1 |
-----BEGIN PGP SIGNED MESSAGE----- |
2 |
Hash: SHA1 |
3 |
|
4 |
|
5 |
Hi everyone, |
6 |
|
7 |
All kernels before Sept 14 are vulnerable to the "IA32 Syscall Entry |
8 |
Point Privilege Escalation" and "IA32 Emulation Stack Underflow". See |
9 |
|
10 |
http://bugs.gentoo.org/show_bug.cgi?id=337645 |
11 |
http://bugs.gentoo.org/show_bug.cgi?id=337659 |
12 |
|
13 |
Also see |
14 |
|
15 |
https://bugs.gentoo.org/show_bug.cgi?id=326885#c10 |
16 |
|
17 |
|
18 |
As a result, certain configurations of hardened-sources are also |
19 |
vulnerable. As a work around until I get the fix into the tree and fast |
20 |
track stabilization, keep the following in mind: |
21 |
|
22 |
|
23 |
1) Whether hardened or not, if you don't have CONFIG_IA32_EMULATION, the |
24 |
exploits fail. |
25 |
|
26 |
|
27 |
2) If you hide kernel symbols in /proc/kallsyms, the proof-of-concept |
28 |
code won't work. You can do that by either not enabling CONFIG_KALLSYMS |
29 |
on non-hardened kernels, or just set CONFIG_GRKERNSEC_HIDESYM=y on |
30 |
hardened. |
31 |
|
32 |
(However, there may still be ways of making the exploit work even |
33 |
without symbol info.) |
34 |
|
35 |
|
36 |
3) On hardened systems, if you enable CONFIG_PAX_MEMORY_UDEREF=y, the |
37 |
exploits fail even with access to symbol info. If possible, I would |
38 |
also recommend enabling CONFIG_PAX_KERNEXEC=y. |
39 |
|
40 |
|
41 |
- -- |
42 |
Anthony G. Basile, Ph.D. |
43 |
Gentoo Developer |
44 |
-----BEGIN PGP SIGNATURE----- |
45 |
Version: GnuPG v2.0.16 (GNU/Linux) |
46 |
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ |
47 |
|
48 |
iEYEARECAAYFAkySiOIACgkQl5yvQNBFVTUZzQCeMolKjTKql6/ShNRtYSH/K1DM |
49 |
thUAmwTJOrYbB1wJ4A+FlPDu78tc55AT |
50 |
=xfQc |
51 |
-----END PGP SIGNATURE----- |