RFR: 8256265: G1: Improve parallelism in regions that failed evacuation [v19]

[Albert Mingkun Yang]

On Mon, 11 Apr 2022 13:39:26 GMT, Hamlin Li <mli at openjdk.org> wrote:

>> Currently G1 assigns a thread per failed evacuated region. This can in effect serialize the whole process as often (particularly with region pinning) there is only one region to fix up.
>> 
>> This patch tries to improve parallelism when walking over the regions in chunks
>> 
>> Latest implementation scans regions in chunks to bring parallelism, it's based on JDK-8278917 which changes to uses prev bitmap to mark evacuation failure objs.
>> 
>> Here's the summary of performance data based on latest implementation, basically, it brings better and stable performance than baseline at "Post Evacuate Cleanup 1/remove self forwardee" phase. (Although some regression is spotted when calculate the results in geomean, becuase one pause time from baseline is far too small than others.)
>> 
>> The performance benefit trend is:
>>  - pause time (Post Evacuate Cleanup 1) is decreased from 76.79% to 2.28% for average time, from 71.61% to 3.04% for geomean, when G1EvacuationFailureALotCSetPercent is changed from 2 to 90 (-XX:ParallelGCThreads=8)
>>  - pause time (Post Evacuate Cleanup 1) is decreased from 63.84% to 15.16% for average time, from 55.41% to 12.45% for geomean, when G1EvacuationFailureALotCSetPercent is changed from 2 to 90 (-XX:ParallelGCThreads=<default=123>)
>> ( Other common Evacuation Failure configurations are:
>> -XX:+G1EvacuationFailureALot -XX:G1EvacuationFailureALotInterval=0 -XX:G1EvacuationFailureALotCount=0 )
>> 
>> For more detailed performance data, please check the related bug.
>
> Hamlin Li has updated the pull request incrementally with one additional commit since the last revision:
> 
>   Albert review

As a followup to the explicit-loop topic raised before, here's a patch exploring that alternative.

Last commit of https://github.com/openjdk/jdk/compare/master...albertnetymk:explicit-loop?expand=1

It contains mostly two changes, explicit loop and chunking logic encapsulation.

1. I think the workflow is easier to follow in the explicit-loop approach, compared with iterator + closure.

2. The chunking logic is not inherent to (evac-failure) regions; instead of it's related to processing evac-failure regions. This way the evac-failure regions class is just a collection of regions, nothing more.

It can probably be further polished, but hopefully it illustrates the gist for now. What do you think?

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

PR: https://git.openjdk.java.net/jdk/pull/7047