On Fri, Aug 04, 2006 at 08:23:55PM +0200, Julien Danjou <[EMAIL PROTECTED]> was
heard to say:
> Package: aptitude
> Version: 0.4.2-1
> Severity: serious
>
> Hello,
>
> There was a problem while autobuilding your package:
>
> > Automatic build of aptitude_0.4.2-1 on avidan by sbuild/i386 0.49
> > Build started at 20060804-1937
> > ******************************************************************************
> ...
> > g++ -g -O2 -I/usr/include/sigc++-2.0 -I/usr/lib/sigc++-2.0/include
> > -D_REENTRANT -o test main.o test_dense_setset.o test_eassert.o
> > test_misc.o test_resolver.o test_setset.o test_tags.o test_temp.o
> > test_threads.o test_wtree.o ../src/generic/apt/libgeneric-apt.a
> > ../src/generic/problemresolver/dummy_universe.o
> > ../src/generic/util/libgeneric-util.a ../src/vscreen/libvscreen.a -lcppunit
> > -lapt-pkg -lncursesw -lsigc-2.0 -lpthread
> > ../src/generic/util/libgeneric-util.a(temp.o): In function `impl':
> > /build/buildd/aptitude-0.4.2/src/generic/util/temp.cc:123: warning: the use
> > of `mktemp' is dangerous, better use `mkstemp'
> > make[3]: Leaving directory `/build/buildd/aptitude-0.4.2/tests'
> > /usr/bin/make check-TESTS
> > make[3]: Entering directory `/build/buildd/aptitude-0.4.2/tests'
> > ......................F.............
> >
> >
> > !!!FAILURES!!!
> > Test Results:
> > Run: 35 Failures: 1 Errors: 0
> >
> >
> > 1) test: TestThreads::testTimedTake (F) line: 117 test_threads.cc
> > assertion failed
> > - Expression: now2.tv_sec > now.tv_sec + 2 || now2.tv_usec >= now.tv_usec
This test basically relies on waiting for a condition variable, and
the failure you're seeing appears to indicate that
pthread_cond_timedwait is terminating in less time than requested. Two
questions:
(1) Is the failure 100% reproducible or is it intermittent?
(2) Does it happen with aptitude 0.4.1?
(3) If it's reproducible, can you trace through the code and see where
it fails? In particular, what value is being returned from
pthread_cond_timedwait?
Thanks,
Daniel
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
