From 6ca738d60c563d5c6cf6253ee4b8e76fa77b2b9e Mon Sep 17 00:00:00 2001 From: Derek Basehore Date: Thu, 3 Apr 2014 14:46:22 -0700 Subject: backing_dev: fix hung task on sync bdi_wakeup_thread_delayed() used the mod_delayed_work() function to schedule work to writeback dirty inodes. The problem with this is that it can delay work that is scheduled for immediate execution, such as the work from sync_inodes_sb(). This can happen since mod_delayed_work() can now steal work from a work_queue. This fixes the problem by using queue_delayed_work() instead. This is a regression caused by commit 839a8e8660b6 ("writeback: replace custom worker pool implementation with unbound workqueue"). The reason that this causes a problem is that laptop-mode will change the delay, dirty_writeback_centisecs, to 60000 (10 minutes) by default. In the case that bdi_wakeup_thread_delayed() races with sync_inodes_sb(), sync will be stopped for 10 minutes and trigger a hung task. Even if dirty_writeback_centisecs is not long enough to cause a hung task, we still don't want to delay sync for that long. We fix the problem by using queue_delayed_work() when we want to schedule writeback sometime in future. This function doesn't change the timer if it is already armed. For the same reason, we also change bdi_writeback_workfn() to immediately queue the work again in the case that the work_list is not empty. The same problem can happen if the sync work is run on the rescue worker. [jack@suse.cz: update changelog, add comment, use bdi_wakeup_thread_delayed()] Signed-off-by: Derek Basehore Reviewed-by: Jan Kara Cc: Alexander Viro Reviewed-by: Tejun Heo Cc: Greg Kroah-Hartman Cc: "Darrick J. Wong" Cc: Derek Basehore Cc: Kees Cook Cc: Benson Leung Cc: Sonny Rao Cc: Luigi Semenzato Cc: Jens Axboe Cc: Dave Chinner Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/backing-dev.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'mm/backing-dev.c') diff --git a/mm/backing-dev.c b/mm/backing-dev.c index ce682f7a4f2..fab8401fc54 100644 --- a/mm/backing-dev.c +++ b/mm/backing-dev.c @@ -288,13 +288,16 @@ int bdi_has_dirty_io(struct backing_dev_info *bdi) * Note, we wouldn't bother setting up the timer, but this function is on the * fast-path (used by '__mark_inode_dirty()'), so we save few context switches * by delaying the wake-up. + * + * We have to be careful not to postpone flush work if it is scheduled for + * earlier. Thus we use queue_delayed_work(). */ void bdi_wakeup_thread_delayed(struct backing_dev_info *bdi) { unsigned long timeout; timeout = msecs_to_jiffies(dirty_writeback_interval * 10); - mod_delayed_work(bdi_wq, &bdi->wb.dwork, timeout); + queue_delayed_work(bdi_wq, &bdi->wb.dwork, timeout); } /* -- cgit v1.2.3-70-g09d2 From 5acda9d12dcf1ad0d9a5a2a7c646de3472fa7555 Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Thu, 3 Apr 2014 14:46:23 -0700 Subject: bdi: avoid oops on device removal After commit 839a8e8660b6 ("writeback: replace custom worker pool implementation with unbound workqueue") when device is removed while we are writing to it we crash in bdi_writeback_workfn() -> set_worker_desc() because bdi->dev is NULL. This can happen because even though bdi_unregister() cancels all pending flushing work, nothing really prevents new ones from being queued from balance_dirty_pages() or other places. Fix the problem by clearing BDI_registered bit in bdi_unregister() and checking it before scheduling of any flushing work. Fixes: 839a8e8660b6777e7fe4e80af1a048aebe2b5977 Reviewed-by: Tejun Heo Signed-off-by: Jan Kara Cc: Derek Basehore Cc: Jens Axboe Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/fs-writeback.c | 23 ++++++++++++++++++----- include/linux/backing-dev.h | 2 +- mm/backing-dev.c | 13 +++++++++---- 3 files changed, 28 insertions(+), 10 deletions(-) (limited to 'mm/backing-dev.c') diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index 6788fe0c976..a16315957ef 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -89,16 +89,29 @@ static inline struct inode *wb_inode(struct list_head *head) #define CREATE_TRACE_POINTS #include +static void bdi_wakeup_thread(struct backing_dev_info *bdi) +{ + spin_lock_bh(&bdi->wb_lock); + if (test_bit(BDI_registered, &bdi->state)) + mod_delayed_work(bdi_wq, &bdi->wb.dwork, 0); + spin_unlock_bh(&bdi->wb_lock); +} + static void bdi_queue_work(struct backing_dev_info *bdi, struct wb_writeback_work *work) { trace_writeback_queue(bdi, work); spin_lock_bh(&bdi->wb_lock); + if (!test_bit(BDI_registered, &bdi->state)) { + if (work->done) + complete(work->done); + goto out_unlock; + } list_add_tail(&work->list, &bdi->work_list); - spin_unlock_bh(&bdi->wb_lock); - mod_delayed_work(bdi_wq, &bdi->wb.dwork, 0); +out_unlock: + spin_unlock_bh(&bdi->wb_lock); } static void @@ -114,7 +127,7 @@ __bdi_start_writeback(struct backing_dev_info *bdi, long nr_pages, work = kzalloc(sizeof(*work), GFP_ATOMIC); if (!work) { trace_writeback_nowork(bdi); - mod_delayed_work(bdi_wq, &bdi->wb.dwork, 0); + bdi_wakeup_thread(bdi); return; } @@ -161,7 +174,7 @@ void bdi_start_background_writeback(struct backing_dev_info *bdi) * writeback as soon as there is no other work to do. */ trace_writeback_wake_background(bdi); - mod_delayed_work(bdi_wq, &bdi->wb.dwork, 0); + bdi_wakeup_thread(bdi); } /* @@ -1017,7 +1030,7 @@ void bdi_writeback_workfn(struct work_struct *work) current->flags |= PF_SWAPWRITE; if (likely(!current_is_workqueue_rescuer() || - list_empty(&bdi->bdi_list))) { + !test_bit(BDI_registered, &bdi->state))) { /* * The normal path. Keep writing back @bdi until its * work_list is empty. Note that this path is also taken diff --git a/include/linux/backing-dev.h b/include/linux/backing-dev.h index 24819001f5c..e488e9459a9 100644 --- a/include/linux/backing-dev.h +++ b/include/linux/backing-dev.h @@ -95,7 +95,7 @@ struct backing_dev_info { unsigned int max_ratio, max_prop_frac; struct bdi_writeback wb; /* default writeback info for this bdi */ - spinlock_t wb_lock; /* protects work_list */ + spinlock_t wb_lock; /* protects work_list & wb.dwork scheduling */ struct list_head work_list; diff --git a/mm/backing-dev.c b/mm/backing-dev.c index fab8401fc54..09d9591b770 100644 --- a/mm/backing-dev.c +++ b/mm/backing-dev.c @@ -297,7 +297,10 @@ void bdi_wakeup_thread_delayed(struct backing_dev_info *bdi) unsigned long timeout; timeout = msecs_to_jiffies(dirty_writeback_interval * 10); - queue_delayed_work(bdi_wq, &bdi->wb.dwork, timeout); + spin_lock_bh(&bdi->wb_lock); + if (test_bit(BDI_registered, &bdi->state)) + queue_delayed_work(bdi_wq, &bdi->wb.dwork, timeout); + spin_unlock_bh(&bdi->wb_lock); } /* @@ -310,9 +313,6 @@ static void bdi_remove_from_list(struct backing_dev_info *bdi) spin_unlock_bh(&bdi_lock); synchronize_rcu_expedited(); - - /* bdi_list is now unused, clear it to mark @bdi dying */ - INIT_LIST_HEAD(&bdi->bdi_list); } int bdi_register(struct backing_dev_info *bdi, struct device *parent, @@ -363,6 +363,11 @@ static void bdi_wb_shutdown(struct backing_dev_info *bdi) */ bdi_remove_from_list(bdi); + /* Make sure nobody queues further work */ + spin_lock_bh(&bdi->wb_lock); + clear_bit(BDI_registered, &bdi->state); + spin_unlock_bh(&bdi->wb_lock); + /* * Drain work list and shutdown the delayed_work. At this point, * @bdi->bdi_list is empty telling bdi_Writeback_workfn() that @bdi -- cgit v1.2.3-70-g09d2