On Tue, Apr 30, 2024 at 03:35:09PM -0400, Darius Rad wrote:
> The transmit queue is stopped when the number of free queue entries is less
> than 2+MAX_SKB_FRAGS, in start_xmit(). If the queue length (QUEUE_NUM_MAX)
> is less than then this, transmission will immediately trigger a netdev
> watchdog timeout. Report this condition earlier and more directly.
>
> Signed-off-by: Darius Rad <[email protected]>
> ---
> 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 115c3c5414f2..72ee8473b61c 100644
> --- a/drivers/net/virtio_net.c
> +++ b/drivers/net/virtio_net.c
> @@ -4917,6 +4917,9 @@ static int virtnet_probe(struct virtio_device *vdev)
> set_bit(guest_offloads[i], &vi->guest_offloads);
> vi->guest_offloads_capable = vi->guest_offloads;
>
> + if (virtqueue_get_vring_size(vi->sq->vq) < 2 + MAX_SKB_FRAGS)
> + netdev_warn_once(dev, "not enough queue entries, expect xmit timeout\n");
> +
How about actually fixing it though? E.g. by linearizing...
It also bothers me that there's practically
/proc/sys/net/core/max_skb_frags
and if that's low then things could actually work.
Finally, while originally it was just 17 typically, now it's
configurable. So it's possible that you change the config to make big
tcp work better and device stops working while it worked fine
previously.
> pr_debug("virtnet: registered device %s with %d RX and TX vq's\n",
> dev->name, max_queue_pairs);
>
> --
> 2.39.2
On Wed, May 1, 2024 at 4:07 AM Michael S. Tsirkin <[email protected]> wrote:
>
> On Tue, Apr 30, 2024 at 03:35:09PM -0400, Darius Rad wrote:
> > The transmit queue is stopped when the number of free queue entries is less
> > than 2+MAX_SKB_FRAGS, in start_xmit(). If the queue length (QUEUE_NUM_MAX)
> > is less than then this, transmission will immediately trigger a netdev
> > watchdog timeout. Report this condition earlier and more directly.
> >
> > Signed-off-by: Darius Rad <[email protected]>
> > ---
> > 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 115c3c5414f2..72ee8473b61c 100644
> > --- a/drivers/net/virtio_net.c
> > +++ b/drivers/net/virtio_net.c
> > @@ -4917,6 +4917,9 @@ static int virtnet_probe(struct virtio_device *vdev)
> > set_bit(guest_offloads[i], &vi->guest_offloads);
> > vi->guest_offloads_capable = vi->guest_offloads;
> >
> > + if (virtqueue_get_vring_size(vi->sq->vq) < 2 + MAX_SKB_FRAGS)
> > + netdev_warn_once(dev, "not enough queue entries, expect xmit timeout\n");
> > +
>
> How about actually fixing it though? E.g. by linearizing...
Actually, the linearing is only needed for the case when the indirect
descriptor is not supported.
>
> It also bothers me that there's practically
> /proc/sys/net/core/max_skb_frags
> and if that's low then things could actually work.
Probably not as it won't exceed MAX_SKB_FRAGS.
>
> Finally, while originally it was just 17 typically, now it's
> configurable. So it's possible that you change the config to make big
> tcp
Note that virtio-net doesn't fully support big TCP.
> work better and device stops working while it worked fine
> previously.
For this patch, I guess not as we had:
if (sq->vq->num_free < 2+MAX_SKB_FRAGS)
in the tx path. So it won't even work before this patch.
Thanks
>
>
> > pr_debug("virtnet: registered device %s with %d RX and TX vq's\n",
> > dev->name, max_queue_pairs);
> >
> > --
> > 2.39.2
>