Re: [RFC PATCH v1 04/26] char-socket: add 'reconnecting' property

[Adalbert Lazăr]

On Wed, 15 Apr 2020 12:46:57 +0200, Marc-André Lureau <address@hidden> wrote:
> Hi
> 
> On Wed, Apr 15, 2020 at 3:03 AM Adalbert Lazăr <address@hidden> wrote:
> >
> > This is used by the VM introspection object to check if the connection
> > will be reestablished in case it disconnects from some reason.
> >
> > The closing of the socket is used by any of the three parties involved,
> > KVM, the introspection tool and QEMU (eg. on force-reset), to signal
> > the other parties that the session is over. As such, it is very important
> > that the socket will reconnect.
> >
> > CC: "Marc-André Lureau" <address@hidden>
> > CC: Paolo Bonzini <address@hidden>
> > Signed-off-by: Adalbert Lazăr <address@hidden>
> > ---
> >  chardev/char-socket.c | 11 +++++++++++
> >  1 file changed, 11 insertions(+)
> >
> > diff --git a/chardev/char-socket.c b/chardev/char-socket.c
> > index fd0106ab85..22ab242748 100644
> > --- a/chardev/char-socket.c
> > +++ b/chardev/char-socket.c
> > @@ -1492,6 +1492,13 @@ char_socket_get_connected(Object *obj, Error **errp)
> >      return s->state == TCP_CHARDEV_STATE_CONNECTED;
> >  }
> >
> > +static bool char_socket_get_reconnecting(Object *obj, Error **errp)
> > +{
> > +    SocketChardev *s = SOCKET_CHARDEV(obj);
> > +
> > +    return s->reconnect_time > 0;
> > +}
> > +
> >  static int tcp_chr_reconnect_time(Chardev *chr, int secs)
> >  {
> >      SocketChardev *s = SOCKET_CHARDEV(chr);
> > @@ -1528,6 +1535,10 @@ static void char_socket_class_init(ObjectClass *oc, 
> > void *data)
> >
> >      object_class_property_add_bool(oc, "connected", 
> > char_socket_get_connected,
> >                                     NULL, &error_abort);
> > +
> > +    object_class_property_add_bool(oc, "reconnecting",
> > +                                   char_socket_get_reconnecting,
> > +                                   NULL, &error_abort);
> 
> That may be misleading, as the socket connection may be established
> and this will return true if reconnect_time > 0. Why not have a
> "reconnect-time" property instead?
> 

Sure.
Thanks.