8345181: (ch) Windows asynchronous channels may return temporary direct buffers to the buffer cache twice (win)

[bplb]

Due to operations on different threads, the Windows asynchronous channels might try to return the same buffer to the temporary direct buffer cache twice resulting in an attempt to free the same address more than once. This was not a problem before #22339 was integrated, as previously a cleaner had been used to free temporary buffers, and the cleaner did not permit duplicative freeing of the same memory. This change introduces checks in the Windows asynchronous channels to prevent duplicative returns of a buffer to the cache.

---

Progress

• Change must not contain extraneous whitespace
• Commit message must refer to an issue
• Change must be properly reviewed (2 reviews required, with at least 1 Reviewer, 1 Author)

Issue

• JDK-8345181: (ch) Windows asynchronous channels may return temporary direct buffers to the buffer cache twice (win) (Bug - P2)

Reviewers

• Alan Bateman (@AlanBateman - Reviewer)
• Michael McMahon (@Michael-Mc-Mahon - Reviewer)

Reviewing

Using git

Checkout this PR locally:
$ git fetch https://git.openjdk.org/jdk.git pull/22554/head:pull/22554
$ git checkout pull/22554

Update a local copy of the PR:
$ git checkout pull/22554
$ git pull https://git.openjdk.org/jdk.git pull/22554/head

Using Skara CLI tools

Checkout this PR locally:
$ git pr checkout 22554

View PR using the GUI difftool:
$ git pr show -t 22554

Using diff file

Download this PR as a diff file:
https://git.openjdk.org/jdk/pull/22554.diff

Using Webrev

Link to Webrev Comment

[bridgekeeper]

👋 Welcome back bpb! 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]

@bplb 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:

8345181: (ch) Windows asynchronous channels may return temporary direct buffers to the buffer cache twice (win)

Reviewed-by: alanb, michaelm

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 8 new commits pushed to the master branch:

• 1d3dcdf: 8345433: (fs) Use stream to load FileTypeDetectors in Files.probeContentType
• 9740c1d: 8333572: Update jpackage.md with info related to WiX v4 and v5 support
• de3a218: 8345502: Remove doIntersectionPrivilege methods
• faa109e: 8340580: Characters in Document diagrams not in the middle
• 6aa7667: 8339535: JVM crashes with -Xshare:dump -XX:+SegmentedCodeCache
• e1695f6: 8345472: Fix incorrect format instruction for floating point max/min patterns
• baa6957: 8345507: Fix build of static launcher
• 79eb77b: 8345486: Reevaluate the classes in java.lang.classfile.components package

Please see this link for an up-to-date comparison between the source branch of this pull request and the master branch.
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]

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

• nio

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]

Webrevs

• 00: Full (a736b7c2)

[bplb]

In further testing, with this change, CI test tiers 1-3 and 500 repeats of the subset of tests where the failure was observed passed on Windows.

[AlanBateman]

Just more context here. On Windows, when a file or socket handle is associated with an I/O completion port then the outcome may be handled on the initiating thread when the I/O op completes or fails immediately, or I/O completion status is queued. There are cases where it gets handled on the initiating thread but an I/O completion status is also queued. The buffer handled around this is tricky and it seems that since JDK 7, a substituted buffer can been returned to the TL buffer cache more than once. The changes in this PR will ensure can't handle. Future work could be done in this area to simplify several things but not now.

[AlanBateman]

/reviewers 2

[openjdk]

@AlanBateman
The total number of required reviews for this PR (including the jcheck configuration and the last /reviewers command) is now set to 2 (with at least 1 Reviewer, 1 Author).

[AlanBateman]

I provided the change for this PR so I can't really be the reviewer too. So I've set reviewer to 2. @Michael-Mc-Mahon is going to help review too.

[Michael-Mc-Mahon]

Seems pretty straightforward. Looks fine.

[bplb]

/integrate

[openjdk]

Going to push as commit ea73e05.
Since your change was applied there have been 11 commits pushed to the master branch:

• ee0f88c: 8341923: java.util.Locale class specification improvements
• a72cab8: 8345148: Fix for JDK-8337317 is incomplete
• 1a73c76: 8345524: CHECK_FOR_FILES should properly handle directories
• 1d3dcdf: 8345433: (fs) Use stream to load FileTypeDetectors in Files.probeContentType
• 9740c1d: 8333572: Update jpackage.md with info related to WiX v4 and v5 support
• de3a218: 8345502: Remove doIntersectionPrivilege methods
• faa109e: 8340580: Characters in Document diagrams not in the middle
• 6aa7667: 8339535: JVM crashes with -Xshare:dump -XX:+SegmentedCodeCache
• e1695f6: 8345472: Fix incorrect format instruction for floating point max/min patterns
• baa6957: 8345507: Fix build of static launcher
• ... and 1 more: https://git.openjdk.org/jdk/compare/f3b4350e0f14d3b0c551e0d24563788f379111d6...master

Your commit was automatically rebased without conflicts.

[openjdk]

@bplb Pushed as commit ea73e05.

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

[bplb]

Thanks @AlanBateman for the patch and @Michael-Mc-Mahon for the approval!