Re: [gpsd-dev] OS X dropped. How I think about our platform priorities

[Eric S. Raymond]

Frank Nicholas <address@hidden>:
> On Wed, Feb 18, 2015 at 1:31 PM, Greg Troxel <address@hidden> wrote:
> > Your characterization of what's going on seems off; see my recent mail.
> >
> > And I don't think there's any justification for claiming that this is
> > necessarily a mac bug, even though I think it's 90% likely.

Greg: I never received a patch from you with an updated diagnosis.  I would
like to have it, but don't intend to hold the release for that purpose. Not
when *you* say it's 90% likely an OS X bug.

> I did a clean install of Ubuntu Desktop LTS on a PC that has Intel WiFI &
> Bluetooth built-in.  Ubuntu was patched to current.  I cloned and built the
> current gpsd.  After working through the Linux bluetooth BS, my results on
> the Linux PC match ** EXACTLY ** my results on the Mac.  My issues are NOT
> related to running on Mac OS X.

I believe you - the bug entry says "This bug has been observed on both
USB and Bluetooth".  However, Greg Troxel's bug report *does* seem to
be Mac-specific.  We have *a* Mac bug even if it's not your Mac bug,
and as Gary Miller points out we have nobody with strong enough Mac-fu 
to fix it.
 
> My issue may be related to the documented (in gpsd docs) issue with some
> bluetooth GPS receivers.  To get my bluetooth GPS to work at all, I seem to
> have to remove power form the GPS (no battery in the GPS), wait (several
> minutes), and possibly power down the Ubuntu system - not reboot (this step
> is not confirmed to be required, as I may have not been waiting long enough
> after the GPS power removal).

Then there is something very simple you can try, which is what I would do:

1. Start with the device and Ubuntu reset to a clean state.

2. Launch gpsd with the Bluetooth device on the command line and the -b
   option to prevent autobauding.

3. Observe it operating.

4. Power down the device and terminate gpsd

5. Repeat step 2.

If you find the device comes up on step 5, then we know that -b is an
effective workaround and we can finger bad Bluetooth firmware as the 
cause of the problem.
-- 
                <a href="http://www.catb.org/~esr/";>Eric S. Raymond</a>