svcrdma: Eliminate RPCRDMA_SQ_DEPTH_MULT
authorChuck Lever <chuck.lever@oracle.com>
Sun, 9 Apr 2017 17:05:52 +0000 (13:05 -0400)
committerJ. Bruce Fields <bfields@redhat.com>
Tue, 25 Apr 2017 21:25:54 +0000 (17:25 -0400)
commitb623589dbacbc786c2fffc85113a1dc1a331e2ca
tree6cc0b33b761469c882e208438b2688d32926b84c
parent6e6092ca305ad785c605d7e313727aad96c228a5
svcrdma: Eliminate RPCRDMA_SQ_DEPTH_MULT

The Send Queue depth is temporarily reduced to 1 SQE per credit. The
new rdma_rw API does an internal computation, during QP creation, to
increase the depth of the Send Queue to handle RDMA Read and Write
operations.

This change has to come before the NFSD code paths are updated to
use the rdma_rw API. Without this patch, rdma_rw_init_qp() increases
the size of the SQ too much, resulting in memory allocation failures
during QP creation.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
include/linux/sunrpc/svc_rdma.h
net/sunrpc/xprtrdma/svc_rdma.c
net/sunrpc/xprtrdma/svc_rdma_transport.c