perf sample: Make user_regs and intr_regs optional

The struct dump_regs contains 512 bytes of cache_regs, meaning the two
values in perf_sample contribute 1088 bytes of its total 1384 bytes
size. Initializing this much memory has a cost reported by Tavian
Barnes <tavianator@tavianator.com> as about 2.5% when running `perf
script --itrace=i0`:
https://lore.kernel.org/lkml/d841b97b3ad2ca8bcab07e4293375fb7c32dfce7.1736618095.git.tavianator@tavianator.com/

Adrian Hunter <adrian.hunter@intel.com> replied that the zero
initialization was necessary and couldn't simply be removed.

This patch aims to strike a middle ground of still zeroing the
perf_sample, but removing 79% of its size by make user_regs and
intr_regs optional pointers to zalloc-ed memory. To support the
allocation accessors are created for user_regs and intr_regs. To
support correct cleanup perf_sample__init and perf_sample__exit
functions are created and added throughout the code base.

Signed-off-by: Ian Rogers <irogers@google.com>
Link: https://lore.kernel.org/r/20250113194345.1537821-1-irogers@google.com
Signed-off-by: Namhyung Kim <namhyung@kernel.org>

1 files changed, 22 insertions, 9 deletions

diff --git a/tools/perf/util/cs-etm.c b/tools/perf/util/cs-etm.c
index 0bf9e5c27b59..30f4bb3e7fa3 100644
--- a/tools/perf/util/cs-etm.c
+++ b/tools/perf/util/cs-etm.c
@@ -506,20 +506,27 @@ static int cs_etm__process_aux_output_hw_id(struct perf_session *session,
 	evsel = evlist__event2evsel(session->evlist, event);
 	if (!evsel)
 		return -EINVAL;
+	perf_sample__init(&sample, /*all=*/false);
 	err = evsel__parse_sample(evsel, event, &sample);
 	if (err)
-		return err;
+		goto out;
 	cpu = sample.cpu;
 	if (cpu == -1) {
 		/* no CPU in the sample - possibly recorded with an old version of perf */
 		pr_err("CS_ETM: no CPU AUX_OUTPUT_HW_ID sample. Use compatible perf to record.");
-		return -EINVAL;
+		err = -EINVAL;
+		goto out;
 	}
 
-	if (FIELD_GET(CS_AUX_HW_ID_MINOR_VERSION_MASK, hw_id) == 0)
-		return cs_etm__process_trace_id_v0(etm, cpu, hw_id);
+	if (FIELD_GET(CS_AUX_HW_ID_MINOR_VERSION_MASK, hw_id) == 0) {
+		err = cs_etm__process_trace_id_v0(etm, cpu, hw_id);
+		goto out;
+	}
 
-	return cs_etm__process_trace_id_v0_1(etm, cpu, hw_id);
+	err = cs_etm__process_trace_id_v0_1(etm, cpu, hw_id);
+out:
+	perf_sample__exit(&sample);
+	return err;
 }
 
 void cs_etm__etmq_set_traceid_queue_timestamp(struct cs_etm_queue *etmq,
@@ -1560,8 +1567,9 @@ static int cs_etm__synth_instruction_sample(struct cs_etm_queue *etmq,
 	int ret = 0;
 	struct cs_etm_auxtrace *etm = etmq->etm;
 	union perf_event *event = tidq->event_buf;
-	struct perf_sample sample = {.ip = 0,};
+	struct perf_sample sample;
 
+	perf_sample__init(&sample, /*all=*/true);
 	event->sample.header.type = PERF_RECORD_SAMPLE;
 	event->sample.header.misc = cs_etm__cpu_mode(etmq, addr, tidq->el);
 	event->sample.header.size = sizeof(struct perf_event_header);
@@ -1598,6 +1606,7 @@ static int cs_etm__synth_instruction_sample(struct cs_etm_queue *etmq,
 			"CS ETM Trace: failed to deliver instruction event, error %d\n",
 			ret);
 
+	perf_sample__exit(&sample);
 	return ret;
 }
 
@@ -3151,9 +3160,10 @@ static int cs_etm__queue_aux_records_cb(struct perf_session *session, union perf
 	evsel = evlist__event2evsel(session->evlist, event);
 	if (!evsel)
 		return -EINVAL;
+	perf_sample__init(&sample, /*all=*/false);
 	ret = evsel__parse_sample(evsel, event, &sample);
 	if (ret)
-		return ret;
+		goto out;
 
 	/*
 	 * Loop through the auxtrace index to find the buffer that matches up with this aux event.
@@ -3168,7 +3178,7 @@ static int cs_etm__queue_aux_records_cb(struct perf_session *session, union perf
 			 * 1 ('not found')
 			 */
 			if (ret != 1)
-				return ret;
+				goto out;
 		}
 	}
 
@@ -3178,7 +3188,10 @@ static int cs_etm__queue_aux_records_cb(struct perf_session *session, union perf
 	 */
 	pr_err("CS ETM: Couldn't find auxtrace buffer for aux_offset: %#"PRI_lx64
 	       " tid: %d cpu: %d\n", event->aux.aux_offset, sample.tid, sample.cpu);
-	return 0;
+	ret = 0;
+out:
+	perf_sample__exit(&sample);
+	return ret;
 }
 
 static int cs_etm__queue_aux_records(struct perf_session *session)