From patchwork Fri Nov 6 17:20:21 2015 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Jens Axboe X-Patchwork-Id: 7570231 X-Patchwork-Delegate: axboe@kernel.dk Return-Path: X-Original-To: patchwork-linux-block@patchwork.kernel.org Delivered-To: patchwork-parsemail@patchwork2.web.kernel.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.136]) by patchwork2.web.kernel.org (Postfix) with ESMTP id 8FD81C05C6 for ; Fri, 6 Nov 2015 17:22:42 +0000 (UTC) Received: from mail.kernel.org (localhost [127.0.0.1]) by mail.kernel.org (Postfix) with ESMTP id 8516720619 for ; Fri, 6 Nov 2015 17:22:41 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.kernel.org (Postfix) with ESMTP id 53D76205B4 for ; Fri, 6 Nov 2015 17:22:40 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S933023AbbKFRVi (ORCPT ); Fri, 6 Nov 2015 12:21:38 -0500 Received: from mx0a-00082601.pphosted.com ([67.231.145.42]:42650 "EHLO mx0a-00082601.pphosted.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1161797AbbKFRVC (ORCPT ); Fri, 6 Nov 2015 12:21:02 -0500 Received: from pps.filterd (m0044010.ppops.net [127.0.0.1]) by mx0a-00082601.pphosted.com (8.15.0.59/8.15.0.59) with SMTP id tA6HK1ft022208; Fri, 6 Nov 2015 09:20:31 -0800 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=fb.com; h=from : to : cc : subject : date : message-id : in-reply-to : references : mime-version : content-type; s=facebook; bh=YIsLcS5QTruq7qBwcbbMIxw1hJZO7GRxu7hSU+yvY6c=; b=Jj0pQ1YJ1o3Kp8j2+ivBVqB2TrFjNdq0jcoir1SCF+cipWGC9o4l7GJqHNJkLeBFv8Tv 6wVD99cmZdVgCn1i0jnluHVOke3wKw/SYw7pJQbHOApJZyGAIxyCaWkB+qbl5F0qJHEd oJUQBcDNxNUurMmfdZkwrlhk+EAJ3CeaH5o= Received: from mail.thefacebook.com ([199.201.64.23]) by mx0a-00082601.pphosted.com with ESMTP id 1xxfbh0fw4-2 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=NOT); Fri, 06 Nov 2015 09:20:31 -0800 Received: from localhost.localdomain (192.168.54.13) by mail.thefacebook.com (192.168.16.15) with Microsoft SMTP Server (TLS) id 14.3.248.2; Fri, 6 Nov 2015 09:20:30 -0800 From: Jens Axboe To: , CC: , , Jens Axboe Subject: [PATCH 3/5] block: add block polling support Date: Fri, 6 Nov 2015 10:20:21 -0700 Message-ID: <1446830423-25027-4-git-send-email-axboe@fb.com> X-Mailer: git-send-email 2.4.1.168.g1ea28e1 In-Reply-To: <1446830423-25027-1-git-send-email-axboe@fb.com> References: <1446830423-25027-1-git-send-email-axboe@fb.com> MIME-Version: 1.0 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=2015-11-06_11:, , signatures=0 Sender: linux-block-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-block@vger.kernel.org X-Spam-Status: No, score=-6.8 required=5.0 tests=BAYES_00,DKIM_SIGNED, RCVD_IN_DNSWL_HI,T_DKIM_INVALID,T_RP_MATCHES_RCVD,UNPARSEABLE_RELAY autolearn=unavailable version=3.3.1 X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on mail.kernel.org X-Virus-Scanned: ClamAV using ClamSMTP Add basic support for polling for specific IO to complete. This uses the cookie that blk-mq passes back, which enables the block layer to pass this cookie to the driver to spin for a specific request. This will be combined with request latency tracking, so we can make qualified decisions about when to poll and when not to. For now, for benchmark purposes, we add a sysfs file that controls whether polling is enabled or not. Signed-off-by: Jens Axboe --- block/blk-core.c | 41 +++++++++++++++++++++++++++++++++++++++++ block/blk-mq-sysfs.c | 10 ++++++++++ block/blk-sysfs.c | 35 +++++++++++++++++++++++++++++++++++ include/linux/blk-mq.h | 10 ++++++++++ include/linux/blkdev.h | 3 +++ 5 files changed, 99 insertions(+) diff --git a/block/blk-core.c b/block/blk-core.c index e93df6d386a0..fa36b4ff7d63 100644 --- a/block/blk-core.c +++ b/block/blk-core.c @@ -3312,6 +3312,47 @@ void blk_finish_plug(struct blk_plug *plug) } EXPORT_SYMBOL(blk_finish_plug); +bool blk_poll(struct request_queue *q, blk_qc_t cookie) +{ + struct blk_plug *plug; + long state; + + if (!q->mq_ops || !q->mq_ops->poll || !blk_qc_t_valid(cookie) || + !test_bit(QUEUE_FLAG_POLL, &q->queue_flags)) + return false; + + plug = current->plug; + if (plug) + blk_flush_plug_list(plug, false); + + state = current->state; + while (!need_resched()) { + unsigned int queue_num = blk_qc_t_to_queue_num(cookie); + struct blk_mq_hw_ctx *hctx = q->queue_hw_ctx[queue_num]; + int ret; + + hctx->poll_invoked++; + + ret = q->mq_ops->poll(hctx, blk_qc_t_to_tag(cookie)); + if (ret > 0) { + hctx->poll_success++; + set_current_state(TASK_RUNNING); + return true; + } + + if (signal_pending_state(state, current)) + set_current_state(TASK_RUNNING); + + if (current->state == TASK_RUNNING) + return true; + if (ret < 0) + break; + cpu_relax(); + } + + return false; +} + #ifdef CONFIG_PM /** * blk_pm_runtime_init - Block layer runtime PM initialization routine diff --git a/block/blk-mq-sysfs.c b/block/blk-mq-sysfs.c index 6f57a110289c..1cf18784c5cf 100644 --- a/block/blk-mq-sysfs.c +++ b/block/blk-mq-sysfs.c @@ -174,6 +174,11 @@ static ssize_t blk_mq_sysfs_rq_list_show(struct blk_mq_ctx *ctx, char *page) return ret; } +static ssize_t blk_mq_hw_sysfs_poll_show(struct blk_mq_hw_ctx *hctx, char *page) +{ + return sprintf(page, "invoked=%lu, success=%lu\n", hctx->poll_invoked, hctx->poll_success); +} + static ssize_t blk_mq_hw_sysfs_queued_show(struct blk_mq_hw_ctx *hctx, char *page) { @@ -295,6 +300,10 @@ static struct blk_mq_hw_ctx_sysfs_entry blk_mq_hw_sysfs_cpus = { .attr = {.name = "cpu_list", .mode = S_IRUGO }, .show = blk_mq_hw_sysfs_cpus_show, }; +static struct blk_mq_hw_ctx_sysfs_entry blk_mq_hw_sysfs_poll = { + .attr = {.name = "io_poll", .mode = S_IRUGO }, + .show = blk_mq_hw_sysfs_poll_show, +}; static struct attribute *default_hw_ctx_attrs[] = { &blk_mq_hw_sysfs_queued.attr, @@ -304,6 +313,7 @@ static struct attribute *default_hw_ctx_attrs[] = { &blk_mq_hw_sysfs_tags.attr, &blk_mq_hw_sysfs_cpus.attr, &blk_mq_hw_sysfs_active.attr, + &blk_mq_hw_sysfs_poll.attr, NULL, }; diff --git a/block/blk-sysfs.c b/block/blk-sysfs.c index 31849e328b45..565b8dac5782 100644 --- a/block/blk-sysfs.c +++ b/block/blk-sysfs.c @@ -317,6 +317,34 @@ queue_rq_affinity_store(struct request_queue *q, const char *page, size_t count) return ret; } +static ssize_t queue_poll_show(struct request_queue *q, char *page) +{ + return queue_var_show(test_bit(QUEUE_FLAG_POLL, &q->queue_flags), page); +} + +static ssize_t queue_poll_store(struct request_queue *q, const char *page, + size_t count) +{ + unsigned long poll_on; + ssize_t ret; + + if (!q->mq_ops || !q->mq_ops->poll) + return -EINVAL; + + ret = queue_var_store(&poll_on, page, count); + if (ret < 0) + return ret; + + spin_lock_irq(q->queue_lock); + if (poll_on) + queue_flag_set(QUEUE_FLAG_POLL, q); + else + queue_flag_clear(QUEUE_FLAG_POLL, q); + spin_unlock_irq(q->queue_lock); + + return ret; +} + static struct queue_sysfs_entry queue_requests_entry = { .attr = {.name = "nr_requests", .mode = S_IRUGO | S_IWUSR }, .show = queue_requests_show, @@ -442,6 +470,12 @@ static struct queue_sysfs_entry queue_random_entry = { .store = queue_store_random, }; +static struct queue_sysfs_entry queue_poll_entry = { + .attr = {.name = "io_poll", .mode = S_IRUGO | S_IWUSR }, + .show = queue_poll_show, + .store = queue_poll_store, +}; + static struct attribute *default_attrs[] = { &queue_requests_entry.attr, &queue_ra_entry.attr, @@ -466,6 +500,7 @@ static struct attribute *default_attrs[] = { &queue_rq_affinity_entry.attr, &queue_iostats_entry.attr, &queue_random_entry.attr, + &queue_poll_entry.attr, NULL, }; diff --git a/include/linux/blk-mq.h b/include/linux/blk-mq.h index 83cc9d4e5455..daf17d70aeca 100644 --- a/include/linux/blk-mq.h +++ b/include/linux/blk-mq.h @@ -59,6 +59,9 @@ struct blk_mq_hw_ctx { struct blk_mq_cpu_notifier cpu_notifier; struct kobject kobj; + + unsigned long poll_invoked; + unsigned long poll_success; }; struct blk_mq_tag_set { @@ -97,6 +100,8 @@ typedef void (exit_request_fn)(void *, struct request *, unsigned int, typedef void (busy_iter_fn)(struct blk_mq_hw_ctx *, struct request *, void *, bool); typedef void (busy_tag_iter_fn)(struct request *, void *, bool); +typedef int (poll_fn)(struct blk_mq_hw_ctx *, unsigned int); + struct blk_mq_ops { /* @@ -114,6 +119,11 @@ struct blk_mq_ops { */ timeout_fn *timeout; + /* + * Called to poll for completion of a specific tag. + */ + poll_fn *poll; + softirq_done_fn *complete; /* diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 5ee0f5243025..3fe27f8d91f0 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -487,6 +487,7 @@ struct request_queue { #define QUEUE_FLAG_DEAD 19 /* queue tear-down finished */ #define QUEUE_FLAG_INIT_DONE 20 /* queue is initialized */ #define QUEUE_FLAG_NO_SG_MERGE 21 /* don't attempt to merge SG segments*/ +#define QUEUE_FLAG_POLL 22 /* IO polling enabled if set */ #define QUEUE_FLAG_DEFAULT ((1 << QUEUE_FLAG_IO_STAT) | \ (1 << QUEUE_FLAG_STACKABLE) | \ @@ -814,6 +815,8 @@ extern int blk_execute_rq(struct request_queue *, struct gendisk *, extern void blk_execute_rq_nowait(struct request_queue *, struct gendisk *, struct request *, int, rq_end_io_fn *); +bool blk_poll(struct request_queue *q, blk_qc_t cookie); + static inline struct request_queue *bdev_get_queue(struct block_device *bdev) { return bdev->bd_disk->queue; /* this is never NULL */