8325028: (ch) Pipe channels should lazily set socket to non-blocking mode on first use by virtual thread

[vpa1977]

Hi all,

This pull request contains a backport of commit d1099033 from the openjdk/jdk repository.

The commit being backported was authored by Alan Bateman on 8 Feb 2024 and was reviewed by Brian Burkhalter.

This change fixes a bug in jruby[1] that broken pipeline functions of Open3[1][2].
This change is high risk as there is no jtreg test directly testing the pipe behaviour.
Patch applies cleanly to 21u.
The patch was tested against JRuby test suite in Ubuntu. Without the patch the following tests[3] fail:

1) Failure:
TestOpen3#test_pipeline_start [/tmp/autopkgtest.Dz4xQk/autopkgtest_tmp/test/mri/test_open3.rb:303]:
Failed assertion, no message given.

  2) Failure:
TestOpen3#test_pipeline_w [/tmp/autopkgtest.Dz4xQk/autopkgtest_tmp/test/mri/test_open3.rb:288]:
<true> expected but was
<false>.

  3) Failure:
TestOpen3#test_capture3_flip [/tmp/autopkgtest.Dz4xQk/autopkgtest_tmp/test/mri/test_open3.rb:190]:

With the patch, they pass. Also the reproducer

require 'open3'

RUBY = 'jruby'
Open3.pipeline_r([RUBY, '-e', 'print "1"'],
                     [RUBY, '-e', 'print STDIN.read + "2"']) {|o,ts|
      print(o.read)
    }
---

outputs 12 as expected in multiple runs.
There is no regression in tier1 and tier2 tests ran on Ubuntu.

Thanks!

[1] jruby/jruby#8069
[2] https://bugs.launchpad.net/ubuntu/+source/jruby/+bug/2054943
[3] https://bugs.launchpad.net/ubuntu/+source/openjdk-21/+bug/2054943

---

Progress

• Change must not contain extraneous whitespace
• Commit message must refer to an issue
• JDK-8325028 needs maintainer approval

Issue

• JDK-8325028: (ch) Pipe channels should lazily set socket to non-blocking mode on first use by virtual thread (Enhancement - P4 - Approved)

Reviewing

Using git

Checkout this PR locally:
$ git fetch https://git.openjdk.org/jdk21u-dev.git pull/290/head:pull/290
$ git checkout pull/290

Update a local copy of the PR:
$ git checkout pull/290
$ git pull https://git.openjdk.org/jdk21u-dev.git pull/290/head

Using Skara CLI tools

Checkout this PR locally:
$ git pr checkout 290

View PR using the GUI difftool:
$ git pr show -t 290

Using diff file

Download this PR as a diff file:
https://git.openjdk.org/jdk21u-dev/pull/290.diff

Webrev

Link to Webrev Comment

[bridgekeeper]

👋 Welcome back vpetko! 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]

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

[openjdk]

⚠️ @vpa1977 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.

[mlbridge]

Webrevs

• 01: Full - Incremental (086bc625)
• 00: Full (6ac72c28)

[GoeLin]

@vpa1977 please also backport this to jdk22u.

[vpa1977]

@vpa1977 please also backport this to jdk22u.
I apologise for forgetting this. I have created openjdk/jdk22u#79 .

[GoeLin]

Thanks. I also ran it through our nightly testing. No issues found.

[vpa1977]

/approval request

[openjdk]

@vpa1977
8325028: The approval request has been created successfully.

[openjdk]

@vpa1977 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:

8325028: (ch) Pipe channels should lazily set socket to non-blocking mode on first use by virtual thread

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 24 new commits pushed to the master branch:

• 7c63791: 8319955: Improve dependencies removal during class unloading
• 2a3c4fc: 8318109: Writing JFR records while a CHT has taken its lock asserts in rank checking
• 5a94e48: 8317600: VtableStubs::stub_containing() table load not ordered wrt to stores
• 41c8606: 8316959: Improve InlineCacheBuffer pending queue management
• f341621: 8323685: PrintSystemDictionaryAtExit has mutex rank assert
• e63addb: 8315503: G1: Code root scan causes long GC pauses due to imbalanced iteration
• 686399a: 8318599: HttpURLConnection cache issues leading to crashes in JGSS w/ native GSS introduced by 8303809
• 6c5500b: 8318986: Improve GenericWaitBarrier performance
• f0f2e70: 8326936: RISC-V: Shenandoah GC crashes due to incorrect atomic memory operations
• 7743b6c: Merge
• ... and 14 more: https://git.openjdk.org/jdk21u-dev/compare/18254641606816215189a06a281e1763fd86eecf...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.

➡️ 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).

[GoeLin]

Hi @vpa1977
please have a look at https://wiki.openjdk.org/display/JDKUpdates/How+to+contribute+or+backport+a+fix to write a proper fix request comment. Removed the tag for now.

[vpa1977]

Hi @vpa1977 please have a look at https://wiki.openjdk.org/display/JDKUpdates/How+to+contribute+or+backport+a+fix to write a proper fix request comment. Removed the tag for now.

Thank you!!!
I have added the comment, but did not set the tag. Please advice if i should add more details before tagging the issue?

[vpa1977]

/integrate

[openjdk]

@vpa1977
Your change (at version 6ac72c2) is now ready to be sponsored by a Committer.

[shipilev]

Hey @vpa1977, the last GHA test run was back in Feburary. Please merge from master and see that GHAs are still green? After that, I can sponsor.

[vpa1977]

@shipilev Thank you!!! I've merged master and the tests passed.

[vpa1977]

/sponsor

[openjdk]

@vpa1977 Only Committers are allowed to sponsor changes.

[vpa1977]

/integrate

[openjdk]

@vpa1977
Your change (at version 086bc62) is now ready to be sponsored by a Committer.

[shipilev]

/sponsor

[openjdk]

Going to push as commit 667513a.
Since your change was applied there have been 25 commits pushed to the master branch:

• 515783b: 8328166: Epsilon: 'EpsilonHeap::allocate_work' misuses the parameter 'size' as size in bytes
• 7c63791: 8319955: Improve dependencies removal during class unloading
• 2a3c4fc: 8318109: Writing JFR records while a CHT has taken its lock asserts in rank checking
• 5a94e48: 8317600: VtableStubs::stub_containing() table load not ordered wrt to stores
• 41c8606: 8316959: Improve InlineCacheBuffer pending queue management
• f341621: 8323685: PrintSystemDictionaryAtExit has mutex rank assert
• e63addb: 8315503: G1: Code root scan causes long GC pauses due to imbalanced iteration
• 686399a: 8318599: HttpURLConnection cache issues leading to crashes in JGSS w/ native GSS introduced by 8303809
• 6c5500b: 8318986: Improve GenericWaitBarrier performance
• f0f2e70: 8326936: RISC-V: Shenandoah GC crashes due to incorrect atomic memory operations
• ... and 15 more: https://git.openjdk.org/jdk21u-dev/compare/18254641606816215189a06a281e1763fd86eecf...master

Your commit was automatically rebased without conflicts.

[openjdk]

@shipilev @vpa1977 Pushed as commit 667513a.

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