8326616: tools/javac/patterns/Exhaustiveness.java intermittently Timeout signalled after 480 seconds #20836

lahodaj · 2024-09-03T12:31:18Z

This patch revisits the fix for JDK-8325215 (PR: #17949). The problem in that bug was along the following lines.

Consider this code:

    sealed interface B permits S1, S2 {}
    final class S1 implements B {}
    final class S2 implements B {}
    record R(B b, B s) {}
    private void test(R r) {
        switch (r) {
            case R(B _, S1 _) -> {}
            case R(S1 _, B _) -> {}
            case R(S2 _, S2 _) -> {}
        }
    }

The switch is exhaustive, but before the fix for JDK-8325215, javac considered it not exhaustive. If case R(S1 _, B _) is preplaced with case R(S1 _, S2 _), javac will consider the switch to be exhaustive.

This problem is because to prove that the switch is exhaustive, it is necessary to merge R(S1 _, B _) and R(S2 _, S2 _) into R(B _, S2 _), which then can be merged with R(B _, S1 _) to form R(B _, B_), and then just R _, which means the switch is exhaustive.

But, before JDK-8325215, javac could merge R(S1 _, B _) and R(S2 _, S2 _) into R(B _, S2 _), because it was looking at the two patterns, set the mismatching component to 0, and was looking at the patterns in all other components, if they are precisely the same - but they are not, so javac failed to merge the patterns. Note that it is permitted to replace a type pattern with a more specific type pattern (i.e. replace B _ with S2 _).

The solution in JDK-8325215 was to expand all type patterns in all the patterns in the set into their possible permitted subtypes. I.e. in the above case R(S1 _, B _) was expanded to R(S1 _, B _), R(S1 _, S1 _) and R(S1 _, S2 _). As a consequence, the merging could proceed, and javac would consider the switch to be exhaustive.

The problem with this solution is that it sometimes leads to a very big pattern set, which the javac then processes very slowly/too slowly.

The proposal alternate fix for the original problem proposed herein is to avoid the creation of the very big pattern set, and rather permit javac to merge R(S1 _, B _) and R(S2 _, S2 _), because B is a supertype of (more general pattern than) S2. This is a bit tricky, as the code that searches for merge candidates is using hashes for the patterns, and the hashes speed up the search very significantly (and the use of the hashes is not compatible with the use of the subtyping relation). So, the proposal is to use hashing when possible, and use the subtyping relation when no more patterns can be merged.

I ran the Exhaustiveness test many times with this change, without a timeout so far.

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-8326616: tools/javac/patterns/Exhaustiveness.java intermittently Timeout signalled after 480 seconds (Bug - P4)

Reviewers

Aggelos Biboudis (@biboudis - Reviewer)

Reviewing

Using git

Checkout this PR locally:
$ git fetch https://git.openjdk.org/jdk.git pull/20836/head:pull/20836
$ git checkout pull/20836

Update a local copy of the PR:
$ git checkout pull/20836
$ git pull https://git.openjdk.org/jdk.git pull/20836/head

Using Skara CLI tools

Checkout this PR locally:
$ git pr checkout 20836

View PR using the GUI difftool:
$ git pr show -t 20836

Using diff file

Download this PR as a diff file:
https://git.openjdk.org/jdk/pull/20836.diff

Webrev

Link to Webrev Comment

…out signalled after 480 seconds

bridgekeeper · 2024-09-03T12:32:06Z

👋 Welcome back jlahoda! 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 · 2024-09-03T12:33:30Z

@lahodaj 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:

8326616: tools/javac/patterns/Exhaustiveness.java intermittently Timeout signalled after 480 seconds

Reviewed-by: abimpoudis

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 140 new commits pushed to the master branch:

a35fd38: 8339368: Switch targets are not inflated in CodeModel if no StackMap
7db4d46: 8330159: [C2] Remove or clarify Compile::init_start
9e0ccb8: 8339548: GHA: RISC-V: Use Debian snapshot archive for bootstrap
8fb8cd8: 8339347: keytool -importpass insists prompting the user even if there is no terminal
9e1af8c: 8339285: Test fails with assert(depth < max_critical_stack_depth) failed: can't have more than 10 critical frames
48d7943: 8339342: FieldAllocationCount is mostly unused
e203df4: 8338100: C2: assert(!n_loop->is_member(get_loop(lca))) failed: control must not be back in the loop
98020e4: 8338133: Cleanup direct use of new HtmlTree
b895d7c: 8332423: [PPC64] Remove C1_MacroAssembler::call_c_with_frame_resize
59c4649: 8329959: Update DigestMD5Client.java - fix typo in javadoc string
... and 130 more: https://git.openjdk.org/jdk/compare/e63418ee017def80689c88671e5d124b2d453fda...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 · 2024-09-03T12:33:56Z

@lahodaj 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 · 2024-09-03T12:36:53Z

Webrevs

01: Full - Incremental (78497255)
00: Full (acab89a4)

src/jdk.compiler/share/classes/com/sun/tools/javac/comp/Flow.java

biboudis · 2024-09-03T15:25:55Z

src/jdk.compiler/share/classes/com/sun/tools/javac/comp/Flow.java

+                                                    !(rpOne.nested[i] instanceof BindingPattern bpOne) ||
+                                                    !(rpOther.nested[i] instanceof BindingPattern bpOther) ||
+                                                    !types.isSubtype(types.erasure(bpOne.type), types.erasure(bpOther.type))) {
+                                                    continue NEXT_PATTERN;


~~No tests in the patterns folder executes this block (continue NEXT_PATTERN;). I am sure it works, can you include a small test that leads to this line?~~

biboudis

LGTM

biboudis · 2024-09-04T12:48:32Z

src/jdk.compiler/share/classes/com/sun/tools/javac/comp/Flow.java

@@ -1054,7 +1064,7 @@ private Set<PatternDescription> reduceNestedPatterns(Set<PatternDescription> pat
                             .stream()
                             //error recovery, ignore patterns with incorrect number of nested patterns:
                             .filter(pd -> pd.nested.length == nestedPatternsCount)
-                             .collect(groupingBy(pd -> pd.hashCode(mismatchingCandidateFin)));
+                             .collect(groupingBy(pd -> useHashes ? pd.hashCode(mismatchingCandidateFin) : 0));


groupByHashes can be renamed into groupByOnlyMismatchingCandidateDifference or something else. The ByHashes is not applicable any more in the general case.

I've used groupEquivalenceCandidates - please let me know if that makes sense.

biboudis

Thanks Jan. Nice work!

lahodaj · 2024-09-09T05:33:05Z

/integrate

openjdk · 2024-09-09T05:34:12Z

Going to push as commit a18d9d8.
Since your change was applied there have been 153 commits pushed to the master branch:

79d7613: 8338153: java/awt/Checkbox/CheckboxCheckerScalingTest.java test failed on linux machine
f0e84b7: 8339703: Problem list serviceability/sa/TestJhsdbJstackUpcall.java for generational ZGC
deeb09a: 8339307: jhsdb jstack could not trace FFM upcall frame
fbe2629: 8339635: StringConcatFactory optimization for CompactStrings off
8e580ec: 8338123: Linker crash when building a downcall handle with many arguments in x64
5b72bbf: 8339519: Remove size field from instructions
0df10bb: 8339466: Enumerate shared stubs and define static fields and names via declarations
9ebc2ec: 8339317: Optimize ClassFile writeBuffer
d2b36f0: 8339642: Reduce overheads in InvokerBytecodeGenerator
cb00333: 8339640: Reduce construction overheads in StringConcatFactory$InlineHiddenClassStrategy
... and 143 more: https://git.openjdk.org/jdk/compare/e63418ee017def80689c88671e5d124b2d453fda...master

Your commit was automatically rebased without conflicts.

openjdk · 2024-09-09T05:34:24Z

@lahodaj Pushed as commit a18d9d8.

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

lahodaj added 3 commits August 30, 2024 13:37

8326616: tools/javac/patterns/Exhaustiveness.java intermittently Time…

fb2a774

…out signalled after 480 seconds

Cleanup.

a2ab4ba

Cleanup.

acab89a

openjdk bot added the rfr Pull request is ready for review label Sep 3, 2024

openjdk bot added the compiler compiler-dev@openjdk.org label Sep 3, 2024

biboudis reviewed Sep 3, 2024

View reviewed changes

src/jdk.compiler/share/classes/com/sun/tools/javac/comp/Flow.java Outdated Show resolved Hide resolved

biboudis reviewed Sep 3, 2024

View reviewed changes

biboudis approved these changes Sep 4, 2024

View reviewed changes

openjdk bot added the ready Pull request is ready to be integrated label Sep 4, 2024

Reflecting review feedback - adjusting javadoc and name.

7849725

openjdk bot removed the ready Pull request is ready to be integrated label Sep 6, 2024

biboudis approved these changes Sep 6, 2024

View reviewed changes

openjdk bot added the ready Pull request is ready to be integrated label Sep 6, 2024

openjdk bot added the integrated Pull request has been integrated label Sep 9, 2024

openjdk bot closed this Sep 9, 2024

openjdk bot removed ready Pull request is ready to be integrated rfr Pull request is ready for review labels Sep 9, 2024

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

8326616: tools/javac/patterns/Exhaustiveness.java intermittently Timeout signalled after 480 seconds #20836

8326616: tools/javac/patterns/Exhaustiveness.java intermittently Timeout signalled after 480 seconds #20836

Uh oh!

lahodaj commented Sep 3, 2024 •

edited by openjdk bot

Loading

Uh oh!

bridgekeeper bot commented Sep 3, 2024

Uh oh!

openjdk bot commented Sep 3, 2024 •

edited

Loading

Uh oh!

openjdk bot commented Sep 3, 2024

Uh oh!

mlbridge bot commented Sep 3, 2024 •

edited

Loading

Uh oh!

Uh oh!

biboudis Sep 3, 2024 •

edited

Loading

Uh oh!

biboudis left a comment

Uh oh!

biboudis Sep 4, 2024

Uh oh!

lahodaj Sep 6, 2024

Uh oh!

biboudis Sep 6, 2024

Uh oh!

biboudis left a comment

Uh oh!

lahodaj commented Sep 9, 2024

Uh oh!

openjdk bot commented Sep 9, 2024

Uh oh!

openjdk bot commented Sep 9, 2024

Uh oh!

Uh oh!

8326616: tools/javac/patterns/Exhaustiveness.java intermittently Timeout signalled after 480 seconds #20836

8326616: tools/javac/patterns/Exhaustiveness.java intermittently Timeout signalled after 480 seconds #20836

Uh oh!

Conversation

lahodaj commented Sep 3, 2024 • edited by openjdk bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Progress

Issue

Reviewers

Reviewing

Webrev

Uh oh!

bridgekeeper bot commented Sep 3, 2024

Uh oh!

openjdk bot commented Sep 3, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

openjdk bot commented Sep 3, 2024

Uh oh!

mlbridge bot commented Sep 3, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Webrevs

Uh oh!

Uh oh!

biboudis Sep 3, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

biboudis left a comment

Choose a reason for hiding this comment

Uh oh!

biboudis Sep 4, 2024

Choose a reason for hiding this comment

Uh oh!

lahodaj Sep 6, 2024

Choose a reason for hiding this comment

Uh oh!

biboudis Sep 6, 2024

Choose a reason for hiding this comment

Uh oh!

biboudis left a comment

Choose a reason for hiding this comment

Uh oh!

lahodaj commented Sep 9, 2024

Uh oh!

openjdk bot commented Sep 9, 2024

Uh oh!

openjdk bot commented Sep 9, 2024

Uh oh!

Uh oh!

lahodaj commented Sep 3, 2024 •

edited by openjdk bot

Loading

openjdk bot commented Sep 3, 2024 •

edited

Loading

mlbridge bot commented Sep 3, 2024 •

edited

Loading

biboudis Sep 3, 2024 •

edited

Loading