[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAGXu5j+_biDq6xzgB0mAEm159mFc5fABfqOT4-Yew9t-YdNuvg@mail.gmail.com>
Date: Tue, 27 Feb 2018 11:33:34 -0800
From: Kees Cook <keescook@...omium.org>
To: Laura Abbott <labbott@...hat.com>
Cc: P J P <ppandit@...hat.com>,
Kernel Hardening <kernel-hardening@...ts.openwall.com>, Florian Weimer <fweimer@...hat.com>,
P J P <pjp@...oraproject.org>
Subject: Re: [PATCH 1/1] Add an option to build kernel with -finit-local-vars
On Tue, Feb 27, 2018 at 11:22 AM, Laura Abbott <labbott@...hat.com> wrote:
> On 02/27/2018 03:15 AM, P J P wrote:
>> Add a configuration option to build kernel with -finit-local-vars
>> compiler option.[*] It'll zero initialize the automatic kernel
>> function variables, thus helping to reduce kernel information
>> leakage issues.
>
> I think this would make the existing structleak plugin
> (scripts/gcc-plugins/structleak_plugin.c) obsolete. This isn't
> a bad thing but we'd need to figure out a deprecation strategy.
It would be nice to make it obsolete, but I don't think that'll happen
right away. We still have issues with structure padding,
passed-by-reference init, and possibly performance. I wouldn't want to
rule anything out until we can have both more complete coverage and
better benchmarks (e.g. this is wipe-before, not wipe-after, so there
may be cache effects, etc).
-Kees
--
Kees Cook
Pixel Security
Powered by blists - more mailing lists
Confused about mailing lists and their use?
Read about mailing lists on Wikipedia
and check out these
guidelines on proper formatting of your messages.
