Commit fe76fc6a authored Jan 15, 2019 by Ming Lei Committed by Linus Torvalds Jan 15, 2019

sbitmap: Protect swap_lock from hardirq



Because we may call blk_mq_get_driver_tag() directly from
blk_mq_dispatch_rq_list() without holding any lock, then HARDIRQ may
come and the above DEADLOCK is triggered.

Commit ab53dcfb3e7b ("sbitmap: Protect swap_lock from hardirq") tries to
fix this issue by using 'spin_lock_bh', which isn't enough because we
complete request from hardirq context direclty in case of multiqueue.

Cc: Clark Williams <williams@redhat.com>
Fixes: ab53dcfb3e7b ("sbitmap: Protect swap_lock from hardirq")
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Ming Lei <ming.lei@redhat.com>
Cc: Guenter Roeck <linux@roeck-us.net>
Cc: Steven Rostedt (VMware) <rostedt@goodmis.org>
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

parent 37198768

lib/sbitmap.c

+3 −2

Original line number	Original line	Diff line number	Diff line
	@@ -27,8 +27,9 @@ static inline bool sbitmap_deferred_clear(struct sbitmap *sb, int index)
	{		{
	unsigned long mask, val;		unsigned long mask, val;
	bool ret = false;		bool ret = false;
			unsigned long flags;

	spin_lock_bh(&sb->map[index].swap_lock);		spin_lock_irqsave(&sb->map[index].swap_lock, flags);

	if (!sb->map[index].cleared)		if (!sb->map[index].cleared)
	goto out_unlock;		goto out_unlock;
	@@ -49,7 +50,7 @@ static inline bool sbitmap_deferred_clear(struct sbitmap *sb, int index)

	ret = true;		ret = true;
	out_unlock:		out_unlock:
	spin_unlock_bh(&sb->map[index].swap_lock);		spin_unlock_irqrestore(&sb->map[index].swap_lock, flags);
	return ret;		return ret;
	}		}

Admin message