15328602ac817be9cf1fc476988394416e797876 - platform/libcore

commit	15328602ac817be9cf1fc476988394416e797876	[log] [tgz]
author	Richard Uhler <ruhler@google.com>	Tue Jan 05 15:09:00 2016 -0800
committer	Richard Uhler <ruhler@google.com>	Tue Jan 26 14:59:50 2016 -0800
tree	645768c92b2a643f36939338c91e38db4bf9dd8c
parent	88a650cb86ae25465e7974edbb1e18bf6cf35bb5 [diff]

Improve performance of Reference processing.

By using synchronization on the ReferenceQueue instead of volatile
modifiers, and adding internal support for ReferenceQueue batch
enqueue and remove.

ReferenceBenchmark improvements on hammerhead:
                            Alloc   250 =>  150
                  AllocAndEnqueue  1060 =>  612
              AllocEnqueueAndPoll  1300 =>  725
            AllocEnqueueAndRemove  1350 =>  812
    AllocImplicitEnqueueAndRemove  7000 => 1758
                     Finalization 10000 => 6095

Bug: 24404957
Change-Id: I6c3e83f7c34f5c22b001e513b50c932a0d54fd43

benchmarks/src/benchmarks/ReferenceBenchmark.java[diff]
libart/src/main/java/java/lang/Daemons.java[diff]
ojluni/src/main/java/java/lang/ref/Reference.java[diff]
ojluni/src/main/java/java/lang/ref/ReferenceQueue.java[diff]

4 files changed

tree: 645768c92b2a643f36939338c91e38db4bf9dd8c