linux-kernel - Re: [PATCH 2/2] perf stat: Fix a segfault with --per-cluster --metric-only

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Date: Fri, 28 Jun 2024 09:47:30 -0300
From: Arnaldo Carvalho de Melo <acme@...nel.org>
To: Namhyung Kim <namhyung@...nel.org>
Cc: Ian Rogers <irogers@...gle.com>, Kan Liang <kan.liang@...ux.intel.com>,
	Jiri Olsa <jolsa@...nel.org>,
	Adrian Hunter <adrian.hunter@...el.com>,
	Peter Zijlstra <peterz@...radead.org>,
	Ingo Molnar <mingo@...nel.org>, LKML <linux-kernel@...r.kernel.org>,
	linux-perf-users@...r.kernel.org,
	Yicong Yang <yangyicong@...ilicon.com>
Subject: Re: [PATCH 2/2] perf stat: Fix a segfault with --per-cluster
 --metric-only

On Thu, Jun 27, 2024 at 01:03:53PM -0700, Namhyung Kim wrote:
> The new --per-cluster option was added recently but it forgot to update
> the aggr_header fields which are used for --metric-only option.  And it
> resulted in a segfault due to NULL string in fputs().

Before:

acme@...ber:~$ sudo ~acme/bin/perf stat -a -x : --per-cluster -M tma_core_bound --metric-only true
Segmentation fault
acme@...ber:~$ 
acme@...ber:~$ sudo su -
root@...ber:~# gdb perf
(gdb) run stat -a -x : --per-cluster -M tma_core_bound --metric-only true
Starting program: /root/bin/perf stat -a -x : --per-cluster -M tma_core_bound --metric-only true

Program received signal SIGSEGV, Segmentation fault.
0x00007ffff6f7f8dd in __strlen_avx2 () from /lib64/libc.so.6
(gdb) bt
#0  0x00007ffff6f7f8dd in __strlen_avx2 () from /lib64/libc.so.6
#1  0x00007ffff6e97a3a in fputs () from /lib64/libc.so.6
#2  0x000000000056b805 in print_metric_headers ()
#3  0x000000000056e084 in evlist.print_counters ()
#4  0x0000000000432513 in cmd_stat ()
#5  0x00000000004c5fb9 in run_builtin ()
#6  0x00000000004c62c9 in handle_internal_command ()
#7  0x0000000000410e57 in main ()
(gdb)

After:

acme@...ber:~$ sudo ~acme/bin/perf stat -a -x : --per-cluster -M tma_core_bound --metric-only true
cluster:cpus:%  tma_core_bound:%  tma_core_bound:
S0-D0-CLS0:2:18.2::::
S0-D0-CLS8:2:26.7::::
S0-D0-CLS16:2:14.2::::
S0-D0-CLS24:2:10.6::::
S0-D0-CLS32:2:0.6::::
S0-D0-CLS40:2:42.5::::
S0-D0-CLS48:2:21.1::::
S0-D0-CLS56:2:36.8::::
S0-D0-CLS64:0:::::::1.0:
S0-D0-CLS72:0:::::::0.8:
S0-D0-CLS80:0:::::::1.0:
acme@...ber:~$

Tested-by: Arnaldo Carvalho de Melo <acme@...hat.com>

- Arnaldo
 
> Fixes: cbc917a1b03b ("perf stat: Support per-cluster aggregation")
> Cc: Yicong Yang <yangyicong@...ilicon.com>
> Signed-off-by: Namhyung Kim <namhyung@...nel.org>
> ---
>  tools/perf/util/stat-display.c | 3 +++
>  1 file changed, 3 insertions(+)
> 
> diff --git a/tools/perf/util/stat-display.c b/tools/perf/util/stat-display.c
> index e8673c9f6b49..462227f663cb 100644
> --- a/tools/perf/util/stat-display.c
> +++ b/tools/perf/util/stat-display.c
> @@ -38,6 +38,7 @@
>  static int aggr_header_lens[] = {
>  	[AGGR_CORE] 	= 18,
>  	[AGGR_CACHE]	= 22,
> +	[AGGR_CLUSTER]	= 20,
>  	[AGGR_DIE] 	= 12,
>  	[AGGR_SOCKET] 	= 6,
>  	[AGGR_NODE] 	= 6,
> @@ -49,6 +50,7 @@ static int aggr_header_lens[] = {
>  static const char *aggr_header_csv[] = {
>  	[AGGR_CORE] 	= 	"core%scpus%s",
>  	[AGGR_CACHE]	= 	"cache%scpus%s",
> +	[AGGR_CLUSTER]	= 	"cluster%scpus%s",
>  	[AGGR_DIE] 	= 	"die%scpus%s",
>  	[AGGR_SOCKET] 	= 	"socket%scpus%s",
>  	[AGGR_NONE] 	= 	"cpu%s",
> @@ -60,6 +62,7 @@ static const char *aggr_header_csv[] = {
>  static int aggr_header_num[] = {
>  	[AGGR_CORE] 	= 	2,
>  	[AGGR_CACHE]	= 	2,
> +	[AGGR_CLUSTER]	= 	2,
>  	[AGGR_DIE] 	= 	2,
>  	[AGGR_SOCKET] 	= 	2,
>  	[AGGR_NONE] 	= 	1,
> -- 
> 2.45.2.803.g4e1b14247a-goog