2022-04-25 06:24:32

by Ravi Bangoria

[permalink] [raw]
Subject: [PATCH 3/6] perf/tool/amd/ibs: Warn about sampling period skew

Samples without an L3 miss are discarded and counter is reset with
random value (between 1-15 for fetch pmu and 1-127 for op pmu) when
IBS L3 miss filtering is enabled. This causes a sampling period skew
but there is no way to reconstruct aggregated sampling period. So
print a warning at perf record if user sets l3missonly=1.

Ex:
# perf record -c 10000 -C 0 -e ibs_op/l3missonly=1/
WARNING: Hw internally resets sampling period when L3 Miss Filtering is enabled
and tagged operation does not cause L3 Miss. This causes sampling period skew.

Signed-off-by: Ravi Bangoria <[email protected]>
---
tools/perf/arch/x86/util/evsel.c | 31 +++++++++++++++++++++++++++++++
tools/perf/util/evsel.c | 7 +++++++
tools/perf/util/evsel.h | 1 +
3 files changed, 39 insertions(+)

diff --git a/tools/perf/arch/x86/util/evsel.c b/tools/perf/arch/x86/util/evsel.c
index ac2899a25b7a..acaabdba5db8 100644
--- a/tools/perf/arch/x86/util/evsel.c
+++ b/tools/perf/arch/x86/util/evsel.c
@@ -4,6 +4,8 @@
#include "util/evsel.h"
#include "util/env.h"
#include "linux/string.h"
+#include "util/pmu.h"
+#include "util/debug.h"

void arch_evsel__set_sample_weight(struct evsel *evsel)
{
@@ -29,3 +31,32 @@ void arch_evsel__fixup_new_cycles(struct perf_event_attr *attr)

free(env.cpuid);
}
+
+void arch_evsel__warn_ambiguity(struct evsel *evsel, struct perf_event_attr *attr)
+{
+ struct perf_env *env = evsel__env(evsel);
+ struct perf_pmu *evsel_pmu = evsel__find_pmu(evsel);
+ struct perf_pmu *ibs_fetch_pmu = perf_pmu__find("ibs_fetch");
+ struct perf_pmu *ibs_op_pmu = perf_pmu__find("ibs_op");
+ static int warned_once;
+
+ if (warned_once || !perf_env__cpuid(env) || !env->cpuid ||
+ !strstarts(env->cpuid, "AuthenticAMD") || !evsel_pmu)
+ return;
+
+ if (ibs_fetch_pmu && ibs_fetch_pmu->type == evsel_pmu->type) {
+ if (attr->config & (1ULL << 59)) {
+ pr_warning(
+"WARNING: Hw internally resets sampling period when L3 Miss Filtering is enabled\n"
+"and tagged operation does not cause L3 Miss. This causes sampling period skew.\n");
+ warned_once = 1;
+ }
+ } else if (ibs_op_pmu && ibs_op_pmu->type == evsel_pmu->type) {
+ if (attr->config & (1ULL << 16)) {
+ pr_warning(
+"WARNING: Hw internally resets sampling period when L3 Miss Filtering is enabled\n"
+"and tagged operation does not cause L3 Miss. This causes sampling period skew.\n");
+ warned_once = 1;
+ }
+ }
+}
diff --git a/tools/perf/util/evsel.c b/tools/perf/util/evsel.c
index 2a1729e7aee4..4f8b72d4a521 100644
--- a/tools/perf/util/evsel.c
+++ b/tools/perf/util/evsel.c
@@ -1064,6 +1064,11 @@ void __weak arch_evsel__fixup_new_cycles(struct perf_event_attr *attr __maybe_un
{
}

+void __weak arch_evsel__warn_ambiguity(struct evsel *evsel __maybe_unused,
+ struct perf_event_attr *attr __maybe_unused)
+{
+}
+
static void evsel__set_default_freq_period(struct record_opts *opts,
struct perf_event_attr *attr)
{
@@ -1339,6 +1344,8 @@ void evsel__config(struct evsel *evsel, struct record_opts *opts,
*/
if (evsel__is_dummy_event(evsel))
evsel__reset_sample_bit(evsel, BRANCH_STACK);
+
+ arch_evsel__warn_ambiguity(evsel, attr);
}

int evsel__set_filter(struct evsel *evsel, const char *filter)
diff --git a/tools/perf/util/evsel.h b/tools/perf/util/evsel.h
index 041b42d33bf5..195ae30ec45b 100644
--- a/tools/perf/util/evsel.h
+++ b/tools/perf/util/evsel.h
@@ -281,6 +281,7 @@ void evsel__set_sample_id(struct evsel *evsel, bool use_sample_identifier);

void arch_evsel__set_sample_weight(struct evsel *evsel);
void arch_evsel__fixup_new_cycles(struct perf_event_attr *attr);
+void arch_evsel__warn_ambiguity(struct evsel *evsel, struct perf_event_attr *attr);

int evsel__set_filter(struct evsel *evsel, const char *filter);
int evsel__append_tp_filter(struct evsel *evsel, const char *filter);
--
2.27.0


2022-04-26 19:40:02

by Ravi Bangoria

[permalink] [raw]
Subject: Re: [PATCH 3/6] perf/tool/amd/ibs: Warn about sampling period skew



On 26-Apr-22 3:39 PM, Robert Richter wrote:
> On 25.04.22 10:13:20, Ravi Bangoria wrote:
>
>> @@ -29,3 +31,32 @@ void arch_evsel__fixup_new_cycles(struct perf_event_attr *attr)
>>
>> free(env.cpuid);
>> }
>> +
>> +void arch_evsel__warn_ambiguity(struct evsel *evsel, struct perf_event_attr *attr)
>
> Have an 'ibs_' string in the name?

No, this is not ibs specific. Any arch can define it's own version.

>
>> +{
>> + struct perf_env *env = evsel__env(evsel);
>> + struct perf_pmu *evsel_pmu = evsel__find_pmu(evsel);
>> + struct perf_pmu *ibs_fetch_pmu = perf_pmu__find("ibs_fetch");
>> + struct perf_pmu *ibs_op_pmu = perf_pmu__find("ibs_op");
>> + static int warned_once;
>> +
>> + if (warned_once || !perf_env__cpuid(env) || !env->cpuid ||
>> + !strstarts(env->cpuid, "AuthenticAMD") || !evsel_pmu)
>> + return;
>> +
>> + if (ibs_fetch_pmu && ibs_fetch_pmu->type == evsel_pmu->type) {
>> + if (attr->config & (1ULL << 59)) {
>> + pr_warning(
>> +"WARNING: Hw internally resets sampling period when L3 Miss Filtering is enabled\n"
>> +"and tagged operation does not cause L3 Miss. This causes sampling period skew.\n");
>> + warned_once = 1;
>> + }
>> + } else if (ibs_op_pmu && ibs_op_pmu->type == evsel_pmu->type) {
>> + if (attr->config & (1ULL << 16)) {
>> + pr_warning(
>> +"WARNING: Hw internally resets sampling period when L3 Miss Filtering is enabled\n"
>> +"and tagged operation does not cause L3 Miss. This causes sampling period skew.\n");
>
> Avoid duplicate code. You could move it to the end of the function and
> early return if the config bit is not set.

Sure.

Thanks,
Ravi

2022-04-27 07:18:24

by Robert Richter

[permalink] [raw]
Subject: Re: [PATCH 3/6] perf/tool/amd/ibs: Warn about sampling period skew

On 25.04.22 10:13:20, Ravi Bangoria wrote:

> @@ -29,3 +31,32 @@ void arch_evsel__fixup_new_cycles(struct perf_event_attr *attr)
>
> free(env.cpuid);
> }
> +
> +void arch_evsel__warn_ambiguity(struct evsel *evsel, struct perf_event_attr *attr)

Have an 'ibs_' string in the name?

> +{
> + struct perf_env *env = evsel__env(evsel);
> + struct perf_pmu *evsel_pmu = evsel__find_pmu(evsel);
> + struct perf_pmu *ibs_fetch_pmu = perf_pmu__find("ibs_fetch");
> + struct perf_pmu *ibs_op_pmu = perf_pmu__find("ibs_op");
> + static int warned_once;
> +
> + if (warned_once || !perf_env__cpuid(env) || !env->cpuid ||
> + !strstarts(env->cpuid, "AuthenticAMD") || !evsel_pmu)
> + return;
> +
> + if (ibs_fetch_pmu && ibs_fetch_pmu->type == evsel_pmu->type) {
> + if (attr->config & (1ULL << 59)) {
> + pr_warning(
> +"WARNING: Hw internally resets sampling period when L3 Miss Filtering is enabled\n"
> +"and tagged operation does not cause L3 Miss. This causes sampling period skew.\n");
> + warned_once = 1;
> + }
> + } else if (ibs_op_pmu && ibs_op_pmu->type == evsel_pmu->type) {
> + if (attr->config & (1ULL << 16)) {
> + pr_warning(
> +"WARNING: Hw internally resets sampling period when L3 Miss Filtering is enabled\n"
> +"and tagged operation does not cause L3 Miss. This causes sampling period skew.\n");

Avoid duplicate code. You could move it to the end of the function and
early return if the config bit is not set.

> + warned_once = 1;
> + }
> + }
> +}

-Robert