Patch Detail
get:
Show a patch.
patch:
Update a patch.
put:
Update a patch.
GET /api/patches/131144/?format=api
{ "id": 131144, "url": "http://patchwork.dpdk.org/api/patches/131144/?format=api", "web_url": "http://patchwork.dpdk.org/project/dpdk/patch/20230904082455.3864024-1-artemyko@nvidia.com/", "project": { "id": 1, "url": "http://patchwork.dpdk.org/api/projects/1/?format=api", "name": "DPDK", "link_name": "dpdk", "list_id": "dev.dpdk.org", "list_email": "dev@dpdk.org", "web_url": "http://core.dpdk.org", "scm_url": "git://dpdk.org/dpdk", "webscm_url": "http://git.dpdk.org/dpdk", "list_archive_url": "https://inbox.dpdk.org/dev", "list_archive_url_format": "https://inbox.dpdk.org/dev/{}", "commit_url_format": "" }, "msgid": "<20230904082455.3864024-1-artemyko@nvidia.com>", "list_archive_url": "https://inbox.dpdk.org/dev/20230904082455.3864024-1-artemyko@nvidia.com", "date": "2023-09-04T08:24:54", "name": "[v2] eal: fix memory initialization deadlock", "commit_ref": null, "pull_url": null, "state": "superseded", "archived": true, "hash": "c9ca7d7bc422f7f886ac7bbc9e259674107b3e7e", "submitter": { "id": 3172, "url": "http://patchwork.dpdk.org/api/people/3172/?format=api", "name": "Artemy Kovalyov", "email": "artemyko@nvidia.com" }, "delegate": { "id": 1, "url": "http://patchwork.dpdk.org/api/users/1/?format=api", "username": "tmonjalo", "first_name": "Thomas", "last_name": "Monjalon", "email": "thomas@monjalon.net" }, "mbox": "http://patchwork.dpdk.org/project/dpdk/patch/20230904082455.3864024-1-artemyko@nvidia.com/mbox/", "series": [ { "id": 29414, "url": "http://patchwork.dpdk.org/api/series/29414/?format=api", "web_url": "http://patchwork.dpdk.org/project/dpdk/list/?series=29414", "date": "2023-09-04T08:24:54", "name": "[v2] eal: fix memory initialization deadlock", "version": 2, "mbox": "http://patchwork.dpdk.org/series/29414/mbox/" } ], "comments": "http://patchwork.dpdk.org/api/patches/131144/comments/", "check": "fail", "checks": "http://patchwork.dpdk.org/api/patches/131144/checks/", "tags": {}, "related": [], "headers": { "Return-Path": "<dev-bounces@dpdk.org>", "X-Original-To": "patchwork@inbox.dpdk.org", "Delivered-To": "patchwork@inbox.dpdk.org", "Received": [ "from mails.dpdk.org (mails.dpdk.org [217.70.189.124])\n\tby inbox.dpdk.org (Postfix) with ESMTP id D5761424EE;\n\tMon, 4 Sep 2023 10:25:25 +0200 (CEST)", "from mails.dpdk.org (localhost [127.0.0.1])\n\tby mails.dpdk.org (Postfix) with ESMTP id 6A1B0402AF;\n\tMon, 4 Sep 2023 10:25:25 +0200 (CEST)", "from NAM02-SN1-obe.outbound.protection.outlook.com\n (mail-sn1nam02on2041.outbound.protection.outlook.com [40.107.96.41])\n by mails.dpdk.org (Postfix) with ESMTP id A0C76400EF;\n Mon, 4 Sep 2023 10:25:23 +0200 (CEST)", "from SN7PR18CA0005.namprd18.prod.outlook.com (2603:10b6:806:f3::24)\n by CH3PR12MB7763.namprd12.prod.outlook.com (2603:10b6:610:145::10)\n with Microsoft SMTP Server (version=TLS1_2,\n cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.20.6699.35; Mon, 4 Sep\n 2023 08:25:20 +0000", "from SA2PEPF00001509.namprd04.prod.outlook.com\n (2603:10b6:806:f3:cafe::3c) by SN7PR18CA0005.outlook.office365.com\n (2603:10b6:806:f3::24) with Microsoft SMTP Server (version=TLS1_2,\n cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.20.6745.33 via Frontend\n Transport; Mon, 4 Sep 2023 08:25:20 +0000", "from mail.nvidia.com (216.228.117.161) by\n SA2PEPF00001509.mail.protection.outlook.com (10.167.242.41) with Microsoft\n SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id\n 15.20.6768.25 via Frontend Transport; Mon, 4 Sep 2023 08:25:20 +0000", "from rnnvmail205.nvidia.com (10.129.68.10) by mail.nvidia.com\n (10.129.200.67) with Microsoft SMTP Server (version=TLS1_2,\n cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.2.986.5; Mon, 4 Sep 2023\n 01:25:07 -0700", "from rnnvmail202.nvidia.com (10.129.68.7) by rnnvmail205.nvidia.com\n (10.129.68.10) with Microsoft SMTP Server (version=TLS1_2,\n cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.2.986.37; Mon, 4 Sep 2023\n 01:25:07 -0700", "from nvidia.com (10.127.8.9) by mail.nvidia.com (10.129.68.7) with\n Microsoft SMTP Server (version=TLS1_2,\n cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.2.986.37 via Frontend\n Transport; Mon, 4 Sep 2023 01:25:05 -0700" ], "ARC-Seal": "i=1; a=rsa-sha256; s=arcselector9901; d=microsoft.com; cv=none;\n b=CapAvqgK9zwcTwSfAcfp1bl6bXuoOKhpYaN1twbo+euwqsewHyzRhvERmYXrWTu7lNVOA8olP+v7JdDQoeX3Pc44a15KeCNboOIG531hPB5rLEuUz9nTofS0nLd9pC/2oArcxOwx1EK5uWIqHB4pKIAIQJFjzzG/E6+DyJeWIEV1pEJlT0WoBn6wVE7Jo1E5jojLZ7eJcxlk0tKaXRF0JmFhW4cNyBkcOPQ0pRZm28ZbucJtAvxK+GfsA7+TJ23AIZ1iK6dEgTkdqB0b2kLghGT5m7wadtF8ZDlCAwm4GWZuc8Se8UH30ax0g73XPQPisdd3zc7JZqtM9Th90RGtRg==", "ARC-Message-Signature": "i=1; a=rsa-sha256; c=relaxed/relaxed; d=microsoft.com;\n s=arcselector9901;\n 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;\n bh=lsrQZZcPKst8RJgvHzIJ7i8p7gzqgSZCUmCDi5xuKqk=;\n b=BdHjGiIWCK5Q+JExr8b32s2GA7VqXuMaRwDLyvibQ/MAxLL7CZIiqcbp4E7uWiivjCLxiPvRrOHdPVYWPfUNrqj98pyqSrdwY2fM52WnATDM3PvZPIufgvPavY6ORTfJ3PJCE96ziE2TNV5d9qSMzvY2ssfpGb9x3FpnROExy3EFS7a7eY9rKlwOLAI5mX9FpedD7m62vVKLKtONPW3MM1TYNSD8OvO8fExdQQWvmib4gp7laG2rWTYBs9XDBZblfMmBNUie5yDjuzAXRnQcuWXGx6xwBA470gG/iqaryYvRTEmJsPaUGNF85bUep1FapHBN0ECCHVDawtEYo1hLig==", "ARC-Authentication-Results": "i=1; mx.microsoft.com 1; spf=pass (sender ip is\n 216.228.117.161) smtp.rcpttodomain=dpdk.org smtp.mailfrom=nvidia.com;\n dmarc=pass (p=reject sp=reject pct=100) action=none header.from=nvidia.com;\n dkim=none (message not signed); arc=none", "DKIM-Signature": "v=1; a=rsa-sha256; c=relaxed/relaxed; d=Nvidia.com;\n s=selector2;\n h=From:Date:Subject:Message-ID:Content-Type:MIME-Version:X-MS-Exchange-SenderADCheck;\n bh=lsrQZZcPKst8RJgvHzIJ7i8p7gzqgSZCUmCDi5xuKqk=;\n b=mTrcEH/GH3zpoX3krfHun5CnR0jyj3jPFmRQlCdlhQd+yJY4/MhQwwv0Lkae8tMIjMTEkT8aeVUwGzJcsu8AssxHOCrHzXFT2UDUBYg3cR3bOWEjtfOQfgyqC8geZvxZD0LM64ngSEC89zxpLBfyzxs97/Uv6gtxh+5FOcvtpq2RVHVx07ZnOcV404inqUhpRaKFWpmKBHF0Scr1+LOXHYan0MOeMRGaYMlQxwPrj5ZUGbrre4GVqpJmKO0pkuDtw6hHDb1qFktGWLcRGw5TfA1GZN2Lb5EhJ0e2rzYjxHCpU44ULT6ZjDlJgW+XHoc/P2P4wKOAWnm+9+RS5WhzFg==", "X-MS-Exchange-Authentication-Results": "spf=pass (sender IP is 216.228.117.161)\n smtp.mailfrom=nvidia.com;\n dkim=none (message not signed)\n header.d=none;dmarc=pass action=none header.from=nvidia.com;", "Received-SPF": "Pass (protection.outlook.com: domain of nvidia.com designates\n 216.228.117.161 as permitted sender) receiver=protection.outlook.com;\n client-ip=216.228.117.161; helo=mail.nvidia.com; pr=C", "From": "Artemy Kovalyov <artemyko@nvidia.com>", "To": "<dev@dpdk.org>", "CC": "Thomas Monjalon <thomas@monjalon.net>, Ophir Munk <ophirmu@nvidia.com>,\n <stable@dpdk.org>, Anatoly Burakov <anatoly.burakov@intel.com>, =?utf-8?q?M?=\n\t=?utf-8?q?orten_Br=C3=B8rup?= <mb@smartsharesystems.com>,\n \"Stephen Hemminger\" <stephen@networkplumber.org>", "Subject": "[PATCH v2] eal: fix memory initialization deadlock", "Date": "Mon, 4 Sep 2023 11:24:54 +0300", "Message-ID": "<20230904082455.3864024-1-artemyko@nvidia.com>", "X-Mailer": "git-send-email 2.25.1", "In-Reply-To": "<20230830103303.2428995-1-artemyko@nvidia.com>", "References": "<20230830103303.2428995-1-artemyko@nvidia.com>", "MIME-Version": "1.0", "Content-Transfer-Encoding": "8bit", "Content-Type": "text/plain", "X-NV-OnPremToCloud": "ExternallySecured", "X-EOPAttributedMessage": "0", "X-MS-PublicTrafficType": "Email", "X-MS-TrafficTypeDiagnostic": "SA2PEPF00001509:EE_|CH3PR12MB7763:EE_", "X-MS-Office365-Filtering-Correlation-Id": "9fa25b35-8e32-4108-61ce-08dbad207a3c", "X-LD-Processed": "43083d15-7273-40c1-b7db-39efd9ccc17a,ExtAddr", "X-MS-Exchange-SenderADCheck": "1", "X-MS-Exchange-AntiSpam-Relay": "0", "X-Microsoft-Antispam": "BCL:0;", "X-Microsoft-Antispam-Message-Info": "\n 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", "X-Forefront-Antispam-Report": "CIP:216.228.117.161; CTRY:US; LANG:en; SCL:1;\n SRV:;\n IPV:NLI; SFV:NSPM; H:mail.nvidia.com; PTR:dc6edge2.nvidia.com; CAT:NONE;\n SFS:(13230031)(4636009)(39860400002)(396003)(376002)(136003)(346002)(82310400011)(451199024)(186009)(1800799009)(46966006)(40470700004)(36840700001)(426003)(40460700003)(316002)(41300700001)(5660300002)(336012)(6286002)(6916009)(2616005)(83380400001)(26005)(86362001)(36756003)(8676002)(55016003)(2906002)(36860700001)(47076005)(4326008)(1076003)(40480700001)(8936002)(6666004)(7636003)(356005)(82740400003)(478600001)(7696005)(54906003)(70586007)(70206006);\n DIR:OUT; SFP:1101;", "X-OriginatorOrg": "Nvidia.com", "X-MS-Exchange-CrossTenant-OriginalArrivalTime": "04 Sep 2023 08:25:20.2987 (UTC)", "X-MS-Exchange-CrossTenant-Network-Message-Id": "\n 9fa25b35-8e32-4108-61ce-08dbad207a3c", "X-MS-Exchange-CrossTenant-Id": "43083d15-7273-40c1-b7db-39efd9ccc17a", "X-MS-Exchange-CrossTenant-OriginalAttributedTenantConnectingIp": "\n TenantId=43083d15-7273-40c1-b7db-39efd9ccc17a; Ip=[216.228.117.161];\n Helo=[mail.nvidia.com]", "X-MS-Exchange-CrossTenant-AuthSource": "\n SA2PEPF00001509.namprd04.prod.outlook.com", "X-MS-Exchange-CrossTenant-AuthAs": "Anonymous", "X-MS-Exchange-CrossTenant-FromEntityHeader": "HybridOnPrem", "X-MS-Exchange-Transport-CrossTenantHeadersStamped": "CH3PR12MB7763", "X-BeenThere": "dev@dpdk.org", "X-Mailman-Version": "2.1.29", "Precedence": "list", "List-Id": "DPDK patches and discussions <dev.dpdk.org>", "List-Unsubscribe": "<https://mails.dpdk.org/options/dev>,\n <mailto:dev-request@dpdk.org?subject=unsubscribe>", "List-Archive": "<http://mails.dpdk.org/archives/dev/>", "List-Post": "<mailto:dev@dpdk.org>", "List-Help": "<mailto:dev-request@dpdk.org?subject=help>", "List-Subscribe": "<https://mails.dpdk.org/listinfo/dev>,\n <mailto:dev-request@dpdk.org?subject=subscribe>", "Errors-To": "dev-bounces@dpdk.org" }, "content": "The issue arose due to the change in the DPDK read-write lock\nimplementation. That change added a new flag, RTE_RWLOCK_WAIT, designed\nto prevent new read locks while a write lock is in the queue. However,\nthis change has led to a scenario where a recursive read lock, where a\nlock is acquired twice by the same execution thread, can initiate a\nsequence of events resulting in a deadlock:\n\nProcess 1 takes the first read lock.\nProcess 2 attempts to take a write lock, triggering RTE_RWLOCK_WAIT due\nto the presence of a read lock. This makes process 2 enter a wait loop\nuntil the read lock is released.\nProcess 1 tries to take a second read lock. However, since a write lock\nis waiting (due to RTE_RWLOCK_WAIT), it also enters a wait loop until\nthe write lock is acquired and then released.\n\nBoth processes end up in a blocked state, unable to proceed, resulting\nin a deadlock scenario.\n\nFollowing these changes, the RW-lock no longer supports\nrecursion, implying that a single thread shouldn't obtain a read lock if\nit already possesses one. The problem arises during initialization: the\nrte_eal_init() function acquires the memory_hotplug_lock, and later on,\nthe sequence of calls rte_eal_memory_init() -> eal_memalloc_init() ->\nrte_memseg_list_walk() acquires it again without releasing it. This\nscenario introduces the risk of a potential deadlock when concurrent\nwrite locks are applied to the same memory_hotplug_lock. To address this\nwe resolved the issue by replacing rte_memseg_list_walk() with\nrte_memseg_list_walk_thread_unsafe().\n\nBugzilla ID: 1277\nFixes: 832cecc03d77 (\"rwlock: prevent readers from starving writers\")\nCc: stable@dpdk.org\n\nSigned-off-by: Artemy Kovalyov <artemyko@nvidia.com>\n---\nv2:\nchanged walk to thread-unsafe version in eal_dynmem_hugepage_init() 32-bit flow\n---\n lib/eal/common/eal_common_dynmem.c | 5 ++++-\n lib/eal/include/generic/rte_rwlock.h | 4 ++++\n lib/eal/linux/eal_memalloc.c | 7 +++++--\n 3 files changed, 13 insertions(+), 3 deletions(-)", "diff": "diff --git a/lib/eal/common/eal_common_dynmem.c b/lib/eal/common/eal_common_dynmem.c\nindex bdbbe233a0..0d5da40096 100644\n--- a/lib/eal/common/eal_common_dynmem.c\n+++ b/lib/eal/common/eal_common_dynmem.c\n@@ -251,7 +251,10 @@ eal_dynmem_hugepage_init(void)\n \t\t */\n \t\tmemset(&dummy, 0, sizeof(dummy));\n \t\tdummy.hugepage_sz = hpi->hugepage_sz;\n-\t\tif (rte_memseg_list_walk(hugepage_count_walk, &dummy) < 0)\n+\t\t/* memory_hotplug_lock is taken in rte_eal_init(), so it's\n+\t\t * safe to call thread-unsafe version.\n+\t\t */\n+\t\tif (rte_memseg_list_walk_thread_unsafe(hugepage_count_walk, &dummy) < 0)\n \t\t\treturn -1;\n \n \t\tfor (i = 0; i < RTE_DIM(dummy.num_pages); i++) {\ndiff --git a/lib/eal/include/generic/rte_rwlock.h b/lib/eal/include/generic/rte_rwlock.h\nindex 9e083bbc61..c98fc7d083 100644\n--- a/lib/eal/include/generic/rte_rwlock.h\n+++ b/lib/eal/include/generic/rte_rwlock.h\n@@ -80,6 +80,10 @@ rte_rwlock_init(rte_rwlock_t *rwl)\n /**\n * Take a read lock. Loop until the lock is held.\n *\n+ * @note The RW lock isn't recursive, so calling this function on the same\n+ * lock twice without releasing it could potentially result in a deadlock\n+ * scenario when a write lock is involved.\n+ *\n * @param rwl\n * A pointer to a rwlock structure.\n */\ndiff --git a/lib/eal/linux/eal_memalloc.c b/lib/eal/linux/eal_memalloc.c\nindex f8b1588cae..3705b41f5f 100644\n--- a/lib/eal/linux/eal_memalloc.c\n+++ b/lib/eal/linux/eal_memalloc.c\n@@ -1740,7 +1740,10 @@ eal_memalloc_init(void)\n \t\teal_get_internal_configuration();\n \n \tif (rte_eal_process_type() == RTE_PROC_SECONDARY)\n-\t\tif (rte_memseg_list_walk(secondary_msl_create_walk, NULL) < 0)\n+\t\t/* memory_hotplug_lock is taken in rte_eal_init(), so it's\n+\t\t * safe to call thread-unsafe version.\n+\t\t */\n+\t\tif (rte_memseg_list_walk_thread_unsafe(secondary_msl_create_walk, NULL) < 0)\n \t\t\treturn -1;\n \tif (rte_eal_process_type() == RTE_PROC_PRIMARY &&\n \t\t\tinternal_conf->in_memory) {\n@@ -1778,7 +1781,7 @@ eal_memalloc_init(void)\n \t}\n \n \t/* initialize all of the fd lists */\n-\tif (rte_memseg_list_walk(fd_list_create_walk, NULL))\n+\tif (rte_memseg_list_walk_thread_unsafe(fd_list_create_walk, NULL))\n \t\treturn -1;\n \treturn 0;\n }\n", "prefixes": [ "v2" ] }