perf: Use exec_comm instead of comm when printing

Process names are much clearer as the first line in a flamegraph
If we ever want to upstream this, it needs to be a separate option

Author: Daniel Shaulov

tools/perf/builtin-script.c

@@ -635,11 +635,11 @@ static int perf_sample__fprintf_start(struct perf_sample *sample,
 
 	if (PRINT_FIELD(COMM)) {
 		if (latency_format)
-			printed += fprintf(fp, "%8.8s ", thread__comm_str(thread));
+			printed += fprintf(fp, "%8.8s ", thread__exec_comm_str(thread));
 		else if (PRINT_FIELD(IP) && evsel__has_callchain(evsel) && symbol_conf.use_callchain)
-			printed += fprintf(fp, "%s ", thread__comm_str(thread));
+			printed += fprintf(fp, "%s ", thread__exec_comm_str(thread));
 		else
-			printed += fprintf(fp, "%16s ", thread__comm_str(thread));
+			printed += fprintf(fp, "%16s ", thread__exec_comm_str(thread));
 	}
 
 	if (PRINT_FIELD(PID) && PRINT_FIELD(TID))

tools/perf/util/thread.c

@@ -308,6 +308,27 @@ const char *thread__comm_str(struct thread *thread)
 	return str;
 }
 
+static const char *__thread__exec_comm_str(const struct thread *thread)
+{
+	const struct comm *comm = thread__exec_comm(thread);
+
+	if (!comm)
+		return NULL;
+
+	return comm__str(comm);
+}
+
+const char *thread__exec_comm_str(struct thread *thread)
+{
+	const char *str;
+
+	down_read(&thread->comm_lock);
+	str = __thread__exec_comm_str(thread);
+	up_read(&thread->comm_lock);
+
+	return str;
+}
+
 /* CHECKME: it should probably better return the max comm len from its comm list */
 int thread__comm_len(struct thread *thread)
 {

tools/perf/util/thread.h

@@ -90,6 +90,7 @@ int thread__comm_len(struct thread *thread);
 struct comm *thread__comm(const struct thread *thread);
 struct comm *thread__exec_comm(const struct thread *thread);
 const char *thread__comm_str(struct thread *thread);
+const char *thread__exec_comm_str(struct thread *thread);
 int thread__insert_map(struct thread *thread, struct map *map);
 int thread__fork(struct thread *thread, struct thread *parent, u64 timestamp, bool do_maps_clone);
 size_t thread__fprintf(struct thread *thread, FILE *fp);