perf pmu: Lazily compute default config

The default config is computed during creation of the PMU and may do
things like scanning sysfs, when the PMU may just be used as part of
scanning. Change default_config to perf_event_attr_init_default, a
callback that is used when a default config needs initializing. This
avoids holding onto the memory for a perf_event_attr and copying.

On a tigerlake laptop running the pmu-scan benchmark:

Before:
Running 'internals/pmu-scan' benchmark:
Computing performance of sysfs PMU event scan for 100 times
  Average core PMU scanning took: 28.780 usec (+- 0.503 usec)
  Average PMU scanning took: 283.480 usec (+- 18.471 usec)
Number of openat syscalls: 30,227

After:
Running 'internals/pmu-scan' benchmark:
Computing performance of sysfs PMU event scan for 100 times
  Average core PMU scanning took: 27.880 usec (+- 0.169 usec)
  Average PMU scanning took: 245.260 usec (+- 15.758 usec)
Number of openat syscalls: 28,914

Over 3 runs it is a nearly 12% reduction in execution time and a 4.3%
of openat calls.

Signed-off-by: Ian Rogers <irogers@google.com>
Reviewed-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: Ravi Bangoria <ravi.bangoria@amd.com>
Cc: James Clark <james.clark@arm.com>
Cc: Suzuki K Poulose <suzuki.poulose@arm.com>
Cc: Yang Jihong <yangjihong1@huawei.com>
Cc: Will Deacon <will@kernel.org>
Cc: Leo Yan <leo.yan@linaro.org>
Cc: Mike Leach <mike.leach@linaro.org>
Cc: Jing Zhang <renyu.zj@linux.alibaba.com>
Cc: Kajol Jain <kjain@linux.ibm.com>
Cc: Thomas Richter <tmricht@linux.ibm.com>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: John Garry <john.g.garry@oracle.com>
Cc: linux-arm-kernel@lists.infradead.org
Cc: coresight@lists.linaro.org
Link: https://lore.kernel.org/r/20231012175645.1849503-8-irogers@google.com
Signed-off-by: Namhyung Kim <namhyung@kernel.org>

6 files changed, 17 insertions, 14 deletions

diff --git a/tools/perf/util/arm-spe.h b/tools/perf/util/arm-spe.h
index 98d3235781c3..4f4900c18f3e 100644
--- a/tools/perf/util/arm-spe.h
+++ b/tools/perf/util/arm-spe.h
@@ -27,5 +27,7 @@ struct auxtrace_record *arm_spe_recording_init(int *err,
 int arm_spe_process_auxtrace_info(union perf_event *event,
 				  struct perf_session *session);
 
-struct perf_event_attr *arm_spe_pmu_default_config(struct perf_pmu *arm_spe_pmu);
+void arm_spe_pmu_default_config(const struct perf_pmu *arm_spe_pmu,
+				struct perf_event_attr *attr);
+
 #endif
diff --git a/tools/perf/util/cs-etm.h b/tools/perf/util/cs-etm.h
index 7cca37887917..4696267a32f0 100644
--- a/tools/perf/util/cs-etm.h
+++ b/tools/perf/util/cs-etm.h
@@ -242,7 +242,7 @@ struct cs_etm_packet_queue {
 
 int cs_etm__process_auxtrace_info(union perf_event *event,
 				  struct perf_session *session);
-struct perf_event_attr *cs_etm_get_default_config(struct perf_pmu *pmu);
+void cs_etm_get_default_config(const struct perf_pmu *pmu, struct perf_event_attr *attr);
 
 enum cs_etm_pid_fmt {
 	CS_ETM_PIDFMT_NONE,
diff --git a/tools/perf/util/intel-pt.h b/tools/perf/util/intel-pt.h
index c7d6068e3a6b..18fd0be52e6c 100644
--- a/tools/perf/util/intel-pt.h
+++ b/tools/perf/util/intel-pt.h
@@ -42,6 +42,7 @@ struct auxtrace_record *intel_pt_recording_init(int *err);
 int intel_pt_process_auxtrace_info(union perf_event *event,
 				   struct perf_session *session);
 
-struct perf_event_attr *intel_pt_pmu_default_config(struct perf_pmu *pmu);
+void intel_pt_pmu_default_config(const struct perf_pmu *intel_pt_pmu,
+				 struct perf_event_attr *attr);
 
 #endif
diff --git a/tools/perf/util/parse-events.c b/tools/perf/util/parse-events.c
index 23c027cf20ae..aa2f5c6fc7fc 100644
--- a/tools/perf/util/parse-events.c
+++ b/tools/perf/util/parse-events.c
@@ -1418,11 +1418,10 @@ int parse_events_add_pmu(struct parse_events_state *parse_state,
 	}
 	fix_raw(&parsed_terms, pmu);
 
-	if (pmu->default_config) {
-		memcpy(&attr, pmu->default_config, sizeof(struct perf_event_attr));
-	} else {
-		memset(&attr, 0, sizeof(attr));
-	}
+	memset(&attr, 0, sizeof(attr));
+	if (pmu->perf_event_attr_init_default)
+		pmu->perf_event_attr_init_default(pmu, &attr);
+
 	attr.type = pmu->type;
 
 	if (list_empty(&parsed_terms.terms)) {
@@ -1466,7 +1465,8 @@ int parse_events_add_pmu(struct parse_events_state *parse_state,
 	 * When using default config, record which bits of attr->config were
 	 * changed by the user.
 	 */
-	if (pmu->default_config && get_config_chgs(pmu, &parsed_terms, &config_terms)) {
+	if (pmu->perf_event_attr_init_default &&
+	    get_config_chgs(pmu, &parsed_terms, &config_terms)) {
 		parse_events_terms__exit(&parsed_terms);
 		return -ENOMEM;
 	}
diff --git a/tools/perf/util/pmu.c b/tools/perf/util/pmu.c
index 8ef675ea7bdd..a967d25e899b 100644
--- a/tools/perf/util/pmu.c
+++ b/tools/perf/util/pmu.c
@@ -1402,7 +1402,7 @@ int perf_pmu__config(struct perf_pmu *pmu, struct perf_event_attr *attr,
 		     struct parse_events_terms *head_terms,
 		     struct parse_events_error *err)
 {
-	bool zero = !!pmu->default_config;
+	bool zero = !!pmu->perf_event_attr_init_default;
 
 	return perf_pmu__config_terms(pmu, attr, head_terms, zero, err);
 }
@@ -2064,7 +2064,6 @@ void perf_pmu__delete(struct perf_pmu *pmu)
 
 	perf_cpu_map__put(pmu->cpus);
 
-	zfree(&pmu->default_config);
 	zfree(&pmu->name);
 	zfree(&pmu->alias_name);
 	zfree(&pmu->id);
diff --git a/tools/perf/util/pmu.h b/tools/perf/util/pmu.h
index 5a05131aa4ce..d2895d415f08 100644
--- a/tools/perf/util/pmu.h
+++ b/tools/perf/util/pmu.h
@@ -92,10 +92,11 @@ struct perf_pmu {
 	 */
 	int max_precise;
 	/**
-	 * @default_config: Optional default perf_event_attr determined in
-	 * architecture specific code.
+	 * @perf_event_attr_init_default: Optional function to default
+	 * initialize PMU specific parts of the perf_event_attr.
 	 */
-	struct perf_event_attr *default_config;
+	void (*perf_event_attr_init_default)(const struct perf_pmu *pmu,
+					     struct perf_event_attr *attr);
 	/**
 	 * @cpus: Empty or the contents of either of:
 	 * <sysfs>/bus/event_source/devices/<name>/cpumask.