ARM: 2.6.3[45] PCI regression (IXP4xx and PXA?)

[FUJITA Tomonori]

On Thu, 19 Aug 2010 18:53:38 +0200
Krzysztof Halasa <khc at pm.waw.pl> wrote:

> FUJITA Tomonori <fujita.tomonori at lab.ntt.co.jp> writes:
> 
> >> I'd rather have the arch (aka the bus) be able to filter the mask,
> >> better than having to deal with multiple masks in the generic code.
> >> Besides, in embedded-land, you never know how many busses are stacked
> >> before you reach the device, ie, you'd end up having to AND quite a few
> >> masks before getting there in some cases.
> >
> > You mean that you like to permit architectures to modify
> > dev->coherent_dma_mask behind a device? If so, I'm against it because
> > it means dev->coherent_dma_mask has two meanings. That's confusing.
> 
> Well, I think it may be the only really correct solution, and in fact
> it's arch-independent.
> 
> The coherent_dma_mask would mean one thing: address space shared between
> the CPU(s) and the device.

linux/device.h defines that it's the device dma mask.

Except for ARM, coherent_dma_mask represents the device dma mask.

We sometimes want to know the device dma mask. Moving to your
definition means that we lose that information.


> This usually equals device's address space - only because CPU and
> bridges next to it have wide (logical) address busses. It's not always
> the case, though, and may be not the case on any arch.
> 
> We should make sure we got it right (including drivers), since any
> reduction of the dma*mask would be irreversible (new masks would be
> ANDed with the existing masks).
> 
> > I think that having the generic place for bus'
> > dma mask would be better rather than architecture specific
> > places.
> 
> Definitely, if possible. BTW the dmabounce (and equivalent code on other
> archs, including probably swiotlb on x86-64) could probably be merged as
> well. I don't know the internals very well, though. At least it may be
> worth it looking at them.

btw, swiotlb is used by x86_64, ia64, and powerpc.

I'm sure that I can convert ARM to use it instead of dmabounce. But
well, even a bug fix patch for dmabounce haven't been merged so I'm
not sure that ARM people would accept such change.

http://marc.info/?l=linux-arm-kernel&m=128047064008554&w=2


> > Adding a new API to set bus' dma mask would make sense too.
> 
> Not sure. Which bus? There could be many :-)
> In practice - 64-bit PCIe -> 32-bit PCI -> 24-bit ISA - etc.
> Or, like with IXP/PXA - 26-bit PCI -> 32-bit device.

Seems that we are not on the same page. As I said before, have you
seen how POWERPC uses max_direct_dma_addr in dev_archdata struct? I
was talking about moving it to the generic place and the API to set
it.