linux-kernel - Re: [PATCH] kdb: prefer strlcpy to strncpy

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]

Message-ID: <20180531082428.c5z7cpdetdrhy5wd@holly.lan>
Date:   Thu, 31 May 2018 09:24:28 +0100
From:   Daniel Thompson <daniel.thompson@...aro.org>
To:     Geert Uytterhoeven <geert@...ux-m68k.org>
Cc:     Nick Desaulniers <nick.desaulniers@...il.com>,
        Arnd Bergmann <arnd@...db.de>,
        Jason Wessel <jason.wessel@...driver.com>,
        Randy Dunlap <rdunlap@...radead.org>,
        Baolin Wang <baolin.wang@...aro.org>,
        "Eric W. Biederman" <ebiederm@...ssion.com>,
        kgdb-bugreport@...ts.sourceforge.net,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
        ebiggers@...gle.com
Subject: Re: [PATCH] kdb: prefer strlcpy to strncpy

On Wed, May 30, 2018 at 10:47:13PM +0200, Geert Uytterhoeven wrote:
> H Daniel,
> 
> On Wed, May 30, 2018 at 4:34 PM, Daniel Thompson
> <daniel.thompson@...aro.org> wrote:
> > On Tue, May 29, 2018 at 07:01:35PM -0700, Nick Desaulniers wrote:
> >> On Tue, May 29, 2018 at 12:57 AM, Arnd Bergmann <arnd@...db.de> wrote:
> >> > On Tue, May 29, 2018 at 7:57 AM, Nick Desaulniers
> >> > <nick.desaulniers@...il.com> wrote:
> >> >> Fixes stringop-truncation and stringop-overflow warnings from gcc-8.
> 
> >> Eric points out that this will leak kernel memory if size is less than
> >> sizeof src.
> >
> > Don't quite understand what this means (there's no allocation here, how
> > can there be a leak?) but the symbol completion certainly won't work if
> > we truncate the copy here.
> 
> Not leak an is memory leak, but leak as in information leak of uninitialized
> data to userspace (if the buffer is ever copied to userspace).

I see... I saw "leak", I saw "memory" and was perhaps too quick to link
the two together.

The underlying bug is a buffer overflow (so a good catch and I look
forward to a v2) but, with or without Nick's change, I can't see a leak
in either sense of the word in the code that Arnd was commenting on[1].


Daniel.


[1] Clearly the undefined behaviour post-overflow *could* be a leak but
    I stopped analyzing after the overflow.