RDS: IB: ack more receive completions to improve performance

For better performance, we split the receive completion IRQ handler. That
lets us acknowledge several WCE events in one call. We also limit the WC
to max 32 to avoid latency. Acknowledging several completions in one call
instead of several calls each time will provide better performance since
less mutual exclusion locks are being performed.

In next patch, send completion is also split which re-uses the poll_cq()
and hence the code is moved to ib_cm.c

Signed-off-by: Santosh Shilimkar <ssantosh@kernel.org>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>

1 files changed, 68 insertions, 2 deletions

diff --git a/net/rds/ib_cm.c b/net/rds/ib_cm.c
index 9043f5c04787..28e0979720b2 100644
--- a/net/rds/ib_cm.c
+++ b/net/rds/ib_cm.c
@@ -216,6 +216,72 @@ static void rds_ib_cq_event_handler(struct ib_event *event, void *data)
 		 event->event, ib_event_msg(event->event), data);
 }
 
+/* Plucking the oldest entry from the ring can be done concurrently with
+ * the thread refilling the ring.  Each ring operation is protected by
+ * spinlocks and the transient state of refilling doesn't change the
+ * recording of which entry is oldest.
+ *
+ * This relies on IB only calling one cq comp_handler for each cq so that
+ * there will only be one caller of rds_recv_incoming() per RDS connection.
+ */
+static void rds_ib_cq_comp_handler_recv(struct ib_cq *cq, void *context)
+{
+	struct rds_connection *conn = context;
+	struct rds_ib_connection *ic = conn->c_transport_data;
+
+	rdsdebug("conn %p cq %p\n", conn, cq);
+
+	rds_ib_stats_inc(s_ib_evt_handler_call);
+
+	tasklet_schedule(&ic->i_recv_tasklet);
+}
+
+static void poll_cq(struct rds_ib_connection *ic, struct ib_cq *cq,
+		    struct ib_wc *wcs,
+		    struct rds_ib_ack_state *ack_state)
+{
+	int nr;
+	int i;
+	struct ib_wc *wc;
+
+	while ((nr = ib_poll_cq(cq, RDS_IB_WC_MAX, wcs)) > 0) {
+		for (i = 0; i < nr; i++) {
+			wc = wcs + i;
+			rdsdebug("wc wr_id 0x%llx status %u byte_len %u imm_data %u\n",
+				 (unsigned long long)wc->wr_id, wc->status,
+				 wc->byte_len, be32_to_cpu(wc->ex.imm_data));
+			rds_ib_recv_cqe_handler(ic, wc, ack_state);
+		}
+	}
+}
+
+static void rds_ib_tasklet_fn_recv(unsigned long data)
+{
+	struct rds_ib_connection *ic = (struct rds_ib_connection *)data;
+	struct rds_connection *conn = ic->conn;
+	struct rds_ib_device *rds_ibdev = ic->rds_ibdev;
+	struct rds_ib_ack_state state;
+
+	BUG_ON(!rds_ibdev);
+
+	rds_ib_stats_inc(s_ib_tasklet_call);
+
+	memset(&state, 0, sizeof(state));
+	poll_cq(ic, ic->i_recv_cq, ic->i_recv_wc, &state);
+	ib_req_notify_cq(ic->i_recv_cq, IB_CQ_SOLICITED);
+	poll_cq(ic, ic->i_recv_cq, ic->i_recv_wc, &state);
+
+	if (state.ack_next_valid)
+		rds_ib_set_ack(ic, state.ack_next, state.ack_required);
+	if (state.ack_recv_valid && state.ack_recv > ic->i_ack_recv) {
+		rds_send_drop_acked(conn, state.ack_recv, NULL);
+		ic->i_ack_recv = state.ack_recv;
+	}
+
+	if (rds_conn_up(conn))
+		rds_ib_attempt_ack(ic);
+}
+
 static void rds_ib_qp_event_handler(struct ib_event *event, void *data)
 {
 	struct rds_connection *conn = data;
@@ -282,7 +348,7 @@ static int rds_ib_setup_qp(struct rds_connection *conn)
 	}
 
 	cq_attr.cqe = ic->i_recv_ring.w_nr;
-	ic->i_recv_cq = ib_create_cq(dev, rds_ib_recv_cq_comp_handler,
+	ic->i_recv_cq = ib_create_cq(dev, rds_ib_cq_comp_handler_recv,
 				     rds_ib_cq_event_handler, conn,
 				     &cq_attr);
 	if (IS_ERR(ic->i_recv_cq)) {
@@ -743,7 +809,7 @@ int rds_ib_conn_alloc(struct rds_connection *conn, gfp_t gfp)
 	}
 
 	INIT_LIST_HEAD(&ic->ib_node);
-	tasklet_init(&ic->i_recv_tasklet, rds_ib_recv_tasklet_fn,
+	tasklet_init(&ic->i_recv_tasklet, rds_ib_tasklet_fn_recv,
 		     (unsigned long) ic);
 	mutex_init(&ic->i_recv_mutex);
 #ifndef KERNEL_HAS_ATOMIC64