On 2011-05-20 11.21, John Rose wrote:
> On May 19, 2011, at 10:06 PM, Charles Oliver Nutter wrote:
>
>> I've confimed that the i386 build performs just as poorly. Have not
>> investigate at inlining or assembly level yet...and it's late, so I'm
>> going to get some sleep.
>
> It's likely that method handle inlining is failing.
>
> Try -XX:+UnlockDiag* -XX:-UseRicochetFrames for differential testing.
>
> If you can find a microbenchmark which shows differential slowdown, we can
> debug it.
When trying the above all my tests fail with this:
java.lang.RuntimeException: java.lang.InternalError: NYI
at
seph.lang.interpreter.MessageInterpreter.evaluateFully(MessageInterpreter.java:107)
at seph.lang.Runtime.evaluateStream(Runtime.java:115)
at seph.lang.Runtime.evaluateString(Runtime.java:127)
at
seph.lang.code.BasicSanityTest.recursive_odd_and_even_that_should_blow_the_stack(BasicSanityTest.java:215)
Caused by: java.lang.InternalError: NYI
at
java.lang.invoke.AdapterMethodHandle.makeReturnConversion(AdapterMethodHandle.java:250)
at
java.lang.invoke.AdapterMethodHandle.makePairwiseConvert(AdapterMethodHandle.java:206)
at
java.lang.invoke.MethodHandleImpl.convertArguments(MethodHandleImpl.java:749)
at
java.lang.invoke.MethodHandleImpl$GuardWithTest.make(MethodHandleImpl.java:926)
at
java.lang.invoke.MethodHandleImpl.makeGuardWithTest(MethodHandleImpl.java:1066)
at
java.lang.invoke.MethodHandles.guardWithTest(MethodHandles.java:2072)
at java.lang.invoke.SwitchPoint.guardWithTest(SwitchPoint.java:145)
at
seph.lang.compiler.Bootstrap.initialSetup_intrinsic_if(Bootstrap.java:1298)
at seph$gen$abstraction$11.activateWith(Unknown Source)
at seph$gen$abstraction$11.activateWith(Unknown Source)
at seph.lang.ast.NamedMessage.go(NamedMessage.java:162)
at seph.lang.ast.NamedMessage.sendTo(NamedMessage.java:143)
at
seph.lang.interpreter.MessageInterpreter.evaluateFully(MessageInterpreter.java:92)
=(
Cheers
--
Ola Bini (http://olabini.com)
Ioke - JRuby - ThoughtWorks
"Yields falsehood when quined" yields falsehood when quined.
_______________________________________________
mlvm-dev mailing list
mlvm-dev@openjdk.java.net
http://mail.openjdk.java.net/mailman/listinfo/mlvm-dev
