Skip to contentSkip to conversation

perf tools: Assorted random updates and fixes

PATCH Applied Namhyung Kim 2 participants 9 messages Jun 21, 2024 – Jun 26, 2024 perf events

lore
Share & tools
lore

Namhyung Kim

author maintainer
#0 lore raw

cover perf tools: Assorted random updates and fixes

Hello,

I found some misc updates and fixes are still remaining. Just sending out together. Please take a look!

Thanks,
Namhyung

Namhyung Kim (4):
perf report: Fix condition in sort__sym_cmp() perf symbol: Simplify kernel module checking perf tools: Fix a compiler warning of NULL pointer perf mem: Fix a segfault with NULL event->name

 tools/perf/util/dsos.c       | 3 +++
 tools/perf/util/mem-events.c | 2 +-
 tools/perf/util/sort.c       | 2 +-
 tools/perf/util/symbol.c     | 5 +----
 4 files changed, 6 insertions(+), 6 deletions(-)
Signature · 1 line
2.45.2.741.gdbec12cfda-goog
author maintainer
#1 lore raw

PATCH 1/4 perf report: Fix condition in sort__sym_cmp()

It's expected that both hist entries are in the same hists when comparing two. But the current code in the function checks one without dso sort key and other with the key. This would make the condition true in any case.

I guess the intention of the original commit was to add '!' for the right side too. But as it should be the same, let's just remove it.

Fixes 69849fc5d2119 ("perf hists: Move sort__has_dso into struct perf_hpp_list")
---
 tools/perf/util/sort.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
tools/perf/util/sort.c +1 -1
index cd39ea972193..ab7c7ff35f9b 100644
@@ -334,7 +334,7 @@ sort__sym_cmp(struct hist_entry *left, struct hist_entry *right)
 	 * comparing symbol address alone is not enough since it's a
 	 * relative address within a dso.
 	 */
-	if (!hists__has(left->hists, dso) || hists__has(right->hists, dso)) {
+	if (!hists__has(left->hists, dso)) {
 		ret = sort__dso_cmp(left, right);
 		if (ret != 0)
 			return ret;
Signature · 1 line
2.45.2.741.gdbec12cfda-goog
author maintainer
#2 lore raw

PATCH 2/4 perf symbol: Simplify kernel module checking

In reply to Namhyung Kim

In dso__load(), it checks if the dso is a kernel module by looking the symtab type. Actually dso has 'is_kmod' field to check that easily and dso__set_module_info() set the symtab type and the is_kmod bit. So it should have the same result to check the is_kmod bit.

---
 tools/perf/util/symbol.c | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)
tools/perf/util/symbol.c +1 -4
index 8e2da48eeaf0..25dfad79b5d7 100644
@@ -1816,10 +1816,7 @@ int dso__load(struct dso *dso, struct map *map)
 		goto out;
 	}
 
-	kmod = dso__symtab_type(dso) == DSO_BINARY_TYPE__SYSTEM_PATH_KMODULE ||
-		dso__symtab_type(dso) == DSO_BINARY_TYPE__SYSTEM_PATH_KMODULE_COMP ||
-		dso__symtab_type(dso) == DSO_BINARY_TYPE__GUEST_KMODULE ||
-		dso__symtab_type(dso) == DSO_BINARY_TYPE__GUEST_KMODULE_COMP;
+	kmod = dso__is_kmod(dso);
 
 	if (dso__kernel(dso) && !kmod) {
 		if (dso__kernel(dso) == DSO_SPACE__KERNEL)
Signature · 1 line
2.45.2.741.gdbec12cfda-goog
author maintainer
#3 lore raw

PATCH 3/4 perf tools: Fix a compiler warning of NULL pointer

In reply to Namhyung Kim

A compiler warning on the second argument of bsearch() should not be NULL, but there's a case we might pass it. Let's return early if we don't have any DSOs to search in __dsos__find_by_longname_id().

util/dsos.c:184:8: runtime error: null pointer passed as argument 2, which is declared to never be null

---
 tools/perf/util/dsos.c | 3 +++
 1 file changed, 3 insertions(+)
tools/perf/util/dsos.c +3 -0
index ab3d0c01dd63..d5d78bdc56b2 100644
@@ -164,6 +164,9 @@ static struct dso *__dsos__find_by_longname_id(struct dsos *dsos,
 	};
 	struct dso **res;
 
+	if (dsos->dsos == NULL)
+		return NULL;
+
 	if (!dsos->sorted) {
 		if (!write_locked) {
 			struct dso *dso;
Signature · 1 line
2.45.2.741.gdbec12cfda-goog
author maintainer
#4 lore raw

PATCH 4/4 perf mem: Fix a segfault with NULL event->name

In reply to Namhyung Kim

Guilherme reported a crash in perf mem record. It's because the perf_mem_event->name was NULL on his machine. It should just return a NULL string when it has no format string in the name.

But I'm not sure why it returns TRUE if it doesn't have event_name in perf_pmu__mem_events_supported().

The backtrace at the crash is below:

Program received signal SIGSEGV, Segmentation fault. __strchrnul_avx2 () at ../sysdeps/x86_64/multiarch/strchr-avx2.S:67
67 vmovdqu (%rdi), %ymm2
(gdb) bt

  #0  __strchrnul_avx2 () at ../sysdeps/x86_64/multiarch/strchr-avx2.S:67
  #1  0x00007ffff6c982de in __find_specmb (format=0x0) at printf-parse.h:82
  #2  __printf_buffer (buf=buf@entry=0x7fffffffc760, format=format@entry=0x0, ap=ap@entry=0x7fffffffc880,
      mode_flags=mode_flags@entry=0) at vfprintf-internal.c:649
  #3  0x00007ffff6cb7840 in __vsnprintf_internal (string=<optimized out>, maxlen=<optimized out>, format=0x0,
      args=0x7fffffffc880, mode_flags=mode_flags@entry=0) at vsnprintf.c:96
  #4  0x00007ffff6cb787f in ___vsnprintf (string=<optimized out>, maxlen=<optimized out>, format=<optimized out>,
      args=<optimized out>) at vsnprintf.c:103
  #5  0x00005555557b9391 in scnprintf (buf=0x555555fe9320 <mem_loads_name> "", size=100, fmt=0x0)
      at ../lib/vsprintf.c:21
  #6  0x00005555557b74c3 in perf_pmu__mem_events_name (i=0, pmu=0x555556832180) at util/mem-events.c:106
  #7  0x00005555557b7ab9 in perf_mem_events__record_args (rec_argv=0x55555684c000, argv_nr=0x7fffffffca20)
      at util/mem-events.c:252
  #8  0x00005555555e370d in __cmd_record (argc=3, argv=0x7fffffffd760, mem=0x7fffffffcd80) at builtin-mem.c:156
  #9  0x00005555555e49c4 in cmd_mem (argc=4, argv=0x7fffffffd760) at builtin-mem.c:514
  #10 0x000055555569716c in run_builtin (p=0x555555fcde80 <commands+672>, argc=8, argv=0x7fffffffd760) at perf.c:349
  #11 0x0000555555697402 in handle_internal_command (argc=8, argv=0x7fffffffd760) at perf.c:402
  #12 0x0000555555697560 in run_argv (argcp=0x7fffffffd59c, argv=0x7fffffffd590) at perf.c:446
  #13 0x00005555556978a6 in main (argc=8, argv=0x7fffffffd760) at perf.c:562
---
 tools/perf/util/mem-events.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
tools/perf/util/mem-events.c +1 -1
index 6dda47bb774f..429079329e48 100644
@@ -86,7 +86,7 @@ static const char *perf_pmu__mem_events_name(int i, struct perf_pmu *pmu)
 		return NULL;
 
 	e = &pmu->mem_events[i];
-	if (!e)
+	if (!e || !e->name)
 		return NULL;
 
 	if (i == PERF_MEM_EVENTS__LOAD || i == PERF_MEM_EVENTS__LOAD_STORE) {
Signature · 1 line
2.45.2.741.gdbec12cfda-goog
#5 lore raw

Re: [PATCH 4/4] perf mem: Fix a segfault with NULL event->name

Quoted 6 lines · Namhyung Kim

Guilherme reported a crash in perf mem record. It's because the perf_mem_event->name was NULL on his machine. It should just return a NULL string when it has no format string in the name.

But I'm not sure why it returns TRUE if it doesn't have event_name in perf_pmu__mem_events_supported().

AMD doesn't have the event_name.

struct perf_mem_event perf_mem_events_amd[PERF_MEM_EVENTS__MAX] = {
	E(NULL,		NULL,		NULL,	false,	0),
	E(NULL,		NULL,		NULL,	false,	0),
	E("mem-ldst",	"%s//",		NULL,	false,	0),
};

It looks like as long as it's a "ibs_op" PMU, the mem event is always available. The "ibs_op" check has been moved into perf_pmu__arch_init(). So the event_name is empty.

The "e->tag" in the perf_pmu__mem_events_init() can help to skip the
case E(NULL, NULL, NULL, false, 0).
So the perf_pmu__mem_events_supported() returns TRUE for the !event_name.

Thanks,
Kan

Quoted 47 lines

The backtrace at the crash is below:

Program received signal SIGSEGV, Segmentation fault. __strchrnul_avx2 () at ../sysdeps/x86_64/multiarch/strchr-avx2.S:67
67 vmovdqu (%rdi), %ymm2
(gdb) bt

  #0  __strchrnul_avx2 () at ../sysdeps/x86_64/multiarch/strchr-avx2.S:67
  #1  0x00007ffff6c982de in __find_specmb (format=0x0) at printf-parse.h:82
  #2  __printf_buffer (buf=buf@entry=0x7fffffffc760, format=format@entry=0x0, ap=ap@entry=0x7fffffffc880,
      mode_flags=mode_flags@entry=0) at vfprintf-internal.c:649
  #3  0x00007ffff6cb7840 in __vsnprintf_internal (string=<optimized out>, maxlen=<optimized out>, format=0x0,
      args=0x7fffffffc880, mode_flags=mode_flags@entry=0) at vsnprintf.c:96
  #4  0x00007ffff6cb787f in ___vsnprintf (string=<optimized out>, maxlen=<optimized out>, format=<optimized out>,
      args=<optimized out>) at vsnprintf.c:103
  #5  0x00005555557b9391 in scnprintf (buf=0x555555fe9320 <mem_loads_name> "", size=100, fmt=0x0)
      at ../lib/vsprintf.c:21
  #6  0x00005555557b74c3 in perf_pmu__mem_events_name (i=0, pmu=0x555556832180) at util/mem-events.c:106
  #7  0x00005555557b7ab9 in perf_mem_events__record_args (rec_argv=0x55555684c000, argv_nr=0x7fffffffca20)
      at util/mem-events.c:252
  #8  0x00005555555e370d in __cmd_record (argc=3, argv=0x7fffffffd760, mem=0x7fffffffcd80) at builtin-mem.c:156
  #9  0x00005555555e49c4 in cmd_mem (argc=4, argv=0x7fffffffd760) at builtin-mem.c:514
  #10 0x000055555569716c in run_builtin (p=0x555555fcde80 <commands+672>, argc=8, argv=0x7fffffffd760) at perf.c:349
  #11 0x0000555555697402 in handle_internal_command (argc=8, argv=0x7fffffffd760) at perf.c:402
  #12 0x0000555555697560 in run_argv (argcp=0x7fffffffd59c, argv=0x7fffffffd590) at perf.c:446
  #13 0x00005555556978a6 in main (argc=8, argv=0x7fffffffd760) at perf.c:562
---
 tools/perf/util/mem-events.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
tools/perf/util/mem-events.c +1 -1
index 6dda47bb774f..429079329e48 100644
@@ -86,7 +86,7 @@ static const char *perf_pmu__mem_events_name(int i, struct perf_pmu *pmu)
 		return NULL;
 
 	e = &pmu->mem_events[i];
-	if (!e)
+	if (!e || !e->name)
 		return NULL;
 
 	if (i == PERF_MEM_EVENTS__LOAD || i == PERF_MEM_EVENTS__LOAD_STORE) {
author maintainer
#6 lore raw

Re: [PATCH 4/4] perf mem: Fix a segfault with NULL event->name

Hi Kan,

Quoted 19 lines · Kan
Quoted 6 lines · Namhyung Kim

Guilherme reported a crash in perf mem record. It's because the perf_mem_event->name was NULL on his machine. It should just return a NULL string when it has no format string in the name.

But I'm not sure why it returns TRUE if it doesn't have event_name in perf_pmu__mem_events_supported().

AMD doesn't have the event_name.

struct perf_mem_event perf_mem_events_amd[PERF_MEM_EVENTS__MAX] = {

        E(NULL,         NULL,           NULL,   false,  0),
        E(NULL,         NULL,           NULL,   false,  0),
        E("mem-ldst",   "%s//",         NULL,   false,  0),

};

Quoted 7 lines

It looks like as long as it's a "ibs_op" PMU, the mem event is always available. The "ibs_op" check has been moved into perf_pmu__arch_init(). So the event_name is empty.

The "e->tag" in the perf_pmu__mem_events_init() can help to skip the
case E(NULL, NULL, NULL, false, 0).
So the perf_pmu__mem_events_supported() returns TRUE for the !event_name.

Ok, thanks for the explanation!

Namhyung

Reviewed-by on the v1 series
#7 lore raw
In reply to Namhyung Kim
Quoted 9 lines · Namhyung Kim

Hello,

I found some misc updates and fixes are still remaining. Just sending out together. Please take a look!

Thanks,
Namhyung

Quoted 5 lines

Namhyung Kim (4):
perf report: Fix condition in sort__sym_cmp() perf symbol: Simplify kernel module checking perf tools: Fix a compiler warning of NULL pointer perf mem: Fix a segfault with NULL event->name

Looks good to me.

Thanks,
Kan

Quoted 7 lines
 tools/perf/util/dsos.c       | 3 +++
 tools/perf/util/mem-events.c | 2 +-
 tools/perf/util/sort.c       | 2 +-
 tools/perf/util/symbol.c     | 5 +----
 4 files changed, 6 insertions(+), 6 deletions(-)
author maintainer Applied
#8 lore raw
In reply to Namhyung Kim
Quoted 8 lines · Namhyung Kim

I found some misc updates and fixes are still remaining. Just sending out together. Please take a look!

Thanks,
Namhyung

[...]

Applied to perf-tools-next, thanks!

Best regards,
Namhyung