[PATCH 1/5] blk mq: take ref to q when running it
STALE2709d
From: <hidden>
Date: 2016-05-25 07:54:59
Also in:
linux-scsi
Subsystem:
block layer, the rest · Maintainers:
Jens Axboe, Linus Torvalds
From: Mike Christie <redacted> If the __blk_mq_run_hw_queue is a result of a async run or retry then we do not have a reference to the queue. At this time, blk_cleanup_queue could begin tearing it down while the queue_rq function is being run. This patch just has us do a blk_queue_enter call. Signed-off-by: Mike Christie <redacted> --- block/blk-mq.c | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-)
diff --git a/block/blk-mq.c b/block/blk-mq.c
index 7df9c92..5958a02 100644
--- a/block/blk-mq.c
+++ b/block/blk-mq.c@@ -738,8 +738,13 @@ static void __blk_mq_run_hw_queue(struct blk_mq_hw_ctx *hctx) WARN_ON(!cpumask_test_cpu(raw_smp_processor_id(), hctx->cpumask)); - if (unlikely(test_bit(BLK_MQ_S_STOPPED, &hctx->state))) + if (blk_queue_enter(q, false)) { + blk_mq_run_hw_queue(hctx, true); return; + } + + if (unlikely(test_bit(BLK_MQ_S_STOPPED, &hctx->state))) + goto exit_queue; hctx->run++;
@@ -832,6 +837,9 @@ static void __blk_mq_run_hw_queue(struct blk_mq_hw_ctx *hctx) **/ blk_mq_run_hw_queue(hctx, true); } + +exit_queue: + blk_queue_exit(q); } /*
--
2.7.2