8334396: RISC-V: verify perf of ReverseBytesI/L

[Hamlin-Li]

Hi,
Can you help to review the patch?

The test data below shows that, if zbb is not supported on current hardware, then intrinsic brings perf regression rather than benefit.
So, ReverseBytesI/L should only be enabled when zbb is supported.

Test data

existing benchmark:
test/micro/org/openjdk/bench/java/lang/Longs.java, reverseBytes()
test/micro/org/openjdk/bench/java/lang/Integers.java, reverseBytes()

all tests running below with rvv disabled.
tested on K230-CanMV

1. -XX:-UseZbb, intrinsic enabled by default
   o.o.b.j.lang.Integers.reverseBytes N/A 500 avgt 6128.125 ns/op
   o.o.b.j.lang.Longs.reverseBytes N/A 500 avgt 10807.930 ns/op

2. -XX:+UseZbb, intrinsic enabled by default
   o.o.b.j.lang.Integers.reverseBytes N/A 500 avgt 1788.990 ns/op
   o.o.b.j.lang.Longs.reverseBytes N/A 500 avgt 1113.734 ns/op

3. -XX:-UseZbb, and disable ReverseBytesI/L instrinsic
   o.o.b.j.lang.Integers.reverseBytes N/A 500 avgt 3552.902 ns/op
   o.o.b.j.lang.Longs.reverseBytes N/A 500 avgt 4586.980 ns/op

---

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-8334396: RISC-V: verify perf of ReverseBytesI/L (Sub-task - P4)

Reviewers

• Fei Yang (@RealFYang - Reviewer)
• Robbin Ehn (@robehn - Reviewer)

Reviewing

Using git

Checkout this PR locally:
$ git fetch https://git.openjdk.org/jdk.git pull/19750/head:pull/19750
$ git checkout pull/19750

Update a local copy of the PR:
$ git checkout pull/19750
$ git pull https://git.openjdk.org/jdk.git pull/19750/head

Using Skara CLI tools

Checkout this PR locally:
$ git pr checkout 19750

View PR using the GUI difftool:
$ git pr show -t 19750

Using diff file

Download this PR as a diff file:
https://git.openjdk.org/jdk/pull/19750.diff

Webrev

Link to Webrev Comment

[bridgekeeper]

👋 Welcome back mli! 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]

@Hamlin-Li 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:

8334396: RISC-V: verify perf of ReverseBytesI/L

Reviewed-by: fyang, rehn

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 117 new commits pushed to the master branch:

• c6f3bf4: 8334026: Provide a diagnostic PrintMemoryMapAtExit switch on Linux
• cabd104: 8334164: The fix for JDK-8322811 should use _filename.is_set() rather than strcmp()
• d7dad50: 8334544: C2: wrong control assigned in PhaseIdealLoop::clone_assertion_predicate_for_unswitched_loops()
• ff30240: 8334239: Introduce macro for ubsan method/function exclusions
• 2d4185f: 8332717: ZGC: Division by zero in heuristics
• fad6644: 8333754: Add a Test against ECDSA and ECDH NIST Test vector
• b211929: 8334570: Problem list gc/TestAlwaysPreTouchBehavior.java
• 4e58d8c: 8309821: Link to hidden classes section in Class specification for Class::isHidden
• 78682fe: 8329288: Update Visual Studio visibility support for POSIX functions
• bcf4bb4: 8333344: JMX attaching of Subject does not work when security manager not allowed
• ... and 107 more: https://git.openjdk.org/jdk/compare/83b34410e326c47f357a37c3a337b7dedb8cbbda...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]

@Hamlin-Li 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]

Webrevs

• 00: Full (dca85a5c)

[RealFYang]

Good catch! Did you check the friends ReverseBytesS/ReverseBytesUS? It's very likely that they bear a similar peformance issue.

[Hamlin-Li]

Good catch! Did you check the friends ReverseBytesS/ReverseBytesUS? It's very likely that they bear a similar peformance issue.

I think so too, but not yet check ReverseBytesS/ReverseBytesUS, I don't find the existing micro test for them, so I plan to put them in other pr's. And I created an umbrella bug at: https://bugs.openjdk.org/browse/JDK-8334395, am planning to go through (verify and fix) the intrinsics which might lead to potential performance regression gradually.

[RealFYang]

All right then! Thanks.

[robehn]

Hi, a Q.

The cost in AD file is:
ins_cost(ALU_COST * 2);

Without ZBB it looks like it should be 17x.
As we don't know if the alternative is alwasy cheaper, so why not set the cost correctly and the compiler would do the right thing ?
Meaning we can still have instrinsic but with correct cost?

[Hamlin-Li]

Hi, a Q.

The cost in AD file is: ins_cost(ALU_COST * 2);

Without ZBB it looks like it should be 17x. As we don't know if the alternative is alwasy cheaper, so why not set the cost correctly and the compiler would do the right thing ? Meaning we can still have instrinsic but with correct cost?

Seems the compiler will not decide whether to pick an intrinsic or just JIT it based on the cost of the supplied instrinsic. And I think the test result confirm this point.
So, if one intrinsic is poorer than JIT version in performance, what we can do is to remove it.

[robehn]

Ok, thanks.

[Hamlin-Li]

Thanks @robehn @RealFYang for your reviewing.

[Hamlin-Li]

/integrate

[openjdk]

Going to push as commit 6420846.
Since your change was applied there have been 117 commits pushed to the master branch:

• c6f3bf4: 8334026: Provide a diagnostic PrintMemoryMapAtExit switch on Linux
• cabd104: 8334164: The fix for JDK-8322811 should use _filename.is_set() rather than strcmp()
• d7dad50: 8334544: C2: wrong control assigned in PhaseIdealLoop::clone_assertion_predicate_for_unswitched_loops()
• ff30240: 8334239: Introduce macro for ubsan method/function exclusions
• 2d4185f: 8332717: ZGC: Division by zero in heuristics
• fad6644: 8333754: Add a Test against ECDSA and ECDH NIST Test vector
• b211929: 8334570: Problem list gc/TestAlwaysPreTouchBehavior.java
• 4e58d8c: 8309821: Link to hidden classes section in Class specification for Class::isHidden
• 78682fe: 8329288: Update Visual Studio visibility support for POSIX functions
• bcf4bb4: 8333344: JMX attaching of Subject does not work when security manager not allowed
• ... and 107 more: https://git.openjdk.org/jdk/compare/83b34410e326c47f357a37c3a337b7dedb8cbbda...master

Your commit was automatically rebased without conflicts.

[openjdk]

@Hamlin-Li Pushed as commit 6420846.

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