From mboxrd@z Thu Jan 1 00:00:00 1970 Authentication-Results: smtp.subspace.kernel.org; arc=pass header.oldest-pass=0 smtp.remote-ip=40.93.195.31 ARC-Seal: i=2; d=subspace.kernel.org; s=arc20261015; a=rsa-sha256; cv=pass; t=1793808506; b=jEprynAs5uhOOobLsFDuMiinVgSEWmvJT6BXNEEih63ZStEhVgeB8Ja3msOlYtqAlplZ D+7VrVy5AotgkNPW+uY36A2uM+m4l3VQeeKayrkbiSvVBQTRvJnap+fBoAKBr9a55J29t O3MKWcV8M1Rz3NVaDh9skjEUr53aQFheCTz5Wv4hi34lfO2ckKndWHs9XUQy3KvKWAAPq QEH0nnnaQx6yR1M7zd7pdcy5k5LYeVhnV5+4nFyRHRPxC0xyuqc1M1SFvxInxrdK7nkv5 /Iippxqu8SRaaES7FiKl+Kn/kYV58qJE0E7/4Uq1dUSuLRvoPtl6b+aSnFvlSqXwl3Q== ARC-Message-Signature: i=2; d=subspace.kernel.org; s=arc20261015; a=rsa-sha256; c=relaxed/relaxed; t=1793808506; h=DMARC-Filter:DKIM-Signature:Received:Received: X-MS-Exchange-Authentication-Results:Received-SPF:Received:Received: Received:Received:From:To:CC:Subject:Date:Message-ID:X-Mailer: In-Reply-To:References:MIME-Version:Content-Transfer-Encoding: Content-Type:X-NV-OnPremToCloud:X-EOPAttributedMessage: X-MS-PublicTrafficType:X-MS-TrafficTypeDiagnostic: X-MS-Office365-Filtering-Correlation-Id:X-MS-Exchange-SenderADCheck: X-MS-Exchange-AntiSpam-Relay:X-Microsoft-Antispam: X-Microsoft-Antispam-Message-Info:X-Forefront-Antispam-Report: X-MS-Exchange-AntiSpam-MessageData-ChunkCount: X-MS-Exchange-AntiSpam-MessageData-0:X-OriginatorOrg: X-MS-Exchange-CrossTenant-OriginalArrivalTime: X-MS-Exchange-CrossTenant-Network-Message-Id: X-MS-Exchange-CrossTenant-Id: X-MS-Exchange-CrossTenant-OriginalAttributedTenantConnectingIp: X-MS-Exchange-CrossTenant-AuthSource:X-MS-Exchange-CrossTenant-AuthAs: X-MS-Exchange-CrossTenant-FromEntityHeader: X-MS-Exchange-Transport-CrossTenantHeadersStamped; bh=6tj+5d0UfNnMCWkAJDG8Qeo+KwXLzoRZ7s4oS0H+lhU=; b=c4Z7TgsZlPE4N4Gkwh4u2hEDCBdva6h+raHQMvzGQgUbxfoabMt91lRzSkvG/lJ6l1DR 3TPQBor7CUlFOkp0Sai/UKuOE3N1occeq8A7ATVIOWUqsQ4TX+wZE7sR/Lq5WvMTlUfWU JF7oqMu/8MpAOMf9NO1Ih4xqrkd4stNFMV4jWAma/Uy8e0SF7cCjLOesU7vIz3Be3v1TY wrPALm01uRHp0DvTUIKjyQOixkbZtwV9G379wnVJCsmWrEx3lKtC7QgmBuCKcoKfp+qhg +5dfQo19Wqm16Mxjl3PGeAVqSirvl97APT0qyLYuSBq1wm6kxYngj909q6c/TvKhXrA== ARC-Authentication-Results: i=2; smtp.subspace.kernel.org; dmarc=pass header.from=nvidia.com; spf=fail smtp.mailfrom=nvidia.com; dkim=pass header.d=Nvidia.com header.i=@Nvidia.com header.a=rsa-sha256 header.s=selector2 header.b=GDwQMHeC; arc=pass header.oldest-pass=0 smtp.remote-ip=40.93.195.31 Received: from SN4PR2101CU001.outbound.protection.outlook.com (mail-southcentralusazon11012031.outbound.protection.outlook.com [40.93.195.31]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange ECDHE (secp384r1) server-signature ECDSA (prime256v1) server-digest SHA256) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id E1351606A332; Wed, 4 Nov 2026 16:08:24 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.4.2 smtp.subspace.kernel.org E1351606A332 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=reject dis=none) header.from=nvidia.com Authentication-Results: smtp.subspace.kernel.org; spf=fail smtp.mailfrom=nvidia.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key, unprotected) header.d=Nvidia.com header.i=@Nvidia.com header.a=rsa-sha256 header.s=selector2 header.b=GDwQMHeC ARC-Seal: i=1; a=rsa-sha256; s=arcselector10001; d=microsoft.com; cv=none; b=Jnyad5gBxRxYR3fxrGylEEQPi2yNAE8jH2hN8xyxbh7z3iJS2l1l9hT0ZbVPdoopRQ3R7U148jOJDmRxBlEZ3+KpKSYROjTBiXMwr8wAE8owtWHPA2IPOUFs3bBAOzKFKA57XzpdwiCMikCe6hGdeHTxjBZSvXrvrxxMxEHZmn5qyx7GVZ0dssQPSWTGEKKqa00Isj37YNRU9G6iW/FAiForlDQIXwKIDixOrKFybLeyZkX7dx4JqOZjzpaDCBT/8FRwBQM1bV18s4Ja2bUQRPAAUMGIl0wUr6ZkxuQ8Jff2kTKcz1YHHeTw3iZGvMwb/cdrUw+R11i6bcQ/p0vWTw== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=microsoft.com; s=arcselector10001; h=From:Date:Subject:Message-ID:Content-Type:MIME-Version:X-MS-Exchange-AntiSpam-MessageData-ChunkCount:X-MS-Exchange-AntiSpam-MessageData-0:X-MS-Exchange-AntiSpam-MessageData-1; bh=6tj+5d0UfNnMCWkAJDG8Qeo+KwXLzoRZ7s4oS0H+lhU=; b=MUdCduqfXOr+srG7t4eZEvLe6FEtLqr7rUOWlg5aEf4dHbapPRNquEcLmbjkdYS949SrBTqTN583EDT4pzdwel6EJLWQ6AIh4rflskZE3ijQL+Xau7J8xKCcuHR6jBMcOPTN9SXql1HGkVrv8fQweXbloNYJArwWign21Qyl8BYsjaDGpL+orNoB6kwjw1GyXbsqyiWNTIZXHNCLZCQQlnPfCtaevpVtmCyAjAmz3vZLJbd6nt2KcZyxTEaJqfWHs/BrQJqTCHpc8WvaA+IWvpXTNiaAKQkkCfmrWpvd7eU1RefOMk1SOTlw2ReYD7ooOrQRnh0bHEti+x1U1XwMTg== ARC-Authentication-Results: i=1; mx.microsoft.com 1; spf=pass (sender ip is 216.228.117.161) smtp.rcpttodomain=kernel.org smtp.mailfrom=nvidia.com; dmarc=pass (p=reject sp=reject pct=100) action=none header.from=nvidia.com; dkim=none (message not signed); arc=none (0) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=Nvidia.com; s=selector2; h=From:Date:Subject:Message-ID:Content-Type:MIME-Version:X-MS-Exchange-SenderADCheck; bh=6tj+5d0UfNnMCWkAJDG8Qeo+KwXLzoRZ7s4oS0H+lhU=; b=GDwQMHeCKijM4MMjFQyDqrDtj07xPKuvjz49h6f6VRdufKnDGHqyuE+FpcndCCBg1ezYH/M0oMMjthuuBvCpg2eXTuPfiA77MAkckx1LwjGIShckx1xyxQz0FDwn7ia3ffvQPlZn13Jp6eSY0gqf8ao5YM8lYRsk+aryskLoiHuQpqPCNtZ785V60uXAMSeKqa1XxGo0sX3lV/XWPuXohG3LTUNVCuFOs+m4KXHI+Uk6RYeMhy6afwxKtJHV1+G2aWoLvxzaD9Rc0mFmLIPyWF+C++Uff2NYueGSFO1Fg3/l01Db4m73Fa0mUW2wZwwX3TSiEB2zWFZnpKIs4njGfg== Received: from PH7P220CA0100.NAMP220.PROD.OUTLOOK.COM (2603:10b6:510:32d::14) by SN7PR12MB8025.namprd12.prod.outlook.com (2603:10b6:806:340::12) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.21.592.10; Wed, 4 Nov 2026 16:08:12 +0000 Received: from CY4PEPF000392D6.namprd03.prod.outlook.com (2603:10b6:510:32d:cafe::a9) by PH7P220CA0100.outlook.office365.com (2603:10b6:510:32d::14) with Microsoft SMTP Server (version=TLS1_3, cipher=TLS_AES_256_GCM_SHA384) id 15.21.592.9 via Frontend Transport; Wed, 4 Nov 2026 16:08:12 +0000 X-MS-Exchange-Authentication-Results: mx.microsoft.com 1; spf=pass (sender IP is 216.228.117.161) smtp.mailfrom=nvidia.com; dkim=none (message not signed) header.d=none;dmarc=pass action=none header.from=nvidia.com; Received-SPF: Pass (protection.outlook.com: domain of nvidia.com designates 216.228.117.161 as permitted sender) receiver=protection.outlook.com; client-ip=216.228.117.161; helo=mail.nvidia.com; pr=C Received: from mail.nvidia.com (216.228.117.161) by CY4PEPF000392D6.mail.protection.outlook.com (10.167.247.116) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.21.592.5 via Frontend Transport; Wed, 4 Nov 2026 16:08:12 +0000 Received: from rnnvmail204.nvidia.com (10.129.68.6) by mail.nvidia.com (10.129.200.67) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.2.2562.56; Wed, 4 Nov 2026 08:07:47 -0800 Received: from rnnvmail204.nvidia.com (10.129.68.6) by rnnvmail204.nvidia.com (10.129.68.6) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.2.2562.56; Wed, 4 Nov 2026 08:07:46 -0800 Received: from build-bwicaksono-noble-20260804.internal (10.127.8.11) by mail.nvidia.com (10.129.68.6) with Microsoft SMTP Server id 15.2.2562.56 via Frontend Transport; Wed, 4 Nov 2026 08:07:45 -0800 From: Besar Wicaksono To: , , , CC: , , , , , , , , , , , , , , , , , , , , Besar Wicaksono Subject: [PATCH v3 1/7] perf jevents: Add support for NVIDIA Tegra410 UCF PMU aliasing Date: Wed, 4 Nov 2026 16:07:10 +0000 Message-ID: <20261104160716.2910873-2-bwicaksono@nvidia.com> X-Mailer: git-send-email 2.43.0 In-Reply-To: <20261104160716.2910873-1-bwicaksono@nvidia.com> References: <20261104160716.2910873-1-bwicaksono@nvidia.com> X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Post: List-Help: List-Subscribe: List-Unsubscribe: List-Owner: Precedence: list MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain X-NV-OnPremToCloud: ExternallySecured X-EOPAttributedMessage: 0 X-MS-PublicTrafficType: Email X-MS-TrafficTypeDiagnostic: CY4PEPF000392D6:EE_|SN7PR12MB8025:EE_ X-MS-Office365-Filtering-Correlation-Id: 435f6855-d669-46cb-31db-08df3a8dd395 X-MS-Exchange-SenderADCheck: 1 X-MS-Exchange-AntiSpam-Relay: 0 X-Microsoft-Antispam: BCL:0;ARA:13230040|376014|7416014|23010399003|1800799024|36860700016|82310400026|11063799006|5023799004|56012099006|261021121427099003|261009223027099003|261022075727099003|261026121527099003|261022105027099003|261028144727099003|261026140227099003|261102123227099003|10067099003|18002099003|22082099003|6133799003; X-Microsoft-Antispam-Message-Info: 52Gif9BZms66Xb09aFZX7v9OelF1Qf+TdY8aTt9Er441w3khIR/Cl5ULUgij2fnpfOILHZhUcuH8qgTGG6N8j0eBKIVDsqUEXO90U2Ed1M+fQEKY3IINdR5e+fkcj20rYa3c9XIfJkJCiZ9iKD7/Y/3KIclErw8GAdnAjUJ/kcjxOJF/06iWvdvOBwbz2uDvo/KBJ4fuhUEgmZxlAbh+16ojQcdfzCcalVcyjzUiSb1UIALUetmBvOz3B4j9szzUvkrxq4i21JHOJjAVD8gjQusV3whMsQB1sOVnhsm6zn27mUFWLgiE9yDIDlzJH1D6l5HJoetAyxqKeJtCWLTKWnojFWoQ1y+6kfHgrvguxPk7Uz4Zf6KZ7wktRhQGPuOOLvbx0sW4I8ch+Vq3jkREOuiCderCJ1mVhZTB/CeB/+uEE2zxKK9rZKt1g6lQ7hpuSPu6V5PLilFcPZxZvavlxAJFqHy9DY8ObVp8HVavMqLsGkFXOIjW6D4zCUi+DD9k/fjmC/bqgRDrD3kh/NQLTND5RlNKdb1/lGC+efU6ars7iJqJVU9yW6jFyzT+G2yCs6tSOzF7IBw7v7lWKCjIKk0jtkpjG+NKx4Me8P9itWH+gFyxzZCK0Eyk32bQvJ1G5fBItfUSUOW5xWFr+StpE6HQSHUQEz8amJBn/3VWP+n63L0oL2h7kyrftFVqmW+VhJ/Ncny7mi16shyWyt8yog== X-Forefront-Antispam-Report: CIP:216.228.117.161;CTRY:US;LANG:en;SCL:1;SRV:;IPV:NLI;SFV:NSPM;H:mail.nvidia.com;PTR:dc6edge2.nvidia.com;CAT:NONE;SFS:(13230040)(376014)(7416014)(23010399003)(1800799024)(36860700016)(82310400026)(11063799006)(5023799004)(56012099006)(261021121427099003)(261009223027099003)(261022075727099003)(261026121527099003)(261022105027099003)(261028144727099003)(261026140227099003)(261102123227099003)(10067099003)(18002099003)(22082099003)(6133799003);DIR:OUT;SFP:1101; X-MS-Exchange-AntiSpam-MessageData-ChunkCount: 1 X-MS-Exchange-AntiSpam-MessageData-0: jowNJfTax0qWf4isbOU0vHobcU991UpC7wrHuiBciQnAypPPfizfSmsz1W/4bU8HaHYUlvbZ/8PNrGjoMHfK4cNB6FnN8r8BZfJSkh9Wd2tUMeeQZ1BCTgRs2SSXVdzZe9VIfFmwMJyuG4HROzUKAY8GRXnbSsfrqDJ3I9oRospAC/2ZRRUHJf0d151oZYrsj+46AJfeqXtsi6YwA45cET0OTq1MiBGyMJ+v+TRsbxuBBBeC4gmYayVg1PoQJ9u7uUtd+j0Z4ieoEHGtc1tnxZVURuKbO/q8LZZnE1ZGXY2tFimmzyvvt0b9GBDS+kueuN4wUOB5YITmI+3nrOOU9PUckkkqYtdA7jzMgifY0B7yBli5d6yFAsBV4Q3vxDqQgHMTIWwc1agXGiAYa8gM/g11azEAYUd2741v4QD73F3dF83XW0DuOlBX2MlduNco X-OriginatorOrg: Nvidia.com X-MS-Exchange-CrossTenant-OriginalArrivalTime: 04 Nov 2026 16:08:12.2182 (UTC) X-MS-Exchange-CrossTenant-Network-Message-Id: 435f6855-d669-46cb-31db-08df3a8dd395 X-MS-Exchange-CrossTenant-Id: 43083d15-7273-40c1-b7db-39efd9ccc17a X-MS-Exchange-CrossTenant-OriginalAttributedTenantConnectingIp: TenantId=43083d15-7273-40c1-b7db-39efd9ccc17a;Ip=[216.228.117.161];Helo=[mail.nvidia.com] X-MS-Exchange-CrossTenant-AuthSource: CY4PEPF000392D6.namprd03.prod.outlook.com X-MS-Exchange-CrossTenant-AuthAs: Anonymous X-MS-Exchange-CrossTenant-FromEntityHeader: HybridOnPrem X-MS-Exchange-Transport-CrossTenantHeadersStamped: SN7PR12MB8025 Add alias support for NVIDIA Tegra410 UCF PMU events and metrics. Signed-off-by: Besar Wicaksono --- .../arch/arm64/nvidia/t410/uncore-ucf.json | 764 ++++++++++++++++++ 1 file changed, 764 insertions(+) create mode 100644 tools/perf/pmu-events/arch/arm64/nvidia/t410/uncore-ucf= .json diff --git a/tools/perf/pmu-events/arch/arm64/nvidia/t410/uncore-ucf.json b= /tools/perf/pmu-events/arch/arm64/nvidia/t410/uncore-ucf.json new file mode 100644 index 000000000000..9cb964f7c53d --- /dev/null +++ b/tools/perf/pmu-events/arch/arm64/nvidia/t410/uncore-ucf.json @@ -0,0 +1,764 @@ +[ + { + "EventCode": "0x100000000", + "EventName": "UCF_CYCLES", + "BriefDescription": "Counts UCF cycles using the dedicated PMU cyc= le counter.", + "PublicDescription": "Counts UCF cycles using the PMU's dedicated = cycle counter.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "EventCode": "0x1d", + "EventName": "UCF_BUS_CYCLES", + "BriefDescription": "Counts UCF cycles using a programmable event = counter.", + "PublicDescription": "Counts UCF cycles using a programmable event= counter. The count is identical to UCF_CYCLES. Source and destination filt= ers are ignored.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "EventCode": "0x111", + "EventName": "UCF_SLC_ACCESS_RD", + "BriefDescription": "Counts read requests from a requester to the = SLC.", + "PublicDescription": "Counts read requests from a requester to the= SLC. A read is a request for which the completer returns data: dataless Ma= keReadUnique and snoop requests are not reads, and prefetches issued by the= cores' L1/L2 caches count as reads rather than as prefetch requests. Filte= rs: src_loc_cpu, src_loc_noncpu and src_rem select the request initiator: l= ocal CPU, local non-CPU device or remote socket. dst_loc_cmem, dst_loc_gmem= , dst_loc_other and dst_rem select the memory type of the request address: = local CMEM, local GMEM, local CXL memory or remote memory; local or remote = is determined by the home socket of the address. An unset group counts all.= ", + "Unit": "nvidia_ucf_pmu_" + }, + { + "EventCode": "0x112", + "EventName": "UCF_SLC_ACCESS_WR", + "BriefDescription": "Counts write requests from a requester to the= SLC.", + "PublicDescription": "Counts write requests from a requester to th= e SLC. A write is a request in which the requester sends data: dataless Wri= teEvictOrEvict is not a write, and CHI Write*Zero is a write because it eve= ntually consumes DRAM bandwidth although it sends no data over the interfac= e. Filters: src_loc_cpu, src_loc_noncpu and src_rem select the request init= iator: local CPU, local non-CPU device or remote socket. dst_loc_cmem, dst_= loc_gmem, dst_loc_other and dst_rem select the memory type of the request a= ddress: local CMEM, local GMEM, local CXL memory or remote memory; local or= remote is determined by the home socket of the address. An unset group cou= nts all.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "EventCode": "0x183", + "EventName": "UCF_SLC_ACCESS_DATALESS", + "BriefDescription": "Counts CHI Dataless requests to the SLC.", + "PublicDescription": "Counts CHI Dataless requests to the SLC. Dat= aless MakeReadUnique and WriteEvictOrEvict requests are not counted. Filter= s: src_loc_cpu, src_loc_noncpu and src_rem select the request initiator: lo= cal CPU, local non-CPU device or remote socket. dst_loc_cmem, dst_loc_gmem,= dst_loc_other and dst_rem select the memory type of the request address: l= ocal CMEM, local GMEM, local CXL memory or remote memory; local or remote i= s determined by the home socket of the address. An unset group counts all.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "EventCode": "0x184", + "EventName": "UCF_SLC_ACCESS_ATOMIC", + "BriefDescription": "Counts CHI Atomic requests to the SLC.", + "PublicDescription": "Counts CHI Atomic requests to the SLC. Filte= rs: src_loc_cpu, src_loc_noncpu and src_rem select the request initiator: l= ocal CPU, local non-CPU device or remote socket. dst_loc_cmem, dst_loc_gmem= , dst_loc_other and dst_rem select the memory type of the request address: = local CMEM, local GMEM, local CXL memory or remote memory; local or remote = is determined by the home socket of the address. An unset group counts all.= ", + "Unit": "nvidia_ucf_pmu_" + }, + { + "EventCode": "0x113", + "EventName": "UCF_SLC_BYTES_RD", + "BriefDescription": "Counts bytes transferred by the read requests= counted by UCF_SLC_ACCESS_RD.", + "PublicDescription": "Counts the bytes transferred by the read req= uests counted by UCF_SLC_ACCESS_RD, including read data sent by Direct Memo= ry Transfer (DMT) or Direct Cache Transfer (DCT). For transfers of an odd n= umber of bytes, the count is rounded to the nearest power of 2 greater than= the bytes transferred. Because reads can be smaller than 64 bytes, the cou= nt may not equal UCF_SLC_ACCESS_RD times 64. Source and destination filters= work as described for UCF_SLC_ACCESS_RD.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "EventCode": "0x114", + "EventName": "UCF_SLC_BYTES_WR", + "BriefDescription": "Counts bytes transferred by the write request= s counted by UCF_SLC_ACCESS_WR.", + "PublicDescription": "Counts the bytes transferred by the write re= quests counted by UCF_SLC_ACCESS_WR, including write data sent by Direct Wr= ite Transfer (DWT). For transfers of an odd number of bytes, the count is r= ounded to the nearest power of 2. Data returned to the SLC by a requester's= cache in response to a snoop is not counted. Because writes can be smaller= than 64 bytes, the count may not equal UCF_SLC_ACCESS_WR times 64. Source = and destination filters work as described for UCF_SLC_ACCESS_WR.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "EventCode": "0x119", + "EventName": "UCF_SLC_HIT_RD", + "BriefDescription": "Counts inbound read requests that hit in the = SLC's L3 cache or in a peer requester cache.", + "PublicDescription": "Counts inbound read requests that hit in the= SLC's L3 cache or in a peer requester cache, including one in a remote soc= ket. Some hits are still refilled from memory (see UCF_SLC_REFILL_RD). Filt= ers: src_loc_cpu, src_loc_noncpu and src_rem select the request initiator: = local CPU, local non-CPU device or remote socket. dst_loc_cmem, dst_loc_gme= m, dst_loc_other and dst_rem select the memory type of the request address:= local CMEM, local GMEM, local CXL memory or remote memory; local or remote= is determined by the home socket of the address. An unset group counts all= .", + "Unit": "nvidia_ucf_pmu_" + }, + { + "EventCode": "0x109", + "EventName": "UCF_SLC_REFILL_RD", + "BriefDescription": "Counts inbound read requests not supplied by = the local socket's L1, L2 or L3 caches.", + "PublicDescription": "Counts inbound read requests not supplied by= the local socket's L1, L2 or L3 caches. For remote-memory addresses, it al= so counts each Shared-to-Unique permission upgrade (dataless MakeReadUnique= ) of a cache line shared by requesters in different sockets. A refill is no= t the same as a miss: the SLC may fetch data from memory even when a reques= ter cache holds clean data, and such accesses are hits that are still refil= led. Filters: src_loc_cpu, src_loc_noncpu and src_rem select the request in= itiator: local CPU, local non-CPU device or remote socket. dst_loc_cmem, ds= t_loc_gmem, dst_loc_other and dst_rem select the memory type of the request= address: local CMEM, local GMEM, local CXL memory or remote memory; local = or remote is determined by the home socket of the address. An unset group c= ounts all.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "EventCode": "0x10a", + "EventName": "UCF_SLC_REFILL_WR", + "BriefDescription": "Counts inbound write requests not supplied by= the local socket's L1, L2 or L3 caches.", + "PublicDescription": "Counts inbound write requests not supplied b= y the local socket's L1, L2 or L3 caches. A write refill occurs only for a = partial write that needs the existing data; dataless write requests do not = cause a refill and are not counted. Filters: src_loc_cpu, src_loc_noncpu an= d src_rem select the request initiator: local CPU, local non-CPU device or = remote socket. dst_loc_cmem, dst_loc_gmem, dst_loc_other and dst_rem select= the memory type of the request address: local CMEM, local GMEM, local CXL = memory or remote memory; local or remote is determined by the home socket o= f the address. An unset group counts all.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "EventCode": "0xf0", + "EventName": "UCF_SLC_ALLOCATE", + "BriefDescription": "Counts read, write, atomic or prefetch access= es that allocate a line in the SLC's L3 cache.", + "PublicDescription": "Counts read, write, atomic or prefetch acces= ses that allocate a line in the SLC's L3 cache. Filters: src_loc_cpu, src_l= oc_noncpu and src_rem select the request initiator: local CPU, local non-CP= U device or remote socket. dst_loc_cmem, dst_loc_gmem, dst_loc_other and ds= t_rem select the memory type of the request address: local CMEM, local GMEM= , local CXL memory or remote memory; local or remote is determined by the h= ome socket of the address. An unset group counts all.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "EventCode": "0xf3", + "EventName": "UCF_SLC_WB", + "BriefDescription": "Counts SLC eviction and coherence write-backs= , and non-allocating write-backs from upstream clients.", + "PublicDescription": "Counts SLC write-backs: eviction write-backs= , coherence write-backs, and non-allocating write-backs from upstream clien= ts. Evictions include L3 cache and snoop filter evictions. A coherence writ= e-back occurs when data is dirty in the L3 cache or another requester node = cache, but a requesting agent needs clean data. A home SLC in another socke= t, or the GL2 of a connected local or remote GPU (the point of coherence fo= r GPU memory), can evict a line and snoop this SLC. A data-bearing snoop re= sponse from this SLC is not counted by this event, though it uses this SLC'= s downstream bandwidth. For a home SLC eviction, the home SLC's events coun= t the downstream write traffic it causes. Filters: src_loc_cpu, src_loc_non= cpu and src_rem select the initiator of the request that caused the write-b= ack: local CPU, local non-CPU device or remote socket. This need not be the= agent that produced or modified the data. dst_loc_cmem, dst_loc_gmem, dst_= loc_other and dst_rem select the memory type of the memory written back: lo= cal CMEM, local GMEM, local CXL memory or remote memory; local or remote is= determined by the home socket of the address. An unset group counts all.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "EventCode": "0x180", + "EventName": "UCF_LOCAL_SNOOP", + "BriefDescription": "Counts snoop requests sent by this SLC to loc= al fully coherent requester nodes or to an SLC in another socket.", + "PublicDescription": "Counts snoop requests sent by this SLC to fu= lly coherent requester nodes in the local socket or to an SLC in another so= cket, including an SLC that this SLC tracks as a sharing node. Invalidating= , forwarding, and non-forwarding snoops are included. Each snoop request is= counted separately, even when one transaction snoops several clients for t= he same cache line. Snoops generated for atomic requests or dataless reques= ts are not counted. Filters: src_loc_cpu, src_loc_noncpu and src_rem select= the initiator of the request that caused the snoop: local CPU, local non-C= PU device or remote socket. dst_loc_cmem, dst_loc_gmem, dst_loc_other and d= st_rem select the memory type of the request address: local CMEM, local GME= M, local CXL memory or remote memory; local or remote is determined by the = home socket of the address. An unset group counts all.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "EventCode": "0x181", + "EventName": "UCF_EXT_SNP_ACCESS", + "BriefDescription": "Counts snoop requests that a home SLC or GL2 = in another socket sends to this SLC.", + "PublicDescription": "Counts snoop requests that a home SLC or GL2= in another socket sends to this SLC. Snoops are counted whether or not the= response carries data. Filters: every counted snoop matches src_rem and ds= t_rem, since it comes from a remote initiator for an address homed in anoth= er socket. An unset source or destination group counts all.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "EventCode": "0x182", + "EventName": "UCF_EXT_SNP_EVICT", + "BriefDescription": "Counts snoop requests from a home SLC or GL2 = in another socket that invalidate an L3 cache or snoop filter entry in this= SLC.", + "PublicDescription": "Counts snoop requests from a home SLC or GL2= in another socket that invalidate an L3 cache entry or snoop filter entry = in this SLC. Snoops are counted whether or not the response carries data, s= o this event cannot give the number or volume of data-bearing snoop respons= es (see UCF_SLC_WB). Filters: every counted snoop matches src_rem and dst_r= em, since it comes from a remote initiator for an address homed in another = socket. An unset source or destination group counts all.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "EventCode": "0x121", + "EventName": "UCF_MEM_ACCESS_RD", + "BriefDescription": "Counts read requests from the SLC toward loca= l CMEM, local GMEM, local CXL memory or remote memory.", + "PublicDescription": "Counts read requests from the SLC toward loc= al CMEM, local GMEM, local CXL memory or remote memory. For remote memory i= t can overcount: it also increments for each Shared-to-Unique permission up= grade (dataless MakeReadUnique) of a cache line shared by requesters in dif= ferent sockets. To count reads from remote sources without this overcount, = count UCF_SLC_ACCESS_RD on the memory's home socket with src_rem, dst_loc_c= mem, dst_loc_gmem and dst_loc_other set. Filters: src_loc_cpu, src_loc_nonc= pu and src_rem select the request initiator: local CPU, local non-CPU devic= e or remote socket. dst_loc_cmem, dst_loc_gmem, dst_loc_other and dst_rem s= elect the memory type of the request address: local CMEM, local GMEM, local= CXL memory or remote memory; local or remote is determined by the home soc= ket of the address. An unset group counts all.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "EventCode": "0x122", + "EventName": "UCF_MEM_ACCESS_WR", + "BriefDescription": "Counts write requests from the SLC toward loc= al CMEM, local GMEM, local CXL memory or remote memory.", + "PublicDescription": "Counts write requests from the SLC toward lo= cal CMEM, local GMEM, local CXL memory or remote memory. It includes SLC wr= ite-backs, which are attributed as described for UCF_SLC_WB. Data-bearing s= noop responses to evictions by a home SLC in another socket or a connected = GPU's GL2 are not counted (see UCF_SLC_WB). Filters: src_loc_cpu, src_loc_n= oncpu and src_rem select the request initiator: local CPU, local non-CPU de= vice or remote socket. dst_loc_cmem, dst_loc_gmem, dst_loc_other and dst_re= m select the memory type of the request address: local CMEM, local GMEM, lo= cal CXL memory or remote memory; local or remote is determined by the home = socket of the address. An unset group counts all.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "EventCode": "0x123", + "EventName": "UCF_MEM_BYTES_RD", + "BriefDescription": "Counts bytes transferred by the read requests= counted by UCF_MEM_ACCESS_RD.", + "PublicDescription": "Counts bytes transferred by the read request= s counted by UCF_MEM_ACCESS_RD. Source and destination filters work as desc= ribed for UCF_MEM_ACCESS_RD.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "EventCode": "0x124", + "EventName": "UCF_MEM_BYTES_WR", + "BriefDescription": "Counts bytes transferred by the write request= s counted by UCF_MEM_ACCESS_WR.", + "PublicDescription": "Counts bytes transferred by the write reques= ts counted by UCF_MEM_ACCESS_WR, including write-back data attributed as de= scribed for UCF_SLC_WB. Data-bearing snoop responses to evictions by a home= SLC in another socket or a connected GPU's GL2 are not counted (see UCF_SL= C_WB). Source and destination filters work as described for UCF_MEM_ACCESS_= WR.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_freq", + "MetricGroup": "UCF_Frequency", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "(UCF_CYCLES / aggr_nr(UCF_CYCLES)) / 1e9 / duration= _time", + "ScaleUnit": "1GHz", + "BriefDescription": "UCF clock frequency.", + "PublicDescription": "Average UCF clock frequency over the measure= ment interval.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_slc_rd_bandwidth_all_src_to_all_dst", + "MetricGroup": "UCF_BW_Upstream_RD;UCF_BW_All_Src_To_All_Dst", + "MetricExpr": "UCF_SLC_BYTES_RD@src_loc_cpu=3D1,src_loc_noncpu=3D1= ,src_rem=3D1,dst_loc_cmem=3D1,dst_loc_gmem=3D1,dst_loc_other=3D1,dst_rem=3D= 1@ / 1e6 / duration_time", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average requester-to-SLC read bandwidth for r= equests from all sources to all destinations.", + "PublicDescription": "Average requester-to-SLC read bandwidth of r= equests from local CPUs, local non-CPU devices and remote sockets for local= CMEM, local GMEM, local CXL memory and remote memory addresses.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_mem_rd_bandwidth_all_src_to_all_dst", + "MetricGroup": "UCF_BW_Downstream_RD;UCF_BW_All_Src_To_All_Dst", + "MetricExpr": "UCF_MEM_BYTES_RD@src_loc_cpu=3D1,src_loc_noncpu=3D1= ,src_rem=3D1,dst_loc_cmem=3D1,dst_loc_gmem=3D1,dst_loc_other=3D1,dst_rem=3D= 1@ / 1e6 / duration_time", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average SLC-to-memory read bandwidth for all = memory destinations, for requests from all initiators.", + "PublicDescription": "Average SLC-to-memory read bandwidth for loc= al CMEM, local GMEM, local CXL memory and remote memory, for requests initi= ated by local CPUs, local non-CPU devices and remote sockets.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_slc_wr_bandwidth_all_src_to_all_dst", + "MetricGroup": "UCF_BW_Upstream_WR;UCF_BW_All_Src_To_All_Dst", + "MetricExpr": "UCF_SLC_BYTES_WR@src_loc_cpu=3D1,src_loc_noncpu=3D1= ,src_rem=3D1,dst_loc_cmem=3D1,dst_loc_gmem=3D1,dst_loc_other=3D1,dst_rem=3D= 1@ / 1e6 / duration_time", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average requester-to-SLC write bandwidth for = requests from all sources to all destinations.", + "PublicDescription": "Average requester-to-SLC write bandwidth of = requests from local CPUs, local non-CPU devices and remote sockets for loca= l CMEM, local GMEM, local CXL memory and remote memory addresses.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_mem_wr_bandwidth_all_src_to_all_dst", + "MetricGroup": "UCF_BW_Downstream_WR;UCF_BW_All_Src_To_All_Dst", + "MetricExpr": "UCF_MEM_BYTES_WR@src_loc_cpu=3D1,src_loc_noncpu=3D1= ,src_rem=3D1,dst_loc_cmem=3D1,dst_loc_gmem=3D1,dst_loc_other=3D1,dst_rem=3D= 1@ / 1e6 / duration_time", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average SLC-to-memory write bandwidth for loc= al and remote memory, for traffic from all initiators.", + "PublicDescription": "Average SLC-to-memory write bandwidth for lo= cal CMEM, local GMEM, local CXL memory and remote memory, for traffic initi= ated by local CPUs, local non-CPU devices and remote sockets. Includes writ= e-backs, attributed as described for UCF_SLC_WB. Data-bearing snoop respons= es to evictions by a home SLC in another socket or a connected GPU's GL2 ar= e not counted (see UCF_SLC_WB).", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_slc_bandwidth_all_src_to_all_dst", + "MetricGroup": "UCF_BW_All_Src_To_All_Dst", + "MetricExpr": "ucf_avg_slc_rd_bandwidth_all_src_to_all_dst + ucf_a= vg_slc_wr_bandwidth_all_src_to_all_dst", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average requester-to-SLC read and write bandw= idth for requests from all sources to all destinations.", + "PublicDescription": "Average requester-to-SLC read and write band= width of requests from local CPUs, local non-CPU devices and remote sockets= for local CMEM, local GMEM, local CXL memory and remote memory addresses.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_mem_bandwidth_all_src_to_all_dst", + "MetricGroup": "UCF_BW_All_Src_To_All_Dst", + "MetricExpr": "ucf_avg_mem_rd_bandwidth_all_src_to_all_dst + ucf_a= vg_mem_wr_bandwidth_all_src_to_all_dst", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average SLC-to-memory read and write bandwidt= h for local and remote memory, for traffic from all initiators.", + "PublicDescription": "Average SLC-to-memory read and write bandwid= th for local CMEM, local GMEM, local CXL memory and remote memory, for traf= fic initiated by local CPUs, local non-CPU devices and remote sockets. Writ= es include write-backs, attributed as described for UCF_SLC_WB. Data-bearin= g snoop responses to evictions by a home SLC in another socket or a connect= ed GPU's GL2 are not counted (see UCF_SLC_WB).", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_slc_rd_bandwidth_all_src_to_local_cmem", + "MetricGroup": "UCF_BW_Upstream_RD;UCF_BW_Local_CMEM;UCF_BW_All_Sr= c_To_Local_CMEM", + "MetricExpr": "UCF_SLC_BYTES_RD@src_loc_cpu=3D1,src_loc_noncpu=3D1= ,src_rem=3D1,dst_loc_cmem=3D1@ / 1e6 / duration_time", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average requester-to-SLC read bandwidth for r= equests from all sources to local CMEM addresses.", + "PublicDescription": "Average requester-to-SLC read bandwidth of r= equests from local CPUs, local non-CPU devices and remote sockets for local= CMEM addresses.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_mem_rd_bandwidth_all_src_to_local_cmem", + "MetricGroup": "UCF_BW_Downstream_RD;UCF_BW_Local_CMEM;UCF_BW_All_= Src_To_Local_CMEM", + "MetricExpr": "UCF_MEM_BYTES_RD@src_loc_cpu=3D1,src_loc_noncpu=3D1= ,src_rem=3D1,dst_loc_cmem=3D1@ / 1e6 / duration_time", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average SLC-to-memory read bandwidth for loca= l CMEM, for requests from all initiators.", + "PublicDescription": "Average SLC-to-memory read bandwidth for loc= al CMEM, for requests initiated by local CPUs, local non-CPU devices and re= mote sockets.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_slc_wr_bandwidth_all_src_to_local_cmem", + "MetricGroup": "UCF_BW_Upstream_WR;UCF_BW_Local_CMEM;UCF_BW_All_Sr= c_To_Local_CMEM", + "MetricExpr": "UCF_SLC_BYTES_WR@src_loc_cpu=3D1,src_loc_noncpu=3D1= ,src_rem=3D1,dst_loc_cmem=3D1@ / 1e6 / duration_time", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average requester-to-SLC write bandwidth for = requests from all sources to local CMEM addresses.", + "PublicDescription": "Average requester-to-SLC write bandwidth of = requests from local CPUs, local non-CPU devices and remote sockets for loca= l CMEM addresses.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_mem_wr_bandwidth_all_src_to_local_cmem", + "MetricGroup": "UCF_BW_Downstream_WR;UCF_BW_Local_CMEM;UCF_BW_All_= Src_To_Local_CMEM", + "MetricExpr": "UCF_MEM_BYTES_WR@src_loc_cpu=3D1,src_loc_noncpu=3D1= ,src_rem=3D1,dst_loc_cmem=3D1@ / 1e6 / duration_time", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average SLC-to-memory write bandwidth for loc= al CMEM, for traffic from all initiators.", + "PublicDescription": "Average SLC-to-memory write bandwidth for lo= cal CMEM, for traffic initiated by local CPUs, local non-CPU devices and re= mote sockets. Includes write-backs, attributed as described for UCF_SLC_WB.= ", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_slc_bandwidth_all_src_to_local_cmem", + "MetricGroup": "UCF_BW_All_Src_To_Local_CMEM", + "MetricExpr": "ucf_avg_slc_rd_bandwidth_all_src_to_local_cmem + uc= f_avg_slc_wr_bandwidth_all_src_to_local_cmem", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average requester-to-SLC read and write bandw= idth for requests from all sources to local CMEM addresses.", + "PublicDescription": "Average requester-to-SLC read and write band= width of requests from local CPUs, local non-CPU devices and remote sockets= for local CMEM addresses.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_mem_bandwidth_all_src_to_local_cmem", + "MetricGroup": "UCF_BW_All_Src_To_Local_CMEM", + "MetricExpr": "ucf_avg_mem_rd_bandwidth_all_src_to_local_cmem + uc= f_avg_mem_wr_bandwidth_all_src_to_local_cmem", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average SLC-to-memory read and write bandwidt= h for local CMEM, for traffic from all initiators.", + "PublicDescription": "Average SLC-to-memory read and write bandwid= th for local CMEM, for traffic initiated by local CPUs, local non-CPU devic= es and remote sockets. Writes include write-backs, attributed as described = for UCF_SLC_WB.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_slc_rd_bandwidth_local_src_to_local_cmem", + "MetricGroup": "UCF_BW_Upstream_RD;UCF_BW_Local_CMEM;UCF_BW_Local_= Src_To_Local_CMEM", + "MetricExpr": "ucf_avg_slc_rd_bandwidth_local_cpu_to_local_cmem + = ucf_avg_slc_rd_bandwidth_local_noncpu_to_local_cmem", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average requester-to-SLC read bandwidth for l= ocal CPU and local non-CPU requests to local CMEM addresses.", + "PublicDescription": "Average requester-to-SLC read bandwidth of r= equests from local CPUs and local non-CPU devices such as the GPU and PCIe = devices for local CMEM addresses.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_mem_rd_bandwidth_local_src_to_local_cmem", + "MetricGroup": "UCF_BW_Downstream_RD;UCF_BW_Local_CMEM;UCF_BW_Loca= l_Src_To_Local_CMEM", + "MetricExpr": "ucf_avg_mem_rd_bandwidth_local_cpu_to_local_cmem + = ucf_avg_mem_rd_bandwidth_local_noncpu_to_local_cmem", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average SLC-to-memory read bandwidth for loca= l CMEM, for requests initiated by local CPUs and local non-CPU devices.", + "PublicDescription": "Average SLC-to-memory read bandwidth for loc= al CMEM, for requests initiated by local CPUs and local non-CPU devices suc= h as the GPU and PCIe devices.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_slc_wr_bandwidth_local_src_to_local_cmem", + "MetricGroup": "UCF_BW_Upstream_WR;UCF_BW_Local_CMEM;UCF_BW_Local_= Src_To_Local_CMEM", + "MetricExpr": "ucf_avg_slc_wr_bandwidth_local_cpu_to_local_cmem + = ucf_avg_slc_wr_bandwidth_local_noncpu_to_local_cmem", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average requester-to-SLC write bandwidth for = local CPU and local non-CPU requests to local CMEM addresses.", + "PublicDescription": "Average requester-to-SLC write bandwidth of = requests from local CPUs and local non-CPU devices such as the GPU and PCIe= devices for local CMEM addresses.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_mem_wr_bandwidth_local_src_to_local_cmem", + "MetricGroup": "UCF_BW_Downstream_WR;UCF_BW_Local_CMEM;UCF_BW_Loca= l_Src_To_Local_CMEM", + "MetricExpr": "ucf_avg_mem_wr_bandwidth_local_cpu_to_local_cmem + = ucf_avg_mem_wr_bandwidth_local_noncpu_to_local_cmem", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average SLC-to-memory write bandwidth for loc= al CMEM, for traffic initiated by local CPUs and local non-CPU devices.", + "PublicDescription": "Average SLC-to-memory write bandwidth for lo= cal CMEM, for traffic initiated by local CPUs and local non-CPU devices suc= h as the GPU and PCIe devices. Includes write-backs, attributed as describe= d for UCF_SLC_WB.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_slc_rd_bandwidth_local_src_to_remote_mem", + "MetricGroup": "UCF_BW_Upstream_RD;UCF_BW_Remote_Memory;UCF_BW_Loc= al_Src_To_Remote_Memory", + "MetricExpr": "ucf_avg_slc_rd_bandwidth_local_cpu_to_remote_mem + = ucf_avg_slc_rd_bandwidth_local_noncpu_to_remote_mem", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average requester-to-SLC read bandwidth for l= ocal CPU and local non-CPU requests to remote memory addresses.", + "PublicDescription": "Average requester-to-SLC read bandwidth of r= equests from local CPUs and local non-CPU devices such as the GPU and PCIe = devices for remote memory addresses.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_mem_rd_bandwidth_local_src_to_remote_mem", + "MetricGroup": "UCF_BW_Downstream_RD;UCF_BW_Remote_Memory;UCF_BW_L= ocal_Src_To_Remote_Memory", + "MetricExpr": "ucf_avg_mem_rd_bandwidth_local_cpu_to_remote_mem + = ucf_avg_mem_rd_bandwidth_local_noncpu_to_remote_mem", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average SLC-to-memory read bandwidth for remo= te memory, for requests initiated by local CPUs and local non-CPU devices.", + "PublicDescription": "Average SLC-to-memory read bandwidth for rem= ote memory, for requests initiated by local CPUs and local non-CPU devices = such as the GPU and PCIe devices.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_slc_wr_bandwidth_local_src_to_remote_mem", + "MetricGroup": "UCF_BW_Upstream_WR;UCF_BW_Remote_Memory;UCF_BW_Loc= al_Src_To_Remote_Memory", + "MetricExpr": "ucf_avg_slc_wr_bandwidth_local_cpu_to_remote_mem + = ucf_avg_slc_wr_bandwidth_local_noncpu_to_remote_mem", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average requester-to-SLC write bandwidth for = local CPU and local non-CPU requests to remote memory addresses.", + "PublicDescription": "Average requester-to-SLC write bandwidth of = requests from local CPUs and local non-CPU devices such as the GPU and PCIe= devices for remote memory addresses.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_mem_wr_bandwidth_local_src_to_remote_mem", + "MetricGroup": "UCF_BW_Downstream_WR;UCF_BW_Remote_Memory;UCF_BW_L= ocal_Src_To_Remote_Memory", + "MetricExpr": "ucf_avg_mem_wr_bandwidth_local_cpu_to_remote_mem + = ucf_avg_mem_wr_bandwidth_local_noncpu_to_remote_mem", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average SLC-to-memory write bandwidth for rem= ote memory, for traffic initiated by local CPUs and local non-CPU devices.", + "PublicDescription": "Average SLC-to-memory write bandwidth for re= mote memory, for traffic initiated by local CPUs and local non-CPU devices = such as the GPU and PCIe devices. Includes write-backs, attributed as descr= ibed for UCF_SLC_WB. Data-bearing snoop responses to evictions by a home SL= C in another socket or a connected GPU's GL2 are not counted (see UCF_SLC_W= B).", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_slc_rd_bandwidth_local_cpu_to_local_cmem", + "MetricGroup": "UCF_BW_Upstream_RD;UCF_BW_Local_CPU;UCF_BW_Local_C= MEM;UCF_BW_Local_CPU_To_Local_CMEM", + "MetricExpr": "UCF_SLC_BYTES_RD@src_loc_cpu=3D1,dst_loc_cmem=3D1@ = / 1e6 / duration_time", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average requester-to-SLC read bandwidth for l= ocal CPU requests to local CMEM addresses.", + "PublicDescription": "Average requester-to-SLC read bandwidth of r= equests from local CPUs for local CMEM addresses.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_mem_rd_bandwidth_local_cpu_to_local_cmem", + "MetricGroup": "UCF_BW_Downstream_RD;UCF_BW_Local_CPU;UCF_BW_Local= _CMEM;UCF_BW_Local_CPU_To_Local_CMEM", + "MetricExpr": "UCF_MEM_BYTES_RD@src_loc_cpu=3D1,dst_loc_cmem=3D1@ = / 1e6 / duration_time", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average SLC-to-memory read bandwidth for loca= l CMEM, for requests initiated by local CPUs.", + "PublicDescription": "Average SLC-to-memory read bandwidth for loc= al CMEM, for requests initiated by local CPUs.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_slc_wr_bandwidth_local_cpu_to_local_cmem", + "MetricGroup": "UCF_BW_Upstream_WR;UCF_BW_Local_CPU;UCF_BW_Local_C= MEM;UCF_BW_Local_CPU_To_Local_CMEM", + "MetricExpr": "UCF_SLC_BYTES_WR@src_loc_cpu=3D1,dst_loc_cmem=3D1@ = / 1e6 / duration_time", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average requester-to-SLC write bandwidth for = local CPU requests to local CMEM addresses.", + "PublicDescription": "Average requester-to-SLC write bandwidth of = requests from local CPUs for local CMEM addresses.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_mem_wr_bandwidth_local_cpu_to_local_cmem", + "MetricGroup": "UCF_BW_Downstream_WR;UCF_BW_Local_CPU;UCF_BW_Local= _CMEM;UCF_BW_Local_CPU_To_Local_CMEM", + "MetricExpr": "UCF_MEM_BYTES_WR@src_loc_cpu=3D1,dst_loc_cmem=3D1@ = / 1e6 / duration_time", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average SLC-to-memory write bandwidth for loc= al CMEM, for traffic initiated by local CPUs.", + "PublicDescription": "Average SLC-to-memory write bandwidth for lo= cal CMEM, for traffic initiated by local CPUs. Includes write-backs, attrib= uted as described for UCF_SLC_WB.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_slc_rd_bandwidth_local_cpu_to_local_gmem", + "MetricGroup": "UCF_BW_Upstream_RD;UCF_BW_Local_CPU;UCF_BW_Local_G= MEM;UCF_BW_Local_CPU_To_Local_GMEM", + "MetricExpr": "UCF_SLC_BYTES_RD@src_loc_cpu=3D1,dst_loc_gmem=3D1@ = / 1e6 / duration_time", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average requester-to-SLC read bandwidth for l= ocal CPU requests to local GMEM addresses.", + "PublicDescription": "Average requester-to-SLC read bandwidth of r= equests from local CPUs for local GMEM addresses.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_mem_rd_bandwidth_local_cpu_to_local_gmem", + "MetricGroup": "UCF_BW_Downstream_RD;UCF_BW_Local_CPU;UCF_BW_Local= _GMEM;UCF_BW_Local_CPU_To_Local_GMEM", + "MetricExpr": "UCF_MEM_BYTES_RD@src_loc_cpu=3D1,dst_loc_gmem=3D1@ = / 1e6 / duration_time", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average SLC-to-memory read bandwidth for loca= l GMEM, for requests initiated by local CPUs.", + "PublicDescription": "Average SLC-to-memory read bandwidth for loc= al GMEM, for requests initiated by local CPUs.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_slc_wr_bandwidth_local_cpu_to_local_gmem", + "MetricGroup": "UCF_BW_Upstream_WR;UCF_BW_Local_CPU;UCF_BW_Local_G= MEM;UCF_BW_Local_CPU_To_Local_GMEM", + "MetricExpr": "UCF_SLC_BYTES_WR@src_loc_cpu=3D1,dst_loc_gmem=3D1@ = / 1e6 / duration_time", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average requester-to-SLC write bandwidth for = local CPU requests to local GMEM addresses.", + "PublicDescription": "Average requester-to-SLC write bandwidth of = requests from local CPUs for local GMEM addresses.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_mem_wr_bandwidth_local_cpu_to_local_gmem", + "MetricGroup": "UCF_BW_Downstream_WR;UCF_BW_Local_CPU;UCF_BW_Local= _GMEM;UCF_BW_Local_CPU_To_Local_GMEM", + "MetricExpr": "UCF_MEM_BYTES_WR@src_loc_cpu=3D1,dst_loc_gmem=3D1@ = / 1e6 / duration_time", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average SLC-to-memory write bandwidth for loc= al GMEM, for traffic initiated by local CPUs.", + "PublicDescription": "Average SLC-to-memory write bandwidth for lo= cal GMEM, for traffic initiated by local CPUs. Includes write-backs, attrib= uted as described for UCF_SLC_WB. Data-bearing snoop responses to evictions= by a home SLC in another socket or a connected GPU's GL2 are not counted (= see UCF_SLC_WB).", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_slc_rd_bandwidth_local_cpu_to_local_cxlmem", + "MetricGroup": "UCF_BW_Upstream_RD;UCF_BW_Local_CPU;UCF_BW_Local_C= XLMEM;UCF_BW_Local_CPU_To_Local_CXLMEM", + "MetricExpr": "UCF_SLC_BYTES_RD@src_loc_cpu=3D1,dst_loc_other=3D1@= / 1e6 / duration_time", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average requester-to-SLC read bandwidth for l= ocal CPU requests to local CXL memory addresses.", + "PublicDescription": "Average requester-to-SLC read bandwidth of r= equests from local CPUs for local CXL memory addresses.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_mem_rd_bandwidth_local_cpu_to_local_cxlmem", + "MetricGroup": "UCF_BW_Downstream_RD;UCF_BW_Local_CPU;UCF_BW_Local= _CXLMEM;UCF_BW_Local_CPU_To_Local_CXLMEM", + "MetricExpr": "UCF_MEM_BYTES_RD@src_loc_cpu=3D1,dst_loc_other=3D1@= / 1e6 / duration_time", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average SLC-to-memory read bandwidth for loca= l CXL memory, for requests initiated by local CPUs.", + "PublicDescription": "Average SLC-to-memory read bandwidth for loc= al CXL memory, for requests initiated by local CPUs.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_slc_wr_bandwidth_local_cpu_to_local_cxlmem", + "MetricGroup": "UCF_BW_Upstream_WR;UCF_BW_Local_CPU;UCF_BW_Local_C= XLMEM;UCF_BW_Local_CPU_To_Local_CXLMEM", + "MetricExpr": "UCF_SLC_BYTES_WR@src_loc_cpu=3D1,dst_loc_other=3D1@= / 1e6 / duration_time", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average requester-to-SLC write bandwidth for = local CPU requests to local CXL memory addresses.", + "PublicDescription": "Average requester-to-SLC write bandwidth of = requests from local CPUs for local CXL memory addresses.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_mem_wr_bandwidth_local_cpu_to_local_cxlmem", + "MetricGroup": "UCF_BW_Downstream_WR;UCF_BW_Local_CPU;UCF_BW_Local= _CXLMEM;UCF_BW_Local_CPU_To_Local_CXLMEM", + "MetricExpr": "UCF_MEM_BYTES_WR@src_loc_cpu=3D1,dst_loc_other=3D1@= / 1e6 / duration_time", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average SLC-to-memory write bandwidth for loc= al CXL memory, for traffic initiated by local CPUs.", + "PublicDescription": "Average SLC-to-memory write bandwidth for lo= cal CXL memory, for traffic initiated by local CPUs. Includes write-backs, = attributed as described for UCF_SLC_WB.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_slc_rd_bandwidth_local_cpu_to_remote_mem", + "MetricGroup": "UCF_BW_Upstream_RD;UCF_BW_Local_CPU;UCF_BW_Remote_= Memory;UCF_BW_Local_CPU_To_Remote_Memory", + "MetricExpr": "UCF_SLC_BYTES_RD@src_loc_cpu=3D1,dst_rem=3D1@ / 1e6= / duration_time", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average requester-to-SLC read bandwidth for l= ocal CPU requests to remote memory addresses.", + "PublicDescription": "Average requester-to-SLC read bandwidth of r= equests from local CPUs for remote memory addresses.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_mem_rd_bandwidth_local_cpu_to_remote_mem", + "MetricGroup": "UCF_BW_Downstream_RD;UCF_BW_Local_CPU;UCF_BW_Remot= e_Memory;UCF_BW_Local_CPU_To_Remote_Memory", + "MetricExpr": "UCF_MEM_BYTES_RD@src_loc_cpu=3D1,dst_rem=3D1@ / 1e6= / duration_time", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average SLC-to-memory read bandwidth for remo= te memory, for requests initiated by local CPUs.", + "PublicDescription": "Average SLC-to-memory read bandwidth for rem= ote memory, for requests initiated by local CPUs.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_slc_wr_bandwidth_local_cpu_to_remote_mem", + "MetricGroup": "UCF_BW_Upstream_WR;UCF_BW_Local_CPU;UCF_BW_Remote_= Memory;UCF_BW_Local_CPU_To_Remote_Memory", + "MetricExpr": "UCF_SLC_BYTES_WR@src_loc_cpu=3D1,dst_rem=3D1@ / 1e6= / duration_time", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average requester-to-SLC write bandwidth for = local CPU requests to remote memory addresses.", + "PublicDescription": "Average requester-to-SLC write bandwidth of = requests from local CPUs for remote memory addresses.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_mem_wr_bandwidth_local_cpu_to_remote_mem", + "MetricGroup": "UCF_BW_Downstream_WR;UCF_BW_Local_CPU;UCF_BW_Remot= e_Memory;UCF_BW_Local_CPU_To_Remote_Memory", + "MetricExpr": "UCF_MEM_BYTES_WR@src_loc_cpu=3D1,dst_rem=3D1@ / 1e6= / duration_time", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average SLC-to-memory write bandwidth for rem= ote memory, for traffic initiated by local CPUs.", + "PublicDescription": "Average SLC-to-memory write bandwidth for re= mote memory, for traffic initiated by local CPUs. Includes write-backs, att= ributed as described for UCF_SLC_WB. Data-bearing snoop responses to evicti= ons by a home SLC in another socket or a connected GPU's GL2 are not counte= d (see UCF_SLC_WB).", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_slc_rd_bandwidth_local_noncpu_to_local_cmem= ", + "MetricGroup": "UCF_BW_Upstream_RD;UCF_BW_Local_NonCPU;UCF_BW_Loca= l_CMEM;UCF_BW_Local_NonCPU_To_Local_CMEM", + "MetricExpr": "UCF_SLC_BYTES_RD@src_loc_noncpu=3D1,dst_loc_cmem=3D= 1@ / 1e6 / duration_time", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average requester-to-SLC read bandwidth for l= ocal non-CPU requests to local CMEM addresses.", + "PublicDescription": "Average requester-to-SLC read bandwidth of r= equests from local non-CPU devices such as the GPU and PCIe devices for loc= al CMEM addresses.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_mem_rd_bandwidth_local_noncpu_to_local_cmem= ", + "MetricGroup": "UCF_BW_Downstream_RD;UCF_BW_Local_NonCPU;UCF_BW_Lo= cal_CMEM;UCF_BW_Local_NonCPU_To_Local_CMEM", + "MetricExpr": "UCF_MEM_BYTES_RD@src_loc_noncpu=3D1,dst_loc_cmem=3D= 1@ / 1e6 / duration_time", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average SLC-to-memory read bandwidth for loca= l CMEM, for requests initiated by local non-CPU devices.", + "PublicDescription": "Average SLC-to-memory read bandwidth for loc= al CMEM, for requests initiated by local non-CPU devices such as the GPU an= d PCIe devices.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_slc_wr_bandwidth_local_noncpu_to_local_cmem= ", + "MetricGroup": "UCF_BW_Upstream_WR;UCF_BW_Local_NonCPU;UCF_BW_Loca= l_CMEM;UCF_BW_Local_NonCPU_To_Local_CMEM", + "MetricExpr": "UCF_SLC_BYTES_WR@src_loc_noncpu=3D1,dst_loc_cmem=3D= 1@ / 1e6 / duration_time", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average requester-to-SLC write bandwidth for = local non-CPU requests to local CMEM addresses.", + "PublicDescription": "Average requester-to-SLC write bandwidth of = requests from local non-CPU devices such as the GPU and PCIe devices for lo= cal CMEM addresses.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_mem_wr_bandwidth_local_noncpu_to_local_cmem= ", + "MetricGroup": "UCF_BW_Downstream_WR;UCF_BW_Local_NonCPU;UCF_BW_Lo= cal_CMEM;UCF_BW_Local_NonCPU_To_Local_CMEM", + "MetricExpr": "UCF_MEM_BYTES_WR@src_loc_noncpu=3D1,dst_loc_cmem=3D= 1@ / 1e6 / duration_time", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average SLC-to-memory write bandwidth for loc= al CMEM, for traffic initiated by local non-CPU devices.", + "PublicDescription": "Average SLC-to-memory write bandwidth for lo= cal CMEM, for traffic initiated by local non-CPU devices such as the GPU an= d PCIe devices. Includes write-backs, attributed as described for UCF_SLC_W= B.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_slc_rd_bandwidth_local_noncpu_to_local_gmem= ", + "MetricGroup": "UCF_BW_Upstream_RD;UCF_BW_Local_NonCPU;UCF_BW_Loca= l_GMEM;UCF_BW_Local_NonCPU_To_Local_GMEM", + "MetricExpr": "UCF_SLC_BYTES_RD@src_loc_noncpu=3D1,dst_loc_gmem=3D= 1@ / 1e6 / duration_time", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average requester-to-SLC read bandwidth for l= ocal non-CPU requests to local GMEM addresses.", + "PublicDescription": "Average requester-to-SLC read bandwidth of r= equests from local non-CPU devices such as the GPU and PCIe devices for loc= al GMEM addresses.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_mem_rd_bandwidth_local_noncpu_to_local_gmem= ", + "MetricGroup": "UCF_BW_Downstream_RD;UCF_BW_Local_NonCPU;UCF_BW_Lo= cal_GMEM;UCF_BW_Local_NonCPU_To_Local_GMEM", + "MetricExpr": "UCF_MEM_BYTES_RD@src_loc_noncpu=3D1,dst_loc_gmem=3D= 1@ / 1e6 / duration_time", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average SLC-to-memory read bandwidth for loca= l GMEM, for requests initiated by local non-CPU devices.", + "PublicDescription": "Average SLC-to-memory read bandwidth for loc= al GMEM, for requests initiated by local non-CPU devices such as the GPU an= d PCIe devices.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_slc_wr_bandwidth_local_noncpu_to_local_gmem= ", + "MetricGroup": "UCF_BW_Upstream_WR;UCF_BW_Local_NonCPU;UCF_BW_Loca= l_GMEM;UCF_BW_Local_NonCPU_To_Local_GMEM", + "MetricExpr": "UCF_SLC_BYTES_WR@src_loc_noncpu=3D1,dst_loc_gmem=3D= 1@ / 1e6 / duration_time", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average requester-to-SLC write bandwidth for = local non-CPU requests to local GMEM addresses.", + "PublicDescription": "Average requester-to-SLC write bandwidth of = requests from local non-CPU devices such as the GPU and PCIe devices for lo= cal GMEM addresses.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_mem_wr_bandwidth_local_noncpu_to_local_gmem= ", + "MetricGroup": "UCF_BW_Downstream_WR;UCF_BW_Local_NonCPU;UCF_BW_Lo= cal_GMEM;UCF_BW_Local_NonCPU_To_Local_GMEM", + "MetricExpr": "UCF_MEM_BYTES_WR@src_loc_noncpu=3D1,dst_loc_gmem=3D= 1@ / 1e6 / duration_time", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average SLC-to-memory write bandwidth for loc= al GMEM, for traffic initiated by local non-CPU devices.", + "PublicDescription": "Average SLC-to-memory write bandwidth for lo= cal GMEM, for traffic initiated by local non-CPU devices such as the GPU an= d PCIe devices. Includes write-backs, attributed as described for UCF_SLC_W= B. Data-bearing snoop responses to evictions by a home SLC in another socke= t or a connected GPU's GL2 are not counted (see UCF_SLC_WB).", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_slc_rd_bandwidth_local_noncpu_to_local_cxlm= em", + "MetricGroup": "UCF_BW_Upstream_RD;UCF_BW_Local_NonCPU;UCF_BW_Loca= l_CXLMEM;UCF_BW_Local_NonCPU_To_Local_CXLMEM", + "MetricExpr": "UCF_SLC_BYTES_RD@src_loc_noncpu=3D1,dst_loc_other= =3D1@ / 1e6 / duration_time", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average requester-to-SLC read bandwidth for l= ocal non-CPU requests to local CXL memory addresses.", + "PublicDescription": "Average requester-to-SLC read bandwidth of r= equests from local non-CPU devices such as the GPU and PCIe devices for loc= al CXL memory addresses.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_mem_rd_bandwidth_local_noncpu_to_local_cxlm= em", + "MetricGroup": "UCF_BW_Downstream_RD;UCF_BW_Local_NonCPU;UCF_BW_Lo= cal_CXLMEM;UCF_BW_Local_NonCPU_To_Local_CXLMEM", + "MetricExpr": "UCF_MEM_BYTES_RD@src_loc_noncpu=3D1,dst_loc_other= =3D1@ / 1e6 / duration_time", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average SLC-to-memory read bandwidth for loca= l CXL memory, for requests initiated by local non-CPU devices.", + "PublicDescription": "Average SLC-to-memory read bandwidth for loc= al CXL memory, for requests initiated by local non-CPU devices such as the = GPU and PCIe devices.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_slc_wr_bandwidth_local_noncpu_to_local_cxlm= em", + "MetricGroup": "UCF_BW_Upstream_WR;UCF_BW_Local_NonCPU;UCF_BW_Loca= l_CXLMEM;UCF_BW_Local_NonCPU_To_Local_CXLMEM", + "MetricExpr": "UCF_SLC_BYTES_WR@src_loc_noncpu=3D1,dst_loc_other= =3D1@ / 1e6 / duration_time", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average requester-to-SLC write bandwidth for = local non-CPU requests to local CXL memory addresses.", + "PublicDescription": "Average requester-to-SLC write bandwidth of = requests from local non-CPU devices such as the GPU and PCIe devices for lo= cal CXL memory addresses.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_mem_wr_bandwidth_local_noncpu_to_local_cxlm= em", + "MetricGroup": "UCF_BW_Downstream_WR;UCF_BW_Local_NonCPU;UCF_BW_Lo= cal_CXLMEM;UCF_BW_Local_NonCPU_To_Local_CXLMEM", + "MetricExpr": "UCF_MEM_BYTES_WR@src_loc_noncpu=3D1,dst_loc_other= =3D1@ / 1e6 / duration_time", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average SLC-to-memory write bandwidth for loc= al CXL memory, for traffic initiated by local non-CPU devices.", + "PublicDescription": "Average SLC-to-memory write bandwidth for lo= cal CXL memory, for traffic initiated by local non-CPU devices such as the = GPU and PCIe devices. Includes write-backs, attributed as described for UCF= _SLC_WB.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_slc_rd_bandwidth_local_noncpu_to_remote_mem= ", + "MetricGroup": "UCF_BW_Upstream_RD;UCF_BW_Local_NonCPU;UCF_BW_Remo= te_Memory;UCF_BW_Local_NonCPU_To_Remote_Memory", + "MetricExpr": "UCF_SLC_BYTES_RD@src_loc_noncpu=3D1,dst_rem=3D1@ / = 1e6 / duration_time", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average requester-to-SLC read bandwidth for l= ocal non-CPU requests to remote memory addresses.", + "PublicDescription": "Average requester-to-SLC read bandwidth of r= equests from local non-CPU devices such as the GPU and PCIe devices for rem= ote memory addresses.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_mem_rd_bandwidth_local_noncpu_to_remote_mem= ", + "MetricGroup": "UCF_BW_Downstream_RD;UCF_BW_Local_NonCPU;UCF_BW_Re= mote_Memory;UCF_BW_Local_NonCPU_To_Remote_Memory", + "MetricExpr": "UCF_MEM_BYTES_RD@src_loc_noncpu=3D1,dst_rem=3D1@ / = 1e6 / duration_time", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average SLC-to-memory read bandwidth for remo= te memory, for requests initiated by local non-CPU devices.", + "PublicDescription": "Average SLC-to-memory read bandwidth for rem= ote memory, for requests initiated by local non-CPU devices such as the GPU= and PCIe devices.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_slc_wr_bandwidth_local_noncpu_to_remote_mem= ", + "MetricGroup": "UCF_BW_Upstream_WR;UCF_BW_Local_NonCPU;UCF_BW_Remo= te_Memory;UCF_BW_Local_NonCPU_To_Remote_Memory", + "MetricExpr": "UCF_SLC_BYTES_WR@src_loc_noncpu=3D1,dst_rem=3D1@ / = 1e6 / duration_time", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average requester-to-SLC write bandwidth for = local non-CPU requests to remote memory addresses.", + "PublicDescription": "Average requester-to-SLC write bandwidth of = requests from local non-CPU devices such as the GPU and PCIe devices for re= mote memory addresses.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_mem_wr_bandwidth_local_noncpu_to_remote_mem= ", + "MetricGroup": "UCF_BW_Downstream_WR;UCF_BW_Local_NonCPU;UCF_BW_Re= mote_Memory;UCF_BW_Local_NonCPU_To_Remote_Memory", + "MetricExpr": "UCF_MEM_BYTES_WR@src_loc_noncpu=3D1,dst_rem=3D1@ / = 1e6 / duration_time", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average SLC-to-memory write bandwidth for rem= ote memory, for traffic initiated by local non-CPU devices.", + "PublicDescription": "Average SLC-to-memory write bandwidth for re= mote memory, for traffic initiated by local non-CPU devices such as the GPU= and PCIe devices. Includes write-backs, attributed as described for UCF_SL= C_WB. Data-bearing snoop responses to evictions by a home SLC in another so= cket or a connected GPU's GL2 are not counted (see UCF_SLC_WB).", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_slc_rd_bandwidth_remote_to_local_cmem", + "MetricGroup": "UCF_BW_Upstream_RD;UCF_BW_Remote_Source;UCF_BW_Loc= al_CMEM;UCF_BW_Remote_To_Local_CMEM", + "MetricExpr": "UCF_SLC_BYTES_RD@src_rem=3D1,dst_loc_cmem=3D1@ / 1e= 6 / duration_time", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average requester-to-SLC read bandwidth for r= emote-socket requests to local CMEM addresses.", + "PublicDescription": "Average requester-to-SLC read bandwidth of r= equests from a remote socket for local CMEM addresses.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_mem_rd_bandwidth_remote_to_local_cmem", + "MetricGroup": "UCF_BW_Downstream_RD;UCF_BW_Remote_Source;UCF_BW_L= ocal_CMEM;UCF_BW_Remote_To_Local_CMEM", + "MetricExpr": "UCF_MEM_BYTES_RD@src_rem=3D1,dst_loc_cmem=3D1@ / 1e= 6 / duration_time", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average SLC-to-memory read bandwidth for loca= l CMEM, for requests initiated by a remote socket.", + "PublicDescription": "Average SLC-to-memory read bandwidth for loc= al CMEM, for requests initiated by a remote socket.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_slc_wr_bandwidth_remote_to_local_cmem", + "MetricGroup": "UCF_BW_Upstream_WR;UCF_BW_Remote_Source;UCF_BW_Loc= al_CMEM;UCF_BW_Remote_To_Local_CMEM", + "MetricExpr": "UCF_SLC_BYTES_WR@src_rem=3D1,dst_loc_cmem=3D1@ / 1e= 6 / duration_time", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average requester-to-SLC write bandwidth for = remote-socket requests to local CMEM addresses.", + "PublicDescription": "Average requester-to-SLC write bandwidth of = requests from a remote socket for local CMEM addresses.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_mem_wr_bandwidth_remote_to_local_cmem", + "MetricGroup": "UCF_BW_Downstream_WR;UCF_BW_Remote_Source;UCF_BW_L= ocal_CMEM;UCF_BW_Remote_To_Local_CMEM", + "MetricExpr": "UCF_MEM_BYTES_WR@src_rem=3D1,dst_loc_cmem=3D1@ / 1e= 6 / duration_time", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average SLC-to-memory write bandwidth for loc= al CMEM, for traffic initiated by a remote socket.", + "PublicDescription": "Average SLC-to-memory write bandwidth for lo= cal CMEM, for traffic initiated by a remote socket. Includes write-backs, a= ttributed as described for UCF_SLC_WB.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_slc_rd_bandwidth_remote_to_local_gmem", + "MetricGroup": "UCF_BW_Upstream_RD;UCF_BW_Remote_Source;UCF_BW_Loc= al_GMEM;UCF_BW_Remote_To_Local_GMEM", + "MetricExpr": "UCF_SLC_BYTES_RD@src_rem=3D1,dst_loc_gmem=3D1@ / 1e= 6 / duration_time", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average requester-to-SLC read bandwidth for r= emote-socket requests to local GMEM addresses.", + "PublicDescription": "Average requester-to-SLC read bandwidth of r= equests from a remote socket for local GMEM addresses.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_mem_rd_bandwidth_remote_to_local_gmem", + "MetricGroup": "UCF_BW_Downstream_RD;UCF_BW_Remote_Source;UCF_BW_L= ocal_GMEM;UCF_BW_Remote_To_Local_GMEM", + "MetricExpr": "UCF_MEM_BYTES_RD@src_rem=3D1,dst_loc_gmem=3D1@ / 1e= 6 / duration_time", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average SLC-to-memory read bandwidth for loca= l GMEM, for requests initiated by a remote socket.", + "PublicDescription": "Average SLC-to-memory read bandwidth for loc= al GMEM, for requests initiated by a remote socket.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_slc_wr_bandwidth_remote_to_local_gmem", + "MetricGroup": "UCF_BW_Upstream_WR;UCF_BW_Remote_Source;UCF_BW_Loc= al_GMEM;UCF_BW_Remote_To_Local_GMEM", + "MetricExpr": "UCF_SLC_BYTES_WR@src_rem=3D1,dst_loc_gmem=3D1@ / 1e= 6 / duration_time", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average requester-to-SLC write bandwidth for = remote-socket requests to local GMEM addresses.", + "PublicDescription": "Average requester-to-SLC write bandwidth of = requests from a remote socket for local GMEM addresses.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_mem_wr_bandwidth_remote_to_local_gmem", + "MetricGroup": "UCF_BW_Downstream_WR;UCF_BW_Remote_Source;UCF_BW_L= ocal_GMEM;UCF_BW_Remote_To_Local_GMEM", + "MetricExpr": "UCF_MEM_BYTES_WR@src_rem=3D1,dst_loc_gmem=3D1@ / 1e= 6 / duration_time", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average SLC-to-memory write bandwidth for loc= al GMEM, for traffic initiated by a remote socket.", + "PublicDescription": "Average SLC-to-memory write bandwidth for lo= cal GMEM, for traffic initiated by a remote socket. Includes write-backs, a= ttributed as described for UCF_SLC_WB. Data-bearing snoop responses to evic= tions by a home SLC in another socket or a connected GPU's GL2 are not coun= ted (see UCF_SLC_WB).", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_slc_rd_bandwidth_remote_to_local_cxlmem", + "MetricGroup": "UCF_BW_Upstream_RD;UCF_BW_Remote_Source;UCF_BW_Loc= al_CXLMEM;UCF_BW_Remote_To_Local_CXLMEM", + "MetricExpr": "UCF_SLC_BYTES_RD@src_rem=3D1,dst_loc_other=3D1@ / 1= e6 / duration_time", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average requester-to-SLC read bandwidth for r= emote-socket requests to local CXL memory addresses.", + "PublicDescription": "Average requester-to-SLC read bandwidth of r= equests from a remote socket for local CXL memory addresses.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_mem_rd_bandwidth_remote_to_local_cxlmem", + "MetricGroup": "UCF_BW_Downstream_RD;UCF_BW_Remote_Source;UCF_BW_L= ocal_CXLMEM;UCF_BW_Remote_To_Local_CXLMEM", + "MetricExpr": "UCF_MEM_BYTES_RD@src_rem=3D1,dst_loc_other=3D1@ / 1= e6 / duration_time", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average SLC-to-memory read bandwidth for loca= l CXL memory, for requests initiated by a remote socket.", + "PublicDescription": "Average SLC-to-memory read bandwidth for loc= al CXL memory, for requests initiated by a remote socket.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_slc_wr_bandwidth_remote_to_local_cxlmem", + "MetricGroup": "UCF_BW_Upstream_WR;UCF_BW_Remote_Source;UCF_BW_Loc= al_CXLMEM;UCF_BW_Remote_To_Local_CXLMEM", + "MetricExpr": "UCF_SLC_BYTES_WR@src_rem=3D1,dst_loc_other=3D1@ / 1= e6 / duration_time", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average requester-to-SLC write bandwidth for = remote-socket requests to local CXL memory addresses.", + "PublicDescription": "Average requester-to-SLC write bandwidth of = requests from a remote socket for local CXL memory addresses.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_mem_wr_bandwidth_remote_to_local_cxlmem", + "MetricGroup": "UCF_BW_Downstream_WR;UCF_BW_Remote_Source;UCF_BW_L= ocal_CXLMEM;UCF_BW_Remote_To_Local_CXLMEM", + "MetricExpr": "UCF_MEM_BYTES_WR@src_rem=3D1,dst_loc_other=3D1@ / 1= e6 / duration_time", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average SLC-to-memory write bandwidth for loc= al CXL memory, for traffic initiated by a remote socket.", + "PublicDescription": "Average SLC-to-memory write bandwidth for lo= cal CXL memory, for traffic initiated by a remote socket. Includes write-ba= cks, attributed as described for UCF_SLC_WB.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_slc_rd_bandwidth_remote_to_remote_mem", + "MetricGroup": "UCF_BW_Upstream_RD;UCF_BW_Remote_Source;UCF_BW_Rem= ote_Memory;UCF_BW_Remote_To_Remote_Memory", + "MetricExpr": "UCF_SLC_BYTES_RD@src_rem=3D1,dst_rem=3D1@ / 1e6 / d= uration_time", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average requester-to-SLC read bandwidth for r= emote-socket requests to remote memory addresses.", + "PublicDescription": "Average requester-to-SLC read bandwidth of r= equests from a remote socket for remote memory addresses.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_mem_rd_bandwidth_remote_to_remote_mem", + "MetricGroup": "UCF_BW_Downstream_RD;UCF_BW_Remote_Source;UCF_BW_R= emote_Memory;UCF_BW_Remote_To_Remote_Memory", + "MetricExpr": "UCF_MEM_BYTES_RD@src_rem=3D1,dst_rem=3D1@ / 1e6 / d= uration_time", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average SLC-to-memory read bandwidth for remo= te memory, for requests initiated by a remote socket.", + "PublicDescription": "Average SLC-to-memory read bandwidth for rem= ote memory, for requests initiated by a remote socket.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_slc_wr_bandwidth_remote_to_remote_mem", + "MetricGroup": "UCF_BW_Upstream_WR;UCF_BW_Remote_Source;UCF_BW_Rem= ote_Memory;UCF_BW_Remote_To_Remote_Memory", + "MetricExpr": "UCF_SLC_BYTES_WR@src_rem=3D1,dst_rem=3D1@ / 1e6 / d= uration_time", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average requester-to-SLC write bandwidth for = remote-socket requests to remote memory addresses.", + "PublicDescription": "Average requester-to-SLC write bandwidth of = requests from a remote socket for remote memory addresses.", + "Unit": "nvidia_ucf_pmu_" + }, + { + "MetricName": "ucf_avg_mem_wr_bandwidth_remote_to_remote_mem", + "MetricGroup": "UCF_BW_Downstream_WR;UCF_BW_Remote_Source;UCF_BW_R= emote_Memory;UCF_BW_Remote_To_Remote_Memory", + "MetricExpr": "UCF_MEM_BYTES_WR@src_rem=3D1,dst_rem=3D1@ / 1e6 / d= uration_time", + "ScaleUnit": "1MB/s", + "BriefDescription": "Average SLC-to-memory write bandwidth for rem= ote memory, for traffic initiated by a remote socket.", + "PublicDescription": "Average SLC-to-memory write bandwidth for re= mote memory, for traffic initiated by a remote socket. Includes write-backs= , attributed as described for UCF_SLC_WB. Data-bearing snoop responses to e= victions by a home SLC in another socket or a connected GPU's GL2 are not c= ounted (see UCF_SLC_WB).", + "Unit": "nvidia_ucf_pmu_" + } +] --=20 2.43.0