vmxnet3: restore Tx data ring support

Tx data ring support was removed in a previous change that
added multi-seg transmit.  This change adds it back.

According to the original commit (2e849373), 64B pkt
rate with l2fwd improved by ~20% on an Ivy Bridge
server at which point we start to hit some bottleneck
on the rx side.

I also re-did the same test on a different setup (Haswell
processor, ~2.3GHz clock rate) on top of the master
and still observed ~17% performance gains.

Fixes: 7ba5de417e3c ("vmxnet3: support multi-segment transmit")

Signed-off-by: Yong Wang <yongwang@vmware.com>
Acked-by: Stephen Hemminger <stephen@networkplumber.org>

diff --git a/doc/guides/rel_notes/release_16_04.rst b/doc/guides/rel_notes/release_16_04.rst
index 06879e0..522c463 100644 (file)
--- a/doc/guides/rel_notes/release_16_04.rst
+++ b/doc/guides/rel_notes/release_16_04.rst
@@ -47,6 +47,11 @@ This section should contain new features added in this release. Sample format:
   A new function ``rte_pktmbuf_alloc_bulk()`` has been added to allow the user
   to allocate a bulk of mbufs.
 
+* **Restored vmxnet3 Tx data ring.**
+
+  Tx data ring has been shown to improve small pkt forwarding performance
+  on vSphere environment.
+
 * **Virtio 1.0.**
 
   Enabled virtio 1.0 support for virtio pmd driver.

diff --git a/drivers/net/vmxnet3/vmxnet3_rxtx.c b/drivers/net/vmxnet3/vmxnet3_rxtx.c
index 8a8e441..5f20745 100644 (file)
--- a/drivers/net/vmxnet3/vmxnet3_rxtx.c
+++ b/drivers/net/vmxnet3/vmxnet3_rxtx.c
@@ -342,6 +342,7 @@ vmxnet3_xmit_pkts(void *tx_queue, struct rte_mbuf **tx_pkts,
                uint32_t first2fill, avail, dw2;
                struct rte_mbuf *txm = tx_pkts[nb_tx];
                struct rte_mbuf *m_seg = txm;
+               int copy_size = 0;
 
                /* Is this packet execessively fragmented, then drop */
                if (unlikely(txm->nb_segs > VMXNET3_MAX_TXD_PER_PKT)) {
@@ -359,6 +360,14 @@ vmxnet3_xmit_pkts(void *tx_queue, struct rte_mbuf **tx_pkts,
                        break;
                }
 
+               if (txm->nb_segs == 1 && rte_pktmbuf_pkt_len(txm) <= VMXNET3_HDR_COPY_SIZE) {
+                       struct Vmxnet3_TxDataDesc *tdd;
+
+                       tdd = txq->data_ring.base + txq->cmd_ring.next2fill;
+                       copy_size = rte_pktmbuf_pkt_len(txm);
+                       rte_memcpy(tdd->data, rte_pktmbuf_mtod(txm, char *), copy_size);
+               }
+
                /* use the previous gen bit for the SOP desc */
                dw2 = (txq->cmd_ring.gen ^ 0x1) << VMXNET3_TXD_GEN_SHIFT;
                first2fill = txq->cmd_ring.next2fill;
@@ -371,7 +380,13 @@ vmxnet3_xmit_pkts(void *tx_queue, struct rte_mbuf **tx_pkts,
                           transmit buffer size (16K) is greater than
                           maximum sizeof mbuf segment size. */
                        gdesc = txq->cmd_ring.base + txq->cmd_ring.next2fill;
-                       gdesc->txd.addr = rte_mbuf_data_dma_addr(m_seg);
+                       if (copy_size)
+                               gdesc->txd.addr = rte_cpu_to_le_64(txq->data_ring.basePA +
+                                                               txq->cmd_ring.next2fill *
+                                                               sizeof(struct Vmxnet3_TxDataDesc));
+                       else
+                               gdesc->txd.addr = rte_mbuf_data_dma_addr(m_seg);
+
                        gdesc->dword[2] = dw2 | m_seg->data_len;
                        gdesc->dword[3] = 0;