From mboxrd@z Thu Jan 1 00:00:00 1970 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 7F125C61DFD for ; Wed, 2 Sep 2026 06:31:42 +0000 (UTC) Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id 379AC10E46E; Wed, 2 Sep 2026 06:31:42 +0000 (UTC) Authentication-Results: gabe.freedesktop.org; dkim=pass (2048-bit key; unprotected) header.d=intel.com header.i=@intel.com header.b="d5AehIhZ"; dkim-atps=neutral Received: from mgamail.intel.com (mgamail.intel.com [192.198.163.10]) by gabe.freedesktop.org (Postfix) with ESMTPS id BE14A10E46E for ; Wed, 2 Sep 2026 06:31:40 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1788330701; x=1819866701; h=message-id:date:subject:to:cc:references:from: in-reply-to:mime-version; bh=EyxPIxL4PEjvcjCItXNv9k9fN+Vc8l7jTGtiCc6ZdTU=; b=d5AehIhZG08bEotT4tPcLW9ofWvcUFp8qTPXKkULFoP1bR+BUm68kdPk 37wutAx7ufWIAYsCm/v+2LYGJ+RZ6GkeGhG32itkWpc4+PIf7b6H32G8/ TZKnzeBT/vx9/CT1FoRtRrRxFWcb6xsSAyZ0vmr3tWqu2v+xGjp3zjCEA tZ9BnWNBaINXG+IrwUhVriWf03kunlFSCR67HH0GktrHKUH6uM8+7kkZR YxRTDc5JlHkg9NRosRvBFiI2cEGh4OezdxgZkqTEd3zWvoEu4Ru49f+Fu imbCvBcflJo+UHTpY/pxvqaXGwQ9HrqWUs8DhXMsAj8KrHpLLtBe17snC w==; X-CSE-ConnectionGUID: QbCDuaitRG+75AeTFoyg4Q== X-CSE-MsgGUID: CB7sV9aTQLWnu4Kmt9ZPaA== X-IronPort-AV: E=McAfee;i="6800,10657,11893"; a="100131520" X-IronPort-AV: E=Sophos;i="6.25,257,1779174000"; d="scan'208,217";a="100131520" Received: from orviesa007.jf.intel.com ([10.64.159.147]) by fmvoesa104.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 01 Sep 2026 23:31:40 -0700 X-CSE-ConnectionGUID: aFvxNUP6Rqq8U16/hWAy5Q== X-CSE-MsgGUID: dkmO6NGtRcqB6B1eEvg3UA== X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="6.25,257,1779174000"; d="scan'208,217";a="269337702" Received: from fmsmsx901.amr.corp.intel.com ([10.18.126.90]) by orviesa007.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 01 Sep 2026 23:31:39 -0700 Received: from FMSMSX901.amr.corp.intel.com (10.18.126.90) 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 15.2.2562.46; Tue, 1 Sep 2026 23:31:39 -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 15.2.2562.46 via Frontend Transport; Tue, 1 Sep 2026 23:31:39 -0700 Received: from DM5PR21CU001.outbound.protection.outlook.com (52.101.62.42) 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 15.2.2562.46; Tue, 1 Sep 2026 23:31:27 -0700 ARC-Seal: i=1; a=rsa-sha256; s=arcselector10001; d=microsoft.com; cv=none; b=npWG2+LdPtcA3SBe6Qk1mjkABRBhZ3zgW+WPrgQUghHWoImc19MJANuSsisbitT8/JvEIuwCuQgfA98Li7nipF7s4b5E17Fz6aYuNYo4I19GiNKEcQSTzNic6Uj/ZqDa9mncuKXk/tewPRl8JFZgCiCEMON+7nG+sWuVMUazW4eiTHwHh3rjaVW8/RodjpNE5QmrJbqNRqcNR3yMykeImseFNsDj/lFW0mayAm19eahhn5FYQJhtQ6TDYLIG9ia1TbRuMUXtkjGcQcavIfHQ/vWgn5opfv68qgpSR9qTuS1oL5JgRVtFhNGWKdWVspCjbKnCPd4hKj/TLGuSMLRscQ== 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=izqssHYdOatR+ZGyIatO1AY2QGfe5nrkcuxiXqEMOVQ=; b=w+VUg8OSiNOSRVeyRhxPsxlnKcE3aqW8P5KzymTRgIL1aAt2o6HWeORHztXtHbvKXZsC9LwBQuMEt+Hjjo+AwFHYN+SBQhVo3BVxaUsr5CFn1KWj8xJClt9ld91bID12S7/AwxewK5qQBgPv+VrEvbNbeOHMAv2ah+1p8DdkrM42OEJmowfkU7DrjtNQB+9sTVNDJPM1q+3c4jjUmFfHfv7qZ1Fb0K7/J3pbIGUmN6BWWK+HFKq81uNDLhg1WpSIvn5iyfsBN15LwGo0/DQIJ4Vr6mM/73urk9PkWfwysSumGUWc1CNVMYvU3J+USy8i8yPSykEV/WCxpdZ2hAHGMA== 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 MN0PR11MB6207.namprd11.prod.outlook.com (2603:10b6:208:3c5::21) by PH7PR11MB5914.namprd11.prod.outlook.com (2603:10b6:510:138::7) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.21.360.13; Wed, 2 Sep 2026 06:31:03 +0000 Received: from MN0PR11MB6207.namprd11.prod.outlook.com ([fe80::52eb:929f:a8b2:139d]) by MN0PR11MB6207.namprd11.prod.outlook.com ([fe80::52eb:929f:a8b2:139d%4]) with mapi id 15.21.0360.008; Wed, 2 Sep 2026 06:31:03 +0000 Content-Type: multipart/alternative; boundary="------------if3MC01eUyHM9gCt1BCNyua1" Message-ID: <17780d34-b06b-4db7-888e-e3734530ae31@intel.com> Date: Wed, 2 Sep 2026 12:00:56 +0530 User-Agent: Mozilla Thunderbird Subject: Re: [RFC PATCH 2/5] drm/xe/xe_ras: Handle page offline requests for device memory ecc errors To: Riana Tauro CC: , , , , , , , , , "intel-xe@lists.freedesktop.org" References: <20260825063615.3697317-7-riana.tauro@intel.com> <20260825063615.3697317-9-riana.tauro@intel.com> Content-Language: en-US From: "Mallesh, Koujalagi" In-Reply-To: <20260825063615.3697317-9-riana.tauro@intel.com> X-ClientProxiedBy: MA5P287CA0224.INDP287.PROD.OUTLOOK.COM (2603:1096:a01:1b4::16) To MN0PR11MB6207.namprd11.prod.outlook.com (2603:10b6:208:3c5::21) MIME-Version: 1.0 X-MS-PublicTrafficType: Email X-MS-TrafficTypeDiagnostic: MN0PR11MB6207:EE_|PH7PR11MB5914:EE_ X-MS-Office365-Filtering-Correlation-Id: ecf0fb51-e4b2-4bd8-7a5e-08df08bbc2f4 X-MS-Exchange-SenderADCheck: 1 X-MS-Exchange-AntiSpam-Relay: 0 X-Microsoft-Antispam: BCL:0; ARA:13230040|1800799024|366016|23010399003|376014|18002099003|22082099003|13003099007|8096899003|56012099006|6133799003|4143699003|10067099003|11063799006; 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:MN0PR11MB6207.namprd11.prod.outlook.com; PTR:; CAT:NONE; SFS:(13230040)(1800799024)(366016)(23010399003)(376014)(18002099003)(22082099003)(13003099007)(8096899003)(56012099006)(6133799003)(4143699003)(10067099003)(11063799006); DIR:OUT; SFP:1101; X-MS-Exchange-AntiSpam-MessageData-ChunkCount: 1 X-MS-Exchange-AntiSpam-MessageData-0: =?utf-8?B?UXIxL3R5SVJhaUR5bnpCOU9JVjBqY2dJVmlZLzhWVitsMXVGRmcrdFBDc3ZK?= =?utf-8?B?YmZPSDZSZzhBTlJaSFB0MEVzMFo3amhsNDArU3ZXemdlOUZWNFVaUDl5UGZt?= =?utf-8?B?TWZMTE1JQ3l2SDFkRGQvaEpkVnE2WWRIK3hWTDB5amEyNnNBK2JkcnVJOWpY?= =?utf-8?B?OTVjTnljY0d3YTJUOXQxZ3pjeFZwb1VDUzRzb0dpUURWYmYxSjVRNWMrZDFN?= =?utf-8?B?VHQwc1pOSm5nbUpROTRQQzlENG9zUjd5Q2lvVENvUzdsSll2cENVd2J1R2ZB?= =?utf-8?B?eWQxbEZEM3FmbjRoRlJaRXhnSmRkTUJhWWl5bGY5NFd5QTAwTVl0eWpvZkVX?= =?utf-8?B?aW1nQjY2SllRc1JqdXdRU2FCK3lJcTVEV0kwbFJHUUp4aUZLcjZVbnVyVUVt?= =?utf-8?B?UTI0NE1odVlXRjUwYXVxVVgvSW12Ymt3WGN1WVRmZmlSYmk0QkxxVDRudzJS?= =?utf-8?B?UE9KUHA3Z3U3NU9QZ2NNeWJ6WEdoOVg2c0xOU0hQT094V1ZVUzg4bXZCTS9l?= =?utf-8?B?Z2hPRks3cnRLMnhBWGFJY053bEE2eDUvbHBtV092eVpabm0vazNZaFM2cHls?= =?utf-8?B?M1Y4RFlIR2hoNy9aWTJxc1Q3Sk01aGpNRkxmcVlkY3pDK1NzZU9DeGRDUHBO?= =?utf-8?B?cG9ZamcrS2NNM3lPcE9kWWtrakFlcXl5WW4wMkE5emZtT2J0dnhyK25hMDVF?= =?utf-8?B?YmhvTGw5QWhlbEVmRXNMc3Y4bDFnM1k4WDQxV3poeG5QVHV0NTRlUmt4Ujhl?= =?utf-8?B?ZTFNcFhNNWdJY3JXOWFWNmZ3MGNPQU9tMU5tL3hnVzg4cVZsdnZLMWsybnZO?= =?utf-8?B?QWNzZDJlem9OcWlETlZQeTZTbkFyRURLdml3d1F5ZURlMVczQXdCNThtMDRF?= =?utf-8?B?eEFqZS9ncUszcUJCc1BoQ2J2NHhRNHRCZFUrMXU5SDZodXpaZGVIcGE2Nm5r?= =?utf-8?B?N3hOc3piV29Na1BReHRWNm9PeTFhS05CN09SY0VRbzNqRTFWZkYycVVnWUkz?= =?utf-8?B?T014VnBNa29ZbTFKQ0N1SU9mUTFXMHJsaGNCMmZ6ZHVrWVVKazRvaTFYOVhK?= =?utf-8?B?akIvODVibCtxSjJ0bVpXbVlsTUhMd0E3Yk51SVZRcHF2RzZxbVhLMnhSYk5J?= =?utf-8?B?TVFLeVhja3B1WmhGWlNnVTB6aFRvcnJVWFdJUEFIOE9iMFl6eVZKTVBvWHpK?= =?utf-8?B?RnFQUlVTSDRKQzU2S3NobVZkYTZuQTkva1dQRXROYytRMXBjbVFiZHZFYnBF?= =?utf-8?B?UjAwNVlPalptRWFlc2lZbFNWd0lnNmx4Vzg1QzNWdDl2SnBiRGdUQnU4Z21Q?= =?utf-8?B?NUIxUVNSVVY5NVNqWCsyMkJRZGFZUU9OSm9md242bUd3WGJYZis3KzVVUjF6?= =?utf-8?B?eGhMVGd3bjh1dk5ydUhMZThHb1BoZklxTitjcWdVTi84OUFzQndScEYxUEFN?= =?utf-8?B?MzhaeWVQRnVHWUVueDgvTkhtK09HSjRYNFFnRVVoeE81OGoxSHhnOTNVMFpi?= =?utf-8?B?c3BxLzQ4dVgvemliZUdLTmt5bkxJTnJKUCs0WnowOVN1S3ZXRThsV1FPN09u?= =?utf-8?B?VHNEak5VQ0VjM0VQVHBiZGUzMEVxS1pxMnpYeHQzdklqUmtDZFVndmN3TU5Q?= =?utf-8?B?UGxVM25UMXpSYktpeHQvTUxaYTJTZlN2K0JsY2hiQmFJN0lSK3M3T2tQSllG?= =?utf-8?B?NkR0VFRrYTFGTTJHWWZhY3JSZUZqbWFJWkZmQlhsclgzaGF3SGorMGVhWVFG?= =?utf-8?B?NlZIajBWYkQ3ZHd6NDdab2Z6Ynh1NXBLL3EwMjZ3RURsM3BxY2E1U3hZTmQv?= =?utf-8?B?Vkt5U3MrdDhrMlpLVTVWVUc5TGpEMktyL1pycGsyWjYxbHozaUxwUU9SNTFv?= =?utf-8?B?Z0szM2NoV3Z2UlVCMVF4a1J1OEU1NTl4cUhQOXY1NHBTK1BMcUpkWGV5WXhV?= =?utf-8?B?cThZU0VnZUNTdTdPSWhvN1ZHSEV4bTdJZXY3NVBNdFZtdHlsbzB3cGRxS3hx?= =?utf-8?B?V0dWSE9OWTVCcXMyK3NSOVY5SmtOUGl5UFJGK2NwYnNOYU1TdFh1S0F1eTRs?= =?utf-8?B?Z1AwK0JRK29BR0pQdU5xT1h3bVRsekduWTlYeHRpOGlmTS9JVWtpZG9wcUpQ?= =?utf-8?B?SHhXcU85RjIvRUNuek92akdYbmlrV1NXTHgxT2haYURCWGJ6TW9BbUlDQitY?= =?utf-8?B?K3pJSnZFeFFMNG05YklobkFTWDFjSWZQeEY3L056NTQ2YXZDRTlheW1uOGxX?= =?utf-8?B?ZC94MWRmWThwSHpHUjZWenl5eTc0QS85TFNlSnNqcjlXSi9CbW8xcFpremFH?= =?utf-8?B?UUtoQXMzR2lLNERzSzlmNkUxQ2hPWU50cmRoMVZFZjFRRmRJTVRxOFowRnlw?= =?utf-8?Q?dI3pUwwQ8uprOIDM=3D?= X-Exchange-RoutingPolicyChecked: pr2cGchR7BzV3n1fz81rxWYzHjZfLEETSeb7S6b/QORu3ge5mz15mXPU0OCMA8h1ujh1VP6vwpFVhz1xwWnL6Eq5m6uc7OmlQHmPAE/noEoikZRneGHCCfhovXZWsUHs18+/RZFD3q35w1kGhbc4bOCPw/BmRhkofBfF9b7cOlQHszvmwBld2GLCGdwvOXbPPeIRF9a+Jol3uTlQjrgsWOC8DwoevK3kkFsfx+MKISokZWNuxYnRe5A7zLfP49EPLetl/Siz12gxBAXNOAjk6q8c8wZrjIPaZStHpDERUYcUlESkgF83bCYlYbI9VwVu0xd25ge6Zz6Nb5bkoaM4DA== X-MS-Exchange-CrossTenant-Network-Message-Id: ecf0fb51-e4b2-4bd8-7a5e-08df08bbc2f4 X-MS-Exchange-CrossTenant-AuthSource: MN0PR11MB6207.namprd11.prod.outlook.com X-MS-Exchange-CrossTenant-AuthAs: Internal X-MS-Exchange-CrossTenant-OriginalArrivalTime: 02 Sep 2026 06:31:03.4669 (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: W+NkN0j0516rnyUa2flu552Hl4zroImfy4MU4ukBAbdvX2UyZNIkApmNX2VxP3AoLVBPWIXqCyFe4Ea4WSYRcKeAPRYcTRTy+SmDSq4VJmg= X-MS-Exchange-Transport-CrossTenantHeadersStamped: PH7PR11MB5914 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" --------------if3MC01eUyHM9gCt1BCNyua1 Content-Type: text/plain; charset="UTF-8"; format=flowed Content-Transfer-Encoding: 8bit On 25-08-2026 12:06 pm, Riana Tauro wrote: > This will be integrated with the related address-fault handling flow > once this patch is merged. > https://lore.kernel.org/intel-xe/20260818104055.3833974-14-tejas.upadhyay@intel.com/ > Sending for initial comments. > > Add basic support for sending page offline/decline requests to system > controller and use it for device memory ECC error handling. > Pages that belong to critical BOs cannot be handled by offlining and > require a SBR (Secondary Bus Reset). > Pages that are configured for log-only handling are not marked as bad by > firmware. > > For all other valid page addresses, the first occurrence of error > indicates a poison error and the page is offlined only by software. > Firmware avoids permanently marking the page as bad. The second occurrence > of an error indicates a Double-bit ECC error and the firmware > permanently marks the page as bad. > > Cc: Tejas Upadhyay > Cc: Himal Prasad Ghimiray > Signed-off-by: Riana Tauro > --- > drivers/gpu/drm/xe/xe_ras.c | 121 +++++++++++++++++- > drivers/gpu/drm/xe/xe_ras_types.h | 35 +++++ > drivers/gpu/drm/xe/xe_sysctrl_mailbox_types.h | 2 + > 3 files changed, 153 insertions(+), 5 deletions(-) > > diff --git a/drivers/gpu/drm/xe/xe_ras.c b/drivers/gpu/drm/xe/xe_ras.c > index d25d25f77531..c643c7137a42 100644 > --- a/drivers/gpu/drm/xe/xe_ras.c > +++ b/drivers/gpu/drm/xe/xe_ras.c > @@ -3,6 +3,7 @@ > * Copyright © 2026 Intel Corporation > */ > > +#include "xe_bo.h" > #include "xe_debugfs.h" > #include "xe_device.h" > #include "xe_drm_ras.h" > @@ -200,6 +201,115 @@ static inline const char *comp_to_str(u8 component) > return xe_ras_components[component]; > } > > +static int send_page_offline_cmd(struct xe_device *xe, u64 page_address, > + enum xe_ras_page_action action) > +{ > + struct xe_sysctrl_mailbox_command command = {0}; > + struct xe_ras_page_offline_request request = {0}; > + struct xe_ras_page_offline_response response = {0}; > + size_t rlen; > + int ret; > + > + if (!xe->info.has_sysctrl) > + return 0; > + > + if (action >= XE_RAS_PAGE_ACTION_MAX) { > + xe_log_err(xe, DEVICE_MEMORY, -EINVAL, "Invalid page offline action %d\n", action); > + return -EINVAL; > + } > + > + request.page_address = page_address; > + request.action = action; > + > + xe_sysctrl_create_command(&command, XE_SYSCTRL_GROUP_GFSP, XE_SYSCTRL_CMD_PAGE_OFFLINE, > + &request, sizeof(request), &response, sizeof(response)); > + > + ret = xe_sysctrl_send_command(&xe->sc, &command, &rlen); > + if (ret) { > + xe_log_err_fatal(xe, SYSCTRL, ret, "failed to send page offline command\n"); > + return ret; > + } > + > + if (rlen != sizeof(response)) { > + xe_log_err(xe, SYSCTRL, -EINVAL, > + "unexpected page offline response length %zu (expected %zu)\n", > + rlen, sizeof(response)); > + return -EINVAL; > + } > + > + ret = ras_status_to_errno(response.status); > + if (ret) { > + xe_log_err(xe, SYSCTRL, ret, "page offline command failed with status %u\n", > + response.status); > + return ret; > + } > + > + return ret; > +} > + > +static int handle_page_offline(struct xe_device *xe, u64 page_address, bool send_cmd) > +{ > + enum xe_ras_page_action action; > + int ret = 0; > + > + if (!IS_ALIGNED(page_address, XE_PAGE_SIZE)) { > + xe_log_err(xe, SYSCTRL, -EINVAL, "Unaligned physical page address: 0x%llx\n", > + page_address); > + return -EINVAL; > + } > + > + /* > + * TODO: Call function to handle address fault > + * ret = xe_ttm_vram_handle_addr_fault(xe, page_address); > + */ > + > + /* > + * Handle return code from address fault handling function: > + * 0: Page soft offlined, decline to firmware > + * -EIO: Address belongs to a critical BO/stolen area that cannot be offlined > + * -EOPNOTSUPP: Address is valid and can be offlined but user policy is not to offline > + * -EXIST: Address is soft offlined but yet to be offlined by firmware for second occurrence nit: -EEXIST? > + */ > + > + switch (ret) { > + case 0: > + action = XE_RAS_PAGE_ACTION_DECLINE; > + xe_log_err(xe, DEVICE_MEMORY, 0, I know, it's switch case using ret, please make use of errno as ret from xe_ttm_vram_handle_addr_fault (). and make it consistency across all below xe_log_err. > + "Poison detected at physical address 0x%llx, page software offlined\n", > + page_address); > + break; > + /* User policy set to decline page offlining */ > + case -EOPNOTSUPP: > + action = XE_RAS_PAGE_ACTION_DECLINE; > + break; > + case -EIO: > + xe_log_err(xe, DEVICE_MEMORY, -EIO, > + "Physical page address belongs to critical BO: 0x%llx\n", page_address); > + return ret; > + case -EEXIST: > + action = XE_RAS_PAGE_ACTION_OFFLINE; > + xe_log_err(xe, DEVICE_MEMORY, -EEXIST, > + "Double-bit ECC error detected at physical address 0x%llx, page already software offlined\n", > + page_address); > + break; > + default: > + xe_log_err_fatal(xe, DEVICE_MEMORY, ret, "Failed to handle address fault 0x%llx\n", > + page_address); > + return 0; hmm, In default case, we need to use return ret; right? > + } > + > + if (send_cmd) { > + ret = send_page_offline_cmd(xe, page_address, action); > + if (ret) > + xe_log_err_fatal(xe, SYSCTRL, ret, > + "Failed to offline page for physical address 0x%llx\n", > + page_address); > + return ret; 'return ret' should be inside {} > + } > + > + return 0; > +} > + > static bool ras_counter_is_valid(struct xe_device *xe, struct xe_ras_error_class *counter) > { > u8 severity = counter->common.severity; > @@ -367,11 +477,12 @@ static u8 handle_soc_internal_errors(struct xe_device *xe, struct xe_ras_error_a > static u8 handle_device_memory_errors(struct xe_device *xe, struct xe_ras_error_array *arr) > { > struct xe_ras_memory_error *info = (void *)arr->details; > + int ret; > > /* > * For memory errors, the recovery action depends on the error category > * > - * TODO: Double-bit ECC errors: Page offlining > + * Double-bit ECC errors: Page offlining > * Poison and data parity errors: Log only > * For any other memory errors, request a reset as recovery mechanism > */ > @@ -383,10 +494,10 @@ static u8 handle_device_memory_errors(struct xe_device *xe, struct xe_ras_error_ > xe_info(xe, "[RAS]: Data parity error detected\n"); > break; > case XE_RAS_MEMORY_DB_ECC: > - xe_info(xe, "[RAS]: Double-bit ECC error detected at sw address 0x%llx\n", > - info->sw_address); > - /* TODO: Add page offlining for Double-bit ECC error */ > - fallthrough; > + ret = handle_page_offline(xe, info->sw_address, true); In case of soft page offline, we need not required reset right? however when send_page_offline_cmd return err that case, reset is required. is that correct behavior? Thanks, -/Mallesh > + if (ret) > + return XE_RAS_RECOVERY_ACTION_RESET; > + break; > default: > return XE_RAS_RECOVERY_ACTION_RESET; > } > diff --git a/drivers/gpu/drm/xe/xe_ras_types.h b/drivers/gpu/drm/xe/xe_ras_types.h > index 99b2466e2062..2fac968879b6 100644 > --- a/drivers/gpu/drm/xe/xe_ras_types.h > +++ b/drivers/gpu/drm/xe/xe_ras_types.h > @@ -17,6 +17,19 @@ > #define XE_RAS_MEMORY_POISON BIT(2) > #define XE_RAS_MEMORY_DATA_PARITY BIT(5) > > +/** > + * enum xe_ras_page_action - Page offline actions for page offline request > + * > + * @XE_RAS_PAGE_ACTION_OFFLINE: Instruct firmware to offline the page > + * @XE_RAS_PAGE_ACTION_DECLINE: Instruct firmware to remove the page from queue > + * @XE_RAS_PAGE_ACTION_MAX: Max value > + */ > +enum xe_ras_page_action { > + XE_RAS_PAGE_ACTION_OFFLINE, > + XE_RAS_PAGE_ACTION_DECLINE, > + XE_RAS_PAGE_ACTION_MAX > +}; > + > /** > * enum xe_ras_recovery_action - RAS recovery actions > * > @@ -245,6 +258,28 @@ struct xe_ras_memory_error { > u32 reserved2[10]; > } __packed; > > +/** > + * struct xe_ras_page_offline_request - Request for page offline command > + */ > +struct xe_ras_page_offline_request { > + /** @page_address: Page address (4KB aligned) */ > + u64 page_address; > + /** @action: Action to be performed, see &enum xe_ras_page_action */ > + u32 action; > + /** @reserved: Reserved for future use */ > + u32 reserved; > +} __packed; > + > +/** > + * struct xe_ras_page_offline_response - Response from page offline command > + */ > +struct xe_ras_page_offline_response { > + /** @status: Status of the page offline request */ > + u32 status; > + /** @reserved: Reserved for future use */ > + u32 reserved; > +} __packed; > + > /** > * struct xe_ras_get_health_request - Request structure for obtaining gpu health > */ > diff --git a/drivers/gpu/drm/xe/xe_sysctrl_mailbox_types.h b/drivers/gpu/drm/xe/xe_sysctrl_mailbox_types.h > index d0341538ad05..3363f48da2b7 100644 > --- a/drivers/gpu/drm/xe/xe_sysctrl_mailbox_types.h > +++ b/drivers/gpu/drm/xe/xe_sysctrl_mailbox_types.h > @@ -26,6 +26,7 @@ enum xe_sysctrl_group { > * @XE_SYSCTRL_CMD_GET_COUNTER: Get error counter value > * @XE_SYSCTRL_CMD_CLEAR_COUNTER: Clear error counter value > * @XE_SYSCTRL_CMD_GET_PENDING_EVENT: Retrieve pending event > + * @XE_SYSCTRL_CMD_PAGE_OFFLINE: Instruct firmware to offline/decline a page > * @XE_SYSCTRL_CMD_GET_HEALTH: Retrieve gpu health > * @XE_SYSCTRL_CMD_SET_HEALTH: Set gpu health > */ > @@ -34,6 +35,7 @@ enum xe_sysctrl_gfsp_cmd { > XE_SYSCTRL_CMD_GET_COUNTER = 0x03, > XE_SYSCTRL_CMD_CLEAR_COUNTER = 0x04, > XE_SYSCTRL_CMD_GET_PENDING_EVENT = 0x07, > + XE_SYSCTRL_CMD_PAGE_OFFLINE = 0x08, > XE_SYSCTRL_CMD_GET_HEALTH = 0x0B, > XE_SYSCTRL_CMD_SET_HEALTH = 0x0C, > }; --------------if3MC01eUyHM9gCt1BCNyua1 Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: 8bit


On 25-08-2026 12:06 pm, Riana Tauro wrote:
This will be integrated with the related address-fault handling flow
once this patch is merged.
https://lore.kernel.org/intel-xe/20260818104055.3833974-14-tejas.upadhyay@intel.com/
Sending for initial comments.

Add basic support for sending page offline/decline requests to system
controller and use it for device memory ECC error handling.
Pages that belong to critical BOs cannot be handled by offlining and
require a SBR (Secondary Bus Reset).
Pages that are configured for log-only handling are not marked as bad by
firmware.

For all other valid page addresses, the first occurrence of error
indicates a poison error and the page is offlined only by software.
Firmware avoids permanently marking the page as bad. The second occurrence
of an error indicates a Double-bit ECC error and the firmware
permanently marks the page as bad.

Cc: Tejas Upadhyay <tejas.upadhyay@intel.com>
Cc: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Signed-off-by: Riana Tauro <riana.tauro@intel.com>
---
 drivers/gpu/drm/xe/xe_ras.c                   | 121 +++++++++++++++++-
 drivers/gpu/drm/xe/xe_ras_types.h             |  35 +++++
 drivers/gpu/drm/xe/xe_sysctrl_mailbox_types.h |   2 +
 3 files changed, 153 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_ras.c b/drivers/gpu/drm/xe/xe_ras.c
index d25d25f77531..c643c7137a42 100644
--- a/drivers/gpu/drm/xe/xe_ras.c
+++ b/drivers/gpu/drm/xe/xe_ras.c
@@ -3,6 +3,7 @@
  * Copyright © 2026 Intel Corporation
  */
 
+#include "xe_bo.h"
 #include "xe_debugfs.h"
 #include "xe_device.h"
 #include "xe_drm_ras.h"
@@ -200,6 +201,115 @@ static inline const char *comp_to_str(u8 component)
 	return xe_ras_components[component];
 }
 
+static int send_page_offline_cmd(struct xe_device *xe, u64 page_address,
+				 enum xe_ras_page_action action)
+{
+	struct xe_sysctrl_mailbox_command command = {0};
+	struct xe_ras_page_offline_request request = {0};
+	struct xe_ras_page_offline_response response = {0};
+	size_t rlen;
+	int ret;
+
+	if (!xe->info.has_sysctrl)
+		return 0;
+
+	if (action >= XE_RAS_PAGE_ACTION_MAX) {
+		xe_log_err(xe, DEVICE_MEMORY, -EINVAL, "Invalid page offline action %d\n", action);
+		return -EINVAL;
+	}
+
+	request.page_address = page_address;
+	request.action = action;
+
+	xe_sysctrl_create_command(&command, XE_SYSCTRL_GROUP_GFSP, XE_SYSCTRL_CMD_PAGE_OFFLINE,
+				  &request, sizeof(request), &response, sizeof(response));
+
+	ret = xe_sysctrl_send_command(&xe->sc, &command, &rlen);
+	if (ret) {
+		xe_log_err_fatal(xe, SYSCTRL, ret, "failed to send page offline command\n");
+		return ret;
+	}
+
+	if (rlen != sizeof(response)) {
+		xe_log_err(xe, SYSCTRL, -EINVAL,
+			   "unexpected page offline response length %zu (expected %zu)\n",
+			   rlen, sizeof(response));
+		return -EINVAL;
+	}
+
+	ret = ras_status_to_errno(response.status);
+	if (ret) {
+		xe_log_err(xe, SYSCTRL, ret, "page offline command failed with status %u\n",
+			   response.status);
+		return ret;
+	}
+
+	return ret;
+}
+
+static int handle_page_offline(struct xe_device *xe, u64 page_address, bool send_cmd)
+{
+	enum xe_ras_page_action action;
+	int ret = 0;
+
+	if (!IS_ALIGNED(page_address, XE_PAGE_SIZE)) {
+		xe_log_err(xe, SYSCTRL, -EINVAL, "Unaligned physical page address: 0x%llx\n",
+			   page_address);
+		return -EINVAL;
+	}
+
+	/*
+	 * TODO: Call function to handle address fault
+	 * ret = xe_ttm_vram_handle_addr_fault(xe, page_address);
+	 */
+
+	/*
+	 * Handle return code from address fault handling function:
+	 *  0: Page soft offlined, decline to firmware
+	 * -EIO: Address belongs to a critical BO/stolen area that cannot be offlined
+	 * -EOPNOTSUPP: Address is valid and can be offlined but user policy is not to offline
+	 * -EXIST: Address is soft offlined but yet to be offlined by firmware for second occurrence
nit: -EEXIST? 
+	 */
+
+	switch (ret) {
+	case 0:
+		action = XE_RAS_PAGE_ACTION_DECLINE;
+		xe_log_err(xe, DEVICE_MEMORY, 0,

I know, it's switch case using ret, please make use of errno as ret from xe_ttm_vram_handle_addr_fault ().

and make it consistency across all below xe_log_err.

+			   "Poison detected at physical address 0x%llx, page software offlined\n",
+			   page_address);
+		break;
+	/* User policy set to decline page offlining */
+	case -EOPNOTSUPP:
+		action = XE_RAS_PAGE_ACTION_DECLINE;
+		break;
+	case -EIO:
+		xe_log_err(xe, DEVICE_MEMORY, -EIO,
+			   "Physical page address belongs to critical BO: 0x%llx\n", page_address);
+		return ret;
+	case -EEXIST:
+		action = XE_RAS_PAGE_ACTION_OFFLINE;
+		xe_log_err(xe, DEVICE_MEMORY, -EEXIST,
+			   "Double-bit ECC error detected at physical address 0x%llx, page already software offlined\n",
+			   page_address);
+		break;
+	default:
+		xe_log_err_fatal(xe, DEVICE_MEMORY, ret, "Failed to handle address fault 0x%llx\n",
+				 page_address);
+		return 0;
hmm, In default case, we need to use return ret; right? 
+	}
+
+	if (send_cmd) {
+		ret = send_page_offline_cmd(xe, page_address, action);
+		if (ret)
+			xe_log_err_fatal(xe, SYSCTRL, ret,
+					 "Failed to offline page for physical address 0x%llx\n",
+					 page_address);
+		return ret;
'return ret' should be inside {}  
+	}
+
+	return 0;
+}
+
 static bool ras_counter_is_valid(struct xe_device *xe, struct xe_ras_error_class *counter)
 {
 	u8 severity = counter->common.severity;
@@ -367,11 +477,12 @@ static u8 handle_soc_internal_errors(struct xe_device *xe, struct xe_ras_error_a
 static u8 handle_device_memory_errors(struct xe_device *xe, struct xe_ras_error_array *arr)
 {
 	struct xe_ras_memory_error *info = (void *)arr->details;
+	int ret;
 
 	/*
 	 * For memory errors, the recovery action depends on the error category
 	 *
-	 * TODO: Double-bit ECC errors: Page offlining
+	 * Double-bit ECC errors: Page offlining
 	 * Poison and data parity errors: Log only
 	 * For any other memory errors, request a reset as recovery mechanism
 	 */
@@ -383,10 +494,10 @@ static u8 handle_device_memory_errors(struct xe_device *xe, struct xe_ras_error_
 		xe_info(xe, "[RAS]: Data parity error detected\n");
 		break;
 	case XE_RAS_MEMORY_DB_ECC:
-		xe_info(xe, "[RAS]: Double-bit ECC error detected at sw address 0x%llx\n",
-			info->sw_address);
-		/* TODO: Add page offlining for Double-bit ECC error */
-		fallthrough;
+		ret = handle_page_offline(xe, info->sw_address, true);

In case of soft page offline, we need not required reset right? however when send_page_offline_cmd

return err that case, reset is required. is that correct behavior?

Thanks,

-/Mallesh

+		if (ret)
+			return XE_RAS_RECOVERY_ACTION_RESET;
+		break;
 	default:
 		return XE_RAS_RECOVERY_ACTION_RESET;
 	}
diff --git a/drivers/gpu/drm/xe/xe_ras_types.h b/drivers/gpu/drm/xe/xe_ras_types.h
index 99b2466e2062..2fac968879b6 100644
--- a/drivers/gpu/drm/xe/xe_ras_types.h
+++ b/drivers/gpu/drm/xe/xe_ras_types.h
@@ -17,6 +17,19 @@
 #define XE_RAS_MEMORY_POISON			BIT(2)
 #define XE_RAS_MEMORY_DATA_PARITY		BIT(5)
 
+/**
+ * enum xe_ras_page_action - Page offline actions for page offline request
+ *
+ * @XE_RAS_PAGE_ACTION_OFFLINE: Instruct firmware to offline the page
+ * @XE_RAS_PAGE_ACTION_DECLINE: Instruct firmware to remove the page from queue
+ * @XE_RAS_PAGE_ACTION_MAX: Max value
+ */
+enum xe_ras_page_action {
+	XE_RAS_PAGE_ACTION_OFFLINE,
+	XE_RAS_PAGE_ACTION_DECLINE,
+	XE_RAS_PAGE_ACTION_MAX
+};
+
 /**
  * enum xe_ras_recovery_action - RAS recovery actions
  *
@@ -245,6 +258,28 @@ struct xe_ras_memory_error {
 	u32 reserved2[10];
 } __packed;
 
+/**
+ * struct xe_ras_page_offline_request - Request for page offline command
+ */
+struct xe_ras_page_offline_request {
+	/** @page_address: Page address (4KB aligned) */
+	u64 page_address;
+	/** @action: Action to be performed, see &enum xe_ras_page_action */
+	u32 action;
+	/** @reserved: Reserved for future use */
+	u32 reserved;
+} __packed;
+
+/**
+ * struct xe_ras_page_offline_response - Response from page offline command
+ */
+struct xe_ras_page_offline_response {
+	/** @status: Status of the page offline request */
+	u32 status;
+	/** @reserved: Reserved for future use */
+	u32 reserved;
+} __packed;
+
 /**
  * struct xe_ras_get_health_request - Request structure for obtaining gpu health
  */
diff --git a/drivers/gpu/drm/xe/xe_sysctrl_mailbox_types.h b/drivers/gpu/drm/xe/xe_sysctrl_mailbox_types.h
index d0341538ad05..3363f48da2b7 100644
--- a/drivers/gpu/drm/xe/xe_sysctrl_mailbox_types.h
+++ b/drivers/gpu/drm/xe/xe_sysctrl_mailbox_types.h
@@ -26,6 +26,7 @@ enum xe_sysctrl_group {
  * @XE_SYSCTRL_CMD_GET_COUNTER: Get error counter value
  * @XE_SYSCTRL_CMD_CLEAR_COUNTER: Clear error counter value
  * @XE_SYSCTRL_CMD_GET_PENDING_EVENT: Retrieve pending event
+ * @XE_SYSCTRL_CMD_PAGE_OFFLINE: Instruct firmware to offline/decline a page
  * @XE_SYSCTRL_CMD_GET_HEALTH: Retrieve gpu health
  * @XE_SYSCTRL_CMD_SET_HEALTH: Set gpu health
  */
@@ -34,6 +35,7 @@ enum xe_sysctrl_gfsp_cmd {
 	XE_SYSCTRL_CMD_GET_COUNTER		= 0x03,
 	XE_SYSCTRL_CMD_CLEAR_COUNTER		= 0x04,
 	XE_SYSCTRL_CMD_GET_PENDING_EVENT	= 0x07,
+	XE_SYSCTRL_CMD_PAGE_OFFLINE             = 0x08,
 	XE_SYSCTRL_CMD_GET_HEALTH		= 0x0B,
 	XE_SYSCTRL_CMD_SET_HEALTH		= 0x0C,
 };
--------------if3MC01eUyHM9gCt1BCNyua1--