Re: RFC: powerd Patch & proposed future changes
- From: Aragon Gouveia <aragon@xxxxxxxxxxx>
- Date: Mon, 06 Jul 2009 14:53:39 +0200
Hi,
Rene Schickbauer wrote:
Yesterday i submitted a patch for powerd to set maximum allowed CPU speed
for adaptive modes (to keep the system cool and using less power).
Thank you for your work.
Would it also make sense to have powerd run an (optional) user configureable
script on ac state change? I'm thinking about things like dimming TFT
backlight, on EEE PC turning of the webcam and so on.
No. As pointed out, devd(8) does this already.
Another option that could make sense in powerd is checking the battery state
and running a user configureable script when ac-state is set to battery and
battery falls below a configured threshold. The script could do a number of
things like warning the user, scheduling a shutdown and so on in order to
give the user a fair chance to save his/her work and do a clean shutdown (or
just plugin the ac adapter).
As for the notification script at low capacity, in theory devd(8) should be able to do this as well. A notebook ought to generate a battery event when capacity hits its factory set "warning" and "low" marks. Of course I say this without having tested it on mine. :)
powerd currently only adjusts CPU speed, but having a *second* programm
monitor the same kernel variables to work on another part of the same
problem does not seem to make sense.
BTW, i'm also thinking of having the option to have powerd log the battery
status (ac mode + load + charge level) every 5 Minutes or so to syslog. That
way, a second script (log parser) may be able to determine information about
the battery - like how long does it take to charge, rough capacity
estimation and possible degradation of battery.
Seems unnecessary. That information is already available with acpiconf(8) and via sysctl(8) (see hw.acpi.acline, hw.acpi.battery), so someone is able to write their own logger if they wish.
IMHO, I think powerd should be kept doing only work that's most useful. As I've experienced with power saving, the moment you start doing more and more work to save power, you also end up consuming more power. Keep it simple.
Regards,
Aragon
_______________________________________________
freebsd-current@xxxxxxxxxxx mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to "freebsd-current-unsubscribe@xxxxxxxxxxx"
- References:
- RFC: powerd Patch & proposed future changes
- From: Rene Schickbauer
- RFC: powerd Patch & proposed future changes
- Prev by Date: Re: RFC: powerd Patch & proposed future changes
- Next by Date: Re: RFC: powerd Patch & proposed future changes
- Previous by thread: Re: RFC: powerd Patch & proposed future changes
- Next by thread: Re: RFC: powerd Patch & proposed future changes
- Index(es):
Relevant Pages
|