RFR (S): 8211065: Private method check in linkResolver is incorrect

David Holmes david.holmes at oracle.com
Wed Oct 10 00:18:40 UTC 2018


Thanks for the review Lois!

David

Lois wrote:
> On 10/7/2018 11:10 PM, David Holmes wrote:
> 
>> webrev: http://cr.openjdk.java.net/~dholmes/8211065/webrev/
>> bug: https://bugs.openjdk.java.net/browse/JDK-8211065
>>
>> I had added a check, as part of the nestmates works, to ensure that 
>> the selected and resolved methods, in the case of private methods, 
>> were the same. That check was both in the wrong place and ill-formed.
>>
>> The check has now been deleted.
>>
>> A nestmate test that incorrectly expected NoSuchMethodError when a 
>> private method in a subclass was deleted (but still called) has been 
>> updated to expect the superclass method to be invoked.
>>
>> A new test was added that previously highlighted an incorrect 
>> assertion within the check. Now it serves as a non-nestmate case of 
>> deleting a subclass method.
>>
>> This will be backported to 11 as well.
>>
>> Thanks,
>> David
> 
> Hi David,
> 
> I've looked at the latest JVM 11 spec and agree, this check should be 
> removed.  Looks good.
> 
> Thanks,
> Lois



More information about the hotspot-runtime-dev mailing list