perf parse-event: Release cpu_map if evsel alloc failed
Namhyung Kim
maintainerPATCH v2 1/2 perf parse-event: Release cpu_map if evsel alloc failed
--- tools/perf/util/parse-events.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)
tools/perf/util/parse-events.c +3 -1
@@ -360,8 +360,10 @@ __add_event(struct list_head *list, int *idx,event_attr_init (attr); evsel =evsel__new_idx (attr, *idx); -if (!evsel) +if (!evsel) { +perf_cpu_map__put (cpus);return NULL ; + } (*idx)++; evsel->core.cpus =perf_cpu_map__get (cpus);
Signature · 1 line
2.28.0.618.gf4bc123cb7-goog
Namhyung Kim
maintainerPATCH v2 2/2 perf parse-event: Fix cpu map leaks
Like evlist cpu map, evsel's cpu map should have a proper refcount. As it's created with a refcount, we don't need to get an extra count. Thanks to Arnaldo for the simpler suggestion.
This fixes the following ASAN report:
Direct leak of 840 byte(s) in 70 object(s) allocated from:
#0 0x7fe36703f628 in malloc (/lib/x86_64-linux-gnu/libasan.so.5+0x107628)
#1 0x559fbbf611ca in cpu_map__trim_new /home/namhyung/project/linux/tools/lib/perf/cpumap.c:79
#2 0x559fbbf6229c in perf_cpu_map__new /home/namhyung/project/linux/tools/lib/perf/cpumap.c:237
#3 0x559fbbcc6c6d in __add_event util/parse-events.c:357
#4 0x559fbbcc6c6d in add_event_tool util/parse-events.c:408
#5 0x559fbbcc6c6d in parse_events_add_tool util/parse-events.c:1414
#6 0x559fbbd8474d in parse_events_parse util/parse-events.y:439
#7 0x559fbbcc95da in parse_events__scanner util/parse-events.c:2096
#8 0x559fbbcc95da in __parse_events util/parse-events.c:2141
#9 0x559fbbc2788b in check_parse_id tests/pmu-events.c:406
#10 0x559fbbc2788b in check_parse_id tests/pmu-events.c:393
#11 0x559fbbc2788b in check_parse_fake tests/pmu-events.c:436
#12 0x559fbbc2788b in metric_parse_fake tests/pmu-events.c:553
#13 0x559fbbc27e2d in test_parsing_fake tests/pmu-events.c:599
#14 0x559fbbc27e2d in test_parsing_fake tests/pmu-events.c:574
#15 0x559fbbc0109b in run_test tests/builtin-test.c:410
#16 0x559fbbc0109b in test_and_print tests/builtin-test.c:440
#17 0x559fbbc03e69 in __cmd_test tests/builtin-test.c:695
#18 0x559fbbc03e69 in cmd_test tests/builtin-test.c:807
#19 0x559fbbc691f4 in run_builtin /home/namhyung/project/linux/tools/perf/perf.c:312
#20 0x559fbbb071a8 in handle_internal_command /home/namhyung/project/linux/tools/perf/perf.c:364
#21 0x559fbbb071a8 in run_argv /home/namhyung/project/linux/tools/perf/perf.c:408
#22 0x559fbbb071a8 in main /home/namhyung/project/linux/tools/perf/perf.c:538
#23 0x7fe366b68cc9 in __libc_start_main ../csu/libc-start.c:308And I've failed which commit introduced this bug as the code was heavily changed since then. ;-/
--- tools/perf/util/parse-events.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
tools/perf/util/parse-events.c +2 -2
@@ -353,7 +353,7 @@ __add_event(struct list_head *list, int *idx, const char *cpu_list) {struct evsel *evsel; -struct perf_cpu_map *cpus = pmu ? pmu->cpus : +struct perf_cpu_map *cpus = pmu ?perf_cpu_map__get (pmu->cpus) : cpu_list ?perf_cpu_map__new (cpu_list) :NULL ;if (init_attr)
@@ -366,7 +366,7 @@ __add_event(struct list_head *list, int *idx, } (*idx)++; - evsel->core.cpus =perf_cpu_map__get (cpus); + evsel->core.cpus = cpus; evsel->core.own_cpus =perf_cpu_map__get (cpus); evsel->core.system_wide = pmu ? pmu->is_uncore : false; evsel->auto_merge_stats = auto_merge_stats;
Signature · 1 line
2.28.0.618.gf4bc123cb7-goog
Arnaldo Carvalho de Melo
maintainerRe: [PATCH v2 2/2] perf parse-event: Fix cpu map leaks
Em Thu, Sep 17, 2020 at 03:02:19PM +0900, Namhyung Kim escreveu:
Quoted 5 lines
Like evlist cpu map, evsel's cpu map should have a proper refcount. As it's created with a refcount, we don't need to get an extra count. Thanks to Arnaldo for the simpler suggestion.
This fixes the following ASAN report:
This one should come first, and the patch summary is:
"perf parse-events Fix cpu map refcounting"
Then the next patch fixes the refcoutnt leak on the error path.
I'm fixing this up,
Thanks,
- Arnaldo
Quoted 60 lines
Direct leak of 840 byte(s) in 70 object(s) allocated from:
#0 0x7fe36703f628 in malloc (/lib/x86_64-linux-gnu/libasan.so.5+0x107628) #1 0x559fbbf611ca in cpu_map__trim_new /home/namhyung/project/linux/tools/lib/perf/cpumap.c:79 #2 0x559fbbf6229c in perf_cpu_map__new /home/namhyung/project/linux/tools/lib/perf/cpumap.c:237 #3 0x559fbbcc6c6d in __add_event util/parse-events.c:357 #4 0x559fbbcc6c6d in add_event_tool util/parse-events.c:408 #5 0x559fbbcc6c6d in parse_events_add_tool util/parse-events.c:1414 #6 0x559fbbd8474d in parse_events_parse util/parse-events.y:439 #7 0x559fbbcc95da in parse_events__scanner util/parse-events.c:2096 #8 0x559fbbcc95da in __parse_events util/parse-events.c:2141 #9 0x559fbbc2788b in check_parse_id tests/pmu-events.c:406 #10 0x559fbbc2788b in check_parse_id tests/pmu-events.c:393 #11 0x559fbbc2788b in check_parse_fake tests/pmu-events.c:436 #12 0x559fbbc2788b in metric_parse_fake tests/pmu-events.c:553 #13 0x559fbbc27e2d in test_parsing_fake tests/pmu-events.c:599 #14 0x559fbbc27e2d in test_parsing_fake tests/pmu-events.c:574 #15 0x559fbbc0109b in run_test tests/builtin-test.c:410 #16 0x559fbbc0109b in test_and_print tests/builtin-test.c:440 #17 0x559fbbc03e69 in __cmd_test tests/builtin-test.c:695 #18 0x559fbbc03e69 in cmd_test tests/builtin-test.c:807 #19 0x559fbbc691f4 in run_builtin /home/namhyung/project/linux/tools/perf/perf.c:312 #20 0x559fbbb071a8 in handle_internal_command /home/namhyung/project/linux/tools/perf/perf.c:364 #21 0x559fbbb071a8 in run_argv /home/namhyung/project/linux/tools/perf/perf.c:408 #22 0x559fbbb071a8 in main /home/namhyung/project/linux/tools/perf/perf.c:538 #23 0x7fe366b68cc9 in __libc_start_main ../csu/libc-start.c:308And I've failed which commit introduced this bug as the code was heavily changed since then. ;-/
Acked-by Jiri Olsa <[email protected]>Signed-off-by Namhyung Kim <[email protected]>--- tools/perf/util/parse-events.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)tools/perf/util/parse-events.c +2
-2@@ -353,7 +353,7 @@ __add_event(struct list_head *list, int *idx, const char *cpu_list) {struct evsel *evsel; -struct perf_cpu_map *cpus = pmu ? pmu->cpus : +struct perf_cpu_map *cpus = pmu ?perf_cpu_map__get (pmu->cpus) : cpu_list ?perf_cpu_map__new (cpu_list) :NULL ;if (init_attr)@@ -366,7 +366,7 @@ __add_event(struct list_head *list, int *idx, } (*idx)++; - evsel->core.cpus =perf_cpu_map__get (cpus); + evsel->core.cpus = cpus; evsel->core.own_cpus =perf_cpu_map__get (cpus); evsel->core.system_wide = pmu ? pmu->is_uncore : false; evsel->auto_merge_stats = auto_merge_stats;Signature · 1 line
2.28.0.618.gf4bc123cb7-goog
Signature · 1 line
- Arnaldo
Quoted 10 lines · Namhyung Kim
Signed-off-by Namhyung Kim <[email protected]>--- tools/perf/util/parse-events.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)tools/perf/util/parse-events.c +0
-0
tools/perf/util/parse-events.c __add_event(struct list_head *list, int *idx, · line 364
Quoted 6 lines
tools/perf/util/parse-events.c +2
-1@@ -360,8 +360,10 @@ __add_event(struct list_head *list, int *idx,event_attr_init (attr); evsel =evsel__new_idx (attr, *idx); -if (!evsel) +if (!evsel) { +perf_cpu_map__put (cpus);
if there's pmu defined, we don't we get on perf_cpu_map:
struct perf_cpu_map *cpus = pmu ? pmu->cpus :
cpu_list ? perf_cpu_map__new(cpu_list) : NULL;jirka
Quoted 8 lines
tools/perf/util/parse-events.c +1
-0return NULL ; + } (*idx)++; evsel->core.cpus =perf_cpu_map__get (cpus);Signature · 1 line
2.28.0.618.gf4bc123cb7-goog
Namhyung Kim
maintainerQuoted 21 lines · Jiri Olsa
Quoted 10 lines · Namhyung Kim
Signed-off-by Namhyung Kim <[email protected]>--- tools/perf/util/parse-events.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)tools/perf/util/parse-events.c +0
-0tools/perf/util/parse-events.c __add_event(struct list_head *list, int *idx, · line 364
Quoted 6 lines
tools/perf/util/parse-events.c +2
-1@@ -360,8 +360,10 @@ __add_event(struct list_head *list, int *idx,event_attr_init (attr); evsel =evsel__new_idx (attr, *idx); -if (!evsel) +if (!evsel) { +perf_cpu_map__put (cpus);if there's pmu defined, we don't we get on
perf_cpu_map:
Quoted 2 lines
struct perf_cpu_map *cpus = pmu ? pmu->cpus :
cpu_list ? perf_cpu_map__new(cpu_list) : NULL;Right, but it's changed in the next patch. Maybe I need to exchange them.
Thanks
Namhyung
Quoted 10 lines
Quoted 9 lines
tools/perf/util/parse-events.c +1
-1return NULL ; + } (*idx)++; evsel->core.cpus =perf_cpu_map__get (cpus); --2.28.0.618.gf4bc123cb7-goog
Arnaldo Carvalho de Melo
maintainerEm Thu, Sep 17, 2020 at 03:10:58PM +0200, Jiri Olsa escreveu:
Quoted 20 lines · Namhyung Kim
Quoted 10 lines
Signed-off-by Namhyung Kim <[email protected]>--- tools/perf/util/parse-events.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)tools/perf/util/parse-events.c +0
-0tools/perf/util/parse-events.c __add_event(struct list_head *list, int *idx, · line 364
Quoted 6 lines
tools/perf/util/parse-events.c +2
-1@@ -360,8 +360,10 @@ __add_event(struct list_head *list, int *idx,event_attr_init (attr); evsel =evsel__new_idx (attr, *idx); -if (!evsel) +if (!evsel) { +perf_cpu_map__put (cpus);if there's pmu defined, we don't we get on
perf_cpu_map:
Quoted 2 lines
struct perf_cpu_map *cpus = pmu ? pmu->cpus :
cpu_list ? perf_cpu_map__new(cpu_list) : NULL;Yeah, I fixed this up by applying first the patch that grabs the pmu->cpus refcount and makes the first attribution of evsel->core.cpus be directly to 'cpus', as it will either have the refcount for pmu->cpus or the newly created, refcount set at 1 cpu map.o
Then the error path fix goes in, just dropping the reference to whatever cpu map we have a refcount for.
- Arnaldo
Quoted 11 lines
jirka
Quoted 8 lines
tools/perf/util/parse-events.c +1
-0return NULL ; + } (*idx)++; evsel->core.cpus =perf_cpu_map__get (cpus);Signature · 1 line
2.28.0.618.gf4bc123cb7-goog
Signature · 1 line
- Arnaldo
Namhyung Kim
maintainerQuoted 25 lines · Arnaldo Carvalho de Melo
Em Thu, Sep 17, 2020 at 03:10:58PM +0200, Jiri Olsa escreveu:
Quoted 20 lines · Namhyung Kim
Quoted 10 lines
Signed-off-by Namhyung Kim <[email protected]>--- tools/perf/util/parse-events.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)tools/perf/util/parse-events.c +0
-0tools/perf/util/parse-events.c __add_event(struct list_head *list, int *idx, · line 364
Quoted 6 lines
tools/perf/util/parse-events.c +2
-1@@ -360,8 +360,10 @@ __add_event(struct list_head *list, int *idx,event_attr_init (attr); evsel =evsel__new_idx (attr, *idx); -if (!evsel) +if (!evsel) { +perf_cpu_map__put (cpus);if there's pmu defined, we don't we get on
perf_cpu_map:Quoted 2 lines
struct perf_cpu_map *cpus = pmu ? pmu->cpus : cpu_list ? perf_cpu_map__new(cpu_list) : NULL;
Quoted 7 lines
Yeah, I fixed this up by applying first the patch that grabs the
pmu->cpusrefcount and makes the first attribution ofevsel->core.cpusbe directly to 'cpus', as it will either have the refcount forpmu->cpusor the newly created, refcount set at 1 cpu map.oThen the error path fix goes in, just dropping the reference to whatever cpu map we have a refcount for.
Sorry for messing this up and thanks for fixing it. I'll take more care later..
Thanks
Namhyung
Quoted 14 lines
Quoted 10 lines
Quoted 9 lines
tools/perf/util/parse-events.c +1
-1return NULL ; + } (*idx)++; evsel->core.cpus =perf_cpu_map__get (cpus); --2.28.0.618.gf4bc123cb7-goog
--
- Arnaldo