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 6985BC5B543 for ; Thu, 29 May 2025 22:42:47 +0000 (UTC) Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id 2C9F410E7A3; Thu, 29 May 2025 22:42:42 +0000 (UTC) Authentication-Results: gabe.freedesktop.org; dkim=pass (2048-bit key; unprotected) header.d=intel.com header.i=@intel.com header.b="Umupf7wJ"; dkim-atps=neutral Received: from mgamail.intel.com (mgamail.intel.com [192.198.163.9]) by gabe.freedesktop.org (Postfix) with ESMTPS id 6095810E7A4 for ; Thu, 29 May 2025 22:42:40 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1748558560; x=1780094560; h=date:from:to:cc:subject:message-id:references: content-transfer-encoding:in-reply-to:mime-version; bh=016lwPvGI7E740p3gj9IuTDWovXcDtMp/E4FsUkqEnI=; b=Umupf7wJgsPL6gQHFEjuh8cWrRG56IwEd3QDPxUMgG48mt7NS9+IzOfw +w7xw0nT0KPf8hc3q1CRYxDolEMpJLTB860qMPAcqyvkrsEgayyl5398c ZyTZZxWduCDqNQ/af8BN66KinB6dWesyreydDJv9mb1Y/XuiNH6iPfAcc hkJXJMqGNJ6CC2/3IEgHwkYZi0N+iTDY5x/l5G5JN8A8bqNhmuyNaJv7i lhBn9xiQWm4/XmQQcMnYT216p96NRt4uh57AdB7eNI7qaWdZHz6FW8C4B zd7yiJD0EAF9Zzxl0GFqhLN/mY+tGBxv9hMNysGW5zC8e5mWOfXkwm6iM A==; X-CSE-ConnectionGUID: e4fm6tsPRqqrSXy+KtFlwQ== X-CSE-MsgGUID: P00plUyuSA+D6x/0PbFSRA== X-IronPort-AV: E=McAfee;i="6700,10204,11448"; a="61305286" X-IronPort-AV: E=Sophos;i="6.16,194,1744095600"; d="scan'208";a="61305286" Received: from fmviesa001.fm.intel.com ([10.60.135.141]) by fmvoesa103.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 29 May 2025 15:42:39 -0700 X-CSE-ConnectionGUID: Q3Qd2hM8Si+jZ+6vF+toHw== X-CSE-MsgGUID: KtZvKRW6QqW8b9b7OoQo3g== X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="6.16,194,1744095600"; d="scan'208";a="174590082" Received: from orsmsx901.amr.corp.intel.com ([10.22.229.23]) by fmviesa001.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 29 May 2025 15:42:34 -0700 Received: from ORSMSX901.amr.corp.intel.com (10.22.229.23) by ORSMSX901.amr.corp.intel.com (10.22.229.23) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.2.1544.25; Thu, 29 May 2025 15:42:32 -0700 Received: from orsedg603.ED.cps.intel.com (10.7.248.4) by ORSMSX901.amr.corp.intel.com (10.22.229.23) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.2.1544.25 via Frontend Transport; Thu, 29 May 2025 15:42:32 -0700 Received: from NAM10-BN7-obe.outbound.protection.outlook.com (40.107.92.61) by edgegateway.intel.com (134.134.137.100) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.1.2507.55; Thu, 29 May 2025 15:42:31 -0700 ARC-Seal: i=1; a=rsa-sha256; s=arcselector10001; d=microsoft.com; cv=none; b=B2HMPIO+7gRrwRrPrZb1rOTEPLZSNG4NdsbJG2b7ARwfoga5Ol/Tu3iX88W3hVYT3eb4kbtVhueGHf3EaazUfCOQ7C+bg978JeTPeAA1ipvGjSMIydipWYC3jC4W6huuOIY8hMz/qUIIYQRIHZN50g9sci7H0xCep63Cs5l75ZRx2PXyiUpK+levdh5GfQbmWdrMnPQntL7ET6lYa+dar2081roTCQOO8p0tP8iwcGmDgumwFOO90gwLAIIGYoRRJkroVBFQGsoo3JiDCvT6h5HyX52iNsuhCfxTBahkQonJxSSphVcZ9XDF6oQ7fwHJhLODXhKiPJBNkieK0cWOIQ== 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=y0ZCMe9LPRD8iNBH7nUdJ3gd+l1bXr3WpLL5twQZZKk=; b=pSqutGDSTiQSa+d0g8WHRga/TaTbtN2XuEausMVF8oXuPzDmyC0i8guNJGj+xRN/gcdVh4bVrQU9IpHGhqD9xgDSovuoTme2mOXsaHrwtJzhAxTUa2i2XBhx9HWDIuci3Y3lmXTpOog3cWXTG5uyyRtdyvmY7o1smvpOh/rpwVCkkEt5t/6lOJBbKXYA8ydZdxy7MSEsDzhREYkeTuiWwFRmnoQPvzwnbe6hkLxUeSAF+OnNBZfLzDIsbk1PsrEThzuu+QHZsaUi2eH9Z9kQzwfSVtRLKnFBztHSP13dOKsb3td/wjT3J5xXjjyPfuTT13fstOFeDZMqYwJT+0K0rA== 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 PH7PR11MB6522.namprd11.prod.outlook.com (2603:10b6:510:212::12) by MW4PR11MB6569.namprd11.prod.outlook.com (2603:10b6:303:1e1::14) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.20.8769.24; Thu, 29 May 2025 22:42:10 +0000 Received: from PH7PR11MB6522.namprd11.prod.outlook.com ([fe80::9e94:e21f:e11a:332]) by PH7PR11MB6522.namprd11.prod.outlook.com ([fe80::9e94:e21f:e11a:332%5]) with mapi id 15.20.8769.029; Thu, 29 May 2025 22:42:10 +0000 Date: Thu, 29 May 2025 15:43:40 -0700 From: Matthew Brost To: Himal Prasad Ghimiray CC: Subject: Re: [PATCH v3 10/19] drm/xe: Implement madvise ioctl for xe Message-ID: References: <20250527164003.1068118-1-himal.prasad.ghimiray@intel.com> <20250527164003.1068118-11-himal.prasad.ghimiray@intel.com> Content-Type: text/plain; charset="utf-8" Content-Disposition: inline Content-Transfer-Encoding: 8bit In-Reply-To: <20250527164003.1068118-11-himal.prasad.ghimiray@intel.com> X-ClientProxiedBy: BN9PR03CA0219.namprd03.prod.outlook.com (2603:10b6:408:f8::14) To PH7PR11MB6522.namprd11.prod.outlook.com (2603:10b6:510:212::12) MIME-Version: 1.0 X-MS-PublicTrafficType: Email X-MS-TrafficTypeDiagnostic: PH7PR11MB6522:EE_|MW4PR11MB6569:EE_ X-MS-Office365-Filtering-Correlation-Id: 5304215a-3202-4a13-c195-08dd9f020c36 X-MS-Exchange-SenderADCheck: 1 X-MS-Exchange-AntiSpam-Relay: 0 X-Microsoft-Antispam: BCL:0;ARA:13230040|376014|366016|1800799024; X-Microsoft-Antispam-Message-Info: =?utf-8?B?ZW0xRGVqSFF0Zm5QdGFHYysvTEhEV3BXSitCTjd4R3BTa1lTam15U1lDc2w5?= =?utf-8?B?OU9hWUsvMVFMVjJNKzhBL29aMk5ZTDdWb2V3UGpsMytqYmJRVm9JZWRPbnNj?= =?utf-8?B?OFdtNDZHVlRtMmJvNVdCdm5DRWI2MXJlckdEcDdzL21pTjkwZ2JJaWtjRUJW?= =?utf-8?B?a280WGFYWTg5N0RVdWNDNk5wekpXeThKbmliUVYrM3ZWSU0xY2RGRXZlTzJo?= =?utf-8?B?UHFaS1B1cDNxa1NOT3lxMkc3cUlQT1VmbjB6b2FRWHZ5Mk9uZUZLNUh4RldB?= =?utf-8?B?bldLR3dwM3JMM2ZKWTV3aVd6eE14NDV4QjM0OGhVekZvV1lZM2xxczZxc1JK?= =?utf-8?B?dGVzcEtuV2RjMHBTYU5MNmgvNUwyaEF1Z1J2cDM1TC9TWWJ1bGVnRUtlSXpO?= =?utf-8?B?eFRHZHl5NExCVWR6TEpBeHhHRjRrQUpuUUxMNGVMR0pxYkJwY1ZLazJUWWsz?= =?utf-8?B?VTFmTWVMQ2Rtc0puMUVqQkZjZlRoZERiWFYrclpHcHBNanNCUEFIUy9vSzc0?= =?utf-8?B?aXpoUDRGQ0YzUDJ5N0Fwa1NYU3pIMzNhRHRtbVdpR2o0dWxxYjlzN1I4OUVk?= =?utf-8?B?REVZNlF4WVMyY0dTOW9lOG5lOUY1ZGlpRitoU1o1TjVwSXA5ZnA2ajJsTGNO?= =?utf-8?B?UmZKZExGamlQb09MZEoyUStBdklFTW5DSnc1cFNEZXI1VnllQndtK0NVTm56?= =?utf-8?B?MHhOUktxblh3Z29LdEVVUUJTNTUrU1dDQ2UwblEzbXllSDVjbHBueFYvWlF2?= =?utf-8?B?OXAydmR4L09uM3U4ZytRL2ZHWXR0ekE1MHUrZDZXY3phSFM5RHRUcHQ0bVR1?= =?utf-8?B?TDNMYWw3UmhDMXV6ZDBrSmh6YW1jd2c4Z1lvMHJMRWx2ZmVHeVZzV2ZyQ2xS?= =?utf-8?B?QklTZ2tFUHBJaHd6L05ZdTFUd2gzMXQvbDZ4eE5BVVAvdW5reEE1R3FqUEFm?= =?utf-8?B?WndCRjArcGNJWWM2V0xYQWNURDJLMGhPeStwRDRUVXo3WUJMUjFzdmJISkpk?= =?utf-8?B?OVA1TWZrbDdEK1ZwenZlbTFWUnhGbmdFb0FyRW8vYkJtZGxLSmJNb1p6SkYy?= =?utf-8?B?K3dHcm9BY0V3UFBEbzdGUjJHSVBDTGZkWlkxN0llUU5pb2xvNUd4Q3Z3RFY3?= =?utf-8?B?MkNJZTFnWVFaYlBreXU2TVhEaTNoaXBjUVRnaHNBdHlPM3JGZmJJZ0dKQ3Vs?= =?utf-8?B?RE1ZU1B2RVZxTzQ1OGFDYjFsZ2xSQWhEdDVieEZPZ1VyRUJjandQRHM2akpl?= =?utf-8?B?WHlkTlpqZkxweVVSRm54M0VWK1QvT0hFMU9CMEtGSXdZMDRDZEdrOXRzenNj?= =?utf-8?B?dzhPV1BVZVpQSFZpeGJUQ3lMY25XenZqOWxYT0Yrb1FZcGRESGYyMDltbUI0?= =?utf-8?B?cS90a3l6aWxUcmduZGFtdjJTWjgwQ3VpdWxOTEVWaWxoY3VNS01zNWEzSUMz?= =?utf-8?B?Q2lDSjRsVE5TVjBHb3dBTDdPRXFnQlZrQit1Skl6dURGa1ZJR2NlRm92aTB5?= =?utf-8?B?QVU1UGsvQ0J3Q1poT3VBdlFmdzhDbDZUQ0pnS1BkZ2xFUkZ6WHRwMDRmYnlk?= =?utf-8?B?RTBXc3NTWEllZFhrQ1UrV0M4S1p2M2Jqczc0bHNHN3dHTXJQYXZnUkIxS2xN?= =?utf-8?B?VWZYTjRVdC85Y2x1Z1hzNjV4UjExQ0xBQXN0by9LWk9RMlZuZy9kRjRaQ0NY?= =?utf-8?B?VTlXQ2JlMEh0aUo2Z2gzRWFQSVk1WW52eEU2Q0JuMXlCd29TQzVLRmRhWmRR?= =?utf-8?B?MVJnWE9YWWhvOVJVaGp6RDZtcDlaOHpHMWd3YUtzY2ZEZkRnVjFwZ0Q4Ukh4?= =?utf-8?Q?kO6xGszOwIVfOlFJYMlPBwNhZod+WsOWNN6/8=3D?= X-Forefront-Antispam-Report: CIP:255.255.255.255; CTRY:; LANG:en; SCL:1; SRV:; IPV:NLI; SFV:NSPM; H:PH7PR11MB6522.namprd11.prod.outlook.com; PTR:; CAT:NONE; SFS:(13230040)(376014)(366016)(1800799024); DIR:OUT; SFP:1101; X-MS-Exchange-AntiSpam-MessageData-ChunkCount: 1 X-MS-Exchange-AntiSpam-MessageData-0: =?utf-8?B?ZlRNUGNXbFlKRDRweThLQmNwSXlSZ21ML2dNVnpTK3ZBZm1mMUJXYmQrTmVR?= =?utf-8?B?bEtqcCtxUVpQQ2RySms2cG9ZRlFvUUptZEg5ZzBxRUFYcEoxRDlHM3Q0R0dp?= =?utf-8?B?NytQeXQvTTZpR3JoK3pMYThOZDlrTmE2K3FFdGRRdFZlaE13aXdSbXhvRUk4?= =?utf-8?B?T3Y4ZjgzMCtyQWozWkwyRk1XeGV0SmF4TDZ6QWhhY3ZOWVExVFZERmJ4N1lI?= =?utf-8?B?RlhNeGw2bi9VMGlSVGR2eDQxVjFIVXlrYjdQL0duRjM3QnZXUldOL3pKOEVR?= =?utf-8?B?Zm5WRVprYUtIc2hWTFErWm9wZWVzaU52cEN1SEZRd1QxallGZzAxWXRTUkpy?= =?utf-8?B?ei96dUxONHR0R0JDK2hWOTU0MXhURXFBQXRWV0VMWm03dWUzRXB6L0ZuTnh4?= =?utf-8?B?THJVbFRSQWJ4VDZPQldVUi9udnBYTEozYWtURklJakR2bkFNMm9BejBBK3J4?= =?utf-8?B?dXZ1eGdQQXN0UTB5ZU9QN3RrdXU2MFJBbEtLdWRNbU5TTkRJbGE2bUlZaUdI?= =?utf-8?B?OGI0NlB3WXRqdm5zRzhtRTdSbmVxNnZPQWE2dXlYTG12OTRxemFYa2F6TGwr?= =?utf-8?B?bHQ4dlZtMVdHdXVwMUkxRUg2T0ltMlc3K3ZDQ3ovVmFTZEdDWHM2R2lVZlFB?= =?utf-8?B?U1ZOajFuT0QySEVCeURpSjZMcFBybUI4cHZRS2trNGhDUXp6SlpickJadyt4?= =?utf-8?B?NXdVZmQ0R1VOQ05PODR2RndGeVM0eFliOUpIcHU4REEzOThGMzJ6MCt0Uzh6?= =?utf-8?B?ekd4NXZUQTVSQlQzVUswZXQ3RVpvbkw4WHRnU0xkYUdpc2cwRVhRelhsM2Rl?= =?utf-8?B?a1NGSmVYeGFyRkphS29iRHRZdFkzN3VtV0NmVE90NVRhR2RtTVpSRFpheWlG?= =?utf-8?B?MkZncGd5TXloam5raWI4b3QrSXN0aktUcUZuYmU0MXFUUWQxQTk5TEJIbUF1?= =?utf-8?B?aUJ0RzBITHUyZ1A2cnJzbE1IQkY0N2RFNnZ6UUtGMWNRL2p6bXA1bkVrTm03?= =?utf-8?B?MWdzUTN4VG5zbXRCRjBmTERMbWtFTXV4RUMrMEppZWZtK1J6VzFTTU81SzZ5?= =?utf-8?B?cWoxTnhPOTRYRUwvRGFqSjdPa0ZROEhPYkpNUGh4bEx0SHlPTWJ6Nit4Qjdv?= =?utf-8?B?NGhGMWpMYXZMVUtqRE94eTNkeGJhYThnY3I3TUNYcGo1ZU00bTQ2enhTUjUz?= =?utf-8?B?SDJzSzludmtBSnVHZDcvU1M5TDFtWkRRNjAybnF3WnVZU01ISU1uWG9rb1RG?= =?utf-8?B?YURJQXp6SzBzSk11UlpNOXBtY2lPd28vdTU4KzdyMnVlc09OVzdYMUZQemor?= =?utf-8?B?dzArNUpOWnFnVTYrbk1YTEhod3dUTVFITGtYc2cxaGRWcjlEQU1nQWtNRlRU?= =?utf-8?B?U2luK05QbHpUbnNzRnFGNG1GeHErbkIyV2luRFJDNnlOdzNhdllIMWVDb2dE?= =?utf-8?B?TWYyL0k5eDk1VE0rcGh0NkFaQ0YyNHN0N1FtOEFiNy83RXRCb1FWL2Y5UXZ1?= =?utf-8?B?TUR4NVB2L0xzNVFBMDIvaWgvQVY1R09qT2xMYXhRSlN3MDdCeVhLVW5yZVBh?= =?utf-8?B?TzhtOFVaQTM3OHZQNE1PRzVmVFQ0NUh3RWJ2WDg4cXRyRVhYdWJmYTFLb0xa?= =?utf-8?B?MnM2UjhPb2ZHMjNNZGFweXNkYVVuVTA4b3N3QmtwMjh1NDBjWWFsM2FFRHBS?= =?utf-8?B?K2l1SnBlV01RWGgxQUVSN0hISkxHT1FFd0VSdHRFZUJKR3o2eG5IWkFLQyt4?= =?utf-8?B?alhldnV6RXY4SDMwRGNibTQvRHRTUFFXUXVhZXVmVEhsSFdiQmJiUWluQTNS?= =?utf-8?B?am5GWm1IQXJoSVBwSXRJTG1JdmlXUjJ6Vmk1SUJEVXdaQXRQWlVWU25GdG1P?= =?utf-8?B?b2o4V3Vad2ZHOW1JUm82cXJ4aUhlSitWSTVWWXNoK3EvdHFrNjQwM0tqcTVH?= =?utf-8?B?QjdaRjVWYTdzb3M5UkkxMkpVZU9LRUFybkhiU0RIRTlmQ3NhQWd4eEpjM1Vl?= =?utf-8?B?MUJXT21NNDNKSGlDempvSmNzYnAzVUFId2k0M005OFFobXhHR0xiNlhlbGdi?= =?utf-8?B?SWJJVlRHTlVtSTg2Z3ZGMWNaN2t0dXBQVng1UWJRUG9VU3RoblBJV05ScXFj?= =?utf-8?B?bHIydWFzcGozVXVZdmw1QjJZN1hnWlN6bnpoM29XK3M3R20waTkwRDdzSkVv?= =?utf-8?B?WlE9PQ==?= X-MS-Exchange-CrossTenant-Network-Message-Id: 5304215a-3202-4a13-c195-08dd9f020c36 X-MS-Exchange-CrossTenant-AuthSource: PH7PR11MB6522.namprd11.prod.outlook.com X-MS-Exchange-CrossTenant-AuthAs: Internal X-MS-Exchange-CrossTenant-OriginalArrivalTime: 29 May 2025 22:42:10.3107 (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: uZVymyx0ClJiFzM7B9/iwc0JXuVa1L+c530lA0D4zIwaBMtg4A2j4N0vaFfnFlX397Ovi8L/ixNVtoC9X18V/A== X-MS-Exchange-Transport-CrossTenantHeadersStamped: MW4PR11MB6569 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 Tue, May 27, 2025 at 10:09:54PM +0530, Himal Prasad Ghimiray wrote: > This driver-specific ioctl enables UMDs to control the memory attributes > for GPU VMAs within a specified input range. If the start or end > addresses fall within an existing VMA, the VMA is split accordingly. The > attributes of the VMA are modified as provided by the users. The old > mappings of the VMAs are invalidated, and TLB invalidation is performed > if necessary. > > v2(Matthew brost) > - xe_vm_in_fault_mode can't be enabled by Mesa, hence allow ioctl in non > fault mode too > - fix tlb invalidation skip for same ranges in multiple op > - use helper for tlb invalidation > - use xe_svm_notifier_lock/unlock helper > - s/lockdep_assert_held/lockdep_assert_held_write > - Add kernel-doc > > Signed-off-by: Himal Prasad Ghimiray > --- > drivers/gpu/drm/xe/Makefile | 1 + > drivers/gpu/drm/xe/xe_device.c | 2 + > drivers/gpu/drm/xe/xe_vm_madvise.c | 264 +++++++++++++++++++++++++++++ > drivers/gpu/drm/xe/xe_vm_madvise.h | 15 ++ > 4 files changed, 282 insertions(+) > create mode 100644 drivers/gpu/drm/xe/xe_vm_madvise.c > create mode 100644 drivers/gpu/drm/xe/xe_vm_madvise.h > > diff --git a/drivers/gpu/drm/xe/Makefile b/drivers/gpu/drm/xe/Makefile > index c5d6681645ed..dc64bdcddfdc 100644 > --- a/drivers/gpu/drm/xe/Makefile > +++ b/drivers/gpu/drm/xe/Makefile > @@ -117,6 +117,7 @@ xe-y += xe_bb.o \ > xe_uc.o \ > xe_uc_fw.o \ > xe_vm.o \ > + xe_vm_madvise.o \ > xe_vram.o \ > xe_vram_freq.o \ > xe_vsec.o \ > diff --git a/drivers/gpu/drm/xe/xe_device.c b/drivers/gpu/drm/xe/xe_device.c > index d4b6e623aa48..b9791c614749 100644 > --- a/drivers/gpu/drm/xe/xe_device.c > +++ b/drivers/gpu/drm/xe/xe_device.c > @@ -61,6 +61,7 @@ > #include "xe_ttm_stolen_mgr.h" > #include "xe_ttm_sys_mgr.h" > #include "xe_vm.h" > +#include "xe_vm_madvise.h" > #include "xe_vram.h" > #include "xe_vsec.h" > #include "xe_wait_user_fence.h" > @@ -197,6 +198,7 @@ static const struct drm_ioctl_desc xe_ioctls[] = { > DRM_IOCTL_DEF_DRV(XE_WAIT_USER_FENCE, xe_wait_user_fence_ioctl, > DRM_RENDER_ALLOW), > DRM_IOCTL_DEF_DRV(XE_OBSERVATION, xe_observation_ioctl, DRM_RENDER_ALLOW), > + DRM_IOCTL_DEF_DRV(XE_MADVISE, xe_vm_madvise_ioctl, DRM_RENDER_ALLOW), > }; > > static long xe_drm_ioctl(struct file *file, unsigned int cmd, unsigned long arg) > diff --git a/drivers/gpu/drm/xe/xe_vm_madvise.c b/drivers/gpu/drm/xe/xe_vm_madvise.c > new file mode 100644 > index 000000000000..f7edefe5f6cf > --- /dev/null > +++ b/drivers/gpu/drm/xe/xe_vm_madvise.c > @@ -0,0 +1,264 @@ > +// SPDX-License-Identifier: MIT > +/* > + * Copyright © 2024 Intel Corporation > + */ > + > +#include "xe_vm_madvise.h" > + > +#include > +#include > +#include > + > +#include "xe_bo.h" > +#include "xe_gt_tlb_invalidation.h" > +#include "xe_pt.h" > +#include "xe_svm.h" > + > +static struct xe_vma **get_vmas(struct xe_vm *vm, int *num_vmas, > + u64 addr, u64 range) > +{ > + struct xe_vma **vmas, **__vmas; > + struct drm_gpuva *gpuva; > + int max_vmas = 8; > + > + lockdep_assert_held(&vm->lock); > + > + *num_vmas = 0; > + vmas = kmalloc_array(max_vmas, sizeof(*vmas), GFP_KERNEL); > + if (!vmas) > + return NULL; > + > + vm_dbg(&vm->xe->drm, "VMA's in range: start=0x%016llx, end=0x%016llx", addr, addr + range); > + > + drm_gpuvm_for_each_va_range(gpuva, &vm->gpuvm, addr, addr + range) { > + struct xe_vma *vma = gpuva_to_vma(gpuva); > + > + if (*num_vmas == max_vmas) { > + max_vmas <<= 1; > + __vmas = krealloc(vmas, max_vmas * sizeof(*vmas), GFP_KERNEL); > + if (!__vmas) { > + kfree(vmas); > + return NULL; > + } > + vmas = __vmas; > + } > + > + vmas[*num_vmas] = vma; > + (*num_vmas)++; > + } > + > + vm_dbg(&vm->xe->drm, "*num_vmas = %d\n", *num_vmas); > + > + if (!*num_vmas) { > + kfree(vmas); > + return NULL; > + } > + > + return vmas; > +} > + > +static int madvise_preferred_mem_loc(struct xe_device *xe, struct xe_vm *vm, > + struct xe_vma **vmas, int num_vmas, > + struct drm_xe_madvise_ops ops) > +{ > + /* Implementation pending */ > + return 0; > +} > + > +static int madvise_atomic(struct xe_device *xe, struct xe_vm *vm, > + struct xe_vma **vmas, int num_vmas, > + struct drm_xe_madvise_ops ops) > +{ > + /* Implementation pending */ > + return 0; > +} > + > +static int madvise_pat_index(struct xe_device *xe, struct xe_vm *vm, > + struct xe_vma **vmas, int num_vmas, > + struct drm_xe_madvise_ops ops) > +{ > + /* Implementation pending */ > + return 0; > +} > + > +static int madvise_purgeable_state(struct xe_device *xe, struct xe_vm *vm, > + struct xe_vma **vmas, int num_vmas, > + struct drm_xe_madvise_ops ops) > +{ > + /* Implementation pending */ > + return 0; > +} > + > +typedef int (*madvise_func)(struct xe_device *xe, struct xe_vm *vm, > + struct xe_vma **vmas, int num_vmas, struct drm_xe_madvise_ops ops); > + See my latest replies in patch #19, if possible, making these functions so they can't fail would be best. > +static const madvise_func madvise_funcs[] = { > + [DRM_XE_VMA_ATTR_PREFERRED_LOC] = madvise_preferred_mem_loc, > + [DRM_XE_VMA_ATTR_ATOMIC] = madvise_atomic, > + [DRM_XE_VMA_ATTR_PAT] = madvise_pat_index, > + [DRM_XE_VMA_ATTR_PURGEABLE_STATE] = madvise_purgeable_state, > +}; > + > +static void xe_zap_ptes_in_madvise_range(struct xe_vm *vm, u64 start, u64 end, u8 *tile_mask) > +{ > + struct drm_gpuva *gpuva; > + struct xe_tile *tile; > + u8 id; > + > + lockdep_assert_held_write(&vm->lock); > + > + if (dma_resv_wait_timeout(xe_vm_resv(vm), DMA_RESV_USAGE_BOOKKEEP, > + false, MAX_SCHEDULE_TIMEOUT) <= 0) > + XE_WARN_ON(1); > + > + *tile_mask = xe_svm_ranges_zap_ptes_in_range(vm, start, end); > + > + drm_gpuvm_for_each_va_range(gpuva, &vm->gpuvm, start, end) { > + struct xe_vma *vma = gpuva_to_vma(gpuva); > + > + if (xe_vma_is_cpu_addr_mirror(vma)) > + continue; > + > + if (xe_vma_is_userptr(vma)) { > + WARN_ON_ONCE(!mmu_interval_check_retry > + (&to_userptr_vma(vma)->userptr.notifier, > + to_userptr_vma(vma)->userptr.notifier_seq)); > + > + WARN_ON_ONCE(!dma_resv_test_signaled(xe_vm_resv(xe_vma_vm(vma)), > + DMA_RESV_USAGE_BOOKKEEP)); > + } > + I think the similar code in xe_vm_invalidate_vma is a bit stale and contains remnants from when we issued the userptr TLB invalidation outside of the notifier lock in the MMU notifier. There, it should be: if (xe_vma_is_userptr()) lockdep_assert(userptr notifier lock held); extra sanity checks So with that, you need the userptr notifier lock in read mode here. Also, I think the first WARN_ON is likely to always trigger here, as the CPU pages are likely valid, so that one should be removed. In xe_vm_invalidate_vma, the extra tests are protected by the PROVE_LOCKING Kconfig option. So if you keep the dma-resv check, I’d recommend hiding it behind PROVE_LOCKING. Note: speaking of the userptr notifier lock, Matt Auld has a patch series [1] to unify this with the GPU SVM notifier lock. If that lands before your changes, then you should use the SVM notifier lock in read mode for userptr here. [1] https://patchwork.freedesktop.org/series/146553/ > + if (xe_vma_bo(vma)) > + xe_bo_lock(xe_vma_bo(vma), false); > + Ah yes, you do need the BO lock here to prevent races from BO moves issuing a zap. Maybe we can update the xe_pt_zap_ptes lockdep assertions for userptr/BOs, along with some comments? We could do that independently of this series. I suggest this because I had forgotten how this worked in the previous revision. Maybe I can post this independently. > + for_each_tile(tile, vm->xe, id) { > + if (xe_pt_zap_ptes(tile, vma)) > + *tile_mask |= BIT(id); > + } > + > + if (xe_vma_bo(vma)) > + xe_bo_unlock(xe_vma_bo(vma)); > + } > +} > + > +static int xe_vm_invalidate_madvise_range(struct xe_vm *vm, u64 start, u64 end) > +{ > + u8 tile_mask = 0; > + > + xe_zap_ptes_in_madvise_range(vm, start, end, &tile_mask); > + if (!tile_mask) > + return 0; > + > + xe_device_wmb(vm->xe); > + > + return xe_vm_range_tilemask_tlb_invalidation(vm, start, end, tile_mask); > +} > + > +static int input_ranges_same(struct drm_xe_madvise_ops *old, > + struct drm_xe_madvise_ops *new) > +{ > + return (new->start == old->start && new->range == old->range); > +} > + > +/** > + * xe_vm_madvise_ioctl - Handle MADVise ioctl for a VM > + * @dev: DRM device pointer > + * @data: Pointer to ioctl data (drm_xe_madvise*) > + * @file: DRM file pointer > + * > + * Handles the MADVISE ioctl to provide memory advice for vma's within > + * input range. > + * > + * Return: 0 on success or a negative error code on failure. > + */ > +int xe_vm_madvise_ioctl(struct drm_device *dev, void *data, struct drm_file *file) > +{ > + struct xe_device *xe = to_xe_device(dev); > + struct xe_file *xef = to_xe_file(file); > + struct drm_xe_madvise_ops *advs_ops; > + struct drm_xe_madvise *args = data; > + struct xe_vm *vm; > + struct xe_vma **vmas = NULL; > + int num_vmas, err = 0; > + int i, j, attr_type; > + bool needs_invalidation; > + > + if (XE_IOCTL_DBG(xe, args->num_ops < 1)) > + return -EINVAL; > + > + vm = xe_vm_lookup(xef, args->vm_id); > + if (XE_IOCTL_DBG(xe, !vm)) > + return -EINVAL; > + > + down_write(&vm->lock); > + > + if (XE_IOCTL_DBG(xe, xe_vm_is_closed_or_banned(vm))) { > + err = -ENOENT; > + goto unlock_vm; > + } > + > + if (args->num_ops > 1) { > + u64 __user *madvise_user = u64_to_user_ptr(args->vector_of_ops); > + > + advs_ops = kvmalloc_array(args->num_ops, sizeof(struct drm_xe_madvise_ops), > + GFP_KERNEL | __GFP_ACCOUNT | > + __GFP_RETRY_MAYFAIL | __GFP_NOWARN); > + if (!advs_ops) { > + err = args->num_ops > 1 ? -ENOBUFS : -ENOMEM; > + goto unlock_vm; > + } > + > + err = __copy_from_user(advs_ops, madvise_user, > + sizeof(struct drm_xe_madvise_ops) * > + args->num_ops); > + if (XE_IOCTL_DBG(xe, err)) { > + err = -EFAULT; > + goto free_advs_ops; > + } > + } else { > + advs_ops = &args->ops; > + } > + See my reply in patch #19, I think we should validate user input ahead of the below loop rather than failing mid-loop on bad user input. > + for (i = 0; i < args->num_ops; i++) { > + xe_vm_alloc_madvise_vma(vm, advs_ops[i].start, advs_ops[i].range); > + > + vmas = get_vmas(vm, &num_vmas, advs_ops[i].start, advs_ops[i].range); > + if (!vmas) { > + err = -ENOMEM; > + goto free_advs_ops; > + } > + > + attr_type = array_index_nospec(advs_ops[i].type, ARRAY_SIZE(madvise_funcs)); > + err = madvise_funcs[attr_type](xe, vm, vmas, num_vmas, advs_ops[i]); > + > + kfree(vmas); > + vmas = NULL; > + > + if (err) > + goto free_advs_ops; > + } > + > + for (i = 0; i < args->num_ops; i++) { > + needs_invalidation = true; > + for (j = i + 1; j < args->num_ops; ++j) { > + if (input_ranges_same(&advs_ops[j], &advs_ops[i])) { > + needs_invalidation = false; > + break; > + } > + } I'd drop this extra check. The invalidation code already short ciruits on the tile_present / tile_invalidated bits in the range or VMA so I don't think an extra short circuit here buys us a ton. Matt > + if (needs_invalidation) { > + err = xe_vm_invalidate_madvise_range(vm, advs_ops[i].start, > + advs_ops[i].start + advs_ops[i].range); > + if (err) > + goto free_advs_ops; > + } > + } > + > +free_advs_ops: > + if (args->num_ops > 1) > + kvfree(advs_ops); > +unlock_vm: > + up_write(&vm->lock); > + xe_vm_put(vm); > + return err; > +} > diff --git a/drivers/gpu/drm/xe/xe_vm_madvise.h b/drivers/gpu/drm/xe/xe_vm_madvise.h > new file mode 100644 > index 000000000000..c5cdd058c322 > --- /dev/null > +++ b/drivers/gpu/drm/xe/xe_vm_madvise.h > @@ -0,0 +1,15 @@ > +/* SPDX-License-Identifier: MIT */ > +/* > + * Copyright © 2024 Intel Corporation > + */ > + > +#ifndef _XE_VM_MADVISE_H_ > +#define _XE_VM_MADVISE_H_ > + > +struct drm_device; > +struct drm_file; > + > +int xe_vm_madvise_ioctl(struct drm_device *dev, void *data, > + struct drm_file *file); > + > +#endif > -- > 2.34.1 >