Re: [Xen-devel] [PATCH] [RFC] MSI and interrupt mapping

[Espen Skoglund <espen.skoglund@xxxxxxxxxxxxx>]

[Keir Fraser]
> On 04/02/2009 17:47, "Espen Skoglund" <espen.skoglund@xxxxxxxxxxxxx> wrote:
>> More generally speaking, the current state of affairs with regards
>> to interrupt management in Xen can be a bit confusing from the
>> source code point of view.  Essentially, all irqs except some
>> legacy interrupts are assumed to be IOAPIC irqs.  This assumption
>> breaks down with the introduction of MSIs and per-domain pirq
>> tables.  It might therefore be worthwile to rename a few things and
>> slightly restructure the Xen code to make the distinction between
>> IOAPIC irqs and other interrupts clearer.  I could volunteer to
>> have a stab at such a cleanup if this is something people want.
>> 
>> Comments?

> Sketch out what you mean in a bit more detail and I'll think about
> it. The stuff in arch/x86/irq.c at least makes reasonable sense to
> me right now (of course ;-).

Get rid of the global irq_vector[] and vector_irq[] arrays.  Where
needed (e.g., in the IOAPIC code) use the dom0 pirq arrays instead.
>From a quick glance, the stuff in arch/x86/irq.c makes sense as it is;
the exception perhaps being the dump_irqs() function which does not
take MSIs into account.

        eSk



_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel