From nobody Mon Apr 6 18:54:58 2026 Received: from CO1PR03CU002.outbound.protection.outlook.com (mail-westus2azon11010063.outbound.protection.outlook.com [52.101.46.63]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 31E8F3A641C for ; Wed, 18 Mar 2026 09:22:27 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=fail smtp.client-ip=52.101.46.63 ARC-Seal: i=2; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1773825749; cv=fail; b=aKJJQ/ow1mQz0fLZcT0IsORiaehzQeRkIFZQPnQvXu2+unmJrp8U4Q5cfgKDbztawZCLUnNaWLzfUDGsgK4gyh7TEfZhaKrSJkwfbPzidBX0kT2vQQ4IF1gay4Y3xLarIp7X8EZn9T0HCaNVG8NN9V+rNXLpFFuwNEXTbZt+TV4= ARC-Message-Signature: i=2; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1773825749; c=relaxed/simple; bh=fvyjIxVsfKrCv/uKq4BmRvyARi+Kk8Lvbjf8m90ADLk=; h=From:To:Cc:Subject:Date:Message-ID:Content-Type:MIME-Version; b=DrTFO9JHWCUplWAR3EP5tprnGQJcHAIUgft0NF3YCdvAjn6s++vK41uc339ootB1Jhty0Wv1hVUYVEFiB887gC4sBpvlMj9N5wxGg2dmcUXWjUnTfyOKxMdyp7S1Yz9bTgE8gEhYVKgPx9CYV+2dvabAseIW3X2YOc3LkfV1+S0= ARC-Authentication-Results: i=2; smtp.subspace.kernel.org; dmarc=pass (p=reject dis=none) header.from=nvidia.com; spf=fail smtp.mailfrom=nvidia.com; dkim=pass (2048-bit key) header.d=Nvidia.com header.i=@Nvidia.com header.b=fmN07o4P; arc=fail smtp.client-ip=52.101.46.63 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=reject dis=none) header.from=nvidia.com Authentication-Results: smtp.subspace.kernel.org; spf=fail smtp.mailfrom=nvidia.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=Nvidia.com header.i=@Nvidia.com header.b="fmN07o4P" ARC-Seal: i=1; a=rsa-sha256; s=arcselector10001; d=microsoft.com; cv=none; b=W+927upuf6XM2xhaeoUG+mRHBzhSG0Usd7000scVz34+wWCUe98jp9yJWrxU1lfnpFcA25Oo9YBN9mBlWw4PZ050J9JOVawlf88qZ2xWzbPHE8IA2aIVQUZWDlcmbWkKnCxZX05TYkPm7btbW57MZCVKZnFqeS9tszSAKCZwxq8wXv4nWRAbD3gZ6yb3HpcMXqRsyY/X/phH7rP4MM0P2uQPYqnDnwW/WhuoQbamUT7oDuAfYKaybT0KgAk/wAYi7vE+rIqOt/1j+LYoT0DPjRtfwS3gEAGTXONJxP0mTmmAJXKTFdrs41H+NnhAhr8tpOAO19a6hXZkVJaaJLOaMg== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=microsoft.com; s=arcselector10001; h=From:Date:Subject:Message-ID:Content-Type:MIME-Version:X-MS-Exchange-AntiSpam-MessageData-ChunkCount:X-MS-Exchange-AntiSpam-MessageData-0:X-MS-Exchange-AntiSpam-MessageData-1; bh=SWrkafHm5sNWnXy4gYBVW8qktbHsGpEtShPdN60JvlQ=; b=LNusz/orSz3wY9/e8SzsDMHnfr9xeF/aSOjSW7erL2HqNxeyam7HpO4ckruaME7meyM2Rm1F+GIblgDh9voAGYuNHRezWhBGRwsUL8KMHhpAicRhMXaZpc9wEneE3j9ShbW73grXtkNZ133ORwHrRlwc3PCDcvU+gmfhZib7LMMjxxUyepIUlTVpSkmgspE8tCKBY0xadYdj/3Fjn2wMXVk/007TdH2HYwyvppKDEuDMDrrT+pK9LaQUMetYH1chcJvPz+DRoOzW4y0JCTFcGh/obKstrXx1RE28iXj+9yRdD2wVuFa5cN27lfaAEaswsTUZkfZcbqsdDglzfJ8bVA== ARC-Authentication-Results: i=1; mx.microsoft.com 1; spf=pass smtp.mailfrom=nvidia.com; dmarc=pass action=none header.from=nvidia.com; dkim=pass header.d=nvidia.com; arc=none DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=Nvidia.com; s=selector2; h=From:Date:Subject:Message-ID:Content-Type:MIME-Version:X-MS-Exchange-SenderADCheck; bh=SWrkafHm5sNWnXy4gYBVW8qktbHsGpEtShPdN60JvlQ=; b=fmN07o4PhbAs2Uhj4h80tZoyLnJXPU/nCo90NlZ2ivae05NzLfAFh2ymlp+NsQPvn1zxMeAxbG/30uiyPWqWOrS8y78yWgsB65hEkmD9yOrFK2TZzQ+yE11PiaHbo+/P2CDuXe7T77mlr8weUba93O04FTyD3RHCHr2SNRYTOiFHjUQ3YMl6Qt9gC9QJ6W3QGYvyREPKr5SQr90/Xk02fUJzq4AG1n1QEGOnXCIIB/f+9OQ/ce8UoatTzURm56M3hSi5df9C52fuwpbmcilIYJQJYj68UyREjnsxpB8l1FgxGOLtv/z6eLAMCEBYnI3PNr9dq3y1lqCevLKAtZBnuA== Authentication-Results: dkim=none (message not signed) header.d=none;dmarc=none action=none header.from=nvidia.com; Received: from LV8PR12MB9620.namprd12.prod.outlook.com (2603:10b6:408:2a1::19) by DS0PR12MB999105.namprd12.prod.outlook.com (2603:10b6:8:301::11) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.20.9745.9; Wed, 18 Mar 2026 09:22:24 +0000 Received: from LV8PR12MB9620.namprd12.prod.outlook.com ([fe80::299d:f5e0:3550:1528]) by LV8PR12MB9620.namprd12.prod.outlook.com ([fe80::299d:f5e0:3550:1528%5]) with mapi id 15.20.9723.010; Wed, 18 Mar 2026 09:22:24 +0000 From: Andrea Righi To: Ingo Molnar , Peter Zijlstra , Vincent Guittot Cc: Dietmar Eggemann , Steven Rostedt , Ben Segall , Mel Gorman , Valentin Schneider , Joel Fernandes , linux-kernel@vger.kernel.org Subject: [PATCH] sched/fair: Prefer fully-idle SMT cores in asym-capacity idle selection Date: Wed, 18 Mar 2026 10:22:14 +0100 Message-ID: <20260318092214.130908-1-arighi@nvidia.com> X-Mailer: git-send-email 2.53.0 Content-Transfer-Encoding: quoted-printable X-ClientProxiedBy: ZR2P278CA0086.CHEP278.PROD.OUTLOOK.COM (2603:10a6:910:65::16) To LV8PR12MB9620.namprd12.prod.outlook.com (2603:10b6:408:2a1::19) Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 X-MS-PublicTrafficType: Email X-MS-TrafficTypeDiagnostic: LV8PR12MB9620:EE_|DS0PR12MB999105:EE_ X-MS-Office365-Filtering-Correlation-Id: 6a51e531-defe-43cd-3c14-08de84cfdd48 X-MS-Exchange-SenderADCheck: 1 X-MS-Exchange-AntiSpam-Relay: 0 X-Microsoft-Antispam: BCL:0;ARA:13230040|366016|1800799024|376014|7416014|18002099003|56012099003; X-Microsoft-Antispam-Message-Info: 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 X-Forefront-Antispam-Report: CIP:255.255.255.255;CTRY:;LANG:en;SCL:1;SRV:;IPV:NLI;SFV:NSPM;H:LV8PR12MB9620.namprd12.prod.outlook.com;PTR:;CAT:NONE;SFS:(13230040)(366016)(1800799024)(376014)(7416014)(18002099003)(56012099003);DIR:OUT;SFP:1101; X-MS-Exchange-AntiSpam-MessageData-ChunkCount: 1 X-MS-Exchange-AntiSpam-MessageData-0: =?us-ascii?Q?wS+b1VvIt+t6ciO1vxTbwpxb/FA9qHAIHwaI/2OUe0Gkg0+tYbBJinD3CZJB?= =?us-ascii?Q?CFueGaR0YsvN4GuQyYqhE/T6Okzi/3zJClVk9auwIXdaHBMOcym7c85FoTiY?= =?us-ascii?Q?zhHgUr6DaYrjUY6fYGJ6BdB3YHN5KpDd4hKSH5Yo9IBkCTQTOWJ39kM3ahFn?= =?us-ascii?Q?v05oOeQiGSDUonsy94+CvzBZpzP0OuAA7824v1+6mfew+ur5v3z1mS6uLb4q?= =?us-ascii?Q?8R0Dynqi5lxal6BTXNPZEqBSALqgd/ipLcvVA25h8tkYh3L3ayUGc554I6iT?= =?us-ascii?Q?INR/VMuJ5GJfRD2pY+LFQaOUrZE0DsZ4QlaY2CtCkiN5SLHVzSC2G1H0t0zq?= =?us-ascii?Q?Zwmepmtp4V132tCtXQXgboZBe8FbPr+fJgmPq3v0/N2SmkIB7CSVnN3m0jPM?= =?us-ascii?Q?rcOq1k2jPqnVtke14LbHIEJpOoyLcvlLUkWu6iJh3RI5jy8qyUDESjk8RLO4?= =?us-ascii?Q?8KcjwmaGK2qgesXdZY/TMEIxnFNCoQFWGV+8+LlnRFCnfnTbrBlIumHLeEf9?= =?us-ascii?Q?Bft+5lFam9Bq/ADeKTdsm9HJhogHIVUubGK/X0JsDuI7T5nuWbKxyG9FV8rM?= =?us-ascii?Q?BISmbR4pUv6c1VznNZapQyarDwOIyBxbgoEnRBFEoVJKIZ/5+KDuak/Ift+D?= =?us-ascii?Q?IsPytoLg8CjcNimwaSKJLSbIYaQ9C1mkGWVCWiPEve5DPrYi46kDWpZm1N2d?= =?us-ascii?Q?QGWl/dxThhy+7FbG9FRLTEK94p5erQ0Q6Jb8p/yoBxzWyKyFnbuo0SaW1Cw+?= =?us-ascii?Q?XqurvEci8shcgmn887D7/SSLfTGkn8My6DFmXHOW5jFomFKqJiMwhr5EjmKV?= =?us-ascii?Q?x6oZ2C07ZmeNKAW/VJcIG6jsTdWU8xeCaYrEnKzqtsGVb9qyirGlG/BevqZ5?= =?us-ascii?Q?GCQ1Z7NMfgmZ3ASsacJoLYZ2e5PbG6jpt/cXs5qCeTvYTGJFGGJN1TE34q9k?= =?us-ascii?Q?5GSeWRjWHlswwPbe3DU1X2iPV8N4W1+7e6MVc1+IOA7CCcXXlF/809D6WjDM?= =?us-ascii?Q?ZL0KrytFIxPH8XYsPbmxYfIfjbH5xcqqxrbrUDqXwJmY7ocKORskoVqBZ7h+?= =?us-ascii?Q?MJfSIkxXqBBBJ1ts1Crk/tnvJj4dLfaSBIrNF3lwxIj/orI2kJSLiT9jqeFk?= =?us-ascii?Q?59er6dXMq3XrScjFcLfCu9Dj0hcHowDG8E/EkoFfCuhir/NU2g5qnWOY7kWv?= =?us-ascii?Q?avHMPnIUkc8Q8qTD5Bjmu6iYCWBTV6EeiWoK/2aZ3L7acz89nyp6izqdnH8P?= =?us-ascii?Q?uyNastNfLSbmPLNbwVG0vWG7hVhUEmBPaa9rRgG/aU4SsE9i1SykNIr+DBt+?= =?us-ascii?Q?3a5kl9DU2o2ThjQZdp07d4ByrycG8fmj3kMQJZVynB6YxvnmSyPKJms+MyrS?= =?us-ascii?Q?9uVQCzMBYNnomGSTLryohlxTeo3XX5ZSdNqJ2GBAsc/o9SzcdADaM75Jp3kL?= =?us-ascii?Q?/0aFGOzaquZljAnlviyEJxbb+DFGa3wJer1blIulfNPAr7YaRL1y4qLDAPVL?= =?us-ascii?Q?tj8K3A9+cXgxBqa9+Iw1kDbpYTI1CUJ3RC6CTB8AfYL5hQD0OBk1Atxg9J5S?= =?us-ascii?Q?O8N2exQyAypqg/3tUjgNYWLkf27wwkprwFWG3pvqJzwaQq/csupcFxnReVjp?= =?us-ascii?Q?wKtFxMV1T3TBaqNE7lTyGetgC06/jcZ49uHPS25gtk6NIiYyEuyTEPHDIWrb?= =?us-ascii?Q?MfHXJhSEeDj4dwAPGU8YallMDXBpfga/q818mqEx0ZxGH7kBbQKE5Y178MNx?= =?us-ascii?Q?FqHTpbU98A=3D=3D?= X-OriginatorOrg: Nvidia.com X-MS-Exchange-CrossTenant-Network-Message-Id: 6a51e531-defe-43cd-3c14-08de84cfdd48 X-MS-Exchange-CrossTenant-AuthSource: LV8PR12MB9620.namprd12.prod.outlook.com X-MS-Exchange-CrossTenant-AuthAs: Internal X-MS-Exchange-CrossTenant-OriginalArrivalTime: 18 Mar 2026 09:22:23.9857 (UTC) X-MS-Exchange-CrossTenant-FromEntityHeader: Hosted X-MS-Exchange-CrossTenant-Id: 43083d15-7273-40c1-b7db-39efd9ccc17a X-MS-Exchange-CrossTenant-MailboxType: HOSTED X-MS-Exchange-CrossTenant-UserPrincipalName: EPnmPjyloj6Z3E3leuW2xAa5fneBeofxYPsARSNEljV98jHoKaCUYJUl6zKHlnK7OtzL8yxUfiGtprPxYOPMvw== X-MS-Exchange-Transport-CrossTenantHeadersStamped: DS0PR12MB999105 Content-Type: text/plain; charset="utf-8" On systems with asymmetric CPU capacity (e.g., ACPI/CPPC reporting different per-core frequencies), the wakeup path uses select_idle_capacity() and prioritizes idle CPUs with higher capacity for better task placement. However, when those CPUs belong to SMT cores, their effective capacity can be much lower than the nominal capacity when the sibling thread is busy: SMT siblings compete for shared resources, so a "high capacity" CPU that is idle but whose sibling is busy does not deliver its full capacity. This effective capacity reduction cannot be modeled by the static capacity value alone. Introduce SMT awareness in the asym-capacity idle selection policy: when SMT is active prefer fully-idle SMT cores over partially-idle ones. A two-phase selection first tries only CPUs on fully idle cores, then falls back to any idle CPU if none fit. Prioritizing fully-idle SMT cores yields better task placement because the effective capacity of partially-idle SMT cores is reduced; always preferring them when available leads to more accurate capacity usage on task wakeup. On an SMT system with asymmetric CPU capacities, SMT-aware idle selection has been shown to improve throughput by around 15-18% for CPU-bound workloads, running an amount of tasks equal to the amount of SMT cores. Signed-off-by: Andrea Righi --- kernel/sched/fair.c | 24 +++++++++++++++++++++--- 1 file changed, 21 insertions(+), 3 deletions(-) diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 0a35a82e47920..0f97c44d4606b 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -7945,9 +7945,13 @@ static int select_idle_cpu(struct task_struct *p, st= ruct sched_domain *sd, bool * Scan the asym_capacity domain for idle CPUs; pick the first idle one on= which * the task fits. If no CPU is big enough, but there are idle ones, try to * maximize capacity. + * + * When @smt_idle_only is true (asym + SMT), only consider CPUs on cores w= hose + * SMT siblings are all idle, to avoid stacking and sharing SMT resources. */ static int -select_idle_capacity(struct task_struct *p, struct sched_domain *sd, int t= arget) +select_idle_capacity(struct task_struct *p, struct sched_domain *sd, int t= arget, + bool smt_idle_only) { unsigned long task_util, util_min, util_max, best_cap =3D 0; int fits, best_fits =3D 0; @@ -7967,6 +7971,9 @@ select_idle_capacity(struct task_struct *p, struct sc= hed_domain *sd, int target) if (!choose_idle_cpu(cpu, p)) continue; =20 + if (smt_idle_only && !is_core_idle(cpu)) + continue; + fits =3D util_fits_cpu(task_util, util_min, util_max, cpu); =20 /* This CPU fits with all requirements */ @@ -8102,8 +8109,19 @@ static int select_idle_sibling(struct task_struct *p= , int prev, int target) * capacity path. */ if (sd) { - i =3D select_idle_capacity(p, sd, target); - return ((unsigned)i < nr_cpumask_bits) ? i : target; + /* + * When asym + SMT and the hint says idle cores exist, + * try idle cores first to avoid stacking on SMT; else + * scan all idle CPUs. + */ + if (sched_smt_active() && test_idle_cores(target)) { + i =3D select_idle_capacity(p, sd, target, true); + if ((unsigned int)i >=3D nr_cpumask_bits) + i =3D select_idle_capacity(p, sd, target, false); + } else { + i =3D select_idle_capacity(p, sd, target, false); + } + return ((unsigned int)i < nr_cpumask_bits) ? i : target; } } =20 --=20 2.53.0