blk-mq: don't grab ->q_usage_counter in blk_mq_sched_bio_merge

blk_mq_sched_bio_merge is only called from blk-mq.c:blk_attempt_bio_merge(),
which is called when queue usage counter is grabbed already:

1) blk_mq_get_new_requests()

2) blk_mq_get_request()
- cached request in current plug owns one queue usage counter

So don't grab ->q_usage_counter in blk_mq_sched_bio_merge(), and more
importantly this nest way causes hang in blk_mq_freeze_queue_wait().

Cc: Christoph Hellwig <hch@lst.de>
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211111085134.345235-2-ming.lei@redhat.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>

diff --git a/block/blk-mq-sched.c b/block/blk-mq-sched.c
index 4be652f..ba21449 100644 (file)
--- a/block/blk-mq-sched.c
+++ b/block/blk-mq-sched.c
@@ -370,9 +370,6 @@ bool blk_mq_sched_bio_merge(struct request_queue *q, struct bio *bio,
        bool ret = false;
        enum hctx_type type;
 
-       if (bio_queue_enter(bio))
-               return false;
-
        if (e && e->type->ops.bio_merge) {
                ret = e->type->ops.bio_merge(q, bio, nr_segs);
                goto out_put;
@@ -397,7 +394,6 @@ bool blk_mq_sched_bio_merge(struct request_queue *q, struct bio *bio,
 
        spin_unlock(&ctx->lock);
 out_put:
-       blk_queue_exit(q);
        return ret;
 }