[PATCH v4 4/7] Add support for 750CL Holly board

Josh Boyer jwboyer at linux.vnet.ibm.com
Wed May 9 03:01:26 EST 2007


On Tue, 2007-05-08 at 11:45 -0500, Nathan Lynch wrote:
> Hi Josh-
> 
> I see this has been picked up by Paul already, but there are a couple
> minor things I noticed as I was reading the patch.

That's fine.  Comments are still welcome since I do plan on adding a bit
more a bit down the road.  I'll make sure to note them and fix them up.

> > +int holly_exclude_device(u_char bus, u_char devfn)
> > +{
> > +	if (bus == 0 && PCI_SLOT(devfn) == 0)
> > +		return PCIBIOS_DEVICE_NOT_FOUND;
> > +	else
> > +		return PCIBIOS_SUCCESSFUL;
> > +}
> 
> This function could be static, I think?  Looks like it's assigned to
> ppc_md.pci_exclude_device and called through that.

Yes, could be I think.

> > +static void __init holly_init_IRQ(void)
> > +{
> > +	struct mpic *mpic;
> > +	phys_addr_t mpic_paddr = 0;
> > +	struct device_node *tsi_pic;
> > +#ifdef CONFIG_PCI
> > +	unsigned int cascade_pci_irq;
> > +	struct device_node *tsi_pci;
> > +	struct device_node *cascade_node = NULL;
> > +#endif
> > +
> > +	tsi_pic = of_find_node_by_type(NULL, "open-pic");
> > +	if (tsi_pic) {
> > +		unsigned int size;
> > +		const void *prop = of_get_property(tsi_pic, "reg", &size);
> > +		mpic_paddr = of_translate_address(tsi_pic, prop);
> > +	}
> 
> You don't actually need 'size' there; it's legal to pass NULL to
> of_get_property.

Yep.

josh




More information about the Linuxppc-dev mailing list