On Thu, 4 May 2006 20:32:58 +0200
Manuel Bouyer <bouyer@antioche.eu.org> wrote:

> On Thu, May 04, 2006 at 02:36:48PM +0200, Marcin Jessa wrote:
> > Hi guys.
> > 
> > I set up RAID 0 with RaidFrame to run XEN on my 
> > CURRENT 3.99.18 system following 
> > http://www.netbsd.org/guide/en/chap-rf.html 
> > and
> > http://mail-index.netbsd.org/port-xen/2006/03/01/0010.html
> > 
> > The RaidFrame itself works fine but trying to boot GENERIC XEN0 kernel
> > with grub I get following error:
> > kernel(hd0,e)/xen.gz dom0_mem=524288
> >    [Multiboot-elf, <0x100000:0x3f080:0x35f80>, shtab=0x175078,
> > entry=0x100000] module (hd0,e)/netbsd-XEN0 root=/dev/hda1 ro console=pc
> > Error 14: Filesystem compatibility error, cannot read whole file
> 
> Is your RaidFrame volume raid0 or raid1 ? grub can only handle raid1
> (i.e. mirror)

Raid1. I may not have expressed myself clearly enough in my initial email.
Grub can boot my XEN0 fine. The problem/issue I was struggeling
with was the kernel size of the GENERIC XEN0 kernel.
GRUB was unable to boot an 8.5M uncompressed kernel image. 
But it booted a smaller, custom XEN0 kernel just fine.
So I was wondering about the restrictions on the kernel size GRUB can take.
And suggested a SoC project to improve our native bootloader code 
to be able to handle XEN so one didn't have to deal with silliness GRUB.
Since I have you "on the line" Manuel. Is XEN3 code usable yet ?
If not, any idea if Linux would work as a XENU domain with XEN2 on 
CURRENT NetBSD?


Cheers & Thanks,
Marcin