8348239: SA does not know about DeoptimizeObjectsALotThread #23279

plummercj · 2025-01-23T20:15:35Z

If you run hotspot with -XX:+DeoptimizeObjectsALot, it will create one or more DeoptimizeObjectsALotThread threads. This is one of many JavaThread subclasses that SA needs to know about, but in this case it does not. When SA tries to create a mirror of one of these threads, it fails with:

 stderr: [java.lang.RuntimeException: Unable to deduce type of thread from address 0x0000ffff34144e30 (expected type JavaThread, CompilerThread, MonitorDeflationThread, AttachListenerThread, StringDedupThread, NotificationThread, ServiceThread or JvmtiAgentThread)
at jdk.hotspot.agent/sun.jvm.hotspot.runtime.Threads.createJavaThreadWrapper(Threads.java:196)
at jdk.hotspot.agent/sun.jvm.hotspot.runtime.Threads.getJavaThreadAt(Threads.java:178)

The following can be used to reproduce this failure. Most SA tests will fail:

make test TEST=serviceability/sa TEST_VM_OPTS=-XX:+DeoptimizeObjectsALot

I had to move the native DeoptimizeObjectsALotThread declaration to compileBroker.hpp to make it visible to vmStructs.cpp.

I tested with the above "make test" command and ran all svc tier1, tier2, tier3, and tier5 tests.

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-8348239: SA does not know about DeoptimizeObjectsALotThread (Bug - P4)

Reviewers

Dean Long (@dean-long - Reviewer) Review applies to 8ba5e23a
David Holmes (@dholmes-ora - Reviewer) Review applies to 8ba5e23a
Leonid Mesnik (@lmesnik - Reviewer) Review applies to 8ba5e23a
Kevin Walls (@kevinjwalls - Reviewer)

Reviewing

Using git

Checkout this PR locally:
$ git fetch https://git.openjdk.org/jdk.git pull/23279/head:pull/23279
$ git checkout pull/23279

Update a local copy of the PR:
$ git checkout pull/23279
$ git pull https://git.openjdk.org/jdk.git pull/23279/head

Using Skara CLI tools

Checkout this PR locally:
$ git pr checkout 23279

View PR using the GUI difftool:
$ git pr show -t 23279

Using diff file

Download this PR as a diff file:
https://git.openjdk.org/jdk/pull/23279.diff

Using Webrev

Link to Webrev Comment

bridgekeeper · 2025-01-23T20:16:29Z

👋 Welcome back cjplummer! 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 · 2025-01-23T20:17:41Z

@plummercj 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:

8348239: SA does not know about DeoptimizeObjectsALotThread

Reviewed-by: kevinw, dlong, dholmes, lmesnik

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 63 new commits pushed to the master branch:

dec9367: 8347498: JDK 24 RDP2 L10n resource files update
bef5f56: 8291027: Some of TimeZone methods marked 'synchronized' unnecessarily
8e8f800: 8344943: Mark not subclassable classes final in java.base exported classes
76f792b: 8348203: [JVMCI] Make eager JVMCI initialization observable in the debugger
909cef5: 8339891: Several sun/security/ssl/SSLSessionImpl/* tests override test.java.opts
56a1877: 8348170: Unnecessary Hashtable usage in CSS.styleConstantToCssMap
9768f60: 8345249: Apply some conservative cleanups in FileURLConnection
c5a69b6: 8331723: Serial: Remove the unused parameter of the method SerialHeap::gc_prologue
9c55e25: 8347981: RISC-V: Add Zfa zli imm loads
e20bd01: 8344361: Restore null return for invalid services from legacy providers
... and 53 more: https://git.openjdk.org/jdk/compare/7a6ff9206a0a4d116dec542819d51daa558d200d...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 · 2025-01-23T20:18:33Z

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

hotspot
serviceability

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 · 2025-01-23T20:21:23Z

Webrevs

01: Full - Incremental (5779c9eb)
00: Full (8ba5e23a)

dean-long · 2025-01-23T22:01:14Z

src/jdk.hotspot.agent/share/classes/sun/jvm/hotspot/runtime/Thread.java

  public boolean   isServiceThread()             { return false; }
  public boolean   isMonitorDeflationThread()    { return false; }
  public boolean   isAttachListenerThread()      { return false; }
+  public boolean   isDeoptimizeObjectsALotThread() { return false; }


Is this method used anywhere?

No, but I added it to be consistent with all the other JavaThread subclasses. Actually none of these are used except for isCompilerThread and isCodeCacheSweeperThread. The latter reference is irrelevant since there is no longer a CodeCacheSweeperThread, and the former reference should be replaced with isHiddenFromExternalView(). These are all things documented to fix as part of JDK-8348347, but in the meantime I though it best to be consistent with existing code.

dean-long

Overall the changes seem fine, however it would be nice if the vmStructs.cpp registration mechanism was modular and extensible, instead of monolithic. That would make it easier to register private types, or even types outside of libjvm.

plummercj · 2025-01-23T22:30:48Z

Overall the changes seem fine, however it would be nice if the vmStructs.cpp registration mechanism was modular and extensible, instead of monolithic. That would make it easier to register private types, or even types outside of libjvm.

I'm not too sure what you are suggesting here.

dholmes-ora

Okay as a quick fix but I look forward to all this going away.

Thanks

kevinjwalls · 2025-01-24T10:19:30Z

src/jdk.hotspot.agent/share/classes/sun/jvm/hotspot/runtime/Threads.java

-      StringDedupThread, AttachListenerThread and ServiceThread.
-      The latter seven subclasses of the former. Most operations
+      StringDedupThread, AttachListenerThread, DeoptimizeObjectsALotThread and
+      ServiceThread. The latter seven subclasses of the former. Most operations


The latter eight are subclasses of the former? 8-) Or just "JavaThread, or its subclasses..."

Yes, it should be updated to eight, but I think what would be better is "returns objects of type JavaThraed or one of its subclasses..."

Ok, it's fixed now. I stuck with a somewhat minimal update mostly just to resolve the incorrect count. This comment may have further cleanup as part of JDK-8348347.

dean-long · 2025-01-24T22:52:19Z

Overall the changes seem fine, however it would be nice if the vmStructs.cpp registration mechanism was modular and extensible, instead of monolithic. That would make it easier to register private types, or even types outside of libjvm.

I'm not too sure what you are suggesting here.

For example, instead of a single table, allow multiple tables. They could be registered with with an API similar to JNI RegisterNatives.

plummercj · 2025-01-24T23:23:43Z

Overall the changes seem fine, however it would be nice if the vmStructs.cpp registration mechanism was modular and extensible, instead of monolithic. That would make it easier to register private types, or even types outside of libjvm.

I'm not too sure what you are suggesting here.

For example, instead of a single table, allow multiple tables. They could be registered with with an API similar to JNI RegisterNatives.

I don't see how that relates to this change. It sounds more like a general suggestion to overhaul of how vmStructs is implemented.

dean-long · 2025-01-25T01:26:50Z

I don't see how that relates to this change. It sounds more like a general suggestion to overhaul of how vmStructs is implemented.

It relates because if something like that was available then we could register types that are local to a .cpp file without moving them to a .hpp first.

plummercj · 2025-01-27T19:44:10Z

Thanks for the reviews David, Leonid, Dean, and Kevin

/integrate

openjdk · 2025-01-27T19:45:51Z

Going to push as commit 21feef3.
Since your change was applied there have been 85 commits pushed to the master branch:

8cc1304: 8287788: Implement a better allocator for downcalls
039e73f: 8346736: Java Security Standard Algorithm Names spec should include key algorithm names
aba60a9: 8189441: Define algorithm names for keys derived from KeyAgreement
03106eb: 8344119: CUPSPrinter does not respect PostScript printer definition specification in case of reading ImageableArea values from PPD files
ad01dfb: 8346920: Serial: Support allocation in old generation when heap is almost full
1d8ccb8: 8342465: Improve API documentation for java.lang.classfile
7d6055a: 8348429: Update cross-compilation devkits to Fedora 41/gcc 13.2
f1e0797: 8348586: Optionally silence make warnings about non-control variables
ffeb9b5: 8342807: Update links in java.base to use https://
afcc2b0: 8348562: ZGC: segmentation fault due to missing node type check in barrier elision analysis
... and 75 more: https://git.openjdk.org/jdk/compare/7a6ff9206a0a4d116dec542819d51daa558d200d...master

Your commit was automatically rebased without conflicts.

openjdk · 2025-01-27T19:46:11Z

@plummercj Pushed as commit 21feef3.

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

Add DeoptimizeObjectsALotThread support

8ba5e23

openjdk bot changed the title ~~8348239~~ 8348239: SA does not know about DeoptimizeObjectsALotThread Jan 23, 2025

openjdk bot added the rfr Pull request is ready for review label Jan 23, 2025

openjdk bot added serviceability serviceability-dev@openjdk.org hotspot hotspot-dev@openjdk.org labels Jan 23, 2025

dean-long reviewed Jan 23, 2025

View reviewed changes

dean-long approved these changes Jan 23, 2025

View reviewed changes

openjdk bot added the ready Pull request is ready to be integrated label Jan 23, 2025

dholmes-ora approved these changes Jan 24, 2025

View reviewed changes

lmesnik approved these changes Jan 24, 2025

View reviewed changes

kevinjwalls reviewed Jan 24, 2025

View reviewed changes

kevinjwalls approved these changes Jan 24, 2025

View reviewed changes

Cleanup comment

5779c9e

openjdk bot removed the ready Pull request is ready to be integrated label Jan 24, 2025

kevinjwalls approved these changes Jan 24, 2025

View reviewed changes

openjdk bot added the ready Pull request is ready to be integrated label Jan 24, 2025

openjdk bot added the integrated Pull request has been integrated label Jan 27, 2025

openjdk bot closed this Jan 27, 2025

openjdk bot removed ready Pull request is ready to be integrated rfr Pull request is ready for review labels Jan 27, 2025

8348239: SA does not know about DeoptimizeObjectsALotThread #23279

8348239: SA does not know about DeoptimizeObjectsALotThread #23279

Uh oh!

Conversation

plummercj commented Jan 23, 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 Jan 23, 2025

Uh oh!

openjdk bot commented Jan 23, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

openjdk bot commented Jan 23, 2025

Uh oh!

mlbridge bot commented Jan 23, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Webrevs

Uh oh!

dean-long Jan 23, 2025

Choose a reason for hiding this comment

Uh oh!

plummercj Jan 23, 2025

Choose a reason for hiding this comment

Uh oh!

dean-long left a comment

Choose a reason for hiding this comment

Uh oh!

plummercj commented Jan 23, 2025

Uh oh!

dholmes-ora left a comment

Choose a reason for hiding this comment

Uh oh!

kevinjwalls Jan 24, 2025

Choose a reason for hiding this comment

Uh oh!

plummercj Jan 24, 2025

Choose a reason for hiding this comment

Uh oh!

plummercj Jan 24, 2025

Choose a reason for hiding this comment

Uh oh!

dean-long commented Jan 24, 2025

Uh oh!

plummercj commented Jan 24, 2025

Uh oh!

dean-long commented Jan 25, 2025

Uh oh!

plummercj commented Jan 27, 2025

Uh oh!

openjdk bot commented Jan 27, 2025

Uh oh!

openjdk bot commented Jan 27, 2025

Uh oh!

Reviewers

Assignees

Labels

Milestone

Development

Uh oh!

5 participants

plummercj commented Jan 23, 2025 •

edited by openjdk bot

Loading

openjdk bot commented Jan 23, 2025 •

edited

Loading

mlbridge bot commented Jan 23, 2025 •

edited

Loading