Re: [rvm-research] Curious about OpenJDK on JikesRVM
Xi Yang <[email protected]>
| Newsgroups | gmane.comp.java.jikes.rvm.devel |
|---|---|
| Message-ID | <CAG2O1R1MW-haTufUj+TDGg=ZyLxDzhtObRD0cKBp=ACu4AcEpg@mail.gmail.com> |
Hi all, Here is the patch to allow us to replace fields and methods of the class when generating the bootimage. https://jira.codehaus.org/browse/RVM-1065 With the help of this patch, we could support OpenJDK without copying any file from OpenJDK repository. Regards. On 16 August 2013 17:45, Erik Brangs <[email protected]> wrote: > Hi, > > On 15.08.2013 20:58, Xi Yang wrote: >> I would like to drop all classes who should be included in JDK >> libraries, and as Maxine does, replace some methods and fields. >> However, I am not sure this is a good idea for GNU Classpath. >> Classpath may already assume any VM implementation would replace those >> classes in "libraryInterface/Common"? > > AFAIK the Classpath implementations of things like java.lang.Class just > defer to the corresponding stub class (i.e. VMClass in this case) so we > could just write the appropriate classes if we wanted to. I suppose that > there'll be a detailed discussion on this when the OpenJDK support is > ready for mainline so we don't have to worry about it for now. > >> Beside the interface, I think there are two points Jikes did not do it >> very well. One is the list of JDK classes in bootimage, another is the >> the list of classes who should be inited in certain order before >> dynamic loading becoming functional. I think this is a good chance to >> fix them all. Instead of trying many times, and add classes in the >> list manually, we can have a script to figure out candidates. We can >> put why we need to init class X, and the dependence between class >> X,Y,Z for JDK libraries. > > That would indeed be very useful. It would certainly reduce the time > needed to port to new versions of the class libraries. That being said, > I have not looked at this problem at all so I don't know how difficult > it is. > > The only thing I can say from memory is that you would likely need to > define a comprehensive set of test programs to run on each image to > ensure that all features work. For example, you will need to check that > the -Xbootclasspath options work (see e.g. > http://jira.codehaus.org/browse/RVM-887 for an issue relating to that). > > > Kind regards, > > Erik Brangs > > ------------------------------------------------------------------------------ > Get 100% visibility into Java/.NET code with AppDynamics Lite! > It's a free troubleshooting tool designed for production. > Get down to code-level detail for bottlenecks, with <2% overhead. > Download for free and get started troubleshooting in minutes. > http://pubads.g.doubleclick.net/gampad/clk?id=48897031&iu=/4140/ostg.clktrk > _______________________________________________ > Jikesrvm-researchers mailing list > [email protected] > https://lists.sourceforge.net/lists/listinfo/jikesrvm-researchers ------------------------------------------------------------------------------ Shape the Mobile Experience: Free Subscription Software experts and developers: Be at the forefront of tech innovation. Intel(R) Software Adrenaline delivers strategic insight and game-changing conversations that shape the rapidly evolving mobile landscape. Sign up now. http://pubads.g.doubleclick.net/gampad/clk?id=63431311&iu=/4140/ostg.clktrk