From: Omar Sandoval <osandov@fb.com>
Date: Tue, 27 Feb 2018 16:56:42 -0800
Subject: [PATCH] block: clear ctx pending bit under ctx lock
Git-commit: e9a99a638800af25c7ed006c96fd1dabb99254b7
Patch-mainline: v4.17-rc1
References: bsc#1104967,FATE#325924

When we insert a request, we set the software queue pending bit while
holding the software queue lock. However, we clear it outside of the
lock, so it's possible that a concurrent insert could reset the bit
after we clear it but before we empty the request list. Afterwards, the
bit would still be set but the software queue wouldn't have any requests
in it, leading us to do a spurious run in the future. This is mostly a
benign/theoretical issue, but it makes the following change easier to
justify.

Signed-off-by: Omar Sandoval <osandov@fb.com>
Acked-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Acked-by: Hannes Reinecke <hare@suse.com>
---
 block/blk-mq.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/block/blk-mq.c b/block/blk-mq.c
index 16e83e6df404..9594a0e9f65b 100644
--- a/block/blk-mq.c
+++ b/block/blk-mq.c
@@ -986,9 +986,9 @@ static bool flush_busy_ctx(struct sbitmap *sb, unsigned int bitnr, void *data)
 	struct blk_mq_hw_ctx *hctx = flush_data->hctx;
 	struct blk_mq_ctx *ctx = hctx->ctxs[bitnr];
 
-	sbitmap_clear_bit(sb, bitnr);
 	spin_lock(&ctx->lock);
 	list_splice_tail_init(&ctx->rq_list, flush_data->list);
+	sbitmap_clear_bit(sb, bitnr);
 	spin_unlock(&ctx->lock);
 	return true;
 }
-- 
2.16.4