RFR: 8252003: remove usage of PropertyResolvingWrapper in vmTestbase/nsk/jvmti [v2]

Serguei Spitsyn sspitsyn at openjdk.java.net
Wed Sep 30 21:40:27 UTC 2020


On Wed, 30 Sep 2020 21:17:59 GMT, Igor Ignatyev <iignatyev at openjdk.org> wrote:

>> Marked as reviewed by sspitsyn (Reviewer).
>
>> What are the meaning of these properties ?:
>> -Dnsk.jvmti.test.property=value_of_nsk.jvmti.test.property
>> -Dnsk.jvmti.test.property=initial_value_of_nsk.jvmti.test.property
>> -Dnsk.jvmti.test.property.empty.new=initial_value_of_nsk.jvmti.test.property.empty.new
> 
> these are properties used by jvmti tests, the tests compare the actual value w/ expected values hardcoded in the test
> code. the purpose of these properties is different in different tests, ranging from just checking values by an agent to
> changing the value by the agent and checking that the value was changed.
>  
>> 
>> Could you, please, explain this change? :
>> 
>> 
>  ... scenarios/general_functions/GF08/gf08t001/TestDriver.java
> @@ -87,8 +87,8 @@ public static void main(String[] args) throws Exception {
>         nsk.jvmti.scenarios.general_functions.GF08.gf08t.main(new String[] {
>                 "gf08t001",
>                 nsk.jvmti.scenarios.general_functions.GF08.gf08t001.class.getName(),
>  -               keyPhrase,
>  -              "gc"});
>  +              "gc",
>  +              keyPhrase});
>     }
> }
> 
> it's the same changes I described [above](https://github.com/openjdk/jdk/pull/370#discussion_r497655837), in two words,
> due to how jtreg handles spaces and `"` in the arguments, the positions of `phrase` and  `verboseType` have been
> switched.

Okay, thanks!

-------------

PR: https://git.openjdk.java.net/jdk/pull/370


More information about the serviceability-dev mailing list