RDMA/rtrs-clt: Write path fast memory registration

With fast memory registration in write path, we can reduce
the memory consumption by using less max_send_sge, support IO bigger
than 116 KB (29 segments * 4 KB) without splitting, and it also
make the IO path more symmetric.

To avoid some times MR reg failed, waiting for the invalidation to finish
before the new mr reg. Introduce a refcount, only finish the request
when both local invalidation and io reply are there.

Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Jack Wang <[email protected]>
Signed-off-by: Md Haris Iqbal <[email protected]>
Signed-off-by: Dima Stepanov <[email protected]>
Signed-off-by: Jason Gunthorpe <[email protected]>

0 files changed, 0 insertions, 0 deletions