[8u] RFR: JDK-8141551: C2 can not handle returns with incompatible interface arrays
Volker Simonis
volker.simonis at gmail.com
Mon Apr 18 16:29:59 UTC 2016
On Mon, Apr 18, 2016 at 6:17 PM, Vladimir Kozlov
<vladimir.kozlov at oracle.com> wrote:
> On 4/18/16 5:39 AM, Volker Simonis wrote:
>>
>> Hi Shafi,
>>
>> please not that I'm not a formal jdk8u reviewer.
>>
>> The change to the source files looks good. Where did the change not
>> applied cleanly?
>
>
> Looks like it is new() vs new(Compiler) Node allocation known difference.
> I verified that changes are matching jdk 9 changes except new() difference.
>
Hi, Shavi, Vladimir,
thanks for the explanation. Then I think everything looks fine if you
verify that the JTREG test are passing.
Regards,
Volker
>>
>> I'm only a little worried about the test because it uses the
>> '@modules' tag. But if it runs and executed fine with jdk8u it's fine.
>
>
>
> The test should not have @modules for jdk8u since it is not supported there.
>
> JPRT does not run jtreg tests when testing jdk8u sources (only one very
> simple ExecuteInternalVMTests.java):
>
> http://hg.openjdk.java.net/jdk8u/jdk8u-dev/hotspot/file/d025821b6b6d/test/TEST.groups#l131
>
> You should run jtreg tests yourself using your jdk8u as test jdk.
>
> Please, verify results of your testing before pushing.
>
> Thanks,
> Vladimir
>
>
>>
>> Thanks for backporting this to jdk8,
>> Volker
>>
>>
>> On Mon, Apr 18, 2016 at 12:26 PM, Shafi Ahmad <shafi.s.ahmad at oracle.com>
>> wrote:
>>>
>>> Hi,
>>>
>>>
>>>
>>> Please review the backport of bug: "JDK-8141551: C2 can not handle
>>> returns with incompatible interface arrays" to jdk8u.
>>>
>>> Please note the backport is not clean.
>>>
>>>
>>>
>>> Webrev: http://cr.openjdk.java.net/~rpatil/8141551/webrev.00/
>>>
>>> Jdk9 bug: https://bugs.openjdk.java.net/browse/JDK-8141551
>>>
>>> Original patch pushed to jdk9:
>>> http://hg.openjdk.java.net/jdk9/jdk9/hotspot/rev/b425a78e8512
>>>
>>>
>>>
>>> Test: Run jprt.
>>>
>>>
>>>
>>> Regards,
>>>
>>> Shafi
>>>
>>>
>
More information about the hotspot-dev
mailing list