[PATCH v6 1/4] RDMA/umem: Support importing dma-buf as user memory region

[Jason Gunthorpe]

On Tue, Nov 03, 2020 at 09:43:17PM +0100, Daniel Vetter wrote:

> > > Yeah definitely don't call dma_buf_map_attachment and expect a page back. In practice you'll get a page back fairly often, but I don't think
> > > we want to bake that in, maybe we eventually get to non-hacky dma_addr_t only sgl.
> > >
> > > What I'm wondering is whether dma_buf_attach shouldn't reject such devices directly, instead of each importer having to do that.
> >
> > Come back here to see if consensus can be reached on who should do the check. My
> > thinking is that it could be over restrictive for dma_buf_attach to always reject
> > dma_virt_ops. According to dma-buf documentation the back storage would be
> > moved to system area upon mapping unless p2p is requested and can be supported
> > by the exporter. The sg_list for system memory would have struct page present.
> 
> So I'm not clear on what this dma_virt_ops stuff is for, but if it's
> an entirely virtual device with cpu access, then you shouldn't do
> dma_buf_map_attachment, and then peek at the struct page in the sgl.

Yes, so I think the answer is it is rdma device driver error to call these
new APIs and touch the struct page side of the SGL.

After Christophs series removign dma_virt_ops we could make that more
explicit, like was done for the pci p2p case.


> As a third option, if it's something about the connectivity between
> the importing and exporting device, then this should be checked in the
> ->attach callback the exporter can provide, like the p2p check. The

Drivers doing p2p are supposed to be calling the p2p distance stuff
and p2p dma map stuff which already has these checks.

Doing p2p and skipping all that in the dma buf side we already knew
was a hacky thing.

Jason