8303215: Make thread stacks not use huge pages #3106

tkiriyama · 2025-10-21T11:11:57Z

Hi All,

I would like to backport the fix for bug JDK-8303215 from JDK 17 to JDK 11. This bug addresses an issue with Transparent Huge Pages (THP) and thread stack allocation.
Backport is unclean,

1. Backport:
The fix in os_linux.cpp cannot be cleanly backported.
Since jdk11 does not include JDK-8244752 and JDK-8256155, it differs from the fix in jdk17. It uses setup_large_page_size() instead of scan_default_large_page_size().

2. Bug Reproduction:
The bug was successfully reproduced in Linux x86_64 with THP enabled (/sys/kernel/mm/transparent_hugepage/enabled = always).
Reproduction was confirmed using the following dummy thread program:

public class ThreadStackHugePageTest {
        public static void main(String[] args) {
                final int NUM_THREADS = 1000;
                final long SLEEP_SECONDS = 3600;
                final long SLEEP_MILISECONDS = SLEEP_SECONDS * 1000;

                List<Thread> threads = new ArrayList<>();

                for (int i = 0; i < NUM_THREADS; i++) {
                        Runnable task = () -> {
                                try {
                                        Thread.sleep(SLEEP_MILISECONDS);
                                } catch (InterruptedException e) {
                                        Thread.currentThread().interrupt();
                                }
                        };

                        Thread thread = new Thread(task);
                        threads.add(thread);

                        thread.start();
                }

                System.out.println("all threads are started");

                for (Thread thread : threads) {
                        try {
                                thread.join();
                        } catch (InterruptedException e) {
                                Thread.currentThread().interrupt();
                        }
                }
        }
}

The fix significantly reduces the number of anonymous regions with fully occupied RSS.

3. Regression Testing:
I ran the test/hotspot tests on Linux x64 as a regression test. No failures were found.

Thank you.

Progress

Change must not contain extraneous whitespace
Commit message must refer to an issue
Change must be properly reviewed (2 reviews required, with at least 2 Reviewers)
JDK-8303215 needs maintainer approval

Issue

JDK-8303215: Make thread stacks not use huge pages (Enhancement - P3 - Approved)

Reviewers

Paul Hohensee (@phohensee - Reviewer)
Thomas Stuefe (@tstuefe - Reviewer)

Reviewing

Using git

Checkout this PR locally:
$ git fetch https://git.openjdk.org/jdk11u-dev.git pull/3106/head:pull/3106
$ git checkout pull/3106

Update a local copy of the PR:
$ git checkout pull/3106
$ git pull https://git.openjdk.org/jdk11u-dev.git pull/3106/head

Using Skara CLI tools

Checkout this PR locally:
$ git pr checkout 3106

View PR using the GUI difftool:
$ git pr show -t 3106

Using diff file

Download this PR as a diff file:
https://git.openjdk.org/jdk11u-dev/pull/3106.diff

Using Webrev

Link to Webrev Comment

bridgekeeper · 2025-10-21T11:12:39Z

👋 Welcome back tkiriyama! 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 · 2025-10-21T11:14:02Z

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

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

8303215: Make thread stacks not use huge pages

Reviewed-by: phh, stuefe

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 42 new commits pushed to the master branch:

1d8535d: 8305186: Reference.waitForReferenceProcessing should be more accessible to tests
8fe22bd: Merge
be7966c: 8372534: Update Libpng to 1.6.51
... and 39 more: https://git.openjdk.org/jdk11u-dev/compare/b638bb35cc124122e84ba0583c1218a08c60bf93...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.

As you do not have Committer status in this project an existing Committer must agree to sponsor your change. Possible candidates are the reviewers of this PR (@phohensee, @tstuefe) but any other Committer may sponsor as well.

➡️ To flag this PR as ready for integration with the above commit message, type /integrate in a new comment. (Afterwards, your sponsor types /sponsor in a new comment to perform the integration).

mlbridge · 2025-10-21T11:17:38Z

Webrevs

00: Full (54a6c045)

phohensee · 2025-10-21T12:03:56Z

Hi, @tkiriyama. As the bot says, you need to change the PR title to "Backport 59d9d9fcb93c26dd8931d70934b889245b050acc" (I looked it up).

There are two follow-up issues, and they in turn also have follow-up issues. Please check them to see if they need to be included.

openjdk · 2025-10-22T09:35:52Z

This backport pull request has now been updated with issue from the original commit.

tkiriyama · 2025-10-22T09:42:37Z

Hi, @phohensee ,
Sorry, I fixed the title as you suggested.
Are you referring to JDK-8312182 and JDK-8310687?
They have not been verified yet, but they might also be needed for jdk11. If so, I'll create new pull requests.

phohensee · 2025-10-23T07:05:56Z

Yes, those two. But they in turn have related issues.

tkiriyama · 2025-10-23T12:22:32Z

I see, but these verifications will take time.
Can we integrate this fix first? No regression was found.

phohensee

Ok.

openjdk · 2025-10-27T20:51:42Z

⚠️ @tkiriyama This change is now ready for you to apply for maintainer approval. This can be done directly in each associated issue or by using the /approval command.

tkiriyama · 2025-10-28T12:05:53Z

/approval request I would like to backport JDK-8303215 fix done in jdk17 to jdk11. Even though it is not a clean backport, it has been verified that there is no risk by testing test/hotspot.

openjdk · 2025-10-28T12:07:26Z

@tkiriyama
8303215: The approval request has been created successfully.

bridgekeeper · 2025-11-25T16:06:11Z

@tkiriyama 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 issue a /touch or /keepalive command to the pull request. Feel free to ask for assistance if you need help with progressing this pull request towards integration!

tkiriyama · 2025-12-10T05:24:00Z

/touch

openjdk · 2025-12-10T05:25:51Z

@tkiriyama The pull request is being re-evaluated and the inactivity timeout has been reset.

tkiriyama · 2025-12-10T05:31:24Z

Hi, @theRealAph , @jerboaa , @gnu-andrew ,
Would you please approve my Fix Request for the 8303215 backport as a maintainer?

jerboaa · 2025-12-10T09:26:40Z

/reviewers 2 Reviewer

jerboaa · 2025-12-10T09:28:14Z

@tstuefe Could you please help getting a second review on this one? This seems a moderate risk patch for 11u and you've done some work in that area.

openjdk · 2025-12-10T09:28:57Z

@jerboaa
The total number of required reviews for this PR (including the jcheck configuration and the last /reviewers command) is now set to 2 (with at least 2 Reviewers).

tstuefe

This is ok

jerboaa · 2025-12-12T13:22:50Z

@tkiriyama Please enable testing on your jdk11u-dev fork and trigger a test run. I don't see any GHA runs. Thanks! Once that's done feel free to re-apply for approval.

tkiriyama · 2025-12-15T07:41:12Z

Hi, @jerboaa,
Thank you for your review and pointing that out. I ran the GHA.
Some errors occurred on the Windows platform, but I believe this fix does not affect execution on Windows and can be ignored.
Would you please approve this PR?

jerboaa · 2025-12-15T09:44:38Z

Hi, @jerboaa, Thank you for your review and pointing that out. I ran the GHA. Some errors occurred on the Windows platform, but I believe this fix does not affect execution on Windows and can be ignored.

Agreed. GHA failures on Windows are infra related:

Downloading MSYS2...
  Unexpected HTTP response: 504
  Waiting 10 seconds before trying again
  Unexpected HTTP response: 504
  Waiting 16 seconds before trying again
  Error: Unexpected HTTP response: 504

Would you please approve this PR?

@tkiriyama Please issue the /approval request command again first.

tkiriyama · 2025-12-16T05:47:01Z

/approval request I would like to backport JDK-8303215 fix done in jdk17 to jdk11. Even though it is not a clean backport, it has been verified that there is no risk by testing test/hotspot.

openjdk · 2025-12-16T05:48:36Z

@tkiriyama
8303215: The approval request was already up to date.

tkiriyama · 2025-12-16T05:52:28Z

@jerboaa
Understood. Please review the newly generated request.

tkiriyama · 2026-01-09T10:58:01Z

Hi, @jerboaa,
Would you please approve my Fix Request ?

jerboaa · 2026-01-12T10:31:05Z

Hi, @jerboaa, Would you please approve my Fix Request ?

It's approved.

tkiriyama · 2026-01-14T09:22:30Z

/integrate

openjdk · 2026-01-14T09:25:06Z

@tkiriyama
Your change (at version 54a6c04) is now ready to be sponsored by a Committer.

jerboaa · 2026-01-14T09:26:52Z

/sponsor

openjdk · 2026-01-14T09:27:15Z

Going to push as commit ff759fc.
Since your change was applied there have been 42 commits pushed to the master branch:

1d8535d: 8305186: Reference.waitForReferenceProcessing should be more accessible to tests
8fe22bd: Merge
be7966c: 8372534: Update Libpng to 1.6.51
... and 39 more: https://git.openjdk.org/jdk11u-dev/compare/b638bb35cc124122e84ba0583c1218a08c60bf93...master

Your commit was automatically rebased without conflicts.

openjdk · 2026-01-14T09:27:28Z

@jerboaa @tkiriyama Pushed as commit ff759fc.

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

8303215: Make thread stacks not use huge pages

54a6c04

openjdk bot added the rfr Pull request is ready for review label Oct 21, 2025

tkiriyama mentioned this pull request Oct 21, 2025

8303215: Make thread stacks not use huge pages #2208

Closed

4 tasks

tkiriyama changed the title ~~8303215: Make thread stacks not use huge pages~~ Backport 59d9d9fcb93c26dd8931d70934b889245b050acc Oct 22, 2025

openjdk bot changed the title ~~Backport 59d9d9fcb93c26dd8931d70934b889245b050acc~~ 8303215: Make thread stacks not use huge pages Oct 22, 2025

openjdk bot added the backport Port of a pull request already in a different code base label Oct 22, 2025

phohensee approved these changes Oct 27, 2025

View reviewed changes

openjdk bot added the approval Requires approval; will be removed when approval is received label Oct 28, 2025

openjdk bot removed the approval Requires approval; will be removed when approval is received label Dec 12, 2025

tstuefe approved these changes Dec 12, 2025

View reviewed changes

openjdk bot added the approval Requires approval; will be removed when approval is received label Dec 16, 2025

openjdk bot added ready Pull request is ready to be integrated and removed approval Requires approval; will be removed when approval is received labels Jan 9, 2026

openjdk bot added the sponsor Pull request is ready to be sponsored label Jan 14, 2026

openjdk bot added the integrated Pull request has been integrated label Jan 14, 2026

openjdk bot closed this Jan 14, 2026

openjdk bot removed ready Pull request is ready to be integrated rfr Pull request is ready for review sponsor Pull request is ready to be sponsored labels Jan 14, 2026

8303215: Make thread stacks not use huge pages #3106

8303215: Make thread stacks not use huge pages #3106

Uh oh!

Conversation

tkiriyama commented Oct 21, 2025 • edited by openjdk bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Progress

Issue

Reviewers

Reviewing

Uh oh!

bridgekeeper bot commented Oct 21, 2025

Uh oh!

openjdk bot commented Oct 21, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

mlbridge bot commented Oct 21, 2025

Webrevs

Uh oh!

phohensee commented Oct 21, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

openjdk bot commented Oct 22, 2025

Uh oh!

tkiriyama commented Oct 22, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

phohensee commented Oct 23, 2025

Uh oh!

tkiriyama commented Oct 23, 2025

Uh oh!

phohensee left a comment

Choose a reason for hiding this comment

Uh oh!

openjdk bot commented Oct 27, 2025

Uh oh!

tkiriyama commented Oct 28, 2025

Uh oh!

openjdk bot commented Oct 28, 2025

Uh oh!

bridgekeeper bot commented Nov 25, 2025

Uh oh!

tkiriyama commented Dec 10, 2025

Uh oh!

openjdk bot commented Dec 10, 2025

Uh oh!

tkiriyama commented Dec 10, 2025

Uh oh!

jerboaa commented Dec 10, 2025

Uh oh!

jerboaa commented Dec 10, 2025

Uh oh!

openjdk bot commented Dec 10, 2025

Uh oh!

tstuefe left a comment

Choose a reason for hiding this comment

Uh oh!

jerboaa commented Dec 12, 2025

Uh oh!

tkiriyama commented Dec 15, 2025

Uh oh!

jerboaa commented Dec 15, 2025

Uh oh!

tkiriyama commented Dec 16, 2025

Uh oh!

openjdk bot commented Dec 16, 2025

Uh oh!

tkiriyama commented Dec 16, 2025

Uh oh!

tkiriyama commented Jan 9, 2026

Uh oh!

jerboaa commented Jan 12, 2026

Uh oh!

tkiriyama commented Jan 14, 2026

Uh oh!

openjdk bot commented Jan 14, 2026

Uh oh!

jerboaa commented Jan 14, 2026

Uh oh!

openjdk bot commented Jan 14, 2026

tkiriyama commented Oct 21, 2025 •

edited by openjdk bot

Loading

openjdk bot commented Oct 21, 2025 •

edited

Loading

phohensee commented Oct 21, 2025 •

edited

Loading

tkiriyama commented Oct 22, 2025 •

edited

Loading