| Summary: | sys-kernel/hardened-sources-2.6.32-r89: Unbootable kernel with CONFIG_GRKERNSEC_SETXID=y | ||
|---|---|---|---|
| Product: | Gentoo Linux | Reporter: | Torbjörn Svensson <azoff> |
| Component: | [OLD] Core system | Assignee: | The Gentoo Linux Hardened Kernel Team (OBSOLETE) <hardened-kernel+disabled> |
| Status: | RESOLVED FIXED | ||
| Severity: | critical | CC: | hardened |
| Priority: | Normal | ||
| Version: | unspecified | ||
| Hardware: | AMD64 | ||
| OS: | Linux | ||
| Whiteboard: | |||
| Package list: | Runtime testing required: | --- | |
|
Description
Torbjörn Svensson
2012-02-18 14:02:34 UTC
> pipacs in #grsecurity asked me to try a later version of spender's patch, and
> it worked like a charm.
>
> Please bump to a later version of the grsec patch.
>
> Both sys-kernel/hardened-sources-2.6.32-r90 and
> sys-kernel/hardened-sources-2.6.32-r91 seams to be working fine.
>
Thanks for the report. Unfortunately, hardened-sources-2.6.32-r89 was fastrack stabilized with hardened-sources-3.2.2-r1 to deals with a information leak out of /proc. So I wonder if the same problem is in 3.2.2-r1. Can you test that one with as close a config as possible and see if you hit the same hangup. If so, I'll fastrack 2.6.32-r90 and 3.2.5.
I've just tested sys-kernel/hardened-sources-3.2.2-r1 and it works. In spender's changelog for stable I found this (probably related): commit cf5731b0b8ac026402e004a62cc3b0c522b5d9d2 Author: Brad Spengler <spender@grsecurity.net> Date: Tue Feb 7 17:21:00 2012 -0500 Add current_is_single_threaded() fix I applied to the test branch but forgot to apply to stable when backporting GRKERNSEC_SETXID I've just tested sys-kernel/hardened-sources-2.6.32-r89 with the "if(!mm) return true;" fix for current_is_single_threaded(void) and it boots! Have no idea what other impact it has though. Just for the exercise, I also added a WARN_ON(!mm); and I got 38 hits during a normal boot up, so I'm quite surprised that no one else has run into this issue. (In reply to comment #2) > I've just tested sys-kernel/hardened-sources-3.2.2-r1 and it works. > > In spender's changelog for stable I found this (probably related): > > commit cf5731b0b8ac026402e004a62cc3b0c522b5d9d2 > Author: Brad Spengler <spender@grsecurity.net> > Date: Tue Feb 7 17:21:00 2012 -0500 > > Add current_is_single_threaded() fix I applied to the test branch but > forgot to apply to stable when backporting GRKERNSEC_SETXID > > > > I've just tested sys-kernel/hardened-sources-2.6.32-r89 with the "if(!mm) > return true;" fix for current_is_single_threaded(void) and it boots! Have no > idea what other impact it has though. > > Just for the exercise, I also added a WARN_ON(!mm); and I got 38 hits during a > normal boot up, so I'm quite surprised that no one else has run into this > issue. I'm surprised too! I do test these kernel, but I can't test every combination of feature and hardware. Usually the community comes forward. Okay I'm leaving 3.2.2-r1 alone, stabilizing 2.6.32-r90 and removing 2.6.32-r89.
> Okay I'm leaving 3.2.2-r1 alone, stabilizing 2.6.32-r90 and removing
> 2.6.32-r89.
Done.
|
