linux-kernel - Re: LTP: list of failures on 32bit and compat mode

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 for Android: free password hash cracker in your pocket

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

Message-ID: <ZC7P9BgW1CTFAMjB@yuki>
Date:   Thu, 6 Apr 2023 15:58:12 +0200
From:   Cyril Hrubis <chrubis@...e.cz>
To:     Arnd Bergmann <arnd@...db.de>
Cc:     Petr Vorel <pvorel@...e.cz>,
        Naresh Kamboju <naresh.kamboju@...aro.org>,
        open list <linux-kernel@...r.kernel.org>,
        LTP List <ltp@...ts.linux.it>, llvm@...ts.linux.dev,
        Nathan Chancellor <nathan@...nel.org>,
        Anders Roxell <anders.roxell@...aro.org>,
        Daniel Díaz <daniel.diaz@...aro.org>,
        Benjamin Copeland <ben.copeland@...aro.org>,
        Tudor Cretu <tudor.cretu@....com>
Subject: Re: LTP: list of failures on 32bit and compat mode

Hi!
> > Thanks! I've just searched in musl as well, because it didn't make sense to me
> > it'd be a code for LTP.
> >
> > "to catch the fault on ltp" I wonder if it's not actually musl bug.
> 
> No, musl is fine here. The problem is that ltp passes an invalid pointer,
> expecting to get -EFAULT from the kernel when that faults in
> copy_to_user().
> 
> There is nothing wrong with musl sanitizing the data behind that
> pointer, but then you get a signal instead of the EFAULT error.

That's actually quite common, usually we fix that by running the test in
a child and treating SEGFAULT as a PASS.

-- 
Cyril Hrubis
chrubis@...e.cz