From e1d8aa9f1dd655a3534b22fcfbecb70cdb125766 Mon Sep 17 00:00:00 2001 From: Frederic Weisbecker Date: Mon, 12 Jan 2009 23:15:46 +0100 Subject: tracing: add a new workqueue tracer Impact: new tracer The workqueue tracer provides some statistical informations about each cpu workqueue thread such as the number of the works inserted and executed since their creation. It can help to evaluate the amount of work each of them have to perform. For example it can help a developer to decide whether he should choose a per cpu workqueue instead of a singlethreaded one. It only traces statistical informations for now but it will probably later provide event tracing too. Such a tracer could help too, and be improved, to help rt priority sorted workqueue development. To have a snapshot of the workqueues state at any time, just do cat /debugfs/tracing/trace_stat/workqueues Ie: 1 125 125 reiserfs/1 1 0 0 scsi_tgtd/1 1 0 0 aio/1 1 0 0 ata/1 1 114 114 kblockd/1 1 0 0 kintegrityd/1 1 2147 2147 events/1 0 0 0 kpsmoused 0 105 105 reiserfs/0 0 0 0 scsi_tgtd/0 0 0 0 aio/0 0 0 0 ata_aux 0 0 0 ata/0 0 0 0 cqueue 0 0 0 kacpi_notify 0 0 0 kacpid 0 149 149 kblockd/0 0 0 0 kintegrityd/0 0 1000 1000 khelper 0 2270 2270 events/0 Changes in V2: _ Drop the static array based on NR_CPU and dynamically allocate the stat array with num_possible_cpus() and other cpu mask facilities.... _ Trace workqueue insertion at a bit lower level (insert_work instead of queue_work) to handle even the workqueue barriers. Signed-off-by: Frederic Weisbecker Signed-off-by: Steven Rostedt Signed-off-by: Ingo Molnar --- kernel/workqueue.c | 16 +++++++++++++++- 1 file changed, 15 insertions(+), 1 deletion(-) (limited to 'kernel/workqueue.c') diff --git a/kernel/workqueue.c b/kernel/workqueue.c index 2f445833ae3..1fc2bc20603 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -33,6 +33,7 @@ #include #include #include +#include /* * The per-CPU workqueue (if single thread, we always use the first @@ -125,9 +126,13 @@ struct cpu_workqueue_struct *get_wq_data(struct work_struct *work) return (void *) (atomic_long_read(&work->data) & WORK_STRUCT_WQ_DATA_MASK); } +DEFINE_TRACE(workqueue_insertion); + static void insert_work(struct cpu_workqueue_struct *cwq, struct work_struct *work, struct list_head *head) { + trace_workqueue_insertion(cwq->thread, work); + set_wq_data(work, cwq); /* * Ensure that we get the right work->data if we see the @@ -259,6 +264,8 @@ int queue_delayed_work_on(int cpu, struct workqueue_struct *wq, } EXPORT_SYMBOL_GPL(queue_delayed_work_on); +DEFINE_TRACE(workqueue_execution); + static void run_workqueue(struct cpu_workqueue_struct *cwq) { spin_lock_irq(&cwq->lock); @@ -284,7 +291,7 @@ static void run_workqueue(struct cpu_workqueue_struct *cwq) */ struct lockdep_map lockdep_map = work->lockdep_map; #endif - + trace_workqueue_execution(cwq->thread, work); cwq->current_work = work; list_del_init(cwq->worklist.next); spin_unlock_irq(&cwq->lock); @@ -765,6 +772,8 @@ init_cpu_workqueue(struct workqueue_struct *wq, int cpu) return cwq; } +DEFINE_TRACE(workqueue_creation); + static int create_workqueue_thread(struct cpu_workqueue_struct *cwq, int cpu) { struct sched_param param = { .sched_priority = MAX_RT_PRIO-1 }; @@ -787,6 +796,8 @@ static int create_workqueue_thread(struct cpu_workqueue_struct *cwq, int cpu) sched_setscheduler_nocheck(p, SCHED_FIFO, ¶m); cwq->thread = p; + trace_workqueue_creation(cwq->thread, cpu); + return 0; } @@ -868,6 +879,8 @@ struct workqueue_struct *__create_workqueue_key(const char *name, } EXPORT_SYMBOL_GPL(__create_workqueue_key); +DEFINE_TRACE(workqueue_destruction); + static void cleanup_workqueue_thread(struct cpu_workqueue_struct *cwq) { /* @@ -891,6 +904,7 @@ static void cleanup_workqueue_thread(struct cpu_workqueue_struct *cwq) * checks list_empty(), and a "normal" queue_work() can't use * a dead CPU. */ + trace_workqueue_destruction(cwq->thread); kthread_stop(cwq->thread); cwq->thread = NULL; } -- cgit v1.2.3-70-g09d2 From aa85ea5b89c36c51200d795dd788139bd9b8cf50 Mon Sep 17 00:00:00 2001 From: Rusty Russell Date: Mon, 30 Mar 2009 22:05:15 -0600 Subject: cpumask: use new cpumask_ functions in core code. Impact: cleanup Time to clean up remaining laggards using the old cpu_ functions. Signed-off-by: Rusty Russell Cc: Greg Kroah-Hartman Cc: Ingo Molnar Cc: Trond.Myklebust@netapp.com --- drivers/base/cpu.c | 2 +- include/linux/cpuset.h | 4 ++-- kernel/workqueue.c | 6 +++--- mm/allocpercpu.c | 2 +- mm/vmstat.c | 2 +- net/sunrpc/svc.c | 2 +- 6 files changed, 9 insertions(+), 9 deletions(-) (limited to 'kernel/workqueue.c') diff --git a/drivers/base/cpu.c b/drivers/base/cpu.c index 5b257a57bc5..e62a4ccea54 100644 --- a/drivers/base/cpu.c +++ b/drivers/base/cpu.c @@ -119,7 +119,7 @@ static ssize_t print_cpus_map(char *buf, const struct cpumask *map) #define print_cpus_func(type) \ static ssize_t print_cpus_##type(struct sysdev_class *class, char *buf) \ { \ - return print_cpus_map(buf, &cpu_##type##_map); \ + return print_cpus_map(buf, cpu_##type##_mask); \ } \ static struct sysdev_class_attribute attr_##type##_map = \ _SYSDEV_CLASS_ATTR(type, 0444, print_cpus_##type, NULL) diff --git a/include/linux/cpuset.h b/include/linux/cpuset.h index 90c6074a36c..2e0d79678de 100644 --- a/include/linux/cpuset.h +++ b/include/linux/cpuset.h @@ -90,12 +90,12 @@ static inline void cpuset_init_smp(void) {} static inline void cpuset_cpus_allowed(struct task_struct *p, struct cpumask *mask) { - *mask = cpu_possible_map; + cpumask_copy(mask, cpu_possible_mask); } static inline void cpuset_cpus_allowed_locked(struct task_struct *p, struct cpumask *mask) { - *mask = cpu_possible_map; + cpumask_copy(mask, cpu_possible_mask); } static inline nodemask_t cpuset_mems_allowed(struct task_struct *p) diff --git a/kernel/workqueue.c b/kernel/workqueue.c index 1f0c509b40d..9aedd9fd825 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -416,7 +416,7 @@ void flush_workqueue(struct workqueue_struct *wq) might_sleep(); lock_map_acquire(&wq->lockdep_map); lock_map_release(&wq->lockdep_map); - for_each_cpu_mask_nr(cpu, *cpu_map) + for_each_cpu(cpu, cpu_map) flush_cpu_workqueue(per_cpu_ptr(wq->cpu_wq, cpu)); } EXPORT_SYMBOL_GPL(flush_workqueue); @@ -547,7 +547,7 @@ static void wait_on_work(struct work_struct *work) wq = cwq->wq; cpu_map = wq_cpu_map(wq); - for_each_cpu_mask_nr(cpu, *cpu_map) + for_each_cpu(cpu, cpu_map) wait_on_cpu_work(per_cpu_ptr(wq->cpu_wq, cpu), work); } @@ -911,7 +911,7 @@ void destroy_workqueue(struct workqueue_struct *wq) list_del(&wq->list); spin_unlock(&workqueue_lock); - for_each_cpu_mask_nr(cpu, *cpu_map) + for_each_cpu(cpu, cpu_map) cleanup_workqueue_thread(per_cpu_ptr(wq->cpu_wq, cpu)); cpu_maps_update_done(); diff --git a/mm/allocpercpu.c b/mm/allocpercpu.c index 1882923bc70..139d5b7b662 100644 --- a/mm/allocpercpu.c +++ b/mm/allocpercpu.c @@ -143,7 +143,7 @@ void free_percpu(void *__pdata) { if (unlikely(!__pdata)) return; - __percpu_depopulate_mask(__pdata, &cpu_possible_map); + __percpu_depopulate_mask(__pdata, cpu_possible_mask); kfree(__percpu_disguise(__pdata)); } EXPORT_SYMBOL_GPL(free_percpu); diff --git a/mm/vmstat.c b/mm/vmstat.c index 91149746bb8..8cd81ea1ddc 100644 --- a/mm/vmstat.c +++ b/mm/vmstat.c @@ -27,7 +27,7 @@ static void sum_vm_events(unsigned long *ret, const struct cpumask *cpumask) memset(ret, 0, NR_VM_EVENT_ITEMS * sizeof(unsigned long)); - for_each_cpu_mask_nr(cpu, *cpumask) { + for_each_cpu(cpu, cpumask) { struct vm_event_state *this = &per_cpu(vm_event_states, cpu); for (i = 0; i < NR_VM_EVENT_ITEMS; i++) diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c index c51fed4d1af..bb507e2bb94 100644 --- a/net/sunrpc/svc.c +++ b/net/sunrpc/svc.c @@ -312,7 +312,7 @@ svc_pool_map_set_cpumask(struct task_struct *task, unsigned int pidx) switch (m->mode) { case SVC_POOL_PERCPU: { - set_cpus_allowed_ptr(task, &cpumask_of_cpu(node)); + set_cpus_allowed_ptr(task, cpumask_of(node)); break; } case SVC_POOL_PERNODE: -- cgit v1.2.3-70-g09d2 From 2355b70fd59cb5be7de2052a9edeee7afb7ff099 Mon Sep 17 00:00:00 2001 From: Lai Jiangshan Date: Thu, 2 Apr 2009 16:58:24 -0700 Subject: workqueue: avoid recursion in run_workqueue() 1) lockdep will complain when run_workqueue() performs recursion. 2) The recursive implementation of run_workqueue() means that flush_workqueue() and its documentation are inconsistent. This may hide deadlocks and other bugs. 3) The recursion in run_workqueue() will poison cwq->current_work, but flush_work() and __cancel_work_timer(), etcetera need a reliable cwq->current_work. Signed-off-by: Lai Jiangshan Acked-by: Oleg Nesterov Cc: Peter Zijlstra Cc: Ingo Molnar Cc: Frederic Weisbecker Cc: Eric Dumazet Cc: Rusty Russell Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/workqueue.c | 41 +++++++++++------------------------------ 1 file changed, 11 insertions(+), 30 deletions(-) (limited to 'kernel/workqueue.c') diff --git a/kernel/workqueue.c b/kernel/workqueue.c index 9aedd9fd825..32f8e0d2bf5 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -48,8 +48,6 @@ struct cpu_workqueue_struct { struct workqueue_struct *wq; struct task_struct *thread; - - int run_depth; /* Detect run_workqueue() recursion depth */ } ____cacheline_aligned; /* @@ -262,13 +260,6 @@ EXPORT_SYMBOL_GPL(queue_delayed_work_on); static void run_workqueue(struct cpu_workqueue_struct *cwq) { spin_lock_irq(&cwq->lock); - cwq->run_depth++; - if (cwq->run_depth > 3) { - /* morton gets to eat his hat */ - printk("%s: recursion depth exceeded: %d\n", - __func__, cwq->run_depth); - dump_stack(); - } while (!list_empty(&cwq->worklist)) { struct work_struct *work = list_entry(cwq->worklist.next, struct work_struct, entry); @@ -311,7 +302,6 @@ static void run_workqueue(struct cpu_workqueue_struct *cwq) spin_lock_irq(&cwq->lock); cwq->current_work = NULL; } - cwq->run_depth--; spin_unlock_irq(&cwq->lock); } @@ -368,29 +358,20 @@ static void insert_wq_barrier(struct cpu_workqueue_struct *cwq, static int flush_cpu_workqueue(struct cpu_workqueue_struct *cwq) { - int active; + int active = 0; + struct wq_barrier barr; - if (cwq->thread == current) { - /* - * Probably keventd trying to flush its own queue. So simply run - * it by hand rather than deadlocking. - */ - run_workqueue(cwq); - active = 1; - } else { - struct wq_barrier barr; + WARN_ON(cwq->thread == current); - active = 0; - spin_lock_irq(&cwq->lock); - if (!list_empty(&cwq->worklist) || cwq->current_work != NULL) { - insert_wq_barrier(cwq, &barr, &cwq->worklist); - active = 1; - } - spin_unlock_irq(&cwq->lock); - - if (active) - wait_for_completion(&barr.done); + spin_lock_irq(&cwq->lock); + if (!list_empty(&cwq->worklist) || cwq->current_work != NULL) { + insert_wq_barrier(cwq, &barr, &cwq->worklist); + active = 1; } + spin_unlock_irq(&cwq->lock); + + if (active) + wait_for_completion(&barr.done); return active; } -- cgit v1.2.3-70-g09d2