Messages in this thread |  | | Date | Fri, 13 Dec 2013 09:58:48 -0800 | Subject | Re: [PATCH 1/9] Known exploit detection | From | Kees Cook <> |
| |
On Thu, Dec 12, 2013 at 5:42 PM, Greg Kroah-Hartman <gregkh@linuxfoundation.org> wrote: > On Thu, Dec 12, 2013 at 07:25:23PM -0500, Dave Jones wrote: >> On Thu, Dec 12, 2013 at 01:13:41PM -0800, Kees Cook wrote: >> >> > - who will keep adding these triggers going forward? >> >> also.. >> >> - Who will test the existing triggers are doing the right thing when related code changes. > > And: > - how do you determine an "expoit attempt" from "userspace program > doing something stupid" / "corrupted filesytem mounted"? > > I really don't like this, it means that our normal error handling for > userspace data will suddenly all have CVE entries on them over time. > How is that helpful to anyone?
These locations tend to be very hard to reach accidentally, or userspace never exercised the path (which is why the flaws go unnoticed usually). Having userspace trip over them is unlikely, so we'd want to know about that anyway. If something actually turns noisy, we drop it. But in at least the i915 case, it took seriously careful work to hit the flawed code path.
> Think ahead in 10-20 years, what is the code paths going to look like > then? Horrible...
If we have that many CVE in the moving 5 year window, then we should certainly feel worse about that reality than just having a few extra lines in the code. :)
Keeping this up at the memory-corruption or permissions bypass level means we won't annotate the vast majority of CVEs that are usually low priority issues.
-Kees
-- Kees Cook Chrome OS Security
|  |