From nobody Wed Dec 17 14:20:40 2025 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id E5FFAEE49A5 for ; Sun, 20 Aug 2023 15:26:08 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S231449AbjHTP0H (ORCPT ); Sun, 20 Aug 2023 11:26:07 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:45046 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S231407AbjHTPZ7 (ORCPT ); Sun, 20 Aug 2023 11:25:59 -0400 Received: from mail-wm1-x330.google.com (mail-wm1-x330.google.com [IPv6:2a00:1450:4864:20::330]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 112484684 for ; Sun, 20 Aug 2023 08:23:08 -0700 (PDT) Received: by mail-wm1-x330.google.com with SMTP id 5b1f17b1804b1-3feddbb670cso13234645e9.3 for ; Sun, 20 Aug 2023 08:23:07 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=layalina-io.20221208.gappssmtp.com; s=20221208; t=1692544986; x=1693149786; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:from:to:cc:subject:date :message-id:reply-to; bh=SjJzyXCDZRnO5S3i8PaQSpY49m3CqdcltrMAbk+AgrI=; b=H9sJpLmtQexSNnmzR9lDVSDxw7LVQBCygZsvaSp+cGqHTMKeI/TpC00lNhqXVmbpMa 5ed8pfohi1I2zsWjUwxFcS9rvYK0lhAt1MUWJ2sy/PJrA2Slkfl1Pfe4/LpF5evX8win CpwDSQd6RfC8kFl5vkhyb9AEsej7hxtYETrd8qUGQOh0fOSayAf/vYAIW5s8mdi/2i6P SISc1WVE2hm7zbkl5uqSRlLeqYaRkOCoaEFmbgK4r+n/SqXx2lp0oRZLsIkDXIhLVNT1 +GCgWDVifQ0+5XpX2Mogre4B8VEQPq2t6WGs1tIMPJzNjk15hmPE0BiTmon+3DzveZez ip4A== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20221208; t=1692544986; x=1693149786; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:x-gm-message-state:from:to:cc :subject:date:message-id:reply-to; bh=SjJzyXCDZRnO5S3i8PaQSpY49m3CqdcltrMAbk+AgrI=; b=d67gRdcGJlN4HQSliAr17xQ/bcXJS5Dfq5Cg88o/aWPZB/If2j8IGqo+biHlekdypk IAWZHkpznAydR023Uh2POw4M4u8fp28Tz75qDgw3HXr4l6EHIEYdQSalC8oj7VpFFXj4 sr6bHTarNy1maRpeINcspLkV6TQ2QD3bcvCtQospVdkZsZCPCPasF0NHgJkRavncSOqS FpNL2emJ4DGx750gQjuILTH8B4EW7biK0NcizuBhsl/QNwjgW+AVnRSVBPbf9kVi5uiQ 0ZTuZu5hLqMQ3iRiu1iuJfbjB/zFt62ZTkgda9hnWdxOpXCuwkIZkv3S7fkVI0JSIYvz Z+hA== X-Gm-Message-State: AOJu0Yw2JT4NCoeTYwgv3alHdkYlXlGdeLFaH+yVkBW3tTWvsxCA/uj8 KfnEHju2AIPU0KLrUcWk0FgIGA== X-Google-Smtp-Source: AGHT+IGgfmwaMPu6+ziabfVirvrWGn78n4Vx3osiOq1I/zZd75YriD9oasKLB3zsIJfNjonG6rcz2A== X-Received: by 2002:adf:f10c:0:b0:31c:4c68:e73a with SMTP id r12-20020adff10c000000b0031c4c68e73amr356189wro.66.1692544986616; Sun, 20 Aug 2023 08:23:06 -0700 (PDT) Received: from airbuntu.. (host109-151-228-137.range109-151.btcentralplus.com. [109.151.228.137]) by smtp.gmail.com with ESMTPSA id c3-20020adfe703000000b0031773a8e5c4sm9527466wrm.37.2023.08.20.08.23.05 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Sun, 20 Aug 2023 08:23:06 -0700 (PDT) From: Qais Yousef To: stable@vger.kernel.org Cc: Juri Lelli , Waiman Long , Tejun Heo , Dietmar Eggemann , Peter Zijlstra , Vincent Guittot , Ingo Molnar , Hao Luo , John Stultz , cgroups@vger.kernel.org, linux-kernel@vger.kernel.org, Qais Yousef Subject: [PATCH 3/6] sched/cpuset: Keep track of SCHED_DEADLINE task in cpusets Date: Sun, 20 Aug 2023 16:22:55 +0100 Message-Id: <20230820152258.518128-4-qyousef@layalina.io> X-Mailer: git-send-email 2.34.1 In-Reply-To: <20230820152258.518128-1-qyousef@layalina.io> References: <20230820152258.518128-1-qyousef@layalina.io> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Type: text/plain; charset="utf-8" From: Juri Lelli commit 6c24849f5515e4966d94fa5279bdff4acf2e9489 upstream. Qais reported that iterating over all tasks when rebuilding root domains for finding out which ones are DEADLINE and need their bandwidth correctly restored on such root domains can be a costly operation (10+ ms delays on suspend-resume). To fix the problem keep track of the number of DEADLINE tasks belonging to each cpuset and then use this information (followup patch) to only perform the above iteration if DEADLINE tasks are actually present in the cpuset for which a corresponding root domain is being rebuilt. Reported-by: Qais Yousef (Google) Link: https://lore.kernel.org/lkml/20230206221428.2125324-1-qyousef@layalin= a.io/ Signed-off-by: Juri Lelli Reviewed-by: Waiman Long Signed-off-by: Tejun Heo (cherry picked from commit 6c24849f5515e4966d94fa5279bdff4acf2e9489) [Conflict in kernel/cgroup/cpuset.c and kernel/sched/deadline.c due to pulling new code. Reject new code/fields.] Signed-off-by: Qais Yousef (Google) --- include/linux/cpuset.h | 4 ++++ kernel/cgroup/cgroup.c | 4 ++++ kernel/cgroup/cpuset.c | 25 +++++++++++++++++++++++++ kernel/sched/deadline.c | 13 +++++++++++++ 4 files changed, 46 insertions(+) diff --git a/include/linux/cpuset.h b/include/linux/cpuset.h index 5a94f5143924..82fb7e24d1cb 100644 --- a/include/linux/cpuset.h +++ b/include/linux/cpuset.h @@ -56,6 +56,8 @@ extern void cpuset_init_smp(void); extern void cpuset_force_rebuild(void); extern void cpuset_update_active_cpus(void); extern void cpuset_wait_for_hotplug(void); +extern void inc_dl_tasks_cs(struct task_struct *task); +extern void dec_dl_tasks_cs(struct task_struct *task); extern void cpuset_lock(void); extern void cpuset_unlock(void); extern void cpuset_cpus_allowed(struct task_struct *p, struct cpumask *mas= k); @@ -179,6 +181,8 @@ static inline void cpuset_update_active_cpus(void) =20 static inline void cpuset_wait_for_hotplug(void) { } =20 +static inline void inc_dl_tasks_cs(struct task_struct *task) { } +static inline void dec_dl_tasks_cs(struct task_struct *task) { } static inline void cpuset_lock(void) { } static inline void cpuset_unlock(void) { } =20 diff --git a/kernel/cgroup/cgroup.c b/kernel/cgroup/cgroup.c index 6ccdbce17399..be467aea457e 100644 --- a/kernel/cgroup/cgroup.c +++ b/kernel/cgroup/cgroup.c @@ -56,6 +56,7 @@ #include #include #include +#include #include #include =20 @@ -6467,6 +6468,9 @@ void cgroup_exit(struct task_struct *tsk) list_add_tail(&tsk->cg_list, &cset->dying_tasks); cset->nr_tasks--; =20 + if (dl_task(tsk)) + dec_dl_tasks_cs(tsk); + WARN_ON_ONCE(cgroup_task_frozen(tsk)); if (unlikely(!(tsk->flags & PF_KTHREAD) && test_bit(CGRP_FREEZE, &task_dfl_cgroup(tsk)->flags))) diff --git a/kernel/cgroup/cpuset.c b/kernel/cgroup/cpuset.c index 0eff9fb4e6d3..00821ed3d300 100644 --- a/kernel/cgroup/cpuset.c +++ b/kernel/cgroup/cpuset.c @@ -162,6 +162,12 @@ struct cpuset { int use_parent_ecpus; int child_ecpus_count; =20 + /* + * number of SCHED_DEADLINE tasks attached to this cpuset, so that we + * know when to rebuild associated root domain bandwidth information. + */ + int nr_deadline_tasks; + /* Handle for cpuset.cpus.partition */ struct cgroup_file partition_file; }; @@ -209,6 +215,20 @@ static inline struct cpuset *parent_cs(struct cpuset *= cs) return css_cs(cs->css.parent); } =20 +void inc_dl_tasks_cs(struct task_struct *p) +{ + struct cpuset *cs =3D task_cs(p); + + cs->nr_deadline_tasks++; +} + +void dec_dl_tasks_cs(struct task_struct *p) +{ + struct cpuset *cs =3D task_cs(p); + + cs->nr_deadline_tasks--; +} + /* bits in struct cpuset flags field */ typedef enum { CS_ONLINE, @@ -2210,6 +2230,11 @@ static int cpuset_can_attach(struct cgroup_taskset *= tset) ret =3D security_task_setscheduler(task); if (ret) goto out_unlock; + + if (dl_task(task)) { + cs->nr_deadline_tasks++; + cpuset_attach_old_cs->nr_deadline_tasks--; + } } =20 /* diff --git a/kernel/sched/deadline.c b/kernel/sched/deadline.c index b3e206498395..fced55d6e8da 100644 --- a/kernel/sched/deadline.c +++ b/kernel/sched/deadline.c @@ -17,6 +17,7 @@ */ #include "sched.h" #include "pelt.h" +#include =20 struct dl_bandwidth def_dl_bandwidth; =20 @@ -2446,6 +2447,12 @@ static void switched_from_dl(struct rq *rq, struct t= ask_struct *p) if (task_on_rq_queued(p) && p->dl.dl_runtime) task_non_contending(p); =20 + /* + * In case a task is setscheduled out from SCHED_DEADLINE we need to + * keep track of that on its cpuset (for correct bandwidth tracking). + */ + dec_dl_tasks_cs(p); + if (!task_on_rq_queued(p)) { /* * Inactive timer is armed. However, p is leaving DEADLINE and @@ -2486,6 +2493,12 @@ static void switched_to_dl(struct rq *rq, struct tas= k_struct *p) if (hrtimer_try_to_cancel(&p->dl.inactive_timer) =3D=3D 1) put_task_struct(p); =20 + /* + * In case a task is setscheduled to SCHED_DEADLINE we need to keep + * track of that on its cpuset (for correct bandwidth tracking). + */ + inc_dl_tasks_cs(p); + /* If p is not queued we will update its parameters at next wakeup. */ if (!task_on_rq_queued(p)) { add_rq_bw(&p->dl, &rq->dl); --=20 2.34.1