[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]
Re: adt7xxx drivers (Was Re: Device name length restriction?)
-----BEGIN PGP SIGNED MESSAGE-----
On Sep 9, 2008, at 1:54 PM, Paul Goyette wrote:
On Tue, 9 Sep 2008, David Brownlee wrote:
For a GENERIC kernel the table size is amost certainly offset by
reducing duplication, and if someone is that size crazy then they
can have an ifdef to disable some table entries. Merged driver
definitely seems like a good idea to me...
These drivers don't really get included in GENERIC kernels (at
least, not on amd64/i386) since the i2c bus isn't really probe-
able, and the devices all share a common i2c address (0x2e). This
really wouldn't affect GENERIC kernels at all.
Now, I'm not at all sure about OpenFirmware ports, where there's an
external source of configuration information that doesn't require
probing of the i2c bus. I don't know if any of these devices are
included in those GENERIC kernels. Michael, any guidance here?
macppc's GENERIC contains this:
adt7467c* at ki2c?
adm1030c* at ki2c?
Both drivers use different, ki2c and OpenFirmware specific probe and
attach functions.( see sys/arch/macppc/dev/adt7467_ki2c.c for example )
So far, I've identified ten different chips that all could be
easily handled by a single driver, with trivial detection of the
specific chip based on two on-board registers.
adt 7460, 7463, 7466, 7467, 7468, 7473, 7475, 7476, & 7490
All of these are identified externally as "dBCool (tm) thermal
monitor and controller" and all made by ON Semiconductor (they
bought the product line from Analog Devices at the end of 2007).
If you can easily unify them I'm all for it. It would be nice if the
ki2c attachment remains intact though.
The ADM 1030 driver needs more testing, I wrote it blindly from the
datasheet and someone with an iBook G3 tested it. I never had the
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.7 (Darwin)
-----END PGP SIGNATURE-----
Main Index |
Thread Index |