On Mon, 2012-05-14 at 08:48 +0200, Rik Theys wrote:
Hi,
On 05/12/2012 11:32 PM, Ben Hutchings wrote:
On Sat, 2012-05-12 at 16:25 -0500, Jonathan Nieder wrote:
Ben Hutchings wrote:
Which shows that the segfault is always at the same code address:
[ 56.663596] lvm[540]: segfault at ffffffffff600400 ip ffffffffff600400 sp 00007fff25461ec8 error 5
[ 76.174282] exe[541]: segfault at ffffffffff600400 ip ffffffffff600400 sp 00007fffa69b3388 error 5
[ 78.307062] exe[542]: segfault at ffffffffff600400 ip ffffffffff600400 sp 00007fff33270d08 error 5
[ 87.775183] exe[543]: segfault at ffffffffff600400 ip ffffffffff600400 sp 00007ffffb125068 error 5
[ 97.937356] exe[545]: segfault at ffffffffff600400 ip ffffffffff600400 sp 00007fffb53be498 error 5
[ 108.789157] lvm[547]: segfault at ffffffffff600400 ip ffffffffff600400 sp 00007fff0e012348 error 5
This address is not accessible in user-mode, and probably isn't used by
the kernel either.
Nice lead. Looks like
http://thread.gmane.org/gmane.linux.kernel/1248253/focus=1254330
Agreed. Rik, which version of the kernel is the hypervisor from?
The hypervisor is CentOS 6.2 with kernel version
2.6.32-220.7.1.el6.x86_64 and qemu-kvm-0.12.1.2-2.209.el6_2.4.x86_64.
OK, so it doesn't look we have a bug to fix.
Based on that email thread I think you can work around this with
'vsyscall=native' on the guest's kernel command line. The down-side of
this is that it makes it easier to exploit some types of bug for
privilege escalation.