c-d.hailfinger.devel.2006 at gmx.net
Wed May 23 00:41:43 CEST 2007
On 11.05.2007 21:29, yhlu wrote:
> On 5/11/07, Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006 at gmx.net> wrote:
>> Hi ST,
>> On 08.05.2007 22:45, ST wrote:
>>>> could someone with a GA-M57SLI please post a lspci -vvvnx to the
>>>> list, preferably both under LB and factory BIOS? I'm trying to
>>>> verify some code I wrote.
>>> Attached to this mail you will find the output of lspci -vvvnx under the
>>> original bios version f8 and under linuxbios updated from today (2007-05-08).
>> A few interesting differences between the dumps:
>> * LinuxBIOS has subsystem vendor/id 1022:2b80 for all devices
>> -> Bug or feature?
> you could change it in MB Option.lb. I never boot that MB to normal
> BIOS to check subsystem vendor/id and just use values for AMD cheetah.
Normal BIOS subsystem vendor/device ids are a problem because we
use them in flashrom to look up flash enable GPIO pins. I have a
patch which changes the subsystem vendor id for the GA-M57SLI,
BUT... the subsystem device id is not constant for a given board,
the GA-M57SLI has ~7 different subsystem device ids. Can we assign
different sub-devids to different devices? The code seems to
be unable to cope with that.
More information about the coreboot