Re: [PATCH V2 00/23] MMCONFIG refactoring and support for ARM64 PCI hostbridge init based on ACPI

[Lorenzo Pieralisi]

On Fri, Dec 18, 2015 at 06:56:39PM +0000, okaya@xxxxxxxxxxxxxx wrote:

[...]

> Here is what I have as an IO resource.
> 
> 		QWORDIO(			//Consumed-And-produced resource
> 			ResourceProducer,	// bit 0 of general flags is 0
> 			MinFixed,		// Range is fixed
> 			MaxFixed,		// Range is fixed
> 			PosDecode,
> 			EntireRange,
> 			0x0000,			// Granularity
> 			0x1000,		// Min, 0 is not accepted
> 			0x10FFF,		// Max
> 			0x8FFFFFEF000,	// Translation
> 			0x10000,		// Range Length
> 			,, PI00
> 		 )
> 
> I don't have any type specified.
> 
> I agree with Lorenzo's assessment. The min and max values represent the
> PCI IO bus addresses. The translation offset is added to these values to
> figure out the CPU view of the PCI IO range.
> 
> The endpoints BAR addresses are programmed with IO addresses ranging
> between 0x1000 and 0x10FFF for this example above.
> 
> Here is another question. Chris Covington and I asked this question on a
> private email to you but we didn't hear back.
> 
> We were referring to a Linaro IO hack patch as we were not sure whether
> this was a limitation of the hack or a general expectation for ARM64 PCI
> in general.
> 
> I'll repeat it here.
> 
> I have multiple root ports with the same IO port configuration in the
> current ACPI table.
> 
> Root port 0 = IO range 0x1000-0x10FFF
> Root port 1 = IO range 0x1000-0x10FFF
> Root port 2 = IO range 0x1000-0x10FFF

It is fine. You end up mapping for each of those a 4k window of the
virtual address space allocated to IO and that's what you will have in
the kernel PCI resources (not in the HW BARs though). If that was a problem
it would be even for the current DT host controllers eg:

arch/arm64/boot/dts/apm/apm-storm.dtsi

it should not be (again I will let Arnd comment on this since he may be
aware of issues encountered on other arches/platforms).

Lorenzo

> 
> Each root port can have the same IO address range configuration, are we
> expecting IO port numbers to be unique across the whole system for ARM64?
> 
> Something like
> 
> Root port 0 = IO range 0x1000-0x10FFF
> Root port 1 = IO range 0x11000-0x20FFF
> Root port 2 = IO range 0x21000-0x30FFF
> 
> since the IO addresses are being remapped into PCI IO range printed during
> boot.
> 
> PCI I/O : 0xffff7ffffae00000 - 0xffff7ffffbe00000   (    16 MB)
> 
> and each root port would remap to 64k of the 16MB range.
> 
> >
> > Any comments are very appreciated!
> >
> > Tomasz
> > --
> > To unsubscribe from this list: send the line "unsubscribe linux-pci" in
> > the body of a message to majordomo@xxxxxxxxxxxxxxx
> > More majordomo info at  http://vger.kernel.org/majordomo-info.html
> >
> 
> 
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/