linux-kernel - Re: [Linaro-mm-sig] [PATCH v3] Add udmabuf misc device

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, 29 May 2018 14:29:43 +0200
From:   Daniel Vetter <daniel.vetter@...ll.ch>
To:     Gerd Hoffmann <kraxel@...hat.com>
Cc:     dri-devel <dri-devel@...ts.freedesktop.org>,
        David Airlie <airlied@...ux.ie>,
        Tomeu Vizoso <tomeu.vizoso@...labora.com>,
        Sumit Semwal <sumit.semwal@...aro.org>,
        Shuah Khan <shuah@...nel.org>,
        open list <linux-kernel@...r.kernel.org>,
        "open list:DMA BUFFER SHARING FRAMEWORK" 
        <linux-media@...r.kernel.org>,
        "moderated list:DMA BUFFER SHARING FRAMEWORK" 
        <linaro-mm-sig@...ts.linaro.org>,
        "open list:KERNEL SELFTEST FRAMEWORK" 
        <linux-kselftest@...r.kernel.org>
Subject: Re: [Linaro-mm-sig] [PATCH v3] Add udmabuf misc device

On Tue, May 29, 2018 at 12:48 PM, Gerd Hoffmann <kraxel@...hat.com> wrote:
>   Hi,
>
>> > > qemu test branch:
>> > >   https://git.kraxel.org/cgit/qemu/log/?h=sirius/udmabuf
>         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
>
>> > > + if (!shmem_mapping(file_inode(ubuf->filp)->i_mapping))
>> > > +         goto err_free_ubuf;
>>
>> Can/should we test here that the memfd has a locked down size here?
>
> Makes sense.  Suggested way to check that?  unstatic memfd_get_seals()
> function (mm/shmem.c)?  Or is there some better way?
>
> Also which seals should we require?  Is F_SEAL_SHRINK enough?

Yes I think that's enough.

Hm ... I think we also need to prevent the F_SEAL_WRITE, because
there's no way to stop dma from tampering with the buffer once it's a
dma-buf. Otherwise evil userspace could create a memfd, F_SEAL_SHRINK
it, make a dma-buf out of it, F_SEAL_WRITE it, hand it to some
unsuspecting priviledged service and then pull it over the table with
a few dma-buf writes.

>> On that: Link to userspace patches/git tree using this would be nice.
>
> See above.

Ow, I was blind :-)

Thanks, Daniel
-- 
Daniel Vetter
Software Engineer, Intel Corporation
+41 (0) 79 365 57 48 - http://blog.ffwll.ch