On 05/19/2017 15:56, Bjorn Helgaas wrote:
> pci_add_resource_offset() is for host bridge windows where the bridge
> translates CPU addresses to PCI bus addresses by adding an offset. To my
> knowledge, no host bridge translates bus numbers, so this is only useful
> for MEM and IO windows. In any event, host->busn_offset is never set to
> anything other than zero, so pci_add_resource() is sufficient.
>
> a2e50f53d535 ("MIPS: PCI: Add a hook for IORESOURCE_BUS in
> pci_controller/bridge_controller") also added busn_resource itself. This
> is currently unused but may be used by future SGI IP27 fixes, so I left it
> there.
>
> Signed-off-by: Bjorn Helgaas <bhelgaas@xxxxxxxxxx>
> CC: Joshua Kinard <kumba@xxxxxxxxxx>
> ---
> arch/mips/include/asm/pci.h | 1 -
> arch/mips/pci/pci-legacy.c | 3 +--
> 2 files changed, 1 insertion(+), 3 deletions(-)
>
> diff --git a/arch/mips/include/asm/pci.h b/arch/mips/include/asm/pci.h
> index 1000c1b4c875..52f551ee492d 100644
> --- a/arch/mips/include/asm/pci.h
> +++ b/arch/mips/include/asm/pci.h
> @@ -39,7 +39,6 @@ struct pci_controller {
> unsigned long io_offset;
> unsigned long io_map_base;
> struct resource *busn_resource;
> - unsigned long busn_offset;
>
> #ifndef CONFIG_PCI_DOMAINS_GENERIC
> unsigned int index;
> diff --git a/arch/mips/pci/pci-legacy.c b/arch/mips/pci/pci-legacy.c
> index 3a84f6c0c840..174575a9a112 100644
> --- a/arch/mips/pci/pci-legacy.c
> +++ b/arch/mips/pci/pci-legacy.c
> @@ -86,8 +86,7 @@ static void pcibios_scanbus(struct pci_controller *hose)
> hose->mem_resource, hose->mem_offset);
> pci_add_resource_offset(&resources,
> hose->io_resource, hose->io_offset);
> - pci_add_resource_offset(&resources,
> - hose->busn_resource, hose->busn_offset);
> + pci_add_resource(&resources, hose->busn_resource);
> bus = pci_scan_root_bus(NULL, next_busno, hose->pci_ops, hose,
> &resources);
> hose->bus = bus;
>
>
I tested this change out on my Octane (SGI-IP30) and Origin 200 (SGI-IP27), and
it appears to work fine on both systems (for now).
The Octane has two PCI host bridges in it right now (BaseIO & PCI Card cage),
while the Origin 200 only has one. Haven't tested it on my bigger Onyx2 w/
multiple host bridges, but that machine has other problems, so I am ignoring it
for now.
I wasn't aware that pci_add_resource existed back when I wrote the patch. The
symptom that was addressed was some conditional in the PCI core -- I forget
where -- wasn't enumerating multiple PCI busses without having an
IORESOURCE_BUS struct passed to it. I think it found the first bus and then
moved on to other subsystems, which usually led to a non-booting machine.
Acked-by: Joshua Kinard <kumba@xxxxxxxxxx>
