blk-mq: update hctx->nr_active in blk_mq_end_request_batch()

In case of shared tags and none io sched, batched completion still may
be run into, and hctx->nr_active is accounted when getting driver tag,
so it has to be updated in blk_mq_end_request_batch().

Otherwise, hctx->nr_active may become same with queue depth, then
hctx_may_queue() always return false, then io hang is caused.

Fixes the issue by updating the counter in batched way.
Reported-by: Shinichiro Kawasaki <shinichiro.kawasaki@wdc.com>
Fixes: f794f335 ("block: add support for blk_mq_end_request_batch()")
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Link: https://lore.kernel.org/r/20211102153619.3627505-4-ming.lei@redhat.comSigned-off-by: Jens Axboe <axboe@kernel.dk>

block/blk-mq.c

@@ -818,6 +818,13 @@ static inline void blk_mq_flush_tag_batch(struct blk_mq_hw_ctx *hctx,
 {
 	struct request_queue *q = hctx->queue;
 
+	/*
+	 * All requests should have been marked as RQF_MQ_INFLIGHT, so
+	 * update hctx->nr_active in batch
+	 */
+	if (hctx->flags & BLK_MQ_F_TAG_QUEUE_SHARED)
+		__blk_mq_sub_active_requests(hctx, nr_tags);
+
 	blk_mq_put_tags(hctx->tags, tag_array, nr_tags);
 	percpu_ref_put_many(&q->q_usage_counter, nr_tags);
 }

block/blk-mq.h

@@ -225,12 +225,18 @@ static inline void __blk_mq_inc_active_requests(struct blk_mq_hw_ctx *hctx)
 		atomic_inc(&hctx->nr_active);
 }
 
-static inline void __blk_mq_dec_active_requests(struct blk_mq_hw_ctx *hctx)
+static inline void __blk_mq_sub_active_requests(struct blk_mq_hw_ctx *hctx,
+		int val)
 {
 	if (blk_mq_is_shared_tags(hctx->flags))
-		atomic_dec(&hctx->queue->nr_active_requests_shared_tags);
+		atomic_sub(val, &hctx->queue->nr_active_requests_shared_tags);
 	else
-		atomic_dec(&hctx->nr_active);
+		atomic_sub(val, &hctx->nr_active);
+}
+
+static inline void __blk_mq_dec_active_requests(struct blk_mq_hw_ctx *hctx)
+{
+	__blk_mq_sub_active_requests(hctx, 1);
 }
 
 static inline int __blk_mq_active_requests(struct blk_mq_hw_ctx *hctx)