From patchwork Thu Sep 15 16:23:18 2016 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Shaohua Li X-Patchwork-Id: 9334211 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 DA32A607FD for ; Thu, 15 Sep 2016 16:25:07 +0000 (UTC) Received: from mail.wl.linuxfoundation.org (localhost [127.0.0.1]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id CE6BB2993F for ; Thu, 15 Sep 2016 16:25:07 +0000 (UTC) Received: by mail.wl.linuxfoundation.org (Postfix, from userid 486) id C088F29A53; Thu, 15 Sep 2016 16:25:07 +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=-6.8 required=2.0 tests=BAYES_00,DKIM_SIGNED, RCVD_IN_DNSWL_HI,T_DKIM_INVALID autolearn=unavailable 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 09CC029A11 for ; Thu, 15 Sep 2016 16:25:06 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753307AbcIOQY7 (ORCPT ); Thu, 15 Sep 2016 12:24:59 -0400 Received: from mx0a-00082601.pphosted.com ([67.231.145.42]:33485 "EHLO mx0a-00082601.pphosted.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754891AbcIOQXW (ORCPT ); Thu, 15 Sep 2016 12:23:22 -0400 Received: from pps.filterd (m0044010.ppops.net [127.0.0.1]) by mx0a-00082601.pphosted.com (8.16.0.17/8.16.0.17) with SMTP id u8FGFE6a022591 for ; Thu, 15 Sep 2016 09:23:22 -0700 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=hwzDIk2YwKrVknbG6N6wGqnZQFosnvEyhraRDPKxoc8=; b=N+FAw/p8RjWXNoZN/q1sF1DYe013Ye6IqovcGaly+JQdZ4+q7CEjIctBcEsSvQSiSxiL 3Lh4mqaH+IpsYpf24s8EBxaGKjFWzRAjhYOdyaEy0cQWK2hQZ6p0/MdWyWvtsllx9mpQ AzwDx6X06rYMSZ9M8o17aGLoPxWYN6YCj10= Received: from mail.thefacebook.com ([199.201.64.23]) by mx0a-00082601.pphosted.com with ESMTP id 25fww30vgw-6 (version=TLSv1 cipher=ECDHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Thu, 15 Sep 2016 09:23:22 -0700 Received: from mx-out.facebook.com (192.168.52.123) by PRN-CHUB03.TheFacebook.com (192.168.16.13) with Microsoft SMTP Server (TLS) id 14.3.294.0; Thu, 15 Sep 2016 09:23:21 -0700 Received: from facebook.com (2401:db00:11:d0a2:face:0:39:0) by mx-out.facebook.com (10.223.100.99) with ESMTP id b76616447b6011e680c424be05956610-7b8e2a50 for ; Thu, 15 Sep 2016 09:23:20 -0700 Received: by devbig084.prn1.facebook.com (Postfix, from userid 11222) id 0399B4800724; Thu, 15 Sep 2016 09:23:18 -0700 (PDT) From: Shaohua Li To: , CC: , , , , Subject: [PATCH V2 11/11] blk-throttle: ignore idle cgroup limit Date: Thu, 15 Sep 2016 09:23:18 -0700 Message-ID: <8c687c3d686fda706084b748e757e8b9e1cffd47.1473953743.git.shli@fb.com> X-Mailer: git-send-email 2.8.0.rc2 In-Reply-To: References: X-FB-Internal: Safe MIME-Version: 1.0 X-Proofpoint-Spam-Reason: safe X-FB-Internal: Safe X-Proofpoint-Virus-Version: vendor=fsecure engine=2.50.10432:, , definitions=2016-09-15_08:, , signatures=0 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 Last patch introduces a way to detect idle cgroup. We use it to make upgrade/downgrade decision. Signed-off-by: Shaohua Li --- block/blk-throttle.c | 30 ++++++++++++++++++------------ 1 file changed, 18 insertions(+), 12 deletions(-) diff --git a/block/blk-throttle.c b/block/blk-throttle.c index 0810e1b..276e71e 100644 --- a/block/blk-throttle.c +++ b/block/blk-throttle.c @@ -145,8 +145,6 @@ struct throtl_grp { unsigned long last_check_time; - unsigned long last_dispatch_time[2]; - /* When did we start a new slice */ unsigned long slice_start[2]; unsigned long slice_end[2]; @@ -488,8 +486,6 @@ static void throtl_pd_online(struct blkg_policy_data *pd) * Update has_rules[] after a new group is brought online. */ tg_update_has_rules(tg); - tg->last_dispatch_time[READ] = jiffies; - tg->last_dispatch_time[WRITE] = jiffies; } static void blk_throtl_update_valid_limit(struct throtl_data *td) @@ -1661,9 +1657,8 @@ static bool throtl_upgrade_check_one(struct throtl_grp *tg) return true; if (time_after_eq(jiffies, - tg->last_dispatch_time[READ] + tg->td->throtl_slice) && - time_after_eq(jiffies, - tg->last_dispatch_time[WRITE] + tg->td->throtl_slice)) + tg_last_high_overflow_time(tg) + tg->td->throtl_slice) && + throtl_tg_is_idle(tg)) return true; return false; } @@ -1712,6 +1707,19 @@ static bool throtl_can_upgrade(struct throtl_data *td, return true; } +static void throtl_upgrade_check(struct throtl_grp *tg) +{ + if (tg->td->limit_index != LIMIT_HIGH) + return; + + if (!time_after_eq(jiffies, + __tg_last_high_overflow_time(tg) + tg->td->throtl_slice)) + return; + + if (throtl_can_upgrade(tg->td, NULL)) + throtl_upgrade_state(tg->td); +} + static void throtl_upgrade_state(struct throtl_data *td) { struct cgroup_subsys_state *pos_css; @@ -1746,15 +1754,13 @@ static bool throtl_downgrade_check_one(struct throtl_grp *tg) struct throtl_data *td = tg->td; unsigned long now = jiffies; - if (time_after_eq(now, tg->last_dispatch_time[READ] + td->throtl_slice) && - time_after_eq(now, tg->last_dispatch_time[WRITE] + td->throtl_slice)) - return false; /* * If cgroup is below high limit, consider downgrade and throttle other * cgroups */ if (time_after_eq(now, td->high_upgrade_time + td->throtl_slice) && - time_after_eq(now, tg_last_high_overflow_time(tg) + td->throtl_slice)) + time_after_eq(now, tg_last_high_overflow_time(tg) + td->throtl_slice) && + !throtl_tg_is_idle(tg)) return true; return false; } @@ -1885,10 +1891,10 @@ bool blk_throtl_bio(struct request_queue *q, struct blkcg_gq *blkg, again: while (true) { - tg->last_dispatch_time[rw] = jiffies; if (tg->last_high_overflow_time[rw] == 0) tg->last_high_overflow_time[rw] = jiffies; throtl_downgrade_check(tg); + throtl_upgrade_check(tg); /* throtl is FIFO - if bios are already queued, should queue */ if (sq->nr_queued[rw]) break;