public inbox for linux-pm@vger.kernel.org
 help / color / mirror / Atom feed
* Re: [RFC] [0/4] Voltage Framework
       [not found]   ` <46166A8E.8030602@dev.rtsoft.ru>
@ 2007-04-06 18:05     ` dmitry pervushin
  0 siblings, 0 replies; only message in thread
From: dmitry pervushin @ 2007-04-06 18:05 UTC (permalink / raw)
  To: Dmitry Krivoschokov; +Cc: linux-pm, ARM-kernel-linux, Harald Welte

On Птн, 2007-04-06 at 19:43 +0400, Dmitry Krivoschokov wrote:
[skipped]
> The struct resource was initially intended for IO resources only,
> (which looks wrong to me, at least its naming, it could be called
> as io_resources). But voltage is not an IO resource, so struct resource
> is not appropriate place for that
You object on naming of the structure (have you ever tried to do output
on interrupt line :) )?

The struct resource layout is not well suitable to keep information
about clocks/voltage and ever gpio lines, but all of these things are
resources. Resources of device. Well, resources that belongs to
platform_device.

I don't like idea to keep two separate lists of resources - one for I/O
ports/memory windows/IRQ lines and the another one to keep pm resources.
Why should they be separated ?
  
> >This might sound odd, but after all, why not look at power [voltage] as
> >a resource just loke memory, IO ports, GPIO or IRQ lines?
> >
> I'd consider voltage and clocks as a power resources, every computer
> (say CMOS) system (or device) consumes (requires for) voltage and clocks,
> so it looks natural that device driver will check what resources
> are needed for device on this particular platform.
> 
> >
> >This keeps the drivers clean from having to know any device specifics.
> >
> >
> Yep.
> 
> >It also keeps the actual device driver of any peripheral independent of
> >the PMU driver, since the latter would sit behind the 'struct resource'
> >abstraction.  
> >A device driver would then just call something like
> >requrest_power(struct resource *) whic hides all the magic of
> >calling back into the individual PMU driver.
> >
> It assumes some infrastructure to deal with power resources, like
> we have for interrupts now.
> 
> Actually, all Linux PM-related problems are discussed on
> linux-pm@lists.linux-foundation.org list, here it's offtopic.
> Continue the discassion on linux-pm, if you will.
I added linux-pm to Cc, so let's continue there.
> 
> 
> Thanks,
> Dmitry
> 
> 
> -------------------------------------------------------------------
> List admin: http://lists.arm.linux.org.uk/mailman/listinfo/linux-arm-kernel
> FAQ:        http://www.arm.linux.org.uk/mailinglists/faq.php
> Etiquette:  http://www.arm.linux.org.uk/mailinglists/etiquette.php

_______________________________________________
linux-pm mailing list
linux-pm@lists.linux-foundation.org
https://lists.linux-foundation.org/mailman/listinfo/linux-pm

^ permalink raw reply	[flat|nested] only message in thread

only message in thread, other threads:[~2007-04-06 18:05 UTC | newest]

Thread overview: (only message) (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
     [not found] <1174469715.19019.65.camel@amit-laptop>
     [not found] ` <20070406073159.GE4081@prithivi.gnumonks.org>
     [not found]   ` <46166A8E.8030602@dev.rtsoft.ru>
2007-04-06 18:05     ` [RFC] [0/4] Voltage Framework dmitry pervushin

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox