Re: MPC7448_HPC2 question

[Zang Roy-r61911 <tie-fei.zang@freescale.com>]

On Wed, 2006-10-25 at 13:20, Benjamin Herrenschmidt wrote:
> Any reason why that:
> 
> void mpc7448_hpc2_fixup_irq(struct pci_dev *dev)
> {
>         struct pci_controller *hose;
>         struct device_node *node;
>         const unsigned int *interrupt;
>         int busnr;
>         int len;
>         u8 slot;
>         u8 pin;
> 
>         /* Lookup the hose */
>         busnr = dev->bus->number;
>         hose = pci_bus_to_hose(busnr);
>         if (!hose)
>                 printk(KERN_ERR "No pci hose found\n");
> 
>         /* Check it has an OF node associated */
>         node = (struct device_node *) hose->arch_data;
>         if (!node)
>                 printk(KERN_ERR "No pci node found\n");
> 
>         interrupt = get_property(node, "interrupt-map", &len);
>         slot = find_slot_by_devfn(interrupt, dev->devfn);
>         pci_read_config_byte(dev, PCI_INTERRUPT_PIN, &pin);
>         if (pin == 0 || pin > 4)
>                 pin = 1;
>         pin--;
>         dev->irq  = interrupt[slot*4*7 + pin*7 + 5];
>         DBG("TSI_PCI: dev->irq = 0x%x\n", dev->irq);
> }
> 
> Cannot be replaced by:
> 
>         pci_read_irq_line(dev);
> 
> I'm about to make pci_read_irq_line() called by default by the PCI
> code (with a new ppc_md.pci_irq_fixup for platforms that really need
> something else done) and so while "fixing" all platforms, I stumbled
> accross the code above which is a bit shocking :)
> 
> Cheers,
> 
> Ben.
> 
> 
> 

Ben, 
	Last week, I had tried to do the replacement. While, it did not work. I
had though some update of pci_read_irq_line() should be needed.
	Any way, I will investigate it and give the patch ASAP :-).

Roy