Re: Patch: FYI:

[Tom Tromey <tromey at redhat dot com>]

>>>>> "Ulrich" == Ulrich Weigand <Ulrich.Weigand@de.ibm.com> writes:

Ulrich> This does not catch an ArrayOutOfBounds exception thrown by classAt
Ulrich> (e.g. because the target does not define HAVE_BACKTRACE like on s390),
Ulrich> which causes initialization failures due to uncaught exceptions.

Thanks.  Andrew has an improved version of this patch which I'll check
in once the trunk is unfrozen (I prefer to put changes on the branch
and trunk simultaneously).

Ulrich> This results in a number of libjava testsuite regressions on
Ulrich> s390 and s390x.  Both CVS head and 3.3 branch are affected, it
Ulrich> seems.

Would you like to be the official s390 tester for the gcj 3.3 release?
There has been a lot of work (by you :-) to make gcj work on s390.  I
don't think I have access to an s390 (yet), and we need someone to run
tests to make sure that everything is ok on that platform.  Ideally
the testing would include Mauve, Jacks, and the other packages we care
about; see:

    http://gcc.gnu.org/java/gcj-3.3-status.html

If you can do this I'll add an entry (or two?) for s390 with your name
next to it.

Tom