Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752469AbaKCIkx (ORCPT ); Mon, 3 Nov 2014 03:40:53 -0500 Received: from mail-ob0-f173.google.com ([209.85.214.173]:36499 "EHLO mail-ob0-f173.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751797AbaKCIks (ORCPT ); Mon, 3 Nov 2014 03:40:48 -0500 MIME-Version: 1.0 In-Reply-To: <1415003008-16683-3-git-send-email-hch@lst.de> References: <1415003008-16683-1-git-send-email-hch@lst.de> <1415003008-16683-3-git-send-email-hch@lst.de> Date: Mon, 3 Nov 2014 16:40:47 +0800 Message-ID: Subject: Re: [PATCH 2/2] blk-mq: allow direct dispatch to a driver specific workqueue From: Ming Lei To: Christoph Hellwig Cc: Jens Axboe , Richard Weinberger , ceph-devel@vger.kernel.org, Linux Kernel Mailing List Content-Type: text/plain; charset=UTF-8 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Hi Christoph, On Mon, Nov 3, 2014 at 4:23 PM, Christoph Hellwig wrote: > We have various block drivers that need to execute long term blocking > operations during I/O submission like file system or network I/O. > > Currently these drivers just queue up work to an internal workqueue > from their request_fn. With blk-mq we can make sure they always get > called on their own workqueue directly for I/O submission by: > > 1) adding a flag to prevent inline submission of I/O, and > 2) allowing the driver to pass in a workqueue in the tag_set that > will be used instead of kblockd. The above two aren't enough because the big problem is that drivers need a per-request work structure instead of 'hctx->run_work', otherwise there are at most NR_CPUS concurrent submissions. So the per-request work structure should be exposed to blk-mq too for the kind of usage, such as .blk_mq_req_work(req) callback in case of BLK_MQ_F_WORKQUEUE. Thanks, > > Signed-off-by: Christoph Hellwig > --- > block/blk-core.c | 2 +- > block/blk-mq.c | 12 +++++++++--- > block/blk.h | 1 + > include/linux/blk-mq.h | 4 ++++ > 4 files changed, 15 insertions(+), 4 deletions(-) > > diff --git a/block/blk-core.c b/block/blk-core.c > index 0421b53..7f7249f 100644 > --- a/block/blk-core.c > +++ b/block/blk-core.c > @@ -61,7 +61,7 @@ struct kmem_cache *blk_requestq_cachep; > /* > * Controlling structure to kblockd > */ > -static struct workqueue_struct *kblockd_workqueue; > +struct workqueue_struct *kblockd_workqueue; > > void blk_queue_congestion_threshold(struct request_queue *q) > { > diff --git a/block/blk-mq.c b/block/blk-mq.c > index 22e50a5..3d27d22 100644 > --- a/block/blk-mq.c > +++ b/block/blk-mq.c > @@ -804,12 +804,13 @@ void blk_mq_run_hw_queue(struct blk_mq_hw_ctx *hctx, bool async) > if (unlikely(test_bit(BLK_MQ_S_STOPPED, &hctx->state))) > return; > > - if (!async && cpumask_test_cpu(smp_processor_id(), hctx->cpumask)) { > + if (!async && !(hctx->flags & BLK_MQ_F_WORKQUEUE) && > + cpumask_test_cpu(smp_processor_id(), hctx->cpumask)) { > __blk_mq_run_hw_queue(hctx); > return; > } > > - kblockd_schedule_delayed_work_on(blk_mq_hctx_next_cpu(hctx), > + queue_delayed_work_on(blk_mq_hctx_next_cpu(hctx), hctx->wq, > &hctx->run_work, 0); > } > > @@ -908,7 +909,7 @@ static void blk_mq_delay_work_fn(struct work_struct *work) > > void blk_mq_delay_queue(struct blk_mq_hw_ctx *hctx, unsigned long msecs) > { > - kblockd_schedule_delayed_work_on(blk_mq_hctx_next_cpu(hctx), > + queue_delayed_work_on(blk_mq_hctx_next_cpu(hctx), hctx->wq, > &hctx->delay_work, msecs_to_jiffies(msecs)); > } > EXPORT_SYMBOL(blk_mq_delay_queue); > @@ -1581,6 +1582,11 @@ static int blk_mq_init_hctx(struct request_queue *q, > hctx->flags = set->flags; > hctx->cmd_size = set->cmd_size; > > + if (set->wq) > + hctx->wq = set->wq; > + else > + hctx->wq = kblockd_workqueue; > + > blk_mq_init_cpu_notifier(&hctx->cpu_notifier, > blk_mq_hctx_notify, hctx); > blk_mq_register_cpu_notifier(&hctx->cpu_notifier); > diff --git a/block/blk.h b/block/blk.h > index 43b0361..fb46ad0 100644 > --- a/block/blk.h > +++ b/block/blk.h > @@ -25,6 +25,7 @@ struct blk_flush_queue { > spinlock_t mq_flush_lock; > }; > > +extern struct workqueue_struct *kblockd_workqueue; > extern struct kmem_cache *blk_requestq_cachep; > extern struct kmem_cache *request_cachep; > extern struct kobj_type blk_queue_ktype; > diff --git a/include/linux/blk-mq.h b/include/linux/blk-mq.h > index 5a901d0..ebe4699 100644 > --- a/include/linux/blk-mq.h > +++ b/include/linux/blk-mq.h > @@ -37,6 +37,8 @@ struct blk_mq_hw_ctx { > unsigned int queue_num; > struct blk_flush_queue *fq; > > + struct workqueue_struct *wq; > + > void *driver_data; > > struct blk_mq_ctxmap ctx_map; > @@ -64,6 +66,7 @@ struct blk_mq_hw_ctx { > > struct blk_mq_tag_set { > struct blk_mq_ops *ops; > + struct workqueue_struct *wq; > unsigned int nr_hw_queues; > unsigned int queue_depth; /* max hw supported */ > unsigned int reserved_tags; > @@ -156,6 +159,7 @@ enum { > BLK_MQ_F_SG_MERGE = 1 << 2, > BLK_MQ_F_SYSFS_UP = 1 << 3, > BLK_MQ_F_DEFER_ISSUE = 1 << 4, > + BLK_MQ_F_WORKQUEUE = 1 << 5, > > BLK_MQ_S_STOPPED = 0, > BLK_MQ_S_TAG_ACTIVE = 1, > -- > 1.9.1 > -- Ming Lei -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/