XS review: 8098557: Eager initialization of a super interface with a default method (original) (raw)

harold seigel harold.seigel at oracle.com
Fri Nov 20 23:40:47 UTC 2015

I like choice b).

Thanks, Harold

On 11/20/2015 4:55 PM, Karen Kinnear wrote:

Harold -

Thank you for the quick review! The text not in the () is cut-and-pasted from JVMS 5.5. The () is the clarification. Would you prefer: a) that I reference JVMS 5.5 Initialization or b) the following text? "If C is an interface that contains a default method, it requires initialization. This initialization is triggered by initialization of a dependent subclass, but not by initialization of a dependent sub interface. See JVMS 5.5” thanks, Karen

On Nov 20, 2015, at 4:41 PM, harold seigel <harold.seigel at oracle.com> wrote:

Hi Karen, Your changes look good, but I didn't quite understand this new comment in instanceKlass.cpp: + // If C is an interface that declares a non-abstract, non-static method, + // the initialization of a class (not an interface) that implements C directly or + // indirectly. Could you re-phrase it? (No need to see a new webrev.) Thanks, Harold

On 11/20/2015 2:29 PM, Karen Kinnear wrote: Please review fix for 8098557: Summary: Initialization of a super interface with a default method should be triggered by initialization of a dependent subclass, but not of a dependent subinterface. open webrev at http://cr.openjdk.java.net/~acorn/8098557/webrev bug link at https://bugs.openjdk.java.net/browse/JDK-8098557 tests: jprt reported example, new test all platforms: vm quick & non-cool linux-x64: defmeth, SelectionResolution, jtreg hotspot, hotspot/closed, compiler, compiler/closed jtreg java/util, jtreg tools/javac/lambdaShapes jtreg jdk in progress jcks - working on thanks, Karen

More information about the hotspot-runtime-dev mailing list