On Tue, Aug 24, 2021 at 10:05:20AM -0700, Bart Van Assche wrote:
> The block layer may call the I/O scheduler .finish_request() callback
> without having called the .insert_requests() callback. Make sure that the
> mq-deadline I/O statistics are correct if the block layer inserts an I/O
> request that bypasses the I/O scheduler. This patch prevents that lower
> priority I/O is delayed longer than necessary for mixed I/O priority
> workloads.
>
> Cc: Niklas Cassel <Niklas.Cassel@xxxxxxx>
> Cc: Damien Le Moal <damien.lemoal@xxxxxxx>
> Cc: Hannes Reinecke <hare@xxxxxxx>
> Reported-by: Niklas Cassel <Niklas.Cassel@xxxxxxx>
> Fixes: 08a9ad8bf607 ("block/mq-deadline: Add cgroup support")
> Signed-off-by: Bart Van Assche <bvanassche@xxxxxxx>
> ---
> block/mq-deadline.c | 16 +++++++++++-----
> 1 file changed, 11 insertions(+), 5 deletions(-)
>
> diff --git a/block/mq-deadline.c b/block/mq-deadline.c
> index dbd74bae9f11..28f2e7655a5f 100644
> --- a/block/mq-deadline.c
> +++ b/block/mq-deadline.c
> @@ -713,6 +713,7 @@ static void dd_insert_request(struct blk_mq_hw_ctx *hctx, struct request *rq,
>
> prio = ioprio_class_to_prio[ioprio_class];
> dd_count(dd, inserted, prio);
> + rq->elv.priv[0] = (void *)(uintptr_t)1;
>
> if (blk_mq_sched_try_insert_merge(q, rq, &free)) {
> blk_mq_free_requests(&free);
> @@ -761,12 +762,10 @@ static void dd_insert_requests(struct blk_mq_hw_ctx *hctx,
> spin_unlock(&dd->lock);
> }
>
> -/*
> - * Nothing to do here. This is defined only to ensure that .finish_request
> - * method is called upon request completion.
> - */
> +/* Callback from inside blk_mq_rq_ctx_init(). */
> static void dd_prepare_request(struct request *rq)
> {
> + rq->elv.priv[0] = NULL;
> }
>
> /*
> @@ -793,7 +792,14 @@ static void dd_finish_request(struct request *rq)
> const enum dd_prio prio = ioprio_class_to_prio[ioprio_class];
> struct dd_per_prio *per_prio = &dd->per_prio[prio];
>
> - dd_count(dd, completed, prio);
> + /*
> + * The block layer core may call dd_finish_request() without having
> + * called dd_insert_requests(). Hence only update statistics for
> + * requests for which dd_insert_requests() has been called. See also
> + * blk_mq_request_bypass_insert().
> + */
> + if (rq->elv.priv[0])
> + dd_count(dd, completed, prio);
>
> if (blk_queue_is_zoned(q)) {
> unsigned long flags;
Reviewed-by: Niklas Cassel <niklas.cassel@xxxxxxx>
Tested-by: Niklas Cassel <niklas.cassel@xxxxxxx>
