Dying battery

[Al Johnson]

On Wednesday 07 January 2009, Chris Samuel wrote:
> On Wed, 7 Jan 2009 2:26:00 pm The Digital Pioneer wrote:
> > GTA02 with SHR and mwester kernel. How often does mwester release a new
> > kernel? Maybe it could use updating...
>
> I think that's a userspace work around to issue:
>
> AT%SLEEP=2
>
> to the modem, not a kernel fix.

There is a parameter to control this in /etc/frameworkd.conf but I think SHR 
applies it by default. There is a section that should look something like 
this:

[ogsmd]
modemtype = ti_calypso
ti_calypso_deep_sleep = never

The second parameter controls whether ogsmd uses the AT%SLEEP=2 workaround 
permanently (deep_sleep = never) or if it uses it only when it detects #1024 
type behaviour (deep_sleep = auto? check the source...) In FSO the 
autodetection is default (parameter not present in config), but I think SHR 
adds the parameter to their config file.

I don't think this is the problem though. I have seen this happen while using 
the parameter above, and with debug logging of frameworkd enabled to confirm 
#1024 wasn't happening. I suspect either incorrect battery level indication 
in the GUI, some runaway process eating CPU, or some other cause of power 
hogging. Sadly I haven't been able to track it down yet.