perf/arm: adjust hwevents mappings on boot
Subsystems (3)
PATCH v2 1/1 perf/arm: adjust hwevents mappings on boot
The mapping of perf_events generic hardware events to actual PMU events on ARM PMUv3 may not always be correct. This is in particular true for the PERF_COUNT_HW_BRANCH_INSTRUCTIONS event. Although the mapping points to an architected event, it may not always be available. This can be seen with a simple:
$ perf stat -e branches sleep 0 Performance counter stats for 'sleep 0':
<not supported> branches
0.001401081 seconds time elapsed
Yet the hardware does have an event that could be used for branches. This patch fixes the problem by dynamically validating the generic hardware events against the supported architected events. If a mapping is wrong it can be replaced it with another. This is done for the event above at boot time and the kernel will log the remapping:
armv8_pmuv3: hwevent HW_BRANCH_INSTRUCTIONS remapped from 0xc to 0x21
And with that:
$ perf stat -e branches sleep 0
Performance counter stats for 'sleep 0':
166,739 branches
0.000832163 seconds time elapsed--- arch/arm64/kernel/perf_event.c | 41 +++++++++++++++++++++++++++++++++- 1 file changed, 40 insertions(+), 1 deletion(-)
arch/arm64/kernel/perf_event.c +40 -1
@@ -39,7 +39,7 @@* be supported on any given implementation. Unsupported events will * be disabled at run-time based on the PMCEID registers. */ -static const unsigned armv8_pmuv3_perf_map[PERF_COUNT_HW_MAX ] = { +static unsigned armv8_pmuv3_perf_map[PERF_COUNT_HW_MAX ] = {PERF_MAP_ALL_UNSUPPORTED , [PERF_COUNT_HW_CPU_CYCLES ] =ARMV8_PMUV3_PERFCTR_CPU_CYCLES , [PERF_COUNT_HW_INSTRUCTIONS ] =ARMV8_PMUV3_PERFCTR_INST_RETIRED ,
@@ -1222,6 +1222,42 @@ static void armv8_pmu_register_sysctl_table(void)register_sysctl ("kernel" , armv8_pmu_sysctl_table); } +static void armv8pmu_fixup_perf_map (struct arm_pmu *cpu_pmu) +{ +int i, code; +unsigned *map = armv8_pmuv3_perf_map; + +for (i =0 ; i <PERF_COUNT_HW_MAX ; i++) { +retry: + code = map[i]; +if (code ==HW_OP_UNSUPPORTED ) +continue ; + +if (test_bit (map[i], cpu_pmu->pmceid_bitmap)) +continue ; +/* +* mapping does not exist, +* let's see if we can fix it +*/ +switch (i) { +case PERF_COUNT_HW_BRANCH_INSTRUCTIONS : +if (code ==ARMV8_PMUV3_PERFCTR_PC_WRITE_RETIRED ) { + map[i] =ARMV8_PMUV3_PERFCTR_BR_RETIRED ; +pr_info ("armv8_pmuv3: hwevent " +"HW_BRANCH_INSTRUCTIONS remapped " +" from 0x%x to 0x%x\n" , code, map[i]); + goto retry; + } +break ; +default : +pr_info ("armv8_pmuv3: hwevent %d not supported\n" , i); + map[i] =HW_OP_UNSUPPORTED ; +break ; + } + + } +} + static int armv8_pmu_init(struct arm_pmu *cpu_pmu, char *name, int (*map_event)(struct perf_event *event), const struct attribute_group *events,
@@ -1249,6 +1285,9 @@ static int armv8_pmu_init(struct arm_pmu *cpu_pmu, char *name, cpu_pmu->name = name; cpu_pmu->map_event = map_event; + +armv8pmu_fixup_perf_map (cpu_pmu); + cpu_pmu->attr_groups[ARMPMU_ATTR_GROUP_EVENTS ] = events ? events : &armv8_pmuv3_events_attr_group; cpu_pmu->attr_groups[ARMPMU_ATTR_GROUP_FORMATS] = format ?
Signature · 1 line
2.35.1.1021.g381101b075-goog
Will Deacon
maintainerQuoted 72 lines · Stephane Eranian
The mapping of
perf_eventsgeneric hardware events to actual PMU events on ARM PMUv3 may not always be correct. This is in particular true for thePERF_COUNT_HW_BRANCH_INSTRUCTIONSevent. Although the mapping points to an architected event, it may not always be available. This can be seen with a simple:$ perf stat -e branches sleep 0 Performance counter stats for 'sleep 0':<not supported> branches
0.001401081 seconds time elapsedYet the hardware does have an event that could be used for branches. This patch fixes the problem by dynamically validating the generic hardware events against the supported architected events. If a mapping is wrong it can be replaced it with another. This is done for the event above at boot time and the kernel will log the remapping:
armv8_pmuv3: hweventHW_BRANCH_INSTRUCTIONSremapped from0xcto0x21And with that:
$ perf stat -e branches sleep 0Performance counter stats for 'sleep 0':
166,739 branches 0.000832163 seconds time elapsedSigned-off-by Stephane Eranian <[email protected]>--- arch/arm64/kernel/perf_event.c | 41 +++++++++++++++++++++++++++++++++- 1 file changed, 40 insertions(+), 1 deletion(-)arch/arm64/kernel/perf_event.c +20
-1@@ -39,7 +39,7 @@* be supported on any given implementation. Unsupported events will * be disabled at run-time based on the PMCEID registers. */ -static const unsigned armv8_pmuv3_perf_map[PERF_COUNT_HW_MAX ] = { +static unsigned armv8_pmuv3_perf_map[PERF_COUNT_HW_MAX ] = {PERF_MAP_ALL_UNSUPPORTED , [PERF_COUNT_HW_CPU_CYCLES ] =ARMV8_PMUV3_PERFCTR_CPU_CYCLES , [PERF_COUNT_HW_INSTRUCTIONS ] =ARMV8_PMUV3_PERFCTR_INST_RETIRED ,@@ -1222,6 +1222,42 @@ static void armv8_pmu_register_sysctl_table(void)register_sysctl ("kernel" , armv8_pmu_sysctl_table); } +static void armv8pmu_fixup_perf_map (struct arm_pmu *cpu_pmu) +{ +int i, code; +unsigned *map = armv8_pmuv3_perf_map; + +for (i =0 ; i <PERF_COUNT_HW_MAX ; i++) { +retry: + code = map[i]; +if (code ==HW_OP_UNSUPPORTED ) +continue ; + +if (test_bit (map[i], cpu_pmu->pmceid_bitmap)) +continue ; +/* +* mapping does not exist, +* let's see if we can fix it +*/ +switch (i) { +case PERF_COUNT_HW_BRANCH_INSTRUCTIONS :
arch/arm64/kernel/perf_event.c static void armv8_pmu_register_sysctl_table(void) · line 1253
Quoted 10 lines
arch/arm64/kernel/perf_event.c +10
-0+if (code ==ARMV8_PMUV3_PERFCTR_PC_WRITE_RETIRED ) { + map[i] =ARMV8_PMUV3_PERFCTR_BR_RETIRED ; +pr_info ("armv8_pmuv3: hwevent " +"HW_BRANCH_INSTRUCTIONS remapped " +" from 0x%x to 0x%x\n" , code, map[i]); + goto retry; + } +break ; +default : +pr_info ("armv8_pmuv3: hwevent %d not supported\n" , i);
If a CPU supports neither ARMV8_PMUV3_PERFCTR_PC_WRITE_RETIRED nor ARMV8_PMUV3_PERFCTR_BR_RETIRED, won't we get a funny series of messages here? I think I'd prefer to drop the prints altogether.
Will
Quoted 80 lines · Will Deacon
Quoted 78 lines · Stephane Eranian
The mapping of
perf_eventsgeneric hardware events to actual PMU events on ARM PMUv3 may not always be correct. This is in particular true for thePERF_COUNT_HW_BRANCH_INSTRUCTIONSevent. Although the mapping points to an architected event, it may not always be available. This can be seen with a simple:$ perf stat -e branches sleep 0 Performance counter stats for 'sleep 0':<not supported> branches
0.001401081 seconds time elapsedYet the hardware does have an event that could be used for branches. This patch fixes the problem by dynamically validating the generic hardware events against the supported architected events. If a mapping is wrong it can be replaced it with another. This is done for the event above at boot time and the kernel will log the remapping:
armv8_pmuv3: hweventHW_BRANCH_INSTRUCTIONSremapped from0xcto0x21And with that:
$ perf stat -e branches sleep 0Performance counter stats for 'sleep 0':
166,739 branches 0.000832163 seconds time elapsedSigned-off-by Stephane Eranian <[email protected]>--- arch/arm64/kernel/perf_event.c | 41 +++++++++++++++++++++++++++++++++- 1 file changed, 40 insertions(+), 1 deletion(-)arch/arm64/kernel/perf_event.c +26
-1@@ -39,7 +39,7 @@* be supported on any given implementation. Unsupported events will * be disabled at run-time based on the PMCEID registers. */ -static const unsigned armv8_pmuv3_perf_map[PERF_COUNT_HW_MAX ] = { +static unsigned armv8_pmuv3_perf_map[PERF_COUNT_HW_MAX ] = {PERF_MAP_ALL_UNSUPPORTED , [PERF_COUNT_HW_CPU_CYCLES ] =ARMV8_PMUV3_PERFCTR_CPU_CYCLES , [PERF_COUNT_HW_INSTRUCTIONS ] =ARMV8_PMUV3_PERFCTR_INST_RETIRED ,@@ -1222,6 +1222,42 @@ static void armv8_pmu_register_sysctl_table(void)register_sysctl ("kernel" , armv8_pmu_sysctl_table); } +static void armv8pmu_fixup_perf_map (struct arm_pmu *cpu_pmu) +{ +int i, code; +unsigned *map = armv8_pmuv3_perf_map; + +for (i =0 ; i <PERF_COUNT_HW_MAX ; i++) { +retry: + code = map[i]; +if (code ==HW_OP_UNSUPPORTED ) +continue ; + +if (test_bit (map[i], cpu_pmu->pmceid_bitmap)) +continue ; +/* +* mapping does not exist, +* let's see if we can fix it +*/ +switch (i) { +case PERF_COUNT_HW_BRANCH_INSTRUCTIONS : +if (code ==ARMV8_PMUV3_PERFCTR_PC_WRITE_RETIRED ) { + map[i] =ARMV8_PMUV3_PERFCTR_BR_RETIRED ; +pr_info ("armv8_pmuv3: hwevent " +"HW_BRANCH_INSTRUCTIONS remapped " +" from 0x%x to 0x%x\n" , code, map[i]); + goto retry;
Quoted 9 lines
arch/arm64/kernel/perf_event.c static void armv8_pmu_register_sysctl_table(void) · line 1253
Quoted 4 lines
arch/arm64/kernel/perf_event.c +4
-0+ } +break ; +default : +pr_info ("armv8_pmuv3: hwevent %d not supported\n" , i);If a CPU supports neither
ARMV8_PMUV3_PERFCTR_PC_WRITE_RETIREDnorARMV8_PMUV3_PERFCTR_BR_RETIRED, won't we get a funny series of messages here? I think I'd prefer to drop the prints altogether.
Ok, let me clean this up.
Quoted 1 line
Will