From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from kanga.kvack.org (kanga.kvack.org [205.233.56.17]) by smtp.lore.kernel.org (Postfix) with ESMTP id A8C10C3600C for ; Thu, 3 Apr 2025 08:49:00 +0000 (UTC) Received: by kanga.kvack.org (Postfix) id 7BD0F280003; Thu, 3 Apr 2025 04:48:58 -0400 (EDT) Received: by kanga.kvack.org (Postfix, from userid 40) id 76926280001; Thu, 3 Apr 2025 04:48:58 -0400 (EDT) X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id 549F0280003; Thu, 3 Apr 2025 04:48:58 -0400 (EDT) X-Delivered-To: linux-mm@kvack.org Received: from relay.hostedemail.com (smtprelay0012.hostedemail.com [216.40.44.12]) by kanga.kvack.org (Postfix) with ESMTP id 2114E280001 for ; Thu, 3 Apr 2025 04:48:58 -0400 (EDT) Received: from smtpin01.hostedemail.com (a10.router.float.18 [10.200.18.1]) by unirelay06.hostedemail.com (Postfix) with ESMTP id 21536BA5DE for ; Thu, 3 Apr 2025 08:48:59 +0000 (UTC) X-FDA: 83292107598.01.B557DB3 Received: from mgamail.intel.com (mgamail.intel.com [198.175.65.15]) by imf24.hostedemail.com (Postfix) with ESMTP id 21DDD180006 for ; Thu, 3 Apr 2025 08:48:53 +0000 (UTC) Authentication-Results: imf24.hostedemail.com; dkim=pass header.d=intel.com header.s=Intel header.b=epGmKjAW; arc=reject ("signature check failed: fail, {[1] = sig:microsoft.com:reject}"); dmarc=pass (policy=none) header.from=intel.com; spf=pass (imf24.hostedemail.com: domain of oliver.sang@intel.com designates 198.175.65.15 as permitted sender) smtp.mailfrom=oliver.sang@intel.com ARC-Message-Signature: i=2; a=rsa-sha256; c=relaxed/relaxed; d=hostedemail.com; s=arc-20220608; t=1743670135; h=from:from:sender:reply-to:subject:subject:date:date: message-id:message-id:to:to:cc:cc:mime-version:mime-version: content-type:content-type: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references:dkim-signature; bh=TlxGD2E4AeVQoa5K9mDUoaD4f/wVRxsVtfff/Tenizk=; b=DTJToVQvSv/4LSzLRUdZ+/86Eq5dBksmZHfTF5nmGLUSIZEupEuJr0AMqV0W6oRPOtQM6n cSctKwmMuJiuP9ol+F5b4a7fbfWL/oJdEFRZRhXU/WMEKRMVtwZgBFK5HDGaVARpyilKcq hRPenW+G5fv6YguXa9W9uWP7/UUT4Z8= ARC-Authentication-Results: i=2; imf24.hostedemail.com; dkim=pass header.d=intel.com header.s=Intel header.b=epGmKjAW; arc=reject ("signature check failed: fail, {[1] = sig:microsoft.com:reject}"); dmarc=pass (policy=none) header.from=intel.com; spf=pass (imf24.hostedemail.com: domain of oliver.sang@intel.com designates 198.175.65.15 as permitted sender) smtp.mailfrom=oliver.sang@intel.com ARC-Seal: i=2; s=arc-20220608; d=hostedemail.com; t=1743670135; a=rsa-sha256; cv=fail; b=kDnP8HixsGRdXCg9FSF84A4T3Ouqxz6J+ojgx9qinkjde2MHZXTRCEu+guvCHeCFry2MdX CAAJBLvVipuMQbZQdPXniG/AXjpciLX/Li1IPKfiaSv8Po0SgiSz0/ANtlvdc0bSZmLzYg xi/B01ny8xNW2VajkFIq/VzgOPg76/w= DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1743670134; x=1775206134; h=date:from:to:cc:subject:message-id:references: content-transfer-encoding:in-reply-to:mime-version; bh=O7weTwxEd0F4QukTm+KtcRKBC6+4gcik3ns0u5q/AXs=; b=epGmKjAWAqFfB1F5iRswe5x+PM4Hw/xA9Yrg9tME17ADRVdNi/qefjF2 OSJ4LRKQZxOFIY9rB5HTnItBY4wcpuO9HOaU16hLc2N4z9RiKqvmiZMue sXdojD5/N2zIR2531JS5SFojkv0a6muAgNYXgjo92XuF7/qyE9ChyJGL2 ZiIlMAAjMbCAdk/Kvy7ROe3u5gru+Uc/pBL8p5Rrddy9RSl78FeHSkg5K B66jY/jEmvFeAs7RthNepkisdV1QlztAeCdkYKCAIuKi8UsR/5TjbNkVU tgGn8JZv8fxcvuFIJJQ1Uqp19OBBLG9e43NFLBqw0OKLJblPew7eooxkl A==; X-CSE-ConnectionGUID: iephDZXAQxKGCTRmUsJfbA== X-CSE-MsgGUID: uxpKyOC/RP+xldBxg8iJEA== X-IronPort-AV: E=McAfee;i="6700,10204,11392"; a="48726790" X-IronPort-AV: E=Sophos;i="6.15,184,1739865600"; d="scan'208";a="48726790" Received: from orviesa008.jf.intel.com ([10.64.159.148]) by orvoesa107.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 03 Apr 2025 01:48:52 -0700 X-CSE-ConnectionGUID: Z/Yt1of8RcWKre86s2V0yQ== X-CSE-MsgGUID: NsjDmMxbRNS5orvbH3gaag== X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="6.15,184,1739865600"; d="scan'208";a="127861980" Received: from orsmsx903.amr.corp.intel.com ([10.22.229.25]) by orviesa008.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 03 Apr 2025 01:48:52 -0700 Received: from ORSMSX901.amr.corp.intel.com (10.22.229.23) by ORSMSX903.amr.corp.intel.com (10.22.229.25) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.2.1544.14; Thu, 3 Apr 2025 01:48:51 -0700 Received: from orsedg603.ED.cps.intel.com (10.7.248.4) by ORSMSX901.amr.corp.intel.com (10.22.229.23) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.2.1544.14 via Frontend Transport; Thu, 3 Apr 2025 01:48:51 -0700 Received: from NAM10-MW2-obe.outbound.protection.outlook.com (104.47.55.44) by edgegateway.intel.com (134.134.137.100) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.1.2507.44; Thu, 3 Apr 2025 01:48:50 -0700 ARC-Seal: i=1; a=rsa-sha256; s=arcselector10001; d=microsoft.com; cv=none; b=mQYKToBWlBoRP4I1dVAYq9v5q/DNRoNGp8YWoqlglrm60QyfF5BbBp0PfC0+7vzaphAZs8AH64Jog8NCzcnnyoEB+/vaU+8VBT64PaHLrQIKcO7yJGStGBFHOMSN/noKPl1poxKhR/I0gNl3fLa5Qt4+fDLXXvz7fKSRXcI/m7+YCXpIgwJRz407m+aoZAU4NTd1x6BTij/F44smrxsYMVl2x/9ZiqIxa9ZnhMk34c9AsyTQ7EwpVpauLDI2BGrOgV+lJiXHYqi2Ucw+KyvuxlzfAV30WHEmEE8rC5Vw/yxx/9DUtSvwXfDuwYYtq2Q05mJgQ8Uv9ow8bIIrY7aVfw== 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=j+vqAo+4MEYquFr6W/VRn0Oj6uRloc3yoiaAGk5qHHM=; b=PWf11uuQeFv3I2EmhDy8BIcyz+oAqSno6OzOszQlwc23lyXFGRRE5cJgUq+wCEz6lSxGvpjZMZvt9fZM7PLQ1DiIYImDB+enfX+6tGTmDZ/akUZKCAszaR8lTn9oIRheHOeNV6yotivtsgCtPqixOUNyvIQBQi6yNzrozYuP91bBbNjguSS3pfD2SRwqaHznpF4MF9J9fzcBlGLhEh3IAXi3AibMT8qSoYITm+n+yB/GXa+987OYbX9hPTQY5h45rqhjjnYGTcTXq1PHT5SE/FvkSg5BZKWC5miRBx/XpBiFYHbmun1miJoED9F2K8FzdDolgpVonzYGeO7buk04lg== ARC-Authentication-Results: i=1; mx.microsoft.com 1; spf=pass smtp.mailfrom=intel.com; dmarc=pass action=none header.from=intel.com; dkim=pass header.d=intel.com; arc=none Received: from LV3PR11MB8603.namprd11.prod.outlook.com (2603:10b6:408:1b6::9) by MW3PR11MB4732.namprd11.prod.outlook.com (2603:10b6:303:2c::11) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.20.8534.44; Thu, 3 Apr 2025 08:48:46 +0000 Received: from LV3PR11MB8603.namprd11.prod.outlook.com ([fe80::4622:29cf:32b:7e5c]) by LV3PR11MB8603.namprd11.prod.outlook.com ([fe80::4622:29cf:32b:7e5c%4]) with mapi id 15.20.8534.043; Thu, 3 Apr 2025 08:48:46 +0000 Date: Thu, 3 Apr 2025 16:48:36 +0800 From: Oliver Sang To: Johannes Weiner CC: , , Andrew Morton , Vlastimil Babka , "Brendan Jackman" , , Subject: Re: [linux-next:master] [mm] c2f6ea38fc: vm-scalability.throughput 56.4% regression Message-ID: References: <202503271547.fc08b188-lkp@intel.com> <20250402195042.GC198651@cmpxchg.org> Content-Type: text/plain; charset="iso-8859-1" Content-Disposition: inline Content-Transfer-Encoding: 8bit In-Reply-To: <20250402195042.GC198651@cmpxchg.org> X-ClientProxiedBy: KL1PR01CA0001.apcprd01.prod.exchangelabs.com (2603:1096:820::13) To LV3PR11MB8603.namprd11.prod.outlook.com (2603:10b6:408:1b6::9) MIME-Version: 1.0 X-MS-PublicTrafficType: Email X-MS-TrafficTypeDiagnostic: LV3PR11MB8603:EE_|MW3PR11MB4732:EE_ X-MS-Office365-Filtering-Correlation-Id: 8fd94ca3-d3c6-4f50-cc8c-08dd728c582f X-MS-Exchange-SenderADCheck: 1 X-MS-Exchange-AntiSpam-Relay: 0 X-Microsoft-Antispam: BCL:0;ARA:13230040|366016|376014|1800799024; X-Microsoft-Antispam-Message-Info: =?iso-8859-1?Q?V0/GsV2K9st3JD6adVHZ0zUSloMT6agtpC4LfFBskt24EtGb38M9Q296Y8?= =?iso-8859-1?Q?pDb/wWVZ+qiAgKsTHfyQjBqk9c7KJJaeOuA4lgJkcF3AvnW1r6PtGd7QXo?= =?iso-8859-1?Q?KrokZ3LPx+pwXBUrGTk2d5NrCSzN/9q0Ce818hdyOndTphn+kUpOyaXQpt?= =?iso-8859-1?Q?Vcnooj//fy2t9yGhZeMxg1TfWz+sMLy6q0zEMaY9I7WMu0vhTZx73B8S4W?= =?iso-8859-1?Q?LxiDvmXWrVmCr1RnF4TmS51rvzmSeTypsVJB25a6+5pToppRoSEa4mrmZL?= =?iso-8859-1?Q?cBTDCpf1cmaLaWm9582Jb5ZUr2YQGDPJb+JEClSIToAtzZD5OECSxGNZ97?= =?iso-8859-1?Q?RfZ8HZgAEcKFbsV/dMwlsLGT3sn/SGQyWsh0MfvRBQ6/pTEn+rDLgsGF39?= =?iso-8859-1?Q?AamyO9xBKHWhnn8MM6X/Pplb2jbS9MhfX0DMMAxNHMSTeJy4Dbxnb+CZQA?= =?iso-8859-1?Q?1nGwwSftzNIUNn0xD9MVTBIcXOu2t7FVeHi/YAcGhOM8p3VcbBvj2rklst?= =?iso-8859-1?Q?yABFc2PqZrFKgWd/bx2kSQkvDuzDDQR07vD/LKynu3KWCquzCDPpVNxxCM?= =?iso-8859-1?Q?IJDurBbnSdRre925MrBdk3E3Fo97JtXVxOSudzRjA05WTPH+15wBl3LUAK?= =?iso-8859-1?Q?5rB/wX4Krwu4P8yDSrYkDhRpeT3kahIidJ8LHPAkIhhKVuDJMqBnudIOyy?= =?iso-8859-1?Q?Qg8s/MNuu2czgO5wniJF31L7uYbs9iHe9nUGgggeZsZLQeM2sY6DX2YAqI?= =?iso-8859-1?Q?ZJF7ZhPqenJ7qRWrhvd5H/M1Yt+cxVr4Smx/qxCaIMHTKerHrRDmkpktiA?= =?iso-8859-1?Q?hskIUp7P1GsbVdAIfPzAkdETXQtAtuo4JLnnMTAQgrveqiB41XRG02q2xI?= =?iso-8859-1?Q?tbC7Mm9jGHgumcfEMPIHcJRgyhHM8Gra3A6nUrJT2P5Uky1EvtDqGemNZL?= =?iso-8859-1?Q?aK9N2Ldr/yyXrSBlabjpLwb04QJp5jjcacQoaZ9HhPtwEStk7mAx9/di8M?= =?iso-8859-1?Q?ZohzHkprjV6vX2jfhwN3mvBa+b6fEcUt6VEbQMBZ+a9TL8Ny7UHV5AsJ03?= =?iso-8859-1?Q?nw04JHcr+TPgEaKYnFPGjiPzBUl/fhMiAb0nCQS5JEx3A/FHWqM1Uq1rVs?= =?iso-8859-1?Q?J4i0wMUoffE7K+bthr2A8R42eqdorWVMUkjxaK+7oGCtzifSK6rzzKg08F?= =?iso-8859-1?Q?nkLgcX9RonCN7BPezNlvWSlGb0dNIbLuB9ReLGj0DT8bLIdbDZkEXIqfpO?= =?iso-8859-1?Q?SHfCghuH4aI/jLeHeqNXd2CFjEF4ZiQyfFmY9l8qlMFLEJP8kyKIKTzRuG?= =?iso-8859-1?Q?Av+O8Rl4n97/cLhXoFUFylC7cFpaL/c20rMsfBPAayxFTviBXko7G8k9hV?= =?iso-8859-1?Q?u5GJjpFzvB9pUUqkf5XzQzpviBo69Gx7Bk6F5WbLfrhtgb4xymyXJpdzRh?= =?iso-8859-1?Q?EdMn/oD3sWmc09W0YrggmECrzL9ZyMi+CHybAA=3D=3D?= X-Forefront-Antispam-Report: CIP:255.255.255.255;CTRY:;LANG:en;SCL:1;SRV:;IPV:NLI;SFV:NSPM;H:LV3PR11MB8603.namprd11.prod.outlook.com;PTR:;CAT:NONE;SFS:(13230040)(366016)(376014)(1800799024);DIR:OUT;SFP:1101; X-MS-Exchange-AntiSpam-MessageData-ChunkCount: 1 X-MS-Exchange-AntiSpam-MessageData-0: =?iso-8859-1?Q?EeyeydsPOjJtEHranRJ81MFXgcdA+htfq3/a1PftxPMnuTNct2QGAYL4nv?= =?iso-8859-1?Q?j8wpVcq9TlZyCMMC7JVc3jCuv9sP6WUoJsV/pG39zk9lAy9W7V3O0aDa2L?= =?iso-8859-1?Q?JyOa/6G3nY+Sg2/L2cMMAmCRuhNCqIX9+ObRd/MUS2ZPf6Eaic6dqzhkEL?= =?iso-8859-1?Q?kmeR5NKfN+v05Bckx5o1bLNMaVLX1wT9Kq4LqBHNwrv6LqyZOXQKr9rSOy?= =?iso-8859-1?Q?9fPBUziBFTj0DNk9wxYmKM85jcQkHkRyHhLe74g1wJ8uWUZU20jjOylGNY?= =?iso-8859-1?Q?eQmKT4ctw/azmZGJzck6Z95CBYTeNrMDUDd0LMN3ioDDFRSwFSzyqdsbi5?= =?iso-8859-1?Q?O7pPmklnb3fyYNBrlc6St9SeGDyLGsotxMyCqW8DEObsNVi5M5TpGNe6T2?= =?iso-8859-1?Q?rd7TIxD9qbP0BlNXkm7JsenAPa3coGuqSLxWLW5wCTfOPOwxEdjmyLi3kN?= =?iso-8859-1?Q?X+epdf5ViQXlf2+AYRgXpg0nY5GJL5TXS+gKDkFkzMujFEhbZ9HS2u2i8k?= =?iso-8859-1?Q?Vy6/2kiLvLJAaJ98PynClqR6IWAbWWnuWMlgHVLSwqPTxIHxHoZbouFRLV?= =?iso-8859-1?Q?kN3PmX+qcakRCSyji1C2bRP+zX4fpZGiwg9mreoWBNVju0Y3WRUEtIrNUJ?= =?iso-8859-1?Q?U7eviYmOFFkCim406jemXrjKGwILbdPSH4eeHIbz7I/sM7TbilCraSRzzU?= =?iso-8859-1?Q?ehuPxDTUSxtt3C8hPe9JkdwODLugSXkYEtblqGAYDgbgjiufouxbvccw/A?= =?iso-8859-1?Q?SrY2hLQmZ5/dbxF3FIJlk2yBPzXDjk8SPdxQGxHF54KbR5YV6zS0pgxsxN?= =?iso-8859-1?Q?dKJHbmmwQVnv9So/9DTxBAXoCJ4KZRxhYaYIwgaQ5J0YOQZT/kKgLn+wpz?= =?iso-8859-1?Q?LJYfX6DaYU4YJogREqic7ebi2fQ2PQPJt98ROpBVnGSjV4G/pPp3/Lms6z?= =?iso-8859-1?Q?hF+inI2Kv2UtVVGUZi6Xtv3f0xzl1ROzL6mgWt6s7P4KkiZzE+kuzff8Xx?= =?iso-8859-1?Q?1PyqWfZ0V/DDCK2NpkDIMbUBsuEtYfJuW3CthYJFXCOzlpwzy12MZbdyod?= =?iso-8859-1?Q?g+ZocJbK8SJ9z+7CTbevLOcbhrF67azqFnGmSMrJODtAEtG20EaNoOyJtH?= =?iso-8859-1?Q?6a4TMzuQe+RcYVTnt5/4QLCL5M2KA0kecdxFLU5VNJdWGdpyThC8Qq4T6n?= =?iso-8859-1?Q?CFHWnYREPrYzQJ35uuB5lriRrl0moCiEQSQY0QK9onbmuUnYXta6o0Iygi?= =?iso-8859-1?Q?g9JkDN3TPKCUXlHcUa7ZlJRmw3UIFpBlXFkJOXxdP9sOgU6I0fPce74yx7?= =?iso-8859-1?Q?leuXrKkhJZ1H7ClUB186MjBBdjqnSR6naK0nPFeB6YhwDXbRVXDACmrQkf?= =?iso-8859-1?Q?b29qLf54/vHZUYwOSyEYESk3uai34ToBRYD/QhGDeHc+8Rwj7k0kqPR8M9?= =?iso-8859-1?Q?0Ya24KUgr598Ja3YpBu0LIpbmYKDrho+old/YVSmNAluspXVcWC47/4nJj?= =?iso-8859-1?Q?hoOPhy8oojwKsPapCHjpnkXLSEhLsD2Wb/WeAEgmppXLqFeWK2iWKHb0S1?= =?iso-8859-1?Q?5AczWanatHus4ewTgy/xLzVnkKXjmnbHEld8b4xgoQpz6SyNSKBj68WUdc?= =?iso-8859-1?Q?s43lFKY4EHaoR1Wr8FhooBI1y6hk3muZlE?= X-MS-Exchange-CrossTenant-Network-Message-Id: 8fd94ca3-d3c6-4f50-cc8c-08dd728c582f X-MS-Exchange-CrossTenant-AuthSource: LV3PR11MB8603.namprd11.prod.outlook.com X-MS-Exchange-CrossTenant-AuthAs: Internal X-MS-Exchange-CrossTenant-OriginalArrivalTime: 03 Apr 2025 08:48:46.5444 (UTC) X-MS-Exchange-CrossTenant-FromEntityHeader: Hosted X-MS-Exchange-CrossTenant-Id: 46c98d88-e344-4ed4-8496-4ed7712e255d X-MS-Exchange-CrossTenant-MailboxType: HOSTED X-MS-Exchange-CrossTenant-UserPrincipalName: MV6UTIk76Zb3hMioygNB+H7bsILw6zMDTaRPfjaMpLa8BcvqKfSR6E/fwIM60NML6nYPhWkhvF+yb2oIEu4Fjw== X-MS-Exchange-Transport-CrossTenantHeadersStamped: MW3PR11MB4732 X-OriginatorOrg: intel.com X-Rspamd-Server: rspam11 X-Rspamd-Queue-Id: 21DDD180006 X-Stat-Signature: n7dbqta4qs98ey1zkke9tmie7hzgrsj8 X-Rspam-User: X-HE-Tag: 1743670133-592073 X-HE-Meta: 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 Vv7Iv4Jf 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 X-Bogosity: Ham, tests=bogofilter, spamicity=0.000000, version=1.2.4 Sender: owner-linux-mm@kvack.org Precedence: bulk X-Loop: owner-majordomo@kvack.org List-ID: List-Subscribe: List-Unsubscribe: hi, Johannes Weiner, On Wed, Apr 02, 2025 at 03:50:42PM -0400, Johannes Weiner wrote: > Hello, > > On Thu, Mar 27, 2025 at 04:20:41PM +0800, kernel test robot wrote: > > kernel test robot noticed a 56.4% regression of vm-scalability.throughput on: > > > > commit: c2f6ea38fc1b640aa7a2e155cc1c0410ff91afa2 ("mm: page_alloc: don't steal single pages from biggest buddy") > > https://git.kernel.org/cgit/linux/kernel/git/next/linux-next.git master > > > > testcase: vm-scalability > > config: x86_64-rhel-9.4 > > compiler: gcc-12 > > test machine: 224 threads 4 sockets Intel(R) Xeon(R) Platinum 8380H CPU @ 2.90GHz (Cooper Lake) with 192G memory > > parameters: > > > > runtime: 300s > > test: lru-file-mmap-read > > cpufreq_governor: performance > > Thanks for the report. > > Would you be able to re-test with the below patch applied? > > There are more details in the thread here: > https://lore.kernel.org/all/20250402194425.GB198651@cmpxchg.org/ > > It's on top of the following upstream commit: we applied below patch upon acc4d5ff0b then run the same tests on acc4d5ff0b and 2c847f27c3 (acc4d5ff0b+patch). we noticed acc4d5ff0b has very similar performance results as c2f6ea38fc, then 2c847f27c3 has a much better results comparing to f3b92176f4 (parent of c2f6ea38fc). so if that acc4d5ff0b and c2f6ea38fc has similar performance is expected, it seems to us your patch not only recover the regression, but also make a big improvement. ========================================================================================= compiler/cpufreq_governor/kconfig/rootfs/runtime/tbox_group/test/testcase: gcc-12/performance/x86_64-rhel-9.4/debian-12-x86_64-20240206.cgz/300s/lkp-cpl-4sp2/lru-file-mmap-read/vm-scalability commit: f3b92176f4 ("tools/selftests: add guard region test for /proc/$pid/pagemap") c2f6ea38fc ("mm: page_alloc: don't steal single pages from biggest buddy") acc4d5ff0b ("Merge tag 'net-6.15-rc0' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net") 2c847f27c3 ("mm: page_alloc: speed up fallbacks in rmqueue_bulk()") <--- your patch f3b92176f4f7100f c2f6ea38fc1b640aa7a2e155cc1 acc4d5ff0b61eb1715c498b6536 2c847f27c37da65a93d23c237c5 ---------------- --------------------------- --------------------------- --------------------------- %stddev %change %stddev %change %stddev %change %stddev \ | \ | \ | \ 25525364 ± 3% -56.4% 11135467 -57.8% 10779336 +31.6% 33581409 vm-scalability.throughput full comparison is as below FYI. please feel free to add Tested-by: kernel test robot if above tests has no problem. thanks ========================================================================================= compiler/cpufreq_governor/kconfig/rootfs/runtime/tbox_group/test/testcase: gcc-12/performance/x86_64-rhel-9.4/debian-12-x86_64-20240206.cgz/300s/lkp-cpl-4sp2/lru-file-mmap-read/vm-scalability commit: f3b92176f4 ("tools/selftests: add guard region test for /proc/$pid/pagemap") c2f6ea38fc ("mm: page_alloc: don't steal single pages from biggest buddy") acc4d5ff0b ("Merge tag 'net-6.15-rc0' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net") 2c847f27c3 (mm: page_alloc: speed up fallbacks in rmqueue_bulk()) f3b92176f4f7100f c2f6ea38fc1b640aa7a2e155cc1 acc4d5ff0b61eb1715c498b6536 2c847f27c37da65a93d23c237c5 ---------------- --------------------------- --------------------------- --------------------------- %stddev %change %stddev %change %stddev %change %stddev \ | \ | \ | \ 1.702e+10 ± 12% -53.7% 7.876e+09 ± 5% -24.4% 1.287e+10 ± 57% -48.4% 8.785e+09 ± 6% cpuidle..time 3890512 ± 5% -48.0% 2022625 ± 19% -53.7% 1799850 ± 17% +5.4% 4101844 ± 9% cpuidle..usage 320.71 ± 5% +18.2% 379.03 +31.2% 420.83 ± 13% -28.7% 228.56 uptime.boot 26286 ± 8% -33.8% 17404 ± 4% +0.9% 26520 ± 47% -30.9% 18161 ± 3% uptime.idle 28.91 ± 7% -59.9% 11.58 ± 7% -44.0% 16.20 ± 45% -23.7% 22.05 ± 6% vmstat.cpu.id 166.61 ± 2% +25.4% 208.98 +18.1% 196.82 ± 8% +8.6% 180.94 vmstat.procs.r 3642 ± 3% +3.8% 3780 ± 4% -2.4% 3556 ± 2% +34.1% 4883 ± 7% vmstat.system.cs 550402 ± 6% -39.8% 331592 ± 2% -44.5% 305715 ± 9% +59.8% 879806 ± 2% vmstat.system.in 2742 ± 27% -13.5% 2372 ± 10% -23.6% 2095 ± 11% -99.9% 3.25 ± 62% perf-c2c.DRAM.local 7978 ± 21% +28.4% 10242 ± 9% +15.9% 9242 ± 12% -99.8% 15.50 ± 24% perf-c2c.DRAM.remote 4782 ± 22% +23.4% 5901 ± 8% +13.9% 5445 ± 12% -99.9% 6.88 ± 50% perf-c2c.HITM.local 2840 ± 24% +25.8% 3574 ± 10% +15.8% 3289 ± 14% -99.8% 7.00 ± 43% perf-c2c.HITM.remote 7623 ± 22% +24.3% 9475 ± 9% +14.6% 8735 ± 12% -99.8% 13.88 ± 24% perf-c2c.HITM.total 28.49 ± 8% -17.4 11.09 ± 7% -12.7 15.75 ± 47% -7.3 21.18 ± 6% mpstat.cpu.all.idle% 0.00 ± 50% +0.0 0.00 ± 94% -0.0 0.00 ± 59% -0.0 0.00 ±100% mpstat.cpu.all.iowait% 0.31 ± 4% -0.0 0.28 -0.0 0.27 ± 8% +0.1 0.42 mpstat.cpu.all.irq% 0.05 ± 2% -0.0 0.04 ± 8% -0.0 0.04 ± 13% +0.0 0.06 ± 4% mpstat.cpu.all.soft% 68.92 ± 3% +17.8 86.74 +13.3 82.25 ± 8% +6.1 75.00 mpstat.cpu.all.sys% 2.23 ± 6% -0.4 1.85 -0.5 1.69 ± 9% +1.1 3.34 mpstat.cpu.all.usr% 765416 ± 4% +13.2% 866765 ± 3% +15.0% 880593 ± 7% -24.7% 576338 ± 5% meminfo.Active(anon) 16677 ± 3% +6.3% 17724 ± 7% +181.7% 46975 ±108% +18.9% 19822 ± 6% meminfo.AnonHugePages 1731797 ± 3% +7.2% 1855684 ± 2% +7.9% 1868615 ± 5% -26.0% 1280873 meminfo.Committed_AS 781325 ± 5% +1.2% 790721 ± 3% +1.8% 795332 ± 11% -35.2% 506126 ± 3% meminfo.Inactive(anon) 1.435e+08 +13.2% 1.623e+08 +6.0% 1.52e+08 ± 9% +0.7% 1.445e+08 meminfo.Mapped 20964191 ± 2% -44.7% 11597836 ± 4% +9.0% 22842984 ± 65% -3.0% 20338164 ± 3% meminfo.MemFree 2690300 ± 2% +5.8% 2847331 -4.7% 2564772 ± 9% +26.6% 3406653 meminfo.PageTables 852601 ± 7% +12.8% 962045 ± 4% +15.4% 983850 ± 10% -55.8% 377229 ± 3% meminfo.Shmem 274.21 ± 6% +20.6% 330.73 +20.7% 330.85 -34.1% 180.74 time.elapsed_time 274.21 ± 6% +20.6% 330.73 +20.7% 330.85 -34.1% 180.74 time.elapsed_time.max 327511 +46.7% 480523 ± 7% +36.9% 448208 ± 5% +16.8% 382566 ± 14% time.involuntary_context_switches 1348 ± 3% +344.1% 5987 ± 3% +355.6% 6142 ± 8% +280.8% 5134 ± 18% time.major_page_faults 10706144 ± 4% -73.6% 2825142 ± 18% -70.7% 3137988 ± 22% -61.2% 4148705 ± 11% time.maximum_resident_set_size 93328020 -34.2% 61436664 -36.8% 59002722 +1.9% 95067700 time.minor_page_faults 15802 ± 2% +24.3% 19641 +24.8% 19721 +9.6% 17317 time.percent_of_cpu_this_job_got 41920 ± 4% +51.6% 63539 +52.3% 63860 -28.5% 29977 time.system_time 1352 +5.2% 1422 +2.7% 1388 -2.3% 1320 time.user_time 3779 -11.1% 3360 ± 2% -11.8% 3331 ± 2% -7.3% 3501 time.voluntary_context_switches 0.12 ± 2% -51.3% 0.06 ± 3% -48.3% 0.06 ± 6% -43.5% 0.07 ± 8% vm-scalability.free_time 121005 ± 3% -58.8% 49845 -60.2% 48099 +21.8% 147354 vm-scalability.median 4346 ± 16% -2970.6 1376 ± 17% -3164.2 1182 ± 6% -3226.7 1119 ± 8% vm-scalability.stddev% 25525364 ± 3% -56.4% 11135467 -57.8% 10779336 +31.6% 33581409 vm-scalability.throughput 274.21 ± 6% +20.6% 330.73 +20.7% 330.85 -34.1% 180.74 vm-scalability.time.elapsed_time 274.21 ± 6% +20.6% 330.73 +20.7% 330.85 -34.1% 180.74 vm-scalability.time.elapsed_time.max 327511 +46.7% 480523 ± 7% +36.9% 448208 ± 5% +16.8% 382566 ± 14% vm-scalability.time.involuntary_context_switches 1348 ± 3% +344.1% 5987 ± 3% +355.6% 6142 ± 8% +280.8% 5134 ± 18% vm-scalability.time.major_page_faults 10706144 ± 4% -73.6% 2825142 ± 18% -70.7% 3137988 ± 22% -61.2% 4148705 ± 11% vm-scalability.time.maximum_resident_set_size 93328020 -34.2% 61436664 -36.8% 59002722 +1.9% 95067700 vm-scalability.time.minor_page_faults 15802 ± 2% +24.3% 19641 +24.8% 19721 +9.6% 17317 vm-scalability.time.percent_of_cpu_this_job_got 41920 ± 4% +51.6% 63539 +52.3% 63860 -28.5% 29977 vm-scalability.time.system_time 1352 +5.2% 1422 +2.7% 1388 -2.3% 1320 vm-scalability.time.user_time 4.832e+09 -30.7% 3.346e+09 -33.0% 3.239e+09 +0.0% 4.832e+09 vm-scalability.workload 4896915 ± 5% -39.0% 2988750 ± 8% +19.0% 5827893 ± 64% +1.9% 4992149 ± 7% numa-meminfo.node0.MemFree 665421 ± 3% +6.6% 709460 -4.4% 635976 ± 9% +28.3% 853921 ± 3% numa-meminfo.node0.PageTables 12227553 ± 3% +12.0% 13694479 ± 4% +3.8% 12691397 ± 9% +2.4% 12522980 ± 4% numa-meminfo.node1.Active 35323186 ± 3% +15.9% 40935721 ± 2% +7.7% 38058602 ± 9% +3.8% 36672727 ± 2% numa-meminfo.node1.Mapped 5320950 ± 6% -43.3% 3017277 ± 5% +13.3% 6030412 ± 66% -1.2% 5257151 ± 7% numa-meminfo.node1.MemFree 668594 ± 2% +5.7% 706819 -4.8% 636211 ± 9% +27.4% 851702 numa-meminfo.node1.PageTables 677517 ± 8% +13.1% 766377 ± 5% +2.6% 694945 ± 9% +1.0% 684143 ± 8% numa-meminfo.node2.KReclaimable 35949507 ± 3% +13.6% 40822797 ± 3% +7.3% 38557650 ± 10% +1.3% 36413009 ± 2% numa-meminfo.node2.Mapped 5273037 ± 8% -42.7% 3019409 ± 4% +14.1% 6015871 ± 64% -0.1% 5268423 ± 3% numa-meminfo.node2.MemFree 667862 ± 2% +5.9% 707231 -5.3% 632171 ± 9% +27.5% 851776 ± 2% numa-meminfo.node2.PageTables 677517 ± 8% +13.1% 766377 ± 5% +2.6% 694945 ± 9% +1.0% 684143 ± 8% numa-meminfo.node2.SReclaimable 850745 ± 7% +11.1% 945010 ± 4% +1.5% 863702 ± 7% +0.3% 853639 ± 6% numa-meminfo.node2.Slab 12355499 +11.5% 13779785 ± 4% +7.2% 13245341 ± 8% +1.3% 12511162 ± 3% numa-meminfo.node3.Active 11930599 ± 2% +11.0% 13243137 ± 3% +6.8% 12737603 ± 8% +3.4% 12331424 ± 3% numa-meminfo.node3.Active(file) 457041 ± 42% +7.5% 491478 ± 17% +5.4% 481915 ± 22% -65.0% 159787 ± 64% numa-meminfo.node3.Inactive(anon) 35933100 ± 4% +14.0% 40967998 +6.6% 38295485 ± 9% +0.8% 36206962 ± 3% numa-meminfo.node3.Mapped 5570339 ± 4% -46.1% 3002472 ± 3% +7.6% 5991739 ± 64% -8.3% 5105868 ± 5% numa-meminfo.node3.MemFree 685019 ± 5% +3.1% 706516 -7.7% 632019 ± 9% +22.0% 835716 ± 3% numa-meminfo.node3.PageTables 649934 ± 25% +16.1% 754493 ± 9% +14.9% 746997 ± 14% -67.8% 209052 ± 33% numa-meminfo.node3.Shmem 53331064 ± 4% -40.0% 32001591 ± 2% -42.3% 30773977 ± 2% -1.1% 52762661 numa-numastat.node0.local_node 11230540 ± 10% -14.4% 9615282 -14.9% 9553492 ± 3% +15.0% 12914957 ± 9% numa-numastat.node0.numa_foreign 53443666 ± 4% -40.0% 32082061 ± 2% -42.3% 30857092 ± 2% -1.1% 52844556 numa-numastat.node0.numa_hit 18920287 ± 9% -43.6% 10668582 ± 4% -46.0% 10214237 ± 4% -31.2% 13023140 ± 11% numa-numastat.node0.numa_miss 19025905 ± 9% -43.5% 10749724 ± 3% -45.9% 10299624 ± 4% -31.1% 13105066 ± 11% numa-numastat.node0.other_node 56535277 ± 4% -42.5% 32511952 ± 3% -45.1% 31014584 -4.9% 53743966 ± 2% numa-numastat.node1.local_node 14228056 ± 6% -29.9% 9967849 ± 3% -33.3% 9488657 ± 3% -2.5% 13871144 ± 16% numa-numastat.node1.numa_foreign 56618771 ± 4% -42.4% 32596415 ± 3% -45.1% 31109797 -4.9% 53829937 ± 2% numa-numastat.node1.numa_hit 10146402 ± 5% -3.8% 9758929 ± 2% -3.4% 9799142 ± 6% +19.3% 12105407 ± 6% numa-numastat.node1.numa_miss 10233076 ± 5% -3.8% 9843695 ± 2% -3.3% 9894436 ± 6% +19.2% 12193463 ± 6% numa-numastat.node1.other_node 53165000 ± 5% -38.0% 32981697 ± 2% -41.0% 31364931 ± 2% -0.4% 52973679 numa-numastat.node2.local_node 13182856 ± 7% -22.6% 10202650 ± 5% -26.7% 9659290 ± 3% -4.8% 12544230 ± 6% numa-numastat.node2.numa_foreign 53265107 ± 5% -37.9% 33065351 ± 2% -41.0% 31452697 ± 2% -0.4% 53065111 numa-numastat.node2.numa_hit 12626193 ± 4% -23.6% 9641387 ± 5% -25.8% 9367817 ± 3% +6.9% 13500636 ± 8% numa-numastat.node2.numa_miss 12723206 ± 4% -23.5% 9727091 ± 5% -25.7% 9454042 ± 3% +6.8% 13593388 ± 8% numa-numastat.node2.other_node 53553158 ± 4% -38.8% 32791369 -40.7% 31754688 ± 2% -0.3% 53375310 numa-numastat.node3.local_node 14822055 ± 13% -32.0% 10075025 ± 4% -33.3% 9887745 ± 4% -17.6% 12217967 ± 6% numa-numastat.node3.numa_foreign 53612301 ± 4% -38.7% 32888921 -40.6% 31838271 ± 2% -0.3% 53456894 numa-numastat.node3.numa_hit 11766242 ± 12% -16.8% 9789841 ± 4% -21.7% 9207177 ± 2% +9.8% 12916504 ± 7% numa-numastat.node3.numa_miss 11826951 ± 12% -16.4% 9886570 ± 4% -21.5% 9289581 ± 2% +9.9% 13002401 ± 7% numa-numastat.node3.other_node 5.95 ± 3% -22.1% 4.63 ± 4% -27.4% 4.32 ± 9% -11.6% 5.26 ± 2% perf-stat.i.MPKI 3.24e+10 ± 3% -41.7% 1.888e+10 -48.4% 1.673e+10 ± 13% +14.0% 3.694e+10 perf-stat.i.branch-instructions 0.33 +0.0 0.34 +0.2 0.52 ± 32% -0.0 0.29 perf-stat.i.branch-miss-rate% 95838227 ± 2% -45.2% 52482057 -43.6% 54055222 ± 16% -22.3% 74458420 perf-stat.i.branch-misses 66.91 +3.8 70.68 -1.8 65.13 ± 11% +0.3 67.21 perf-stat.i.cache-miss-rate% 6.536e+08 ± 4% -56.4% 2.852e+08 -61.7% 2.506e+08 ± 14% -3.8% 6.29e+08 perf-stat.i.cache-misses 1.004e+09 ± 4% -59.5% 4.069e+08 -64.4% 3.575e+08 ± 13% -4.6% 9.578e+08 perf-stat.i.cache-references 3578 ± 3% +6.3% 3804 ± 4% -1.2% 3536 ± 3% +33.6% 4781 ± 7% perf-stat.i.context-switches 5.02 ± 3% +111.1% 10.59 +99.2% 9.99 ± 11% -9.7% 4.53 perf-stat.i.cpi 225243 +1.0% 227600 +1.1% 227710 -0.2% 224716 perf-stat.i.cpu-clock 5.886e+11 ± 2% +23.7% 7.28e+11 +12.4% 6.617e+11 ± 13% +12.2% 6.606e+11 perf-stat.i.cpu-cycles 290.25 ± 2% -7.7% 267.91 ± 2% -12.4% 254.28 ± 3% +12.3% 326.01 perf-stat.i.cpu-migrations 859.86 ± 2% +201.8% 2594 +196.7% 2551 ± 7% +10.0% 945.58 ± 2% perf-stat.i.cycles-between-cache-misses 1.172e+11 ± 2% -42.6% 6.733e+10 -49.1% 5.971e+10 ± 13% +11.2% 1.303e+11 perf-stat.i.instructions 0.33 ± 3% -35.8% 0.21 -19.3% 0.27 ± 25% -5.6% 0.32 ± 2% perf-stat.i.ipc 5.53 ± 7% +240.6% 18.83 ± 3% +218.6% 17.61 ± 17% +401.8% 27.73 ± 18% perf-stat.i.major-faults 2.83 ± 12% -81.7% 0.52 ± 3% -84.1% 0.45 ± 14% +62.6% 4.60 perf-stat.i.metric.K/sec 346126 ± 7% -43.9% 194284 -50.6% 171036 ± 13% +49.7% 518309 perf-stat.i.minor-faults 346131 ± 7% -43.9% 194303 -50.6% 171054 ± 13% +49.8% 518337 perf-stat.i.page-faults 225243 +1.0% 227600 +1.1% 227710 -0.2% 224716 perf-stat.i.task-clock 5.70 ± 2% -26.5% 4.19 -27.4% 4.13 -15.4% 4.82 perf-stat.overall.MPKI 0.30 -0.0 0.28 +0.0 0.33 ± 2% -0.1 0.20 perf-stat.overall.branch-miss-rate% 64.60 +5.2 69.79 +5.1 69.73 +1.0 65.63 perf-stat.overall.cache-miss-rate% 5.28 ± 2% +116.9% 11.46 +123.9% 11.83 -3.7% 5.09 perf-stat.overall.cpi 927.91 +194.9% 2736 +208.6% 2863 +13.7% 1055 perf-stat.overall.cycles-between-cache-misses 0.19 ± 2% -53.9% 0.09 -55.4% 0.08 +3.8% 0.20 perf-stat.overall.ipc 6654 ± 3% -0.9% 6597 -0.4% 6628 -25.1% 4982 perf-stat.overall.path-length 3.23e+10 ± 3% -42.4% 1.859e+10 -47.4% 1.701e+10 ± 8% +15.7% 3.736e+10 perf-stat.ps.branch-instructions 97953773 ± 2% -46.6% 52288618 -42.0% 56848683 ± 10% -23.3% 75098323 perf-stat.ps.branch-misses 6.662e+08 ± 4% -58.2% 2.782e+08 -62.3% 2.514e+08 ± 9% -4.6% 6.354e+08 perf-stat.ps.cache-misses 1.031e+09 ± 4% -61.3% 3.986e+08 -65.0% 3.606e+08 ± 9% -6.1% 9.682e+08 perf-stat.ps.cache-references 3562 ± 3% +3.9% 3702 ± 4% -1.9% 3494 ± 2% +34.1% 4775 ± 8% perf-stat.ps.context-switches 6.177e+11 ± 2% +23.2% 7.612e+11 +16.5% 7.195e+11 ± 8% +8.6% 6.707e+11 perf-stat.ps.cpu-cycles 285.57 ± 2% -10.4% 255.87 ± 2% -15.4% 241.61 ± 4% +13.5% 324.22 perf-stat.ps.cpu-migrations 1.169e+11 ± 2% -43.2% 6.643e+10 -48.0% 6.083e+10 ± 9% +12.7% 1.318e+11 perf-stat.ps.instructions 4.92 ± 6% +266.3% 18.02 ± 3% +254.3% 17.43 ± 13% +471.3% 28.11 ± 17% perf-stat.ps.major-faults 344062 ± 6% -45.4% 188002 -50.5% 170284 ± 9% +52.4% 524213 perf-stat.ps.minor-faults 344067 ± 6% -45.4% 188020 -50.5% 170301 ± 9% +52.4% 524242 perf-stat.ps.page-faults 3.215e+13 ± 3% -31.3% 2.208e+13 -33.2% 2.147e+13 -25.1% 2.407e+13 perf-stat.total.instructions 1226884 ± 5% -36.0% 785027 ± 8% +23.2% 1511805 ± 65% +2.8% 1261046 ± 7% numa-vmstat.node0.nr_free_pages 166318 ± 4% +5.8% 175958 -5.1% 157824 ± 10% +28.0% 212909 ± 3% numa-vmstat.node0.nr_page_table_pages 11230540 ± 10% -14.4% 9615282 -14.9% 9553492 ± 3% +15.0% 12914957 ± 9% numa-vmstat.node0.numa_foreign 53443303 ± 4% -40.0% 32082034 ± 2% -42.3% 30856944 ± 2% -1.1% 52844258 numa-vmstat.node0.numa_hit 53330701 ± 4% -40.0% 32001564 ± 2% -42.3% 30773829 ± 2% -1.1% 52762363 numa-vmstat.node0.numa_local 18920287 ± 9% -43.6% 10668582 ± 4% -46.0% 10214237 ± 4% -31.2% 13023140 ± 11% numa-vmstat.node0.numa_miss 19025905 ± 9% -43.5% 10749724 ± 3% -45.9% 10299624 ± 4% -31.1% 13105066 ± 11% numa-vmstat.node0.numa_other 2404851 ± 17% -81.8% 436913 ± 19% -80.5% 468574 ± 24% -29.4% 1697340 ± 14% numa-vmstat.node0.workingset_nodereclaim 3035160 ± 3% +12.0% 3400384 ± 4% +3.3% 3133854 ± 10% +4.6% 3174439 ± 7% numa-vmstat.node1.nr_active_file 1332733 ± 6% -40.3% 795144 ± 6% +17.7% 1568458 ± 66% -0.4% 1327571 ± 7% numa-vmstat.node1.nr_free_pages 8817580 ± 3% +15.4% 10171927 ± 2% +6.9% 9428593 ± 10% +3.6% 9132555 ± 2% numa-vmstat.node1.nr_mapped 166937 ± 2% +5.0% 175300 -5.4% 157885 ± 10% +27.2% 212407 numa-vmstat.node1.nr_page_table_pages 3039792 ± 3% +12.1% 3406582 ± 4% +3.4% 3142072 ± 10% +4.6% 3179577 ± 7% numa-vmstat.node1.nr_zone_active_file 14228056 ± 6% -29.9% 9967849 ± 3% -33.3% 9488657 ± 3% -2.5% 13871144 ± 16% numa-vmstat.node1.numa_foreign 56618372 ± 4% -42.4% 32596203 ± 3% -45.1% 31109326 -4.9% 53829441 ± 2% numa-vmstat.node1.numa_hit 56534878 ± 4% -42.5% 32511740 ± 3% -45.1% 31014114 -4.9% 53743471 ± 2% numa-vmstat.node1.numa_local 10146402 ± 5% -3.8% 9758929 ± 2% -3.4% 9799142 ± 6% +19.3% 12105407 ± 6% numa-vmstat.node1.numa_miss 10233076 ± 5% -3.8% 9843695 ± 2% -3.3% 9894436 ± 6% +19.2% 12193463 ± 6% numa-vmstat.node1.numa_other 954127 ± 20% -73.1% 256611 ± 34% -71.6% 270621 ± 33% +25.2% 1195021 ± 24% numa-vmstat.node1.workingset_nodereclaim 3046975 ± 3% +12.3% 3422217 ± 3% +3.6% 3156615 ± 12% +0.8% 3070205 ± 5% numa-vmstat.node2.nr_active_file 1320090 ± 8% -39.7% 796670 ± 4% +18.4% 1562543 ± 64% +0.9% 1331463 ± 4% numa-vmstat.node2.nr_free_pages 8977022 ± 3% +13.0% 10145019 ± 3% +6.4% 9553399 ± 11% +1.0% 9070716 ± 2% numa-vmstat.node2.nr_mapped 166832 ± 2% +5.2% 175431 -6.0% 156887 ± 10% +27.4% 212464 ± 2% numa-vmstat.node2.nr_page_table_pages 169424 ± 8% +12.7% 191016 ± 5% +2.2% 173094 ± 9% +0.9% 170930 ± 8% numa-vmstat.node2.nr_slab_reclaimable 3051630 ± 3% +12.4% 3431556 ± 3% +3.6% 3162379 ± 12% +0.9% 3079502 ± 5% numa-vmstat.node2.nr_zone_active_file 13182856 ± 7% -22.6% 10202650 ± 5% -26.7% 9659290 ± 3% -4.8% 12544230 ± 6% numa-vmstat.node2.numa_foreign 53264088 ± 5% -37.9% 33064447 ± 2% -41.0% 31451863 ± 2% -0.4% 53064445 numa-vmstat.node2.numa_hit 53163982 ± 5% -38.0% 32980793 ± 2% -41.0% 31364093 ± 2% -0.4% 52973013 numa-vmstat.node2.numa_local 12626193 ± 4% -23.6% 9641387 ± 5% -25.8% 9367817 ± 3% +6.9% 13500636 ± 8% numa-vmstat.node2.numa_miss 12723206 ± 4% -23.5% 9727091 ± 5% -25.7% 9454042 ± 3% +6.8% 13593388 ± 8% numa-vmstat.node2.numa_other 919386 ± 31% -64.5% 326738 ± 18% -68.2% 292149 ± 38% -5.5% 869137 ± 9% numa-vmstat.node2.workingset_nodereclaim 3025992 +9.3% 3308587 ± 3% +5.2% 3184016 ± 9% +2.4% 3097552 ± 4% numa-vmstat.node3.nr_active_file 1395961 ± 3% -43.3% 791816 ± 3% +11.4% 1555626 ± 65% -7.6% 1289705 ± 5% numa-vmstat.node3.nr_free_pages 113754 ± 41% +8.1% 122994 ± 17% +6.0% 120611 ± 22% -64.9% 39920 ± 63% numa-vmstat.node3.nr_inactive_anon 8971283 ± 4% +13.5% 10181092 +5.8% 9488578 ± 9% +0.5% 9016607 ± 3% numa-vmstat.node3.nr_mapped 171114 ± 5% +2.4% 175296 -8.3% 156889 ± 10% +21.8% 208408 ± 3% numa-vmstat.node3.nr_page_table_pages 163305 ± 25% +16.0% 189436 ± 9% +15.4% 188460 ± 15% -67.9% 52479 ± 33% numa-vmstat.node3.nr_shmem 3030697 +9.4% 3314830 ± 3% +5.3% 3189831 ± 9% +2.4% 3102693 ± 4% numa-vmstat.node3.nr_zone_active_file 113675 ± 41% +8.2% 122992 ± 17% +6.1% 120610 ± 22% -64.9% 39918 ± 63% numa-vmstat.node3.nr_zone_inactive_anon 14822055 ± 13% -32.0% 10075025 ± 4% -33.3% 9887745 ± 4% -17.6% 12217967 ± 6% numa-vmstat.node3.numa_foreign 53610436 ± 4% -38.7% 32888516 -40.6% 31837112 ± 2% -0.3% 53455092 numa-vmstat.node3.numa_hit 53551293 ± 4% -38.8% 32790963 -40.7% 31753529 ± 2% -0.3% 53373507 numa-vmstat.node3.numa_local 11766242 ± 12% -16.8% 9789841 ± 4% -21.7% 9207177 ± 2% +9.8% 12916504 ± 7% numa-vmstat.node3.numa_miss 11826951 ± 12% -16.4% 9886570 ± 4% -21.5% 9289581 ± 2% +9.9% 13002401 ± 7% numa-vmstat.node3.numa_other 925252 ± 26% -66.0% 314477 ± 22% -67.3% 302227 ± 15% +5.2% 973634 ± 17% numa-vmstat.node3.workingset_nodereclaim 26042583 ± 5% +24.2% 32343080 ± 10% +43.8% 37446150 ± 5% -38.5% 16024952 ± 14% sched_debug.cfs_rq:/.avg_vruntime.avg 28289592 ± 6% +20.6% 34111958 ± 11% +38.7% 39248002 ± 6% -40.0% 16973252 ± 15% sched_debug.cfs_rq:/.avg_vruntime.max 5569284 ± 52% -96.3% 205900 ± 37% -62.4% 2094368 ±140% -93.8% 346140 ±170% sched_debug.cfs_rq:/.avg_vruntime.min 2663785 ± 13% +37.7% 3669196 ± 6% +53.8% 4096457 ± 5% -26.9% 1948436 ± 14% sched_debug.cfs_rq:/.avg_vruntime.stddev 0.60 ± 7% +41.6% 0.85 ± 2% +39.7% 0.84 ± 6% +6.2% 0.64 ± 12% sched_debug.cfs_rq:/.h_nr_queued.avg 1.55 ± 8% +29.4% 2.01 ± 5% +26.7% 1.96 ± 8% +7.5% 1.67 ± 4% sched_debug.cfs_rq:/.h_nr_queued.max 0.29 ± 6% -29.3% 0.21 ± 9% -31.6% 0.20 ± 6% -34.5% 0.19 ± 10% sched_debug.cfs_rq:/.h_nr_queued.stddev 0.60 ± 7% +41.6% 0.85 +39.7% 0.84 ± 6% +6.2% 0.64 ± 12% sched_debug.cfs_rq:/.h_nr_runnable.avg 1.52 ± 9% +32.2% 2.01 ± 5% +28.1% 1.94 ± 8% +7.8% 1.64 ± 5% sched_debug.cfs_rq:/.h_nr_runnable.max 0.29 ± 6% -29.5% 0.21 ± 10% -32.0% 0.20 ± 6% -35.0% 0.19 ± 8% sched_debug.cfs_rq:/.h_nr_runnable.stddev 27841 ± 24% -17.5% 22967 ± 11% -16.0% 23393 ± 14% -33.7% 18446 ± 17% sched_debug.cfs_rq:/.load.avg 137171 ± 12% -8.6% 125407 ± 6% -4.3% 131315 ± 10% -24.4% 103716 ± 13% sched_debug.cfs_rq:/.load.stddev 33.98 ± 16% -16.6% 28.35 ± 6% -15.9% 28.58 ± 9% +4.9% 35.65 ± 76% sched_debug.cfs_rq:/.load_avg.avg 26042586 ± 5% +24.2% 32343084 ± 10% +43.8% 37446154 ± 5% -38.5% 16024954 ± 14% sched_debug.cfs_rq:/.min_vruntime.avg 28289592 ± 6% +20.6% 34111958 ± 11% +38.7% 39248002 ± 6% -40.0% 16973252 ± 15% sched_debug.cfs_rq:/.min_vruntime.max 5569284 ± 52% -96.3% 205900 ± 37% -62.4% 2094368 ±140% -93.8% 346140 ±170% sched_debug.cfs_rq:/.min_vruntime.min 2663785 ± 13% +37.7% 3669195 ± 6% +53.8% 4096457 ± 5% -26.9% 1948437 ± 14% sched_debug.cfs_rq:/.min_vruntime.stddev 0.59 ± 7% +38.8% 0.82 +37.2% 0.81 ± 6% +4.4% 0.62 ± 13% sched_debug.cfs_rq:/.nr_queued.avg 0.26 ± 6% -59.4% 0.11 ± 26% -62.7% 0.10 ± 15% -56.9% 0.11 ± 31% sched_debug.cfs_rq:/.nr_queued.stddev 226.57 ± 21% -2.1% 221.83 ± 38% -27.4% 164.57 ± 6% +36.5% 309.30 ± 43% sched_debug.cfs_rq:/.removed.load_avg.max 30.84 ± 20% -18.5% 25.14 ± 35% -29.3% 21.81 ± 29% +33.3% 41.10 ± 46% sched_debug.cfs_rq:/.removed.load_avg.stddev 116.17 ± 22% +2.8% 119.39 ± 60% -27.6% 84.10 ± 8% +30.4% 151.49 ± 39% sched_debug.cfs_rq:/.removed.runnable_avg.max 116.17 ± 22% -9.9% 104.72 ± 39% -27.6% 84.10 ± 8% +24.2% 144.27 ± 39% sched_debug.cfs_rq:/.removed.util_avg.max 625.72 ± 7% +41.3% 883.96 +39.1% 870.63 ± 6% +7.6% 673.15 ± 13% sched_debug.cfs_rq:/.runnable_avg.avg 1492 ± 5% +31.8% 1967 ± 3% +27.8% 1906 ± 4% +10.9% 1654 ± 4% sched_debug.cfs_rq:/.runnable_avg.max 284.33 ± 6% -29.2% 201.27 ± 8% -33.0% 190.51 ± 6% -35.7% 182.69 ± 11% sched_debug.cfs_rq:/.runnable_avg.stddev 613.28 ± 7% +38.6% 850.28 +36.7% 838.66 ± 6% +6.5% 653.13 ± 13% sched_debug.cfs_rq:/.util_avg.avg 1227 ± 3% +12.3% 1377 ± 3% +10.1% 1351 ± 5% +11.2% 1365 ± 9% sched_debug.cfs_rq:/.util_avg.max 250.81 ± 4% -54.7% 113.67 ± 16% -59.3% 102.11 ± 11% -48.2% 129.80 ± 10% sched_debug.cfs_rq:/.util_avg.stddev 472.34 ± 6% +14.7% 541.81 ± 9% +17.6% 555.66 ± 8% -8.8% 430.79 ± 20% sched_debug.cfs_rq:/.util_est.avg 1293 ± 8% +34.5% 1739 ± 5% +25.4% 1621 ± 8% +11.3% 1438 ± 11% sched_debug.cfs_rq:/.util_est.max 307.05 ± 7% -1.6% 302.05 ± 6% -8.5% 280.94 ± 5% -24.9% 230.63 ± 10% sched_debug.cfs_rq:/.util_est.stddev 113130 ± 21% +55.7% 176123 ± 28% +113.7% 241712 ± 47% +40.5% 159001 ± 45% sched_debug.cpu.avg_idle.min 170684 ± 6% +6.9% 182523 ± 8% +30.8% 223234 ± 16% -31.2% 117363 ± 12% sched_debug.cpu.clock.avg 170713 ± 6% +7.0% 182635 ± 8% +30.8% 223304 ± 16% -31.2% 117399 ± 12% sched_debug.cpu.clock.max 170647 ± 6% +6.9% 182402 ± 8% +30.8% 223148 ± 16% -31.2% 117320 ± 12% sched_debug.cpu.clock.min 17.82 ± 12% +263.4% 64.74 ± 14% +163.0% 46.85 ± 28% +24.7% 22.22 ± 21% sched_debug.cpu.clock.stddev 170095 ± 6% +7.0% 182084 ± 8% +31.0% 222761 ± 16% -31.3% 116885 ± 12% sched_debug.cpu.clock_task.avg 170366 ± 6% +7.0% 182343 ± 8% +30.9% 222981 ± 16% -31.2% 117187 ± 12% sched_debug.cpu.clock_task.max 157562 ± 7% +7.5% 169450 ± 9% +33.7% 210623 ± 17% -33.5% 104794 ± 13% sched_debug.cpu.clock_task.min 3198 ± 4% +52.1% 4863 +50.0% 4798 ± 6% +13.6% 3634 ± 13% sched_debug.cpu.curr->pid.avg 9729 ± 3% -0.4% 9694 ± 5% +5.8% 10289 ± 4% -19.4% 7840 ± 6% sched_debug.cpu.curr->pid.max 1544 ± 5% -41.8% 898.71 ± 9% -42.5% 887.59 ± 6% -44.5% 858.09 ± 11% sched_debug.cpu.curr->pid.stddev 0.00 ± 26% +198.9% 0.00 ± 15% +156.3% 0.00 ± 12% -4.8% 0.00 ± 12% sched_debug.cpu.next_balance.stddev 0.56 ± 4% +52.3% 0.85 +50.2% 0.84 ± 6% +14.1% 0.64 ± 13% sched_debug.cpu.nr_running.avg 1.68 ± 13% +22.8% 2.07 ± 6% +21.5% 2.04 ± 10% +1.5% 1.71 ± 7% sched_debug.cpu.nr_running.max 0.30 ± 8% -27.8% 0.21 ± 9% -30.3% 0.21 ± 8% -34.5% 0.19 ± 9% sched_debug.cpu.nr_running.stddev 2958 ± 4% +5.1% 3109 ± 11% +16.1% 3433 ± 7% -17.1% 2453 ± 16% sched_debug.cpu.nr_switches.avg 1028 ± 8% -13.5% 889.74 ± 7% -4.3% 984.66 ± 8% -37.5% 643.32 ± 12% sched_debug.cpu.nr_switches.min 2121 ± 7% +15.2% 2444 ± 15% +23.7% 2623 ± 10% +1.0% 2142 ± 17% sched_debug.cpu.nr_switches.stddev 170653 ± 6% +6.9% 182400 ± 8% +30.8% 223145 ± 16% -31.3% 117323 ± 12% sched_debug.cpu_clk 169804 ± 6% +6.9% 181548 ± 8% +30.9% 222295 ± 16% -31.4% 116473 ± 12% sched_debug.ktime 171435 ± 6% +6.9% 183225 ± 8% +30.6% 223900 ± 16% -31.1% 118087 ± 12% sched_debug.sched_clk 4828005 ± 2% -36.7% 3055295 -39.3% 2930398 -3.9% 4637517 proc-vmstat.allocstall_movable 19547 ± 3% +181.7% 55069 +174.3% 53609 +230.1% 64518 proc-vmstat.allocstall_normal 1.308e+08 ± 18% +81.5% 2.373e+08 ± 18% +71.5% 2.243e+08 ± 13% +234.1% 4.369e+08 ± 18% proc-vmstat.compact_daemon_free_scanned 1.503e+08 ± 13% -82.4% 26384666 ± 37% -83.2% 25185230 ± 16% -81.0% 28501389 ± 16% proc-vmstat.compact_daemon_migrate_scanned 2891 ± 12% -97.4% 76.17 ± 29% -96.9% 90.75 ± 19% -92.7% 211.62 ± 43% proc-vmstat.compact_daemon_wake 1520826 ± 19% -99.9% 1855 ± 38% -99.8% 2497 ± 67% -99.7% 5024 ±100% proc-vmstat.compact_fail 1.762e+08 ± 14% +35.6% 2.388e+08 ± 18% +28.4% 2.262e+08 ± 13% +150.7% 4.416e+08 ± 17% proc-vmstat.compact_free_scanned 33090251 ± 10% -40.6% 19664055 ± 31% -40.4% 19726165 ± 13% -25.3% 24731462 ± 11% proc-vmstat.compact_isolated 1.826e+10 ± 5% -99.7% 49399799 ± 29% -99.7% 54234660 ± 36% -99.6% 79280517 ± 47% proc-vmstat.compact_migrate_scanned 2450182 ± 21% -99.4% 15045 ± 15% -99.3% 16582 ± 29% -98.8% 28362 ± 47% proc-vmstat.compact_stall 929355 ± 26% -98.6% 13190 ± 13% -98.5% 14085 ± 22% -97.5% 23338 ± 36% proc-vmstat.compact_success 3716 ± 15% -96.4% 134.00 ± 18% -94.6% 200.50 ± 33% -89.7% 382.62 ± 36% proc-vmstat.kswapd_low_wmark_hit_quickly 191553 ± 3% +13.4% 217163 ± 3% +15.2% 220698 ± 7% -24.8% 144034 ± 4% proc-vmstat.nr_active_anon 12143003 ± 2% +8.5% 13179985 ± 2% +3.0% 12506962 ± 8% +1.0% 12267487 proc-vmstat.nr_active_file 174346 +0.2% 174651 -0.3% 173784 +1.6% 177080 proc-vmstat.nr_anon_pages 41767951 +5.4% 44014874 -1.1% 41312209 ± 8% -0.5% 41549076 proc-vmstat.nr_file_pages 5235973 ± 2% -43.9% 2939596 ± 2% +10.5% 5788020 ± 65% -1.6% 5151177 ± 4% proc-vmstat.nr_free_pages 195319 ± 5% +1.1% 197415 ± 2% +1.8% 198737 ± 12% -35.1% 126699 ± 3% proc-vmstat.nr_inactive_anon 28515253 +4.1% 29688629 -3.0% 27651676 ± 9% -0.8% 28282751 proc-vmstat.nr_inactive_file 75.88 ± 13% -20.3% 60.50 ± 29% -28.6% 54.18 ± 37% +82.1% 138.19 ± 20% proc-vmstat.nr_isolated_file 42800 +2.6% 43922 +1.3% 43348 +1.3% 43362 proc-vmstat.nr_kernel_stack 35859536 +12.9% 40468962 +5.7% 37892928 ± 9% +0.4% 36017924 proc-vmstat.nr_mapped 672657 ± 2% +5.5% 709985 -4.9% 639808 ± 9% +26.3% 849359 proc-vmstat.nr_page_table_pages 213344 ± 7% +12.8% 240725 ± 4% +15.5% 246461 ± 10% -55.7% 94464 ± 3% proc-vmstat.nr_shmem 746816 ± 2% +4.2% 777915 -3.5% 720723 ± 9% +4.6% 781061 proc-vmstat.nr_slab_reclaimable 192129 ± 3% +13.1% 217232 ± 3% +14.9% 220760 ± 7% -25.0% 144112 ± 4% proc-vmstat.nr_zone_active_anon 12164300 ± 2% +8.6% 13206949 ± 2% +3.0% 12532232 ± 8% +1.0% 12291549 proc-vmstat.nr_zone_active_file 194816 ± 5% +1.3% 197412 ± 2% +2.0% 198733 ± 12% -35.0% 126681 ± 3% proc-vmstat.nr_zone_inactive_anon 28495099 +4.1% 29661915 -3.0% 27626694 ± 9% -0.8% 28258909 proc-vmstat.nr_zone_inactive_file 53463509 ± 2% -25.4% 39860806 -27.8% 38589184 -3.6% 51548300 ± 3% proc-vmstat.numa_foreign 46231 ± 15% -74.8% 11635 ± 61% -75.0% 11553 ± 35% -64.9% 16238 ± 54% proc-vmstat.numa_hint_faults 36770 ± 26% -84.6% 5645 ± 68% -81.9% 6670 ± 52% -72.7% 10054 ± 71% proc-vmstat.numa_hint_faults_local 2.169e+08 ± 3% -39.8% 1.306e+08 -42.3% 1.253e+08 -1.7% 2.132e+08 proc-vmstat.numa_hit 2.166e+08 ± 3% -39.8% 1.303e+08 -42.3% 1.249e+08 -1.7% 2.129e+08 proc-vmstat.numa_local 53459124 ± 2% -25.4% 39858740 -27.8% 38588374 -3.6% 51545689 ± 3% proc-vmstat.numa_miss 53809139 ± 2% -25.3% 40207081 -27.6% 38937684 -3.6% 51894319 ± 3% proc-vmstat.numa_other 6545 ± 80% -92.1% 517.83 ± 48% -83.5% 1081 ± 68% -94.8% 340.12 ± 18% proc-vmstat.numa_pages_migrated 136643 ± 40% -70.2% 40747 ± 85% -55.3% 61018 ± 62% -53.5% 63487 ± 84% proc-vmstat.numa_pte_updates 3746 ± 15% -96.0% 149.50 ± 17% -94.2% 218.12 ± 31% -89.2% 403.12 ± 33% proc-vmstat.pageoutrun 94282999 -24.3% 71368473 -26.6% 69232966 +0.8% 95049121 proc-vmstat.pgactivate 10012599 ± 4% -43.2% 5686084 ± 3% -45.4% 5462427 ± 3% -13.4% 8674039 ± 4% proc-vmstat.pgalloc_dma32 1.069e+09 -34.4% 7.014e+08 -36.9% 6.748e+08 +0.1% 1.07e+09 proc-vmstat.pgalloc_normal 94463121 -33.7% 62613232 -36.2% 60238525 +1.5% 95858739 proc-vmstat.pgfault 1.095e+09 -34.6% 7.165e+08 -37.0% 6.898e+08 -0.4% 1.091e+09 proc-vmstat.pgfree 1339 ± 3% +346.2% 5975 ± 3% +358.0% 6133 ± 8% +282.7% 5124 ± 18% proc-vmstat.pgmajfault 74970 ± 36% -95.3% 3527 ± 22% -94.9% 3850 ± 22% -93.7% 4693 ± 34% proc-vmstat.pgmigrate_fail 16249857 ± 11% -39.5% 9827004 ± 31% -39.3% 9858267 ± 13% -24.0% 12351659 ± 11% proc-vmstat.pgmigrate_success 2211 +6.1% 2345 +4.6% 2313 ± 2% +53.8% 3401 ± 39% proc-vmstat.pgpgin 20426382 ± 10% -45.8% 11078695 -48.3% 10560458 ± 2% -2.5% 19913760 proc-vmstat.pgrefill 53930 ± 5% -2.1% 52803 +0.5% 54210 ± 6% -27.2% 39267 proc-vmstat.pgreuse 1.191e+09 -26.3% 8.784e+08 -29.2% 8.432e+08 +10.1% 1.312e+09 proc-vmstat.pgscan_direct 1.394e+09 -33.9% 9.209e+08 -36.6% 8.844e+08 +0.2% 1.397e+09 proc-vmstat.pgscan_file 3200 ±122% +251.2% 11238 ± 8% +311.7% 13175 ± 15% +60.3% 5128 ± 11% proc-vmstat.pgscan_khugepaged 2.027e+08 ± 8% -79.0% 42500310 ± 3% -79.7% 41212659 ± 4% -57.9% 85419213 ± 5% proc-vmstat.pgscan_kswapd 8913 ± 5% -27.3% 6476 ± 3% -28.5% 6370 ± 4% -35.2% 5772 ± 2% proc-vmstat.pgskip_normal 8.692e+08 -28.3% 6.229e+08 -31.3% 5.971e+08 +10.5% 9.601e+08 proc-vmstat.pgsteal_direct 1.041e+09 -36.9% 6.566e+08 -39.5% 6.3e+08 -1.0% 1.03e+09 proc-vmstat.pgsteal_file 1.714e+08 ± 7% -80.3% 33771596 ± 4% -80.8% 32889815 ± 5% -59.0% 70248436 ± 6% proc-vmstat.pgsteal_kswapd 24999520 ± 10% -76.1% 5984550 ± 5% -78.5% 5362533 ± 5% -26.9% 18265931 ± 6% proc-vmstat.slabs_scanned 5158406 ± 6% -74.7% 1306603 ± 4% -74.9% 1293326 ± 16% -8.9% 4699442 ± 2% proc-vmstat.workingset_nodereclaim 3779163 ± 3% +7.9% 4077693 -0.5% 3758804 ± 9% +4.4% 3943980 ± 2% proc-vmstat.workingset_nodes 0.78 ± 54% +265.2% 2.84 ± 51% +1249.1% 10.49 ±182% -12.5% 0.68 ±138% perf-sched.sch_delay.avg.ms.__cond_resched.__alloc_frozen_pages_noprof.alloc_pages_mpol.alloc_pages_noprof.__pmd_alloc 1.45 ± 30% +133.0% 3.39 ± 27% +160.4% 3.78 ± 29% -25.1% 1.09 ± 37% perf-sched.sch_delay.avg.ms.__cond_resched.__alloc_frozen_pages_noprof.alloc_pages_mpol.alloc_pages_noprof.pte_alloc_one 0.98 ± 53% +77.2% 1.74 ±136% -100.0% 0.00 -100.0% 0.00 perf-sched.sch_delay.avg.ms.__cond_resched.__alloc_frozen_pages_noprof.alloc_pages_mpol.folio_alloc_mpol_noprof.vma_alloc_folio_noprof 2.39 ± 5% +135.5% 5.63 ± 11% +145.0% 5.86 ± 17% -2.1% 2.34 ± 10% perf-sched.sch_delay.avg.ms.__cond_resched.__alloc_frozen_pages_noprof.alloc_pages_mpol.folio_alloc_noprof.page_cache_ra_order 2.62 ± 56% +78.5% 4.68 ± 24% +133.0% 6.11 ± 39% -12.8% 2.29 ± 30% perf-sched.sch_delay.avg.ms.__cond_resched.__alloc_frozen_pages_noprof.alloc_pages_mpol.folio_alloc_noprof.page_cache_ra_unbounded 2.32 ± 14% -100.0% 0.00 -100.0% 0.00 -100.0% 0.00 perf-sched.sch_delay.avg.ms.__cond_resched.__alloc_pages_direct_compact.__alloc_pages_slowpath.constprop.0 2.56 ± 10% +126.4% 5.80 ± 10% +93.6% 4.96 ± 14% -22.2% 1.99 ± 12% perf-sched.sch_delay.avg.ms.__cond_resched.__alloc_pages_slowpath.constprop.0.__alloc_frozen_pages_noprof 2.04 ± 15% +244.6% 7.04 ±115% +324.0% 8.66 ±166% +16.6% 2.38 ± 14% perf-sched.sch_delay.avg.ms.__cond_resched.__kmalloc_noprof.ifs_alloc.isra.0 1.52 ± 38% -100.0% 0.00 -100.0% 0.00 -100.0% 0.00 perf-sched.sch_delay.avg.ms.__cond_resched.__reset_isolation_suitable.compact_zone.compact_zone_order.try_to_compact_pages 2.40 ± 24% -24.7% 1.80 ±107% +46.8% 3.52 ± 38% -60.1% 0.96 ± 82% perf-sched.sch_delay.avg.ms.__cond_resched.__tlb_batch_free_encoded_pages.tlb_finish_mmu.exit_mmap.__mmput 0.00 ±223% +1.9e+06% 27.79 ±210% -100.0% 0.00 -100.0% 0.00 perf-sched.sch_delay.avg.ms.__cond_resched.__vmalloc_area_node.__vmalloc_node_range_noprof.alloc_thread_stack_node.dup_task_struct 1.54 ± 27% +148.8% 3.83 ± 34% +215.4% 4.86 ± 30% -64.0% 0.55 ± 42% perf-sched.sch_delay.avg.ms.__cond_resched.__wait_for_common.affine_move_task.__set_cpus_allowed_ptr.__sched_setaffinity 1.63 ± 49% +165.4% 4.33 ± 60% +532.2% 10.32 ±146% +56.2% 2.55 ± 14% perf-sched.sch_delay.avg.ms.__cond_resched.down_read.page_cache_ra_order.filemap_fault.__do_fault 2.27 ± 87% -100.0% 0.00 -100.0% 0.00 -100.0% 0.00 perf-sched.sch_delay.avg.ms.__cond_resched.down_read.split_huge_page_to_list_to_order.migrate_pages_batch.migrate_pages_sync 2.44 ± 21% +114.5% 5.24 ± 10% +124.0% 5.48 ± 17% -1.5% 2.41 ± 13% perf-sched.sch_delay.avg.ms.__cond_resched.down_read.xfs_ilock_for_iomap.xfs_read_iomap_begin.iomap_iter 0.01 ±223% +21626.2% 2.35 ± 83% +22370.0% 2.43 ±126% -100.0% 0.00 perf-sched.sch_delay.avg.ms.__cond_resched.down_write.unlink_anon_vmas.free_pgtables.exit_mmap 0.01 ± 64% +14883.3% 1.05 ±135% +12248.2% 0.86 ±115% +598.2% 0.05 ± 97% perf-sched.sch_delay.avg.ms.__cond_resched.down_write_killable.exec_mmap.begin_new_exec.load_elf_binary 4.10 ± 35% -100.0% 0.00 -99.8% 0.01 ±264% -99.1% 0.04 ± 84% perf-sched.sch_delay.avg.ms.__cond_resched.isolate_freepages_block.fast_isolate_freepages.isolate_freepages.compaction_alloc 3.31 ± 61% -100.0% 0.00 -100.0% 0.00 -100.0% 0.00 perf-sched.sch_delay.avg.ms.__cond_resched.isolate_freepages_block.isolate_freepages.compaction_alloc.migrate_folio_unmap 2.20 ± 12% -100.0% 0.00 -100.0% 0.00 -100.0% 0.00 perf-sched.sch_delay.avg.ms.__cond_resched.isolate_migratepages.compact_zone.compact_zone_order.try_to_compact_pages 2.18 ± 14% -100.0% 0.00 -100.0% 0.00 -86.0% 0.30 ±264% perf-sched.sch_delay.avg.ms.__cond_resched.isolate_migratepages_block.isolate_migratepages.compact_zone.compact_zone_order 0.50 ±215% +18617.1% 94.12 ±210% +389.8% 2.46 ±112% -100.0% 0.00 perf-sched.sch_delay.avg.ms.__cond_resched.kmem_cache_alloc_noprof.alloc_pid.copy_process.kernel_clone 0.74 ±150% +236.4% 2.50 ± 62% +399.0% 3.71 ± 69% -2.1% 0.73 ±169% perf-sched.sch_delay.avg.ms.__cond_resched.kmem_cache_alloc_noprof.mas_alloc_nodes.mas_preallocate.vma_shrink 2.69 ± 23% -100.0% 0.00 +2113.9% 59.50 ±264% -65.8% 0.92 ±226% perf-sched.sch_delay.avg.ms.__cond_resched.migrate_pages_batch.migrate_pages_sync.migrate_pages.compact_zone 0.85 ± 51% +90.7% 1.62 ± 84% +66.6% 1.42 ± 28% -99.7% 0.00 ±141% perf-sched.sch_delay.avg.ms.__cond_resched.mutex_lock_killable.pcpu_alloc_noprof.mm_init.alloc_bprm 0.05 ±223% +4260.6% 2.05 ±100% +5335.9% 2.55 ± 59% -89.9% 0.00 ±202% perf-sched.sch_delay.avg.ms.__cond_resched.mutex_lock_killable.pcpu_alloc_noprof.mm_init.dup_mm 2.96 ± 14% -100.0% 0.00 -100.0% 0.00 -100.0% 0.00 perf-sched.sch_delay.avg.ms.__cond_resched.rmap_walk_anon.try_to_migrate.migrate_folio_unmap.migrate_pages_batch 1.66 ± 21% -100.0% 0.00 -100.0% 0.00 +24.0% 2.06 ± 12% perf-sched.sch_delay.avg.ms.__cond_resched.shadow_lru_isolate.__list_lru_walk_one.list_lru_walk_one_irq.do_shrink_slab 3.61 ± 18% -34.7% 2.36 ± 73% -10.3% 3.24 ± 52% -36.1% 2.31 ± 35% perf-sched.sch_delay.avg.ms.__cond_resched.shmem_inode_acct_blocks.shmem_alloc_and_add_folio.shmem_get_folio_gfp.shmem_write_begin 2.61 ± 5% +81.1% 4.73 ± 10% +91.8% 5.01 ± 14% -18.0% 2.14 ± 12% perf-sched.sch_delay.avg.ms.__cond_resched.shrink_folio_list.evict_folios.try_to_shrink_lruvec.shrink_one 1.59 ± 11% +171.8% 4.31 ± 46% +238.6% 5.37 ± 22% -47.8% 0.83 ± 58% perf-sched.sch_delay.avg.ms.__cond_resched.stop_one_cpu.sched_exec.bprm_execve.part 2.68 ± 33% +37.0% 3.67 ± 9% +64.8% 4.42 ± 62% -41.5% 1.57 ± 25% perf-sched.sch_delay.avg.ms.__cond_resched.try_to_shrink_lruvec.shrink_one.shrink_many.shrink_node 0.63 ± 76% +3515.1% 22.67 ±195% +1455.1% 9.75 ±189% +58.3% 0.99 ± 76% perf-sched.sch_delay.avg.ms.__cond_resched.wp_page_copy.__handle_mm_fault.handle_mm_fault.do_user_addr_fault 6.17 ± 89% +97.1% 12.16 ± 37% -18.1% 5.05 ± 66% -96.5% 0.22 ± 90% perf-sched.sch_delay.avg.ms.devkmsg_read.vfs_read.ksys_read.do_syscall_64 0.38 ± 17% +150.2% 0.94 ± 38% +256.4% 1.34 ± 37% -16.1% 0.32 ± 37% perf-sched.sch_delay.avg.ms.do_wait.kernel_wait4.do_syscall_64.entry_SYSCALL_64_after_hwframe 0.86 ± 93% -100.0% 0.00 -98.5% 0.01 ±229% -70.3% 0.25 ±107% perf-sched.sch_delay.avg.ms.io_schedule.migration_entry_wait_on_locked.migration_entry_wait.do_swap_page 0.77 ± 19% +94.0% 1.50 ± 57% +168.3% 2.08 ± 47% -67.1% 0.26 ± 78% perf-sched.sch_delay.avg.ms.irq_thread.kthread.ret_from_fork.ret_from_fork_asm 2.53 ± 15% +133.5% 5.91 ± 41% +142.1% 6.12 ± 32% -6.5% 2.36 ± 10% perf-sched.sch_delay.avg.ms.irqentry_exit_to_user_mode.asm_exc_page_fault.[unknown] 0.92 ± 41% +60.5% 1.48 ±100% +191.6% 2.69 ± 54% -30.9% 0.64 ±103% perf-sched.sch_delay.avg.ms.irqentry_exit_to_user_mode.asm_exc_page_fault.[unknown].[unknown] 2.82 ± 16% +130.6% 6.49 ± 40% +88.1% 5.30 ± 15% -2.7% 2.74 ± 7% perf-sched.sch_delay.avg.ms.irqentry_exit_to_user_mode.asm_sysvec_apic_timer_interrupt.[unknown] 0.38 ±105% +73.0% 0.66 ± 66% -100.0% 0.00 -100.0% 0.00 perf-sched.sch_delay.avg.ms.pipe_read.vfs_read.ksys_read.do_syscall_64 1.31 ±141% +251.7% 4.62 ±100% -100.0% 0.00 -100.0% 0.00 perf-sched.sch_delay.avg.ms.pipe_write.vfs_write.ksys_write.do_syscall_64 2.23 ± 46% +14.1% 2.54 ± 42% -100.0% 0.00 -100.0% 0.00 perf-sched.sch_delay.avg.ms.schedule_hrtimeout_range.do_poll.constprop.0.do_sys_poll 0.79 ± 43% +1820.1% 15.24 ±188% -100.0% 0.00 -100.0% 0.00 perf-sched.sch_delay.avg.ms.schedule_hrtimeout_range.ep_poll.do_epoll_wait.__x64_sys_epoll_wait 6.19 ±169% -100.0% 0.00 -100.0% 0.00 -100.0% 0.00 ±264% perf-sched.sch_delay.avg.ms.schedule_preempt_disabled.__mutex_lock.constprop.0.__drain_all_pages 0.97 ± 14% +55.7% 1.51 ± 90% +136.6% 2.30 ± 72% -68.0% 0.31 ± 70% perf-sched.sch_delay.avg.ms.schedule_timeout.__wait_for_common.wait_for_completion_state.kernel_clone 3.88 ±140% +59.5% 6.18 ±158% +30.2% 5.05 ±171% -92.5% 0.29 ± 67% perf-sched.sch_delay.avg.ms.schedule_timeout.kcompactd.kthread.ret_from_fork 0.11 ±130% -100.0% 0.00 -93.9% 0.01 ±132% -87.2% 0.01 ± 89% perf-sched.sch_delay.avg.ms.schedule_timeout.kswapd_try_to_sleep.kswapd.kthread 0.79 ± 15% +70.1% 1.34 ± 61% +164.1% 2.08 ± 80% -70.4% 0.23 ± 43% perf-sched.sch_delay.avg.ms.schedule_timeout.rcu_gp_fqs_loop.rcu_gp_kthread.kthread 0.59 ± 23% +224.0% 1.91 ± 26% +158.7% 1.52 ± 37% -55.4% 0.26 ± 11% perf-sched.sch_delay.avg.ms.smpboot_thread_fn.kthread.ret_from_fork.ret_from_fork_asm 0.16 ± 65% +272.6% 0.61 ±198% -5.7% 0.16 ±163% -81.2% 0.03 ± 56% perf-sched.sch_delay.avg.ms.syscall_exit_to_user_mode.do_syscall_64.entry_SYSCALL_64_after_hwframe.[unknown] 1.36 ± 29% +9.2% 1.49 ±117% +29.4% 1.76 ± 85% -67.8% 0.44 ±104% perf-sched.sch_delay.avg.ms.wait_for_partner.fifo_open.do_dentry_open.vfs_open 1.26 ± 24% +181.3% 3.54 ± 23% +186.8% 3.61 ± 17% -62.1% 0.48 ± 7% perf-sched.sch_delay.avg.ms.worker_thread.kthread.ret_from_fork.ret_from_fork_asm 2.38 ± 60% +527.9% 14.95 ± 78% +6008.8% 145.45 ±240% -11.9% 2.10 ±132% perf-sched.sch_delay.max.ms.__cond_resched.__alloc_frozen_pages_noprof.alloc_pages_mpol.alloc_pages_noprof.__pmd_alloc 4.40 ± 38% +250.7% 15.42 ± 19% +381.9% 21.18 ± 61% -33.0% 2.94 ± 39% perf-sched.sch_delay.max.ms.__cond_resched.__alloc_frozen_pages_noprof.alloc_pages_mpol.alloc_pages_noprof.pte_alloc_one 1.34 ± 46% +118.8% 2.93 ±147% -100.0% 0.00 -100.0% 0.00 perf-sched.sch_delay.max.ms.__cond_resched.__alloc_frozen_pages_noprof.alloc_pages_mpol.folio_alloc_mpol_noprof.vma_alloc_folio_noprof 208.26 ± 51% +146.1% 512.53 ± 45% +235.7% 699.12 ± 43% -60.2% 82.98 ± 43% perf-sched.sch_delay.max.ms.__cond_resched.__alloc_frozen_pages_noprof.alloc_pages_mpol.folio_alloc_noprof.page_cache_ra_order 18.01 ± 57% -100.0% 0.00 -100.0% 0.00 -100.0% 0.00 perf-sched.sch_delay.max.ms.__cond_resched.__alloc_pages_direct_compact.__alloc_pages_slowpath.constprop.0 128.58 ± 48% +493.6% 763.29 ± 40% +203.4% 390.06 ± 60% -60.1% 51.28 ± 51% perf-sched.sch_delay.max.ms.__cond_resched.__alloc_pages_slowpath.constprop.0.__alloc_frozen_pages_noprof 1.13 ±207% +550.5% 7.37 ± 84% +675.5% 8.78 ± 64% -57.8% 0.48 ±249% perf-sched.sch_delay.max.ms.__cond_resched.__anon_vma_prepare.__vmf_anon_prepare.do_anonymous_page.__handle_mm_fault 6.41 ± 42% -100.0% 0.00 -100.0% 0.00 -100.0% 0.00 perf-sched.sch_delay.max.ms.__cond_resched.__reset_isolation_suitable.compact_zone.compact_zone_order.try_to_compact_pages 6.74 ± 39% +49.0% 10.05 ±123% +97.5% 13.32 ± 40% -57.0% 2.90 ± 73% perf-sched.sch_delay.max.ms.__cond_resched.__tlb_batch_free_encoded_pages.tlb_finish_mmu.exit_mmap.__mmput 0.00 ±223% +1.2e+07% 186.87 ±216% -100.0% 0.00 -100.0% 0.00 perf-sched.sch_delay.max.ms.__cond_resched.__vmalloc_area_node.__vmalloc_node_range_noprof.alloc_thread_stack_node.dup_task_struct 3.42 ± 46% +1591.8% 57.78 ±149% +260.1% 12.30 ± 35% -53.8% 1.58 ±100% perf-sched.sch_delay.max.ms.__cond_resched.change_pmd_range.isra.0.change_pud_range 4.63 ± 77% -100.0% 0.00 -100.0% 0.00 -100.0% 0.00 perf-sched.sch_delay.max.ms.__cond_resched.down_read.split_huge_page_to_list_to_order.migrate_pages_batch.migrate_pages_sync 20.61 ± 74% +914.8% 209.17 ±121% +1062.6% 239.64 ± 80% +85.9% 38.32 ±108% perf-sched.sch_delay.max.ms.__cond_resched.down_read.xfs_ilock_for_iomap.xfs_read_iomap_begin.iomap_iter 0.01 ±223% +43838.5% 4.76 ± 83% +37171.5% 4.04 ±129% -100.0% 0.00 perf-sched.sch_delay.max.ms.__cond_resched.down_write.unlink_anon_vmas.free_pgtables.exit_mmap 0.03 ±168% +25398.1% 8.84 ±113% +23276.6% 8.10 ± 53% +2781.7% 1.00 ±141% perf-sched.sch_delay.max.ms.__cond_resched.down_write_killable.exec_mmap.begin_new_exec.load_elf_binary 9.80 ± 40% -65.4% 3.39 ±129% -87.7% 1.21 ±162% -58.8% 4.04 ± 55% perf-sched.sch_delay.max.ms.__cond_resched.generic_perform_write.shmem_file_write_iter.vfs_write.ksys_write 23.01 ±114% -100.0% 0.00 -100.0% 0.01 ±264% -99.8% 0.06 ± 64% perf-sched.sch_delay.max.ms.__cond_resched.isolate_freepages_block.fast_isolate_freepages.isolate_freepages.compaction_alloc 5.92 ± 48% -100.0% 0.00 -100.0% 0.00 -100.0% 0.00 perf-sched.sch_delay.max.ms.__cond_resched.isolate_freepages_block.isolate_freepages.compaction_alloc.migrate_folio_unmap 25.32 ± 28% -100.0% 0.00 -100.0% 0.00 -100.0% 0.00 perf-sched.sch_delay.max.ms.__cond_resched.isolate_migratepages.compact_zone.compact_zone_order.try_to_compact_pages 17.84 ± 26% -100.0% 0.00 -100.0% 0.00 -97.9% 0.37 ±264% perf-sched.sch_delay.max.ms.__cond_resched.isolate_migratepages_block.isolate_migratepages.compact_zone.compact_zone_order 0.51 ±210% +36510.2% 187.87 ±210% +764.1% 4.43 ± 93% -100.0% 0.00 perf-sched.sch_delay.max.ms.__cond_resched.kmem_cache_alloc_noprof.alloc_pid.copy_process.kernel_clone 0.74 ±150% +731.1% 6.18 ± 68% +755.9% 6.37 ± 95% +16.2% 0.86 ±153% perf-sched.sch_delay.max.ms.__cond_resched.kmem_cache_alloc_noprof.mas_alloc_nodes.mas_preallocate.vma_shrink 9.68 ± 42% -100.0% 0.00 +1716.6% 175.82 ±264% -79.7% 1.96 ±156% perf-sched.sch_delay.max.ms.__cond_resched.migrate_pages_batch.migrate_pages_sync.migrate_pages.compact_zone 1.89 ± 44% +441.0% 10.22 ± 79% +292.0% 7.41 ± 47% -99.8% 0.00 ±155% perf-sched.sch_delay.max.ms.__cond_resched.mutex_lock_killable.pcpu_alloc_noprof.mm_init.alloc_bprm 0.05 ±223% +10820.9% 5.13 ±101% +13454.5% 6.37 ± 47% -89.9% 0.00 ±202% perf-sched.sch_delay.max.ms.__cond_resched.mutex_lock_killable.pcpu_alloc_noprof.mm_init.dup_mm 7.52 ± 19% -100.0% 0.00 -100.0% 0.00 -100.0% 0.00 perf-sched.sch_delay.max.ms.__cond_resched.rmap_walk_anon.try_to_migrate.migrate_folio_unmap.migrate_pages_batch 26.92 ± 77% -100.0% 0.00 -100.0% 0.00 -61.7% 10.31 ± 23% perf-sched.sch_delay.max.ms.__cond_resched.shadow_lru_isolate.__list_lru_walk_one.list_lru_walk_one_irq.do_shrink_slab 200.50 ± 36% +180.0% 561.35 ± 52% +249.3% 700.26 ± 27% -61.2% 77.75 ± 64% perf-sched.sch_delay.max.ms.__cond_resched.shrink_folio_list.evict_folios.try_to_shrink_lruvec.shrink_one 3.26 ± 29% -28.7% 2.32 ± 60% -26.5% 2.39 ± 78% -66.5% 1.09 ±102% perf-sched.sch_delay.max.ms.__cond_resched.smpboot_thread_fn.kthread.ret_from_fork.ret_from_fork_asm 1.21 ± 71% +17474.9% 213.15 ±210% +11182.9% 136.84 ±239% +127.5% 2.76 ± 94% perf-sched.sch_delay.max.ms.__cond_resched.wp_page_copy.__handle_mm_fault.handle_mm_fault.do_user_addr_fault 6.98 ± 71% +81.7% 12.68 ± 30% -21.7% 5.46 ± 60% -94.7% 0.37 ± 85% perf-sched.sch_delay.max.ms.devkmsg_read.vfs_read.ksys_read.do_syscall_64 5.86 ± 47% +207.0% 18.00 ± 94% +426.3% 30.86 ± 61% +9.5% 6.42 ± 50% perf-sched.sch_delay.max.ms.do_wait.kernel_wait4.do_syscall_64.entry_SYSCALL_64_after_hwframe 15.65 ±168% -100.0% 0.00 -99.9% 0.02 ±240% -88.3% 1.83 ±109% perf-sched.sch_delay.max.ms.io_schedule.migration_entry_wait_on_locked.migration_entry_wait.do_swap_page 2.60 ± 31% +273.0% 9.68 ± 82% +416.2% 13.40 ± 42% -19.7% 2.08 ± 83% perf-sched.sch_delay.max.ms.irq_thread.kthread.ret_from_fork.ret_from_fork_asm 1.87 ± 38% +213.0% 5.85 ± 93% +651.4% 14.04 ± 17% +8.3% 2.02 ± 88% perf-sched.sch_delay.max.ms.irqentry_exit_to_user_mode.asm_exc_page_fault.[unknown].[unknown] 368.10 ±127% +154.7% 937.69 ± 43% -100.0% 0.00 -100.0% 0.00 perf-sched.sch_delay.max.ms.pipe_read.vfs_read.ksys_read.do_syscall_64 1.31 ±141% +356.6% 6.00 ± 82% -100.0% 0.00 -100.0% 0.00 perf-sched.sch_delay.max.ms.pipe_write.vfs_write.ksys_write.do_syscall_64 19.65 ±105% -22.5% 15.23 ± 57% -100.0% 0.00 -100.0% 0.00 perf-sched.sch_delay.max.ms.schedule_hrtimeout_range.do_poll.constprop.0.do_sys_poll 4.84 ± 74% +3625.1% 180.43 ±206% -100.0% 0.00 -100.0% 0.00 perf-sched.sch_delay.max.ms.schedule_hrtimeout_range.ep_poll.do_epoll_wait.__x64_sys_epoll_wait 62.50 ±201% -100.0% 0.00 -100.0% 0.00 -100.0% 0.00 ±264% perf-sched.sch_delay.max.ms.schedule_preempt_disabled.__mutex_lock.constprop.0.__drain_all_pages 121.47 ±162% +52.3% 185.00 ±204% -34.2% 79.91 ±215% -98.3% 2.12 ± 54% perf-sched.sch_delay.max.ms.schedule_timeout.kcompactd.kthread.ret_from_fork 0.86 ±151% -100.0% 0.00 -98.7% 0.01 ±154% -96.2% 0.03 ± 92% perf-sched.sch_delay.max.ms.schedule_timeout.kswapd_try_to_sleep.kswapd.kthread 58.28 ±121% +153.5% 147.75 ±103% +490.6% 344.22 ±119% -86.4% 7.95 ± 89% perf-sched.sch_delay.max.ms.schedule_timeout.rcu_gp_fqs_loop.rcu_gp_kthread.kthread 184.61 ± 15% +245.5% 637.81 ± 19% +212.8% 577.38 ± 28% -68.2% 58.76 ± 22% perf-sched.sch_delay.max.ms.smpboot_thread_fn.kthread.ret_from_fork.ret_from_fork_asm 184.34 ± 18% +264.7% 672.25 ± 31% +364.8% 856.74 ± 55% -65.4% 63.72 ± 24% perf-sched.sch_delay.max.ms.worker_thread.kthread.ret_from_fork.ret_from_fork_asm 1.66 ± 9% +93.7% 3.21 ± 20% +78.5% 2.96 ± 14% -20.9% 1.31 ± 13% perf-sched.total_sch_delay.average.ms 662.36 ± 55% +73.5% 1148 ± 9% +89.2% 1253 ± 27% -71.9% 185.98 ± 87% perf-sched.total_sch_delay.max.ms 133.24 ± 5% -7.9% 122.73 ± 12% -12.3% 116.85 ± 9% -23.6% 101.83 ± 8% perf-sched.total_wait_and_delay.average.ms 20314 ± 10% -3.3% 19639 ± 14% -0.9% 20122 ± 17% +44.6% 29380 ± 8% perf-sched.total_wait_and_delay.count.ms 131.58 ± 5% -9.2% 119.52 ± 12% -13.4% 113.89 ± 9% -23.6% 100.51 ± 9% perf-sched.total_wait_time.average.ms 4.79 ± 5% +135.3% 11.27 ± 11% +144.9% 11.72 ± 17% -2.1% 4.68 ± 10% perf-sched.wait_and_delay.avg.ms.__cond_resched.__alloc_frozen_pages_noprof.alloc_pages_mpol.folio_alloc_noprof.page_cache_ra_order 5.31 ± 6% +123.3% 11.86 ± 12% +106.7% 10.98 ± 14% -7.5% 4.91 ± 16% perf-sched.wait_and_delay.avg.ms.__cond_resched.__alloc_pages_slowpath.constprop.0.__alloc_frozen_pages_noprof 4.08 ± 15% -100.0% 0.00 -100.0% 0.00 +16.6% 4.76 ± 14% perf-sched.wait_and_delay.avg.ms.__cond_resched.__kmalloc_noprof.ifs_alloc.isra.0 6.19 ± 13% +69.9% 10.51 ± 19% +96.3% 12.14 ± 20% -32.9% 4.15 ± 13% perf-sched.wait_and_delay.avg.ms.__cond_resched.__wait_for_common.affine_move_task.__set_cpus_allowed_ptr.__sched_setaffinity 3.23 ± 75% +224.3% 10.49 ± 10% +238.7% 10.95 ± 17% +49.0% 4.82 ± 13% perf-sched.wait_and_delay.avg.ms.__cond_resched.down_read.xfs_ilock_for_iomap.xfs_read_iomap_begin.iomap_iter 4.40 ± 12% -100.0% 0.00 -100.0% 0.00 -100.0% 0.00 perf-sched.wait_and_delay.avg.ms.__cond_resched.isolate_migratepages.compact_zone.compact_zone_order.try_to_compact_pages 4.58 ± 20% -100.0% 0.00 -100.0% 0.00 -100.0% 0.00 perf-sched.wait_and_delay.avg.ms.__cond_resched.isolate_migratepages_block.isolate_migratepages.compact_zone.compact_zone_order 2.61 ± 49% -100.0% 0.00 -100.0% 0.00 +85.2% 4.84 ± 42% perf-sched.wait_and_delay.avg.ms.__cond_resched.shadow_lru_isolate.__list_lru_walk_one.list_lru_walk_one_irq.do_shrink_slab 5.21 ± 5% +132.3% 12.11 ± 8% +141.9% 12.61 ± 12% +5.9% 5.52 ± 11% perf-sched.wait_and_delay.avg.ms.__cond_resched.shrink_folio_list.evict_folios.try_to_shrink_lruvec.shrink_one 4.92 ± 22% +130.6% 11.35 ± 29% +183.4% 13.95 ± 61% -75.7% 1.19 ±140% perf-sched.wait_and_delay.avg.ms.__cond_resched.try_to_shrink_lruvec.shrink_one.shrink_many.shrink_node 4.40 ± 18% +241.9% 15.06 ± 39% +331.2% 18.99 ± 26% -69.1% 1.36 ±180% perf-sched.wait_and_delay.avg.ms.do_wait.kernel_wait4.do_syscall_64.entry_SYSCALL_64_after_hwframe 400.86 +2.5% 410.98 ± 3% -28.8% 285.40 ± 19% -22.0% 312.53 ± 7% perf-sched.wait_and_delay.avg.ms.irq_thread.kthread.ret_from_fork.ret_from_fork_asm 4.99 ± 16% +136.4% 11.80 ± 41% +145.4% 12.25 ± 32% -5.3% 4.73 ± 10% perf-sched.wait_and_delay.avg.ms.irqentry_exit_to_user_mode.asm_exc_page_fault.[unknown] 5.66 ± 16% -100.0% 0.00 -100.0% 0.00 +1.4% 5.74 ± 8% perf-sched.wait_and_delay.avg.ms.irqentry_exit_to_user_mode.asm_sysvec_apic_timer_interrupt.[unknown] 49.92 ± 40% -58.5% 20.72 ± 31% -100.0% 0.00 -100.0% 0.00 perf-sched.wait_and_delay.avg.ms.pipe_read.vfs_read.ksys_read.do_syscall_64 467.69 ± 7% -5.3% 442.80 ± 7% -100.0% 0.00 -100.0% 0.00 perf-sched.wait_and_delay.avg.ms.schedule_hrtimeout_range.do_poll.constprop.0.do_sys_poll 387.51 ± 4% +6.8% 413.92 ± 5% -100.0% 0.00 -100.0% 0.00 perf-sched.wait_and_delay.avg.ms.schedule_hrtimeout_range.ep_poll.do_epoll_wait.__x64_sys_epoll_wait 5.67 ± 7% +233.8% 18.94 ± 43% +218.4% 18.07 ± 59% -100.0% 0.00 perf-sched.wait_and_delay.avg.ms.schedule_timeout.__wait_for_common.wait_for_completion_state.kernel_clone 308.58 ± 7% +52.0% 469.04 ± 3% +47.9% 456.38 ± 3% +42.4% 439.45 ± 2% perf-sched.wait_and_delay.avg.ms.schedule_timeout.kcompactd.kthread.ret_from_fork 5.96 ± 4% +34.5% 8.01 ± 22% +53.9% 9.16 ± 27% -22.3% 4.63 ± 3% perf-sched.wait_and_delay.avg.ms.schedule_timeout.rcu_gp_fqs_loop.rcu_gp_kthread.kthread 621.78 ± 7% +35.1% 839.89 ± 10% +42.5% 886.03 ± 10% -8.9% 566.75 ± 8% perf-sched.wait_and_delay.avg.ms.smpboot_thread_fn.kthread.ret_from_fork.ret_from_fork_asm 3130 ± 11% +5.8% 3310 ± 20% -3.6% 3018 ± 16% +99.8% 6255 ± 18% perf-sched.wait_and_delay.count.__cond_resched.__alloc_frozen_pages_noprof.alloc_pages_mpol.folio_alloc_noprof.page_cache_ra_order 1626 ± 6% -23.8% 1239 ± 21% -30.0% 1138 ± 15% -6.1% 1528 ± 9% perf-sched.wait_and_delay.count.__cond_resched.__alloc_pages_slowpath.constprop.0.__alloc_frozen_pages_noprof 148.00 ± 23% -100.0% 0.00 -100.0% 0.00 +150.2% 370.25 ± 21% perf-sched.wait_and_delay.count.__cond_resched.__kmalloc_noprof.ifs_alloc.isra.0 1042 ± 7% -17.3% 861.83 ± 15% -21.0% 823.50 ± 14% +22.3% 1275 ± 6% perf-sched.wait_and_delay.count.__cond_resched.__wait_for_common.affine_move_task.__set_cpus_allowed_ptr.__sched_setaffinity 70.50 ± 72% +505.7% 427.00 ± 29% +478.0% 407.50 ± 21% +931.9% 727.50 ± 16% perf-sched.wait_and_delay.count.__cond_resched.down_read.xfs_ilock_for_iomap.xfs_read_iomap_begin.iomap_iter 394.83 ± 11% -100.0% 0.00 -100.0% 0.00 -100.0% 0.00 perf-sched.wait_and_delay.count.__cond_resched.isolate_migratepages.compact_zone.compact_zone_order.try_to_compact_pages 696.83 ± 25% -100.0% 0.00 -100.0% 0.00 -100.0% 0.00 perf-sched.wait_and_delay.count.__cond_resched.isolate_migratepages_block.isolate_migratepages.compact_zone.compact_zone_order 0.17 ±223% +2800.0% 4.83 ± 45% +2225.0% 3.88 ± 52% +50.0% 0.25 ±173% perf-sched.wait_and_delay.count.__cond_resched.mutex_lock_killable.pcpu_alloc_noprof.mm_init.dup_mm 3434 ± 7% -11.2% 3049 ± 20% -17.8% 2823 ± 15% +44.1% 4949 ± 14% perf-sched.wait_and_delay.count.__cond_resched.shrink_folio_list.evict_folios.try_to_shrink_lruvec.shrink_one 50.00 ± 18% -74.0% 13.00 ± 65% -65.5% 17.25 ± 40% -72.8% 13.62 ± 99% perf-sched.wait_and_delay.count.__cond_resched.smpboot_thread_fn.kthread.ret_from_fork.ret_from_fork_asm 135.83 ± 6% -89.7% 14.00 ±223% -90.9% 12.38 ±264% -88.2% 16.00 ±264% perf-sched.wait_and_delay.count.__cond_resched.stop_one_cpu.sched_exec.bprm_execve.part 10.00 +13.3% 11.33 ± 19% +83.8% 18.38 ± 11% +48.8% 14.88 ± 7% perf-sched.wait_and_delay.count.irq_thread.kthread.ret_from_fork.ret_from_fork_asm 206.83 ± 10% +18.2% 244.50 ± 19% +9.3% 226.00 ± 17% +50.7% 311.62 ± 15% perf-sched.wait_and_delay.count.irqentry_exit_to_user_mode.asm_exc_page_fault.[unknown] 168.50 ± 16% -100.0% 0.00 -100.0% 0.00 +237.2% 568.25 ± 19% perf-sched.wait_and_delay.count.irqentry_exit_to_user_mode.asm_sysvec_apic_timer_interrupt.[unknown] 1644 ± 45% +103.2% 3340 ± 25% -100.0% 0.00 -100.0% 0.00 perf-sched.wait_and_delay.count.pipe_read.vfs_read.ksys_read.do_syscall_64 26.00 ± 7% -9.6% 23.50 ± 15% -100.0% 0.00 -100.0% 0.00 perf-sched.wait_and_delay.count.schedule_hrtimeout_range.do_poll.constprop.0.do_sys_poll 13.00 +7.7% 14.00 ± 15% -100.0% 0.00 -100.0% 0.00 perf-sched.wait_and_delay.count.schedule_hrtimeout_range.ep_poll.do_epoll_wait.__x64_sys_epoll_wait 115.67 ± 3% -6.8% 107.83 ± 19% -18.6% 94.12 ± 43% -100.0% 0.00 perf-sched.wait_and_delay.count.schedule_timeout.__wait_for_common.wait_for_completion_state.kernel_clone 985.67 ± 2% -7.3% 914.17 ± 24% -14.2% 845.25 ± 24% +21.4% 1196 ± 4% perf-sched.wait_and_delay.count.schedule_timeout.rcu_gp_fqs_loop.rcu_gp_kthread.kthread 2500 ± 15% -49.3% 1268 ± 11% -51.6% 1210 ± 11% +27.4% 3185 ± 17% perf-sched.wait_and_delay.count.smpboot_thread_fn.kthread.ret_from_fork.ret_from_fork_asm 1105 ± 80% +118.9% 2420 ± 58% +199.3% 3309 ± 38% +112.7% 2351 ± 45% perf-sched.wait_and_delay.count.syscall_exit_to_user_mode.do_syscall_64.entry_SYSCALL_64_after_hwframe.[unknown] 416.51 ± 51% +146.1% 1025 ± 45% +235.7% 1398 ± 43% -60.2% 165.95 ± 43% perf-sched.wait_and_delay.max.ms.__cond_resched.__alloc_frozen_pages_noprof.alloc_pages_mpol.folio_alloc_noprof.page_cache_ra_order 494.23 ± 70% +158.3% 1276 ± 27% +100.5% 991.08 ± 42% +84.8% 913.47 ± 34% perf-sched.wait_and_delay.max.ms.__cond_resched.__alloc_pages_slowpath.constprop.0.__alloc_frozen_pages_noprof 43.31 ± 65% -100.0% 0.00 -100.0% 0.00 -19.4% 34.90 ± 68% perf-sched.wait_and_delay.max.ms.__cond_resched.__kmalloc_noprof.ifs_alloc.isra.0 32.05 ±114% +1205.3% 418.33 ±121% +1395.4% 479.28 ± 80% +139.1% 76.64 ±108% perf-sched.wait_and_delay.max.ms.__cond_resched.down_read.xfs_ilock_for_iomap.xfs_read_iomap_begin.iomap_iter 50.65 ± 28% -100.0% 0.00 -100.0% 0.00 -100.0% 0.00 perf-sched.wait_and_delay.max.ms.__cond_resched.isolate_migratepages.compact_zone.compact_zone_order.try_to_compact_pages 196.51 ±182% -100.0% 0.00 -100.0% 0.00 -100.0% 0.00 perf-sched.wait_and_delay.max.ms.__cond_resched.isolate_migratepages_block.isolate_migratepages.compact_zone.compact_zone_order 167.85 ±223% +511.9% 1027 +443.6% 912.41 ± 38% +49.2% 250.43 ±173% perf-sched.wait_and_delay.max.ms.__cond_resched.mutex_lock_killable.pcpu_alloc_noprof.mm_init.dup_mm 401.00 ± 36% +238.8% 1358 ± 29% +247.9% 1395 ± 19% +313.1% 1656 ± 46% perf-sched.wait_and_delay.max.ms.__cond_resched.shrink_folio_list.evict_folios.try_to_shrink_lruvec.shrink_one 3117 ± 18% -31.6% 2131 ± 67% -16.8% 2593 ± 41% -58.2% 1304 ± 67% perf-sched.wait_and_delay.max.ms.__cond_resched.smpboot_thread_fn.kthread.ret_from_fork.ret_from_fork_asm 19.38 ± 38% -62.3% 7.31 ±223% -73.0% 5.23 ±264% -91.2% 1.70 ±264% perf-sched.wait_and_delay.max.ms.__cond_resched.stop_one_cpu.sched_exec.bprm_execve.part 105.15 ±102% +558.6% 692.52 ± 66% +598.5% 734.40 ± 63% -95.2% 5.04 ±135% perf-sched.wait_and_delay.max.ms.__cond_resched.try_to_shrink_lruvec.shrink_one.shrink_many.shrink_node 44.08 ± 33% +151.5% 110.86 ± 31% +660.4% 335.24 ±119% -40.4% 26.29 ±203% perf-sched.wait_and_delay.max.ms.do_wait.kernel_wait4.do_syscall_64.entry_SYSCALL_64_after_hwframe 54.90 ± 67% -100.0% 0.00 -100.0% 0.00 +281.8% 209.58 ±147% perf-sched.wait_and_delay.max.ms.irqentry_exit_to_user_mode.asm_sysvec_apic_timer_interrupt.[unknown] 1475 ± 5% +51.9% 2241 ± 19% -100.0% 0.00 -100.0% 0.00 perf-sched.wait_and_delay.max.ms.pipe_read.vfs_read.ksys_read.do_syscall_64 1030 ± 4% -0.3% 1027 -100.0% 0.00 -100.0% 0.00 perf-sched.wait_and_delay.max.ms.schedule_hrtimeout_range.do_poll.constprop.0.do_sys_poll 505.03 +19.3% 602.66 ± 30% -100.0% 0.00 -100.0% 0.00 perf-sched.wait_and_delay.max.ms.schedule_hrtimeout_range.ep_poll.do_epoll_wait.__x64_sys_epoll_wait 44.18 ± 28% +522.5% 275.00 ±119% +453.2% 244.39 ± 92% -100.0% 0.00 perf-sched.wait_and_delay.max.ms.schedule_timeout.__wait_for_common.wait_for_completion_state.kernel_clone 119.88 ±117% +153.2% 303.58 ± 99% +347.4% 536.35 ± 98% -82.9% 20.48 ± 70% perf-sched.wait_and_delay.max.ms.schedule_timeout.rcu_gp_fqs_loop.rcu_gp_kthread.kthread 1.44 ± 29% +1635.4% 24.97 ±130% +714.9% 11.73 ±187% -30.0% 1.01 ± 52% perf-sched.wait_time.avg.ms.__cond_resched.__alloc_frozen_pages_noprof.alloc_pages_mpol.alloc_pages_noprof.pte_alloc_one 0.98 ± 53% +75.6% 1.73 ±138% -100.0% 0.00 -100.0% 0.00 perf-sched.wait_time.avg.ms.__cond_resched.__alloc_frozen_pages_noprof.alloc_pages_mpol.folio_alloc_mpol_noprof.vma_alloc_folio_noprof 2.40 ± 5% +135.1% 5.63 ± 11% +144.7% 5.86 ± 17% -2.2% 2.34 ± 10% perf-sched.wait_time.avg.ms.__cond_resched.__alloc_frozen_pages_noprof.alloc_pages_mpol.folio_alloc_noprof.page_cache_ra_order 2.62 ± 56% +78.5% 4.68 ± 24% +133.0% 6.11 ± 39% -12.8% 2.29 ± 30% perf-sched.wait_time.avg.ms.__cond_resched.__alloc_frozen_pages_noprof.alloc_pages_mpol.folio_alloc_noprof.page_cache_ra_unbounded 2.32 ± 14% -100.0% 0.00 -100.0% 0.00 -100.0% 0.00 perf-sched.wait_time.avg.ms.__cond_resched.__alloc_pages_direct_compact.__alloc_pages_slowpath.constprop.0 2.75 ± 5% +120.5% 6.06 ± 16% +118.9% 6.01 ± 18% +6.3% 2.92 ± 23% perf-sched.wait_time.avg.ms.__cond_resched.__alloc_pages_slowpath.constprop.0.__alloc_frozen_pages_noprof 0.18 ±223% +1644.4% 3.11 ± 99% +1583.1% 3.00 ± 84% -100.0% 0.00 perf-sched.wait_time.avg.ms.__cond_resched.__get_user_pages.get_user_pages_remote.get_arg_page.copy_string_kernel 2.04 ± 15% +244.6% 7.04 ±115% +323.9% 8.66 ±166% +16.6% 2.38 ± 14% perf-sched.wait_time.avg.ms.__cond_resched.__kmalloc_noprof.ifs_alloc.isra.0 1.52 ± 38% -100.0% 0.00 -100.0% 0.00 -100.0% 0.00 perf-sched.wait_time.avg.ms.__cond_resched.__reset_isolation_suitable.compact_zone.compact_zone_order.try_to_compact_pages 0.00 ±223% +9.9e+06% 149.06 ± 94% -100.0% 0.00 -100.0% 0.00 perf-sched.wait_time.avg.ms.__cond_resched.__vmalloc_area_node.__vmalloc_node_range_noprof.alloc_thread_stack_node.dup_task_struct 4.65 ± 9% +43.8% 6.68 ± 13% +56.8% 7.29 ± 16% -22.6% 3.59 ± 9% perf-sched.wait_time.avg.ms.__cond_resched.__wait_for_common.affine_move_task.__set_cpus_allowed_ptr.__sched_setaffinity 1.63 ± 49% +165.4% 4.33 ± 60% +532.2% 10.32 ±146% +56.2% 2.55 ± 14% perf-sched.wait_time.avg.ms.__cond_resched.down_read.page_cache_ra_order.filemap_fault.__do_fault 28.89 ±104% -100.0% 0.00 -100.0% 0.00 -100.0% 0.00 perf-sched.wait_time.avg.ms.__cond_resched.down_read.split_huge_page_to_list_to_order.migrate_pages_batch.migrate_pages_sync 2.45 ± 21% +114.3% 5.24 ± 10% +123.8% 5.48 ± 17% -1.5% 2.41 ± 13% perf-sched.wait_time.avg.ms.__cond_resched.down_read.xfs_ilock_for_iomap.xfs_read_iomap_begin.iomap_iter 0.00 ±203% +40515.4% 0.88 ±155% +34769.2% 0.76 ±133% +1619.2% 0.04 ±130% perf-sched.wait_time.avg.ms.__cond_resched.down_write_killable.exec_mmap.begin_new_exec.load_elf_binary 131.00 ±166% -98.4% 2.13 ±155% -99.3% 0.92 ±195% +20.4% 157.67 ±128% perf-sched.wait_time.avg.ms.__cond_resched.generic_perform_write.shmem_file_write_iter.vfs_write.ksys_write 45.09 ±137% -100.0% 0.00 -100.0% 0.01 ±264% -87.7% 5.56 ±262% perf-sched.wait_time.avg.ms.__cond_resched.isolate_freepages_block.fast_isolate_freepages.isolate_freepages.compaction_alloc 3.35 ± 59% -100.0% 0.00 -100.0% 0.00 -100.0% 0.00 perf-sched.wait_time.avg.ms.__cond_resched.isolate_freepages_block.isolate_freepages.compaction_alloc.migrate_folio_unmap 2.20 ± 12% -100.0% 0.00 -100.0% 0.00 -100.0% 0.00 perf-sched.wait_time.avg.ms.__cond_resched.isolate_migratepages.compact_zone.compact_zone_order.try_to_compact_pages 2.40 ± 27% -100.0% 0.00 -100.0% 0.00 -87.3% 0.30 ±264% perf-sched.wait_time.avg.ms.__cond_resched.isolate_migratepages_block.isolate_migratepages.compact_zone.compact_zone_order 0.74 ±150% +235.3% 2.49 ± 62% +398.9% 3.71 ± 69% -2.1% 0.73 ±169% perf-sched.wait_time.avg.ms.__cond_resched.kmem_cache_alloc_noprof.mas_alloc_nodes.mas_preallocate.vma_shrink 20.77 ± 70% -100.0% 0.00 +403.9% 104.64 ±175% +84.7% 38.36 ±132% perf-sched.wait_time.avg.ms.__cond_resched.migrate_pages_batch.migrate_pages_sync.migrate_pages.compact_zone 2.96 ± 14% -100.0% 0.00 -100.0% 0.00 -100.0% 0.00 perf-sched.wait_time.avg.ms.__cond_resched.rmap_walk_anon.try_to_migrate.migrate_folio_unmap.migrate_pages_batch 1.66 ± 21% -100.0% 0.00 -100.0% 0.00 +67.5% 2.78 ± 70% perf-sched.wait_time.avg.ms.__cond_resched.shadow_lru_isolate.__list_lru_walk_one.list_lru_walk_one_irq.do_shrink_slab 175.04 ± 74% -98.4% 2.81 ± 89% -96.6% 5.90 ±134% -78.7% 37.24 ±157% perf-sched.wait_time.avg.ms.__cond_resched.shmem_inode_acct_blocks.shmem_alloc_and_add_folio.shmem_get_folio_gfp.shmem_write_begin 2.60 ± 4% +183.8% 7.38 ± 9% +192.2% 7.60 ± 13% +29.9% 3.38 ± 13% perf-sched.wait_time.avg.ms.__cond_resched.shrink_folio_list.evict_folios.try_to_shrink_lruvec.shrink_one 0.39 ± 48% +579.8% 2.66 ± 59% +891.3% 3.89 ± 29% +16.7% 0.46 ± 49% perf-sched.wait_time.avg.ms.__cond_resched.stop_one_cpu.sched_exec.bprm_execve.part 2.24 ± 15% +242.5% 7.68 ± 40% +325.2% 9.53 ± 65% -30.0% 1.57 ± 25% perf-sched.wait_time.avg.ms.__cond_resched.try_to_shrink_lruvec.shrink_one.shrink_many.shrink_node 4.03 ± 21% +250.4% 14.12 ± 40% +338.2% 17.65 ± 25% -17.1% 3.34 ± 35% perf-sched.wait_time.avg.ms.do_wait.kernel_wait4.do_syscall_64.entry_SYSCALL_64_after_hwframe 0.86 ± 93% -100.0% 0.00 -93.7% 0.05 ±208% +625.9% 6.24 ±256% perf-sched.wait_time.avg.ms.io_schedule.migration_entry_wait_on_locked.migration_entry_wait.do_swap_page 400.08 +2.3% 409.48 ± 3% -29.2% 283.32 ± 19% -21.9% 312.27 ± 6% perf-sched.wait_time.avg.ms.irq_thread.kthread.ret_from_fork.ret_from_fork_asm 2.46 ± 18% +139.4% 5.90 ± 41% +148.7% 6.12 ± 32% -4.0% 2.36 ± 10% perf-sched.wait_time.avg.ms.irqentry_exit_to_user_mode.asm_exc_page_fault.[unknown] 0.90 ± 41% +58.7% 1.43 ±100% +926.1% 9.22 ±188% -36.1% 0.57 ±107% perf-sched.wait_time.avg.ms.irqentry_exit_to_user_mode.asm_exc_page_fault.[unknown].[unknown] 2.84 ± 16% +128.4% 6.49 ± 40% +86.8% 5.31 ± 15% +5.4% 3.00 ± 18% perf-sched.wait_time.avg.ms.irqentry_exit_to_user_mode.asm_sysvec_apic_timer_interrupt.[unknown] 49.54 ± 40% -59.5% 20.06 ± 31% -100.0% 0.00 -100.0% 0.00 perf-sched.wait_time.avg.ms.pipe_read.vfs_read.ksys_read.do_syscall_64 465.46 ± 7% -5.4% 440.26 ± 7% -100.0% 0.00 -100.0% 0.00 perf-sched.wait_time.avg.ms.schedule_hrtimeout_range.do_poll.constprop.0.do_sys_poll 386.71 ± 3% +3.1% 398.68 ± 5% -100.0% 0.00 -100.0% 0.00 perf-sched.wait_time.avg.ms.schedule_hrtimeout_range.ep_poll.do_epoll_wait.__x64_sys_epoll_wait 98.02 ± 46% -100.0% 0.00 -100.0% 0.00 -100.0% 0.00 ±264% perf-sched.wait_time.avg.ms.schedule_preempt_disabled.__mutex_lock.constprop.0.__drain_all_pages 4.70 ± 9% +270.7% 17.43 ± 40% +269.6% 17.37 ± 42% -7.8% 4.33 ± 20% perf-sched.wait_time.avg.ms.schedule_timeout.__wait_for_common.wait_for_completion_state.kernel_clone 304.70 ± 7% +51.9% 462.85 ± 2% +48.1% 451.33 ± 2% +44.1% 439.16 ± 2% perf-sched.wait_time.avg.ms.schedule_timeout.kcompactd.kthread.ret_from_fork 0.12 ±120% -100.0% 0.00 -91.4% 0.01 ±136% -41.3% 0.07 ±130% perf-sched.wait_time.avg.ms.schedule_timeout.kswapd_try_to_sleep.kswapd.kthread 5.17 ± 2% +29.1% 6.67 ± 14% +37.1% 7.09 ± 12% -15.0% 4.39 perf-sched.wait_time.avg.ms.schedule_timeout.rcu_gp_fqs_loop.rcu_gp_kthread.kthread 621.19 ± 7% +34.9% 837.98 ± 10% +42.4% 884.51 ± 10% -8.8% 566.49 ± 8% perf-sched.wait_time.avg.ms.smpboot_thread_fn.kthread.ret_from_fork.ret_from_fork_asm 167.70 ±222% +410.0% 855.18 ± 44% +355.8% 764.33 ± 56% -23.9% 127.67 ±259% perf-sched.wait_time.max.ms.__cond_resched.__alloc_frozen_pages_noprof.alloc_pages_mpol.alloc_pages_noprof.__pmd_alloc 4.40 ± 38% +7795.2% 347.10 ±135% +3073.1% 139.50 ±234% -37.6% 2.74 ± 53% perf-sched.wait_time.max.ms.__cond_resched.__alloc_frozen_pages_noprof.alloc_pages_mpol.alloc_pages_noprof.pte_alloc_one 1.34 ± 46% +118.7% 2.93 ±147% -100.0% 0.00 -100.0% 0.00 perf-sched.wait_time.max.ms.__cond_resched.__alloc_frozen_pages_noprof.alloc_pages_mpol.folio_alloc_mpol_noprof.vma_alloc_folio_noprof 208.26 ± 51% +146.1% 512.53 ± 45% +235.7% 699.12 ± 43% -60.2% 82.98 ± 43% perf-sched.wait_time.max.ms.__cond_resched.__alloc_frozen_pages_noprof.alloc_pages_mpol.folio_alloc_noprof.page_cache_ra_order 18.01 ± 57% -100.0% 0.00 -100.0% 0.00 -100.0% 0.00 perf-sched.wait_time.max.ms.__cond_resched.__alloc_pages_direct_compact.__alloc_pages_slowpath.constprop.0 0.54 ±223% +931.2% 5.52 ± 90% +980.9% 5.79 ± 78% -100.0% 0.00 perf-sched.wait_time.max.ms.__cond_resched.__get_user_pages.get_user_pages_remote.get_arg_page.copy_string_kernel 6.41 ± 42% -100.0% 0.00 -100.0% 0.00 -100.0% 0.00 perf-sched.wait_time.max.ms.__cond_resched.__reset_isolation_suitable.compact_zone.compact_zone_order.try_to_compact_pages 336.71 ±139% +98.4% 668.00 ± 70% +124.2% 755.05 ± 56% +12.2% 377.89 ±127% perf-sched.wait_time.max.ms.__cond_resched.__tlb_batch_free_encoded_pages.tlb_finish_mmu.exit_mmap.__mmput 0.00 ±223% +4.6e+07% 697.43 ± 69% -100.0% 0.00 -100.0% 0.00 perf-sched.wait_time.max.ms.__cond_resched.__vmalloc_area_node.__vmalloc_node_range_noprof.alloc_thread_stack_node.dup_task_struct 3.42 ± 46% +1591.8% 57.78 ±149% +260.1% 12.30 ± 35% -53.8% 1.58 ±100% perf-sched.wait_time.max.ms.__cond_resched.change_pmd_range.isra.0.change_pud_range 106.54 ±138% -100.0% 0.00 -100.0% 0.00 -100.0% 0.00 perf-sched.wait_time.max.ms.__cond_resched.down_read.split_huge_page_to_list_to_order.migrate_pages_batch.migrate_pages_sync 20.61 ± 74% +914.8% 209.17 ±121% +1062.6% 239.64 ± 80% +85.9% 38.32 ±108% perf-sched.wait_time.max.ms.__cond_resched.down_read.xfs_ilock_for_iomap.xfs_read_iomap_begin.iomap_iter 0.03 ±202% +24357.2% 7.34 ±134% +25094.6% 7.56 ± 55% +3114.6% 0.96 ±149% perf-sched.wait_time.max.ms.__cond_resched.down_write_killable.exec_mmap.begin_new_exec.load_elf_binary 1310 ±138% -99.7% 3.39 ±129% -99.9% 1.21 ±162% -32.3% 887.78 ± 98% perf-sched.wait_time.max.ms.__cond_resched.generic_perform_write.shmem_file_write_iter.vfs_write.ksys_write 215.44 ± 60% -100.0% 0.00 -100.0% 0.01 ±264% -89.7% 22.14 ±263% perf-sched.wait_time.max.ms.__cond_resched.isolate_freepages_block.fast_isolate_freepages.isolate_freepages.compaction_alloc 6.01 ± 47% -100.0% 0.00 -100.0% 0.00 -100.0% 0.00 perf-sched.wait_time.max.ms.__cond_resched.isolate_freepages_block.isolate_freepages.compaction_alloc.migrate_folio_unmap 25.32 ± 28% -100.0% 0.00 -100.0% 0.00 -100.0% 0.00 perf-sched.wait_time.max.ms.__cond_resched.isolate_migratepages.compact_zone.compact_zone_order.try_to_compact_pages 181.35 ±201% -100.0% 0.00 -100.0% 0.00 -99.8% 0.37 ±264% perf-sched.wait_time.max.ms.__cond_resched.isolate_migratepages_block.isolate_migratepages.compact_zone.compact_zone_order 335.43 ±141% +156.3% 859.60 ± 44% +165.6% 891.01 ± 37% -100.0% 0.00 perf-sched.wait_time.max.ms.__cond_resched.kmem_cache_alloc_noprof.alloc_pid.copy_process.kernel_clone 0.74 ±150% +731.1% 6.18 ± 68% +755.9% 6.37 ± 95% +16.2% 0.86 ±153% perf-sched.wait_time.max.ms.__cond_resched.kmem_cache_alloc_noprof.mas_alloc_nodes.mas_preallocate.vma_shrink 304.80 ± 58% -100.0% 0.00 -27.5% 220.96 ±209% -62.1% 115.59 ±150% perf-sched.wait_time.max.ms.__cond_resched.migrate_pages_batch.migrate_pages_sync.migrate_pages.compact_zone 167.80 ±223% +511.9% 1026 +444.1% 913.01 ± 37% +49.2% 250.43 ±173% perf-sched.wait_time.max.ms.__cond_resched.mutex_lock_killable.pcpu_alloc_noprof.mm_init.dup_mm 7.52 ± 19% -100.0% 0.00 -100.0% 0.00 -100.0% 0.00 perf-sched.wait_time.max.ms.__cond_resched.rmap_walk_anon.try_to_migrate.migrate_folio_unmap.migrate_pages_batch 26.92 ± 77% -100.0% 0.00 -100.0% 0.00 +399.1% 134.36 ±243% perf-sched.wait_time.max.ms.__cond_resched.shadow_lru_isolate.__list_lru_walk_one.list_lru_walk_one_irq.do_shrink_slab 2273 ± 72% -99.3% 15.34 ±105% -96.4% 81.35 ±220% -81.7% 416.33 ±143% perf-sched.wait_time.max.ms.__cond_resched.shmem_inode_acct_blocks.shmem_alloc_and_add_folio.shmem_get_folio_gfp.shmem_write_begin 200.50 ± 36% +513.0% 1228 ± 31% +450.3% 1103 ± 10% +726.1% 1656 ± 46% perf-sched.wait_time.max.ms.__cond_resched.shrink_folio_list.evict_folios.try_to_shrink_lruvec.shrink_one 3117 ± 18% -28.0% 2245 ± 57% -16.8% 2593 ± 41% -58.2% 1304 ± 67% perf-sched.wait_time.max.ms.__cond_resched.smpboot_thread_fn.kthread.ret_from_fork.ret_from_fork_asm 8.07 ± 34% +93.9% 15.65 ± 57% +175.8% 22.27 ± 26% -51.9% 3.89 ± 47% perf-sched.wait_time.max.ms.__cond_resched.stop_one_cpu.sched_exec.bprm_execve.part 25.24 ±105% +2617.2% 685.79 ± 68% +2508.9% 658.46 ± 68% -45.0% 13.87 ±142% perf-sched.wait_time.max.ms.__cond_resched.try_to_shrink_lruvec.shrink_one.shrink_many.shrink_node 334.70 ±141% +107.0% 692.98 ± 70% +165.6% 888.93 ± 37% -61.8% 127.70 ±258% perf-sched.wait_time.max.ms.__cond_resched.wp_page_copy.__handle_mm_fault.handle_mm_fault.do_user_addr_fault 42.20 ± 40% +160.7% 110.01 ± 32% +690.0% 333.38 ±120% +32.9% 56.09 ± 74% perf-sched.wait_time.max.ms.do_wait.kernel_wait4.do_syscall_64.entry_SYSCALL_64_after_hwframe 15.65 ±168% -100.0% 0.00 -98.3% 0.27 ±241% +706.1% 126.19 ±261% perf-sched.wait_time.max.ms.io_schedule.migration_entry_wait_on_locked.migration_entry_wait.do_swap_page 1.87 ± 38% +213.0% 5.85 ± 93% +7228.8% 136.97 ±238% -2.4% 1.82 ± 92% perf-sched.wait_time.max.ms.irqentry_exit_to_user_mode.asm_exc_page_fault.[unknown].[unknown] 1475 ± 5% +51.9% 2241 ± 19% -100.0% 0.00 -100.0% 0.00 perf-sched.wait_time.max.ms.pipe_read.vfs_read.ksys_read.do_syscall_64 1015 ± 2% -0.1% 1014 -100.0% 0.00 -100.0% 0.00 perf-sched.wait_time.max.ms.schedule_hrtimeout_range.do_poll.constprop.0.do_sys_poll 502.63 +1.2% 508.90 -100.0% 0.00 -100.0% 0.00 perf-sched.wait_time.max.ms.schedule_hrtimeout_range.ep_poll.do_epoll_wait.__x64_sys_epoll_wait 365.80 ± 22% -100.0% 0.00 -100.0% 0.00 -100.0% 0.00 ±264% perf-sched.wait_time.max.ms.schedule_preempt_disabled.__mutex_lock.constprop.0.__drain_all_pages 35.15 ± 40% +673.9% 272.04 ±121% +587.8% 241.76 ± 86% +60.7% 56.48 ± 57% perf-sched.wait_time.max.ms.schedule_timeout.__wait_for_common.wait_for_completion_state.kernel_clone 0.90 ±143% -100.0% 0.00 -98.2% 0.02 ±142% -73.2% 0.24 ±169% perf-sched.wait_time.max.ms.schedule_timeout.kswapd_try_to_sleep.kswapd.kthread 64.94 ±111% +165.2% 172.21 ± 90% +259.3% 233.31 ± 98% -80.6% 12.59 ± 57% perf-sched.wait_time.max.ms.schedule_timeout.rcu_gp_fqs_loop.rcu_gp_kthread.kthread 21.49 ± 11% -21.5 0.00 -21.5 0.00 -20.0 1.47 ±158% perf-profile.calltrace.cycles-pp.__alloc_pages_direct_compact.__alloc_pages_slowpath.__alloc_frozen_pages_noprof.alloc_pages_mpol.folio_alloc_noprof 20.75 ± 11% -20.8 0.00 -20.8 0.00 -19.3 1.46 ±158% perf-profile.calltrace.cycles-pp.try_to_compact_pages.__alloc_pages_direct_compact.__alloc_pages_slowpath.__alloc_frozen_pages_noprof.alloc_pages_mpol 20.75 ± 11% -20.7 0.00 -20.7 0.00 -19.3 1.46 ±158% perf-profile.calltrace.cycles-pp.compact_zone.compact_zone_order.try_to_compact_pages.__alloc_pages_direct_compact.__alloc_pages_slowpath 20.75 ± 11% -20.7 0.00 -20.7 0.00 -19.3 1.46 ±158% perf-profile.calltrace.cycles-pp.compact_zone_order.try_to_compact_pages.__alloc_pages_direct_compact.__alloc_pages_slowpath.__alloc_frozen_pages_noprof 19.28 ± 11% -19.3 0.00 -19.3 0.00 -18.1 1.22 ±162% perf-profile.calltrace.cycles-pp.isolate_migratepages.compact_zone.compact_zone_order.try_to_compact_pages.__alloc_pages_direct_compact 14.19 ± 7% -14.2 0.00 -14.2 0.00 -9.7 4.53 ± 76% perf-profile.calltrace.cycles-pp.get_page_from_freelist.__alloc_frozen_pages_noprof.alloc_pages_mpol.folio_alloc_noprof.page_cache_ra_order 13.84 ± 7% -13.8 0.00 -13.8 0.00 -9.8 4.08 ± 78% perf-profile.calltrace.cycles-pp.rmqueue.get_page_from_freelist.__alloc_frozen_pages_noprof.alloc_pages_mpol.folio_alloc_noprof 12.76 ± 16% -12.8 0.00 -12.8 0.00 -12.1 0.65 ±164% perf-profile.calltrace.cycles-pp.isolate_migratepages_block.isolate_migratepages.compact_zone.compact_zone_order.try_to_compact_pages 11.50 ± 8% -11.5 0.00 -11.5 0.00 -8.4 3.10 ± 77% perf-profile.calltrace.cycles-pp.__rmqueue_pcplist.rmqueue.get_page_from_freelist.__alloc_frozen_pages_noprof.alloc_pages_mpol 11.47 ± 8% -11.5 0.00 -11.5 0.00 -8.4 3.06 ± 77% perf-profile.calltrace.cycles-pp.rmqueue_bulk.__rmqueue_pcplist.rmqueue.get_page_from_freelist.__alloc_frozen_pages_noprof 6.99 ± 20% -5.6 1.35 ± 11% -7.0 0.00 -6.9 0.10 ±264% perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.folio_lruvec_lock_irqsave.folio_batch_move_lru.__folio_batch_add_and_move 5.65 ± 78% -5.5 0.17 ±223% -5.5 0.15 ±264% +48.3 53.97 ± 22% perf-profile.calltrace.cycles-pp.asm_sysvec_apic_timer_interrupt.acpi_safe_halt.acpi_idle_do_entry.acpi_idle_enter.cpuidle_enter_state 6.86 ± 17% -5.4 1.48 ± 10% -6.9 0.00 -6.9 0.00 perf-profile.calltrace.cycles-pp.__folio_batch_add_and_move.filemap_add_folio.page_cache_ra_order.filemap_fault.__do_fault 6.84 ± 18% -5.4 1.47 ± 10% -6.8 0.00 -6.8 0.00 perf-profile.calltrace.cycles-pp.folio_batch_move_lru.__folio_batch_add_and_move.filemap_add_folio.page_cache_ra_order.filemap_fault 6.49 ± 18% -5.1 1.37 ± 11% -6.5 0.00 -6.5 0.00 perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.folio_lruvec_lock_irqsave.folio_batch_move_lru.__folio_batch_add_and_move.filemap_add_folio 6.50 ± 18% -5.1 1.37 ± 10% -6.5 0.00 -6.5 0.00 perf-profile.calltrace.cycles-pp.folio_lruvec_lock_irqsave.folio_batch_move_lru.__folio_batch_add_and_move.filemap_add_folio.page_cache_ra_order 7.18 ± 16% -5.1 2.12 ± 6% -5.3 1.92 ± 7% +0.3 7.47 ± 95% perf-profile.calltrace.cycles-pp.filemap_add_folio.page_cache_ra_order.filemap_fault.__do_fault.do_read_fault 90.95 ± 2% -3.7 87.21 -3.6 87.32 -62.6 28.37 ± 68% perf-profile.calltrace.cycles-pp.page_cache_ra_order.filemap_fault.__do_fault.do_read_fault.do_pte_missing 3.00 ± 14% -3.0 0.00 -3.0 0.00 -2.1 0.94 ± 86% perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.rmqueue.get_page_from_freelist.__alloc_frozen_pages_noprof.alloc_pages_mpol 3.00 ± 14% -3.0 0.00 -3.0 0.00 -2.1 0.93 ± 86% perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.rmqueue.get_page_from_freelist.__alloc_frozen_pages_noprof 2.95 ± 76% -2.9 0.09 ±223% -2.9 0.08 ±264% +30.8 33.78 ± 29% perf-profile.calltrace.cycles-pp.acpi_idle_do_entry.acpi_idle_enter.cpuidle_enter_state.cpuidle_enter.cpuidle_idle_call 2.95 ± 76% -2.9 0.09 ±223% -2.9 0.08 ±264% +30.9 33.80 ± 29% perf-profile.calltrace.cycles-pp.acpi_idle_enter.cpuidle_enter_state.cpuidle_enter.cpuidle_idle_call.do_idle 2.95 ± 76% -2.9 0.09 ±223% -2.9 0.08 ±264% +31.2 34.10 ± 30% perf-profile.calltrace.cycles-pp.cpuidle_enter_state.cpuidle_enter.cpuidle_idle_call.do_idle.cpu_startup_entry 2.95 ± 76% -2.9 0.09 ±223% -2.9 0.08 ±264% +31.2 34.16 ± 30% perf-profile.calltrace.cycles-pp.cpuidle_enter.cpuidle_idle_call.do_idle.cpu_startup_entry.start_secondary 2.95 ± 76% -2.9 0.09 ±223% -2.9 0.08 ±264% +32.1 35.04 ± 31% perf-profile.calltrace.cycles-pp.cpuidle_idle_call.do_idle.cpu_startup_entry.start_secondary.common_startup_64 2.95 ± 76% -2.9 0.09 ±223% -2.9 0.08 ±264% +32.2 35.16 ± 31% perf-profile.calltrace.cycles-pp.do_idle.cpu_startup_entry.start_secondary.common_startup_64 2.95 ± 76% -2.9 0.09 ±223% -2.9 0.08 ±264% +32.2 35.19 ± 31% perf-profile.calltrace.cycles-pp.cpu_startup_entry.start_secondary.common_startup_64 2.95 ± 76% -2.9 0.09 ±223% -2.9 0.08 ±264% +32.2 35.19 ± 31% perf-profile.calltrace.cycles-pp.start_secondary.common_startup_64 2.95 ± 76% -2.8 0.10 ±223% -2.9 0.08 ±264% +32.5 35.44 ± 31% perf-profile.calltrace.cycles-pp.common_startup_64 2.02 ± 21% -1.0 1.06 ± 8% -1.1 0.92 ± 10% -1.0 1.03 ±128% perf-profile.calltrace.cycles-pp._raw_spin_lock_irq.evict_folios.try_to_shrink_lruvec.shrink_one.shrink_many 2.01 ± 21% -1.0 1.05 ± 8% -1.1 0.91 ± 10% -1.0 1.02 ±128% perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irq.evict_folios.try_to_shrink_lruvec.shrink_one 1.33 ± 7% -0.9 0.45 ± 44% -1.3 0.07 ±264% +3.5 4.88 ± 64% perf-profile.calltrace.cycles-pp.read_pages.page_cache_ra_order.filemap_fault.__do_fault.do_read_fault 1.32 ± 7% -0.9 0.45 ± 44% -1.3 0.07 ±264% +3.5 4.87 ± 64% perf-profile.calltrace.cycles-pp.iomap_readahead.read_pages.page_cache_ra_order.filemap_fault.__do_fault 1.91 ± 23% -0.8 1.08 ± 8% -1.0 0.93 ± 7% -1.3 0.59 ±142% perf-profile.calltrace.cycles-pp.folio_referenced.shrink_folio_list.evict_folios.try_to_shrink_lruvec.shrink_one 1.90 ± 23% -0.8 1.08 ± 8% -1.0 0.92 ± 7% -1.3 0.58 ±143% perf-profile.calltrace.cycles-pp.__rmap_walk_file.folio_referenced.shrink_folio_list.evict_folios.try_to_shrink_lruvec 1.82 ± 24% -0.8 1.04 ± 8% -0.9 0.90 ± 7% -1.3 0.56 ±143% perf-profile.calltrace.cycles-pp.folio_referenced_one.__rmap_walk_file.folio_referenced.shrink_folio_list.evict_folios 1.24 ± 6% -0.7 0.58 ± 5% -0.7 0.54 ± 3% +2.3 3.59 ± 60% perf-profile.calltrace.cycles-pp.do_rw_once 2.27 ± 8% -0.4 1.84 ± 3% -0.4 1.89 ± 7% +13.9 16.12 ± 35% perf-profile.calltrace.cycles-pp.kthread.ret_from_fork.ret_from_fork_asm 2.27 ± 8% -0.4 1.84 ± 3% -0.4 1.89 ± 7% +13.9 16.12 ± 35% perf-profile.calltrace.cycles-pp.ret_from_fork.ret_from_fork_asm 2.27 ± 8% -0.4 1.84 ± 3% -0.4 1.89 ± 7% +13.9 16.12 ± 35% perf-profile.calltrace.cycles-pp.ret_from_fork_asm 0.00 +0.0 0.00 +0.0 0.00 +3.8 3.75 ± 80% perf-profile.calltrace.cycles-pp.truncate_folio_batch_exceptionals.truncate_inode_pages_range.xfs_fs_evict_inode.evict.do_unlinkat 0.00 +0.0 0.00 +0.0 0.00 +3.8 3.83 ± 78% perf-profile.calltrace.cycles-pp.update_process_times.tick_nohz_handler.__hrtimer_run_queues.hrtimer_interrupt.__sysvec_apic_timer_interrupt 0.00 +0.0 0.00 +0.0 0.00 +4.6 4.55 ± 77% perf-profile.calltrace.cycles-pp.tick_nohz_handler.__hrtimer_run_queues.hrtimer_interrupt.__sysvec_apic_timer_interrupt.sysvec_apic_timer_interrupt 0.00 +0.0 0.00 +0.0 0.00 +6.2 6.17 ± 83% perf-profile.calltrace.cycles-pp.migrate_pages_batch.migrate_pages_sync.migrate_pages.compact_zone.compact_node 0.00 +0.0 0.00 +0.0 0.00 +6.2 6.18 ± 83% perf-profile.calltrace.cycles-pp.migrate_pages_sync.migrate_pages.compact_zone.compact_node.kcompactd 0.00 +0.0 0.00 +0.0 0.00 +6.6 6.63 ± 74% perf-profile.calltrace.cycles-pp.__hrtimer_run_queues.hrtimer_interrupt.__sysvec_apic_timer_interrupt.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt 0.00 +0.0 0.00 +0.0 0.00 +7.4 7.39 ± 74% perf-profile.calltrace.cycles-pp.hrtimer_interrupt.__sysvec_apic_timer_interrupt.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt.acpi_safe_halt 0.00 +0.0 0.00 +0.0 0.00 +7.4 7.41 ± 74% perf-profile.calltrace.cycles-pp.__sysvec_apic_timer_interrupt.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt.acpi_safe_halt.acpi_idle_do_entry 0.00 +0.0 0.00 +0.0 0.00 +8.0 8.02 ± 82% perf-profile.calltrace.cycles-pp.migrate_pages.compact_zone.compact_node.kcompactd.kthread 0.00 +0.0 0.00 +0.0 0.00 +8.5 8.54 ± 82% perf-profile.calltrace.cycles-pp.compact_node.kcompactd.kthread.ret_from_fork.ret_from_fork_asm 0.00 +0.0 0.00 +0.0 0.00 +8.5 8.54 ± 82% perf-profile.calltrace.cycles-pp.compact_zone.compact_node.kcompactd.kthread.ret_from_fork 0.00 +0.0 0.00 +0.0 0.00 +9.5 9.47 ± 85% perf-profile.calltrace.cycles-pp.truncate_inode_pages_range.xfs_fs_evict_inode.evict.do_unlinkat.__x64_sys_unlinkat 0.00 +0.0 0.00 +0.0 0.00 +9.5 9.48 ± 85% perf-profile.calltrace.cycles-pp.evict.do_unlinkat.__x64_sys_unlinkat.do_syscall_64.entry_SYSCALL_64_after_hwframe 0.00 +0.0 0.00 +0.0 0.00 +9.5 9.48 ± 85% perf-profile.calltrace.cycles-pp.xfs_fs_evict_inode.evict.do_unlinkat.__x64_sys_unlinkat.do_syscall_64 0.00 +0.0 0.00 +0.0 0.00 +9.5 9.48 ± 85% perf-profile.calltrace.cycles-pp.__x64_sys_unlinkat.do_syscall_64.entry_SYSCALL_64_after_hwframe.unlinkat 0.00 +0.0 0.00 +0.0 0.00 +9.5 9.48 ± 85% perf-profile.calltrace.cycles-pp.do_unlinkat.__x64_sys_unlinkat.do_syscall_64.entry_SYSCALL_64_after_hwframe.unlinkat 0.00 +0.0 0.00 +0.0 0.00 +9.5 9.48 ± 85% perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.unlinkat 0.00 +0.0 0.00 +0.0 0.00 +9.5 9.48 ± 85% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.unlinkat 0.00 +0.0 0.00 +0.0 0.00 +9.5 9.48 ± 85% perf-profile.calltrace.cycles-pp.unlinkat 0.00 +0.0 0.00 +0.0 0.00 +9.7 9.68 ± 72% perf-profile.calltrace.cycles-pp.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt.acpi_safe_halt.acpi_idle_do_entry.acpi_idle_enter 0.00 +0.0 0.00 +0.0 0.00 +10.2 10.19 ± 63% perf-profile.calltrace.cycles-pp.kcompactd.kthread.ret_from_fork.ret_from_fork_asm 0.00 +0.0 0.00 +0.0 0.00 +11.2 11.16 ± 73% perf-profile.calltrace.cycles-pp.acpi_safe_halt.acpi_idle_do_entry.acpi_idle_enter.cpuidle_enter_state.cpuidle_enter 0.00 +0.0 0.00 +0.7 0.66 ± 4% +0.0 0.00 perf-profile.calltrace.cycles-pp.prep_move_freepages_block.try_to_claim_block.rmqueue_bulk.__rmqueue_pcplist.rmqueue 0.00 +0.0 0.00 +0.7 0.74 ± 4% +0.0 0.00 perf-profile.calltrace.cycles-pp.try_to_claim_block.rmqueue_bulk.__rmqueue_pcplist.rmqueue.get_page_from_freelist 0.00 +0.0 0.00 +1.2 1.15 ± 12% +5.9 5.90 ±104% perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.folio_lruvec_lock_irqsave.folio_batch_move_lru.folio_add_lru 0.00 +0.0 0.00 +1.2 1.17 ± 12% +6.1 6.06 ±103% perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.folio_lruvec_lock_irqsave.folio_batch_move_lru.folio_add_lru.filemap_add_folio 0.00 +0.0 0.00 +1.2 1.17 ± 12% +6.1 6.08 ±103% perf-profile.calltrace.cycles-pp.folio_lruvec_lock_irqsave.folio_batch_move_lru.folio_add_lru.filemap_add_folio.page_cache_ra_order 0.00 +0.0 0.00 +1.3 1.26 ± 12% +6.9 6.88 ± 98% perf-profile.calltrace.cycles-pp.folio_batch_move_lru.folio_add_lru.filemap_add_folio.page_cache_ra_order.filemap_fault 0.00 +0.0 0.00 +1.3 1.27 ± 12% +6.9 6.93 ± 98% perf-profile.calltrace.cycles-pp.folio_add_lru.filemap_add_folio.page_cache_ra_order.filemap_fault.__do_fault 0.00 +0.4 0.37 ± 71% +0.6 0.55 ± 4% +0.0 0.00 perf-profile.calltrace.cycles-pp.kmem_cache_alloc_lru_noprof.xas_alloc.xas_create.xas_store.__filemap_add_folio 0.00 +0.4 0.37 ± 71% +0.6 0.55 ± 4% +0.0 0.00 perf-profile.calltrace.cycles-pp.xas_alloc.xas_create.xas_store.__filemap_add_folio.filemap_add_folio 0.00 +0.4 0.37 ± 71% +0.6 0.55 ± 4% +0.0 0.00 perf-profile.calltrace.cycles-pp.xas_create.xas_store.__filemap_add_folio.filemap_add_folio.page_cache_ra_order 0.00 +0.5 0.46 ± 45% +0.6 0.56 ± 4% +0.0 0.00 perf-profile.calltrace.cycles-pp.xas_store.__filemap_add_folio.filemap_add_folio.page_cache_ra_order.filemap_fault 0.00 +0.6 0.63 ± 7% +0.6 0.64 ± 4% +0.3 0.32 ±101% perf-profile.calltrace.cycles-pp.__filemap_add_folio.filemap_add_folio.page_cache_ra_order.filemap_fault.__do_fault 0.00 +0.6 0.64 ± 6% +0.4 0.42 ± 57% +0.0 0.00 perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.unreserve_highatomic_pageblock.__alloc_pages_slowpath.__alloc_frozen_pages_noprof.alloc_pages_mpol 0.00 +0.6 0.64 ± 6% +0.4 0.42 ± 57% +0.0 0.00 perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.unreserve_highatomic_pageblock.__alloc_pages_slowpath.__alloc_frozen_pages_noprof 0.00 +0.6 0.64 ± 5% +0.4 0.42 ± 57% +0.0 0.00 perf-profile.calltrace.cycles-pp.unreserve_highatomic_pageblock.__alloc_pages_slowpath.__alloc_frozen_pages_noprof.alloc_pages_mpol.folio_alloc_noprof 0.00 +0.7 0.66 ± 2% +0.0 0.00 +0.0 0.00 perf-profile.calltrace.cycles-pp.prep_move_freepages_block.try_to_steal_block.rmqueue_bulk.__rmqueue_pcplist.rmqueue 0.00 +0.7 0.70 ± 3% +0.7 0.71 ± 2% +0.0 0.00 perf-profile.calltrace.cycles-pp.try_to_free_pages.__alloc_pages_slowpath.__alloc_frozen_pages_noprof.alloc_pages_mpol.alloc_pages_noprof 0.00 +0.8 0.75 ± 2% +0.0 0.00 +0.0 0.00 perf-profile.calltrace.cycles-pp.try_to_steal_block.rmqueue_bulk.__rmqueue_pcplist.rmqueue.get_page_from_freelist 0.00 +1.3 1.32 ± 7% +1.4 1.41 ± 11% +0.0 0.00 perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.free_one_page.free_unref_folios.shrink_folio_list 0.00 +1.3 1.32 ± 7% +1.4 1.41 ± 11% +0.0 0.00 perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.free_one_page.free_unref_folios.shrink_folio_list.evict_folios 0.00 +1.3 1.32 ± 7% +1.4 1.41 ± 11% +0.0 0.00 perf-profile.calltrace.cycles-pp.free_one_page.free_unref_folios.shrink_folio_list.evict_folios.try_to_shrink_lruvec 91.27 ± 2% +1.8 93.05 +2.0 93.23 -62.9 28.41 ± 68% perf-profile.calltrace.cycles-pp.__do_fault.do_read_fault.do_pte_missing.__handle_mm_fault.handle_mm_fault 91.27 ± 2% +1.8 93.05 +2.0 93.23 -62.9 28.41 ± 68% perf-profile.calltrace.cycles-pp.filemap_fault.__do_fault.do_read_fault.do_pte_missing.__handle_mm_fault 82.42 ± 2% +2.1 84.56 +2.5 84.90 -66.4 15.99 ± 70% perf-profile.calltrace.cycles-pp.folio_alloc_noprof.page_cache_ra_order.filemap_fault.__do_fault.do_read_fault 82.39 ± 2% +2.2 84.54 +2.5 84.89 -66.5 15.85 ± 71% perf-profile.calltrace.cycles-pp.alloc_pages_mpol.folio_alloc_noprof.page_cache_ra_order.filemap_fault.__do_fault 82.38 ± 2% +2.2 84.54 +2.5 84.88 -66.6 15.84 ± 71% perf-profile.calltrace.cycles-pp.__alloc_frozen_pages_noprof.alloc_pages_mpol.folio_alloc_noprof.page_cache_ra_order.filemap_fault 0.88 ± 11% +2.2 3.08 +2.2 3.06 -0.7 0.14 ±173% perf-profile.calltrace.cycles-pp.__alloc_frozen_pages_noprof.alloc_pages_mpol.alloc_pages_noprof.pte_alloc_one.do_read_fault 0.88 ± 11% +2.2 3.08 +2.2 3.06 -0.7 0.14 ±173% perf-profile.calltrace.cycles-pp.alloc_pages_mpol.alloc_pages_noprof.pte_alloc_one.do_read_fault.do_pte_missing 0.88 ± 11% +2.2 3.08 +2.2 3.06 -0.7 0.14 ±173% perf-profile.calltrace.cycles-pp.alloc_pages_noprof.pte_alloc_one.do_read_fault.do_pte_missing.__handle_mm_fault 0.88 ± 11% +2.2 3.08 +2.2 3.06 -0.7 0.14 ±173% perf-profile.calltrace.cycles-pp.pte_alloc_one.do_read_fault.do_pte_missing.__handle_mm_fault.handle_mm_fault 0.00 +2.4 2.37 ± 2% +2.3 2.32 +0.0 0.00 perf-profile.calltrace.cycles-pp.get_page_from_freelist.__alloc_pages_slowpath.__alloc_frozen_pages_noprof.alloc_pages_mpol.alloc_pages_noprof 0.00 +2.8 2.78 ± 10% +2.8 2.82 ± 9% +0.0 0.00 perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.rmqueue.get_page_from_freelist.__alloc_pages_slowpath 0.00 +2.8 2.78 ± 10% +2.8 2.82 ± 9% +0.0 0.00 perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.rmqueue.get_page_from_freelist.__alloc_pages_slowpath.__alloc_frozen_pages_noprof 0.00 +3.1 3.08 +3.1 3.05 +0.0 0.00 perf-profile.calltrace.cycles-pp.__alloc_pages_slowpath.__alloc_frozen_pages_noprof.alloc_pages_mpol.alloc_pages_noprof.pte_alloc_one 93.16 ± 2% +3.5 96.62 +3.6 96.74 -60.9 32.28 ± 66% perf-profile.calltrace.cycles-pp.do_access 92.48 ± 2% +3.8 96.30 +4.0 96.46 -63.0 29.51 ± 67% perf-profile.calltrace.cycles-pp.asm_exc_page_fault.do_access 92.40 ± 2% +3.8 96.24 +4.0 96.40 -63.1 29.35 ± 67% perf-profile.calltrace.cycles-pp.do_user_addr_fault.exc_page_fault.asm_exc_page_fault.do_access 92.40 ± 2% +3.8 96.25 +4.0 96.40 -63.1 29.35 ± 67% perf-profile.calltrace.cycles-pp.exc_page_fault.asm_exc_page_fault.do_access 92.36 ± 2% +3.9 96.23 +4.0 96.39 -63.1 29.25 ± 67% perf-profile.calltrace.cycles-pp.handle_mm_fault.do_user_addr_fault.exc_page_fault.asm_exc_page_fault.do_access 92.33 ± 2% +3.9 96.21 +4.0 96.37 -63.2 29.17 ± 67% perf-profile.calltrace.cycles-pp.do_pte_missing.__handle_mm_fault.handle_mm_fault.do_user_addr_fault.exc_page_fault 92.33 ± 2% +3.9 96.21 +4.0 96.37 -63.2 29.16 ± 67% perf-profile.calltrace.cycles-pp.do_read_fault.do_pte_missing.__handle_mm_fault.handle_mm_fault.do_user_addr_fault 92.35 ± 2% +4.0 96.31 +4.0 96.38 -63.1 29.21 ± 67% perf-profile.calltrace.cycles-pp.__handle_mm_fault.handle_mm_fault.do_user_addr_fault.exc_page_fault.asm_exc_page_fault 17.83 ± 9% +5.3 23.11 +5.5 23.31 -15.3 2.51 ± 67% perf-profile.calltrace.cycles-pp.smp_call_function_many_cond.on_each_cpu_cond_mask.arch_tlbbatch_flush.try_to_unmap_flush.shrink_folio_list 17.84 ± 9% +5.3 23.12 +5.5 23.32 -15.3 2.51 ± 67% perf-profile.calltrace.cycles-pp.on_each_cpu_cond_mask.arch_tlbbatch_flush.try_to_unmap_flush.shrink_folio_list.evict_folios 17.84 ± 9% +5.3 23.12 +5.5 23.32 -15.3 2.52 ± 67% perf-profile.calltrace.cycles-pp.arch_tlbbatch_flush.try_to_unmap_flush.shrink_folio_list.evict_folios.try_to_shrink_lruvec 17.84 ± 9% +5.3 23.12 +5.5 23.32 -15.3 2.52 ± 67% perf-profile.calltrace.cycles-pp.try_to_unmap_flush.shrink_folio_list.evict_folios.try_to_shrink_lruvec.shrink_one 0.00 +5.7 5.72 ± 10% +5.8 5.80 ± 9% +0.0 0.00 perf-profile.calltrace.cycles-pp.__alloc_pages_slowpath.__alloc_frozen_pages_noprof.alloc_pages_mpol.folio_alloc_noprof.page_cache_ra_unbounded 0.00 +5.7 5.73 ± 10% +5.8 5.80 ± 9% +0.0 0.00 perf-profile.calltrace.cycles-pp.__alloc_frozen_pages_noprof.alloc_pages_mpol.folio_alloc_noprof.page_cache_ra_unbounded.filemap_fault 0.00 +5.7 5.73 ± 10% +5.8 5.80 ± 9% +0.0 0.00 perf-profile.calltrace.cycles-pp.alloc_pages_mpol.folio_alloc_noprof.page_cache_ra_unbounded.filemap_fault.__do_fault 0.00 +5.7 5.73 ± 10% +5.8 5.80 ± 9% +0.0 0.00 perf-profile.calltrace.cycles-pp.folio_alloc_noprof.page_cache_ra_unbounded.filemap_fault.__do_fault.do_read_fault 0.00 +5.7 5.75 ± 10% +5.8 5.83 ± 9% +0.0 0.00 perf-profile.calltrace.cycles-pp.page_cache_ra_unbounded.filemap_fault.__do_fault.do_read_fault.do_pte_missing 22.03 ± 7% +7.3 29.35 ± 2% +8.1 30.12 ± 2% -19.2 2.79 ± 82% perf-profile.calltrace.cycles-pp.free_frozen_page_commit.free_unref_folios.shrink_folio_list.evict_folios.try_to_shrink_lruvec 21.90 ± 7% +7.4 29.28 ± 2% +8.1 30.05 ± 2% -19.2 2.70 ± 83% perf-profile.calltrace.cycles-pp.free_pcppages_bulk.free_frozen_page_commit.free_unref_folios.shrink_folio_list.evict_folios 21.68 ± 7% +7.5 29.17 ± 2% +8.3 29.94 ± 2% -19.3 2.36 ± 87% perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.free_pcppages_bulk.free_frozen_page_commit.free_unref_folios.shrink_folio_list 21.65 ± 7% +7.5 29.16 ± 2% +8.3 29.93 ± 2% -19.4 2.23 ± 93% perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.free_pcppages_bulk.free_frozen_page_commit.free_unref_folios 22.13 ± 7% +8.7 30.85 ± 2% +9.6 31.71 ± 2% -19.0 3.08 ± 80% perf-profile.calltrace.cycles-pp.free_unref_folios.shrink_folio_list.evict_folios.try_to_shrink_lruvec.shrink_one 45.19 ± 8% +11.0 56.15 +11.7 56.85 -36.3 8.90 ± 70% perf-profile.calltrace.cycles-pp.try_to_free_pages.__alloc_pages_slowpath.__alloc_frozen_pages_noprof.alloc_pages_mpol.folio_alloc_noprof 45.07 ± 8% +11.7 56.81 +12.5 57.52 -36.3 8.82 ± 70% perf-profile.calltrace.cycles-pp.do_try_to_free_pages.try_to_free_pages.__alloc_pages_slowpath.__alloc_frozen_pages_noprof.alloc_pages_mpol 45.06 ± 8% +11.7 56.80 +12.5 57.52 -36.2 8.81 ± 70% perf-profile.calltrace.cycles-pp.shrink_node.do_try_to_free_pages.try_to_free_pages.__alloc_pages_slowpath.__alloc_frozen_pages_noprof 45.04 ± 8% +11.7 56.79 +12.5 57.51 -36.2 8.80 ± 70% perf-profile.calltrace.cycles-pp.shrink_many.shrink_node.do_try_to_free_pages.try_to_free_pages.__alloc_pages_slowpath 45.03 ± 8% +11.8 56.79 +12.5 57.51 -36.2 8.79 ± 70% perf-profile.calltrace.cycles-pp.shrink_one.shrink_many.shrink_node.do_try_to_free_pages.try_to_free_pages 46.17 ± 7% +12.2 58.42 +13.0 59.20 -35.0 11.16 ± 65% perf-profile.calltrace.cycles-pp.evict_folios.try_to_shrink_lruvec.shrink_one.shrink_many.shrink_node 44.27 ± 8% +12.4 56.63 +13.1 57.40 -37.1 7.18 ± 77% perf-profile.calltrace.cycles-pp.try_to_shrink_lruvec.shrink_one.shrink_many.shrink_node.do_try_to_free_pages 43.67 ± 7% +13.4 57.06 +14.3 58.02 -34.5 9.12 ± 64% perf-profile.calltrace.cycles-pp.shrink_folio_list.evict_folios.try_to_shrink_lruvec.shrink_one.shrink_many 68.15 ± 2% +16.1 84.30 +16.5 84.64 -56.9 11.27 ± 72% perf-profile.calltrace.cycles-pp.__alloc_pages_slowpath.__alloc_frozen_pages_noprof.alloc_pages_mpol.folio_alloc_noprof.page_cache_ra_order 10.91 ± 10% +20.1 31.00 +19.8 30.74 -8.3 2.66 ± 80% perf-profile.calltrace.cycles-pp._raw_spin_lock_irqsave.rmqueue_bulk.__rmqueue_pcplist.rmqueue.get_page_from_freelist 10.90 ± 10% +20.1 30.99 +19.8 30.74 -8.3 2.64 ± 80% perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.rmqueue_bulk.__rmqueue_pcplist.rmqueue 0.42 ± 72% +31.5 31.88 +31.2 31.61 -0.3 0.09 ±264% perf-profile.calltrace.cycles-pp.rmqueue_bulk.__rmqueue_pcplist.rmqueue.get_page_from_freelist.__alloc_pages_slowpath 0.43 ± 72% +31.5 31.90 +31.2 31.63 -0.3 0.09 ±264% perf-profile.calltrace.cycles-pp.__rmqueue_pcplist.rmqueue.get_page_from_freelist.__alloc_pages_slowpath.__alloc_frozen_pages_noprof 0.91 ± 19% +31.6 32.48 +31.3 32.20 -0.7 0.21 ±188% perf-profile.calltrace.cycles-pp.get_page_from_freelist.__alloc_pages_slowpath.__alloc_frozen_pages_noprof.alloc_pages_mpol.folio_alloc_noprof 0.84 ± 20% +33.9 34.72 +33.6 34.47 -0.7 0.12 ±264% perf-profile.calltrace.cycles-pp.rmqueue.get_page_from_freelist.__alloc_pages_slowpath.__alloc_frozen_pages_noprof.alloc_pages_mpol 21.49 ± 11% -21.5 0.00 -21.5 0.00 -19.9 1.56 ±147% perf-profile.children.cycles-pp.__alloc_pages_direct_compact 20.91 ± 11% -20.9 0.03 ±223% -20.9 0.06 ±211% -9.2 11.71 ± 55% perf-profile.children.cycles-pp.compact_zone 20.76 ± 11% -20.8 0.00 -20.8 0.00 -19.2 1.54 ±147% perf-profile.children.cycles-pp.try_to_compact_pages 20.75 ± 11% -20.8 0.00 -20.8 0.00 -19.2 1.54 ±147% perf-profile.children.cycles-pp.compact_zone_order 19.29 ± 11% -19.3 0.00 -19.3 0.00 -17.4 1.89 ± 99% perf-profile.children.cycles-pp.isolate_migratepages 12.77 ± 16% -12.8 0.00 -12.8 0.00 -11.5 1.28 ± 78% perf-profile.children.cycles-pp.isolate_migratepages_block 7.53 ± 17% -5.7 1.87 ± 10% -7.2 0.32 ± 8% -7.3 0.22 ±128% perf-profile.children.cycles-pp.__folio_batch_add_and_move 7.54 ± 17% -5.7 1.88 ± 10% -5.9 1.61 ± 11% +0.0 7.56 ± 90% perf-profile.children.cycles-pp.folio_batch_move_lru 7.14 ± 18% -5.4 1.76 ± 10% -5.6 1.50 ± 11% -0.7 6.48 ± 99% perf-profile.children.cycles-pp.folio_lruvec_lock_irqsave 7.20 ± 16% -5.1 2.14 ± 6% -5.3 1.94 ± 7% +0.3 7.47 ± 95% perf-profile.children.cycles-pp.filemap_add_folio 3.83 ± 22% -3.8 0.00 -3.8 0.00 -3.4 0.40 ±151% perf-profile.children.cycles-pp.fast_find_migrateblock 90.95 ± 2% -3.6 87.33 -3.6 87.37 -62.6 28.40 ± 67% perf-profile.children.cycles-pp.page_cache_ra_order 2.96 ± 75% -2.7 0.28 ± 62% -2.7 0.24 ± 73% +32.2 35.19 ± 31% perf-profile.children.cycles-pp.start_secondary 3.10 ± 71% -2.7 0.42 ± 41% -2.7 0.38 ± 46% +29.7 32.76 ± 29% perf-profile.children.cycles-pp.asm_sysvec_apic_timer_interrupt 2.96 ± 75% -2.7 0.29 ± 65% -2.7 0.25 ± 75% +31.0 34.01 ± 29% perf-profile.children.cycles-pp.acpi_safe_halt 2.96 ± 75% -2.7 0.29 ± 65% -2.7 0.25 ± 75% +31.1 34.02 ± 29% perf-profile.children.cycles-pp.acpi_idle_do_entry 2.96 ± 75% -2.7 0.29 ± 65% -2.7 0.25 ± 75% +31.1 34.04 ± 29% perf-profile.children.cycles-pp.acpi_idle_enter 2.96 ± 75% -2.7 0.29 ± 65% -2.7 0.25 ± 75% +31.4 34.35 ± 29% perf-profile.children.cycles-pp.cpuidle_enter_state 2.96 ± 75% -2.7 0.29 ± 65% -2.7 0.25 ± 75% +31.4 34.41 ± 29% perf-profile.children.cycles-pp.cpuidle_enter 2.96 ± 75% -2.7 0.29 ± 65% -2.7 0.25 ± 75% +32.3 35.30 ± 30% perf-profile.children.cycles-pp.cpuidle_idle_call 2.96 ± 75% -2.7 0.29 ± 65% -2.7 0.25 ± 75% +32.5 35.44 ± 31% perf-profile.children.cycles-pp.common_startup_64 2.96 ± 75% -2.7 0.29 ± 65% -2.7 0.25 ± 75% +32.5 35.44 ± 31% perf-profile.children.cycles-pp.cpu_startup_entry 2.96 ± 75% -2.7 0.29 ± 65% -2.7 0.25 ± 75% +32.5 35.44 ± 31% perf-profile.children.cycles-pp.do_idle 2.80 ± 18% -1.3 1.46 ± 8% -1.5 1.28 ± 8% +0.1 2.93 ± 65% perf-profile.children.cycles-pp._raw_spin_lock_irq 1.17 ± 18% -1.2 0.00 -1.2 0.00 -1.0 0.15 ± 77% perf-profile.children.cycles-pp.get_pfnblock_flags_mask 2.42 ± 22% -0.9 1.49 ± 8% -1.1 1.30 ± 7% -1.1 1.27 ± 91% perf-profile.children.cycles-pp.__rmap_walk_file 1.35 ± 7% -0.8 0.53 ± 5% -0.9 0.49 ± 4% +3.5 4.90 ± 64% perf-profile.children.cycles-pp.iomap_readahead 1.35 ± 7% -0.8 0.54 ± 5% -0.9 0.49 ± 4% +3.5 4.90 ± 64% perf-profile.children.cycles-pp.read_pages 1.49 ± 6% -0.8 0.70 ± 5% -0.8 0.65 ± 4% +3.0 4.47 ± 58% perf-profile.children.cycles-pp.do_rw_once 1.26 ± 7% -0.8 0.50 ± 5% -0.8 0.46 ± 4% +3.4 4.67 ± 63% perf-profile.children.cycles-pp.iomap_readpage_iter 1.08 ± 8% -0.6 0.44 ± 6% -0.7 0.40 ± 5% +2.9 4.00 ± 63% perf-profile.children.cycles-pp.zero_user_segments 1.08 ± 7% -0.6 0.44 ± 6% -0.7 0.40 ± 5% +2.9 3.98 ± 63% perf-profile.children.cycles-pp.memset_orig 0.75 ± 37% -0.6 0.13 ±109% -0.7 0.08 ±110% +0.9 1.64 ±103% perf-profile.children.cycles-pp.shrink_slab_memcg 0.71 ± 39% -0.6 0.12 ±110% -0.6 0.07 ±110% +0.9 1.61 ±105% perf-profile.children.cycles-pp.do_shrink_slab 2.27 ± 8% -0.4 1.84 ± 3% -0.4 1.89 ± 7% +13.9 16.12 ± 35% perf-profile.children.cycles-pp.kthread 1.07 ± 21% -0.4 0.64 ± 8% -0.5 0.55 ± 6% -0.5 0.53 ± 94% perf-profile.children.cycles-pp.lru_gen_look_around 2.27 ± 8% -0.4 1.85 ± 3% -0.4 1.90 ± 7% +13.9 16.13 ± 35% perf-profile.children.cycles-pp.ret_from_fork 2.27 ± 8% -0.4 1.86 ± 3% -0.4 1.90 ± 7% +13.9 16.13 ± 35% perf-profile.children.cycles-pp.ret_from_fork_asm 0.36 ± 14% -0.3 0.03 ±223% -0.3 0.06 ±210% +9.4 9.74 ± 62% perf-profile.children.cycles-pp.migrate_pages_batch 0.36 ± 14% -0.3 0.03 ±223% -0.3 0.06 ±210% +9.4 9.75 ± 62% perf-profile.children.cycles-pp.migrate_pages 0.32 ± 18% -0.3 0.03 ±223% -0.3 0.06 ±210% +7.1 7.40 ± 64% perf-profile.children.cycles-pp.migrate_pages_sync 0.28 ± 25% -0.3 0.03 ±223% -0.2 0.06 ±211% +9.9 10.19 ± 63% perf-profile.children.cycles-pp.kcompactd 1.42 ± 14% -0.2 1.18 ± 4% -0.3 1.16 ± 4% -0.3 1.11 ± 21% perf-profile.children.cycles-pp._raw_spin_lock 0.21 ± 13% -0.2 0.00 -0.2 0.01 ±264% +1.6 1.78 ± 34% perf-profile.children.cycles-pp.migrate_folio_unmap 0.32 ± 10% -0.2 0.11 ± 6% -0.2 0.10 ± 6% +0.5 0.79 ± 60% perf-profile.children.cycles-pp.__filemap_remove_folio 0.21 ± 12% -0.2 0.00 -0.2 0.01 ±264% +1.4 1.56 ± 31% perf-profile.children.cycles-pp.isolate_freepages 0.21 ± 12% -0.2 0.00 -0.2 0.01 ±264% +1.5 1.67 ± 32% perf-profile.children.cycles-pp.compaction_alloc 0.19 ± 15% -0.2 0.00 -0.2 0.00 -0.2 0.03 ±109% perf-profile.children.cycles-pp.folio_mapping 0.48 ± 18% -0.2 0.29 ± 7% -0.2 0.27 ± 5% -0.1 0.39 ± 79% perf-profile.children.cycles-pp.try_to_unmap 0.33 ± 9% -0.2 0.14 ± 6% -0.2 0.13 ± 4% +0.4 0.72 ± 58% perf-profile.children.cycles-pp.isolate_folios 0.32 ± 9% -0.2 0.14 ± 7% -0.2 0.13 ± 5% +0.4 0.71 ± 58% perf-profile.children.cycles-pp.scan_folios 0.45 ± 18% -0.2 0.28 ± 6% -0.2 0.26 ± 4% -0.1 0.37 ± 79% perf-profile.children.cycles-pp.try_to_unmap_one 0.26 ± 15% -0.2 0.09 ± 12% -0.2 0.08 ± 21% +0.6 0.87 ± 83% perf-profile.children.cycles-pp.asm_sysvec_call_function 0.24 ± 8% -0.2 0.07 ± 6% -0.2 0.06 ± 7% +0.5 0.74 ± 27% perf-profile.children.cycles-pp.lru_add 0.23 ± 8% -0.2 0.08 ± 10% -0.2 0.07 ± 4% +0.3 0.58 ± 30% perf-profile.children.cycles-pp.lru_gen_add_folio 0.59 ± 10% -0.1 0.45 ± 6% -0.1 0.44 ± 4% -0.1 0.49 ± 81% perf-profile.children.cycles-pp.__drain_all_pages 0.21 ± 5% -0.1 0.07 ± 6% -0.1 0.07 ± 6% +0.6 0.77 ± 43% perf-profile.children.cycles-pp.iomap_release_folio 0.22 ± 8% -0.1 0.09 ± 6% -0.1 0.08 ± 5% +1.0 1.21 ± 31% perf-profile.children.cycles-pp.__free_one_page 0.20 ± 9% -0.1 0.08 ± 8% -0.1 0.07 +1.0 1.20 ± 40% perf-profile.children.cycles-pp.lru_gen_del_folio 0.11 ± 6% -0.1 0.00 -0.1 0.00 +0.7 0.79 ± 30% perf-profile.children.cycles-pp.kfree 0.18 ± 6% -0.1 0.07 ± 6% -0.1 0.07 ± 4% +0.3 0.52 ± 61% perf-profile.children.cycles-pp.filemap_map_pages 0.23 ± 7% -0.1 0.13 ± 5% -0.1 0.13 ± 4% +10.1 10.32 ± 70% perf-profile.children.cycles-pp.sysvec_apic_timer_interrupt 0.18 ± 7% -0.1 0.07 ± 15% -0.1 0.06 ± 8% +0.5 0.69 ± 52% perf-profile.children.cycles-pp.__lruvec_stat_mod_folio 0.25 ± 13% -0.1 0.15 ± 8% -0.1 0.14 ± 4% +0.7 0.90 ± 54% perf-profile.children.cycles-pp.folio_remove_rmap_ptes 0.14 ± 11% -0.1 0.04 ± 72% -0.1 0.03 ± 78% +0.5 0.63 ± 35% perf-profile.children.cycles-pp.__mod_memcg_lruvec_state 0.08 ± 26% -0.1 0.00 -0.1 0.00 +1.2 1.25 ± 60% perf-profile.children.cycles-pp.workingset_update_node 0.20 ± 6% -0.1 0.12 ± 4% -0.1 0.11 ± 4% +7.5 7.74 ± 71% perf-profile.children.cycles-pp.__sysvec_apic_timer_interrupt 0.19 ± 7% -0.1 0.11 ± 4% -0.1 0.11 ± 4% +7.5 7.73 ± 71% perf-profile.children.cycles-pp.hrtimer_interrupt 0.14 ± 19% -0.1 0.08 ± 8% -0.1 0.07 ± 8% -0.1 0.09 ± 92% perf-profile.children.cycles-pp.get_pfn_folio 0.06 ± 7% -0.1 0.00 -0.1 0.00 +0.5 0.56 ± 27% perf-profile.children.cycles-pp.ifs_free 0.10 ± 18% -0.1 0.04 ± 44% -0.1 0.03 ± 78% +0.0 0.14 ± 70% perf-profile.children.cycles-pp.__flush_smp_call_function_queue 0.16 ± 4% -0.1 0.09 ± 5% -0.1 0.09 ± 5% +6.8 6.94 ± 71% perf-profile.children.cycles-pp.__hrtimer_run_queues 0.12 ± 20% -0.1 0.06 ± 9% -0.1 0.06 ± 12% +0.0 0.13 ± 75% perf-profile.children.cycles-pp.sysvec_call_function 0.51 ± 5% -0.1 0.45 ± 6% -0.1 0.44 ± 4% -0.0 0.48 ± 81% perf-profile.children.cycles-pp.drain_pages_zone 0.10 ± 17% -0.1 0.04 ± 44% -0.1 0.03 ± 78% +0.0 0.11 ± 76% perf-profile.children.cycles-pp.__sysvec_call_function 0.06 ± 73% -0.1 0.00 -0.1 0.00 +1.2 1.22 ± 40% perf-profile.children.cycles-pp.isolate_freepages_block 0.12 ± 6% -0.0 0.08 ± 6% -0.0 0.07 ± 6% +4.7 4.79 ± 73% perf-profile.children.cycles-pp.tick_nohz_handler 0.11 ± 6% -0.0 0.07 ± 8% -0.0 0.07 ± 7% +3.9 4.05 ± 74% perf-profile.children.cycles-pp.update_process_times 0.04 ± 44% -0.0 0.00 -0.0 0.00 +0.0 0.08 ± 23% perf-profile.children.cycles-pp.task_tick_fair 0.04 ± 44% -0.0 0.00 -0.0 0.00 +3.0 3.02 ± 68% perf-profile.children.cycles-pp.handle_softirqs 0.08 ± 5% -0.0 0.05 ± 7% -0.0 0.05 ± 6% +1.4 1.51 ± 71% perf-profile.children.cycles-pp.sched_tick 0.03 ± 70% -0.0 0.00 -0.0 0.00 +2.1 2.13 ± 66% perf-profile.children.cycles-pp.__irq_exit_rcu 0.02 ± 99% -0.0 0.00 -0.0 0.00 +0.2 0.20 ± 26% perf-profile.children.cycles-pp.__mod_node_page_state 0.02 ± 99% -0.0 0.00 -0.0 0.00 +0.8 0.87 ± 60% perf-profile.children.cycles-pp.__slab_free 0.02 ± 99% -0.0 0.00 -0.0 0.00 +1.7 1.70 ± 61% perf-profile.children.cycles-pp.rcu_core 0.02 ±141% -0.0 0.00 -0.0 0.00 +2.3 2.32 ± 81% perf-profile.children.cycles-pp.folios_put_refs 0.01 ±223% -0.0 0.00 -0.0 0.00 +0.4 0.38 ± 45% perf-profile.children.cycles-pp.native_irq_return_iret 0.01 ±223% -0.0 0.00 -0.0 0.00 +1.5 1.53 ± 61% perf-profile.children.cycles-pp.rcu_do_batch 0.00 +0.0 0.00 +0.0 0.00 +0.1 0.09 ± 38% perf-profile.children.cycles-pp.cgroup_rstat_updated 0.00 +0.0 0.00 +0.0 0.00 +0.1 0.10 ± 11% perf-profile.children.cycles-pp.__mod_lruvec_state 0.00 +0.0 0.00 +0.0 0.00 +0.1 0.11 ± 13% perf-profile.children.cycles-pp.__mem_cgroup_uncharge_folios 0.00 +0.0 0.00 +0.0 0.00 +0.1 0.14 ± 71% perf-profile.children.cycles-pp.timekeeping_max_deferment 0.00 +0.0 0.00 +0.0 0.00 +0.2 0.16 ± 17% perf-profile.children.cycles-pp._raw_spin_trylock 0.00 +0.0 0.00 +0.0 0.00 +0.2 0.16 ± 63% perf-profile.children.cycles-pp.update_rq_clock_task 0.00 +0.0 0.00 +0.0 0.00 +0.2 0.17 ± 56% perf-profile.children.cycles-pp.xas_start 0.00 +0.0 0.00 +0.0 0.00 +0.2 0.17 ± 28% perf-profile.children.cycles-pp.filemap_unaccount_folio 0.00 +0.0 0.00 +0.0 0.00 +0.2 0.18 ± 39% perf-profile.children.cycles-pp.asm_sysvec_call_function_single 0.00 +0.0 0.00 +0.0 0.00 +0.2 0.21 ± 73% perf-profile.children.cycles-pp.lapic_next_deadline 0.00 +0.0 0.00 +0.0 0.00 +0.2 0.21 ± 70% perf-profile.children.cycles-pp.rcu_cblist_dequeue 0.00 +0.0 0.00 +0.0 0.00 +0.2 0.22 ± 58% perf-profile.children.cycles-pp.mem_cgroup_from_slab_obj 0.00 +0.0 0.00 +0.0 0.00 +0.2 0.22 ± 69% perf-profile.children.cycles-pp.update_sg_lb_stats 0.00 +0.0 0.00 +0.0 0.00 +0.2 0.23 ± 18% perf-profile.children.cycles-pp.folio_unlock 0.00 +0.0 0.00 +0.0 0.00 +0.2 0.23 ± 66% perf-profile.children.cycles-pp.__intel_pmu_enable_all 0.00 +0.0 0.00 +0.0 0.00 +0.2 0.25 ± 71% perf-profile.children.cycles-pp.irqtime_account_irq 0.00 +0.0 0.00 +0.0 0.00 +0.3 0.27 ± 39% perf-profile.children.cycles-pp.__split_unmapped_folio 0.00 +0.0 0.00 +0.0 0.00 +0.3 0.29 ± 70% perf-profile.children.cycles-pp.update_sd_lb_stats 0.00 +0.0 0.00 +0.0 0.00 +0.3 0.29 ± 74% perf-profile.children.cycles-pp.ktime_get_update_offsets_now 0.00 +0.0 0.00 +0.0 0.00 +0.3 0.30 ± 71% perf-profile.children.cycles-pp.sched_balance_find_src_group 0.00 +0.0 0.00 +0.0 0.00 +0.3 0.31 ± 85% perf-profile.children.cycles-pp.raw_spin_rq_lock_nested 0.00 +0.0 0.00 +0.0 0.00 +0.3 0.34 ± 72% perf-profile.children.cycles-pp.clockevents_program_event 0.00 +0.0 0.00 +0.0 0.00 +0.3 0.35 ± 64% perf-profile.children.cycles-pp.perf_rotate_context 0.00 +0.0 0.00 +0.0 0.00 +0.4 0.36 ± 69% perf-profile.children.cycles-pp.rcu_pending 0.00 +0.0 0.00 +0.0 0.00 +0.4 0.36 ± 70% perf-profile.children.cycles-pp.tick_nohz_next_event 0.00 +0.0 0.00 +0.0 0.00 +0.4 0.37 ± 77% perf-profile.children.cycles-pp.calc_global_load_tick 0.00 +0.0 0.00 +0.0 0.00 +0.4 0.40 ± 70% perf-profile.children.cycles-pp.rcu_sched_clock_irq 0.00 +0.0 0.00 +0.0 0.00 +0.4 0.41 ± 75% perf-profile.children.cycles-pp.arch_scale_freq_tick 0.00 +0.0 0.00 +0.0 0.00 +0.4 0.43 ± 74% perf-profile.children.cycles-pp.sched_balance_rq 0.00 +0.0 0.00 +0.0 0.00 +0.4 0.45 ± 76% perf-profile.children.cycles-pp.xas_clear_mark 0.00 +0.0 0.00 +0.0 0.00 +0.5 0.45 ± 69% perf-profile.children.cycles-pp.tick_nohz_get_sleep_length 0.00 +0.0 0.00 +0.0 0.00 +0.5 0.48 ± 52% perf-profile.children.cycles-pp.xas_load 0.00 +0.0 0.00 +0.0 0.00 +0.5 0.48 ± 76% perf-profile.children.cycles-pp.__memcg_slab_free_hook 0.00 +0.0 0.00 +0.0 0.00 +0.5 0.50 ± 49% perf-profile.children.cycles-pp.__free_frozen_pages 0.00 +0.0 0.00 +0.0 0.00 +0.6 0.59 ± 50% perf-profile.children.cycles-pp.ast_primary_plane_helper_atomic_update 0.00 +0.0 0.00 +0.0 0.00 +0.6 0.59 ± 50% perf-profile.children.cycles-pp.drm_fb_memcpy 0.00 +0.0 0.00 +0.0 0.00 +0.6 0.59 ± 50% perf-profile.children.cycles-pp.memcpy_toio 0.00 +0.0 0.00 +0.0 0.00 +0.6 0.59 ± 50% perf-profile.children.cycles-pp.ast_mode_config_helper_atomic_commit_tail 0.00 +0.0 0.00 +0.0 0.00 +0.6 0.59 ± 50% perf-profile.children.cycles-pp.commit_tail 0.00 +0.0 0.00 +0.0 0.00 +0.6 0.59 ± 50% perf-profile.children.cycles-pp.drm_atomic_commit 0.00 +0.0 0.00 +0.0 0.00 +0.6 0.59 ± 50% perf-profile.children.cycles-pp.drm_atomic_helper_commit 0.00 +0.0 0.00 +0.0 0.00 +0.6 0.59 ± 50% perf-profile.children.cycles-pp.drm_atomic_helper_commit_tail 0.00 +0.0 0.00 +0.0 0.00 +0.6 0.60 ± 49% perf-profile.children.cycles-pp.drm_atomic_helper_dirtyfb 0.00 +0.0 0.00 +0.0 0.00 +0.6 0.60 ± 49% perf-profile.children.cycles-pp.drm_fb_helper_damage_work 0.00 +0.0 0.00 +0.0 0.00 +0.6 0.60 ± 49% perf-profile.children.cycles-pp.drm_fbdev_shmem_helper_fb_dirty 0.00 +0.0 0.00 +0.0 0.00 +0.6 0.63 ± 45% perf-profile.children.cycles-pp.process_one_work 0.00 +0.0 0.00 +0.0 0.00 +0.6 0.64 ± 80% perf-profile.children.cycles-pp.xas_get_order 0.00 +0.0 0.00 +0.0 0.00 +0.6 0.64 ± 44% perf-profile.children.cycles-pp.worker_thread 0.00 +0.0 0.00 +0.0 0.00 +0.7 0.65 ± 88% perf-profile.children.cycles-pp.__page_cache_release 0.00 +0.0 0.00 +0.0 0.00 +0.7 0.69 ± 95% perf-profile.children.cycles-pp.truncate_cleanup_folio 0.00 +0.0 0.00 +0.0 0.00 +0.7 0.70 ± 77% perf-profile.children.cycles-pp.list_lru_del 0.00 +0.0 0.00 +0.0 0.00 +0.7 0.70 ± 70% perf-profile.children.cycles-pp.menu_select 0.00 +0.0 0.00 +0.0 0.00 +1.0 0.96 ± 79% perf-profile.children.cycles-pp.list_lru_del_obj 0.00 +0.0 0.00 +0.0 0.00 +1.0 0.99 ± 77% perf-profile.children.cycles-pp.__folio_migrate_mapping 0.00 +0.0 0.00 +0.0 0.00 +1.0 1.02 ± 79% perf-profile.children.cycles-pp.sched_balance_domains 0.00 +0.0 0.00 +0.0 0.00 +1.0 1.02 ± 72% perf-profile.children.cycles-pp.xas_find 0.00 +0.0 0.00 +0.0 0.00 +1.0 1.03 ± 72% perf-profile.children.cycles-pp.run_ksoftirqd 0.00 +0.0 0.00 +0.0 0.00 +1.1 1.06 ± 71% perf-profile.children.cycles-pp.smpboot_thread_fn 0.00 +0.0 0.00 +0.0 0.00 +1.1 1.10 ± 73% perf-profile.children.cycles-pp.kmem_cache_free 0.00 +0.0 0.00 +0.0 0.00 +1.6 1.57 ± 76% perf-profile.children.cycles-pp.get_jiffies_update 0.00 +0.0 0.00 +0.0 0.00 +1.6 1.59 ± 75% perf-profile.children.cycles-pp.tmigr_requires_handle_remote 0.00 +0.0 0.00 +0.0 0.00 +1.8 1.83 ± 68% perf-profile.children.cycles-pp.ktime_get 0.00 +0.0 0.00 +0.0 0.00 +1.9 1.87 ± 66% perf-profile.children.cycles-pp.perf_mux_hrtimer_handler 0.00 +0.0 0.00 +0.0 0.00 +2.4 2.40 ± 83% perf-profile.children.cycles-pp.find_lock_entries 0.00 +0.0 0.00 +0.0 0.00 +3.8 3.77 ± 81% perf-profile.children.cycles-pp.truncate_folio_batch_exceptionals 0.00 +0.0 0.00 +0.0 0.00 +4.5 4.49 ± 78% perf-profile.children.cycles-pp.copy_mc_fragile 0.00 +0.0 0.00 +0.0 0.00 +4.5 4.52 ± 78% perf-profile.children.cycles-pp.folio_mc_copy 0.00 +0.0 0.00 +0.0 0.00 +5.8 5.83 ± 79% perf-profile.children.cycles-pp.__migrate_folio 0.00 +0.0 0.00 +0.0 0.00 +5.9 5.93 ± 78% perf-profile.children.cycles-pp.move_to_new_folio 0.00 +0.0 0.00 +0.0 0.00 +8.5 8.54 ± 82% perf-profile.children.cycles-pp.compact_node 0.00 +0.0 0.00 +0.0 0.00 +9.5 9.48 ± 85% perf-profile.children.cycles-pp.evict 0.00 +0.0 0.00 +0.0 0.00 +9.5 9.48 ± 85% perf-profile.children.cycles-pp.truncate_inode_pages_range 0.00 +0.0 0.00 +0.0 0.00 +9.5 9.48 ± 85% perf-profile.children.cycles-pp.xfs_fs_evict_inode 0.00 +0.0 0.00 +0.0 0.00 +9.5 9.48 ± 85% perf-profile.children.cycles-pp.__x64_sys_unlinkat 0.00 +0.0 0.00 +0.0 0.00 +9.5 9.48 ± 85% perf-profile.children.cycles-pp.do_unlinkat 0.00 +0.0 0.00 +0.0 0.00 +9.5 9.48 ± 85% perf-profile.children.cycles-pp.unlinkat 0.00 +0.0 0.00 +0.0 0.03 ±264% +0.8 0.76 ± 36% perf-profile.children.cycles-pp.__folio_split 0.00 +0.0 0.00 +0.8 0.79 ± 4% +0.0 0.00 perf-profile.children.cycles-pp.try_to_claim_block 0.00 +0.0 0.00 +1.3 1.28 ± 12% +7.4 7.38 ± 89% perf-profile.children.cycles-pp.folio_add_lru 0.00 +0.0 0.01 ±223% +0.0 0.00 +0.2 0.25 ± 41% perf-profile.children.cycles-pp.rest_init 0.00 +0.0 0.01 ±223% +0.0 0.00 +0.2 0.25 ± 41% perf-profile.children.cycles-pp.start_kernel 0.00 +0.0 0.01 ±223% +0.0 0.00 +0.2 0.25 ± 41% perf-profile.children.cycles-pp.x86_64_start_kernel 0.00 +0.0 0.01 ±223% +0.0 0.00 +0.2 0.25 ± 41% perf-profile.children.cycles-pp.x86_64_start_reservations 0.00 +0.0 0.02 ±142% +0.1 0.06 ± 20% +0.0 0.00 perf-profile.children.cycles-pp.copy_process 0.00 +0.0 0.02 ±142% +0.1 0.06 ± 24% +0.0 0.00 perf-profile.children.cycles-pp.kernel_clone 0.10 ± 17% +0.0 0.12 ± 8% +0.0 0.12 ± 11% +0.4 0.51 ± 55% perf-profile.children.cycles-pp.vfs_write 0.00 +0.0 0.03 ±100% +0.1 0.05 ± 6% +0.0 0.00 perf-profile.children.cycles-pp.__pud_alloc 0.08 ± 14% +0.0 0.12 ± 8% +0.0 0.12 ± 11% -0.1 0.00 perf-profile.children.cycles-pp.record__pushfn 0.08 ± 14% +0.0 0.12 ± 8% +0.0 0.12 ± 11% -0.1 0.00 perf-profile.children.cycles-pp.writen 0.07 ± 18% +0.0 0.11 ± 9% +0.0 0.11 ± 12% -0.1 0.00 perf-profile.children.cycles-pp.generic_perform_write 0.07 ± 20% +0.0 0.12 ± 10% +0.0 0.11 ± 12% -0.1 0.00 perf-profile.children.cycles-pp.shmem_file_write_iter 0.13 ± 10% +0.0 0.17 ± 7% +0.0 0.16 ± 9% -0.1 0.00 perf-profile.children.cycles-pp.perf_mmap__push 0.14 ± 10% +0.0 0.18 ± 6% +0.0 0.17 ± 8% -0.1 0.00 perf-profile.children.cycles-pp.record__mmap_read_evlist 0.00 +0.1 0.05 ± 7% +0.0 0.04 ± 62% +0.0 0.01 ±173% perf-profile.children.cycles-pp.load_elf_binary 0.00 +0.1 0.05 ± 7% +0.0 0.04 ± 62% +0.0 0.01 ±174% perf-profile.children.cycles-pp.exec_binprm 0.00 +0.1 0.05 ± 8% +0.0 0.04 ± 62% +0.0 0.01 ±174% perf-profile.children.cycles-pp.bprm_execve 0.03 ±100% +0.1 0.09 ± 11% +0.1 0.09 ± 14% -0.0 0.00 perf-profile.children.cycles-pp.shmem_get_folio_gfp 0.03 ±100% +0.1 0.09 ± 11% +0.1 0.09 ± 14% -0.0 0.00 perf-profile.children.cycles-pp.shmem_write_begin 0.03 ±100% +0.1 0.09 ± 9% +0.1 0.09 ± 15% -0.0 0.00 perf-profile.children.cycles-pp.shmem_alloc_and_add_folio 0.00 +0.1 0.06 ± 25% +0.1 0.06 ± 12% +0.0 0.00 perf-profile.children.cycles-pp.alloc_anon_folio 0.01 ±223% +0.1 0.07 ± 14% +0.1 0.07 ± 17% -0.0 0.00 perf-profile.children.cycles-pp.shmem_alloc_folio 0.00 +0.1 0.07 ± 18% +0.1 0.06 ± 13% +0.0 0.00 perf-profile.children.cycles-pp.copy_string_kernel 0.00 +0.1 0.07 ± 26% +0.1 0.07 ± 10% +0.0 0.00 perf-profile.children.cycles-pp.do_anonymous_page 0.00 +0.1 0.08 ± 16% +0.1 0.07 ± 9% +0.0 0.00 perf-profile.children.cycles-pp.__get_user_pages 0.00 +0.1 0.08 ± 16% +0.1 0.07 ± 9% +0.0 0.00 perf-profile.children.cycles-pp.get_arg_page 0.00 +0.1 0.08 ± 16% +0.1 0.07 ± 9% +0.0 0.00 perf-profile.children.cycles-pp.get_user_pages_remote 0.00 +0.1 0.08 ± 14% +0.1 0.09 ± 28% +0.0 0.00 perf-profile.children.cycles-pp.do_sync_mmap_readahead 0.27 ± 24% +0.1 0.40 ± 8% +0.1 0.40 ± 4% +12.9 13.13 ± 53% perf-profile.children.cycles-pp.do_syscall_64 0.27 ± 24% +0.1 0.40 ± 8% +0.1 0.40 ± 4% +12.9 13.13 ± 53% perf-profile.children.cycles-pp.entry_SYSCALL_64_after_hwframe 0.00 +0.1 0.13 ± 11% +0.1 0.12 ± 8% +0.0 0.00 perf-profile.children.cycles-pp.vma_alloc_folio_noprof 0.00 +0.2 0.16 ± 12% +0.2 0.15 ± 9% +0.0 0.03 ±135% perf-profile.children.cycles-pp.__x64_sys_execve 0.00 +0.2 0.16 ± 12% +0.2 0.15 ± 9% +0.0 0.03 ±135% perf-profile.children.cycles-pp.do_execveat_common 0.00 +0.2 0.16 ± 12% +0.2 0.15 ± 9% +0.0 0.03 ±135% perf-profile.children.cycles-pp.execve 0.46 ± 22% +0.2 0.65 ± 6% +0.2 0.66 ± 3% +4.3 4.75 ± 56% perf-profile.children.cycles-pp.xas_store 0.01 ±223% +0.2 0.20 ± 9% +0.1 0.07 ± 17% -0.0 0.00 perf-profile.children.cycles-pp.folio_alloc_mpol_noprof 0.30 ± 32% +0.3 0.60 ± 7% +0.3 0.62 ± 4% +0.8 1.09 ± 26% perf-profile.children.cycles-pp.xas_create 0.29 ± 34% +0.4 0.64 ± 7% +0.4 0.65 ± 4% +0.1 0.42 ± 67% perf-profile.children.cycles-pp.__filemap_add_folio 0.35 ± 17% +0.4 0.70 ± 2% +0.4 0.71 ± 4% -0.3 0.00 perf-profile.children.cycles-pp.prep_move_freepages_block 0.14 ± 70% +0.4 0.55 ± 8% +0.4 0.57 ± 4% -0.1 0.02 ±129% perf-profile.children.cycles-pp.xas_alloc 0.22 ± 51% +0.4 0.62 ± 8% +0.4 0.64 ± 5% +0.1 0.33 ± 69% perf-profile.children.cycles-pp.___slab_alloc 0.14 ± 70% +0.5 0.60 ± 8% +0.5 0.61 ± 5% -0.1 0.02 ±129% perf-profile.children.cycles-pp.kmem_cache_alloc_lru_noprof 0.14 ± 72% +0.5 0.60 ± 8% +0.5 0.62 ± 6% -0.1 0.00 perf-profile.children.cycles-pp.allocate_slab 0.05 ±223% +0.7 0.71 ± 5% +0.6 0.60 ± 8% -0.0 0.00 perf-profile.children.cycles-pp.unreserve_highatomic_pageblock 0.00 +0.8 0.80 +0.0 0.00 +0.0 0.00 perf-profile.children.cycles-pp.try_to_steal_block 0.00 +1.5 1.50 ± 7% +1.6 1.60 ± 10% +0.1 0.09 ± 86% perf-profile.children.cycles-pp.free_one_page 91.27 ± 2% +1.9 93.12 +2.0 93.23 -62.8 28.44 ± 67% perf-profile.children.cycles-pp.filemap_fault 91.27 ± 2% +1.9 93.12 +2.0 93.24 -62.8 28.44 ± 67% perf-profile.children.cycles-pp.__do_fault 0.88 ± 11% +2.2 3.12 +2.2 3.10 -0.6 0.25 ± 82% perf-profile.children.cycles-pp.pte_alloc_one 0.89 ± 11% +2.4 3.29 +2.4 3.28 -0.6 0.25 ± 83% perf-profile.children.cycles-pp.alloc_pages_noprof 93.43 ± 2% +3.3 96.74 +3.4 96.86 -61.7 31.68 ± 66% perf-profile.children.cycles-pp.do_access 92.33 ± 2% +4.0 96.28 +4.0 96.37 -63.1 29.21 ± 67% perf-profile.children.cycles-pp.do_read_fault 92.33 ± 2% +4.0 96.31 +4.1 96.40 -63.1 29.22 ± 67% perf-profile.children.cycles-pp.do_pte_missing 92.49 ± 2% +4.0 96.52 +4.1 96.60 -62.9 29.61 ± 67% perf-profile.children.cycles-pp.asm_exc_page_fault 92.41 ± 2% +4.0 96.46 +4.1 96.54 -63.0 29.44 ± 67% perf-profile.children.cycles-pp.exc_page_fault 92.41 ± 2% +4.0 96.46 +4.1 96.54 -63.0 29.43 ± 67% perf-profile.children.cycles-pp.do_user_addr_fault 92.37 ± 2% +4.1 96.52 +4.2 96.59 -63.0 29.33 ± 67% perf-profile.children.cycles-pp.handle_mm_fault 92.35 ± 2% +4.2 96.51 +4.2 96.59 -63.1 29.29 ± 67% perf-profile.children.cycles-pp.__handle_mm_fault 0.30 ± 15% +5.5 5.78 ± 10% +5.6 5.86 ± 9% -0.3 0.00 perf-profile.children.cycles-pp.page_cache_ra_unbounded 17.97 ± 9% +5.6 23.56 +5.8 23.77 -15.1 2.84 ± 67% perf-profile.children.cycles-pp.try_to_unmap_flush 17.97 ± 9% +5.6 23.56 +5.8 23.77 -15.1 2.84 ± 67% perf-profile.children.cycles-pp.arch_tlbbatch_flush 17.97 ± 9% +5.6 23.58 +5.8 23.79 -15.1 2.84 ± 67% perf-profile.children.cycles-pp.on_each_cpu_cond_mask 17.97 ± 9% +5.6 23.58 +5.8 23.79 -15.1 2.84 ± 67% perf-profile.children.cycles-pp.smp_call_function_many_cond 82.68 ± 2% +7.8 90.43 +8.1 90.79 -66.7 16.00 ± 70% perf-profile.children.cycles-pp.folio_alloc_noprof 22.07 ± 7% +7.9 29.92 ± 2% +8.6 30.70 ± 2% -18.1 3.95 ± 44% perf-profile.children.cycles-pp.free_frozen_page_commit 22.20 ± 7% +7.9 30.09 ± 2% +8.7 30.86 ± 2% -18.0 4.16 ± 49% perf-profile.children.cycles-pp.free_pcppages_bulk 22.13 ± 7% +9.3 31.42 ± 2% +10.2 32.29 ± 2% -17.9 4.22 ± 40% perf-profile.children.cycles-pp.free_unref_folios 83.71 ± 2% +10.8 94.50 +11.2 94.86 -67.6 16.14 ± 71% perf-profile.children.cycles-pp.alloc_pages_mpol 83.71 ± 2% +10.8 94.52 +11.2 94.88 -67.6 16.13 ± 71% perf-profile.children.cycles-pp.__alloc_frozen_pages_noprof 47.00 ± 8% +12.0 58.98 +12.7 59.67 -34.1 12.93 ± 64% perf-profile.children.cycles-pp.shrink_node 46.98 ± 8% +12.0 58.97 +12.7 59.66 -34.1 12.92 ± 64% perf-profile.children.cycles-pp.shrink_many 46.97 ± 8% +12.0 58.96 +12.7 59.65 -34.1 12.91 ± 64% perf-profile.children.cycles-pp.shrink_one 45.19 ± 8% +12.0 57.22 +12.7 57.89 -36.3 8.90 ± 70% perf-profile.children.cycles-pp.try_to_free_pages 45.07 ± 8% +12.1 57.18 +12.8 57.86 -36.3 8.82 ± 70% perf-profile.children.cycles-pp.do_try_to_free_pages 46.21 ± 7% +12.6 58.81 +13.3 59.55 -35.0 11.25 ± 64% perf-profile.children.cycles-pp.try_to_shrink_lruvec 46.18 ± 7% +12.6 58.79 +13.4 59.54 -35.0 11.19 ± 64% perf-profile.children.cycles-pp.evict_folios 43.67 ± 7% +13.7 57.42 +14.7 58.34 -34.5 9.14 ± 64% perf-profile.children.cycles-pp.shrink_folio_list 16.71 ± 7% +18.9 35.65 +18.7 35.46 -12.1 4.63 ± 77% perf-profile.children.cycles-pp.rmqueue 17.15 ± 6% +18.9 36.10 +18.7 35.90 -12.0 5.15 ± 75% perf-profile.children.cycles-pp.get_page_from_freelist 52.45 ± 6% +19.0 71.46 +19.2 71.60 -35.9 16.55 ± 72% perf-profile.children.cycles-pp.native_queued_spin_lock_slowpath 12.76 ± 7% +20.1 32.82 ± 2% +19.8 32.60 -9.3 3.47 ± 75% perf-profile.children.cycles-pp.__rmqueue_pcplist 12.73 ± 7% +20.1 32.81 ± 2% +19.8 32.58 -9.3 3.43 ± 76% perf-profile.children.cycles-pp.rmqueue_bulk 48.52 ± 6% +20.4 68.92 +20.7 69.26 -34.1 14.39 ± 69% perf-profile.children.cycles-pp._raw_spin_lock_irqsave 68.16 ± 2% +26.1 94.26 +26.5 94.61 -56.9 11.28 ± 72% perf-profile.children.cycles-pp.__alloc_pages_slowpath 11.45 ± 16% -11.4 0.00 -11.4 0.00 -10.5 0.91 ±102% perf-profile.self.cycles-pp.isolate_migratepages_block 2.95 ± 75% -2.7 0.29 ± 66% -2.7 0.25 ± 75% +20.7 23.67 ± 10% perf-profile.self.cycles-pp.acpi_safe_halt 1.17 ± 17% -1.2 0.00 -1.2 0.00 -1.0 0.14 ± 75% perf-profile.self.cycles-pp.get_pfnblock_flags_mask 1.24 ± 6% -0.7 0.58 ± 5% -0.7 0.54 ± 4% +3.1 4.33 ± 58% perf-profile.self.cycles-pp.do_rw_once 1.07 ± 7% -0.6 0.44 ± 6% -0.7 0.40 ± 5% +2.9 3.93 ± 63% perf-profile.self.cycles-pp.memset_orig 0.69 ± 6% -0.4 0.32 ± 5% -0.4 0.30 ± 5% +0.6 1.27 ± 61% perf-profile.self.cycles-pp.do_access 0.18 ± 9% -0.1 0.06 ± 9% -0.1 0.05 +0.6 0.81 ± 17% perf-profile.self.cycles-pp._raw_spin_lock_irqsave 0.14 ± 5% -0.1 0.05 ± 8% -0.1 0.04 ± 37% +0.9 1.02 ± 29% perf-profile.self.cycles-pp.xas_create 0.13 ± 12% -0.1 0.04 ± 71% -0.1 0.03 ± 77% +0.4 0.54 ± 44% perf-profile.self.cycles-pp.__mod_memcg_lruvec_state 0.18 ± 8% -0.1 0.09 ± 6% -0.1 0.09 ± 8% +0.3 0.50 ± 62% perf-profile.self.cycles-pp.rmqueue_bulk 0.16 ± 8% -0.1 0.08 ± 6% -0.1 0.08 ± 6% +1.0 1.13 ± 34% perf-profile.self.cycles-pp.__free_one_page 0.08 ± 5% -0.1 0.00 -0.1 0.00 +0.1 0.17 ± 8% perf-profile.self.cycles-pp.free_frozen_page_commit 0.12 ± 8% -0.1 0.04 ± 71% -0.1 0.01 ±264% +0.9 1.00 ± 47% perf-profile.self.cycles-pp.lru_gen_del_folio 0.14 ± 7% -0.1 0.06 ± 9% -0.1 0.05 +0.3 0.39 ± 35% perf-profile.self.cycles-pp.lru_gen_add_folio 0.08 ± 7% -0.1 0.00 -0.1 0.00 +0.6 0.68 ± 37% perf-profile.self.cycles-pp._raw_spin_lock 0.23 ± 14% -0.1 0.16 ± 5% -0.1 0.15 ± 6% +0.1 0.29 ± 76% perf-profile.self.cycles-pp.get_page_from_freelist 0.14 ± 18% -0.1 0.08 ± 9% -0.1 0.07 ± 7% -0.1 0.08 ± 91% perf-profile.self.cycles-pp.get_pfn_folio 0.19 ± 12% -0.1 0.12 ± 7% -0.1 0.12 ± 5% +0.5 0.66 ± 54% perf-profile.self.cycles-pp.folio_remove_rmap_ptes 0.06 ± 7% -0.1 0.00 -0.1 0.00 +0.5 0.55 ± 28% perf-profile.self.cycles-pp.ifs_free 0.10 ± 17% -0.1 0.04 ± 44% -0.1 0.01 ±173% -0.0 0.08 ± 89% perf-profile.self.cycles-pp.try_to_unmap_one 0.04 ± 45% -0.0 0.00 -0.0 0.00 +0.2 0.24 ± 42% perf-profile.self.cycles-pp.free_pcppages_bulk 0.03 ± 70% -0.0 0.00 -0.0 0.00 +0.2 0.19 ± 56% perf-profile.self.cycles-pp.__lruvec_stat_mod_folio 0.02 ± 99% -0.0 0.00 -0.0 0.00 +0.2 0.18 ± 24% perf-profile.self.cycles-pp.lru_add 0.02 ± 99% -0.0 0.00 -0.0 0.00 +1.6 1.64 ± 73% perf-profile.self.cycles-pp.xas_store 0.02 ±141% -0.0 0.00 -0.0 0.00 +0.2 0.19 ± 27% perf-profile.self.cycles-pp.__mod_node_page_state 0.01 ±223% -0.0 0.00 -0.0 0.00 +0.4 0.38 ± 45% perf-profile.self.cycles-pp.native_irq_return_iret 0.01 ±223% -0.0 0.00 -0.0 0.00 +0.5 0.46 ± 41% perf-profile.self.cycles-pp.free_unref_folios 0.01 ±223% -0.0 0.00 -0.0 0.00 +0.5 0.54 ± 53% perf-profile.self.cycles-pp.folios_put_refs 0.01 ±223% -0.0 0.00 -0.0 0.00 +0.9 0.86 ± 59% perf-profile.self.cycles-pp.__slab_free 0.00 +0.0 0.00 +0.0 0.00 +0.1 0.06 ± 17% perf-profile.self.cycles-pp.folio_add_lru 0.00 +0.0 0.00 +0.0 0.00 +0.1 0.09 ± 38% perf-profile.self.cycles-pp.cgroup_rstat_updated 0.00 +0.0 0.00 +0.0 0.00 +0.1 0.11 ± 37% perf-profile.self.cycles-pp.asm_sysvec_apic_timer_interrupt 0.00 +0.0 0.00 +0.0 0.00 +0.1 0.13 ± 64% perf-profile.self.cycles-pp.kfree 0.00 +0.0 0.00 +0.0 0.00 +0.1 0.14 ± 70% perf-profile.self.cycles-pp.timekeeping_max_deferment 0.00 +0.0 0.00 +0.0 0.00 +0.1 0.14 ± 48% perf-profile.self.cycles-pp.workingset_update_node 0.00 +0.0 0.00 +0.0 0.00 +0.2 0.15 ± 20% perf-profile.self.cycles-pp._raw_spin_trylock 0.00 +0.0 0.00 +0.0 0.00 +0.2 0.16 ± 58% perf-profile.self.cycles-pp.xas_start 0.00 +0.0 0.00 +0.0 0.00 +0.2 0.19 ± 72% perf-profile.self.cycles-pp.__memcg_slab_free_hook 0.00 +0.0 0.00 +0.0 0.00 +0.2 0.20 ± 70% perf-profile.self.cycles-pp.kmem_cache_free 0.00 +0.0 0.00 +0.0 0.00 +0.2 0.21 ± 73% perf-profile.self.cycles-pp.lapic_next_deadline 0.00 +0.0 0.00 +0.0 0.00 +0.2 0.21 ± 70% perf-profile.self.cycles-pp.rcu_cblist_dequeue 0.00 +0.0 0.00 +0.0 0.00 +0.2 0.22 ± 58% perf-profile.self.cycles-pp.mem_cgroup_from_slab_obj 0.00 +0.0 0.00 +0.0 0.00 +0.2 0.22 ± 18% perf-profile.self.cycles-pp.folio_unlock 0.00 +0.0 0.00 +0.0 0.00 +0.2 0.23 ± 66% perf-profile.self.cycles-pp.__intel_pmu_enable_all 0.00 +0.0 0.00 +0.0 0.00 +0.3 0.27 ± 75% perf-profile.self.cycles-pp.ktime_get_update_offsets_now 0.00 +0.0 0.00 +0.0 0.00 +0.3 0.30 ± 47% perf-profile.self.cycles-pp.xas_load 0.00 +0.0 0.00 +0.0 0.00 +0.4 0.36 ± 77% perf-profile.self.cycles-pp.calc_global_load_tick 0.00 +0.0 0.00 +0.0 0.00 +0.4 0.37 ± 48% perf-profile.self.cycles-pp._raw_spin_lock_irq 0.00 +0.0 0.00 +0.0 0.00 +0.4 0.41 ± 75% perf-profile.self.cycles-pp.arch_scale_freq_tick 0.00 +0.0 0.00 +0.0 0.00 +0.4 0.43 ± 76% perf-profile.self.cycles-pp.xas_clear_mark 0.00 +0.0 0.00 +0.0 0.00 +0.5 0.52 ± 80% perf-profile.self.cycles-pp.truncate_folio_batch_exceptionals 0.00 +0.0 0.00 +0.0 0.00 +0.6 0.59 ± 50% perf-profile.self.cycles-pp.memcpy_toio 0.00 +0.0 0.00 +0.0 0.00 +0.6 0.60 ± 81% perf-profile.self.cycles-pp.xas_get_order 0.00 +0.0 0.00 +0.0 0.00 +0.6 0.62 ± 72% perf-profile.self.cycles-pp.tick_nohz_handler 0.00 +0.0 0.00 +0.0 0.00 +0.6 0.63 ± 79% perf-profile.self.cycles-pp.sched_balance_domains 0.00 +0.0 0.00 +0.0 0.00 +0.7 0.66 ± 78% perf-profile.self.cycles-pp.list_lru_del 0.00 +0.0 0.00 +0.0 0.00 +0.8 0.84 ± 76% perf-profile.self.cycles-pp.xas_find 0.00 +0.0 0.00 +0.0 0.00 +0.9 0.86 ± 40% perf-profile.self.cycles-pp.isolate_freepages_block 0.00 +0.0 0.00 +0.0 0.00 +0.9 0.92 ± 88% perf-profile.self.cycles-pp.find_lock_entries 0.00 +0.0 0.00 +0.0 0.00 +1.6 1.57 ± 76% perf-profile.self.cycles-pp.get_jiffies_update 0.00 +0.0 0.00 +0.0 0.00 +1.7 1.70 ± 68% perf-profile.self.cycles-pp.ktime_get 0.00 +0.0 0.00 +0.0 0.00 +4.5 4.45 ± 78% perf-profile.self.cycles-pp.copy_mc_fragile 0.00 +0.0 0.00 +0.1 0.09 ± 5% +0.0 0.00 perf-profile.self.cycles-pp.try_to_claim_block 0.00 +0.1 0.09 +0.0 0.00 +0.0 0.00 perf-profile.self.cycles-pp.try_to_steal_block 0.35 ± 17% +0.4 0.70 ± 2% +0.4 0.71 ± 4% -0.3 0.00 perf-profile.self.cycles-pp.prep_move_freepages_block 17.81 ± 9% +5.7 23.46 +5.9 23.67 -15.1 2.72 ± 67% perf-profile.self.cycles-pp.smp_call_function_many_cond 52.45 ± 6% +19.0 71.46 +19.2 71.60 -35.9 16.55 ± 72% perf-profile.self.cycles-pp.native_queued_spin_lock_slowpath > > commit acc4d5ff0b61eb1715c498b6536c38c1feb7f3c1 (origin/master, origin/HEAD) > Merge: 3491aa04787f f278b6d5bb46 > Author: Linus Torvalds > Date: Tue Apr 1 20:00:51 2025 -0700 > > Merge tag 'net-6.15-rc0' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net > > Thanks! > > --- > > From 13433454403e0c6f99ccc3b76c609034fe47e41c Mon Sep 17 00:00:00 2001 > From: Johannes Weiner > Date: Wed, 2 Apr 2025 14:23:53 -0400 > Subject: [PATCH] mm: page_alloc: speed up fallbacks in rmqueue_bulk() > > Not-yet-signed-off-by: Johannes Weiner > --- > mm/page_alloc.c | 100 +++++++++++++++++++++++++++++++++++------------- > 1 file changed, 74 insertions(+), 26 deletions(-) > > diff --git a/mm/page_alloc.c b/mm/page_alloc.c > index f51aa6051a99..03b0d45ed45a 100644 > --- a/mm/page_alloc.c > +++ b/mm/page_alloc.c > @@ -2194,11 +2194,11 @@ try_to_claim_block(struct zone *zone, struct page *page, > * The use of signed ints for order and current_order is a deliberate > * deviation from the rest of this file, to make the for loop > * condition simpler. > - * > - * Return the stolen page, or NULL if none can be found. > */ > + > +/* Try to claim a whole foreign block, take a page, expand the remainder */ > static __always_inline struct page * > -__rmqueue_fallback(struct zone *zone, int order, int start_migratetype, > +__rmqueue_claim(struct zone *zone, int order, int start_migratetype, > unsigned int alloc_flags) > { > struct free_area *area; > @@ -2236,14 +2236,26 @@ __rmqueue_fallback(struct zone *zone, int order, int start_migratetype, > page = try_to_claim_block(zone, page, current_order, order, > start_migratetype, fallback_mt, > alloc_flags); > - if (page) > - goto got_one; > + if (page) { > + trace_mm_page_alloc_extfrag(page, order, current_order, > + start_migratetype, fallback_mt); > + return page; > + } > } > > - if (alloc_flags & ALLOC_NOFRAGMENT) > - return NULL; > + return NULL; > +} > + > +/* Try to steal a single page from a foreign block */ > +static __always_inline struct page * > +__rmqueue_steal(struct zone *zone, int order, int start_migratetype) > +{ > + struct free_area *area; > + int current_order; > + struct page *page; > + int fallback_mt; > + bool claim_block; > > - /* No luck claiming pageblock. Find the smallest fallback page */ > for (current_order = order; current_order < NR_PAGE_ORDERS; current_order++) { > area = &(zone->free_area[current_order]); > fallback_mt = find_suitable_fallback(area, current_order, > @@ -2253,25 +2265,28 @@ __rmqueue_fallback(struct zone *zone, int order, int start_migratetype, > > page = get_page_from_free_area(area, fallback_mt); > page_del_and_expand(zone, page, order, current_order, fallback_mt); > - goto got_one; > + trace_mm_page_alloc_extfrag(page, order, current_order, > + start_migratetype, fallback_mt); > + return page; > } > > return NULL; > - > -got_one: > - trace_mm_page_alloc_extfrag(page, order, current_order, > - start_migratetype, fallback_mt); > - > - return page; > } > > +enum rmqueue_mode { > + RMQUEUE_NORMAL, > + RMQUEUE_CMA, > + RMQUEUE_CLAIM, > + RMQUEUE_STEAL, > +}; > + > /* > * Do the hard work of removing an element from the buddy allocator. > * Call me with the zone->lock already held. > */ > static __always_inline struct page * > __rmqueue(struct zone *zone, unsigned int order, int migratetype, > - unsigned int alloc_flags) > + unsigned int alloc_flags, enum rmqueue_mode *mode) > { > struct page *page; > > @@ -2290,16 +2305,47 @@ __rmqueue(struct zone *zone, unsigned int order, int migratetype, > } > } > > - page = __rmqueue_smallest(zone, order, migratetype); > - if (unlikely(!page)) { > - if (alloc_flags & ALLOC_CMA) > + /* > + * Try the different freelists, native then foreign. > + * > + * The fallback logic is expensive and rmqueue_bulk() calls in > + * a loop with the zone->lock held, meaning the freelists are > + * not subject to any outside changes. Remember in *mode where > + * we found pay dirt, to save us the search on the next call. > + */ > + switch (*mode) { > + case RMQUEUE_NORMAL: > + page = __rmqueue_smallest(zone, order, migratetype); > + if (page) > + return page; > + fallthrough; > + case RMQUEUE_CMA: > + if (alloc_flags & ALLOC_CMA) { > page = __rmqueue_cma_fallback(zone, order); > - > - if (!page) > - page = __rmqueue_fallback(zone, order, migratetype, > - alloc_flags); > + if (page) { > + *mode = RMQUEUE_CMA; > + return page; > + } > + } > + fallthrough; > + case RMQUEUE_CLAIM: > + page = __rmqueue_claim(zone, order, migratetype, alloc_flags); > + if (page) { > + /* Replenished native freelist, back to normal mode */ > + *mode = RMQUEUE_NORMAL; > + return page; > + } > + fallthrough; > + case RMQUEUE_STEAL: > + if (!(alloc_flags & ALLOC_NOFRAGMENT)) { > + page = __rmqueue_steal(zone, order, migratetype); > + if (page) { > + *mode = RMQUEUE_STEAL; > + return page; > + } > + } > } > - return page; > + return NULL; > } > > /* > @@ -2311,6 +2357,7 @@ static int rmqueue_bulk(struct zone *zone, unsigned int order, > unsigned long count, struct list_head *list, > int migratetype, unsigned int alloc_flags) > { > + enum rmqueue_mode rmqm = RMQUEUE_NORMAL; > unsigned long flags; > int i; > > @@ -2321,7 +2368,7 @@ static int rmqueue_bulk(struct zone *zone, unsigned int order, > } > for (i = 0; i < count; ++i) { > struct page *page = __rmqueue(zone, order, migratetype, > - alloc_flags); > + alloc_flags, &rmqm); > if (unlikely(page == NULL)) > break; > > @@ -2934,6 +2981,7 @@ struct page *rmqueue_buddy(struct zone *preferred_zone, struct zone *zone, > { > struct page *page; > unsigned long flags; > + enum rmqueue_mode rmqm = RMQUEUE_NORMAL; > > do { > page = NULL; > @@ -2945,7 +2993,7 @@ struct page *rmqueue_buddy(struct zone *preferred_zone, struct zone *zone, > if (alloc_flags & ALLOC_HIGHATOMIC) > page = __rmqueue_smallest(zone, order, MIGRATE_HIGHATOMIC); > if (!page) { > - page = __rmqueue(zone, order, migratetype, alloc_flags); > + page = __rmqueue(zone, order, migratetype, alloc_flags, &rmqm); > > /* > * If the allocation fails, allow OOM handling and > -- > 2.49.0