On 10/01/18 15:35, Christoph Hellwig wrote:

On Wed, Jan 10, 2018 at 12:16:15PM +0000, Robin Murphy wrote:

quoted

On 10/01/18 08:09, Christoph Hellwig wrote:

quoted

To properly reject too small DMA masks based on the addressability of the
bounce buffer.

I reckon this is self-evident enough that it should simply be squashed into
the previous patch.

x86 didn't wire it up before, so I want a clear blaimpoint for this
change instead of mixing it up.

That almost makes sense, if x86 were using this generic swiotlb_dma_ops 
already. AFAICS it's only ia64, unicore and tile who end up using it, 
and they all had swiotlb_dma_supported hooked up to begin with. Am I 
missing something?

If regressions are going to happen, they'll surely point at whichever 
commit pulls the ops into the relevant arch code - there doesn't seem to 
be a great deal of value in having a piecemeal history of said ops 
*before* that point.

Robin.