Re: [Mactel-linux-devel] SF.net SVN: mactel-linux: [97] trunk/kernel/mactel-patches-2.6.21

[Nicolas B. <ni...@bo...>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Kai Weber wrote:
> * nbo...@us... <nbo...@us...>:
> 
> Hi Nicolas, do you consider the comment from Dmitry Torokhov?
> http://lkml.org/lkml/2007/3/22/193
> 
> I think, he is right on this.
> 
> "Please consider implemention open and close methods for the input
> device and start/stop timer from there - there is no point of checking
> hardware state if noone is listening to events."

Yeah yeah I agree, he's right... Thank you for pointing me out this
message, for some reason I never received it. Will fix that.

> In this blog post the author thinks, that a interrupt is used in
> AppleSMC for sensor monitoring
> (http://blog.technologeek.org/2006/12/27/44). Have you infos about that?

Interesting post, unfortunately he does not provide anything proving
what he's stating in his article. (not saying he never contacted me to
give me more details about his findings)

For example, for me, this sentence: "It is highly probable that the OS X
driver is taking over the fan control to keep the machine both cool and
quiet." contradicts basic principles of physics, I don't understand how
you can run the fans more slowly, and have a lower temperature...

Anyway, it's very highly likely there are some interrupts in the SMC,
which are not handled by the Linux driver. These are probably used for
the accelerometer device, to be able to park the head of the hard drive
quickly in case of shock of free fall (the old PowerBook AMS supports
these interrupts, and I don't see why Apple would have removed this
important feature).

Best regards,

Nicolas
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.3 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFGEwO401ajQnpJXgERAlLCAJ44neMKd4TvtOYYkwOpbxr2Dp/O9ACfaZbV
NPH0UEpfMD03N2NtvmyLRyw=
=DY9z
-----END PGP SIGNATURE-----