linux-kernel - Re: [RFC PATCH] kbuild: check uniqueness of basename of modules

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]

Message-ID: <201905151131.EBB45E5@keescook>
Date:   Wed, 15 May 2019 11:38:26 -0700
From:   Kees Cook <keescook@...omium.org>
To:     Masahiro Yamada <yamada.masahiro@...ionext.com>
Cc:     Linux Kbuild mailing list <linux-kbuild@...r.kernel.org>,
        Andrew Morton <akpm@...ux-foundation.org>,
        Sam Ravnborg <sam@...nborg.org>, Arnd Bergmann <arnd@...db.de>,
        Greg KH <gregkh@...uxfoundation.org>,
        Jessica Yu <jeyu@...nel.org>,
        Lucas De Marchi <lucas.de.marchi@...il.com>,
        Linus Torvalds <torvalds@...ux-foundation.org>,
        Rusty Russell <rusty@...tcorp.com.au>,
        Michal Marek <michal.lkml@...kovi.net>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>
Subject: Re: [RFC PATCH] kbuild: check uniqueness of basename of modules

On Thu, May 16, 2019 at 02:55:02AM +0900, Masahiro Yamada wrote:
> 
> On Thu, May 16, 2019 at 1:20 AM Kees Cook <keescook@...omium.org> wrote:
> >
> > On Wed, May 15, 2019 at 04:53:15PM +0900, Masahiro Yamada wrote:
> > > On Wed, May 15, 2019 at 4:40 PM Masahiro Yamada
> > > <yamada.masahiro@...ionext.com> wrote:
> > > >
> > > > [...]
> > > > diff --git a/scripts/modules-check.sh b/scripts/modules-check.sh
> > > > new file mode 100755
> > > > index 000000000000..944e68bd22b0
> > > > --- /dev/null
> > > > +++ b/scripts/modules-check.sh
> > > > @@ -0,0 +1,18 @@
> > > > +#!/bin/sh
> > > > +# SPDX-License-Identifier: GPL-2.0
> > > > +
> > > > +# Warn if two or more modules have the same basename
> > > > +check_same_name_modules()
> > > > +{
> > > > +       same_name_modules=$(cat modules.order modules.builtin | \
> > > > +                               xargs basename -a | sort | uniq -d)
> >
> > While probably it'll never be a problem, just for robustness, I'd add "--"
> > to the end basename to terminate argument interpretation:
> >
> >     xargs basename -a -- | sort | ...
> 
> 
> Sorry for my ignorance, but could you
> teach me the effect of "--" ?
> 
> 
> I sometimes use "--" as a separator
> when there is ambiguity in arguments
> for example, "git log <revision> -- <path>"
> 
> 
> In this case, what is intended by "--"?

It means "end of arguments" so that whatever xargs passes into the
program aren't interpretted as an argument. In this case, if there was
a module path somehow ever named --weird/build/path/foo.o, xargs would
launch basename as:

	basename -a --weird/build/path/foo.o

and basename would fail since it didn't recognize the argument. Having
"--" will stop argument parsing:

	basename -a -- --weird/build/path/foo.o

This is just a robustness suggestion that I always recommend for xargs
piping, since this can turn into a security flaw (though not here) when
an argument may have behavioral side-effects. So, it's just a thing that
always jumps out at me, though in this particular case I don't think
we could ever see it cause a problem, but better to always write these
xargs patterns as safely as possible.

-- 
Kees Cook