Re: Tracking performance regressions in 1.1.7-pre
Laurent Martelli <[email protected]>
| Newsgroups | gmane.comp.java.vm.sablevm.devel |
|---|---|
| Message-ID | <[email protected]> |
>>>>> "Grzegorz" == Grzegorz B Prokopski <[email protected]> writes: [...] >> Revision 2934 of classpath does not compile: >> >> laurent@stan:~/src/sablevm-classpath-staging+2934$ make Making >> all in lib make[1]: Entering directory >> `/home/laurent/src/sablevm-classpath-staging+2934/lib' >> top_builddir=.. /bin/bash ./gen-classlist.sh standard >> /usr/bin/gcj --bootclasspath '' --classpath ..:../vm/current:.: >> -C -d . @classes Grzegorz> ^^^^^^^^^^^^^^ >> ../java/lang/Class.java:1193: error: Duplicate variable >> declaration: `String name' was `<unknown> name' (<built-in>:0). >> private String name; Grzegorz> You have to use jikes. GCJ has never been guaranteed to Grzegorz> work. There's a configure switch to force which compiler Grzegorz> to use. OK it compiles now. If you have not already done that, I think it would be a good idea to make jikes the default compiler. The result is that 2934 is OK for performance. My test application is an AOP framework, so there's a custom classloader and lots of java.lang.reflect uses. If that can give a you a clue about where to search. -- Laurent Martelli [email protected] Java Aspect Components http://www.aopsys.com/ http://jac.objectweb.org