From nobody Tue Feb 10 05:48:29 2026 Received: from mail-wm1-f41.google.com (mail-wm1-f41.google.com [209.85.128.41]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 9D0E845BEC for ; Fri, 29 Nov 2024 16:18:03 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.128.41 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1732897085; cv=none; b=teOo8UxZ0JWc5ne/QlA6UTQ+7ZkHB6mRGzSHIOYdL8ORFb/zA2mfMSYJH20SI1AsW9liCypinmsUAR/UlcTm7+tBeN9jqhUUyIJk/WxeNKyaNcKhew61YZWXdL7TE1xLt5qaOPpxc4E/HJkpIuYO3qoeNb6ld8DVvXXqQaBKeu0= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1732897085; c=relaxed/simple; bh=mVjbS47HdiYvofNh5GZkckIEOg0gcnZo8m6/Giv5pm8=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=H8V4A9s/7hyak4nqvt/FObOEMHrk2p0pVaJKz3h5cE6IKqi5tXc6CMfkO4qX/clDnRJCVHC59SuUI0/VDrJHmjgqeaGACt8/Il+ow3A2Aru1nPsJJU3aypRX+UUhoFKFVLrHZh/ac7GGkFH3nP7xQ39d5Wz0aeoNHoN+fCUSJLA= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=linaro.org; spf=pass smtp.mailfrom=linaro.org; dkim=pass (2048-bit key) header.d=linaro.org header.i=@linaro.org header.b=lH0KGkGP; arc=none smtp.client-ip=209.85.128.41 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=linaro.org Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=linaro.org Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=linaro.org header.i=@linaro.org header.b="lH0KGkGP" Received: by mail-wm1-f41.google.com with SMTP id 5b1f17b1804b1-434a45f05feso22577385e9.3 for ; Fri, 29 Nov 2024 08:18:03 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=linaro.org; s=google; t=1732897082; x=1733501882; darn=vger.kernel.org; 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=FlFKgVHb4nQ5bY+wYMwnva3c054rOZjazLjMOFEFCGI=; b=lH0KGkGP99GfCa07Z3ETGm90TOk/Gse9KXb1es3clYOwdljXrcL891qh9UT0BidLo0 vFBGgod/HtMS8qNiSnk6dn/F+u0EH8wIk8atXrqakyA4AijJBk2dabS6kt7Lsh9rq8yE +f8KCVO4TljXYUjFiAoQG8jnRyxYwG2jZ4ZXvl2HMkSlzVXS5IFGh/1waxNcTpVlkPpf qsEuZQAqEG3DQZr8iRnpoRbwxcCs3r39Tt5dncK0k+ma0WVUp06FGg99CR9Adwfy52jm rJEwvfHD8wHKiXCaqrEBeelyZpMZvRwukP6NjMPhPhuc781XHmFtE6tuItfmNcIa7rDn Mo0Q== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1732897082; x=1733501882; 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=FlFKgVHb4nQ5bY+wYMwnva3c054rOZjazLjMOFEFCGI=; b=M1I2LoE3Cy474pVZncRszK9DH+KEZyDJASClVrcUMEs+4zgeomxeYS+wTqx/kwzPUV jQpAtDmicixO3vtLbcbKyxRjBjFtjBVFmSPesgBPacJnEQsAKUmdUllsdAdJdzOmjDPl N3fxbhDeTJ0JLQ/Tui1Xm5togdsRLkgnhmpjFZeO6agfKNrqEyfLsQMR4RV1RMUX5sP9 YcYyvLVfdU3XvfzC708L2AvBUmonDYDmyBvmI8OjMQWMGy/mpgp64si5Eo8sWehC4pkf ngiEbX54ziW7LA7wZ5Ds3CeAdaI3A1dQjRgawAj0xdAq/Tx9AjlJeCJblffEBA7OCiMA Sl1g== X-Forwarded-Encrypted: i=1; AJvYcCUpEjLzsc9kBvnbSWHHKRIqTJWBUjN30dgsipIldUs9whBDNdtVw94RZMQt+Ol9KtjWvQWuoBmBoV4mmpI=@vger.kernel.org X-Gm-Message-State: AOJu0Yw3YlEBMKvZwaQZah2G3u8s/xkXtznREQx/Ny6x8tUaHeiRC438 ipwedq7XyDjfWOnYh3UoRgdbVh75ewVHusCcL/ob2YrniJOnttrGhELnWOSyD/4= X-Gm-Gg: ASbGncs89hFeszVs/86YM5DtmbJTHhfvXQ3NjnD4qSAfdayCfsCUaLpvlB7pzSUFuHZ jFG4+UcXGJQ2jmmOjA4hUM97a/hlrN2ADp+SWi6KS9CKJPfqZSpYpNB/j9QlEFkM5EchOaCG0cv A4vH9Z3krMobAeigqDbd1Mr8UFtHR1vf5Qq2xCoBSJx7DvquQIk3ZC8V77HSrWj6fZT6vH6SAPc Zjy1q9euVEKoWKBY3zv3EK/HT1BHcEMlY2pnL9Ei6gsBdsOvpXtxD2XpzM= X-Google-Smtp-Source: AGHT+IHxNip/mKKsi5cIrDGLz8EFEzYYZIENgG+Z67N0gzEjPnvbF3TSg7XY3Euzjs1+MkxlWP7oFg== X-Received: by 2002:a05:600c:4e8a:b0:431:562a:54be with SMTP id 5b1f17b1804b1-434a9dc36c1mr131953595e9.9.1732897081748; Fri, 29 Nov 2024 08:18:01 -0800 (PST) Received: from vingu-cube.. ([2a01:e0a:f:6020:c54b:5636:4db3:1028]) by smtp.gmail.com with ESMTPSA id 5b1f17b1804b1-434b0dbf95fsm56771665e9.15.2024.11.29.08.18.00 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Fri, 29 Nov 2024 08:18:01 -0800 (PST) From: Vincent Guittot To: mingo@redhat.com, peterz@infradead.org, juri.lelli@redhat.com, dietmar.eggemann@arm.com, rostedt@goodmis.org, bsegall@google.com, mgorman@suse.de, vschneid@redhat.com, linux-kernel@vger.kernel.org Cc: kprateek.nayak@amd.com, pauld@redhat.com, efault@gmx.de, luis.machado@arm.com, Vincent Guittot Subject: [PATCH 01/10 v2] sched/eevdf: More PELT vs DELAYED_DEQUEUE Date: Fri, 29 Nov 2024 17:17:47 +0100 Message-ID: <20241129161756.3081386-2-vincent.guittot@linaro.org> X-Mailer: git-send-email 2.43.0 In-Reply-To: <20241129161756.3081386-1-vincent.guittot@linaro.org> References: <20241129161756.3081386-1-vincent.guittot@linaro.org> 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" From: Peter Zijlstra Vincent and Dietmar noted that while commit fc1892becd56 ("sched/eevdf: Fixup PELT vs DELAYED_DEQUEUE") fixes the entity runnable stats, it does not adjust the cfs_rq runnable stats, which are based off of h_nr_running. Track h_nr_delayed such that we can discount those and adjust the signal. Fixes: fc1892becd56 ("sched/eevdf: Fixup PELT vs DELAYED_DEQUEUE") Reported-by: Dietmar Eggemann Closes: https://lore.kernel.org/lkml/a9a45193-d0c6-4ba2-a822-464ad30b550e@a= rm.com/ Reported-by: Vincent Guittot Closes: https://lore.kernel.org/lkml/CAKfTPtCNUvWE_GX5LyvTF-WdxUT=3DZgvZZv-= 4t=3DeWntg5uOFqiQ@mail.gmail.com/ Signed-off-by: Peter Zijlstra (Intel) [ Fixes checkpatch warnings ] Signed-off-by: Vincent Guittot Link: https://lkml.kernel.org/r/20240906104525.GG4928@noisy.programming.kic= ks-ass.net Tested-by: K Prateek Nayak --- kernel/sched/debug.c | 1 + kernel/sched/fair.c | 51 +++++++++++++++++++++++++++++++++++++++----- kernel/sched/pelt.c | 2 +- kernel/sched/sched.h | 8 +++++-- 4 files changed, 54 insertions(+), 8 deletions(-) diff --git a/kernel/sched/debug.c b/kernel/sched/debug.c index a48b2a701ec2..a1be00a988bf 100644 --- a/kernel/sched/debug.c +++ b/kernel/sched/debug.c @@ -845,6 +845,7 @@ void print_cfs_rq(struct seq_file *m, int cpu, struct c= fs_rq *cfs_rq) SEQ_printf(m, " .%-30s: %Ld.%06ld\n", "spread", SPLIT_NS(spread)); SEQ_printf(m, " .%-30s: %d\n", "nr_running", cfs_rq->nr_running); SEQ_printf(m, " .%-30s: %d\n", "h_nr_running", cfs_rq->h_nr_running); + SEQ_printf(m, " .%-30s: %d\n", "h_nr_delayed", cfs_rq->h_nr_delayed); SEQ_printf(m, " .%-30s: %d\n", "idle_nr_running", cfs_rq->idle_nr_running); SEQ_printf(m, " .%-30s: %d\n", "idle_h_nr_running", diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 3356315d7e64..f4ed2ed070e4 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -5465,9 +5465,33 @@ static void clear_buddies(struct cfs_rq *cfs_rq, str= uct sched_entity *se) =20 static __always_inline void return_cfs_rq_runtime(struct cfs_rq *cfs_rq); =20 -static inline void finish_delayed_dequeue_entity(struct sched_entity *se) +static void set_delayed(struct sched_entity *se) +{ + se->sched_delayed =3D 1; + for_each_sched_entity(se) { + struct cfs_rq *cfs_rq =3D cfs_rq_of(se); + + cfs_rq->h_nr_delayed++; + if (cfs_rq_throttled(cfs_rq)) + break; + } +} + +static void clear_delayed(struct sched_entity *se) { se->sched_delayed =3D 0; + for_each_sched_entity(se) { + struct cfs_rq *cfs_rq =3D cfs_rq_of(se); + + cfs_rq->h_nr_delayed--; + if (cfs_rq_throttled(cfs_rq)) + break; + } +} + +static inline void finish_delayed_dequeue_entity(struct sched_entity *se) +{ + clear_delayed(se); if (sched_feat(DELAY_ZERO) && se->vlag > 0) se->vlag =3D 0; } @@ -5497,7 +5521,7 @@ dequeue_entity(struct cfs_rq *cfs_rq, struct sched_en= tity *se, int flags) if (cfs_rq->next =3D=3D se) cfs_rq->next =3D NULL; update_load_avg(cfs_rq, se, 0); - se->sched_delayed =3D 1; + set_delayed(se); return false; } } @@ -5910,7 +5934,7 @@ static bool throttle_cfs_rq(struct cfs_rq *cfs_rq) struct rq *rq =3D rq_of(cfs_rq); struct cfs_bandwidth *cfs_b =3D tg_cfs_bandwidth(cfs_rq->tg); struct sched_entity *se; - long task_delta, idle_task_delta, dequeue =3D 1; + long task_delta, idle_task_delta, delayed_delta, dequeue =3D 1; long rq_h_nr_running =3D rq->cfs.h_nr_running; =20 raw_spin_lock(&cfs_b->lock); @@ -5943,6 +5967,7 @@ static bool throttle_cfs_rq(struct cfs_rq *cfs_rq) =20 task_delta =3D cfs_rq->h_nr_running; idle_task_delta =3D cfs_rq->idle_h_nr_running; + delayed_delta =3D cfs_rq->h_nr_delayed; for_each_sched_entity(se) { struct cfs_rq *qcfs_rq =3D cfs_rq_of(se); int flags; @@ -5966,6 +5991,7 @@ static bool throttle_cfs_rq(struct cfs_rq *cfs_rq) =20 qcfs_rq->h_nr_running -=3D task_delta; qcfs_rq->idle_h_nr_running -=3D idle_task_delta; + qcfs_rq->h_nr_delayed -=3D delayed_delta; =20 if (qcfs_rq->load.weight) { /* Avoid re-evaluating load for this entity: */ @@ -5988,6 +6014,7 @@ static bool throttle_cfs_rq(struct cfs_rq *cfs_rq) =20 qcfs_rq->h_nr_running -=3D task_delta; qcfs_rq->idle_h_nr_running -=3D idle_task_delta; + qcfs_rq->h_nr_delayed -=3D delayed_delta; } =20 /* At this point se is NULL and we are at root level*/ @@ -6013,7 +6040,7 @@ void unthrottle_cfs_rq(struct cfs_rq *cfs_rq) struct rq *rq =3D rq_of(cfs_rq); struct cfs_bandwidth *cfs_b =3D tg_cfs_bandwidth(cfs_rq->tg); struct sched_entity *se; - long task_delta, idle_task_delta; + long task_delta, idle_task_delta, delayed_delta; long rq_h_nr_running =3D rq->cfs.h_nr_running; =20 se =3D cfs_rq->tg->se[cpu_of(rq)]; @@ -6049,6 +6076,7 @@ void unthrottle_cfs_rq(struct cfs_rq *cfs_rq) =20 task_delta =3D cfs_rq->h_nr_running; idle_task_delta =3D cfs_rq->idle_h_nr_running; + delayed_delta =3D cfs_rq->h_nr_delayed; for_each_sched_entity(se) { struct cfs_rq *qcfs_rq =3D cfs_rq_of(se); =20 @@ -6066,6 +6094,7 @@ void unthrottle_cfs_rq(struct cfs_rq *cfs_rq) =20 qcfs_rq->h_nr_running +=3D task_delta; qcfs_rq->idle_h_nr_running +=3D idle_task_delta; + qcfs_rq->h_nr_delayed +=3D delayed_delta; =20 /* end evaluation on encountering a throttled cfs_rq */ if (cfs_rq_throttled(qcfs_rq)) @@ -6083,6 +6112,7 @@ void unthrottle_cfs_rq(struct cfs_rq *cfs_rq) =20 qcfs_rq->h_nr_running +=3D task_delta; qcfs_rq->idle_h_nr_running +=3D idle_task_delta; + qcfs_rq->h_nr_delayed +=3D delayed_delta; =20 /* end evaluation on encountering a throttled cfs_rq */ if (cfs_rq_throttled(qcfs_rq)) @@ -6936,7 +6966,7 @@ requeue_delayed_entity(struct sched_entity *se) } =20 update_load_avg(cfs_rq, se, 0); - se->sched_delayed =3D 0; + clear_delayed(se); } =20 /* @@ -6950,6 +6980,7 @@ enqueue_task_fair(struct rq *rq, struct task_struct *= p, int flags) struct cfs_rq *cfs_rq; struct sched_entity *se =3D &p->se; int idle_h_nr_running =3D task_has_idle_policy(p); + int h_nr_delayed =3D 0; int task_new =3D !(flags & ENQUEUE_WAKEUP); int rq_h_nr_running =3D rq->cfs.h_nr_running; u64 slice =3D 0; @@ -6976,6 +7007,9 @@ enqueue_task_fair(struct rq *rq, struct task_struct *= p, int flags) if (p->in_iowait) cpufreq_update_util(rq, SCHED_CPUFREQ_IOWAIT); =20 + if (task_new) + h_nr_delayed =3D !!se->sched_delayed; + for_each_sched_entity(se) { if (se->on_rq) { if (se->sched_delayed) @@ -6998,6 +7032,7 @@ enqueue_task_fair(struct rq *rq, struct task_struct *= p, int flags) =20 cfs_rq->h_nr_running++; cfs_rq->idle_h_nr_running +=3D idle_h_nr_running; + cfs_rq->h_nr_delayed +=3D h_nr_delayed; =20 if (cfs_rq_is_idle(cfs_rq)) idle_h_nr_running =3D 1; @@ -7021,6 +7056,7 @@ enqueue_task_fair(struct rq *rq, struct task_struct *= p, int flags) =20 cfs_rq->h_nr_running++; cfs_rq->idle_h_nr_running +=3D idle_h_nr_running; + cfs_rq->h_nr_delayed +=3D h_nr_delayed; =20 if (cfs_rq_is_idle(cfs_rq)) idle_h_nr_running =3D 1; @@ -7083,6 +7119,7 @@ static int dequeue_entities(struct rq *rq, struct sch= ed_entity *se, int flags) struct task_struct *p =3D NULL; int idle_h_nr_running =3D 0; int h_nr_running =3D 0; + int h_nr_delayed =3D 0; struct cfs_rq *cfs_rq; u64 slice =3D 0; =20 @@ -7090,6 +7127,8 @@ static int dequeue_entities(struct rq *rq, struct sch= ed_entity *se, int flags) p =3D task_of(se); h_nr_running =3D 1; idle_h_nr_running =3D task_has_idle_policy(p); + if (!task_sleep && !task_delayed) + h_nr_delayed =3D !!se->sched_delayed; } else { cfs_rq =3D group_cfs_rq(se); slice =3D cfs_rq_min_slice(cfs_rq); @@ -7107,6 +7146,7 @@ static int dequeue_entities(struct rq *rq, struct sch= ed_entity *se, int flags) =20 cfs_rq->h_nr_running -=3D h_nr_running; cfs_rq->idle_h_nr_running -=3D idle_h_nr_running; + cfs_rq->h_nr_delayed -=3D h_nr_delayed; =20 if (cfs_rq_is_idle(cfs_rq)) idle_h_nr_running =3D h_nr_running; @@ -7145,6 +7185,7 @@ static int dequeue_entities(struct rq *rq, struct sch= ed_entity *se, int flags) =20 cfs_rq->h_nr_running -=3D h_nr_running; cfs_rq->idle_h_nr_running -=3D idle_h_nr_running; + cfs_rq->h_nr_delayed -=3D h_nr_delayed; =20 if (cfs_rq_is_idle(cfs_rq)) idle_h_nr_running =3D h_nr_running; diff --git a/kernel/sched/pelt.c b/kernel/sched/pelt.c index fc07382361a8..fee75cc2c47b 100644 --- a/kernel/sched/pelt.c +++ b/kernel/sched/pelt.c @@ -321,7 +321,7 @@ int __update_load_avg_cfs_rq(u64 now, struct cfs_rq *cf= s_rq) { if (___update_load_sum(now, &cfs_rq->avg, scale_load_down(cfs_rq->load.weight), - cfs_rq->h_nr_running, + cfs_rq->h_nr_running - cfs_rq->h_nr_delayed, cfs_rq->curr !=3D NULL)) { =20 ___update_load_avg(&cfs_rq->avg, 1); diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index 090dd4b38fa2..77a05ca0cce7 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -649,6 +649,7 @@ struct cfs_rq { unsigned int h_nr_running; /* SCHED_{NORMAL,BATCH,IDLE} */ unsigned int idle_nr_running; /* SCHED_IDLE */ unsigned int idle_h_nr_running; /* SCHED_IDLE */ + unsigned int h_nr_delayed; =20 s64 avg_vruntime; u64 avg_load; @@ -897,8 +898,11 @@ struct dl_rq { =20 static inline void se_update_runnable(struct sched_entity *se) { - if (!entity_is_task(se)) - se->runnable_weight =3D se->my_q->h_nr_running; + if (!entity_is_task(se)) { + struct cfs_rq *cfs_rq =3D se->my_q; + + se->runnable_weight =3D cfs_rq->h_nr_running - cfs_rq->h_nr_delayed; + } } =20 static inline long se_runnable(struct sched_entity *se) --=20 2.43.0