Re: [Qemu-devel] [PATCH v2 1/1] virtio: serial: expose a 'guest_writable' callback for users

[Amit Shah]

On (Thu) 09 Oct 2014 [13:18:16], Markus Armbruster wrote:
> Amit Shah <address@hidden> writes:
> 
> > Users of virtio-serial may want to know when a port becomes writable.  A
> > port can stop accepting writes if the guest port is open but not being
> > read from.  In this case, data gets queued up in the virtqueue, and
> > after the vq is full, writes to the port do not succeed.
> >
> > When the guest reads off a vq element, and adds a new one for the host
> > to put data in, we can tell users the port is available for more writes,
> > via the new ->guest_writable() callback.
> >
> > Signed-off-by: Amit Shah <address@hidden>
> >
> > ---
> > v2: check for port != NULL (Peter Maydell)
> > ---
> >  hw/char/virtio-serial-bus.c       | 27 +++++++++++++++++++++++++++
> >  include/hw/virtio/virtio-serial.h |  3 +++
> >  2 files changed, 30 insertions(+)
> >
> > diff --git a/hw/char/virtio-serial-bus.c b/hw/char/virtio-serial-bus.c
> > index 3931085..1c7acbf 100644
> > --- a/hw/char/virtio-serial-bus.c
> > +++ b/hw/char/virtio-serial-bus.c
> > @@ -465,6 +465,33 @@ static void handle_output(VirtIODevice *vdev, 
> > VirtQueue *vq)
> >  
> >  static void handle_input(VirtIODevice *vdev, VirtQueue *vq)
> >  {
> > +    /*
> > +     * Users of virtio-serial would like to know when guest becomes
> > +     * writable again -- i.e. if a vq had stuff queued up and the
> > +     * guest wasn't reading at all, the host would not be able to
> > +     * write to the vq anymore.  Once the guest reads off something,
> > +     * we can start queueing things up again.
> > +     */
> > +    VirtIOSerial *vser;
> > +    VirtIOSerialPort *port;
> > +    VirtIOSerialPortClass *vsc;
> > +
> > +    vser = VIRTIO_SERIAL(vdev);
> > +    port = find_port_by_vq(vser, vq);
> > +
> > +    if (!port) {
> > +        return;
> > +    }
> > +    vsc = VIRTIO_SERIAL_PORT_GET_CLASS(port);
> > +
> > +    /*
> > +     * If guest_connected is false, this call is being made by the
> > +     * early-boot queueing up of descriptors, which is just noise for
> > +     * the host apps -- don't disturb them in that case.
> > +    */
> > +    if (port->guest_connected && port->host_connected && 
> > vsc->guest_writable) {
> > +        vsc->guest_writable(port);
> > +    }
> >  }
> >  
> >  static uint32_t get_features(VirtIODevice *vdev, uint32_t features)
> > diff --git a/include/hw/virtio/virtio-serial.h 
> > b/include/hw/virtio/virtio-serial.h
> > index a679e54..b434f78 100644
> > --- a/include/hw/virtio/virtio-serial.h
> > +++ b/include/hw/virtio/virtio-serial.h
> > @@ -98,6 +98,9 @@ typedef struct VirtIOSerialPortClass {
> >          /* Guest is now ready to accept data (virtqueues set up). */
> >      void (*guest_ready)(VirtIOSerialPort *port);
> >  
> > +        /* Guest vq became writable again */
> > +    void (*guest_writable)(VirtIOSerialPort *port);
> > +
> >      /*
> >       * Guest wrote some data to the port. This data is handed over to
> >       * the app via this callback.  The app can return a size less than
> 
> The code should work, but whether it makes sense is hard to judge for
> virtio noobs like me without a user of guest_writable.  The conditional
> guarding vsc->guest_writable(port) in particular.

Right.  This was originally requested by the spice folks, and they
don't yet have a user implemented (waiting for the spice-char
implementation).  But Peter came up with a user; so I posted this w/o
the spice part of it.  But looks like Peter has lost the code for his
user, so this patch will have to wait ;-)

> virtio_add_queue()'s callback being undocumented doesn't exactly help,
> either.  Fun: the parameter is called handle_output, the argument is
> handle_input.  Clear as mud!

Yea - some things in virtio are from the guest's POV so it makes these
things really confusing in qemu.

                Amit