[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Date: Mon, 30 Jul 2018 07:35:11 -0700
From: Sandeep Patil <sspatil@...gle.com>
To: "Theodore Y. Ts'o" <tytso@....edu>,
Steven Rostedt <rostedt@...dmis.org>, Jann Horn <jannh@...gle.com>,
salyzyn@...gle.com, Nick Desaulniers <ndesaulniers@...gle.com>,
Golden_Miller83@...tonmail.ch, Greg KH <greg@...ah.com>,
Kees Cook <keescook@...gle.com>, salyzyn@...roid.com,
kernel list <linux-kernel@...r.kernel.org>,
Ingo Molnar <mingo@...hat.com>, kernel-team@...roid.com,
stable@...r.kernel.org,
Kernel Hardening <kernel-hardening@...ts.openwall.com>,
Jeffrey Vander Stoep <jeffv@...gle.com>
Subject: Re: [PATCH] tracing: do not leak kernel addresses
On Fri, Jul 27, 2018 at 08:04:28PM -0400, Theodore Y. Ts'o wrote:
> On Fri, Jul 27, 2018 at 03:05:43PM -0700, Sandeep Patil wrote:
> > On Fri, Jul 27, 2018 at 04:21:14PM -0400, Theodore Y. Ts'o wrote:
> > > On Fri, Jul 27, 2018 at 04:11:03PM -0400, Steven Rostedt wrote:
> > > > That said, I would assume that
> > > > other Android utilities are using other debugfs files for system
> > > > status and such.
> >
> > As of today, I think a lot of information in 'bugreports' is read
> > out of debugfs (including things like binder stats). We do have a plan
> > to change that.
>
> Hmm, if it's only for bugreports, maybe it can be only mounted when
> about root processes getting tricked into reading from debugfs.
Yes, that's an interesting idea. May be a quicker way to get ourselves
rid of relying on debugfs. We need some platform cleanup to remove
all debugfs accessing code that's not "debug only" first. That work has been
ongoing ..
>
>
> > Indeed, I think it can. However, the problem is the last time I tried to
> > remove this a whole bunch of things just broke. So, it wasn't about losing
> > a functionality here and there. Agree, we need to clean up platform to not use
> > debugfs first. Then we can expect Apps or other native processes to not rely
> > on debugfs at all.
>
> Is Android controlling access to debugfs files via SELinux? If so,
> then access to debugfs can be gradually cranked down as use cases are
> removed.
Yes, that's what we've done now, so we know where the code is that depends on
it and working on moving it out. New domains aren't allowed to rely on
debugfs now.
- ssp
>
> - Ted
>
>
>
Powered by blists - more mailing lists
Confused about mailing lists and their use?
Read about mailing lists on Wikipedia
and check out these
guidelines on proper formatting of your messages.
