From: Namhyung Kim <namhyung@kernel.org>
To: Ian Rogers <irogers@google.com>
Cc: zhaimingbing <zhaimingbing@cmss.chinamobile.com>,
Peter Zijlstra <peterz@infradead.org>,
Ingo Molnar <mingo@redhat.com>,
Arnaldo Carvalho de Melo <acme@kernel.org>,
Mark Rutland <mark.rutland@arm.com>,
Alexander Shishkin <alexander.shishkin@linux.intel.com>,
Jiri Olsa <jolsa@kernel.org>,
Adrian Hunter <adrian.hunter@intel.com>,
Kan Liang <kan.liang@linux.intel.com>,
linux-perf-users@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH v1] perf lock: More strdup argument freeing
Date: Wed, 1 May 2024 12:46:38 -0700 [thread overview]
Message-ID: <CAM9d7chFv38q+a9m_OE-dP1aSUpvd35tM1V7DA_t756O6hrTUA@mail.gmail.com> (raw)
In-Reply-To: <20240430184156.1824083-1-irogers@google.com>
On Tue, Apr 30, 2024 at 11:42 AM Ian Rogers <irogers@google.com> wrote:
>
> Leak sanitizer complains about the strdup-ed arguments not being
> freed. rec_argv is reordered and duplicates inserted, meaning making
> all its contents strdup-ed and freeing them all leads to double frees
> or leaks. Add an extra array to track strup-ed arguments and free
> them. This makes address sanitier running `perf test` "kernel lock
> contention analysis test" memory leak free.
>
> Signed-off-by: Ian Rogers <irogers@google.com>
> ---
> tools/perf/builtin-lock.c | 44 +++++++++++++++++++++++----------------
> 1 file changed, 26 insertions(+), 18 deletions(-)
>
> diff --git a/tools/perf/builtin-lock.c b/tools/perf/builtin-lock.c
> index 230461280e45..26c059397cdf 100644
> --- a/tools/perf/builtin-lock.c
> +++ b/tools/perf/builtin-lock.c
> @@ -2230,10 +2230,11 @@ static int __cmd_record(int argc, const char **argv)
> const char *callgraph_args[] = {
> "--call-graph", "fp," __stringify(CONTENTION_STACK_DEPTH),
> };
> - unsigned int rec_argc, i, j, ret;
> + unsigned int rec_argc, i, j, dups = 0, ret;
> unsigned int nr_tracepoints;
> unsigned int nr_callgraph_args = 0;
> const char **rec_argv;
> + char **to_free;
> bool has_lock_stat = true;
>
> for (i = 0; i < ARRAY_SIZE(lock_tracepoints); i++) {
> @@ -2270,28 +2271,25 @@ static int __cmd_record(int argc, const char **argv)
> /* factor of 2 is for -e in front of each tracepoint */
> rec_argc += 2 * nr_tracepoints;
>
> - rec_argv = calloc(rec_argc + 1, sizeof(char *));
> + rec_argv = calloc(rec_argc + 1, sizeof(*rec_argv));
> if (!rec_argv)
> return -ENOMEM;
>
> - for (i = 0; i < ARRAY_SIZE(record_args); i++)
> - rec_argv[i] = strdup(record_args[i]);
> -
> - for (j = 0; j < nr_tracepoints; j++) {
> - const char *ev_name;
> + to_free = calloc(rec_argc, sizeof(*to_free));
> + if (!to_free)
> + return -ENOMEM;
Need to free rec_argv. 'goto out' would be fine.
>
> - if (has_lock_stat)
> - ev_name = strdup(lock_tracepoints[j].name);
> - else
> - ev_name = strdup(contention_tracepoints[j].name);
> -
> - if (!ev_name) {
> - free(rec_argv);
> - return -ENOMEM;
> - }
>
> + for (i = 0; i < ARRAY_SIZE(record_args);) {
> + to_free[dups] = strdup(record_args[i]);
> + rec_argv[i++] = to_free[dups++];
> + }
> + for (j = 0; j < nr_tracepoints; j++) {
> + to_free[dups] = strdup(has_lock_stat
> + ? lock_tracepoints[j].name
> + : contention_tracepoints[j].name);
> rec_argv[i++] = "-e";
> - rec_argv[i++] = ev_name;
> + rec_argv[i++] = to_free[dups++];
Now I'm curious why we copy the string in the first place.
Maybe not needed..?
Thanks,
Namhyung
> }
>
> for (j = 0; j < nr_callgraph_args; j++, i++)
> @@ -2302,7 +2300,17 @@ static int __cmd_record(int argc, const char **argv)
>
> BUG_ON(i != rec_argc);
>
> - ret = cmd_record(i, rec_argv);
> + for (i = 0; i < dups; i++) {
> + if (to_free[i] == NULL) {
> + ret = -ENOMEM;
> + goto out;
> + }
> + }
> + ret = cmd_record(rec_argc, rec_argv);
> +out:
> + for (i = 0; i < dups; i++)
> + zfree(&to_free[i]);
> + free(to_free);
> free(rec_argv);
> return ret;
> }
> --
> 2.45.0.rc0.197.gbae5840b3b-goog
>
prev parent reply other threads:[~2024-05-01 19:46 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-04-30 18:41 [PATCH v1] perf lock: More strdup argument freeing Ian Rogers
2024-05-01 14:22 ` James Clark
2024-05-01 19:46 ` Namhyung Kim [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=CAM9d7chFv38q+a9m_OE-dP1aSUpvd35tM1V7DA_t756O6hrTUA@mail.gmail.com \
--to=namhyung@kernel.org \
--cc=acme@kernel.org \
--cc=adrian.hunter@intel.com \
--cc=alexander.shishkin@linux.intel.com \
--cc=irogers@google.com \
--cc=jolsa@kernel.org \
--cc=kan.liang@linux.intel.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-perf-users@vger.kernel.org \
--cc=mark.rutland@arm.com \
--cc=mingo@redhat.com \
--cc=peterz@infradead.org \
--cc=zhaimingbing@cmss.chinamobile.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).