From 23fb064bb96f001ecb8682129f7ee1bc1ca691bc Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 21 Jul 2009 21:18:35 +0900 Subject: percpu: kill legacy percpu allocator With ia64 converted, there's no arch left which still uses legacy percpu allocator. Kill it. Signed-off-by: Tejun Heo Delightedly-acked-by: Rusty Russell Cc: Ingo Molnar Cc: Christoph Lameter --- kernel/module.c | 150 -------------------------------------------------------- 1 file changed, 150 deletions(-) (limited to 'kernel') diff --git a/kernel/module.c b/kernel/module.c index 8b7d8805819d..64787cddeb5e 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -370,8 +370,6 @@ EXPORT_SYMBOL_GPL(find_module); #ifdef CONFIG_SMP -#ifndef CONFIG_HAVE_LEGACY_PER_CPU_AREA - static void *percpu_modalloc(unsigned long size, unsigned long align, const char *name) { @@ -395,154 +393,6 @@ static void percpu_modfree(void *freeme) free_percpu(freeme); } -#else /* ... CONFIG_HAVE_LEGACY_PER_CPU_AREA */ - -/* Number of blocks used and allocated. */ -static unsigned int pcpu_num_used, pcpu_num_allocated; -/* Size of each block. -ve means used. */ -static int *pcpu_size; - -static int split_block(unsigned int i, unsigned short size) -{ - /* Reallocation required? */ - if (pcpu_num_used + 1 > pcpu_num_allocated) { - int *new; - - new = krealloc(pcpu_size, sizeof(new[0])*pcpu_num_allocated*2, - GFP_KERNEL); - if (!new) - return 0; - - pcpu_num_allocated *= 2; - pcpu_size = new; - } - - /* Insert a new subblock */ - memmove(&pcpu_size[i+1], &pcpu_size[i], - sizeof(pcpu_size[0]) * (pcpu_num_used - i)); - pcpu_num_used++; - - pcpu_size[i+1] -= size; - pcpu_size[i] = size; - return 1; -} - -static inline unsigned int block_size(int val) -{ - if (val < 0) - return -val; - return val; -} - -static void *percpu_modalloc(unsigned long size, unsigned long align, - const char *name) -{ - unsigned long extra; - unsigned int i; - void *ptr; - int cpu; - - if (align > PAGE_SIZE) { - printk(KERN_WARNING "%s: per-cpu alignment %li > %li\n", - name, align, PAGE_SIZE); - align = PAGE_SIZE; - } - - ptr = __per_cpu_start; - for (i = 0; i < pcpu_num_used; ptr += block_size(pcpu_size[i]), i++) { - /* Extra for alignment requirement. */ - extra = ALIGN((unsigned long)ptr, align) - (unsigned long)ptr; - BUG_ON(i == 0 && extra != 0); - - if (pcpu_size[i] < 0 || pcpu_size[i] < extra + size) - continue; - - /* Transfer extra to previous block. */ - if (pcpu_size[i-1] < 0) - pcpu_size[i-1] -= extra; - else - pcpu_size[i-1] += extra; - pcpu_size[i] -= extra; - ptr += extra; - - /* Split block if warranted */ - if (pcpu_size[i] - size > sizeof(unsigned long)) - if (!split_block(i, size)) - return NULL; - - /* add the per-cpu scanning areas */ - for_each_possible_cpu(cpu) - kmemleak_alloc(ptr + per_cpu_offset(cpu), size, 0, - GFP_KERNEL); - - /* Mark allocated */ - pcpu_size[i] = -pcpu_size[i]; - return ptr; - } - - printk(KERN_WARNING "Could not allocate %lu bytes percpu data\n", - size); - return NULL; -} - -static void percpu_modfree(void *freeme) -{ - unsigned int i; - void *ptr = __per_cpu_start + block_size(pcpu_size[0]); - int cpu; - - /* First entry is core kernel percpu data. */ - for (i = 1; i < pcpu_num_used; ptr += block_size(pcpu_size[i]), i++) { - if (ptr == freeme) { - pcpu_size[i] = -pcpu_size[i]; - goto free; - } - } - BUG(); - - free: - /* remove the per-cpu scanning areas */ - for_each_possible_cpu(cpu) - kmemleak_free(freeme + per_cpu_offset(cpu)); - - /* Merge with previous? */ - if (pcpu_size[i-1] >= 0) { - pcpu_size[i-1] += pcpu_size[i]; - pcpu_num_used--; - memmove(&pcpu_size[i], &pcpu_size[i+1], - (pcpu_num_used - i) * sizeof(pcpu_size[0])); - i--; - } - /* Merge with next? */ - if (i+1 < pcpu_num_used && pcpu_size[i+1] >= 0) { - pcpu_size[i] += pcpu_size[i+1]; - pcpu_num_used--; - memmove(&pcpu_size[i+1], &pcpu_size[i+2], - (pcpu_num_used - (i+1)) * sizeof(pcpu_size[0])); - } -} - -static int percpu_modinit(void) -{ - pcpu_num_used = 2; - pcpu_num_allocated = 2; - pcpu_size = kmalloc(sizeof(pcpu_size[0]) * pcpu_num_allocated, - GFP_KERNEL); - /* Static in-kernel percpu data (used). */ - pcpu_size[0] = -(__per_cpu_end-__per_cpu_start); - /* Free room. */ - pcpu_size[1] = PERCPU_ENOUGH_ROOM + pcpu_size[0]; - if (pcpu_size[1] < 0) { - printk(KERN_ERR "No per-cpu room for modules.\n"); - pcpu_num_used = 1; - } - - return 0; -} -__initcall(percpu_modinit); - -#endif /* CONFIG_HAVE_LEGACY_PER_CPU_AREA */ - static unsigned int find_pcpusec(Elf_Ehdr *hdr, Elf_Shdr *sechdrs, const char *secstrings) -- cgit v1.2.3-71-gd317 From e800879d50c5a528d40191528557b1bdfbccbd42 Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Sat, 3 Oct 2009 19:48:23 +0900 Subject: this_cpu: Use this_cpu operations in RCU RCU does not do dynamic allocations but it increments per cpu variables a lot. These instructions results in a move to a register and then back to memory. This patch will make it use the inc/dec instructions on x86 that do not need a register. Acked-by: Tejun Heo Acked-by: Paul E. McKenney Signed-off-by: Christoph Lameter Signed-off-by: Tejun Heo --- kernel/rcutorture.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'kernel') diff --git a/kernel/rcutorture.c b/kernel/rcutorture.c index 233768f21f97..178967b6434e 100644 --- a/kernel/rcutorture.c +++ b/kernel/rcutorture.c @@ -731,13 +731,13 @@ static void rcu_torture_timer(unsigned long unused) /* Should not happen, but... */ pipe_count = RCU_TORTURE_PIPE_LEN; } - ++__get_cpu_var(rcu_torture_count)[pipe_count]; + __this_cpu_inc(per_cpu_var(rcu_torture_count)[pipe_count]); completed = cur_ops->completed() - completed; if (completed > RCU_TORTURE_PIPE_LEN) { /* Should not happen, but... */ completed = RCU_TORTURE_PIPE_LEN; } - ++__get_cpu_var(rcu_torture_batch)[completed]; + __this_cpu_inc(per_cpu_var(rcu_torture_batch)[completed]); preempt_enable(); cur_ops->readunlock(idx); } @@ -786,13 +786,13 @@ rcu_torture_reader(void *arg) /* Should not happen, but... */ pipe_count = RCU_TORTURE_PIPE_LEN; } - ++__get_cpu_var(rcu_torture_count)[pipe_count]; + __this_cpu_inc(per_cpu_var(rcu_torture_count)[pipe_count]); completed = cur_ops->completed() - completed; if (completed > RCU_TORTURE_PIPE_LEN) { /* Should not happen, but... */ completed = RCU_TORTURE_PIPE_LEN; } - ++__get_cpu_var(rcu_torture_batch)[completed]; + __this_cpu_inc(per_cpu_var(rcu_torture_batch)[completed]); preempt_enable(); cur_ops->readunlock(idx); schedule(); -- cgit v1.2.3-71-gd317 From 9288f99aa52d90a5b82573c4b769c97c55af2f56 Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Wed, 7 Oct 2009 19:17:45 -0400 Subject: this_cpu: Use this_cpu_xx for ftrace this_cpu_xx can reduce the instruction count here and also avoid address arithmetic. Signed-off-by: Christoph Lameter Acked-by: Ingo Molnar Signed-off-by: Tejun Heo --- kernel/trace/trace.c | 8 ++++---- kernel/trace/trace.h | 2 +- 2 files changed, 5 insertions(+), 5 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 45068269ebb1..8439cdcada94 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -86,17 +86,17 @@ static int dummy_set_flag(u32 old_flags, u32 bit, int set) */ static int tracing_disabled = 1; -DEFINE_PER_CPU(local_t, ftrace_cpu_disabled); +DEFINE_PER_CPU(int, ftrace_cpu_disabled); static inline void ftrace_disable_cpu(void) { preempt_disable(); - local_inc(&__get_cpu_var(ftrace_cpu_disabled)); + __this_cpu_inc(per_cpu_var(ftrace_cpu_disabled)); } static inline void ftrace_enable_cpu(void) { - local_dec(&__get_cpu_var(ftrace_cpu_disabled)); + __this_cpu_dec(per_cpu_var(ftrace_cpu_disabled)); preempt_enable(); } @@ -1085,7 +1085,7 @@ trace_function(struct trace_array *tr, struct ftrace_entry *entry; /* If we are reading the ring buffer, don't trace */ - if (unlikely(local_read(&__get_cpu_var(ftrace_cpu_disabled)))) + if (unlikely(__this_cpu_read(per_cpu_var(ftrace_cpu_disabled)))) return; event = trace_buffer_lock_reserve(buffer, TRACE_FN, sizeof(*entry), diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h index 405cb850b75d..542f45554883 100644 --- a/kernel/trace/trace.h +++ b/kernel/trace/trace.h @@ -413,7 +413,7 @@ extern int DYN_FTRACE_TEST_NAME(void); extern int ring_buffer_expanded; extern bool tracing_selftest_disabled; -DECLARE_PER_CPU(local_t, ftrace_cpu_disabled); +DECLARE_PER_CPU(int, ftrace_cpu_disabled); #ifdef CONFIG_FTRACE_STARTUP_TEST extern int trace_selftest_startup_function(struct tracer *trace, -- cgit v1.2.3-71-gd317 From dec54bf538326a1503dd780c9f2811f495af95c5 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 13 Oct 2009 23:23:02 +0900 Subject: this_cpu: Use this_cpu_xx in trace_functions_graph.c ftrace_cpu_disabled usage in trace_functions_graph.c were left out during this_cpu_xx conversion in commit 9288f99a causing compile failure. Convert them. Signed-off-by: Tejun Heo Reported-by: Stephen Rothwell Cc: Christoph Lameter --- kernel/trace/trace_functions_graph.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/trace_functions_graph.c b/kernel/trace/trace_functions_graph.c index 45e6c01b2e4d..90a6daa10962 100644 --- a/kernel/trace/trace_functions_graph.c +++ b/kernel/trace/trace_functions_graph.c @@ -176,7 +176,7 @@ static int __trace_graph_entry(struct trace_array *tr, struct ring_buffer *buffer = tr->buffer; struct ftrace_graph_ent_entry *entry; - if (unlikely(local_read(&__get_cpu_var(ftrace_cpu_disabled)))) + if (unlikely(__this_cpu_read(per_cpu_var(ftrace_cpu_disabled)))) return 0; event = trace_buffer_lock_reserve(buffer, TRACE_GRAPH_ENT, @@ -240,7 +240,7 @@ static void __trace_graph_return(struct trace_array *tr, struct ring_buffer *buffer = tr->buffer; struct ftrace_graph_ret_entry *entry; - if (unlikely(local_read(&__get_cpu_var(ftrace_cpu_disabled)))) + if (unlikely(__this_cpu_read(per_cpu_var(ftrace_cpu_disabled)))) return; event = trace_buffer_lock_reserve(buffer, TRACE_GRAPH_RET, -- cgit v1.2.3-71-gd317 From 1871e52c76dd95895caeb772f845a1718dcbcd75 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Thu, 29 Oct 2009 22:34:13 +0900 Subject: percpu: make percpu symbols under kernel/ and mm/ unique This patch updates percpu related symbols under kernel/ and mm/ such that percpu symbols are unique and don't clash with local symbols. This serves two purposes of decreasing the possibility of global percpu symbol collision and allowing dropping per_cpu__ prefix from percpu symbols. * kernel/lockdep.c: s/lock_stats/cpu_lock_stats/ * kernel/sched.c: s/init_rq_rt/init_rt_rq_var/ (any better idea?) s/sched_group_cpus/sched_groups/ * kernel/softirq.c: s/ksoftirqd/run_ksoftirqd/a * kernel/softlockup.c: s/(*)_timestamp/softlockup_\1_ts/ s/watchdog_task/softlockup_watchdog/ s/timestamp/ts/ for local variables * kernel/time/timer_stats: s/lookup_lock/tstats_lookup_lock/ * mm/slab.c: s/reap_work/slab_reap_work/ s/reap_node/slab_reap_node/ * mm/vmstat.c: local variable changed to avoid collision with vmstat_work Partly based on Rusty Russell's "alloc_percpu: rename percpu vars which cause name clashes" patch. Signed-off-by: Tejun Heo Acked-by: (slab/vmstat) Christoph Lameter Reviewed-by: Christoph Lameter Cc: Rusty Russell Cc: Ingo Molnar Cc: Thomas Gleixner Cc: Andrew Morton Cc: Nick Piggin --- kernel/lockdep.c | 11 +++++----- kernel/sched.c | 8 +++---- kernel/softirq.c | 4 ++-- kernel/softlockup.c | 54 +++++++++++++++++++++++------------------------ kernel/time/timer_stats.c | 11 +++++----- mm/slab.c | 18 ++++++++-------- mm/vmstat.c | 7 +++--- 7 files changed, 57 insertions(+), 56 deletions(-) (limited to 'kernel') diff --git a/kernel/lockdep.c b/kernel/lockdep.c index 3815ac1d58b2..8631320a50d0 100644 --- a/kernel/lockdep.c +++ b/kernel/lockdep.c @@ -140,7 +140,8 @@ static inline struct lock_class *hlock_class(struct held_lock *hlock) } #ifdef CONFIG_LOCK_STAT -static DEFINE_PER_CPU(struct lock_class_stats[MAX_LOCKDEP_KEYS], lock_stats); +static DEFINE_PER_CPU(struct lock_class_stats[MAX_LOCKDEP_KEYS], + cpu_lock_stats); static int lock_point(unsigned long points[], unsigned long ip) { @@ -186,7 +187,7 @@ struct lock_class_stats lock_stats(struct lock_class *class) memset(&stats, 0, sizeof(struct lock_class_stats)); for_each_possible_cpu(cpu) { struct lock_class_stats *pcs = - &per_cpu(lock_stats, cpu)[class - lock_classes]; + &per_cpu(cpu_lock_stats, cpu)[class - lock_classes]; for (i = 0; i < ARRAY_SIZE(stats.contention_point); i++) stats.contention_point[i] += pcs->contention_point[i]; @@ -213,7 +214,7 @@ void clear_lock_stats(struct lock_class *class) for_each_possible_cpu(cpu) { struct lock_class_stats *cpu_stats = - &per_cpu(lock_stats, cpu)[class - lock_classes]; + &per_cpu(cpu_lock_stats, cpu)[class - lock_classes]; memset(cpu_stats, 0, sizeof(struct lock_class_stats)); } @@ -223,12 +224,12 @@ void clear_lock_stats(struct lock_class *class) static struct lock_class_stats *get_lock_stats(struct lock_class *class) { - return &get_cpu_var(lock_stats)[class - lock_classes]; + return &get_cpu_var(cpu_lock_stats)[class - lock_classes]; } static void put_lock_stats(struct lock_class_stats *stats) { - put_cpu_var(lock_stats); + put_cpu_var(cpu_lock_stats); } static void lock_release_holdtime(struct held_lock *hlock) diff --git a/kernel/sched.c b/kernel/sched.c index 1535f3884b88..854ab418fd42 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -298,7 +298,7 @@ static DEFINE_PER_CPU_SHARED_ALIGNED(struct cfs_rq, init_tg_cfs_rq); #ifdef CONFIG_RT_GROUP_SCHED static DEFINE_PER_CPU(struct sched_rt_entity, init_sched_rt_entity); -static DEFINE_PER_CPU_SHARED_ALIGNED(struct rt_rq, init_rt_rq); +static DEFINE_PER_CPU_SHARED_ALIGNED(struct rt_rq, init_rt_rq_var); #endif /* CONFIG_RT_GROUP_SCHED */ #else /* !CONFIG_USER_SCHED */ #define root_task_group init_task_group @@ -8199,14 +8199,14 @@ enum s_alloc { */ #ifdef CONFIG_SCHED_SMT static DEFINE_PER_CPU(struct static_sched_domain, cpu_domains); -static DEFINE_PER_CPU(struct static_sched_group, sched_group_cpus); +static DEFINE_PER_CPU(struct static_sched_group, sched_groups); static int cpu_to_cpu_group(int cpu, const struct cpumask *cpu_map, struct sched_group **sg, struct cpumask *unused) { if (sg) - *sg = &per_cpu(sched_group_cpus, cpu).sg; + *sg = &per_cpu(sched_groups, cpu).sg; return cpu; } #endif /* CONFIG_SCHED_SMT */ @@ -9470,7 +9470,7 @@ void __init sched_init(void) #elif defined CONFIG_USER_SCHED init_tg_rt_entry(&root_task_group, &rq->rt, NULL, i, 0, NULL); init_tg_rt_entry(&init_task_group, - &per_cpu(init_rt_rq, i), + &per_cpu(init_rt_rq_var, i), &per_cpu(init_sched_rt_entity, i), i, 1, root_task_group.rt_se[i]); #endif diff --git a/kernel/softirq.c b/kernel/softirq.c index f8749e5216e0..0740dfd55c51 100644 --- a/kernel/softirq.c +++ b/kernel/softirq.c @@ -697,7 +697,7 @@ void __init softirq_init(void) open_softirq(HI_SOFTIRQ, tasklet_hi_action); } -static int ksoftirqd(void * __bind_cpu) +static int run_ksoftirqd(void * __bind_cpu) { set_current_state(TASK_INTERRUPTIBLE); @@ -810,7 +810,7 @@ static int __cpuinit cpu_callback(struct notifier_block *nfb, switch (action) { case CPU_UP_PREPARE: case CPU_UP_PREPARE_FROZEN: - p = kthread_create(ksoftirqd, hcpu, "ksoftirqd/%d", hotcpu); + p = kthread_create(run_ksoftirqd, hcpu, "ksoftirqd/%d", hotcpu); if (IS_ERR(p)) { printk("ksoftirqd for %i failed\n", hotcpu); return NOTIFY_BAD; diff --git a/kernel/softlockup.c b/kernel/softlockup.c index 81324d12eb35..d22579087e27 100644 --- a/kernel/softlockup.c +++ b/kernel/softlockup.c @@ -22,9 +22,9 @@ static DEFINE_SPINLOCK(print_lock); -static DEFINE_PER_CPU(unsigned long, touch_timestamp); -static DEFINE_PER_CPU(unsigned long, print_timestamp); -static DEFINE_PER_CPU(struct task_struct *, watchdog_task); +static DEFINE_PER_CPU(unsigned long, softlockup_touch_ts); /* touch timestamp */ +static DEFINE_PER_CPU(unsigned long, softlockup_print_ts); /* print timestamp */ +static DEFINE_PER_CPU(struct task_struct *, softlockup_watchdog); static int __read_mostly did_panic; int __read_mostly softlockup_thresh = 60; @@ -70,12 +70,12 @@ static void __touch_softlockup_watchdog(void) { int this_cpu = raw_smp_processor_id(); - __raw_get_cpu_var(touch_timestamp) = get_timestamp(this_cpu); + __raw_get_cpu_var(softlockup_touch_ts) = get_timestamp(this_cpu); } void touch_softlockup_watchdog(void) { - __raw_get_cpu_var(touch_timestamp) = 0; + __raw_get_cpu_var(softlockup_touch_ts) = 0; } EXPORT_SYMBOL(touch_softlockup_watchdog); @@ -85,7 +85,7 @@ void touch_all_softlockup_watchdogs(void) /* Cause each CPU to re-update its timestamp rather than complain */ for_each_online_cpu(cpu) - per_cpu(touch_timestamp, cpu) = 0; + per_cpu(softlockup_touch_ts, cpu) = 0; } EXPORT_SYMBOL(touch_all_softlockup_watchdogs); @@ -104,28 +104,28 @@ int proc_dosoftlockup_thresh(struct ctl_table *table, int write, void softlockup_tick(void) { int this_cpu = smp_processor_id(); - unsigned long touch_timestamp = per_cpu(touch_timestamp, this_cpu); - unsigned long print_timestamp; + unsigned long touch_ts = per_cpu(softlockup_touch_ts, this_cpu); + unsigned long print_ts; struct pt_regs *regs = get_irq_regs(); unsigned long now; /* Is detection switched off? */ - if (!per_cpu(watchdog_task, this_cpu) || softlockup_thresh <= 0) { + if (!per_cpu(softlockup_watchdog, this_cpu) || softlockup_thresh <= 0) { /* Be sure we don't false trigger if switched back on */ - if (touch_timestamp) - per_cpu(touch_timestamp, this_cpu) = 0; + if (touch_ts) + per_cpu(softlockup_touch_ts, this_cpu) = 0; return; } - if (touch_timestamp == 0) { + if (touch_ts == 0) { __touch_softlockup_watchdog(); return; } - print_timestamp = per_cpu(print_timestamp, this_cpu); + print_ts = per_cpu(softlockup_print_ts, this_cpu); /* report at most once a second */ - if (print_timestamp == touch_timestamp || did_panic) + if (print_ts == touch_ts || did_panic) return; /* do not print during early bootup: */ @@ -140,18 +140,18 @@ void softlockup_tick(void) * Wake up the high-prio watchdog task twice per * threshold timespan. */ - if (now > touch_timestamp + softlockup_thresh/2) - wake_up_process(per_cpu(watchdog_task, this_cpu)); + if (now > touch_ts + softlockup_thresh/2) + wake_up_process(per_cpu(softlockup_watchdog, this_cpu)); /* Warn about unreasonable delays: */ - if (now <= (touch_timestamp + softlockup_thresh)) + if (now <= (touch_ts + softlockup_thresh)) return; - per_cpu(print_timestamp, this_cpu) = touch_timestamp; + per_cpu(softlockup_print_ts, this_cpu) = touch_ts; spin_lock(&print_lock); printk(KERN_ERR "BUG: soft lockup - CPU#%d stuck for %lus! [%s:%d]\n", - this_cpu, now - touch_timestamp, + this_cpu, now - touch_ts, current->comm, task_pid_nr(current)); print_modules(); print_irqtrace_events(current); @@ -209,32 +209,32 @@ cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu) switch (action) { case CPU_UP_PREPARE: case CPU_UP_PREPARE_FROZEN: - BUG_ON(per_cpu(watchdog_task, hotcpu)); + BUG_ON(per_cpu(softlockup_watchdog, hotcpu)); p = kthread_create(watchdog, hcpu, "watchdog/%d", hotcpu); if (IS_ERR(p)) { printk(KERN_ERR "watchdog for %i failed\n", hotcpu); return NOTIFY_BAD; } - per_cpu(touch_timestamp, hotcpu) = 0; - per_cpu(watchdog_task, hotcpu) = p; + per_cpu(softlockup_touch_ts, hotcpu) = 0; + per_cpu(softlockup_watchdog, hotcpu) = p; kthread_bind(p, hotcpu); break; case CPU_ONLINE: case CPU_ONLINE_FROZEN: - wake_up_process(per_cpu(watchdog_task, hotcpu)); + wake_up_process(per_cpu(softlockup_watchdog, hotcpu)); break; #ifdef CONFIG_HOTPLUG_CPU case CPU_UP_CANCELED: case CPU_UP_CANCELED_FROZEN: - if (!per_cpu(watchdog_task, hotcpu)) + if (!per_cpu(softlockup_watchdog, hotcpu)) break; /* Unbind so it can run. Fall thru. */ - kthread_bind(per_cpu(watchdog_task, hotcpu), + kthread_bind(per_cpu(softlockup_watchdog, hotcpu), cpumask_any(cpu_online_mask)); case CPU_DEAD: case CPU_DEAD_FROZEN: - p = per_cpu(watchdog_task, hotcpu); - per_cpu(watchdog_task, hotcpu) = NULL; + p = per_cpu(softlockup_watchdog, hotcpu); + per_cpu(softlockup_watchdog, hotcpu) = NULL; kthread_stop(p); break; #endif /* CONFIG_HOTPLUG_CPU */ diff --git a/kernel/time/timer_stats.c b/kernel/time/timer_stats.c index ee5681f8d7ec..63b117e9eba1 100644 --- a/kernel/time/timer_stats.c +++ b/kernel/time/timer_stats.c @@ -86,7 +86,7 @@ static DEFINE_SPINLOCK(table_lock); /* * Per-CPU lookup locks for fast hash lookup: */ -static DEFINE_PER_CPU(spinlock_t, lookup_lock); +static DEFINE_PER_CPU(spinlock_t, tstats_lookup_lock); /* * Mutex to serialize state changes with show-stats activities: @@ -245,7 +245,7 @@ void timer_stats_update_stats(void *timer, pid_t pid, void *startf, if (likely(!timer_stats_active)) return; - lock = &per_cpu(lookup_lock, raw_smp_processor_id()); + lock = &per_cpu(tstats_lookup_lock, raw_smp_processor_id()); input.timer = timer; input.start_func = startf; @@ -348,9 +348,10 @@ static void sync_access(void) int cpu; for_each_online_cpu(cpu) { - spin_lock_irqsave(&per_cpu(lookup_lock, cpu), flags); + spinlock_t *lock = &per_cpu(tstats_lookup_lock, cpu); + spin_lock_irqsave(lock, flags); /* nothing */ - spin_unlock_irqrestore(&per_cpu(lookup_lock, cpu), flags); + spin_unlock_irqrestore(lock, flags); } } @@ -408,7 +409,7 @@ void __init init_timer_stats(void) int cpu; for_each_possible_cpu(cpu) - spin_lock_init(&per_cpu(lookup_lock, cpu)); + spin_lock_init(&per_cpu(tstats_lookup_lock, cpu)); } static int __init init_tstats_procfs(void) diff --git a/mm/slab.c b/mm/slab.c index 7dfa481c96ba..211b1746c63c 100644 --- a/mm/slab.c +++ b/mm/slab.c @@ -685,7 +685,7 @@ int slab_is_available(void) return g_cpucache_up >= EARLY; } -static DEFINE_PER_CPU(struct delayed_work, reap_work); +static DEFINE_PER_CPU(struct delayed_work, slab_reap_work); static inline struct array_cache *cpu_cache_get(struct kmem_cache *cachep) { @@ -826,7 +826,7 @@ __setup("noaliencache", noaliencache_setup); * objects freed on different nodes from which they were allocated) and the * flushing of remote pcps by calling drain_node_pages. */ -static DEFINE_PER_CPU(unsigned long, reap_node); +static DEFINE_PER_CPU(unsigned long, slab_reap_node); static void init_reap_node(int cpu) { @@ -836,17 +836,17 @@ static void init_reap_node(int cpu) if (node == MAX_NUMNODES) node = first_node(node_online_map); - per_cpu(reap_node, cpu) = node; + per_cpu(slab_reap_node, cpu) = node; } static void next_reap_node(void) { - int node = __get_cpu_var(reap_node); + int node = __get_cpu_var(slab_reap_node); node = next_node(node, node_online_map); if (unlikely(node >= MAX_NUMNODES)) node = first_node(node_online_map); - __get_cpu_var(reap_node) = node; + __get_cpu_var(slab_reap_node) = node; } #else @@ -863,7 +863,7 @@ static void next_reap_node(void) */ static void __cpuinit start_cpu_timer(int cpu) { - struct delayed_work *reap_work = &per_cpu(reap_work, cpu); + struct delayed_work *reap_work = &per_cpu(slab_reap_work, cpu); /* * When this gets called from do_initcalls via cpucache_init(), @@ -1027,7 +1027,7 @@ static void __drain_alien_cache(struct kmem_cache *cachep, */ static void reap_alien(struct kmem_cache *cachep, struct kmem_list3 *l3) { - int node = __get_cpu_var(reap_node); + int node = __get_cpu_var(slab_reap_node); if (l3->alien) { struct array_cache *ac = l3->alien[node]; @@ -1286,9 +1286,9 @@ static int __cpuinit cpuup_callback(struct notifier_block *nfb, * anything expensive but will only modify reap_work * and reschedule the timer. */ - cancel_rearming_delayed_work(&per_cpu(reap_work, cpu)); + cancel_rearming_delayed_work(&per_cpu(slab_reap_work, cpu)); /* Now the cache_reaper is guaranteed to be not running. */ - per_cpu(reap_work, cpu).work.func = NULL; + per_cpu(slab_reap_work, cpu).work.func = NULL; break; case CPU_DOWN_FAILED: case CPU_DOWN_FAILED_FROZEN: diff --git a/mm/vmstat.c b/mm/vmstat.c index c81321f9feec..dad2327e4580 100644 --- a/mm/vmstat.c +++ b/mm/vmstat.c @@ -883,11 +883,10 @@ static void vmstat_update(struct work_struct *w) static void __cpuinit start_cpu_timer(int cpu) { - struct delayed_work *vmstat_work = &per_cpu(vmstat_work, cpu); + struct delayed_work *work = &per_cpu(vmstat_work, cpu); - INIT_DELAYED_WORK_DEFERRABLE(vmstat_work, vmstat_update); - schedule_delayed_work_on(cpu, vmstat_work, - __round_jiffies_relative(HZ, cpu)); + INIT_DELAYED_WORK_DEFERRABLE(work, vmstat_update); + schedule_delayed_work_on(cpu, work, __round_jiffies_relative(HZ, cpu)); } /* -- cgit v1.2.3-71-gd317 From 9705f69ed0a5ef593f45e618bcb3cbfdbf391f64 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Thu, 29 Oct 2009 22:34:13 +0900 Subject: percpu: make percpu symbols in tracer unique This patch updates percpu related symbols in kernel tracer such that percpu symbols are unique and don't clash with local symbols. This serves two purposes of decreasing the possibility of global percpu symbol collision and allowing dropping per_cpu__ prefix from percpu symbols. * kernel/trace/trace.c: s/max_data/max_tr_data/ * kernel/trace/trace_hw_branches: s/tracer/hwb_tracer/, s/buffer/hwb_buffer/ Partly based on Rusty Russell's "alloc_percpu: rename percpu vars which cause name clashes" patch. Signed-off-by: Tejun Heo Acked-by: Steven Rostedt Cc: Rusty Russell Cc: Frederic Weisbecker Cc: Ingo Molnar --- kernel/trace/trace.c | 4 ++-- kernel/trace/trace_hw_branches.c | 51 ++++++++++++++++++++-------------------- 2 files changed, 28 insertions(+), 27 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 8439cdcada94..85a5ed70b5b2 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -203,7 +203,7 @@ cycle_t ftrace_now(int cpu) */ static struct trace_array max_tr; -static DEFINE_PER_CPU(struct trace_array_cpu, max_data); +static DEFINE_PER_CPU(struct trace_array_cpu, max_tr_data); /* tracer_enabled is used to toggle activation of a tracer */ static int tracer_enabled = 1; @@ -4426,7 +4426,7 @@ __init static int tracer_alloc_buffers(void) /* Allocate the first page for all buffers */ for_each_tracing_cpu(i) { global_trace.data[i] = &per_cpu(global_trace_cpu, i); - max_tr.data[i] = &per_cpu(max_data, i); + max_tr.data[i] = &per_cpu(max_tr_data, i); } trace_init_cmdlines(); diff --git a/kernel/trace/trace_hw_branches.c b/kernel/trace/trace_hw_branches.c index 23b63859130e..adaf7a39d0dc 100644 --- a/kernel/trace/trace_hw_branches.c +++ b/kernel/trace/trace_hw_branches.c @@ -20,10 +20,10 @@ #define BTS_BUFFER_SIZE (1 << 13) -static DEFINE_PER_CPU(struct bts_tracer *, tracer); -static DEFINE_PER_CPU(unsigned char[BTS_BUFFER_SIZE], buffer); +static DEFINE_PER_CPU(struct bts_tracer *, hwb_tracer); +static DEFINE_PER_CPU(unsigned char[BTS_BUFFER_SIZE], hwb_buffer); -#define this_tracer per_cpu(tracer, smp_processor_id()) +#define this_tracer per_cpu(hwb_tracer, smp_processor_id()) static int trace_hw_branches_enabled __read_mostly; static int trace_hw_branches_suspended __read_mostly; @@ -32,12 +32,13 @@ static struct trace_array *hw_branch_trace __read_mostly; static void bts_trace_init_cpu(int cpu) { - per_cpu(tracer, cpu) = - ds_request_bts_cpu(cpu, per_cpu(buffer, cpu), BTS_BUFFER_SIZE, - NULL, (size_t)-1, BTS_KERNEL); + per_cpu(hwb_tracer, cpu) = + ds_request_bts_cpu(cpu, per_cpu(hwb_buffer, cpu), + BTS_BUFFER_SIZE, NULL, (size_t)-1, + BTS_KERNEL); - if (IS_ERR(per_cpu(tracer, cpu))) - per_cpu(tracer, cpu) = NULL; + if (IS_ERR(per_cpu(hwb_tracer, cpu))) + per_cpu(hwb_tracer, cpu) = NULL; } static int bts_trace_init(struct trace_array *tr) @@ -51,7 +52,7 @@ static int bts_trace_init(struct trace_array *tr) for_each_online_cpu(cpu) { bts_trace_init_cpu(cpu); - if (likely(per_cpu(tracer, cpu))) + if (likely(per_cpu(hwb_tracer, cpu))) trace_hw_branches_enabled = 1; } trace_hw_branches_suspended = 0; @@ -67,9 +68,9 @@ static void bts_trace_reset(struct trace_array *tr) get_online_cpus(); for_each_online_cpu(cpu) { - if (likely(per_cpu(tracer, cpu))) { - ds_release_bts(per_cpu(tracer, cpu)); - per_cpu(tracer, cpu) = NULL; + if (likely(per_cpu(hwb_tracer, cpu))) { + ds_release_bts(per_cpu(hwb_tracer, cpu)); + per_cpu(hwb_tracer, cpu) = NULL; } } trace_hw_branches_enabled = 0; @@ -83,8 +84,8 @@ static void bts_trace_start(struct trace_array *tr) get_online_cpus(); for_each_online_cpu(cpu) - if (likely(per_cpu(tracer, cpu))) - ds_resume_bts(per_cpu(tracer, cpu)); + if (likely(per_cpu(hwb_tracer, cpu))) + ds_resume_bts(per_cpu(hwb_tracer, cpu)); trace_hw_branches_suspended = 0; put_online_cpus(); } @@ -95,8 +96,8 @@ static void bts_trace_stop(struct trace_array *tr) get_online_cpus(); for_each_online_cpu(cpu) - if (likely(per_cpu(tracer, cpu))) - ds_suspend_bts(per_cpu(tracer, cpu)); + if (likely(per_cpu(hwb_tracer, cpu))) + ds_suspend_bts(per_cpu(hwb_tracer, cpu)); trace_hw_branches_suspended = 1; put_online_cpus(); } @@ -114,16 +115,16 @@ static int __cpuinit bts_hotcpu_handler(struct notifier_block *nfb, bts_trace_init_cpu(cpu); if (trace_hw_branches_suspended && - likely(per_cpu(tracer, cpu))) - ds_suspend_bts(per_cpu(tracer, cpu)); + likely(per_cpu(hwb_tracer, cpu))) + ds_suspend_bts(per_cpu(hwb_tracer, cpu)); } break; case CPU_DOWN_PREPARE: /* The notification is sent with interrupts enabled. */ - if (likely(per_cpu(tracer, cpu))) { - ds_release_bts(per_cpu(tracer, cpu)); - per_cpu(tracer, cpu) = NULL; + if (likely(per_cpu(hwb_tracer, cpu))) { + ds_release_bts(per_cpu(hwb_tracer, cpu)); + per_cpu(hwb_tracer, cpu) = NULL; } } @@ -256,8 +257,8 @@ static void trace_bts_prepare(struct trace_iterator *iter) get_online_cpus(); for_each_online_cpu(cpu) - if (likely(per_cpu(tracer, cpu))) - ds_suspend_bts(per_cpu(tracer, cpu)); + if (likely(per_cpu(hwb_tracer, cpu))) + ds_suspend_bts(per_cpu(hwb_tracer, cpu)); /* * We need to collect the trace on the respective cpu since ftrace * implicitly adds the record for the current cpu. @@ -266,8 +267,8 @@ static void trace_bts_prepare(struct trace_iterator *iter) on_each_cpu(trace_bts_cpu, iter->tr, 1); for_each_online_cpu(cpu) - if (likely(per_cpu(tracer, cpu))) - ds_resume_bts(per_cpu(tracer, cpu)); + if (likely(per_cpu(hwb_tracer, cpu))) + ds_resume_bts(per_cpu(hwb_tracer, cpu)); put_online_cpus(); } -- cgit v1.2.3-71-gd317 From 1e5ad9679016275d422e36b12a98b0927d76f556 Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Mon, 2 Nov 2009 10:45:36 -0700 Subject: resources: when allocate_resource() fails, leave resource untouched When "allocate_resource(root, new, size, ...)" fails, we currently clobber "new". This is inconvenient for the caller, who might care about the original contents of the resource. For example, when pci_bus_alloc_resource() fails, the "can't allocate mem resource %pR" message from pci_assign_resources() currently contains junk for the resource start/end. This patch delays the "new" update until we're about to return success. Acked-by: Linus Torvalds Acked-by: Yinghai Lu Signed-off-by: Bjorn Helgaas Signed-off-by: Jesse Barnes --- kernel/resource.c | 26 ++++++++++++++------------ 1 file changed, 14 insertions(+), 12 deletions(-) (limited to 'kernel') diff --git a/kernel/resource.c b/kernel/resource.c index fb11a58b9594..dc15686b7a77 100644 --- a/kernel/resource.c +++ b/kernel/resource.c @@ -308,35 +308,37 @@ static int find_resource(struct resource *root, struct resource *new, void *alignf_data) { struct resource *this = root->child; + resource_size_t start, end; - new->start = root->start; + start = root->start; /* * Skip past an allocated resource that starts at 0, since the assignment * of this->start - 1 to new->end below would cause an underflow. */ if (this && this->start == 0) { - new->start = this->end + 1; + start = this->end + 1; this = this->sibling; } for(;;) { if (this) - new->end = this->start - 1; + end = this->start - 1; else - new->end = root->end; - if (new->start < min) - new->start = min; - if (new->end > max) - new->end = max; - new->start = ALIGN(new->start, align); + end = root->end; + if (start < min) + start = min; + if (end > max) + end = max; + start = ALIGN(start, align); if (alignf) alignf(alignf_data, new, size, align); - if (new->start < new->end && new->end - new->start >= size - 1) { - new->end = new->start + size - 1; + if (start < end && end - start >= size - 1) { + new->start = start; + new->end = start + size - 1; return 0; } if (!this) break; - new->start = this->end + 1; + start = this->end + 1; this = this->sibling; } return -EBUSY; -- cgit v1.2.3-71-gd317 From dc186ad741c12ae9ecac8b89e317ef706fdaf8f6 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Mon, 16 Nov 2009 01:09:48 +0900 Subject: workqueue: Add debugobjects support Add debugobject support to track the life time of work_structs. While at it, remove duplicate definition of INIT_DELAYED_WORK_ON_STACK(). Signed-off-by: Thomas Gleixner Signed-off-by: Tejun Heo --- arch/x86/kernel/smpboot.c | 4 +- include/linux/workqueue.h | 38 ++++++++++---- kernel/workqueue.c | 131 ++++++++++++++++++++++++++++++++++++++++++++-- lib/Kconfig.debug | 8 +++ 4 files changed, 166 insertions(+), 15 deletions(-) (limited to 'kernel') diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c index 565ebc65920e..ba43dfed353d 100644 --- a/arch/x86/kernel/smpboot.c +++ b/arch/x86/kernel/smpboot.c @@ -687,7 +687,7 @@ static int __cpuinit do_boot_cpu(int apicid, int cpu) .done = COMPLETION_INITIALIZER_ONSTACK(c_idle.done), }; - INIT_WORK(&c_idle.work, do_fork_idle); + INIT_WORK_ON_STACK(&c_idle.work, do_fork_idle); alternatives_smp_switch(1); @@ -713,6 +713,7 @@ static int __cpuinit do_boot_cpu(int apicid, int cpu) if (IS_ERR(c_idle.idle)) { printk("failed fork for CPU %d\n", cpu); + destroy_work_on_stack(&c_idle.work); return PTR_ERR(c_idle.idle); } @@ -831,6 +832,7 @@ do_rest: smpboot_restore_warm_reset_vector(); } + destroy_work_on_stack(&c_idle.work); return boot_error; } diff --git a/include/linux/workqueue.h b/include/linux/workqueue.h index cf24c20de9e4..9466e860d8c2 100644 --- a/include/linux/workqueue.h +++ b/include/linux/workqueue.h @@ -25,6 +25,7 @@ typedef void (*work_func_t)(struct work_struct *work); struct work_struct { atomic_long_t data; #define WORK_STRUCT_PENDING 0 /* T if work item pending execution */ +#define WORK_STRUCT_STATIC 1 /* static initializer (debugobjects) */ #define WORK_STRUCT_FLAG_MASK (3UL) #define WORK_STRUCT_WQ_DATA_MASK (~WORK_STRUCT_FLAG_MASK) struct list_head entry; @@ -35,6 +36,7 @@ struct work_struct { }; #define WORK_DATA_INIT() ATOMIC_LONG_INIT(0) +#define WORK_DATA_STATIC_INIT() ATOMIC_LONG_INIT(2) struct delayed_work { struct work_struct work; @@ -63,7 +65,7 @@ struct execute_work { #endif #define __WORK_INITIALIZER(n, f) { \ - .data = WORK_DATA_INIT(), \ + .data = WORK_DATA_STATIC_INIT(), \ .entry = { &(n).entry, &(n).entry }, \ .func = (f), \ __WORK_INIT_LOCKDEP_MAP(#n, &(n)) \ @@ -91,6 +93,14 @@ struct execute_work { #define PREPARE_DELAYED_WORK(_work, _func) \ PREPARE_WORK(&(_work)->work, (_func)) +#ifdef CONFIG_DEBUG_OBJECTS_WORK +extern void __init_work(struct work_struct *work, int onstack); +extern void destroy_work_on_stack(struct work_struct *work); +#else +static inline void __init_work(struct work_struct *work, int onstack) { } +static inline void destroy_work_on_stack(struct work_struct *work) { } +#endif + /* * initialize all of a work item in one go * @@ -99,24 +109,36 @@ struct execute_work { * to generate better code. */ #ifdef CONFIG_LOCKDEP -#define INIT_WORK(_work, _func) \ +#define __INIT_WORK(_work, _func, _onstack) \ do { \ static struct lock_class_key __key; \ \ + __init_work((_work), _onstack); \ (_work)->data = (atomic_long_t) WORK_DATA_INIT(); \ lockdep_init_map(&(_work)->lockdep_map, #_work, &__key, 0);\ INIT_LIST_HEAD(&(_work)->entry); \ PREPARE_WORK((_work), (_func)); \ } while (0) #else -#define INIT_WORK(_work, _func) \ +#define __INIT_WORK(_work, _func, _onstack) \ do { \ + __init_work((_work), _onstack); \ (_work)->data = (atomic_long_t) WORK_DATA_INIT(); \ INIT_LIST_HEAD(&(_work)->entry); \ PREPARE_WORK((_work), (_func)); \ } while (0) #endif +#define INIT_WORK(_work, _func) \ + do { \ + __INIT_WORK((_work), (_func), 0); \ + } while (0) + +#define INIT_WORK_ON_STACK(_work, _func) \ + do { \ + __INIT_WORK((_work), (_func), 1); \ + } while (0) + #define INIT_DELAYED_WORK(_work, _func) \ do { \ INIT_WORK(&(_work)->work, (_func)); \ @@ -125,22 +147,16 @@ struct execute_work { #define INIT_DELAYED_WORK_ON_STACK(_work, _func) \ do { \ - INIT_WORK(&(_work)->work, (_func)); \ + INIT_WORK_ON_STACK(&(_work)->work, (_func)); \ init_timer_on_stack(&(_work)->timer); \ } while (0) -#define INIT_DELAYED_WORK_DEFERRABLE(_work, _func) \ +#define INIT_DELAYED_WORK_DEFERRABLE(_work, _func) \ do { \ INIT_WORK(&(_work)->work, (_func)); \ init_timer_deferrable(&(_work)->timer); \ } while (0) -#define INIT_DELAYED_WORK_ON_STACK(_work, _func) \ - do { \ - INIT_WORK(&(_work)->work, (_func)); \ - init_timer_on_stack(&(_work)->timer); \ - } while (0) - /** * work_pending - Find out whether a work item is currently pending * @work: The work item in question diff --git a/kernel/workqueue.c b/kernel/workqueue.c index 12328147132c..ddad63fbb61b 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -68,6 +68,116 @@ struct workqueue_struct { #endif }; +#ifdef CONFIG_DEBUG_OBJECTS_WORK + +static struct debug_obj_descr work_debug_descr; + +/* + * fixup_init is called when: + * - an active object is initialized + */ +static int work_fixup_init(void *addr, enum debug_obj_state state) +{ + struct work_struct *work = addr; + + switch (state) { + case ODEBUG_STATE_ACTIVE: + cancel_work_sync(work); + debug_object_init(work, &work_debug_descr); + return 1; + default: + return 0; + } +} + +/* + * fixup_activate is called when: + * - an active object is activated + * - an unknown object is activated (might be a statically initialized object) + */ +static int work_fixup_activate(void *addr, enum debug_obj_state state) +{ + struct work_struct *work = addr; + + switch (state) { + + case ODEBUG_STATE_NOTAVAILABLE: + /* + * This is not really a fixup. The work struct was + * statically initialized. We just make sure that it + * is tracked in the object tracker. + */ + if (test_bit(WORK_STRUCT_STATIC, work_data_bits(work))) { + debug_object_init(work, &work_debug_descr); + debug_object_activate(work, &work_debug_descr); + return 0; + } + WARN_ON_ONCE(1); + return 0; + + case ODEBUG_STATE_ACTIVE: + WARN_ON(1); + + default: + return 0; + } +} + +/* + * fixup_free is called when: + * - an active object is freed + */ +static int work_fixup_free(void *addr, enum debug_obj_state state) +{ + struct work_struct *work = addr; + + switch (state) { + case ODEBUG_STATE_ACTIVE: + cancel_work_sync(work); + debug_object_free(work, &work_debug_descr); + return 1; + default: + return 0; + } +} + +static struct debug_obj_descr work_debug_descr = { + .name = "work_struct", + .fixup_init = work_fixup_init, + .fixup_activate = work_fixup_activate, + .fixup_free = work_fixup_free, +}; + +static inline void debug_work_activate(struct work_struct *work) +{ + debug_object_activate(work, &work_debug_descr); +} + +static inline void debug_work_deactivate(struct work_struct *work) +{ + debug_object_deactivate(work, &work_debug_descr); +} + +void __init_work(struct work_struct *work, int onstack) +{ + if (onstack) + debug_object_init_on_stack(work, &work_debug_descr); + else + debug_object_init(work, &work_debug_descr); +} +EXPORT_SYMBOL_GPL(__init_work); + +void destroy_work_on_stack(struct work_struct *work) +{ + debug_object_free(work, &work_debug_descr); +} +EXPORT_SYMBOL_GPL(destroy_work_on_stack); + +#else +static inline void debug_work_activate(struct work_struct *work) { } +static inline void debug_work_deactivate(struct work_struct *work) { } +#endif + /* Serializes the accesses to the list of workqueues. */ static DEFINE_SPINLOCK(workqueue_lock); static LIST_HEAD(workqueues); @@ -145,6 +255,7 @@ static void __queue_work(struct cpu_workqueue_struct *cwq, { unsigned long flags; + debug_work_activate(work); spin_lock_irqsave(&cwq->lock, flags); insert_work(cwq, work, &cwq->worklist); spin_unlock_irqrestore(&cwq->lock, flags); @@ -280,6 +391,7 @@ static void run_workqueue(struct cpu_workqueue_struct *cwq) struct lockdep_map lockdep_map = work->lockdep_map; #endif trace_workqueue_execution(cwq->thread, work); + debug_work_deactivate(work); cwq->current_work = work; list_del_init(cwq->worklist.next); spin_unlock_irq(&cwq->lock); @@ -350,11 +462,18 @@ static void wq_barrier_func(struct work_struct *work) static void insert_wq_barrier(struct cpu_workqueue_struct *cwq, struct wq_barrier *barr, struct list_head *head) { - INIT_WORK(&barr->work, wq_barrier_func); + /* + * debugobject calls are safe here even with cwq->lock locked + * as we know for sure that this will not trigger any of the + * checks and call back into the fixup functions where we + * might deadlock. + */ + INIT_WORK_ON_STACK(&barr->work, wq_barrier_func); __set_bit(WORK_STRUCT_PENDING, work_data_bits(&barr->work)); init_completion(&barr->done); + debug_work_activate(&barr->work); insert_work(cwq, &barr->work, head); } @@ -372,8 +491,10 @@ static int flush_cpu_workqueue(struct cpu_workqueue_struct *cwq) } spin_unlock_irq(&cwq->lock); - if (active) + if (active) { wait_for_completion(&barr.done); + destroy_work_on_stack(&barr.work); + } return active; } @@ -451,6 +572,7 @@ out: return 0; wait_for_completion(&barr.done); + destroy_work_on_stack(&barr.work); return 1; } EXPORT_SYMBOL_GPL(flush_work); @@ -485,6 +607,7 @@ static int try_to_grab_pending(struct work_struct *work) */ smp_rmb(); if (cwq == get_wq_data(work)) { + debug_work_deactivate(work); list_del_init(&work->entry); ret = 1; } @@ -507,8 +630,10 @@ static void wait_on_cpu_work(struct cpu_workqueue_struct *cwq, } spin_unlock_irq(&cwq->lock); - if (unlikely(running)) + if (unlikely(running)) { wait_for_completion(&barr.done); + destroy_work_on_stack(&barr.work); + } } static void wait_on_work(struct work_struct *work) diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index 234ceb10861f..c91f0519d493 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -298,6 +298,14 @@ config DEBUG_OBJECTS_TIMERS timer routines to track the life time of timer objects and validate the timer operations. +config DEBUG_OBJECTS_WORK + bool "Debug work objects" + depends on DEBUG_OBJECTS + help + If you say Y here, additional code will be inserted into the + work queue routines to track the life time of work objects and + validate the work operations. + config DEBUG_OBJECTS_ENABLE_DEFAULT int "debug_objects bootup default value (0-1)" range 0 1 -- cgit v1.2.3-71-gd317 From 456b565cc52fbcdaa2e19ffdf40d9dd3b726d603 Mon Sep 17 00:00:00 2001 From: Simon Kagstrom Date: Fri, 16 Oct 2009 14:09:18 +0200 Subject: core: Add kernel message dumper to call on oopses and panics The core functionality is implemented as per Linus suggestion from http://lists.infradead.org/pipermail/linux-mtd/2009-October/027620.html (with the kmsg_dump implementation by Linus). A struct kmsg_dumper has been added which contains a callback to dump the kernel log buffers on crashes. The kmsg_dump function gets called from oops_exit() and panic() and invokes this callbacks with the crash reason. [dwmw2: Fix log_end handling] Signed-off-by: Simon Kagstrom Reviewed-by: Anders Grafstrom Reviewed-by: Linus Torvalds Acked-by: Ingo Molnar Signed-off-by: Artem Bityutskiy Signed-off-by: David Woodhouse --- include/linux/kmsg_dump.h | 44 +++++++++++++++++ kernel/panic.c | 3 ++ kernel/printk.c | 119 ++++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 166 insertions(+) create mode 100644 include/linux/kmsg_dump.h (limited to 'kernel') diff --git a/include/linux/kmsg_dump.h b/include/linux/kmsg_dump.h new file mode 100644 index 000000000000..7f089ec5ef32 --- /dev/null +++ b/include/linux/kmsg_dump.h @@ -0,0 +1,44 @@ +/* + * linux/include/kmsg_dump.h + * + * Copyright (C) 2009 Net Insight AB + * + * Author: Simon Kagstrom + * + * 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. + */ +#ifndef _LINUX_KMSG_DUMP_H +#define _LINUX_KMSG_DUMP_H + +#include + +enum kmsg_dump_reason { + KMSG_DUMP_OOPS, + KMSG_DUMP_PANIC, +}; + +/** + * struct kmsg_dumper - kernel crash message dumper structure + * @dump: The callback which gets called on crashes. The buffer is passed + * as two sections, where s1 (length l1) contains the older + * messages and s2 (length l2) contains the newer. + * @list: Entry in the dumper list (private) + * @registered: Flag that specifies if this is already registered + */ +struct kmsg_dumper { + void (*dump)(struct kmsg_dumper *dumper, enum kmsg_dump_reason reason, + const char *s1, unsigned long l1, + const char *s2, unsigned long l2); + struct list_head list; + int registered; +}; + +void kmsg_dump(enum kmsg_dump_reason reason); + +int kmsg_dump_register(struct kmsg_dumper *dumper); + +int kmsg_dump_unregister(struct kmsg_dumper *dumper); + +#endif /* _LINUX_KMSG_DUMP_H */ diff --git a/kernel/panic.c b/kernel/panic.c index bcdef26e3332..8c43226a544d 100644 --- a/kernel/panic.c +++ b/kernel/panic.c @@ -10,6 +10,7 @@ */ #include #include +#include #include #include #include @@ -74,6 +75,7 @@ NORET_TYPE void panic(const char * fmt, ...) dump_stack(); #endif + kmsg_dump(KMSG_DUMP_PANIC); /* * If we have crashed and we have a crash kernel loaded let it handle * everything else. @@ -338,6 +340,7 @@ void oops_exit(void) { do_oops_enter_exit(); print_oops_end_marker(); + kmsg_dump(KMSG_DUMP_OOPS); } #ifdef WANT_WARN_ON_SLOWPATH diff --git a/kernel/printk.c b/kernel/printk.c index f38b07f78a4e..051d1f50648f 100644 --- a/kernel/printk.c +++ b/kernel/printk.c @@ -33,6 +33,7 @@ #include #include #include +#include #include @@ -1405,3 +1406,121 @@ bool printk_timed_ratelimit(unsigned long *caller_jiffies, } EXPORT_SYMBOL(printk_timed_ratelimit); #endif + +static DEFINE_SPINLOCK(dump_list_lock); +static LIST_HEAD(dump_list); + +/** + * kmsg_dump_register - register a kernel log dumper. + * @dump: pointer to the kmsg_dumper structure + * + * Adds a kernel log dumper to the system. The dump callback in the + * structure will be called when the kernel oopses or panics and must be + * set. Returns zero on success and %-EINVAL or %-EBUSY otherwise. + */ +int kmsg_dump_register(struct kmsg_dumper *dumper) +{ + unsigned long flags; + int err = -EBUSY; + + /* The dump callback needs to be set */ + if (!dumper->dump) + return -EINVAL; + + spin_lock_irqsave(&dump_list_lock, flags); + /* Don't allow registering multiple times */ + if (!dumper->registered) { + dumper->registered = 1; + list_add_tail(&dumper->list, &dump_list); + err = 0; + } + spin_unlock_irqrestore(&dump_list_lock, flags); + + return err; +} +EXPORT_SYMBOL_GPL(kmsg_dump_register); + +/** + * kmsg_dump_unregister - unregister a kmsg dumper. + * @dump: pointer to the kmsg_dumper structure + * + * Removes a dump device from the system. Returns zero on success and + * %-EINVAL otherwise. + */ +int kmsg_dump_unregister(struct kmsg_dumper *dumper) +{ + unsigned long flags; + int err = -EINVAL; + + spin_lock_irqsave(&dump_list_lock, flags); + if (dumper->registered) { + dumper->registered = 0; + list_del(&dumper->list); + err = 0; + } + spin_unlock_irqrestore(&dump_list_lock, flags); + + return err; +} +EXPORT_SYMBOL_GPL(kmsg_dump_unregister); + +static const char const *kmsg_reasons[] = { + [KMSG_DUMP_OOPS] = "oops", + [KMSG_DUMP_PANIC] = "panic", +}; + +static const char *kmsg_to_str(enum kmsg_dump_reason reason) +{ + if (reason >= ARRAY_SIZE(kmsg_reasons) || reason < 0) + return "unknown"; + + return kmsg_reasons[reason]; +} + +/** + * kmsg_dump - dump kernel log to kernel message dumpers. + * @reason: the reason (oops, panic etc) for dumping + * + * Iterate through each of the dump devices and call the oops/panic + * callbacks with the log buffer. + */ +void kmsg_dump(enum kmsg_dump_reason reason) +{ + unsigned long end; + unsigned chars; + struct kmsg_dumper *dumper; + const char *s1, *s2; + unsigned long l1, l2; + unsigned long flags; + + /* Theoretically, the log could move on after we do this, but + there's not a lot we can do about that. The new messages + will overwrite the start of what we dump. */ + spin_lock_irqsave(&logbuf_lock, flags); + end = log_end & LOG_BUF_MASK; + chars = logged_chars; + spin_unlock_irqrestore(&logbuf_lock, flags); + + if (logged_chars > end) { + s1 = log_buf + log_buf_len - logged_chars + end; + l1 = logged_chars - end; + + s2 = log_buf; + l2 = end; + } else { + s1 = ""; + l1 = 0; + + s2 = log_buf + end - logged_chars; + l2 = logged_chars; + } + + if (!spin_trylock_irqsave(&dump_list_lock, flags)) { + printk(KERN_ERR "dump_kmsg: dump list lock is held during %s, skipping dump\n", + kmsg_to_str(reason)); + return; + } + list_for_each_entry(dumper, &dump_list, list) + dumper->dump(dumper, reason, s1, l1, s2, l2); + spin_unlock_irqrestore(&dump_list_lock, flags); +} -- cgit v1.2.3-71-gd317 From 595dd3d8bf953254d8d2f30f99c54fe09c470040 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Tue, 1 Dec 2009 10:52:02 -0800 Subject: kmsg_dump: fix build for CONFIG_PRINTK=n kmsg_dump() fails to build when CONFIG_PRINTK=n; provide stubs for the kmsg_dump*() functions when CONFIG_PRINTK=n. kernel/printk.c: In function 'kmsg_dump': kernel/printk.c:1501: error: 'log_buf_len' undeclared (first use in this function) kernel/printk.c:1502: error: 'logged_chars' undeclared (first use in this function) kernel/printk.c:1506: error: 'log_buf' undeclared (first use in this function) Signed-off-by: Randy Dunlap Acked-by: Simon Kagstrom Signed-off-by: David Woodhouse --- include/linux/kmsg_dump.h | 16 ++++++++++++++++ kernel/printk.c | 2 +- 2 files changed, 17 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/include/linux/kmsg_dump.h b/include/linux/kmsg_dump.h index 7f089ec5ef32..e32aa268efac 100644 --- a/include/linux/kmsg_dump.h +++ b/include/linux/kmsg_dump.h @@ -35,10 +35,26 @@ struct kmsg_dumper { int registered; }; +#ifdef CONFIG_PRINTK void kmsg_dump(enum kmsg_dump_reason reason); int kmsg_dump_register(struct kmsg_dumper *dumper); int kmsg_dump_unregister(struct kmsg_dumper *dumper); +#else +static inline void kmsg_dump(enum kmsg_dump_reason reason) +{ +} + +static inline int kmsg_dump_register(struct kmsg_dumper *dumper) +{ + return -EINVAL; +} + +static inline int kmsg_dump_unregister(struct kmsg_dumper *dumper) +{ + return -EINVAL; +} +#endif #endif /* _LINUX_KMSG_DUMP_H */ diff --git a/kernel/printk.c b/kernel/printk.c index 051d1f50648f..2a564570f822 100644 --- a/kernel/printk.c +++ b/kernel/printk.c @@ -1405,7 +1405,6 @@ bool printk_timed_ratelimit(unsigned long *caller_jiffies, return false; } EXPORT_SYMBOL(printk_timed_ratelimit); -#endif static DEFINE_SPINLOCK(dump_list_lock); static LIST_HEAD(dump_list); @@ -1524,3 +1523,4 @@ void kmsg_dump(enum kmsg_dump_reason reason) dumper->dump(dumper, reason, s1, l1, s2, l2); spin_unlock_irqrestore(&dump_list_lock, flags); } +#endif -- cgit v1.2.3-71-gd317 From 2f0993e0fb663c49e4d1e02654f6203246be4817 Mon Sep 17 00:00:00 2001 From: Frederic Weisbecker Date: Sat, 5 Dec 2009 07:06:10 +0100 Subject: hw-breakpoints: Drop callback and task parameters from modify helper Drop the callback and task parameters from modify_user_hw_breakpoint(). For now we have no user that need to modify a breakpoint to the point of changing its handler or its task context. Signed-off-by: Frederic Weisbecker Cc: "K. Prasad" --- arch/x86/kernel/ptrace.c | 4 ++-- include/linux/hw_breakpoint.h | 9 ++------- kernel/hw_breakpoint.c | 7 +++---- 3 files changed, 7 insertions(+), 13 deletions(-) (limited to 'kernel') diff --git a/arch/x86/kernel/ptrace.c b/arch/x86/kernel/ptrace.c index 04d182a7cfdb..dbb395572ae2 100644 --- a/arch/x86/kernel/ptrace.c +++ b/arch/x86/kernel/ptrace.c @@ -618,7 +618,7 @@ ptrace_modify_breakpoint(struct perf_event *bp, int len, int type, attr.bp_type = gen_type; attr.disabled = disabled; - return modify_user_hw_breakpoint(bp, &attr, bp->callback, tsk); + return modify_user_hw_breakpoint(bp, &attr); } /* @@ -740,7 +740,7 @@ static int ptrace_set_breakpoint_addr(struct task_struct *tsk, int nr, attr = bp->attr; attr.bp_addr = addr; - bp = modify_user_hw_breakpoint(bp, &attr, bp->callback, tsk); + bp = modify_user_hw_breakpoint(bp, &attr); } /* * CHECKME: the previous code returned -EIO if the addr wasn't a diff --git a/include/linux/hw_breakpoint.h b/include/linux/hw_breakpoint.h index a03daed08c59..d33096e0dbd4 100644 --- a/include/linux/hw_breakpoint.h +++ b/include/linux/hw_breakpoint.h @@ -57,10 +57,7 @@ register_user_hw_breakpoint(struct perf_event_attr *attr, /* FIXME: only change from the attr, and don't unregister */ extern struct perf_event * -modify_user_hw_breakpoint(struct perf_event *bp, - struct perf_event_attr *attr, - perf_callback_t triggered, - struct task_struct *tsk); +modify_user_hw_breakpoint(struct perf_event *bp, struct perf_event_attr *attr); /* * Kernel breakpoints are not associated with any particular thread. @@ -97,9 +94,7 @@ register_user_hw_breakpoint(struct perf_event_attr *attr, struct task_struct *tsk) { return NULL; } static inline struct perf_event * modify_user_hw_breakpoint(struct perf_event *bp, - struct perf_event_attr *attr, - perf_callback_t triggered, - struct task_struct *tsk) { return NULL; } + struct perf_event_attr *attr) { return NULL; } static inline struct perf_event * register_wide_hw_breakpoint_cpu(struct perf_event_attr *attr, perf_callback_t triggered, diff --git a/kernel/hw_breakpoint.c b/kernel/hw_breakpoint.c index cf5ee1628411..2d10b012828f 100644 --- a/kernel/hw_breakpoint.c +++ b/kernel/hw_breakpoint.c @@ -312,9 +312,7 @@ EXPORT_SYMBOL_GPL(register_user_hw_breakpoint); * @tsk: pointer to 'task_struct' of the process to which the address belongs */ struct perf_event * -modify_user_hw_breakpoint(struct perf_event *bp, struct perf_event_attr *attr, - perf_callback_t triggered, - struct task_struct *tsk) +modify_user_hw_breakpoint(struct perf_event *bp, struct perf_event_attr *attr) { /* * FIXME: do it without unregistering @@ -323,7 +321,8 @@ modify_user_hw_breakpoint(struct perf_event *bp, struct perf_event_attr *attr, */ unregister_hw_breakpoint(bp); - return perf_event_create_kernel_counter(attr, -1, tsk->pid, triggered); + return perf_event_create_kernel_counter(attr, -1, bp->ctx->task->pid, + bp->callback); } EXPORT_SYMBOL_GPL(modify_user_hw_breakpoint); -- cgit v1.2.3-71-gd317 From b326e9560a28fc3e950637ef51847ed8f05c1335 Mon Sep 17 00:00:00 2001 From: Frederic Weisbecker Date: Sat, 5 Dec 2009 09:44:31 +0100 Subject: hw-breakpoints: Use overflow handler instead of the event callback struct perf_event::event callback was called when a breakpoint triggers. But this is a rather opaque callback, pretty tied-only to the breakpoint API and not really integrated into perf as it triggers even when we don't overflow. We prefer to use overflow_handler() as it fits into the perf events rules, being called only when we overflow. Reported-by: Peter Zijlstra Signed-off-by: Frederic Weisbecker Cc: Paul Mackerras Cc: Arnaldo Carvalho de Melo Cc: "K. Prasad" --- arch/x86/kernel/hw_breakpoint.c | 5 ++--- arch/x86/kernel/ptrace.c | 9 ++++++--- include/linux/hw_breakpoint.h | 25 +++++++++++-------------- include/linux/perf_event.h | 13 +++++++------ kernel/hw_breakpoint.c | 17 +++++------------ kernel/perf_event.c | 24 +++++++++--------------- kernel/trace/trace_ksym.c | 5 +++-- samples/hw_breakpoint/data_breakpoint.c | 7 +++++-- 8 files changed, 48 insertions(+), 57 deletions(-) (limited to 'kernel') diff --git a/arch/x86/kernel/hw_breakpoint.c b/arch/x86/kernel/hw_breakpoint.c index d42f65ac4927..05d5fec64a94 100644 --- a/arch/x86/kernel/hw_breakpoint.c +++ b/arch/x86/kernel/hw_breakpoint.c @@ -362,8 +362,7 @@ int arch_validate_hwbkpt_settings(struct perf_event *bp, return ret; } - if (bp->callback) - ret = arch_store_info(bp); + ret = arch_store_info(bp); if (ret < 0) return ret; @@ -519,7 +518,7 @@ static int __kprobes hw_breakpoint_handler(struct die_args *args) break; } - (bp->callback)(bp, args->regs); + perf_bp_event(bp, args->regs); rcu_read_unlock(); } diff --git a/arch/x86/kernel/ptrace.c b/arch/x86/kernel/ptrace.c index dbb395572ae2..b361d28061d0 100644 --- a/arch/x86/kernel/ptrace.c +++ b/arch/x86/kernel/ptrace.c @@ -555,7 +555,9 @@ static int genregs_set(struct task_struct *target, return ret; } -static void ptrace_triggered(struct perf_event *bp, void *data) +static void ptrace_triggered(struct perf_event *bp, int nmi, + struct perf_sample_data *data, + struct pt_regs *regs) { int i; struct thread_struct *thread = &(current->thread); @@ -599,7 +601,7 @@ ptrace_modify_breakpoint(struct perf_event *bp, int len, int type, { int err; int gen_len, gen_type; - DEFINE_BREAKPOINT_ATTR(attr); + struct perf_event_attr attr; /* * We shoud have at least an inactive breakpoint at this @@ -721,9 +723,10 @@ static int ptrace_set_breakpoint_addr(struct task_struct *tsk, int nr, { struct perf_event *bp; struct thread_struct *t = &tsk->thread; - DEFINE_BREAKPOINT_ATTR(attr); + struct perf_event_attr attr; if (!t->ptrace_bps[nr]) { + hw_breakpoint_init(&attr); /* * Put stub len and type to register (reserve) an inactive but * correct bp diff --git a/include/linux/hw_breakpoint.h b/include/linux/hw_breakpoint.h index d33096e0dbd4..4d14a384a01e 100644 --- a/include/linux/hw_breakpoint.h +++ b/include/linux/hw_breakpoint.h @@ -20,19 +20,16 @@ enum { #ifdef CONFIG_HAVE_HW_BREAKPOINT -/* As it's for in-kernel or ptrace use, we want it to be pinned */ -#define DEFINE_BREAKPOINT_ATTR(name) \ -struct perf_event_attr name = { \ - .type = PERF_TYPE_BREAKPOINT, \ - .size = sizeof(name), \ - .pinned = 1, \ -}; - static inline void hw_breakpoint_init(struct perf_event_attr *attr) { attr->type = PERF_TYPE_BREAKPOINT; attr->size = sizeof(*attr); + /* + * As it's for in-kernel or ptrace use, we want it to be pinned + * and to call its callback every hits. + */ attr->pinned = 1; + attr->sample_period = 1; } static inline unsigned long hw_breakpoint_addr(struct perf_event *bp) @@ -52,7 +49,7 @@ static inline int hw_breakpoint_len(struct perf_event *bp) extern struct perf_event * register_user_hw_breakpoint(struct perf_event_attr *attr, - perf_callback_t triggered, + perf_overflow_handler_t triggered, struct task_struct *tsk); /* FIXME: only change from the attr, and don't unregister */ @@ -64,12 +61,12 @@ modify_user_hw_breakpoint(struct perf_event *bp, struct perf_event_attr *attr); */ extern struct perf_event * register_wide_hw_breakpoint_cpu(struct perf_event_attr *attr, - perf_callback_t triggered, + perf_overflow_handler_t triggered, int cpu); extern struct perf_event ** register_wide_hw_breakpoint(struct perf_event_attr *attr, - perf_callback_t triggered); + perf_overflow_handler_t triggered); extern int register_perf_hw_breakpoint(struct perf_event *bp); extern int __register_perf_hw_breakpoint(struct perf_event *bp); @@ -90,18 +87,18 @@ static inline struct arch_hw_breakpoint *counter_arch_bp(struct perf_event *bp) static inline struct perf_event * register_user_hw_breakpoint(struct perf_event_attr *attr, - perf_callback_t triggered, + perf_overflow_handler_t triggered, struct task_struct *tsk) { return NULL; } static inline struct perf_event * modify_user_hw_breakpoint(struct perf_event *bp, struct perf_event_attr *attr) { return NULL; } static inline struct perf_event * register_wide_hw_breakpoint_cpu(struct perf_event_attr *attr, - perf_callback_t triggered, + perf_overflow_handler_t triggered, int cpu) { return NULL; } static inline struct perf_event ** register_wide_hw_breakpoint(struct perf_event_attr *attr, - perf_callback_t triggered) { return NULL; } + perf_overflow_handler_t triggered) { return NULL; } static inline int register_perf_hw_breakpoint(struct perf_event *bp) { return -ENOSYS; } static inline int diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h index 84bd28a0ffab..d2f2667430da 100644 --- a/include/linux/perf_event.h +++ b/include/linux/perf_event.h @@ -565,10 +565,13 @@ struct perf_pending_entry { void (*func)(struct perf_pending_entry *); }; -typedef void (*perf_callback_t)(struct perf_event *, void *); - struct perf_sample_data; +typedef void (*perf_callback_t)(struct perf_event *, void *); +typedef void (*perf_overflow_handler_t)(struct perf_event *, int, + struct perf_sample_data *, + struct pt_regs *regs); + /** * struct perf_event - performance event kernel representation: */ @@ -660,9 +663,7 @@ struct perf_event { struct pid_namespace *ns; u64 id; - void (*overflow_handler)(struct perf_event *event, - int nmi, struct perf_sample_data *data, - struct pt_regs *regs); + perf_overflow_handler_t overflow_handler; #ifdef CONFIG_EVENT_PROFILE struct event_filter *filter; @@ -779,7 +780,7 @@ extern struct perf_event * perf_event_create_kernel_counter(struct perf_event_attr *attr, int cpu, pid_t pid, - perf_callback_t callback); + perf_overflow_handler_t callback); extern u64 perf_event_read_value(struct perf_event *event, u64 *enabled, u64 *running); diff --git a/kernel/hw_breakpoint.c b/kernel/hw_breakpoint.c index 2d10b012828f..b600fc27f161 100644 --- a/kernel/hw_breakpoint.c +++ b/kernel/hw_breakpoint.c @@ -259,7 +259,7 @@ void release_bp_slot(struct perf_event *bp) } -int __register_perf_hw_breakpoint(struct perf_event *bp) +int register_perf_hw_breakpoint(struct perf_event *bp) { int ret; @@ -276,19 +276,12 @@ int __register_perf_hw_breakpoint(struct perf_event *bp) * This is a quick hack that will be removed soon, once we remove * the tmp breakpoints from ptrace */ - if (!bp->attr.disabled || bp->callback == perf_bp_event) + if (!bp->attr.disabled || !bp->overflow_handler) ret = arch_validate_hwbkpt_settings(bp, bp->ctx->task); return ret; } -int register_perf_hw_breakpoint(struct perf_event *bp) -{ - bp->callback = perf_bp_event; - - return __register_perf_hw_breakpoint(bp); -} - /** * register_user_hw_breakpoint - register a hardware breakpoint for user space * @attr: breakpoint attributes @@ -297,7 +290,7 @@ int register_perf_hw_breakpoint(struct perf_event *bp) */ struct perf_event * register_user_hw_breakpoint(struct perf_event_attr *attr, - perf_callback_t triggered, + perf_overflow_handler_t triggered, struct task_struct *tsk) { return perf_event_create_kernel_counter(attr, -1, tsk->pid, triggered); @@ -322,7 +315,7 @@ modify_user_hw_breakpoint(struct perf_event *bp, struct perf_event_attr *attr) unregister_hw_breakpoint(bp); return perf_event_create_kernel_counter(attr, -1, bp->ctx->task->pid, - bp->callback); + bp->overflow_handler); } EXPORT_SYMBOL_GPL(modify_user_hw_breakpoint); @@ -347,7 +340,7 @@ EXPORT_SYMBOL_GPL(unregister_hw_breakpoint); */ struct perf_event ** register_wide_hw_breakpoint(struct perf_event_attr *attr, - perf_callback_t triggered) + perf_overflow_handler_t triggered) { struct perf_event **cpu_events, **pevent, *bp; long err; diff --git a/kernel/perf_event.c b/kernel/perf_event.c index 6b7ddba1dd64..fd43ff4ac860 100644 --- a/kernel/perf_event.c +++ b/kernel/perf_event.c @@ -4286,15 +4286,8 @@ static void bp_perf_event_destroy(struct perf_event *event) static const struct pmu *bp_perf_event_init(struct perf_event *bp) { int err; - /* - * The breakpoint is already filled if we haven't created the counter - * through perf syscall - * FIXME: manage to get trigerred to NULL if it comes from syscalls - */ - if (!bp->callback) - err = register_perf_hw_breakpoint(bp); - else - err = __register_perf_hw_breakpoint(bp); + + err = register_perf_hw_breakpoint(bp); if (err) return ERR_PTR(err); @@ -4390,7 +4383,7 @@ perf_event_alloc(struct perf_event_attr *attr, struct perf_event_context *ctx, struct perf_event *group_leader, struct perf_event *parent_event, - perf_callback_t callback, + perf_overflow_handler_t overflow_handler, gfp_t gfpflags) { const struct pmu *pmu; @@ -4433,10 +4426,10 @@ perf_event_alloc(struct perf_event_attr *attr, event->state = PERF_EVENT_STATE_INACTIVE; - if (!callback && parent_event) - callback = parent_event->callback; + if (!overflow_handler && parent_event) + overflow_handler = parent_event->overflow_handler; - event->callback = callback; + event->overflow_handler = overflow_handler; if (attr->disabled) event->state = PERF_EVENT_STATE_OFF; @@ -4776,7 +4769,8 @@ err_put_context: */ struct perf_event * perf_event_create_kernel_counter(struct perf_event_attr *attr, int cpu, - pid_t pid, perf_callback_t callback) + pid_t pid, + perf_overflow_handler_t overflow_handler) { struct perf_event *event; struct perf_event_context *ctx; @@ -4793,7 +4787,7 @@ perf_event_create_kernel_counter(struct perf_event_attr *attr, int cpu, } event = perf_event_alloc(attr, cpu, ctx, NULL, - NULL, callback, GFP_KERNEL); + NULL, overflow_handler, GFP_KERNEL); if (IS_ERR(event)) { err = PTR_ERR(event); goto err_put_context; diff --git a/kernel/trace/trace_ksym.c b/kernel/trace/trace_ksym.c index ddfa0fd43bc0..acb87d4a4ac1 100644 --- a/kernel/trace/trace_ksym.c +++ b/kernel/trace/trace_ksym.c @@ -79,11 +79,12 @@ void ksym_collect_stats(unsigned long hbp_hit_addr) } #endif /* CONFIG_PROFILE_KSYM_TRACER */ -void ksym_hbp_handler(struct perf_event *hbp, void *data) +void ksym_hbp_handler(struct perf_event *hbp, int nmi, + struct perf_sample_data *data, + struct pt_regs *regs) { struct ring_buffer_event *event; struct ksym_trace_entry *entry; - struct pt_regs *regs = data; struct ring_buffer *buffer; int pc; diff --git a/samples/hw_breakpoint/data_breakpoint.c b/samples/hw_breakpoint/data_breakpoint.c index 29525500df00..c69cbe9b2426 100644 --- a/samples/hw_breakpoint/data_breakpoint.c +++ b/samples/hw_breakpoint/data_breakpoint.c @@ -41,7 +41,9 @@ module_param_string(ksym, ksym_name, KSYM_NAME_LEN, S_IRUGO); MODULE_PARM_DESC(ksym, "Kernel symbol to monitor; this module will report any" " write operations on the kernel symbol"); -static void sample_hbp_handler(struct perf_event *temp, void *data) +static void sample_hbp_handler(struct perf_event *bp, int nmi, + struct perf_sample_data *data, + struct pt_regs *regs) { printk(KERN_INFO "%s value is changed\n", ksym_name); dump_stack(); @@ -51,8 +53,9 @@ static void sample_hbp_handler(struct perf_event *temp, void *data) static int __init hw_break_module_init(void) { int ret; - DEFINE_BREAKPOINT_ATTR(attr); + struct perf_event_attr attr; + hw_breakpoint_init(&attr); attr.bp_addr = kallsyms_lookup_name(ksym_name); attr.bp_len = HW_BREAKPOINT_LEN_4; attr.bp_type = HW_BREAKPOINT_W | HW_BREAKPOINT_R; -- cgit v1.2.3-71-gd317 From 109d71c6dd52ec08878c2c67eb4c0bd67fcbc80b Mon Sep 17 00:00:00 2001 From: Frank Rowand Date: Thu, 19 Nov 2009 13:42:06 -0800 Subject: lockstat: Fix min, max times in /proc/lock_stats Fix min, max times in /proc/lock_stats (1) When collecting lock hold and wait times, if the current minimum time is zero, it will be replaced by the next time. (2) When aggregating minimum and maximum lock hold and wait times accross cpus, the values are added, instead of selecting the minimum and maximum. Signed-off-by: Frank Rowand Signed-off-by: Peter Zijlstra LKML-Reference: <4B05BBAE.2050005@am.sony.com> Signed-off-by: Ingo Molnar --- kernel/lockdep.c | 13 ++++++++++--- 1 file changed, 10 insertions(+), 3 deletions(-) (limited to 'kernel') diff --git a/kernel/lockdep.c b/kernel/lockdep.c index f5dcd36d3151..7a3ae56b3a7f 100644 --- a/kernel/lockdep.c +++ b/kernel/lockdep.c @@ -168,7 +168,7 @@ static void lock_time_inc(struct lock_time *lt, u64 time) if (time > lt->max) lt->max = time; - if (time < lt->min || !lt->min) + if (time < lt->min || !lt->nr) lt->min = time; lt->total += time; @@ -177,8 +177,15 @@ static void lock_time_inc(struct lock_time *lt, u64 time) static inline void lock_time_add(struct lock_time *src, struct lock_time *dst) { - dst->min += src->min; - dst->max += src->max; + if (!src->nr) + return; + + if (src->max > dst->max) + dst->max = src->max; + + if (src->min < dst->min || !dst->nr) + dst->min = src->min; + dst->total += src->total; dst->nr += src->nr; } -- cgit v1.2.3-71-gd317 From e1b8090bdf125f8b2e192149547fead7f302a89c Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Sun, 6 Dec 2009 20:41:16 +0100 Subject: cpumask: Fix generate_sched_domains() for UP Commit acc3f5d7cabbfd6cec71f0c1f9900621fa2d6ae7 ("cpumask: Partition_sched_domains takes array of cpumask_var_t") changed the function signature of generate_sched_domains() for the CONFIG_SMP=y case, but forgot to update the corresponding function for the CONFIG_SMP=n case, causing: kernel/cpuset.c:2073: warning: passing argument 1 of 'generate_sched_domains' from incompatible pointer type Signed-off-by: Geert Uytterhoeven Cc: Rusty Russell Cc: Linus Torvalds LKML-Reference: Signed-off-by: Ingo Molnar --- kernel/cpuset.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/cpuset.c b/kernel/cpuset.c index 3cf2183b472d..43fb7e800028 100644 --- a/kernel/cpuset.c +++ b/kernel/cpuset.c @@ -737,7 +737,7 @@ static void do_rebuild_sched_domains(struct work_struct *unused) { } -static int generate_sched_domains(struct cpumask **domains, +static int generate_sched_domains(cpumask_var_t **domains, struct sched_domain_attr **attributes) { *domains = NULL; -- cgit v1.2.3-71-gd317 From 6ad4c18884e864cf4c77f9074d3d1816063f99cd Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 25 Nov 2009 13:31:39 +0100 Subject: sched: Fix balance vs hotplug race Since (e761b77: cpu hotplug, sched: Introduce cpu_active_map and redo sched domain managment) we have cpu_active_mask which is suppose to rule scheduler migration and load-balancing, except it never (fully) did. The particular problem being solved here is a crash in try_to_wake_up() where select_task_rq() ends up selecting an offline cpu because select_task_rq_fair() trusts the sched_domain tree to reflect the current state of affairs, similarly select_task_rq_rt() trusts the root_domain. However, the sched_domains are updated from CPU_DEAD, which is after the cpu is taken offline and after stop_machine is done. Therefore it can race perfectly well with code assuming the domains are right. Cure this by building the domains from cpu_active_mask on CPU_DOWN_PREPARE. Signed-off-by: Peter Zijlstra LKML-Reference: Signed-off-by: Ingo Molnar --- include/linux/cpumask.h | 2 ++ kernel/cpu.c | 18 +++++++++++++----- kernel/cpuset.c | 16 +++++++++------- kernel/sched.c | 32 +++++++++++++++++--------------- 4 files changed, 41 insertions(+), 27 deletions(-) (limited to 'kernel') diff --git a/include/linux/cpumask.h b/include/linux/cpumask.h index 789cf5f920ce..d77b54733c5b 100644 --- a/include/linux/cpumask.h +++ b/include/linux/cpumask.h @@ -84,6 +84,7 @@ extern const struct cpumask *const cpu_active_mask; #define num_online_cpus() cpumask_weight(cpu_online_mask) #define num_possible_cpus() cpumask_weight(cpu_possible_mask) #define num_present_cpus() cpumask_weight(cpu_present_mask) +#define num_active_cpus() cpumask_weight(cpu_active_mask) #define cpu_online(cpu) cpumask_test_cpu((cpu), cpu_online_mask) #define cpu_possible(cpu) cpumask_test_cpu((cpu), cpu_possible_mask) #define cpu_present(cpu) cpumask_test_cpu((cpu), cpu_present_mask) @@ -92,6 +93,7 @@ extern const struct cpumask *const cpu_active_mask; #define num_online_cpus() 1 #define num_possible_cpus() 1 #define num_present_cpus() 1 +#define num_active_cpus() 1 #define cpu_online(cpu) ((cpu) == 0) #define cpu_possible(cpu) ((cpu) == 0) #define cpu_present(cpu) ((cpu) == 0) diff --git a/kernel/cpu.c b/kernel/cpu.c index 6ba0f1ecb212..b21688640377 100644 --- a/kernel/cpu.c +++ b/kernel/cpu.c @@ -212,6 +212,8 @@ static int __ref _cpu_down(unsigned int cpu, int tasks_frozen) err = __raw_notifier_call_chain(&cpu_chain, CPU_DOWN_PREPARE | mod, hcpu, -1, &nr_calls); if (err == NOTIFY_BAD) { + set_cpu_active(cpu, true); + nr_calls--; __raw_notifier_call_chain(&cpu_chain, CPU_DOWN_FAILED | mod, hcpu, nr_calls, NULL); @@ -223,11 +225,11 @@ static int __ref _cpu_down(unsigned int cpu, int tasks_frozen) /* Ensure that we are not runnable on dying cpu */ cpumask_copy(old_allowed, ¤t->cpus_allowed); - set_cpus_allowed_ptr(current, - cpumask_of(cpumask_any_but(cpu_online_mask, cpu))); + set_cpus_allowed_ptr(current, cpu_active_mask); err = __stop_machine(take_cpu_down, &tcd_param, cpumask_of(cpu)); if (err) { + set_cpu_active(cpu, true); /* CPU didn't die: tell everyone. Can't complain. */ if (raw_notifier_call_chain(&cpu_chain, CPU_DOWN_FAILED | mod, hcpu) == NOTIFY_BAD) @@ -292,9 +294,6 @@ int __ref cpu_down(unsigned int cpu) err = _cpu_down(cpu, 0); - if (cpu_online(cpu)) - set_cpu_active(cpu, true); - out: cpu_maps_update_done(); stop_machine_destroy(); @@ -387,6 +386,15 @@ int disable_nonboot_cpus(void) * with the userspace trying to use the CPU hotplug at the same time */ cpumask_clear(frozen_cpus); + + for_each_online_cpu(cpu) { + if (cpu == first_cpu) + continue; + set_cpu_active(cpu, false); + } + + synchronize_sched(); + printk("Disabling non-boot CPUs ...\n"); for_each_online_cpu(cpu) { if (cpu == first_cpu) diff --git a/kernel/cpuset.c b/kernel/cpuset.c index 43fb7e800028..ba401fab459f 100644 --- a/kernel/cpuset.c +++ b/kernel/cpuset.c @@ -872,7 +872,7 @@ static int update_cpumask(struct cpuset *cs, struct cpuset *trialcs, if (retval < 0) return retval; - if (!cpumask_subset(trialcs->cpus_allowed, cpu_online_mask)) + if (!cpumask_subset(trialcs->cpus_allowed, cpu_active_mask)) return -EINVAL; } retval = validate_change(cs, trialcs); @@ -2010,7 +2010,7 @@ static void scan_for_empty_cpusets(struct cpuset *root) } /* Continue past cpusets with all cpus, mems online */ - if (cpumask_subset(cp->cpus_allowed, cpu_online_mask) && + if (cpumask_subset(cp->cpus_allowed, cpu_active_mask) && nodes_subset(cp->mems_allowed, node_states[N_HIGH_MEMORY])) continue; @@ -2019,7 +2019,7 @@ static void scan_for_empty_cpusets(struct cpuset *root) /* Remove offline cpus and mems from this cpuset. */ mutex_lock(&callback_mutex); cpumask_and(cp->cpus_allowed, cp->cpus_allowed, - cpu_online_mask); + cpu_active_mask); nodes_and(cp->mems_allowed, cp->mems_allowed, node_states[N_HIGH_MEMORY]); mutex_unlock(&callback_mutex); @@ -2057,8 +2057,10 @@ static int cpuset_track_online_cpus(struct notifier_block *unused_nb, switch (phase) { case CPU_ONLINE: case CPU_ONLINE_FROZEN: - case CPU_DEAD: - case CPU_DEAD_FROZEN: + case CPU_DOWN_PREPARE: + case CPU_DOWN_PREPARE_FROZEN: + case CPU_DOWN_FAILED: + case CPU_DOWN_FAILED_FROZEN: break; default: @@ -2067,7 +2069,7 @@ static int cpuset_track_online_cpus(struct notifier_block *unused_nb, cgroup_lock(); mutex_lock(&callback_mutex); - cpumask_copy(top_cpuset.cpus_allowed, cpu_online_mask); + cpumask_copy(top_cpuset.cpus_allowed, cpu_active_mask); mutex_unlock(&callback_mutex); scan_for_empty_cpusets(&top_cpuset); ndoms = generate_sched_domains(&doms, &attr); @@ -2114,7 +2116,7 @@ static int cpuset_track_online_nodes(struct notifier_block *self, void __init cpuset_init_smp(void) { - cpumask_copy(top_cpuset.cpus_allowed, cpu_online_mask); + cpumask_copy(top_cpuset.cpus_allowed, cpu_active_mask); top_cpuset.mems_allowed = node_states[N_HIGH_MEMORY]; hotcpu_notifier(cpuset_track_online_cpus, 0); diff --git a/kernel/sched.c b/kernel/sched.c index aa31244caa9f..281da29d0801 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -4134,7 +4134,7 @@ static int load_balance(int this_cpu, struct rq *this_rq, unsigned long flags; struct cpumask *cpus = __get_cpu_var(load_balance_tmpmask); - cpumask_copy(cpus, cpu_online_mask); + cpumask_copy(cpus, cpu_active_mask); /* * When power savings policy is enabled for the parent domain, idle @@ -4297,7 +4297,7 @@ load_balance_newidle(int this_cpu, struct rq *this_rq, struct sched_domain *sd) int all_pinned = 0; struct cpumask *cpus = __get_cpu_var(load_balance_tmpmask); - cpumask_copy(cpus, cpu_online_mask); + cpumask_copy(cpus, cpu_active_mask); /* * When power savings policy is enabled for the parent domain, idle @@ -4694,7 +4694,7 @@ int select_nohz_load_balancer(int stop_tick) cpumask_set_cpu(cpu, nohz.cpu_mask); /* time for ilb owner also to sleep */ - if (cpumask_weight(nohz.cpu_mask) == num_online_cpus()) { + if (cpumask_weight(nohz.cpu_mask) == num_active_cpus()) { if (atomic_read(&nohz.load_balancer) == cpu) atomic_set(&nohz.load_balancer, -1); return 0; @@ -7093,7 +7093,7 @@ int set_cpus_allowed_ptr(struct task_struct *p, const struct cpumask *new_mask) int ret = 0; rq = task_rq_lock(p, &flags); - if (!cpumask_intersects(new_mask, cpu_online_mask)) { + if (!cpumask_intersects(new_mask, cpu_active_mask)) { ret = -EINVAL; goto out; } @@ -7115,7 +7115,7 @@ int set_cpus_allowed_ptr(struct task_struct *p, const struct cpumask *new_mask) if (cpumask_test_cpu(task_cpu(p), new_mask)) goto out; - if (migrate_task(p, cpumask_any_and(cpu_online_mask, new_mask), &req)) { + if (migrate_task(p, cpumask_any_and(cpu_active_mask, new_mask), &req)) { /* Need help from migration thread: drop lock and wait. */ struct task_struct *mt = rq->migration_thread; @@ -7269,19 +7269,19 @@ static void move_task_off_dead_cpu(int dead_cpu, struct task_struct *p) again: /* Look for allowed, online CPU in same node. */ - for_each_cpu_and(dest_cpu, nodemask, cpu_online_mask) + for_each_cpu_and(dest_cpu, nodemask, cpu_active_mask) if (cpumask_test_cpu(dest_cpu, &p->cpus_allowed)) goto move; /* Any allowed, online CPU? */ - dest_cpu = cpumask_any_and(&p->cpus_allowed, cpu_online_mask); + dest_cpu = cpumask_any_and(&p->cpus_allowed, cpu_active_mask); if (dest_cpu < nr_cpu_ids) goto move; /* No more Mr. Nice Guy. */ if (dest_cpu >= nr_cpu_ids) { cpuset_cpus_allowed_locked(p, &p->cpus_allowed); - dest_cpu = cpumask_any_and(cpu_online_mask, &p->cpus_allowed); + dest_cpu = cpumask_any_and(cpu_active_mask, &p->cpus_allowed); /* * Don't tell them about moving exiting tasks or @@ -7310,7 +7310,7 @@ move: */ static void migrate_nr_uninterruptible(struct rq *rq_src) { - struct rq *rq_dest = cpu_rq(cpumask_any(cpu_online_mask)); + struct rq *rq_dest = cpu_rq(cpumask_any(cpu_active_mask)); unsigned long flags; local_irq_save(flags); @@ -7564,7 +7564,7 @@ static ctl_table *sd_alloc_ctl_cpu_table(int cpu) static struct ctl_table_header *sd_sysctl_header; static void register_sched_domain_sysctl(void) { - int i, cpu_num = num_online_cpus(); + int i, cpu_num = num_possible_cpus(); struct ctl_table *entry = sd_alloc_ctl_entry(cpu_num + 1); char buf[32]; @@ -7574,7 +7574,7 @@ static void register_sched_domain_sysctl(void) if (entry == NULL) return; - for_each_online_cpu(i) { + for_each_possible_cpu(i) { snprintf(buf, 32, "cpu%d", i); entry->procname = kstrdup(buf, GFP_KERNEL); entry->mode = 0555; @@ -9100,7 +9100,7 @@ match1: if (doms_new == NULL) { ndoms_cur = 0; doms_new = &fallback_doms; - cpumask_andnot(doms_new[0], cpu_online_mask, cpu_isolated_map); + cpumask_andnot(doms_new[0], cpu_active_mask, cpu_isolated_map); WARN_ON_ONCE(dattr_new); } @@ -9231,8 +9231,10 @@ static int update_sched_domains(struct notifier_block *nfb, switch (action) { case CPU_ONLINE: case CPU_ONLINE_FROZEN: - case CPU_DEAD: - case CPU_DEAD_FROZEN: + case CPU_DOWN_PREPARE: + case CPU_DOWN_PREPARE_FROZEN: + case CPU_DOWN_FAILED: + case CPU_DOWN_FAILED_FROZEN: partition_sched_domains(1, NULL, NULL); return NOTIFY_OK; @@ -9279,7 +9281,7 @@ void __init sched_init_smp(void) #endif get_online_cpus(); mutex_lock(&sched_domains_mutex); - arch_init_sched_domains(cpu_online_mask); + arch_init_sched_domains(cpu_active_mask); cpumask_andnot(non_isolated_cpus, cpu_possible_mask, cpu_isolated_map); if (cpumask_empty(non_isolated_cpus)) cpumask_set_cpu(smp_processor_id(), non_isolated_cpus); -- cgit v1.2.3-71-gd317 From 56053170ea2a2c0dc17420e9b94aa3ca51d80408 Mon Sep 17 00:00:00 2001 From: Frederic Weisbecker Date: Mon, 7 Dec 2009 06:46:48 +0100 Subject: hw-breakpoints: Fix task-bound breakpoint slot allocation Whatever the context nature of a breakpoint, we always perform the following constraint checks before allocating it a slot: - Check the number of pinned breakpoint bound the concerned cpus - Check the max number of task-bound breakpoints that are belonging to a task. - Add both and see if we have a reamining slot for the new breakpoint This is the right thing to do when we are about to register a cpu-only bound breakpoint. But not if we are dealing with a task bound breakpoint. What we want in this case is: - Check the number of pinned breakpoint bound the concerned cpus - Check the number of breakpoints that already belong to the task in which the breakpoint to register is bound to. - Add both This fixes a regression that makes the "firefox -g" command fail to register breakpoints once we deal with a secondary thread. Reported-by: Walt Signed-off-by: Frederic Weisbecker Cc: Prasad --- kernel/hw_breakpoint.c | 74 ++++++++++++++++++++++++++++++-------------------- 1 file changed, 45 insertions(+), 29 deletions(-) (limited to 'kernel') diff --git a/kernel/hw_breakpoint.c b/kernel/hw_breakpoint.c index b600fc27f161..02b492504a5a 100644 --- a/kernel/hw_breakpoint.c +++ b/kernel/hw_breakpoint.c @@ -83,15 +83,51 @@ static unsigned int max_task_bp_pinned(int cpu) return 0; } +static int task_bp_pinned(struct task_struct *tsk) +{ + struct perf_event_context *ctx = tsk->perf_event_ctxp; + struct list_head *list; + struct perf_event *bp; + unsigned long flags; + int count = 0; + + if (WARN_ONCE(!ctx, "No perf context for this task")) + return 0; + + list = &ctx->event_list; + + spin_lock_irqsave(&ctx->lock, flags); + + /* + * The current breakpoint counter is not included in the list + * at the open() callback time + */ + list_for_each_entry(bp, list, event_entry) { + if (bp->attr.type == PERF_TYPE_BREAKPOINT) + count++; + } + + spin_unlock_irqrestore(&ctx->lock, flags); + + return count; +} + /* * Report the number of pinned/un-pinned breakpoints we have in * a given cpu (cpu > -1) or in all of them (cpu = -1). */ -static void fetch_bp_busy_slots(struct bp_busy_slots *slots, int cpu) +static void +fetch_bp_busy_slots(struct bp_busy_slots *slots, struct perf_event *bp) { + int cpu = bp->cpu; + struct task_struct *tsk = bp->ctx->task; + if (cpu >= 0) { slots->pinned = per_cpu(nr_cpu_bp_pinned, cpu); - slots->pinned += max_task_bp_pinned(cpu); + if (!tsk) + slots->pinned += max_task_bp_pinned(cpu); + else + slots->pinned += task_bp_pinned(tsk); slots->flexible = per_cpu(nr_bp_flexible, cpu); return; @@ -101,7 +137,10 @@ static void fetch_bp_busy_slots(struct bp_busy_slots *slots, int cpu) unsigned int nr; nr = per_cpu(nr_cpu_bp_pinned, cpu); - nr += max_task_bp_pinned(cpu); + if (!tsk) + nr += max_task_bp_pinned(cpu); + else + nr += task_bp_pinned(tsk); if (nr > slots->pinned) slots->pinned = nr; @@ -118,33 +157,10 @@ static void fetch_bp_busy_slots(struct bp_busy_slots *slots, int cpu) */ static void toggle_bp_task_slot(struct task_struct *tsk, int cpu, bool enable) { - int count = 0; - struct perf_event *bp; - struct perf_event_context *ctx = tsk->perf_event_ctxp; unsigned int *tsk_pinned; - struct list_head *list; - unsigned long flags; - - if (WARN_ONCE(!ctx, "No perf context for this task")) - return; - - list = &ctx->event_list; - - spin_lock_irqsave(&ctx->lock, flags); - - /* - * The current breakpoint counter is not included in the list - * at the open() callback time - */ - list_for_each_entry(bp, list, event_entry) { - if (bp->attr.type == PERF_TYPE_BREAKPOINT) - count++; - } - - spin_unlock_irqrestore(&ctx->lock, flags); + int count = 0; - if (WARN_ONCE(count < 0, "No breakpoint counter found in the counter list")) - return; + count = task_bp_pinned(tsk); tsk_pinned = per_cpu(task_bp_pinned, cpu); if (enable) { @@ -233,7 +249,7 @@ int reserve_bp_slot(struct perf_event *bp) mutex_lock(&nr_bp_mutex); - fetch_bp_busy_slots(&slots, bp->cpu); + fetch_bp_busy_slots(&slots, bp); /* Flexible counters need to keep at least one slot */ if (slots.pinned + (!!slots.flexible) == HBP_NUM) { -- cgit v1.2.3-71-gd317 From c521efd1700a8c0f7ce26f011f5eaecca17fabfa Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Mon, 7 Dec 2009 09:06:24 -0500 Subject: tracing: Add pipe_close interface An ftrace plugin can add a pipe_open interface when the user opens trace_pipe. But if the plugin allocates something within the pipe_open it can not free it because there exists no pipe_close. The hook to the trace file open has a corresponding close. The closing of the trace_pipe file should also have a corresponding close. Signed-off-by: Steven Rostedt --- kernel/trace/trace.c | 4 ++++ kernel/trace/trace.h | 2 ++ 2 files changed, 6 insertions(+) (limited to 'kernel') diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 874f2893cff0..f804b407d438 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -2898,6 +2898,10 @@ static int tracing_release_pipe(struct inode *inode, struct file *file) else cpumask_clear_cpu(iter->cpu_file, tracing_reader_cpumask); + + if (iter->trace->pipe_open) + iter->trace->pipe_close(iter); + mutex_unlock(&trace_types_lock); free_cpumask_var(iter->started); diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h index 1d7f4830a80d..7fa33cab6962 100644 --- a/kernel/trace/trace.h +++ b/kernel/trace/trace.h @@ -272,6 +272,7 @@ struct tracer_flags { * @pipe_open: called when the trace_pipe file is opened * @wait_pipe: override how the user waits for traces on trace_pipe * @close: called when the trace file is released + * @pipe_close: called when the trace_pipe file is released * @read: override the default read callback on trace_pipe * @splice_read: override the default splice_read callback on trace_pipe * @selftest: selftest to run on boot (see trace_selftest.c) @@ -290,6 +291,7 @@ struct tracer { void (*pipe_open)(struct trace_iterator *iter); void (*wait_pipe)(struct trace_iterator *iter); void (*close)(struct trace_iterator *iter); + void (*pipe_close)(struct trace_iterator *iter); ssize_t (*read)(struct trace_iterator *iter, struct file *filp, char __user *ubuf, size_t cnt, loff_t *ppos); -- cgit v1.2.3-71-gd317 From 6ab8886326a1b9a3a8d164d8174e3c20703a03a2 Mon Sep 17 00:00:00 2001 From: Stephen Rothwell Date: Tue, 8 Dec 2009 18:25:15 +1100 Subject: perf: hw_breakpoints: Fix percpu namespace clash Today's linux-next build failed with: kernel/hw_breakpoint.c:86: error: 'task_bp_pinned' redeclared as different kind of symbol ... Caused by commit dd17c8f72993f9461e9c19250e3f155d6d99df22 ("percpu: remove per_cpu__ prefix") from the percpu tree interacting with commit 56053170ea2a2c0dc17420e9b94aa3ca51d80408 ("hw-breakpoints: Fix task-bound breakpoint slot allocation") from the tip tree. Signed-off-by: Stephen Rothwell Acked-by: Frederic Weisbecker Cc: Peter Zijlstra Cc: Tejun Heo Cc: Rusty Russell Cc: Christoph Lameter Cc: Andrew Morton Cc: Linus Torvalds LKML-Reference: <20091208182515.bb6dda4a.sfr@canb.auug.org.au> Signed-off-by: Ingo Molnar --- kernel/hw_breakpoint.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) (limited to 'kernel') diff --git a/kernel/hw_breakpoint.c b/kernel/hw_breakpoint.c index 02b492504a5a..03a0773ac2b2 100644 --- a/kernel/hw_breakpoint.c +++ b/kernel/hw_breakpoint.c @@ -52,7 +52,7 @@ static DEFINE_PER_CPU(unsigned int, nr_cpu_bp_pinned); /* Number of pinned task breakpoints in a cpu */ -static DEFINE_PER_CPU(unsigned int, task_bp_pinned[HBP_NUM]); +static DEFINE_PER_CPU(unsigned int, nr_task_bp_pinned[HBP_NUM]); /* Number of non-pinned cpu/task breakpoints in a cpu */ static DEFINE_PER_CPU(unsigned int, nr_bp_flexible); @@ -73,7 +73,7 @@ static DEFINE_MUTEX(nr_bp_mutex); static unsigned int max_task_bp_pinned(int cpu) { int i; - unsigned int *tsk_pinned = per_cpu(task_bp_pinned, cpu); + unsigned int *tsk_pinned = per_cpu(nr_task_bp_pinned, cpu); for (i = HBP_NUM -1; i >= 0; i--) { if (tsk_pinned[i] > 0) @@ -162,7 +162,7 @@ static void toggle_bp_task_slot(struct task_struct *tsk, int cpu, bool enable) count = task_bp_pinned(tsk); - tsk_pinned = per_cpu(task_bp_pinned, cpu); + tsk_pinned = per_cpu(nr_task_bp_pinned, cpu); if (enable) { tsk_pinned[count]++; if (count > 0) @@ -209,7 +209,7 @@ static void toggle_bp_slot(struct perf_event *bp, bool enable) * - If attached to a single cpu, check: * * (per_cpu(nr_bp_flexible, cpu) || (per_cpu(nr_cpu_bp_pinned, cpu) - * + max(per_cpu(task_bp_pinned, cpu)))) < HBP_NUM + * + max(per_cpu(nr_task_bp_pinned, cpu)))) < HBP_NUM * * -> If there are already non-pinned counters in this cpu, it means * there is already a free slot for them. @@ -220,7 +220,7 @@ static void toggle_bp_slot(struct perf_event *bp, bool enable) * - If attached to every cpus, check: * * (per_cpu(nr_bp_flexible, *) || (max(per_cpu(nr_cpu_bp_pinned, *)) - * + max(per_cpu(task_bp_pinned, *)))) < HBP_NUM + * + max(per_cpu(nr_task_bp_pinned, *)))) < HBP_NUM * * -> This is roughly the same, except we check the number of per cpu * bp for every cpu and we keep the max one. Same for the per tasks @@ -232,7 +232,7 @@ static void toggle_bp_slot(struct perf_event *bp, bool enable) * - If attached to a single cpu, check: * * ((per_cpu(nr_bp_flexible, cpu) > 1) + per_cpu(nr_cpu_bp_pinned, cpu) - * + max(per_cpu(task_bp_pinned, cpu))) < HBP_NUM + * + max(per_cpu(nr_task_bp_pinned, cpu))) < HBP_NUM * * -> Same checks as before. But now the nr_bp_flexible, if any, must keep * one register at least (or they will never be fed). @@ -240,7 +240,7 @@ static void toggle_bp_slot(struct perf_event *bp, bool enable) * - If attached to every cpus, check: * * ((per_cpu(nr_bp_flexible, *) > 1) + max(per_cpu(nr_cpu_bp_pinned, *)) - * + max(per_cpu(task_bp_pinned, *))) < HBP_NUM + * + max(per_cpu(nr_task_bp_pinned, *))) < HBP_NUM */ int reserve_bp_slot(struct perf_event *bp) { -- cgit v1.2.3-71-gd317 From 722d0172377a5697919b9f7e5beb95165b1dec4e Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Tue, 8 Dec 2009 13:19:42 +0100 Subject: futex: Take mmap_sem for get_user_pages in fault_in_user_writeable get_user_pages() must be called with mmap_sem held. Signed-off-by: Andi Kleen Cc: stable@kernel.org Cc: Andrew Morton Cc: Nick Piggin Cc: Darren Hart Cc: Peter Zijlstra LKML-Reference: <20091208121942.GA21298@basil.fritz.box> Signed-off-by: Thomas Gleixner --- kernel/futex.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/futex.c b/kernel/futex.c index fb65e822fc41..d73ef1f3e55d 100644 --- a/kernel/futex.c +++ b/kernel/futex.c @@ -304,8 +304,14 @@ void put_futex_key(int fshared, union futex_key *key) */ static int fault_in_user_writeable(u32 __user *uaddr) { - int ret = get_user_pages(current, current->mm, (unsigned long)uaddr, - 1, 1, 0, NULL, NULL); + struct mm_struct *mm = current->mm; + int ret; + + down_read(&mm->mmap_sem); + ret = get_user_pages(current, mm, (unsigned long)uaddr, + 1, 1, 0, NULL, NULL); + up_read(&mm->mmap_sem); + return ret < 0 ? ret : 0; } -- cgit v1.2.3-71-gd317 From a7c312bed772c11138409c3a98531e85d690302e Mon Sep 17 00:00:00 2001 From: Masami Hiramatsu Date: Tue, 8 Dec 2009 17:03:16 -0500 Subject: trace-kprobe: Support delete probe syntax Support delete probe syntax. The syntax is "-:[group/]event". Signed-off-by: Masami Hiramatsu Cc: Steven Rostedt Cc: Jim Keniston Cc: Ananth N Mavinakayanahalli Cc: Christoph Hellwig Cc: Frank Ch. Eigler Cc: Frederic Weisbecker Cc: Jason Baron Cc: K.Prasad Cc: Peter Zijlstra Cc: Srikar Dronamraju Cc: Arnaldo Carvalho de Melo Cc: systemtap Cc: DLE LKML-Reference: <20091208220316.10142.39192.stgit@dhcp-100-2-132.bos.redhat.com> Signed-off-by: Ingo Molnar Cc: Steven Rostedt Cc: Jim Keniston Cc: Ananth N Mavinakayanahalli Cc: Christoph Hellwig Cc: Frank Ch. Eigler Cc: Frederic Weisbecker Cc: Jason Baron Cc: K.Prasad Cc: Peter Zijlstra Cc: Srikar Dronamraju Cc: Arnaldo Carvalho de Melo --- kernel/trace/trace_kprobe.c | 37 ++++++++++++++++++++++++++++--------- 1 file changed, 28 insertions(+), 9 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c index aff5f80b59b8..bf05fb49a6f3 100644 --- a/kernel/trace/trace_kprobe.c +++ b/kernel/trace/trace_kprobe.c @@ -606,23 +606,22 @@ static int create_trace_probe(int argc, char **argv) */ struct trace_probe *tp; int i, ret = 0; - int is_return = 0; + int is_return = 0, is_delete = 0; char *symbol = NULL, *event = NULL, *arg = NULL, *group = NULL; unsigned long offset = 0; void *addr = NULL; char buf[MAX_EVENT_NAME_LEN]; - if (argc < 2) { - pr_info("Probe point is not specified.\n"); - return -EINVAL; - } - + /* argc must be >= 1 */ if (argv[0][0] == 'p') is_return = 0; else if (argv[0][0] == 'r') is_return = 1; + else if (argv[0][0] == '-') + is_delete = 1; else { - pr_info("Probe definition must be started with 'p' or 'r'.\n"); + pr_info("Probe definition must be started with 'p', 'r' or" + " '-'.\n"); return -EINVAL; } @@ -642,7 +641,29 @@ static int create_trace_probe(int argc, char **argv) return -EINVAL; } } + if (!group) + group = KPROBE_EVENT_SYSTEM; + if (is_delete) { + if (!event) { + pr_info("Delete command needs an event name.\n"); + return -EINVAL; + } + tp = find_probe_event(event, group); + if (!tp) { + pr_info("Event %s/%s doesn't exist.\n", group, event); + return -ENOENT; + } + /* delete an event */ + unregister_trace_probe(tp); + free_trace_probe(tp); + return 0; + } + + if (argc < 2) { + pr_info("Probe point is not specified.\n"); + return -EINVAL; + } if (isdigit(argv[1][0])) { if (is_return) { pr_info("Return probe point must be a symbol.\n"); @@ -671,8 +692,6 @@ static int create_trace_probe(int argc, char **argv) argc -= 2; argv += 2; /* setup a probe */ - if (!group) - group = KPROBE_EVENT_SYSTEM; if (!event) { /* Make a new event name */ if (symbol) -- cgit v1.2.3-71-gd317 From 44234adcdce38f83c56e05f808ce656175b4beeb Mon Sep 17 00:00:00 2001 From: Frederic Weisbecker Date: Wed, 9 Dec 2009 09:25:48 +0100 Subject: hw-breakpoints: Modify breakpoints without unregistering them Currently, when ptrace needs to modify a breakpoint, like disabling it, changing its address, type or len, it calls modify_user_hw_breakpoint(). This latter will perform the heavy and racy task of unregistering the old breakpoint and registering a new one. This is racy as someone else might steal the reserved breakpoint slot under us, which is undesired as the breakpoint is only supposed to be modified, sometimes in the middle of a debugging workflow. We don't want our slot to be stolen in the middle. So instead of unregistering/registering the breakpoint, just disable it while we modify its breakpoint fields and re-enable it after if necessary. Signed-off-by: Frederic Weisbecker Cc: Peter Zijlstra Cc: Prasad LKML-Reference: <1260347148-5519-1-git-send-regression-fweisbec@gmail.com> Signed-off-by: Ingo Molnar --- arch/x86/kernel/ptrace.c | 57 ++++++++++++++++++++----------------------- include/linux/hw_breakpoint.h | 4 +-- include/linux/perf_event.h | 5 +++- kernel/hw_breakpoint.c | 42 +++++++++++++++++++++++-------- kernel/perf_event.c | 4 +-- 5 files changed, 66 insertions(+), 46 deletions(-) (limited to 'kernel') diff --git a/arch/x86/kernel/ptrace.c b/arch/x86/kernel/ptrace.c index b361d28061d0..7079ddaf0731 100644 --- a/arch/x86/kernel/ptrace.c +++ b/arch/x86/kernel/ptrace.c @@ -595,7 +595,7 @@ static unsigned long ptrace_get_dr7(struct perf_event *bp[]) return dr7; } -static struct perf_event * +static int ptrace_modify_breakpoint(struct perf_event *bp, int len, int type, struct task_struct *tsk, int disabled) { @@ -609,11 +609,11 @@ ptrace_modify_breakpoint(struct perf_event *bp, int len, int type, * written the address register first */ if (!bp) - return ERR_PTR(-EINVAL); + return -EINVAL; err = arch_bp_generic_fields(len, type, &gen_len, &gen_type); if (err) - return ERR_PTR(err); + return err; attr = bp->attr; attr.bp_len = gen_len; @@ -658,28 +658,17 @@ restore: if (!second_pass) continue; - thread->ptrace_bps[i] = NULL; - bp = ptrace_modify_breakpoint(bp, len, type, + rc = ptrace_modify_breakpoint(bp, len, type, tsk, 1); - if (IS_ERR(bp)) { - rc = PTR_ERR(bp); - thread->ptrace_bps[i] = NULL; + if (rc) break; - } - thread->ptrace_bps[i] = bp; } continue; } - bp = ptrace_modify_breakpoint(bp, len, type, tsk, 0); - - /* Incorrect bp, or we have a bug in bp API */ - if (IS_ERR(bp)) { - rc = PTR_ERR(bp); - thread->ptrace_bps[i] = NULL; + rc = ptrace_modify_breakpoint(bp, len, type, tsk, 0); + if (rc) break; - } - thread->ptrace_bps[i] = bp; } /* * Make a second pass to free the remaining unused breakpoints @@ -737,26 +726,32 @@ static int ptrace_set_breakpoint_addr(struct task_struct *tsk, int nr, attr.disabled = 1; bp = register_user_hw_breakpoint(&attr, ptrace_triggered, tsk); + + /* + * CHECKME: the previous code returned -EIO if the addr wasn't + * a valid task virtual addr. The new one will return -EINVAL in + * this case. + * -EINVAL may be what we want for in-kernel breakpoints users, + * but -EIO looks better for ptrace, since we refuse a register + * writing for the user. And anyway this is the previous + * behaviour. + */ + if (IS_ERR(bp)) + return PTR_ERR(bp); + + t->ptrace_bps[nr] = bp; } else { + int err; + bp = t->ptrace_bps[nr]; - t->ptrace_bps[nr] = NULL; attr = bp->attr; attr.bp_addr = addr; - bp = modify_user_hw_breakpoint(bp, &attr); + err = modify_user_hw_breakpoint(bp, &attr); + if (err) + return err; } - /* - * CHECKME: the previous code returned -EIO if the addr wasn't a - * valid task virtual addr. The new one will return -EINVAL in this - * case. - * -EINVAL may be what we want for in-kernel breakpoints users, but - * -EIO looks better for ptrace, since we refuse a register writing - * for the user. And anyway this is the previous behaviour. - */ - if (IS_ERR(bp)) - return PTR_ERR(bp); - t->ptrace_bps[nr] = bp; return 0; } diff --git a/include/linux/hw_breakpoint.h b/include/linux/hw_breakpoint.h index 42da1ce19ec0..69f07a9f1277 100644 --- a/include/linux/hw_breakpoint.h +++ b/include/linux/hw_breakpoint.h @@ -55,7 +55,7 @@ register_user_hw_breakpoint(struct perf_event_attr *attr, struct task_struct *tsk); /* FIXME: only change from the attr, and don't unregister */ -extern struct perf_event * +extern int modify_user_hw_breakpoint(struct perf_event *bp, struct perf_event_attr *attr); /* @@ -91,7 +91,7 @@ static inline struct perf_event * register_user_hw_breakpoint(struct perf_event_attr *attr, perf_overflow_handler_t triggered, struct task_struct *tsk) { return NULL; } -static inline struct perf_event * +static inline int modify_user_hw_breakpoint(struct perf_event *bp, struct perf_event_attr *attr) { return NULL; } static inline struct perf_event * diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h index bf3329413e18..64a53f74c9a9 100644 --- a/include/linux/perf_event.h +++ b/include/linux/perf_event.h @@ -872,6 +872,8 @@ extern void perf_output_copy(struct perf_output_handle *handle, const void *buf, unsigned int len); extern int perf_swevent_get_recursion_context(void); extern void perf_swevent_put_recursion_context(int rctx); +extern void perf_event_enable(struct perf_event *event); +extern void perf_event_disable(struct perf_event *event); #else static inline void perf_event_task_sched_in(struct task_struct *task, int cpu) { } @@ -902,7 +904,8 @@ static inline void perf_event_fork(struct task_struct *tsk) { } static inline void perf_event_init(void) { } static inline int perf_swevent_get_recursion_context(void) { return -1; } static inline void perf_swevent_put_recursion_context(int rctx) { } - +static inline void perf_event_enable(struct perf_event *event) { } +static inline void perf_event_disable(struct perf_event *event) { } #endif #define perf_output_put(handle, x) \ diff --git a/kernel/hw_breakpoint.c b/kernel/hw_breakpoint.c index 03a0773ac2b2..366eedf949c0 100644 --- a/kernel/hw_breakpoint.c +++ b/kernel/hw_breakpoint.c @@ -320,18 +320,40 @@ EXPORT_SYMBOL_GPL(register_user_hw_breakpoint); * @triggered: callback to trigger when we hit the breakpoint * @tsk: pointer to 'task_struct' of the process to which the address belongs */ -struct perf_event * -modify_user_hw_breakpoint(struct perf_event *bp, struct perf_event_attr *attr) +int modify_user_hw_breakpoint(struct perf_event *bp, struct perf_event_attr *attr) { - /* - * FIXME: do it without unregistering - * - We don't want to lose our slot - * - If the new bp is incorrect, don't lose the older one - */ - unregister_hw_breakpoint(bp); + u64 old_addr = bp->attr.bp_addr; + int old_type = bp->attr.bp_type; + int old_len = bp->attr.bp_len; + int err = 0; + + perf_event_disable(bp); + + bp->attr.bp_addr = attr->bp_addr; + bp->attr.bp_type = attr->bp_type; + bp->attr.bp_len = attr->bp_len; + + if (attr->disabled) + goto end; - return perf_event_create_kernel_counter(attr, -1, bp->ctx->task->pid, - bp->overflow_handler); + err = arch_validate_hwbkpt_settings(bp, bp->ctx->task); + if (!err) + perf_event_enable(bp); + + if (err) { + bp->attr.bp_addr = old_addr; + bp->attr.bp_type = old_type; + bp->attr.bp_len = old_len; + if (!bp->attr.disabled) + perf_event_enable(bp); + + return err; + } + +end: + bp->attr.disabled = attr->disabled; + + return 0; } EXPORT_SYMBOL_GPL(modify_user_hw_breakpoint); diff --git a/kernel/perf_event.c b/kernel/perf_event.c index fd43ff4ac860..3b0cf86eee84 100644 --- a/kernel/perf_event.c +++ b/kernel/perf_event.c @@ -567,7 +567,7 @@ static void __perf_event_disable(void *info) * is the current context on this CPU and preemption is disabled, * hence we can't get into perf_event_task_sched_out for this context. */ -static void perf_event_disable(struct perf_event *event) +void perf_event_disable(struct perf_event *event) { struct perf_event_context *ctx = event->ctx; struct task_struct *task = ctx->task; @@ -971,7 +971,7 @@ static void __perf_event_enable(void *info) * perf_event_for_each_child or perf_event_for_each as described * for perf_event_disable. */ -static void perf_event_enable(struct perf_event *event) +void perf_event_enable(struct perf_event *event) { struct perf_event_context *ctx = event->ctx; struct task_struct *task = ctx->task; -- cgit v1.2.3-71-gd317 From aa5452d70c0d559310598b243b8b1033c10056e7 Mon Sep 17 00:00:00 2001 From: Xiao Guangrong Date: Wed, 9 Dec 2009 11:28:13 +0800 Subject: perf_event: Clean up __perf_event_init_context() Clean up the code a bit: - define 'perf_cpu_context' variable with 'static' - use kzalloc() instead of kmalloc() and memset() Signed-off-by: Xiao Guangrong Reviewed-by: Frederic Weisbecker Cc: Peter Zijlstra Cc: Paul Mackerras LKML-Reference: <4B1F194D.7080306@cn.fujitsu.com> Signed-off-by: Ingo Molnar --- kernel/perf_event.c | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) (limited to 'kernel') diff --git a/kernel/perf_event.c b/kernel/perf_event.c index 3b0cf86eee84..2b06c45bfba9 100644 --- a/kernel/perf_event.c +++ b/kernel/perf_event.c @@ -36,7 +36,7 @@ /* * Each CPU has a list of per CPU events: */ -DEFINE_PER_CPU(struct perf_cpu_context, perf_cpu_context); +static DEFINE_PER_CPU(struct perf_cpu_context, perf_cpu_context); int perf_max_events __read_mostly = 1; static int perf_reserved_percpu __read_mostly; @@ -1579,7 +1579,6 @@ static void __perf_event_init_context(struct perf_event_context *ctx, struct task_struct *task) { - memset(ctx, 0, sizeof(*ctx)); spin_lock_init(&ctx->lock); mutex_init(&ctx->mutex); INIT_LIST_HEAD(&ctx->group_list); @@ -1654,7 +1653,7 @@ static struct perf_event_context *find_get_context(pid_t pid, int cpu) } if (!ctx) { - ctx = kmalloc(sizeof(struct perf_event_context), GFP_KERNEL); + ctx = kzalloc(sizeof(struct perf_event_context), GFP_KERNEL); err = -ENOMEM; if (!ctx) goto errout; @@ -5105,7 +5104,7 @@ int perf_event_init_task(struct task_struct *child) * First allocate and initialize a context for the child. */ - child_ctx = kmalloc(sizeof(struct perf_event_context), GFP_KERNEL); + child_ctx = kzalloc(sizeof(struct perf_event_context), GFP_KERNEL); if (!child_ctx) return -ENOMEM; -- cgit v1.2.3-71-gd317 From b93f7978ad6b46133e9453b90ccc057dc2429e75 Mon Sep 17 00:00:00 2001 From: Xiao Guangrong Date: Wed, 9 Dec 2009 11:29:44 +0800 Subject: perf_event: Allocate children's perf_event_ctxp at the right time In current code, children task will allocate memory for 'child->perf_event_ctxp' if the parent is counted, we can do it only if the parent allowed children inherit it. It can save memory and reduce overhead. Signed-off-by: Xiao Guangrong Reviewed-by: Frederic Weisbecker Cc: Peter Zijlstra Cc: Paul Mackerras LKML-Reference: <4B1F19A8.5040805@cn.fujitsu.com> Signed-off-by: Ingo Molnar --- kernel/perf_event.c | 37 ++++++++++++++++++++++--------------- 1 file changed, 22 insertions(+), 15 deletions(-) (limited to 'kernel') diff --git a/kernel/perf_event.c b/kernel/perf_event.c index 2b06c45bfba9..77641ae6b23f 100644 --- a/kernel/perf_event.c +++ b/kernel/perf_event.c @@ -5083,7 +5083,7 @@ again: */ int perf_event_init_task(struct task_struct *child) { - struct perf_event_context *child_ctx, *parent_ctx; + struct perf_event_context *child_ctx = NULL, *parent_ctx; struct perf_event_context *cloned_ctx; struct perf_event *event; struct task_struct *parent = current; @@ -5098,20 +5098,6 @@ int perf_event_init_task(struct task_struct *child) if (likely(!parent->perf_event_ctxp)) return 0; - /* - * This is executed from the parent task context, so inherit - * events that have been marked for cloning. - * First allocate and initialize a context for the child. - */ - - child_ctx = kzalloc(sizeof(struct perf_event_context), GFP_KERNEL); - if (!child_ctx) - return -ENOMEM; - - __perf_event_init_context(child_ctx, child); - child->perf_event_ctxp = child_ctx; - get_task_struct(child); - /* * If the parent's context is a clone, pin it so it won't get * swapped under us. @@ -5142,6 +5128,26 @@ int perf_event_init_task(struct task_struct *child) continue; } + if (!child->perf_event_ctxp) { + /* + * This is executed from the parent task context, so + * inherit events that have been marked for cloning. + * First allocate and initialize a context for the + * child. + */ + + child_ctx = kzalloc(sizeof(struct perf_event_context), + GFP_KERNEL); + if (!child_ctx) { + ret = -ENOMEM; + goto exit; + } + + __perf_event_init_context(child_ctx, child); + child->perf_event_ctxp = child_ctx; + get_task_struct(child); + } + ret = inherit_group(event, parent, parent_ctx, child, child_ctx); if (ret) { @@ -5170,6 +5176,7 @@ int perf_event_init_task(struct task_struct *child) get_ctx(child_ctx->parent_ctx); } +exit: mutex_unlock(&parent_ctx->mutex); perf_unpin_context(parent_ctx); -- cgit v1.2.3-71-gd317 From ec89a06fd4e12301f11ab039ee07d2353a18addc Mon Sep 17 00:00:00 2001 From: Xiao Guangrong Date: Wed, 9 Dec 2009 11:30:36 +0800 Subject: perf_event: Cleanup for cpu_clock_perf_event_update() Using atomic64_xchg() instead of atomic64_read() and atomic64_set(). Signed-off-by: Xiao Guangrong Reviewed-by: Frederic Weisbecker Cc: Peter Zijlstra Cc: Paul Mackerras LKML-Reference: <4B1F19DC.90204@cn.fujitsu.com> Signed-off-by: Ingo Molnar --- kernel/perf_event.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/perf_event.c b/kernel/perf_event.c index 77641ae6b23f..94e1b28333ae 100644 --- a/kernel/perf_event.c +++ b/kernel/perf_event.c @@ -4079,8 +4079,7 @@ static void cpu_clock_perf_event_update(struct perf_event *event) u64 now; now = cpu_clock(cpu); - prev = atomic64_read(&event->hw.prev_count); - atomic64_set(&event->hw.prev_count, now); + prev = atomic64_xchg(&event->hw.prev_count, now); atomic64_add(now - prev, &event->count); } -- cgit v1.2.3-71-gd317 From 3160568371da441b7f2fb57f2f1225404207e8f2 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 8 Dec 2009 20:24:16 +0000 Subject: sched: Protect task->cpus_allowed access in sched_getaffinity() sched_getaffinity() is not protected against a concurrent modification of the tasks affinity. Serialize the access with task_rq_lock(task). Signed-off-by: Thomas Gleixner Acked-by: Peter Zijlstra LKML-Reference: <20091208202026.769251187@linutronix.de> Signed-off-by: Ingo Molnar --- kernel/sched.c | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'kernel') diff --git a/kernel/sched.c b/kernel/sched.c index 281da29d0801..c4635f74540c 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -6631,6 +6631,8 @@ SYSCALL_DEFINE3(sched_setaffinity, pid_t, pid, unsigned int, len, long sched_getaffinity(pid_t pid, struct cpumask *mask) { struct task_struct *p; + unsigned long flags; + struct rq *rq; int retval; get_online_cpus(); @@ -6645,7 +6647,9 @@ long sched_getaffinity(pid_t pid, struct cpumask *mask) if (retval) goto out_unlock; + rq = task_rq_lock(p, &flags); cpumask_and(mask, &p->cpus_allowed, cpu_online_mask); + task_rq_unlock(rq, &flags); out_unlock: read_unlock(&tasklist_lock); -- cgit v1.2.3-71-gd317 From dba091b9e3522b9d32fc9975e48d3b69633b45f0 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Wed, 9 Dec 2009 09:32:03 +0100 Subject: sched: Protect sched_rr_get_param() access to task->sched_class sched_rr_get_param calls task->sched_class->get_rr_interval(task) without protection against a concurrent sched_setscheduler() call which modifies task->sched_class. Serialize the access with task_rq_lock(task) and hand the rq pointer into get_rr_interval() as it's needed at least in the sched_fair implementation. Signed-off-by: Thomas Gleixner Acked-by: Peter Zijlstra LKML-Reference: Signed-off-by: Ingo Molnar --- include/linux/sched.h | 3 ++- kernel/sched.c | 6 +++++- kernel/sched_fair.c | 6 +----- kernel/sched_idletask.c | 2 +- kernel/sched_rt.c | 2 +- 5 files changed, 10 insertions(+), 9 deletions(-) (limited to 'kernel') diff --git a/include/linux/sched.h b/include/linux/sched.h index 89115ec7d43f..9b2402725088 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1111,7 +1111,8 @@ struct sched_class { void (*prio_changed) (struct rq *this_rq, struct task_struct *task, int oldprio, int running); - unsigned int (*get_rr_interval) (struct task_struct *task); + unsigned int (*get_rr_interval) (struct rq *rq, + struct task_struct *task); #ifdef CONFIG_FAIR_GROUP_SCHED void (*moved_group) (struct task_struct *p); diff --git a/kernel/sched.c b/kernel/sched.c index c4635f74540c..68db5a2e6545 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -6887,6 +6887,8 @@ SYSCALL_DEFINE2(sched_rr_get_interval, pid_t, pid, { struct task_struct *p; unsigned int time_slice; + unsigned long flags; + struct rq *rq; int retval; struct timespec t; @@ -6903,7 +6905,9 @@ SYSCALL_DEFINE2(sched_rr_get_interval, pid_t, pid, if (retval) goto out_unlock; - time_slice = p->sched_class->get_rr_interval(p); + rq = task_rq_lock(p, &flags); + time_slice = p->sched_class->get_rr_interval(rq, p); + task_rq_unlock(rq, &flags); read_unlock(&tasklist_lock); jiffies_to_timespec(time_slice, &t); diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c index f61837ad336d..613c1c749677 100644 --- a/kernel/sched_fair.c +++ b/kernel/sched_fair.c @@ -2014,21 +2014,17 @@ static void moved_group_fair(struct task_struct *p) } #endif -unsigned int get_rr_interval_fair(struct task_struct *task) +unsigned int get_rr_interval_fair(struct rq *rq, struct task_struct *task) { struct sched_entity *se = &task->se; - unsigned long flags; - struct rq *rq; unsigned int rr_interval = 0; /* * Time slice is 0 for SCHED_OTHER tasks that are on an otherwise * idle runqueue: */ - rq = task_rq_lock(task, &flags); if (rq->cfs.load.weight) rr_interval = NS_TO_JIFFIES(sched_slice(&rq->cfs, se)); - task_rq_unlock(rq, &flags); return rr_interval; } diff --git a/kernel/sched_idletask.c b/kernel/sched_idletask.c index b133a28fcde3..33d5384a73a8 100644 --- a/kernel/sched_idletask.c +++ b/kernel/sched_idletask.c @@ -97,7 +97,7 @@ static void prio_changed_idle(struct rq *rq, struct task_struct *p, check_preempt_curr(rq, p, 0); } -unsigned int get_rr_interval_idle(struct task_struct *task) +unsigned int get_rr_interval_idle(struct rq *rq, struct task_struct *task) { return 0; } diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c index 5c5fef378415..aecbd9c6b20c 100644 --- a/kernel/sched_rt.c +++ b/kernel/sched_rt.c @@ -1721,7 +1721,7 @@ static void set_curr_task_rt(struct rq *rq) dequeue_pushable_task(rq, p); } -unsigned int get_rr_interval_rt(struct task_struct *task) +unsigned int get_rr_interval_rt(struct rq *rq, struct task_struct *task) { /* * Time slice is 0 for SCHED_FIFO tasks -- cgit v1.2.3-71-gd317 From 6b314d0e11924c803bf8cd944e87fd58cdb5088c Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 2 Dec 2009 18:58:05 +0100 Subject: sched: Remove sysctl.sched_features Since we've had a much saner debugfs interface to this, remove the sysctl one. Signed-off-by: Peter Zijlstra LKML-Reference: [ v2: build fix ] Signed-off-by: Ingo Molnar --- include/linux/sched.h | 1 - kernel/sysctl.c | 8 -------- 2 files changed, 9 deletions(-) (limited to 'kernel') diff --git a/include/linux/sched.h b/include/linux/sched.h index 9b2402725088..ca72ed42ac34 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1905,7 +1905,6 @@ extern unsigned int sysctl_sched_shares_ratelimit; extern unsigned int sysctl_sched_shares_thresh; extern unsigned int sysctl_sched_child_runs_first; #ifdef CONFIG_SCHED_DEBUG -extern unsigned int sysctl_sched_features; extern unsigned int sysctl_sched_migration_cost; extern unsigned int sysctl_sched_nr_migrate; extern unsigned int sysctl_sched_time_avg; diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 4dbf93a52ee9..e5cc53514caa 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -314,14 +314,6 @@ static struct ctl_table kern_table[] = { .strategy = &sysctl_intvec, .extra1 = &zero, }, - { - .ctl_name = CTL_UNNUMBERED, - .procname = "sched_features", - .data = &sysctl_sched_features, - .maxlen = sizeof(unsigned int), - .mode = 0644, - .proc_handler = &proc_dointvec, - }, { .ctl_name = CTL_UNNUMBERED, .procname = "sched_migration_cost", -- cgit v1.2.3-71-gd317 From 970b13bacba14a8cef6f642861947df1d175b0b3 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 25 Nov 2009 13:31:39 +0100 Subject: sched: Consolidate select_task_rq() callers Small cleanup. Signed-off-by: Peter Zijlstra LKML-Reference: [ v2: build fix ] Signed-off-by: Ingo Molnar --- kernel/sched.c | 14 +++++++++++--- 1 file changed, 11 insertions(+), 3 deletions(-) (limited to 'kernel') diff --git a/kernel/sched.c b/kernel/sched.c index 68db5a2e6545..01fd131b47a4 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -2323,6 +2323,14 @@ void task_oncpu_function_call(struct task_struct *p, preempt_enable(); } +#ifdef CONFIG_SMP +static inline +int select_task_rq(struct task_struct *p, int sd_flags, int wake_flags) +{ + return p->sched_class->select_task_rq(p, sd_flags, wake_flags); +} +#endif + /*** * try_to_wake_up - wake up a thread * @p: the to-be-woken-up thread @@ -2376,7 +2384,7 @@ static int try_to_wake_up(struct task_struct *p, unsigned int state, p->state = TASK_WAKING; task_rq_unlock(rq, &flags); - cpu = p->sched_class->select_task_rq(p, SD_BALANCE_WAKE, wake_flags); + cpu = select_task_rq(p, SD_BALANCE_WAKE, wake_flags); if (cpu != orig_cpu) { local_irq_save(flags); rq = cpu_rq(cpu); @@ -2593,7 +2601,7 @@ void sched_fork(struct task_struct *p, int clone_flags) p->sched_class = &fair_sched_class; #ifdef CONFIG_SMP - cpu = p->sched_class->select_task_rq(p, SD_BALANCE_FORK, 0); + cpu = select_task_rq(p, SD_BALANCE_FORK, 0); #endif local_irq_save(flags); update_rq_clock(cpu_rq(cpu)); @@ -3156,7 +3164,7 @@ out: void sched_exec(void) { int new_cpu, this_cpu = get_cpu(); - new_cpu = current->sched_class->select_task_rq(current, SD_BALANCE_EXEC, 0); + new_cpu = select_task_rq(current, SD_BALANCE_EXEC, 0); put_cpu(); if (new_cpu != this_cpu) sched_migrate_task(current, new_cpu); -- cgit v1.2.3-71-gd317 From 5afcdab706d6002cb02b567ba46e650215e694e8 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Fri, 27 Nov 2009 14:12:25 +0100 Subject: sched: Remove rq->clock coupling from set_task_cpu() set_task_cpu() should be rq invariant and only touch task state, it currently fails to do so, which opens up a few races, since not all callers hold both rq->locks. Remove the relyance on rq->clock, as any site calling set_task_cpu() should also do a remote clock update, which should ensure the observed time between these two cpus is monotonic, as per kernel/sched_clock.c:sched_clock_remote(). Therefore we can simply remove the clock_offset bits and be happy. Signed-off-by: Peter Zijlstra LKML-Reference: Signed-off-by: Ingo Molnar --- kernel/sched.c | 13 +------------ 1 file changed, 1 insertion(+), 12 deletions(-) (limited to 'kernel') diff --git a/kernel/sched.c b/kernel/sched.c index 01fd131b47a4..1f9c6d99f15d 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -2060,23 +2060,12 @@ task_hot(struct task_struct *p, u64 now, struct sched_domain *sd) void set_task_cpu(struct task_struct *p, unsigned int new_cpu) { int old_cpu = task_cpu(p); - struct rq *old_rq = cpu_rq(old_cpu), *new_rq = cpu_rq(new_cpu); + struct rq *old_rq = cpu_rq(old_cpu); struct cfs_rq *old_cfsrq = task_cfs_rq(p), *new_cfsrq = cpu_cfs_rq(old_cfsrq, new_cpu); - u64 clock_offset; - - clock_offset = old_rq->clock - new_rq->clock; trace_sched_migrate_task(p, new_cpu); -#ifdef CONFIG_SCHEDSTATS - if (p->se.wait_start) - p->se.wait_start -= clock_offset; - if (p->se.sleep_start) - p->se.sleep_start -= clock_offset; - if (p->se.block_start) - p->se.block_start -= clock_offset; -#endif if (old_cpu != new_cpu) { p->se.nr_migrations++; #ifdef CONFIG_SCHEDSTATS -- cgit v1.2.3-71-gd317 From ab19cb23313733c55e0517607844b86720b35f5f Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Fri, 27 Nov 2009 15:44:43 +0100 Subject: sched: Clean up ttwu() rq locking Since set_task_clock() doesn't rely on rq->clock anymore we can simplyfy the mess in ttwu(). Optimize things a bit by not fiddling with the IRQ state there. Signed-off-by: Peter Zijlstra LKML-Reference: Signed-off-by: Ingo Molnar --- kernel/sched.c | 13 +++++-------- 1 file changed, 5 insertions(+), 8 deletions(-) (limited to 'kernel') diff --git a/kernel/sched.c b/kernel/sched.c index 1f9c6d99f15d..c92670f8e097 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -2371,17 +2371,14 @@ static int try_to_wake_up(struct task_struct *p, unsigned int state, if (task_contributes_to_load(p)) rq->nr_uninterruptible--; p->state = TASK_WAKING; - task_rq_unlock(rq, &flags); + __task_rq_unlock(rq); cpu = select_task_rq(p, SD_BALANCE_WAKE, wake_flags); - if (cpu != orig_cpu) { - local_irq_save(flags); - rq = cpu_rq(cpu); - update_rq_clock(rq); + if (cpu != orig_cpu) set_task_cpu(p, cpu); - local_irq_restore(flags); - } - rq = task_rq_lock(p, &flags); + + rq = __task_rq_lock(p); + update_rq_clock(rq); WARN_ON(p->state != TASK_WAKING); cpu = task_cpu(p); -- cgit v1.2.3-71-gd317 From cd29fe6f2637cc2ccbda5ac65f5332d6bf5fa3c6 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Fri, 27 Nov 2009 17:32:46 +0100 Subject: sched: Sanitize fork() handling Currently we try to do task placement in wake_up_new_task() after we do the load-balance pass in sched_fork(). This yields complicated semantics in that we have to deal with tasks on different RQs and the set_task_cpu() calls in copy_process() and sched_fork() Rename ->task_new() to ->task_fork() and call it from sched_fork() before the balancing, this gives the policy a clear point to place the task. Signed-off-by: Peter Zijlstra LKML-Reference: Signed-off-by: Ingo Molnar --- include/linux/sched.h | 2 +- kernel/sched.c | 47 ++++++++++++++++++----------------------------- kernel/sched_fair.c | 28 +++++++++++++++------------- 3 files changed, 34 insertions(+), 43 deletions(-) (limited to 'kernel') diff --git a/include/linux/sched.h b/include/linux/sched.h index ca72ed42ac34..31d9dec78675 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1102,7 +1102,7 @@ struct sched_class { void (*set_curr_task) (struct rq *rq); void (*task_tick) (struct rq *rq, struct task_struct *p, int queued); - void (*task_new) (struct rq *rq, struct task_struct *p); + void (*task_fork) (struct task_struct *p); void (*switched_from) (struct rq *this_rq, struct task_struct *task, int running); diff --git a/kernel/sched.c b/kernel/sched.c index c92670f8e097..33c903573132 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -1811,6 +1811,20 @@ static void cfs_rq_set_shares(struct cfs_rq *cfs_rq, unsigned long shares) static void calc_load_account_active(struct rq *this_rq); +static inline void __set_task_cpu(struct task_struct *p, unsigned int cpu) +{ + set_task_rq(p, cpu); +#ifdef CONFIG_SMP + /* + * After ->cpu is set up to a new value, task_rq_lock(p, ...) can be + * successfuly executed on another CPU. We must ensure that updates of + * per-task data have been completed by this moment. + */ + smp_wmb(); + task_thread_info(p)->cpu = cpu; +#endif +} + #include "sched_stats.h" #include "sched_idletask.c" #include "sched_fair.c" @@ -1967,20 +1981,6 @@ inline int task_curr(const struct task_struct *p) return cpu_curr(task_cpu(p)) == p; } -static inline void __set_task_cpu(struct task_struct *p, unsigned int cpu) -{ - set_task_rq(p, cpu); -#ifdef CONFIG_SMP - /* - * After ->cpu is set up to a new value, task_rq_lock(p, ...) can be - * successfuly executed on another CPU. We must ensure that updates of - * per-task data have been completed by this moment. - */ - smp_wmb(); - task_thread_info(p)->cpu = cpu; -#endif -} - static inline void check_class_changed(struct rq *rq, struct task_struct *p, const struct sched_class *prev_class, int oldprio, int running) @@ -2552,7 +2552,6 @@ static void __sched_fork(struct task_struct *p) void sched_fork(struct task_struct *p, int clone_flags) { int cpu = get_cpu(); - unsigned long flags; __sched_fork(p); @@ -2586,13 +2585,13 @@ void sched_fork(struct task_struct *p, int clone_flags) if (!rt_prio(p->prio)) p->sched_class = &fair_sched_class; + if (p->sched_class->task_fork) + p->sched_class->task_fork(p); + #ifdef CONFIG_SMP cpu = select_task_rq(p, SD_BALANCE_FORK, 0); #endif - local_irq_save(flags); - update_rq_clock(cpu_rq(cpu)); set_task_cpu(p, cpu); - local_irq_restore(flags); #if defined(CONFIG_SCHEDSTATS) || defined(CONFIG_TASK_DELAY_ACCT) if (likely(sched_info_on())) @@ -2625,17 +2624,7 @@ void wake_up_new_task(struct task_struct *p, unsigned long clone_flags) rq = task_rq_lock(p, &flags); BUG_ON(p->state != TASK_RUNNING); update_rq_clock(rq); - - if (!p->sched_class->task_new || !current->se.on_rq) { - activate_task(rq, p, 0); - } else { - /* - * Let the scheduling class do new task startup - * management (if any): - */ - p->sched_class->task_new(rq, p); - inc_nr_running(rq); - } + activate_task(rq, p, 0); trace_sched_wakeup_new(rq, p, 1); check_preempt_curr(rq, p, WF_FORK); #ifdef CONFIG_SMP diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c index 613c1c749677..44ec80ccfa85 100644 --- a/kernel/sched_fair.c +++ b/kernel/sched_fair.c @@ -1922,28 +1922,30 @@ static void task_tick_fair(struct rq *rq, struct task_struct *curr, int queued) } /* - * Share the fairness runtime between parent and child, thus the - * total amount of pressure for CPU stays equal - new tasks - * get a chance to run but frequent forkers are not allowed to - * monopolize the CPU. Note: the parent runqueue is locked, - * the child is not running yet. + * called on fork with the child task as argument from the parent's context + * - child not yet on the tasklist + * - preemption disabled */ -static void task_new_fair(struct rq *rq, struct task_struct *p) +static void task_fork_fair(struct task_struct *p) { - struct cfs_rq *cfs_rq = task_cfs_rq(p); + struct cfs_rq *cfs_rq = task_cfs_rq(current); struct sched_entity *se = &p->se, *curr = cfs_rq->curr; int this_cpu = smp_processor_id(); + struct rq *rq = this_rq(); + unsigned long flags; + + spin_lock_irqsave(&rq->lock, flags); - sched_info_queued(p); + if (unlikely(task_cpu(p) != this_cpu)) + __set_task_cpu(p, this_cpu); update_curr(cfs_rq); + if (curr) se->vruntime = curr->vruntime; place_entity(cfs_rq, se, 1); - /* 'curr' will be NULL if the child belongs to a different group */ - if (sysctl_sched_child_runs_first && this_cpu == task_cpu(p) && - curr && entity_before(curr, se)) { + if (sysctl_sched_child_runs_first && curr && entity_before(curr, se)) { /* * Upon rescheduling, sched_class::put_prev_task() will place * 'current' within the tree based on its new key value. @@ -1952,7 +1954,7 @@ static void task_new_fair(struct rq *rq, struct task_struct *p) resched_task(rq->curr); } - enqueue_task_fair(rq, p, 0); + spin_unlock_irqrestore(&rq->lock, flags); } /* @@ -2052,7 +2054,7 @@ static const struct sched_class fair_sched_class = { .set_curr_task = set_curr_task_fair, .task_tick = task_tick_fair, - .task_new = task_new_fair, + .task_fork = task_fork_fair, .prio_changed = prio_changed_fair, .switched_to = switched_to_fair, -- cgit v1.2.3-71-gd317 From a65ac745e47e91f9d98dbf07f22ed0492e34d998 Mon Sep 17 00:00:00 2001 From: Jupyung Lee Date: Tue, 17 Nov 2009 18:51:40 +0900 Subject: sched: Move update_curr() in check_preempt_wakeup() to avoid redundant call If a RT task is woken up while a non-RT task is running, check_preempt_wakeup() is called to check whether the new task can preempt the old task. The function returns quickly without going deeper because it is apparent that a RT task can always preempt a non-RT task. In this situation, check_preempt_wakeup() always calls update_curr() to update vruntime value of the currently running task. However, the function call is unnecessary and redundant at that moment because (1) a non-RT task can always be preempted by a RT task regardless of its vruntime value, and (2) update_curr() will be called shortly when the context switch between two occurs. By moving update_curr() in check_preempt_wakeup(), we can avoid redundant call to update_curr(), slightly reducing the time taken to wake up RT tasks. Signed-off-by: Jupyung Lee [ Place update_curr() right before the wake_preempt_entity() call, which is the only thing that relies on the updated vruntime ] Signed-off-by: Peter Zijlstra LKML-Reference: <1258451500-6714-1-git-send-email-jupyung@gmail.com> Signed-off-by: Ingo Molnar --- kernel/sched_fair.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c index 44ec80ccfa85..4dec18579c9a 100644 --- a/kernel/sched_fair.c +++ b/kernel/sched_fair.c @@ -1651,8 +1651,6 @@ static void check_preempt_wakeup(struct rq *rq, struct task_struct *p, int wake_ int sync = wake_flags & WF_SYNC; int scale = cfs_rq->nr_running >= sched_nr_latency; - update_curr(cfs_rq); - if (unlikely(rt_prio(p->prio))) { resched_task(curr); return; @@ -1710,6 +1708,8 @@ static void check_preempt_wakeup(struct rq *rq, struct task_struct *p, int wake_ BUG_ON(!pse); + update_curr(cfs_rq); + if (wakeup_preempt_entity(se, pse) == 1) { resched_task(curr); /* -- cgit v1.2.3-71-gd317 From 3a7e73a2e26fffdbc46ba95fc0425418984f5140 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Sat, 28 Nov 2009 18:51:02 +0100 Subject: sched: Clean up check_preempt_wakeup() Streamline the wakeup preemption code a bit, unifying the preempt path so that they all do the same. Signed-off-by: Peter Zijlstra LKML-Reference: Signed-off-by: Ingo Molnar --- kernel/sched_fair.c | 73 ++++++++++++++++++++++++----------------------------- 1 file changed, 33 insertions(+), 40 deletions(-) (limited to 'kernel') diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c index 4dec18579c9a..76b5792c4198 100644 --- a/kernel/sched_fair.c +++ b/kernel/sched_fair.c @@ -1651,10 +1651,8 @@ static void check_preempt_wakeup(struct rq *rq, struct task_struct *p, int wake_ int sync = wake_flags & WF_SYNC; int scale = cfs_rq->nr_running >= sched_nr_latency; - if (unlikely(rt_prio(p->prio))) { - resched_task(curr); - return; - } + if (unlikely(rt_prio(p->prio))) + goto preempt; if (unlikely(p->sched_class != &fair_sched_class)) return; @@ -1680,52 +1678,47 @@ static void check_preempt_wakeup(struct rq *rq, struct task_struct *p, int wake_ return; /* Idle tasks are by definition preempted by everybody. */ - if (unlikely(curr->policy == SCHED_IDLE)) { - resched_task(curr); - return; - } + if (unlikely(curr->policy == SCHED_IDLE)) + goto preempt; - if ((sched_feat(WAKEUP_SYNC) && sync) || - (sched_feat(WAKEUP_OVERLAP) && - (se->avg_overlap < sysctl_sched_migration_cost && - pse->avg_overlap < sysctl_sched_migration_cost))) { - resched_task(curr); - return; - } + if (sched_feat(WAKEUP_SYNC) && sync) + goto preempt; - if (sched_feat(WAKEUP_RUNNING)) { - if (pse->avg_running < se->avg_running) { - set_next_buddy(pse); - resched_task(curr); - return; - } - } + if (sched_feat(WAKEUP_OVERLAP) && + se->avg_overlap < sysctl_sched_migration_cost && + pse->avg_overlap < sysctl_sched_migration_cost) + goto preempt; + + if (sched_feat(WAKEUP_RUNNING) && pse->avg_running < se->avg_running) + goto preempt; if (!sched_feat(WAKEUP_PREEMPT)) return; + update_curr(cfs_rq); find_matching_se(&se, &pse); - BUG_ON(!pse); + if (wakeup_preempt_entity(se, pse) == 1) + goto preempt; - update_curr(cfs_rq); + return; - if (wakeup_preempt_entity(se, pse) == 1) { - resched_task(curr); - /* - * Only set the backward buddy when the current task is still - * on the rq. This can happen when a wakeup gets interleaved - * with schedule on the ->pre_schedule() or idle_balance() - * point, either of which can * drop the rq lock. - * - * Also, during early boot the idle thread is in the fair class, - * for obvious reasons its a bad idea to schedule back to it. - */ - if (unlikely(!se->on_rq || curr == rq->idle)) - return; - if (sched_feat(LAST_BUDDY) && scale && entity_is_task(se)) - set_last_buddy(se); - } +preempt: + resched_task(curr); + /* + * Only set the backward buddy when the current task is still + * on the rq. This can happen when a wakeup gets interleaved + * with schedule on the ->pre_schedule() or idle_balance() + * point, either of which can * drop the rq lock. + * + * Also, during early boot the idle thread is in the fair class, + * for obvious reasons its a bad idea to schedule back to it. + */ + if (unlikely(!se->on_rq || curr == rq->idle)) + return; + + if (sched_feat(LAST_BUDDY) && scale && entity_is_task(se)) + set_last_buddy(se); } static struct task_struct *pick_next_task_fair(struct rq *rq) -- cgit v1.2.3-71-gd317 From 6cecd084d0fd27bb1e498e2829fd45846d806856 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Mon, 30 Nov 2009 13:00:37 +0100 Subject: sched: Discard some old bits WAKEUP_RUNNING was an experiment, not sure why that ever ended up being merged... Signed-off-by: Peter Zijlstra LKML-Reference: Signed-off-by: Ingo Molnar --- include/linux/sched.h | 2 -- kernel/sched.c | 17 +++++++---------- kernel/sched_debug.c | 1 - kernel/sched_fair.c | 3 --- kernel/sched_features.h | 5 ----- 5 files changed, 7 insertions(+), 21 deletions(-) (limited to 'kernel') diff --git a/include/linux/sched.h b/include/linux/sched.h index 31d9dec78675..4b1ebd3280c6 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1152,8 +1152,6 @@ struct sched_entity { u64 start_runtime; u64 avg_wakeup; - u64 avg_running; - #ifdef CONFIG_SCHEDSTATS u64 wait_start; u64 wait_max; diff --git a/kernel/sched.c b/kernel/sched.c index 33c903573132..0170735bdafc 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -2493,7 +2493,6 @@ static void __sched_fork(struct task_struct *p) p->se.avg_overlap = 0; p->se.start_runtime = 0; p->se.avg_wakeup = sysctl_sched_wakeup_granularity; - p->se.avg_running = 0; #ifdef CONFIG_SCHEDSTATS p->se.wait_start = 0; @@ -5379,13 +5378,14 @@ static inline void schedule_debug(struct task_struct *prev) #endif } -static void put_prev_task(struct rq *rq, struct task_struct *p) +static void put_prev_task(struct rq *rq, struct task_struct *prev) { - u64 runtime = p->se.sum_exec_runtime - p->se.prev_sum_exec_runtime; + if (prev->state == TASK_RUNNING) { + u64 runtime = prev->se.sum_exec_runtime; - update_avg(&p->se.avg_running, runtime); + runtime -= prev->se.prev_sum_exec_runtime; + runtime = min_t(u64, runtime, 2*sysctl_sched_migration_cost); - if (p->state == TASK_RUNNING) { /* * In order to avoid avg_overlap growing stale when we are * indeed overlapping and hence not getting put to sleep, grow @@ -5395,12 +5395,9 @@ static void put_prev_task(struct rq *rq, struct task_struct *p) * correlates to the amount of cache footprint a task can * build up. */ - runtime = min_t(u64, runtime, 2*sysctl_sched_migration_cost); - update_avg(&p->se.avg_overlap, runtime); - } else { - update_avg(&p->se.avg_running, 0); + update_avg(&prev->se.avg_overlap, runtime); } - p->sched_class->put_prev_task(rq, p); + prev->sched_class->put_prev_task(rq, prev); } /* diff --git a/kernel/sched_debug.c b/kernel/sched_debug.c index 6988cf08f705..5fda66615fee 100644 --- a/kernel/sched_debug.c +++ b/kernel/sched_debug.c @@ -399,7 +399,6 @@ void proc_sched_show_task(struct task_struct *p, struct seq_file *m) PN(se.sum_exec_runtime); PN(se.avg_overlap); PN(se.avg_wakeup); - PN(se.avg_running); nr_switches = p->nvcsw + p->nivcsw; diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c index 76b5792c4198..e9f5daee12c7 100644 --- a/kernel/sched_fair.c +++ b/kernel/sched_fair.c @@ -1689,9 +1689,6 @@ static void check_preempt_wakeup(struct rq *rq, struct task_struct *p, int wake_ pse->avg_overlap < sysctl_sched_migration_cost) goto preempt; - if (sched_feat(WAKEUP_RUNNING) && pse->avg_running < se->avg_running) - goto preempt; - if (!sched_feat(WAKEUP_PREEMPT)) return; diff --git a/kernel/sched_features.h b/kernel/sched_features.h index 0d94083582c7..d5059fd761d9 100644 --- a/kernel/sched_features.h +++ b/kernel/sched_features.h @@ -53,11 +53,6 @@ SCHED_FEAT(WAKEUP_SYNC, 0) */ SCHED_FEAT(WAKEUP_OVERLAP, 0) -/* - * Wakeup preemption towards tasks that run short - */ -SCHED_FEAT(WAKEUP_RUNNING, 0) - /* * Use the SYNC wakeup hint, pipes and the likes use this to indicate * the remote end is likely to consume the data we just wrote, and -- cgit v1.2.3-71-gd317 From fb58bac5c75bfff8bbf7d02071a10a62f32fe28b Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Tue, 1 Dec 2009 12:21:47 +0100 Subject: sched: Remove unnecessary RCU exclusion As Nick pointed out, and realized by myself when doing: sched: Fix balance vs hotplug race the patch: sched: for_each_domain() vs RCU is wrong, sched_domains are freed after synchronize_sched(), which means disabling preemption is enough. Reported-by: Nick Piggin Signed-off-by: Peter Zijlstra LKML-Reference: Signed-off-by: Ingo Molnar --- kernel/sched_fair.c | 9 ++------- 1 file changed, 2 insertions(+), 7 deletions(-) (limited to 'kernel') diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c index e9f5daee12c7..c163a285bf05 100644 --- a/kernel/sched_fair.c +++ b/kernel/sched_fair.c @@ -1403,7 +1403,6 @@ static int select_task_rq_fair(struct task_struct *p, int sd_flag, int wake_flag new_cpu = prev_cpu; } - rcu_read_lock(); for_each_domain(cpu, tmp) { /* * If power savings logic is enabled for a domain, see if we @@ -1484,10 +1483,8 @@ static int select_task_rq_fair(struct task_struct *p, int sd_flag, int wake_flag update_shares(tmp); } - if (affine_sd && wake_affine(affine_sd, p, sync)) { - new_cpu = cpu; - goto out; - } + if (affine_sd && wake_affine(affine_sd, p, sync)) + return cpu; while (sd) { int load_idx = sd->forkexec_idx; @@ -1528,8 +1525,6 @@ static int select_task_rq_fair(struct task_struct *p, int sd_flag, int wake_flag /* while loop will break here if sd == NULL */ } -out: - rcu_read_unlock(); return new_cpu; } #endif /* CONFIG_SMP */ -- cgit v1.2.3-71-gd317 From cd8ad40de36c2fe75f3b731bd70198b385895246 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Thu, 3 Dec 2009 18:00:07 +0100 Subject: sched: cgroup: Implement different treatment for idle shares When setting the weight for a per-cpu task-group, we have to put in a phantom weight when there is no work on that cpu, otherwise we'll not service that cpu when new work gets placed there until we again update the per-cpu weights. We used to add these phantom weights to the total, so that the idle per-cpu shares don't get inflated, this however causes the non-idle parts to get deflated, causing unexpected weight distibutions. Reverse this, so that the non-idle shares are correct but the idle shares are inflated. Reported-by: Yasunori Goto Tested-by: Yasunori Goto Signed-off-by: Peter Zijlstra LKML-Reference: <1257934048.23203.76.camel@twins> Signed-off-by: Ingo Molnar --- kernel/sched.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/sched.c b/kernel/sched.c index 0170735bdafc..71eb0622f548 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -1614,7 +1614,7 @@ static void update_group_shares_cpu(struct task_group *tg, int cpu, */ static int tg_shares_up(struct task_group *tg, void *data) { - unsigned long weight, rq_weight = 0, shares = 0; + unsigned long weight, rq_weight = 0, sum_weight = 0, shares = 0; unsigned long *usd_rq_weight; struct sched_domain *sd = data; unsigned long flags; @@ -1630,6 +1630,7 @@ static int tg_shares_up(struct task_group *tg, void *data) weight = tg->cfs_rq[i]->load.weight; usd_rq_weight[i] = weight; + rq_weight += weight; /* * If there are currently no tasks on the cpu pretend there * is one of average load so that when a new task gets to @@ -1638,10 +1639,13 @@ static int tg_shares_up(struct task_group *tg, void *data) if (!weight) weight = NICE_0_LOAD; - rq_weight += weight; + sum_weight += weight; shares += tg->cfs_rq[i]->shares; } + if (!rq_weight) + rq_weight = sum_weight; + if ((!shares && rq_weight) || shares > tg->shares) shares = tg->shares; -- cgit v1.2.3-71-gd317 From 57785df5ac53c70da9fb53696130f3c551bfe1f9 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Fri, 4 Dec 2009 09:59:02 +0100 Subject: sched: Fix task priority bug 83f9ac removed a call to effective_prio() in wake_up_new_task(), which leads to tasks running at MAX_PRIO. This is caused by the idle thread being set to MAX_PRIO before forking off init. O(1) used that to make sure idle was always preempted, CFS uses check_preempt_curr_idle() for that so we can savely remove this bit of legacy code. Reported-by: Mike Galbraith Tested-by: Mike Galbraith Signed-off-by: Peter Zijlstra LKML-Reference: <1259754383.4003.610.camel@laptop> Signed-off-by: Ingo Molnar --- kernel/sched.c | 6 ------ 1 file changed, 6 deletions(-) (limited to 'kernel') diff --git a/kernel/sched.c b/kernel/sched.c index 71eb0622f548..3878f5018007 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -3158,10 +3158,6 @@ static void pull_task(struct rq *src_rq, struct task_struct *p, deactivate_task(src_rq, p, 0); set_task_cpu(p, this_cpu); activate_task(this_rq, p, 0); - /* - * Note that idle threads have a prio of MAX_PRIO, for this test - * to be always true for them. - */ check_preempt_curr(this_rq, p, 0); } @@ -6992,7 +6988,6 @@ void __cpuinit init_idle(struct task_struct *idle, int cpu) __sched_fork(idle); idle->se.exec_start = sched_clock(); - idle->prio = idle->normal_prio = MAX_PRIO; cpumask_copy(&idle->cpus_allowed, cpumask_of(cpu)); __set_task_cpu(idle, cpu); @@ -7696,7 +7691,6 @@ migration_call(struct notifier_block *nfb, unsigned long action, void *hcpu) spin_lock_irq(&rq->lock); update_rq_clock(rq); deactivate_task(rq, rq->idle, 0); - rq->idle->static_prio = MAX_PRIO; __setscheduler(rq, rq->idle, SCHED_NORMAL, 0); rq->idle->sched_class = &idle_sched_class; migrate_dead_tasks(cpu); -- cgit v1.2.3-71-gd317 From 0bcdcf28c979869f44e05121b96ff2cfb05bd8e6 Mon Sep 17 00:00:00 2001 From: Christian Ehrhardt Date: Mon, 30 Nov 2009 12:16:46 +0100 Subject: sched: Fix missing sched tunable recalculation on cpu add/remove Based on Peter Zijlstras patch suggestion this enables recalculation of the scheduler tunables in response of a change in the number of cpus. It also adds a max of eight cpus that are considered in that scaling. Signed-off-by: Christian Ehrhardt Signed-off-by: Peter Zijlstra LKML-Reference: <1259579808-11357-2-git-send-email-ehrhardt@linux.vnet.ibm.com> Signed-off-by: Ingo Molnar --- kernel/sched.c | 29 ++++++++++++++++------------- kernel/sched_fair.c | 16 ++++++++++++++++ 2 files changed, 32 insertions(+), 13 deletions(-) (limited to 'kernel') diff --git a/kernel/sched.c b/kernel/sched.c index 3878f5018007..b54ecf84b6be 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -814,6 +814,7 @@ const_debug unsigned int sysctl_sched_nr_migrate = 32; * default: 0.25ms */ unsigned int sysctl_sched_shares_ratelimit = 250000; +unsigned int normalized_sysctl_sched_shares_ratelimit = 250000; /* * Inject some fuzzyness into changing the per-cpu group shares @@ -1814,6 +1815,7 @@ static void cfs_rq_set_shares(struct cfs_rq *cfs_rq, unsigned long shares) #endif static void calc_load_account_active(struct rq *this_rq); +static void update_sysctl(void); static inline void __set_task_cpu(struct task_struct *p, unsigned int cpu) { @@ -7028,22 +7030,23 @@ cpumask_var_t nohz_cpu_mask; * * This idea comes from the SD scheduler of Con Kolivas: */ -static inline void sched_init_granularity(void) +static void update_sysctl(void) { - unsigned int factor = 1 + ilog2(num_online_cpus()); - const unsigned long limit = 200000000; - - sysctl_sched_min_granularity *= factor; - if (sysctl_sched_min_granularity > limit) - sysctl_sched_min_granularity = limit; - - sysctl_sched_latency *= factor; - if (sysctl_sched_latency > limit) - sysctl_sched_latency = limit; + unsigned int cpus = min(num_online_cpus(), 8U); + unsigned int factor = 1 + ilog2(cpus); - sysctl_sched_wakeup_granularity *= factor; +#define SET_SYSCTL(name) \ + (sysctl_##name = (factor) * normalized_sysctl_##name) + SET_SYSCTL(sched_min_granularity); + SET_SYSCTL(sched_latency); + SET_SYSCTL(sched_wakeup_granularity); + SET_SYSCTL(sched_shares_ratelimit); +#undef SET_SYSCTL +} - sysctl_sched_shares_ratelimit *= factor; +static inline void sched_init_granularity(void) +{ + update_sysctl(); } #ifdef CONFIG_SMP diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c index c163a285bf05..71b3458245e5 100644 --- a/kernel/sched_fair.c +++ b/kernel/sched_fair.c @@ -35,12 +35,14 @@ * run vmstat and monitor the context-switches (cs) field) */ unsigned int sysctl_sched_latency = 5000000ULL; +unsigned int normalized_sysctl_sched_latency = 5000000ULL; /* * Minimal preemption granularity for CPU-bound tasks: * (default: 1 msec * (1 + ilog(ncpus)), units: nanoseconds) */ unsigned int sysctl_sched_min_granularity = 1000000ULL; +unsigned int normalized_sysctl_sched_min_granularity = 1000000ULL; /* * is kept at sysctl_sched_latency / sysctl_sched_min_granularity @@ -70,6 +72,7 @@ unsigned int __read_mostly sysctl_sched_compat_yield; * have immediate wakeup/sleep latencies. */ unsigned int sysctl_sched_wakeup_granularity = 1000000UL; +unsigned int normalized_sysctl_sched_wakeup_granularity = 1000000UL; const_debug unsigned int sysctl_sched_migration_cost = 500000UL; @@ -1890,6 +1893,17 @@ move_one_task_fair(struct rq *this_rq, int this_cpu, struct rq *busiest, return 0; } + +static void rq_online_fair(struct rq *rq) +{ + update_sysctl(); +} + +static void rq_offline_fair(struct rq *rq) +{ + update_sysctl(); +} + #endif /* CONFIG_SMP */ /* @@ -2035,6 +2049,8 @@ static const struct sched_class fair_sched_class = { .load_balance = load_balance_fair, .move_one_task = move_one_task_fair, + .rq_online = rq_online_fair, + .rq_offline = rq_offline_fair, #endif .set_curr_task = set_curr_task_fair, -- cgit v1.2.3-71-gd317 From 1983a922a1bc843806b9a36cf3a370b242783140 Mon Sep 17 00:00:00 2001 From: Christian Ehrhardt Date: Mon, 30 Nov 2009 12:16:47 +0100 Subject: sched: Make tunable scaling style configurable As scaling now takes place on all kind of cpu add/remove events a user that configures values via proc should be able to configure if his set values are still rescaled or kept whatever happens. As the comments state that log2 was just a second guess that worked the interface is not just designed for on/off, but to choose a scaling type. Currently this allows none, log and linear, but more important it allwos us to keep the interface even if someone has an even better idea how to scale the values. Signed-off-by: Christian Ehrhardt Signed-off-by: Peter Zijlstra LKML-Reference: <1259579808-11357-3-git-send-email-ehrhardt@linux.vnet.ibm.com> Signed-off-by: Ingo Molnar --- include/linux/sched.h | 11 ++++++++++- kernel/sched.c | 15 ++++++++++++++- kernel/sched_debug.c | 10 ++++++++++ kernel/sched_fair.c | 13 +++++++++++++ kernel/sysctl.c | 14 ++++++++++++++ 5 files changed, 61 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/include/linux/sched.h b/include/linux/sched.h index 4b1ebd3280c6..ee9f200d12d3 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1902,13 +1902,22 @@ extern unsigned int sysctl_sched_wakeup_granularity; extern unsigned int sysctl_sched_shares_ratelimit; extern unsigned int sysctl_sched_shares_thresh; extern unsigned int sysctl_sched_child_runs_first; + +enum sched_tunable_scaling { + SCHED_TUNABLESCALING_NONE, + SCHED_TUNABLESCALING_LOG, + SCHED_TUNABLESCALING_LINEAR, + SCHED_TUNABLESCALING_END, +}; +extern enum sched_tunable_scaling sysctl_sched_tunable_scaling; + #ifdef CONFIG_SCHED_DEBUG extern unsigned int sysctl_sched_migration_cost; extern unsigned int sysctl_sched_nr_migrate; extern unsigned int sysctl_sched_time_avg; extern unsigned int sysctl_timer_migration; -int sched_nr_latency_handler(struct ctl_table *table, int write, +int sched_proc_update_handler(struct ctl_table *table, int write, void __user *buffer, size_t *length, loff_t *ppos); #endif diff --git a/kernel/sched.c b/kernel/sched.c index b54ecf84b6be..116efed962c6 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -7033,7 +7033,20 @@ cpumask_var_t nohz_cpu_mask; static void update_sysctl(void) { unsigned int cpus = min(num_online_cpus(), 8U); - unsigned int factor = 1 + ilog2(cpus); + unsigned int factor; + + switch (sysctl_sched_tunable_scaling) { + case SCHED_TUNABLESCALING_NONE: + factor = 1; + break; + case SCHED_TUNABLESCALING_LINEAR: + factor = cpus; + break; + case SCHED_TUNABLESCALING_LOG: + default: + factor = 1 + ilog2(cpus); + break; + } #define SET_SYSCTL(name) \ (sysctl_##name = (factor) * normalized_sysctl_##name) diff --git a/kernel/sched_debug.c b/kernel/sched_debug.c index 5fda66615fee..0fc5287fe80f 100644 --- a/kernel/sched_debug.c +++ b/kernel/sched_debug.c @@ -309,6 +309,12 @@ static void print_cpu(struct seq_file *m, int cpu) print_rq(m, rq, cpu); } +static const char *sched_tunable_scaling_names[] = { + "none", + "logaritmic", + "linear" +}; + static int sched_debug_show(struct seq_file *m, void *v) { u64 now = ktime_to_ns(ktime_get()); @@ -334,6 +340,10 @@ static int sched_debug_show(struct seq_file *m, void *v) #undef PN #undef P + SEQ_printf(m, " .%-40s: %d (%s)\n", "sysctl_sched_tunable_scaling", + sysctl_sched_tunable_scaling, + sched_tunable_scaling_names[sysctl_sched_tunable_scaling]); + for_each_online_cpu(cpu) print_cpu(m, cpu); diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c index 71b3458245e5..455106d318a8 100644 --- a/kernel/sched_fair.c +++ b/kernel/sched_fair.c @@ -21,6 +21,7 @@ */ #include +#include /* * Targeted preemption latency for CPU-bound tasks: @@ -37,6 +38,18 @@ unsigned int sysctl_sched_latency = 5000000ULL; unsigned int normalized_sysctl_sched_latency = 5000000ULL; +/* + * The initial- and re-scaling of tunables is configurable + * (default SCHED_TUNABLESCALING_LOG = *(1+ilog(ncpus)) + * + * Options are: + * SCHED_TUNABLESCALING_NONE - unscaled, always *1 + * SCHED_TUNABLESCALING_LOG - scaled logarithmical, *1+ilog(ncpus) + * SCHED_TUNABLESCALING_LINEAR - scaled linear, *ncpus + */ +enum sched_tunable_scaling sysctl_sched_tunable_scaling + = SCHED_TUNABLESCALING_LOG; + /* * Minimal preemption granularity for CPU-bound tasks: * (default: 1 msec * (1 + ilog(ncpus)), units: nanoseconds) diff --git a/kernel/sysctl.c b/kernel/sysctl.c index e5cc53514caa..d10406e5fdfe 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -251,6 +251,8 @@ static int min_sched_granularity_ns = 100000; /* 100 usecs */ static int max_sched_granularity_ns = NSEC_PER_SEC; /* 1 second */ static int min_wakeup_granularity_ns; /* 0 usecs */ static int max_wakeup_granularity_ns = NSEC_PER_SEC; /* 1 second */ +static int min_sched_tunable_scaling = SCHED_TUNABLESCALING_NONE; +static int max_sched_tunable_scaling = SCHED_TUNABLESCALING_END-1; #endif static struct ctl_table kern_table[] = { @@ -304,6 +306,18 @@ static struct ctl_table kern_table[] = { .mode = 0644, .proc_handler = &proc_dointvec, }, + { + .ctl_name = CTL_UNNUMBERED, + .procname = "sched_tunable_scaling", + .data = &sysctl_sched_tunable_scaling, + .maxlen = sizeof(enum sched_tunable_scaling), + .mode = 0644, + .proc_handler = &proc_dointvec_minmax, + .strategy = &sysctl_intvec, + .extra1 = &min_sched_tunable_scaling, + .extra2 = &max_sched_tunable_scaling, + }, + { .ctl_name = CTL_UNNUMBERED, .procname = "sched_shares_thresh", -- cgit v1.2.3-71-gd317 From acb4a848da821a095ae9e4d8b22ae2d9633ba5cd Mon Sep 17 00:00:00 2001 From: Christian Ehrhardt Date: Mon, 30 Nov 2009 12:16:48 +0100 Subject: sched: Update normalized values on user updates via proc The normalized values are also recalculated in case the scaling factor changes. This patch updates the internally used scheduler tuning values that are normalized to one cpu in case a user sets new values via sysfs. Together with patch 2 of this series this allows to let user configured values scale (or not) to cpu add/remove events taking place later. Signed-off-by: Christian Ehrhardt Signed-off-by: Peter Zijlstra LKML-Reference: <1259579808-11357-4-git-send-email-ehrhardt@linux.vnet.ibm.com> [ v2: fix warning ] Signed-off-by: Ingo Molnar --- kernel/sched.c | 12 ++++++++++-- kernel/sched_fair.c | 11 ++++++++++- kernel/sysctl.c | 14 +++++++++----- 3 files changed, 29 insertions(+), 8 deletions(-) (limited to 'kernel') diff --git a/kernel/sched.c b/kernel/sched.c index 116efed962c6..0a60e8e9b094 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -1816,6 +1816,7 @@ static void cfs_rq_set_shares(struct cfs_rq *cfs_rq, unsigned long shares) static void calc_load_account_active(struct rq *this_rq); static void update_sysctl(void); +static int get_update_sysctl_factor(void); static inline void __set_task_cpu(struct task_struct *p, unsigned int cpu) { @@ -7030,9 +7031,9 @@ cpumask_var_t nohz_cpu_mask; * * This idea comes from the SD scheduler of Con Kolivas: */ -static void update_sysctl(void) +static int get_update_sysctl_factor(void) { - unsigned int cpus = min(num_online_cpus(), 8U); + unsigned int cpus = min(num_online_cpus(), 8); unsigned int factor; switch (sysctl_sched_tunable_scaling) { @@ -7048,6 +7049,13 @@ static void update_sysctl(void) break; } + return factor; +} + +static void update_sysctl(void) +{ + unsigned int factor = get_update_sysctl_factor(); + #define SET_SYSCTL(name) \ (sysctl_##name = (factor) * normalized_sysctl_##name) SET_SYSCTL(sched_min_granularity); diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c index 455106d318a8..804a411838f1 100644 --- a/kernel/sched_fair.c +++ b/kernel/sched_fair.c @@ -399,11 +399,12 @@ static struct sched_entity *__pick_last_entity(struct cfs_rq *cfs_rq) */ #ifdef CONFIG_SCHED_DEBUG -int sched_nr_latency_handler(struct ctl_table *table, int write, +int sched_proc_update_handler(struct ctl_table *table, int write, void __user *buffer, size_t *lenp, loff_t *ppos) { int ret = proc_dointvec_minmax(table, write, buffer, lenp, ppos); + int factor = get_update_sysctl_factor(); if (ret || !write) return ret; @@ -411,6 +412,14 @@ int sched_nr_latency_handler(struct ctl_table *table, int write, sched_nr_latency = DIV_ROUND_UP(sysctl_sched_latency, sysctl_sched_min_granularity); +#define WRT_SYSCTL(name) \ + (normalized_sysctl_##name = sysctl_##name / (factor)) + WRT_SYSCTL(sched_min_granularity); + WRT_SYSCTL(sched_latency); + WRT_SYSCTL(sched_wakeup_granularity); + WRT_SYSCTL(sched_shares_ratelimit); +#undef WRT_SYSCTL + return 0; } #endif diff --git a/kernel/sysctl.c b/kernel/sysctl.c index d10406e5fdfe..b9e5a45f1e28 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -253,6 +253,8 @@ static int min_wakeup_granularity_ns; /* 0 usecs */ static int max_wakeup_granularity_ns = NSEC_PER_SEC; /* 1 second */ static int min_sched_tunable_scaling = SCHED_TUNABLESCALING_NONE; static int max_sched_tunable_scaling = SCHED_TUNABLESCALING_END-1; +static int min_sched_shares_ratelimit = 100000; /* 100 usec */ +static int max_sched_shares_ratelimit = NSEC_PER_SEC; /* 1 second */ #endif static struct ctl_table kern_table[] = { @@ -271,7 +273,7 @@ static struct ctl_table kern_table[] = { .data = &sysctl_sched_min_granularity, .maxlen = sizeof(unsigned int), .mode = 0644, - .proc_handler = &sched_nr_latency_handler, + .proc_handler = &sched_proc_update_handler, .strategy = &sysctl_intvec, .extra1 = &min_sched_granularity_ns, .extra2 = &max_sched_granularity_ns, @@ -282,7 +284,7 @@ static struct ctl_table kern_table[] = { .data = &sysctl_sched_latency, .maxlen = sizeof(unsigned int), .mode = 0644, - .proc_handler = &sched_nr_latency_handler, + .proc_handler = &sched_proc_update_handler, .strategy = &sysctl_intvec, .extra1 = &min_sched_granularity_ns, .extra2 = &max_sched_granularity_ns, @@ -293,7 +295,7 @@ static struct ctl_table kern_table[] = { .data = &sysctl_sched_wakeup_granularity, .maxlen = sizeof(unsigned int), .mode = 0644, - .proc_handler = &proc_dointvec_minmax, + .proc_handler = &sched_proc_update_handler, .strategy = &sysctl_intvec, .extra1 = &min_wakeup_granularity_ns, .extra2 = &max_wakeup_granularity_ns, @@ -304,7 +306,9 @@ static struct ctl_table kern_table[] = { .data = &sysctl_sched_shares_ratelimit, .maxlen = sizeof(unsigned int), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = &sched_proc_update_handler, + .extra1 = &min_sched_shares_ratelimit, + .extra2 = &max_sched_shares_ratelimit, }, { .ctl_name = CTL_UNNUMBERED, @@ -312,7 +316,7 @@ static struct ctl_table kern_table[] = { .data = &sysctl_sched_tunable_scaling, .maxlen = sizeof(enum sched_tunable_scaling), .mode = 0644, - .proc_handler = &proc_dointvec_minmax, + .proc_handler = &sched_proc_update_handler, .strategy = &sysctl_intvec, .extra1 = &min_sched_tunable_scaling, .extra2 = &max_sched_tunable_scaling, -- cgit v1.2.3-71-gd317 From 822a6961112f0c9101d3359d8524604c3309ee6c Mon Sep 17 00:00:00 2001 From: Frederic Weisbecker Date: Tue, 8 Dec 2009 10:00:04 +0100 Subject: tracing/kprobes: Fix field creation's bad error handling When we define the common event fields in kprobe, we invert the error handling and return immediately in case of success. Then we omit to define specific kprobes fields (ip and nargs), and specific kretprobes fields (func, ret_ip, nargs). And we only define them when we fail to create common fields. The most visible consequence is that we can't create filter for k(ret)probes specific fields. This patch re-invert the success/error handling to fix it. Reported-by: Lai Jiangshan Signed-off-by: Frederic Weisbecker Acked-by: Masami Hiramatsu Cc: Steven Rostedt Cc: Li Zefan LKML-Reference: <1260263815-5167-1-git-send-regression-fweisbec@gmail.com> Signed-off-by: Ingo Molnar --- kernel/trace/trace_kprobe.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c index bf05fb49a6f3..b52d397e57eb 100644 --- a/kernel/trace/trace_kprobe.c +++ b/kernel/trace/trace_kprobe.c @@ -1133,7 +1133,7 @@ static int kprobe_event_define_fields(struct ftrace_event_call *event_call) struct trace_probe *tp = (struct trace_probe *)event_call->data; ret = trace_define_common_fields(event_call); - if (!ret) + if (ret) return ret; DEFINE_FIELD(unsigned long, ip, FIELD_STRING_IP, 0); @@ -1151,7 +1151,7 @@ static int kretprobe_event_define_fields(struct ftrace_event_call *event_call) struct trace_probe *tp = (struct trace_probe *)event_call->data; ret = trace_define_common_fields(event_call); - if (!ret) + if (ret) return ret; DEFINE_FIELD(unsigned long, func, FIELD_STRING_FUNC, 0); -- cgit v1.2.3-71-gd317 From 29bf4a5e3fed3dde3eb629a0cb1762c1e9217458 Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Wed, 9 Dec 2009 12:37:43 -0500 Subject: tracing: Only call pipe_close if pipe_close is defined This fixes a cut and paste error that had pipe_close get called if pipe_open was defined (not pipe_close). Reported-by: Kosaki Motohiro LKML-Reference: <20091209153204.F4CD.A69D9226@jp.fujitsu.com> Signed-off-by: Steven Rostedt --- kernel/trace/trace.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index f804b407d438..dc937e1baa91 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -2899,7 +2899,7 @@ static int tracing_release_pipe(struct inode *inode, struct file *file) cpumask_clear_cpu(iter->cpu_file, tracing_reader_cpumask); - if (iter->trace->pipe_open) + if (iter->trace->pipe_close) iter->trace->pipe_close(iter); mutex_unlock(&trace_types_lock); -- cgit v1.2.3-71-gd317 From a63ce5b306855bccdacba95c03bfc293316c8ae3 Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Mon, 7 Dec 2009 09:11:39 -0500 Subject: tracing: Buffer the output of seq_file in case of filled buffer If the seq_read fills the buffer it will call s_start again on the next itertation with the same position. This causes a problem with the function_graph tracer because it consumes the iteration in order to determine leaf functions. What happens is that the iterator stores the entry, and the function graph plugin will look at the next entry. If that next entry is a return of the same function and task, then the function is a leaf and the function_graph plugin calls ring_buffer_read which moves the ring buffer iterator forward (the trace iterator still points to the function start entry). The copying of the trace_seq to the seq_file buffer will fail if the seq_file buffer is full. The seq_read will not show this entry. The next read by userspace will cause seq_read to again call s_start which will reuse the trace iterator entry (the function start entry). But the function return entry was already consumed. The function graph plugin will think that this entry is a nested function and not a leaf. To solve this, the trace code now checks the return status of the seq_printf (trace_print_seq). If the writing to the seq_file buffer fails, we set a flag in the iterator (leftover) and we do not reset the trace_seq buffer. On the next call to s_start, we check the leftover flag, and if it is set, we just reuse the trace_seq buffer and do not call into the plugin print functions. Before this patch: 2) | fput() { 2) | __fput() { 2) 0.550 us | inotify_inode_queue_event(); 2) | __fsnotify_parent() { 2) 0.540 us | inotify_dentry_parent_queue_event(); After the patch: 2) | fput() { 2) | __fput() { 2) 0.550 us | inotify_inode_queue_event(); 2) 0.548 us | __fsnotify_parent(); 2) 0.540 us | inotify_dentry_parent_queue_event(); [ Updated the patch to fix a missing return 0 from the trace_print_seq() stub when CONFIG_TRACING is disabled. Reported-by: Ingo Molnar ] Reported-by: Jiri Olsa Cc: Frederic Weisbecker Signed-off-by: Steven Rostedt --- include/linux/ftrace_event.h | 1 + include/linux/trace_seq.h | 5 +++-- kernel/trace/trace.c | 35 ++++++++++++++++++++++++++++++++--- kernel/trace/trace_output.c | 14 +++++++++++--- 4 files changed, 47 insertions(+), 8 deletions(-) (limited to 'kernel') diff --git a/include/linux/ftrace_event.h b/include/linux/ftrace_event.h index 47bbdf9c38d0..38f8d6553831 100644 --- a/include/linux/ftrace_event.h +++ b/include/linux/ftrace_event.h @@ -57,6 +57,7 @@ struct trace_iterator { /* The below is zeroed out in pipe_read */ struct trace_seq seq; struct trace_entry *ent; + int leftover; int cpu; u64 ts; diff --git a/include/linux/trace_seq.h b/include/linux/trace_seq.h index 09077f6ed128..fad6857bc0f3 100644 --- a/include/linux/trace_seq.h +++ b/include/linux/trace_seq.h @@ -33,7 +33,7 @@ extern int trace_seq_vprintf(struct trace_seq *s, const char *fmt, va_list args) __attribute__ ((format (printf, 2, 0))); extern int trace_seq_bprintf(struct trace_seq *s, const char *fmt, const u32 *binary); -extern void trace_print_seq(struct seq_file *m, struct trace_seq *s); +extern int trace_print_seq(struct seq_file *m, struct trace_seq *s); extern ssize_t trace_seq_to_user(struct trace_seq *s, char __user *ubuf, size_t cnt); extern int trace_seq_puts(struct trace_seq *s, const char *str); @@ -55,8 +55,9 @@ trace_seq_bprintf(struct trace_seq *s, const char *fmt, const u32 *binary) return 0; } -static inline void trace_print_seq(struct seq_file *m, struct trace_seq *s) +static inline int trace_print_seq(struct seq_file *m, struct trace_seq *s) { + return 0; } static inline ssize_t trace_seq_to_user(struct trace_seq *s, char __user *ubuf, size_t cnt) diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index dc937e1baa91..484114d70743 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -1516,6 +1516,8 @@ static void *s_next(struct seq_file *m, void *v, loff_t *pos) int i = (int)*pos; void *ent; + WARN_ON_ONCE(iter->leftover); + (*pos)++; /* can't go backwards */ @@ -1614,8 +1616,16 @@ static void *s_start(struct seq_file *m, loff_t *pos) ; } else { - l = *pos - 1; - p = s_next(m, p, &l); + /* + * If we overflowed the seq_file before, then we want + * to just reuse the trace_seq buffer again. + */ + if (iter->leftover) + p = iter; + else { + l = *pos - 1; + p = s_next(m, p, &l); + } } trace_event_read_lock(); @@ -1923,6 +1933,7 @@ static enum print_line_t print_trace_line(struct trace_iterator *iter) static int s_show(struct seq_file *m, void *v) { struct trace_iterator *iter = v; + int ret; if (iter->ent == NULL) { if (iter->tr) { @@ -1942,9 +1953,27 @@ static int s_show(struct seq_file *m, void *v) if (!(trace_flags & TRACE_ITER_VERBOSE)) print_func_help_header(m); } + } else if (iter->leftover) { + /* + * If we filled the seq_file buffer earlier, we + * want to just show it now. + */ + ret = trace_print_seq(m, &iter->seq); + + /* ret should this time be zero, but you never know */ + iter->leftover = ret; + } else { print_trace_line(iter); - trace_print_seq(m, &iter->seq); + ret = trace_print_seq(m, &iter->seq); + /* + * If we overflow the seq_file buffer, then it will + * ask us for this data again at start up. + * Use that instead. + * ret is 0 if seq_file write succeeded. + * -1 otherwise. + */ + iter->leftover = ret; } return 0; diff --git a/kernel/trace/trace_output.c b/kernel/trace/trace_output.c index b6c12c6a1bcd..e5cf90fef34e 100644 --- a/kernel/trace/trace_output.c +++ b/kernel/trace/trace_output.c @@ -23,13 +23,21 @@ static struct hlist_head event_hash[EVENT_HASHSIZE] __read_mostly; static int next_event_type = __TRACE_LAST_TYPE + 1; -void trace_print_seq(struct seq_file *m, struct trace_seq *s) +int trace_print_seq(struct seq_file *m, struct trace_seq *s) { int len = s->len >= PAGE_SIZE ? PAGE_SIZE - 1 : s->len; + int ret; + + ret = seq_write(m, s->buffer, len); - seq_write(m, s->buffer, len); + /* + * Only reset this buffer if we successfully wrote to the + * seq_file buffer. + */ + if (!ret) + trace_seq_init(s); - trace_seq_init(s); + return ret; } enum print_line_t trace_print_bprintk_msg_only(struct trace_iterator *iter) -- cgit v1.2.3-71-gd317 From d184b31c0e403580aafb3f8955ecc185a3d04801 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Wed, 25 Nov 2009 16:10:14 +0100 Subject: tracing: Add full state to trace_seq The trace_seq buffer might fill up, and right now one needs to check the return value of each printf into the buffer to check for that. Instead, have the buffer keep track of whether it is full or not, and reject more input if it is full or would have overflowed with an input that wasn't added. Cc: Lai Jiangshan Signed-off-by: Johannes Berg Signed-off-by: Steven Rostedt --- include/linux/trace_seq.h | 2 ++ kernel/trace/trace_output.c | 61 +++++++++++++++++++++++++++++++++++++-------- 2 files changed, 52 insertions(+), 11 deletions(-) (limited to 'kernel') diff --git a/include/linux/trace_seq.h b/include/linux/trace_seq.h index fad6857bc0f3..5cf397ceb726 100644 --- a/include/linux/trace_seq.h +++ b/include/linux/trace_seq.h @@ -14,6 +14,7 @@ struct trace_seq { unsigned char buffer[PAGE_SIZE]; unsigned int len; unsigned int readpos; + int full; }; static inline void @@ -21,6 +22,7 @@ trace_seq_init(struct trace_seq *s) { s->len = 0; s->readpos = 0; + s->full = 0; } /* diff --git a/kernel/trace/trace_output.c b/kernel/trace/trace_output.c index e5cf90fef34e..8e46b3323cdc 100644 --- a/kernel/trace/trace_output.c +++ b/kernel/trace/trace_output.c @@ -93,7 +93,7 @@ trace_seq_printf(struct trace_seq *s, const char *fmt, ...) va_list ap; int ret; - if (!len) + if (s->full || !len) return 0; va_start(ap, fmt); @@ -101,8 +101,10 @@ trace_seq_printf(struct trace_seq *s, const char *fmt, ...) va_end(ap); /* If we can't write it all, don't bother writing anything */ - if (ret >= len) + if (ret >= len) { + s->full = 1; return 0; + } s->len += ret; @@ -127,14 +129,16 @@ trace_seq_vprintf(struct trace_seq *s, const char *fmt, va_list args) int len = (PAGE_SIZE - 1) - s->len; int ret; - if (!len) + if (s->full || !len) return 0; ret = vsnprintf(s->buffer + s->len, len, fmt, args); /* If we can't write it all, don't bother writing anything */ - if (ret >= len) + if (ret >= len) { + s->full = 1; return 0; + } s->len += ret; @@ -147,14 +151,16 @@ int trace_seq_bprintf(struct trace_seq *s, const char *fmt, const u32 *binary) int len = (PAGE_SIZE - 1) - s->len; int ret; - if (!len) + if (s->full || !len) return 0; ret = bstr_printf(s->buffer + s->len, len, fmt, binary); /* If we can't write it all, don't bother writing anything */ - if (ret >= len) + if (ret >= len) { + s->full = 1; return 0; + } s->len += ret; @@ -175,9 +181,14 @@ int trace_seq_puts(struct trace_seq *s, const char *str) { int len = strlen(str); - if (len > ((PAGE_SIZE - 1) - s->len)) + if (s->full) return 0; + if (len > ((PAGE_SIZE - 1) - s->len)) { + s->full = 1; + return 0; + } + memcpy(s->buffer + s->len, str, len); s->len += len; @@ -186,9 +197,14 @@ int trace_seq_puts(struct trace_seq *s, const char *str) int trace_seq_putc(struct trace_seq *s, unsigned char c) { - if (s->len >= (PAGE_SIZE - 1)) + if (s->full) return 0; + if (s->len >= (PAGE_SIZE - 1)) { + s->full = 1; + return 0; + } + s->buffer[s->len++] = c; return 1; @@ -196,9 +212,14 @@ int trace_seq_putc(struct trace_seq *s, unsigned char c) int trace_seq_putmem(struct trace_seq *s, const void *mem, size_t len) { - if (len > ((PAGE_SIZE - 1) - s->len)) + if (s->full) return 0; + if (len > ((PAGE_SIZE - 1) - s->len)) { + s->full = 1; + return 0; + } + memcpy(s->buffer + s->len, mem, len); s->len += len; @@ -211,6 +232,9 @@ int trace_seq_putmem_hex(struct trace_seq *s, const void *mem, size_t len) const unsigned char *data = mem; int i, j; + if (s->full) + return 0; + #ifdef __BIG_ENDIAN for (i = 0, j = 0; i < len; i++) { #else @@ -228,8 +252,13 @@ void *trace_seq_reserve(struct trace_seq *s, size_t len) { void *ret; - if (len > ((PAGE_SIZE - 1) - s->len)) + if (s->full) + return 0; + + if (len > ((PAGE_SIZE - 1) - s->len)) { + s->full = 1; return NULL; + } ret = s->buffer + s->len; s->len += len; @@ -241,8 +270,14 @@ int trace_seq_path(struct trace_seq *s, struct path *path) { unsigned char *p; - if (s->len >= (PAGE_SIZE - 1)) + if (s->full) return 0; + + if (s->len >= (PAGE_SIZE - 1)) { + s->full = 1; + return 0; + } + p = d_path(path, s->buffer + s->len, PAGE_SIZE - s->len); if (!IS_ERR(p)) { p = mangle_path(s->buffer + s->len, p, "\n"); @@ -255,6 +290,7 @@ int trace_seq_path(struct trace_seq *s, struct path *path) return 1; } + s->full = 1; return 0; } @@ -381,6 +417,9 @@ int seq_print_user_ip(struct trace_seq *s, struct mm_struct *mm, unsigned long vmstart = 0; int ret = 1; + if (s->full) + return 0; + if (mm) { const struct vm_area_struct *vma; -- cgit v1.2.3-71-gd317 From be1eca39319689aed7d3aedb9c3bece9469fe10f Mon Sep 17 00:00:00 2001 From: Jiri Olsa Date: Tue, 24 Nov 2009 13:57:38 +0100 Subject: tracing: Fix function graph trace_pipe to properly display failed entries There is a case where the graph tracer might get confused and omits displaying of a single record. This applies mostly with the trace_pipe since it is unlikely that the trace_seq buffer will overflow with the trace file. As the function_graph tracer goes through the trace entries keeping a pointer to the current record: current -> func1 ENTRY func2 ENTRY func2 RETURN func1 RETURN When an function ENTRY is encountered, it moves the pointer to the next entry to check if the function is a nested or leaf function. func1 ENTRY current -> func2 ENTRY func2 RETURN func1 RETURN If the rest of the writing of the function fills the trace_seq buffer, then the trace_pipe read will ignore this entry. The next read will Now start at the current location, but the first entry (func1) will be discarded. This patch keeps a copy of the current entry in the iterator private storage and will keep track of when the trace_seq buffer fills. When the trace_seq buffer fills, it will reuse the copy of the entry in the next iteration. [ This patch has been largely modified by Steven Rostedt in order to clean it up and simplify it. The original idea and concept was from Jirka and for that, this patch will go under his name to give him the credit he deserves. But because this was modify by Steven Rostedt anything wrong with the patch should be blamed on Steven. ] Signed-off-by: Jiri Olsa Cc: Frederic Weisbecker LKML-Reference: <1259067458-27143-1-git-send-email-jolsa@redhat.com> Signed-off-by: Steven Rostedt --- kernel/trace/trace_functions_graph.c | 165 +++++++++++++++++++++++++++-------- 1 file changed, 131 insertions(+), 34 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/trace_functions_graph.c b/kernel/trace/trace_functions_graph.c index 45e6c01b2e4d..a43d009c561a 100644 --- a/kernel/trace/trace_functions_graph.c +++ b/kernel/trace/trace_functions_graph.c @@ -14,9 +14,20 @@ #include "trace.h" #include "trace_output.h" -struct fgraph_data { +struct fgraph_cpu_data { pid_t last_pid; int depth; + int ignore; +}; + +struct fgraph_data { + struct fgraph_cpu_data *cpu_data; + + /* Place to preserve last processed entry. */ + struct ftrace_graph_ent_entry ent; + struct ftrace_graph_ret_entry ret; + int failed; + int cpu; }; #define TRACE_GRAPH_INDENT 2 @@ -384,7 +395,7 @@ verif_pid(struct trace_seq *s, pid_t pid, int cpu, struct fgraph_data *data) if (!data) return TRACE_TYPE_HANDLED; - last_pid = &(per_cpu_ptr(data, cpu)->last_pid); + last_pid = &(per_cpu_ptr(data->cpu_data, cpu)->last_pid); if (*last_pid == pid) return TRACE_TYPE_HANDLED; @@ -435,26 +446,49 @@ static struct ftrace_graph_ret_entry * get_return_for_leaf(struct trace_iterator *iter, struct ftrace_graph_ent_entry *curr) { - struct ring_buffer_iter *ring_iter; + struct fgraph_data *data = iter->private; + struct ring_buffer_iter *ring_iter = NULL; struct ring_buffer_event *event; struct ftrace_graph_ret_entry *next; - ring_iter = iter->buffer_iter[iter->cpu]; + /* + * If the previous output failed to write to the seq buffer, + * then we just reuse the data from before. + */ + if (data && data->failed) { + curr = &data->ent; + next = &data->ret; + } else { - /* First peek to compare current entry and the next one */ - if (ring_iter) - event = ring_buffer_iter_peek(ring_iter, NULL); - else { - /* We need to consume the current entry to see the next one */ - ring_buffer_consume(iter->tr->buffer, iter->cpu, NULL); - event = ring_buffer_peek(iter->tr->buffer, iter->cpu, - NULL); - } + ring_iter = iter->buffer_iter[iter->cpu]; + + /* First peek to compare current entry and the next one */ + if (ring_iter) + event = ring_buffer_iter_peek(ring_iter, NULL); + else { + /* + * We need to consume the current entry to see + * the next one. + */ + ring_buffer_consume(iter->tr->buffer, iter->cpu, NULL); + event = ring_buffer_peek(iter->tr->buffer, iter->cpu, + NULL); + } - if (!event) - return NULL; + if (!event) + return NULL; + + next = ring_buffer_event_data(event); - next = ring_buffer_event_data(event); + if (data) { + /* + * Save current and next entries for later reference + * if the output fails. + */ + data->ent = *curr; + data->ret = *next; + } + } if (next->ent.type != TRACE_GRAPH_RET) return NULL; @@ -640,7 +674,7 @@ print_graph_entry_leaf(struct trace_iterator *iter, if (data) { int cpu = iter->cpu; - int *depth = &(per_cpu_ptr(data, cpu)->depth); + int *depth = &(per_cpu_ptr(data->cpu_data, cpu)->depth); /* * Comments display at + 1 to depth. Since @@ -688,7 +722,7 @@ print_graph_entry_nested(struct trace_iterator *iter, if (data) { int cpu = iter->cpu; - int *depth = &(per_cpu_ptr(data, cpu)->depth); + int *depth = &(per_cpu_ptr(data->cpu_data, cpu)->depth); *depth = call->depth; } @@ -782,19 +816,34 @@ static enum print_line_t print_graph_entry(struct ftrace_graph_ent_entry *field, struct trace_seq *s, struct trace_iterator *iter) { - int cpu = iter->cpu; + struct fgraph_data *data = iter->private; struct ftrace_graph_ent *call = &field->graph_ent; struct ftrace_graph_ret_entry *leaf_ret; + static enum print_line_t ret; + int cpu = iter->cpu; if (print_graph_prologue(iter, s, TRACE_GRAPH_ENT, call->func)) return TRACE_TYPE_PARTIAL_LINE; leaf_ret = get_return_for_leaf(iter, field); if (leaf_ret) - return print_graph_entry_leaf(iter, field, leaf_ret, s); + ret = print_graph_entry_leaf(iter, field, leaf_ret, s); else - return print_graph_entry_nested(iter, field, s, cpu); + ret = print_graph_entry_nested(iter, field, s, cpu); + if (data) { + /* + * If we failed to write our output, then we need to make + * note of it. Because we already consumed our entry. + */ + if (s->full) { + data->failed = 1; + data->cpu = cpu; + } else + data->failed = 0; + } + + return ret; } static enum print_line_t @@ -810,7 +859,7 @@ print_graph_return(struct ftrace_graph_ret *trace, struct trace_seq *s, if (data) { int cpu = iter->cpu; - int *depth = &(per_cpu_ptr(data, cpu)->depth); + int *depth = &(per_cpu_ptr(data->cpu_data, cpu)->depth); /* * Comments display at + 1 to depth. This is the @@ -873,7 +922,7 @@ print_graph_comment(struct trace_seq *s, struct trace_entry *ent, int i; if (data) - depth = per_cpu_ptr(data, iter->cpu)->depth; + depth = per_cpu_ptr(data->cpu_data, iter->cpu)->depth; if (print_graph_prologue(iter, s, 0, 0)) return TRACE_TYPE_PARTIAL_LINE; @@ -941,8 +990,33 @@ print_graph_comment(struct trace_seq *s, struct trace_entry *ent, enum print_line_t print_graph_function(struct trace_iterator *iter) { + struct ftrace_graph_ent_entry *field; + struct fgraph_data *data = iter->private; struct trace_entry *entry = iter->ent; struct trace_seq *s = &iter->seq; + int cpu = iter->cpu; + int ret; + + if (data && per_cpu_ptr(data->cpu_data, cpu)->ignore) { + per_cpu_ptr(data->cpu_data, cpu)->ignore = 0; + return TRACE_TYPE_HANDLED; + } + + /* + * If the last output failed, there's a possibility we need + * to print out the missing entry which would never go out. + */ + if (data && data->failed) { + field = &data->ent; + iter->cpu = data->cpu; + ret = print_graph_entry(field, s, iter); + if (ret == TRACE_TYPE_HANDLED && iter->cpu != cpu) { + per_cpu_ptr(data->cpu_data, iter->cpu)->ignore = 1; + ret = TRACE_TYPE_NO_CONSUME; + } + iter->cpu = cpu; + return ret; + } switch (entry->type) { case TRACE_GRAPH_ENT: { @@ -952,7 +1026,7 @@ print_graph_function(struct trace_iterator *iter) * sizeof(struct ftrace_graph_ent_entry) is very small, * it can be safely saved at the stack. */ - struct ftrace_graph_ent_entry *field, saved; + struct ftrace_graph_ent_entry saved; trace_assign_type(field, entry); saved = *field; return print_graph_entry(&saved, s, iter); @@ -1030,31 +1104,54 @@ static void print_graph_headers(struct seq_file *s) static void graph_trace_open(struct trace_iterator *iter) { /* pid and depth on the last trace processed */ - struct fgraph_data *data = alloc_percpu(struct fgraph_data); + struct fgraph_data *data; int cpu; + iter->private = NULL; + + data = kzalloc(sizeof(*data), GFP_KERNEL); if (!data) - pr_warning("function graph tracer: not enough memory\n"); - else - for_each_possible_cpu(cpu) { - pid_t *pid = &(per_cpu_ptr(data, cpu)->last_pid); - int *depth = &(per_cpu_ptr(data, cpu)->depth); - *pid = -1; - *depth = 0; - } + goto out_err; + + data->cpu_data = alloc_percpu(struct fgraph_cpu_data); + if (!data->cpu_data) + goto out_err_free; + + for_each_possible_cpu(cpu) { + pid_t *pid = &(per_cpu_ptr(data->cpu_data, cpu)->last_pid); + int *depth = &(per_cpu_ptr(data->cpu_data, cpu)->depth); + int *ignore = &(per_cpu_ptr(data->cpu_data, cpu)->ignore); + *pid = -1; + *depth = 0; + *ignore = 0; + } iter->private = data; + + return; + + out_err_free: + kfree(data); + out_err: + pr_warning("function graph tracer: not enough memory\n"); } static void graph_trace_close(struct trace_iterator *iter) { - free_percpu(iter->private); + struct fgraph_data *data = iter->private; + + if (data) { + free_percpu(data->cpu_data); + kfree(data); + } } static struct tracer graph_trace __read_mostly = { .name = "function_graph", .open = graph_trace_open, + .pipe_open = graph_trace_open, .close = graph_trace_close, + .pipe_close = graph_trace_close, .wait_pipe = poll_wait_pipe, .init = graph_trace_init, .reset = graph_trace_reset, -- cgit v1.2.3-71-gd317 From f2942487ffb0c0a80b2312f667ea30dd55a24bb0 Mon Sep 17 00:00:00 2001 From: Carsten Emde Date: Sun, 6 Dec 2009 14:02:44 +0100 Subject: tracing: Remove comparing of NULL to va_list in trace_array_vprintk() Olof Johansson stated the following: Comparing a va_list with NULL is bogus. It's supposed to be treated like an opaque type and only be manipulated with va_* accessors. Olof noticed that this code broke the ARM builds: kernel/trace/trace.c: In function 'trace_array_vprintk': kernel/trace/trace.c:1364: error: invalid operands to binary == (have 'va_list' and 'void *') kernel/trace/trace.c: In function 'tracing_mark_write': kernel/trace/trace.c:3349: error: incompatible type for argument 3 of 'trace_vprintk' This patch partly reverts c13d2f7c3231e873f30db92b96c8caa48f100f33 and re-installs the original mark_printk() mechanism. Reported-by: Olof Johansson Signed-off-by: Carsten Emde LKML-Reference: <4B1BAB74.104@osadl.org> Signed-off-by: Steven Rostedt --- kernel/trace/trace.c | 18 ++++++++++++------ 1 file changed, 12 insertions(+), 6 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 484114d70743..88bd9ae2a9ed 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -1361,11 +1361,7 @@ int trace_array_vprintk(struct trace_array *tr, pause_graph_tracing(); raw_local_irq_save(irq_flags); __raw_spin_lock(&trace_buf_lock); - if (args == NULL) { - strncpy(trace_buf, fmt, TRACE_BUF_SIZE); - len = strlen(trace_buf); - } else - len = vsnprintf(trace_buf, TRACE_BUF_SIZE, fmt, args); + len = vsnprintf(trace_buf, TRACE_BUF_SIZE, fmt, args); size = sizeof(*entry) + len + 1; buffer = tr->buffer; @@ -3353,6 +3349,16 @@ tracing_entries_write(struct file *filp, const char __user *ubuf, return cnt; } +static int mark_printk(const char *fmt, ...) +{ + int ret; + va_list args; + va_start(args, fmt); + ret = trace_vprintk(0, fmt, args); + va_end(args); + return ret; +} + static ssize_t tracing_mark_write(struct file *filp, const char __user *ubuf, size_t cnt, loff_t *fpos) @@ -3379,7 +3385,7 @@ tracing_mark_write(struct file *filp, const char __user *ubuf, } else buf[cnt] = '\0'; - cnt = trace_vprintk(0, buf, NULL); + cnt = mark_printk("%s", buf); kfree(buf); *fpos += cnt; -- cgit v1.2.3-71-gd317 From 21140f4d3387aa2213f1deea0128df1dbf924379 Mon Sep 17 00:00:00 2001 From: Xiao Guangrong Date: Thu, 10 Dec 2009 14:00:51 +0800 Subject: perf_event: Fix perf_swevent_hrtimer() variable initialization fix: [] ? printk+0x1d/0x24 [] ? perf_prepare_sample+0x269/0x280 [] warn_slowpath_common+0x71/0xd0 [] ? perf_prepare_sample+0x269/0x280 [] warn_slowpath_null+0x1a/0x20 [] perf_prepare_sample+0x269/0x280 [] ? cpu_clock+0x53/0x90 [] __perf_event_overflow+0x2a8/0x300 [] perf_event_overflow+0x1b/0x30 [] perf_swevent_hrtimer+0x7f/0x120 This is because 'data.raw' variable not initialize. Signed-off-by: Xiao Guangrong Acked-by: Peter Zijlstra Cc: Frederic Weisbecker Cc: Paul Mackerras LKML-Reference: <4B208E93.1010801@cn.fujitsu.com> Signed-off-by: Ingo Molnar --- kernel/perf_event.c | 1 + 1 file changed, 1 insertion(+) (limited to 'kernel') diff --git a/kernel/perf_event.c b/kernel/perf_event.c index 94e1b28333ae..3a5d6c4786bb 100644 --- a/kernel/perf_event.c +++ b/kernel/perf_event.c @@ -4010,6 +4010,7 @@ static enum hrtimer_restart perf_swevent_hrtimer(struct hrtimer *hrtimer) event->pmu->read(event); data.addr = 0; + data.raw = NULL; data.period = event->hw.last_period; regs = get_irq_regs(); /* -- cgit v1.2.3-71-gd317 From ea5b41f9d595be354f7a50e56b28c2d72e6e88a5 Mon Sep 17 00:00:00 2001 From: "Luck, Tony" Date: Wed, 9 Dec 2009 14:29:36 -0800 Subject: lockdep: Avoid out of bounds array reference in save_trace() ia64 found this the hard way (because we currently have a stub for save_stack_trace() that does nothing). But it would be a good idea to be cautious in case a real save_stack_trace() bailed out with an error before it set trace->nr_entries. Signed-off-by: Tony Luck Acked-by: Peter Zijlstra Cc: luming.yu@intel.com LKML-Reference: <4b2024d085302c2a2@agluck-desktop.sc.intel.com> Signed-off-by: Ingo Molnar --- kernel/lockdep.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/lockdep.c b/kernel/lockdep.c index 7a3ae56b3a7f..4f8df01dbe51 100644 --- a/kernel/lockdep.c +++ b/kernel/lockdep.c @@ -386,7 +386,8 @@ static int save_trace(struct stack_trace *trace) * complete trace that maxes out the entries provided will be reported * as incomplete, friggin useless */ - if (trace->entries[trace->nr_entries-1] == ULONG_MAX) + if (trace->nr_entries != 0 && + trace->entries[trace->nr_entries-1] == ULONG_MAX) trace->nr_entries--; trace->max_entries = trace->nr_entries; -- cgit v1.2.3-71-gd317 From 4ca3ef71f54655af98b66e8ff308a47a2a580a53 Mon Sep 17 00:00:00 2001 From: Mike Galbraith Date: Thu, 10 Dec 2009 09:25:53 +0100 Subject: sched: Fix build warning in get_update_sysctl_factor() Signed-off-by: Mike Galbraith Acked-by: Peter Zijlstra Signed-off-by: Ingo Molnar LKML-Reference: --- kernel/sched.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/sched.c b/kernel/sched.c index 0a60e8e9b094..3de3deab8095 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -7033,7 +7033,7 @@ cpumask_var_t nohz_cpu_mask; */ static int get_update_sysctl_factor(void) { - unsigned int cpus = min(num_online_cpus(), 8); + unsigned int cpus = min_t(int, num_online_cpus(), 8); unsigned int factor; switch (sysctl_sched_tunable_scaling) { -- cgit v1.2.3-71-gd317 From 41d2e494937715d3150e5c75d01f0e75ae899337 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Fri, 13 Nov 2009 17:05:44 +0100 Subject: hrtimer: Tune hrtimer_interrupt hang logic The hrtimer_interrupt hang logic adjusts min_delta_ns based on the execution time of the hrtimer callbacks. This is error-prone for virtual machines, where a guest vcpu can be scheduled out during the execution of the callbacks (and the callbacks themselves can do operations that translate to blocking operations in the hypervisor), which in can lead to large min_delta_ns rendering the system unusable. Replace the current heuristics with something more reliable. Allow the interrupt code to try 3 times to catch up with the lost time. If that fails use the total time spent in the interrupt handler to defer the next timer interrupt so the system can catch up with other things which got delayed. Limit that deferment to 100ms. The retry events and the maximum time spent in the interrupt handler are recorded and exposed via /proc/timer_list Inspired by a patch from Marcelo. Reported-by: Michael Tokarev Signed-off-by: Thomas Gleixner Tested-by: Marcelo Tosatti Cc: kvm@vger.kernel.org --- include/linux/hrtimer.h | 13 +++++-- kernel/hrtimer.c | 97 ++++++++++++++++++++++++++++-------------------- kernel/time/timer_list.c | 5 ++- 3 files changed, 70 insertions(+), 45 deletions(-) (limited to 'kernel') diff --git a/include/linux/hrtimer.h b/include/linux/hrtimer.h index 9bace4b9f4fe..040b6796ab4d 100644 --- a/include/linux/hrtimer.h +++ b/include/linux/hrtimer.h @@ -162,10 +162,11 @@ struct hrtimer_clock_base { * @expires_next: absolute time of the next event which was scheduled * via clock_set_next_event() * @hres_active: State of high resolution mode - * @check_clocks: Indictator, when set evaluate time source and clock - * event devices whether high resolution mode can be - * activated. - * @nr_events: Total number of timer interrupt events + * @hang_detected: The last hrtimer interrupt detected a hang + * @nr_events: Total number of hrtimer interrupt events + * @nr_retries: Total number of hrtimer interrupt retries + * @nr_hangs: Total number of hrtimer interrupt hangs + * @max_hang_time: Maximum time spent in hrtimer_interrupt */ struct hrtimer_cpu_base { spinlock_t lock; @@ -173,7 +174,11 @@ struct hrtimer_cpu_base { #ifdef CONFIG_HIGH_RES_TIMERS ktime_t expires_next; int hres_active; + int hang_detected; unsigned long nr_events; + unsigned long nr_retries; + unsigned long nr_hangs; + ktime_t max_hang_time; #endif }; diff --git a/kernel/hrtimer.c b/kernel/hrtimer.c index ede527708123..931a4d99bc55 100644 --- a/kernel/hrtimer.c +++ b/kernel/hrtimer.c @@ -557,7 +557,7 @@ hrtimer_force_reprogram(struct hrtimer_cpu_base *cpu_base, int skip_equal) static int hrtimer_reprogram(struct hrtimer *timer, struct hrtimer_clock_base *base) { - ktime_t *expires_next = &__get_cpu_var(hrtimer_bases).expires_next; + struct hrtimer_cpu_base *cpu_base = &__get_cpu_var(hrtimer_bases); ktime_t expires = ktime_sub(hrtimer_get_expires(timer), base->offset); int res; @@ -582,7 +582,16 @@ static int hrtimer_reprogram(struct hrtimer *timer, if (expires.tv64 < 0) return -ETIME; - if (expires.tv64 >= expires_next->tv64) + if (expires.tv64 >= cpu_base->expires_next.tv64) + return 0; + + /* + * If a hang was detected in the last timer interrupt then we + * do not schedule a timer which is earlier than the expiry + * which we enforced in the hang detection. We want the system + * to make progress. + */ + if (cpu_base->hang_detected) return 0; /* @@ -590,7 +599,7 @@ static int hrtimer_reprogram(struct hrtimer *timer, */ res = tick_program_event(expires, 0); if (!IS_ERR_VALUE(res)) - *expires_next = expires; + cpu_base->expires_next = expires; return res; } @@ -1217,30 +1226,6 @@ static void __run_hrtimer(struct hrtimer *timer, ktime_t *now) #ifdef CONFIG_HIGH_RES_TIMERS -static int force_clock_reprogram; - -/* - * After 5 iteration's attempts, we consider that hrtimer_interrupt() - * is hanging, which could happen with something that slows the interrupt - * such as the tracing. Then we force the clock reprogramming for each future - * hrtimer interrupts to avoid infinite loops and use the min_delta_ns - * threshold that we will overwrite. - * The next tick event will be scheduled to 3 times we currently spend on - * hrtimer_interrupt(). This gives a good compromise, the cpus will spend - * 1/4 of their time to process the hrtimer interrupts. This is enough to - * let it running without serious starvation. - */ - -static inline void -hrtimer_interrupt_hanging(struct clock_event_device *dev, - ktime_t try_time) -{ - force_clock_reprogram = 1; - dev->min_delta_ns = (unsigned long)try_time.tv64 * 3; - printk(KERN_WARNING "hrtimer: interrupt too slow, " - "forcing clock min delta to %llu ns\n", - (unsigned long long) dev->min_delta_ns); -} /* * High resolution timer interrupt * Called with interrupts disabled @@ -1249,21 +1234,15 @@ void hrtimer_interrupt(struct clock_event_device *dev) { struct hrtimer_cpu_base *cpu_base = &__get_cpu_var(hrtimer_bases); struct hrtimer_clock_base *base; - ktime_t expires_next, now; - int nr_retries = 0; - int i; + ktime_t expires_next, now, entry_time, delta; + int i, retries = 0; BUG_ON(!cpu_base->hres_active); cpu_base->nr_events++; dev->next_event.tv64 = KTIME_MAX; - retry: - /* 5 retries is enough to notice a hang */ - if (!(++nr_retries % 5)) - hrtimer_interrupt_hanging(dev, ktime_sub(ktime_get(), now)); - - now = ktime_get(); - + entry_time = now = ktime_get(); +retry: expires_next.tv64 = KTIME_MAX; spin_lock(&cpu_base->lock); @@ -1325,10 +1304,48 @@ void hrtimer_interrupt(struct clock_event_device *dev) spin_unlock(&cpu_base->lock); /* Reprogramming necessary ? */ - if (expires_next.tv64 != KTIME_MAX) { - if (tick_program_event(expires_next, force_clock_reprogram)) - goto retry; + if (expires_next.tv64 == KTIME_MAX || + !tick_program_event(expires_next, 0)) { + cpu_base->hang_detected = 0; + return; } + + /* + * The next timer was already expired due to: + * - tracing + * - long lasting callbacks + * - being scheduled away when running in a VM + * + * We need to prevent that we loop forever in the hrtimer + * interrupt routine. We give it 3 attempts to avoid + * overreacting on some spurious event. + */ + now = ktime_get(); + cpu_base->nr_retries++; + if (++retries < 3) + goto retry; + /* + * Give the system a chance to do something else than looping + * here. We stored the entry time, so we know exactly how long + * we spent here. We schedule the next event this amount of + * time away. + */ + cpu_base->nr_hangs++; + cpu_base->hang_detected = 1; + delta = ktime_sub(now, entry_time); + if (delta.tv64 > cpu_base->max_hang_time.tv64) + cpu_base->max_hang_time = delta; + /* + * Limit it to a sensible value as we enforce a longer + * delay. Give the CPU at least 100ms to catch up. + */ + if (delta.tv64 > 100 * NSEC_PER_MSEC) + expires_next = ktime_add_ns(now, 100 * NSEC_PER_MSEC); + else + expires_next = ktime_add(now, delta); + tick_program_event(expires_next, 1); + printk_once(KERN_WARNING "hrtimer: interrupt took %llu ns\n", + ktime_to_ns(delta)); } /* diff --git a/kernel/time/timer_list.c b/kernel/time/timer_list.c index 665c76edbf17..9d80db4747d4 100644 --- a/kernel/time/timer_list.c +++ b/kernel/time/timer_list.c @@ -150,6 +150,9 @@ static void print_cpu(struct seq_file *m, int cpu, u64 now) P_ns(expires_next); P(hres_active); P(nr_events); + P(nr_retries); + P(nr_hangs); + P_ns(max_hang_time); #endif #undef P #undef P_ns @@ -254,7 +257,7 @@ static int timer_list_show(struct seq_file *m, void *v) u64 now = ktime_to_ns(ktime_get()); int cpu; - SEQ_printf(m, "Timer List Version: v0.4\n"); + SEQ_printf(m, "Timer List Version: v0.5\n"); SEQ_printf(m, "HRTIMER_MAX_CLOCK_BASES: %d\n", HRTIMER_MAX_CLOCK_BASES); SEQ_printf(m, "now at %Ld nsecs\n", (unsigned long long)now); -- cgit v1.2.3-71-gd317 From 5f201907dfe4ad42c44006ddfcec00ed12e59497 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Thu, 10 Dec 2009 10:56:29 +0100 Subject: hrtimer: move timer stats helper functions to hrtimer.c There is no reason to make timer_stats_hrtimer_set_start_info and friends visible to the rest of the kernel. So move all of them to hrtimer.c. Also make timer_stats_hrtimer_set_start_info a static inline function so it gets inlined and we avoid another function call. Based on a patch by Thomas Gleixner. Signed-off-by: Heiko Carstens LKML-Reference: <20091210095629.GC4144@osiris.boeblingen.de.ibm.com> Signed-off-by: Thomas Gleixner --- include/linux/hrtimer.h | 43 ------------------------------------------- kernel/hrtimer.c | 24 ++++++++++++++++++++---- 2 files changed, 20 insertions(+), 47 deletions(-) (limited to 'kernel') diff --git a/include/linux/hrtimer.h b/include/linux/hrtimer.h index 040b6796ab4d..af634e95871d 100644 --- a/include/linux/hrtimer.h +++ b/include/linux/hrtimer.h @@ -440,47 +440,4 @@ extern u64 ktime_divns(const ktime_t kt, s64 div); /* Show pending timers: */ extern void sysrq_timer_list_show(void); -/* - * Timer-statistics info: - */ -#ifdef CONFIG_TIMER_STATS - -extern void timer_stats_update_stats(void *timer, pid_t pid, void *startf, - void *timerf, char *comm, - unsigned int timer_flag); - -static inline void timer_stats_account_hrtimer(struct hrtimer *timer) -{ - if (likely(!timer_stats_active)) - return; - timer_stats_update_stats(timer, timer->start_pid, timer->start_site, - timer->function, timer->start_comm, 0); -} - -extern void __timer_stats_hrtimer_set_start_info(struct hrtimer *timer, - void *addr); - -static inline void timer_stats_hrtimer_set_start_info(struct hrtimer *timer) -{ - __timer_stats_hrtimer_set_start_info(timer, __builtin_return_address(0)); -} - -static inline void timer_stats_hrtimer_clear_start_info(struct hrtimer *timer) -{ - timer->start_site = NULL; -} -#else -static inline void timer_stats_account_hrtimer(struct hrtimer *timer) -{ -} - -static inline void timer_stats_hrtimer_set_start_info(struct hrtimer *timer) -{ -} - -static inline void timer_stats_hrtimer_clear_start_info(struct hrtimer *timer) -{ -} -#endif - #endif diff --git a/kernel/hrtimer.c b/kernel/hrtimer.c index 931a4d99bc55..d2f9239dc6ba 100644 --- a/kernel/hrtimer.c +++ b/kernel/hrtimer.c @@ -756,17 +756,33 @@ static inline void hrtimer_init_timer_hres(struct hrtimer *timer) { } #endif /* CONFIG_HIGH_RES_TIMERS */ -#ifdef CONFIG_TIMER_STATS -void __timer_stats_hrtimer_set_start_info(struct hrtimer *timer, void *addr) +static inline void timer_stats_hrtimer_set_start_info(struct hrtimer *timer) { +#ifdef CONFIG_TIMER_STATS if (timer->start_site) return; - - timer->start_site = addr; + timer->start_site = __builtin_return_address(0); memcpy(timer->start_comm, current->comm, TASK_COMM_LEN); timer->start_pid = current->pid; +#endif } + +static inline void timer_stats_hrtimer_clear_start_info(struct hrtimer *timer) +{ +#ifdef CONFIG_TIMER_STATS + timer->start_site = NULL; #endif +} + +static inline void timer_stats_account_hrtimer(struct hrtimer *timer) +{ +#ifdef CONFIG_TIMER_STATS + if (likely(!timer_stats_active)) + return; + timer_stats_update_stats(timer, timer->start_pid, timer->start_site, + timer->function, timer->start_comm, 0); +#endif +} /* * Counterpart to lock_hrtimer_base above: -- cgit v1.2.3-71-gd317 From dfc12eb26a285df316be68a808af86964f3bff86 Mon Sep 17 00:00:00 2001 From: Phil Carmody Date: Thu, 10 Dec 2009 14:29:37 +0200 Subject: sched: Fix memory leak in two error corner cases If the second in each of these pairs of allocations fails, then the first one will not be freed in the error route out. Found by a static code analysis tool. Signed-off-by: Phil Carmody Acked-by: Peter Zijlstra LKML-Reference: <1260448177-28448-1-git-send-email-ext-phil.2.carmody@nokia.com> Signed-off-by: Ingo Molnar --- kernel/sched.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/sched.c b/kernel/sched.c index 3de3deab8095..36cc05a76947 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -9855,13 +9855,15 @@ int alloc_fair_sched_group(struct task_group *tg, struct task_group *parent) se = kzalloc_node(sizeof(struct sched_entity), GFP_KERNEL, cpu_to_node(i)); if (!se) - goto err; + goto err_free_rq; init_tg_cfs_entry(tg, cfs_rq, se, i, 0, parent->se[i]); } return 1; + err_free_rq: + kfree(cfs_rq); err: return 0; } @@ -9943,13 +9945,15 @@ int alloc_rt_sched_group(struct task_group *tg, struct task_group *parent) rt_se = kzalloc_node(sizeof(struct sched_rt_entity), GFP_KERNEL, cpu_to_node(i)); if (!rt_se) - goto err; + goto err_free_rq; init_tg_rt_entry(tg, rt_rq, rt_se, i, 0, parent->rt_se[i]); } return 1; + err_free_rq: + kfree(rt_rq); err: return 0; } -- cgit v1.2.3-71-gd317 From 5e855db5d8fec44e6604eb245aa9077bbd3f0d05 Mon Sep 17 00:00:00 2001 From: Xiao Guangrong Date: Thu, 10 Dec 2009 17:08:54 +0800 Subject: perf_event: Fix variable initialization in other codepaths Signed-off-by: Xiao Guangrong Cc: Peter Zijlstra Cc: Frederic Weisbecker Cc: Paul Mackerras LKML-Reference: <4B20BAA6.7010609@cn.fujitsu.com> Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/perf_event.c | 4 ++++ kernel/perf_event.c | 1 + 2 files changed, 5 insertions(+) (limited to 'kernel') diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c index d35f26076ae5..1342f236e32a 100644 --- a/arch/x86/kernel/cpu/perf_event.c +++ b/arch/x86/kernel/cpu/perf_event.c @@ -1632,6 +1632,7 @@ static void intel_pmu_drain_bts_buffer(struct cpu_hw_events *cpuc) data.period = event->hw.last_period; data.addr = 0; + data.raw = NULL; regs.ip = 0; /* @@ -1749,6 +1750,7 @@ static int p6_pmu_handle_irq(struct pt_regs *regs) u64 val; data.addr = 0; + data.raw = NULL; cpuc = &__get_cpu_var(cpu_hw_events); @@ -1794,6 +1796,7 @@ static int intel_pmu_handle_irq(struct pt_regs *regs) u64 ack, status; data.addr = 0; + data.raw = NULL; cpuc = &__get_cpu_var(cpu_hw_events); @@ -1857,6 +1860,7 @@ static int amd_pmu_handle_irq(struct pt_regs *regs) u64 val; data.addr = 0; + data.raw = NULL; cpuc = &__get_cpu_var(cpu_hw_events); diff --git a/kernel/perf_event.c b/kernel/perf_event.c index 3a5d6c4786bb..d891ec4a8100 100644 --- a/kernel/perf_event.c +++ b/kernel/perf_event.c @@ -4300,6 +4300,7 @@ void perf_bp_event(struct perf_event *bp, void *data) struct perf_sample_data sample; struct pt_regs *regs = data; + sample.raw = NULL; sample.addr = bp->attr.bp_addr; if (!perf_exclude_event(bp, regs)) -- cgit v1.2.3-71-gd317 From b9889ed1ddeca5a3f3569c8de7354e9e97d803ae Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Thu, 10 Dec 2009 20:32:39 +0100 Subject: sched: Remove forced2_migrations stats This build warning: kernel/sched.c: In function 'set_task_cpu': kernel/sched.c:2070: warning: unused variable 'old_rq' Made me realize that the forced2_migrations stat looks pretty pointless (and a misnomer) - remove it. Cc: Peter Zijlstra Cc: Mike Galbraith LKML-Reference: Signed-off-by: Ingo Molnar --- include/linux/sched.h | 1 - kernel/sched.c | 6 ------ kernel/sched_debug.c | 2 -- 3 files changed, 9 deletions(-) (limited to 'kernel') diff --git a/include/linux/sched.h b/include/linux/sched.h index ee9f200d12d3..87b89a827f0c 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1174,7 +1174,6 @@ struct sched_entity { u64 nr_failed_migrations_running; u64 nr_failed_migrations_hot; u64 nr_forced_migrations; - u64 nr_forced2_migrations; u64 nr_wakeups; u64 nr_wakeups_sync; diff --git a/kernel/sched.c b/kernel/sched.c index 36cc05a76947..bc68037f3199 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -2067,7 +2067,6 @@ task_hot(struct task_struct *p, u64 now, struct sched_domain *sd) void set_task_cpu(struct task_struct *p, unsigned int new_cpu) { int old_cpu = task_cpu(p); - struct rq *old_rq = cpu_rq(old_cpu); struct cfs_rq *old_cfsrq = task_cfs_rq(p), *new_cfsrq = cpu_cfs_rq(old_cfsrq, new_cpu); @@ -2075,10 +2074,6 @@ void set_task_cpu(struct task_struct *p, unsigned int new_cpu) if (old_cpu != new_cpu) { p->se.nr_migrations++; -#ifdef CONFIG_SCHEDSTATS - if (task_hot(p, old_rq->clock, NULL)) - schedstat_inc(p, se.nr_forced2_migrations); -#endif perf_sw_event(PERF_COUNT_SW_CPU_MIGRATIONS, 1, 1, NULL, 0); } @@ -2521,7 +2516,6 @@ static void __sched_fork(struct task_struct *p) p->se.nr_failed_migrations_running = 0; p->se.nr_failed_migrations_hot = 0; p->se.nr_forced_migrations = 0; - p->se.nr_forced2_migrations = 0; p->se.nr_wakeups = 0; p->se.nr_wakeups_sync = 0; diff --git a/kernel/sched_debug.c b/kernel/sched_debug.c index 0fc5287fe80f..5ae24fc65d75 100644 --- a/kernel/sched_debug.c +++ b/kernel/sched_debug.c @@ -432,7 +432,6 @@ void proc_sched_show_task(struct task_struct *p, struct seq_file *m) P(se.nr_failed_migrations_running); P(se.nr_failed_migrations_hot); P(se.nr_forced_migrations); - P(se.nr_forced2_migrations); P(se.nr_wakeups); P(se.nr_wakeups_sync); P(se.nr_wakeups_migrate); @@ -508,7 +507,6 @@ void proc_sched_set_task(struct task_struct *p) p->se.nr_failed_migrations_running = 0; p->se.nr_failed_migrations_hot = 0; p->se.nr_forced_migrations = 0; - p->se.nr_forced2_migrations = 0; p->se.nr_wakeups = 0; p->se.nr_wakeups_sync = 0; p->se.nr_wakeups_migrate = 0; -- cgit v1.2.3-71-gd317 From d954fbf0ff6b5fdfb32350e85a2f15d3db976506 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Thu, 10 Dec 2009 23:46:52 +0100 Subject: tracing: Fix wrong usage of strstrip in trace_ksyms MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit strstrip returns a pointer to the first non space character, but the code in parse_ksym_trace_str() ignores that. strstrip is now must_check and therefor we get the correct warning: kernel/trace/trace_ksym.c:294: warning: ignoring return value of ‘strstrip’, declared with attribute warn_unused_result We are really not interested in leading whitespace here. Fix that and cleanup the dozen kfree() exit pathes. Signed-off-by: Thomas Gleixner Cc: Xiao Guangrong Cc: Steven Rostedt --- kernel/trace/trace_ksym.c | 49 ++++++++++++++++++++--------------------------- 1 file changed, 21 insertions(+), 28 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/trace_ksym.c b/kernel/trace/trace_ksym.c index ddfa0fd43bc0..64e7a5bd6692 100644 --- a/kernel/trace/trace_ksym.c +++ b/kernel/trace/trace_ksym.c @@ -277,21 +277,20 @@ static ssize_t ksym_trace_filter_write(struct file *file, { struct trace_ksym *entry; struct hlist_node *node; - char *input_string, *ksymname = NULL; + char *buf, *input_string, *ksymname = NULL; unsigned long ksym_addr = 0; int ret, op, changed = 0; - input_string = kzalloc(count + 1, GFP_KERNEL); - if (!input_string) + buf = kzalloc(count + 1, GFP_KERNEL); + if (!buf) return -ENOMEM; - if (copy_from_user(input_string, buffer, count)) { - kfree(input_string); - return -EFAULT; - } - input_string[count] = '\0'; + ret = -EFAULT; + if (copy_from_user(buf, buffer, count)) + goto out; - strstrip(input_string); + buf[count] = '\0'; + input_string = strstrip(buf); /* * Clear all breakpoints if: @@ -302,15 +301,13 @@ static ssize_t ksym_trace_filter_write(struct file *file, if (!input_string[0] || !strcmp(input_string, "0") || !strcmp(input_string, "*:---")) { __ksym_trace_reset(); - kfree(input_string); - return count; + ret = 0; + goto out; } ret = op = parse_ksym_trace_str(input_string, &ksymname, &ksym_addr); - if (ret < 0) { - kfree(input_string); - return ret; - } + if (ret < 0) + goto out; mutex_lock(&ksym_tracer_mutex); @@ -321,7 +318,7 @@ static ssize_t ksym_trace_filter_write(struct file *file, if (entry->attr.bp_type != op) changed = 1; else - goto out; + goto out_unlock; break; } } @@ -336,28 +333,24 @@ static ssize_t ksym_trace_filter_write(struct file *file, if (IS_ERR(entry->ksym_hbp)) ret = PTR_ERR(entry->ksym_hbp); else - goto out; + goto out_unlock; } /* Error or "symbol:---" case: drop it */ ksym_filter_entry_count--; hlist_del_rcu(&(entry->ksym_hlist)); synchronize_rcu(); kfree(entry); - goto out; + goto out_unlock; } else { /* Check for malformed request: (4) */ - if (op == 0) - goto out; - ret = process_new_ksym_entry(ksymname, op, ksym_addr); + if (op) + ret = process_new_ksym_entry(ksymname, op, ksym_addr); } -out: +out_unlock: mutex_unlock(&ksym_tracer_mutex); - - kfree(input_string); - - if (!ret) - ret = count; - return ret; +out: + kfree(buf); + return !ret ? count : ret; } static const struct file_operations ksym_tracing_fops = { -- cgit v1.2.3-71-gd317 From 184210154b9aa570099183f6c062ac4eb11190b7 Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Thu, 10 Dec 2009 22:54:27 -0500 Subject: ring-buffer: Use sync sched protection on ring buffer resizing There was a comment in the ring buffer code that says the calling layers should prevent tracing or reading of the ring buffer while resizing. I have discovered that the tracers do not honor this arrangement. This patch moves the disabling and synchronizing the ring buffer to a higher layer during resizing. This guarantees that no writes are occurring while the resize takes place. Signed-off-by: Steven Rostedt --- kernel/trace/ring_buffer.c | 25 +++++++++---------------- 1 file changed, 9 insertions(+), 16 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c index a1ca4956ab5e..0d64c51ab4df 100644 --- a/kernel/trace/ring_buffer.c +++ b/kernel/trace/ring_buffer.c @@ -1193,9 +1193,6 @@ rb_remove_pages(struct ring_buffer_per_cpu *cpu_buffer, unsigned nr_pages) struct list_head *p; unsigned i; - atomic_inc(&cpu_buffer->record_disabled); - synchronize_sched(); - spin_lock_irq(&cpu_buffer->reader_lock); rb_head_page_deactivate(cpu_buffer); @@ -1214,9 +1211,6 @@ rb_remove_pages(struct ring_buffer_per_cpu *cpu_buffer, unsigned nr_pages) spin_unlock_irq(&cpu_buffer->reader_lock); rb_check_pages(cpu_buffer); - - atomic_dec(&cpu_buffer->record_disabled); - } static void @@ -1227,9 +1221,6 @@ rb_insert_pages(struct ring_buffer_per_cpu *cpu_buffer, struct list_head *p; unsigned i; - atomic_inc(&cpu_buffer->record_disabled); - synchronize_sched(); - spin_lock_irq(&cpu_buffer->reader_lock); rb_head_page_deactivate(cpu_buffer); @@ -1245,8 +1236,6 @@ rb_insert_pages(struct ring_buffer_per_cpu *cpu_buffer, spin_unlock_irq(&cpu_buffer->reader_lock); rb_check_pages(cpu_buffer); - - atomic_dec(&cpu_buffer->record_disabled); } /** @@ -1254,11 +1243,6 @@ rb_insert_pages(struct ring_buffer_per_cpu *cpu_buffer, * @buffer: the buffer to resize. * @size: the new size. * - * The tracer is responsible for making sure that the buffer is - * not being used while changing the size. - * Note: We may be able to change the above requirement by using - * RCU synchronizations. - * * Minimum size is 2 * BUF_PAGE_SIZE. * * Returns -1 on failure. @@ -1290,6 +1274,11 @@ int ring_buffer_resize(struct ring_buffer *buffer, unsigned long size) if (size == buffer_size) return size; + atomic_inc(&buffer->record_disabled); + + /* Make sure all writers are done with this buffer. */ + synchronize_sched(); + mutex_lock(&buffer->mutex); get_online_cpus(); @@ -1352,6 +1341,8 @@ int ring_buffer_resize(struct ring_buffer *buffer, unsigned long size) put_online_cpus(); mutex_unlock(&buffer->mutex); + atomic_dec(&buffer->record_disabled); + return size; free_pages: @@ -1361,6 +1352,7 @@ int ring_buffer_resize(struct ring_buffer *buffer, unsigned long size) } put_online_cpus(); mutex_unlock(&buffer->mutex); + atomic_dec(&buffer->record_disabled); return -ENOMEM; /* @@ -1370,6 +1362,7 @@ int ring_buffer_resize(struct ring_buffer *buffer, unsigned long size) out_fail: put_online_cpus(); mutex_unlock(&buffer->mutex); + atomic_dec(&buffer->record_disabled); return -1; } EXPORT_SYMBOL_GPL(ring_buffer_resize); -- cgit v1.2.3-71-gd317 From dd7f59435782a02ceb6d16b9ce823dd3345d75ec Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Thu, 10 Dec 2009 23:20:52 -0500 Subject: ring-buffer: Move resize integrity check under reader lock While using an application that does splice on the ftrace ring buffer at start up, I triggered an integrity check failure. Looking into this, I discovered that resizing the buffer performs an integrity check after the buffer is resized. This check unfortunately is preformed after it releases the reader lock. If a reader is reading the buffer it may cause the integrity check to trigger a false failure. This patch simply moves the integrity checker under the protection of the ring buffer reader lock. Signed-off-by: Steven Rostedt --- kernel/trace/ring_buffer.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c index 0d64c51ab4df..eccb4cf1e998 100644 --- a/kernel/trace/ring_buffer.c +++ b/kernel/trace/ring_buffer.c @@ -1208,9 +1208,9 @@ rb_remove_pages(struct ring_buffer_per_cpu *cpu_buffer, unsigned nr_pages) return; rb_reset_cpu(cpu_buffer); - spin_unlock_irq(&cpu_buffer->reader_lock); - rb_check_pages(cpu_buffer); + + spin_unlock_irq(&cpu_buffer->reader_lock); } static void @@ -1233,9 +1233,9 @@ rb_insert_pages(struct ring_buffer_per_cpu *cpu_buffer, list_add_tail(&bpage->list, cpu_buffer->pages); } rb_reset_cpu(cpu_buffer); - spin_unlock_irq(&cpu_buffer->reader_lock); - rb_check_pages(cpu_buffer); + + spin_unlock_irq(&cpu_buffer->reader_lock); } /** -- cgit v1.2.3-71-gd317 From 84667d4849b0e0a939a76f9f62d45fa3b4d59692 Mon Sep 17 00:00:00 2001 From: Jason Wessel Date: Fri, 11 Dec 2009 08:43:13 -0600 Subject: kgdb: Read buffer overflow Roel Kluin reported an error found with Parfait. Where we want to ensure that that kgdb_info[-1] never gets accessed. Also check to ensure any negative tid does not exceed the size of the shadow CPU array, else report critical debug context because it is an internal kgdb failure. Reported-by: Roel Kluin Signed-off-by: Jason Wessel --- kernel/kgdb.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/kgdb.c b/kernel/kgdb.c index 7d7014634022..29357a9ccfb2 100644 --- a/kernel/kgdb.c +++ b/kernel/kgdb.c @@ -541,12 +541,17 @@ static struct task_struct *getthread(struct pt_regs *regs, int tid) */ if (tid == 0 || tid == -1) tid = -atomic_read(&kgdb_active) - 2; - if (tid < 0) { + if (tid < -1 && tid > -NR_CPUS - 2) { if (kgdb_info[-tid - 2].task) return kgdb_info[-tid - 2].task; else return idle_task(-tid - 2); } + if (tid <= 0) { + printk(KERN_ERR "KGDB: Internal thread select error\n"); + dump_stack(); + return NULL; + } /* * find_task_by_pid_ns() does not take the tasklist lock anymore -- cgit v1.2.3-71-gd317 From 028e7b175970be8fca58bfd7d61cc375babe40b7 Mon Sep 17 00:00:00 2001 From: Jason Wessel Date: Fri, 11 Dec 2009 08:43:17 -0600 Subject: kgdb: allow for cpu switch when single stepping The kgdb core should not assume that a single step operation of a kernel thread will complete on the same CPU. The single step flag is set at the "thread" level and it is possible in a multi cpu system that a kernel thread can get scheduled on another cpu the next time it is run. As a further safety net in case a slave cpu is hung, the debug master cpu will try 100 times before giving up and assuming control of the slave cpus is no longer possible. It is more useful to be able to get some information out of kgdb instead of spinning forever. Signed-off-by: Jason Wessel --- kernel/kgdb.c | 20 +++++++++++++++----- 1 file changed, 15 insertions(+), 5 deletions(-) (limited to 'kernel') diff --git a/kernel/kgdb.c b/kernel/kgdb.c index 29357a9ccfb2..ca21fe98e8de 100644 --- a/kernel/kgdb.c +++ b/kernel/kgdb.c @@ -129,6 +129,7 @@ struct task_struct *kgdb_usethread; struct task_struct *kgdb_contthread; int kgdb_single_step; +pid_t kgdb_sstep_pid; /* Our I/O buffers. */ static char remcom_in_buffer[BUFMAX]; @@ -1400,6 +1401,7 @@ kgdb_handle_exception(int evector, int signo, int ecode, struct pt_regs *regs) struct kgdb_state kgdb_var; struct kgdb_state *ks = &kgdb_var; unsigned long flags; + int sstep_tries = 100; int error = 0; int i, cpu; @@ -1430,13 +1432,14 @@ acquirelock: cpu_relax(); /* - * Do not start the debugger connection on this CPU if the last - * instance of the exception handler wanted to come into the - * debugger on a different CPU via a single step + * For single stepping, try to only enter on the processor + * that was single stepping. To gaurd against a deadlock, the + * kernel will only try for the value of sstep_tries before + * giving up and continuing on. */ if (atomic_read(&kgdb_cpu_doing_single_step) != -1 && - atomic_read(&kgdb_cpu_doing_single_step) != cpu) { - + (kgdb_info[cpu].task && + kgdb_info[cpu].task->pid != kgdb_sstep_pid) && --sstep_tries) { atomic_set(&kgdb_active, -1); touch_softlockup_watchdog(); clocksource_touch_watchdog(); @@ -1529,6 +1532,13 @@ acquirelock: } kgdb_restore: + if (atomic_read(&kgdb_cpu_doing_single_step) != -1) { + int sstep_cpu = atomic_read(&kgdb_cpu_doing_single_step); + if (kgdb_info[sstep_cpu].task) + kgdb_sstep_pid = kgdb_info[sstep_cpu].task->pid; + else + kgdb_sstep_pid = 0; + } /* Free kgdb_active */ atomic_set(&kgdb_active, -1); touch_softlockup_watchdog(); -- cgit v1.2.3-71-gd317 From d625e9c0d706eb43afbf52634d5cecacae1d57cc Mon Sep 17 00:00:00 2001 From: Jason Wessel Date: Mon, 27 Apr 2009 13:20:21 -0500 Subject: kgdb: continue and warn on signal passing from gdb On some architectures for the segv trap, gdb wants to pass the signal back on continue. For kgdb this is not the default behavior, because it can cause the kernel to crash if you arbitrarily pass back a exception outside of kgdb. Instead of causing instability, pass a message back to gdb about the supported kgdb signal passing and execute a standard kgdb continue operation. Signed-off-by: Jason Wessel --- kernel/kgdb.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/kgdb.c b/kernel/kgdb.c index ca21fe98e8de..8584eac55e30 100644 --- a/kernel/kgdb.c +++ b/kernel/kgdb.c @@ -1210,8 +1210,10 @@ static int gdb_cmd_exception_pass(struct kgdb_state *ks) return 1; } else { - error_packet(remcom_out_buffer, -EINVAL); - return 0; + kgdb_msg_write("KGDB only knows signal 9 (pass)" + " and 15 (pass and disconnect)\n" + "Executing a continue without signal passing\n", 0); + remcom_in_buffer[0] = 'c'; } /* Indicate fall through */ -- cgit v1.2.3-71-gd317 From 7f8b7ed6f825c729332b8190aca55c6bf95b158e Mon Sep 17 00:00:00 2001 From: Jason Wessel Date: Fri, 11 Dec 2009 08:43:20 -0600 Subject: kgdb: Always process the whole breakpoint list on activate or deactivate This patch fixes 2 edge cases in using kgdb in conjunction with gdb. 1) kgdb_deactivate_sw_breakpoints() should process the entire array of breakpoints. The failure to do so results in breakpoints that you cannot remove, because a break point can only be removed if its state flag is set to BP_SET. The easy way to duplicate this problem is to plant a break point in a kernel module and then unload the kernel module. 2) kgdb_activate_sw_breakpoints() should process the entire array of breakpoints. The failure to do so results in missed breakpoints when a breakpoint cannot be activated. Signed-off-by: Jason Wessel --- kernel/kgdb.c | 23 +++++++++++++++-------- 1 file changed, 15 insertions(+), 8 deletions(-) (limited to 'kernel') diff --git a/kernel/kgdb.c b/kernel/kgdb.c index 8584eac55e30..2eb517e23514 100644 --- a/kernel/kgdb.c +++ b/kernel/kgdb.c @@ -625,7 +625,8 @@ static void kgdb_flush_swbreak_addr(unsigned long addr) static int kgdb_activate_sw_breakpoints(void) { unsigned long addr; - int error = 0; + int error; + int ret = 0; int i; for (i = 0; i < KGDB_MAX_BREAKPOINTS; i++) { @@ -635,13 +636,16 @@ static int kgdb_activate_sw_breakpoints(void) addr = kgdb_break[i].bpt_addr; error = kgdb_arch_set_breakpoint(addr, kgdb_break[i].saved_instr); - if (error) - return error; + if (error) { + ret = error; + printk(KERN_INFO "KGDB: BP install failed: %lx", addr); + continue; + } kgdb_flush_swbreak_addr(addr); kgdb_break[i].state = BP_ACTIVE; } - return 0; + return ret; } static int kgdb_set_sw_break(unsigned long addr) @@ -688,7 +692,8 @@ static int kgdb_set_sw_break(unsigned long addr) static int kgdb_deactivate_sw_breakpoints(void) { unsigned long addr; - int error = 0; + int error; + int ret = 0; int i; for (i = 0; i < KGDB_MAX_BREAKPOINTS; i++) { @@ -697,13 +702,15 @@ static int kgdb_deactivate_sw_breakpoints(void) addr = kgdb_break[i].bpt_addr; error = kgdb_arch_remove_breakpoint(addr, kgdb_break[i].saved_instr); - if (error) - return error; + if (error) { + printk(KERN_INFO "KGDB: BP remove failed: %lx\n", addr); + ret = error; + } kgdb_flush_swbreak_addr(addr); kgdb_break[i].state = BP_SET; } - return 0; + return ret; } static int kgdb_remove_sw_break(unsigned long addr) -- cgit v1.2.3-71-gd317 From 03889384cee7a198a79447c1ea6aca2c8e54d155 Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Fri, 11 Dec 2009 09:48:22 -0500 Subject: tracing: Add trace_dump_stack() I've been asked a few times about how to find out what is calling some location in the kernel. One way is to use dynamic function tracing and implement the func_stack_trace. But this only finds out who is calling a particular function. It does not tell you who is calling that function and entering a specific if conditional. I have myself implemented a quick version of trace_dump_stack() for this purpose a few times, and just needed it now. This is when I realized that this would be a good tool to have in the kernel like trace_printk(). Using trace_dump_stack() is similar to dump_stack() except that it writes to the trace buffer instead and can be used in critical locations. For example: @@ -5485,8 +5485,12 @@ need_resched_nonpreemptible: if (prev->state && !(preempt_count() & PREEMPT_ACTIVE)) { if (unlikely(signal_pending_state(prev->state, prev))) prev->state = TASK_RUNNING; - else + else { deactivate_task(rq, prev, 1); + trace_printk("Deactivating task %s:%d\n", + prev->comm, prev->pid); + trace_dump_stack(); + } switch_count = &prev->nvcsw; } Produces: <...>-3249 [001] 296.105269: schedule: Deactivating task ntpd:3249 <...>-3249 [001] 296.105270: => schedule => schedule_hrtimeout_range => poll_schedule_timeout => do_select => core_sys_select => sys_select => system_call_fastpath Signed-off-by: Steven Rostedt --- include/linux/kernel.h | 2 ++ kernel/trace/trace.c | 16 ++++++++++++++++ 2 files changed, 18 insertions(+) (limited to 'kernel') diff --git a/include/linux/kernel.h b/include/linux/kernel.h index 3fa4c590cf12..5ad4199fb073 100644 --- a/include/linux/kernel.h +++ b/include/linux/kernel.h @@ -492,6 +492,8 @@ extern int __trace_printk(unsigned long ip, const char *fmt, ...) __attribute__ ((format (printf, 2, 3))); +extern void trace_dump_stack(void); + /* * The double __builtin_constant_p is because gcc will give us an error * if we try to allocate the static variable to fmt if it is not a diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 88bd9ae2a9ed..f531301b7a3b 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -1151,6 +1151,22 @@ void __trace_stack(struct trace_array *tr, unsigned long flags, int skip, __ftrace_trace_stack(tr->buffer, flags, skip, pc); } +/** + * trace_dump_stack - record a stack back trace in the trace buffer + */ +void trace_dump_stack(void) +{ + unsigned long flags; + + if (tracing_disabled || tracing_selftest_running) + return 0; + + local_save_flags(flags); + + /* skipping 3 traces, seems to get us at the caller of this function */ + __ftrace_trace_stack(global_trace.buffer, flags, 3, preempt_count()); +} + void ftrace_trace_userstack(struct ring_buffer *buffer, unsigned long flags, int pc) { -- cgit v1.2.3-71-gd317 From cc51a0fca66658ea710db566ba17e80e3f7d4957 Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Fri, 11 Dec 2009 11:54:51 -0500 Subject: tracing: Add stack trace to irqsoff tracer The irqsoff and friends tracers help in finding causes of latency in the kernel. The also work with the function tracer to show what was happening when interrupts or preemption are disabled. But the function tracer has a bit of an overhead and can cause exagerated readings. Currently, when tracing with /proc/sys/kernel/ftrace_enabled = 0, where the function tracer is disabled, the information that is provided can end up being useless. For example, a 2 and a half millisecond latency only showed: # tracer: preemptirqsoff # # preemptirqsoff latency trace v1.1.5 on 2.6.32 # -------------------------------------------------------------------- # latency: 2463 us, #4/4, CPU#2 | (M:preempt VP:0, KP:0, SP:0 HP:0 #P:4) # ----------------- # | task: -4242 (uid:0 nice:0 policy:0 rt_prio:0) # ----------------- # => started at: _spin_lock_irqsave # => ended at: remove_wait_queue # # # _------=> CPU# # / _-----=> irqs-off # | / _----=> need-resched # || / _---=> hardirq/softirq # ||| / _--=> preempt-depth # |||| /_--=> lock-depth # |||||/ delay # cmd pid |||||| time | caller # \ / |||||| \ | / hackbenc-4242 2d.... 0us!: trace_hardirqs_off <-_spin_lock_irqsave hackbenc-4242 2...1. 2463us+: _spin_unlock_irqrestore <-remove_wait_queue hackbenc-4242 2...1. 2466us : trace_preempt_on <-remove_wait_queue The above lets us know that hackbench with pid 2463 grabbed a spin lock somewhere and enabled preemption at remove_wait_queue. This helps a little but where this actually happened is not informative. This patch adds the stack dump to the end of the irqsoff tracer. This provides the following output: hackbenc-4242 2d.... 0us!: trace_hardirqs_off <-_spin_lock_irqsave hackbenc-4242 2...1. 2463us+: _spin_unlock_irqrestore <-remove_wait_queue hackbenc-4242 2...1. 2466us : trace_preempt_on <-remove_wait_queue hackbenc-4242 2...1. 2467us : => sub_preempt_count => _spin_unlock_irqrestore => remove_wait_queue => free_poll_entry => poll_freewait => do_sys_poll => sys_poll => system_call_fastpath Now we see that the culprit of this latency was the free_poll_entry code. Signed-off-by: Steven Rostedt --- kernel/trace/trace_irqsoff.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'kernel') diff --git a/kernel/trace/trace_irqsoff.c b/kernel/trace/trace_irqsoff.c index 3aa7eaa2114c..2974bc7538c7 100644 --- a/kernel/trace/trace_irqsoff.c +++ b/kernel/trace/trace_irqsoff.c @@ -151,6 +151,8 @@ check_critical_timing(struct trace_array *tr, goto out_unlock; trace_function(tr, CALLER_ADDR0, parent_ip, flags, pc); + /* Skip 5 functions to get to the irq/preempt enable function */ + __trace_stack(tr, flags, 5, pc); if (data->critical_sequence != max_sequence) goto out_unlock; -- cgit v1.2.3-71-gd317 From 5ec93d1154fd1e269162398f8e70efc7e004485d Mon Sep 17 00:00:00 2001 From: Alan Cox Date: Mon, 30 Nov 2009 13:18:45 +0000 Subject: tty: Move the leader test in disassociate There are two call points, both want to check that tty->signal->leader is set. Move the test into disassociate_ctty() as that will make locking changes easier in a bit Signed-off-by: Alan Cox Signed-off-by: Greg Kroah-Hartman --- drivers/char/tty_io.c | 5 +++-- kernel/exit.c | 2 +- 2 files changed, 4 insertions(+), 3 deletions(-) (limited to 'kernel') diff --git a/drivers/char/tty_io.c b/drivers/char/tty_io.c index cc941a3504d7..a19fef2093f1 100644 --- a/drivers/char/tty_io.c +++ b/drivers/char/tty_io.c @@ -707,6 +707,8 @@ void disassociate_ctty(int on_exit) struct tty_struct *tty; struct pid *tty_pgrp = NULL; + if (!current->signal->leader) + return; tty = get_current_tty(); if (tty) { @@ -772,8 +774,7 @@ void no_tty(void) { struct task_struct *tsk = current; lock_kernel(); - if (tsk->signal->leader) - disassociate_ctty(0); + disassociate_ctty(0); unlock_kernel(); proc_clear_tty(tsk); } diff --git a/kernel/exit.c b/kernel/exit.c index 1143012951e9..6f50ef55a6f3 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -971,7 +971,7 @@ NORET_TYPE void do_exit(long code) exit_thread(); cgroup_exit(tsk, 1); - if (group_dead && tsk->signal->leader) + if (group_dead) disassociate_ctty(1); module_put(task_thread_info(tsk)->exec_domain->module); -- cgit v1.2.3-71-gd317 From 87d9b4e1c52867a45331a9a5495f6448e0c68b23 Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Tue, 8 Dec 2009 11:14:20 +0800 Subject: tracing: Extract duplicate ftrace_raw_init_event_foo() Use a generic trace_event_raw_init() function for all event's raw_init callbacks (but kprobes) instead of defining the same version for each of these. This shrinks the kernel code: text data bss dec hex filename 5355293 1961928 7103260 14420481 dc0a01 vmlinux.o.old 5346802 1961864 7103260 14411926 dbe896 vmlinux.o raw_init can't be removed, because ftrace events and kprobe events use different raw_init callbacks. Though it's possible to totally remove raw_init, I choose to leave it as it is for now. Signed-off-by: Li Zefan Acked-by: Steven Rostedt Cc: Jason Baron Cc: Ingo Molnar LKML-Reference: <4B1DC48C.7080603@cn.fujitsu.com> Signed-off-by: Frederic Weisbecker --- include/linux/ftrace_event.h | 1 + include/linux/syscalls.h | 4 ++-- include/trace/ftrace.h | 35 ++++------------------------------- kernel/trace/trace_events.c | 14 ++++++++++++++ 4 files changed, 21 insertions(+), 33 deletions(-) (limited to 'kernel') diff --git a/include/linux/ftrace_event.h b/include/linux/ftrace_event.h index 38f8d6553831..ea44b8911094 100644 --- a/include/linux/ftrace_event.h +++ b/include/linux/ftrace_event.h @@ -158,6 +158,7 @@ enum { FILTER_PTR_STRING, }; +extern int trace_event_raw_init(struct ftrace_event_call *call); extern int trace_define_common_fields(struct ftrace_event_call *call); extern int trace_define_field(struct ftrace_event_call *call, const char *type, const char *name, int offset, int size, diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h index bc70c5810fec..94ac28437bef 100644 --- a/include/linux/syscalls.h +++ b/include/linux/syscalls.h @@ -145,7 +145,7 @@ struct perf_event_attr; .name = "sys_enter"#sname, \ .system = "syscalls", \ .event = &enter_syscall_print_##sname, \ - .raw_init = init_syscall_trace, \ + .raw_init = trace_event_raw_init, \ .show_format = syscall_enter_format, \ .define_fields = syscall_enter_define_fields, \ .regfunc = reg_event_syscall_enter, \ @@ -167,7 +167,7 @@ struct perf_event_attr; .name = "sys_exit"#sname, \ .system = "syscalls", \ .event = &exit_syscall_print_##sname, \ - .raw_init = init_syscall_trace, \ + .raw_init = trace_event_raw_init, \ .show_format = syscall_exit_format, \ .define_fields = syscall_exit_define_fields, \ .regfunc = reg_event_syscall_exit, \ diff --git a/include/trace/ftrace.h b/include/trace/ftrace.h index c4eca380204e..6055b0604c86 100644 --- a/include/trace/ftrace.h +++ b/include/trace/ftrace.h @@ -623,23 +623,12 @@ static void ftrace_profile_disable_##name(struct ftrace_event_call *unused)\ * .trace = ftrace_raw_output_, <-- stage 2 * }; * - * static int ftrace_raw_init_event_(struct ftrace_event_call *unused) - * { - * int id; - * - * id = register_ftrace_event(&ftrace_event_type_); - * if (!id) - * return -ENODEV; - * event_.id = id; - * return 0; - * } - * * static struct ftrace_event_call __used * __attribute__((__aligned__(4))) * __attribute__((section("_ftrace_events"))) event_ = { * .name = "", * .system = "", - * .raw_init = ftrace_raw_init_event_, + * .raw_init = trace_event_raw_init, * .regfunc = ftrace_reg_event_, * .unregfunc = ftrace_unreg_event_, * .show_format = ftrace_format_, @@ -647,9 +636,6 @@ static void ftrace_profile_disable_##name(struct ftrace_event_call *unused)\ * */ -#undef TP_FMT -#define TP_FMT(fmt, args...) fmt "\n", ##args - #ifdef CONFIG_EVENT_PROFILE #define _TRACE_PROFILE_INIT(call) \ @@ -744,19 +730,7 @@ static void ftrace_raw_unreg_event_##call(struct ftrace_event_call *unused)\ \ static struct trace_event ftrace_event_type_##call = { \ .trace = ftrace_raw_output_##call, \ -}; \ - \ -static int ftrace_raw_init_event_##call(struct ftrace_event_call *unused)\ -{ \ - int id; \ - \ - id = register_ftrace_event(&ftrace_event_type_##call); \ - if (!id) \ - return -ENODEV; \ - event_##call.id = id; \ - INIT_LIST_HEAD(&event_##call.fields); \ - return 0; \ -} +}; #undef DEFINE_EVENT_PRINT #define DEFINE_EVENT_PRINT(template, name, proto, args, print) \ @@ -776,7 +750,7 @@ __attribute__((section("_ftrace_events"))) event_##call = { \ .name = #call, \ .system = __stringify(TRACE_SYSTEM), \ .event = &ftrace_event_type_##call, \ - .raw_init = ftrace_raw_init_event_##call, \ + .raw_init = trace_event_raw_init, \ .regfunc = ftrace_raw_reg_event_##call, \ .unregfunc = ftrace_raw_unreg_event_##call, \ .show_format = ftrace_format_##template, \ @@ -793,7 +767,7 @@ __attribute__((section("_ftrace_events"))) event_##call = { \ .name = #call, \ .system = __stringify(TRACE_SYSTEM), \ .event = &ftrace_event_type_##call, \ - .raw_init = ftrace_raw_init_event_##call, \ + .raw_init = trace_event_raw_init, \ .regfunc = ftrace_raw_reg_event_##call, \ .unregfunc = ftrace_raw_unreg_event_##call, \ .show_format = ftrace_format_##call, \ @@ -953,7 +927,6 @@ end: \ perf_swevent_put_recursion_context(rctx); \ end_recursion: \ local_irq_restore(irq_flags); \ - \ } #undef DEFINE_EVENT diff --git a/kernel/trace/trace_events.c b/kernel/trace/trace_events.c index 1d18315dc836..8ed66e0d476b 100644 --- a/kernel/trace/trace_events.c +++ b/kernel/trace/trace_events.c @@ -105,6 +105,20 @@ void trace_destroy_fields(struct ftrace_event_call *call) } } +int trace_event_raw_init(struct ftrace_event_call *call) +{ + int id; + + id = register_ftrace_event(call->event); + if (!id) + return -ENODEV; + call->id = id; + INIT_LIST_HEAD(&call->fields); + + return 0; +} +EXPORT_SYMBOL_GPL(trace_event_raw_init); + static void ftrace_event_enable_disable(struct ftrace_event_call *call, int enable) { -- cgit v1.2.3-71-gd317 From 614a71a26ba3d97e9fa85649db69a682b78e407d Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Tue, 8 Dec 2009 11:14:36 +0800 Subject: tracing: Pull up calls to trace_define_common_fields() Call trace_define_common_fields() in event_create_dir() only. This avoids trace events to handle it from their define_fields callbacks and shrinks the kernel code size: text data bss dec hex filename 5346802 1961864 7103260 14411926 dbe896 vmlinux.o.old 5345151 1961864 7103260 14410275 dbe223 vmlinux.o Signed-off-by: Li Zefan Acked-by: Steven Rostedt Cc: Ingo Molnar Cc: Jason Baron Cc: Masami Hiramatsu LKML-Reference: <4B1DC49C.8000107@cn.fujitsu.com> Signed-off-by: Frederic Weisbecker --- include/linux/ftrace_event.h | 1 - include/trace/ftrace.h | 4 ---- kernel/trace/trace_events.c | 7 ++++--- kernel/trace/trace_export.c | 4 ---- kernel/trace/trace_kprobe.c | 8 -------- kernel/trace/trace_syscalls.c | 8 -------- 6 files changed, 4 insertions(+), 28 deletions(-) (limited to 'kernel') diff --git a/include/linux/ftrace_event.h b/include/linux/ftrace_event.h index ea44b8911094..db97c64ce0e9 100644 --- a/include/linux/ftrace_event.h +++ b/include/linux/ftrace_event.h @@ -159,7 +159,6 @@ enum { }; extern int trace_event_raw_init(struct ftrace_event_call *call); -extern int trace_define_common_fields(struct ftrace_event_call *call); extern int trace_define_field(struct ftrace_event_call *call, const char *type, const char *name, int offset, int size, int is_signed, int filter_type); diff --git a/include/trace/ftrace.h b/include/trace/ftrace.h index 6055b0604c86..2af2f7a2c1bd 100644 --- a/include/trace/ftrace.h +++ b/include/trace/ftrace.h @@ -436,10 +436,6 @@ ftrace_define_fields_##call(struct ftrace_event_call *event_call) \ struct ftrace_raw_##call field; \ int ret; \ \ - ret = trace_define_common_fields(event_call); \ - if (ret) \ - return ret; \ - \ tstruct; \ \ return ret; \ diff --git a/kernel/trace/trace_events.c b/kernel/trace/trace_events.c index 8ed66e0d476b..97b0b3aa166d 100644 --- a/kernel/trace/trace_events.c +++ b/kernel/trace/trace_events.c @@ -78,7 +78,7 @@ EXPORT_SYMBOL_GPL(trace_define_field); if (ret) \ return ret; -int trace_define_common_fields(struct ftrace_event_call *call) +static int trace_define_common_fields(struct ftrace_event_call *call) { int ret; struct trace_entry ent; @@ -91,7 +91,6 @@ int trace_define_common_fields(struct ftrace_event_call *call) return ret; } -EXPORT_SYMBOL_GPL(trace_define_common_fields); void trace_destroy_fields(struct ftrace_event_call *call) { @@ -927,7 +926,9 @@ event_create_dir(struct ftrace_event_call *call, struct dentry *d_events, id); if (call->define_fields) { - ret = call->define_fields(call); + ret = trace_define_common_fields(call); + if (!ret) + ret = call->define_fields(call); if (ret < 0) { pr_warning("Could not initialize trace point" " events/%s\n", call->name); diff --git a/kernel/trace/trace_export.c b/kernel/trace/trace_export.c index dff8c84ddf17..458e5bfe26d0 100644 --- a/kernel/trace/trace_export.c +++ b/kernel/trace/trace_export.c @@ -184,10 +184,6 @@ ftrace_define_fields_##name(struct ftrace_event_call *event_call) \ struct struct_name field; \ int ret; \ \ - ret = trace_define_common_fields(event_call); \ - if (ret) \ - return ret; \ - \ tstruct; \ \ return ret; \ diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c index aff5f80b59b8..e3c80e925896 100644 --- a/kernel/trace/trace_kprobe.c +++ b/kernel/trace/trace_kprobe.c @@ -1113,10 +1113,6 @@ static int kprobe_event_define_fields(struct ftrace_event_call *event_call) struct kprobe_trace_entry field; struct trace_probe *tp = (struct trace_probe *)event_call->data; - ret = trace_define_common_fields(event_call); - if (!ret) - return ret; - DEFINE_FIELD(unsigned long, ip, FIELD_STRING_IP, 0); DEFINE_FIELD(int, nargs, FIELD_STRING_NARGS, 1); /* Set argument names as fields */ @@ -1131,10 +1127,6 @@ static int kretprobe_event_define_fields(struct ftrace_event_call *event_call) struct kretprobe_trace_entry field; struct trace_probe *tp = (struct trace_probe *)event_call->data; - ret = trace_define_common_fields(event_call); - if (!ret) - return ret; - DEFINE_FIELD(unsigned long, func, FIELD_STRING_FUNC, 0); DEFINE_FIELD(unsigned long, ret_ip, FIELD_STRING_RETIP, 0); DEFINE_FIELD(int, nargs, FIELD_STRING_NARGS, 1); diff --git a/kernel/trace/trace_syscalls.c b/kernel/trace/trace_syscalls.c index 57501d90096a..b957edd0ca3b 100644 --- a/kernel/trace/trace_syscalls.c +++ b/kernel/trace/trace_syscalls.c @@ -217,10 +217,6 @@ int syscall_enter_define_fields(struct ftrace_event_call *call) int i; int offset = offsetof(typeof(trace), args); - ret = trace_define_common_fields(call); - if (ret) - return ret; - ret = trace_define_field(call, SYSCALL_FIELD(int, nr), FILTER_OTHER); if (ret) return ret; @@ -241,10 +237,6 @@ int syscall_exit_define_fields(struct ftrace_event_call *call) struct syscall_trace_exit trace; int ret; - ret = trace_define_common_fields(call); - if (ret) - return ret; - ret = trace_define_field(call, SYSCALL_FIELD(int, nr), FILTER_OTHER); if (ret) return ret; -- cgit v1.2.3-71-gd317 From 3b8e4273814a7f9e9a74ece517d9206fea919aaa Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Tue, 8 Dec 2009 11:14:52 +0800 Subject: tracing: Move a printk out of ftrace_raw_reg_event_foo() Move the printk from each ftrace_raw_reg_event_foo() to its caller ftrace_event_enable_disable(). This avoids each regfunc trace event callbacks to handle a same error report that can be carried from the caller. See how much space this saves: text data bss dec hex filename 5345151 1961864 7103260 14410275 dbe223 vmlinux.o.old 5331487 1961864 7103260 14396611 dbacc3 vmlinux.o Signed-off-by: Li Zefan Acked-by: Steven Rostedt Cc: Jason Baron LKML-Reference: <4B1DC4AC.802@cn.fujitsu.com> [start cmdline record before calling regfunc to avoid lost window of pid to comm resolution] Signed-off-by: Frederic Weisbecker --- include/trace/ftrace.h | 16 ++-------------- kernel/trace/trace_events.c | 20 +++++++++++++++----- kernel/trace/trace_syscalls.c | 10 ++-------- 3 files changed, 19 insertions(+), 27 deletions(-) (limited to 'kernel') diff --git a/include/trace/ftrace.h b/include/trace/ftrace.h index 2af2f7a2c1bd..0c21af85211c 100644 --- a/include/trace/ftrace.h +++ b/include/trace/ftrace.h @@ -555,13 +555,7 @@ static void ftrace_profile_disable_##name(struct ftrace_event_call *unused)\ * * static int ftrace_reg_event_(struct ftrace_event_call *unused) * { - * int ret; - * - * ret = register_trace_(ftrace_event_); - * if (!ret) - * pr_info("event trace: Could not activate trace point " - * "probe to "); - * return ret; + * return register_trace_(ftrace_event_); * } * * static void ftrace_unreg_event_(struct ftrace_event_call *unused) @@ -710,13 +704,7 @@ static void ftrace_raw_event_##call(proto) \ \ static int ftrace_raw_reg_event_##call(struct ftrace_event_call *unused)\ { \ - int ret; \ - \ - ret = register_trace_##call(ftrace_raw_event_##call); \ - if (ret) \ - pr_info("event trace: Could not activate trace point " \ - "probe to %s\n", #call); \ - return ret; \ + return register_trace_##call(ftrace_raw_event_##call); \ } \ \ static void ftrace_raw_unreg_event_##call(struct ftrace_event_call *unused)\ diff --git a/kernel/trace/trace_events.c b/kernel/trace/trace_events.c index 97b0b3aa166d..189b09baf4fb 100644 --- a/kernel/trace/trace_events.c +++ b/kernel/trace/trace_events.c @@ -118,9 +118,11 @@ int trace_event_raw_init(struct ftrace_event_call *call) } EXPORT_SYMBOL_GPL(trace_event_raw_init); -static void ftrace_event_enable_disable(struct ftrace_event_call *call, +static int ftrace_event_enable_disable(struct ftrace_event_call *call, int enable) { + int ret = 0; + switch (enable) { case 0: if (call->enabled) { @@ -131,12 +133,20 @@ static void ftrace_event_enable_disable(struct ftrace_event_call *call, break; case 1: if (!call->enabled) { - call->enabled = 1; tracing_start_cmdline_record(); - call->regfunc(call); + ret = call->regfunc(call); + if (ret) { + tracing_stop_cmdline_record(); + pr_info("event trace: Could not enable event " + "%s\n", call->name); + break; + } + call->enabled = 1; } break; } + + return ret; } static void ftrace_clear_events(void) @@ -415,7 +425,7 @@ event_enable_write(struct file *filp, const char __user *ubuf, size_t cnt, case 0: case 1: mutex_lock(&event_mutex); - ftrace_event_enable_disable(call, val); + ret = ftrace_event_enable_disable(call, val); mutex_unlock(&event_mutex); break; @@ -425,7 +435,7 @@ event_enable_write(struct file *filp, const char __user *ubuf, size_t cnt, *ppos += cnt; - return cnt; + return ret ? ret : cnt; } static ssize_t diff --git a/kernel/trace/trace_syscalls.c b/kernel/trace/trace_syscalls.c index b957edd0ca3b..75289f372dd2 100644 --- a/kernel/trace/trace_syscalls.c +++ b/kernel/trace/trace_syscalls.c @@ -325,10 +325,7 @@ int reg_event_syscall_enter(struct ftrace_event_call *call) mutex_lock(&syscall_trace_lock); if (!sys_refcount_enter) ret = register_trace_sys_enter(ftrace_syscall_enter); - if (ret) { - pr_info("event trace: Could not activate" - "syscall entry trace point"); - } else { + if (!ret) { set_bit(num, enabled_enter_syscalls); sys_refcount_enter++; } @@ -362,10 +359,7 @@ int reg_event_syscall_exit(struct ftrace_event_call *call) mutex_lock(&syscall_trace_lock); if (!sys_refcount_exit) ret = register_trace_sys_exit(ftrace_syscall_exit); - if (ret) { - pr_info("event trace: Could not activate" - "syscall exit trace point"); - } else { + if (!ret) { set_bit(num, enabled_exit_syscalls); sys_refcount_exit++; } -- cgit v1.2.3-71-gd317 From 311d16da575f53c3367099579736c1d233efe0dc Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Tue, 8 Dec 2009 11:15:11 +0800 Subject: ftrace: Return EINVAL when writing invalid val to set_ftrace_filter Currently it doesn't warn user on invald value: # echo nonexist_symbol > set_ftrace_filter or: # echo 'nonexist_symbol:mod:fuse' > set_ftrace_filter Better make it return failure. Signed-off-by: Li Zefan Acked-by: Steven Rostedt LKML-Reference: <4B1DC4BF.2070003@cn.fujitsu.com> Signed-off-by: Frederic Weisbecker --- kernel/trace/ftrace.c | 22 ++++++++++++++++------ 1 file changed, 16 insertions(+), 6 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index e51a1bcb7bed..08a3fb5b3187 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -1724,7 +1724,7 @@ ftrace_match_record(struct dyn_ftrace *rec, char *regex, int len, int type) return ftrace_match(str, regex, len, type); } -static void ftrace_match_records(char *buff, int len, int enable) +static int ftrace_match_records(char *buff, int len, int enable) { unsigned int search_len; struct ftrace_page *pg; @@ -1733,6 +1733,7 @@ static void ftrace_match_records(char *buff, int len, int enable) char *search; int type; int not; + int found = 0; flag = enable ? FTRACE_FL_FILTER : FTRACE_FL_NOTRACE; type = filter_parse_regex(buff, len, &search, ¬); @@ -1750,6 +1751,7 @@ static void ftrace_match_records(char *buff, int len, int enable) rec->flags &= ~flag; else rec->flags |= flag; + found = 1; } /* * Only enable filtering if we have a function that @@ -1759,6 +1761,8 @@ static void ftrace_match_records(char *buff, int len, int enable) ftrace_filtered = 1; } while_for_each_ftrace_rec(); mutex_unlock(&ftrace_lock); + + return found; } static int @@ -1780,7 +1784,7 @@ ftrace_match_module_record(struct dyn_ftrace *rec, char *mod, return 1; } -static void ftrace_match_module_records(char *buff, char *mod, int enable) +static int ftrace_match_module_records(char *buff, char *mod, int enable) { unsigned search_len = 0; struct ftrace_page *pg; @@ -1789,6 +1793,7 @@ static void ftrace_match_module_records(char *buff, char *mod, int enable) char *search = buff; unsigned long flag; int not = 0; + int found = 0; flag = enable ? FTRACE_FL_FILTER : FTRACE_FL_NOTRACE; @@ -1819,12 +1824,15 @@ static void ftrace_match_module_records(char *buff, char *mod, int enable) rec->flags &= ~flag; else rec->flags |= flag; + found = 1; } if (enable && (rec->flags & FTRACE_FL_FILTER)) ftrace_filtered = 1; } while_for_each_ftrace_rec(); mutex_unlock(&ftrace_lock); + + return found; } /* @@ -1853,8 +1861,9 @@ ftrace_mod_callback(char *func, char *cmd, char *param, int enable) if (!strlen(mod)) return -EINVAL; - ftrace_match_module_records(func, mod, enable); - return 0; + if (ftrace_match_module_records(func, mod, enable)) + return 0; + return -EINVAL; } static struct ftrace_func_command ftrace_mod_cmd = { @@ -2151,8 +2160,9 @@ static int ftrace_process_regex(char *buff, int len, int enable) func = strsep(&next, ":"); if (!next) { - ftrace_match_records(func, len, enable); - return 0; + if (ftrace_match_records(func, len, enable)) + return 0; + return ret; } /* command found */ -- cgit v1.2.3-71-gd317 From 313254a9400d388b46150c0f355e216418a2f598 Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Tue, 8 Dec 2009 11:15:30 +0800 Subject: ftrace: Call trace_parser_clear() properly I found a weird behavior: # echo 'fuse:*' > set_ftrace_filter bash: echo: write error: Invalid argument # cat set_ftrace_filter fuse_dev_fasync fuse_dev_poll fuse_copy_do We should call trace_parser_clear() no matter ftrace_process_regex() returns 0 or -errno, otherwise we will actually take the unaccepted records from ftrace_regex_release(). Signed-off-by: Li Zefan Acked-by: Steven Rostedt LKML-Reference: <4B1DC4D2.3000406@cn.fujitsu.com> Signed-off-by: Frederic Weisbecker --- kernel/trace/ftrace.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index 08a3fb5b3187..ff8aecdc6dd6 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -2208,10 +2208,9 @@ ftrace_regex_write(struct file *file, const char __user *ubuf, !trace_parser_cont(parser)) { ret = ftrace_process_regex(parser->buffer, parser->idx, enable); + trace_parser_clear(parser); if (ret) goto out_unlock; - - trace_parser_clear(parser); } ret = read; -- cgit v1.2.3-71-gd317 From 91baf6285be7282cfa487de92f836c50749dffb9 Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Tue, 8 Dec 2009 11:15:45 +0800 Subject: function-graph: Allow writing the same val to set_graph_function # echo 'do_open' > set_graph_function # echo 'do_open' >> set_graph_function bash: echo: write error: Invalid argument Make it valid to write the same value to set_graph_function, which is consistent with set_ftrace_filter interface. Signed-off-by: Li Zefan Acked-by: Steven Rostedt LKML-reference: <4B1DC4E1.1060303@cn.fujitsu.com> Signed-off-by: Frederic Weisbecker --- kernel/trace/ftrace.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index ff8aecdc6dd6..7968762c8167 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -2552,10 +2552,9 @@ ftrace_set_func(unsigned long *array, int *idx, char *buffer) exists = true; break; } - if (!exists) { + if (!exists) array[(*idx)++] = rec->ip; - found = 1; - } + found = 1; } } while_for_each_ftrace_rec(); -- cgit v1.2.3-71-gd317 From fdb372ed4cadbfe9dbba0e932a77d0523682e690 Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Tue, 8 Dec 2009 11:15:59 +0800 Subject: tracing: Use seq file for trace_options Code simplification for reading trace_options. Signed-off-by: Li Zefan Acked-by: Steven Rostedt LKML-reference: <4B1DC4EF.3090106@cn.fujitsu.com> Signed-off-by: Frederic Weisbecker --- kernel/trace/trace.c | 60 +++++++++++++++------------------------------------- 1 file changed, 17 insertions(+), 43 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 88bd9ae2a9ed..a6c41cc63285 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -2316,67 +2316,32 @@ static const struct file_operations tracing_cpumask_fops = { .write = tracing_cpumask_write, }; -static ssize_t -tracing_trace_options_read(struct file *filp, char __user *ubuf, - size_t cnt, loff_t *ppos) +static int tracing_trace_options_show(struct seq_file *m, void *v) { struct tracer_opt *trace_opts; u32 tracer_flags; - int len = 0; - char *buf; - int r = 0; int i; - - /* calculate max size */ - for (i = 0; trace_options[i]; i++) { - len += strlen(trace_options[i]); - len += 3; /* "no" and newline */ - } - mutex_lock(&trace_types_lock); tracer_flags = current_trace->flags->val; trace_opts = current_trace->flags->opts; - /* - * Increase the size with names of options specific - * of the current tracer. - */ - for (i = 0; trace_opts[i].name; i++) { - len += strlen(trace_opts[i].name); - len += 3; /* "no" and newline */ - } - - /* +1 for \0 */ - buf = kmalloc(len + 1, GFP_KERNEL); - if (!buf) { - mutex_unlock(&trace_types_lock); - return -ENOMEM; - } - for (i = 0; trace_options[i]; i++) { if (trace_flags & (1 << i)) - r += sprintf(buf + r, "%s\n", trace_options[i]); + seq_printf(m, "%s\n", trace_options[i]); else - r += sprintf(buf + r, "no%s\n", trace_options[i]); + seq_printf(m, "no%s\n", trace_options[i]); } for (i = 0; trace_opts[i].name; i++) { if (tracer_flags & trace_opts[i].bit) - r += sprintf(buf + r, "%s\n", - trace_opts[i].name); + seq_printf(m, "%s\n", trace_opts[i].name); else - r += sprintf(buf + r, "no%s\n", - trace_opts[i].name); + seq_printf(m, "no%s\n", trace_opts[i].name); } mutex_unlock(&trace_types_lock); - WARN_ON(r >= len + 1); - - r = simple_read_from_buffer(ubuf, cnt, ppos, buf, r); - - kfree(buf); - return r; + return 0; } /* Try to assign a tracer specific option */ @@ -2471,9 +2436,18 @@ tracing_trace_options_write(struct file *filp, const char __user *ubuf, return cnt; } +static int tracing_trace_options_open(struct inode *inode, struct file *file) +{ + if (tracing_disabled) + return -ENODEV; + return single_open(file, tracing_trace_options_show, NULL); +} + static const struct file_operations tracing_iter_fops = { - .open = tracing_open_generic, - .read = tracing_trace_options_read, + .open = tracing_trace_options_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, .write = tracing_trace_options_write, }; -- cgit v1.2.3-71-gd317 From 13f16d209161c95e92aef40e350cc6cf56ac440b Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Tue, 8 Dec 2009 11:16:11 +0800 Subject: tracing: Use seq file for trace_clock The buffer for the output is as small as 64 bytes, so it'll overflow if we add more clock type. Use seq file instead. Signed-off-by: Li Zefan Acked-by: Steven Rostedt LKML-Reference: <4B1DC4FB.5030407@cn.fujitsu.com> Signed-off-by: Frederic Weisbecker --- kernel/trace/trace.c | 24 +++++++++++++++--------- 1 file changed, 15 insertions(+), 9 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index a6c41cc63285..886268e0d8ee 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -3366,21 +3366,18 @@ tracing_mark_write(struct file *filp, const char __user *ubuf, return cnt; } -static ssize_t tracing_clock_read(struct file *filp, char __user *ubuf, - size_t cnt, loff_t *ppos) +static int tracing_clock_show(struct seq_file *m, void *v) { - char buf[64]; - int bufiter = 0; int i; for (i = 0; i < ARRAY_SIZE(trace_clocks); i++) - bufiter += snprintf(buf + bufiter, sizeof(buf) - bufiter, + seq_printf(m, "%s%s%s%s", i ? " " : "", i == trace_clock_id ? "[" : "", trace_clocks[i].name, i == trace_clock_id ? "]" : ""); - bufiter += snprintf(buf + bufiter, sizeof(buf) - bufiter, "\n"); + seq_putc(m, '\n'); - return simple_read_from_buffer(ubuf, cnt, ppos, buf, bufiter); + return 0; } static ssize_t tracing_clock_write(struct file *filp, const char __user *ubuf, @@ -3422,6 +3419,13 @@ static ssize_t tracing_clock_write(struct file *filp, const char __user *ubuf, return cnt; } +static int tracing_clock_open(struct inode *inode, struct file *file) +{ + if (tracing_disabled) + return -ENODEV; + return single_open(file, tracing_clock_show, NULL); +} + static const struct file_operations tracing_max_lat_fops = { .open = tracing_open_generic, .read = tracing_max_lat_read, @@ -3460,8 +3464,10 @@ static const struct file_operations tracing_mark_fops = { }; static const struct file_operations trace_clock_fops = { - .open = tracing_open_generic, - .read = tracing_clock_read, + .open = tracing_clock_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, .write = tracing_clock_write, }; -- cgit v1.2.3-71-gd317 From 2cbafd68b826f8e0471875cf33cdfb8a1478aef1 Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Tue, 8 Dec 2009 11:16:26 +0800 Subject: tracing: Remove useless trace option Since commit 4d9493c90f8e6e1b164aede3814010a290161abb ("ftrace: remove add-hoc code"), option "sched-tree" has become useless. Signed-off-by: Li Zefan Acked-by: Steven Rostedt LKML-Reference: <4B1DC50A.7040402@cn.fujitsu.com> Signed-off-by: Frederic Weisbecker --- kernel/trace/trace.c | 1 - kernel/trace/trace.h | 23 +++++++++++------------ 2 files changed, 11 insertions(+), 13 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 886268e0d8ee..898409d60422 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -313,7 +313,6 @@ static const char *trace_options[] = { "bin", "block", "stacktrace", - "sched-tree", "trace_printk", "ftrace_preempt", "branch", diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h index 7fa33cab6962..1b18cb240c16 100644 --- a/kernel/trace/trace.h +++ b/kernel/trace/trace.h @@ -597,18 +597,17 @@ enum trace_iterator_flags { TRACE_ITER_BIN = 0x40, TRACE_ITER_BLOCK = 0x80, TRACE_ITER_STACKTRACE = 0x100, - TRACE_ITER_SCHED_TREE = 0x200, - TRACE_ITER_PRINTK = 0x400, - TRACE_ITER_PREEMPTONLY = 0x800, - TRACE_ITER_BRANCH = 0x1000, - TRACE_ITER_ANNOTATE = 0x2000, - TRACE_ITER_USERSTACKTRACE = 0x4000, - TRACE_ITER_SYM_USEROBJ = 0x8000, - TRACE_ITER_PRINTK_MSGONLY = 0x10000, - TRACE_ITER_CONTEXT_INFO = 0x20000, /* Print pid/cpu/time */ - TRACE_ITER_LATENCY_FMT = 0x40000, - TRACE_ITER_SLEEP_TIME = 0x80000, - TRACE_ITER_GRAPH_TIME = 0x100000, + TRACE_ITER_PRINTK = 0x200, + TRACE_ITER_PREEMPTONLY = 0x400, + TRACE_ITER_BRANCH = 0x800, + TRACE_ITER_ANNOTATE = 0x1000, + TRACE_ITER_USERSTACKTRACE = 0x2000, + TRACE_ITER_SYM_USEROBJ = 0x4000, + TRACE_ITER_PRINTK_MSGONLY = 0x8000, + TRACE_ITER_CONTEXT_INFO = 0x10000, /* Print pid/cpu/time */ + TRACE_ITER_LATENCY_FMT = 0x20000, + TRACE_ITER_SLEEP_TIME = 0x40000, + TRACE_ITER_GRAPH_TIME = 0x80000, }; /* -- cgit v1.2.3-71-gd317 From 8d18eaaff5acaa58369be342c86e607643ce10c7 Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Tue, 8 Dec 2009 11:17:06 +0800 Subject: tracing: Simplify trace_option_write() - remove duplicate code inside trace_options_write() - extract duplicate code in trace_options_write() and set_tracer_option() Signed-off-by: Li Zefan Acked-by: Steven Rostedt LKML-Reference: <4B1DC532.9010802@cn.fujitsu.com> Signed-off-by: Frederic Weisbecker --- kernel/trace/trace.c | 85 +++++++++++++++++++--------------------------------- 1 file changed, 31 insertions(+), 54 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 898409d60422..05076008f371 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -2343,38 +2343,39 @@ static int tracing_trace_options_show(struct seq_file *m, void *v) return 0; } +static int __set_tracer_option(struct tracer *trace, + struct tracer_flags *tracer_flags, + struct tracer_opt *opts, int neg) +{ + int ret; + + ret = trace->set_flag(tracer_flags->val, opts->bit, !neg); + if (ret) + return ret; + + if (neg) + tracer_flags->val &= ~opts->bit; + else + tracer_flags->val |= opts->bit; + return 0; +} + /* Try to assign a tracer specific option */ static int set_tracer_option(struct tracer *trace, char *cmp, int neg) { struct tracer_flags *tracer_flags = trace->flags; struct tracer_opt *opts = NULL; - int ret = 0, i = 0; - int len; + int i; for (i = 0; tracer_flags->opts[i].name; i++) { opts = &tracer_flags->opts[i]; - len = strlen(opts->name); - if (strncmp(cmp, opts->name, len) == 0) { - ret = trace->set_flag(tracer_flags->val, - opts->bit, !neg); - break; - } + if (strcmp(cmp, opts->name) == 0) + return __set_tracer_option(trace, trace->flags, + opts, neg); } - /* Not found */ - if (!tracer_flags->opts[i].name) - return -EINVAL; - /* Refused to handle */ - if (ret) - return ret; - - if (neg) - tracer_flags->val &= ~opts->bit; - else - tracer_flags->val |= opts->bit; - - return 0; + return -EINVAL; } static void set_tracer_flags(unsigned int mask, int enabled) @@ -2394,7 +2395,7 @@ tracing_trace_options_write(struct file *filp, const char __user *ubuf, size_t cnt, loff_t *ppos) { char buf[64]; - char *cmp = buf; + char *cmp; int neg = 0; int ret; int i; @@ -2406,16 +2407,15 @@ tracing_trace_options_write(struct file *filp, const char __user *ubuf, return -EFAULT; buf[cnt] = 0; + cmp = strstrip(buf); - if (strncmp(buf, "no", 2) == 0) { + if (strncmp(cmp, "no", 2) == 0) { neg = 1; cmp += 2; } for (i = 0; trace_options[i]; i++) { - int len = strlen(trace_options[i]); - - if (strncmp(cmp, trace_options[i], len) == 0) { + if (strcmp(cmp, trace_options[i]) == 0) { set_tracer_flags(1 << i, !neg); break; } @@ -3927,39 +3927,16 @@ trace_options_write(struct file *filp, const char __user *ubuf, size_t cnt, if (ret < 0) return ret; - ret = 0; - switch (val) { - case 0: - /* do nothing if already cleared */ - if (!(topt->flags->val & topt->opt->bit)) - break; - - mutex_lock(&trace_types_lock); - if (current_trace->set_flag) - ret = current_trace->set_flag(topt->flags->val, - topt->opt->bit, 0); - mutex_unlock(&trace_types_lock); - if (ret) - return ret; - topt->flags->val &= ~topt->opt->bit; - break; - case 1: - /* do nothing if already set */ - if (topt->flags->val & topt->opt->bit) - break; + if (val != 0 && val != 1) + return -EINVAL; + if (!!(topt->flags->val & topt->opt->bit) != val) { mutex_lock(&trace_types_lock); - if (current_trace->set_flag) - ret = current_trace->set_flag(topt->flags->val, - topt->opt->bit, 1); + ret = __set_tracer_option(current_trace, topt->flags, + topt->opt, val); mutex_unlock(&trace_types_lock); if (ret) return ret; - topt->flags->val |= topt->opt->bit; - break; - - default: - return -EINVAL; } *ppos += cnt; -- cgit v1.2.3-71-gd317 From e00bf2ec60605eb95687b7a0c3b83c87c48541dc Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Tue, 8 Dec 2009 11:17:29 +0800 Subject: tracing: Change event->profile_count to be int type Like total_profile_count, struct ftrace_event_call::profile_count is protected by event_mutex, so it doesn't need to be atomic_t. Signed-off-by: Li Zefan Acked-by: Steven Rostedt Cc: Jason Baron Cc: Masami Hiramatsu Cc: Peter Zijlstra LKML-Reference: <4B1DC549.5010705@cn.fujitsu.com> Signed-off-by: Frederic Weisbecker --- include/linux/ftrace_event.h | 2 +- include/linux/syscalls.h | 2 -- include/trace/ftrace.h | 1 - kernel/trace/trace_event_profile.c | 6 +++--- kernel/trace/trace_kprobe.c | 1 - 5 files changed, 4 insertions(+), 8 deletions(-) (limited to 'kernel') diff --git a/include/linux/ftrace_event.h b/include/linux/ftrace_event.h index db97c64ce0e9..2233c98d80df 100644 --- a/include/linux/ftrace_event.h +++ b/include/linux/ftrace_event.h @@ -131,7 +131,7 @@ struct ftrace_event_call { void *mod; void *data; - atomic_t profile_count; + int profile_count; int (*profile_enable)(struct ftrace_event_call *); void (*profile_disable)(struct ftrace_event_call *); }; diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h index 94ac28437bef..72d69860d901 100644 --- a/include/linux/syscalls.h +++ b/include/linux/syscalls.h @@ -102,12 +102,10 @@ struct perf_event_attr; #ifdef CONFIG_EVENT_PROFILE #define TRACE_SYS_ENTER_PROFILE_INIT(sname) \ - .profile_count = ATOMIC_INIT(-1), \ .profile_enable = prof_sysenter_enable, \ .profile_disable = prof_sysenter_disable, #define TRACE_SYS_EXIT_PROFILE_INIT(sname) \ - .profile_count = ATOMIC_INIT(-1), \ .profile_enable = prof_sysexit_enable, \ .profile_disable = prof_sysexit_disable, #else diff --git a/include/trace/ftrace.h b/include/trace/ftrace.h index 0c21af85211c..73523151a731 100644 --- a/include/trace/ftrace.h +++ b/include/trace/ftrace.h @@ -629,7 +629,6 @@ static void ftrace_profile_disable_##name(struct ftrace_event_call *unused)\ #ifdef CONFIG_EVENT_PROFILE #define _TRACE_PROFILE_INIT(call) \ - .profile_count = ATOMIC_INIT(-1), \ .profile_enable = ftrace_profile_enable_##call, \ .profile_disable = ftrace_profile_disable_##call, diff --git a/kernel/trace/trace_event_profile.c b/kernel/trace/trace_event_profile.c index d9c60f80aa0d..9e25573242cf 100644 --- a/kernel/trace/trace_event_profile.c +++ b/kernel/trace/trace_event_profile.c @@ -25,7 +25,7 @@ static int ftrace_profile_enable_event(struct ftrace_event_call *event) char *buf; int ret = -ENOMEM; - if (atomic_inc_return(&event->profile_count)) + if (event->profile_count++ > 0) return 0; if (!total_profile_count) { @@ -56,7 +56,7 @@ fail_buf_nmi: perf_trace_buf = NULL; } fail_buf: - atomic_dec(&event->profile_count); + event->profile_count--; return ret; } @@ -83,7 +83,7 @@ static void ftrace_profile_disable_event(struct ftrace_event_call *event) { char *buf, *nmi_buf; - if (!atomic_add_negative(-1, &event->profile_count)) + if (--event->profile_count > 0) return; event->profile_disable(event); diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c index e3c80e925896..6ed223447a3f 100644 --- a/kernel/trace/trace_kprobe.c +++ b/kernel/trace/trace_kprobe.c @@ -1426,7 +1426,6 @@ static int register_probe_event(struct trace_probe *tp) call->unregfunc = probe_event_disable; #ifdef CONFIG_EVENT_PROFILE - atomic_set(&call->profile_count, -1); call->profile_enable = probe_profile_enable; call->profile_disable = probe_profile_disable; #endif -- cgit v1.2.3-71-gd317 From 472bbe02c92a7a8299d7b16946277d98bb8f4bb7 Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Tue, 8 Dec 2009 11:17:51 +0800 Subject: tracing/power: Remove two exports trace_power_start and trace_power_end are used in arch/x86/kernel/power.c, and this file can't be compiled as a module, so these two tracepoints don't need to be exported. Signed-off-by: Li Zefan Acked-by: Arjan van de Ven Acked-by: Steven Rostedt LKML-Reference: <4B1DC55F.7060305@cn.fujitsu.com> Signed-off-by: Frederic Weisbecker --- kernel/trace/power-traces.c | 2 -- 1 file changed, 2 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/power-traces.c b/kernel/trace/power-traces.c index e06c6e3d56a3..9f4f565b01e6 100644 --- a/kernel/trace/power-traces.c +++ b/kernel/trace/power-traces.c @@ -14,7 +14,5 @@ #define CREATE_TRACE_POINTS #include -EXPORT_TRACEPOINT_SYMBOL_GPL(power_start); -EXPORT_TRACEPOINT_SYMBOL_GPL(power_end); EXPORT_TRACEPOINT_SYMBOL_GPL(power_frequency); -- cgit v1.2.3-71-gd317 From 16620e0f1990fa6d896a639449c4b3d678458464 Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Tue, 8 Dec 2009 11:18:16 +0800 Subject: ksym_tracer: Fix bad cast Fix this warning: kernel/trace/trace_ksym.c: In function 'ksym_trace_filter_read': kernel/trace/trace_ksym.c:239: warning: cast to pointer from integer of different size Signed-off-by: Li Zefan Acked-by: Steven Rostedt Cc: "K.Prasad" LKML-Reference: <4B1DC578.9020909@cn.fujitsu.com> [remove the strstrip fix as tglx already fixed that] Signed-off-by: Frederic Weisbecker --- kernel/trace/trace_ksym.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/trace_ksym.c b/kernel/trace/trace_ksym.c index 64e7a5bd6692..48f1c6c248c6 100644 --- a/kernel/trace/trace_ksym.c +++ b/kernel/trace/trace_ksym.c @@ -235,7 +235,8 @@ static ssize_t ksym_trace_filter_read(struct file *filp, char __user *ubuf, mutex_lock(&ksym_tracer_mutex); hlist_for_each_entry(entry, node, &ksym_filter_head, ksym_hlist) { - ret = trace_seq_printf(s, "%pS:", (void *)entry->attr.bp_addr); + ret = trace_seq_printf(s, "%pS:", + (void *)(unsigned long)entry->attr.bp_addr); if (entry->attr.bp_type == HW_BREAKPOINT_R) ret = trace_seq_puts(s, "r--\n"); else if (entry->attr.bp_type == HW_BREAKPOINT_W) @@ -298,8 +299,8 @@ static ssize_t ksym_trace_filter_write(struct file *file, * 2: echo 0 > ksym_trace_filter * 3: echo "*:---" > ksym_trace_filter */ - if (!input_string[0] || !strcmp(input_string, "0") || - !strcmp(input_string, "*:---")) { + if (!buf[0] || !strcmp(buf, "0") || + !strcmp(buf, "*:---")) { __ksym_trace_reset(); ret = 0; goto out; -- cgit v1.2.3-71-gd317 From b7b40ade58e621851896aa261452df99d4e9d99b Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Mon, 9 Nov 2009 21:01:59 +0100 Subject: locking: Reorder functions in spinlock.c Separate spin_lock and rw_lock functions. Preempt-RT needs to exclude the rw_lock functions from being compiled. The reordering allows to do that with a single #ifdef. No functional change. Signed-off-by: Thomas Gleixner Acked-by: Peter Zijlstra Acked-by: Ingo Molnar --- kernel/spinlock.c | 246 +++++++++++++++++++++++++++--------------------------- 1 file changed, 123 insertions(+), 123 deletions(-) (limited to 'kernel') diff --git a/kernel/spinlock.c b/kernel/spinlock.c index 41e042219ff6..e6e136318437 100644 --- a/kernel/spinlock.c +++ b/kernel/spinlock.c @@ -113,41 +113,6 @@ BUILD_LOCK_OPS(write, rwlock); #endif -#ifdef CONFIG_DEBUG_LOCK_ALLOC - -void __lockfunc _spin_lock_nested(spinlock_t *lock, int subclass) -{ - preempt_disable(); - spin_acquire(&lock->dep_map, subclass, 0, _RET_IP_); - LOCK_CONTENDED(lock, _raw_spin_trylock, _raw_spin_lock); -} -EXPORT_SYMBOL(_spin_lock_nested); - -unsigned long __lockfunc _spin_lock_irqsave_nested(spinlock_t *lock, - int subclass) -{ - unsigned long flags; - - local_irq_save(flags); - preempt_disable(); - spin_acquire(&lock->dep_map, subclass, 0, _RET_IP_); - LOCK_CONTENDED_FLAGS(lock, _raw_spin_trylock, _raw_spin_lock, - _raw_spin_lock_flags, &flags); - return flags; -} -EXPORT_SYMBOL(_spin_lock_irqsave_nested); - -void __lockfunc _spin_lock_nest_lock(spinlock_t *lock, - struct lockdep_map *nest_lock) -{ - preempt_disable(); - spin_acquire_nest(&lock->dep_map, 0, 0, nest_lock, _RET_IP_); - LOCK_CONTENDED(lock, _raw_spin_trylock, _raw_spin_lock); -} -EXPORT_SYMBOL(_spin_lock_nest_lock); - -#endif - #ifndef CONFIG_INLINE_SPIN_TRYLOCK int __lockfunc _spin_trylock(spinlock_t *lock) { @@ -156,28 +121,20 @@ int __lockfunc _spin_trylock(spinlock_t *lock) EXPORT_SYMBOL(_spin_trylock); #endif -#ifndef CONFIG_INLINE_READ_TRYLOCK -int __lockfunc _read_trylock(rwlock_t *lock) -{ - return __read_trylock(lock); -} -EXPORT_SYMBOL(_read_trylock); -#endif - -#ifndef CONFIG_INLINE_WRITE_TRYLOCK -int __lockfunc _write_trylock(rwlock_t *lock) +#ifndef CONFIG_INLINE_SPIN_TRYLOCK_BH +int __lockfunc _spin_trylock_bh(spinlock_t *lock) { - return __write_trylock(lock); + return __spin_trylock_bh(lock); } -EXPORT_SYMBOL(_write_trylock); +EXPORT_SYMBOL(_spin_trylock_bh); #endif -#ifndef CONFIG_INLINE_READ_LOCK -void __lockfunc _read_lock(rwlock_t *lock) +#ifndef CONFIG_INLINE_SPIN_LOCK +void __lockfunc _spin_lock(spinlock_t *lock) { - __read_lock(lock); + __spin_lock(lock); } -EXPORT_SYMBOL(_read_lock); +EXPORT_SYMBOL(_spin_lock); #endif #ifndef CONFIG_INLINE_SPIN_LOCK_IRQSAVE @@ -204,84 +161,76 @@ void __lockfunc _spin_lock_bh(spinlock_t *lock) EXPORT_SYMBOL(_spin_lock_bh); #endif -#ifndef CONFIG_INLINE_READ_LOCK_IRQSAVE -unsigned long __lockfunc _read_lock_irqsave(rwlock_t *lock) -{ - return __read_lock_irqsave(lock); -} -EXPORT_SYMBOL(_read_lock_irqsave); -#endif - -#ifndef CONFIG_INLINE_READ_LOCK_IRQ -void __lockfunc _read_lock_irq(rwlock_t *lock) +#ifndef CONFIG_INLINE_SPIN_UNLOCK +void __lockfunc _spin_unlock(spinlock_t *lock) { - __read_lock_irq(lock); + __spin_unlock(lock); } -EXPORT_SYMBOL(_read_lock_irq); +EXPORT_SYMBOL(_spin_unlock); #endif -#ifndef CONFIG_INLINE_READ_LOCK_BH -void __lockfunc _read_lock_bh(rwlock_t *lock) +#ifndef CONFIG_INLINE_SPIN_UNLOCK_IRQRESTORE +void __lockfunc _spin_unlock_irqrestore(spinlock_t *lock, unsigned long flags) { - __read_lock_bh(lock); + __spin_unlock_irqrestore(lock, flags); } -EXPORT_SYMBOL(_read_lock_bh); +EXPORT_SYMBOL(_spin_unlock_irqrestore); #endif -#ifndef CONFIG_INLINE_WRITE_LOCK_IRQSAVE -unsigned long __lockfunc _write_lock_irqsave(rwlock_t *lock) +#ifndef CONFIG_INLINE_SPIN_UNLOCK_IRQ +void __lockfunc _spin_unlock_irq(spinlock_t *lock) { - return __write_lock_irqsave(lock); + __spin_unlock_irq(lock); } -EXPORT_SYMBOL(_write_lock_irqsave); +EXPORT_SYMBOL(_spin_unlock_irq); #endif -#ifndef CONFIG_INLINE_WRITE_LOCK_IRQ -void __lockfunc _write_lock_irq(rwlock_t *lock) +#ifndef CONFIG_INLINE_SPIN_UNLOCK_BH +void __lockfunc _spin_unlock_bh(spinlock_t *lock) { - __write_lock_irq(lock); + __spin_unlock_bh(lock); } -EXPORT_SYMBOL(_write_lock_irq); +EXPORT_SYMBOL(_spin_unlock_bh); #endif -#ifndef CONFIG_INLINE_WRITE_LOCK_BH -void __lockfunc _write_lock_bh(rwlock_t *lock) +#ifndef CONFIG_INLINE_READ_TRYLOCK +int __lockfunc _read_trylock(rwlock_t *lock) { - __write_lock_bh(lock); + return __read_trylock(lock); } -EXPORT_SYMBOL(_write_lock_bh); +EXPORT_SYMBOL(_read_trylock); #endif -#ifndef CONFIG_INLINE_SPIN_LOCK -void __lockfunc _spin_lock(spinlock_t *lock) +#ifndef CONFIG_INLINE_READ_LOCK +void __lockfunc _read_lock(rwlock_t *lock) { - __spin_lock(lock); + __read_lock(lock); } -EXPORT_SYMBOL(_spin_lock); +EXPORT_SYMBOL(_read_lock); #endif -#ifndef CONFIG_INLINE_WRITE_LOCK -void __lockfunc _write_lock(rwlock_t *lock) +#ifndef CONFIG_INLINE_READ_LOCK_IRQSAVE +unsigned long __lockfunc _read_lock_irqsave(rwlock_t *lock) { - __write_lock(lock); + return __read_lock_irqsave(lock); } -EXPORT_SYMBOL(_write_lock); +EXPORT_SYMBOL(_read_lock_irqsave); #endif -#ifndef CONFIG_INLINE_SPIN_UNLOCK -void __lockfunc _spin_unlock(spinlock_t *lock) +#ifndef CONFIG_INLINE_READ_LOCK_IRQ +void __lockfunc _read_lock_irq(rwlock_t *lock) { - __spin_unlock(lock); + __read_lock_irq(lock); } -EXPORT_SYMBOL(_spin_unlock); +EXPORT_SYMBOL(_read_lock_irq); #endif -#ifndef CONFIG_INLINE_WRITE_UNLOCK -void __lockfunc _write_unlock(rwlock_t *lock) +#ifndef CONFIG_INLINE_READ_LOCK_BH +void __lockfunc _read_lock_bh(rwlock_t *lock) { - __write_unlock(lock); + __read_lock_bh(lock); } -EXPORT_SYMBOL(_write_unlock); +EXPORT_SYMBOL(_read_lock_bh); #endif #ifndef CONFIG_INLINE_READ_UNLOCK @@ -292,30 +241,6 @@ void __lockfunc _read_unlock(rwlock_t *lock) EXPORT_SYMBOL(_read_unlock); #endif -#ifndef CONFIG_INLINE_SPIN_UNLOCK_IRQRESTORE -void __lockfunc _spin_unlock_irqrestore(spinlock_t *lock, unsigned long flags) -{ - __spin_unlock_irqrestore(lock, flags); -} -EXPORT_SYMBOL(_spin_unlock_irqrestore); -#endif - -#ifndef CONFIG_INLINE_SPIN_UNLOCK_IRQ -void __lockfunc _spin_unlock_irq(spinlock_t *lock) -{ - __spin_unlock_irq(lock); -} -EXPORT_SYMBOL(_spin_unlock_irq); -#endif - -#ifndef CONFIG_INLINE_SPIN_UNLOCK_BH -void __lockfunc _spin_unlock_bh(spinlock_t *lock) -{ - __spin_unlock_bh(lock); -} -EXPORT_SYMBOL(_spin_unlock_bh); -#endif - #ifndef CONFIG_INLINE_READ_UNLOCK_IRQRESTORE void __lockfunc _read_unlock_irqrestore(rwlock_t *lock, unsigned long flags) { @@ -340,6 +265,54 @@ void __lockfunc _read_unlock_bh(rwlock_t *lock) EXPORT_SYMBOL(_read_unlock_bh); #endif +#ifndef CONFIG_INLINE_WRITE_TRYLOCK +int __lockfunc _write_trylock(rwlock_t *lock) +{ + return __write_trylock(lock); +} +EXPORT_SYMBOL(_write_trylock); +#endif + +#ifndef CONFIG_INLINE_WRITE_LOCK +void __lockfunc _write_lock(rwlock_t *lock) +{ + __write_lock(lock); +} +EXPORT_SYMBOL(_write_lock); +#endif + +#ifndef CONFIG_INLINE_WRITE_LOCK_IRQSAVE +unsigned long __lockfunc _write_lock_irqsave(rwlock_t *lock) +{ + return __write_lock_irqsave(lock); +} +EXPORT_SYMBOL(_write_lock_irqsave); +#endif + +#ifndef CONFIG_INLINE_WRITE_LOCK_IRQ +void __lockfunc _write_lock_irq(rwlock_t *lock) +{ + __write_lock_irq(lock); +} +EXPORT_SYMBOL(_write_lock_irq); +#endif + +#ifndef CONFIG_INLINE_WRITE_LOCK_BH +void __lockfunc _write_lock_bh(rwlock_t *lock) +{ + __write_lock_bh(lock); +} +EXPORT_SYMBOL(_write_lock_bh); +#endif + +#ifndef CONFIG_INLINE_WRITE_UNLOCK +void __lockfunc _write_unlock(rwlock_t *lock) +{ + __write_unlock(lock); +} +EXPORT_SYMBOL(_write_unlock); +#endif + #ifndef CONFIG_INLINE_WRITE_UNLOCK_IRQRESTORE void __lockfunc _write_unlock_irqrestore(rwlock_t *lock, unsigned long flags) { @@ -364,12 +337,39 @@ void __lockfunc _write_unlock_bh(rwlock_t *lock) EXPORT_SYMBOL(_write_unlock_bh); #endif -#ifndef CONFIG_INLINE_SPIN_TRYLOCK_BH -int __lockfunc _spin_trylock_bh(spinlock_t *lock) +#ifdef CONFIG_DEBUG_LOCK_ALLOC + +void __lockfunc _spin_lock_nested(spinlock_t *lock, int subclass) { - return __spin_trylock_bh(lock); + preempt_disable(); + spin_acquire(&lock->dep_map, subclass, 0, _RET_IP_); + LOCK_CONTENDED(lock, _raw_spin_trylock, _raw_spin_lock); } -EXPORT_SYMBOL(_spin_trylock_bh); +EXPORT_SYMBOL(_spin_lock_nested); + +unsigned long __lockfunc _spin_lock_irqsave_nested(spinlock_t *lock, + int subclass) +{ + unsigned long flags; + + local_irq_save(flags); + preempt_disable(); + spin_acquire(&lock->dep_map, subclass, 0, _RET_IP_); + LOCK_CONTENDED_FLAGS(lock, _raw_spin_trylock, _raw_spin_lock, + _raw_spin_lock_flags, &flags); + return flags; +} +EXPORT_SYMBOL(_spin_lock_irqsave_nested); + +void __lockfunc _spin_lock_nest_lock(spinlock_t *lock, + struct lockdep_map *nest_lock) +{ + preempt_disable(); + spin_acquire_nest(&lock->dep_map, 0, 0, nest_lock, _RET_IP_); + LOCK_CONTENDED(lock, _raw_spin_trylock, _raw_spin_lock); +} +EXPORT_SYMBOL(_spin_lock_nest_lock); + #endif notrace int in_lock_functions(unsigned long addr) -- cgit v1.2.3-71-gd317 From 445c89514be242b1b0080056d50bdc1b72adeb5c Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Wed, 2 Dec 2009 19:49:50 +0100 Subject: locking: Convert raw_spinlock to arch_spinlock The raw_spin* namespace was taken by lockdep for the architecture specific implementations. raw_spin_* would be the ideal name space for the spinlocks which are not converted to sleeping locks in preempt-rt. Linus suggested to convert the raw_ to arch_ locks and cleanup the name space instead of using an artifical name like core_spin, atomic_spin or whatever No functional change. Signed-off-by: Thomas Gleixner Acked-by: Peter Zijlstra Acked-by: David S. Miller Acked-by: Ingo Molnar Cc: linux-arch@vger.kernel.org --- arch/alpha/include/asm/spinlock.h | 6 +++--- arch/alpha/include/asm/spinlock_types.h | 2 +- arch/arm/include/asm/spinlock.h | 6 +++--- arch/arm/include/asm/spinlock_types.h | 2 +- arch/blackfin/include/asm/spinlock.h | 10 +++++----- arch/blackfin/include/asm/spinlock_types.h | 2 +- arch/cris/include/arch-v32/arch/spinlock.h | 12 ++++++------ arch/ia64/include/asm/spinlock.h | 26 +++++++++++++------------- arch/ia64/include/asm/spinlock_types.h | 2 +- arch/m32r/include/asm/spinlock.h | 6 +++--- arch/m32r/include/asm/spinlock_types.h | 2 +- arch/mips/include/asm/spinlock.h | 10 +++++----- arch/mips/include/asm/spinlock_types.h | 2 +- arch/parisc/include/asm/atomic.h | 6 +++--- arch/parisc/include/asm/spinlock.h | 8 ++++---- arch/parisc/include/asm/spinlock_types.h | 4 ++-- arch/parisc/lib/bitops.c | 2 +- arch/powerpc/include/asm/rtas.h | 2 +- arch/powerpc/include/asm/spinlock.h | 14 +++++++------- arch/powerpc/include/asm/spinlock_types.h | 2 +- arch/powerpc/kernel/rtas.c | 2 +- arch/powerpc/lib/locks.c | 4 ++-- arch/powerpc/platforms/pasemi/setup.c | 2 +- arch/s390/include/asm/spinlock.h | 16 ++++++++-------- arch/s390/include/asm/spinlock_types.h | 2 +- arch/s390/lib/spinlock.c | 8 ++++---- arch/sh/include/asm/spinlock.h | 6 +++--- arch/sh/include/asm/spinlock_types.h | 2 +- arch/sparc/include/asm/spinlock_32.h | 6 +++--- arch/sparc/include/asm/spinlock_64.h | 8 ++++---- arch/sparc/include/asm/spinlock_types.h | 2 +- arch/x86/include/asm/paravirt.h | 12 ++++++------ arch/x86/include/asm/paravirt_types.h | 14 +++++++------- arch/x86/include/asm/spinlock.h | 30 +++++++++++++++--------------- arch/x86/include/asm/spinlock_types.h | 4 ++-- arch/x86/kernel/dumpstack.c | 2 +- arch/x86/kernel/paravirt-spinlocks.c | 2 +- arch/x86/kernel/tsc_sync.c | 2 +- arch/x86/xen/spinlock.c | 16 ++++++++-------- include/asm-generic/bitops/atomic.h | 6 +++--- include/linux/spinlock.h | 4 ++-- include/linux/spinlock_types.h | 2 +- include/linux/spinlock_types_up.h | 4 ++-- include/linux/spinlock_up.h | 8 ++++---- kernel/lockdep.c | 2 +- kernel/trace/ring_buffer.c | 4 ++-- kernel/trace/trace.c | 18 +++++++++--------- kernel/trace/trace_clock.c | 4 ++-- kernel/trace/trace_sched_wakeup.c | 4 ++-- kernel/trace/trace_stack.c | 4 ++-- lib/spinlock_debug.c | 2 +- 51 files changed, 164 insertions(+), 164 deletions(-) (limited to 'kernel') diff --git a/arch/alpha/include/asm/spinlock.h b/arch/alpha/include/asm/spinlock.h index e38fb95cb335..bdb26a1940b4 100644 --- a/arch/alpha/include/asm/spinlock.h +++ b/arch/alpha/include/asm/spinlock.h @@ -17,13 +17,13 @@ #define __raw_spin_unlock_wait(x) \ do { cpu_relax(); } while ((x)->lock) -static inline void __raw_spin_unlock(raw_spinlock_t * lock) +static inline void __raw_spin_unlock(arch_spinlock_t * lock) { mb(); lock->lock = 0; } -static inline void __raw_spin_lock(raw_spinlock_t * lock) +static inline void __raw_spin_lock(arch_spinlock_t * lock) { long tmp; @@ -43,7 +43,7 @@ static inline void __raw_spin_lock(raw_spinlock_t * lock) : "m"(lock->lock) : "memory"); } -static inline int __raw_spin_trylock(raw_spinlock_t *lock) +static inline int __raw_spin_trylock(arch_spinlock_t *lock) { return !test_and_set_bit(0, &lock->lock); } diff --git a/arch/alpha/include/asm/spinlock_types.h b/arch/alpha/include/asm/spinlock_types.h index 8141eb5ebf0d..bb94a51e53d2 100644 --- a/arch/alpha/include/asm/spinlock_types.h +++ b/arch/alpha/include/asm/spinlock_types.h @@ -7,7 +7,7 @@ typedef struct { volatile unsigned int lock; -} raw_spinlock_t; +} arch_spinlock_t; #define __RAW_SPIN_LOCK_UNLOCKED { 0 } diff --git a/arch/arm/include/asm/spinlock.h b/arch/arm/include/asm/spinlock.h index c13681ac1ede..4e7712ee9394 100644 --- a/arch/arm/include/asm/spinlock.h +++ b/arch/arm/include/asm/spinlock.h @@ -23,7 +23,7 @@ #define __raw_spin_lock_flags(lock, flags) __raw_spin_lock(lock) -static inline void __raw_spin_lock(raw_spinlock_t *lock) +static inline void __raw_spin_lock(arch_spinlock_t *lock) { unsigned long tmp; @@ -43,7 +43,7 @@ static inline void __raw_spin_lock(raw_spinlock_t *lock) smp_mb(); } -static inline int __raw_spin_trylock(raw_spinlock_t *lock) +static inline int __raw_spin_trylock(arch_spinlock_t *lock) { unsigned long tmp; @@ -63,7 +63,7 @@ static inline int __raw_spin_trylock(raw_spinlock_t *lock) } } -static inline void __raw_spin_unlock(raw_spinlock_t *lock) +static inline void __raw_spin_unlock(arch_spinlock_t *lock) { smp_mb(); diff --git a/arch/arm/include/asm/spinlock_types.h b/arch/arm/include/asm/spinlock_types.h index 43e83f6d2ee5..5e9d3eadd167 100644 --- a/arch/arm/include/asm/spinlock_types.h +++ b/arch/arm/include/asm/spinlock_types.h @@ -7,7 +7,7 @@ typedef struct { volatile unsigned int lock; -} raw_spinlock_t; +} arch_spinlock_t; #define __RAW_SPIN_LOCK_UNLOCKED { 0 } diff --git a/arch/blackfin/include/asm/spinlock.h b/arch/blackfin/include/asm/spinlock.h index b0c7f0ee4b03..fc16b4c5309b 100644 --- a/arch/blackfin/include/asm/spinlock.h +++ b/arch/blackfin/include/asm/spinlock.h @@ -24,29 +24,29 @@ asmlinkage void __raw_write_lock_asm(volatile int *ptr); asmlinkage int __raw_write_trylock_asm(volatile int *ptr); asmlinkage void __raw_write_unlock_asm(volatile int *ptr); -static inline int __raw_spin_is_locked(raw_spinlock_t *lock) +static inline int __raw_spin_is_locked(arch_spinlock_t *lock) { return __raw_spin_is_locked_asm(&lock->lock); } -static inline void __raw_spin_lock(raw_spinlock_t *lock) +static inline void __raw_spin_lock(arch_spinlock_t *lock) { __raw_spin_lock_asm(&lock->lock); } #define __raw_spin_lock_flags(lock, flags) __raw_spin_lock(lock) -static inline int __raw_spin_trylock(raw_spinlock_t *lock) +static inline int __raw_spin_trylock(arch_spinlock_t *lock) { return __raw_spin_trylock_asm(&lock->lock); } -static inline void __raw_spin_unlock(raw_spinlock_t *lock) +static inline void __raw_spin_unlock(arch_spinlock_t *lock) { __raw_spin_unlock_asm(&lock->lock); } -static inline void __raw_spin_unlock_wait(raw_spinlock_t *lock) +static inline void __raw_spin_unlock_wait(arch_spinlock_t *lock) { while (__raw_spin_is_locked(lock)) cpu_relax(); diff --git a/arch/blackfin/include/asm/spinlock_types.h b/arch/blackfin/include/asm/spinlock_types.h index be75762c0610..03b377abf5c0 100644 --- a/arch/blackfin/include/asm/spinlock_types.h +++ b/arch/blackfin/include/asm/spinlock_types.h @@ -15,7 +15,7 @@ typedef struct { volatile unsigned int lock; -} raw_spinlock_t; +} arch_spinlock_t; #define __RAW_SPIN_LOCK_UNLOCKED { 0 } diff --git a/arch/cris/include/arch-v32/arch/spinlock.h b/arch/cris/include/arch-v32/arch/spinlock.h index 367a53ea10c5..e253457765f2 100644 --- a/arch/cris/include/arch-v32/arch/spinlock.h +++ b/arch/cris/include/arch-v32/arch/spinlock.h @@ -9,12 +9,12 @@ extern void cris_spin_unlock(void *l, int val); extern void cris_spin_lock(void *l); extern int cris_spin_trylock(void *l); -static inline int __raw_spin_is_locked(raw_spinlock_t *x) +static inline int __raw_spin_is_locked(arch_spinlock_t *x) { return *(volatile signed char *)(&(x)->slock) <= 0; } -static inline void __raw_spin_unlock(raw_spinlock_t *lock) +static inline void __raw_spin_unlock(arch_spinlock_t *lock) { __asm__ volatile ("move.d %1,%0" \ : "=m" (lock->slock) \ @@ -22,24 +22,24 @@ static inline void __raw_spin_unlock(raw_spinlock_t *lock) : "memory"); } -static inline void __raw_spin_unlock_wait(raw_spinlock_t *lock) +static inline void __raw_spin_unlock_wait(arch_spinlock_t *lock) { while (__raw_spin_is_locked(lock)) cpu_relax(); } -static inline int __raw_spin_trylock(raw_spinlock_t *lock) +static inline int __raw_spin_trylock(arch_spinlock_t *lock) { return cris_spin_trylock((void *)&lock->slock); } -static inline void __raw_spin_lock(raw_spinlock_t *lock) +static inline void __raw_spin_lock(arch_spinlock_t *lock) { cris_spin_lock((void *)&lock->slock); } static inline void -__raw_spin_lock_flags(raw_spinlock_t *lock, unsigned long flags) +__raw_spin_lock_flags(arch_spinlock_t *lock, unsigned long flags) { __raw_spin_lock(lock); } diff --git a/arch/ia64/include/asm/spinlock.h b/arch/ia64/include/asm/spinlock.h index 239ecdc9516d..9fbdf7e61087 100644 --- a/arch/ia64/include/asm/spinlock.h +++ b/arch/ia64/include/asm/spinlock.h @@ -38,7 +38,7 @@ #define TICKET_BITS 15 #define TICKET_MASK ((1 << TICKET_BITS) - 1) -static __always_inline void __ticket_spin_lock(raw_spinlock_t *lock) +static __always_inline void __ticket_spin_lock(arch_spinlock_t *lock) { int *p = (int *)&lock->lock, ticket, serve; @@ -58,7 +58,7 @@ static __always_inline void __ticket_spin_lock(raw_spinlock_t *lock) } } -static __always_inline int __ticket_spin_trylock(raw_spinlock_t *lock) +static __always_inline int __ticket_spin_trylock(arch_spinlock_t *lock) { int tmp = ACCESS_ONCE(lock->lock); @@ -67,7 +67,7 @@ static __always_inline int __ticket_spin_trylock(raw_spinlock_t *lock) return 0; } -static __always_inline void __ticket_spin_unlock(raw_spinlock_t *lock) +static __always_inline void __ticket_spin_unlock(arch_spinlock_t *lock) { unsigned short *p = (unsigned short *)&lock->lock + 1, tmp; @@ -75,7 +75,7 @@ static __always_inline void __ticket_spin_unlock(raw_spinlock_t *lock) ACCESS_ONCE(*p) = (tmp + 2) & ~1; } -static __always_inline void __ticket_spin_unlock_wait(raw_spinlock_t *lock) +static __always_inline void __ticket_spin_unlock_wait(arch_spinlock_t *lock) { int *p = (int *)&lock->lock, ticket; @@ -89,53 +89,53 @@ static __always_inline void __ticket_spin_unlock_wait(raw_spinlock_t *lock) } } -static inline int __ticket_spin_is_locked(raw_spinlock_t *lock) +static inline int __ticket_spin_is_locked(arch_spinlock_t *lock) { long tmp = ACCESS_ONCE(lock->lock); return !!(((tmp >> TICKET_SHIFT) ^ tmp) & TICKET_MASK); } -static inline int __ticket_spin_is_contended(raw_spinlock_t *lock) +static inline int __ticket_spin_is_contended(arch_spinlock_t *lock) { long tmp = ACCESS_ONCE(lock->lock); return ((tmp - (tmp >> TICKET_SHIFT)) & TICKET_MASK) > 1; } -static inline int __raw_spin_is_locked(raw_spinlock_t *lock) +static inline int __raw_spin_is_locked(arch_spinlock_t *lock) { return __ticket_spin_is_locked(lock); } -static inline int __raw_spin_is_contended(raw_spinlock_t *lock) +static inline int __raw_spin_is_contended(arch_spinlock_t *lock) { return __ticket_spin_is_contended(lock); } #define __raw_spin_is_contended __raw_spin_is_contended -static __always_inline void __raw_spin_lock(raw_spinlock_t *lock) +static __always_inline void __raw_spin_lock(arch_spinlock_t *lock) { __ticket_spin_lock(lock); } -static __always_inline int __raw_spin_trylock(raw_spinlock_t *lock) +static __always_inline int __raw_spin_trylock(arch_spinlock_t *lock) { return __ticket_spin_trylock(lock); } -static __always_inline void __raw_spin_unlock(raw_spinlock_t *lock) +static __always_inline void __raw_spin_unlock(arch_spinlock_t *lock) { __ticket_spin_unlock(lock); } -static __always_inline void __raw_spin_lock_flags(raw_spinlock_t *lock, +static __always_inline void __raw_spin_lock_flags(arch_spinlock_t *lock, unsigned long flags) { __raw_spin_lock(lock); } -static inline void __raw_spin_unlock_wait(raw_spinlock_t *lock) +static inline void __raw_spin_unlock_wait(arch_spinlock_t *lock) { __ticket_spin_unlock_wait(lock); } diff --git a/arch/ia64/include/asm/spinlock_types.h b/arch/ia64/include/asm/spinlock_types.h index 474e46f1ab4a..447ccc6ca7a8 100644 --- a/arch/ia64/include/asm/spinlock_types.h +++ b/arch/ia64/include/asm/spinlock_types.h @@ -7,7 +7,7 @@ typedef struct { volatile unsigned int lock; -} raw_spinlock_t; +} arch_spinlock_t; #define __RAW_SPIN_LOCK_UNLOCKED { 0 } diff --git a/arch/m32r/include/asm/spinlock.h b/arch/m32r/include/asm/spinlock.h index dded923883b2..0c0164225bc0 100644 --- a/arch/m32r/include/asm/spinlock.h +++ b/arch/m32r/include/asm/spinlock.h @@ -36,7 +36,7 @@ * __raw_spin_trylock() tries to get the lock and returns a result. * On the m32r, the result value is 1 (= Success) or 0 (= Failure). */ -static inline int __raw_spin_trylock(raw_spinlock_t *lock) +static inline int __raw_spin_trylock(arch_spinlock_t *lock) { int oldval; unsigned long tmp1, tmp2; @@ -69,7 +69,7 @@ static inline int __raw_spin_trylock(raw_spinlock_t *lock) return (oldval > 0); } -static inline void __raw_spin_lock(raw_spinlock_t *lock) +static inline void __raw_spin_lock(arch_spinlock_t *lock) { unsigned long tmp0, tmp1; @@ -111,7 +111,7 @@ static inline void __raw_spin_lock(raw_spinlock_t *lock) ); } -static inline void __raw_spin_unlock(raw_spinlock_t *lock) +static inline void __raw_spin_unlock(arch_spinlock_t *lock) { mb(); lock->slock = 1; diff --git a/arch/m32r/include/asm/spinlock_types.h b/arch/m32r/include/asm/spinlock_types.h index 83f52105c0e4..17d15bd6322d 100644 --- a/arch/m32r/include/asm/spinlock_types.h +++ b/arch/m32r/include/asm/spinlock_types.h @@ -7,7 +7,7 @@ typedef struct { volatile int slock; -} raw_spinlock_t; +} arch_spinlock_t; #define __RAW_SPIN_LOCK_UNLOCKED { 1 } diff --git a/arch/mips/include/asm/spinlock.h b/arch/mips/include/asm/spinlock.h index 5b60a09a0f08..0f16d0673b4a 100644 --- a/arch/mips/include/asm/spinlock.h +++ b/arch/mips/include/asm/spinlock.h @@ -34,7 +34,7 @@ * becomes equal to the the initial value of the tail. */ -static inline int __raw_spin_is_locked(raw_spinlock_t *lock) +static inline int __raw_spin_is_locked(arch_spinlock_t *lock) { unsigned int counters = ACCESS_ONCE(lock->lock); @@ -45,7 +45,7 @@ static inline int __raw_spin_is_locked(raw_spinlock_t *lock) #define __raw_spin_unlock_wait(x) \ while (__raw_spin_is_locked(x)) { cpu_relax(); } -static inline int __raw_spin_is_contended(raw_spinlock_t *lock) +static inline int __raw_spin_is_contended(arch_spinlock_t *lock) { unsigned int counters = ACCESS_ONCE(lock->lock); @@ -53,7 +53,7 @@ static inline int __raw_spin_is_contended(raw_spinlock_t *lock) } #define __raw_spin_is_contended __raw_spin_is_contended -static inline void __raw_spin_lock(raw_spinlock_t *lock) +static inline void __raw_spin_lock(arch_spinlock_t *lock) { int my_ticket; int tmp; @@ -134,7 +134,7 @@ static inline void __raw_spin_lock(raw_spinlock_t *lock) smp_llsc_mb(); } -static inline void __raw_spin_unlock(raw_spinlock_t *lock) +static inline void __raw_spin_unlock(arch_spinlock_t *lock) { int tmp; @@ -174,7 +174,7 @@ static inline void __raw_spin_unlock(raw_spinlock_t *lock) } } -static inline unsigned int __raw_spin_trylock(raw_spinlock_t *lock) +static inline unsigned int __raw_spin_trylock(arch_spinlock_t *lock) { int tmp, tmp2, tmp3; diff --git a/arch/mips/include/asm/spinlock_types.h b/arch/mips/include/asm/spinlock_types.h index adeedaa116c1..2e1060892d3b 100644 --- a/arch/mips/include/asm/spinlock_types.h +++ b/arch/mips/include/asm/spinlock_types.h @@ -12,7 +12,7 @@ typedef struct { * bits 15..28: ticket */ unsigned int lock; -} raw_spinlock_t; +} arch_spinlock_t; #define __RAW_SPIN_LOCK_UNLOCKED { 0 } diff --git a/arch/parisc/include/asm/atomic.h b/arch/parisc/include/asm/atomic.h index 8bc9e96699b2..3a4ea778d4b6 100644 --- a/arch/parisc/include/asm/atomic.h +++ b/arch/parisc/include/asm/atomic.h @@ -27,18 +27,18 @@ # define ATOMIC_HASH_SIZE 4 # define ATOMIC_HASH(a) (&(__atomic_hash[ (((unsigned long) (a))/L1_CACHE_BYTES) & (ATOMIC_HASH_SIZE-1) ])) -extern raw_spinlock_t __atomic_hash[ATOMIC_HASH_SIZE] __lock_aligned; +extern arch_spinlock_t __atomic_hash[ATOMIC_HASH_SIZE] __lock_aligned; /* Can't use raw_spin_lock_irq because of #include problems, so * this is the substitute */ #define _atomic_spin_lock_irqsave(l,f) do { \ - raw_spinlock_t *s = ATOMIC_HASH(l); \ + arch_spinlock_t *s = ATOMIC_HASH(l); \ local_irq_save(f); \ __raw_spin_lock(s); \ } while(0) #define _atomic_spin_unlock_irqrestore(l,f) do { \ - raw_spinlock_t *s = ATOMIC_HASH(l); \ + arch_spinlock_t *s = ATOMIC_HASH(l); \ __raw_spin_unlock(s); \ local_irq_restore(f); \ } while(0) diff --git a/arch/parisc/include/asm/spinlock.h b/arch/parisc/include/asm/spinlock.h index fae03e136fa8..69e8dca26744 100644 --- a/arch/parisc/include/asm/spinlock.h +++ b/arch/parisc/include/asm/spinlock.h @@ -5,7 +5,7 @@ #include #include -static inline int __raw_spin_is_locked(raw_spinlock_t *x) +static inline int __raw_spin_is_locked(arch_spinlock_t *x) { volatile unsigned int *a = __ldcw_align(x); return *a == 0; @@ -15,7 +15,7 @@ static inline int __raw_spin_is_locked(raw_spinlock_t *x) #define __raw_spin_unlock_wait(x) \ do { cpu_relax(); } while (__raw_spin_is_locked(x)) -static inline void __raw_spin_lock_flags(raw_spinlock_t *x, +static inline void __raw_spin_lock_flags(arch_spinlock_t *x, unsigned long flags) { volatile unsigned int *a; @@ -33,7 +33,7 @@ static inline void __raw_spin_lock_flags(raw_spinlock_t *x, mb(); } -static inline void __raw_spin_unlock(raw_spinlock_t *x) +static inline void __raw_spin_unlock(arch_spinlock_t *x) { volatile unsigned int *a; mb(); @@ -42,7 +42,7 @@ static inline void __raw_spin_unlock(raw_spinlock_t *x) mb(); } -static inline int __raw_spin_trylock(raw_spinlock_t *x) +static inline int __raw_spin_trylock(arch_spinlock_t *x) { volatile unsigned int *a; int ret; diff --git a/arch/parisc/include/asm/spinlock_types.h b/arch/parisc/include/asm/spinlock_types.h index 3f72f47cf4b2..735caafb81f5 100644 --- a/arch/parisc/include/asm/spinlock_types.h +++ b/arch/parisc/include/asm/spinlock_types.h @@ -9,10 +9,10 @@ typedef struct { volatile unsigned int lock[4]; # define __RAW_SPIN_LOCK_UNLOCKED { { 1, 1, 1, 1 } } #endif -} raw_spinlock_t; +} arch_spinlock_t; typedef struct { - raw_spinlock_t lock; + arch_spinlock_t lock; volatile int counter; } raw_rwlock_t; diff --git a/arch/parisc/lib/bitops.c b/arch/parisc/lib/bitops.c index e3eb739fab19..fdd7f583de54 100644 --- a/arch/parisc/lib/bitops.c +++ b/arch/parisc/lib/bitops.c @@ -12,7 +12,7 @@ #include #ifdef CONFIG_SMP -raw_spinlock_t __atomic_hash[ATOMIC_HASH_SIZE] __lock_aligned = { +arch_spinlock_t __atomic_hash[ATOMIC_HASH_SIZE] __lock_aligned = { [0 ... (ATOMIC_HASH_SIZE-1)] = __RAW_SPIN_LOCK_UNLOCKED }; #endif diff --git a/arch/powerpc/include/asm/rtas.h b/arch/powerpc/include/asm/rtas.h index 168fce726201..20de73c36682 100644 --- a/arch/powerpc/include/asm/rtas.h +++ b/arch/powerpc/include/asm/rtas.h @@ -58,7 +58,7 @@ struct rtas_t { unsigned long entry; /* physical address pointer */ unsigned long base; /* physical address pointer */ unsigned long size; - raw_spinlock_t lock; + arch_spinlock_t lock; struct rtas_args args; struct device_node *dev; /* virtual address pointer */ }; diff --git a/arch/powerpc/include/asm/spinlock.h b/arch/powerpc/include/asm/spinlock.h index 198266cf9e2d..c0d44c92ff0e 100644 --- a/arch/powerpc/include/asm/spinlock.h +++ b/arch/powerpc/include/asm/spinlock.h @@ -54,7 +54,7 @@ * This returns the old value in the lock, so we succeeded * in getting the lock if the return value is 0. */ -static inline unsigned long arch_spin_trylock(raw_spinlock_t *lock) +static inline unsigned long arch_spin_trylock(arch_spinlock_t *lock) { unsigned long tmp, token; @@ -73,7 +73,7 @@ static inline unsigned long arch_spin_trylock(raw_spinlock_t *lock) return tmp; } -static inline int __raw_spin_trylock(raw_spinlock_t *lock) +static inline int __raw_spin_trylock(arch_spinlock_t *lock) { CLEAR_IO_SYNC; return arch_spin_trylock(lock) == 0; @@ -96,7 +96,7 @@ static inline int __raw_spin_trylock(raw_spinlock_t *lock) #if defined(CONFIG_PPC_SPLPAR) || defined(CONFIG_PPC_ISERIES) /* We only yield to the hypervisor if we are in shared processor mode */ #define SHARED_PROCESSOR (get_lppaca()->shared_proc) -extern void __spin_yield(raw_spinlock_t *lock); +extern void __spin_yield(arch_spinlock_t *lock); extern void __rw_yield(raw_rwlock_t *lock); #else /* SPLPAR || ISERIES */ #define __spin_yield(x) barrier() @@ -104,7 +104,7 @@ extern void __rw_yield(raw_rwlock_t *lock); #define SHARED_PROCESSOR 0 #endif -static inline void __raw_spin_lock(raw_spinlock_t *lock) +static inline void __raw_spin_lock(arch_spinlock_t *lock) { CLEAR_IO_SYNC; while (1) { @@ -120,7 +120,7 @@ static inline void __raw_spin_lock(raw_spinlock_t *lock) } static inline -void __raw_spin_lock_flags(raw_spinlock_t *lock, unsigned long flags) +void __raw_spin_lock_flags(arch_spinlock_t *lock, unsigned long flags) { unsigned long flags_dis; @@ -140,7 +140,7 @@ void __raw_spin_lock_flags(raw_spinlock_t *lock, unsigned long flags) } } -static inline void __raw_spin_unlock(raw_spinlock_t *lock) +static inline void __raw_spin_unlock(arch_spinlock_t *lock) { SYNC_IO; __asm__ __volatile__("# __raw_spin_unlock\n\t" @@ -149,7 +149,7 @@ static inline void __raw_spin_unlock(raw_spinlock_t *lock) } #ifdef CONFIG_PPC64 -extern void __raw_spin_unlock_wait(raw_spinlock_t *lock); +extern void __raw_spin_unlock_wait(arch_spinlock_t *lock); #else #define __raw_spin_unlock_wait(lock) \ do { while (__raw_spin_is_locked(lock)) cpu_relax(); } while (0) diff --git a/arch/powerpc/include/asm/spinlock_types.h b/arch/powerpc/include/asm/spinlock_types.h index 74236c9f05b1..4312e5baaf88 100644 --- a/arch/powerpc/include/asm/spinlock_types.h +++ b/arch/powerpc/include/asm/spinlock_types.h @@ -7,7 +7,7 @@ typedef struct { volatile unsigned int slock; -} raw_spinlock_t; +} arch_spinlock_t; #define __RAW_SPIN_LOCK_UNLOCKED { 0 } diff --git a/arch/powerpc/kernel/rtas.c b/arch/powerpc/kernel/rtas.c index bf90361bb70f..579069c12152 100644 --- a/arch/powerpc/kernel/rtas.c +++ b/arch/powerpc/kernel/rtas.c @@ -978,7 +978,7 @@ int __init early_init_dt_scan_rtas(unsigned long node, return 1; } -static raw_spinlock_t timebase_lock; +static arch_spinlock_t timebase_lock; static u64 timebase = 0; void __cpuinit rtas_give_timebase(void) diff --git a/arch/powerpc/lib/locks.c b/arch/powerpc/lib/locks.c index 79d0fa3a470d..b06294cde499 100644 --- a/arch/powerpc/lib/locks.c +++ b/arch/powerpc/lib/locks.c @@ -25,7 +25,7 @@ #include #include -void __spin_yield(raw_spinlock_t *lock) +void __spin_yield(arch_spinlock_t *lock) { unsigned int lock_value, holder_cpu, yield_count; @@ -82,7 +82,7 @@ void __rw_yield(raw_rwlock_t *rw) } #endif -void __raw_spin_unlock_wait(raw_spinlock_t *lock) +void __raw_spin_unlock_wait(arch_spinlock_t *lock) { while (lock->slock) { HMT_low(); diff --git a/arch/powerpc/platforms/pasemi/setup.c b/arch/powerpc/platforms/pasemi/setup.c index a4619347aa7e..be36fece41d7 100644 --- a/arch/powerpc/platforms/pasemi/setup.c +++ b/arch/powerpc/platforms/pasemi/setup.c @@ -71,7 +71,7 @@ static void pas_restart(char *cmd) } #ifdef CONFIG_SMP -static raw_spinlock_t timebase_lock; +static arch_spinlock_t timebase_lock; static unsigned long timebase; static void __devinit pas_give_timebase(void) diff --git a/arch/s390/include/asm/spinlock.h b/arch/s390/include/asm/spinlock.h index c9af0d19c7ab..6121fa4b83d9 100644 --- a/arch/s390/include/asm/spinlock.h +++ b/arch/s390/include/asm/spinlock.h @@ -57,12 +57,12 @@ _raw_compare_and_swap(volatile unsigned int *lock, do { while (__raw_spin_is_locked(lock)) \ _raw_spin_relax(lock); } while (0) -extern void _raw_spin_lock_wait(raw_spinlock_t *); -extern void _raw_spin_lock_wait_flags(raw_spinlock_t *, unsigned long flags); -extern int _raw_spin_trylock_retry(raw_spinlock_t *); -extern void _raw_spin_relax(raw_spinlock_t *lock); +extern void _raw_spin_lock_wait(arch_spinlock_t *); +extern void _raw_spin_lock_wait_flags(arch_spinlock_t *, unsigned long flags); +extern int _raw_spin_trylock_retry(arch_spinlock_t *); +extern void _raw_spin_relax(arch_spinlock_t *lock); -static inline void __raw_spin_lock(raw_spinlock_t *lp) +static inline void __raw_spin_lock(arch_spinlock_t *lp) { int old; @@ -72,7 +72,7 @@ static inline void __raw_spin_lock(raw_spinlock_t *lp) _raw_spin_lock_wait(lp); } -static inline void __raw_spin_lock_flags(raw_spinlock_t *lp, +static inline void __raw_spin_lock_flags(arch_spinlock_t *lp, unsigned long flags) { int old; @@ -83,7 +83,7 @@ static inline void __raw_spin_lock_flags(raw_spinlock_t *lp, _raw_spin_lock_wait_flags(lp, flags); } -static inline int __raw_spin_trylock(raw_spinlock_t *lp) +static inline int __raw_spin_trylock(arch_spinlock_t *lp) { int old; @@ -93,7 +93,7 @@ static inline int __raw_spin_trylock(raw_spinlock_t *lp) return _raw_spin_trylock_retry(lp); } -static inline void __raw_spin_unlock(raw_spinlock_t *lp) +static inline void __raw_spin_unlock(arch_spinlock_t *lp) { _raw_compare_and_swap(&lp->owner_cpu, lp->owner_cpu, 0); } diff --git a/arch/s390/include/asm/spinlock_types.h b/arch/s390/include/asm/spinlock_types.h index 654abc40de04..a93638eee3f7 100644 --- a/arch/s390/include/asm/spinlock_types.h +++ b/arch/s390/include/asm/spinlock_types.h @@ -7,7 +7,7 @@ typedef struct { volatile unsigned int owner_cpu; -} __attribute__ ((aligned (4))) raw_spinlock_t; +} __attribute__ ((aligned (4))) arch_spinlock_t; #define __RAW_SPIN_LOCK_UNLOCKED { 0 } diff --git a/arch/s390/lib/spinlock.c b/arch/s390/lib/spinlock.c index f7e0d30250b7..d4cbf71a6077 100644 --- a/arch/s390/lib/spinlock.c +++ b/arch/s390/lib/spinlock.c @@ -39,7 +39,7 @@ static inline void _raw_yield_cpu(int cpu) _raw_yield(); } -void _raw_spin_lock_wait(raw_spinlock_t *lp) +void _raw_spin_lock_wait(arch_spinlock_t *lp) { int count = spin_retry; unsigned int cpu = ~smp_processor_id(); @@ -59,7 +59,7 @@ void _raw_spin_lock_wait(raw_spinlock_t *lp) } EXPORT_SYMBOL(_raw_spin_lock_wait); -void _raw_spin_lock_wait_flags(raw_spinlock_t *lp, unsigned long flags) +void _raw_spin_lock_wait_flags(arch_spinlock_t *lp, unsigned long flags) { int count = spin_retry; unsigned int cpu = ~smp_processor_id(); @@ -82,7 +82,7 @@ void _raw_spin_lock_wait_flags(raw_spinlock_t *lp, unsigned long flags) } EXPORT_SYMBOL(_raw_spin_lock_wait_flags); -int _raw_spin_trylock_retry(raw_spinlock_t *lp) +int _raw_spin_trylock_retry(arch_spinlock_t *lp) { unsigned int cpu = ~smp_processor_id(); int count; @@ -97,7 +97,7 @@ int _raw_spin_trylock_retry(raw_spinlock_t *lp) } EXPORT_SYMBOL(_raw_spin_trylock_retry); -void _raw_spin_relax(raw_spinlock_t *lock) +void _raw_spin_relax(arch_spinlock_t *lock) { unsigned int cpu = lock->owner_cpu; if (cpu != 0) diff --git a/arch/sh/include/asm/spinlock.h b/arch/sh/include/asm/spinlock.h index a28c9f0053fd..5a05b3fcefbe 100644 --- a/arch/sh/include/asm/spinlock.h +++ b/arch/sh/include/asm/spinlock.h @@ -34,7 +34,7 @@ * * We make no fairness assumptions. They have a cost. */ -static inline void __raw_spin_lock(raw_spinlock_t *lock) +static inline void __raw_spin_lock(arch_spinlock_t *lock) { unsigned long tmp; unsigned long oldval; @@ -54,7 +54,7 @@ static inline void __raw_spin_lock(raw_spinlock_t *lock) ); } -static inline void __raw_spin_unlock(raw_spinlock_t *lock) +static inline void __raw_spin_unlock(arch_spinlock_t *lock) { unsigned long tmp; @@ -67,7 +67,7 @@ static inline void __raw_spin_unlock(raw_spinlock_t *lock) ); } -static inline int __raw_spin_trylock(raw_spinlock_t *lock) +static inline int __raw_spin_trylock(arch_spinlock_t *lock) { unsigned long tmp, oldval; diff --git a/arch/sh/include/asm/spinlock_types.h b/arch/sh/include/asm/spinlock_types.h index b4d244e7b60c..37712c32ba99 100644 --- a/arch/sh/include/asm/spinlock_types.h +++ b/arch/sh/include/asm/spinlock_types.h @@ -7,7 +7,7 @@ typedef struct { volatile unsigned int lock; -} raw_spinlock_t; +} arch_spinlock_t; #define __RAW_SPIN_LOCK_UNLOCKED { 1 } diff --git a/arch/sparc/include/asm/spinlock_32.h b/arch/sparc/include/asm/spinlock_32.h index 857630cff636..b2d8a67f727e 100644 --- a/arch/sparc/include/asm/spinlock_32.h +++ b/arch/sparc/include/asm/spinlock_32.h @@ -15,7 +15,7 @@ #define __raw_spin_unlock_wait(lock) \ do { while (__raw_spin_is_locked(lock)) cpu_relax(); } while (0) -static inline void __raw_spin_lock(raw_spinlock_t *lock) +static inline void __raw_spin_lock(arch_spinlock_t *lock) { __asm__ __volatile__( "\n1:\n\t" @@ -35,7 +35,7 @@ static inline void __raw_spin_lock(raw_spinlock_t *lock) : "g2", "memory", "cc"); } -static inline int __raw_spin_trylock(raw_spinlock_t *lock) +static inline int __raw_spin_trylock(arch_spinlock_t *lock) { unsigned int result; __asm__ __volatile__("ldstub [%1], %0" @@ -45,7 +45,7 @@ static inline int __raw_spin_trylock(raw_spinlock_t *lock) return (result == 0); } -static inline void __raw_spin_unlock(raw_spinlock_t *lock) +static inline void __raw_spin_unlock(arch_spinlock_t *lock) { __asm__ __volatile__("stb %%g0, [%0]" : : "r" (lock) : "memory"); } diff --git a/arch/sparc/include/asm/spinlock_64.h b/arch/sparc/include/asm/spinlock_64.h index 43e514783582..38e16c40efc4 100644 --- a/arch/sparc/include/asm/spinlock_64.h +++ b/arch/sparc/include/asm/spinlock_64.h @@ -27,7 +27,7 @@ do { rmb(); \ } while((lp)->lock) -static inline void __raw_spin_lock(raw_spinlock_t *lock) +static inline void __raw_spin_lock(arch_spinlock_t *lock) { unsigned long tmp; @@ -46,7 +46,7 @@ static inline void __raw_spin_lock(raw_spinlock_t *lock) : "memory"); } -static inline int __raw_spin_trylock(raw_spinlock_t *lock) +static inline int __raw_spin_trylock(arch_spinlock_t *lock) { unsigned long result; @@ -59,7 +59,7 @@ static inline int __raw_spin_trylock(raw_spinlock_t *lock) return (result == 0UL); } -static inline void __raw_spin_unlock(raw_spinlock_t *lock) +static inline void __raw_spin_unlock(arch_spinlock_t *lock) { __asm__ __volatile__( " stb %%g0, [%0]" @@ -68,7 +68,7 @@ static inline void __raw_spin_unlock(raw_spinlock_t *lock) : "memory"); } -static inline void __raw_spin_lock_flags(raw_spinlock_t *lock, unsigned long flags) +static inline void __raw_spin_lock_flags(arch_spinlock_t *lock, unsigned long flags) { unsigned long tmp1, tmp2; diff --git a/arch/sparc/include/asm/spinlock_types.h b/arch/sparc/include/asm/spinlock_types.h index 37cbe01c585b..41d9a8fec13d 100644 --- a/arch/sparc/include/asm/spinlock_types.h +++ b/arch/sparc/include/asm/spinlock_types.h @@ -7,7 +7,7 @@ typedef struct { volatile unsigned char lock; -} raw_spinlock_t; +} arch_spinlock_t; #define __RAW_SPIN_LOCK_UNLOCKED { 0 } diff --git a/arch/x86/include/asm/paravirt.h b/arch/x86/include/asm/paravirt.h index efb38994859c..5655f75f10b7 100644 --- a/arch/x86/include/asm/paravirt.h +++ b/arch/x86/include/asm/paravirt.h @@ -731,34 +731,34 @@ static inline void __set_fixmap(unsigned /* enum fixed_addresses */ idx, #if defined(CONFIG_SMP) && defined(CONFIG_PARAVIRT_SPINLOCKS) -static inline int __raw_spin_is_locked(struct raw_spinlock *lock) +static inline int __raw_spin_is_locked(struct arch_spinlock *lock) { return PVOP_CALL1(int, pv_lock_ops.spin_is_locked, lock); } -static inline int __raw_spin_is_contended(struct raw_spinlock *lock) +static inline int __raw_spin_is_contended(struct arch_spinlock *lock) { return PVOP_CALL1(int, pv_lock_ops.spin_is_contended, lock); } #define __raw_spin_is_contended __raw_spin_is_contended -static __always_inline void __raw_spin_lock(struct raw_spinlock *lock) +static __always_inline void __raw_spin_lock(struct arch_spinlock *lock) { PVOP_VCALL1(pv_lock_ops.spin_lock, lock); } -static __always_inline void __raw_spin_lock_flags(struct raw_spinlock *lock, +static __always_inline void __raw_spin_lock_flags(struct arch_spinlock *lock, unsigned long flags) { PVOP_VCALL2(pv_lock_ops.spin_lock_flags, lock, flags); } -static __always_inline int __raw_spin_trylock(struct raw_spinlock *lock) +static __always_inline int __raw_spin_trylock(struct arch_spinlock *lock) { return PVOP_CALL1(int, pv_lock_ops.spin_trylock, lock); } -static __always_inline void __raw_spin_unlock(struct raw_spinlock *lock) +static __always_inline void __raw_spin_unlock(struct arch_spinlock *lock) { PVOP_VCALL1(pv_lock_ops.spin_unlock, lock); } diff --git a/arch/x86/include/asm/paravirt_types.h b/arch/x86/include/asm/paravirt_types.h index 9357473c8da0..b1e70d51e40c 100644 --- a/arch/x86/include/asm/paravirt_types.h +++ b/arch/x86/include/asm/paravirt_types.h @@ -318,14 +318,14 @@ struct pv_mmu_ops { phys_addr_t phys, pgprot_t flags); }; -struct raw_spinlock; +struct arch_spinlock; struct pv_lock_ops { - int (*spin_is_locked)(struct raw_spinlock *lock); - int (*spin_is_contended)(struct raw_spinlock *lock); - void (*spin_lock)(struct raw_spinlock *lock); - void (*spin_lock_flags)(struct raw_spinlock *lock, unsigned long flags); - int (*spin_trylock)(struct raw_spinlock *lock); - void (*spin_unlock)(struct raw_spinlock *lock); + int (*spin_is_locked)(struct arch_spinlock *lock); + int (*spin_is_contended)(struct arch_spinlock *lock); + void (*spin_lock)(struct arch_spinlock *lock); + void (*spin_lock_flags)(struct arch_spinlock *lock, unsigned long flags); + int (*spin_trylock)(struct arch_spinlock *lock); + void (*spin_unlock)(struct arch_spinlock *lock); }; /* This contains all the paravirt structures: we get a convenient diff --git a/arch/x86/include/asm/spinlock.h b/arch/x86/include/asm/spinlock.h index 4e77853321db..204b524fcf57 100644 --- a/arch/x86/include/asm/spinlock.h +++ b/arch/x86/include/asm/spinlock.h @@ -58,7 +58,7 @@ #if (NR_CPUS < 256) #define TICKET_SHIFT 8 -static __always_inline void __ticket_spin_lock(raw_spinlock_t *lock) +static __always_inline void __ticket_spin_lock(arch_spinlock_t *lock) { short inc = 0x0100; @@ -77,7 +77,7 @@ static __always_inline void __ticket_spin_lock(raw_spinlock_t *lock) : "memory", "cc"); } -static __always_inline int __ticket_spin_trylock(raw_spinlock_t *lock) +static __always_inline int __ticket_spin_trylock(arch_spinlock_t *lock) { int tmp, new; @@ -96,7 +96,7 @@ static __always_inline int __ticket_spin_trylock(raw_spinlock_t *lock) return tmp; } -static __always_inline void __ticket_spin_unlock(raw_spinlock_t *lock) +static __always_inline void __ticket_spin_unlock(arch_spinlock_t *lock) { asm volatile(UNLOCK_LOCK_PREFIX "incb %0" : "+m" (lock->slock) @@ -106,7 +106,7 @@ static __always_inline void __ticket_spin_unlock(raw_spinlock_t *lock) #else #define TICKET_SHIFT 16 -static __always_inline void __ticket_spin_lock(raw_spinlock_t *lock) +static __always_inline void __ticket_spin_lock(arch_spinlock_t *lock) { int inc = 0x00010000; int tmp; @@ -127,7 +127,7 @@ static __always_inline void __ticket_spin_lock(raw_spinlock_t *lock) : "memory", "cc"); } -static __always_inline int __ticket_spin_trylock(raw_spinlock_t *lock) +static __always_inline int __ticket_spin_trylock(arch_spinlock_t *lock) { int tmp; int new; @@ -149,7 +149,7 @@ static __always_inline int __ticket_spin_trylock(raw_spinlock_t *lock) return tmp; } -static __always_inline void __ticket_spin_unlock(raw_spinlock_t *lock) +static __always_inline void __ticket_spin_unlock(arch_spinlock_t *lock) { asm volatile(UNLOCK_LOCK_PREFIX "incw %0" : "+m" (lock->slock) @@ -158,14 +158,14 @@ static __always_inline void __ticket_spin_unlock(raw_spinlock_t *lock) } #endif -static inline int __ticket_spin_is_locked(raw_spinlock_t *lock) +static inline int __ticket_spin_is_locked(arch_spinlock_t *lock) { int tmp = ACCESS_ONCE(lock->slock); return !!(((tmp >> TICKET_SHIFT) ^ tmp) & ((1 << TICKET_SHIFT) - 1)); } -static inline int __ticket_spin_is_contended(raw_spinlock_t *lock) +static inline int __ticket_spin_is_contended(arch_spinlock_t *lock) { int tmp = ACCESS_ONCE(lock->slock); @@ -174,33 +174,33 @@ static inline int __ticket_spin_is_contended(raw_spinlock_t *lock) #ifndef CONFIG_PARAVIRT_SPINLOCKS -static inline int __raw_spin_is_locked(raw_spinlock_t *lock) +static inline int __raw_spin_is_locked(arch_spinlock_t *lock) { return __ticket_spin_is_locked(lock); } -static inline int __raw_spin_is_contended(raw_spinlock_t *lock) +static inline int __raw_spin_is_contended(arch_spinlock_t *lock) { return __ticket_spin_is_contended(lock); } #define __raw_spin_is_contended __raw_spin_is_contended -static __always_inline void __raw_spin_lock(raw_spinlock_t *lock) +static __always_inline void __raw_spin_lock(arch_spinlock_t *lock) { __ticket_spin_lock(lock); } -static __always_inline int __raw_spin_trylock(raw_spinlock_t *lock) +static __always_inline int __raw_spin_trylock(arch_spinlock_t *lock) { return __ticket_spin_trylock(lock); } -static __always_inline void __raw_spin_unlock(raw_spinlock_t *lock) +static __always_inline void __raw_spin_unlock(arch_spinlock_t *lock) { __ticket_spin_unlock(lock); } -static __always_inline void __raw_spin_lock_flags(raw_spinlock_t *lock, +static __always_inline void __raw_spin_lock_flags(arch_spinlock_t *lock, unsigned long flags) { __raw_spin_lock(lock); @@ -208,7 +208,7 @@ static __always_inline void __raw_spin_lock_flags(raw_spinlock_t *lock, #endif /* CONFIG_PARAVIRT_SPINLOCKS */ -static inline void __raw_spin_unlock_wait(raw_spinlock_t *lock) +static inline void __raw_spin_unlock_wait(arch_spinlock_t *lock) { while (__raw_spin_is_locked(lock)) cpu_relax(); diff --git a/arch/x86/include/asm/spinlock_types.h b/arch/x86/include/asm/spinlock_types.h index 845f81c87091..2ae7637ed524 100644 --- a/arch/x86/include/asm/spinlock_types.h +++ b/arch/x86/include/asm/spinlock_types.h @@ -5,9 +5,9 @@ # error "please don't include this file directly" #endif -typedef struct raw_spinlock { +typedef struct arch_spinlock { unsigned int slock; -} raw_spinlock_t; +} arch_spinlock_t; #define __RAW_SPIN_LOCK_UNLOCKED { 0 } diff --git a/arch/x86/kernel/dumpstack.c b/arch/x86/kernel/dumpstack.c index b8ce165dde5d..0862d9d89c92 100644 --- a/arch/x86/kernel/dumpstack.c +++ b/arch/x86/kernel/dumpstack.c @@ -188,7 +188,7 @@ void dump_stack(void) } EXPORT_SYMBOL(dump_stack); -static raw_spinlock_t die_lock = __RAW_SPIN_LOCK_UNLOCKED; +static arch_spinlock_t die_lock = __RAW_SPIN_LOCK_UNLOCKED; static int die_owner = -1; static unsigned int die_nest_count; diff --git a/arch/x86/kernel/paravirt-spinlocks.c b/arch/x86/kernel/paravirt-spinlocks.c index 3a7c5a44082e..a0f39e090684 100644 --- a/arch/x86/kernel/paravirt-spinlocks.c +++ b/arch/x86/kernel/paravirt-spinlocks.c @@ -8,7 +8,7 @@ #include static inline void -default_spin_lock_flags(raw_spinlock_t *lock, unsigned long flags) +default_spin_lock_flags(arch_spinlock_t *lock, unsigned long flags) { __raw_spin_lock(lock); } diff --git a/arch/x86/kernel/tsc_sync.c b/arch/x86/kernel/tsc_sync.c index eed156851f5d..9f908b9d1abe 100644 --- a/arch/x86/kernel/tsc_sync.c +++ b/arch/x86/kernel/tsc_sync.c @@ -33,7 +33,7 @@ static __cpuinitdata atomic_t stop_count; * we want to have the fastest, inlined, non-debug version * of a critical section, to be able to prove TSC time-warps: */ -static __cpuinitdata raw_spinlock_t sync_lock = __RAW_SPIN_LOCK_UNLOCKED; +static __cpuinitdata arch_spinlock_t sync_lock = __RAW_SPIN_LOCK_UNLOCKED; static __cpuinitdata cycles_t last_tsc; static __cpuinitdata cycles_t max_warp; diff --git a/arch/x86/xen/spinlock.c b/arch/x86/xen/spinlock.c index 36a5141108df..24ded31b5aec 100644 --- a/arch/x86/xen/spinlock.c +++ b/arch/x86/xen/spinlock.c @@ -120,14 +120,14 @@ struct xen_spinlock { unsigned short spinners; /* count of waiting cpus */ }; -static int xen_spin_is_locked(struct raw_spinlock *lock) +static int xen_spin_is_locked(struct arch_spinlock *lock) { struct xen_spinlock *xl = (struct xen_spinlock *)lock; return xl->lock != 0; } -static int xen_spin_is_contended(struct raw_spinlock *lock) +static int xen_spin_is_contended(struct arch_spinlock *lock) { struct xen_spinlock *xl = (struct xen_spinlock *)lock; @@ -136,7 +136,7 @@ static int xen_spin_is_contended(struct raw_spinlock *lock) return xl->spinners != 0; } -static int xen_spin_trylock(struct raw_spinlock *lock) +static int xen_spin_trylock(struct arch_spinlock *lock) { struct xen_spinlock *xl = (struct xen_spinlock *)lock; u8 old = 1; @@ -181,7 +181,7 @@ static inline void unspinning_lock(struct xen_spinlock *xl, struct xen_spinlock __get_cpu_var(lock_spinners) = prev; } -static noinline int xen_spin_lock_slow(struct raw_spinlock *lock, bool irq_enable) +static noinline int xen_spin_lock_slow(struct arch_spinlock *lock, bool irq_enable) { struct xen_spinlock *xl = (struct xen_spinlock *)lock; struct xen_spinlock *prev; @@ -254,7 +254,7 @@ out: return ret; } -static inline void __xen_spin_lock(struct raw_spinlock *lock, bool irq_enable) +static inline void __xen_spin_lock(struct arch_spinlock *lock, bool irq_enable) { struct xen_spinlock *xl = (struct xen_spinlock *)lock; unsigned timeout; @@ -291,12 +291,12 @@ static inline void __xen_spin_lock(struct raw_spinlock *lock, bool irq_enable) spin_time_accum_total(start_spin); } -static void xen_spin_lock(struct raw_spinlock *lock) +static void xen_spin_lock(struct arch_spinlock *lock) { __xen_spin_lock(lock, false); } -static void xen_spin_lock_flags(struct raw_spinlock *lock, unsigned long flags) +static void xen_spin_lock_flags(struct arch_spinlock *lock, unsigned long flags) { __xen_spin_lock(lock, !raw_irqs_disabled_flags(flags)); } @@ -317,7 +317,7 @@ static noinline void xen_spin_unlock_slow(struct xen_spinlock *xl) } } -static void xen_spin_unlock(struct raw_spinlock *lock) +static void xen_spin_unlock(struct arch_spinlock *lock) { struct xen_spinlock *xl = (struct xen_spinlock *)lock; diff --git a/include/asm-generic/bitops/atomic.h b/include/asm-generic/bitops/atomic.h index c8946465e63a..dcf0afad4a7f 100644 --- a/include/asm-generic/bitops/atomic.h +++ b/include/asm-generic/bitops/atomic.h @@ -15,18 +15,18 @@ # define ATOMIC_HASH_SIZE 4 # define ATOMIC_HASH(a) (&(__atomic_hash[ (((unsigned long) a)/L1_CACHE_BYTES) & (ATOMIC_HASH_SIZE-1) ])) -extern raw_spinlock_t __atomic_hash[ATOMIC_HASH_SIZE] __lock_aligned; +extern arch_spinlock_t __atomic_hash[ATOMIC_HASH_SIZE] __lock_aligned; /* Can't use raw_spin_lock_irq because of #include problems, so * this is the substitute */ #define _atomic_spin_lock_irqsave(l,f) do { \ - raw_spinlock_t *s = ATOMIC_HASH(l); \ + arch_spinlock_t *s = ATOMIC_HASH(l); \ local_irq_save(f); \ __raw_spin_lock(s); \ } while(0) #define _atomic_spin_unlock_irqrestore(l,f) do { \ - raw_spinlock_t *s = ATOMIC_HASH(l); \ + arch_spinlock_t *s = ATOMIC_HASH(l); \ __raw_spin_unlock(s); \ local_irq_restore(f); \ } while(0) diff --git a/include/linux/spinlock.h b/include/linux/spinlock.h index a9aaa709fb93..5ef7a4c060b5 100644 --- a/include/linux/spinlock.h +++ b/include/linux/spinlock.h @@ -8,7 +8,7 @@ * * on SMP builds: * - * asm/spinlock_types.h: contains the raw_spinlock_t/raw_rwlock_t and the + * asm/spinlock_types.h: contains the arch_spinlock_t/raw_rwlock_t and the * initializers * * linux/spinlock_types.h: @@ -75,7 +75,7 @@ #define __lockfunc __attribute__((section(".spinlock.text"))) /* - * Pull the raw_spinlock_t and raw_rwlock_t definitions: + * Pull the arch_spinlock_t and raw_rwlock_t definitions: */ #include diff --git a/include/linux/spinlock_types.h b/include/linux/spinlock_types.h index f979d5d8a160..d4af2d7a86ea 100644 --- a/include/linux/spinlock_types.h +++ b/include/linux/spinlock_types.h @@ -18,7 +18,7 @@ #include typedef struct { - raw_spinlock_t raw_lock; + arch_spinlock_t raw_lock; #ifdef CONFIG_GENERIC_LOCKBREAK unsigned int break_lock; #endif diff --git a/include/linux/spinlock_types_up.h b/include/linux/spinlock_types_up.h index 04135b0e198e..34d36691c4ec 100644 --- a/include/linux/spinlock_types_up.h +++ b/include/linux/spinlock_types_up.h @@ -16,13 +16,13 @@ typedef struct { volatile unsigned int slock; -} raw_spinlock_t; +} arch_spinlock_t; #define __RAW_SPIN_LOCK_UNLOCKED { 1 } #else -typedef struct { } raw_spinlock_t; +typedef struct { } arch_spinlock_t; #define __RAW_SPIN_LOCK_UNLOCKED { } diff --git a/include/linux/spinlock_up.h b/include/linux/spinlock_up.h index d4841ed8215b..8ee2ac1bf636 100644 --- a/include/linux/spinlock_up.h +++ b/include/linux/spinlock_up.h @@ -20,19 +20,19 @@ #ifdef CONFIG_DEBUG_SPINLOCK #define __raw_spin_is_locked(x) ((x)->slock == 0) -static inline void __raw_spin_lock(raw_spinlock_t *lock) +static inline void __raw_spin_lock(arch_spinlock_t *lock) { lock->slock = 0; } static inline void -__raw_spin_lock_flags(raw_spinlock_t *lock, unsigned long flags) +__raw_spin_lock_flags(arch_spinlock_t *lock, unsigned long flags) { local_irq_save(flags); lock->slock = 0; } -static inline int __raw_spin_trylock(raw_spinlock_t *lock) +static inline int __raw_spin_trylock(arch_spinlock_t *lock) { char oldval = lock->slock; @@ -41,7 +41,7 @@ static inline int __raw_spin_trylock(raw_spinlock_t *lock) return oldval > 0; } -static inline void __raw_spin_unlock(raw_spinlock_t *lock) +static inline void __raw_spin_unlock(arch_spinlock_t *lock) { lock->slock = 1; } diff --git a/kernel/lockdep.c b/kernel/lockdep.c index 429540c70d3f..7cc50c62af59 100644 --- a/kernel/lockdep.c +++ b/kernel/lockdep.c @@ -73,7 +73,7 @@ module_param(lock_stat, int, 0644); * to use a raw spinlock - we really dont want the spinlock * code to recurse back into the lockdep code... */ -static raw_spinlock_t lockdep_lock = (raw_spinlock_t)__RAW_SPIN_LOCK_UNLOCKED; +static arch_spinlock_t lockdep_lock = (arch_spinlock_t)__RAW_SPIN_LOCK_UNLOCKED; static int graph_lock(void) { diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c index a1ca4956ab5e..5ac8ee0a9e35 100644 --- a/kernel/trace/ring_buffer.c +++ b/kernel/trace/ring_buffer.c @@ -423,7 +423,7 @@ struct ring_buffer_per_cpu { int cpu; struct ring_buffer *buffer; spinlock_t reader_lock; /* serialize readers */ - raw_spinlock_t lock; + arch_spinlock_t lock; struct lock_class_key lock_key; struct list_head *pages; struct buffer_page *head_page; /* read from head */ @@ -998,7 +998,7 @@ rb_allocate_cpu_buffer(struct ring_buffer *buffer, int cpu) cpu_buffer->buffer = buffer; spin_lock_init(&cpu_buffer->reader_lock); lockdep_set_class(&cpu_buffer->reader_lock, buffer->reader_lock_key); - cpu_buffer->lock = (raw_spinlock_t)__RAW_SPIN_LOCK_UNLOCKED; + cpu_buffer->lock = (arch_spinlock_t)__RAW_SPIN_LOCK_UNLOCKED; bpage = kzalloc_node(ALIGN(sizeof(*bpage), cache_line_size()), GFP_KERNEL, cpu_to_node(cpu)); diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index c82dfd92fdfd..7d56cecc2c6e 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -493,15 +493,15 @@ static ssize_t trace_seq_to_buffer(struct trace_seq *s, void *buf, size_t cnt) * protected by per_cpu spinlocks. But the action of the swap * needs its own lock. * - * This is defined as a raw_spinlock_t in order to help + * This is defined as a arch_spinlock_t in order to help * with performance when lockdep debugging is enabled. * * It is also used in other places outside the update_max_tr * so it needs to be defined outside of the * CONFIG_TRACER_MAX_TRACE. */ -static raw_spinlock_t ftrace_max_lock = - (raw_spinlock_t)__RAW_SPIN_LOCK_UNLOCKED; +static arch_spinlock_t ftrace_max_lock = + (arch_spinlock_t)__RAW_SPIN_LOCK_UNLOCKED; #ifdef CONFIG_TRACER_MAX_TRACE unsigned long __read_mostly tracing_max_latency; @@ -802,7 +802,7 @@ static unsigned map_pid_to_cmdline[PID_MAX_DEFAULT+1]; static unsigned map_cmdline_to_pid[SAVED_CMDLINES]; static char saved_cmdlines[SAVED_CMDLINES][TASK_COMM_LEN]; static int cmdline_idx; -static raw_spinlock_t trace_cmdline_lock = __RAW_SPIN_LOCK_UNLOCKED; +static arch_spinlock_t trace_cmdline_lock = __RAW_SPIN_LOCK_UNLOCKED; /* temporary disable recording */ static atomic_t trace_record_cmdline_disabled __read_mostly; @@ -1251,8 +1251,8 @@ ftrace_special(unsigned long arg1, unsigned long arg2, unsigned long arg3) */ int trace_vbprintk(unsigned long ip, const char *fmt, va_list args) { - static raw_spinlock_t trace_buf_lock = - (raw_spinlock_t)__RAW_SPIN_LOCK_UNLOCKED; + static arch_spinlock_t trace_buf_lock = + (arch_spinlock_t)__RAW_SPIN_LOCK_UNLOCKED; static u32 trace_buf[TRACE_BUF_SIZE]; struct ftrace_event_call *call = &event_bprint; @@ -1334,7 +1334,7 @@ int trace_array_printk(struct trace_array *tr, int trace_array_vprintk(struct trace_array *tr, unsigned long ip, const char *fmt, va_list args) { - static raw_spinlock_t trace_buf_lock = __RAW_SPIN_LOCK_UNLOCKED; + static arch_spinlock_t trace_buf_lock = __RAW_SPIN_LOCK_UNLOCKED; static char trace_buf[TRACE_BUF_SIZE]; struct ftrace_event_call *call = &event_print; @@ -4307,8 +4307,8 @@ trace_printk_seq(struct trace_seq *s) static void __ftrace_dump(bool disable_tracing) { - static raw_spinlock_t ftrace_dump_lock = - (raw_spinlock_t)__RAW_SPIN_LOCK_UNLOCKED; + static arch_spinlock_t ftrace_dump_lock = + (arch_spinlock_t)__RAW_SPIN_LOCK_UNLOCKED; /* use static because iter can be a bit big for the stack */ static struct trace_iterator iter; unsigned int old_userobj; diff --git a/kernel/trace/trace_clock.c b/kernel/trace/trace_clock.c index 878c03f386ba..206ec3d4b3c2 100644 --- a/kernel/trace/trace_clock.c +++ b/kernel/trace/trace_clock.c @@ -71,10 +71,10 @@ u64 notrace trace_clock(void) /* keep prev_time and lock in the same cacheline. */ static struct { u64 prev_time; - raw_spinlock_t lock; + arch_spinlock_t lock; } trace_clock_struct ____cacheline_aligned_in_smp = { - .lock = (raw_spinlock_t)__RAW_SPIN_LOCK_UNLOCKED, + .lock = (arch_spinlock_t)__RAW_SPIN_LOCK_UNLOCKED, }; u64 notrace trace_clock_global(void) diff --git a/kernel/trace/trace_sched_wakeup.c b/kernel/trace/trace_sched_wakeup.c index 26185d727676..4cf7e83ec235 100644 --- a/kernel/trace/trace_sched_wakeup.c +++ b/kernel/trace/trace_sched_wakeup.c @@ -28,8 +28,8 @@ static int wakeup_current_cpu; static unsigned wakeup_prio = -1; static int wakeup_rt; -static raw_spinlock_t wakeup_lock = - (raw_spinlock_t)__RAW_SPIN_LOCK_UNLOCKED; +static arch_spinlock_t wakeup_lock = + (arch_spinlock_t)__RAW_SPIN_LOCK_UNLOCKED; static void __wakeup_reset(struct trace_array *tr); diff --git a/kernel/trace/trace_stack.c b/kernel/trace/trace_stack.c index 8504ac71e4e8..9a82d568fdec 100644 --- a/kernel/trace/trace_stack.c +++ b/kernel/trace/trace_stack.c @@ -27,8 +27,8 @@ static struct stack_trace max_stack_trace = { }; static unsigned long max_stack_size; -static raw_spinlock_t max_stack_lock = - (raw_spinlock_t)__RAW_SPIN_LOCK_UNLOCKED; +static arch_spinlock_t max_stack_lock = + (arch_spinlock_t)__RAW_SPIN_LOCK_UNLOCKED; static int stack_trace_disabled __read_mostly; static DEFINE_PER_CPU(int, trace_active); diff --git a/lib/spinlock_debug.c b/lib/spinlock_debug.c index 9c4b0256490b..2acd501b3826 100644 --- a/lib/spinlock_debug.c +++ b/lib/spinlock_debug.c @@ -23,7 +23,7 @@ void __spin_lock_init(spinlock_t *lock, const char *name, debug_check_no_locks_freed((void *)lock, sizeof(*lock)); lockdep_init_map(&lock->dep_map, name, key, 0); #endif - lock->raw_lock = (raw_spinlock_t)__RAW_SPIN_LOCK_UNLOCKED; + lock->raw_lock = (arch_spinlock_t)__RAW_SPIN_LOCK_UNLOCKED; lock->magic = SPINLOCK_MAGIC; lock->owner = SPINLOCK_OWNER_INIT; lock->owner_cpu = -1; -- cgit v1.2.3-71-gd317 From edc35bd72e2079b25f99c5da7d7a65dbbffc4a26 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Thu, 3 Dec 2009 12:38:57 +0100 Subject: locking: Rename __RAW_SPIN_LOCK_UNLOCKED to __ARCH_SPIN_LOCK_UNLOCKED Further name space cleanup. No functional change Signed-off-by: Thomas Gleixner Acked-by: Peter Zijlstra Acked-by: David S. Miller Acked-by: Ingo Molnar Cc: linux-arch@vger.kernel.org --- arch/alpha/include/asm/spinlock_types.h | 2 +- arch/arm/include/asm/spinlock_types.h | 2 +- arch/blackfin/include/asm/spinlock_types.h | 2 +- arch/ia64/include/asm/spinlock_types.h | 2 +- arch/m32r/include/asm/spinlock_types.h | 2 +- arch/mips/include/asm/spinlock_types.h | 2 +- arch/parisc/include/asm/spinlock_types.h | 6 +++--- arch/parisc/lib/bitops.c | 2 +- arch/powerpc/include/asm/spinlock_types.h | 2 +- arch/powerpc/kernel/rtas.c | 2 +- arch/s390/include/asm/spinlock_types.h | 2 +- arch/sh/include/asm/spinlock_types.h | 2 +- arch/sparc/include/asm/spinlock_types.h | 2 +- arch/x86/include/asm/spinlock_types.h | 2 +- arch/x86/kernel/dumpstack.c | 2 +- arch/x86/kernel/tsc_sync.c | 2 +- include/linux/spinlock_types.h | 4 ++-- include/linux/spinlock_types_up.h | 4 ++-- kernel/lockdep.c | 2 +- kernel/trace/ring_buffer.c | 2 +- kernel/trace/trace.c | 10 +++++----- kernel/trace/trace_clock.c | 2 +- kernel/trace/trace_sched_wakeup.c | 2 +- kernel/trace/trace_stack.c | 2 +- lib/spinlock_debug.c | 2 +- 25 files changed, 33 insertions(+), 33 deletions(-) (limited to 'kernel') diff --git a/arch/alpha/include/asm/spinlock_types.h b/arch/alpha/include/asm/spinlock_types.h index bb94a51e53d2..08975ee0a100 100644 --- a/arch/alpha/include/asm/spinlock_types.h +++ b/arch/alpha/include/asm/spinlock_types.h @@ -9,7 +9,7 @@ typedef struct { volatile unsigned int lock; } arch_spinlock_t; -#define __RAW_SPIN_LOCK_UNLOCKED { 0 } +#define __ARCH_SPIN_LOCK_UNLOCKED { 0 } typedef struct { volatile unsigned int lock; diff --git a/arch/arm/include/asm/spinlock_types.h b/arch/arm/include/asm/spinlock_types.h index 5e9d3eadd167..9622e126a8de 100644 --- a/arch/arm/include/asm/spinlock_types.h +++ b/arch/arm/include/asm/spinlock_types.h @@ -9,7 +9,7 @@ typedef struct { volatile unsigned int lock; } arch_spinlock_t; -#define __RAW_SPIN_LOCK_UNLOCKED { 0 } +#define __ARCH_SPIN_LOCK_UNLOCKED { 0 } typedef struct { volatile unsigned int lock; diff --git a/arch/blackfin/include/asm/spinlock_types.h b/arch/blackfin/include/asm/spinlock_types.h index 03b377abf5c0..c8a3928a58c5 100644 --- a/arch/blackfin/include/asm/spinlock_types.h +++ b/arch/blackfin/include/asm/spinlock_types.h @@ -17,7 +17,7 @@ typedef struct { volatile unsigned int lock; } arch_spinlock_t; -#define __RAW_SPIN_LOCK_UNLOCKED { 0 } +#define __ARCH_SPIN_LOCK_UNLOCKED { 0 } typedef struct { volatile unsigned int lock; diff --git a/arch/ia64/include/asm/spinlock_types.h b/arch/ia64/include/asm/spinlock_types.h index 447ccc6ca7a8..6a11b65fa66d 100644 --- a/arch/ia64/include/asm/spinlock_types.h +++ b/arch/ia64/include/asm/spinlock_types.h @@ -9,7 +9,7 @@ typedef struct { volatile unsigned int lock; } arch_spinlock_t; -#define __RAW_SPIN_LOCK_UNLOCKED { 0 } +#define __ARCH_SPIN_LOCK_UNLOCKED { 0 } typedef struct { volatile unsigned int read_counter : 31; diff --git a/arch/m32r/include/asm/spinlock_types.h b/arch/m32r/include/asm/spinlock_types.h index 17d15bd6322d..5873a8701107 100644 --- a/arch/m32r/include/asm/spinlock_types.h +++ b/arch/m32r/include/asm/spinlock_types.h @@ -9,7 +9,7 @@ typedef struct { volatile int slock; } arch_spinlock_t; -#define __RAW_SPIN_LOCK_UNLOCKED { 1 } +#define __ARCH_SPIN_LOCK_UNLOCKED { 1 } typedef struct { volatile int lock; diff --git a/arch/mips/include/asm/spinlock_types.h b/arch/mips/include/asm/spinlock_types.h index 2e1060892d3b..b4c5efaadb9c 100644 --- a/arch/mips/include/asm/spinlock_types.h +++ b/arch/mips/include/asm/spinlock_types.h @@ -14,7 +14,7 @@ typedef struct { unsigned int lock; } arch_spinlock_t; -#define __RAW_SPIN_LOCK_UNLOCKED { 0 } +#define __ARCH_SPIN_LOCK_UNLOCKED { 0 } typedef struct { volatile unsigned int lock; diff --git a/arch/parisc/include/asm/spinlock_types.h b/arch/parisc/include/asm/spinlock_types.h index 735caafb81f5..396d2746ca57 100644 --- a/arch/parisc/include/asm/spinlock_types.h +++ b/arch/parisc/include/asm/spinlock_types.h @@ -4,10 +4,10 @@ typedef struct { #ifdef CONFIG_PA20 volatile unsigned int slock; -# define __RAW_SPIN_LOCK_UNLOCKED { 1 } +# define __ARCH_SPIN_LOCK_UNLOCKED { 1 } #else volatile unsigned int lock[4]; -# define __RAW_SPIN_LOCK_UNLOCKED { { 1, 1, 1, 1 } } +# define __ARCH_SPIN_LOCK_UNLOCKED { { 1, 1, 1, 1 } } #endif } arch_spinlock_t; @@ -16,6 +16,6 @@ typedef struct { volatile int counter; } raw_rwlock_t; -#define __RAW_RW_LOCK_UNLOCKED { __RAW_SPIN_LOCK_UNLOCKED, 0 } +#define __RAW_RW_LOCK_UNLOCKED { __ARCH_SPIN_LOCK_UNLOCKED, 0 } #endif diff --git a/arch/parisc/lib/bitops.c b/arch/parisc/lib/bitops.c index fdd7f583de54..353963d42059 100644 --- a/arch/parisc/lib/bitops.c +++ b/arch/parisc/lib/bitops.c @@ -13,7 +13,7 @@ #ifdef CONFIG_SMP arch_spinlock_t __atomic_hash[ATOMIC_HASH_SIZE] __lock_aligned = { - [0 ... (ATOMIC_HASH_SIZE-1)] = __RAW_SPIN_LOCK_UNLOCKED + [0 ... (ATOMIC_HASH_SIZE-1)] = __ARCH_SPIN_LOCK_UNLOCKED }; #endif diff --git a/arch/powerpc/include/asm/spinlock_types.h b/arch/powerpc/include/asm/spinlock_types.h index 4312e5baaf88..f5f39d82711f 100644 --- a/arch/powerpc/include/asm/spinlock_types.h +++ b/arch/powerpc/include/asm/spinlock_types.h @@ -9,7 +9,7 @@ typedef struct { volatile unsigned int slock; } arch_spinlock_t; -#define __RAW_SPIN_LOCK_UNLOCKED { 0 } +#define __ARCH_SPIN_LOCK_UNLOCKED { 0 } typedef struct { volatile signed int lock; diff --git a/arch/powerpc/kernel/rtas.c b/arch/powerpc/kernel/rtas.c index 579069c12152..57dfa414cfb8 100644 --- a/arch/powerpc/kernel/rtas.c +++ b/arch/powerpc/kernel/rtas.c @@ -42,7 +42,7 @@ #include struct rtas_t rtas = { - .lock = __RAW_SPIN_LOCK_UNLOCKED + .lock = __ARCH_SPIN_LOCK_UNLOCKED }; EXPORT_SYMBOL(rtas); diff --git a/arch/s390/include/asm/spinlock_types.h b/arch/s390/include/asm/spinlock_types.h index a93638eee3f7..e25c0370f6cd 100644 --- a/arch/s390/include/asm/spinlock_types.h +++ b/arch/s390/include/asm/spinlock_types.h @@ -9,7 +9,7 @@ typedef struct { volatile unsigned int owner_cpu; } __attribute__ ((aligned (4))) arch_spinlock_t; -#define __RAW_SPIN_LOCK_UNLOCKED { 0 } +#define __ARCH_SPIN_LOCK_UNLOCKED { 0 } typedef struct { volatile unsigned int lock; diff --git a/arch/sh/include/asm/spinlock_types.h b/arch/sh/include/asm/spinlock_types.h index 37712c32ba99..a3be2db960ed 100644 --- a/arch/sh/include/asm/spinlock_types.h +++ b/arch/sh/include/asm/spinlock_types.h @@ -9,7 +9,7 @@ typedef struct { volatile unsigned int lock; } arch_spinlock_t; -#define __RAW_SPIN_LOCK_UNLOCKED { 1 } +#define __ARCH_SPIN_LOCK_UNLOCKED { 1 } typedef struct { volatile unsigned int lock; diff --git a/arch/sparc/include/asm/spinlock_types.h b/arch/sparc/include/asm/spinlock_types.h index 41d9a8fec13d..c145e63a5d66 100644 --- a/arch/sparc/include/asm/spinlock_types.h +++ b/arch/sparc/include/asm/spinlock_types.h @@ -9,7 +9,7 @@ typedef struct { volatile unsigned char lock; } arch_spinlock_t; -#define __RAW_SPIN_LOCK_UNLOCKED { 0 } +#define __ARCH_SPIN_LOCK_UNLOCKED { 0 } typedef struct { volatile unsigned int lock; diff --git a/arch/x86/include/asm/spinlock_types.h b/arch/x86/include/asm/spinlock_types.h index 2ae7637ed524..696f8364a4f3 100644 --- a/arch/x86/include/asm/spinlock_types.h +++ b/arch/x86/include/asm/spinlock_types.h @@ -9,7 +9,7 @@ typedef struct arch_spinlock { unsigned int slock; } arch_spinlock_t; -#define __RAW_SPIN_LOCK_UNLOCKED { 0 } +#define __ARCH_SPIN_LOCK_UNLOCKED { 0 } typedef struct { unsigned int lock; diff --git a/arch/x86/kernel/dumpstack.c b/arch/x86/kernel/dumpstack.c index 0862d9d89c92..5b75afac8a38 100644 --- a/arch/x86/kernel/dumpstack.c +++ b/arch/x86/kernel/dumpstack.c @@ -188,7 +188,7 @@ void dump_stack(void) } EXPORT_SYMBOL(dump_stack); -static arch_spinlock_t die_lock = __RAW_SPIN_LOCK_UNLOCKED; +static arch_spinlock_t die_lock = __ARCH_SPIN_LOCK_UNLOCKED; static int die_owner = -1; static unsigned int die_nest_count; diff --git a/arch/x86/kernel/tsc_sync.c b/arch/x86/kernel/tsc_sync.c index 9f908b9d1abe..f1714697a09a 100644 --- a/arch/x86/kernel/tsc_sync.c +++ b/arch/x86/kernel/tsc_sync.c @@ -33,7 +33,7 @@ static __cpuinitdata atomic_t stop_count; * we want to have the fastest, inlined, non-debug version * of a critical section, to be able to prove TSC time-warps: */ -static __cpuinitdata arch_spinlock_t sync_lock = __RAW_SPIN_LOCK_UNLOCKED; +static __cpuinitdata arch_spinlock_t sync_lock = __ARCH_SPIN_LOCK_UNLOCKED; static __cpuinitdata cycles_t last_tsc; static __cpuinitdata cycles_t max_warp; diff --git a/include/linux/spinlock_types.h b/include/linux/spinlock_types.h index d4af2d7a86ea..7dadce303ebf 100644 --- a/include/linux/spinlock_types.h +++ b/include/linux/spinlock_types.h @@ -43,14 +43,14 @@ typedef struct { #ifdef CONFIG_DEBUG_SPINLOCK # define __SPIN_LOCK_UNLOCKED(lockname) \ - (spinlock_t) { .raw_lock = __RAW_SPIN_LOCK_UNLOCKED, \ + (spinlock_t) { .raw_lock = __ARCH_SPIN_LOCK_UNLOCKED, \ .magic = SPINLOCK_MAGIC, \ .owner = SPINLOCK_OWNER_INIT, \ .owner_cpu = -1, \ SPIN_DEP_MAP_INIT(lockname) } #else # define __SPIN_LOCK_UNLOCKED(lockname) \ - (spinlock_t) { .raw_lock = __RAW_SPIN_LOCK_UNLOCKED, \ + (spinlock_t) { .raw_lock = __ARCH_SPIN_LOCK_UNLOCKED, \ SPIN_DEP_MAP_INIT(lockname) } #endif diff --git a/include/linux/spinlock_types_up.h b/include/linux/spinlock_types_up.h index 34d36691c4ec..10db021f4875 100644 --- a/include/linux/spinlock_types_up.h +++ b/include/linux/spinlock_types_up.h @@ -18,13 +18,13 @@ typedef struct { volatile unsigned int slock; } arch_spinlock_t; -#define __RAW_SPIN_LOCK_UNLOCKED { 1 } +#define __ARCH_SPIN_LOCK_UNLOCKED { 1 } #else typedef struct { } arch_spinlock_t; -#define __RAW_SPIN_LOCK_UNLOCKED { } +#define __ARCH_SPIN_LOCK_UNLOCKED { } #endif diff --git a/kernel/lockdep.c b/kernel/lockdep.c index 7cc50c62af59..2389e3f85cf6 100644 --- a/kernel/lockdep.c +++ b/kernel/lockdep.c @@ -73,7 +73,7 @@ module_param(lock_stat, int, 0644); * to use a raw spinlock - we really dont want the spinlock * code to recurse back into the lockdep code... */ -static arch_spinlock_t lockdep_lock = (arch_spinlock_t)__RAW_SPIN_LOCK_UNLOCKED; +static arch_spinlock_t lockdep_lock = (arch_spinlock_t)__ARCH_SPIN_LOCK_UNLOCKED; static int graph_lock(void) { diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c index 5ac8ee0a9e35..fb7a0fa508b9 100644 --- a/kernel/trace/ring_buffer.c +++ b/kernel/trace/ring_buffer.c @@ -998,7 +998,7 @@ rb_allocate_cpu_buffer(struct ring_buffer *buffer, int cpu) cpu_buffer->buffer = buffer; spin_lock_init(&cpu_buffer->reader_lock); lockdep_set_class(&cpu_buffer->reader_lock, buffer->reader_lock_key); - cpu_buffer->lock = (arch_spinlock_t)__RAW_SPIN_LOCK_UNLOCKED; + cpu_buffer->lock = (arch_spinlock_t)__ARCH_SPIN_LOCK_UNLOCKED; bpage = kzalloc_node(ALIGN(sizeof(*bpage), cache_line_size()), GFP_KERNEL, cpu_to_node(cpu)); diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 7d56cecc2c6e..63bc1cc38219 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -501,7 +501,7 @@ static ssize_t trace_seq_to_buffer(struct trace_seq *s, void *buf, size_t cnt) * CONFIG_TRACER_MAX_TRACE. */ static arch_spinlock_t ftrace_max_lock = - (arch_spinlock_t)__RAW_SPIN_LOCK_UNLOCKED; + (arch_spinlock_t)__ARCH_SPIN_LOCK_UNLOCKED; #ifdef CONFIG_TRACER_MAX_TRACE unsigned long __read_mostly tracing_max_latency; @@ -802,7 +802,7 @@ static unsigned map_pid_to_cmdline[PID_MAX_DEFAULT+1]; static unsigned map_cmdline_to_pid[SAVED_CMDLINES]; static char saved_cmdlines[SAVED_CMDLINES][TASK_COMM_LEN]; static int cmdline_idx; -static arch_spinlock_t trace_cmdline_lock = __RAW_SPIN_LOCK_UNLOCKED; +static arch_spinlock_t trace_cmdline_lock = __ARCH_SPIN_LOCK_UNLOCKED; /* temporary disable recording */ static atomic_t trace_record_cmdline_disabled __read_mostly; @@ -1252,7 +1252,7 @@ ftrace_special(unsigned long arg1, unsigned long arg2, unsigned long arg3) int trace_vbprintk(unsigned long ip, const char *fmt, va_list args) { static arch_spinlock_t trace_buf_lock = - (arch_spinlock_t)__RAW_SPIN_LOCK_UNLOCKED; + (arch_spinlock_t)__ARCH_SPIN_LOCK_UNLOCKED; static u32 trace_buf[TRACE_BUF_SIZE]; struct ftrace_event_call *call = &event_bprint; @@ -1334,7 +1334,7 @@ int trace_array_printk(struct trace_array *tr, int trace_array_vprintk(struct trace_array *tr, unsigned long ip, const char *fmt, va_list args) { - static arch_spinlock_t trace_buf_lock = __RAW_SPIN_LOCK_UNLOCKED; + static arch_spinlock_t trace_buf_lock = __ARCH_SPIN_LOCK_UNLOCKED; static char trace_buf[TRACE_BUF_SIZE]; struct ftrace_event_call *call = &event_print; @@ -4308,7 +4308,7 @@ trace_printk_seq(struct trace_seq *s) static void __ftrace_dump(bool disable_tracing) { static arch_spinlock_t ftrace_dump_lock = - (arch_spinlock_t)__RAW_SPIN_LOCK_UNLOCKED; + (arch_spinlock_t)__ARCH_SPIN_LOCK_UNLOCKED; /* use static because iter can be a bit big for the stack */ static struct trace_iterator iter; unsigned int old_userobj; diff --git a/kernel/trace/trace_clock.c b/kernel/trace/trace_clock.c index 206ec3d4b3c2..433e2eda2d01 100644 --- a/kernel/trace/trace_clock.c +++ b/kernel/trace/trace_clock.c @@ -74,7 +74,7 @@ static struct { arch_spinlock_t lock; } trace_clock_struct ____cacheline_aligned_in_smp = { - .lock = (arch_spinlock_t)__RAW_SPIN_LOCK_UNLOCKED, + .lock = (arch_spinlock_t)__ARCH_SPIN_LOCK_UNLOCKED, }; u64 notrace trace_clock_global(void) diff --git a/kernel/trace/trace_sched_wakeup.c b/kernel/trace/trace_sched_wakeup.c index 4cf7e83ec235..e347853564e9 100644 --- a/kernel/trace/trace_sched_wakeup.c +++ b/kernel/trace/trace_sched_wakeup.c @@ -29,7 +29,7 @@ static unsigned wakeup_prio = -1; static int wakeup_rt; static arch_spinlock_t wakeup_lock = - (arch_spinlock_t)__RAW_SPIN_LOCK_UNLOCKED; + (arch_spinlock_t)__ARCH_SPIN_LOCK_UNLOCKED; static void __wakeup_reset(struct trace_array *tr); diff --git a/kernel/trace/trace_stack.c b/kernel/trace/trace_stack.c index 9a82d568fdec..728c35221483 100644 --- a/kernel/trace/trace_stack.c +++ b/kernel/trace/trace_stack.c @@ -28,7 +28,7 @@ static struct stack_trace max_stack_trace = { static unsigned long max_stack_size; static arch_spinlock_t max_stack_lock = - (arch_spinlock_t)__RAW_SPIN_LOCK_UNLOCKED; + (arch_spinlock_t)__ARCH_SPIN_LOCK_UNLOCKED; static int stack_trace_disabled __read_mostly; static DEFINE_PER_CPU(int, trace_active); diff --git a/lib/spinlock_debug.c b/lib/spinlock_debug.c index 2acd501b3826..f73004137141 100644 --- a/lib/spinlock_debug.c +++ b/lib/spinlock_debug.c @@ -23,7 +23,7 @@ void __spin_lock_init(spinlock_t *lock, const char *name, debug_check_no_locks_freed((void *)lock, sizeof(*lock)); lockdep_init_map(&lock->dep_map, name, key, 0); #endif - lock->raw_lock = (arch_spinlock_t)__RAW_SPIN_LOCK_UNLOCKED; + lock->raw_lock = (arch_spinlock_t)__ARCH_SPIN_LOCK_UNLOCKED; lock->magic = SPINLOCK_MAGIC; lock->owner = SPINLOCK_OWNER_INIT; lock->owner_cpu = -1; -- cgit v1.2.3-71-gd317 From 0199c4e68d1f02894bdefe4b5d9e9ee4aedd8d62 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Wed, 2 Dec 2009 20:01:25 +0100 Subject: locking: Convert __raw_spin* functions to arch_spin* Name space cleanup. No functional change. Signed-off-by: Thomas Gleixner Acked-by: Peter Zijlstra Acked-by: David S. Miller Acked-by: Ingo Molnar Cc: linux-arch@vger.kernel.org --- arch/alpha/include/asm/spinlock.h | 18 ++++++------ arch/arm/include/asm/spinlock.h | 20 ++++++------- arch/blackfin/include/asm/spinlock.h | 20 ++++++------- arch/cris/include/arch-v32/arch/spinlock.h | 46 +++++++++++++++--------------- arch/ia64/include/asm/bitops.h | 2 +- arch/ia64/include/asm/spinlock.h | 26 ++++++++--------- arch/m32r/include/asm/spinlock.h | 28 +++++++++--------- arch/mips/include/asm/spinlock.h | 36 +++++++++++------------ arch/parisc/include/asm/atomic.h | 4 +-- arch/parisc/include/asm/spinlock.h | 44 ++++++++++++++-------------- arch/powerpc/include/asm/spinlock.h | 32 ++++++++++----------- arch/powerpc/kernel/rtas.c | 12 ++++---- arch/powerpc/lib/locks.c | 4 +-- arch/powerpc/platforms/pasemi/setup.c | 8 +++--- arch/s390/include/asm/spinlock.h | 34 +++++++++++----------- arch/s390/lib/spinlock.c | 22 +++++++------- arch/sh/include/asm/spinlock.h | 26 ++++++++--------- arch/sparc/include/asm/spinlock_32.h | 20 ++++++------- arch/sparc/include/asm/spinlock_64.h | 18 ++++++------ arch/x86/include/asm/paravirt.h | 14 ++++----- arch/x86/include/asm/spinlock.h | 26 ++++++++--------- arch/x86/kernel/dumpstack.c | 6 ++-- arch/x86/kernel/paravirt-spinlocks.c | 2 +- arch/x86/kernel/tsc_sync.c | 8 +++--- include/asm-generic/bitops/atomic.h | 4 +-- include/linux/spinlock.h | 22 +++++++------- include/linux/spinlock_up.h | 26 ++++++++--------- kernel/lockdep.c | 18 ++++++------ kernel/mutex-debug.h | 4 +-- kernel/spinlock.c | 4 +-- kernel/trace/ring_buffer.c | 12 ++++---- kernel/trace/trace.c | 32 ++++++++++----------- kernel/trace/trace_clock.c | 4 +-- kernel/trace/trace_sched_wakeup.c | 12 ++++---- kernel/trace/trace_selftest.c | 4 +-- kernel/trace/trace_stack.c | 12 ++++---- lib/spinlock_debug.c | 8 +++--- 37 files changed, 319 insertions(+), 319 deletions(-) (limited to 'kernel') diff --git a/arch/alpha/include/asm/spinlock.h b/arch/alpha/include/asm/spinlock.h index bdb26a1940b4..4dac79f504c3 100644 --- a/arch/alpha/include/asm/spinlock.h +++ b/arch/alpha/include/asm/spinlock.h @@ -12,18 +12,18 @@ * We make no fairness assumptions. They have a cost. */ -#define __raw_spin_lock_flags(lock, flags) __raw_spin_lock(lock) -#define __raw_spin_is_locked(x) ((x)->lock != 0) -#define __raw_spin_unlock_wait(x) \ +#define arch_spin_lock_flags(lock, flags) arch_spin_lock(lock) +#define arch_spin_is_locked(x) ((x)->lock != 0) +#define arch_spin_unlock_wait(x) \ do { cpu_relax(); } while ((x)->lock) -static inline void __raw_spin_unlock(arch_spinlock_t * lock) +static inline void arch_spin_unlock(arch_spinlock_t * lock) { mb(); lock->lock = 0; } -static inline void __raw_spin_lock(arch_spinlock_t * lock) +static inline void arch_spin_lock(arch_spinlock_t * lock) { long tmp; @@ -43,7 +43,7 @@ static inline void __raw_spin_lock(arch_spinlock_t * lock) : "m"(lock->lock) : "memory"); } -static inline int __raw_spin_trylock(arch_spinlock_t *lock) +static inline int arch_spin_trylock(arch_spinlock_t *lock) { return !test_and_set_bit(0, &lock->lock); } @@ -169,8 +169,8 @@ static inline void __raw_write_unlock(raw_rwlock_t * lock) #define __raw_read_lock_flags(lock, flags) __raw_read_lock(lock) #define __raw_write_lock_flags(lock, flags) __raw_write_lock(lock) -#define _raw_spin_relax(lock) cpu_relax() -#define _raw_read_relax(lock) cpu_relax() -#define _raw_write_relax(lock) cpu_relax() +#define arch_spin_relax(lock) cpu_relax() +#define arch_read_relax(lock) cpu_relax() +#define arch_write_relax(lock) cpu_relax() #endif /* _ALPHA_SPINLOCK_H */ diff --git a/arch/arm/include/asm/spinlock.h b/arch/arm/include/asm/spinlock.h index 4e7712ee9394..de62eb098f68 100644 --- a/arch/arm/include/asm/spinlock.h +++ b/arch/arm/include/asm/spinlock.h @@ -17,13 +17,13 @@ * Locked value: 1 */ -#define __raw_spin_is_locked(x) ((x)->lock != 0) -#define __raw_spin_unlock_wait(lock) \ - do { while (__raw_spin_is_locked(lock)) cpu_relax(); } while (0) +#define arch_spin_is_locked(x) ((x)->lock != 0) +#define arch_spin_unlock_wait(lock) \ + do { while (arch_spin_is_locked(lock)) cpu_relax(); } while (0) -#define __raw_spin_lock_flags(lock, flags) __raw_spin_lock(lock) +#define arch_spin_lock_flags(lock, flags) arch_spin_lock(lock) -static inline void __raw_spin_lock(arch_spinlock_t *lock) +static inline void arch_spin_lock(arch_spinlock_t *lock) { unsigned long tmp; @@ -43,7 +43,7 @@ static inline void __raw_spin_lock(arch_spinlock_t *lock) smp_mb(); } -static inline int __raw_spin_trylock(arch_spinlock_t *lock) +static inline int arch_spin_trylock(arch_spinlock_t *lock) { unsigned long tmp; @@ -63,7 +63,7 @@ static inline int __raw_spin_trylock(arch_spinlock_t *lock) } } -static inline void __raw_spin_unlock(arch_spinlock_t *lock) +static inline void arch_spin_unlock(arch_spinlock_t *lock) { smp_mb(); @@ -220,8 +220,8 @@ static inline int __raw_read_trylock(raw_rwlock_t *rw) #define __raw_read_lock_flags(lock, flags) __raw_read_lock(lock) #define __raw_write_lock_flags(lock, flags) __raw_write_lock(lock) -#define _raw_spin_relax(lock) cpu_relax() -#define _raw_read_relax(lock) cpu_relax() -#define _raw_write_relax(lock) cpu_relax() +#define arch_spin_relax(lock) cpu_relax() +#define arch_read_relax(lock) cpu_relax() +#define arch_write_relax(lock) cpu_relax() #endif /* __ASM_SPINLOCK_H */ diff --git a/arch/blackfin/include/asm/spinlock.h b/arch/blackfin/include/asm/spinlock.h index fc16b4c5309b..62d49540e02b 100644 --- a/arch/blackfin/include/asm/spinlock.h +++ b/arch/blackfin/include/asm/spinlock.h @@ -24,31 +24,31 @@ asmlinkage void __raw_write_lock_asm(volatile int *ptr); asmlinkage int __raw_write_trylock_asm(volatile int *ptr); asmlinkage void __raw_write_unlock_asm(volatile int *ptr); -static inline int __raw_spin_is_locked(arch_spinlock_t *lock) +static inline int arch_spin_is_locked(arch_spinlock_t *lock) { return __raw_spin_is_locked_asm(&lock->lock); } -static inline void __raw_spin_lock(arch_spinlock_t *lock) +static inline void arch_spin_lock(arch_spinlock_t *lock) { __raw_spin_lock_asm(&lock->lock); } -#define __raw_spin_lock_flags(lock, flags) __raw_spin_lock(lock) +#define arch_spin_lock_flags(lock, flags) arch_spin_lock(lock) -static inline int __raw_spin_trylock(arch_spinlock_t *lock) +static inline int arch_spin_trylock(arch_spinlock_t *lock) { return __raw_spin_trylock_asm(&lock->lock); } -static inline void __raw_spin_unlock(arch_spinlock_t *lock) +static inline void arch_spin_unlock(arch_spinlock_t *lock) { __raw_spin_unlock_asm(&lock->lock); } -static inline void __raw_spin_unlock_wait(arch_spinlock_t *lock) +static inline void arch_spin_unlock_wait(arch_spinlock_t *lock) { - while (__raw_spin_is_locked(lock)) + while (arch_spin_is_locked(lock)) cpu_relax(); } @@ -92,9 +92,9 @@ static inline void __raw_write_unlock(raw_rwlock_t *rw) __raw_write_unlock_asm(&rw->lock); } -#define _raw_spin_relax(lock) cpu_relax() -#define _raw_read_relax(lock) cpu_relax() -#define _raw_write_relax(lock) cpu_relax() +#define arch_spin_relax(lock) cpu_relax() +#define arch_read_relax(lock) cpu_relax() +#define arch_write_relax(lock) cpu_relax() #endif diff --git a/arch/cris/include/arch-v32/arch/spinlock.h b/arch/cris/include/arch-v32/arch/spinlock.h index e253457765f2..a2e8a394d555 100644 --- a/arch/cris/include/arch-v32/arch/spinlock.h +++ b/arch/cris/include/arch-v32/arch/spinlock.h @@ -9,12 +9,12 @@ extern void cris_spin_unlock(void *l, int val); extern void cris_spin_lock(void *l); extern int cris_spin_trylock(void *l); -static inline int __raw_spin_is_locked(arch_spinlock_t *x) +static inline int arch_spin_is_locked(arch_spinlock_t *x) { return *(volatile signed char *)(&(x)->slock) <= 0; } -static inline void __raw_spin_unlock(arch_spinlock_t *lock) +static inline void arch_spin_unlock(arch_spinlock_t *lock) { __asm__ volatile ("move.d %1,%0" \ : "=m" (lock->slock) \ @@ -22,26 +22,26 @@ static inline void __raw_spin_unlock(arch_spinlock_t *lock) : "memory"); } -static inline void __raw_spin_unlock_wait(arch_spinlock_t *lock) +static inline void arch_spin_unlock_wait(arch_spinlock_t *lock) { - while (__raw_spin_is_locked(lock)) + while (arch_spin_is_locked(lock)) cpu_relax(); } -static inline int __raw_spin_trylock(arch_spinlock_t *lock) +static inline int arch_spin_trylock(arch_spinlock_t *lock) { return cris_spin_trylock((void *)&lock->slock); } -static inline void __raw_spin_lock(arch_spinlock_t *lock) +static inline void arch_spin_lock(arch_spinlock_t *lock) { cris_spin_lock((void *)&lock->slock); } static inline void -__raw_spin_lock_flags(arch_spinlock_t *lock, unsigned long flags) +arch_spin_lock_flags(arch_spinlock_t *lock, unsigned long flags) { - __raw_spin_lock(lock); + arch_spin_lock(lock); } /* @@ -68,64 +68,64 @@ static inline int __raw_write_can_lock(raw_rwlock_t *x) static inline void __raw_read_lock(raw_rwlock_t *rw) { - __raw_spin_lock(&rw->slock); + arch_spin_lock(&rw->slock); while (rw->lock == 0); rw->lock--; - __raw_spin_unlock(&rw->slock); + arch_spin_unlock(&rw->slock); } static inline void __raw_write_lock(raw_rwlock_t *rw) { - __raw_spin_lock(&rw->slock); + arch_spin_lock(&rw->slock); while (rw->lock != RW_LOCK_BIAS); rw->lock = 0; - __raw_spin_unlock(&rw->slock); + arch_spin_unlock(&rw->slock); } static inline void __raw_read_unlock(raw_rwlock_t *rw) { - __raw_spin_lock(&rw->slock); + arch_spin_lock(&rw->slock); rw->lock++; - __raw_spin_unlock(&rw->slock); + arch_spin_unlock(&rw->slock); } static inline void __raw_write_unlock(raw_rwlock_t *rw) { - __raw_spin_lock(&rw->slock); + arch_spin_lock(&rw->slock); while (rw->lock != RW_LOCK_BIAS); rw->lock = RW_LOCK_BIAS; - __raw_spin_unlock(&rw->slock); + arch_spin_unlock(&rw->slock); } static inline int __raw_read_trylock(raw_rwlock_t *rw) { int ret = 0; - __raw_spin_lock(&rw->slock); + arch_spin_lock(&rw->slock); if (rw->lock != 0) { rw->lock--; ret = 1; } - __raw_spin_unlock(&rw->slock); + arch_spin_unlock(&rw->slock); return ret; } static inline int __raw_write_trylock(raw_rwlock_t *rw) { int ret = 0; - __raw_spin_lock(&rw->slock); + arch_spin_lock(&rw->slock); if (rw->lock == RW_LOCK_BIAS) { rw->lock = 0; ret = 1; } - __raw_spin_unlock(&rw->slock); + arch_spin_unlock(&rw->slock); return 1; } #define _raw_read_lock_flags(lock, flags) _raw_read_lock(lock) #define _raw_write_lock_flags(lock, flags) _raw_write_lock(lock) -#define _raw_spin_relax(lock) cpu_relax() -#define _raw_read_relax(lock) cpu_relax() -#define _raw_write_relax(lock) cpu_relax() +#define arch_spin_relax(lock) cpu_relax() +#define arch_read_relax(lock) cpu_relax() +#define arch_write_relax(lock) cpu_relax() #endif /* __ASM_ARCH_SPINLOCK_H */ diff --git a/arch/ia64/include/asm/bitops.h b/arch/ia64/include/asm/bitops.h index 57a2787bc9fb..6ebc229a1c51 100644 --- a/arch/ia64/include/asm/bitops.h +++ b/arch/ia64/include/asm/bitops.h @@ -127,7 +127,7 @@ clear_bit_unlock (int nr, volatile void *addr) * @addr: Address to start counting from * * Similarly to clear_bit_unlock, the implementation uses a store - * with release semantics. See also __raw_spin_unlock(). + * with release semantics. See also arch_spin_unlock(). */ static __inline__ void __clear_bit_unlock(int nr, void *addr) diff --git a/arch/ia64/include/asm/spinlock.h b/arch/ia64/include/asm/spinlock.h index 9fbdf7e61087..b06165f6352f 100644 --- a/arch/ia64/include/asm/spinlock.h +++ b/arch/ia64/include/asm/spinlock.h @@ -17,7 +17,7 @@ #include #include -#define __raw_spin_lock_init(x) ((x)->lock = 0) +#define arch_spin_lock_init(x) ((x)->lock = 0) /* * Ticket locks are conceptually two parts, one indicating the current head of @@ -103,39 +103,39 @@ static inline int __ticket_spin_is_contended(arch_spinlock_t *lock) return ((tmp - (tmp >> TICKET_SHIFT)) & TICKET_MASK) > 1; } -static inline int __raw_spin_is_locked(arch_spinlock_t *lock) +static inline int arch_spin_is_locked(arch_spinlock_t *lock) { return __ticket_spin_is_locked(lock); } -static inline int __raw_spin_is_contended(arch_spinlock_t *lock) +static inline int arch_spin_is_contended(arch_spinlock_t *lock) { return __ticket_spin_is_contended(lock); } -#define __raw_spin_is_contended __raw_spin_is_contended +#define arch_spin_is_contended arch_spin_is_contended -static __always_inline void __raw_spin_lock(arch_spinlock_t *lock) +static __always_inline void arch_spin_lock(arch_spinlock_t *lock) { __ticket_spin_lock(lock); } -static __always_inline int __raw_spin_trylock(arch_spinlock_t *lock) +static __always_inline int arch_spin_trylock(arch_spinlock_t *lock) { return __ticket_spin_trylock(lock); } -static __always_inline void __raw_spin_unlock(arch_spinlock_t *lock) +static __always_inline void arch_spin_unlock(arch_spinlock_t *lock) { __ticket_spin_unlock(lock); } -static __always_inline void __raw_spin_lock_flags(arch_spinlock_t *lock, +static __always_inline void arch_spin_lock_flags(arch_spinlock_t *lock, unsigned long flags) { - __raw_spin_lock(lock); + arch_spin_lock(lock); } -static inline void __raw_spin_unlock_wait(arch_spinlock_t *lock) +static inline void arch_spin_unlock_wait(arch_spinlock_t *lock) { __ticket_spin_unlock_wait(lock); } @@ -285,8 +285,8 @@ static inline int __raw_read_trylock(raw_rwlock_t *x) return (u32)ia64_cmpxchg4_acq((__u32 *)(x), new.word, old.word) == old.word; } -#define _raw_spin_relax(lock) cpu_relax() -#define _raw_read_relax(lock) cpu_relax() -#define _raw_write_relax(lock) cpu_relax() +#define arch_spin_relax(lock) cpu_relax() +#define arch_read_relax(lock) cpu_relax() +#define arch_write_relax(lock) cpu_relax() #endif /* _ASM_IA64_SPINLOCK_H */ diff --git a/arch/m32r/include/asm/spinlock.h b/arch/m32r/include/asm/spinlock.h index 0c0164225bc0..8acac950a43c 100644 --- a/arch/m32r/include/asm/spinlock.h +++ b/arch/m32r/include/asm/spinlock.h @@ -24,19 +24,19 @@ * We make no fairness assumptions. They have a cost. */ -#define __raw_spin_is_locked(x) (*(volatile int *)(&(x)->slock) <= 0) -#define __raw_spin_lock_flags(lock, flags) __raw_spin_lock(lock) -#define __raw_spin_unlock_wait(x) \ - do { cpu_relax(); } while (__raw_spin_is_locked(x)) +#define arch_spin_is_locked(x) (*(volatile int *)(&(x)->slock) <= 0) +#define arch_spin_lock_flags(lock, flags) arch_spin_lock(lock) +#define arch_spin_unlock_wait(x) \ + do { cpu_relax(); } while (arch_spin_is_locked(x)) /** - * __raw_spin_trylock - Try spin lock and return a result + * arch_spin_trylock - Try spin lock and return a result * @lock: Pointer to the lock variable * - * __raw_spin_trylock() tries to get the lock and returns a result. + * arch_spin_trylock() tries to get the lock and returns a result. * On the m32r, the result value is 1 (= Success) or 0 (= Failure). */ -static inline int __raw_spin_trylock(arch_spinlock_t *lock) +static inline int arch_spin_trylock(arch_spinlock_t *lock) { int oldval; unsigned long tmp1, tmp2; @@ -50,7 +50,7 @@ static inline int __raw_spin_trylock(arch_spinlock_t *lock) * } */ __asm__ __volatile__ ( - "# __raw_spin_trylock \n\t" + "# arch_spin_trylock \n\t" "ldi %1, #0; \n\t" "mvfc %2, psw; \n\t" "clrpsw #0x40 -> nop; \n\t" @@ -69,7 +69,7 @@ static inline int __raw_spin_trylock(arch_spinlock_t *lock) return (oldval > 0); } -static inline void __raw_spin_lock(arch_spinlock_t *lock) +static inline void arch_spin_lock(arch_spinlock_t *lock) { unsigned long tmp0, tmp1; @@ -84,7 +84,7 @@ static inline void __raw_spin_lock(arch_spinlock_t *lock) * } */ __asm__ __volatile__ ( - "# __raw_spin_lock \n\t" + "# arch_spin_lock \n\t" ".fillinsn \n" "1: \n\t" "mvfc %1, psw; \n\t" @@ -111,7 +111,7 @@ static inline void __raw_spin_lock(arch_spinlock_t *lock) ); } -static inline void __raw_spin_unlock(arch_spinlock_t *lock) +static inline void arch_spin_unlock(arch_spinlock_t *lock) { mb(); lock->slock = 1; @@ -319,8 +319,8 @@ static inline int __raw_write_trylock(raw_rwlock_t *lock) #define __raw_read_lock_flags(lock, flags) __raw_read_lock(lock) #define __raw_write_lock_flags(lock, flags) __raw_write_lock(lock) -#define _raw_spin_relax(lock) cpu_relax() -#define _raw_read_relax(lock) cpu_relax() -#define _raw_write_relax(lock) cpu_relax() +#define arch_spin_relax(lock) cpu_relax() +#define arch_read_relax(lock) cpu_relax() +#define arch_write_relax(lock) cpu_relax() #endif /* _ASM_M32R_SPINLOCK_H */ diff --git a/arch/mips/include/asm/spinlock.h b/arch/mips/include/asm/spinlock.h index 0f16d0673b4a..95edebaaf22a 100644 --- a/arch/mips/include/asm/spinlock.h +++ b/arch/mips/include/asm/spinlock.h @@ -34,33 +34,33 @@ * becomes equal to the the initial value of the tail. */ -static inline int __raw_spin_is_locked(arch_spinlock_t *lock) +static inline int arch_spin_is_locked(arch_spinlock_t *lock) { unsigned int counters = ACCESS_ONCE(lock->lock); return ((counters >> 14) ^ counters) & 0x1fff; } -#define __raw_spin_lock_flags(lock, flags) __raw_spin_lock(lock) -#define __raw_spin_unlock_wait(x) \ - while (__raw_spin_is_locked(x)) { cpu_relax(); } +#define arch_spin_lock_flags(lock, flags) arch_spin_lock(lock) +#define arch_spin_unlock_wait(x) \ + while (arch_spin_is_locked(x)) { cpu_relax(); } -static inline int __raw_spin_is_contended(arch_spinlock_t *lock) +static inline int arch_spin_is_contended(arch_spinlock_t *lock) { unsigned int counters = ACCESS_ONCE(lock->lock); return (((counters >> 14) - counters) & 0x1fff) > 1; } -#define __raw_spin_is_contended __raw_spin_is_contended +#define arch_spin_is_contended arch_spin_is_contended -static inline void __raw_spin_lock(arch_spinlock_t *lock) +static inline void arch_spin_lock(arch_spinlock_t *lock) { int my_ticket; int tmp; if (R10000_LLSC_WAR) { __asm__ __volatile__ ( - " .set push # __raw_spin_lock \n" + " .set push # arch_spin_lock \n" " .set noreorder \n" " \n" "1: ll %[ticket], %[ticket_ptr] \n" @@ -94,7 +94,7 @@ static inline void __raw_spin_lock(arch_spinlock_t *lock) [my_ticket] "=&r" (my_ticket)); } else { __asm__ __volatile__ ( - " .set push # __raw_spin_lock \n" + " .set push # arch_spin_lock \n" " .set noreorder \n" " \n" " ll %[ticket], %[ticket_ptr] \n" @@ -134,7 +134,7 @@ static inline void __raw_spin_lock(arch_spinlock_t *lock) smp_llsc_mb(); } -static inline void __raw_spin_unlock(arch_spinlock_t *lock) +static inline void arch_spin_unlock(arch_spinlock_t *lock) { int tmp; @@ -142,7 +142,7 @@ static inline void __raw_spin_unlock(arch_spinlock_t *lock) if (R10000_LLSC_WAR) { __asm__ __volatile__ ( - " # __raw_spin_unlock \n" + " # arch_spin_unlock \n" "1: ll %[ticket], %[ticket_ptr] \n" " addiu %[ticket], %[ticket], 1 \n" " ori %[ticket], %[ticket], 0x2000 \n" @@ -153,7 +153,7 @@ static inline void __raw_spin_unlock(arch_spinlock_t *lock) [ticket] "=&r" (tmp)); } else { __asm__ __volatile__ ( - " .set push # __raw_spin_unlock \n" + " .set push # arch_spin_unlock \n" " .set noreorder \n" " \n" " ll %[ticket], %[ticket_ptr] \n" @@ -174,13 +174,13 @@ static inline void __raw_spin_unlock(arch_spinlock_t *lock) } } -static inline unsigned int __raw_spin_trylock(arch_spinlock_t *lock) +static inline unsigned int arch_spin_trylock(arch_spinlock_t *lock) { int tmp, tmp2, tmp3; if (R10000_LLSC_WAR) { __asm__ __volatile__ ( - " .set push # __raw_spin_trylock \n" + " .set push # arch_spin_trylock \n" " .set noreorder \n" " \n" "1: ll %[ticket], %[ticket_ptr] \n" @@ -204,7 +204,7 @@ static inline unsigned int __raw_spin_trylock(arch_spinlock_t *lock) [now_serving] "=&r" (tmp3)); } else { __asm__ __volatile__ ( - " .set push # __raw_spin_trylock \n" + " .set push # arch_spin_trylock \n" " .set noreorder \n" " \n" " ll %[ticket], %[ticket_ptr] \n" @@ -483,8 +483,8 @@ static inline int __raw_write_trylock(raw_rwlock_t *rw) #define __raw_read_lock_flags(lock, flags) __raw_read_lock(lock) #define __raw_write_lock_flags(lock, flags) __raw_write_lock(lock) -#define _raw_spin_relax(lock) cpu_relax() -#define _raw_read_relax(lock) cpu_relax() -#define _raw_write_relax(lock) cpu_relax() +#define arch_spin_relax(lock) cpu_relax() +#define arch_read_relax(lock) cpu_relax() +#define arch_write_relax(lock) cpu_relax() #endif /* _ASM_SPINLOCK_H */ diff --git a/arch/parisc/include/asm/atomic.h b/arch/parisc/include/asm/atomic.h index 3a4ea778d4b6..716634d1f546 100644 --- a/arch/parisc/include/asm/atomic.h +++ b/arch/parisc/include/asm/atomic.h @@ -34,12 +34,12 @@ extern arch_spinlock_t __atomic_hash[ATOMIC_HASH_SIZE] __lock_aligned; #define _atomic_spin_lock_irqsave(l,f) do { \ arch_spinlock_t *s = ATOMIC_HASH(l); \ local_irq_save(f); \ - __raw_spin_lock(s); \ + arch_spin_lock(s); \ } while(0) #define _atomic_spin_unlock_irqrestore(l,f) do { \ arch_spinlock_t *s = ATOMIC_HASH(l); \ - __raw_spin_unlock(s); \ + arch_spin_unlock(s); \ local_irq_restore(f); \ } while(0) diff --git a/arch/parisc/include/asm/spinlock.h b/arch/parisc/include/asm/spinlock.h index 69e8dca26744..235e7e386e2a 100644 --- a/arch/parisc/include/asm/spinlock.h +++ b/arch/parisc/include/asm/spinlock.h @@ -5,17 +5,17 @@ #include #include -static inline int __raw_spin_is_locked(arch_spinlock_t *x) +static inline int arch_spin_is_locked(arch_spinlock_t *x) { volatile unsigned int *a = __ldcw_align(x); return *a == 0; } -#define __raw_spin_lock(lock) __raw_spin_lock_flags(lock, 0) -#define __raw_spin_unlock_wait(x) \ - do { cpu_relax(); } while (__raw_spin_is_locked(x)) +#define arch_spin_lock(lock) arch_spin_lock_flags(lock, 0) +#define arch_spin_unlock_wait(x) \ + do { cpu_relax(); } while (arch_spin_is_locked(x)) -static inline void __raw_spin_lock_flags(arch_spinlock_t *x, +static inline void arch_spin_lock_flags(arch_spinlock_t *x, unsigned long flags) { volatile unsigned int *a; @@ -33,7 +33,7 @@ static inline void __raw_spin_lock_flags(arch_spinlock_t *x, mb(); } -static inline void __raw_spin_unlock(arch_spinlock_t *x) +static inline void arch_spin_unlock(arch_spinlock_t *x) { volatile unsigned int *a; mb(); @@ -42,7 +42,7 @@ static inline void __raw_spin_unlock(arch_spinlock_t *x) mb(); } -static inline int __raw_spin_trylock(arch_spinlock_t *x) +static inline int arch_spin_trylock(arch_spinlock_t *x) { volatile unsigned int *a; int ret; @@ -73,9 +73,9 @@ static __inline__ void __raw_read_lock(raw_rwlock_t *rw) { unsigned long flags; local_irq_save(flags); - __raw_spin_lock_flags(&rw->lock, flags); + arch_spin_lock_flags(&rw->lock, flags); rw->counter++; - __raw_spin_unlock(&rw->lock); + arch_spin_unlock(&rw->lock); local_irq_restore(flags); } @@ -85,9 +85,9 @@ static __inline__ void __raw_read_unlock(raw_rwlock_t *rw) { unsigned long flags; local_irq_save(flags); - __raw_spin_lock_flags(&rw->lock, flags); + arch_spin_lock_flags(&rw->lock, flags); rw->counter--; - __raw_spin_unlock(&rw->lock); + arch_spin_unlock(&rw->lock); local_irq_restore(flags); } @@ -98,9 +98,9 @@ static __inline__ int __raw_read_trylock(raw_rwlock_t *rw) unsigned long flags; retry: local_irq_save(flags); - if (__raw_spin_trylock(&rw->lock)) { + if (arch_spin_trylock(&rw->lock)) { rw->counter++; - __raw_spin_unlock(&rw->lock); + arch_spin_unlock(&rw->lock); local_irq_restore(flags); return 1; } @@ -111,7 +111,7 @@ static __inline__ int __raw_read_trylock(raw_rwlock_t *rw) return 0; /* Wait until we have a realistic chance at the lock */ - while (__raw_spin_is_locked(&rw->lock) && rw->counter >= 0) + while (arch_spin_is_locked(&rw->lock) && rw->counter >= 0) cpu_relax(); goto retry; @@ -124,10 +124,10 @@ static __inline__ void __raw_write_lock(raw_rwlock_t *rw) unsigned long flags; retry: local_irq_save(flags); - __raw_spin_lock_flags(&rw->lock, flags); + arch_spin_lock_flags(&rw->lock, flags); if (rw->counter != 0) { - __raw_spin_unlock(&rw->lock); + arch_spin_unlock(&rw->lock); local_irq_restore(flags); while (rw->counter != 0) @@ -144,7 +144,7 @@ retry: static __inline__ void __raw_write_unlock(raw_rwlock_t *rw) { rw->counter = 0; - __raw_spin_unlock(&rw->lock); + arch_spin_unlock(&rw->lock); } /* Note that we have to ensure interrupts are disabled in case we're @@ -155,13 +155,13 @@ static __inline__ int __raw_write_trylock(raw_rwlock_t *rw) int result = 0; local_irq_save(flags); - if (__raw_spin_trylock(&rw->lock)) { + if (arch_spin_trylock(&rw->lock)) { if (rw->counter == 0) { rw->counter = -1; result = 1; } else { /* Read-locked. Oh well. */ - __raw_spin_unlock(&rw->lock); + arch_spin_unlock(&rw->lock); } } local_irq_restore(flags); @@ -190,8 +190,8 @@ static __inline__ int __raw_write_can_lock(raw_rwlock_t *rw) #define __raw_read_lock_flags(lock, flags) __raw_read_lock(lock) #define __raw_write_lock_flags(lock, flags) __raw_write_lock(lock) -#define _raw_spin_relax(lock) cpu_relax() -#define _raw_read_relax(lock) cpu_relax() -#define _raw_write_relax(lock) cpu_relax() +#define arch_spin_relax(lock) cpu_relax() +#define arch_read_relax(lock) cpu_relax() +#define arch_write_relax(lock) cpu_relax() #endif /* __ASM_SPINLOCK_H */ diff --git a/arch/powerpc/include/asm/spinlock.h b/arch/powerpc/include/asm/spinlock.h index c0d44c92ff0e..cdcaf6b97087 100644 --- a/arch/powerpc/include/asm/spinlock.h +++ b/arch/powerpc/include/asm/spinlock.h @@ -28,7 +28,7 @@ #include #include -#define __raw_spin_is_locked(x) ((x)->slock != 0) +#define arch_spin_is_locked(x) ((x)->slock != 0) #ifdef CONFIG_PPC64 /* use 0x800000yy when locked, where yy == CPU number */ @@ -54,7 +54,7 @@ * This returns the old value in the lock, so we succeeded * in getting the lock if the return value is 0. */ -static inline unsigned long arch_spin_trylock(arch_spinlock_t *lock) +static inline unsigned long __arch_spin_trylock(arch_spinlock_t *lock) { unsigned long tmp, token; @@ -73,10 +73,10 @@ static inline unsigned long arch_spin_trylock(arch_spinlock_t *lock) return tmp; } -static inline int __raw_spin_trylock(arch_spinlock_t *lock) +static inline int arch_spin_trylock(arch_spinlock_t *lock) { CLEAR_IO_SYNC; - return arch_spin_trylock(lock) == 0; + return __arch_spin_trylock(lock) == 0; } /* @@ -104,11 +104,11 @@ extern void __rw_yield(raw_rwlock_t *lock); #define SHARED_PROCESSOR 0 #endif -static inline void __raw_spin_lock(arch_spinlock_t *lock) +static inline void arch_spin_lock(arch_spinlock_t *lock) { CLEAR_IO_SYNC; while (1) { - if (likely(arch_spin_trylock(lock) == 0)) + if (likely(__arch_spin_trylock(lock) == 0)) break; do { HMT_low(); @@ -120,13 +120,13 @@ static inline void __raw_spin_lock(arch_spinlock_t *lock) } static inline -void __raw_spin_lock_flags(arch_spinlock_t *lock, unsigned long flags) +void arch_spin_lock_flags(arch_spinlock_t *lock, unsigned long flags) { unsigned long flags_dis; CLEAR_IO_SYNC; while (1) { - if (likely(arch_spin_trylock(lock) == 0)) + if (likely(__arch_spin_trylock(lock) == 0)) break; local_save_flags(flags_dis); local_irq_restore(flags); @@ -140,19 +140,19 @@ void __raw_spin_lock_flags(arch_spinlock_t *lock, unsigned long flags) } } -static inline void __raw_spin_unlock(arch_spinlock_t *lock) +static inline void arch_spin_unlock(arch_spinlock_t *lock) { SYNC_IO; - __asm__ __volatile__("# __raw_spin_unlock\n\t" + __asm__ __volatile__("# arch_spin_unlock\n\t" LWSYNC_ON_SMP: : :"memory"); lock->slock = 0; } #ifdef CONFIG_PPC64 -extern void __raw_spin_unlock_wait(arch_spinlock_t *lock); +extern void arch_spin_unlock_wait(arch_spinlock_t *lock); #else -#define __raw_spin_unlock_wait(lock) \ - do { while (__raw_spin_is_locked(lock)) cpu_relax(); } while (0) +#define arch_spin_unlock_wait(lock) \ + do { while (arch_spin_is_locked(lock)) cpu_relax(); } while (0) #endif /* @@ -290,9 +290,9 @@ static inline void __raw_write_unlock(raw_rwlock_t *rw) #define __raw_read_lock_flags(lock, flags) __raw_read_lock(lock) #define __raw_write_lock_flags(lock, flags) __raw_write_lock(lock) -#define _raw_spin_relax(lock) __spin_yield(lock) -#define _raw_read_relax(lock) __rw_yield(lock) -#define _raw_write_relax(lock) __rw_yield(lock) +#define arch_spin_relax(lock) __spin_yield(lock) +#define arch_read_relax(lock) __rw_yield(lock) +#define arch_write_relax(lock) __rw_yield(lock) #endif /* __KERNEL__ */ #endif /* __ASM_SPINLOCK_H */ diff --git a/arch/powerpc/kernel/rtas.c b/arch/powerpc/kernel/rtas.c index 57dfa414cfb8..fd0d29493fd6 100644 --- a/arch/powerpc/kernel/rtas.c +++ b/arch/powerpc/kernel/rtas.c @@ -80,13 +80,13 @@ static unsigned long lock_rtas(void) local_irq_save(flags); preempt_disable(); - __raw_spin_lock_flags(&rtas.lock, flags); + arch_spin_lock_flags(&rtas.lock, flags); return flags; } static void unlock_rtas(unsigned long flags) { - __raw_spin_unlock(&rtas.lock); + arch_spin_unlock(&rtas.lock); local_irq_restore(flags); preempt_enable(); } @@ -987,10 +987,10 @@ void __cpuinit rtas_give_timebase(void) local_irq_save(flags); hard_irq_disable(); - __raw_spin_lock(&timebase_lock); + arch_spin_lock(&timebase_lock); rtas_call(rtas_token("freeze-time-base"), 0, 1, NULL); timebase = get_tb(); - __raw_spin_unlock(&timebase_lock); + arch_spin_unlock(&timebase_lock); while (timebase) barrier(); @@ -1002,8 +1002,8 @@ void __cpuinit rtas_take_timebase(void) { while (!timebase) barrier(); - __raw_spin_lock(&timebase_lock); + arch_spin_lock(&timebase_lock); set_tb(timebase >> 32, timebase & 0xffffffff); timebase = 0; - __raw_spin_unlock(&timebase_lock); + arch_spin_unlock(&timebase_lock); } diff --git a/arch/powerpc/lib/locks.c b/arch/powerpc/lib/locks.c index b06294cde499..ee395e392115 100644 --- a/arch/powerpc/lib/locks.c +++ b/arch/powerpc/lib/locks.c @@ -82,7 +82,7 @@ void __rw_yield(raw_rwlock_t *rw) } #endif -void __raw_spin_unlock_wait(arch_spinlock_t *lock) +void arch_spin_unlock_wait(arch_spinlock_t *lock) { while (lock->slock) { HMT_low(); @@ -92,4 +92,4 @@ void __raw_spin_unlock_wait(arch_spinlock_t *lock) HMT_medium(); } -EXPORT_SYMBOL(__raw_spin_unlock_wait); +EXPORT_SYMBOL(arch_spin_unlock_wait); diff --git a/arch/powerpc/platforms/pasemi/setup.c b/arch/powerpc/platforms/pasemi/setup.c index be36fece41d7..242f8095c2df 100644 --- a/arch/powerpc/platforms/pasemi/setup.c +++ b/arch/powerpc/platforms/pasemi/setup.c @@ -80,11 +80,11 @@ static void __devinit pas_give_timebase(void) local_irq_save(flags); hard_irq_disable(); - __raw_spin_lock(&timebase_lock); + arch_spin_lock(&timebase_lock); mtspr(SPRN_TBCTL, TBCTL_FREEZE); isync(); timebase = get_tb(); - __raw_spin_unlock(&timebase_lock); + arch_spin_unlock(&timebase_lock); while (timebase) barrier(); @@ -97,10 +97,10 @@ static void __devinit pas_take_timebase(void) while (!timebase) smp_rmb(); - __raw_spin_lock(&timebase_lock); + arch_spin_lock(&timebase_lock); set_tb(timebase >> 32, timebase & 0xffffffff); timebase = 0; - __raw_spin_unlock(&timebase_lock); + arch_spin_unlock(&timebase_lock); } struct smp_ops_t pas_smp_ops = { diff --git a/arch/s390/include/asm/spinlock.h b/arch/s390/include/asm/spinlock.h index 6121fa4b83d9..a94c146657a9 100644 --- a/arch/s390/include/asm/spinlock.h +++ b/arch/s390/include/asm/spinlock.h @@ -52,27 +52,27 @@ _raw_compare_and_swap(volatile unsigned int *lock, * (the type definitions are in asm/spinlock_types.h) */ -#define __raw_spin_is_locked(x) ((x)->owner_cpu != 0) -#define __raw_spin_unlock_wait(lock) \ - do { while (__raw_spin_is_locked(lock)) \ - _raw_spin_relax(lock); } while (0) +#define arch_spin_is_locked(x) ((x)->owner_cpu != 0) +#define arch_spin_unlock_wait(lock) \ + do { while (arch_spin_is_locked(lock)) \ + arch_spin_relax(lock); } while (0) -extern void _raw_spin_lock_wait(arch_spinlock_t *); -extern void _raw_spin_lock_wait_flags(arch_spinlock_t *, unsigned long flags); -extern int _raw_spin_trylock_retry(arch_spinlock_t *); -extern void _raw_spin_relax(arch_spinlock_t *lock); +extern void arch_spin_lock_wait(arch_spinlock_t *); +extern void arch_spin_lock_wait_flags(arch_spinlock_t *, unsigned long flags); +extern int arch_spin_trylock_retry(arch_spinlock_t *); +extern void arch_spin_relax(arch_spinlock_t *lock); -static inline void __raw_spin_lock(arch_spinlock_t *lp) +static inline void arch_spin_lock(arch_spinlock_t *lp) { int old; old = _raw_compare_and_swap(&lp->owner_cpu, 0, ~smp_processor_id()); if (likely(old == 0)) return; - _raw_spin_lock_wait(lp); + arch_spin_lock_wait(lp); } -static inline void __raw_spin_lock_flags(arch_spinlock_t *lp, +static inline void arch_spin_lock_flags(arch_spinlock_t *lp, unsigned long flags) { int old; @@ -80,20 +80,20 @@ static inline void __raw_spin_lock_flags(arch_spinlock_t *lp, old = _raw_compare_and_swap(&lp->owner_cpu, 0, ~smp_processor_id()); if (likely(old == 0)) return; - _raw_spin_lock_wait_flags(lp, flags); + arch_spin_lock_wait_flags(lp, flags); } -static inline int __raw_spin_trylock(arch_spinlock_t *lp) +static inline int arch_spin_trylock(arch_spinlock_t *lp) { int old; old = _raw_compare_and_swap(&lp->owner_cpu, 0, ~smp_processor_id()); if (likely(old == 0)) return 1; - return _raw_spin_trylock_retry(lp); + return arch_spin_trylock_retry(lp); } -static inline void __raw_spin_unlock(arch_spinlock_t *lp) +static inline void arch_spin_unlock(arch_spinlock_t *lp) { _raw_compare_and_swap(&lp->owner_cpu, lp->owner_cpu, 0); } @@ -188,7 +188,7 @@ static inline int __raw_write_trylock(raw_rwlock_t *rw) return _raw_write_trylock_retry(rw); } -#define _raw_read_relax(lock) cpu_relax() -#define _raw_write_relax(lock) cpu_relax() +#define arch_read_relax(lock) cpu_relax() +#define arch_write_relax(lock) cpu_relax() #endif /* __ASM_SPINLOCK_H */ diff --git a/arch/s390/lib/spinlock.c b/arch/s390/lib/spinlock.c index d4cbf71a6077..f4596452f072 100644 --- a/arch/s390/lib/spinlock.c +++ b/arch/s390/lib/spinlock.c @@ -39,7 +39,7 @@ static inline void _raw_yield_cpu(int cpu) _raw_yield(); } -void _raw_spin_lock_wait(arch_spinlock_t *lp) +void arch_spin_lock_wait(arch_spinlock_t *lp) { int count = spin_retry; unsigned int cpu = ~smp_processor_id(); @@ -51,15 +51,15 @@ void _raw_spin_lock_wait(arch_spinlock_t *lp) _raw_yield_cpu(~owner); count = spin_retry; } - if (__raw_spin_is_locked(lp)) + if (arch_spin_is_locked(lp)) continue; if (_raw_compare_and_swap(&lp->owner_cpu, 0, cpu) == 0) return; } } -EXPORT_SYMBOL(_raw_spin_lock_wait); +EXPORT_SYMBOL(arch_spin_lock_wait); -void _raw_spin_lock_wait_flags(arch_spinlock_t *lp, unsigned long flags) +void arch_spin_lock_wait_flags(arch_spinlock_t *lp, unsigned long flags) { int count = spin_retry; unsigned int cpu = ~smp_processor_id(); @@ -72,7 +72,7 @@ void _raw_spin_lock_wait_flags(arch_spinlock_t *lp, unsigned long flags) _raw_yield_cpu(~owner); count = spin_retry; } - if (__raw_spin_is_locked(lp)) + if (arch_spin_is_locked(lp)) continue; local_irq_disable(); if (_raw_compare_and_swap(&lp->owner_cpu, 0, cpu) == 0) @@ -80,30 +80,30 @@ void _raw_spin_lock_wait_flags(arch_spinlock_t *lp, unsigned long flags) local_irq_restore(flags); } } -EXPORT_SYMBOL(_raw_spin_lock_wait_flags); +EXPORT_SYMBOL(arch_spin_lock_wait_flags); -int _raw_spin_trylock_retry(arch_spinlock_t *lp) +int arch_spin_trylock_retry(arch_spinlock_t *lp) { unsigned int cpu = ~smp_processor_id(); int count; for (count = spin_retry; count > 0; count--) { - if (__raw_spin_is_locked(lp)) + if (arch_spin_is_locked(lp)) continue; if (_raw_compare_and_swap(&lp->owner_cpu, 0, cpu) == 0) return 1; } return 0; } -EXPORT_SYMBOL(_raw_spin_trylock_retry); +EXPORT_SYMBOL(arch_spin_trylock_retry); -void _raw_spin_relax(arch_spinlock_t *lock) +void arch_spin_relax(arch_spinlock_t *lock) { unsigned int cpu = lock->owner_cpu; if (cpu != 0) _raw_yield_cpu(~cpu); } -EXPORT_SYMBOL(_raw_spin_relax); +EXPORT_SYMBOL(arch_spin_relax); void _raw_read_lock_wait(raw_rwlock_t *rw) { diff --git a/arch/sh/include/asm/spinlock.h b/arch/sh/include/asm/spinlock.h index 5a05b3fcefbe..da1c6491ed4b 100644 --- a/arch/sh/include/asm/spinlock.h +++ b/arch/sh/include/asm/spinlock.h @@ -23,10 +23,10 @@ * Your basic SMP spinlocks, allowing only a single CPU anywhere */ -#define __raw_spin_is_locked(x) ((x)->lock <= 0) -#define __raw_spin_lock_flags(lock, flags) __raw_spin_lock(lock) -#define __raw_spin_unlock_wait(x) \ - do { while (__raw_spin_is_locked(x)) cpu_relax(); } while (0) +#define arch_spin_is_locked(x) ((x)->lock <= 0) +#define arch_spin_lock_flags(lock, flags) arch_spin_lock(lock) +#define arch_spin_unlock_wait(x) \ + do { while (arch_spin_is_locked(x)) cpu_relax(); } while (0) /* * Simple spin lock operations. There are two variants, one clears IRQ's @@ -34,14 +34,14 @@ * * We make no fairness assumptions. They have a cost. */ -static inline void __raw_spin_lock(arch_spinlock_t *lock) +static inline void arch_spin_lock(arch_spinlock_t *lock) { unsigned long tmp; unsigned long oldval; __asm__ __volatile__ ( "1: \n\t" - "movli.l @%2, %0 ! __raw_spin_lock \n\t" + "movli.l @%2, %0 ! arch_spin_lock \n\t" "mov %0, %1 \n\t" "mov #0, %0 \n\t" "movco.l %0, @%2 \n\t" @@ -54,12 +54,12 @@ static inline void __raw_spin_lock(arch_spinlock_t *lock) ); } -static inline void __raw_spin_unlock(arch_spinlock_t *lock) +static inline void arch_spin_unlock(arch_spinlock_t *lock) { unsigned long tmp; __asm__ __volatile__ ( - "mov #1, %0 ! __raw_spin_unlock \n\t" + "mov #1, %0 ! arch_spin_unlock \n\t" "mov.l %0, @%1 \n\t" : "=&z" (tmp) : "r" (&lock->lock) @@ -67,13 +67,13 @@ static inline void __raw_spin_unlock(arch_spinlock_t *lock) ); } -static inline int __raw_spin_trylock(arch_spinlock_t *lock) +static inline int arch_spin_trylock(arch_spinlock_t *lock) { unsigned long tmp, oldval; __asm__ __volatile__ ( "1: \n\t" - "movli.l @%2, %0 ! __raw_spin_trylock \n\t" + "movli.l @%2, %0 ! arch_spin_trylock \n\t" "mov %0, %1 \n\t" "mov #0, %0 \n\t" "movco.l %0, @%2 \n\t" @@ -219,8 +219,8 @@ static inline int __raw_write_trylock(raw_rwlock_t *rw) #define __raw_read_lock_flags(lock, flags) __raw_read_lock(lock) #define __raw_write_lock_flags(lock, flags) __raw_write_lock(lock) -#define _raw_spin_relax(lock) cpu_relax() -#define _raw_read_relax(lock) cpu_relax() -#define _raw_write_relax(lock) cpu_relax() +#define arch_spin_relax(lock) cpu_relax() +#define arch_read_relax(lock) cpu_relax() +#define arch_write_relax(lock) cpu_relax() #endif /* __ASM_SH_SPINLOCK_H */ diff --git a/arch/sparc/include/asm/spinlock_32.h b/arch/sparc/include/asm/spinlock_32.h index b2d8a67f727e..9b0f2f53c81c 100644 --- a/arch/sparc/include/asm/spinlock_32.h +++ b/arch/sparc/include/asm/spinlock_32.h @@ -10,12 +10,12 @@ #include -#define __raw_spin_is_locked(lock) (*((volatile unsigned char *)(lock)) != 0) +#define arch_spin_is_locked(lock) (*((volatile unsigned char *)(lock)) != 0) -#define __raw_spin_unlock_wait(lock) \ - do { while (__raw_spin_is_locked(lock)) cpu_relax(); } while (0) +#define arch_spin_unlock_wait(lock) \ + do { while (arch_spin_is_locked(lock)) cpu_relax(); } while (0) -static inline void __raw_spin_lock(arch_spinlock_t *lock) +static inline void arch_spin_lock(arch_spinlock_t *lock) { __asm__ __volatile__( "\n1:\n\t" @@ -35,7 +35,7 @@ static inline void __raw_spin_lock(arch_spinlock_t *lock) : "g2", "memory", "cc"); } -static inline int __raw_spin_trylock(arch_spinlock_t *lock) +static inline int arch_spin_trylock(arch_spinlock_t *lock) { unsigned int result; __asm__ __volatile__("ldstub [%1], %0" @@ -45,7 +45,7 @@ static inline int __raw_spin_trylock(arch_spinlock_t *lock) return (result == 0); } -static inline void __raw_spin_unlock(arch_spinlock_t *lock) +static inline void arch_spin_unlock(arch_spinlock_t *lock) { __asm__ __volatile__("stb %%g0, [%0]" : : "r" (lock) : "memory"); } @@ -176,13 +176,13 @@ static inline int arch_read_trylock(raw_rwlock_t *rw) #define __raw_write_unlock(rw) do { (rw)->lock = 0; } while(0) -#define __raw_spin_lock_flags(lock, flags) __raw_spin_lock(lock) +#define arch_spin_lock_flags(lock, flags) arch_spin_lock(lock) #define __raw_read_lock_flags(rw, flags) __raw_read_lock(rw) #define __raw_write_lock_flags(rw, flags) __raw_write_lock(rw) -#define _raw_spin_relax(lock) cpu_relax() -#define _raw_read_relax(lock) cpu_relax() -#define _raw_write_relax(lock) cpu_relax() +#define arch_spin_relax(lock) cpu_relax() +#define arch_read_relax(lock) cpu_relax() +#define arch_write_relax(lock) cpu_relax() #define __raw_read_can_lock(rw) (!((rw)->lock & 0xff)) #define __raw_write_can_lock(rw) (!(rw)->lock) diff --git a/arch/sparc/include/asm/spinlock_64.h b/arch/sparc/include/asm/spinlock_64.h index 38e16c40efc4..7cf58a2fcda4 100644 --- a/arch/sparc/include/asm/spinlock_64.h +++ b/arch/sparc/include/asm/spinlock_64.h @@ -21,13 +21,13 @@ * the spinner sections must be pre-V9 branches. */ -#define __raw_spin_is_locked(lp) ((lp)->lock != 0) +#define arch_spin_is_locked(lp) ((lp)->lock != 0) -#define __raw_spin_unlock_wait(lp) \ +#define arch_spin_unlock_wait(lp) \ do { rmb(); \ } while((lp)->lock) -static inline void __raw_spin_lock(arch_spinlock_t *lock) +static inline void arch_spin_lock(arch_spinlock_t *lock) { unsigned long tmp; @@ -46,7 +46,7 @@ static inline void __raw_spin_lock(arch_spinlock_t *lock) : "memory"); } -static inline int __raw_spin_trylock(arch_spinlock_t *lock) +static inline int arch_spin_trylock(arch_spinlock_t *lock) { unsigned long result; @@ -59,7 +59,7 @@ static inline int __raw_spin_trylock(arch_spinlock_t *lock) return (result == 0UL); } -static inline void __raw_spin_unlock(arch_spinlock_t *lock) +static inline void arch_spin_unlock(arch_spinlock_t *lock) { __asm__ __volatile__( " stb %%g0, [%0]" @@ -68,7 +68,7 @@ static inline void __raw_spin_unlock(arch_spinlock_t *lock) : "memory"); } -static inline void __raw_spin_lock_flags(arch_spinlock_t *lock, unsigned long flags) +static inline void arch_spin_lock_flags(arch_spinlock_t *lock, unsigned long flags) { unsigned long tmp1, tmp2; @@ -222,9 +222,9 @@ static int inline arch_write_trylock(raw_rwlock_t *lock) #define __raw_read_can_lock(rw) (!((rw)->lock & 0x80000000UL)) #define __raw_write_can_lock(rw) (!(rw)->lock) -#define _raw_spin_relax(lock) cpu_relax() -#define _raw_read_relax(lock) cpu_relax() -#define _raw_write_relax(lock) cpu_relax() +#define arch_spin_relax(lock) cpu_relax() +#define arch_read_relax(lock) cpu_relax() +#define arch_write_relax(lock) cpu_relax() #endif /* !(__ASSEMBLY__) */ diff --git a/arch/x86/include/asm/paravirt.h b/arch/x86/include/asm/paravirt.h index 5655f75f10b7..dd59a85a918f 100644 --- a/arch/x86/include/asm/paravirt.h +++ b/arch/x86/include/asm/paravirt.h @@ -731,34 +731,34 @@ static inline void __set_fixmap(unsigned /* enum fixed_addresses */ idx, #if defined(CONFIG_SMP) && defined(CONFIG_PARAVIRT_SPINLOCKS) -static inline int __raw_spin_is_locked(struct arch_spinlock *lock) +static inline int arch_spin_is_locked(struct arch_spinlock *lock) { return PVOP_CALL1(int, pv_lock_ops.spin_is_locked, lock); } -static inline int __raw_spin_is_contended(struct arch_spinlock *lock) +static inline int arch_spin_is_contended(struct arch_spinlock *lock) { return PVOP_CALL1(int, pv_lock_ops.spin_is_contended, lock); } -#define __raw_spin_is_contended __raw_spin_is_contended +#define arch_spin_is_contended arch_spin_is_contended -static __always_inline void __raw_spin_lock(struct arch_spinlock *lock) +static __always_inline void arch_spin_lock(struct arch_spinlock *lock) { PVOP_VCALL1(pv_lock_ops.spin_lock, lock); } -static __always_inline void __raw_spin_lock_flags(struct arch_spinlock *lock, +static __always_inline void arch_spin_lock_flags(struct arch_spinlock *lock, unsigned long flags) { PVOP_VCALL2(pv_lock_ops.spin_lock_flags, lock, flags); } -static __always_inline int __raw_spin_trylock(struct arch_spinlock *lock) +static __always_inline int arch_spin_trylock(struct arch_spinlock *lock) { return PVOP_CALL1(int, pv_lock_ops.spin_trylock, lock); } -static __always_inline void __raw_spin_unlock(struct arch_spinlock *lock) +static __always_inline void arch_spin_unlock(struct arch_spinlock *lock) { PVOP_VCALL1(pv_lock_ops.spin_unlock, lock); } diff --git a/arch/x86/include/asm/spinlock.h b/arch/x86/include/asm/spinlock.h index 204b524fcf57..ab9055fd57d9 100644 --- a/arch/x86/include/asm/spinlock.h +++ b/arch/x86/include/asm/spinlock.h @@ -174,43 +174,43 @@ static inline int __ticket_spin_is_contended(arch_spinlock_t *lock) #ifndef CONFIG_PARAVIRT_SPINLOCKS -static inline int __raw_spin_is_locked(arch_spinlock_t *lock) +static inline int arch_spin_is_locked(arch_spinlock_t *lock) { return __ticket_spin_is_locked(lock); } -static inline int __raw_spin_is_contended(arch_spinlock_t *lock) +static inline int arch_spin_is_contended(arch_spinlock_t *lock) { return __ticket_spin_is_contended(lock); } -#define __raw_spin_is_contended __raw_spin_is_contended +#define arch_spin_is_contended arch_spin_is_contended -static __always_inline void __raw_spin_lock(arch_spinlock_t *lock) +static __always_inline void arch_spin_lock(arch_spinlock_t *lock) { __ticket_spin_lock(lock); } -static __always_inline int __raw_spin_trylock(arch_spinlock_t *lock) +static __always_inline int arch_spin_trylock(arch_spinlock_t *lock) { return __ticket_spin_trylock(lock); } -static __always_inline void __raw_spin_unlock(arch_spinlock_t *lock) +static __always_inline void arch_spin_unlock(arch_spinlock_t *lock) { __ticket_spin_unlock(lock); } -static __always_inline void __raw_spin_lock_flags(arch_spinlock_t *lock, +static __always_inline void arch_spin_lock_flags(arch_spinlock_t *lock, unsigned long flags) { - __raw_spin_lock(lock); + arch_spin_lock(lock); } #endif /* CONFIG_PARAVIRT_SPINLOCKS */ -static inline void __raw_spin_unlock_wait(arch_spinlock_t *lock) +static inline void arch_spin_unlock_wait(arch_spinlock_t *lock) { - while (__raw_spin_is_locked(lock)) + while (arch_spin_is_locked(lock)) cpu_relax(); } @@ -298,9 +298,9 @@ static inline void __raw_write_unlock(raw_rwlock_t *rw) #define __raw_read_lock_flags(lock, flags) __raw_read_lock(lock) #define __raw_write_lock_flags(lock, flags) __raw_write_lock(lock) -#define _raw_spin_relax(lock) cpu_relax() -#define _raw_read_relax(lock) cpu_relax() -#define _raw_write_relax(lock) cpu_relax() +#define arch_spin_relax(lock) cpu_relax() +#define arch_read_relax(lock) cpu_relax() +#define arch_write_relax(lock) cpu_relax() /* The {read|write|spin}_lock() on x86 are full memory barriers. */ static inline void smp_mb__after_lock(void) { } diff --git a/arch/x86/kernel/dumpstack.c b/arch/x86/kernel/dumpstack.c index 5b75afac8a38..0a0aa1cec8f1 100644 --- a/arch/x86/kernel/dumpstack.c +++ b/arch/x86/kernel/dumpstack.c @@ -207,11 +207,11 @@ unsigned __kprobes long oops_begin(void) /* racy, but better than risking deadlock. */ raw_local_irq_save(flags); cpu = smp_processor_id(); - if (!__raw_spin_trylock(&die_lock)) { + if (!arch_spin_trylock(&die_lock)) { if (cpu == die_owner) /* nested oops. should stop eventually */; else - __raw_spin_lock(&die_lock); + arch_spin_lock(&die_lock); } die_nest_count++; die_owner = cpu; @@ -231,7 +231,7 @@ void __kprobes oops_end(unsigned long flags, struct pt_regs *regs, int signr) die_nest_count--; if (!die_nest_count) /* Nest count reaches zero, release the lock. */ - __raw_spin_unlock(&die_lock); + arch_spin_unlock(&die_lock); raw_local_irq_restore(flags); oops_exit(); diff --git a/arch/x86/kernel/paravirt-spinlocks.c b/arch/x86/kernel/paravirt-spinlocks.c index a0f39e090684..676b8c77a976 100644 --- a/arch/x86/kernel/paravirt-spinlocks.c +++ b/arch/x86/kernel/paravirt-spinlocks.c @@ -10,7 +10,7 @@ static inline void default_spin_lock_flags(arch_spinlock_t *lock, unsigned long flags) { - __raw_spin_lock(lock); + arch_spin_lock(lock); } struct pv_lock_ops pv_lock_ops = { diff --git a/arch/x86/kernel/tsc_sync.c b/arch/x86/kernel/tsc_sync.c index f1714697a09a..0aa5fed8b9e6 100644 --- a/arch/x86/kernel/tsc_sync.c +++ b/arch/x86/kernel/tsc_sync.c @@ -62,13 +62,13 @@ static __cpuinit void check_tsc_warp(void) * previous TSC that was measured (possibly on * another CPU) and update the previous TSC timestamp. */ - __raw_spin_lock(&sync_lock); + arch_spin_lock(&sync_lock); prev = last_tsc; rdtsc_barrier(); now = get_cycles(); rdtsc_barrier(); last_tsc = now; - __raw_spin_unlock(&sync_lock); + arch_spin_unlock(&sync_lock); /* * Be nice every now and then (and also check whether @@ -87,10 +87,10 @@ static __cpuinit void check_tsc_warp(void) * we saw a time-warp of the TSC going backwards: */ if (unlikely(prev > now)) { - __raw_spin_lock(&sync_lock); + arch_spin_lock(&sync_lock); max_warp = max(max_warp, prev - now); nr_warps++; - __raw_spin_unlock(&sync_lock); + arch_spin_unlock(&sync_lock); } } WARN(!(now-start), diff --git a/include/asm-generic/bitops/atomic.h b/include/asm-generic/bitops/atomic.h index dcf0afad4a7f..ecc44a8e2b44 100644 --- a/include/asm-generic/bitops/atomic.h +++ b/include/asm-generic/bitops/atomic.h @@ -22,12 +22,12 @@ extern arch_spinlock_t __atomic_hash[ATOMIC_HASH_SIZE] __lock_aligned; #define _atomic_spin_lock_irqsave(l,f) do { \ arch_spinlock_t *s = ATOMIC_HASH(l); \ local_irq_save(f); \ - __raw_spin_lock(s); \ + arch_spin_lock(s); \ } while(0) #define _atomic_spin_unlock_irqrestore(l,f) do { \ arch_spinlock_t *s = ATOMIC_HASH(l); \ - __raw_spin_unlock(s); \ + arch_spin_unlock(s); \ local_irq_restore(f); \ } while(0) diff --git a/include/linux/spinlock.h b/include/linux/spinlock.h index 5ef7a4c060b5..de3a022489c6 100644 --- a/include/linux/spinlock.h +++ b/include/linux/spinlock.h @@ -14,7 +14,7 @@ * linux/spinlock_types.h: * defines the generic type and initializers * - * asm/spinlock.h: contains the __raw_spin_*()/etc. lowlevel + * asm/spinlock.h: contains the arch_spin_*()/etc. lowlevel * implementations, mostly inline assembly code * * (also included on UP-debug builds:) @@ -34,7 +34,7 @@ * defines the generic type and initializers * * linux/spinlock_up.h: - * contains the __raw_spin_*()/etc. version of UP + * contains the arch_spin_*()/etc. version of UP * builds. (which are NOPs on non-debug, non-preempt * builds) * @@ -103,17 +103,17 @@ do { \ do { *(lock) = __SPIN_LOCK_UNLOCKED(lock); } while (0) #endif -#define spin_is_locked(lock) __raw_spin_is_locked(&(lock)->raw_lock) +#define spin_is_locked(lock) arch_spin_is_locked(&(lock)->raw_lock) #ifdef CONFIG_GENERIC_LOCKBREAK #define spin_is_contended(lock) ((lock)->break_lock) #else -#ifdef __raw_spin_is_contended -#define spin_is_contended(lock) __raw_spin_is_contended(&(lock)->raw_lock) +#ifdef arch_spin_is_contended +#define spin_is_contended(lock) arch_spin_is_contended(&(lock)->raw_lock) #else #define spin_is_contended(lock) (((void)(lock), 0)) -#endif /*__raw_spin_is_contended*/ +#endif /*arch_spin_is_contended*/ #endif /* The lock does not imply full memory barrier. */ @@ -125,7 +125,7 @@ static inline void smp_mb__after_lock(void) { smp_mb(); } * spin_unlock_wait - wait until the spinlock gets unlocked * @lock: the spinlock in question. */ -#define spin_unlock_wait(lock) __raw_spin_unlock_wait(&(lock)->raw_lock) +#define spin_unlock_wait(lock) arch_spin_unlock_wait(&(lock)->raw_lock) #ifdef CONFIG_DEBUG_SPINLOCK extern void _raw_spin_lock(spinlock_t *lock); @@ -133,11 +133,11 @@ static inline void smp_mb__after_lock(void) { smp_mb(); } extern int _raw_spin_trylock(spinlock_t *lock); extern void _raw_spin_unlock(spinlock_t *lock); #else -# define _raw_spin_lock(lock) __raw_spin_lock(&(lock)->raw_lock) +# define _raw_spin_lock(lock) arch_spin_lock(&(lock)->raw_lock) # define _raw_spin_lock_flags(lock, flags) \ - __raw_spin_lock_flags(&(lock)->raw_lock, *(flags)) -# define _raw_spin_trylock(lock) __raw_spin_trylock(&(lock)->raw_lock) -# define _raw_spin_unlock(lock) __raw_spin_unlock(&(lock)->raw_lock) + arch_spin_lock_flags(&(lock)->raw_lock, *(flags)) +# define _raw_spin_trylock(lock) arch_spin_trylock(&(lock)->raw_lock) +# define _raw_spin_unlock(lock) arch_spin_unlock(&(lock)->raw_lock) #endif /* diff --git a/include/linux/spinlock_up.h b/include/linux/spinlock_up.h index 8ee2ac1bf636..1d3bcc3cf7c6 100644 --- a/include/linux/spinlock_up.h +++ b/include/linux/spinlock_up.h @@ -18,21 +18,21 @@ */ #ifdef CONFIG_DEBUG_SPINLOCK -#define __raw_spin_is_locked(x) ((x)->slock == 0) +#define arch_spin_is_locked(x) ((x)->slock == 0) -static inline void __raw_spin_lock(arch_spinlock_t *lock) +static inline void arch_spin_lock(arch_spinlock_t *lock) { lock->slock = 0; } static inline void -__raw_spin_lock_flags(arch_spinlock_t *lock, unsigned long flags) +arch_spin_lock_flags(arch_spinlock_t *lock, unsigned long flags) { local_irq_save(flags); lock->slock = 0; } -static inline int __raw_spin_trylock(arch_spinlock_t *lock) +static inline int arch_spin_trylock(arch_spinlock_t *lock) { char oldval = lock->slock; @@ -41,7 +41,7 @@ static inline int __raw_spin_trylock(arch_spinlock_t *lock) return oldval > 0; } -static inline void __raw_spin_unlock(arch_spinlock_t *lock) +static inline void arch_spin_unlock(arch_spinlock_t *lock) { lock->slock = 1; } @@ -57,20 +57,20 @@ static inline void __raw_spin_unlock(arch_spinlock_t *lock) #define __raw_write_unlock(lock) do { (void)(lock); } while (0) #else /* DEBUG_SPINLOCK */ -#define __raw_spin_is_locked(lock) ((void)(lock), 0) +#define arch_spin_is_locked(lock) ((void)(lock), 0) /* for sched.c and kernel_lock.c: */ -# define __raw_spin_lock(lock) do { (void)(lock); } while (0) -# define __raw_spin_lock_flags(lock, flags) do { (void)(lock); } while (0) -# define __raw_spin_unlock(lock) do { (void)(lock); } while (0) -# define __raw_spin_trylock(lock) ({ (void)(lock); 1; }) +# define arch_spin_lock(lock) do { (void)(lock); } while (0) +# define arch_spin_lock_flags(lock, flags) do { (void)(lock); } while (0) +# define arch_spin_unlock(lock) do { (void)(lock); } while (0) +# define arch_spin_trylock(lock) ({ (void)(lock); 1; }) #endif /* DEBUG_SPINLOCK */ -#define __raw_spin_is_contended(lock) (((void)(lock), 0)) +#define arch_spin_is_contended(lock) (((void)(lock), 0)) #define __raw_read_can_lock(lock) (((void)(lock), 1)) #define __raw_write_can_lock(lock) (((void)(lock), 1)) -#define __raw_spin_unlock_wait(lock) \ - do { cpu_relax(); } while (__raw_spin_is_locked(lock)) +#define arch_spin_unlock_wait(lock) \ + do { cpu_relax(); } while (arch_spin_is_locked(lock)) #endif /* __LINUX_SPINLOCK_UP_H */ diff --git a/kernel/lockdep.c b/kernel/lockdep.c index 2389e3f85cf6..5feaddcdbe49 100644 --- a/kernel/lockdep.c +++ b/kernel/lockdep.c @@ -77,7 +77,7 @@ static arch_spinlock_t lockdep_lock = (arch_spinlock_t)__ARCH_SPIN_LOCK_UNLOCKED static int graph_lock(void) { - __raw_spin_lock(&lockdep_lock); + arch_spin_lock(&lockdep_lock); /* * Make sure that if another CPU detected a bug while * walking the graph we dont change it (while the other @@ -85,7 +85,7 @@ static int graph_lock(void) * dropped already) */ if (!debug_locks) { - __raw_spin_unlock(&lockdep_lock); + arch_spin_unlock(&lockdep_lock); return 0; } /* prevent any recursions within lockdep from causing deadlocks */ @@ -95,11 +95,11 @@ static int graph_lock(void) static inline int graph_unlock(void) { - if (debug_locks && !__raw_spin_is_locked(&lockdep_lock)) + if (debug_locks && !arch_spin_is_locked(&lockdep_lock)) return DEBUG_LOCKS_WARN_ON(1); current->lockdep_recursion--; - __raw_spin_unlock(&lockdep_lock); + arch_spin_unlock(&lockdep_lock); return 0; } @@ -111,7 +111,7 @@ static inline int debug_locks_off_graph_unlock(void) { int ret = debug_locks_off(); - __raw_spin_unlock(&lockdep_lock); + arch_spin_unlock(&lockdep_lock); return ret; } @@ -1170,9 +1170,9 @@ unsigned long lockdep_count_forward_deps(struct lock_class *class) this.class = class; local_irq_save(flags); - __raw_spin_lock(&lockdep_lock); + arch_spin_lock(&lockdep_lock); ret = __lockdep_count_forward_deps(&this); - __raw_spin_unlock(&lockdep_lock); + arch_spin_unlock(&lockdep_lock); local_irq_restore(flags); return ret; @@ -1197,9 +1197,9 @@ unsigned long lockdep_count_backward_deps(struct lock_class *class) this.class = class; local_irq_save(flags); - __raw_spin_lock(&lockdep_lock); + arch_spin_lock(&lockdep_lock); ret = __lockdep_count_backward_deps(&this); - __raw_spin_unlock(&lockdep_lock); + arch_spin_unlock(&lockdep_lock); local_irq_restore(flags); return ret; diff --git a/kernel/mutex-debug.h b/kernel/mutex-debug.h index 6b2d735846a5..7bebbd15b342 100644 --- a/kernel/mutex-debug.h +++ b/kernel/mutex-debug.h @@ -43,13 +43,13 @@ static inline void mutex_clear_owner(struct mutex *lock) \ DEBUG_LOCKS_WARN_ON(in_interrupt()); \ local_irq_save(flags); \ - __raw_spin_lock(&(lock)->raw_lock); \ + arch_spin_lock(&(lock)->raw_lock); \ DEBUG_LOCKS_WARN_ON(l->magic != l); \ } while (0) #define spin_unlock_mutex(lock, flags) \ do { \ - __raw_spin_unlock(&(lock)->raw_lock); \ + arch_spin_unlock(&(lock)->raw_lock); \ local_irq_restore(flags); \ preempt_check_resched(); \ } while (0) diff --git a/kernel/spinlock.c b/kernel/spinlock.c index e6e136318437..fbb5f8b78357 100644 --- a/kernel/spinlock.c +++ b/kernel/spinlock.c @@ -53,7 +53,7 @@ void __lockfunc __##op##_lock(locktype##_t *lock) \ if (!(lock)->break_lock) \ (lock)->break_lock = 1; \ while (!op##_can_lock(lock) && (lock)->break_lock) \ - _raw_##op##_relax(&lock->raw_lock); \ + arch_##op##_relax(&lock->raw_lock); \ } \ (lock)->break_lock = 0; \ } \ @@ -73,7 +73,7 @@ unsigned long __lockfunc __##op##_lock_irqsave(locktype##_t *lock) \ if (!(lock)->break_lock) \ (lock)->break_lock = 1; \ while (!op##_can_lock(lock) && (lock)->break_lock) \ - _raw_##op##_relax(&lock->raw_lock); \ + arch_##op##_relax(&lock->raw_lock); \ } \ (lock)->break_lock = 0; \ return flags; \ diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c index fb7a0fa508b9..f58c9ad15830 100644 --- a/kernel/trace/ring_buffer.c +++ b/kernel/trace/ring_buffer.c @@ -2834,7 +2834,7 @@ rb_get_reader_page(struct ring_buffer_per_cpu *cpu_buffer) int ret; local_irq_save(flags); - __raw_spin_lock(&cpu_buffer->lock); + arch_spin_lock(&cpu_buffer->lock); again: /* @@ -2923,7 +2923,7 @@ rb_get_reader_page(struct ring_buffer_per_cpu *cpu_buffer) goto again; out: - __raw_spin_unlock(&cpu_buffer->lock); + arch_spin_unlock(&cpu_buffer->lock); local_irq_restore(flags); return reader; @@ -3286,9 +3286,9 @@ ring_buffer_read_start(struct ring_buffer *buffer, int cpu) synchronize_sched(); spin_lock_irqsave(&cpu_buffer->reader_lock, flags); - __raw_spin_lock(&cpu_buffer->lock); + arch_spin_lock(&cpu_buffer->lock); rb_iter_reset(iter); - __raw_spin_unlock(&cpu_buffer->lock); + arch_spin_unlock(&cpu_buffer->lock); spin_unlock_irqrestore(&cpu_buffer->reader_lock, flags); return iter; @@ -3408,11 +3408,11 @@ void ring_buffer_reset_cpu(struct ring_buffer *buffer, int cpu) if (RB_WARN_ON(cpu_buffer, local_read(&cpu_buffer->committing))) goto out; - __raw_spin_lock(&cpu_buffer->lock); + arch_spin_lock(&cpu_buffer->lock); rb_reset_cpu(cpu_buffer); - __raw_spin_unlock(&cpu_buffer->lock); + arch_spin_unlock(&cpu_buffer->lock); out: spin_unlock_irqrestore(&cpu_buffer->reader_lock, flags); diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 63bc1cc38219..bb6b5e7fa2a2 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -555,13 +555,13 @@ update_max_tr(struct trace_array *tr, struct task_struct *tsk, int cpu) return; WARN_ON_ONCE(!irqs_disabled()); - __raw_spin_lock(&ftrace_max_lock); + arch_spin_lock(&ftrace_max_lock); tr->buffer = max_tr.buffer; max_tr.buffer = buf; __update_max_tr(tr, tsk, cpu); - __raw_spin_unlock(&ftrace_max_lock); + arch_spin_unlock(&ftrace_max_lock); } /** @@ -581,7 +581,7 @@ update_max_tr_single(struct trace_array *tr, struct task_struct *tsk, int cpu) return; WARN_ON_ONCE(!irqs_disabled()); - __raw_spin_lock(&ftrace_max_lock); + arch_spin_lock(&ftrace_max_lock); ftrace_disable_cpu(); @@ -603,7 +603,7 @@ update_max_tr_single(struct trace_array *tr, struct task_struct *tsk, int cpu) WARN_ON_ONCE(ret && ret != -EAGAIN && ret != -EBUSY); __update_max_tr(tr, tsk, cpu); - __raw_spin_unlock(&ftrace_max_lock); + arch_spin_unlock(&ftrace_max_lock); } #endif /* CONFIG_TRACER_MAX_TRACE */ @@ -915,7 +915,7 @@ static void trace_save_cmdline(struct task_struct *tsk) * nor do we want to disable interrupts, * so if we miss here, then better luck next time. */ - if (!__raw_spin_trylock(&trace_cmdline_lock)) + if (!arch_spin_trylock(&trace_cmdline_lock)) return; idx = map_pid_to_cmdline[tsk->pid]; @@ -940,7 +940,7 @@ static void trace_save_cmdline(struct task_struct *tsk) memcpy(&saved_cmdlines[idx], tsk->comm, TASK_COMM_LEN); - __raw_spin_unlock(&trace_cmdline_lock); + arch_spin_unlock(&trace_cmdline_lock); } void trace_find_cmdline(int pid, char comm[]) @@ -958,14 +958,14 @@ void trace_find_cmdline(int pid, char comm[]) } preempt_disable(); - __raw_spin_lock(&trace_cmdline_lock); + arch_spin_lock(&trace_cmdline_lock); map = map_pid_to_cmdline[pid]; if (map != NO_CMDLINE_MAP) strcpy(comm, saved_cmdlines[map]); else strcpy(comm, "<...>"); - __raw_spin_unlock(&trace_cmdline_lock); + arch_spin_unlock(&trace_cmdline_lock); preempt_enable(); } @@ -1283,7 +1283,7 @@ int trace_vbprintk(unsigned long ip, const char *fmt, va_list args) /* Lockdep uses trace_printk for lock tracing */ local_irq_save(flags); - __raw_spin_lock(&trace_buf_lock); + arch_spin_lock(&trace_buf_lock); len = vbin_printf(trace_buf, TRACE_BUF_SIZE, fmt, args); if (len > TRACE_BUF_SIZE || len < 0) @@ -1304,7 +1304,7 @@ int trace_vbprintk(unsigned long ip, const char *fmt, va_list args) ring_buffer_unlock_commit(buffer, event); out_unlock: - __raw_spin_unlock(&trace_buf_lock); + arch_spin_unlock(&trace_buf_lock); local_irq_restore(flags); out: @@ -1360,7 +1360,7 @@ int trace_array_vprintk(struct trace_array *tr, pause_graph_tracing(); raw_local_irq_save(irq_flags); - __raw_spin_lock(&trace_buf_lock); + arch_spin_lock(&trace_buf_lock); len = vsnprintf(trace_buf, TRACE_BUF_SIZE, fmt, args); size = sizeof(*entry) + len + 1; @@ -1378,7 +1378,7 @@ int trace_array_vprintk(struct trace_array *tr, ring_buffer_unlock_commit(buffer, event); out_unlock: - __raw_spin_unlock(&trace_buf_lock); + arch_spin_unlock(&trace_buf_lock); raw_local_irq_restore(irq_flags); unpause_graph_tracing(); out: @@ -2279,7 +2279,7 @@ tracing_cpumask_write(struct file *filp, const char __user *ubuf, mutex_lock(&tracing_cpumask_update_lock); local_irq_disable(); - __raw_spin_lock(&ftrace_max_lock); + arch_spin_lock(&ftrace_max_lock); for_each_tracing_cpu(cpu) { /* * Increase/decrease the disabled counter if we are @@ -2294,7 +2294,7 @@ tracing_cpumask_write(struct file *filp, const char __user *ubuf, atomic_dec(&global_trace.data[cpu]->disabled); } } - __raw_spin_unlock(&ftrace_max_lock); + arch_spin_unlock(&ftrace_max_lock); local_irq_enable(); cpumask_copy(tracing_cpumask, tracing_cpumask_new); @@ -4318,7 +4318,7 @@ static void __ftrace_dump(bool disable_tracing) /* only one dump */ local_irq_save(flags); - __raw_spin_lock(&ftrace_dump_lock); + arch_spin_lock(&ftrace_dump_lock); if (dump_ran) goto out; @@ -4393,7 +4393,7 @@ static void __ftrace_dump(bool disable_tracing) } out: - __raw_spin_unlock(&ftrace_dump_lock); + arch_spin_unlock(&ftrace_dump_lock); local_irq_restore(flags); } diff --git a/kernel/trace/trace_clock.c b/kernel/trace/trace_clock.c index 433e2eda2d01..84a3a7ba072a 100644 --- a/kernel/trace/trace_clock.c +++ b/kernel/trace/trace_clock.c @@ -94,7 +94,7 @@ u64 notrace trace_clock_global(void) if (unlikely(in_nmi())) goto out; - __raw_spin_lock(&trace_clock_struct.lock); + arch_spin_lock(&trace_clock_struct.lock); /* * TODO: if this happens often then maybe we should reset @@ -106,7 +106,7 @@ u64 notrace trace_clock_global(void) trace_clock_struct.prev_time = now; - __raw_spin_unlock(&trace_clock_struct.lock); + arch_spin_unlock(&trace_clock_struct.lock); out: raw_local_irq_restore(flags); diff --git a/kernel/trace/trace_sched_wakeup.c b/kernel/trace/trace_sched_wakeup.c index e347853564e9..0271742abb8d 100644 --- a/kernel/trace/trace_sched_wakeup.c +++ b/kernel/trace/trace_sched_wakeup.c @@ -143,7 +143,7 @@ probe_wakeup_sched_switch(struct rq *rq, struct task_struct *prev, goto out; local_irq_save(flags); - __raw_spin_lock(&wakeup_lock); + arch_spin_lock(&wakeup_lock); /* We could race with grabbing wakeup_lock */ if (unlikely(!tracer_enabled || next != wakeup_task)) @@ -169,7 +169,7 @@ probe_wakeup_sched_switch(struct rq *rq, struct task_struct *prev, out_unlock: __wakeup_reset(wakeup_trace); - __raw_spin_unlock(&wakeup_lock); + arch_spin_unlock(&wakeup_lock); local_irq_restore(flags); out: atomic_dec(&wakeup_trace->data[cpu]->disabled); @@ -193,9 +193,9 @@ static void wakeup_reset(struct trace_array *tr) tracing_reset_online_cpus(tr); local_irq_save(flags); - __raw_spin_lock(&wakeup_lock); + arch_spin_lock(&wakeup_lock); __wakeup_reset(tr); - __raw_spin_unlock(&wakeup_lock); + arch_spin_unlock(&wakeup_lock); local_irq_restore(flags); } @@ -225,7 +225,7 @@ probe_wakeup(struct rq *rq, struct task_struct *p, int success) goto out; /* interrupts should be off from try_to_wake_up */ - __raw_spin_lock(&wakeup_lock); + arch_spin_lock(&wakeup_lock); /* check for races. */ if (!tracer_enabled || p->prio >= wakeup_prio) @@ -255,7 +255,7 @@ probe_wakeup(struct rq *rq, struct task_struct *p, int success) trace_function(wakeup_trace, CALLER_ADDR1, CALLER_ADDR2, flags, pc); out_locked: - __raw_spin_unlock(&wakeup_lock); + arch_spin_unlock(&wakeup_lock); out: atomic_dec(&wakeup_trace->data[cpu]->disabled); } diff --git a/kernel/trace/trace_selftest.c b/kernel/trace/trace_selftest.c index dc98309e839a..280fea470d67 100644 --- a/kernel/trace/trace_selftest.c +++ b/kernel/trace/trace_selftest.c @@ -67,7 +67,7 @@ static int trace_test_buffer(struct trace_array *tr, unsigned long *count) /* Don't allow flipping of max traces now */ local_irq_save(flags); - __raw_spin_lock(&ftrace_max_lock); + arch_spin_lock(&ftrace_max_lock); cnt = ring_buffer_entries(tr->buffer); @@ -85,7 +85,7 @@ static int trace_test_buffer(struct trace_array *tr, unsigned long *count) break; } tracing_on(); - __raw_spin_unlock(&ftrace_max_lock); + arch_spin_unlock(&ftrace_max_lock); local_irq_restore(flags); if (count) diff --git a/kernel/trace/trace_stack.c b/kernel/trace/trace_stack.c index 728c35221483..678a5120ee30 100644 --- a/kernel/trace/trace_stack.c +++ b/kernel/trace/trace_stack.c @@ -54,7 +54,7 @@ static inline void check_stack(void) return; local_irq_save(flags); - __raw_spin_lock(&max_stack_lock); + arch_spin_lock(&max_stack_lock); /* a race could have already updated it */ if (this_size <= max_stack_size) @@ -103,7 +103,7 @@ static inline void check_stack(void) } out: - __raw_spin_unlock(&max_stack_lock); + arch_spin_unlock(&max_stack_lock); local_irq_restore(flags); } @@ -171,9 +171,9 @@ stack_max_size_write(struct file *filp, const char __user *ubuf, return ret; local_irq_save(flags); - __raw_spin_lock(&max_stack_lock); + arch_spin_lock(&max_stack_lock); *ptr = val; - __raw_spin_unlock(&max_stack_lock); + arch_spin_unlock(&max_stack_lock); local_irq_restore(flags); return count; @@ -207,7 +207,7 @@ t_next(struct seq_file *m, void *v, loff_t *pos) static void *t_start(struct seq_file *m, loff_t *pos) { local_irq_disable(); - __raw_spin_lock(&max_stack_lock); + arch_spin_lock(&max_stack_lock); if (*pos == 0) return SEQ_START_TOKEN; @@ -217,7 +217,7 @@ static void *t_start(struct seq_file *m, loff_t *pos) static void t_stop(struct seq_file *m, void *p) { - __raw_spin_unlock(&max_stack_lock); + arch_spin_unlock(&max_stack_lock); local_irq_enable(); } diff --git a/lib/spinlock_debug.c b/lib/spinlock_debug.c index f73004137141..1304fe094546 100644 --- a/lib/spinlock_debug.c +++ b/lib/spinlock_debug.c @@ -106,7 +106,7 @@ static void __spin_lock_debug(spinlock_t *lock) for (;;) { for (i = 0; i < loops; i++) { - if (__raw_spin_trylock(&lock->raw_lock)) + if (arch_spin_trylock(&lock->raw_lock)) return; __delay(1); } @@ -128,14 +128,14 @@ static void __spin_lock_debug(spinlock_t *lock) void _raw_spin_lock(spinlock_t *lock) { debug_spin_lock_before(lock); - if (unlikely(!__raw_spin_trylock(&lock->raw_lock))) + if (unlikely(!arch_spin_trylock(&lock->raw_lock))) __spin_lock_debug(lock); debug_spin_lock_after(lock); } int _raw_spin_trylock(spinlock_t *lock) { - int ret = __raw_spin_trylock(&lock->raw_lock); + int ret = arch_spin_trylock(&lock->raw_lock); if (ret) debug_spin_lock_after(lock); @@ -151,7 +151,7 @@ int _raw_spin_trylock(spinlock_t *lock) void _raw_spin_unlock(spinlock_t *lock) { debug_spin_unlock(lock); - __raw_spin_unlock(&lock->raw_lock); + arch_spin_unlock(&lock->raw_lock); } static void rwlock_bug(rwlock_t *lock, const char *msg) -- cgit v1.2.3-71-gd317 From c2f21ce2e31286a0a32f8da0a7856e9ca1122ef3 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Wed, 2 Dec 2009 20:02:59 +0100 Subject: locking: Implement new raw_spinlock Now that the raw_spin name space is freed up, we can implement raw_spinlock and the related functions which are used to annotate the locks which are not converted to sleeping spinlocks in preempt-rt. A side effect is that only such locks can be used with the low level lock fsunctions which circumvent lockdep. For !rt spin_* functions are mapped to the raw_spin* implementations. Signed-off-by: Thomas Gleixner Acked-by: Peter Zijlstra Acked-by: Ingo Molnar --- include/linux/spinlock.h | 258 +++++++++++++++++++++++++++++---------- include/linux/spinlock_api_smp.h | 51 ++++---- include/linux/spinlock_api_up.h | 2 +- include/linux/spinlock_types.h | 49 ++++++-- kernel/mutex-debug.h | 12 +- kernel/sched.c | 2 +- kernel/spinlock.c | 34 +++--- lib/spinlock_debug.c | 24 ++-- 8 files changed, 297 insertions(+), 135 deletions(-) (limited to 'kernel') diff --git a/include/linux/spinlock.h b/include/linux/spinlock.h index 53bc2213b414..ef5a55d96b9b 100644 --- a/include/linux/spinlock.h +++ b/include/linux/spinlock.h @@ -80,7 +80,7 @@ #include /* - * Pull the __raw*() functions/declarations (UP-nondebug doesnt need them): + * Pull the arch_spin*() functions/declarations (UP-nondebug doesnt need them): */ #ifdef CONFIG_SMP # include @@ -89,30 +89,30 @@ #endif #ifdef CONFIG_DEBUG_SPINLOCK - extern void __spin_lock_init(spinlock_t *lock, const char *name, - struct lock_class_key *key); -# define spin_lock_init(lock) \ + extern void __raw_spin_lock_init(raw_spinlock_t *lock, const char *name, + struct lock_class_key *key); +# define raw_spin_lock_init(lock) \ do { \ static struct lock_class_key __key; \ \ - __spin_lock_init((lock), #lock, &__key); \ + __raw_spin_lock_init((lock), #lock, &__key); \ } while (0) #else -# define spin_lock_init(lock) \ - do { *(lock) = __SPIN_LOCK_UNLOCKED(lock); } while (0) +# define raw_spin_lock_init(lock) \ + do { *(lock) = __RAW_SPIN_LOCK_UNLOCKED(lock); } while (0) #endif -#define spin_is_locked(lock) arch_spin_is_locked(&(lock)->raw_lock) +#define raw_spin_is_locked(lock) arch_spin_is_locked(&(lock)->raw_lock) #ifdef CONFIG_GENERIC_LOCKBREAK -#define spin_is_contended(lock) ((lock)->break_lock) +#define raw_spin_is_contended(lock) ((lock)->break_lock) #else #ifdef arch_spin_is_contended -#define spin_is_contended(lock) arch_spin_is_contended(&(lock)->raw_lock) +#define raw_spin_is_contended(lock) arch_spin_is_contended(&(lock)->raw_lock) #else -#define spin_is_contended(lock) (((void)(lock), 0)) +#define raw_spin_is_contended(lock) (((void)(lock), 0)) #endif /*arch_spin_is_contended*/ #endif @@ -122,22 +122,37 @@ static inline void smp_mb__after_lock(void) { smp_mb(); } #endif /** - * spin_unlock_wait - wait until the spinlock gets unlocked + * raw_spin_unlock_wait - wait until the spinlock gets unlocked * @lock: the spinlock in question. */ -#define spin_unlock_wait(lock) arch_spin_unlock_wait(&(lock)->raw_lock) +#define raw_spin_unlock_wait(lock) arch_spin_unlock_wait(&(lock)->raw_lock) #ifdef CONFIG_DEBUG_SPINLOCK - extern void _raw_spin_lock(spinlock_t *lock); + extern void _raw_spin_lock(raw_spinlock_t *lock); #define _raw_spin_lock_flags(lock, flags) _raw_spin_lock(lock) - extern int _raw_spin_trylock(spinlock_t *lock); - extern void _raw_spin_unlock(spinlock_t *lock); + extern int _raw_spin_trylock(raw_spinlock_t *lock); + extern void _raw_spin_unlock(raw_spinlock_t *lock); #else -# define _raw_spin_lock(lock) arch_spin_lock(&(lock)->raw_lock) -# define _raw_spin_lock_flags(lock, flags) \ - arch_spin_lock_flags(&(lock)->raw_lock, *(flags)) -# define _raw_spin_trylock(lock) arch_spin_trylock(&(lock)->raw_lock) -# define _raw_spin_unlock(lock) arch_spin_unlock(&(lock)->raw_lock) +static inline void _raw_spin_lock(raw_spinlock_t *lock) +{ + arch_spin_lock(&lock->raw_lock); +} + +static inline void +_raw_spin_lock_flags(raw_spinlock_t *lock, unsigned long *flags) +{ + arch_spin_lock_flags(&lock->raw_lock, *flags); +} + +static inline int _raw_spin_trylock(raw_spinlock_t *lock) +{ + return arch_spin_trylock(&(lock)->raw_lock); +} + +static inline void _raw_spin_unlock(raw_spinlock_t *lock) +{ + arch_spin_unlock(&lock->raw_lock); +} #endif /* @@ -146,38 +161,38 @@ static inline void smp_mb__after_lock(void) { smp_mb(); } * various methods are defined as nops in the case they are not * required. */ -#define spin_trylock(lock) __cond_lock(lock, _spin_trylock(lock)) +#define raw_spin_trylock(lock) __cond_lock(lock, _spin_trylock(lock)) -#define spin_lock(lock) _spin_lock(lock) +#define raw_spin_lock(lock) _spin_lock(lock) #ifdef CONFIG_DEBUG_LOCK_ALLOC -# define spin_lock_nested(lock, subclass) _spin_lock_nested(lock, subclass) -# define spin_lock_nest_lock(lock, nest_lock) \ +# define raw_spin_lock_nested(lock, subclass) _spin_lock_nested(lock, subclass) +# define raw_spin_lock_nest_lock(lock, nest_lock) \ do { \ typecheck(struct lockdep_map *, &(nest_lock)->dep_map);\ _spin_lock_nest_lock(lock, &(nest_lock)->dep_map); \ } while (0) #else -# define spin_lock_nested(lock, subclass) _spin_lock(lock) -# define spin_lock_nest_lock(lock, nest_lock) _spin_lock(lock) +# define raw_spin_lock_nested(lock, subclass) _spin_lock(lock) +# define raw_spin_lock_nest_lock(lock, nest_lock) _spin_lock(lock) #endif #if defined(CONFIG_SMP) || defined(CONFIG_DEBUG_SPINLOCK) -#define spin_lock_irqsave(lock, flags) \ +#define raw_spin_lock_irqsave(lock, flags) \ do { \ typecheck(unsigned long, flags); \ flags = _spin_lock_irqsave(lock); \ } while (0) #ifdef CONFIG_DEBUG_LOCK_ALLOC -#define spin_lock_irqsave_nested(lock, flags, subclass) \ +#define raw_spin_lock_irqsave_nested(lock, flags, subclass) \ do { \ typecheck(unsigned long, flags); \ flags = _spin_lock_irqsave_nested(lock, subclass); \ } while (0) #else -#define spin_lock_irqsave_nested(lock, flags, subclass) \ +#define raw_spin_lock_irqsave_nested(lock, flags, subclass) \ do { \ typecheck(unsigned long, flags); \ flags = _spin_lock_irqsave(lock); \ @@ -186,45 +201,178 @@ static inline void smp_mb__after_lock(void) { smp_mb(); } #else -#define spin_lock_irqsave(lock, flags) \ +#define raw_spin_lock_irqsave(lock, flags) \ do { \ typecheck(unsigned long, flags); \ _spin_lock_irqsave(lock, flags); \ } while (0) -#define spin_lock_irqsave_nested(lock, flags, subclass) \ - spin_lock_irqsave(lock, flags) +#define raw_spin_lock_irqsave_nested(lock, flags, subclass) \ + raw_spin_lock_irqsave(lock, flags) #endif -#define spin_lock_irq(lock) _spin_lock_irq(lock) -#define spin_lock_bh(lock) _spin_lock_bh(lock) -#define spin_unlock(lock) _spin_unlock(lock) -#define spin_unlock_irq(lock) _spin_unlock_irq(lock) +#define raw_spin_lock_irq(lock) _spin_lock_irq(lock) +#define raw_spin_lock_bh(lock) _spin_lock_bh(lock) +#define raw_spin_unlock(lock) _spin_unlock(lock) +#define raw_spin_unlock_irq(lock) _spin_unlock_irq(lock) -#define spin_unlock_irqrestore(lock, flags) \ - do { \ - typecheck(unsigned long, flags); \ +#define raw_spin_unlock_irqrestore(lock, flags) \ + do { \ + typecheck(unsigned long, flags); \ _spin_unlock_irqrestore(lock, flags); \ } while (0) -#define spin_unlock_bh(lock) _spin_unlock_bh(lock) +#define raw_spin_unlock_bh(lock) _spin_unlock_bh(lock) -#define spin_trylock_bh(lock) __cond_lock(lock, _spin_trylock_bh(lock)) +#define raw_spin_trylock_bh(lock) __cond_lock(lock, _spin_trylock_bh(lock)) -#define spin_trylock_irq(lock) \ +#define raw_spin_trylock_irq(lock) \ ({ \ local_irq_disable(); \ - spin_trylock(lock) ? \ + raw_spin_trylock(lock) ? \ 1 : ({ local_irq_enable(); 0; }); \ }) -#define spin_trylock_irqsave(lock, flags) \ +#define raw_spin_trylock_irqsave(lock, flags) \ ({ \ local_irq_save(flags); \ - spin_trylock(lock) ? \ + raw_spin_trylock(lock) ? \ 1 : ({ local_irq_restore(flags); 0; }); \ }) +/** + * raw_spin_can_lock - would raw_spin_trylock() succeed? + * @lock: the spinlock in question. + */ +#define raw_spin_can_lock(lock) (!raw_spin_is_locked(lock)) + +/* Include rwlock functions */ +#include + +/* + * Pull the _spin_*()/_read_*()/_write_*() functions/declarations: + */ +#if defined(CONFIG_SMP) || defined(CONFIG_DEBUG_SPINLOCK) +# include +#else +# include +#endif + +/* + * Map the spin_lock functions to the raw variants for PREEMPT_RT=n + */ + +static inline raw_spinlock_t *spinlock_check(spinlock_t *lock) +{ + return &lock->rlock; +} + +#define spin_lock_init(_lock) \ +do { \ + spinlock_check(_lock); \ + raw_spin_lock_init(&(_lock)->rlock); \ +} while (0) + +static inline void spin_lock(spinlock_t *lock) +{ + raw_spin_lock(&lock->rlock); +} + +static inline void spin_lock_bh(spinlock_t *lock) +{ + raw_spin_lock_bh(&lock->rlock); +} + +static inline int spin_trylock(spinlock_t *lock) +{ + return raw_spin_trylock(&lock->rlock); +} + +#define spin_lock_nested(lock, subclass) \ +do { \ + raw_spin_lock_nested(spinlock_check(lock), subclass); \ +} while (0) + +#define spin_lock_nest_lock(lock, nest_lock) \ +do { \ + raw_spin_lock_nest_lock(spinlock_check(lock), nest_lock); \ +} while (0) + +static inline void spin_lock_irq(spinlock_t *lock) +{ + raw_spin_lock_irq(&lock->rlock); +} + +#define spin_lock_irqsave(lock, flags) \ +do { \ + raw_spin_lock_irqsave(spinlock_check(lock), flags); \ +} while (0) + +#define spin_lock_irqsave_nested(lock, flags, subclass) \ +do { \ + raw_spin_lock_irqsave_nested(spinlock_check(lock), flags, subclass); \ +} while (0) + +static inline void spin_unlock(spinlock_t *lock) +{ + raw_spin_unlock(&lock->rlock); +} + +static inline void spin_unlock_bh(spinlock_t *lock) +{ + raw_spin_unlock_bh(&lock->rlock); +} + +static inline void spin_unlock_irq(spinlock_t *lock) +{ + raw_spin_unlock_irq(&lock->rlock); +} + +static inline void spin_unlock_irqrestore(spinlock_t *lock, unsigned long flags) +{ + raw_spin_unlock_irqrestore(&lock->rlock, flags); +} + +static inline int spin_trylock_bh(spinlock_t *lock) +{ + return raw_spin_trylock_bh(&lock->rlock); +} + +static inline int spin_trylock_irq(spinlock_t *lock) +{ + return raw_spin_trylock_irq(&lock->rlock); +} + +#define spin_trylock_irqsave(lock, flags) \ +({ \ + raw_spin_trylock_irqsave(spinlock_check(lock), flags); \ +}) + +static inline void spin_unlock_wait(spinlock_t *lock) +{ + raw_spin_unlock_wait(&lock->rlock); +} + +static inline int spin_is_locked(spinlock_t *lock) +{ + return raw_spin_is_locked(&lock->rlock); +} + +static inline int spin_is_contended(spinlock_t *lock) +{ + return raw_spin_is_contended(&lock->rlock); +} + +static inline int spin_can_lock(spinlock_t *lock) +{ + return raw_spin_can_lock(&lock->rlock); +} + +static inline void assert_spin_locked(spinlock_t *lock) +{ + assert_raw_spin_locked(&lock->rlock); +} + /* * Pull the atomic_t declaration: * (asm-mips/atomic.h needs above definitions) @@ -242,22 +390,4 @@ extern int _atomic_dec_and_lock(atomic_t *atomic, spinlock_t *lock); #define atomic_dec_and_lock(atomic, lock) \ __cond_lock(lock, _atomic_dec_and_lock(atomic, lock)) -/** - * spin_can_lock - would spin_trylock() succeed? - * @lock: the spinlock in question. - */ -#define spin_can_lock(lock) (!spin_is_locked(lock)) - -/* Include rwlock functions */ -#include - -/* - * Pull the _spin_*()/_read_*()/_write_*() functions/declarations: - */ -#if defined(CONFIG_SMP) || defined(CONFIG_DEBUG_SPINLOCK) -# include -#else -# include -#endif - #endif /* __LINUX_SPINLOCK_H */ diff --git a/include/linux/spinlock_api_smp.h b/include/linux/spinlock_api_smp.h index a2b2c9df91de..eabe5068d138 100644 --- a/include/linux/spinlock_api_smp.h +++ b/include/linux/spinlock_api_smp.h @@ -17,26 +17,29 @@ int in_lock_functions(unsigned long addr); -#define assert_spin_locked(x) BUG_ON(!spin_is_locked(x)) +#define assert_raw_spin_locked(x) BUG_ON(!raw_spin_is_locked(x)) -void __lockfunc _spin_lock(spinlock_t *lock) __acquires(lock); -void __lockfunc _spin_lock_nested(spinlock_t *lock, int subclass) +void __lockfunc _spin_lock(raw_spinlock_t *lock) __acquires(lock); +void __lockfunc _spin_lock_nested(raw_spinlock_t *lock, int subclass) __acquires(lock); -void __lockfunc _spin_lock_nest_lock(spinlock_t *lock, struct lockdep_map *map) +void __lockfunc +_spin_lock_nest_lock(raw_spinlock_t *lock, struct lockdep_map *map) __acquires(lock); -void __lockfunc _spin_lock_bh(spinlock_t *lock) __acquires(lock); -void __lockfunc _spin_lock_irq(spinlock_t *lock) __acquires(lock); +void __lockfunc _spin_lock_bh(raw_spinlock_t *lock) __acquires(lock); +void __lockfunc _spin_lock_irq(raw_spinlock_t *lock) __acquires(lock); -unsigned long __lockfunc _spin_lock_irqsave(spinlock_t *lock) +unsigned long __lockfunc _spin_lock_irqsave(raw_spinlock_t *lock) __acquires(lock); -unsigned long __lockfunc _spin_lock_irqsave_nested(spinlock_t *lock, int subclass) +unsigned long __lockfunc +_spin_lock_irqsave_nested(raw_spinlock_t *lock, int subclass) __acquires(lock); -int __lockfunc _spin_trylock(spinlock_t *lock); -int __lockfunc _spin_trylock_bh(spinlock_t *lock); -void __lockfunc _spin_unlock(spinlock_t *lock) __releases(lock); -void __lockfunc _spin_unlock_bh(spinlock_t *lock) __releases(lock); -void __lockfunc _spin_unlock_irq(spinlock_t *lock) __releases(lock); -void __lockfunc _spin_unlock_irqrestore(spinlock_t *lock, unsigned long flags) +int __lockfunc _spin_trylock(raw_spinlock_t *lock); +int __lockfunc _spin_trylock_bh(raw_spinlock_t *lock); +void __lockfunc _spin_unlock(raw_spinlock_t *lock) __releases(lock); +void __lockfunc _spin_unlock_bh(raw_spinlock_t *lock) __releases(lock); +void __lockfunc _spin_unlock_irq(raw_spinlock_t *lock) __releases(lock); +void __lockfunc +_spin_unlock_irqrestore(raw_spinlock_t *lock, unsigned long flags) __releases(lock); #ifdef CONFIG_INLINE_SPIN_LOCK @@ -79,7 +82,7 @@ void __lockfunc _spin_unlock_irqrestore(spinlock_t *lock, unsigned long flags) #define _spin_unlock_irqrestore(lock, flags) __spin_unlock_irqrestore(lock, flags) #endif -static inline int __spin_trylock(spinlock_t *lock) +static inline int __spin_trylock(raw_spinlock_t *lock) { preempt_disable(); if (_raw_spin_trylock(lock)) { @@ -97,7 +100,7 @@ static inline int __spin_trylock(spinlock_t *lock) */ #if !defined(CONFIG_GENERIC_LOCKBREAK) || defined(CONFIG_DEBUG_LOCK_ALLOC) -static inline unsigned long __spin_lock_irqsave(spinlock_t *lock) +static inline unsigned long __spin_lock_irqsave(raw_spinlock_t *lock) { unsigned long flags; @@ -117,7 +120,7 @@ static inline unsigned long __spin_lock_irqsave(spinlock_t *lock) return flags; } -static inline void __spin_lock_irq(spinlock_t *lock) +static inline void __spin_lock_irq(raw_spinlock_t *lock) { local_irq_disable(); preempt_disable(); @@ -125,7 +128,7 @@ static inline void __spin_lock_irq(spinlock_t *lock) LOCK_CONTENDED(lock, _raw_spin_trylock, _raw_spin_lock); } -static inline void __spin_lock_bh(spinlock_t *lock) +static inline void __spin_lock_bh(raw_spinlock_t *lock) { local_bh_disable(); preempt_disable(); @@ -133,7 +136,7 @@ static inline void __spin_lock_bh(spinlock_t *lock) LOCK_CONTENDED(lock, _raw_spin_trylock, _raw_spin_lock); } -static inline void __spin_lock(spinlock_t *lock) +static inline void __spin_lock(raw_spinlock_t *lock) { preempt_disable(); spin_acquire(&lock->dep_map, 0, 0, _RET_IP_); @@ -142,14 +145,14 @@ static inline void __spin_lock(spinlock_t *lock) #endif /* CONFIG_PREEMPT */ -static inline void __spin_unlock(spinlock_t *lock) +static inline void __spin_unlock(raw_spinlock_t *lock) { spin_release(&lock->dep_map, 1, _RET_IP_); _raw_spin_unlock(lock); preempt_enable(); } -static inline void __spin_unlock_irqrestore(spinlock_t *lock, +static inline void __spin_unlock_irqrestore(raw_spinlock_t *lock, unsigned long flags) { spin_release(&lock->dep_map, 1, _RET_IP_); @@ -158,7 +161,7 @@ static inline void __spin_unlock_irqrestore(spinlock_t *lock, preempt_enable(); } -static inline void __spin_unlock_irq(spinlock_t *lock) +static inline void __spin_unlock_irq(raw_spinlock_t *lock) { spin_release(&lock->dep_map, 1, _RET_IP_); _raw_spin_unlock(lock); @@ -166,7 +169,7 @@ static inline void __spin_unlock_irq(spinlock_t *lock) preempt_enable(); } -static inline void __spin_unlock_bh(spinlock_t *lock) +static inline void __spin_unlock_bh(raw_spinlock_t *lock) { spin_release(&lock->dep_map, 1, _RET_IP_); _raw_spin_unlock(lock); @@ -174,7 +177,7 @@ static inline void __spin_unlock_bh(spinlock_t *lock) local_bh_enable_ip((unsigned long)__builtin_return_address(0)); } -static inline int __spin_trylock_bh(spinlock_t *lock) +static inline int __spin_trylock_bh(raw_spinlock_t *lock) { local_bh_disable(); preempt_disable(); diff --git a/include/linux/spinlock_api_up.h b/include/linux/spinlock_api_up.h index 04e1d3164576..3a9e27adecf9 100644 --- a/include/linux/spinlock_api_up.h +++ b/include/linux/spinlock_api_up.h @@ -16,7 +16,7 @@ #define in_lock_functions(ADDR) 0 -#define assert_spin_locked(lock) do { (void)(lock); } while (0) +#define assert_raw_spin_locked(lock) do { (void)(lock); } while (0) /* * In the UP-nondebug case there's no real locking going on, so the diff --git a/include/linux/spinlock_types.h b/include/linux/spinlock_types.h index 7dadce303ebf..851b7783720d 100644 --- a/include/linux/spinlock_types.h +++ b/include/linux/spinlock_types.h @@ -17,7 +17,7 @@ #include -typedef struct { +typedef struct raw_spinlock { arch_spinlock_t raw_lock; #ifdef CONFIG_GENERIC_LOCKBREAK unsigned int break_lock; @@ -29,7 +29,7 @@ typedef struct { #ifdef CONFIG_DEBUG_LOCK_ALLOC struct lockdep_map dep_map; #endif -} spinlock_t; +} raw_spinlock_t; #define SPINLOCK_MAGIC 0xdead4ead @@ -42,18 +42,45 @@ typedef struct { #endif #ifdef CONFIG_DEBUG_SPINLOCK -# define __SPIN_LOCK_UNLOCKED(lockname) \ - (spinlock_t) { .raw_lock = __ARCH_SPIN_LOCK_UNLOCKED, \ - .magic = SPINLOCK_MAGIC, \ - .owner = SPINLOCK_OWNER_INIT, \ - .owner_cpu = -1, \ - SPIN_DEP_MAP_INIT(lockname) } +# define SPIN_DEBUG_INIT(lockname) \ + .magic = SPINLOCK_MAGIC, \ + .owner_cpu = -1, \ + .owner = SPINLOCK_OWNER_INIT, #else -# define __SPIN_LOCK_UNLOCKED(lockname) \ - (spinlock_t) { .raw_lock = __ARCH_SPIN_LOCK_UNLOCKED, \ - SPIN_DEP_MAP_INIT(lockname) } +# define SPIN_DEBUG_INIT(lockname) #endif +#define __RAW_SPIN_LOCK_INITIALIZER(lockname) \ + { \ + .raw_lock = __ARCH_SPIN_LOCK_UNLOCKED, \ + SPIN_DEBUG_INIT(lockname) \ + SPIN_DEP_MAP_INIT(lockname) } + +#define __RAW_SPIN_LOCK_UNLOCKED(lockname) \ + (raw_spinlock_t) __RAW_SPIN_LOCK_INITIALIZER(lockname) + +#define DEFINE_RAW_SPINLOCK(x) raw_spinlock_t x = __RAW_SPIN_LOCK_UNLOCKED(x) + +typedef struct spinlock { + union { + struct raw_spinlock rlock; + +#ifdef CONFIG_DEBUG_LOCK_ALLOC +# define LOCK_PADSIZE (offsetof(struct raw_spinlock, dep_map)) + struct { + u8 __padding[LOCK_PADSIZE]; + struct lockdep_map dep_map; + }; +#endif + }; +} spinlock_t; + +#define __SPIN_LOCK_INITIALIZER(lockname) \ + { { .rlock = __RAW_SPIN_LOCK_INITIALIZER(lockname) } } + +#define __SPIN_LOCK_UNLOCKED(lockname) \ + (spinlock_t ) __SPIN_LOCK_INITIALIZER(lockname) + /* * SPIN_LOCK_UNLOCKED defeats lockdep state tracking and is hence * deprecated. diff --git a/kernel/mutex-debug.h b/kernel/mutex-debug.h index 7bebbd15b342..57d527a16f9d 100644 --- a/kernel/mutex-debug.h +++ b/kernel/mutex-debug.h @@ -43,13 +43,13 @@ static inline void mutex_clear_owner(struct mutex *lock) \ DEBUG_LOCKS_WARN_ON(in_interrupt()); \ local_irq_save(flags); \ - arch_spin_lock(&(lock)->raw_lock); \ + arch_spin_lock(&(lock)->rlock.raw_lock);\ DEBUG_LOCKS_WARN_ON(l->magic != l); \ } while (0) -#define spin_unlock_mutex(lock, flags) \ - do { \ - arch_spin_unlock(&(lock)->raw_lock); \ - local_irq_restore(flags); \ - preempt_check_resched(); \ +#define spin_unlock_mutex(lock, flags) \ + do { \ + arch_spin_unlock(&(lock)->rlock.raw_lock); \ + local_irq_restore(flags); \ + preempt_check_resched(); \ } while (0) diff --git a/kernel/sched.c b/kernel/sched.c index fd05861b2111..e6acf2d7b753 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -884,7 +884,7 @@ static inline void finish_lock_switch(struct rq *rq, struct task_struct *prev) { #ifdef CONFIG_DEBUG_SPINLOCK /* this is a valid case when another task releases the spinlock */ - rq->lock.owner = current; + rq->lock.rlock.owner = current; #endif /* * If we are tracking spinlock dependencies then we have to diff --git a/kernel/spinlock.c b/kernel/spinlock.c index fbb5f8b78357..54eb7dd3c608 100644 --- a/kernel/spinlock.c +++ b/kernel/spinlock.c @@ -32,6 +32,8 @@ * include/linux/spinlock_api_smp.h */ #else +#define raw_read_can_lock(l) read_can_lock(l) +#define raw_write_can_lock(l) write_can_lock(l) /* * We build the __lock_function inlines here. They are too large for * inlining all over the place, but here is only one user per function @@ -52,7 +54,7 @@ void __lockfunc __##op##_lock(locktype##_t *lock) \ \ if (!(lock)->break_lock) \ (lock)->break_lock = 1; \ - while (!op##_can_lock(lock) && (lock)->break_lock) \ + while (!raw_##op##_can_lock(lock) && (lock)->break_lock)\ arch_##op##_relax(&lock->raw_lock); \ } \ (lock)->break_lock = 0; \ @@ -72,7 +74,7 @@ unsigned long __lockfunc __##op##_lock_irqsave(locktype##_t *lock) \ \ if (!(lock)->break_lock) \ (lock)->break_lock = 1; \ - while (!op##_can_lock(lock) && (lock)->break_lock) \ + while (!raw_##op##_can_lock(lock) && (lock)->break_lock)\ arch_##op##_relax(&lock->raw_lock); \ } \ (lock)->break_lock = 0; \ @@ -107,14 +109,14 @@ void __lockfunc __##op##_lock_bh(locktype##_t *lock) \ * __[spin|read|write]_lock_irqsave() * __[spin|read|write]_lock_bh() */ -BUILD_LOCK_OPS(spin, spinlock); +BUILD_LOCK_OPS(spin, raw_spinlock); BUILD_LOCK_OPS(read, rwlock); BUILD_LOCK_OPS(write, rwlock); #endif #ifndef CONFIG_INLINE_SPIN_TRYLOCK -int __lockfunc _spin_trylock(spinlock_t *lock) +int __lockfunc _spin_trylock(raw_spinlock_t *lock) { return __spin_trylock(lock); } @@ -122,7 +124,7 @@ EXPORT_SYMBOL(_spin_trylock); #endif #ifndef CONFIG_INLINE_SPIN_TRYLOCK_BH -int __lockfunc _spin_trylock_bh(spinlock_t *lock) +int __lockfunc _spin_trylock_bh(raw_spinlock_t *lock) { return __spin_trylock_bh(lock); } @@ -130,7 +132,7 @@ EXPORT_SYMBOL(_spin_trylock_bh); #endif #ifndef CONFIG_INLINE_SPIN_LOCK -void __lockfunc _spin_lock(spinlock_t *lock) +void __lockfunc _spin_lock(raw_spinlock_t *lock) { __spin_lock(lock); } @@ -138,7 +140,7 @@ EXPORT_SYMBOL(_spin_lock); #endif #ifndef CONFIG_INLINE_SPIN_LOCK_IRQSAVE -unsigned long __lockfunc _spin_lock_irqsave(spinlock_t *lock) +unsigned long __lockfunc _spin_lock_irqsave(raw_spinlock_t *lock) { return __spin_lock_irqsave(lock); } @@ -146,7 +148,7 @@ EXPORT_SYMBOL(_spin_lock_irqsave); #endif #ifndef CONFIG_INLINE_SPIN_LOCK_IRQ -void __lockfunc _spin_lock_irq(spinlock_t *lock) +void __lockfunc _spin_lock_irq(raw_spinlock_t *lock) { __spin_lock_irq(lock); } @@ -154,7 +156,7 @@ EXPORT_SYMBOL(_spin_lock_irq); #endif #ifndef CONFIG_INLINE_SPIN_LOCK_BH -void __lockfunc _spin_lock_bh(spinlock_t *lock) +void __lockfunc _spin_lock_bh(raw_spinlock_t *lock) { __spin_lock_bh(lock); } @@ -162,7 +164,7 @@ EXPORT_SYMBOL(_spin_lock_bh); #endif #ifndef CONFIG_INLINE_SPIN_UNLOCK -void __lockfunc _spin_unlock(spinlock_t *lock) +void __lockfunc _spin_unlock(raw_spinlock_t *lock) { __spin_unlock(lock); } @@ -170,7 +172,7 @@ EXPORT_SYMBOL(_spin_unlock); #endif #ifndef CONFIG_INLINE_SPIN_UNLOCK_IRQRESTORE -void __lockfunc _spin_unlock_irqrestore(spinlock_t *lock, unsigned long flags) +void __lockfunc _spin_unlock_irqrestore(raw_spinlock_t *lock, unsigned long flags) { __spin_unlock_irqrestore(lock, flags); } @@ -178,7 +180,7 @@ EXPORT_SYMBOL(_spin_unlock_irqrestore); #endif #ifndef CONFIG_INLINE_SPIN_UNLOCK_IRQ -void __lockfunc _spin_unlock_irq(spinlock_t *lock) +void __lockfunc _spin_unlock_irq(raw_spinlock_t *lock) { __spin_unlock_irq(lock); } @@ -186,7 +188,7 @@ EXPORT_SYMBOL(_spin_unlock_irq); #endif #ifndef CONFIG_INLINE_SPIN_UNLOCK_BH -void __lockfunc _spin_unlock_bh(spinlock_t *lock) +void __lockfunc _spin_unlock_bh(raw_spinlock_t *lock) { __spin_unlock_bh(lock); } @@ -339,7 +341,7 @@ EXPORT_SYMBOL(_write_unlock_bh); #ifdef CONFIG_DEBUG_LOCK_ALLOC -void __lockfunc _spin_lock_nested(spinlock_t *lock, int subclass) +void __lockfunc _spin_lock_nested(raw_spinlock_t *lock, int subclass) { preempt_disable(); spin_acquire(&lock->dep_map, subclass, 0, _RET_IP_); @@ -347,7 +349,7 @@ void __lockfunc _spin_lock_nested(spinlock_t *lock, int subclass) } EXPORT_SYMBOL(_spin_lock_nested); -unsigned long __lockfunc _spin_lock_irqsave_nested(spinlock_t *lock, +unsigned long __lockfunc _spin_lock_irqsave_nested(raw_spinlock_t *lock, int subclass) { unsigned long flags; @@ -361,7 +363,7 @@ unsigned long __lockfunc _spin_lock_irqsave_nested(spinlock_t *lock, } EXPORT_SYMBOL(_spin_lock_irqsave_nested); -void __lockfunc _spin_lock_nest_lock(spinlock_t *lock, +void __lockfunc _spin_lock_nest_lock(raw_spinlock_t *lock, struct lockdep_map *nest_lock) { preempt_disable(); diff --git a/lib/spinlock_debug.c b/lib/spinlock_debug.c index 0cea0bf6114e..e705848cc33c 100644 --- a/lib/spinlock_debug.c +++ b/lib/spinlock_debug.c @@ -13,8 +13,8 @@ #include #include -void __spin_lock_init(spinlock_t *lock, const char *name, - struct lock_class_key *key) +void __raw_spin_lock_init(raw_spinlock_t *lock, const char *name, + struct lock_class_key *key) { #ifdef CONFIG_DEBUG_LOCK_ALLOC /* @@ -29,7 +29,7 @@ void __spin_lock_init(spinlock_t *lock, const char *name, lock->owner_cpu = -1; } -EXPORT_SYMBOL(__spin_lock_init); +EXPORT_SYMBOL(__raw_spin_lock_init); void __rwlock_init(rwlock_t *lock, const char *name, struct lock_class_key *key) @@ -49,7 +49,7 @@ void __rwlock_init(rwlock_t *lock, const char *name, EXPORT_SYMBOL(__rwlock_init); -static void spin_bug(spinlock_t *lock, const char *msg) +static void spin_bug(raw_spinlock_t *lock, const char *msg) { struct task_struct *owner = NULL; @@ -73,7 +73,7 @@ static void spin_bug(spinlock_t *lock, const char *msg) #define SPIN_BUG_ON(cond, lock, msg) if (unlikely(cond)) spin_bug(lock, msg) static inline void -debug_spin_lock_before(spinlock_t *lock) +debug_spin_lock_before(raw_spinlock_t *lock) { SPIN_BUG_ON(lock->magic != SPINLOCK_MAGIC, lock, "bad magic"); SPIN_BUG_ON(lock->owner == current, lock, "recursion"); @@ -81,16 +81,16 @@ debug_spin_lock_before(spinlock_t *lock) lock, "cpu recursion"); } -static inline void debug_spin_lock_after(spinlock_t *lock) +static inline void debug_spin_lock_after(raw_spinlock_t *lock) { lock->owner_cpu = raw_smp_processor_id(); lock->owner = current; } -static inline void debug_spin_unlock(spinlock_t *lock) +static inline void debug_spin_unlock(raw_spinlock_t *lock) { SPIN_BUG_ON(lock->magic != SPINLOCK_MAGIC, lock, "bad magic"); - SPIN_BUG_ON(!spin_is_locked(lock), lock, "already unlocked"); + SPIN_BUG_ON(!raw_spin_is_locked(lock), lock, "already unlocked"); SPIN_BUG_ON(lock->owner != current, lock, "wrong owner"); SPIN_BUG_ON(lock->owner_cpu != raw_smp_processor_id(), lock, "wrong CPU"); @@ -98,7 +98,7 @@ static inline void debug_spin_unlock(spinlock_t *lock) lock->owner_cpu = -1; } -static void __spin_lock_debug(spinlock_t *lock) +static void __spin_lock_debug(raw_spinlock_t *lock) { u64 i; u64 loops = loops_per_jiffy * HZ; @@ -125,7 +125,7 @@ static void __spin_lock_debug(spinlock_t *lock) } } -void _raw_spin_lock(spinlock_t *lock) +void _raw_spin_lock(raw_spinlock_t *lock) { debug_spin_lock_before(lock); if (unlikely(!arch_spin_trylock(&lock->raw_lock))) @@ -133,7 +133,7 @@ void _raw_spin_lock(spinlock_t *lock) debug_spin_lock_after(lock); } -int _raw_spin_trylock(spinlock_t *lock) +int _raw_spin_trylock(raw_spinlock_t *lock) { int ret = arch_spin_trylock(&lock->raw_lock); @@ -148,7 +148,7 @@ int _raw_spin_trylock(spinlock_t *lock) return ret; } -void _raw_spin_unlock(spinlock_t *lock) +void _raw_spin_unlock(raw_spinlock_t *lock) { debug_spin_unlock(lock); arch_spin_unlock(&lock->raw_lock); -- cgit v1.2.3-71-gd317 From 9828ea9d75c38fe3dce05d00566eed61c85732e6 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Thu, 3 Dec 2009 20:55:53 +0100 Subject: locking: Further name space cleanups The name space hierarchy for the internal lock functions is now a bit backwards. raw_spin* functions map to _spin* which use __spin*, while we would like to have _raw_spin* and __raw_spin*. _raw_spin* is already used by lock debugging, so rename those funtions to do_raw_spin* to free up the _raw_spin* name space. No functional change. Signed-off-by: Thomas Gleixner Acked-by: Peter Zijlstra Acked-by: Ingo Molnar --- include/linux/rwlock.h | 32 ++++++++++++++++---------------- include/linux/rwlock_api_smp.h | 40 ++++++++++++++++++++-------------------- include/linux/spinlock.h | 16 ++++++++-------- include/linux/spinlock_api_smp.h | 24 ++++++++++++------------ kernel/sched.c | 2 +- kernel/spinlock.c | 12 ++++++------ lib/kernel_lock.c | 18 +++++++++--------- lib/spinlock_debug.c | 18 +++++++++--------- 8 files changed, 81 insertions(+), 81 deletions(-) (limited to 'kernel') diff --git a/include/linux/rwlock.h b/include/linux/rwlock.h index 5725b034defe..bd799bc6d086 100644 --- a/include/linux/rwlock.h +++ b/include/linux/rwlock.h @@ -29,25 +29,25 @@ do { \ #endif #ifdef CONFIG_DEBUG_SPINLOCK - extern void _raw_read_lock(rwlock_t *lock); -#define _raw_read_lock_flags(lock, flags) _raw_read_lock(lock) - extern int _raw_read_trylock(rwlock_t *lock); - extern void _raw_read_unlock(rwlock_t *lock); - extern void _raw_write_lock(rwlock_t *lock); -#define _raw_write_lock_flags(lock, flags) _raw_write_lock(lock) - extern int _raw_write_trylock(rwlock_t *lock); - extern void _raw_write_unlock(rwlock_t *lock); + extern void do_raw_read_lock(rwlock_t *lock); +#define do_raw_read_lock_flags(lock, flags) do_raw_read_lock(lock) + extern int do_raw_read_trylock(rwlock_t *lock); + extern void do_raw_read_unlock(rwlock_t *lock); + extern void do_raw_write_lock(rwlock_t *lock); +#define do_raw_write_lock_flags(lock, flags) do_raw_write_lock(lock) + extern int do_raw_write_trylock(rwlock_t *lock); + extern void do_raw_write_unlock(rwlock_t *lock); #else -# define _raw_read_lock(rwlock) arch_read_lock(&(rwlock)->raw_lock) -# define _raw_read_lock_flags(lock, flags) \ +# define do_raw_read_lock(rwlock) arch_read_lock(&(rwlock)->raw_lock) +# define do_raw_read_lock_flags(lock, flags) \ arch_read_lock_flags(&(lock)->raw_lock, *(flags)) -# define _raw_read_trylock(rwlock) arch_read_trylock(&(rwlock)->raw_lock) -# define _raw_read_unlock(rwlock) arch_read_unlock(&(rwlock)->raw_lock) -# define _raw_write_lock(rwlock) arch_write_lock(&(rwlock)->raw_lock) -# define _raw_write_lock_flags(lock, flags) \ +# define do_raw_read_trylock(rwlock) arch_read_trylock(&(rwlock)->raw_lock) +# define do_raw_read_unlock(rwlock) arch_read_unlock(&(rwlock)->raw_lock) +# define do_raw_write_lock(rwlock) arch_write_lock(&(rwlock)->raw_lock) +# define do_raw_write_lock_flags(lock, flags) \ arch_write_lock_flags(&(lock)->raw_lock, *(flags)) -# define _raw_write_trylock(rwlock) arch_write_trylock(&(rwlock)->raw_lock) -# define _raw_write_unlock(rwlock) arch_write_unlock(&(rwlock)->raw_lock) +# define do_raw_write_trylock(rwlock) arch_write_trylock(&(rwlock)->raw_lock) +# define do_raw_write_unlock(rwlock) arch_write_unlock(&(rwlock)->raw_lock) #endif #define read_can_lock(rwlock) arch_read_can_lock(&(rwlock)->raw_lock) diff --git a/include/linux/rwlock_api_smp.h b/include/linux/rwlock_api_smp.h index 090f876f828d..b3ba5ae6a8c4 100644 --- a/include/linux/rwlock_api_smp.h +++ b/include/linux/rwlock_api_smp.h @@ -113,7 +113,7 @@ void __lockfunc _write_unlock_irqrestore(rwlock_t *lock, unsigned long flags) static inline int __read_trylock(rwlock_t *lock) { preempt_disable(); - if (_raw_read_trylock(lock)) { + if (do_raw_read_trylock(lock)) { rwlock_acquire_read(&lock->dep_map, 0, 1, _RET_IP_); return 1; } @@ -124,7 +124,7 @@ static inline int __read_trylock(rwlock_t *lock) static inline int __write_trylock(rwlock_t *lock) { preempt_disable(); - if (_raw_write_trylock(lock)) { + if (do_raw_write_trylock(lock)) { rwlock_acquire(&lock->dep_map, 0, 1, _RET_IP_); return 1; } @@ -143,7 +143,7 @@ static inline void __read_lock(rwlock_t *lock) { preempt_disable(); rwlock_acquire_read(&lock->dep_map, 0, 0, _RET_IP_); - LOCK_CONTENDED(lock, _raw_read_trylock, _raw_read_lock); + LOCK_CONTENDED(lock, do_raw_read_trylock, do_raw_read_lock); } static inline unsigned long __read_lock_irqsave(rwlock_t *lock) @@ -153,8 +153,8 @@ static inline unsigned long __read_lock_irqsave(rwlock_t *lock) local_irq_save(flags); preempt_disable(); rwlock_acquire_read(&lock->dep_map, 0, 0, _RET_IP_); - LOCK_CONTENDED_FLAGS(lock, _raw_read_trylock, _raw_read_lock, - _raw_read_lock_flags, &flags); + LOCK_CONTENDED_FLAGS(lock, do_raw_read_trylock, do_raw_read_lock, + do_raw_read_lock_flags, &flags); return flags; } @@ -163,7 +163,7 @@ static inline void __read_lock_irq(rwlock_t *lock) local_irq_disable(); preempt_disable(); rwlock_acquire_read(&lock->dep_map, 0, 0, _RET_IP_); - LOCK_CONTENDED(lock, _raw_read_trylock, _raw_read_lock); + LOCK_CONTENDED(lock, do_raw_read_trylock, do_raw_read_lock); } static inline void __read_lock_bh(rwlock_t *lock) @@ -171,7 +171,7 @@ static inline void __read_lock_bh(rwlock_t *lock) local_bh_disable(); preempt_disable(); rwlock_acquire_read(&lock->dep_map, 0, 0, _RET_IP_); - LOCK_CONTENDED(lock, _raw_read_trylock, _raw_read_lock); + LOCK_CONTENDED(lock, do_raw_read_trylock, do_raw_read_lock); } static inline unsigned long __write_lock_irqsave(rwlock_t *lock) @@ -181,8 +181,8 @@ static inline unsigned long __write_lock_irqsave(rwlock_t *lock) local_irq_save(flags); preempt_disable(); rwlock_acquire(&lock->dep_map, 0, 0, _RET_IP_); - LOCK_CONTENDED_FLAGS(lock, _raw_write_trylock, _raw_write_lock, - _raw_write_lock_flags, &flags); + LOCK_CONTENDED_FLAGS(lock, do_raw_write_trylock, do_raw_write_lock, + do_raw_write_lock_flags, &flags); return flags; } @@ -191,7 +191,7 @@ static inline void __write_lock_irq(rwlock_t *lock) local_irq_disable(); preempt_disable(); rwlock_acquire(&lock->dep_map, 0, 0, _RET_IP_); - LOCK_CONTENDED(lock, _raw_write_trylock, _raw_write_lock); + LOCK_CONTENDED(lock, do_raw_write_trylock, do_raw_write_lock); } static inline void __write_lock_bh(rwlock_t *lock) @@ -199,14 +199,14 @@ static inline void __write_lock_bh(rwlock_t *lock) local_bh_disable(); preempt_disable(); rwlock_acquire(&lock->dep_map, 0, 0, _RET_IP_); - LOCK_CONTENDED(lock, _raw_write_trylock, _raw_write_lock); + LOCK_CONTENDED(lock, do_raw_write_trylock, do_raw_write_lock); } static inline void __write_lock(rwlock_t *lock) { preempt_disable(); rwlock_acquire(&lock->dep_map, 0, 0, _RET_IP_); - LOCK_CONTENDED(lock, _raw_write_trylock, _raw_write_lock); + LOCK_CONTENDED(lock, do_raw_write_trylock, do_raw_write_lock); } #endif /* CONFIG_PREEMPT */ @@ -214,21 +214,21 @@ static inline void __write_lock(rwlock_t *lock) static inline void __write_unlock(rwlock_t *lock) { rwlock_release(&lock->dep_map, 1, _RET_IP_); - _raw_write_unlock(lock); + do_raw_write_unlock(lock); preempt_enable(); } static inline void __read_unlock(rwlock_t *lock) { rwlock_release(&lock->dep_map, 1, _RET_IP_); - _raw_read_unlock(lock); + do_raw_read_unlock(lock); preempt_enable(); } static inline void __read_unlock_irqrestore(rwlock_t *lock, unsigned long flags) { rwlock_release(&lock->dep_map, 1, _RET_IP_); - _raw_read_unlock(lock); + do_raw_read_unlock(lock); local_irq_restore(flags); preempt_enable(); } @@ -236,7 +236,7 @@ static inline void __read_unlock_irqrestore(rwlock_t *lock, unsigned long flags) static inline void __read_unlock_irq(rwlock_t *lock) { rwlock_release(&lock->dep_map, 1, _RET_IP_); - _raw_read_unlock(lock); + do_raw_read_unlock(lock); local_irq_enable(); preempt_enable(); } @@ -244,7 +244,7 @@ static inline void __read_unlock_irq(rwlock_t *lock) static inline void __read_unlock_bh(rwlock_t *lock) { rwlock_release(&lock->dep_map, 1, _RET_IP_); - _raw_read_unlock(lock); + do_raw_read_unlock(lock); preempt_enable_no_resched(); local_bh_enable_ip((unsigned long)__builtin_return_address(0)); } @@ -253,7 +253,7 @@ static inline void __write_unlock_irqrestore(rwlock_t *lock, unsigned long flags) { rwlock_release(&lock->dep_map, 1, _RET_IP_); - _raw_write_unlock(lock); + do_raw_write_unlock(lock); local_irq_restore(flags); preempt_enable(); } @@ -261,7 +261,7 @@ static inline void __write_unlock_irqrestore(rwlock_t *lock, static inline void __write_unlock_irq(rwlock_t *lock) { rwlock_release(&lock->dep_map, 1, _RET_IP_); - _raw_write_unlock(lock); + do_raw_write_unlock(lock); local_irq_enable(); preempt_enable(); } @@ -269,7 +269,7 @@ static inline void __write_unlock_irq(rwlock_t *lock) static inline void __write_unlock_bh(rwlock_t *lock) { rwlock_release(&lock->dep_map, 1, _RET_IP_); - _raw_write_unlock(lock); + do_raw_write_unlock(lock); preempt_enable_no_resched(); local_bh_enable_ip((unsigned long)__builtin_return_address(0)); } diff --git a/include/linux/spinlock.h b/include/linux/spinlock.h index ef5a55d96b9b..0cbc58acf689 100644 --- a/include/linux/spinlock.h +++ b/include/linux/spinlock.h @@ -128,28 +128,28 @@ static inline void smp_mb__after_lock(void) { smp_mb(); } #define raw_spin_unlock_wait(lock) arch_spin_unlock_wait(&(lock)->raw_lock) #ifdef CONFIG_DEBUG_SPINLOCK - extern void _raw_spin_lock(raw_spinlock_t *lock); -#define _raw_spin_lock_flags(lock, flags) _raw_spin_lock(lock) - extern int _raw_spin_trylock(raw_spinlock_t *lock); - extern void _raw_spin_unlock(raw_spinlock_t *lock); + extern void do_raw_spin_lock(raw_spinlock_t *lock); +#define do_raw_spin_lock_flags(lock, flags) do_raw_spin_lock(lock) + extern int do_raw_spin_trylock(raw_spinlock_t *lock); + extern void do_raw_spin_unlock(raw_spinlock_t *lock); #else -static inline void _raw_spin_lock(raw_spinlock_t *lock) +static inline void do_raw_spin_lock(raw_spinlock_t *lock) { arch_spin_lock(&lock->raw_lock); } static inline void -_raw_spin_lock_flags(raw_spinlock_t *lock, unsigned long *flags) +do_raw_spin_lock_flags(raw_spinlock_t *lock, unsigned long *flags) { arch_spin_lock_flags(&lock->raw_lock, *flags); } -static inline int _raw_spin_trylock(raw_spinlock_t *lock) +static inline int do_raw_spin_trylock(raw_spinlock_t *lock) { return arch_spin_trylock(&(lock)->raw_lock); } -static inline void _raw_spin_unlock(raw_spinlock_t *lock) +static inline void do_raw_spin_unlock(raw_spinlock_t *lock) { arch_spin_unlock(&lock->raw_lock); } diff --git a/include/linux/spinlock_api_smp.h b/include/linux/spinlock_api_smp.h index eabe5068d138..1be1fc57fc4b 100644 --- a/include/linux/spinlock_api_smp.h +++ b/include/linux/spinlock_api_smp.h @@ -85,7 +85,7 @@ _spin_unlock_irqrestore(raw_spinlock_t *lock, unsigned long flags) static inline int __spin_trylock(raw_spinlock_t *lock) { preempt_disable(); - if (_raw_spin_trylock(lock)) { + if (do_raw_spin_trylock(lock)) { spin_acquire(&lock->dep_map, 0, 1, _RET_IP_); return 1; } @@ -109,13 +109,13 @@ static inline unsigned long __spin_lock_irqsave(raw_spinlock_t *lock) spin_acquire(&lock->dep_map, 0, 0, _RET_IP_); /* * On lockdep we dont want the hand-coded irq-enable of - * _raw_spin_lock_flags() code, because lockdep assumes + * do_raw_spin_lock_flags() code, because lockdep assumes * that interrupts are not re-enabled during lock-acquire: */ #ifdef CONFIG_LOCKDEP - LOCK_CONTENDED(lock, _raw_spin_trylock, _raw_spin_lock); + LOCK_CONTENDED(lock, do_raw_spin_trylock, do_raw_spin_lock); #else - _raw_spin_lock_flags(lock, &flags); + do_raw_spin_lock_flags(lock, &flags); #endif return flags; } @@ -125,7 +125,7 @@ static inline void __spin_lock_irq(raw_spinlock_t *lock) local_irq_disable(); preempt_disable(); spin_acquire(&lock->dep_map, 0, 0, _RET_IP_); - LOCK_CONTENDED(lock, _raw_spin_trylock, _raw_spin_lock); + LOCK_CONTENDED(lock, do_raw_spin_trylock, do_raw_spin_lock); } static inline void __spin_lock_bh(raw_spinlock_t *lock) @@ -133,14 +133,14 @@ static inline void __spin_lock_bh(raw_spinlock_t *lock) local_bh_disable(); preempt_disable(); spin_acquire(&lock->dep_map, 0, 0, _RET_IP_); - LOCK_CONTENDED(lock, _raw_spin_trylock, _raw_spin_lock); + LOCK_CONTENDED(lock, do_raw_spin_trylock, do_raw_spin_lock); } static inline void __spin_lock(raw_spinlock_t *lock) { preempt_disable(); spin_acquire(&lock->dep_map, 0, 0, _RET_IP_); - LOCK_CONTENDED(lock, _raw_spin_trylock, _raw_spin_lock); + LOCK_CONTENDED(lock, do_raw_spin_trylock, do_raw_spin_lock); } #endif /* CONFIG_PREEMPT */ @@ -148,7 +148,7 @@ static inline void __spin_lock(raw_spinlock_t *lock) static inline void __spin_unlock(raw_spinlock_t *lock) { spin_release(&lock->dep_map, 1, _RET_IP_); - _raw_spin_unlock(lock); + do_raw_spin_unlock(lock); preempt_enable(); } @@ -156,7 +156,7 @@ static inline void __spin_unlock_irqrestore(raw_spinlock_t *lock, unsigned long flags) { spin_release(&lock->dep_map, 1, _RET_IP_); - _raw_spin_unlock(lock); + do_raw_spin_unlock(lock); local_irq_restore(flags); preempt_enable(); } @@ -164,7 +164,7 @@ static inline void __spin_unlock_irqrestore(raw_spinlock_t *lock, static inline void __spin_unlock_irq(raw_spinlock_t *lock) { spin_release(&lock->dep_map, 1, _RET_IP_); - _raw_spin_unlock(lock); + do_raw_spin_unlock(lock); local_irq_enable(); preempt_enable(); } @@ -172,7 +172,7 @@ static inline void __spin_unlock_irq(raw_spinlock_t *lock) static inline void __spin_unlock_bh(raw_spinlock_t *lock) { spin_release(&lock->dep_map, 1, _RET_IP_); - _raw_spin_unlock(lock); + do_raw_spin_unlock(lock); preempt_enable_no_resched(); local_bh_enable_ip((unsigned long)__builtin_return_address(0)); } @@ -181,7 +181,7 @@ static inline int __spin_trylock_bh(raw_spinlock_t *lock) { local_bh_disable(); preempt_disable(); - if (_raw_spin_trylock(lock)) { + if (do_raw_spin_trylock(lock)) { spin_acquire(&lock->dep_map, 0, 1, _RET_IP_); return 1; } diff --git a/kernel/sched.c b/kernel/sched.c index e6acf2d7b753..91c65dd91435 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -6684,7 +6684,7 @@ SYSCALL_DEFINE0(sched_yield) */ __release(rq->lock); spin_release(&rq->lock.dep_map, 1, _THIS_IP_); - _raw_spin_unlock(&rq->lock); + do_raw_spin_unlock(&rq->lock); preempt_enable_no_resched(); schedule(); diff --git a/kernel/spinlock.c b/kernel/spinlock.c index 54eb7dd3c608..795240b81224 100644 --- a/kernel/spinlock.c +++ b/kernel/spinlock.c @@ -48,7 +48,7 @@ void __lockfunc __##op##_lock(locktype##_t *lock) \ { \ for (;;) { \ preempt_disable(); \ - if (likely(_raw_##op##_trylock(lock))) \ + if (likely(do_raw_##op##_trylock(lock))) \ break; \ preempt_enable(); \ \ @@ -67,7 +67,7 @@ unsigned long __lockfunc __##op##_lock_irqsave(locktype##_t *lock) \ for (;;) { \ preempt_disable(); \ local_irq_save(flags); \ - if (likely(_raw_##op##_trylock(lock))) \ + if (likely(do_raw_##op##_trylock(lock))) \ break; \ local_irq_restore(flags); \ preempt_enable(); \ @@ -345,7 +345,7 @@ void __lockfunc _spin_lock_nested(raw_spinlock_t *lock, int subclass) { preempt_disable(); spin_acquire(&lock->dep_map, subclass, 0, _RET_IP_); - LOCK_CONTENDED(lock, _raw_spin_trylock, _raw_spin_lock); + LOCK_CONTENDED(lock, do_raw_spin_trylock, do_raw_spin_lock); } EXPORT_SYMBOL(_spin_lock_nested); @@ -357,8 +357,8 @@ unsigned long __lockfunc _spin_lock_irqsave_nested(raw_spinlock_t *lock, local_irq_save(flags); preempt_disable(); spin_acquire(&lock->dep_map, subclass, 0, _RET_IP_); - LOCK_CONTENDED_FLAGS(lock, _raw_spin_trylock, _raw_spin_lock, - _raw_spin_lock_flags, &flags); + LOCK_CONTENDED_FLAGS(lock, do_raw_spin_trylock, do_raw_spin_lock, + do_raw_spin_lock_flags, &flags); return flags; } EXPORT_SYMBOL(_spin_lock_irqsave_nested); @@ -368,7 +368,7 @@ void __lockfunc _spin_lock_nest_lock(raw_spinlock_t *lock, { preempt_disable(); spin_acquire_nest(&lock->dep_map, 0, 0, nest_lock, _RET_IP_); - LOCK_CONTENDED(lock, _raw_spin_trylock, _raw_spin_lock); + LOCK_CONTENDED(lock, do_raw_spin_trylock, do_raw_spin_lock); } EXPORT_SYMBOL(_spin_lock_nest_lock); diff --git a/lib/kernel_lock.c b/lib/kernel_lock.c index 5526b46aba94..fdd23cdb53f3 100644 --- a/lib/kernel_lock.c +++ b/lib/kernel_lock.c @@ -36,12 +36,12 @@ static __cacheline_aligned_in_smp DEFINE_SPINLOCK(kernel_flag); * If it successfully gets the lock, it should increment * the preemption count like any spinlock does. * - * (This works on UP too - _raw_spin_trylock will never + * (This works on UP too - do_raw_spin_trylock will never * return false in that case) */ int __lockfunc __reacquire_kernel_lock(void) { - while (!_raw_spin_trylock(&kernel_flag)) { + while (!do_raw_spin_trylock(&kernel_flag)) { if (need_resched()) return -EAGAIN; cpu_relax(); @@ -52,27 +52,27 @@ int __lockfunc __reacquire_kernel_lock(void) void __lockfunc __release_kernel_lock(void) { - _raw_spin_unlock(&kernel_flag); + do_raw_spin_unlock(&kernel_flag); preempt_enable_no_resched(); } /* * These are the BKL spinlocks - we try to be polite about preemption. * If SMP is not on (ie UP preemption), this all goes away because the - * _raw_spin_trylock() will always succeed. + * do_raw_spin_trylock() will always succeed. */ #ifdef CONFIG_PREEMPT static inline void __lock_kernel(void) { preempt_disable(); - if (unlikely(!_raw_spin_trylock(&kernel_flag))) { + if (unlikely(!do_raw_spin_trylock(&kernel_flag))) { /* * If preemption was disabled even before this * was called, there's nothing we can be polite * about - just spin. */ if (preempt_count() > 1) { - _raw_spin_lock(&kernel_flag); + do_raw_spin_lock(&kernel_flag); return; } @@ -85,7 +85,7 @@ static inline void __lock_kernel(void) while (spin_is_locked(&kernel_flag)) cpu_relax(); preempt_disable(); - } while (!_raw_spin_trylock(&kernel_flag)); + } while (!do_raw_spin_trylock(&kernel_flag)); } } @@ -96,7 +96,7 @@ static inline void __lock_kernel(void) */ static inline void __lock_kernel(void) { - _raw_spin_lock(&kernel_flag); + do_raw_spin_lock(&kernel_flag); } #endif @@ -106,7 +106,7 @@ static inline void __unlock_kernel(void) * the BKL is not covered by lockdep, so we open-code the * unlocking sequence (and thus avoid the dep-chain ops): */ - _raw_spin_unlock(&kernel_flag); + do_raw_spin_unlock(&kernel_flag); preempt_enable(); } diff --git a/lib/spinlock_debug.c b/lib/spinlock_debug.c index e705848cc33c..4755b98b6dfb 100644 --- a/lib/spinlock_debug.c +++ b/lib/spinlock_debug.c @@ -125,7 +125,7 @@ static void __spin_lock_debug(raw_spinlock_t *lock) } } -void _raw_spin_lock(raw_spinlock_t *lock) +void do_raw_spin_lock(raw_spinlock_t *lock) { debug_spin_lock_before(lock); if (unlikely(!arch_spin_trylock(&lock->raw_lock))) @@ -133,7 +133,7 @@ void _raw_spin_lock(raw_spinlock_t *lock) debug_spin_lock_after(lock); } -int _raw_spin_trylock(raw_spinlock_t *lock) +int do_raw_spin_trylock(raw_spinlock_t *lock) { int ret = arch_spin_trylock(&lock->raw_lock); @@ -148,7 +148,7 @@ int _raw_spin_trylock(raw_spinlock_t *lock) return ret; } -void _raw_spin_unlock(raw_spinlock_t *lock) +void do_raw_spin_unlock(raw_spinlock_t *lock) { debug_spin_unlock(lock); arch_spin_unlock(&lock->raw_lock); @@ -193,13 +193,13 @@ static void __read_lock_debug(rwlock_t *lock) } #endif -void _raw_read_lock(rwlock_t *lock) +void do_raw_read_lock(rwlock_t *lock) { RWLOCK_BUG_ON(lock->magic != RWLOCK_MAGIC, lock, "bad magic"); arch_read_lock(&lock->raw_lock); } -int _raw_read_trylock(rwlock_t *lock) +int do_raw_read_trylock(rwlock_t *lock) { int ret = arch_read_trylock(&lock->raw_lock); @@ -212,7 +212,7 @@ int _raw_read_trylock(rwlock_t *lock) return ret; } -void _raw_read_unlock(rwlock_t *lock) +void do_raw_read_unlock(rwlock_t *lock) { RWLOCK_BUG_ON(lock->magic != RWLOCK_MAGIC, lock, "bad magic"); arch_read_unlock(&lock->raw_lock); @@ -268,14 +268,14 @@ static void __write_lock_debug(rwlock_t *lock) } #endif -void _raw_write_lock(rwlock_t *lock) +void do_raw_write_lock(rwlock_t *lock) { debug_write_lock_before(lock); arch_write_lock(&lock->raw_lock); debug_write_lock_after(lock); } -int _raw_write_trylock(rwlock_t *lock) +int do_raw_write_trylock(rwlock_t *lock) { int ret = arch_write_trylock(&lock->raw_lock); @@ -290,7 +290,7 @@ int _raw_write_trylock(rwlock_t *lock) return ret; } -void _raw_write_unlock(rwlock_t *lock) +void do_raw_write_unlock(rwlock_t *lock) { debug_write_unlock(lock); arch_write_unlock(&lock->raw_lock); -- cgit v1.2.3-71-gd317 From 9c1721aa4994f6625decbd915241f3a94ee2fe67 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Thu, 3 Dec 2009 21:52:18 +0100 Subject: locking: Cleanup the name space completely Make the name space hierarchy of locking functions consistent: raw_spin* -> _raw_spin* -> __raw_spin* No functional change. Signed-off-by: Thomas Gleixner Acked-by: Peter Zijlstra Acked-by: Ingo Molnar --- include/linux/rwlock.h | 50 +++++----- include/linux/rwlock_api_smp.h | 113 ++++++++++++----------- include/linux/spinlock.h | 37 ++++---- include/linux/spinlock_api_smp.h | 79 ++++++++-------- include/linux/spinlock_api_up.h | 64 +++++++------ kernel/spinlock.c | 192 +++++++++++++++++++-------------------- 6 files changed, 274 insertions(+), 261 deletions(-) (limited to 'kernel') diff --git a/include/linux/rwlock.h b/include/linux/rwlock.h index bd799bc6d086..71e0b00b6f2c 100644 --- a/include/linux/rwlock.h +++ b/include/linux/rwlock.h @@ -38,7 +38,7 @@ do { \ extern int do_raw_write_trylock(rwlock_t *lock); extern void do_raw_write_unlock(rwlock_t *lock); #else -# define do_raw_read_lock(rwlock) arch_read_lock(&(rwlock)->raw_lock) +# define do_raw_read_lock(rwlock) arch_read_lock(&(rwlock)->raw_lock) # define do_raw_read_lock_flags(lock, flags) \ arch_read_lock_flags(&(lock)->raw_lock, *(flags)) # define do_raw_read_trylock(rwlock) arch_read_trylock(&(rwlock)->raw_lock) @@ -58,23 +58,23 @@ do { \ * regardless of whether CONFIG_SMP or CONFIG_PREEMPT are set. The various * methods are defined as nops in the case they are not required. */ -#define read_trylock(lock) __cond_lock(lock, _read_trylock(lock)) -#define write_trylock(lock) __cond_lock(lock, _write_trylock(lock)) +#define read_trylock(lock) __cond_lock(lock, _raw_read_trylock(lock)) +#define write_trylock(lock) __cond_lock(lock, _raw_write_trylock(lock)) -#define write_lock(lock) _write_lock(lock) -#define read_lock(lock) _read_lock(lock) +#define write_lock(lock) _raw_write_lock(lock) +#define read_lock(lock) _raw_read_lock(lock) #if defined(CONFIG_SMP) || defined(CONFIG_DEBUG_SPINLOCK) #define read_lock_irqsave(lock, flags) \ do { \ typecheck(unsigned long, flags); \ - flags = _read_lock_irqsave(lock); \ + flags = _raw_read_lock_irqsave(lock); \ } while (0) #define write_lock_irqsave(lock, flags) \ do { \ typecheck(unsigned long, flags); \ - flags = _write_lock_irqsave(lock); \ + flags = _raw_write_lock_irqsave(lock); \ } while (0) #else @@ -82,38 +82,38 @@ do { \ #define read_lock_irqsave(lock, flags) \ do { \ typecheck(unsigned long, flags); \ - _read_lock_irqsave(lock, flags); \ + _raw_read_lock_irqsave(lock, flags); \ } while (0) #define write_lock_irqsave(lock, flags) \ do { \ typecheck(unsigned long, flags); \ - _write_lock_irqsave(lock, flags); \ + _raw_write_lock_irqsave(lock, flags); \ } while (0) #endif -#define read_lock_irq(lock) _read_lock_irq(lock) -#define read_lock_bh(lock) _read_lock_bh(lock) -#define write_lock_irq(lock) _write_lock_irq(lock) -#define write_lock_bh(lock) _write_lock_bh(lock) -#define read_unlock(lock) _read_unlock(lock) -#define write_unlock(lock) _write_unlock(lock) -#define read_unlock_irq(lock) _read_unlock_irq(lock) -#define write_unlock_irq(lock) _write_unlock_irq(lock) - -#define read_unlock_irqrestore(lock, flags) \ - do { \ - typecheck(unsigned long, flags); \ - _read_unlock_irqrestore(lock, flags); \ +#define read_lock_irq(lock) _raw_read_lock_irq(lock) +#define read_lock_bh(lock) _raw_read_lock_bh(lock) +#define write_lock_irq(lock) _raw_write_lock_irq(lock) +#define write_lock_bh(lock) _raw_write_lock_bh(lock) +#define read_unlock(lock) _raw_read_unlock(lock) +#define write_unlock(lock) _raw_write_unlock(lock) +#define read_unlock_irq(lock) _raw_read_unlock_irq(lock) +#define write_unlock_irq(lock) _raw_write_unlock_irq(lock) + +#define read_unlock_irqrestore(lock, flags) \ + do { \ + typecheck(unsigned long, flags); \ + _raw_read_unlock_irqrestore(lock, flags); \ } while (0) -#define read_unlock_bh(lock) _read_unlock_bh(lock) +#define read_unlock_bh(lock) _raw_read_unlock_bh(lock) #define write_unlock_irqrestore(lock, flags) \ do { \ typecheck(unsigned long, flags); \ - _write_unlock_irqrestore(lock, flags); \ + _raw_write_unlock_irqrestore(lock, flags); \ } while (0) -#define write_unlock_bh(lock) _write_unlock_bh(lock) +#define write_unlock_bh(lock) _raw_write_unlock_bh(lock) #define write_trylock_irqsave(lock, flags) \ ({ \ diff --git a/include/linux/rwlock_api_smp.h b/include/linux/rwlock_api_smp.h index b3ba5ae6a8c4..9c9f0495d37c 100644 --- a/include/linux/rwlock_api_smp.h +++ b/include/linux/rwlock_api_smp.h @@ -15,102 +15,106 @@ * Released under the General Public License (GPL). */ -void __lockfunc _read_lock(rwlock_t *lock) __acquires(lock); -void __lockfunc _write_lock(rwlock_t *lock) __acquires(lock); -void __lockfunc _read_lock_bh(rwlock_t *lock) __acquires(lock); -void __lockfunc _write_lock_bh(rwlock_t *lock) __acquires(lock); -void __lockfunc _read_lock_irq(rwlock_t *lock) __acquires(lock); -void __lockfunc _write_lock_irq(rwlock_t *lock) __acquires(lock); -unsigned long __lockfunc _read_lock_irqsave(rwlock_t *lock) +void __lockfunc _raw_read_lock(rwlock_t *lock) __acquires(lock); +void __lockfunc _raw_write_lock(rwlock_t *lock) __acquires(lock); +void __lockfunc _raw_read_lock_bh(rwlock_t *lock) __acquires(lock); +void __lockfunc _raw_write_lock_bh(rwlock_t *lock) __acquires(lock); +void __lockfunc _raw_read_lock_irq(rwlock_t *lock) __acquires(lock); +void __lockfunc _raw_write_lock_irq(rwlock_t *lock) __acquires(lock); +unsigned long __lockfunc _raw_read_lock_irqsave(rwlock_t *lock) __acquires(lock); -unsigned long __lockfunc _write_lock_irqsave(rwlock_t *lock) +unsigned long __lockfunc _raw_write_lock_irqsave(rwlock_t *lock) __acquires(lock); -int __lockfunc _read_trylock(rwlock_t *lock); -int __lockfunc _write_trylock(rwlock_t *lock); -void __lockfunc _read_unlock(rwlock_t *lock) __releases(lock); -void __lockfunc _write_unlock(rwlock_t *lock) __releases(lock); -void __lockfunc _read_unlock_bh(rwlock_t *lock) __releases(lock); -void __lockfunc _write_unlock_bh(rwlock_t *lock) __releases(lock); -void __lockfunc _read_unlock_irq(rwlock_t *lock) __releases(lock); -void __lockfunc _write_unlock_irq(rwlock_t *lock) __releases(lock); -void __lockfunc _read_unlock_irqrestore(rwlock_t *lock, unsigned long flags) +int __lockfunc _raw_read_trylock(rwlock_t *lock); +int __lockfunc _raw_write_trylock(rwlock_t *lock); +void __lockfunc _raw_read_unlock(rwlock_t *lock) __releases(lock); +void __lockfunc _raw_write_unlock(rwlock_t *lock) __releases(lock); +void __lockfunc _raw_read_unlock_bh(rwlock_t *lock) __releases(lock); +void __lockfunc _raw_write_unlock_bh(rwlock_t *lock) __releases(lock); +void __lockfunc _raw_read_unlock_irq(rwlock_t *lock) __releases(lock); +void __lockfunc _raw_write_unlock_irq(rwlock_t *lock) __releases(lock); +void __lockfunc +_raw_read_unlock_irqrestore(rwlock_t *lock, unsigned long flags) __releases(lock); -void __lockfunc _write_unlock_irqrestore(rwlock_t *lock, unsigned long flags) +void __lockfunc +_raw_write_unlock_irqrestore(rwlock_t *lock, unsigned long flags) __releases(lock); #ifdef CONFIG_INLINE_READ_LOCK -#define _read_lock(lock) __read_lock(lock) +#define _raw_read_lock(lock) __raw_read_lock(lock) #endif #ifdef CONFIG_INLINE_WRITE_LOCK -#define _write_lock(lock) __write_lock(lock) +#define _raw_write_lock(lock) __raw_write_lock(lock) #endif #ifdef CONFIG_INLINE_READ_LOCK_BH -#define _read_lock_bh(lock) __read_lock_bh(lock) +#define _raw_read_lock_bh(lock) __raw_read_lock_bh(lock) #endif #ifdef CONFIG_INLINE_WRITE_LOCK_BH -#define _write_lock_bh(lock) __write_lock_bh(lock) +#define _raw_write_lock_bh(lock) __raw_write_lock_bh(lock) #endif #ifdef CONFIG_INLINE_READ_LOCK_IRQ -#define _read_lock_irq(lock) __read_lock_irq(lock) +#define _raw_read_lock_irq(lock) __raw_read_lock_irq(lock) #endif #ifdef CONFIG_INLINE_WRITE_LOCK_IRQ -#define _write_lock_irq(lock) __write_lock_irq(lock) +#define _raw_write_lock_irq(lock) __raw_write_lock_irq(lock) #endif #ifdef CONFIG_INLINE_READ_LOCK_IRQSAVE -#define _read_lock_irqsave(lock) __read_lock_irqsave(lock) +#define _raw_read_lock_irqsave(lock) __raw_read_lock_irqsave(lock) #endif #ifdef CONFIG_INLINE_WRITE_LOCK_IRQSAVE -#define _write_lock_irqsave(lock) __write_lock_irqsave(lock) +#define _raw_write_lock_irqsave(lock) __raw_write_lock_irqsave(lock) #endif #ifdef CONFIG_INLINE_READ_TRYLOCK -#define _read_trylock(lock) __read_trylock(lock) +#define _raw_read_trylock(lock) __raw_read_trylock(lock) #endif #ifdef CONFIG_INLINE_WRITE_TRYLOCK -#define _write_trylock(lock) __write_trylock(lock) +#define _raw_write_trylock(lock) __raw_write_trylock(lock) #endif #ifdef CONFIG_INLINE_READ_UNLOCK -#define _read_unlock(lock) __read_unlock(lock) +#define _raw_read_unlock(lock) __raw_read_unlock(lock) #endif #ifdef CONFIG_INLINE_WRITE_UNLOCK -#define _write_unlock(lock) __write_unlock(lock) +#define _raw_write_unlock(lock) __raw_write_unlock(lock) #endif #ifdef CONFIG_INLINE_READ_UNLOCK_BH -#define _read_unlock_bh(lock) __read_unlock_bh(lock) +#define _raw_read_unlock_bh(lock) __raw_read_unlock_bh(lock) #endif #ifdef CONFIG_INLINE_WRITE_UNLOCK_BH -#define _write_unlock_bh(lock) __write_unlock_bh(lock) +#define _raw_write_unlock_bh(lock) __raw_write_unlock_bh(lock) #endif #ifdef CONFIG_INLINE_READ_UNLOCK_IRQ -#define _read_unlock_irq(lock) __read_unlock_irq(lock) +#define _raw_read_unlock_irq(lock) __raw_read_unlock_irq(lock) #endif #ifdef CONFIG_INLINE_WRITE_UNLOCK_IRQ -#define _write_unlock_irq(lock) __write_unlock_irq(lock) +#define _raw_write_unlock_irq(lock) __raw_write_unlock_irq(lock) #endif #ifdef CONFIG_INLINE_READ_UNLOCK_IRQRESTORE -#define _read_unlock_irqrestore(lock, flags) __read_unlock_irqrestore(lock, flags) +#define _raw_read_unlock_irqrestore(lock, flags) \ + __raw_read_unlock_irqrestore(lock, flags) #endif #ifdef CONFIG_INLINE_WRITE_UNLOCK_IRQRESTORE -#define _write_unlock_irqrestore(lock, flags) __write_unlock_irqrestore(lock, flags) +#define _raw_write_unlock_irqrestore(lock, flags) \ + __raw_write_unlock_irqrestore(lock, flags) #endif -static inline int __read_trylock(rwlock_t *lock) +static inline int __raw_read_trylock(rwlock_t *lock) { preempt_disable(); if (do_raw_read_trylock(lock)) { @@ -121,7 +125,7 @@ static inline int __read_trylock(rwlock_t *lock) return 0; } -static inline int __write_trylock(rwlock_t *lock) +static inline int __raw_write_trylock(rwlock_t *lock) { preempt_disable(); if (do_raw_write_trylock(lock)) { @@ -139,14 +143,14 @@ static inline int __write_trylock(rwlock_t *lock) */ #if !defined(CONFIG_GENERIC_LOCKBREAK) || defined(CONFIG_DEBUG_LOCK_ALLOC) -static inline void __read_lock(rwlock_t *lock) +static inline void __raw_read_lock(rwlock_t *lock) { preempt_disable(); rwlock_acquire_read(&lock->dep_map, 0, 0, _RET_IP_); LOCK_CONTENDED(lock, do_raw_read_trylock, do_raw_read_lock); } -static inline unsigned long __read_lock_irqsave(rwlock_t *lock) +static inline unsigned long __raw_read_lock_irqsave(rwlock_t *lock) { unsigned long flags; @@ -158,7 +162,7 @@ static inline unsigned long __read_lock_irqsave(rwlock_t *lock) return flags; } -static inline void __read_lock_irq(rwlock_t *lock) +static inline void __raw_read_lock_irq(rwlock_t *lock) { local_irq_disable(); preempt_disable(); @@ -166,7 +170,7 @@ static inline void __read_lock_irq(rwlock_t *lock) LOCK_CONTENDED(lock, do_raw_read_trylock, do_raw_read_lock); } -static inline void __read_lock_bh(rwlock_t *lock) +static inline void __raw_read_lock_bh(rwlock_t *lock) { local_bh_disable(); preempt_disable(); @@ -174,7 +178,7 @@ static inline void __read_lock_bh(rwlock_t *lock) LOCK_CONTENDED(lock, do_raw_read_trylock, do_raw_read_lock); } -static inline unsigned long __write_lock_irqsave(rwlock_t *lock) +static inline unsigned long __raw_write_lock_irqsave(rwlock_t *lock) { unsigned long flags; @@ -186,7 +190,7 @@ static inline unsigned long __write_lock_irqsave(rwlock_t *lock) return flags; } -static inline void __write_lock_irq(rwlock_t *lock) +static inline void __raw_write_lock_irq(rwlock_t *lock) { local_irq_disable(); preempt_disable(); @@ -194,7 +198,7 @@ static inline void __write_lock_irq(rwlock_t *lock) LOCK_CONTENDED(lock, do_raw_write_trylock, do_raw_write_lock); } -static inline void __write_lock_bh(rwlock_t *lock) +static inline void __raw_write_lock_bh(rwlock_t *lock) { local_bh_disable(); preempt_disable(); @@ -202,7 +206,7 @@ static inline void __write_lock_bh(rwlock_t *lock) LOCK_CONTENDED(lock, do_raw_write_trylock, do_raw_write_lock); } -static inline void __write_lock(rwlock_t *lock) +static inline void __raw_write_lock(rwlock_t *lock) { preempt_disable(); rwlock_acquire(&lock->dep_map, 0, 0, _RET_IP_); @@ -211,21 +215,22 @@ static inline void __write_lock(rwlock_t *lock) #endif /* CONFIG_PREEMPT */ -static inline void __write_unlock(rwlock_t *lock) +static inline void __raw_write_unlock(rwlock_t *lock) { rwlock_release(&lock->dep_map, 1, _RET_IP_); do_raw_write_unlock(lock); preempt_enable(); } -static inline void __read_unlock(rwlock_t *lock) +static inline void __raw_read_unlock(rwlock_t *lock) { rwlock_release(&lock->dep_map, 1, _RET_IP_); do_raw_read_unlock(lock); preempt_enable(); } -static inline void __read_unlock_irqrestore(rwlock_t *lock, unsigned long flags) +static inline void +__raw_read_unlock_irqrestore(rwlock_t *lock, unsigned long flags) { rwlock_release(&lock->dep_map, 1, _RET_IP_); do_raw_read_unlock(lock); @@ -233,7 +238,7 @@ static inline void __read_unlock_irqrestore(rwlock_t *lock, unsigned long flags) preempt_enable(); } -static inline void __read_unlock_irq(rwlock_t *lock) +static inline void __raw_read_unlock_irq(rwlock_t *lock) { rwlock_release(&lock->dep_map, 1, _RET_IP_); do_raw_read_unlock(lock); @@ -241,7 +246,7 @@ static inline void __read_unlock_irq(rwlock_t *lock) preempt_enable(); } -static inline void __read_unlock_bh(rwlock_t *lock) +static inline void __raw_read_unlock_bh(rwlock_t *lock) { rwlock_release(&lock->dep_map, 1, _RET_IP_); do_raw_read_unlock(lock); @@ -249,7 +254,7 @@ static inline void __read_unlock_bh(rwlock_t *lock) local_bh_enable_ip((unsigned long)__builtin_return_address(0)); } -static inline void __write_unlock_irqrestore(rwlock_t *lock, +static inline void __raw_write_unlock_irqrestore(rwlock_t *lock, unsigned long flags) { rwlock_release(&lock->dep_map, 1, _RET_IP_); @@ -258,7 +263,7 @@ static inline void __write_unlock_irqrestore(rwlock_t *lock, preempt_enable(); } -static inline void __write_unlock_irq(rwlock_t *lock) +static inline void __raw_write_unlock_irq(rwlock_t *lock) { rwlock_release(&lock->dep_map, 1, _RET_IP_); do_raw_write_unlock(lock); @@ -266,7 +271,7 @@ static inline void __write_unlock_irq(rwlock_t *lock) preempt_enable(); } -static inline void __write_unlock_bh(rwlock_t *lock) +static inline void __raw_write_unlock_bh(rwlock_t *lock) { rwlock_release(&lock->dep_map, 1, _RET_IP_); do_raw_write_unlock(lock); diff --git a/include/linux/spinlock.h b/include/linux/spinlock.h index 0cbc58acf689..86088213334a 100644 --- a/include/linux/spinlock.h +++ b/include/linux/spinlock.h @@ -161,20 +161,22 @@ static inline void do_raw_spin_unlock(raw_spinlock_t *lock) * various methods are defined as nops in the case they are not * required. */ -#define raw_spin_trylock(lock) __cond_lock(lock, _spin_trylock(lock)) +#define raw_spin_trylock(lock) __cond_lock(lock, _raw_spin_trylock(lock)) -#define raw_spin_lock(lock) _spin_lock(lock) +#define raw_spin_lock(lock) _raw_spin_lock(lock) #ifdef CONFIG_DEBUG_LOCK_ALLOC -# define raw_spin_lock_nested(lock, subclass) _spin_lock_nested(lock, subclass) +# define raw_spin_lock_nested(lock, subclass) \ + _raw_spin_lock_nested(lock, subclass) + # define raw_spin_lock_nest_lock(lock, nest_lock) \ do { \ typecheck(struct lockdep_map *, &(nest_lock)->dep_map);\ - _spin_lock_nest_lock(lock, &(nest_lock)->dep_map); \ + _raw_spin_lock_nest_lock(lock, &(nest_lock)->dep_map); \ } while (0) #else -# define raw_spin_lock_nested(lock, subclass) _spin_lock(lock) -# define raw_spin_lock_nest_lock(lock, nest_lock) _spin_lock(lock) +# define raw_spin_lock_nested(lock, subclass) _raw_spin_lock(lock) +# define raw_spin_lock_nest_lock(lock, nest_lock) _raw_spin_lock(lock) #endif #if defined(CONFIG_SMP) || defined(CONFIG_DEBUG_SPINLOCK) @@ -182,20 +184,20 @@ static inline void do_raw_spin_unlock(raw_spinlock_t *lock) #define raw_spin_lock_irqsave(lock, flags) \ do { \ typecheck(unsigned long, flags); \ - flags = _spin_lock_irqsave(lock); \ + flags = _raw_spin_lock_irqsave(lock); \ } while (0) #ifdef CONFIG_DEBUG_LOCK_ALLOC #define raw_spin_lock_irqsave_nested(lock, flags, subclass) \ do { \ typecheck(unsigned long, flags); \ - flags = _spin_lock_irqsave_nested(lock, subclass); \ + flags = _raw_spin_lock_irqsave_nested(lock, subclass); \ } while (0) #else #define raw_spin_lock_irqsave_nested(lock, flags, subclass) \ do { \ typecheck(unsigned long, flags); \ - flags = _spin_lock_irqsave(lock); \ + flags = _raw_spin_lock_irqsave(lock); \ } while (0) #endif @@ -204,7 +206,7 @@ static inline void do_raw_spin_unlock(raw_spinlock_t *lock) #define raw_spin_lock_irqsave(lock, flags) \ do { \ typecheck(unsigned long, flags); \ - _spin_lock_irqsave(lock, flags); \ + _raw_spin_lock_irqsave(lock, flags); \ } while (0) #define raw_spin_lock_irqsave_nested(lock, flags, subclass) \ @@ -212,19 +214,20 @@ static inline void do_raw_spin_unlock(raw_spinlock_t *lock) #endif -#define raw_spin_lock_irq(lock) _spin_lock_irq(lock) -#define raw_spin_lock_bh(lock) _spin_lock_bh(lock) -#define raw_spin_unlock(lock) _spin_unlock(lock) -#define raw_spin_unlock_irq(lock) _spin_unlock_irq(lock) +#define raw_spin_lock_irq(lock) _raw_spin_lock_irq(lock) +#define raw_spin_lock_bh(lock) _raw_spin_lock_bh(lock) +#define raw_spin_unlock(lock) _raw_spin_unlock(lock) +#define raw_spin_unlock_irq(lock) _raw_spin_unlock_irq(lock) #define raw_spin_unlock_irqrestore(lock, flags) \ do { \ typecheck(unsigned long, flags); \ - _spin_unlock_irqrestore(lock, flags); \ + _raw_spin_unlock_irqrestore(lock, flags); \ } while (0) -#define raw_spin_unlock_bh(lock) _spin_unlock_bh(lock) +#define raw_spin_unlock_bh(lock) _raw_spin_unlock_bh(lock) -#define raw_spin_trylock_bh(lock) __cond_lock(lock, _spin_trylock_bh(lock)) +#define raw_spin_trylock_bh(lock) \ + __cond_lock(lock, _raw_spin_trylock_bh(lock)) #define raw_spin_trylock_irq(lock) \ ({ \ diff --git a/include/linux/spinlock_api_smp.h b/include/linux/spinlock_api_smp.h index 1be1fc57fc4b..e253ccd7a604 100644 --- a/include/linux/spinlock_api_smp.h +++ b/include/linux/spinlock_api_smp.h @@ -19,70 +19,71 @@ int in_lock_functions(unsigned long addr); #define assert_raw_spin_locked(x) BUG_ON(!raw_spin_is_locked(x)) -void __lockfunc _spin_lock(raw_spinlock_t *lock) __acquires(lock); -void __lockfunc _spin_lock_nested(raw_spinlock_t *lock, int subclass) - __acquires(lock); +void __lockfunc _raw_spin_lock(raw_spinlock_t *lock) __acquires(lock); +void __lockfunc _raw_spin_lock_nested(raw_spinlock_t *lock, int subclass) + __acquires(lock); void __lockfunc -_spin_lock_nest_lock(raw_spinlock_t *lock, struct lockdep_map *map) - __acquires(lock); -void __lockfunc _spin_lock_bh(raw_spinlock_t *lock) __acquires(lock); -void __lockfunc _spin_lock_irq(raw_spinlock_t *lock) __acquires(lock); - -unsigned long __lockfunc _spin_lock_irqsave(raw_spinlock_t *lock) - __acquires(lock); +_raw_spin_lock_nest_lock(raw_spinlock_t *lock, struct lockdep_map *map) + __acquires(lock); +void __lockfunc _raw_spin_lock_bh(raw_spinlock_t *lock) __acquires(lock); +void __lockfunc _raw_spin_lock_irq(raw_spinlock_t *lock) + __acquires(lock); + +unsigned long __lockfunc _raw_spin_lock_irqsave(raw_spinlock_t *lock) + __acquires(lock); unsigned long __lockfunc -_spin_lock_irqsave_nested(raw_spinlock_t *lock, int subclass) - __acquires(lock); -int __lockfunc _spin_trylock(raw_spinlock_t *lock); -int __lockfunc _spin_trylock_bh(raw_spinlock_t *lock); -void __lockfunc _spin_unlock(raw_spinlock_t *lock) __releases(lock); -void __lockfunc _spin_unlock_bh(raw_spinlock_t *lock) __releases(lock); -void __lockfunc _spin_unlock_irq(raw_spinlock_t *lock) __releases(lock); +_raw_spin_lock_irqsave_nested(raw_spinlock_t *lock, int subclass) + __acquires(lock); +int __lockfunc _raw_spin_trylock(raw_spinlock_t *lock); +int __lockfunc _raw_spin_trylock_bh(raw_spinlock_t *lock); +void __lockfunc _raw_spin_unlock(raw_spinlock_t *lock) __releases(lock); +void __lockfunc _raw_spin_unlock_bh(raw_spinlock_t *lock) __releases(lock); +void __lockfunc _raw_spin_unlock_irq(raw_spinlock_t *lock) __releases(lock); void __lockfunc -_spin_unlock_irqrestore(raw_spinlock_t *lock, unsigned long flags) - __releases(lock); +_raw_spin_unlock_irqrestore(raw_spinlock_t *lock, unsigned long flags) + __releases(lock); #ifdef CONFIG_INLINE_SPIN_LOCK -#define _spin_lock(lock) __spin_lock(lock) +#define _raw_spin_lock(lock) __raw_spin_lock(lock) #endif #ifdef CONFIG_INLINE_SPIN_LOCK_BH -#define _spin_lock_bh(lock) __spin_lock_bh(lock) +#define _raw_spin_lock_bh(lock) __raw_spin_lock_bh(lock) #endif #ifdef CONFIG_INLINE_SPIN_LOCK_IRQ -#define _spin_lock_irq(lock) __spin_lock_irq(lock) +#define _raw_spin_lock_irq(lock) __raw_spin_lock_irq(lock) #endif #ifdef CONFIG_INLINE_SPIN_LOCK_IRQSAVE -#define _spin_lock_irqsave(lock) __spin_lock_irqsave(lock) +#define _raw_spin_lock_irqsave(lock) __raw_spin_lock_irqsave(lock) #endif #ifdef CONFIG_INLINE_SPIN_TRYLOCK -#define _spin_trylock(lock) __spin_trylock(lock) +#define _raw_spin_trylock(lock) __raw_spin_trylock(lock) #endif #ifdef CONFIG_INLINE_SPIN_TRYLOCK_BH -#define _spin_trylock_bh(lock) __spin_trylock_bh(lock) +#define _raw_spin_trylock_bh(lock) __raw_spin_trylock_bh(lock) #endif #ifdef CONFIG_INLINE_SPIN_UNLOCK -#define _spin_unlock(lock) __spin_unlock(lock) +#define _raw_spin_unlock(lock) __raw_spin_unlock(lock) #endif #ifdef CONFIG_INLINE_SPIN_UNLOCK_BH -#define _spin_unlock_bh(lock) __spin_unlock_bh(lock) +#define _raw_spin_unlock_bh(lock) __raw_spin_unlock_bh(lock) #endif #ifdef CONFIG_INLINE_SPIN_UNLOCK_IRQ -#define _spin_unlock_irq(lock) __spin_unlock_irq(lock) +#define _raw_spin_unlock_irq(lock) __raw_spin_unlock_irq(lock) #endif #ifdef CONFIG_INLINE_SPIN_UNLOCK_IRQRESTORE -#define _spin_unlock_irqrestore(lock, flags) __spin_unlock_irqrestore(lock, flags) +#define _raw_spin_unlock_irqrestore(lock, flags) __raw_spin_unlock_irqrestore(lock, flags) #endif -static inline int __spin_trylock(raw_spinlock_t *lock) +static inline int __raw_spin_trylock(raw_spinlock_t *lock) { preempt_disable(); if (do_raw_spin_trylock(lock)) { @@ -100,7 +101,7 @@ static inline int __spin_trylock(raw_spinlock_t *lock) */ #if !defined(CONFIG_GENERIC_LOCKBREAK) || defined(CONFIG_DEBUG_LOCK_ALLOC) -static inline unsigned long __spin_lock_irqsave(raw_spinlock_t *lock) +static inline unsigned long __raw_spin_lock_irqsave(raw_spinlock_t *lock) { unsigned long flags; @@ -120,7 +121,7 @@ static inline unsigned long __spin_lock_irqsave(raw_spinlock_t *lock) return flags; } -static inline void __spin_lock_irq(raw_spinlock_t *lock) +static inline void __raw_spin_lock_irq(raw_spinlock_t *lock) { local_irq_disable(); preempt_disable(); @@ -128,7 +129,7 @@ static inline void __spin_lock_irq(raw_spinlock_t *lock) LOCK_CONTENDED(lock, do_raw_spin_trylock, do_raw_spin_lock); } -static inline void __spin_lock_bh(raw_spinlock_t *lock) +static inline void __raw_spin_lock_bh(raw_spinlock_t *lock) { local_bh_disable(); preempt_disable(); @@ -136,7 +137,7 @@ static inline void __spin_lock_bh(raw_spinlock_t *lock) LOCK_CONTENDED(lock, do_raw_spin_trylock, do_raw_spin_lock); } -static inline void __spin_lock(raw_spinlock_t *lock) +static inline void __raw_spin_lock(raw_spinlock_t *lock) { preempt_disable(); spin_acquire(&lock->dep_map, 0, 0, _RET_IP_); @@ -145,14 +146,14 @@ static inline void __spin_lock(raw_spinlock_t *lock) #endif /* CONFIG_PREEMPT */ -static inline void __spin_unlock(raw_spinlock_t *lock) +static inline void __raw_spin_unlock(raw_spinlock_t *lock) { spin_release(&lock->dep_map, 1, _RET_IP_); do_raw_spin_unlock(lock); preempt_enable(); } -static inline void __spin_unlock_irqrestore(raw_spinlock_t *lock, +static inline void __raw_spin_unlock_irqrestore(raw_spinlock_t *lock, unsigned long flags) { spin_release(&lock->dep_map, 1, _RET_IP_); @@ -161,7 +162,7 @@ static inline void __spin_unlock_irqrestore(raw_spinlock_t *lock, preempt_enable(); } -static inline void __spin_unlock_irq(raw_spinlock_t *lock) +static inline void __raw_spin_unlock_irq(raw_spinlock_t *lock) { spin_release(&lock->dep_map, 1, _RET_IP_); do_raw_spin_unlock(lock); @@ -169,7 +170,7 @@ static inline void __spin_unlock_irq(raw_spinlock_t *lock) preempt_enable(); } -static inline void __spin_unlock_bh(raw_spinlock_t *lock) +static inline void __raw_spin_unlock_bh(raw_spinlock_t *lock) { spin_release(&lock->dep_map, 1, _RET_IP_); do_raw_spin_unlock(lock); @@ -177,7 +178,7 @@ static inline void __spin_unlock_bh(raw_spinlock_t *lock) local_bh_enable_ip((unsigned long)__builtin_return_address(0)); } -static inline int __spin_trylock_bh(raw_spinlock_t *lock) +static inline int __raw_spin_trylock_bh(raw_spinlock_t *lock) { local_bh_disable(); preempt_disable(); diff --git a/include/linux/spinlock_api_up.h b/include/linux/spinlock_api_up.h index 3a9e27adecf9..af1f47229e70 100644 --- a/include/linux/spinlock_api_up.h +++ b/include/linux/spinlock_api_up.h @@ -40,7 +40,8 @@ do { preempt_enable(); __release(lock); (void)(lock); } while (0) #define __UNLOCK_BH(lock) \ - do { preempt_enable_no_resched(); local_bh_enable(); __release(lock); (void)(lock); } while (0) + do { preempt_enable_no_resched(); local_bh_enable(); \ + __release(lock); (void)(lock); } while (0) #define __UNLOCK_IRQ(lock) \ do { local_irq_enable(); __UNLOCK(lock); } while (0) @@ -48,34 +49,37 @@ #define __UNLOCK_IRQRESTORE(lock, flags) \ do { local_irq_restore(flags); __UNLOCK(lock); } while (0) -#define _spin_lock(lock) __LOCK(lock) -#define _spin_lock_nested(lock, subclass) __LOCK(lock) -#define _read_lock(lock) __LOCK(lock) -#define _write_lock(lock) __LOCK(lock) -#define _spin_lock_bh(lock) __LOCK_BH(lock) -#define _read_lock_bh(lock) __LOCK_BH(lock) -#define _write_lock_bh(lock) __LOCK_BH(lock) -#define _spin_lock_irq(lock) __LOCK_IRQ(lock) -#define _read_lock_irq(lock) __LOCK_IRQ(lock) -#define _write_lock_irq(lock) __LOCK_IRQ(lock) -#define _spin_lock_irqsave(lock, flags) __LOCK_IRQSAVE(lock, flags) -#define _read_lock_irqsave(lock, flags) __LOCK_IRQSAVE(lock, flags) -#define _write_lock_irqsave(lock, flags) __LOCK_IRQSAVE(lock, flags) -#define _spin_trylock(lock) ({ __LOCK(lock); 1; }) -#define _read_trylock(lock) ({ __LOCK(lock); 1; }) -#define _write_trylock(lock) ({ __LOCK(lock); 1; }) -#define _spin_trylock_bh(lock) ({ __LOCK_BH(lock); 1; }) -#define _spin_unlock(lock) __UNLOCK(lock) -#define _read_unlock(lock) __UNLOCK(lock) -#define _write_unlock(lock) __UNLOCK(lock) -#define _spin_unlock_bh(lock) __UNLOCK_BH(lock) -#define _write_unlock_bh(lock) __UNLOCK_BH(lock) -#define _read_unlock_bh(lock) __UNLOCK_BH(lock) -#define _spin_unlock_irq(lock) __UNLOCK_IRQ(lock) -#define _read_unlock_irq(lock) __UNLOCK_IRQ(lock) -#define _write_unlock_irq(lock) __UNLOCK_IRQ(lock) -#define _spin_unlock_irqrestore(lock, flags) __UNLOCK_IRQRESTORE(lock, flags) -#define _read_unlock_irqrestore(lock, flags) __UNLOCK_IRQRESTORE(lock, flags) -#define _write_unlock_irqrestore(lock, flags) __UNLOCK_IRQRESTORE(lock, flags) +#define _raw_spin_lock(lock) __LOCK(lock) +#define _raw_spin_lock_nested(lock, subclass) __LOCK(lock) +#define _raw_read_lock(lock) __LOCK(lock) +#define _raw_write_lock(lock) __LOCK(lock) +#define _raw_spin_lock_bh(lock) __LOCK_BH(lock) +#define _raw_read_lock_bh(lock) __LOCK_BH(lock) +#define _raw_write_lock_bh(lock) __LOCK_BH(lock) +#define _raw_spin_lock_irq(lock) __LOCK_IRQ(lock) +#define _raw_read_lock_irq(lock) __LOCK_IRQ(lock) +#define _raw_write_lock_irq(lock) __LOCK_IRQ(lock) +#define _raw_spin_lock_irqsave(lock, flags) __LOCK_IRQSAVE(lock, flags) +#define _raw_read_lock_irqsave(lock, flags) __LOCK_IRQSAVE(lock, flags) +#define _raw_write_lock_irqsave(lock, flags) __LOCK_IRQSAVE(lock, flags) +#define _raw_spin_trylock(lock) ({ __LOCK(lock); 1; }) +#define _raw_read_trylock(lock) ({ __LOCK(lock); 1; }) +#define _raw_write_trylock(lock) ({ __LOCK(lock); 1; }) +#define _raw_spin_trylock_bh(lock) ({ __LOCK_BH(lock); 1; }) +#define _raw_spin_unlock(lock) __UNLOCK(lock) +#define _raw_read_unlock(lock) __UNLOCK(lock) +#define _raw_write_unlock(lock) __UNLOCK(lock) +#define _raw_spin_unlock_bh(lock) __UNLOCK_BH(lock) +#define _raw_write_unlock_bh(lock) __UNLOCK_BH(lock) +#define _raw_read_unlock_bh(lock) __UNLOCK_BH(lock) +#define _raw_spin_unlock_irq(lock) __UNLOCK_IRQ(lock) +#define _raw_read_unlock_irq(lock) __UNLOCK_IRQ(lock) +#define _raw_write_unlock_irq(lock) __UNLOCK_IRQ(lock) +#define _raw_spin_unlock_irqrestore(lock, flags) \ + __UNLOCK_IRQRESTORE(lock, flags) +#define _raw_read_unlock_irqrestore(lock, flags) \ + __UNLOCK_IRQRESTORE(lock, flags) +#define _raw_write_unlock_irqrestore(lock, flags) \ + __UNLOCK_IRQRESTORE(lock, flags) #endif /* __LINUX_SPINLOCK_API_UP_H */ diff --git a/kernel/spinlock.c b/kernel/spinlock.c index 795240b81224..be6517fb9c14 100644 --- a/kernel/spinlock.c +++ b/kernel/spinlock.c @@ -44,7 +44,7 @@ * towards that other CPU that it should break the lock ASAP. */ #define BUILD_LOCK_OPS(op, locktype) \ -void __lockfunc __##op##_lock(locktype##_t *lock) \ +void __lockfunc __raw_##op##_lock(locktype##_t *lock) \ { \ for (;;) { \ preempt_disable(); \ @@ -60,7 +60,7 @@ void __lockfunc __##op##_lock(locktype##_t *lock) \ (lock)->break_lock = 0; \ } \ \ -unsigned long __lockfunc __##op##_lock_irqsave(locktype##_t *lock) \ +unsigned long __lockfunc __raw_##op##_lock_irqsave(locktype##_t *lock) \ { \ unsigned long flags; \ \ @@ -81,12 +81,12 @@ unsigned long __lockfunc __##op##_lock_irqsave(locktype##_t *lock) \ return flags; \ } \ \ -void __lockfunc __##op##_lock_irq(locktype##_t *lock) \ +void __lockfunc __raw_##op##_lock_irq(locktype##_t *lock) \ { \ - _##op##_lock_irqsave(lock); \ + _raw_##op##_lock_irqsave(lock); \ } \ \ -void __lockfunc __##op##_lock_bh(locktype##_t *lock) \ +void __lockfunc __raw_##op##_lock_bh(locktype##_t *lock) \ { \ unsigned long flags; \ \ @@ -95,7 +95,7 @@ void __lockfunc __##op##_lock_bh(locktype##_t *lock) \ /* irq-disabling. We use the generic preemption-aware */ \ /* function: */ \ /**/ \ - flags = _##op##_lock_irqsave(lock); \ + flags = _raw_##op##_lock_irqsave(lock); \ local_bh_disable(); \ local_irq_restore(flags); \ } \ @@ -116,240 +116,240 @@ BUILD_LOCK_OPS(write, rwlock); #endif #ifndef CONFIG_INLINE_SPIN_TRYLOCK -int __lockfunc _spin_trylock(raw_spinlock_t *lock) +int __lockfunc _raw_spin_trylock(raw_spinlock_t *lock) { - return __spin_trylock(lock); + return __raw_spin_trylock(lock); } -EXPORT_SYMBOL(_spin_trylock); +EXPORT_SYMBOL(_raw_spin_trylock); #endif #ifndef CONFIG_INLINE_SPIN_TRYLOCK_BH -int __lockfunc _spin_trylock_bh(raw_spinlock_t *lock) +int __lockfunc _raw_spin_trylock_bh(raw_spinlock_t *lock) { - return __spin_trylock_bh(lock); + return __raw_spin_trylock_bh(lock); } -EXPORT_SYMBOL(_spin_trylock_bh); +EXPORT_SYMBOL(_raw_spin_trylock_bh); #endif #ifndef CONFIG_INLINE_SPIN_LOCK -void __lockfunc _spin_lock(raw_spinlock_t *lock) +void __lockfunc _raw_spin_lock(raw_spinlock_t *lock) { - __spin_lock(lock); + __raw_spin_lock(lock); } -EXPORT_SYMBOL(_spin_lock); +EXPORT_SYMBOL(_raw_spin_lock); #endif #ifndef CONFIG_INLINE_SPIN_LOCK_IRQSAVE -unsigned long __lockfunc _spin_lock_irqsave(raw_spinlock_t *lock) +unsigned long __lockfunc _raw_spin_lock_irqsave(raw_spinlock_t *lock) { - return __spin_lock_irqsave(lock); + return __raw_spin_lock_irqsave(lock); } -EXPORT_SYMBOL(_spin_lock_irqsave); +EXPORT_SYMBOL(_raw_spin_lock_irqsave); #endif #ifndef CONFIG_INLINE_SPIN_LOCK_IRQ -void __lockfunc _spin_lock_irq(raw_spinlock_t *lock) +void __lockfunc _raw_spin_lock_irq(raw_spinlock_t *lock) { - __spin_lock_irq(lock); + __raw_spin_lock_irq(lock); } -EXPORT_SYMBOL(_spin_lock_irq); +EXPORT_SYMBOL(_raw_spin_lock_irq); #endif #ifndef CONFIG_INLINE_SPIN_LOCK_BH -void __lockfunc _spin_lock_bh(raw_spinlock_t *lock) +void __lockfunc _raw_spin_lock_bh(raw_spinlock_t *lock) { - __spin_lock_bh(lock); + __raw_spin_lock_bh(lock); } -EXPORT_SYMBOL(_spin_lock_bh); +EXPORT_SYMBOL(_raw_spin_lock_bh); #endif #ifndef CONFIG_INLINE_SPIN_UNLOCK -void __lockfunc _spin_unlock(raw_spinlock_t *lock) +void __lockfunc _raw_spin_unlock(raw_spinlock_t *lock) { - __spin_unlock(lock); + __raw_spin_unlock(lock); } -EXPORT_SYMBOL(_spin_unlock); +EXPORT_SYMBOL(_raw_spin_unlock); #endif #ifndef CONFIG_INLINE_SPIN_UNLOCK_IRQRESTORE -void __lockfunc _spin_unlock_irqrestore(raw_spinlock_t *lock, unsigned long flags) +void __lockfunc _raw_spin_unlock_irqrestore(raw_spinlock_t *lock, unsigned long flags) { - __spin_unlock_irqrestore(lock, flags); + __raw_spin_unlock_irqrestore(lock, flags); } -EXPORT_SYMBOL(_spin_unlock_irqrestore); +EXPORT_SYMBOL(_raw_spin_unlock_irqrestore); #endif #ifndef CONFIG_INLINE_SPIN_UNLOCK_IRQ -void __lockfunc _spin_unlock_irq(raw_spinlock_t *lock) +void __lockfunc _raw_spin_unlock_irq(raw_spinlock_t *lock) { - __spin_unlock_irq(lock); + __raw_spin_unlock_irq(lock); } -EXPORT_SYMBOL(_spin_unlock_irq); +EXPORT_SYMBOL(_raw_spin_unlock_irq); #endif #ifndef CONFIG_INLINE_SPIN_UNLOCK_BH -void __lockfunc _spin_unlock_bh(raw_spinlock_t *lock) +void __lockfunc _raw_spin_unlock_bh(raw_spinlock_t *lock) { - __spin_unlock_bh(lock); + __raw_spin_unlock_bh(lock); } -EXPORT_SYMBOL(_spin_unlock_bh); +EXPORT_SYMBOL(_raw_spin_unlock_bh); #endif #ifndef CONFIG_INLINE_READ_TRYLOCK -int __lockfunc _read_trylock(rwlock_t *lock) +int __lockfunc _raw_read_trylock(rwlock_t *lock) { - return __read_trylock(lock); + return __raw_read_trylock(lock); } -EXPORT_SYMBOL(_read_trylock); +EXPORT_SYMBOL(_raw_read_trylock); #endif #ifndef CONFIG_INLINE_READ_LOCK -void __lockfunc _read_lock(rwlock_t *lock) +void __lockfunc _raw_read_lock(rwlock_t *lock) { - __read_lock(lock); + __raw_read_lock(lock); } -EXPORT_SYMBOL(_read_lock); +EXPORT_SYMBOL(_raw_read_lock); #endif #ifndef CONFIG_INLINE_READ_LOCK_IRQSAVE -unsigned long __lockfunc _read_lock_irqsave(rwlock_t *lock) +unsigned long __lockfunc _raw_read_lock_irqsave(rwlock_t *lock) { - return __read_lock_irqsave(lock); + return __raw_read_lock_irqsave(lock); } -EXPORT_SYMBOL(_read_lock_irqsave); +EXPORT_SYMBOL(_raw_read_lock_irqsave); #endif #ifndef CONFIG_INLINE_READ_LOCK_IRQ -void __lockfunc _read_lock_irq(rwlock_t *lock) +void __lockfunc _raw_read_lock_irq(rwlock_t *lock) { - __read_lock_irq(lock); + __raw_read_lock_irq(lock); } -EXPORT_SYMBOL(_read_lock_irq); +EXPORT_SYMBOL(_raw_read_lock_irq); #endif #ifndef CONFIG_INLINE_READ_LOCK_BH -void __lockfunc _read_lock_bh(rwlock_t *lock) +void __lockfunc _raw_read_lock_bh(rwlock_t *lock) { - __read_lock_bh(lock); + __raw_read_lock_bh(lock); } -EXPORT_SYMBOL(_read_lock_bh); +EXPORT_SYMBOL(_raw_read_lock_bh); #endif #ifndef CONFIG_INLINE_READ_UNLOCK -void __lockfunc _read_unlock(rwlock_t *lock) +void __lockfunc _raw_read_unlock(rwlock_t *lock) { - __read_unlock(lock); + __raw_read_unlock(lock); } -EXPORT_SYMBOL(_read_unlock); +EXPORT_SYMBOL(_raw_read_unlock); #endif #ifndef CONFIG_INLINE_READ_UNLOCK_IRQRESTORE -void __lockfunc _read_unlock_irqrestore(rwlock_t *lock, unsigned long flags) +void __lockfunc _raw_read_unlock_irqrestore(rwlock_t *lock, unsigned long flags) { - __read_unlock_irqrestore(lock, flags); + __raw_read_unlock_irqrestore(lock, flags); } -EXPORT_SYMBOL(_read_unlock_irqrestore); +EXPORT_SYMBOL(_raw_read_unlock_irqrestore); #endif #ifndef CONFIG_INLINE_READ_UNLOCK_IRQ -void __lockfunc _read_unlock_irq(rwlock_t *lock) +void __lockfunc _raw_read_unlock_irq(rwlock_t *lock) { - __read_unlock_irq(lock); + __raw_read_unlock_irq(lock); } -EXPORT_SYMBOL(_read_unlock_irq); +EXPORT_SYMBOL(_raw_read_unlock_irq); #endif #ifndef CONFIG_INLINE_READ_UNLOCK_BH -void __lockfunc _read_unlock_bh(rwlock_t *lock) +void __lockfunc _raw_read_unlock_bh(rwlock_t *lock) { - __read_unlock_bh(lock); + __raw_read_unlock_bh(lock); } -EXPORT_SYMBOL(_read_unlock_bh); +EXPORT_SYMBOL(_raw_read_unlock_bh); #endif #ifndef CONFIG_INLINE_WRITE_TRYLOCK -int __lockfunc _write_trylock(rwlock_t *lock) +int __lockfunc _raw_write_trylock(rwlock_t *lock) { - return __write_trylock(lock); + return __raw_write_trylock(lock); } -EXPORT_SYMBOL(_write_trylock); +EXPORT_SYMBOL(_raw_write_trylock); #endif #ifndef CONFIG_INLINE_WRITE_LOCK -void __lockfunc _write_lock(rwlock_t *lock) +void __lockfunc _raw_write_lock(rwlock_t *lock) { - __write_lock(lock); + __raw_write_lock(lock); } -EXPORT_SYMBOL(_write_lock); +EXPORT_SYMBOL(_raw_write_lock); #endif #ifndef CONFIG_INLINE_WRITE_LOCK_IRQSAVE -unsigned long __lockfunc _write_lock_irqsave(rwlock_t *lock) +unsigned long __lockfunc _raw_write_lock_irqsave(rwlock_t *lock) { - return __write_lock_irqsave(lock); + return __raw_write_lock_irqsave(lock); } -EXPORT_SYMBOL(_write_lock_irqsave); +EXPORT_SYMBOL(_raw_write_lock_irqsave); #endif #ifndef CONFIG_INLINE_WRITE_LOCK_IRQ -void __lockfunc _write_lock_irq(rwlock_t *lock) +void __lockfunc _raw_write_lock_irq(rwlock_t *lock) { - __write_lock_irq(lock); + __raw_write_lock_irq(lock); } -EXPORT_SYMBOL(_write_lock_irq); +EXPORT_SYMBOL(_raw_write_lock_irq); #endif #ifndef CONFIG_INLINE_WRITE_LOCK_BH -void __lockfunc _write_lock_bh(rwlock_t *lock) +void __lockfunc _raw_write_lock_bh(rwlock_t *lock) { - __write_lock_bh(lock); + __raw_write_lock_bh(lock); } -EXPORT_SYMBOL(_write_lock_bh); +EXPORT_SYMBOL(_raw_write_lock_bh); #endif #ifndef CONFIG_INLINE_WRITE_UNLOCK -void __lockfunc _write_unlock(rwlock_t *lock) +void __lockfunc _raw_write_unlock(rwlock_t *lock) { - __write_unlock(lock); + __raw_write_unlock(lock); } -EXPORT_SYMBOL(_write_unlock); +EXPORT_SYMBOL(_raw_write_unlock); #endif #ifndef CONFIG_INLINE_WRITE_UNLOCK_IRQRESTORE -void __lockfunc _write_unlock_irqrestore(rwlock_t *lock, unsigned long flags) +void __lockfunc _raw_write_unlock_irqrestore(rwlock_t *lock, unsigned long flags) { - __write_unlock_irqrestore(lock, flags); + __raw_write_unlock_irqrestore(lock, flags); } -EXPORT_SYMBOL(_write_unlock_irqrestore); +EXPORT_SYMBOL(_raw_write_unlock_irqrestore); #endif #ifndef CONFIG_INLINE_WRITE_UNLOCK_IRQ -void __lockfunc _write_unlock_irq(rwlock_t *lock) +void __lockfunc _raw_write_unlock_irq(rwlock_t *lock) { - __write_unlock_irq(lock); + __raw_write_unlock_irq(lock); } -EXPORT_SYMBOL(_write_unlock_irq); +EXPORT_SYMBOL(_raw_write_unlock_irq); #endif #ifndef CONFIG_INLINE_WRITE_UNLOCK_BH -void __lockfunc _write_unlock_bh(rwlock_t *lock) +void __lockfunc _raw_write_unlock_bh(rwlock_t *lock) { - __write_unlock_bh(lock); + __raw_write_unlock_bh(lock); } -EXPORT_SYMBOL(_write_unlock_bh); +EXPORT_SYMBOL(_raw_write_unlock_bh); #endif #ifdef CONFIG_DEBUG_LOCK_ALLOC -void __lockfunc _spin_lock_nested(raw_spinlock_t *lock, int subclass) +void __lockfunc _raw_spin_lock_nested(raw_spinlock_t *lock, int subclass) { preempt_disable(); spin_acquire(&lock->dep_map, subclass, 0, _RET_IP_); LOCK_CONTENDED(lock, do_raw_spin_trylock, do_raw_spin_lock); } -EXPORT_SYMBOL(_spin_lock_nested); +EXPORT_SYMBOL(_raw_spin_lock_nested); -unsigned long __lockfunc _spin_lock_irqsave_nested(raw_spinlock_t *lock, +unsigned long __lockfunc _raw_spin_lock_irqsave_nested(raw_spinlock_t *lock, int subclass) { unsigned long flags; @@ -361,16 +361,16 @@ unsigned long __lockfunc _spin_lock_irqsave_nested(raw_spinlock_t *lock, do_raw_spin_lock_flags, &flags); return flags; } -EXPORT_SYMBOL(_spin_lock_irqsave_nested); +EXPORT_SYMBOL(_raw_spin_lock_irqsave_nested); -void __lockfunc _spin_lock_nest_lock(raw_spinlock_t *lock, +void __lockfunc _raw_spin_lock_nest_lock(raw_spinlock_t *lock, struct lockdep_map *nest_lock) { preempt_disable(); spin_acquire_nest(&lock->dep_map, 0, 0, nest_lock, _RET_IP_); LOCK_CONTENDED(lock, do_raw_spin_trylock, do_raw_spin_lock); } -EXPORT_SYMBOL(_spin_lock_nest_lock); +EXPORT_SYMBOL(_raw_spin_lock_nest_lock); #endif -- cgit v1.2.3-71-gd317 From a26724591edba5acc528d41f3906a972590e8f54 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 17 Nov 2009 14:46:14 +0100 Subject: plist: Make plist debugging raw_spinlock aware plists are used with spinlocks and raw_spinlocks. Change the plist debugging to handle both types. Signed-off-by: Thomas Gleixner Acked-by: Peter Zijlstra Acked-by: Ingo Molnar --- include/linux/plist.h | 43 +++++++++++++++++++++++++++++++++++++------ kernel/futex.c | 6 +++--- lib/plist.c | 8 +++++--- 3 files changed, 45 insertions(+), 12 deletions(-) (limited to 'kernel') diff --git a/include/linux/plist.h b/include/linux/plist.h index 45926d77d6ac..8227f717c70f 100644 --- a/include/linux/plist.h +++ b/include/linux/plist.h @@ -81,7 +81,8 @@ struct plist_head { struct list_head prio_list; struct list_head node_list; #ifdef CONFIG_DEBUG_PI_LIST - spinlock_t *lock; + raw_spinlock_t *rawlock; + spinlock_t *spinlock; #endif }; @@ -91,9 +92,11 @@ struct plist_node { }; #ifdef CONFIG_DEBUG_PI_LIST -# define PLIST_HEAD_LOCK_INIT(_lock) .lock = _lock +# define PLIST_HEAD_LOCK_INIT(_lock) .spinlock = _lock +# define PLIST_HEAD_LOCK_INIT_RAW(_lock) .rawlock = _lock #else # define PLIST_HEAD_LOCK_INIT(_lock) +# define PLIST_HEAD_LOCK_INIT_RAW(_lock) #endif #define _PLIST_HEAD_INIT(head) \ @@ -107,10 +110,21 @@ struct plist_node { */ #define PLIST_HEAD_INIT(head, _lock) \ { \ - _PLIST_HEAD_INIT(head), \ + _PLIST_HEAD_INIT(head), \ PLIST_HEAD_LOCK_INIT(&(_lock)) \ } +/** + * PLIST_HEAD_INIT_RAW - static struct plist_head initializer + * @head: struct plist_head variable name + * @_lock: lock to initialize for this list + */ +#define PLIST_HEAD_INIT_RAW(head, _lock) \ +{ \ + _PLIST_HEAD_INIT(head), \ + PLIST_HEAD_LOCK_INIT_RAW(&(_lock)) \ +} + /** * PLIST_NODE_INIT - static struct plist_node initializer * @node: struct plist_node variable name @@ -119,13 +133,13 @@ struct plist_node { #define PLIST_NODE_INIT(node, __prio) \ { \ .prio = (__prio), \ - .plist = { _PLIST_HEAD_INIT((node).plist) }, \ + .plist = { _PLIST_HEAD_INIT((node).plist) }, \ } /** * plist_head_init - dynamic struct plist_head initializer * @head: &struct plist_head pointer - * @lock: list spinlock, remembered for debugging + * @lock: spinlock protecting the list (debugging) */ static inline void plist_head_init(struct plist_head *head, spinlock_t *lock) @@ -133,7 +147,24 @@ plist_head_init(struct plist_head *head, spinlock_t *lock) INIT_LIST_HEAD(&head->prio_list); INIT_LIST_HEAD(&head->node_list); #ifdef CONFIG_DEBUG_PI_LIST - head->lock = lock; + head->spinlock = lock; + head->rawlock = NULL; +#endif +} + +/** + * plist_head_init_raw - dynamic struct plist_head initializer + * @head: &struct plist_head pointer + * @lock: raw_spinlock protecting the list (debugging) + */ +static inline void +plist_head_init_raw(struct plist_head *head, raw_spinlock_t *lock) +{ + INIT_LIST_HEAD(&head->prio_list); + INIT_LIST_HEAD(&head->node_list); +#ifdef CONFIG_DEBUG_PI_LIST + head->rawlock = lock; + head->spinlock = NULL; #endif } diff --git a/kernel/futex.c b/kernel/futex.c index d73ef1f3e55d..6af474df17bb 100644 --- a/kernel/futex.c +++ b/kernel/futex.c @@ -1010,7 +1010,7 @@ void requeue_futex(struct futex_q *q, struct futex_hash_bucket *hb1, plist_add(&q->list, &hb2->chain); q->lock_ptr = &hb2->lock; #ifdef CONFIG_DEBUG_PI_LIST - q->list.plist.lock = &hb2->lock; + q->list.plist.spinlock = &hb2->lock; #endif } get_futex_key_refs(key2); @@ -1046,7 +1046,7 @@ void requeue_pi_wake_futex(struct futex_q *q, union futex_key *key, q->lock_ptr = &hb->lock; #ifdef CONFIG_DEBUG_PI_LIST - q->list.plist.lock = &hb->lock; + q->list.plist.spinlock = &hb->lock; #endif wake_up_state(q->task, TASK_NORMAL); @@ -1394,7 +1394,7 @@ static inline void queue_me(struct futex_q *q, struct futex_hash_bucket *hb) plist_node_init(&q->list, prio); #ifdef CONFIG_DEBUG_PI_LIST - q->list.plist.lock = &hb->lock; + q->list.plist.spinlock = &hb->lock; #endif plist_add(&q->list, &hb->chain); q->task = current; diff --git a/lib/plist.c b/lib/plist.c index d6c64a824e1d..1471988d9190 100644 --- a/lib/plist.c +++ b/lib/plist.c @@ -54,9 +54,11 @@ static void plist_check_list(struct list_head *top) static void plist_check_head(struct plist_head *head) { - WARN_ON(!head->lock); - if (head->lock) - WARN_ON_SMP(!spin_is_locked(head->lock)); + WARN_ON(!head->rawlock && !head->spinlock); + if (head->rawlock) + WARN_ON_SMP(!raw_spin_is_locked(head->rawlock)); + if (head->spinlock) + WARN_ON_SMP(!spin_is_locked(head->spinlock)); plist_check_list(&head->prio_list); plist_check_list(&head->node_list); } -- cgit v1.2.3-71-gd317 From 05fa785cf80c9b7c0254c3056037147aed3ea16b Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 17 Nov 2009 14:28:38 +0100 Subject: sched: Convert rq->lock to raw_spinlock Convert locks which cannot be sleeping locks in preempt-rt to raw_spinlocks. Signed-off-by: Thomas Gleixner Acked-by: Peter Zijlstra Acked-by: Ingo Molnar --- kernel/sched.c | 183 ++++++++++++++++++++++++------------------------ kernel/sched_debug.c | 4 +- kernel/sched_fair.c | 4 +- kernel/sched_idletask.c | 4 +- kernel/sched_rt.c | 14 ++-- 5 files changed, 106 insertions(+), 103 deletions(-) (limited to 'kernel') diff --git a/kernel/sched.c b/kernel/sched.c index 91c65dd91435..3dbe6178ebfd 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -525,7 +525,7 @@ static struct root_domain def_root_domain; */ struct rq { /* runqueue lock: */ - spinlock_t lock; + raw_spinlock_t lock; /* * nr_running and cpu_load should be in the same cacheline because @@ -685,7 +685,7 @@ inline void update_rq_clock(struct rq *rq) */ int runqueue_is_locked(int cpu) { - return spin_is_locked(&cpu_rq(cpu)->lock); + return raw_spin_is_locked(&cpu_rq(cpu)->lock); } /* @@ -884,7 +884,7 @@ static inline void finish_lock_switch(struct rq *rq, struct task_struct *prev) { #ifdef CONFIG_DEBUG_SPINLOCK /* this is a valid case when another task releases the spinlock */ - rq->lock.rlock.owner = current; + rq->lock.owner = current; #endif /* * If we are tracking spinlock dependencies then we have to @@ -893,7 +893,7 @@ static inline void finish_lock_switch(struct rq *rq, struct task_struct *prev) */ spin_acquire(&rq->lock.dep_map, 0, 0, _THIS_IP_); - spin_unlock_irq(&rq->lock); + raw_spin_unlock_irq(&rq->lock); } #else /* __ARCH_WANT_UNLOCKED_CTXSW */ @@ -917,9 +917,9 @@ static inline void prepare_lock_switch(struct rq *rq, struct task_struct *next) next->oncpu = 1; #endif #ifdef __ARCH_WANT_INTERRUPTS_ON_CTXSW - spin_unlock_irq(&rq->lock); + raw_spin_unlock_irq(&rq->lock); #else - spin_unlock(&rq->lock); + raw_spin_unlock(&rq->lock); #endif } @@ -949,10 +949,10 @@ static inline struct rq *__task_rq_lock(struct task_struct *p) { for (;;) { struct rq *rq = task_rq(p); - spin_lock(&rq->lock); + raw_spin_lock(&rq->lock); if (likely(rq == task_rq(p))) return rq; - spin_unlock(&rq->lock); + raw_spin_unlock(&rq->lock); } } @@ -969,10 +969,10 @@ static struct rq *task_rq_lock(struct task_struct *p, unsigned long *flags) for (;;) { local_irq_save(*flags); rq = task_rq(p); - spin_lock(&rq->lock); + raw_spin_lock(&rq->lock); if (likely(rq == task_rq(p))) return rq; - spin_unlock_irqrestore(&rq->lock, *flags); + raw_spin_unlock_irqrestore(&rq->lock, *flags); } } @@ -981,19 +981,19 @@ void task_rq_unlock_wait(struct task_struct *p) struct rq *rq = task_rq(p); smp_mb(); /* spin-unlock-wait is not a full memory barrier */ - spin_unlock_wait(&rq->lock); + raw_spin_unlock_wait(&rq->lock); } static void __task_rq_unlock(struct rq *rq) __releases(rq->lock) { - spin_unlock(&rq->lock); + raw_spin_unlock(&rq->lock); } static inline void task_rq_unlock(struct rq *rq, unsigned long *flags) __releases(rq->lock) { - spin_unlock_irqrestore(&rq->lock, *flags); + raw_spin_unlock_irqrestore(&rq->lock, *flags); } /* @@ -1006,7 +1006,7 @@ static struct rq *this_rq_lock(void) local_irq_disable(); rq = this_rq(); - spin_lock(&rq->lock); + raw_spin_lock(&rq->lock); return rq; } @@ -1053,10 +1053,10 @@ static enum hrtimer_restart hrtick(struct hrtimer *timer) WARN_ON_ONCE(cpu_of(rq) != smp_processor_id()); - spin_lock(&rq->lock); + raw_spin_lock(&rq->lock); update_rq_clock(rq); rq->curr->sched_class->task_tick(rq, rq->curr, 1); - spin_unlock(&rq->lock); + raw_spin_unlock(&rq->lock); return HRTIMER_NORESTART; } @@ -1069,10 +1069,10 @@ static void __hrtick_start(void *arg) { struct rq *rq = arg; - spin_lock(&rq->lock); + raw_spin_lock(&rq->lock); hrtimer_restart(&rq->hrtick_timer); rq->hrtick_csd_pending = 0; - spin_unlock(&rq->lock); + raw_spin_unlock(&rq->lock); } /* @@ -1179,7 +1179,7 @@ static void resched_task(struct task_struct *p) { int cpu; - assert_spin_locked(&task_rq(p)->lock); + assert_raw_spin_locked(&task_rq(p)->lock); if (test_tsk_need_resched(p)) return; @@ -1201,10 +1201,10 @@ static void resched_cpu(int cpu) struct rq *rq = cpu_rq(cpu); unsigned long flags; - if (!spin_trylock_irqsave(&rq->lock, flags)) + if (!raw_spin_trylock_irqsave(&rq->lock, flags)) return; resched_task(cpu_curr(cpu)); - spin_unlock_irqrestore(&rq->lock, flags); + raw_spin_unlock_irqrestore(&rq->lock, flags); } #ifdef CONFIG_NO_HZ @@ -1273,7 +1273,7 @@ static void sched_rt_avg_update(struct rq *rq, u64 rt_delta) #else /* !CONFIG_SMP */ static void resched_task(struct task_struct *p) { - assert_spin_locked(&task_rq(p)->lock); + assert_raw_spin_locked(&task_rq(p)->lock); set_tsk_need_resched(p); } @@ -1600,11 +1600,11 @@ static void update_group_shares_cpu(struct task_group *tg, int cpu, struct rq *rq = cpu_rq(cpu); unsigned long flags; - spin_lock_irqsave(&rq->lock, flags); + raw_spin_lock_irqsave(&rq->lock, flags); tg->cfs_rq[cpu]->rq_weight = boost ? 0 : rq_weight; tg->cfs_rq[cpu]->shares = boost ? 0 : shares; __set_se_shares(tg->se[cpu], shares); - spin_unlock_irqrestore(&rq->lock, flags); + raw_spin_unlock_irqrestore(&rq->lock, flags); } } @@ -1706,9 +1706,9 @@ static void update_shares_locked(struct rq *rq, struct sched_domain *sd) if (root_task_group_empty()) return; - spin_unlock(&rq->lock); + raw_spin_unlock(&rq->lock); update_shares(sd); - spin_lock(&rq->lock); + raw_spin_lock(&rq->lock); } static void update_h_load(long cpu) @@ -1748,7 +1748,7 @@ static inline int _double_lock_balance(struct rq *this_rq, struct rq *busiest) __acquires(busiest->lock) __acquires(this_rq->lock) { - spin_unlock(&this_rq->lock); + raw_spin_unlock(&this_rq->lock); double_rq_lock(this_rq, busiest); return 1; @@ -1769,14 +1769,16 @@ static int _double_lock_balance(struct rq *this_rq, struct rq *busiest) { int ret = 0; - if (unlikely(!spin_trylock(&busiest->lock))) { + if (unlikely(!raw_spin_trylock(&busiest->lock))) { if (busiest < this_rq) { - spin_unlock(&this_rq->lock); - spin_lock(&busiest->lock); - spin_lock_nested(&this_rq->lock, SINGLE_DEPTH_NESTING); + raw_spin_unlock(&this_rq->lock); + raw_spin_lock(&busiest->lock); + raw_spin_lock_nested(&this_rq->lock, + SINGLE_DEPTH_NESTING); ret = 1; } else - spin_lock_nested(&busiest->lock, SINGLE_DEPTH_NESTING); + raw_spin_lock_nested(&busiest->lock, + SINGLE_DEPTH_NESTING); } return ret; } @@ -1790,7 +1792,7 @@ static int double_lock_balance(struct rq *this_rq, struct rq *busiest) { if (unlikely(!irqs_disabled())) { /* printk() doesn't work good under rq->lock */ - spin_unlock(&this_rq->lock); + raw_spin_unlock(&this_rq->lock); BUG_ON(1); } @@ -1800,7 +1802,7 @@ static int double_lock_balance(struct rq *this_rq, struct rq *busiest) static inline void double_unlock_balance(struct rq *this_rq, struct rq *busiest) __releases(busiest->lock) { - spin_unlock(&busiest->lock); + raw_spin_unlock(&busiest->lock); lock_set_subclass(&this_rq->lock.dep_map, 0, _RET_IP_); } #endif @@ -2023,13 +2025,13 @@ void kthread_bind(struct task_struct *p, unsigned int cpu) return; } - spin_lock_irqsave(&rq->lock, flags); + raw_spin_lock_irqsave(&rq->lock, flags); update_rq_clock(rq); set_task_cpu(p, cpu); p->cpus_allowed = cpumask_of_cpu(cpu); p->rt.nr_cpus_allowed = 1; p->flags |= PF_THREAD_BOUND; - spin_unlock_irqrestore(&rq->lock, flags); + raw_spin_unlock_irqrestore(&rq->lock, flags); } EXPORT_SYMBOL(kthread_bind); @@ -2781,10 +2783,10 @@ static inline void post_schedule(struct rq *rq) if (rq->post_schedule) { unsigned long flags; - spin_lock_irqsave(&rq->lock, flags); + raw_spin_lock_irqsave(&rq->lock, flags); if (rq->curr->sched_class->post_schedule) rq->curr->sched_class->post_schedule(rq); - spin_unlock_irqrestore(&rq->lock, flags); + raw_spin_unlock_irqrestore(&rq->lock, flags); rq->post_schedule = 0; } @@ -3066,15 +3068,15 @@ static void double_rq_lock(struct rq *rq1, struct rq *rq2) { BUG_ON(!irqs_disabled()); if (rq1 == rq2) { - spin_lock(&rq1->lock); + raw_spin_lock(&rq1->lock); __acquire(rq2->lock); /* Fake it out ;) */ } else { if (rq1 < rq2) { - spin_lock(&rq1->lock); - spin_lock_nested(&rq2->lock, SINGLE_DEPTH_NESTING); + raw_spin_lock(&rq1->lock); + raw_spin_lock_nested(&rq2->lock, SINGLE_DEPTH_NESTING); } else { - spin_lock(&rq2->lock); - spin_lock_nested(&rq1->lock, SINGLE_DEPTH_NESTING); + raw_spin_lock(&rq2->lock); + raw_spin_lock_nested(&rq1->lock, SINGLE_DEPTH_NESTING); } } update_rq_clock(rq1); @@ -3091,9 +3093,9 @@ static void double_rq_unlock(struct rq *rq1, struct rq *rq2) __releases(rq1->lock) __releases(rq2->lock) { - spin_unlock(&rq1->lock); + raw_spin_unlock(&rq1->lock); if (rq1 != rq2) - spin_unlock(&rq2->lock); + raw_spin_unlock(&rq2->lock); else __release(rq2->lock); } @@ -4186,14 +4188,15 @@ redo: if (unlikely(sd->nr_balance_failed > sd->cache_nice_tries+2)) { - spin_lock_irqsave(&busiest->lock, flags); + raw_spin_lock_irqsave(&busiest->lock, flags); /* don't kick the migration_thread, if the curr * task on busiest cpu can't be moved to this_cpu */ if (!cpumask_test_cpu(this_cpu, &busiest->curr->cpus_allowed)) { - spin_unlock_irqrestore(&busiest->lock, flags); + raw_spin_unlock_irqrestore(&busiest->lock, + flags); all_pinned = 1; goto out_one_pinned; } @@ -4203,7 +4206,7 @@ redo: busiest->push_cpu = this_cpu; active_balance = 1; } - spin_unlock_irqrestore(&busiest->lock, flags); + raw_spin_unlock_irqrestore(&busiest->lock, flags); if (active_balance) wake_up_process(busiest->migration_thread); @@ -4385,10 +4388,10 @@ redo: /* * Should not call ttwu while holding a rq->lock */ - spin_unlock(&this_rq->lock); + raw_spin_unlock(&this_rq->lock); if (active_balance) wake_up_process(busiest->migration_thread); - spin_lock(&this_rq->lock); + raw_spin_lock(&this_rq->lock); } else sd->nr_balance_failed = 0; @@ -5257,11 +5260,11 @@ void scheduler_tick(void) sched_clock_tick(); - spin_lock(&rq->lock); + raw_spin_lock(&rq->lock); update_rq_clock(rq); update_cpu_load(rq); curr->sched_class->task_tick(rq, curr, 0); - spin_unlock(&rq->lock); + raw_spin_unlock(&rq->lock); perf_event_task_tick(curr, cpu); @@ -5455,7 +5458,7 @@ need_resched_nonpreemptible: if (sched_feat(HRTICK)) hrtick_clear(rq); - spin_lock_irq(&rq->lock); + raw_spin_lock_irq(&rq->lock); update_rq_clock(rq); clear_tsk_need_resched(prev); @@ -5491,7 +5494,7 @@ need_resched_nonpreemptible: cpu = smp_processor_id(); rq = cpu_rq(cpu); } else - spin_unlock_irq(&rq->lock); + raw_spin_unlock_irq(&rq->lock); post_schedule(rq); @@ -6980,7 +6983,7 @@ void __cpuinit init_idle(struct task_struct *idle, int cpu) struct rq *rq = cpu_rq(cpu); unsigned long flags; - spin_lock_irqsave(&rq->lock, flags); + raw_spin_lock_irqsave(&rq->lock, flags); __sched_fork(idle); idle->se.exec_start = sched_clock(); @@ -6992,7 +6995,7 @@ void __cpuinit init_idle(struct task_struct *idle, int cpu) #if defined(CONFIG_SMP) && defined(__ARCH_WANT_UNLOCKED_CTXSW) idle->oncpu = 1; #endif - spin_unlock_irqrestore(&rq->lock, flags); + raw_spin_unlock_irqrestore(&rq->lock, flags); /* Set the preempt count _outside_ the spinlocks! */ #if defined(CONFIG_PREEMPT) @@ -7209,10 +7212,10 @@ static int migration_thread(void *data) struct migration_req *req; struct list_head *head; - spin_lock_irq(&rq->lock); + raw_spin_lock_irq(&rq->lock); if (cpu_is_offline(cpu)) { - spin_unlock_irq(&rq->lock); + raw_spin_unlock_irq(&rq->lock); break; } @@ -7224,7 +7227,7 @@ static int migration_thread(void *data) head = &rq->migration_queue; if (list_empty(head)) { - spin_unlock_irq(&rq->lock); + raw_spin_unlock_irq(&rq->lock); schedule(); set_current_state(TASK_INTERRUPTIBLE); continue; @@ -7233,14 +7236,14 @@ static int migration_thread(void *data) list_del_init(head->next); if (req->task != NULL) { - spin_unlock(&rq->lock); + raw_spin_unlock(&rq->lock); __migrate_task(req->task, cpu, req->dest_cpu); } else if (likely(cpu == (badcpu = smp_processor_id()))) { req->dest_cpu = RCU_MIGRATION_GOT_QS; - spin_unlock(&rq->lock); + raw_spin_unlock(&rq->lock); } else { req->dest_cpu = RCU_MIGRATION_MUST_SYNC; - spin_unlock(&rq->lock); + raw_spin_unlock(&rq->lock); WARN_ONCE(1, "migration_thread() on CPU %d, expected %d\n", badcpu, cpu); } local_irq_enable(); @@ -7363,14 +7366,14 @@ void sched_idle_next(void) * Strictly not necessary since rest of the CPUs are stopped by now * and interrupts disabled on the current cpu. */ - spin_lock_irqsave(&rq->lock, flags); + raw_spin_lock_irqsave(&rq->lock, flags); __setscheduler(rq, p, SCHED_FIFO, MAX_RT_PRIO-1); update_rq_clock(rq); activate_task(rq, p, 0); - spin_unlock_irqrestore(&rq->lock, flags); + raw_spin_unlock_irqrestore(&rq->lock, flags); } /* @@ -7406,9 +7409,9 @@ static void migrate_dead(unsigned int dead_cpu, struct task_struct *p) * that's OK. No task can be added to this CPU, so iteration is * fine. */ - spin_unlock_irq(&rq->lock); + raw_spin_unlock_irq(&rq->lock); move_task_off_dead_cpu(dead_cpu, p); - spin_lock_irq(&rq->lock); + raw_spin_lock_irq(&rq->lock); put_task_struct(p); } @@ -7674,13 +7677,13 @@ migration_call(struct notifier_block *nfb, unsigned long action, void *hcpu) /* Update our root-domain */ rq = cpu_rq(cpu); - spin_lock_irqsave(&rq->lock, flags); + raw_spin_lock_irqsave(&rq->lock, flags); if (rq->rd) { BUG_ON(!cpumask_test_cpu(cpu, rq->rd->span)); set_rq_online(rq); } - spin_unlock_irqrestore(&rq->lock, flags); + raw_spin_unlock_irqrestore(&rq->lock, flags); break; #ifdef CONFIG_HOTPLUG_CPU @@ -7705,13 +7708,13 @@ migration_call(struct notifier_block *nfb, unsigned long action, void *hcpu) put_task_struct(rq->migration_thread); rq->migration_thread = NULL; /* Idle task back to normal (off runqueue, low prio) */ - spin_lock_irq(&rq->lock); + raw_spin_lock_irq(&rq->lock); update_rq_clock(rq); deactivate_task(rq, rq->idle, 0); __setscheduler(rq, rq->idle, SCHED_NORMAL, 0); rq->idle->sched_class = &idle_sched_class; migrate_dead_tasks(cpu); - spin_unlock_irq(&rq->lock); + raw_spin_unlock_irq(&rq->lock); cpuset_unlock(); migrate_nr_uninterruptible(rq); BUG_ON(rq->nr_running != 0); @@ -7721,30 +7724,30 @@ migration_call(struct notifier_block *nfb, unsigned long action, void *hcpu) * they didn't take sched_hotcpu_mutex. Just wake up * the requestors. */ - spin_lock_irq(&rq->lock); + raw_spin_lock_irq(&rq->lock); while (!list_empty(&rq->migration_queue)) { struct migration_req *req; req = list_entry(rq->migration_queue.next, struct migration_req, list); list_del_init(&req->list); - spin_unlock_irq(&rq->lock); + raw_spin_unlock_irq(&rq->lock); complete(&req->done); - spin_lock_irq(&rq->lock); + raw_spin_lock_irq(&rq->lock); } - spin_unlock_irq(&rq->lock); + raw_spin_unlock_irq(&rq->lock); break; case CPU_DYING: case CPU_DYING_FROZEN: /* Update our root-domain */ rq = cpu_rq(cpu); - spin_lock_irqsave(&rq->lock, flags); + raw_spin_lock_irqsave(&rq->lock, flags); if (rq->rd) { BUG_ON(!cpumask_test_cpu(cpu, rq->rd->span)); set_rq_offline(rq); } - spin_unlock_irqrestore(&rq->lock, flags); + raw_spin_unlock_irqrestore(&rq->lock, flags); break; #endif } @@ -7974,7 +7977,7 @@ static void rq_attach_root(struct rq *rq, struct root_domain *rd) struct root_domain *old_rd = NULL; unsigned long flags; - spin_lock_irqsave(&rq->lock, flags); + raw_spin_lock_irqsave(&rq->lock, flags); if (rq->rd) { old_rd = rq->rd; @@ -8000,7 +8003,7 @@ static void rq_attach_root(struct rq *rq, struct root_domain *rd) if (cpumask_test_cpu(rq->cpu, cpu_active_mask)) set_rq_online(rq); - spin_unlock_irqrestore(&rq->lock, flags); + raw_spin_unlock_irqrestore(&rq->lock, flags); if (old_rd) free_rootdomain(old_rd); @@ -9357,7 +9360,7 @@ static void init_rt_rq(struct rt_rq *rt_rq, struct rq *rq) #ifdef CONFIG_SMP rt_rq->rt_nr_migratory = 0; rt_rq->overloaded = 0; - plist_head_init(&rt_rq->pushable_tasks, &rq->lock); + plist_head_init_raw(&rt_rq->pushable_tasks, &rq->lock); #endif rt_rq->rt_time = 0; @@ -9523,7 +9526,7 @@ void __init sched_init(void) struct rq *rq; rq = cpu_rq(i); - spin_lock_init(&rq->lock); + raw_spin_lock_init(&rq->lock); rq->nr_running = 0; rq->calc_load_active = 0; rq->calc_load_update = jiffies + LOAD_FREQ; @@ -10115,9 +10118,9 @@ static void set_se_shares(struct sched_entity *se, unsigned long shares) struct rq *rq = cfs_rq->rq; unsigned long flags; - spin_lock_irqsave(&rq->lock, flags); + raw_spin_lock_irqsave(&rq->lock, flags); __set_se_shares(se, shares); - spin_unlock_irqrestore(&rq->lock, flags); + raw_spin_unlock_irqrestore(&rq->lock, flags); } static DEFINE_MUTEX(shares_mutex); @@ -10717,9 +10720,9 @@ static u64 cpuacct_cpuusage_read(struct cpuacct *ca, int cpu) /* * Take rq->lock to make 64-bit read safe on 32-bit platforms. */ - spin_lock_irq(&cpu_rq(cpu)->lock); + raw_spin_lock_irq(&cpu_rq(cpu)->lock); data = *cpuusage; - spin_unlock_irq(&cpu_rq(cpu)->lock); + raw_spin_unlock_irq(&cpu_rq(cpu)->lock); #else data = *cpuusage; #endif @@ -10735,9 +10738,9 @@ static void cpuacct_cpuusage_write(struct cpuacct *ca, int cpu, u64 val) /* * Take rq->lock to make 64-bit write safe on 32-bit platforms. */ - spin_lock_irq(&cpu_rq(cpu)->lock); + raw_spin_lock_irq(&cpu_rq(cpu)->lock); *cpuusage = val; - spin_unlock_irq(&cpu_rq(cpu)->lock); + raw_spin_unlock_irq(&cpu_rq(cpu)->lock); #else *cpuusage = val; #endif @@ -10971,9 +10974,9 @@ void synchronize_sched_expedited(void) init_completion(&req->done); req->task = NULL; req->dest_cpu = RCU_MIGRATION_NEED_QS; - spin_lock_irqsave(&rq->lock, flags); + raw_spin_lock_irqsave(&rq->lock, flags); list_add(&req->list, &rq->migration_queue); - spin_unlock_irqrestore(&rq->lock, flags); + raw_spin_unlock_irqrestore(&rq->lock, flags); wake_up_process(rq->migration_thread); } for_each_online_cpu(cpu) { @@ -10981,11 +10984,11 @@ void synchronize_sched_expedited(void) req = &per_cpu(rcu_migration_req, cpu); rq = cpu_rq(cpu); wait_for_completion(&req->done); - spin_lock_irqsave(&rq->lock, flags); + raw_spin_lock_irqsave(&rq->lock, flags); if (unlikely(req->dest_cpu == RCU_MIGRATION_MUST_SYNC)) need_full_sync = 1; req->dest_cpu = RCU_MIGRATION_IDLE; - spin_unlock_irqrestore(&rq->lock, flags); + raw_spin_unlock_irqrestore(&rq->lock, flags); } rcu_expedited_state = RCU_EXPEDITED_STATE_IDLE; synchronize_sched_expedited_count++; diff --git a/kernel/sched_debug.c b/kernel/sched_debug.c index 5ae24fc65d75..67f95aada4b9 100644 --- a/kernel/sched_debug.c +++ b/kernel/sched_debug.c @@ -184,7 +184,7 @@ void print_cfs_rq(struct seq_file *m, int cpu, struct cfs_rq *cfs_rq) SEQ_printf(m, " .%-30s: %Ld.%06ld\n", "exec_clock", SPLIT_NS(cfs_rq->exec_clock)); - spin_lock_irqsave(&rq->lock, flags); + raw_spin_lock_irqsave(&rq->lock, flags); if (cfs_rq->rb_leftmost) MIN_vruntime = (__pick_next_entity(cfs_rq))->vruntime; last = __pick_last_entity(cfs_rq); @@ -192,7 +192,7 @@ void print_cfs_rq(struct seq_file *m, int cpu, struct cfs_rq *cfs_rq) max_vruntime = last->vruntime; min_vruntime = cfs_rq->min_vruntime; rq0_min_vruntime = cpu_rq(0)->cfs.min_vruntime; - spin_unlock_irqrestore(&rq->lock, flags); + raw_spin_unlock_irqrestore(&rq->lock, flags); SEQ_printf(m, " .%-30s: %Ld.%06ld\n", "MIN_vruntime", SPLIT_NS(MIN_vruntime)); SEQ_printf(m, " .%-30s: %Ld.%06ld\n", "min_vruntime", diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c index 804a411838f1..5bedf6e3ebf3 100644 --- a/kernel/sched_fair.c +++ b/kernel/sched_fair.c @@ -1955,7 +1955,7 @@ static void task_fork_fair(struct task_struct *p) struct rq *rq = this_rq(); unsigned long flags; - spin_lock_irqsave(&rq->lock, flags); + raw_spin_lock_irqsave(&rq->lock, flags); if (unlikely(task_cpu(p) != this_cpu)) __set_task_cpu(p, this_cpu); @@ -1975,7 +1975,7 @@ static void task_fork_fair(struct task_struct *p) resched_task(rq->curr); } - spin_unlock_irqrestore(&rq->lock, flags); + raw_spin_unlock_irqrestore(&rq->lock, flags); } /* diff --git a/kernel/sched_idletask.c b/kernel/sched_idletask.c index 33d5384a73a8..5f93b570d383 100644 --- a/kernel/sched_idletask.c +++ b/kernel/sched_idletask.c @@ -34,10 +34,10 @@ static struct task_struct *pick_next_task_idle(struct rq *rq) static void dequeue_task_idle(struct rq *rq, struct task_struct *p, int sleep) { - spin_unlock_irq(&rq->lock); + raw_spin_unlock_irq(&rq->lock); printk(KERN_ERR "bad: scheduling from the idle thread!\n"); dump_stack(); - spin_lock_irq(&rq->lock); + raw_spin_lock_irq(&rq->lock); } static void put_prev_task_idle(struct rq *rq, struct task_struct *prev) diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c index aecbd9c6b20c..a8325a7ff94c 100644 --- a/kernel/sched_rt.c +++ b/kernel/sched_rt.c @@ -454,9 +454,9 @@ static void disable_runtime(struct rq *rq) { unsigned long flags; - spin_lock_irqsave(&rq->lock, flags); + raw_spin_lock_irqsave(&rq->lock, flags); __disable_runtime(rq); - spin_unlock_irqrestore(&rq->lock, flags); + raw_spin_unlock_irqrestore(&rq->lock, flags); } static void __enable_runtime(struct rq *rq) @@ -486,9 +486,9 @@ static void enable_runtime(struct rq *rq) { unsigned long flags; - spin_lock_irqsave(&rq->lock, flags); + raw_spin_lock_irqsave(&rq->lock, flags); __enable_runtime(rq); - spin_unlock_irqrestore(&rq->lock, flags); + raw_spin_unlock_irqrestore(&rq->lock, flags); } static int balance_runtime(struct rt_rq *rt_rq) @@ -524,7 +524,7 @@ static int do_sched_rt_period_timer(struct rt_bandwidth *rt_b, int overrun) struct rt_rq *rt_rq = sched_rt_period_rt_rq(rt_b, i); struct rq *rq = rq_of_rt_rq(rt_rq); - spin_lock(&rq->lock); + raw_spin_lock(&rq->lock); if (rt_rq->rt_time) { u64 runtime; @@ -545,7 +545,7 @@ static int do_sched_rt_period_timer(struct rt_bandwidth *rt_b, int overrun) if (enqueue) sched_rt_rq_enqueue(rt_rq); - spin_unlock(&rq->lock); + raw_spin_unlock(&rq->lock); } return idle; @@ -1246,7 +1246,7 @@ static struct rq *find_lock_lowest_rq(struct task_struct *task, struct rq *rq) task_running(rq, task) || !task->se.on_rq)) { - spin_unlock(&lowest_rq->lock); + raw_spin_unlock(&lowest_rq->lock); lowest_rq = NULL; break; } -- cgit v1.2.3-71-gd317 From 0986b11b12ad2baed338385f79fd0dec58a23fc6 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 17 Nov 2009 15:32:06 +0100 Subject: sched: Convert rt_runtime_lock to raw_spinlock Convert locks which cannot be sleeping locks in preempt-rt to raw_spinlocks. Signed-off-by: Thomas Gleixner Acked-by: Peter Zijlstra Acked-by: Ingo Molnar --- kernel/sched.c | 28 ++++++++++++++-------------- kernel/sched_rt.c | 46 +++++++++++++++++++++++----------------------- 2 files changed, 37 insertions(+), 37 deletions(-) (limited to 'kernel') diff --git a/kernel/sched.c b/kernel/sched.c index 3dbe6178ebfd..01c5016e57f1 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -141,7 +141,7 @@ struct rt_prio_array { struct rt_bandwidth { /* nests inside the rq lock: */ - spinlock_t rt_runtime_lock; + raw_spinlock_t rt_runtime_lock; ktime_t rt_period; u64 rt_runtime; struct hrtimer rt_period_timer; @@ -178,7 +178,7 @@ void init_rt_bandwidth(struct rt_bandwidth *rt_b, u64 period, u64 runtime) rt_b->rt_period = ns_to_ktime(period); rt_b->rt_runtime = runtime; - spin_lock_init(&rt_b->rt_runtime_lock); + raw_spin_lock_init(&rt_b->rt_runtime_lock); hrtimer_init(&rt_b->rt_period_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); @@ -200,7 +200,7 @@ static void start_rt_bandwidth(struct rt_bandwidth *rt_b) if (hrtimer_active(&rt_b->rt_period_timer)) return; - spin_lock(&rt_b->rt_runtime_lock); + raw_spin_lock(&rt_b->rt_runtime_lock); for (;;) { unsigned long delta; ktime_t soft, hard; @@ -217,7 +217,7 @@ static void start_rt_bandwidth(struct rt_bandwidth *rt_b) __hrtimer_start_range_ns(&rt_b->rt_period_timer, soft, delta, HRTIMER_MODE_ABS_PINNED, 0); } - spin_unlock(&rt_b->rt_runtime_lock); + raw_spin_unlock(&rt_b->rt_runtime_lock); } #ifdef CONFIG_RT_GROUP_SCHED @@ -470,7 +470,7 @@ struct rt_rq { u64 rt_time; u64 rt_runtime; /* Nests inside the rq lock: */ - spinlock_t rt_runtime_lock; + raw_spinlock_t rt_runtime_lock; #ifdef CONFIG_RT_GROUP_SCHED unsigned long rt_nr_boosted; @@ -9366,7 +9366,7 @@ static void init_rt_rq(struct rt_rq *rt_rq, struct rq *rq) rt_rq->rt_time = 0; rt_rq->rt_throttled = 0; rt_rq->rt_runtime = 0; - spin_lock_init(&rt_rq->rt_runtime_lock); + raw_spin_lock_init(&rt_rq->rt_runtime_lock); #ifdef CONFIG_RT_GROUP_SCHED rt_rq->rt_nr_boosted = 0; @@ -10305,18 +10305,18 @@ static int tg_set_bandwidth(struct task_group *tg, if (err) goto unlock; - spin_lock_irq(&tg->rt_bandwidth.rt_runtime_lock); + raw_spin_lock_irq(&tg->rt_bandwidth.rt_runtime_lock); tg->rt_bandwidth.rt_period = ns_to_ktime(rt_period); tg->rt_bandwidth.rt_runtime = rt_runtime; for_each_possible_cpu(i) { struct rt_rq *rt_rq = tg->rt_rq[i]; - spin_lock(&rt_rq->rt_runtime_lock); + raw_spin_lock(&rt_rq->rt_runtime_lock); rt_rq->rt_runtime = rt_runtime; - spin_unlock(&rt_rq->rt_runtime_lock); + raw_spin_unlock(&rt_rq->rt_runtime_lock); } - spin_unlock_irq(&tg->rt_bandwidth.rt_runtime_lock); + raw_spin_unlock_irq(&tg->rt_bandwidth.rt_runtime_lock); unlock: read_unlock(&tasklist_lock); mutex_unlock(&rt_constraints_mutex); @@ -10421,15 +10421,15 @@ static int sched_rt_global_constraints(void) if (sysctl_sched_rt_runtime == 0) return -EBUSY; - spin_lock_irqsave(&def_rt_bandwidth.rt_runtime_lock, flags); + raw_spin_lock_irqsave(&def_rt_bandwidth.rt_runtime_lock, flags); for_each_possible_cpu(i) { struct rt_rq *rt_rq = &cpu_rq(i)->rt; - spin_lock(&rt_rq->rt_runtime_lock); + raw_spin_lock(&rt_rq->rt_runtime_lock); rt_rq->rt_runtime = global_rt_runtime(); - spin_unlock(&rt_rq->rt_runtime_lock); + raw_spin_unlock(&rt_rq->rt_runtime_lock); } - spin_unlock_irqrestore(&def_rt_bandwidth.rt_runtime_lock, flags); + raw_spin_unlock_irqrestore(&def_rt_bandwidth.rt_runtime_lock, flags); return 0; } diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c index a8325a7ff94c..d2ea2828164e 100644 --- a/kernel/sched_rt.c +++ b/kernel/sched_rt.c @@ -327,7 +327,7 @@ static int do_balance_runtime(struct rt_rq *rt_rq) weight = cpumask_weight(rd->span); - spin_lock(&rt_b->rt_runtime_lock); + raw_spin_lock(&rt_b->rt_runtime_lock); rt_period = ktime_to_ns(rt_b->rt_period); for_each_cpu(i, rd->span) { struct rt_rq *iter = sched_rt_period_rt_rq(rt_b, i); @@ -336,7 +336,7 @@ static int do_balance_runtime(struct rt_rq *rt_rq) if (iter == rt_rq) continue; - spin_lock(&iter->rt_runtime_lock); + raw_spin_lock(&iter->rt_runtime_lock); /* * Either all rqs have inf runtime and there's nothing to steal * or __disable_runtime() below sets a specific rq to inf to @@ -358,14 +358,14 @@ static int do_balance_runtime(struct rt_rq *rt_rq) rt_rq->rt_runtime += diff; more = 1; if (rt_rq->rt_runtime == rt_period) { - spin_unlock(&iter->rt_runtime_lock); + raw_spin_unlock(&iter->rt_runtime_lock); break; } } next: - spin_unlock(&iter->rt_runtime_lock); + raw_spin_unlock(&iter->rt_runtime_lock); } - spin_unlock(&rt_b->rt_runtime_lock); + raw_spin_unlock(&rt_b->rt_runtime_lock); return more; } @@ -386,8 +386,8 @@ static void __disable_runtime(struct rq *rq) s64 want; int i; - spin_lock(&rt_b->rt_runtime_lock); - spin_lock(&rt_rq->rt_runtime_lock); + raw_spin_lock(&rt_b->rt_runtime_lock); + raw_spin_lock(&rt_rq->rt_runtime_lock); /* * Either we're all inf and nobody needs to borrow, or we're * already disabled and thus have nothing to do, or we have @@ -396,7 +396,7 @@ static void __disable_runtime(struct rq *rq) if (rt_rq->rt_runtime == RUNTIME_INF || rt_rq->rt_runtime == rt_b->rt_runtime) goto balanced; - spin_unlock(&rt_rq->rt_runtime_lock); + raw_spin_unlock(&rt_rq->rt_runtime_lock); /* * Calculate the difference between what we started out with @@ -418,7 +418,7 @@ static void __disable_runtime(struct rq *rq) if (iter == rt_rq || iter->rt_runtime == RUNTIME_INF) continue; - spin_lock(&iter->rt_runtime_lock); + raw_spin_lock(&iter->rt_runtime_lock); if (want > 0) { diff = min_t(s64, iter->rt_runtime, want); iter->rt_runtime -= diff; @@ -427,13 +427,13 @@ static void __disable_runtime(struct rq *rq) iter->rt_runtime -= want; want -= want; } - spin_unlock(&iter->rt_runtime_lock); + raw_spin_unlock(&iter->rt_runtime_lock); if (!want) break; } - spin_lock(&rt_rq->rt_runtime_lock); + raw_spin_lock(&rt_rq->rt_runtime_lock); /* * We cannot be left wanting - that would mean some runtime * leaked out of the system. @@ -445,8 +445,8 @@ balanced: * runtime - in which case borrowing doesn't make sense. */ rt_rq->rt_runtime = RUNTIME_INF; - spin_unlock(&rt_rq->rt_runtime_lock); - spin_unlock(&rt_b->rt_runtime_lock); + raw_spin_unlock(&rt_rq->rt_runtime_lock); + raw_spin_unlock(&rt_b->rt_runtime_lock); } } @@ -472,13 +472,13 @@ static void __enable_runtime(struct rq *rq) for_each_leaf_rt_rq(rt_rq, rq) { struct rt_bandwidth *rt_b = sched_rt_bandwidth(rt_rq); - spin_lock(&rt_b->rt_runtime_lock); - spin_lock(&rt_rq->rt_runtime_lock); + raw_spin_lock(&rt_b->rt_runtime_lock); + raw_spin_lock(&rt_rq->rt_runtime_lock); rt_rq->rt_runtime = rt_b->rt_runtime; rt_rq->rt_time = 0; rt_rq->rt_throttled = 0; - spin_unlock(&rt_rq->rt_runtime_lock); - spin_unlock(&rt_b->rt_runtime_lock); + raw_spin_unlock(&rt_rq->rt_runtime_lock); + raw_spin_unlock(&rt_b->rt_runtime_lock); } } @@ -496,9 +496,9 @@ static int balance_runtime(struct rt_rq *rt_rq) int more = 0; if (rt_rq->rt_time > rt_rq->rt_runtime) { - spin_unlock(&rt_rq->rt_runtime_lock); + raw_spin_unlock(&rt_rq->rt_runtime_lock); more = do_balance_runtime(rt_rq); - spin_lock(&rt_rq->rt_runtime_lock); + raw_spin_lock(&rt_rq->rt_runtime_lock); } return more; @@ -528,7 +528,7 @@ static int do_sched_rt_period_timer(struct rt_bandwidth *rt_b, int overrun) if (rt_rq->rt_time) { u64 runtime; - spin_lock(&rt_rq->rt_runtime_lock); + raw_spin_lock(&rt_rq->rt_runtime_lock); if (rt_rq->rt_throttled) balance_runtime(rt_rq); runtime = rt_rq->rt_runtime; @@ -539,7 +539,7 @@ static int do_sched_rt_period_timer(struct rt_bandwidth *rt_b, int overrun) } if (rt_rq->rt_time || rt_rq->rt_nr_running) idle = 0; - spin_unlock(&rt_rq->rt_runtime_lock); + raw_spin_unlock(&rt_rq->rt_runtime_lock); } else if (rt_rq->rt_nr_running) idle = 0; @@ -624,11 +624,11 @@ static void update_curr_rt(struct rq *rq) rt_rq = rt_rq_of_se(rt_se); if (sched_rt_runtime(rt_rq) != RUNTIME_INF) { - spin_lock(&rt_rq->rt_runtime_lock); + raw_spin_lock(&rt_rq->rt_runtime_lock); rt_rq->rt_time += delta_exec; if (sched_rt_runtime_exceeded(rt_rq)) resched_task(curr); - spin_unlock(&rt_rq->rt_runtime_lock); + raw_spin_unlock(&rt_rq->rt_runtime_lock); } } } -- cgit v1.2.3-71-gd317 From fe841226bd954fba4fd79f037a876053fe9c3217 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 17 Nov 2009 17:59:15 +0100 Subject: sched: Convert cpupri lock to raw_spinlock Convert locks which cannot be sleeping locks in preempt-rt to raw_spinlocks. Signed-off-by: Thomas Gleixner Acked-by: Peter Zijlstra Acked-by: Ingo Molnar --- kernel/sched_cpupri.c | 10 +++++----- kernel/sched_cpupri.h | 2 +- 2 files changed, 6 insertions(+), 6 deletions(-) (limited to 'kernel') diff --git a/kernel/sched_cpupri.c b/kernel/sched_cpupri.c index 0f052fc674d5..597b33099dfa 100644 --- a/kernel/sched_cpupri.c +++ b/kernel/sched_cpupri.c @@ -135,26 +135,26 @@ void cpupri_set(struct cpupri *cp, int cpu, int newpri) if (likely(newpri != CPUPRI_INVALID)) { struct cpupri_vec *vec = &cp->pri_to_cpu[newpri]; - spin_lock_irqsave(&vec->lock, flags); + raw_spin_lock_irqsave(&vec->lock, flags); cpumask_set_cpu(cpu, vec->mask); vec->count++; if (vec->count == 1) set_bit(newpri, cp->pri_active); - spin_unlock_irqrestore(&vec->lock, flags); + raw_spin_unlock_irqrestore(&vec->lock, flags); } if (likely(oldpri != CPUPRI_INVALID)) { struct cpupri_vec *vec = &cp->pri_to_cpu[oldpri]; - spin_lock_irqsave(&vec->lock, flags); + raw_spin_lock_irqsave(&vec->lock, flags); vec->count--; if (!vec->count) clear_bit(oldpri, cp->pri_active); cpumask_clear_cpu(cpu, vec->mask); - spin_unlock_irqrestore(&vec->lock, flags); + raw_spin_unlock_irqrestore(&vec->lock, flags); } *currpri = newpri; @@ -180,7 +180,7 @@ int cpupri_init(struct cpupri *cp, bool bootmem) for (i = 0; i < CPUPRI_NR_PRIORITIES; i++) { struct cpupri_vec *vec = &cp->pri_to_cpu[i]; - spin_lock_init(&vec->lock); + raw_spin_lock_init(&vec->lock); vec->count = 0; if (!zalloc_cpumask_var(&vec->mask, gfp)) goto cleanup; diff --git a/kernel/sched_cpupri.h b/kernel/sched_cpupri.h index 9a7e859b8fbf..7cb5bb6b95be 100644 --- a/kernel/sched_cpupri.h +++ b/kernel/sched_cpupri.h @@ -12,7 +12,7 @@ /* values 2-101 are RT priorities 0-99 */ struct cpupri_vec { - spinlock_t lock; + raw_spinlock_t lock; int count; cpumask_var_t mask; }; -- cgit v1.2.3-71-gd317 From 1d615482547584b9a8bb6316a58fed6ce90dd9ff Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 17 Nov 2009 14:54:03 +0100 Subject: sched: Convert pi_lock to raw_spinlock Convert locks which cannot be sleeping locks in preempt-rt to raw_spinlocks. Signed-off-by: Thomas Gleixner Acked-by: Peter Zijlstra Acked-by: Ingo Molnar --- include/linux/init_task.h | 2 +- include/linux/sched.h | 2 +- kernel/exit.c | 2 +- kernel/fork.c | 4 ++-- kernel/futex.c | 38 +++++++++++++++---------------- kernel/rtmutex-debug.c | 4 ++-- kernel/rtmutex.c | 58 +++++++++++++++++++++++------------------------ kernel/sched.c | 12 +++++----- 8 files changed, 61 insertions(+), 61 deletions(-) (limited to 'kernel') diff --git a/include/linux/init_task.h b/include/linux/init_task.h index 8d10aa7fd4c9..abec69b63d7e 100644 --- a/include/linux/init_task.h +++ b/include/linux/init_task.h @@ -165,7 +165,7 @@ extern struct cred init_cred; .journal_info = NULL, \ .cpu_timers = INIT_CPU_TIMERS(tsk.cpu_timers), \ .fs_excl = ATOMIC_INIT(0), \ - .pi_lock = __SPIN_LOCK_UNLOCKED(tsk.pi_lock), \ + .pi_lock = __RAW_SPIN_LOCK_UNLOCKED(tsk.pi_lock), \ .timer_slack_ns = 50000, /* 50 usec default slack */ \ .pids = { \ [PIDTYPE_PID] = INIT_PID_LINK(PIDTYPE_PID), \ diff --git a/include/linux/sched.h b/include/linux/sched.h index 294eb2f80144..41a9ea322dce 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1409,7 +1409,7 @@ struct task_struct { #endif /* Protection of the PI data structures: */ - spinlock_t pi_lock; + raw_spinlock_t pi_lock; #ifdef CONFIG_RT_MUTEXES /* PI waiters blocked on a rt_mutex held by this task */ diff --git a/kernel/exit.c b/kernel/exit.c index 6f50ef55a6f3..5962d7ccf243 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -933,7 +933,7 @@ NORET_TYPE void do_exit(long code) * an exiting task cleaning up the robust pi futexes. */ smp_mb(); - spin_unlock_wait(&tsk->pi_lock); + raw_spin_unlock_wait(&tsk->pi_lock); if (unlikely(in_atomic())) printk(KERN_INFO "note: %s[%d] exited with preempt_count %d\n", diff --git a/kernel/fork.c b/kernel/fork.c index 1415dc4598ae..9bd91447e052 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -939,9 +939,9 @@ SYSCALL_DEFINE1(set_tid_address, int __user *, tidptr) static void rt_mutex_init_task(struct task_struct *p) { - spin_lock_init(&p->pi_lock); + raw_spin_lock_init(&p->pi_lock); #ifdef CONFIG_RT_MUTEXES - plist_head_init(&p->pi_waiters, &p->pi_lock); + plist_head_init_raw(&p->pi_waiters, &p->pi_lock); p->pi_blocked_on = NULL; #endif } diff --git a/kernel/futex.c b/kernel/futex.c index 6af474df17bb..320b369d20b5 100644 --- a/kernel/futex.c +++ b/kernel/futex.c @@ -403,9 +403,9 @@ static void free_pi_state(struct futex_pi_state *pi_state) * and has cleaned up the pi_state already */ if (pi_state->owner) { - spin_lock_irq(&pi_state->owner->pi_lock); + raw_spin_lock_irq(&pi_state->owner->pi_lock); list_del_init(&pi_state->list); - spin_unlock_irq(&pi_state->owner->pi_lock); + raw_spin_unlock_irq(&pi_state->owner->pi_lock); rt_mutex_proxy_unlock(&pi_state->pi_mutex, pi_state->owner); } @@ -470,18 +470,18 @@ void exit_pi_state_list(struct task_struct *curr) * pi_state_list anymore, but we have to be careful * versus waiters unqueueing themselves: */ - spin_lock_irq(&curr->pi_lock); + raw_spin_lock_irq(&curr->pi_lock); while (!list_empty(head)) { next = head->next; pi_state = list_entry(next, struct futex_pi_state, list); key = pi_state->key; hb = hash_futex(&key); - spin_unlock_irq(&curr->pi_lock); + raw_spin_unlock_irq(&curr->pi_lock); spin_lock(&hb->lock); - spin_lock_irq(&curr->pi_lock); + raw_spin_lock_irq(&curr->pi_lock); /* * We dropped the pi-lock, so re-check whether this * task still owns the PI-state: @@ -495,15 +495,15 @@ void exit_pi_state_list(struct task_struct *curr) WARN_ON(list_empty(&pi_state->list)); list_del_init(&pi_state->list); pi_state->owner = NULL; - spin_unlock_irq(&curr->pi_lock); + raw_spin_unlock_irq(&curr->pi_lock); rt_mutex_unlock(&pi_state->pi_mutex); spin_unlock(&hb->lock); - spin_lock_irq(&curr->pi_lock); + raw_spin_lock_irq(&curr->pi_lock); } - spin_unlock_irq(&curr->pi_lock); + raw_spin_unlock_irq(&curr->pi_lock); } static int @@ -558,7 +558,7 @@ lookup_pi_state(u32 uval, struct futex_hash_bucket *hb, * change of the task flags, we do this protected by * p->pi_lock: */ - spin_lock_irq(&p->pi_lock); + raw_spin_lock_irq(&p->pi_lock); if (unlikely(p->flags & PF_EXITING)) { /* * The task is on the way out. When PF_EXITPIDONE is @@ -567,7 +567,7 @@ lookup_pi_state(u32 uval, struct futex_hash_bucket *hb, */ int ret = (p->flags & PF_EXITPIDONE) ? -ESRCH : -EAGAIN; - spin_unlock_irq(&p->pi_lock); + raw_spin_unlock_irq(&p->pi_lock); put_task_struct(p); return ret; } @@ -586,7 +586,7 @@ lookup_pi_state(u32 uval, struct futex_hash_bucket *hb, WARN_ON(!list_empty(&pi_state->list)); list_add(&pi_state->list, &p->pi_state_list); pi_state->owner = p; - spin_unlock_irq(&p->pi_lock); + raw_spin_unlock_irq(&p->pi_lock); put_task_struct(p); @@ -794,16 +794,16 @@ static int wake_futex_pi(u32 __user *uaddr, u32 uval, struct futex_q *this) } } - spin_lock_irq(&pi_state->owner->pi_lock); + raw_spin_lock_irq(&pi_state->owner->pi_lock); WARN_ON(list_empty(&pi_state->list)); list_del_init(&pi_state->list); - spin_unlock_irq(&pi_state->owner->pi_lock); + raw_spin_unlock_irq(&pi_state->owner->pi_lock); - spin_lock_irq(&new_owner->pi_lock); + raw_spin_lock_irq(&new_owner->pi_lock); WARN_ON(!list_empty(&pi_state->list)); list_add(&pi_state->list, &new_owner->pi_state_list); pi_state->owner = new_owner; - spin_unlock_irq(&new_owner->pi_lock); + raw_spin_unlock_irq(&new_owner->pi_lock); spin_unlock(&pi_state->pi_mutex.wait_lock); rt_mutex_unlock(&pi_state->pi_mutex); @@ -1529,18 +1529,18 @@ retry: * itself. */ if (pi_state->owner != NULL) { - spin_lock_irq(&pi_state->owner->pi_lock); + raw_spin_lock_irq(&pi_state->owner->pi_lock); WARN_ON(list_empty(&pi_state->list)); list_del_init(&pi_state->list); - spin_unlock_irq(&pi_state->owner->pi_lock); + raw_spin_unlock_irq(&pi_state->owner->pi_lock); } pi_state->owner = newowner; - spin_lock_irq(&newowner->pi_lock); + raw_spin_lock_irq(&newowner->pi_lock); WARN_ON(!list_empty(&pi_state->list)); list_add(&pi_state->list, &newowner->pi_state_list); - spin_unlock_irq(&newowner->pi_lock); + raw_spin_unlock_irq(&newowner->pi_lock); return 0; /* diff --git a/kernel/rtmutex-debug.c b/kernel/rtmutex-debug.c index 5fcb4fe645e2..ddabb54bb5c8 100644 --- a/kernel/rtmutex-debug.c +++ b/kernel/rtmutex-debug.c @@ -37,8 +37,8 @@ do { \ if (rt_trace_on) { \ rt_trace_on = 0; \ console_verbose(); \ - if (spin_is_locked(¤t->pi_lock)) \ - spin_unlock(¤t->pi_lock); \ + if (raw_spin_is_locked(¤t->pi_lock)) \ + raw_spin_unlock(¤t->pi_lock); \ } \ } while (0) diff --git a/kernel/rtmutex.c b/kernel/rtmutex.c index 29bd4baf9e75..d33da470f9da 100644 --- a/kernel/rtmutex.c +++ b/kernel/rtmutex.c @@ -138,9 +138,9 @@ static void rt_mutex_adjust_prio(struct task_struct *task) { unsigned long flags; - spin_lock_irqsave(&task->pi_lock, flags); + raw_spin_lock_irqsave(&task->pi_lock, flags); __rt_mutex_adjust_prio(task); - spin_unlock_irqrestore(&task->pi_lock, flags); + raw_spin_unlock_irqrestore(&task->pi_lock, flags); } /* @@ -195,7 +195,7 @@ static int rt_mutex_adjust_prio_chain(struct task_struct *task, /* * Task can not go away as we did a get_task() before ! */ - spin_lock_irqsave(&task->pi_lock, flags); + raw_spin_lock_irqsave(&task->pi_lock, flags); waiter = task->pi_blocked_on; /* @@ -232,7 +232,7 @@ static int rt_mutex_adjust_prio_chain(struct task_struct *task, lock = waiter->lock; if (!spin_trylock(&lock->wait_lock)) { - spin_unlock_irqrestore(&task->pi_lock, flags); + raw_spin_unlock_irqrestore(&task->pi_lock, flags); cpu_relax(); goto retry; } @@ -253,13 +253,13 @@ static int rt_mutex_adjust_prio_chain(struct task_struct *task, plist_add(&waiter->list_entry, &lock->wait_list); /* Release the task */ - spin_unlock_irqrestore(&task->pi_lock, flags); + raw_spin_unlock_irqrestore(&task->pi_lock, flags); put_task_struct(task); /* Grab the next task */ task = rt_mutex_owner(lock); get_task_struct(task); - spin_lock_irqsave(&task->pi_lock, flags); + raw_spin_lock_irqsave(&task->pi_lock, flags); if (waiter == rt_mutex_top_waiter(lock)) { /* Boost the owner */ @@ -277,7 +277,7 @@ static int rt_mutex_adjust_prio_chain(struct task_struct *task, __rt_mutex_adjust_prio(task); } - spin_unlock_irqrestore(&task->pi_lock, flags); + raw_spin_unlock_irqrestore(&task->pi_lock, flags); top_waiter = rt_mutex_top_waiter(lock); spin_unlock(&lock->wait_lock); @@ -288,7 +288,7 @@ static int rt_mutex_adjust_prio_chain(struct task_struct *task, goto again; out_unlock_pi: - spin_unlock_irqrestore(&task->pi_lock, flags); + raw_spin_unlock_irqrestore(&task->pi_lock, flags); out_put_task: put_task_struct(task); @@ -313,9 +313,9 @@ static inline int try_to_steal_lock(struct rt_mutex *lock, if (pendowner == task) return 1; - spin_lock_irqsave(&pendowner->pi_lock, flags); + raw_spin_lock_irqsave(&pendowner->pi_lock, flags); if (task->prio >= pendowner->prio) { - spin_unlock_irqrestore(&pendowner->pi_lock, flags); + raw_spin_unlock_irqrestore(&pendowner->pi_lock, flags); return 0; } @@ -325,7 +325,7 @@ static inline int try_to_steal_lock(struct rt_mutex *lock, * priority. */ if (likely(!rt_mutex_has_waiters(lock))) { - spin_unlock_irqrestore(&pendowner->pi_lock, flags); + raw_spin_unlock_irqrestore(&pendowner->pi_lock, flags); return 1; } @@ -333,7 +333,7 @@ static inline int try_to_steal_lock(struct rt_mutex *lock, next = rt_mutex_top_waiter(lock); plist_del(&next->pi_list_entry, &pendowner->pi_waiters); __rt_mutex_adjust_prio(pendowner); - spin_unlock_irqrestore(&pendowner->pi_lock, flags); + raw_spin_unlock_irqrestore(&pendowner->pi_lock, flags); /* * We are going to steal the lock and a waiter was @@ -350,10 +350,10 @@ static inline int try_to_steal_lock(struct rt_mutex *lock, * might be task: */ if (likely(next->task != task)) { - spin_lock_irqsave(&task->pi_lock, flags); + raw_spin_lock_irqsave(&task->pi_lock, flags); plist_add(&next->pi_list_entry, &task->pi_waiters); __rt_mutex_adjust_prio(task); - spin_unlock_irqrestore(&task->pi_lock, flags); + raw_spin_unlock_irqrestore(&task->pi_lock, flags); } return 1; } @@ -420,7 +420,7 @@ static int task_blocks_on_rt_mutex(struct rt_mutex *lock, unsigned long flags; int chain_walk = 0, res; - spin_lock_irqsave(&task->pi_lock, flags); + raw_spin_lock_irqsave(&task->pi_lock, flags); __rt_mutex_adjust_prio(task); waiter->task = task; waiter->lock = lock; @@ -434,17 +434,17 @@ static int task_blocks_on_rt_mutex(struct rt_mutex *lock, task->pi_blocked_on = waiter; - spin_unlock_irqrestore(&task->pi_lock, flags); + raw_spin_unlock_irqrestore(&task->pi_lock, flags); if (waiter == rt_mutex_top_waiter(lock)) { - spin_lock_irqsave(&owner->pi_lock, flags); + raw_spin_lock_irqsave(&owner->pi_lock, flags); plist_del(&top_waiter->pi_list_entry, &owner->pi_waiters); plist_add(&waiter->pi_list_entry, &owner->pi_waiters); __rt_mutex_adjust_prio(owner); if (owner->pi_blocked_on) chain_walk = 1; - spin_unlock_irqrestore(&owner->pi_lock, flags); + raw_spin_unlock_irqrestore(&owner->pi_lock, flags); } else if (debug_rt_mutex_detect_deadlock(waiter, detect_deadlock)) chain_walk = 1; @@ -483,7 +483,7 @@ static void wakeup_next_waiter(struct rt_mutex *lock) struct task_struct *pendowner; unsigned long flags; - spin_lock_irqsave(¤t->pi_lock, flags); + raw_spin_lock_irqsave(¤t->pi_lock, flags); waiter = rt_mutex_top_waiter(lock); plist_del(&waiter->list_entry, &lock->wait_list); @@ -500,7 +500,7 @@ static void wakeup_next_waiter(struct rt_mutex *lock) rt_mutex_set_owner(lock, pendowner, RT_MUTEX_OWNER_PENDING); - spin_unlock_irqrestore(¤t->pi_lock, flags); + raw_spin_unlock_irqrestore(¤t->pi_lock, flags); /* * Clear the pi_blocked_on variable and enqueue a possible @@ -509,7 +509,7 @@ static void wakeup_next_waiter(struct rt_mutex *lock) * waiter with higher priority than pending-owner->normal_prio * is blocked on the unboosted (pending) owner. */ - spin_lock_irqsave(&pendowner->pi_lock, flags); + raw_spin_lock_irqsave(&pendowner->pi_lock, flags); WARN_ON(!pendowner->pi_blocked_on); WARN_ON(pendowner->pi_blocked_on != waiter); @@ -523,7 +523,7 @@ static void wakeup_next_waiter(struct rt_mutex *lock) next = rt_mutex_top_waiter(lock); plist_add(&next->pi_list_entry, &pendowner->pi_waiters); } - spin_unlock_irqrestore(&pendowner->pi_lock, flags); + raw_spin_unlock_irqrestore(&pendowner->pi_lock, flags); wake_up_process(pendowner); } @@ -541,15 +541,15 @@ static void remove_waiter(struct rt_mutex *lock, unsigned long flags; int chain_walk = 0; - spin_lock_irqsave(¤t->pi_lock, flags); + raw_spin_lock_irqsave(¤t->pi_lock, flags); plist_del(&waiter->list_entry, &lock->wait_list); waiter->task = NULL; current->pi_blocked_on = NULL; - spin_unlock_irqrestore(¤t->pi_lock, flags); + raw_spin_unlock_irqrestore(¤t->pi_lock, flags); if (first && owner != current) { - spin_lock_irqsave(&owner->pi_lock, flags); + raw_spin_lock_irqsave(&owner->pi_lock, flags); plist_del(&waiter->pi_list_entry, &owner->pi_waiters); @@ -564,7 +564,7 @@ static void remove_waiter(struct rt_mutex *lock, if (owner->pi_blocked_on) chain_walk = 1; - spin_unlock_irqrestore(&owner->pi_lock, flags); + raw_spin_unlock_irqrestore(&owner->pi_lock, flags); } WARN_ON(!plist_node_empty(&waiter->pi_list_entry)); @@ -592,15 +592,15 @@ void rt_mutex_adjust_pi(struct task_struct *task) struct rt_mutex_waiter *waiter; unsigned long flags; - spin_lock_irqsave(&task->pi_lock, flags); + raw_spin_lock_irqsave(&task->pi_lock, flags); waiter = task->pi_blocked_on; if (!waiter || waiter->list_entry.prio == task->prio) { - spin_unlock_irqrestore(&task->pi_lock, flags); + raw_spin_unlock_irqrestore(&task->pi_lock, flags); return; } - spin_unlock_irqrestore(&task->pi_lock, flags); + raw_spin_unlock_irqrestore(&task->pi_lock, flags); /* gets dropped in rt_mutex_adjust_prio_chain()! */ get_task_struct(task); diff --git a/kernel/sched.c b/kernel/sched.c index 01c5016e57f1..18cceeecce35 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -6323,7 +6323,7 @@ recheck: * make sure no PI-waiters arrive (or leave) while we are * changing the priority of the task: */ - spin_lock_irqsave(&p->pi_lock, flags); + raw_spin_lock_irqsave(&p->pi_lock, flags); /* * To be able to change p->policy safely, the apropriate * runqueue lock must be held. @@ -6333,7 +6333,7 @@ recheck: if (unlikely(oldpolicy != -1 && oldpolicy != p->policy)) { policy = oldpolicy = -1; __task_rq_unlock(rq); - spin_unlock_irqrestore(&p->pi_lock, flags); + raw_spin_unlock_irqrestore(&p->pi_lock, flags); goto recheck; } update_rq_clock(rq); @@ -6357,7 +6357,7 @@ recheck: check_class_changed(rq, p, prev_class, oldprio, running); } __task_rq_unlock(rq); - spin_unlock_irqrestore(&p->pi_lock, flags); + raw_spin_unlock_irqrestore(&p->pi_lock, flags); rt_mutex_adjust_pi(p); @@ -9624,7 +9624,7 @@ void __init sched_init(void) #endif #ifdef CONFIG_RT_MUTEXES - plist_head_init(&init_task.pi_waiters, &init_task.pi_lock); + plist_head_init_raw(&init_task.pi_waiters, &init_task.pi_lock); #endif /* @@ -9749,13 +9749,13 @@ void normalize_rt_tasks(void) continue; } - spin_lock(&p->pi_lock); + raw_spin_lock(&p->pi_lock); rq = __task_rq_lock(p); normalize_task(rq, p); __task_rq_unlock(rq); - spin_unlock(&p->pi_lock); + raw_spin_unlock(&p->pi_lock); } while_each_thread(g, p); read_unlock_irqrestore(&tasklist_lock, flags); -- cgit v1.2.3-71-gd317 From d209d74d52ab39dc071656533cac095294f70de7 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 17 Nov 2009 18:22:11 +0100 Subject: rtmutes: Convert rtmutex.lock to raw_spinlock Convert locks which cannot be sleeping locks in preempt-rt to raw_spinlocks. Signed-off-by: Thomas Gleixner Acked-by: Peter Zijlstra Acked-by: Ingo Molnar --- include/linux/rtmutex.h | 6 +++--- kernel/futex.c | 6 +++--- kernel/rtmutex.c | 48 ++++++++++++++++++++++++------------------------ 3 files changed, 30 insertions(+), 30 deletions(-) (limited to 'kernel') diff --git a/include/linux/rtmutex.h b/include/linux/rtmutex.h index f19b00b7d530..281d8fd775e8 100644 --- a/include/linux/rtmutex.h +++ b/include/linux/rtmutex.h @@ -24,7 +24,7 @@ * @owner: the mutex owner */ struct rt_mutex { - spinlock_t wait_lock; + raw_spinlock_t wait_lock; struct plist_head wait_list; struct task_struct *owner; #ifdef CONFIG_DEBUG_RT_MUTEXES @@ -63,8 +63,8 @@ struct hrtimer_sleeper; #endif #define __RT_MUTEX_INITIALIZER(mutexname) \ - { .wait_lock = __SPIN_LOCK_UNLOCKED(mutexname.wait_lock) \ - , .wait_list = PLIST_HEAD_INIT(mutexname.wait_list, mutexname.wait_lock) \ + { .wait_lock = __RAW_SPIN_LOCK_UNLOCKED(mutexname.wait_lock) \ + , .wait_list = PLIST_HEAD_INIT_RAW(mutexname.wait_list, mutexname.wait_lock) \ , .owner = NULL \ __DEBUG_RT_MUTEX_INITIALIZER(mutexname)} diff --git a/kernel/futex.c b/kernel/futex.c index 320b369d20b5..8e3c3ffe1b9a 100644 --- a/kernel/futex.c +++ b/kernel/futex.c @@ -760,7 +760,7 @@ static int wake_futex_pi(u32 __user *uaddr, u32 uval, struct futex_q *this) if (!pi_state) return -EINVAL; - spin_lock(&pi_state->pi_mutex.wait_lock); + raw_spin_lock(&pi_state->pi_mutex.wait_lock); new_owner = rt_mutex_next_owner(&pi_state->pi_mutex); /* @@ -789,7 +789,7 @@ static int wake_futex_pi(u32 __user *uaddr, u32 uval, struct futex_q *this) else if (curval != uval) ret = -EINVAL; if (ret) { - spin_unlock(&pi_state->pi_mutex.wait_lock); + raw_spin_unlock(&pi_state->pi_mutex.wait_lock); return ret; } } @@ -805,7 +805,7 @@ static int wake_futex_pi(u32 __user *uaddr, u32 uval, struct futex_q *this) pi_state->owner = new_owner; raw_spin_unlock_irq(&new_owner->pi_lock); - spin_unlock(&pi_state->pi_mutex.wait_lock); + raw_spin_unlock(&pi_state->pi_mutex.wait_lock); rt_mutex_unlock(&pi_state->pi_mutex); return 0; diff --git a/kernel/rtmutex.c b/kernel/rtmutex.c index d33da470f9da..a9604815786a 100644 --- a/kernel/rtmutex.c +++ b/kernel/rtmutex.c @@ -231,7 +231,7 @@ static int rt_mutex_adjust_prio_chain(struct task_struct *task, goto out_unlock_pi; lock = waiter->lock; - if (!spin_trylock(&lock->wait_lock)) { + if (!raw_spin_trylock(&lock->wait_lock)) { raw_spin_unlock_irqrestore(&task->pi_lock, flags); cpu_relax(); goto retry; @@ -240,7 +240,7 @@ static int rt_mutex_adjust_prio_chain(struct task_struct *task, /* Deadlock detection */ if (lock == orig_lock || rt_mutex_owner(lock) == top_task) { debug_rt_mutex_deadlock(deadlock_detect, orig_waiter, lock); - spin_unlock(&lock->wait_lock); + raw_spin_unlock(&lock->wait_lock); ret = deadlock_detect ? -EDEADLK : 0; goto out_unlock_pi; } @@ -280,7 +280,7 @@ static int rt_mutex_adjust_prio_chain(struct task_struct *task, raw_spin_unlock_irqrestore(&task->pi_lock, flags); top_waiter = rt_mutex_top_waiter(lock); - spin_unlock(&lock->wait_lock); + raw_spin_unlock(&lock->wait_lock); if (!detect_deadlock && waiter != top_waiter) goto out_put_task; @@ -459,12 +459,12 @@ static int task_blocks_on_rt_mutex(struct rt_mutex *lock, */ get_task_struct(owner); - spin_unlock(&lock->wait_lock); + raw_spin_unlock(&lock->wait_lock); res = rt_mutex_adjust_prio_chain(owner, detect_deadlock, lock, waiter, task); - spin_lock(&lock->wait_lock); + raw_spin_lock(&lock->wait_lock); return res; } @@ -575,11 +575,11 @@ static void remove_waiter(struct rt_mutex *lock, /* gets dropped in rt_mutex_adjust_prio_chain()! */ get_task_struct(owner); - spin_unlock(&lock->wait_lock); + raw_spin_unlock(&lock->wait_lock); rt_mutex_adjust_prio_chain(owner, 0, lock, NULL, current); - spin_lock(&lock->wait_lock); + raw_spin_lock(&lock->wait_lock); } /* @@ -672,14 +672,14 @@ __rt_mutex_slowlock(struct rt_mutex *lock, int state, break; } - spin_unlock(&lock->wait_lock); + raw_spin_unlock(&lock->wait_lock); debug_rt_mutex_print_deadlock(waiter); if (waiter->task) schedule_rt_mutex(lock); - spin_lock(&lock->wait_lock); + raw_spin_lock(&lock->wait_lock); set_current_state(state); } @@ -700,11 +700,11 @@ rt_mutex_slowlock(struct rt_mutex *lock, int state, debug_rt_mutex_init_waiter(&waiter); waiter.task = NULL; - spin_lock(&lock->wait_lock); + raw_spin_lock(&lock->wait_lock); /* Try to acquire the lock again: */ if (try_to_take_rt_mutex(lock)) { - spin_unlock(&lock->wait_lock); + raw_spin_unlock(&lock->wait_lock); return 0; } @@ -731,7 +731,7 @@ rt_mutex_slowlock(struct rt_mutex *lock, int state, */ fixup_rt_mutex_waiters(lock); - spin_unlock(&lock->wait_lock); + raw_spin_unlock(&lock->wait_lock); /* Remove pending timer: */ if (unlikely(timeout)) @@ -758,7 +758,7 @@ rt_mutex_slowtrylock(struct rt_mutex *lock) { int ret = 0; - spin_lock(&lock->wait_lock); + raw_spin_lock(&lock->wait_lock); if (likely(rt_mutex_owner(lock) != current)) { @@ -770,7 +770,7 @@ rt_mutex_slowtrylock(struct rt_mutex *lock) fixup_rt_mutex_waiters(lock); } - spin_unlock(&lock->wait_lock); + raw_spin_unlock(&lock->wait_lock); return ret; } @@ -781,7 +781,7 @@ rt_mutex_slowtrylock(struct rt_mutex *lock) static void __sched rt_mutex_slowunlock(struct rt_mutex *lock) { - spin_lock(&lock->wait_lock); + raw_spin_lock(&lock->wait_lock); debug_rt_mutex_unlock(lock); @@ -789,13 +789,13 @@ rt_mutex_slowunlock(struct rt_mutex *lock) if (!rt_mutex_has_waiters(lock)) { lock->owner = NULL; - spin_unlock(&lock->wait_lock); + raw_spin_unlock(&lock->wait_lock); return; } wakeup_next_waiter(lock); - spin_unlock(&lock->wait_lock); + raw_spin_unlock(&lock->wait_lock); /* Undo pi boosting if necessary: */ rt_mutex_adjust_prio(current); @@ -970,8 +970,8 @@ EXPORT_SYMBOL_GPL(rt_mutex_destroy); void __rt_mutex_init(struct rt_mutex *lock, const char *name) { lock->owner = NULL; - spin_lock_init(&lock->wait_lock); - plist_head_init(&lock->wait_list, &lock->wait_lock); + raw_spin_lock_init(&lock->wait_lock); + plist_head_init_raw(&lock->wait_list, &lock->wait_lock); debug_rt_mutex_init(lock, name); } @@ -1032,7 +1032,7 @@ int rt_mutex_start_proxy_lock(struct rt_mutex *lock, { int ret; - spin_lock(&lock->wait_lock); + raw_spin_lock(&lock->wait_lock); mark_rt_mutex_waiters(lock); @@ -1040,7 +1040,7 @@ int rt_mutex_start_proxy_lock(struct rt_mutex *lock, /* We got the lock for task. */ debug_rt_mutex_lock(lock); rt_mutex_set_owner(lock, task, 0); - spin_unlock(&lock->wait_lock); + raw_spin_unlock(&lock->wait_lock); rt_mutex_deadlock_account_lock(lock, task); return 1; } @@ -1056,7 +1056,7 @@ int rt_mutex_start_proxy_lock(struct rt_mutex *lock, */ ret = 0; } - spin_unlock(&lock->wait_lock); + raw_spin_unlock(&lock->wait_lock); debug_rt_mutex_print_deadlock(waiter); @@ -1106,7 +1106,7 @@ int rt_mutex_finish_proxy_lock(struct rt_mutex *lock, { int ret; - spin_lock(&lock->wait_lock); + raw_spin_lock(&lock->wait_lock); set_current_state(TASK_INTERRUPTIBLE); @@ -1124,7 +1124,7 @@ int rt_mutex_finish_proxy_lock(struct rt_mutex *lock, */ fixup_rt_mutex_waiters(lock); - spin_unlock(&lock->wait_lock); + raw_spin_unlock(&lock->wait_lock); /* * Readjust priority, when we did not get the lock. We might have been -- cgit v1.2.3-71-gd317 From 9f5a5621e78cf48d86682a71ceb3fcdbde38b222 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 17 Nov 2009 15:40:01 +0100 Subject: smp: Convert smplocks to raw_spinlocks Convert locks which cannot be sleeping locks in preempt-rt to raw_spinlocks. Signed-off-by: Thomas Gleixner Acked-by: Peter Zijlstra Acked-by: Ingo Molnar --- kernel/smp.c | 32 ++++++++++++++++---------------- 1 file changed, 16 insertions(+), 16 deletions(-) (limited to 'kernel') diff --git a/kernel/smp.c b/kernel/smp.c index a8c76069cf50..6e7c7fdcd9b4 100644 --- a/kernel/smp.c +++ b/kernel/smp.c @@ -16,11 +16,11 @@ static DEFINE_PER_CPU(struct call_single_queue, call_single_queue); static struct { struct list_head queue; - spinlock_t lock; + raw_spinlock_t lock; } call_function __cacheline_aligned_in_smp = { .queue = LIST_HEAD_INIT(call_function.queue), - .lock = __SPIN_LOCK_UNLOCKED(call_function.lock), + .lock = __RAW_SPIN_LOCK_UNLOCKED(call_function.lock), }; enum { @@ -35,7 +35,7 @@ struct call_function_data { struct call_single_queue { struct list_head list; - spinlock_t lock; + raw_spinlock_t lock; }; static DEFINE_PER_CPU(struct call_function_data, cfd_data); @@ -80,7 +80,7 @@ static int __cpuinit init_call_single_data(void) for_each_possible_cpu(i) { struct call_single_queue *q = &per_cpu(call_single_queue, i); - spin_lock_init(&q->lock); + raw_spin_lock_init(&q->lock); INIT_LIST_HEAD(&q->list); } @@ -141,10 +141,10 @@ void generic_exec_single(int cpu, struct call_single_data *data, int wait) unsigned long flags; int ipi; - spin_lock_irqsave(&dst->lock, flags); + raw_spin_lock_irqsave(&dst->lock, flags); ipi = list_empty(&dst->list); list_add_tail(&data->list, &dst->list); - spin_unlock_irqrestore(&dst->lock, flags); + raw_spin_unlock_irqrestore(&dst->lock, flags); /* * The list addition should be visible before sending the IPI @@ -201,9 +201,9 @@ void generic_smp_call_function_interrupt(void) refs = atomic_dec_return(&data->refs); WARN_ON(refs < 0); if (!refs) { - spin_lock(&call_function.lock); + raw_spin_lock(&call_function.lock); list_del_rcu(&data->csd.list); - spin_unlock(&call_function.lock); + raw_spin_unlock(&call_function.lock); } if (refs) @@ -230,9 +230,9 @@ void generic_smp_call_function_single_interrupt(void) */ WARN_ON_ONCE(!cpu_online(smp_processor_id())); - spin_lock(&q->lock); + raw_spin_lock(&q->lock); list_replace_init(&q->list, &list); - spin_unlock(&q->lock); + raw_spin_unlock(&q->lock); while (!list_empty(&list)) { struct call_single_data *data; @@ -449,14 +449,14 @@ void smp_call_function_many(const struct cpumask *mask, cpumask_clear_cpu(this_cpu, data->cpumask); atomic_set(&data->refs, cpumask_weight(data->cpumask)); - spin_lock_irqsave(&call_function.lock, flags); + raw_spin_lock_irqsave(&call_function.lock, flags); /* * Place entry at the _HEAD_ of the list, so that any cpu still * observing the entry in generic_smp_call_function_interrupt() * will not miss any other list entries: */ list_add_rcu(&data->csd.list, &call_function.queue); - spin_unlock_irqrestore(&call_function.lock, flags); + raw_spin_unlock_irqrestore(&call_function.lock, flags); /* * Make the list addition visible before sending the ipi. @@ -501,20 +501,20 @@ EXPORT_SYMBOL(smp_call_function); void ipi_call_lock(void) { - spin_lock(&call_function.lock); + raw_spin_lock(&call_function.lock); } void ipi_call_unlock(void) { - spin_unlock(&call_function.lock); + raw_spin_unlock(&call_function.lock); } void ipi_call_lock_irq(void) { - spin_lock_irq(&call_function.lock); + raw_spin_lock_irq(&call_function.lock); } void ipi_call_unlock_irq(void) { - spin_unlock_irq(&call_function.lock); + raw_spin_unlock_irq(&call_function.lock); } -- cgit v1.2.3-71-gd317 From 239007b8440abff689632f50cdf0f2b9e895b534 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 17 Nov 2009 16:46:45 +0100 Subject: genirq: Convert irq_desc.lock to raw_spinlock Convert locks which cannot be sleeping locks in preempt-rt to raw_spinlocks. Signed-off-by: Thomas Gleixner Acked-by: Peter Zijlstra Acked-by: Ingo Molnar --- arch/alpha/kernel/irq.c | 4 +- arch/arm/include/asm/mach/irq.h | 4 +- arch/arm/kernel/irq.c | 12 ++--- arch/arm/mach-ns9xxx/irq.c | 8 +-- arch/avr32/kernel/irq.c | 4 +- arch/blackfin/kernel/irqchip.c | 6 +-- arch/blackfin/kernel/traps.c | 4 +- arch/cris/kernel/irq.c | 4 +- arch/frv/kernel/irq.c | 4 +- arch/h8300/kernel/irq.c | 4 +- arch/ia64/kernel/iosapic.c | 6 +-- arch/ia64/kernel/irq.c | 4 +- arch/ia64/kernel/irq_ia64.c | 4 +- arch/m32r/kernel/irq.c | 4 +- arch/microblaze/kernel/irq.c | 4 +- arch/mips/kernel/irq.c | 4 +- arch/mips/vr41xx/common/icu.c | 92 ++++++++++++++++----------------- arch/mn10300/kernel/irq.c | 4 +- arch/parisc/kernel/irq.c | 4 +- arch/powerpc/kernel/irq.c | 8 +-- arch/powerpc/platforms/52xx/media5200.c | 8 +-- arch/powerpc/platforms/cell/interrupt.c | 8 +-- arch/powerpc/platforms/iseries/irq.c | 4 +- arch/powerpc/platforms/pseries/xics.c | 4 +- arch/powerpc/sysdev/fsl_msi.c | 4 +- arch/powerpc/sysdev/uic.c | 8 +-- arch/sh/kernel/irq.c | 4 +- arch/sparc/kernel/irq_64.c | 8 +-- arch/um/kernel/irq.c | 4 +- arch/x86/kernel/apic/io_apic.c | 4 +- arch/x86/kernel/irq.c | 14 ++--- arch/xtensa/kernel/irq.c | 4 +- include/linux/irq.h | 2 +- kernel/irq/autoprobe.c | 20 +++---- kernel/irq/chip.c | 86 +++++++++++++++--------------- kernel/irq/handle.c | 22 ++++---- kernel/irq/internals.h | 2 +- kernel/irq/manage.c | 50 +++++++++--------- kernel/irq/migration.c | 2 +- kernel/irq/numa_migrate.c | 8 +-- kernel/irq/pm.c | 8 +-- kernel/irq/proc.c | 4 +- kernel/irq/spurious.c | 14 ++--- 43 files changed, 240 insertions(+), 240 deletions(-) (limited to 'kernel') diff --git a/arch/alpha/kernel/irq.c b/arch/alpha/kernel/irq.c index c0de072b8305..5f2cf23c4648 100644 --- a/arch/alpha/kernel/irq.c +++ b/arch/alpha/kernel/irq.c @@ -81,7 +81,7 @@ show_interrupts(struct seq_file *p, void *v) #endif if (irq < ACTUAL_NR_IRQS) { - spin_lock_irqsave(&irq_desc[irq].lock, flags); + raw_spin_lock_irqsave(&irq_desc[irq].lock, flags); action = irq_desc[irq].action; if (!action) goto unlock; @@ -105,7 +105,7 @@ show_interrupts(struct seq_file *p, void *v) seq_putc(p, '\n'); unlock: - spin_unlock_irqrestore(&irq_desc[irq].lock, flags); + raw_spin_unlock_irqrestore(&irq_desc[irq].lock, flags); } else if (irq == ACTUAL_NR_IRQS) { #ifdef CONFIG_SMP seq_puts(p, "IPI: "); diff --git a/arch/arm/include/asm/mach/irq.h b/arch/arm/include/asm/mach/irq.h index acac5302e4ea..8920b2d6e3b8 100644 --- a/arch/arm/include/asm/mach/irq.h +++ b/arch/arm/include/asm/mach/irq.h @@ -26,9 +26,9 @@ extern int show_fiq_list(struct seq_file *, void *); */ #define do_bad_IRQ(irq,desc) \ do { \ - spin_lock(&desc->lock); \ + raw_spin_lock(&desc->lock); \ handle_bad_irq(irq, desc); \ - spin_unlock(&desc->lock); \ + raw_spin_unlock(&desc->lock); \ } while(0) #endif diff --git a/arch/arm/kernel/irq.c b/arch/arm/kernel/irq.c index c9a8619f3856..b7cb45bb91e8 100644 --- a/arch/arm/kernel/irq.c +++ b/arch/arm/kernel/irq.c @@ -69,7 +69,7 @@ int show_interrupts(struct seq_file *p, void *v) } if (i < NR_IRQS) { - spin_lock_irqsave(&irq_desc[i].lock, flags); + raw_spin_lock_irqsave(&irq_desc[i].lock, flags); action = irq_desc[i].action; if (!action) goto unlock; @@ -84,7 +84,7 @@ int show_interrupts(struct seq_file *p, void *v) seq_putc(p, '\n'); unlock: - spin_unlock_irqrestore(&irq_desc[i].lock, flags); + raw_spin_unlock_irqrestore(&irq_desc[i].lock, flags); } else if (i == NR_IRQS) { #ifdef CONFIG_FIQ show_fiq_list(p, v); @@ -139,7 +139,7 @@ void set_irq_flags(unsigned int irq, unsigned int iflags) } desc = irq_desc + irq; - spin_lock_irqsave(&desc->lock, flags); + raw_spin_lock_irqsave(&desc->lock, flags); desc->status |= IRQ_NOREQUEST | IRQ_NOPROBE | IRQ_NOAUTOEN; if (iflags & IRQF_VALID) desc->status &= ~IRQ_NOREQUEST; @@ -147,7 +147,7 @@ void set_irq_flags(unsigned int irq, unsigned int iflags) desc->status &= ~IRQ_NOPROBE; if (!(iflags & IRQF_NOAUTOEN)) desc->status &= ~IRQ_NOAUTOEN; - spin_unlock_irqrestore(&desc->lock, flags); + raw_spin_unlock_irqrestore(&desc->lock, flags); } void __init init_IRQ(void) @@ -166,9 +166,9 @@ static void route_irq(struct irq_desc *desc, unsigned int irq, unsigned int cpu) { pr_debug("IRQ%u: moving from cpu%u to cpu%u\n", irq, desc->node, cpu); - spin_lock_irq(&desc->lock); + raw_spin_lock_irq(&desc->lock); desc->chip->set_affinity(irq, cpumask_of(cpu)); - spin_unlock_irq(&desc->lock); + raw_spin_unlock_irq(&desc->lock); } /* diff --git a/arch/arm/mach-ns9xxx/irq.c b/arch/arm/mach-ns9xxx/irq.c index feb0e54a91de..038f24d47023 100644 --- a/arch/arm/mach-ns9xxx/irq.c +++ b/arch/arm/mach-ns9xxx/irq.c @@ -66,7 +66,7 @@ static void handle_prio_irq(unsigned int irq, struct irq_desc *desc) struct irqaction *action; irqreturn_t action_ret; - spin_lock(&desc->lock); + raw_spin_lock(&desc->lock); BUG_ON(desc->status & IRQ_INPROGRESS); @@ -78,7 +78,7 @@ static void handle_prio_irq(unsigned int irq, struct irq_desc *desc) goto out_mask; desc->status |= IRQ_INPROGRESS; - spin_unlock(&desc->lock); + raw_spin_unlock(&desc->lock); action_ret = handle_IRQ_event(irq, action); @@ -87,7 +87,7 @@ static void handle_prio_irq(unsigned int irq, struct irq_desc *desc) * Maybe this function should go to kernel/irq/chip.c? */ note_interrupt(irq, desc, action_ret); - spin_lock(&desc->lock); + raw_spin_lock(&desc->lock); desc->status &= ~IRQ_INPROGRESS; if (desc->status & IRQ_DISABLED) @@ -97,7 +97,7 @@ out_mask: /* ack unconditionally to unmask lower prio irqs */ desc->chip->ack(irq); - spin_unlock(&desc->lock); + raw_spin_unlock(&desc->lock); } #define handle_irq handle_prio_irq #endif diff --git a/arch/avr32/kernel/irq.c b/arch/avr32/kernel/irq.c index 9f572229d318..c8ab63e3a675 100644 --- a/arch/avr32/kernel/irq.c +++ b/arch/avr32/kernel/irq.c @@ -51,7 +51,7 @@ int show_interrupts(struct seq_file *p, void *v) } if (i < NR_IRQS) { - spin_lock_irqsave(&irq_desc[i].lock, flags); + raw_spin_lock_irqsave(&irq_desc[i].lock, flags); action = irq_desc[i].action; if (!action) goto unlock; @@ -66,7 +66,7 @@ int show_interrupts(struct seq_file *p, void *v) seq_putc(p, '\n'); unlock: - spin_unlock_irqrestore(&irq_desc[i].lock, flags); + raw_spin_unlock_irqrestore(&irq_desc[i].lock, flags); } return 0; diff --git a/arch/blackfin/kernel/irqchip.c b/arch/blackfin/kernel/irqchip.c index db9f9c91f11f..64cff54a8a58 100644 --- a/arch/blackfin/kernel/irqchip.c +++ b/arch/blackfin/kernel/irqchip.c @@ -23,7 +23,7 @@ void ack_bad_irq(unsigned int irq) static struct irq_desc bad_irq_desc = { .handle_irq = handle_bad_irq, - .lock = __SPIN_LOCK_UNLOCKED(irq_desc->lock), + .lock = __RAW_SPIN_LOCK_UNLOCKED(bad_irq_desc.lock), }; #ifdef CONFIG_CPUMASK_OFFSTACK @@ -39,7 +39,7 @@ int show_interrupts(struct seq_file *p, void *v) unsigned long flags; if (i < NR_IRQS) { - spin_lock_irqsave(&irq_desc[i].lock, flags); + raw_spin_lock_irqsave(&irq_desc[i].lock, flags); action = irq_desc[i].action; if (!action) goto skip; @@ -53,7 +53,7 @@ int show_interrupts(struct seq_file *p, void *v) seq_putc(p, '\n'); skip: - spin_unlock_irqrestore(&irq_desc[i].lock, flags); + raw_spin_unlock_irqrestore(&irq_desc[i].lock, flags); } else if (i == NR_IRQS) { seq_printf(p, "NMI: "); for_each_online_cpu(j) diff --git a/arch/blackfin/kernel/traps.c b/arch/blackfin/kernel/traps.c index 78cb3d38f899..9636bace00e8 100644 --- a/arch/blackfin/kernel/traps.c +++ b/arch/blackfin/kernel/traps.c @@ -1140,7 +1140,7 @@ void show_regs(struct pt_regs *fp) if (fp->ipend & ~0x3F) { for (i = 0; i < (NR_IRQS - 1); i++) { if (!in_atomic) - spin_lock_irqsave(&irq_desc[i].lock, flags); + raw_spin_lock_irqsave(&irq_desc[i].lock, flags); action = irq_desc[i].action; if (!action) @@ -1155,7 +1155,7 @@ void show_regs(struct pt_regs *fp) verbose_printk("\n"); unlock: if (!in_atomic) - spin_unlock_irqrestore(&irq_desc[i].lock, flags); + raw_spin_unlock_irqrestore(&irq_desc[i].lock, flags); } } diff --git a/arch/cris/kernel/irq.c b/arch/cris/kernel/irq.c index 0ca7d9892cc6..b5ce0724a88f 100644 --- a/arch/cris/kernel/irq.c +++ b/arch/cris/kernel/irq.c @@ -52,7 +52,7 @@ int show_interrupts(struct seq_file *p, void *v) } if (i < NR_IRQS) { - spin_lock_irqsave(&irq_desc[i].lock, flags); + raw_spin_lock_irqsave(&irq_desc[i].lock, flags); action = irq_desc[i].action; if (!action) goto skip; @@ -71,7 +71,7 @@ int show_interrupts(struct seq_file *p, void *v) seq_putc(p, '\n'); skip: - spin_unlock_irqrestore(&irq_desc[i].lock, flags); + raw_spin_unlock_irqrestore(&irq_desc[i].lock, flags); } return 0; } diff --git a/arch/frv/kernel/irq.c b/arch/frv/kernel/irq.c index af3e824b91b3..62d1aba615dc 100644 --- a/arch/frv/kernel/irq.c +++ b/arch/frv/kernel/irq.c @@ -69,7 +69,7 @@ int show_interrupts(struct seq_file *p, void *v) } if (i < NR_IRQS) { - spin_lock_irqsave(&irq_desc[i].lock, flags); + raw_spin_lock_irqsave(&irq_desc[i].lock, flags); action = irq_desc[i].action; if (action) { seq_printf(p, "%3d: ", i); @@ -85,7 +85,7 @@ int show_interrupts(struct seq_file *p, void *v) seq_putc(p, '\n'); } - spin_unlock_irqrestore(&irq_desc[i].lock, flags); + raw_spin_unlock_irqrestore(&irq_desc[i].lock, flags); } else if (i == NR_IRQS) { seq_printf(p, "Err: %10u\n", atomic_read(&irq_err_count)); } diff --git a/arch/h8300/kernel/irq.c b/arch/h8300/kernel/irq.c index 5c913d472119..c25dc2c2b1da 100644 --- a/arch/h8300/kernel/irq.c +++ b/arch/h8300/kernel/irq.c @@ -186,7 +186,7 @@ int show_interrupts(struct seq_file *p, void *v) seq_puts(p, " CPU0"); if (i < NR_IRQS) { - spin_lock_irqsave(&irq_desc[i].lock, flags); + raw_spin_lock_irqsave(&irq_desc[i].lock, flags); action = irq_desc[i].action; if (!action) goto unlock; @@ -200,7 +200,7 @@ int show_interrupts(struct seq_file *p, void *v) seq_printf(p, ", %s", action->name); seq_putc(p, '\n'); unlock: - spin_unlock_irqrestore(&irq_desc[i].lock, flags); + raw_spin_unlock_irqrestore(&irq_desc[i].lock, flags); } return 0; } diff --git a/arch/ia64/kernel/iosapic.c b/arch/ia64/kernel/iosapic.c index dab4d393908c..95ac77aeae9b 100644 --- a/arch/ia64/kernel/iosapic.c +++ b/arch/ia64/kernel/iosapic.c @@ -793,12 +793,12 @@ iosapic_register_intr (unsigned int gsi, goto unlock_iosapic_lock; } - spin_lock(&irq_desc[irq].lock); + raw_spin_lock(&irq_desc[irq].lock); dest = get_target_cpu(gsi, irq); dmode = choose_dmode(); err = register_intr(gsi, irq, dmode, polarity, trigger); if (err < 0) { - spin_unlock(&irq_desc[irq].lock); + raw_spin_unlock(&irq_desc[irq].lock); irq = err; goto unlock_iosapic_lock; } @@ -817,7 +817,7 @@ iosapic_register_intr (unsigned int gsi, (polarity == IOSAPIC_POL_HIGH ? "high" : "low"), cpu_logical_id(dest), dest, irq_to_vector(irq)); - spin_unlock(&irq_desc[irq].lock); + raw_spin_unlock(&irq_desc[irq].lock); unlock_iosapic_lock: spin_unlock_irqrestore(&iosapic_lock, flags); return irq; diff --git a/arch/ia64/kernel/irq.c b/arch/ia64/kernel/irq.c index 7d8951229e7c..94ee9d067cbd 100644 --- a/arch/ia64/kernel/irq.c +++ b/arch/ia64/kernel/irq.c @@ -71,7 +71,7 @@ int show_interrupts(struct seq_file *p, void *v) } if (i < NR_IRQS) { - spin_lock_irqsave(&irq_desc[i].lock, flags); + raw_spin_lock_irqsave(&irq_desc[i].lock, flags); action = irq_desc[i].action; if (!action) goto skip; @@ -91,7 +91,7 @@ int show_interrupts(struct seq_file *p, void *v) seq_putc(p, '\n'); skip: - spin_unlock_irqrestore(&irq_desc[i].lock, flags); + raw_spin_unlock_irqrestore(&irq_desc[i].lock, flags); } else if (i == NR_IRQS) seq_printf(p, "ERR: %10u\n", atomic_read(&irq_err_count)); return 0; diff --git a/arch/ia64/kernel/irq_ia64.c b/arch/ia64/kernel/irq_ia64.c index dd9d7b54f1a1..70e4bad23432 100644 --- a/arch/ia64/kernel/irq_ia64.c +++ b/arch/ia64/kernel/irq_ia64.c @@ -345,7 +345,7 @@ static irqreturn_t smp_irq_move_cleanup_interrupt(int irq, void *dev_id) desc = irq_desc + irq; cfg = irq_cfg + irq; - spin_lock(&desc->lock); + raw_spin_lock(&desc->lock); if (!cfg->move_cleanup_count) goto unlock; @@ -358,7 +358,7 @@ static irqreturn_t smp_irq_move_cleanup_interrupt(int irq, void *dev_id) spin_unlock_irqrestore(&vector_lock, flags); cfg->move_cleanup_count--; unlock: - spin_unlock(&desc->lock); + raw_spin_unlock(&desc->lock); } return IRQ_HANDLED; } diff --git a/arch/m32r/kernel/irq.c b/arch/m32r/kernel/irq.c index 8dfd31e87c4c..3c71f776872c 100644 --- a/arch/m32r/kernel/irq.c +++ b/arch/m32r/kernel/irq.c @@ -40,7 +40,7 @@ int show_interrupts(struct seq_file *p, void *v) } if (i < NR_IRQS) { - spin_lock_irqsave(&irq_desc[i].lock, flags); + raw_spin_lock_irqsave(&irq_desc[i].lock, flags); action = irq_desc[i].action; if (!action) goto skip; @@ -59,7 +59,7 @@ int show_interrupts(struct seq_file *p, void *v) seq_putc(p, '\n'); skip: - spin_unlock_irqrestore(&irq_desc[i].lock, flags); + raw_spin_unlock_irqrestore(&irq_desc[i].lock, flags); } return 0; } diff --git a/arch/microblaze/kernel/irq.c b/arch/microblaze/kernel/irq.c index 7d5ddd62d4d2..0f06034d1fe0 100644 --- a/arch/microblaze/kernel/irq.c +++ b/arch/microblaze/kernel/irq.c @@ -68,7 +68,7 @@ int show_interrupts(struct seq_file *p, void *v) } if (i < nr_irq) { - spin_lock_irqsave(&irq_desc[i].lock, flags); + raw_spin_lock_irqsave(&irq_desc[i].lock, flags); action = irq_desc[i].action; if (!action) goto skip; @@ -89,7 +89,7 @@ int show_interrupts(struct seq_file *p, void *v) seq_putc(p, '\n'); skip: - spin_unlock_irqrestore(&irq_desc[i].lock, flags); + raw_spin_unlock_irqrestore(&irq_desc[i].lock, flags); } return 0; } diff --git a/arch/mips/kernel/irq.c b/arch/mips/kernel/irq.c index 7b845ba9dff4..8b0b4181219f 100644 --- a/arch/mips/kernel/irq.c +++ b/arch/mips/kernel/irq.c @@ -99,7 +99,7 @@ int show_interrupts(struct seq_file *p, void *v) } if (i < NR_IRQS) { - spin_lock_irqsave(&irq_desc[i].lock, flags); + raw_spin_lock_irqsave(&irq_desc[i].lock, flags); action = irq_desc[i].action; if (!action) goto skip; @@ -118,7 +118,7 @@ int show_interrupts(struct seq_file *p, void *v) seq_putc(p, '\n'); skip: - spin_unlock_irqrestore(&irq_desc[i].lock, flags); + raw_spin_unlock_irqrestore(&irq_desc[i].lock, flags); } else if (i == NR_IRQS) { seq_putc(p, '\n'); seq_printf(p, "ERR: %10u\n", atomic_read(&irq_err_count)); diff --git a/arch/mips/vr41xx/common/icu.c b/arch/mips/vr41xx/common/icu.c index 6d39e222b170..6153b6a05ccf 100644 --- a/arch/mips/vr41xx/common/icu.c +++ b/arch/mips/vr41xx/common/icu.c @@ -159,9 +159,9 @@ void vr41xx_enable_piuint(uint16_t mask) if (current_cpu_type() == CPU_VR4111 || current_cpu_type() == CPU_VR4121) { - spin_lock_irqsave(&desc->lock, flags); + raw_spin_lock_irqsave(&desc->lock, flags); icu1_set(MPIUINTREG, mask); - spin_unlock_irqrestore(&desc->lock, flags); + raw_spin_unlock_irqrestore(&desc->lock, flags); } } @@ -174,9 +174,9 @@ void vr41xx_disable_piuint(uint16_t mask) if (current_cpu_type() == CPU_VR4111 || current_cpu_type() == CPU_VR4121) { - spin_lock_irqsave(&desc->lock, flags); + raw_spin_lock_irqsave(&desc->lock, flags); icu1_clear(MPIUINTREG, mask); - spin_unlock_irqrestore(&desc->lock, flags); + raw_spin_unlock_irqrestore(&desc->lock, flags); } } @@ -189,9 +189,9 @@ void vr41xx_enable_aiuint(uint16_t mask) if (current_cpu_type() == CPU_VR4111 || current_cpu_type() == CPU_VR4121) { - spin_lock_irqsave(&desc->lock, flags); + raw_spin_lock_irqsave(&desc->lock, flags); icu1_set(MAIUINTREG, mask); - spin_unlock_irqrestore(&desc->lock, flags); + raw_spin_unlock_irqrestore(&desc->lock, flags); } } @@ -204,9 +204,9 @@ void vr41xx_disable_aiuint(uint16_t mask) if (current_cpu_type() == CPU_VR4111 || current_cpu_type() == CPU_VR4121) { - spin_lock_irqsave(&desc->lock, flags); + raw_spin_lock_irqsave(&desc->lock, flags); icu1_clear(MAIUINTREG, mask); - spin_unlock_irqrestore(&desc->lock, flags); + raw_spin_unlock_irqrestore(&desc->lock, flags); } } @@ -219,9 +219,9 @@ void vr41xx_enable_kiuint(uint16_t mask) if (current_cpu_type() == CPU_VR4111 || current_cpu_type() == CPU_VR4121) { - spin_lock_irqsave(&desc->lock, flags); + raw_spin_lock_irqsave(&desc->lock, flags); icu1_set(MKIUINTREG, mask); - spin_unlock_irqrestore(&desc->lock, flags); + raw_spin_unlock_irqrestore(&desc->lock, flags); } } @@ -234,9 +234,9 @@ void vr41xx_disable_kiuint(uint16_t mask) if (current_cpu_type() == CPU_VR4111 || current_cpu_type() == CPU_VR4121) { - spin_lock_irqsave(&desc->lock, flags); + raw_spin_lock_irqsave(&desc->lock, flags); icu1_clear(MKIUINTREG, mask); - spin_unlock_irqrestore(&desc->lock, flags); + raw_spin_unlock_irqrestore(&desc->lock, flags); } } @@ -247,9 +247,9 @@ void vr41xx_enable_macint(uint16_t mask) struct irq_desc *desc = irq_desc + ETHERNET_IRQ; unsigned long flags; - spin_lock_irqsave(&desc->lock, flags); + raw_spin_lock_irqsave(&desc->lock, flags); icu1_set(MMACINTREG, mask); - spin_unlock_irqrestore(&desc->lock, flags); + raw_spin_unlock_irqrestore(&desc->lock, flags); } EXPORT_SYMBOL(vr41xx_enable_macint); @@ -259,9 +259,9 @@ void vr41xx_disable_macint(uint16_t mask) struct irq_desc *desc = irq_desc + ETHERNET_IRQ; unsigned long flags; - spin_lock_irqsave(&desc->lock, flags); + raw_spin_lock_irqsave(&desc->lock, flags); icu1_clear(MMACINTREG, mask); - spin_unlock_irqrestore(&desc->lock, flags); + raw_spin_unlock_irqrestore(&desc->lock, flags); } EXPORT_SYMBOL(vr41xx_disable_macint); @@ -271,9 +271,9 @@ void vr41xx_enable_dsiuint(uint16_t mask) struct irq_desc *desc = irq_desc + DSIU_IRQ; unsigned long flags; - spin_lock_irqsave(&desc->lock, flags); + raw_spin_lock_irqsave(&desc->lock, flags); icu1_set(MDSIUINTREG, mask); - spin_unlock_irqrestore(&desc->lock, flags); + raw_spin_unlock_irqrestore(&desc->lock, flags); } EXPORT_SYMBOL(vr41xx_enable_dsiuint); @@ -283,9 +283,9 @@ void vr41xx_disable_dsiuint(uint16_t mask) struct irq_desc *desc = irq_desc + DSIU_IRQ; unsigned long flags; - spin_lock_irqsave(&desc->lock, flags); + raw_spin_lock_irqsave(&desc->lock, flags); icu1_clear(MDSIUINTREG, mask); - spin_unlock_irqrestore(&desc->lock, flags); + raw_spin_unlock_irqrestore(&desc->lock, flags); } EXPORT_SYMBOL(vr41xx_disable_dsiuint); @@ -295,9 +295,9 @@ void vr41xx_enable_firint(uint16_t mask) struct irq_desc *desc = irq_desc + FIR_IRQ; unsigned long flags; - spin_lock_irqsave(&desc->lock, flags); + raw_spin_lock_irqsave(&desc->lock, flags); icu2_set(MFIRINTREG, mask); - spin_unlock_irqrestore(&desc->lock, flags); + raw_spin_unlock_irqrestore(&desc->lock, flags); } EXPORT_SYMBOL(vr41xx_enable_firint); @@ -307,9 +307,9 @@ void vr41xx_disable_firint(uint16_t mask) struct irq_desc *desc = irq_desc + FIR_IRQ; unsigned long flags; - spin_lock_irqsave(&desc->lock, flags); + raw_spin_lock_irqsave(&desc->lock, flags); icu2_clear(MFIRINTREG, mask); - spin_unlock_irqrestore(&desc->lock, flags); + raw_spin_unlock_irqrestore(&desc->lock, flags); } EXPORT_SYMBOL(vr41xx_disable_firint); @@ -322,9 +322,9 @@ void vr41xx_enable_pciint(void) if (current_cpu_type() == CPU_VR4122 || current_cpu_type() == CPU_VR4131 || current_cpu_type() == CPU_VR4133) { - spin_lock_irqsave(&desc->lock, flags); + raw_spin_lock_irqsave(&desc->lock, flags); icu2_write(MPCIINTREG, PCIINT0); - spin_unlock_irqrestore(&desc->lock, flags); + raw_spin_unlock_irqrestore(&desc->lock, flags); } } @@ -338,9 +338,9 @@ void vr41xx_disable_pciint(void) if (current_cpu_type() == CPU_VR4122 || current_cpu_type() == CPU_VR4131 || current_cpu_type() == CPU_VR4133) { - spin_lock_irqsave(&desc->lock, flags); + raw_spin_lock_irqsave(&desc->lock, flags); icu2_write(MPCIINTREG, 0); - spin_unlock_irqrestore(&desc->lock, flags); + raw_spin_unlock_irqrestore(&desc->lock, flags); } } @@ -354,9 +354,9 @@ void vr41xx_enable_scuint(void) if (current_cpu_type() == CPU_VR4122 || current_cpu_type() == CPU_VR4131 || current_cpu_type() == CPU_VR4133) { - spin_lock_irqsave(&desc->lock, flags); + raw_spin_lock_irqsave(&desc->lock, flags); icu2_write(MSCUINTREG, SCUINT0); - spin_unlock_irqrestore(&desc->lock, flags); + raw_spin_unlock_irqrestore(&desc->lock, flags); } } @@ -370,9 +370,9 @@ void vr41xx_disable_scuint(void) if (current_cpu_type() == CPU_VR4122 || current_cpu_type() == CPU_VR4131 || current_cpu_type() == CPU_VR4133) { - spin_lock_irqsave(&desc->lock, flags); + raw_spin_lock_irqsave(&desc->lock, flags); icu2_write(MSCUINTREG, 0); - spin_unlock_irqrestore(&desc->lock, flags); + raw_spin_unlock_irqrestore(&desc->lock, flags); } } @@ -386,9 +386,9 @@ void vr41xx_enable_csiint(uint16_t mask) if (current_cpu_type() == CPU_VR4122 || current_cpu_type() == CPU_VR4131 || current_cpu_type() == CPU_VR4133) { - spin_lock_irqsave(&desc->lock, flags); + raw_spin_lock_irqsave(&desc->lock, flags); icu2_set(MCSIINTREG, mask); - spin_unlock_irqrestore(&desc->lock, flags); + raw_spin_unlock_irqrestore(&desc->lock, flags); } } @@ -402,9 +402,9 @@ void vr41xx_disable_csiint(uint16_t mask) if (current_cpu_type() == CPU_VR4122 || current_cpu_type() == CPU_VR4131 || current_cpu_type() == CPU_VR4133) { - spin_lock_irqsave(&desc->lock, flags); + raw_spin_lock_irqsave(&desc->lock, flags); icu2_clear(MCSIINTREG, mask); - spin_unlock_irqrestore(&desc->lock, flags); + raw_spin_unlock_irqrestore(&desc->lock, flags); } } @@ -418,9 +418,9 @@ void vr41xx_enable_bcuint(void) if (current_cpu_type() == CPU_VR4122 || current_cpu_type() == CPU_VR4131 || current_cpu_type() == CPU_VR4133) { - spin_lock_irqsave(&desc->lock, flags); + raw_spin_lock_irqsave(&desc->lock, flags); icu2_write(MBCUINTREG, BCUINTR); - spin_unlock_irqrestore(&desc->lock, flags); + raw_spin_unlock_irqrestore(&desc->lock, flags); } } @@ -434,9 +434,9 @@ void vr41xx_disable_bcuint(void) if (current_cpu_type() == CPU_VR4122 || current_cpu_type() == CPU_VR4131 || current_cpu_type() == CPU_VR4133) { - spin_lock_irqsave(&desc->lock, flags); + raw_spin_lock_irqsave(&desc->lock, flags); icu2_write(MBCUINTREG, 0); - spin_unlock_irqrestore(&desc->lock, flags); + raw_spin_unlock_irqrestore(&desc->lock, flags); } } @@ -486,7 +486,7 @@ static inline int set_sysint1_assign(unsigned int irq, unsigned char assign) pin = SYSINT1_IRQ_TO_PIN(irq); - spin_lock_irq(&desc->lock); + raw_spin_lock_irq(&desc->lock); intassign0 = icu1_read(INTASSIGN0); intassign1 = icu1_read(INTASSIGN1); @@ -525,7 +525,7 @@ static inline int set_sysint1_assign(unsigned int irq, unsigned char assign) intassign1 |= (uint16_t)assign << 9; break; default: - spin_unlock_irq(&desc->lock); + raw_spin_unlock_irq(&desc->lock); return -EINVAL; } @@ -533,7 +533,7 @@ static inline int set_sysint1_assign(unsigned int irq, unsigned char assign) icu1_write(INTASSIGN0, intassign0); icu1_write(INTASSIGN1, intassign1); - spin_unlock_irq(&desc->lock); + raw_spin_unlock_irq(&desc->lock); return 0; } @@ -546,7 +546,7 @@ static inline int set_sysint2_assign(unsigned int irq, unsigned char assign) pin = SYSINT2_IRQ_TO_PIN(irq); - spin_lock_irq(&desc->lock); + raw_spin_lock_irq(&desc->lock); intassign2 = icu1_read(INTASSIGN2); intassign3 = icu1_read(INTASSIGN3); @@ -593,7 +593,7 @@ static inline int set_sysint2_assign(unsigned int irq, unsigned char assign) intassign3 |= (uint16_t)assign << 12; break; default: - spin_unlock_irq(&desc->lock); + raw_spin_unlock_irq(&desc->lock); return -EINVAL; } @@ -601,7 +601,7 @@ static inline int set_sysint2_assign(unsigned int irq, unsigned char assign) icu1_write(INTASSIGN2, intassign2); icu1_write(INTASSIGN3, intassign3); - spin_unlock_irq(&desc->lock); + raw_spin_unlock_irq(&desc->lock); return 0; } diff --git a/arch/mn10300/kernel/irq.c b/arch/mn10300/kernel/irq.c index 4c3c58ef5cda..e2d5ed891f37 100644 --- a/arch/mn10300/kernel/irq.c +++ b/arch/mn10300/kernel/irq.c @@ -215,7 +215,7 @@ int show_interrupts(struct seq_file *p, void *v) /* display information rows, one per active CPU */ case 1 ... NR_IRQS - 1: - spin_lock_irqsave(&irq_desc[i].lock, flags); + raw_spin_lock_irqsave(&irq_desc[i].lock, flags); action = irq_desc[i].action; if (action) { @@ -235,7 +235,7 @@ int show_interrupts(struct seq_file *p, void *v) seq_putc(p, '\n'); } - spin_unlock_irqrestore(&irq_desc[i].lock, flags); + raw_spin_unlock_irqrestore(&irq_desc[i].lock, flags); break; /* polish off with NMI and error counters */ diff --git a/arch/parisc/kernel/irq.c b/arch/parisc/kernel/irq.c index 2e7610cb33d5..f47465e8d040 100644 --- a/arch/parisc/kernel/irq.c +++ b/arch/parisc/kernel/irq.c @@ -180,7 +180,7 @@ int show_interrupts(struct seq_file *p, void *v) if (i < NR_IRQS) { struct irqaction *action; - spin_lock_irqsave(&irq_desc[i].lock, flags); + raw_spin_lock_irqsave(&irq_desc[i].lock, flags); action = irq_desc[i].action; if (!action) goto skip; @@ -224,7 +224,7 @@ int show_interrupts(struct seq_file *p, void *v) seq_putc(p, '\n'); skip: - spin_unlock_irqrestore(&irq_desc[i].lock, flags); + raw_spin_unlock_irqrestore(&irq_desc[i].lock, flags); } return 0; diff --git a/arch/powerpc/kernel/irq.c b/arch/powerpc/kernel/irq.c index f6dca4f4b295..9040330b0530 100644 --- a/arch/powerpc/kernel/irq.c +++ b/arch/powerpc/kernel/irq.c @@ -210,7 +210,7 @@ int show_interrupts(struct seq_file *p, void *v) if (!desc) return 0; - spin_lock_irqsave(&desc->lock, flags); + raw_spin_lock_irqsave(&desc->lock, flags); action = desc->action; if (!action || !action->handler) @@ -237,7 +237,7 @@ int show_interrupts(struct seq_file *p, void *v) seq_putc(p, '\n'); skip: - spin_unlock_irqrestore(&desc->lock, flags); + raw_spin_unlock_irqrestore(&desc->lock, flags); return 0; } @@ -1112,7 +1112,7 @@ static int virq_debug_show(struct seq_file *m, void *private) if (!desc) continue; - spin_lock_irqsave(&desc->lock, flags); + raw_spin_lock_irqsave(&desc->lock, flags); if (desc->action && desc->action->handler) { seq_printf(m, "%5d ", i); @@ -1131,7 +1131,7 @@ static int virq_debug_show(struct seq_file *m, void *private) seq_printf(m, "%s\n", p); } - spin_unlock_irqrestore(&desc->lock, flags); + raw_spin_unlock_irqrestore(&desc->lock, flags); } return 0; diff --git a/arch/powerpc/platforms/52xx/media5200.c b/arch/powerpc/platforms/52xx/media5200.c index cc0c854291d7..0bac3a3dbecf 100644 --- a/arch/powerpc/platforms/52xx/media5200.c +++ b/arch/powerpc/platforms/52xx/media5200.c @@ -86,9 +86,9 @@ void media5200_irq_cascade(unsigned int virq, struct irq_desc *desc) u32 status, enable; /* Mask off the cascaded IRQ */ - spin_lock(&desc->lock); + raw_spin_lock(&desc->lock); desc->chip->mask(virq); - spin_unlock(&desc->lock); + raw_spin_unlock(&desc->lock); /* Ask the FPGA for IRQ status. If 'val' is 0, then no irqs * are pending. 'ffs()' is 1 based */ @@ -104,11 +104,11 @@ void media5200_irq_cascade(unsigned int virq, struct irq_desc *desc) } /* Processing done; can reenable the cascade now */ - spin_lock(&desc->lock); + raw_spin_lock(&desc->lock); desc->chip->ack(virq); if (!(desc->status & IRQ_DISABLED)) desc->chip->unmask(virq); - spin_unlock(&desc->lock); + raw_spin_unlock(&desc->lock); } static int media5200_irq_map(struct irq_host *h, unsigned int virq, diff --git a/arch/powerpc/platforms/cell/interrupt.c b/arch/powerpc/platforms/cell/interrupt.c index 7267effc8078..6829cf7e2bda 100644 --- a/arch/powerpc/platforms/cell/interrupt.c +++ b/arch/powerpc/platforms/cell/interrupt.c @@ -237,7 +237,7 @@ extern int noirqdebug; static void handle_iic_irq(unsigned int irq, struct irq_desc *desc) { - spin_lock(&desc->lock); + raw_spin_lock(&desc->lock); desc->status &= ~(IRQ_REPLAY | IRQ_WAITING); @@ -265,18 +265,18 @@ static void handle_iic_irq(unsigned int irq, struct irq_desc *desc) goto out_eoi; desc->status &= ~IRQ_PENDING; - spin_unlock(&desc->lock); + raw_spin_unlock(&desc->lock); action_ret = handle_IRQ_event(irq, action); if (!noirqdebug) note_interrupt(irq, desc, action_ret); - spin_lock(&desc->lock); + raw_spin_lock(&desc->lock); } while ((desc->status & (IRQ_PENDING | IRQ_DISABLED)) == IRQ_PENDING); desc->status &= ~IRQ_INPROGRESS; out_eoi: desc->chip->eoi(irq); - spin_unlock(&desc->lock); + raw_spin_unlock(&desc->lock); } static int iic_host_map(struct irq_host *h, unsigned int virq, diff --git a/arch/powerpc/platforms/iseries/irq.c b/arch/powerpc/platforms/iseries/irq.c index 07762259c60a..86c4b29eea89 100644 --- a/arch/powerpc/platforms/iseries/irq.c +++ b/arch/powerpc/platforms/iseries/irq.c @@ -217,9 +217,9 @@ void __init iSeries_activate_IRQs() struct irq_desc *desc = irq_to_desc(irq); if (desc && desc->chip && desc->chip->startup) { - spin_lock_irqsave(&desc->lock, flags); + raw_spin_lock_irqsave(&desc->lock, flags); desc->chip->startup(irq); - spin_unlock_irqrestore(&desc->lock, flags); + raw_spin_unlock_irqrestore(&desc->lock, flags); } } } diff --git a/arch/powerpc/platforms/pseries/xics.c b/arch/powerpc/platforms/pseries/xics.c index 7d01b58f3989..b9b9e11609ec 100644 --- a/arch/powerpc/platforms/pseries/xics.c +++ b/arch/powerpc/platforms/pseries/xics.c @@ -906,7 +906,7 @@ void xics_migrate_irqs_away(void) || desc->chip->set_affinity == NULL) continue; - spin_lock_irqsave(&desc->lock, flags); + raw_spin_lock_irqsave(&desc->lock, flags); status = rtas_call(ibm_get_xive, 1, 3, xics_status, irq); if (status) { @@ -930,7 +930,7 @@ void xics_migrate_irqs_away(void) cpumask_setall(irq_to_desc(virq)->affinity); desc->chip->set_affinity(virq, cpu_all_mask); unlock: - spin_unlock_irqrestore(&desc->lock, flags); + raw_spin_unlock_irqrestore(&desc->lock, flags); } } #endif diff --git a/arch/powerpc/sysdev/fsl_msi.c b/arch/powerpc/sysdev/fsl_msi.c index 62e50258cdef..c6e11b077108 100644 --- a/arch/powerpc/sysdev/fsl_msi.c +++ b/arch/powerpc/sysdev/fsl_msi.c @@ -173,7 +173,7 @@ static void fsl_msi_cascade(unsigned int irq, struct irq_desc *desc) u32 intr_index; u32 have_shift = 0; - spin_lock(&desc->lock); + raw_spin_lock(&desc->lock); if ((msi_data->feature & FSL_PIC_IP_MASK) == FSL_PIC_IP_IPIC) { if (desc->chip->mask_ack) desc->chip->mask_ack(irq); @@ -225,7 +225,7 @@ static void fsl_msi_cascade(unsigned int irq, struct irq_desc *desc) break; } unlock: - spin_unlock(&desc->lock); + raw_spin_unlock(&desc->lock); } static int __devinit fsl_of_msi_probe(struct of_device *dev, diff --git a/arch/powerpc/sysdev/uic.c b/arch/powerpc/sysdev/uic.c index 7d10074b3304..6f220a913e42 100644 --- a/arch/powerpc/sysdev/uic.c +++ b/arch/powerpc/sysdev/uic.c @@ -225,12 +225,12 @@ void uic_irq_cascade(unsigned int virq, struct irq_desc *desc) int src; int subvirq; - spin_lock(&desc->lock); + raw_spin_lock(&desc->lock); if (desc->status & IRQ_LEVEL) desc->chip->mask(virq); else desc->chip->mask_ack(virq); - spin_unlock(&desc->lock); + raw_spin_unlock(&desc->lock); msr = mfdcr(uic->dcrbase + UIC_MSR); if (!msr) /* spurious interrupt */ @@ -242,12 +242,12 @@ void uic_irq_cascade(unsigned int virq, struct irq_desc *desc) generic_handle_irq(subvirq); uic_irq_ret: - spin_lock(&desc->lock); + raw_spin_lock(&desc->lock); if (desc->status & IRQ_LEVEL) desc->chip->ack(virq); if (!(desc->status & IRQ_DISABLED) && desc->chip->unmask) desc->chip->unmask(virq); - spin_unlock(&desc->lock); + raw_spin_unlock(&desc->lock); } static struct uic * __init uic_init_one(struct device_node *node) diff --git a/arch/sh/kernel/irq.c b/arch/sh/kernel/irq.c index e1913f28f418..d2d41d046657 100644 --- a/arch/sh/kernel/irq.c +++ b/arch/sh/kernel/irq.c @@ -76,7 +76,7 @@ int show_interrupts(struct seq_file *p, void *v) if (!desc) return 0; - spin_lock_irqsave(&desc->lock, flags); + raw_spin_lock_irqsave(&desc->lock, flags); for_each_online_cpu(j) any_count |= kstat_irqs_cpu(i, j); action = desc->action; @@ -97,7 +97,7 @@ int show_interrupts(struct seq_file *p, void *v) seq_putc(p, '\n'); out: - spin_unlock_irqrestore(&desc->lock, flags); + raw_spin_unlock_irqrestore(&desc->lock, flags); return 0; } #endif diff --git a/arch/sparc/kernel/irq_64.c b/arch/sparc/kernel/irq_64.c index ce996f97855f..8d6882bb480a 100644 --- a/arch/sparc/kernel/irq_64.c +++ b/arch/sparc/kernel/irq_64.c @@ -176,7 +176,7 @@ int show_interrupts(struct seq_file *p, void *v) } if (i < NR_IRQS) { - spin_lock_irqsave(&irq_desc[i].lock, flags); + raw_spin_lock_irqsave(&irq_desc[i].lock, flags); action = irq_desc[i].action; if (!action) goto skip; @@ -195,7 +195,7 @@ int show_interrupts(struct seq_file *p, void *v) seq_putc(p, '\n'); skip: - spin_unlock_irqrestore(&irq_desc[i].lock, flags); + raw_spin_unlock_irqrestore(&irq_desc[i].lock, flags); } else if (i == NR_IRQS) { seq_printf(p, "NMI: "); for_each_online_cpu(j) @@ -785,14 +785,14 @@ void fixup_irqs(void) for (irq = 0; irq < NR_IRQS; irq++) { unsigned long flags; - spin_lock_irqsave(&irq_desc[irq].lock, flags); + raw_spin_lock_irqsave(&irq_desc[irq].lock, flags); if (irq_desc[irq].action && !(irq_desc[irq].status & IRQ_PER_CPU)) { if (irq_desc[irq].chip->set_affinity) irq_desc[irq].chip->set_affinity(irq, irq_desc[irq].affinity); } - spin_unlock_irqrestore(&irq_desc[irq].lock, flags); + raw_spin_unlock_irqrestore(&irq_desc[irq].lock, flags); } tick_ops->disable_irq(); diff --git a/arch/um/kernel/irq.c b/arch/um/kernel/irq.c index 039270b9b73b..89474ba0741e 100644 --- a/arch/um/kernel/irq.c +++ b/arch/um/kernel/irq.c @@ -34,7 +34,7 @@ int show_interrupts(struct seq_file *p, void *v) } if (i < NR_IRQS) { - spin_lock_irqsave(&irq_desc[i].lock, flags); + raw_spin_lock_irqsave(&irq_desc[i].lock, flags); action = irq_desc[i].action; if (!action) goto skip; @@ -53,7 +53,7 @@ int show_interrupts(struct seq_file *p, void *v) seq_putc(p, '\n'); skip: - spin_unlock_irqrestore(&irq_desc[i].lock, flags); + raw_spin_unlock_irqrestore(&irq_desc[i].lock, flags); } else if (i == NR_IRQS) seq_putc(p, '\n'); diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c index d5d498fbee4b..11a5851f1f50 100644 --- a/arch/x86/kernel/apic/io_apic.c +++ b/arch/x86/kernel/apic/io_apic.c @@ -2431,7 +2431,7 @@ asmlinkage void smp_irq_move_cleanup_interrupt(void) continue; cfg = irq_cfg(irq); - spin_lock(&desc->lock); + raw_spin_lock(&desc->lock); if (vector == cfg->vector && cpumask_test_cpu(me, cfg->domain)) goto unlock; @@ -2450,7 +2450,7 @@ asmlinkage void smp_irq_move_cleanup_interrupt(void) } __get_cpu_var(vector_irq)[vector] = -1; unlock: - spin_unlock(&desc->lock); + raw_spin_unlock(&desc->lock); } irq_exit(); diff --git a/arch/x86/kernel/irq.c b/arch/x86/kernel/irq.c index 664bcb7384ac..91fd0c70a18a 100644 --- a/arch/x86/kernel/irq.c +++ b/arch/x86/kernel/irq.c @@ -149,7 +149,7 @@ int show_interrupts(struct seq_file *p, void *v) if (!desc) return 0; - spin_lock_irqsave(&desc->lock, flags); + raw_spin_lock_irqsave(&desc->lock, flags); for_each_online_cpu(j) any_count |= kstat_irqs_cpu(i, j); action = desc->action; @@ -170,7 +170,7 @@ int show_interrupts(struct seq_file *p, void *v) seq_putc(p, '\n'); out: - spin_unlock_irqrestore(&desc->lock, flags); + raw_spin_unlock_irqrestore(&desc->lock, flags); return 0; } @@ -294,12 +294,12 @@ void fixup_irqs(void) continue; /* interrupt's are disabled at this point */ - spin_lock(&desc->lock); + raw_spin_lock(&desc->lock); affinity = desc->affinity; if (!irq_has_action(irq) || cpumask_equal(affinity, cpu_online_mask)) { - spin_unlock(&desc->lock); + raw_spin_unlock(&desc->lock); continue; } @@ -326,7 +326,7 @@ void fixup_irqs(void) if (!(desc->status & IRQ_MOVE_PCNTXT) && desc->chip->unmask) desc->chip->unmask(irq); - spin_unlock(&desc->lock); + raw_spin_unlock(&desc->lock); if (break_affinity && set_affinity) printk("Broke affinity for irq %i\n", irq); @@ -356,10 +356,10 @@ void fixup_irqs(void) irq = __get_cpu_var(vector_irq)[vector]; desc = irq_to_desc(irq); - spin_lock(&desc->lock); + raw_spin_lock(&desc->lock); if (desc->chip->retrigger) desc->chip->retrigger(irq); - spin_unlock(&desc->lock); + raw_spin_unlock(&desc->lock); } } } diff --git a/arch/xtensa/kernel/irq.c b/arch/xtensa/kernel/irq.c index a1badb32fcda..8cd38484e130 100644 --- a/arch/xtensa/kernel/irq.c +++ b/arch/xtensa/kernel/irq.c @@ -90,7 +90,7 @@ int show_interrupts(struct seq_file *p, void *v) } if (i < NR_IRQS) { - spin_lock_irqsave(&irq_desc[i].lock, flags); + raw_spin_lock_irqsave(&irq_desc[i].lock, flags); action = irq_desc[i].action; if (!action) goto skip; @@ -109,7 +109,7 @@ int show_interrupts(struct seq_file *p, void *v) seq_putc(p, '\n'); skip: - spin_unlock_irqrestore(&irq_desc[i].lock, flags); + raw_spin_unlock_irqrestore(&irq_desc[i].lock, flags); } else if (i == NR_IRQS) { seq_printf(p, "NMI: "); for_each_online_cpu(j) diff --git a/include/linux/irq.h b/include/linux/irq.h index a287cfc0b1a6..451481c082b5 100644 --- a/include/linux/irq.h +++ b/include/linux/irq.h @@ -192,7 +192,7 @@ struct irq_desc { unsigned int irq_count; /* For detecting broken IRQs */ unsigned long last_unhandled; /* Aging timer for unhandled count */ unsigned int irqs_unhandled; - spinlock_t lock; + raw_spinlock_t lock; #ifdef CONFIG_SMP cpumask_var_t affinity; unsigned int node; diff --git a/kernel/irq/autoprobe.c b/kernel/irq/autoprobe.c index 1de9700f416e..2295a31ef110 100644 --- a/kernel/irq/autoprobe.c +++ b/kernel/irq/autoprobe.c @@ -45,7 +45,7 @@ unsigned long probe_irq_on(void) * flush such a longstanding irq before considering it as spurious. */ for_each_irq_desc_reverse(i, desc) { - spin_lock_irq(&desc->lock); + raw_spin_lock_irq(&desc->lock); if (!desc->action && !(desc->status & IRQ_NOPROBE)) { /* * An old-style architecture might still have @@ -61,7 +61,7 @@ unsigned long probe_irq_on(void) desc->chip->set_type(i, IRQ_TYPE_PROBE); desc->chip->startup(i); } - spin_unlock_irq(&desc->lock); + raw_spin_unlock_irq(&desc->lock); } /* Wait for longstanding interrupts to trigger. */ @@ -73,13 +73,13 @@ unsigned long probe_irq_on(void) * happened in the previous stage, it may have masked itself) */ for_each_irq_desc_reverse(i, desc) { - spin_lock_irq(&desc->lock); + raw_spin_lock_irq(&desc->lock); if (!desc->action && !(desc->status & IRQ_NOPROBE)) { desc->status |= IRQ_AUTODETECT | IRQ_WAITING; if (desc->chip->startup(i)) desc->status |= IRQ_PENDING; } - spin_unlock_irq(&desc->lock); + raw_spin_unlock_irq(&desc->lock); } /* @@ -91,7 +91,7 @@ unsigned long probe_irq_on(void) * Now filter out any obviously spurious interrupts */ for_each_irq_desc(i, desc) { - spin_lock_irq(&desc->lock); + raw_spin_lock_irq(&desc->lock); status = desc->status; if (status & IRQ_AUTODETECT) { @@ -103,7 +103,7 @@ unsigned long probe_irq_on(void) if (i < 32) mask |= 1 << i; } - spin_unlock_irq(&desc->lock); + raw_spin_unlock_irq(&desc->lock); } return mask; @@ -129,7 +129,7 @@ unsigned int probe_irq_mask(unsigned long val) int i; for_each_irq_desc(i, desc) { - spin_lock_irq(&desc->lock); + raw_spin_lock_irq(&desc->lock); status = desc->status; if (status & IRQ_AUTODETECT) { @@ -139,7 +139,7 @@ unsigned int probe_irq_mask(unsigned long val) desc->status = status & ~IRQ_AUTODETECT; desc->chip->shutdown(i); } - spin_unlock_irq(&desc->lock); + raw_spin_unlock_irq(&desc->lock); } mutex_unlock(&probing_active); @@ -171,7 +171,7 @@ int probe_irq_off(unsigned long val) unsigned int status; for_each_irq_desc(i, desc) { - spin_lock_irq(&desc->lock); + raw_spin_lock_irq(&desc->lock); status = desc->status; if (status & IRQ_AUTODETECT) { @@ -183,7 +183,7 @@ int probe_irq_off(unsigned long val) desc->status = status & ~IRQ_AUTODETECT; desc->chip->shutdown(i); } - spin_unlock_irq(&desc->lock); + raw_spin_unlock_irq(&desc->lock); } mutex_unlock(&probing_active); diff --git a/kernel/irq/chip.c b/kernel/irq/chip.c index ba566c261adc..ecc3fa28f666 100644 --- a/kernel/irq/chip.c +++ b/kernel/irq/chip.c @@ -34,7 +34,7 @@ void dynamic_irq_init(unsigned int irq) } /* Ensure we don't have left over values from a previous use of this irq */ - spin_lock_irqsave(&desc->lock, flags); + raw_spin_lock_irqsave(&desc->lock, flags); desc->status = IRQ_DISABLED; desc->chip = &no_irq_chip; desc->handle_irq = handle_bad_irq; @@ -51,7 +51,7 @@ void dynamic_irq_init(unsigned int irq) cpumask_clear(desc->pending_mask); #endif #endif - spin_unlock_irqrestore(&desc->lock, flags); + raw_spin_unlock_irqrestore(&desc->lock, flags); } /** @@ -68,9 +68,9 @@ void dynamic_irq_cleanup(unsigned int irq) return; } - spin_lock_irqsave(&desc->lock, flags); + raw_spin_lock_irqsave(&desc->lock, flags); if (desc->action) { - spin_unlock_irqrestore(&desc->lock, flags); + raw_spin_unlock_irqrestore(&desc->lock, flags); WARN(1, KERN_ERR "Destroying IRQ%d without calling free_irq\n", irq); return; @@ -82,7 +82,7 @@ void dynamic_irq_cleanup(unsigned int irq) desc->chip = &no_irq_chip; desc->name = NULL; clear_kstat_irqs(desc); - spin_unlock_irqrestore(&desc->lock, flags); + raw_spin_unlock_irqrestore(&desc->lock, flags); } @@ -104,10 +104,10 @@ int set_irq_chip(unsigned int irq, struct irq_chip *chip) if (!chip) chip = &no_irq_chip; - spin_lock_irqsave(&desc->lock, flags); + raw_spin_lock_irqsave(&desc->lock, flags); irq_chip_set_defaults(chip); desc->chip = chip; - spin_unlock_irqrestore(&desc->lock, flags); + raw_spin_unlock_irqrestore(&desc->lock, flags); return 0; } @@ -133,9 +133,9 @@ int set_irq_type(unsigned int irq, unsigned int type) if (type == IRQ_TYPE_NONE) return 0; - spin_lock_irqsave(&desc->lock, flags); + raw_spin_lock_irqsave(&desc->lock, flags); ret = __irq_set_trigger(desc, irq, type); - spin_unlock_irqrestore(&desc->lock, flags); + raw_spin_unlock_irqrestore(&desc->lock, flags); return ret; } EXPORT_SYMBOL(set_irq_type); @@ -158,9 +158,9 @@ int set_irq_data(unsigned int irq, void *data) return -EINVAL; } - spin_lock_irqsave(&desc->lock, flags); + raw_spin_lock_irqsave(&desc->lock, flags); desc->handler_data = data; - spin_unlock_irqrestore(&desc->lock, flags); + raw_spin_unlock_irqrestore(&desc->lock, flags); return 0; } EXPORT_SYMBOL(set_irq_data); @@ -183,11 +183,11 @@ int set_irq_msi(unsigned int irq, struct msi_desc *entry) return -EINVAL; } - spin_lock_irqsave(&desc->lock, flags); + raw_spin_lock_irqsave(&desc->lock, flags); desc->msi_desc = entry; if (entry) entry->irq = irq; - spin_unlock_irqrestore(&desc->lock, flags); + raw_spin_unlock_irqrestore(&desc->lock, flags); return 0; } @@ -214,9 +214,9 @@ int set_irq_chip_data(unsigned int irq, void *data) return -EINVAL; } - spin_lock_irqsave(&desc->lock, flags); + raw_spin_lock_irqsave(&desc->lock, flags); desc->chip_data = data; - spin_unlock_irqrestore(&desc->lock, flags); + raw_spin_unlock_irqrestore(&desc->lock, flags); return 0; } @@ -241,12 +241,12 @@ void set_irq_nested_thread(unsigned int irq, int nest) if (!desc) return; - spin_lock_irqsave(&desc->lock, flags); + raw_spin_lock_irqsave(&desc->lock, flags); if (nest) desc->status |= IRQ_NESTED_THREAD; else desc->status &= ~IRQ_NESTED_THREAD; - spin_unlock_irqrestore(&desc->lock, flags); + raw_spin_unlock_irqrestore(&desc->lock, flags); } EXPORT_SYMBOL_GPL(set_irq_nested_thread); @@ -343,7 +343,7 @@ void handle_nested_irq(unsigned int irq) might_sleep(); - spin_lock_irq(&desc->lock); + raw_spin_lock_irq(&desc->lock); kstat_incr_irqs_this_cpu(irq, desc); @@ -352,17 +352,17 @@ void handle_nested_irq(unsigned int irq) goto out_unlock; desc->status |= IRQ_INPROGRESS; - spin_unlock_irq(&desc->lock); + raw_spin_unlock_irq(&desc->lock); action_ret = action->thread_fn(action->irq, action->dev_id); if (!noirqdebug) note_interrupt(irq, desc, action_ret); - spin_lock_irq(&desc->lock); + raw_spin_lock_irq(&desc->lock); desc->status &= ~IRQ_INPROGRESS; out_unlock: - spin_unlock_irq(&desc->lock); + raw_spin_unlock_irq(&desc->lock); } EXPORT_SYMBOL_GPL(handle_nested_irq); @@ -384,7 +384,7 @@ handle_simple_irq(unsigned int irq, struct irq_desc *desc) struct irqaction *action; irqreturn_t action_ret; - spin_lock(&desc->lock); + raw_spin_lock(&desc->lock); if (unlikely(desc->status & IRQ_INPROGRESS)) goto out_unlock; @@ -396,16 +396,16 @@ handle_simple_irq(unsigned int irq, struct irq_desc *desc) goto out_unlock; desc->status |= IRQ_INPROGRESS; - spin_unlock(&desc->lock); + raw_spin_unlock(&desc->lock); action_ret = handle_IRQ_event(irq, action); if (!noirqdebug) note_interrupt(irq, desc, action_ret); - spin_lock(&desc->lock); + raw_spin_lock(&desc->lock); desc->status &= ~IRQ_INPROGRESS; out_unlock: - spin_unlock(&desc->lock); + raw_spin_unlock(&desc->lock); } /** @@ -424,7 +424,7 @@ handle_level_irq(unsigned int irq, struct irq_desc *desc) struct irqaction *action; irqreturn_t action_ret; - spin_lock(&desc->lock); + raw_spin_lock(&desc->lock); mask_ack_irq(desc, irq); if (unlikely(desc->status & IRQ_INPROGRESS)) @@ -441,13 +441,13 @@ handle_level_irq(unsigned int irq, struct irq_desc *desc) goto out_unlock; desc->status |= IRQ_INPROGRESS; - spin_unlock(&desc->lock); + raw_spin_unlock(&desc->lock); action_ret = handle_IRQ_event(irq, action); if (!noirqdebug) note_interrupt(irq, desc, action_ret); - spin_lock(&desc->lock); + raw_spin_lock(&desc->lock); desc->status &= ~IRQ_INPROGRESS; if (unlikely(desc->status & IRQ_ONESHOT)) @@ -455,7 +455,7 @@ handle_level_irq(unsigned int irq, struct irq_desc *desc) else if (!(desc->status & IRQ_DISABLED) && desc->chip->unmask) desc->chip->unmask(irq); out_unlock: - spin_unlock(&desc->lock); + raw_spin_unlock(&desc->lock); } EXPORT_SYMBOL_GPL(handle_level_irq); @@ -475,7 +475,7 @@ handle_fasteoi_irq(unsigned int irq, struct irq_desc *desc) struct irqaction *action; irqreturn_t action_ret; - spin_lock(&desc->lock); + raw_spin_lock(&desc->lock); if (unlikely(desc->status & IRQ_INPROGRESS)) goto out; @@ -497,18 +497,18 @@ handle_fasteoi_irq(unsigned int irq, struct irq_desc *desc) desc->status |= IRQ_INPROGRESS; desc->status &= ~IRQ_PENDING; - spin_unlock(&desc->lock); + raw_spin_unlock(&desc->lock); action_ret = handle_IRQ_event(irq, action); if (!noirqdebug) note_interrupt(irq, desc, action_ret); - spin_lock(&desc->lock); + raw_spin_lock(&desc->lock); desc->status &= ~IRQ_INPROGRESS; out: desc->chip->eoi(irq); - spin_unlock(&desc->lock); + raw_spin_unlock(&desc->lock); } /** @@ -530,7 +530,7 @@ out: void handle_edge_irq(unsigned int irq, struct irq_desc *desc) { - spin_lock(&desc->lock); + raw_spin_lock(&desc->lock); desc->status &= ~(IRQ_REPLAY | IRQ_WAITING); @@ -576,17 +576,17 @@ handle_edge_irq(unsigned int irq, struct irq_desc *desc) } desc->status &= ~IRQ_PENDING; - spin_unlock(&desc->lock); + raw_spin_unlock(&desc->lock); action_ret = handle_IRQ_event(irq, action); if (!noirqdebug) note_interrupt(irq, desc, action_ret); - spin_lock(&desc->lock); + raw_spin_lock(&desc->lock); } while ((desc->status & (IRQ_PENDING | IRQ_DISABLED)) == IRQ_PENDING); desc->status &= ~IRQ_INPROGRESS; out_unlock: - spin_unlock(&desc->lock); + raw_spin_unlock(&desc->lock); } /** @@ -643,7 +643,7 @@ __set_irq_handler(unsigned int irq, irq_flow_handler_t handle, int is_chained, } chip_bus_lock(irq, desc); - spin_lock_irqsave(&desc->lock, flags); + raw_spin_lock_irqsave(&desc->lock, flags); /* Uninstall? */ if (handle == handle_bad_irq) { @@ -661,7 +661,7 @@ __set_irq_handler(unsigned int irq, irq_flow_handler_t handle, int is_chained, desc->depth = 0; desc->chip->startup(irq); } - spin_unlock_irqrestore(&desc->lock, flags); + raw_spin_unlock_irqrestore(&desc->lock, flags); chip_bus_sync_unlock(irq, desc); } EXPORT_SYMBOL_GPL(__set_irq_handler); @@ -692,9 +692,9 @@ void __init set_irq_noprobe(unsigned int irq) return; } - spin_lock_irqsave(&desc->lock, flags); + raw_spin_lock_irqsave(&desc->lock, flags); desc->status |= IRQ_NOPROBE; - spin_unlock_irqrestore(&desc->lock, flags); + raw_spin_unlock_irqrestore(&desc->lock, flags); } void __init set_irq_probe(unsigned int irq) @@ -707,7 +707,7 @@ void __init set_irq_probe(unsigned int irq) return; } - spin_lock_irqsave(&desc->lock, flags); + raw_spin_lock_irqsave(&desc->lock, flags); desc->status &= ~IRQ_NOPROBE; - spin_unlock_irqrestore(&desc->lock, flags); + raw_spin_unlock_irqrestore(&desc->lock, flags); } diff --git a/kernel/irq/handle.c b/kernel/irq/handle.c index 17c71bb565c6..814940e7f485 100644 --- a/kernel/irq/handle.c +++ b/kernel/irq/handle.c @@ -80,7 +80,7 @@ static struct irq_desc irq_desc_init = { .chip = &no_irq_chip, .handle_irq = handle_bad_irq, .depth = 1, - .lock = __SPIN_LOCK_UNLOCKED(irq_desc_init.lock), + .lock = __RAW_SPIN_LOCK_UNLOCKED(irq_desc_init.lock), }; void __ref init_kstat_irqs(struct irq_desc *desc, int node, int nr) @@ -108,7 +108,7 @@ static void init_one_irq_desc(int irq, struct irq_desc *desc, int node) { memcpy(desc, &irq_desc_init, sizeof(struct irq_desc)); - spin_lock_init(&desc->lock); + raw_spin_lock_init(&desc->lock); desc->irq = irq; #ifdef CONFIG_SMP desc->node = node; @@ -130,7 +130,7 @@ static void init_one_irq_desc(int irq, struct irq_desc *desc, int node) /* * Protect the sparse_irqs: */ -DEFINE_SPINLOCK(sparse_irq_lock); +DEFINE_RAW_SPINLOCK(sparse_irq_lock); struct irq_desc **irq_desc_ptrs __read_mostly; @@ -141,7 +141,7 @@ static struct irq_desc irq_desc_legacy[NR_IRQS_LEGACY] __cacheline_aligned_in_sm .chip = &no_irq_chip, .handle_irq = handle_bad_irq, .depth = 1, - .lock = __SPIN_LOCK_UNLOCKED(irq_desc_init.lock), + .lock = __RAW_SPIN_LOCK_UNLOCKED(irq_desc_init.lock), } }; @@ -212,7 +212,7 @@ struct irq_desc * __ref irq_to_desc_alloc_node(unsigned int irq, int node) if (desc) return desc; - spin_lock_irqsave(&sparse_irq_lock, flags); + raw_spin_lock_irqsave(&sparse_irq_lock, flags); /* We have to check it to avoid races with another CPU */ desc = irq_desc_ptrs[irq]; @@ -234,7 +234,7 @@ struct irq_desc * __ref irq_to_desc_alloc_node(unsigned int irq, int node) irq_desc_ptrs[irq] = desc; out_unlock: - spin_unlock_irqrestore(&sparse_irq_lock, flags); + raw_spin_unlock_irqrestore(&sparse_irq_lock, flags); return desc; } @@ -247,7 +247,7 @@ struct irq_desc irq_desc[NR_IRQS] __cacheline_aligned_in_smp = { .chip = &no_irq_chip, .handle_irq = handle_bad_irq, .depth = 1, - .lock = __SPIN_LOCK_UNLOCKED(irq_desc->lock), + .lock = __RAW_SPIN_LOCK_UNLOCKED(irq_desc->lock), } }; @@ -473,7 +473,7 @@ unsigned int __do_IRQ(unsigned int irq) return 1; } - spin_lock(&desc->lock); + raw_spin_lock(&desc->lock); if (desc->chip->ack) desc->chip->ack(irq); /* @@ -517,13 +517,13 @@ unsigned int __do_IRQ(unsigned int irq) for (;;) { irqreturn_t action_ret; - spin_unlock(&desc->lock); + raw_spin_unlock(&desc->lock); action_ret = handle_IRQ_event(irq, action); if (!noirqdebug) note_interrupt(irq, desc, action_ret); - spin_lock(&desc->lock); + raw_spin_lock(&desc->lock); if (likely(!(desc->status & IRQ_PENDING))) break; desc->status &= ~IRQ_PENDING; @@ -536,7 +536,7 @@ out: * disabled while the handler was running. */ desc->chip->end(irq); - spin_unlock(&desc->lock); + raw_spin_unlock(&desc->lock); return 1; } diff --git a/kernel/irq/internals.h b/kernel/irq/internals.h index 1b5d742c6a77..b2821f070a3d 100644 --- a/kernel/irq/internals.h +++ b/kernel/irq/internals.h @@ -18,7 +18,7 @@ extern void __enable_irq(struct irq_desc *desc, unsigned int irq, bool resume); extern struct lock_class_key irq_desc_lock_class; extern void init_kstat_irqs(struct irq_desc *desc, int node, int nr); extern void clear_kstat_irqs(struct irq_desc *desc); -extern spinlock_t sparse_irq_lock; +extern raw_spinlock_t sparse_irq_lock; #ifdef CONFIG_SPARSE_IRQ /* irq_desc_ptrs allocated at boot time */ diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c index 7305b297d1eb..eb6078ca60c7 100644 --- a/kernel/irq/manage.c +++ b/kernel/irq/manage.c @@ -46,9 +46,9 @@ void synchronize_irq(unsigned int irq) cpu_relax(); /* Ok, that indicated we're done: double-check carefully. */ - spin_lock_irqsave(&desc->lock, flags); + raw_spin_lock_irqsave(&desc->lock, flags); status = desc->status; - spin_unlock_irqrestore(&desc->lock, flags); + raw_spin_unlock_irqrestore(&desc->lock, flags); /* Oops, that failed? */ } while (status & IRQ_INPROGRESS); @@ -114,7 +114,7 @@ int irq_set_affinity(unsigned int irq, const struct cpumask *cpumask) if (!desc->chip->set_affinity) return -EINVAL; - spin_lock_irqsave(&desc->lock, flags); + raw_spin_lock_irqsave(&desc->lock, flags); #ifdef CONFIG_GENERIC_PENDING_IRQ if (desc->status & IRQ_MOVE_PCNTXT) { @@ -134,7 +134,7 @@ int irq_set_affinity(unsigned int irq, const struct cpumask *cpumask) } #endif desc->status |= IRQ_AFFINITY_SET; - spin_unlock_irqrestore(&desc->lock, flags); + raw_spin_unlock_irqrestore(&desc->lock, flags); return 0; } @@ -181,11 +181,11 @@ int irq_select_affinity_usr(unsigned int irq) unsigned long flags; int ret; - spin_lock_irqsave(&desc->lock, flags); + raw_spin_lock_irqsave(&desc->lock, flags); ret = setup_affinity(irq, desc); if (!ret) irq_set_thread_affinity(desc); - spin_unlock_irqrestore(&desc->lock, flags); + raw_spin_unlock_irqrestore(&desc->lock, flags); return ret; } @@ -231,9 +231,9 @@ void disable_irq_nosync(unsigned int irq) return; chip_bus_lock(irq, desc); - spin_lock_irqsave(&desc->lock, flags); + raw_spin_lock_irqsave(&desc->lock, flags); __disable_irq(desc, irq, false); - spin_unlock_irqrestore(&desc->lock, flags); + raw_spin_unlock_irqrestore(&desc->lock, flags); chip_bus_sync_unlock(irq, desc); } EXPORT_SYMBOL(disable_irq_nosync); @@ -308,9 +308,9 @@ void enable_irq(unsigned int irq) return; chip_bus_lock(irq, desc); - spin_lock_irqsave(&desc->lock, flags); + raw_spin_lock_irqsave(&desc->lock, flags); __enable_irq(desc, irq, false); - spin_unlock_irqrestore(&desc->lock, flags); + raw_spin_unlock_irqrestore(&desc->lock, flags); chip_bus_sync_unlock(irq, desc); } EXPORT_SYMBOL(enable_irq); @@ -347,7 +347,7 @@ int set_irq_wake(unsigned int irq, unsigned int on) /* wakeup-capable irqs can be shared between drivers that * don't need to have the same sleep mode behaviors. */ - spin_lock_irqsave(&desc->lock, flags); + raw_spin_lock_irqsave(&desc->lock, flags); if (on) { if (desc->wake_depth++ == 0) { ret = set_irq_wake_real(irq, on); @@ -368,7 +368,7 @@ int set_irq_wake(unsigned int irq, unsigned int on) } } - spin_unlock_irqrestore(&desc->lock, flags); + raw_spin_unlock_irqrestore(&desc->lock, flags); return ret; } EXPORT_SYMBOL(set_irq_wake); @@ -484,12 +484,12 @@ static int irq_wait_for_interrupt(struct irqaction *action) static void irq_finalize_oneshot(unsigned int irq, struct irq_desc *desc) { chip_bus_lock(irq, desc); - spin_lock_irq(&desc->lock); + raw_spin_lock_irq(&desc->lock); if (!(desc->status & IRQ_DISABLED) && (desc->status & IRQ_MASKED)) { desc->status &= ~IRQ_MASKED; desc->chip->unmask(irq); } - spin_unlock_irq(&desc->lock); + raw_spin_unlock_irq(&desc->lock); chip_bus_sync_unlock(irq, desc); } @@ -514,9 +514,9 @@ irq_thread_check_affinity(struct irq_desc *desc, struct irqaction *action) return; } - spin_lock_irq(&desc->lock); + raw_spin_lock_irq(&desc->lock); cpumask_copy(mask, desc->affinity); - spin_unlock_irq(&desc->lock); + raw_spin_unlock_irq(&desc->lock); set_cpus_allowed_ptr(current, mask); free_cpumask_var(mask); @@ -545,7 +545,7 @@ static int irq_thread(void *data) atomic_inc(&desc->threads_active); - spin_lock_irq(&desc->lock); + raw_spin_lock_irq(&desc->lock); if (unlikely(desc->status & IRQ_DISABLED)) { /* * CHECKME: We might need a dedicated @@ -555,9 +555,9 @@ static int irq_thread(void *data) * retriggers the interrupt itself --- tglx */ desc->status |= IRQ_PENDING; - spin_unlock_irq(&desc->lock); + raw_spin_unlock_irq(&desc->lock); } else { - spin_unlock_irq(&desc->lock); + raw_spin_unlock_irq(&desc->lock); action->thread_fn(action->irq, action->dev_id); @@ -679,7 +679,7 @@ __setup_irq(unsigned int irq, struct irq_desc *desc, struct irqaction *new) /* * The following block of code has to be executed atomically */ - spin_lock_irqsave(&desc->lock, flags); + raw_spin_lock_irqsave(&desc->lock, flags); old_ptr = &desc->action; old = *old_ptr; if (old) { @@ -775,7 +775,7 @@ __setup_irq(unsigned int irq, struct irq_desc *desc, struct irqaction *new) __enable_irq(desc, irq, false); } - spin_unlock_irqrestore(&desc->lock, flags); + raw_spin_unlock_irqrestore(&desc->lock, flags); /* * Strictly no need to wake it up, but hung_task complains @@ -802,7 +802,7 @@ mismatch: ret = -EBUSY; out_thread: - spin_unlock_irqrestore(&desc->lock, flags); + raw_spin_unlock_irqrestore(&desc->lock, flags); if (new->thread) { struct task_struct *t = new->thread; @@ -844,7 +844,7 @@ static struct irqaction *__free_irq(unsigned int irq, void *dev_id) if (!desc) return NULL; - spin_lock_irqsave(&desc->lock, flags); + raw_spin_lock_irqsave(&desc->lock, flags); /* * There can be multiple actions per IRQ descriptor, find the right @@ -856,7 +856,7 @@ static struct irqaction *__free_irq(unsigned int irq, void *dev_id) if (!action) { WARN(1, "Trying to free already-free IRQ %d\n", irq); - spin_unlock_irqrestore(&desc->lock, flags); + raw_spin_unlock_irqrestore(&desc->lock, flags); return NULL; } @@ -884,7 +884,7 @@ static struct irqaction *__free_irq(unsigned int irq, void *dev_id) desc->chip->disable(irq); } - spin_unlock_irqrestore(&desc->lock, flags); + raw_spin_unlock_irqrestore(&desc->lock, flags); unregister_handler_proc(irq, action); diff --git a/kernel/irq/migration.c b/kernel/irq/migration.c index fcb6c96f2627..241962280836 100644 --- a/kernel/irq/migration.c +++ b/kernel/irq/migration.c @@ -27,7 +27,7 @@ void move_masked_irq(int irq) if (!desc->chip->set_affinity) return; - assert_spin_locked(&desc->lock); + assert_raw_spin_locked(&desc->lock); /* * If there was a valid mask to work with, please diff --git a/kernel/irq/numa_migrate.c b/kernel/irq/numa_migrate.c index 3fd30197da2e..26bac9d8f860 100644 --- a/kernel/irq/numa_migrate.c +++ b/kernel/irq/numa_migrate.c @@ -42,7 +42,7 @@ static bool init_copy_one_irq_desc(int irq, struct irq_desc *old_desc, "for migration.\n", irq); return false; } - spin_lock_init(&desc->lock); + raw_spin_lock_init(&desc->lock); desc->node = node; lockdep_set_class(&desc->lock, &irq_desc_lock_class); init_copy_kstat_irqs(old_desc, desc, node, nr_cpu_ids); @@ -67,7 +67,7 @@ static struct irq_desc *__real_move_irq_desc(struct irq_desc *old_desc, irq = old_desc->irq; - spin_lock_irqsave(&sparse_irq_lock, flags); + raw_spin_lock_irqsave(&sparse_irq_lock, flags); /* We have to check it to avoid races with another CPU */ desc = irq_desc_ptrs[irq]; @@ -91,7 +91,7 @@ static struct irq_desc *__real_move_irq_desc(struct irq_desc *old_desc, } irq_desc_ptrs[irq] = desc; - spin_unlock_irqrestore(&sparse_irq_lock, flags); + raw_spin_unlock_irqrestore(&sparse_irq_lock, flags); /* free the old one */ free_one_irq_desc(old_desc, desc); @@ -100,7 +100,7 @@ static struct irq_desc *__real_move_irq_desc(struct irq_desc *old_desc, return desc; out_unlock: - spin_unlock_irqrestore(&sparse_irq_lock, flags); + raw_spin_unlock_irqrestore(&sparse_irq_lock, flags); return desc; } diff --git a/kernel/irq/pm.c b/kernel/irq/pm.c index a0bb09e79867..0d4005d85b03 100644 --- a/kernel/irq/pm.c +++ b/kernel/irq/pm.c @@ -28,9 +28,9 @@ void suspend_device_irqs(void) for_each_irq_desc(irq, desc) { unsigned long flags; - spin_lock_irqsave(&desc->lock, flags); + raw_spin_lock_irqsave(&desc->lock, flags); __disable_irq(desc, irq, true); - spin_unlock_irqrestore(&desc->lock, flags); + raw_spin_unlock_irqrestore(&desc->lock, flags); } for_each_irq_desc(irq, desc) @@ -56,9 +56,9 @@ void resume_device_irqs(void) if (!(desc->status & IRQ_SUSPENDED)) continue; - spin_lock_irqsave(&desc->lock, flags); + raw_spin_lock_irqsave(&desc->lock, flags); __enable_irq(desc, irq, true); - spin_unlock_irqrestore(&desc->lock, flags); + raw_spin_unlock_irqrestore(&desc->lock, flags); } } EXPORT_SYMBOL_GPL(resume_device_irqs); diff --git a/kernel/irq/proc.c b/kernel/irq/proc.c index 0832145fea97..6f50eccc79c0 100644 --- a/kernel/irq/proc.c +++ b/kernel/irq/proc.c @@ -179,7 +179,7 @@ static int name_unique(unsigned int irq, struct irqaction *new_action) unsigned long flags; int ret = 1; - spin_lock_irqsave(&desc->lock, flags); + raw_spin_lock_irqsave(&desc->lock, flags); for (action = desc->action ; action; action = action->next) { if ((action != new_action) && action->name && !strcmp(new_action->name, action->name)) { @@ -187,7 +187,7 @@ static int name_unique(unsigned int irq, struct irqaction *new_action) break; } } - spin_unlock_irqrestore(&desc->lock, flags); + raw_spin_unlock_irqrestore(&desc->lock, flags); return ret; } diff --git a/kernel/irq/spurious.c b/kernel/irq/spurious.c index e49ea1c5232d..89fb90ae534f 100644 --- a/kernel/irq/spurious.c +++ b/kernel/irq/spurious.c @@ -28,7 +28,7 @@ static int try_one_irq(int irq, struct irq_desc *desc) struct irqaction *action; int ok = 0, work = 0; - spin_lock(&desc->lock); + raw_spin_lock(&desc->lock); /* Already running on another processor */ if (desc->status & IRQ_INPROGRESS) { /* @@ -37,13 +37,13 @@ static int try_one_irq(int irq, struct irq_desc *desc) */ if (desc->action && (desc->action->flags & IRQF_SHARED)) desc->status |= IRQ_PENDING; - spin_unlock(&desc->lock); + raw_spin_unlock(&desc->lock); return ok; } /* Honour the normal IRQ locking */ desc->status |= IRQ_INPROGRESS; action = desc->action; - spin_unlock(&desc->lock); + raw_spin_unlock(&desc->lock); while (action) { /* Only shared IRQ handlers are safe to call */ @@ -56,7 +56,7 @@ static int try_one_irq(int irq, struct irq_desc *desc) } local_irq_disable(); /* Now clean up the flags */ - spin_lock(&desc->lock); + raw_spin_lock(&desc->lock); action = desc->action; /* @@ -68,9 +68,9 @@ static int try_one_irq(int irq, struct irq_desc *desc) * Perform real IRQ processing for the IRQ we deferred */ work = 1; - spin_unlock(&desc->lock); + raw_spin_unlock(&desc->lock); handle_IRQ_event(irq, action); - spin_lock(&desc->lock); + raw_spin_lock(&desc->lock); desc->status &= ~IRQ_PENDING; } desc->status &= ~IRQ_INPROGRESS; @@ -80,7 +80,7 @@ static int try_one_irq(int irq, struct irq_desc *desc) */ if (work && desc->chip && desc->chip->end) desc->chip->end(irq); - spin_unlock(&desc->lock); + raw_spin_unlock(&desc->lock); return ok; } -- cgit v1.2.3-71-gd317 From ecb49d1a639acbacfc3771cae5ec07bed5df3847 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 17 Nov 2009 16:36:54 +0100 Subject: hrtimers: Convert to raw_spinlocks Convert locks which cannot be sleeping locks in preempt-rt to raw_spinlocks. Signed-off-by: Thomas Gleixner Acked-by: Peter Zijlstra Acked-by: Ingo Molnar --- include/linux/hrtimer.h | 2 +- kernel/hrtimer.c | 50 +++++++++++++++++++++++------------------------ kernel/time/timer_list.c | 6 +++--- kernel/time/timer_stats.c | 17 ++++++++-------- 4 files changed, 38 insertions(+), 37 deletions(-) (limited to 'kernel') diff --git a/include/linux/hrtimer.h b/include/linux/hrtimer.h index af634e95871d..5d86fb2309d2 100644 --- a/include/linux/hrtimer.h +++ b/include/linux/hrtimer.h @@ -169,7 +169,7 @@ struct hrtimer_clock_base { * @max_hang_time: Maximum time spent in hrtimer_interrupt */ struct hrtimer_cpu_base { - spinlock_t lock; + raw_spinlock_t lock; struct hrtimer_clock_base clock_base[HRTIMER_MAX_CLOCK_BASES]; #ifdef CONFIG_HIGH_RES_TIMERS ktime_t expires_next; diff --git a/kernel/hrtimer.c b/kernel/hrtimer.c index d2f9239dc6ba..0086628b6e97 100644 --- a/kernel/hrtimer.c +++ b/kernel/hrtimer.c @@ -127,11 +127,11 @@ struct hrtimer_clock_base *lock_hrtimer_base(const struct hrtimer *timer, for (;;) { base = timer->base; if (likely(base != NULL)) { - spin_lock_irqsave(&base->cpu_base->lock, *flags); + raw_spin_lock_irqsave(&base->cpu_base->lock, *flags); if (likely(base == timer->base)) return base; /* The timer has migrated to another CPU: */ - spin_unlock_irqrestore(&base->cpu_base->lock, *flags); + raw_spin_unlock_irqrestore(&base->cpu_base->lock, *flags); } cpu_relax(); } @@ -208,13 +208,13 @@ again: /* See the comment in lock_timer_base() */ timer->base = NULL; - spin_unlock(&base->cpu_base->lock); - spin_lock(&new_base->cpu_base->lock); + raw_spin_unlock(&base->cpu_base->lock); + raw_spin_lock(&new_base->cpu_base->lock); if (cpu != this_cpu && hrtimer_check_target(timer, new_base)) { cpu = this_cpu; - spin_unlock(&new_base->cpu_base->lock); - spin_lock(&base->cpu_base->lock); + raw_spin_unlock(&new_base->cpu_base->lock); + raw_spin_lock(&base->cpu_base->lock); timer->base = base; goto again; } @@ -230,7 +230,7 @@ lock_hrtimer_base(const struct hrtimer *timer, unsigned long *flags) { struct hrtimer_clock_base *base = timer->base; - spin_lock_irqsave(&base->cpu_base->lock, *flags); + raw_spin_lock_irqsave(&base->cpu_base->lock, *flags); return base; } @@ -628,12 +628,12 @@ static void retrigger_next_event(void *arg) base = &__get_cpu_var(hrtimer_bases); /* Adjust CLOCK_REALTIME offset */ - spin_lock(&base->lock); + raw_spin_lock(&base->lock); base->clock_base[CLOCK_REALTIME].offset = timespec_to_ktime(realtime_offset); hrtimer_force_reprogram(base, 0); - spin_unlock(&base->lock); + raw_spin_unlock(&base->lock); } /* @@ -694,9 +694,9 @@ static inline int hrtimer_enqueue_reprogram(struct hrtimer *timer, { if (base->cpu_base->hres_active && hrtimer_reprogram(timer, base)) { if (wakeup) { - spin_unlock(&base->cpu_base->lock); + raw_spin_unlock(&base->cpu_base->lock); raise_softirq_irqoff(HRTIMER_SOFTIRQ); - spin_lock(&base->cpu_base->lock); + raw_spin_lock(&base->cpu_base->lock); } else __raise_softirq_irqoff(HRTIMER_SOFTIRQ); @@ -790,7 +790,7 @@ static inline void timer_stats_account_hrtimer(struct hrtimer *timer) static inline void unlock_hrtimer_base(const struct hrtimer *timer, unsigned long *flags) { - spin_unlock_irqrestore(&timer->base->cpu_base->lock, *flags); + raw_spin_unlock_irqrestore(&timer->base->cpu_base->lock, *flags); } /** @@ -1123,7 +1123,7 @@ ktime_t hrtimer_get_next_event(void) unsigned long flags; int i; - spin_lock_irqsave(&cpu_base->lock, flags); + raw_spin_lock_irqsave(&cpu_base->lock, flags); if (!hrtimer_hres_active()) { for (i = 0; i < HRTIMER_MAX_CLOCK_BASES; i++, base++) { @@ -1140,7 +1140,7 @@ ktime_t hrtimer_get_next_event(void) } } - spin_unlock_irqrestore(&cpu_base->lock, flags); + raw_spin_unlock_irqrestore(&cpu_base->lock, flags); if (mindelta.tv64 < 0) mindelta.tv64 = 0; @@ -1222,11 +1222,11 @@ static void __run_hrtimer(struct hrtimer *timer, ktime_t *now) * they get migrated to another cpu, therefore its safe to unlock * the timer base. */ - spin_unlock(&cpu_base->lock); + raw_spin_unlock(&cpu_base->lock); trace_hrtimer_expire_entry(timer, now); restart = fn(timer); trace_hrtimer_expire_exit(timer); - spin_lock(&cpu_base->lock); + raw_spin_lock(&cpu_base->lock); /* * Note: We clear the CALLBACK bit after enqueue_hrtimer and @@ -1261,7 +1261,7 @@ void hrtimer_interrupt(struct clock_event_device *dev) retry: expires_next.tv64 = KTIME_MAX; - spin_lock(&cpu_base->lock); + raw_spin_lock(&cpu_base->lock); /* * We set expires_next to KTIME_MAX here with cpu_base->lock * held to prevent that a timer is enqueued in our queue via @@ -1317,7 +1317,7 @@ retry: * against it. */ cpu_base->expires_next = expires_next; - spin_unlock(&cpu_base->lock); + raw_spin_unlock(&cpu_base->lock); /* Reprogramming necessary ? */ if (expires_next.tv64 == KTIME_MAX || @@ -1457,7 +1457,7 @@ void hrtimer_run_queues(void) gettime = 0; } - spin_lock(&cpu_base->lock); + raw_spin_lock(&cpu_base->lock); while ((node = base->first)) { struct hrtimer *timer; @@ -1469,7 +1469,7 @@ void hrtimer_run_queues(void) __run_hrtimer(timer, &base->softirq_time); } - spin_unlock(&cpu_base->lock); + raw_spin_unlock(&cpu_base->lock); } } @@ -1625,7 +1625,7 @@ static void __cpuinit init_hrtimers_cpu(int cpu) struct hrtimer_cpu_base *cpu_base = &per_cpu(hrtimer_bases, cpu); int i; - spin_lock_init(&cpu_base->lock); + raw_spin_lock_init(&cpu_base->lock); for (i = 0; i < HRTIMER_MAX_CLOCK_BASES; i++) cpu_base->clock_base[i].cpu_base = cpu_base; @@ -1683,16 +1683,16 @@ static void migrate_hrtimers(int scpu) * The caller is globally serialized and nobody else * takes two locks at once, deadlock is not possible. */ - spin_lock(&new_base->lock); - spin_lock_nested(&old_base->lock, SINGLE_DEPTH_NESTING); + raw_spin_lock(&new_base->lock); + raw_spin_lock_nested(&old_base->lock, SINGLE_DEPTH_NESTING); for (i = 0; i < HRTIMER_MAX_CLOCK_BASES; i++) { migrate_hrtimer_list(&old_base->clock_base[i], &new_base->clock_base[i]); } - spin_unlock(&old_base->lock); - spin_unlock(&new_base->lock); + raw_spin_unlock(&old_base->lock); + raw_spin_unlock(&new_base->lock); /* Check, if we got expired work to do */ __hrtimer_peek_ahead_timers(); diff --git a/kernel/time/timer_list.c b/kernel/time/timer_list.c index 9d80db4747d4..28265636b6c2 100644 --- a/kernel/time/timer_list.c +++ b/kernel/time/timer_list.c @@ -84,7 +84,7 @@ print_active_timers(struct seq_file *m, struct hrtimer_clock_base *base, next_one: i = 0; - spin_lock_irqsave(&base->cpu_base->lock, flags); + raw_spin_lock_irqsave(&base->cpu_base->lock, flags); curr = base->first; /* @@ -100,13 +100,13 @@ next_one: timer = rb_entry(curr, struct hrtimer, node); tmp = *timer; - spin_unlock_irqrestore(&base->cpu_base->lock, flags); + raw_spin_unlock_irqrestore(&base->cpu_base->lock, flags); print_timer(m, timer, &tmp, i, now); next++; goto next_one; } - spin_unlock_irqrestore(&base->cpu_base->lock, flags); + raw_spin_unlock_irqrestore(&base->cpu_base->lock, flags); } static void diff --git a/kernel/time/timer_stats.c b/kernel/time/timer_stats.c index 63b117e9eba1..2f3b585b8d7d 100644 --- a/kernel/time/timer_stats.c +++ b/kernel/time/timer_stats.c @@ -86,7 +86,7 @@ static DEFINE_SPINLOCK(table_lock); /* * Per-CPU lookup locks for fast hash lookup: */ -static DEFINE_PER_CPU(spinlock_t, tstats_lookup_lock); +static DEFINE_PER_CPU(raw_spinlock_t, tstats_lookup_lock); /* * Mutex to serialize state changes with show-stats activities: @@ -238,7 +238,7 @@ void timer_stats_update_stats(void *timer, pid_t pid, void *startf, /* * It doesnt matter which lock we take: */ - spinlock_t *lock; + raw_spinlock_t *lock; struct entry *entry, input; unsigned long flags; @@ -253,7 +253,7 @@ void timer_stats_update_stats(void *timer, pid_t pid, void *startf, input.pid = pid; input.timer_flag = timer_flag; - spin_lock_irqsave(lock, flags); + raw_spin_lock_irqsave(lock, flags); if (!timer_stats_active) goto out_unlock; @@ -264,7 +264,7 @@ void timer_stats_update_stats(void *timer, pid_t pid, void *startf, atomic_inc(&overflow_count); out_unlock: - spin_unlock_irqrestore(lock, flags); + raw_spin_unlock_irqrestore(lock, flags); } static void print_name_offset(struct seq_file *m, unsigned long addr) @@ -348,10 +348,11 @@ static void sync_access(void) int cpu; for_each_online_cpu(cpu) { - spinlock_t *lock = &per_cpu(tstats_lookup_lock, cpu); - spin_lock_irqsave(lock, flags); + raw_spinlock_t *lock = &per_cpu(tstats_lookup_lock, cpu); + + raw_spin_lock_irqsave(lock, flags); /* nothing */ - spin_unlock_irqrestore(lock, flags); + raw_spin_unlock_irqrestore(lock, flags); } } @@ -409,7 +410,7 @@ void __init init_timer_stats(void) int cpu; for_each_possible_cpu(cpu) - spin_lock_init(&per_cpu(tstats_lookup_lock, cpu)); + raw_spin_lock_init(&per_cpu(tstats_lookup_lock, cpu)); } static int __init init_tstats_procfs(void) -- cgit v1.2.3-71-gd317 From e625cce1b73fb38b74e5387226534f7bcbfc36fe Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 17 Nov 2009 18:02:06 +0100 Subject: perf_event: Convert to raw_spinlock Convert locks which cannot be sleeping locks in preempt-rt to raw_spinlocks. Signed-off-by: Thomas Gleixner Acked-by: Peter Zijlstra Acked-by: Ingo Molnar --- include/linux/perf_event.h | 2 +- kernel/hw_breakpoint.c | 4 +- kernel/perf_event.c | 106 ++++++++++++++++++++++----------------------- 3 files changed, 56 insertions(+), 56 deletions(-) (limited to 'kernel') diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h index 64a53f74c9a9..da7bdc23f279 100644 --- a/include/linux/perf_event.h +++ b/include/linux/perf_event.h @@ -681,7 +681,7 @@ struct perf_event_context { * Protect the states of the events in the list, * nr_active, and the list: */ - spinlock_t lock; + raw_spinlock_t lock; /* * Protect the list of events. Locking either mutex or lock * is sufficient to ensure the list doesn't change; to change diff --git a/kernel/hw_breakpoint.c b/kernel/hw_breakpoint.c index 366eedf949c0..dbcbf6a33a08 100644 --- a/kernel/hw_breakpoint.c +++ b/kernel/hw_breakpoint.c @@ -96,7 +96,7 @@ static int task_bp_pinned(struct task_struct *tsk) list = &ctx->event_list; - spin_lock_irqsave(&ctx->lock, flags); + raw_spin_lock_irqsave(&ctx->lock, flags); /* * The current breakpoint counter is not included in the list @@ -107,7 +107,7 @@ static int task_bp_pinned(struct task_struct *tsk) count++; } - spin_unlock_irqrestore(&ctx->lock, flags); + raw_spin_unlock_irqrestore(&ctx->lock, flags); return count; } diff --git a/kernel/perf_event.c b/kernel/perf_event.c index e73e53c7582f..9052d6c8c9fd 100644 --- a/kernel/perf_event.c +++ b/kernel/perf_event.c @@ -203,14 +203,14 @@ perf_lock_task_context(struct task_struct *task, unsigned long *flags) * if so. If we locked the right context, then it * can't get swapped on us any more. */ - spin_lock_irqsave(&ctx->lock, *flags); + raw_spin_lock_irqsave(&ctx->lock, *flags); if (ctx != rcu_dereference(task->perf_event_ctxp)) { - spin_unlock_irqrestore(&ctx->lock, *flags); + raw_spin_unlock_irqrestore(&ctx->lock, *flags); goto retry; } if (!atomic_inc_not_zero(&ctx->refcount)) { - spin_unlock_irqrestore(&ctx->lock, *flags); + raw_spin_unlock_irqrestore(&ctx->lock, *flags); ctx = NULL; } } @@ -231,7 +231,7 @@ static struct perf_event_context *perf_pin_task_context(struct task_struct *task ctx = perf_lock_task_context(task, &flags); if (ctx) { ++ctx->pin_count; - spin_unlock_irqrestore(&ctx->lock, flags); + raw_spin_unlock_irqrestore(&ctx->lock, flags); } return ctx; } @@ -240,9 +240,9 @@ static void perf_unpin_context(struct perf_event_context *ctx) { unsigned long flags; - spin_lock_irqsave(&ctx->lock, flags); + raw_spin_lock_irqsave(&ctx->lock, flags); --ctx->pin_count; - spin_unlock_irqrestore(&ctx->lock, flags); + raw_spin_unlock_irqrestore(&ctx->lock, flags); put_ctx(ctx); } @@ -427,7 +427,7 @@ static void __perf_event_remove_from_context(void *info) if (ctx->task && cpuctx->task_ctx != ctx) return; - spin_lock(&ctx->lock); + raw_spin_lock(&ctx->lock); /* * Protect the list operation against NMI by disabling the * events on a global level. @@ -449,7 +449,7 @@ static void __perf_event_remove_from_context(void *info) } perf_enable(); - spin_unlock(&ctx->lock); + raw_spin_unlock(&ctx->lock); } @@ -488,12 +488,12 @@ retry: task_oncpu_function_call(task, __perf_event_remove_from_context, event); - spin_lock_irq(&ctx->lock); + raw_spin_lock_irq(&ctx->lock); /* * If the context is active we need to retry the smp call. */ if (ctx->nr_active && !list_empty(&event->group_entry)) { - spin_unlock_irq(&ctx->lock); + raw_spin_unlock_irq(&ctx->lock); goto retry; } @@ -504,7 +504,7 @@ retry: */ if (!list_empty(&event->group_entry)) list_del_event(event, ctx); - spin_unlock_irq(&ctx->lock); + raw_spin_unlock_irq(&ctx->lock); } /* @@ -535,7 +535,7 @@ static void __perf_event_disable(void *info) if (ctx->task && cpuctx->task_ctx != ctx) return; - spin_lock(&ctx->lock); + raw_spin_lock(&ctx->lock); /* * If the event is on, turn it off. @@ -551,7 +551,7 @@ static void __perf_event_disable(void *info) event->state = PERF_EVENT_STATE_OFF; } - spin_unlock(&ctx->lock); + raw_spin_unlock(&ctx->lock); } /* @@ -584,12 +584,12 @@ void perf_event_disable(struct perf_event *event) retry: task_oncpu_function_call(task, __perf_event_disable, event); - spin_lock_irq(&ctx->lock); + raw_spin_lock_irq(&ctx->lock); /* * If the event is still active, we need to retry the cross-call. */ if (event->state == PERF_EVENT_STATE_ACTIVE) { - spin_unlock_irq(&ctx->lock); + raw_spin_unlock_irq(&ctx->lock); goto retry; } @@ -602,7 +602,7 @@ void perf_event_disable(struct perf_event *event) event->state = PERF_EVENT_STATE_OFF; } - spin_unlock_irq(&ctx->lock); + raw_spin_unlock_irq(&ctx->lock); } static int @@ -770,7 +770,7 @@ static void __perf_install_in_context(void *info) cpuctx->task_ctx = ctx; } - spin_lock(&ctx->lock); + raw_spin_lock(&ctx->lock); ctx->is_active = 1; update_context_time(ctx); @@ -820,7 +820,7 @@ static void __perf_install_in_context(void *info) unlock: perf_enable(); - spin_unlock(&ctx->lock); + raw_spin_unlock(&ctx->lock); } /* @@ -856,12 +856,12 @@ retry: task_oncpu_function_call(task, __perf_install_in_context, event); - spin_lock_irq(&ctx->lock); + raw_spin_lock_irq(&ctx->lock); /* * we need to retry the smp call. */ if (ctx->is_active && list_empty(&event->group_entry)) { - spin_unlock_irq(&ctx->lock); + raw_spin_unlock_irq(&ctx->lock); goto retry; } @@ -872,7 +872,7 @@ retry: */ if (list_empty(&event->group_entry)) add_event_to_ctx(event, ctx); - spin_unlock_irq(&ctx->lock); + raw_spin_unlock_irq(&ctx->lock); } /* @@ -917,7 +917,7 @@ static void __perf_event_enable(void *info) cpuctx->task_ctx = ctx; } - spin_lock(&ctx->lock); + raw_spin_lock(&ctx->lock); ctx->is_active = 1; update_context_time(ctx); @@ -959,7 +959,7 @@ static void __perf_event_enable(void *info) } unlock: - spin_unlock(&ctx->lock); + raw_spin_unlock(&ctx->lock); } /* @@ -985,7 +985,7 @@ void perf_event_enable(struct perf_event *event) return; } - spin_lock_irq(&ctx->lock); + raw_spin_lock_irq(&ctx->lock); if (event->state >= PERF_EVENT_STATE_INACTIVE) goto out; @@ -1000,10 +1000,10 @@ void perf_event_enable(struct perf_event *event) event->state = PERF_EVENT_STATE_OFF; retry: - spin_unlock_irq(&ctx->lock); + raw_spin_unlock_irq(&ctx->lock); task_oncpu_function_call(task, __perf_event_enable, event); - spin_lock_irq(&ctx->lock); + raw_spin_lock_irq(&ctx->lock); /* * If the context is active and the event is still off, @@ -1020,7 +1020,7 @@ void perf_event_enable(struct perf_event *event) __perf_event_mark_enabled(event, ctx); out: - spin_unlock_irq(&ctx->lock); + raw_spin_unlock_irq(&ctx->lock); } static int perf_event_refresh(struct perf_event *event, int refresh) @@ -1042,7 +1042,7 @@ void __perf_event_sched_out(struct perf_event_context *ctx, { struct perf_event *event; - spin_lock(&ctx->lock); + raw_spin_lock(&ctx->lock); ctx->is_active = 0; if (likely(!ctx->nr_events)) goto out; @@ -1055,7 +1055,7 @@ void __perf_event_sched_out(struct perf_event_context *ctx, } perf_enable(); out: - spin_unlock(&ctx->lock); + raw_spin_unlock(&ctx->lock); } /* @@ -1193,8 +1193,8 @@ void perf_event_task_sched_out(struct task_struct *task, * order we take the locks because no other cpu could * be trying to lock both of these tasks. */ - spin_lock(&ctx->lock); - spin_lock_nested(&next_ctx->lock, SINGLE_DEPTH_NESTING); + raw_spin_lock(&ctx->lock); + raw_spin_lock_nested(&next_ctx->lock, SINGLE_DEPTH_NESTING); if (context_equiv(ctx, next_ctx)) { /* * XXX do we need a memory barrier of sorts @@ -1208,8 +1208,8 @@ void perf_event_task_sched_out(struct task_struct *task, perf_event_sync_stat(ctx, next_ctx); } - spin_unlock(&next_ctx->lock); - spin_unlock(&ctx->lock); + raw_spin_unlock(&next_ctx->lock); + raw_spin_unlock(&ctx->lock); } rcu_read_unlock(); @@ -1251,7 +1251,7 @@ __perf_event_sched_in(struct perf_event_context *ctx, struct perf_event *event; int can_add_hw = 1; - spin_lock(&ctx->lock); + raw_spin_lock(&ctx->lock); ctx->is_active = 1; if (likely(!ctx->nr_events)) goto out; @@ -1306,7 +1306,7 @@ __perf_event_sched_in(struct perf_event_context *ctx, } perf_enable(); out: - spin_unlock(&ctx->lock); + raw_spin_unlock(&ctx->lock); } /* @@ -1370,7 +1370,7 @@ static void perf_ctx_adjust_freq(struct perf_event_context *ctx) struct hw_perf_event *hwc; u64 interrupts, freq; - spin_lock(&ctx->lock); + raw_spin_lock(&ctx->lock); list_for_each_entry_rcu(event, &ctx->event_list, event_entry) { if (event->state != PERF_EVENT_STATE_ACTIVE) continue; @@ -1425,7 +1425,7 @@ static void perf_ctx_adjust_freq(struct perf_event_context *ctx) perf_enable(); } } - spin_unlock(&ctx->lock); + raw_spin_unlock(&ctx->lock); } /* @@ -1438,7 +1438,7 @@ static void rotate_ctx(struct perf_event_context *ctx) if (!ctx->nr_events) return; - spin_lock(&ctx->lock); + raw_spin_lock(&ctx->lock); /* * Rotate the first entry last (works just fine for group events too): */ @@ -1449,7 +1449,7 @@ static void rotate_ctx(struct perf_event_context *ctx) } perf_enable(); - spin_unlock(&ctx->lock); + raw_spin_unlock(&ctx->lock); } void perf_event_task_tick(struct task_struct *curr, int cpu) @@ -1498,7 +1498,7 @@ static void perf_event_enable_on_exec(struct task_struct *task) __perf_event_task_sched_out(ctx); - spin_lock(&ctx->lock); + raw_spin_lock(&ctx->lock); list_for_each_entry(event, &ctx->group_list, group_entry) { if (!event->attr.enable_on_exec) @@ -1516,7 +1516,7 @@ static void perf_event_enable_on_exec(struct task_struct *task) if (enabled) unclone_ctx(ctx); - spin_unlock(&ctx->lock); + raw_spin_unlock(&ctx->lock); perf_event_task_sched_in(task, smp_processor_id()); out: @@ -1542,10 +1542,10 @@ static void __perf_event_read(void *info) if (ctx->task && cpuctx->task_ctx != ctx) return; - spin_lock(&ctx->lock); + raw_spin_lock(&ctx->lock); update_context_time(ctx); update_event_times(event); - spin_unlock(&ctx->lock); + raw_spin_unlock(&ctx->lock); event->pmu->read(event); } @@ -1563,10 +1563,10 @@ static u64 perf_event_read(struct perf_event *event) struct perf_event_context *ctx = event->ctx; unsigned long flags; - spin_lock_irqsave(&ctx->lock, flags); + raw_spin_lock_irqsave(&ctx->lock, flags); update_context_time(ctx); update_event_times(event); - spin_unlock_irqrestore(&ctx->lock, flags); + raw_spin_unlock_irqrestore(&ctx->lock, flags); } return atomic64_read(&event->count); @@ -1579,7 +1579,7 @@ static void __perf_event_init_context(struct perf_event_context *ctx, struct task_struct *task) { - spin_lock_init(&ctx->lock); + raw_spin_lock_init(&ctx->lock); mutex_init(&ctx->mutex); INIT_LIST_HEAD(&ctx->group_list); INIT_LIST_HEAD(&ctx->event_list); @@ -1649,7 +1649,7 @@ static struct perf_event_context *find_get_context(pid_t pid, int cpu) ctx = perf_lock_task_context(task, &flags); if (ctx) { unclone_ctx(ctx); - spin_unlock_irqrestore(&ctx->lock, flags); + raw_spin_unlock_irqrestore(&ctx->lock, flags); } if (!ctx) { @@ -1987,7 +1987,7 @@ static int perf_event_period(struct perf_event *event, u64 __user *arg) if (!value) return -EINVAL; - spin_lock_irq(&ctx->lock); + raw_spin_lock_irq(&ctx->lock); if (event->attr.freq) { if (value > sysctl_perf_event_sample_rate) { ret = -EINVAL; @@ -2000,7 +2000,7 @@ static int perf_event_period(struct perf_event *event, u64 __user *arg) event->hw.sample_period = value; } unlock: - spin_unlock_irq(&ctx->lock); + raw_spin_unlock_irq(&ctx->lock); return ret; } @@ -4992,7 +4992,7 @@ void perf_event_exit_task(struct task_struct *child) * reading child->perf_event_ctxp, we wait until it has * incremented the context's refcount before we do put_ctx below. */ - spin_lock(&child_ctx->lock); + raw_spin_lock(&child_ctx->lock); child->perf_event_ctxp = NULL; /* * If this context is a clone; unclone it so it can't get @@ -5001,7 +5001,7 @@ void perf_event_exit_task(struct task_struct *child) */ unclone_ctx(child_ctx); update_context_time(child_ctx); - spin_unlock_irqrestore(&child_ctx->lock, flags); + raw_spin_unlock_irqrestore(&child_ctx->lock, flags); /* * Report the task dead after unscheduling the events so that we @@ -5292,11 +5292,11 @@ perf_set_reserve_percpu(struct sysdev_class *class, perf_reserved_percpu = val; for_each_online_cpu(cpu) { cpuctx = &per_cpu(perf_cpu_context, cpu); - spin_lock_irq(&cpuctx->ctx.lock); + raw_spin_lock_irq(&cpuctx->ctx.lock); mpt = min(perf_max_events - cpuctx->ctx.nr_events, perf_max_events - perf_reserved_percpu); cpuctx->max_pertask = mpt; - spin_unlock_irq(&cpuctx->ctx.lock); + raw_spin_unlock_irq(&cpuctx->ctx.lock); } spin_unlock(&perf_resource_lock); -- cgit v1.2.3-71-gd317 From d192c47f25daa98996c7eae543d8a27e41247ec2 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 8 Dec 2009 12:49:26 +0100 Subject: clockevents: Make tick_device_lock static Signed-off-by: Thomas Gleixner Acked-by: Peter Zijlstra Acked-by: Ingo Molnar --- kernel/time/tick-common.c | 2 +- kernel/time/tick-internal.h | 1 - 2 files changed, 1 insertion(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/time/tick-common.c b/kernel/time/tick-common.c index 83c4417b6a3c..af39cf1cfa50 100644 --- a/kernel/time/tick-common.c +++ b/kernel/time/tick-common.c @@ -34,7 +34,7 @@ DEFINE_PER_CPU(struct tick_device, tick_cpu_device); ktime_t tick_next_period; ktime_t tick_period; int tick_do_timer_cpu __read_mostly = TICK_DO_TIMER_BOOT; -DEFINE_SPINLOCK(tick_device_lock); +static DEFINE_SPINLOCK(tick_device_lock); /* * Debugging: see timer_list.c diff --git a/kernel/time/tick-internal.h b/kernel/time/tick-internal.h index b1c05bf75ee0..290eefbc1f60 100644 --- a/kernel/time/tick-internal.h +++ b/kernel/time/tick-internal.h @@ -6,7 +6,6 @@ #define TICK_DO_TIMER_BOOT -2 DECLARE_PER_CPU(struct tick_device, tick_cpu_device); -extern spinlock_t tick_device_lock; extern ktime_t tick_next_period; extern ktime_t tick_period; extern int tick_do_timer_cpu __read_mostly; -- cgit v1.2.3-71-gd317 From b5f91da0a6973bb6f9ff3b91b0e92c0773a458f3 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 8 Dec 2009 12:40:31 +0100 Subject: clockevents: Convert to raw_spinlock Convert locks which cannot be sleeping locks in preempt-rt to raw_spinlocks. Signed-off-by: Thomas Gleixner Acked-by: Peter Zijlstra Acked-by: Ingo Molnar --- kernel/time/clockevents.c | 14 +++++++------- kernel/time/tick-broadcast.c | 42 +++++++++++++++++++++--------------------- kernel/time/tick-common.c | 20 ++++++++++---------- 3 files changed, 38 insertions(+), 38 deletions(-) (limited to 'kernel') diff --git a/kernel/time/clockevents.c b/kernel/time/clockevents.c index 20a8920029ee..3d5fc0fd1cca 100644 --- a/kernel/time/clockevents.c +++ b/kernel/time/clockevents.c @@ -30,7 +30,7 @@ static LIST_HEAD(clockevents_released); static RAW_NOTIFIER_HEAD(clockevents_chain); /* Protection for the above */ -static DEFINE_SPINLOCK(clockevents_lock); +static DEFINE_RAW_SPINLOCK(clockevents_lock); /** * clockevents_delta2ns - Convert a latch value (device ticks) to nanoseconds @@ -141,9 +141,9 @@ int clockevents_register_notifier(struct notifier_block *nb) unsigned long flags; int ret; - spin_lock_irqsave(&clockevents_lock, flags); + raw_spin_lock_irqsave(&clockevents_lock, flags); ret = raw_notifier_chain_register(&clockevents_chain, nb); - spin_unlock_irqrestore(&clockevents_lock, flags); + raw_spin_unlock_irqrestore(&clockevents_lock, flags); return ret; } @@ -185,13 +185,13 @@ void clockevents_register_device(struct clock_event_device *dev) BUG_ON(dev->mode != CLOCK_EVT_MODE_UNUSED); BUG_ON(!dev->cpumask); - spin_lock_irqsave(&clockevents_lock, flags); + raw_spin_lock_irqsave(&clockevents_lock, flags); list_add(&dev->list, &clockevent_devices); clockevents_do_notify(CLOCK_EVT_NOTIFY_ADD, dev); clockevents_notify_released(); - spin_unlock_irqrestore(&clockevents_lock, flags); + raw_spin_unlock_irqrestore(&clockevents_lock, flags); } EXPORT_SYMBOL_GPL(clockevents_register_device); @@ -241,7 +241,7 @@ void clockevents_notify(unsigned long reason, void *arg) struct list_head *node, *tmp; unsigned long flags; - spin_lock_irqsave(&clockevents_lock, flags); + raw_spin_lock_irqsave(&clockevents_lock, flags); clockevents_do_notify(reason, arg); switch (reason) { @@ -256,7 +256,7 @@ void clockevents_notify(unsigned long reason, void *arg) default: break; } - spin_unlock_irqrestore(&clockevents_lock, flags); + raw_spin_unlock_irqrestore(&clockevents_lock, flags); } EXPORT_SYMBOL_GPL(clockevents_notify); #endif diff --git a/kernel/time/tick-broadcast.c b/kernel/time/tick-broadcast.c index c2ec25087a35..b3bafd5fc66d 100644 --- a/kernel/time/tick-broadcast.c +++ b/kernel/time/tick-broadcast.c @@ -31,7 +31,7 @@ static struct tick_device tick_broadcast_device; /* FIXME: Use cpumask_var_t. */ static DECLARE_BITMAP(tick_broadcast_mask, NR_CPUS); static DECLARE_BITMAP(tmpmask, NR_CPUS); -static DEFINE_SPINLOCK(tick_broadcast_lock); +static DEFINE_RAW_SPINLOCK(tick_broadcast_lock); static int tick_broadcast_force; #ifdef CONFIG_TICK_ONESHOT @@ -96,7 +96,7 @@ int tick_device_uses_broadcast(struct clock_event_device *dev, int cpu) unsigned long flags; int ret = 0; - spin_lock_irqsave(&tick_broadcast_lock, flags); + raw_spin_lock_irqsave(&tick_broadcast_lock, flags); /* * Devices might be registered with both periodic and oneshot @@ -122,7 +122,7 @@ int tick_device_uses_broadcast(struct clock_event_device *dev, int cpu) tick_broadcast_clear_oneshot(cpu); } } - spin_unlock_irqrestore(&tick_broadcast_lock, flags); + raw_spin_unlock_irqrestore(&tick_broadcast_lock, flags); return ret; } @@ -161,13 +161,13 @@ static void tick_do_broadcast(struct cpumask *mask) */ static void tick_do_periodic_broadcast(void) { - spin_lock(&tick_broadcast_lock); + raw_spin_lock(&tick_broadcast_lock); cpumask_and(to_cpumask(tmpmask), cpu_online_mask, tick_get_broadcast_mask()); tick_do_broadcast(to_cpumask(tmpmask)); - spin_unlock(&tick_broadcast_lock); + raw_spin_unlock(&tick_broadcast_lock); } /* @@ -212,7 +212,7 @@ static void tick_do_broadcast_on_off(unsigned long *reason) unsigned long flags; int cpu, bc_stopped; - spin_lock_irqsave(&tick_broadcast_lock, flags); + raw_spin_lock_irqsave(&tick_broadcast_lock, flags); cpu = smp_processor_id(); td = &per_cpu(tick_cpu_device, cpu); @@ -263,7 +263,7 @@ static void tick_do_broadcast_on_off(unsigned long *reason) tick_broadcast_setup_oneshot(bc); } out: - spin_unlock_irqrestore(&tick_broadcast_lock, flags); + raw_spin_unlock_irqrestore(&tick_broadcast_lock, flags); } /* @@ -299,7 +299,7 @@ void tick_shutdown_broadcast(unsigned int *cpup) unsigned long flags; unsigned int cpu = *cpup; - spin_lock_irqsave(&tick_broadcast_lock, flags); + raw_spin_lock_irqsave(&tick_broadcast_lock, flags); bc = tick_broadcast_device.evtdev; cpumask_clear_cpu(cpu, tick_get_broadcast_mask()); @@ -309,7 +309,7 @@ void tick_shutdown_broadcast(unsigned int *cpup) clockevents_shutdown(bc); } - spin_unlock_irqrestore(&tick_broadcast_lock, flags); + raw_spin_unlock_irqrestore(&tick_broadcast_lock, flags); } void tick_suspend_broadcast(void) @@ -317,13 +317,13 @@ void tick_suspend_broadcast(void) struct clock_event_device *bc; unsigned long flags; - spin_lock_irqsave(&tick_broadcast_lock, flags); + raw_spin_lock_irqsave(&tick_broadcast_lock, flags); bc = tick_broadcast_device.evtdev; if (bc) clockevents_shutdown(bc); - spin_unlock_irqrestore(&tick_broadcast_lock, flags); + raw_spin_unlock_irqrestore(&tick_broadcast_lock, flags); } int tick_resume_broadcast(void) @@ -332,7 +332,7 @@ int tick_resume_broadcast(void) unsigned long flags; int broadcast = 0; - spin_lock_irqsave(&tick_broadcast_lock, flags); + raw_spin_lock_irqsave(&tick_broadcast_lock, flags); bc = tick_broadcast_device.evtdev; @@ -351,7 +351,7 @@ int tick_resume_broadcast(void) break; } } - spin_unlock_irqrestore(&tick_broadcast_lock, flags); + raw_spin_unlock_irqrestore(&tick_broadcast_lock, flags); return broadcast; } @@ -405,7 +405,7 @@ static void tick_handle_oneshot_broadcast(struct clock_event_device *dev) ktime_t now, next_event; int cpu; - spin_lock(&tick_broadcast_lock); + raw_spin_lock(&tick_broadcast_lock); again: dev->next_event.tv64 = KTIME_MAX; next_event.tv64 = KTIME_MAX; @@ -443,7 +443,7 @@ again: if (tick_broadcast_set_event(next_event, 0)) goto again; } - spin_unlock(&tick_broadcast_lock); + raw_spin_unlock(&tick_broadcast_lock); } /* @@ -457,7 +457,7 @@ void tick_broadcast_oneshot_control(unsigned long reason) unsigned long flags; int cpu; - spin_lock_irqsave(&tick_broadcast_lock, flags); + raw_spin_lock_irqsave(&tick_broadcast_lock, flags); /* * Periodic mode does not care about the enter/exit of power @@ -492,7 +492,7 @@ void tick_broadcast_oneshot_control(unsigned long reason) } out: - spin_unlock_irqrestore(&tick_broadcast_lock, flags); + raw_spin_unlock_irqrestore(&tick_broadcast_lock, flags); } /* @@ -563,13 +563,13 @@ void tick_broadcast_switch_to_oneshot(void) struct clock_event_device *bc; unsigned long flags; - spin_lock_irqsave(&tick_broadcast_lock, flags); + raw_spin_lock_irqsave(&tick_broadcast_lock, flags); tick_broadcast_device.mode = TICKDEV_MODE_ONESHOT; bc = tick_broadcast_device.evtdev; if (bc) tick_broadcast_setup_oneshot(bc); - spin_unlock_irqrestore(&tick_broadcast_lock, flags); + raw_spin_unlock_irqrestore(&tick_broadcast_lock, flags); } @@ -581,7 +581,7 @@ void tick_shutdown_broadcast_oneshot(unsigned int *cpup) unsigned long flags; unsigned int cpu = *cpup; - spin_lock_irqsave(&tick_broadcast_lock, flags); + raw_spin_lock_irqsave(&tick_broadcast_lock, flags); /* * Clear the broadcast mask flag for the dead cpu, but do not @@ -589,7 +589,7 @@ void tick_shutdown_broadcast_oneshot(unsigned int *cpup) */ cpumask_clear_cpu(cpu, tick_get_broadcast_oneshot_mask()); - spin_unlock_irqrestore(&tick_broadcast_lock, flags); + raw_spin_unlock_irqrestore(&tick_broadcast_lock, flags); } /* diff --git a/kernel/time/tick-common.c b/kernel/time/tick-common.c index af39cf1cfa50..b6b898d2eeef 100644 --- a/kernel/time/tick-common.c +++ b/kernel/time/tick-common.c @@ -34,7 +34,7 @@ DEFINE_PER_CPU(struct tick_device, tick_cpu_device); ktime_t tick_next_period; ktime_t tick_period; int tick_do_timer_cpu __read_mostly = TICK_DO_TIMER_BOOT; -static DEFINE_SPINLOCK(tick_device_lock); +static DEFINE_RAW_SPINLOCK(tick_device_lock); /* * Debugging: see timer_list.c @@ -209,7 +209,7 @@ static int tick_check_new_device(struct clock_event_device *newdev) int cpu, ret = NOTIFY_OK; unsigned long flags; - spin_lock_irqsave(&tick_device_lock, flags); + raw_spin_lock_irqsave(&tick_device_lock, flags); cpu = smp_processor_id(); if (!cpumask_test_cpu(cpu, newdev->cpumask)) @@ -268,7 +268,7 @@ static int tick_check_new_device(struct clock_event_device *newdev) if (newdev->features & CLOCK_EVT_FEAT_ONESHOT) tick_oneshot_notify(); - spin_unlock_irqrestore(&tick_device_lock, flags); + raw_spin_unlock_irqrestore(&tick_device_lock, flags); return NOTIFY_STOP; out_bc: @@ -278,7 +278,7 @@ out_bc: if (tick_check_broadcast_device(newdev)) ret = NOTIFY_STOP; - spin_unlock_irqrestore(&tick_device_lock, flags); + raw_spin_unlock_irqrestore(&tick_device_lock, flags); return ret; } @@ -311,7 +311,7 @@ static void tick_shutdown(unsigned int *cpup) struct clock_event_device *dev = td->evtdev; unsigned long flags; - spin_lock_irqsave(&tick_device_lock, flags); + raw_spin_lock_irqsave(&tick_device_lock, flags); td->mode = TICKDEV_MODE_PERIODIC; if (dev) { /* @@ -322,7 +322,7 @@ static void tick_shutdown(unsigned int *cpup) clockevents_exchange_device(dev, NULL); td->evtdev = NULL; } - spin_unlock_irqrestore(&tick_device_lock, flags); + raw_spin_unlock_irqrestore(&tick_device_lock, flags); } static void tick_suspend(void) @@ -330,9 +330,9 @@ static void tick_suspend(void) struct tick_device *td = &__get_cpu_var(tick_cpu_device); unsigned long flags; - spin_lock_irqsave(&tick_device_lock, flags); + raw_spin_lock_irqsave(&tick_device_lock, flags); clockevents_shutdown(td->evtdev); - spin_unlock_irqrestore(&tick_device_lock, flags); + raw_spin_unlock_irqrestore(&tick_device_lock, flags); } static void tick_resume(void) @@ -341,7 +341,7 @@ static void tick_resume(void) unsigned long flags; int broadcast = tick_resume_broadcast(); - spin_lock_irqsave(&tick_device_lock, flags); + raw_spin_lock_irqsave(&tick_device_lock, flags); clockevents_set_mode(td->evtdev, CLOCK_EVT_MODE_RESUME); if (!broadcast) { @@ -350,7 +350,7 @@ static void tick_resume(void) else tick_resume_oneshot(); } - spin_unlock_irqrestore(&tick_device_lock, flags); + raw_spin_unlock_irqrestore(&tick_device_lock, flags); } /* -- cgit v1.2.3-71-gd317 From d4703aefdbc8f9f347f6dcefcddd791294314eb7 Mon Sep 17 00:00:00 2001 From: Rusty Russell Date: Tue, 15 Dec 2009 16:28:32 -0600 Subject: module: handle ppc64 relocating kcrctabs when CONFIG_RELOCATABLE=y powerpc applies relocations to the kcrctab. They're absolute symbols, but it's not completely unreasonable: other archs may too, but the relocation is often 0. http://lists.ozlabs.org/pipermail/linuxppc-dev/2009-November/077972.html Inspired-by: Neil Horman Signed-off-by: Rusty Russell Tested-by: Neil Horman Acked-by: Paul Mackerras --- arch/powerpc/include/asm/module.h | 5 +++++ arch/powerpc/kernel/vmlinux.lds.S | 3 +++ kernel/module.c | 28 +++++++++++++++++++++------- 3 files changed, 29 insertions(+), 7 deletions(-) (limited to 'kernel') diff --git a/arch/powerpc/include/asm/module.h b/arch/powerpc/include/asm/module.h index 08454880a2c0..0192a4ee2bc2 100644 --- a/arch/powerpc/include/asm/module.h +++ b/arch/powerpc/include/asm/module.h @@ -87,5 +87,10 @@ struct exception_table_entry; void sort_ex_table(struct exception_table_entry *start, struct exception_table_entry *finish); +#ifdef CONFIG_MODVERSIONS +#define ARCH_RELOCATES_KCRCTAB + +extern const unsigned long reloc_start[]; +#endif #endif /* __KERNEL__ */ #endif /* _ASM_POWERPC_MODULE_H */ diff --git a/arch/powerpc/kernel/vmlinux.lds.S b/arch/powerpc/kernel/vmlinux.lds.S index 27735a7ac12b..dcd01c82e701 100644 --- a/arch/powerpc/kernel/vmlinux.lds.S +++ b/arch/powerpc/kernel/vmlinux.lds.S @@ -38,6 +38,9 @@ jiffies = jiffies_64 + 4; #endif SECTIONS { + . = 0; + reloc_start = .; + . = KERNELBASE; /* diff --git a/kernel/module.c b/kernel/module.c index 12afc5a3ddd3..a65dc787a27b 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -880,11 +880,23 @@ static int try_to_force_load(struct module *mod, const char *reason) } #ifdef CONFIG_MODVERSIONS +/* If the arch applies (non-zero) relocations to kernel kcrctab, unapply it. */ +static unsigned long maybe_relocated(unsigned long crc, + const struct module *crc_owner) +{ +#ifdef ARCH_RELOCATES_KCRCTAB + if (crc_owner == NULL) + return crc - (unsigned long)reloc_start; +#endif + return crc; +} + static int check_version(Elf_Shdr *sechdrs, unsigned int versindex, const char *symname, struct module *mod, - const unsigned long *crc) + const unsigned long *crc, + const struct module *crc_owner) { unsigned int i, num_versions; struct modversion_info *versions; @@ -905,10 +917,10 @@ static int check_version(Elf_Shdr *sechdrs, if (strcmp(versions[i].name, symname) != 0) continue; - if (versions[i].crc == *crc) + if (versions[i].crc == maybe_relocated(*crc, crc_owner)) return 1; DEBUGP("Found checksum %lX vs module %lX\n", - *crc, versions[i].crc); + maybe_relocated(*crc, crc_owner), versions[i].crc); goto bad_version; } @@ -931,7 +943,8 @@ static inline int check_modstruct_version(Elf_Shdr *sechdrs, if (!find_symbol(MODULE_SYMBOL_PREFIX "module_layout", NULL, &crc, true, false)) BUG(); - return check_version(sechdrs, versindex, "module_layout", mod, crc); + return check_version(sechdrs, versindex, "module_layout", mod, crc, + NULL); } /* First part is kernel version, which we ignore if module has crcs. */ @@ -949,7 +962,8 @@ static inline int check_version(Elf_Shdr *sechdrs, unsigned int versindex, const char *symname, struct module *mod, - const unsigned long *crc) + const unsigned long *crc, + const struct module *crc_owner) { return 1; } @@ -984,8 +998,8 @@ static const struct kernel_symbol *resolve_symbol(Elf_Shdr *sechdrs, /* use_module can fail due to OOM, or module initialization or unloading */ if (sym) { - if (!check_version(sechdrs, versindex, name, mod, crc) || - !use_module(mod, owner)) + if (!check_version(sechdrs, versindex, name, mod, crc, owner) + || !use_module(mod, owner)) sym = NULL; } return sym; -- cgit v1.2.3-71-gd317 From e36c54582c6f14adc9e10473e2aec2cc4f0acc03 Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Mon, 14 Dec 2009 15:58:33 -0500 Subject: tracing: Fix return of trace_dump_stack() The trace_dump_stack() returned a value for a void function. Also, added the missing stub for trace_dump_stack() when tracing is not configured. Reported-by: Ingo Molnar LKML-Reference: <20091214162713.GA31060@elte.hu> Signed-off-by: Steven Rostedt Signed-off-by: Ingo Molnar --- include/linux/kernel.h | 1 + kernel/trace/trace.c | 2 +- 2 files changed, 2 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/include/linux/kernel.h b/include/linux/kernel.h index 5ad4199fb073..f1dc752da0d2 100644 --- a/include/linux/kernel.h +++ b/include/linux/kernel.h @@ -527,6 +527,7 @@ trace_printk(const char *fmt, ...) __attribute__ ((format (printf, 1, 2))); static inline void tracing_start(void) { } static inline void tracing_stop(void) { } static inline void ftrace_off_permanent(void) { } +static inline void trace_dump_stack(void) { } static inline int trace_printk(const char *fmt, ...) { diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index bd7b969a729a..ee61915935d5 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -1158,7 +1158,7 @@ void trace_dump_stack(void) unsigned long flags; if (tracing_disabled || tracing_selftest_running) - return 0; + return; local_save_flags(flags); -- cgit v1.2.3-71-gd317 From 0f624e7e5625f4c30c836b7a5decfe2553582391 Mon Sep 17 00:00:00 2001 From: Paul Mackerras Date: Tue, 15 Dec 2009 19:40:32 +1100 Subject: perf_event: Fix incorrect range check on cpu number It is quite legitimate for CPUs to be numbered sparsely, meaning that it possible for an online CPU to have a number which is greater than the total count of possible CPUs. Currently find_get_context() has a sanity check on the cpu number where it checks it against num_possible_cpus(). This test can fail for a legitimate cpu number if the cpu_possible_mask is sparsely populated. This fixes the problem by checking the CPU number against nr_cpumask_bits instead, since that is the appropriate check to ensure that the cpu number is same to pass to cpu_isset() subsequently. Reported-by: Michael Neuling Signed-off-by: Paul Mackerras Tested-by: Michael Neuling Acked-by: Peter Zijlstra Cc: LKML-Reference: <20091215084032.GA18661@brick.ozlabs.ibm.com> Signed-off-by: Ingo Molnar --- kernel/perf_event.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/perf_event.c b/kernel/perf_event.c index d891ec4a8100..8823b0885183 100644 --- a/kernel/perf_event.c +++ b/kernel/perf_event.c @@ -1603,7 +1603,7 @@ static struct perf_event_context *find_get_context(pid_t pid, int cpu) if (perf_paranoid_cpu() && !capable(CAP_SYS_ADMIN)) return ERR_PTR(-EACCES); - if (cpu < 0 || cpu > num_possible_cpus()) + if (cpu < 0 || cpu >= nr_cpumask_bits) return ERR_PTR(-EINVAL); /* -- cgit v1.2.3-71-gd317 From 4b731d50ff3df6b9141a6c12b088e8eb0109e83c Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Mon, 14 Dec 2009 17:57:34 -0800 Subject: bsdacct: fix uid/gid misreporting commit d8e180dcd5bbbab9cd3ff2e779efcf70692ef541 "bsdacct: switch credentials for writing to the accounting file" introduced credential switching during final acct data collecting. However, uid/gid pair continued to be collected from current which became credentials of who created acct file, not who exits. Addresses http://bugzilla.kernel.org/show_bug.cgi?id=14676 Signed-off-by: Alexey Dobriyan Reported-by: Juho K. Juopperi Acked-by: Serge Hallyn Acked-by: David Howells Reviewed-by: Michal Schmidt Cc: James Morris Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/acct.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/acct.c b/kernel/acct.c index 9a4715a2f6bf..a6605ca921b6 100644 --- a/kernel/acct.c +++ b/kernel/acct.c @@ -536,7 +536,8 @@ static void do_acct_process(struct bsd_acct_struct *acct, do_div(elapsed, AHZ); ac.ac_btime = get_seconds() - elapsed; /* we really need to bite the bullet and change layout */ - current_uid_gid(&ac.ac_uid, &ac.ac_gid); + ac.ac_uid = orig_cred->uid; + ac.ac_gid = orig_cred->gid; #if ACCT_VERSION==2 ac.ac_ahz = AHZ; #endif -- cgit v1.2.3-71-gd317 From 06808b0827e1cd14eedc96bac2655d5b37ac246c Mon Sep 17 00:00:00 2001 From: Lee Schermerhorn Date: Mon, 14 Dec 2009 17:58:21 -0800 Subject: hugetlb: derive huge pages nodes allowed from task mempolicy This patch derives a "nodes_allowed" node mask from the numa mempolicy of the task modifying the number of persistent huge pages to control the allocation, freeing and adjusting of surplus huge pages when the pool page count is modified via the new sysctl or sysfs attribute "nr_hugepages_mempolicy". The nodes_allowed mask is derived as follows: * For "default" [NULL] task mempolicy, a NULL nodemask_t pointer is produced. This will cause the hugetlb subsystem to use node_online_map as the "nodes_allowed". This preserves the behavior before this patch. * For "preferred" mempolicy, including explicit local allocation, a nodemask with the single preferred node will be produced. "local" policy will NOT track any internode migrations of the task adjusting nr_hugepages. * For "bind" and "interleave" policy, the mempolicy's nodemask will be used. * Other than to inform the construction of the nodes_allowed node mask, the actual mempolicy mode is ignored. That is, all modes behave like interleave over the resulting nodes_allowed mask with no "fallback". See the updated documentation [next patch] for more information about the implications of this patch. Examples: Starting with: Node 0 HugePages_Total: 0 Node 1 HugePages_Total: 0 Node 2 HugePages_Total: 0 Node 3 HugePages_Total: 0 Default behavior [with or without this patch] balances persistent hugepage allocation across nodes [with sufficient contiguous memory]: sysctl vm.nr_hugepages[_mempolicy]=32 yields: Node 0 HugePages_Total: 8 Node 1 HugePages_Total: 8 Node 2 HugePages_Total: 8 Node 3 HugePages_Total: 8 Of course, we only have nr_hugepages_mempolicy with the patch, but with default mempolicy, nr_hugepages_mempolicy behaves the same as nr_hugepages. Applying mempolicy--e.g., with numactl [using '-m' a.k.a. '--membind' because it allows multiple nodes to be specified and it's easy to type]--we can allocate huge pages on individual nodes or sets of nodes. So, starting from the condition above, with 8 huge pages per node, add 8 more to node 2 using: numactl -m 2 sysctl vm.nr_hugepages_mempolicy=40 This yields: Node 0 HugePages_Total: 8 Node 1 HugePages_Total: 8 Node 2 HugePages_Total: 16 Node 3 HugePages_Total: 8 The incremental 8 huge pages were restricted to node 2 by the specified mempolicy. Similarly, we can use mempolicy to free persistent huge pages from specified nodes: numactl -m 0,1 sysctl vm.nr_hugepages_mempolicy=32 yields: Node 0 HugePages_Total: 4 Node 1 HugePages_Total: 4 Node 2 HugePages_Total: 16 Node 3 HugePages_Total: 8 The 8 huge pages freed were balanced over nodes 0 and 1. [rientjes@google.com: accomodate reworked NODEMASK_ALLOC] Signed-off-by: David Rientjes Signed-off-by: Lee Schermerhorn Acked-by: Mel Gorman Reviewed-by: Andi Kleen Cc: KAMEZAWA Hiroyuki Cc: Randy Dunlap Cc: Nishanth Aravamudan Cc: Adam Litke Cc: Andy Whitcroft Cc: Eric Whitney Cc: Christoph Lameter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/hugetlb.h | 6 +++ include/linux/mempolicy.h | 3 ++ kernel/sysctl.c | 15 +++++++- mm/hugetlb.c | 97 ++++++++++++++++++++++++++++++++++++++++------- mm/mempolicy.c | 47 +++++++++++++++++++++++ 5 files changed, 153 insertions(+), 15 deletions(-) (limited to 'kernel') diff --git a/include/linux/hugetlb.h b/include/linux/hugetlb.h index 41a59afc70fa..78b4bc64c006 100644 --- a/include/linux/hugetlb.h +++ b/include/linux/hugetlb.h @@ -23,6 +23,12 @@ void reset_vma_resv_huge_pages(struct vm_area_struct *vma); int hugetlb_sysctl_handler(struct ctl_table *, int, void __user *, size_t *, loff_t *); int hugetlb_overcommit_handler(struct ctl_table *, int, void __user *, size_t *, loff_t *); int hugetlb_treat_movable_handler(struct ctl_table *, int, void __user *, size_t *, loff_t *); + +#ifdef CONFIG_NUMA +int hugetlb_mempolicy_sysctl_handler(struct ctl_table *, int, + void __user *, size_t *, loff_t *); +#endif + int copy_hugetlb_page_range(struct mm_struct *, struct mm_struct *, struct vm_area_struct *); int follow_hugetlb_page(struct mm_struct *, struct vm_area_struct *, struct page **, struct vm_area_struct **, diff --git a/include/linux/mempolicy.h b/include/linux/mempolicy.h index 085c903fe0f1..1cc966cd3e5f 100644 --- a/include/linux/mempolicy.h +++ b/include/linux/mempolicy.h @@ -201,6 +201,7 @@ extern void mpol_fix_fork_child_flag(struct task_struct *p); extern struct zonelist *huge_zonelist(struct vm_area_struct *vma, unsigned long addr, gfp_t gfp_flags, struct mempolicy **mpol, nodemask_t **nodemask); +extern bool init_nodemask_of_mempolicy(nodemask_t *mask); extern unsigned slab_node(struct mempolicy *policy); extern enum zone_type policy_zone; @@ -328,6 +329,8 @@ static inline struct zonelist *huge_zonelist(struct vm_area_struct *vma, return node_zonelist(0, gfp_flags); } +static inline bool init_nodemask_of_mempolicy(nodemask_t *m) { return false; } + static inline int do_migrate_pages(struct mm_struct *mm, const nodemask_t *from_nodes, const nodemask_t *to_nodes, int flags) diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 554ac4894f0f..60fc93131095 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -1051,7 +1051,7 @@ static struct ctl_table vm_table[] = { .extra2 = &one_hundred, }, #ifdef CONFIG_HUGETLB_PAGE - { + { .procname = "nr_hugepages", .data = NULL, .maxlen = sizeof(unsigned long), @@ -1059,7 +1059,18 @@ static struct ctl_table vm_table[] = { .proc_handler = hugetlb_sysctl_handler, .extra1 = (void *)&hugetlb_zero, .extra2 = (void *)&hugetlb_infinity, - }, + }, +#ifdef CONFIG_NUMA + { + .procname = "nr_hugepages_mempolicy", + .data = NULL, + .maxlen = sizeof(unsigned long), + .mode = 0644, + .proc_handler = &hugetlb_mempolicy_sysctl_handler, + .extra1 = (void *)&hugetlb_zero, + .extra2 = (void *)&hugetlb_infinity, + }, +#endif { .procname = "hugetlb_shm_group", .data = &sysctl_hugetlb_shm_group, diff --git a/mm/hugetlb.c b/mm/hugetlb.c index 324d1abae876..1125d818ea06 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -1330,29 +1330,71 @@ static struct hstate *kobj_to_hstate(struct kobject *kobj) return NULL; } -static ssize_t nr_hugepages_show(struct kobject *kobj, +static ssize_t nr_hugepages_show_common(struct kobject *kobj, struct kobj_attribute *attr, char *buf) { struct hstate *h = kobj_to_hstate(kobj); return sprintf(buf, "%lu\n", h->nr_huge_pages); } -static ssize_t nr_hugepages_store(struct kobject *kobj, - struct kobj_attribute *attr, const char *buf, size_t count) +static ssize_t nr_hugepages_store_common(bool obey_mempolicy, + struct kobject *kobj, struct kobj_attribute *attr, + const char *buf, size_t len) { int err; - unsigned long input; + unsigned long count; struct hstate *h = kobj_to_hstate(kobj); + NODEMASK_ALLOC(nodemask_t, nodes_allowed); - err = strict_strtoul(buf, 10, &input); + err = strict_strtoul(buf, 10, &count); if (err) return 0; - h->max_huge_pages = set_max_huge_pages(h, input, &node_online_map); + if (!(obey_mempolicy && init_nodemask_of_mempolicy(nodes_allowed))) { + NODEMASK_FREE(nodes_allowed); + nodes_allowed = &node_online_map; + } + h->max_huge_pages = set_max_huge_pages(h, count, nodes_allowed); - return count; + if (nodes_allowed != &node_online_map) + NODEMASK_FREE(nodes_allowed); + + return len; +} + +static ssize_t nr_hugepages_show(struct kobject *kobj, + struct kobj_attribute *attr, char *buf) +{ + return nr_hugepages_show_common(kobj, attr, buf); +} + +static ssize_t nr_hugepages_store(struct kobject *kobj, + struct kobj_attribute *attr, const char *buf, size_t len) +{ + return nr_hugepages_store_common(false, kobj, attr, buf, len); } HSTATE_ATTR(nr_hugepages); +#ifdef CONFIG_NUMA + +/* + * hstate attribute for optionally mempolicy-based constraint on persistent + * huge page alloc/free. + */ +static ssize_t nr_hugepages_mempolicy_show(struct kobject *kobj, + struct kobj_attribute *attr, char *buf) +{ + return nr_hugepages_show_common(kobj, attr, buf); +} + +static ssize_t nr_hugepages_mempolicy_store(struct kobject *kobj, + struct kobj_attribute *attr, const char *buf, size_t len) +{ + return nr_hugepages_store_common(true, kobj, attr, buf, len); +} +HSTATE_ATTR(nr_hugepages_mempolicy); +#endif + + static ssize_t nr_overcommit_hugepages_show(struct kobject *kobj, struct kobj_attribute *attr, char *buf) { @@ -1408,6 +1450,9 @@ static struct attribute *hstate_attrs[] = { &free_hugepages_attr.attr, &resv_hugepages_attr.attr, &surplus_hugepages_attr.attr, +#ifdef CONFIG_NUMA + &nr_hugepages_mempolicy_attr.attr, +#endif NULL, }; @@ -1574,9 +1619,9 @@ static unsigned int cpuset_mems_nr(unsigned int *array) } #ifdef CONFIG_SYSCTL -int hugetlb_sysctl_handler(struct ctl_table *table, int write, - void __user *buffer, - size_t *length, loff_t *ppos) +static int hugetlb_sysctl_handler_common(bool obey_mempolicy, + struct ctl_table *table, int write, + void __user *buffer, size_t *length, loff_t *ppos) { struct hstate *h = &default_hstate; unsigned long tmp; @@ -1588,13 +1633,39 @@ int hugetlb_sysctl_handler(struct ctl_table *table, int write, table->maxlen = sizeof(unsigned long); proc_doulongvec_minmax(table, write, buffer, length, ppos); - if (write) - h->max_huge_pages = set_max_huge_pages(h, tmp, - &node_online_map); + if (write) { + NODEMASK_ALLOC(nodemask_t, nodes_allowed); + if (!(obey_mempolicy && + init_nodemask_of_mempolicy(nodes_allowed))) { + NODEMASK_FREE(nodes_allowed); + nodes_allowed = &node_states[N_HIGH_MEMORY]; + } + h->max_huge_pages = set_max_huge_pages(h, tmp, nodes_allowed); + + if (nodes_allowed != &node_states[N_HIGH_MEMORY]) + NODEMASK_FREE(nodes_allowed); + } return 0; } +int hugetlb_sysctl_handler(struct ctl_table *table, int write, + void __user *buffer, size_t *length, loff_t *ppos) +{ + + return hugetlb_sysctl_handler_common(false, table, write, + buffer, length, ppos); +} + +#ifdef CONFIG_NUMA +int hugetlb_mempolicy_sysctl_handler(struct ctl_table *table, int write, + void __user *buffer, size_t *length, loff_t *ppos) +{ + return hugetlb_sysctl_handler_common(true, table, write, + buffer, length, ppos); +} +#endif /* CONFIG_NUMA */ + int hugetlb_treat_movable_handler(struct ctl_table *table, int write, void __user *buffer, size_t *length, loff_t *ppos) diff --git a/mm/mempolicy.c b/mm/mempolicy.c index 0f89eabbaf3e..f11fdad06204 100644 --- a/mm/mempolicy.c +++ b/mm/mempolicy.c @@ -1568,6 +1568,53 @@ struct zonelist *huge_zonelist(struct vm_area_struct *vma, unsigned long addr, } return zl; } + +/* + * init_nodemask_of_mempolicy + * + * If the current task's mempolicy is "default" [NULL], return 'false' + * to indicate default policy. Otherwise, extract the policy nodemask + * for 'bind' or 'interleave' policy into the argument nodemask, or + * initialize the argument nodemask to contain the single node for + * 'preferred' or 'local' policy and return 'true' to indicate presence + * of non-default mempolicy. + * + * We don't bother with reference counting the mempolicy [mpol_get/put] + * because the current task is examining it's own mempolicy and a task's + * mempolicy is only ever changed by the task itself. + * + * N.B., it is the caller's responsibility to free a returned nodemask. + */ +bool init_nodemask_of_mempolicy(nodemask_t *mask) +{ + struct mempolicy *mempolicy; + int nid; + + if (!(mask && current->mempolicy)) + return false; + + mempolicy = current->mempolicy; + switch (mempolicy->mode) { + case MPOL_PREFERRED: + if (mempolicy->flags & MPOL_F_LOCAL) + nid = numa_node_id(); + else + nid = mempolicy->v.preferred_node; + init_nodemask_of_node(mask, nid); + break; + + case MPOL_BIND: + /* Fall through */ + case MPOL_INTERLEAVE: + *mask = mempolicy->v.nodes; + break; + + default: + BUG(); + } + + return true; +} #endif /* Allocate a page in interleaved policy. -- cgit v1.2.3-71-gd317 From 70da2340fbc68e91e701762f785479ab495a0869 Mon Sep 17 00:00:00 2001 From: Amerigo Wang Date: Mon, 14 Dec 2009 17:59:52 -0800 Subject: 'sysctl_max_map_count' should be non-negative Jan Engelhardt reported we have this problem: setting max_map_count to a value large enough results in programs dying at first try. This is on 2.6.31.6: 15:59 borg:/proc/sys/vm # echo $[1<<31-1] >max_map_count 15:59 borg:/proc/sys/vm # cat max_map_count 1073741824 15:59 borg:/proc/sys/vm # echo $[1<<31] >max_map_count 15:59 borg:/proc/sys/vm # cat max_map_count Killed This is because we have a chance to make 'max_map_count' negative. but it's meaningless. Make it only accept non-negative values. Reported-by: Jan Engelhardt Signed-off-by: WANG Cong Cc: Ingo Molnar Cc: Peter Zijlstra Cc: James Morris Cc: Alexey Dobriyan Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/sysctl.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 60fc93131095..45e4bef0012a 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -1131,7 +1131,8 @@ static struct ctl_table vm_table[] = { .data = &sysctl_max_map_count, .maxlen = sizeof(sysctl_max_map_count), .mode = 0644, - .proc_handler = proc_dointvec + .proc_handler = proc_dointvec, + .extra1 = &zero, }, #else { -- cgit v1.2.3-71-gd317 From c0f68c2fab4898bcc4671a8fb941f428856b4ad5 Mon Sep 17 00:00:00 2001 From: Xiao Guangrong Date: Mon, 14 Dec 2009 18:00:16 -0800 Subject: generic-ipi: cleanup for generic_smp_call_function_interrupt() Use smp_processor_id() instead of get_cpu() and put_cpu() in generic_smp_call_function_interrupt(), It's no need to disable preempt, because we must call generic_smp_call_function_interrupt() with interrupts disabled. Signed-off-by: Xiao Guangrong Acked-by: Ingo Molnar Cc: Jens Axboe Cc: Nick Piggin Cc: Peter Zijlstra Cc: Rusty Russell Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/smp.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/smp.c b/kernel/smp.c index a8c76069cf50..00a1d0ede532 100644 --- a/kernel/smp.c +++ b/kernel/smp.c @@ -171,7 +171,7 @@ void generic_exec_single(int cpu, struct call_single_data *data, int wait) void generic_smp_call_function_interrupt(void) { struct call_function_data *data; - int cpu = get_cpu(); + int cpu = smp_processor_id(); /* * Shouldn't receive this interrupt on a cpu that is not yet online. @@ -212,7 +212,6 @@ void generic_smp_call_function_interrupt(void) csd_unlock(&data->csd); } - put_cpu(); } /* -- cgit v1.2.3-71-gd317 From dfc6a736d452a8c308190b618b065c2257d370ff Mon Sep 17 00:00:00 2001 From: H Hartley Sweeten Date: Mon, 14 Dec 2009 18:00:22 -0800 Subject: kernel/sys.c: fix "warning: do-while statement is not a compound statement" noise do_each_thread/while_each_thread wrap a block of code that is in this format: for (...) do ... while If curly braces do not surround the inner loop the following warning is generated by sparse: warning: do-while statement is not a compound statement Fix the warning by adding the braces. Signed-off-by: H Hartley Sweeten Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/sys.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'kernel') diff --git a/kernel/sys.c b/kernel/sys.c index 585d6cd10040..20ccfb5da6af 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -189,10 +189,10 @@ SYSCALL_DEFINE3(setpriority, int, which, int, who, int, niceval) !(user = find_user(who))) goto out_unlock; /* No processes for this user */ - do_each_thread(g, p) + do_each_thread(g, p) { if (__task_cred(p)->uid == who) error = set_one_prio(p, niceval, error); - while_each_thread(g, p); + } while_each_thread(g, p); if (who != cred->uid) free_uid(user); /* For find_user() */ break; @@ -252,13 +252,13 @@ SYSCALL_DEFINE2(getpriority, int, which, int, who) !(user = find_user(who))) goto out_unlock; /* No processes for this user */ - do_each_thread(g, p) + do_each_thread(g, p) { if (__task_cred(p)->uid == who) { niceval = 20 - task_nice(p); if (niceval > retval) retval = niceval; } - while_each_thread(g, p); + } while_each_thread(g, p); if (who != cred->uid) free_uid(user); /* for find_user() */ break; -- cgit v1.2.3-71-gd317 From 5ada918b82399eef3afd6a71e3637697d6bd719f Mon Sep 17 00:00:00 2001 From: Bernhard Walle Date: Mon, 14 Dec 2009 18:00:43 -0800 Subject: vt: introduce and use vt_kmsg_redirect() function The kernel offers with TIOCL_GETKMSGREDIRECT ioctl() the possibility to redirect the kernel messages to a specific console. However, since it's not possible to switch to the kernel message console after a panic(), it would be nice if the kernel would print the panic message on the current console. This patch series adds a new interface to access the global kmsg_redirect variable by a function to be able to use it in code where CONFIG_VT_CONSOLE is not set (kernel/panic.c). This patch: Instead of using and exporting a global value kmsg_redirect, introduce a function vt_kmsg_redirect() that both can set and return the console where messages are printed. Change all users of kmsg_redirect (the VT code itself and kernel/power.c) to the new interface. The main advantage is that vt_kmsg_redirect() can also be used when CONFIG_VT_CONSOLE is not set. Signed-off-by: Bernhard Walle Cc: Alan Cox Cc: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/char/vt.c | 43 +++++++++++++++++++++++++++++++++++++------ include/linux/tty.h | 2 -- include/linux/vt.h | 15 +++++++++++++++ kernel/power/console.c | 7 +++---- 4 files changed, 55 insertions(+), 12 deletions(-) (limited to 'kernel') diff --git a/drivers/char/vt.c b/drivers/char/vt.c index 1e3d728dbf7e..e43fbc66aef0 100644 --- a/drivers/char/vt.c +++ b/drivers/char/vt.c @@ -184,12 +184,10 @@ static DECLARE_WORK(console_work, console_callback); * fg_console is the current virtual console, * last_console is the last used one, * want_console is the console we want to switch to, - * kmsg_redirect is the console for kernel messages, */ int fg_console; int last_console; int want_console = -1; -int kmsg_redirect; /* * For each existing display, we have a pointer to console currently visible @@ -2434,6 +2432,37 @@ struct tty_driver *console_driver; #ifdef CONFIG_VT_CONSOLE +/** + * vt_kmsg_redirect() - Sets/gets the kernel message console + * @new: The new virtual terminal number or -1 if the console should stay + * unchanged + * + * By default, the kernel messages are always printed on the current virtual + * console. However, the user may modify that default with the + * TIOCL_SETKMSGREDIRECT ioctl call. + * + * This function sets the kernel message console to be @new. It returns the old + * virtual console number. The virtual terminal number 0 (both as parameter and + * return value) means no redirection (i.e. always printed on the currently + * active console). + * + * The parameter -1 means that only the current console is returned, but the + * value is not modified. You may use the macro vt_get_kmsg_redirect() in that + * case to make the code more understandable. + * + * When the kernel is compiled without CONFIG_VT_CONSOLE, this function ignores + * the parameter and always returns 0. + */ +int vt_kmsg_redirect(int new) +{ + static int kmsg_con; + + if (new != -1) + return xchg(&kmsg_con, new); + else + return kmsg_con; +} + /* * Console on virtual terminal * @@ -2448,6 +2477,7 @@ static void vt_console_print(struct console *co, const char *b, unsigned count) const ushort *start; ushort cnt = 0; ushort myx; + int kmsg_console; /* console busy or not yet initialized */ if (!printable) @@ -2455,8 +2485,9 @@ static void vt_console_print(struct console *co, const char *b, unsigned count) if (!spin_trylock(&printing_lock)) return; - if (kmsg_redirect && vc_cons_allocated(kmsg_redirect - 1)) - vc = vc_cons[kmsg_redirect - 1].d; + kmsg_console = vt_get_kmsg_redirect(); + if (kmsg_console && vc_cons_allocated(kmsg_console - 1)) + vc = vc_cons[kmsg_console - 1].d; /* read `x' only after setting currcons properly (otherwise the `x' macro will read the x of the foreground console). */ @@ -2613,7 +2644,7 @@ int tioclinux(struct tty_struct *tty, unsigned long arg) ret = set_vesa_blanking(p); break; case TIOCL_GETKMSGREDIRECT: - data = kmsg_redirect; + data = vt_get_kmsg_redirect(); ret = __put_user(data, p); break; case TIOCL_SETKMSGREDIRECT: @@ -2623,7 +2654,7 @@ int tioclinux(struct tty_struct *tty, unsigned long arg) if (get_user(data, p+1)) ret = -EFAULT; else - kmsg_redirect = data; + vt_kmsg_redirect(data); } break; case TIOCL_GETFGCONSOLE: diff --git a/include/linux/tty.h b/include/linux/tty.h index 405a9035fe40..ef3a2947b102 100644 --- a/include/linux/tty.h +++ b/include/linux/tty.h @@ -350,8 +350,6 @@ extern void tty_write_flush(struct tty_struct *); extern struct ktermios tty_std_termios; -extern int kmsg_redirect; - extern void console_init(void); extern int vcs_init(void); diff --git a/include/linux/vt.h b/include/linux/vt.h index 7ffa11f06232..3fb9944e50a6 100644 --- a/include/linux/vt.h +++ b/include/linux/vt.h @@ -84,4 +84,19 @@ struct vt_setactivate { #define VT_SETACTIVATE 0x560F /* Activate and set the mode of a console */ +#ifdef CONFIG_VT_CONSOLE + +extern int vt_kmsg_redirect(int new); + +#else + +static inline int vt_kmsg_redirect(int new) +{ + return 0; +} + +#endif + +#define vt_get_kmsg_redirect() vt_kmsg_redirect(-1) + #endif /* _LINUX_VT_H */ diff --git a/kernel/power/console.c b/kernel/power/console.c index 5187136fe1de..218e5af90156 100644 --- a/kernel/power/console.c +++ b/kernel/power/console.c @@ -6,7 +6,7 @@ #include #include -#include +#include #include #include "power.h" @@ -21,8 +21,7 @@ int pm_prepare_console(void) if (orig_fgconsole < 0) return 1; - orig_kmsg = kmsg_redirect; - kmsg_redirect = SUSPEND_CONSOLE; + orig_kmsg = vt_kmsg_redirect(SUSPEND_CONSOLE); return 0; } @@ -30,7 +29,7 @@ void pm_restore_console(void) { if (orig_fgconsole >= 0) { vt_move_to_console(orig_fgconsole, 0); - kmsg_redirect = orig_kmsg; + vt_kmsg_redirect(orig_kmsg); } } #endif -- cgit v1.2.3-71-gd317 From e7d2860b690d4f3bed6824757c540579638e3d1e Mon Sep 17 00:00:00 2001 From: André Goddard Rosa Date: Mon, 14 Dec 2009 18:01:06 -0800 Subject: tree-wide: convert open calls to remove spaces to skip_spaces() lib function MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Makes use of skip_spaces() defined in lib/string.c for removing leading spaces from strings all over the tree. It decreases lib.a code size by 47 bytes and reuses the function tree-wide: text data bss dec hex filename 64688 584 592 65864 10148 (TOTALS-BEFORE) 64641 584 592 65817 10119 (TOTALS-AFTER) Also, while at it, if we see (*str && isspace(*str)), we can be sure to remove the first condition (*str) as the second one (isspace(*str)) also evaluates to 0 whenever *str == 0, making it redundant. In other words, "a char equals zero is never a space". Julia Lawall tried the semantic patch (http://coccinelle.lip6.fr) below, and found occurrences of this pattern on 3 more files: drivers/leds/led-class.c drivers/leds/ledtrig-timer.c drivers/video/output.c @@ expression str; @@ ( // ignore skip_spaces cases while (*str && isspace(*str)) { \(str++;\|++str;\) } | - *str && isspace(*str) ) Signed-off-by: André Goddard Rosa Cc: Julia Lawall Cc: Martin Schwidefsky Cc: Jeff Dike Cc: Ingo Molnar Cc: Thomas Gleixner Cc: "H. Peter Anvin" Cc: Richard Purdie Cc: Neil Brown Cc: Kyle McMartin Cc: Henrique de Moraes Holschuh Cc: David Howells Cc: Cc: Samuel Ortiz Cc: Patrick McHardy Cc: Takashi Iwai Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/s390/kernel/debug.c | 3 ++- arch/um/drivers/mconsole_kern.c | 16 +++++++--------- arch/x86/kernel/cpu/mtrr/if.c | 11 ++++------- drivers/leds/led-class.c | 2 +- drivers/leds/ledtrig-timer.c | 4 ++-- drivers/md/dm-table.c | 6 ++---- drivers/md/md.c | 4 ++-- drivers/parisc/pdc_stable.c | 9 +++------ drivers/platform/x86/thinkpad_acpi.c | 7 ++----- drivers/pnp/interface.c | 36 ++++++++++------------------------- drivers/s390/block/dasd_proc.c | 5 +++-- drivers/video/backlight/lcd.c | 4 ++-- drivers/video/display/display-sysfs.c | 2 +- drivers/video/output.c | 2 +- fs/cachefiles/daemon.c | 4 ++-- fs/ext4/super.c | 7 ++----- kernel/params.c | 8 +++----- lib/argv_split.c | 13 +++---------- lib/dynamic_debug.c | 4 ++-- lib/vsprintf.c | 15 ++++----------- net/irda/irnet/irnet.h | 1 + net/irda/irnet/irnet_ppp.c | 8 +++----- net/netfilter/xt_recent.c | 3 +-- sound/pci/hda/hda_hwdep.c | 7 +++---- 24 files changed, 66 insertions(+), 115 deletions(-) (limited to 'kernel') diff --git a/arch/s390/kernel/debug.c b/arch/s390/kernel/debug.c index 071c81f179ef..0168472b2fdf 100644 --- a/arch/s390/kernel/debug.c +++ b/arch/s390/kernel/debug.c @@ -18,6 +18,7 @@ #include #include #include +#include #include #include #include @@ -1178,7 +1179,7 @@ debug_get_uint(char *buf) { int rc; - for(; isspace(*buf); buf++); + buf = skip_spaces(buf); rc = simple_strtoul(buf, &buf, 10); if(*buf){ rc = -EINVAL; diff --git a/arch/um/drivers/mconsole_kern.c b/arch/um/drivers/mconsole_kern.c index f0fa47f10e6c..51069245b79a 100644 --- a/arch/um/drivers/mconsole_kern.c +++ b/arch/um/drivers/mconsole_kern.c @@ -6,6 +6,7 @@ #include #include +#include #include #include #include @@ -131,7 +132,7 @@ void mconsole_proc(struct mc_request *req) char *ptr = req->request.data, *buf; ptr += strlen("proc"); - while (isspace(*ptr)) ptr++; + ptr = skip_spaces(ptr); proc = get_fs_type("proc"); if (proc == NULL) { @@ -212,8 +213,7 @@ void mconsole_proc(struct mc_request *req) char *ptr = req->request.data; ptr += strlen("proc"); - while (isspace(*ptr)) - ptr++; + ptr = skip_spaces(ptr); snprintf(path, sizeof(path), "/proc/%s", ptr); fd = sys_open(path, 0, 0); @@ -560,8 +560,7 @@ void mconsole_config(struct mc_request *req) int err; ptr += strlen("config"); - while (isspace(*ptr)) - ptr++; + ptr = skip_spaces(ptr); dev = mconsole_find_dev(ptr); if (dev == NULL) { mconsole_reply(req, "Bad configuration option", 1, 0); @@ -588,7 +587,7 @@ void mconsole_remove(struct mc_request *req) int err, start, end, n; ptr += strlen("remove"); - while (isspace(*ptr)) ptr++; + ptr = skip_spaces(ptr); dev = mconsole_find_dev(ptr); if (dev == NULL) { mconsole_reply(req, "Bad remove option", 1, 0); @@ -712,7 +711,7 @@ void mconsole_sysrq(struct mc_request *req) char *ptr = req->request.data; ptr += strlen("sysrq"); - while (isspace(*ptr)) ptr++; + ptr = skip_spaces(ptr); /* * With 'b', the system will shut down without a chance to reply, @@ -757,8 +756,7 @@ void mconsole_stack(struct mc_request *req) */ ptr += strlen("stack"); - while (isspace(*ptr)) - ptr++; + ptr = skip_spaces(ptr); /* * Should really check for multiple pids or reject bad args here diff --git a/arch/x86/kernel/cpu/mtrr/if.c b/arch/x86/kernel/cpu/mtrr/if.c index 3c1b12d461d1..e006e56f699c 100644 --- a/arch/x86/kernel/cpu/mtrr/if.c +++ b/arch/x86/kernel/cpu/mtrr/if.c @@ -4,6 +4,7 @@ #include #include #include +#include #include #define LINE_SIZE 80 @@ -133,8 +134,7 @@ mtrr_write(struct file *file, const char __user *buf, size_t len, loff_t * ppos) return -EINVAL; base = simple_strtoull(line + 5, &ptr, 0); - while (isspace(*ptr)) - ptr++; + ptr = skip_spaces(ptr); if (strncmp(ptr, "size=", 5)) return -EINVAL; @@ -142,14 +142,11 @@ mtrr_write(struct file *file, const char __user *buf, size_t len, loff_t * ppos) size = simple_strtoull(ptr + 5, &ptr, 0); if ((base & 0xfff) || (size & 0xfff)) return -EINVAL; - while (isspace(*ptr)) - ptr++; + ptr = skip_spaces(ptr); if (strncmp(ptr, "type=", 5)) return -EINVAL; - ptr += 5; - while (isspace(*ptr)) - ptr++; + ptr = skip_spaces(ptr + 5); for (i = 0; i < MTRR_NUM_TYPES; ++i) { if (strcmp(ptr, mtrr_strings[i])) diff --git a/drivers/leds/led-class.c b/drivers/leds/led-class.c index f2cc13d76810..782f95822eab 100644 --- a/drivers/leds/led-class.c +++ b/drivers/leds/led-class.c @@ -50,7 +50,7 @@ static ssize_t led_brightness_store(struct device *dev, unsigned long state = simple_strtoul(buf, &after, 10); size_t count = after - buf; - if (*after && isspace(*after)) + if (isspace(*after)) count++; if (count == size) { diff --git a/drivers/leds/ledtrig-timer.c b/drivers/leds/ledtrig-timer.c index 3b83406de752..38b3378be442 100644 --- a/drivers/leds/ledtrig-timer.c +++ b/drivers/leds/ledtrig-timer.c @@ -83,7 +83,7 @@ static ssize_t led_delay_on_store(struct device *dev, unsigned long state = simple_strtoul(buf, &after, 10); size_t count = after - buf; - if (*after && isspace(*after)) + if (isspace(*after)) count++; if (count == size) { @@ -127,7 +127,7 @@ static ssize_t led_delay_off_store(struct device *dev, unsigned long state = simple_strtoul(buf, &after, 10); size_t count = after - buf; - if (*after && isspace(*after)) + if (isspace(*after)) count++; if (count == size) { diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c index 1a6cb3c7822e..91976e8fae5f 100644 --- a/drivers/md/dm-table.c +++ b/drivers/md/dm-table.c @@ -12,6 +12,7 @@ #include #include #include +#include #include #include #include @@ -600,11 +601,8 @@ int dm_split_args(int *argc, char ***argvp, char *input) return -ENOMEM; while (1) { - start = end; - /* Skip whitespace */ - while (*start && isspace(*start)) - start++; + start = skip_spaces(end); if (!*start) break; /* success, we hit the end */ diff --git a/drivers/md/md.c b/drivers/md/md.c index e1f3c1715cca..6f9148623a4d 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -39,6 +39,7 @@ #include /* for invalidate_bdev */ #include #include +#include #include #include #include @@ -3439,8 +3440,7 @@ bitmap_store(mddev_t *mddev, const char *buf, size_t len) } if (*end && !isspace(*end)) break; bitmap_dirty_bits(mddev->bitmap, chunk, end_chunk); - buf = end; - while (isspace(*buf)) buf++; + buf = skip_spaces(end); } bitmap_unplug(mddev->bitmap); /* flush the bits to disk */ out: diff --git a/drivers/parisc/pdc_stable.c b/drivers/parisc/pdc_stable.c index 13a64bc081b6..0bc5d474b168 100644 --- a/drivers/parisc/pdc_stable.c +++ b/drivers/parisc/pdc_stable.c @@ -779,12 +779,9 @@ static ssize_t pdcs_auto_write(struct kobject *kobj, read_unlock(&pathentry->rw_lock); DPRINTK("%s: flags before: 0x%X\n", __func__, flags); - - temp = in; - - while (*temp && isspace(*temp)) - temp++; - + + temp = skip_spaces(in); + c = *temp++ - '0'; if ((c != 0) && (c != 1)) goto parse_error; diff --git a/drivers/platform/x86/thinkpad_acpi.c b/drivers/platform/x86/thinkpad_acpi.c index 0ed84806f8ae..cf61d6a8ef6f 100644 --- a/drivers/platform/x86/thinkpad_acpi.c +++ b/drivers/platform/x86/thinkpad_acpi.c @@ -1006,11 +1006,8 @@ static int parse_strtoul(const char *buf, { char *endp; - while (*buf && isspace(*buf)) - buf++; - *value = simple_strtoul(buf, &endp, 0); - while (*endp && isspace(*endp)) - endp++; + *value = simple_strtoul(skip_spaces(buf), &endp, 0); + endp = skip_spaces(endp); if (*endp || *value > max) return -EINVAL; diff --git a/drivers/pnp/interface.c b/drivers/pnp/interface.c index c3f1c8e9d254..68b0c04987e4 100644 --- a/drivers/pnp/interface.c +++ b/drivers/pnp/interface.c @@ -310,8 +310,7 @@ static ssize_t pnp_set_current_resources(struct device *dmdev, goto done; } - while (isspace(*buf)) - ++buf; + buf = skip_spaces(buf); if (!strnicmp(buf, "disable", 7)) { retval = pnp_disable_dev(dev); goto done; @@ -353,19 +352,13 @@ static ssize_t pnp_set_current_resources(struct device *dmdev, pnp_init_resources(dev); mutex_lock(&pnp_res_mutex); while (1) { - while (isspace(*buf)) - ++buf; + buf = skip_spaces(buf); if (!strnicmp(buf, "io", 2)) { - buf += 2; - while (isspace(*buf)) - ++buf; + buf = skip_spaces(buf + 2); start = simple_strtoul(buf, &buf, 0); - while (isspace(*buf)) - ++buf; + buf = skip_spaces(buf); if (*buf == '-') { - buf += 1; - while (isspace(*buf)) - ++buf; + buf = skip_spaces(buf + 1); end = simple_strtoul(buf, &buf, 0); } else end = start; @@ -373,16 +366,11 @@ static ssize_t pnp_set_current_resources(struct device *dmdev, continue; } if (!strnicmp(buf, "mem", 3)) { - buf += 3; - while (isspace(*buf)) - ++buf; + buf = skip_spaces(buf + 3); start = simple_strtoul(buf, &buf, 0); - while (isspace(*buf)) - ++buf; + buf = skip_spaces(buf); if (*buf == '-') { - buf += 1; - while (isspace(*buf)) - ++buf; + buf = skip_spaces(buf + 1); end = simple_strtoul(buf, &buf, 0); } else end = start; @@ -390,17 +378,13 @@ static ssize_t pnp_set_current_resources(struct device *dmdev, continue; } if (!strnicmp(buf, "irq", 3)) { - buf += 3; - while (isspace(*buf)) - ++buf; + buf = skip_spaces(buf + 3); start = simple_strtoul(buf, &buf, 0); pnp_add_irq_resource(dev, start, 0); continue; } if (!strnicmp(buf, "dma", 3)) { - buf += 3; - while (isspace(*buf)) - ++buf; + buf = skip_spaces(buf + 3); start = simple_strtoul(buf, &buf, 0); pnp_add_dma_resource(dev, start, 0); continue; diff --git a/drivers/s390/block/dasd_proc.c b/drivers/s390/block/dasd_proc.c index 5f23eca82804..6315fbd8e68b 100644 --- a/drivers/s390/block/dasd_proc.c +++ b/drivers/s390/block/dasd_proc.c @@ -14,6 +14,7 @@ #define KMSG_COMPONENT "dasd" #include +#include #include #include #include @@ -272,10 +273,10 @@ dasd_statistics_write(struct file *file, const char __user *user_buf, DBF_EVENT(DBF_DEBUG, "/proc/dasd/statictics: '%s'\n", buffer); /* check for valid verbs */ - for (str = buffer; isspace(*str); str++); + str = skip_spaces(buffer); if (strncmp(str, "set", 3) == 0 && isspace(str[3])) { /* 'set xxx' was given */ - for (str = str + 4; isspace(*str); str++); + str = skip_spaces(str + 4); if (strcmp(str, "on") == 0) { /* switch on statistics profiling */ dasd_profile_level = DASD_PROFILE_ON; diff --git a/drivers/video/backlight/lcd.c b/drivers/video/backlight/lcd.c index a482dd7b0311..9b3be74cee5a 100644 --- a/drivers/video/backlight/lcd.c +++ b/drivers/video/backlight/lcd.c @@ -101,7 +101,7 @@ static ssize_t lcd_store_power(struct device *dev, int power = simple_strtoul(buf, &endp, 0); size_t size = endp - buf; - if (*endp && isspace(*endp)) + if (isspace(*endp)) size++; if (size != count) return -EINVAL; @@ -140,7 +140,7 @@ static ssize_t lcd_store_contrast(struct device *dev, int contrast = simple_strtoul(buf, &endp, 0); size_t size = endp - buf; - if (*endp && isspace(*endp)) + if (isspace(*endp)) size++; if (size != count) return -EINVAL; diff --git a/drivers/video/display/display-sysfs.c b/drivers/video/display/display-sysfs.c index 4830b1bf51e5..80abbf323b99 100644 --- a/drivers/video/display/display-sysfs.c +++ b/drivers/video/display/display-sysfs.c @@ -67,7 +67,7 @@ static ssize_t display_store_contrast(struct device *dev, contrast = simple_strtoul(buf, &endp, 0); size = endp - buf; - if (*endp && isspace(*endp)) + if (isspace(*endp)) size++; if (size != count) diff --git a/drivers/video/output.c b/drivers/video/output.c index 5e6439ae7394..5137aa016b83 100644 --- a/drivers/video/output.c +++ b/drivers/video/output.c @@ -50,7 +50,7 @@ static ssize_t video_output_store_state(struct device *dev, int request_state = simple_strtoul(buf,&endp,0); size_t size = endp - buf; - if (*endp && isspace(*endp)) + if (isspace(*endp)) size++; if (size != count) return -EINVAL; diff --git a/fs/cachefiles/daemon.c b/fs/cachefiles/daemon.c index 4618516dd994..c2413561ea75 100644 --- a/fs/cachefiles/daemon.c +++ b/fs/cachefiles/daemon.c @@ -21,6 +21,7 @@ #include #include #include +#include #include #include "internal.h" @@ -257,8 +258,7 @@ static ssize_t cachefiles_daemon_write(struct file *file, if (args == data) goto error; *args = '\0'; - for (args++; isspace(*args); args++) - continue; + args = skip_spaces(++args); } /* run the appropriate command handler */ diff --git a/fs/ext4/super.c b/fs/ext4/super.c index 768c111a77ec..827bde1f2594 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -2137,11 +2137,8 @@ static int parse_strtoul(const char *buf, { char *endp; - while (*buf && isspace(*buf)) - buf++; - *value = simple_strtoul(buf, &endp, 0); - while (*endp && isspace(*endp)) - endp++; + *value = simple_strtoul(skip_spaces(buf), &endp, 0); + endp = skip_spaces(endp); if (*endp || *value > max) return -EINVAL; diff --git a/kernel/params.c b/kernel/params.c index d656c276508d..cf1b69183127 100644 --- a/kernel/params.c +++ b/kernel/params.c @@ -24,6 +24,7 @@ #include #include #include +#include #if 0 #define DEBUGP printk @@ -122,9 +123,7 @@ static char *next_arg(char *args, char **param, char **val) next = args + i; /* Chew up trailing spaces. */ - while (isspace(*next)) - next++; - return next; + return skip_spaces(next); } /* Args looks like "foo=bar,bar2 baz=fuz wiz". */ @@ -139,8 +138,7 @@ int parse_args(const char *name, DEBUGP("Parsing ARGS: %s\n", args); /* Chew leading spaces */ - while (isspace(*args)) - args++; + args = skip_spaces(args); while (*args) { int ret; diff --git a/lib/argv_split.c b/lib/argv_split.c index 5205a8dae5bc..4b1b083f219c 100644 --- a/lib/argv_split.c +++ b/lib/argv_split.c @@ -4,17 +4,10 @@ #include #include +#include #include #include -static const char *skip_sep(const char *cp) -{ - while (*cp && isspace(*cp)) - cp++; - - return cp; -} - static const char *skip_arg(const char *cp) { while (*cp && !isspace(*cp)) @@ -28,7 +21,7 @@ static int count_argc(const char *str) int count = 0; while (*str) { - str = skip_sep(str); + str = skip_spaces(str); if (*str) { count++; str = skip_arg(str); @@ -82,7 +75,7 @@ char **argv_split(gfp_t gfp, const char *str, int *argcp) argvp = argv; while (*str) { - str = skip_sep(str); + str = skip_spaces(str); if (*str) { const char *p = str; diff --git a/lib/dynamic_debug.c b/lib/dynamic_debug.c index e22c148e4b7f..f93502915988 100644 --- a/lib/dynamic_debug.c +++ b/lib/dynamic_debug.c @@ -21,6 +21,7 @@ #include #include #include +#include #include #include #include @@ -209,8 +210,7 @@ static int ddebug_tokenize(char *buf, char *words[], int maxwords) char *end; /* Skip leading whitespace */ - while (*buf && isspace(*buf)) - buf++; + buf = skip_spaces(buf); if (!*buf) break; /* oh, it was trailing whitespace */ diff --git a/lib/vsprintf.c b/lib/vsprintf.c index c50733a690f0..7857d4dd62d3 100644 --- a/lib/vsprintf.c +++ b/lib/vsprintf.c @@ -1766,13 +1766,6 @@ EXPORT_SYMBOL_GPL(bprintf); #endif /* CONFIG_BINARY_PRINTF */ -static noinline char *skip_space(const char *str) -{ - while (isspace(*str)) - ++str; - return (char *)str; -} - /** * vsscanf - Unformat a buffer into a list of arguments * @buf: input buffer @@ -1794,8 +1787,8 @@ int vsscanf(const char *buf, const char *fmt, va_list args) * white space, including none, in the input. */ if (isspace(*fmt)) { - fmt = skip_space(fmt); - str = skip_space(str); + fmt = skip_spaces(++fmt); + str = skip_spaces(str); } /* anything that is not a conversion must match exactly */ @@ -1865,7 +1858,7 @@ int vsscanf(const char *buf, const char *fmt, va_list args) if (field_width == -1) field_width = INT_MAX; /* first, skip leading white space in buffer */ - str = skip_space(str); + str = skip_spaces(str); /* now copy until next white space */ while (*str && !isspace(*str) && field_width--) @@ -1907,7 +1900,7 @@ int vsscanf(const char *buf, const char *fmt, va_list args) /* have some sort of integer conversion. * first, skip white space in buffer. */ - str = skip_space(str); + str = skip_spaces(str); digit = *str; if (is_sign && digit == '-') diff --git a/net/irda/irnet/irnet.h b/net/irda/irnet/irnet.h index b001c361ad30..4300df35d37d 100644 --- a/net/irda/irnet/irnet.h +++ b/net/irda/irnet/irnet.h @@ -249,6 +249,7 @@ #include #include #include /* isspace() */ +#include /* skip_spaces() */ #include #include diff --git a/net/irda/irnet/irnet_ppp.c b/net/irda/irnet/irnet_ppp.c index 7dea882dbb75..156020d138b5 100644 --- a/net/irda/irnet/irnet_ppp.c +++ b/net/irda/irnet/irnet_ppp.c @@ -76,9 +76,8 @@ irnet_ctrl_write(irnet_socket * ap, /* Look at the next command */ start = next; - /* Scrap whitespaces before the command */ - while(isspace(*start)) - start++; + /* Scrap whitespaces before the command */ + start = skip_spaces(start); /* ',' is our command separator */ next = strchr(start, ','); @@ -133,8 +132,7 @@ irnet_ctrl_write(irnet_socket * ap, char * endp; /* Scrap whitespaces before the command */ - while(isspace(*begp)) - begp++; + begp = skip_spaces(begp); /* Convert argument to a number (last arg is the base) */ addr = simple_strtoul(begp, &endp, 16); diff --git a/net/netfilter/xt_recent.c b/net/netfilter/xt_recent.c index eb0ceb846527..fc70a49c0afd 100644 --- a/net/netfilter/xt_recent.c +++ b/net/netfilter/xt_recent.c @@ -482,8 +482,7 @@ static ssize_t recent_old_proc_write(struct file *file, if (copy_from_user(buf, input, size)) return -EFAULT; - while (isspace(*c)) - c++; + c = skip_spaces(c); if (size - (c - buf) < 5) return c - buf; diff --git a/sound/pci/hda/hda_hwdep.c b/sound/pci/hda/hda_hwdep.c index d24328661c6a..40ccb419b6e9 100644 --- a/sound/pci/hda/hda_hwdep.c +++ b/sound/pci/hda/hda_hwdep.c @@ -24,6 +24,7 @@ #include #include #include +#include #include #include #include "hda_codec.h" @@ -428,8 +429,7 @@ static int parse_hints(struct hda_codec *codec, const char *buf) char *key, *val; struct hda_hint *hint; - while (isspace(*buf)) - buf++; + buf = skip_spaces(buf); if (!*buf || *buf == '#' || *buf == '\n') return 0; if (*buf == '=') @@ -444,8 +444,7 @@ static int parse_hints(struct hda_codec *codec, const char *buf) return -EINVAL; } *val++ = 0; - while (isspace(*val)) - val++; + val = skip_spaces(val); remove_trail_spaces(key); remove_trail_spaces(val); hint = get_hint(codec, key); -- cgit v1.2.3-71-gd317 From f13c12c634e124d5d31f912b969d542a016d6105 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Tue, 15 Dec 2009 19:43:11 +0100 Subject: perf_events: Fix perf_event_attr layout The miss-alignment of bp_addr created a 32bit hole, causing different structure packings on 32 and 64 bit machines. Fix that by moving __reserve_2 into that hole. Further, remove the useless struct and redundant __bp_reserve muck. Signed-off-by: Peter Zijlstra Cc: Frederic Weisbecker Cc: Arnaldo Carvalho de Melo LKML-Reference: <1260902591.8023.781.camel@laptop> Signed-off-by: Ingo Molnar --- include/linux/perf_event.h | 12 +++--------- kernel/perf_event.c | 2 +- 2 files changed, 4 insertions(+), 10 deletions(-) (limited to 'kernel') diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h index 64a53f74c9a9..5fcbf7d2712a 100644 --- a/include/linux/perf_event.h +++ b/include/linux/perf_event.h @@ -211,17 +211,11 @@ struct perf_event_attr { __u32 wakeup_watermark; /* bytes before wakeup */ }; - struct { /* Hardware breakpoint info */ - __u64 bp_addr; - __u32 bp_type; - __u32 bp_len; - __u64 __bp_reserved_1; - __u64 __bp_reserved_2; - }; - __u32 __reserved_2; - __u64 __reserved_3; + __u64 bp_addr; + __u32 bp_type; + __u32 bp_len; }; /* diff --git a/kernel/perf_event.c b/kernel/perf_event.c index 8823b0885183..0dd8e5d02c66 100644 --- a/kernel/perf_event.c +++ b/kernel/perf_event.c @@ -4564,7 +4564,7 @@ static int perf_copy_attr(struct perf_event_attr __user *uattr, if (attr->type >= PERF_TYPE_MAX) return -EINVAL; - if (attr->__reserved_1 || attr->__reserved_2 || attr->__reserved_3) + if (attr->__reserved_1 || attr->__reserved_2) return -EINVAL; if (attr->sample_type & ~(PERF_SAMPLE_MAX-1)) -- cgit v1.2.3-71-gd317 From f065f41f48569122b5bcddbd1ba2354f7cc29fdc Mon Sep 17 00:00:00 2001 From: Barry Song <21cnbao@gmail.com> Date: Tue, 15 Dec 2009 16:45:34 -0800 Subject: timecompare: fix half-Y2K38 problem in timecompare_update while calculating offset ktime will overflow from 03:14:07 UTC on Tuesday, 19 January 2038, ktime_add() in timecompare_update() will overflow a half earlier. As a result, wrong offset will be gotten, then cause some strange problems. Signed-off-by: Barry Song <21cnbao@gmail.com> Cc: Thomas Gleixner Cc: Ingo Molnar Cc: Patrick Ohly Cc: David S. Miller Cc: John Stultz Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/time/timecompare.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/time/timecompare.c b/kernel/time/timecompare.c index 96ff643a5a59..12f5c55090be 100644 --- a/kernel/time/timecompare.c +++ b/kernel/time/timecompare.c @@ -89,7 +89,7 @@ int timecompare_offset(struct timecompare *sync, * source time */ sample.offset = - ktime_to_ns(ktime_add(end, start)) / 2 - + (ktime_to_ns(end) + ktime_to_ns(start)) / 2 - ts; /* simple insertion sort based on duration */ -- cgit v1.2.3-71-gd317 From 28dfef8febe48f59cf1e7596e1992a6a1893ca24 Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Tue, 15 Dec 2009 16:46:48 -0800 Subject: const: constify remaining pipe_buf_operations Signed-off-by: Alexey Dobriyan Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/relay.c | 2 +- kernel/trace/trace.c | 4 ++-- net/core/skbuff.c | 2 +- 3 files changed, 4 insertions(+), 4 deletions(-) (limited to 'kernel') diff --git a/kernel/relay.c b/kernel/relay.c index 760c26209a3c..c705a41b4ba3 100644 --- a/kernel/relay.c +++ b/kernel/relay.c @@ -1198,7 +1198,7 @@ static void relay_pipe_buf_release(struct pipe_inode_info *pipe, relay_consume_bytes(rbuf, buf->private); } -static struct pipe_buf_operations relay_pipe_buf_ops = { +static const struct pipe_buf_operations relay_pipe_buf_ops = { .can_merge = 0, .map = generic_pipe_buf_map, .unmap = generic_pipe_buf_unmap, diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index bb6b5e7fa2a2..31118ae16f03 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -3133,7 +3133,7 @@ static void tracing_spd_release_pipe(struct splice_pipe_desc *spd, __free_page(spd->pages[idx]); } -static struct pipe_buf_operations tracing_pipe_buf_ops = { +static const struct pipe_buf_operations tracing_pipe_buf_ops = { .can_merge = 0, .map = generic_pipe_buf_map, .unmap = generic_pipe_buf_unmap, @@ -3617,7 +3617,7 @@ static void buffer_pipe_buf_get(struct pipe_inode_info *pipe, } /* Pipe buffer operations for a buffer. */ -static struct pipe_buf_operations buffer_pipe_buf_ops = { +static const struct pipe_buf_operations buffer_pipe_buf_ops = { .can_merge = 0, .map = generic_pipe_buf_map, .unmap = generic_pipe_buf_unmap, diff --git a/net/core/skbuff.c b/net/core/skbuff.c index bfa3e7865a8c..93c4e060c91e 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -93,7 +93,7 @@ static int sock_pipe_buf_steal(struct pipe_inode_info *pipe, /* Pipe buffer operations for a socket. */ -static struct pipe_buf_operations sock_pipe_buf_ops = { +static const struct pipe_buf_operations sock_pipe_buf_ops = { .can_merge = 0, .map = generic_pipe_buf_map, .unmap = generic_pipe_buf_unmap, -- cgit v1.2.3-71-gd317 From 569b846df54ffb2827b83ce3244c5f032394cba4 Mon Sep 17 00:00:00 2001 From: KAMEZAWA Hiroyuki Date: Tue, 15 Dec 2009 16:47:03 -0800 Subject: memcg: coalesce uncharge during unmap/truncate In massive parallel enviroment, res_counter can be a performance bottleneck. One strong techinque to reduce lock contention is reducing calls by coalescing some amount of calls into one. Considering charge/uncharge chatacteristic, - charge is done one by one via demand-paging. - uncharge is done by - in chunk at munmap, truncate, exit, execve... - one by one via vmscan/paging. It seems we have a chance to coalesce uncharges for improving scalability at unmap/truncation. This patch is a for coalescing uncharge. For avoiding scattering memcg's structure to functions under /mm, this patch adds memcg batch uncharge information to the task. A reason for per-task batching is for making use of caller's context information. We do batched uncharge (deleyed uncharge) when truncation/unmap occurs but do direct uncharge when uncharge is called by memory reclaim (vmscan.c). The degree of coalescing depends on callers - at invalidate/trucate... pagevec size - at unmap ....ZAP_BLOCK_SIZE (memory itself will be freed in this degree.) Then, we'll not coalescing too much. On x86-64 8cpu server, I tested overheads of memcg at page fault by running a program which does map/fault/unmap in a loop. Running a task per a cpu by taskset and see sum of the number of page faults in 60secs. [without memcg config] 40156968 page-faults # 0.085 M/sec ( +- 0.046% ) 27.67 cache-miss/faults [root cgroup] 36659599 page-faults # 0.077 M/sec ( +- 0.247% ) 31.58 miss/faults [in a child cgroup] 18444157 page-faults # 0.039 M/sec ( +- 0.133% ) 69.96 miss/faults [child with this patch] 27133719 page-faults # 0.057 M/sec ( +- 0.155% ) 47.16 miss/faults We can see some amounts of improvement. (root cgroup doesn't affected by this patch) Another patch for "charge" will follow this and above will be improved more. Changelog(since 2009/10/02): - renamed filed of memcg_batch (as pages to bytes, memsw to memsw_bytes) - some clean up and commentary/description updates. - added initialize code to copy_process(). (possible bug fix) Changelog(old): - fixed !CONFIG_MEM_CGROUP case. - rebased onto the latest mmotm + softlimit fix patches. - unified patch for callers - added commetns. - make ->do_batch as bool. - removed css_get() at el. We don't need it. Signed-off-by: KAMEZAWA Hiroyuki Cc: Balbir Singh Cc: Daisuke Nishimura Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/memcontrol.h | 13 +++++++ include/linux/sched.h | 8 ++++ kernel/fork.c | 4 ++ mm/memcontrol.c | 96 +++++++++++++++++++++++++++++++++++++++++++--- mm/memory.c | 2 + mm/truncate.c | 6 +++ 6 files changed, 123 insertions(+), 6 deletions(-) (limited to 'kernel') diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h index bf9213b2db8f..91300c972e76 100644 --- a/include/linux/memcontrol.h +++ b/include/linux/memcontrol.h @@ -54,6 +54,11 @@ extern void mem_cgroup_rotate_lru_list(struct page *page, enum lru_list lru); extern void mem_cgroup_del_lru(struct page *page); extern void mem_cgroup_move_lists(struct page *page, enum lru_list from, enum lru_list to); + +/* For coalescing uncharge for reducing memcg' overhead*/ +extern void mem_cgroup_uncharge_start(void); +extern void mem_cgroup_uncharge_end(void); + extern void mem_cgroup_uncharge_page(struct page *page); extern void mem_cgroup_uncharge_cache_page(struct page *page); extern int mem_cgroup_shmem_charge_fallback(struct page *page, @@ -151,6 +156,14 @@ static inline void mem_cgroup_cancel_charge_swapin(struct mem_cgroup *ptr) { } +static inline void mem_cgroup_uncharge_start(void) +{ +} + +static inline void mem_cgroup_uncharge_end(void) +{ +} + static inline void mem_cgroup_uncharge_page(struct page *page) { } diff --git a/include/linux/sched.h b/include/linux/sched.h index 5c858f38e81a..f4c145410a8d 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1544,6 +1544,14 @@ struct task_struct { unsigned long trace_recursion; #endif /* CONFIG_TRACING */ unsigned long stack_start; +#ifdef CONFIG_CGROUP_MEM_RES_CTLR /* memcg uses this to do batch job */ + struct memcg_batch_info { + int do_batch; /* incremented when batch uncharge started */ + struct mem_cgroup *memcg; /* target memcg of uncharge */ + unsigned long bytes; /* uncharged usage */ + unsigned long memsw_bytes; /* uncharged mem+swap usage */ + } memcg_batch; +#endif }; /* Future-safe accessor for struct task_struct's cpus_allowed. */ diff --git a/kernel/fork.c b/kernel/fork.c index 9bd91447e052..b6cbd33dde80 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -1127,6 +1127,10 @@ static struct task_struct *copy_process(unsigned long clone_flags, #ifdef CONFIG_DEBUG_MUTEXES p->blocked_on = NULL; /* not blocked yet */ #endif +#ifdef CONFIG_CGROUP_MEM_RES_CTLR + p->memcg_batch.do_batch = 0; + p->memcg_batch.memcg = NULL; +#endif p->bts = NULL; diff --git a/mm/memcontrol.c b/mm/memcontrol.c index 7b5b108c1c6b..a730c91b8e69 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -1827,6 +1827,50 @@ void mem_cgroup_cancel_charge_swapin(struct mem_cgroup *mem) css_put(&mem->css); } +static void +__do_uncharge(struct mem_cgroup *mem, const enum charge_type ctype) +{ + struct memcg_batch_info *batch = NULL; + bool uncharge_memsw = true; + /* If swapout, usage of swap doesn't decrease */ + if (!do_swap_account || ctype == MEM_CGROUP_CHARGE_TYPE_SWAPOUT) + uncharge_memsw = false; + /* + * do_batch > 0 when unmapping pages or inode invalidate/truncate. + * In those cases, all pages freed continously can be expected to be in + * the same cgroup and we have chance to coalesce uncharges. + * But we do uncharge one by one if this is killed by OOM(TIF_MEMDIE) + * because we want to do uncharge as soon as possible. + */ + if (!current->memcg_batch.do_batch || test_thread_flag(TIF_MEMDIE)) + goto direct_uncharge; + + batch = ¤t->memcg_batch; + /* + * In usual, we do css_get() when we remember memcg pointer. + * But in this case, we keep res->usage until end of a series of + * uncharges. Then, it's ok to ignore memcg's refcnt. + */ + if (!batch->memcg) + batch->memcg = mem; + /* + * In typical case, batch->memcg == mem. This means we can + * merge a series of uncharges to an uncharge of res_counter. + * If not, we uncharge res_counter ony by one. + */ + if (batch->memcg != mem) + goto direct_uncharge; + /* remember freed charge and uncharge it later */ + batch->bytes += PAGE_SIZE; + if (uncharge_memsw) + batch->memsw_bytes += PAGE_SIZE; + return; +direct_uncharge: + res_counter_uncharge(&mem->res, PAGE_SIZE); + if (uncharge_memsw) + res_counter_uncharge(&mem->memsw, PAGE_SIZE); + return; +} /* * uncharge if !page_mapped(page) @@ -1875,12 +1919,8 @@ __mem_cgroup_uncharge_common(struct page *page, enum charge_type ctype) break; } - if (!mem_cgroup_is_root(mem)) { - res_counter_uncharge(&mem->res, PAGE_SIZE); - if (do_swap_account && - (ctype != MEM_CGROUP_CHARGE_TYPE_SWAPOUT)) - res_counter_uncharge(&mem->memsw, PAGE_SIZE); - } + if (!mem_cgroup_is_root(mem)) + __do_uncharge(mem, ctype); if (ctype == MEM_CGROUP_CHARGE_TYPE_SWAPOUT) mem_cgroup_swap_statistics(mem, true); mem_cgroup_charge_statistics(mem, pc, false); @@ -1926,6 +1966,50 @@ void mem_cgroup_uncharge_cache_page(struct page *page) __mem_cgroup_uncharge_common(page, MEM_CGROUP_CHARGE_TYPE_CACHE); } +/* + * Batch_start/batch_end is called in unmap_page_range/invlidate/trucate. + * In that cases, pages are freed continuously and we can expect pages + * are in the same memcg. All these calls itself limits the number of + * pages freed at once, then uncharge_start/end() is called properly. + * This may be called prural(2) times in a context, + */ + +void mem_cgroup_uncharge_start(void) +{ + current->memcg_batch.do_batch++; + /* We can do nest. */ + if (current->memcg_batch.do_batch == 1) { + current->memcg_batch.memcg = NULL; + current->memcg_batch.bytes = 0; + current->memcg_batch.memsw_bytes = 0; + } +} + +void mem_cgroup_uncharge_end(void) +{ + struct memcg_batch_info *batch = ¤t->memcg_batch; + + if (!batch->do_batch) + return; + + batch->do_batch--; + if (batch->do_batch) /* If stacked, do nothing. */ + return; + + if (!batch->memcg) + return; + /* + * This "batch->memcg" is valid without any css_get/put etc... + * bacause we hide charges behind us. + */ + if (batch->bytes) + res_counter_uncharge(&batch->memcg->res, batch->bytes); + if (batch->memsw_bytes) + res_counter_uncharge(&batch->memcg->memsw, batch->memsw_bytes); + /* forget this pointer (for sanity check) */ + batch->memcg = NULL; +} + #ifdef CONFIG_SWAP /* * called after __delete_from_swap_cache() and drop "page" account. diff --git a/mm/memory.c b/mm/memory.c index a54b2c498444..aed45eaf8ac9 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -956,6 +956,7 @@ static unsigned long unmap_page_range(struct mmu_gather *tlb, details = NULL; BUG_ON(addr >= end); + mem_cgroup_uncharge_start(); tlb_start_vma(tlb, vma); pgd = pgd_offset(vma->vm_mm, addr); do { @@ -968,6 +969,7 @@ static unsigned long unmap_page_range(struct mmu_gather *tlb, zap_work, details); } while (pgd++, addr = next, (addr != end && *zap_work > 0)); tlb_end_vma(tlb, vma); + mem_cgroup_uncharge_end(); return addr; } diff --git a/mm/truncate.c b/mm/truncate.c index 2c147a7e5f2c..342deee22684 100644 --- a/mm/truncate.c +++ b/mm/truncate.c @@ -272,6 +272,7 @@ void truncate_inode_pages_range(struct address_space *mapping, pagevec_release(&pvec); break; } + mem_cgroup_uncharge_start(); for (i = 0; i < pagevec_count(&pvec); i++) { struct page *page = pvec.pages[i]; @@ -286,6 +287,7 @@ void truncate_inode_pages_range(struct address_space *mapping, unlock_page(page); } pagevec_release(&pvec); + mem_cgroup_uncharge_end(); } } EXPORT_SYMBOL(truncate_inode_pages_range); @@ -327,6 +329,7 @@ unsigned long invalidate_mapping_pages(struct address_space *mapping, pagevec_init(&pvec, 0); while (next <= end && pagevec_lookup(&pvec, mapping, next, PAGEVEC_SIZE)) { + mem_cgroup_uncharge_start(); for (i = 0; i < pagevec_count(&pvec); i++) { struct page *page = pvec.pages[i]; pgoff_t index; @@ -354,6 +357,7 @@ unsigned long invalidate_mapping_pages(struct address_space *mapping, break; } pagevec_release(&pvec); + mem_cgroup_uncharge_end(); cond_resched(); } return ret; @@ -428,6 +432,7 @@ int invalidate_inode_pages2_range(struct address_space *mapping, while (next <= end && !wrapped && pagevec_lookup(&pvec, mapping, next, min(end - next, (pgoff_t)PAGEVEC_SIZE - 1) + 1)) { + mem_cgroup_uncharge_start(); for (i = 0; i < pagevec_count(&pvec); i++) { struct page *page = pvec.pages[i]; pgoff_t page_index; @@ -477,6 +482,7 @@ int invalidate_inode_pages2_range(struct address_space *mapping, unlock_page(page); } pagevec_release(&pvec); + mem_cgroup_uncharge_end(); cond_resched(); } return ret; -- cgit v1.2.3-71-gd317 From 6580807da14c423f0d0a708108e6df6ebc8bc83d Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Tue, 15 Dec 2009 16:47:16 -0800 Subject: ptrace: copy_process() should disable stepping If the tracee calls fork() after PTRACE_SINGLESTEP, the forked child starts with TIF_SINGLESTEP/X86_EFLAGS_TF bits copied from ptraced parent. This is not right, especially when the new child is not auto-attaced: in this case it is killed by SIGTRAP. Change copy_process() to call user_disable_single_step(). Tested on x86. Test-case: #include #include #include #include #include #include int main(void) { int pid, status; if (!(pid = fork())) { assert(ptrace(PTRACE_TRACEME) == 0); kill(getpid(), SIGSTOP); if (!fork()) { /* kernel bug: this child will be killed by SIGTRAP */ printf("Hello world\n"); return 43; } wait(&status); return WEXITSTATUS(status); } for (;;) { assert(pid == wait(&status)); if (WIFEXITED(status)) break; assert(ptrace(PTRACE_SINGLESTEP, pid, 0,0) == 0); } assert(WEXITSTATUS(status) == 43); return 0; } Signed-off-by: Oleg Nesterov Acked-by: Roland McGrath Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/fork.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/fork.c b/kernel/fork.c index b6cbd33dde80..202a0ba63d3c 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -1210,9 +1210,10 @@ static struct task_struct *copy_process(unsigned long clone_flags, p->sas_ss_sp = p->sas_ss_size = 0; /* - * Syscall tracing should be turned off in the child regardless - * of CLONE_PTRACE. + * Syscall tracing and stepping should be turned off in the + * child regardless of CLONE_PTRACE. */ + user_disable_single_step(p); clear_tsk_thread_flag(p, TIF_SYSCALL_TRACE); #ifdef TIF_SYSCALL_EMU clear_tsk_thread_flag(p, TIF_SYSCALL_EMU); -- cgit v1.2.3-71-gd317 From 614c517d7c00af1b26ded20646b329397d6f51a1 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Tue, 15 Dec 2009 16:47:22 -0800 Subject: signals: SEND_SIG_NOINFO should be considered as SI_FROMUSER() No changes in compiled code. The patch adds the new helper, si_fromuser() and changes check_kill_permission() to use this helper. The real effect of this patch is that from now we "officially" consider SEND_SIG_NOINFO signal as "from user-space" signals. This is already true if we look at the code which uses SEND_SIG_NOINFO, except __send_signal() has another opinion - see the next patch. The naming of these special SEND_SIG_XXX siginfo's is really bad imho. From __send_signal()'s pov they mean SEND_SIG_NOINFO from user SEND_SIG_PRIV from kernel SEND_SIG_FORCED no info Signed-off-by: Oleg Nesterov Cc: Roland McGrath Reviewed-by: Sukadev Bhattiprolu Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/sched.h | 5 ----- kernel/signal.c | 16 +++++++++++++--- 2 files changed, 13 insertions(+), 8 deletions(-) (limited to 'kernel') diff --git a/include/linux/sched.h b/include/linux/sched.h index f4c145410a8d..57b3516f055b 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -2102,11 +2102,6 @@ static inline int kill_cad_pid(int sig, int priv) #define SEND_SIG_PRIV ((struct siginfo *) 1) #define SEND_SIG_FORCED ((struct siginfo *) 2) -static inline int is_si_special(const struct siginfo *info) -{ - return info <= SEND_SIG_FORCED; -} - /* * True if we are on the alternate signal stack. */ diff --git a/kernel/signal.c b/kernel/signal.c index 6b982f2cf524..a0ba428954b6 100644 --- a/kernel/signal.c +++ b/kernel/signal.c @@ -607,6 +607,17 @@ static int rm_from_queue(unsigned long mask, struct sigpending *s) return 1; } +static inline int is_si_special(const struct siginfo *info) +{ + return info <= SEND_SIG_FORCED; +} + +static inline bool si_fromuser(const struct siginfo *info) +{ + return info == SEND_SIG_NOINFO || + (!is_si_special(info) && SI_FROMUSER(info)); +} + /* * Bad permissions for sending the signal * - the caller must hold at least the RCU read lock @@ -621,7 +632,7 @@ static int check_kill_permission(int sig, struct siginfo *info, if (!valid_signal(sig)) return -EINVAL; - if (info != SEND_SIG_NOINFO && (is_si_special(info) || SI_FROMKERNEL(info))) + if (!si_fromuser(info)) return 0; error = audit_signal_info(sig, t); /* Let audit system see the signal */ @@ -1186,8 +1197,7 @@ int kill_pid_info_as_uid(int sig, struct siginfo *info, struct pid *pid, goto out_unlock; } pcred = __task_cred(p); - if ((info == SEND_SIG_NOINFO || - (!is_si_special(info) && SI_FROMUSER(info))) && + if (si_fromuser(info) && euid != pcred->suid && euid != pcred->uid && uid != pcred->suid && uid != pcred->uid) { ret = -EPERM; -- cgit v1.2.3-71-gd317 From dd34200adc01c5217ef09b55905b5c2312d65535 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Tue, 15 Dec 2009 16:47:24 -0800 Subject: signals: send_signal: use si_fromuser() to detect from_ancestor_ns Change send_signal() to use si_fromuser(). From now SEND_SIG_NOINFO triggers the "from_ancestor_ns" check. This fixes reparent_thread()->group_send_sig_info(pdeath_signal) behaviour, before this patch send_signal() does not detect the cross-namespace case when the child of the dying parent belongs to the sub-namespace. This patch can affect the behaviour of send_sig(), kill_pgrp() and kill_pid() when the caller sends the signal to the sub-namespace with "priv == 0" but surprisingly all callers seem to use them correctly, including disassociate_ctty(on_exit). Except: drivers/staging/comedi/drivers/addi-data/*.c incorrectly use send_sig(priv => 0). But his is minor and should be fixed anyway. Reported-by: Daniel Lezcano Signed-off-by: Oleg Nesterov Cc: Roland McGrath Reviewed-by: Sukadev Bhattiprolu Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/signal.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) (limited to 'kernel') diff --git a/kernel/signal.c b/kernel/signal.c index a0ba428954b6..b65879d4e08f 100644 --- a/kernel/signal.c +++ b/kernel/signal.c @@ -960,9 +960,8 @@ static int send_signal(int sig, struct siginfo *info, struct task_struct *t, int from_ancestor_ns = 0; #ifdef CONFIG_PID_NS - if (!is_si_special(info) && SI_FROMUSER(info) && - task_pid_nr_ns(current, task_active_pid_ns(t)) <= 0) - from_ancestor_ns = 1; + from_ancestor_ns = si_fromuser(info) && + !task_pid_nr_ns(current, task_active_pid_ns(t)); #endif return __send_signal(sig, info, t, group, from_ancestor_ns); -- cgit v1.2.3-71-gd317 From 7486e5d9fc773cb67c29381567bed5236fc9573c Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Tue, 15 Dec 2009 16:47:24 -0800 Subject: signals: cosmetic, collect_signal: use SI_USER Trivial, s/0/SI_USER/ in collect_signal() for grep. This is a bit confusing, we don't know the source of this signal. But we don't care, and "info->si_code = 0" is imho worse. Signed-off-by: Oleg Nesterov Cc: Roland McGrath Cc: Sukadev Bhattiprolu Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/signal.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/signal.c b/kernel/signal.c index b65879d4e08f..d7c7f3cd4da8 100644 --- a/kernel/signal.c +++ b/kernel/signal.c @@ -423,7 +423,7 @@ still_pending: */ info->si_signo = sig; info->si_errno = 0; - info->si_code = 0; + info->si_code = SI_USER; info->si_pid = 0; info->si_uid = 0; } -- cgit v1.2.3-71-gd317 From ad09750b51150ca87531b8790a379214a974c167 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Tue, 15 Dec 2009 16:47:25 -0800 Subject: signals: kill force_sig_specific() Kill force_sig_specific(), this trivial wrapper has no callers. Signed-off-by: Oleg Nesterov Cc: Roland McGrath Cc: Sukadev Bhattiprolu Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/sched.h | 1 - kernel/signal.c | 6 ------ 2 files changed, 7 deletions(-) (limited to 'kernel') diff --git a/include/linux/sched.h b/include/linux/sched.h index 57b3516f055b..244c287a5ac1 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -2083,7 +2083,6 @@ extern int kill_proc_info(int, struct siginfo *, pid_t); extern int do_notify_parent(struct task_struct *, int); extern void __wake_up_parent(struct task_struct *p, struct task_struct *parent); extern void force_sig(int, struct task_struct *); -extern void force_sig_specific(int, struct task_struct *); extern int send_sig(int, struct task_struct *, int); extern void zap_other_threads(struct task_struct *p); extern struct sigqueue *sigqueue_alloc(void); diff --git a/kernel/signal.c b/kernel/signal.c index d7c7f3cd4da8..4a9d763f8922 100644 --- a/kernel/signal.c +++ b/kernel/signal.c @@ -1062,12 +1062,6 @@ force_sig_info(int sig, struct siginfo *info, struct task_struct *t) return ret; } -void -force_sig_specific(int sig, struct task_struct *t) -{ - force_sig_info(sig, SEND_SIG_FORCED, t); -} - /* * Nuke all other threads in the group. */ -- cgit v1.2.3-71-gd317 From 1be53963b0519bd3681749a9bed8b83aeb005cca Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Tue, 15 Dec 2009 16:47:26 -0800 Subject: signals: check ->group_stop_count after tracehook_get_signal() Move the call to do_signal_stop() down, after tracehook call. This makes ->group_stop_count condition visible to tracers before do_signal_stop() will participate in this group-stop. Currently the patch has no effect, tracehook_get_signal() always returns 0. Signed-off-by: Oleg Nesterov Acked-by: Roland McGrath Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/signal.c | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-) (limited to 'kernel') diff --git a/kernel/signal.c b/kernel/signal.c index 4a9d763f8922..1814e68e4de3 100644 --- a/kernel/signal.c +++ b/kernel/signal.c @@ -1840,11 +1840,6 @@ relock: for (;;) { struct k_sigaction *ka; - - if (unlikely(signal->group_stop_count > 0) && - do_signal_stop(0)) - goto relock; - /* * Tracing can induce an artifical signal and choose sigaction. * The return value in @signr determines the default action, @@ -1856,6 +1851,10 @@ relock: if (unlikely(signr != 0)) ka = return_ka; else { + if (unlikely(signal->group_stop_count > 0) && + do_signal_stop(0)) + goto relock; + signr = dequeue_signal(current, ¤t->blocked, info); -- cgit v1.2.3-71-gd317 From 7be6d991bca63bbcdc5bc3b09789f367a3486537 Mon Sep 17 00:00:00 2001 From: André Goddard Rosa Date: Tue, 15 Dec 2009 16:47:39 -0800 Subject: pid: tighten pidmap spinlock critical section by removing kfree() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Avoid calling kfree() under pidmap spinlock, calling it afterwards. Normally kfree() is fast, but sometimes it can be slow, so avoid calling it under the spinlock if we can do it. Signed-off-by: André Goddard Rosa Cc: Pekka Enberg Cc: Oleg Nesterov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/pid.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) (limited to 'kernel') diff --git a/kernel/pid.c b/kernel/pid.c index d3f722d20f9c..55fd5900bac8 100644 --- a/kernel/pid.c +++ b/kernel/pid.c @@ -141,11 +141,12 @@ static int alloc_pidmap(struct pid_namespace *pid_ns) * installing it: */ spin_lock_irq(&pidmap_lock); - if (map->page) - kfree(page); - else + if (!map->page) { map->page = page; + page = NULL; + } spin_unlock_irq(&pidmap_lock); + kfree(page); if (unlikely(!map->page)) break; } -- cgit v1.2.3-71-gd317 From 417e315247ebc199d357855bb08d2a5264400565 Mon Sep 17 00:00:00 2001 From: André Goddard Rosa Date: Tue, 15 Dec 2009 16:47:40 -0800 Subject: pid: reduce code size by using a pointer to iterate over array MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit It decreases code size by 16 bytes on my gcc 4.4.1 on Core 2: text data bss dec hex filename 4314 2216 8 6538 198a kernel/pid.o-BEFORE 4298 2216 8 6522 197a kernel/pid.o-AFTER Signed-off-by: André Goddard Rosa Cc: Pekka Enberg Cc: Oleg Nesterov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/pid.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) (limited to 'kernel') diff --git a/kernel/pid.c b/kernel/pid.c index 55fd5900bac8..2e17c9c92cbe 100644 --- a/kernel/pid.c +++ b/kernel/pid.c @@ -269,12 +269,11 @@ struct pid *alloc_pid(struct pid_namespace *ns) for (type = 0; type < PIDTYPE_MAX; ++type) INIT_HLIST_HEAD(&pid->tasks[type]); + upid = pid->numbers + ns->level; spin_lock_irq(&pidmap_lock); - for (i = ns->level; i >= 0; i--) { - upid = &pid->numbers[i]; + for ( ; upid >= pid->numbers; --upid) hlist_add_head_rcu(&upid->pid_chain, &pid_hash[pid_hashfn(upid->nr, upid->ns)]); - } spin_unlock_irq(&pidmap_lock); out: -- cgit v1.2.3-71-gd317 From 06a7f711246b081afc21fff859f1003f1f2a0fbc Mon Sep 17 00:00:00 2001 From: Amerigo Wang Date: Tue, 15 Dec 2009 16:47:46 -0800 Subject: kexec: premit reduction of the reserved memory size Implement shrinking the reserved memory for crash kernel, if it is more than enough. For example, if you have already reserved 128M, now you just want 100M, you can do: # echo $((100*1024*1024)) > /sys/kernel/kexec_crash_size Note, you can only do this before loading the crash kernel. Signed-off-by: WANG Cong Cc: Neil Horman Acked-by: Eric W. Biederman Cc: Andi Kleen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/kexec.h | 2 ++ kernel/kexec.c | 59 +++++++++++++++++++++++++++++++++++++++++++++++++++ kernel/ksysfs.c | 21 ++++++++++++++++++ 3 files changed, 82 insertions(+) (limited to 'kernel') diff --git a/include/linux/kexec.h b/include/linux/kexec.h index adc34f2c6eff..c356b6914ffd 100644 --- a/include/linux/kexec.h +++ b/include/linux/kexec.h @@ -206,6 +206,8 @@ extern size_t vmcoreinfo_max_size; int __init parse_crashkernel(char *cmdline, unsigned long long system_ram, unsigned long long *crash_size, unsigned long long *crash_base); +int crash_shrink_memory(unsigned long new_size); +size_t crash_get_memory_size(void); #else /* !CONFIG_KEXEC */ struct pt_regs; diff --git a/kernel/kexec.c b/kernel/kexec.c index f336e2107f98..433e9fcc1fc5 100644 --- a/kernel/kexec.c +++ b/kernel/kexec.c @@ -31,6 +31,7 @@ #include #include #include +#include #include #include @@ -1082,6 +1083,64 @@ void crash_kexec(struct pt_regs *regs) } } +size_t crash_get_memory_size(void) +{ + size_t size; + mutex_lock(&kexec_mutex); + size = crashk_res.end - crashk_res.start + 1; + mutex_unlock(&kexec_mutex); + return size; +} + +static void free_reserved_phys_range(unsigned long begin, unsigned long end) +{ + unsigned long addr; + + for (addr = begin; addr < end; addr += PAGE_SIZE) { + ClearPageReserved(pfn_to_page(addr >> PAGE_SHIFT)); + init_page_count(pfn_to_page(addr >> PAGE_SHIFT)); + free_page((unsigned long)__va(addr)); + totalram_pages++; + } +} + +int crash_shrink_memory(unsigned long new_size) +{ + int ret = 0; + unsigned long start, end; + + mutex_lock(&kexec_mutex); + + if (kexec_crash_image) { + ret = -ENOENT; + goto unlock; + } + start = crashk_res.start; + end = crashk_res.end; + + if (new_size >= end - start + 1) { + ret = -EINVAL; + if (new_size == end - start + 1) + ret = 0; + goto unlock; + } + + start = roundup(start, PAGE_SIZE); + end = roundup(start + new_size, PAGE_SIZE); + + free_reserved_phys_range(end, crashk_res.end); + + if (start == end) { + crashk_res.end = end; + release_resource(&crashk_res); + } else + crashk_res.end = end - 1; + +unlock: + mutex_unlock(&kexec_mutex); + return ret; +} + static u32 *append_elf_note(u32 *buf, char *name, unsigned type, void *data, size_t data_len) { diff --git a/kernel/ksysfs.c b/kernel/ksysfs.c index 528dd78e7e7e..3feaf5a74514 100644 --- a/kernel/ksysfs.c +++ b/kernel/ksysfs.c @@ -100,6 +100,26 @@ static ssize_t kexec_crash_loaded_show(struct kobject *kobj, } KERNEL_ATTR_RO(kexec_crash_loaded); +static ssize_t kexec_crash_size_show(struct kobject *kobj, + struct kobj_attribute *attr, char *buf) +{ + return sprintf(buf, "%zu\n", crash_get_memory_size()); +} +static ssize_t kexec_crash_size_store(struct kobject *kobj, + struct kobj_attribute *attr, + const char *buf, size_t count) +{ + unsigned long cnt; + int ret; + + if (strict_strtoul(buf, 0, &cnt)) + return -EINVAL; + + ret = crash_shrink_memory(cnt); + return ret < 0 ? ret : count; +} +KERNEL_ATTR_RW(kexec_crash_size); + static ssize_t vmcoreinfo_show(struct kobject *kobj, struct kobj_attribute *attr, char *buf) { @@ -147,6 +167,7 @@ static struct attribute * kernel_attrs[] = { #ifdef CONFIG_KEXEC &kexec_loaded_attr.attr, &kexec_crash_loaded_attr.attr, + &kexec_crash_size_attr.attr, &vmcoreinfo_attr.attr, #endif NULL -- cgit v1.2.3-71-gd317 From f4c4176f21533e22bcc292030da72bcfa105f5b8 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 16 Dec 2009 17:55:54 +0100 Subject: perf events: Allow per-task-per-cpu counters In order to allow for per-task-per-cpu counters, useful for scalability when profiling task hierarchies, we allow installing events with event->cpu != -1 in task contexts. __perf_event_sched_in() already skips events where ->cpu mis-matches the current cpu, fix up __perf_install_in_context() and __perf_event_enable() to also respect this filter. This does lead to vary hard to interpret enabled/running times for such counters, but I don't see a simple solution for that. Signed-off-by: Peter Zijlstra Cc: Arnaldo Carvalho de Melo Cc: fweisbec@gmail.com Cc: Paul Mackerras LKML-Reference: <20091216165904.831451147@chello.nl> Signed-off-by: Ingo Molnar --- kernel/perf_event.c | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) (limited to 'kernel') diff --git a/kernel/perf_event.c b/kernel/perf_event.c index 0dd8e5d02c66..2e0aaa34fc7e 100644 --- a/kernel/perf_event.c +++ b/kernel/perf_event.c @@ -782,6 +782,9 @@ static void __perf_install_in_context(void *info) add_event_to_ctx(event, ctx); + if (event->cpu != -1 && event->cpu != smp_processor_id()) + goto unlock; + /* * Don't put the event on if it is disabled or if * it is in a group and the group isn't on. @@ -925,6 +928,9 @@ static void __perf_event_enable(void *info) goto unlock; __perf_event_mark_enabled(event, ctx); + if (event->cpu != -1 && event->cpu != smp_processor_id()) + goto unlock; + /* * If the event is in a group and isn't the group leader, * then don't put it on unless the group is on. @@ -1595,10 +1601,7 @@ static struct perf_event_context *find_get_context(pid_t pid, int cpu) unsigned long flags; int err; - /* - * If cpu is not a wildcard then this is a percpu event: - */ - if (cpu != -1) { + if (pid == -1 && cpu != -1) { /* Must be root to operate on a CPU event: */ if (perf_paranoid_cpu() && !capable(CAP_SYS_ADMIN)) return ERR_PTR(-EACCES); -- cgit v1.2.3-71-gd317 From 61cf693159d6a968a7014e24905143f71ed8ddcf Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Wed, 16 Dec 2009 12:28:44 +0100 Subject: [sysctl] Fix breakage on systems with older glibc As predicted during code review, the sysctl(2) changes made systems with old glibc nearly unusable. About every command gives a: warning: process `ls' used the deprecated sysctl system call with 1.4 warning in the log. I see this on a SUSE 10.0 system with glibc 2.3.5. Don't warn for this common case. Signed-off-by: Andi Kleen Signed-off-by: Linus Torvalds --- kernel/sysctl_binary.c | 7 +++++++ 1 file changed, 7 insertions(+) (limited to 'kernel') diff --git a/kernel/sysctl_binary.c b/kernel/sysctl_binary.c index b75dbf40f573..112533d5fc08 100644 --- a/kernel/sysctl_binary.c +++ b/kernel/sysctl_binary.c @@ -1399,6 +1399,13 @@ static void deprecated_sysctl_warning(const int *name, int nlen) { int i; + /* + * CTL_KERN/KERN_VERSION is used by older glibc and cannot + * ever go away. + */ + if (name[0] == CTL_KERN && name[1] == KERN_VERSION) + return; + if (printk_ratelimit()) { printk(KERN_INFO "warning: process `%s' used the deprecated sysctl " -- cgit v1.2.3-71-gd317