[coreboot] [LinuxBIOS] Intel refactoring and microcode updates

[Corey Osgood]

joe at smittys.pointclark.net wrote:
> Quoting Corey Osgood <corey.osgood at gmail.com>:
>
>> joe at smittys.pointclark.net wrote:
>>> Quoting Uwe Hermann <uwe at hermann-uwe.de>:
>>>
>>>
>>>> On Fri, Jan 11, 2008 at 03:54:18PM +0100, Peter Stuge wrote:
>>>>
>>>>> On Fri, Jan 11, 2008 at 09:42:00AM -0500, Corey Osgood wrote:
>>>>>
>>>>>> Peter Stuge wrote:
>>>>>>
>>>>>>> I think we need to make it configurable.
>>>>>>>
>>>>>> I don't like that. With a factory bios, you expect the correct
>>>>>> microcode update for your CPU to be present, no matter what CPU you
>>>>>> put in a socket.
>>>>>>
>>>>> (Actually no, not always.)
>>>>>
>>>>>
>>>>>
>>>>>> We should be able to do the same.
>>>>>>
>>>>> I agree, but we should also try to be even more flexible. I think we
>>>>> should allow inclusion of 0<=n<=all microcode updates. Definately an
>>>>> advanced option, but still.
>>>>>
>>>> Yep, that's what I meant. It's fine if the default is "all microcode
>>>> updates", but there should be an option for advanced users to only use
>>>> the one(s) you really want or need in order to save time and space.
>>>>
>>>>
>>>> Uwe.
>>>>
>>>>
>>> That makes perfect sense.....I like the advanced option idea:-)
>>>
>>> Thanks - Joe
>>
>> This seems like it would get very messy, very quickly. The only way I
>> can come up with to do it is #if's for every single ID, perhaps in some
>> intel_custom folder, so it doesn't make a mess of the existing stuff.
>>
>> Anyways, I'll get on top of testing this weekend, I have slot 1 (440zx),
>> socket 370 (i810), and socket 479 (i830) boards I can test with,
>> assuming they all still work. If those work, I'll bet the farm that the
>> rest of them do. Everything passes abuild, btw.
>>
>> -Corey
>>
> Yeh, even though it is a good idea it probibly would complicate things
> alot. For now it may be better to just commit a piece at a time for
> each processor tested. I'm really interested in how the socket 479
> (Pentium M) testing goes. Let us know:-)
>
>
> Thanks - Joe

I know ;) Uwe, do you have a 440bx board you can try this on? Mine has
officially bit the dust. I can say that i810/socket 370 works as
expected with 3 different cores. I'll get to the socket 479 if I have
time tomorrow, I've been tinkering with the i810 most of the last couple
days.

Thanks,
Corey