perf machine: Protect the machine->threads with a rwlock

In addition to using refcounts for the struct thread lifetime
management, we need to protect access to machine->threads from
concurrent access.

That happens in 'perf top', where a thread processes events, inserting
and deleting entries from that rb_tree while another thread decays
hist_entries, that end up dropping references and ultimately deleting
threads from the rb_tree and releasing its resources when no further
hist_entry (or other data structures, like in 'perf sched') references
it.

So the rule is the same for refcounts + protected trees in the kernel,
get the tree lock, find object, bump the refcount, drop the tree lock,
return, use object, drop the refcount if no more use of it is needed,
keep it if storing it in some other data structure, drop when releasing
that data structure.

I.e. pair "t = machine__find(new)_thread()" with a "thread__put(t)", and
"perf_event__preprocess_sample(&al)" with "addr_location__put(&al)".

The addr_location__put() one is because as we return references to
several data structures, we may end up adding more reference counting
for the other data structures and then we'll drop it at
addr_location__put() time.

Acked-by: David Ahern <dsahern@gmail.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: Don Zickus <dzickus@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/n/tip-bs9rt4n0jw3hi9f3zxyy3xln@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>

9 files changed, 39 insertions, 19 deletions

diff --git a/tools/perf/tests/code-reading.c b/tools/perf/tests/code-reading.c
index ca0e480e741b..e2a432b67d52 100644
--- a/tools/perf/tests/code-reading.c
+++ b/tools/perf/tests/code-reading.c
@@ -248,6 +248,7 @@ static int process_sample_event(struct machine *machine,
 	struct perf_sample sample;
 	struct thread *thread;
 	u8 cpumode;
+	int ret;
 
 	if (perf_evlist__parse_sample(evlist, event, &sample)) {
 		pr_debug("perf_evlist__parse_sample failed\n");
@@ -262,7 +263,9 @@ static int process_sample_event(struct machine *machine,
 
 	cpumode = event->header.misc & PERF_RECORD_MISC_CPUMODE_MASK;
 
-	return read_object_code(sample.ip, READLEN, cpumode, thread, state);
+	ret = read_object_code(sample.ip, READLEN, cpumode, thread, state);
+	thread__put(thread);
+	return ret;
 }
 
 static int process_event(struct machine *machine, struct perf_evlist *evlist,
@@ -457,13 +460,13 @@ static int do_test_code_reading(bool try_kcore)
 	thread = machine__findnew_thread(machine, pid, pid);
 	if (!thread) {
 		pr_debug("machine__findnew_thread failed\n");
-		goto out_err;
+		goto out_put;
 	}
 
 	cpus = cpu_map__new(NULL);
 	if (!cpus) {
 		pr_debug("cpu_map__new failed\n");
-		goto out_err;
+		goto out_put;
 	}
 
 	while (1) {
@@ -472,7 +475,7 @@ static int do_test_code_reading(bool try_kcore)
 		evlist = perf_evlist__new();
 		if (!evlist) {
 			pr_debug("perf_evlist__new failed\n");
-			goto out_err;
+			goto out_put;
 		}
 
 		perf_evlist__set_maps(evlist, cpus, threads);
@@ -485,7 +488,7 @@ static int do_test_code_reading(bool try_kcore)
 		ret = parse_events(evlist, str, NULL);
 		if (ret < 0) {
 			pr_debug("parse_events failed\n");
-			goto out_err;
+			goto out_put;
 		}
 
 		perf_evlist__config(evlist, &opts);
@@ -506,7 +509,7 @@ static int do_test_code_reading(bool try_kcore)
 				continue;
 			}
 			pr_debug("perf_evlist__open failed\n");
-			goto out_err;
+			goto out_put;
 		}
 		break;
 	}
@@ -514,7 +517,7 @@ static int do_test_code_reading(bool try_kcore)
 	ret = perf_evlist__mmap(evlist, UINT_MAX, false);
 	if (ret < 0) {
 		pr_debug("perf_evlist__mmap failed\n");
-		goto out_err;
+		goto out_put;
 	}
 
 	perf_evlist__enable(evlist);
@@ -525,7 +528,7 @@ static int do_test_code_reading(bool try_kcore)
 
 	ret = process_events(machine, evlist, &state);
 	if (ret < 0)
-		goto out_err;
+		goto out_put;
 
 	if (!have_vmlinux && !have_kcore && !try_kcore)
 		err = TEST_CODE_READING_NO_KERNEL_OBJ;
@@ -535,7 +538,10 @@ static int do_test_code_reading(bool try_kcore)
 		err = TEST_CODE_READING_NO_ACCESS;
 	else
 		err = TEST_CODE_READING_OK;
+out_put:
+	thread__put(thread);
 out_err:
+
 	if (evlist) {
 		perf_evlist__delete(evlist);
 	} else {
diff --git a/tools/perf/tests/dwarf-unwind.c b/tools/perf/tests/dwarf-unwind.c
index 0bf06bec68c7..9b748e1ad46e 100644
--- a/tools/perf/tests/dwarf-unwind.c
+++ b/tools/perf/tests/dwarf-unwind.c
@@ -170,6 +170,7 @@ int test__dwarf_unwind(void)
 	}
 
 	err = krava_1(thread);
+	thread__put(thread);
 
  out:
 	machine__delete_threads(machine);
diff --git a/tools/perf/tests/hists_common.c b/tools/perf/tests/hists_common.c
index a62c09134516..456f884eb27b 100644
--- a/tools/perf/tests/hists_common.c
+++ b/tools/perf/tests/hists_common.c
@@ -96,6 +96,7 @@ struct machine *setup_fake_machine(struct machines *machines)
 			goto out;
 
 		thread__set_comm(thread, fake_threads[i].comm, 0);
+		thread__put(thread);
 	}
 
 	for (i = 0; i < ARRAY_SIZE(fake_mmap_info); i++) {
diff --git a/tools/perf/tests/hists_cumulate.c b/tools/perf/tests/hists_cumulate.c
index b08a95a5ca1a..620f626e5b35 100644
--- a/tools/perf/tests/hists_cumulate.c
+++ b/tools/perf/tests/hists_cumulate.c
@@ -105,8 +105,10 @@ static int add_hist_entries(struct hists *hists, struct machine *machine)
 			goto out;
 
 		if (hist_entry_iter__add(&iter, &al, evsel, &sample,
-					 PERF_MAX_STACK_DEPTH, NULL) < 0)
+					 PERF_MAX_STACK_DEPTH, NULL) < 0) {
+			addr_location__put(&al);
 			goto out;
+		}
 
 		fake_samples[i].thread = al.thread;
 		fake_samples[i].map = al.map;
diff --git a/tools/perf/tests/hists_filter.c b/tools/perf/tests/hists_filter.c
index 108488cd71fa..82e1ee52e024 100644
--- a/tools/perf/tests/hists_filter.c
+++ b/tools/perf/tests/hists_filter.c
@@ -82,8 +82,10 @@ static int add_hist_entries(struct perf_evlist *evlist,
 				goto out;
 
 			if (hist_entry_iter__add(&iter, &al, evsel, &sample,
-						 PERF_MAX_STACK_DEPTH, NULL) < 0)
+						 PERF_MAX_STACK_DEPTH, NULL) < 0) {
+				addr_location__put(&al);
 				goto out;
+			}
 
 			fake_samples[i].thread = al.thread;
 			fake_samples[i].map = al.map;
diff --git a/tools/perf/tests/hists_link.c b/tools/perf/tests/hists_link.c
index 34c61e4d3352..8c102b011424 100644
--- a/tools/perf/tests/hists_link.c
+++ b/tools/perf/tests/hists_link.c
@@ -91,8 +91,10 @@ static int add_hist_entries(struct perf_evlist *evlist, struct machine *machine)
 
 			he = __hists__add_entry(hists, &al, NULL,
 						NULL, NULL, 1, 1, 0, true);
-			if (he == NULL)
+			if (he == NULL) {
+				addr_location__put(&al);
 				goto out;
+			}
 
 			fake_common_samples[k].thread = al.thread;
 			fake_common_samples[k].map = al.map;
@@ -115,8 +117,10 @@ static int add_hist_entries(struct perf_evlist *evlist, struct machine *machine)
 
 			he = __hists__add_entry(hists, &al, NULL,
 						NULL, NULL, 1, 1, 0, true);
-			if (he == NULL)
+			if (he == NULL) {
+				addr_location__put(&al);
 				goto out;
+			}
 
 			fake_samples[i][k].thread = al.thread;
 			fake_samples[i][k].map = al.map;
diff --git a/tools/perf/tests/hists_output.c b/tools/perf/tests/hists_output.c
index d8a23db80094..fd7ec4f9aeb4 100644
--- a/tools/perf/tests/hists_output.c
+++ b/tools/perf/tests/hists_output.c
@@ -71,8 +71,10 @@ static int add_hist_entries(struct hists *hists, struct machine *machine)
 			goto out;
 
 		if (hist_entry_iter__add(&iter, &al, evsel, &sample,
-					 PERF_MAX_STACK_DEPTH, NULL) < 0)
+					 PERF_MAX_STACK_DEPTH, NULL) < 0) {
+			addr_location__put(&al);
 			goto out;
+		}
 
 		fake_samples[i].thread = al.thread;
 		fake_samples[i].map = al.map;
diff --git a/tools/perf/tests/mmap-thread-lookup.c b/tools/perf/tests/mmap-thread-lookup.c
index 2113f1c8611f..264e215c0d36 100644
--- a/tools/perf/tests/mmap-thread-lookup.c
+++ b/tools/perf/tests/mmap-thread-lookup.c
@@ -191,6 +191,8 @@ static int mmap_events(synth_cb synth)
 				      PERF_RECORD_MISC_USER, MAP__FUNCTION,
 				      (unsigned long) (td->map + 1), &al);
 
+		thread__put(thread);
+
 		if (!al.map) {
 			pr_debug("failed, couldn't find map\n");
 			err = -1;
diff --git a/tools/perf/tests/thread-mg-share.c b/tools/perf/tests/thread-mg-share.c
index b028499dd3cf..dc05bd62b4a3 100644
--- a/tools/perf/tests/thread-mg-share.c
+++ b/tools/perf/tests/thread-mg-share.c
@@ -64,22 +64,22 @@ int test__thread_mg_share(void)
 	TEST_ASSERT_VAL("map groups don't match", other_mg == other_leader->mg);
 
 	/* release thread group */
-	thread__delete(leader);
+	thread__put(leader);
 	TEST_ASSERT_VAL("wrong refcnt", mg->refcnt == 3);
 
-	thread__delete(t1);
+	thread__put(t1);
 	TEST_ASSERT_VAL("wrong refcnt", mg->refcnt == 2);
 
-	thread__delete(t2);
+	thread__put(t2);
 	TEST_ASSERT_VAL("wrong refcnt", mg->refcnt == 1);
 
-	thread__delete(t3);
+	thread__put(t3);
 
 	/* release other group  */
-	thread__delete(other_leader);
+	thread__put(other_leader);
 	TEST_ASSERT_VAL("wrong refcnt", other_mg->refcnt == 1);
 
-	thread__delete(other);
+	thread__put(other);
 
 	/*
 	 * Cannot call machine__delete_threads(machine) now,