nvme: use queuedata for nvme_req_qid

The request's rq_disk isn't set for passthrough IO commands, so tracing
uses qid 0 for these which incorrectly decodes as an admin command. Use
the request_queue's queuedata instead since that value is always set for
the IO queues, and never set for the admin queue.

Signed-off-by: Keith Busch <kbusch@kernel.org>
Signed-off-by: Christoph Hellwig <hch@lst.de>
This commit is contained in:
Keith Busch 2020-10-15 11:36:29 -07:00 коммит произвёл Christoph Hellwig
Родитель a87da50f39
Коммит 643c476d6f
1 изменённых файлов: 1 добавлений и 1 удалений

Просмотреть файл

@ -176,7 +176,7 @@ static inline struct nvme_request *nvme_req(struct request *req)
static inline u16 nvme_req_qid(struct request *req)
{
if (!req->rq_disk)
if (!req->q->queuedata)
return 0;
return blk_mq_unique_tag_to_hwq(blk_mq_unique_tag(req)) + 1;
}