From: Jason Wang <jasowang@xxxxxxxxxx>
We already have counters for sent packets and sent bytes.
Use them to reduce the number of u64_stats_update_begin/end().
Take care not to bother with stats update when called
speculatively.
Based on a patch by Jason Wang.
Cc: Rusty Russell <rusty@xxxxxxxxxxxxxxx>
Signed-off-by: Jason Wang <jasowang@xxxxxxxxxx>
Signed-off-by: Michael S. Tsirkin <mst@xxxxxxxxxx>
---
drivers/net/virtio_net.c | 16 +++++++++++-----
1 file changed, 11 insertions(+), 5 deletions(-)
diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index b83d39d..c2b69f8 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -233,16 +233,22 @@ static unsigned int free_old_xmit_skbs(struct netdev_queue *txq,
(skb = virtqueue_get_buf(sq->vq, &len)) != NULL) {
pr_debug("Sent skb %p\n", skb);
- u64_stats_update_begin(&stats->tx_syncp);
- stats->tx_bytes += skb->len;
bytes += skb->len;
- stats->tx_packets++;
- u64_stats_update_end(&stats->tx_syncp);
+ packets++;
dev_kfree_skb_any(skb);
- packets++;
}
+ /* Avoid overhead when no packets have been processed
+ * happens when called speculatively from start_xmit. */
+ if (!packets)
+ return 0;
+
+ u64_stats_update_begin(&stats->tx_syncp);
+ stats->tx_bytes += bytes;
+ stats->tx_packets += packets;
+ u64_stats_update_end(&stats->tx_syncp);
+
netdev_tx_completed_queue(txq, packets, bytes);
if (sq->vq->num_free >= 2+MAX_SKB_FRAGS)
--
MST
_______________________________________________
Virtualization mailing list
Virtualization@xxxxxxxxxxxxxxxxxxxxxxxxxx
https://lists.linuxfoundation.org/mailman/listinfo/virtualization
