Re: [PATCH 2/2] blk-mq: Improve performance of non-mq IO schedulers with multiple HW queues

[Ming Lei <ming.lei@redhat.com>]

On Fri, Dec 18, 2020 at 10:44:12PM +0100, Jan Kara wrote:
> Currently when non-mq aware IO scheduler (BFQ, mq-deadline) is used for
> a queue with multiple HW queues, the performance it rather bad. The
> problem is that these IO schedulers use queue-wide locking and their
> dispatch function does not respect the hctx it is passed in and returns
> any request it finds appropriate. Thus locality of request access is
> broken and dispatch from multiple CPUs just contends on IO scheduler
> locks. For these IO schedulers there's little point in dispatching from
> multiple CPUs. Instead dispatch always only from a single CPU to limit
> contention.
> 
> Below is a comparison of dbench runs on XFS filesystem where the storage
> is a raid card with 64 HW queues and to it attached a single rotating
> disk. BFQ is used as IO scheduler:
> 
>       clients           MQ                     SQ             MQ-Patched
> Amean 1      39.12 (0.00%)       43.29 * -10.67%*       36.09 *   7.74%*
> Amean 2     128.58 (0.00%)      101.30 *  21.22%*       96.14 *  25.23%*
> Amean 4     577.42 (0.00%)      494.47 *  14.37%*      508.49 *  11.94%*
> Amean 8     610.95 (0.00%)      363.86 *  40.44%*      362.12 *  40.73%*
> Amean 16    391.78 (0.00%)      261.49 *  33.25%*      282.94 *  27.78%*
> Amean 32    324.64 (0.00%)      267.71 *  17.54%*      233.00 *  28.23%*
> Amean 64    295.04 (0.00%)      253.02 *  14.24%*      242.37 *  17.85%*
> Amean 512 10281.61 (0.00%)    10211.16 *   0.69%*    10447.53 *  -1.61%*
> 
> Numbers are times so lower is better. MQ is stock 5.10-rc6 kernel. SQ is
> the same kernel with megaraid_sas.host_tagset_enable=0 so that the card
> advertises just a single HW queue. MQ-Patched is a kernel with this
> patch applied.
> 
> You can see multiple hardware queues heavily hurt performance in
> combination with BFQ. The patch restores the performance.
> 
> Signed-off-by: Jan Kara <jack@suse.cz>
> ---
>  block/blk-mq.c           | 37 +++++++++++++++++++++++++++++++++++++
>  block/kyber-iosched.c    |  1 +
>  include/linux/elevator.h |  2 ++
>  3 files changed, 40 insertions(+)
> 
> diff --git a/block/blk-mq.c b/block/blk-mq.c
> index 57d0461f2be5..6d80054c231b 100644
> --- a/block/blk-mq.c
> +++ b/block/blk-mq.c
> @@ -1663,6 +1663,31 @@ void blk_mq_run_hw_queue(struct blk_mq_hw_ctx *hctx, bool async)
>  }
>  EXPORT_SYMBOL(blk_mq_run_hw_queue);
>  
> +static struct blk_mq_hw_ctx *blk_mq_sq_iosched_hctx(struct request_queue *q)
> +{
> +	struct elevator_queue *e = q->elevator;
> +	struct blk_mq_hw_ctx *hctx;
> +
> +	/*
> +	 * The queue has multiple hardware queues but uses IO scheduler that
> +	 * does not respect hardware queues when dispatching? This is not a
> +	 * great setup but it can be sensible when we have a single rotational
> +	 * disk behind a raid card. Just don't bother with multiple HW queues
> +	 * and dispatch from hctx for the current CPU since running multiple
> +	 * queues just causes lock contention inside the scheduler and
> +	 * pointless cache bouncing because the hctx is not respected by the IO
> +	 * scheduler's dispatch function anyway.
> +	 */
> +	if (q->nr_hw_queues > 1 && e && e->type->ops.dispatch_request &&
> +	    !(e->type->elevator_features & ELEVATOR_F_MQ_AWARE)) {
> +		hctx = blk_mq_map_queue_type(q, HCTX_TYPE_DEFAULT,
> +					     raw_smp_processor_id());
> +		if (!blk_mq_hctx_stopped(hctx))
> +			return hctx;
> +	}
> +	return NULL;
> +}
> +
>  /**
>   * blk_mq_run_hw_queues - Run all hardware queues in a request queue.
>   * @q: Pointer to the request queue to run.
> @@ -1673,6 +1698,12 @@ void blk_mq_run_hw_queues(struct request_queue *q, bool async)
>  	struct blk_mq_hw_ctx *hctx;
>  	int i;
>  
> +	hctx = blk_mq_sq_iosched_hctx(q);
> +	if (hctx) {
> +		blk_mq_run_hw_queue(hctx, async);
> +		return;
> +	}
> +

This approach looks reasonable, just wondering which code path is wrt.
blk_mq_run_hw_queues() improvement by this patch.

Since ed5dd6a67d5e ("scsi: core: Only re-run queue in scsi_end_request() if device
queue is busy") is merged, blk_mq_run_hw_queues() is only called from scsi_end_request()
when the scsi device is busy for megaraid.

Another one is bfq_schedule_dispatch(), in which blk_mq_run_hw_queues()
is still be called, if that is the reason, maybe it is easier to optimize
bfq_schedule_dispatch() by avoiding to call blk_mq_run_hw_queues().


thanks,
Ming