On 12 Mar 2002, Eric W. Biederman wrote:

> > I need to get new southbridge code that works like the superio code set
> > up, but this is something we need right now. We need the southbridge code
> > to work like the superio code I think since there may be systems out there
> > with more than one southbridge.
>
> And with hypertransport and intel's hub architecture this is common.
> I though that was working already..

yeah, sorry, I got done superio and got tired, and did not wrap up
southbridge. My fault.

> On the long term agenda, I want to build a table similiar to what
> pci_enumerate generates at compile time.  This should allows
> us to call pci_enumerate much later or not at all if all hardware
> is onboard.  Plus we can enumerate onboard devices that are not pci.

now that's a nice idea.

> How much of IDE enable do you need?

we're not sure, it still doesn't work :-)

It actually looks like all we need is to enable bit 0 in the Command
register. Although it doesn't work, who knows why.

> On that subject I think it needs to be a goal that LinuxBIOS do enough
> that a loaded OS shouldn't need to configure resources.  We shouldn't
> prevent the OS but allowing it is a good thing.

true. Although as Jeff points out the corner cases are a mess. I always
hoped that Linux would do the hard work. it is so close.

> Hmm. Could we have the prefix be CONFIG or LBCONFIG, when you
> get up t CONFIG_LINUXBIOS that is fairly tedious to type.

Sold!

How about just CONFIG

ron

Reply via email to