On Wed, Mar 29, 2023 at 03:27:03PM +0800, Xuan Zhuo wrote:
> Maybe one new thread is better.
>
> Thanks.
I don't know but do not post same message twice please
without explanation. if you repost put "PATCH repost" in
the subject.
___
Virtualization
Ican't parse the subject at all. I think the subject from v2
was fine - we are skipping event index updates on get buf.
Or maybe go higher level and describe the effect of the patch:
virtio_ring: reduce interrupt rate with event idx enabled
On Wed, Mar 29, 2023 at 03:21:35PM +0800, Albert Huang
On Wed, 29 Mar 2023 17:33:23 +0800, =?utf-8?b?6buE5p2w?=
wrote:
> Xuan Zhuo 于2023年3月29日周三 17:27写道:
> >
> > On Wed, 29 Mar 2023 15:28:41 +0800, Albert Huang
> > wrote:
> > > From: "huangjie.albert"
> > >
> > > in virtio_net, if we disable the napi_tx, when we triger a tx interrupt,
> > > the
On Wed, 29 Mar 2023 15:28:41 +0800, Albert Huang
wrote:
> From: "huangjie.albert"
>
> in virtio_net, if we disable the napi_tx, when we triger a tx interrupt,
> the vq->event_triggered will be set to true. It will no longer be set to
> false. Unless we explicitly call
Maybe one new thread is better.
Thanks.
On Wed, 29 Mar 2023 15:21:35 +0800, Albert Huang
wrote:
> From: "huangjie.albert"
>
> in virtio_net, if we disable the napi_tx, when we triger a tx interrupt,
> the vq->event_triggered will be set to true. It will no longer be set to
> false. Unless we