jdk
jdk copied to clipboard
8322732: ForkJoinPool may underutilize cores in async mode
This set of changes address causes of poor utilization with small numbers of cores due to overly aggressive contention avoidance. A number of further adjustments were needed to still avoid most contention effects in deployments with large numbers of cores
Progress
- [ ] Change must be properly reviewed (1 review required, with at least 1 Reviewer)
- [x] Change must not contain extraneous whitespace
- [x] Commit message must refer to an issue
Issue
- JDK-8322732: ForkJoinPool may underutilize cores in async mode (Bug - P4)
Reviewing
Using git
Checkout this PR locally:
$ git fetch https://git.openjdk.org/jdk.git pull/19131/head:pull/19131
$ git checkout pull/19131
Update a local copy of the PR:
$ git checkout pull/19131
$ git pull https://git.openjdk.org/jdk.git pull/19131/head
Using Skara CLI tools
Checkout this PR locally:
$ git pr checkout 19131
View PR using the GUI difftool:
$ git pr show -t 19131
Using diff file
Download this PR as a diff file:
https://git.openjdk.org/jdk/pull/19131.diff
Webrev
:wave: Welcome back dl! A progress list of the required criteria for merging this PR into master
will be added to the body of your pull request. There are additional pull request commands available for use with this pull request.
❗ This change is not yet ready to be integrated. See the Progress checklist in the description for automated requirements.
@DougLea The following label will be automatically applied to this pull request:
-
core-libs
When this pull request is ready to be reviewed, an "RFR" email will be sent to the corresponding mailing list. If you would like to change these labels, use the /label pull request command.
Webrevs
- 04: Full - Incremental (72224779)
- 03: Full - Incremental (28341747)
- 02: Full - Incremental (f1fc4f3e)
- 01: Full - Incremental (8689ac12)
- 00: Full (7e64cdfc)
Maybe I don't quite understand, or I don't have proof, But wouldn't it be better if invokeAll in FIFO mode (asyncMode) traverses the Future list as a FIFO (currently it traverses in LIFO order)
@sunmisc invokeAll doesn't specify an exec or join order; forward exec snf inverse normally uses fewest resources.