Re: ✘gpsd .23.2~rc1

[Greg Troxel]

"Gary E. Miller" <gem@rellim.com> writes:

> I just bumped the gpsd version in got to 3.23.2~rc1.  There is
> always more to do on gpsd, but this seems like a good pause point.

I have done a first pass of testing, all on NetBSD 9 amd64 with gcc
7.4.0.

Building, testing and distribution creation all work.

Building the created tarball with pkgsrc works, and tests pass with the
pkgsrc build.

With the installed package, gpsd works with a M8030 USB GPS mouse; I see
sensible content on xgpsd, gpspipe -r and -w, and I see a sensible SHM
offset in gpsd (ntpd fudge is set up for the wrong gps mouse, but it's
only 24 ms off, which I interpet as all ok, and yes I know about pps but
this receiver doesn't do that).

gpspipe -w to a file and gpsprof -r from that produces a sensible plot
(given a short time and a mouse antenna indoors not with a ground
plane).

I have not tried NTRIP.

cgps seems to not work; it prints
  
{"class":"VERSION","release":"3.23.2~rc1","rev":"3.23.2~rc1","proto_major":3,"proto_minor":15}
and then just sits there.  I don't really remember if this used to
work.  (I usually use xgps to look at gpspipe -w to log, for later
gpsprof, so I don't know if this is new.)

gpsmon prints VERSION, DEVICES, WATCH and then dumps core, but the crash
is in curses.  IMHO this should not hold the release (or even cause churn
to find it); it seems likely it's a local issue with curses lib.   But
if it crashes for a bunch of other people, that's interesting.


So if gpsmon and cgps work for most others, I'd say this is ok to
release.


I'm testing the package build on a NetBSD 9 RPI3, but it's starting to
build ruby so will report back later.


I will also kick off a build on netbsd-9 i386.

signature.asc
Description: PGP signature