IRQ-Tables once again

Eric W. Biederman ebiederman at lnxi.com
Thu Jan 22 15:14:00 CET 2004


ron minnich <rminnich at lanl.gov> writes:

> On 22 Jan 2004, Eric W. Biederman wrote:
> 
> > > no. In 2.4, in the right circumstances, the kernel can assign interrupts 
> > > just fine. 
> > 
> > However the statement is still true that it is desirable that we
> > assign irqs to support non linux operating systems.
> > 
> 
> yes, that is why we've started plugging support in: we needed it for EPIA
> for Plan 9 ...

And to be very clear the reason it is needed for the EPIA
is that VIA is slightly out of spec.  VIA chipsets instead
of just providing a scratch irq register, actually interpret
the value of that register for irq routing.

So while a general purpose solution should work in the EPIA case
the via chipsets may still require special casing.  Although
hopefully we can abstract an irq router well enough we won't
have to worry about it.

Eric



More information about the coreboot mailing list