From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1750845AbdALWBB (ORCPT ); Thu, 12 Jan 2017 17:01:01 -0500 Received: from mx0a-00082601.pphosted.com ([67.231.145.42]:60126 "EHLO mx0a-00082601.pphosted.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750797AbdALWA7 (ORCPT ); Thu, 12 Jan 2017 17:00:59 -0500 Date: Thu, 12 Jan 2017 14:59:22 -0700 From: Jens Axboe To: Bart Van Assche CC: "linux-kernel@vger.kernel.org" , "linux-block@vger.kernel.org" , "osandov@osandov.com" Subject: Re: [PATCH 08/10] blk-mq-sched: add framework for MQ capable IO schedulers Message-ID: <20170112215922.GB25197@kernel.dk> References: <1484170803-9311-1-git-send-email-axboe@fb.com> <1484170803-9311-9-git-send-email-axboe@fb.com> <1484257502.2720.21.camel@sandisk.com> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Disposition: inline In-Reply-To: <1484257502.2720.21.camel@sandisk.com> X-Originating-IP: [192.168.54.13] X-Proofpoint-Spam-Reason: safe X-FB-Internal: Safe X-Proofpoint-Virus-Version: vendor=fsecure engine=2.50.10432:,, definitions=2017-01-12_15:,, signatures=0 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Thu, Jan 12 2017, Bart Van Assche wrote: > On Wed, 2017-01-11 at 14:40 -0700, Jens Axboe wrote: > > @@ -451,11 +456,11 @@ void blk_insert_flush(struct request *rq) > > * processed directly without going through flush machinery. Queue > > * for normal execution. > > */ > > - if ((policy & REQ_FSEQ_DATA) && > > - !(policy & (REQ_FSEQ_PREFLUSH | REQ_FSEQ_POSTFLUSH))) { > > - if (q->mq_ops) { > > - blk_mq_insert_request(rq, false, true, false); > > - } else > > + if (((policy & REQ_FSEQ_DATA) && > > + !(policy & (REQ_FSEQ_PREFLUSH | REQ_FSEQ_POSTFLUSH)))) { > > + if (q->mq_ops) > > + blk_mq_sched_insert_request(rq, false, true, false); > > + else > > list_add_tail(&rq->queuelist, &q->queue_head); > > return; > > } > > Not that it really matters, but this change adds a pair of parentheses -- > "if (e)" is changed into "if ((e))". Is this necessary? I fixed that up earlier today, as I noticed the same. So that's gone in the current -git tree. > > +void blk_mq_sched_free_hctx_data(struct request_queue *q, > > + void (*exit)(struct blk_mq_hw_ctx *)) > > +{ > > + struct blk_mq_hw_ctx *hctx; > > + int i; > > + > > + queue_for_each_hw_ctx(q, hctx, i) { > > + if (exit) > > + exit(hctx); > > + kfree(hctx->sched_data); > > + hctx->sched_data = NULL; > > + } > > +} > > +EXPORT_SYMBOL_GPL(blk_mq_sched_free_hctx_data); > > + > > +int blk_mq_sched_init_hctx_data(struct request_queue *q, size_t size, > > + int (*init)(struct blk_mq_hw_ctx *), > > + void (*exit)(struct blk_mq_hw_ctx *)) > > +{ > > + struct blk_mq_hw_ctx *hctx; > > + int ret; > > + int i; > > + > > + queue_for_each_hw_ctx(q, hctx, i) { > > + hctx->sched_data = kmalloc_node(size, GFP_KERNEL, hctx->numa_node); > > + if (!hctx->sched_data) { > > + ret = -ENOMEM; > > + goto error; > > + } > > + > > + if (init) { > > + ret = init(hctx); > > + if (ret) { > > + /* > > + * We don't want to give exit() a partially > > + * initialized sched_data. init() must clean up > > + * if it fails. > > + */ > > + kfree(hctx->sched_data); > > + hctx->sched_data = NULL; > > + goto error; > > + } > > + } > > + } > > + > > + return 0; > > +error: > > + blk_mq_sched_free_hctx_data(q, exit); > > + return ret; > > +} > > If one of the init() calls by blk_mq_sched_init_hctx_data() fails then > blk_mq_sched_free_hctx_data() will call exit() even for hctx's for which > init() has not been called. How about changing "if (exit)" into "if (exit && > hctx->sched_data)" such that exit() is only called for hctx's for which > init() has been called? Good point, I'll make that change to the exit function. > > +struct request *blk_mq_sched_get_request(struct request_queue *q, > > + struct bio *bio, > > + unsigned int op, > > + struct blk_mq_alloc_data *data) > > +{ > > + struct elevator_queue *e = q->elevator; > > + struct blk_mq_hw_ctx *hctx; > > + struct blk_mq_ctx *ctx; > > + struct request *rq; > > + > > + blk_queue_enter_live(q); > > + ctx = blk_mq_get_ctx(q); > > + hctx = blk_mq_map_queue(q, ctx->cpu); > > + > > + blk_mq_set_alloc_data(data, q, 0, ctx, hctx); > > + > > + if (e) { > > + data->flags |= BLK_MQ_REQ_INTERNAL; > > + if (e->type->ops.mq.get_request) > > + rq = e->type->ops.mq.get_request(q, op, data); > > + else > > + rq = __blk_mq_alloc_request(data, op); > > + } else { > > + rq = __blk_mq_alloc_request(data, op); > > + if (rq) { > > + rq->tag = rq->internal_tag; > > + rq->internal_tag = -1; > > + } > > + } > > + > > + if (rq) { > > + rq->elv.icq = NULL; > > + if (e && e->type->icq_cache) > > + blk_mq_sched_assign_ioc(q, rq, bio); > > + data->hctx->queued++; > > + return rq; > > + } > > + > > + blk_queue_exit(q); > > + return NULL; > > +} > > The "rq->tag = rq->internal_tag; rq->internal_tag = -1;" occurs not only > here but also in blk_mq_alloc_request_hctx(). Has it been considered to move > that code into __blk_mq_alloc_request()? Yes, it's in two locations. I wanted to keep it out of __blk_mq_alloc_request(), so we can still use that for normal tag allocations. But maybe it's better for __blk_mq_alloc_request() to just do: if (flags & BLK_MQ_REQ_INTERNAL) { rq->tag = -1; rq->internal_tag = tag; } else { rq->tag = tag; rq->internal_tag = -1; } and handle it directly in there. What do you think? > @@ -223,14 +225,17 @@ struct request *__blk_mq_alloc_request(struct blk_mq_alloc_data *data, > > > > tag = blk_mq_get_tag(data); > > if (tag != BLK_MQ_TAG_FAIL) { > > - rq = data->hctx->tags->rqs[tag]; > > + struct blk_mq_tags *tags = blk_mq_tags_from_data(data); > > + > > + rq = tags->rqs[tag]; > > > > if (blk_mq_tag_busy(data->hctx)) { > > rq->rq_flags = RQF_MQ_INFLIGHT; > > atomic_inc(&data->hctx->nr_active); > > } > > > > - rq->tag = tag; > > + rq->tag = -1; > > + rq->internal_tag = tag; > > blk_mq_rq_ctx_init(data->q, data->ctx, rq, op); > > return rq; > > } > > How about using the following code for tag assignment instead of "rq->tag = > -1; rq->internal_tag = tag"? > > if (data->flags & BLK_MQ_REQ_INTERNAL) { > rq->tag = -1; > rq->internal_tag = tag; > } else { > rq->tag = tag; > rq->internal_tag = -1; > } Hah, nevermind, I should have read further. I guess we agree, I'll make that change. > > @@ -313,6 +313,9 @@ struct request *blk_mq_alloc_request_hctx(struct request_queue *q, int rw, > > goto out_queue_exit; > > } > > > > + rq->tag = rq->internal_tag; > > + rq->internal_tag = -1; > > + > > return rq; > > > > out_queue_exit: > > @@ -321,10 +324,10 @@ struct request *blk_mq_alloc_request_hctx(struct request_queue *q, int rw, > > } > > EXPORT_SYMBOL_GPL(blk_mq_alloc_request_hctx); > > Should something like "WARN_ON_ONCE(flags & BLK_MQ_REQ_INTERNAL)" be added > at the start of this function to avoid that BLK_MQ_REQ_INTERNAL is passed in > from outside the block layer? Yes, seems like a prudent safety check. I'll add it, thanks. -- Jens Axboe