8379501: [lworld] Clean up Valhalla flags by Arraying · Pull Request #2226 · openjdk/valhalla

Arraying · 2026-03-13T16:14:27Z

Hi all,

This PR cleans up Valhalla flags. To summarize (for the exhaustive changes, please refer to the diff):

Many flags have their level bumped down. No product flags remain in Valhalla.
The granular flattening control flags have received consistent names.
Some granular flattening control flags have been experimentalized.
Setting granular flattening control flags will give a VM warning is flattening is not enabled.
Flat array printing has been debugized.
Descriptions are updated.
Updated all tests to use the new flags.

Testing: tiers 1-3.

Progress

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

Issue

JDK-8379501: [lworld] Clean up Valhalla flags (Enhancement - P3)

Reviewers

Frederic Parain (@fparain - Committer) ⚠️ Review applies to da8b93a6
Marc Chevalier (@marc-chevalier - Committer)

Reviewing

Using git

Checkout this PR locally:
$ git fetch https://git.openjdk.org/valhalla.git pull/2226/head:pull/2226
$ git checkout pull/2226

Update a local copy of the PR:
$ git checkout pull/2226
$ git pull https://git.openjdk.org/valhalla.git pull/2226/head

Using Skara CLI tools

Checkout this PR locally:
$ git pr checkout 2226

View PR using the GUI difftool:
$ git pr show -t 2226

Using diff file

Download this PR as a diff file:
https://git.openjdk.org/valhalla/pull/2226.diff

Using Webrev

Link to Webrev Comment

bridgekeeper · 2026-03-13T16:15:35Z

👋 Welcome back phubner! A progress list of the required criteria for merging this PR into lworld will be added to the body of your pull request. There are additional pull request commands available for use with this pull request.

openjdk · 2026-03-13T16:16:39Z

@Arraying 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:

8379501: [lworld] Clean up Valhalla flags

Reviewed-by: fparain, mchevalier

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 108 new commits pushed to the lworld branch:

f12276a: 8379913: [lworld] AppCDS test asserts with -Xcomp --enable-preview
e0c8622: 8369808: [lworld] 8366837 partially reverted duing 26+15 merge
ee1e278: 8379800: [lworld] Do not run TestOopsInReturnConvention.java and TestVirtualThreads.java with Xcomp
... and 105 more: https://git.openjdk.org/valhalla/compare/1257fbdfb3db4bde784c8dab830f2220c2b13544...lworld

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 lworld branch, type /integrate in a new comment.

mlbridge · 2026-03-13T16:34:18Z

Webrevs

stefank

Thank you! The less usage of the word "inline" to describe values, the better IMO. I've skimmed through the patch and I think it looks good. I've added a couple of nits that you might want to consider.

stefank · 2026-03-16T06:58:18Z

    if (!FLAG_IS_DEFAULT(flag)) {                                                                            \
-      warning("Valhalla-specific flag \"%s\" has no effect when --enable-preview is not specified.", #flag); \
+      warning("Preview-specific flag \"%s\" has no effect when --enable-preview is not specified.", #flag);  \


Is this done because we don't want to mention Valhalla to the users?

It's less that we want to explicitly hide Valhalla, it's more that we've transitioned from Valhalla vs non-Valhalla to preview vs non-preview, so our warnings should reflect that as well.

stefank · 2026-03-16T07:04:14Z

@@ -38,106 +38,113 @@ public class InlineTypes {
                         "--add-exports", "java.base/jdk.internal.value=ALL-UNNAMED",
                         "--add-exports", "java.base/jdk.internal.vm.annotation=ALL-UNNAMED",
                         "--add-exports", "java.base/jdk.internal.misc=ALL-UNNAMED",
+                         "-XX:+UnlockDiagnosticVMOptions",


Unrelated to this PR, but I saw this flag:
"-XX:+IgnoreUnrecognizedVMOptions",

I would consider removing from all the tests as a follow-up PR. We've been bitten by that flag so many times in our testing that we shouldn't use it at all, unless there's a written motivation why it is needed. My 2c.

Fair point, seems like a very slippery slope indeed. I've filed JDK-8380140 for compiler since they're the only ones who use the flag in Valhalla. Let's have them take a look and make an informed decision.

We could probably remove the from this test because all the Flattening flags are diagnostic and available in product mode. We've added -XX:+IgnoreUnrecognizedVMOptions for tests that pass develop options. I suppose if the test is trying to test the develop option there's probably a @requires vm.debug that should be on these tests instead.
This could remove these I think and leave the others for another day.

Are there Experimental options, requiring adding UnlockExperimentalVMOptions ?

This PR already has a large and invasive scope, I am not super comfortable changing anything that's not strictly necessary. Yes, there is lot's of cleanup that can be done (and flag duplication in some tests), but I would prefer to keep this focused on what's needed for the CSR.

The options listed in this scenario (and other scenarios) do not require experimental VM options. Only null-free flattening is experimental, the rest is diagnostic.

Co-authored-by: Stefan Karlsson <stefan.karlsson@oracle.com>

This reverts commit 768fd17.

This reverts commit 382ea88.

Arraying · 2026-03-20T08:57:03Z

Based on several rounds of offline feedback, I've updated this PR to get rid of the inline renames and changed the scope of this PR to the bare minimum. Further work needs to be done to make the names consistent.

stefank

Removing earlier approval to let others review this properly.

fparain

LGTM

marc-chevalier

Looks fine to me. Thanks.

coleenp

I have a couple of questions about the whole change. Thanks for making the options diagnostic.

coleenp · 2026-03-26T17:51:11Z

@@ -38,106 +38,113 @@ public class InlineTypes {
                         "--add-exports", "java.base/jdk.internal.value=ALL-UNNAMED",
                         "--add-exports", "java.base/jdk.internal.vm.annotation=ALL-UNNAMED",
                         "--add-exports", "java.base/jdk.internal.misc=ALL-UNNAMED",
+                         "-XX:+UnlockDiagnosticVMOptions",


We could probably remove the from this test because all the Flattening flags are diagnostic and available in product mode. We've added -XX:+IgnoreUnrecognizedVMOptions for tests that pass develop options. I suppose if the test is trying to test the develop option there's probably a @requires vm.debug that should be on these tests instead.
This could remove these I think and leave the others for another day.

coleenp · 2026-03-26T17:51:53Z

@@ -38,106 +38,113 @@ public class InlineTypes {
                         "--add-exports", "java.base/jdk.internal.value=ALL-UNNAMED",
                         "--add-exports", "java.base/jdk.internal.vm.annotation=ALL-UNNAMED",
                         "--add-exports", "java.base/jdk.internal.misc=ALL-UNNAMED",
+                         "-XX:+UnlockDiagnosticVMOptions",


Are there Experimental options, requiring adding UnlockExperimentalVMOptions ?

marc-chevalier

Still looks good to me. If tests are happy, it's fine! Also, if we find later that some unlock flag is missing in some weird test at some weird tier, it will not be a big disruption, and an easy fix. No need to be too scared.

Arraying · 2026-03-27T12:19:58Z

Thanks for the reviews everybody.
/integrate

openjdk · 2026-03-27T12:21:49Z

Going to push as commit 559e0e4.
Since your change was applied there have been 108 commits pushed to the lworld branch:

f12276a: 8379913: [lworld] AppCDS test asserts with -Xcomp --enable-preview
e0c8622: 8369808: [lworld] 8366837 partially reverted duing 26+15 merge
ee1e278: 8379800: [lworld] Do not run TestOopsInReturnConvention.java and TestVirtualThreads.java with Xcomp
... and 105 more: https://git.openjdk.org/valhalla/compare/1257fbdfb3db4bde784c8dab830f2220c2b13544...lworld

Your commit was automatically rebased without conflicts.

openjdk · 2026-03-27T12:21:58Z

@Arraying Pushed as commit 559e0e4.

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

Arraying added 6 commits March 13, 2026 14:44

Make many flags diagnostic flags.

ba62442

Better messages and ergonomics.

7ef7bba

Make sure tests have proper diagnostics.

13dc9cb

Proper naming of flattening flags and textual cleanups.

0d09933

Rename PrintInlineLayout to PrintValueLayout.

382ea88

Rename InlineTypeX to ValueTypeX.

768fd17

openjdk bot changed the title ~~8379501~~ 8379501: [lworld] Clean up Valhalla flags Mar 13, 2026

Fix Copyright.

fec5517

Arraying marked this pull request as ready for review March 13, 2026 16:28

openjdk bot added the rfr Pull request is ready for review label Mar 13, 2026

stefank reviewed Mar 16, 2026

View reviewed changes

Arraying and others added 2 commits March 16, 2026 10:56

Update src/hotspot/share/opto/parse2.cpp

d956285

Co-authored-by: Stefan Karlsson <stefan.karlsson@oracle.com>

Update src/hotspot/share/runtime/arguments.cpp

7fccbce

Co-authored-by: Stefan Karlsson <stefan.karlsson@oracle.com>

stefank approved these changes Mar 16, 2026

View reviewed changes

openjdk bot added the ready Pull request is ready to be integrated label Mar 16, 2026

Arraying added 3 commits March 18, 2026 17:18

Revert "Rename InlineTypeX to ValueTypeX."

843f359

This reverts commit 768fd17.

Revert "Rename PrintInlineLayout to PrintValueLayout."

cfb2114

This reverts commit 382ea88.

Adjust level of flattening flags.

6ec2e46

Arraying marked this pull request as draft March 18, 2026 17:20

openjdk bot removed ready Pull request is ready to be integrated rfr Pull request is ready for review labels Mar 18, 2026

Arraying added 3 commits March 19, 2026 13:56

Diagnosticize compiler flag.

57dab69

Experimentalize some flattening flags.

bc91532

Fix test failure.

c40662e

Arraying marked this pull request as ready for review March 20, 2026 08:57

openjdk bot added the ready Pull request is ready to be integrated label Mar 20, 2026

openjdk bot added the rfr Pull request is ready for review label Mar 20, 2026

stefank suggested changes Mar 20, 2026

View reviewed changes

openjdk bot removed the ready Pull request is ready to be integrated label Mar 20, 2026

fparain reviewed Mar 23, 2026

View reviewed changes

Comment thread src/hotspot/share/runtime/globals.hpp Outdated

Arraying added 2 commits March 24, 2026 14:53

Merge branch 'lworld' into JDK-8379501

f57be49

Experimental -> Diagnostic.

da8b93a

fparain approved these changes Mar 24, 2026

View reviewed changes

openjdk bot added the ready Pull request is ready to be integrated label Mar 24, 2026

marc-chevalier approved these changes Mar 26, 2026

View reviewed changes

Add missed diagnostic flags.

2b51358

coleenp reviewed Mar 26, 2026

View reviewed changes

Fix more tests.

b8bead8

marc-chevalier approved these changes Mar 27, 2026

View reviewed changes

openjdk bot added the integrated Pull request has been integrated label Mar 27, 2026

openjdk bot closed this Mar 27, 2026

openjdk bot removed ready Pull request is ready to be integrated rfr Pull request is ready for review labels Mar 27, 2026

Conversation

Arraying commented Mar 13, 2026 • 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 Mar 13, 2026

Uh oh!

openjdk bot commented Mar 13, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

mlbridge bot commented Mar 13, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Webrevs

Uh oh!

stefank left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

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!

Arraying commented Mar 20, 2026

Uh oh!

stefank left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

fparain left a comment

Choose a reason for hiding this comment

Uh oh!

marc-chevalier left a comment

Choose a reason for hiding this comment

Uh oh!

coleenp left a comment

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!

marc-chevalier left a comment

Choose a reason for hiding this comment

Uh oh!

Arraying commented Mar 27, 2026

Uh oh!

openjdk bot commented Mar 27, 2026

Uh oh!

openjdk bot commented Mar 27, 2026

Uh oh!

Reviewers

Assignees

Labels

Milestone

Development

Uh oh!

5 participants

Arraying commented Mar 13, 2026 •

edited by openjdk bot

Loading

openjdk bot commented Mar 13, 2026 •

edited

Loading

mlbridge bot commented Mar 13, 2026 •

edited

Loading