[jdk21u-dev] RFR: 8350285: Shenandoah: Regression caused by ShenandoahLock under extreme contention

Daniel Huang duke at openjdk.org
Mon Jun 30 22:43:56 UTC 2025


Backport for ShenandoahLock performance regression issue. The fix involves sleeping for a very short duration every 3 yields, with the number of yields picked through manual testing.

Clean backport, ran GHA sanity checks and locally tested `tier1`, `tier2`, and `hotspot_gc_shenandoah`. `test/jdk/java/nio/channels/FileChannel/directio/DirectIOTest.java` sometimes fails locally, but it also sometimes failed before the backport.
`test/jdk/java/nio/channels/DatagramChannel/SendReceiveMaxSize.java` fails locally, but it also fails locally before the backport.

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

Commit messages:
 - Backport bd8ad309b59bceb3073a8d6411cca74e73508885

Changes: https://git.openjdk.org/jdk21u-dev/pull/1933/files
  Webrev: https://webrevs.openjdk.org/?repo=jdk21u-dev&pr=1933&range=00
  Issue: https://bugs.openjdk.org/browse/JDK-8350285
  Stats: 18 lines in 2 files changed: 15 ins; 0 del; 3 mod
  Patch: https://git.openjdk.org/jdk21u-dev/pull/1933.diff
  Fetch: git fetch https://git.openjdk.org/jdk21u-dev.git pull/1933/head:pull/1933

PR: https://git.openjdk.org/jdk21u-dev/pull/1933


More information about the jdk-updates-dev mailing list