Re: [PATCH] path: use xmalloc in add_to_trie

From: Jeff King <peff@peff.net>
Date: Oct 24, 2017, 19:26 UTC
Message-ID: <20171024192658.4rvxvb34y64lt2tc@sigill.intra.peff.net>
In-Reply-To: <CAGZ79kZM_hVi8BeTmL5u_QK5+=1g3gcPsccEe1AJVpDLUWz_9Q@mail.gmail.com>

On Tue, Oct 24, 2017 at 09:13:28AM -0700, Stefan Beller wrote:

Show 16 quoted lines

> On Tue, Oct 24, 2017 at 8:15 AM, Andrey Okoshkin <a.okoshkin@samsung.com> wrote:
> > Add usage of xmalloc() instead of malloc() in add_to_trie() as xmalloc wraps
> > and checks memory allocation result.
> >
> > Signed-off-by: Andrey Okoshkin <a.okoshkin@samsung.com>
> > ---
> > Hello,
> > I'm not sure but it looks like there is a missing check of the malloc result.
> > memcpy() may crash with SIGSEGV due to the memory allocation failure.
> > make_trie_node() uses xmalloc() and xcalloc() - so I believe add_to_trie()
> > also should use it.
> 
> Good catch! Thanks for spotting.
> 
> Trying to find similar occurrences via git grep "= malloc" did not
> yield other places that need the same fix.

Don't forget realloc and calloc (though I don't think there are any cases that need touching there; just a hint for anybody doing auditing).

-Peff