[geos-devel] build/test report on NetBSD/sparc

[Greg Troxel]

Sandro Santilli <strk at keybit.net> writes:

> Great. If you want to find the actual culprit you'll see
> that testrunner is a shell script calling XMLTester with a long
> argument list. Maybe you could reduce the list to find the offending
> test and remove it ?

It's so painful to build, but I'll see how it goes.   Modern software
just doesn't fit in 24 MB of RAM :-(

It would be nice if XMLTester got invoked once per test and there was
clearer intermediate status.  But perhaps my case is degenerate.  (Are
there test results from Raspberry Pi, Beaglebone, etc.?  Of course those
have lots of RAM compared to retrocomputing.)

> Also, did the tests under test/unit all succeed ?

I didn't have cunit installed - starting the build of that now.

> Probably. Would you be interested in setting that machine up as a build
> slave for geos, shall the osgeo buildbot ever get back up ?
> http://trac.osgeo.org/osgeo/ticket/1214

Sorry, can't do that (policy issues), plus it's way too slow.   I think
it took 4 days for the first build (with ccache, so the 2nd one wasn't
so bad).
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 194 bytes
Desc: not available
URL: <http://lists.osgeo.org/pipermail/geos-devel/attachments/20130923/10b07f93/attachment.pgp>