From nobody Sat Feb 7 18:20:15 2026 Received: from mail-dy1-f201.google.com (mail-dy1-f201.google.com [74.125.82.201]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 4474122DA1C for ; Thu, 5 Feb 2026 23:22:31 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=74.125.82.201 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1770333751; cv=none; b=hKChWt0yZQVzuqPzvFKvBh4zchypA82nY4xupADdIe2dAGeEtWXCGjLR18xqfczWw+JAK3cNSjeiQWPTEOKe0eHHKA0XDkcMvTYTdpcGEbx2nxuhX8oirKD6gn/IQGcNkegHmtG1MVP3zVo+eMeHT5ect+BCKSt+4i3TY88RiG4= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1770333751; c=relaxed/simple; bh=2jVRN+9SAEgkWqRItLFggdWT/DICf75mh3jcx/EBOhM=; h=Date:Mime-Version:Message-ID:Subject:From:To:Cc:Content-Type; b=AKoRsEVbmcId9D6ic2jZtX1ydXaQQ6WoIw/qNskG3jmDuBPBzTEdZXbTbQKh3ZuPVK/Z3EBYpuw+WiEq9AgVzyex8qaqOQwFCluCSMVRLqvHHt1il0yob8idvbnOb34TsFHk+QtkzrToicyXDSPww499ScH9zI1dJHEtEjzgvu8= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=reject dis=none) header.from=google.com; spf=pass smtp.mailfrom=flex--ctshao.bounces.google.com; dkim=pass (2048-bit key) header.d=google.com header.i=@google.com header.b=CzH38+B5; arc=none smtp.client-ip=74.125.82.201 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=reject dis=none) header.from=google.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=flex--ctshao.bounces.google.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=google.com header.i=@google.com header.b="CzH38+B5" Received: by mail-dy1-f201.google.com with SMTP id 5a478bee46e88-2b6b9c1249fso205448eec.1 for ; Thu, 05 Feb 2026 15:22:31 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20230601; t=1770333750; x=1770938550; darn=vger.kernel.org; h=cc:to:from:subject:message-id:mime-version:date:from:to:cc:subject :date:message-id:reply-to; bh=6JRP+ONgL8h2DzadRmrt/AdG85INIGqNpNsrfSRjVYE=; b=CzH38+B58cTniR2YMQ7ShitKu/aYHGpXcJzVqTiUZvIEob5jQqpUrNqMJACNL2SPWL R9LozGZ8k2kqYRe1XfK+MVSeepFSu5ssYTyLHMffk3YOKkhaXoJAO3l50ZOhyITZrk0l 0uO7Q5pDd9xJEjucXF45KAR9Ipa6nDl+DRs0O/1ohIYpF1vhASrMe2iqHGt6hRW0wMTJ Uh5ST+dO1psRNKFAngyMh+Ap592GH+UnnGkX3oJD1NqPT4a7mjiQw8g8PGesaEGjOIDy QEzh6oHdAH4kU9X00ihmb26ntsZgyjby27OMNN+hfDPSVD40jHlUvmSIyyvh48y71voA 8aTA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1770333750; x=1770938550; h=cc:to:from:subject:message-id:mime-version:date:x-gm-message-state :from:to:cc:subject:date:message-id:reply-to; bh=6JRP+ONgL8h2DzadRmrt/AdG85INIGqNpNsrfSRjVYE=; b=nBdy5pTghaCFmyldTuQpxio6IyfG6ZYno+ojeKRG+/fYjV6pL21KAXOTCwBoGGmbcx cEOIrA1LjilmFWXQMbryB1AMwQQuVDC5yyj64pCtkUCUXz4lV3bAMmqX28JXarfNGJvV JSgBthvZUrZRspGaUngIazH7IfQBNGT97A2/brtJ612NlkFnPdCdzHmO/wDJHd798/S9 PJNwQUrqj/MNr+kdqv4N8FZFVY0torYxOsRV7K+CgEx1nhI8pQQ7zb+Ly9bBSCVLEfoT cBX5KpP+Mg3+jgOX47R2kqdL+KGnKwKmWhBTfJ/40VC1RkAbJIGfC5TtNtpwJpYjozOg ZD3g== X-Gm-Message-State: AOJu0Yy3F4AkPPsGPsr5jonWcmSM2USQ115LA56QZ56zqeINu57AYB+8 NA+9CHb24Ww2kcCwIsneS909KxkL7WTJtpjcWB2bDrw2HEa0ESWExp+79NfePpNzg8fMIXCF1Yu fnXQoXgik/cOubgwMM6kSXbhrph70BrpAMXvjX+xmHYHVV7GOj0EnRL2m+qYjxETs0k8cyt66Cl Ta+gFHTLr8PhjUEk9AoKby3TBpZJ8AkNn+2ONlTOigaTeT X-Received: from dyclx4.prod.google.com ([2002:a05:693c:2184:b0:2b7:4bcf:315a]) (user=ctshao job=prod-delivery.src-stubby-dispatcher) by 2002:a05:7301:1284:b0:2a4:3593:c7e1 with SMTP id 5a478bee46e88-2b85683fe9bmr316308eec.33.1770333750219; Thu, 05 Feb 2026 15:22:30 -0800 (PST) Date: Thu, 5 Feb 2026 15:21:55 -0800 Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: Mime-Version: 1.0 X-Mailer: git-send-email 2.53.0.rc2.204.g2597b5adb4-goog Message-ID: <20260205232220.1980168-1-ctshao@google.com> Subject: [PATCH v2] perf pmu intel: Adjust cpumaks for sub-NUMA clusters on Emeraldrapids From: Chun-Tse Shao To: linux-kernel@vger.kernel.org Cc: Chun-Tse Shao , peterz@infradead.org, mingo@redhat.com, acme@kernel.org, namhyung@kernel.org, mark.rutland@arm.com, alexander.shishkin@linux.intel.com, jolsa@kernel.org, irogers@google.com, adrian.hunter@intel.com, james.clark@linaro.org, ravi.bangoria@amd.com, linux-perf-users@vger.kernel.org Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" Similar to GNR [1], Emeraldrapids supports sub-NUMA clusters as well. Adjust cpumasks as the logic for GNR in [1]. Tested on Emeraldrapids with SNC2 enabled: $ perf stat --per-node -e 'UNC_CHA_CLOCKTICKS,UNC_M_CLOCKTICKS' -a -- sle= ep 1 Performance counter stats for 'system wide': N0 30 72125876670 UNC_CHA_CLOCKTICKS N0 4 8815163648 UNC_M_CLOCKTICKS N1 30 72124958844 UNC_CHA_CLOCKTICKS N1 4 8815014974 UNC_M_CLOCKTICKS N2 30 72121049022 UNC_CHA_CLOCKTICKS N2 4 8814592626 UNC_M_CLOCKTICKS N3 30 72117133854 UNC_CHA_CLOCKTICKS N3 4 8814012840 UNC_M_CLOCKTICKS 1.001574118 seconds time elapsed [1] lore.kernel.org/20250515181417.491401-1-irogers@google.com Signed-off-by: Chun-Tse Shao --- v2: Split EMR and GNR in the SNC2 IMC cpu map. v1: lore.kernel.org/20260108184430.1210223-1-ctshao@google.com tools/perf/arch/x86/util/pmu.c | 95 ++++++++++++++++++++++------------ 1 file changed, 63 insertions(+), 32 deletions(-) diff --git a/tools/perf/arch/x86/util/pmu.c b/tools/perf/arch/x86/util/pmu.c index a3f96221758d..dbe876cd1b08 100644 --- a/tools/perf/arch/x86/util/pmu.c +++ b/tools/perf/arch/x86/util/pmu.c @@ -22,20 +22,29 @@ #include "util/env.h" #include "util/header.h" -static bool x86__is_intel_graniterapids(void) +static bool x86__is_snc_supported(void) { - static bool checked_if_graniterapids; - static bool is_graniterapids; + static bool checked_if_snc_supported; + static bool is_supported; - if (!checked_if_graniterapids) { - const char *graniterapids_cpuid =3D "GenuineIntel-6-A[DE]"; + if (!checked_if_snc_supported) { + + /* Emeraldrapids and Graniterapids support SNC configuration. */ + static const char *const supported_cpuids[] =3D { + "GenuineIntel-6-CF", /* Emeraldrapids */ + "GenuineIntel-6-A[DE]", /* Graniterapids */ + }; char *cpuid =3D get_cpuid_str((struct perf_cpu){0}); - is_graniterapids =3D cpuid && strcmp_cpuid_str(graniterapids_cpuid, cpui= d) =3D=3D 0; + for (size_t i =3D 0; i < ARRAY_SIZE(supported_cpuids); i++) { + is_supported =3D cpuid && strcmp_cpuid_str(supported_cpuids[i], cpuid) = =3D=3D 0; + if (is_supported) + break; + } free(cpuid); - checked_if_graniterapids =3D true; + checked_if_snc_supported =3D true; } - return is_graniterapids; + return checked_if_snc_supported; } static struct perf_cpu_map *read_sysfs_cpu_map(const char *sysfs_path) @@ -64,6 +73,7 @@ static int snc_nodes_per_l3_cache(void) read_sysfs_cpu_map("devices/system/cpu/cpu0/cache/index3/shared_cpu_lis= t"); snc_nodes =3D perf_cpu_map__nr(cache_cpus) / perf_cpu_map__nr(node_cpus); + perf_cpu_map__put(cache_cpus); perf_cpu_map__put(node_cpus); checked_snc =3D true; @@ -137,23 +147,42 @@ static int uncore_imc_snc(struct perf_pmu *pmu) // Compute the IMC SNC using lookup tables. unsigned int imc_num; int snc_nodes =3D snc_nodes_per_l3_cache(); - const u8 snc2_map[] =3D {1, 1, 0, 0, 1, 1, 0, 0}; - const u8 snc3_map[] =3D {1, 1, 0, 0, 2, 2, 1, 1, 0, 0, 2, 2}; - const u8 *snc_map; - size_t snc_map_len; - - switch (snc_nodes) { - case 2: - snc_map =3D snc2_map; - snc_map_len =3D ARRAY_SIZE(snc2_map); - break; - case 3: - snc_map =3D snc3_map; - snc_map_len =3D ARRAY_SIZE(snc3_map); - break; - default: - /* Error or no lookup support for SNC with >3 nodes. */ - return 0; + char *cpuid; + static const u8 emr_snc2_map[] =3D { 0, 0, 1, 1 }; + static const u8 gnr_snc2_map[] =3D { 1, 1, 0, 0 }; + static const u8 snc3_map[] =3D { 1, 1, 0, 0, 2, 2 }; + static const u8 *snc_map =3D NULL; + static size_t snc_map_len; + + /* snc_map is not inited yet. We only look up once to avoid expensive ope= rations. */ + if (!snc_map) { + switch (snc_nodes) { + case 2: + cpuid =3D get_cpuid_str((struct perf_cpu){ 0 }); + if (cpuid) { + if (strcmp_cpuid_str("GenuineIntel-6-CF", cpuid) =3D=3D 0) { + snc_map =3D emr_snc2_map; + snc_map_len =3D ARRAY_SIZE(emr_snc2_map); + } else if (strcmp_cpuid_str("GenuineIntel-6-A[DE]", cpuid) =3D=3D 0) { + snc_map =3D gnr_snc2_map; + snc_map_len =3D ARRAY_SIZE(gnr_snc2_map); + } + free(cpuid); + } + break; + case 3: + snc_map =3D snc3_map; + snc_map_len =3D ARRAY_SIZE(snc3_map); + break; + default: + /* Error or no lookup support for SNC with >3 nodes. */ + return 0; + } + + if (!snc_map) { + pr_warning("Unexpected: can not find snc map config"); + return 0; + } } /* Compute SNC for PMU. */ @@ -161,11 +190,12 @@ static int uncore_imc_snc(struct perf_pmu *pmu) pr_warning("Unexpected: unable to compute IMC number '%s'\n", pmu->name); return 0; } - if (imc_num >=3D snc_map_len) { + if (imc_num >=3D snc_map_len * perf_cpu_map__nr(pmu->cpus)) { pr_warning("Unexpected IMC %d for SNC%d mapping\n", imc_num, snc_nodes); return 0; } - return snc_map[imc_num]; + + return snc_map[imc_num % snc_map_len]; } static int uncore_cha_imc_compute_cpu_adjust(int pmu_snc) @@ -205,7 +235,7 @@ static int uncore_cha_imc_compute_cpu_adjust(int pmu_sn= c) return cpu_adjust[pmu_snc]; } -static void gnr_uncore_cha_imc_adjust_cpumask_for_snc(struct perf_pmu *pmu= , bool cha) +static void uncore_cha_imc_adjust_cpumask_for_snc(struct perf_pmu *pmu, bo= ol cha) { // With sub-NUMA clustering (SNC) there is a NUMA node per SNC in the // topology. For example, a two socket graniterapids machine may be set @@ -304,11 +334,12 @@ void perf_pmu__arch_init(struct perf_pmu *pmu) pmu->mem_events =3D perf_mem_events_intel_aux; else pmu->mem_events =3D perf_mem_events_intel; - } else if (x86__is_intel_graniterapids()) { + } else if (x86__is_snc_supported()) { if (starts_with(pmu->name, "uncore_cha_")) - gnr_uncore_cha_imc_adjust_cpumask_for_snc(pmu, /*cha=3D*/true); - else if (starts_with(pmu->name, "uncore_imc_")) - gnr_uncore_cha_imc_adjust_cpumask_for_snc(pmu, /*cha=3D*/false); + uncore_cha_imc_adjust_cpumask_for_snc(pmu, /*cha=3D*/true); + else if (starts_with(pmu->name, "uncore_imc_") && + !starts_with(pmu->name, "uncore_imc_free_running")) + uncore_cha_imc_adjust_cpumask_for_snc(pmu, /*cha=3D*/false); } } } -- 2.53.0.rc2.204.g2597b5adb4-goog