Re: Kernel oops with kvm-38, 100% cpu usage with kvm-35, 36

[Avi Kivity <avi-atKUWr5tajBWk0Htik3J/w@public.gmane.org>]

Paquet Daniel wrote:
> Hello,
>
>   I have a problem with kvm-38, there is a kernel oops and kvm dies with the guest os without even completely booting. KVM dies with reason 18240.
>
> I start my guests with : kvm -no-rtc -hda disks/windows-xp.qcow2 -localtime -m 512 -net nic,model=ne2k_pci,macaddr=00:00:00:00:00:01 -net tap,ifname=tap0 &
>
> I was trying kvm-38 because my machines are useless with kvm-35 and 36, they start to boot then they jump at 100% cpu and hang for 5-6 sec then restart to run for 3-4 sec then hang.. and so on... Sometimes they run okay like 1 time on 6 tries, what I do is I close the guest machine then unload the kvm modules and reload them and restart the guest machine, it doesn't work all the time.
>
> Sometimes I dont even unload and reload the modules and the guest machine starts okay. I am running a vanilla kernel 2.6.22.3 on debian.
>
>  Here is the logs from the kernel.
> ##################################################
> Sep 12 09:37:42 res17 kernel: vmwrite error: reg 681c value c03a3228 (err -1069927896)
>   

This should be fixed now.


-- 
Any sufficiently difficult bug is indistinguishable from a feature.


-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/