On Fri, 10 Jan 2025 09:41:34 GMT, Alan Bateman <[email protected]> wrote:
>> Per Minborg has updated the pull request incrementally with two additional
>> commits since the last revision:
>>
>> - Clean up benchmark
>> - Fix allocation problem in benchmark
>
> src/java.base/share/classes/jdk/internal/foreign/CaptureStateUtil.java line
> 50:
>
>> 48: private static final long SIZE =
>> Linker.Option.captureStateLayout().byteSize();
>> 49:
>> 50: private static final TerminatingThreadLocal<MemorySegment> TL = new
>> TerminatingThreadLocal<>() {
>
> TerminatingThreadLocal is carrier-local so the usage here will require a
> detailed walk through to make sure that a virtual thread cannot be preempted
> when it has access to this memory segment. Identifying preemption is hard and
> will often during startup/first-use. So while tempting to use
> TerminatingThreadLocal here, I think we will have to do more work on it
> before it can more widely used in java.base.
Right. So, I have added this issue: https://bugs.openjdk.org/browse/JDK-8347537
-------------
PR Review Comment: https://git.openjdk.org/jdk/pull/22391#discussion_r1912954825
