Re: [linux-pm] [PATCH V3 00/10] PM: Create the AVS(Adaptive Voltage Scaling)

[Kevin Hilman <khilman@ti.com>]

"Menon, Nishanth" <nm@ti.com> writes:

> On Wed, May 9, 2012 at 1:29 PM, Kevin Hilman <khilman@ti.com> wrote:
>> "Woodruff, Richard" <r-woodruff2@ti.com> writes:
>>
>>>> From: Hilman, Kevin
>>>> Sent: Tuesday, May 08, 2012 5:17 PM
>>>
>>>> A basic OMAP AVS driver has been in mainline for a long time, yet we
>>>> have not seen support submitted for all of these features.
>>>
>>> 1.5/3.5 is a feature.
>>
>> And I'm still waiting for it to be submitted upstream.
>>
>>> ABB is requirement for a production useable driver. Higher speed rated
>>> OMAP4 and all OMAP5 added these to be useable.
>>
>> ditto
>>
>>> Yes this is effort. Point of mentioning is to raise awareness of need.
>>
>> I'm well aware of the need.
>>
>>> Yet to be added feature has different meaning than functional gap.
>>
>> And both need to be submitted upstream.
>
> SR 1.5: http://marc.info/?l=linux-omap&m=129933897910785&w=2
> ABB: http://marc.info/?l=linux-omap&m=130939399209099&w=2
>
> I am not sure what you mean "need to be submitted upstream"?

You're right.  I should've said re-submitted and merged.  Both have been
submitted (and reviewed) but no follow up submissions after review, and
thus they're still out of tree.

> Just tired of seeing things perpetually change without considering
> even how to handle features that are mandatory for SoC even with code
> posted upstream to show exactly what it takes.. 

I'm sorry, but this is not perpetual change.  

This driver has been upstream in its current (admittedly
feature-limited) form for a long time, the only thing changing in
$SUBJECT series is the location of the driver.  Why all the fuss about
the missing features now?

> I think you do mean merged upstream in this context.

Correct.

Frameworks always have limitations.  The way they get extended/expanded
etc. is by the submission/review/merging of support for new
features/requirements.  The process for that is the same as any feature
in any part of the kernel.

Evolution, not intelligent design[1].

All of that being said, I'm not sure why this thread was hijacked for
this debate in the first place.  The point of $SUBJECT series is simply
to move and *existing* framework from arch/arm out to drivers.  The only
changes done are cleanups to make this move possible.

I for one would welcome extending this framework to ensure it supports
all the SoC features.  I just don't want those features to be a
prerequisite for this move from arch/arm to drivers.

Please, let's get this moved to drivers, and then add support for the
missing features.

Thanks,

Kevin

[1] http://kerneltrap.org/Linux/Kernel_Evolution