* [PATCH v5 1/2] perf pmu intel: Generalize SNC cpumask adjustment for multiple platforms
@ 2026-04-07 20:38 Chun-Tse Shao
2026-04-07 20:38 ` [PATCH v5 2/2] perf pmu intel: Adjust cpumaks for sub-NUMA clusters on Emeraldrapids Chun-Tse Shao
0 siblings, 1 reply; 2+ messages in thread
From: Chun-Tse Shao @ 2026-04-07 20:38 UTC (permalink / raw)
To: linux-kernel
Cc: Chun-Tse Shao, peterz, mingo, acme, namhyung, mark.rutland,
alexander.shishkin, jolsa, irogers, adrian.hunter, james.clark,
zide.chen, ravi.bangoria, linux-perf-users
Prepare for supporting more Intel platforms with sub-NUMA clustering by
generalizing the GNR specific logic.
Signed-off-by: Chun-Tse Shao <ctshao@google.com>
---
v5:
Split patch.
v4: lore.kernel.org/20260402205300.1953706-1-ctshao@google.com
Rebase.
v3: lore.kernel.org/20260212223942.3832857-1-ctshao@google.com
Fix a typo.
v2: lore.kernel.org/20260205232220.1980168-1-ctshao@google.com
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 | 44 +++++++++++++++++++++-------------
1 file changed, 27 insertions(+), 17 deletions(-)
diff --git a/tools/perf/arch/x86/util/pmu.c b/tools/perf/arch/x86/util/pmu.c
index 0661e0f0b02d..938be36ec0f7 100644
--- a/tools/perf/arch/x86/util/pmu.c
+++ b/tools/perf/arch/x86/util/pmu.c
@@ -23,20 +23,28 @@
#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 = "GenuineIntel-6-A[DE]";
+ if (!checked_if_snc_supported) {
+
+ /* Graniterapids supports SNC configuration. */
+ static const char *const supported_cpuids[] = {
+ "GenuineIntel-6-A[DE]", /* Graniterapids */
+ };
char *cpuid = get_cpuid_str((struct perf_cpu){0});
- is_graniterapids = cpuid && strcmp_cpuid_str(graniterapids_cpuid, cpuid) == 0;
+ for (size_t i = 0; i < ARRAY_SIZE(supported_cpuids); i++) {
+ is_supported = cpuid && strcmp_cpuid_str(supported_cpuids[i], cpuid) == 0;
+ if (is_supported)
+ break;
+ }
free(cpuid);
- checked_if_graniterapids = true;
+ checked_if_snc_supported = true;
}
- return is_graniterapids;
+ return is_supported;
}
static struct perf_cpu_map *read_sysfs_cpu_map(const char *sysfs_path)
@@ -133,8 +141,8 @@ static int uncore_imc_snc(struct perf_pmu *pmu)
// Compute the IMC SNC using lookup tables.
unsigned int imc_num;
int snc_nodes = snc_nodes_per_l3_cache();
- const u8 snc2_map[] = {1, 1, 0, 0, 1, 1, 0, 0};
- const u8 snc3_map[] = {1, 1, 0, 0, 2, 2, 1, 1, 0, 0, 2, 2};
+ const u8 snc2_map[] = {1, 1, 0, 0};
+ const u8 snc3_map[] = {1, 1, 0, 0, 2, 2};
const u8 *snc_map;
size_t snc_map_len;
@@ -157,11 +165,12 @@ 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 >= snc_map_len) {
+ if (imc_num >= 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)
@@ -201,7 +210,7 @@ static int uncore_cha_imc_compute_cpu_adjust(int pmu_snc)
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, bool 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
@@ -301,11 +310,12 @@ void perf_pmu__arch_init(struct perf_pmu *pmu)
pmu->mem_events = perf_mem_events_intel_aux;
else
pmu->mem_events = perf_mem_events_intel;
- } else if (x86__is_intel_graniterapids()) {
+ } else if (x86__is_snc_supported()) {
if (strstarts(pmu->name, "uncore_cha_"))
- gnr_uncore_cha_imc_adjust_cpumask_for_snc(pmu, /*cha=*/true);
- else if (strstarts(pmu->name, "uncore_imc_"))
- gnr_uncore_cha_imc_adjust_cpumask_for_snc(pmu, /*cha=*/false);
+ uncore_cha_imc_adjust_cpumask_for_snc(pmu, /*cha=*/true);
+ else if (strstarts(pmu->name, "uncore_imc_") &&
+ !strstarts(pmu->name, "uncore_imc_free_running"))
+ uncore_cha_imc_adjust_cpumask_for_snc(pmu, /*cha=*/false);
}
}
}
--
2.53.0.1213.gd9a14994de-goog
^ permalink raw reply related [flat|nested] 2+ messages in thread* [PATCH v5 2/2] perf pmu intel: Adjust cpumaks for sub-NUMA clusters on Emeraldrapids
2026-04-07 20:38 [PATCH v5 1/2] perf pmu intel: Generalize SNC cpumask adjustment for multiple platforms Chun-Tse Shao
@ 2026-04-07 20:38 ` Chun-Tse Shao
0 siblings, 0 replies; 2+ messages in thread
From: Chun-Tse Shao @ 2026-04-07 20:38 UTC (permalink / raw)
To: linux-kernel
Cc: Chun-Tse Shao, Zide Chen, Ian Rogers, peterz, mingo, acme,
namhyung, mark.rutland, alexander.shishkin, jolsa, adrian.hunter,
james.clark, ravi.bangoria, linux-perf-users
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 -- 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@google.com
Reviewed-by: Zide Chen <zide.chen@intel.com>
Reviewed-by: Ian Rogers <irogers@google.com>
Signed-off-by: Chun-Tse Shao <ctshao@google.com>
---
tools/perf/arch/x86/util/pmu.c | 56 +++++++++++++++++++++++-----------
1 file changed, 38 insertions(+), 18 deletions(-)
diff --git a/tools/perf/arch/x86/util/pmu.c b/tools/perf/arch/x86/util/pmu.c
index 938be36ec0f7..3743f5145505 100644
--- a/tools/perf/arch/x86/util/pmu.c
+++ b/tools/perf/arch/x86/util/pmu.c
@@ -30,8 +30,9 @@ static bool x86__is_snc_supported(void)
if (!checked_if_snc_supported) {
- /* Graniterapids supports SNC configuration. */
+ /* Emeraldrapids Graniterapids support SNC configuration. */
static const char *const supported_cpuids[] = {
+ "GenuineIntel-6-CF", /* Emeraldrapids */
"GenuineIntel-6-A[DE]", /* Graniterapids */
};
char *cpuid = get_cpuid_str((struct perf_cpu){0});
@@ -141,23 +142,42 @@ static int uncore_imc_snc(struct perf_pmu *pmu)
// Compute the IMC SNC using lookup tables.
unsigned int imc_num;
int snc_nodes = snc_nodes_per_l3_cache();
- const u8 snc2_map[] = {1, 1, 0, 0};
- const u8 snc3_map[] = {1, 1, 0, 0, 2, 2};
- const u8 *snc_map;
- size_t snc_map_len;
-
- switch (snc_nodes) {
- case 2:
- snc_map = snc2_map;
- snc_map_len = ARRAY_SIZE(snc2_map);
- break;
- case 3:
- snc_map = snc3_map;
- snc_map_len = 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[] = { 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 };
+ static const u8 *snc_map;
+ static size_t snc_map_len;
+
+ /* snc_map is not inited yet. We only look up once to avoid expensive operations. */
+ if (!snc_map) {
+ switch (snc_nodes) {
+ case 2:
+ cpuid = get_cpuid_str((struct perf_cpu){ 0 });
+ if (cpuid) {
+ if (strcmp_cpuid_str("GenuineIntel-6-CF", cpuid) == 0) {
+ snc_map = emr_snc2_map;
+ snc_map_len = ARRAY_SIZE(emr_snc2_map);
+ } else if (strcmp_cpuid_str("GenuineIntel-6-A[DE]", cpuid) == 0) {
+ snc_map = gnr_snc2_map;
+ snc_map_len = ARRAY_SIZE(gnr_snc2_map);
+ }
+ free(cpuid);
+ }
+ break;
+ case 3:
+ snc_map = snc3_map;
+ snc_map_len = 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. */
--
2.53.0.1213.gd9a14994de-goog
^ permalink raw reply related [flat|nested] 2+ messages in thread
end of thread, other threads:[~2026-04-07 20:39 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-04-07 20:38 [PATCH v5 1/2] perf pmu intel: Generalize SNC cpumask adjustment for multiple platforms Chun-Tse Shao
2026-04-07 20:38 ` [PATCH v5 2/2] perf pmu intel: Adjust cpumaks for sub-NUMA clusters on Emeraldrapids Chun-Tse Shao
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox