Re: powernow-k8: support acpi

[Pavel Machek <pavel@ucw.cz>]

Hi!

>  > This is new version of powernow-k8 driver. It adds SMP support, and
>  > support for getting tables through ACPI. (ACPI support is really
>  > important, because many machines have broken "legacy" tables). Please
>  > apply,
> 
> First off, apologies for sitting on this for so long.
> I just made a first-pass attempt at merging this. Most of
> (maybe all?) the simpler bits are in (whitespace changes, typos etc).
> I'll push that out to bkbits soon, so the powernow-k8 diff in the
> next -mm that Andrew builds should be slightly smaller at least.

Adding of data->, too? Could you mail me your latest powernow-k8.[ch]?

> The ACPI stuff I'm pondering.
> Bruno, how does this stuff look to you? Your acpi perflib stuff is
> already in the pending pool (see the diff in the current -mm tree at
> http://www.kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.5-rc3/2.6.5-rc3-mm4/broken-out/bk-cpufreq.patch)
> Would this driver be better off using that ?

It already uses some kind of library as its *way* smaller than it used
to be...

>  > @@ -31,55 +33,47 @@
>  >  #include <asm/io.h>
>  >  #include <asm/delay.h>
>  >  
>  > +#ifdef CONFIG_ACPI
>  > +#define CONFIG_X86_POWERNOW_K8_ACPI
>  > +#endif
>  > +#ifdef CONFIG_X86_POWERNOW_K8_ACPI
>  > +
>  > +#include <linux/acpi.h>
>  > +#include <acpi/processor.h>
>  > +
>  > +#endif /* CONFIG_X86_POWERNOW_K8_ACPI */
> 
> This bit just looks *wrong* to me.  This looks like something
> that really should be sorted out in Kconfig.

No problem... Or perhaps just CONFIG_X86_POWERNOW_K8_ACPI should be
replaced with CONFIG_ACPI in the driver?
							Pavel
-- 
When do you have a heart between your knees?
[Johanka's followup: and *two* hearts?]