Re: [PATCH] intel-iommu: Obey coherent_dma_mask for alloc_coherent on passthrough

[Alex Williamson <alex.williamson@hp.com>]

On Fri, 2009-11-06 at 12:34 +0900, FUJITA Tomonori wrote:
> On Thu, 05 Nov 2009 20:19:52 -0700
> Alex Williamson <alex.williamson@hp.com> wrote:
> > On Fri, 2009-11-06 at 11:41 +0900, FUJITA Tomonori wrote:
> >
> > > This is fine for 2.6.32 but we'll cleanly fix this by using
> > > swiotlb_dma_ops later, right?
> > 
> > I'm open to suggestions.  I don't really understand why we dropped
> > swiotlb for passthrough mode in 2.6.32 to start with.  It seems like we
> > now have a couple corner cases where we have to either hope for the best
> > or effectively ignore the request to use passthrough.  Thanks,
> 
> I think that the cleanest solution is setting up swiotlb_dma_ops for
> passthrough devices (and devices not behind pci, etc). Calgary IOMMU
> does the same for years.

intel-iommu was using swiotlb for the global dma_ops when in passthrough
mode until 19943b0e (2.6.31 and earlier).  I would say the next step
would be to use per device dma_ops so we can point passthrough devices
to swiotlb, as you suggest, but that seems to work against part of what
19943b0e was trying to accomplish.

Alex