RFR: 8143038: [TESTBUG] TestOptionsWithRanges: allow excluding only a subset of tested values specified for a flag

gerard ziemski gerard.ziemski at oracle.com
Fri Nov 20 18:37:43 UTC 2015


hi Dmitry,

Looks good!



On 11/20/2015 05:23 AM, Dmitry Dmitriev wrote:
> Hello,
>
> I implement Gerard suggestion about API names. Also, I update code to the hs-rt tip revision.
> webrev.01: http://cr.openjdk.java.net/~ddmitriev/8143038/webrev.01/
> <http://cr.openjdk.java.net/%7Eddmitriev/8143038/webrev.01/>
>
> Thanks,
> Dmitry
>
> On 19.11.2015 18:22, gerard ziemski wrote:
>> hi Dmitry,
>>
>> Looks good.
>>
>> They only feedback I have is that I'm not too crazy about the name "notTest*Range" for API. How about
>> "excludeTest*Range" instead?
>>
>>
>> cheers
>>
>> On 11/18/2015 02:00 PM, Dmitry Dmitriev wrote:
>>> Hello,
>>>
>>> Please review enhancement to the command line option validation test framework.
>>>
>>> Currently TestOptionsWithRanges test have several options excluded from testing, because these options with big or small
>>> values can cause problem with testing(create a huge number of threads or consume too much memory). This enhancement
>>> added ability to exclude max or min values from testing and TestOptionsWithRanges was corrected to use this enhancement.
>>>
>>> Also, in this fix I remove adding negative out-of-range values for the unsigned options, because negative values are
>>> actual invalid(not out-of-range) values for unsigned options.
>>>
>>> JBS: https://bugs.openjdk.java.net/browse/JDK-8143038
>>> webrev.00: http://cr.openjdk.java.net/~ddmitriev/8143038/webrev.00/
>>> <http://cr.openjdk.java.net/%7Eddmitriev/8143038/webrev.00/>
>>> Testing: tested on all platforms. One of the test was failed on win-64 platform, but this is not relate to this change.
>>> Filled https://bugs.openjdk.java.net/browse/JDK-8143249 for that(which seems related to the JDK-8069282)
>>>
>>> Thanks,
>>> Dmitry
>>>
>
>


More information about the hotspot-dev mailing list