Edge trait

[wks]

Note that this commit does not add support for compressed OOPs.

See also: mmtk/mmtk-core#606

[wks]

At this stage, I simply set OpenJDK::VMEdge to Address so that it does not change the behaviour.

Microbenchmark shows the performance impact is negligible.

Benchmark: Create a 22-level binary search tree, run System.gc() 10 times for warmup, and run System.gc() 100 times to record the GC time. Repeat the experiment for 5 times, alternating between two builds (master and this PR) and two plans (SemiSpace and Immix). Collect all 500 GC times for each build+plan combination and see their distributions. The code is available at https://github.com/wks/tracetest

It is run on elk.moma. The command is: MMTK_THREADS=1 RUST_BACKTRACE=1 MMTK_PLAN=$plan /path/to/java -XX:+UseThirdPartyHeap -server -XX:-TieredCompilation -Xcomp -XX:MetaspaceSize=100M -Xm{s,x}500M TraceTest $depth $gcs $warmups.

plan	GC	mean:1	mean:2	+-	median:1	median:2	+-	sd:1	sd:2
SemiSpace	500	130.850331838ms	130.49742917ms	99.73030051735986%	130.25169ms	129.9837085ms	99.79425871556829%	1.9529565621904863ms	1.7784804514583266ms
Immix	500	183.50406285ms	185.00186711199999ms	100.81622403272037%	182.564638ms	183.971054ms	100.77036605522697%	4.910330121951998ms	4.992484211050324ms

Distribution:

Given that there is no performance degradation, I'll continue with other bindings. I'll also confirm the performance results by running DaCapo benchmarks.

[wks]

And the results on DaCapo Chopin. Run on elk.moma, 20 iterations.

Results for SemiSpace (on squirrel.moma)

Results for Immix (on squirrel.moma)

The mean STW times of most benchmarks are less than 1% different from master. Batik has huge error bars because GC only happened on average 3 times for SemiSpace and 1.6 times for Immix. H2 is a bit strange because in the 20 invocations, only 4 has results for SemiSpace and 2 has results for Immix, and I think the rest of the invocations failed silently for some reasons.

STW time distribution:

[qinsoon]

LGTM

[wenyuzhao]

LGTM.

Sorry I haven't fully addressed my compressed pointer performance issue, so I haven't try using this change to support compressed pointers. I'll probably do that later after this PR is merged.