8310130: C2: assert(false) failed: scalar_input is neither phi nor a matchin reduction #14494

eme64 · 2023-06-15T14:42:14Z

Removed a spurious assert before optimization bailout.

I assumed that the "scalar input" of a Reduction node must always be either a Phi or another Reduction node. But that is incorrect, partial vectorization can lead to a reduction node chain where we have vector reductions and scalar reductions at the same time. In those cases, we cannot move the UnorderedReductions out of the loop, so a optimization bailout is appropriate.

I assessed the other asserts in PhaseIdealLoop::move_unordered_reduction_out_of_loop, and I think they are all justified. However, one assert would have lead to a continue in production, which would not break out of the nested loop correctly. I changed it to a return, so that would be a bailout from the optimization. This assert should not be triggered because in SuperWord::mark_reductions we forbid that a reduction node has any uses inside the loop except for the successor node in the reduction chain.

I have one regression test delivered by the fuzzer, and one that I constructed myself after understanding the issue.
Testing up to tier6 and stress testing.

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-8310130: C2: assert(false) failed: scalar_input is neither phi nor a matchin reduction (Bug - P3)

Reviewers

Vladimir Kozlov (@vnkozlov - Reviewer)
Christian Hagedorn (@chhagedorn - Reviewer) ⚠️ Review applies to 53e8913e

Reviewing

Using git

Checkout this PR locally:
$ git fetch https://git.openjdk.org/jdk.git pull/14494/head:pull/14494
$ git checkout pull/14494

Update a local copy of the PR:
$ git checkout pull/14494
$ git pull https://git.openjdk.org/jdk.git pull/14494/head

Using Skara CLI tools

Checkout this PR locally:
$ git pr checkout 14494

View PR using the GUI difftool:
$ git pr show -t 14494

Using diff file

Download this PR as a diff file:
https://git.openjdk.org/jdk/pull/14494.diff

Webrev

Link to Webrev Comment

…matchin reduction

bridgekeeper · 2023-06-15T14:43:23Z

👋 Welcome back epeter! 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 · 2023-06-15T14:44:26Z

@eme64 The following label will be automatically applied to this pull request:

hotspot-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 · 2023-06-19T08:53:15Z

Webrevs

vnkozlov

Looks fine.

openjdk · 2023-06-20T17:17:04Z

@eme64 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:

8310130: C2: assert(false) failed: scalar_input is neither phi nor a matchin reduction

Reviewed-by: kvn, 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 165 new commits pushed to the master branch:

48e61c1: 8310728: Enable Zc:inline flag in Visual Studio build
56a73a6: 8309591: Socket.setOption(TCP_QUICKACK) uses wrong level
afdaa2a: 8309109: AArch64: [TESTBUG] compiler/intrinsics/sha/cli/TestUseSHA3IntrinsicsOptionOnSupportedCPU.java fails on Neoverse N2 and V1
7f09435: 8310982: jdk/internal/util/ArchTest.java fails after JDK-8308452 failed with Method isARM()
f07e396: 8310914: Remove 2 malformed java/foreign ProblemList entries
a97f98f: 8310585: GetThreadState spec mentions undefined JVMTI_THREAD_STATE_MONITOR_WAITING
2cf5c2f: 8310238: [test bug] javax/swing/JTableHeader/6889007/bug6889007.java fails
269852b: 8294316: SA core file support is broken on macosx-x64 starting with macOS 12.x
4bce38c: 8310830: typo in the parameter name in @throws of ClassDesc::ofDescriptor
315242b: 8310975: java.util.FormatItemModifier should not be protected
... and 155 more: https://git.openjdk.org/jdk/compare/83d92672d4c2637fc37ddd873533c85a9b083904...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.

chhagedorn

Looks good!

chhagedorn · 2023-06-21T07:04:40Z

test/hotspot/jtreg/compiler/loopopts/superword/TestUnorderedReductionPartialVectorization.java

+        TestFramework.runWithFlags("-Xbatch", "-XX:-TieredCompilation",
+                                   "-XX:CompileCommand=compileonly,compiler.loopopts.superword.TestUnorderedReductionPartialVectorization::test*");


It should probably also trigger without the flags by just specifying TestFramework.run() as test1() is not using other methods and the IR framework will implicitly use -Xbatch and wait for the compilation of test1() to be finished.

@chhagedorn I was able to remove them, thanks for the catch!

eme64 · 2023-06-26T08:36:04Z

Just saw a x86 (32bit) IR rule failure for compiler.loopopts.superword.TestUnorderedReductionPartialVectorization.test1.
Sees the reason is that OR_REDUCTION_V only works with long if we have * @requires vm.bits == "64". I'll add that to the test.

vnkozlov

Update is good

eme64 · 2023-06-28T05:50:31Z

Thanks @vnkozlov @chhagedorn for the reviews!
/integrate

openjdk · 2023-06-28T05:52:31Z

Going to push as commit 526dba1.
Since your change was applied there have been 165 commits pushed to the master branch:

48e61c1: 8310728: Enable Zc:inline flag in Visual Studio build
56a73a6: 8309591: Socket.setOption(TCP_QUICKACK) uses wrong level
afdaa2a: 8309109: AArch64: [TESTBUG] compiler/intrinsics/sha/cli/TestUseSHA3IntrinsicsOptionOnSupportedCPU.java fails on Neoverse N2 and V1
7f09435: 8310982: jdk/internal/util/ArchTest.java fails after JDK-8308452 failed with Method isARM()
f07e396: 8310914: Remove 2 malformed java/foreign ProblemList entries
a97f98f: 8310585: GetThreadState spec mentions undefined JVMTI_THREAD_STATE_MONITOR_WAITING
2cf5c2f: 8310238: [test bug] javax/swing/JTableHeader/6889007/bug6889007.java fails
269852b: 8294316: SA core file support is broken on macosx-x64 starting with macOS 12.x
4bce38c: 8310830: typo in the parameter name in @throws of ClassDesc::ofDescriptor
315242b: 8310975: java.util.FormatItemModifier should not be protected
... and 155 more: https://git.openjdk.org/jdk/compare/83d92672d4c2637fc37ddd873533c85a9b083904...master

Your commit was automatically rebased without conflicts.

openjdk · 2023-06-28T05:52:42Z

@eme64 Pushed as commit 526dba1.

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

8310130: C2: assert(false) failed: scalar_input is neither phi nor a …

6a498d2

…matchin reduction

openjdk bot added the hotspot-compiler hotspot-compiler-dev@openjdk.org label Jun 15, 2023

eme64 added 5 commits June 16, 2023 11:02

improved tests

ebfed03

fix bailout of another assert

197e096

fix whitespace

692a870

fixed IR rule of new test

f3a128c

make test restrictions tighter

bd53d95

eme64 marked this pull request as ready for review June 19, 2023 08:47

openjdk bot added the rfr Pull request is ready for review label Jun 19, 2023

vnkozlov approved these changes Jun 20, 2023

View reviewed changes

openjdk bot added the ready Pull request is ready to be integrated label Jun 20, 2023

chhagedorn approved these changes Jun 21, 2023

View reviewed changes

removed unnecessary flags from test

53e8913

chhagedorn approved these changes Jun 23, 2023

View reviewed changes

require 64 bit for test with OR_REDUCTION_V

056f00a

vnkozlov approved these changes Jun 27, 2023

View reviewed changes

openjdk bot added the integrated Pull request has been integrated label Jun 28, 2023

openjdk bot closed this Jun 28, 2023

openjdk bot removed ready Pull request is ready to be integrated rfr Pull request is ready for review labels Jun 28, 2023

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

8310130: C2: assert(false) failed: scalar_input is neither phi nor a matchin reduction #14494

8310130: C2: assert(false) failed: scalar_input is neither phi nor a matchin reduction #14494

eme64 commented Jun 15, 2023 •

edited by openjdk bot

Loading

bridgekeeper bot commented Jun 15, 2023

openjdk bot commented Jun 15, 2023

mlbridge bot commented Jun 19, 2023 •

edited

Loading

vnkozlov left a comment

openjdk bot commented Jun 20, 2023 •

edited

Loading

chhagedorn left a comment

chhagedorn Jun 21, 2023

eme64 Jun 22, 2023

chhagedorn Jun 23, 2023

eme64 commented Jun 26, 2023 •

edited

Loading

vnkozlov left a comment

eme64 commented Jun 28, 2023

openjdk bot commented Jun 28, 2023

openjdk bot commented Jun 28, 2023

		TestFramework.runWithFlags("-Xbatch", "-XX:-TieredCompilation",
		"-XX:CompileCommand=compileonly,compiler.loopopts.superword.TestUnorderedReductionPartialVectorization::test*");

8310130: C2: assert(false) failed: scalar_input is neither phi nor a matchin reduction #14494

8310130: C2: assert(false) failed: scalar_input is neither phi nor a matchin reduction #14494

Conversation

eme64 commented Jun 15, 2023 • edited by openjdk bot Loading

Progress

Issue

Reviewers

Reviewing

Webrev

bridgekeeper bot commented Jun 15, 2023

openjdk bot commented Jun 15, 2023

mlbridge bot commented Jun 19, 2023 • edited Loading

Webrevs

vnkozlov left a comment

Choose a reason for hiding this comment

openjdk bot commented Jun 20, 2023 • edited Loading

chhagedorn left a comment

Choose a reason for hiding this comment

chhagedorn Jun 21, 2023

Choose a reason for hiding this comment

eme64 Jun 22, 2023

Choose a reason for hiding this comment

chhagedorn Jun 23, 2023

Choose a reason for hiding this comment

eme64 commented Jun 26, 2023 • edited Loading

vnkozlov left a comment

Choose a reason for hiding this comment

eme64 commented Jun 28, 2023

openjdk bot commented Jun 28, 2023

openjdk bot commented Jun 28, 2023

eme64 commented Jun 15, 2023 •

edited by openjdk bot

Loading

mlbridge bot commented Jun 19, 2023 •

edited

Loading

openjdk bot commented Jun 20, 2023 •

edited

Loading

eme64 commented Jun 26, 2023 •

edited

Loading