| 1 |
On Dec 13, 2003, at 13:39, Peter S. Mazinger wrote: |
| 2 |
> How about porting this to uClibc? |
| 3 |
|
| 4 |
I haven't looked at uClibc's code, so I can't say whether the glibc |
| 5 |
patch's techniques apply, but we'd be willing to investigate the issue. |
| 6 |
|
| 7 |
> What are the benefits used along with PaX and propolice? |
| 8 |
|
| 9 |
As I understand it, what PaX provides, among other things, is |
| 10 |
protection from execution on the heap, while propolice prevents |
| 11 |
exploitation of stack overflows. Our patch is in the same class of |
| 12 |
protective techniques as propolice, as it prevents vulnerability |
| 13 |
exploitation rather than execution of injected code. Thus, it's |
| 14 |
another layer of defense, because even with PaX enabled you could |
| 15 |
exploit a heap overflow and execute some exploit payload elsewhere in |
| 16 |
memory or otherwise subvert control flow. |
| 17 |
|
| 18 |
> Peter |
| 19 |
|
| 20 |
-- |
| 21 |
William Robertson |
| 22 |
Reliable Software Group, UC Santa Barbara |
| 23 |
http://www.cs.ucsb.edu/~wkr/ |
| 24 |
|
| 25 |
|
| 26 |
-- |
| 27 |
gentoo-hardened@g.o mailing list |
Archives