From patchwork Wed Jan 23 15:26:52 2019 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Vincent Guittot X-Patchwork-Id: 156382 Delivered-To: patch@linaro.org Received: by 2002:a02:48:0:0:0:0:0 with SMTP id 69csp841408jaa; Wed, 23 Jan 2019 07:27:10 -0800 (PST) X-Google-Smtp-Source: ALg8bN5UCsqF6Gab4F1C/kTVzv/I/RZZzujQggIdbBn27E/xtWUQSmig2vhIFin1vq9T7GXUHfqp X-Received: by 2002:a17:902:aa4c:: with SMTP id c12mr2629146plr.48.1548257230622; Wed, 23 Jan 2019 07:27:10 -0800 (PST) ARC-Seal: i=1; a=rsa-sha256; t=1548257230; cv=none; d=google.com; s=arc-20160816; b=P3daxxZuxH8h1eFLjlPMz0fO/UdtNNeR3AMqSgt53fZToAinb18bzakv3X8AmRTbAI F2OQqOwTZGFe+2xCrP622+n9kOipbg40SwqiEB71AdlXLzqHZ3X+c2lWW5ustwvxhd/h nhneLkxKEvwRpN1jd4LUQZ2IYDdxmrkxedAE9rKfplI1/K5sBrM1H1Rsw5vWNakq6215 sIt3zwBwmLd1a3eS7M4St8gHhx7f4m31dD1N49QXO4F1Z90Wfe3e/gfLQAyurp0rER3/ 8A4iTg9T+NcK3D/SfF8Ki/aiE4IHDxrk6p4uUN8S+xW18cP2gsQ0iwc6htf6aHgCYzDA djyQ== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=arc-20160816; h=list-id:precedence:sender:references:in-reply-to:message-id:date :subject:cc:to:from:dkim-signature; bh=hReVLUTWGkU8mA22pHwBBkB0pAs98jcygffTpDVc1AM=; b=QoJFIfFFSI5YjB9N+9JwtRqSm5XJhfNNUjrHCBEP43Y7mU16DhHuQ1Z1PY31+y+b1t d+mEZSQvZpZ0CqNBgFdw5K695Pdi7vs8uofIu5D5PpgZuF+Xg8AL/iXm1UD9Yw2SUmjf qr968Mfs3X/zYAI0xSuzgKSSvNYkXF/33jv33jPN/7gRlopONKK/SRk+MInadWqPrvV8 70qnEUE0rVggArdRMqR2o2oST+F63IMIZml6f5c+0FsELlidiNKsz/H9MgiH2+ewUtz9 hdgsLmfXo8qje06uD52EdobLpcWMkvjDmWQlwf/1ZiI+sKNOT00UBVU45y0IKI9HCm6C X15w== ARC-Authentication-Results: i=1; mx.google.com; dkim=pass header.i=@linaro.org header.s=google header.b=H5xZtLXo; spf=pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) smtp.mailfrom=linux-kernel-owner@vger.kernel.org; dmarc=pass (p=NONE sp=NONE dis=NONE) header.from=linaro.org Return-Path: Received: from vger.kernel.org (vger.kernel.org. [209.132.180.67]) by mx.google.com with ESMTP id c7si18827418plz.118.2019.01.23.07.27.10; Wed, 23 Jan 2019 07:27:10 -0800 (PST) Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67; Authentication-Results: mx.google.com; dkim=pass header.i=@linaro.org header.s=google header.b=H5xZtLXo; spf=pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) smtp.mailfrom=linux-kernel-owner@vger.kernel.org; dmarc=pass (p=NONE sp=NONE dis=NONE) header.from=linaro.org Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1726243AbfAWP1I (ORCPT + 31 others); Wed, 23 Jan 2019 10:27:08 -0500 Received: from mail-wm1-f66.google.com ([209.85.128.66]:56260 "EHLO mail-wm1-f66.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1725862AbfAWP1F (ORCPT ); Wed, 23 Jan 2019 10:27:05 -0500 Received: by mail-wm1-f66.google.com with SMTP id y139so2364114wmc.5 for ; Wed, 23 Jan 2019 07:27:03 -0800 (PST) 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=hReVLUTWGkU8mA22pHwBBkB0pAs98jcygffTpDVc1AM=; b=H5xZtLXokiLacrs6wDPekNuZfdOP9oqcYTVEw8J0HQAGoGxxyrgpFxpldRIbRdApTj cIKDzK2pmHR4AQtkSNmSDC2SHei2MUZBMt97zvXp9goIwxMofKBwRrvsYJdv2bNOWAXW RHtWGK+EFbzelUcb/mu78bsLyRNodM3lUHlRc= 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=hReVLUTWGkU8mA22pHwBBkB0pAs98jcygffTpDVc1AM=; b=SgOqq+cXzcgEcWdtINcXwDl8yyR6pWdGIPbQU5fb9Kb6BzSYjL0u6fsLm4jDoipc5/ tikX1u+ZB3AugVKeaf7iLx0TF5g/N4asaFX/7lDxyKZF563EcH48jiuUettaaX8VWZh4 2QrSFSPcz0PXxUJxz01vqjky610ST9Fr+vyNU0rk5Dy587r2fVvUUsrPaBHyqeus3XvU XUnu2WAhjZGek6k0CweQrYfT8CPdly0h7RizL/qVQnfAu/OrOrUzAN4IzhIy0wG3LYcd 5moGW6/5fy3OmasPkVNN++w41Hw+kct3N341UbWsbaXgAHThZqxUB+jpCls7ZLfjOb1L FvVg== X-Gm-Message-State: AJcUukd3zBvWxPDmOY6MvI74PONkHBBRlFHYti0TIzrSi6KYFtRGrNZj A0myrtsyWjpVEYgeKnn0uS+9Uw== X-Received: by 2002:a1c:7d06:: with SMTP id y6mr3086261wmc.7.1548257223075; Wed, 23 Jan 2019 07:27:03 -0800 (PST) Received: from localhost.localdomain ([2a01:e0a:f:6020:5ca6:2921:648b:56b6]) by smtp.gmail.com with ESMTPSA id b13sm92938536wrn.28.2019.01.23.07.27.01 (version=TLS1_2 cipher=ECDHE-RSA-AES128-SHA bits=128/128); Wed, 23 Jan 2019 07:27:01 -0800 (PST) From: Vincent Guittot To: peterz@infradead.org, mingo@kernel.org, linux-kernel@vger.kernel.org Cc: rjw@rjwysocki.net, dietmar.eggemann@arm.com, Morten.Rasmussen@arm.com, patrick.bellasi@arm.com, pjt@google.com, bsegall@google.com, thara.gopinath@linaro.org, pkondeti@codeaurora.org, quentin.perret@arm.com, srinivas.pandruvada@linux.intel.com, Vincent Guittot Subject: [PATCH v9 1/3] sched/fair: move rq_of helper function Date: Wed, 23 Jan 2019 16:26:52 +0100 Message-Id: <1548257214-13745-2-git-send-email-vincent.guittot@linaro.org> X-Mailer: git-send-email 2.7.4 In-Reply-To: <1548257214-13745-1-git-send-email-vincent.guittot@linaro.org> References: <1548257214-13745-1-git-send-email-vincent.guittot@linaro.org> Sender: linux-kernel-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Move rq_of() helper function so it can be used in pelt.c Signed-off-by: Vincent Guittot --- kernel/sched/fair.c | 13 ------------- kernel/sched/sched.h | 16 ++++++++++++++++ 2 files changed, 16 insertions(+), 13 deletions(-) -- 2.7.4 diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index e2ff4b6..cdddc86 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -248,13 +248,6 @@ const struct sched_class fair_sched_class; */ #ifdef CONFIG_FAIR_GROUP_SCHED - -/* cpu runqueue to which this cfs_rq is attached */ -static inline struct rq *rq_of(struct cfs_rq *cfs_rq) -{ - return cfs_rq->rq; -} - static inline struct task_struct *task_of(struct sched_entity *se) { SCHED_WARN_ON(!entity_is_task(se)); @@ -410,12 +403,6 @@ static inline struct task_struct *task_of(struct sched_entity *se) return container_of(se, struct task_struct, se); } -static inline struct rq *rq_of(struct cfs_rq *cfs_rq) -{ - return container_of(cfs_rq, struct rq, cfs); -} - - #define for_each_sched_entity(se) \ for (; se; se = NULL) diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index d27c1a5..d57a590 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -951,6 +951,22 @@ struct rq { #endif }; +#ifdef CONFIG_FAIR_GROUP_SCHED + +/* cpu runqueue to which this cfs_rq is attached */ +static inline struct rq *rq_of(struct cfs_rq *cfs_rq) +{ + return cfs_rq->rq; +} + +#else /* !CONFIG_FAIR_GROUP_SCHED */ + +static inline struct rq *rq_of(struct cfs_rq *cfs_rq) +{ + return container_of(cfs_rq, struct rq, cfs); +} +#endif + static inline int cpu_of(struct rq *rq) { #ifdef CONFIG_SMP From patchwork Wed Jan 23 15:26:53 2019 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Vincent Guittot X-Patchwork-Id: 156384 Delivered-To: patch@linaro.org Received: by 2002:a02:48:0:0:0:0:0 with SMTP id 69csp841572jaa; Wed, 23 Jan 2019 07:27:20 -0800 (PST) X-Google-Smtp-Source: ALg8bN4GPvvxX3cW8eX81tadhVq2Rk723fJPE05xlFXkSUGqULrttHtRvKdqF39TvWCMjocc9oSv X-Received: by 2002:a63:f844:: with SMTP id v4mr2325462pgj.82.1548257240038; Wed, 23 Jan 2019 07:27:20 -0800 (PST) ARC-Seal: i=1; a=rsa-sha256; t=1548257240; cv=none; d=google.com; s=arc-20160816; b=k8HyjlxQbH7nX5SUnX0TgumqttHNw2r0FZuqpWt4lmHusBdGfC4cx+97FKinowtT/M FSZaYdmlVnmnYSUPD0F+QsJs+kym/e+raQ6qQyySsvH2/O7NcqnnM8TeoNJZIrY+bt9E +t0ZZimX8G4WOmVP/pIT6Db9UzgnDN20/5BUEBpROtp1noFFM6OA5cl15kVvaoYsY1yB hKHk0juQ/k+RuCw7e5noux4k/JNFDXyS5gcLnPbNhcWeHcpRs3i0GTxD4zvxTFBJoXNG zDlRuZLJUTOAigklzPTjzHmcR8ZIfr4AxngRkzQniQ0QWk3TuKACLLP76OjP9Tt1tDUb OKtA== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=arc-20160816; h=list-id:precedence:sender:references:in-reply-to:message-id:date :subject:cc:to:from:dkim-signature; bh=TdRZtevLFO+WBJ0JFEfXKV8TxxLNpHdLKk8ymk7oQYY=; b=aDqjABk02ZKuaqJfSMxlW4Fqr43nK+Ci+H+ouNRr6UcxUgn9kLbIIuyuF25r4Tus05 sMbSc+3XoRJNKIQUKQQqFsqDSI/zRCoV34MdxTHGZw8sPr+U2ti2843E0svnP9jDwEV2 BS5tOBrC0/Wz8dFYG6H30J4ZuIKK1EPyiJNhHt7kg7Km5QgaYfKNmJHwV7ei+uS8MIJW o2g3UpguawCAPOxAHWCtL46AOP0zL+tiFBsIUlIm/ZLIjriFAtwyHO+Z/2FeRw8k3b/x 83wZ36T0Iezt+MLaindcwJi3Bv1OJfSPbbRZYTy0h10mqrXEk69gaiid7MnD0YQOIqgK GQvA== ARC-Authentication-Results: i=1; mx.google.com; dkim=pass header.i=@linaro.org header.s=google header.b=JJHFpfVr; spf=pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) smtp.mailfrom=linux-kernel-owner@vger.kernel.org; dmarc=pass (p=NONE sp=NONE dis=NONE) header.from=linaro.org Return-Path: Received: from vger.kernel.org (vger.kernel.org. [209.132.180.67]) by mx.google.com with ESMTP id 59si19429468ple.291.2019.01.23.07.27.19; Wed, 23 Jan 2019 07:27:20 -0800 (PST) Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67; Authentication-Results: mx.google.com; dkim=pass header.i=@linaro.org header.s=google header.b=JJHFpfVr; spf=pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) smtp.mailfrom=linux-kernel-owner@vger.kernel.org; dmarc=pass (p=NONE sp=NONE dis=NONE) header.from=linaro.org Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1726337AbfAWP1R (ORCPT + 31 others); Wed, 23 Jan 2019 10:27:17 -0500 Received: from mail-wr1-f65.google.com ([209.85.221.65]:33346 "EHLO mail-wr1-f65.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1725966AbfAWP1H (ORCPT ); Wed, 23 Jan 2019 10:27:07 -0500 Received: by mail-wr1-f65.google.com with SMTP id p7so2987363wru.0 for ; Wed, 23 Jan 2019 07:27:05 -0800 (PST) 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=TdRZtevLFO+WBJ0JFEfXKV8TxxLNpHdLKk8ymk7oQYY=; b=JJHFpfVr6uCAxARSygBxES7itx0cDkJR+k4rxOU40s8V93obusIGM2uiikS04QcQco dpRb8VYEYJYkcKObHWRMLO2Nuy9VwgmD0SKELJGnFDnemdd/x67pT+pChtCLJ662dbN8 vF5xOmFv5jbU6NqIOMDI6zRFP8GSLPSjxe/kY= 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=TdRZtevLFO+WBJ0JFEfXKV8TxxLNpHdLKk8ymk7oQYY=; b=kJHzzTJaduTXskUmkOUxzVAIMnOaNZLO1noFMbhYqjjwpHZ8A7asIn58sCfYSTEMz1 5tR5Ksl59ScHrKTsqfWNO2jd0WJf6WBBsC7z1ZT8CExgPd9mXy1KEoX3oX9qEC4rmpkd psCgRgaW7ic4koc3Z+3d4/qESxlnTIH6Z9+aCdoM9E0venhwD/THUrMQQfK3VVpDTIBG jPlojmEDyn5a5p5z3pBaHKUFJ1ISgqfYEF8+hGipFqbUJn456Cvl6FrkO1P+T7QV3FaK biDlIT+RwhYa9akvHoVGmhw27E7oyKO6uZTzAvkBry9+2DlwkrE8GzqNYol5+NfiLrMX kcPg== X-Gm-Message-State: AJcUukd8lDccJRiU5883D9ftUrX+e0/NJAI8lzrWuZsMAU/YKfMsQn4l CdaMDdvv1/cUfkooclg6LypS9g== X-Received: by 2002:a05:6000:1144:: with SMTP id d4mr3201058wrx.136.1548257224485; Wed, 23 Jan 2019 07:27:04 -0800 (PST) Received: from localhost.localdomain ([2a01:e0a:f:6020:5ca6:2921:648b:56b6]) by smtp.gmail.com with ESMTPSA id b13sm92938536wrn.28.2019.01.23.07.27.03 (version=TLS1_2 cipher=ECDHE-RSA-AES128-SHA bits=128/128); Wed, 23 Jan 2019 07:27:03 -0800 (PST) From: Vincent Guittot To: peterz@infradead.org, mingo@kernel.org, linux-kernel@vger.kernel.org Cc: rjw@rjwysocki.net, dietmar.eggemann@arm.com, Morten.Rasmussen@arm.com, patrick.bellasi@arm.com, pjt@google.com, bsegall@google.com, thara.gopinath@linaro.org, pkondeti@codeaurora.org, quentin.perret@arm.com, srinivas.pandruvada@linux.intel.com, Vincent Guittot Subject: [PATCH v9 2/3] sched/fair: update scale invariance of PELT Date: Wed, 23 Jan 2019 16:26:53 +0100 Message-Id: <1548257214-13745-3-git-send-email-vincent.guittot@linaro.org> X-Mailer: git-send-email 2.7.4 In-Reply-To: <1548257214-13745-1-git-send-email-vincent.guittot@linaro.org> References: <1548257214-13745-1-git-send-email-vincent.guittot@linaro.org> Sender: linux-kernel-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org The current implementation of load tracking invariance scales the contribution with current frequency and uarch performance (only for utilization) of the CPU. One main result of this formula is that the figures are capped by current capacity of CPU. Another one is that the load_avg is not invariant because not scaled with uarch. The util_avg of a periodic task that runs r time slots every p time slots varies in the range : U * (1-y^r)/(1-y^p) * y^i < Utilization < U * (1-y^r)/(1-y^p) with U is the max util_avg value = SCHED_CAPACITY_SCALE At a lower capacity, the range becomes: U * C * (1-y^r')/(1-y^p) * y^i' < Utilization < U * C * (1-y^r')/(1-y^p) with C reflecting the compute capacity ratio between current capacity and max capacity. so C tries to compensate changes in (1-y^r') but it can't be accurate. Instead of scaling the contribution value of PELT algo, we should scale the running time. The PELT signal aims to track the amount of computation of tasks and/or rq so it seems more correct to scale the running time to reflect the effective amount of computation done since the last update. In order to be fully invariant, we need to apply the same amount of running time and idle time whatever the current capacity. Because running at lower capacity implies that the task will run longer, we have to ensure that the same amount of idle time will be applied when system becomes idle and no idle time has been "stolen". But reaching the maximum utilization value (SCHED_CAPACITY_SCALE) means that the task is seen as an always-running task whatever the capacity of the CPU (even at max compute capacity). In this case, we can discard this "stolen" idle times which becomes meaningless. In order to achieve this time scaling, a new clock_pelt is created per rq. The increase of this clock scales with current capacity when something is running on rq and synchronizes with clock_task when rq is idle. With this mechanism, we ensure the same running and idle time whatever the current capacity. This also enables to simplify the pelt algorithm by removing all references of uarch and frequency and applying the same contribution to utilization and loads. Furthermore, the scaling is done only once per update of clock (update_rq_clock_task()) instead of during each update of sched_entities and cfs/rt/dl_rq of the rq like the current implementation. This is interesting when cgroup are involved as shown in the results below: On a hikey (octo Arm64 platform). Performance cpufreq governor and only shallowest c-state to remove variance generated by those power features so we only track the impact of pelt algo. each test runs 16 times ./perf bench sched pipe (higher is better) kernel tip/sched/core + patch ops/seconds ops/seconds diff cgroup root 59652(+/- 0.18%) 59876(+/- 0.24%) +0.38% level1 55608(+/- 0.27%) 55923(+/- 0.24%) +0.57% level2 52115(+/- 0.29%) 52564(+/- 0.22%) +0.86% hackbench -l 1000 (lower is better) kernel tip/sched/core + patch duration(sec) duration(sec) diff cgroup root 4.453(+/- 2.37%) 4.383(+/- 2.88%) -1.57% level1 4.859(+/- 8.50%) 4.830(+/- 7.07%) -0.60% level2 5.063(+/- 9.83%) 4.928(+/- 9.66%) -2.66% Then, the responsiveness of PELT is improved when CPU is not running at max capacity with this new algorithm. I have put below some examples of duration to reach some typical load values according to the capacity of the CPU with current implementation and with this patch. These values has been computed based on the geometric series and the half period value: Util (%) max capacity half capacity(mainline) half capacity(w/ patch) 972 (95%) 138ms not reachable 276ms 486 (47.5%) 30ms 138ms 60ms 256 (25%) 13ms 32ms 26ms On my hikey (octo Arm64 platform) with schedutil governor, the time to reach max OPP when starting from a null utilization, decreases from 223ms with current scale invariance down to 121ms with the new algorithm. Signed-off-by: Vincent Guittot --- include/linux/sched.h | 23 +++------- kernel/sched/core.c | 1 + kernel/sched/deadline.c | 6 +-- kernel/sched/fair.c | 45 ++++++++++--------- kernel/sched/pelt.c | 45 ++++++++++--------- kernel/sched/pelt.h | 114 ++++++++++++++++++++++++++++++++++++++++++++++-- kernel/sched/rt.c | 6 +-- kernel/sched/sched.h | 5 ++- 8 files changed, 177 insertions(+), 68 deletions(-) -- 2.7.4 diff --git a/include/linux/sched.h b/include/linux/sched.h index d2f90fa..44eb8a7 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -356,12 +356,6 @@ struct util_est { * For cfs_rq, it is the aggregated load_avg of all runnable and * blocked sched_entities. * - * load_avg may also take frequency scaling into account: - * - * load_avg = runnable% * scale_load_down(load) * freq% - * - * where freq% is the CPU frequency normalized to the highest frequency. - * * [util_avg definition] * * util_avg = running% * SCHED_CAPACITY_SCALE @@ -370,17 +364,14 @@ struct util_est { * a CPU. For cfs_rq, it is the aggregated util_avg of all runnable * and blocked sched_entities. * - * util_avg may also factor frequency scaling and CPU capacity scaling: - * - * util_avg = running% * SCHED_CAPACITY_SCALE * freq% * capacity% - * - * where freq% is the same as above, and capacity% is the CPU capacity - * normalized to the greatest capacity (due to uarch differences, etc). + * load_avg and util_avg don't direcly factor frequency scaling and CPU + * capacity scaling. The scaling is done through the rq_clock_pelt that + * is used for computing those signals (see update_rq_clock_pelt()) * - * N.B., the above ratios (runnable%, running%, freq%, and capacity%) - * themselves are in the range of [0, 1]. To do fixed point arithmetics, - * we therefore scale them to as large a range as necessary. This is for - * example reflected by util_avg's SCHED_CAPACITY_SCALE. + * N.B., the above ratios (runnable% and running%) themselves are in the + * range of [0, 1]. To do fixed point arithmetics, we therefore scale them + * to as large a range as necessary. This is for example reflected by + * util_avg's SCHED_CAPACITY_SCALE. * * [Overflow issue] * diff --git a/kernel/sched/core.c b/kernel/sched/core.c index d4d3514..205cf2a 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -180,6 +180,7 @@ static void update_rq_clock_task(struct rq *rq, s64 delta) if ((irq_delta + steal) && sched_feat(NONTASK_CAPACITY)) update_irq_load_avg(rq, irq_delta + steal); #endif + update_rq_clock_pelt(rq, delta); } void update_rq_clock(struct rq *rq) diff --git a/kernel/sched/deadline.c b/kernel/sched/deadline.c index fb8b7b5..6a73e41 100644 --- a/kernel/sched/deadline.c +++ b/kernel/sched/deadline.c @@ -1767,7 +1767,7 @@ pick_next_task_dl(struct rq *rq, struct task_struct *prev, struct rq_flags *rf) deadline_queue_push_tasks(rq); if (rq->curr->sched_class != &dl_sched_class) - update_dl_rq_load_avg(rq_clock_task(rq), rq, 0); + update_dl_rq_load_avg(rq_clock_pelt(rq), rq, 0); return p; } @@ -1776,7 +1776,7 @@ static void put_prev_task_dl(struct rq *rq, struct task_struct *p) { update_curr_dl(rq); - update_dl_rq_load_avg(rq_clock_task(rq), rq, 1); + update_dl_rq_load_avg(rq_clock_pelt(rq), rq, 1); if (on_dl_rq(&p->dl) && p->nr_cpus_allowed > 1) enqueue_pushable_dl_task(rq, p); } @@ -1793,7 +1793,7 @@ static void task_tick_dl(struct rq *rq, struct task_struct *p, int queued) { update_curr_dl(rq); - update_dl_rq_load_avg(rq_clock_task(rq), rq, 1); + update_dl_rq_load_avg(rq_clock_pelt(rq), rq, 1); /* * Even when we have runtime, update_curr_dl() might have resulted in us * not being the leftmost task anymore. In that case NEED_RESCHED will diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index cdddc86..a9cee4e 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -673,9 +673,8 @@ static u64 sched_vslice(struct cfs_rq *cfs_rq, struct sched_entity *se) return calc_delta_fair(sched_slice(cfs_rq, se), se); } -#ifdef CONFIG_SMP #include "pelt.h" -#include "sched-pelt.h" +#ifdef CONFIG_SMP static int select_idle_sibling(struct task_struct *p, int prev_cpu, int cpu); static unsigned long task_h_load(struct task_struct *p); @@ -763,7 +762,7 @@ void post_init_entity_util_avg(struct sched_entity *se) * such that the next switched_to_fair() has the * expected state. */ - se->avg.last_update_time = cfs_rq_clock_task(cfs_rq); + se->avg.last_update_time = cfs_rq_clock_pelt(cfs_rq); return; } } @@ -3109,7 +3108,7 @@ void set_task_rq_fair(struct sched_entity *se, p_last_update_time = prev->avg.last_update_time; n_last_update_time = next->avg.last_update_time; #endif - __update_load_avg_blocked_se(p_last_update_time, cpu_of(rq_of(prev)), se); + __update_load_avg_blocked_se(p_last_update_time, se); se->avg.last_update_time = n_last_update_time; } @@ -3244,11 +3243,11 @@ update_tg_cfs_runnable(struct cfs_rq *cfs_rq, struct sched_entity *se, struct cf /* * runnable_sum can't be lower than running_sum - * As running sum is scale with CPU capacity wehreas the runnable sum - * is not we rescale running_sum 1st + * Rescale running sum to be in the same range as runnable sum + * running_sum is in [0 : LOAD_AVG_MAX << SCHED_CAPACITY_SHIFT] + * runnable_sum is in [0 : LOAD_AVG_MAX] */ - running_sum = se->avg.util_sum / - arch_scale_cpu_capacity(NULL, cpu_of(rq_of(cfs_rq))); + running_sum = se->avg.util_sum >> SCHED_CAPACITY_SHIFT; runnable_sum = max(runnable_sum, running_sum); load_sum = (s64)se_weight(se) * runnable_sum; @@ -3351,7 +3350,7 @@ static inline void add_tg_cfs_propagate(struct cfs_rq *cfs_rq, long runnable_sum /** * update_cfs_rq_load_avg - update the cfs_rq's load/util averages - * @now: current time, as per cfs_rq_clock_task() + * @now: current time, as per cfs_rq_clock_pelt() * @cfs_rq: cfs_rq to update * * The cfs_rq avg is the direct sum of all its entities (blocked and runnable) @@ -3396,7 +3395,7 @@ update_cfs_rq_load_avg(u64 now, struct cfs_rq *cfs_rq) decayed = 1; } - decayed |= __update_load_avg_cfs_rq(now, cpu_of(rq_of(cfs_rq)), cfs_rq); + decayed |= __update_load_avg_cfs_rq(now, cfs_rq); #ifndef CONFIG_64BIT smp_wmb(); @@ -3486,9 +3485,7 @@ static void detach_entity_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *s /* Update task and its cfs_rq load average */ static inline void update_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) { - u64 now = cfs_rq_clock_task(cfs_rq); - struct rq *rq = rq_of(cfs_rq); - int cpu = cpu_of(rq); + u64 now = cfs_rq_clock_pelt(cfs_rq); int decayed; /* @@ -3496,7 +3493,7 @@ static inline void update_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *s * track group sched_entity load average for task_h_load calc in migration */ if (se->avg.last_update_time && !(flags & SKIP_AGE_LOAD)) - __update_load_avg_se(now, cpu, cfs_rq, se); + __update_load_avg_se(now, cfs_rq, se); decayed = update_cfs_rq_load_avg(now, cfs_rq); decayed |= propagate_entity_load_avg(se); @@ -3548,7 +3545,7 @@ void sync_entity_load_avg(struct sched_entity *se) u64 last_update_time; last_update_time = cfs_rq_last_update_time(cfs_rq); - __update_load_avg_blocked_se(last_update_time, cpu_of(rq_of(cfs_rq)), se); + __update_load_avg_blocked_se(last_update_time, se); } /* @@ -7015,6 +7012,12 @@ done: __maybe_unused; if (new_tasks > 0) goto again; + /* + * rq is about to be idle, check if we need to update the + * lost_idle_time of clock_pelt + */ + update_idle_rq_clock_pelt(rq); + return NULL; } @@ -7657,7 +7660,7 @@ static void update_blocked_averages(int cpu) if (throttled_hierarchy(cfs_rq)) continue; - if (update_cfs_rq_load_avg(cfs_rq_clock_task(cfs_rq), cfs_rq)) + if (update_cfs_rq_load_avg(cfs_rq_clock_pelt(cfs_rq), cfs_rq)) update_tg_load_avg(cfs_rq, 0); /* Propagate pending load changes to the parent, if any: */ @@ -7671,8 +7674,8 @@ static void update_blocked_averages(int cpu) } curr_class = rq->curr->sched_class; - update_rt_rq_load_avg(rq_clock_task(rq), rq, curr_class == &rt_sched_class); - update_dl_rq_load_avg(rq_clock_task(rq), rq, curr_class == &dl_sched_class); + update_rt_rq_load_avg(rq_clock_pelt(rq), rq, curr_class == &rt_sched_class); + update_dl_rq_load_avg(rq_clock_pelt(rq), rq, curr_class == &dl_sched_class); update_irq_load_avg(rq, 0); /* Don't need periodic decay once load/util_avg are null */ if (others_have_blocked(rq)) @@ -7742,11 +7745,11 @@ static inline void update_blocked_averages(int cpu) rq_lock_irqsave(rq, &rf); update_rq_clock(rq); - update_cfs_rq_load_avg(cfs_rq_clock_task(cfs_rq), cfs_rq); + update_cfs_rq_load_avg(cfs_rq_clock_pelt(cfs_rq), cfs_rq); curr_class = rq->curr->sched_class; - update_rt_rq_load_avg(rq_clock_task(rq), rq, curr_class == &rt_sched_class); - update_dl_rq_load_avg(rq_clock_task(rq), rq, curr_class == &dl_sched_class); + update_rt_rq_load_avg(rq_clock_pelt(rq), rq, curr_class == &rt_sched_class); + update_dl_rq_load_avg(rq_clock_pelt(rq), rq, curr_class == &dl_sched_class); update_irq_load_avg(rq, 0); #ifdef CONFIG_NO_HZ_COMMON rq->last_blocked_load_update_tick = jiffies; diff --git a/kernel/sched/pelt.c b/kernel/sched/pelt.c index 90fb5bc..befce29 100644 --- a/kernel/sched/pelt.c +++ b/kernel/sched/pelt.c @@ -26,7 +26,6 @@ #include #include "sched.h" -#include "sched-pelt.h" #include "pelt.h" /* @@ -106,16 +105,12 @@ static u32 __accumulate_pelt_segments(u64 periods, u32 d1, u32 d3) * n=1 */ static __always_inline u32 -accumulate_sum(u64 delta, int cpu, struct sched_avg *sa, +accumulate_sum(u64 delta, struct sched_avg *sa, unsigned long load, unsigned long runnable, int running) { - unsigned long scale_freq, scale_cpu; u32 contrib = (u32)delta; /* p == 0 -> delta < 1024 */ u64 periods; - scale_freq = arch_scale_freq_capacity(cpu); - scale_cpu = arch_scale_cpu_capacity(NULL, cpu); - delta += sa->period_contrib; periods = delta / 1024; /* A period is 1024us (~1ms) */ @@ -137,13 +132,12 @@ accumulate_sum(u64 delta, int cpu, struct sched_avg *sa, } sa->period_contrib = delta; - contrib = cap_scale(contrib, scale_freq); if (load) sa->load_sum += load * contrib; if (runnable) sa->runnable_load_sum += runnable * contrib; if (running) - sa->util_sum += contrib * scale_cpu; + sa->util_sum += contrib << SCHED_CAPACITY_SHIFT; return periods; } @@ -177,7 +171,7 @@ accumulate_sum(u64 delta, int cpu, struct sched_avg *sa, * = u_0 + u_1*y + u_2*y^2 + ... [re-labeling u_i --> u_{i+1}] */ static __always_inline int -___update_load_sum(u64 now, int cpu, struct sched_avg *sa, +___update_load_sum(u64 now, struct sched_avg *sa, unsigned long load, unsigned long runnable, int running) { u64 delta; @@ -221,7 +215,7 @@ ___update_load_sum(u64 now, int cpu, struct sched_avg *sa, * Step 1: accumulate *_sum since last_update_time. If we haven't * crossed period boundaries, finish. */ - if (!accumulate_sum(delta, cpu, sa, load, runnable, running)) + if (!accumulate_sum(delta, sa, load, runnable, running)) return 0; return 1; @@ -267,9 +261,9 @@ ___update_load_avg(struct sched_avg *sa, unsigned long load, unsigned long runna * runnable_load_avg = \Sum se->avg.runable_load_avg */ -int __update_load_avg_blocked_se(u64 now, int cpu, struct sched_entity *se) +int __update_load_avg_blocked_se(u64 now, struct sched_entity *se) { - if (___update_load_sum(now, cpu, &se->avg, 0, 0, 0)) { + if (___update_load_sum(now, &se->avg, 0, 0, 0)) { ___update_load_avg(&se->avg, se_weight(se), se_runnable(se)); return 1; } @@ -277,9 +271,9 @@ int __update_load_avg_blocked_se(u64 now, int cpu, struct sched_entity *se) return 0; } -int __update_load_avg_se(u64 now, int cpu, struct cfs_rq *cfs_rq, struct sched_entity *se) +int __update_load_avg_se(u64 now, struct cfs_rq *cfs_rq, struct sched_entity *se) { - if (___update_load_sum(now, cpu, &se->avg, !!se->on_rq, !!se->on_rq, + if (___update_load_sum(now, &se->avg, !!se->on_rq, !!se->on_rq, cfs_rq->curr == se)) { ___update_load_avg(&se->avg, se_weight(se), se_runnable(se)); @@ -290,9 +284,9 @@ int __update_load_avg_se(u64 now, int cpu, struct cfs_rq *cfs_rq, struct sched_e return 0; } -int __update_load_avg_cfs_rq(u64 now, int cpu, struct cfs_rq *cfs_rq) +int __update_load_avg_cfs_rq(u64 now, struct cfs_rq *cfs_rq) { - if (___update_load_sum(now, cpu, &cfs_rq->avg, + if (___update_load_sum(now, &cfs_rq->avg, scale_load_down(cfs_rq->load.weight), scale_load_down(cfs_rq->runnable_weight), cfs_rq->curr != NULL)) { @@ -317,7 +311,7 @@ int __update_load_avg_cfs_rq(u64 now, int cpu, struct cfs_rq *cfs_rq) int update_rt_rq_load_avg(u64 now, struct rq *rq, int running) { - if (___update_load_sum(now, rq->cpu, &rq->avg_rt, + if (___update_load_sum(now, &rq->avg_rt, running, running, running)) { @@ -340,7 +334,7 @@ int update_rt_rq_load_avg(u64 now, struct rq *rq, int running) int update_dl_rq_load_avg(u64 now, struct rq *rq, int running) { - if (___update_load_sum(now, rq->cpu, &rq->avg_dl, + if (___update_load_sum(now, &rq->avg_dl, running, running, running)) { @@ -365,22 +359,31 @@ int update_dl_rq_load_avg(u64 now, struct rq *rq, int running) int update_irq_load_avg(struct rq *rq, u64 running) { int ret = 0; + + /* + * We can't use clock_pelt because irq time is not accounted in + * clock_task. Instead we directly scale the running time to + * reflect the real amount of computation + */ + running = cap_scale(running, arch_scale_freq_capacity(cpu_of(rq))); + running = cap_scale(running, arch_scale_cpu_capacity(NULL, cpu_of(rq))); + /* * We know the time that has been used by interrupt since last update * but we don't when. Let be pessimistic and assume that interrupt has * happened just before the update. This is not so far from reality * because interrupt will most probably wake up task and trig an update - * of rq clock during which the metric si updated. + * of rq clock during which the metric is updated. * We start to decay with normal context time and then we add the * interrupt context time. * We can safely remove running from rq->clock because * rq->clock += delta with delta >= running */ - ret = ___update_load_sum(rq->clock - running, rq->cpu, &rq->avg_irq, + ret = ___update_load_sum(rq->clock - running, &rq->avg_irq, 0, 0, 0); - ret += ___update_load_sum(rq->clock, rq->cpu, &rq->avg_irq, + ret += ___update_load_sum(rq->clock, &rq->avg_irq, 1, 1, 1); diff --git a/kernel/sched/pelt.h b/kernel/sched/pelt.h index 7e56b48..7489d5f 100644 --- a/kernel/sched/pelt.h +++ b/kernel/sched/pelt.h @@ -1,8 +1,9 @@ #ifdef CONFIG_SMP +#include "sched-pelt.h" -int __update_load_avg_blocked_se(u64 now, int cpu, struct sched_entity *se); -int __update_load_avg_se(u64 now, int cpu, struct cfs_rq *cfs_rq, struct sched_entity *se); -int __update_load_avg_cfs_rq(u64 now, int cpu, struct cfs_rq *cfs_rq); +int __update_load_avg_blocked_se(u64 now, struct sched_entity *se); +int __update_load_avg_se(u64 now, struct cfs_rq *cfs_rq, struct sched_entity *se); +int __update_load_avg_cfs_rq(u64 now, struct cfs_rq *cfs_rq); int update_rt_rq_load_avg(u64 now, struct rq *rq, int running); int update_dl_rq_load_avg(u64 now, struct rq *rq, int running); @@ -42,6 +43,101 @@ static inline void cfs_se_util_change(struct sched_avg *avg) WRITE_ONCE(avg->util_est.enqueued, enqueued); } +/* + * The clock_pelt scales the time to reflect the effective amount of + * computation done during the running delta time but then sync back to + * clock_task when rq is idle. + * + * + * absolute time | 1| 2| 3| 4| 5| 6| 7| 8| 9|10|11|12|13|14|15|16 + * @ max capacity ------******---------------******--------------- + * @ half capacity ------************---------************--------- + * clock pelt | 1| 2| 3| 4| 7| 8| 9| 10| 11|14|15|16 + * + */ +static inline void update_rq_clock_pelt(struct rq *rq, s64 delta) +{ + if (unlikely(is_idle_task(rq->curr))) { + /* The rq is idle, we can sync to clock_task */ + rq->clock_pelt = rq_clock_task(rq); + return; + } + + /* + * When a rq runs at a lower compute capacity, it will need + * more time to do the same amount of work than at max + * capacity. In order to be invariant, we scale the delta to + * reflect how much work has been really done. + * Running longer results in stealing idle time that will + * disturb the load signal compared to max capacity. This + * stolen idle time will be automatically reflected when the + * rq will be idle and the clock will be synced with + * rq_clock_task. + */ + + /* + * Scale the elapsed time to reflect the real amount of + * computation + */ + delta = cap_scale(delta, arch_scale_cpu_capacity(NULL, cpu_of(rq))); + delta = cap_scale(delta, arch_scale_freq_capacity(cpu_of(rq))); + + rq->clock_pelt += delta; +} + +/* + * When rq becomes idle, we have to check if it has lost idle time + * because it was fully busy. A rq is fully used when the /Sum util_sum + * is greater or equal to: + * (LOAD_AVG_MAX - 1024 + rq->cfs.avg.period_contrib) << SCHED_CAPACITY_SHIFT; + * For optimization and computing rounding purpose, we don't take into account + * the position in the current window (period_contrib) and we use the higher + * bound of util_sum to decide. + */ +static inline void update_idle_rq_clock_pelt(struct rq *rq) +{ + u32 divider = ((LOAD_AVG_MAX - 1024) << SCHED_CAPACITY_SHIFT) - LOAD_AVG_MAX; + u32 util_sum = rq->cfs.avg.util_sum; + util_sum += rq->avg_rt.util_sum; + util_sum += rq->avg_dl.util_sum; + + /* + * Reflecting stolen time makes sense only if the idle + * phase would be present at max capacity. As soon as the + * utilization of a rq has reached the maximum value, it is + * considered as an always runnig rq without idle time to + * steal. This potential idle time is considered as lost in + * this case. We keep track of this lost idle time compare to + * rq's clock_task. + */ + if (util_sum >= divider) + rq->lost_idle_time += rq_clock_task(rq) - rq->clock_pelt; +} + +static inline u64 rq_clock_pelt(struct rq *rq) +{ + lockdep_assert_held(&rq->lock); + assert_clock_updated(rq); + + return rq->clock_pelt - rq->lost_idle_time; +} + +#ifdef CONFIG_CFS_BANDWIDTH +/* rq->task_clock normalized against any time this cfs_rq has spent throttled */ +static inline u64 cfs_rq_clock_pelt(struct cfs_rq *cfs_rq) +{ + if (unlikely(cfs_rq->throttle_count)) + return cfs_rq->throttled_clock_task - cfs_rq->throttled_clock_task_time; + + return rq_clock_pelt(rq_of(cfs_rq)) - cfs_rq->throttled_clock_task_time; +} +#else +static inline u64 cfs_rq_clock_pelt(struct cfs_rq *cfs_rq) +{ + return rq_clock_pelt(rq_of(cfs_rq)); +} +#endif + #else static inline int @@ -67,6 +163,18 @@ update_irq_load_avg(struct rq *rq, u64 running) { return 0; } + +static inline u64 rq_clock_pelt(struct rq *rq) +{ + return rq_clock_task(rq); +} + +static inline void +update_rq_clock_pelt(struct rq *rq, s64 delta) { } + +static inline void +update_idle_rq_clock_pelt(struct rq *rq) { } + #endif diff --git a/kernel/sched/rt.c b/kernel/sched/rt.c index e4f398a..90fa23d 100644 --- a/kernel/sched/rt.c +++ b/kernel/sched/rt.c @@ -1587,7 +1587,7 @@ pick_next_task_rt(struct rq *rq, struct task_struct *prev, struct rq_flags *rf) * rt task */ if (rq->curr->sched_class != &rt_sched_class) - update_rt_rq_load_avg(rq_clock_task(rq), rq, 0); + update_rt_rq_load_avg(rq_clock_pelt(rq), rq, 0); return p; } @@ -1596,7 +1596,7 @@ static void put_prev_task_rt(struct rq *rq, struct task_struct *p) { update_curr_rt(rq); - update_rt_rq_load_avg(rq_clock_task(rq), rq, 1); + update_rt_rq_load_avg(rq_clock_pelt(rq), rq, 1); /* * The previous task needs to be made eligible for pushing @@ -2325,7 +2325,7 @@ static void task_tick_rt(struct rq *rq, struct task_struct *p, int queued) struct sched_rt_entity *rt_se = &p->rt; update_curr_rt(rq); - update_rt_rq_load_avg(rq_clock_task(rq), rq, 1); + update_rt_rq_load_avg(rq_clock_pelt(rq), rq, 1); watchdog(rq, p); diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index d57a590..24007af 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -861,7 +861,10 @@ struct rq { unsigned int clock_update_flags; u64 clock; - u64 clock_task; + /* Ensure that all clocks are in the same cache line */ + u64 clock_task ____cacheline_aligned; + u64 clock_pelt; + unsigned long lost_idle_time; atomic_t nr_iowait; From patchwork Wed Jan 23 15:26:54 2019 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Vincent Guittot X-Patchwork-Id: 156383 Delivered-To: patch@linaro.org Received: by 2002:a02:48:0:0:0:0:0 with SMTP id 69csp841497jaa; Wed, 23 Jan 2019 07:27:15 -0800 (PST) X-Google-Smtp-Source: ALg8bN6jWv/tIyt3LsmEu5psVDdX4h75XTiKOH23WUo28N29sDoSV/f4Ng9TI9WgqdOorYaLAfDE X-Received: by 2002:a63:5107:: with SMTP id f7mr2248420pgb.218.1548257235058; Wed, 23 Jan 2019 07:27:15 -0800 (PST) ARC-Seal: i=1; a=rsa-sha256; t=1548257235; cv=none; d=google.com; s=arc-20160816; b=fPLgAbQNcf8C9QB4+LS8BzvLWVVjphaReGH31nfqtwm4hzOTjVjgynFMl1phwvIEFW 0uwpLYJ78p5Hz4NLK3BPXEcNEvhmepzYhizgUpfwqWEH3ScGTxc5n1HuEbTowN0P7dwg imCpWIT8cpkI4MR7w7BUrjSTvGt8PaVE0JlRE0mzr0gmmLd1vdrNCf64vBa7T+aXUFmF qEMgm9VB8q4CnK/zU2N9ADTR8IjEoMzn1k6jPIrDAtUKlIXwHRR11FvtIoAJqvGo2jXR ofnsUKE25SuWrRnn0i0KLnNp/vA++iDhNEPbiosFNCHwGXxdtluy1bEqllNnbG27V92q WHiA== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=arc-20160816; h=list-id:precedence:sender:references:in-reply-to:message-id:date :subject:cc:to:from:dkim-signature; bh=47RagOthpiQeiOYWnjegSBd4696NQ+py1cx66QJ0Pp4=; b=GR85XIJbg6FLQ3wJuKXEyW5Jz48kqlOyJYt5n5q6JENYTwUbuSecmDzz6gcYmhiPwe v5mOfDt6vPxAgxBttJF3ZxSwoK58xNZJQ+wvFoJjgevpNBixX+zvQWeGAW5y6ej5FXLi FE7aPdmcPtRn1JOPE0RBO51PCM22C4sPU6cRFSR7nTj2q8OWE5OlQnI2OZaj9eHwZ5kL 5TokkmEWpIUfsfcG9cGLISFaur5RGAQ6A1qgKm2bIaGbWSV3E5qejZvVJ+MpCu0snmKE nchHnOSVr0CENoxqpucHfS5RqkDvmXGcYHQRVYepfPbktZN1XVbEg+DG8fMmqrGZo1T3 g3Fw== ARC-Authentication-Results: i=1; mx.google.com; dkim=pass header.i=@linaro.org header.s=google header.b=hBG6moWP; spf=pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) smtp.mailfrom=linux-kernel-owner@vger.kernel.org; dmarc=pass (p=NONE sp=NONE dis=NONE) header.from=linaro.org Return-Path: Received: from vger.kernel.org (vger.kernel.org. [209.132.180.67]) by mx.google.com with ESMTP id d82si18546730pfj.124.2019.01.23.07.27.14; Wed, 23 Jan 2019 07:27:15 -0800 (PST) Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67; Authentication-Results: mx.google.com; dkim=pass header.i=@linaro.org header.s=google header.b=hBG6moWP; spf=pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) smtp.mailfrom=linux-kernel-owner@vger.kernel.org; dmarc=pass (p=NONE sp=NONE dis=NONE) header.from=linaro.org Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1726296AbfAWP1N (ORCPT + 31 others); Wed, 23 Jan 2019 10:27:13 -0500 Received: from mail-wr1-f66.google.com ([209.85.221.66]:40948 "EHLO mail-wr1-f66.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1725994AbfAWP1H (ORCPT ); Wed, 23 Jan 2019 10:27:07 -0500 Received: by mail-wr1-f66.google.com with SMTP id p4so2941070wrt.7 for ; Wed, 23 Jan 2019 07:27:06 -0800 (PST) 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=47RagOthpiQeiOYWnjegSBd4696NQ+py1cx66QJ0Pp4=; b=hBG6moWP86hsWSE1cXP8aTZsfEg1gG1D6/H2Er1xHE0339O/Uw/CRh3s7nmv7H3xEX 1ScsFq6ZwpklaK1Qmz9Git57ockm7VqItlkozkXfuwAY4nQvze0iOf9fEBMdSh3xXEY1 5lE2f6TMQo8Z7Bo/Hx0TilHhf5AnsZ9iVT6j8= 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=47RagOthpiQeiOYWnjegSBd4696NQ+py1cx66QJ0Pp4=; b=uIQppo7CKCMV4oMaReH/7QpcWdWENHLqrVcyTqeTJOW5g3tzLwzL6I9VsQ/DWavTdP sTPFatSPoypOJCt2fU1jxrnWWutN2CtHXeOb+XF+5Sqb32pTYYWpv+UZt12t1hWB0xLT n7pJhVRX0glgxo9/vm4aM9nGX2DFLgazmHD6QFTZ1BeAlvxCzdKzPTHkTI124fhSdfQD IipciQ2zOMYq3TOA/02oyQgTvu+5h+e9wyXvQLmsbEde36bRSm9OWfYQSF5cvrygyRFz fAJd3oCaaSK216ZU8cPSFdZ02ogGuXN+04YXzscpm2/w6F6Pi34qlB6ST7J3yEefFFzK GHsA== X-Gm-Message-State: AJcUukdgN2AO11QzAmSXMCgiTyqnwq9BOj21YDARd9H5FWZqd64KnkYV dqtdJa7jm1DsoagXrKOprWzRsA== X-Received: by 2002:a5d:4202:: with SMTP id n2mr3237888wrq.260.1548257225394; Wed, 23 Jan 2019 07:27:05 -0800 (PST) Received: from localhost.localdomain ([2a01:e0a:f:6020:5ca6:2921:648b:56b6]) by smtp.gmail.com with ESMTPSA id b13sm92938536wrn.28.2019.01.23.07.27.04 (version=TLS1_2 cipher=ECDHE-RSA-AES128-SHA bits=128/128); Wed, 23 Jan 2019 07:27:04 -0800 (PST) From: Vincent Guittot To: peterz@infradead.org, mingo@kernel.org, linux-kernel@vger.kernel.org Cc: rjw@rjwysocki.net, dietmar.eggemann@arm.com, Morten.Rasmussen@arm.com, patrick.bellasi@arm.com, pjt@google.com, bsegall@google.com, thara.gopinath@linaro.org, pkondeti@codeaurora.org, quentin.perret@arm.com, srinivas.pandruvada@linux.intel.com, Vincent Guittot Subject: [PATCH v9 3/3] sched/pelt: skip updating util_est when utilization is higher than cpu's capacity Date: Wed, 23 Jan 2019 16:26:54 +0100 Message-Id: <1548257214-13745-4-git-send-email-vincent.guittot@linaro.org> X-Mailer: git-send-email 2.7.4 In-Reply-To: <1548257214-13745-1-git-send-email-vincent.guittot@linaro.org> References: <1548257214-13745-1-git-send-email-vincent.guittot@linaro.org> Sender: linux-kernel-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org util_est is mainly meant to be a lower-bound for tasks utilization. That's why task_util_est() returns the actual util_avg when it's higher than the estimated utilization. With new invaraince signal and without any special check on samples collection, if a task is limited because of thermal capping for example, we could end up overestimating its utilization and thus perhaps generating an unwanted frequency spike when the capping is relaxed... and (even worst) it will take some more activations for the estimated utilization to converge back to the actual utilization. Since we cannot easily know if there is idle time in a CPU when a task completes an activation with a utilization higher then the CPU capacity, we skip the sampling when utilization is higher than cpu's capacity. Suggested-by: Patrick Bellasi Signed-off-by: Vincent Guittot --- kernel/sched/fair.c | 14 +++++++++----- kernel/sched/sched.h | 7 +++++++ 2 files changed, 16 insertions(+), 5 deletions(-) -- 2.7.4 diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index a9cee4e..47e63441 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -3638,6 +3638,7 @@ util_est_dequeue(struct cfs_rq *cfs_rq, struct task_struct *p, bool task_sleep) { long last_ewma_diff; struct util_est ue; + int cpu; if (!sched_feat(UTIL_EST)) return; @@ -3672,6 +3673,14 @@ util_est_dequeue(struct cfs_rq *cfs_rq, struct task_struct *p, bool task_sleep) return; /* + * To avoid overestimation of actual task utilization, skip updates if + * we cannot grant there is idle time in this CPU. + */ + cpu = cpu_of(rq_of(cfs_rq)); + if (task_util(p) > capacity_orig_of(cpu)) + return; + + /* * Update Task's estimated utilization * * When *p completes an activation we can consolidate another sample @@ -5542,11 +5551,6 @@ static unsigned long capacity_of(int cpu) return cpu_rq(cpu)->cpu_capacity; } -static unsigned long capacity_orig_of(int cpu) -{ - return cpu_rq(cpu)->cpu_capacity_orig; -} - static unsigned long cpu_avg_load_per_task(int cpu) { struct rq *rq = cpu_rq(cpu); diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index 24007af..957c3f3 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -2230,6 +2230,13 @@ static inline void cpufreq_update_util(struct rq *rq, unsigned int flags) {} # define arch_scale_freq_invariant() false #endif +#ifdef CONFIG_SMP +static inline unsigned long capacity_orig_of(int cpu) +{ + return cpu_rq(cpu)->cpu_capacity_orig; +} +#endif + #ifdef CONFIG_CPU_FREQ_GOV_SCHEDUTIL /** * enum schedutil_type - CPU utilization type