Openwall GNU/*/Linux - a small security-enhanced Linux distro for servers
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Fri, 15 Apr 2016 12:28:05 -0700
From: Kees Cook <keescook@...omium.org>
To: Lasse Collin <lasse.collin@...aani.org>
Cc: Ingo Molnar <mingo@...nel.org>, Yinghai Lu <yinghai@...nel.org>, Baoquan He <bhe@...hat.com>, 
	Ard Biesheuvel <ard.biesheuvel@...aro.org>, Matt Redfearn <matt.redfearn@...tec.com>, 
	"x86@...nel.org" <x86@...nel.org>, "H. Peter Anvin" <hpa@...or.com>, Ingo Molnar <mingo@...hat.com>, 
	Borislav Petkov <bp@...en8.de>, Vivek Goyal <vgoyal@...hat.com>, Andy Lutomirski <luto@...nel.org>, 
	Andrew Morton <akpm@...ux-foundation.org>, Dave Young <dyoung@...hat.com>, 
	"kernel-hardening@...ts.openwall.com" <kernel-hardening@...ts.openwall.com>, LKML <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH v5 13/21] x86, boot: Report overlap failures in memcpy

On Fri, Apr 15, 2016 at 7:42 AM, Lasse Collin <lasse.collin@...aani.org> wrote:
> On 2016-04-14 Kees Cook wrote:
>> From: Yinghai Lu <yinghai@...nel.org>
>>
>> parse_elf is using a local memcpy to move sections to their final
>> running position. However, this memcpy only supports non-overlapping
>> arguments (or dest < src).
>
> The same copy of memcpy is used by the decompressors too.
>
>> To avoid future hard-to-debug surprises, this adds checking in memcpy
>> to detect the unhandled condition (which should not be happening
>> currently).
>
> It's already a minor surprise that memcpy is expected to work for
> overlapping buffers at all. It could be good to have a comment about
> it because "scroll" and parse_elf seem to rely on it.
>
> On the other hand, the new code and error message take quite a few bytes
> of space, so a complete memmove can be smaller:
>
> void *memmove(void *dest, const void *src, size_t n)
> {
>         unsigned char *d = dest;
>         const unsigned char *s = src;
>
>         if (d <= s || d - s >= n)
>                 return __memcpy(dest, src, n);
>
>         while (n-- > 0)
>                 d[n] = s[n];
>
>         return dest;
> }
>
> Note that memmove is needed by lib/decompress_unxz.c. It contains its
> own small version inside a "#ifndef memmove" block. That #ifndef should
> be taken into account when adding a memmove symbol. Changing
> decompress_unxz.c is fine but then one needs to think about other
> archs too.

Awesome, thanks! I'd much prefer to fully fix this instead of just
throwing a warning. I'll get this added.

-Kees

-- 
Kees Cook
Chrome OS & Brillo Security

Powered by blists - more mailing lists

Your e-mail address:

Confused about mailing lists and their use? Read about mailing lists on Wikipedia and check out these guidelines on proper formatting of your messages.