[coreboot] Problems porting H8dmr_fam10 to H8qme-2+
Myles Watson
mylesgw at gmail.com
Tue Dec 22 19:30:10 CET 2009
On Tue, Dec 22, 2009 at 9:22 AM, Knut Kujat <knuku at gap.upv.es> wrote:
> Hello,
>
> as Myles suggested to disable siblings to see if I can pass through this
> weird exception and the impossibility to do so because of the compile error
> I changed the physical cpu option to 1 and it worked! But increasing it back
> to 2 or 4 made the exception come back again.
> I told you, Myles, I increased stack size to 4000 that was a filthy lie
> because I thought I'm increasing it to 4000 what I didn't see was that the
> same option was repeated at the end of the Options.lb file with
> STACK_SIZE=8000
>
It's always good to check targets/vendor/board/build/fallback/ldoptions to
see what's really being used.
> (So I don't know why the printks started working). Now fooling around with
> stack size and setting it up to 10000 all 4 cpus started working and I got a
> grub menu :) in text mode :( so I have a graphics Initializing faild and
> Linux doesn't boot up completly.
>
Great. I think we're getting to where we should add your board to the
tree. Then we can see the device tree too.
> I attached a complete log file, it is not so complete because the first
> lines of linux boot up are missing because I had to change serial speed on
> minicom. Thats because I'm having trouble of setting a speed and getting a
> total different one.
>
> Now I thing that my device tree is not completely working and thats why
> linux got some collusion at the beginning ??
>
It device 0:02.3 isn't getting a driver. 1:06.0 is not found.
PCI: 08:01.0 Hypertransport link capability not foundPCI: pci_scan_bus for
bus 08
That doesn't look good.
PCI: Left over static devices:
PCI: 08:01.0
PCI: 08:01.1
PCI: 08:02.0
> And I have no idea why graphic mode doesn't work since it looks like it
> finds vga without any problem.
>
VGA: PCI: 00:18.0 (aka node 0) link 2 has VGA device
That doesn't look right. I would think it's on link 0. I don't know why
that's set wrong, but it would explain why it's not working.
Thanks,
Myles
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.coreboot.org/pipermail/coreboot/attachments/20091222/3bd5118c/attachment.html>
More information about the coreboot
mailing list