Kunal Sareen

> These are some results from booting OpenJDK. I was using ` /usr/bin/time --verbose build/jdk-mmtk-dense-chunk-map/jdk/bin/java -XX:+UseThirdPartyHeap` for each build, and ran it once. I also tried with [what Kunal proposed](https://mmtk.zulipchat.com/#narrow/stream/262674-mmtk-openjdk/topic/Slow.20startup.3F/near/269544013)...

A totally empty class for me using the PR has the same performance? ```console Command being timed: "./build/linux-x86_64-normal-server-release/images/jdk/bin/java Empty.java" User time (seconds): 0.65 System time (seconds): 0.03 Maximum resident set...

Even attach time of both threads is the same in my build. Just checking -- your builds were built using `make images` or `make product-bundles`, right?

> I was using exploded image. The attach time of the first thread is always 0.001s, but the attach time for the second thread depends on the MMTk initialization speed....

Closed as above

Closed with https://github.com/mmtk/mmtk-core/pull/897

See discussions here: https://mmtk.zulipchat.com/#narrow/stream/262673-mmtk-core/topic/Generational.20GC.20performance Results: Ryzen 9 5950X http://squirrel.anu.edu.au/plotty/kunals/bench-test/p/S2cc6q Intel i9 9900K http://squirrel.anu.edu.au/plotty/kunals/bench-test/p/4SQyvC What we need to do @qinsoon is to increase the min nursery size. That will fix some...

Steve also mentioned that we should change all references to "Appel-style nursery" to "variable-size nursery" in comments (as "variable-style nursery" is more descriptive) which I will do at some point.

Closed with https://github.com/mmtk/mmtk-openjdk/pull/216

Very bit rotted and not useful PR anymore.