[lkml]   [2013]   [Oct]   [2]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRe: [PATCH v6 0/7] Kernel base address randomization
On Tue, Oct 1, 2013 at 10:30 PM, H. Peter Anvin <> wrote:
> On 10/01/2013 10:25 PM, Ingo Molnar wrote:
>> I mean, for example in an oops message we print data in words: the RIP,
>> other registers and stack contents. If any of these values lies within the
>> randomization range then we could de-randomize it.
>> So instead of exposing randomized values, we could expose de-randomized
>> values.
>> ( This isn't fool-proof: if some data value happens to lie within the
>> random range spuriously then we'll incorrectly transform it. In the
>> context of oops messages this should not be a big practical problem
>> though. )
> I don't agree that this isn't a big practical problem. I often find it
> necessary to pick out "things that look like pointers". Overall,
> derandomization would make it possible to get really confused when you
> have things like half a pointer overwritten.

I think reflecting the reality of the system is the correct way to go.
Attempting to do the derandomization on the live system seems
extremely fragile. It's much cleaner to have a "true" view of the
running system and work from there. I don't want to have to wonder if
my kernel is lying to me about where things are in memory any more
than it already does. :)


Kees Cook
Chrome OS Security

 \ /
  Last update: 2013-10-02 08:01    [W:0.071 / U:1.640 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site