8341097: GHA: Demote Mac x86 jobs to build only

[shipilev]

See the discussion in the bug. I think we can stop testing Mac x86 in GHA, leaving only the build jobs.

Additional testing:

• GHA passes

---

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-8341097: GHA: Demote Mac x86 jobs to build only (Enhancement - P4)

Reviewers

• Kim Barrett (@kimbarrett - Reviewer)
• Magnus Ihse Bursie (@magicus - Reviewer) Review applies to e8a1ab21
• Phil Race (@prrace - Reviewer)

Reviewing

Using git

Checkout this PR locally:
$ git fetch https://git.openjdk.org/jdk.git pull/21257/head:pull/21257
$ git checkout pull/21257

Update a local copy of the PR:
$ git checkout pull/21257
$ git pull https://git.openjdk.org/jdk.git pull/21257/head

Using Skara CLI tools

Checkout this PR locally:
$ git pr checkout 21257

View PR using the GUI difftool:
$ git pr show -t 21257

Using diff file

Download this PR as a diff file:
https://git.openjdk.org/jdk/pull/21257.diff

Using Webrev

Link to Webrev Comment

[bridgekeeper]

👋 Welcome back shade! 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]

@shipilev This change now passes all automated pre-integration checks.

ℹ️ This project also has non-automated pre-integration requirements. Please see the file CONTRIBUTING.md for details.

After integration, the commit message for the final commit will be:

8341097: GHA: Demote Mac x86 jobs to build only

Reviewed-by: kbarrett, prr, ihse

You can use pull request commands such as /summary, /contributor and /issue to adjust it as needed.

At the time when this comment was updated there had been 52 new commits pushed to the master branch:

• 0a5ef82: 8347162: Update problemlist CR for vmTestbase/nsk/jdi/VMOutOfMemoryException
• ea49537: 8310340: assert(_thread->is_interp_only_mode() || stub_caller) failed: expected a stub-caller
• 2e00816: 8346671: java/nio/file/Files/probeContentType/Basic.java fails on Windows 2025
• 88fa3b2: 8346998: Test nsk/jvmti/ResourceExhausted/resexhausted003 fails with java.lang.OutOfMemoryError when CDS is off
• 6ee2bd2: 8347147: [REDO] AccessFlags can be u2 in metadata
• a641932: 8346310: Duplicate !HAS_PENDING_EXCEPTION check in DynamicArchive::dump_at_exit
• 49ee4df: 8166983: Remove old/legacy unused tzdata files
• 3fe0818: 8346099: JFR: Query for 'jfr view' can't handle wildcard with multiple event types
• 55bcf4c: 8346047: JFR: Incorrect percentile value in 'jfr view'
• ae3fc46: 8345580: Remove const from Node::_idx which is modified
• ... and 42 more: https://git.openjdk.org/jdk/compare/f1d85ab3e61f923b4e120cf30e16109e04505b53...master

As there are no conflicts, your changes will automatically be rebased on top of these commits when integrating. If you prefer to avoid this automatic rebasing, please check the documentation for the /integrate command for further details.

➡️ To integrate this PR with the above commit message to the master branch, type /integrate in a new comment.

[openjdk]

@shipilev The following label will be automatically applied to this pull request:

• build

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.

[mlbridge]

Webrevs

• 03: Full (ee1ecf58)
• 02: Full (e652716a)
• 01: Full (cb106c45)
• 00: Full (e8a1ab21)

[kimbarrett]

This change seems premature. Various companies (Oracle, SAP, others?) are still
supporting macos-x64 on mainline. Removing this seems to be just asking for more
frequent backouts due to x64-specific changes that aren't compatible with clang.

[shipilev]

This change seems premature. Various companies (Oracle, SAP, others?) are still supporting macos-x64 on mainline.

Maybe premature, maybe not. We always run against the runner availability in GHA. AFAIU, MacOS 13 will EOL 3 years after release, so around Oct 2025. I don't mind waiting until next year, but want to point out that the clock is ticking on GHA support for Mac x86, and runner availability will likely get gradually worse. Given that GHA runs for every PR, I prefer to fix GHA relabililty/turnaround issues before they become pain points.

Removing this seems to be just asking for more frequent backouts due to x64-specific changes that aren't compatible with clang.

I would assume that clang incompatibilities would be caught as build error/warnings during the build. This PR suggests to only disable the test jobs. Build jobs are still on, and at least some parts of runtime are verified when building the full JDK, e.g. when building jmods/jimages.

[magicus]

I think this sounds reasonable. Just as we keep adding to GHA testing, we also need to find any opportunity we can to keep it slim.

If macos-x64 had been added at this point, I don't think anyone would have suggested that we added running tests for it, so then we should not keep them just out of old habit.

[bridgekeeper]

@shipilev This pull request has been inactive for more than 4 weeks and will be automatically closed if another 4 weeks passes without any activity. To avoid this, simply add a new comment to the pull request. Feel free to ask for assistance if you need help with progressing this pull request towards integration!

[shipilev]

Not now, bot. I'll probably delay this until the JDK 25 fork.

[shipilev]

As far as I can tell -- and this is very much like reading the tea leaves:

• macos-11 runners started brownouts and got completely removed in June 2024
• macos-12 runners started brownouts in Nov 2024, and would be removed in Dec 2024

Pessimistically assuming the same 5-month interval, we may expect the brownouts and removals for macos-13 runners somewhere in Apr 2025. If this happens, it would start affecting July 2025 CPU releases. But it should still be fine for JDK 24 GA.

I say we integrate this PR in early "JDK 25" mainline, and let update releases pick it up, as the situation develops.

[shipilev]

Rebased, retested, JDK 25 is now open.

@kimbarrett -- checking now that 2+ months have passed, and seeing @magicus's opinion, are you still thinking this is premature?

[shipilev]

Not anecdotally anymore: GHA now publishes richer Actions stats, should be at "Actions" in your personal fork. Here is mine, sorted by queue wait time last month. macos-x86 are the largest waiters, which IMO underlines the reality that macos-x86 workers are not in abundance. macos-aarch64 are also not great, but at least they match the current hardware that we want to test for.

[shipilev]

Friendly ping :)

[prrace]

I don't have any objection to this, but
(1) I'm surprised Microsoft (github) doesn't have x64 systems that can run macOS14/15
(2) I expect we'll actually support macOS x64 for as long as Apple do .. and since in 2023 you could still get x64 macs I expect it'll be a number of years before support is dropped.
But since you can no longer get them, if MS missed the last order date, they will soon be struggling to have enough to meet demand, which is why the demotion makes sense.

[kimbarrett]

Looks good.

[shipilev]

Thanks! Here goes.

/integrate

[openjdk]

Going to push as commit d66737e.
Since your change was applied there have been 62 commits pushed to the master branch:

• dff5719: 8347126: gc/stress/TestStressG1Uncommit.java gets OOM-killed
• 82e16ba: 8347268: [ubsan] logOutput.cpp:357:21: runtime error: applying non-zero offset 1 to null pointer
• 33f9be8: 8347083: Incomplete logging in nsk/jvmti/ResourceExhausted/resexhausted00* tests
• 8eddf67: 8346887: DrawFocusRect() may cause an assertion failure
• a46ae70: 8339728: [Accessibility,Windows,JAWS] Bug in the getKeyChar method of the AccessBridge class
• 97dd06c: 8347299: Add annotations to test cases in LicenseTest
• 0a35ebf: 8347297: Skip the RuntimeImageSymbolicLinksTest test on Windows when it is executed outside of the jtreg
• b3e8736: 8347296: WinInstallerUiTest fails in local test runs if the path to test work directory is longer that regular
• 1ade96b: 8347042: Remove an extra parenthesis in macro definition in jfrTraceIdMacros.hpp
• db7fa6a: 8347289: HKDF delayed provider selection failed with non-extractable PRK
• ... and 52 more: https://git.openjdk.org/jdk/compare/f1d85ab3e61f923b4e120cf30e16109e04505b53...master

Your commit was automatically rebased without conflicts.

[openjdk]

@shipilev Pushed as commit d66737e.

💡 You may see a message that your pull request was closed with unmerged commits. This can be safely ignored.