svcrdma: Pre-allocate svc_rdma_recv_ctxt objects

The original reason for allocating svc_rdma_recv_ctxt objects during
Receive completion was to ensure the objects were allocated on the
NUMA node closest to the underlying IB device.

Since commit c5d68d25bd6b ("svcrdma: Clean up allocation of
svc_rdma_recv_ctxt"), however, the device's favored node is
explicitly passed to the memory allocator.

To enable switching Receive completion to soft IRQ context, move
memory allocation out of completion handling, since it can be
costly, and it can sleep.

A limited number of objects is now allocated at "accept" time.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>

1 files changed, 21 insertions, 11 deletions

diff --git a/net/sunrpc/xprtrdma/svc_rdma_recvfrom.c b/net/sunrpc/xprtrdma/svc_rdma_recvfrom.c
index 3b05f90a3e50..c8c1c534070b 100644
--- a/net/sunrpc/xprtrdma/svc_rdma_recvfrom.c
+++ b/net/sunrpc/xprtrdma/svc_rdma_recvfrom.c
@@ -204,18 +204,11 @@ struct svc_rdma_recv_ctxt *svc_rdma_recv_ctxt_get(struct svcxprt_rdma *rdma)
 
 	node = llist_del_first(&rdma->sc_recv_ctxts);
 	if (!node)
-		goto out_empty;
-	ctxt = llist_entry(node, struct svc_rdma_recv_ctxt, rc_node);
+		return NULL;
 
-out:
+	ctxt = llist_entry(node, struct svc_rdma_recv_ctxt, rc_node);
 	ctxt->rc_page_count = 0;
 	return ctxt;
-
-out_empty:
-	ctxt = svc_rdma_recv_ctxt_alloc(rdma);
-	if (!ctxt)
-		return NULL;
-	goto out;
 }
 
 /**
@@ -277,7 +270,7 @@ static bool svc_rdma_refresh_recvs(struct svcxprt_rdma *rdma,
 		rdma->sc_pending_recvs++;
 	}
 	if (!recv_chain)
-		return false;
+		return true;
 
 	ret = ib_post_recv(rdma->sc_qp, recv_chain, &bad_wr);
 	if (ret)
@@ -301,10 +294,27 @@ err_free:
  * svc_rdma_post_recvs - Post initial set of Recv WRs
  * @rdma: fresh svcxprt_rdma
  *
- * Returns true if successful, otherwise false.
+ * Return values:
+ *   %true: Receive Queue initialization successful
+ *   %false: memory allocation or DMA error
  */
 bool svc_rdma_post_recvs(struct svcxprt_rdma *rdma)
 {
+	unsigned int total;
+
+	/* For each credit, allocate enough recv_ctxts for one
+	 * posted Receive and one RPC in process.
+	 */
+	total = (rdma->sc_max_requests * 2) + rdma->sc_recv_batch;
+	while (total--) {
+		struct svc_rdma_recv_ctxt *ctxt;
+
+		ctxt = svc_rdma_recv_ctxt_alloc(rdma);
+		if (!ctxt)
+			return false;
+		llist_add(&ctxt->rc_node, &rdma->sc_recv_ctxts);
+	}
+
 	return svc_rdma_refresh_recvs(rdma, rdma->sc_max_requests);
 }