linux-kernel - Re: [PATCH] lib: Reduce user_access_begin() boundaries in strncpy_from_user() and strnlen

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:   Thu, 23 Jan 2020 10:47:42 -0800
From:   Linus Torvalds <torvalds@...ux-foundation.org>
To:     Christophe Leroy <christophe.leroy@....fr>,
        linux-arch <linux-arch@...r.kernel.org>
Cc:     Andrew Morton <akpm@...ux-foundation.org>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
        linuxppc-dev <linuxppc-dev@...ts.ozlabs.org>,
        Linux-MM <linux-mm@...ck.org>
Subject: Re: [PATCH] lib: Reduce user_access_begin() boundaries in
 strncpy_from_user() and strnlen_user()

On Thu, Jan 23, 2020 at 12:34 AM Christophe Leroy
<christophe.leroy@....fr> wrote:
>
> The range passed to user_access_begin() by strncpy_from_user() and
> strnlen_user() starts at 'src' and goes up to the limit of userspace
> allthough reads will be limited by the 'count' param.
>
> On 32 bits powerpc (book3s/32) access has to be granted for each 256Mbytes
> segment and the cost increases with the number of segments to unlock.
>
> Limit the range with 'count' param.

Ack. I'm tempted to take this for 5.5 too, just so that the
unquestionably trivial fixes are in that baseline, and the
infrastructure is ready for any architecture that has issues like
this.

Adding 'linux-arch' to the participants, to see if other architectures
are at all looking at actually implementing the whole
user_access_begin/end() dance too..

               Linus