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 A0386C54F4C for ; Tue, 28 Jul 2026 06:26:42 +0000 (UTC) Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id 47E5910E520; Tue, 28 Jul 2026 06:26:42 +0000 (UTC) Authentication-Results: gabe.freedesktop.org; dkim=pass (2048-bit key; unprotected) header.d=intel.com header.i=@intel.com header.b="VwZG/pNy"; dkim-atps=neutral Received: from mgamail.intel.com (mgamail.intel.com [192.198.163.11]) by gabe.freedesktop.org (Postfix) with ESMTPS id 84DEB10E520 for ; Tue, 28 Jul 2026 06:26:40 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1785220001; x=1816756001; h=message-id:date:subject:to:cc:references:from: in-reply-to:content-transfer-encoding:mime-version; bh=Gbtx7FVNuu0ktdCnz3DhtWA7lWdnDXJpGaAJswvIGp0=; b=VwZG/pNy4b+cgSApUCSh1NbuJPW4ftAtxiNXj0UNaunrclU4bpJSFBAW w1iuvUo51i4DdGQG0JiwqUyjiGX8TIUC1bcyb62H1Re7uygeXAuj+Y3BS NrfdhlKr4x/i3DGFXGZ+gDtKCmIRN2NKd5RJhJgMHbbHCVFPrFe/BYcyf gipm2trWRj60wBT/cy8ddVaFz7QCa8rCmYtXlHdwpwsjvjIL8Ujkanebs nrdbr8yBDZmnBdsjxH9UbCbeL0TtQBBM1uLWpc6MMOfdDiM2JKk0GFHpD PPyxAF96ZKFRXy2NcZOnJAabSTUY33seUDih9aKOLD9dGdyKrYzU361V+ A==; X-CSE-ConnectionGUID: iaKb/ctqSOWfzzMIIVuSUw== X-CSE-MsgGUID: sb2OITU/QkWdW61Q5Wc3dg== X-IronPort-AV: E=McAfee;i="6800,10657,11858"; a="96383680" X-IronPort-AV: E=Sophos;i="6.25,189,1779174000"; d="scan'208";a="96383680" Received: from orviesa002.jf.intel.com ([10.64.159.142]) by fmvoesa105.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 27 Jul 2026 23:26:40 -0700 X-CSE-ConnectionGUID: cMCZY3FdSYG4QeOAtURZOQ== X-CSE-MsgGUID: 0U/Rl/uoQeeVEkTtvyC5uQ== X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="6.25,189,1779174000"; d="scan'208";a="289575679" Received: from fmsmsx903.amr.corp.intel.com ([10.18.126.92]) by orviesa002.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 27 Jul 2026 23:26:40 -0700 Received: from FMSMSX901.amr.corp.intel.com (10.18.126.90) by fmsmsx903.amr.corp.intel.com (10.18.126.92) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.2.2562.43; Mon, 27 Jul 2026 23:26:39 -0700 Received: from fmsedg901.ED.cps.intel.com (10.1.192.143) by FMSMSX901.amr.corp.intel.com (10.18.126.90) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.2.2562.43 via Frontend Transport; Mon, 27 Jul 2026 23:26:39 -0700 Received: from BN8PR05CU002.outbound.protection.outlook.com (52.101.57.29) by edgegateway.intel.com (192.55.55.81) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.2.2562.43; Mon, 27 Jul 2026 23:26:38 -0700 ARC-Seal: i=1; a=rsa-sha256; s=arcselector10001; d=microsoft.com; cv=none; b=qN0yVvms12omepXzRBX06EEm6LI2KTqAt54+a3I19MaY6YxxcXjFlJIHxysKdtWIwmP5Jbay0GCrsHp7yTfnHLnsBrtN545RCgyiRzYunLb3ODzlC4FGLflLJ8beNq2PBmS/ZhaRDWKQ6ptZa7iKCYPj5N86TjNoDmIyR6Riip3Yqa6UszP67aE14bC2yUKNPi9tr3UzOdyG81LoPtmTeOan/34GFSnpncDCPvUeQVkOaA4ocmrkxNiNs0XOGs8VjcjNntuffEWi3YQQ4pWEehprrWm76dBGMJhijD0Y1JRBvfaR4G7ePmUsfTBkU/NgUDD5LLYjEyUi02tBpN2noQ== 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=IN0aTFZb+dZvw0CLKqX9CS2DdfkRTRhU0WnCHvpsLFE=; b=fzO1Qqafg96yVy+kzNwg0X3Qm3vlwCAE/izrDfVoIqtNefrSRnP5vkmARbuhBbZCKpVXG8/YhUCVhD8K4jDe5Ei6Ep4b3tpYF9iU51PKHsPBGbAVvat599HzcyAVk9Y/Oc5wJolDHkzERVQXFx8fWbArWVZ2gW9WP1VwszD3cneSGemqB4cbD+2gE0ZPrmR6H69xD1U1EoLByTWF22+XYIMPRZ8hoWaabkhcX1RySgh18VNjCP3znkYoe/lrRCFPmzwp/Nn7+gWh3uHDXRRdLtcnutyuijDmEEz++1ePKqnuh59edKOP6OW4gVpvM6PiDV/2fC/fGPQ6G93+cSrCMw== 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 Authentication-Results: dkim=none (message not signed) header.d=none;dmarc=none action=none header.from=intel.com; Received: from DS4PPF691668CDD.namprd11.prod.outlook.com (2603:10b6:f:fc02::2a) by IA4PR11MB9132.namprd11.prod.outlook.com (2603:10b6:208:56d::15) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.21.245.13; Tue, 28 Jul 2026 06:26:37 +0000 Received: from DS4PPF691668CDD.namprd11.prod.outlook.com ([fe80::5a5c:9c7a:cd1d:d687]) by DS4PPF691668CDD.namprd11.prod.outlook.com ([fe80::5a5c:9c7a:cd1d:d687%6]) with mapi id 15.21.0245.012; Tue, 28 Jul 2026 06:26:36 +0000 Message-ID: Date: Tue, 28 Jul 2026 11:56:29 +0530 User-Agent: Mozilla Thunderbird Subject: Re: [PATCH v2] drm/xe/debugfs: Add knob to wedge on first page fault To: Arvind Yadav , CC: , , References: <20260717082243.337548-1-arvind.yadav@intel.com> Content-Language: en-US From: "Ghimiray, Himal Prasad" In-Reply-To: <20260717082243.337548-1-arvind.yadav@intel.com> Content-Type: text/plain; charset="UTF-8"; format=flowed Content-Transfer-Encoding: 8bit X-ClientProxiedBy: MA0PR01CA0013.INDPRD01.PROD.OUTLOOK.COM (2603:1096:a01:80::13) To DS4PPF691668CDD.namprd11.prod.outlook.com (2603:10b6:f:fc02::2a) MIME-Version: 1.0 X-MS-PublicTrafficType: Email X-MS-TrafficTypeDiagnostic: DS4PPF691668CDD:EE_|IA4PR11MB9132:EE_ X-MS-Office365-Filtering-Correlation-Id: 4d4a1a01-7aa0-4588-b167-08deec712d07 X-MS-Exchange-SenderADCheck: 1 X-MS-Exchange-AntiSpam-Relay: 0 X-Microsoft-Antispam: BCL:0; ARA:13230040|23010399003|1800799024|376014|366016|10067099003|5023799004|11063799006|56012099006|6133799003|10086099003|18092099006|22082099003|18002099003; X-Microsoft-Antispam-Message-Info: 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 X-Forefront-Antispam-Report: CIP:255.255.255.255; CTRY:; LANG:en; SCL:1; SRV:; IPV:NLI; SFV:NSPM; H:DS4PPF691668CDD.namprd11.prod.outlook.com; PTR:; CAT:NONE; SFS:(13230040)(23010399003)(1800799024)(376014)(366016)(10067099003)(5023799004)(11063799006)(56012099006)(6133799003)(10086099003)(18092099006)(22082099003)(18002099003); DIR:OUT; SFP:1101; X-MS-Exchange-AntiSpam-MessageData-ChunkCount: 1 X-MS-Exchange-AntiSpam-MessageData-0: =?utf-8?B?eW5zSzNFcFQ2SDdpYkZOVlBuSWxQYWhwQjlESlp0UmpveUhCTXFLeDk3SW1G?= =?utf-8?B?Ny9JeTNPM3VOcS9KeVBZMXc5T3BSb3NoR3JkQTNzL2REUDZ1TGF1OTZYais2?= =?utf-8?B?bmE2WTlzUUJBTXZUWVpxbWZJNytwVkZyV2MxdkVqbG55QVQvNzFmaFFJdUgw?= =?utf-8?B?b0pQYXh3emdsWU9yaU9uYk1RT3dpS2Q3ejJZM1prNHZSYlBkQk5HOEZHY09t?= =?utf-8?B?WW1pZUVvcWVqOWltdHJ6WHJqQzRSYlVRNmJ1S3VYbjMySEhmUjhmUldhdGNS?= =?utf-8?B?T0pVRGdWOFQrWk4rTHBuREdjSFNSQnB3emkvWjhhOEJLR2k2ZmE0VWozeE5D?= =?utf-8?B?TzFPNUFIYkFPRitzNDE5Q0JqTU1MQWJrZG9zSXJwMWkwUE1CWDFJVWRVUk50?= =?utf-8?B?cE5qdnNOREV2ZDRIQUg2L2grQWduZHN5SjlJc09CVHdUV1Q2VUREMWl6aUdx?= =?utf-8?B?MWVVWjdFaCsrNmJTcjRsdmRXaDl1dERMd3VGczA0K2FQYzN2WUdjaTZuK05L?= =?utf-8?B?ZkFnYW43anBESVJHSXBkRVJtWFFQbTFYSmlYM2ErcWJCWEtsZ01iNE9UZHRa?= =?utf-8?B?UG1qNmt2U2lKNDZROEdUYkVpWFczclBtK2NBWmpyUThNUzZuQ3ZuRW1BejNq?= =?utf-8?B?L2Y2QzY4SVNtT3hWSGx1ZnlkdHlnV3J4QW9JQWpEYlQwSEY3dFhobkxCajYr?= =?utf-8?B?YkRMcUNTZytVVUxpYnZ1NVNaM0NsblJZWWc4M29JaFhRWVk2UngzUVByYTVh?= =?utf-8?B?Q3BYKytZOVdxR2FYNTRFUnlHY24wOW9lNFNBWTdJVVVneG1EU2lNMVNDbVds?= =?utf-8?B?NmdEbVZIajU2Wm5tZFpONWE3RkQzNEFTN0FYbHRpOGhNb0VFSzhRTWE0Mnl4?= =?utf-8?B?UG5xTUJqSFZpQ2p0MWgvR1psRXZJMU9qbDRSUFczYUhJUWxGTnhtZEV1Wkhw?= =?utf-8?B?VWJmNkNmK1VVckR2UEhTYlpvSDdiQmwrWmZEcE1sY3NBWDZWZWVJSmExWHlt?= =?utf-8?B?UTBaRDlsTjMzU1VDY2l1eWsvMnJ5Z1BLdjhLUHJJeXN1aTRyVlkva0QrOE1m?= =?utf-8?B?Q0tiWVVycXlydHBsdERKdHlheG9tZC9IbGQwbEo1T0gzdWQxT2gyUi9UZlJk?= =?utf-8?B?VzBIM21adnpWcC9MU2kvY0lSdVdwK0FqMDZXL0N2K1krR3N3VDdBejlvczJ3?= =?utf-8?B?aGIvZHRzYTlqOGVoZkpjQ3pldTVueEFLQlo0a1ZXeHRZNnZsdFlUVjNpc0Fl?= =?utf-8?B?QUVHaEpvMmdIWlNzcEl6ckExYllscGFjaEhVMTk0VTJ5NWpmY3BETUMvWm0r?= =?utf-8?B?ZXVqYnRsSHdmZllxbG50UWJVdkZsODRjS2VTeWZQS0lCcFBQV2NMK0lzM3Va?= =?utf-8?B?Vk9KbWdCVktjQ2FMdTZiZW04Y254d2ptTVVFTXVuZWM5WHN6b2l4RjJxOXp1?= =?utf-8?B?YW9Lbk1JQWxET1M3bzhBY05jWGF1ZXFPOGpVbEplaHo0ZE9mdXVQOE1LUGpK?= =?utf-8?B?L1U0TFc0L3lDNlJLU0VOQ09ybVplUXVyVGlDcHdobUpEcUIrZEFmWFU2bmZ1?= =?utf-8?B?N0VZMjZ3SDZCN2J5YjBzYS9NaG9PdmlXY1NyL3VkZytzbDdlWFFhcjczU296?= =?utf-8?B?elpuSDVsTHE4cjNZNEt3NHQvbGFmckpyaldxNWw5eVRBRm1vUDVGUVc0NW1M?= =?utf-8?B?NEpabEJWa29WTEozN3Q1ZjJ1bXArdVNsd2RHV3E2Wmd0R0hpR0JnQ3pvS3VI?= =?utf-8?B?TlpSYlFzWms3RmVIMDVmRDRKSm5pVnBEVUY4YVFyN1dqcTBIQ2VwWnZ3WWhy?= =?utf-8?B?c0hzSE9BUmFxS2poTFBLSTRhMHdSU05kUS9YbUVWKytKT0ZOZHk4RDVPeG9S?= =?utf-8?B?RXd6QmppTDNINjZyNWk4TXBtV1NESnl6ZXlVZFJLMWdOaEdkYisvajJLVmMy?= =?utf-8?B?MTh2eEg0K25GY29zWmNlQktHMUtrSkN3em1MdVhTak5idXp6NWl6R0lVMEVa?= =?utf-8?B?NTZNTnZzaGJVTWJNRlN0Y0dkZG9QMlh1L0J4K003TXFSbHJydzdBMjN5MCtJ?= =?utf-8?B?OHRvNnlRV2VSR1pVS0tFYTJOTnNqQWw5UDFHRGQ0Y0RqdS8xZTdvNC8yajVr?= =?utf-8?B?ejJnZnRDREhKMDdnbUxTdXVHazMvdHA0UzRWd0xwTWN6VHZ3QzlveDNPT2Z2?= =?utf-8?B?bUlJQTJqdGJWNENZUEVYbmZ6NGtTQ2dHa1QzemwySVgzSDdsNmFxdzFHQWY3?= =?utf-8?B?QXhCZ3ZzcnlkazI4L3ZGU0RPczhCSVRxbjhvSVhzMHY5TlZFcHdEcjNsTk1X?= =?utf-8?B?a1g5R1JDSmR1YVhOTXNrSkRUZzVRUVRtbWp1cVVQNEFmcUh5TVFHTlpwc29u?= =?utf-8?Q?xoQvoQaobTZaNq3Q=3D?= X-Exchange-RoutingPolicyChecked: bBo79H43QCoJcId5qg5i6gk9Tv5iKRymgPelIF9s2ws5hpmnGkBi7fLKkwZmb7zufzpdeX5y8CKmb3GQFKWV7mTqomIBr0WwKiuz6Tc3PJIwdqf5THIPMUasD4JoCY9ZbV16uXW7767/BJwLXLxtxNOTnGWC8iHs131AepmOX2mfWc3QWoNpfMwEjb19bHVrk378tgiKcIYyHIwHLxJpxx/hr9SqjMck+qsaRQMP4L8C3Wqp3WUpzj84Q3QprAVkIWhOmWBn8sUhUTLz8XMjcgWQa1WbNiwm7wccnvmytyJB31hKpIX9XVnUzxbSvKSMIa/Mafvvx3ZoxAlDrIUD0w== X-MS-Exchange-CrossTenant-Network-Message-Id: 4d4a1a01-7aa0-4588-b167-08deec712d07 X-MS-Exchange-CrossTenant-AuthSource: DS4PPF691668CDD.namprd11.prod.outlook.com X-MS-Exchange-CrossTenant-AuthAs: Internal X-MS-Exchange-CrossTenant-OriginalArrivalTime: 28 Jul 2026 06:26:36.8584 (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: yUORsOW0hKNw8s/FLHhSpveGlOKnxhlIELDL0O1kgbc/rnflfRJum0EGfA7KyIMtpR8cMMGPQLgzJblOA9SkyIumvYgk1HHpGkL5NmE2+Pc= X-MS-Exchange-Transport-CrossTenantHeadersStamped: IA4PR11MB9132 X-OriginatorOrg: intel.com X-BeenThere: intel-xe@lists.freedesktop.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Intel Xe graphics driver List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: intel-xe-bounces@lists.freedesktop.org Sender: "Intel-xe" On 17-07-2026 13:52, Arvind Yadav wrote: > Add a debugfs knob to wedge the device on the first GPU page fault. > > This is intended for debug where preserving the faulting HW > state is more useful than servicing the fault. When enabled, > the first page fault is printed and the device is declared wedged > without servicing or acknowledging the fault. > > echo 1 > /sys/kernel/debug/dri//wedge_on_first_pagefault > > v2: > - Move first-pagefault debug state under xe->wedged. (Matt) > - Queue the wedge operation on the GT ordered workqueue. (Matt) > - Drop xe_pagefault_save_to_vm() from the debug path. (Matt) > - Cancel pending first-pagefault wedge work in pagefault fini. > - Re-arm pf_wedge_triggered before enabling wedge_on_first_pagefault. > (Sashiko) > - Use release/acquire ordering between debugfs enable and PF worker. > (Sashiko) > - Do not clear the latch on disable. (Sashiko) > LGTM Reviewed-by: Himal Prasad Ghimiray > Cc: Thomas Hellström > Cc: Matthew Brost > Cc: Himal Prasad Ghimiray > Signed-off-by: Arvind Yadav > --- > drivers/gpu/drm/xe/xe_debugfs.c | 48 ++++++++++++++++++++++++++++ > drivers/gpu/drm/xe/xe_device_types.h | 17 ++++++++++ > drivers/gpu/drm/xe/xe_pagefault.c | 34 ++++++++++++++++++++ > 3 files changed, 99 insertions(+) > > diff --git a/drivers/gpu/drm/xe/xe_debugfs.c b/drivers/gpu/drm/xe/xe_debugfs.c > index 5a3877fcb0f0..e65e233fdfcd 100644 > --- a/drivers/gpu/drm/xe/xe_debugfs.c > +++ b/drivers/gpu/drm/xe/xe_debugfs.c > @@ -614,6 +614,49 @@ static const struct file_operations disable_late_binding_fops = { > .write = disable_late_binding_set, > }; > > +static ssize_t wedge_on_first_pagefault_show(struct file *f, char __user *ubuf, > + size_t size, loff_t *pos) > +{ > + struct xe_device *xe = file_inode(f)->i_private; > + char buf[8]; > + int len; > + > + len = scnprintf(buf, sizeof(buf), "%d\n", > + READ_ONCE(xe->wedged.on_first_pagefault) ? 1 : 0); > + > + return simple_read_from_buffer(ubuf, size, pos, buf, len); > +} > + > +static ssize_t wedge_on_first_pagefault_set(struct file *f, > + const char __user *ubuf, > + size_t size, loff_t *pos) > +{ > + struct xe_device *xe = file_inode(f)->i_private; > + bool val; > + int ret; > + > + ret = kstrtobool_from_user(ubuf, size, &val); > + if (ret) > + return ret; > + > + if (val) { > + atomic_set(&xe->wedged.pf_triggered, 0); > + > + /* Clear the latch before making the enable visible. */ > + smp_store_release(&xe->wedged.on_first_pagefault, true); > + } else { > + WRITE_ONCE(xe->wedged.on_first_pagefault, false); > + } > + > + return size; > +} > + > +static const struct file_operations wedge_on_first_pagefault_fops = { > + .owner = THIS_MODULE, > + .read = wedge_on_first_pagefault_show, > + .write = wedge_on_first_pagefault_set, > +}; > + > void xe_debugfs_register(struct xe_device *xe) > { > struct ttm_device *bdev = &xe->ttm; > @@ -665,6 +708,11 @@ void xe_debugfs_register(struct xe_device *xe) > debugfs_create_file("disable_late_binding", 0600, root, xe, > &disable_late_binding_fops); > > + /* Debug: wedge on first page fault to freeze HW state for inspection. */ > + if (xe->info.has_usm) > + debugfs_create_file("wedge_on_first_pagefault", 0600, root, xe, > + &wedge_on_first_pagefault_fops); > + > /* > * Don't expose page reclaim configuration file if not supported by the > * hardware initially. > diff --git a/drivers/gpu/drm/xe/xe_device_types.h b/drivers/gpu/drm/xe/xe_device_types.h > index 022e08205897..f0862ca9cb5d 100644 > --- a/drivers/gpu/drm/xe/xe_device_types.h > +++ b/drivers/gpu/drm/xe/xe_device_types.h > @@ -496,6 +496,23 @@ struct xe_device { > unsigned long method; > /** @wedged.inconsistent_reset: Inconsistent reset policy state between GTs */ > bool inconsistent_reset; > + /** > + * @wedged.on_first_pagefault: Debug knob to wedge the device on the > + * first GPU page fault instead of servicing it, freezing HW state > + * for inspection. Only exposed on USM-capable devices. > + */ > + bool on_first_pagefault; > + /** > + * @wedged.pf_triggered: One-shot latch for @on_first_pagefault to > + * ensure a single wedge is triggered per arming. > + */ > + atomic_t pf_triggered; > + /** > + * @wedged.pf_worker: Declares the device wedged for > + * @on_first_pagefault from the GT ordered workqueue, where > + * stopping GuC submission is safe. > + */ > + struct work_struct pf_worker; > } wedged; > > /** @devres_group: devres group */ > diff --git a/drivers/gpu/drm/xe/xe_pagefault.c b/drivers/gpu/drm/xe/xe_pagefault.c > index dd3c068e1a39..7580baeb277a 100644 > --- a/drivers/gpu/drm/xe/xe_pagefault.c > +++ b/drivers/gpu/drm/xe/xe_pagefault.c > @@ -288,6 +288,14 @@ static void xe_pagefault_save_to_vm(struct xe_device *xe, struct xe_pagefault *p > xe_vm_put(vm); > } > > +static void xe_pagefault_wedge_work(struct work_struct *w) > +{ > + struct xe_device *xe = container_of(w, struct xe_device, wedged.pf_worker); > + > + /* GuC submission stop/start must run from the GT ordered workqueue. */ > + xe_device_declare_wedged(xe); > +} > + > static void xe_pagefault_queue_work(struct work_struct *w) > { > struct xe_pagefault_queue *pf_queue = > @@ -299,11 +307,34 @@ static void xe_pagefault_queue_work(struct work_struct *w) > threshold = jiffies + msecs_to_jiffies(USM_QUEUE_MAX_RUNTIME_MS); > > while (xe_pagefault_queue_pop(pf_queue, &pf)) { > + struct xe_device *xe; > int err; > > if (!pf.gt) /* Fault squashed during reset */ > continue; > > + xe = gt_to_xe(pf.gt); > + > + /* > + * Debug mode: if armed via debugfs, wedge on first fault instead of > + * servicing. Print the fault, then declare the device wedged from the > + * GT ordered workqueue (where stopping GuC submission is safe). > + */ > + if (unlikely(smp_load_acquire(&xe->wedged.on_first_pagefault))) { > + if (!atomic_xchg(&xe->wedged.pf_triggered, 1)) { > + xe_pagefault_print(&pf); > + xe_gt_err(pf.gt, > + "wedge_on_first_pagefault: freezing HW after first fault\n"); > + queue_work(pf.gt->ordered_wq, &xe->wedged.pf_worker); > + } > + > + /* > + * Do not service or ack. Stop processing this queue so the > + * captured state stays close to the first fault. > + */ > + return; > + } > + > err = xe_pagefault_service(&pf); > if (err) { > xe_pagefault_save_to_vm(gt_to_xe(pf.gt), &pf); > @@ -380,6 +411,7 @@ static void xe_pagefault_fini(void *arg) > struct xe_device *xe = arg; > > destroy_workqueue(xe->usm.pf_wq); > + cancel_work_sync(&xe->wedged.pf_worker); > } > > /** > @@ -397,6 +429,8 @@ int xe_pagefault_init(struct xe_device *xe) > if (!xe->info.has_usm) > return 0; > > + INIT_WORK(&xe->wedged.pf_worker, xe_pagefault_wedge_work); > + > xe->usm.pf_wq = alloc_workqueue("xe_page_fault_work_queue", > WQ_UNBOUND | WQ_HIGHPRI, > XE_PAGEFAULT_QUEUE_COUNT);