wireless mcast updates, ic->ic_update_mcast

[Adrian Chadd]

Cool. Wrap it up in bootverbose. Hm, or what about a print_once kind of thing?


As far as I can briefly tell, the drivers -should- work with multicast
as long as they're treating it like broadcast traffic and not
requiring that you program a multicast filter for specific link-local
things. The problem is that noone has gone through and verified that
is indeed the case.


-a


On 24 June 2014 11:01, Warren Block <wblock at wonkity.com> wrote:
> On Tue, 24 Jun 2014, Sean Bruno wrote:
>
>> On Sat, 2014-06-21 at 12:47 -0700, Adrian Chadd wrote:
>>>
>>> Well, it depends on what the hardware may want or desire to function
>>> correctly. I've no idea what theose chips require for multicast
>>> behaviour.
>>
>>
>> I think that was kind of my point of bringing this up.  Only a handful
>> of cards really implement the appropriate code.  Most of the rest,
>> implement stub functions that hide a console print message indicating
>> missing functionality.
>>
>> I propose nuking most of these stub functions and wrapping the console
>> printf in a bootverbose.
>
>
> FWIW, I agree.  That little error-that's-not-an-error is what new users see,
> and it obviously must be the problem.  So not only is it not
> helping, it often distracts from real problems.