lkml.org 
[lkml]   [2011]   [Nov]   [29]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [PATCH RFC 0/5] uprobes: kill xol vma
> 
> On top of this series, not for inclusion yet, just to explain what
> I mean. May be someone can test it ;)
>
> This series kills xol_vma. Instead we use the per_cpu-like xol slots.
>
> This is much more simple and efficient. And this of course solves
> many problems we currently have with xol_vma.
>
> For example, we simply can not trust it. We do not know what actually
> we are going to execute in UTASK_SSTEP mode. An application can unmap
> this area and then do mmap(PROT_EXEC|PROT_WRITE, MAP_FIXED) to fool
> uprobes.
>
> The only disadvantage is that this adds a bit more arch-dependant
> code.
>
> The main question, can this work? I know very little in this area.
> And I am not sure if this can be ported to other architectures.

Nice idea. I think this will help us in implementing boosted uprobes if
tweak a bit. (i.e having a jump after the actual instruction that gets
us back to the actual instruction stream). The current method of using a
first cum-first-serve slot reservation doesnt work for booster because
we have had to clear the slot in the post processing.

I will apply your patches and test and let you know how it goes. (in a day
or two).

--
Thanks and Regards
Srikar



\
 
 \ /
  Last update: 2011-11-29 11:35    [W:0.404 / U:0.140 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site