Re: Oracle/Sun and the JVM

2010-01-31 Thread Francis ANDRE
Hi Stepan Sure... I was just thinking of my own requirements... but if the bytecode is extended, it should be done in a way to provide solution for the most of us and specially what you are requiring... I am also requesting the support of unsigned integral (for the JVM not for Java!) and the

Re: more from yesterday

2010-01-31 Thread Charles Oliver Nutter
On Thu, Jan 28, 2010 at 7:56 PM, Fredrik Öhrström wrote: > Thread local garbage collection is a JRockit bleeding edge feature that > allow separate threads in a Java server to have their own heap. There is > of course also a global heap to which shared objects are relocated to. > One important ben

Re: indy and tailc

2010-01-31 Thread Charles Oliver Nutter
On Wed, Jan 27, 2010 at 10:16 AM, John Rose wrote: > They also provide a way to do something like bytecode-level templating, since > CP elements can be varied while the rest of the code is held fixed.  I hope > somebody experiments with loop customization at some point, using anonymous > classe

Re: hg: mlvm/mlvm/hotspot: dynopt, trustfinal: update to match JDK7 pushes

2010-01-31 Thread Charles Oliver Nutter
I think I lost track of patches for a while...trustfinal is the optimization to not repeatedly access finals, correct? I'll have a look at the patch, but perhaps a description of when final accesses will be elided? On Wed, Jan 27, 2010 at 8:42 AM, wrote: > Changeset: 2b8bce60ef94 > Author:    jr