From patchwork Thu Oct 19 11:21:13 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Christian Loehle X-Patchwork-Id: 155444 Return-Path: Delivered-To: ouuuleilei@gmail.com Received: by 2002:a05:612c:2010:b0:403:3b70:6f57 with SMTP id fe16csp310053vqb; Thu, 19 Oct 2023 04:21:29 -0700 (PDT) X-Google-Smtp-Source: AGHT+IGTW7JM4pikfOS4V1QStkYFYZ8mHu5/SP8Aj4ihuV4fk5koml/oSm9Nc/Fh9X6anLnL4bxG X-Received: by 2002:a05:6a20:8f1e:b0:15f:b5:bfee with SMTP id b30-20020a056a208f1e00b0015f00b5bfeemr2023071pzk.5.1697714489521; Thu, 19 Oct 2023 04:21:29 -0700 (PDT) ARC-Seal: i=1; a=rsa-sha256; t=1697714489; cv=none; d=google.com; s=arc-20160816; b=g08Ul6YbojSL0porWnjpKJEDbtytkabJIITnW1TW4pvCHz37LJN8FsYnsfArmZQOfp P6Pa3bJIe/wy+s5hqHuwahLOIa87cs4RxC+Tmgp7L/jeSzHq6+KgmVnF+T6eCf1JvTHS qVSieFcywgr5PCudJcCHn8I4I3zlo/N3zvHFTFPFZFWx1Hf/7ccXNnLJPaWGhOvINZuV BRMksh9Ky/0WRYxmvtB9ERp1T8GNRvqN0Ejxjw6CRx71GNFpWPSdfI2F4E0zk/VUtOmZ RxQZJBZ3HEJHGpAP3u/6g2PtjS/5XyuvwL4PCnrZVU0FnHxV9igIjbDI/F8DkedZifea Dw/w== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=arc-20160816; h=list-id:precedence:content-transfer-encoding:subject:from:to :content-language:user-agent:mime-version:date:message-id; bh=oRRCb9H3yh6pQVVO8Y1K2ATTdUYTnzI0VEpLbM+3Oso=; fh=qUtrYitbc9BTpOxRN2dK4NxGRZWccnftlAmeDuIWb+Y=; b=YX6CcNBnCEEgztUsMsWqjtMqlmmCTH5AWPt0eGPn3f70rEtPIfniR9vjRrpfc3nMgd cp9hkOTPVPjSDx/G2H/s6cTmi0UHsp/V/DMgtlaGTQ4OPm6KGMuOQdH4FPBboMUxdw95 K3TyflIkSm3GLt1YgugIdKaXdLll+7+j+DxJM559+Og6OFvcWHHTNaEnifTMwG2hky/d EeZCBDMKF3EaxpSDJ7dQdtHK/1LGzdfJ+5jtELhWDgut4ltGZ7x+SSH7a5QueiLcwwzW ymhaYpbus9D15r/nhX7yq8WtKnxxWWnYfGRBJbk/LN3CY5gGloiqKP6jVGEUzw8BL5gZ p2Lg== ARC-Authentication-Results: i=1; mx.google.com; spf=pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.37 as permitted sender) smtp.mailfrom=linux-kernel-owner@vger.kernel.org; dmarc=fail (p=NONE sp=NONE dis=NONE) header.from=arm.com Received: from snail.vger.email (snail.vger.email. [23.128.96.37]) by mx.google.com with ESMTPS id qe18-20020a17090b4f9200b0027d0adde858si1957486pjb.48.2023.10.19.04.21.29 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Thu, 19 Oct 2023 04:21:29 -0700 (PDT) Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.37 as permitted sender) client-ip=23.128.96.37; Authentication-Results: mx.google.com; spf=pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.37 as permitted sender) smtp.mailfrom=linux-kernel-owner@vger.kernel.org; dmarc=fail (p=NONE sp=NONE dis=NONE) header.from=arm.com Received: from out1.vger.email (depot.vger.email [IPv6:2620:137:e000::3:0]) by snail.vger.email (Postfix) with ESMTP id 937CF82164DC; Thu, 19 Oct 2023 04:21:27 -0700 (PDT) X-Virus-Status: Clean X-Virus-Scanned: clamav-milter 0.103.10 at snail.vger.email Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1345329AbjJSLVW (ORCPT + 25 others); Thu, 19 Oct 2023 07:21:22 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:53828 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S233051AbjJSLVV (ORCPT ); Thu, 19 Oct 2023 07:21:21 -0400 Received: from foss.arm.com (foss.arm.com [217.140.110.172]) by lindbergh.monkeyblade.net (Postfix) with ESMTP id 771EABE for ; Thu, 19 Oct 2023 04:21:18 -0700 (PDT) Received: from usa-sjc-imap-foss1.foss.arm.com (unknown [10.121.207.14]) by usa-sjc-mx-foss1.foss.arm.com (Postfix) with ESMTP id BE4452F4; Thu, 19 Oct 2023 04:21:58 -0700 (PDT) Received: from [10.57.35.228] (unknown [10.57.35.228]) by usa-sjc-imap-foss1.foss.arm.com (Postfix) with ESMTPSA id 14CC43F762; Thu, 19 Oct 2023 04:21:15 -0700 (PDT) Message-ID: Date: Thu, 19 Oct 2023 12:21:13 +0100 MIME-Version: 1.0 User-Agent: Mozilla Thunderbird Content-Language: en-US To: agk@redhat.com, snitzer@kernel.org, dm-devel@lists.linux.dev, linux-kernel@vger.kernel.org From: Christian Loehle Subject: [PATCH] dm: delay: use kthread instead of timers and wq X-Spam-Status: No, score=-4.2 required=5.0 tests=BAYES_00,RCVD_IN_DNSWL_MED, SPF_HELO_NONE,SPF_NONE autolearn=ham autolearn_force=no version=3.4.6 X-Spam-Checker-Version: SpamAssassin 3.4.6 (2021-04-09) on lindbergh.monkeyblade.net Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org X-Greylist: Sender passed SPF test, not delayed by milter-greylist-4.6.4 (snail.vger.email [0.0.0.0]); Thu, 19 Oct 2023 04:21:27 -0700 (PDT) X-getmail-retrieved-from-mailbox: INBOX X-GMAIL-THRID: 1780182668550117114 X-GMAIL-MSGID: 1780182668550117114 The current design of timers and wq to realize the delays is insufficient especially for delays below ~50ms. The design is replaced with a kthread to flush the expired delays, trading some CPU time (in some cases) for better delay accuracy and delays closer to what the user requested for smaller delays. Since bios can't be completed in interrupt context using a kthread is probably the most reasonable way to approach this. Testing with echo "0 2097152 zero" | dmsetup create dm-zeros for i in $(seq 0 20); do echo "0 2097152 delay /dev/mapper/dm-zeros 0 $i" | dmsetup create dm-delay-${i}ms; done Some performance numbers for comparison beaglebone black (single core) CONFIG_HZ_1000=y: fio --name=1msread --rw=randread --bs=4k --runtime=60 --time_based --filename=/dev/mapper/dm-delay-1ms Theoretical maximum: 1000 IOPS Previous: 250 IOPS Kthread: 500 IOPS fio --name=10msread --rw=randread --bs=4k --runtime=60 --time_based --filename=/dev/mapper/dm-delay-10ms Theoretical maximum: 1000 IOPS Previous: 45 IOPS Kthread: 50 IOPS fio --name=1mswrite --rw=randwrite --direct=1 --bs=4k --runtime=60 --time_based --filename=/dev/mapper/dm-delay-1ms Theoretical maximum: 1000 IOPS Previous: 498 IOPS Kthread: 1000 IOPS fio --name=10mswrite --rw=randwrite --direct=1 --bs=4k --runtime=60 --time_based --filename=/dev/mapper/dm-delay-10ms Theoretical maximum: 100 IOPS Previous: 90 IOPS Kthread: 100 IOPS fio --name=10mswriteasync --rw=randwrite --direct=1 --bs=4k --runtime=60 --time_based --filename=/dev/mapper/dm-delay-10ms --numjobs=32 --iodepth=64 --ioengine=libaio --group_reporting Previous: 13.3k IOPS Kthread: 13.3k IOPS (This one is just to prove the new design isn't impacting throughput, not really about delays) Signed-off-by: Christian Loehle --- drivers/md/dm-delay.c | 83 ++++++++++++------------------------------- 1 file changed, 22 insertions(+), 61 deletions(-) diff --git a/drivers/md/dm-delay.c b/drivers/md/dm-delay.c index 7433525e5985..a37f28294f7e 100644 --- a/drivers/md/dm-delay.c +++ b/drivers/md/dm-delay.c @@ -13,6 +13,7 @@ #include #include #include +#include #include @@ -26,11 +27,9 @@ struct delay_class { }; struct delay_c { - struct timer_list delay_timer; - struct mutex timer_lock; - struct workqueue_struct *kdelayd_wq; struct work_struct flush_expired_bios; struct list_head delayed_bios; + struct task_struct *worker; atomic_t may_delay; struct delay_class read; @@ -48,42 +47,25 @@ struct dm_delay_info { }; static DEFINE_MUTEX(delayed_bios_lock); +static void flush_delayed_bios(struct delay_c *dc, int flush_all); -static void handle_delayed_timer(struct timer_list *t) +static int flush_worker_fn(void *data) { - struct delay_c *dc = from_timer(dc, t, delay_timer); + struct delay_c *dc = (struct delay_c *)data; - queue_work(dc->kdelayd_wq, &dc->flush_expired_bios); -} - -static void queue_timeout(struct delay_c *dc, unsigned long expires) -{ - mutex_lock(&dc->timer_lock); - - if (!timer_pending(&dc->delay_timer) || expires < dc->delay_timer.expires) - mod_timer(&dc->delay_timer, expires); - - mutex_unlock(&dc->timer_lock); -} - -static void flush_bios(struct bio *bio) -{ - struct bio *n; - - while (bio) { - n = bio->bi_next; - bio->bi_next = NULL; - dm_submit_bio_remap(bio, NULL); - bio = n; + while (1) { + flush_delayed_bios(dc, 0); + if (unlikely(list_empty(&dc->delayed_bios))) + set_current_state(TASK_INTERRUPTIBLE); + schedule(); } + + return 0; } -static struct bio *flush_delayed_bios(struct delay_c *dc, int flush_all) +static void flush_delayed_bios(struct delay_c *dc, int flush_all) { struct dm_delay_info *delayed, *next; - unsigned long next_expires = 0; - unsigned long start_timer = 0; - struct bio_list flush_bios = { }; mutex_lock(&delayed_bios_lock); list_for_each_entry_safe(delayed, next, &dc->delayed_bios, list) { @@ -91,23 +73,12 @@ static struct bio *flush_delayed_bios(struct delay_c *dc, int flush_all) struct bio *bio = dm_bio_from_per_bio_data(delayed, sizeof(struct dm_delay_info)); list_del(&delayed->list); - bio_list_add(&flush_bios, bio); + dm_submit_bio_remap(bio, NULL); delayed->class->ops--; - continue; } - - if (!start_timer) { - start_timer = 1; - next_expires = delayed->expires; - } else - next_expires = min(next_expires, delayed->expires); } mutex_unlock(&delayed_bios_lock); - if (start_timer) - queue_timeout(dc, next_expires); - - return bio_list_get(&flush_bios); } static void flush_expired_bios(struct work_struct *work) @@ -115,24 +86,21 @@ static void flush_expired_bios(struct work_struct *work) struct delay_c *dc; dc = container_of(work, struct delay_c, flush_expired_bios); - flush_bios(flush_delayed_bios(dc, 0)); + flush_delayed_bios(dc, 0); } static void delay_dtr(struct dm_target *ti) { struct delay_c *dc = ti->private; - if (dc->kdelayd_wq) - destroy_workqueue(dc->kdelayd_wq); - if (dc->read.dev) dm_put_device(ti, dc->read.dev); if (dc->write.dev) dm_put_device(ti, dc->write.dev); if (dc->flush.dev) dm_put_device(ti, dc->flush.dev); - - mutex_destroy(&dc->timer_lock); + if (dc->worker) + kthread_stop(dc->worker); kfree(dc); } @@ -188,10 +156,11 @@ static int delay_ctr(struct dm_target *ti, unsigned int argc, char **argv) } ti->private = dc; - timer_setup(&dc->delay_timer, handle_delayed_timer, 0); INIT_WORK(&dc->flush_expired_bios, flush_expired_bios); INIT_LIST_HEAD(&dc->delayed_bios); - mutex_init(&dc->timer_lock); + dc->worker = kthread_create(&flush_worker_fn, dc, "dm-delay-flush-worker"); + if (!dc->worker) + goto bad; atomic_set(&dc->may_delay, 1); dc->argc = argc; @@ -224,13 +193,6 @@ static int delay_ctr(struct dm_target *ti, unsigned int argc, char **argv) goto bad; out: - dc->kdelayd_wq = alloc_workqueue("kdelayd", WQ_MEM_RECLAIM, 0); - if (!dc->kdelayd_wq) { - ret = -EINVAL; - DMERR("Couldn't start kdelayd"); - goto bad; - } - ti->num_flush_bios = 1; ti->num_discard_bios = 1; ti->accounts_remapped_io = true; @@ -260,7 +222,7 @@ static int delay_bio(struct delay_c *dc, struct delay_class *c, struct bio *bio) list_add_tail(&delayed->list, &dc->delayed_bios); mutex_unlock(&delayed_bios_lock); - queue_timeout(dc, expires); + wake_up_process(dc->worker); return DM_MAPIO_SUBMITTED; } @@ -270,8 +232,7 @@ static void delay_presuspend(struct dm_target *ti) struct delay_c *dc = ti->private; atomic_set(&dc->may_delay, 0); - del_timer_sync(&dc->delay_timer); - flush_bios(flush_delayed_bios(dc, 1)); + flush_delayed_bios(dc, 1); } static void delay_resume(struct dm_target *ti)