Re: [PATCH] arm64: Add architecture support for PCI

From: Catalin Marinas
Date: Tue Feb 04 2014 - 11:42:03 EST

On Tue, Feb 04, 2014 at 08:44:36AM +0000, Arnd Bergmann wrote:
> On Monday 03 February 2014 21:36:58 Liviu Dudau wrote:
> > On Mon, Feb 03, 2014 at 08:05:56PM +0000, Arnd Bergmann wrote:
> > > 2 MB was a compromise on arm32 to allow up to 32 PCI host bridges but not
> > > take up too much virtual space. On arm64 it should be at least as big.
> > > Could be more than that, although I don't see a reason why it should be,
> > > unless we expect to see systems with tons of host bridges, or buses
> > > that exceed 64KB of I/O space.
> >
> > I will increase the size to 2MB for v2.
> Thinking about this some more, I'd go a little higher. In case of
> pci_mv, we already register a 1MB region for one logical host
> (which has multiple I/O spaces behind an emulated bridge), so
> going to 16MB or more would let us handle multiple 1MB windows
> for some amount of future proofing.
> Maybe Catalin can assign us some virtual address space to use,
> with the constraints that it should be 16MB or more in an
> area that doesn't require additional kernel page table pages.

See below, 16MB and could be extended further if needed.