linux-kernel - Re: [PATCH] riscv: fix syscall_get_arguments() and syscall_set

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:   Fri, 29 Mar 2019 16:32:08 -0400
From:   Steven Rostedt <rostedt@...dmis.org>
To:     "Dmitry V. Levin" <ldv@...linux.org>
Cc:     David Abdurachmanov <david.abdurachmanov@...il.com>,
        Will Drewry <wad@...omium.org>,
        Kees Cook <keescook@...omium.org>,
        Palmer Dabbelt <palmer@...ive.com>,
        linux-kernel@...r.kernel.org,
        Andy Lutomirski <luto@...capital.net>,
        Ingo Molnar <mingo@...hat.com>, linux-riscv@...ts.infradead.org
Subject: Re: [PATCH] riscv: fix syscall_get_arguments() and
 syscall_set_arguments()

On Fri, 29 Mar 2019 21:11:09 +0300
"Dmitry V. Levin" <ldv@...linux.org> wrote:

> On Fri, Mar 29, 2019 at 01:56:35PM -0400, Steven Rostedt wrote:
> > On Fri, 29 Mar 2019 18:52:18 +0100
> > David Abdurachmanov <david.abdurachmanov@...il.com> wrote:
> >   
> > > I have alternative version posted in December part of SECCOMP
> > > patchset which is based on arm64 implementation.
> > > 
> > > http://lists.infradead.org/pipermail/linux-riscv/2018-December/002450.html
> > > 
> > > I noticed that SECCOMP wasn't working properly if filters were
> > > checking syscall arguments, because populated arguments were wrong.
> > > 
> > > Btw, I plan to send v2 of SECCOMP patchset soonish.  
> > 
> > Please do. I want to get my patch series out, which will require these
> > changes.  
> 
> Sorry, I haven't seen the alternative patch posted by David before.
> Apparently, besides fixing the bug it also introduces new sanity checks
> of "i" and "n" arguments in syscall_get_arguments() and
> syscall_set_arguments().
> 
> Given that your patchset removes these arguments completely,
> I see little sense in adding new checks that are going to be removed
> by the subsequent commit in the series.

I agree. I'm going to pull in Dmitry's patches as my patches are going
to rewrite most the code anyway, and no need for the extra churn of the
sanity checks that are going to become irrelevant immediately afterward.

-- Steve