8315604: IGV: dump and visualize node bottom and phase types #15881

robcasloz · 2023-09-22T08:18:46Z

This changeset dumps node type information in IGV graphs. Type information includes node bottom types (Node::bottom_type()) and node phase types (GVN, IGVN, CCP) as IGV node properties bottom_type and phase_type. It also adds a filter (Show types) to display type information in the node labels:

The filter presents a concise version of the complete type information (for example, removing long interface lists from reference types). It shows the phase type of each node, followed by the bottom type (prefixed with B:) if it differs from the phase type (see, for example, node 124 AddI in the picture above) or is the only type available.

On the HotSpot side, the size of the stringStream buffer used to extract type information (and other properties such as dump_spec) is increased from 128 to 512 characters to prevent truncations. The additional type information and the buffer size change increase the size of the dumped graphs by around 10% on average.

Thanks to Christian Hagedorn and Toby Holenstein for providing feedback on earlier versions of the changeset.

Testing

tier1-3 (windows-x64, linux-x64, linux-aarch64, macosx-x64, macosx-aarch64; release and debug mode)
tested that thousands of graphs are correctly opened and visualized with IGV using different filter configurations
tested that graphs dumped interactively from GDB are correctly opened and visualized
tested that graphs dumped from earlier JVM versions are correctly handled by the new filter

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-8315604: IGV: dump and visualize node bottom and phase types (Enhancement - P4)

Reviewers

Tobias Hartmann (@TobiHartmann - Reviewer) ⚠️ Review applies to 61183662
Christian Hagedorn (@chhagedorn - Reviewer) ⚠️ Review applies to 61183662
Tobias Holenstein (@tobiasholenstein - Reviewer)

Contributors

Tobias Holenstein <tholenstein@openjdk.org>

Reviewing

Using git

Checkout this PR locally:
$ git fetch https://git.openjdk.org/jdk.git pull/15881/head:pull/15881
$ git checkout pull/15881

Update a local copy of the PR:
$ git checkout pull/15881
$ git pull https://git.openjdk.org/jdk.git pull/15881/head

Using Skara CLI tools

Checkout this PR locally:
$ git pr checkout 15881

View PR using the GUI difftool:
$ git pr show -t 15881

Using diff file

Download this PR as a diff file:
https://git.openjdk.org/jdk/pull/15881.diff

Webrev

Link to Webrev Comment

bridgekeeper · 2023-09-22T08:19:46Z

👋 Welcome back rcastanedalo! 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 · 2023-09-22T08:22:25Z

@robcasloz 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 · 2023-09-22T08:58:32Z

Webrevs

TobiHartmann

This is a very nice improvement! The VM changes look good to me.

openjdk · 2023-09-25T05:55:09Z

@robcasloz 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:

8315604: IGV: dump and visualize node bottom and phase types

Co-authored-by: Tobias Holenstein <tholenstein@openjdk.org>
Reviewed-by: thartmann, chagedorn, tholenstein

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 340 new commits pushed to the master branch:

bd918f4: 8316401: sun/tools/jhsdb/JStackStressTest.java failed with "InternalError: We should have found a thread that owns the anonymous lock"
c45308a: 8301327: convert assert to guarantee in Handle_IDiv_Exception
bd7bb67: 8317257: RISC-V: llvm build broken
0259da9: 8301686: TLS 1.3 handshake fails if server_name doesn't match resuming session
355811a: 8316559: Refactor some util/Calendar tests to JUnit
ecb5e8a: 8317264: Pattern.Bound has static fields that should be static final.
f2c221d: 8317119: Remove unused imports in the java.util.stream package
09dad0e: 8313810: BoxLayout uses
instead of list for layout options
73a47f0: 4622866: javax.swing.text.Document.remove(int, int) has a misleading picture
cfcbfc6: 8316879: RegionMatches1Tests fails if CompactStrings are disabled after JDK-8302163
... and 330 more: https://git.openjdk.org/jdk/compare/024133b089d911dcc3ea70dfdaa6b150b14a9eb4...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

That's a nice improvement! Cannot say much about the filter code - maybe someone else should have a look at that as well.

src/hotspot/share/opto/idealGraphPrinter.cpp

robcasloz · 2023-09-26T12:54:25Z

This is a very nice improvement! The VM changes look good to me.

Thanks for reviewing, Tobias!

tobiasholenstein · 2023-09-26T13:23:00Z

Thanks for working on this @robcasloz . That’s a feature I was always missing!

Suggestion: I would find it better for the mind to process if the second line has a different font from the first line in a widget. (e.g. different color, like dark gray or italic) What do you think?

E.g. adding the following code in FigureWidget.java

for (int i=1; i<labelWidgets.size(); i++) {
    labelWidgets.get(i).setFont(Diagram.FONT.deriveFont(Font.ITALIC));
    labelWidgets.get(i).setForeground(Color.DARK_GRAY);
}

robcasloz · 2023-09-26T16:40:50Z

Thanks for working on this @robcasloz . That’s a feature I was always missing!

Suggestion: I would find it better for the mind to process if the second line has a different font from the first line in a widget. (e.g. different color, like dark gray or italic) What do you think?

E.g. adding the following code in FigureWidget.java
for (int i=1; i<labelWidgets.size(); i++) {
    labelWidgets.get(i).setFont(Diagram.FONT.deriveFont(Font.ITALIC));
    labelWidgets.get(i).setForeground(Color.DARK_GRAY);
}

Great suggestion, done, thanks!

robcasloz · 2023-09-26T16:41:47Z

/contributor add @tobiasholenstein

openjdk · 2023-09-26T16:58:57Z

@robcasloz
Contributor Tobias Holenstein <tholenstein@openjdk.org> successfully added.

robcasloz · 2023-09-29T08:23:04Z

The latest commits change the order and syntax used when displaying both types (when they are available and differ from each other) after offline feedback from @chhagedorn and @tobiasholenstein. See the updated pull request description for details. They also improve the display of tuple types and pretty-print inequalities.

@tobiasholenstein could you please review the IGV parts of this change?

tobiasholenstein

Thanks for the update @robcasloz . The code looks good to me!

robcasloz · 2023-10-02T06:55:32Z

Thanks Toby, Christian, and Tobias for reviewing!

robcasloz · 2023-10-02T06:55:37Z

/integrate

openjdk · 2023-10-02T07:04:42Z

Going to push as commit 207819a.
Since your change was applied there have been 352 commits pushed to the master branch:

8fcf70e: 5066247: Refine the spec of equals() and hashCode() for j.text.Format classes
93f662f: 8317335: Build on windows fails after 8316645
b8fa6c2: 8316186: RISC-V: Remove PlatformCmpxchg<4>
fb055e7: 8316645: RISC-V: Remove dependency on libatomic by adding cmpxchg 1b
009f5e1: 8317141: Remove unused validIndex method from URLClassPath$JarLoader
47569a2: 8295919: java.security.MessageDigest.isEqual does not adhere to @implNote
5a6aa56: 8303959: tools/jpackage/share/RuntimePackageTest.java fails with java.lang.AssertionError missing files
014c95a: 8317126: Redundant entries in Windows tzmappings file
fa0697a: 8316998: Remove redundant type arguments in the java.util.stream package
49376e4: 8316000: File.setExecutable silently fails if file does not exist
... and 342 more: https://git.openjdk.org/jdk/compare/024133b089d911dcc3ea70dfdaa6b150b14a9eb4...master

Your commit was automatically rebased without conflicts.

openjdk · 2023-10-02T07:04:56Z

@robcasloz Pushed as commit 207819a.

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

robcasloz added 6 commits September 11, 2023 14:37

Initial implementation

ae8376e

Print bottom and phase type info only for interesting nodes

4e74889

Add filter for each piece of type info (bottom, phase)

5aefbcd

Simplify logic to append type info to extra line

69750d7

Unify both sources of type info into a single filter

776f67e

Show bottom type as default, and phase type if different

6118366

openjdk bot added the hotspot-compiler hotspot-compiler-dev@openjdk.org label Sep 22, 2023

robcasloz marked this pull request as ready for review September 22, 2023 08:51

openjdk bot added the rfr Pull request is ready for review label Sep 22, 2023

TobiHartmann approved these changes Sep 25, 2023

View reviewed changes

openjdk bot added the ready Pull request is ready to be integrated label Sep 25, 2023

chhagedorn approved these changes Sep 25, 2023

View reviewed changes

src/hotspot/share/opto/idealGraphPrinter.cpp Outdated Show resolved Hide resolved

Access Compile pointer via IdealGraphPrinter::C

ecbc8c2

Format extra label using italics and light gray as suggested by Toby

7ff67b8

Update copyright year

2a9268e

robcasloz added 3 commits September 29, 2023 09:39

Fix simplification of tuple types

4f2a5d3

Show phase type first, then bottom type if different

c9cb87c

Pretty-print inequalities

cd7555d

tobiasholenstein approved these changes Sep 29, 2023

View reviewed changes

openjdk bot added the integrated Pull request has been integrated label Oct 2, 2023

openjdk bot closed this Oct 2, 2023

openjdk bot removed ready Pull request is ready to be integrated rfr Pull request is ready for review labels Oct 2, 2023

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

8315604: IGV: dump and visualize node bottom and phase types #15881

8315604: IGV: dump and visualize node bottom and phase types #15881

robcasloz commented Sep 22, 2023 •

edited by openjdk bot

Loading

bridgekeeper bot commented Sep 22, 2023

openjdk bot commented Sep 22, 2023

mlbridge bot commented Sep 22, 2023 •

edited

Loading

TobiHartmann left a comment

openjdk bot commented Sep 25, 2023 •

edited

Loading

chhagedorn left a comment

robcasloz commented Sep 26, 2023

tobiasholenstein commented Sep 26, 2023

robcasloz commented Sep 26, 2023

robcasloz commented Sep 26, 2023

openjdk bot commented Sep 26, 2023

robcasloz commented Sep 29, 2023

tobiasholenstein left a comment

robcasloz commented Oct 2, 2023

robcasloz commented Oct 2, 2023

openjdk bot commented Oct 2, 2023

openjdk bot commented Oct 2, 2023

8315604: IGV: dump and visualize node bottom and phase types #15881

8315604: IGV: dump and visualize node bottom and phase types #15881

Conversation

robcasloz commented Sep 22, 2023 • edited by openjdk bot Loading

Testing

Progress

Issue

Reviewers

Contributors

Reviewing

Webrev

bridgekeeper bot commented Sep 22, 2023

openjdk bot commented Sep 22, 2023

mlbridge bot commented Sep 22, 2023 • edited Loading

Webrevs

TobiHartmann left a comment

Choose a reason for hiding this comment

openjdk bot commented Sep 25, 2023 • edited Loading

chhagedorn left a comment

Choose a reason for hiding this comment

robcasloz commented Sep 26, 2023

tobiasholenstein commented Sep 26, 2023

robcasloz commented Sep 26, 2023

robcasloz commented Sep 26, 2023

openjdk bot commented Sep 26, 2023

robcasloz commented Sep 29, 2023

tobiasholenstein left a comment

Choose a reason for hiding this comment

robcasloz commented Oct 2, 2023

robcasloz commented Oct 2, 2023

openjdk bot commented Oct 2, 2023

openjdk bot commented Oct 2, 2023

robcasloz commented Sep 22, 2023 •

edited by openjdk bot

Loading

mlbridge bot commented Sep 22, 2023 •

edited

Loading

openjdk bot commented Sep 25, 2023 •

edited

Loading