PCI: 02:06.0 30 <- [0x00fff80000 - 0x00fff7ffff] rom
It seems it can not be processed by pci_set_resource properly.
YH
-----Original Message----- From: Li-Ta Lo [mailto:ollie@lanl.gov] Sent: Wednesday, January 12, 2005 10:10 PM To: YhLu Cc: Sagiv Yefet; LinuxBIOS Subject: RE: Running with VGA
On Wed, 2005-01-12 at 23:07, YhLu wrote:
You assign the value to the new resource.
I wonder if it will confuse the resource allocator. Because the range is
out
of touch of allocator.
I don't think so but I am not sure neither. The resource is flagged as FIXED and ASSIGNED. The resource allocator should leave that resource as is. If it does not work, we can just write dev->rom_address to PCI_ROM_ADDRESS directly in pci_get_rom_resource() and the pci_rom_probe() can still be unchanged.
Could you try the code?
Ollie
YH
-----Original Message----- From: Li-Ta Lo [mailto:ollie@lanl.gov] Sent: Wednesday, January 12, 2005 9:48 PM To: YhLu Cc: Sagiv Yefet; LinuxBIOS Subject: RE: Running with VGA
On Wed, 2005-01-12 at 21:02, YhLu wrote:
Committed.
Also I added readme in /drivers/pci/onboard/onboard.c
YH
I made some modification which should utilize the resource allocation code and minimize the impact of the embedded ROM image. I was doing it blindly, please test it on your platform.
Ollie
On Thu, 2005-01-13 at 10:49, YhLu wrote:
PCI: 02:06.0 30 <- [0x00fff80000 - 0x00fff7ffff] rom
It seems it can not be processed by pci_set_resource properly.
YH
Is 0xffff8000 written to register 0x30? That resource is dummy so it should not be process by the resource allocator.
Anyway, probably we should give up (ab)using the allocator. I made another version to use dev->rom_address in pci_rom_probe().
Ollie