no stacktrace available

[Boehm, Hans]

If I read the configuration code correctly, we silently
configure the GC without thread support on systems that
aren't explicitly handled by configure.ac.  I suspect this
will cause "make check" in the gc directory to successfully
run the single-threaded gc, but will cause libgcj to fail
for anything nontrivial.

I don't think that's what we should be doing.  I think the
case statement for "pthreads" should have an "*)" clause
which just fails.

I now changed this in the upstream source.

Hans

> -----Original Message-----
> From: tromey@redhat.com [mailto:tromey@redhat.com]
> Sent: Wednesday, December 15, 2004 9:37 AM
> To: Andrew Haley
> Cc: Boehm, Hans; Bart Locanthi; Michael Koch; java@gcc.gnu.org
> Subject: Re: no stacktrace available
> 
> 
> >>>>> "Andrew" == Andrew Haley <aph@redhat.com> writes:
> 
> Andrew> The libgcj tests are designed to test, as far as 
> possible, just one
> Andrew> thing like division, array access, and so on.  Maybe 
> we should also
> Andrew> have a smoke test for the gc.
> 
> For some reason we don't seem to regularly run "make check" in the GC
> build directory.  Seems like that would work for this.
> 
> Tom
>