|
From: Bart V. A. <bar...@gm...> - 2008-03-03 11:18:21
|
On Mon, Mar 3, 2008 at 12:06 PM, Tom Hughes <to...@co...> wrote: > In message <e2e...@ma...> > > > Bart Van Assche <bar...@gm...> wrote: > > > On Mon, Mar 3, 2008 at 11:42 AM, Tom Hughes <th...@cy...> wrote: > >> > >> Nightly build on dellow ( x86_64, Fedora 8 ) started at 2008-03-01 03:10:04 GMT > >> Results differ from 24 hours ago > >> ... > >> pth_broadcast: valgrind ./pth_broadcast > >> sh: line 1: 19837 Killed VALGRIND_LIB=/tmp/vgtest/2008-03-01/valgrind/.in_place VALGRIND_LIB_INNER=/tmp/vgtest/2008-03-01/valgrind/.in_place /tmp/vgtest/2008-03-01/valgrind/./coregrind/valgrind --command-line-only=yes --memcheck:leak-check=no --tool=exp-drd ./pth_broadcast > pth_broadcast.stdout.out 2> pth_broadcast.stderr.out > >> *** pth_broadcast failed (stderr) *** > > > > Hello Tom, > > > > I haven't seen this test being killed ever before. Could this be > > caused by the regression test framework ? If not, can you please post > > pth_broadcast.stderr.out ? > > It was killed by me because I had about five days worth of valgrind > tests hanging around chewing cpu on that machine. > > It doesn't actually seem to be hanging under manual testing at the > moment. That's unfortunate. I have added the statement "alarm(100);" inside pth_broadcast.c, such that the test is aborted if it did not complete within 100s. Bart. |