This is the mail archive of the
java-patches@gcc.gnu.org
mailing list for the Java project.
Re: URLClassLoader update
- From: Mark Wielaard <mark at klomp dot org>
- To: java-patches at gcc dot gnu dot org
- Date: 01 Dec 2002 17:38:56 +0100
- Subject: Re: URLClassLoader update
- References: <1038267557.25590.63.camel@elsschot>
Hi,
On Tue, 2002-11-26 at 00:39, Mark Wielaard wrote:
> With this patch all new relevant Mauve URLClassLoader tests pass.
>
> There seems to be one regression though in the libjava.loader testsuite.
> FAIL: TestEarlyGC.exe output
> Which prints nine times "finalized" on my system and according to the
> TestEarlyGC.out should not print anything. The problem is that the
> actualy code is obfuscated (there is a big byte array inside the source
> that represents the byte code of a class). Is the actual java source
> somewhere? That would make debugging a bit easier.
>
> I currently don't have an old build around to check against (disk space
> is low at the moment), but on my other (powerpc) this test does not
> fail. I did notice that this test seems to fail for some people e.g.
> http://gcc.gnu.org/ml/gcc-testresults/2002-11/msg00925.html
> http://gcc.gnu.org/ml/gcc-testresults/2002-11/msg00857.html
>
> So if this isn't actually a regression, would it be OK to check this in?
I finally tested the changes on my powerpc machine and the results can
be found here:
http://gcc.gnu.org/ml/gcc-testresults/2002-12/msg00015.html
Note that the TestEarlyGC failure is not a regression on powerpc.
It failed already a while back:
http://gcc.gnu.org/ml/gcc-testresults/2002-11/msg01092.html
I would still like to see the actual TestEarlyGC source code though.
There does seem to be one regression (Mauve test):
FAIL: gnu.testlet.java.util.ResourceBundle.getBundle: with locale of France (number 5)
But this seems to be a bug in ResourceBundle which caches results
according to the ClassLoader used but doesn't take into account that the
requested or default) Locale can change. I will look into this.
Any comments on the actual patches and wether it is OK to check them in
is appreciated.
Cheers,
Mark