From: Tiwei Bie <tiwei.bie@intel.com>
To: dev@dpdk.org
Subject: [dpdk-dev] [RFC 16/29] net/virtio: avoid touching packet data
Date: Wed, 21 Jun 2017 10:57:52 +0800 [thread overview]
Message-ID: <1498013885-102779-17-git-send-email-tiwei.bie@intel.com> (raw)
In-Reply-To: <1498013885-102779-1-git-send-email-tiwei.bie@intel.com>
For performance testing purpose, avoid touching the packet data
when receiving packets.
Signed-off-by: Tiwei Bie <tiwei.bie@intel.com>
---
drivers/net/virtio/virtio_rxtx.c | 8 ++++++++
1 file changed, 8 insertions(+)
diff --git a/drivers/net/virtio/virtio_rxtx.c b/drivers/net/virtio/virtio_rxtx.c
index 3be64da..93d564f 100644
--- a/drivers/net/virtio/virtio_rxtx.c
+++ b/drivers/net/virtio/virtio_rxtx.c
@@ -139,7 +139,9 @@ virtqueue_dequeue_burst_rx_1_0(struct virtqueue *vq, struct rte_mbuf **rx_pkts,
}
rte_prefetch0(cookie);
+#if 0
rte_packet_prefetch(rte_pktmbuf_mtod(cookie, void *));
+#endif
rx_pkts[i] = cookie;
vq->vq_used_cons_idx++;
vq_ring_free_chain(vq, desc_idx);
@@ -174,7 +176,9 @@ virtqueue_dequeue_burst_rx_1_1(struct virtqueue *vq, struct rte_mbuf **rx_pkts,
vq->vq_descx[used_idx].cookie = NULL;
rte_prefetch0(cookie);
+#if 0
rte_packet_prefetch(rte_pktmbuf_mtod(cookie, void *));
+#endif
rx_pkts[i] = cookie;
vq->vq_used_cons_idx++;
@@ -568,7 +572,9 @@ static void
virtio_update_packet_stats(struct virtnet_stats *stats, struct rte_mbuf *mbuf)
{
uint32_t s = mbuf->pkt_len;
+#if 0
struct ether_addr *ea;
+#endif
if (s == 64) {
stats->size_bins[1]++;
@@ -587,6 +593,7 @@ virtio_update_packet_stats(struct virtnet_stats *stats, struct rte_mbuf *mbuf)
stats->size_bins[7]++;
}
+#if 0
ea = rte_pktmbuf_mtod(mbuf, struct ether_addr *);
if (is_multicast_ether_addr(ea)) {
if (is_broadcast_ether_addr(ea))
@@ -594,6 +601,7 @@ virtio_update_packet_stats(struct virtnet_stats *stats, struct rte_mbuf *mbuf)
else
stats->multicast++;
}
+#endif
}
/* Optionally fill offload information in structure */
--
2.7.4
next prev parent reply other threads:[~2017-06-21 2:59 UTC|newest]
Thread overview: 30+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-06-21 2:57 [dpdk-dev] [RFC 00/29] latest virtio1.1 prototype Tiwei Bie
2017-06-21 2:57 ` [dpdk-dev] [RFC 01/29] net/virtio: vring init for 1.1 Tiwei Bie
2017-06-21 2:57 ` [dpdk-dev] [RFC 02/29] net/virtio: implement 1.1 guest Tx Tiwei Bie
2017-06-21 2:57 ` [dpdk-dev] [RFC 03/29] net/virtio-user: add option to enable 1.1 Tiwei Bie
2017-06-21 2:57 ` [dpdk-dev] [RFC 04/29] vhost: enable 1.1 for testing Tiwei Bie
2017-06-21 2:57 ` [dpdk-dev] [RFC 05/29] vhost: set desc addr for 1.1 Tiwei Bie
2017-06-21 2:57 ` [dpdk-dev] [RFC 06/29] vhost: implement virtio 1.1 dequeue path Tiwei Bie
2017-06-21 2:57 ` [dpdk-dev] [RFC 07/29] vhost: mark desc being used Tiwei Bie
2017-06-21 2:57 ` [dpdk-dev] [RFC 08/29] xxx: batch the desc_hw update? Tiwei Bie
2017-06-21 2:57 ` [dpdk-dev] [RFC 09/29] xxx: virtio: remove overheads Tiwei Bie
2017-06-21 2:57 ` [dpdk-dev] [RFC 10/29] vhost: prefetch desc Tiwei Bie
2017-06-21 2:57 ` [dpdk-dev] [RFC 11/29] add virtio 1.1 test guide Tiwei Bie
2017-06-21 2:57 ` [dpdk-dev] [RFC 12/29] testpmd: add s-txonly Tiwei Bie
2017-06-21 2:57 ` [dpdk-dev] [RFC 13/29] net/virtio: implement the Rx code path Tiwei Bie
2017-06-21 2:57 ` [dpdk-dev] [RFC 14/29] vhost: a rough implementation on enqueue " Tiwei Bie
2017-06-21 2:57 ` [dpdk-dev] [RFC 15/29] vhost: descriptor length should include vhost header Tiwei Bie
2017-06-21 2:57 ` Tiwei Bie [this message]
2017-06-21 2:57 ` [dpdk-dev] [RFC 17/29] net/virtio: fix virtio1.1 feature negotiation Tiwei Bie
2017-06-21 2:57 ` [dpdk-dev] [RFC 18/29] net/virtio: the Rx support for virtio1.1 has been added now Tiwei Bie
2017-06-21 2:57 ` [dpdk-dev] [RFC 19/29] vhost: VIRTIO_NET_F_MRG_RXBUF is not supported for now Tiwei Bie
2017-06-21 2:57 ` [dpdk-dev] [RFC 20/29] vhost: fix vring addr setup Tiwei Bie
2017-06-21 2:57 ` [dpdk-dev] [RFC 21/29] net/virtio: free mbuf when need to use Tiwei Bie
2017-06-21 2:57 ` [dpdk-dev] [RFC 22/29] vhost: don't copy descs during Rx Tiwei Bie
2017-06-21 2:57 ` [dpdk-dev] [RFC 23/29] vhost: fix mbuf leak Tiwei Bie
2017-06-21 2:58 ` [dpdk-dev] [RFC 24/29] net/virtio: cleanup txd when free count below threshold Tiwei Bie
2017-06-21 2:58 ` [dpdk-dev] [RFC 25/29] net/virtio: refill descs for vhost in batch Tiwei Bie
2017-06-21 2:58 ` [dpdk-dev] [RFC 26/29] vhost: remove dead code Tiwei Bie
2017-06-21 2:58 ` [dpdk-dev] [RFC 27/29] vhost: various optimizations for Tx Tiwei Bie
2017-06-21 2:58 ` [dpdk-dev] [RFC 28/29] vhost: make the code more readable Tiwei Bie
2017-06-21 2:58 ` [dpdk-dev] [RFC 29/29] vhost: update and return descs in batch Tiwei Bie
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1498013885-102779-17-git-send-email-tiwei.bie@intel.com \
--to=tiwei.bie@intel.com \
--cc=dev@dpdk.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).