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 A7B88C2BBCA for ; Thu, 20 Jun 2024 23:44:01 +0000 (UTC) Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id 227F810E870; Thu, 20 Jun 2024 23:44:01 +0000 (UTC) Authentication-Results: gabe.freedesktop.org; dkim=pass (2048-bit key; unprotected) header.d=intel.com header.i=@intel.com header.b="X7lg2NPk"; dkim-atps=neutral Received: from mgamail.intel.com (mgamail.intel.com [198.175.65.20]) by gabe.freedesktop.org (Postfix) with ESMTPS id CCF8510E870 for ; Thu, 20 Jun 2024 23:43:59 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1718927040; x=1750463040; h=date:from:to:cc:subject:message-id:references: content-transfer-encoding:in-reply-to:mime-version; bh=NrHNl75fLMoTjdECeaG6z1BoPNdrHMEjktfel1eJtRc=; b=X7lg2NPkpVkUgkHK8u/4BROy1kaDNXNOKctbIZaCv3vl5DeBV1XmRvbA EAd3gr1rE5SGMh8fIuWFJpCnuVLqe6xM8DN/PYlbQKxPCiSJ3TevlLChb 4qtFY/OXPqCK51GvDb+un7k0vJhDGizTROK+OHsn30Avb/tmbrPJW1IAq R+rjedftxTc7ktVCV8ciCUjz0styE/diqNDNGYNje8l5eJwTuiVII0bTG /sSSgYlbjFDcTZdepXwJbJ1nxgjijoWp7th5axB+yK7Nd18324Bd5jxQY PzC4SUPVHaRO9l2xhp0BMDlugREIrWMBFbA5MTfRGXQ05HyilHeF2PpKf Q==; X-CSE-ConnectionGUID: 2o1LiBlbQCmyBwWwTeHsvw== X-CSE-MsgGUID: oZzTQLfsSweNaHszRJbHqA== X-IronPort-AV: E=McAfee;i="6700,10204,11109"; a="15774624" X-IronPort-AV: E=Sophos;i="6.08,253,1712646000"; d="scan'208";a="15774624" Received: from orviesa010.jf.intel.com ([10.64.159.150]) by orvoesa112.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 20 Jun 2024 16:43:59 -0700 X-CSE-ConnectionGUID: BhTbV4sQTO2myF7XVtONQw== X-CSE-MsgGUID: pBxoDxubRFOk8B1KRQ4McQ== X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="6.08,253,1712646000"; d="scan'208";a="42311444" Received: from orsmsx601.amr.corp.intel.com ([10.22.229.14]) by orviesa010.jf.intel.com with ESMTP/TLS/AES256-GCM-SHA384; 20 Jun 2024 16:43:59 -0700 Received: from orsmsx610.amr.corp.intel.com (10.22.229.23) by ORSMSX601.amr.corp.intel.com (10.22.229.14) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256) id 15.1.2507.39; Thu, 20 Jun 2024 16:43:58 -0700 Received: from orsedg603.ED.cps.intel.com (10.7.248.4) 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 15.1.2507.39 via Frontend Transport; Thu, 20 Jun 2024 16:43:58 -0700 Received: from NAM12-MW2-obe.outbound.protection.outlook.com (104.47.66.42) 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 15.1.2507.39; Thu, 20 Jun 2024 16:43:58 -0700 ARC-Seal: i=1; a=rsa-sha256; s=arcselector9901; d=microsoft.com; cv=none; b=Tv82rDqNw6BbQ7MAN9GQGXiK+ivhKbEm//gh3gRaLHH1YkpZQFVCG5T2lFFrN3Oqzk1oxqG2z4nMSEmS9LaB3nYvQEtB/FD1203QwFBYZti3WjYM4uVk4WhvHxgUGnhgzDuU6rGeEiUFfFFeDq+MSpZ306vMowVMUvGqNaYbUvjU4AlBJNoukhsb3OduA/utOSeMvPDEbx7w9XdjE2WCwNrpzh+SQDEcMuKoL3BX4cvgvglXem8mi5wWkORL5FeOe9V4uKCIZ83Z/BrAD5YUtb7eQKcbBxXC11hadtXbVbC0MVE281/D9FfTRLPZYtizZhQ7rmc2b7K9IJ/MBfiOIA== 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=Peza8V+zWL/SaU6Z+lj9vIODbIl2ekcdoCvSlquLOXM=; b=FgL6JPxJV7VtGvEQafcXqGKsQz3j9/DMkTAUnj00iQ3lYWKJhChCIvYrJPZ+epAL1o7jKggD7lUYzTPj4+hVNvu05WSuDMKgeGrroXBZKxDsETegYwFws5hdBT8bR4JlubC09ESPkz9MYXPa3fU4a1vez/ebIzdHmyxTpfmfXyHXXCHPHyOMHJxZTw6Qll7u5rIHQpdWYLAincCwIJZvCDu/DiDTBIWU1KVdtGOEHl7ACC6o6gmj3hNfZeCoxOo7q2uLJQGvRg/d/avVogGbx1CmhilBR6kO3q/KbrmeQ/KI1FJ3Gq2reL9JCglXJyMngSBMU7dnHZX2FfRmGVW5IA== 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 DS0PR11MB7408.namprd11.prod.outlook.com (2603:10b6:8:136::15) by SJ0PR11MB4813.namprd11.prod.outlook.com (2603:10b6:a03:2df::14) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.20.7698.22; Thu, 20 Jun 2024 23:43:52 +0000 Received: from DS0PR11MB7408.namprd11.prod.outlook.com ([fe80::6387:4b73:8906:7543]) by DS0PR11MB7408.namprd11.prod.outlook.com ([fe80::6387:4b73:8906:7543%5]) with mapi id 15.20.7698.017; Thu, 20 Jun 2024 23:43:52 +0000 Date: Thu, 20 Jun 2024 16:43:50 -0700 From: Umesh Nerlige Ramappa To: Ashutosh Dixit CC: Subject: Re: [PATCH i-g-t 26/28] xe/oa: Regenerate oa-lnl.xml now parsing all counters Message-ID: References: <20240620200054.3550653-1-ashutosh.dixit@intel.com> <20240620200054.3550653-27-ashutosh.dixit@intel.com> Content-Type: text/plain; charset="utf-8"; format=flowed Content-Disposition: inline Content-Transfer-Encoding: 8bit In-Reply-To: <20240620200054.3550653-27-ashutosh.dixit@intel.com> X-ClientProxiedBy: MW4PR04CA0038.namprd04.prod.outlook.com (2603:10b6:303:6a::13) To DS0PR11MB7408.namprd11.prod.outlook.com (2603:10b6:8:136::15) MIME-Version: 1.0 X-MS-PublicTrafficType: Email X-MS-TrafficTypeDiagnostic: DS0PR11MB7408:EE_|SJ0PR11MB4813:EE_ X-MS-Office365-Filtering-Correlation-Id: 39c4a93a-9a5e-4ca1-9ac9-08dc9182d728 X-MS-Exchange-SenderADCheck: 1 X-MS-Exchange-AntiSpam-Relay: 0 X-Microsoft-Antispam: BCL:0;ARA:13230037|1800799021|366013|376011; X-Microsoft-Antispam-Message-Info: =?utf-8?B?bEdVNGxQQy9LMzJPb2dBWVJpcDdwRW9xRmRmYVNkYmU2ZEFzeGZFUytySSs5?= =?utf-8?B?NDZVUm5EVkV0NTRITGQ1SXdGaXdhSUk4K3JNMzIrb0REQmhuMnRjd3dlZTlt?= =?utf-8?B?eDBteGZIUU51REhmVUlNcnAzcEtLOEhPUlF4TG5sa3QxbFF2N3hTWFMrc2FQ?= =?utf-8?B?V0FGZEl0RkFiUWR1c3dDblBCZVhkNFFCcy92ZE9SeHdOUWp6SWp2L2Zwcmxa?= =?utf-8?B?aVVoWWNsV05QT3Awcmo0NWdnUklFQUhsVm1OYVRYeEhpUkpDWFcxOThxZHVm?= =?utf-8?B?K2FWcE41Qmx1VzkvTVE3WmtzUUtEc3Y0ODk2VXZLUDlZZEp0Mk1DVzJEWnhi?= =?utf-8?B?cHZZY1R2c1A4bVJ3UzVYL21EUUx4L0g5bCs2VGtTM1gvUkdjTnJIcmJCZ3A4?= =?utf-8?B?KzY2dkR1TWxMV0VvSjJCaGZCVzRUYThzNXRGMWc0clFkRFlQRE9xYk1iWmlr?= =?utf-8?B?TjlkRStGamlpOHBXS3BnMHRPbzFrUFBiOUhqbFQwUWU1aUYxemZCYWRYb25C?= =?utf-8?B?QU9vam5YRnVLTmpicjZ2T280MS8xclJGcmVGY2M5emkvM2MvS1h5bmpBRmlu?= =?utf-8?B?c3RRdVA4a3RtY1ZWUzZTSTJSc0JJdUppVUZCQUk0enFQYTROQi9VWjdycVY2?= =?utf-8?B?UDZCeWd3ZEhCUXVrN09oTjRLajgyZy9wVFdLSkVJeTJ0a3k1TWlFeGdobmZz?= =?utf-8?B?VW5LYmM4a1JYeHZ4RkV3WHN4akFrQ1RIUUF5aDZaVHpDUTFUUmo1b25Vd1FJ?= =?utf-8?B?TSsxKzl0UUJYaDkrd1hMM1dUQ2xhQUFnQmFWUUV6SGxPb0R4czh6UE5mTnJC?= =?utf-8?B?NUFybFU3TmhUUlMyaEd5WXo4QzUxTS91SWx0VHBRU08yTmg2QStQQ2FxaHhC?= =?utf-8?B?d2NSMmM1RXJ6VUZ3QW9CK3ludFpLdmUvQWUzYkVvVitISkx5S21wRVYzYU8v?= =?utf-8?B?OExJVG01dTg2dU9SQkZwT3EyVE9SMVcvMUVMalRjQTRKR3QvWmtBdFpPck5Q?= =?utf-8?B?RGk0c2NrZldVR1BtMk5KUHAzRVFCaDUwR0dvQVJsYjE3MFNNdElRajZNUXNy?= =?utf-8?B?YldHRlRyK1FzdmlRMXdzUnlZVVhnQmU4Um02ajlydXg5Rm85R0xtT0xWRlFG?= =?utf-8?B?bTNjU0hZUkxyM3kxdFVZRVE2YzZJcmRpdUlkNVExTWlZaTc2NytBSnhTdmNK?= =?utf-8?B?eUd1QmlTSjk3ZlhxbitKOTdtOUVYRHNnZGNvb1JudTRmTEE4c1liNU5tOFV5?= =?utf-8?B?eDRPZnhMazhGb2llVWJSWVJUcXBkWDNCMzdqMWtlZW5GOXIwVmgvTnFqaXQv?= =?utf-8?B?MTNyamhJK0d1RUp4Y2ZnRlg4UWdTRkx0RENmbmRjVzViUkF5RGVYMWRnRktz?= =?utf-8?B?UjRXMkU5MTVGb2pQdjFRUjRYTzZ0cW9LUHYyMmhDMzI1OXZRQVJxRlZYZmxy?= =?utf-8?B?ZkVYNFNsRmNrWTBjMmNHKzVnaGlhdWJoUjRhT29aOGZPdG5hWFJoZllORGZ3?= =?utf-8?B?dm5WSjhDNE95em9MSGJsSUc5SlVoblU0cmx6bFJYTkdpRDhhR0x5V2lISkxo?= =?utf-8?B?c2FMQnVlM2pvcjF3QzN5VVlEdUdTUW1EcFBrQ01DWForQ2I1dUhJeDRSbG1v?= =?utf-8?B?VFhxZjd3V1pnRUhSYXBZelhweHBnY1lMQnNOd3JadG1iSlZVOWd5czMvdkZv?= =?utf-8?B?R2RPblRzbVVFMUhRQkNDS08wRVRjeFNObVFid3JNWWNObTBhczVWVDNSQXFE?= =?utf-8?Q?nuVI8KZarzEhc038PpzFi8U3yAjuQc8DWsYNfIV?= X-Forefront-Antispam-Report: CIP:255.255.255.255; CTRY:; LANG:en; SCL:1; SRV:; IPV:NLI; SFV:NSPM; H:DS0PR11MB7408.namprd11.prod.outlook.com; PTR:; CAT:NONE; SFS:(13230037)(1800799021)(366013)(376011); DIR:OUT; SFP:1101; X-MS-Exchange-AntiSpam-MessageData-ChunkCount: 1 X-MS-Exchange-AntiSpam-MessageData-0: =?utf-8?B?NU1kdmVVNjhDaUhNVWx5L2lCNzZvY20vczkzTGJoa29veFhKb3ZTOEh6UU5Z?= =?utf-8?B?V2VSRlRMZS95WHlsSTVZOXo1Y3ROOXpTSEFyelFaQlErU0NNdjFRaHZFZ0p0?= =?utf-8?B?ZjFtZVQ1cmUvMGEydkcyR0xSYkMzSFFmMmdJOHEva3FCN2I5aWlkRFpXMU1s?= =?utf-8?B?UW1rQ2NzU0NvVnNFbzBKOWxRcEdxTTlhWk9EOFV2b2c5WllEVGJyRW84SVI0?= =?utf-8?B?Qlg1bDk1eERCU29SRXV5QytKK3Z3eEhSZTV2Zm5iakpWcTFnRVBtRmRZbzZw?= =?utf-8?B?VlVDQm5NVkI3eVJMOE92cWliK05VTXE5UE1OUjZUencxTVZtM3FibHRKeE1J?= =?utf-8?B?K0lIRlpHbWVXaUN2cmIwVFloQnFCM1N5UWNiTGJ4U2hCSkNlbVd1MSt0aGN3?= =?utf-8?B?WVFxTHYzaG1xQkJZbUQ5czdxanZDc296VzJXZnlBZ2doYWJOcThqME1ENmpB?= =?utf-8?B?OUpUWU4zb0dBWGNpdWVDMnhVeHRncFZxb04xSEJLUlprd0o0d0orNHc2RWhu?= =?utf-8?B?SldwbzBMR01ZKzVYTnRqZ3JYTllZbjNqZUQ4VHZ2Q3RlWXd4Ym9EejJHSGNm?= =?utf-8?B?eWpwVFlDZjN3eUdvU0VXT0Fjc1ZMR1N6ZTBSMXowUXBqZERsQzRCU0tGcFZO?= =?utf-8?B?R2tJNVlMTDZ4MXBmVGF4RmdTQVNwSFEyc0wreERqaUdJUE4xeEVWR3dlS1F0?= =?utf-8?B?cmcwL2dZM2lHVXBlWWdyWldOVHpPakhnUVQ2a3djbDV4S29PR2UzMzJsTjRL?= =?utf-8?B?T2x5a1VEUU1STlpZeFFkM0lLQWluT1B5bzRuaEsydi9nS1RvRERPa1NRME1o?= =?utf-8?B?Um00QjIzUy9oY0lPMkJOdjFZMWU0Rk5MZjNCNXlTODJVZER1OEwxWm1VMWZp?= =?utf-8?B?endHVEV0czQ5N2hLeFlYYXNxUkFDbm9Dd2Q3QmEvcEVoSkY5V2xHakRQVkxT?= =?utf-8?B?S0kwTGExQWNBR3FYMGFQSElLRGZNeHR5N2xyMkNrRmFMR3MwZHZDVHpybjY3?= =?utf-8?B?VnQxbzRsaHY4dzhwekVhWHl6MnpLYnlPT2UzUkJPOTZyNG1XaTNDWnp0N2Fp?= =?utf-8?B?b21MSk9NMHBrR1kxeDlkUU8vcTI4a2JMSE5SdTgrS3RKVWZ6V0NzV2t0YlJj?= =?utf-8?B?RHlncFhYZlE1dDBXemdpdnc1b0lvYndQSFVKM2RTNlVjUVRNNWg0RmZPb3Z2?= =?utf-8?B?c0ZNUGd4WDhzUWdmbjBoMWlNZlFQLzMwb0Z3QjkrODRHMU90WHViRmUzWk9L?= =?utf-8?B?V1pmOUl5UnY0T2h6dEQ4eXFNZUlKWVBaVUsxMVNJOUxPd0Z3SWExSWVQbElH?= =?utf-8?B?dkFaaytZb0FhU1J6a1JVdWg5SnRneTZGVDFYSjVQRGdvL3pKM3A2YlhFUERU?= =?utf-8?B?M29BWi9OSlR0R2lxejkxVno4Wk43Z2krMTBpSENtbHU0UHcxWVNHdEp4N0NK?= =?utf-8?B?bXROTGVQazdJL2FPYWs3QlFURmVhdmJoSk90ZU04d0E5VDIxWWdsdncrRlVw?= =?utf-8?B?aHN5WHplQkRpZFZsQ0xuOWFlclRWa2hWM3Q2dm5qLzJzcmpOVE1EWEpNNkxK?= =?utf-8?B?cGxnTHhSZXJhUEJYSlJXbjgwSnlLdExicllndThYU0s2RTZWTXVWNXF3NjdQ?= =?utf-8?B?T1NRdjF0RkxMaldFVkJzaEcvdDUxRThmdGg4c0hqdGtaTk03dkgwMWJQaGRW?= =?utf-8?B?RlRmclpMcnc5TDlST3dseE95SzFvNFZsUmt0M0xmMXkxZ0RLWE52R0x1ZXE5?= =?utf-8?B?NjZsUk45S05scDhQT2pEZVNCNERuSTEyMkhQZHBnNGpvZkhPUHRwcXltZWtt?= =?utf-8?B?dENMTXM1WUt0RWdYNjRQWlE1Y01MODlQbEJ0V2NrdGFpb3hyL3BWZ1VxQU5C?= =?utf-8?B?QmNUMjlDSS9IR3JEVC8yRkZobFJWZXhxbnRta2x3VXNPOVRYQTJzMk9KSEVU?= =?utf-8?B?QXFVU2sxeDZzNXhjUDdYN2k2WnlsK3VDNVZuVUowSUN4cU9mZnRjMVJLaFpi?= =?utf-8?B?TDc1OEVWSGI3dmNVbXF1cWk3dnVUSEJMQXdlNEllMUlqOHlVUTA4Qk5KbDcy?= =?utf-8?B?TEhDTmpxZXoxeExWSUlMUGVTbmNMZkR0U2pycXJORVAzVDRvNFBaTmZvQjJ0?= =?utf-8?B?UVBrUzJ2b3Urc3locHBJc2U4ZzdCUHVGSGpjakhXZjdHRDZ6TDEzSWR5bWpS?= =?utf-8?Q?SoVDQ0uy7jfx8h8BHWO/gv4=3D?= X-MS-Exchange-CrossTenant-Network-Message-Id: 39c4a93a-9a5e-4ca1-9ac9-08dc9182d728 X-MS-Exchange-CrossTenant-AuthSource: DS0PR11MB7408.namprd11.prod.outlook.com X-MS-Exchange-CrossTenant-AuthAs: Internal X-MS-Exchange-CrossTenant-OriginalArrivalTime: 20 Jun 2024 23:43:52.4494 (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: 48VK4EVScg+F1pB7hV6fk4mfnGoFubFgnZILsJrTnbr8c5j7b9AfP2quBi8WbBWF01LyQLVEuOhvh1B2buFPsbHvpwNNqpKbp+4u6ooZKdA= X-MS-Exchange-Transport-CrossTenantHeadersStamped: SJ0PR11MB4813 X-OriginatorOrg: intel.com X-BeenThere: igt-dev@lists.freedesktop.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Development mailing list for IGT GPU Tools List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: igt-dev-bounces@lists.freedesktop.org Sender: "igt-dev" On Thu, Jun 20, 2024 at 01:00:51PM -0700, Ashutosh Dixit wrote: >From: José Roberto de Souza > >Signed-off-by: José Roberto de Souza Reviewed-by: Umesh Nerlige Ramappa >--- > lib/xe/oa-configs/oa-lnl.xml | 6929 +++++++++++++++++++++++++++------- > 1 file changed, 5557 insertions(+), 1372 deletions(-) > >diff --git a/lib/xe/oa-configs/oa-lnl.xml b/lib/xe/oa-configs/oa-lnl.xml >index 3576ca2827..8ddee11de2 100644 >--- a/lib/xe/oa-configs/oa-lnl.xml >+++ b/lib/xe/oa-configs/oa-lnl.xml >@@ -1,11 +1,11 @@ > >- >+ > chipset="LNL" > symbol_name="RenderBasic" > underscore_name="render_basic" > mdapi_supported_apis="VK OGL OCL IO" >- oa_format="256B_GENERIC_NOA16" >+ oa_format="576B_PEC64LL" > hw_config_guid="12f20772-0044-44ff-bcc0-d2bc252d140e" > > > @@ -21,6 +21,839 @@ > mdapi_supported_apis="" > mdapi_hw_unit_type="gpu" > /> >+ + symbol_name="GpuCoreClocks" >+ underscore_name="gpu_core_clocks" >+ description="The total number of GPU core clocks elapsed during the measurement." >+ data_type="uint64" >+ units="cycles" >+ semantic_type="event" >+ equation="GPU_CLOCK 0 READ" >+ mdapi_group="GPU" >+ mdapi_usage_flags="Tier1 Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="AvgGpuCoreFrequency" >+ underscore_name="avg_gpu_core_frequency" >+ description="Average GPU Core Frequency in the measurement." >+ data_type="uint64" >+ max_equation="$GpuMaxFrequency" >+ units="hz" >+ semantic_type="event" >+ equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV" >+ mdapi_group="GPU" >+ mdapi_usage_flags="Tier1 Overview System Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="GPU_BUSY" >+ underscore_name="gpu_busy" >+ description="Percentage of time in which GPU is not idle including all GPU engines" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 32 READ 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Front End" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="IA_VERTEX" >+ underscore_name="ia_vertex" >+ description="Number of vertices in a draw" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 33 READ" >+ mdapi_group="Geometry" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="GPGPU_THREADGROUP_COUNT" >+ underscore_name="gpgpu_threadgroup_count" >+ description="Number of GPGPU threadgroups dispatched" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 0 READ" >+ mdapi_group="Thread Dispatcher" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="ASYNC_GPGPU_THREADGROUP_COUNT" >+ underscore_name="async_gpgpu_threadgroup_count" >+ description="Number of Async GPGPU threadgroups dispatched" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 1 READ" >+ mdapi_group="Thread Dispatcher" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="RASTERIZER_SAMPLE_OUTPUT" >+ underscore_name="rasterizer_sample_output" >+ description="Number of lit samples emitted by Rasterizer" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 34 READ" >+ mdapi_group="Rasterizer" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="PIXEL_BLEND" >+ underscore_name="pixel_blend" >+ description="Number of blended pixels written to all render targets" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 35 READ" >+ mdapi_group="Color Pipe" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="PIXEL_WRITE" >+ underscore_name="pixel_write" >+ description="Number of pixels written to all render targets" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 36 READ" >+ mdapi_group="Color Pipe" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="POSTPS_ALPHA_TEST_FAIL" >+ underscore_name="postps_alpha_test_fail" >+ description="Number of 2x2 that were lit from Rasterizer but failed the alpha test" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 2 READ" >+ mdapi_group="Color Pipe" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="SHADER_PIX_KILL" >+ underscore_name="shader_pix_kill" >+ description="Number of pixels that were lit from Rasterizer but killed by the Pixel Shader" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 3 READ" >+ mdapi_group="Color Pipe" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="L3_READ" >+ underscore_name="l3_read" >+ description="Number of Device Cache 64B read requests" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 37 READ" >+ mdapi_group="Device Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="L3_WRITE" >+ underscore_name="l3_write" >+ description="Number of Device Cache 64B write requests" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 38 READ" >+ mdapi_group="Device Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="COMMAND_PARSER_COMPUTE_ENGINE_BUSY" >+ underscore_name="command_parser_compute_engine_busy" >+ description="Percentage of time in which there is a context loaded and active on the compute queue" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 39 READ $ComputeEngineTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Front End" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="COMMAND_PARSER_RENDER_ENGINE_BUSY" >+ underscore_name="command_parser_render_engine_busy" >+ description="Percentage of time in which there is a context loaded and active on the 3D queue" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 40 READ 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Front End" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="COMMAND_PARSER_COPY_ENGINE_BUSY" >+ underscore_name="command_parser_copy_engine_busy" >+ description="Percentage of time in which there is a context loaded and active on the copy queue." >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 41 READ 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Front End" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="CLIPPER_INPUT_VERTEX" >+ underscore_name="clipper_input_vertex" >+ description="Number of Clipper input vertices" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 42 READ" >+ mdapi_group="Geometry" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="CLIPPER_PRIMITIVE_CULL" >+ underscore_name="clipper_primitive_cull" >+ description="Number of Clipper early cull primitives" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 43 READ" >+ mdapi_group="Geometry" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="CLIPPER_PRIMITIVE_FAR_NEAR_CLIP" >+ underscore_name="clipper_primitive_far_near_clip" >+ description="Number of primitives clipped by Clipper due to near/far planes" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 44 READ" >+ mdapi_group="Geometry" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="CLIPPER_PRIMITIVE_OUTPUT" >+ underscore_name="clipper_primitive_output" >+ description="Number of primitives going out of Clipper, must clip plus the trivial accept. Must clip primitives are counted one time only." >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 45 READ" >+ mdapi_group="Geometry" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="IA_PRIMITIVE" >+ underscore_name="ia_primitive" >+ description="Number of primitives in a draw" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 46 READ" >+ mdapi_group="Geometry" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="STREAMOUT_OUTPUT_VERTEX_COUNT" >+ underscore_name="streamout_output_vertex_count" >+ description="Number of vertices written by SOL" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 47 READ" >+ mdapi_group="Geometry" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="STRIPSFAN_OBJECT_COUNT" >+ underscore_name="stripsfan_object_count" >+ description="Number of objects exiting Stripsfan stage" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 48 READ" >+ mdapi_group="Geometry" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="STRIPSFAN_OBJECT_CULL" >+ underscore_name="stripsfan_object_cull" >+ description="Number of simple culled objects in Stripsfan stage" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 49 READ" >+ mdapi_group="Geometry" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="STREAMOUT_PRIMITIVE_WRITE_COUNT" >+ underscore_name="streamout_primitive_write_count" >+ description="Number of objects data written to memory, read from pipe 0 only" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 50 READ" >+ mdapi_group="Geometry" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="SLM_BANK_CONFLICT_COUNT" >+ underscore_name="slm_bank_conflict_count" >+ description="Number of SLM accesses resulting in a bank conflict" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 4 READ" >+ mdapi_group="L1 Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="COMPRESSOR_INPUT" >+ underscore_name="compressor_input" >+ description="Number of 256B writes to compressible surfaces at the input of the compressor" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 51 READ" >+ mdapi_group="Memory" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="COMPRESSOR_OUTPUT" >+ underscore_name="compressor_output" >+ description="Number of 256B writes to compressible surfaces at the output of the compressor" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 52 READ" >+ mdapi_group="Memory" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="TLB_MISS" >+ underscore_name="tlb_miss" >+ description="Number of misses from all TLBs and all engines" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 53 READ" >+ mdapi_group="Memory" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="SAMPLER_2X2_READ" >+ underscore_name="sampler_2_x2_read" >+ description="Number of 2x2 texel block requested from Sampler" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 5 READ" >+ mdapi_group="Sampler" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="SAMPLER_TEXTURE_CACHE_ACCESS" >+ underscore_name="sampler_texture_cache_access" >+ description="Number of Sampler L1 requests" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 6 READ" >+ mdapi_group="Sampler" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="SAMPLER_TEXTURE_CACHE_MISS" >+ underscore_name="sampler_texture_cache_miss" >+ description="Number of Sampler L1 misses" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 21 READ" >+ mdapi_group="Sampler" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="URB_READ" >+ underscore_name="urb_read" >+ description="Number of URB reads" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 55 READ" >+ mdapi_group="URB" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="URB_WRITE" >+ underscore_name="urb_write" >+ description="Number of URB writes" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 56 READ" >+ mdapi_group="URB" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="ICACHE_HIT" >+ underscore_name="icache_hit" >+ description="Number of Instruction Cache hits" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 22 READ" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="ICACHE_MISS" >+ underscore_name="icache_miss" >+ description="Number of Instruction Cache misses" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 23 READ" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_ACTIVE" >+ underscore_name="xve_active" >+ description="Percentage of time in which at least one pipe is active in XVE" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 24 READ $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_INST_EXECUTED_ALU0_ALL" >+ underscore_name="xve_inst_executed_alu0_all" >+ description="Number of execution slots taken by instructions executed by ALL threads on ALU0 pipe" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 25 READ" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_INST_EXECUTED_ALU1_ALL" >+ underscore_name="xve_inst_executed_alu1_all" >+ description="Number of execution slots taken by instructions executed by ALL threads on ALU1 pipe" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 26 READ" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_INST_EXECUTED_SEND_ALL" >+ underscore_name="xve_inst_executed_send_all" >+ description="Number of instruction (GRF) dispatches executed by ALL threads on SEND Pipe" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 7 READ" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_INST_EXECUTED_ALU0_CS" >+ underscore_name="xve_inst_executed_alu0_cs" >+ description="Number of execution slots taken by instructions executed by CS threads on ALU0 pipe" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 8 READ" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_INST_EXECUTED_ALU0_PS" >+ underscore_name="xve_inst_executed_alu0_ps" >+ description="Number of execution slots taken by instructions executed by PS threads on ALU0 pipe" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 9 READ" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_INST_EXECUTED_ALU0_VS" >+ underscore_name="xve_inst_executed_alu0_vs" >+ description="Number of execution slots taken by instructions executed by VS threads on ALU0 pipe" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 10 READ" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_INST_EXECUTED_ALU1_CS" >+ underscore_name="xve_inst_executed_alu1_cs" >+ description="Number of execution slots taken by instructions executed by CS threads on ALU1 pipe" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 11 READ" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_INST_EXECUTED_ALU1_PS" >+ underscore_name="xve_inst_executed_alu1_ps" >+ description="Number of execution slots taken by instructions executed by PS threads on ALU1 pipe" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 12 READ" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_INST_EXECUTED_ALU1_VS" >+ underscore_name="xve_inst_executed_alu1_vs" >+ description="Number of execution slots taken by instructions executed by VS threads on ALU1 pipe" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 13 READ" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_SHARED_FUNCTION_ACCESS_HOLD" >+ underscore_name="xve_shared_function_access_hold" >+ description="Percentage of time in which XVE requests stalled by Shared Function units" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 27 READ $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="RT_BVH_CACHE_MISS" >+ underscore_name="rt_bvh_cache_miss" >+ description="Number of BVH cache misses for RayQuery only" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 28 READ" >+ mdapi_group="Ray Tracing" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="RT_INPUT_MESSAGE_RAY_COUNT" >+ underscore_name="rt_input_message_ray_count" >+ description="Number of valid SIMD lanes in the TraceRay message" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 29 READ" >+ mdapi_group="Ray Tracing" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="LOAD_STORE_CACHE_HIT" >+ underscore_name="load_store_cache_hit" >+ description="Number of Load Store Cache hits" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 30 READ" >+ mdapi_group="L1 Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="LOAD_STORE_CACHE_ACCESS" >+ underscore_name="load_store_cache_access" >+ description="Number of Load Store Cache accesses" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 31 READ" >+ mdapi_group="L1 Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_THREADS_OCCUPANCY_ALL" >+ underscore_name="xve_threads_occupancy_all" >+ description="Percentage of thread slots occupied by ALL threads" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 14 READ 8 UMUL $VectorEngineThreadsCount $VectorEngineTotalCount UMUL UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_INST_ISSUED_ALL" >+ underscore_name="xve_inst_issued_all" >+ description="Number of instructions issued (decoded) to any pipe" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 15 READ" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="GPU_MEMORY_BYTE_READ" >+ underscore_name="gpu_memory_byte_read" >+ description="Number of device local memory (HBM, GDDR, LPDDR, etc.) read bytes" >+ data_type="uint64" >+ units="bytes" >+ semantic_type="event" >+ equation="PEC 57 READ 2 UMUL" >+ mdapi_group="Memory" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="GPU_MEMORY_BYTE_WRITE" >+ underscore_name="gpu_memory_byte_write" >+ description="Number of device local memory (HBM, GDDR, LPDDR, etc.) write bytes" >+ data_type="uint64" >+ units="bytes" >+ semantic_type="event" >+ equation="PEC 58 READ 2 UMUL" >+ mdapi_group="Memory" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="LOAD_STORE_CACHE_BYTE_READ" >+ underscore_name="load_store_cache_byte_read" >+ description="Number of bytes read out of the Load Store Cache, excluding SLM accesses" >+ data_type="uint64" >+ units="bytes" >+ semantic_type="event" >+ equation="PEC 16 READ" >+ mdapi_group="L1 Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="LOAD_STORE_CACHE_BYTE_WRITE" >+ underscore_name="load_store_cache_byte_write" >+ description="Number of bytes written to the Load Store Cache, excluding SLM accesses" >+ data_type="uint64" >+ units="bytes" >+ semantic_type="event" >+ equation="PEC 17 READ" >+ mdapi_group="L1 Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="SLM_BYTE_READ" >+ underscore_name="slm_byte_read" >+ description="Number of bytes read from SLM" >+ data_type="uint64" >+ units="bytes" >+ semantic_type="event" >+ equation="PEC 18 READ" >+ mdapi_group="L1 Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="SLM_BYTE_WRITE" >+ underscore_name="slm_byte_write" >+ description="Number of bytes written to SLM" >+ data_type="uint64" >+ units="bytes" >+ semantic_type="event" >+ equation="PEC 19 READ" >+ mdapi_group="L1 Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="GPU_MEMORY_BYTE_READ_RATE" >+ underscore_name="gpu_memory_byte_read_rate" >+ description="Device local memory (HBM, GDDR, LPDDR, etc.) read bandwidth" >+ data_type="float" >+ units="gbps" >+ semantic_type="throughput" >+ equation="PEC 57 READ 2 UMUL $GpuTime FDIV" >+ mdapi_group="Memory" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="GPU_MEMORY_BYTE_WRITE_RATE" >+ underscore_name="gpu_memory_byte_write_rate" >+ description="Device local memory (HBM, GDDR, LPDDR, etc.) write bandwidth" >+ data_type="float" >+ units="gbps" >+ semantic_type="throughput" >+ equation="PEC 58 READ 2 UMUL $GpuTime FDIV" >+ mdapi_group="Memory" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_STALL" >+ underscore_name="xve_stall" >+ description="Percentage of time in which any thread loaded but not even a single pipe is active in XVE" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 20 READ $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_INST_EXECUTED_ALU0_ALL_UTILIZATION" >+ underscore_name="xve_inst_executed_alu0_all_utilization" >+ description="Percentage of time taken by instructions executed by ALL threads on ALU0 pipe" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 25 READ $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_INST_EXECUTED_ALU1_ALL_UTILIZATION" >+ underscore_name="xve_inst_executed_alu1_all_utilization" >+ description="Percentage of time taken by instructions executed by ALL threads on ALU1 pipe" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 26 READ $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> > > > >@@ -263,7 +1096,7 @@ > symbol_name="ComputeBasic" > underscore_name="compute_basic" > mdapi_supported_apis="VK OGL OCL IO" >- oa_format="256B_GENERIC_NOA16" >+ oa_format="576B_PEC64LL" > hw_config_guid="72539be3-67d9-4101-b067-766b7811d2bb" > > > @@ -279,6 +1112,709 @@ > mdapi_supported_apis="" > mdapi_hw_unit_type="gpu" > /> >+ + symbol_name="GpuCoreClocks" >+ underscore_name="gpu_core_clocks" >+ description="The total number of GPU core clocks elapsed during the measurement." >+ data_type="uint64" >+ units="cycles" >+ semantic_type="event" >+ equation="GPU_CLOCK 0 READ" >+ mdapi_group="GPU" >+ mdapi_usage_flags="Tier1 Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="AvgGpuCoreFrequency" >+ underscore_name="avg_gpu_core_frequency" >+ description="Average GPU Core Frequency in the measurement." >+ data_type="uint64" >+ max_equation="$GpuMaxFrequency" >+ units="hz" >+ semantic_type="event" >+ equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV" >+ mdapi_group="GPU" >+ mdapi_usage_flags="Tier1 Overview System Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="GPU_BUSY" >+ underscore_name="gpu_busy" >+ description="Percentage of time in which GPU is not idle including all GPU engines" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 32 READ 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Front End" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="IA_VERTEX" >+ underscore_name="ia_vertex" >+ description="Number of vertices in a draw" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 33 READ" >+ mdapi_group="Geometry" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="GPGPU_THREADGROUP_COUNT" >+ underscore_name="gpgpu_threadgroup_count" >+ description="Number of GPGPU threadgroups dispatched" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 0 READ" >+ mdapi_group="Thread Dispatcher" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="ASYNC_GPGPU_THREADGROUP_COUNT" >+ underscore_name="async_gpgpu_threadgroup_count" >+ description="Number of Async GPGPU threadgroups dispatched" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 1 READ" >+ mdapi_group="Thread Dispatcher" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="RASTERIZER_SAMPLE_OUTPUT" >+ underscore_name="rasterizer_sample_output" >+ description="Number of lit samples emitted by Rasterizer" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 34 READ" >+ mdapi_group="Rasterizer" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="ICACHE_HIT" >+ underscore_name="icache_hit" >+ description="Number of Instruction Cache hits" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 2 READ" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="ICACHE_MISS" >+ underscore_name="icache_miss" >+ description="Number of Instruction Cache misses" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 3 READ" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_ACTIVE" >+ underscore_name="xve_active" >+ description="Percentage of time in which at least one pipe is active in XVE" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 4 READ $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_INST_EXECUTED_ALU0_ALL" >+ underscore_name="xve_inst_executed_alu0_all" >+ description="Number of execution slots taken by instructions executed by ALL threads on ALU0 pipe" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 5 READ" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_INST_EXECUTED_ALU1_ALL" >+ underscore_name="xve_inst_executed_alu1_all" >+ description="Number of execution slots taken by instructions executed by ALL threads on ALU1 pipe" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 6 READ" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_INST_EXECUTED_SEND_ALL" >+ underscore_name="xve_inst_executed_send_all" >+ description="Number of instruction (GRF) dispatches executed by ALL threads on SEND Pipe" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 7 READ" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_INST_ISSUED_ALL" >+ underscore_name="xve_inst_issued_all" >+ description="Number of instructions issued (decoded) to any pipe" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 21 READ" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_SHARED_FUNCTION_ACCESS_HOLD" >+ underscore_name="xve_shared_function_access_hold" >+ description="Percentage of time in which XVE requests stalled by Shared Function units" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 22 READ $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_STALL" >+ underscore_name="xve_stall" >+ description="Percentage of time in which any thread loaded but not even a single pipe is active in XVE" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 23 READ $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_THREADS_OCCUPANCY_ALL" >+ underscore_name="xve_threads_occupancy_all" >+ description="Percentage of thread slots occupied by ALL threads" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 8 READ 8 UMUL $VectorEngineThreadsCount $VectorEngineTotalCount UMUL UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_INST_EXECUTED_ALU2_ALL" >+ underscore_name="xve_inst_executed_alu2_all" >+ description="Number of execution slots taken by instructions executed in ALU2 pipe" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 24 READ" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_MULTIPLE_PIPE_ACTIVE" >+ underscore_name="xve_multiple_pipe_active" >+ description="Percentage of time in which at least two pipes are actively executing a Gen ISA instruction among ALU0, ALU1 and ALU2 pipes" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 25 READ $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="LOAD_STORE_CACHE_PARTIAL_WRITE_COUNT" >+ underscore_name="load_store_cache_partial_write_count" >+ description="Number of writes to the Load Store Cache that don't fill a subsector" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 26 READ" >+ mdapi_group="L1 Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="SLM_BANK_CONFLICT_COUNT" >+ underscore_name="slm_bank_conflict_count" >+ description="Number of SLM accesses resulting in a bank conflict" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 27 READ" >+ mdapi_group="L1 Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="SLM_BYTE_READ" >+ underscore_name="slm_byte_read" >+ description="Number of bytes read from SLM" >+ data_type="uint64" >+ units="bytes" >+ semantic_type="event" >+ equation="PEC 28 READ" >+ mdapi_group="L1 Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="SLM_BYTE_WRITE" >+ underscore_name="slm_byte_write" >+ description="Number of bytes written to SLM" >+ data_type="uint64" >+ units="bytes" >+ semantic_type="event" >+ equation="PEC 29 READ" >+ mdapi_group="L1 Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="LOAD_STORE_CACHE_BYTE_READ" >+ underscore_name="load_store_cache_byte_read" >+ description="Number of bytes read out of the Load Store Cache, excluding SLM accesses" >+ data_type="uint64" >+ units="bytes" >+ semantic_type="event" >+ equation="PEC 30 READ" >+ mdapi_group="L1 Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="LOAD_STORE_CACHE_BYTE_WRITE" >+ underscore_name="load_store_cache_byte_write" >+ description="Number of bytes written to the Load Store Cache, excluding SLM accesses" >+ data_type="uint64" >+ units="bytes" >+ semantic_type="event" >+ equation="PEC 31 READ" >+ mdapi_group="L1 Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="LOAD_STORE_CACHE_ACCESS" >+ underscore_name="load_store_cache_access" >+ description="Number of Load Store Cache accesses" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 9 READ" >+ mdapi_group="L1 Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="LOAD_STORE_CACHE_HIT" >+ underscore_name="load_store_cache_hit" >+ description="Number of Load Store Cache hits" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 10 READ" >+ mdapi_group="L1 Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="L3_ATOMIC_ACCESS" >+ underscore_name="l3_atomic_access" >+ description="Number of atomic accesses to Device Cache bank" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 35 READ" >+ mdapi_group="Device Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="L3_HIT" >+ underscore_name="l3_hit" >+ description="Number of Device Cache accesses (except atomics) which hits in the Device Cache cache" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 36 READ" >+ mdapi_group="Device Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="L3_MISS" >+ underscore_name="l3_miss" >+ description="Number of Device Cache accesses (except atomics) which miss in the Device Cache cache" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 37 READ" >+ mdapi_group="Device Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="L3_READ" >+ underscore_name="l3_read" >+ description="Number of Device Cache 64B read requests" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 38 READ" >+ mdapi_group="Device Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="L3_WRITE" >+ underscore_name="l3_write" >+ description="Number of Device Cache 64B write requests" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 39 READ" >+ mdapi_group="Device Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="COMPRESSOR_INPUT" >+ underscore_name="compressor_input" >+ description="Number of 256B writes to compressible surfaces at the input of the compressor" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 41 READ" >+ mdapi_group="Memory" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="COMPRESSOR_OUTPUT" >+ underscore_name="compressor_output" >+ description="Number of 256B writes to compressible surfaces at the output of the compressor" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 42 READ" >+ mdapi_group="Memory" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="GPU_MEMORY_BYTE_READ" >+ underscore_name="gpu_memory_byte_read" >+ description="Number of device local memory (HBM, GDDR, LPDDR, etc.) read bytes" >+ data_type="uint64" >+ units="bytes" >+ semantic_type="event" >+ equation="PEC 43 READ 2 UMUL" >+ mdapi_group="Memory" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="GPU_MEMORY_BYTE_WRITE" >+ underscore_name="gpu_memory_byte_write" >+ description="Number of device local memory (HBM, GDDR, LPDDR, etc.) write bytes" >+ data_type="uint64" >+ units="bytes" >+ semantic_type="event" >+ equation="PEC 44 READ 2 UMUL" >+ mdapi_group="Memory" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="GPU_MEMORY_BYTE_READ_RATE" >+ underscore_name="gpu_memory_byte_read_rate" >+ description="Device local memory (HBM, GDDR, LPDDR, etc.) read bandwidth" >+ data_type="float" >+ units="gbps" >+ semantic_type="throughput" >+ equation="PEC 43 READ 2 UMUL $GpuTime FDIV" >+ mdapi_group="Memory" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="GPU_MEMORY_BYTE_WRITE_RATE" >+ underscore_name="gpu_memory_byte_write_rate" >+ description="Device local memory (HBM, GDDR, LPDDR, etc.) write bandwidth" >+ data_type="float" >+ units="gbps" >+ semantic_type="throughput" >+ equation="PEC 44 READ 2 UMUL $GpuTime FDIV" >+ mdapi_group="Memory" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="GPU_MEMORY_L3_READ" >+ underscore_name="gpu_memory_l3_read" >+ description="Number of GTI memory reads from Device Cache caused by Device Cache misses" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 45 READ" >+ mdapi_group="Memory" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="GPU_MEMORY_L3_WRITE" >+ underscore_name="gpu_memory_l3_write" >+ description="Number of GTI memory writes from Device Cache caused by Device Cache invalidations" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 46 READ" >+ mdapi_group="Memory" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="TLB_MISS" >+ underscore_name="tlb_miss" >+ description="Number of misses from all TLBs and all engines" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 48 READ" >+ mdapi_group="Memory" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="ASYNC_GPGPU_THREAD_EXIT_COUNT" >+ underscore_name="async_gpgpu_thread_exit_count" >+ description="Number of Async GPGPU EOT messages received" >+ data_type="uint64" >+ units="messages" >+ semantic_type="event" >+ equation="PEC 11 READ" >+ mdapi_group="Thread Dispatcher" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="GPGPU_DISPATCH" >+ underscore_name="gpgpu_dispatch" >+ description="Percentage of time in which spent dispatching GPGPU threads to XVEs" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 12 READ $XeCoreTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Thread Dispatcher" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="COMMAND_PARSER_COMPUTE_ENGINE_BUSY" >+ underscore_name="command_parser_compute_engine_busy" >+ description="Percentage of time in which there is a context loaded and active on the compute queue" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 49 READ $ComputeEngineTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Front End" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="COMMAND_PARSER_COMPUTE_ENGINE_DISPATCH_KERNEL_COUNT" >+ underscore_name="command_parser_compute_engine_dispatch_kernel_count" >+ description="Number of compute walker commands parsed on the compute engine" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 50 READ" >+ mdapi_group="Front End" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="COMMAND_PARSER_COPY_ENGINE_BUSY" >+ underscore_name="command_parser_copy_engine_busy" >+ description="Percentage of time in which there is a context loaded and active on the copy queue." >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 51 READ 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Front End" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="COMMAND_PARSER_FLUSH_COUNT" >+ underscore_name="command_parser_flush_count" >+ description="Number of Stalling flushes at the top of the pipe" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 52 READ" >+ mdapi_group="Front End" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="COMMAND_PARSER_RENDER_ENGINE_BUSY" >+ underscore_name="command_parser_render_engine_busy" >+ description="Percentage of time in which there is a context loaded and active on the 3D queue" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 53 READ 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Front End" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="COMMAND_PARSER_RENDER_ENGINE_DISPATCH_KERNEL_COUNT" >+ underscore_name="command_parser_render_engine_dispatch_kernel_count" >+ description="Number of compute walker commands parsed on the 3D engine" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 54 READ" >+ mdapi_group="Front End" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_PIPE_ALU0_AND_ALU1_ACTIVE" >+ underscore_name="xve_pipe_alu0_and_alu1_active" >+ description="Percentage of time in which ALU0 and ALU1 pipes are both actively executing a Gen ISA instruction" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 13 READ $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_PIPE_ALU0_AND_ALU2_ACTIVE" >+ underscore_name="xve_pipe_alu0_and_alu2_active" >+ description="Percentage of time in which ALU0 and ALU2 pipes are both actively executing a Gen ISA instruction" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 14 READ $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_INST_EXECUTED_ALU0_ALL_UTILIZATION" >+ underscore_name="xve_inst_executed_alu0_all_utilization" >+ description="Percentage of time taken by instructions executed by ALL threads on ALU0 pipe" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 5 READ $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_INST_EXECUTED_ALU1_ALL_UTILIZATION" >+ underscore_name="xve_inst_executed_alu1_all_utilization" >+ description="Percentage of time taken by instructions executed by ALL threads on ALU1 pipe" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 6 READ $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_INST_EXECUTED_ALU2_ALL_UTILIZATION" >+ underscore_name="xve_inst_executed_alu2_all_utilization" >+ description="Percentage of time taken by instructions executed in ALU2 pipe" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 24 READ $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> > > > >@@ -481,7 +2017,7 @@ > symbol_name="DepthProfile" > underscore_name="depth_profile" > mdapi_supported_apis="VK OGL OCL IO" >- oa_format="256B_GENERIC_NOA16" >+ oa_format="576B_PEC64LL" > hw_config_guid="cae0b2ca-c52a-4111-b8bf-5af070677d57" > > > @@ -497,6 +2033,332 @@ > mdapi_supported_apis="" > mdapi_hw_unit_type="gpu" > /> >+ + symbol_name="GpuCoreClocks" >+ underscore_name="gpu_core_clocks" >+ description="The total number of GPU core clocks elapsed during the measurement." >+ data_type="uint64" >+ units="cycles" >+ semantic_type="event" >+ equation="GPU_CLOCK 0 READ" >+ mdapi_group="GPU" >+ mdapi_usage_flags="Tier1 Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="AvgGpuCoreFrequency" >+ underscore_name="avg_gpu_core_frequency" >+ description="Average GPU Core Frequency in the measurement." >+ data_type="uint64" >+ max_equation="$GpuMaxFrequency" >+ units="hz" >+ semantic_type="event" >+ equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV" >+ mdapi_group="GPU" >+ mdapi_usage_flags="Tier1 Overview System Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="GPU_BUSY" >+ underscore_name="gpu_busy" >+ description="Percentage of time in which GPU is not idle including all GPU engines" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 32 READ 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Front End" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="IA_VERTEX" >+ underscore_name="ia_vertex" >+ description="Number of vertices in a draw" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 33 READ" >+ mdapi_group="Geometry" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="GPGPU_THREADGROUP_COUNT" >+ underscore_name="gpgpu_threadgroup_count" >+ description="Number of GPGPU threadgroups dispatched" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 0 READ" >+ mdapi_group="Thread Dispatcher" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="ASYNC_GPGPU_THREADGROUP_COUNT" >+ underscore_name="async_gpgpu_threadgroup_count" >+ description="Number of Async GPGPU threadgroups dispatched" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 1 READ" >+ mdapi_group="Thread Dispatcher" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="RASTERIZER_SAMPLE_OUTPUT" >+ underscore_name="rasterizer_sample_output" >+ description="Number of lit samples emitted by Rasterizer" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 34 READ" >+ mdapi_group="Rasterizer" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="EARLY_DEPTH_STENCIL_TEST_FAIL_NP" >+ underscore_name="early_depth_stencil_test_fail_np" >+ description="Number of non-promoted 2x2 that failed Depth / Stencil before the Pixel Shader that were previously ambiguous at HiZ" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 35 READ" >+ mdapi_group="Depth Pipe" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="EARLY_DEPTH_STENCIL_TEST_FAIL_P" >+ underscore_name="early_depth_stencil_test_fail_p" >+ description="Number of promoted 2x2 that failed Depth / Stencil that were previously ambiguous at HiZ" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 36 READ" >+ mdapi_group="Depth Pipe" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="HIZ_DEPTH_TEST_AMBIG_NP" >+ underscore_name="hiz_depth_test_ambig_np" >+ description="Number of non-promoted 2x2 that are ambiguous by hierarchical depth test" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 37 READ" >+ mdapi_group="Depth Pipe" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="HIZ_DEPTH_TEST_AMBIG_P" >+ underscore_name="hiz_depth_test_ambig_p" >+ description="Number of promoted 2x2 that are ambiguous by hierarchical depth test" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 38 READ" >+ mdapi_group="Depth Pipe" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="HIZ_DEPTH_TEST_FAIL_NP" >+ underscore_name="hiz_depth_test_fail_np" >+ description="Number of non-promoted 2x2 that failed the hierarchical depth test" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 39 READ" >+ mdapi_group="Depth Pipe" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="HIZ_DEPTH_TEST_FAIL_P" >+ underscore_name="hiz_depth_test_fail_p" >+ description="Number of promoted 2x2 that failed the hierarchical depth test" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 40 READ" >+ mdapi_group="Depth Pipe" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="HIZ_DEPTH_TEST_PASS_P" >+ underscore_name="hiz_depth_test_pass_p" >+ description="Number of promoted 2x2 passed by hierarchical depth test" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 41 READ" >+ mdapi_group="Depth Pipe" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="HIZ_SUBSPAN_LATENCY_FIFOFULL" >+ underscore_name="hiz_subspan_latency_fifofull" >+ description="Percentage of time in which for which HiZ latency hiding structure full" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 42 READ $SliceTotalCount 2 UMUL UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Depth Pipe" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="IZ_OUTPUT_READY" >+ underscore_name="iz_output_ready" >+ description="Percentage of time in which IZ has requests to different clients" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 43 READ $SliceTotalCount 2 UMUL UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Depth Pipe" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="IZ_SUBSPAN_LATENCY_FIFOFULL" >+ underscore_name="iz_subspan_latency_fifofull" >+ description="Percentage of time in which for which IZ latency hiding structure full" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 44 READ $SliceTotalCount 2 UMUL UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Depth Pipe" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="POSTPS_DEPTH_STENCIL_TEST_FAIL" >+ underscore_name="postps_depth_stencil_test_fail" >+ description="Number of 2x2 that were lit from Rasterizer but failed the depth or stencil test post shader" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 45 READ" >+ mdapi_group="Depth Pipe" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="RENDER_CACHE_ALLOC" >+ underscore_name="render_cache_alloc" >+ description="Number of Render Cache allocations" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 46 READ" >+ mdapi_group="Color Pipe" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="RENDER_CACHE_HIT" >+ underscore_name="render_cache_hit" >+ description="Number of Render Cache hits" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 47 READ" >+ mdapi_group="Color Pipe" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="RENDER_CACHE_READ" >+ underscore_name="render_cache_read" >+ description="Number of Render Cache reads" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 48 READ" >+ mdapi_group="Color Pipe" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="RENDER_CACHE_WRITE" >+ underscore_name="render_cache_write" >+ description="Number of Render Cache writes" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 49 READ" >+ mdapi_group="Color Pipe" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="URB_CROSS_SLICE_READ" >+ underscore_name="urb_cross_slice_read" >+ description="Number of URB reads occurred from cross slices" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 50 READ" >+ mdapi_group="URB" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="PIXEL_2X2_LIT_POST_RASTERIZER_EARLY_DEPTH" >+ underscore_name="pixel_2_x2_lit_post_rasterizer_early_depth" >+ description="Number of promoted 2x2 that are lit from Rasterizer" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 51 READ" >+ mdapi_group="Rasterizer" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="PIXEL_2X2_LIT_POST_RASTERIZER_LATE_DEPTH" >+ underscore_name="pixel_2_x2_lit_post_rasterizer_late_depth" >+ description="Number of non-promoted 2x2 that are lit from Rasterizer" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 52 READ" >+ mdapi_group="Rasterizer" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> > > > >@@ -642,7 +2504,7 @@ > symbol_name="DeviceCacheProfile" > underscore_name="device_cache_profile" > mdapi_supported_apis="VK OGL OCL IO" >- oa_format="256B_GENERIC_NOA16" >+ oa_format="576B_PEC64LL" > hw_config_guid="70cc56cf-3a37-4f6c-8911-4455fe441b2a" > > > @@ -650,11 +2512,415 @@ > underscore_name="gpu_time" > description="Time elapsed on the GPU during the measurement." > data_type="uint64" >- units="ns" >+ units="ns" >+ semantic_type="event" >+ equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV" >+ mdapi_group="GPU" >+ mdapi_usage_flags="Tier1 Overview System Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="GpuCoreClocks" >+ underscore_name="gpu_core_clocks" >+ description="The total number of GPU core clocks elapsed during the measurement." >+ data_type="uint64" >+ units="cycles" >+ semantic_type="event" >+ equation="GPU_CLOCK 0 READ" >+ mdapi_group="GPU" >+ mdapi_usage_flags="Tier1 Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="AvgGpuCoreFrequency" >+ underscore_name="avg_gpu_core_frequency" >+ description="Average GPU Core Frequency in the measurement." >+ data_type="uint64" >+ max_equation="$GpuMaxFrequency" >+ units="hz" >+ semantic_type="event" >+ equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV" >+ mdapi_group="GPU" >+ mdapi_usage_flags="Tier1 Overview System Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="GPU_BUSY" >+ underscore_name="gpu_busy" >+ description="Percentage of time in which GPU is not idle including all GPU engines" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 32 READ 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Front End" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="IA_VERTEX" >+ underscore_name="ia_vertex" >+ description="Number of vertices in a draw" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 33 READ" >+ mdapi_group="Geometry" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="GPGPU_THREADGROUP_COUNT" >+ underscore_name="gpgpu_threadgroup_count" >+ description="Number of GPGPU threadgroups dispatched" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 0 READ" >+ mdapi_group="Thread Dispatcher" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="ASYNC_GPGPU_THREADGROUP_COUNT" >+ underscore_name="async_gpgpu_threadgroup_count" >+ description="Number of Async GPGPU threadgroups dispatched" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 1 READ" >+ mdapi_group="Thread Dispatcher" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="RASTERIZER_SAMPLE_OUTPUT" >+ underscore_name="rasterizer_sample_output" >+ description="Number of lit samples emitted by Rasterizer" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 34 READ" >+ mdapi_group="Rasterizer" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="AMFS_L3_ACCESS" >+ underscore_name="amfs_l3_access" >+ description="Number of AMFS access to Device Cache" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 35 READ" >+ mdapi_group="Device Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="AMFS_L3_ATOMIC" >+ underscore_name="amfs_l3_atomic" >+ description="Number of AMFS atomics to Device Cache" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 36 READ" >+ mdapi_group="Device Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="AMFS_L3_HIT" >+ underscore_name="amfs_l3_hit" >+ description="Number of AMFS hits (except atomics) in Device Cache" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 37 READ" >+ mdapi_group="Device Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="AMFS_L3_WRITE" >+ underscore_name="amfs_l3_write" >+ description="Number of AMFS writes to Device Cache" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 38 READ" >+ mdapi_group="Device Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="COLOR_L3_ACCESS" >+ underscore_name="color_l3_access" >+ description="Number of Device Cache requests resulting from color local cache misses" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 39 READ" >+ mdapi_group="Device Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="COLOR_L3_HIT" >+ underscore_name="color_l3_hit" >+ description="Number of Device Cache requests resulting from color local cache miss which hits in Device Cache" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 40 READ" >+ mdapi_group="Device Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="ICACHE_L3_HIT" >+ underscore_name="icache_l3_hit" >+ description="Number of XVE instruction cache requests that hit the Device Cache" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 41 READ" >+ mdapi_group="Device Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="ICACHE_L3_READ" >+ underscore_name="icache_l3_read" >+ description="Number of Device Cache read requests coming from XVE Instruction Cache" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 42 READ" >+ mdapi_group="Device Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="LOAD_STORE_CACHE_L3_HIT" >+ underscore_name="load_store_cache_l3_hit" >+ description="Number of Load Store Cache requests (except atomics) to Device Cache that are hit" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 43 READ" >+ mdapi_group="Device Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="LOAD_STORE_CACHE_L3_READ" >+ underscore_name="load_store_cache_l3_read" >+ description="Number of cacheline read requests from the Load Store Cache to Device Cache" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 44 READ" >+ mdapi_group="Device Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="LOAD_STORE_CACHE_L3_WRITE" >+ underscore_name="load_store_cache_l3_write" >+ description="Number of cacheline write requests from the Load Store Cache to Device Cache" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 45 READ" >+ mdapi_group="Device Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="SAMPLER_L3_HIT" >+ underscore_name="sampler_l3_hit" >+ description="Number of Device Cache hit requests resulting from sampler local cache miss which hits in Device Cache" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 46 READ" >+ mdapi_group="Device Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="SAMPLER_L3_READ" >+ underscore_name="sampler_l3_read" >+ description="Number of Device Cache read requests resulting from sampler local cache miss" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 47 READ" >+ mdapi_group="Device Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="Z_L3_ACCESS" >+ underscore_name="z_l3_access" >+ description="Number of Device Cache requests resulting from Z local cache misses" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 48 READ" >+ mdapi_group="Device Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="Z_L3_HIT" >+ underscore_name="z_l3_hit" >+ description="Number of Device Cache requests resulting from Z local cache miss which hits in Device Cache" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 49 READ" >+ mdapi_group="Device Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="L3_ATOMIC_ACCESS" >+ underscore_name="l3_atomic_access" >+ description="Number of atomic accesses to Device Cache bank" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 54 READ" >+ mdapi_group="Device Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="L3_HIT" >+ underscore_name="l3_hit" >+ description="Number of Device Cache accesses (except atomics) which hits in the Device Cache cache" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 55 READ" >+ mdapi_group="Device Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="L3_MISS" >+ underscore_name="l3_miss" >+ description="Number of Device Cache accesses (except atomics) which miss in the Device Cache cache" >+ data_type="uint64" >+ units="events" > semantic_type="event" >- equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV" >- mdapi_group="GPU" >- mdapi_usage_flags="Tier1 Overview System Frame Batch Draw" >+ equation="PEC 56 READ" >+ mdapi_group="Device Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="L3_READ" >+ underscore_name="l3_read" >+ description="Number of Device Cache 64B read requests" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 57 READ" >+ mdapi_group="Device Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="L3_WRITE" >+ underscore_name="l3_write" >+ description="Number of Device Cache 64B write requests" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 59 READ" >+ mdapi_group="Device Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="GPU_MEMORY_L3_READ" >+ underscore_name="gpu_memory_l3_read" >+ description="Number of GTI memory reads from Device Cache caused by Device Cache misses" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 60 READ" >+ mdapi_group="Memory" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="GPU_MEMORY_L3_WRITE" >+ underscore_name="gpu_memory_l3_write" >+ description="Number of GTI memory writes from Device Cache caused by Device Cache invalidations" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 61 READ" >+ mdapi_group="Memory" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="GPU_MEMORY_READ" >+ underscore_name="gpu_memory_read" >+ description="Number of GTI memory reads" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 62 READ" >+ mdapi_group="Memory" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="GPU_MEMORY_WRITE" >+ underscore_name="gpu_memory_write" >+ description="Number of GTI memory writes" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 63 READ" >+ mdapi_group="Memory" >+ mdapi_usage_flags="Frame Batch Draw" > mdapi_supported_apis="" > mdapi_hw_unit_type="gpu" > /> >@@ -863,7 +3129,7 @@ > symbol_name="MemoryProfile" > underscore_name="memory_profile" > mdapi_supported_apis="VK OGL OCL IO" >- oa_format="256B_GENERIC_NOA16" >+ oa_format="576B_PEC64LL" > hw_config_guid="01ffd51d-a6d6-4ff4-93d0-619f04c3a547" > > > @@ -871,11 +3137,662 @@ > underscore_name="gpu_time" > description="Time elapsed on the GPU during the measurement." > data_type="uint64" >- units="ns" >+ units="ns" >+ semantic_type="event" >+ equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV" >+ mdapi_group="GPU" >+ mdapi_usage_flags="Tier1 Overview System Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="GpuCoreClocks" >+ underscore_name="gpu_core_clocks" >+ description="The total number of GPU core clocks elapsed during the measurement." >+ data_type="uint64" >+ units="cycles" >+ semantic_type="event" >+ equation="GPU_CLOCK 0 READ" >+ mdapi_group="GPU" >+ mdapi_usage_flags="Tier1 Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="AvgGpuCoreFrequency" >+ underscore_name="avg_gpu_core_frequency" >+ description="Average GPU Core Frequency in the measurement." >+ data_type="uint64" >+ max_equation="$GpuMaxFrequency" >+ units="hz" >+ semantic_type="event" >+ equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV" >+ mdapi_group="GPU" >+ mdapi_usage_flags="Tier1 Overview System Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="GPU_BUSY" >+ underscore_name="gpu_busy" >+ description="Percentage of time in which GPU is not idle including all GPU engines" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 32 READ 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Front End" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="IA_VERTEX" >+ underscore_name="ia_vertex" >+ description="Number of vertices in a draw" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 33 READ" >+ mdapi_group="Geometry" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="GPGPU_THREADGROUP_COUNT" >+ underscore_name="gpgpu_threadgroup_count" >+ description="Number of GPGPU threadgroups dispatched" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 0 READ" >+ mdapi_group="Thread Dispatcher" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="ASYNC_GPGPU_THREADGROUP_COUNT" >+ underscore_name="async_gpgpu_threadgroup_count" >+ description="Number of Async GPGPU threadgroups dispatched" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 1 READ" >+ mdapi_group="Thread Dispatcher" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="RASTERIZER_SAMPLE_OUTPUT" >+ underscore_name="rasterizer_sample_output" >+ description="Number of lit samples emitted by Rasterizer" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 34 READ" >+ mdapi_group="Rasterizer" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="COMPRESSOR_INPUT" >+ underscore_name="compressor_input" >+ description="Number of 256B writes to compressible surfaces at the input of the compressor" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 35 READ" >+ mdapi_group="Memory" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="COMPRESSOR_OUTPUT" >+ underscore_name="compressor_output" >+ description="Number of 256B writes to compressible surfaces at the output of the compressor" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 36 READ" >+ mdapi_group="Memory" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="GPU_MEMORY_32B_TRANSACTION_READ" >+ underscore_name="gpu_memory_32_b_transaction_read" >+ description="Number of device local memory (HBM, GDDR, LPDDR, etc.) reads (32B)" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 37 READ" >+ mdapi_group="Memory" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="GPU_MEMORY_32B_TRANSACTION_WRITE" >+ underscore_name="gpu_memory_32_b_transaction_write" >+ description="Number of device local memory (HBM, GDDR, LPDDR, etc.) writes (32B)" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 38 READ" >+ mdapi_group="Memory" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="GPU_MEMORY_64B_TRANSACTION_READ" >+ underscore_name="gpu_memory_64_b_transaction_read" >+ description="Number of device local memory (HBM, GDDR, LPDDR, etc.) reads (64B)" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 39 READ" >+ mdapi_group="Memory" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="GPU_MEMORY_64B_TRANSACTION_WRITE" >+ underscore_name="gpu_memory_64_b_transaction_write" >+ description="Number of device local memory (HBM, GDDR, LPDDR, etc.) writes (64B)" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 40 READ" >+ mdapi_group="Memory" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="GPU_MEMORY_BYTE_READ" >+ underscore_name="gpu_memory_byte_read" >+ description="Number of device local memory (HBM, GDDR, LPDDR, etc.) read bytes" >+ data_type="uint64" >+ units="bytes" >+ semantic_type="event" >+ equation="PEC 41 READ 2 UMUL" >+ mdapi_group="Memory" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="GPU_MEMORY_BYTE_WRITE" >+ underscore_name="gpu_memory_byte_write" >+ description="Number of device local memory (HBM, GDDR, LPDDR, etc.) write bytes" >+ data_type="uint64" >+ units="bytes" >+ semantic_type="event" >+ equation="PEC 42 READ 2 UMUL" >+ mdapi_group="Memory" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="GPU_MEMORY_BYTE_READ_RATE" >+ underscore_name="gpu_memory_byte_read_rate" >+ description="Device local memory (HBM, GDDR, LPDDR, etc.) read bandwidth" >+ data_type="float" >+ units="gbps" >+ semantic_type="throughput" >+ equation="PEC 41 READ 2 UMUL $GpuTime FDIV" >+ mdapi_group="Memory" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="GPU_MEMORY_BYTE_WRITE_RATE" >+ underscore_name="gpu_memory_byte_write_rate" >+ description="Device local memory (HBM, GDDR, LPDDR, etc.) write bandwidth" >+ data_type="float" >+ units="gbps" >+ semantic_type="throughput" >+ equation="PEC 42 READ 2 UMUL $GpuTime FDIV" >+ mdapi_group="Memory" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="GPU_MEMORY_L3_READ" >+ underscore_name="gpu_memory_l3_read" >+ description="Number of GTI memory reads from Device Cache caused by Device Cache misses" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 43 READ" >+ mdapi_group="Memory" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="GPU_MEMORY_L3_WRITE" >+ underscore_name="gpu_memory_l3_write" >+ description="Number of GTI memory writes from Device Cache caused by Device Cache invalidations" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 44 READ" >+ mdapi_group="Memory" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="TLB_MISS" >+ underscore_name="tlb_miss" >+ description="Number of misses from all TLBs and all engines" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 46 READ" >+ mdapi_group="Memory" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="L3_ATOMIC_ACCESS" >+ underscore_name="l3_atomic_access" >+ description="Number of atomic accesses to Device Cache bank" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 47 READ" >+ mdapi_group="Device Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="L3_HIT" >+ underscore_name="l3_hit" >+ description="Number of Device Cache accesses (except atomics) which hits in the Device Cache cache" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 49 READ" >+ mdapi_group="Device Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="L3_MISS" >+ underscore_name="l3_miss" >+ description="Number of Device Cache accesses (except atomics) which miss in the Device Cache cache" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 51 READ" >+ mdapi_group="Device Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="L3_READ" >+ underscore_name="l3_read" >+ description="Number of Device Cache 64B read requests" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 53 READ" >+ mdapi_group="Device Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="L3_WRITE" >+ underscore_name="l3_write" >+ description="Number of Device Cache 64B write requests" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 56 READ" >+ mdapi_group="Device Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="LOAD_STORE_CACHE_ACCESS" >+ underscore_name="load_store_cache_access" >+ description="Number of Load Store Cache accesses" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 2 READ" >+ mdapi_group="L1 Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="LOAD_STORE_CACHE_BYTE_READ" >+ underscore_name="load_store_cache_byte_read" >+ description="Number of bytes read out of the Load Store Cache, excluding SLM accesses" >+ data_type="uint64" >+ units="bytes" >+ semantic_type="event" >+ equation="PEC 3 READ" >+ mdapi_group="L1 Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="LOAD_STORE_CACHE_BYTE_WRITE" >+ underscore_name="load_store_cache_byte_write" >+ description="Number of bytes written to the Load Store Cache, excluding SLM accesses" >+ data_type="uint64" >+ units="bytes" >+ semantic_type="event" >+ equation="PEC 4 READ" >+ mdapi_group="L1 Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="LOAD_STORE_CACHE_HIT" >+ underscore_name="load_store_cache_hit" >+ description="Number of Load Store Cache hits" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 5 READ" >+ mdapi_group="L1 Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="LOAD_STORE_CACHE_INPUT_AVAILABLE" >+ underscore_name="load_store_cache_input_available" >+ description="Percentage of time in which the Load Store Cache has input available" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 6 READ $XeCoreTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="L1 Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="LOAD_STORE_CACHE_NUMBER_OF_BANK_ACCESS_COUNT" >+ underscore_name="load_store_cache_number_of_bank_access_count" >+ description="Number of Load Store Cache banks accessed in a clock" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 21 READ" >+ mdapi_group="L1 Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="LOAD_STORE_CACHE_OUTPUT_READY" >+ underscore_name="load_store_cache_output_ready" >+ description="Percentage of time in which the Load Store Cache has output ready" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 22 READ $XeCoreTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="L1 Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="LOAD_STORE_CACHE_PARTIAL_WRITE_COUNT" >+ underscore_name="load_store_cache_partial_write_count" >+ description="Number of writes to the Load Store Cache that don't fill a subsector" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 23 READ" >+ mdapi_group="L1 Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="SLM_ACCESS_COUNT" >+ underscore_name="slm_access_count" >+ description="Number of SLM accesses" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 24 READ" >+ mdapi_group="L1 Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="SLM_BANK_CONFLICT_COUNT" >+ underscore_name="slm_bank_conflict_count" >+ description="Number of SLM accesses resulting in a bank conflict" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 25 READ" >+ mdapi_group="L1 Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="SLM_BYTE_READ" >+ underscore_name="slm_byte_read" >+ description="Number of bytes read from SLM" >+ data_type="uint64" >+ units="bytes" >+ semantic_type="event" >+ equation="PEC 26 READ" >+ mdapi_group="L1 Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="SLM_BYTE_WRITE" >+ underscore_name="slm_byte_write" >+ description="Number of bytes written to SLM" >+ data_type="uint64" >+ units="bytes" >+ semantic_type="event" >+ equation="PEC 27 READ" >+ mdapi_group="L1 Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_LOAD_STORE_CACHE_ATOMIC_MESSAGE_COUNT" >+ underscore_name="xve_load_store_cache_atomic_message_count" >+ description="Number of atomic operations sent by XVEs to the Load Store Cache" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 28 READ" >+ mdapi_group="L1 Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_LOAD_STORE_CACHE_FENCE_MESSAGE_COUNT" >+ underscore_name="xve_load_store_cache_fence_message_count" >+ description="Number of fence messages sent by XVEs to the Load Store Cache" >+ data_type="uint64" >+ units="messages" >+ semantic_type="event" >+ equation="PEC 29 READ" >+ mdapi_group="L1 Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_LOAD_STORE_CACHE_READ_MESSAGE_COUNT" >+ underscore_name="xve_load_store_cache_read_message_count" >+ description="Number of read messages sent by XVEs to the Load Store Cache" >+ data_type="uint64" >+ units="messages" >+ semantic_type="event" >+ equation="PEC 30 READ" >+ mdapi_group="L1 Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_LOAD_STORE_CACHE_REGISTER_REQUEST_COUNT" >+ underscore_name="xve_load_store_cache_register_request_count" >+ description="Number of message payload transactions sent by XVEs to the Load Store Cache" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 31 READ" >+ mdapi_group="L1 Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_LOAD_STORE_CACHE_REGISTER_RESPONSE_COUNT" >+ underscore_name="xve_load_store_cache_register_response_count" >+ description="Number of message payload transactions sent from the Load Store Cache back to XVEs" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 7 READ" >+ mdapi_group="L1 Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_LOAD_STORE_CACHE_WRITE_MESSAGE_COUNT" >+ underscore_name="xve_load_store_cache_write_message_count" >+ description="Number of write messages sent by XVEs to the Load Store Cache" >+ data_type="uint64" >+ units="messages" >+ semantic_type="event" >+ equation="PEC 8 READ" >+ mdapi_group="L1 Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_SLM_ATOMIC_MESSAGE_COUNT" >+ underscore_name="xve_slm_atomic_message_count" >+ description="Number of SLM atomic operations sent by XVEs" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 9 READ" >+ mdapi_group="L1 Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_SLM_FENCE_MESSAGE_COUNT" >+ underscore_name="xve_slm_fence_message_count" >+ description="Number of SLM fence operations sent by XVEs" >+ data_type="uint64" >+ units="events" > semantic_type="event" >- equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV" >- mdapi_group="GPU" >- mdapi_usage_flags="Tier1 Overview System Frame Batch Draw" >+ equation="PEC 10 READ" >+ mdapi_group="L1 Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_SLM_READ_MESSAGE_COUNT" >+ underscore_name="xve_slm_read_message_count" >+ description="Number of SLM read messages sent by XVEs" >+ data_type="uint64" >+ units="messages" >+ semantic_type="event" >+ equation="PEC 11 READ" >+ mdapi_group="L1 Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_SLM_WRITE_MESSAGE_COUNT" >+ underscore_name="xve_slm_write_message_count" >+ description="Number of SLM write messages sent by XVEs" >+ data_type="uint64" >+ units="messages" >+ semantic_type="event" >+ equation="PEC 12 READ" >+ mdapi_group="L1 Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="COPY_ENGINE_READ_REQUEST" >+ underscore_name="copy_engine_read_request" >+ description="Number of read requests sent by the Main Copy Engine" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 57 READ" >+ mdapi_group="Copy Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="COPY_ENGINE_REQUEST_STALL" >+ underscore_name="copy_engine_request_stall" >+ description="Percentage of time in which copy engine is stalled on the memory path due to stall on the read or write path when there are commands pending within copy engine" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 58 READ $CopyEngineTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Copy Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="COPY_ENGINE_WRITE_REQUEST" >+ underscore_name="copy_engine_write_request" >+ description="Number of write requests sent by the Main Copy Engine" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 59 READ" >+ mdapi_group="Copy Engine" >+ mdapi_usage_flags="Frame Batch Draw" > mdapi_supported_apis="" > mdapi_hw_unit_type="gpu" > /> >@@ -992,457 +3909,691 @@ > > > >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- - chipset="LNL" >- symbol_name="RenderPipeProfile" >- underscore_name="render_pipe_profile" >- mdapi_supported_apis="VK OGL OCL IO" >- oa_format="256B_GENERIC_NOA16" >- hw_config_guid="6791470f-f1e6-4f77-9013-bcb0c4be6f57" >- > >- - symbol_name="GpuTime" >- underscore_name="gpu_time" >- description="Time elapsed on the GPU during the measurement." >- data_type="uint64" >- units="ns" >- semantic_type="event" >- equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV" >- mdapi_group="GPU" >- mdapi_usage_flags="Tier1 Overview System Frame Batch Draw" >- mdapi_supported_apis="" >- mdapi_hw_unit_type="gpu" >- /> >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >+ >+ >+ >+ >+ >+ >+ >+ >+ >+ >+ >+ > > > >- + chipset="LNL" >- symbol_name="RTProfile" >- underscore_name="rt_profile" >+ symbol_name="RenderPipeProfile" >+ underscore_name="render_pipe_profile" > mdapi_supported_apis="VK OGL OCL IO" >- oa_format="256B_GENERIC_NOA16" >- hw_config_guid="d66ce633-4ae5-4de2-8678-9705adfd45bc" >+ oa_format="576B_PEC64LL" >+ hw_config_guid="6791470f-f1e6-4f77-9013-bcb0c4be6f57" > > > symbol_name="GpuTime" > underscore_name="gpu_time" > description="Time elapsed on the GPU during the measurement." > data_type="uint64" >- units="ns" >+ units="ns" >+ semantic_type="event" >+ equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV" >+ mdapi_group="GPU" >+ mdapi_usage_flags="Tier1 Overview System Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="GpuCoreClocks" >+ underscore_name="gpu_core_clocks" >+ description="The total number of GPU core clocks elapsed during the measurement." >+ data_type="uint64" >+ units="cycles" >+ semantic_type="event" >+ equation="GPU_CLOCK 0 READ" >+ mdapi_group="GPU" >+ mdapi_usage_flags="Tier1 Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="AvgGpuCoreFrequency" >+ underscore_name="avg_gpu_core_frequency" >+ description="Average GPU Core Frequency in the measurement." >+ data_type="uint64" >+ max_equation="$GpuMaxFrequency" >+ units="hz" >+ semantic_type="event" >+ equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV" >+ mdapi_group="GPU" >+ mdapi_usage_flags="Tier1 Overview System Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="GPU_BUSY" >+ underscore_name="gpu_busy" >+ description="Percentage of time in which GPU is not idle including all GPU engines" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 32 READ 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Front End" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="IA_VERTEX" >+ underscore_name="ia_vertex" >+ description="Number of vertices in a draw" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 33 READ" >+ mdapi_group="Geometry" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="GPGPU_THREADGROUP_COUNT" >+ underscore_name="gpgpu_threadgroup_count" >+ description="Number of GPGPU threadgroups dispatched" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 0 READ" >+ mdapi_group="Thread Dispatcher" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="ASYNC_GPGPU_THREADGROUP_COUNT" >+ underscore_name="async_gpgpu_threadgroup_count" >+ description="Number of Async GPGPU threadgroups dispatched" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 1 READ" >+ mdapi_group="Thread Dispatcher" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="RASTERIZER_SAMPLE_OUTPUT" >+ underscore_name="rasterizer_sample_output" >+ description="Number of lit samples emitted by Rasterizer" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 34 READ" >+ mdapi_group="Rasterizer" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="CLIPPER_INPUT_AVAILABLE" >+ underscore_name="clipper_input_available" >+ description="Percentage of time in which Clipper has input available (from Vertex Shader or SOL)" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 35 READ $SliceTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Geometry" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="CLIPPER_OUTPUT_READY" >+ underscore_name="clipper_output_ready" >+ description="Percentage of time in which Clipper output ready" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 36 READ $SliceTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Geometry" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="CLIPPER_TRANSACTION_OUTPUT" >+ underscore_name="clipper_transaction_output" >+ description="Number of elements pushed by Clipper into Stripsfan stage" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 37 READ" >+ mdapi_group="Geometry" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="STREAMOUT_PRIMITIVE_STORAGE_NEED" >+ underscore_name="streamout_primitive_storage_need" >+ description="Number of objects data that needed beyond the current allocated space, read from pipe 0 only" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 38 READ" >+ mdapi_group="Geometry" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="STRIPSFAN_OUTPUT_READY" >+ underscore_name="stripsfan_output_ready" >+ description="Percentage of time in which in which geometry pipeline output is ready" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 39 READ $SliceTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Geometry" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="VERTEX_FETCH_INPUT_AVAILABLE" >+ underscore_name="vertex_fetch_input_available" >+ description="Percentage of time in which Vertex Fetch input is available" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 40 READ $SliceTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Geometry" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="VERTEX_FETCH_OUTPUT_READY" >+ underscore_name="vertex_fetch_output_ready" >+ description="Percentage of time in which Vertex Fetch output is ready" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 41 READ $SliceTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Geometry" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="VS_OUTPUT_READY" >+ underscore_name="vs_output_ready" >+ description="Percentage of time in which Vertex Shader output is ready" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 42 READ $SliceTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Geometry" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="COLOR_PIPE_CACHE_LATENCY1_STALL" >+ underscore_name="color_pipe_cache_latency1_stall" >+ description="Percentage of time in which Color Pipeline stalled due to MultiSampling Cache latency hiding structure full" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 43 READ $SliceTotalCount 2 UMUL UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Color Pipe" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="COLOR_PIPE_CACHE_LATENCY2_STALL" >+ underscore_name="color_pipe_cache_latency2_stall" >+ description="Percentage of time in which Color Pipeline stalled due to Render Cache latency hiding structure full" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 44 READ 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Color Pipe" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="PIXEL_POST_PROCESS_INPUT_AVAILABLE" >+ underscore_name="pixel_post_process_input_available" >+ description="Percentage of time in which Color Pipeline input ready" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 45 READ $SliceTotalCount 2 UMUL UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Color Pipe" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="PIXEL_POST_PROCESS_OUTPUT_READY" >+ underscore_name="pixel_post_process_output_ready" >+ description="Percentage of time in which Color Pipeline pixel output ready" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 46 READ $SliceTotalCount 2 UMUL UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Color Pipe" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="PS_OUTPUT_AVAILABLE" >+ underscore_name="ps_output_available" >+ description="Percentage of time in which Pixel Shader data is available" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 2 READ $XeCoreTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Color Pipe" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="AMFS_STALL_ALL_INPUT" >+ underscore_name="amfs_stall_all_input" >+ description="Percentage of time in which AMFS stalls at both of the color pipe inputs" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 49 READ $SliceTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="AMFS" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="AMFS_STALL_ANY_INPUT" >+ underscore_name="amfs_stall_any_input" >+ description="Percentage of time in which AMFS stalls at any of the color pipe inputs" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 50 READ $SliceTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="AMFS" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="SAMPLER_ACTIVE" >+ underscore_name="sampler_active" >+ description="Percentage of time in which Sampler pipeline active" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 3 READ $XeCoreTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Sampler" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="SAMPLER_INPUT_AVAILABLE" >+ underscore_name="sampler_input_available" >+ description="Percentage of time in which Sampler input is available" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 4 READ $XeCoreTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Sampler" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="SAMPLER_MEMORY_LATENCY_STALL" >+ underscore_name="sampler_memory_latency_stall" >+ description="Percentage of time in which Sampler stalled due to latency hiding structure full" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 5 READ $XeCoreTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Sampler" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="SAMPLER_OUTPUT_READY" >+ underscore_name="sampler_output_ready" >+ description="Percentage of time in which Sampler output is ready" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 6 READ $XeCoreTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Sampler" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="RASTERIZER_INPUT_AVAILABLE" >+ underscore_name="rasterizer_input_available" >+ description="Percentage of time in which Rasterizer input is available" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 51 READ $SliceTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Rasterizer" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="RASTERIZER_OUTPUT_READY" >+ underscore_name="rasterizer_output_ready" >+ description="Percentage of time in which where rasterizer output is ready" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 52 READ $SliceTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Rasterizer" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="RASTERIZER_TRANSACTION_OUTPUT" >+ underscore_name="rasterizer_transaction_output" >+ description="Number of transactions pushed from Rasterizer to the Z pipe" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 53 READ" >+ mdapi_group="Rasterizer" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="TASK_THREAD_EXIT_COUNT" >+ underscore_name="task_thread_exit_count" >+ description="Number of Task Shader EOT messages received" >+ data_type="uint64" >+ units="messages" >+ semantic_type="event" >+ equation="PEC 21 READ" >+ mdapi_group="Thread Dispatcher" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="TASK_THREADGROUP_COUNT" >+ underscore_name="task_threadgroup_count" >+ description="Number of Task Shader threadgroups dispatched" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 22 READ" >+ mdapi_group="Thread Dispatcher" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="THREAD_DISPATCH_INPUT_AVAILABLE" >+ underscore_name="thread_dispatch_input_available" >+ description="Percentage of time in which Thread Dispatch input is available" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 23 READ $XeCoreTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Thread Dispatcher" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="THREAD_DISPATCH_PS_ACTIVE" >+ underscore_name="thread_dispatch_ps_active" >+ description="Percentage of time in which Pixel Shader threads are ready for dispatch" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 24 READ $XeCoreTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Thread Dispatcher" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="THREAD_DISPATCH_QUEUE0_ACTIVE" >+ underscore_name="thread_dispatch_queue0_active" >+ description="Percentage of time in which non-Pixel Shader threads are ready for dispatch in a particular Xe core" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 25 READ $XeCoreTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Thread Dispatcher" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="THREAD_DISPATCH_QUEUE0_STALL" >+ underscore_name="thread_dispatch_queue0_stall" >+ description="Percentage of time in which queue 0 is stalled waiting for threads to be available" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 26 READ $XeCoreTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Thread Dispatcher" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="THREAD_DISPATCH_QUEUE1_ACTIVE" >+ underscore_name="thread_dispatch_queue1_active" >+ description="Percentage of time in which non-Pixel Shader threads are ready for dispatch in a particular Xe core" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 27 READ $XeCoreTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Thread Dispatcher" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="THREAD_DISPATCH_QUEUE1_STALL" >+ underscore_name="thread_dispatch_queue1_stall" >+ description="Percentage of time in which queue 1 is stalled waiting for threads to be available" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 28 READ $XeCoreTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Thread Dispatcher" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="THREADGROUP_DISPATCH_QUEUE0_RESOURCE_STALL" >+ underscore_name="threadgroup_dispatch_queue0_resource_stall" >+ description="Percentage of time in which Thread Spawner queue 0 is stalled waiting for resources to be available (SLM, Barrier, BTD stack)" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 29 READ $XeCoreTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Thread Dispatcher" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="THREADGROUP_DISPATCH_QUEUE1_RESOURCE_STALL" >+ underscore_name="threadgroup_dispatch_queue1_resource_stall" >+ description="Percentage of time in which Thread Spawner queue 1 is stalled waiting for resources to be available (SLM, Barrier, BTD stack)" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 30 READ $XeCoreTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Thread Dispatcher" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="COMMAND_PARSER_RENDER_ENGINE_DRAW_COUNT" >+ underscore_name="command_parser_render_engine_draw_count" >+ description="Number of DRAW and MESH commands parsed on the 3D engine" >+ data_type="uint64" >+ units="events" > semantic_type="event" >- equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV" >- mdapi_group="GPU" >- mdapi_usage_flags="Tier1 Overview System Frame Batch Draw" >+ equation="PEC 54 READ" >+ mdapi_group="Front End" >+ mdapi_usage_flags="Frame Batch Draw" > mdapi_supported_apis="" > mdapi_hw_unit_type="gpu" > /> >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- - chipset="LNL" >- symbol_name="VectorEngineProfile" >- underscore_name="vector_engine_profile" >- mdapi_supported_apis="VK OGL OCL IO" >- oa_format="256B_GENERIC_NOA16" >- hw_config_guid="e7338926-568c-4424-bc44-f90318a6b472" >- > >- - symbol_name="GpuTime" >- underscore_name="gpu_time" >- description="Time elapsed on the GPU during the measurement." >+ + symbol_name="AMFS_4X4_SHADING_REQUEST" >+ underscore_name="amfs_4_x4_shading_request" >+ description="Number of Shading Request (evaluate) messages processed by AMFS" > data_type="uint64" >- units="ns" >+ units="messages" > semantic_type="event" >- equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV" >- mdapi_group="GPU" >- mdapi_usage_flags="Tier1 Overview System Frame Batch Draw" >+ equation="PEC 55 READ" >+ mdapi_group="AMFS" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="AMFS_CACHE_HIT" >+ underscore_name="amfs_cache_hit" >+ description="Number of hits in the AMFS cache" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 56 READ" >+ mdapi_group="AMFS" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="AMFS_CACHE_MISS" >+ underscore_name="amfs_cache_miss" >+ description="Number of cache misses in AMFS" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 57 READ" >+ mdapi_group="AMFS" >+ mdapi_usage_flags="Frame Batch Draw" > mdapi_supported_apis="" > mdapi_hw_unit_type="gpu" > /> >@@ -1451,66 +4602,36 @@ > > > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >- > > > >@@ -1529,144 +4650,154 @@ > > > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >- >- >- >- >- >- >- >+ >+ >+ >+ >+ >+ >+ >+ >+ >+ >+ >+ >+ >+ >+ >+ >+ >+ > > > >- + chipset="LNL" >- symbol_name="VectorEngineStalls" >- underscore_name="vector_engine_stalls" >+ symbol_name="RTProfile" >+ underscore_name="rt_profile" > mdapi_supported_apis="VK OGL OCL IO" >- oa_format="256B_GENERIC_NOA16" >- hw_config_guid="4ccd6535-fb9a-440f-b0f5-882879dc4cb0" >+ oa_format="576B_PEC64LL" >+ hw_config_guid="d66ce633-4ae5-4de2-8678-9705adfd45bc" > > > symbol_name="GpuTime" >@@ -1681,591 +4812,1430 @@ > mdapi_supported_apis="" > mdapi_hw_unit_type="gpu" > /> >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- - chipset="LNL" >- symbol_name="TestOa" >- underscore_name="test_oa" >- mdapi_supported_apis="VK OGL OCL IO" >- oa_format="256B_GENERIC_NOA16" >- hw_config_guid="24c33f2c-f8d9-40d9-b2db-64e863b4d625" >- > >- - symbol_name="GpuTime" >- underscore_name="gpu_time" >- description="Time elapsed on the GPU during the measurement." >+ + symbol_name="GpuCoreClocks" >+ underscore_name="gpu_core_clocks" >+ description="The total number of GPU core clocks elapsed during the measurement." >+ data_type="uint64" >+ units="cycles" >+ semantic_type="event" >+ equation="GPU_CLOCK 0 READ" >+ mdapi_group="GPU" >+ mdapi_usage_flags="Tier1 Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="AvgGpuCoreFrequency" >+ underscore_name="avg_gpu_core_frequency" >+ description="Average GPU Core Frequency in the measurement." >+ data_type="uint64" >+ max_equation="$GpuMaxFrequency" >+ units="hz" >+ semantic_type="event" >+ equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV" >+ mdapi_group="GPU" >+ mdapi_usage_flags="Tier1 Overview System Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="GPU_BUSY" >+ underscore_name="gpu_busy" >+ description="Percentage of time in which GPU is not idle including all GPU engines" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 32 READ 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Front End" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="IA_VERTEX" >+ underscore_name="ia_vertex" >+ description="Number of vertices in a draw" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 33 READ" >+ mdapi_group="Geometry" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="GPGPU_THREADGROUP_COUNT" >+ underscore_name="gpgpu_threadgroup_count" >+ description="Number of GPGPU threadgroups dispatched" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 0 READ" >+ mdapi_group="Thread Dispatcher" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="ASYNC_GPGPU_THREADGROUP_COUNT" >+ underscore_name="async_gpgpu_threadgroup_count" >+ description="Number of Async GPGPU threadgroups dispatched" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 1 READ" >+ mdapi_group="Thread Dispatcher" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="RASTERIZER_SAMPLE_OUTPUT" >+ underscore_name="rasterizer_sample_output" >+ description="Number of lit samples emitted by Rasterizer" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 34 READ" >+ mdapi_group="Rasterizer" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="RT_ANY_HIT_THREAD_RAY_DISPATCH" >+ underscore_name="rt_any_hit_thread_ray_dispatch" >+ description="Number of any hit threads dispatched for RayQuery only" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 2 READ" >+ mdapi_group="Ray Tracing" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="RT_CALLABLE_THREAD_RAY_DISPATCH" >+ underscore_name="rt_callable_thread_ray_dispatch" >+ description="Number of callable threads dispatched for RayQuery only" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 3 READ" >+ mdapi_group="Ray Tracing" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="RT_CLOSEST_HIT_THREAD_RAY_DISPATCH" >+ underscore_name="rt_closest_hit_thread_ray_dispatch" >+ description="Number of closest hit threads dispatched for RayQuery only" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 4 READ" >+ mdapi_group="Ray Tracing" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="RT_INPUT_MESSAGE_SPAWN_COUNT" >+ underscore_name="rt_input_message_spawn_count" >+ description="Number of valid SIMD lanes in the Spawn message" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 5 READ" >+ mdapi_group="Ray Tracing" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="RT_INSTANCE_NODE_RAY_COUNT" >+ underscore_name="rt_instance_node_ray_count" >+ description="Number of instance BVH nodes processed by the traversal function" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 6 READ" >+ mdapi_group="Ray Tracing" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="RT_INTERNAL_NODE_RAY_COUNT" >+ underscore_name="rt_internal_node_ray_count" >+ description="Number of internal BVH nodes processed by the traversal function" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 21 READ" >+ mdapi_group="Ray Tracing" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="RT_INTERSECTION_THREAD_RAY_DISPATCH" >+ underscore_name="rt_intersection_thread_ray_dispatch" >+ description="Number of intersection threads dispatched for RayQuery only" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 22 READ" >+ mdapi_group="Ray Tracing" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="RT_MESSAGE_STALL" >+ underscore_name="rt_message_stall" >+ description="Percentage of time in which Ray Tracing message input is stalled while accepting input" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 23 READ $XeCoreTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Ray Tracing" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="RT_MISS_THREAD_RAY_DISPATCH" >+ underscore_name="rt_miss_thread_ray_dispatch" >+ description="Number of miss threads dispatched for RayQuery only" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 24 READ" >+ mdapi_group="Ray Tracing" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="RT_PROCEDURAL_NODE_RAY_COUNT" >+ underscore_name="rt_procedural_node_ray_count" >+ description="Number of procedural BVH nodes processed by the traversal function" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 25 READ" >+ mdapi_group="Ray Tracing" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="RT_QUAD_LEAF_RAY_COUNT" >+ underscore_name="rt_quad_leaf_ray_count" >+ description="Number of triangle BVH nodes processed by the traversal function" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 26 READ" >+ mdapi_group="Ray Tracing" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="RT_QUAD_TEST_RAY_COUNT" >+ underscore_name="rt_quad_test_ray_count" >+ description="Number of nodes processed that use ray-quad intersection pipeline" > data_type="uint64" >- units="ns" >+ units="events" > semantic_type="event" >- equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV" >- mdapi_group="GPU" >- mdapi_usage_flags="Tier1 Overview System Frame Batch Draw" >+ equation="PEC 27 READ" >+ mdapi_group="Ray Tracing" >+ mdapi_usage_flags="Frame Batch Draw" > mdapi_supported_apis="" > mdapi_hw_unit_type="gpu" > /> >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- - chipset="LNL" >- symbol_name="MetricSet1" >- underscore_name="metric_set1" >- mdapi_supported_apis="VK OGL OCL IO" >- oa_format="256B_GENERIC_NOA16" >- hw_config_guid="0af24f48-05bc-467b-b561-33dec252e651" >- > >- - symbol_name="GpuTime" >- underscore_name="gpu_time" >- description="Time elapsed on the GPU during the measurement." >+ + symbol_name="RT_REQUEST_COLLISION" >+ underscore_name="rt_request_collision" >+ description="Percentage of time in which Ray Tracing Leaf has two Traversal inputs valid" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 28 READ $XeCoreTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Ray Tracing" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="RT_TRANSFORM_RAY_COUNT" >+ underscore_name="rt_transform_ray_count" >+ description="Number of only HW transform accesses" > data_type="uint64" >- units="ns" >+ units="events" > semantic_type="event" >- equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV" >- mdapi_group="GPU" >- mdapi_usage_flags="Tier1 Overview System Frame Batch Draw" >+ equation="PEC 29 READ" >+ mdapi_group="Ray Tracing" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="RT_TRAVERSAL_INPUT_RAY_COUNT" >+ underscore_name="rt_traversal_input_ray_count" >+ description="Number of Ray Tracing Traversal input rays" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 30 READ" >+ mdapi_group="Ray Tracing" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="RT_TRAVERSAL_OUTPUT_RAY_COUNT" >+ underscore_name="rt_traversal_output_ray_count" >+ description="Number of Ray Tracing Traversal output rays" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 31 READ" >+ mdapi_group="Ray Tracing" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="RT_TRAVERSAL_STALL" >+ underscore_name="rt_traversal_stall" >+ description="Percentage of time in which Ray Tracing Frontend is stalled by Traversal" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 7 READ $XeCoreTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Ray Tracing" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="RT_TRAVERSAL_STEP_RAY_COUNT" >+ underscore_name="rt_traversal_step_ray_count" >+ description="Number of BVH nodes processed" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 8 READ" >+ mdapi_group="Ray Tracing" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_INST_EXECUTED_ALU0_RT_AHS" >+ underscore_name="xve_inst_executed_alu0_rt_ahs" >+ description="Number of execution slots taken by instructions executed by RT_AHS threads on ALU0 pipe" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 9 READ" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_INST_EXECUTED_ALU0_RT_CHS" >+ underscore_name="xve_inst_executed_alu0_rt_chs" >+ description="Number of execution slots taken by instructions executed by RT_CHS threads on ALU0 pipe" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 10 READ" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_INST_EXECUTED_ALU0_RT_MS" >+ underscore_name="xve_inst_executed_alu0_rt_ms" >+ description="Number of execution slots taken by instructions executed by RT_MS threads on ALU0 pipe" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 11 READ" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_INST_EXECUTED_ALU1_RT_AHS" >+ underscore_name="xve_inst_executed_alu1_rt_ahs" >+ description="Number of execution slots taken by instructions executed by RT_AHS threads on ALU1 pipe" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 12 READ" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_INST_EXECUTED_ALU1_RT_CHS" >+ underscore_name="xve_inst_executed_alu1_rt_chs" >+ description="Number of execution slots taken by instructions executed by RT_CHS threads on ALU1 pipe" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 13 READ" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_INST_EXECUTED_ALU1_RT_MS" >+ underscore_name="xve_inst_executed_alu1_rt_ms" >+ description="Number of execution slots taken by instructions executed by RT_MS threads on ALU1 pipe" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 14 READ" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_THREADS_OCCUPANCY_ALL" >+ underscore_name="xve_threads_occupancy_all" >+ description="Percentage of thread slots occupied by ALL threads" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 15 READ 8 UMUL $VectorEngineThreadsCount $VectorEngineTotalCount UMUL UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="RT_BVH_CACHE_MISS" >+ underscore_name="rt_bvh_cache_miss" >+ description="Number of BVH cache misses for RayQuery only" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 16 READ" >+ mdapi_group="Ray Tracing" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="RT_INPUT_MESSAGE_RAY_COUNT" >+ underscore_name="rt_input_message_ray_count" >+ description="Number of valid SIMD lanes in the TraceRay message" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 17 READ" >+ mdapi_group="Ray Tracing" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="RT_LOAD_STORE_CACHE_READ_MESSAGE_COUNT" >+ underscore_name="rt_load_store_cache_read_message_count" >+ description="Number of read messages sent from Ray Tracing unit to the Load Store Cache" >+ data_type="uint64" >+ units="messages" >+ semantic_type="event" >+ equation="PEC 18 READ" >+ mdapi_group="L1 Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="RT_LOAD_STORE_CACHE_WRITE_FROM_MESSAGE_COUNT" >+ underscore_name="rt_load_store_cache_write_from_message_count" >+ description="Number of write messages sent from Ray Tracing unit to the Load Store Cache" >+ data_type="uint64" >+ units="messages" >+ semantic_type="event" >+ equation="PEC 19 READ" >+ mdapi_group="L1 Cache" >+ mdapi_usage_flags="Frame Batch Draw" > mdapi_supported_apis="" > mdapi_hw_unit_type="gpu" > /> > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > > > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >- >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >+ >+ >+ >+ >+ >+ >+ >+ >+ >+ >+ >+ > > > >- + chipset="LNL" >- symbol_name="MetricSet2" >- underscore_name="metric_set2" >+ symbol_name="VectorEngineProfile" >+ underscore_name="vector_engine_profile" > mdapi_supported_apis="VK OGL OCL IO" >- oa_format="256B_GENERIC_NOA16" >- hw_config_guid="914ad901-832a-4ccd-a21e-ddc7690ed213" >+ oa_format="576B_PEC64LL" >+ hw_config_guid="e7338926-568c-4424-bc44-f90318a6b472" > > > symbol_name="GpuTime" > underscore_name="gpu_time" > description="Time elapsed on the GPU during the measurement." > data_type="uint64" >- units="ns" >+ units="ns" >+ semantic_type="event" >+ equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV" >+ mdapi_group="GPU" >+ mdapi_usage_flags="Tier1 Overview System Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="GpuCoreClocks" >+ underscore_name="gpu_core_clocks" >+ description="The total number of GPU core clocks elapsed during the measurement." >+ data_type="uint64" >+ units="cycles" >+ semantic_type="event" >+ equation="GPU_CLOCK 0 READ" >+ mdapi_group="GPU" >+ mdapi_usage_flags="Tier1 Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="AvgGpuCoreFrequency" >+ underscore_name="avg_gpu_core_frequency" >+ description="Average GPU Core Frequency in the measurement." >+ data_type="uint64" >+ max_equation="$GpuMaxFrequency" >+ units="hz" >+ semantic_type="event" >+ equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV" >+ mdapi_group="GPU" >+ mdapi_usage_flags="Tier1 Overview System Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="GPU_BUSY" >+ underscore_name="gpu_busy" >+ description="Percentage of time in which GPU is not idle including all GPU engines" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 32 READ 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Front End" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="IA_VERTEX" >+ underscore_name="ia_vertex" >+ description="Number of vertices in a draw" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 33 READ" >+ mdapi_group="Geometry" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="GPGPU_THREADGROUP_COUNT" >+ underscore_name="gpgpu_threadgroup_count" >+ description="Number of GPGPU threadgroups dispatched" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 0 READ" >+ mdapi_group="Thread Dispatcher" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="ASYNC_GPGPU_THREADGROUP_COUNT" >+ underscore_name="async_gpgpu_threadgroup_count" >+ description="Number of Async GPGPU threadgroups dispatched" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 1 READ" >+ mdapi_group="Thread Dispatcher" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="RASTERIZER_SAMPLE_OUTPUT" >+ underscore_name="rasterizer_sample_output" >+ description="Number of lit samples emitted by Rasterizer" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 34 READ" >+ mdapi_group="Rasterizer" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="ICACHE_HIT" >+ underscore_name="icache_hit" >+ description="Number of Instruction Cache hits" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 2 READ" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="ICACHE_MISS" >+ underscore_name="icache_miss" >+ description="Number of Instruction Cache misses" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 3 READ" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_ACTIVE" >+ underscore_name="xve_active" >+ description="Percentage of time in which at least one pipe is active in XVE" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 4 READ $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_INST_EXECUTED_ALU0_ALL" >+ underscore_name="xve_inst_executed_alu0_all" >+ description="Number of execution slots taken by instructions executed by ALL threads on ALU0 pipe" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 5 READ" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_INST_EXECUTED_ALU1_ALL" >+ underscore_name="xve_inst_executed_alu1_all" >+ description="Number of execution slots taken by instructions executed by ALL threads on ALU1 pipe" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 6 READ" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_INST_EXECUTED_ALU2_ALL" >+ underscore_name="xve_inst_executed_alu2_all" >+ description="Number of execution slots taken by instructions executed in ALU2 pipe" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 21 READ" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_INST_EXECUTED_BARRIER" >+ underscore_name="xve_inst_executed_barrier" >+ description="Number of sync.bar and sync.host instructions executed" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 22 READ" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_INST_EXECUTED_BITCONV" >+ underscore_name="xve_inst_executed_bitconv" >+ description="Number of execution slots taken by bit manipulation instructions" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 23 READ" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_INST_EXECUTED_CONTROL_ALL" >+ underscore_name="xve_inst_executed_control_all" >+ description="Number of instructions executed on JEU Pipe" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 24 READ" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_INST_EXECUTED_FP16" >+ underscore_name="xve_inst_executed_fp16" >+ description="Number of execution slots taken by FP16 ALU instructions" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 25 READ" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_INST_EXECUTED_FP32" >+ underscore_name="xve_inst_executed_fp32" >+ description="Number of execution slots taken by FP32 ALU instructions" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 26 READ" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_INST_EXECUTED_FP64" >+ underscore_name="xve_inst_executed_fp64" >+ description="Number of execution slots taken by FP64 ALU instructions" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 27 READ" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_INST_EXECUTED_FP64_2ND" >+ underscore_name="xve_inst_executed_fp64_2_nd" >+ description="Number of execution slots taken by FP64 ALU instructions executed in the second pipe" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 28 READ" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_INST_EXECUTED_INT16" >+ underscore_name="xve_inst_executed_int16" >+ description="Number of execution slots taken by INT16 ALU instructions" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 29 READ" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_INST_EXECUTED_INT32" >+ underscore_name="xve_inst_executed_int32" >+ description="Number of execution slots taken by INT32 ALU instructions" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 30 READ" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_INST_EXECUTED_INT64" >+ underscore_name="xve_inst_executed_int64" >+ description="Number of execution slots taken by INT64 instructions" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 31 READ" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_INST_EXECUTED_MATH" >+ underscore_name="xve_inst_executed_math" >+ description="Number of execution slots taken by extended math instructions" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 7 READ" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_INST_EXECUTED_NONDIVERGENT" >+ underscore_name="xve_inst_executed_nondivergent" >+ description="Number of execution slots taken by non-divergent instructions" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 8 READ" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_INST_EXECUTED_PREDICATION" >+ underscore_name="xve_inst_executed_predication" >+ description="Number of execution slots taken by instructions with predication mask enabled" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 9 READ" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_INST_EXECUTED_SEND_ALL" >+ underscore_name="xve_inst_executed_send_all" >+ description="Number of instruction (GRF) dispatches executed by ALL threads on SEND Pipe" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 10 READ" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_INST_EXECUTED_XMX_BF16" >+ underscore_name="xve_inst_executed_xmx_bf16" >+ description="Number of execution slots taken by BF16 XMX instructions" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 11 READ" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_INST_EXECUTED_XMX_FP16" >+ underscore_name="xve_inst_executed_xmx_fp16" >+ description="Number of execution slots taken by FP16 XMX instructions" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 12 READ" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_INST_EXECUTED_XMX_INT2" >+ underscore_name="xve_inst_executed_xmx_int2" >+ description="Number of execution slots taken by INT2 XMX instructions" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 13 READ" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_INST_EXECUTED_XMX_INT4" >+ underscore_name="xve_inst_executed_xmx_int4" >+ description="Number of execution slots taken by INT4 XMX instructions" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 14 READ" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_INST_EXECUTED_XMX_INT8" >+ underscore_name="xve_inst_executed_xmx_int8" >+ description="Number of execution slots taken by INT8 XMX instructions" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 15 READ" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_INST_ISSUED_ALL" >+ underscore_name="xve_inst_issued_all" >+ description="Number of instructions issued (decoded) to any pipe" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 16 READ" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_MULTIPLE_PIPE_ACTIVE" >+ underscore_name="xve_multiple_pipe_active" >+ description="Percentage of time in which at least two pipes are actively executing a Gen ISA instruction among ALU0, ALU1 and ALU2 pipes" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 17 READ $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_PIPE_ALU0_AND_ALU1_ACTIVE" >+ underscore_name="xve_pipe_alu0_and_alu1_active" >+ description="Percentage of time in which ALU0 and ALU1 pipes are both actively executing a Gen ISA instruction" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 18 READ $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_PIPE_ALU0_AND_ALU2_ACTIVE" >+ underscore_name="xve_pipe_alu0_and_alu2_active" >+ description="Percentage of time in which ALU0 and ALU2 pipes are both actively executing a Gen ISA instruction" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 19 READ $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_THREADS_OCCUPANCY_ALL" >+ underscore_name="xve_threads_occupancy_all" >+ description="Percentage of thread slots occupied by ALL threads" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 20 READ 8 UMUL $VectorEngineThreadsCount $VectorEngineTotalCount UMUL UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="L3_ATOMIC_ACCESS" >+ underscore_name="l3_atomic_access" >+ description="Number of atomic accesses to Device Cache bank" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 35 READ" >+ mdapi_group="Device Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="L3_HIT" >+ underscore_name="l3_hit" >+ description="Number of Device Cache accesses (except atomics) which hits in the Device Cache cache" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 37 READ" >+ mdapi_group="Device Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="L3_MISS" >+ underscore_name="l3_miss" >+ description="Number of Device Cache accesses (except atomics) which miss in the Device Cache cache" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 39 READ" >+ mdapi_group="Device Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="L3_READ" >+ underscore_name="l3_read" >+ description="Number of Device Cache 64B read requests" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 41 READ" >+ mdapi_group="Device Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="L3_WRITE" >+ underscore_name="l3_write" >+ description="Number of Device Cache 64B write requests" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 44 READ" >+ mdapi_group="Device Cache" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="COMMAND_PARSER_COMPUTE_ENGINE_BUSY" >+ underscore_name="command_parser_compute_engine_busy" >+ description="Percentage of time in which there is a context loaded and active on the compute queue" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 45 READ $ComputeEngineTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Front End" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="COMMAND_PARSER_COMPUTE_ENGINE_DISPATCH_KERNEL_COUNT" >+ underscore_name="command_parser_compute_engine_dispatch_kernel_count" >+ description="Number of compute walker commands parsed on the compute engine" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 46 READ" >+ mdapi_group="Front End" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="COMMAND_PARSER_RENDER_ENGINE_BUSY" >+ underscore_name="command_parser_render_engine_busy" >+ description="Percentage of time in which there is a context loaded and active on the 3D queue" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 47 READ 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Front End" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="COMMAND_PARSER_RENDER_ENGINE_DISPATCH_KERNEL_COUNT" >+ underscore_name="command_parser_render_engine_dispatch_kernel_count" >+ description="Number of compute walker commands parsed on the 3D engine" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 48 READ" >+ mdapi_group="Front End" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="COMPRESSOR_INPUT" >+ underscore_name="compressor_input" >+ description="Number of 256B writes to compressible surfaces at the input of the compressor" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 49 READ" >+ mdapi_group="Memory" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="COMPRESSOR_OUTPUT" >+ underscore_name="compressor_output" >+ description="Number of 256B writes to compressible surfaces at the output of the compressor" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 50 READ" >+ mdapi_group="Memory" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="GPU_MEMORY_BYTE_READ" >+ underscore_name="gpu_memory_byte_read" >+ description="Number of device local memory (HBM, GDDR, LPDDR, etc.) read bytes" >+ data_type="uint64" >+ units="bytes" >+ semantic_type="event" >+ equation="PEC 51 READ 2 UMUL" >+ mdapi_group="Memory" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="GPU_MEMORY_BYTE_WRITE" >+ underscore_name="gpu_memory_byte_write" >+ description="Number of device local memory (HBM, GDDR, LPDDR, etc.) write bytes" >+ data_type="uint64" >+ units="bytes" > semantic_type="event" >- equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV" >- mdapi_group="GPU" >- mdapi_usage_flags="Tier1 Overview System Frame Batch Draw" >+ equation="PEC 52 READ 2 UMUL" >+ mdapi_group="Memory" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="GPU_MEMORY_BYTE_READ_RATE" >+ underscore_name="gpu_memory_byte_read_rate" >+ description="Device local memory (HBM, GDDR, LPDDR, etc.) read bandwidth" >+ data_type="float" >+ units="gbps" >+ semantic_type="throughput" >+ equation="PEC 51 READ 2 UMUL $GpuTime FDIV" >+ mdapi_group="Memory" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="GPU_MEMORY_BYTE_WRITE_RATE" >+ underscore_name="gpu_memory_byte_write_rate" >+ description="Device local memory (HBM, GDDR, LPDDR, etc.) write bandwidth" >+ data_type="float" >+ units="gbps" >+ semantic_type="throughput" >+ equation="PEC 52 READ 2 UMUL $GpuTime FDIV" >+ mdapi_group="Memory" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="TLB_MISS" >+ underscore_name="tlb_miss" >+ description="Number of misses from all TLBs and all engines" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 53 READ" >+ mdapi_group="Memory" >+ mdapi_usage_flags="Frame Batch Draw" > mdapi_supported_apis="" > mdapi_hw_unit_type="gpu" > /> > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > > > >@@ -2273,41 +6243,41 @@ > > > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > > > >@@ -2315,70 +6285,24 @@ > > > >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >+ >+ >+ >+ >+ >+ >+ >+ > > > >- + chipset="LNL" >- symbol_name="MetricSet3" >- underscore_name="metric_set3" >+ symbol_name="VectorEngineStalls" >+ underscore_name="vector_engine_stalls" > mdapi_supported_apis="VK OGL OCL IO" >- oa_format="256B_GENERIC_NOA16" >- hw_config_guid="3b3c9f55-e34b-423d-8c05-7b1af03b6685" >+ oa_format="576B_PEC64LL" >+ hw_config_guid="4ccd6535-fb9a-440f-b0f5-882879dc4cb0" > > > symbol_name="GpuTime" >@@ -2393,273 +6317,465 @@ > mdapi_supported_apis="" > mdapi_hw_unit_type="gpu" > /> >+ + symbol_name="GpuCoreClocks" >+ underscore_name="gpu_core_clocks" >+ description="The total number of GPU core clocks elapsed during the measurement." >+ data_type="uint64" >+ units="cycles" >+ semantic_type="event" >+ equation="GPU_CLOCK 0 READ" >+ mdapi_group="GPU" >+ mdapi_usage_flags="Tier1 Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="AvgGpuCoreFrequency" >+ underscore_name="avg_gpu_core_frequency" >+ description="Average GPU Core Frequency in the measurement." >+ data_type="uint64" >+ max_equation="$GpuMaxFrequency" >+ units="hz" >+ semantic_type="event" >+ equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV" >+ mdapi_group="GPU" >+ mdapi_usage_flags="Tier1 Overview System Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="GPU_BUSY" >+ underscore_name="gpu_busy" >+ description="Percentage of time in which GPU is not idle including all GPU engines" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 32 READ 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Front End" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="IA_VERTEX" >+ underscore_name="ia_vertex" >+ description="Number of vertices in a draw" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 33 READ" >+ mdapi_group="Geometry" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="GPGPU_THREADGROUP_COUNT" >+ underscore_name="gpgpu_threadgroup_count" >+ description="Number of GPGPU threadgroups dispatched" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 0 READ" >+ mdapi_group="Thread Dispatcher" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="ASYNC_GPGPU_THREADGROUP_COUNT" >+ underscore_name="async_gpgpu_threadgroup_count" >+ description="Number of Async GPGPU threadgroups dispatched" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 1 READ" >+ mdapi_group="Thread Dispatcher" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="RASTERIZER_SAMPLE_OUTPUT" >+ underscore_name="rasterizer_sample_output" >+ description="Number of lit samples emitted by Rasterizer" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 34 READ" >+ mdapi_group="Rasterizer" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_STALL_ALUWR" >+ underscore_name="xve_stall_aluwr" >+ description="Percentage of time in which XVE stalled, with at least one thread waiting for ALU to write GRF/ACC register. Multiple stall reasons can qualify during the same cycle" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 2 READ $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_STALL_BARRIER" >+ underscore_name="xve_stall_barrier" >+ description="Percentage of time in which XVE stalled, with at least one thread waiting for Gateway to write Notify register. Multiple stall reasons can qualify during the same cycle" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 3 READ $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_STALL_CONTROL" >+ underscore_name="xve_stall_control" >+ description="Percentage of time in which XVE stalled, with at least one thread waiting for JEU to complete branch instruction. Multiple stall reasons can qualify during the same cycle" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 4 READ $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_STALL_INSTFETCH" >+ underscore_name="xve_stall_instfetch" >+ description="Percentage of time in which XVE stalled, with at least one thread waiting for Instruction Fetch. Multiple stall reasons can qualify during the same cycle" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 5 READ $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_STALL_OTHER" >+ underscore_name="xve_stall_other" >+ description="Percentage of time in which XVE stalled, with at least one thread waiting on any other dependency (Flag/EoT etc). Multiple stall reasons can qualify during the same cycle" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 6 READ $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_STALL_PIPESTALL" >+ underscore_name="xve_stall_pipestall" >+ description="Percentage of time in which XVE stalled, with at least one thread ready to be scheduled (Grf conf/send holds etc). Multiple stall reasons can qualify during the same cycle" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 21 READ $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_STALL_PS_DEPENDENCY" >+ underscore_name="xve_stall_ps_dependency" >+ description="Percentage of time in which XVE stalled, with at least one thread waiting on Pixel Shader dependency" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 22 READ $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_STALL_SBID" >+ underscore_name="xve_stall_sbid" >+ description="Percentage of time in which XVE stalled, with at least one thread waiting for Scoreboard token to be available. Multiple stall reasons can qualify during the same cycle" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 23 READ $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_STALL_SENDWR" >+ underscore_name="xve_stall_sendwr" >+ description="Percentage of time in which XVE stalled, with at least one thread waiting for SEND message to be dispatched from XVE. Multiple stall reasons can qualify during the same cycle" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 24 READ $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="THREAD_DISPATCH_QUEUE0_ACTIVE" >+ underscore_name="thread_dispatch_queue0_active" >+ description="Percentage of time in which non-Pixel Shader threads are ready for dispatch in a particular Xe core" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 25 READ $XeCoreTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Thread Dispatcher" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="THREAD_DISPATCH_QUEUE0_STALL" >+ underscore_name="thread_dispatch_queue0_stall" >+ description="Percentage of time in which queue 0 is stalled waiting for threads to be available" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 26 READ $XeCoreTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Thread Dispatcher" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="THREAD_DISPATCH_QUEUE1_ACTIVE" >+ underscore_name="thread_dispatch_queue1_active" >+ description="Percentage of time in which non-Pixel Shader threads are ready for dispatch in a particular Xe core" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 27 READ $XeCoreTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Thread Dispatcher" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="THREAD_DISPATCH_QUEUE1_STALL" >+ underscore_name="thread_dispatch_queue1_stall" >+ description="Percentage of time in which queue 1 is stalled waiting for threads to be available" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 28 READ $XeCoreTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Thread Dispatcher" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="THREADGROUP_DISPATCH_QUEUE0_RESOURCE_STALL" >+ underscore_name="threadgroup_dispatch_queue0_resource_stall" >+ description="Percentage of time in which Thread Spawner queue 0 is stalled waiting for resources to be available (SLM, Barrier, BTD stack)" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 29 READ $XeCoreTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Thread Dispatcher" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="THREADGROUP_DISPATCH_QUEUE1_RESOURCE_STALL" >+ underscore_name="threadgroup_dispatch_queue1_resource_stall" >+ description="Percentage of time in which Thread Spawner queue 1 is stalled waiting for resources to be available (SLM, Barrier, BTD stack)" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 30 READ $XeCoreTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Thread Dispatcher" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_ACTIVE" >+ underscore_name="xve_active" >+ description="Percentage of time in which at least one pipe is active in XVE" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 31 READ $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_SHARED_FUNCTION_ACCESS_HOLD" >+ underscore_name="xve_shared_function_access_hold" >+ description="Percentage of time in which XVE requests stalled by Shared Function units" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 7 READ $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_STALL" >+ underscore_name="xve_stall" >+ description="Percentage of time in which any thread loaded but not even a single pipe is active in XVE" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 8 READ $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_THREADS_OCCUPANCY_ALL" >+ underscore_name="xve_threads_occupancy_all" >+ description="Percentage of thread slots occupied by ALL threads" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 9 READ 8 UMUL $VectorEngineThreadsCount $VectorEngineTotalCount UMUL UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_MULTIPLE_PIPE_ACTIVE" >+ underscore_name="xve_multiple_pipe_active" >+ description="Percentage of time in which at least two pipes are actively executing a Gen ISA instruction among ALU0, ALU1 and ALU2 pipes" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 10 READ $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_PIPE_ALU0_AND_ALU1_ACTIVE" >+ underscore_name="xve_pipe_alu0_and_alu1_active" >+ description="Percentage of time in which ALU0 and ALU1 pipes are both actively executing a Gen ISA instruction" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 11 READ $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="XVE_PIPE_ALU0_AND_ALU2_ACTIVE" >+ underscore_name="xve_pipe_alu0_and_alu2_active" >+ description="Percentage of time in which ALU0 and ALU2 pipes are both actively executing a Gen ISA instruction" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 12 READ $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Vector Engine" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >+ >+ >+ >+ > > > >- + chipset="LNL" >- symbol_name="MetricSet4" >- underscore_name="metric_set4" >+ symbol_name="TestOa" >+ underscore_name="test_oa" > mdapi_supported_apis="VK OGL OCL IO" >- oa_format="256B_GENERIC_NOA16" >- hw_config_guid="e6f21065-a6f5-4ea4-a545-80d6aebaab56" >+ oa_format="576B_PEC64LL" >+ hw_config_guid="24c33f2c-f8d9-40d9-b2db-64e863b4d625" > > > symbol_name="GpuTime" >@@ -2674,125 +6790,194 @@ > mdapi_supported_apis="" > mdapi_hw_unit_type="gpu" > /> >+ + symbol_name="GpuCoreClocks" >+ underscore_name="gpu_core_clocks" >+ description="The total number of GPU core clocks elapsed during the measurement." >+ data_type="uint64" >+ units="cycles" >+ semantic_type="event" >+ equation="GPU_CLOCK 0 READ" >+ mdapi_group="GPU" >+ mdapi_usage_flags="Tier1 Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="AvgGpuCoreFrequency" >+ underscore_name="avg_gpu_core_frequency" >+ description="Average GPU Core Frequency in the measurement." >+ data_type="uint64" >+ max_equation="$GpuMaxFrequency" >+ units="hz" >+ semantic_type="event" >+ equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV" >+ mdapi_group="GPU" >+ mdapi_usage_flags="Tier1 Overview System Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="GPU_BUSY" >+ underscore_name="gpu_busy" >+ description="Percentage of time in which GPU is not idle including all GPU engines" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 32 READ 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Front End" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="IA_VERTEX" >+ underscore_name="ia_vertex" >+ description="Number of vertices in a draw" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 33 READ" >+ mdapi_group="Geometry" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="GPGPU_THREADGROUP_COUNT" >+ underscore_name="gpgpu_threadgroup_count" >+ description="Number of GPGPU threadgroups dispatched" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 0 READ" >+ mdapi_group="Thread Dispatcher" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="ASYNC_GPGPU_THREADGROUP_COUNT" >+ underscore_name="async_gpgpu_threadgroup_count" >+ description="Number of Async GPGPU threadgroups dispatched" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 1 READ" >+ mdapi_group="Thread Dispatcher" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="RASTERIZER_SAMPLE_OUTPUT" >+ underscore_name="rasterizer_sample_output" >+ description="Number of lit samples emitted by Rasterizer" >+ data_type="uint64" >+ units="events" >+ semantic_type="event" >+ equation="PEC 34 READ" >+ mdapi_group="Rasterizer" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="TEST_EVENT1" >+ underscore_name="test_event1" >+ description="Number of GPU (gt) clock cycles at every clock period" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 2 READ $XeCoreTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Test" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="TEST_EVENT1_CYCLES" >+ underscore_name="test_event1_cycles" >+ description="Number of GPU (gt) clock cycles at every clock period" >+ data_type="uint64" >+ units="cycles" >+ semantic_type="event" >+ equation="PEC 2 READ" >+ mdapi_group="Test" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="TEST_EVENT1_CYCLES_AVERAGE" >+ underscore_name="test_event1_cycles_average" >+ description="Average number of GPU (gt) clock cycles at every clock period" >+ data_type="uint64" >+ units="cycles" >+ semantic_type="event" >+ equation="PEC 2 READ $XeCoreTotalCount UDIV" >+ mdapi_group="Test" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="TEST_EVENT2" >+ underscore_name="test_event2" >+ description="Number of GPU (gt) clock cycles at every other clock period" >+ data_type="float" >+ units="percent" >+ semantic_type="duration" >+ equation="PEC 35 READ $SliceTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV" >+ mdapi_group="Test" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> >+ + symbol_name="TEST_EVENT2_CYCLES" >+ underscore_name="test_event2_cycles" >+ description="Number of GPU (gt) clock cycles at every other clock period" >+ data_type="uint64" >+ units="cycles" >+ semantic_type="event" >+ equation="PEC 35 READ" >+ mdapi_group="Test" >+ mdapi_usage_flags="Frame Batch Draw" >+ mdapi_supported_apis="" >+ mdapi_hw_unit_type="gpu" >+ /> > >- >+ > >- >+ > >- >+ > >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >+ > >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- >- > > > >-- >2.41.0 >