Re: [rvm-research] Recompilation of Method
Eliot Moss <[email protected]>
| Newsgroups | gmane.comp.java.jikes.rvm.devel |
|---|---|
| Message-ID | <[email protected]> |
On 5/12/2015 7:32 AM, Abhinav Jangda wrote: > Hello Everyone, > > I am doing a research project over JikesRVM. For this I need to recompile a method whenever it is > called again. As JikesRVM has a JIT, so I think there must be some place where JikesRVM is getting > the compiled code of method from its cache and calling it. Can anyone please tell me where this > thing is happening? > Moreover, I am working with BaselineCompiler not with OptCompiler. The "cache" is not probed on each method call. Rather, a virtual dispatch is done (for instance methods) or the address of a static method is obtained, and the method is called. I don't recall whether the Baseline compiler hard codes calls to static methods or not. For methods that are not yet loaded / compiled, the compiler puts in a call to as stub that forces the necessary loading and compiling actions and then redirects to the compiled method, fixing up so that future calls do not need to go through the stub. So, in effect, there is not really a cache in the sense I take from you email. However, it is easy to replace the addresses of methods in JikesRVM's equivalent of vtables ... Regards -- Eliot Moss ------------------------------------------------------------------------------ One dashboard for servers and applications across Physical-Virtual-Cloud Widest out-of-the-box monitoring support with 50+ applications Performance metrics, stats and reports that give you Actionable Insights Deep dive visibility with transaction tracing using APM Insight. http://ad.doubleclick.net/ddm/clk/290420510;117567292;y