writeback: optimize periodic bdi thread wakeups

Whe the first inode for a bdi is marked dirty, we wake up the bdi thread which
should take care of the periodic background write-out. However, the write-out
will actually start only 'dirty_writeback_interval' centisecs later, so we can
delay the wake-up.

This change was requested by Nick Piggin who pointed out that if we delay the
wake-up, we weed out 2 unnecessary contex switches, which matters because
'__mark_inode_dirty()' is a hot-path function.

This patch introduces a new function - 'bdi_wakeup_thread_delayed()', which
sets up a timer to wake-up the bdi thread and returns. So the wake-up is
delayed.

We also delete the timer in bdi threads just before writing-back. And
synchronously delete it when unregistering bdi. At the unregister point the bdi
does not have any users, so no one can arm it again.

Since now we take 'bdi->wb_lock' in the timer, which can execute in softirq
context, we have to use 'spin_lock_bh()' for 'bdi->wb_lock'. This patch makes
this change as well.

This patch also moves the 'bdi_wb_init()' function down in the file to avoid
forward-declaration of 'bdi_wakeup_thread_delayed()'.

Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>

1 files changed, 11 insertions, 25 deletions

diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c
index 55f6e46e06f1..bfa2df2c7ce2 100644
--- a/fs/fs-writeback.c
+++ b/fs/fs-writeback.c
@@ -76,7 +76,7 @@ static void bdi_queue_work(struct backing_dev_info *bdi,
 {
 	trace_writeback_queue(bdi, work);
 
-	spin_lock(&bdi->wb_lock);
+	spin_lock_bh(&bdi->wb_lock);
 	list_add_tail(&work->list, &bdi->work_list);
 	if (bdi->wb.task) {
 		wake_up_process(bdi->wb.task);
@@ -88,7 +88,7 @@ static void bdi_queue_work(struct backing_dev_info *bdi,
 		trace_writeback_nothread(bdi, work);
 		wake_up_process(default_backing_dev_info.wb.task);
 	}
-	spin_unlock(&bdi->wb_lock);
+	spin_unlock_bh(&bdi->wb_lock);
 }
 
 static void
@@ -704,13 +704,13 @@ get_next_work_item(struct backing_dev_info *bdi)
 {
 	struct wb_writeback_work *work = NULL;
 
-	spin_lock(&bdi->wb_lock);
+	spin_lock_bh(&bdi->wb_lock);
 	if (!list_empty(&bdi->work_list)) {
 		work = list_entry(bdi->work_list.next,
 				  struct wb_writeback_work, list);
 		list_del_init(&work->list);
 	}
-	spin_unlock(&bdi->wb_lock);
+	spin_unlock_bh(&bdi->wb_lock);
 	return work;
 }
 
@@ -810,6 +810,12 @@ int bdi_writeback_thread(void *data)
 	trace_writeback_thread_start(bdi);
 
 	while (!kthread_should_stop()) {
+		/*
+		 * Remove own delayed wake-up timer, since we are already awake
+		 * and we'll take care of the preriodic write-back.
+		 */
+		del_timer(&wb->wakeup_timer);
+
 		pages_written = wb_do_writeback(wb, 0);
 
 		trace_writeback_pages_written(pages_written);
@@ -868,26 +874,6 @@ void wakeup_flusher_threads(long nr_pages)
 	rcu_read_unlock();
 }
 
-/*
- * This function is used when the first inode for this bdi is marked dirty. It
- * wakes-up the corresponding bdi thread which should then take care of the
- * periodic background write-out of dirty inodes.
- */
-static void wakeup_bdi_thread(struct backing_dev_info *bdi)
-{
-	spin_lock(&bdi->wb_lock);
-	if (bdi->wb.task)
-		wake_up_process(bdi->wb.task);
-	else
-		/*
-		 * When bdi tasks are inactive for long time, they are killed.
-		 * In this case we have to wake-up the forker thread which
-		 * should create and run the bdi thread.
-		 */
-		wake_up_process(default_backing_dev_info.wb.task);
-	spin_unlock(&bdi->wb_lock);
-}
-
 static noinline void block_dump___mark_inode_dirty(struct inode *inode)
 {
 	if (inode->i_ino || strcmp(inode->i_sb->s_id, "bdev")) {
@@ -1019,7 +1005,7 @@ out:
 	spin_unlock(&inode_lock);
 
 	if (wakeup_bdi)
-		wakeup_bdi_thread(bdi);
+		bdi_wakeup_thread_delayed(bdi);
 }
 EXPORT_SYMBOL(__mark_inode_dirty);