8347997: assert(false) failed: EA: missing memory path #23284

vnkozlov · 2025-01-24T00:07:43Z

C2's Escape Analysis does not recognize pattern where one input of memory Phi node is MergeMem node and an other is RAW store. This pattern is created by Continuation pinning intrinsic. As result EA complains about strange memory graph.

I suggest to add second MergeMem between Store and Phi nodes by calling reset_memory(). EA recognize such patter and removes allocations.

I checked generated assembler pinning code and it is the same as before. The only difference in the test is eliminated allocations.

I moved Uncommon code up to avoid resetting memory - it is already done at the beginning of this intrinsic code.

We should not use uncommon_trap_exact() for Deoptimization::Action_none - It is used for other actions to prevent changing them to Action_none.

Tested tier1-5, hs-xcomp, hs-comp-stress
Added new regression test based on reproducer from bug report.

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-8347997: assert(false) failed: EA: missing memory path (Bug - P3)

Reviewers

Tobias Hartmann (@TobiHartmann - Reviewer)
Christian Hagedorn (@chhagedorn - Reviewer)

Reviewing

Using git

Checkout this PR locally:
$ git fetch https://git.openjdk.org/jdk.git pull/23284/head:pull/23284
$ git checkout pull/23284

Update a local copy of the PR:
$ git checkout pull/23284
$ git pull https://git.openjdk.org/jdk.git pull/23284/head

Using Skara CLI tools

Checkout this PR locally:
$ git pr checkout 23284

View PR using the GUI difftool:
$ git pr show -t 23284

Using diff file

Download this PR as a diff file:
https://git.openjdk.org/jdk/pull/23284.diff

Using Webrev

Link to Webrev Comment

bridgekeeper · 2025-01-24T00:08:10Z

👋 Welcome back kvn! 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 · 2025-01-24T00:08:24Z

@vnkozlov 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:

8347997: assert(false) failed: EA: missing memory path

Reviewed-by: thartmann, chagedorn

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 63 new commits pushed to the master branch:

55c3e78: 8348631: Crash in PredictedCallGenerator::generate after JDK-8347006
98a93e1: 8348800: Many serviceability/sa tests failing after JDK-8348239
5e81fa6: 8348892: Properly fix compilation error for zip_util.c on Windows
3a564ed: 8347955: TimeZone methods to stream the available timezone IDs
1efae9a: 8348888: tier1 closed build failure on Windows after JDK-8348348
c018a60: 8344637: Fix Page8 of manual test java/awt/print/PrinterJob/PrintTextTest.java on Linux and Windows
c3c3888: 8336760: [JVMCI] -XX:+PrintCompilation should also print "hosted" JVMCI compilations
9f4d3de: 8347718: Unexpected NullPointerException in C2 compiled code due to ReduceAllocationMerges
a224f12: 8348205: Improve cutover time selection when building available currencies set
8103256: 8348348: Remove unnecessary #ifdef STATIC_BUILD around DEF_STATIC_JNI_OnLoad from zip_util.c
... and 53 more: https://git.openjdk.org/jdk/compare/7f16a0875ced8669b9d2131c67496a66e74ea36f...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 · 2025-01-24T00:09:07Z

@vnkozlov The following labels will be automatically applied to this pull request:

graal
hotspot-compiler

When this pull request is ready to be reviewed, an "RFR" email will be sent to the corresponding mailing lists. If you would like to change these labels, use the /label pull request command.

mlbridge · 2025-01-24T00:14:57Z

Webrevs

AlanBateman · 2025-01-24T07:39:48Z

If you want, the reproducer can be simplified to just invoke pin/unpin, avoids needing the method handle code to invoke them reflectively, e.g.

import jdk.internal.vm.Continuation;

    static class FailsEA {
        final Object o;

        public FailsEA() throws Throwable {
            o = new Object();
            Continuation.pin();
            Continuation.unpin();
        }
    }

    static class Crashes {
        final Object o;

        public Crashes() throws Throwable {
            Continuation.pin();
            Continuation.unpin();
            o = new Object();
        }
    }

If you add @modules java.base/jdk.internal.vm to the test description then jtreg will compile and run the test with this package exported, no need be open the package with the @run tag.

vnkozlov · 2025-01-24T17:57:46Z

Thank you @AlanBateman for suggestions about the test. I implemented them and verified that the test still catches the failure without fix.

AlanBateman · 2025-01-24T18:28:36Z

Thank you @AlanBateman for suggestions about the test. I implemented them and verified that the test still catches the failure without fix.

Good, much simpler.

TobiHartmann

Looks good to me otherwise.

The only difference in the test is eliminated allocations.

Sounds like a good opportunity for an IR framework test but could be done as follow-up (starter) RFE.

test/hotspot/jtreg/compiler/intrinsics/TestContinuationPinningAndEA.java

TobiHartmann · 2025-01-28T06:25:16Z

test/hotspot/jtreg/compiler/intrinsics/TestContinuationPinningAndEA.java

+
+public class TestContinuationPinningAndEA {
+
+  static class FailsEA {


Should use 4-whitespace indentation since it's Java code.

chhagedorn

Otherwise, the fix looks good to me, too!

chhagedorn · 2025-01-28T06:49:35Z

src/hotspot/share/opto/library_call.cpp

@@ -3772,6 +3772,18 @@ bool LibraryCallKit::inline_native_Continuation_pinning(bool unpin) {
  Node* test_pin_count_over_underflow = _gvn.transform(new BoolNode(pin_count_cmp, BoolTest::eq));
  IfNode* iff_pin_count_over_underflow = create_and_map_if(control(), test_pin_count_over_underflow, PROB_MIN, COUNT_UNKNOWN);

+  // True branch, pin count over/underflow.


Maybe you can add a comment here about why we do the trap first (as described in the PR description).

…AndEA.java Co-authored-by: Tobias Hartmann <tobias.hartmann@oracle.com>

vnkozlov · 2025-01-28T18:03:39Z

Sounds like a good opportunity for an IR framework test but could be done as follow-up (starter) RFE.

I created JDK-8348887

vnkozlov · 2025-01-28T18:04:37Z

Thank you @TobiHartmann and @chhagedorn for reviews. I addressed your comments.

chhagedorn

Update looks good, thanks!

vnkozlov · 2025-01-29T17:22:37Z

/integrate

vnkozlov · 2025-01-29T17:23:08Z

Thank you, @TobiHartmann and @chhagedorn for reviews

openjdk · 2025-01-29T17:24:17Z

Going to push as commit 6b581d2.
Since your change was applied there have been 75 commits pushed to the master branch:

4662363: 8348687: [BACKOUT] C2: Non-fluid StringBuilder pattern bails out in OptoStringConcat
d266ca9: 8348752: Enable -XX:+AOTClassLinking by default when -XX:AOTMode is specified
cbc89a7: 8348898: Remove unused OctalDigits to clean up code
96fefed: 8319850: PrintInlining should print which methods are late inlines
51cce6e: 8318577: Windows Look-and-Feel JProgressBarUI does not render correctly on 2x UI scale
6bfae3a: 8333386: TestAbortOnVMOperationTimeout test fails for client VM
f98d9a3: 8348870: Eliminate array bound checks in DecimalDigits
fe6d9ab: 8348582: Set -fstack-protector when building with clang
fff97d6: 8343938: TestStressBailout triggers "Should not be locked when freed" assert
69ec453: 8347985: Deprecate java.management Permission classes for removal
... and 65 more: https://git.openjdk.org/jdk/compare/7f16a0875ced8669b9d2131c67496a66e74ea36f...master

Your commit was automatically rebased without conflicts.

openjdk · 2025-01-29T17:24:32Z

@vnkozlov Pushed as commit 6b581d2.

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

8347997: assert(false) failed: EA: missing memory path

679d49c

openjdk bot added graal graal-dev@openjdk.org hotspot-compiler hotspot-compiler-dev@openjdk.org labels Jan 24, 2025

Remove trailing space

dd1b1b9

openjdk bot added the rfr Pull request is ready for review label Jan 24, 2025

Update test

fad42ec

TobiHartmann approved these changes Jan 28, 2025

View reviewed changes

openjdk bot added the ready Pull request is ready to be integrated label Jan 28, 2025

chhagedorn approved these changes Jan 28, 2025

View reviewed changes

Update test/hotspot/jtreg/compiler/intrinsics/TestContinuationPinning…

fee3b72

…AndEA.java Co-authored-by: Tobias Hartmann <tobias.hartmann@oracle.com>

openjdk bot removed the ready Pull request is ready to be integrated label Jan 28, 2025

Address comments

e502568

TobiHartmann approved these changes Jan 28, 2025

View reviewed changes

openjdk bot added the ready Pull request is ready to be integrated label Jan 28, 2025

chhagedorn approved these changes Jan 29, 2025

View reviewed changes

openjdk bot added the integrated Pull request has been integrated label Jan 29, 2025

openjdk bot closed this Jan 29, 2025

openjdk bot removed the ready Pull request is ready to be integrated label Jan 29, 2025

openjdk bot removed the rfr Pull request is ready for review label Jan 29, 2025

graalvmbot mentioned this pull request Feb 12, 2025

[GR-62075] Update SyncPort info. oracle/graal#10675

Merged

vnkozlov deleted the 8347997 branch February 13, 2025 16:16


		public class TestContinuationPinningAndEA {

		static class FailsEA {

8347997: assert(false) failed: EA: missing memory path #23284

8347997: assert(false) failed: EA: missing memory path #23284

Uh oh!

Conversation

vnkozlov commented Jan 24, 2025 • edited by openjdk bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Progress

Issue

Reviewers

Reviewing

Uh oh!

bridgekeeper bot commented Jan 24, 2025

Uh oh!

openjdk bot commented Jan 24, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

openjdk bot commented Jan 24, 2025

Uh oh!

mlbridge bot commented Jan 24, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Webrevs

Uh oh!

AlanBateman commented Jan 24, 2025

Uh oh!

vnkozlov commented Jan 24, 2025

Uh oh!

AlanBateman commented Jan 24, 2025

Uh oh!

TobiHartmann left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

TobiHartmann Jan 28, 2025

Choose a reason for hiding this comment

Uh oh!

chhagedorn left a comment • 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!

chhagedorn Jan 28, 2025

Choose a reason for hiding this comment

Uh oh!

vnkozlov commented Jan 28, 2025

Uh oh!

vnkozlov commented Jan 28, 2025

Uh oh!

chhagedorn left a comment

Choose a reason for hiding this comment

Uh oh!

vnkozlov commented Jan 29, 2025

Uh oh!

vnkozlov commented Jan 29, 2025

Uh oh!

openjdk bot commented Jan 29, 2025

Uh oh!

openjdk bot commented Jan 29, 2025

Uh oh!

Uh oh!

vnkozlov commented Jan 24, 2025 •

edited by openjdk bot

Loading

openjdk bot commented Jan 24, 2025 •

edited

Loading

mlbridge bot commented Jan 24, 2025 •

edited

Loading

chhagedorn left a comment •

edited

Loading