8366658: Add missing locks when accessing the VirtualMemoryTracker instance in tests and MemMapPrinter #27038

jdksjolen · 2025-09-02T07:07:51Z

Hi,

This PR adds the missing locks for the tests and MemMapPrinter, also fixes a small bug in walk_virtual_memory. See inline review comments.

Passes tier1.

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-8366658: Add missing locks when accessing the VirtualMemoryTracker instance in tests and MemMapPrinter (Bug - P4)

Reviewers

Paul Hübner (@Arraying - Author) Review applies to 20d2f5de
Afshin Zafari (@afshin-zafari - Reviewer)

Reviewing

Using git

Checkout this PR locally:
$ git fetch https://git.openjdk.org/jdk.git pull/27038/head:pull/27038
$ git checkout pull/27038

Update a local copy of the PR:
$ git checkout pull/27038
$ git pull https://git.openjdk.org/jdk.git pull/27038/head

Using Skara CLI tools

Checkout this PR locally:
$ git pr checkout 27038

View PR using the GUI difftool:
$ git pr show -t 27038

Using diff file

Download this PR as a diff file:
https://git.openjdk.org/jdk/pull/27038.diff

Using Webrev

Link to Webrev Comment

bridgekeeper · 2025-09-02T07:08:26Z

👋 Welcome back jsjolen! 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 · 2025-09-02T07:09:14Z

@jdksjolen 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:

8366658: Add missing locks when accessing the VirtualMemoryTracker instance in tests and MemMapPrinter

Reviewed-by: azafari, phubner

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 242 new commits pushed to the master branch:

02d7281: 8367483: C2 crash in PhaseValues::type: assert(t != nullptr) failed: must set before get - missing notification for CastX2P(SubL(x, y))
15dd96f: 8144124: [macosx] The tabs can't be aligned when we pressing the key of 'R','B','L','C' or 'T'.
d13769d: 8367501: RISC-V: build broken after JDK-8365926
... and 239 more: https://git.openjdk.org/jdk/compare/124575b4c2b52328a8efddb40e67057a53b44a04...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 · 2025-09-02T07:10:12Z

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

hotspot-runtime

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.

jdksjolen · 2025-09-04T07:40:54Z

src/hotspot/share/nmt/virtualMemoryTracker.hpp

 class VirtualMemoryWalker : public StackObj {
- public:
-   virtual bool do_allocation_site(const ReservedMemoryRegion* rgn) { return false; }
+public:
+  virtual bool do_allocation_site(const ReservedMemoryRegion* rgn) { return false; }
 };


Indentation fix

jdksjolen · 2025-09-04T07:41:23Z

test/hotspot/gtest/runtime/test_virtualMemoryTracker.cpp

  LOG("In reserved region " PTR_FORMAT ", size %X:", p2i(rmr.base()), rmr.size());
-  VirtualMemoryTracker::Instance::tree()->visit_committed_regions(rmr, [&](CommittedMemoryRegion& region) {
+  vmt.tree()->visit_committed_regions(rmr, [&](CommittedMemoryRegion& region) {
    LOG("   committed region: " PTR_FORMAT ", size %X", p2i(region.base()), region.size());


Here the VMT was passed along but never actually used for printing

jdksjolen · 2025-09-04T07:42:31Z

src/hotspot/share/nmt/virtualMemoryTracker.cpp

 bool VirtualMemoryTracker::walk_virtual_memory(VirtualMemoryWalker* walker) {
-  MemTracker::NmtVirtualMemoryLocker nvml;
+  bool ret = true;
  tree()->visit_reserved_regions([&](ReservedMemoryRegion& rgn) {
    if (!walker->do_allocation_site(&rgn)) {
+      ret = false;
      return false;
    }
    return true;
  });
-  return true;
+  return ret;
 }


A VirtualMemoryWalker returned true or false depending on whether or not its operation succeeded, and we are meant to pass that along as the return value of walk_virtual_memory. That was missing, now fixed.

Nice catch!

jdksjolen · 2025-09-04T07:42:54Z

src/hotspot/share/nmt/memTracker.hpp

+  static inline bool walk_virtual_memory(VirtualMemoryWalker* walker) {
+    assert_post_init();
+    if (!enabled()) return false;
+    MemTracker::NmtVirtualMemoryLocker nvml;
+    return VirtualMemoryTracker::Instance::walk_virtual_memory(walker);
+  }
+


Added this so that we can do the usual checks and lock taking in the place where we typically take the lock.

mlbridge · 2025-09-11T10:14:50Z

Webrevs

src/hotspot/share/nmt/virtualMemoryTracker.hpp

dholmes-ora · 2025-09-12T02:20:43Z

test/hotspot/gtest/runtime/test_committed_virtualmemory.cpp

+    {
+      MemTracker::NmtVirtualMemoryLocker nvml;
+      VirtualMemoryTracker::Instance::add_reserved_region(stack_end, stack_size, CALLER_PC, mtThreadStack);
+    }

    // snapshot current stack usage
-    VirtualMemoryTracker::Instance::snapshot_thread_stacks();
+    {
+      MemTracker::NmtVirtualMemoryLocker nvml;
+      VirtualMemoryTracker::Instance::snapshot_thread_stacks();
+    }


Why two separate locking blocks?

No reason, let's merge them.

Arraying · 2025-09-12T08:06:04Z

src/hotspot/share/nmt/memTracker.hpp

    }
  }

+  static inline bool walk_virtual_memory(VirtualMemoryWalker* walker) {


Nit: should probably be in memTracker.inline.hpp?

Arraying · 2025-09-12T08:06:52Z

src/hotspot/share/nmt/memTracker.hpp

+    return VirtualMemoryTracker::Instance::walk_virtual_memory(walker);
+  }
+
  static inline MemoryFileTracker::MemoryFile* register_file(const char* descriptive_name) {


Same here. I wonder why we have the inline header when it only contains MemTracker::check_exceeds_limit...

Hmm, so do I. I've never used the inline header myself. I think moving stuff over to the inline header is something we should think about in a separate PR.

I'm okay with that.

Arraying · 2025-09-12T08:09:01Z

src/hotspot/share/nmt/virtualMemoryTracker.cpp

 bool VirtualMemoryTracker::walk_virtual_memory(VirtualMemoryWalker* walker) {
-  MemTracker::NmtVirtualMemoryLocker nvml;
+  bool ret = true;
  tree()->visit_reserved_regions([&](ReservedMemoryRegion& rgn) {
    if (!walker->do_allocation_site(&rgn)) {
+      ret = false;
      return false;
    }
    return true;
  });
-  return true;
+  return ret;
 }


Nice catch!

Arraying

Happy to address the inline header refactoring separately.

Co-authored-by: David Holmes <62092539+dholmes-ora@users.noreply.github.com>

openjdk · 2025-09-12T09:14:01Z

⚠️ @jdksjolen This pull request contains merges that bring in commits not present in the target repository. Since this is not a "merge style" pull request, these changes will be squashed when this pull request in integrated. If this is your intention, then please ignore this message. If you want to preserve the commit structure, you must change the title of this pull request to Merge <project>:<branch> where <project> is the name of another project in the OpenJDK organization (for example Merge jdk:master).

afshin-zafari

LGTM. Thanks.

jdksjolen · 2025-09-15T11:22:31Z

/integrate

Thanks!

openjdk · 2025-09-15T11:24:27Z

Going to push as commit a7dc011.
Since your change was applied there have been 263 commits pushed to the master branch:

1cb1267: 8361711: Add library name configurability to PKCS11Test.java
3db3c06: 8366852: java/awt/Choice/ChoiceMouseWheelTest/ChoiceMouseWheelTest.java test is failing
c16462b: 8367576: JvmtiThreadState::_debuggable is unused
... and 260 more: https://git.openjdk.org/jdk/compare/124575b4c2b52328a8efddb40e67057a53b44a04...master

Your commit was automatically rebased without conflicts.

openjdk · 2025-09-15T11:24:37Z

@jdksjolen Pushed as commit a7dc011.

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

Add missing locks

7caa6fe

openjdk bot changed the title ~~8366658~~ 8366658: Add missing locks when accessing the VirtualMemoryTracker instance in tests and MemMapPrinter Sep 2, 2025

openjdk bot added the hotspot-runtime hotspot-runtime-dev@openjdk.org label Sep 2, 2025

Fix

6904d36

jdksjolen force-pushed the missing-locks branch from 43b7e19 to 6904d36 Compare September 4, 2025 07:31

Fix

e31b4d6

jdksjolen commented Sep 4, 2025

View reviewed changes

Missing locks

20d2f5d

jdksjolen marked this pull request as ready for review September 11, 2025 10:08

openjdk bot added the rfr Pull request is ready for review label Sep 11, 2025

dholmes-ora reviewed Sep 12, 2025

View reviewed changes

src/hotspot/share/nmt/virtualMemoryTracker.hpp Outdated Show resolved Hide resolved

dholmes-ora reviewed Sep 12, 2025

View reviewed changes

Arraying reviewed Sep 12, 2025

View reviewed changes

Arraying approved these changes Sep 12, 2025

View reviewed changes

jdksjolen and others added 3 commits September 12, 2025 11:08

Update src/hotspot/share/nmt/virtualMemoryTracker.hpp

1cb8353

Co-authored-by: David Holmes <62092539+dholmes-ora@users.noreply.github.com>

Merge lock blocks

79218b6

Merge remote-tracking branch 'origin/missing-locks' into missing-locks

2b4f53e

afshin-zafari approved these changes Sep 12, 2025

View reviewed changes

openjdk bot added the ready Pull request is ready to be integrated label Sep 12, 2025

openjdk bot added the integrated Pull request has been integrated label Sep 15, 2025

openjdk bot closed this Sep 15, 2025

openjdk bot removed ready Pull request is ready to be integrated rfr Pull request is ready for review labels Sep 15, 2025

8366658: Add missing locks when accessing the VirtualMemoryTracker instance in tests and MemMapPrinter #27038

8366658: Add missing locks when accessing the VirtualMemoryTracker instance in tests and MemMapPrinter #27038

Uh oh!

Conversation

jdksjolen commented Sep 2, 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 Sep 2, 2025

Uh oh!

openjdk bot commented Sep 2, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

openjdk bot commented Sep 2, 2025

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

mlbridge bot commented Sep 11, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Webrevs

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Arraying left a comment

Choose a reason for hiding this comment

Uh oh!

openjdk bot commented Sep 12, 2025

Uh oh!

afshin-zafari left a comment

Choose a reason for hiding this comment

Uh oh!

jdksjolen commented Sep 15, 2025

Uh oh!

openjdk bot commented Sep 15, 2025

Uh oh!

openjdk bot commented Sep 15, 2025

Uh oh!

Reviewers

Assignees

Labels

Milestone

Development

Uh oh!

4 participants

jdksjolen commented Sep 2, 2025 •

edited by openjdk bot

Loading

openjdk bot commented Sep 2, 2025 •

edited

Loading

mlbridge bot commented Sep 11, 2025 •

edited

Loading