8381550: Shenandoah: Fix confusing end of degenerated cycle log message

[pf0n]

A degenerated global cycle in Generational Shenandoah will show up as a Degenerated Young GC instead of a Degenerated Global GC in the gc logs. Changed the hardcoded message by using the generation type of the gc to distinguish between a young and global gc.

Before change:

[0.314s][info][gc,start    ] GC(1) Pause Degenerated GC (Global) (Update Refs)
[0.314s][info][gc,task     ] GC(1) Using 16 of 16 workers for stw degenerated gc
[0.316s][info][gc,free     ] Whole heap stats: Total free: 237M, Total used: 128K, Max free in a single region: 256K, Max humongous: 236M; Frag stats: External: 1%, Internal: 0%;  Mutator freeset stats: Partition count: 952, Reserved: 237M, Max free available in a single region: 256K; Collector freeset stats: Partition count: 52, Reserved: 13247K, Max free available in a single region: 256K; OldCollector freeset stats: Partition count: 2, Reserved: 512K, Max free available in a single region: 256K;
[0.316s][info][gc,plab     ] GC(1) Cycle complete, promotions reserved: 273196, promotions expended: 0, failed count: 0, failed bytes: 0
[0.316s][info][gc,ergo     ] GC(1) Good progress for free space: 237M, need 2621K
[0.316s][info][gc,ergo     ] GC(1) Good progress for used space: 768K, need 256K
[0.316s][info][gc          ] GC(1) Pause Degenerated GC (Global) (Update Refs) 5M->4M(256M) 1.976ms
[0.316s][info][gc,ergo     ] GC(1) At end of Degenerated Young GC: GCU: 0.4%, MU: 4.4% during period of 0.316s
[0.316s][info][gc,ergo     ] GC(1) At end of Degenerated Young GC: Young generation used: 4800K, used regions: 5120K, humongous waste: 255K, soft capacity: 256M, max capacity: 255M, available: 237M
[0.316s][info][gc,ergo     ] GC(1) At end of Degenerated Young GC: Old generation used: 0B, used regions: 0B, humongous waste: 0B, soft capacity: 256M, max capacity: 512K, available: 512K

After change:

[0.310s][info][gc,start    ] GC(1) Pause Degenerated GC (Global) (Update Refs)
[0.310s][info][gc,task     ] GC(1) Using 16 of 16 workers for stw degenerated gc
[0.312s][info][gc,free     ] Whole heap stats: Total free: 238M, Total used: 65488B, Max free in a single region: 256K, Max humongous: 237M; Frag stats: External: 1%, Internal: 0%;  Mutator freeset stats: Partition count: 954, Reserved: 238M, Max free available in a single region: 256K; Collector freeset stats: Partition count: 52, Reserved: 13312K, Max free available in a single region: 256K; OldCollector freeset stats: Partition count: 1, Reserved: 256K, Max free available in a single region: 256K;
[0.312s][info][gc,plab     ] GC(1) Cycle complete, promotions reserved: 195187, promotions expended: 0, failed count: 0, failed bytes: 0
[0.312s][info][gc,ergo     ] GC(1) Good progress for free space: 238M, need 2621K
[0.312s][info][gc,ergo     ] GC(1) Good progress for used space: 768K, need 256K
[0.312s][info][gc          ] GC(1) Pause Degenerated GC (Global) (Update Refs) 5M->4M(256M) 1.639ms
[0.312s][info][gc,ergo     ] GC(1) At end of Degenerated Global GC: GCU: 0.3%, MU: 4.4% during period of 0.312s
[0.312s][info][gc,ergo     ] GC(1) At end of Degenerated Global GC: Young generation used: 4415K, used regions: 4608K, humongous waste: 255K, soft capacity: 256M, max capacity: 255M, available: 238M
[0.312s][info][gc,ergo     ] GC(1) At end of Degenerated Global GC: Old generation used: 0B, used regions: 0B, humongous waste: 0B, soft capacity: 256M, max capacity: 256K, available: 256K

Tested with GHA, local tier 1 tests, and created a test that produces a degenerated global cycle (did not push this per @earthling-amzn).

---

• I confirm that I make this contribution in accordance with the OpenJDK Interim AI Policy.

---

Progress

• Change must be properly reviewed (1 review required, with at least 1 Reviewer)
• Change must not contain extraneous whitespace
• Commit message must refer to an issue

Issue

• JDK-8381550: Shenandoah: Fix confusing end of degenerated cycle log message (Bug - P4)

Reviewing

Using git

Checkout this PR locally:
$ git fetch https://git.openjdk.org/jdk.git pull/30684/head:pull/30684
$ git checkout pull/30684

Update a local copy of the PR:
$ git checkout pull/30684
$ git pull https://git.openjdk.org/jdk.git pull/30684/head

Using Skara CLI tools

Checkout this PR locally:
$ git pr checkout 30684

View PR using the GUI difftool:
$ git pr show -t 30684

Using diff file

Download this PR as a diff file:
https://git.openjdk.org/jdk/pull/30684.diff

Using Webrev

Link to Webrev Comment

[bridgekeeper]

👋 Welcome back pf0n! 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.

[openjdk]

❗ This change is not yet ready to be integrated.
See the Progress checklist in the description for automated requirements.

[openjdk]

@pf0n The following labels will be automatically applied to this pull request:

• hotspot-gc
• shenandoah

When this pull request is ready to be reviewed, an "RFR" email will be sent to the corresponding mailing lists. If you would like to change these labels, use the /label pull request command.

[mlbridge]

Webrevs

• 00: Full (be9e9911)

[ysramakrishna]

I think you may be able to refactor and common up some of the label generation (pun not intended) to reduce some of the repetition?

Looks good otherwise.

[earthling-amzn]

It would be nice if we didn't need to duplicate this switch logic to compute what is essentially the same value, but I can't see a way around it that wouldn't turn into a huge refactor.

[pf0n]

Yeah I initially didn't want the repetitive logic of the switch statements, but I noticed that the method record_degenerated calls ShenandoahMmuTracker::update_utilization which has the prefix At end of, while log_heap_status doesn't, so I can't pass in the same value for the record_degenerated method to avoid the repetition.

[earthling-amzn]

Can we also fix this from shenandoahGenerationalControlThread.cpp:

msg = "At end of Interrupted Concurrent GLOBAL GC";

It shouldn't be yelling about GLOBAL GCs.