Subject: Re: Documentation on power management
To: Ken Nakata <kenn@synap.ne.jp>
From: Ken Hornstein <kenh@cmf.nrl.navy.mil>
List: current-users
Date: 01/07/1999 16:36:04
(Can you tell I'm behind?)

>Great work!  I just like to comment on a discripancy between this page 
>and apmd(8) man page, and request for feature addition to apmd.
>
>Apmd(8) man page says:
>
>     apmd periodically polls the APM driver for the current power state.  If
>     the battery charge level changes substantially or the external power sta-
>     tus changes, the new status is logged.  The polling rate defaults to once
>     per 10 minutes, but may be specified using the -t command-line flag.
>
>but your web page says:
>
>    2. The APM kernel driver gets the request event the next time it
>       polls the APM BIOS (typically once per second) and queues the
>       standby/suspend request to the apm daemon.
>[...]
>Or, are these talking about two entirely different polling periods?

They're talking about two different polling periods.

The _kernel_ has to poll the APM BIOS for status changes.  When it
sees a status change, it (in some cases) queues it to the apm device,
and in some cases just knows about it (I might be wrong on particulars
here, but you get the idea).

apmd listens on the device (which is the interface to the _kernel_, not
the BIOS) for power status changes and also does an occasional poll as well.

I believe that power status changes (such as line->AC transisions) are
sent immediately to apmd, where battery level changes are not (and that's
what apmd really polls for).

>It would be great if we could set logging threshholds for battery
>charge level and estimated battery life.
>[...]
>So, I think there should be user-configurable logging options, maybe
>not just thresholds for battery charge level and estimated battery
>life but even more flexible options (e.g. one might want to disable
>apmd logging entirely, or another might not want any message until
>battery level goes down to "low" after which point much more frequent
>update on battery status is desired, etc.).

I think this sounds reasonable to me.  You have my permission to work
on it :-)

(If you don't have CVS access, I'll work with you to get any changes you
make into the tree).

--Ken