From: Derek Basehore <[email protected]>
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 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.
[[email protected]: update changelog, add comment, use bdi_wakeup_thread_delayed()]
Signed-off-by: Derek Basehore <[email protected]>
Reviewed-by: Jan Kara <[email protected]>
Cc: Alexander Viro <[email protected]>
Reviewed-by: Tejun Heo <[email protected]>
Cc: Greg Kroah-Hartman <[email protected]>
Cc: "Darrick J. Wong" <[email protected]>
Cc: Derek Basehore <[email protected]>
Cc: Kees Cook <[email protected]>
Cc: Benson Leung <[email protected]>
Cc: Sonny Rao <[email protected]>
Cc: Luigi Semenzato <[email protected]>
Cc: Jens Axboe <[email protected]>
Cc: Dave Chinner <[email protected]>
Cc: <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
---
fs/fs-writeback.c | 8 ++++----
mm/backing-dev.c | 5 ++++-
2 files changed, 8 insertions(+), 5 deletions(-)
diff -puN fs/fs-writeback.c~backing_dev-fix-hung-task-on-sync fs/fs-writeback.c
--- a/fs/fs-writeback.c~backing_dev-fix-hung-task-on-sync
+++ a/fs/fs-writeback.c
@@ -1039,10 +1039,10 @@ void bdi_writeback_workfn(struct work_st
trace_writeback_pages_written(pages_written);
}
- if (!list_empty(&bdi->work_list) ||
- (wb_has_dirty_io(wb) && dirty_writeback_interval))
- queue_delayed_work(bdi_wq, &wb->dwork,
- msecs_to_jiffies(dirty_writeback_interval * 10));
+ if (!list_empty(&bdi->work_list))
+ mod_delayed_work(bdi_wq, &wb->dwork, 0);
+ else if (wb_has_dirty_io(wb) && dirty_writeback_interval)
+ bdi_wakeup_thread_delayed(bdi);
current->flags &= ~PF_SWAPWRITE;
}
diff -puN mm/backing-dev.c~backing_dev-fix-hung-task-on-sync mm/backing-dev.c
--- a/mm/backing-dev.c~backing_dev-fix-hung-task-on-sync
+++ a/mm/backing-dev.c
@@ -288,13 +288,16 @@ int bdi_has_dirty_io(struct backing_dev_
* 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);
}
/*
_
--
To unsubscribe from this list: send the line "unsubscribe stable" in
the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
