RFR: 8315923: pretouch_memory by atomic-add-0 fragments huge pages unexpectedly [v21]

[Liming Liu]

On Thu, 11 Jan 2024 08:32:28 GMT, Thomas Stuefe <stuefe at openjdk.org> wrote:

>> Liming Liu has updated the pull request incrementally with one additional commit since the last revision:
>> 
>>   Use pthread instead
>
> test/hotspot/jtreg/runtime/os/TestTransparentHugePageUsage.java line 96:
> 
>> 94:           .map(e -> Long.valueOf(e.getKey().substring(e.getValue().start(1), e.getValue().end(1))));
>> 95:       if (!usage.isPresent()) throw new RuntimeException("The usage of THP was not found.");
>> 96:       if (usage.get() == 0) throw new RuntimeException("The usage of THP should not be zero.");
> 
> The effect we would see without your patch would be small pages that are then converted to huge pages by khugepaged in its own time, right? So, maybe test that AnonHugePages == Size for the heap VMA?

Unluckily, the size of AnonHugePages is still smaller than the heap size.

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

PR Review Comment: https://git.openjdk.org/jdk/pull/15781#discussion_r1449961373