Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
MIME-Version: 1.0
References: <20230329072841.45304-1-huangjie.albert@bytedance.com> <1680082024.6148505-2-xuanzhuo@linux.alibaba.com>
In-Reply-To: <1680082024.6148505-2-xuanzhuo@linux.alibaba.com>
From:   =?UTF-8?B?6buE5p2w?= <huangjie.albert@bytedance.com>
Date:   Wed, 29 Mar 2023 17:33:23 +0800
Message-ID: <CABKxMyNXm4W+-yECstbnqy6U4_GYoOtfZ7h6+v3URzmVaJLOBQ@mail.gmail.com>
Subject: Re: [External] Re: [PATCH v3] virtio_ring: interrupt disable flag
 updated to vq even with event_triggered is set
To:     Xuan Zhuo <xuanzhuo@linux.alibaba.com>
Cc:     virtualization@lists.linux-foundation.org,
        linux-kernel@vger.kernel.org, Jason Wang <jasowang@redhat.com>,
        "Michael S . Tsirkin" <mst@redhat.com>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

Xuan Zhuo <xuanzhuo@linux.alibaba.com> =E4=BA=8E2023=E5=B9=B43=E6=9C=8829=
=E6=97=A5=E5=91=A8=E4=B8=89 17:27=E5=86=99=E9=81=93=EF=BC=9A
>
> On Wed, 29 Mar 2023 15:28:41 +0800, Albert Huang <huangjie.albert@bytedan=
ce.com> wrote:
> > From: "huangjie.albert" <huangjie.albert@bytedance.com>
> >
> > 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 t=
o
> > false. Unless we explicitly call virtqueue_enable_cb_delayed or
> > virtqueue_enable_cb_prepare.
> >
> > If we disable the napi_tx, it will only be called when the tx ring
> > buffer is relatively small.
> >
> > Because event_triggered is true. Therefore, VRING_AVAIL_F_NO_INTERRUPT =
or
> > VRING_PACKED_EVENT_FLAG_DISABLE will not be set. So we update
> > vring_used_event(&vq->split.vring) or vq->packed.vring.driver->off_wrap
> > every time we call virtqueue_get_buf_ctx.This bring more interruptions.
> >
> > To summarize:
> > 1) event_triggered was set to true in vring_interrupt()
> > 2) after this nothing will happen for virtqueue_disable_cb() so
> >    VRING_AVAIL_F_NO_INTERRUPT is not set in avail_flags_shadow
> > 3) virtqueue_get_buf_ctx_split() will still think the cb is enabled
> >    then it tries to publish new event
> >
> > To fix:
> > update VRING_AVAIL_F_NO_INTERRUPT or VRING_PACKED_EVENT_FLAG_DISABLE to=
 vq
> > when we call virtqueue_disable_cb even the event_triggered is set to tr=
ue.
> >
> > Tested with iperf:
> > iperf3 tcp stream:
> > vm1 -----------------> vm2
> > vm2 just receives tcp data stream from vm1, and sends the ack to vm1,
> > there are many tx interrupts in vm2.
> > but without event_triggered there are just a few tx interrupts.
> >
> > v2->v3:
> > -update the interrupt disable flag even with the event_triggered is set=
,
> > -instead of checking whether event_triggered is set in
> > -virtqueue_get_buf_ctx_{packed/split}, will cause the drivers  which ha=
ve
> > -not called virtqueue_{enable/disable}_cb to miss notifications.
> >
> > Fixes: 8d622d21d248 ("virtio: fix up virtio_disable_cb")
> > Signed-off-by: huangjie.albert <huangjie.albert@bytedance.com>
> > Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
> > Signed-off-by: Jason Wang <jasowang@redhat.com>
> > ---
> >  drivers/virtio/virtio_ring.c | 24 +++++++++++++++++-------
> >  1 file changed, 17 insertions(+), 7 deletions(-)
> >
> > diff --git a/drivers/virtio/virtio_ring.c b/drivers/virtio/virtio_ring.=
c
> > index 307e139cb11d..ad74463a48ee 100644
> > --- a/drivers/virtio/virtio_ring.c
> > +++ b/drivers/virtio/virtio_ring.c
> > @@ -812,6 +812,14 @@ static void virtqueue_disable_cb_split(struct virt=
queue *_vq)
> >
> >       if (!(vq->split.avail_flags_shadow & VRING_AVAIL_F_NO_INTERRUPT))=
 {
> >               vq->split.avail_flags_shadow |=3D VRING_AVAIL_F_NO_INTERR=
UPT;
> > +
> > +             /*
> > +              * If device triggered an event already it won't trigger =
one again:
> > +              * no need to disable.
> > +              */
> > +             if (vq->event_triggered)
> > +                     return;
> > +
> >               if (vq->event)
> >                       /* TODO: this is a hack. Figure out a cleaner val=
ue to write. */
> >                       vring_used_event(&vq->split.vring) =3D 0x0;
> > @@ -1544,8 +1552,16 @@ static void virtqueue_disable_cb_packed(struct v=
irtqueue *_vq)
> >  {
> >       struct vring_virtqueue *vq =3D to_vvq(_vq);
> >
> > -     if (vq->packed.event_flags_shadow !=3D VRING_PACKED_EVENT_FLAG_DI=
SABLE) {
> > +     if (!(vq->packed.event_flags_shadow & VRING_PACKED_EVENT_FLAG_DIS=
ABLE)) {
>
> This seems to be another problem.
>
> Thanks.
>

Here, we are only making this change to maintain consistency with
virtqueue_disable_cb_split.
Is there any concern with doing so?

Thanks.

>
> >               vq->packed.event_flags_shadow =3D VRING_PACKED_EVENT_FLAG=
_DISABLE;
> > +
> > +             /*
> > +              * If device triggered an event already it won't trigger =
one again:
> > +              * no need to disable.
> > +              */
> > +             if (vq->event_triggered)
> > +                     return;
> > +
> >               vq->packed.vring.driver->flags =3D
> >                       cpu_to_le16(vq->packed.event_flags_shadow);
> >       }
> > @@ -2063,12 +2079,6 @@ void virtqueue_disable_cb(struct virtqueue *_vq)
> >  {
> >       struct vring_virtqueue *vq =3D to_vvq(_vq);
> >
> > -     /* If device triggered an event already it won't trigger one agai=
n:
> > -      * no need to disable.
> > -      */
> > -     if (vq->event_triggered)
> > -             return;
> > -
> >       if (vq->packed_ring)
> >               virtqueue_disable_cb_packed(_vq);
> >       else
> > --
> > 2.31.1
> >