8269870: PS: Membar in PSPromotionManager::copy_unmarked_to_survivor_space could be relaxed by albertnetymk · Pull Request #19764 · openjdk/jdk

albertnetymk · 2024-06-18T10:17:53Z

The acquire-release mem-barrier was probably used/required before but not any more. Inspiration taken from G1ParScanThreadState::do_oop_evac.

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-8269870: PS: Membar in PSPromotionManager::copy_unmarked_to_survivor_space could be relaxed (Enhancement - P4)

Reviewers

Hamlin Li (@Hamlin-Li - Reviewer) ⚠️ Review applies to c1b5336f
Kim Barrett (@kimbarrett - Reviewer)
Thomas Schatzl (@tschatzl - Reviewer)

Contributors

Hamlin Li <mli@openjdk.org>

Reviewing

Using git

Checkout this PR locally:
$ git fetch https://git.openjdk.org/jdk.git pull/19764/head:pull/19764
$ git checkout pull/19764

Update a local copy of the PR:
$ git checkout pull/19764
$ git pull https://git.openjdk.org/jdk.git pull/19764/head

Using Skara CLI tools

Checkout this PR locally:
$ git pr checkout 19764

View PR using the GUI difftool:
$ git pr show -t 19764

Using diff file

Download this PR as a diff file:
https://git.openjdk.org/jdk/pull/19764.diff

Webrev

Link to Webrev Comment

bridgekeeper · 2024-06-18T10:18:42Z

👋 Welcome back ayang! 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 · 2024-06-18T10:19:01Z

@albertnetymk 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:

8269870: PS: Membar in PSPromotionManager::copy_unmarked_to_survivor_space could be relaxed

Co-authored-by: Hamlin Li <mli@openjdk.org>
Reviewed-by: mli, kbarrett, tschatzl

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 75 new commits pushed to the master branch:

a5f401f: 8334650: Add debug information about whether an Assertion Predicate is for the init or last value
25c3845: 8333133: Simplify QuickSort::sort
c66f785: 8334505: RISC-V: Several tests fail when MaxVectorSize does not match VM_Version::_initial_vector_length
f101e15: 8333583: Crypto-XDH.generateSecret regression after JDK-8329538
b3bf31a: 8333542: Breakpoint in parallel code does not work
86b0cf2: 8334653: ISO 4217 Amendment 177 Update
861aefc: 8334418: Update IANA Language Subtag Registry to Version 2024-06-14
f8bf470: 8334810: Redo: Un-ProblemList LocaleProvidersRun and CalendarDataRegression
933eaba: 8334629: [BACKOUT] PhaseIdealLoop::conditional_move is too conservative
7429c37: 8334598: Default classlist in JDK is not deterministic after JDK-8293980
... and 65 more: https://git.openjdk.org/jdk/compare/78682fe78e18268b1857855c3595b4d118808c66...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.

albertnetymk · 2024-06-18T10:19:35Z

/contributor add mli

openjdk · 2024-06-18T10:19:36Z

@albertnetymk
Contributor Hamlin Li <mli@openjdk.org> successfully added.

openjdk · 2024-06-18T10:20:12Z

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

hotspot-gc

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 · 2024-06-18T10:24:46Z

Webrevs

Hamlin-Li

Looks good.

Hamlin-Li · 2024-06-18T13:52:50Z

Thanks for working on this. :)

I suppose with the removal of OrderAccess::acquire();, it brings even more benefit?

openjdk · 2024-06-19T19:43:52Z

@albertnetymk Please do not rebase or force-push to an active PR as it invalidates existing review comments. Note for future reference, the bots always squash all changes into a single commit automatically as part of the integration. See OpenJDK Developers’ Guide for more information.

albertnetymk · 2024-06-19T19:44:40Z

Missed the acquire in the cas-fail path in the first version. Fixed now.

kimbarrett · 2024-06-26T00:30:47Z

src/hotspot/share/gc/parallel/psPromotionManager.inline.hpp

  // Now we have to CAS in the header.
-  // Make copy visible to threads reading the forwardee.
-  oop forwardee = o->forward_to_atomic(new_obj, test_mark, memory_order_release);
+  oop forwardee = o->forward_to_atomic(new_obj, test_mark, memory_order_relaxed);


I think this needs a comment similar to what's in G1ParScanThreadState::do_copy_to_suvivor_space, e.g.

jdk/src/hotspot/share/gc/g1/g1ParScanThreadState.cpp

Lines 504 to 507 in f101e15

// Because the forwarding is done with memory_order_relaxed there is no

// ordering with the above copy. Clients that get the forwardee must not

// examine its contents without other synchronization, since the contents

// may not be up to date for them.

kimbarrett

Looks good.

tschatzl

Lgtm.

Fwiw, now that it is established that nobody is looking at the object's contents, PR#17227 can be resurrected too.

albertnetymk · 2024-06-26T07:39:54Z

Thanks for review.

/integrate

openjdk · 2024-06-26T07:40:36Z

Going to push as commit b88af94.
Since your change was applied there have been 75 commits pushed to the master branch:

a5f401f: 8334650: Add debug information about whether an Assertion Predicate is for the init or last value
25c3845: 8333133: Simplify QuickSort::sort
c66f785: 8334505: RISC-V: Several tests fail when MaxVectorSize does not match VM_Version::_initial_vector_length
f101e15: 8333583: Crypto-XDH.generateSecret regression after JDK-8329538
b3bf31a: 8333542: Breakpoint in parallel code does not work
86b0cf2: 8334653: ISO 4217 Amendment 177 Update
861aefc: 8334418: Update IANA Language Subtag Registry to Version 2024-06-14
f8bf470: 8334810: Redo: Un-ProblemList LocaleProvidersRun and CalendarDataRegression
933eaba: 8334629: [BACKOUT] PhaseIdealLoop::conditional_move is too conservative
7429c37: 8334598: Default classlist in JDK is not deterministic after JDK-8293980
... and 65 more: https://git.openjdk.org/jdk/compare/78682fe78e18268b1857855c3595b4d118808c66...master

Your commit was automatically rebased without conflicts.

openjdk · 2024-06-26T07:40:45Z

@albertnetymk Pushed as commit b88af94.

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

openjdk bot changed the title ~~8269870~~ 8269870: PS: Membar in PSPromotionManager::copy_unmarked_to_survivor_space could be relaxed Jun 18, 2024

openjdk bot added the rfr Pull request is ready for review label Jun 18, 2024

openjdk bot added the hotspot-gc hotspot-gc-dev@openjdk.org label Jun 18, 2024

Hamlin-Li approved these changes Jun 18, 2024

View reviewed changes

openjdk bot added the ready Pull request is ready to be integrated label Jun 18, 2024

pgc-relax-forward

fb4a255

albertnetymk force-pushed the pgc-relax-forward branch from c1b5336 to fb4a255 Compare June 19, 2024 19:43

kimbarrett suggested changes Jun 26, 2024

View reviewed changes

review

58e3807

kimbarrett approved these changes Jun 26, 2024

View reviewed changes

tschatzl approved these changes Jun 26, 2024

View reviewed changes

openjdk bot added the integrated Pull request has been integrated label Jun 26, 2024

openjdk bot closed this Jun 26, 2024

openjdk bot removed ready Pull request is ready to be integrated rfr Pull request is ready for review labels Jun 26, 2024

albertnetymk deleted the pgc-relax-forward branch June 26, 2024 07:40

	// Because the forwarding is done with memory_order_relaxed there is no
	// ordering with the above copy. Clients that get the forwardee must not
	// examine its contents without other synchronization, since the contents
	// may not be up to date for them.

Conversation

albertnetymk commented Jun 18, 2024 • edited by openjdk bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Progress

Issue

Reviewers

Contributors

Reviewing

Webrev

Uh oh!

bridgekeeper bot commented Jun 18, 2024

Uh oh!

openjdk bot commented Jun 18, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

albertnetymk commented Jun 18, 2024

Uh oh!

openjdk bot commented Jun 18, 2024

Uh oh!

openjdk bot commented Jun 18, 2024

Uh oh!

mlbridge bot commented Jun 18, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Webrevs

Uh oh!

Hamlin-Li left a comment

Choose a reason for hiding this comment

Uh oh!

Hamlin-Li commented Jun 18, 2024

Uh oh!

openjdk bot commented Jun 19, 2024

Uh oh!

albertnetymk commented Jun 19, 2024

Uh oh!

kimbarrett Jun 26, 2024

Choose a reason for hiding this comment

Uh oh!

kimbarrett left a comment

Choose a reason for hiding this comment

Uh oh!

tschatzl left a comment

Choose a reason for hiding this comment

Uh oh!

albertnetymk commented Jun 26, 2024

Uh oh!

openjdk bot commented Jun 26, 2024

Uh oh!

openjdk bot commented Jun 26, 2024

Uh oh!

Reviewers

Assignees

Labels

Milestone

Development

Uh oh!

4 participants

albertnetymk commented Jun 18, 2024 •

edited by openjdk bot

Loading

openjdk bot commented Jun 18, 2024 •

edited

Loading

mlbridge bot commented Jun 18, 2024 •

edited

Loading