Re: Problems with gpsd under BullsEye

[Don Rolph]

As to your comment:

"> The gpsd 3.23.2 build environment is incompatible with the darlier

> development code.

Huh?  That would be news to me.  As part of gpsd CI, there are pipelines
running a wide variety of tool chains."

see error messages below:

[ 54%] Linking C executable il2p_test
/root/direwolf/src/dwgpsd.c:64:2: error: #error libgps API version might be incompatible.
   64 | #error libgps API version might be incompatible.
      |  ^~~~~
make[2]: *** [src/CMakeFiles/atest.dir/build.make:433: src/CMakeFiles/atest.dir/dwgpsd.c.o] Error 1
make[2]: *** Waiting for unfinished jobs....
[ 54%] Built target il2p_test
make[1]: *** [CMakeFiles/Makefile2:1670: src/CMakeFiles/atest.dir/all] Error 2
make: *** [Makefile:182: all] Error 2
root@atgp4:~/direwolf/build# ^C
root@atgp4:~/direwolf/build#

On Fri, Dec 31, 2021 at 2:09 PM Gary E. Miller <gem@rellim.com> wrote:

Yo Don!

On Fri, 31 Dec 2021 11:52:51 -0500
Don Rolph <don.rolph@gmail.com> wrote:

> I have been poking to try to understand some instabilities I have
> observed with Dire Wolf using gpsd under Raspbian BullsEye.

Not much we can do for 18 month old software.

> - a flood of packets through the sound card when achieving gps fix
> with the MT3333 GPS

How does a packet o through a sound card?

> - an apparent failure to update the gps coordinates used by Dire Wolf
> when using a u-blox GPS puck

What is Dire Wolf?

> As noted below, if I substitute the gpsd 3.23.2 version for the
> default BullsEye gpsd 3.22-4 version all issues I have observed have
> apparently been resolved.

So file a bug report with the BullsEye people.

> I will include this in my OS build instructions for BullsEye.

Is there something wrong with the existing build instructions?

> I successfully build a gpsd V3.23.2 executable.
>
> I then use apt-get install to add the gpsd components.
>
> I then replaced the /usr/sbin/gpds v3.22-4 with the gpsd 3.23.2
> executable.

Mixing binaries is not a good idea.  In some modes of opteration, gsd
passes paccked binary structirs to clients.  THat is not stable over
tool chains.

> - the autobauding problem for 9600 baud gps in V3.22-4 is fixed in
> gpsd 3.23.2

We knew that.

> The gpsd 3.22-4 executable seems to have instabilities.

Not our bonary.  You need to report that to the people that built it.

> The gpsd 3.23.2 build environment is incompatible with the darlier
> development code.

Huh?  That would be news to me.  As part of gpsd CI, there are pipelines
running a wide variety of tool chains.

RGDS
GARY
---------------------------------------------------------------------------
Gary E. Miller Rellim 109 NW Wilmington Ave., Suite E, Bend, OR 97703
        gem@rellim.com  Tel:+1 541 382 8588

            Veritas liberabit vos. -- Quid est veritas?
    "If you can't measure it, you can't improve it." - Lord Kelvin

--

73,

AB1PH

Don Rolph