From patchwork Tue May 30 15:13:28 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Slava Ovsiienko X-Patchwork-Id: 127728 X-Patchwork-Delegate: rasland@nvidia.com Return-Path: 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]) by inbox.dpdk.org (Postfix) with ESMTP id 4AB8B42BE6; Tue, 30 May 2023 17:14:03 +0200 (CEST) Received: from mails.dpdk.org (localhost [127.0.0.1]) by mails.dpdk.org (Postfix) with ESMTP id 213BF410EE; Tue, 30 May 2023 17:14:03 +0200 (CEST) Received: from NAM12-DM6-obe.outbound.protection.outlook.com (mail-dm6nam12on2040.outbound.protection.outlook.com [40.107.243.40]) by mails.dpdk.org (Postfix) with ESMTP id 138CC410DD; Tue, 30 May 2023 17:14:02 +0200 (CEST) ARC-Seal: i=1; a=rsa-sha256; s=arcselector9901; d=microsoft.com; cv=none; b=BGpqJFSB43I9ZZrJfxW54mEBhfjypQrS1mw0kC+gJb2xf4us+AQkd20Pu80xzQ881ChdiMLa21DUbRA9ZCMBzPPeOrh1szgZY1Fn5IwDcLlYJacggl7BpuC2fdvAkcTkyeab+AcZJXsJWJvCBXLVHuoJ94siCP2GT3x9y7RUIj7o+ZMiFyv3lNeFvszOJDItJDYzMxZHbL7F9Z21zEWsgMuy6IwEYY+IeM0POLuGe1cz7050QlePPFyRP2rVbR0dJJbXphNe5fBi4zSGBet2vAECaw2j4J8WsPb830X4GYcyc4YjVTllD7OiwH4PtnD0pbik9jldT/EGRvGjJhxZCQ== 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=HwGT6959ozBROXk5P4Gnl5tD587a4a8qP+hB2JStP/s=; b=KESqG2dnjD7rUR+IZlU+byKz+OwNYd1so9bTT8QnDi5qamU2t3rl/s41APS9CUXE8QwzT7abAnc2HZA528PUyqZE/yIGX6hf+o6I8tTeukHzl1/N7+slbgcEOEq28N7sap0SipGNZFNBtSVn/9GPRPIFFjS9wDg1laSudWsbIz35PiwtEWH68nAcC293Rex3IFq8KhV5AH7vkKtVaz8/Rcl8QDlUrsmY6bQxa5Mytx0tooDSlEJvG2FVU0XyvLbn+3aEcrp2hC4HyNBdNWNHzFbDsV11CRAa9oXajFKcORLyuOWsS8Ve3MZ2EFgKnwwlC4fIE3x3QUxyVreqMJY4oA== ARC-Authentication-Results: i=1; mx.microsoft.com 1; spf=pass (sender ip is 216.228.117.161) smtp.rcpttodomain=dpdk.org smtp.mailfrom=nvidia.com; dmarc=pass (p=reject sp=reject pct=100) action=none header.from=nvidia.com; dkim=none (message not signed); arc=none DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=Nvidia.com; s=selector2; h=From:Date:Subject:Message-ID:Content-Type:MIME-Version:X-MS-Exchange-SenderADCheck; bh=HwGT6959ozBROXk5P4Gnl5tD587a4a8qP+hB2JStP/s=; b=mXQBo3FTPwtY1bXfvRhPcbr2XDyitSn45ocO8/hRMhGI+zHqTQ6CJCYIzFZNIXQMKMufiRGim8RD97U8UsQnJwzRYBHkbnXwtusOLC8Ux9D3DlvOXi4UfGyS+IoGfSk9neuEI2mlmhmz/EdbVArPn4LO0SUTtSKIXXa1sTu9nfEGWLghqcjqNJYa/eYXU+01sNYwD6ruA77sqajnYX+fS0ncXvpk00buhxYAa4XNQ+uBFEGKajHm9o2mVBxlAO8iw5/XOabSTaQBVFaawSqvTUNYEBkUME6Tlc6siwuHmz7NHaOSITzXlmBFDutbQ/eUIp09fSt+2suyq7/ByA/FMw== Received: from MW4PR04CA0357.namprd04.prod.outlook.com (2603:10b6:303:8a::32) by PH7PR12MB6834.namprd12.prod.outlook.com (2603:10b6:510:1b4::18) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.20.6433.21; Tue, 30 May 2023 15:13:59 +0000 Received: from CO1NAM11FT112.eop-nam11.prod.protection.outlook.com (2603:10b6:303:8a:cafe::78) by MW4PR04CA0357.outlook.office365.com (2603:10b6:303:8a::32) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.20.6433.23 via Frontend Transport; Tue, 30 May 2023 15:13:59 +0000 X-MS-Exchange-Authentication-Results: spf=pass (sender IP is 216.228.117.161) smtp.mailfrom=nvidia.com; dkim=none (message not signed) header.d=none;dmarc=pass action=none header.from=nvidia.com; Received-SPF: Pass (protection.outlook.com: domain of nvidia.com designates 216.228.117.161 as permitted sender) receiver=protection.outlook.com; client-ip=216.228.117.161; helo=mail.nvidia.com; pr=C Received: from mail.nvidia.com (216.228.117.161) by CO1NAM11FT112.mail.protection.outlook.com (10.13.174.213) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.20.6455.22 via Frontend Transport; Tue, 30 May 2023 15:13:58 +0000 Received: from rnnvmail201.nvidia.com (10.129.68.8) by mail.nvidia.com (10.129.200.67) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.2.986.5; Tue, 30 May 2023 08:13:43 -0700 Received: from nvidia.com (10.126.230.35) by rnnvmail201.nvidia.com (10.129.68.8) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.2.986.37; Tue, 30 May 2023 08:13:42 -0700 From: Viacheslav Ovsiienko To: CC: , , , Subject: [PATCH 1/1] net/mlx5: fix device removal event handling Date: Tue, 30 May 2023 18:13:28 +0300 Message-ID: <20230530151328.29738-1-viacheslavo@nvidia.com> X-Mailer: git-send-email 2.18.1 MIME-Version: 1.0 X-Originating-IP: [10.126.230.35] X-ClientProxiedBy: rnnvmail203.nvidia.com (10.129.68.9) To rnnvmail201.nvidia.com (10.129.68.8) X-EOPAttributedMessage: 0 X-MS-PublicTrafficType: Email X-MS-TrafficTypeDiagnostic: CO1NAM11FT112:EE_|PH7PR12MB6834:EE_ X-MS-Office365-Filtering-Correlation-Id: 8e275505-aafe-4639-8e72-08db61207e6c 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:216.228.117.161; CTRY:US; LANG:en; SCL:1; SRV:; IPV:NLI; SFV:NSPM; H:mail.nvidia.com; PTR:dc6edge2.nvidia.com; CAT:NONE; SFS:(13230028)(4636009)(136003)(396003)(39860400002)(346002)(376002)(451199021)(40470700004)(36840700001)(46966006)(54906003)(5660300002)(70206006)(86362001)(70586007)(450100002)(478600001)(8936002)(8676002)(1076003)(26005)(7696005)(82310400005)(6666004)(316002)(4326008)(41300700001)(6916009)(186003)(6286002)(16526019)(40460700003)(83380400001)(426003)(336012)(2906002)(47076005)(2616005)(36860700001)(55016003)(36756003)(40480700001)(7636003)(82740400003)(356005); DIR:OUT; SFP:1101; X-OriginatorOrg: Nvidia.com X-MS-Exchange-CrossTenant-OriginalArrivalTime: 30 May 2023 15:13:58.9386 (UTC) X-MS-Exchange-CrossTenant-Network-Message-Id: 8e275505-aafe-4639-8e72-08db61207e6c X-MS-Exchange-CrossTenant-Id: 43083d15-7273-40c1-b7db-39efd9ccc17a X-MS-Exchange-CrossTenant-OriginalAttributedTenantConnectingIp: TenantId=43083d15-7273-40c1-b7db-39efd9ccc17a; Ip=[216.228.117.161]; Helo=[mail.nvidia.com] X-MS-Exchange-CrossTenant-AuthSource: CO1NAM11FT112.eop-nam11.prod.protection.outlook.com X-MS-Exchange-CrossTenant-AuthAs: Anonymous X-MS-Exchange-CrossTenant-FromEntityHeader: HybridOnPrem X-MS-Exchange-Transport-CrossTenantHeadersStamped: PH7PR12MB6834 X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org On the device removal kernel notifies user space application with queueing the IBV_DEVICE_FATAL_EVENT and triggering appropriate file descriptor. Mellanox kernel driver stack emits this event twice from different layers (mlx5 and uverbs). The IB port index is not applicable in the event structure and should be ignored for IBV_DEVICE_FATAL_EVENT events. Also, on the older kernels (at least from OFED 4.9) there might be race conditions causing the event queue close before application fetches the IBV_DEVICE_FATAL_EVENT message with ibv_get_async_event() API. To provide the reliable device removal event detection the patch: - ignores the IB port index for the IBV_DEVICE_FATAL_EVENT - introduces the flag to notify PMD about removal only once - acks event with ibv_ack_async_event after actual handling - checks for EIO error, making sure queue is not closed yet Fixes: 40d9f906f4e2 ("net/mlx5: fix device removal handler for multiport") Cc: stable@dpdk.org Signed-off-by: Viacheslav Ovsiienko --- drivers/net/mlx5/linux/mlx5_ethdev_os.c | 34 +++++++++++++++++-------- drivers/net/mlx5/mlx5.h | 1 + 2 files changed, 25 insertions(+), 10 deletions(-) diff --git a/drivers/net/mlx5/linux/mlx5_ethdev_os.c b/drivers/net/mlx5/linux/mlx5_ethdev_os.c index 55801534d1..639e629fe4 100644 --- a/drivers/net/mlx5/linux/mlx5_ethdev_os.c +++ b/drivers/net/mlx5/linux/mlx5_ethdev_os.c @@ -746,6 +746,7 @@ mlx5_dev_interrupt_device_fatal(struct mlx5_dev_ctx_shared *sh) for (i = 0; i < sh->max_port; ++i) { struct rte_eth_dev *dev; + struct mlx5_priv *priv; if (sh->port[i].ih_port_id >= RTE_MAX_ETHPORTS) { /* @@ -756,9 +757,14 @@ mlx5_dev_interrupt_device_fatal(struct mlx5_dev_ctx_shared *sh) } dev = &rte_eth_devices[sh->port[i].ih_port_id]; MLX5_ASSERT(dev); - if (dev->data->dev_conf.intr_conf.rmv) + priv = dev->data->dev_private; + MLX5_ASSERT(priv); + if (!priv->rmv_notified && dev->data->dev_conf.intr_conf.rmv) { + /* Notify driver about removal only once. */ + priv->rmv_notified = 1; rte_eth_dev_callback_process (dev, RTE_ETH_EVENT_INTR_RMV, NULL); + } } } @@ -830,21 +836,29 @@ mlx5_dev_interrupt_handler(void *cb_arg) struct rte_eth_dev *dev; uint32_t tmp; - if (mlx5_glue->get_async_event(sh->cdev->ctx, &event)) + if (mlx5_glue->get_async_event(sh->cdev->ctx, &event)) { + if (errno == EIO) { + DRV_LOG(DEBUG, + "IBV async event queue closed on: %s", + sh->ibdev_name); + mlx5_dev_interrupt_device_fatal(sh); + } break; - /* Retrieve and check IB port index. */ - tmp = (uint32_t)event.element.port_num; - if (!tmp && event.event_type == IBV_EVENT_DEVICE_FATAL) { + } + if (event.event_type == IBV_EVENT_DEVICE_FATAL) { /* - * The DEVICE_FATAL event is called once for - * entire device without port specifying. - * We should notify all existing ports. + * The DEVICE_FATAL event can be called by kernel + * twice - from mlx5 and uverbs layers, and port + * index is not applicable. We should notify all + * existing ports. */ - mlx5_glue->ack_async_event(&event); mlx5_dev_interrupt_device_fatal(sh); + mlx5_glue->ack_async_event(&event); continue; } - MLX5_ASSERT(tmp && (tmp <= sh->max_port)); + /* Retrieve and check IB port index. */ + tmp = (uint32_t)event.element.port_num; + MLX5_ASSERT(tmp <= sh->max_port); if (!tmp) { /* Unsupported device level event. */ mlx5_glue->ack_async_event(&event); diff --git a/drivers/net/mlx5/mlx5.h b/drivers/net/mlx5/mlx5.h index 021049ad2b..6aae8fe3f4 100644 --- a/drivers/net/mlx5/mlx5.h +++ b/drivers/net/mlx5/mlx5.h @@ -1743,6 +1743,7 @@ struct mlx5_priv { unsigned int mtr_en:1; /* Whether support meter. */ unsigned int mtr_reg_share:1; /* Whether support meter REG_C share. */ unsigned int lb_used:1; /* Loopback queue is referred to. */ + unsigned int rmv_notified:1; /* Notified about removal event */ uint32_t mark_enabled:1; /* If mark action is enabled on rxqs. */ uint32_t num_lag_ports:4; /* Number of ports can be bonded. */ uint16_t domain_id; /* Switch domain identifier. */