On 2/4/2020 2:34 PM, Erik Hons wrote: > This fixes an issue with free running clocks. When the master clock id > changes the measured master/local ratio should be reset. If it isn't, > and the local clock becomes master, then the last measured ratio to > the old master continues to be used in peer delay calculations. If a > bad master/local ration calculation triggers this state change the > port can get "stuck" in not "asCapable" mode. >
Seems fairly straight forward to me. > Signed-off-by: Erik Hons <erik.h...@ni.com> > --- > clock.c | 1 + > 1 file changed, 1 insertion(+) > > diff --git a/clock.c b/clock.c > index 146576a..a48656c 100644 > --- a/clock.c > +++ b/clock.c > @@ -1774,6 +1774,7 @@ static void handle_state_decision_event(struct clock *c) > tsproc_set_delay(c->tsproc, c->initial_delay); > c->ingress_ts = tmv_zero(); > c->path_delay = c->initial_delay; > + c->master_local_rr = 1.0; > c->nrr = 1.0; > fresh_best = 1; > } > _______________________________________________ Linuxptp-devel mailing list Linuxptp-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/linuxptp-devel
