linux-kernel - Re: [PATCH] regmap: don't corrupt work buffer in _regmap_raw

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: <5162F390.5050701@wwwdotorg.org>
Date:	Mon, 08 Apr 2013 10:42:56 -0600
From:	Stephen Warren <swarren@...dotorg.org>
To:	Mark Brown <broonie@...nsource.wolfsonmicro.com>
CC:	Paul Bolle <pebolle@...cali.nl>, linux-kernel@...r.kernel.org,
	Stephen Warren <swarren@...dia.com>,
	Laxman Dewangan <ldewangan@...dia.com>
Subject: Re: [PATCH] regmap: don't corrupt work buffer in _regmap_raw_write()

On 04/08/2013 06:01 AM, Mark Brown wrote:
> On Mon, Apr 08, 2013 at 12:24:22PM +0200, Paul Bolle wrote:
> 
>> 0) This patch ended up as mainline commit 
>> bc8ce4afd7ee7e1421c935d24b1f879f82afdd4e, which is part of
>> v3.9-rc6.
> 
> Numbering your paragraphs and using full commit IDs like this
> doesn't help with legibility...
> 
>> 2) The following (draft) patch silences this warning. I'm a bit 
>> uncertain what the regmap_parse_*() functions are meant to do. So
>> I'd like to first ask whether something along these lines is
>> acceptable.
> 
> Nope, this breaks v3.9 - in that version they both modify in place
> and return the parsed value.  In v3.10 it already does what you
> have because the in place and return value versions have been
> split.
> 
> We need to either revert the commit from v3.9 on the basis that
> nobody noticed the issue until some other work so it can't be that
> bad or come up with a more invasive fix, at this point in the cycle
> I'm more inclined to do the latter.

I assume you mean "former" not "latter" above?

Revert might well be simplest and just fine. The issue this patch
fixes ended up getting accidentally "fixed" by the side-effects of
calling .parse_val() before commit 8a819ff "regmap: core: Split out in
place value parsing", and that commit only appears in 3.10, so we
should be able to revert this patch without issue in 3.9.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/