Re: pvops-2.6.32 - Interrupt routing problem

[Bastian Blank <waldi@debian.org>]

On Fri, Mar 19, 2010 at 10:15:23PM +0800, Zhang, Xiantao wrote:
> Bastian Blank wrote:
> > I did some digging and found similar code that disallows
> > reprogramming of IO-APIC pins in
> > arch/x86/kernel/apic/io_apic.c:io_apic_set_pci_routing
> > So this behaviour is correct.
> > So the question is: Is it wrong for the kernel to do register_gsi
> > calls early in bootup for interrupts it does not know anything about?
> 
> I remembered you mentioned that you hacked hypervisor and allows
> re-programme for the special gsi and everying goes well. If you only
> hacked hypervisor in your case, and I think the hypervisor patch
> should be enough.

Only if we can distinguish between the two registrations, IMHO we can't.
So the kernel must not do them in the first case or we need to add
something to distiguish. IMHO the first solution is easier and should
work.

> Could you attach the full log ?

No, I have none. But there is nothing new in it over the last one
anyway.

Bastian

-- 
No one wants war.
		-- Kirk, "Errand of Mercy", stardate 3201.7