6 files changed, 89 insertions, 14 deletions

diff --git a/tools/perf/util/event.c b/tools/perf/util/event.c
index 9d12aa6dd485..dbcaea1a8180 100644
--- a/tools/perf/util/event.c
+++ b/tools/perf/util/event.c
@@ -699,7 +699,7 @@ void thread__find_addr_map(struct thread *thread,
 			   enum map_type type, u64 addr,
 			   struct addr_location *al)
 {
-	struct map_groups *mg = &thread->mg;
+	struct map_groups *mg = thread->mg;
 	bool load_map = false;
 
 	al->machine = machine;
diff --git a/tools/perf/util/machine.c b/tools/perf/util/machine.c
index a53cd0b8c151..98ec56dc890b 100644
--- a/tools/perf/util/machine.c
+++ b/tools/perf/util/machine.c
@@ -316,6 +316,17 @@ static struct thread *__machine__findnew_thread(struct machine *machine,
 		rb_link_node(&th->rb_node, parent, p);
 		rb_insert_color(&th->rb_node, &machine->threads);
 		machine->last_match = th;
+
+		/*
+		 * We have to initialize map_groups separately
+		 * after rb tree is updated.
+		 *
+		 * The reason is that we call machine__findnew_thread
+		 * within thread__init_map_groups to find the thread
+		 * leader and that would screwed the rb tree.
+		 */
+		if (thread__init_map_groups(th, machine))
+			return NULL;
 	}
 
 	return th;
diff --git a/tools/perf/util/map.c b/tools/perf/util/map.c
index 39cd2d0faff6..ba5f5c0c838b 100644
--- a/tools/perf/util/map.c
+++ b/tools/perf/util/map.c
@@ -323,6 +323,7 @@ void map_groups__init(struct map_groups *mg)
 		INIT_LIST_HEAD(&mg->removed_maps[i]);
 	}
 	mg->machine = NULL;
+	mg->refcnt = 1;
 }
 
 static void maps__delete(struct rb_root *maps)
@@ -358,6 +359,28 @@ void map_groups__exit(struct map_groups *mg)
 	}
 }
 
+struct map_groups *map_groups__new(void)
+{
+	struct map_groups *mg = malloc(sizeof(*mg));
+
+	if (mg != NULL)
+		map_groups__init(mg);
+
+	return mg;
+}
+
+void map_groups__delete(struct map_groups *mg)
+{
+	map_groups__exit(mg);
+	free(mg);
+}
+
+void map_groups__put(struct map_groups *mg)
+{
+	if (--mg->refcnt == 0)
+		map_groups__delete(mg);
+}
+
 void map_groups__flush(struct map_groups *mg)
 {
 	int type;
diff --git a/tools/perf/util/map.h b/tools/perf/util/map.h
index f00f058afb3b..d6445b27d672 100644
--- a/tools/perf/util/map.h
+++ b/tools/perf/util/map.h
@@ -59,8 +59,20 @@ struct map_groups {
 	struct rb_root	 maps[MAP__NR_TYPES];
 	struct list_head removed_maps[MAP__NR_TYPES];
 	struct machine	 *machine;
+	int		 refcnt;
 };
 
+struct map_groups *map_groups__new(void);
+void map_groups__delete(struct map_groups *mg);
+
+static inline struct map_groups *map_groups__get(struct map_groups *mg)
+{
+	++mg->refcnt;
+	return mg;
+}
+
+void map_groups__put(struct map_groups *mg);
+
 static inline struct kmap *map__kmap(struct map *map)
 {
 	return (struct kmap *)(map + 1);
diff --git a/tools/perf/util/thread.c b/tools/perf/util/thread.c
index 3ce0498bdae6..2fde0d5e40b5 100644
--- a/tools/perf/util/thread.c
+++ b/tools/perf/util/thread.c
@@ -8,6 +8,22 @@
 #include "debug.h"
 #include "comm.h"
 
+int thread__init_map_groups(struct thread *thread, struct machine *machine)
+{
+	struct thread *leader;
+	pid_t pid = thread->pid_;
+
+	if (pid == thread->tid) {
+		thread->mg = map_groups__new();
+	} else {
+		leader = machine__findnew_thread(machine, pid, pid);
+		if (leader)
+			thread->mg = map_groups__get(leader->mg);
+	}
+
+	return thread->mg ? 0 : -1;
+}
+
 struct thread *thread__new(pid_t pid, pid_t tid)
 {
 	char *comm_str;
@@ -15,7 +31,6 @@ struct thread *thread__new(pid_t pid, pid_t tid)
 	struct thread *thread = zalloc(sizeof(*thread));
 
 	if (thread != NULL) {
-		map_groups__init(&thread->mg);
 		thread->pid_ = pid;
 		thread->tid = tid;
 		thread->ppid = -1;
@@ -45,7 +60,8 @@ void thread__delete(struct thread *thread)
 {
 	struct comm *comm, *tmp;
 
-	map_groups__exit(&thread->mg);
+	map_groups__put(thread->mg);
+	thread->mg = NULL;
 	list_for_each_entry_safe(comm, tmp, &thread->comm_list, list) {
 		list_del(&comm->list);
 		comm__free(comm);
@@ -111,18 +127,35 @@ int thread__comm_len(struct thread *thread)
 size_t thread__fprintf(struct thread *thread, FILE *fp)
 {
 	return fprintf(fp, "Thread %d %s\n", thread->tid, thread__comm_str(thread)) +
-	       map_groups__fprintf(&thread->mg, verbose, fp);
+	       map_groups__fprintf(thread->mg, verbose, fp);
 }
 
 void thread__insert_map(struct thread *thread, struct map *map)
 {
-	map_groups__fixup_overlappings(&thread->mg, map, verbose, stderr);
-	map_groups__insert(&thread->mg, map);
+	map_groups__fixup_overlappings(thread->mg, map, verbose, stderr);
+	map_groups__insert(thread->mg, map);
+}
+
+static int thread__clone_map_groups(struct thread *thread,
+				    struct thread *parent)
+{
+	int i;
+
+	/* This is new thread, we share map groups for process. */
+	if (thread->pid_ == parent->pid_)
+		return 0;
+
+	/* But this one is new process, copy maps. */
+	for (i = 0; i < MAP__NR_TYPES; ++i)
+		if (map_groups__clone(thread->mg, parent->mg, i) < 0)
+			return -ENOMEM;
+
+	return 0;
 }
 
 int thread__fork(struct thread *thread, struct thread *parent, u64 timestamp)
 {
-	int i, err;
+	int err;
 
 	if (parent->comm_set) {
 		const char *comm = thread__comm_str(parent);
@@ -134,13 +167,8 @@ int thread__fork(struct thread *thread, struct thread *parent, u64 timestamp)
 		thread->comm_set = true;
 	}
 
-	for (i = 0; i < MAP__NR_TYPES; ++i)
-		if (map_groups__clone(&thread->mg, &parent->mg, i) < 0)
-			return -ENOMEM;
-
 	thread->ppid = parent->tid;
-
-	return 0;
+	return thread__clone_map_groups(thread, parent);
 }
 
 void thread__find_cpumode_addr_location(struct thread *thread,
diff --git a/tools/perf/util/thread.h b/tools/perf/util/thread.h
index 9b29f085aede..3c0c2724f82c 100644
--- a/tools/perf/util/thread.h
+++ b/tools/perf/util/thread.h
@@ -13,7 +13,7 @@ struct thread {
 		struct rb_node	 rb_node;
 		struct list_head node;
 	};
-	struct map_groups	mg;
+	struct map_groups	*mg;
 	pid_t			pid_; /* Not all tools update this */
 	pid_t			tid;
 	pid_t			ppid;
@@ -30,6 +30,7 @@ struct machine;
 struct comm;
 
 struct thread *thread__new(pid_t pid, pid_t tid);
+int thread__init_map_groups(struct thread *thread, struct machine *machine);
 void thread__delete(struct thread *thread);
 static inline void thread__exited(struct thread *thread)
 {