All of lore.kernel.org
 help / color / mirror / Atom feed
From: Thierry Herbelot <thierry.herbelot@6wind.com>
To: dev@dpdk.org
Cc: Thibaut Collet <thibaut.collet@6wind.com>,
	stable@dpdk.org, Thomas Monjalon <thomas@monjalon.net>
Subject: [dpdk-dev] [PATCH 19.11 08/12] virtio: fix rx stats with vectorized functions
Date: Wed,  7 Aug 2019 16:37:27 +0200	[thread overview]
Message-ID: <cdaee805ad0cc271863597388a292c516562d4d8.1565188248.git.thierry.herbelot@6wind.com> (raw)
In-Reply-To: <cover.1565188248.git.thierry.herbelot@6wind.com>
In-Reply-To: <cover.1565188248.git.thierry.herbelot@6wind.com>

From: Thibaut Collet <thibaut.collet@6wind.com>

With vectorized functions, only the rx stats for number of packets is
incremented.
Update also the other statistics.
Performance impact is about 2%

Fixes: fc3d66212fed ("virtio: add vector Rx")
Cc: stable at dpdk.org

Signed-off-by: Thibaut Collet <thibaut.collet@6wind.com>
---
 drivers/net/virtio/virtio_rxtx.c             | 2 +-
 drivers/net/virtio/virtio_rxtx.h             | 2 ++
 drivers/net/virtio/virtio_rxtx_simple_neon.c | 6 ++++++
 drivers/net/virtio/virtio_rxtx_simple_sse.c  | 6 ++++++
 4 files changed, 15 insertions(+), 1 deletion(-)

diff --git a/drivers/net/virtio/virtio_rxtx.c b/drivers/net/virtio/virtio_rxtx.c
index 27ead19fbe81..6dd62bf51863 100644
--- a/drivers/net/virtio/virtio_rxtx.c
+++ b/drivers/net/virtio/virtio_rxtx.c
@@ -1083,7 +1083,7 @@ virtio_discard_rxbuf_inorder(struct virtqueue *vq, struct rte_mbuf *m)
 	}
 }
 
-static inline void
+void
 virtio_update_packet_stats(struct virtnet_stats *stats, struct rte_mbuf *mbuf)
 {
 	uint32_t s = mbuf->pkt_len;
diff --git a/drivers/net/virtio/virtio_rxtx.h b/drivers/net/virtio/virtio_rxtx.h
index 685cc4f8104c..1eb8dae227ee 100644
--- a/drivers/net/virtio/virtio_rxtx.h
+++ b/drivers/net/virtio/virtio_rxtx.h
@@ -59,5 +59,7 @@ struct virtnet_ctl {
 };
 
 int virtio_rxq_vec_setup(struct virtnet_rx *rxvq);
+void virtio_update_packet_stats(struct virtnet_stats *stats,
+				struct rte_mbuf *mbuf);
 
 #endif /* _VIRTIO_RXTX_H_ */
diff --git a/drivers/net/virtio/virtio_rxtx_simple_neon.c b/drivers/net/virtio/virtio_rxtx_simple_neon.c
index cdc2a4d28ed5..e4b18cba0ce5 100644
--- a/drivers/net/virtio/virtio_rxtx_simple_neon.c
+++ b/drivers/net/virtio/virtio_rxtx_simple_neon.c
@@ -47,6 +47,7 @@ virtio_recv_pkts_vec(void *rx_queue, struct rte_mbuf **rx_pkts,
 	struct vring_used_elem *rused;
 	struct rte_mbuf **sw_ring;
 	struct rte_mbuf **sw_ring_end;
+	struct rte_mbuf **ref_rx_pkts;
 	uint16_t nb_pkts_received = 0;
 
 	uint8x16_t shuf_msk1 = {
@@ -105,6 +106,7 @@ virtio_recv_pkts_vec(void *rx_queue, struct rte_mbuf **rx_pkts,
 			virtqueue_notify(vq);
 	}
 
+	ref_rx_pkts = rx_pkts;
 	for (nb_pkts_received = 0;
 		nb_pkts_received < nb_used;) {
 		uint64x2_t desc[RTE_VIRTIO_DESC_PER_LOOP / 2];
@@ -204,5 +206,9 @@ virtio_recv_pkts_vec(void *rx_queue, struct rte_mbuf **rx_pkts,
 	vq->vq_used_cons_idx += nb_pkts_received;
 	vq->vq_free_cnt += nb_pkts_received;
 	rxvq->stats.packets += nb_pkts_received;
+	for (nb_used = 0; nb_used < nb_pkts_received; nb_used++) {
+		rxvq->stats.bytes += ref_rx_pkts[nb_used]->pkt_len;
+		virtio_update_packet_stats(&rxvq->stats, ref_rx_pkts[nb_used]);
+	}
 	return nb_pkts_received;
 }
diff --git a/drivers/net/virtio/virtio_rxtx_simple_sse.c b/drivers/net/virtio/virtio_rxtx_simple_sse.c
index af76708d66ae..c757e8c9d601 100644
--- a/drivers/net/virtio/virtio_rxtx_simple_sse.c
+++ b/drivers/net/virtio/virtio_rxtx_simple_sse.c
@@ -48,6 +48,7 @@ virtio_recv_pkts_vec(void *rx_queue, struct rte_mbuf **rx_pkts,
 	struct vring_used_elem *rused;
 	struct rte_mbuf **sw_ring;
 	struct rte_mbuf **sw_ring_end;
+	struct rte_mbuf **ref_rx_pkts;
 	uint16_t nb_pkts_received = 0;
 	__m128i shuf_msk1, shuf_msk2, len_adjust;
 
@@ -107,6 +108,7 @@ virtio_recv_pkts_vec(void *rx_queue, struct rte_mbuf **rx_pkts,
 			virtqueue_notify(vq);
 	}
 
+	ref_rx_pkts = rx_pkts;
 	for (nb_pkts_received = 0;
 		nb_pkts_received < nb_used;) {
 		__m128i desc[RTE_VIRTIO_DESC_PER_LOOP / 2];
@@ -190,5 +192,9 @@ virtio_recv_pkts_vec(void *rx_queue, struct rte_mbuf **rx_pkts,
 	vq->vq_used_cons_idx += nb_pkts_received;
 	vq->vq_free_cnt += nb_pkts_received;
 	rxvq->stats.packets += nb_pkts_received;
+	for (nb_used = 0; nb_used < nb_pkts_received; nb_used++) {
+		rxvq->stats.bytes += ref_rx_pkts[nb_used]->pkt_len;
+		virtio_update_packet_stats(&rxvq->stats, ref_rx_pkts[nb_used]);
+	}
 	return nb_pkts_received;
 }
-- 
2.11.0


  parent reply	other threads:[~2019-08-07 14:39 UTC|newest]

Thread overview: 46+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-08-07 14:37 [dpdk-dev] [PATCH 19.11 00/12] Miscellaneous fixes Thierry Herbelot
2019-08-07 14:37 ` [dpdk-dev] [PATCH 19.11 01/12] net/ixgbevf: fix stats update after a PF reset Thierry Herbelot
2019-08-07 14:37 ` [dpdk-dev] [PATCH 19.11 02/12] ethdev: fix description of tx descriptor status Thierry Herbelot
2019-08-07 14:37 ` [dpdk-dev] [PATCH 19.11 03/12] net/e1000: fix Tx descriptor status api (igb) Thierry Herbelot
2019-08-07 14:37 ` [dpdk-dev] [PATCH 19.11 04/12] net/e1000: fix Tx descriptor status api (em) Thierry Herbelot
2019-08-07 14:37 ` [dpdk-dev] [PATCH 19.11 05/12] net/ixgbe: fix Tx descriptor status api Thierry Herbelot
2019-08-07 14:37 ` [dpdk-dev] [PATCH 19.11 06/12] net/i40e: " Thierry Herbelot
2019-08-07 14:37 ` [dpdk-dev] [PATCH 19.11 07/12] net/i40e: set speed to undefined for default case in link update Thierry Herbelot
2019-08-07 14:37 ` Thierry Herbelot [this message]
2019-08-07 14:37 ` [dpdk-dev] [PATCH 19.11 09/12] virtio: get all pending rx packets with vectorized functions Thierry Herbelot
2019-08-07 14:37 ` [dpdk-dev] [PATCH 19.11 10/12] drivers/crypto/openssl: use a local copy for the session contexts Thierry Herbelot
2019-08-07 14:37 ` [dpdk-dev] [PATCH 19.11 11/12] drivers/crypto/dpaa_sec: update DPAA iova table in dpaa_mem_vtop Thierry Herbelot
2019-08-07 14:37 ` [dpdk-dev] [PATCH 19.11 12/12] drivers/crypto/octeontx: enable unbinding for the OcteonTx crypto engines Thierry Herbelot
2019-08-07 15:09 ` [dpdk-dev] [PATCH 19.11 V2 00/12] Miscellaneous fixes Thierry Herbelot
2019-08-08  8:22   ` [dpdk-dev] [PATCH 19.11 V3 " Thierry Herbelot
2019-08-08 13:19     ` Thierry Herbelot
2019-08-08 14:34       ` Thomas Monjalon
2019-10-10 13:01         ` David Marchand
2019-10-10 13:12           ` Thierry Herbelot
2019-08-08  8:22   ` [dpdk-dev] [PATCH 19.11 V3 01/12] net/ixgbevf: fix stats update after a PF reset Thierry Herbelot
2019-08-08  8:22   ` [dpdk-dev] [PATCH 19.11 V3 02/12] ethdev: fix description of tx descriptor status Thierry Herbelot
2019-08-08 10:37     ` Andrew Rybchenko
2019-08-08  8:22   ` [dpdk-dev] [PATCH 19.11 V3 03/12] net/e1000: fix Tx descriptor status api (igb) Thierry Herbelot
2019-08-08  8:22   ` [dpdk-dev] [PATCH 19.11 V3 04/12] net/e1000: fix Tx descriptor status api (em) Thierry Herbelot
2019-08-08  8:22   ` [dpdk-dev] [PATCH 19.11 V3 05/12] net/ixgbe: fix Tx descriptor status api Thierry Herbelot
2019-08-08  8:22   ` [dpdk-dev] [PATCH 19.11 V3 06/12] net/i40e: " Thierry Herbelot
2019-08-08  8:22   ` [dpdk-dev] [PATCH 19.11 V3 07/12] net/i40e: set speed to undefined for default case in link update Thierry Herbelot
2019-08-08  8:22   ` [dpdk-dev] [PATCH 19.11 V3 08/12] virtio: fix rx stats with vectorized functions Thierry Herbelot
2019-08-08  8:22   ` [dpdk-dev] [PATCH 19.11 V3 09/12] virtio: get all pending rx packets " Thierry Herbelot
2019-08-08  8:22   ` [dpdk-dev] [PATCH 19.11 V3 10/12] drivers/crypto/openssl: use a local copy for the session contexts Thierry Herbelot
2019-08-08  8:22   ` [dpdk-dev] [PATCH 19.11 V3 11/12] drivers/crypto/dpaa_sec: update DPAA iova table in dpaa_mem_vtop Thierry Herbelot
2019-08-08  8:22   ` [dpdk-dev] [PATCH 19.11 V3 12/12] drivers/crypto/octeontx: enable unbinding for the OcteonTx crypto engines Thierry Herbelot
2019-08-07 15:09 ` [dpdk-dev] [PATCH 19.11 V2 01/12] net/ixgbevf: fix stats update after a PF reset Thierry Herbelot
2019-08-07 15:09 ` [dpdk-dev] [PATCH 19.11 V2 02/12] ethdev: fix description of tx descriptor status Thierry Herbelot
2019-08-07 15:09 ` [dpdk-dev] [PATCH 19.11 V2 03/12] net/e1000: fix Tx descriptor status api (igb) Thierry Herbelot
2019-08-07 15:09 ` [dpdk-dev] [PATCH 19.11 V2 04/12] net/e1000: fix Tx descriptor status api (em) Thierry Herbelot
2019-08-07 15:09 ` [dpdk-dev] [PATCH 19.11 V2 05/12] net/ixgbe: fix Tx descriptor status api Thierry Herbelot
2019-08-07 15:09 ` [dpdk-dev] [PATCH 19.11 V2 06/12] net/i40e: " Thierry Herbelot
2019-08-07 15:09 ` [dpdk-dev] [PATCH 19.11 V2 07/12] net/i40e: set speed to undefined for default case in link update Thierry Herbelot
2019-08-07 15:09 ` [dpdk-dev] [PATCH 19.11 V2 08/12] virtio: fix rx stats with vectorized functions Thierry Herbelot
2019-08-08  5:15   ` [dpdk-dev] [dpdk-stable] " Tiwei Bie
2019-08-08  7:35     ` Thibaut Collet
2019-08-07 15:09 ` [dpdk-dev] [PATCH 19.11 V2 09/12] virtio: get all pending rx packets " Thierry Herbelot
2019-08-07 15:09 ` [dpdk-dev] [PATCH 19.11 V2 10/12] drivers/crypto/openssl: use a local copy for the session contexts Thierry Herbelot
2019-08-07 15:09 ` [dpdk-dev] [PATCH 19.11 V2 11/12] drivers/crypto/dpaa_sec: update DPAA iova table in dpaa_mem_vtop Thierry Herbelot
2019-08-07 15:09 ` [dpdk-dev] [PATCH 19.11 V2 12/12] drivers/crypto/octeontx: enable unbinding for the OcteonTx crypto engines Thierry Herbelot

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=cdaee805ad0cc271863597388a292c516562d4d8.1565188248.git.thierry.herbelot@6wind.com \
    --to=thierry.herbelot@6wind.com \
    --cc=dev@dpdk.org \
    --cc=stable@dpdk.org \
    --cc=thibaut.collet@6wind.com \
    --cc=thomas@monjalon.net \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.