linux-kernel - Re: [PATCH] perf probe: check for dup and fdopen failures

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

Message-Id: <20160813085554.3223c0c346a07ea0542c0d02@kernel.org>
Date:	Sat, 13 Aug 2016 08:55:54 +0900
From:	Masami Hiramatsu <mhiramat@...nel.org>
To:	Arnaldo Carvalho de Melo <acme@...nel.org>
Cc:	Colin King <colin.king@...onical.com>,
	Peter Zijlstra <peterz@...radead.org>,
	Ingo Molnar <mingo@...nel.org>,
	Alexander Shishkin <alexander.shishkin@...ux.intel.com>,
	Masami Hiramatsu <mhiramat@...nel.org>,
	Namhyung Kim <namhyung@...nel.org>,
	Jiri Olsa <jolsa@...nel.org>, Wang Nan <wangnan0@...wei.com>,
	linux-kernel@...r.kernel.org
Subject: Re: [PATCH] perf probe: check for dup and fdopen failures

On Fri, 12 Aug 2016 18:55:17 -0300
Arnaldo Carvalho de Melo <acme@...nel.org> wrote:

> Em Fri, Aug 12, 2016 at 10:44:56PM +0100, Colin King escreveu:
> > From: Colin Ian King <colin.king@...onical.com>
> > 
> > dup and fdopen can potentially fail, so add some extra
> > error handling checks rather than assuming they always work.

Thanks, I forgot to handle that.

> > 
> > Signed-off-by: Colin Ian King <colin.king@...onical.com>
> > ---
> >  tools/perf/util/probe-file.c | 16 ++++++++++++----
> >  1 file changed, 12 insertions(+), 4 deletions(-)
> > 
> > diff --git a/tools/perf/util/probe-file.c b/tools/perf/util/probe-file.c
> > index 9aed9c3..f3df939 100644
> > --- a/tools/perf/util/probe-file.c
> > +++ b/tools/perf/util/probe-file.c
> > @@ -133,7 +133,7 @@ int probe_file__open_both(int *kfd, int *ufd, int flag)
> >  /* Get raw string list of current kprobe_events  or uprobe_events */
> >  struct strlist *probe_file__get_rawlist(int fd)
> >  {
> > -	int ret, idx;
> > +	int ret, idx, fddup;
> >  	FILE *fp;
> >  	char buf[MAX_CMDLEN];
> >  	char *p;
> > @@ -144,7 +144,12 @@ struct strlist *probe_file__get_rawlist(int fd)
> >  
> >  	sl = strlist__new(NULL, NULL);
> >  
> > -	fp = fdopen(dup(fd), "r");
> > +	fddup = dup(fd);
> > +	if (fddup < 0)
> > +		return NULL;
> > +	fp = fdopen(fddup, "r");
> > +	if (!fp)
> 
> here we leak fddup, no?
> 
> > +		return NULL;
> 
> I.e. this should be:
> 
> 	if (!fp) {
> 		close(fddup);
> 		return NULL;
> 	}
> 
> Ack?

Correct, it should be closed. So both fixes look good to me.

Acked-by: Masami Hiramatsu <mhiramat@...nel.org>

Thanks!


> 
> - Arnaldo
> 
> >  	while (!feof(fp)) {
> >  		p = fgets(buf, MAX_CMDLEN, fp);
> >  		if (!p)
> > @@ -447,10 +452,13 @@ static int probe_cache__load(struct probe_cache *pcache)
> >  {
> >  	struct probe_cache_entry *entry = NULL;
> >  	char buf[MAX_CMDLEN], *p;
> > -	int ret = 0;
> > +	int ret = 0, fddup;
> >  	FILE *fp;
> >  
> > -	fp = fdopen(dup(pcache->fd), "r");
> > +	fddup = dup(pcache->fd);
> > +	if (fddup < 0)
> > +		return -errno;
> > +	fp = fdopen(fddup, "r");
> >  	if (!fp)
> >  		return -EINVAL;
> >  
> > -- 
> > 2.8.1


-- 
Masami Hiramatsu <mhiramat@...nel.org>