* [PATCH v3] virtio-net: lower min ring num_free for efficiency
@ 2019-08-20 2:51 ? jiang
2019-09-02 15:01 ` Michael S. Tsirkin
2019-09-02 15:01 ` Michael S. Tsirkin
0 siblings, 2 replies; 3+ messages in thread
From: ? jiang @ 2019-08-20 2:51 UTC (permalink / raw)
To: mst@redhat.com
Cc: xdp-newbies@vger.kernel.org, songliubraving@fb.com,
jakub.kicinski@netronome.com, hawk@kernel.org,
daniel@iogearbox.net, jiangran.jr@alibaba-inc.com,
netdev@vger.kernel.org, john.fastabend@gmail.com, ast@kernel.org,
kafai@fb.com, ? jiang, yhs@fb.com, bpf@vger.kernel.org,
virtualization@lists.linux-foundation.org, davem@davemloft.net,
linux-kernel@vger.kernel.org
This change lowers ring buffer reclaim threshold from 1/2*queue to budget
for better performance. According to our test with qemu + dpdk, packet
dropping happens when the guest is not able to provide free buffer in
avail ring timely with default 1/2*queue. The value in the patch has been
tested and does show better performance.
Test setup: iperf3 to generate packets to guest (total 30mins, pps 400k, UDP)
avg packets drop before: 2842
avg packets drop after: 360(-87.3%)
Further, current code suffers from a starvation problem: the amount of
work done by try_fill_recv is not bounded by the budget parameter, thus
(with large queues) once in a while userspace gets blocked for a long
time while queue is being refilled. Trigger refills earlier to make sure
the amount of work to do is limited.
Signed-off-by: jiangkidd <jiangkidd@hotmail.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
---
drivers/net/virtio_net.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index 0d4115c9e20b..bc08be7925eb 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -1331,7 +1331,7 @@ static int virtnet_receive(struct receive_queue *rq, int budget,
}
}
- if (rq->vq->num_free > virtqueue_get_vring_size(rq->vq) / 2) {
+ if (rq->vq->num_free > min((unsigned int)budget, virtqueue_get_vring_size(rq->vq)) / 2) {
if (!try_fill_recv(vi, rq, GFP_ATOMIC))
schedule_delayed_work(&vi->refill, 0);
}
--
2.11.0
^ permalink raw reply related [flat|nested] 3+ messages in thread* Re: [PATCH v3] virtio-net: lower min ring num_free for efficiency
2019-08-20 2:51 [PATCH v3] virtio-net: lower min ring num_free for efficiency ? jiang
@ 2019-09-02 15:01 ` Michael S. Tsirkin
2019-09-02 15:01 ` Michael S. Tsirkin
1 sibling, 0 replies; 3+ messages in thread
From: Michael S. Tsirkin @ 2019-09-02 15:01 UTC (permalink / raw)
To: ? jiang
Cc: xdp-newbies@vger.kernel.org, songliubraving@fb.com,
jakub.kicinski@netronome.com, hawk@kernel.org,
daniel@iogearbox.net, jiangran.jr@alibaba-inc.com,
netdev@vger.kernel.org, john.fastabend@gmail.com, ast@kernel.org,
kafai@fb.com, yhs@fb.com, bpf@vger.kernel.org,
virtualization@lists.linux-foundation.org, davem@davemloft.net,
linux-kernel@vger.kernel.org
On Tue, Aug 20, 2019 at 02:51:23AM +0000, ? jiang wrote:
> This change lowers ring buffer reclaim threshold from 1/2*queue to budget
> for better performance. According to our test with qemu + dpdk, packet
> dropping happens when the guest is not able to provide free buffer in
> avail ring timely with default 1/2*queue. The value in the patch has been
> tested and does show better performance.
>
> Test setup: iperf3 to generate packets to guest (total 30mins, pps 400k, UDP)
> avg packets drop before: 2842
> avg packets drop after: 360(-87.3%)
>
> Further, current code suffers from a starvation problem: the amount of
> work done by try_fill_recv is not bounded by the budget parameter, thus
> (with large queues) once in a while userspace gets blocked for a long
> time while queue is being refilled. Trigger refills earlier to make sure
> the amount of work to do is limited.
>
> Signed-off-by: jiangkidd <jiangkidd@hotmail.com>
> Acked-by: Jason Wang <jasowang@redhat.com>
> Acked-by: Michael S. Tsirkin <mst@redhat.com>
Dave, could you merge this please?
Either net or net-next at your discretion.
> ---
> drivers/net/virtio_net.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
> index 0d4115c9e20b..bc08be7925eb 100644
> --- a/drivers/net/virtio_net.c
> +++ b/drivers/net/virtio_net.c
> @@ -1331,7 +1331,7 @@ static int virtnet_receive(struct receive_queue *rq, int budget,
> }
> }
>
> - if (rq->vq->num_free > virtqueue_get_vring_size(rq->vq) / 2) {
> + if (rq->vq->num_free > min((unsigned int)budget, virtqueue_get_vring_size(rq->vq)) / 2) {
> if (!try_fill_recv(vi, rq, GFP_ATOMIC))
> schedule_delayed_work(&vi->refill, 0);
> }
> --
> 2.11.0
^ permalink raw reply [flat|nested] 3+ messages in thread* Re: [PATCH v3] virtio-net: lower min ring num_free for efficiency
2019-08-20 2:51 [PATCH v3] virtio-net: lower min ring num_free for efficiency ? jiang
2019-09-02 15:01 ` Michael S. Tsirkin
@ 2019-09-02 15:01 ` Michael S. Tsirkin
1 sibling, 0 replies; 3+ messages in thread
From: Michael S. Tsirkin @ 2019-09-02 15:01 UTC (permalink / raw)
To: ? jiang
Cc: jasowang@redhat.com, davem@davemloft.net, ast@kernel.org,
daniel@iogearbox.net, jakub.kicinski@netronome.com,
hawk@kernel.org, john.fastabend@gmail.com, kafai@fb.com,
songliubraving@fb.com, yhs@fb.com,
virtualization@lists.linux-foundation.org, netdev@vger.kernel.org,
linux-kernel@vger.kernel.org, xdp-newbies@vger.kernel.org,
bpf@vger.kernel.org, jiangran.jr@alibaba-inc.com
On Tue, Aug 20, 2019 at 02:51:23AM +0000, ? jiang wrote:
> This change lowers ring buffer reclaim threshold from 1/2*queue to budget
> for better performance. According to our test with qemu + dpdk, packet
> dropping happens when the guest is not able to provide free buffer in
> avail ring timely with default 1/2*queue. The value in the patch has been
> tested and does show better performance.
>
> Test setup: iperf3 to generate packets to guest (total 30mins, pps 400k, UDP)
> avg packets drop before: 2842
> avg packets drop after: 360(-87.3%)
>
> Further, current code suffers from a starvation problem: the amount of
> work done by try_fill_recv is not bounded by the budget parameter, thus
> (with large queues) once in a while userspace gets blocked for a long
> time while queue is being refilled. Trigger refills earlier to make sure
> the amount of work to do is limited.
>
> Signed-off-by: jiangkidd <jiangkidd@hotmail.com>
> Acked-by: Jason Wang <jasowang@redhat.com>
> Acked-by: Michael S. Tsirkin <mst@redhat.com>
Dave, could you merge this please?
Either net or net-next at your discretion.
> ---
> drivers/net/virtio_net.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
> index 0d4115c9e20b..bc08be7925eb 100644
> --- a/drivers/net/virtio_net.c
> +++ b/drivers/net/virtio_net.c
> @@ -1331,7 +1331,7 @@ static int virtnet_receive(struct receive_queue *rq, int budget,
> }
> }
>
> - if (rq->vq->num_free > virtqueue_get_vring_size(rq->vq) / 2) {
> + if (rq->vq->num_free > min((unsigned int)budget, virtqueue_get_vring_size(rq->vq)) / 2) {
> if (!try_fill_recv(vi, rq, GFP_ATOMIC))
> schedule_delayed_work(&vi->refill, 0);
> }
> --
> 2.11.0
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2019-09-02 15:01 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2019-08-20 2:51 [PATCH v3] virtio-net: lower min ring num_free for efficiency ? jiang
2019-09-02 15:01 ` Michael S. Tsirkin
2019-09-02 15:01 ` Michael S. Tsirkin
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.