[openib-general] uverbs: reg_mr

[Michael S. Tsirkin]

Hi, Roland!
I'd like to suggest extending ibv_reg_mr to pass the
virtual address for RDMA, in addition to the CPU virtual address.

Currently it seems they are assumed to be the same, but there
does not appear to be an actual reason to enforce this limitation.

For example, I am now writing a simple test application for rdma,
and it would be convenient to simply rdma to virtual address 0,
and have the HCA land it at the proper address.
As it is, I have to pass the virtual address along with the RKey.

-- 
MST - Michael S. Tsirkin