[dpdk-dev] [PATCH v3 5/5] net/vhost: remove limit of vhost RX burst size

[Zhiyong Yang <zhiyong.yang@intel.com>]

vhost removes limit of RX burst size(32 pkts) and supports to make
an best effort to receive pkts.

Cc: yuanhan.liu@linux.intel.com
Cc: maxime.coquelin@redhat.com
Signed-off-by: Zhiyong Yang <zhiyong.yang@intel.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
---
 doc/guides/rel_notes/release_17_05.rst |  4 ++++
 drivers/net/vhost/rte_eth_vhost.c      | 17 +++++++++++++++--
 2 files changed, 19 insertions(+), 2 deletions(-)

diff --git a/doc/guides/rel_notes/release_17_05.rst b/doc/guides/rel_notes/release_17_05.rst
index 0c33f7b..226ecd6 100644
--- a/doc/guides/rel_notes/release_17_05.rst
+++ b/doc/guides/rel_notes/release_17_05.rst
@@ -127,6 +127,10 @@ New Features
   performance enhancements viz. configurable TX data ring, Receive
   Data Ring, ability to register memory regions.
 
+* **Kept consistent PMD batching behaviour.**
+
+  Removed the limit of fm10k/i40e/ixgbe TX burst size and vhost RX/TX burst size in
+  order to support the same policy "make an best effort to RX/TX pkts" for PMDs.
 
 Resolved Issues
 ---------------
diff --git a/drivers/net/vhost/rte_eth_vhost.c b/drivers/net/vhost/rte_eth_vhost.c
index 2a38b19..7f5cd7e 100644
--- a/drivers/net/vhost/rte_eth_vhost.c
+++ b/drivers/net/vhost/rte_eth_vhost.c
@@ -392,6 +392,7 @@ eth_vhost_rx(void *q, struct rte_mbuf **bufs, uint16_t nb_bufs)
 {
 	struct vhost_queue *r = q;
 	uint16_t i, nb_rx = 0;
+	uint16_t nb_receive = nb_bufs;
 
 	if (unlikely(rte_atomic32_read(&r->allow_queuing) == 0))
 		return 0;
@@ -402,8 +403,20 @@ eth_vhost_rx(void *q, struct rte_mbuf **bufs, uint16_t nb_bufs)
 		goto out;
 
 	/* Dequeue packets from guest TX queue */
-	nb_rx = rte_vhost_dequeue_burst(r->vid,
-			r->virtqueue_id, r->mb_pool, bufs, nb_bufs);
+	while (nb_receive) {
+		uint16_t nb_pkts;
+		uint16_t num = (uint16_t)RTE_MIN(nb_receive,
+						 VHOST_MAX_PKT_BURST);
+
+		nb_pkts = rte_vhost_dequeue_burst(r->vid, r->virtqueue_id,
+						  r->mb_pool, &bufs[nb_rx],
+						  num);
+
+		nb_rx += nb_pkts;
+		nb_receive -= nb_pkts;
+		if (nb_pkts < num)
+			break;
+	}
 
 	r->stats.pkts += nb_rx;
 
-- 
2.7.4