ARM Interpreter

[Andrew Haley]

Edward Nevill wrote:

> http://mail.openjdk.java.net/pipermail/distro-pkg-dev/2009-July/006543.html
> 
>> Matthias Klose wrote:
>>> +#if defined(PRODUCT) && defined(HOTSPOT_ASM)
>> So my first question is why is everything conditionalized on PRODUCT?
>> Not being able to build not-PRODUCT means no assertions, no debug
>> helpers and no diagnostic options.  I'd like to see all these
>> conditionals replaced with "#ifdef HOTSPOT_ASM".
>>
>>> +#endif
> 
> The reason everything is conditionalised on PRODUCT is that the Asm loop
> will not work with a non product build. This is because all the offsets in
> the structures change on a non product build because of the addition of
> debug info into the structures.
> 
> You can still do a non PRODUCT build for debug purposes and get all the
> assertions etc, its just you wont get the Asm loop at the same time, it will
> revert to the C++ loop.

We should be able to fix this so that it works with non-product builds
at some later stage.

>>> diff -r 914f2022b402 ports/hotspot/src/cpu/zero/vm/mkbc.c
>> This file isn't part of HotSpot, it's a generator for some other files.
>> Can it be placed into contrib?
> 
> I'm not sure where the best place for this is. The mkbc.c generates the mkbc
> tool which generates the bytecodes.s file from the bc.def file.
> 
> Ideally this would all be part of the build and the bytecodes.s file would
> not need to be checked in because it would be generated from bc.def.
> 
> However, currently I have included the bytecodes.s file manually generated
> because I am unclear how to handle the mkbc tool in the OpenJDK / IcedTea
> build.

IMO this is reasonable enough.

Andrew.