RFR: 8325553: Parallel: Use per-marker cache for marking stats during Full GC [v4]

Albert Mingkun Yang ayang at openjdk.org
Fri Feb 23 15:43:17 UTC 2024


> Add cache to avoid too much contentious writes to global marking stats.
> 
> Test: tier1-6; using the attached bm, ~30% reduction in pause-time is observed. Only marginal improvement (probably just noise) is observed in BigRamTester.

Albert Mingkun Yang has updated the pull request with a new target base due to a merge or a rebase. The incremental webrev excludes the unrelated changes brought in by the merge/rebase. The pull request contains seven additional commits since the last revision:

 - serial-flush
 - Merge branch 'master' into pgc-full-gc-cache
 - review
 - Merge branch 'master' into pgc-full-gc-cache
 - review
 - Merge branch 'master' into pgc-full-gc-cache
 - pgc-full-gc-cache

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

Changes:
  - all: https://git.openjdk.org/jdk/pull/17788/files
  - new: https://git.openjdk.org/jdk/pull/17788/files/b99ef273..6b06cadd

Webrevs:
 - full: https://webrevs.openjdk.org/?repo=jdk&pr=17788&range=03
 - incr: https://webrevs.openjdk.org/?repo=jdk&pr=17788&range=02-03

  Stats: 13001 lines in 379 files changed: 7497 ins; 2904 del; 2600 mod
  Patch: https://git.openjdk.org/jdk/pull/17788.diff
  Fetch: git fetch https://git.openjdk.org/jdk.git pull/17788/head:pull/17788

PR: https://git.openjdk.org/jdk/pull/17788


More information about the hotspot-gc-dev mailing list