block: splice plug list to local context (109b8129) · Commits · 戴 / test

If the request_fn ends up blocking, we could be re-entering the plug flush. Since the list is protected by explicitly not allowing schedule events, this isn't a terribly good idea. Additionally, it can cause us to recurse. As request_fn called by __blk_run_queue is allowed to 'schedule()' (after dropping the queue lock of course), it is possible to get a recursive call: schedule -> blk_flush_plug -> __blk_finish_plug -> flush_plug_list -> __blk_run_queue -> request_fn -> schedule We must make sure that the second schedule does not call into blk_flush_plug again. So instead of leaving the list of requests on blk_plug->list, move them to a separate list leaving blk_plug->list empty. Signed-off-by:

Jens Axboe <jaxboe@fusionio.com>

block/blk-core.c

+9 −5

Original line number	Diff line number	Diff line
		@@ -2673,19 +2673,24 @@ static void flush_plug_list(struct blk_plug *plug)
		struct request_queue *q;
		unsigned long flags;
		struct request *rq;
		LIST_HEAD(list);

		BUG_ON(plug->magic != PLUG_MAGIC);

		if (list_empty(&plug->list))
		return;

		if (plug->should_sort)
		list_sort(NULL, &plug->list, plug_rq_cmp);
		list_splice_init(&plug->list, &list);

		if (plug->should_sort) {
		list_sort(NULL, &list, plug_rq_cmp);
		plug->should_sort = 0;
		}

		q = NULL;
		local_irq_save(flags);
		while (!list_empty(&plug->list)) {
		rq = list_entry_rq(plug->list.next);
		while (!list_empty(&list)) {
		rq = list_entry_rq(list.next);
		list_del_init(&rq->queuelist);
		BUG_ON(!(rq->cmd_flags & REQ_ON_PLUG));
		BUG_ON(!rq->q);
		@@ -2713,7 +2718,6 @@ static void flush_plug_list(struct blk_plug *plug)
		spin_unlock(q->queue_lock);
		}

		BUG_ON(!list_empty(&plug->list));
		local_irq_restore(flags);
		}

Admin message