Re: 32-bit FreeBSD under 64-bit KVM

[Aurelien Jarno <aurelien-rXXEIb44qovR7s880joybQ@public.gmane.org>]

On Tue, Feb 27, 2007 at 09:08:50AM +0200, Avi Kivity wrote:
> Aurelien Jarno wrote:
> >Hi all,
> >
> >While I am able to run a 64-bit FreeBSD under a 64-bit KVM perfectly, I
> >am having problems with running a 32-bit FreeBSD under the same KVM:
> >
> >INIT: version 2.86 booting
> >INIT: PANIC: segmentation violation! sleeping for 30 seconds.
> >pid 34 (init), uid 0: exited on signal 11
> >INIT: PANIC: segmentation violation! sleeping for 30 seconds.
> >INIT: PANIC: segmentation violation! sleeping for 30 seconds.
> >INIT: PANIC: segmentation violation! sleeping for 30 seconds.
> >INIT: PANIC: segmentation violation! sleeping for 30 seconds.
> >INIT: PANIC: segmentation violation! sleeping for 30 seconds.
> >INIT: PANIC: segmentation violation! sleeping for 30 seconds.
> >INIT: PANIC: segmentation violation! sleeping for 30 seconds.
> >
> >Basically the kernel boots correctly, but then the userland fails to
> >work. I am using KVM-15 on an AMD64 CPU.
> >
> >I am able to run the same OS with -no-kvm, also I am able to run a
> >32-bit Linux under a 64-bit KVM. It seems the 32-bit support is at least
> >not totally broken, though I haven't made extensive tests.
> >
> >Does somebody is experiencing the same problems? Do you have a solution?
> >
> >  
> 
> Can you try turning the '#undef AUDIT' to '#define AUDIT' in mmu.c?  
> Also, change 'int dbg = 1' to 'int dbg = 0' in the same file.  Note kvm 
> will be very slower after this.  dmesg will show any issues (issue a 
> 'echo 9 > /proc/sysrq-trigger' to make sure all messages are shown)
> 

Yep I confirm it is very slow, especially when your tty0 device is
actually a serial port running at 9600 bauds!

I also got some problems, as it seems logging so much data slowed down
the machine, including the emulated machine. The segmentation violation
message disappeared, but the OS was like in an endless loop. So the
problem _may_ be due to a race condition.

After a lot of tries, I found that I was able to get the segmentation
violation by sending the kernel log to another machine. The full log is
available on http://temp.aurel32.net/kvm-gnu_kfreebsd32.log.bz2. But I
don't know what to look for in a such big log.

Bye,
Aurelien

-- 
  .''`.  Aurelien Jarno	            | GPG: 1024D/F1BCDB73
 : :' :  Debian developer           | Electrical Engineer
 `. `'   aurel32-8fiUuRrzOP0dnm+yROfE0A@public.gmane.org         | aurelien-rXXEIb44qovR7s880joybQ@public.gmane.org
   `-    people.debian.org/~aurel32 | www.aurel32.net

-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV