RFR(S) - JDK-8013875: Incorrect vtable index being set during methodHandle creation for static methods

Bharadwaj Yadavalli bharadwaj.yadavalli at oracle.com
Mon May 6 11:43:02 PDT 2013


[Updated request for review]

Changes to address https://jbs.oracle.com/bugs/browse/JDK-8006267 
included modifications to methodHandles code.

John Rose pointed out that these changes are incorrect as static methods 
do not use vtable or itable indices. I made the necessary changes with 
his help.

I ran jtreg compiler, runtime and java/lang/invoke tests, JPRT, nashorn 
tests and vm.quick.testlist with no new failures.

Please review the changes.

Thanks,

Bharadwaj

JBS: https://jbs.oracle.com/bugs/browse/JDK-8013875
Webrev: http://cr.openjdk.java.net/~bharadwaj/8013875/webrev/

On 5/5/2013 11:27 PM, Bharadwaj Yadavalli wrote:
>
> On 5/5/2013 8:22 AM, Bharadwaj Yadavalli wrote:
>>
>> On 5/3/2013 11:53 PM, Christian Thalinger wrote:
>>> On May 3, 2013, at 6:32 PM, Bharadwaj Yadavalli 
>>> <bharadwaj.yadavalli at oracle.com> wrote:
>>>
>>>> I ran jtreg compiler and runtime; JPRT. 
>>> Since it changes method handle code you should at least run the 
>>> jtreg tests in jdk/test/java/lang/invoke and some Nashorn tests.
>>
>> How do I run Nashorn tests?
>>
>
> Never mind. I figured out. I ran both j.l.invoke (no new failures) and 
> nashorn test suite (no failures). I am running a few other tests as well.
>
> Will send an update once those tests complete successfully.
>
> Thanks,
>
> Bharadwaj
>
>



More information about the hotspot-compiler-dev mailing list