[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20161109034209.i5ok4cy2apjp3mpk@x>
Date: Tue, 8 Nov 2016 19:42:09 -0800
From: Josh Triplett <josh@...htriplett.org>
To: Arnd Bergmann <arnd@...db.de>
Cc: Andrew Morton <akpm@...ux-foundation.org>,
Kees Cook <keescook@...omium.org>,
Johannes Weiner <hannes@...xchg.org>,
Ingo Molnar <mingo@...nel.org>,
Andy Lutomirski <luto@...nel.org>,
Petr Mladek <pmladek@...e.com>,
Thomas Garnier <thgarnie@...gle.com>,
Ard Biesheuvel <ard.biesheuvel@...aro.org>,
Nicolas Pitre <nicolas.pitre@...aro.org>,
Zefan Li <lizefan@...wei.com>,
Li Bin <huawei.libin@...wei.com>,
"Eric W. Biederman" <ebiederm@...ssion.com>,
Dmitry Vyukov <dvyukov@...gle.com>,
Ralf Baechle <ralf@...ux-mips.org>,
Alex Thorlton <athorlton@....com>,
Michal Hocko <mhocko@...e.com>,
Mateusz Guzik <mguzik@...hat.com>,
Cyrill Gorcunov <gorcunov@...nvz.org>,
John Stultz <john.stultz@...aro.org>,
Al Viro <viro@...iv.linux.org.uk>, Zach Brown <zab@...hat.com>,
Anna Schumaker <Anna.Schumaker@...app.com>,
Dave Hansen <dave.hansen@...el.com>,
linux-kernel@...r.kernel.org, linux-api@...r.kernel.org,
Nicolas Pitre <nico@...aro.org>
Subject: Re: [PATCH 0/2] Support compiling out the prctl syscall
On Wed, Nov 09, 2016 at 01:26:53AM +0100, Arnd Bergmann wrote:
> On Tuesday, November 8, 2016 4:17:47 PM CET Josh Triplett wrote:
> > Some embedded systems can do without the prctl syscall, saving some space.
> >
> > This also avoids regular increases in tinyconfig size as people add more
> > non-optional functionality to prctl (observed via the 0-day kernel
> > infrastructure).
> >
> > The first patch moves prctl and all its helpers into a separate source file,
> > kernel/prctl.c, to allow making it optional via the Makefile without any
> > ifdefs. The second patch adds the Kconfig option to make prctl optional.
> >
> > Josh Triplett (2):
> > kernel: Move prctl and helpers from kernel/sys.c to new kernel/prctl.c
> > kernel: Support compiling out the prctl syscall
> >
>
> Both of these look really nice to me. Another syscall that comes to
> mind as a candidate to be made optional is ptrace, and I've added
> Nicolas Pitre to Cc here, as he has looked into this before.
I've looked at that as well; it's a rather more complex patch, as
several other bits of the kernel hook into ptrace or otherwise depend on
ptrace. (And ptrace_may_access will need to stick around even when
ptrace doesn't.)
- Josh Triplett
Powered by blists - more mailing lists