[PATCH v3] blk-mq: check on cpu id when there is only one ctx mapping

[PATCH v3] blk-mq: check on cpu id when there is only one ctx mapping

[Ed Tsai]

commit f168420c62e7 ("blk-mq: don't redirect completion for hctx withs
only one ctx mapping") When nvme applies a 1:1 mapping of hctx and ctx,
there will be no remote request.

But for ufs, the submission and completion queues could be asymmetric.
(e.g. Multiple SQs share one CQ) Therefore, 1:1 mapping of hctx and
ctx won't complete request on the submission cpu. In this situation,
this nr_ctx check could violate the QUEUE_FLAG_SAME_FORCE, as a result,
check on cpu id when there is only one ctx mapping.

Signed-off-by: Ed Tsai <ed.tsai@mediatek.com>
Signed-off-by: Po-Wen Kao <powen.kao@mediatek.com>
Suggested-by: Keith Busch <kbusch@kernel.org>

---
 block/blk-mq.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/block/blk-mq.c b/block/blk-mq.c
index 41f7e2b500bd..ecc30ebe9483 100644
--- a/block/blk-mq.c
+++ b/block/blk-mq.c
@@ -1176,7 +1176,8 @@ bool blk_mq_complete_request_remote(struct request *rq)
 	 * or a polled request, always complete locally,
 	 * it's pointless to redirect the completion.
 	 */
-	if (rq->mq_hctx->nr_ctx == 1 ||
+	if ((rq->mq_hctx->nr_ctx == 1 &&
+	     rq->mq_ctx->cpu == raw_smp_processor_id()) ||
 		rq->cmd_flags & REQ_POLLED)
 		return false;
 
-- 
2.18.0

Re: [PATCH v3] blk-mq: check on cpu id when there is only one ctx mapping

[Christoph Hellwig]

This looks fine to me:

Reviewed-by: Christoph Hellwig <hch@lst.de>
> -	if (rq->mq_hctx->nr_ctx == 1 ||
> +	if ((rq->mq_hctx->nr_ctx == 1 &&
> +	     rq->mq_ctx->cpu == raw_smp_processor_id()) ||
>  		rq->cmd_flags & REQ_POLLED)

But please fix the incorrct indentation in the existing code here
please while you're at it (maybe Jens can just fold it in):

	if ((rq->mq_hctx->nr_ctx == 1 &&
	     rq->mq_ctx->cpu == raw_smp_processor_id()) ||
 	    (rq->cmd_flags & REQ_POLLED))

Re: [PATCH v3] blk-mq: check on cpu id when there is only one ctx mapping

[Jens Axboe]

On Wed, 14 Jun 2023 08:25:29 +0800, Ed Tsai wrote:
> commit f168420c62e7 ("blk-mq: don't redirect completion for hctx withs
> only one ctx mapping") When nvme applies a 1:1 mapping of hctx and ctx,
> there will be no remote request.
> 
> But for ufs, the submission and completion queues could be asymmetric.
> (e.g. Multiple SQs share one CQ) Therefore, 1:1 mapping of hctx and
> ctx won't complete request on the submission cpu. In this situation,
> this nr_ctx check could violate the QUEUE_FLAG_SAME_FORCE, as a result,
> check on cpu id when there is only one ctx mapping.
> 
> [...]

Applied, thanks!

[1/1] blk-mq: check on cpu id when there is only one ctx mapping
      commit: 30654614f3d27230200b1650f6025a2ce67900b4

Best regards,
-- 
Jens Axboe