From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from SA9PR02CU001.outbound.protection.outlook.com (mail-southcentralusazon11013006.outbound.protection.outlook.com [40.93.196.6]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 77BD9388E68; Wed, 22 Jul 2026 02:13:40 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=fail smtp.client-ip=40.93.196.6 ARC-Seal:i=2; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1784686422; cv=fail; b=bQyScu61UbFNAqLevmGMxOXc1OE/Bm36Iuop/Z6naG1QOyxkwzStePpGVA64vU6mwEAbP2nXqdR91yGvVX0UC1kU5bWWaHTAp0rVdm6bV/2+VPX782Hx6TLxQCH9nzN6d8UhwR+UWzwp3xbuAAkmxNSQks7jJc/6yTuHabcj55I= ARC-Message-Signature:i=2; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1784686422; c=relaxed/simple; bh=PVPJGqbAeH60CiLxG/vegQpow0sXbcGFg4yJbq/dXBs=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: Content-Type:MIME-Version; b=qFUCGJh2zhi2gxmA4v7S1E8ohyRb36ihNHBJjJhdWz3RV2UPX0O9H3Ardj45MpOGgrvzmsBMdASR+6Kwo5r0ufgIC2zSJfBFWeCOGyohY3DLHsiT428m9SYe2IvkD1n6lHJuLp6XDOa0Tz6M6kWtotg0M0yZ/bkUseHrsYEn9mM= ARC-Authentication-Results:i=2; smtp.subspace.kernel.org; dmarc=pass (p=reject dis=none) header.from=nvidia.com; spf=fail smtp.mailfrom=nvidia.com; dkim=pass (2048-bit key) header.d=Nvidia.com header.i=@Nvidia.com header.b=UlKYz/lj; arc=fail smtp.client-ip=40.93.196.6 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) header.d=Nvidia.com header.i=@Nvidia.com header.b="UlKYz/lj" ARC-Seal: i=1; a=rsa-sha256; s=arcselector10001; d=microsoft.com; cv=none; b=Dxr4bZFbcLRmsx6yFWhUMmUS0PDJ9UN1W1nxbemxRVMMh+fkDoFLnV55FrMZHHI7X2QQFk36PVEEnAmcFm4Fqb8wdjbxIGHtVKexSEQ2Y5alp6vPkJeHsH2szDzPFv0KAIRc+JSrRl60W73HrYevLujuHmReaHealCuaFGe+WXydnW/aSLPX+ny7f+UoUdXcRX0MyRwIp15JW5pEal/nzLFzHVWkQ9ousr4EsLRtppFVVLtSYVdG3vb3OlWrr/LwSjULqqYohavc26vapgwtINvEzOz5mBnbJqP5CZdI2899nbPphki8FTwq+7LBGN/fGEO5XLAI7ghb/Tc581dg0w== 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=oJZLGHcDNEvLcg0fyvZny0rcCFqacEclqhQuhC+wFJk=; b=yf+oLNN6xQDE+PawSoAsKC+gI/n8bKQGjThLP0ta9Idnu6blMar88Y0v+ARsACu75hRPS+ZYM0Ekj3F6CmX7XFBdWYHs6rOinLsVznUCRTTFyk1taGsJuLzcFrUPVCgFz/Ymh1ZHVDTvZGln1VR0QsBbIXHaF2fSMwhQNOC4JqXzlpq5w1EH78yC5oqmTalRkccJtOt74sj5/afmIGzZsd+oijteFe2SgEqYwLaNCFUdzKu6FTlz0pNV3RBxhkqIn7IzpLvae/TCqezI8C6ct3EVDfz1GrmdI8+6ycn9z4mA1KzEELKttrwwXY0DtBdyuHnklXRxqe6cyoGVjOnyDg== ARC-Authentication-Results: i=1; mx.microsoft.com 1; spf=pass smtp.mailfrom=nvidia.com; dmarc=pass action=none header.from=nvidia.com; dkim=pass header.d=nvidia.com; arc=none 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=oJZLGHcDNEvLcg0fyvZny0rcCFqacEclqhQuhC+wFJk=; b=UlKYz/ljMpMmaz3LKrR5qAM1jwKn0a9l46NkLFpCIZE33i6VtC2/eRktlwWuahhOUow8YUKh5e4zeuJUn81/F/moWTHL4EzmnlfHojDtvZmgAZQ0wWXN7vDxfdDzZ6CVxn06HPhkXNPrdb+x1zYqHwKJZ/mEJqSVByJLCJyFneldkDY24/RINvjCwQw47EKc08VnMxG4fbmiCpr75jT3YldMORQ6SE+H8+JbGpZay6Rfga0dUQxDEUlG0vbe6ZD20IXChQPOwr0428wWsZiOZC3gaFEjfo18s+R+DMIiQ3lMNKA2YH9BXTMUtGLwHhh3VEW6X8cpacugILYIgTxDPQ== Authentication-Results: dkim=none (message not signed) header.d=none;dmarc=none action=none header.from=nvidia.com; Received: from BL0PR12MB2370.namprd12.prod.outlook.com (2603:10b6:207:47::27) by SA0PR12MB7002.namprd12.prod.outlook.com (2603:10b6:806:2c0::18) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.21.245.10; Wed, 22 Jul 2026 02:13:34 +0000 Received: from BL0PR12MB2370.namprd12.prod.outlook.com ([fe80::86cf:c3ec:2cf5:74c8]) by BL0PR12MB2370.namprd12.prod.outlook.com ([fe80::86cf:c3ec:2cf5:74c8%5]) with mapi id 15.21.0245.009; Wed, 22 Jul 2026 02:13:34 +0000 From: Richard Cheng To: tony.luck@intel.com, reinette.chatre@intel.com, x86@kernel.org Cc: Dave.Martin@arm.com, james.morse@arm.com, babu.moger@amd.com, shuah@kernel.org, linux-kernel@vger.kernel.org, linux-kselftest@vger.kernel.org, newtonl@nvidia.com, kristinc@nvidia.com, kobak@nvidia.com, kaihengf@nvidia.com, fenghuay@nvidia.com, ltrager@nvidia.com, Richard Cheng Subject: [PATCH 3/5] selftests/resctrl: Support the MPAM memory bandwidth monitoring layout Date: Wed, 22 Jul 2026 10:13:01 +0800 Message-ID: <20260722021303.9471-4-icheng@nvidia.com> X-Mailer: git-send-email 2.50.1 In-Reply-To: <20260722021303.9471-1-icheng@nvidia.com> References: <20260722021303.9471-1-icheng@nvidia.com> Content-Transfer-Encoding: 8bit Content-Type: text/plain X-ClientProxiedBy: TP0P295CA0036.TWNP295.PROD.OUTLOOK.COM (2603:1096:910:4::15) To BL0PR12MB2370.namprd12.prod.outlook.com (2603:10b6:207:47::27) Precedence: bulk X-Mailing-List: linux-kselftest@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 X-MS-PublicTrafficType: Email X-MS-TrafficTypeDiagnostic: BL0PR12MB2370:EE_|SA0PR12MB7002:EE_ X-MS-Office365-Filtering-Correlation-Id: e8c092c1-e5b4-4b6c-208b-08dee796d4f0 X-MS-Exchange-SenderADCheck: 1 X-MS-Exchange-AntiSpam-Relay: 0 X-Microsoft-Antispam: BCL:0;ARA:13230040|366016|23010399003|376014|1800799024|6133799003|5023799004|56012099006|11063799006|10067099003|3023799007|22082099003|18002099003; X-Microsoft-Antispam-Message-Info: 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 X-Forefront-Antispam-Report: CIP:255.255.255.255;CTRY:;LANG:en;SCL:1;SRV:;IPV:NLI;SFV:NSPM;H:BL0PR12MB2370.namprd12.prod.outlook.com;PTR:;CAT:NONE;SFS:(13230040)(366016)(23010399003)(376014)(1800799024)(6133799003)(5023799004)(56012099006)(11063799006)(10067099003)(3023799007)(22082099003)(18002099003);DIR:OUT;SFP:1101; X-MS-Exchange-AntiSpam-MessageData-ChunkCount: 1 X-MS-Exchange-AntiSpam-MessageData-0: =?us-ascii?Q?Y1AqHgiBOuvgyFNg/7rrMPIPw5WKhAqja8wYqOwHw4CKUCwhTYhWofZNUzpx?= =?us-ascii?Q?kpZBDKMMWO1BLV1VxhXoujCrf8QarhJ1dQWCXfotYLWTWF/8VCKCtuwtmF7J?= =?us-ascii?Q?NJNpC4eB90VPu6LYTMeWJOGAVFj5zkK/dKhzPIT0jCnwVRH0Zdsn4CZOXX0e?= =?us-ascii?Q?zNSmOuA23Njw3pzjnNl3clxaHTugeBoUdNR2Nb1lm8Ispf+F7zTTvClDwk/C?= =?us-ascii?Q?HxOogmeYP9YEd6LlsxaFjCww86o0d95oxKs79Bao5PPpzEfBMkU8RunInJWr?= =?us-ascii?Q?C/1WOBhCOqFmZiLPNt532/w3cWbbkSPpunSD2ZehiiYVW1X651P5XNCnhZGW?= =?us-ascii?Q?aEpLoKNWJfYpakZGCw8FESVnGn9D4Rb1lJR0sZftdYoz59xhuRX4nK29Xxev?= =?us-ascii?Q?SbMHtWKTsJu/OrcuY39oDv1WoOYB3q3nDkZDxI6V0DGW7xJvfcGXFBBIqS33?= =?us-ascii?Q?JM+lPDUj4my5I2o8QJ3D5eUsjW+JNITX0HjibDaVDsUfbrfYT2mdghDuOLUs?= =?us-ascii?Q?CFo/GsSLgiylWJ7G/CwrMagHIulo6y3CcMjhoX09VOMri+7nB6fM/1BCK6tq?= =?us-ascii?Q?63jOLCOL6soH77tSAIfDzuvvs0J0161y5JOjZeTw0N5G2wBBCfJZ9Ywf5CRC?= =?us-ascii?Q?zKaoqqN2Yu+XIFRq6QQz56EHyDdy9+xFCBT4q8Ny3c7uMztYQE+Pjt4Tpw7n?= =?us-ascii?Q?5+clsDIQeYtyxMjpApCKNYxUAKZ9GniYud7XU2kl0kX5dxZMsQFO5nAWqsjN?= =?us-ascii?Q?1+jLoV64lkwOESyfIOYWoJyqXWvlLnQvfgrKJ4xoBwlkBBp/hW32Wv58g5cs?= =?us-ascii?Q?HUzIDcqiwccTsDOGHJx75okk4noW5IoQ104prRO0C30iYb2fLvB0VD1C/k38?= =?us-ascii?Q?vqgWdAM1FsF+hDRwrV9vt44yNXb42NlXrCfBBoki92kDAK7tG49grDXBpI+F?= =?us-ascii?Q?tXAzY956Vx08QnnbkZiznzqKgjhLyKsqRLsmEw1acUUSw3fU3XkeJEQHjP8E?= =?us-ascii?Q?qyhx/K2OVIq3s2vW46Yb5LrBmp5v8pkQJAw1IrIUC3+lzr1F7wBEYXxSNQV3?= =?us-ascii?Q?y45SM++p1NcSV562jbwj8OZkgiCtNRp5Lcxyx0hbWzwkIYEL9xYoBheN8e8i?= =?us-ascii?Q?gDd951qx7Bm7ZX9NLN7VA1zqMaZ5hy2ACvn4+XUtTP/ZpBTC72Tm/GjZM1Fe?= =?us-ascii?Q?sw3p3BPId/t8bgEG541RPYefkqAUkv5xzDWtqTgIE1bAcwypDnsofJuia6/W?= =?us-ascii?Q?UhRNFvX59IfQxz89m1yTwovzzzg5TwyyjgMH6itZ5a39AzB8GPgXYkumBCaV?= =?us-ascii?Q?W9Jc0hin9E6CXQAR+h0YBpiGap8RWUvRX4LbqypAB5OuBrtj97uhChOYu0li?= =?us-ascii?Q?yXspoxnicoQ2ZyOHsNLFDfkWeZF19Hzw660MrHvzcJo+pNqc3WCmdMlu1GiO?= =?us-ascii?Q?9pomwz6STCB8mqrf1yvDjM3R8ZvmUJQ/AdKRgAZJnL8LEZ15W79mHfrBrJIL?= =?us-ascii?Q?JEmN42G1LZOpSR1OP2Thu/rrbDR7JQEdUTjZQgd8AEUTkJByDNIURw4jb7aQ?= =?us-ascii?Q?sKHNeT+zRIZGJ8l6J7zlkTGftAnf9E99KEj7DjlNILKp3C/pTcZsSZJuyus3?= =?us-ascii?Q?u7vglUZ8dv4srbIvQPqHRM//z4lQcmlvg1BIsOyRpltJcRIcnurEcsI5WO/h?= =?us-ascii?Q?NAMq0a/Mdf7xvhmP2JwnMGCHWYl8k83soi/kPxcZhvvEQt4RU7HHlVjZSgr4?= =?us-ascii?Q?UibTF+dvkg=3D=3D?= X-OriginatorOrg: Nvidia.com X-MS-Exchange-CrossTenant-Network-Message-Id: e8c092c1-e5b4-4b6c-208b-08dee796d4f0 X-MS-Exchange-CrossTenant-AuthSource: BL0PR12MB2370.namprd12.prod.outlook.com X-MS-Exchange-CrossTenant-AuthAs: Internal X-MS-Exchange-CrossTenant-OriginalArrivalTime: 22 Jul 2026 02:13:34.2721 (UTC) X-MS-Exchange-CrossTenant-FromEntityHeader: Hosted X-MS-Exchange-CrossTenant-Id: 43083d15-7273-40c1-b7db-39efd9ccc17a X-MS-Exchange-CrossTenant-MailboxType: HOSTED X-MS-Exchange-CrossTenant-UserPrincipalName: dzSB3ulAiCDqXQawpR55nMGwZKQQ4KhooEsSbExnTS1Bi+ricu3wlNdJOkEb182Ab40QDcndw4y9vr+0ksw/JQ== X-MS-Exchange-Transport-CrossTenantHeadersStamped: SA0PR12MB7002 Arm MPAM systems running resctrl expose memory bandwidth differently from x86: - MB control/monitor domains are keyed by NUMA node id, which need not match the L3 cache id (on a Grace/Vera system L3 ids are {1,2} while MB domains are {0,1,2,10,...}). - Monitoring lives in a separate MB_MON resource that provides only mbm_total_bytes; there is no local/total split and no L3_MON bandwidth event. - In "mbm_event" (ABMC) assignment mode a hardware counter must be explicitly assigned to a group+domain+event before the bandwidth file returns a value (it reads "Unassigned" otherwise), and removing a group does not release its counter, so it must be unassigned explicitly or it leaks. Teach the test framework this layout: - get_domain_id() resolves MB domains via the CPU's NUMA node when the MB_MON resource exists. x86 (no MB_MON) is unaffected. - initialize_mem_bw_resctrl() falls back to mon_MB_/mbm_total_bytes when mbm_local_bytes is absent, and assigns an ABMC counter when the group's mbm_MB_assignments file is present. Note the assignment parser requires a trailing newline; without it the write fails with EINVAL. - mem_bw_ref_cleanup() unassigns the counter again. - The MBA feature check accepts MB_MON's mbm_total_bytes as the monitoring source. Comparing a reference PMU total (read+write) with mbm_total_bytes is a like-for-like comparison; the iMC path keeps comparing read bandwidth with mbm_local_bytes as before. No effect yet on any vendor: the MBA test is still gated to Intel. Signed-off-by: Richard Cheng --- tools/testing/selftests/resctrl/mba_test.c | 3 +- tools/testing/selftests/resctrl/resctrl_val.c | 69 +++++++++++++++++-- tools/testing/selftests/resctrl/resctrlfs.c | 45 +++++++++++- 3 files changed, 111 insertions(+), 6 deletions(-) diff --git a/tools/testing/selftests/resctrl/mba_test.c b/tools/testing/selftests/resctrl/mba_test.c index 8b891d3e6650..6c85107c32ee 100644 --- a/tools/testing/selftests/resctrl/mba_test.c +++ b/tools/testing/selftests/resctrl/mba_test.c @@ -211,7 +211,8 @@ static int mba_run_test(const struct resctrl_test *test, const struct user_param static bool mba_feature_check(const struct resctrl_test *test) { return test_resource_feature_check(test) && - resctrl_mon_feature_exists("L3_MON", "mbm_local_bytes") && + (resctrl_mon_feature_exists("L3_MON", "mbm_local_bytes") || + resctrl_mon_feature_exists("MB_MON", "mbm_total_bytes")) && mem_bw_ref_available(); } diff --git a/tools/testing/selftests/resctrl/resctrl_val.c b/tools/testing/selftests/resctrl/resctrl_val.c index a2b919bd4839..0c3079d31340 100644 --- a/tools/testing/selftests/resctrl/resctrl_val.c +++ b/tools/testing/selftests/resctrl/resctrl_val.c @@ -20,6 +20,21 @@ #define CON_MBM_LOCAL_BYTES_PATH \ "%s/%s/mon_data/mon_L3_%02d/mbm_local_bytes" +/* + * MPAM exposes only total (read+write) bandwidth, under MB_MON, and per the + * NUMA-node-keyed MB domain rather than the L3 cache id. + */ +#define CON_MBM_TOTAL_BYTES_PATH \ + "%s/%s/mon_data/mon_MB_%02d/mbm_total_bytes" + +/* + * In MPAM "mbm_event" (ABMC) counter-assignment mode a hardware counter must + * be explicitly assigned to a group+domain+event before the bandwidth file + * reads a value; otherwise it returns "Unassigned". + */ +#define MBM_MB_ASSIGN_PATH \ + "%s/%s/mbm_MB_assignments" + struct membw_read_format { __u64 value; /* The value of the event */ __u64 time_enabled; /* if PERF_FORMAT_TOTAL_TIME_ENABLED */ @@ -58,6 +73,10 @@ static struct mem_bw_counter bw_counters[MAX_BW_COUNTERS]; static const struct resctrl_test *current_test; static const struct mem_bw_backend *mem_bw_backend; +/* MPAM ABMC counter assignment state, see initialize_mem_bw_resctrl(). */ +static char mbm_assign_path[1024]; +static int mbm_assign_domain = -1; + static void read_mem_bw_initialize_perf_event_attr(int i) { memset(&bw_counters[i].pe, 0, @@ -452,21 +471,63 @@ static int get_mem_bw_ref(float *bw_ref) * initialize_mem_bw_resctrl: Appropriately populate "mbm_total_path" * @param: Parameters passed to resctrl_val() * @domain_id: Domain ID (cache ID; for MB, L3 cache ID) + * + * x86 resctrl reports per-RMID local bandwidth under L3_MON. MPAM exposes only + * total (read+write) bandwidth under MB_MON, indexed by the NUMA-node-keyed MB + * domain, and in "mbm_event" (ABMC) mode a hardware counter must be assigned to + * the group+domain before the file reads a value. */ void initialize_mem_bw_resctrl(const struct resctrl_val_param *param, int domain_id) { - sprintf(mbm_total_path, CON_MBM_LOCAL_BYTES_PATH, RESCTRL_PATH, - param->ctrlgrp, domain_id); + const char *grp = param->ctrlgrp ? param->ctrlgrp : ""; + + if (resctrl_mon_feature_exists("L3_MON", "mbm_local_bytes")) { + sprintf(mbm_total_path, CON_MBM_LOCAL_BYTES_PATH, RESCTRL_PATH, + grp, domain_id); + return; + } + + /* MPAM: total bandwidth under MB_MON. */ + sprintf(mbm_total_path, CON_MBM_TOTAL_BYTES_PATH, RESCTRL_PATH, + grp, domain_id); + + /* Assign an ABMC counter for this group+domain if in mbm_event mode. */ + snprintf(mbm_assign_path, sizeof(mbm_assign_path), MBM_MB_ASSIGN_PATH, + RESCTRL_PATH, grp); + if (access(mbm_assign_path, W_OK) == 0) { + FILE *fp = fopen(mbm_assign_path, "w"); + + if (fp) { + /* The assignment parser requires a trailing newline. */ + fprintf(fp, "mbm_total_bytes:%d=e\n", domain_id); + fclose(fp); + mbm_assign_domain = domain_id; + } + } else { + /* Not in counter-assignment mode (or no MB_MON); nothing to do. */ + mbm_assign_path[0] = '\0'; + } } /* - * mem_bw_ref_cleanup - Reset the selected reference-bandwidth backend + * mem_bw_ref_cleanup - Release an assigned ABMC counter and reset backend state * - * Safe to call when no backend was selected. + * Group removal does not reclaim assigned MPAM counters, so the test must + * unassign explicitly. Safe to call when nothing was assigned. */ void mem_bw_ref_cleanup(void) { + if (mbm_assign_domain >= 0 && mbm_assign_path[0]) { + FILE *fp = fopen(mbm_assign_path, "w"); + + if (fp) { + fprintf(fp, "mbm_total_bytes:%d=_\n", mbm_assign_domain); + fclose(fp); + } + } + mbm_assign_domain = -1; + mbm_assign_path[0] = '\0'; mem_bw_backend = NULL; } diff --git a/tools/testing/selftests/resctrl/resctrlfs.c b/tools/testing/selftests/resctrl/resctrlfs.c index b9c1bfb6cc02..893f6748c68c 100644 --- a/tools/testing/selftests/resctrl/resctrlfs.c +++ b/tools/testing/selftests/resctrl/resctrlfs.c @@ -121,11 +121,40 @@ static int get_resource_cache_level(const char *resource) return get_cache_level(resource); } +/* + * cpu_to_numa_node - NUMA node a logical CPU belongs to + * @cpu_no: CPU number + * + * Reads the "node" symlink in the CPU's sysfs directory. + * + * Return: NUMA node id (>= 0) on success, < 0 on failure. + */ +static int cpu_to_numa_node(int cpu_no) +{ + char cpu_dir[1024]; + struct dirent *ep; + int node = -1; + DIR *dp; + + snprintf(cpu_dir, sizeof(cpu_dir), "%s%d", PHYS_ID_PATH, cpu_no); + dp = opendir(cpu_dir); + if (!dp) + return -1; + while ((ep = readdir(dp))) { + if (!strncmp(ep->d_name, "node", 4) && isdigit(ep->d_name[4])) { + node = atoi(ep->d_name + 4); + break; + } + } + closedir(dp); + return node; +} + /* * get_domain_id - Get resctrl domain ID for a specified CPU * @resource: resource name * @cpu_no: CPU number - * @domain_id: domain ID (cache ID; for MB, L3 cache ID) + * @domain_id: domain ID (cache ID; for MB, L3 cache ID, or NUMA node on MPAM) * * Return: >= 0 on success, < 0 on failure. */ @@ -135,6 +164,20 @@ int get_domain_id(const char *resource, int cpu_no, int *domain_id) int cache_num; FILE *fp; + /* + * On MPAM the MB monitoring/allocation domains are keyed by NUMA node + * (surfaced via a separate MB_MON group) and need not match the L3 cache + * id used on x86, so resolve the MB domain from the CPU's NUMA node. + */ + if (!strcmp(resource, "MB") && resctrl_resource_exists("MB_MON")) { + int node = cpu_to_numa_node(cpu_no); + + if (node >= 0) { + *domain_id = node; + return 0; + } + } + cache_num = get_resource_cache_level(resource); if (cache_num < 0) return cache_num; -- 2.43.0