Fix up power managment in 2.6

[Pavel Machek <pavel@ucw.cz>]

[-- Attachment #1: Type: text/plain, Size: 1171 bytes --]

Hi!

Patrick fucked up power managment in 2.6.0-test4, without discussing
it with anybody. He
* changed userland interface
* changed in-kernel interface, in such way that all drivers needing
power
managment need to be changed. (And he messed it up, driver no longer
can do something with both interrupts disabled and enabled on resume).
* DID NOT BOTHER TO TEST HIS CHANGES and broke both suspend-to-disk
and suspend-to-ram. His "cleanups" are unneccessary and clearly buggy
(he changed prototype but failed to change assembly function), and he
ignores non-i386 architectures.

Unfortunately that diff is big enough and does so many things at once
that it is virtually impossible to fix. No usefull work can be done on
power managment in this state.

Please take attached patch, and apply with -R, to bring tree back to
-test3 state. It should be a non-brainer, as Patrick's patch should
not go in in the first place. [If you can do some bk magic to exclude
Patrick's power managment merge, that would be even better].
	                                                        Pavel

-- 
When do you have a heart between your knees?
[Johanka's followup: and *two* hearts?]

[-- Attachment #2: delme.gz --]
[-- Type: application/octet-stream, Size: 39494 bytes --]