8331081: 'internal proprietary API' diagnostics if --system is configured to an earlier JDK version

[cushon]

This change fixes a bug preventing javac from emitting 'compiler.warn.sun.proprietary' diagnostics if --system is set to a non-default value. The diagnostics are currently emitted for values of --release, and for the default value of --system.

---

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-8331081: 'internal proprietary API' diagnostics if --system is configured to an earlier JDK version (Bug - P4)

Reviewers

• Jan Lahoda (@lahodaj - Reviewer)

Reviewing

Using git

Checkout this PR locally:
$ git fetch https://git.openjdk.org/jdk.git pull/19069/head:pull/19069
$ git checkout pull/19069

Update a local copy of the PR:
$ git checkout pull/19069
$ git pull https://git.openjdk.org/jdk.git pull/19069/head

Using Skara CLI tools

Checkout this PR locally:
$ git pr checkout 19069

View PR using the GUI difftool:
$ git pr show -t 19069

Using diff file

Download this PR as a diff file:
https://git.openjdk.org/jdk/pull/19069.diff

Webrev

Link to Webrev Comment

[bridgekeeper]

👋 Welcome back cushon! 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]

@cushon 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:

8331081: 'internal proprietary API' diagnostics if --system is configured to an earlier JDK version

Reviewed-by: jlahoda

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 2 new commits pushed to the master branch:

• a0c5714: 8332071: Convert package.html files in java.management.rmi to package-info.java
• afed7d0: 8329538: Accelerate P256 on x86_64 using Montgomery intrinsic

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]

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

• compiler

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

• 02: Full - Incremental (b64951ff)
• 01: Full - Incremental (fef11730)
• 00: Full (4cb9915e)

[lahodaj]

Traditionally, I don't think javac was ever (supposed to) produce these warnings when the system classes where overridden (using -bootclasspath in JDK <9, and --system in JDK >= 9).

I believe the idea was that when someone provides their own the system classes, we shall not produce a warning based on some hardcoded knowledge, which may or may not be relevant for the externally provided system classes. That might be overly cautious, especially given that since JDK 9, it is only about jdk.unsupported, and it is harder to imagine a usecase where the warning would not be reasonable for that module.

On the code level, I would suggest to re-organize the code so that user packages would not accumulate in supplementaryFlags - that seems unnecessary. If the module owning c is neither noModule nor jdk_unsupported, there's no point in putting anything in the map the computation, I think. (Not too terrible, but some user programs use a lot of packages.)

Also, I wonder why not simply use --release 11. The set of usecases for which --system needs to be used instead of --release, and for which the warning is important seems quite limited to me.

[cushon]

Traditionally, I don't think javac was ever (supposed to) produce these warnings when the system classes where overridden (using -bootclasspath in JDK <9, and --system in JDK >= 9).

I believe the idea was that when someone provides their own the system classes, we shall not produce a warning based on some hardcoded knowledge, which may or may not be relevant for the externally provided system classes. That might be overly cautious, especially given that since JDK 9, it is only about jdk.unsupported, and it is harder to imagine a usecase where the warning would not be reasonable for that module.

Understood, thanks. The motivation here is that I'm stuck using --system for some use-cases (more on that below), and all else being equal I'd like to discourage new uses of jdk.unsupported APIs, so supporting this diagnostic would be helpful for that.

On the code level, I would suggest to re-organize the code so that user packages would not accumulate in supplementaryFlags - that seems unnecessary. If the module owning c is neither noModule nor jdk_unsupported, there's no point in putting anything in the map the computation, I think. (Not too terrible, but some user programs use a lot of packages.)

Thanks, I can clean that up.

Also, I wonder why not simply use --release 11. The set of usecases for which --system needs to be used instead of --release, and for which the warning is important seems quite limited to me.

--release can't be used together with --add-exports=, and can't be used to compile code that uses internal APIs for Java 8 and earlier (JDK-8206937). That use-case may be limited and I understand that at least the first part is deliberate, but that's use-case for --system I have.

[cushon]

On the code level, I would suggest to re-organize the code so that user packages would not accumulate in supplementaryFlags - that seems unnecessary. If the module owning c is neither noModule nor jdk_unsupported, there's no point in putting anything in the map the computation, I think. (Not too terrible, but some user programs use a lot of packages.)

Thanks, I can clean that up.

Done

[lahodaj]

Looks good to me.

[cushon]

/integrate

[openjdk]

Going to push as commit 2170e99.
Since your change was applied there have been 2 commits pushed to the master branch:

• a0c5714: 8332071: Convert package.html files in java.management.rmi to package-info.java
• afed7d0: 8329538: Accelerate P256 on x86_64 using Montgomery intrinsic

Your commit was automatically rebased without conflicts.

[openjdk]

@cushon Pushed as commit 2170e99.

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