[1/2] net/sfc: support ops to check ready descriptors in ESSB Rx

Message ID 1530286891-8833-1-git-send-email-arybchenko@solarflare.com (mailing list archive)
State Accepted, archived
Delegated to: Ferruh Yigit
Headers
Series [1/2] net/sfc: support ops to check ready descriptors in ESSB Rx |

Checks

Context Check Description
ci/checkpatch success coding style OK
ci/Intel-compilation success Compilation OK

Commit Message

Andrew Rybchenko June 29, 2018, 3:41 p.m. UTC
  Implement EF10 ESSB Rx datapath callback to get number of pending
descriptors.

Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Ivan Malov <ivan.malov@oktetlabs.ru>
---
 drivers/net/sfc/sfc_ef10_essb_rx.c | 27 +++++++++++++++++++++------
 1 file changed, 21 insertions(+), 6 deletions(-)
  

Comments

Ferruh Yigit July 4, 2018, 6:30 p.m. UTC | #1
On 6/29/2018 4:41 PM, Andrew Rybchenko wrote:
> Implement EF10 ESSB Rx datapath callback to get number of pending
> descriptors.
> 
> Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
> Reviewed-by: Ivan Malov <ivan.malov@oktetlabs.ru>

Series applied to dpdk-next-net/master, thanks.
  

Patch

diff --git a/drivers/net/sfc/sfc_ef10_essb_rx.c b/drivers/net/sfc/sfc_ef10_essb_rx.c
index dcb5cb24c..257e40db1 100644
--- a/drivers/net/sfc/sfc_ef10_essb_rx.c
+++ b/drivers/net/sfc/sfc_ef10_essb_rx.c
@@ -404,13 +404,28 @@  sfc_ef10_essb_recv_pkts(void *rx_queue, struct rte_mbuf **rx_pkts,
 
 static sfc_dp_rx_qdesc_npending_t sfc_ef10_essb_rx_qdesc_npending;
 static unsigned int
-sfc_ef10_essb_rx_qdesc_npending(__rte_unused struct sfc_dp_rxq *dp_rxq)
+sfc_ef10_essb_rx_qdesc_npending(struct sfc_dp_rxq *dp_rxq)
 {
-	/*
-	 * Correct implementation requires EvQ polling and events
-	 * processing.
-	 */
-	return -ENOTSUP;
+	struct sfc_ef10_essb_rxq *rxq = sfc_ef10_essb_rxq_by_dp_rxq(dp_rxq);
+	const unsigned int evq_old_read_ptr = rxq->evq_read_ptr;
+	efx_qword_t rx_ev;
+
+	if (unlikely(rxq->flags & (SFC_EF10_ESSB_RXQ_NOT_RUNNING |
+				   SFC_EF10_ESSB_RXQ_EXCEPTION)))
+		return rxq->bufs_pending;
+
+	while (sfc_ef10_essb_rx_event_get(rxq, &rx_ev)) {
+		/*
+		 * DROP_EVENT is an internal to the NIC, software should
+		 * never see it and, therefore, may ignore it.
+		 */
+		sfc_ef10_essb_rx_process_ev(rxq, rx_ev);
+	}
+
+	sfc_ef10_ev_qclear(rxq->evq_hw_ring, rxq->evq_ptr_mask,
+			   evq_old_read_ptr, rxq->evq_read_ptr);
+
+	return rxq->bufs_pending;
 }
 
 static sfc_dp_rx_qdesc_status_t sfc_ef10_essb_rx_qdesc_status;