8332507: compilation result depends on compilation order #19400

vicente-romero-oracle · 2024-05-24T18:59:56Z

Wildcards and separate compilation haven't been friends since like: forever. Javac has had a long standing issue of not being able to restore wildcard bounds when they are read from the classfile. Part of this is due to a restriction of not reading another class file while a given class file is being read. Without breaking this restriction we can "wait" until a class symbol is completed from a class file to have access to richer information and being able to set the right bounds to wildcards. The fix seems simple enough. Basically the proposal is to register listeners that will be invoked once a given class symbol is completely read from a class file. This tactic has been used in type inference and other areas of the compiler where another layer of indirection has been a must to fix similar problems.

It will be a great pleasure if this proposal can be the solution to this long standing issue,

TIA

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

Warning

⚠️ Found leading lowercase letter in issue title for 8332507: compilation result depends on compilation order

Issue

JDK-8332507: compilation result depends on compilation order (Bug - P4)(⚠️ The fixVersion in this issue is [24] but the fixVersion in .jcheck/conf is 23, a new backport will be created when this pr is integrated.)

Reviewers

Maurizio Cimadamore (@mcimadamore - Reviewer) ⚠️ Review applies to ba4d99fe

Reviewing

Using git

Checkout this PR locally:
$ git fetch https://git.openjdk.org/jdk.git pull/19400/head:pull/19400
$ git checkout pull/19400

Update a local copy of the PR:
$ git checkout pull/19400
$ git pull https://git.openjdk.org/jdk.git pull/19400/head

Using Skara CLI tools

Checkout this PR locally:
$ git pr checkout 19400

View PR using the GUI difftool:
$ git pr show -t 19400

Using diff file

Download this PR as a diff file:
https://git.openjdk.org/jdk/pull/19400.diff

Webrev

Link to Webrev Comment

bridgekeeper · 2024-05-24T19:00:28Z

👋 Welcome back vromero! 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 · 2024-05-24T19:01:05Z

@vicente-romero-oracle 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:

8332507: compilation result depends on compilation order

Reviewed-by: mcimadamore

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 164 new commits pushed to the master branch:

5abc029: 8331877: JFR: Remove JIInliner framework
d9e7b7e: 8210471: GZIPInputStream constructor could leak an un-end()ed Inflater
1e04ee6: 8331579: Reference to primitive type fails without error or warning
32ee252: 8333169: javac NullPointerException record.type
e930bc1: 8329537: Nested and enclosing classes should be linked separately in breadcrumb navigation
79a78f0: 8333129: Move ShrinkHeapInSteps flag to Serial GC
2f2dc22: 8330981: ZGC: Should not dedup strings in the finalizer graph
d481215: 8333005: Deadlock when setting or updating the inline cache
7ab74c5: 8333200: Test containers/docker/TestPids.java fails Limit value -1 is not accepted as unlimited
a7864af: 8182774: Verify code in javap
... and 154 more: https://git.openjdk.org/jdk/compare/9bfae8891e6efa58c557bd6dac61de111a16f71e...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.

openjdk · 2024-05-24T19:01:32Z

@vicente-romero-oracle 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 · 2024-05-24T19:04:12Z

Webrevs

mcimadamore · 2024-05-28T13:50:41Z

src/jdk.compiler/share/classes/com/sun/tools/javac/jvm/ClassReader.java

+                }
+                if (actuals.length() != (formals == null ? 0 : formals.length())) {
+                    final List<Type> actualsCp = actuals;
+                    addSymbolReadListener(t, () -> {


I wonder if we could use the completion mechanism for this - class reader should already sign itself up as a completer for the class symbols it generates?

I wonder if we could use the completion mechanism for this - class reader should already sign itself up as a completer for the class symbols it generates?

pretty good suggestion, thanks! I have uploaded another iteration that leverages the completer mechanism

mcimadamore · 2024-05-31T11:53:20Z

src/jdk.compiler/share/classes/com/sun/tools/javac/jvm/ClassReader.java

+                    if (actuals.isEmpty())
+                        actuals = formals;
+                    else
+                        updateBounds(actuals, formals);


Is it important we do this here - or can we always do this in the completer (to simplify the code some more)

thanks for the suggestion, will give it a try

mcimadamore

Looks good - I left an optional suggestion

mcimadamore · 2024-05-31T14:29:52Z

src/jdk.compiler/share/classes/com/sun/tools/javac/jvm/ClassReader.java

+
+                        @Override
+                        public List<Type> getTypeArguments() {
+                            List<Type> formalsCp = ((ClassType)t.type.tsym.type).typarams_field;


This works. You might use another boolean flag up top to say whether you actually need to adjust the type arguments (which is only needed the first time) ?

sure sorry I forgot about that, done

vicente-romero-oracle · 2024-05-31T17:34:20Z

/integrate

openjdk · 2024-05-31T17:34:53Z

Going to push as commit e650bdf.
Since your change was applied there have been 167 commits pushed to the master branch:

e4fbb15: 8320215: HeapDumper can use DumpWriter buffer during merge
681137c: 8333131: Source launcher should work with service loader SPI
914423e: 8332899: RISC-V: add comment and make the code more readable (if possible) in MacroAssembler::movptr
5abc029: 8331877: JFR: Remove JIInliner framework
d9e7b7e: 8210471: GZIPInputStream constructor could leak an un-end()ed Inflater
1e04ee6: 8331579: Reference to primitive type fails without error or warning
32ee252: 8333169: javac NullPointerException record.type
e930bc1: 8329537: Nested and enclosing classes should be linked separately in breadcrumb navigation
79a78f0: 8333129: Move ShrinkHeapInSteps flag to Serial GC
2f2dc22: 8330981: ZGC: Should not dedup strings in the finalizer graph
... and 157 more: https://git.openjdk.org/jdk/compare/9bfae8891e6efa58c557bd6dac61de111a16f71e...master

Your commit was automatically rebased without conflicts.

openjdk · 2024-05-31T17:34:59Z

@vicente-romero-oracle Pushed as commit e650bdf.

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

8332507: compilation result depends on compilation order

d428dc6

openjdk bot added the rfr Pull request is ready for review label May 24, 2024

openjdk bot added the compiler compiler-dev@openjdk.org label May 24, 2024

mcimadamore reviewed May 28, 2024

View reviewed changes

vicente-romero-oracle added 2 commits May 28, 2024 11:19

addressing review comments, levarage the existing complete hooks

81ebe29

removing unused imports

ba4d99f

mcimadamore reviewed May 31, 2024

View reviewed changes

mcimadamore approved these changes May 31, 2024

View reviewed changes

openjdk bot added the ready Pull request is ready to be integrated label May 31, 2024

vicente-romero-oracle added 2 commits May 31, 2024 09:54

review comments

6014d8c

review comments

b22d741

mcimadamore reviewed May 31, 2024

View reviewed changes

review comments

d88e855

openjdk bot added the integrated Pull request has been integrated label May 31, 2024

openjdk bot closed this May 31, 2024

openjdk bot removed ready Pull request is ready to be integrated rfr Pull request is ready for review labels May 31, 2024

vicente-romero-oracle deleted the JDK-8332507 branch May 31, 2024 17:35

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

8332507: compilation result depends on compilation order #19400

8332507: compilation result depends on compilation order #19400

vicente-romero-oracle commented May 24, 2024 •

edited by openjdk bot

Loading

bridgekeeper bot commented May 24, 2024

openjdk bot commented May 24, 2024 •

edited

Loading

openjdk bot commented May 24, 2024

mlbridge bot commented May 24, 2024 •

edited

Loading

mcimadamore May 28, 2024

vicente-romero-oracle May 28, 2024

mcimadamore May 31, 2024

vicente-romero-oracle May 31, 2024

mcimadamore left a comment

mcimadamore May 31, 2024

vicente-romero-oracle May 31, 2024 •

edited

Loading

vicente-romero-oracle commented May 31, 2024

openjdk bot commented May 31, 2024

openjdk bot commented May 31, 2024

8332507: compilation result depends on compilation order #19400

8332507: compilation result depends on compilation order #19400

Conversation

vicente-romero-oracle commented May 24, 2024 • edited by openjdk bot Loading

Progress

Warning

Issue

Reviewers

Reviewing

Webrev

bridgekeeper bot commented May 24, 2024

openjdk bot commented May 24, 2024 • edited Loading

openjdk bot commented May 24, 2024

mlbridge bot commented May 24, 2024 • edited Loading

Webrevs

mcimadamore May 28, 2024

Choose a reason for hiding this comment

vicente-romero-oracle May 28, 2024

Choose a reason for hiding this comment

mcimadamore May 31, 2024

Choose a reason for hiding this comment

vicente-romero-oracle May 31, 2024

Choose a reason for hiding this comment

mcimadamore left a comment

Choose a reason for hiding this comment

mcimadamore May 31, 2024

Choose a reason for hiding this comment

vicente-romero-oracle May 31, 2024 • edited Loading

Choose a reason for hiding this comment

vicente-romero-oracle commented May 31, 2024

openjdk bot commented May 31, 2024

openjdk bot commented May 31, 2024

vicente-romero-oracle commented May 24, 2024 •

edited by openjdk bot

Loading

openjdk bot commented May 24, 2024 •

edited

Loading

mlbridge bot commented May 24, 2024 •

edited

Loading

vicente-romero-oracle May 31, 2024 •

edited

Loading