I have a reproducible kernel panic running the ltrace program. The
panic only occurs when running the uml kernel in skas mode. mode=tt
does not panic.
Running as a normal (non root) user:
$ ltrace strace ls
Kernel panic: Kernel mode signal 5
Is anyone else able to reproduce this?
UML is debian woody, running vanilla kernel 2.4.22 with
uml-patch-2.4.22-5.bz2. Also tested is vanilla kernel 2.4.23 with
uml-patch-2.4.23-1.bz2. The kernel panic occures on both of these
Host is debian woody 2.4.22 with preemt, lowlatency and skas3 patches
The panic occurs because the syscall number gets set to -1. I have not
been able to track down exactly where this occurs, and any help on this
would be warmly received.
Dion's Maxim: If you are ever surprised at just how stupid people can be,
then you haven't understood Dion's Maxim.
From: Sven 'Darkman' Michels <sven@da...> - 2004-01-20 11:54:29
> I have a reproducible kernel panic running the ltrace program. The
> panic only occurs when running the uml kernel in skas mode. mode=tt
> does not panic.
> Running as a normal (non root) user:
> $ ltrace strace ls
> results in:
> Kernel panic: Kernel mode signal 5
> Is anyone else able to reproduce this?
Yes, but not everytime:
Host: 2.4.22 with skas patch
First test: uml crashed.
Second: uml didn't crash, finished the ltrace just fine
(i did the ltrace strace ls two times, just to be sure)
Third test: uml crashed.
(reboot between every test).
So it may be some kind of race condition?
Get latest updates about Open Source Projects, Conferences and News.