linux-kernel - Re: [PATCH v5 01/13] taint: Introduce a new taint flag (insecure)

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]

Date:   Tue, 5 Feb 2019 13:21:46 -0800
From:   Andrew Morton <akpm@...ux-foundation.org>
To:     Andy Lutomirski <luto@...nel.org>
Cc:     "Chang S. Bae" <chang.seok.bae@...el.com>,
        Thomas Gleixner <tglx@...utronix.de>,
        Ingo Molnar <mingo@...nel.org>,
        "H . Peter Anvin" <hpa@...or.com>, Andi Kleen <ak@...ux.intel.com>,
        Markus T Metzger <markus.t.metzger@...el.com>,
        Ravi Shankar <ravi.v.shankar@...el.com>,
        LKML <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH v5 01/13] taint: Introduce a new taint flag (insecure)

On Fri, 1 Feb 2019 18:42:29 -0800 Andy Lutomirski <luto@...nel.org> wrote:

> On Fri, Feb 1, 2019 at 12:54 PM Chang S. Bae <chang.seok.bae@...el.com> wrote:
> >
> > For testing (or root-only) purposes, the new flag will serve to tag the
> > kernel taint accurately.
> >
> > When adding a new feature support, patches need to be incrementally
> > applied and tested with temporal parameters. Currently, there is no flag
> > for this usage.
> 
> I think this should be reviewed by someone like akpm.  akpm, for
> background, this is part of an x86 patch series.  If only part of the
> series is applied, the kernel will be blatantly insecure (but still
> functional and useful for testing and bisection), and this taint flag
> will be set if this kernel is booted.  With the whole series applied,
> there are no users of the taint flag in the kernel.
> 
> Do you think this is a good idea?

What does "temporal parameters" mean?  A complete description of this
testing process would help.

I sounds a bit strange.  You mean it assumes that people will partially
apply the series to test its functionality?  That would be inconvenient.

- Can the new and now-unused taint flag be removed again at
  end-of-series?

- It would be a lot more convenient if we had some means of testing
  after the whole series is applied, on a permanent basis - some
  debugfs flag, perhaps?