> I don't have a philosophical position on exactly how rdma_create_qp() > should work going forward, but I agree that ULPs have depended on the > current behavior and will need to be updated if the QPs returned by > rdma_create_qp() are not in INIT state. I stand ready to fix things up in the > RPC/RDMA consumers should that be needed. > > In fact it looks like some consumers might already assume the corrected CMA > behavior. Maybe the RPC/RDMA consumers can safely be modified now? > Let me know how to proceed if this is the case. > I haven't done the same ULP audit as Chuck, but I do know that the cma.c patch caused failures on both iSer and NFS/RDMA. The revert fixed ALL the issues. Let me know how to help test any successor patch. Mike
