diff mbox

[dpdk-dev,7/8] net/vmxnet3: ignore emtpy segments in reception

Message ID 20180328154349.24976-8-didier.pallard@6wind.com (mailing list archive)
State Accepted, archived
Delegated to: Ferruh Yigit
Headers

Checks

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

Commit Message

Didier Pallard March 28, 2018, 3:43 p.m. UTC 
  When several TCP fragments are contained in a packet that is only one mbuf
segment long, vmxnet3 receives an empty segment following first one, that
contains offload information. In current version, this segment is
propagated as is to upper application.
Remove those empty segments directly when receiving buffers, they may
generate unneeded extra processing in the upper application.

Signed-off-by: Didier Pallard <didier.pallard@6wind.com>
---
 drivers/net/vmxnet3/vmxnet3_rxtx.c | 13 +++++++++----
 1 file changed, 9 insertions(+), 4 deletions(-)
diff mbox

Patch

diff --git a/drivers/net/vmxnet3/vmxnet3_rxtx.c b/drivers/net/vmxnet3/vmxnet3_rxtx.c
index 1f273f88e..1d344b26e 100644
--- a/drivers/net/vmxnet3/vmxnet3_rxtx.c
+++ b/drivers/net/vmxnet3/vmxnet3_rxtx.c
@@ -925,18 +925,23 @@  vmxnet3_recv_pkts(void *rx_queue, struct rte_mbuf **rx_pkts, uint16_t nb_pkts)
 			}
 
 			rxq->start_seg = rxm;
+			rxq->last_seg = rxm;
 			vmxnet3_rx_offload(hw, rcd, rxm, 1);
 		} else {
 			struct rte_mbuf *start = rxq->start_seg;
 
 			RTE_ASSERT(rxd->btype == VMXNET3_RXD_BTYPE_BODY);
 
-			start->pkt_len += rxm->data_len;
-			start->nb_segs++;
+			if (rxm->data_len) {
+				start->pkt_len += rxm->data_len;
+				start->nb_segs++;
 
-			rxq->last_seg->next = rxm;
+				rxq->last_seg->next = rxm;
+				rxq->last_seg = rxm;
+			} else {
+				rte_pktmbuf_free_seg(rxm);
+			}
 		}
-		rxq->last_seg = rxm;
 
 		if (rcd->eop) {
 			struct rte_mbuf *start = rxq->start_seg;