lkml.org 
[lkml]   [1997]   [Aug]   [7]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: other platforms and Linux compile...


On Thu, 7 Aug 1997, Albert D. Cahalan wrote:

> Doesn't that eliminate most stack smashing exploits?
> You can only hit function pointers, not the return address.
> Most overwrites would fill empty stack space until the
> stack limit is exceeded.
>
> This looks better than preventing stack execution.
>
> With the PowerPC binary compatibility, that would mean
> the stack grows up on normal PowerMac Linux too, right?

No, it grows down on all PPC API I know of (SYSV and PowerOpen).

> I think I recall a flag in the Intel CPU to do that too.

You are probably getting confused by the expand-down flag in the segment
descriptors. It just controls whether the valid address are either between
0 and the limit or between the limit and the maximum size. On Intel
the stack grows down (pushes and calls decrement the stack pointer, pops
and ret increment it).

Gabriel.

\
 
 \ /
  Last update: 2005-03-22 13:40    [W:0.601 / U:0.056 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site