[PATCH RESEND v7 2/2] perf pmu intel: Adjust cpumasks for sub-NUMA clusters on Sapphire Rapids and Emerald Rapids
From: Chun-Tse Shao
Date: Wed May 27 2026 - 18:19:46 EST
Similar to GNR [1], Sapphire Rapids and Emerald Rapids support sub-NUMA
clusters as well. Adjust cpumasks using the same logic as GNR in [1].
Tested on Emerald Rapids with SNC2 enabled:
$ perf stat --per-node -e 'UNC_CHA_CLOCKTICKS,UNC_M_CLOCKTICKS' -a -- sleep 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@xxxxxxxxxx
Reviewed-by: Zide Chen <zide.chen@xxxxxxxxx>
Reviewed-by: Ian Rogers <irogers@xxxxxxxxxx>
Signed-off-by: Chun-Tse Shao <ctshao@xxxxxxxxxx>
Assisted-by: Gemini:gemini-3.1-pro-preview
---
tools/perf/arch/x86/util/pmu.c | 13 +++++++++++--
1 file changed, 11 insertions(+), 2 deletions(-)
diff --git a/tools/perf/arch/x86/util/pmu.c b/tools/perf/arch/x86/util/pmu.c
index 9b00d5720fb7..2c24ef3140da 100644
--- a/tools/perf/arch/x86/util/pmu.c
+++ b/tools/perf/arch/x86/util/pmu.c
@@ -23,6 +23,8 @@
#include "util/env.h"
#include "util/header.h"
+#define GENUINE_INTEL_SPR "GenuineIntel-6-8F"
+#define GENUINE_INTEL_EMR "GenuineIntel-6-CF"
#define GENUINE_INTEL_GNR "GenuineIntel-6-A[DE]"
static bool cached_snc_supported;
@@ -30,8 +32,10 @@ static pthread_once_t snc_support_once = PTHREAD_ONCE_INIT;
static void init_snc_support(void)
{
- /* Graniterapids supports SNC configuration. */
+ /* Sapphirerapids Emeraldrapids Graniterapids support SNC configuration. */
static const char *const supported_cpuids[] = {
+ GENUINE_INTEL_SPR, /* Sapphirerapids */
+ GENUINE_INTEL_EMR, /* Emeraldrapids */
GENUINE_INTEL_GNR, /* Graniterapids */
};
char *cpuid = get_cpuid_str((struct perf_cpu){0});
@@ -161,6 +165,7 @@ static void init_snc_map(void)
{
int snc_nodes = snc_nodes_per_l3_cache();
char *cpuid;
+ static const u8 spr_emr_snc2_map[] = { 0, 0, 1, 1 };
static const u8 gnr_snc2_map[] = { 1, 1, 0, 0 };
static const u8 snc3_map[] = { 1, 1, 0, 0, 2, 2 };
@@ -168,7 +173,11 @@ static void init_snc_map(void)
case 2:
cpuid = get_cpuid_str((struct perf_cpu){ 0 });
if (cpuid) {
- if (strcmp_cpuid_str(GENUINE_INTEL_GNR, cpuid) == 0) {
+ if (strcmp_cpuid_str(GENUINE_INTEL_SPR, cpuid) == 0 ||
+ strcmp_cpuid_str(GENUINE_INTEL_EMR, cpuid) == 0) {
+ cached_imc_snc_map = spr_emr_snc2_map;
+ cached_imc_snc_map_len = ARRAY_SIZE(spr_emr_snc2_map);
+ } else if (strcmp_cpuid_str(GENUINE_INTEL_GNR, cpuid) == 0) {
cached_imc_snc_map = gnr_snc2_map;
cached_imc_snc_map_len = ARRAY_SIZE(gnr_snc2_map);
}
--
2.54.0.823.g6e5bcc1fc9-goog