RFR(T) : 8244491 : make runtime/cds/appcds/TestZGCWithCDS.java test more robust

David Holmes david.holmes at oracle.com
Wed May 6 00:54:34 UTC 2020


Hi Igor,

This seems fine to me. Thanks for fixing it so promptly!

David

On 6/05/2020 10:41 am, Igor Ignatyev wrote:
> http://cr.openjdk.java.net/~iignatyev//8244491/webrev.00
>> 30 lines changed: 4 ins; 20 del; 6 mod;
> 
> 
> Hi all,
> 
> could you please review this patch for TestZGCWithCDS.java test?
> from JBS:
>> runtime/cds/appcds/TestZGCWithCDS.java test makes assumptions about default GC behavior, in some cases, this might lead to false-positive results, so it's better to explicitly select GC whose behavior satisfies the needs of the test.
> the patch updates the test to explicitly set GC in all spawned JVMs, in cases where we used to use a default GC, we now use SerialGC. this effectively made 'Run with ZGC turned off' the equivalent of 'Run with +UseCompressedOops +UseCompressedClassPointers', so the former got removed.
> 
> JBS: https://bugs.openjdk.java.net/browse/JDK-8244491
> webrev: http://cr.openjdk.java.net/~iignatyev//8244491/webrev.00
> testing: TestZGCWithCDS.java test
> 
> Thanks,
> -- Igor
> 
> 


More information about the hotspot-runtime-dev mailing list