From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mout.gmx.net (mout.gmx.net [212.227.17.21]) (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 CF87B3BB115; Sun, 18 Oct 2026 06:49:50 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=212.227.17.21 ARC-Seal:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1792306193; cv=none; b=XlUZB3KdpxMR/KQBqOg7gjmt0QxtYOAT4D7xOV5P5a58OwbRgXDH0h+419Hpf6q12/otcLbuK8gT9VP2DWp/x21sB2FyiDrbbSD9oUzbX0D7Y3RWraIVc6/UFAWvRjerJXq28RKaqjWgNAJDcaUwJQhf9Q58eUH3drUtfDSACNM= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1792306193; c=relaxed/simple; bh=q/ntWjpRoBwAFRXgvSzE5667UQ5q5AK7Ik5KKRUhs2k=; h=Message-ID:Subject:From:To:Cc:Date:In-Reply-To:References: Content-Type:MIME-Version; b=gFyCV2Qu/t0d8UJf9dQn33omNDTZJWn4UaCPDDuQkCdkAXgf/gsaDCLQFLxYMx1kVmc3qoaWdmpS4beB+Vih2UVPT9qpz6rIKk5l09L1BXX8BKBNzvo/HV5vV2j2MS+D9e2fegjZDvA0s2iO6o86uusKZT1gUZfxaBd9oMpyuFY= ARC-Authentication-Results:i=1; smtp.subspace.kernel.org; dmarc=pass (p=quarantine dis=none) header.from=gmx.de; spf=pass smtp.mailfrom=gmx.de; dkim=pass (2048-bit key) header.d=gmx.de header.i=efault@gmx.de header.b=Ff5qagFP; arc=none smtp.client-ip=212.227.17.21 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=quarantine dis=none) header.from=gmx.de Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=gmx.de Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=gmx.de header.i=efault@gmx.de header.b="Ff5qagFP" DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmx.de; s=s31663417; t=1792306174; x=1792910974; i=efault@gmx.de; bh=a9qFg4rh5rJY/94lKvbhTGLmNg2V/yvXn7/WIyaspCo=; h=X-UI-Sender-Class:Message-ID:Subject:From:To:Cc:Date:In-Reply-To: References:Content-Type:Content-Transfer-Encoding:MIME-Version:cc: content-transfer-encoding:content-type:date:from:message-id: mime-version:reply-to:subject:to; b=Ff5qagFPpkA+0Jlj9rmsc35uY1Pp3VVaXAN7SgCbVRYxtSqeoOKj6topdYmV4T/c ANGK4t/IFVMQCBLhodvKcizh6nKN/LD0JJzLtJJ4AnClAP3GkN59KjPlHQrwCvdHf 1bEb57VJK/29fXmNC+w4Oy6BZ/BYmHVL7l46ZqwpbDnUnJXxpjfgbyBGBsOd8jamS lOsMZoi4rr53Hq78dX2GE7/2WZIPLIRajwEOwoED99jNCXHVpYfEuyWcrzLax8esI YoePmeL3EoFxT8OvNP4pvbdZnTxs5b/508zfLvinDrwdQOEOHVuBr7Gf5LpjCxHgv scr0A3c9Cf4JvfyEBw== X-UI-Sender-Class: 724b4f7f-cbec-4199-ad4e-598c01a50d3a Received: from client.hidden.invalid by mail.gmx.net (mrgmx105 [212.227.17.168]) with ESMTPSA (Nemesis) id 1MAwXh-1xTYyV2oQb-008pPw; Sun, 18 Oct 2026 08:49:34 +0200 Message-ID: Subject: Re: [PATCH v3 7/7] sched/eevdf: Move to a single runqueue From: Mike Galbraith To: Jakub =?UTF-8?Q?Oko=C5=84ski?= , Peter Zijlstra , mingo@kernel.org Cc: longman@redhat.com, chenridong@huaweicloud.com, juri.lelli@redhat.com, vincent.guittot@linaro.org, dietmar.eggemann@arm.com, rostedt@goodmis.org, bsegall@google.com, mgorman@suse.de, vschneid@redhat.com, tj@kernel.org, hannes@cmpxchg.org, mkoutny@suse.com, cgroups@vger.kernel.org, linux-kernel@vger.kernel.org, jstultz@google.com, kprateek.nayak@amd.com, qyousef@layalina.io Date: Sun, 18 Oct 2026 08:49:29 +0200 In-Reply-To: <58dc8e8a-7260-43ce-aebb-8d9c29a6afc9@okonski.org> References: <20260605105513.354837583@infradead.org> <20260605124052.227463677@infradead.org> <58dc8e8a-7260-43ce-aebb-8d9c29a6afc9@okonski.org> Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable User-Agent: Evolution 3.56.2 Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 X-Provags-ID: V03:K1:9iNAocpG5x1CJAmMXwVqsaAqdzptXRDOt1yko+dKLlbY5IwsXZV oG4luLShD2wb+EaulXzYgO0utVPpU7zbGV97FQD1NHvNoQe+bmjWghaVslZQ4JFumYWxd5M 5X3eAJT2dU+PMUqdpPV/I6JasOJcs2oeQZpOlnVcMt4hrUS+h/aZPQtO5GxJAzw0cFCW9sZ We0UOdBc88kmCYETUmH1w== X-Spam-Flag: NO UI-OutboundReport: notjunk:1;M01:P0:yaoA2VtcRP4=;4mGPFCdxrx40IOU2VngsZhU1KuW Nph/vB4Y9wR7aKfPpZFofXBcXxO9quUHswkWb2kavXAqjksyDFYeAZHk/zCFHdFk1tTJjOGU4 psd9q9aI7QcmrIPoJI+FDcdAtWxLT3HHXqjvv1ewrOKn0DmG6cRoloWO9S8ein0hX0tA0O7YN CpApAnhgZJ4X2RDn5Q0cMP9aJZsgviibg1zmCvbl8gLXk/vzwxTKZqoLBj6JbYPTKHCwv+ALp Lk6NX/gH3quV1CqLvjv2ah8jCMbz+92cOp7F9BwBS8jd5WazuF64dF6wwUEFK9cAAznaB4Ug3 GFlPpvJQQtodEs5FeSLkFiBb/x2e9agXr8Da63UyZ6Q33JIyKrbVzy4JYavTrdDpuRy/TsNfI Bc/oZO8ByxZDP9G5PTEjAZqhf6+QtEG6E2btpZSu1NC0A0JPEw3MFBX/tfED/MQErJ3niotD2 vkEoXJ0mqSqqjyZeuOObSITDgaYRwDwrY3UGJzeLP+rMiW8Fcjfedkxp9LMpKpZy6tzRhlja5 nkw2ICjfjhw+kF3onttGbHQkBMzL5QTWVD4Vw1ko9GtA+JxiBnGsLGI10K5MwLfMgS3zLN9hQ /ROphbsQ99gk3XqRk2k1wRdSh7QQ6TqFjaFsiXYqvxK3JTHZIoD4jOO3hyzW05xoFhkp8x2xj xo9kmNZrA0ikRdtmTCtsfDjhPRnLHshJdAo3T43s7gqmvTDeZSYlADeBqujaJy/iMUo8LQi/v 7LVLIUOG3LQcxcCUCCXNXTemcpfzLc0Co1aKF3xy4Ovugh+EH/PDFh+m1dE4/BL+7qHfog6xZ kjZOQm4yTvuifqS2M9J+s0jjnzBG+E+MUz3/IGU8ConlFM22ZfSV763KQVNItlEuwdp6c7Pd0 C/srrwhE2/UUWuzZF8E5z5jjsw3BbBYMGu5XYx9exSVNEFW2NleDtI2SKrWKp24rHxQleVFYS I832RybXqixONtLV2RYUGyNkQRPJHdAI58YHLjAdxVCUiX81tVMK345c+MgLzVB/kyYQ9gCWg dCmkTCqnhRQR0uXYmTFbXNy9rh4SErJgtLGT+9jHF5rc/0XZFovkQU0qJxmPCX+t0jkt05l1+ JwHN/nGBC9Z6PsNR1k7enm9kBUgbnpZA1Umjwtf4kOd3kf0K8e8QEV5tgBMQ4XM79yEaP/NLX ntWzwAGpwTyrNFRMbe3SXrKq2IG0KRoeDqgSJtemssbWlXycRzC/EBftK8g8hNE26tCdwzmDF 3LrlqS+HB5u38x+s9ir6asXcVDQR63LIeR6hyQg3273PvsdHLffXqBSlSMN8KyjFtsCbVNObU O+dPWVeSoVGB9Xn+Y/Os3+ow4/u94F7Ygw36tpNx9LTsRlx7UkPxFbz/wkUqeIURQ8ry2WOr+ TrfvMXJ4qViHmSex6z0v4ITl8kWS/nIzKS1H9bMWAqEuTQA1IEhgq2A76zSjvRI6Ke35TS+01 tUjx4gJ1x3CwvhH7tRyyXUj0XD6FQQppUjbss6T/BbZpG4bFQQIOMWrsuWJQtp/ZAE6wcE+e6 oBZvL5hZX/OPAjnUDf9XdRXak2zx6bXcb/6c6FQFAwWmAxtFWBCj7Wh+NbnYYdVE2OQz0ECz7 xErMPNEwdQzUmAuiIFN+1SG5MvP3fT0jJ5SWhoD8Yx4CIr9nCqvv+bJl7VxL3zM83VeD6d75Y dCWMi9qiV/+jMMcTQJfpHU074G0zen3rpND1c70xP5hewL/7z+zpAnxw7rtDFP8X2RWkIM93l AaU6YU1e+9RX37lWINtLBnP7O3Id108n8UgU7/ZHxjYOnkg6jZ9vJam1Fx84QV6A3lCGNTKi/ 4xq5kmLO59DVzgw5BMvnUd1yKSPMSOuFV2ro+wfPyZMHzjzqrv30RdKgxOsrt2kET5omKGtqO jvJOU0o8bOD71/nZjD9ocuqPbZNra6MsjNEerNOtUds4gSwlc0pEYDRPgVUy7Seb3esYJQaz4 Rv8FWfxQzslqPuBuu4cu2WNo0dDNhAvsm0vt0Z7vWmeyu4qJT11Kg1Bykq4Sc0RsCTkER+3ug pdRx+WkpfocA0qUTsCkHe+U11VmUsfTRwPNSUxyBIJhXFQBN1CUPdN/pAc+1vvwLza5k/bi26 PWbJu6Yi2m22KVyjO5WK1WDhoHxFz93nyXa40BggUOIQYXEe3bMoYJoC17YDRxMFzEJXcDFhX 8krpkTbPg0qyhymdAVNxhz6UIJHtZ8AfRHNZM8COstWHqF5tclL4mLtfCpmbTbj9RCiic+JPD gUANh5KxUcsz1Nkvasu3O3PTH3tVndy+EzdolnUAYI6tNvXTPaAbk9GRD1Yhm3NU0+3dS0Rv+ 9W0qhG/jJomxHaKWLVb56/nF0YEr4SJw8D+TzbELOy8LwbYwGvPK2VMi4HT8uEWPR5ZdEF8x+ bl6HN9fDT4UinfwKA9dh+igEaiuAHyAFgINdYeiXJ1xhVsoD6e7IxFuEvUDnanpmEtSrQppR7 Ofd3Ar3biiLkS662BnEGtEJu6S/buY1NN97vKv3xdw+PMJvAyuNsPtQAGK34VGgfAi5WlEMks z8oq7x9O7fllK4kdlS4FnfE+8mNipaT/6jlIDPq3woVsqW1YKM56X28FGb/Z17HhW1qP2r5Xx uZSf/PtlQySNWBjKoi9lolPnEjvEadwn7g87H30sRhRQ1wWJY0NDCpM4BkFqQ3xglyHaWmfe4 cPp49RhKpVEsvMDqYp8eRbh6N7dXI4x8gV+5WNdrz9MElXEoiFXnSVR3QEZZFudqKQMVvI7DE VObRNMrQjexTOryI6WAehL9IeNEZLEHe0CVZD3AaWgInlhN3i7PQ40bCqsaUaviTQ4tJ3R8xu oh9jSr1xJZ1lfMm5IC+VPUWeqgIh7QnFnKF9QXI6/dqnalmvzl4KQCZr1wkbYkTqh46FXjAma 49gtjdRMCvFCXT0gvhcPwnJ3u9JuF+E9Uuo+p21eWGx1w0RpW8I9HT6b8xC2S+UwKk0dgxDKq LeiRg/K+eF3NEOFhRT2Vg8ofp9xdFdO11Z7o8jh5nDIETcHO/vuGpk3F/ILGJ1lv/8t/VXCak jUg9uOHbib6victbfZm4P1a6AIx+Ypgqx3M0sBTbjv79YGBrt9nMtwukNgFDbsc9tOajzp+00 uBYO843qIpKdamG39JQc3y1J9JIRuYL5F02VXPnowDqwpx6OJt+4zqK8qoydLvluxAVUmQJHo jW8GW+57+p1RPUQe4qykfXj53aR+Cnezg7EP2oGOJa3Mz97EzeAIpFY/vq3V9+z9a6jvRD63n eR0MLZ1CCZILnKPCp7glJXvLgLuHZ3kFSkP9zidY7zjkJBSL7lMaBmF68HJhEgQmEFcnEdRIa jkO3e1tPL5OnAUQ+Fc8aOsOtFrmSba4t7VBcGb1f9EXhKQyRTWWm3lPwy00LZqHWkexHM0sY1 sneb137yNNZ7IKtInrt/Uqm64ei24nQceT0S8keCNTbhOHb8EcVYWpNYVIXoglV3b9Uj+rk16 MxiLg66eFkF7Gp3Ay6OVQN0wBgK0XxJeG0PFZxBj+73n4/dJBVrmQ9I7V0fKz4h9nHaVCAmtY OI8YTl8qSi56Fbdnz8af/2yjynG4I0CYuwF3+uiaFTWkND9QUty4mffkTDHaauMnSa5abm0K1 5zWFzMXAPELjaa2TdNXlTN2YRSFCdqCUhWyNuIEfhnWBTQeBs6jyaeHDeP1y+P+KwSP7ICdTS T6jGuOa2U7R/AHgkomN2BlL5lNn52mYzaSTd5IlmuyV6He29s9XDrtlKEfSWs8srIq4+GupeM UY/rEx3BsKftXK1CfZA96jo7B6wfIa79p5O/lmKmwfZjzhLVaFOKe2n3TFHqjsuS8SJATAjNL mC7l51OVSFbgNr9RQu8krBQBxqTkqn0BcfcmKRanDog8d7xI0So78cahS5MA4OCcywvioFg56 gzDJ4OUTLrm9CARChifALVmE+yZu+pYWlZzgDAuzuP03elA0/hJmVDXrewUe4UW023CHAmZSq jYuYnOFlqpNYMjLH1/kH1t0MxBat4do2B+IxumPJoCW2qk73ixiebW3N2FPxT3CsS9wHLkfcV xQOYLs0+j1IquuyeWkPP2lJ4okaZIMHTy+02ISxA7DSz+Z9HZnY3VjWNna2CNSLdG8BpEjkW4 +LXRnbNmGh8BwoSzYXEn308o82ZIIVg6+kcwM8A3D2lz4glk/4PX97PmoXQMQMuc7Stc9c69/ NAxIyztoiXL6GU+tokHqNI8rX//YAJtBxuhC4ZFxJSRNUMlCBY+3HnDneTf7kC3dMXzeWHcNr JfubiK3bDMaOTSSRHPYTRhPa/rqU7U0sXVpGDYOrpcPkJXc5jxTYIZk+lpbCCmgCIqBNf5vpJ wfOdheJsaRZRCZeMIlDMutsFECp+4kkgJ8nlhVTd32724qYR13D+JHG2xHyYJ3R5PUJOx7Pkv Er+1oUby7aGeN/BxNk0i/pWOcYxAevYxnr6R6keeYBReq+YQhp9+DJ9R9lDpPCQ9Unbku5OXi MA6aVkgv4Ijb7Dhsm2c4bWAlCf/6dj6u/Wr24ih32STwRX6TjSi8aYarnmXVaAPA59ptssx1r as2o/zrns5Ckvwwyw0/f8uYXXLEgFvN3mjg3Kn6IXwom0+0fwTjQy8R3d1G0ViPFrU+K6TgOz YzJLG/5AkBpVR9SMZhgR1R0QiMYrAYvWz3RGG5Ezuqux2Op8oe2zv0cxx8N6u/fMuS0h2YJ/H ewjyz9VK/EfJcf0qLdtVAZmTXLKFedrI91QQ/cMgmbLf4vS4eLiPsDfHd9tp5kPffolrZFC5M 7u+QQ8kO838MhMCIBqhTsfF+h+gofzrEnflgcHf728qpDTpy3EmmL/UGgZdfvmPGRFz/kdmo1 Xd0jAZWOue/x7G8wjS0kY7yY5IiHOPfIN9eMan1U6Bk4aGtPrvKBv4HXu+1bH0QQiBF+LcKRl cME/PEgwvQEbaAt7Iw5IspkBF1o76laR7t4xWYpllhJL64mhASdNAy9QQ9PnHFbEBymTdQfFR NCG2get84ME4RP5w5NKdQM2lPi1U7IrUglh63iULihxMzE53/5DwXe6l8Lcm0SG4y13mJncrL Flvc4soFlmMp06CiUTU/XRpmQ3P5PdSE2XfsjZl2u5TUUSa5XkMW0tbv2tz1PhrT7J53tvExY ZQch5idsyZ4+4Kg8dCQhzSPb8Os9QvqR5s6O3VjS/cPAaVIpoCBYBuOikQECTDVFOwKCTrzfC mUi1L541W+vU/JUiQt9vbJMtbu7DFoCuvoIWL4FQ7w7jvuddy5KyAkne0HwSn+wl+pCETO2uG k+UzExA1tLYLxyh4ukQbk1bFZf69vo4FTIbvsiqBTmSm5Lgp1PJLwAckr6p6BtE0Ycf2XhcnE Q+OFQuA2FB4OXy/NP99gNGvnXtY6srlZJGgRY1a7L9FJNrz4V3aNxTesFriOmfHpypwTj1ELv PtU6kG7XkrfN2jy+LfEI9k3JUedZHKzP4/xZYrJ9/kFx3hMZuS1JLkDcOSUonkbuCFisNBCB7 9zPt4rXElUPKLLxqRnUWkSUVcrUr7sjDlQlAJtXwFWUdvv4EBCRsXmD4Sss44rSyhbNYSMyfU GMzmSzSq/J9M1pe6bwYFV4ebyaoqzs/xbMxP03c7aarsKu71u/pI19TgGAW8pSzdiz29FeFdp Qw0F7lick/RJL0ozRKs3UQKX6M2GfTJl0RBdEZyTTLg16D95kT/IsKDkSY1/JawEw4Y6fWkW2 YWjubAdk0CFmjKcNTuLmZTzB/YuwvKP4x8kmcmzrqqbwh71Kkuovum6yQDTw9UoVlY2BuBBIP wyA3wUlzLRCf3IUUWL6d4bcemenJwCzmSObe68oCXXCw3dYBpdoMPMAZB9AkBFAzH42OBjnaW uZVWKcW1kJrBEZgJ4zAJK5uJapIMOtFCDAwJMINq49rUUHVFNv6I6BlxCQxgH52h19/IKu7Hd N1YtOReRoWXSvbaRlbCSf+BAX1lqR4lkq2oIKTfQfTwlP9uFCcG9h4Bqdt5QAtUCGU7+3Q2m3 xDHoPXT5UQjF7tu5aThscXdwquJQjhr8ReUjQmYtaQ5e5EsFGQR78Jen8fY5TbYbw/X081in8 C6y8qAZqgbjTfouPxsEwnITAPAkUu4rqtY1IVu8g7IedfTFAoCINnndcG0HYvPZVVVTG8tbmE LhKCiE9fKFoVCbGyDws7KtQDT+zDfpMuhLz8BAq7vdVhzKgvGL/TZpVKwIaBx4TMR/zyi4zmq sSaZECUPRrXmOFlOPBbEDKOReMbxdS1af/942hgPzUqwmwbFdb+miFiBG/+AVB0Q3Cz7w9J9A AaWXBxx9kWOe4CB+1h09tN9chRwaIkYeKhofdSI41i0vEgikZF54jxKNrsPStW/tQgnQ3GLf3 lnYa19jTurQwo7/fJHBwtcsDm3MNYkNMclgJWdqnYptE1vEBi30SdiiQZFAyU7+O+18POUy12 OTNeCskqIUcwaFCYbVMZDiW1PS2lGL513aUeLxlwRQt7I2Madpx1dWHiCXXM1MDmVWzXZ62jY Sqxl1D930Q1hbD6kKnS60QoEN/FfeoZvtbntkM60lpVcOEXr/3Kk+xO4npFnuRW9turdudPZY al560AGVFs79ItVQ6vLHMfflUJvfI= Greetings! On Fri, 2026-10-16 at 15:18 +0200, Jakub Oko=C5=84ski wrote: > Hi Peter, >=20 > On 6/5/26 14:40, Peter Zijlstra wrote: >=20 > > Change fair/cgroup to a single runqueue. > >=20 > > [...] > >=20 > > --- a/kernel/sched/fair.c > > +++ b/kernel/sched/fair.c > > @@ -9916,7 +9837,6 @@ static void wakeup_preempt_fair(struct r > > =C2=A0 if (!sched_feat(WAKEUP_PREEMPTION)) > > =C2=A0 return; > > =C2=A0 > > - find_matching_se(&se, &pse); > > =C2=A0 WARN_ON_ONCE(!pse); > > =C2=A0 > > =C2=A0 cse_is_idle =3D se_is_idle(se); >=20 > I think this changes how cpu.idle affects wakeup preemption. My box claims this to be unrelated to the jitter you're seeing, as it is present in 7.2.11, with and without the sched-flat set applied. > With the single runqueue, find_matching_se() is gone and se_is_idle() > reduces to task_has_idle_policy(). Wouldn't this mean that SCHED_NORMAL > tasks in a cpu.idle cgroup are no longer treated as idle for > wakeup_preempt_fair()? >=20 > I tested my existing 7.2.9 desktop against 7.3-rc6 with the setup below. > cyclictest doesn't reproduce it because it just sleeps and wakes, no CPU > burn and therefore its accrued EEVDF lag keeps the latency low? >=20 > cc -O2 -o wakelat wakelat.c > systemd-run --user --collect --unit=3Dspin -p AllowedCPUs=3D7 -p CPUWeigh= t=3Didle sh -c 'while :; do :; done' > systemd-run --user --collect --wait --pipe --same-dir -p AllowedCPUs=3D7 = ./wakelat 1000 15 200 | > =C2=A0 mlr --inidx --ifs space --opprint --barred --right label lat \ > =C2=A0=C2=A0=C2=A0 then stats1 -a count,p50,p90,p99,max -f lat then renam= e -r '^lat_,' \ > =C2=A0=C2=A0=C2=A0 then put -q 'emit mapsum({"kernel": system("uname -r")= }, $*)' > systemctl --user stop spin >=20 > wakelat wakes on a fixed 1 ms schedule, records how late each wakeup > was, then busy-loops for 200 micros (wall clock). >=20 > +-----------+-------+-----+-----+------+------+ > > =C2=A0=C2=A0 kernel | count | p50 | p90 |=C2=A0 p99 |=C2=A0 max | > +-----------+-------+-----+-----+------+------+ > > =C2=A0=C2=A0=C2=A0 7.2.9 | 14998 |=C2=A0 11 |=C2=A0 11 |=C2=A0=C2=A0 12= | 1869 | > > 7.3.0-rc6 | 13853 |=C2=A0 11 |=C2=A0 12 | 2029 | 2116 | > +-----------+-------+-----+-----+------+------+ >=20 > count is out of 15000 periods. rc6 had 1147 wakeups delayed by over 1 ms. Thanks a bunch for the proggy. I've been looking for a clear signal with various loads, and stuffing wakelat+awk into one of my scripts provided a perfectly clear signal, which I bisected and confirmed. Note: below is box pointing it's little silicon finger at the source of aforementioned jitter, not necessarily at breakage. Reverting the fingered commit removed the jitter from 7.2.11, and force reverting 68e37487810a + 333238da9a19 removed it from master as well. 333238da9a193ffc58792995f3e951e4cb87bfd2 is the first 'bad' commit commit 333238da9a193ffc58792995f3e951e4cb87bfd2 Author: Vincent Guittot Date: Fri Aug 14 15:52:41 2026 +0200 sched: Update time before requeueing delayed entities =20 In order to compute the right lag, it is required to update time to 'no= w'. Without this, the delayed entity might appear younger than it really is= and receive less compensation for having waited. =20 Signed-off-by: Vincent Guittot Signed-off-by: Peter Zijlstra (Intel) kernel/sched/fair.c | 2 ++ 1 file changed, 2 insertions(+)