DPDK patches and discussions
 help / color / mirror / Atom feed
From: Jiayu Hu <jiayu.hu@intel.com>
To: dev@dpdk.org
Cc: maxime.coquelin@redhat.com, chenbo.xia@intel.com,
	cheng1.jiang@intel.com, yinan.wang@intel.com,
	Jiayu Hu <jiayu.hu@intel.com>
Subject: [dpdk-dev] [PATCH 1/2] vhost: cleanup async enqueue
Date: Fri, 11 Dec 2020 04:21:39 -0500	[thread overview]
Message-ID: <1607678500-172518-2-git-send-email-jiayu.hu@intel.com> (raw)
In-Reply-To: <1607678500-172518-1-git-send-email-jiayu.hu@intel.com>

This patch removes unncessary check and function calls, and it changes
appropriate types for internal variables and fixes typos.

Signed-off-by: Jiayu Hu <jiayu.hu@intel.com>
---
 lib/librte_vhost/rte_vhost_async.h |  6 +++---
 lib/librte_vhost/virtio_net.c      | 16 ++++++++--------
 2 files changed, 11 insertions(+), 11 deletions(-)

diff --git a/lib/librte_vhost/rte_vhost_async.h b/lib/librte_vhost/rte_vhost_async.h
index c73bd7c..3be4ee4 100644
--- a/lib/librte_vhost/rte_vhost_async.h
+++ b/lib/librte_vhost/rte_vhost_async.h
@@ -147,8 +147,8 @@ __rte_experimental
 int rte_vhost_async_channel_unregister(int vid, uint16_t queue_id);
 
 /**
- * This function submit enqueue data to async engine. This function has
- * no guranttee to the transfer completion upon return. Applications
+ * This function submits enqueue data to async engine. This function has
+ * no guarantee to the transfer completion upon return. Applications
  * should poll transfer status by rte_vhost_poll_enqueue_completed()
  *
  * @param vid
@@ -167,7 +167,7 @@ uint16_t rte_vhost_submit_enqueue_burst(int vid, uint16_t queue_id,
 		struct rte_mbuf **pkts, uint16_t count);
 
 /**
- * This function check async completion status for a specific vhost
+ * This function checks async completion status for a specific vhost
  * device queue. Packets which finish copying (enqueue) operation
  * will be returned in an array.
  *
diff --git a/lib/librte_vhost/virtio_net.c b/lib/librte_vhost/virtio_net.c
index 6c51286..fc654be 100644
--- a/lib/librte_vhost/virtio_net.c
+++ b/lib/librte_vhost/virtio_net.c
@@ -1128,8 +1128,11 @@ async_mbuf_to_desc(struct virtio_net *dev, struct vhost_virtqueue *vq,
 	}
 
 out:
-	async_fill_iter(src_it, tlen, src_iovec, tvec_idx);
-	async_fill_iter(dst_it, tlen, dst_iovec, tvec_idx);
+	if (tlen) {
+		async_fill_iter(src_it, tlen, src_iovec, tvec_idx);
+		async_fill_iter(dst_it, tlen, dst_iovec, tvec_idx);
+	} else
+		src_it->count = 0;
 
 	return error;
 }
@@ -1492,10 +1495,9 @@ virtio_dev_rx_async_submit_split(struct virtio_net *dev,
 	struct rte_vhost_iov_iter *src_it = it_pool;
 	struct rte_vhost_iov_iter *dst_it = it_pool + 1;
 	uint16_t n_free_slot, slot_idx = 0;
-	uint16_t pkt_err = 0;
 	uint16_t segs_await = 0;
 	struct async_inflight_info *pkts_info = vq->async_pkts_info;
-	int n_pkts = 0;
+	uint32_t n_pkts = 0, pkt_err = 0;
 
 	avail_head = __atomic_load_n(&vq->avail->idx, __ATOMIC_ACQUIRE);
 
@@ -1553,11 +1555,9 @@ virtio_dev_rx_async_submit_split(struct virtio_net *dev,
 		/*
 		 * conditions to trigger async device transfer:
 		 * - buffered packet number reaches transfer threshold
-		 * - this is the last packet in the burst enqueue
 		 * - unused async iov number is less than max vhost vector
 		 */
 		if (pkt_burst_idx >= VHOST_ASYNC_BATCH_THRESHOLD ||
-			(pkt_idx == count - 1 && pkt_burst_idx) ||
 			(VHOST_MAX_ASYNC_VEC / 2 - segs_await <
 			BUF_VECTOR_MAX)) {
 			n_pkts = vq->async_ops.transfer_data(dev->vid,
@@ -1569,7 +1569,7 @@ virtio_dev_rx_async_submit_split(struct virtio_net *dev,
 			segs_await = 0;
 			vq->async_pkts_inflight_n += pkt_burst_idx;
 
-			if (unlikely(n_pkts < (int)pkt_burst_idx)) {
+			if (unlikely(n_pkts < pkt_burst_idx)) {
 				/*
 				 * log error packets number here and do actual
 				 * error processing when applications poll
@@ -1589,7 +1589,7 @@ virtio_dev_rx_async_submit_split(struct virtio_net *dev,
 				queue_id, tdes, 0, pkt_burst_idx);
 		vq->async_pkts_inflight_n += pkt_burst_idx;
 
-		if (unlikely(n_pkts < (int)pkt_burst_idx))
+		if (unlikely(n_pkts < pkt_burst_idx))
 			pkt_err = pkt_burst_idx - n_pkts;
 	}
 
-- 
2.7.4


  reply	other threads:[~2020-12-11  2:50 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-12-11  9:21 [dpdk-dev] [PATCH 0/2] Enhance Async Enqueue for Small Packets Jiayu Hu
2020-12-11  9:21 ` Jiayu Hu [this message]
2020-12-11  9:21 ` [dpdk-dev] [PATCH 2/2] vhost: enhance async enqueue for small packets Jiayu Hu
2020-12-22  9:46 ` [dpdk-dev] [Patch v2 0/2] Enhance Async Enqueue for Small Packets Jiayu Hu
2020-12-22  9:46   ` [dpdk-dev] [Patch v2 1/2] vhost: cleanup async enqueue Jiayu Hu
2020-12-22  9:46   ` [dpdk-dev] [Patch v2 2/2] vhost: enhance async enqueue for small packets Jiayu Hu
2020-12-25  8:28   ` [dpdk-dev] [PATCH v3 0/2] Enhance Async Enqueue for Small Packets Jiayu Hu
2020-12-25  8:28     ` [dpdk-dev] [PATCH v3 1/2] vhost: cleanup async enqueue Jiayu Hu
2020-12-25  8:28     ` [dpdk-dev] [PATCH v3 2/2] vhost: enhance async enqueue for small packets Jiayu Hu
2021-01-05 11:41     ` [dpdk-dev] [PATCH v3 0/2] Enhance Async Enqueue for Small Packets Wang, Yinan
2021-01-07 10:45     ` Maxime Coquelin
2021-01-11 12:16     ` [dpdk-dev] [PATCH v4 " Jiayu Hu
2021-01-11 12:16       ` [dpdk-dev] [PATCH v4 1/2] vhost: cleanup async enqueue Jiayu Hu
2021-01-11 11:04         ` Maxime Coquelin
2021-01-11 14:04           ` Maxime Coquelin
2021-01-11 12:16       ` [dpdk-dev] [PATCH v4 2/2] vhost: enhance async enqueue for small packets Jiayu Hu
2021-01-11 12:44         ` Maxime Coquelin
2021-01-11 15:02       ` [dpdk-dev] [PATCH v4 0/2] Enhance Async Enqueue for Small Packets Maxime Coquelin

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=1607678500-172518-2-git-send-email-jiayu.hu@intel.com \
    --to=jiayu.hu@intel.com \
    --cc=chenbo.xia@intel.com \
    --cc=cheng1.jiang@intel.com \
    --cc=dev@dpdk.org \
    --cc=maxime.coquelin@redhat.com \
    --cc=yinan.wang@intel.com \
    /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).