Withdrawn: CODETOOLS-7902944: JMH: Avoid direct weak references in Blackhole.consume(Object)

[duke]

On Wed, 26 May 2021 10:54:28 GMT, Aleksey Shipilev <shade at openjdk.org> wrote:

> CODETOOLS-7901867 wrapped the stored object reference with a `WeakReference`. This unfortunately fails with inline type boxes that throw the exceptions like:
> 
> 
> java.lang.IllegalArgumentException: cannot reference a primitive type: org.openjdk.PrimitiveClasses$Value
> at java.base/java.lang.ref.Reference.<init>(Reference.java:495)
> at java.base/java.lang.ref.Reference.<init>(Reference.java:489)
> at java.base/java.lang.ref.WeakReference.<init>(WeakReference.java:62)
> at org.openjdk.jmh.infra.Blackhole.consumeFull(Blackhole.java:515)
> at org.openjdk.jmh.infra.Blackhole.consume(Blackhole.java:315)
> at org.openjdk.jmh_generated.PrimitiveClasses_test_jmhTest.test_avgt_jmhStub(PrimitiveClasses_test_jmhTest.java:190)
> at org.openjdk.jmh_generated.PrimitiveClasses_test_jmhTest.test_AverageTime(PrimitiveClasses_test_jmhTest.java:152)
> at java.base/jdk.internal.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
> at java.base/jdk.internal.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:78)
> at java.base/jdk.internal.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
> at java.base/java.lang.reflect.Method.invoke(Method.java:568)
> at org.openjdk.jmh.runner.BenchmarkHandler$BenchmarkTask.call(BenchmarkHandler.java:470)
> at org.openjdk.jmh.runner.BenchmarkHandler$BenchmarkTask.call(BenchmarkHandler.java:453)
> at java.base/java.util.concurrent.FutureTask.run(FutureTask.java:264)
> at java.base/java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:515)
> at java.base/java.util.concurrent.FutureTask.run(FutureTask.java:264)
> at java.base/java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1135)
> at java.base/java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:635)
> at java.base/java.lang.Thread.run(Thread.java:831)
> 
> 
> Wrapping in `WeakReference` is not very reliable, but it is an important defense-in-depth strategy: if it hoards a large object, then a GC in the middle of benchmark would drop it. To handle inline types, it might make some sense to wrap the instance with another holder.

This pull request has been closed without being integrated.

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

PR: https://git.openjdk.org/jmh/pull/42