From patchwork Fri Aug 4 05:35:11 2017 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Paolo Valente X-Patchwork-Id: 9880513 Return-Path: Received: from mail.wl.linuxfoundation.org (pdx-wl-mail.web.codeaurora.org [172.30.200.125]) by pdx-korg-patchwork.web.codeaurora.org (Postfix) with ESMTP id 2749560360 for ; Fri, 4 Aug 2017 05:35:34 +0000 (UTC) Received: from mail.wl.linuxfoundation.org (localhost [127.0.0.1]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id 2267C28911 for ; Fri, 4 Aug 2017 05:35:34 +0000 (UTC) Received: by mail.wl.linuxfoundation.org (Postfix, from userid 486) id 17330289A2; Fri, 4 Aug 2017 05:35:34 +0000 (UTC) X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on pdx-wl-mail.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-6.5 required=2.0 tests=BAYES_00,DKIM_SIGNED, DKIM_VALID, DKIM_VALID_AU, RCVD_IN_DNSWL_HI, RCVD_IN_SORBS_SPAM autolearn=ham version=3.3.1 Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id A044328906 for ; Fri, 4 Aug 2017 05:35:33 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751284AbdHDFfb (ORCPT ); Fri, 4 Aug 2017 01:35:31 -0400 Received: from mail-wm0-f47.google.com ([74.125.82.47]:37770 "EHLO mail-wm0-f47.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751324AbdHDFf3 (ORCPT ); Fri, 4 Aug 2017 01:35:29 -0400 Received: by mail-wm0-f47.google.com with SMTP id t201so11054368wmt.0 for ; Thu, 03 Aug 2017 22:35:29 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=linaro.org; s=google; h=from:to:cc:subject:date:message-id:in-reply-to:references; bh=CIRX8Y9MElfdXj5FRrqnkXLtIJPsU+MC4PYFAK8m4tc=; b=CYw9TBxAY6RA4b76kYjwVaNYFukBImYStCoS0V6J1NsfD0+ehVRqKAwBQCTb8bh3R7 B3hHqGcfRRjHcX/+SUm8t5w6JTvhz6JZ+sFHLSJ9aR1EVOdIeyOtNdhIN2T2aeG1Prqv GDQeHDv33LToIeUdUJiK803M38fLyjc7DUPHQ= X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:from:to:cc:subject:date:message-id:in-reply-to :references; bh=CIRX8Y9MElfdXj5FRrqnkXLtIJPsU+MC4PYFAK8m4tc=; b=muPKXKZ4na/eGv323KBu/RAjgX7bk0vY8m5z/aFsgdjMU4uj0iZ8ddcXDhj+StC/zH c8SZNAjngOXQisufO/jR8kmdB/wtpekRloDe4bGs/B8SEskT8z40KpX9sWkw0/PowERZ jWO5+WE2wY0A3CrHcE0BgYWrNqkuDUBXklMf8s7r0rWA3K18hvvJ/oTyGEQkcDXV4/ws GY+RLSK4Fg7iTmnuvKY5WSNLYtCna6s/+Z8m1yr+XFmznWJYfvqqKR6hFQFNfwdJHv/V vYWDhwoUspT6u4uJwSYKBlgFx2Y/Be9udf5upx+zpYGYFU+AEgek2XDQXgZpoxG+zu43 4eIw== X-Gm-Message-State: AHYfb5isPHqTYx5L+SuqPpqIO8WGuPE4maYRKvJotbAzAlup1XlqjaKE TVbEFMEifeTz8zn7 X-Received: by 10.28.170.18 with SMTP id t18mr512304wme.6.1501824928496; Thu, 03 Aug 2017 22:35:28 -0700 (PDT) Received: from localhost.localdomain ([185.14.11.62]) by smtp.gmail.com with ESMTPSA id 196sm3082142wmg.36.2017.08.03.22.35.26 (version=TLS1 cipher=AES128-SHA bits=128/128); Thu, 03 Aug 2017 22:35:27 -0700 (PDT) From: Paolo Valente To: Jens Axboe Cc: linux-block@vger.kernel.org, linux-kernel@vger.kernel.org, ulf.hansson@linaro.org, broonie@kernel.org, lucmiccio@gmail.com, Paolo Valente Subject: [PATCH BUGFIX/IMPROVEMENT V2 2/2] block, bfq: boost throughput with flash-based non-queueing devices Date: Fri, 4 Aug 2017 07:35:11 +0200 Message-Id: <20170804053511.2389-3-paolo.valente@linaro.org> X-Mailer: git-send-email 2.10.0 In-Reply-To: <20170804053511.2389-1-paolo.valente@linaro.org> References: <20170804053511.2389-1-paolo.valente@linaro.org> Sender: linux-block-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-block@vger.kernel.org X-Virus-Scanned: ClamAV using ClamSMTP When a queue associated with a process remains empty, there are cases where throughput gets boosted if the device is idled to await the arrival of a new I/O request for that queue. Currently, BFQ assumes that one of these cases is when the device has no internal queueing (regardless of the properties of the I/O being served). Unfortunately, this condition has proved to be too general. So, this commit refines it as "the device has no internal queueing and is rotational". This refinement provides a significant throughput boost with random I/O, on flash-based storage without internal queueing. For example, on a HiKey board, throughput increases by up to 125%, growing, e.g., from 6.9MB/s to 15.6MB/s with two or three random readers in parallel. Signed-off-by: Paolo Valente Signed-off-by: Luca Miccio --- block/bfq-iosched.c | 29 +++++++++++++++++++---------- 1 file changed, 19 insertions(+), 10 deletions(-) diff --git a/block/bfq-iosched.c b/block/bfq-iosched.c index ccdc9e6..509f399 100644 --- a/block/bfq-iosched.c +++ b/block/bfq-iosched.c @@ -3114,7 +3114,10 @@ static bool bfq_may_expire_for_budg_timeout(struct bfq_queue *bfqq) static bool bfq_bfqq_may_idle(struct bfq_queue *bfqq) { struct bfq_data *bfqd = bfqq->bfqd; - bool idling_boosts_thr, idling_boosts_thr_without_issues, + bool rot_without_queueing = + !blk_queue_nonrot(bfqd->queue) && !bfqd->hw_tag, + bfqq_sequential_and_IO_bound, + idling_boosts_thr, idling_boosts_thr_without_issues, idling_needed_for_service_guarantees, asymmetric_scenario; @@ -3133,28 +3136,34 @@ static bool bfq_bfqq_may_idle(struct bfq_queue *bfqq) bfq_class_idle(bfqq)) return false; + bfqq_sequential_and_IO_bound = !BFQQ_SEEKY(bfqq) && + bfq_bfqq_IO_bound(bfqq) && bfq_bfqq_has_short_ttime(bfqq); + /* * The next variable takes into account the cases where idling * boosts the throughput. * * The value of the variable is computed considering, first, that * idling is virtually always beneficial for the throughput if: - * (a) the device is not NCQ-capable, or - * (b) regardless of the presence of NCQ, the device is rotational - * and the request pattern for bfqq is I/O-bound and sequential. + * (a) the device is not NCQ-capable and rotational, or + * (b) regardless of the presence of NCQ, the device is rotational and + * the request pattern for bfqq is I/O-bound and sequential, or + * (c) regardless of whether it is rotational, the device is + * not NCQ-capable and the request pattern for bfqq is + * I/O-bound and sequential. * * Secondly, and in contrast to the above item (b), idling an * NCQ-capable flash-based device would not boost the * throughput even with sequential I/O; rather it would lower * the throughput in proportion to how fast the device * is. Accordingly, the next variable is true if any of the - * above conditions (a) and (b) is true, and, in particular, - * happens to be false if bfqd is an NCQ-capable flash-based - * device. + * above conditions (a), (b) or (c) is true, and, in + * particular, happens to be false if bfqd is an NCQ-capable + * flash-based device. */ - idling_boosts_thr = !bfqd->hw_tag || - (!blk_queue_nonrot(bfqd->queue) && bfq_bfqq_IO_bound(bfqq) && - bfq_bfqq_has_short_ttime(bfqq)); + idling_boosts_thr = rot_without_queueing || + ((!blk_queue_nonrot(bfqd->queue) || !bfqd->hw_tag) && + bfqq_sequential_and_IO_bound); /* * The value of the next variable,