[PATCH net-next v3 0/5] virtio-net tx napi

[PATCH net-next v3 0/5] virtio-net tx napi

[Willem de Bruijn]

From: Willem de Bruijn <willemb@google.com>

Add napi for virtio-net transmit completion processing.

Changes:
  v2 -> v3:
    - convert __netif_tx_trylock to __netif_tx_lock on tx napi poll
          ensure that the handler always cleans, to avoid deadlock
    - unconditionally clean in start_xmit
          avoid adding an unnecessary "if (use_napi)" branch
    - remove virtqueue_disable_cb in patch 5/5
          a noop in the common event_idx based loop
    - document affinity_hint_set constraint

  v1 -> v2:
    - disable by default
    - disable unless affinity_hint_set
          because cache misses add up to a third higher cycle cost,
	  e.g., in TCP_RR tests. This is not limited to the patch
	  that enables tx completion cleaning in rx napi.
    - use trylock to avoid contention between tx and rx napi
    - keep interrupts masked during xmit_more (new patch 5/5)
          this improves cycles especially for multi UDP_STREAM, which
	  does not benefit from cleaning tx completions on rx napi.
    - move free_old_xmit_skbs (new patch 3/5)
          to avoid forward declaration

    not changed:
    - deduplicate virnet_poll_tx and virtnet_poll_txclean
          they look similar, but have differ too much to make it
	  worthwhile.
    - delay netif_wake_subqueue for more than 2 + MAX_SKB_FRAGS
          evaluated, but made no difference
    - patch 1/5

  RFC -> v1:
    - dropped vhost interrupt moderation patch:
          not needed and likely expensive at light load
    - remove tx napi weight
        - always clean all tx completions
        - use boolean to toggle tx-napi, instead
    - only clean tx in rx if tx-napi is enabled
        - then clean tx before rx
    - fix: add missing braces in virtnet_freeze_down
    - testing: add 4KB TCP_RR + UDP test results

Based on previous patchsets by Jason Wang:

  [RFC V7 PATCH 0/7] enable tx interrupts for virtio-net
  http://lkml.iu.edu/hypermail/linux/kernel/1505.3/00245.html


Before commit b0c39dbdc204 ("virtio_net: don't free buffers in xmit
ring") the virtio-net driver would free transmitted packets on
transmission of new packets in ndo_start_xmit and, to catch the edge
case when no new packet is sent, also in a timer at 10HZ.

A timer can cause long stalls. VIRTIO_F_NOTIFY_ON_EMPTY avoids stalls
due to low free descriptor count. It does not address a stalls due to
low socket SO_SNDBUF. Increasing timer frequency decreases that stall
time, but increases interrupt rate and, thus, cycle count.

Currently, with no timer, packets are freed only at ndo_start_xmit.
Latency of consume_skb is now unbounded. To avoid a deadlock if a sock
reaches SO_SNDBUF, packets are orphaned on tx. This breaks TCP small
queues.

Reenable TCP small queues by removing the orphan. Instead of using a
timer, convert the driver to regular tx napi. This does not have the
unresolved stall issue and does not have any frequency to tune.

By keeping interrupts enabled by default, napi increases tx
interrupt rate. VIRTIO_F_EVENT_IDX avoids sending an interrupt if
one is already unacknowledged, so makes this more feasible today.
Combine that with an optimization that brings interrupt rate
back in line with the existing version for most workloads:

Tx completion cleaning on rx interrupts elides most explicit tx
interrupts by relying on the fact that many rx interrupts fire.

Tested by running {1, 10, 100} {TCP, UDP} STREAM, RR, 4K_RR benchmarks
from a guest to a server on the host, on an x86_64 Haswell. The guest
runs 4 vCPUs pinned to 4 cores. vhost and the test server are
pinned to a core each.

All results are the median of 5 runs, with variance well < 10%.
Used neper (github.com/google/neper) as test process.

Napi increases single stream throughput, but increases cycle cost.
The optimizations bring this down. The previous patchset saw a
regression with UDP_STREAM, which does not benefit from cleaning tx
interrupts in rx napi. This regression is now gone for 10x, 100x.
Remaining difference is higher 1x TCP_STREAM, lower 1x UDP_STREAM.

The latest results are with process, rx napi and tx napi affine to
the same core. All numbers are lower than the previous patchset.


             upstream     napi
TCP_STREAM:
1x:
  Mbps          27816    39805
  Gcycles         274      285

10x:
  Mbps          42947    42531
  Gcycles         300      296

100x:
  Mbps          31830    28042
  Gcycles         279      269

TCP_RR Latency (us):
1x:
  p50              21       21
  p99              27       27
  Gcycles         180      167

10x:
  p50              40       39
  p99              52       52
  Gcycles         214      211

100x:
  p50             281      241
  p99             411      337
  Gcycles         218      226

TCP_RR 4K:
1x:
  p50              28       29
  p99              34       36
  Gcycles         177      167

10x:
  p50              70       71
  p99              85      134
  Gcycles         213      214

100x:
  p50             442      611
  p99             802      785
  Gcycles         237      216

UDP_STREAM:
1x:
  Mbps          29468    26800
  Gcycles         284      293

10x:
  Mbps          29891    29978
  Gcycles         285      312

100x:
  Mbps          30269    30304
  Gcycles         318      316

UDP_RR:
1x:
  p50              19       19
  p99              23       23
  Gcycles         180      173

10x:
  p50              35       40
  p99              54       64
  Gcycles         245      237

100x:
  p50             234      286
  p99             484      473
  Gcycles         224      214

Note that GSO is enabled, so 4K RR still translates to one packet
per request.

Lower throughput at 100x vs 10x can be (at least in part)
explained by looking at bytes per packet sent (nstat). It likely
also explains the lower throughput of 1x for some variants.

upstream:

 N=1   bytes/pkt=16581
 N=10  bytes/pkt=61513
 N=100 bytes/pkt=51558

at_rx:

 N=1   bytes/pkt=65204
 N=10  bytes/pkt=65148
 N=100 bytes/pkt=56840

Willem de Bruijn (5):
  virtio-net: napi helper functions
  virtio-net: transmit napi
  virtio-net: move free_old_xmit_skbs
  virtio-net: clean tx descriptors from rx napi
  virtio-net: keep tx interrupts disabled unless kick

 drivers/net/virtio_net.c | 193 ++++++++++++++++++++++++++++++++---------------
 1 file changed, 132 insertions(+), 61 deletions(-)

-- 
2.12.2.816.g2cccc81164-goog

[PATCH net-next v3 1/5] virtio-net: napi helper functions

[Willem de Bruijn]

From: Willem de Bruijn <willemb@google.com>

Prepare virtio-net for tx napi by converting existing napi code to
use helper functions. This also deduplicates some logic.

Signed-off-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
---
 drivers/net/virtio_net.c | 65 ++++++++++++++++++++++++++----------------------
 1 file changed, 35 insertions(+), 30 deletions(-)

diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index 666ada6130ab..b9c1df29892c 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -239,6 +239,26 @@ static struct page *get_a_page(struct receive_queue *rq, gfp_t gfp_mask)
 	return p;
 }
 
+static void virtqueue_napi_schedule(struct napi_struct *napi,
+				    struct virtqueue *vq)
+{
+	if (napi_schedule_prep(napi)) {
+		virtqueue_disable_cb(vq);
+		__napi_schedule(napi);
+	}
+}
+
+static void virtqueue_napi_complete(struct napi_struct *napi,
+				    struct virtqueue *vq, int processed)
+{
+	int opaque;
+
+	opaque = virtqueue_enable_cb_prepare(vq);
+	if (napi_complete_done(napi, processed) &&
+	    unlikely(virtqueue_poll(vq, opaque)))
+		virtqueue_napi_schedule(napi, vq);
+}
+
 static void skb_xmit_done(struct virtqueue *vq)
 {
 	struct virtnet_info *vi = vq->vdev->priv;
@@ -936,27 +956,20 @@ static void skb_recv_done(struct virtqueue *rvq)
 	struct virtnet_info *vi = rvq->vdev->priv;
 	struct receive_queue *rq = &vi->rq[vq2rxq(rvq)];
 
-	/* Schedule NAPI, Suppress further interrupts if successful. */
-	if (napi_schedule_prep(&rq->napi)) {
-		virtqueue_disable_cb(rvq);
-		__napi_schedule(&rq->napi);
-	}
+	virtqueue_napi_schedule(&rq->napi, rvq);
 }
 
-static void virtnet_napi_enable(struct receive_queue *rq)
+static void virtnet_napi_enable(struct virtqueue *vq, struct napi_struct *napi)
 {
-	napi_enable(&rq->napi);
+	napi_enable(napi);
 
 	/* If all buffers were filled by other side before we napi_enabled, we
-	 * won't get another interrupt, so process any outstanding packets
-	 * now.  virtnet_poll wants re-enable the queue, so we disable here.
-	 * We synchronize against interrupts via NAPI_STATE_SCHED */
-	if (napi_schedule_prep(&rq->napi)) {
-		virtqueue_disable_cb(rq->vq);
-		local_bh_disable();
-		__napi_schedule(&rq->napi);
-		local_bh_enable();
-	}
+	 * won't get another interrupt, so process any outstanding packets now.
+	 * Call local_bh_enable after to trigger softIRQ processing.
+	 */
+	local_bh_disable();
+	virtqueue_napi_schedule(napi, vq);
+	local_bh_enable();
 }
 
 static void refill_work(struct work_struct *work)
@@ -971,7 +984,7 @@ static void refill_work(struct work_struct *work)
 
 		napi_disable(&rq->napi);
 		still_empty = !try_fill_recv(vi, rq, GFP_KERNEL);
-		virtnet_napi_enable(rq);
+		virtnet_napi_enable(rq->vq, &rq->napi);
 
 		/* In theory, this can happen: if we don't get any buffers in
 		 * we will *never* try to fill again.
@@ -1011,21 +1024,13 @@ static int virtnet_poll(struct napi_struct *napi, int budget)
 {
 	struct receive_queue *rq =
 		container_of(napi, struct receive_queue, napi);
-	unsigned int r, received;
+	unsigned int received;
 
 	received = virtnet_receive(rq, budget);
 
 	/* Out of packets? */
-	if (received < budget) {
-		r = virtqueue_enable_cb_prepare(rq->vq);
-		if (napi_complete_done(napi, received)) {
-			if (unlikely(virtqueue_poll(rq->vq, r)) &&
-			    napi_schedule_prep(napi)) {
-				virtqueue_disable_cb(rq->vq);
-				__napi_schedule(napi);
-			}
-		}
-	}
+	if (received < budget)
+		virtqueue_napi_complete(napi, rq->vq, received);
 
 	return received;
 }
@@ -1040,7 +1045,7 @@ static int virtnet_open(struct net_device *dev)
 			/* Make sure we have some buffers: if oom use wq. */
 			if (!try_fill_recv(vi, &vi->rq[i], GFP_KERNEL))
 				schedule_delayed_work(&vi->refill, 0);
-		virtnet_napi_enable(&vi->rq[i]);
+		virtnet_napi_enable(vi->rq[i].vq, &vi->rq[i].napi);
 	}
 
 	return 0;
@@ -1747,7 +1752,7 @@ static int virtnet_restore_up(struct virtio_device *vdev)
 				schedule_delayed_work(&vi->refill, 0);
 
 		for (i = 0; i < vi->max_queue_pairs; i++)
-			virtnet_napi_enable(&vi->rq[i]);
+			virtnet_napi_enable(vi->rq[i].vq, &vi->rq[i].napi);
 	}
 
 	netif_device_attach(vi->dev);
-- 
2.12.2.816.g2cccc81164-goog

[PATCH net-next v3 1/5] virtio-net: napi helper functions

[Willem de Bruijn]

From: Willem de Bruijn <willemb@google.com>

Prepare virtio-net for tx napi by converting existing napi code to
use helper functions. This also deduplicates some logic.

Signed-off-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
---
 drivers/net/virtio_net.c | 65 ++++++++++++++++++++++++++----------------------
 1 file changed, 35 insertions(+), 30 deletions(-)

diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index 666ada6130ab..b9c1df29892c 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -239,6 +239,26 @@ static struct page *get_a_page(struct receive_queue *rq, gfp_t gfp_mask)
 	return p;
 }
 
+static void virtqueue_napi_schedule(struct napi_struct *napi,
+				    struct virtqueue *vq)
+{
+	if (napi_schedule_prep(napi)) {
+		virtqueue_disable_cb(vq);
+		__napi_schedule(napi);
+	}
+}
+
+static void virtqueue_napi_complete(struct napi_struct *napi,
+				    struct virtqueue *vq, int processed)
+{
+	int opaque;
+
+	opaque = virtqueue_enable_cb_prepare(vq);
+	if (napi_complete_done(napi, processed) &&
+	    unlikely(virtqueue_poll(vq, opaque)))
+		virtqueue_napi_schedule(napi, vq);
+}
+
 static void skb_xmit_done(struct virtqueue *vq)
 {
 	struct virtnet_info *vi = vq->vdev->priv;
@@ -936,27 +956,20 @@ static void skb_recv_done(struct virtqueue *rvq)
 	struct virtnet_info *vi = rvq->vdev->priv;
 	struct receive_queue *rq = &vi->rq[vq2rxq(rvq)];
 
-	/* Schedule NAPI, Suppress further interrupts if successful. */
-	if (napi_schedule_prep(&rq->napi)) {
-		virtqueue_disable_cb(rvq);
-		__napi_schedule(&rq->napi);
-	}
+	virtqueue_napi_schedule(&rq->napi, rvq);
 }
 
-static void virtnet_napi_enable(struct receive_queue *rq)
+static void virtnet_napi_enable(struct virtqueue *vq, struct napi_struct *napi)
 {
-	napi_enable(&rq->napi);
+	napi_enable(napi);
 
 	/* If all buffers were filled by other side before we napi_enabled, we
-	 * won't get another interrupt, so process any outstanding packets
-	 * now.  virtnet_poll wants re-enable the queue, so we disable here.
-	 * We synchronize against interrupts via NAPI_STATE_SCHED */
-	if (napi_schedule_prep(&rq->napi)) {
-		virtqueue_disable_cb(rq->vq);
-		local_bh_disable();
-		__napi_schedule(&rq->napi);
-		local_bh_enable();
-	}
+	 * won't get another interrupt, so process any outstanding packets now.
+	 * Call local_bh_enable after to trigger softIRQ processing.
+	 */
+	local_bh_disable();
+	virtqueue_napi_schedule(napi, vq);
+	local_bh_enable();
 }
 
 static void refill_work(struct work_struct *work)
@@ -971,7 +984,7 @@ static void refill_work(struct work_struct *work)
 
 		napi_disable(&rq->napi);
 		still_empty = !try_fill_recv(vi, rq, GFP_KERNEL);
-		virtnet_napi_enable(rq);
+		virtnet_napi_enable(rq->vq, &rq->napi);
 
 		/* In theory, this can happen: if we don't get any buffers in
 		 * we will *never* try to fill again.
@@ -1011,21 +1024,13 @@ static int virtnet_poll(struct napi_struct *napi, int budget)
 {
 	struct receive_queue *rq =
 		container_of(napi, struct receive_queue, napi);
-	unsigned int r, received;
+	unsigned int received;
 
 	received = virtnet_receive(rq, budget);
 
 	/* Out of packets? */
-	if (received < budget) {
-		r = virtqueue_enable_cb_prepare(rq->vq);
-		if (napi_complete_done(napi, received)) {
-			if (unlikely(virtqueue_poll(rq->vq, r)) &&
-			    napi_schedule_prep(napi)) {
-				virtqueue_disable_cb(rq->vq);
-				__napi_schedule(napi);
-			}
-		}
-	}
+	if (received < budget)
+		virtqueue_napi_complete(napi, rq->vq, received);
 
 	return received;
 }
@@ -1040,7 +1045,7 @@ static int virtnet_open(struct net_device *dev)
 			/* Make sure we have some buffers: if oom use wq. */
 			if (!try_fill_recv(vi, &vi->rq[i], GFP_KERNEL))
 				schedule_delayed_work(&vi->refill, 0);
-		virtnet_napi_enable(&vi->rq[i]);
+		virtnet_napi_enable(vi->rq[i].vq, &vi->rq[i].napi);
 	}
 
 	return 0;
@@ -1747,7 +1752,7 @@ static int virtnet_restore_up(struct virtio_device *vdev)
 				schedule_delayed_work(&vi->refill, 0);
 
 		for (i = 0; i < vi->max_queue_pairs; i++)
-			virtnet_napi_enable(&vi->rq[i]);
+			virtnet_napi_enable(vi->rq[i].vq, &vi->rq[i].napi);
 	}
 
 	netif_device_attach(vi->dev);
-- 
2.12.2.816.g2cccc81164-goog

[PATCH net-next v3 2/5] virtio-net: transmit napi

[Willem de Bruijn]

From: Willem de Bruijn <willemb@google.com>

Convert virtio-net to a standard napi tx completion path. This enables
better TCP pacing using TCP small queues and increases single stream
throughput.

The virtio-net driver currently cleans tx descriptors on transmission
of new packets in ndo_start_xmit. Latency depends on new traffic, so
is unbounded. To avoid deadlock when a socket reaches its snd limit,
packets are orphaned on tranmission. This breaks socket backpressure,
including TSQ.

Napi increases the number of interrupts generated compared to the
current model, which keeps interrupts disabled as long as the ring
has enough free descriptors. Keep tx napi optional and disabled for
now. Follow-on patches will reduce the interrupt cost.

Signed-off-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
---
 drivers/net/virtio_net.c | 76 ++++++++++++++++++++++++++++++++++++++++++------
 1 file changed, 67 insertions(+), 9 deletions(-)

diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index b9c1df29892c..356d18481ee4 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -33,9 +33,10 @@
 static int napi_weight = NAPI_POLL_WEIGHT;
 module_param(napi_weight, int, 0444);
 
-static bool csum = true, gso = true;
+static bool csum = true, gso = true, napi_tx;
 module_param(csum, bool, 0444);
 module_param(gso, bool, 0444);
+module_param(napi_tx, bool, 0644);
 
 /* FIXME: MTU in config. */
 #define GOOD_PACKET_LEN (ETH_HLEN + VLAN_HLEN + ETH_DATA_LEN)
@@ -86,6 +87,8 @@ struct send_queue {
 
 	/* Name of the send queue: output.$index */
 	char name[40];
+
+	struct napi_struct napi;
 };
 
 /* Internal representation of a receive virtqueue */
@@ -262,12 +265,16 @@ static void virtqueue_napi_complete(struct napi_struct *napi,
 static void skb_xmit_done(struct virtqueue *vq)
 {
 	struct virtnet_info *vi = vq->vdev->priv;
+	struct napi_struct *napi = &vi->sq[vq2txq(vq)].napi;
 
 	/* Suppress further interrupts. */
 	virtqueue_disable_cb(vq);
 
-	/* We were probably waiting for more output buffers. */
-	netif_wake_subqueue(vi->dev, vq2txq(vq));
+	if (napi->weight)
+		virtqueue_napi_schedule(napi, vq);
+	else
+		/* We were probably waiting for more output buffers. */
+		netif_wake_subqueue(vi->dev, vq2txq(vq));
 }
 
 static unsigned int mergeable_ctx_to_buf_truesize(unsigned long mrg_ctx)
@@ -972,6 +979,24 @@ static void virtnet_napi_enable(struct virtqueue *vq, struct napi_struct *napi)
 	local_bh_enable();
 }
 
+static void virtnet_napi_tx_enable(struct virtnet_info *vi,
+				   struct virtqueue *vq,
+				   struct napi_struct *napi)
+{
+	if (!napi->weight)
+		return;
+
+	/* Tx napi touches cachelines on the cpu handling tx interrupts. Only
+	 * enable the feature if this is likely affine with the transmit path.
+	 */
+	if (!vi->affinity_hint_set) {
+		napi->weight = 0;
+		return;
+	}
+
+	return virtnet_napi_enable(vq, napi);
+}
+
 static void refill_work(struct work_struct *work)
 {
 	struct virtnet_info *vi =
@@ -1046,6 +1071,7 @@ static int virtnet_open(struct net_device *dev)
 			if (!try_fill_recv(vi, &vi->rq[i], GFP_KERNEL))
 				schedule_delayed_work(&vi->refill, 0);
 		virtnet_napi_enable(vi->rq[i].vq, &vi->rq[i].napi);
+		virtnet_napi_tx_enable(vi, vi->sq[i].vq, &vi->sq[i].napi);
 	}
 
 	return 0;
@@ -1081,6 +1107,24 @@ static void free_old_xmit_skbs(struct send_queue *sq)
 	u64_stats_update_end(&stats->tx_syncp);
 }
 
+static int virtnet_poll_tx(struct napi_struct *napi, int budget)
+{
+	struct send_queue *sq = container_of(napi, struct send_queue, napi);
+	struct virtnet_info *vi = sq->vq->vdev->priv;
+	struct netdev_queue *txq = netdev_get_tx_queue(vi->dev, vq2txq(sq->vq));
+
+	__netif_tx_lock(txq, raw_smp_processor_id());
+	free_old_xmit_skbs(sq);
+	__netif_tx_unlock(txq);
+
+	virtqueue_napi_complete(napi, sq->vq, 0);
+
+	if (sq->vq->num_free >= 2 + MAX_SKB_FRAGS)
+		netif_tx_wake_queue(txq);
+
+	return 0;
+}
+
 static int xmit_skb(struct send_queue *sq, struct sk_buff *skb)
 {
 	struct virtio_net_hdr_mrg_rxbuf *hdr;
@@ -1130,6 +1174,7 @@ static netdev_tx_t start_xmit(struct sk_buff *skb, struct net_device *dev)
 	int err;
 	struct netdev_queue *txq = netdev_get_tx_queue(dev, qnum);
 	bool kick = !skb->xmit_more;
+	bool use_napi = sq->napi.weight;
 
 	/* Free up any pending old buffers before queueing new ones. */
 	free_old_xmit_skbs(sq);
@@ -1152,8 +1197,10 @@ static netdev_tx_t start_xmit(struct sk_buff *skb, struct net_device *dev)
 	}
 
 	/* Don't wait up for transmitted skbs to be freed. */
-	skb_orphan(skb);
-	nf_reset(skb);
+	if (!use_napi) {
+		skb_orphan(skb);
+		nf_reset(skb);
+	}
 
 	/* If running out of space, stop queue to avoid getting packets that we
 	 * are then unable to transmit.
@@ -1167,7 +1214,8 @@ static netdev_tx_t start_xmit(struct sk_buff *skb, struct net_device *dev)
 	 */
 	if (sq->vq->num_free < 2+MAX_SKB_FRAGS) {
 		netif_stop_subqueue(dev, qnum);
-		if (unlikely(!virtqueue_enable_cb_delayed(sq->vq))) {
+		if (!use_napi &&
+		    unlikely(!virtqueue_enable_cb_delayed(sq->vq))) {
 			/* More just got used, free them then recheck. */
 			free_old_xmit_skbs(sq);
 			if (sq->vq->num_free >= 2+MAX_SKB_FRAGS) {
@@ -1371,8 +1419,10 @@ static int virtnet_close(struct net_device *dev)
 	/* Make sure refill_work doesn't re-enable napi! */
 	cancel_delayed_work_sync(&vi->refill);
 
-	for (i = 0; i < vi->max_queue_pairs; i++)
+	for (i = 0; i < vi->max_queue_pairs; i++) {
 		napi_disable(&vi->rq[i].napi);
+		napi_disable(&vi->sq[i].napi);
+	}
 
 	return 0;
 }
@@ -1727,8 +1777,10 @@ static void virtnet_freeze_down(struct virtio_device *vdev)
 	cancel_delayed_work_sync(&vi->refill);
 
 	if (netif_running(vi->dev)) {
-		for (i = 0; i < vi->max_queue_pairs; i++)
+		for (i = 0; i < vi->max_queue_pairs; i++) {
 			napi_disable(&vi->rq[i].napi);
+			napi_disable(&vi->sq[i].napi);
+		}
 	}
 }
 
@@ -1751,8 +1803,11 @@ static int virtnet_restore_up(struct virtio_device *vdev)
 			if (!try_fill_recv(vi, &vi->rq[i], GFP_KERNEL))
 				schedule_delayed_work(&vi->refill, 0);
 
-		for (i = 0; i < vi->max_queue_pairs; i++)
+		for (i = 0; i < vi->max_queue_pairs; i++) {
 			virtnet_napi_enable(vi->rq[i].vq, &vi->rq[i].napi);
+			virtnet_napi_tx_enable(vi, vi->sq[i].vq,
+					       &vi->sq[i].napi);
+		}
 	}
 
 	netif_device_attach(vi->dev);
@@ -1957,6 +2012,7 @@ static void virtnet_free_queues(struct virtnet_info *vi)
 	for (i = 0; i < vi->max_queue_pairs; i++) {
 		napi_hash_del(&vi->rq[i].napi);
 		netif_napi_del(&vi->rq[i].napi);
+		netif_napi_del(&vi->sq[i].napi);
 	}
 
 	/* We called napi_hash_del() before netif_napi_del(),
@@ -2142,6 +2198,8 @@ static int virtnet_alloc_queues(struct virtnet_info *vi)
 		vi->rq[i].pages = NULL;
 		netif_napi_add(vi->dev, &vi->rq[i].napi, virtnet_poll,
 			       napi_weight);
+		netif_napi_add(vi->dev, &vi->sq[i].napi, virtnet_poll_tx,
+			       napi_tx ? napi_weight : 0);
 
 		sg_init_table(vi->rq[i].sg, ARRAY_SIZE(vi->rq[i].sg));
 		ewma_pkt_len_init(&vi->rq[i].mrg_avg_pkt_len);
-- 
2.12.2.816.g2cccc81164-goog

[PATCH net-next v3 2/5] virtio-net: transmit napi

[Willem de Bruijn]

From: Willem de Bruijn <willemb@google.com>

Convert virtio-net to a standard napi tx completion path. This enables
better TCP pacing using TCP small queues and increases single stream
throughput.

The virtio-net driver currently cleans tx descriptors on transmission
of new packets in ndo_start_xmit. Latency depends on new traffic, so
is unbounded. To avoid deadlock when a socket reaches its snd limit,
packets are orphaned on tranmission. This breaks socket backpressure,
including TSQ.

Napi increases the number of interrupts generated compared to the
current model, which keeps interrupts disabled as long as the ring
has enough free descriptors. Keep tx napi optional and disabled for
now. Follow-on patches will reduce the interrupt cost.

Signed-off-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
---
 drivers/net/virtio_net.c | 76 ++++++++++++++++++++++++++++++++++++++++++------
 1 file changed, 67 insertions(+), 9 deletions(-)

diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index b9c1df29892c..356d18481ee4 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -33,9 +33,10 @@
 static int napi_weight = NAPI_POLL_WEIGHT;
 module_param(napi_weight, int, 0444);
 
-static bool csum = true, gso = true;
+static bool csum = true, gso = true, napi_tx;
 module_param(csum, bool, 0444);
 module_param(gso, bool, 0444);
+module_param(napi_tx, bool, 0644);
 
 /* FIXME: MTU in config. */
 #define GOOD_PACKET_LEN (ETH_HLEN + VLAN_HLEN + ETH_DATA_LEN)
@@ -86,6 +87,8 @@ struct send_queue {
 
 	/* Name of the send queue: output.$index */
 	char name[40];
+
+	struct napi_struct napi;
 };
 
 /* Internal representation of a receive virtqueue */
@@ -262,12 +265,16 @@ static void virtqueue_napi_complete(struct napi_struct *napi,
 static void skb_xmit_done(struct virtqueue *vq)
 {
 	struct virtnet_info *vi = vq->vdev->priv;
+	struct napi_struct *napi = &vi->sq[vq2txq(vq)].napi;
 
 	/* Suppress further interrupts. */
 	virtqueue_disable_cb(vq);
 
-	/* We were probably waiting for more output buffers. */
-	netif_wake_subqueue(vi->dev, vq2txq(vq));
+	if (napi->weight)
+		virtqueue_napi_schedule(napi, vq);
+	else
+		/* We were probably waiting for more output buffers. */
+		netif_wake_subqueue(vi->dev, vq2txq(vq));
 }
 
 static unsigned int mergeable_ctx_to_buf_truesize(unsigned long mrg_ctx)
@@ -972,6 +979,24 @@ static void virtnet_napi_enable(struct virtqueue *vq, struct napi_struct *napi)
 	local_bh_enable();
 }
 
+static void virtnet_napi_tx_enable(struct virtnet_info *vi,
+				   struct virtqueue *vq,
+				   struct napi_struct *napi)
+{
+	if (!napi->weight)
+		return;
+
+	/* Tx napi touches cachelines on the cpu handling tx interrupts. Only
+	 * enable the feature if this is likely affine with the transmit path.
+	 */
+	if (!vi->affinity_hint_set) {
+		napi->weight = 0;
+		return;
+	}
+
+	return virtnet_napi_enable(vq, napi);
+}
+
 static void refill_work(struct work_struct *work)
 {
 	struct virtnet_info *vi =
@@ -1046,6 +1071,7 @@ static int virtnet_open(struct net_device *dev)
 			if (!try_fill_recv(vi, &vi->rq[i], GFP_KERNEL))
 				schedule_delayed_work(&vi->refill, 0);
 		virtnet_napi_enable(vi->rq[i].vq, &vi->rq[i].napi);
+		virtnet_napi_tx_enable(vi, vi->sq[i].vq, &vi->sq[i].napi);
 	}
 
 	return 0;
@@ -1081,6 +1107,24 @@ static void free_old_xmit_skbs(struct send_queue *sq)
 	u64_stats_update_end(&stats->tx_syncp);
 }
 
+static int virtnet_poll_tx(struct napi_struct *napi, int budget)
+{
+	struct send_queue *sq = container_of(napi, struct send_queue, napi);
+	struct virtnet_info *vi = sq->vq->vdev->priv;
+	struct netdev_queue *txq = netdev_get_tx_queue(vi->dev, vq2txq(sq->vq));
+
+	__netif_tx_lock(txq, raw_smp_processor_id());
+	free_old_xmit_skbs(sq);
+	__netif_tx_unlock(txq);
+
+	virtqueue_napi_complete(napi, sq->vq, 0);
+
+	if (sq->vq->num_free >= 2 + MAX_SKB_FRAGS)
+		netif_tx_wake_queue(txq);
+
+	return 0;
+}
+
 static int xmit_skb(struct send_queue *sq, struct sk_buff *skb)
 {
 	struct virtio_net_hdr_mrg_rxbuf *hdr;
@@ -1130,6 +1174,7 @@ static netdev_tx_t start_xmit(struct sk_buff *skb, struct net_device *dev)
 	int err;
 	struct netdev_queue *txq = netdev_get_tx_queue(dev, qnum);
 	bool kick = !skb->xmit_more;
+	bool use_napi = sq->napi.weight;
 
 	/* Free up any pending old buffers before queueing new ones. */
 	free_old_xmit_skbs(sq);
@@ -1152,8 +1197,10 @@ static netdev_tx_t start_xmit(struct sk_buff *skb, struct net_device *dev)
 	}
 
 	/* Don't wait up for transmitted skbs to be freed. */
-	skb_orphan(skb);
-	nf_reset(skb);
+	if (!use_napi) {
+		skb_orphan(skb);
+		nf_reset(skb);
+	}
 
 	/* If running out of space, stop queue to avoid getting packets that we
 	 * are then unable to transmit.
@@ -1167,7 +1214,8 @@ static netdev_tx_t start_xmit(struct sk_buff *skb, struct net_device *dev)
 	 */
 	if (sq->vq->num_free < 2+MAX_SKB_FRAGS) {
 		netif_stop_subqueue(dev, qnum);
-		if (unlikely(!virtqueue_enable_cb_delayed(sq->vq))) {
+		if (!use_napi &&
+		    unlikely(!virtqueue_enable_cb_delayed(sq->vq))) {
 			/* More just got used, free them then recheck. */
 			free_old_xmit_skbs(sq);
 			if (sq->vq->num_free >= 2+MAX_SKB_FRAGS) {
@@ -1371,8 +1419,10 @@ static int virtnet_close(struct net_device *dev)
 	/* Make sure refill_work doesn't re-enable napi! */
 	cancel_delayed_work_sync(&vi->refill);
 
-	for (i = 0; i < vi->max_queue_pairs; i++)
+	for (i = 0; i < vi->max_queue_pairs; i++) {
 		napi_disable(&vi->rq[i].napi);
+		napi_disable(&vi->sq[i].napi);
+	}
 
 	return 0;
 }
@@ -1727,8 +1777,10 @@ static void virtnet_freeze_down(struct virtio_device *vdev)
 	cancel_delayed_work_sync(&vi->refill);
 
 	if (netif_running(vi->dev)) {
-		for (i = 0; i < vi->max_queue_pairs; i++)
+		for (i = 0; i < vi->max_queue_pairs; i++) {
 			napi_disable(&vi->rq[i].napi);
+			napi_disable(&vi->sq[i].napi);
+		}
 	}
 }
 
@@ -1751,8 +1803,11 @@ static int virtnet_restore_up(struct virtio_device *vdev)
 			if (!try_fill_recv(vi, &vi->rq[i], GFP_KERNEL))
 				schedule_delayed_work(&vi->refill, 0);
 
-		for (i = 0; i < vi->max_queue_pairs; i++)
+		for (i = 0; i < vi->max_queue_pairs; i++) {
 			virtnet_napi_enable(vi->rq[i].vq, &vi->rq[i].napi);
+			virtnet_napi_tx_enable(vi, vi->sq[i].vq,
+					       &vi->sq[i].napi);
+		}
 	}
 
 	netif_device_attach(vi->dev);
@@ -1957,6 +2012,7 @@ static void virtnet_free_queues(struct virtnet_info *vi)
 	for (i = 0; i < vi->max_queue_pairs; i++) {
 		napi_hash_del(&vi->rq[i].napi);
 		netif_napi_del(&vi->rq[i].napi);
+		netif_napi_del(&vi->sq[i].napi);
 	}
 
 	/* We called napi_hash_del() before netif_napi_del(),
@@ -2142,6 +2198,8 @@ static int virtnet_alloc_queues(struct virtnet_info *vi)
 		vi->rq[i].pages = NULL;
 		netif_napi_add(vi->dev, &vi->rq[i].napi, virtnet_poll,
 			       napi_weight);
+		netif_napi_add(vi->dev, &vi->sq[i].napi, virtnet_poll_tx,
+			       napi_tx ? napi_weight : 0);
 
 		sg_init_table(vi->rq[i].sg, ARRAY_SIZE(vi->rq[i].sg));
 		ewma_pkt_len_init(&vi->rq[i].mrg_avg_pkt_len);
-- 
2.12.2.816.g2cccc81164-goog

Re: [PATCH net-next v3 2/5] virtio-net: transmit napi

[Jason Wang]

On 2017年04月25日 01:49, Willem de Bruijn wrote:
> @@ -1371,8 +1419,10 @@ static int virtnet_close(struct net_device *dev)
>   	/* Make sure refill_work doesn't re-enable napi! */
>   	cancel_delayed_work_sync(&vi->refill);
>   
> -	for (i = 0; i < vi->max_queue_pairs; i++)
> +	for (i = 0; i < vi->max_queue_pairs; i++) {
>   		napi_disable(&vi->rq[i].napi);
> +		napi_disable(&vi->sq[i].napi);
> +	}

Looks like this will wait for ever if napi_tx is false because we never 
enable the NAPI so we will wait for NAPI_STATE_SCHED to be cleared.

Thanks
_______________________________________________
Virtualization mailing list
Virtualization@lists.linux-foundation.org
https://lists.linuxfoundation.org/mailman/listinfo/virtualization

Re: [PATCH net-next v3 2/5] virtio-net: transmit napi

[Jason Wang]

On 2017年04月25日 01:49, Willem de Bruijn wrote:
> @@ -1371,8 +1419,10 @@ static int virtnet_close(struct net_device *dev)
>   	/* Make sure refill_work doesn't re-enable napi! */
>   	cancel_delayed_work_sync(&vi->refill);
>   
> -	for (i = 0; i < vi->max_queue_pairs; i++)
> +	for (i = 0; i < vi->max_queue_pairs; i++) {
>   		napi_disable(&vi->rq[i].napi);
> +		napi_disable(&vi->sq[i].napi);
> +	}

Looks like this will wait for ever if napi_tx is false because we never 
enable the NAPI so we will wait for NAPI_STATE_SCHED to be cleared.

Thanks

Re: [PATCH net-next v3 2/5] virtio-net: transmit napi

[Willem de Bruijn]

On Tue, Apr 25, 2017 at 4:36 AM, Jason Wang <jasowang@redhat.com> wrote:
>
>
> On 2017年04月25日 01:49, Willem de Bruijn wrote:
>>
>> @@ -1371,8 +1419,10 @@ static int virtnet_close(struct net_device *dev)
>>         /* Make sure refill_work doesn't re-enable napi! */
>>         cancel_delayed_work_sync(&vi->refill);
>>   -     for (i = 0; i < vi->max_queue_pairs; i++)
>> +       for (i = 0; i < vi->max_queue_pairs; i++) {
>>                 napi_disable(&vi->rq[i].napi);
>> +               napi_disable(&vi->sq[i].napi);
>> +       }
>
>
> Looks like this will wait for ever if napi_tx is false because we never
> enable the NAPI so we will wait for NAPI_STATE_SCHED to be cleared.

Indeed, thanks! I'll send a fix.
_______________________________________________
Virtualization mailing list
Virtualization@lists.linux-foundation.org
https://lists.linuxfoundation.org/mailman/listinfo/virtualization

[PATCH net-next v3 3/5] virtio-net: move free_old_xmit_skbs

[Willem de Bruijn]

From: Willem de Bruijn <willemb@google.com>

An upcoming patch will call free_old_xmit_skbs indirectly from
virtnet_poll. Move the function above this to avoid having to
introduce a forward declaration.

This is a pure move: no code changes.

Signed-off-by: Willem de Bruijn <willemb@google.com>
---
 drivers/net/virtio_net.c | 60 ++++++++++++++++++++++++------------------------
 1 file changed, 30 insertions(+), 30 deletions(-)

diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index 356d18481ee4..4ec79e5d7a86 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -1045,6 +1045,36 @@ static int virtnet_receive(struct receive_queue *rq, int budget)
 	return received;
 }
 
+static void free_old_xmit_skbs(struct send_queue *sq)
+{
+	struct sk_buff *skb;
+	unsigned int len;
+	struct virtnet_info *vi = sq->vq->vdev->priv;
+	struct virtnet_stats *stats = this_cpu_ptr(vi->stats);
+	unsigned int packets = 0;
+	unsigned int bytes = 0;
+
+	while ((skb = virtqueue_get_buf(sq->vq, &len)) != NULL) {
+		pr_debug("Sent skb %p\n", skb);
+
+		bytes += skb->len;
+		packets++;
+
+		dev_kfree_skb_any(skb);
+	}
+
+	/* Avoid overhead when no packets have been processed
+	 * happens when called speculatively from start_xmit.
+	 */
+	if (!packets)
+		return;
+
+	u64_stats_update_begin(&stats->tx_syncp);
+	stats->tx_bytes += bytes;
+	stats->tx_packets += packets;
+	u64_stats_update_end(&stats->tx_syncp);
+}
+
 static int virtnet_poll(struct napi_struct *napi, int budget)
 {
 	struct receive_queue *rq =
@@ -1077,36 +1107,6 @@ static int virtnet_open(struct net_device *dev)
 	return 0;
 }
 
-static void free_old_xmit_skbs(struct send_queue *sq)
-{
-	struct sk_buff *skb;
-	unsigned int len;
-	struct virtnet_info *vi = sq->vq->vdev->priv;
-	struct virtnet_stats *stats = this_cpu_ptr(vi->stats);
-	unsigned int packets = 0;
-	unsigned int bytes = 0;
-
-	while ((skb = virtqueue_get_buf(sq->vq, &len)) != NULL) {
-		pr_debug("Sent skb %p\n", skb);
-
-		bytes += skb->len;
-		packets++;
-
-		dev_kfree_skb_any(skb);
-	}
-
-	/* Avoid overhead when no packets have been processed
-	 * happens when called speculatively from start_xmit.
-	 */
-	if (!packets)
-		return;
-
-	u64_stats_update_begin(&stats->tx_syncp);
-	stats->tx_bytes += bytes;
-	stats->tx_packets += packets;
-	u64_stats_update_end(&stats->tx_syncp);
-}
-
 static int virtnet_poll_tx(struct napi_struct *napi, int budget)
 {
 	struct send_queue *sq = container_of(napi, struct send_queue, napi);
-- 
2.12.2.816.g2cccc81164-goog

[PATCH net-next v3 3/5] virtio-net: move free_old_xmit_skbs

[Willem de Bruijn]

From: Willem de Bruijn <willemb@google.com>

An upcoming patch will call free_old_xmit_skbs indirectly from
virtnet_poll. Move the function above this to avoid having to
introduce a forward declaration.

This is a pure move: no code changes.

Signed-off-by: Willem de Bruijn <willemb@google.com>
---
 drivers/net/virtio_net.c | 60 ++++++++++++++++++++++++------------------------
 1 file changed, 30 insertions(+), 30 deletions(-)

diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index 356d18481ee4..4ec79e5d7a86 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -1045,6 +1045,36 @@ static int virtnet_receive(struct receive_queue *rq, int budget)
 	return received;
 }
 
+static void free_old_xmit_skbs(struct send_queue *sq)
+{
+	struct sk_buff *skb;
+	unsigned int len;
+	struct virtnet_info *vi = sq->vq->vdev->priv;
+	struct virtnet_stats *stats = this_cpu_ptr(vi->stats);
+	unsigned int packets = 0;
+	unsigned int bytes = 0;
+
+	while ((skb = virtqueue_get_buf(sq->vq, &len)) != NULL) {
+		pr_debug("Sent skb %p\n", skb);
+
+		bytes += skb->len;
+		packets++;
+
+		dev_kfree_skb_any(skb);
+	}
+
+	/* Avoid overhead when no packets have been processed
+	 * happens when called speculatively from start_xmit.
+	 */
+	if (!packets)
+		return;
+
+	u64_stats_update_begin(&stats->tx_syncp);
+	stats->tx_bytes += bytes;
+	stats->tx_packets += packets;
+	u64_stats_update_end(&stats->tx_syncp);
+}
+
 static int virtnet_poll(struct napi_struct *napi, int budget)
 {
 	struct receive_queue *rq =
@@ -1077,36 +1107,6 @@ static int virtnet_open(struct net_device *dev)
 	return 0;
 }
 
-static void free_old_xmit_skbs(struct send_queue *sq)
-{
-	struct sk_buff *skb;
-	unsigned int len;
-	struct virtnet_info *vi = sq->vq->vdev->priv;
-	struct virtnet_stats *stats = this_cpu_ptr(vi->stats);
-	unsigned int packets = 0;
-	unsigned int bytes = 0;
-
-	while ((skb = virtqueue_get_buf(sq->vq, &len)) != NULL) {
-		pr_debug("Sent skb %p\n", skb);
-
-		bytes += skb->len;
-		packets++;
-
-		dev_kfree_skb_any(skb);
-	}
-
-	/* Avoid overhead when no packets have been processed
-	 * happens when called speculatively from start_xmit.
-	 */
-	if (!packets)
-		return;
-
-	u64_stats_update_begin(&stats->tx_syncp);
-	stats->tx_bytes += bytes;
-	stats->tx_packets += packets;
-	u64_stats_update_end(&stats->tx_syncp);
-}
-
 static int virtnet_poll_tx(struct napi_struct *napi, int budget)
 {
 	struct send_queue *sq = container_of(napi, struct send_queue, napi);
-- 
2.12.2.816.g2cccc81164-goog

[PATCH net-next v3 4/5] virtio-net: clean tx descriptors from rx napi

[Willem de Bruijn]

From: Willem de Bruijn <willemb@google.com>

Amortize the cost of virtual interrupts by doing both rx and tx work
on reception of a receive interrupt if tx napi is enabled. With
VIRTIO_F_EVENT_IDX, this suppresses most explicit tx completion
interrupts for bidirectional workloads.

Signed-off-by: Willem de Bruijn <willemb@google.com>
---
 drivers/net/virtio_net.c | 21 +++++++++++++++++++++
 1 file changed, 21 insertions(+)

diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index 4ec79e5d7a86..9dd978f34c1f 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -1075,12 +1075,33 @@ static void free_old_xmit_skbs(struct send_queue *sq)
 	u64_stats_update_end(&stats->tx_syncp);
 }
 
+static void virtnet_poll_cleantx(struct receive_queue *rq)
+{
+	struct virtnet_info *vi = rq->vq->vdev->priv;
+	unsigned int index = vq2rxq(rq->vq);
+	struct send_queue *sq = &vi->sq[index];
+	struct netdev_queue *txq = netdev_get_tx_queue(vi->dev, index);
+
+	if (!sq->napi.weight)
+		return;
+
+	if (__netif_tx_trylock(txq)) {
+		free_old_xmit_skbs(sq);
+		__netif_tx_unlock(txq);
+	}
+
+	if (sq->vq->num_free >= 2 + MAX_SKB_FRAGS)
+		netif_tx_wake_queue(txq);
+}
+
 static int virtnet_poll(struct napi_struct *napi, int budget)
 {
 	struct receive_queue *rq =
 		container_of(napi, struct receive_queue, napi);
 	unsigned int received;
 
+	virtnet_poll_cleantx(rq);
+
 	received = virtnet_receive(rq, budget);
 
 	/* Out of packets? */
-- 
2.12.2.816.g2cccc81164-goog

[PATCH net-next v3 4/5] virtio-net: clean tx descriptors from rx napi

[Willem de Bruijn]

From: Willem de Bruijn <willemb@google.com>

Amortize the cost of virtual interrupts by doing both rx and tx work
on reception of a receive interrupt if tx napi is enabled. With
VIRTIO_F_EVENT_IDX, this suppresses most explicit tx completion
interrupts for bidirectional workloads.

Signed-off-by: Willem de Bruijn <willemb@google.com>
---
 drivers/net/virtio_net.c | 21 +++++++++++++++++++++
 1 file changed, 21 insertions(+)

diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index 4ec79e5d7a86..9dd978f34c1f 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -1075,12 +1075,33 @@ static void free_old_xmit_skbs(struct send_queue *sq)
 	u64_stats_update_end(&stats->tx_syncp);
 }
 
+static void virtnet_poll_cleantx(struct receive_queue *rq)
+{
+	struct virtnet_info *vi = rq->vq->vdev->priv;
+	unsigned int index = vq2rxq(rq->vq);
+	struct send_queue *sq = &vi->sq[index];
+	struct netdev_queue *txq = netdev_get_tx_queue(vi->dev, index);
+
+	if (!sq->napi.weight)
+		return;
+
+	if (__netif_tx_trylock(txq)) {
+		free_old_xmit_skbs(sq);
+		__netif_tx_unlock(txq);
+	}
+
+	if (sq->vq->num_free >= 2 + MAX_SKB_FRAGS)
+		netif_tx_wake_queue(txq);
+}
+
 static int virtnet_poll(struct napi_struct *napi, int budget)
 {
 	struct receive_queue *rq =
 		container_of(napi, struct receive_queue, napi);
 	unsigned int received;
 
+	virtnet_poll_cleantx(rq);
+
 	received = virtnet_receive(rq, budget);
 
 	/* Out of packets? */
-- 
2.12.2.816.g2cccc81164-goog

[PATCH net-next v3 5/5] virtio-net: keep tx interrupts disabled unless kick

[Willem de Bruijn]

From: Willem de Bruijn <willemb@google.com>

Tx napi mode increases the rate of transmit interrupts. Suppress some
by masking interrupts while more packets are expected. The interrupts
will be reenabled before the last packet is sent.

This optimization reduces the througput drop with tx napi for
unidirectional flows such as UDP_STREAM that do not benefit from
cleaning tx completions in the the receive napi handler.

Signed-off-by: Willem de Bruijn <willemb@google.com>
---
 drivers/net/virtio_net.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index 9dd978f34c1f..003143835766 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -1200,6 +1200,9 @@ static netdev_tx_t start_xmit(struct sk_buff *skb, struct net_device *dev)
 	/* Free up any pending old buffers before queueing new ones. */
 	free_old_xmit_skbs(sq);
 
+	if (use_napi && kick)
+		virtqueue_enable_cb_delayed(sq->vq);
+
 	/* timestamp packet in software */
 	skb_tx_timestamp(skb);
 
-- 
2.12.2.816.g2cccc81164-goog

[PATCH net-next v3 5/5] virtio-net: keep tx interrupts disabled unless kick

[Willem de Bruijn]

From: Willem de Bruijn <willemb@google.com>

Tx napi mode increases the rate of transmit interrupts. Suppress some
by masking interrupts while more packets are expected. The interrupts
will be reenabled before the last packet is sent.

This optimization reduces the througput drop with tx napi for
unidirectional flows such as UDP_STREAM that do not benefit from
cleaning tx completions in the the receive napi handler.

Signed-off-by: Willem de Bruijn <willemb@google.com>
---
 drivers/net/virtio_net.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index 9dd978f34c1f..003143835766 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -1200,6 +1200,9 @@ static netdev_tx_t start_xmit(struct sk_buff *skb, struct net_device *dev)
 	/* Free up any pending old buffers before queueing new ones. */
 	free_old_xmit_skbs(sq);
 
+	if (use_napi && kick)
+		virtqueue_enable_cb_delayed(sq->vq);
+
 	/* timestamp packet in software */
 	skb_tx_timestamp(skb);
 
-- 
2.12.2.816.g2cccc81164-goog

Re: [PATCH net-next v3 5/5] virtio-net: keep tx interrupts disabled unless kick

[Michael S. Tsirkin]

On Mon, Apr 24, 2017 at 01:49:30PM -0400, Willem de Bruijn wrote:
> From: Willem de Bruijn <willemb@google.com>
> 
> Tx napi mode increases the rate of transmit interrupts. Suppress some
> by masking interrupts while more packets are expected. The interrupts
> will be reenabled before the last packet is sent.
> 
> This optimization reduces the througput drop with tx napi for
> unidirectional flows such as UDP_STREAM that do not benefit from
> cleaning tx completions in the the receive napi handler.
> 
> Signed-off-by: Willem de Bruijn <willemb@google.com>
> ---
>  drivers/net/virtio_net.c | 3 +++
>  1 file changed, 3 insertions(+)
> 
> diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
> index 9dd978f34c1f..003143835766 100644
> --- a/drivers/net/virtio_net.c
> +++ b/drivers/net/virtio_net.c
> @@ -1200,6 +1200,9 @@ static netdev_tx_t start_xmit(struct sk_buff *skb, struct net_device *dev)
>  	/* Free up any pending old buffers before queueing new ones. */
>  	free_old_xmit_skbs(sq);
>  
> +	if (use_napi && kick)
> +		virtqueue_enable_cb_delayed(sq->vq);
> +
>  	/* timestamp packet in software */
>  	skb_tx_timestamp(skb);


I have been poking at this code today and I noticed that is
actually does enable cb where the commit log says masking interrupts.
I think the reason is that with even index previously disable cb
actually did nothing while virtqueue_enable_cb_delayed pushed
the event index out some more.
And this likely explains why it does not work well for packed,
where virtqueue_enable_cb_delayed is same as virtqueue_enable_cb.

Right? Or did I miss something?

> -- 
> 2.12.2.816.g2cccc81164-goog

Re: [PATCH net-next v3 5/5] virtio-net: keep tx interrupts disabled unless kick

[Michael S. Tsirkin]

On Mon, Apr 24, 2017 at 01:49:30PM -0400, Willem de Bruijn wrote:
> From: Willem de Bruijn <willemb@google.com>
> 
> Tx napi mode increases the rate of transmit interrupts. Suppress some
> by masking interrupts while more packets are expected. The interrupts
> will be reenabled before the last packet is sent.
> 
> This optimization reduces the througput drop with tx napi for
> unidirectional flows such as UDP_STREAM that do not benefit from
> cleaning tx completions in the the receive napi handler.
> 
> Signed-off-by: Willem de Bruijn <willemb@google.com>
> ---
>  drivers/net/virtio_net.c | 3 +++
>  1 file changed, 3 insertions(+)
> 
> diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
> index 9dd978f34c1f..003143835766 100644
> --- a/drivers/net/virtio_net.c
> +++ b/drivers/net/virtio_net.c
> @@ -1200,6 +1200,9 @@ static netdev_tx_t start_xmit(struct sk_buff *skb, struct net_device *dev)
>  	/* Free up any pending old buffers before queueing new ones. */
>  	free_old_xmit_skbs(sq);
>  
> +	if (use_napi && kick)
> +		virtqueue_enable_cb_delayed(sq->vq);
> +
>  	/* timestamp packet in software */
>  	skb_tx_timestamp(skb);


I have been poking at this code today and I noticed that is
actually does enable cb where the commit log says masking interrupts.
I think the reason is that with even index previously disable cb
actually did nothing while virtqueue_enable_cb_delayed pushed
the event index out some more.
And this likely explains why it does not work well for packed,
where virtqueue_enable_cb_delayed is same as virtqueue_enable_cb.

Right? Or did I miss something?

> -- 
> 2.12.2.816.g2cccc81164-goog

_______________________________________________
Virtualization mailing list
Virtualization@lists.linux-foundation.org
https://lists.linuxfoundation.org/mailman/listinfo/virtualization

Re: [PATCH net-next v3 5/5] virtio-net: keep tx interrupts disabled unless kick

[Willem de Bruijn]

On Tue, Apr 13, 2021 at 1:06 AM Michael S. Tsirkin <mst@redhat.com> wrote:
>
> On Mon, Apr 24, 2017 at 01:49:30PM -0400, Willem de Bruijn wrote:
> > From: Willem de Bruijn <willemb@google.com>
> >
> > Tx napi mode increases the rate of transmit interrupts. Suppress some
> > by masking interrupts while more packets are expected. The interrupts
> > will be reenabled before the last packet is sent.
> >
> > This optimization reduces the througput drop with tx napi for
> > unidirectional flows such as UDP_STREAM that do not benefit from
> > cleaning tx completions in the the receive napi handler.
> >
> > Signed-off-by: Willem de Bruijn <willemb@google.com>
> > ---
> >  drivers/net/virtio_net.c | 3 +++
> >  1 file changed, 3 insertions(+)
> >
> > diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
> > index 9dd978f34c1f..003143835766 100644
> > --- a/drivers/net/virtio_net.c
> > +++ b/drivers/net/virtio_net.c
> > @@ -1200,6 +1200,9 @@ static netdev_tx_t start_xmit(struct sk_buff *skb, struct net_device *dev)
> >       /* Free up any pending old buffers before queueing new ones. */
> >       free_old_xmit_skbs(sq);
> >
> > +     if (use_napi && kick)
> > +             virtqueue_enable_cb_delayed(sq->vq);
> > +
> >       /* timestamp packet in software */
> >       skb_tx_timestamp(skb);
>
>
> I have been poking at this code today and I noticed that is
> actually does enable cb where the commit log says masking interrupts.
> I think the reason is that with even index previously disable cb
> actually did nothing while virtqueue_enable_cb_delayed pushed
> the event index out some more.
> And this likely explains why it does not work well for packed,
> where virtqueue_enable_cb_delayed is same as virtqueue_enable_cb.
>
> Right? Or did I miss something?

This was definitely based on the split queue with event index handling.

When you say does not work well for packed, you mean that with packed
mode we see the consequences of the race condition when accessing vq
without holding __netif_tx_lock, in a way that I did not notice with
split queue with event index, right?

Thanks for looking into this and proposing fixes for this issue and the
known other spurious tx interrupt issue.

Re: [PATCH net-next v3 5/5] virtio-net: keep tx interrupts disabled unless kick

[Willem de Bruijn]

On Tue, Apr 13, 2021 at 1:06 AM Michael S. Tsirkin <mst@redhat.com> wrote:
>
> On Mon, Apr 24, 2017 at 01:49:30PM -0400, Willem de Bruijn wrote:
> > From: Willem de Bruijn <willemb@google.com>
> >
> > Tx napi mode increases the rate of transmit interrupts. Suppress some
> > by masking interrupts while more packets are expected. The interrupts
> > will be reenabled before the last packet is sent.
> >
> > This optimization reduces the througput drop with tx napi for
> > unidirectional flows such as UDP_STREAM that do not benefit from
> > cleaning tx completions in the the receive napi handler.
> >
> > Signed-off-by: Willem de Bruijn <willemb@google.com>
> > ---
> >  drivers/net/virtio_net.c | 3 +++
> >  1 file changed, 3 insertions(+)
> >
> > diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
> > index 9dd978f34c1f..003143835766 100644
> > --- a/drivers/net/virtio_net.c
> > +++ b/drivers/net/virtio_net.c
> > @@ -1200,6 +1200,9 @@ static netdev_tx_t start_xmit(struct sk_buff *skb, struct net_device *dev)
> >       /* Free up any pending old buffers before queueing new ones. */
> >       free_old_xmit_skbs(sq);
> >
> > +     if (use_napi && kick)
> > +             virtqueue_enable_cb_delayed(sq->vq);
> > +
> >       /* timestamp packet in software */
> >       skb_tx_timestamp(skb);
>
>
> I have been poking at this code today and I noticed that is
> actually does enable cb where the commit log says masking interrupts.
> I think the reason is that with even index previously disable cb
> actually did nothing while virtqueue_enable_cb_delayed pushed
> the event index out some more.
> And this likely explains why it does not work well for packed,
> where virtqueue_enable_cb_delayed is same as virtqueue_enable_cb.
>
> Right? Or did I miss something?

This was definitely based on the split queue with event index handling.

When you say does not work well for packed, you mean that with packed
mode we see the consequences of the race condition when accessing vq
without holding __netif_tx_lock, in a way that I did not notice with
split queue with event index, right?

Thanks for looking into this and proposing fixes for this issue and the
known other spurious tx interrupt issue.
_______________________________________________
Virtualization mailing list
Virtualization@lists.linux-foundation.org
https://lists.linuxfoundation.org/mailman/listinfo/virtualization

Re: [PATCH net-next v3 5/5] virtio-net: keep tx interrupts disabled unless kick

[Michael S. Tsirkin]

On Tue, Apr 13, 2021 at 10:27:16AM -0400, Willem de Bruijn wrote:
> On Tue, Apr 13, 2021 at 1:06 AM Michael S. Tsirkin <mst@redhat.com> wrote:
> >
> > On Mon, Apr 24, 2017 at 01:49:30PM -0400, Willem de Bruijn wrote:
> > > From: Willem de Bruijn <willemb@google.com>
> > >
> > > Tx napi mode increases the rate of transmit interrupts. Suppress some
> > > by masking interrupts while more packets are expected. The interrupts
> > > will be reenabled before the last packet is sent.
> > >
> > > This optimization reduces the througput drop with tx napi for
> > > unidirectional flows such as UDP_STREAM that do not benefit from
> > > cleaning tx completions in the the receive napi handler.
> > >
> > > Signed-off-by: Willem de Bruijn <willemb@google.com>
> > > ---
> > >  drivers/net/virtio_net.c | 3 +++
> > >  1 file changed, 3 insertions(+)
> > >
> > > diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
> > > index 9dd978f34c1f..003143835766 100644
> > > --- a/drivers/net/virtio_net.c
> > > +++ b/drivers/net/virtio_net.c
> > > @@ -1200,6 +1200,9 @@ static netdev_tx_t start_xmit(struct sk_buff *skb, struct net_device *dev)
> > >       /* Free up any pending old buffers before queueing new ones. */
> > >       free_old_xmit_skbs(sq);
> > >
> > > +     if (use_napi && kick)
> > > +             virtqueue_enable_cb_delayed(sq->vq);
> > > +
> > >       /* timestamp packet in software */
> > >       skb_tx_timestamp(skb);
> >
> >
> > I have been poking at this code today and I noticed that is
> > actually does enable cb where the commit log says masking interrupts.
> > I think the reason is that with even index previously disable cb
> > actually did nothing while virtqueue_enable_cb_delayed pushed
> > the event index out some more.
> > And this likely explains why it does not work well for packed,
> > where virtqueue_enable_cb_delayed is same as virtqueue_enable_cb.
> >
> > Right? Or did I miss something?
> 
> This was definitely based on the split queue with event index handling.
> 
> When you say does not work well for packed, you mean that with packed
> mode we see the consequences of the race condition when accessing vq
> without holding __netif_tx_lock, in a way that I did not notice with
> split queue with event index, right?

I mean curretly packed does not seem to show same performance gains as
a micro-benchmark. Could be due to enabling interrupts more aggressively
there.

> Thanks for looking into this and proposing fixes for this issue and the
> known other spurious tx interrupt issue.

Re: [PATCH net-next v3 5/5] virtio-net: keep tx interrupts disabled unless kick

[Michael S. Tsirkin]

On Tue, Apr 13, 2021 at 10:27:16AM -0400, Willem de Bruijn wrote:
> On Tue, Apr 13, 2021 at 1:06 AM Michael S. Tsirkin <mst@redhat.com> wrote:
> >
> > On Mon, Apr 24, 2017 at 01:49:30PM -0400, Willem de Bruijn wrote:
> > > From: Willem de Bruijn <willemb@google.com>
> > >
> > > Tx napi mode increases the rate of transmit interrupts. Suppress some
> > > by masking interrupts while more packets are expected. The interrupts
> > > will be reenabled before the last packet is sent.
> > >
> > > This optimization reduces the througput drop with tx napi for
> > > unidirectional flows such as UDP_STREAM that do not benefit from
> > > cleaning tx completions in the the receive napi handler.
> > >
> > > Signed-off-by: Willem de Bruijn <willemb@google.com>
> > > ---
> > >  drivers/net/virtio_net.c | 3 +++
> > >  1 file changed, 3 insertions(+)
> > >
> > > diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
> > > index 9dd978f34c1f..003143835766 100644
> > > --- a/drivers/net/virtio_net.c
> > > +++ b/drivers/net/virtio_net.c
> > > @@ -1200,6 +1200,9 @@ static netdev_tx_t start_xmit(struct sk_buff *skb, struct net_device *dev)
> > >       /* Free up any pending old buffers before queueing new ones. */
> > >       free_old_xmit_skbs(sq);
> > >
> > > +     if (use_napi && kick)
> > > +             virtqueue_enable_cb_delayed(sq->vq);
> > > +
> > >       /* timestamp packet in software */
> > >       skb_tx_timestamp(skb);
> >
> >
> > I have been poking at this code today and I noticed that is
> > actually does enable cb where the commit log says masking interrupts.
> > I think the reason is that with even index previously disable cb
> > actually did nothing while virtqueue_enable_cb_delayed pushed
> > the event index out some more.
> > And this likely explains why it does not work well for packed,
> > where virtqueue_enable_cb_delayed is same as virtqueue_enable_cb.
> >
> > Right? Or did I miss something?
> 
> This was definitely based on the split queue with event index handling.
> 
> When you say does not work well for packed, you mean that with packed
> mode we see the consequences of the race condition when accessing vq
> without holding __netif_tx_lock, in a way that I did not notice with
> split queue with event index, right?

I mean curretly packed does not seem to show same performance gains as
a micro-benchmark. Could be due to enabling interrupts more aggressively
there.

> Thanks for looking into this and proposing fixes for this issue and the
> known other spurious tx interrupt issue.

_______________________________________________
Virtualization mailing list
Virtualization@lists.linux-foundation.org
https://lists.linuxfoundation.org/mailman/listinfo/virtualization

Re: [PATCH net-next v3 0/5] virtio-net tx napi

[Michael S. Tsirkin]

On Mon, Apr 24, 2017 at 01:49:25PM -0400, Willem de Bruijn wrote:
> From: Willem de Bruijn <willemb@google.com>
> 
> Add napi for virtio-net transmit completion processing.


Acked-by: Michael S. Tsirkin <mst@redhat.com>

> Changes:
>   v2 -> v3:
>     - convert __netif_tx_trylock to __netif_tx_lock on tx napi poll
>           ensure that the handler always cleans, to avoid deadlock
>     - unconditionally clean in start_xmit
>           avoid adding an unnecessary "if (use_napi)" branch
>     - remove virtqueue_disable_cb in patch 5/5
>           a noop in the common event_idx based loop
>     - document affinity_hint_set constraint
> 
>   v1 -> v2:
>     - disable by default
>     - disable unless affinity_hint_set
>           because cache misses add up to a third higher cycle cost,
> 	  e.g., in TCP_RR tests. This is not limited to the patch
> 	  that enables tx completion cleaning in rx napi.
>     - use trylock to avoid contention between tx and rx napi
>     - keep interrupts masked during xmit_more (new patch 5/5)
>           this improves cycles especially for multi UDP_STREAM, which
> 	  does not benefit from cleaning tx completions on rx napi.
>     - move free_old_xmit_skbs (new patch 3/5)
>           to avoid forward declaration
> 
>     not changed:
>     - deduplicate virnet_poll_tx and virtnet_poll_txclean
>           they look similar, but have differ too much to make it
> 	  worthwhile.
>     - delay netif_wake_subqueue for more than 2 + MAX_SKB_FRAGS
>           evaluated, but made no difference
>     - patch 1/5
> 
>   RFC -> v1:
>     - dropped vhost interrupt moderation patch:
>           not needed and likely expensive at light load
>     - remove tx napi weight
>         - always clean all tx completions
>         - use boolean to toggle tx-napi, instead
>     - only clean tx in rx if tx-napi is enabled
>         - then clean tx before rx
>     - fix: add missing braces in virtnet_freeze_down
>     - testing: add 4KB TCP_RR + UDP test results
> 
> Based on previous patchsets by Jason Wang:
> 
>   [RFC V7 PATCH 0/7] enable tx interrupts for virtio-net
>   http://lkml.iu.edu/hypermail/linux/kernel/1505.3/00245.html
> 
> 
> Before commit b0c39dbdc204 ("virtio_net: don't free buffers in xmit
> ring") the virtio-net driver would free transmitted packets on
> transmission of new packets in ndo_start_xmit and, to catch the edge
> case when no new packet is sent, also in a timer at 10HZ.
> 
> A timer can cause long stalls. VIRTIO_F_NOTIFY_ON_EMPTY avoids stalls
> due to low free descriptor count. It does not address a stalls due to
> low socket SO_SNDBUF. Increasing timer frequency decreases that stall
> time, but increases interrupt rate and, thus, cycle count.
> 
> Currently, with no timer, packets are freed only at ndo_start_xmit.
> Latency of consume_skb is now unbounded. To avoid a deadlock if a sock
> reaches SO_SNDBUF, packets are orphaned on tx. This breaks TCP small
> queues.
> 
> Reenable TCP small queues by removing the orphan. Instead of using a
> timer, convert the driver to regular tx napi. This does not have the
> unresolved stall issue and does not have any frequency to tune.
> 
> By keeping interrupts enabled by default, napi increases tx
> interrupt rate. VIRTIO_F_EVENT_IDX avoids sending an interrupt if
> one is already unacknowledged, so makes this more feasible today.
> Combine that with an optimization that brings interrupt rate
> back in line with the existing version for most workloads:
> 
> Tx completion cleaning on rx interrupts elides most explicit tx
> interrupts by relying on the fact that many rx interrupts fire.
> 
> Tested by running {1, 10, 100} {TCP, UDP} STREAM, RR, 4K_RR benchmarks
> from a guest to a server on the host, on an x86_64 Haswell. The guest
> runs 4 vCPUs pinned to 4 cores. vhost and the test server are
> pinned to a core each.
> 
> All results are the median of 5 runs, with variance well < 10%.
> Used neper (github.com/google/neper) as test process.
> 
> Napi increases single stream throughput, but increases cycle cost.
> The optimizations bring this down. The previous patchset saw a
> regression with UDP_STREAM, which does not benefit from cleaning tx
> interrupts in rx napi. This regression is now gone for 10x, 100x.
> Remaining difference is higher 1x TCP_STREAM, lower 1x UDP_STREAM.
> 
> The latest results are with process, rx napi and tx napi affine to
> the same core. All numbers are lower than the previous patchset.
> 
> 
>              upstream     napi
> TCP_STREAM:
> 1x:
>   Mbps          27816    39805
>   Gcycles         274      285
> 
> 10x:
>   Mbps          42947    42531
>   Gcycles         300      296
> 
> 100x:
>   Mbps          31830    28042
>   Gcycles         279      269
> 
> TCP_RR Latency (us):
> 1x:
>   p50              21       21
>   p99              27       27
>   Gcycles         180      167
> 
> 10x:
>   p50              40       39
>   p99              52       52
>   Gcycles         214      211
> 
> 100x:
>   p50             281      241
>   p99             411      337
>   Gcycles         218      226
> 
> TCP_RR 4K:
> 1x:
>   p50              28       29
>   p99              34       36
>   Gcycles         177      167
> 
> 10x:
>   p50              70       71
>   p99              85      134
>   Gcycles         213      214
> 
> 100x:
>   p50             442      611
>   p99             802      785
>   Gcycles         237      216
> 
> UDP_STREAM:
> 1x:
>   Mbps          29468    26800
>   Gcycles         284      293
> 
> 10x:
>   Mbps          29891    29978
>   Gcycles         285      312
> 
> 100x:
>   Mbps          30269    30304
>   Gcycles         318      316
> 
> UDP_RR:
> 1x:
>   p50              19       19
>   p99              23       23
>   Gcycles         180      173
> 
> 10x:
>   p50              35       40
>   p99              54       64
>   Gcycles         245      237
> 
> 100x:
>   p50             234      286
>   p99             484      473
>   Gcycles         224      214
> 
> Note that GSO is enabled, so 4K RR still translates to one packet
> per request.
> 
> Lower throughput at 100x vs 10x can be (at least in part)
> explained by looking at bytes per packet sent (nstat). It likely
> also explains the lower throughput of 1x for some variants.
> 
> upstream:
> 
>  N=1   bytes/pkt=16581
>  N=10  bytes/pkt=61513
>  N=100 bytes/pkt=51558
> 
> at_rx:
> 
>  N=1   bytes/pkt=65204
>  N=10  bytes/pkt=65148
>  N=100 bytes/pkt=56840
> 
> Willem de Bruijn (5):
>   virtio-net: napi helper functions
>   virtio-net: transmit napi
>   virtio-net: move free_old_xmit_skbs
>   virtio-net: clean tx descriptors from rx napi
>   virtio-net: keep tx interrupts disabled unless kick
> 
>  drivers/net/virtio_net.c | 193 ++++++++++++++++++++++++++++++++---------------
>  1 file changed, 132 insertions(+), 61 deletions(-)
> 
> -- 
> 2.12.2.816.g2cccc81164-goog

Re: [PATCH net-next v3 0/5] virtio-net tx napi

[Michael S. Tsirkin]

On Mon, Apr 24, 2017 at 01:49:25PM -0400, Willem de Bruijn wrote:
> From: Willem de Bruijn <willemb@google.com>
> 
> Add napi for virtio-net transmit completion processing.


Acked-by: Michael S. Tsirkin <mst@redhat.com>

> Changes:
>   v2 -> v3:
>     - convert __netif_tx_trylock to __netif_tx_lock on tx napi poll
>           ensure that the handler always cleans, to avoid deadlock
>     - unconditionally clean in start_xmit
>           avoid adding an unnecessary "if (use_napi)" branch
>     - remove virtqueue_disable_cb in patch 5/5
>           a noop in the common event_idx based loop
>     - document affinity_hint_set constraint
> 
>   v1 -> v2:
>     - disable by default
>     - disable unless affinity_hint_set
>           because cache misses add up to a third higher cycle cost,
> 	  e.g., in TCP_RR tests. This is not limited to the patch
> 	  that enables tx completion cleaning in rx napi.
>     - use trylock to avoid contention between tx and rx napi
>     - keep interrupts masked during xmit_more (new patch 5/5)
>           this improves cycles especially for multi UDP_STREAM, which
> 	  does not benefit from cleaning tx completions on rx napi.
>     - move free_old_xmit_skbs (new patch 3/5)
>           to avoid forward declaration
> 
>     not changed:
>     - deduplicate virnet_poll_tx and virtnet_poll_txclean
>           they look similar, but have differ too much to make it
> 	  worthwhile.
>     - delay netif_wake_subqueue for more than 2 + MAX_SKB_FRAGS
>           evaluated, but made no difference
>     - patch 1/5
> 
>   RFC -> v1:
>     - dropped vhost interrupt moderation patch:
>           not needed and likely expensive at light load
>     - remove tx napi weight
>         - always clean all tx completions
>         - use boolean to toggle tx-napi, instead
>     - only clean tx in rx if tx-napi is enabled
>         - then clean tx before rx
>     - fix: add missing braces in virtnet_freeze_down
>     - testing: add 4KB TCP_RR + UDP test results
> 
> Based on previous patchsets by Jason Wang:
> 
>   [RFC V7 PATCH 0/7] enable tx interrupts for virtio-net
>   http://lkml.iu.edu/hypermail/linux/kernel/1505.3/00245.html
> 
> 
> Before commit b0c39dbdc204 ("virtio_net: don't free buffers in xmit
> ring") the virtio-net driver would free transmitted packets on
> transmission of new packets in ndo_start_xmit and, to catch the edge
> case when no new packet is sent, also in a timer at 10HZ.
> 
> A timer can cause long stalls. VIRTIO_F_NOTIFY_ON_EMPTY avoids stalls
> due to low free descriptor count. It does not address a stalls due to
> low socket SO_SNDBUF. Increasing timer frequency decreases that stall
> time, but increases interrupt rate and, thus, cycle count.
> 
> Currently, with no timer, packets are freed only at ndo_start_xmit.
> Latency of consume_skb is now unbounded. To avoid a deadlock if a sock
> reaches SO_SNDBUF, packets are orphaned on tx. This breaks TCP small
> queues.
> 
> Reenable TCP small queues by removing the orphan. Instead of using a
> timer, convert the driver to regular tx napi. This does not have the
> unresolved stall issue and does not have any frequency to tune.
> 
> By keeping interrupts enabled by default, napi increases tx
> interrupt rate. VIRTIO_F_EVENT_IDX avoids sending an interrupt if
> one is already unacknowledged, so makes this more feasible today.
> Combine that with an optimization that brings interrupt rate
> back in line with the existing version for most workloads:
> 
> Tx completion cleaning on rx interrupts elides most explicit tx
> interrupts by relying on the fact that many rx interrupts fire.
> 
> Tested by running {1, 10, 100} {TCP, UDP} STREAM, RR, 4K_RR benchmarks
> from a guest to a server on the host, on an x86_64 Haswell. The guest
> runs 4 vCPUs pinned to 4 cores. vhost and the test server are
> pinned to a core each.
> 
> All results are the median of 5 runs, with variance well < 10%.
> Used neper (github.com/google/neper) as test process.
> 
> Napi increases single stream throughput, but increases cycle cost.
> The optimizations bring this down. The previous patchset saw a
> regression with UDP_STREAM, which does not benefit from cleaning tx
> interrupts in rx napi. This regression is now gone for 10x, 100x.
> Remaining difference is higher 1x TCP_STREAM, lower 1x UDP_STREAM.
> 
> The latest results are with process, rx napi and tx napi affine to
> the same core. All numbers are lower than the previous patchset.
> 
> 
>              upstream     napi
> TCP_STREAM:
> 1x:
>   Mbps          27816    39805
>   Gcycles         274      285
> 
> 10x:
>   Mbps          42947    42531
>   Gcycles         300      296
> 
> 100x:
>   Mbps          31830    28042
>   Gcycles         279      269
> 
> TCP_RR Latency (us):
> 1x:
>   p50              21       21
>   p99              27       27
>   Gcycles         180      167
> 
> 10x:
>   p50              40       39
>   p99              52       52
>   Gcycles         214      211
> 
> 100x:
>   p50             281      241
>   p99             411      337
>   Gcycles         218      226
> 
> TCP_RR 4K:
> 1x:
>   p50              28       29
>   p99              34       36
>   Gcycles         177      167
> 
> 10x:
>   p50              70       71
>   p99              85      134
>   Gcycles         213      214
> 
> 100x:
>   p50             442      611
>   p99             802      785
>   Gcycles         237      216
> 
> UDP_STREAM:
> 1x:
>   Mbps          29468    26800
>   Gcycles         284      293
> 
> 10x:
>   Mbps          29891    29978
>   Gcycles         285      312
> 
> 100x:
>   Mbps          30269    30304
>   Gcycles         318      316
> 
> UDP_RR:
> 1x:
>   p50              19       19
>   p99              23       23
>   Gcycles         180      173
> 
> 10x:
>   p50              35       40
>   p99              54       64
>   Gcycles         245      237
> 
> 100x:
>   p50             234      286
>   p99             484      473
>   Gcycles         224      214
> 
> Note that GSO is enabled, so 4K RR still translates to one packet
> per request.
> 
> Lower throughput at 100x vs 10x can be (at least in part)
> explained by looking at bytes per packet sent (nstat). It likely
> also explains the lower throughput of 1x for some variants.
> 
> upstream:
> 
>  N=1   bytes/pkt=16581
>  N=10  bytes/pkt=61513
>  N=100 bytes/pkt=51558
> 
> at_rx:
> 
>  N=1   bytes/pkt=65204
>  N=10  bytes/pkt=65148
>  N=100 bytes/pkt=56840
> 
> Willem de Bruijn (5):
>   virtio-net: napi helper functions
>   virtio-net: transmit napi
>   virtio-net: move free_old_xmit_skbs
>   virtio-net: clean tx descriptors from rx napi
>   virtio-net: keep tx interrupts disabled unless kick
> 
>  drivers/net/virtio_net.c | 193 ++++++++++++++++++++++++++++++++---------------
>  1 file changed, 132 insertions(+), 61 deletions(-)
> 
> -- 
> 2.12.2.816.g2cccc81164-goog

Re: [PATCH net-next v3 0/5] virtio-net tx napi

[David Miller]

From: Willem de Bruijn <willemdebruijn.kernel@gmail.com>
Date: Mon, 24 Apr 2017 13:49:25 -0400

> Add napi for virtio-net transmit completion processing.

Series applied, thanks.