On 1/17/19 10:36 AM, Vivien Didelot wrote: > On Wed, 16 Jan 2019 12:01:00 -0800, Florian Fainelli <f.faine...@gmail.com> > wrote: > >> + int (*port_multicast_toggle)(struct dsa_switch *ds, int port, >> + bool mc_disabled); > > Waw this looks counter-intuitive and error-prone... Would you prefer to make > it > symmetrical to ops->port_vlan_filtering() by implementing > ops->port_multicast() > and passing !obj->u.mc_disabled to it?
I debated doing that, but I think it is better to have the same boolean polarity from end to end, even if it is highly counter intuitive as you said. That is kind of the story of bridge attributes annyway, completely counter intuitive. -- Florian
