Skip to content
Commit 6b1184cd authored by Chuck Lever's avatar Chuck Lever Committed by Anna Schumaker
Browse files

xprtrdma: Allocate RPC/RDMA receive buffer separately from struct rpcrdma_rep



The rr_base field is currently the buffer where RPC replies land.

An RPC/RDMA reply header lands in this buffer. In some cases an RPC
reply header also lands in this buffer, just after the RPC/RDMA
header.

The inline threshold is an agreed-on size limit for RDMA SEND
operations that pass from server and client. The sum of the
RPC/RDMA reply header size and the RPC reply header size must be
less than this threshold.

The largest RDMA RECV that the client should have to handle is the
size of the inline threshold. The receive buffer should thus be the
size of the inline threshold, and not related to RPCRDMA_MAX_SEGS.

RPC replies received via RDMA WRITE (long replies) are caught in
rq_rcv_buf, which is the second half of the RPC send buffer. Ie,
such replies are not involved in any way with rr_base.

Signed-off-by: default avatarChuck Lever <chuck.lever@oracle.com>
Signed-off-by: default avatarAnna Schumaker <Anna.Schumaker@Netapp.com>
parent 85275c87
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment