Hi,

On 10/26/06, Michael Lorenz <macallan@netbsd.org> wrote:
> > #ki2c*        at obio?                        # Keywest I2C
> >
> > You need to uncomment that line too since ki2c is needed by snapper(4)
> > to operate.
>
> That's not enough anymore - what matters is
> deq* at ki2c?
> Since there may be other devices hooked up to the iic bus which we may
> want to use I changed the snapper driver to not talk to the ki2c
> directly anymore but look for deq instead.
> Without deq snapper won't be able to find its mixer.

Oops... I see. I have to update this system to -current (it was
running NetBSD-2 for ages and now it's still with NetBSD 3.1_RC2).
Also it appears that the snapper driver is greatly improved in -current!

Although taking a look to the CVS log it appears that this change
doesn`t affect NetBSD-3-0-1 ...

> It may be worth a try to have both ki2c ( at obio and at uni_n ),
> adm1030c at ki2c and adt7467c at ki2c enabled - chances are that the
> eMac uses one of those thermal monitor / fan controller chips. That
> would give various temperatures, fan speeds etc. via envsys.

That's interesting... I have to do some tests with this -- getting
access to thermal data would be nice. But I think the eMac doesn't
have a fan controller. In fact, there are people who installed one to
reduce the noise caused by the eMac fan... [in my case this is not
much of a problem since the UPS near the eMac causes as twice as noise
as the eMac itself... :-) ]

> Reminds me - I need to clean up and commit the code that allows us to
> use hardware accelerated framebuffer drivers as console. That would
> also eliminate the messed up screen when X exits, at least with drivers
> that know how to change video modes ( like radeonfb... )

That's great!

     -Marco