RFR(xs): 8224710: [TESTBUG] dynamicArchive/ExcludedClasses.java test failed in tier6 testing on Windows
Ioi Lam
IOI.LAM at ORACLE.COM
Wed Jun 12 17:11:25 UTC 2019
Sounds good. Thanks
Ioi
> On Jun 12, 2019, at 8:51 AM, Calvin Cheung <calvin.cheung at oracle.com> wrote:
>
> Hi Jiangli, Ioi,
>
> I've changed it to "DynamicDumpSharedSpaces is unsupported ..." - using the word "unsupported" instead of "not supported".
>
> updated webrev: http://cr.openjdk.java.net/~ccheung/8224710/webrev.01/
>
> thanks,
> Calvin
>
>> On 6/12/19, 8:37 AM, Jiangli Zhou wrote:
>>> On Tue, Jun 11, 2019 at 9:32 PM Ioi Lam<ioi.lam at oracle.com> wrote:
>>> Looks good to me.
>>>
>>>> On 6/11/19 7:26 PM, Jiangli Zhou wrote:
>>>> Hi Calvin,
>>>>
>>>> Test fix looks reasonable. It would be a good idea to also fix the
>>>> original error message and the check in the test so it's grammatically
>>>> more correct:
>>>>
>>>> "DynamicDumpSharedSpaces is not supported ..."
>>>> ^^
>>> The "is" is often omitted in HotSpot's warning messages for conciseness.
>>> Such usage is commonplace in news headlines, etc.
>>>
>>> $ grep -R not.supported | grep -v 'is not' | grep -v assert
>> That doesn't appear to be a good reason to not fix this. There are
>> many places in hotspot use proper grammar. It's not worth the debate.
>>
>> Thanks,
>> Jiangli
>>> Thanks
>>> - Ioi
>>>
>>>> Best Regards,
>>>> Jiangli
>>>>
>>>>> On Tue, Jun 11, 2019 at 4:42 PM Calvin Cheung<calvin.cheung at oracle.com> wrote:
>>>>> bug: https://bugs.openjdk.java.net/browse/JDK-8224710
>>>>>
>>>>> webrev: http://cr.openjdk.java.net/~ccheung/8224710/webrev.00/
>>>>>
>>>>> The test failure was observed on Windows once when the loading of the
>>>>> base archive failed. A simple fix is in CDSTestUtils::isUnableToMap() to
>>>>> account for the failure in loading of the base archive.
>>>>>
>>>>> I haven't been able to reproduce the bug so far but was able to simulate
>>>>> the failure condition by renaming the base archive prior to running the
>>>>> test.
>>>>>
>>>>> Testing: ran the test on Windows and Linux machines many times.
>>>>>
>>>>> thanks,
>>>>> Calvin
More information about the hotspot-runtime-dev
mailing list