perf tools: Allow controlling synthesizing PERF_RECORD_ metadata events during record

Depending on the use case, it might require some kind of synthesizing
and some not. Make it controllable to turn off heavy operations like
MMAP for all tasks.

Currently all users are converted to enable all the synthesis by
default. It'll be updated in the later patch.

Signed-off-by: Namhyung Kim <[e-mail]>
Acked-by: Jiri Olsa <[e-mail]>
Cc: Adrian Hunter <[e-mail]>
Cc: Andi Kleen <[e-mail]>
Cc: Ian Rogers <[e-mail]>
Cc: Ingo Molnar <[e-mail]>
Cc: Jin Yao <[e-mail]>
Cc: Peter Zijlstra <[e-mail]>
Cc: Stephane Eranian <[e-mail]>
Link: https //lore.kernel.org/r/[e-mail]
Signed-off-by: Arnaldo Carvalho de Melo <[e-mail]>

tools/perf/bench/synthesize.c

···
 						NULL,
 						target, threads,
 						process_synthesized_event,
-						data_mmap,
+						true, data_mmap,
 						nr_threads_synthesize);
 		if (err)
 			return err;
···
 						NULL,
 						target, NULL,
 						process_synthesized_event,
-						false,
+						true, false,
 						nr_threads_synthesize);
 		if (err) {
 			perf_session__delete(session);

tools/perf/builtin-kvm.c

···
 	perf_session__set_id_hdr_size(kvm->session);
 	ordered_events__set_copy_on_queue(&kvm->session->ordered_events, true);
 	machine__synthesize_threads(&kvm->session->machines.host, &kvm->opts.target,
-				    kvm->evlist->core.threads, false, 1);
+				    kvm->evlist->core.threads, true, false, 1);
 	err = kvm_live_open_events(kvm);
 	if (err)
 		goto out;

tools/perf/builtin-record.c

···
 	err = perf_event__synthesize_thread_map(&rec->tool, thread_map,
 						 process_synthesized_event,
 						 &rec->session->machines.host,
+						 true,
 						 rec->opts.sample_address);
 	perf_thread_map__put(thread_map);
 	return err;
···
 		f = process_locked_synthesized_event;
 	}
 
-	err = __machine__synthesize_threads(machine, tool, &opts->target, rec->evlist->core.threads,
-					    f, opts->sample_address,
+	err = __machine__synthesize_threads(machine, tool, &opts->target,
+					    rec->evlist->core.threads,
+					    f, true, opts->sample_address,
 					    rec->opts.nr_threads_synthesize);
 
 	if (rec->opts.nr_threads_synthesize > 1)

tools/perf/builtin-top.c

···
 		pr_debug("Couldn't synthesize cgroup events.\n");
 
 	machine__synthesize_threads(&top->session->machines.host, &opts->target,
-				    top->evlist->core.threads, false,
+				    top->evlist->core.threads, true, false,
 				    top->nr_threads_synthesize);
 
 	if (top->nr_threads_synthesize > 1)

tools/perf/builtin-trace.c

···
 		goto out;
 
 	err = __machine__synthesize_threads(trace->host, &trace->tool, &trace->opts.target,
-					    evlist->core.threads, trace__tool_process, false,
-					    1);
+					    evlist->core.threads, trace__tool_process,
+					    true, false, 1);
 out:
 	if (err)
 		symbol__exit();

tools/perf/tests/code-reading.c

···
 	}
 
 	ret = perf_event__synthesize_thread_map(NULL, threads,
-						perf_event__process, machine, false);
+						perf_event__process, machine,
+						true, false);
 	if (ret < 0) {
 		pr_debug("perf_event__synthesize_thread_map failed\n");
 		goto out_err;

tools/perf/tests/mmap-thread-lookup.c

···
 {
 	return perf_event__synthesize_threads(NULL,
 					      perf_event__process,
-					      machine, 0, 1);
+					      machine, 1, 0, 1);
 }
 
 static int synth_process(struct machine *machine)
···
 
 	err = perf_event__synthesize_thread_map(NULL, map,
 						perf_event__process,
-						machine, 0);
+						machine, 1, 0);
 
 	perf_thread_map__put(map);
 	return err;

tools/perf/util/synthetic-events.c

···
 				      union perf_event *fork_event,
 				      union perf_event *namespaces_event,
 				      pid_t pid, int full, perf_event__handler_t process,
-				      struct perf_tool *tool, struct machine *machine, bool mmap_data)
+				      struct perf_tool *tool, struct machine *machine,
+				      bool needs_mmap, bool mmap_data)
 {
 	char filename[PATH_MAX];
 	struct dirent **dirent;
···
 		 * send mmap only for thread group leader
 		 * see thread__init_maps()
 		 */
-		if (pid == tgid &&
+		if (pid == tgid && needs_mmap &&
 		    perf_event__synthesize_mmap_events(tool, mmap_event, pid, tgid,
 						       process, machine, mmap_data))
 			return -1;
···
 			break;
 
 		rc = 0;
-		if (_pid == pid && !kernel_thread) {
+		if (_pid == pid && !kernel_thread && needs_mmap) {
 			/* process the parent's maps too */
 			rc = perf_event__synthesize_mmap_events(tool, mmap_event, pid, tgid,
 						process, machine, mmap_data);
···
 				      struct perf_thread_map *threads,
 				      perf_event__handler_t process,
 				      struct machine *machine,
-				      bool mmap_data)
+				      bool needs_mmap, bool mmap_data)
 {
 	union perf_event *comm_event, *mmap_event, *fork_event;
 	union perf_event *namespaces_event;
···
 					       fork_event, namespaces_event,
 					       perf_thread_map__pid(threads, thread), 0,
 					       process, tool, machine,
-					       mmap_data)) {
+					       needs_mmap, mmap_data)) {
 			err = -1;
 			break;
 		}
···
 						       fork_event, namespaces_event,
 						       comm_event->comm.pid, 0,
 						       process, tool, machine,
-						       mmap_data)) {
+						       needs_mmap, mmap_data)) {
 				err = -1;
 				break;
 			}
···
 static int __perf_event__synthesize_threads(struct perf_tool *tool,
 					    perf_event__handler_t process,
 					    struct machine *machine,
+					    bool needs_mmap,
 					    bool mmap_data,
 					    struct dirent **dirent,
 					    int start,
···
 		 */
 		__event__synthesize_thread(comm_event, mmap_event, fork_event,
 					   namespaces_event, pid, 1, process,
-					   tool, machine, mmap_data);
+					   tool, machine, needs_mmap, mmap_data);
 	}
 	err = 0;
 
···
 	struct perf_tool *tool;
 	perf_event__handler_t process;
 	struct machine *machine;
+	bool needs_mmap;
 	bool mmap_data;
 	struct dirent **dirent;
 	int num;
···
 	struct synthesize_threads_arg *args = arg;
 
 	__perf_event__synthesize_threads(args->tool, args->process,
-					 args->machine, args->mmap_data,
+					 args->machine,
+					 args->needs_mmap, args->mmap_data,
 					 args->dirent,
 					 args->start, args->num);
 	return NULL;
···
 int perf_event__synthesize_threads(struct perf_tool *tool,
 				   perf_event__handler_t process,
 				   struct machine *machine,
-				   bool mmap_data,
+				   bool needs_mmap, bool mmap_data,
 				   unsigned int nr_threads_synthesize)
 {
 	struct synthesize_threads_arg *args = NULL;
···
 
 	if (thread_nr <= 1) {
 		err = __perf_event__synthesize_threads(tool, process,
-						       machine, mmap_data,
+						       machine,
+						       needs_mmap, mmap_data,
 						       dirent, base, n);
 		goto free_dirent;
 	}
···
 		args[i].tool = tool;
 		args[i].process = process;
 		args[i].machine = machine;
+		args[i].needs_mmap = needs_mmap;
 		args[i].mmap_data = mmap_data;
 		args[i].dirent = dirent;
 	}
···
 
 int __machine__synthesize_threads(struct machine *machine, struct perf_tool *tool,
 				  struct target *target, struct perf_thread_map *threads,
-				  perf_event__handler_t process, bool data_mmap,
-				  unsigned int nr_threads_synthesize)
+				  perf_event__handler_t process, bool needs_mmap,
+				  bool data_mmap, unsigned int nr_threads_synthesize)
 {
 	if (target__has_task(target))
-		return perf_event__synthesize_thread_map(tool, threads, process, machine, data_mmap);
+		return perf_event__synthesize_thread_map(tool, threads, process, machine,
+							 needs_mmap, data_mmap);
 	else if (target__has_cpu(target))
-		return perf_event__synthesize_threads(tool, process,
-						      machine, data_mmap,
+		return perf_event__synthesize_threads(tool, process, machine,
+						      needs_mmap, data_mmap,
 						      nr_threads_synthesize);
 	/* command specified */
 	return 0;
 }
 
 int machine__synthesize_threads(struct machine *machine, struct target *target,
-				struct perf_thread_map *threads, bool data_mmap,
-				unsigned int nr_threads_synthesize)
+				struct perf_thread_map *threads, bool needs_mmap,
+				bool data_mmap, unsigned int nr_threads_synthesize)
 {
 	return __machine__synthesize_threads(machine, NULL, target, threads,
-					     perf_event__process, data_mmap,
-					     nr_threads_synthesize);
+					     perf_event__process, needs_mmap,
+					     data_mmap, nr_threads_synthesize);
 }
 
 static struct perf_record_event_update *event_update_event__new(size_t size, u64 type, u64 id)

tools/perf/util/synthetic-events.h

···
 int perf_event__synthesize_stat_round(struct perf_tool *tool, u64 time, u64 type, perf_event__handler_t process, struct machine *machine);
 int perf_event__synthesize_stat(struct perf_tool *tool, u32 cpu, u32 thread, u64 id, struct perf_counts_values *count, perf_event__handler_t process, struct machine *machine);
 int perf_event__synthesize_thread_map2(struct perf_tool *tool, struct perf_thread_map *threads, perf_event__handler_t process, struct machine *machine);
-int perf_event__synthesize_thread_map(struct perf_tool *tool, struct perf_thread_map *threads, perf_event__handler_t process, struct machine *machine, bool mmap_data);
-int perf_event__synthesize_threads(struct perf_tool *tool, perf_event__handler_t process, struct machine *machine, bool mmap_data, unsigned int nr_threads_synthesize);
+int perf_event__synthesize_thread_map(struct perf_tool *tool, struct perf_thread_map *threads, perf_event__handler_t process, struct machine *machine, bool needs_mmap, bool mmap_data);
+int perf_event__synthesize_threads(struct perf_tool *tool, perf_event__handler_t process, struct machine *machine, bool needs_mmap, bool mmap_data, unsigned int nr_threads_synthesize);
 int perf_event__synthesize_tracing_data(struct perf_tool *tool, int fd, struct evlist *evlist, perf_event__handler_t process);
 int perf_event__synth_time_conv(const struct perf_event_mmap_page *pc, struct perf_tool *tool, perf_event__handler_t process, struct machine *machine);
 pid_t perf_event__synthesize_comm(struct perf_tool *tool, union perf_event *event, pid_t pid, perf_event__handler_t process, struct machine *machine);
···
 
 int __machine__synthesize_threads(struct machine *machine, struct perf_tool *tool,
 				  struct target *target, struct perf_thread_map *threads,
-				  perf_event__handler_t process, bool data_mmap,
+				  perf_event__handler_t process, bool needs_mmap, bool data_mmap,
 				  unsigned int nr_threads_synthesize);
 int machine__synthesize_threads(struct machine *machine, struct target *target,
-				struct perf_thread_map *threads, bool data_mmap,
+				struct perf_thread_map *threads, bool needs_mmap, bool data_mmap,
 				unsigned int nr_threads_synthesize);
 
 #ifdef HAVE_AUXTRACE_SUPPORT