8283187: C2: loop candidate for superword not always unrolled fully if superword fails #7822

rwestrel · 2022-03-15T15:45:08Z

IdealLoopTree::policy_unroll_slp_analysis() restricts unrolling of
loops that are found to be candidate for superword but if superword
then fails, unrolling doesn't always resume. SuperWord::output() has
logic for that in one of the failure paths but some other paths are
not covered (for instance in the case of the test case). I propose
that superword explictly returns whether it succeeded or not and that
in case of failure, for a main loop, unrolling be attempted again.

Progress

Change must not contain extraneous whitespace
Commit message must refer to an issue
Change must be properly reviewed

Issue

JDK-8283187: C2: loop candidate for superword not always unrolled fully if superword fails

Reviewers

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

Reviewing

Using git

Checkout this PR locally:
$ git fetch https://git.openjdk.java.net/jdk pull/7822/head:pull/7822
$ git checkout pull/7822

Update a local copy of the PR:
$ git checkout pull/7822
$ git pull https://git.openjdk.java.net/jdk pull/7822/head

Using Skara CLI tools

Checkout this PR locally:
$ git pr checkout 7822

View PR using the GUI difftool:
$ git pr show -t 7822

Using diff file

Download this PR as a diff file:
https://git.openjdk.java.net/jdk/pull/7822.diff

bridgekeeper · 2022-03-15T15:45:30Z

👋 Welcome back roland! 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 · 2022-03-15T15:47:24Z

@rwestrel 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 · 2022-03-15T16:00:24Z

Webrevs

TobiHartmann

Good catch! The fix looks good to me.

test/hotspot/jtreg/compiler/c2/irTests/TestSuperwordFailsUnrolling.java

openjdk · 2022-03-22T09:20:30Z

@rwestrel 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:

8283187: C2: loop candidate for superword not always unrolled fully if superword fails

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 24 new commits pushed to the master branch:

2c43ecb: 8283323: libharfbuzz optimization level results in extreme build times
5905b02: 8276799: Implementation of JEP 422: Linux/RISC-V Port
b05d4cc: 8283350: (tz) Update Timezone Data to 2022a
a35afe8: 8283519: Hsdis with capstone should annotate output
e6f707a: 8283437: Refactor imageio classes javadoc to use @throws instead of @exception
af18b11: 8283574: Use Klass::_id for type checks in the C++ code
5cf580e: 8283587: [BACKOUT] Invalid generic signature for redefined classes
2ef9767: 8270331: [TESTBUG] Error: Not a test or directory containing tests: java/awt/print/PrinterJob/InitToBlack.java
a6740c0: 8283408: Fix a C2 crash when filling arrays with unsafe
1a84d75: 8283594: Improve docs of ElementScanner classes
... and 14 more: https://git.openjdk.java.net/jdk/compare/91fab6ad59d2a4baf58802fc6e6039af3dd8d578...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.

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

rwestrel · 2022-03-22T12:18:54Z

Good catch! The fix looks good to me.

thanks for the review! I applied your suggestions.

chhagedorn

That looks good!

chhagedorn · 2022-03-23T07:07:28Z

src/hotspot/share/opto/superword.cpp

+  if (!cl->is_valid_counted_loop(T_INT)) return false; // skip malformed counted loop

  bool post_loop_allowed = (PostLoopMultiversioning && Matcher::has_predicated_vectors() && cl->is_post_loop());
  if (post_loop_allowed) {
-    if (cl->is_reduction_loop()) return; // no predication mapping
+    if (cl->is_reduction_loop()) return false; // no predication mapping
    Node *limit = cl->limit();
-    if (limit->is_Con()) return; // non constant limits only
+    if (limit->is_Con()) return false; // non constant limits only
    // Now check the limit for expressions we do not handle
    if (limit->is_Add()) {
      Node *in2 = limit->in(2);
      if (in2->is_Con()) {
        int val = in2->get_int();
        // should not try to program these cases
-        if (val < 0) return;
+        if (val < 0) return false;
      }
    }
  }

  // skip any loop that has not been assigned max unroll by analysis
  if (do_optimization) {
-    if (SuperWordLoopUnrollAnalysis && cl->slp_max_unroll() == 0) return;
+    if (SuperWordLoopUnrollAnalysis && cl->slp_max_unroll() == 0) return false;


While at it, you could also add braces for the one liner ifs.

Should be done in the new commit

TobiHartmann

I executed some testing and compiler/c2/irTests/TestSuperwordFailsUnrolling.java fails on Windows only with:

java.lang.Error: failed to install whitebox classes
	at compiler.lib.ir_framework.TestFramework.installWhiteBox(TestFramework.java:529)
	at compiler.lib.ir_framework.TestFramework.start(TestFramework.java:318)
	at compiler.lib.ir_framework.TestFramework.run(TestFramework.java:210)
	at compiler.lib.ir_framework.TestFramework.run(TestFramework.java:199)
	at compiler.c2.irTests.TestSuperwordFailsUnrolling.main(TestSuperwordFailsUnrolling.java:48)
	at java.base/jdk.internal.reflect.DirectMethodHandleAccessor.invoke(DirectMethodHandleAccessor.java:104)
	at java.base/java.lang.reflect.Method.invoke(Method.java:577)
	at com.sun.javatest.regtest.agent.MainWrapper$MainThread.run(MainWrapper.java:127)
	at java.base/java.lang.Thread.run(Thread.java:828)
Caused by: java.nio.file.FileSystemException: sun\\hotspot\\WhiteBox.class: The process cannot access the file because it is being used by another process
	at java.base/sun.nio.fs.WindowsException.translateToIOException(WindowsException.java:92)
	at java.base/sun.nio.fs.WindowsException.rethrowAsIOException(WindowsException.java:103)
	at java.base/sun.nio.fs.WindowsException.rethrowAsIOException(WindowsException.java:108)
	at java.base/sun.nio.fs.WindowsFileSystemProvider.implDelete(WindowsFileSystemProvider.java:275)
	at java.base/sun.nio.fs.AbstractFileSystemProvider.deleteIfExists(AbstractFileSystemProvider.java:110)
	at java.base/java.nio.file.Files.deleteIfExists(Files.java:1191)
	at java.base/java.nio.file.Files.copy(Files.java:3149)
	at jdk.test.lib.helpers.ClassFileInstaller.writeToDisk(ClassFileInstaller.java:296)
	at jdk.test.lib.helpers.ClassFileInstaller.writeClassToDisk(ClassFileInstaller.java:255)
	at jdk.test.lib.helpers.ClassFileInstaller.writeClassToDisk(ClassFileInstaller.java:237)
	at jdk.test.lib.helpers.ClassFileInstaller.writeClassToDisk(ClassFileInstaller.java:234)
	at jdk.test.lib.helpers.ClassFileInstaller.main(ClassFileInstaller.java:97)
	at compiler.lib.ir_framework.TestFramework.installWhiteBox(TestFramework.java:527)
	... 8 more

Weird but reproducible.

TobiHartmann · 2022-03-23T07:29:54Z

Let me try to reproduce with -DSkipWhiteBoxInstall=true.

TobiHartmann · 2022-03-23T07:36:20Z

This fixes the issue:

- * @run main/othervm -Xbootclasspath/a:. -XX:+UnlockDiagnosticVMOptions -XX:+WhiteBoxAPI compiler.c2.irTests.TestSuperwordFailsUnrolling
+ * @run main/othervm -Xbootclasspath/a:. -DSkipWhiteBoxInstall=true -XX:+UnlockDiagnosticVMOptions -XX:+WhiteBoxAPI compiler.c2.irTests.TestSuperwordFailsUnrolling

@chhagedorn Maybe the IR verification framework could detect this automatically?

chhagedorn · 2022-03-23T07:42:59Z

This fixes the issue:

- * @run main/othervm -Xbootclasspath/a:. -XX:+UnlockDiagnosticVMOptions -XX:+WhiteBoxAPI compiler.c2.irTests.TestSuperwordFailsUnrolling
+ * @run main/othervm -Xbootclasspath/a:. -DSkipWhiteBoxInstall=true -XX:+UnlockDiagnosticVMOptions -XX:+WhiteBoxAPI compiler.c2.irTests.TestSuperwordFailsUnrolling

@chhagedorn Maybe the IR verification framework could detect this automatically?

Yes, it would be good to extend the framework to handle this. I added this -DSkipWhiteBoxInstall flag at some point to prevent the problem above but was more of a quick fix. I will follow up with an RFE to make this better without the need to use -DSkipWhiteBoxInstall.

rwestrel · 2022-03-23T13:54:29Z

This fixes the issue:

- * @run main/othervm -Xbootclasspath/a:. -XX:+UnlockDiagnosticVMOptions -XX:+WhiteBoxAPI compiler.c2.irTests.TestSuperwordFailsUnrolling
+ * @run main/othervm -Xbootclasspath/a:. -DSkipWhiteBoxInstall=true -XX:+UnlockDiagnosticVMOptions -XX:+WhiteBoxAPI compiler.c2.irTests.TestSuperwordFailsUnrolling

New commit has this change

rwestrel · 2022-03-23T13:54:51Z

That looks good!

Thanks for the review @chhagedorn

openjdk · 2022-03-23T13:55:48Z

⚠️ @rwestrel This pull request contains merges that bring in commits not present in the target repository. Since this is not a "merge style" pull request, these changes will be squashed when this pull request in integrated. If this is your intention, then please ignore this message. If you want to preserve the commit structure, you must change the title of this pull request to Merge <project>:<branch> where <project> is the name of another project in the OpenJDK organization (for example Merge jdk:master).

TobiHartmann

Looks good.

chhagedorn

Looks good, thanks for the updates!

chhagedorn · 2022-03-23T07:07:28Z

src/hotspot/share/opto/superword.cpp

+  if (!cl->is_valid_counted_loop(T_INT)) return false; // skip malformed counted loop

  bool post_loop_allowed = (PostLoopMultiversioning && Matcher::has_predicated_vectors() && cl->is_post_loop());
  if (post_loop_allowed) {
-    if (cl->is_reduction_loop()) return; // no predication mapping
+    if (cl->is_reduction_loop()) return false; // no predication mapping
    Node *limit = cl->limit();
-    if (limit->is_Con()) return; // non constant limits only
+    if (limit->is_Con()) return false; // non constant limits only
    // Now check the limit for expressions we do not handle
    if (limit->is_Add()) {
      Node *in2 = limit->in(2);
      if (in2->is_Con()) {
        int val = in2->get_int();
        // should not try to program these cases
-        if (val < 0) return;
+        if (val < 0) return false;
      }
    }
  }

  // skip any loop that has not been assigned max unroll by analysis
  if (do_optimization) {
-    if (SuperWordLoopUnrollAnalysis && cl->slp_max_unroll() == 0) return;
+    if (SuperWordLoopUnrollAnalysis && cl->slp_max_unroll() == 0) return false;


While at it, you could also add braces for the one liner ifs.

rwestrel · 2022-03-24T12:52:34Z

/integrate

openjdk · 2022-03-24T12:54:01Z

Going to push as commit 14c20bc.
Since your change was applied there have been 25 commits pushed to the master branch:

1c4f5fc: 8283575: Check for GNU time fails for version >1.7
2c43ecb: 8283323: libharfbuzz optimization level results in extreme build times
5905b02: 8276799: Implementation of JEP 422: Linux/RISC-V Port
b05d4cc: 8283350: (tz) Update Timezone Data to 2022a
a35afe8: 8283519: Hsdis with capstone should annotate output
e6f707a: 8283437: Refactor imageio classes javadoc to use @throws instead of @exception
af18b11: 8283574: Use Klass::_id for type checks in the C++ code
5cf580e: 8283587: [BACKOUT] Invalid generic signature for redefined classes
2ef9767: 8270331: [TESTBUG] Error: Not a test or directory containing tests: java/awt/print/PrinterJob/InitToBlack.java
a6740c0: 8283408: Fix a C2 crash when filling arrays with unsafe
... and 15 more: https://git.openjdk.java.net/jdk/compare/91fab6ad59d2a4baf58802fc6e6039af3dd8d578...master

Your commit was automatically rebased without conflicts.

openjdk · 2022-03-24T12:54:19Z

@rwestrel Pushed as commit 14c20bc.

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

fix & test

130329d

openjdk bot added the hotspot-compiler hotspot-compiler-dev@openjdk.org label Mar 15, 2022

whitespaces

9941937

openjdk bot added the rfr Pull request is ready for review label Mar 15, 2022

TobiHartmann approved these changes Mar 22, 2022

View reviewed changes

test/hotspot/jtreg/compiler/c2/irTests/TestSuperwordFailsUnrolling.java Outdated Show resolved Hide resolved

openjdk bot added the ready Pull request is ready to be integrated label Mar 22, 2022

Update test/hotspot/jtreg/compiler/c2/irTests/TestSuperwordFailsUnrol…

cac6704

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

chhagedorn approved these changes Mar 23, 2022

View reviewed changes

TobiHartmann suggested changes Mar 23, 2022

View reviewed changes

rwestrel added 3 commits March 23, 2022 14:17

Merge branch 'master' into JDK-8283187

e2f18c2

review

b1e8856

Merge commit 'cac6704b8a1791c621918af316250e94047bff43' into JDK-8283187

8ae4fa5

test fix on x86

757f031

TobiHartmann approved these changes Mar 24, 2022

View reviewed changes

chhagedorn approved these changes Mar 24, 2022

View reviewed changes

openjdk bot added the integrated Pull request has been integrated label Mar 24, 2022

openjdk bot closed this Mar 24, 2022

openjdk bot removed ready Pull request is ready to be integrated rfr Pull request is ready for review labels Mar 24, 2022

8283187: C2: loop candidate for superword not always unrolled fully if superword fails #7822

8283187: C2: loop candidate for superword not always unrolled fully if superword fails #7822

Uh oh!

Conversation

rwestrel commented Mar 15, 2022 • 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 Mar 15, 2022

Uh oh!

openjdk bot commented Mar 15, 2022

Uh oh!

mlbridge bot commented Mar 15, 2022 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Webrevs

Uh oh!

TobiHartmann left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

openjdk bot commented Mar 22, 2022 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

rwestrel commented Mar 22, 2022

Uh oh!

chhagedorn left a comment

Choose a reason for hiding this comment

Uh oh!

chhagedorn Mar 23, 2022

Choose a reason for hiding this comment

Uh oh!

rwestrel Mar 23, 2022

Choose a reason for hiding this comment

Uh oh!

TobiHartmann left a comment

Choose a reason for hiding this comment

Uh oh!

TobiHartmann commented Mar 23, 2022

Uh oh!

TobiHartmann commented Mar 23, 2022

Uh oh!

chhagedorn commented Mar 23, 2022

Uh oh!

rwestrel commented Mar 23, 2022

Uh oh!

rwestrel commented Mar 23, 2022

Uh oh!

openjdk bot commented Mar 23, 2022

Uh oh!

TobiHartmann left a comment

Choose a reason for hiding this comment

Uh oh!

chhagedorn left a comment

Choose a reason for hiding this comment

Uh oh!

chhagedorn Mar 23, 2022

Choose a reason for hiding this comment

Uh oh!

rwestrel commented Mar 24, 2022

Uh oh!

openjdk bot commented Mar 24, 2022

Uh oh!

openjdk bot commented Mar 24, 2022

Uh oh!

Uh oh!

rwestrel commented Mar 15, 2022 •

edited by openjdk bot

Loading

mlbridge bot commented Mar 15, 2022 •

edited

Loading

openjdk bot commented Mar 22, 2022 •

edited

Loading