[PATCH] powerpc/pci: Support per-aperture memory offset

[Sethi Varun-B16395]

> -----Original Message-----
> From: Benjamin Herrenschmidt [mailto:benh at kernel.crashing.org]
> Sent: Monday, May 06, 2013 12:20 PM
> To: linuxppc-dev
> Cc: Kumar Gala; Wood Scott-B07421; Sethi Varun-B16395; Thomas Petazzoni;
> Andrew Murray; Bjorn Helgaas; linux-pci at vger.kernel.org
> Subject: [PATCH] powerpc/pci: Support per-aperture memory offset
> 
> The PCI core supports an offset per aperture nowadays but our arch code
> still has a single offset per host bridge representing the difference
> betwen CPU memory addresses and PCI MMIO addresses.
> 
> This is a problem as new machines and hypervisor versions are coming out
> where the 64-bit windows will have a different offset (basically mapped
> 1:1) from the 32-bit windows.
> 
> This fixes it by using separate offsets. In the long run, we probably
> want to get rid of that intermediary struct pci_controller and have those
> directly stored into the pci_host_bridge as they are parsed but this will
> be a more invasive change.
> 
> Signed-off-by: Benjamin Herrenschmidt <benh at kernel.crashing.org>
> ---
> 
> Now, this is a big one but I'd like to still merge it for 3.10 because
> we're having new machine coming up (and new versions of pHyp on existing
> machines) that are going to expose MMIO windows with different offsets
> (basically our 64-bit windows are 1:1 and our 32-bit windows remapped).
> 
> I'm not expecting any major issue with the patch, I've tested it on some
> of our machines here and will test it more during the next couple of days
> the notable thing is the removal of a "workaround" / fallback on 32-bit
> that I suspect only ever mattered for machines long unsupported (PReP ?)
> as I don't think we have anything that doesn't populate the bridge
> resources and expects to work nowadays (other stuff would break anyway).
> 
> This is also why I'm NAK'ing the patch making
> pci_process_bridge_OF_ranges() generic since I need to change it for
> powerpc and this isn't the right long term approach (we should "merge"
> pci_controller & pci_host_bridge instead and directly populate the
> pci_host_bridge apertures).
> 
> If I see no major issue with the patch during the next few days, I'll
> send it to Linus with my next pull request, probably at -rc1.
> 
> Kumar, Scott, Sethi, please test on FSL HW. I'll take care of macs and
> 4xx in addition to the various IBM ppc64 platforms.
> 
[Sethi Varun-B16395] Tested patch on FSL T4240, P4080, P5040 and P1020 boards.

-Varun