C++ Interpreter on 64-bit Linux

[Nagy Mostafa]

Hi Peng,
I had the same problem with the CPP interpreter. The alternative is to use
the Gary Benson's hotspot Zero port which is interpreter-only hotspot.

The way I do it is as follows:

./configure --enable-zero make icedtea-against-ecj

This will build the bootstrap VM only. Now you can edit the code in
"openjdk-ecj" and rebuild like this:


make hotspot


To do a clean:

rm -Rf openjdk-ecj/build/*/hotspot


then do "make hotspot" again


It has been a while since I tried all this, so things might have changed. I
hope this works for you though. More under:

http://markmail.org/search/?q=Nagy+Mostafa#query:Nagy%20Mostafa+page:1+mid:gqklt4gpv3izoqae+state:results


Thanks,

- nagy





On Fri, Jan 29, 2010 at 3:49 PM, Peng Du <imdupeng at gmail.com> wrote:

> Hi,
>
> Same error with 32-bit Linux and Intel Core2. Does that mean the code of
> C++ Interpreter on Linux is currently broken? Anyone knows a solution to
> get the C++ Interpreter working on Linux?
>
> BTW, I am using openjdk-7-b80.
>
> Thanks
>
>
>
> On Thu, 2010-01-28 at 12:25 -0600, Peng Du wrote:
> > Hello,
> >
> > Is HotSpot w/ C++ Interpreter currently working on 64-bit Linux
> > (x86_64)? I tried building HotSpot
> > with CC_INTERP=true but encountered this error:
> >
> > /openjdk/hotspot/src/cpu/x86/vm/interpreter_x86_64.cpp: In member
> > function 'unsigned char*
> > InterpreterGenerator::generate_abstract_entry()':
> > /openjdk/hotspot/src/cpu/x86/vm/interpreter_x86_64.cpp:282: error:
> > 'class InterpreterMacroAssembler' has no member named
> > 'empty_expression_stack'
> > /openjdk/hotspot/src/cpu/x86/vm/interpreter_x86_64.cpp:284: error:
> > 'class InterpreterMacroAssembler' has no member named 'restore_locals'
> > make[6]: *** [interpreter_x86_64.o] Error 1
> > make[6]: *** Waiting for unfinished jobs....
> > /openjdk/hotspot/src/cpu/x86/vm/methodHandles_x86.cpp: In static
> > member function 'static unsigned char*
> >
> MethodHandles::generate_method_handle_interpreter_entry(MacroAssembler*)':
> > /openjdk/hotspot/src/cpu/x86/vm/methodHandles_x86.cpp:95: error:
> > 'throw_WrongMethodType_entry' is not a member of 'Interpreter'
> > /openjdk/hotspot/src/cpu/x86/vm/methodHandles_x86.cpp: In static
> > member function 'static void
> > MethodHandles::generate_method_handle_stub(MacroAssembler*,
> > MethodHandles::EntryKind)':
> > /openjdk/hotspot/src/cpu/x86/vm/methodHandles_x86.cpp:404: error:
> > 'throw_WrongMethodType_entry' is not a member of 'Interpreter'
> > make[6]: *** [methodHandles_x86.o] Error 1
> > make[6]: Leaving directory
> > `/jdk-dist-64-cpp/hotspot/outputdir/linux_amd64_compiler2/product'
> > make[5]: *** [the_vm] Error 2
> > make[5]: Leaving directory
> > `/jdk-dist-64-cpp/hotspot/outputdir/linux_amd64_compiler2/product'
> > make[4]: *** [product] Error 2
> > make[4]: Leaving directory `/jdk-dist-64-cpp/hotspot/outputdir'
> > make[3]: *** [generic_build2] Error 2
> > make[3]: Leaving directory `/openjdk/hotspot/make'
> > make[2]: *** [product] Error 2
> > make[2]: Leaving directory `/openjdk/hotspot/make'
> > make[1]: *** [hotspot-build] Error 2
> > make[1]: Leaving directory `/openjdk'
> > make: *** [build_product_image] Error 2
> >
> >
> > Is there any patch to fix it?
> >
> >
> > Thanks
> >
> >
> >
> >
> >
> >
> >
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.openjdk.java.net/pipermail/hotspot-dev/attachments/20100129/a5a61116/attachment.html