[PATCH] blk-mq: don't dispatch request in blk_mq_request_direct_issue if queue is busy

[PATCH] blk-mq: don't dispatch request in blk_mq_request_direct_issue if queue is busy

[Ming Lei]

If we run into blk_mq_request_direct_issue(), when queue is busy, we
don't want to dispatch this request into hctx->dispatch_list, and
what we need to do is to return the queue busy info to caller, so
that caller can deal with it well.

Fixes: 396eaf21ee ("blk-mq: improve DM's blk-mq IO merging via blk_insert_cloned_request feedback")
Reported-by: Laurence Oberman <loberman@redhat.com>
Reviewed-by: Mike Snitzer <snitzer@redhat.com>
Signed-off-by: Ming Lei <ming.lei@redhat.com>
---
 block/blk-mq.c | 22 ++++++++++------------
 1 file changed, 10 insertions(+), 12 deletions(-)

diff --git a/block/blk-mq.c b/block/blk-mq.c
index 4d4af8d712da..1af7fa70993b 100644
--- a/block/blk-mq.c
+++ b/block/blk-mq.c
@@ -1856,15 +1856,6 @@ static blk_status_t __blk_mq_issue_directly(struct blk_mq_hw_ctx *hctx,
 	return ret;
 }
 
-static void __blk_mq_fallback_to_insert(struct request *rq,
-					bool run_queue, bool bypass_insert)
-{
-	if (!bypass_insert)
-		blk_mq_sched_insert_request(rq, false, run_queue, false);
-	else
-		blk_mq_request_bypass_insert(rq, run_queue);
-}
-
 static blk_status_t __blk_mq_try_issue_directly(struct blk_mq_hw_ctx *hctx,
 						struct request *rq,
 						blk_qc_t *cookie,
@@ -1873,9 +1864,16 @@ static blk_status_t __blk_mq_try_issue_directly(struct blk_mq_hw_ctx *hctx,
 	struct request_queue *q = rq->q;
 	bool run_queue = true;
 
-	/* RCU or SRCU read lock is needed before checking quiesced flag */
+	/*
+	 * RCU or SRCU read lock is needed before checking quiesced flag.
+	 *
+	 * When queue is stopped or quiesced, ignore 'bypass_insert' from
+	 * blk_mq_request_direct_issue(), and return BLK_STS_OK to caller,
+	 * and avoid driver to try to dispatch again.
+	 */
 	if (blk_mq_hctx_stopped(hctx) || blk_queue_quiesced(q)) {
 		run_queue = false;
+		bypass_insert = false;
 		goto insert;
 	}
 
@@ -1892,10 +1890,10 @@ static blk_status_t __blk_mq_try_issue_directly(struct blk_mq_hw_ctx *hctx,
 
 	return __blk_mq_issue_directly(hctx, rq, cookie);
 insert:
-	__blk_mq_fallback_to_insert(rq, run_queue, bypass_insert);
 	if (bypass_insert)
 		return BLK_STS_RESOURCE;
 
+	blk_mq_sched_insert_request(rq, false, run_queue, false);
 	return BLK_STS_OK;
 }
 
@@ -1911,7 +1909,7 @@ static void blk_mq_try_issue_directly(struct blk_mq_hw_ctx *hctx,
 
 	ret = __blk_mq_try_issue_directly(hctx, rq, cookie, false);
 	if (ret == BLK_STS_RESOURCE)
-		__blk_mq_fallback_to_insert(rq, true, false);
+		blk_mq_sched_insert_request(rq, false, true, false);
 	else if (ret != BLK_STS_OK)
 		blk_mq_end_request(rq, ret);
 
-- 
2.9.5

Re: blk-mq: don't dispatch request in blk_mq_request_direct_issue if queue is busy

[Mike Snitzer]

On Wed, Jan 17 2018 at 11:06pm -0500,
Ming Lei <ming.lei@redhat.com> wrote:

> If we run into blk_mq_request_direct_issue(), when queue is busy, we
> don't want to dispatch this request into hctx->dispatch_list, and
> what we need to do is to return the queue busy info to caller, so
> that caller can deal with it well.
> 
> Fixes: 396eaf21ee ("blk-mq: improve DM's blk-mq IO merging via blk_insert_cloned_request feedback")
> Reported-by: Laurence Oberman <loberman@redhat.com>
> Reviewed-by: Mike Snitzer <snitzer@redhat.com>
> Signed-off-by: Ming Lei <ming.lei@redhat.com>
> ---
>  block/blk-mq.c | 22 ++++++++++------------
>  1 file changed, 10 insertions(+), 12 deletions(-)
> 
> diff --git a/block/blk-mq.c b/block/blk-mq.c
> index 4d4af8d712da..1af7fa70993b 100644
> --- a/block/blk-mq.c
> +++ b/block/blk-mq.c
> @@ -1856,15 +1856,6 @@ static blk_status_t __blk_mq_issue_directly(struct blk_mq_hw_ctx *hctx,
>  	return ret;
>  }
>  
> -static void __blk_mq_fallback_to_insert(struct request *rq,
> -					bool run_queue, bool bypass_insert)
> -{
> -	if (!bypass_insert)
> -		blk_mq_sched_insert_request(rq, false, run_queue, false);
> -	else
> -		blk_mq_request_bypass_insert(rq, run_queue);
> -}
> -
>  static blk_status_t __blk_mq_try_issue_directly(struct blk_mq_hw_ctx *hctx,
>  						struct request *rq,
>  						blk_qc_t *cookie,
> @@ -1873,9 +1864,16 @@ static blk_status_t __blk_mq_try_issue_directly(struct blk_mq_hw_ctx *hctx,
>  	struct request_queue *q = rq->q;
>  	bool run_queue = true;
>  
> -	/* RCU or SRCU read lock is needed before checking quiesced flag */
> +	/*
> +	 * RCU or SRCU read lock is needed before checking quiesced flag.
> +	 *
> +	 * When queue is stopped or quiesced, ignore 'bypass_insert' from
> +	 * blk_mq_request_direct_issue(), and return BLK_STS_OK to caller,
> +	 * and avoid driver to try to dispatch again.
> +	 */
>  	if (blk_mq_hctx_stopped(hctx) || blk_queue_quiesced(q)) {
>  		run_queue = false;
> +		bypass_insert = false;
>  		goto insert;
>  	}
>  
> @@ -1892,10 +1890,10 @@ static blk_status_t __blk_mq_try_issue_directly(struct blk_mq_hw_ctx *hctx,
>  
>  	return __blk_mq_issue_directly(hctx, rq, cookie);
>  insert:
> -	__blk_mq_fallback_to_insert(rq, run_queue, bypass_insert);
>  	if (bypass_insert)
>  		return BLK_STS_RESOURCE;
>  
> +	blk_mq_sched_insert_request(rq, false, run_queue, false);
>  	return BLK_STS_OK;
>  }

OK so you're just leveraging blk_mq_sched_insert_request()'s
ability to resort to__blk_mq_insert_request() if !q->elevator.

Re: [PATCH] blk-mq: don't dispatch request in blk_mq_request_direct_issue if queue is busy

[Jens Axboe]

On 1/17/18 9:06 PM, Ming Lei wrote:
> If we run into blk_mq_request_direct_issue(), when queue is busy, we
> don't want to dispatch this request into hctx->dispatch_list, and
> what we need to do is to return the queue busy info to caller, so
> that caller can deal with it well.

Thanks for fixing this up, applied.

-- 
Jens Axboe

Re: blk-mq: don't dispatch request in blk_mq_request_direct_issue if queue is busy

[Laurence Oberman]

On Wed, 2018-01-17 at 23:36 -0500, Mike Snitzer wrote:
> On Wed, Jan 17 2018 at 11:06pm -0500,
> Ming Lei <ming.lei@redhat.com> wrote:
> 
> > If we run into blk_mq_request_direct_issue(), when queue is busy,
> > we
> > don't want to dispatch this request into hctx->dispatch_list, and
> > what we need to do is to return the queue busy info to caller, so
> > that caller can deal with it well.
> > 
> > Fixes: 396eaf21ee ("blk-mq: improve DM's blk-mq IO merging via
> > blk_insert_cloned_request feedback")
> > Reported-by: Laurence Oberman <loberman@redhat.com>
> > Reviewed-by: Mike Snitzer <snitzer@redhat.com>
> > Signed-off-by: Ming Lei <ming.lei@redhat.com>
> > ---
> >  block/blk-mq.c | 22 ++++++++++------------
> >  1 file changed, 10 insertions(+), 12 deletions(-)
> > 
> > diff --git a/block/blk-mq.c b/block/blk-mq.c
> > index 4d4af8d712da..1af7fa70993b 100644
> > --- a/block/blk-mq.c
> > +++ b/block/blk-mq.c
> > @@ -1856,15 +1856,6 @@ static blk_status_t
> > __blk_mq_issue_directly(struct blk_mq_hw_ctx *hctx,
> >  	return ret;
> >  }
> >  
> > -static void __blk_mq_fallback_to_insert(struct request *rq,
> > -					bool run_queue, bool
> > bypass_insert)
> > -{
> > -	if (!bypass_insert)
> > -		blk_mq_sched_insert_request(rq, false, run_queue,
> > false);
> > -	else
> > -		blk_mq_request_bypass_insert(rq, run_queue);
> > -}
> > -
> >  static blk_status_t __blk_mq_try_issue_directly(struct
> > blk_mq_hw_ctx *hctx,
> >  						struct request
> > *rq,
> >  						blk_qc_t *cookie,
> > @@ -1873,9 +1864,16 @@ static blk_status_t
> > __blk_mq_try_issue_directly(struct blk_mq_hw_ctx *hctx,
> >  	struct request_queue *q = rq->q;
> >  	bool run_queue = true;
> >  
> > -	/* RCU or SRCU read lock is needed before checking
> > quiesced flag */
> > +	/*
> > +	 * RCU or SRCU read lock is needed before checking
> > quiesced flag.
> > +	 *
> > +	 * When queue is stopped or quiesced, ignore
> > 'bypass_insert' from
> > +	 * blk_mq_request_direct_issue(), and return BLK_STS_OK to
> > caller,
> > +	 * and avoid driver to try to dispatch again.
> > +	 */
> >  	if (blk_mq_hctx_stopped(hctx) || blk_queue_quiesced(q)) {
> >  		run_queue = false;
> > +		bypass_insert = false;
> >  		goto insert;
> >  	}
> >  
> > @@ -1892,10 +1890,10 @@ static blk_status_t
> > __blk_mq_try_issue_directly(struct blk_mq_hw_ctx *hctx,
> >  
> >  	return __blk_mq_issue_directly(hctx, rq, cookie);
> >  insert:
> > -	__blk_mq_fallback_to_insert(rq, run_queue, bypass_insert);
> >  	if (bypass_insert)
> >  		return BLK_STS_RESOURCE;
> >  
> > +	blk_mq_sched_insert_request(rq, false, run_queue, false);
> >  	return BLK_STS_OK;
> >  }
> 
> OK so you're just leveraging blk_mq_sched_insert_request()'s
> ability to resort to__blk_mq_insert_request() if !q->elevator.

I tested this against Mike's latest combined tree and its stable.
This fixes the list corruption issue.
Many Thanks Ming and Mike.

I will apply it to Bart's latest SRP/SRPT tree tomorrow as its very
late here but it will clearly fix the issue in Bart's tree too.

Tested-by: Laurence Oberman <loberman@redhat.com>