On 08/01/18 11:09 AM, Jason Gunthorpe wrote:

It could, if we had a DMA op for p2p then the drivers that provide
their own ops can implement it appropriately or not at all.

I was thinking of doing something like this. I'll probably rough out a 
patch and send it along today or tomorrow.

quoted

If at all it should be in the dma_map* wrappers, but for that we'd need
a good identifier.  And it still would not solve the whole fake dma
ops issue.

Very long term the IOMMUs under the ops will need to care about this,
so the wrapper is not an optimal place to put it - but I wouldn't
object if it gets it out of RDMA :)

Well, creating the extra op doesn't really change anything to the RDMA 
patch in this series. The point is, for the time being, we need to track 
whether we are doing a P2P or normal mapping using a side channel as we 
don't have a good way of tracking it in the SGL at this time.

Adding an extra DMA op is just one way to allow the existing DMA 
providers to opt-in/out.

Logan