RFR: 8254146: Avoid unnecessary volatile write on new AtomicBoolean(false)
Severin Gehwolf
sgehwolf at openjdk.java.net
Wed Oct 7 10:08:26 UTC 2020
On Mon, 5 Oct 2020 17:29:02 GMT, Christoph Dreis <github.com+6304496+dreis2211 at openjdk.org> wrote:
>> Hi,
>>
>> the following PR optimizes `new AtomicBoolean(boolean)` by avoiding the volatile write in case `false` is passed.
>> Essentially, it changes the ternary operator to a simple `if` without the `else` that would cause the volatile write.
>> The resulting bytecode seems to also benefit from the change:
>> Code:
>> 0: aload_0
>> 1: invokespecial #1 // Method java/lang/Object."<init>":()V
>> 4: aload_0
>> 5: iload_1
>> 6: ifeq 13
>> 9: iconst_1
>> 10: goto 14
>> 13: iconst_0
>> 14: putfield #7 // Field value:I
>> 17: return
>>
>> After:
>> Code:
>> 0: aload_0
>> 1: invokespecial #1 // Method java/lang/Object."<init>":()V
>> 4: iload_1
>> 5: ifeq 13
>> 8: aload_0
>> 9: iconst_1
>> 10: putfield #7 // Field value:I
>> 13: return
>>
>> A simple benchmark that returns `new AtomicBoolean(false)` shows the following results, that brings it on par to `new
>> AtomicBoolean()`: MyBenchmark.empty avgt 10 3,103 ± 0,246 ns/op
>> MyBenchmark.explicitNew avgt 10 2,966 ± 0,071 ns/op
>> MyBenchmark.explicitOld avgt 10 7,738 ± 0,321 ns/op
>>
>> In case you think this is worthwhile I'd be happy if this is sponsored.
>> Cheers,
>> Christoph
>
> I have contributed before the move to GitHub and signed the OCA there. Can anybody tell me what the process is for this
> case?
@dreis2211 Have you tried what the bot suggested?
Once you have signed the OCA, please let us know by writing /signed in a comment in this pull request.
-------------
PR: https://git.openjdk.java.net/jdk/pull/510
More information about the core-libs-dev
mailing list