Re: [Xen-devel] [PATCH net v3 1/1] xen-netback: Handle backend state transitions in a more robust way

[Paul Durrant <Paul.Durrant@xxxxxxxxxx>]

> -----Original Message-----
> From: Paul Durrant
> Sent: 24 September 2013 16:31
> To: Wei Liu
> Cc: xen-devel@xxxxxxxxxxxxx; netdev@xxxxxxxxxxxxxxx; Ian Campbell; Wei Liu;
> David Vrabel
> Subject: RE: [PATCH net v3 1/1] xen-netback: Handle backend state
> transitions in a more robust way
> 
> > -----Original Message-----
> > From: Wei Liu [mailto:wei.liu2@xxxxxxxxxx]
> > Sent: 24 September 2013 16:26
> > To: Paul Durrant
> > Cc: xen-devel@xxxxxxxxxxxxx; netdev@xxxxxxxxxxxxxxx; Ian Campbell; Wei
> Liu;
> > David Vrabel
> > Subject: Re: [PATCH net v3 1/1] xen-netback: Handle backend state
> > transitions in a more robust way
> >
> > On Tue, Sep 24, 2013 at 03:51:22PM +0100, Paul Durrant wrote:
> > > When the frontend state changes metback now specifies its desired state
> > to
> >                                   netback
> > > a new function, set_backend_state(), which transitions through any
> > [...]
> > > +/* Handle backend state transitions:
> > > + *
> > > + * The backend state starts in InitWait and the following transtions are
> >                                                              transitions
> > > + * allowed.
> > >
> > [...]
> > > @@ -363,7 +448,9 @@ static void hotplug_status_changed(struct
> > xenbus_watch *watch,
> > >   if (IS_ERR(str))
> > >           return;
> > >   if (len == sizeof("connected")-1 && !memcmp(str, "connected", len))
> > {
> > > -         xenbus_switch_state(be->dev, XenbusStateConnected);
> > > +         /* Complete any pending state change */
> > > +         xenbus_switch_state(be->dev, be->state);
> > > +
> >
> > The state transition takes place iff hotplug status is "connected", is
> > this desirable? What if hotplug fails?
> >
> 
> The xenbus state will remain in InitWait but the backend state will be
> Connected.
> 
> > If it cycles through connect again it looks like it will trigger that
> > BUG_ON in connect()?
> >
> 
> No, I don't think so because be->state is not the same as dev->state. The
> frontend can go to Closing/Closed (which will take dev->state to
> Closing/Closed) and this should be fine.
> 

Sorry, I misread. You're right. I'll fix.

  Paul

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
http://lists.xen.org/xen-devel