Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 15 Sep 2009 16:22:26 +0100
From:      Matt Dawson <matt@chronos.org.uk>
To:        freebsd-security@freebsd.org
Subject:   Re: Protecting against kernel NULL-pointer derefs
Message-ID:  <200909151622.26589.matt@chronos.org.uk>
In-Reply-To: <8663bk2xcb.fsf@ds4.des.no>
References:  <4AAF4A64.3080906@thedarkside.nl> <4AAF8775.7000002@thedarkside.nl> <8663bk2xcb.fsf@ds4.des.no>

next in thread | previous in thread | raw e-mail | index | archive | help
On Tuesday 15 Sep 2009 13:42:28 Dag-Erling Sm=C3=B8rgrav wrote:

> there must be a better way to make your case than to sow FUD?

Where? To paraphrase yourself: Please define "sowing FUD." There's an issue=
;=20
there have been two previously. Nobody is blaming anyone, blowing it out of=
=20
proportion, leaving FBSD in droves or pointing fingers. We know it's local =
and=20
we're all well aware of the axiom "if someone else has physical access to y=
our=20
box, it isn't your box any more." I don't see or feel any fear, uncertainty=
 or=20
doubt. I just see a concerned but dedicated FBSD user discussing an issue=20
sensibly with the information currently to hand.

Providing it does not seriously affect anything else (Pieter has already as=
ked=20
for information and opinions before the thread went off on this tangent), i=
f=20
setting this #define downgrades arbitrary code execution vulnerabilities an=
d=20
privilege escalations to crashes where system and, more importantly IMHO, h=
ost=20
integrity is preserved then I am all for it. I'd certainly rather have a Do=
S=20
and the risk of cached data loss than a potential information leak or a=20
reputation-destroying spamming session run. That we don't have multiple pla=
ces=20
that this could be overridden/reset similar to the SELinux issue also inspi=
res=20
confidence in Pieter's method. As simple as it seems, it would appear to be=
=20
(sorry, buzzword-that-fits coming up) proactive in its approach, addressing=
=20
and mitigating any future issues of this type and limiting the possible=20
damage.

Also worth thinking about: Do we need to consider -fno-delete-null-pointer-
checks or a downgrade to -O for kernel/world optimisation level for now unt=
il=20
we're sure there are no more of these lurking? Linux found out the hard way=
=20
that code order matters when compiling at >-O and that perfectly acceptable=
=20
code coupled with assumptions made by the compiler can bite you in the=20
backside.
=2D-=20
Matt Dawson
MTD15-RIPE
matt@chronos.org.uk



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200909151622.26589.matt>