On Thu, Jan 8, 2009 at 11:00 AM, Ward Vandewege ward@gnu.org wrote:
On Thu, Jan 08, 2009 at 08:52:44AM -0700, Myles Watson wrote:
On Thu, Jan 8, 2009 at 6:58 AM, Ward Vandewege ward@gnu.org wrote:
On Wed, Jan 07, 2009 at 09:39:39PM -0700, Myles Watson wrote:
The rev that broke the boot was 1089 or 1090. 1089 does not build, so I can't tell for sure. 1088 boots fine. 1090 is broken.
They might have well been the same commit because they were interdependent. It was just an easier way of looking at the changes for me.
Thanks for tracking it down. Could you send me a log from 1088? Hopefully we can track it down quickly.
Sure, see attached.
The biggest difference that I see is that there is no graphics device (PCI 1.1) in the broken one. In 1088 it was found dynamically (not in the tree), but now it isn't found at all.
Since it was supposed to be disabled by: /* this board does not really have vga; disable it (pci device 00:01.1) */ unwanted_vpci = < 80000900 0 >; in the dts, I'm not sure why it disappeared now.
Possible solution:
- Try it again without the unwanted_vpci line in the dts.
I guess I don't understand why the device is gone now. I didn't think the changes I've made would make the device disappear.
I have not tried that yet, but here's a boot log from v1108, which still does not boot but it stops in a different place now, it seems.
Thanks for your patience. One problem is that there is no area reserved for the APIC and the ROM. I think we should add a reserved resource in the domain that goes from 0xfc000000-0xffffffff. That will stop allocations from going there.
I copied the reserved areas from i440bx_emulation. Like the comments say, I know these need to be reserved areas, but I'm not exactly sure where they belong.
Compile tested.
Signed-off-by: Myles Watson mylesgw@gmail.com
Thanks, Myles