From de1482974080ec9ef414bf048b2646b246b63f6e Mon Sep 17 00:00:00 2001 From: Ming Lei Date: Sat, 14 Oct 2017 17:22:29 +0800 Subject: blk-mq: introduce .get_budget and .put_budget in blk_mq_ops For SCSI devices, there is often a per-request-queue depth, which needs to be respected before queuing one request. Currently blk-mq always dequeues the request first, then calls .queue_rq() to dispatch the request to lld. One obvious issue with this approach is that I/O merging may not be successful, because when the per-request-queue depth can't be respected, .queue_rq() has to return BLK_STS_RESOURCE, and then this request has to stay in hctx->dispatch list. This means it never gets a chance to be merged with other IO. This patch introduces .get_budget and .put_budget callback in blk_mq_ops, then we can try to get reserved budget first before dequeuing request. If the budget for queueing I/O can't be satisfied, we don't need to dequeue request at all. Hence the request can be left in the IO scheduler queue, for more merging opportunities. Signed-off-by: Ming Lei Signed-off-by: Jens Axboe --- block/blk-mq.h | 20 +++++++++++++++++++- 1 file changed, 19 insertions(+), 1 deletion(-) (limited to 'block/blk-mq.h') diff --git a/block/blk-mq.h b/block/blk-mq.h index ef15b3414da5..e413b732374e 100644 --- a/block/blk-mq.h +++ b/block/blk-mq.h @@ -30,7 +30,7 @@ void blk_mq_freeze_queue(struct request_queue *q); void blk_mq_free_queue(struct request_queue *q); int blk_mq_update_nr_requests(struct request_queue *q, unsigned int nr); void blk_mq_wake_waiters(struct request_queue *q); -bool blk_mq_dispatch_rq_list(struct request_queue *, struct list_head *); +bool blk_mq_dispatch_rq_list(struct request_queue *, struct list_head *, bool); void blk_mq_flush_busy_ctxs(struct blk_mq_hw_ctx *hctx, struct list_head *list); bool blk_mq_hctx_has_pending(struct blk_mq_hw_ctx *hctx); bool blk_mq_get_driver_tag(struct request *rq, struct blk_mq_hw_ctx **hctx, @@ -137,4 +137,22 @@ static inline bool blk_mq_hw_queue_mapped(struct blk_mq_hw_ctx *hctx) void blk_mq_in_flight(struct request_queue *q, struct hd_struct *part, unsigned int inflight[2]); +static inline void blk_mq_put_dispatch_budget(struct blk_mq_hw_ctx *hctx) +{ + struct request_queue *q = hctx->queue; + + if (q->mq_ops->put_budget) + q->mq_ops->put_budget(hctx); +} + +static inline blk_status_t blk_mq_get_dispatch_budget( + struct blk_mq_hw_ctx *hctx) +{ + struct request_queue *q = hctx->queue; + + if (q->mq_ops->get_budget) + return q->mq_ops->get_budget(hctx); + return BLK_STS_OK; +} + #endif -- cgit v1.2.1 From b347689ffbca745ac457ee27400ce1affd571c6f Mon Sep 17 00:00:00 2001 From: Ming Lei Date: Sat, 14 Oct 2017 17:22:30 +0800 Subject: blk-mq-sched: improve dispatching from sw queue SCSI devices use host-wide tagset, and the shared driver tag space is often quite big. However, there is also a queue depth for each lun( .cmd_per_lun), which is often small, for example, on both lpfc and qla2xxx, .cmd_per_lun is just 3. So lots of requests may stay in sw queue, and we always flush all belonging to same hw queue and dispatch them all to driver. Unfortunately it is easy to cause queue busy because of the small .cmd_per_lun. Once these requests are flushed out, they have to stay in hctx->dispatch, and no bio merge can happen on these requests, and sequential IO performance is harmed. This patch introduces blk_mq_dequeue_from_ctx for dequeuing a request from a sw queue, so that we can dispatch them in scheduler's way. We can then avoid dequeueing too many requests from sw queue, since we don't flush ->dispatch completely. This patch improves dispatching from sw queue by using the .get_budget and .put_budget callbacks. Reviewed-by: Omar Sandoval Signed-off-by: Ming Lei Signed-off-by: Jens Axboe --- block/blk-mq.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'block/blk-mq.h') diff --git a/block/blk-mq.h b/block/blk-mq.h index e413b732374e..522b420dedc0 100644 --- a/block/blk-mq.h +++ b/block/blk-mq.h @@ -35,6 +35,8 @@ void blk_mq_flush_busy_ctxs(struct blk_mq_hw_ctx *hctx, struct list_head *list); bool blk_mq_hctx_has_pending(struct blk_mq_hw_ctx *hctx); bool blk_mq_get_driver_tag(struct request *rq, struct blk_mq_hw_ctx **hctx, bool wait); +struct request *blk_mq_dequeue_from_ctx(struct blk_mq_hw_ctx *hctx, + struct blk_mq_ctx *start); /* * Internal helpers for allocating/freeing the request map -- cgit v1.2.1 From 88022d7201e96b43f1754b0358fc6bcd8dbdcde1 Mon Sep 17 00:00:00 2001 From: Ming Lei Date: Sun, 5 Nov 2017 02:21:12 +0800 Subject: blk-mq: don't handle failure in .get_budget It is enough to just check if we can get the budget via .get_budget(). And we don't need to deal with device state change in .get_budget(). For SCSI, one issue to be fixed is that we have to call scsi_mq_uninit_cmd() to free allocated ressources if SCSI device fails to handle the request. And it isn't enough to simply call blk_mq_end_request() to do that if this request is marked as RQF_DONTPREP. Fixes: 0df21c86bdbf(scsi: implement .get_budget and .put_budget for blk-mq) Signed-off-by: Ming Lei Signed-off-by: Jens Axboe --- block/blk-mq.h | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) (limited to 'block/blk-mq.h') diff --git a/block/blk-mq.h b/block/blk-mq.h index 522b420dedc0..f97aceff76e9 100644 --- a/block/blk-mq.h +++ b/block/blk-mq.h @@ -147,14 +147,13 @@ static inline void blk_mq_put_dispatch_budget(struct blk_mq_hw_ctx *hctx) q->mq_ops->put_budget(hctx); } -static inline blk_status_t blk_mq_get_dispatch_budget( - struct blk_mq_hw_ctx *hctx) +static inline bool blk_mq_get_dispatch_budget(struct blk_mq_hw_ctx *hctx) { struct request_queue *q = hctx->queue; if (q->mq_ops->get_budget) return q->mq_ops->get_budget(hctx); - return BLK_STS_OK; + return true; } #endif -- cgit v1.2.1 From b0850297c749ea79a5717d597931366b3d7f4b09 Mon Sep 17 00:00:00 2001 From: Ming Lei Date: Thu, 2 Nov 2017 23:24:34 +0800 Subject: block: pass 'run_queue' to blk_mq_request_bypass_insert Block flush need this function without running the queue, so add a parameter controlling whether we run it or not. Signed-off-by: Ming Lei Signed-off-by: Jens Axboe --- block/blk-mq.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'block/blk-mq.h') diff --git a/block/blk-mq.h b/block/blk-mq.h index f97aceff76e9..9fffec0ad8e9 100644 --- a/block/blk-mq.h +++ b/block/blk-mq.h @@ -56,7 +56,7 @@ int blk_mq_alloc_rqs(struct blk_mq_tag_set *set, struct blk_mq_tags *tags, */ void __blk_mq_insert_request(struct blk_mq_hw_ctx *hctx, struct request *rq, bool at_head); -void blk_mq_request_bypass_insert(struct request *rq); +void blk_mq_request_bypass_insert(struct request *rq, bool run_queue); void blk_mq_insert_requests(struct blk_mq_hw_ctx *hctx, struct blk_mq_ctx *ctx, struct list_head *list); -- cgit v1.2.1 From 244c65a3ccaa06fd15cc940315606674d3108b2f Mon Sep 17 00:00:00 2001 From: Ming Lei Date: Sat, 4 Nov 2017 12:39:57 -0600 Subject: blk-mq: move blk_mq_put_driver_tag*() into blk-mq.h We need this helper to put the driver tag for flush rq, since we will not share tag in the flush request sequence in the following patch in case that I/O scheduler is applied. Signed-off-by: Ming Lei Signed-off-by: Jens Axboe --- block/blk-mq.h | 33 +++++++++++++++++++++++++++++++++ 1 file changed, 33 insertions(+) (limited to 'block/blk-mq.h') diff --git a/block/blk-mq.h b/block/blk-mq.h index 9fffec0ad8e9..2502f40ccdc0 100644 --- a/block/blk-mq.h +++ b/block/blk-mq.h @@ -2,6 +2,7 @@ #define INT_BLK_MQ_H #include "blk-stat.h" +#include "blk-mq-tag.h" struct blk_mq_tag_set; @@ -156,4 +157,36 @@ static inline bool blk_mq_get_dispatch_budget(struct blk_mq_hw_ctx *hctx) return true; } +static inline void __blk_mq_put_driver_tag(struct blk_mq_hw_ctx *hctx, + struct request *rq) +{ + blk_mq_put_tag(hctx, hctx->tags, rq->mq_ctx, rq->tag); + rq->tag = -1; + + if (rq->rq_flags & RQF_MQ_INFLIGHT) { + rq->rq_flags &= ~RQF_MQ_INFLIGHT; + atomic_dec(&hctx->nr_active); + } +} + +static inline void blk_mq_put_driver_tag_hctx(struct blk_mq_hw_ctx *hctx, + struct request *rq) +{ + if (rq->tag == -1 || rq->internal_tag == -1) + return; + + __blk_mq_put_driver_tag(hctx, rq); +} + +static inline void blk_mq_put_driver_tag(struct request *rq) +{ + struct blk_mq_hw_ctx *hctx; + + if (rq->tag == -1 || rq->internal_tag == -1) + return; + + hctx = blk_mq_map_queue(rq->q, rq->mq_ctx->cpu); + __blk_mq_put_driver_tag(hctx, rq); +} + #endif -- cgit v1.2.1 From 9a95e4ef709533efac4aafcb8bddf73f96db50ed Mon Sep 17 00:00:00 2001 From: Bart Van Assche Date: Thu, 9 Nov 2017 10:49:59 -0800 Subject: block, nvme: Introduce blk_mq_req_flags_t Several block layer and NVMe core functions accept a combination of BLK_MQ_REQ_* flags through the 'flags' argument but there is no verification at compile time whether the right type of block layer flags is passed. Make it possible for sparse to verify this. This patch does not change any functionality. Signed-off-by: Bart Van Assche Reviewed-by: Hannes Reinecke Tested-by: Oleksandr Natalenko Cc: linux-nvme@lists.infradead.org Cc: Christoph Hellwig Cc: Johannes Thumshirn Cc: Ming Lei Signed-off-by: Jens Axboe --- block/blk-mq.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'block/blk-mq.h') diff --git a/block/blk-mq.h b/block/blk-mq.h index 2502f40ccdc0..99a19c5523e2 100644 --- a/block/blk-mq.h +++ b/block/blk-mq.h @@ -111,7 +111,7 @@ static inline void blk_mq_put_ctx(struct blk_mq_ctx *ctx) struct blk_mq_alloc_data { /* input parameter */ struct request_queue *q; - unsigned int flags; + blk_mq_req_flags_t flags; unsigned int shallow_depth; /* input & output parameter */ -- cgit v1.2.1 From 79f720a751cad613620d0237e3b44f89f4a69181 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Fri, 10 Nov 2017 09:13:21 -0700 Subject: blk-mq: only run the hardware queue if IO is pending Currently we are inconsistent in when we decide to run the queue. Using blk_mq_run_hw_queues() we check if the hctx has pending IO before running it, but we don't do that from the individual queue run function, blk_mq_run_hw_queue(). This results in a lot of extra and pointless queue runs, potentially, on flush requests and (much worse) on tag starvation situations. This is observable just looking at top output, with lots of kworkers active. For the !async runs, it just adds to the CPU overhead of blk-mq. Move the has-pending check into the run function instead of having callers do it. Reviewed-by: Christoph Hellwig Signed-off-by: Jens Axboe --- block/blk-mq.h | 2 -- 1 file changed, 2 deletions(-) (limited to 'block/blk-mq.h') diff --git a/block/blk-mq.h b/block/blk-mq.h index 99a19c5523e2..dcf379a892dd 100644 --- a/block/blk-mq.h +++ b/block/blk-mq.h @@ -26,14 +26,12 @@ struct blk_mq_ctx { struct kobject kobj; } ____cacheline_aligned_in_smp; -void blk_mq_run_hw_queue(struct blk_mq_hw_ctx *hctx, bool async); void blk_mq_freeze_queue(struct request_queue *q); void blk_mq_free_queue(struct request_queue *q); int blk_mq_update_nr_requests(struct request_queue *q, unsigned int nr); void blk_mq_wake_waiters(struct request_queue *q); bool blk_mq_dispatch_rq_list(struct request_queue *, struct list_head *, bool); void blk_mq_flush_busy_ctxs(struct blk_mq_hw_ctx *hctx, struct list_head *list); -bool blk_mq_hctx_has_pending(struct blk_mq_hw_ctx *hctx); bool blk_mq_get_driver_tag(struct request *rq, struct blk_mq_hw_ctx **hctx, bool wait); struct request *blk_mq_dequeue_from_ctx(struct blk_mq_hw_ctx *hctx, -- cgit v1.2.1