RFR(S): 8150899 - [TESTBUG] Split hotspot_all job into smaller jobs
Dmitry Fazunenko
dmitry.fazunenko at oracle.com
Fri Apr 1 11:50:24 UTC 2016
Christian,
Moving tests (renaming) is very bad practice, because we lose the
history associated with the tests in JIRA and in the execution system.
Fortunately the story of stress/gc is not so long and we can sacrifice
it. So, I'm okay with moving stress/gc to gc/stress.
Thanks,
Dima
On 31.03.2016 22:10, Christian Tornqvist wrote:
> Hi Dima,
>
> I merged with your change for 8152422. The tests in stress/gc should be
> moved to gc/stress instead, I'll file a bug for this.
>
> Thanks,
> Christian
>
> -----Original Message-----
> From: Dmitry Fazunenko [mailto:dmitry.fazunenko at oracle.com]
> Sent: Thursday, March 31, 2016 12:53 PM
> To: Christian Tornqvist <christian.tornqvist at oracle.com>;
> hotspot-dev at openjdk.java.net
> Subject: Re: RFR(S): 8150899 - [TESTBUG] Split hotspot_all job into smaller
> jobs
>
> Hi Christian,
>
> Please note, that hotspot_gc has been split up recently:
> https://bugs.openjdk.java.net/browse/JDK-8152422
> so, you need to adjust a little your change.
>
> GC tests besides gc/ also comprise stress/gc/ Please update the hostpot_gc
> group to include stress tests as well.
>
> Other changes looks good to me.
>
> Thanks,
> Dima
>
>
>
>
> On 29.03.2016 12:53, Christian Tornqvist wrote:
>> Hi everyone,
>>
>>
>>
>> Please review this small change that splits the hotspot_all group used
>> in our testing into smaller batches. Since the
>> hotspot_compiler/gc/etc.. names were already used by JPRT but didn't
>> reflect a complete set of tests for that component area, these groups
>> have been renamed to hotspot_fast_compiler etc to reflect the intent
>> of these being a faster subset of the subcomponent tests.
>>
>>
>>
>> Webrevs:
>>
>> http://cr.openjdk.java.net/~ctornqvi/webrev/8150899/webrev.hotspot.00/
>>
>> http://cr.openjdk.java.net/~ctornqvi/webrev/8150899/webrev.root.00/
>>
>>
>>
>> Bug is unfortunately not visible externally.
>>
>>
>>
>> Thanks,
>>
>> Christian
>>
>>
>>
>
More information about the hotspot-dev
mailing list