RFR: 8310031: Parallel: Implement better work distribution for large object arrays in old gen

[Ralf Schmelter]

On Wed, 12 Jul 2023 08:05:59 GMT, Richard Reingruber <rrich at openjdk.org> wrote:

> This pr introduces parallel scanning of large object arrays in the old generation containing roots for young collections of Parallel GC. This allows for better distribution of the actual work (following the array references) as opposed to "stealing" from other task queues which can lead to inverse scaling demonstrated by small tests (attached to JDK-8310031) and also observed in gerrit production systems.
> 
> The algorithm to share scanning large arrays is supposed to be a straight
> forward extension of the scheme implemented in
> `PSCardTable::scavenge_contents_parallel`.
> 
> - A worker scans the part of a large array located in its stripe
> 
> - Except for the end of the large array reaching into a stripe which is scanned by the thread owning the previous stripe. This is just what the current implementation does: it skips objects crossing into the stripe.
> 
> - For this it is necessary that large arrays cover at least 3 stripes (see `PSCardTable::large_obj_arr_min_words`)
>   
> The implementation also makes use of the precise card marks for arrays. Only dirty regions are actually scanned.
> 
> #### Performance testing
> 
> ##### BigArrayInOldGenRR.java
> 
> [BigArrayInOldGenRR.java](https://bugs.openjdk.org/secure/attachment/104422/BigArrayInOldGenRR.java) is a micro benchmark that assigns new objects to a large array in a loop. Creating new array elements triggers young collections. In each collection the large array is scanned because of its references to the new elements in the young generation. The benchmark score is the geometric mean of the duration of the last 5 young collections (lower is better).
> 
> [BigArrayInOldGenRR.pdf](https://cr.openjdk.org/~rrich/webrevs/8310031/BigArrayInOldGenRR.pdf)([BigArrayInOldGenRR.ods](https://cr.openjdk.org/~rrich/webrevs/8310031/BigArrayInOldGenRR.ods)) presents the benchmark results with 1 to 64 gc threads.
> 
> Observations
> 
> * JDK22 scales inversely. Adding gc threads prolongues young collections. With 32 threads young collections take ~15x longer than single threaded.
> 
> * Fixed JDK22 scales well. Adding gc theads reduces the duration of young collections. With 32 threads young collections are 5x shorter than single threaded.
> 
> * With just 1 gc thread there is a regression. Young collections are 1.5x longer with the fix. I assume the reason is that the iteration over the array elements is interrupted at the end of a stripe which makes it less efficient. The prize for parallelization is paid without actually doing it. Also ParallelGC will use at lea...

src/hotspot/share/gc/parallel/psCardTable.cpp line 297:

> 295:       if (obj_sz >= large_obj_arr_min_words() && cast_to_oop(obj_addr)->is_objArray() &&
> 296:           obj_addr >= cur_stripe_addr) {
> 297:         // the last condition is not redundant as we can reach here if an obj starts at space_top

I think the comment here must be expanded. As far as I understand you, one can generally assume that the array found here must start in the stripe because either:
- the array actually starts in the stripe. Since it is larger than a stripe it must be the rightmost object.
- the array covers the whole stripe and doesn't start in the stripe. In this case we would never land here since the check !start_array->object_starts_in_range(cur_stripe_addr, cur_stripe_end_addr) in line 258 filters this out already (no object would start in the stripe)
- the array ends in the stripe, but not on the stripe end. In this case the rightmost object is not the array itself..

But when the object start array overlaps with the space_top, the start_array->object_starts_in_range(cur_stripe_addr, cur_stripe_end_addr) could return true, if there is an object at space_top. In this case the array would not start in the stripe and the check ensures that.

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

PR Review Comment: https://git.openjdk.org/jdk/pull/14846#discussion_r1275082104