RFR: 8254693: Add Panama feature to pass heap segments to native code [v2]

[Jorn Vernee]

On Wed, 18 Oct 2023 09:48:47 GMT, Jorn Vernee <jvernee at openjdk.org> wrote:

>> test/micro/org/openjdk/bench/java/lang/foreign/CriticalCalls.java line 101:
>> 
>>> 99:     public int callNotPinned() throws Throwable {
>>> 100:         try (Arena arena = Arena.ofConfined()) {
>>> 101:             MemorySegment nativeArr = arena.allocate(JAVA_INT, arr.length);
>> 
>> This allocation method will zero the segment. I think that's a bit unfair and will bias the results against non-pinned, right?
>
> Good point. I made this benchmark a while ago, before we had the more optimized allocate variants

This had a pretty big impact, actually. Especially on the larger sizes:


Benchmark                     (size)  Mode  Cnt       Score       Error  Units
CriticalCalls.callNotPinned      100  avgt   30      84.818 �     0.729  ns/op
CriticalCalls.callNotPinned    10000  avgt   30    2966.918 �    39.898  ns/op
CriticalCalls.callNotPinned  1000000  avgt   30  952864.052 � 34996.156  ns/op
CriticalCalls.callPinned         100  avgt   30      30.640 �     0.173  ns/op
CriticalCalls.callPinned       10000  avgt   30    2241.403 �    35.473  ns/op
CriticalCalls.callPinned     1000000  avgt   30  221152.247 �  1697.968  ns/op
CriticalCalls.callRecycled       100  avgt   30      40.205 �     0.458  ns/op
CriticalCalls.callRecycled     10000  avgt   30    2845.316 �    13.331  ns/op
CriticalCalls.callRecycled   1000000  avgt   30  287752.178 �  2322.382  ns/op

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

PR Review Comment: https://git.openjdk.org/jdk/pull/16201#discussion_r1363661902