GSM modem doesn't power off when Neo does

Graeme Gregory graeme at
Thu Feb 21 08:52:16 CET 2008

On Thu, 21 Feb 2008 01:35:00 -0600
"Mike (mwester)" <mwester at> wrote:
> If whomever is in charge of this decision is waiting for a set of 
> completely-debugged, fully-implemented patches, all thoroughly
> tested, and just waiting for that person to pick the one that feels
> the best to them -- well, that just 'ain't gonna happen.  Writing and
> debugging this stuff takes effort, especially things like adding
> signal handling to gsmd, or adding the code to the kernel; a little
> direction before anyone invests hours or even days to write and debug
> isn't too much to expect.
The distro team for this stuff, ie me/john knows little about gsmd
so we need some direction from people who do know gsmd. I cant just
check in any old stuff. I know OE well, not the intricacies of every
little detail of OM software.

> I'm feeling more than a little frustrated right now, and it's not
> just due to this one little item -- in general, the OM distro seems
> to be a huge unknown, and I feel that there's little attention being
> given to it right now -- and even less to those who attempt to
> contribute.
The OM distro isnt unknown, I am here, I sit on IRC all day in GMT
timezone. John is there quite often as well. The guys in #openmoko
know to shout at me online when they want something. One thing
to remeber is it is openmoko-devel-image we haven't started polishing
yet. I also have tasks that are not directly OE related so it
might take time for me to look at something.

> I mean, just how bleeding difficult is it to address this problem?  
> Really, how hard can it possibly be for somebody to say, "We of
> Openmoko select proposal X to solve this problem."   Why is there
> always somebody who jumps forward to say NO!! to a proposed solution,
> but there's nobody who can offer a suggestion on what they would
> accept? 

Do you have a solution that works? And isnt a patch to gsmd or kernel?
My policy is to use both those software direct from svn. Im quite happy
for you to put any solution that doesn't patch these in for
further testing. The discussion just seemed to lead round in circles
with eveyone claiming every solution didn't really work until the point
I have no idea. But if you have something that really does do something
its better than the nothing we have now and I would love to get that in for people. At least then we can improve/fix/re-implement
instead of running uphill everytime this discussion starts without end.


More information about the distro-devel mailing list