From nobody Sun Feb 8 21:29:00 2026 Received: from foss.arm.com (foss.arm.com [217.140.110.172]) by smtp.subspace.kernel.org (Postfix) with ESMTP id 6826B15E5CC for ; Tue, 7 May 2024 12:51:01 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=217.140.110.172 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1715086262; cv=none; b=jBAzLGJPjMmoS/sR51UDDn+O/u7iRv9dalT+w3QCeV4GUL/M9bmtNOVDgZY1f9WLnXtSnrXihXqd/dmfDoEhQuKeJBk1SX1o2QRqvOkeeOyn777dw5dVbx6lg6Lu71j52/YhdYNW/xFLpUqrlcrLVlyy01kzE1JOQdTjI2SsJFw= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1715086262; c=relaxed/simple; bh=kCOFTh2w1NJO+h53GHlkLLXq/I3SCktxk8/Z5w+DA04=; h=From:To:Cc:Subject:Date:Message-Id:In-Reply-To:References: MIME-Version; b=i77hMn9ZR8OLmJ58b/sYVLzQLZ2lnngsJG69Lgt0e9gOcbZzYgaTFzOuCwtO3rpCEAoM8sSdaJ2tkfcd0ELfUNrhDPghX3TM/68Yui/2yyZhibTDSqOAiqsvqL7syLwZj/YhCexKDpftryfEC2nP0w8gXM4d7i2Xz1apFzcg1SE= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=arm.com; spf=pass smtp.mailfrom=arm.com; arc=none smtp.client-ip=217.140.110.172 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=arm.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=arm.com 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 D4AC31063; Tue, 7 May 2024 05:51:26 -0700 (PDT) Received: from e130256.cambridge.arm.com (usa-sjc-imap-foss1.foss.arm.com [10.121.207.14]) by usa-sjc-imap-foss1.foss.arm.com (Postfix) with ESMTPSA id D1EF83F793; Tue, 7 May 2024 05:50:58 -0700 (PDT) From: Hongyan Xia To: Ingo Molnar , Peter Zijlstra , Vincent Guittot , Dietmar Eggemann , Juri Lelli , Steven Rostedt , Ben Segall , Mel Gorman , Daniel Bristot de Oliveira , Valentin Schneider Cc: Qais Yousef , Morten Rasmussen , Lukasz Luba , Christian Loehle , pierre.gondois@arm.com, linux-kernel@vger.kernel.org, Hongyan Xia Subject: [RFC PATCH v3 6/6] Propagate negative bias Date: Tue, 7 May 2024 13:50:29 +0100 Message-Id: X-Mailer: git-send-email 2.34.1 In-Reply-To: References: Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" Negative bias is interesting, because dequeuing such a task will actually increase utilization. Solve by applying PELT decay to negative biases as well. This in fact can be implemented easily with some math tricks. Signed-off-by: Hongyan Xia --- kernel/sched/fair.c | 44 ++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 44 insertions(+) diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 0177d7e8f364..7259a61e9ae5 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -4863,6 +4863,45 @@ static inline unsigned long task_util_est_uclamp(str= uct task_struct *p) { return max(task_util_uclamp(p), _task_util_est_uclamp(p)); } + +/* + * Negative biases are tricky. If we remove them right away then dequeuing= a + * uclamp_max task has the interesting effect that dequeuing results in a = higher + * rq utilization. Solve this by applying PELT decay to the bias itself. + * + * Keeping track of a PELT-decayed negative bias is extra overhead. Howeve= r, we + * observe this interesting math property, where y is the decay factor and= p is + * the number of periods elapsed: + * + * util_new =3D util_old * y^p - neg_bias * y^p + * =3D (util_old - neg_bias) * y^p + * + * Therefore, we simply subtract the negative bias from util_avg the momen= t we + * dequeue, then the PELT signal itself is the total of util_avg and the d= ecayed + * negative bias, and we no longer need to track the decayed bias separate= ly. + */ +static void propagate_negative_bias(struct task_struct *p) +{ + if (task_util_bias(p) < 0 && !task_on_rq_migrating(p)) { + unsigned long neg_bias =3D -task_util_bias(p); + struct sched_entity *se =3D &p->se; + struct cfs_rq *cfs_rq; + + p->se.avg.util_avg_bias =3D 0; + + for_each_sched_entity(se) { + u32 divider, neg_sum; + + cfs_rq =3D cfs_rq_of(se); + divider =3D get_pelt_divider(&cfs_rq->avg); + neg_sum =3D neg_bias * divider; + sub_positive(&se->avg.util_avg, neg_bias); + sub_positive(&se->avg.util_sum, neg_sum); + sub_positive(&cfs_rq->avg.util_avg, neg_bias); + sub_positive(&cfs_rq->avg.util_sum, neg_sum); + } + } +} #else static inline long task_util_bias(struct task_struct *p) { @@ -4883,6 +4922,10 @@ static inline unsigned long task_util_est_uclamp(str= uct task_struct *p) { return task_util_est(p); } + +static void propagate_negative_bias(struct task_struct *p) +{ +} #endif =20 static inline void util_est_enqueue(struct cfs_rq *cfs_rq, @@ -6844,6 +6887,7 @@ static void dequeue_task_fair(struct rq *rq, struct t= ask_struct *p, int flags) /* At this point se is NULL and we are at root level*/ sub_nr_running(rq, 1); util_bias_dequeue(&rq->cfs.avg, p); + propagate_negative_bias(p); /* XXX: We should skip the update above and only do it once here. */ cpufreq_update_util(rq, 0); =20 --=20 2.34.1