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 gabe.freedesktop.org (gabe.freedesktop.org [131.252.210.177]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.lore.kernel.org (Postfix) with ESMTPS id B2909C5DF7E for ; Tue, 18 Aug 2026 11:57:55 +0000 (UTC) Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id 4450410E4B8; Tue, 18 Aug 2026 11:57:55 +0000 (UTC) Authentication-Results: gabe.freedesktop.org; dkim=pass (1024-bit key; unprotected) header.d=amd.com header.i=@amd.com header.b="426ioQvp"; dkim-atps=neutral Received: from PH8PR06CU001.outbound.protection.outlook.com (mail-westus3azon11012024.outbound.protection.outlook.com [40.107.209.24]) by gabe.freedesktop.org (Postfix) with ESMTPS id 6DF9010E4B8 for ; Tue, 18 Aug 2026 11:57:53 +0000 (UTC) ARC-Seal: i=1; a=rsa-sha256; s=arcselector10001; d=microsoft.com; cv=none; b=fugTnl2A/PZb9sRa0mLHC+EJR9qwqOYupf5nqNc0dbWVmxveapqJWD8a5L6oLYlctnSAmAIesVGnSm2ElFltbL4VyZ3+kmM/QL7IV7lOIu58zTp3bG2kkfU8WbdhyE0n5KteutkbhdgGYzxVUbhlM1L+725uu3wDSG6zMSgAXHrfjoMwofy89JiQ7zg9SEpU1f/NzbFocUSIWPRbPDqE5AS/KIiVWnZe9KMMo+v1H6HC+1/bPJkePh8h+SoAyNGLwXC4r21AYSIrfPcvUDe7a3zwbTXmORYJnQXO6D3uV4w50AvyX6BMXPuPP2XXyuKUVKwx9x+WtjqsfzggrYwsfw== 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=Bh7qkudzdselXjx14aRskF6pWKQ6VxnHAf7HTBZ4cl0=; b=Vn4dH7XfcEqdMwdCHK4XJ8tGxYb5HY9dnfHQUGnJKXjMPKPmsAYs1PIM1pGHma74a9KVeIA7jT2Kqv6A+GjpEHRJQJXjSNMNzRB1Utt123N6v34QZ2GPZebOL5OG0oqMGQenj1F4T8qdJBeElizGEybqaL8NRDDTM5hN0Y1b4BXgirHL3yif75hEU918po5YhBFlufA3q4a/7VcQpQNP+GKgH+zYCWa+hcGDXfQuAlwklchKhfRRQbyAP6PF2DcTDjFZtEh/w8NNervRBA4YKoLr10gQioAApJr/PsCUUnDBllSZJN/CrZjes4ue3ZwWSbrRlKU3v4xifVXNeVj9uA== ARC-Authentication-Results: i=1; mx.microsoft.com 1; spf=pass (sender ip is 165.204.84.17) smtp.rcpttodomain=lists.freedesktop.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 (0) 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=Bh7qkudzdselXjx14aRskF6pWKQ6VxnHAf7HTBZ4cl0=; b=426ioQvp5x/B7TfjvFdvpGy3GDQmYZKZ0e5pBafsTVtgvWOzT0bpp8zl9F3PtC8Ql4eG1CJw2Znly6+hXoZLghIeqygLyODLbvoAnHpucqeixFuPsbNFToJuJvgOmN8jN5DQXaiVqwgHRgWta8I7zek4cKnIzzQRupT2Q3DNk8k= Received: from DS1PR06CA0022.namprd06.prod.outlook.com (2603:10b6:8:23e::9) by IA0PR12MB8421.namprd12.prod.outlook.com (2603:10b6:208:40f::5) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.21.315.17; Tue, 18 Aug 2026 11:57:48 +0000 Received: from DS3PEPF0000C380.namprd04.prod.outlook.com (2603:10b6:8:23e:cafe::aa) by DS1PR06CA0022.outlook.office365.com (2603:10b6:8:23e::9) with Microsoft SMTP Server (version=TLS1_3, cipher=TLS_AES_256_GCM_SHA384) id 15.21.339.8 via Frontend Transport; Tue, 18 Aug 2026 11:57:48 +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=satlexmb07.amd.com; pr=C Received: from satlexmb07.amd.com (165.204.84.17) by DS3PEPF0000C380.mail.protection.outlook.com (10.167.23.10) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.21.339.3 via Frontend Transport; Tue, 18 Aug 2026 11:57:47 +0000 Received: from ubuntu.amd.com (10.180.168.240) by satlexmb07.amd.com (10.181.42.216) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.2.2562.45; Tue, 18 Aug 2026 06:57:45 -0500 From: Zhu Lingshan To: , , CC: , , Zhu Lingshan Subject: [PATCH] drm/amdgpu: keep the userq manager alive as long as its queues Date: Tue, 18 Aug 2026 19:57:53 +0800 Message-ID: <20260818115753.127358-1-lingshan.zhu@amd.com> X-Mailer: git-send-email 2.53.0 MIME-Version: 1.0 Content-Transfer-Encoding: 8bit Content-Type: text/plain X-Originating-IP: [10.180.168.240] X-ClientProxiedBy: satlexmb07.amd.com (10.181.42.216) To satlexmb07.amd.com (10.181.42.216) X-EOPAttributedMessage: 0 X-MS-PublicTrafficType: Email X-MS-TrafficTypeDiagnostic: DS3PEPF0000C380:EE_|IA0PR12MB8421:EE_ X-MS-Office365-Filtering-Correlation-Id: 1eef7551-3aac-42c9-f2ea-08defd1fec19 X-MS-Exchange-SenderADCheck: 1 X-MS-Exchange-AntiSpam-Relay: 0 X-Microsoft-Antispam: BCL:0; ARA:13230040|36860700016|1800799024|23010399003|376014|82310400026|6133799003|10067099003|11063799006|56012099006|18002099003; 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:NLI; SFV:NSPM; H:satlexmb07.amd.com; PTR:InfoDomainNonexistent; CAT:NONE; SFS:(13230040)(36860700016)(1800799024)(23010399003)(376014)(82310400026)(6133799003)(10067099003)(11063799006)(56012099006)(18002099003); DIR:OUT; SFP:1101; X-MS-Exchange-AntiSpam-MessageData-ChunkCount: 1 X-MS-Exchange-AntiSpam-MessageData-0: tGCF42EdzHOlwu/U6SwmLwhKHK+sqecUpVpxctgoeJX+EjTHGykhiLuNRr6DQyXEnezXsAJp63YurQdA5N46OW5KZP6TpBoyM6Qx+tpulO9J7AngEZjhf4/rTvqcaKfwNBRa3dgwl4opr6Ac2sWq2GnAFz9KwjRno7fVpDpZGOPvsXvmWqeYfP+if2joEki6frA8/a8XDhF+N4bnCjp+RU7rHeMMW1FLrb7sxwl1AfbZAXmb7EDc2OHqCbUduKhAGK4S0n2sSLPosTMp/uUSPPjfgAikLPTf501G5b2PjLDKpasBjtAaTkp1IPJtWOMjbmN/eJ8p3lFwzms/PNTXxjMn7GNwVDiVPQ6AL0tELnKSRDdUbkIlHDRwwdnab4NR9blpN+KEssrW0v8ylxWq91AIbKp3V7sARVeMhrwUPIS1ltbN2U6R/vfzJM9P6SwA X-OriginatorOrg: amd.com X-MS-Exchange-CrossTenant-OriginalArrivalTime: 18 Aug 2026 11:57:47.9392 (UTC) X-MS-Exchange-CrossTenant-Network-Message-Id: 1eef7551-3aac-42c9-f2ea-08defd1fec19 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=[satlexmb07.amd.com] X-MS-Exchange-CrossTenant-AuthSource: DS3PEPF0000C380.namprd04.prod.outlook.com X-MS-Exchange-CrossTenant-AuthAs: Anonymous X-MS-Exchange-CrossTenant-FromEntityHeader: HybridOnPrem X-MS-Exchange-Transport-CrossTenantHeadersStamped: IA0PR12MB8421 X-BeenThere: amd-gfx@lists.freedesktop.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Discussion list for AMD gfx List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: amd-gfx-bounces@lists.freedesktop.org Sender: "amd-gfx" The userq manager is embedded in fpriv, so any code paths trigger kfree(fpriv) would destroy the userq_mgr, even some of its queues are still alive, resuting in userq->userq_mgr use-after-free issues. This commit fixes this problem by introduce a new counter refs representing for the number of its queues, and only free the userq_manager when refs == 0 Signed-off-by: Zhu Lingshan --- drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c | 27 +++++++++++++++++++++++ drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h | 19 ++++++++++++++++ 2 files changed, 46 insertions(+) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c index 24adad7be251..434384b53b85 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c @@ -533,6 +533,17 @@ amdgpu_userq_get_doorbell_index(struct amdgpu_userq_mgr *uq_mgr, return r; } +static void amdgpu_userq_mgr_inc_refs(struct amdgpu_userq_mgr *uq_mgr) +{ + atomic_inc(&uq_mgr->refs); +} + +static void amdgpu_userq_mgr_dec_refs(struct amdgpu_userq_mgr *uq_mgr) +{ + if (atomic_dec_and_test(&uq_mgr->refs)) + wake_up(&uq_mgr->refs_wq); +} + static int amdgpu_userq_destroy(struct amdgpu_userq_mgr *uq_mgr, struct amdgpu_usermode_queue *queue) { @@ -594,6 +605,8 @@ static void amdgpu_userq_kref_destroy(struct kref *kref) r = amdgpu_userq_destroy(uq_mgr, queue); if (r) drm_file_err(uq_mgr->file, "Failed to destroy usermode queue %d\n", r); + + amdgpu_userq_mgr_dec_refs(uq_mgr); } struct amdgpu_usermode_queue *amdgpu_userq_get(struct amdgpu_userq_mgr *uq_mgr, u32 qid) @@ -677,6 +690,7 @@ amdgpu_userq_create(struct drm_file *filp, union drm_amdgpu_userq *args) queue->xcp_id = (fpriv->xcp_id != AMDGPU_XCP_NO_PARTITION) ? fpriv->xcp_id : 0; queue->userq_mgr = uq_mgr; + amdgpu_userq_mgr_inc_refs(uq_mgr); INIT_DELAYED_WORK(&queue->hang_detect_work, amdgpu_userq_hang_detect_work); @@ -789,6 +803,7 @@ amdgpu_userq_create(struct drm_file *filp, union drm_amdgpu_userq *args) free_queue: trace_amdgpu_userq_create_end(queue, r); kfree(queue); + amdgpu_userq_mgr_dec_refs(uq_mgr); err_pm_runtime: pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); return r; @@ -1277,6 +1292,8 @@ int amdgpu_userq_mgr_init(struct amdgpu_userq_mgr *userq_mgr, struct drm_file *f { mutex_init(&userq_mgr->userq_mutex); xa_init_flags(&userq_mgr->userq_xa, XA_FLAGS_ALLOC); + atomic_set(&userq_mgr->refs, 0); + init_waitqueue_head(&userq_mgr->refs_wq); userq_mgr->adev = adev; userq_mgr->file = file_priv; userq_mgr->proc_ctx_allocated = false; @@ -1326,6 +1343,16 @@ void amdgpu_userq_mgr_fini(struct amdgpu_userq_mgr *userq_mgr) amdgpu_userq_put(queue); } + /* + * userq_mgr is embedded in struct amdgpu_fpriv and get + * freed when kfree(fpriv). This wait_evet() blocks + * amdgpu_userq_mgr_fini() which is drm_release path, + * so keep userq_mgr alive while any queues holding it. + * + * This prevents queue->userq_mgr use-after-free issues. + */ + wait_event(userq_mgr->refs_wq, !atomic_read(&userq_mgr->refs)); + xa_destroy(&userq_mgr->userq_xa); /* diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h index 6412a7f7b6ef..7141383ea2ae 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h @@ -24,6 +24,8 @@ #ifndef AMDGPU_USERQ_H_ #define AMDGPU_USERQ_H_ +#include + #include "amdgpu_eviction_fence.h" #define AMDGPU_MAX_USERQ_COUNT 512 @@ -126,6 +128,23 @@ struct amdgpu_userq_mgr { */ struct xarray userq_xa; struct mutex userq_mutex; + + /** + * @refs: + * + * Each queue increase this counter when join this manager, + * and decreash this counter when leave this manager. + */ + atomic_t refs; + + /** + * @refs_wq: + * + * Waited on by amdgpu_userq_mgr_fini() until @refs drops to zero. + * This wait prevents userq->userq_mgr UAF + */ + wait_queue_head_t refs_wq; + struct amdgpu_device *adev; struct delayed_work resume_work; struct drm_file *file; -- 2.53.0