Hi Richard, On Tue, 2018-09-18 at 06:06 -0700, Richard Cochran wrote: > On Thu, Sep 13, 2018 at 08:35:22PM -0700, Richard Cochran wrote: > > > > Yes, I'll get to it as soon as I can. Overall I'm happy with this > > series, but I wanted to sit down and think through the automatic > > switch to MASTER state that we discussed. > Okay, I took your series and hacked the following diff on top of it. > The main difference is that my version doesn't touch clock_poll(). > > Also, there is no need for the master mode to go into the listening > state. > Is there any specific reason you want to transition to the LISTENING state first for the slave mode? We are initializing the delay timer in port_initialize() and the announce timer is also initialized when the device transitions to PS_SLAVE. So, there is nothing to do in slave mode.
If we do remove the PS_LISTENING transition, We don't have any need for the following code as well. > diff --git a/configs/automotive-slave.cfg b/configs/automotive- > slave.cfg > index 9cabe7a..66b6141 100644 > --- a/configs/automotive-slave.cfg > +++ b/configs/automotive-slave.cfg > @@ -28,3 +28,5 @@ BMCA noop > slaveOnly 1 > inhibit_announce 1 > ignore_source_id 1 > +announceReceiptTimeout 2 > +logAnnounceInterval -20 Thanks, Vedang _______________________________________________ Linuxptp-devel mailing list Linuxptp-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/linuxptp-devel
