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 DC818C41535 for ; Fri, 22 Dec 2023 20:03:13 +0000 (UTC) Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id A5B6E10E84B; Fri, 22 Dec 2023 20:03:13 +0000 (UTC) Received: from mgamail.intel.com (mgamail.intel.com [192.55.52.115]) by gabe.freedesktop.org (Postfix) with ESMTPS id BE83C10E849 for ; Fri, 22 Dec 2023 20:03:11 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1703275391; x=1734811391; h=message-id:date:subject:to:cc:references:from: in-reply-to:mime-version; bh=1EFBPCDuKUko7f6vmZ0W8F8D7iB/eXzRR2q/nPPwoNM=; b=OSG53Pfnt3TXRNbYI2EKHlkghqNho7ig2gkm8pKbDCoBi0Liu9BLq+S8 mA0LiDgEX0l5PJtBrROKUdFeX6BsPKKirMBDj61cyRjrBM5EC7bNFXO50 ur993qXoDRtIVRkuY8frZioZswktXxkfGRY+ZFBQGCvuV8R2tyx2+X+p5 QryhDd2gV8ZNpNBz7UTc2w3fu3V2RtLjpnEv/WcfkWYYLAKM3GjCf0R4g HxHJB9TBNzK6JG1Ze5yT/IyXUNheOo8W80dSwM9DIwkbVvpjkuag7LdPn VaKsQUmyua72yD1BR1MJWpwMZTOstZCq4AKS1oLLsabpGZgQNih9eG+aL g==; X-IronPort-AV: E=McAfee;i="6600,9927,10932"; a="395893942" X-IronPort-AV: E=Sophos;i="6.04,297,1695711600"; d="scan'208,217";a="395893942" Received: from fmsmga001.fm.intel.com ([10.253.24.23]) by fmsmga103.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 22 Dec 2023 12:03:10 -0800 X-ExtLoop1: 1 X-IronPort-AV: E=McAfee;i="6600,9927,10932"; a="920729635" X-IronPort-AV: E=Sophos;i="6.04,297,1695711600"; d="scan'208,217";a="920729635" Received: from orsmsx603.amr.corp.intel.com ([10.22.229.16]) by fmsmga001.fm.intel.com with ESMTP/TLS/AES256-GCM-SHA384; 22 Dec 2023 12:03:10 -0800 Received: from orsmsx611.amr.corp.intel.com (10.22.229.24) by ORSMSX603.amr.corp.intel.com (10.22.229.16) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256) id 15.1.2507.35; Fri, 22 Dec 2023 12:03:09 -0800 Received: from orsmsx610.amr.corp.intel.com (10.22.229.23) by ORSMSX611.amr.corp.intel.com (10.22.229.24) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256) id 15.1.2507.35; Fri, 22 Dec 2023 12:03:09 -0800 Received: from ORSEDG602.ED.cps.intel.com (10.7.248.7) by orsmsx610.amr.corp.intel.com (10.22.229.23) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256) id 15.1.2507.35 via Frontend Transport; Fri, 22 Dec 2023 12:03:09 -0800 Received: from NAM11-CO1-obe.outbound.protection.outlook.com (104.47.56.169) by edgegateway.intel.com (134.134.137.103) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.1.2507.35; Fri, 22 Dec 2023 12:03:09 -0800 ARC-Seal: i=1; a=rsa-sha256; s=arcselector9901; d=microsoft.com; cv=none; b=SNaLxToNAtk7Keof7bYBQwggG/cWTfurfk8UEiOfGTMuRr+cTm0gmRD3HhbEJjYfVf+YsjGt+7Grh0kAXzFp9YEXxOlXRvTgOh2M7GVXg251XOwvPF8bffVnntbXGtR0wnvaaeZ0lCf4vsdSvTLtqxMFemio1SGoNyQsQ+XrAZtdMOfKZ8tELCaiIcuQrcwbZpU4Q4hOzmv7LdIqYxTVMD+BcdPQPVGr8/zJeU4jHG2GtxHLIxxNsMtEpK/OiVFVgv9DCj5BU9pQs0Ovv0bIt4qkTsA5GrtGhHE6+oLjnOaI0A4+YZVnn5Wl1tpP007EszwEYunogVcDFmYLWCHXNg== 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=YR0qoxuwHyIcc5f5KiS4KR+yicUXIE1BwbgdtVCrrkM=; b=Vk7BUQY6wNk/GA/ytVJgJ61p9LR2N+j6MeiPQ/uX2eifjS4V+Gjszajs+SY4nP4cckK5b/qJ3ooIIYyBikpyCXH+fHGtkJA+fPad5/6UfRr7Yb1kY9tqAOOPeBxz5jmwHOw+uWPeBQd5jbC71AyJaebuPmKyVNVtTcTgojttgYSK7q0aotE8ewdC8/K2dwN0mhYOKwZtZn+Uh0uAviDVTWanS9jO7pe4FrUw4utAuRw93LUo7cw5eUYvWDZGIbwd/wSJAyH9MVfxk5NQMLAU0Qd5l3NnxdeaP9+xNJtXjoyzhCFaJA5YQAHgx0GJ5KgTU6cFSfUimVzeJtpu8y/t3g== 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 BY5PR11MB4274.namprd11.prod.outlook.com (2603:10b6:a03:1c1::23) by DM3PR11MB8684.namprd11.prod.outlook.com (2603:10b6:0:4a::21) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.20.7113.21; Fri, 22 Dec 2023 20:03:07 +0000 Received: from BY5PR11MB4274.namprd11.prod.outlook.com ([fe80::16a8:c3ae:ef16:3d13]) by BY5PR11MB4274.namprd11.prod.outlook.com ([fe80::16a8:c3ae:ef16:3d13%7]) with mapi id 15.20.7113.019; Fri, 22 Dec 2023 20:03:06 +0000 Content-Type: multipart/alternative; boundary="------------RmHfmYMiEnXfQ8ZBfzMzykdF" Message-ID: <112dd04d-5b84-4998-8800-522d9b6ca996@intel.com> Date: Fri, 22 Dec 2023 12:03:04 -0800 User-Agent: Mozilla Thunderbird Subject: Re: [PATCH v4 5/8] drm/xe/guc: Add PMU counter for total active ticks To: Riana Tauro , References: <20231222074602.817518-1-riana.tauro@intel.com> <20231222074602.817518-6-riana.tauro@intel.com> Content-Language: en-US From: "Belgaumkar, Vinay" In-Reply-To: <20231222074602.817518-6-riana.tauro@intel.com> X-ClientProxiedBy: SJ0PR05CA0143.namprd05.prod.outlook.com (2603:10b6:a03:33d::28) To BY5PR11MB4274.namprd11.prod.outlook.com (2603:10b6:a03:1c1::23) MIME-Version: 1.0 X-MS-PublicTrafficType: Email X-MS-TrafficTypeDiagnostic: BY5PR11MB4274:EE_|DM3PR11MB8684:EE_ X-MS-Office365-Filtering-Correlation-Id: b5753fbf-e2fb-424a-a581-08dc0329036b 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:255.255.255.255; CTRY:; LANG:en; SCL:1; SRV:; IPV:NLI; SFV:NSPM; H:BY5PR11MB4274.namprd11.prod.outlook.com; PTR:; CAT:NONE; SFS:(13230031)(136003)(346002)(396003)(39860400002)(366004)(376002)(230922051799003)(1800799012)(64100799003)(451199024)(186009)(316002)(30864003)(2906002)(2616005)(26005)(8936002)(6486002)(38100700002)(478600001)(83380400001)(66556008)(66476007)(66946007)(6512007)(53546011)(5660300002)(33964004)(6506007)(4326008)(8676002)(31686004)(82960400001)(41300700001)(36756003)(31696002)(86362001)(45980500001)(43740500002); DIR:OUT; SFP:1102; X-MS-Exchange-AntiSpam-MessageData-ChunkCount: 1 X-MS-Exchange-AntiSpam-MessageData-0: =?utf-8?B?b2o2cmxoR2hIWDZxY3BWOVN4Q1VYN0k1ZERiTGYwRnJJY1pEbTduaVhRUmVo?= =?utf-8?B?dFk0ZWcwc3AvRFZMUmF0V1hqNTlGZkdQTEhNUUs4ems2ZVVweTlsWG9zRzdW?= =?utf-8?B?eGFWanRSTEUvbWVtNTJxMnF3cFVLWFVzaXJVSGcrWFozYXNHU3RGc080Y2Fa?= =?utf-8?B?VnpRczhIcEZ0NUFvL3FURWM3YkppNWs2M3hwYnF6REQwREtaWmtBdDJ0Mkd3?= =?utf-8?B?N1REVFFaU2Q2Y2htMllCaml1QVFETVllWElVNGc5UVdkb20vLzM0NmlYQmlN?= =?utf-8?B?Q2E2R01kMHR2aGFzZnFLRmZDdVFod3VOdzBaVlpjdSt2Q2F6YktjZ1dNWkJu?= =?utf-8?B?MlRYZnliWjlaZ0Q4ZjRoWjdTa3Y2Z0NDcEtjUjNCN2Z0TFIrRDVCTGlOYzB6?= =?utf-8?B?cDUzOEJBbCtLemdnaGhyek5XR0dJTjVXUDY4OE8veVpLS3lMdzNyYzhrcTl1?= =?utf-8?B?V0lvNHBselBScGVzSVZwZlU4SEFvRmgzWTBVN3pzdVV3TFFXSVZMempRM0xx?= =?utf-8?B?bFpHUWRNM3pVcmdrSURNTGtrVGxJaHdBV2xzQk93QUhnYTZtWUh4QXlIdnBZ?= =?utf-8?B?STVUdjRINGFUYVR3ZG9RWXFBK2pXcDIyOW93U0QwVkJWQlg4bW5hODFJRERY?= =?utf-8?B?WWdONnM4OHNsN0pJdkROZy9PN0MxSm5FU0ZjQUxqSnhYcE95MlZYc2ZxN3FC?= =?utf-8?B?eGs3aXQrUm5FWEZ6M0RhMXhVcXg4QzVCUllJVERxa24rQVlaUEJYTXd2S01o?= =?utf-8?B?SmhYTGE4UkQ4T0ZTdTFUbktUMWxzbnlIQnJnbCtYem9pVkRPT1gwTFo5akQy?= =?utf-8?B?dWh5bFY0MmpEMEhmcFlnWXJkSFVIR1UxaVVCK1FrSk50MWFpM3BEUTRzd1Z5?= =?utf-8?B?NzBDWS9Sajg2QTVFT1htRmY0RDV0NC9iRXY4T0dJMC80ejQvQ2NOV1JqbktX?= =?utf-8?B?K1I1c3lKM1ljNUJ6a2k5ckdyVVBadVJ0TFBwWncxRHhSbmYvb2ZPbnRXc0dw?= =?utf-8?B?SnNmZGhYZXZqV2RGYUhzR29LcVRjZ3BOWUJEK2ZCWUdaaFFmbU5aOFVjVWov?= =?utf-8?B?SmtBZ2h1djBMUWxmQU9ISXQ2NStPWmZXU2FEQzlzdzN6TGN5a0NIQUtORE9h?= =?utf-8?B?QXR3Yi9kSlZDRXVXajg4THZucEgxSSt3WDlRODBacXFZQ3RESkRpMU5vWCtB?= =?utf-8?B?bGJWTUxNUkludzZyM2V0b0RHOUx6NXNpanZSdWUva0NvYjh2b3FHWGliQzR2?= =?utf-8?B?WWZFVGtlV3hjOGEzYzhZa1Bsc29Sb1l2eGUxR3VVVC9rekR4UlkxKy9XbnNM?= =?utf-8?B?S0JCNC9tTmdDNStBS1BTaEl1RFdPS3hUeDF2c3hib1FYZ2I2bTZCU0JaM1ZG?= =?utf-8?B?aFRMT0NkUEdhMFBKeFl2Wjl3bU5BUXcrZjFob2sveldIUmRHSmphSkZEVUM1?= =?utf-8?B?V3FmZ2c5VkZrNzZiUEY3SVVqWFc0eDg4TXhnL2h0V1UxWStXTXNqSDBDc29C?= =?utf-8?B?d1kzZ21TcXl0RXVpZ0c1YkQ3Tkh4d0pTYXJPYjFrbjBFWThDYllmTzRvSEZ5?= =?utf-8?B?SDdIdElxdjIzNU5ucU83WHp3S0UyVjdwbjBwQ2hEUkVnRSt1dFNNRUdzVzRB?= =?utf-8?B?aGdmdmp6czNWeXFWRldMaUorVllaYWpTYlkzYXpRTE1JUVhWUDBLanJDWkdi?= =?utf-8?B?Z0xITm51bDNNZ1hHSHllc0pxbXA2OTFrUFpHVTVWZWFFbG5jeFBYMzJqSS8v?= =?utf-8?B?bktCbDVlWjJtUjVac3Y3QXNna1V2RVgxVkpubjRQMFlIbi9pU3BuN1RkRzNX?= =?utf-8?B?bTZYdVB2czJvdUViWVF4SXpCbks2eGpFN1Jzcm4vbEVwSHJ0WEdDblVMbS9r?= =?utf-8?B?d3RoY2VaSW1tTGdwK3lvQjdCZHI3VEY2TDYxZ3lqNit4RjRrYXRyNzRwZ1VJ?= =?utf-8?B?aFplVktnWVB1TG9vSmZMMXVRc0RDRTA1RkxMTk0rbW45Tk1BTFFSUXpjNHBY?= =?utf-8?B?R3hmNElTWHlHSmFJVFRVQi9RbWJ5eWpoVU90MVViSUJoVURKb1lBdVFEYkps?= =?utf-8?B?K0R5ZFdhSEVmTDRnc3dkVGYvTjl3R043OXAwbnE2dm5oUG5KR3ViUncycmFh?= =?utf-8?B?WUNvQmtsekJWcXhOSi9iRWkxa3orQU53bEV3WHo1azh5alRmeGdYd3YrT1pY?= =?utf-8?B?ZUE9PQ==?= X-MS-Exchange-CrossTenant-Network-Message-Id: b5753fbf-e2fb-424a-a581-08dc0329036b X-MS-Exchange-CrossTenant-AuthSource: BY5PR11MB4274.namprd11.prod.outlook.com X-MS-Exchange-CrossTenant-AuthAs: Internal X-MS-Exchange-CrossTenant-OriginalArrivalTime: 22 Dec 2023 20:03:06.8447 (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: fRBt2s1RSFzXXJOA5ehI+O7yv1QP7EBPDXAguhtIDknnadwPY1wq2sscBqu6mA2ttyuzr1P4SZJxnRdjZLAM6idAoNdSmMWP41wEadrNriY= X-MS-Exchange-Transport-CrossTenantHeadersStamped: DM3PR11MB8684 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" --------------RmHfmYMiEnXfQ8ZBfzMzykdF Content-Type: text/plain; charset="UTF-8"; format=flowed Content-Transfer-Encoding: 7bit On 12/21/2023 11:45 PM, Riana Tauro wrote: > GuC provides engine busyness ticks as a 64 bit counter which count > as clock ticks. These counters are maintained in a > shared memory buffer and internally updated on a continuous basis. > > GuC also provides a periodically total active ticks that GT has been > active for (GuC loaded and running). > This counter is exposed to the user such that busyness can > be calculated as a percentage using > > busyness % = (engine active ticks/total active ticks) * 100. > > This patch provides a pmu counter for total active ticks. > > This is listed by perf tool as > > sudo ./perf list > xe_0000_03_00.0/total-active-ticks-gt0/ [Kernel PMU event] > > and can be read using > > sudo ./perf stat -e xe_0000_03_00.0/total-active-ticks-gt0/ -I 1000 > time counts unit events > 1.001332764 58942964 xe_0000_03_00.0/total-active-ticks-gt0/ > 2.011421147 21191869 xe_0000_03_00.0/total-active-ticks-gt0/ > 3.013223865 19269012 xe_0000_03_00.0/total-active-ticks-gt0/ > > v2: change commit message and comment for > total active ticks (Umesh, Tvrtko) > > Co-developed-by: Umesh Nerlige Ramappa > Signed-off-by: Umesh Nerlige Ramappa > Signed-off-by: Riana Tauro > --- > drivers/gpu/drm/xe/xe_gt.c | 11 +++ > drivers/gpu/drm/xe/xe_gt.h | 2 +- > drivers/gpu/drm/xe/xe_guc_engine_busyness.c | 72 ++++++++++++++++---- > drivers/gpu/drm/xe/xe_guc_engine_busyness.h | 1 + > drivers/gpu/drm/xe/xe_pmu.c | 74 +++++++++++++++++++-- > include/uapi/drm/xe_drm.h | 23 ++++++- > 6 files changed, 162 insertions(+), 21 deletions(-) > > diff --git a/drivers/gpu/drm/xe/xe_gt.c b/drivers/gpu/drm/xe/xe_gt.c > index 58daadc00363..5825471a3422 100644 > --- a/drivers/gpu/drm/xe/xe_gt.c > +++ b/drivers/gpu/drm/xe/xe_gt.c > @@ -789,3 +789,14 @@ u64 xe_gt_engine_busy_ticks(struct xe_gt *gt, struct xe_hw_engine *hwe) > { > return xe_guc_engine_busyness_ticks(>->uc.guc, hwe); > } > + > +/** > + * xe_gt_total_active_ticks - Return total active ticks > + * @gt: GT structure > + * > + * Returns total active ticks that the GT was active for. > + */ > +u64 xe_gt_total_active_ticks(struct xe_gt *gt) > +{ > + return xe_guc_engine_busyness_active_ticks(>->uc.guc); > +} > diff --git a/drivers/gpu/drm/xe/xe_gt.h b/drivers/gpu/drm/xe/xe_gt.h > index 4303ec48c404..9bac85cdf609 100644 > --- a/drivers/gpu/drm/xe/xe_gt.h > +++ b/drivers/gpu/drm/xe/xe_gt.h > @@ -43,7 +43,7 @@ void xe_gt_reset_async(struct xe_gt *gt); > void xe_gt_sanitize(struct xe_gt *gt); > > u64 xe_gt_engine_busy_ticks(struct xe_gt *gt, struct xe_hw_engine *hwe); > - > +u64 xe_gt_total_active_ticks(struct xe_gt *gt); > /** > * xe_gt_any_hw_engine_by_reset_domain - scan the list of engines and return the > * first that matches the same reset domain as @class > diff --git a/drivers/gpu/drm/xe/xe_guc_engine_busyness.c b/drivers/gpu/drm/xe/xe_guc_engine_busyness.c > index af7b6e768751..24e72555647a 100644 > --- a/drivers/gpu/drm/xe/xe_guc_engine_busyness.c > +++ b/drivers/gpu/drm/xe/xe_guc_engine_busyness.c > @@ -20,53 +20,83 @@ > * timer internal to GuC. The update rate is guaranteed to be at least 2Hz (but with > * a caveat that is not real time, best effort only). > * > + * In addition to the engine busyness ticks, there is also a total time count which > + * is a free running GT timestamp counter. > + * > + * Note that counters should be used as ratios of each other for calculating a > + * percentage. > + * > * engine busyness ticks (ticks_engine) : clock ticks for which engine was active > + * total active ticks (ticks_gt) : total clock ticks > + * > + * engine busyness % = (ticks_engine / ticks_gt) * 100 > */ > > static void guc_engine_busyness_usage_map(struct xe_guc *guc, > struct xe_hw_engine *hwe, > - struct iosys_map *engine_map) > + struct iosys_map *engine_map, > + struct iosys_map *global_map) > { > struct iosys_map *map; > size_t offset; > u32 instance; > u8 guc_class; > > - guc_class = xe_engine_class_to_guc_class(hwe->class); > - instance = hwe->logical_instance; > + if (hwe) { > + guc_class = xe_engine_class_to_guc_class(hwe->class); > + instance = hwe->logical_instance; > + } > > map = &guc->busy.bo->vmap; > > - offset = offsetof(struct guc_engine_observation_data, > - engine_data[guc_class][instance]); > + if (hwe) { > + offset = offsetof(struct guc_engine_observation_data, > + engine_data[guc_class][instance]); > + *engine_map = IOSYS_MAP_INIT_OFFSET(map, offset); > + } > > - *engine_map = IOSYS_MAP_INIT_OFFSET(map, offset); > + *global_map = IOSYS_MAP_INIT_OFFSET(map, 0); > } > > static void guc_engine_busyness_get_usage(struct xe_guc *guc, > struct xe_hw_engine *hwe, > - u64 *ticks_engine) > + u64 *ticks_engine, > + u64 *ticks_gt) > { > - struct iosys_map engine_map; > - u64 engine_ticks = 0; > + struct iosys_map engine_map, global_map; > + u64 engine_ticks = 0, gt_ticks = 0; > int i = 0; > > - guc_engine_busyness_usage_map(guc, hwe, &engine_map); > + guc_engine_busyness_usage_map(guc, hwe, &engine_map, &global_map); > > #define read_engine_usage(map_, field_) \ > iosys_map_rd_field(map_, 0, struct guc_engine_data, field_) > > +#define read_global_field(map_, field_) \ > + iosys_map_rd_field(map_, 0, struct guc_engine_observation_data, field_) > + > do { > - engine_ticks = read_engine_usage(&engine_map, total_execution_ticks); > + if (hwe) > + engine_ticks = read_engine_usage(&engine_map, total_execution_ticks); > + > + gt_ticks = read_global_field(&global_map, gt_timestamp); > > - if (read_engine_usage(&engine_map, total_execution_ticks) == engine_ticks) > + if (hwe && read_engine_usage(&engine_map, total_execution_ticks) != engine_ticks) > + continue; > + > + if (read_global_field(&global_map, gt_timestamp) == gt_ticks) > break; > + > } while (++i < 6); > > #undef read_engine_usage > +#undef read_global_field > > if (ticks_engine) > *ticks_engine = engine_ticks; > + > + if (ticks_gt) > + *ticks_gt = gt_ticks; > } > > static void guc_engine_busyness_enable_stats(struct xe_guc *guc) > @@ -92,6 +122,22 @@ static void guc_engine_busyness_fini(struct drm_device *drm, void *arg) > xe_bo_unpin_map_no_vm(guc->busy.bo); > } > > +/* > + * xe_guc_engine_busyness_active_ticks - Gets the total active ticks > + * @guc: The GuC object > + * > + * Returns total active ticks that the GT has been running for > + * (GuC loaded and running). > + */ > +u64 xe_guc_engine_busyness_active_ticks(struct xe_guc *guc) nit: Could we have the word gt somewhere in this function so its clear what we are asking for? like xe_guc_engine_busyness_active_gt_ticks? Other than that, Reviewed-by: Vinay Belgaumkar > +{ > + u64 ticks_gt; > + > + guc_engine_busyness_get_usage(guc, NULL, NULL, &ticks_gt); > + > + return ticks_gt; > +} > + > /* > * xe_guc_engine_busyness_ticks - Gets current accumulated > * engine busyness ticks > @@ -104,7 +150,7 @@ u64 xe_guc_engine_busyness_ticks(struct xe_guc *guc, struct xe_hw_engine *hwe) > { > u64 ticks_engine; > > - guc_engine_busyness_get_usage(guc, hwe, &ticks_engine); > + guc_engine_busyness_get_usage(guc, hwe, &ticks_engine, NULL); > > return ticks_engine; > } > diff --git a/drivers/gpu/drm/xe/xe_guc_engine_busyness.h b/drivers/gpu/drm/xe/xe_guc_engine_busyness.h > index d70f06209896..57325910ebc4 100644 > --- a/drivers/gpu/drm/xe/xe_guc_engine_busyness.h > +++ b/drivers/gpu/drm/xe/xe_guc_engine_busyness.h > @@ -12,6 +12,7 @@ struct xe_hw_engine; > struct xe_guc; > > int xe_guc_engine_busyness_init(struct xe_guc *guc); > +u64 xe_guc_engine_busyness_active_ticks(struct xe_guc *guc); > u64 xe_guc_engine_busyness_ticks(struct xe_guc *guc, struct xe_hw_engine *hwe); > > #endif > diff --git a/drivers/gpu/drm/xe/xe_pmu.c b/drivers/gpu/drm/xe/xe_pmu.c > index 371ca6d7e215..c2be157a6f5d 100644 > --- a/drivers/gpu/drm/xe/xe_pmu.c > +++ b/drivers/gpu/drm/xe/xe_pmu.c > @@ -51,6 +51,20 @@ static bool is_engine_event(struct perf_event *event) > return config_counter(event->attr.config) < __DRM_XE_PMU_OTHER(0, 0); > } > > +static int > +config_status(struct xe_device *xe, u64 config) > +{ > + unsigned int gt_id = config_gt_id(config); > + > + if (gt_id >= XE_PMU_MAX_GT) > + return -ENOENT; > + > + if (config_counter(config) == DRM_XE_PMU_TOTAL_ACTIVE_TICKS(0)) > + return 0; > + > + return -ENOENT; > +} > + > static int engine_event_status(struct xe_hw_engine *hwe, > enum drm_xe_pmu_engine_sample sample) > { > @@ -113,11 +127,13 @@ static int xe_pmu_event_init(struct perf_event *event) > if (!cpumask_test_cpu(event->cpu, &xe_pmu_cpumask)) > return -EINVAL; > > - if (is_engine_event(event)) { > + if (is_engine_event(event)) > ret = engine_event_init(event); > - if (ret) > - return ret; > - } > + else > + ret = config_status(xe, event->attr.config); > + > + if (ret) > + return ret; > > if (!event->parent) { > drm_dev_get(&xe->drm); > @@ -131,7 +147,8 @@ static u64 __xe_pmu_event_read(struct perf_event *event) > { > struct xe_device *xe = > container_of(event->pmu, typeof(*xe), pmu.base); > - const unsigned int gt_id = config_gt_id(event->attr.config); > + u64 config = event->attr.config; > + const unsigned int gt_id = config_gt_id(config); > struct xe_gt *gt = xe_device_get_gt(xe, gt_id); > u64 val; > > @@ -147,6 +164,11 @@ static u64 __xe_pmu_event_read(struct perf_event *event) > val = xe_gt_engine_busy_ticks(gt, hwe); > else > drm_warn(&xe->drm, "unknown pmu engine event\n"); > + } else { > + if (config_counter(config) == DRM_XE_PMU_TOTAL_ACTIVE_TICKS(0)) > + val = xe_gt_total_active_ticks(gt); > + else > + drm_warn(&xe->drm, "unknown pmu event\n"); > } > > return val; > @@ -256,6 +278,12 @@ static const struct attribute_group xe_pmu_cpumask_attr_group = { > .attrs = xe_cpumask_attrs, > }; > > +#define __event(__counter, __name) \ > +{ \ > + .counter = (__counter), \ > + .name = (__name), \ > +} > + > #define __engine_event(__sample, __name) \ > { \ > .sample = (__sample), \ > @@ -293,6 +321,23 @@ create_event_attributes(struct xe_pmu *pmu) > __engine_event(DRM_XE_PMU_SAMPLE_BUSY_TICKS, "busy-ticks"), > }; > > + static const struct { > + unsigned int counter; > + const char *name; > + } events[] = { > + __event(0, "total-active-ticks"), > + }; > + > + /* Count how many counters we will be exposing. */ > + for_each_gt(gt, xe, j) { > + for (i = 0; i < ARRAY_SIZE(events); i++) { > + u64 config = __DRM_XE_PMU_OTHER(j, events[i].counter); > + > + if (!config_status(xe, config)) > + count++; > + } > + } > + > for_each_gt(gt, xe, j) { > for_each_hw_engine(hwe, gt, id) { > for (i = 0; i < ARRAY_SIZE(engine_events); i++) { > @@ -315,6 +360,25 @@ create_event_attributes(struct xe_pmu *pmu) > xe_iter = xe_attr; > attr_iter = attr; > > + /* Initialize supported non-engine counters */ > + for_each_gt(gt, xe, j) { > + for (i = 0; i < ARRAY_SIZE(events); i++) { > + u64 config = __DRM_XE_PMU_OTHER(j, events[i].counter); > + char *str; > + > + if (config_status(xe, config)) > + continue; > + > + str = kasprintf(GFP_KERNEL, "%s-gt%u", > + events[i].name, j); > + if (!str) > + goto err; > + > + *attr_iter++ = &xe_iter->attr.attr; > + xe_iter = add_xe_attr(xe_iter, str, config); > + } > + } > + > /* Initialize supported engine counters */ > for_each_gt(gt, xe, j) { > for_each_hw_engine(hwe, gt, id) { > diff --git a/include/uapi/drm/xe_drm.h b/include/uapi/drm/xe_drm.h > index f8456cda5cda..3134930b0160 100644 > --- a/include/uapi/drm/xe_drm.h > +++ b/include/uapi/drm/xe_drm.h > @@ -1341,12 +1341,29 @@ struct drm_xe_wait_user_fence { > }; > > /** > - * DOC: XE PMU event config IDs > + * DOC: XE PMU Event Config ID's > * > - * Check 'man perf_event_open' to use the ID's DRM_XE_PMU_XXXX listed in xe_drm.h > + * Check 'man perf_event_open' to use the ID's DRM_XE_PMU_XXXX listed here > * in 'struct perf_event_attr' as part of perf_event_open syscall to read a > * particular event. > * > + * For example to open the DRM_XE_PMU_TOTAL_ACTIVE_TICKS(0): > + * > + * .. code-block:: C > + * > + * struct perf_event_attr attr; > + * long long count; > + * int cpu = 0; > + * int fd; > + * > + * memset(&attr, 0, sizeof(struct perf_event_attr)); > + * attr.type = type; // eg: /sys/bus/event_source/devices/xe_0000_03_00.0/type > + * attr.read_format = PERF_FORMAT_TOTAL_TIME_ENABLED; > + * attr.use_clockid = 1; > + * attr.clockid = CLOCK_MONOTONIC; > + * attr.config = DRM_XE_PMU_TOTAL_ACTIVE_TICKS(0); > + * > + * fd = syscall(__NR_perf_event_open, &attr, -1, cpu, -1, 0); > */ > > /** > @@ -1381,6 +1398,8 @@ enum drm_xe_pmu_engine_sample { > __DRM_XE_PMU_GT_EVENT(gt, __DRM_XE_PMU_ENGINE(class, instance, \ > DRM_XE_PMU_SAMPLE_BUSY_TICKS)) > > +#define DRM_XE_PMU_TOTAL_ACTIVE_TICKS(gt) __DRM_XE_PMU_OTHER(gt, 0) > + > #if defined(__cplusplus) > } > #endif --------------RmHfmYMiEnXfQ8ZBfzMzykdF Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: 7bit


On 12/21/2023 11:45 PM, Riana Tauro wrote:
GuC provides engine busyness ticks as a 64 bit counter which count
as clock ticks. These counters are maintained in a
shared memory buffer and internally updated on a continuous basis.

GuC also provides a periodically total active ticks that GT has been
active for (GuC loaded and running).
This counter is exposed to the user such that busyness can
be calculated as a percentage using

busyness % = (engine active ticks/total active ticks) * 100.

This patch provides a pmu counter for total active ticks.

This is listed by perf tool as

sudo ./perf list
	  xe_0000_03_00.0/total-active-ticks-gt0/            [Kernel PMU event]

and can be read using

sudo ./perf stat -e xe_0000_03_00.0/total-active-ticks-gt0/ -I 1000
        time 	    counts  unit 	events
    1.001332764    58942964    xe_0000_03_00.0/total-active-ticks-gt0/
    2.011421147	   21191869    xe_0000_03_00.0/total-active-ticks-gt0/
    3.013223865	   19269012    xe_0000_03_00.0/total-active-ticks-gt0/

v2: change commit message and comment for
    total active ticks (Umesh, Tvrtko)

Co-developed-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Signed-off-by: Riana Tauro <riana.tauro@intel.com>
---
 drivers/gpu/drm/xe/xe_gt.c                  | 11 +++
 drivers/gpu/drm/xe/xe_gt.h                  |  2 +-
 drivers/gpu/drm/xe/xe_guc_engine_busyness.c | 72 ++++++++++++++++----
 drivers/gpu/drm/xe/xe_guc_engine_busyness.h |  1 +
 drivers/gpu/drm/xe/xe_pmu.c                 | 74 +++++++++++++++++++--
 include/uapi/drm/xe_drm.h                   | 23 ++++++-
 6 files changed, 162 insertions(+), 21 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_gt.c b/drivers/gpu/drm/xe/xe_gt.c
index 58daadc00363..5825471a3422 100644
--- a/drivers/gpu/drm/xe/xe_gt.c
+++ b/drivers/gpu/drm/xe/xe_gt.c
@@ -789,3 +789,14 @@ u64 xe_gt_engine_busy_ticks(struct xe_gt *gt, struct xe_hw_engine *hwe)
 {
 	return xe_guc_engine_busyness_ticks(&gt->uc.guc, hwe);
 }
+
+/**
+ * xe_gt_total_active_ticks - Return total active ticks
+ * @gt: GT structure
+ *
+ * Returns total active ticks that the GT was active for.
+ */
+u64 xe_gt_total_active_ticks(struct xe_gt *gt)
+{
+	return xe_guc_engine_busyness_active_ticks(&gt->uc.guc);
+}
diff --git a/drivers/gpu/drm/xe/xe_gt.h b/drivers/gpu/drm/xe/xe_gt.h
index 4303ec48c404..9bac85cdf609 100644
--- a/drivers/gpu/drm/xe/xe_gt.h
+++ b/drivers/gpu/drm/xe/xe_gt.h
@@ -43,7 +43,7 @@ void xe_gt_reset_async(struct xe_gt *gt);
 void xe_gt_sanitize(struct xe_gt *gt);
 
 u64 xe_gt_engine_busy_ticks(struct xe_gt *gt, struct xe_hw_engine *hwe);
-
+u64 xe_gt_total_active_ticks(struct xe_gt *gt);
 /**
  * xe_gt_any_hw_engine_by_reset_domain - scan the list of engines and return the
  * first that matches the same reset domain as @class
diff --git a/drivers/gpu/drm/xe/xe_guc_engine_busyness.c b/drivers/gpu/drm/xe/xe_guc_engine_busyness.c
index af7b6e768751..24e72555647a 100644
--- a/drivers/gpu/drm/xe/xe_guc_engine_busyness.c
+++ b/drivers/gpu/drm/xe/xe_guc_engine_busyness.c
@@ -20,53 +20,83 @@
  * timer internal to GuC. The update rate is guaranteed to be at least 2Hz (but with
  * a caveat that is not real time, best effort only).
  *
+ * In addition to the engine busyness ticks, there is also a total time count which
+ * is a free running GT timestamp counter.
+ *
+ * Note that counters should be used as ratios of each other for calculating a
+ * percentage.
+ *
  * engine busyness ticks (ticks_engine) : clock ticks for which engine was active
+ * total active ticks (ticks_gt)	: total clock ticks
+ *
+ * engine busyness % = (ticks_engine / ticks_gt) * 100
  */
 
 static void guc_engine_busyness_usage_map(struct xe_guc *guc,
 					  struct xe_hw_engine *hwe,
-					  struct iosys_map *engine_map)
+					  struct iosys_map *engine_map,
+					  struct iosys_map *global_map)
 {
 	struct iosys_map *map;
 	size_t offset;
 	u32 instance;
 	u8 guc_class;
 
-	guc_class = xe_engine_class_to_guc_class(hwe->class);
-	instance = hwe->logical_instance;
+	if (hwe) {
+		guc_class = xe_engine_class_to_guc_class(hwe->class);
+		instance = hwe->logical_instance;
+	}
 
 	map = &guc->busy.bo->vmap;
 
-	offset = offsetof(struct guc_engine_observation_data,
-			  engine_data[guc_class][instance]);
+	if (hwe) {
+		offset = offsetof(struct guc_engine_observation_data,
+				  engine_data[guc_class][instance]);
+		*engine_map = IOSYS_MAP_INIT_OFFSET(map, offset);
+	}
 
-	*engine_map = IOSYS_MAP_INIT_OFFSET(map, offset);
+	*global_map = IOSYS_MAP_INIT_OFFSET(map, 0);
 }
 
 static void guc_engine_busyness_get_usage(struct xe_guc *guc,
 					  struct xe_hw_engine *hwe,
-					  u64 *ticks_engine)
+					  u64 *ticks_engine,
+					  u64 *ticks_gt)
 {
-	struct iosys_map engine_map;
-	u64 engine_ticks = 0;
+	struct iosys_map engine_map, global_map;
+	u64 engine_ticks = 0, gt_ticks = 0;
 	int i = 0;
 
-	guc_engine_busyness_usage_map(guc, hwe, &engine_map);
+	guc_engine_busyness_usage_map(guc, hwe, &engine_map, &global_map);
 
 #define read_engine_usage(map_, field_) \
 	iosys_map_rd_field(map_, 0, struct guc_engine_data, field_)
 
+#define read_global_field(map_, field_) \
+	iosys_map_rd_field(map_, 0, struct guc_engine_observation_data, field_)
+
 	do {
-		engine_ticks = read_engine_usage(&engine_map, total_execution_ticks);
+		if (hwe)
+			engine_ticks = read_engine_usage(&engine_map, total_execution_ticks);
+
+		gt_ticks = read_global_field(&global_map, gt_timestamp);
 
-		if (read_engine_usage(&engine_map, total_execution_ticks) == engine_ticks)
+		if (hwe && read_engine_usage(&engine_map, total_execution_ticks) != engine_ticks)
+			continue;
+
+		if (read_global_field(&global_map, gt_timestamp) == gt_ticks)
 			break;
+
 	} while (++i < 6);
 
 #undef read_engine_usage
+#undef read_global_field
 
 	if (ticks_engine)
 		*ticks_engine = engine_ticks;
+
+	if (ticks_gt)
+		*ticks_gt = gt_ticks;
 }
 
 static void guc_engine_busyness_enable_stats(struct xe_guc *guc)
@@ -92,6 +122,22 @@ static void guc_engine_busyness_fini(struct drm_device *drm, void *arg)
 	xe_bo_unpin_map_no_vm(guc->busy.bo);
 }
 
+/*
+ * xe_guc_engine_busyness_active_ticks - Gets the total active ticks
+ * @guc: The GuC object
+ *
+ * Returns total active ticks that the GT has been running for
+ * (GuC loaded and running).
+ */
+u64 xe_guc_engine_busyness_active_ticks(struct xe_guc *guc)

nit: Could we have the word gt somewhere in this function so its clear what we are asking for? like xe_guc_engine_busyness_active_gt_ticks?

Other than that,

Reviewed-by: Vinay Belgaumkar <vinay.belgaumkar@intel.com> 

+{
+	u64 ticks_gt;
+
+	guc_engine_busyness_get_usage(guc, NULL, NULL, &ticks_gt);
+
+	return ticks_gt;
+}
+
 /*
  * xe_guc_engine_busyness_ticks - Gets current accumulated
  *				  engine busyness ticks
@@ -104,7 +150,7 @@ u64 xe_guc_engine_busyness_ticks(struct xe_guc *guc, struct xe_hw_engine *hwe)
 {
 	u64 ticks_engine;
 
-	guc_engine_busyness_get_usage(guc, hwe, &ticks_engine);
+	guc_engine_busyness_get_usage(guc, hwe, &ticks_engine, NULL);
 
 	return ticks_engine;
 }
diff --git a/drivers/gpu/drm/xe/xe_guc_engine_busyness.h b/drivers/gpu/drm/xe/xe_guc_engine_busyness.h
index d70f06209896..57325910ebc4 100644
--- a/drivers/gpu/drm/xe/xe_guc_engine_busyness.h
+++ b/drivers/gpu/drm/xe/xe_guc_engine_busyness.h
@@ -12,6 +12,7 @@ struct xe_hw_engine;
 struct xe_guc;
 
 int xe_guc_engine_busyness_init(struct xe_guc *guc);
+u64 xe_guc_engine_busyness_active_ticks(struct xe_guc *guc);
 u64 xe_guc_engine_busyness_ticks(struct xe_guc *guc, struct xe_hw_engine *hwe);
 
 #endif
diff --git a/drivers/gpu/drm/xe/xe_pmu.c b/drivers/gpu/drm/xe/xe_pmu.c
index 371ca6d7e215..c2be157a6f5d 100644
--- a/drivers/gpu/drm/xe/xe_pmu.c
+++ b/drivers/gpu/drm/xe/xe_pmu.c
@@ -51,6 +51,20 @@ static bool is_engine_event(struct perf_event *event)
 	return config_counter(event->attr.config) < __DRM_XE_PMU_OTHER(0, 0);
 }
 
+static int
+config_status(struct xe_device *xe, u64 config)
+{
+	unsigned int gt_id = config_gt_id(config);
+
+	if (gt_id >= XE_PMU_MAX_GT)
+		return -ENOENT;
+
+	if (config_counter(config) == DRM_XE_PMU_TOTAL_ACTIVE_TICKS(0))
+		return 0;
+
+	return -ENOENT;
+}
+
 static int engine_event_status(struct xe_hw_engine *hwe,
 			       enum drm_xe_pmu_engine_sample sample)
 {
@@ -113,11 +127,13 @@ static int xe_pmu_event_init(struct perf_event *event)
 	if (!cpumask_test_cpu(event->cpu, &xe_pmu_cpumask))
 		return -EINVAL;
 
-	if (is_engine_event(event)) {
+	if (is_engine_event(event))
 		ret = engine_event_init(event);
-		if (ret)
-			return ret;
-	}
+	else
+		ret = config_status(xe, event->attr.config);
+
+	if (ret)
+		return ret;
 
 	if (!event->parent) {
 		drm_dev_get(&xe->drm);
@@ -131,7 +147,8 @@ static u64 __xe_pmu_event_read(struct perf_event *event)
 {
 	struct xe_device *xe =
 		container_of(event->pmu, typeof(*xe), pmu.base);
-	const unsigned int gt_id = config_gt_id(event->attr.config);
+	u64 config = event->attr.config;
+	const unsigned int gt_id = config_gt_id(config);
 	struct xe_gt *gt = xe_device_get_gt(xe, gt_id);
 	u64 val;
 
@@ -147,6 +164,11 @@ static u64 __xe_pmu_event_read(struct perf_event *event)
 			val = xe_gt_engine_busy_ticks(gt, hwe);
 		else
 			drm_warn(&xe->drm, "unknown pmu engine event\n");
+	} else {
+		if (config_counter(config) == DRM_XE_PMU_TOTAL_ACTIVE_TICKS(0))
+			val = xe_gt_total_active_ticks(gt);
+		else
+			drm_warn(&xe->drm, "unknown pmu event\n");
 	}
 
 	return val;
@@ -256,6 +278,12 @@ static const struct attribute_group xe_pmu_cpumask_attr_group = {
 	.attrs = xe_cpumask_attrs,
 };
 
+#define __event(__counter, __name) \
+{ \
+	.counter = (__counter), \
+	.name = (__name), \
+}
+
 #define __engine_event(__sample, __name) \
 { \
 	.sample = (__sample), \
@@ -293,6 +321,23 @@ create_event_attributes(struct xe_pmu *pmu)
 		__engine_event(DRM_XE_PMU_SAMPLE_BUSY_TICKS, "busy-ticks"),
 	};
 
+	static const struct {
+		unsigned int counter;
+		const char *name;
+	} events[] = {
+		__event(0, "total-active-ticks"),
+	};
+
+	/* Count how many counters we will be exposing. */
+	for_each_gt(gt, xe, j) {
+		for (i = 0; i < ARRAY_SIZE(events); i++) {
+			u64 config = __DRM_XE_PMU_OTHER(j, events[i].counter);
+
+			if (!config_status(xe, config))
+				count++;
+		}
+	}
+
 	for_each_gt(gt, xe, j) {
 		for_each_hw_engine(hwe, gt, id) {
 			for (i = 0; i < ARRAY_SIZE(engine_events); i++) {
@@ -315,6 +360,25 @@ create_event_attributes(struct xe_pmu *pmu)
 	xe_iter = xe_attr;
 	attr_iter = attr;
 
+	/* Initialize supported non-engine counters */
+	for_each_gt(gt, xe, j) {
+		for (i = 0; i < ARRAY_SIZE(events); i++) {
+			u64 config = __DRM_XE_PMU_OTHER(j, events[i].counter);
+			char *str;
+
+			if (config_status(xe, config))
+				continue;
+
+			str = kasprintf(GFP_KERNEL, "%s-gt%u",
+					events[i].name, j);
+			if (!str)
+				goto err;
+
+			*attr_iter++ = &xe_iter->attr.attr;
+			xe_iter = add_xe_attr(xe_iter, str, config);
+		}
+	}
+
 	/* Initialize supported engine counters */
 	for_each_gt(gt, xe, j) {
 		for_each_hw_engine(hwe, gt, id) {
diff --git a/include/uapi/drm/xe_drm.h b/include/uapi/drm/xe_drm.h
index f8456cda5cda..3134930b0160 100644
--- a/include/uapi/drm/xe_drm.h
+++ b/include/uapi/drm/xe_drm.h
@@ -1341,12 +1341,29 @@ struct drm_xe_wait_user_fence {
 };
 
 /**
- * DOC: XE PMU event config IDs
+ * DOC: XE PMU Event Config ID's
  *
- * Check 'man perf_event_open' to use the ID's DRM_XE_PMU_XXXX listed in xe_drm.h
+ * Check 'man perf_event_open' to use the ID's DRM_XE_PMU_XXXX listed here
  * in 'struct perf_event_attr' as part of perf_event_open syscall to read a
  * particular event.
  *
+ * For example to open the DRM_XE_PMU_TOTAL_ACTIVE_TICKS(0):
+ *
+ * .. code-block:: C
+ *
+ *     struct perf_event_attr attr;
+ *     long long count;
+ *     int cpu = 0;
+ *     int fd;
+ *
+ *     memset(&attr, 0, sizeof(struct perf_event_attr));
+ *     attr.type = type; // eg: /sys/bus/event_source/devices/xe_0000_03_00.0/type
+ *     attr.read_format = PERF_FORMAT_TOTAL_TIME_ENABLED;
+ *     attr.use_clockid = 1;
+ *     attr.clockid = CLOCK_MONOTONIC;
+ *     attr.config = DRM_XE_PMU_TOTAL_ACTIVE_TICKS(0);
+ *
+ *     fd = syscall(__NR_perf_event_open, &attr, -1, cpu, -1, 0);
  */
 
 /**
@@ -1381,6 +1398,8 @@ enum drm_xe_pmu_engine_sample {
 	__DRM_XE_PMU_GT_EVENT(gt, __DRM_XE_PMU_ENGINE(class, instance, \
 						      DRM_XE_PMU_SAMPLE_BUSY_TICKS))
 
+#define DRM_XE_PMU_TOTAL_ACTIVE_TICKS(gt)	__DRM_XE_PMU_OTHER(gt, 0)
+
 #if defined(__cplusplus)
 }
 #endif
--------------RmHfmYMiEnXfQ8ZBfzMzykdF--