--Signature_Mon__13_Feb_2006_12_55_11_-0500_KcQvyKTMPfta4_3a
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: quoted-printable

Hello,

> > > (on my G4 I get "snapper0: unable to find i2c", so something's
> > > even more wrong there.)
> >=20
> > That happens when the deq 'driver' doesn't attach for some reason.
> > Its sole purpose is to attach to ki2c and get the mixer's i2c
> > address out of OF so snapper can talk to it.
> >=20
> > Maybe Apple renamed device nodes again.
>=20
> ok, "deq" doesn't attach for me.  but it's in the OFW tree:
>=20
> ff967a00: /pci@f2000000/mac-io@17/i2c@18000/deq
>=20
> name                    64657100 ........ ........ ........   "deq"
> device_type             64657100 ........ ........ ........   "deq"
> i2c-address             0000006a ........ ........ ........   ...j

Looks good.

> I put a full dmesg and ofdump output at
> ftp://ftp.netbsd.org/pub/NetBSD/misc/chs/macppc/

This is odd - the 2nd ki2c doesn't enumerate any child nodes.
ARGH, I see the reason - mac-io/i2c has child nodes representing i2c
devices but they don't have reg properties so ki2c skips them. I'll make
it look at i2c-address if there's no reg.

> is there some reason the old code that searches for the i2c info
> is still in snapper_defer()?  (though #if 0'd)

Not really.

have fun
Michael

--Signature_Mon__13_Feb_2006_12_55_11_-0500_KcQvyKTMPfta4_3a
Content-Type: application/pgp-signature

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.0 (NetBSD)

iQEVAwUBQ/DH/8pnzkX8Yg2nAQI4Ogf+IvJyzRgtHnj56A3FfOcTjywfBJrpAQ05
40GU9N9ynFmT6Lo4BozZ/G2BXEC3RbYbccqRKdNLj//6BkWSlca57vxlRu1AzLo5
LBUo83+2Ou0NFwpDUbHzOoZQpzY3OV8HINwxOjGSkDmrux8FCrHIytpDU8HZaBIu
G+pt117PxJqe0gNxdjehFaVd8qy6CpmNhf3h2YzUEzGE8UtBInBPmTXW8Nrgkk8B
BNLDGVCf2fTTLpQR+tUyTNlMKo+fSGJJAmIBGFlhhe1AyDdOF1YvhLuCn62N7FrG
AD/CC2BFnyQTvBc418G0oHdeQhZsvcN8SbqFfY+dx0ynmPi/C/mqqA==
=5D8a
-----END PGP SIGNATURE-----

--Signature_Mon__13_Feb_2006_12_55_11_-0500_KcQvyKTMPfta4_3a--