8317572: C2 SuperWord: refactor/improve TraceSuperWord, replace VectorizeDebugOption with TraceAutoVectorization #17586

eme64 · 2024-01-26T12:49:50Z

Subtask of #16620

I got approval to remove VectorizeDebugOption: JDK-8320668

I want a more general flag for AutoVectorization, that can trace different components of AutoVectorization.
It should be a CompileCommand, so that it can select which methods it traces for.

TraceSuperWord should still look similar, and select a subset of the TraceAutoVectorization components (those for SuperWord), but still apply to all classes/methods.

With more refactoring later in JDK-8315361, this flag should become more usable and interpretable. Especially, the idea is that different components of the VLoop / VLoopAnalyzer can have tracing enabled / disabled.

How to use the flag:
Get "help", i.e. see all available tags:
./java -Xcomp -XX:CompileCommand=TraceAutoVectorization,*::*,help --version

See "rejections" (i.e. failures where we don't vectorize) and successes (using TraceNewVectors):
./java -Xcomp -XX:CompileCommand=TraceAutoVectorization,*::*,SW_REJECTIONS -XX:+TraceNewVectors --version
The results are currently underwhealming. I will have to track many more failures, and I will do that with the bigger refactoring, when I move around the code and require error code returning everywhere, and then I can use that error code for printing.

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-8317572: C2 SuperWord: refactor/improve TraceSuperWord, replace VectorizeDebugOption with TraceAutoVectorization (Sub-task - P4)

Reviewers

Christian Hagedorn (@chhagedorn - Reviewer)
Vladimir Kozlov (@vnkozlov - Reviewer) ⚠️ Review applies to 28e0893f

Reviewing

Using git

Checkout this PR locally:
$ git fetch https://git.openjdk.org/jdk.git pull/17586/head:pull/17586
$ git checkout pull/17586

Update a local copy of the PR:
$ git checkout pull/17586
$ git pull https://git.openjdk.org/jdk.git pull/17586/head

Using Skara CLI tools

Checkout this PR locally:
$ git pr checkout 17586

View PR using the GUI difftool:
$ git pr show -t 17586

Using diff file

Download this PR as a diff file:
https://git.openjdk.org/jdk/pull/17586.diff

Webrev

Link to Webrev Comment

bridgekeeper · 2024-01-26T12:50:55Z

👋 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 · 2024-01-26T12:52:36Z

@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.

eme64 · 2024-01-26T13:18:48Z

src/hotspot/share/opto/superword.cpp

-    _vector_loop_debug = phase->C->directive()->VectorizeDebugOption;
-  }
-
-#endif


Note: initialization now happens via _vtrace field, and is constructed implicitly.

eme64 · 2024-01-26T13:21:22Z

src/hotspot/share/opto/superword.cpp

-    }
-  }
-#endif
-


This was only printed if _do_vector_loop was on, i.e. if OptionVectorize enabled (kinda odd anyway).
And we already do print_bb, which prints all relevant nodes (enabled with SW_INFO or TraceSuperWord).

eme64 · 2024-01-26T13:21:52Z

src/hotspot/share/opto/superword.cpp

-        _nlist.at(j)->dump();
-      }
-    }
-#endif


We already print the mem slice in mem_slice_preds.

eme64 · 2024-01-26T13:23:34Z

src/hotspot/share/opto/superword.cpp

      uint vlen = p->size();
      uint vlen_in_bytes = 0;
      Node* vn = nullptr;
-      NOT_PRODUCT(if(is_trace_cmov()) {tty->print_cr("VPointer::output: %d executed first, %d executed last in pack", first->_idx, n->_idx); print_pack(p);})


This was behind the wrong flag is_trace_cmov, and I think it was never used anyway.

eme64 · 2024-01-26T13:25:10Z

src/hotspot/share/opto/vectorization.cpp

  _nstack(nstack), _analyze_only(analyze_only), _stack_idx(0)
 #ifndef PRODUCT
-  , _tracer((phase->C->directive()->VectorizeDebugOption & 2) > 0)
+  , _tracer(phase->C->directive()->traceautovectorization_tags().at(TraceAutoVectorizationTag::POINTER_ANALYSIS))


I now do the ugly thing. Later, with the bigger refactoring, I will pass VLoop into the VPointer, and then we can access the flag via VPointer -> VLoop -> VTrace.

eme64 · 2024-01-26T13:25:43Z

src/hotspot/share/opto/vectorization.hpp

+  }
+};
+#endif
+


The idea is that this is going to be a "component" of VLoop, once I do the bigger refactoring.

mlbridge · 2024-01-26T13:34:47Z

Webrevs

chhagedorn

Generally looks good! I have some comments.

src/hotspot/share/opto/traceAutoVectorizationTag.hpp

src/hotspot/share/compiler/compilerDirectives.cpp

src/hotspot/share/opto/superword.cpp

src/hotspot/share/opto/superword.hpp

src/hotspot/share/opto/traceAutoVectorizationTag.hpp

chhagedorn · 2024-01-26T15:09:43Z

src/hotspot/share/opto/superword.cpp

 // Return a memory slice (node list) in predecessor order starting at "start"
 void SuperWord::mem_slice_preds(Node* start, Node* stop, GrowableArray<Node*> &preds) {
  assert(preds.length() == 0, "start empty");
  Node* n = start;


There is still a usage of TraceSuperWord on L927. Should this also be replaced?

It will be removed with #17585 anyway ;)

src/hotspot/share/opto/traceAutoVectorizationTag.hpp

Co-authored-by: Christian Hagedorn <christian.hagedorn@oracle.com>

vnkozlov

It is good. Two comments.

vnkozlov · 2024-01-26T19:02:16Z

src/hotspot/share/compiler/compilerOracle.cpp

          jio_snprintf(errorbuf, buf_size, "Unrecognized intrinsic detected in %s: %s", option2name(option), validator.what());
        }
      }
 #ifndef PRODUCT


Missing #ifdef COMPILER2 for this and PrintIdealPhase.

vnkozlov · 2024-01-26T19:03:47Z

src/hotspot/share/compiler/directivesParser.cpp

          if (!valid) {
            error(VALUE_ERROR, "Unrecognized intrinsic detected in DisableIntrinsic: %s", validator.what());
          }
+        } else if (strncmp(option_key->name, "TraceAutoVectorization", 22) == 0) {


Missing #ifndef PRODUCT and #ifdef COMPILER2 for this and for PrintIdealPhase.

vnkozlov

Looks good.

openjdk · 2024-01-27T20:27:56Z

@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:

8317572: C2 SuperWord: refactor/improve TraceSuperWord, replace VectorizeDebugOption with TraceAutoVectorization

Reviewed-by: chagedorn, kvn

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 4 new commits pushed to the master branch:

72ba817: 8324236: compiler/ciReplay/TestInliningProtectionDomain.java failed with RuntimeException: should only dump inline information for ... expected true, was false
b39b876: 8324304: RISC-V: add hw probe flags
69586e7: 8322996: BoxLockNode creation fails with assert(reg < CHUNK_SIZE) failed: sanity
f0bae79: 8324750: C2: rename Matcher methods using "superword" -> "autovectorization"

Please see this link for an up-to-date comparison between the source branch of this pull request and the master branch.
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, thanks for the updates!

chhagedorn · 2024-01-29T07:47:23Z

src/hotspot/share/opto/traceAutoVectorizationTag.hpp

+#ifndef SHARE_OPTO_TRACE_AUTO_VECTORIZATION_TAG_HPP
+#define SHARE_OPTO_TRACE_AUTO_VECTORIZATION_TAG_HPP


I think for this define, you should keep SHARE_OPTO_TRACEAUTOVECTORIZATIONTAG_HPP to follow the convention of other files where we do not insert underlines in the filename.

src/hotspot/share/opto/traceAutoVectorizationTag.hpp

eme64 · 2024-01-29T10:12:47Z

@chhagedorn @vnkozlov thanks for the reviews and helpful suggestions!
/integrate

openjdk · 2024-01-29T10:13:45Z

Going to push as commit 3066d49.
Since your change was applied there have been 6 commits pushed to the master branch:

7a300b6: 8324213: C1: There is no need for Canonicalizer to handle IfOp
628348d: 8324186: Use "dmb.ishst+dmb.ishld" for release barrier
72ba817: 8324236: compiler/ciReplay/TestInliningProtectionDomain.java failed with RuntimeException: should only dump inline information for ... expected true, was false
b39b876: 8324304: RISC-V: add hw probe flags
69586e7: 8322996: BoxLockNode creation fails with assert(reg < CHUNK_SIZE) failed: sanity
f0bae79: 8324750: C2: rename Matcher methods using "superword" -> "autovectorization"

Your commit was automatically rebased without conflicts.

openjdk · 2024-01-29T10:13:52Z

@eme64 Pushed as commit 3066d49.

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

8317572

48c8f64

openjdk bot changed the title ~~8317572~~ 8317572: C2 SuperWord: refactor/improve TraceSuperWord, replace VectorizeDebugOption with TraceAutoVectorization Jan 26, 2024

openjdk bot added the hotspot-compiler hotspot-compiler-dev@openjdk.org label Jan 26, 2024

eme64 added 2 commits January 26, 2024 13:58

reordering some things

9736ae8

a bit more

0ef53ac

eme64 commented Jan 26, 2024

View reviewed changes

eme64 marked this pull request as ready for review January 26, 2024 13:27

openjdk bot added the rfr Pull request is ready for review label Jan 26, 2024

chhagedorn reviewed Jan 26, 2024

View reviewed changes

eme64 and others added 4 commits January 26, 2024 16:40

Apply suggestions from code review

7772f60

Co-authored-by: Christian Hagedorn <christian.hagedorn@oracle.com>

more for Christian

5706391

move code to StringUtils::CommaSeparatedStringIterator

fc59b6b

fix a test

c9ac434

vnkozlov reviewed Jan 26, 2024

View reviewed changes

eme64 added 2 commits January 27, 2024 06:25

add PRODUCT and COMPILER2 ifdefs for Vladimir

be11c9d

fix missing include

28e0893

vnkozlov approved these changes Jan 27, 2024

View reviewed changes

openjdk bot added the ready Pull request is ready to be integrated label Jan 27, 2024

chhagedorn approved these changes Jan 29, 2024

View reviewed changes

eme64 commented Jan 29, 2024

View reviewed changes

src/hotspot/share/opto/traceAutoVectorizationTag.hpp Outdated Show resolved Hide resolved

src/hotspot/share/opto/traceAutoVectorizationTag.hpp Outdated Show resolved Hide resolved

adjust hpp guards for Christian

ddd01ce

Merge branch 'master' into JDK-8317572

bf5e826

chhagedorn approved these changes Jan 29, 2024

View reviewed changes

openjdk bot added the integrated Pull request has been integrated label Jan 29, 2024

openjdk bot closed this Jan 29, 2024

openjdk bot removed ready Pull request is ready to be integrated rfr Pull request is ready for review labels Jan 29, 2024

		#ifndef SHARE_OPTO_TRACE_AUTO_VECTORIZATION_TAG_HPP
		#define SHARE_OPTO_TRACE_AUTO_VECTORIZATION_TAG_HPP

8317572: C2 SuperWord: refactor/improve TraceSuperWord, replace VectorizeDebugOption with TraceAutoVectorization #17586

8317572: C2 SuperWord: refactor/improve TraceSuperWord, replace VectorizeDebugOption with TraceAutoVectorization #17586

Uh oh!

Conversation

eme64 commented Jan 26, 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 Jan 26, 2024

Uh oh!

openjdk bot commented Jan 26, 2024

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

mlbridge bot commented Jan 26, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Webrevs

Uh oh!

chhagedorn left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

vnkozlov left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

vnkozlov left a comment

Choose a reason for hiding this comment

Uh oh!

openjdk bot commented Jan 27, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

chhagedorn left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

eme64 commented Jan 29, 2024

Uh oh!

openjdk bot commented Jan 29, 2024

Uh oh!

openjdk bot commented Jan 29, 2024

Uh oh!

Reviewers

Assignees

eme64 commented Jan 26, 2024 •

edited by openjdk bot

Loading

mlbridge bot commented Jan 26, 2024 •

edited

Loading

openjdk bot commented Jan 27, 2024 •

edited

Loading