From nobody Fri Dec 19 04:15:06 2025 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 vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 2B3B1C77B78 for ; Wed, 3 May 2023 02:07:51 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S229567AbjECCGq (ORCPT ); Tue, 2 May 2023 22:06:46 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:38814 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S229482AbjECCGn (ORCPT ); Tue, 2 May 2023 22:06:43 -0400 Received: from NAM10-DM6-obe.outbound.protection.outlook.com (mail-dm6nam10on2059.outbound.protection.outlook.com [40.107.93.59]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 4F22119BE for ; Tue, 2 May 2023 19:06:39 -0700 (PDT) ARC-Seal: i=1; a=rsa-sha256; s=arcselector9901; d=microsoft.com; cv=none; b=HD8HeAy/i88DWSkG6wfyuBbUrB2m3xS5Sti6djynH3AjcdEkZRtjWeYCmFssOwOaacygPxLkAw5rLwd8czC0647UfG+F5/m8YsClKotoq6sqdWOcgNyvNfQ0r/Q/Vid9fi2n5stC4V4AAfR969l/UZZboQn7QHaEQF0EHGLYofUXZAnIEPUdIi3Ez49kqfAzsRzcabpvLtw55yG44USozsCDteiOb8Z/VQZAFoJTnIwmnvqM0co0jqHo/4droTZrD65NB6dEhT4ZHzHdyfATjtdBYSwSzWR8YPOdcf9YbXLYBa305Kl+HjPT0TUnQKDTUaLhiISuV7t/3AV8EPQ1UQ== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=microsoft.com; s=arcselector9901; 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=XRFZzU6RWuHttsW+R6U0rWTSI9QV998SO9eAEjJvhs0=; b=kTQj9uT6vf+Z6YdHxnG5iJSzERfWQgDekahSBh0OjHNws7aIl9s1dejIFs7lj57ZrqjHDr8147GN2cUdAavAHMO6k4M79ZnGZvfRz2I716z/X2x4AcqktL/907+K06lI9kgC4bb6VgAOoPTo4c20Jks9AgZzlqLtO5LbU+6ET25e1HZzouRzh7jHvqv9FjwVZnTQNchCKqbFaAiX9uROvq5/6TQtwoLI8vczysZXwDMy26UEj+D6ACaMjqZgjLndbiccf0OgzEZzN5jMyzz6eBDZjTKWhhvinQw0ZGKXlc+nEN/4DK5AFSJpavhgdJ6mG1F6LMsA6bvftEoH0r8a2w== ARC-Authentication-Results: i=1; mx.microsoft.com 1; spf=pass (sender ip is 165.204.84.17) smtp.rcpttodomain=vger.kernel.org smtp.mailfrom=amd.com; dmarc=pass (p=quarantine sp=quarantine pct=100) action=none header.from=amd.com; dkim=none (message not signed); arc=none DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=amd.com; s=selector1; h=From:Date:Subject:Message-ID:Content-Type:MIME-Version:X-MS-Exchange-SenderADCheck; bh=XRFZzU6RWuHttsW+R6U0rWTSI9QV998SO9eAEjJvhs0=; b=1EQOKnSJ8TQhWcOkcnUGphbfkYzMDGvn6/htBYVdutzbRlQXRRUwYexucICLXcb1rofi+5khDgicK27OQ7mRMkLPcax/ecfjUiVmpRNaznwIL5C4iYkeumdEBPkANpnQde+ei0KiGkcA4CU2Z5IVR0lZQ4o+ijEmwKiWmJrkTeM= Received: from DS7PR03CA0021.namprd03.prod.outlook.com (2603:10b6:5:3b8::26) by LV2PR12MB6014.namprd12.prod.outlook.com (2603:10b6:408:170::18) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.20.6340.24; Wed, 3 May 2023 02:06:37 +0000 Received: from DM6NAM11FT100.eop-nam11.prod.protection.outlook.com (2603:10b6:5:3b8:cafe::a7) by DS7PR03CA0021.outlook.office365.com (2603:10b6:5:3b8::26) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.20.6340.31 via Frontend Transport; Wed, 3 May 2023 02:06:37 +0000 X-MS-Exchange-Authentication-Results: spf=pass (sender IP is 165.204.84.17) smtp.mailfrom=amd.com; dkim=none (message not signed) header.d=none;dmarc=pass action=none header.from=amd.com; Received-SPF: Pass (protection.outlook.com: domain of amd.com designates 165.204.84.17 as permitted sender) receiver=protection.outlook.com; client-ip=165.204.84.17; helo=SATLEXMB04.amd.com; pr=C Received: from SATLEXMB04.amd.com (165.204.84.17) by DM6NAM11FT100.mail.protection.outlook.com (10.13.172.247) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256) id 15.20.6363.21 via Frontend Transport; Wed, 3 May 2023 02:06:36 +0000 Received: from BLR-L-RKODSARA.amd.com (10.180.168.240) by SATLEXMB04.amd.com (10.181.40.145) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256) id 15.1.2375.34; Tue, 2 May 2023 21:06:32 -0500 From: Raghavendra K T To: , CC: Ingo Molnar , Peter Zijlstra , "Mel Gorman" , Andrew Morton , "David Hildenbrand" , , Juri Lelli , Vincent Guittot , Bharata B Rao , Raghavendra K T Subject: [RFC PATCH V1 1/2] sched/numa: Introduce per vma scan counter Date: Wed, 3 May 2023 07:35:48 +0530 Message-ID: X-Mailer: git-send-email 2.25.1 In-Reply-To: References: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Originating-IP: [10.180.168.240] X-ClientProxiedBy: SATLEXMB03.amd.com (10.181.40.144) To SATLEXMB04.amd.com (10.181.40.145) X-EOPAttributedMessage: 0 X-MS-PublicTrafficType: Email X-MS-TrafficTypeDiagnostic: DM6NAM11FT100:EE_|LV2PR12MB6014:EE_ X-MS-Office365-Filtering-Correlation-Id: ca225477-f148-458f-8270-08db4b7b06b9 X-MS-Exchange-SenderADCheck: 1 X-MS-Exchange-AntiSpam-Relay: 0 X-Microsoft-Antispam: BCL:0; X-Microsoft-Antispam-Message-Info: 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 X-Forefront-Antispam-Report: CIP:165.204.84.17;CTRY:US;LANG:en;SCL:1;SRV:;IPV:CAL;SFV:NSPM;H:SATLEXMB04.amd.com;PTR:InfoDomainNonexistent;CAT:NONE;SFS:(13230028)(4636009)(376002)(39860400002)(346002)(136003)(396003)(451199021)(36840700001)(40470700004)(46966006)(81166007)(82740400003)(356005)(7696005)(26005)(6666004)(82310400005)(336012)(83380400001)(47076005)(426003)(70586007)(36860700001)(16526019)(186003)(2616005)(8936002)(36756003)(2906002)(40460700003)(5660300002)(8676002)(478600001)(110136005)(70206006)(54906003)(316002)(7416002)(41300700001)(4326008)(40480700001)(36900700001)(2101003);DIR:OUT;SFP:1101; X-OriginatorOrg: amd.com X-MS-Exchange-CrossTenant-OriginalArrivalTime: 03 May 2023 02:06:36.8157 (UTC) X-MS-Exchange-CrossTenant-Network-Message-Id: ca225477-f148-458f-8270-08db4b7b06b9 X-MS-Exchange-CrossTenant-Id: 3dd8961f-e488-4e60-8e11-a82d994e183d X-MS-Exchange-CrossTenant-OriginalAttributedTenantConnectingIp: TenantId=3dd8961f-e488-4e60-8e11-a82d994e183d;Ip=[165.204.84.17];Helo=[SATLEXMB04.amd.com] X-MS-Exchange-CrossTenant-AuthSource: DM6NAM11FT100.eop-nam11.prod.protection.outlook.com X-MS-Exchange-CrossTenant-AuthAs: Anonymous X-MS-Exchange-CrossTenant-FromEntityHeader: HybridOnPrem X-MS-Exchange-Transport-CrossTenantHeadersStamped: LV2PR12MB6014 Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Type: text/plain; charset="utf-8" With the recent numa scan enhancements, only the tasks which had previously accessed vma are allowed to scan. While this has improved significant system time overhead, there are corner cases, which genuinely needs some relaxation for e.g., concern raised by PeterZ where unfairness amongst the theread belonging to disjoint set of VMSs can potentially amplify the side effects of vma regions belonging to some of the tasks being left unscanned. To address this, allow scanning for first few times with a per vma counter. Signed-off-by: Raghavendra K T --- include/linux/mm_types.h | 1 + kernel/sched/fair.c | 30 +++++++++++++++++++++++++++--- 2 files changed, 28 insertions(+), 3 deletions(-) diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h index 3fc9e680f174..f66e6b4e0620 100644 --- a/include/linux/mm_types.h +++ b/include/linux/mm_types.h @@ -479,6 +479,7 @@ struct vma_numab_state { unsigned long next_scan; unsigned long next_pid_reset; unsigned long access_pids[2]; + unsigned int scan_counter; }; =20 /* diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index a29ca11bead2..3c50dc3893eb 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -2928,19 +2928,38 @@ static void reset_ptenuma_scan(struct task_struct *= p) p->mm->numa_scan_offset =3D 0; } =20 +/* Scan 1GB or 4 * scan_size */ +#define VMA_DISJOINT_SET_ACCESS_THRESH 4U + static bool vma_is_accessed(struct vm_area_struct *vma) { unsigned long pids; + unsigned int windows; + unsigned int scan_size =3D READ_ONCE(sysctl_numa_balancing_scan_size); + + if (scan_size < MAX_SCAN_WINDOW) + windows =3D MAX_SCAN_WINDOW / scan_size; + + /* Allow only half of the windows for disjoint set cases */ + windows /=3D 2; + + windows =3D max(VMA_DISJOINT_SET_ACCESS_THRESH, windows); + /* - * Allow unconditional access first two times, so that all the (pages) - * of VMAs get prot_none fault introduced irrespective of accesses. + * Make sure to allow scanning of disjoint vma set for the first + * few times. + * OR At mm level allow unconditional access first two times, so that + * all the (pages) of VMAs get prot_none fault introduced irrespective + * of accesses. * This is also done to avoid any side effect of task scanning * amplifying the unfairness of disjoint set of VMAs' access. */ - if (READ_ONCE(current->mm->numa_scan_seq) < 2) + if (READ_ONCE(vma->numab_state->scan_counter) < windows || + READ_ONCE(current->mm->numa_scan_seq) < 2) return true; =20 pids =3D vma->numab_state->access_pids[0] | vma->numab_state->access_pids= [1]; + return test_bit(hash_32(current->pid, ilog2(BITS_PER_LONG)), &pids); } =20 @@ -3058,6 +3077,8 @@ static void task_numa_work(struct callback_head *work) /* Reset happens after 4 times scan delay of scan start */ vma->numab_state->next_pid_reset =3D vma->numab_state->next_scan + msecs_to_jiffies(VMA_PID_RESET_PERIOD); + + WRITE_ONCE(vma->numab_state->scan_counter, 0); } =20 /* @@ -3084,6 +3105,9 @@ static void task_numa_work(struct callback_head *work) vma->numab_state->access_pids[1] =3D 0; } =20 + WRITE_ONCE(vma->numab_state->scan_counter, + READ_ONCE(vma->numab_state->scan_counter) + 1); + do { start =3D max(start, vma->vm_start); end =3D ALIGN(start + (pages << PAGE_SHIFT), HPAGE_SIZE); --=20 2.34.1 From nobody Fri Dec 19 04:15:06 2025 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 vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 3D03EC77B75 for ; Wed, 3 May 2023 02:07:51 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S229588AbjECCGu (ORCPT ); Tue, 2 May 2023 22:06:50 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:38840 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S229569AbjECCGr (ORCPT ); Tue, 2 May 2023 22:06:47 -0400 Received: from NAM11-DM6-obe.outbound.protection.outlook.com (mail-dm6nam11on2048.outbound.protection.outlook.com [40.107.223.48]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 00D4D2700 for ; Tue, 2 May 2023 19:06:45 -0700 (PDT) ARC-Seal: i=1; a=rsa-sha256; s=arcselector9901; d=microsoft.com; cv=none; b=Z1KuHjiTQrqEax+O7Oi6Qf+ny/e8rxUr02tfWKch0+rm2tLgxd40QJVG3Xo1lbjwh25IzXw3GI6MORyDJz1+zgnSLONiPKfle1gEj0iwLjhk/Ot88VpY7aSwPylDcx89xkpW3xh62+Osu0QS7PCqfYVjQKxXb/DOAkUg/Uo0RD6LHfO5i7o8S10JcchcAtS7GZYSBDLx2cOcBmTOAJcFNiNwRdrGjtWUyNJiVPmQEBEESAzAciPSQk1GypjXYC2vfABIoB1huus9UJ3aFc+m3EVAZhulcC8Oss2XfZ9Y4Nk8sg2yTD0LdSB7++KhoNu9y14DJpVHbOcYPEADSujypg== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=microsoft.com; s=arcselector9901; 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=/NSBlCz8xCMXbjCQo62cKyUj6pVFVaXwQU03xdm2Tgw=; b=a+3pb+bhcBSbo6PS9G1eqM2U/+Gg+1phV3cZv/0GaoMOJXUs8c3WM9vV8xUb+wWM0QYkMDS0MUDA9FBVBJzgDUtSHwB7I9I/arrOpjyfkoqZ4L2BMUqWAFX/DMtC79ASPcXhBDRNi9ESPbEt1WAnaU4w9sV3RAjtx3QF6C20Kwa4USVJ69cCeRvqx2zWW7vgZLdGY3Y1ZS1OGpKjswaVp1JKhpiiw9YkQz2EfJTKqSEcypqfSXFtdkCAHj9CMwXn9ZcMmi2IMGylIXR3SnxWRAI32IJmoMjXRQGk6VgdIVP5P9U1/flyHLK7spyApyilugeuA6Lkz5TmhaxITQiBOg== ARC-Authentication-Results: i=1; mx.microsoft.com 1; spf=pass (sender ip is 165.204.84.17) smtp.rcpttodomain=vger.kernel.org smtp.mailfrom=amd.com; dmarc=pass (p=quarantine sp=quarantine pct=100) action=none header.from=amd.com; dkim=none (message not signed); arc=none DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=amd.com; s=selector1; h=From:Date:Subject:Message-ID:Content-Type:MIME-Version:X-MS-Exchange-SenderADCheck; bh=/NSBlCz8xCMXbjCQo62cKyUj6pVFVaXwQU03xdm2Tgw=; b=Zy/YQ+9l7CXzGRtV9waTkX3XOILR3EOBYLbbyVNz2WFmOqAcleJErve6IS6VEmFtEEPaMHX87BAlGMOhRPCoPTyjES9BXVBTbwsT8UWo5bqGmeosndfkP0dIRNh9WSVV70XapfpS5ir6ZR4PbHUSHpMu7xLRjzAclXIESadsGkc= Received: from DS7PR03CA0321.namprd03.prod.outlook.com (2603:10b6:8:2b::6) by PH7PR12MB7212.namprd12.prod.outlook.com (2603:10b6:510:207::18) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.20.6340.27; Wed, 3 May 2023 02:06:41 +0000 Received: from DM6NAM11FT009.eop-nam11.prod.protection.outlook.com (2603:10b6:8:2b:cafe::e5) by DS7PR03CA0321.outlook.office365.com (2603:10b6:8:2b::6) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.20.6363.22 via Frontend Transport; Wed, 3 May 2023 02:06:40 +0000 X-MS-Exchange-Authentication-Results: spf=pass (sender IP is 165.204.84.17) smtp.mailfrom=amd.com; dkim=none (message not signed) header.d=none;dmarc=pass action=none header.from=amd.com; Received-SPF: Pass (protection.outlook.com: domain of amd.com designates 165.204.84.17 as permitted sender) receiver=protection.outlook.com; client-ip=165.204.84.17; helo=SATLEXMB04.amd.com; pr=C Received: from SATLEXMB04.amd.com (165.204.84.17) by DM6NAM11FT009.mail.protection.outlook.com (10.13.173.20) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256) id 15.20.6363.20 via Frontend Transport; Wed, 3 May 2023 02:06:40 +0000 Received: from BLR-L-RKODSARA.amd.com (10.180.168.240) by SATLEXMB04.amd.com (10.181.40.145) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256) id 15.1.2375.34; Tue, 2 May 2023 21:06:36 -0500 From: Raghavendra K T To: , CC: Ingo Molnar , Peter Zijlstra , "Mel Gorman" , Andrew Morton , "David Hildenbrand" , , Juri Lelli , Vincent Guittot , Bharata B Rao , Raghavendra K T Subject: [RFC PATCH V1 2/2] sched/numa: Introduce per vma numa_scan_seq Date: Wed, 3 May 2023 07:35:49 +0530 Message-ID: X-Mailer: git-send-email 2.25.1 In-Reply-To: References: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Originating-IP: [10.180.168.240] X-ClientProxiedBy: SATLEXMB03.amd.com (10.181.40.144) To SATLEXMB04.amd.com (10.181.40.145) X-EOPAttributedMessage: 0 X-MS-PublicTrafficType: Email X-MS-TrafficTypeDiagnostic: DM6NAM11FT009:EE_|PH7PR12MB7212:EE_ X-MS-Office365-Filtering-Correlation-Id: 99af254d-92aa-494e-3ffa-08db4b7b091a X-MS-Exchange-SenderADCheck: 1 X-MS-Exchange-AntiSpam-Relay: 0 X-Microsoft-Antispam: BCL:0; X-Microsoft-Antispam-Message-Info: 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 X-Forefront-Antispam-Report: CIP:165.204.84.17;CTRY:US;LANG:en;SCL:1;SRV:;IPV:CAL;SFV:NSPM;H:SATLEXMB04.amd.com;PTR:InfoDomainNonexistent;CAT:NONE;SFS:(13230028)(4636009)(136003)(39860400002)(376002)(396003)(346002)(451199021)(36840700001)(46966006)(40470700004)(4326008)(82310400005)(26005)(40460700003)(40480700001)(316002)(7416002)(8936002)(8676002)(16526019)(5660300002)(478600001)(36860700001)(186003)(70206006)(54906003)(70586007)(336012)(47076005)(426003)(83380400001)(41300700001)(110136005)(2616005)(356005)(36756003)(81166007)(2906002)(7696005)(6666004)(82740400003)(2101003)(36900700001);DIR:OUT;SFP:1101; X-OriginatorOrg: amd.com X-MS-Exchange-CrossTenant-OriginalArrivalTime: 03 May 2023 02:06:40.8051 (UTC) X-MS-Exchange-CrossTenant-Network-Message-Id: 99af254d-92aa-494e-3ffa-08db4b7b091a X-MS-Exchange-CrossTenant-Id: 3dd8961f-e488-4e60-8e11-a82d994e183d X-MS-Exchange-CrossTenant-OriginalAttributedTenantConnectingIp: TenantId=3dd8961f-e488-4e60-8e11-a82d994e183d;Ip=[165.204.84.17];Helo=[SATLEXMB04.amd.com] X-MS-Exchange-CrossTenant-AuthSource: DM6NAM11FT009.eop-nam11.prod.protection.outlook.com X-MS-Exchange-CrossTenant-AuthAs: Anonymous X-MS-Exchange-CrossTenant-FromEntityHeader: HybridOnPrem X-MS-Exchange-Transport-CrossTenantHeadersStamped: PH7PR12MB7212 Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Type: text/plain; charset="utf-8" Per vma scan counter was introduced to aid disjoint set vma scanning in corner cases. But that counter needs reset regularly. Reset is achieved after full round of mm scanning by per vma numa_scan_sequence that follows mm->numa_scan_seq. Result: With this patch series we recover mmtest's numa01_THREAD_ALLOC performance as below Base 11-apr-next w/numascan w/o numascan numascan+patch real 1m33.579s 1m2.042s 1m11.738s user 280m46.032s 213m38.647s 231m40.226s sys 0m18.061s 6m54.963s 4m43.174s In summary: it adds back some system overhaed of scanning disjoint vma scanning, But still we are at huge advantage w.r.t base kernel Signed-off-by: Raghavendra K T --- include/linux/mm_types.h | 1 + kernel/sched/fair.c | 18 ++++++++++++++++-- 2 files changed, 17 insertions(+), 2 deletions(-) diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h index f66e6b4e0620..9c0fc83118da 100644 --- a/include/linux/mm_types.h +++ b/include/linux/mm_types.h @@ -479,6 +479,7 @@ struct vma_numab_state { unsigned long next_scan; unsigned long next_pid_reset; unsigned long access_pids[2]; + unsigned int vma_scan_seq; unsigned int scan_counter; }; =20 diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 3c50dc3893eb..dc011a2a31ac 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -2935,6 +2935,7 @@ static bool vma_is_accessed(struct vm_area_struct *vm= a) { unsigned long pids; unsigned int windows; + unsigned int mm_seq, vma_seq; unsigned int scan_size =3D READ_ONCE(sysctl_numa_balancing_scan_size); =20 if (scan_size < MAX_SCAN_WINDOW) @@ -2945,6 +2946,18 @@ static bool vma_is_accessed(struct vm_area_struct *v= ma) =20 windows =3D max(VMA_DISJOINT_SET_ACCESS_THRESH, windows); =20 + mm_seq =3D READ_ONCE(current->mm->numa_scan_seq); + vma_seq =3D READ_ONCE(vma->numab_state->vma_scan_seq); + + if (vma_seq !=3D mm_seq) { + /* + * One more round of whole mm scan was done. Reset the vma scan_counter + * and sync per vma numa_scan_seq. + */ + WRITE_ONCE(vma->numab_state->vma_scan_seq, + READ_ONCE(current->mm->numa_scan_seq)); + WRITE_ONCE(vma->numab_state->scan_counter, 0); + } /* * Make sure to allow scanning of disjoint vma set for the first * few times. @@ -2954,8 +2967,7 @@ static bool vma_is_accessed(struct vm_area_struct *vm= a) * This is also done to avoid any side effect of task scanning * amplifying the unfairness of disjoint set of VMAs' access. */ - if (READ_ONCE(vma->numab_state->scan_counter) < windows || - READ_ONCE(current->mm->numa_scan_seq) < 2) + if (READ_ONCE(vma->numab_state->scan_counter) < windows || mm_seq < 2) return true; =20 pids =3D vma->numab_state->access_pids[0] | vma->numab_state->access_pids= [1]; @@ -3078,6 +3090,8 @@ static void task_numa_work(struct callback_head *work) vma->numab_state->next_pid_reset =3D vma->numab_state->next_scan + msecs_to_jiffies(VMA_PID_RESET_PERIOD); =20 + WRITE_ONCE(vma->numab_state->vma_scan_seq, + READ_ONCE(current->mm->numa_scan_seq)); WRITE_ONCE(vma->numab_state->scan_counter, 0); } =20 --=20 2.34.1