Re: [RFC] enhancing the kernel's graphics subsystem

[Jeff Garzik <jeff@garzik.org>]

Jon Smirl wrote:
> On 5/22/07, Jesse Barnes <jbarnes@virtuousgeek.org> wrote:
>> On Tuesday, May 22, 2007, Jon Smirl wrote:
>> > I've talked to an ATI engineer about VBIOS initialization. The chips
>> > may have different steppings. They flash the right VBIOS that matches
>> > the chip into the ROM on the card.  Given all the various steppings
>> > this is the only sane way to initialize the hardware. I don't believe
>> > generic initialization code that can handle all of the various
>> > steppings exists for any hardware.
>>
>> I think you must have misunderstood.  Last time I worked on ATI, there 
>> was
>> a single ROM image that supported several (maybe even all) of their addin
>> cards.  There's nothing magical about them...  But that's beside the
>> point.
> 
> What about the poke the obscure value into this undocumented IO port
> class of problem where they are using the chip test interface to work
> around bugs in the silicon.
> 
> He also mentioned that all of the OEMs change the standard ROM image
> before loading it onto their cards to handle how they wired things up
> and bugs in other chips.
> 
> Of course I don't have the source for the ROM so I don't know the
> accuracy of what I was told.

There is nothing magic about a ROM.  It's just software, or, often, data 
tables.

	Jeff