xprtrdma: Use workqueue to process RPC/RDMA replies

The reply tasklet is fast, but it's single threaded. After reply
traffic saturates a single CPU, there's no more reply processing
capacity.

Replace the tasklet with a workqueue to spread reply handling across
all CPUs.  This also moves RPC/RDMA reply handling out of the soft
IRQ context and into a context that allows sleeps.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Reviewed-by: Sagi Grimberg <sagig@mellanox.com>
Tested-By: Devesh Sharma <devesh.sharma@avagotech.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
diff --git a/net/sunrpc/xprtrdma/xprt_rdma.h b/net/sunrpc/xprtrdma/xprt_rdma.h
index e6a358f..6ea1dbe 100644
--- a/net/sunrpc/xprtrdma/xprt_rdma.h
+++ b/net/sunrpc/xprtrdma/xprt_rdma.h
@@ -164,6 +164,7 @@
 	unsigned int		rr_len;
 	struct ib_device	*rr_device;
 	struct rpcrdma_xprt	*rr_rxprt;
+	struct work_struct	rr_work;
 	struct list_head	rr_list;
 	struct rpcrdma_regbuf	*rr_rdmabuf;
 };
@@ -430,6 +431,9 @@
 int frwr_alloc_recovery_wq(void);
 void frwr_destroy_recovery_wq(void);
 
+int rpcrdma_alloc_wq(void);
+void rpcrdma_destroy_wq(void);
+
 /*
  * Wrappers for chunk registration, shared by read/write chunk code.
  */