perf inject: Lazy build-id mmap2 event insertion

Add -B option that lazily inserts mmap2 events thereby dropping all
mmap events without samples. This is similar to the behavior of -b
where only build_id events are inserted when a dso is accessed in a
sample.

File size savings can be significant in system-wide mode, consider:

$ perf record -g -a -o perf.data sleep 1
$ perf inject -B -i perf.data -o perf.new.data
$ ls -al perf.data perf.new.data
5147049 perf.data
2248493 perf.new.data

Give test coverage of the new option in pipe test.

Signed-off-by: Ian Rogers <[e-mail]>
Acked-by: Namhyung Kim <[e-mail]>
Cc: Adrian Hunter <[e-mail]>
Cc: Alexander Shishkin <[e-mail]>
Cc: Anne Macedo <[e-mail]>
Cc: Casey Chen <[e-mail]>
Cc: Colin Ian King <[e-mail]>
Cc: Ingo Molnar <[e-mail]>
Cc: Jiri Olsa <[e-mail]>
Cc: Kan Liang <[e-mail]>
Cc: Mark Rutland <[e-mail]>
Cc: Peter Zijlstra <[e-mail]>
Cc: Sun Haiyong <[e-mail]>
Link: https://lore.kernel.org/r/[e-mail]
Signed-off-by: Arnaldo Carvalho de Melo <[e-mail]>

tools/perf/builtin-inject.c

···
 	BID_RWS__INJECT_HEADER_LAZY,
 	BID_RWS__INJECT_HEADER_ALL,
 	BID_RWS__MMAP2_BUILDID_ALL,
+	BID_RWS__MMAP2_BUILDID_LAZY,
 };
 
 struct perf_inject {
···
 		 * Remember the evsel for lazy build id generation. It is used
 		 * for the sample id header type.
 		 */
-		if (inject->build_id_style == BID_RWS__INJECT_HEADER_LAZY &&
+		if ((inject->build_id_style == BID_RWS__INJECT_HEADER_LAZY ||
+		     inject->build_id_style == BID_RWS__MMAP2_BUILDID_LAZY) &&
 		    !inject->mmap_evsel)
 			inject->mmap_evsel = evlist__event2evsel(inject->session->evlist, event);
 
···
 		}
 	}
 	dso__put(dso);
+	if (inject->build_id_style == BID_RWS__MMAP2_BUILDID_LAZY)
+		return 0;
+
 	return perf_event__repipe(tool, event, sample, machine);
 }
 
···
 	return 0;
 }
 
-static int mark_dso_hit(const struct perf_tool *tool,
+static int mark_dso_hit(const struct perf_inject *inject,
+			const struct perf_tool *tool,
 			struct perf_sample *sample,
 			struct machine *machine,
 			const struct evsel *mmap_evsel,
···
 		}
 	}
 	dso = map__dso(map);
-	if (dso && !dso__hit(dso)) {
-		dso__set_hit(dso);
-		tool__inject_build_id(tool, sample, machine,
-				      mmap_evsel, misc, dso__long_name(dso), dso,
-				      map__flags(map));
+	if (inject->build_id_style == BID_RWS__INJECT_HEADER_LAZY) {
+		if (dso && !dso__hit(dso)) {
+			dso__set_hit(dso);
+			tool__inject_build_id(tool, sample, machine,
+					     mmap_evsel, misc, dso__long_name(dso), dso,
+					     map__flags(map));
+		}
+	} else if (inject->build_id_style == BID_RWS__MMAP2_BUILDID_LAZY) {
+		if (!map__hit(map)) {
+			const struct build_id null_bid = { .size = 0 };
+			const struct build_id *bid = dso ? dso__bid(dso) : &null_bid;
+			const char *filename = dso ? dso__long_name(dso) : "";
+
+			map__set_hit(map);
+			perf_event__synthesize_mmap2_build_id(tool, sample, machine,
+								perf_event__repipe,
+								mmap_evsel,
+								misc,
+								sample->pid, sample->tid,
+								map__start(map),
+								map__end(map) - map__start(map),
+								map__pgoff(map),
+								bid,
+								map__prot(map),
+								map__flags(map),
+								filename);
+		}
 	}
 	return 0;
 }
 
 struct mark_dso_hit_args {
+	const struct perf_inject *inject;
 	const struct perf_tool *tool;
 	struct perf_sample *sample;
 	struct machine *machine;
···
 	struct mark_dso_hit_args *args = data;
 	struct map *map = node->ms.map;
 
-	return mark_dso_hit(args->tool, args->sample, args->machine,
+	return mark_dso_hit(args->inject, args->tool, args->sample, args->machine,
 			    args->mmap_evsel, map, /*sample_in_dso=*/false);
 }
 
···
 	struct thread *thread;
 	struct perf_inject *inject = container_of(tool, struct perf_inject, tool);
 	struct mark_dso_hit_args args = {
+		.inject = inject,
 		.tool = tool,
 		/*
 		 * Use the parsed sample data of the sample event, which will
···
 	}
 
 	if (thread__find_map(thread, sample->cpumode, sample->ip, &al)) {
-		mark_dso_hit(tool, sample, machine, args.mmap_evsel, al.map,
+		mark_dso_hit(inject, tool, sample, machine, args.mmap_evsel, al.map,
 			     /*sample_in_dso=*/true);
 	}
 
···
 #endif
 	}
 
-	if (inject->build_id_style == BID_RWS__INJECT_HEADER_LAZY) {
+	if (inject->build_id_style == BID_RWS__INJECT_HEADER_LAZY ||
+	    inject->build_id_style == BID_RWS__MMAP2_BUILDID_LAZY) {
 		inject->tool.sample = perf_event__inject_buildid;
 	} else if (inject->sched_stat) {
 		struct evsel *evsel;
···
 	const char *known_build_ids = NULL;
 	bool build_ids;
 	bool build_id_all;
+	bool mmap2_build_ids;
 	bool mmap2_build_id_all;
 
 	struct option options[] = {
···
 			    "Inject build-ids into the output stream"),
 		OPT_BOOLEAN(0, "buildid-all", &build_id_all,
 			    "Inject build-ids of all DSOs into the output stream"),
+		OPT_BOOLEAN('B', "mmap2-buildids", &mmap2_build_ids,
+			    "Drop unused mmap events, make others mmap2 with build IDs"),
 		OPT_BOOLEAN(0, "mmap2-buildid-all", &mmap2_build_id_all,
 			    "Rewrite all mmap events as mmap2 events with build IDs"),
 		OPT_STRING(0, "known-build-ids", &known_build_ids,
···
 			return -1;
 		}
 	}
+	if (mmap2_build_ids)
+		inject.build_id_style = BID_RWS__MMAP2_BUILDID_LAZY;
 	if (mmap2_build_id_all)
 		inject.build_id_style = BID_RWS__MMAP2_BUILDID_ALL;
 	if (build_ids)
···
 	data.path = inject.input_name;
 
 	ordered_events = inject.jit_mode || inject.sched_stat ||
-		(inject.build_id_style == BID_RWS__INJECT_HEADER_LAZY);
+		inject.build_id_style == BID_RWS__INJECT_HEADER_LAZY ||
+		inject.build_id_style == BID_RWS__MMAP2_BUILDID_LAZY;
 	perf_tool__init(&inject.tool, ordered_events);
 	inject.tool.sample		= perf_event__repipe_sample;
 	inject.tool.read		= perf_event__repipe_sample;
···
 		}
 	}
 
-	if (inject.build_id_style == BID_RWS__INJECT_HEADER_LAZY) {
+	if (inject.build_id_style == BID_RWS__INJECT_HEADER_LAZY ||
+	    inject.build_id_style == BID_RWS__MMAP2_BUILDID_LAZY) {
 		/*
 		 * to make sure the mmap records are ordered correctly
 		 * and so that the correct especially due to jitted code

tools/perf/tests/shell/pipe_test.sh

···
 }
 
 test_record_report
+test_inject_bids -B
 test_inject_bids -b
 test_inject_bids --buildid-all
 test_inject_bids --mmap2-buildid-all

tools/perf/util/map.c

···
 	map__set_mapping_type(map, MAPPING_TYPE__DSO);
 	assert(map__erange_warned(map) == false);
 	assert(map__priv(map) == false);
+	assert(map__hit(map) == false);
 }
 
 struct map *map__new(struct machine *machine, u64 start, u64 len,

tools/perf/util/map.h

···
 	enum mapping_type	mapping_type:8;
 	bool			erange_warned;
 	bool			priv;
+	bool			hit;
 };
 
 struct kmap;
···
 static inline bool map__priv(const struct map *map)
 {
 	return RC_CHK_ACCESS(map)->priv;
+}
+
+static inline bool map__hit(const struct map *map)
+{
+	return RC_CHK_ACCESS(map)->hit;
 }
 
 static inline refcount_t *map__refcnt(struct map *map)
···
 static inline void map__set_priv(struct map *map)
 {
 	RC_CHK_ACCESS(map)->priv = true;
+}
+
+static inline void map__set_hit(struct map *map)
+{
+	RC_CHK_ACCESS(map)->hit = true;
 }
 
 static inline void map__set_erange_warned(struct map *map)