[patch] Create a new device_pm.{c|h} for internal use

Danny Kukawka danny.kukawka at web.de
Tue Jul 3 11:13:04 PDT 2007


On Dienstag, 3. Juli 2007, Richard Hughes wrote:
> On Tue, 2007-07-03 at 19:56 +0200, Danny Kukawka wrote:
> > Uh ... looks not that good to me. Why do you use in pmu.c
> > device_pm_remove_optional_props() to remove 6 properties by try to
> > remove 35.
> > Looks really inefficient, you should only remove the keys you really
> > have used.
>
> Disagree. Not removing a missing property is very cheap, and removing a
> battery is something that is done once in a blue moon. Abstracting out
> the common routine easily outweighs the few us slowdown.

I have to say: I'm not happy about this behaviour, we should only remove what 
we added and nothing more.

I have may overseen something, but where are the mails (with the patches) to 
all these ACPI/PMU/APM commits in the last hour? They are really confusing 
and I miss the discussion about them (at least the commit logs are also not 
that descriptive in some cases).

I hope they introduce no new ACPI related bugs, the code is really sensitive 
as you know.

Danny


More information about the hal mailing list