On Tue, 14 Nov 2023 11:55:52 +0800, Jason Wang <jasowang@xxxxxxxxxx> wrote: > On Tue, Nov 14, 2023 at 11:42 AM Xuan Zhuo <xuanzhuo@xxxxxxxxxxxxxxxxx> wrote: > > > > On Tue, 14 Nov 2023 11:26:42 +0800, Jason Wang <jasowang@xxxxxxxxxx> wrote: > > > On Thu, Nov 9, 2023 at 7:06 PM Xuan Zhuo <xuanzhuo@xxxxxxxxxxxxxxxxx> wrote: > > > > > > > > On Thu, 9 Nov 2023 14:37:38 +0800, Jason Wang <jasowang@xxxxxxxxxx> wrote: > > > > > On Tue, Nov 7, 2023 at 11:12 AM Xuan Zhuo <xuanzhuo@xxxxxxxxxxxxxxxxx> wrote: > > > > > > > > > > > > If the xsk is enabling, the xsk tx will share the send queue. > > > > > > But the xsk requires that the send queue use the premapped mode. > > > > > > So the send queue must support premapped mode. > > > > > > > > > > > > Signed-off-by: Xuan Zhuo <xuanzhuo@xxxxxxxxxxxxxxxxx> > > > > > > --- > > > > > > drivers/net/virtio/main.c | 163 ++++++++++++++++++++++++++++---- > > > > > > drivers/net/virtio/virtio_net.h | 16 ++++ > > > > > > 2 files changed, 163 insertions(+), 16 deletions(-) > > > > > > > > [...] > > > > > > > > > > > I think we need to seek a way to reuse what has been stored by virtio > > > > > core. It should be much more efficient. > > > > > > > > > > > > Yes. > > > > > > > > But that is for net-next branch. > > > > > > > > Can we do that as a fix after that is merged to 6.8? > > > > > > We still have time. I would like to do it from the start. > > > > > > I want to finish the job including new AF_XDP ZC feature. > > Because that this must wait the merge window. > > Base on that, the optimizing work can be done everytime. > > > > If we work from the new virtio prepare, that can be merged to 6.8. > > And the AF_XDP zc must wait 6.9. right? > > It can be part of this series. Or anything I missed? > > My understanding is that, since the information is handy, it just > requires new helpers. So I don't expect it needs a large series. Now, this is pushing to net-next. If we add an new virtio-core helper. That must be pushed to virtio branch. And this patch set must wait that. Thanks. > > Thanks > > > > > Thanks > > >
