Re: [PATCH 6/9] MPIC MSI allocator

[Benjamin Herrenschmidt <benh@kernel.crashing.org>]

On Wed, 2006-12-13 at 12:23 -0600, Olof Johansson wrote:
> On Wed, 13 Dec 2006 21:40:03 +1100 Michael Ellerman <michael@ellerman.id.au> wrote:
> 
> > To support MSI on MPIC we need a way to reserve and allocate hardware irq
> > numbers, this patch implements an allocator for that. It looks like we'll
> > end up with several backends based on the MPIC, so the allocator is attached
> > to the struct mpic, not the msi backend.
> > 
> > Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
> 
> > +static void mpic_msi_auto_reserve_hwirqs(struct mpic *mpic)
> > +{
> > +	irq_hw_number_t hwirq;
> > +	struct irq_host_ops *ops = mpic->irqhost->ops;
> > +	struct device_node *np;
> > +	int flags, index, i;
> > +	struct of_irq oirq;
> > +
> > +	/* Reserve source numbers we know are reserved in the HW */
> 
> How do we know? Reserved on what HW? Sure looks system/platform
> dependent to me.
> 
> > +	for (i = 0;   i < 8;   i++)
> > +		__mpic_msi_reserve_hwirq(mpic, i);
> > +	for (i = 42;  i < 26;  i++)
>              ^^^^^^^^^^^^^^^
> 
> Is this some sort of check to see if we're awake? :)

This code in MPIC currently is only ever used on U3/U4 and so that code
reserves IRQs we know are used for U3/U4 internal IRQs and not available
for HT interrupts (or MSIs). The code, then, does a second pass also
exclusing all IRQs for all devices in the device-tree.

This is a "hack" that works until device-trees provide the right
property we have defined for mpic to indicate what range of irqs is to
be used for MSIs.

Ben.