From b89deed32ccc96098bd6bc953c64bba6b847774f Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Wed, 9 May 2007 02:33:52 -0700 Subject: implement flush_work() A basic problem with flush_scheduled_work() is that it blocks behind _all_ presently-queued works, rather than just the work whcih the caller wants to flush. If the caller holds some lock, and if one of the queued work happens to want that lock as well then accidental deadlocks can occur. One example of this is the phy layer: it wants to flush work while holding rtnl_lock(). But if a linkwatch event happens to be queued, the phy code will deadlock because the linkwatch callback function takes rtnl_lock. So we implement a new function which will flush a *single* work - just the one which the caller wants to free up. Thus we avoid the accidental deadlocks which can arise from unrelated subsystems' callbacks taking shared locks. flush_work() non-blockingly dequeues the work_struct which we want to kill, then it waits for its handler to complete on all CPUs. Add ->current_work to the "struct cpu_workqueue_struct", it points to currently running "struct work_struct". When flush_work(work) detects ->current_work == work, it inserts a barrier at the _head_ of ->worklist (and thus right _after_ that work) and waits for completition. This means that the next work fired on that CPU will be this barrier, or another barrier queued by concurrent flush_work(), so the caller of flush_work() will be woken before any "regular" work has a chance to run. When wait_on_work() unlocks workqueue_mutex (or whatever we choose to protect against CPU hotplug), CPU may go away. But in that case take_over_work() will move a barrier we queued to another CPU, it will be fired sometime, and wait_on_work() will be woken. Actually, we are doing cleanup_workqueue_thread()->kthread_stop() before take_over_work(), so cwq->thread should complete its ->worklist (and thus the barrier), because currently we don't check kthread_should_stop() in run_workqueue(). But even if we did, everything should be ok. [akpm@osdl.org: cleanup] [akpm@osdl.org: add flush_work_keventd() wrapper] Signed-off-by: Oleg Nesterov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/workqueue.c | 95 ++++++++++++++++++++++++++++++++++++++++++++++++++++-- 1 file changed, 92 insertions(+), 3 deletions(-) (limited to 'kernel') diff --git a/kernel/workqueue.c b/kernel/workqueue.c index b7bb37ab03b..918d55267a1 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -46,6 +46,7 @@ struct cpu_workqueue_struct { struct workqueue_struct *wq; struct task_struct *thread; + struct work_struct *current_work; int run_depth; /* Detect run_workqueue() recursion depth */ @@ -120,6 +121,7 @@ static int __run_work(struct cpu_workqueue_struct *cwq, struct work_struct *work && work_pending(work) && !list_empty(&work->entry)) { work_func_t f = work->func; + cwq->current_work = work; list_del_init(&work->entry); spin_unlock_irqrestore(&cwq->lock, flags); @@ -128,6 +130,7 @@ static int __run_work(struct cpu_workqueue_struct *cwq, struct work_struct *work f(work); spin_lock_irqsave(&cwq->lock, flags); + cwq->current_work = NULL; ret = 1; } spin_unlock_irqrestore(&cwq->lock, flags); @@ -166,6 +169,17 @@ int fastcall run_scheduled_work(struct work_struct *work) } EXPORT_SYMBOL(run_scheduled_work); +static void insert_work(struct cpu_workqueue_struct *cwq, + struct work_struct *work, int tail) +{ + set_wq_data(work, cwq); + if (tail) + list_add_tail(&work->entry, &cwq->worklist); + else + list_add(&work->entry, &cwq->worklist); + wake_up(&cwq->more_work); +} + /* Preempt must be disabled. */ static void __queue_work(struct cpu_workqueue_struct *cwq, struct work_struct *work) @@ -173,9 +187,7 @@ static void __queue_work(struct cpu_workqueue_struct *cwq, unsigned long flags; spin_lock_irqsave(&cwq->lock, flags); - set_wq_data(work, cwq); - list_add_tail(&work->entry, &cwq->worklist); - wake_up(&cwq->more_work); + insert_work(cwq, work, 1); spin_unlock_irqrestore(&cwq->lock, flags); } @@ -305,6 +317,7 @@ static void run_workqueue(struct cpu_workqueue_struct *cwq) struct work_struct, entry); work_func_t f = work->func; + cwq->current_work = work; list_del_init(cwq->worklist.next); spin_unlock_irqrestore(&cwq->lock, flags); @@ -325,6 +338,7 @@ static void run_workqueue(struct cpu_workqueue_struct *cwq) } spin_lock_irqsave(&cwq->lock, flags); + cwq->current_work = NULL; } cwq->run_depth--; spin_unlock_irqrestore(&cwq->lock, flags); @@ -449,6 +463,75 @@ void fastcall flush_workqueue(struct workqueue_struct *wq) } EXPORT_SYMBOL_GPL(flush_workqueue); +static void wait_on_work(struct cpu_workqueue_struct *cwq, + struct work_struct *work) +{ + struct wq_barrier barr; + int running = 0; + + spin_lock_irq(&cwq->lock); + if (unlikely(cwq->current_work == work)) { + init_wq_barrier(&barr); + insert_work(cwq, &barr.work, 0); + running = 1; + } + spin_unlock_irq(&cwq->lock); + + if (unlikely(running)) { + mutex_unlock(&workqueue_mutex); + wait_for_completion(&barr.done); + mutex_lock(&workqueue_mutex); + } +} + +/** + * flush_work - block until a work_struct's callback has terminated + * @wq: the workqueue on which the work is queued + * @work: the work which is to be flushed + * + * flush_work() will attempt to cancel the work if it is queued. If the work's + * callback appears to be running, flush_work() will block until it has + * completed. + * + * flush_work() is designed to be used when the caller is tearing down data + * structures which the callback function operates upon. It is expected that, + * prior to calling flush_work(), the caller has arranged for the work to not + * be requeued. + */ +void flush_work(struct workqueue_struct *wq, struct work_struct *work) +{ + struct cpu_workqueue_struct *cwq; + + mutex_lock(&workqueue_mutex); + cwq = get_wq_data(work); + /* Was it ever queued ? */ + if (!cwq) + goto out; + + /* + * This work can't be re-queued, and the lock above protects us + * from take_over_work(), no need to re-check that get_wq_data() + * is still the same when we take cwq->lock. + */ + spin_lock_irq(&cwq->lock); + list_del_init(&work->entry); + work_release(work); + spin_unlock_irq(&cwq->lock); + + if (is_single_threaded(wq)) { + /* Always use first cpu's area. */ + wait_on_work(per_cpu_ptr(wq->cpu_wq, singlethread_cpu), work); + } else { + int cpu; + + for_each_online_cpu(cpu) + wait_on_work(per_cpu_ptr(wq->cpu_wq, cpu), work); + } +out: + mutex_unlock(&workqueue_mutex); +} +EXPORT_SYMBOL_GPL(flush_work); + static struct task_struct *create_workqueue_thread(struct workqueue_struct *wq, int cpu, int freezeable) { @@ -650,6 +733,12 @@ void flush_scheduled_work(void) } EXPORT_SYMBOL(flush_scheduled_work); +void flush_work_keventd(struct work_struct *work) +{ + flush_work(keventd_wq, work); +} +EXPORT_SYMBOL(flush_work_keventd); + /** * cancel_rearming_delayed_workqueue - reliably kill off a delayed work whose handler rearms the delayed work. * @wq: the controlling workqueue structure -- cgit v1.2.3-18-g5258