GGC and finding resources problem? [was: Re: Regression source found?]
"Grzegorz B. Prokopski" <[email protected]>
| Newsgroups | gmane.comp.java.vm.sablevm.devel |
|---|---|
| Organization | Debian http://www.debian.org - The Free Operating System |
| Message-ID | <[email protected]> |
On Sun, 2004-10-17 at 10:40, Laurent Martelli wrote: > >>>>> "Grzegorz" == Grzegorz B Prokopski <[email protected]> writes: > Grzegorz> Laurent, please test the latest staging version of sablevm > Grzegorz> (and you'll need new classpath too, I think) and compare > Grzegorz> performance of your app to what you had with 1.1.6. > > The current staging is good for me :-) Great! Thanks for all the testing you've done! > --with-gc=gengc does not work for me (with current staging, revision 2982): > > java.lang.NullPointerException > at gnu.java.lang.SystemClassLoader.systemFindResource (SystemClassLoader.java:268) > at gnu.java.lang.SystemClassLoader.findResource (SystemClassLoader.java:169) > at java.lang.ClassLoader.getResource (ClassLoader.java:760) > at gnu.java.lang.SystemClassLoader.findClass (SystemClassLoader.java:79) > at java.lang.ClassLoader.loadClass (ClassLoader.java:332) > at java.lang.ClassLoader.loadClass (ClassLoader.java:276) > at java.lang.VirtualMachine.main (VirtualMachine.java:83) Strange. Can you try to write a testcase that would trigger this behaviour and report it to BTS? Does it always happen and always in the same place or is it random? Is it the same with rev. 2960? Is it the same with Carl's GGC tag I merged? svn://svn.sablevm.org/developers/lebsack/tags/svm-ggc-automatic-heap/ (you'll need a matching classpath from around rev. 2949; 2960 should do) I hardly see a normal way a GC algorithm could influence finding resources, unless there's something broken... Cheers, Grzegorz B. Prokopski -- Grzegorz B. Prokopski <[email protected]> Debian GNU/Linux http://www.debian.org SableVM - LGPL'ed Java VM http://www.sablevm.org Why SableVM ?!? http://sablevm.org/wiki/Features