RFR: 8295476: Split G1 cost per byte predictor on gc phase
Thomas Schatzl
tschatzl at openjdk.org
Tue Oct 25 20:07:23 UTC 2022
Hi,
please review this change to the cost-per-byte predictor to be split on gc phase (young-only/mixed) instead of during marking/outside of marking.
The reason is that the distinction between young-only/mixed of these values is higher than marking/outside of marking, although the latter degrades somewhat to the former in case of continuous concurrent marking where both are almost equivalent anyway. I believe that has been the original reason to do the marking/not-marking distinction.
I also removed the implied 10% extra cost for copies during marking (i.e. during the young-only phase). I could not find any particular reason why copying would be more expensive then with the current code base.
The distinction in general makes sense as the data copied during young-only and mixed is different (only short-lived data vs. short+long-lived data).
Testing: gha, examining predictions with some benchmarks
Thanks,
Thomas
-------------
Commit messages:
- rename sequence
- initial version
Changes: https://git.openjdk.org/jdk/pull/10862/files
Webrev: https://webrevs.openjdk.org/?repo=jdk&pr=10862&range=00
Issue: https://bugs.openjdk.org/browse/JDK-8295476
Stats: 37 lines in 3 files changed: 4 ins; 24 del; 9 mod
Patch: https://git.openjdk.org/jdk/pull/10862.diff
Fetch: git fetch https://git.openjdk.org/jdk pull/10862/head:pull/10862
PR: https://git.openjdk.org/jdk/pull/10862
More information about the hotspot-gc-dev
mailing list