pkg/28962: glib-2.6.1 fails test atomic-test on NetBSD/Sparc64

[Gert Doering <gert%kirk.greenie.muc.de@localhost>]

>Number:         28962
>Category:       pkg
>Synopsis:       glib-2.6.1 fails test atomic-test on NetBSD/Sparc64
>Confidential:   no
>Severity:       non-critical
>Priority:       medium
>Responsible:    pkg-manager
>State:          open
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Fri Jan 14 15:18:00 +0000 2005
>Originator:     Gert Doering
>Release:        NetBSD 2.0
>Environment:
        
        
System: NetBSD kirk 2.0 NetBSD 2.0 (KIRK) #2: Tue Jan 11 20:49:13 CET 2005 
gert@kirk:/home/sparc64/obj/home/src-2.0/sys/arch/sparc64/compile/KIRK sparc64
Architecture: sparc64
Machine: sparc64
>Description:
        when running the tests after compiling pkgsrc/devel/glib2
        (PKGSRC_RUN_TEST?=yes in /etc/mk.conf) testing fails one test
        with:

/usr/pkg/bin/gmake  check-TESTS
gmake[4]: Entering directory `/home/obj/pkg/devel/glib2/work/glib-2.6.1/tests'

** ERROR **: file atomic-test.c: line 53 (main): assertion failed: 
(g_atomic_pointer_compare_and_exchange (&atomic_pointer, NULL, biggest_pointer))
aborting...
[1]   Abort trap (core dumped) srcdir=. LIBCHAR...
FAIL: atomic-test

        all other tests are successful.

        The resulting glib2 seems to work (force-installed with 
        "PKGSRC_RUN_TEST?=no"), at least good enough so that "irssi"
        works - but still this test program core dump smells like 
        "deseaster waiting to happen" and the root cause should be 
        fixed.

>How-To-Repeat:
        just run "make" in pkgsrc/devel/glib2 on NetBSD/sparc64
>Fix:
        I don't really understand the code in question, but I assume that
        it might be related to 64/32 bit integer size mixups.

>Unformatted: