review request (L): 6939861: JVM should handle more conversion operations

[John Rose]

On May 5, 2011, at 11:58 PM, Christian Thalinger wrote:

> On May 5, 2011, at 1:16 PM, John Rose wrote:
>> I have finished the last large chunk of JVM work for JDK 7 JSR 292, the implementation of so-called "ricochet frames".  Here it is for review:
>> 
>> 6939861: JVM should handle more conversion operations
>> http://cr.openjdk.java.net/~jrose/6939861/webrev.04/
> 
> src/share/vm/code/codeBlob.cpp:
> 
> In DeoptimizationBlob::create you call trace_new_stub but the code that code factored out is still there.

Thanks.

> src/cpu/x86/vm/methodHandles_x86.cpp:
> 
> +          // This opens space space for the return value.
> 
> One space too much.

Fixed.

> +      __ mov(saved_last_sp, rsp);    // set rsi/r13 for calleee
> 
> One e too much.

Fixed.

> All new header files have the wrong (Sun) copyright header.

I'm backing off from introducing new header files.  The only new file is methodHandles_x86.hpp.

See what you think.  The idea is to allow each port to control its own individual future relative to this refactoring.  The logic is toward the bottom of this diff:

  http://cr.openjdk.java.net/~jrose/6939861/webrev.05/src/share/vm/prims/methodHandles.hpp.udiff.html

Also, this new webrev is relative to hotspot-comp, not bsd-port:

  http://cr.openjdk.java.net/~jrose/6939861/webrev.05

Thanks.  I'm hoping we can put this in today, especially with the reduced dependency on port changes.

-- John