)" + range 2 1024 depends on SMP - default "32" + default "64" source "drivers/cpufreq/Kconfig" diff --git a/arch/sparc64/kernel/Makefile b/arch/sparc64/kernel/Makefile index c749dccacc3..d8d19093d12 100644 --- a/arch/sparc64/kernel/Makefile +++ b/arch/sparc64/kernel/Makefile @@ -8,11 +8,11 @@ EXTRA_CFLAGS := -Werror extra-y := head.o init_task.o vmlinux.lds obj-y := process.o setup.o cpu.o idprom.o \ - traps.o devices.o auxio.o una_asm.o \ + traps.o auxio.o una_asm.o \ irq.o ptrace.o time.o sys_sparc.o signal.o \ unaligned.o central.o pci.o starfire.o semaphore.o \ power.o sbus.o iommu_common.o sparc64_ksyms.o chmc.o \ - visemul.o prom.o of_device.o hvapi.o + visemul.o prom.o of_device.o hvapi.o sstate.o mdesc.o obj-$(CONFIG_STACKTRACE) += stacktrace.o obj-$(CONFIG_PCI) += ebus.o isa.o pci_common.o pci_iommu.o \ diff --git a/arch/sparc64/kernel/devices.c b/arch/sparc64/kernel/devices.c deleted file mode 100644 index 0e03c8e218c..00000000000 --- a/arch/sparc64/kernel/devices.c +++ /dev/null @@ -1,196 +0,0 @@ -/* devices.c: Initial scan of the prom device tree for important - * Sparc device nodes which we need to find. - * - * Copyright (C) 1996 David S. Miller (davem@caip.rutgers.edu) - */ - -#include <linux/kernel.h> -#include <linux/threads.h> -#include <linux/init.h> -#include <linux/ioport.h> -#include <linux/string.h> -#include <linux/spinlock.h> -#include <linux/errno.h> -#include <linux/bootmem.h> - -#include <asm/page.h> -#include <asm/oplib.h> -#include <asm/system.h> -#include <asm/smp.h> -#include <asm/spitfire.h> -#include <asm/timer.h> -#include <asm/cpudata.h> - -/* Used to synchronize accesses to NatSemi SUPER I/O chip configure - * operations in asm/ns87303.h - */ -DEFINE_SPINLOCK(ns87303_lock); - -extern void cpu_probe(void); -extern void central_probe(void); - -static const char *cpu_mid_prop(void) -{ - if (tlb_type == spitfire) - return "upa-portid"; - return "portid"; -} - -static int get_cpu_mid(struct device_node *dp) -{ - struct property *prop; - - if (tlb_type == hypervisor) { - struct linux_prom64_registers *reg; - int len; - - prop = of_find_property(dp, "cpuid", &len); - if (prop && len == 4) - return *(int *) prop->value; - - prop = of_find_property(dp, "reg", NULL); - reg = prop->value; - return (reg[0].phys_addr >> 32) & 0x0fffffffUL; - } else { - const char *prop_name = cpu_mid_prop(); - - prop = of_find_property(dp, prop_name, NULL); - if (prop) - return *(int *) prop->value; - return 0; - } -} - -static int check_cpu_node(struct device_node *dp, int *cur_inst, - int (*compare)(struct device_node *, int, void *), - void *compare_arg, - struct device_node **dev_node, int *mid) -{ - if (!compare(dp, *cur_inst, compare_arg)) { - if (dev_node) - *dev_node = dp; - if (mid) - *mid = get_cpu_mid(dp); - return 0; - } - - (*cur_inst)++; - - return -ENODEV; -} - -static int __cpu_find_by(int (*compare)(struct device_node *, int, void *), - void *compare_arg, - struct device_node **dev_node, int *mid) -{ - struct device_node *dp; - int cur_inst; - - cur_inst = 0; - for_each_node_by_type(dp, "cpu") { - int err = check_cpu_node(dp, &cur_inst, - compare, compare_arg, - dev_node, mid); - if (err == 0) - return 0; - } - - return -ENODEV; -} - -static int cpu_instance_compare(struct device_node *dp, int instance, void *_arg) -{ - int desired_instance = (int) (long) _arg; - - if (instance == desired_instance) - return 0; - return -ENODEV; -} - -int cpu_find_by_instance(int instance, struct device_node **dev_node, int *mid) -{ - return __cpu_find_by(cpu_instance_compare, (void *)(long)instance, - dev_node, mid); -} - -static int cpu_mid_compare(struct device_node *dp, int instance, void *_arg) -{ - int desired_mid = (int) (long) _arg; - int this_mid; - - this_mid = get_cpu_mid(dp); - if (this_mid == desired_mid) - return 0; - return -ENODEV; -} - -int cpu_find_by_mid(int mid, struct device_node **dev_node) -{ - return __cpu_find_by(cpu_mid_compare, (void *)(long)mid, - dev_node, NULL); -} - -void __init device_scan(void) -{ - /* FIX ME FAST... -DaveM */ - ioport_resource.end = 0xffffffffffffffffUL; - - prom_printf("Booting Linux...\n"); - -#ifndef CONFIG_SMP - { - struct device_node *dp; - int err, def; - - err = cpu_find_by_instance(0, &dp, NULL); - if (err) { - prom_printf("No cpu nodes, cannot continue\n"); - prom_halt(); - } - cpu_data(0).clock_tick = - of_getintprop_default(dp, "clock-frequency", 0); - - def = ((tlb_type == hypervisor) ? - (8 * 1024) : - (16 * 1024)); - cpu_data(0).dcache_size = of_getintprop_default(dp, - "dcache-size", - def); - - def = 32; - cpu_data(0).dcache_line_size = - of_getintprop_default(dp, "dcache-line-size", def); - - def = 16 * 1024; - cpu_data(0).icache_size = of_getintprop_default(dp, - "icache-size", - def); - - def = 32; - cpu_data(0).icache_line_size = - of_getintprop_default(dp, "icache-line-size", def); - - def = ((tlb_type == hypervisor) ? - (3 * 1024 * 1024) : - (4 * 1024 * 1024)); - cpu_data(0).ecache_size = of_getintprop_default(dp, - "ecache-size", - def); - - def = 64; - cpu_data(0).ecache_line_size = - of_getintprop_default(dp, "ecache-line-size", def); - printk("CPU[0]: Caches " - "D[sz(%d):line_sz(%d)] " - "I[sz(%d):line_sz(%d)] " - "E[sz(%d):line_sz(%d)]\n", - cpu_data(0).dcache_size, cpu_data(0).dcache_line_size, - cpu_data(0).icache_size, cpu_data(0).icache_line_size, - cpu_data(0).ecache_size, cpu_data(0).ecache_line_size); - } -#endif - - central_probe(); - - cpu_probe(); -} diff --git a/arch/sparc64/kernel/entry.S b/arch/sparc64/kernel/entry.S index 732b77cb71f..8f10dda0f5c 100644 --- a/arch/sparc64/kernel/entry.S +++ b/arch/sparc64/kernel/entry.S @@ -1725,96 +1725,142 @@ real_hard_smp_processor_id: * returns %o0: sysino */ .globl sun4v_devino_to_sysino + .type sun4v_devino_to_sysino,#function sun4v_devino_to_sysino: mov HV_FAST_INTR_DEVINO2SYSINO, %o5 ta HV_FAST_TRAP retl mov %o1, %o0 + .size sun4v_devino_to_sysino, .-sun4v_devino_to_sysino /* %o0: sysino * * returns %o0: intr_enabled (HV_INTR_{DISABLED,ENABLED}) */ .globl sun4v_intr_getenabled + .type sun4v_intr_getenabled,#function sun4v_intr_getenabled: mov HV_FAST_INTR_GETENABLED, %o5 ta HV_FAST_TRAP retl mov %o1, %o0 + .size sun4v_intr_getenabled, .-sun4v_intr_getenabled /* %o0: sysino * %o1: intr_enabled (HV_INTR_{DISABLED,ENABLED}) */ .globl sun4v_intr_setenabled + .type sun4v_intr_setenabled,#function sun4v_intr_setenabled: mov HV_FAST_INTR_SETENABLED, %o5 ta HV_FAST_TRAP retl nop + .size sun4v_intr_setenabled, .-sun4v_intr_setenabled /* %o0: sysino * * returns %o0: intr_state (HV_INTR_STATE_*) */ .globl sun4v_intr_getstate + .type sun4v_intr_getstate,#function sun4v_intr_getstate: mov HV_FAST_INTR_GETSTATE, %o5 ta HV_FAST_TRAP retl mov %o1, %o0 + .size sun4v_intr_getstate, .-sun4v_intr_getstate /* %o0: sysino * %o1: intr_state (HV_INTR_STATE_*) */ .globl sun4v_intr_setstate + .type sun4v_intr_setstate,#function sun4v_intr_setstate: mov HV_FAST_INTR_SETSTATE, %o5 ta HV_FAST_TRAP retl nop + .size sun4v_intr_setstate, .-sun4v_intr_setstate /* %o0: sysino * * returns %o0: cpuid */ .globl sun4v_intr_gettarget + .type sun4v_intr_gettarget,#function sun4v_intr_gettarget: mov HV_FAST_INTR_GETTARGET, %o5 ta HV_FAST_TRAP retl mov %o1, %o0 + .size sun4v_intr_gettarget, .-sun4v_intr_gettarget /* %o0: sysino * %o1: cpuid */ .globl sun4v_intr_settarget + .type sun4v_intr_settarget,#function sun4v_intr_settarget: mov HV_FAST_INTR_SETTARGET, %o5 ta HV_FAST_TRAP retl nop + .size sun4v_intr_settarget, .-sun4v_intr_settarget - /* %o0: type - * %o1: queue paddr - * %o2: num queue entries + /* %o0: cpuid + * %o1: pc + * %o2: rtba + * %o3: arg0 * * returns %o0: status */ - .globl sun4v_cpu_qconf -sun4v_cpu_qconf: - mov HV_FAST_CPU_QCONF, %o5 + .globl sun4v_cpu_start + .type sun4v_cpu_start,#function +sun4v_cpu_start: + mov HV_FAST_CPU_START, %o5 ta HV_FAST_TRAP retl nop + .size sun4v_cpu_start, .-sun4v_cpu_start - /* returns %o0: status + /* %o0: cpuid + * + * returns %o0: status */ + .globl sun4v_cpu_stop + .type sun4v_cpu_stop,#function +sun4v_cpu_stop: + mov HV_FAST_CPU_STOP, %o5 + ta HV_FAST_TRAP + retl + nop + .size sun4v_cpu_stop, .-sun4v_cpu_stop + + /* returns %o0: status */ .globl sun4v_cpu_yield + .type sun4v_cpu_yield, #function sun4v_cpu_yield: mov HV_FAST_CPU_YIELD, %o5 ta HV_FAST_TRAP retl nop + .size sun4v_cpu_yield, .-sun4v_cpu_yield + + /* %o0: type + * %o1: queue paddr + * %o2: num queue entries + * + * returns %o0: status + */ + .globl sun4v_cpu_qconf + .type sun4v_cpu_qconf,#function +sun4v_cpu_qconf: + mov HV_FAST_CPU_QCONF, %o5 + ta HV_FAST_TRAP + retl + nop + .size sun4v_cpu_qconf, .-sun4v_cpu_qconf /* %o0: num cpus in cpu list * %o1: cpu list paddr @@ -1823,11 +1869,13 @@ sun4v_cpu_yield: * returns %o0: status */ .globl sun4v_cpu_mondo_send + .type sun4v_cpu_mondo_send,#function sun4v_cpu_mondo_send: mov HV_FAST_CPU_MONDO_SEND, %o5 ta HV_FAST_TRAP retl nop + .size sun4v_cpu_mondo_send, .-sun4v_cpu_mondo_send /* %o0: CPU ID * @@ -1835,6 +1883,7 @@ sun4v_cpu_mondo_send: * %o0: cpu state as HV_CPU_STATE_* */ .globl sun4v_cpu_state + .type sun4v_cpu_state,#function sun4v_cpu_state: mov HV_FAST_CPU_STATE, %o5 ta HV_FAST_TRAP @@ -1843,6 +1892,37 @@ sun4v_cpu_state: mov %o1, %o0 1: retl nop + .size sun4v_cpu_state, .-sun4v_cpu_state + + /* %o0: virtual address + * %o1: must be zero + * %o2: TTE + * %o3: HV_MMU_* flags + * + * returns %o0: status + */ + .globl sun4v_mmu_map_perm_addr + .type sun4v_mmu_map_perm_addr,#function +sun4v_mmu_map_perm_addr: + mov HV_FAST_MMU_MAP_PERM_ADDR, %o5 + ta HV_FAST_TRAP + retl + nop + .size sun4v_mmu_map_perm_addr, .-sun4v_mmu_map_perm_addr + + /* %o0: number of TSB descriptions + * %o1: TSB descriptions real address + * + * returns %o0: status + */ + .globl sun4v_mmu_tsb_ctx0 + .type sun4v_mmu_tsb_ctx0,#function +sun4v_mmu_tsb_ctx0: + mov HV_FAST_MMU_TSB_CTX0, %o5 + ta HV_FAST_TRAP + retl + nop + .size sun4v_mmu_tsb_ctx0, .-sun4v_mmu_tsb_ctx0 /* %o0: API group number * %o1: pointer to unsigned long major number storage @@ -1851,6 +1931,7 @@ sun4v_cpu_state: * returns %o0: status */ .globl sun4v_get_version + .type sun4v_get_version,#function sun4v_get_version: mov HV_CORE_GET_VER, %o5 mov %o1, %o3 @@ -1859,6 +1940,7 @@ sun4v_get_version: stx %o1, [%o3] retl stx %o2, [%o4] + .size sun4v_get_version, .-sun4v_get_version /* %o0: API group number * %o1: desired major number @@ -1868,18 +1950,49 @@ sun4v_get_version: * returns %o0: status */ .globl sun4v_set_version + .type sun4v_set_version,#function sun4v_set_version: mov HV_CORE_SET_VER, %o5 mov %o3, %o4 ta HV_CORE_TRAP retl stx %o1, [%o4] + .size sun4v_set_version, .-sun4v_set_version + + /* %o0: pointer to unsigned long time + * + * returns %o0: status + */ + .globl sun4v_tod_get + .type sun4v_tod_get,#function +sun4v_tod_get: + mov %o0, %o4 + mov HV_FAST_TOD_GET, %o5 + ta HV_FAST_TRAP + stx %o1, [%o4] + retl + nop + .size sun4v_tod_get, .-sun4v_tod_get + + /* %o0: time + * + * returns %o0: status + */ + .globl sun4v_tod_set + .type sun4v_tod_set,#function +sun4v_tod_set: + mov HV_FAST_TOD_SET, %o5 + ta HV_FAST_TRAP + retl + nop + .size sun4v_tod_set, .-sun4v_tod_set /* %o0: pointer to unsigned long status * * returns %o0: signed character */ .globl sun4v_con_getchar + .type sun4v_con_getchar,#function sun4v_con_getchar: mov %o0, %o4 mov HV_FAST_CONS_GETCHAR, %o5 @@ -1889,17 +2002,20 @@ sun4v_con_getchar: stx %o0, [%o4] retl sra %o1, 0, %o0 + .size sun4v_con_getchar, .-sun4v_con_getchar /* %o0: signed long character * * returns %o0: status */ .globl sun4v_con_putchar + .type sun4v_con_putchar,#function sun4v_con_putchar: mov HV_FAST_CONS_PUTCHAR, %o5 ta HV_FAST_TRAP retl sra %o0, 0, %o0 + .size sun4v_con_putchar, .-sun4v_con_putchar /* %o0: buffer real address * %o1: buffer size @@ -1908,6 +2024,7 @@ sun4v_con_putchar: * returns %o0: status */ .globl sun4v_con_read + .type sun4v_con_read,#function sun4v_con_read: mov %o2, %o4 mov HV_FAST_CONS_READ, %o5 @@ -1922,6 +2039,7 @@ sun4v_con_read: stx %o1, [%o4] 1: retl nop + .size sun4v_con_read, .-sun4v_con_read /* %o0: buffer real address * %o1: buffer size @@ -1930,6 +2048,7 @@ sun4v_con_read: * returns %o0: status */ .globl sun4v_con_write + .type sun4v_con_write,#function sun4v_con_write: mov %o2, %o4 mov HV_FAST_CONS_WRITE, %o5 @@ -1937,3 +2056,445 @@ sun4v_con_write: stx %o1, [%o4] retl nop + .size sun4v_con_write, .-sun4v_con_write + + /* %o0: soft state + * %o1: address of description string + * + * returns %o0: status + */ + .globl sun4v_mach_set_soft_state + .type sun4v_mach_set_soft_state,#function +sun4v_mach_set_soft_state: + mov HV_FAST_MACH_SET_SOFT_STATE, %o5 + ta HV_FAST_TRAP + retl + nop + .size sun4v_mach_set_soft_state, .-sun4v_mach_set_soft_state + + /* %o0: exit code + * + * Does not return. + */ + .globl sun4v_mach_exit + .type sun4v_mach_exit,#function +sun4v_mach_exit: + mov HV_FAST_MACH_EXIT, %o5 + ta HV_FAST_TRAP + retl + nop + .size sun4v_mach_exit, .-sun4v_mach_exit + + /* %o0: buffer real address + * %o1: buffer length + * %o2: pointer to unsigned long real_buf_len + * + * returns %o0: status + */ + .globl sun4v_mach_desc + .type sun4v_mach_desc,#function +sun4v_mach_desc: + mov %o2, %o4 + mov HV_FAST_MACH_DESC, %o5 + ta HV_FAST_TRAP + stx %o1, [%o4] + retl + nop + .size sun4v_mach_desc, .-sun4v_mach_desc + + /* %o0: new timeout in milliseconds + * %o1: pointer to unsigned long orig_timeout + * + * returns %o0: status + */ + .globl sun4v_mach_set_watchdog + .type sun4v_mach_set_watchdog,#function +sun4v_mach_set_watchdog: + mov %o1, %o4 + mov HV_FAST_MACH_SET_WATCHDOG, %o5 + ta HV_FAST_TRAP + stx %o1, [%o4] + retl + nop + .size sun4v_mach_set_watchdog, .-sun4v_mach_set_watchdog + + /* No inputs and does not return. */ + .globl sun4v_mach_sir + .type sun4v_mach_sir,#function +sun4v_mach_sir: + mov %o1, %o4 + mov HV_FAST_MACH_SIR, %o5 + ta HV_FAST_TRAP + stx %o1, [%o4] + retl + nop + .size sun4v_mach_sir, .-sun4v_mach_sir + + /* %o0: channel + * %o1: ra + * %o2: num_entries + * + * returns %o0: status + */ + .globl sun4v_ldc_tx_qconf + .type sun4v_ldc_tx_qconf,#function +sun4v_ldc_tx_qconf: + mov HV_FAST_LDC_TX_QCONF, %o5 + ta HV_FAST_TRAP + retl + nop + .size sun4v_ldc_tx_qconf, .-sun4v_ldc_tx_qconf + + /* %o0: channel + * %o1: pointer to unsigned long ra + * %o2: pointer to unsigned long num_entries + * + * returns %o0: status + */ + .globl sun4v_ldc_tx_qinfo + .type sun4v_ldc_tx_qinfo,#function +sun4v_ldc_tx_qinfo: + mov %o1, %g1 + mov %o2, %g2 + mov HV_FAST_LDC_TX_QINFO, %o5 + ta HV_FAST_TRAP + stx %o1, [%g1] + stx %o2, [%g2] + retl + nop + .size sun4v_ldc_tx_qinfo, .-sun4v_ldc_tx_qinfo + + /* %o0: channel + * %o1: pointer to unsigned long head_off + * %o2: pointer to unsigned long tail_off + * %o2: pointer to unsigned long chan_state + * + * returns %o0: status + */ + .globl sun4v_ldc_tx_get_state + .type sun4v_ldc_tx_get_state,#function +sun4v_ldc_tx_get_state: + mov %o1, %g1 + mov %o2, %g2 + mov %o3, %g3 + mov HV_FAST_LDC_TX_GET_STATE, %o5 + ta HV_FAST_TRAP + stx %o1, [%g1] + stx %o2, [%g2] + stx %o3, [%g3] + retl + nop + .size sun4v_ldc_tx_get_state, .-sun4v_ldc_tx_get_state + + /* %o0: channel + * %o1: tail_off + * + * returns %o0: status + */ + .globl sun4v_ldc_tx_set_qtail + .type sun4v_ldc_tx_set_qtail,#function +sun4v_ldc_tx_set_qtail: + mov HV_FAST_LDC_TX_SET_QTAIL, %o5 + ta HV_FAST_TRAP + retl + nop + .size sun4v_ldc_tx_set_qtail, .-sun4v_ldc_tx_set_qtail + + /* %o0: channel + * %o1: ra + * %o2: num_entries + * + * returns %o0: status + */ + .globl sun4v_ldc_rx_qconf + .type sun4v_ldc_rx_qconf,#function +sun4v_ldc_rx_qconf: + mov HV_FAST_LDC_RX_QCONF, %o5 + ta HV_FAST_TRAP + retl + nop + .size sun4v_ldc_rx_qconf, .-sun4v_ldc_rx_qconf + + /* %o0: channel + * %o1: pointer to unsigned long ra + * %o2: pointer to unsigned long num_entries + * + * returns %o0: status + */ + .globl sun4v_ldc_rx_qinfo + .type sun4v_ldc_rx_qinfo,#function +sun4v_ldc_rx_qinfo: + mov %o1, %g1 + mov %o2, %g2 + mov HV_FAST_LDC_RX_QINFO, %o5 + ta HV_FAST_TRAP + stx %o1, [%g1] + stx %o2, [%g2] + retl + nop + .size sun4v_ldc_rx_qinfo, .-sun4v_ldc_rx_qinfo + + /* %o0: channel + * %o1: pointer to unsigned long head_off + * %o2: pointer to unsigned long tail_off + * %o2: pointer to unsigned long chan_state + * + * returns %o0: status + */ + .globl sun4v_ldc_rx_get_state + .type sun4v_ldc_rx_get_state,#function +sun4v_ldc_rx_get_state: + mov %o1, %g1 + mov %o2, %g2 + mov %o3, %g3 + mov HV_FAST_LDC_RX_GET_STATE, %o5 + ta HV_FAST_TRAP + stx %o1, [%g1] + stx %o2, [%g2] + stx %o3, [%g3] + retl + nop + .size sun4v_ldc_rx_get_state, .-sun4v_ldc_rx_get_state + + /* %o0: channel + * %o1: head_off + * + * returns %o0: status + */ + .globl sun4v_ldc_rx_set_qhead + .type sun4v_ldc_rx_set_qhead,#function +sun4v_ldc_rx_set_qhead: + mov HV_FAST_LDC_RX_SET_QHEAD, %o5 + ta HV_FAST_TRAP + retl + nop + .size sun4v_ldc_rx_set_qhead, .-sun4v_ldc_rx_set_qhead + + /* %o0: channel + * %o1: ra + * %o2: num_entries + * + * returns %o0: status + */ + .globl sun4v_ldc_set_map_table + .type sun4v_ldc_set_map_table,#function +sun4v_ldc_set_map_table: + mov HV_FAST_LDC_SET_MAP_TABLE, %o5 + ta HV_FAST_TRAP + retl + nop + .size sun4v_ldc_set_map_table, .-sun4v_ldc_set_map_table + + /* %o0: channel + * %o1: pointer to unsigned long ra + * %o2: pointer to unsigned long num_entries + * + * returns %o0: status + */ + .globl sun4v_ldc_get_map_table + .type sun4v_ldc_get_map_table,#function +sun4v_ldc_get_map_table: + mov %o1, %g1 + mov %o2, %g2 + mov HV_FAST_LDC_GET_MAP_TABLE, %o5 + ta HV_FAST_TRAP + stx %o1, [%g1] + stx %o2, [%g2] + retl + nop + .size sun4v_ldc_get_map_table, .-sun4v_ldc_get_map_table + + /* %o0: channel + * %o1: dir_code + * %o2: tgt_raddr + * %o3: lcl_raddr + * %o4: len + * %o5: pointer to unsigned long actual_len + * + * returns %o0: status + */ + .globl sun4v_ldc_copy + .type sun4v_ldc_copy,#function +sun4v_ldc_copy: + mov %o5, %g1 + mov HV_FAST_LDC_COPY, %o5 + ta HV_FAST_TRAP + stx %o1, [%g1] + retl + nop + .size sun4v_ldc_copy, .-sun4v_ldc_copy + + /* %o0: channel + * %o1: cookie + * %o2: pointer to unsigned long ra + * %o3: pointer to unsigned long perm + * + * returns %o0: status + */ + .globl sun4v_ldc_mapin + .type sun4v_ldc_mapin,#function +sun4v_ldc_mapin: + mov %o2, %g1 + mov %o3, %g2 + mov HV_FAST_LDC_MAPIN, %o5 + ta HV_FAST_TRAP + stx %o1, [%g1] + stx %o2, [%g2] + retl + nop + .size sun4v_ldc_mapin, .-sun4v_ldc_mapin + + /* %o0: ra + * + * returns %o0: status + */ + .globl sun4v_ldc_unmap + .type sun4v_ldc_unmap,#function +sun4v_ldc_unmap: + mov HV_FAST_LDC_UNMAP, %o5 + ta HV_FAST_TRAP + retl + nop + .size sun4v_ldc_unmap, .-sun4v_ldc_unmap + + /* %o0: cookie + * %o1: mte_cookie + * + * returns %o0: status + */ + .globl sun4v_ldc_revoke + .type sun4v_ldc_revoke,#function +sun4v_ldc_revoke: + mov HV_FAST_LDC_REVOKE, %o5 + ta HV_FAST_TRAP + retl + nop + .size sun4v_ldc_revoke, .-sun4v_ldc_revoke + + /* %o0: device handle + * %o1: device INO + * %o2: pointer to unsigned long cookie + * + * returns %o0: status + */ + .globl sun4v_vintr_get_cookie + .type sun4v_vintr_get_cookie,#function +sun4v_vintr_get_cookie: + mov %o2, %g1 + mov HV_FAST_VINTR_GET_COOKIE, %o5 + ta HV_FAST_TRAP + stx %o1, [%g1] + retl + nop + .size sun4v_vintr_get_cookie, .-sun4v_vintr_get_cookie + + /* %o0: device handle + * %o1: device INO + * %o2: cookie + * + * returns %o0: status + */ + .globl sun4v_vintr_set_cookie + .type sun4v_vintr_set_cookie,#function +sun4v_vintr_set_cookie: + mov HV_FAST_VINTR_SET_COOKIE, %o5 + ta HV_FAST_TRAP + retl + nop + .size sun4v_vintr_set_cookie, .-sun4v_vintr_set_cookie + + /* %o0: device handle + * %o1: device INO + * %o2: pointer to unsigned long valid_state + * + * returns %o0: status + */ + .globl sun4v_vintr_get_valid + .type sun4v_vintr_get_valid,#function +sun4v_vintr_get_valid: + mov %o2, %g1 + mov HV_FAST_VINTR_GET_VALID, %o5 + ta HV_FAST_TRAP + stx %o1, [%g1] + retl + nop + .size sun4v_vintr_get_valid, .-sun4v_vintr_get_valid + + /* %o0: device handle + * %o1: device INO + * %o2: valid_state + * + * returns %o0: status + */ + .globl sun4v_vintr_set_valid + .type sun4v_vintr_set_valid,#function +sun4v_vintr_set_valid: + mov HV_FAST_VINTR_SET_VALID, %o5 + ta HV_FAST_TRAP + retl + nop + .size sun4v_vintr_set_valid, .-sun4v_vintr_set_valid + + /* %o0: device handle + * %o1: device INO + * %o2: pointer to unsigned long state + * + * returns %o0: status + */ + .globl sun4v_vintr_get_state + .type sun4v_vintr_get_state,#function +sun4v_vintr_get_state: + mov %o2, %g1 + mov HV_FAST_VINTR_GET_STATE, %o5 + ta HV_FAST_TRAP + stx %o1, [%g1] + retl + nop + .size sun4v_vintr_get_state, .-sun4v_vintr_get_state + + /* %o0: device handle + * %o1: device INO + * %o2: state + * + * returns %o0: status + */ + .globl sun4v_vintr_set_state + .type sun4v_vintr_set_state,#function +sun4v_vintr_set_state: + mov HV_FAST_VINTR_SET_STATE, %o5 + ta HV_FAST_TRAP + retl + nop + .size sun4v_vintr_set_state, .-sun4v_vintr_set_state + + /* %o0: device handle + * %o1: device INO + * %o2: pointer to unsigned long cpuid + * + * returns %o0: status + */ + .globl sun4v_vintr_get_target + .type sun4v_vintr_get_target,#function +sun4v_vintr_get_target: + mov %o2, %g1 + mov HV_FAST_VINTR_GET_TARGET, %o5 + ta HV_FAST_TRAP + stx %o1, [%g1] + retl + nop + .size sun4v_vintr_get_target, .-sun4v_vintr_get_target + + /* %o0: device handle + * %o1: device INO + * %o2: cpuid + * + * returns %o0: status + */ + .globl sun4v_vintr_set_target + .type sun4v_vintr_set_target,#function +sun4v_vintr_set_target: + mov HV_FAST_VINTR_SET_TARGET, %o5 + ta HV_FAST_TRAP + retl + nop + .size sun4v_vintr_set_target, .-sun4v_vintr_set_target diff --git a/arch/sparc64/kernel/head.S b/arch/sparc64/kernel/head.S index baea10a9819..77259526cb1 100644 --- a/arch/sparc64/kernel/head.S +++ b/arch/sparc64/kernel/head.S @@ -523,7 +523,7 @@ tlb_fixup_done: #else mov 0, %o0 #endif - stb %o0, [%g6 + TI_CPU] + sth %o0, [%g6 + TI_CPU] /* Off we go.... */ call start_kernel @@ -653,33 +653,54 @@ setup_tba: restore sparc64_boot_end: -#include "ktlb.S" -#include "tsb.S" #include "etrap.S" #include "rtrap.S" #include "winfixup.S" #include "entry.S" #include "sun4v_tlb_miss.S" #include "sun4v_ivec.S" +#include "ktlb.S" +#include "tsb.S" /* * The following skip makes sure the trap table in ttable.S is aligned * on a 32K boundary as required by the v9 specs for TBA register. * * We align to a 32K boundary, then we have the 32K kernel TSB, - * then the 32K aligned trap table. + * the 64K kernel 4MB TSB, and then the 32K aligned trap table. */ 1: .skip 0x4000 + _start - 1b +! 0x0000000000408000 + .globl swapper_tsb swapper_tsb: .skip (32 * 1024) -! 0x0000000000408000 + .globl swapper_4m_tsb +swapper_4m_tsb: + .skip (64 * 1024) + +! 0x0000000000420000 + /* Some care needs to be exercised if you try to move the + * location of the trap table relative to other things. For + * one thing there are br* instructions in some of the + * trap table entires which branch back to code in ktlb.S + * Those instructions can only handle a signed 16-bit + * displacement. + * + * There is a binutils bug (bugzilla #4558) which causes + * the relocation overflow checks for such instructions to + * not be done correctly. So bintuils will not notice the + * error and will instead write junk into the relocation and + * you'll have an unbootable kernel. + */ #include "ttable.S" +! 0x0000000000428000 + #include "systbls.S" .data diff --git a/arch/sparc64/kernel/hvapi.c b/arch/sparc64/kernel/hvapi.c index f03ffc829c7..f34f5d6181e 100644 --- a/arch/sparc64/kernel/hvapi.c +++ b/arch/sparc64/kernel/hvapi.c @@ -9,6 +9,7 @@ #include <asm/hypervisor.h> #include <asm/oplib.h> +#include <asm/sstate.h> /* If the hypervisor indicates that the API setting * calls are unsupported, by returning HV_EBADTRAP or @@ -107,7 +108,7 @@ int sun4v_hvapi_register(unsigned long group, unsigned long major, p->minor = actual_minor; ret = 0; } else if (hv_ret == HV_EBADTRAP || - HV_ENOTSUPPORTED) { + hv_ret == HV_ENOTSUPPORTED) { if (p->flags & FLAG_PRE_API) { if (major == 1) { p->major = 1; @@ -179,6 +180,8 @@ void __init sun4v_hvapi_init(void) if (sun4v_hvapi_register(group, major, &minor)) goto bad; + sun4v_sstate_init(); + return; bad: diff --git a/arch/sparc64/kernel/irq.c b/arch/sparc64/kernel/irq.c index 3edc18e1b81..a36f8dd0c02 100644 --- a/arch/sparc64/kernel/irq.c +++ b/arch/sparc64/kernel/irq.c @@ -171,8 +171,6 @@ skip: return 0; } -extern unsigned long real_hard_smp_processor_id(void); - static unsigned int sun4u_compute_tid(unsigned long imap, unsigned long cpuid) { unsigned int tid; @@ -694,9 +692,20 @@ void init_irqwork_curcpu(void) trap_block[cpu].irq_worklist = 0; } -static void __cpuinit register_one_mondo(unsigned long paddr, unsigned long type) +/* Please be very careful with register_one_mondo() and + * sun4v_register_mondo_queues(). + * + * On SMP this gets invoked from the CPU trampoline before + * the cpu has fully taken over the trap table from OBP, + * and it's kernel stack + %g6 thread register state is + * not fully cooked yet. + * + * Therefore you cannot make any OBP calls, not even prom_printf, + * from these two routines. + */ +static void __cpuinit register_one_mondo(unsigned long paddr, unsigned long type, unsigned long qmask) { - unsigned long num_entries = 128; + unsigned long num_entries = (qmask + 1) / 64; unsigned long status; status = sun4v_cpu_qconf(type, paddr, num_entries); @@ -711,44 +720,58 @@ static void __cpuinit sun4v_register_mondo_queues(int this_cpu) { struct trap_per_cpu *tb = &trap_block[this_cpu]; - register_one_mondo(tb->cpu_mondo_pa, HV_CPU_QUEUE_CPU_MONDO); - register_one_mondo(tb->dev_mondo_pa, HV_CPU_QUEUE_DEVICE_MONDO); - register_one_mondo(tb->resum_mondo_pa, HV_CPU_QUEUE_RES_ERROR); - register_one_mondo(tb->nonresum_mondo_pa, HV_CPU_QUEUE_NONRES_ERROR); + register_one_mondo(tb->cpu_mondo_pa, HV_CPU_QUEUE_CPU_MONDO, + tb->cpu_mondo_qmask); + register_one_mondo(tb->dev_mondo_pa, HV_CPU_QUEUE_DEVICE_MONDO, + tb->dev_mondo_qmask); + register_one_mondo(tb->resum_mondo_pa, HV_CPU_QUEUE_RES_ERROR, + tb->resum_qmask); + register_one_mondo(tb->nonresum_mondo_pa, HV_CPU_QUEUE_NONRES_ERROR, + tb->nonresum_qmask); } -static void __cpuinit alloc_one_mondo(unsigned long *pa_ptr, int use_bootmem) +static void __cpuinit alloc_one_mondo(unsigned long *pa_ptr, unsigned long qmask, int use_bootmem) { - void *page; + unsigned long size = PAGE_ALIGN(qmask + 1); + unsigned long order = get_order(size); + void *p = NULL; - if (use_bootmem) - page = alloc_bootmem_low_pages(PAGE_SIZE); - else - page = (void *) get_zeroed_page(GFP_ATOMIC); + if (use_bootmem) { + p = __alloc_bootmem_low(size, size, 0); + } else { + struct page *page = alloc_pages(GFP_ATOMIC | __GFP_ZERO, order); + if (page) + p = page_address(page); + } - if (!page) { + if (!p) { prom_printf("SUN4V: Error, cannot allocate mondo queue.\n"); prom_halt(); } - *pa_ptr = __pa(page); + *pa_ptr = __pa(p); } -static void __cpuinit alloc_one_kbuf(unsigned long *pa_ptr, int use_bootmem) +static void __cpuinit alloc_one_kbuf(unsigned long *pa_ptr, unsigned long qmask, int use_bootmem) { - void *page; + unsigned long size = PAGE_ALIGN(qmask + 1); + unsigned long order = get_order(size); + void *p = NULL; - if (use_bootmem) - page = alloc_bootmem_low_pages(PAGE_SIZE); - else - page = (void *) get_zeroed_page(GFP_ATOMIC); + if (use_bootmem) { + p = __alloc_bootmem_low(size, size, 0); + } else { + struct page *page = alloc_pages(GFP_ATOMIC | __GFP_ZERO, order); + if (page) + p = page_address(page); + } - if (!page) { + if (!p) { prom_printf("SUN4V: Error, cannot allocate kbuf page.\n"); prom_halt(); } - *pa_ptr = __pa(page); + *pa_ptr = __pa(p); } static void __cpuinit init_cpu_send_mondo_info(struct trap_per_cpu *tb, int use_bootmem) @@ -779,12 +802,12 @@ void __cpuinit sun4v_init_mondo_queues(int use_bootmem, int cpu, int alloc, int struct trap_per_cpu *tb = &trap_block[cpu]; if (alloc) { - alloc_one_mondo(&tb->cpu_mondo_pa, use_bootmem); - alloc_one_mondo(&tb->dev_mondo_pa, use_bootmem); - alloc_one_mondo(&tb->resum_mondo_pa, use_bootmem); - alloc_one_kbuf(&tb->resum_kernel_buf_pa, use_bootmem); - alloc_one_mondo(&tb->nonresum_mondo_pa, use_bootmem); - alloc_one_kbuf(&tb->nonresum_kernel_buf_pa, use_bootmem); + alloc_one_mondo(&tb->cpu_mondo_pa, tb->cpu_mondo_qmask, use_bootmem); + alloc_one_mondo(&tb->dev_mondo_pa, tb->dev_mondo_qmask, use_bootmem); + alloc_one_mondo(&tb->resum_mondo_pa, tb->resum_qmask, use_bootmem); + alloc_one_kbuf(&tb->resum_kernel_buf_pa, tb->resum_qmask, use_bootmem); + alloc_one_mondo(&tb->nonresum_mondo_pa, tb->nonresum_qmask, use_bootmem); + alloc_one_kbuf(&tb->nonresum_kernel_buf_pa, tb->nonresum_qmask, use_bootmem); init_cpu_send_mondo_info(tb, use_bootmem); } diff --git a/arch/sparc64/kernel/itlb_miss.S b/arch/sparc64/kernel/itlb_miss.S index ad46e2024f4..5a8377b5495 100644 --- a/arch/sparc64/kernel/itlb_miss.S +++ b/arch/sparc64/kernel/itlb_miss.S @@ -11,12 +11,12 @@ /* ITLB ** ICACHE line 2: TSB compare and TLB load */ bne,pn %xcc, tsb_miss_itlb ! Miss mov FAULT_CODE_ITLB, %g3 - andcc %g5, _PAGE_EXEC_4U, %g0 ! Executable? + sethi %hi(_PAGE_EXEC_4U), %g4 + andcc %g5, %g4, %g0 ! Executable? be,pn %xcc, tsb_do_fault nop ! Delay slot, fill me stxa %g5, [%g0] ASI_ITLB_DATA_IN ! Load TLB retry ! Trap done - nop /* ITLB ** ICACHE line 3: */ nop diff --git a/arch/sparc64/kernel/mdesc.c b/arch/sparc64/kernel/mdesc.c new file mode 100644 index 00000000000..9246c2cf957 --- /dev/null +++ b/arch/sparc64/kernel/mdesc.c @@ -0,0 +1,619 @@ +/* mdesc.c: Sun4V machine description handling. + * + * Copyright (C) 2007 David S. Miller <davem@davemloft.net> + */ +#include <linux/kernel.h> +#include <linux/types.h> +#include <linux/bootmem.h> +#include <linux/log2.h> + +#include <asm/hypervisor.h> +#include <asm/mdesc.h> +#include <asm/prom.h> +#include <asm/oplib.h> +#include <asm/smp.h> + +/* Unlike the OBP device tree, the machine description is a full-on + * DAG. An arbitrary number of ARCs are possible from one + * node to other nodes and thus we can't use the OBP device_node + * data structure to represent these nodes inside of the kernel. + * + * Actually, it isn't even a DAG, because there are back pointers + * which create cycles in the graph. + * + * mdesc_hdr and mdesc_elem describe the layout of the data structure + * we get from the Hypervisor. + */ +struct mdesc_hdr { + u32 version; /* Transport version */ + u32 node_sz; /* node block size */ + u32 name_sz; /* name block size */ + u32 data_sz; /* data block size */ +}; + +struct mdesc_elem { + u8 tag; +#define MD_LIST_END 0x00 +#define MD_NODE 0x4e +#define MD_NODE_END 0x45 +#define MD_NOOP 0x20 +#define MD_PROP_ARC 0x61 +#define MD_PROP_VAL 0x76 +#define MD_PROP_STR 0x73 +#define MD_PROP_DATA 0x64 + u8 name_len; + u16 resv; + u32 name_offset; + union { + struct { + u32 data_len; + u32 data_offset; + } data; + u64 val; + } d; +}; + +static struct mdesc_hdr *main_mdesc; +static struct mdesc_node *allnodes; + +static struct mdesc_node *allnodes_tail; +static unsigned int unique_id; + +static struct mdesc_node **mdesc_hash; +static unsigned int mdesc_hash_size; + +static inline unsigned int node_hashfn(u64 node) +{ + return ((unsigned int) (node ^ (node >> 8) ^ (node >> 16))) + & (mdesc_hash_size - 1); +} + +static inline void hash_node(struct mdesc_node *mp) +{ + struct mdesc_node **head = &mdesc_hash[node_hashfn(mp->node)]; + + mp->hash_next = *head; + *head = mp; + + if (allnodes_tail) { + allnodes_tail->allnodes_next = mp; + allnodes_tail = mp; + } else { + allnodes = allnodes_tail = mp; + } +} + +static struct mdesc_node *find_node(u64 node) +{ + struct mdesc_node *mp = mdesc_hash[node_hashfn(node)]; + + while (mp) { + if (mp->node == node) + return mp; + + mp = mp->hash_next; + } + return NULL; +} + +struct property *md_find_property(const struct mdesc_node *mp, + const char *name, + int *lenp) +{ + struct property *pp; + + for (pp = mp->properties; pp != 0; pp = pp->next) { + if (strcasecmp(pp->name, name) == 0) { + if (lenp) + *lenp = pp->length; + break; + } + } + return pp; +} +EXPORT_SYMBOL(md_find_property); + +/* + * Find a property with a given name for a given node + * and return the value. + */ +const void *md_get_property(const struct mdesc_node *mp, const char *name, + int *lenp) +{ + struct property *pp = md_find_property(mp, name, lenp); + return pp ? pp->value : NULL; +} +EXPORT_SYMBOL(md_get_property); + +struct mdesc_node *md_find_node_by_name(struct mdesc_node *from, + const char *name) +{ + struct mdesc_node *mp; + + mp = from ? from->allnodes_next : allnodes; + for (; mp != NULL; mp = mp->allnodes_next) { + if (strcmp(mp->name, name) == 0) + break; + } + return mp; +} +EXPORT_SYMBOL(md_find_node_by_name); + +static unsigned int mdesc_early_allocated; + +static void * __init mdesc_early_alloc(unsigned long size) +{ + void *ret; + + ret = __alloc_bootmem(size, SMP_CACHE_BYTES, 0UL); + if (ret == NULL) { + prom_printf("MDESC: alloc of %lu bytes failed.\n", size); + prom_halt(); + } + + memset(ret, 0, size); + + mdesc_early_allocated += size; + + return ret; +} + +static unsigned int __init count_arcs(struct mdesc_elem *ep) +{ + unsigned int ret = 0; + + ep++; + while (ep->tag != MD_NODE_END) { + if (ep->tag == MD_PROP_ARC) + ret++; + ep++; + } + return ret; +} + +static void __init mdesc_node_alloc(u64 node, struct mdesc_elem *ep, const char *names) +{ + unsigned int num_arcs = count_arcs(ep); + struct mdesc_node *mp; + + mp = mdesc_early_alloc(sizeof(*mp) + + (num_arcs * sizeof(struct mdesc_arc))); + mp->name = names + ep->name_offset; + mp->node = node; + mp->unique_id = unique_id++; + mp->num_arcs = num_arcs; + + hash_node(mp); +} + +static inline struct mdesc_elem *node_block(struct mdesc_hdr *mdesc) +{ + return (struct mdesc_elem *) (mdesc + 1); +} + +static inline void *name_block(struct mdesc_hdr *mdesc) +{ + return ((void *) node_block(mdesc)) + mdesc->node_sz; +} + +static inline void *data_block(struct mdesc_hdr *mdesc) +{ + return ((void *) name_block(mdesc)) + mdesc->name_sz; +} + +/* In order to avoid recursion (the graph can be very deep) we use a + * two pass algorithm. First we allocate all the nodes and hash them. + * Then we iterate over each node, filling in the arcs and properties. + */ +static void __init build_all_nodes(struct mdesc_hdr *mdesc) +{ + struct mdesc_elem *start, *ep; + struct mdesc_node *mp; + const char *names; + void *data; + u64 last_node; + + start = ep = node_block(mdesc); + last_node = mdesc->node_sz / 16; + + names = name_block(mdesc); + + while (1) { + u64 node = ep - start; + + if (ep->tag == MD_LIST_END) + break; + + if (ep->tag != MD_NODE) { + prom_printf("MDESC: Inconsistent element list.\n"); + prom_halt(); + } + + mdesc_node_alloc(node, ep, names); + + if (ep->d.val >= last_node) { + printk("MDESC: Warning, early break out of node scan.\n"); + printk("MDESC: Next node [%lu] last_node [%lu].\n", + node, last_node); + break; + } + + ep = start + ep->d.val; + } + + data = data_block(mdesc); + for (mp = allnodes; mp; mp = mp->allnodes_next) { + struct mdesc_elem *ep = start + mp->node; + struct property **link = &mp->properties; + unsigned int this_arc = 0; + + ep++; + while (ep->tag != MD_NODE_END) { + switch (ep->tag) { + case MD_PROP_ARC: { + struct mdesc_node *target; + + if (this_arc >= mp->num_arcs) { + prom_printf("MDESC: ARC overrun [%u:%u]\n", + this_arc, mp->num_arcs); + prom_halt(); + } + target = find_node(ep->d.val); + if (!target) { + printk("MDESC: Warning, arc points to " + "missing node, ignoring.\n"); + break; + } + mp->arcs[this_arc].name = + (names + ep->name_offset); + mp->arcs[this_arc].arc = target; + this_arc++; + break; + } + + case MD_PROP_VAL: + case MD_PROP_STR: + case MD_PROP_DATA: { + struct property *p = mdesc_early_alloc(sizeof(*p)); + + p->unique_id = unique_id++; + p->name = (char *) names + ep->name_offset; + if (ep->tag == MD_PROP_VAL) { + p->value = &ep->d.val; + p->length = 8; + } else { + p->value = data + ep->d.data.data_offset; + p->length = ep->d.data.data_len; + } + *link = p; + link = &p->next; + break; + } + + case MD_NOOP: + break; + + default: + printk("MDESC: Warning, ignoring unknown tag type %02x\n", + ep->tag); + } + ep++; + } + } +} + +static unsigned int __init count_nodes(struct mdesc_hdr *mdesc) +{ + struct mdesc_elem *ep = node_block(mdesc); + struct mdesc_elem *end; + unsigned int cnt = 0; + + end = ((void *)ep) + mdesc->node_sz; + while (ep < end) { + if (ep->tag == MD_NODE) + cnt++; + ep++; + } + return cnt; +} + +static void __init report_platform_properties(void) +{ + struct mdesc_node *pn = md_find_node_by_name(NULL, "platform"); + const char *s; + const u64 *v; + + if (!pn) { + prom_printf("No platform node in machine-description.\n"); + prom_halt(); + } + + s = md_get_property(pn, "banner-name", NULL); + printk("PLATFORM: banner-name [%s]\n", s); + s = md_get_property(pn, "name", NULL); + printk("PLATFORM: name [%s]\n", s); + + v = md_get_property(pn, "hostid", NULL); + if (v) + printk("PLATFORM: hostid [%08lx]\n", *v); + v = md_get_property(pn, "serial#", NULL); + if (v) + printk("PLATFORM: serial# [%08lx]\n", *v); + v = md_get_property(pn, "stick-frequency", NULL); + printk("PLATFORM: stick-frequency [%08lx]\n", *v); + v = md_get_property(pn, "mac-address", NULL); + if (v) + printk("PLATFORM: mac-address [%lx]\n", *v); + v = md_get_property(pn, "watchdog-resolution", NULL); + if (v) + printk("PLATFORM: watchdog-resolution [%lu ms]\n", *v); + v = md_get_property(pn, "watchdog-max-timeout", NULL); + if (v) + printk("PLATFORM: watchdog-max-timeout [%lu ms]\n", *v); + v = md_get_property(pn, "max-cpus", NULL); + if (v) + printk("PLATFORM: max-cpus [%lu]\n", *v); +} + +static int inline find_in_proplist(const char *list, const char *match, int len) +{ + while (len > 0) { + int l; + + if (!strcmp(list, match)) + return 1; + l = strlen(list) + 1; + list += l; + len -= l; + } + return 0; +} + +static void __init fill_in_one_cache(cpuinfo_sparc *c, struct mdesc_node *mp) +{ + const u64 *level = md_get_property(mp, "level", NULL); + const u64 *size = md_get_property(mp, "size", NULL); + const u64 *line_size = md_get_property(mp, "line-size", NULL); + const char *type; + int type_len; + + type = md_get_property(mp, "type", &type_len); + + switch (*level) { + case 1: + if (find_in_proplist(type, "instn", type_len)) { + c->icache_size = *size; + c->icache_line_size = *line_size; + } else if (find_in_proplist(type, "data", type_len)) { + c->dcache_size = *size; + c->dcache_line_size = *line_size; + } + break; + + case 2: + c->ecache_size = *size; + c->ecache_line_size = *line_size; + break; + + default: + break; + } + + if (*level == 1) { + unsigned int i; + + for (i = 0; i < mp->num_arcs; i++) { + struct mdesc_node *t = mp->arcs[i].arc; + + if (strcmp(mp->arcs[i].name, "fwd")) + continue; + + if (!strcmp(t->name, "cache")) + fill_in_one_cache(c, t); + } + } +} + +static void __init mark_core_ids(struct mdesc_node *mp, int core_id) +{ + unsigned int i; + + for (i = 0; i < mp->num_arcs; i++) { + struct mdesc_node *t = mp->arcs[i].arc; + const u64 *id; + + if (strcmp(mp->arcs[i].name, "back")) + continue; + + if (!strcmp(t->name, "cpu")) { + id = md_get_property(t, "id", NULL); + if (*id < NR_CPUS) + cpu_data(*id).core_id = core_id; + } else { + unsigned int j; + + for (j = 0; j < t->num_arcs; j++) { + struct mdesc_node *n = t->arcs[j].arc; + + if (strcmp(t->arcs[j].name, "back")) + continue; + + if (strcmp(n->name, "cpu")) + continue; + + id = md_get_property(n, "id", NULL); + if (*id < NR_CPUS) + cpu_data(*id).core_id = core_id; + } + } + } +} + +static void __init set_core_ids(void) +{ + struct mdesc_node *mp; + int idx; + + idx = 1; + md_for_each_node_by_name(mp, "cache") { + const u64 *level = md_get_property(mp, "level", NULL); + const char *type; + int len; + + if (*level != 1) + continue; + + type = md_get_property(mp, "type", &len); + if (!find_in_proplist(type, "instn", len)) + continue; + + mark_core_ids(mp, idx); + + idx++; + } +} + +static void __init get_one_mondo_bits(const u64 *p, unsigned int *mask, unsigned char def) +{ + u64 val; + + if (!p) + goto use_default; + val = *p; + + if (!val || val >= 64) + goto use_default; + + *mask = ((1U << val) * 64U) - 1U; + return; + +use_default: + *mask = ((1U << def) * 64U) - 1U; +} + +static void __init get_mondo_data(struct mdesc_node *mp, struct trap_per_cpu *tb) +{ + const u64 *val; + + val = md_get_property(mp, "q-cpu-mondo-#bits", NULL); + get_one_mondo_bits(val, &tb->cpu_mondo_qmask, 7); + + val = md_get_property(mp, "q-dev-mondo-#bits", NULL); + get_one_mondo_bits(val, &tb->dev_mondo_qmask, 7); + + val = md_get_property(mp, "q-resumable-#bits", NULL); + get_one_mondo_bits(val, &tb->resum_qmask, 6); + + val = md_get_property(mp, "q-nonresumable-#bits", NULL); + get_one_mondo_bits(val, &tb->nonresum_qmask, 2); +} + +static void __init mdesc_fill_in_cpu_data(void) +{ + struct mdesc_node *mp; + + ncpus_probed = 0; + md_for_each_node_by_name(mp, "cpu") { + const u64 *id = md_get_property(mp, "id", NULL); + const u64 *cfreq = md_get_property(mp, "clock-frequency", NULL); + struct trap_per_cpu *tb; + cpuinfo_sparc *c; + unsigned int i; + int cpuid; + + ncpus_probed++; + + cpuid = *id; + +#ifdef CONFIG_SMP + if (cpuid >= NR_CPUS) + continue; +#else + /* On uniprocessor we only want the values for the + * real physical cpu the kernel booted onto, however + * cpu_data() only has one entry at index 0. + */ + if (cpuid != real_hard_smp_processor_id()) + continue; + cpuid = 0; +#endif + + c = &cpu_data(cpuid); + c->clock_tick = *cfreq; + + tb = &trap_block[cpuid]; + get_mondo_data(mp, tb); + + for (i = 0; i < mp->num_arcs; i++) { + struct mdesc_node *t = mp->arcs[i].arc; + unsigned int j; + + if (strcmp(mp->arcs[i].name, "fwd")) + continue; + + if (!strcmp(t->name, "cache")) { + fill_in_one_cache(c, t); + continue; + } + + for (j = 0; j < t->num_arcs; j++) { + struct mdesc_node *n; + + n = t->arcs[j].arc; + if (strcmp(t->arcs[j].name, "fwd")) + continue; + + if (!strcmp(n->name, "cache")) + fill_in_one_cache(c, n); + } + } + +#ifdef CONFIG_SMP + cpu_set(cpuid, cpu_present_map); + cpu_set(cpuid, phys_cpu_present_map); +#endif + + c->core_id = 0; + } + + set_core_ids(); + + smp_fill_in_sib_core_maps(); +} + +void __init sun4v_mdesc_init(void) +{ + unsigned long len, real_len, status; + + (void) sun4v_mach_desc(0UL, 0UL, &len); + + printk("MDESC: Size is %lu bytes.\n", len); + + main_mdesc = mdesc_early_alloc(len); + + status = sun4v_mach_desc(__pa(main_mdesc), len, &real_len); + if (status != HV_EOK || real_len > len) { + prom_printf("sun4v_mach_desc fails, err(%lu), " + "len(%lu), real_len(%lu)\n", + status, len, real_len); + prom_halt(); + } + + len = count_nodes(main_mdesc); + printk("MDESC: %lu nodes.\n", len); + + len = roundup_pow_of_two(len); + + mdesc_hash = mdesc_early_alloc(len * sizeof(struct mdesc_node *)); + mdesc_hash_size = len; + + printk("MDESC: Hash size %lu entries.\n", len); + + build_all_nodes(main_mdesc); + + printk("MDESC: Built graph with %u bytes of memory.\n", + mdesc_early_allocated); + + report_platform_properties(); + mdesc_fill_in_cpu_data(); +} diff --git a/arch/sparc64/kernel/pci.c b/arch/sparc64/kernel/pci.c index d4c077dc5e8..38a32bc95d2 100644 --- a/arch/sparc64/kernel/pci.c +++ b/arch/sparc64/kernel/pci.c @@ -306,6 +306,20 @@ static void __init pci_controller_probe(void) pci_controller_scan(pci_controller_init); } +static int ofpci_verbose; + +static int __init ofpci_debug(char *str) +{ + int val = 0; + + get_option(&str, &val); + if (val) + ofpci_verbose = 1; + return 1; +} + +__setup("ofpci_debug=", ofpci_debug); + static unsigned long pci_parse_of_flags(u32 addr0) { unsigned long flags = 0; @@ -337,7 +351,9 @@ static void pci_parse_of_addrs(struct of_device *op, addrs = of_get_property(node, "assigned-addresses", &proplen); if (!addrs) return; - printk(" parse addresses (%d bytes) @ %p\n", proplen, addrs); + if (ofpci_verbose) + printk(" parse addresses (%d bytes) @ %p\n", + proplen, addrs); op_res = &op->resource[0]; for (; proplen >= 20; proplen -= 20, addrs += 5, op_res++) { struct resource *res; @@ -348,8 +364,9 @@ static void pci_parse_of_addrs(struct of_device *op, if (!flags) continue; i = addrs[0] & 0xff; - printk(" start: %lx, end: %lx, i: %x\n", - op_res->start, op_res->end, i); + if (ofpci_verbose) + printk(" start: %lx, end: %lx, i: %x\n", + op_res->start, op_res->end, i); if (PCI_BASE_ADDRESS_0 <= i && i <= PCI_BASE_ADDRESS_5) { res = &dev->resource[(i - PCI_BASE_ADDRESS_0) >> 2]; @@ -393,8 +410,9 @@ struct pci_dev *of_create_pci_dev(struct pci_pbm_info *pbm, if (type == NULL) type = ""; - printk(" create device, devfn: %x, type: %s hostcontroller(%d)\n", - devfn, type, host_controller); + if (ofpci_verbose) + printk(" create device, devfn: %x, type: %s\n", + devfn, type); dev->bus = bus; dev->sysdata = node; @@ -434,8 +452,9 @@ struct pci_dev *of_create_pci_dev(struct pci_pbm_info *pbm, sprintf(pci_name(dev), "%04x:%02x:%02x.%d", pci_domain_nr(bus), dev->bus->number, PCI_SLOT(devfn), PCI_FUNC(devfn)); } - printk(" class: 0x%x device name: %s\n", - dev->class, pci_name(dev)); + if (ofpci_verbose) + printk(" class: 0x%x device name: %s\n", + dev->class, pci_name(dev)); /* I have seen IDE devices which will not respond to * the bmdma simplex check reads if bus mastering is @@ -469,7 +488,8 @@ struct pci_dev *of_create_pci_dev(struct pci_pbm_info *pbm, } pci_parse_of_addrs(sd->op, node, dev); - printk(" adding to system ...\n"); + if (ofpci_verbose) + printk(" adding to system ...\n"); pci_device_add(dev, bus); @@ -547,7 +567,8 @@ static void __devinit of_scan_pci_bridge(struct pci_pbm_info *pbm, unsigned int flags; u64 size; - printk("of_scan_pci_bridge(%s)\n", node->full_name); + if (ofpci_verbose) + printk("of_scan_pci_bridge(%s)\n", node->full_name); /* parse bus-range property */ busrange = of_get_property(node, "bus-range", &len); @@ -632,7 +653,8 @@ static void __devinit of_scan_pci_bridge(struct pci_pbm_info *pbm, simba_cont: sprintf(bus->name, "PCI Bus %04x:%02x", pci_domain_nr(bus), bus->number); - printk(" bus name: %s\n", bus->name); + if (ofpci_verbose) + printk(" bus name: %s\n", bus->name); pci_of_scan_bus(pbm, node, bus); } @@ -646,12 +668,14 @@ static void __devinit pci_of_scan_bus(struct pci_pbm_info *pbm, int reglen, devfn; struct pci_dev *dev; - printk("PCI: scan_bus[%s] bus no %d\n", - node->full_name, bus->number); + if (ofpci_verbose) + printk("PCI: scan_bus[%s] bus no %d\n", + node->full_name, bus->number); child = NULL; while ((child = of_get_next_child(node, child)) != NULL) { - printk(" * %s\n", child->full_name); + if (ofpci_verbose) + printk(" * %s\n", child->full_name); reg = of_get_property(child, "reg", ®len); if (reg == NULL || reglen < 20) continue; @@ -661,7 +685,9 @@ static void __devinit pci_of_scan_bus(struct pci_pbm_info *pbm, dev = of_create_pci_dev(pbm, child, bus, devfn, 0); if (!dev) continue; - printk("PCI: dev header type: %x\n", dev->hdr_type); + if (ofpci_verbose) + printk("PCI: dev header type: %x\n", + dev->hdr_type); if (dev->hdr_type == PCI_HEADER_TYPE_BRIDGE || dev->hdr_type == PCI_HEADER_TYPE_CARDBUS) diff --git a/arch/sparc64/kernel/pci_sabre.c b/arch/sparc64/kernel/pci_sabre.c index e2377796de8..323d6c27851 100644 --- a/arch/sparc64/kernel/pci_sabre.c +++ b/arch/sparc64/kernel/pci_sabre.c @@ -762,9 +762,10 @@ void sabre_init(struct device_node *dp, char *model_name) /* Of course, Sun has to encode things a thousand * different ways, inconsistently. */ - cpu_find_by_instance(0, &dp, NULL); - if (!strcmp(dp->name, "SUNW,UltraSPARC-IIe")) - hummingbird_p = 1; + for_each_node_by_type(dp, "cpu") { + if (!strcmp(dp->name, "SUNW,UltraSPARC-IIe")) + hummingbird_p = 1; + } } } diff --git a/arch/sparc64/kernel/pci_sun4v.c b/arch/sparc64/kernel/pci_sun4v.c index 044e8ec4c0f..6b3fe2c1d65 100644 --- a/arch/sparc64/kernel/pci_sun4v.c +++ b/arch/sparc64/kernel/pci_sun4v.c @@ -12,6 +12,7 @@ #include <linux/percpu.h> #include <linux/irq.h> #include <linux/msi.h> +#include <linux/log2.h> #include <asm/iommu.h> #include <asm/irq.h> @@ -26,6 +27,9 @@ #include "pci_sun4v.h" +static unsigned long vpci_major = 1; +static unsigned long vpci_minor = 1; + #define PGLIST_NENTS (PAGE_SIZE / sizeof(u64)) struct iommu_batch { @@ -638,9 +642,8 @@ static void pci_sun4v_iommu_init(struct pci_pbm_info *pbm) { struct iommu *iommu = pbm->iommu; struct property *prop; - unsigned long num_tsb_entries, sz; + unsigned long num_tsb_entries, sz, tsbsize; u32 vdma[2], dma_mask, dma_offset; - int tsbsize; prop = of_find_property(pbm->prom_node, "virtual-dma", NULL); if (prop) { @@ -654,31 +657,15 @@ static void pci_sun4v_iommu_init(struct pci_pbm_info *pbm) vdma[1] = 0x80000000; } - dma_mask = vdma[0]; - switch (vdma[1]) { - case 0x20000000: - dma_mask |= 0x1fffffff; - tsbsize = 64; - break; - - case 0x40000000: - dma_mask |= 0x3fffffff; - tsbsize = 128; - break; - - case 0x80000000: - dma_mask |= 0x7fffffff; - tsbsize = 256; - break; - - default: - prom_printf("PCI-SUN4V: strange virtual-dma size.\n"); - prom_halt(); + if ((vdma[0] | vdma[1]) & ~IO_PAGE_MASK) { + prom_printf("PCI-SUN4V: strange virtual-dma[%08x:%08x].\n", + vdma[0], vdma[1]); + prom_halt(); }; - tsbsize *= (8 * 1024); - - num_tsb_entries = tsbsize / sizeof(iopte_t); + dma_mask = (roundup_pow_of_two(vdma[1]) - 1UL); + num_tsb_entries = vdma[1] / IO_PAGE_SIZE; + tsbsize = num_tsb_entries * sizeof(iopte_t); dma_offset = vdma[0]; @@ -689,7 +676,7 @@ static void pci_sun4v_iommu_init(struct pci_pbm_info *pbm) iommu->dma_addr_mask = dma_mask; /* Allocate and initialize the free area map. */ - sz = num_tsb_entries / 8; + sz = (num_tsb_entries + 7) / 8; sz = (sz + 7UL) & ~7UL; iommu->arena.map = kzalloc(sz, GFP_KERNEL); if (!iommu->arena.map) { @@ -1178,6 +1165,7 @@ static void pci_sun4v_pbm_init(struct pci_controller_info *p, struct device_node void sun4v_pci_init(struct device_node *dp, char *model_name) { + static int hvapi_negotiated = 0; struct pci_controller_info *p; struct pci_pbm_info *pbm; struct iommu *iommu; @@ -1186,6 +1174,20 @@ void sun4v_pci_init(struct device_node *dp, char *model_name) u32 devhandle; int i; + if (!hvapi_negotiated++) { + int err = sun4v_hvapi_register(HV_GRP_PCI, + vpci_major, + &vpci_minor); + + if (err) { + prom_printf("SUN4V_PCI: Could not register hvapi, " + "err=%d\n", err); + prom_halt(); + } + printk("SUN4V_PCI: Registered hvapi major[%lu] minor[%lu]\n", + vpci_major, vpci_minor); + } + prop = of_find_property(dp, "reg", NULL); regs = prop->value; diff --git a/arch/sparc64/kernel/power.c b/arch/sparc64/kernel/power.c index 699b24b890d..5d6adea3967 100644 --- a/arch/sparc64/kernel/power.c +++ b/arch/sparc64/kernel/power.c @@ -19,6 +19,7 @@ #include <asm/prom.h> #include <asm/of_device.h> #include <asm/io.h> +#include <asm/sstate.h> #include <linux/unistd.h> @@ -53,6 +54,7 @@ static void (*poweroff_method)(void) = machine_alt_power_off; void machine_power_off(void) { + sstate_poweroff(); if (!serial_console || scons_pwroff) { #ifdef CONFIG_PCI if (power_reg) { diff --git a/arch/sparc64/kernel/process.c b/arch/sparc64/kernel/process.c index 952762bfb4c..f5f97e2c669 100644 --- a/arch/sparc64/kernel/process.c +++ b/arch/sparc64/kernel/process.c @@ -45,6 +45,7 @@ #include <asm/mmu_context.h> #include <asm/unistd.h> #include <asm/hypervisor.h> +#include <asm/sstate.h> /* #define VERBOSE_SHOWREGS */ @@ -106,6 +107,7 @@ extern void (*prom_keyboard)(void); void machine_halt(void) { + sstate_halt(); if (!serial_console && prom_palette) prom_palette (1); if (prom_keyboard) @@ -116,6 +118,7 @@ void machine_halt(void) void machine_alt_power_off(void) { + sstate_poweroff(); if (!serial_console && prom_palette) prom_palette(1); if (prom_keyboard) @@ -128,6 +131,7 @@ void machine_restart(char * cmd) { char *p; + sstate_reboot(); p = strchr (reboot_command, '\n'); if (p) *p = 0; if (!serial_console && prom_palette) diff --git a/arch/sparc64/kernel/prom.c b/arch/sparc64/kernel/prom.c index 02830e4671f..dad4b3ba705 100644 --- a/arch/sparc64/kernel/prom.c +++ b/arch/sparc64/kernel/prom.c @@ -28,6 +28,7 @@ #include <asm/irq.h> #include <asm/asi.h> #include <asm/upa.h> +#include <asm/smp.h> static struct device_node *allnodes; @@ -1665,6 +1666,150 @@ static struct device_node * __init build_tree(struct device_node *parent, phandl return ret; } +static const char *get_mid_prop(void) +{ + return (tlb_type == spitfire ? "upa-portid" : "portid"); +} + +struct device_node *of_find_node_by_cpuid(int cpuid) +{ + struct device_node *dp; + const char *mid_prop = get_mid_prop(); + + for_each_node_by_type(dp, "cpu") { + int id = of_getintprop_default(dp, mid_prop, -1); + const char *this_mid_prop = mid_prop; + + if (id < 0) { + this_mid_prop = "cpuid"; + id = of_getintprop_default(dp, this_mid_prop, -1); + } + + if (id < 0) { + prom_printf("OF: Serious problem, cpu lacks " + "%s property", this_mid_prop); + prom_halt(); + } + if (cpuid == id) + return dp; + } + return NULL; +} + +static void __init of_fill_in_cpu_data(void) +{ + struct device_node *dp; + const char *mid_prop = get_mid_prop(); + + ncpus_probed = 0; + for_each_node_by_type(dp, "cpu") { + int cpuid = of_getintprop_default(dp, mid_prop, -1); + const char *this_mid_prop = mid_prop; + struct device_node *portid_parent; + int portid = -1; + + portid_parent = NULL; + if (cpuid < 0) { + this_mid_prop = "cpuid"; + cpuid = of_getintprop_default(dp, this_mid_prop, -1); + if (cpuid >= 0) { + int limit = 2; + + portid_parent = dp; + while (limit--) { + portid_parent = portid_parent->parent; + if (!portid_parent) + break; + portid = of_getintprop_default(portid_parent, + "portid", -1); + if (portid >= 0) + break; + } + } + } + + if (cpuid < 0) { + prom_printf("OF: Serious problem, cpu lacks " + "%s property", this_mid_prop); + prom_halt(); + } + + ncpus_probed++; + +#ifdef CONFIG_SMP + if (cpuid >= NR_CPUS) + continue; +#else + /* On uniprocessor we only want the values for the + * real physical cpu the kernel booted onto, however + * cpu_data() only has one entry at index 0. + */ + if (cpuid != real_hard_smp_processor_id()) + continue; + cpuid = 0; +#endif + + cpu_data(cpuid).clock_tick = + of_getintprop_default(dp, "clock-frequency", 0); + + if (portid_parent) { + cpu_data(cpuid).dcache_size = + of_getintprop_default(dp, "l1-dcache-size", + 16 * 1024); + cpu_data(cpuid).dcache_line_size = + of_getintprop_default(dp, "l1-dcache-line-size", + 32); + cpu_data(cpuid).icache_size = + of_getintprop_default(dp, "l1-icache-size", + 8 * 1024); + cpu_data(cpuid).icache_line_size = + of_getintprop_default(dp, "l1-icache-line-size", + 32); + cpu_data(cpuid).ecache_size = + of_getintprop_default(dp, "l2-cache-size", 0); + cpu_data(cpuid).ecache_line_size = + of_getintprop_default(dp, "l2-cache-line-size", 0); + if (!cpu_data(cpuid).ecache_size || + !cpu_data(cpuid).ecache_line_size) { + cpu_data(cpuid).ecache_size = + of_getintprop_default(portid_parent, + "l2-cache-size", + (4 * 1024 * 1024)); + cpu_data(cpuid).ecache_line_size = + of_getintprop_default(portid_parent, + "l2-cache-line-size", 64); + } + + cpu_data(cpuid).core_id = portid + 1; + } else { + cpu_data(cpuid).dcache_size = + of_getintprop_default(dp, "dcache-size", 16 * 1024); + cpu_data(cpuid).dcache_line_size = + of_getintprop_default(dp, "dcache-line-size", 32); + + cpu_data(cpuid).icache_size = + of_getintprop_default(dp, "icache-size", 16 * 1024); + cpu_data(cpuid).icache_line_size = + of_getintprop_default(dp, "icache-line-size", 32); + + cpu_data(cpuid).ecache_size = + of_getintprop_default(dp, "ecache-size", + (4 * 1024 * 1024)); + cpu_data(cpuid).ecache_line_size = + of_getintprop_default(dp, "ecache-line-size", 64); + + cpu_data(cpuid).core_id = 0; + } + +#ifdef CONFIG_SMP + cpu_set(cpuid, cpu_present_map); + cpu_set(cpuid, phys_cpu_present_map); +#endif + } + + smp_fill_in_sib_core_maps(); +} + void __init prom_build_devicetree(void) { struct device_node **nextp; @@ -1679,4 +1824,7 @@ void __init prom_build_devicetree(void) &nextp); printk("PROM: Built device tree with %u bytes of memory.\n", prom_early_allocated); + + if (tlb_type != hypervisor) + of_fill_in_cpu_data(); } diff --git a/arch/sparc64/kernel/setup.c b/arch/sparc64/kernel/setup.c index dea9c3c9ec5..de9b4c13f1c 100644 --- a/arch/sparc64/kernel/setup.c +++ b/arch/sparc64/kernel/setup.c @@ -46,11 +46,17 @@ #include <asm/sections.h> #include <asm/setup.h> #include <asm/mmu.h> +#include <asm/ns87303.h> #ifdef CONFIG_IP_PNP #include <net/ipconfig.h> #endif +/* Used to synchronize accesses to NatSemi SUPER I/O chip configure + * operations in asm/ns87303.h + */ +DEFINE_SPINLOCK(ns87303_lock); + struct screen_info screen_info = { 0, 0, /* orig-x, orig-y */ 0, /* unused */ @@ -370,8 +376,6 @@ void __init setup_arch(char **cmdline_p) init_cur_cpu_trap(current_thread_info()); paging_init(); - - smp_setup_cpu_possible_map(); } static int __init set_preferred_console(void) @@ -424,7 +428,7 @@ extern void mmu_info(struct seq_file *); unsigned int dcache_parity_tl1_occurred; unsigned int icache_parity_tl1_occurred; -static int ncpus_probed; +int ncpus_probed; static int show_cpuinfo(struct seq_file *m, void *__unused) { @@ -516,14 +520,6 @@ static int __init topology_init(void) err = -ENOMEM; - /* Count the number of physically present processors in - * the machine, even on uniprocessor, so that /proc/cpuinfo - * output is consistent with 2.4.x - */ - ncpus_probed = 0; - while (!cpu_find_by_instance(ncpus_probed, NULL, NULL)) - ncpus_probed++; - for_each_possible_cpu(i) { struct cpu *p = kzalloc(sizeof(*p), GFP_KERNEL); if (p) { diff --git a/arch/sparc64/kernel/smp.c b/arch/sparc64/kernel/smp.c index 24fdf1d0adc..c550bba3490 100644 --- a/arch/sparc64/kernel/smp.c +++ b/arch/sparc64/kernel/smp.c @@ -40,6 +40,7 @@ #include <asm/tlb.h> #include <asm/sections.h> #include <asm/prom.h> +#include <asm/mdesc.h> extern void calibrate_delay(void); @@ -75,53 +76,6 @@ void smp_bogo(struct seq_file *m) i, cpu_data(i).clock_tick); } -void __init smp_store_cpu_info(int id) -{ - struct device_node *dp; - int def; - - cpu_data(id).udelay_val = loops_per_jiffy; - - cpu_find_by_mid(id, &dp); - cpu_data(id).clock_tick = - of_getintprop_default(dp, "clock-frequency", 0); - - def = ((tlb_type == hypervisor) ? (8 * 1024) : (16 * 1024)); - cpu_data(id).dcache_size = - of_getintprop_default(dp, "dcache-size", def); - - def = 32; - cpu_data(id).dcache_line_size = - of_getintprop_default(dp, "dcache-line-size", def); - - def = 16 * 1024; - cpu_data(id).icache_size = - of_getintprop_default(dp, "icache-size", def); - - def = 32; - cpu_data(id).icache_line_size = - of_getintprop_default(dp, "icache-line-size", def); - - def = ((tlb_type == hypervisor) ? - (3 * 1024 * 1024) : - (4 * 1024 * 1024)); - cpu_data(id).ecache_size = - of_getintprop_default(dp, "ecache-size", def); - - def = 64; - cpu_data(id).ecache_line_size = - of_getintprop_default(dp, "ecache-line-size", def); - - printk("CPU[%d]: Caches " - "D[sz(%d):line_sz(%d)] " - "I[sz(%d):line_sz(%d)] " - "E[sz(%d):line_sz(%d)]\n", - id, - cpu_data(id).dcache_size, cpu_data(id).dcache_line_size, - cpu_data(id).icache_size, cpu_data(id).icache_line_size, - cpu_data(id).ecache_size, cpu_data(id).ecache_line_size); -} - extern void setup_sparc64_timer(void); static volatile unsigned long callin_flag = 0; @@ -145,7 +99,7 @@ void __init smp_callin(void) local_irq_enable(); calibrate_delay(); - smp_store_cpu_info(cpuid); + cpu_data(cpuid).udelay_val = loops_per_jiffy; callin_flag = 1; __asm__ __volatile__("membar #Sync\n\t" "flush %%g6" : : : "memory"); @@ -340,9 +294,8 @@ static int __devinit smp_boot_one_cpu(unsigned int cpu) prom_startcpu_cpuid(cpu, entry, cookie); } else { - struct device_node *dp; + struct device_node *dp = of_find_node_by_cpuid(cpu); - cpu_find_by_mid(cpu, &dp); prom_startcpu(dp->node, entry, cookie); } @@ -447,7 +400,7 @@ static __inline__ void spitfire_xcall_deliver(u64 data0, u64 data1, u64 data2, c static void cheetah_xcall_deliver(u64 data0, u64 data1, u64 data2, cpumask_t mask) { u64 pstate, ver; - int nack_busy_id, is_jbus; + int nack_busy_id, is_jbus, need_more; if (cpus_empty(mask)) return; @@ -463,6 +416,7 @@ static void cheetah_xcall_deliver(u64 data0, u64 data1, u64 data2, cpumask_t mas __asm__ __volatile__("rdpr %%pstate, %0" : "=r" (pstate)); retry: + need_more = 0; __asm__ __volatile__("wrpr %0, %1, %%pstate\n\t" : : "r" (pstate), "i" (PSTATE_IE)); @@ -491,6 +445,10 @@ retry: : /* no outputs */ : "r" (target), "i" (ASI_INTR_W)); nack_busy_id++; + if (nack_busy_id == 32) { + need_more = 1; + break; + } } } @@ -507,6 +465,16 @@ retry: if (dispatch_stat == 0UL) { __asm__ __volatile__("wrpr %0, 0x0, %%pstate" : : "r" (pstate)); + if (unlikely(need_more)) { + int i, cnt = 0; + for_each_cpu_mask(i, mask) { + cpu_clear(i, mask); + cnt++; + if (cnt == 32) + break; + } + goto retry; + } return; } if (!--stuck) @@ -544,6 +512,8 @@ retry: if ((dispatch_stat & check_mask) == 0) cpu_clear(i, mask); this_busy_nack += 2; + if (this_busy_nack == 64) + break; } goto retry; @@ -1191,23 +1161,14 @@ int setup_profiling_timer(unsigned int multiplier) static void __init smp_tune_scheduling(void) { - struct device_node *dp; - int instance; - unsigned int def, smallest = ~0U; - - def = ((tlb_type == hypervisor) ? - (3 * 1024 * 1024) : - (4 * 1024 * 1024)); + unsigned int smallest = ~0U; + int i; - instance = 0; - while (!cpu_find_by_instance(instance, &dp, NULL)) { - unsigned int val; + for (i = 0; i < NR_CPUS; i++) { + unsigned int val = cpu_data(i).ecache_size; - val = of_getintprop_default(dp, "ecache-size", def); - if (val < smallest) + if (val && val < smallest) smallest = val; - - instance++; } /* Any value less than 256K is nonsense. */ @@ -1230,58 +1191,42 @@ void __init smp_prepare_cpus(unsigned int max_cpus) int i; if (num_possible_cpus() > max_cpus) { - int instance, mid; - - instance = 0; - while (!cpu_find_by_instance(instance, NULL, &mid)) { - if (mid != boot_cpu_id) { - cpu_clear(mid, phys_cpu_present_map); - cpu_clear(mid, cpu_present_map); + for_each_possible_cpu(i) { + if (i != boot_cpu_id) { + cpu_clear(i, phys_cpu_present_map); + cpu_clear(i, cpu_present_map); if (num_possible_cpus() <= max_cpus) break; } - instance++; } } - for_each_possible_cpu(i) { - if (tlb_type == hypervisor) { - int j; - - /* XXX get this mapping from machine description */ - for_each_possible_cpu(j) { - if ((j >> 2) == (i >> 2)) - cpu_set(j, cpu_sibling_map[i]); - } - } else { - cpu_set(i, cpu_sibling_map[i]); - } - } - - smp_store_cpu_info(boot_cpu_id); + cpu_data(boot_cpu_id).udelay_val = loops_per_jiffy; smp_tune_scheduling(); } -/* Set this up early so that things like the scheduler can init - * properly. We use the same cpu mask for both the present and - * possible cpu map. - */ -void __init smp_setup_cpu_possible_map(void) +void __devinit smp_prepare_boot_cpu(void) { - int instance, mid; - - instance = 0; - while (!cpu_find_by_instance(instance, NULL, &mid)) { - if (mid < NR_CPUS) { - cpu_set(mid, phys_cpu_present_map); - cpu_set(mid, cpu_present_map); - } - instance++; - } } -void __devinit smp_prepare_boot_cpu(void) +void __devinit smp_fill_in_sib_core_maps(void) { + unsigned int i; + + for_each_possible_cpu(i) { + unsigned int j; + + if (cpu_data(i).core_id == 0) { + cpu_set(i, cpu_sibling_map[i]); + continue; + } + + for_each_possible_cpu(j) { + if (cpu_data(i).core_id == + cpu_data(j).core_id) + cpu_set(j, cpu_sibling_map[i]); + } + } } int __cpuinit __cpu_up(unsigned int cpu) @@ -1337,7 +1282,7 @@ unsigned long __per_cpu_shift __read_mostly; EXPORT_SYMBOL(__per_cpu_base); EXPORT_SYMBOL(__per_cpu_shift); -void __init setup_per_cpu_areas(void) +void __init real_setup_per_cpu_areas(void) { unsigned long goal, size, i; char *ptr; diff --git a/arch/sparc64/kernel/sstate.c b/arch/sparc64/kernel/sstate.c new file mode 100644 index 00000000000..5b6e75b7f05 --- /dev/null +++ b/arch/sparc64/kernel/sstate.c @@ -0,0 +1,104 @@ +/* sstate.c: System soft state support. + * + * Copyright (C) 2007 David S. Miller <davem@davemloft.net> + */ + +#include <linux/kernel.h> +#include <linux/notifier.h> +#include <linux/init.h> + +#include <asm/hypervisor.h> +#include <asm/sstate.h> +#include <asm/oplib.h> +#include <asm/head.h> +#include <asm/io.h> + +static int hv_supports_soft_state; + +static unsigned long kimage_addr_to_ra(const char *p) +{ + unsigned long val = (unsigned long) p; + + return kern_base + (val - KERNBASE); +} + +static void do_set_sstate(unsigned long state, const char *msg) +{ + unsigned long err; + + if (!hv_supports_soft_state) + return; + + err = sun4v_mach_set_soft_state(state, kimage_addr_to_ra(msg)); + if (err) { + printk(KERN_WARNING "SSTATE: Failed to set soft-state to " + "state[%lx] msg[%s], err=%lu\n", + state, msg, err); + } +} + +static const char booting_msg[32] __attribute__((aligned(32))) = + "Linux booting"; +static const char running_msg[32] __attribute__((aligned(32))) = + "Linux running"; +static const char halting_msg[32] __attribute__((aligned(32))) = + "Linux halting"; +static const char poweroff_msg[32] __attribute__((aligned(32))) = + "Linux powering off"; +static const char rebooting_msg[32] __attribute__((aligned(32))) = + "Linux rebooting"; +static const char panicing_msg[32] __attribute__((aligned(32))) = + "Linux panicing"; + +void sstate_booting(void) +{ + do_set_sstate(HV_SOFT_STATE_TRANSITION, booting_msg); +} + +void sstate_running(void) +{ + do_set_sstate(HV_SOFT_STATE_NORMAL, running_msg); +} + +void sstate_halt(void) +{ + do_set_sstate(HV_SOFT_STATE_TRANSITION, halting_msg); +} + +void sstate_poweroff(void) +{ + do_set_sstate(HV_SOFT_STATE_TRANSITION, poweroff_msg); +} + +void sstate_reboot(void) +{ + do_set_sstate(HV_SOFT_STATE_TRANSITION, rebooting_msg); +} + +static int sstate_panic_event(struct notifier_block *n, unsigned long event, void *ptr) +{ + do_set_sstate(HV_SOFT_STATE_TRANSITION, panicing_msg); + + return NOTIFY_DONE; +} + +static struct notifier_block sstate_panic_block = { + .notifier_call = sstate_panic_event, + .priority = INT_MAX, +}; + +void __init sun4v_sstate_init(void) +{ + unsigned long major, minor; + + major = 1; + minor = 0; + if (sun4v_hvapi_register(HV_GRP_SOFT_STATE, major, &minor)) + return; + + hv_supports_soft_state = 1; + + prom_sun4v_guest_soft_state(); + atomic_notifier_chain_register(&panic_notifier_list, + &sstate_panic_block); +} diff --git a/arch/sparc64/kernel/sun4v_ivec.S b/arch/sparc64/kernel/sun4v_ivec.S index 405855dd886..574bc248bca 100644 --- a/arch/sparc64/kernel/sun4v_ivec.S +++ b/arch/sparc64/kernel/sun4v_ivec.S @@ -22,12 +22,12 @@ sun4v_cpu_mondo: be,pn %xcc, sun4v_cpu_mondo_queue_empty nop - /* Get &trap_block[smp_processor_id()] into %g3. */ - ldxa [%g0] ASI_SCRATCHPAD, %g3 - sub %g3, TRAP_PER_CPU_FAULT_INFO, %g3 + /* Get &trap_block[smp_processor_id()] into %g4. */ + ldxa [%g0] ASI_SCRATCHPAD, %g4 + sub %g4, TRAP_PER_CPU_FAULT_INFO, %g4 /* Get CPU mondo queue base phys address into %g7. */ - ldx [%g3 + TRAP_PER_CPU_CPU_MONDO_PA], %g7 + ldx [%g4 + TRAP_PER_CPU_CPU_MONDO_PA], %g7 /* Now get the cross-call arguments and handler PC, same * layout as sun4u: @@ -47,8 +47,7 @@ sun4v_cpu_mondo: add %g2, 0x40 - 0x8 - 0x8, %g2 /* Update queue head pointer. */ - sethi %hi(8192 - 1), %g4 - or %g4, %lo(8192 - 1), %g4 + lduw [%g4 + TRAP_PER_CPU_CPU_MONDO_QMASK], %g4 and %g2, %g4, %g2 mov INTRQ_CPU_MONDO_HEAD, %g4 @@ -71,12 +70,12 @@ sun4v_dev_mondo: be,pn %xcc, sun4v_dev_mondo_queue_empty nop - /* Get &trap_block[smp_processor_id()] into %g3. */ - ldxa [%g0] ASI_SCRATCHPAD, %g3 - sub %g3, TRAP_PER_CPU_FAULT_INFO, %g3 + /* Get &trap_block[smp_processor_id()] into %g4. */ + ldxa [%g0] ASI_SCRATCHPAD, %g4 + sub %g4, TRAP_PER_CPU_FAULT_INFO, %g4 /* Get DEV mondo queue base phys address into %g5. */ - ldx [%g3 + TRAP_PER_CPU_DEV_MONDO_PA], %g5 + ldx [%g4 + TRAP_PER_CPU_DEV_MONDO_PA], %g5 /* Load IVEC into %g3. */ ldxa [%g5 + %g2] ASI_PHYS_USE_EC, %g3 @@ -90,8 +89,7 @@ sun4v_dev_mondo: */ /* Update queue head pointer, this frees up some registers. */ - sethi %hi(8192 - 1), %g4 - or %g4, %lo(8192 - 1), %g4 + lduw [%g4 + TRAP_PER_CPU_DEV_MONDO_QMASK], %g4 and %g2, %g4, %g2 mov INTRQ_DEVICE_MONDO_HEAD, %g4 @@ -143,6 +141,8 @@ sun4v_res_mondo: brnz,pn %g1, sun4v_res_mondo_queue_full nop + lduw [%g3 + TRAP_PER_CPU_RESUM_QMASK], %g4 + /* Remember this entry's offset in %g1. */ mov %g2, %g1 @@ -173,8 +173,6 @@ sun4v_res_mondo: add %g2, 0x08, %g2 /* Update queue head pointer. */ - sethi %hi(8192 - 1), %g4 - or %g4, %lo(8192 - 1), %g4 and %g2, %g4, %g2 mov INTRQ_RESUM_MONDO_HEAD, %g4 @@ -254,6 +252,8 @@ sun4v_nonres_mondo: brnz,pn %g1, sun4v_nonres_mondo_queue_full nop + lduw [%g3 + TRAP_PER_CPU_NONRESUM_QMASK], %g4 + /* Remember this entry's offset in %g1. */ mov %g2, %g1 @@ -284,8 +284,6 @@ sun4v_nonres_mondo: add %g2, 0x08, %g2 /* Update queue head pointer. */ - sethi %hi(8192 - 1), %g4 - or %g4, %lo(8192 - 1), %g4 and %g2, %g4, %g2 mov INTRQ_NONRESUM_MONDO_HEAD, %g4 diff --git a/arch/sparc64/kernel/time.c b/arch/sparc64/kernel/time.c index 2d63d768996..a31a0439244 100644 --- a/arch/sparc64/kernel/time.c +++ b/arch/sparc64/kernel/time.c @@ -680,22 +680,14 @@ static int starfire_set_time(u32 val) static u32 hypervisor_get_time(void) { - register unsigned long func asm("%o5"); - register unsigned long arg0 asm("%o0"); - register unsigned long arg1 asm("%o1"); + unsigned long ret, time; int retries = 10000; retry: - func = HV_FAST_TOD_GET; - arg0 = 0; - arg1 = 0; - __asm__ __volatile__("ta %6" - : "=&r" (func), "=&r" (arg0), "=&r" (arg1) - : "0" (func), "1" (arg0), "2" (arg1), - "i" (HV_FAST_TRAP)); - if (arg0 == HV_EOK) - return arg1; - if (arg0 == HV_EWOULDBLOCK) { + ret = sun4v_tod_get(&time); + if (ret == HV_EOK) + return time; + if (ret == HV_EWOULDBLOCK) { if (--retries > 0) { udelay(100); goto retry; @@ -709,20 +701,14 @@ retry: static int hypervisor_set_time(u32 secs) { - register unsigned long func asm("%o5"); - register unsigned long arg0 asm("%o0"); + unsigned long ret; int retries = 10000; retry: - func = HV_FAST_TOD_SET; - arg0 = secs; - __asm__ __volatile__("ta %4" - : "=&r" (func), "=&r" (arg0) - : "0" (func), "1" (arg0), - "i" (HV_FAST_TRAP)); - if (arg0 == HV_EOK) + ret = sun4v_tod_set(secs); + if (ret == HV_EOK) return 0; - if (arg0 == HV_EWOULDBLOCK) { + if (ret == HV_EWOULDBLOCK) { if (--retries > 0) { udelay(100); goto retry; @@ -862,7 +848,6 @@ fs_initcall(clock_init); static unsigned long sparc64_init_timers(void) { struct device_node *dp; - struct property *prop; unsigned long clock; #ifdef CONFIG_SMP extern void smp_tick_init(void); @@ -879,17 +864,15 @@ static unsigned long sparc64_init_timers(void) if (manuf == 0x17 && impl == 0x13) { /* Hummingbird, aka Ultra-IIe */ tick_ops = &hbtick_operations; - prop = of_find_property(dp, "stick-frequency", NULL); + clock = of_getintprop_default(dp, "stick-frequency", 0); } else { tick_ops = &tick_operations; - cpu_find_by_instance(0, &dp, NULL); - prop = of_find_property(dp, "clock-frequency", NULL); + clock = local_cpu_data().clock_tick; } } else { tick_ops = &stick_operations; - prop = of_find_property(dp, "stick-frequency", NULL); + clock = of_getintprop_default(dp, "stick-frequency", 0); } - clock = *(unsigned int *) prop->value; #ifdef CONFIG_SMP smp_tick_init(); @@ -1365,6 +1348,7 @@ static int hypervisor_set_rtc_time(struct rtc_time *time) return hypervisor_set_time(seconds); } +#ifdef CONFIG_PCI static void bq4802_get_rtc_time(struct rtc_time *time) { unsigned char val = readb(bq4802_regs + 0x0e); @@ -1436,6 +1420,7 @@ static int bq4802_set_rtc_time(struct rtc_time *time) return 0; } +#endif /* CONFIG_PCI */ struct mini_rtc_ops { void (*get_rtc_time)(struct rtc_time *); @@ -1452,10 +1437,12 @@ static struct mini_rtc_ops hypervisor_rtc_ops = { .set_rtc_time = hypervisor_set_rtc_time, }; +#ifdef CONFIG_PCI static struct mini_rtc_ops bq4802_rtc_ops = { .get_rtc_time = bq4802_get_rtc_time, .set_rtc_time = bq4802_set_rtc_time, }; +#endif /* CONFIG_PCI */ static struct mini_rtc_ops *mini_rtc_ops; @@ -1579,8 +1566,10 @@ static int __init rtc_mini_init(void) mini_rtc_ops = &hypervisor_rtc_ops; else if (this_is_starfire) mini_rtc_ops = &starfire_rtc_ops; +#ifdef CONFIG_PCI else if (bq4802_regs) mini_rtc_ops = &bq4802_rtc_ops; +#endif /* CONFIG_PCI */ else return -ENODEV; diff --git a/arch/sparc64/kernel/traps.c b/arch/sparc64/kernel/traps.c index d0fde36395b..00a9e3286c8 100644 --- a/arch/sparc64/kernel/traps.c +++ b/arch/sparc64/kernel/traps.c @@ -795,8 +795,7 @@ extern unsigned int cheetah_deferred_trap_vector[], cheetah_deferred_trap_vector void __init cheetah_ecache_flush_init(void) { unsigned long largest_size, smallest_linesize, order, ver; - struct device_node *dp; - int i, instance, sz; + int i, sz; /* Scan all cpu device tree nodes, note two values: * 1) largest E-cache size @@ -805,18 +804,20 @@ void __init cheetah_ecache_flush_init(void) largest_size = 0UL; smallest_linesize = ~0UL; - instance = 0; - while (!cpu_find_by_instance(instance, &dp, NULL)) { + for (i = 0; i < NR_CPUS; i++) { unsigned long val; - val = of_getintprop_default(dp, "ecache-size", - (2 * 1024 * 1024)); + val = cpu_data(i).ecache_size; + if (!val) + continue; + if (val > largest_size) largest_size = val; - val = of_getintprop_default(dp, "ecache-line-size", 64); + + val = cpu_data(i).ecache_line_size; if (val < smallest_linesize) smallest_linesize = val; - instance++; + } if (largest_size == 0UL || smallest_linesize == ~0UL) { @@ -2564,7 +2565,15 @@ void __init trap_init(void) (TRAP_PER_CPU_TSB_HUGE_TEMP != offsetof(struct trap_per_cpu, tsb_huge_temp)) || (TRAP_PER_CPU_IRQ_WORKLIST != - offsetof(struct trap_per_cpu, irq_worklist))) + offsetof(struct trap_per_cpu, irq_worklist)) || + (TRAP_PER_CPU_CPU_MONDO_QMASK != + offsetof(struct trap_per_cpu, cpu_mondo_qmask)) || + (TRAP_PER_CPU_DEV_MONDO_QMASK != + offsetof(struct trap_per_cpu, dev_mondo_qmask)) || + (TRAP_PER_CPU_RESUM_QMASK != + offsetof(struct trap_per_cpu, resum_qmask)) || + (TRAP_PER_CPU_NONRESUM_QMASK != + offsetof(struct trap_per_cpu, nonresum_qmask))) trap_per_cpu_offsets_are_bolixed_dave(); if ((TSB_CONFIG_TSB != diff --git a/arch/sparc64/mm/init.c b/arch/sparc64/mm/init.c index 6e5b01d779d..3010227fe24 100644 --- a/arch/sparc64/mm/init.c +++ b/arch/sparc64/mm/init.c @@ -23,6 +23,7 @@ #include <linux/kprobes.h> #include <linux/cache.h> #include <linux/sort.h> +#include <linux/percpu.h> #include <asm/head.h> #include <asm/system.h> @@ -43,8 +44,8 @@ #include <asm/tsb.h> #include <asm/hypervisor.h> #include <asm/prom.h> - -extern void device_scan(void); +#include <asm/sstate.h> +#include <asm/mdesc.h> #define MAX_PHYS_ADDRESS (1UL << 42UL) #define KPTE_BITMAP_CHUNK_SZ (256UL * 1024UL * 1024UL) @@ -60,8 +61,11 @@ unsigned long kern_linear_pte_xor[2] __read_mostly; unsigned long kpte_linear_bitmap[KPTE_BITMAP_BYTES / sizeof(unsigned long)]; #ifndef CONFIG_DEBUG_PAGEALLOC -/* A special kernel TSB for 4MB and 256MB linear mappings. */ -struct tsb swapper_4m_tsb[KERNEL_TSB4M_NENTRIES]; +/* A special kernel TSB for 4MB and 256MB linear mappings. + * Space is allocated for this right after the trap table + * in arch/sparc64/kernel/head.S + */ +extern struct tsb swapper_4m_tsb[KERNEL_TSB4M_NENTRIES]; #endif #define MAX_BANKS 32 @@ -190,12 +194,9 @@ inline void flush_dcache_page_impl(struct page *page) } #define PG_dcache_dirty PG_arch_1 -#define PG_dcache_cpu_shift 24UL -#define PG_dcache_cpu_mask (256UL - 1UL) - -#if NR_CPUS > 256 -#error D-cache dirty tracking and thread_info->cpu need fixing for > 256 cpus -#endif +#define PG_dcache_cpu_shift 32UL +#define PG_dcache_cpu_mask \ + ((1UL<<ilog2(roundup_pow_of_two(NR_CPUS)))-1UL) #define dcache_dirty_cpu(page) \ (((page)->flags >> PG_dcache_cpu_shift) & PG_dcache_cpu_mask) @@ -557,26 +558,11 @@ static void __init hypervisor_tlb_lock(unsigned long vaddr, unsigned long pte, unsigned long mmu) { - register unsigned long func asm("%o5"); - register unsigned long arg0 asm("%o0"); - register unsigned long arg1 asm("%o1"); - register unsigned long arg2 asm("%o2"); - register unsigned long arg3 asm("%o3"); - - func = HV_FAST_MMU_MAP_PERM_ADDR; - arg0 = vaddr; - arg1 = 0; - arg2 = pte; - arg3 = mmu; - __asm__ __volatile__("ta 0x80" - : "=&r" (func), "=&r" (arg0), - "=&r" (arg1), "=&r" (arg2), - "=&r" (arg3) - : "0" (func), "1" (arg0), "2" (arg1), - "3" (arg2), "4" (arg3)); - if (arg0 != 0) { + unsigned long ret = sun4v_mmu_map_perm_addr(vaddr, 0, pte, mmu); + + if (ret != 0) { prom_printf("hypervisor_tlb_lock[%lx:%lx:%lx:%lx]: " - "errors with %lx\n", vaddr, 0, pte, mmu, arg0); + "errors with %lx\n", vaddr, 0, pte, mmu, ret); prom_halt(); } } @@ -1313,20 +1299,16 @@ static void __init sun4v_ktsb_init(void) void __cpuinit sun4v_ktsb_register(void) { - register unsigned long func asm("%o5"); - register unsigned long arg0 asm("%o0"); - register unsigned long arg1 asm("%o1"); - unsigned long pa; + unsigned long pa, ret; pa = kern_base + ((unsigned long)&ktsb_descr[0] - KERNBASE); - func = HV_FAST_MMU_TSB_CTX0; - arg0 = NUM_KTSB_DESCR; - arg1 = pa; - __asm__ __volatile__("ta %6" - : "=&r" (func), "=&r" (arg0), "=&r" (arg1) - : "0" (func), "1" (arg0), "2" (arg1), - "i" (HV_FAST_TRAP)); + ret = sun4v_mmu_tsb_ctx0(NUM_KTSB_DESCR, pa); + if (ret != 0) { + prom_printf("hypervisor_mmu_tsb_ctx0[%lx]: " + "errors with %lx\n", pa, ret); + prom_halt(); + } } /* paging_init() sets up the page tables */ @@ -1334,6 +1316,9 @@ void __cpuinit sun4v_ktsb_register(void) extern void cheetah_ecache_flush_init(void); extern void sun4v_patch_tlb_handlers(void); +extern void cpu_probe(void); +extern void central_probe(void); + static unsigned long last_valid_pfn; pgd_t swapper_pg_dir[2048]; @@ -1345,9 +1330,24 @@ void __init paging_init(void) unsigned long end_pfn, pages_avail, shift, phys_base; unsigned long real_end, i; + /* These build time checkes make sure that the dcache_dirty_cpu() + * page->flags usage will work. + * + * When a page gets marked as dcache-dirty, we store the + * cpu number starting at bit 32 in the page->flags. Also, + * functions like clear_dcache_dirty_cpu use the cpu mask + * in 13-bit signed-immediate instruction fields. + */ + BUILD_BUG_ON(FLAGS_RESERVED != 32); + BUILD_BUG_ON(SECTIONS_WIDTH + NODES_WIDTH + ZONES_WIDTH + + ilog2(roundup_pow_of_two(NR_CPUS)) > FLAGS_RESERVED); + BUILD_BUG_ON(NR_CPUS > 4096); + kern_base = (prom_boot_mapping_phys_low >> 22UL) << 22UL; kern_size = (unsigned long)&_end - (unsigned long)KERNBASE; + sstate_booting(); + /* Invalidate both kernel TSBs. */ memset(swapper_tsb, 0x40, sizeof(swapper_tsb)); #ifndef CONFIG_DEBUG_PAGEALLOC @@ -1416,8 +1416,13 @@ void __init paging_init(void) kernel_physical_mapping_init(); + real_setup_per_cpu_areas(); + prom_build_devicetree(); + if (tlb_type == hypervisor) + sun4v_mdesc_init(); + { unsigned long zones_size[MAX_NR_ZONES]; unsigned long zholes_size[MAX_NR_ZONES]; @@ -1434,7 +1439,10 @@ void __init paging_init(void) zholes_size); } - device_scan(); + prom_printf("Booting Linux...\n"); + + central_probe(); + cpu_probe(); } static void __init taint_real_pages(void) diff --git a/arch/sparc64/prom/misc.c b/arch/sparc64/prom/misc.c index 0b4213720d4..f3e0c14e9ee 100644 --- a/arch/sparc64/prom/misc.c +++ b/arch/sparc64/prom/misc.c @@ -15,6 +15,25 @@ #include <asm/oplib.h> #include <asm/system.h> +int prom_service_exists(const char *service_name) +{ + int err = p1275_cmd("test", P1275_ARG(0, P1275_ARG_IN_STRING) | + P1275_INOUT(1, 1), service_name); + + if (err) + return 0; + return 1; +} + +void prom_sun4v_guest_soft_state(void) +{ + const char *svc = "SUNW,soft-state-supported"; + + if (!prom_service_exists(svc)) + return; + p1275_cmd(svc, P1275_INOUT(0, 0)); +} + /* Reset and reboot the machine with the command 'bcommand'. */ void prom_reboot(const char *bcommand) { diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c index 3ca9c610c11..af625147df6 100644 --- a/drivers/ata/libata-core.c +++ b/drivers/ata/libata-core.c @@ -3783,6 +3783,7 @@ static const struct ata_blacklist_entry ata_device_blacklist [] = { { "FUJITSU MHT2060BH", NULL, ATA_HORKAGE_NONCQ }, /* NCQ is broken */ { "Maxtor 6L250S0", "BANC1G10", ATA_HORKAGE_NONCQ }, + { "Maxtor 6B200M0", "BANC1B10", ATA_HORKAGE_NONCQ }, /* NCQ hard hangs device under heavier load, needs hard power cycle */ { "Maxtor 6B250S0", "BANC1B70", ATA_HORKAGE_NONCQ }, /* Blacklist entries taken from Silicon Image 3124/3132 diff --git a/drivers/ata/libata-scsi.c b/drivers/ata/libata-scsi.c index 242c43eef80..b3900cfbd88 100644 --- a/drivers/ata/libata-scsi.c +++ b/drivers/ata/libata-scsi.c @@ -1050,14 +1050,15 @@ static unsigned int ata_scsi_flush_xlat(struct ata_queued_cmd *qc) static void scsi_6_lba_len(const u8 *cdb, u64 *plba, u32 *plen) { u64 lba = 0; - u32 len = 0; + u32 len; VPRINTK("six-byte command\n"); + lba |= ((u64)(cdb[1] & 0x1f)) << 16; lba |= ((u64)cdb[2]) << 8; lba |= ((u64)cdb[3]); - len |= ((u32)cdb[4]); + len = cdb[4]; *plba = lba; *plen = len; diff --git a/drivers/ata/pata_artop.c b/drivers/ata/pata_artop.c index 7b4810c171b..03b6ddd2abd 100644 --- a/drivers/ata/pata_artop.c +++ b/drivers/ata/pata_artop.c @@ -97,7 +97,7 @@ static int artop6260_pre_reset(struct ata_port *ap, unsigned long deadline) * artop6260_cable_detect - identify cable type * @ap: Port * - * Identify the cable type for the ARTOp interface in question + * Identify the cable type for the ARTOP interface in question */ static int artop6260_cable_detect(struct ata_port *ap) diff --git a/drivers/ata/pata_hpt37x.c b/drivers/ata/pata_hpt37x.c index a54c17433ee..6446735a46e 100644 --- a/drivers/ata/pata_hpt37x.c +++ b/drivers/ata/pata_hpt37x.c @@ -26,7 +26,7 @@ #include <linux/libata.h> #define DRV_NAME "pata_hpt37x" -#define DRV_VERSION "0.6.5" +#define DRV_VERSION "0.6.6" struct hpt_clock { u8 xfer_speed; @@ -931,15 +931,6 @@ static int hpt37x_init_one(struct pci_dev *dev, const struct pci_device_id *id) .udma_mask = 0x7f, .port_ops = &hpt372_port_ops }; - /* HPT371, 372 and friends - UDMA100 at 50MHz clock */ - static const struct ata_port_info info_hpt372_50 = { - .sht = &hpt37x_sht, - .flags = ATA_FLAG_SLAVE_POSS|ATA_FLAG_SRST, - .pio_mask = 0x1f, - .mwdma_mask = 0x07, - .udma_mask = 0x3f, - .port_ops = &hpt372_port_ops - }; /* HPT374 - UDMA133 */ static const struct ata_port_info info_hpt374 = { .sht = &hpt37x_sht, @@ -1098,17 +1089,21 @@ static int hpt37x_init_one(struct pci_dev *dev, const struct pci_device_id *id) * use a 50MHz DPLL by choice */ unsigned int f_low, f_high; - int adjust; + int dpll, adjust; - clock_slot = 2; + /* Compute DPLL */ + dpll = 2; if (port->udma_mask & 0xE0) - clock_slot = 3; + dpll = 3; - f_low = (MHz[clock_slot] * chip_table->base) / 192; + f_low = (MHz[clock_slot] * 48) / MHz[dpll]; f_high = f_low + 2; + if (clock_slot > 1) + f_high += 2; /* Select the DPLL clock. */ pci_write_config_byte(dev, 0x5b, 0x21); + pci_write_config_dword(dev, 0x5C, (f_high << 16) | f_low); for(adjust = 0; adjust < 8; adjust++) { if (hpt37x_calibrate_dpll(dev)) @@ -1124,12 +1119,12 @@ static int hpt37x_init_one(struct pci_dev *dev, const struct pci_device_id *id) printk(KERN_WARNING "hpt37x: DPLL did not stabilize.\n"); return -ENODEV; } - if (clock_slot == 3) + if (dpll == 3) private_data = (void *)hpt37x_timings_66; else private_data = (void *)hpt37x_timings_50; - printk(KERN_INFO "hpt37x: Bus clock %dMHz, using DPLL.\n", MHz[clock_slot]); + printk(KERN_INFO "hpt37x: Bus clock %dMHz, using DPLL.\n", MHz[dpll]); } else { private_data = (void *)chip_table->clocks[clock_slot]; /* diff --git a/drivers/ata/pata_it821x.c b/drivers/ata/pata_it821x.c index ff9a6fd3665..b3456d7a592 100644 --- a/drivers/ata/pata_it821x.c +++ b/drivers/ata/pata_it821x.c @@ -1,5 +1,5 @@ /* - * ata-it821x.c - IT821x PATA for new ATA layer + * pata_it821x.c - IT821x PATA for new ATA layer * (C) 2005 Red Hat Inc * Alan Cox <alan@redhat.com> * @@ -65,7 +65,6 @@ * * TODO * - ATAPI and other speed filtering - * - Command filter in smart mode * - RAID configuration ioctls */ diff --git a/drivers/ata/pata_scc.c b/drivers/ata/pata_scc.c index aa138d211b0..61502bc7bf1 100644 --- a/drivers/ata/pata_scc.c +++ b/drivers/ata/pata_scc.c @@ -489,23 +489,26 @@ static unsigned int scc_devchk (struct ata_port *ap, * Note: Original code is ata_bus_post_reset(). */ -static void scc_bus_post_reset (struct ata_port *ap, unsigned int devmask) +static int scc_bus_post_reset(struct ata_port *ap, unsigned int devmask, + unsigned long deadline) { struct ata_ioports *ioaddr = &ap->ioaddr; unsigned int dev0 = devmask & (1 << 0); unsigned int dev1 = devmask & (1 << 1); - unsigned long timeout; + int rc; /* if device 0 was found in ata_devchk, wait for its * BSY bit to clear */ - if (dev0) - ata_busy_sleep(ap, ATA_TMOUT_BOOT_QUICK, ATA_TMOUT_BOOT); + if (dev0) { + rc = ata_wait_ready(ap, deadline); + if (rc && rc != -ENODEV) + return rc; + } /* if device 1 was found in ata_devchk, wait for * register access, then wait for BSY to clear */ - timeout = jiffies + ATA_TMOUT_BOOT; while (dev1) { u8 nsect, lbal; @@ -514,14 +517,15 @@ static void scc_bus_post_reset (struct ata_port *ap, unsigned int devmask) lbal = in_be32(ioaddr->lbal_addr); if ((nsect == 1) && (lbal == 1)) break; - if (time_after(jiffies, timeout)) { - dev1 = 0; - break; - } + if (time_after(jiffies, deadline)) + return -EBUSY; msleep(50); /* give drive a breather */ } - if (dev1) - ata_busy_sleep(ap, ATA_TMOUT_BOOT_QUICK, ATA_TMOUT_BOOT); + if (dev1) { + rc = ata_wait_ready(ap, deadline); + if (rc && rc != -ENODEV) + return rc; + } /* is all this really necessary? */ ap->ops->dev_select(ap, 0); @@ -529,6 +533,8 @@ static void scc_bus_post_reset (struct ata_port *ap, unsigned int devmask) ap->ops->dev_select(ap, 1); if (dev0) ap->ops->dev_select(ap, 0); + + return 0; } /** @@ -537,8 +543,8 @@ static void scc_bus_post_reset (struct ata_port *ap, unsigned int devmask) * Note: Original code is ata_bus_softreset(). */ -static unsigned int scc_bus_softreset (struct ata_port *ap, - unsigned int devmask) +static unsigned int scc_bus_softreset(struct ata_port *ap, unsigned int devmask, + unsigned long deadline) { struct ata_ioports *ioaddr = &ap->ioaddr; @@ -570,7 +576,7 @@ static unsigned int scc_bus_softreset (struct ata_port *ap, if (scc_check_status(ap) == 0xFF) return 0; - scc_bus_post_reset(ap, devmask); + scc_bus_post_reset(ap, devmask, deadline); return 0; } @@ -579,11 +585,13 @@ static unsigned int scc_bus_softreset (struct ata_port *ap, * scc_std_softreset - reset host port via ATA SRST * @ap: port to reset * @classes: resulting classes of attached devices + * @deadline: deadline jiffies for the operation * * Note: Original code is ata_std_softreset(). */ -static int scc_std_softreset (struct ata_port *ap, unsigned int *classes) +static int scc_std_softreset (struct ata_port *ap, unsigned int *classes, + unsigned long deadline) { unsigned int slave_possible = ap->flags & ATA_FLAG_SLAVE_POSS; unsigned int devmask = 0, err_mask; @@ -607,7 +615,7 @@ static int scc_std_softreset (struct ata_port *ap, unsigned int *classes) /* issue bus reset */ DPRINTK("about to softreset, devmask=%x\n", devmask); - err_mask = scc_bus_softreset(ap, devmask); + err_mask = scc_bus_softreset(ap, devmask, deadline); if (err_mask) { ata_port_printk(ap, KERN_ERR, "SRST failed (err_mask=0x%x)\n", err_mask); @@ -676,10 +684,11 @@ static void scc_bmdma_stop (struct ata_queued_cmd *qc) if (reg & INTSTS_BMSINT) { unsigned int classes; + unsigned long deadline = jiffies + ATA_TMOUT_BOOT; printk(KERN_WARNING "%s: Internal Bus Error\n", DRV_NAME); out_be32(bmid_base + SCC_DMA_INTST, INTSTS_BMSINT); /* TBD: SW reset */ - scc_std_softreset(ap, &classes); + scc_std_softreset(ap, &classes, deadline); continue; } @@ -862,9 +871,10 @@ static void scc_bmdma_freeze (struct ata_port *ap) /** * scc_pata_prereset - prepare for reset * @ap: ATA port to be reset + * @deadline: deadline jiffies for the operation */ -static int scc_pata_prereset (struct ata_port *ap, unsigned long deadline) +static int scc_pata_prereset(struct ata_port *ap, unsigned long deadline) { ap->cbl = ATA_CBL_PATA80; return ata_std_prereset(ap, deadline); diff --git a/drivers/ata/pata_via.c b/drivers/ata/pata_via.c index 708dfc0c716..a8462f1e890 100644 --- a/drivers/ata/pata_via.c +++ b/drivers/ata/pata_via.c @@ -621,10 +621,11 @@ static int via_reinit_one(struct pci_dev *pdev) #endif static const struct pci_device_id via[] = { - { PCI_VDEVICE(VIA, PCI_DEVICE_ID_VIA_82C576_1), }, - { PCI_VDEVICE(VIA, PCI_DEVICE_ID_VIA_82C586_1), }, - { PCI_VDEVICE(VIA, PCI_DEVICE_ID_VIA_6410), }, - { PCI_VDEVICE(VIA, PCI_DEVICE_ID_VIA_SATA_EIDE), }, + { PCI_VDEVICE(VIA, 0x0571), }, + { PCI_VDEVICE(VIA, 0x0581), }, + { PCI_VDEVICE(VIA, 0x1571), }, + { PCI_VDEVICE(VIA, 0x3164), }, + { PCI_VDEVICE(VIA, 0x5324), }, { }, }; diff --git a/drivers/ata/sata_mv.c b/drivers/ata/sata_mv.c index 705a020e159..c957e6e54ba 100644 --- a/drivers/ata/sata_mv.c +++ b/drivers/ata/sata_mv.c @@ -21,6 +21,50 @@ * */ +/* + sata_mv TODO list: + + 1) Needs a full errata audit for all chipsets. I implemented most + of the errata workarounds found in the Marvell vendor driver, but + I distinctly remember a couple workarounds (one related to PCI-X) + are still needed. + + 2) Convert to LibATA new EH. Required for hotplug, NCQ, and sane + probing/error handling in general. MUST HAVE. + + 3) Add hotplug support (easy, once new-EH support appears) + + 4) Add NCQ support (easy to intermediate, once new-EH support appears) + + 5) Investigate problems with PCI Message Signalled Interrupts (MSI). + + 6) Add port multiplier support (intermediate) + + 7) Test and verify 3.0 Gbps support + + 8) Develop a low-power-consumption strategy, and implement it. + + 9) [Experiment, low priority] See if ATAPI can be supported using + "unknown FIS" or "vendor-specific FIS" support, or something creative + like that. + + 10) [Experiment, low priority] Investigate interrupt coalescing. + Quite often, especially with PCI Message Signalled Interrupts (MSI), + the overhead reduced by interrupt mitigation is quite often not + worth the latency cost. + + 11) [Experiment, Marvell value added] Is it possible to use target + mode to cross-connect two Linux boxes with Marvell cards? If so, + creating LibATA target mode support would be very interesting. + + Target mode, for those without docs, is the ability to directly + connect two SATA controllers. + + 13) Verify that 7042 is fully supported. I only have a 6042. + +*/ + + #include <linux/kernel.h> #include <linux/module.h> #include <linux/pci.h> diff --git a/drivers/ata/sata_promise.c b/drivers/ata/sata_promise.c index 3a7d9b5332a..2b924a69b36 100644 --- a/drivers/ata/sata_promise.c +++ b/drivers/ata/sata_promise.c @@ -297,7 +297,7 @@ static const struct ata_port_info pdc_port_info[] = { /* board_2057x_pata */ { - .flags = PDC_COMMON_FLAGS | ATA_FLAG_SLAVE_POSS, + .flags = PDC_COMMON_FLAGS | ATA_FLAG_SLAVE_POSS | PDC_FLAG_GEN_II, .pio_mask = 0x1f, /* pio0-4 */ .mwdma_mask = 0x07, /* mwdma0-2 */ diff --git a/drivers/ata/sata_sis.c b/drivers/ata/sata_sis.c index 51d9251b089..221099d1d08 100644 --- a/drivers/ata/sata_sis.c +++ b/drivers/ata/sata_sis.c @@ -255,7 +255,7 @@ static int sis_init_one (struct pci_dev *pdev, const struct pci_device_id *ent) { static int printed_version; struct ata_port_info pi = sis_port_info; - const struct ata_port_info *ppi[] = { &pi, NULL }; + const struct ata_port_info *ppi[] = { &pi, &pi }; struct ata_host *host; u32 genctl, val; u8 pmr; diff --git a/drivers/ata/sata_via.c b/drivers/ata/sata_via.c index 01bba26d8d1..e8b90e7b42d 100644 --- a/drivers/ata/sata_via.c +++ b/drivers/ata/sata_via.c @@ -85,6 +85,9 @@ static const struct pci_device_id svia_pci_tbl[] = { { PCI_VDEVICE(VIA, 0x0591), vt6420 }, { PCI_VDEVICE(VIA, 0x3149), vt6420 }, { PCI_VDEVICE(VIA, 0x3249), vt6421 }, + { PCI_VDEVICE(VIA, 0x5287), vt6420 }, + { PCI_VDEVICE(VIA, 0x5372), vt6420 }, + { PCI_VDEVICE(VIA, 0x7372), vt6420 }, { } /* terminate list */ }; diff --git a/drivers/char/drm/Kconfig b/drivers/char/drm/Kconfig index ef833a1c27e..0b7ffa5191c 100644 --- a/drivers/char/drm/Kconfig +++ b/drivers/char/drm/Kconfig @@ -6,7 +6,7 @@ # config DRM tristate "Direct Rendering Manager (XFree86 4.1.0 and higher DRI support)" - depends on (AGP || AGP=n) && PCI + depends on (AGP || AGP=n) && PCI && !EMULATED_CMPXCHG help Kernel-level support for the Direct Rendering Infrastructure (DRI) introduced in XFree86 4.0. If you say Y here, you need to select diff --git a/drivers/char/random.c b/drivers/char/random.c index 46c1b97748b..0474cac4a84 100644 --- a/drivers/char/random.c +++ b/drivers/char/random.c @@ -760,7 +760,7 @@ static size_t account(struct entropy_store *r, size_t nbytes, int min, static void extract_buf(struct entropy_store *r, __u8 *out) { - int i, x; + int i; __u32 data[16], buf[5 + SHA_WORKSPACE_WORDS]; sha_init(buf); @@ -772,9 +772,11 @@ static void extract_buf(struct entropy_store *r, __u8 *out) * attempts to find previous ouputs), unless the hash * function can be inverted. */ - for (i = 0, x = 0; i < r->poolinfo->poolwords; i += 16, x+=2) { - sha_transform(buf, (__u8 *)r->pool+i, buf + 5); - add_entropy_words(r, &buf[x % 5], 1); + for (i = 0; i < r->poolinfo->poolwords; i += 16) { + /* hash blocks of 16 words = 512 bits */ + sha_transform(buf, (__u8 *)(r->pool + i), buf + 5); + /* feed back portion of the resulting hash */ + add_entropy_words(r, &buf[i % 5], 1); } /* @@ -782,7 +784,7 @@ static void extract_buf(struct entropy_store *r, __u8 *out) * portion of the pool while mixing, and hash one * final time. */ - __add_entropy_words(r, &buf[x % 5], 1, data); + __add_entropy_words(r, &buf[i % 5], 1, data); sha_transform(buf, (__u8 *)data, buf + 5); /* @@ -1018,37 +1020,44 @@ random_poll(struct file *file, poll_table * wait) return mask; } -static ssize_t -random_write(struct file * file, const char __user * buffer, - size_t count, loff_t *ppos) +static int +write_pool(struct entropy_store *r, const char __user *buffer, size_t count) { - int ret = 0; size_t bytes; __u32 buf[16]; const char __user *p = buffer; - size_t c = count; - while (c > 0) { - bytes = min(c, sizeof(buf)); + while (count > 0) { + bytes = min(count, sizeof(buf)); + if (copy_from_user(&buf, p, bytes)) + return -EFAULT; - bytes -= copy_from_user(&buf, p, bytes); - if (!bytes) { - ret = -EFAULT; - break; - } - c -= bytes; + count -= bytes; p += bytes; - add_entropy_words(&input_pool, buf, (bytes + 3) / 4); - } - if (p == buffer) { - return (ssize_t)ret; - } else { - struct inode *inode = file->f_path.dentry->d_inode; - inode->i_mtime = current_fs_time(inode->i_sb); - mark_inode_dirty(inode); - return (ssize_t)(p - buffer); + add_entropy_words(r, buf, (bytes + 3) / 4); } + + return 0; +} + +static ssize_t +random_write(struct file * file, const char __user * buffer, + size_t count, loff_t *ppos) +{ + size_t ret; + struct inode *inode = file->f_path.dentry->d_inode; + + ret = write_pool(&blocking_pool, buffer, count); + if (ret) + return ret; + ret = write_pool(&nonblocking_pool, buffer, count); + if (ret) + return ret; + + inode->i_mtime = current_fs_time(inode->i_sb); + mark_inode_dirty(inode); + return (ssize_t)count; } static int @@ -1087,8 +1096,8 @@ random_ioctl(struct inode * inode, struct file * file, return -EINVAL; if (get_user(size, p++)) return -EFAULT; - retval = random_write(file, (const char __user *) p, - size, &file->f_pos); + retval = write_pool(&input_pool, (const char __user *)p, + size); if (retval < 0) return retval; credit_entropy_store(&input_pool, ent_count); diff --git a/drivers/firewire/Kconfig b/drivers/firewire/Kconfig index 5932c72f9e4..396dade731f 100644 --- a/drivers/firewire/Kconfig +++ b/drivers/firewire/Kconfig @@ -18,7 +18,7 @@ config FIREWIRE your IEEE 1394 adapter. To compile this driver as a module, say M here: the module will be - called fw-core. + called firewire-core. This is the "JUJU" FireWire stack, an alternative implementation designed for robustness and simplicity. You can build either this @@ -34,11 +34,11 @@ config FIREWIRE_OHCI is the only chipset in use, so say Y here. To compile this driver as a module, say M here: The module will be - called fw-ohci. + called firewire-ohci. If you also build ohci1394 of the classic IEEE 1394 driver stack, - blacklist either ohci1394 or fw-ohci to let hotplug load the desired - driver. + blacklist either ohci1394 or firewire-ohci to let hotplug load the + desired driver. config FIREWIRE_SBP2 tristate "Support for storage devices (SBP-2 protocol driver)" @@ -50,12 +50,12 @@ config FIREWIRE_SBP2 like scanners. To compile this driver as a module, say M here: The module will be - called fw-sbp2. + called firewire-sbp2. You should also enable support for disks, CD-ROMs, etc. in the SCSI configuration section. If you also build sbp2 of the classic IEEE 1394 driver stack, - blacklist either sbp2 or fw-sbp2 to let hotplug load the desired - driver. + blacklist either sbp2 or firewire-sbp2 to let hotplug load the + desired driver. diff --git a/drivers/firewire/Makefile b/drivers/firewire/Makefile index fc7d59d4bce..a7c31e9039c 100644 --- a/drivers/firewire/Makefile +++ b/drivers/firewire/Makefile @@ -2,9 +2,11 @@ # Makefile for the Linux IEEE 1394 implementation # -fw-core-y += fw-card.o fw-topology.o fw-transaction.o fw-iso.o \ - fw-device.o fw-cdev.o +firewire-core-y += fw-card.o fw-topology.o fw-transaction.o fw-iso.o \ + fw-device.o fw-cdev.o +firewire-ohci-y += fw-ohci.o +firewire-sbp2-y += fw-sbp2.o -obj-$(CONFIG_FIREWIRE) += fw-core.o -obj-$(CONFIG_FIREWIRE_OHCI) += fw-ohci.o -obj-$(CONFIG_FIREWIRE_SBP2) += fw-sbp2.o +obj-$(CONFIG_FIREWIRE) += firewire-core.o +obj-$(CONFIG_FIREWIRE_OHCI) += firewire-ohci.o +obj-$(CONFIG_FIREWIRE_SBP2) += firewire-sbp2.o diff --git a/drivers/firewire/fw-cdev.c b/drivers/firewire/fw-cdev.c index 0fa5bd54c6a..3ab3585d360 100644 --- a/drivers/firewire/fw-cdev.c +++ b/drivers/firewire/fw-cdev.c @@ -365,7 +365,7 @@ complete_transaction(struct fw_card *card, int rcode, response->response.data, response->response.length); } -static ssize_t ioctl_send_request(struct client *client, void *buffer) +static int ioctl_send_request(struct client *client, void *buffer) { struct fw_device *device = client->device; struct fw_cdev_send_request *request = buffer; diff --git a/drivers/firewire/fw-ohci.c b/drivers/firewire/fw-ohci.c index c17342d3e6f..2e4cfa57126 100644 --- a/drivers/firewire/fw-ohci.c +++ b/drivers/firewire/fw-ohci.c @@ -268,7 +268,7 @@ static int ar_context_add_page(struct ar_context *ctx) dma_sync_single_for_device(dev, ab_bus, PAGE_SIZE, DMA_BIDIRECTIONAL); - ctx->last_buffer->descriptor.branch_address = ab_bus | 1; + ctx->last_buffer->descriptor.branch_address = cpu_to_le32(ab_bus | 1); ctx->last_buffer->next = ab; ctx->last_buffer = ab; @@ -417,7 +417,8 @@ ar_context_init(struct ar_context *ctx, struct fw_ohci *ohci, u32 regs) ctx->current_buffer = ab.next; ctx->pointer = ctx->current_buffer->data; - reg_write(ctx->ohci, COMMAND_PTR(ctx->regs), ab.descriptor.branch_address); + reg_write(ctx->ohci, COMMAND_PTR(ctx->regs), + le32_to_cpu(ab.descriptor.branch_address)); reg_write(ctx->ohci, CONTROL_SET(ctx->regs), CONTEXT_RUN); flush_writes(ctx->ohci); diff --git a/drivers/hwmon/Kconfig b/drivers/hwmon/Kconfig index 4d1cb5b855d..13eea47dceb 100644 --- a/drivers/hwmon/Kconfig +++ b/drivers/hwmon/Kconfig @@ -620,7 +620,7 @@ config SENSORS_HDAPS config SENSORS_APPLESMC tristate "Apple SMC (Motion sensor, light sensor, keyboard backlight)" - depends on HWMON && INPUT && X86 + depends on INPUT && X86 select NEW_LEDS select LEDS_CLASS default n diff --git a/drivers/hwmon/applesmc.c b/drivers/hwmon/applesmc.c index 366f4a1a2cb..fd1281f4220 100644 --- a/drivers/hwmon/applesmc.c +++ b/drivers/hwmon/applesmc.c @@ -1206,11 +1206,13 @@ static int __init applesmc_init(void) } ret = sysfs_create_file(&pdev->dev.kobj, &dev_attr_name.attr); + if (ret) + goto out_device; /* Create key enumeration sysfs files */ ret = sysfs_create_group(&pdev->dev.kobj, &key_enumeration_group); if (ret) - goto out_device; + goto out_name; /* create fan files */ count = applesmc_get_fan_count(); @@ -1310,6 +1312,8 @@ out_fan_1: sysfs_remove_group(&pdev->dev.kobj, &fan_attribute_groups[1]); out_key_enumeration: sysfs_remove_group(&pdev->dev.kobj, &key_enumeration_group); +out_name: + sysfs_remove_file(&pdev->dev.kobj, &dev_attr_name.attr); out_device: platform_device_unregister(pdev); out_driver: @@ -1335,6 +1339,7 @@ static void __exit applesmc_exit(void) sysfs_remove_group(&pdev->dev.kobj, &fan_attribute_groups[0]); sysfs_remove_group(&pdev->dev.kobj, &fan_attribute_groups[1]); sysfs_remove_group(&pdev->dev.kobj, &key_enumeration_group); + sysfs_remove_file(&pdev->dev.kobj, &dev_attr_name.attr); platform_device_unregister(pdev); platform_driver_unregister(&applesmc_driver); release_region(APPLESMC_DATA_PORT, APPLESMC_NR_PORTS); diff --git a/drivers/hwmon/coretemp.c b/drivers/hwmon/coretemp.c index 75e3911810a..0328382df8f 100644 --- a/drivers/hwmon/coretemp.c +++ b/drivers/hwmon/coretemp.c @@ -176,6 +176,22 @@ static int __devinit coretemp_probe(struct platform_device *pdev) goto exit_free; } + /* Check if we have problem with errata AE18 of Core processors: + Readings might stop update when processor visited too deep sleep, + fixed for stepping D0 (6EC). + */ + + if ((c->x86_model == 0xe) && (c->x86_mask < 0xc)) { + /* check for microcode update */ + rdmsr_on_cpu(data->id, MSR_IA32_UCODE_REV, &eax, &edx); + if (edx < 0x39) { + dev_err(&pdev->dev, + "Errata AE18 not fixed, update BIOS or " + "microcode of the CPU!\n"); + goto exit_free; + } + } + /* Some processors have Tjmax 85 following magic should detect it Intel won't disclose the information without signed NDA, but individuals cannot sign it. Catch(ed) 22. @@ -193,6 +209,19 @@ static int __devinit coretemp_probe(struct platform_device *pdev) } } + /* Intel says that above should not work for desktop Core2 processors, + but it seems to work. There is no other way how get the absolute + readings. Warn the user about this. First check if are desktop, + bit 50 of MSR_IA32_PLATFORM_ID should be 0. + */ + + rdmsr_safe_on_cpu(data->id, MSR_IA32_PLATFORM_ID, &eax, &edx); + + if ((c->x86_model == 0xf) && (!(edx & 0x00040000))) { + dev_warn(&pdev->dev, "Using undocumented features, absolute " + "temperature might be wrong!\n"); + } + platform_set_drvdata(pdev, data); if ((err = sysfs_create_group(&pdev->dev.kobj, &coretemp_group))) @@ -330,9 +359,6 @@ static int __init coretemp_init(void) int i, err = -ENODEV; struct pdev_entry *p, *n; - printk(KERN_NOTICE DRVNAME ": This driver uses undocumented features " - "of Core CPU. Temperature might be wrong!\n"); - /* quick check if we run Intel */ if (cpu_data[0].x86_vendor != X86_VENDOR_INTEL) goto exit; diff --git a/drivers/hwmon/ds1621.c b/drivers/hwmon/ds1621.c index c849c0c6ee9..d5ac422d73b 100644 --- a/drivers/hwmon/ds1621.c +++ b/drivers/hwmon/ds1621.c @@ -53,8 +53,8 @@ MODULE_PARM_DESC(polarity, "Output's polarity: 0 = active high, 1 = active low") /* The DS1621 registers */ #define DS1621_REG_TEMP 0xAA /* word, RO */ -#define DS1621_REG_TEMP_MIN 0xA1 /* word, RW */ -#define DS1621_REG_TEMP_MAX 0xA2 /* word, RW */ +#define DS1621_REG_TEMP_MIN 0xA2 /* word, RW */ +#define DS1621_REG_TEMP_MAX 0xA1 /* word, RW */ #define DS1621_REG_CONF 0xAC /* byte, RW */ #define DS1621_COM_START 0xEE /* no data */ #define DS1621_COM_STOP 0x22 /* no data */ @@ -328,9 +328,9 @@ static struct ds1621_data *ds1621_update_client(struct device *dev) /* reset alarms if necessary */ new_conf = data->conf; - if (data->temp < data->temp_min) + if (data->temp > data->temp_min) new_conf &= ~DS1621_ALARM_TEMP_LOW; - if (data->temp > data->temp_max) + if (data->temp < data->temp_max) new_conf &= ~DS1621_ALARM_TEMP_HIGH; if (data->conf != new_conf) ds1621_write_value(client, DS1621_REG_CONF, diff --git a/drivers/hwmon/hwmon-vid.c b/drivers/hwmon/hwmon-vid.c index 5aab23b93e2..f17e771e42f 100644 --- a/drivers/hwmon/hwmon-vid.c +++ b/drivers/hwmon/hwmon-vid.c @@ -132,7 +132,9 @@ int vid_from_reg(int val, u8 vrm) val &= 0x7f; return(val > 0x77 ? 0 : (1500000 - (val * 12500) + 500) / 1000); default: /* report 0 for unknown */ - printk(KERN_INFO "hwmon-vid: requested unknown VRM version\n"); + if (vrm) + printk(KERN_WARNING "hwmon-vid: Requested unsupported " + "VRM version (%u)\n", (unsigned int)vrm); return 0; } } diff --git a/drivers/hwmon/w83627hf.c b/drivers/hwmon/w83627hf.c index a5b774b07cb..12cb40a975d 100644 --- a/drivers/hwmon/w83627hf.c +++ b/drivers/hwmon/w83627hf.c @@ -965,8 +965,10 @@ static int __init w83627hf_find(int sioaddr, unsigned short *addr, case W687THF_DEVID: sio_data->type = w83687thf; break; + case 0xff: /* No device at all */ + goto exit; default: - pr_debug(DRVNAME ": Unsupported chip (DEVID=0x%x)\n", val); + pr_debug(DRVNAME ": Unsupported chip (DEVID=0x%02x)\n", val); goto exit; } diff --git a/drivers/i2c/busses/i2c-pxa.c b/drivers/i2c/busses/i2c-pxa.c index 8a0a99b9364..28e7b91a455 100644 --- a/drivers/i2c/busses/i2c-pxa.c +++ b/drivers/i2c/busses/i2c-pxa.c @@ -837,20 +837,10 @@ static const struct i2c_algorithm i2c_pxa_algorithm = { .functionality = i2c_pxa_functionality, }; -static struct pxa_i2c i2c_pxa = { - .lock = __SPIN_LOCK_UNLOCKED(i2c_pxa.lock), - .adap = { - .owner = THIS_MODULE, - .algo = &i2c_pxa_algorithm, - .name = "pxa2xx-i2c.0", - .retries = 5, - }, -}; - #define res_len(r) ((r)->end - (r)->start + 1) static int i2c_pxa_probe(struct platform_device *dev) { - struct pxa_i2c *i2c = &i2c_pxa; + struct pxa_i2c *i2c; struct resource *res; struct i2c_pxa_platform_data *plat = dev->dev.platform_data; int ret; @@ -864,15 +854,20 @@ static int i2c_pxa_probe(struct platform_device *dev) if (!request_mem_region(res->start, res_len(res), res->name)) return -ENOMEM; - i2c = kmalloc(sizeof(struct pxa_i2c), GFP_KERNEL); + i2c = kzalloc(sizeof(struct pxa_i2c), GFP_KERNEL); if (!i2c) { ret = -ENOMEM; goto emalloc; } - memcpy(i2c, &i2c_pxa, sizeof(struct pxa_i2c)); + i2c->adap.owner = THIS_MODULE; + i2c->adap.algo = &i2c_pxa_algorithm; + i2c->adap.retries = 5; + + spin_lock_init(&i2c->lock); init_waitqueue_head(&i2c->wait); - i2c->adap.name[strlen(i2c->adap.name) - 1] = '0' + dev->id % 10; + + sprintf(i2c->adap.name, "pxa_i2c-i2c.%u", dev->id); i2c->reg_base = ioremap(res->start, res_len(res)); if (!i2c->reg_base) { diff --git a/drivers/ieee1394/eth1394.c b/drivers/ieee1394/eth1394.c index 2296d43a241..5f026b5d785 100644 --- a/drivers/ieee1394/eth1394.c +++ b/drivers/ieee1394/eth1394.c @@ -47,6 +47,7 @@ #include <linux/types.h> #include <linux/delay.h> #include <linux/init.h> +#include <linux/workqueue.h> #include <linux/netdevice.h> #include <linux/inetdevice.h> @@ -235,6 +236,9 @@ static int ether1394_open(struct net_device *dev) /* This is called after an "ifdown" */ static int ether1394_stop(struct net_device *dev) { + /* flush priv->wake */ + flush_scheduled_work(); + netif_stop_queue(dev); return 0; } @@ -531,6 +535,37 @@ static void ether1394_init_dev(struct net_device *dev) } /* + * Wake the queue up after commonly encountered transmit failure conditions are + * hopefully over. Currently only tlabel exhaustion is accounted for. + */ +static void ether1394_wake_queue(struct work_struct *work) +{ + struct eth1394_priv *priv; + struct hpsb_packet *packet; + + priv = container_of(work, struct eth1394_priv, wake); + packet = hpsb_alloc_packet(0); + + /* This is really bad, but unjam the queue anyway. */ + if (!packet) + goto out; + + packet->host = priv->host; + packet->node_id = priv->wake_node; + /* + * A transaction label is all we really want. If we get one, it almost + * always means we can get a lot more because the ieee1394 core recycled + * a whole batch of tlabels, at last. + */ + if (hpsb_get_tlabel(packet) == 0) + hpsb_free_tlabel(packet); + + hpsb_free_packet(packet); +out: + netif_wake_queue(priv->wake_dev); +} + +/* * This function is called every time a card is found. It is generally called * when the module is installed. This is where we add all of our ethernet * devices. One for each host. @@ -564,16 +599,17 @@ static void ether1394_add_host(struct hpsb_host *host) } SET_MODULE_OWNER(dev); -#if 0 - /* FIXME - Is this the correct parent device anyway? */ - SET_NETDEV_DEV(dev, &host->device); -#endif + + /* This used to be &host->device in Linux 2.6.20 and before. */ + SET_NETDEV_DEV(dev, host->device.parent); priv = netdev_priv(dev); INIT_LIST_HEAD(&priv->ip_node_list); spin_lock_init(&priv->lock); priv->host = host; priv->local_fifo = fifo_addr; + INIT_WORK(&priv->wake, ether1394_wake_queue); + priv->wake_dev = dev; hi = hpsb_create_hostinfo(ð1394_highlevel, host, sizeof(*hi)); if (hi == NULL) { @@ -1390,22 +1426,17 @@ static int ether1394_prep_write_packet(struct hpsb_packet *p, u64 addr, void *data, int tx_len) { p->node_id = node; - p->data = NULL; - p->tcode = TCODE_WRITEB; - p->header[1] = host->node_id << 16 | addr >> 32; - p->header[2] = addr & 0xffffffff; + if (hpsb_get_tlabel(p)) + return -EAGAIN; + p->tcode = TCODE_WRITEB; p->header_size = 16; p->expect_response = 1; - - if (hpsb_get_tlabel(p)) { - ETH1394_PRINT_G(KERN_ERR, "Out of tlabels\n"); - return -1; - } p->header[0] = p->node_id << 16 | p->tlabel << 10 | 1 << 8 | TCODE_WRITEB << 4; - + p->header[1] = host->node_id << 16 | addr >> 32; + p->header[2] = addr & 0xffffffff; p->header[3] = tx_len << 16; p->data_size = (tx_len + 3) & ~3; p->data = data; @@ -1451,7 +1482,7 @@ static int ether1394_send_packet(struct packet_task *ptask, unsigned int tx_len) packet = ether1394_alloc_common_packet(priv->host); if (!packet) - return -1; + return -ENOMEM; if (ptask->tx_type == ETH1394_GASP) { int length = tx_len + 2 * sizeof(quadlet_t); @@ -1462,7 +1493,7 @@ static int ether1394_send_packet(struct packet_task *ptask, unsigned int tx_len) ptask->addr, ptask->skb->data, tx_len)) { hpsb_free_packet(packet); - return -1; + return -EAGAIN; } ptask->packet = packet; @@ -1471,7 +1502,7 @@ static int ether1394_send_packet(struct packet_task *ptask, unsigned int tx_len) if (hpsb_send_packet(packet) < 0) { ether1394_free_packet(packet); - return -1; + return -EIO; } return 0; @@ -1514,13 +1545,18 @@ static void ether1394_complete_cb(void *__ptask) ptask->outstanding_pkts--; if (ptask->outstanding_pkts > 0 && !fail) { - int tx_len; + int tx_len, err; /* Add the encapsulation header to the fragment */ tx_len = ether1394_encapsulate(ptask->skb, ptask->max_payload, &ptask->hdr); - if (ether1394_send_packet(ptask, tx_len)) + err = ether1394_send_packet(ptask, tx_len); + if (err) { + if (err == -EAGAIN) + ETH1394_PRINT_G(KERN_ERR, "Out of tlabels\n"); + ether1394_dg_complete(ptask, 1); + } } else { ether1394_dg_complete(ptask, fail); } @@ -1633,10 +1669,18 @@ static int ether1394_tx(struct sk_buff *skb, struct net_device *dev) /* Add the encapsulation header to the fragment */ tx_len = ether1394_encapsulate(skb, max_payload, &ptask->hdr); dev->trans_start = jiffies; - if (ether1394_send_packet(ptask, tx_len)) - goto fail; + if (ether1394_send_packet(ptask, tx_len)) { + if (dest_node == (LOCAL_BUS | ALL_NODES)) + goto fail; + + /* Most failures of ether1394_send_packet are recoverable. */ + netif_stop_queue(dev); + priv->wake_node = dest_node; + schedule_work(&priv->wake); + kmem_cache_free(packet_task_cache, ptask); + return NETDEV_TX_BUSY; + } - netif_wake_queue(dev); return NETDEV_TX_OK; fail: if (ptask) @@ -1650,9 +1694,6 @@ fail: priv->stats.tx_errors++; spin_unlock_irqrestore(&priv->lock, flags); - if (netif_queue_stopped(dev)) - netif_wake_queue(dev); - /* * FIXME: According to a patch from 2003-02-26, "returning non-zero * causes serious problems" here, allegedly. Before that patch, diff --git a/drivers/ieee1394/eth1394.h b/drivers/ieee1394/eth1394.h index a3439ee7cb4..4f3e2dd46f0 100644 --- a/drivers/ieee1394/eth1394.h +++ b/drivers/ieee1394/eth1394.h @@ -66,6 +66,10 @@ struct eth1394_priv { int bc_dgl; /* Outgoing broadcast datagram label */ struct list_head ip_node_list; /* List of IP capable nodes */ struct unit_directory *ud_list[ALL_NODES]; /* Cached unit dir list */ + + struct work_struct wake; /* Wake up after xmit failure */ + struct net_device *wake_dev; /* Stupid backlink for .wake */ + nodeid_t wake_node; /* Destination of failed xmit */ }; diff --git a/drivers/ieee1394/raw1394.c b/drivers/ieee1394/raw1394.c index d382500f421..f1d05eeb9f5 100644 --- a/drivers/ieee1394/raw1394.c +++ b/drivers/ieee1394/raw1394.c @@ -936,6 +936,7 @@ static int handle_async_send(struct file_info *fi, struct pending_request *req) struct hpsb_packet *packet; int header_length = req->req.misc & 0xffff; int expect_response = req->req.misc >> 16; + size_t data_size; if (header_length > req->req.length || header_length < 12 || header_length > FIELD_SIZEOF(struct hpsb_packet, header)) { @@ -945,7 +946,8 @@ static int handle_async_send(struct file_info *fi, struct pending_request *req) return sizeof(struct raw1394_request); } - packet = hpsb_alloc_packet(req->req.length - header_length); + data_size = req->req.length - header_length; + packet = hpsb_alloc_packet(data_size); req->packet = packet; if (!packet) return -ENOMEM; @@ -960,7 +962,7 @@ static int handle_async_send(struct file_info *fi, struct pending_request *req) if (copy_from_user (packet->data, int2ptr(req->req.sendb) + header_length, - packet->data_size)) { + data_size)) { req->req.error = RAW1394_ERROR_MEMFAULT; req->req.length = 0; queue_complete_req(req); @@ -974,7 +976,7 @@ static int handle_async_send(struct file_info *fi, struct pending_request *req) packet->host = fi->host; packet->expect_response = expect_response; packet->header_size = header_length; - packet->data_size = req->req.length - header_length; + packet->data_size = data_size; req->req.length = 0; hpsb_set_packet_complete_task(packet, diff --git a/drivers/ieee1394/sbp2.c b/drivers/ieee1394/sbp2.c index 4cb6fa2bcfb..875eadd5e8f 100644 --- a/drivers/ieee1394/sbp2.c +++ b/drivers/ieee1394/sbp2.c @@ -70,6 +70,7 @@ #include <linux/stringify.h> #include <linux/types.h> #include <linux/wait.h> +#include <linux/workqueue.h> #include <asm/byteorder.h> #include <asm/errno.h> diff --git a/drivers/infiniband/core/cm.c b/drivers/infiniband/core/cm.c index e840434a96d..40c004a2697 100644 --- a/drivers/infiniband/core/cm.c +++ b/drivers/infiniband/core/cm.c @@ -1297,26 +1297,29 @@ static struct cm_id_private * cm_match_req(struct cm_work *work, req_msg = (struct cm_req_msg *)work->mad_recv_wc->recv_buf.mad; - /* Check for duplicate REQ and stale connections. */ + /* Check for possible duplicate REQ. */ spin_lock_irqsave(&cm.lock, flags); timewait_info = cm_insert_remote_id(cm_id_priv->timewait_info); - if (!timewait_info) - timewait_info = cm_insert_remote_qpn(cm_id_priv->timewait_info); - if (timewait_info) { cur_cm_id_priv = cm_get_id(timewait_info->work.local_id, timewait_info->work.remote_id); - cm_cleanup_timewait(cm_id_priv->timewait_info); spin_unlock_irqrestore(&cm.lock, flags); if (cur_cm_id_priv) { cm_dup_req_handler(work, cur_cm_id_priv); cm_deref_id(cur_cm_id_priv); - } else - cm_issue_rej(work->port, work->mad_recv_wc, - IB_CM_REJ_STALE_CONN, CM_MSG_RESPONSE_REQ, - NULL, 0); - listen_cm_id_priv = NULL; - goto out; + } + return NULL; + } + + /* Check for stale connections. */ + timewait_info = cm_insert_remote_qpn(cm_id_priv->timewait_info); + if (timewait_info) { + cm_cleanup_timewait(cm_id_priv->timewait_info); + spin_unlock_irqrestore(&cm.lock, flags); + cm_issue_rej(work->port, work->mad_recv_wc, + IB_CM_REJ_STALE_CONN, CM_MSG_RESPONSE_REQ, + NULL, 0); + return NULL; } /* Find matching listen request. */ diff --git a/drivers/infiniband/hw/ehca/hcp_if.c b/drivers/infiniband/hw/ehca/hcp_if.c index 7f0beec74f7..5766ae3a202 100644 --- a/drivers/infiniband/hw/ehca/hcp_if.c +++ b/drivers/infiniband/hw/ehca/hcp_if.c @@ -331,7 +331,7 @@ u64 hipz_h_alloc_resource_qp(const struct ipz_adapter_handle adapter_handle, 0); qp->ipz_qp_handle.handle = outs[0]; qp->real_qp_num = (u32)outs[1]; - parms->act_nr_send_sges = + parms->act_nr_send_wqes = (u16)EHCA_BMASK_GET(H_ALL_RES_QP_ACT_OUTST_SEND_WR, outs[2]); parms->act_nr_recv_wqes = (u16)EHCA_BMASK_GET(H_ALL_RES_QP_ACT_OUTST_RECV_WR, outs[2]); diff --git a/drivers/infiniband/hw/mlx4/qp.c b/drivers/infiniband/hw/mlx4/qp.c index a824bc5f79f..dc137dec230 100644 --- a/drivers/infiniband/hw/mlx4/qp.c +++ b/drivers/infiniband/hw/mlx4/qp.c @@ -270,9 +270,7 @@ static int create_qp_common(struct mlx4_ib_dev *dev, struct ib_pd *pd, struct ib_qp_init_attr *init_attr, struct ib_udata *udata, int sqpn, struct mlx4_ib_qp *qp) { - struct mlx4_wqe_ctrl_seg *ctrl; int err; - int i; mutex_init(&qp->mutex); spin_lock_init(&qp->sq.lock); @@ -319,20 +317,24 @@ static int create_qp_common(struct mlx4_ib_dev *dev, struct ib_pd *pd, if (err) goto err_mtt; - err = mlx4_ib_db_map_user(to_mucontext(pd->uobject->context), - ucmd.db_addr, &qp->db); - if (err) - goto err_mtt; + if (!init_attr->srq) { + err = mlx4_ib_db_map_user(to_mucontext(pd->uobject->context), + ucmd.db_addr, &qp->db); + if (err) + goto err_mtt; + } } else { err = set_kernel_sq_size(dev, &init_attr->cap, init_attr->qp_type, qp); if (err) goto err; - err = mlx4_ib_db_alloc(dev, &qp->db, 0); - if (err) - goto err; + if (!init_attr->srq) { + err = mlx4_ib_db_alloc(dev, &qp->db, 0); + if (err) + goto err; - *qp->db.db = 0; + *qp->db.db = 0; + } if (mlx4_buf_alloc(dev->dev, qp->buf_size, PAGE_SIZE * 2, &qp->buf)) { err = -ENOMEM; @@ -348,11 +350,6 @@ static int create_qp_common(struct mlx4_ib_dev *dev, struct ib_pd *pd, if (err) goto err_mtt; - for (i = 0; i < qp->sq.max; ++i) { - ctrl = get_send_wqe(qp, i); - ctrl->owner_opcode = cpu_to_be32(1 << 31); - } - qp->sq.wrid = kmalloc(qp->sq.max * sizeof (u64), GFP_KERNEL); qp->rq.wrid = kmalloc(qp->rq.max * sizeof (u64), GFP_KERNEL); @@ -386,7 +383,7 @@ static int create_qp_common(struct mlx4_ib_dev *dev, struct ib_pd *pd, return 0; err_wrid: - if (pd->uobject) + if (pd->uobject && !init_attr->srq) mlx4_ib_db_unmap_user(to_mucontext(pd->uobject->context), &qp->db); else { kfree(qp->sq.wrid); @@ -403,7 +400,7 @@ err_buf: mlx4_buf_free(dev->dev, qp->buf_size, &qp->buf); err_db: - if (!pd->uobject) + if (!pd->uobject && !init_attr->srq) mlx4_ib_db_free(dev, &qp->db); err: @@ -481,14 +478,16 @@ static void destroy_qp_common(struct mlx4_ib_dev *dev, struct mlx4_ib_qp *qp, mlx4_mtt_cleanup(dev->dev, &qp->mtt); if (is_user) { - mlx4_ib_db_unmap_user(to_mucontext(qp->ibqp.uobject->context), - &qp->db); + if (!qp->ibqp.srq) + mlx4_ib_db_unmap_user(to_mucontext(qp->ibqp.uobject->context), + &qp->db); ib_umem_release(qp->umem); } else { kfree(qp->sq.wrid); kfree(qp->rq.wrid); mlx4_buf_free(dev->dev, qp->buf_size, &qp->buf); - mlx4_ib_db_free(dev, &qp->db); + if (!qp->ibqp.srq) + mlx4_ib_db_free(dev, &qp->db); } } @@ -852,7 +851,7 @@ static int __mlx4_ib_modify_qp(struct ib_qp *ibqp, if (ibqp->srq) context->srqn = cpu_to_be32(1 << 24 | to_msrq(ibqp->srq)->msrq.srqn); - if (cur_state == IB_QPS_RESET && new_state == IB_QPS_INIT) + if (!ibqp->srq && cur_state == IB_QPS_RESET && new_state == IB_QPS_INIT) context->db_rec_addr = cpu_to_be64(qp->db.dma); if (cur_state == IB_QPS_INIT && @@ -872,6 +871,21 @@ static int __mlx4_ib_modify_qp(struct ib_qp *ibqp, else sqd_event = 0; + /* + * Before passing a kernel QP to the HW, make sure that the + * ownership bits of the send queue are set so that the + * hardware doesn't start processing stale work requests. + */ + if (!ibqp->uobject && cur_state == IB_QPS_RESET && new_state == IB_QPS_INIT) { + struct mlx4_wqe_ctrl_seg *ctrl; + int i; + + for (i = 0; i < qp->sq.max; ++i) { + ctrl = get_send_wqe(qp, i); + ctrl->owner_opcode = cpu_to_be32(1 << 31); + } + } + err = mlx4_qp_modify(dev->dev, &qp->mtt, to_mlx4_state(cur_state), to_mlx4_state(new_state), context, optpar, sqd_event, &qp->mqp); @@ -919,7 +933,8 @@ static int __mlx4_ib_modify_qp(struct ib_qp *ibqp, qp->rq.tail = 0; qp->sq.head = 0; qp->sq.tail = 0; - *qp->db.db = 0; + if (!ibqp->srq) + *qp->db.db = 0; } out: diff --git a/drivers/infiniband/hw/mthca/mthca_qp.c b/drivers/infiniband/hw/mthca/mthca_qp.c index 027664979fe..eef415b12b2 100644 --- a/drivers/infiniband/hw/mthca/mthca_qp.c +++ b/drivers/infiniband/hw/mthca/mthca_qp.c @@ -2284,10 +2284,10 @@ void mthca_free_err_wqe(struct mthca_dev *dev, struct mthca_qp *qp, int is_send, struct mthca_next_seg *next; /* - * For SRQs, all WQEs generate a CQE, so we're always at the - * end of the doorbell chain. + * For SRQs, all receive WQEs generate a CQE, so we're always + * at the end of the doorbell chain. */ - if (qp->ibqp.srq) { + if (qp->ibqp.srq && !is_send) { *new_wqe = 0; return; } diff --git a/drivers/infiniband/ulp/ipoib/ipoib.h b/drivers/infiniband/ulp/ipoib/ipoib.h index a0b3782c762..285c143115c 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib.h +++ b/drivers/infiniband/ulp/ipoib/ipoib.h @@ -156,7 +156,7 @@ struct ipoib_cm_data { * - and then invoke a Destroy QP or Reset QP. * * We use the second option and wait for a completion on the - * rx_drain_qp before destroying QPs attached to our SRQ. + * same CQ before destroying QPs attached to our SRQ. */ enum ipoib_cm_state { @@ -199,7 +199,6 @@ struct ipoib_cm_dev_priv { struct ib_srq *srq; struct ipoib_cm_rx_buf *srq_ring; struct ib_cm_id *id; - struct ib_qp *rx_drain_qp; /* generates WR described in 10.3.1 */ struct list_head passive_ids; /* state: LIVE */ struct list_head rx_error_list; /* state: ERROR */ struct list_head rx_flush_list; /* state: FLUSH, drain not started */ @@ -429,6 +428,7 @@ int ipoib_vlan_delete(struct net_device *pdev, unsigned short pkey); void ipoib_pkey_poll(struct work_struct *work); int ipoib_pkey_dev_delay_open(struct net_device *dev); +void ipoib_drain_cq(struct net_device *dev); #ifdef CONFIG_INFINIBAND_IPOIB_CM diff --git a/drivers/infiniband/ulp/ipoib/ipoib_cm.c b/drivers/infiniband/ulp/ipoib/ipoib_cm.c index ffec794b791..076a0bbb63d 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_cm.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_cm.c @@ -69,8 +69,9 @@ static struct ib_qp_attr ipoib_cm_err_attr = { #define IPOIB_CM_RX_DRAIN_WRID 0x7fffffff -static struct ib_recv_wr ipoib_cm_rx_drain_wr = { - .wr_id = IPOIB_CM_RX_DRAIN_WRID +static struct ib_send_wr ipoib_cm_rx_drain_wr = { + .wr_id = IPOIB_CM_RX_DRAIN_WRID, + .opcode = IB_WR_SEND, }; static int ipoib_cm_tx_handler(struct ib_cm_id *cm_id, @@ -163,16 +164,22 @@ partial_error: static void ipoib_cm_start_rx_drain(struct ipoib_dev_priv* priv) { - struct ib_recv_wr *bad_wr; + struct ib_send_wr *bad_wr; + struct ipoib_cm_rx *p; - /* rx_drain_qp send queue depth is 1, so + /* We only reserved 1 extra slot in CQ for drain WRs, so * make sure we have at most 1 outstanding WR. */ if (list_empty(&priv->cm.rx_flush_list) || !list_empty(&priv->cm.rx_drain_list)) return; - if (ib_post_recv(priv->cm.rx_drain_qp, &ipoib_cm_rx_drain_wr, &bad_wr)) - ipoib_warn(priv, "failed to post rx_drain wr\n"); + /* + * QPs on flush list are error state. This way, a "flush + * error" WC will be immediately generated for each WR we post. + */ + p = list_entry(priv->cm.rx_flush_list.next, typeof(*p), list); + if (ib_post_send(p->qp, &ipoib_cm_rx_drain_wr, &bad_wr)) + ipoib_warn(priv, "failed to post drain wr\n"); list_splice_init(&priv->cm.rx_flush_list, &priv->cm.rx_drain_list); } @@ -199,10 +206,10 @@ static struct ib_qp *ipoib_cm_create_rx_qp(struct net_device *dev, struct ipoib_dev_priv *priv = netdev_priv(dev); struct ib_qp_init_attr attr = { .event_handler = ipoib_cm_rx_event_handler, - .send_cq = priv->cq, /* does not matter, we never send anything */ + .send_cq = priv->cq, /* For drain WR */ .recv_cq = priv->cq, .srq = priv->cm.srq, - .cap.max_send_wr = 1, /* FIXME: 0 Seems not to work */ + .cap.max_send_wr = 1, /* For drain WR */ .cap.max_send_sge = 1, /* FIXME: 0 Seems not to work */ .sq_sig_type = IB_SIGNAL_ALL_WR, .qp_type = IB_QPT_RC, @@ -242,6 +249,27 @@ static int ipoib_cm_modify_rx_qp(struct net_device *dev, ipoib_warn(priv, "failed to modify QP to RTR: %d\n", ret); return ret; } + + /* + * Current Mellanox HCA firmware won't generate completions + * with error for drain WRs unless the QP has been moved to + * RTS first. This work-around leaves a window where a QP has + * moved to error asynchronously, but this will eventually get + * fixed in firmware, so let's not error out if modify QP + * fails. + */ + qp_attr.qp_state = IB_QPS_RTS; + ret = ib_cm_init_qp_attr(cm_id, &qp_attr, &qp_attr_mask); + if (ret) { + ipoib_warn(priv, "failed to init QP attr for RTS: %d\n", ret); + return 0; + } + ret = ib_modify_qp(qp, &qp_attr, qp_attr_mask); + if (ret) { + ipoib_warn(priv, "failed to modify QP to RTS: %d\n", ret); + return 0; + } + return 0; } @@ -623,38 +651,11 @@ static void ipoib_cm_tx_completion(struct ib_cq *cq, void *tx_ptr) int ipoib_cm_dev_open(struct net_device *dev) { struct ipoib_dev_priv *priv = netdev_priv(dev); - struct ib_qp_init_attr qp_init_attr = { - .send_cq = priv->cq, /* does not matter, we never send anything */ - .recv_cq = priv->cq, - .cap.max_send_wr = 1, /* FIXME: 0 Seems not to work */ - .cap.max_send_sge = 1, /* FIXME: 0 Seems not to work */ - .cap.max_recv_wr = 1, - .cap.max_recv_sge = 1, /* FIXME: 0 Seems not to work */ - .sq_sig_type = IB_SIGNAL_ALL_WR, - .qp_type = IB_QPT_UC, - }; int ret; if (!IPOIB_CM_SUPPORTED(dev->dev_addr)) return 0; - priv->cm.rx_drain_qp = ib_create_qp(priv->pd, &qp_init_attr); - if (IS_ERR(priv->cm.rx_drain_qp)) { - printk(KERN_WARNING "%s: failed to create CM ID\n", priv->ca->name); - ret = PTR_ERR(priv->cm.rx_drain_qp); - return ret; - } - - /* - * We put the QP in error state directly. This way, a "flush - * error" WC will be immediately generated for each WR we post. - */ - ret = ib_modify_qp(priv->cm.rx_drain_qp, &ipoib_cm_err_attr, IB_QP_STATE); - if (ret) { - ipoib_warn(priv, "failed to modify drain QP to error: %d\n", ret); - goto err_qp; - } - priv->cm.id = ib_create_cm_id(priv->ca, ipoib_cm_rx_handler, dev); if (IS_ERR(priv->cm.id)) { printk(KERN_WARNING "%s: failed to create CM ID\n", priv->ca->name); @@ -676,8 +677,6 @@ err_listen: ib_destroy_cm_id(priv->cm.id); err_cm: priv->cm.id = NULL; -err_qp: - ib_destroy_qp(priv->cm.rx_drain_qp); return ret; } @@ -713,7 +712,7 @@ void ipoib_cm_dev_stop(struct net_device *dev) while (!list_empty(&priv->cm.rx_error_list) || !list_empty(&priv->cm.rx_flush_list) || !list_empty(&priv->cm.rx_drain_list)) { - if (!time_after(jiffies, begin + 5 * HZ)) { + if (time_after(jiffies, begin + 5 * HZ)) { ipoib_warn(priv, "RX drain timing out\n"); /* @@ -726,6 +725,7 @@ void ipoib_cm_dev_stop(struct net_device *dev) } spin_unlock_irq(&priv->lock); msleep(1); + ipoib_drain_cq(dev); spin_lock_irq(&priv->lock); } @@ -739,7 +739,6 @@ void ipoib_cm_dev_stop(struct net_device *dev) kfree(p); } - ib_destroy_qp(priv->cm.rx_drain_qp); cancel_delayed_work(&priv->cm.stale_task); } diff --git a/drivers/infiniband/ulp/ipoib/ipoib_ib.c b/drivers/infiniband/ulp/ipoib/ipoib_ib.c index c1aad06eb4e..8404f05b2b6 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_ib.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_ib.c @@ -550,13 +550,30 @@ static int recvs_pending(struct net_device *dev) return pending; } +void ipoib_drain_cq(struct net_device *dev) +{ + struct ipoib_dev_priv *priv = netdev_priv(dev); + int i, n; + do { + n = ib_poll_cq(priv->cq, IPOIB_NUM_WC, priv->ibwc); + for (i = 0; i < n; ++i) { + if (priv->ibwc[i].wr_id & IPOIB_CM_OP_SRQ) + ipoib_cm_handle_rx_wc(dev, priv->ibwc + i); + else if (priv->ibwc[i].wr_id & IPOIB_OP_RECV) + ipoib_ib_handle_rx_wc(dev, priv->ibwc + i); + else + ipoib_ib_handle_tx_wc(dev, priv->ibwc + i); + } + } while (n == IPOIB_NUM_WC); +} + int ipoib_ib_dev_stop(struct net_device *dev, int flush) { struct ipoib_dev_priv *priv = netdev_priv(dev); struct ib_qp_attr qp_attr; unsigned long begin; struct ipoib_tx_buf *tx_req; - int i, n; + int i; clear_bit(IPOIB_FLAG_INITIALIZED, &priv->flags); netif_poll_disable(dev); @@ -611,17 +628,7 @@ int ipoib_ib_dev_stop(struct net_device *dev, int flush) goto timeout; } - do { - n = ib_poll_cq(priv->cq, IPOIB_NUM_WC, priv->ibwc); - for (i = 0; i < n; ++i) { - if (priv->ibwc[i].wr_id & IPOIB_CM_OP_SRQ) - ipoib_cm_handle_rx_wc(dev, priv->ibwc + i); - else if (priv->ibwc[i].wr_id & IPOIB_OP_RECV) - ipoib_ib_handle_rx_wc(dev, priv->ibwc + i); - else - ipoib_ib_handle_tx_wc(dev, priv->ibwc + i); - } - } while (n == IPOIB_NUM_WC); + ipoib_drain_cq(dev); msleep(1); } diff --git a/drivers/input/serio/sa1111ps2.c b/drivers/input/serio/sa1111ps2.c index 559508795af..d31ece8f68e 100644 --- a/drivers/input/serio/sa1111ps2.c +++ b/drivers/input/serio/sa1111ps2.c @@ -170,7 +170,7 @@ static void ps2_close(struct serio *io) /* * Clear the input buffer. */ -static void __init ps2_clear_input(struct ps2if *ps2if) +static void __devinit ps2_clear_input(struct ps2if *ps2if) { int maxread = 100; @@ -228,7 +228,7 @@ static int __init ps2_test(struct ps2if *ps2if) /* * Add one device to this driver. */ -static int ps2_probe(struct sa1111_dev *dev) +static int __devinit ps2_probe(struct sa1111_dev *dev) { struct ps2if *ps2if; struct serio *serio; diff --git a/drivers/macintosh/Kconfig b/drivers/macintosh/Kconfig index 58926da0ae1..f44c94abd88 100644 --- a/drivers/macintosh/Kconfig +++ b/drivers/macintosh/Kconfig @@ -113,7 +113,6 @@ config PMAC_SMU config PMAC_APM_EMU tristate "APM emulation" - select SYS_SUPPORTS_APM_EMULATION select APM_EMULATION depends on ADB_PMU && PM diff --git a/drivers/message/fusion/mptbase.h b/drivers/message/fusion/mptbase.h index d25d3be8fcd..165f81d16d0 100644 --- a/drivers/message/fusion/mptbase.h +++ b/drivers/message/fusion/mptbase.h @@ -436,7 +436,7 @@ typedef struct _MPT_SAS_MGMT { typedef struct _mpt_ioctl_events { u32 event; /* Specified by define above */ u32 eventContext; /* Index or counter */ - int data[2]; /* First 8 bytes of Event Data */ + u32 data[2]; /* First 8 bytes of Event Data */ } MPT_IOCTL_EVENTS; /* diff --git a/drivers/message/fusion/mptscsih.c b/drivers/message/fusion/mptscsih.c index fa0f7761652..3bd94f11e7d 100644 --- a/drivers/message/fusion/mptscsih.c +++ b/drivers/message/fusion/mptscsih.c @@ -2463,11 +2463,11 @@ mptscsih_copy_sense_data(struct scsi_cmnd *sc, MPT_SCSI_HOST *hd, MPT_FRAME_HDR ioc->events[idx].event = MPI_EVENT_SCSI_DEVICE_STATUS_CHANGE; ioc->events[idx].eventContext = ioc->eventContext; - ioc->events[idx].data[0] = (pReq->LUN[1] << 24) || - (MPI_EVENT_SCSI_DEV_STAT_RC_SMART_DATA << 16) || - (sc->device->channel << 8) || sc->device->id; + ioc->events[idx].data[0] = (pReq->LUN[1] << 24) | + (MPI_EVENT_SCSI_DEV_STAT_RC_SMART_DATA << 16) | + (sc->device->channel << 8) | sc->device->id; - ioc->events[idx].data[1] = (sense_data[13] << 8) || sense_data[12]; + ioc->events[idx].data[1] = (sense_data[13] << 8) | sense_data[12]; ioc->eventContext++; if (hd->ioc->pcidev->vendor == diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig index 30fd479fea5..1798a9f9fb2 100644 --- a/drivers/net/Kconfig +++ b/drivers/net/Kconfig @@ -2280,7 +2280,6 @@ config GFAR_NAPI config UCC_GETH tristate "Freescale QE Gigabit Ethernet" depends on QUICC_ENGINE - select UCC_FAST help This driver supports the Gigabit Ethernet mode of the QUICC Engine, which is available on some Freescale SOCs. diff --git a/drivers/net/defxx.c b/drivers/net/defxx.c index 571d82f8008..7df23dc2819 100644 --- a/drivers/net/defxx.c +++ b/drivers/net/defxx.c @@ -566,6 +566,7 @@ static int __devinit dfx_register(struct device *bdev) bp->base.mem = ioremap_nocache(bar_start, bar_len); if (!bp->base.mem) { printk(KERN_ERR "%s: Cannot map MMIO\n", print_name); + err = -ENOMEM; goto err_out_region; } } else { diff --git a/drivers/net/e1000/e1000_main.c b/drivers/net/e1000/e1000_main.c index cbc7febe9cd..9ec35b7a820 100644 --- a/drivers/net/e1000/e1000_main.c +++ b/drivers/net/e1000/e1000_main.c @@ -1325,7 +1325,10 @@ e1000_sw_init(struct e1000_adapter *adapter) spin_lock_init(&adapter->tx_queue_lock); #endif - atomic_set(&adapter->irq_sem, 1); + /* Explicitly disable IRQ since the NIC can be in any state. */ + atomic_set(&adapter->irq_sem, 0); + e1000_irq_disable(adapter); + spin_lock_init(&adapter->stats_lock); set_bit(__E1000_DOWN, &adapter->flags); @@ -1431,6 +1434,10 @@ e1000_open(struct net_device *netdev) /* From here on the code is the same as e1000_up() */ clear_bit(__E1000_DOWN, &adapter->flags); +#ifdef CONFIG_E1000_NAPI + netif_poll_enable(netdev); +#endif + e1000_irq_enable(adapter); /* fire a link status change interrupt to start the watchdog */ diff --git a/drivers/net/ehea/ehea.h b/drivers/net/ehea/ehea.h index 602872dbe15..e85a933a476 100644 --- a/drivers/net/ehea/ehea.h +++ b/drivers/net/ehea/ehea.h @@ -39,7 +39,7 @@ #include <asm/io.h> #define DRV_NAME "ehea" -#define DRV_VERSION "EHEA_0058" +#define DRV_VERSION "EHEA_0061" #define EHEA_MSG_DEFAULT (NETIF_MSG_LINK | NETIF_MSG_TIMER \ | NETIF_MSG_RX_ERR | NETIF_MSG_TX_ERR) diff --git a/drivers/net/ehea/ehea_main.c b/drivers/net/ehea/ehea_main.c index f6e0cb1ada1..152bb2016a2 100644 --- a/drivers/net/ehea/ehea_main.c +++ b/drivers/net/ehea/ehea_main.c @@ -428,7 +428,7 @@ static struct ehea_cqe *ehea_proc_rwqes(struct net_device *dev, } skb_copy_to_linear_data(skb, ((char*)cqe) + 64, cqe->num_bytes_transfered - 4); - ehea_fill_skb(dev, skb, cqe); + ehea_fill_skb(port->netdev, skb, cqe); } else if (rq == 2) { /* RQ2 */ skb = get_skb_by_index(skb_arr_rq2, skb_arr_rq2_len, cqe); diff --git a/drivers/net/mlx4/alloc.c b/drivers/net/mlx4/alloc.c index dfbd5809d74..f8d63d39f59 100644 --- a/drivers/net/mlx4/alloc.c +++ b/drivers/net/mlx4/alloc.c @@ -51,8 +51,8 @@ u32 mlx4_bitmap_alloc(struct mlx4_bitmap *bitmap) if (obj < bitmap->max) { set_bit(obj, bitmap->table); + bitmap->last = (obj + 1) & (bitmap->max - 1); obj |= bitmap->top; - bitmap->last = obj + 1; } else obj = -1; diff --git a/drivers/net/sky2.c b/drivers/net/sky2.c index 832fd69a0e5..adfbe81693a 100644 --- a/drivers/net/sky2.c +++ b/drivers/net/sky2.c @@ -364,7 +364,7 @@ static void sky2_phy_init(struct sky2_hw *hw, unsigned port) /* for SFP-module set SIGDET polarity to low */ ctrl = gm_phy_read(hw, port, PHY_MARV_PHY_CTRL); ctrl |= PHY_M_FIB_SIGD_POL; - gm_phy_write(hw, port, PHY_MARV_CTRL, ctrl); + gm_phy_write(hw, port, PHY_MARV_PHY_CTRL, ctrl); } gm_phy_write(hw, port, PHY_MARV_EXT_ADR, pg); @@ -658,7 +658,7 @@ static void sky2_mac_init(struct sky2_hw *hw, unsigned port) const u8 *addr = hw->dev[port]->dev_addr; sky2_write32(hw, SK_REG(port, GPHY_CTRL), GPC_RST_SET); - sky2_write32(hw, SK_REG(port, GPHY_CTRL), GPC_RST_CLR|GPC_ENA_PAUSE); + sky2_write32(hw, SK_REG(port, GPHY_CTRL), GPC_RST_CLR); sky2_write8(hw, SK_REG(port, GMAC_CTRL), GMC_RST_CLR); @@ -1432,7 +1432,7 @@ static int sky2_xmit_frame(struct sk_buff *skb, struct net_device *dev) tcpsum = offset << 16; /* sum start */ tcpsum |= offset + skb->csum_offset; /* sum write */ - ctrl = CALSUM | WR_SUM | INIT_SUM | LOCK_SUM; + ctrl |= CALSUM | WR_SUM | INIT_SUM | LOCK_SUM; if (ip_hdr(skb)->protocol == IPPROTO_UDP) ctrl |= UDPTCP; diff --git a/drivers/net/sky2.h b/drivers/net/sky2.h index 5efb5afc45b..b8c4a3b5ead 100644 --- a/drivers/net/sky2.h +++ b/drivers/net/sky2.h @@ -1149,7 +1149,7 @@ enum { PHY_M_IS_JABBER = 1<<0, /* Jabber */ PHY_M_DEF_MSK = PHY_M_IS_LSP_CHANGE | PHY_M_IS_LST_CHANGE - | PHY_M_IS_FIFO_ERROR, + | PHY_M_IS_DUP_CHANGE, PHY_M_AN_MSK = PHY_M_IS_AN_ERROR | PHY_M_IS_AN_COMPL, }; @@ -1732,28 +1732,6 @@ enum { /* GPHY_CTRL 32 bit GPHY Control Reg (YUKON only) */ enum { - GPC_SEL_BDT = 1<<28, /* Select Bi-Dir. Transfer for MDC/MDIO */ - GPC_INT_POL_HI = 1<<27, /* IRQ Polarity is Active HIGH */ - GPC_75_OHM = 1<<26, /* Use 75 Ohm Termination instead of 50 */ - GPC_DIS_FC = 1<<25, /* Disable Automatic Fiber/Copper Detection */ - GPC_DIS_SLEEP = 1<<24, /* Disable Energy Detect */ - GPC_HWCFG_M_3 = 1<<23, /* HWCFG_MODE[3] */ - GPC_HWCFG_M_2 = 1<<22, /* HWCFG_MODE[2] */ - GPC_HWCFG_M_1 = 1<<21, /* HWCFG_MODE[1] */ - GPC_HWCFG_M_0 = 1<<20, /* HWCFG_MODE[0] */ - GPC_ANEG_0 = 1<<19, /* ANEG[0] */ - GPC_ENA_XC = 1<<18, /* Enable MDI crossover */ - GPC_DIS_125 = 1<<17, /* Disable 125 MHz clock */ - GPC_ANEG_3 = 1<<16, /* ANEG[3] */ - GPC_ANEG_2 = 1<<15, /* ANEG[2] */ - GPC_ANEG_1 = 1<<14, /* ANEG[1] */ - GPC_ENA_PAUSE = 1<<13, /* Enable Pause (SYM_OR_REM) */ - GPC_PHYADDR_4 = 1<<12, /* Bit 4 of Phy Addr */ - GPC_PHYADDR_3 = 1<<11, /* Bit 3 of Phy Addr */ - GPC_PHYADDR_2 = 1<<10, /* Bit 2 of Phy Addr */ - GPC_PHYADDR_1 = 1<<9, /* Bit 1 of Phy Addr */ - GPC_PHYADDR_0 = 1<<8, /* Bit 0 of Phy Addr */ - /* Bits 7..2: reserved */ GPC_RST_CLR = 1<<1, /* Clear GPHY Reset */ GPC_RST_SET = 1<<0, /* Set GPHY Reset */ }; diff --git a/drivers/net/wireless/hostap/hostap_80211_tx.c b/drivers/net/wireless/hostap/hostap_80211_tx.c index 246fac0e800..3df3c60263d 100644 --- a/drivers/net/wireless/hostap/hostap_80211_tx.c +++ b/drivers/net/wireless/hostap/hostap_80211_tx.c @@ -311,7 +311,7 @@ static struct sk_buff * hostap_tx_encrypt(struct sk_buff *skb, local_info_t *local; struct ieee80211_hdr_4addr *hdr; u16 fc; - int hdr_len, res; + int prefix_len, postfix_len, hdr_len, res; iface = netdev_priv(skb->dev); local = iface->local; @@ -337,10 +337,13 @@ static struct sk_buff * hostap_tx_encrypt(struct sk_buff *skb, if (skb == NULL) return NULL; - if ((skb_headroom(skb) < crypt->ops->extra_mpdu_prefix_len || - skb_tailroom(skb) < crypt->ops->extra_mpdu_postfix_len) && - pskb_expand_head(skb, crypt->ops->extra_mpdu_prefix_len, - crypt->ops->extra_mpdu_postfix_len, GFP_ATOMIC)) { + prefix_len = crypt->ops->extra_mpdu_prefix_len + + crypt->ops->extra_msdu_prefix_len; + postfix_len = crypt->ops->extra_mpdu_postfix_len + + crypt->ops->extra_msdu_postfix_len; + if ((skb_headroom(skb) < prefix_len || + skb_tailroom(skb) < postfix_len) && + pskb_expand_head(skb, prefix_len, postfix_len, GFP_ATOMIC)) { kfree_skb(skb); return NULL; } diff --git a/drivers/net/wireless/prism54/islpci_eth.c b/drivers/net/wireless/prism54/islpci_eth.c index dd070cccf32..f49eb068c7d 100644 --- a/drivers/net/wireless/prism54/islpci_eth.c +++ b/drivers/net/wireless/prism54/islpci_eth.c @@ -378,9 +378,10 @@ islpci_eth_receive(islpci_private *priv) display_buffer((char *) skb->data, skb->len); #endif /* take care of monitor mode and spy monitoring. */ - if (unlikely(priv->iw_mode == IW_MODE_MONITOR)) + if (unlikely(priv->iw_mode == IW_MODE_MONITOR)) { + skb->dev = ndev; discard = islpci_monitor_rx(priv, &skb); - else { + } else { if (unlikely(skb->data[2 * ETH_ALEN] == 0)) { /* The packet has a rx_annex. Read it for spy monitoring, Then * remove it, while keeping the 2 leading MAC addr. diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c index 6ccc2e95930..1cff65fb9c4 100644 --- a/drivers/pci/quirks.c +++ b/drivers/pci/quirks.c @@ -1625,18 +1625,20 @@ DECLARE_PCI_FIXUP_RESUME(PCI_VENDOR_ID_NVIDIA, PCI_DEVICE_ID_NVIDIA_CK804_PCIE, quirk_nvidia_ck804_pcie_aer_ext_cap); #ifdef CONFIG_PCI_MSI -/* The Serverworks PCI-X chipset does not support MSI. We cannot easily rely - * on setting PCI_BUS_FLAGS_NO_MSI in its bus flags because there are actually - * some other busses controlled by the chipset even if Linux is not aware of it. - * Instead of setting the flag on all busses in the machine, simply disable MSI - * globally. +/* Some chipsets do not support MSI. We cannot easily rely on setting + * PCI_BUS_FLAGS_NO_MSI in its bus flags because there are actually + * some other busses controlled by the chipset even if Linux is not + * aware of it. Instead of setting the flag on all busses in the + * machine, simply disable MSI globally. */ -static void __init quirk_svw_msi(struct pci_dev *dev) +static void __init quirk_disable_all_msi(struct pci_dev *dev) { pci_no_msi(); printk(KERN_WARNING "PCI: MSI quirk detected. MSI deactivated.\n"); } -DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_SERVERWORKS, PCI_DEVICE_ID_SERVERWORKS_GCNB_LE, quirk_svw_msi); +DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_SERVERWORKS, PCI_DEVICE_ID_SERVERWORKS_GCNB_LE, quirk_disable_all_msi); +DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_ATI, PCI_DEVICE_ID_ATI_RS400_200, quirk_disable_all_msi); +DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_ATI, PCI_DEVICE_ID_ATI_RS480, quirk_disable_all_msi); /* Disable MSI on chipsets that are known to not support it */ static void __devinit quirk_disable_msi(struct pci_dev *dev) @@ -1649,8 +1651,6 @@ static void __devinit quirk_disable_msi(struct pci_dev *dev) } } DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_8131_BRIDGE, quirk_disable_msi); -DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_ATI, PCI_DEVICE_ID_ATI_RS400_200, quirk_disable_msi); -DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_ATI, PCI_DEVICE_ID_ATI_RS480, quirk_disable_msi); /* Go through the list of Hypertransport capabilities and * return 1 if a HT MSI capability is found and enabled */ diff --git a/drivers/pcmcia/at91_cf.c b/drivers/pcmcia/at91_cf.c index 948efc775a7..eb6abd3f922 100644 --- a/drivers/pcmcia/at91_cf.c +++ b/drivers/pcmcia/at91_cf.c @@ -336,16 +336,21 @@ static int at91_cf_suspend(struct platform_device *pdev, pm_message_t mesg) enable_irq_wake(board->det_pin); if (board->irq_pin) enable_irq_wake(board->irq_pin); - } else { - disable_irq_wake(board->det_pin); - if (board->irq_pin) - disable_irq_wake(board->irq_pin); } return 0; } static int at91_cf_resume(struct platform_device *pdev) { + struct at91_cf_socket *cf = platform_get_drvdata(pdev); + struct at91_cf_data *board = cf->board; + + if (device_may_wakeup(&pdev->dev)) { + disable_irq_wake(board->det_pin); + if (board->irq_pin) + disable_irq_wake(board->irq_pin); + } + pcmcia_socket_dev_resume(&pdev->dev); return 0; } diff --git a/drivers/s390/scsi/zfcp_aux.c b/drivers/s390/scsi/zfcp_aux.c index ddff40c4212..821cde65e36 100644 --- a/drivers/s390/scsi/zfcp_aux.c +++ b/drivers/s390/scsi/zfcp_aux.c @@ -1127,6 +1127,7 @@ zfcp_adapter_dequeue(struct zfcp_adapter *adapter) int retval = 0; unsigned long flags; + zfcp_adapter_scsi_unregister(adapter); device_unregister(&adapter->generic_services); zfcp_sysfs_adapter_remove_files(&adapter->ccw_device->dev); dev_set_drvdata(&adapter->ccw_device->dev, NULL); diff --git a/drivers/s390/scsi/zfcp_ccw.c b/drivers/s390/scsi/zfcp_ccw.c index 81680efa172..1c8f71a5985 100644 --- a/drivers/s390/scsi/zfcp_ccw.c +++ b/drivers/s390/scsi/zfcp_ccw.c @@ -189,9 +189,7 @@ zfcp_ccw_set_online(struct ccw_device *ccw_device) * @ccw_device: pointer to belonging ccw device * * This function gets called by the common i/o layer and sets an adapter - * into state offline. Setting an fcp device offline means that it will be - * unregistered from the SCSI stack and that the adapter will be shut down - * asynchronously. + * into state offline. */ static int zfcp_ccw_set_offline(struct ccw_device *ccw_device) @@ -202,7 +200,6 @@ zfcp_ccw_set_offline(struct ccw_device *ccw_device) adapter = dev_get_drvdata(&ccw_device->dev); zfcp_erp_adapter_shutdown(adapter, 0); zfcp_erp_wait(adapter); - zfcp_adapter_scsi_unregister(adapter); zfcp_erp_thread_kill(adapter); zfcp_adapter_debug_unregister(adapter); up(&zfcp_data.config_sema); diff --git a/drivers/s390/scsi/zfcp_fsf.c b/drivers/s390/scsi/zfcp_fsf.c index a8b02542ac2..0eb31e162b1 100644 --- a/drivers/s390/scsi/zfcp_fsf.c +++ b/drivers/s390/scsi/zfcp_fsf.c @@ -156,44 +156,30 @@ zfcp_fsf_req_free(struct zfcp_fsf_req *fsf_req) kfree(fsf_req); } -/** - * zfcp_fsf_req_dismiss - dismiss a single fsf request - */ -static void zfcp_fsf_req_dismiss(struct zfcp_adapter *adapter, - struct zfcp_fsf_req *fsf_req, - unsigned int counter) -{ - u64 dbg_tmp[2]; - - dbg_tmp[0] = (u64) atomic_read(&adapter->reqs_active); - dbg_tmp[1] = (u64) counter; - debug_event(adapter->erp_dbf, 4, (void *) dbg_tmp, 16); - list_del(&fsf_req->list); - fsf_req->status |= ZFCP_STATUS_FSFREQ_DISMISSED; - zfcp_fsf_req_complete(fsf_req); -} - -/** - * zfcp_fsf_req_dismiss_all - dismiss all remaining fsf requests +/* + * Never ever call this without shutting down the adapter first. + * Otherwise the adapter would continue using and corrupting s390 storage. + * Included BUG_ON() call to ensure this is done. + * ERP is supposed to be the only user of this function. */ void zfcp_fsf_req_dismiss_all(struct zfcp_adapter *adapter) { - struct zfcp_fsf_req *request, *tmp; + struct zfcp_fsf_req *fsf_req, *tmp; unsigned long flags; LIST_HEAD(remove_queue); - unsigned int i, counter; + unsigned int i; + BUG_ON(atomic_test_mask(ZFCP_STATUS_ADAPTER_QDIOUP, &adapter->status)); spin_lock_irqsave(&adapter->req_list_lock, flags); atomic_set(&adapter->reqs_active, 0); - for (i=0; i<REQUEST_LIST_SIZE; i++) + for (i = 0; i < REQUEST_LIST_SIZE; i++) list_splice_init(&adapter->req_list[i], &remove_queue); - spin_unlock_irqrestore(&adapter->req_list_lock, flags); - counter = 0; - list_for_each_entry_safe(request, tmp, &remove_queue, list) { - zfcp_fsf_req_dismiss(adapter, request, counter); - counter++; + list_for_each_entry_safe(fsf_req, tmp, &remove_queue, list) { + list_del(&fsf_req->list); + fsf_req->status |= ZFCP_STATUS_FSFREQ_DISMISSED; + zfcp_fsf_req_complete(fsf_req); } } diff --git a/drivers/s390/scsi/zfcp_scsi.c b/drivers/s390/scsi/zfcp_scsi.c index 16e2d64658a..0acf6db0a08 100644 --- a/drivers/s390/scsi/zfcp_scsi.c +++ b/drivers/s390/scsi/zfcp_scsi.c @@ -569,6 +569,9 @@ zfcp_adapter_scsi_register(struct zfcp_adapter *adapter) int retval = 0; static unsigned int unique_id = 0; + if (adapter->scsi_host) + goto out; + /* register adapter as SCSI host with mid layer of SCSI stack */ adapter->scsi_host = scsi_host_alloc(&zfcp_data.scsi_host_template, sizeof (struct zfcp_adapter *)); diff --git a/drivers/scsi/Kconfig b/drivers/scsi/Kconfig index d28c14e23c3..572034ceb14 100644 --- a/drivers/scsi/Kconfig +++ b/drivers/scsi/Kconfig @@ -1753,23 +1753,9 @@ config SUN3X_ESP The ESP was an on-board SCSI controller used on Sun 3/80 machines. Say Y here to compile in support for it. -config SCSI_ESP_CORE - tristate "ESP Scsi Driver Core" - depends on SCSI - select SCSI_SPI_ATTRS - help - This is a core driver for NCR53c9x based scsi chipsets, - also known as "ESP" for Emulex Scsi Processor or - Enhanced Scsi Processor. This driver does not exist by - itself, there are front-end drivers which, when enabled, - select and enable this driver. One example is SCSI_SUNESP. - These front-end drivers provide probing, DMA, and register - access support for the core driver. - config SCSI_SUNESP tristate "Sparc ESP Scsi Driver" depends on SBUS && SCSI - select SCSI_ESP_CORE help This is the driver for the Sun ESP SCSI host adapter. The ESP chipset is present in most SPARC SBUS-based computers. diff --git a/drivers/scsi/Makefile b/drivers/scsi/Makefile index 51e884fa10b..b1b63279158 100644 --- a/drivers/scsi/Makefile +++ b/drivers/scsi/Makefile @@ -106,8 +106,7 @@ obj-$(CONFIG_MEGARAID_LEGACY) += megaraid.o obj-$(CONFIG_MEGARAID_NEWGEN) += megaraid/ obj-$(CONFIG_MEGARAID_SAS) += megaraid/ obj-$(CONFIG_SCSI_ACARD) += atp870u.o -obj-$(CONFIG_SCSI_ESP_CORE) += esp_scsi.o -obj-$(CONFIG_SCSI_SUNESP) += sun_esp.o +obj-$(CONFIG_SCSI_SUNESP) += esp_scsi.o sun_esp.o obj-$(CONFIG_SCSI_GDTH) += gdth.o obj-$(CONFIG_SCSI_INITIO) += initio.o obj-$(CONFIG_SCSI_INIA100) += a100u2w.o @@ -121,7 +120,7 @@ obj-$(CONFIG_BLK_DEV_3W_XXXX_RAID) += 3w-xxxx.o obj-$(CONFIG_SCSI_3W_9XXX) += 3w-9xxx.o obj-$(CONFIG_SCSI_PPA) += ppa.o obj-$(CONFIG_SCSI_IMM) += imm.o -obj-$(CONFIG_JAZZ_ESP) += NCR53C9x.o jazz_esp.o +obj-$(CONFIG_JAZZ_ESP) += esp_scsi.o jazz_esp.o obj-$(CONFIG_SUN3X_ESP) += NCR53C9x.o sun3x_esp.o obj-$(CONFIG_SCSI_FCAL) += fcal.o obj-$(CONFIG_SCSI_LASI700) += 53c700.o lasi700.o diff --git a/drivers/scsi/aacraid/aachba.c b/drivers/scsi/aacraid/aachba.c index 1e82c69b36b..8dcfe4ec35c 100644 --- a/drivers/scsi/aacraid/aachba.c +++ b/drivers/scsi/aacraid/aachba.c @@ -146,7 +146,7 @@ static char *aac_get_status_string(u32 status); static int nondasd = -1; static int dacmode = -1; -static int commit = -1; +int aac_commit = -1; int startup_timeout = 180; int aif_timeout = 120; @@ -154,7 +154,7 @@ module_param(nondasd, int, S_IRUGO|S_IWUSR); MODULE_PARM_DESC(nondasd, "Control scanning of hba for nondasd devices. 0=off, 1=on"); module_param(dacmode, int, S_IRUGO|S_IWUSR); MODULE_PARM_DESC(dacmode, "Control whether dma addressing is using 64 bit DAC. 0=off, 1=on"); -module_param(commit, int, S_IRUGO|S_IWUSR); +module_param_named(commit, aac_commit, int, S_IRUGO|S_IWUSR); MODULE_PARM_DESC(commit, "Control whether a COMMIT_CONFIG is issued to the adapter for foreign arrays.\nThis is typically needed in systems that do not have a BIOS. 0=off, 1=on"); module_param(startup_timeout, int, S_IRUGO|S_IWUSR); MODULE_PARM_DESC(startup_timeout, "The duration of time in seconds to wait for adapter to have it's kernel up and\nrunning. This is typically adjusted for large systems that do not have a BIOS."); @@ -173,6 +173,9 @@ int expose_physicals = -1; module_param(expose_physicals, int, S_IRUGO|S_IWUSR); MODULE_PARM_DESC(expose_physicals, "Expose physical components of the arrays. -1=protect 0=off, 1=on"); +int aac_reset_devices = 0; +module_param_named(reset_devices, aac_reset_devices, int, S_IRUGO|S_IWUSR); +MODULE_PARM_DESC(reset_devices, "Force an adapter reset at initialization."); static inline int aac_valid_context(struct scsi_cmnd *scsicmd, struct fib *fibptr) { @@ -246,7 +249,7 @@ int aac_get_config_status(struct aac_dev *dev, int commit_flag) aac_fib_complete(fibptr); /* Send a CT_COMMIT_CONFIG to enable discovery of devices */ if (status >= 0) { - if ((commit == 1) || commit_flag) { + if ((aac_commit == 1) || commit_flag) { struct aac_commit_config * dinfo; aac_fib_init(fibptr); dinfo = (struct aac_commit_config *) fib_data(fibptr); @@ -261,7 +264,7 @@ int aac_get_config_status(struct aac_dev *dev, int commit_flag) 1, 1, NULL, NULL); aac_fib_complete(fibptr); - } else if (commit == 0) { + } else if (aac_commit == 0) { printk(KERN_WARNING "aac_get_config_status: Foreign device configurations are being ignored\n"); } @@ -340,7 +343,7 @@ int aac_get_containers(struct aac_dev *dev) static void aac_internal_transfer(struct scsi_cmnd *scsicmd, void *data, unsigned int offset, unsigned int len) { void *buf; - unsigned int transfer_len; + int transfer_len; struct scatterlist *sg = scsicmd->request_buffer; if (scsicmd->use_sg) { @@ -351,7 +354,7 @@ static void aac_internal_transfer(struct scsi_cmnd *scsicmd, void *data, unsigne transfer_len = min(scsicmd->request_bufflen, len + offset); } transfer_len -= offset; - if (buf && transfer_len) + if (buf && transfer_len > 0) memcpy(buf + offset, data, transfer_len); if (scsicmd->use_sg) diff --git a/drivers/scsi/aacraid/aacraid.h b/drivers/scsi/aacraid/aacraid.h index 45ca3e80161..c81edf36913 100644 --- a/drivers/scsi/aacraid/aacraid.h +++ b/drivers/scsi/aacraid/aacraid.h @@ -1823,9 +1823,12 @@ int aac_send_shutdown(struct aac_dev *dev); int aac_probe_container(struct aac_dev *dev, int cid); int _aac_rx_init(struct aac_dev *dev); int aac_rx_select_comm(struct aac_dev *dev, int comm); +int aac_rx_deliver_producer(struct fib * fib); extern int numacb; extern int acbsize; extern char aac_driver_version[]; extern int startup_timeout; extern int aif_timeout; extern int expose_physicals; +extern int aac_reset_devices; +extern int aac_commit; diff --git a/drivers/scsi/aacraid/rx.c b/drivers/scsi/aacraid/rx.c index 291cd14f4e9..ae978a373c5 100644 --- a/drivers/scsi/aacraid/rx.c +++ b/drivers/scsi/aacraid/rx.c @@ -378,7 +378,7 @@ static int aac_rx_check_health(struct aac_dev *dev) * * Will send a fib, returning 0 if successful. */ -static int aac_rx_deliver_producer(struct fib * fib) +int aac_rx_deliver_producer(struct fib * fib) { struct aac_dev *dev = fib->dev; struct aac_queue *q = &dev->queues->queue[AdapNormCmdQueue]; @@ -488,6 +488,8 @@ static int aac_rx_restart_adapter(struct aac_dev *dev, int bled) return -EINVAL; if (rx_readl(dev, MUnit.OMRx[0]) & KERNEL_PANIC) return -ENODEV; + if (startup_timeout < 300) + startup_timeout = 300; return 0; } @@ -542,7 +544,7 @@ int _aac_rx_init(struct aac_dev *dev) dev->a_ops.adapter_sync_cmd = rx_sync_cmd; dev->a_ops.adapter_enable_int = aac_rx_disable_interrupt; dev->OIMR = status = rx_readb (dev, MUnit.OIMR); - if ((((status & 0x0c) != 0x0c) || reset_devices) && + if ((((status & 0x0c) != 0x0c) || aac_reset_devices || reset_devices) && !aac_rx_restart_adapter(dev, 0)) ++restart; /* @@ -594,6 +596,8 @@ int _aac_rx_init(struct aac_dev *dev) } msleep(1); } + if (restart) + aac_commit = 1; /* * Fill in the common function dispatch table. */ diff --git a/drivers/scsi/aacraid/sa.c b/drivers/scsi/aacraid/sa.c index f4b5e9742ab..85b91bc578c 100644 --- a/drivers/scsi/aacraid/sa.c +++ b/drivers/scsi/aacraid/sa.c @@ -5,7 +5,7 @@ * based on the old aacraid driver that is.. * Adaptec aacraid device driver for Linux. * - * Copyright (c) 2000 Adaptec, Inc. (aacraid@adaptec.com) + * Copyright (c) 2000-2007 Adaptec, Inc. (aacraid@adaptec.com) * * 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 @@ -257,6 +257,11 @@ static void aac_sa_start_adapter(struct aac_dev *dev) NULL, NULL, NULL, NULL, NULL); } +static int aac_sa_restart_adapter(struct aac_dev *dev, int bled) +{ + return -EINVAL; +} + /** * aac_sa_check_health * @dev: device to check if healthy @@ -366,7 +371,9 @@ int aac_sa_init(struct aac_dev *dev) dev->a_ops.adapter_notify = aac_sa_notify_adapter; dev->a_ops.adapter_sync_cmd = sa_sync_cmd; dev->a_ops.adapter_check_health = aac_sa_check_health; + dev->a_ops.adapter_restart = aac_sa_restart_adapter; dev->a_ops.adapter_intr = aac_sa_intr; + dev->a_ops.adapter_deliver = aac_rx_deliver_producer; dev->a_ops.adapter_ioremap = aac_sa_ioremap; /* diff --git a/drivers/scsi/aic7xxx/aicasm/aicasm_gram.y b/drivers/scsi/aic7xxx/aicasm/aicasm_gram.y index c328596def3..6066998ed56 100644 --- a/drivers/scsi/aic7xxx/aicasm/aicasm_gram.y +++ b/drivers/scsi/aic7xxx/aicasm/aicasm_gram.y @@ -106,6 +106,7 @@ static void make_expression(expression_t *immed, int value); static void add_conditional(symbol_t *symbol); static void add_version(const char *verstring); static int is_download_const(expression_t *immed); +void yyerror(const char *string); #define SRAM_SYMNAME "SRAM_BASE" #define SCB_SYMNAME "SCB_BASE" diff --git a/drivers/scsi/aic7xxx/aicasm/aicasm_macro_gram.y b/drivers/scsi/aic7xxx/aicasm/aicasm_macro_gram.y index 439f760b34b..ff46aa6801b 100644 --- a/drivers/scsi/aic7xxx/aicasm/aicasm_macro_gram.y +++ b/drivers/scsi/aic7xxx/aicasm/aicasm_macro_gram.y @@ -65,6 +65,7 @@ static symbol_t *macro_symbol; static void add_macro_arg(const char *argtext, int position); +void mmerror(const char *string); %} diff --git a/drivers/scsi/aic94xx/aic94xx_tmf.c b/drivers/scsi/aic94xx/aic94xx_tmf.c index 9a14a6d9727..c0d0b7d7a8c 100644 --- a/drivers/scsi/aic94xx/aic94xx_tmf.c +++ b/drivers/scsi/aic94xx/aic94xx_tmf.c @@ -290,6 +290,7 @@ static void asd_tmf_tasklet_complete(struct asd_ascb *ascb, static inline int asd_clear_nexus(struct sas_task *task) { int res = TMF_RESP_FUNC_FAILED; + int leftover; struct asd_ascb *tascb = task->lldd_task; unsigned long flags; @@ -298,10 +299,12 @@ static inline int asd_clear_nexus(struct sas_task *task) res = asd_clear_nexus_tag(task); else res = asd_clear_nexus_index(task); - wait_for_completion_timeout(&tascb->completion, - AIC94XX_SCB_TIMEOUT); + leftover = wait_for_completion_timeout(&tascb->completion, + AIC94XX_SCB_TIMEOUT); ASD_DPRINTK("came back from clear nexus\n"); spin_lock_irqsave(&task->task_state_lock, flags); + if (leftover < 1) + res = TMF_RESP_FUNC_FAILED; if (task->task_state_flags & SAS_TASK_STATE_DONE) res = TMF_RESP_FUNC_COMPLETE; spin_unlock_irqrestore(&task->task_state_lock, flags); @@ -350,6 +353,7 @@ int asd_abort_task(struct sas_task *task) unsigned long flags; struct asd_ascb *ascb = NULL; struct scb *scb; + int leftover; spin_lock_irqsave(&task->task_state_lock, flags); if (task->task_state_flags & SAS_TASK_STATE_DONE) { @@ -455,9 +459,11 @@ int asd_abort_task(struct sas_task *task) break; case TF_TMF_TASK_DONE + 0xFF00: /* done but not reported yet */ res = TMF_RESP_FUNC_FAILED; - wait_for_completion_timeout(&tascb->completion, - AIC94XX_SCB_TIMEOUT); + leftover = wait_for_completion_timeout(&tascb->completion, + AIC94XX_SCB_TIMEOUT); spin_lock_irqsave(&task->task_state_lock, flags); + if (leftover < 1) + res = TMF_RESP_FUNC_FAILED; if (task->task_state_flags & SAS_TASK_STATE_DONE) res = TMF_RESP_FUNC_COMPLETE; spin_unlock_irqrestore(&task->task_state_lock, flags); diff --git a/drivers/scsi/ipr.c b/drivers/scsi/ipr.c index 4baa79e6867..fa6ff295e56 100644 --- a/drivers/scsi/ipr.c +++ b/drivers/scsi/ipr.c @@ -3954,6 +3954,13 @@ static int __ipr_eh_dev_reset(struct scsi_cmnd * scsi_cmd) spin_unlock_irq(scsi_cmd->device->host->host_lock); ata_do_eh(ap, NULL, NULL, ipr_sata_reset, NULL); spin_lock_irq(scsi_cmd->device->host->host_lock); + + list_for_each_entry(ipr_cmd, &ioa_cfg->pending_q, queue) { + if (ipr_cmd->ioarcb.res_handle == res->cfgte.res_handle) { + rc = -EIO; + break; + } + } } else rc = ipr_device_reset(ioa_cfg, res); res->resetting_device = 0; diff --git a/drivers/scsi/jazz_esp.c b/drivers/scsi/jazz_esp.c index 19dd4b962e1..81e497d9eae 100644 --- a/drivers/scsi/jazz_esp.c +++ b/drivers/scsi/jazz_esp.c @@ -1,307 +1,244 @@ -/* - * jazz_esp.c: Driver for SCSI chip on Mips Magnum Boards (JAZZ architecture) +/* jazz_esp.c: ESP front-end for MIPS JAZZ systems. * - * Copyright (C) 1997 Thomas Bogendoerfer (tsbogend@alpha.franken.de) - * - * jazz_esp is based on David S. Miller's ESP driver and cyber_esp + * Copyright (C) 2007 Thomas Bogendörfer (tsbogend@alpha.frankende) */ -#include <linux/init.h> #include <linux/kernel.h> -#include <linux/delay.h> #include <linux/types.h> -#include <linux/string.h> -#include <linux/slab.h> -#include <linux/blkdev.h> -#include <linux/proc_fs.h> -#include <linux/stat.h> - -#include "scsi.h" -#include <scsi/scsi_host.h> -#include "NCR53C9x.h" +#include <linux/module.h> +#include <linux/init.h> +#include <linux/interrupt.h> +#include <linux/platform_device.h> +#include <linux/dma-mapping.h> #include <asm/irq.h> +#include <asm/io.h> +#include <asm/dma.h> + #include <asm/jazz.h> #include <asm/jazzdma.h> -#include <asm/dma.h> -#include <asm/pgtable.h> - -static int dma_bytes_sent(struct NCR_ESP *esp, int fifo_count); -static int dma_can_transfer(struct NCR_ESP *esp, struct scsi_cmnd *sp); -static void dma_dump_state(struct NCR_ESP *esp); -static void dma_init_read(struct NCR_ESP *esp, __u32 vaddress, int length); -static void dma_init_write(struct NCR_ESP *esp, __u32 vaddress, int length); -static void dma_ints_off(struct NCR_ESP *esp); -static void dma_ints_on(struct NCR_ESP *esp); -static int dma_irq_p(struct NCR_ESP *esp); -static int dma_ports_p(struct NCR_ESP *esp); -static void dma_setup(struct NCR_ESP *esp, __u32 addr, int count, int write); -static void dma_mmu_get_scsi_one (struct NCR_ESP *esp, struct scsi_cmnd *sp); -static void dma_mmu_get_scsi_sgl (struct NCR_ESP *esp, struct scsi_cmnd *sp); -static void dma_mmu_release_scsi_one (struct NCR_ESP *esp, struct scsi_cmnd *sp); -static void dma_mmu_release_scsi_sgl (struct NCR_ESP *esp, struct scsi_cmnd *sp); -static void dma_advance_sg (struct scsi_cmnd *sp); -static void dma_led_off(struct NCR_ESP *); -static void dma_led_on(struct NCR_ESP *); - - -static volatile unsigned char cmd_buffer[16]; - /* This is where all commands are put - * before they are trasfered to the ESP chip - * via PIO. - */ - -static int jazz_esp_release(struct Scsi_Host *shost) -{ - if (shost->irq) - free_irq(shost->irq, NULL); - if (shost->dma_channel != 0xff) - free_dma(shost->dma_channel); - if (shost->io_port && shost->n_io_port) - release_region(shost->io_port, shost->n_io_port); - scsi_unregister(shost); - return 0; -} +#include <scsi/scsi_host.h> -/***************************************************************** Detection */ -static int jazz_esp_detect(struct scsi_host_template *tpnt) -{ - struct NCR_ESP *esp; - struct ConfigDev *esp_dev; - - /* - * first assumption it is there:-) - */ - if (1) { - esp_dev = NULL; - esp = esp_allocate(tpnt, esp_dev, 0); - - /* Do command transfer with programmed I/O */ - esp->do_pio_cmds = 1; - - /* Required functions */ - esp->dma_bytes_sent = &dma_bytes_sent; - esp->dma_can_transfer = &dma_can_transfer; - esp->dma_dump_state = &dma_dump_state; - esp->dma_init_read = &dma_init_read; - esp->dma_init_write = &dma_init_write; - esp->dma_ints_off = &dma_ints_off; - esp->dma_ints_on = &dma_ints_on; - esp->dma_irq_p = &dma_irq_p; - esp->dma_ports_p = &dma_ports_p; - esp->dma_setup = &dma_setup; - - /* Optional functions */ - esp->dma_barrier = NULL; - esp->dma_drain = NULL; - esp->dma_invalidate = NULL; - esp->dma_irq_entry = NULL; - esp->dma_irq_exit = NULL; - esp->dma_poll = NULL; - esp->dma_reset = NULL; - esp->dma_led_off = &dma_led_off; - esp->dma_led_on = &dma_led_on; - - /* virtual DMA functions */ - esp->dma_mmu_get_scsi_one = &dma_mmu_get_scsi_one; - esp->dma_mmu_get_scsi_sgl = &dma_mmu_get_scsi_sgl; - esp->dma_mmu_release_scsi_one = &dma_mmu_release_scsi_one; - esp->dma_mmu_release_scsi_sgl = &dma_mmu_release_scsi_sgl; - esp->dma_advance_sg = &dma_advance_sg; - - - /* SCSI chip speed */ - esp->cfreq = 40000000; +#include "esp_scsi.h" - /* - * we don't give the address of DMA channel, but the number - * of DMA channel, so we can use the jazz DMA functions - * - */ - esp->dregs = (void *) JAZZ_SCSI_DMA; - - /* ESP register base */ - esp->eregs = (struct ESP_regs *)(JAZZ_SCSI_BASE); - - /* Set the command buffer */ - esp->esp_command = (volatile unsigned char *)cmd_buffer; - - /* get virtual dma address for command buffer */ - esp->esp_command_dvma = vdma_alloc(CPHYSADDR(cmd_buffer), sizeof (cmd_buffer)); - - esp->irq = JAZZ_SCSI_IRQ; - request_irq(JAZZ_SCSI_IRQ, esp_intr, IRQF_DISABLED, "JAZZ SCSI", - esp->ehost); - - /* - * FIXME, look if the scsi id is available from NVRAM - */ - esp->scsi_id = 7; - - /* Check for differential SCSI-bus */ - /* What is this stuff? */ - esp->diff = 0; - - esp_initialize(esp); - - printk("ESP: Total of %d ESP hosts found, %d actually in use.\n", nesps,esps_in_use); - esps_running = esps_in_use; - return esps_in_use; - } - return 0; -} +#define DRV_MODULE_NAME "jazz_esp" +#define PFX DRV_MODULE_NAME ": " +#define DRV_VERSION "1.000" +#define DRV_MODULE_RELDATE "May 19, 2007" -/************************************************************* DMA Functions */ -static int dma_bytes_sent(struct NCR_ESP *esp, int fifo_count) +static void jazz_esp_write8(struct esp *esp, u8 val, unsigned long reg) { - return fifo_count; + *(volatile u8 *)(esp->regs + reg) = val; } -static int dma_can_transfer(struct NCR_ESP *esp, struct scsi_cmnd *sp) +static u8 jazz_esp_read8(struct esp *esp, unsigned long reg) { - /* - * maximum DMA size is 1MB - */ - unsigned long sz = sp->SCp.this_residual; - if(sz > 0x100000) - sz = 0x100000; - return sz; + return *(volatile u8 *)(esp->regs + reg); } -static void dma_dump_state(struct NCR_ESP *esp) +static dma_addr_t jazz_esp_map_single(struct esp *esp, void *buf, + size_t sz, int dir) { - - ESPLOG(("esp%d: dma -- enable <%08x> residue <%08x\n", - esp->esp_id, vdma_get_enable((int)esp->dregs), vdma_get_residue((int)esp->dregs))); + return dma_map_single(esp->dev, buf, sz, dir); } -static void dma_init_read(struct NCR_ESP *esp, __u32 vaddress, int length) +static int jazz_esp_map_sg(struct esp *esp, struct scatterlist *sg, + int num_sg, int dir) { - dma_cache_wback_inv ((unsigned long)phys_to_virt(vdma_log2phys(vaddress)), length); - vdma_disable ((int)esp->dregs); - vdma_set_mode ((int)esp->dregs, DMA_MODE_READ); - vdma_set_addr ((int)esp->dregs, vaddress); - vdma_set_count ((int)esp->dregs, length); - vdma_enable ((int)esp->dregs); + return dma_map_sg(esp->dev, sg, num_sg, dir); } -static void dma_init_write(struct NCR_ESP *esp, __u32 vaddress, int length) +static void jazz_esp_unmap_single(struct esp *esp, dma_addr_t addr, + size_t sz, int dir) { - dma_cache_wback_inv ((unsigned long)phys_to_virt(vdma_log2phys(vaddress)), length); - vdma_disable ((int)esp->dregs); - vdma_set_mode ((int)esp->dregs, DMA_MODE_WRITE); - vdma_set_addr ((int)esp->dregs, vaddress); - vdma_set_count ((int)esp->dregs, length); - vdma_enable ((int)esp->dregs); + dma_unmap_single(esp->dev, addr, sz, dir); } -static void dma_ints_off(struct NCR_ESP *esp) +static void jazz_esp_unmap_sg(struct esp *esp, struct scatterlist *sg, + int num_sg, int dir) { - disable_irq(esp->irq); + dma_unmap_sg(esp->dev, sg, num_sg, dir); } -static void dma_ints_on(struct NCR_ESP *esp) +static int jazz_esp_irq_pending(struct esp *esp) { - enable_irq(esp->irq); + if (jazz_esp_read8(esp, ESP_STATUS) & ESP_STAT_INTR) + return 1; + return 0; } -static int dma_irq_p(struct NCR_ESP *esp) +static void jazz_esp_reset_dma(struct esp *esp) { - return (esp_read(esp->eregs->esp_status) & ESP_STAT_INTR); + vdma_disable ((int)esp->dma_regs); } -static int dma_ports_p(struct NCR_ESP *esp) +static void jazz_esp_dma_drain(struct esp *esp) { - int enable = vdma_get_enable((int)esp->dregs); - - return (enable & R4030_CHNL_ENABLE); + /* nothing to do */ } -static void dma_setup(struct NCR_ESP *esp, __u32 addr, int count, int write) +static void jazz_esp_dma_invalidate(struct esp *esp) { - /* - * On the Sparc, DMA_ST_WRITE means "move data from device to memory" - * so when (write) is true, it actually means READ! - */ - if(write){ - dma_init_read(esp, addr, count); - } else { - dma_init_write(esp, addr, count); - } + vdma_disable ((int)esp->dma_regs); } -static void dma_mmu_get_scsi_one (struct NCR_ESP *esp, struct scsi_cmnd *sp) +static void jazz_esp_send_dma_cmd(struct esp *esp, u32 addr, u32 esp_count, + u32 dma_count, int write, u8 cmd) { - sp->SCp.have_data_in = vdma_alloc(CPHYSADDR(sp->SCp.buffer), sp->SCp.this_residual); - sp->SCp.ptr = (char *)((unsigned long)sp->SCp.have_data_in); + BUG_ON(!(cmd & ESP_CMD_DMA)); + + jazz_esp_write8(esp, (esp_count >> 0) & 0xff, ESP_TCLOW); + jazz_esp_write8(esp, (esp_count >> 8) & 0xff, ESP_TCMED); + vdma_disable ((int)esp->dma_regs); + if (write) + vdma_set_mode ((int)esp->dma_regs, DMA_MODE_READ); + else + vdma_set_mode ((int)esp->dma_regs, DMA_MODE_WRITE); + + vdma_set_addr ((int)esp->dma_regs, addr); + vdma_set_count ((int)esp->dma_regs, dma_count); + vdma_enable ((int)esp->dma_regs); + + scsi_esp_cmd(esp, cmd); } -static void dma_mmu_get_scsi_sgl (struct NCR_ESP *esp, struct scsi_cmnd *sp) -{ - int sz = sp->SCp.buffers_residual; - struct scatterlist *sg = (struct scatterlist *) sp->SCp.buffer; - - while (sz >= 0) { - sg[sz].dma_address = vdma_alloc(CPHYSADDR(page_address(sg[sz].page) + sg[sz].offset), sg[sz].length); - sz--; - } - sp->SCp.ptr=(char *)(sp->SCp.buffer->dma_address); -} - -static void dma_mmu_release_scsi_one (struct NCR_ESP *esp, struct scsi_cmnd *sp) +static int jazz_esp_dma_error(struct esp *esp) { - vdma_free(sp->SCp.have_data_in); + u32 enable = vdma_get_enable((int)esp->dma_regs); + + if (enable & (R4030_MEM_INTR|R4030_ADDR_INTR)) + return 1; + + return 0; } -static void dma_mmu_release_scsi_sgl (struct NCR_ESP *esp, struct scsi_cmnd *sp) +static const struct esp_driver_ops jazz_esp_ops = { + .esp_write8 = jazz_esp_write8, + .esp_read8 = jazz_esp_read8, + .map_single = jazz_esp_map_single, + .map_sg = jazz_esp_map_sg, + .unmap_single = jazz_esp_unmap_single, + .unmap_sg = jazz_esp_unmap_sg, + .irq_pending = jazz_esp_irq_pending, + .reset_dma = jazz_esp_reset_dma, + .dma_drain = jazz_esp_dma_drain, + .dma_invalidate = jazz_esp_dma_invalidate, + .send_dma_cmd = jazz_esp_send_dma_cmd, + .dma_error = jazz_esp_dma_error, +}; + +static int __devinit esp_jazz_probe(struct platform_device *dev) { - int sz = sp->use_sg - 1; - struct scatterlist *sg = (struct scatterlist *)sp->request_buffer; - - while(sz >= 0) { - vdma_free(sg[sz].dma_address); - sz--; - } + struct scsi_host_template *tpnt = &scsi_esp_template; + struct Scsi_Host *host; + struct esp *esp; + struct resource *res; + int err; + + host = scsi_host_alloc(tpnt, sizeof(struct esp)); + + err = -ENOMEM; + if (!host) + goto fail; + + host->max_id = 8; + esp = host_to_esp(host); + + esp->host = host; + esp->dev = dev; + esp->ops = &jazz_esp_ops; + + res = platform_get_resource(dev, IORESOURCE_MEM, 0); + if (!res) + goto fail_unlink; + + esp->regs = (void __iomem *)res->start; + if (!esp->regs) + goto fail_unlink; + + res = platform_get_resource(dev, IORESOURCE_MEM, 1); + if (!res) + goto fail_unlink; + + esp->dma_regs = (void __iomem *)res->start; + + esp->command_block = dma_alloc_coherent(esp->dev, 16, + &esp->command_block_dma, + GFP_KERNEL); + if (!esp->command_block) + goto fail_unmap_regs; + + host->irq = platform_get_irq(dev, 0); + err = request_irq(host->irq, scsi_esp_intr, IRQF_SHARED, "ESP", esp); + if (err < 0) + goto fail_unmap_command_block; + + esp->scsi_id = 7; + esp->host->this_id = esp->scsi_id; + esp->scsi_id_mask = (1 << esp->scsi_id); + esp->cfreq = 40000000; + + dev_set_drvdata(&dev->dev, esp); + + err = scsi_esp_register(esp, &dev->dev); + if (err) + goto fail_free_irq; + + return 0; + +fail_free_irq: + free_irq(host->irq, esp); +fail_unmap_command_block: + dma_free_coherent(esp->dev, 16, + esp->command_block, + esp->command_block_dma); +fail_unmap_regs: +fail_unlink: + scsi_host_put(host); +fail: + return err; } -static void dma_advance_sg (struct scsi_cmnd *sp) +static int __devexit esp_jazz_remove(struct platform_device *dev) { - sp->SCp.ptr = (char *)(sp->SCp.buffer->dma_address); + struct esp *esp = dev_get_drvdata(&dev->dev); + unsigned int irq = esp->host->irq; + + scsi_esp_unregister(esp); + + free_irq(irq, esp); + dma_free_coherent(esp->dev, 16, + esp->command_block, + esp->command_block_dma); + + scsi_host_put(esp->host); + + return 0; } -#define JAZZ_HDC_LED 0xe000d100 /* FIXME, find correct address */ +static struct platform_driver esp_jazz_driver = { + .probe = esp_jazz_probe, + .remove = __devexit_p(esp_jazz_remove), + .driver = { + .name = "jazz_esp", + }, +}; -static void dma_led_off(struct NCR_ESP *esp) +static int __init jazz_esp_init(void) { -#if 0 - *(unsigned char *)JAZZ_HDC_LED = 0; -#endif + return platform_driver_register(&esp_jazz_driver); } -static void dma_led_on(struct NCR_ESP *esp) -{ -#if 0 - *(unsigned char *)JAZZ_HDC_LED = 1; -#endif +static void __exit jazz_esp_exit(void) +{ + platform_driver_unregister(&esp_jazz_driver); } -static struct scsi_host_template driver_template = { - .proc_name = "jazz_esp", - .proc_info = esp_proc_info, - .name = "ESP 100/100a/200", - .detect = jazz_esp_detect, - .slave_alloc = esp_slave_alloc, - .slave_destroy = esp_slave_destroy, - .release = jazz_esp_release, - .info = esp_info, - .queuecommand = esp_queue, - .eh_abort_handler = esp_abort, - .eh_bus_reset_handler = esp_reset, - .can_queue = 7, - .this_id = 7, - .sg_tablesize = SG_ALL, - .cmd_per_lun = 1, - .use_clustering = DISABLE_CLUSTERING, -}; -#include "scsi_module.c" +MODULE_DESCRIPTION("JAZZ ESP SCSI driver"); +MODULE_AUTHOR("Thomas Bogendoerfer (tsbogend@alpha.franken.de)"); +MODULE_LICENSE("GPL"); +MODULE_VERSION(DRV_VERSION); + +module_init(jazz_esp_init); +module_exit(jazz_esp_exit); diff --git a/drivers/scsi/libsrp.c b/drivers/scsi/libsrp.c index 5631c199a8e..732446e6396 100644 --- a/drivers/scsi/libsrp.c +++ b/drivers/scsi/libsrp.c @@ -254,6 +254,7 @@ static int srp_indirect_data(struct scsi_cmnd *sc, struct srp_cmd *cmd, sg_init_one(&dummy, md, id->table_desc.len); sg_dma_address(&dummy) = token; + sg_dma_len(&dummy) = id->table_desc.len; err = rdma_io(sc, &dummy, 1, &id->table_desc, 1, DMA_TO_DEVICE, id->table_desc.len); if (err) { diff --git a/drivers/scsi/megaraid/megaraid_sas.c b/drivers/scsi/megaraid/megaraid_sas.c index 7a812677ff8..e2cf12ef368 100644 --- a/drivers/scsi/megaraid/megaraid_sas.c +++ b/drivers/scsi/megaraid/megaraid_sas.c @@ -10,7 +10,7 @@ * 2 of the License, or (at your option) any later version. * * FILE : megaraid_sas.c - * Version : v00.00.03.10-rc1 + * Version : v00.00.03.10-rc5 * * Authors: * (email-id : megaraidlinux@lsi.com) @@ -886,6 +886,7 @@ megasas_queue_command(struct scsi_cmnd *scmd, void (*done) (struct scsi_cmnd *)) goto out_return_cmd; cmd->scmd = scmd; + scmd->SCp.ptr = (char *)cmd; /* * Issue the command to the FW @@ -919,7 +920,7 @@ static int megasas_slave_configure(struct scsi_device *sdev) * The RAID firmware may require extended timeouts. */ if (sdev->channel >= MEGASAS_MAX_PD_CHANNELS) - sdev->timeout = 90 * HZ; + sdev->timeout = MEGASAS_DEFAULT_CMD_TIMEOUT * HZ; return 0; } @@ -981,8 +982,8 @@ static int megasas_generic_reset(struct scsi_cmnd *scmd) instance = (struct megasas_instance *)scmd->device->host->hostdata; - scmd_printk(KERN_NOTICE, scmd, "megasas: RESET -%ld cmd=%x\n", - scmd->serial_number, scmd->cmnd[0]); + scmd_printk(KERN_NOTICE, scmd, "megasas: RESET -%ld cmd=%x retries=%x\n", + scmd->serial_number, scmd->cmnd[0], scmd->retries); if (instance->hw_crit_error) { printk(KERN_ERR "megasas: cannot recover from previous reset " @@ -1000,6 +1001,39 @@ static int megasas_generic_reset(struct scsi_cmnd *scmd) } /** + * megasas_reset_timer - quiesce the adapter if required + * @scmd: scsi cmnd + * + * Sets the FW busy flag and reduces the host->can_queue if the + * cmd has not been completed within the timeout period. + */ +static enum +scsi_eh_timer_return megasas_reset_timer(struct scsi_cmnd *scmd) +{ + struct megasas_cmd *cmd = (struct megasas_cmd *)scmd->SCp.ptr; + struct megasas_instance *instance; + unsigned long flags; + + if (time_after(jiffies, scmd->jiffies_at_alloc + + (MEGASAS_DEFAULT_CMD_TIMEOUT * 2) * HZ)) { + return EH_NOT_HANDLED; + } + + instance = cmd->instance; + if (!(instance->flag & MEGASAS_FW_BUSY)) { + /* FW is busy, throttle IO */ + spin_lock_irqsave(instance->host->host_lock, flags); + + instance->host->can_queue = 16; + instance->last_time = jiffies; + instance->flag |= MEGASAS_FW_BUSY; + + spin_unlock_irqrestore(instance->host->host_lock, flags); + } + return EH_RESET_TIMER; +} + +/** * megasas_reset_device - Device reset handler entry point */ static int megasas_reset_device(struct scsi_cmnd *scmd) @@ -1112,6 +1146,7 @@ static struct scsi_host_template megasas_template = { .eh_device_reset_handler = megasas_reset_device, .eh_bus_reset_handler = megasas_reset_bus_host, .eh_host_reset_handler = megasas_reset_bus_host, + .eh_timed_out = megasas_reset_timer, .bios_param = megasas_bios_param, .use_clustering = ENABLE_CLUSTERING, }; @@ -1215,9 +1250,8 @@ megasas_complete_cmd(struct megasas_instance *instance, struct megasas_cmd *cmd, int exception = 0; struct megasas_header *hdr = &cmd->frame->hdr; - if (cmd->scmd) { - cmd->scmd->SCp.ptr = (char *)0; - } + if (cmd->scmd) + cmd->scmd->SCp.ptr = NULL; switch (hdr->cmd) { @@ -1806,6 +1840,7 @@ static void megasas_complete_cmd_dpc(unsigned long instance_addr) u32 context; struct megasas_cmd *cmd; struct megasas_instance *instance = (struct megasas_instance *)instance_addr; + unsigned long flags; /* If we have already declared adapter dead, donot complete cmds */ if (instance->hw_crit_error) @@ -1828,6 +1863,22 @@ static void megasas_complete_cmd_dpc(unsigned long instance_addr) } *instance->consumer = producer; + + /* + * Check if we can restore can_queue + */ + if (instance->flag & MEGASAS_FW_BUSY + && time_after(jiffies, instance->last_time + 5 * HZ) + && atomic_read(&instance->fw_outstanding) < 17) { + + spin_lock_irqsave(instance->host->host_lock, flags); + instance->flag &= ~MEGASAS_FW_BUSY; + instance->host->can_queue = + instance->max_fw_cmds - MEGASAS_INT_CMDS; + + spin_unlock_irqrestore(instance->host->host_lock, flags); + } + } /** @@ -2398,6 +2449,8 @@ megasas_probe_one(struct pci_dev *pdev, const struct pci_device_id *id) instance->init_id = MEGASAS_DEFAULT_INIT_ID; megasas_dbg_lvl = 0; + instance->flag = 0; + instance->last_time = 0; /* * Initialize MFI Firmware diff --git a/drivers/scsi/megaraid/megaraid_sas.h b/drivers/scsi/megaraid/megaraid_sas.h index e862992ee37..4dffc918a41 100644 --- a/drivers/scsi/megaraid/megaraid_sas.h +++ b/drivers/scsi/megaraid/megaraid_sas.h @@ -18,9 +18,9 @@ /* * MegaRAID SAS Driver meta data */ -#define MEGASAS_VERSION "00.00.03.10-rc1" -#define MEGASAS_RELDATE "Feb 14, 2007" -#define MEGASAS_EXT_VERSION "Wed Feb 14 10:14:25 PST 2007" +#define MEGASAS_VERSION "00.00.03.10-rc5" +#define MEGASAS_RELDATE "May 17, 2007" +#define MEGASAS_EXT_VERSION "Thu May 17 10:09:32 PDT 2007" /* * Device IDs @@ -539,6 +539,8 @@ struct megasas_ctrl_info { #define MEGASAS_DBG_LVL 1 +#define MEGASAS_FW_BUSY 1 + /* * When SCSI mid-layer calls driver's reset routine, driver waits for * MEGASAS_RESET_WAIT_TIME seconds for all outstanding IO to complete. Note @@ -549,8 +551,8 @@ struct megasas_ctrl_info { #define MEGASAS_RESET_WAIT_TIME 180 #define MEGASAS_INTERNAL_CMD_WAIT_TIME 180 #define MEGASAS_RESET_NOTICE_INTERVAL 5 - #define MEGASAS_IOCTL_CMD 0 +#define MEGASAS_DEFAULT_CMD_TIMEOUT 90 /* * FW reports the maximum of number of commands that it can accept (maximum @@ -1073,7 +1075,6 @@ struct megasas_instance { struct megasas_register_set __iomem *reg_set; s8 init_id; - u8 reserved[3]; u16 max_num_sge; u16 max_fw_cmds; @@ -1104,6 +1105,9 @@ struct megasas_instance { struct megasas_instance_template *instancet; struct tasklet_struct isr_tasklet; + + u8 flag; + unsigned long last_time; }; #define MEGASAS_IS_LOGICAL(scp) \ diff --git a/drivers/scsi/pluto.c b/drivers/scsi/pluto.c index 3b2e1a53e6e..d953d43fe2e 100644 --- a/drivers/scsi/pluto.c +++ b/drivers/scsi/pluto.c @@ -4,6 +4,7 @@ * */ +#include <linux/completion.h> #include <linux/kernel.h> #include <linux/delay.h> #include <linux/types.h> @@ -50,16 +51,10 @@ static struct ctrl_inquiry { } *fcs __initdata; static int fcscount __initdata = 0; static atomic_t fcss __initdata = ATOMIC_INIT(0); -DECLARE_MUTEX_LOCKED(fc_sem); +static DECLARE_COMPLETION(fc_detect_complete); static int pluto_encode_addr(Scsi_Cmnd *SCpnt, u16 *addr, fc_channel *fc, fcp_cmnd *fcmd); -static void __init pluto_detect_timeout(unsigned long data) -{ - PLND(("Timeout\n")) - up(&fc_sem); -} - static void __init pluto_detect_done(Scsi_Cmnd *SCpnt) { /* Do nothing */ @@ -69,7 +64,7 @@ static void __init pluto_detect_scsi_done(Scsi_Cmnd *SCpnt) { PLND(("Detect done %08lx\n", (long)SCpnt)) if (atomic_dec_and_test (&fcss)) - up(&fc_sem); + complete(&fc_detect_complete); } int pluto_slave_configure(struct scsi_device *device) @@ -96,7 +91,6 @@ int __init pluto_detect(struct scsi_host_template *tpnt) int i, retry, nplutos; fc_channel *fc; struct scsi_device dev; - DEFINE_TIMER(fc_timer, pluto_detect_timeout, 0, 0); tpnt->proc_name = "pluto"; fcscount = 0; @@ -187,15 +181,11 @@ int __init pluto_detect(struct scsi_host_template *tpnt) } } - fc_timer.expires = jiffies + 10 * HZ; - add_timer(&fc_timer); - - down(&fc_sem); + wait_for_completion_timeout(&fc_detect_complete, 10 * HZ); PLND(("Woken up\n")) if (!atomic_read(&fcss)) break; /* All fc channels have answered us */ } - del_timer_sync(&fc_timer); PLND(("Finished search\n")) for (i = 0, nplutos = 0; i < fcscount; i++) { diff --git a/drivers/scsi/scsi_devinfo.c b/drivers/scsi/scsi_devinfo.c index ce63044b1ec..18dd5cc4d7c 100644 --- a/drivers/scsi/scsi_devinfo.c +++ b/drivers/scsi/scsi_devinfo.c @@ -209,6 +209,7 @@ static struct { {"PIONEER", "CD-ROM DRM-602X", NULL, BLIST_FORCELUN | BLIST_SINGLELUN}, {"PIONEER", "CD-ROM DRM-604X", NULL, BLIST_FORCELUN | BLIST_SINGLELUN}, {"PIONEER", "CD-ROM DRM-624X", NULL, BLIST_FORCELUN | BLIST_SINGLELUN}, + {"Promise", "", NULL, BLIST_SPARSELUN}, {"REGAL", "CDC-4X", NULL, BLIST_MAX5LUN | BLIST_SINGLELUN}, {"SanDisk", "ImageMate CF-SD1", NULL, BLIST_FORCELUN}, {"SEAGATE", "ST34555N", "0930", BLIST_NOTQ}, /* Chokes on tagged INQUIRY */ diff --git a/drivers/scsi/sd.c b/drivers/scsi/sd.c index 00e46662296..3d8c9cb24f9 100644 --- a/drivers/scsi/sd.c +++ b/drivers/scsi/sd.c @@ -1789,7 +1789,7 @@ static void sd_shutdown(struct device *dev) static int sd_suspend(struct device *dev, pm_message_t mesg) { struct scsi_disk *sdkp = scsi_disk_get_from_dev(dev); - int ret; + int ret = 0; if (!sdkp) return 0; /* this can happen */ @@ -1798,30 +1798,34 @@ static int sd_suspend(struct device *dev, pm_message_t mesg) sd_printk(KERN_NOTICE, sdkp, "Synchronizing SCSI cache\n"); ret = sd_sync_cache(sdkp); if (ret) - return ret; + goto done; } if (mesg.event == PM_EVENT_SUSPEND && sdkp->device->manage_start_stop) { sd_printk(KERN_NOTICE, sdkp, "Stopping disk\n"); ret = sd_start_stop_device(sdkp, 0); - if (ret) - return ret; } - return 0; +done: + scsi_disk_put(sdkp); + return ret; } static int sd_resume(struct device *dev) { struct scsi_disk *sdkp = scsi_disk_get_from_dev(dev); + int ret = 0; if (!sdkp->device->manage_start_stop) - return 0; + goto done; sd_printk(KERN_NOTICE, sdkp, "Starting disk\n"); + ret = sd_start_stop_device(sdkp, 1); - return sd_start_stop_device(sdkp, 1); +done: + scsi_disk_put(sdkp); + return ret; } /** diff --git a/drivers/scsi/stex.c b/drivers/scsi/stex.c index 69be1324b11..9ac83abc402 100644 --- a/drivers/scsi/stex.c +++ b/drivers/scsi/stex.c @@ -32,11 +32,12 @@ #include <scsi/scsi_cmnd.h> #include <scsi/scsi_host.h> #include <scsi/scsi_tcq.h> +#include <scsi/scsi_dbg.h> #define DRV_NAME "stex" -#define ST_DRIVER_VERSION "3.1.0.1" +#define ST_DRIVER_VERSION "3.6.0000.1" #define ST_VER_MAJOR 3 -#define ST_VER_MINOR 1 +#define ST_VER_MINOR 6 #define ST_OEM 0 #define ST_BUILD_VER 1 @@ -113,10 +114,6 @@ enum { SG_CF_64B = 0x40, /* 64 bit item */ SG_CF_HOST = 0x20, /* sg in host memory */ - ST_MAX_ARRAY_SUPPORTED = 16, - ST_MAX_TARGET_NUM = (ST_MAX_ARRAY_SUPPORTED+1), - ST_MAX_LUN_PER_TARGET = 16, - st_shasta = 0, st_vsc = 1, st_vsc1 = 2, @@ -586,7 +583,7 @@ stex_queuecommand(struct scsi_cmnd *cmd, void (* done)(struct scsi_cmnd *)) u16 tag; host = cmd->device->host; id = cmd->device->id; - lun = cmd->device->channel; /* firmware lun issue work around */ + lun = cmd->device->lun; hba = (struct st_hba *) &host->hostdata[0]; switch (cmd->cmnd[0]) { @@ -605,8 +602,26 @@ stex_queuecommand(struct scsi_cmnd *cmd, void (* done)(struct scsi_cmnd *)) stex_invalid_field(cmd, done); return 0; } + case REPORT_LUNS: + /* + * The shasta firmware does not report actual luns in the + * target, so fail the command to force sequential lun scan. + * Also, the console device does not support this command. + */ + if (hba->cardtype == st_shasta || id == host->max_id - 1) { + stex_invalid_field(cmd, done); + return 0; + } + break; + case TEST_UNIT_READY: + if (id == host->max_id - 1) { + cmd->result = DID_OK << 16 | COMMAND_COMPLETE << 8; + done(cmd); + return 0; + } + break; case INQUIRY: - if (id != ST_MAX_ARRAY_SUPPORTED) + if (id != host->max_id - 1) break; if (lun == 0 && (cmd->cmnd[1] & INQUIRY_EVPD) == 0) { stex_direct_copy(cmd, console_inq_page, @@ -624,7 +639,7 @@ stex_queuecommand(struct scsi_cmnd *cmd, void (* done)(struct scsi_cmnd *)) ver.oem = ST_OEM; ver.build = ST_BUILD_VER; ver.signature[0] = PASSTHRU_SIGNATURE; - ver.console_id = ST_MAX_ARRAY_SUPPORTED; + ver.console_id = host->max_id - 1; ver.host_no = hba->host->host_no; cmd->result = stex_direct_copy(cmd, &ver, sizeof(ver)) ? DID_OK << 16 | COMMAND_COMPLETE << 8 : @@ -645,13 +660,8 @@ stex_queuecommand(struct scsi_cmnd *cmd, void (* done)(struct scsi_cmnd *)) req = stex_alloc_req(hba); - if (hba->cardtype == st_yosemite) { - req->lun = lun * (ST_MAX_TARGET_NUM - 1) + id; - req->target = 0; - } else { - req->lun = lun; - req->target = id; - } + req->lun = lun; + req->target = id; /* cdb */ memcpy(req->cdb, cmd->cmnd, STEX_CDB_LENGTH); @@ -767,18 +777,6 @@ static void stex_ys_commands(struct st_hba *hba, ccb->srb_status = SRB_STATUS_SELECTION_TIMEOUT; else ccb->srb_status = SRB_STATUS_SUCCESS; - } else if (ccb->cmd->cmnd[0] == REPORT_LUNS) { - u8 *report_lun_data = (u8 *)hba->copy_buffer; - - count = STEX_EXTRA_SIZE; - stex_internal_copy(ccb->cmd, report_lun_data, - &count, ccb->sg_count, ST_FROM_CMD); - if (report_lun_data[2] || report_lun_data[3]) { - report_lun_data[2] = 0x00; - report_lun_data[3] = 0x08; - stex_internal_copy(ccb->cmd, report_lun_data, - &count, ccb->sg_count, ST_TO_CMD); - } } } @@ -995,6 +993,11 @@ static int stex_abort(struct scsi_cmnd *cmd) u32 data; int result = SUCCESS; unsigned long flags; + + printk(KERN_INFO DRV_NAME + "(%s): aborting command\n", pci_name(hba->pdev)); + scsi_print_command(cmd); + base = hba->mmio_base; spin_lock_irqsave(host->host_lock, flags); if (tag < host->can_queue && hba->ccb[tag].cmd == cmd) @@ -1051,7 +1054,12 @@ static void stex_hard_reset(struct st_hba *hba) pci_read_config_byte(bus->self, PCI_BRIDGE_CONTROL, &pci_bctl); pci_bctl |= PCI_BRIDGE_CTL_BUS_RESET; pci_write_config_byte(bus->self, PCI_BRIDGE_CONTROL, pci_bctl); - msleep(1); + + /* + * 1 ms may be enough for 8-port controllers. But 16-port controllers + * require more time to finish bus reset. Use 100 ms here for safety + */ + msleep(100); pci_bctl &= ~PCI_BRIDGE_CTL_BUS_RESET; pci_write_config_byte(bus->self, PCI_BRIDGE_CONTROL, pci_bctl); @@ -1075,6 +1083,10 @@ static int stex_reset(struct scsi_cmnd *cmd) unsigned long before; hba = (struct st_hba *) &cmd->device->host->hostdata[0]; + printk(KERN_INFO DRV_NAME + "(%s): resetting host\n", pci_name(hba->pdev)); + scsi_print_command(cmd); + hba->mu_status = MU_STATE_RESETTING; if (hba->cardtype == st_shasta) @@ -1194,7 +1206,7 @@ stex_probe(struct pci_dev *pdev, const struct pci_device_id *id) goto out_scsi_host_put; } - hba->mmio_base = ioremap(pci_resource_start(pdev, 0), + hba->mmio_base = ioremap_nocache(pci_resource_start(pdev, 0), pci_resource_len(pdev, 0)); if ( !hba->mmio_base) { printk(KERN_ERR DRV_NAME "(%s): memory map failed\n", @@ -1229,12 +1241,18 @@ stex_probe(struct pci_dev *pdev, const struct pci_device_id *id) hba->copy_buffer = hba->dma_mem + MU_BUFFER_SIZE; hba->mu_status = MU_STATE_STARTING; - /* firmware uses id/lun pair for a logical drive, but lun would be - always 0 if CONFIG_SCSI_MULTI_LUN not configured, so we use - channel to map lun here */ - host->max_channel = ST_MAX_LUN_PER_TARGET - 1; - host->max_id = ST_MAX_TARGET_NUM; - host->max_lun = 1; + if (hba->cardtype == st_shasta) { + host->max_lun = 8; + host->max_id = 16 + 1; + } else if (hba->cardtype == st_yosemite) { + host->max_lun = 128; + host->max_id = 1 + 1; + } else { + /* st_vsc and st_vsc1 */ + host->max_lun = 1; + host->max_id = 128 + 1; + } + host->max_channel = 0; host->unique_id = host->host_no; host->max_cmd_len = STEX_CDB_LENGTH; diff --git a/drivers/serial/amba-pl010.c b/drivers/serial/amba-pl010.c index 1a9a24b8263..00d1255e4c1 100644 --- a/drivers/serial/amba-pl010.c +++ b/drivers/serial/amba-pl010.c @@ -167,8 +167,9 @@ static void pl010_rx_chars(struct uart_amba_port *uap) ignore_char: status = readb(uap->port.membase + UART01x_FR); } + spin_unlock(&port->lock); tty_flip_buffer_push(tty); - return; + spin_lock(&port->lock); } static void pl010_tx_chars(struct uart_amba_port *uap) diff --git a/drivers/serial/amba-pl011.c b/drivers/serial/amba-pl011.c index 44639e71372..954073c6ce3 100644 --- a/drivers/serial/amba-pl011.c +++ b/drivers/serial/amba-pl011.c @@ -153,8 +153,9 @@ static void pl011_rx_chars(struct uart_amba_port *uap) ignore_char: status = readw(uap->port.membase + UART01x_FR); } + spin_unlock(&uap->port.lock); tty_flip_buffer_push(tty); - return; + spin_lock(&uap->port.lock); } static void pl011_tx_chars(struct uart_amba_port *uap) diff --git a/drivers/serial/serial_ks8695.c b/drivers/serial/serial_ks8695.c index 698763b28dd..8721afe1ae4 100644 --- a/drivers/serial/serial_ks8695.c +++ b/drivers/serial/serial_ks8695.c @@ -589,7 +589,7 @@ static int __init ks8695_console_setup(struct console *co, char *options) return uart_set_options(port, co, baud, parity, bits, flow); } -extern struct uart_driver ks8695_reg; +static struct uart_driver ks8695_reg; static struct console ks8695_console = { .name = SERIAL_KS8695_DEVNAME, diff --git a/drivers/serial/suncore.c b/drivers/serial/suncore.c index e35d9ab359f..b45ba5392dd 100644 --- a/drivers/serial/suncore.c +++ b/drivers/serial/suncore.c @@ -30,9 +30,9 @@ void sunserial_console_termios(struct console *con) { char mode[16], buf[16], *s; - char *mode_prop = "ttyX-mode"; - char *cd_prop = "ttyX-ignore-cd"; - char *dtr_prop = "ttyX-rts-dtr-off"; + char mode_prop[] = "ttyX-mode"; + char cd_prop[] = "ttyX-ignore-cd"; + char dtr_prop[] = "ttyX-rts-dtr-off"; char *ssp_console_modes_prop = "ssp-console-modes"; int baud, bits, stop, cflag; char parity; diff --git a/drivers/serial/sunzilog.c b/drivers/serial/sunzilog.c index 0985193dc57..15b6e1cb040 100644 --- a/drivers/serial/sunzilog.c +++ b/drivers/serial/sunzilog.c @@ -1239,7 +1239,7 @@ static inline struct console *SUNZILOG_CONSOLE(void) #define SUNZILOG_CONSOLE() (NULL) #endif -static void __init sunzilog_init_kbdms(struct uart_sunzilog_port *up, int channel) +static void __devinit sunzilog_init_kbdms(struct uart_sunzilog_port *up, int channel) { int baud, brg; @@ -1259,7 +1259,7 @@ static void __init sunzilog_init_kbdms(struct uart_sunzilog_port *up, int channe } #ifdef CONFIG_SERIO -static void __init sunzilog_register_serio(struct uart_sunzilog_port *up) +static void __devinit sunzilog_register_serio(struct uart_sunzilog_port *up) { struct serio *serio = &up->serio; diff --git a/drivers/usb/core/hcd.c b/drivers/usb/core/hcd.c index e277258df38..8969e42434b 100644 --- a/drivers/usb/core/hcd.c +++ b/drivers/usb/core/hcd.c @@ -1681,7 +1681,7 @@ void usb_remove_hcd(struct usb_hcd *hcd) spin_unlock_irq (&hcd_root_hub_lock); #ifdef CONFIG_PM - flush_workqueue(ksuspend_usb_wq); + cancel_work_sync(&hcd->wakeup_work); #endif mutex_lock(&usb_bus_list_lock); diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c index caaa46f2dec..24f10a19dbd 100644 --- a/drivers/usb/core/hub.c +++ b/drivers/usb/core/hub.c @@ -1158,6 +1158,30 @@ static void release_address(struct usb_device *udev) } } +#ifdef CONFIG_USB_SUSPEND + +static void usb_stop_pm(struct usb_device *udev) +{ + /* Synchronize with the ksuspend thread to prevent any more + * autosuspend requests from being submitted, and decrement + * the parent's count of unsuspended children. + */ + usb_pm_lock(udev); + if (udev->parent && !udev->discon_suspended) + usb_autosuspend_device(udev->parent); + usb_pm_unlock(udev); + + /* Stop any autosuspend requests already submitted */ + cancel_rearming_delayed_work(&udev->autosuspend); +} + +#else + +static inline void usb_stop_pm(struct usb_device *udev) +{ } + +#endif + /** * usb_disconnect - disconnect a device (usbcore-internal) * @pdev: pointer to device being disconnected @@ -1224,13 +1248,7 @@ void usb_disconnect(struct usb_device **pdev) *pdev = NULL; spin_unlock_irq(&device_state_lock); - /* Decrement the parent's count of unsuspended children */ - if (udev->parent) { - usb_pm_lock(udev); - if (!udev->discon_suspended) - usb_autosuspend_device(udev->parent); - usb_pm_unlock(udev); - } + usb_stop_pm(udev); put_device(&udev->dev); } diff --git a/drivers/usb/core/usb.c b/drivers/usb/core/usb.c index 80627b6a2bf..4a6299bd004 100644 --- a/drivers/usb/core/usb.c +++ b/drivers/usb/core/usb.c @@ -184,10 +184,6 @@ static void usb_release_dev(struct device *dev) udev = to_usb_device(dev); -#ifdef CONFIG_USB_SUSPEND - cancel_delayed_work(&udev->autosuspend); - flush_workqueue(ksuspend_usb_wq); -#endif usb_destroy_configuration(udev); usb_put_hcd(bus_to_hcd(udev->bus)); kfree(udev->product); diff --git a/drivers/video/neofb.c b/drivers/video/neofb.c index bd30aba242d..731d7a5c5aa 100644 --- a/drivers/video/neofb.c +++ b/drivers/video/neofb.c @@ -1286,34 +1286,36 @@ static int neofb_setcolreg(u_int regno, u_int red, u_int green, u_int blue, if (regno >= fb->cmap.len || regno > 255) return -EINVAL; - switch (fb->var.bits_per_pixel) { - case 8: + if (fb->var.bits_per_pixel <= 8) { outb(regno, 0x3c8); outb(red >> 10, 0x3c9); outb(green >> 10, 0x3c9); outb(blue >> 10, 0x3c9); - break; - case 16: - ((u32 *) fb->pseudo_palette)[regno] = + } else if (regno < 16) { + switch (fb->var.bits_per_pixel) { + case 16: + ((u32 *) fb->pseudo_palette)[regno] = ((red & 0xf800)) | ((green & 0xfc00) >> 5) | ((blue & 0xf800) >> 11); - break; - case 24: - ((u32 *) fb->pseudo_palette)[regno] = + break; + case 24: + ((u32 *) fb->pseudo_palette)[regno] = ((red & 0xff00) << 8) | ((green & 0xff00)) | ((blue & 0xff00) >> 8); - break; + break; #ifdef NO_32BIT_SUPPORT_YET - case 32: - ((u32 *) fb->pseudo_palette)[regno] = + case 32: + ((u32 *) fb->pseudo_palette)[regno] = ((transp & 0xff00) << 16) | ((red & 0xff00) << 8) | ((green & 0xff00)) | ((blue & 0xff00) >> 8); - break; + break; #endif - default: - return 1; + default: + return 1; + } } + return 0; } diff --git a/fs/afs/internal.h b/fs/afs/internal.h index 2dac3ad2c44..2c55dd94a1d 100644 --- a/fs/afs/internal.h +++ b/fs/afs/internal.h @@ -17,6 +17,8 @@ #include <linux/rxrpc.h> #include <linux/key.h> #include <linux/workqueue.h> +#include <linux/sched.h> + #include "afs.h" #include "afs_vl.h" diff --git a/fs/compat_ioctl.c b/fs/compat_ioctl.c index 65643def318..6b44cdc96fa 100644 --- a/fs/compat_ioctl.c +++ b/fs/compat_ioctl.c @@ -1194,6 +1194,7 @@ static int vt_check(struct file *file) { struct tty_struct *tty; struct inode *inode = file->f_path.dentry->d_inode; + struct vc_data *vc; if (file->f_op->ioctl != tty_ioctl) return -EINVAL; @@ -1204,12 +1205,16 @@ static int vt_check(struct file *file) if (tty->driver->ioctl != vt_ioctl) return -EINVAL; - + + vc = (struct vc_data *)tty->driver_data; + if (!vc_cons_allocated(vc->vc_num)) /* impossible? */ + return -ENOIOCTLCMD; + /* * To have permissions to do most of the vt ioctls, we either have - * to be the owner of the tty, or super-user. + * to be the owner of the tty, or have CAP_SYS_TTY_CONFIG. */ - if (current->signal->tty == tty || capable(CAP_SYS_ADMIN)) + if (current->signal->tty == tty || capable(CAP_SYS_TTY_CONFIG)) return 1; return 0; } @@ -1310,16 +1315,28 @@ static int do_unimap_ioctl(unsigned int fd, unsigned int cmd, unsigned long arg, struct unimapdesc32 tmp; struct unimapdesc32 __user *user_ud = compat_ptr(arg); int perm = vt_check(file); - - if (perm < 0) return perm; + struct vc_data *vc; + + if (perm < 0) + return perm; if (copy_from_user(&tmp, user_ud, sizeof tmp)) return -EFAULT; + if (tmp.entries) + if (!access_ok(VERIFY_WRITE, compat_ptr(tmp.entries), + tmp.entry_ct*sizeof(struct unipair))) + return -EFAULT; + vc = ((struct tty_struct *)file->private_data)->driver_data; switch (cmd) { case PIO_UNIMAP: - if (!perm) return -EPERM; - return con_set_unimap(vc_cons[fg_console].d, tmp.entry_ct, compat_ptr(tmp.entries)); + if (!perm) + return -EPERM; + return con_set_unimap(vc, tmp.entry_ct, + compat_ptr(tmp.entries)); case GIO_UNIMAP: - return con_get_unimap(vc_cons[fg_console].d, tmp.entry_ct, &(user_ud->entry_ct), compat_ptr(tmp.entries)); + if (!perm && fg_console != vc->vc_num) + return -EPERM; + return con_get_unimap(vc, tmp.entry_ct, &(user_ud->entry_ct), + compat_ptr(tmp.entries)); } return 0; } diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c index 0c542ec92d5..00eee87510f 100644 --- a/fs/nfs/direct.c +++ b/fs/nfs/direct.c @@ -168,7 +168,7 @@ static inline struct nfs_direct_req *nfs_direct_req_alloc(void) return dreq; } -static void nfs_direct_req_release(struct kref *kref) +static void nfs_direct_req_free(struct kref *kref) { struct nfs_direct_req *dreq = container_of(kref, struct nfs_direct_req, kref); @@ -177,6 +177,11 @@ static void nfs_direct_req_release(struct kref *kref) kmem_cache_free(nfs_direct_cachep, dreq); } +static void nfs_direct_req_release(struct nfs_direct_req *dreq) +{ + kref_put(&dreq->kref, nfs_direct_req_free); +} + /* * Collects and returns the final error value/byte-count. */ @@ -196,7 +201,6 @@ static ssize_t nfs_direct_wait(struct nfs_direct_req *dreq) result = dreq->count; out: - kref_put(&dreq->kref, nfs_direct_req_release); return (ssize_t) result; } @@ -214,7 +218,7 @@ static void nfs_direct_complete(struct nfs_direct_req *dreq) } complete_all(&dreq->completion); - kref_put(&dreq->kref, nfs_direct_req_release); + nfs_direct_req_release(dreq); } /* @@ -369,6 +373,7 @@ static ssize_t nfs_direct_read(struct kiocb *iocb, unsigned long user_addr, size if (!result) result = nfs_direct_wait(dreq); rpc_clnt_sigunmask(clnt, &oldset); + nfs_direct_req_release(dreq); return result; } @@ -716,6 +721,7 @@ static ssize_t nfs_direct_write(struct kiocb *iocb, unsigned long user_addr, siz if (!result) result = nfs_direct_wait(dreq); rpc_clnt_sigunmask(clnt, &oldset); + nfs_direct_req_release(dreq); return result; } diff --git a/fs/ntfs/inode.c b/fs/ntfs/inode.c index 074791ce4ab..b532a730cec 100644 --- a/fs/ntfs/inode.c +++ b/fs/ntfs/inode.c @@ -140,7 +140,7 @@ static int ntfs_init_locked_inode(struct inode *vi, ntfs_attr *na) if (!ni->name) return -ENOMEM; memcpy(ni->name, na->name, i); - ni->name[i] = 0; + ni->name[na->name_len] = 0; } return 0; } diff --git a/fs/ocfs2/aops.c b/fs/ocfs2/aops.c index 8e7cafb5fc6..0023b31e48a 100644 --- a/fs/ocfs2/aops.c +++ b/fs/ocfs2/aops.c @@ -222,7 +222,10 @@ static int ocfs2_readpage(struct file *file, struct page *page) goto out; } - down_read(&OCFS2_I(inode)->ip_alloc_sem); + if (down_read_trylock(&OCFS2_I(inode)->ip_alloc_sem) == 0) { + ret = AOP_TRUNCATED_PAGE; + goto out_meta_unlock; + } /* * i_size might have just been updated as we grabed the meta lock. We @@ -235,10 +238,7 @@ static int ocfs2_readpage(struct file *file, struct page *page) * XXX sys_readahead() seems to get that wrong? */ if (start >= i_size_read(inode)) { - char *addr = kmap(page); - memset(addr, 0, PAGE_SIZE); - flush_dcache_page(page); - kunmap(page); + zero_user_page(page, 0, PAGE_SIZE, KM_USER0); SetPageUptodate(page); ret = 0; goto out_alloc; @@ -258,6 +258,7 @@ static int ocfs2_readpage(struct file *file, struct page *page) ocfs2_data_unlock(inode, 0); out_alloc: up_read(&OCFS2_I(inode)->ip_alloc_sem); +out_meta_unlock: ocfs2_meta_unlock(inode, 0); out: if (unlock) diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c index 9395b4fa547..ac6c96431bb 100644 --- a/fs/ocfs2/file.c +++ b/fs/ocfs2/file.c @@ -326,6 +326,7 @@ static int ocfs2_truncate_file(struct inode *inode, (unsigned long long)OCFS2_I(inode)->ip_blkno, (unsigned long long)new_i_size); + unmap_mapping_range(inode->i_mapping, new_i_size + PAGE_SIZE - 1, 0, 1); truncate_inode_pages(inode->i_mapping, new_i_size); fe = (struct ocfs2_dinode *) di_bh->b_data; @@ -1418,36 +1419,6 @@ out: return total ? total : ret; } -static int ocfs2_check_iovec(const struct iovec *iov, size_t *counted, - unsigned long *nr_segs) -{ - size_t ocount; /* original count */ - unsigned long seg; - - ocount = 0; - for (seg = 0; seg < *nr_segs; seg++) { - const struct iovec *iv = &iov[seg]; - - /* - * If any segment has a negative length, or the cumulative - * length ever wraps negative then return -EINVAL. - */ - ocount += iv->iov_len; - if (unlikely((ssize_t)(ocount|iv->iov_len) < 0)) - return -EINVAL; - if (access_ok(VERIFY_READ, iv->iov_base, iv->iov_len)) - continue; - if (seg == 0) - return -EFAULT; - *nr_segs = seg; - ocount -= iv->iov_len; /* This segment is no good */ - break; - } - - *counted = ocount; - return 0; -} - static ssize_t ocfs2_file_aio_write(struct kiocb *iocb, const struct iovec *iov, unsigned long nr_segs, @@ -1470,7 +1441,7 @@ static ssize_t ocfs2_file_aio_write(struct kiocb *iocb, if (iocb->ki_left == 0) return 0; - ret = ocfs2_check_iovec(iov, &ocount, &nr_segs); + ret = generic_segment_checks(iov, &nr_segs, &ocount, VERIFY_READ); if (ret) return ret; diff --git a/fs/ocfs2/localalloc.c b/fs/ocfs2/localalloc.c index 4dedd978910..545f7892cdf 100644 --- a/fs/ocfs2/localalloc.c +++ b/fs/ocfs2/localalloc.c @@ -471,9 +471,6 @@ int ocfs2_reserve_local_alloc_bits(struct ocfs2_super *osb, mutex_lock(&local_alloc_inode->i_mutex); - ac->ac_inode = local_alloc_inode; - ac->ac_which = OCFS2_AC_USE_LOCAL; - if (osb->local_alloc_state != OCFS2_LA_ENABLED) { status = -ENOSPC; goto bail; @@ -511,10 +508,14 @@ int ocfs2_reserve_local_alloc_bits(struct ocfs2_super *osb, } } + ac->ac_inode = local_alloc_inode; + ac->ac_which = OCFS2_AC_USE_LOCAL; get_bh(osb->local_alloc_bh); ac->ac_bh = osb->local_alloc_bh; status = 0; bail: + if (status < 0 && local_alloc_inode) + iput(local_alloc_inode); mlog_exit(status); return status; diff --git a/fs/ramfs/file-nommu.c b/fs/ramfs/file-nommu.c index 3b481d557ed..9345a46ffb3 100644 --- a/fs/ramfs/file-nommu.c +++ b/fs/ramfs/file-nommu.c @@ -179,7 +179,7 @@ static int ramfs_nommu_resize(struct inode *inode, loff_t newsize, loff_t size) return ret; } - ret = vmtruncate(inode, size); + ret = vmtruncate(inode, newsize); return ret; } diff --git a/fs/xfs/linux-2.6/xfs_aops.c b/fs/xfs/linux-2.6/xfs_aops.c index 4475588e973..7361861e3aa 100644 --- a/fs/xfs/linux-2.6/xfs_aops.c +++ b/fs/xfs/linux-2.6/xfs_aops.c @@ -701,7 +701,7 @@ xfs_is_delayed_page( else if (buffer_delay(bh)) acceptable = (type == IOMAP_DELAY); else if (buffer_dirty(bh) && buffer_mapped(bh)) - acceptable = (type == 0); + acceptable = (type == IOMAP_NEW); else break; } while ((bh = bh->b_this_page) != head); @@ -810,7 +810,7 @@ xfs_convert_page( page_dirty--; count++; } else { - type = 0; + type = IOMAP_NEW; if (buffer_mapped(bh) && all_bh && startio) { lock_buffer(bh); xfs_add_to_ioend(inode, bh, offset, @@ -968,8 +968,8 @@ xfs_page_state_convert( bh = head = page_buffers(page); offset = page_offset(page); - flags = -1; - type = IOMAP_READ; + flags = BMAPI_READ; + type = IOMAP_NEW; /* TODO: cleanup count and page_dirty */ @@ -999,14 +999,14 @@ xfs_page_state_convert( * * Third case, an unmapped buffer was found, and we are * in a path where we need to write the whole page out. - */ + */ if (buffer_unwritten(bh) || buffer_delay(bh) || ((buffer_uptodate(bh) || PageUptodate(page)) && !buffer_mapped(bh) && (unmapped || startio))) { - /* + /* * Make sure we don't use a read-only iomap */ - if (flags == BMAPI_READ) + if (flags == BMAPI_READ) iomap_valid = 0; if (buffer_unwritten(bh)) { @@ -1055,7 +1055,7 @@ xfs_page_state_convert( * That means it must already have extents allocated * underneath it. Map the extent by reading it. */ - if (!iomap_valid || type != IOMAP_READ) { + if (!iomap_valid || flags != BMAPI_READ) { flags = BMAPI_READ; size = xfs_probe_cluster(inode, page, bh, head, 1); @@ -1066,7 +1066,15 @@ xfs_page_state_convert( iomap_valid = xfs_iomap_valid(&iomap, offset); } - type = IOMAP_READ; + /* + * We set the type to IOMAP_NEW in case we are doing a + * small write at EOF that is extending the file but + * without needing an allocation. We need to update the + * file size on I/O completion in this case so it is + * the same case as having just allocated a new extent + * that we are writing into for the first time. + */ + type = IOMAP_NEW; if (!test_and_set_bit(BH_Lock, &bh->b_state)) { ASSERT(buffer_mapped(bh)); if (iomap_valid) diff --git a/include/asm-alpha/bitops.h b/include/asm-alpha/bitops.h index 4b6ef7f21b9..3a0cbeb03fa 100644 --- a/include/asm-alpha/bitops.h +++ b/include/asm-alpha/bitops.h @@ -313,32 +313,29 @@ static inline int ffs(int word) * fls: find last bit set. */ #if defined(CONFIG_ALPHA_EV6) && defined(CONFIG_ALPHA_EV67) -static inline int fls(int word) +static inline int fls64(unsigned long word) { - return 64 - __kernel_ctlz(word & 0xffffffff); + return 64 - __kernel_ctlz(word); } #else -#include <asm-generic/bitops/fls.h> -#endif -#include <asm-generic/bitops/fls64.h> +extern const unsigned char __flsm1_tab[256]; -/* Compute powers of two for the given integer. */ -static inline long floor_log2(unsigned long word) +static inline int fls64(unsigned long x) { -#if defined(CONFIG_ALPHA_EV6) && defined(CONFIG_ALPHA_EV67) - return 63 - __kernel_ctlz(word); -#else - long bit; - for (bit = -1; word ; bit++) - word >>= 1; - return bit; -#endif + unsigned long t, a, r; + + t = __kernel_cmpbge (x, 0x0101010101010101); + a = __flsm1_tab[t]; + t = __kernel_extbl (x, a); + r = a*8 + __flsm1_tab[t] + (x != 0); + + return r; } +#endif -static inline long ceil_log2(unsigned long word) +static inline int fls(int x) { - long bit = floor_log2(word); - return bit + (word > (1UL << bit)); + return fls64((unsigned int) x); } /* @@ -353,9 +350,20 @@ static inline unsigned long hweight64(unsigned long w) return __kernel_ctpop(w); } -#define hweight32(x) (unsigned int) hweight64((x) & 0xfffffffful) -#define hweight16(x) (unsigned int) hweight64((x) & 0xfffful) -#define hweight8(x) (unsigned int) hweight64((x) & 0xfful) +static inline unsigned int hweight32(unsigned int w) +{ + return hweight64(w); +} + +static inline unsigned int hweight16(unsigned int w) +{ + return hweight64(w & 0xffff); +} + +static inline unsigned int hweight8(unsigned int w) +{ + return hweight64(w & 0xff); +} #else #include <asm-generic/bitops/hweight.h> #endif diff --git a/include/asm-alpha/thread_info.h b/include/asm-alpha/thread_info.h index f4defc2bd3f..48a22e3e6f3 100644 --- a/include/asm-alpha/thread_info.h +++ b/include/asm-alpha/thread_info.h @@ -76,12 +76,14 @@ register struct thread_info *__current_thread_info __asm__("$8"); #define TIF_UAC_NOFIX 7 #define TIF_UAC_SIGBUS 8 #define TIF_MEMDIE 9 +#define TIF_RESTORE_SIGMASK 10 /* restore signal mask in do_signal */ #define _TIF_SYSCALL_TRACE (1<<TIF_SYSCALL_TRACE) #define _TIF_NOTIFY_RESUME (1<<TIF_NOTIFY_RESUME) #define _TIF_SIGPENDING (1<<TIF_SIGPENDING) #define _TIF_NEED_RESCHED (1<<TIF_NEED_RESCHED) #define _TIF_POLLING_NRFLAG (1<<TIF_POLLING_NRFLAG) +#define _TIF_RESTORE_SIGMASK (1<<TIF_RESTORE_SIGMASK) /* Work to do on interrupt/exception return. */ #define _TIF_WORK_MASK (_TIF_NOTIFY_RESUME \ diff --git a/include/asm-alpha/unistd.h b/include/asm-alpha/unistd.h index e58a427012d..29bf2fdc91c 100644 --- a/include/asm-alpha/unistd.h +++ b/include/asm-alpha/unistd.h @@ -233,6 +233,20 @@ #define __NR_osf_memcntl 260 /* not implemented */ #define __NR_osf_fdatasync 261 /* not implemented */ +/* + * Ignore legacy syscalls that we don't use. + */ +#define __IGNORE_alarm +#define __IGNORE_creat +#define __IGNORE_getegid +#define __IGNORE_geteuid +#define __IGNORE_getgid +#define __IGNORE_getpid +#define __IGNORE_getppid +#define __IGNORE_getuid +#define __IGNORE_pause +#define __IGNORE_time +#define __IGNORE_utime /* * Linux-specific system calls begin at 300 @@ -387,10 +401,42 @@ #define __NR_inotify_init 444 #define __NR_inotify_add_watch 445 #define __NR_inotify_rm_watch 446 +#define __NR_fdatasync 447 +#define __NR_kexec_load 448 +#define __NR_migrate_pages 449 +#define __NR_openat 450 +#define __NR_mkdirat 451 +#define __NR_mknodat 452 +#define __NR_fchownat 453 +#define __NR_futimesat 454 +#define __NR_fstatat64 455 +#define __NR_unlinkat 456 +#define __NR_renameat 457 +#define __NR_linkat 458 +#define __NR_symlinkat 459 +#define __NR_readlinkat 460 +#define __NR_fchmodat 461 +#define __NR_faccessat 462 +#define __NR_pselect6 463 +#define __NR_ppoll 464 +#define __NR_unshare 465 +#define __NR_set_robust_list 466 +#define __NR_get_robust_list 467 +#define __NR_splice 468 +#define __NR_sync_file_range 469 +#define __NR_tee 470 +#define __NR_vmsplice 471 +#define __NR_move_pages 472 +#define __NR_getcpu 473 +#define __NR_epoll_pwait 474 +#define __NR_utimensat 475 +#define __NR_signalfd 476 +#define __NR_timerfd 477 +#define __NR_eventfd 478 #ifdef __KERNEL__ -#define NR_SYSCALLS 447 +#define NR_SYSCALLS 479 #define __ARCH_WANT_IPC_PARSE_VERSION #define __ARCH_WANT_OLD_READDIR diff --git a/include/asm-arm/arch-ixp4xx/nas100d.h b/include/asm-arm/arch-ixp4xx/nas100d.h index 84467a5190d..131e0a1d0df 100644 --- a/include/asm-arm/arch-ixp4xx/nas100d.h +++ b/include/asm-arm/arch-ixp4xx/nas100d.h @@ -10,7 +10,7 @@ * based on ixdp425.h: * Copyright 2004 (c) MontaVista, Software, Inc. * - * This file is licensed under the terms of the GNU General Public + * This file is licensed under the terms of the GNU General Public * License version 2. This program is licensed "as is" without any * warranty of any kind, whether express or implied. */ @@ -36,31 +36,11 @@ #define NAS100D_PCI_INTD_PIN 8 #define NAS100D_PCI_INTE_PIN 7 -/* GPIO */ - -#define NAS100D_GPIO0 0 -#define NAS100D_GPIO1 1 -#define NAS100D_GPIO2 2 -#define NAS100D_GPIO3 3 -#define NAS100D_GPIO4 4 -#define NAS100D_GPIO5 5 -#define NAS100D_GPIO6 6 -#define NAS100D_GPIO7 7 -#define NAS100D_GPIO8 8 -#define NAS100D_GPIO9 9 -#define NAS100D_GPIO10 10 -#define NAS100D_GPIO11 11 -#define NAS100D_GPIO12 12 -#define NAS100D_GPIO13 13 -#define NAS100D_GPIO14 14 -#define NAS100D_GPIO15 15 - - /* Buttons */ -#define NAS100D_PB_GPIO NAS100D_GPIO14 -#define NAS100D_RB_GPIO NAS100D_GPIO4 -#define NAS100D_PO_GPIO NAS100D_GPIO12 /* power off */ +#define NAS100D_PB_GPIO 14 +#define NAS100D_RB_GPIO 4 +#define NAS100D_PO_GPIO 12 /* power off */ #define NAS100D_PB_IRQ IRQ_IXP4XX_GPIO14 #define NAS100D_RB_IRQ IRQ_IXP4XX_GPIO4 diff --git a/include/asm-arm/arch-ixp4xx/nslu2.h b/include/asm-arm/arch-ixp4xx/nslu2.h index 6b437f7c995..850fdc5b45d 100644 --- a/include/asm-arm/arch-ixp4xx/nslu2.h +++ b/include/asm-arm/arch-ixp4xx/nslu2.h @@ -9,7 +9,7 @@ * based on ixdp425.h: * Copyright 2004 (c) MontaVista, Software, Inc. * - * This file is licensed under the terms of the GNU General Public + * This file is licensed under the terms of the GNU General Public * License version 2. This program is licensed "as is" without any * warranty of any kind, whether express or implied. */ @@ -34,36 +34,14 @@ #define NSLU2_PCI_INTC_PIN 9 #define NSLU2_PCI_INTD_PIN 8 - /* NSLU2 Timer */ #define NSLU2_FREQ 66000000 -#define NSLU2_CLOCK_TICK_RATE (((NSLU2_FREQ / HZ & ~IXP4XX_OST_RELOAD_MASK) + 1) * HZ) -#define NSLU2_CLOCK_TICKS_PER_USEC ((NSLU2_CLOCK_TICK_RATE + USEC_PER_SEC/2) / USEC_PER_SEC) - -/* GPIO */ - -#define NSLU2_GPIO0 0 -#define NSLU2_GPIO1 1 -#define NSLU2_GPIO2 2 -#define NSLU2_GPIO3 3 -#define NSLU2_GPIO4 4 -#define NSLU2_GPIO5 5 -#define NSLU2_GPIO6 6 -#define NSLU2_GPIO7 7 -#define NSLU2_GPIO8 8 -#define NSLU2_GPIO9 9 -#define NSLU2_GPIO10 10 -#define NSLU2_GPIO11 11 -#define NSLU2_GPIO12 12 -#define NSLU2_GPIO13 13 -#define NSLU2_GPIO14 14 -#define NSLU2_GPIO15 15 /* Buttons */ -#define NSLU2_PB_GPIO NSLU2_GPIO5 -#define NSLU2_PO_GPIO NSLU2_GPIO8 /* power off */ -#define NSLU2_RB_GPIO NSLU2_GPIO12 +#define NSLU2_PB_GPIO 5 +#define NSLU2_PO_GPIO 8 /* power off */ +#define NSLU2_RB_GPIO 12 #define NSLU2_PB_IRQ IRQ_IXP4XX_GPIO5 #define NSLU2_RB_IRQ IRQ_IXP4XX_GPIO12 @@ -79,16 +57,16 @@ /* LEDs */ -#define NSLU2_LED_RED NSLU2_GPIO0 -#define NSLU2_LED_GRN NSLU2_GPIO1 +#define NSLU2_LED_RED_GPIO 0 +#define NSLU2_LED_GRN_GPIO 1 -#define NSLU2_LED_RED_BM (1L << NSLU2_LED_RED) -#define NSLU2_LED_GRN_BM (1L << NSLU2_LED_GRN) +#define NSLU2_LED_RED_BM (1L << NSLU2_LED_RED_GPIO) +#define NSLU2_LED_GRN_BM (1L << NSLU2_LED_GRN_GPIO) -#define NSLU2_LED_DISK1 NSLU2_GPIO3 -#define NSLU2_LED_DISK2 NSLU2_GPIO2 +#define NSLU2_LED_DISK1_GPIO 3 +#define NSLU2_LED_DISK2_GPIO 2 -#define NSLU2_LED_DISK1_BM (1L << NSLU2_GPIO2) -#define NSLU2_LED_DISK2_BM (1L << NSLU2_GPIO3) +#define NSLU2_LED_DISK1_BM (1L << NSLU2_LED_DISK1_GPIO) +#define NSLU2_LED_DISK2_BM (1L << NSLU2_LED_DISK2_GPIO) diff --git a/include/asm-arm/arch-ixp4xx/platform.h b/include/asm-arm/arch-ixp4xx/platform.h index ab194e5f665..2a44d3d6798 100644 --- a/include/asm-arm/arch-ixp4xx/platform.h +++ b/include/asm-arm/arch-ixp4xx/platform.h @@ -113,6 +113,7 @@ extern unsigned long ixp4xx_timer_freq; extern void ixp4xx_map_io(void); extern void ixp4xx_init_irq(void); extern void ixp4xx_sys_init(void); +extern void ixp4xx_timer_init(void); extern struct sys_timer ixp4xx_timer; extern void ixp4xx_pci_preinit(void); struct pci_sys_data; diff --git a/include/asm-arm/arch-s3c2410/map.h b/include/asm-arm/arch-s3c2410/map.h index 4505aefbad1..19e77f03804 100644 --- a/include/asm-arm/arch-s3c2410/map.h +++ b/include/asm-arm/arch-s3c2410/map.h @@ -153,6 +153,10 @@ #define S3C2440_PA_AC97 (0x5B000000) #define S3C2440_SZ_AC97 SZ_1M +/* S3C2443 High-speed SD/MMC */ +#define S3C2443_PA_HSMMC (0x4A800000) +#define S3C2443_SZ_HSMMC (256) + /* ISA style IO, for each machine to sort out mappings for, if it * implements it. We reserve two 16M regions for ISA. */ diff --git a/include/asm-arm/arch-s3c2410/regs-gpioj.h b/include/asm-arm/arch-s3c2410/regs-gpioj.h index 02131a5a1d3..0362332faaf 100644 --- a/include/asm-arm/arch-s3c2410/regs-gpioj.h +++ b/include/asm-arm/arch-s3c2410/regs-gpioj.h @@ -98,5 +98,9 @@ #define S3C2440_GPJ12_OUTP (0x01 << 24) #define S3C2440_GPJ12_CAMRESET (0x02 << 24) +#define S3C2443_GPJ13 S3C2410_GPIONO(S3C2440_GPIO_BANKJ, 13) +#define S3C2443_GPJ14 S3C2410_GPIONO(S3C2440_GPIO_BANKJ, 14) +#define S3C2443_GPJ15 S3C2410_GPIONO(S3C2440_GPIO_BANKJ, 15) + #endif /* __ASM_ARCH_REGS_GPIOJ_H */ diff --git a/include/asm-arm/arch-s3c2410/regs-s3c2412.h b/include/asm-arm/arch-s3c2410/regs-s3c2412.h new file mode 100644 index 00000000000..8ca6a3bc855 --- /dev/null +++ b/include/asm-arm/arch-s3c2410/regs-s3c2412.h @@ -0,0 +1,21 @@ +/* linux/include/asm-arm/arch-s3c2410/regs-s3c2412.h + * + * Copyright 2007 Simtec Electronics + * http://armlinux.simtec.co.uk/ + * Ben Dooks <ben@simtec.co.uk> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * S3C2412 specific register definitions +*/ + +#ifndef __ASM_ARCH_REGS_S3C2412_H +#define __ASM_ARCH_REGS_S3C2412_H "s3c2412" + +#define S3C2412_SWRST (S3C24XX_VA_CLKPWR + 0x30) +#define S3C2412_SWRST_RESET (0x533C2412) + +#endif /* __ASM_ARCH_REGS_S3C2412_H */ + diff --git a/include/asm-arm/ioctls.h b/include/asm-arm/ioctls.h index bb9a7aa10c1..a91d8a1523c 100644 --- a/include/asm-arm/ioctls.h +++ b/include/asm-arm/ioctls.h @@ -46,6 +46,10 @@ #define TIOCSBRK 0x5427 /* BSD compatibility */ #define TIOCCBRK 0x5428 /* BSD compatibility */ #define TIOCGSID 0x5429 /* Return the session ID of FD */ +#define TCGETS2 _IOR('T',0x2A, struct termios2) +#define TCSETS2 _IOW('T',0x2B, struct termios2) +#define TCSETSW2 _IOW('T',0x2C, struct termios2) +#define TCSETSF2 _IOW('T',0x2D, struct termios2) #define TIOCGPTN _IOR('T',0x30, unsigned int) /* Get Pty Number (of pty-mux device) */ #define TIOCSPTLCK _IOW('T',0x31, int) /* Lock/unlock Pty */ diff --git a/include/asm-arm/mach/arch.h b/include/asm-arm/mach/arch.h index fd2f9bf4dcc..c59fad18e73 100644 --- a/include/asm-arm/mach/arch.h +++ b/include/asm-arm/mach/arch.h @@ -49,7 +49,7 @@ struct machine_desc { */ #define MACHINE_START(_type,_name) \ static const struct machine_desc __mach_desc_##_type \ - __attribute_used__ \ + __used \ __attribute__((__section__(".arch.info.init"))) = { \ .nr = MACH_TYPE_##_type, \ .name = _name, diff --git a/include/asm-arm/setup.h b/include/asm-arm/setup.h index e5407392afc..7bbf105463f 100644 --- a/include/asm-arm/setup.h +++ b/include/asm-arm/setup.h @@ -185,7 +185,7 @@ struct tagtable { #ifdef __KERNEL__ -#define __tag __attribute_used__ __attribute__((__section__(".taglist.init"))) +#define __tag __used __attribute__((__section__(".taglist.init"))) #define __tagtable(tag, fn) \ static struct tagtable __tagtable_##fn __tag = { tag, fn } @@ -218,7 +218,7 @@ struct early_params { }; #define __early_param(name,fn) \ -static struct early_params __early_##fn __attribute_used__ \ +static struct early_params __early_##fn __used \ __attribute__((__section__(".early_param.init"))) = { name, fn } #endif /* __KERNEL__ */ diff --git a/include/asm-arm/termbits.h b/include/asm-arm/termbits.h index a3f4fe1742d..f784d11f40b 100644 --- a/include/asm-arm/termbits.h +++ b/include/asm-arm/termbits.h @@ -15,6 +15,17 @@ struct termios { cc_t c_cc[NCCS]; /* control characters */ }; +struct termios2 { + tcflag_t c_iflag; /* input mode flags */ + tcflag_t c_oflag; /* output mode flags */ + tcflag_t c_cflag; /* control mode flags */ + tcflag_t c_lflag; /* local mode flags */ + cc_t c_line; /* line discipline */ + cc_t c_cc[NCCS]; /* control characters */ + speed_t c_ispeed; /* input speed */ + speed_t c_ospeed; /* output speed */ +}; + struct ktermios { tcflag_t c_iflag; /* input mode flags */ tcflag_t c_oflag; /* output mode flags */ @@ -128,6 +139,7 @@ struct ktermios { #define HUPCL 0002000 #define CLOCAL 0004000 #define CBAUDEX 0010000 +#define BOTHER 0010000 #define B57600 0010001 #define B115200 0010002 #define B230400 0010003 @@ -143,10 +155,12 @@ struct ktermios { #define B3000000 0010015 #define B3500000 0010016 #define B4000000 0010017 -#define CIBAUD 002003600000 /* input baud rate (not used) */ +#define CIBAUD 002003600000 /* input baud rate */ #define CMSPAR 010000000000 /* mark or space (stick) parity */ #define CRTSCTS 020000000000 /* flow control */ +#define IBSHIFT 16 + /* c_lflag bits */ #define ISIG 0000001 #define ICANON 0000002 diff --git a/include/asm-arm/termios.h b/include/asm-arm/termios.h index 329c324c404..293e3f1bc3f 100644 --- a/include/asm-arm/termios.h +++ b/include/asm-arm/termios.h @@ -82,8 +82,10 @@ struct termio { copy_to_user((termio)->c_cc, (termios)->c_cc, NCC); \ }) -#define user_termios_to_kernel_termios(k, u) copy_from_user(k, u, sizeof(struct termios)) -#define kernel_termios_to_user_termios(u, k) copy_to_user(u, k, sizeof(struct termios)) +#define user_termios_to_kernel_termios(k, u) copy_from_user(k, u, sizeof(struct termios2)) +#define kernel_termios_to_user_termios(u, k) copy_to_user(u, k, sizeof(struct termios2)) +#define user_termios_to_kernel_termios_1(k, u) copy_from_user(k, u, sizeof(struct termios)) +#define kernel_termios_to_user_termios_1(u, k) copy_to_user(u, k, sizeof(struct termios)) #endif /* __KERNEL__ */ diff --git a/include/asm-arm/tlbflush.h b/include/asm-arm/tlbflush.h index ccd0de010e8..71be4fded7e 100644 --- a/include/asm-arm/tlbflush.h +++ b/include/asm-arm/tlbflush.h @@ -138,6 +138,19 @@ # define v6wbi_always_flags (-1UL) #endif +#ifdef CONFIG_CPU_TLB_V7 +# define v7wbi_possible_flags v6wbi_tlb_flags +# define v7wbi_always_flags v6wbi_tlb_flags +# ifdef _TLB +# define MULTI_TLB 1 +# else +# define _TLB v7wbi +# endif +#else +# define v7wbi_possible_flags 0 +# define v7wbi_always_flags (-1UL) +#endif + #ifndef _TLB #error Unknown TLB model #endif diff --git a/include/asm-arm26/setup.h b/include/asm-arm26/setup.h index 10fd07c7666..e8256230647 100644 --- a/include/asm-arm26/setup.h +++ b/include/asm-arm26/setup.h @@ -173,7 +173,7 @@ struct tagtable { int (*parse)(const struct tag *); }; -#define __tag __attribute_used__ __attribute__((__section__(".taglist"))) +#define __tag __used __attribute__((__section__(".taglist"))) #define __tagtable(tag, fn) \ static struct tagtable __tagtable_##fn __tag = { tag, fn } diff --git a/include/asm-m68k/mmzone.h b/include/asm-m68k/mmzone.h new file mode 100644 index 00000000000..e1f1ec7b700 --- /dev/null +++ b/include/asm-m68k/mmzone.h @@ -0,0 +1,9 @@ +#ifndef _ASM_M68K_MMZONE_H_ +#define _ASM_M68K_MMZONE_H_ + +extern pg_data_t pg_data_map[]; + +#define NODE_DATA(nid) (&pg_data_map[nid]) +#define NODE_MEM_MAP(nid) (NODE_DATA(nid)->node_mem_map) + +#endif /* _ASM_M68K_MMZONE_H_ */ diff --git a/include/asm-m68k/module.h b/include/asm-m68k/module.h index c6d75af2d8d..382d20a6fc1 100644 --- a/include/asm-m68k/module.h +++ b/include/asm-m68k/module.h @@ -1,7 +1,39 @@ #ifndef _ASM_M68K_MODULE_H #define _ASM_M68K_MODULE_H -struct mod_arch_specific { }; + +struct mod_arch_specific { + struct m68k_fixup_info *fixup_start, *fixup_end; +}; + +#define MODULE_ARCH_INIT { \ + .fixup_start = __start_fixup, \ + .fixup_end = __stop_fixup, \ +} + #define Elf_Shdr Elf32_Shdr #define Elf_Sym Elf32_Sym #define Elf_Ehdr Elf32_Ehdr + + +enum m68k_fixup_type { + m68k_fixup_memoffset, + m68k_fixup_vnode_shift, +}; + +struct m68k_fixup_info { + enum m68k_fixup_type type; + void *addr; +}; + +#define m68k_fixup(type, addr) \ + " .section \".m68k_fixup\",\"aw\"\n" \ + " .long " #type "," #addr "\n" \ + " .previous\n" + +extern struct m68k_fixup_info __start_fixup[], __stop_fixup[]; + +struct module; +extern void module_fixup(struct module *mod, struct m68k_fixup_info *start, + struct m68k_fixup_info *end); + #endif /* _ASM_M68K_MODULE_H */ diff --git a/include/asm-m68k/motorola_pgtable.h b/include/asm-m68k/motorola_pgtable.h index 61e4406ed96..b5b78c01eb6 100644 --- a/include/asm-m68k/motorola_pgtable.h +++ b/include/asm-m68k/motorola_pgtable.h @@ -130,7 +130,7 @@ static inline void pgd_set(pgd_t *pgdp, pmd_t *pmdp) #define pte_present(pte) (pte_val(pte) & (_PAGE_PRESENT | _PAGE_PROTNONE)) #define pte_clear(mm,addr,ptep) ({ pte_val(*(ptep)) = 0; }) -#define pte_page(pte) (mem_map + ((unsigned long)(__va(pte_val(pte)) - PAGE_OFFSET) >> PAGE_SHIFT)) +#define pte_page(pte) virt_to_page(__va(pte_val(pte))) #define pte_pfn(pte) (pte_val(pte) >> PAGE_SHIFT) #define pfn_pte(pfn, prot) __pte(((pfn) << PAGE_SHIFT) | pgprot_val(prot)) @@ -143,7 +143,7 @@ static inline void pgd_set(pgd_t *pgdp, pmd_t *pmdp) while (--__i >= 0) \ *__ptr++ = 0; \ }) -#define pmd_page(pmd) (mem_map + ((unsigned long)(__va(pmd_val(pmd)) - PAGE_OFFSET) >> PAGE_SHIFT)) +#define pmd_page(pmd) virt_to_page(__va(pmd_val(pmd))) #define pgd_none(pgd) (!pgd_val(pgd)) @@ -223,10 +223,10 @@ static inline pte_t *pte_offset_kernel(pmd_t *pmdp, unsigned long address) return (pte_t *)__pmd_page(*pmdp) + ((address >> PAGE_SHIFT) & (PTRS_PER_PTE - 1)); } -#define pte_offset_map(pmdp,address) ((pte_t *)kmap(pmd_page(*pmdp)) + ((address >> PAGE_SHIFT) & (PTRS_PER_PTE - 1))) +#define pte_offset_map(pmdp,address) ((pte_t *)__pmd_page(*pmdp) + (((address) >> PAGE_SHIFT) & (PTRS_PER_PTE - 1))) #define pte_offset_map_nested(pmdp, address) pte_offset_map(pmdp, address) -#define pte_unmap(pte) kunmap(pte) -#define pte_unmap_nested(pte) kunmap(pte) +#define pte_unmap(pte) ((void)0) +#define pte_unmap_nested(pte) ((void)0) /* * Allocate and free page tables. The xxx_kernel() versions are diff --git a/include/asm-m68k/page.h b/include/asm-m68k/page.h index fcc165ddd09..9e6d0d6debd 100644 --- a/include/asm-m68k/page.h +++ b/include/asm-m68k/page.h @@ -27,6 +27,8 @@ #ifndef __ASSEMBLY__ +#include <asm/module.h> + #define get_user_page(vaddr) __get_free_page(GFP_KERNEL) #define free_user_page(page, addr) free_page(addr) @@ -114,18 +116,33 @@ typedef struct { unsigned long pgprot; } pgprot_t; #ifndef __ASSEMBLY__ +extern unsigned long m68k_memoffset; + #ifndef CONFIG_SUN3 #define WANT_PAGE_VIRTUAL -#ifdef CONFIG_SINGLE_MEMORY_CHUNK -extern unsigned long m68k_memoffset; -#define __pa(vaddr) ((unsigned long)(vaddr)+m68k_memoffset) -#define __va(paddr) ((void *)((unsigned long)(paddr)-m68k_memoffset)) -#else -#define __pa(vaddr) virt_to_phys((void *)(vaddr)) -#define __va(paddr) phys_to_virt((unsigned long)(paddr)) -#endif +static inline unsigned long ___pa(void *vaddr) +{ + unsigned long paddr; + asm ( + "1: addl #0,%0\n" + m68k_fixup(%c2, 1b+2) + : "=r" (paddr) + : "0" (vaddr), "i" (m68k_fixup_memoffset)); + return paddr; +} +#define __pa(vaddr) ___pa((void *)(vaddr)) +static inline void *__va(unsigned long paddr) +{ + void *vaddr; + asm ( + "1: subl #0,%0\n" + m68k_fixup(%c2, 1b+2) + : "=r" (vaddr) + : "0" (paddr), "i" (m68k_fixup_memoffset)); + return vaddr; +} #else /* !CONFIG_SUN3 */ /* This #define is a horrible hack to suppress lots of warnings. --m */ @@ -161,11 +178,47 @@ static inline void *__va(unsigned long x) #define virt_to_pfn(kaddr) (__pa(kaddr) >> PAGE_SHIFT) #define pfn_to_virt(pfn) __va((pfn) << PAGE_SHIFT) -#define virt_to_page(kaddr) (mem_map + (((unsigned long)(kaddr)-PAGE_OFFSET) >> PAGE_SHIFT)) -#define page_to_virt(page) ((((page) - mem_map) << PAGE_SHIFT) + PAGE_OFFSET) +extern int m68k_virt_to_node_shift; + +#ifdef CONFIG_SINGLE_MEMORY_CHUNK +#define __virt_to_node(addr) (&pg_data_map[0]) +#else +extern struct pglist_data *pg_data_table[]; + +static inline __attribute_const__ int __virt_to_node_shift(void) +{ + int shift; + + asm ( + "1: moveq #0,%0\n" + m68k_fixup(%c1, 1b) + : "=d" (shift) + : "i" (m68k_fixup_vnode_shift)); + return shift; +} + +#define __virt_to_node(addr) (pg_data_table[(unsigned long)(addr) >> __virt_to_node_shift()]) +#endif -#define pfn_to_page(pfn) virt_to_page(pfn_to_virt(pfn)) -#define page_to_pfn(page) virt_to_pfn(page_to_virt(page)) +#define virt_to_page(addr) ({ \ + pfn_to_page(virt_to_pfn(addr)); \ +}) +#define page_to_virt(page) ({ \ + pfn_to_virt(page_to_pfn(page)); \ +}) + +#define pfn_to_page(pfn) ({ \ + unsigned long __pfn = (pfn); \ + struct pglist_data *pgdat; \ + pgdat = __virt_to_node((unsigned long)pfn_to_virt(__pfn)); \ + pgdat->node_mem_map + (__pfn - pgdat->node_start_pfn); \ +}) +#define page_to_pfn(_page) ({ \ + struct page *__p = (_page); \ + struct pglist_data *pgdat; \ + pgdat = &pg_data_map[page_to_nid(__p)]; \ + ((__p) - pgdat->node_mem_map) + pgdat->node_start_pfn; \ +}) #define virt_addr_valid(kaddr) ((void *)(kaddr) >= (void *)PAGE_OFFSET && (void *)(kaddr) < high_memory) #define pfn_valid(pfn) virt_addr_valid(pfn_to_virt(pfn)) diff --git a/include/asm-m68k/pgalloc.h b/include/asm-m68k/pgalloc.h index a9cfb4b99d8..4cb1a57ab76 100644 --- a/include/asm-m68k/pgalloc.h +++ b/include/asm-m68k/pgalloc.h @@ -8,11 +8,12 @@ #include <asm/virtconvert.h> - #ifdef CONFIG_SUN3 #include <asm/sun3_pgalloc.h> #else #include <asm/motorola_pgalloc.h> #endif +extern void m68k_setup_node(int node); + #endif /* M68K_PGALLOC_H */ diff --git a/include/asm-m68k/pgtable.h b/include/asm-m68k/pgtable.h index 555b87a1f7e..778a4c538eb 100644 --- a/include/asm-m68k/pgtable.h +++ b/include/asm-m68k/pgtable.h @@ -107,22 +107,7 @@ extern void *empty_zero_page; /* 64-bit machines, beware! SRB. */ #define SIZEOF_PTR_LOG2 2 -/* - * Check if the addr/len goes up to the end of a physical - * memory chunk. Used for DMA functions. - */ -#ifdef CONFIG_SINGLE_MEMORY_CHUNK -/* - * It makes no sense to consider whether we cross a memory boundary if - * we support just one physical chunk of memory. - */ -static inline int mm_end_of_chunk(unsigned long addr, int len) -{ - return 0; -} -#else -int mm_end_of_chunk (unsigned long addr, int len); -#endif +#define mm_end_of_chunk(addr, len) 0 extern void kernel_set_cachemode(void *addr, unsigned long size, int cmode); diff --git a/include/asm-m68k/sun3_pgtable.h b/include/asm-m68k/sun3_pgtable.h index 5156a28a18d..b9e62c1e7ae 100644 --- a/include/asm-m68k/sun3_pgtable.h +++ b/include/asm-m68k/sun3_pgtable.h @@ -132,8 +132,8 @@ static inline void pte_clear (struct mm_struct *mm, unsigned long addr, pte_t *p #define pfn_pte(pfn, pgprot) \ ({ pte_t __pte; pte_val(__pte) = pfn | pgprot_val(pgprot); __pte; }) -#define pte_page(pte) (mem_map+((__pte_page(pte) - PAGE_OFFSET) >> PAGE_SHIFT)) -#define pmd_page(pmd) (mem_map+((__pmd_page(pmd) - PAGE_OFFSET) >> PAGE_SHIFT)) +#define pte_page(pte) virt_to_page(__pte_page(pte)) +#define pmd_page(pmd) virt_to_page(__pmd_page(pmd)) static inline int pmd_none2 (pmd_t *pmd) { return !pmd_val (*pmd); } diff --git a/include/asm-m68k/virtconvert.h b/include/asm-m68k/virtconvert.h index 83a87c9b1a1..dea32fbc7e5 100644 --- a/include/asm-m68k/virtconvert.h +++ b/include/asm-m68k/virtconvert.h @@ -8,56 +8,35 @@ #ifdef __KERNEL__ #include <linux/compiler.h> +#include <linux/mmzone.h> #include <asm/setup.h> #include <asm/page.h> -#ifdef CONFIG_AMIGA -#include <asm/amigahw.h> -#endif - /* * Change virtual addresses to physical addresses and vv. */ -#ifndef CONFIG_SUN3 -extern unsigned long mm_vtop(unsigned long addr) __attribute_const__; -extern unsigned long mm_ptov(unsigned long addr) __attribute_const__; -#else -static inline unsigned long mm_vtop(unsigned long vaddr) -{ - return __pa(vaddr); -} - -static inline unsigned long mm_ptov(unsigned long paddr) -{ - return (unsigned long)__va(paddr); -} -#endif - -#ifdef CONFIG_SINGLE_MEMORY_CHUNK -static inline unsigned long virt_to_phys(void *vaddr) -{ - return (unsigned long)vaddr - PAGE_OFFSET + m68k_memory[0].addr; -} - -static inline void * phys_to_virt(unsigned long paddr) -{ - return (void *)(paddr - m68k_memory[0].addr + PAGE_OFFSET); -} -#else static inline unsigned long virt_to_phys(void *address) { - return mm_vtop((unsigned long)address); + return __pa(address); } static inline void *phys_to_virt(unsigned long address) { - return (void *) mm_ptov(address); + return __va(address); } -#endif /* Permanent address of a page. */ -#define __page_address(page) (PAGE_OFFSET + (((page) - mem_map) << PAGE_SHIFT)) -#define page_to_phys(page) virt_to_phys((void *)__page_address(page)) +#ifdef CONFIG_SINGLE_MEMORY_CHUNK +#define page_to_phys(page) \ + __pa(PAGE_OFFSET + (((page) - pg_data_map[0].node_mem_map) << PAGE_SHIFT)) +#else +#define page_to_phys(_page) ({ \ + struct page *__page = _page; \ + struct pglist_data *pgdat; \ + pgdat = pg_data_table[page_to_nid(__page)]; \ + page_to_pfn(__page) << PAGE_SHIFT; \ +}) +#endif /* * IO bus memory addresses are 1:1 with the physical address, diff --git a/include/asm-sparc/atomic.h b/include/asm-sparc/atomic.h index 731fa56e0c3..bdca5416d8b 100644 --- a/include/asm-sparc/atomic.h +++ b/include/asm-sparc/atomic.h @@ -2,6 +2,7 @@ * * Copyright (C) 1996 David S. Miller (davem@caip.rutgers.edu) * Copyright (C) 2000 Anton Blanchard (anton@linuxcare.com.au) + * Copyright (C) 2007 Kyle McMartin (kyle@parisc-linux.org) * * Additions by Keith M Wesolowski (wesolows@foobazco.org) based * on asm-parisc/atomic.h Copyright (C) 2000 Philipp Rumpf <prumpf@tux.org>. @@ -10,11 +11,48 @@ #ifndef __ARCH_SPARC_ATOMIC__ #define __ARCH_SPARC_ATOMIC__ +#include <linux/types.h> typedef struct { volatile int counter; } atomic_t; #ifdef __KERNEL__ +/* Emulate cmpxchg() the same way we emulate atomics, + * by hashing the object address and indexing into an array + * of spinlocks to get a bit of performance... + * + * See arch/sparc/lib/atomic32.c for implementation. + * + * Cribbed from <asm-parisc/atomic.h> + */ +#define __HAVE_ARCH_CMPXCHG 1 + +/* bug catcher for when unsupported size is used - won't link */ +extern void __cmpxchg_called_with_bad_pointer(void); +/* we only need to support cmpxchg of a u32 on sparc */ +extern unsigned long __cmpxchg_u32(volatile u32 *m, u32 old, u32 new_); + +/* don't worry...optimizer will get rid of most of this */ +static __inline__ unsigned long +__cmpxchg(volatile void *ptr, unsigned long old, unsigned long new_, int size) +{ + switch(size) { + case 4: + return __cmpxchg_u32((u32 *)ptr, (u32)old, (u32)new_); + default: + __cmpxchg_called_with_bad_pointer(); + break; + } + return old; +} + +#define cmpxchg(ptr,o,n) ({ \ + __typeof__(*(ptr)) _o_ = (o); \ + __typeof__(*(ptr)) _n_ = (n); \ + (__typeof__(*(ptr))) __cmpxchg((ptr), (unsigned long)_o_, \ + (unsigned long)_n_, sizeof(*(ptr))); \ +}) + #define ATOMIC_INIT(i) { (i) } extern int __atomic_add_return(int, atomic_t *); diff --git a/include/asm-sparc64/bugs.h b/include/asm-sparc64/bugs.h index 120422fdb02..bf39d86c0c9 100644 --- a/include/asm-sparc64/bugs.h +++ b/include/asm-sparc64/bugs.h @@ -1,9 +1,8 @@ -/* $Id: bugs.h,v 1.1 1996/12/26 13:25:20 davem Exp $ - * include/asm-sparc64/bugs.h: Sparc probes for various bugs. +/* bugs.h: Sparc64 probes for various bugs. * - * Copyright (C) 1996 David S. Miller (davem@caip.rutgers.edu) + * Copyright (C) 1996, 2007 David S. Miller (davem@davemloft.net) */ - +#include <asm/sstate.h> extern unsigned long loops_per_jiffy; @@ -12,4 +11,5 @@ static void __init check_bugs(void) #ifndef CONFIG_SMP cpu_data(0).udelay_val = loops_per_jiffy; #endif + sstate_running(); } diff --git a/include/asm-sparc64/cpudata.h b/include/asm-sparc64/cpudata.h index e89922d6718..03c385de761 100644 --- a/include/asm-sparc64/cpudata.h +++ b/include/asm-sparc64/cpudata.h @@ -17,11 +17,11 @@ typedef struct { /* Dcache line 1 */ unsigned int __softirq_pending; /* must be 1st, see rtrap.S */ - unsigned int __pad0_1; - unsigned int __pad0_2; - unsigned int __pad1; + unsigned int __pad0; unsigned long clock_tick; /* %tick's per second */ unsigned long udelay_val; + unsigned int __pad1; + unsigned int __pad2; /* Dcache line 2, rarely used */ unsigned int dcache_size; @@ -30,8 +30,8 @@ typedef struct { unsigned int icache_line_size; unsigned int ecache_size; unsigned int ecache_line_size; + int core_id; unsigned int __pad3; - unsigned int __pad4; } cpuinfo_sparc; DECLARE_PER_CPU(cpuinfo_sparc, __cpu_data); @@ -76,12 +76,18 @@ struct trap_per_cpu { /* Dcache line 8: IRQ work list, and keep trap_block a power-of-2 in size. */ unsigned int irq_worklist; - unsigned int __pad1; - unsigned long __pad2[3]; + unsigned int cpu_mondo_qmask; + unsigned int dev_mondo_qmask; + unsigned int resum_qmask; + unsigned int nonresum_qmask; + unsigned int __pad2[3]; } __attribute__((aligned(64))); extern struct trap_per_cpu trap_block[NR_CPUS]; extern void init_cur_cpu_trap(struct thread_info *); extern void setup_tba(void); +extern int ncpus_probed; + +extern unsigned long real_hard_smp_processor_id(void); struct cpuid_patch_entry { unsigned int addr; @@ -122,6 +128,10 @@ extern struct sun4v_2insn_patch_entry __sun4v_2insn_patch, #define TRAP_PER_CPU_TSB_HUGE 0xd0 #define TRAP_PER_CPU_TSB_HUGE_TEMP 0xd8 #define TRAP_PER_CPU_IRQ_WORKLIST 0xe0 +#define TRAP_PER_CPU_CPU_MONDO_QMASK 0xe4 +#define TRAP_PER_CPU_DEV_MONDO_QMASK 0xe8 +#define TRAP_PER_CPU_RESUM_QMASK 0xec +#define TRAP_PER_CPU_NONRESUM_QMASK 0xf0 #define TRAP_BLOCK_SZ_SHIFT 8 @@ -192,7 +202,7 @@ extern struct sun4v_2insn_patch_entry __sun4v_2insn_patch, * the calculations done by the macro mid-stream. */ #define LOAD_PER_CPU_BASE(DEST, THR, REG1, REG2, REG3) \ - ldub [THR + TI_CPU], REG1; \ + lduh [THR + TI_CPU], REG1; \ sethi %hi(__per_cpu_shift), REG3; \ sethi %hi(__per_cpu_base), REG2; \ ldx [REG3 + %lo(__per_cpu_shift)], REG3; \ diff --git a/include/asm-sparc64/hypervisor.h b/include/asm-sparc64/hypervisor.h index a5558c87556..5cdb1ff0483 100644 --- a/include/asm-sparc64/hypervisor.h +++ b/include/asm-sparc64/hypervisor.h @@ -73,6 +73,8 @@ #define HV_ENOTSUPPORTED 13 /* Function not supported */ #define HV_ENOMAP 14 /* No mapping found */ #define HV_ETOOMANY 15 /* Too many items specified */ +#define HV_ECHANNEL 16 /* Invalid LDC channel */ +#define HV_EBUSY 17 /* Resource busy */ /* mach_exit() * TRAP: HV_FAST_TRAP @@ -95,6 +97,10 @@ */ #define HV_FAST_MACH_EXIT 0x00 +#ifndef __ASSEMBLY__ +extern void sun4v_mach_exit(unsigned long exit_core); +#endif + /* Domain services. */ /* mach_desc() @@ -120,7 +126,13 @@ */ #define HV_FAST_MACH_DESC 0x01 -/* mach_exit() +#ifndef __ASSEMBLY__ +extern unsigned long sun4v_mach_desc(unsigned long buffer_pa, + unsigned long buf_len, + unsigned long *real_buf_len); +#endif + +/* mach_sir() * TRAP: HV_FAST_TRAP * FUNCTION: HV_FAST_MACH_SIR * ERRORS: This service does not return. @@ -135,53 +147,66 @@ */ #define HV_FAST_MACH_SIR 0x02 -/* mach_set_soft_state() +#ifndef __ASSEMBLY__ +extern void sun4v_mach_sir(void); +#endif + +/* mach_set_watchdog() * TRAP: HV_FAST_TRAP - * FUNCTION: HV_FAST_MACH_SET_SOFT_STATE - * ARG0: software state - * ARG1: software state description pointer + * FUNCTION: HV_FAST_MACH_SET_WATCHDOG + * ARG0: timeout in milliseconds * RET0: status - * ERRORS: EINVAL software state not valid or software state - * description is not NULL terminated - * ENORADDR software state description pointer is not a - * valid real address - * EBADALIGNED software state description is not correctly - * aligned + * RET1: time remaining in milliseconds * - * This allows the guest to report it's soft state to the hypervisor. There - * are two primary components to this state. The first part states whether - * the guest software is running or not. The second containts optional - * details specific to the software. + * A guest uses this API to set a watchdog timer. Once the gues has set + * the timer, it must call the timer service again either to disable or + * postpone the expiration. If the timer expires before being reset or + * disabled, then the hypervisor take a platform specific action leading + * to guest termination within a bounded time period. The platform action + * may include recovery actions such as reporting the expiration to a + * Service Processor, and/or automatically restarting the gues. * - * The software state argument is defined below in HV_SOFT_STATE_*, and - * indicates whether the guest is operating normally or in a transitional - * state. + * The 'timeout' parameter is specified in milliseconds, however the + * implementated granularity is given by the 'watchdog-resolution' + * property in the 'platform' node of the guest's machine description. + * The largest allowed timeout value is specified by the + * 'watchdog-max-timeout' property of the 'platform' node. * - * The software state description argument is a real address of a data buffer - * of size 32-bytes aligned on a 32-byte boundary. It is treated as a NULL - * terminated 7-bit ASCII string of up to 31 characters not including the - * NULL termination. - */ -#define HV_FAST_MACH_SET_SOFT_STATE 0x03 -#define HV_SOFT_STATE_NORMAL 0x01 -#define HV_SOFT_STATE_TRANSITION 0x02 - -/* mach_get_soft_state() - * TRAP: HV_FAST_TRAP - * FUNCTION: HV_FAST_MACH_GET_SOFT_STATE - * ARG0: software state description pointer - * RET0: status - * RET1: software state - * ERRORS: ENORADDR software state description pointer is not a - * valid real address - * EBADALIGNED software state description is not correctly - * aligned + * If the 'timeout' argument is not zero, the watchdog timer is set to + * expire after a minimum of 'timeout' milliseconds. * - * Retrieve the current value of the guest's software state. The rules - * for the software state pointer are the same as for mach_set_soft_state() - * above. + * If the 'timeout' argument is zero, the watchdog timer is disabled. + * + * If the 'timeout' value exceeds the value of the 'max-watchdog-timeout' + * property, the hypervisor leaves the watchdog timer state unchanged, + * and returns a status of EINVAL. + * + * The 'time remaining' return value is valid regardless of whether the + * return status is EOK or EINVAL. A non-zero return value indicates the + * number of milliseconds that were remaining until the timer was to expire. + * If less than one millisecond remains, the return value is '1'. If the + * watchdog timer was disabled at the time of the call, the return value is + * zero. + * + * If the hypervisor cannot support the exact timeout value requested, but + * can support a larger timeout value, the hypervisor may round the actual + * timeout to a value larger than the requested timeout, consequently the + * 'time remaining' return value may be larger than the previously requested + * timeout value. + * + * Any guest OS debugger should be aware that the watchdog service may be in + * use. Consequently, it is recommended that the watchdog service is + * disabled upon debugger entry (e.g. reaching a breakpoint), and then + * re-enabled upon returning to normal execution. The API has been designed + * with this in mind, and the 'time remaining' result of the disable call may + * be used directly as the timeout argument of the re-enable call. */ -#define HV_FAST_MACH_GET_SOFT_STATE 0x04 +#define HV_FAST_MACH_SET_WATCHDOG 0x05 + +#ifndef __ASSEMBLY__ +extern unsigned long sun4v_mach_set_watchdog(unsigned long timeout, + unsigned long *orig_timeout); +#endif /* CPU services. * @@ -206,8 +231,8 @@ * FUNCTION: HV_FAST_CPU_START * ARG0: CPU ID * ARG1: PC - * ARG1: RTBA - * ARG1: target ARG0 + * ARG2: RTBA + * ARG3: target ARG0 * RET0: status * ERRORS: ENOCPU Invalid CPU ID * EINVAL Target CPU ID is not in the stopped state @@ -224,6 +249,13 @@ */ #define HV_FAST_CPU_START 0x10 +#ifndef __ASSEMBLY__ +extern unsigned long sun4v_cpu_start(unsigned long cpuid, + unsigned long pc, + unsigned long rtba, + unsigned long arg0); +#endif + /* cpu_stop() * TRAP: HV_FAST_TRAP * FUNCTION: HV_FAST_CPU_STOP @@ -245,6 +277,10 @@ */ #define HV_FAST_CPU_STOP 0x11 +#ifndef __ASSEMBLY__ +extern unsigned long sun4v_cpu_stop(unsigned long cpuid); +#endif + /* cpu_yield() * TRAP: HV_FAST_TRAP * FUNCTION: HV_FAST_CPU_YIELD @@ -588,6 +624,11 @@ struct hv_fault_status { */ #define HV_FAST_MMU_TSB_CTX0 0x20 +#ifndef __ASSEMBLY__ +extern unsigned long sun4v_mmu_tsb_ctx0(unsigned long num_descriptions, + unsigned long tsb_desc_ra); +#endif + /* mmu_tsb_ctxnon0() * TRAP: HV_FAST_TRAP * FUNCTION: HV_FAST_MMU_TSB_CTXNON0 @@ -694,6 +735,13 @@ struct hv_fault_status { */ #define HV_FAST_MMU_MAP_PERM_ADDR 0x25 +#ifndef __ASSEMBLY__ +extern unsigned long sun4v_mmu_map_perm_addr(unsigned long vaddr, + unsigned long set_to_zero, + unsigned long tte, + unsigned long flags); +#endif + /* mmu_fault_area_conf() * TRAP: HV_FAST_TRAP * FUNCTION: HV_FAST_MMU_FAULT_AREA_CONF @@ -892,6 +940,10 @@ struct hv_fault_status { */ #define HV_FAST_TOD_GET 0x50 +#ifndef __ASSEMBLY__ +extern unsigned long sun4v_tod_get(unsigned long *time); +#endif + /* tod_set() * TRAP: HV_FAST_TRAP * FUNCTION: HV_FAST_TOD_SET @@ -905,6 +957,10 @@ struct hv_fault_status { */ #define HV_FAST_TOD_SET 0x51 +#ifndef __ASSEMBLY__ +extern unsigned long sun4v_tod_set(unsigned long time); +#endif + /* Console services */ /* con_getchar() @@ -988,6 +1044,59 @@ extern unsigned long sun4v_con_write(unsigned long buffer, unsigned long *bytes_written); #endif +/* mach_set_soft_state() + * TRAP: HV_FAST_TRAP + * FUNCTION: HV_FAST_MACH_SET_SOFT_STATE + * ARG0: software state + * ARG1: software state description pointer + * RET0: status + * ERRORS: EINVAL software state not valid or software state + * description is not NULL terminated + * ENORADDR software state description pointer is not a + * valid real address + * EBADALIGNED software state description is not correctly + * aligned + * + * This allows the guest to report it's soft state to the hypervisor. There + * are two primary components to this state. The first part states whether + * the guest software is running or not. The second containts optional + * details specific to the software. + * + * The software state argument is defined below in HV_SOFT_STATE_*, and + * indicates whether the guest is operating normally or in a transitional + * state. + * + * The software state description argument is a real address of a data buffer + * of size 32-bytes aligned on a 32-byte boundary. It is treated as a NULL + * terminated 7-bit ASCII string of up to 31 characters not including the + * NULL termination. + */ +#define HV_FAST_MACH_SET_SOFT_STATE 0x70 +#define HV_SOFT_STATE_NORMAL 0x01 +#define HV_SOFT_STATE_TRANSITION 0x02 + +#ifndef __ASSEMBLY__ +extern unsigned long sun4v_mach_set_soft_state(unsigned long soft_state, + unsigned long msg_string_ra); +#endif + +/* mach_get_soft_state() + * TRAP: HV_FAST_TRAP + * FUNCTION: HV_FAST_MACH_GET_SOFT_STATE + * ARG0: software state description pointer + * RET0: status + * RET1: software state + * ERRORS: ENORADDR software state description pointer is not a + * valid real address + * EBADALIGNED software state description is not correctly + * aligned + * + * Retrieve the current value of the guest's software state. The rules + * for the software state pointer are the same as for mach_set_soft_state() + * above. + */ +#define HV_FAST_MACH_GET_SOFT_STATE 0x71 + /* Trap trace services. * * The hypervisor provides a trap tracing capability for privileged @@ -1379,6 +1488,113 @@ extern unsigned long sun4v_intr_gettarget(unsigned long sysino); extern unsigned long sun4v_intr_settarget(unsigned long sysino, unsigned long cpuid); #endif +/* vintr_get_cookie() + * TRAP: HV_FAST_TRAP + * FUNCTION: HV_FAST_VINTR_GET_COOKIE + * ARG0: device handle + * ARG1: device ino + * RET0: status + * RET1: cookie + */ +#define HV_FAST_VINTR_GET_COOKIE 0xa7 + +/* vintr_set_cookie() + * TRAP: HV_FAST_TRAP + * FUNCTION: HV_FAST_VINTR_SET_COOKIE + * ARG0: device handle + * ARG1: device ino + * ARG2: cookie + * RET0: status + */ +#define HV_FAST_VINTR_SET_COOKIE 0xa8 + +/* vintr_get_valid() + * TRAP: HV_FAST_TRAP + * FUNCTION: HV_FAST_VINTR_GET_VALID + * ARG0: device handle + * ARG1: device ino + * RET0: status + * RET1: valid state + */ +#define HV_FAST_VINTR_GET_VALID 0xa9 + +/* vintr_set_valid() + * TRAP: HV_FAST_TRAP + * FUNCTION: HV_FAST_VINTR_SET_VALID + * ARG0: device handle + * ARG1: device ino + * ARG2: valid state + * RET0: status + */ +#define HV_FAST_VINTR_SET_VALID 0xaa + +/* vintr_get_state() + * TRAP: HV_FAST_TRAP + * FUNCTION: HV_FAST_VINTR_GET_STATE + * ARG0: device handle + * ARG1: device ino + * RET0: status + * RET1: state + */ +#define HV_FAST_VINTR_GET_STATE 0xab + +/* vintr_set_state() + * TRAP: HV_FAST_TRAP + * FUNCTION: HV_FAST_VINTR_SET_STATE + * ARG0: device handle + * ARG1: device ino + * ARG2: state + * RET0: status + */ +#define HV_FAST_VINTR_SET_STATE 0xac + +/* vintr_get_target() + * TRAP: HV_FAST_TRAP + * FUNCTION: HV_FAST_VINTR_GET_TARGET + * ARG0: device handle + * ARG1: device ino + * RET0: status + * RET1: cpuid + */ +#define HV_FAST_VINTR_GET_TARGET 0xad + +/* vintr_set_target() + * TRAP: HV_FAST_TRAP + * FUNCTION: HV_FAST_VINTR_SET_TARGET + * ARG0: device handle + * ARG1: device ino + * ARG2: cpuid + * RET0: status + */ +#define HV_FAST_VINTR_SET_TARGET 0xae + +#ifndef __ASSEMBLY__ +extern unsigned long sun4v_vintr_get_cookie(unsigned long dev_handle, + unsigned long dev_ino, + unsigned long *cookie); +extern unsigned long sun4v_vintr_set_cookie(unsigned long dev_handle, + unsigned long dev_ino, + unsigned long cookie); +extern unsigned long sun4v_vintr_get_valid(unsigned long dev_handle, + unsigned long dev_ino, + unsigned long *valid); +extern unsigned long sun4v_vintr_set_valid(unsigned long dev_handle, + unsigned long dev_ino, + unsigned long valid); +extern unsigned long sun4v_vintr_get_state(unsigned long dev_handle, + unsigned long dev_ino, + unsigned long *state); +extern unsigned long sun4v_vintr_set_state(unsigned long dev_handle, + unsigned long dev_ino, + unsigned long state); +extern unsigned long sun4v_vintr_get_target(unsigned long dev_handle, + unsigned long dev_ino, + unsigned long *cpuid); +extern unsigned long sun4v_vintr_set_target(unsigned long dev_handle, + unsigned long dev_ino, + unsigned long cpuid); +#endif + /* PCI IO services. * * See the terminology descriptions in the device interrupt services @@ -2037,6 +2253,346 @@ extern unsigned long sun4v_intr_settarget(unsigned long sysino, unsigned long cp */ #define HV_FAST_PCI_MSG_SETVALID 0xd3 +/* Logical Domain Channel services. */ + +#define LDC_CHANNEL_DOWN 0 +#define LDC_CHANNEL_UP 1 +#define LDC_CHANNEL_RESETTING 2 + +/* ldc_tx_qconf() + * TRAP: HV_FAST_TRAP + * FUNCTION: HV_FAST_LDC_TX_QCONF + * ARG0: channel ID + * ARG1: real address base of queue + * ARG2: num entries in queue + * RET0: status + * + * Configure transmit queue for the LDC endpoint specified by the + * given channel ID, to be placed at the given real address, and + * be of the given num entries. Num entries must be a power of two. + * The real address base of the queue must be aligned on the queue + * size. Each queue entry is 64-bytes, so for example, a 32 entry + * queue must be aligned on a 2048 byte real address boundary. + * + * Upon configuration of a valid transmit queue the head and tail + * pointers are set to a hypervisor specific identical value indicating + * that the queue initially is empty. + * + * The endpoint's transmit queue is un-configured if num entries is zero. + * + * The maximum number of entries for each queue for a specific cpu may be + * determined from the machine description. A transmit queue may be + * specified even in the event that the LDC is down (peer endpoint has no + * receive queue specified). Transmission will begin as soon as the peer + * endpoint defines a receive queue. + * + * It is recommended that a guest wait for a transmit queue to empty prior + * to reconfiguring it, or un-configuring it. Re or un-configuring of a + * non-empty transmit queue behaves exactly as defined above, however it + * is undefined as to how many of the pending entries in the original queue + * will be delivered prior to the re-configuration taking effect. + * Furthermore, as the queue configuration causes a reset of the head and + * tail pointers there is no way for a guest to determine how many entries + * have been sent after the configuration operation. + */ +#define HV_FAST_LDC_TX_QCONF 0xe0 + +/* ldc_tx_qinfo() + * TRAP: HV_FAST_TRAP + * FUNCTION: HV_FAST_LDC_TX_QINFO + * ARG0: channel ID + * RET0: status + * RET1: real address base of queue + * RET2: num entries in queue + * + * Return the configuration info for the transmit queue of LDC endpoint + * defined by the given channel ID. The real address is the currently + * defined real address base of the defined queue, and num entries is the + * size of the queue in terms of number of entries. + * + * If the specified channel ID is a valid endpoint number, but no transmit + * queue has been defined this service will return success, but with num + * entries set to zero and the real address will have an undefined value. + */ +#define HV_FAST_LDC_TX_QINFO 0xe1 + +/* ldc_tx_get_state() + * TRAP: HV_FAST_TRAP + * FUNCTION: HV_FAST_LDC_TX_GET_STATE + * ARG0: channel ID + * RET0: status + * RET1: head offset + * RET2: tail offset + * RET3: channel state + * + * Return the transmit state, and the head and tail queue pointers, for + * the transmit queue of the LDC endpoint defined by the given channel ID. + * The head and tail values are the byte offset of the head and tail + * positions of the transmit queue for the specified endpoint. + */ +#define HV_FAST_LDC_TX_GET_STATE 0xe2 + +/* ldc_tx_set_qtail() + * TRAP: HV_FAST_TRAP + * FUNCTION: HV_FAST_LDC_TX_SET_QTAIL + * ARG0: channel ID + * ARG1: tail offset + * RET0: status + * + * Update the tail pointer for the transmit queue associated with the LDC + * endpoint defined by the given channel ID. The tail offset specified + * must be aligned on a 64 byte boundary, and calculated so as to increase + * the number of pending entries on the transmit queue. Any attempt to + * decrease the number of pending transmit queue entires is considered + * an invalid tail offset and will result in an EINVAL error. + * + * Since the tail of the transmit queue may not be moved backwards, the + * transmit queue may be flushed by configuring a new transmit queue, + * whereupon the hypervisor will configure the initial transmit head and + * tail pointers to be equal. + */ +#define HV_FAST_LDC_TX_SET_QTAIL 0xe3 + +/* ldc_rx_qconf() + * TRAP: HV_FAST_TRAP + * FUNCTION: HV_FAST_LDC_RX_QCONF + * ARG0: channel ID + * ARG1: real address base of queue + * ARG2: num entries in queue + * RET0: status + * + * Configure receive queue for the LDC endpoint specified by the + * given channel ID, to be placed at the given real address, and + * be of the given num entries. Num entries must be a power of two. + * The real address base of the queue must be aligned on the queue + * size. Each queue entry is 64-bytes, so for example, a 32 entry + * queue must be aligned on a 2048 byte real address boundary. + * + * The endpoint's transmit queue is un-configured if num entries is zero. + * + * If a valid receive queue is specified for a local endpoint the LDC is + * in the up state for the purpose of transmission to this endpoint. + * + * The maximum number of entries for each queue for a specific cpu may be + * determined from the machine description. + * + * As receive queue configuration causes a reset of the queue's head and + * tail pointers there is no way for a gues to determine how many entries + * have been received between a preceeding ldc_get_rx_state() API call + * and the completion of the configuration operation. It should be noted + * that datagram delivery is not guarenteed via domain channels anyway, + * and therefore any higher protocol should be resilient to datagram + * loss if necessary. However, to overcome this specific race potential + * it is recommended, for example, that a higher level protocol be employed + * to ensure either retransmission, or ensure that no datagrams are pending + * on the peer endpoint's transmit queue prior to the configuration process. + */ +#define HV_FAST_LDC_RX_QCONF 0xe4 + +/* ldc_rx_qinfo() + * TRAP: HV_FAST_TRAP + * FUNCTION: HV_FAST_LDC_RX_QINFO + * ARG0: channel ID + * RET0: status + * RET1: real address base of queue + * RET2: num entries in queue + * + * Return the configuration info for the receive queue of LDC endpoint + * defined by the given channel ID. The real address is the currently + * defined real address base of the defined queue, and num entries is the + * size of the queue in terms of number of entries. + * + * If the specified channel ID is a valid endpoint number, but no receive + * queue has been defined this service will return success, but with num + * entries set to zero and the real address will have an undefined value. + */ +#define HV_FAST_LDC_RX_QINFO 0xe5 + +/* ldc_rx_get_state() + * TRAP: HV_FAST_TRAP + * FUNCTION: HV_FAST_LDC_RX_GET_STATE + * ARG0: channel ID + * RET0: status + * RET1: head offset + * RET2: tail offset + * RET3: channel state + * + * Return the receive state, and the head and tail queue pointers, for + * the receive queue of the LDC endpoint defined by the given channel ID. + * The head and tail values are the byte offset of the head and tail + * positions of the receive queue for the specified endpoint. + */ +#define HV_FAST_LDC_RX_GET_STATE 0xe6 + +/* ldc_rx_set_qhead() + * TRAP: HV_FAST_TRAP + * FUNCTION: HV_FAST_LDC_RX_SET_QHEAD + * ARG0: channel ID + * ARG1: head offset + * RET0: status + * + * Update the head pointer for the receive queue associated with the LDC + * endpoint defined by the given channel ID. The head offset specified + * must be aligned on a 64 byte boundary, and calculated so as to decrease + * the number of pending entries on the receive queue. Any attempt to + * increase the number of pending receive queue entires is considered + * an invalid head offset and will result in an EINVAL error. + * + * The receive queue may be flushed by setting the head offset equal + * to the current tail offset. + */ +#define HV_FAST_LDC_RX_SET_QHEAD 0xe7 + +/* LDC Map Table Entry. Each slot is defined by a translation table + * entry, as specified by the LDC_MTE_* bits below, and a 64-bit + * hypervisor invalidation cookie. + */ +#define LDC_MTE_PADDR 0x0fffffffffffe000 /* pa[55:13] */ +#define LDC_MTE_COPY_W 0x0000000000000400 /* copy write access */ +#define LDC_MTE_COPY_R 0x0000000000000200 /* copy read access */ +#define LDC_MTE_IOMMU_W 0x0000000000000100 /* IOMMU write access */ +#define LDC_MTE_IOMMU_R 0x0000000000000080 /* IOMMU read access */ +#define LDC_MTE_EXEC 0x0000000000000040 /* execute */ +#define LDC_MTE_WRITE 0x0000000000000020 /* read */ +#define LDC_MTE_READ 0x0000000000000010 /* write */ +#define LDC_MTE_SZALL 0x000000000000000f /* page size bits */ +#define LDC_MTE_SZ16GB 0x0000000000000007 /* 16GB page */ +#define LDC_MTE_SZ2GB 0x0000000000000006 /* 2GB page */ +#define LDC_MTE_SZ256MB 0x0000000000000005 /* 256MB page */ +#define LDC_MTE_SZ32MB 0x0000000000000004 /* 32MB page */ +#define LDC_MTE_SZ4MB 0x0000000000000003 /* 4MB page */ +#define LDC_MTE_SZ512K 0x0000000000000002 /* 512K page */ +#define LDC_MTE_SZ64K 0x0000000000000001 /* 64K page */ +#define LDC_MTE_SZ8K 0x0000000000000000 /* 8K page */ + +#ifndef __ASSEMBLY__ +struct ldc_mtable_entry { + unsigned long mte; + unsigned long cookie; +}; +#endif + +/* ldc_set_map_table() + * TRAP: HV_FAST_TRAP + * FUNCTION: HV_FAST_LDC_SET_MAP_TABLE + * ARG0: channel ID + * ARG1: table real address + * ARG2: num entries + * RET0: status + * + * Register the MTE table at the given table real address, with the + * specified num entries, for the LDC indicated by the given channel + * ID. + */ +#define HV_FAST_LDC_SET_MAP_TABLE 0xea + +/* ldc_get_map_table() + * TRAP: HV_FAST_TRAP + * FUNCTION: HV_FAST_LDC_GET_MAP_TABLE + * ARG0: channel ID + * RET0: status + * RET1: table real address + * RET2: num entries + * + * Return the configuration of the current mapping table registered + * for the given channel ID. + */ +#define HV_FAST_LDC_GET_MAP_TABLE 0xeb + +#define LDC_COPY_IN 0 +#define LDC_COPY_OUT 1 + +/* ldc_copy() + * TRAP: HV_FAST_TRAP + * FUNCTION: HV_FAST_LDC_COPY + * ARG0: channel ID + * ARG1: LDC_COPY_* direction code + * ARG2: target real address + * ARG3: local real address + * ARG4: length in bytes + * RET0: status + * RET1: actual length in bytes + */ +#define HV_FAST_LDC_COPY 0xec + +#define LDC_MEM_READ 1 +#define LDC_MEM_WRITE 2 +#define LDC_MEM_EXEC 4 + +/* ldc_mapin() + * TRAP: HV_FAST_TRAP + * FUNCTION: HV_FAST_LDC_MAPIN + * ARG0: channel ID + * ARG1: cookie + * RET0: status + * RET1: real address + * RET2: LDC_MEM_* permissions + */ +#define HV_FAST_LDC_MAPIN 0xed + +/* ldc_unmap() + * TRAP: HV_FAST_TRAP + * FUNCTION: HV_FAST_LDC_UNMAP + * ARG0: real address + * RET0: status + */ +#define HV_FAST_LDC_UNMAP 0xee + +/* ldc_revoke() + * TRAP: HV_FAST_TRAP + * FUNCTION: HV_FAST_LDC_REVOKE + * ARG0: cookie + * ARG1: ldc_mtable_entry cookie + * RET0: status + */ +#define HV_FAST_LDC_REVOKE 0xef + +#ifndef __ASSEMBLY__ +extern unsigned long sun4v_ldc_tx_qconf(unsigned long channel, + unsigned long ra, + unsigned long num_entries); +extern unsigned long sun4v_ldc_tx_qinfo(unsigned long channel, + unsigned long *ra, + unsigned long *num_entries); +extern unsigned long sun4v_ldc_tx_get_state(unsigned long channel, + unsigned long *head_off, + unsigned long *tail_off, + unsigned long *chan_state); +extern unsigned long sun4v_ldc_tx_set_qtail(unsigned long channel, + unsigned long tail_off); +extern unsigned long sun4v_ldc_rx_qconf(unsigned long channel, + unsigned long ra, + unsigned long num_entries); +extern unsigned long sun4v_ldc_rx_qinfo(unsigned long channel, + unsigned long *ra, + unsigned long *num_entries); +extern unsigned long sun4v_ldc_rx_get_state(unsigned long channel, + unsigned long *head_off, + unsigned long *tail_off, + unsigned long *chan_state); +extern unsigned long sun4v_ldc_rx_set_qhead(unsigned long channel, + unsigned long head_off); +extern unsigned long sun4v_ldc_set_map_table(unsigned long channel, + unsigned long ra, + unsigned long num_entries); +extern unsigned long sun4v_ldc_get_map_table(unsigned long channel, + unsigned long *ra, + unsigned long *num_entries); +extern unsigned long sun4v_ldc_copy(unsigned long channel, + unsigned long dir_code, + unsigned long tgt_raddr, + unsigned long lcl_raddr, + unsigned long len, + unsigned long *actual_len); +extern unsigned long sun4v_ldc_mapin(unsigned long channel, + unsigned long cookie, + unsigned long *ra, + unsigned long *perm); +extern unsigned long sun4v_ldc_unmap(unsigned long ra); +extern unsigned long sun4v_ldc_revoke(unsigned long cookie, + unsigned long mte_cookie); +#endif + /* Performance counter services. */ #define HV_PERF_JBUS_PERF_CTRL_REG 0x00 @@ -2204,6 +2760,7 @@ extern void sun4v_hvapi_unregister(unsigned long group); extern int sun4v_hvapi_get(unsigned long group, unsigned long *major, unsigned long *minor); +extern void sun4v_hvapi_init(void); #endif #endif /* !(_SPARC64_HYPERVISOR_H) */ diff --git a/include/asm-sparc64/kdebug.h b/include/asm-sparc64/kdebug.h index 627e3396a5f..9974c7b0aeb 100644 --- a/include/asm-sparc64/kdebug.h +++ b/include/asm-sparc64/kdebug.h @@ -32,7 +32,6 @@ enum die_val { DIE_TRAP, DIE_TRAP_TL1, DIE_CALL, - DIE_PAGE_FAULT, }; #endif diff --git a/include/asm-sparc64/mdesc.h b/include/asm-sparc64/mdesc.h new file mode 100644 index 00000000000..124eb8ca237 --- /dev/null +++ b/include/asm-sparc64/mdesc.h @@ -0,0 +1,39 @@ +#ifndef _SPARC64_MDESC_H +#define _SPARC64_MDESC_H + +#include <linux/types.h> +#include <asm/prom.h> + +struct mdesc_node; +struct mdesc_arc { + const char *name; + struct mdesc_node *arc; +}; + +struct mdesc_node { + const char *name; + u64 node; + unsigned int unique_id; + unsigned int num_arcs; + struct property *properties; + struct mdesc_node *hash_next; + struct mdesc_node *allnodes_next; + struct mdesc_arc arcs[0]; +}; + +extern struct mdesc_node *md_find_node_by_name(struct mdesc_node *from, + const char *name); +#define md_for_each_node_by_name(__mn, __name) \ + for (__mn = md_find_node_by_name(NULL, __name); __mn; \ + __mn = md_find_node_by_name(__mn, __name)) + +extern struct property *md_find_property(const struct mdesc_node *mp, + const char *name, + int *lenp); +extern const void *md_get_property(const struct mdesc_node *mp, + const char *name, + int *lenp); + +extern void sun4v_mdesc_init(void); + +#endif diff --git a/include/asm-sparc64/oplib.h b/include/asm-sparc64/oplib.h index 6a0da3b1695..992f9f7a476 100644 --- a/include/asm-sparc64/oplib.h +++ b/include/asm-sparc64/oplib.h @@ -316,11 +316,8 @@ extern int prom_setprop(int node, const char *prop_name, char *prop_value, extern int prom_pathtoinode(const char *path); extern int prom_inst2pkg(int); - -/* CPU probing helpers. */ -struct device_node; -int cpu_find_by_instance(int instance, struct device_node **dev_node, int *mid); -int cpu_find_by_mid(int mid, struct device_node **prom_node); +extern int prom_service_exists(const char *service_name); +extern void prom_sun4v_guest_soft_state(void); /* Client interface level routines. */ extern void prom_set_trap_table(unsigned long tba); diff --git a/include/asm-sparc64/percpu.h b/include/asm-sparc64/percpu.h index ced8cbde046..88db872ce2f 100644 --- a/include/asm-sparc64/percpu.h +++ b/include/asm-sparc64/percpu.h @@ -5,7 +5,8 @@ #ifdef CONFIG_SMP -extern void setup_per_cpu_areas(void); +#define setup_per_cpu_areas() do { } while (0) +extern void real_setup_per_cpu_areas(void); extern unsigned long __per_cpu_base; extern unsigned long __per_cpu_shift; @@ -34,6 +35,7 @@ do { \ } while (0) #else /* ! SMP */ +#define real_setup_per_cpu_areas() do { } while (0) #define DEFINE_PER_CPU(type, name) \ __typeof__(type) per_cpu__##name diff --git a/include/asm-sparc64/prom.h b/include/asm-sparc64/prom.h index ddad5f99ac7..b4df3042add 100644 --- a/include/asm-sparc64/prom.h +++ b/include/asm-sparc64/prom.h @@ -90,6 +90,7 @@ extern struct device_node *of_find_compatible_node(struct device_node *from, const char *type, const char *compat); extern struct device_node *of_find_node_by_path(const char *path); extern struct device_node *of_find_node_by_phandle(phandle handle); +extern struct device_node *of_find_node_by_cpuid(int cpuid); extern struct device_node *of_get_parent(const struct device_node *node); extern struct device_node *of_get_next_child(const struct device_node *node, struct device_node *prev); diff --git a/include/asm-sparc64/smp.h b/include/asm-sparc64/smp.h index 869d16fb907..f76e1492add 100644 --- a/include/asm-sparc64/smp.h +++ b/include/asm-sparc64/smp.h @@ -41,7 +41,7 @@ extern cpumask_t cpu_sibling_map[NR_CPUS]; extern int hard_smp_processor_id(void); #define raw_smp_processor_id() (current_thread_info()->cpu) -extern void smp_setup_cpu_possible_map(void); +extern void smp_fill_in_sib_core_maps(void); extern unsigned char boot_cpu_id; #endif /* !(__ASSEMBLY__) */ @@ -49,7 +49,7 @@ extern unsigned char boot_cpu_id; #else #define hard_smp_processor_id() 0 -#define smp_setup_cpu_possible_map() do { } while (0) +#define smp_fill_in_sib_core_maps() do { } while (0) #define boot_cpu_id (0) #endif /* !(CONFIG_SMP) */ diff --git a/include/asm-sparc64/sstate.h b/include/asm-sparc64/sstate.h new file mode 100644 index 00000000000..a7c35dbcb28 --- /dev/null +++ b/include/asm-sparc64/sstate.h @@ -0,0 +1,13 @@ +#ifndef _SPARC64_SSTATE_H +#define _SPARC64_SSTATE_H + +extern void sstate_booting(void); +extern void sstate_running(void); +extern void sstate_halt(void); +extern void sstate_poweroff(void); +extern void sstate_panic(void); +extern void sstate_reboot(void); + +extern void sun4v_sstate_init(void); + +#endif /* _SPARC64_SSTATE_H */ diff --git a/include/asm-sparc64/thread_info.h b/include/asm-sparc64/thread_info.h index 2ebf7f27bf9..98252cd44dd 100644 --- a/include/asm-sparc64/thread_info.h +++ b/include/asm-sparc64/thread_info.h @@ -38,8 +38,8 @@ struct thread_info { /* D$ line 1 */ struct task_struct *task; unsigned long flags; - __u8 cpu; __u8 fpsaved[7]; + __u8 pad; unsigned long ksp; /* D$ line 2 */ @@ -49,7 +49,7 @@ struct thread_info { int preempt_count; /* 0 => preemptable, <0 => BUG */ __u8 new_child; __u8 syscall_noerror; - __u16 __pad; + __u16 cpu; unsigned long *utraps; @@ -83,8 +83,7 @@ struct thread_info { #define TI_CURRENT_DS (TI_FLAGS + TI_FLAG_BYTE_CURRENT_DS) #define TI_FPDEPTH (TI_FLAGS + TI_FLAG_BYTE_FPDEPTH) #define TI_WSAVED (TI_FLAGS + TI_FLAG_BYTE_WSAVED) -#define TI_CPU 0x00000010 -#define TI_FPSAVED 0x00000011 +#define TI_FPSAVED 0x00000010 #define TI_KSP 0x00000018 #define TI_FAULT_ADDR 0x00000020 #define TI_KREGS 0x00000028 @@ -92,6 +91,7 @@ struct thread_info { #define TI_PRE_COUNT 0x00000038 #define TI_NEW_CHILD 0x0000003c #define TI_SYS_NOERROR 0x0000003d +#define TI_CPU 0x0000003e #define TI_UTRAPS 0x00000040 #define TI_REG_WINDOW 0x00000048 #define TI_RWIN_SPTRS 0x000003c8 diff --git a/include/asm-sparc64/topology.h b/include/asm-sparc64/topology.h index 98a6c613589..e0d450d600e 100644 --- a/include/asm-sparc64/topology.h +++ b/include/asm-sparc64/topology.h @@ -6,4 +6,7 @@ #include <asm-generic/topology.h> +#define topology_core_id(cpu) (cpu_data(cpu).core_id) +#define topology_thread_siblings(cpu) (cpu_sibling_map[cpu]) + #endif /* _ASM_SPARC64_TOPOLOGY_H */ diff --git a/include/asm-sparc64/tsb.h b/include/asm-sparc64/tsb.h index ab55ffcb7bf..76e4299dd9b 100644 --- a/include/asm-sparc64/tsb.h +++ b/include/asm-sparc64/tsb.h @@ -271,7 +271,7 @@ extern struct tsb_phys_patch_entry __tsb_phys_patch, __tsb_phys_patch_end; #define KERN_TSB4M_LOOKUP_TL1(TAG, REG1, REG2, REG3, REG4, OK_LABEL) \ sethi %hi(swapper_4m_tsb), REG1; \ or REG1, %lo(swapper_4m_tsb), REG1; \ - and TAG, (KERNEL_TSB_NENTRIES - 1), REG2; \ + and TAG, (KERNEL_TSB4M_NENTRIES - 1), REG2; \ sllx REG2, 4, REG2; \ add REG1, REG2, REG2; \ KTSB_LOAD_QUAD(REG2, REG3); \ diff --git a/include/linux/timer.h b/include/linux/timer.h index e0c5c16c992..c661710d362 100644 --- a/include/linux/timer.h +++ b/include/linux/timer.h @@ -69,6 +69,12 @@ extern int __mod_timer(struct timer_list *timer, unsigned long expires); extern int mod_timer(struct timer_list *timer, unsigned long expires); /* + * The jiffies value which is added to now, when there is no timer + * in the timer wheel: + */ +#define NEXT_TIMER_MAX_DELTA ((1UL << 30) - 1) + +/* * Return when the next timer-wheel timeout occurs (in absolute jiffies), * locks the timer base: */ diff --git a/kernel/kallsyms.c b/kernel/kallsyms.c index f1bda23140b..fed54418626 100644 --- a/kernel/kallsyms.c +++ b/kernel/kallsyms.c @@ -257,7 +257,8 @@ const char *kallsyms_lookup(unsigned long addr, pos = get_symbol_pos(addr, symbolsize, offset); /* Grab name */ kallsyms_expand_symbol(get_symbol_offset(pos), namebuf); - *modname = NULL; + if (modname) + *modname = NULL; return namebuf; } diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c index 3e7ebc4646b..52db9e3c526 100644 --- a/kernel/time/tick-sched.c +++ b/kernel/time/tick-sched.c @@ -247,6 +247,21 @@ void tick_nohz_stop_sched_tick(void) if (cpu == tick_do_timer_cpu) tick_do_timer_cpu = -1; + ts->idle_sleeps++; + + /* + * delta_jiffies >= NEXT_TIMER_MAX_DELTA signals that + * there is no timer pending or at least extremly far + * into the future (12 days for HZ=1000). In this case + * we simply stop the tick timer: + */ + if (unlikely(delta_jiffies >= NEXT_TIMER_MAX_DELTA)) { + ts->idle_expires.tv64 = KTIME_MAX; + if (ts->nohz_mode == NOHZ_MODE_HIGHRES) + hrtimer_cancel(&ts->sched_timer); + goto out; + } + /* * calculate the expiry time for the next timer wheel * timer @@ -254,7 +269,6 @@ void tick_nohz_stop_sched_tick(void) expires = ktime_add_ns(last_update, tick_period.tv64 * delta_jiffies); ts->idle_expires = expires; - ts->idle_sleeps++; if (ts->nohz_mode == NOHZ_MODE_HIGHRES) { hrtimer_start(&ts->sched_timer, expires, diff --git a/kernel/timer.c b/kernel/timer.c index 5ec5490f8d8..1a69705c2fb 100644 --- a/kernel/timer.c +++ b/kernel/timer.c @@ -666,7 +666,7 @@ static inline void __run_timers(tvec_base_t *base) static unsigned long __next_timer_interrupt(tvec_base_t *base) { unsigned long timer_jiffies = base->timer_jiffies; - unsigned long expires = timer_jiffies + (LONG_MAX >> 1); + unsigned long expires = timer_jiffies + NEXT_TIMER_MAX_DELTA; int index, slot, array, found = 0; struct timer_list *nte; tvec_t *varray[4]; @@ -752,6 +752,14 @@ static unsigned long cmp_next_hrtimer_event(unsigned long now, tsdelta = ktime_to_timespec(hr_delta); delta = timespec_to_jiffies(&tsdelta); + + /* + * Limit the delta to the max value, which is checked in + * tick_nohz_stop_sched_tick(): + */ + if (delta > NEXT_TIMER_MAX_DELTA) + delta = NEXT_TIMER_MAX_DELTA; + /* * Take rounding errors in to account and make sure, that it * expires in the next tick. Otherwise we go into an endless diff --git a/mm/page_alloc.c b/mm/page_alloc.c index d8970623c56..bd8e33582d2 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -2689,7 +2689,7 @@ static void __init_refok alloc_node_mem_map(struct pglist_data *pgdat) map = alloc_bootmem_node(pgdat, size); pgdat->node_mem_map = map + (pgdat->node_start_pfn - start); } -#ifdef CONFIG_FLATMEM +#ifndef CONFIG_NEED_MULTIPLE_NODES /* * With no DISCONTIG, the global mem_map is just set as node 0's */ diff --git a/mm/slub.c b/mm/slub.c index 3e5aefcb407..238c5a6ee89 100644 --- a/mm/slub.c +++ b/mm/slub.c @@ -2435,6 +2435,7 @@ void __init kmem_cache_init(void) */ create_kmalloc_cache(&kmalloc_caches[0], "kmem_cache_node", sizeof(struct kmem_cache_node), GFP_KERNEL); + kmalloc_caches[0].refcount = -1; #endif /* Able to allocate the per node structures */ @@ -2482,6 +2483,12 @@ static int slab_unmergeable(struct kmem_cache *s) if (s->ctor) return 1; + /* + * We may have set a slab to be unmergeable during bootstrap. + */ + if (s->refcount < 0) + return 1; + return 0; } diff --git a/net/ieee80211/ieee80211_module.c b/net/ieee80211/ieee80211_module.c index 7ec6610841b..17ad278696e 100644 --- a/net/ieee80211/ieee80211_module.c +++ b/net/ieee80211/ieee80211_module.c @@ -140,7 +140,7 @@ struct net_device *alloc_ieee80211(int sizeof_priv) dev = alloc_etherdev(sizeof(struct ieee80211_device) + sizeof_priv); if (!dev) { - IEEE80211_ERROR("Unable to network device.\n"); + IEEE80211_ERROR("Unable to allocate network device.\n"); goto failed; } ieee = netdev_priv(dev); diff --git a/net/ieee80211/softmac/ieee80211softmac_module.c b/net/ieee80211/softmac/ieee80211softmac_module.c index e9cdc6615dd..c308756c2f9 100644 --- a/net/ieee80211/softmac/ieee80211softmac_module.c +++ b/net/ieee80211/softmac/ieee80211softmac_module.c @@ -33,7 +33,10 @@ struct net_device *alloc_ieee80211softmac(int sizeof_priv) struct ieee80211softmac_device *softmac; struct net_device *dev; - dev = alloc_ieee80211(sizeof(struct ieee80211softmac_device) + sizeof_priv); + dev = alloc_ieee80211(sizeof(*softmac) + sizeof_priv); + if (!dev) + return NULL; + softmac = ieee80211_priv(dev); softmac->dev = dev; softmac->ieee = netdev_priv(dev); |