linux-kernel - Re: memcpy: detected field-spanning write (size 168) of single field "&device->entry" at drivers/firmware/google/coreboot

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] [thread-next>] [day] [month] [year] [list]

Date:   Tue, 3 Jan 2023 11:31:02 -0800
From:   Brian Norris <briannorris@...omium.org>
To:     Guenter Roeck <groeck@...gle.com>
Cc:     Julius Werner <jwerner@...omium.org>,
        Paul Menzel <pmenzel@...gen.mpg.de>,
        LKML <linux-kernel@...r.kernel.org>,
        Greg KH <gregkh@...uxfoundation.org>,
        Guenter Roeck <groeck@...omium.org>,
        Stephen Boyd <swboyd@...omium.org>,
        Jack Rosenthal <jrosenth@...omium.org>, coreboot@...eboot.org,
        Kees Cook <keescook@...omium.org>
Subject: Re: memcpy: detected field-spanning write (size 168) of single field
 "&device->entry" at drivers/firmware/google/coreboot_table.c:103 (size 8)

On Thu, Dec 29, 2022 at 12:28:14PM -0800, Guenter Roeck wrote:
> On Thu, Dec 29, 2022 at 6:43 AM Julius Werner <jwerner@...omium.org> wrote:
> >
> > I can confirm that this warning is a false positive, at least. We're
> > intentionally copying bytes from beyond the end of the header
> > structure in this case.
> >
> > I don't know what kind of kernel system detects this stuff at runtime
> > and how to silence it. Probably need to add a void pointer cast or
> > something?
> >
> 
> This is part of kernel hardening code. Kees Cook might know what to do about it.

One could probably throw in casts, like this example did:

  0d043351e5ba ext4: fix fortify warning in fs/ext4/fast_commit.c:1551

Or one could probably imitate this example, and insert an appropriate
flexible array (possibly with yet another union?):

  b43088f30db1 s390/zcrypt: fix warning about field-spanning write

Side mostly-unrelated note: coreboot_table_populate() doesn't do any
bounds checking that the individual entry copies don't overflow the
table buffer size. We're _probably_ not that interested in recovering
from a malicious (or even buggy) Coreboot, but it does seem like an area
of improvement.

Brian

> 
> Guenter
> 
> > On Thu, Dec 29, 2022 at 11:46 AM Paul Menzel <pmenzel@...gen.mpg.de> wrote:
> > >
> > > Dear Linux folks,
> > >
> > >
> > > Running Linux v6.2-rc1+ on a motherboard using coreboot as firmware, the
> > > warning below is shown.
> > >
> > > ```
> > > [    1.630244] ------------[ cut here ]------------
> > > [    1.630249] memcpy: detected field-spanning write (size 168) of
> > > single field "&device->entry" at
> > > drivers/firmware/google/coreboot_table.c:103 (size 8)
> > > [    1.630299] WARNING: CPU: 1 PID: 150 at
> > > drivers/firmware/google/coreboot_table.c:103
> > > coreboot_table_probe+0x1ea/0x210 [coreboot_table]
[...]