8288445: AArch64: C2 compilation fails with guarantee(!true || (true && (shift != 0))) failed: impossible encoding

[dean-long]

The range for aarch64 vector right-shift is 1 to the element width. This issue fixes the problem in the back-end. There is a separate problem in the front-end that shift by 0 is not always optimized out.

---

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-8288445: AArch64: C2 compilation fails with guarantee(!true || (true && (shift != 0))) failed: impossible encoding

Reviewers

• Tobias Hartmann (@TobiHartmann - Reviewer) ⚠️ Review applies to 0f927ce7
• Hao Sun (@shqking - Author)
• Ningsheng Jian (@nsjian - Committer)

Reviewing

Using git

Checkout this PR locally:
$ git fetch https://git.openjdk.org/jdk19 pull/40/head:pull/40
$ git checkout pull/40

Update a local copy of the PR:
$ git checkout pull/40
$ git pull https://git.openjdk.org/jdk19 pull/40/head

Using Skara CLI tools

Checkout this PR locally:
$ git pr checkout 40

View PR using the GUI difftool:
$ git pr show -t 40

Using diff file

Download this PR as a diff file:
https://git.openjdk.org/jdk19/pull/40.diff

[bridgekeeper]

👋 Welcome back dlong! 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]

@dean-long 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

• 04: Full - Incremental (f4761fbf)
• 03: Full - Incremental (583c54c0)
• 02: Full - Incremental (0f927ce7)
• 01: Full - Incremental (e432a308)
• 00: Full (d0c2bf33)

[TobiHartmann]

What instruction will the zero-shift be matched with then?

[shqking]

Instead of introducing immI_positive, I wonder if we can generate orr dst src for zero shift count, just as the SVE part does. E.g., https://github.com/openjdk/jdk19/blob/master/src/hotspot/cpu/aarch64/aarch64_sve.ad#L3667

[dean-long]

What instruction will the zero-shift be matched with then?

The shift instruction that takes a vector register shift count rather than an immediate.

[dean-long]

Instead of introducing immI_positive, I wonder if we can generate orr dst src for zero shift count, just as the SVE part does

I considered using left-shift by 0, but I don't think it's worth it to optimize this case in the back-end. If we really want to optimize shift by 0, I think it should be done in the front-end.

[TobiHartmann]

Looks good to me.

[openjdk]

@dean-long 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:

8288445: AArch64: C2 compilation fails with guarantee(!true || (true && (shift != 0))) failed: impossible encoding

Reviewed-by: thartmann, haosun, njian

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 38 new commits pushed to the master branch:

• 9c92da5: 8247407: tools/jlink/plugins/CompressorPluginTest.java test failing
• 7ac40f3: 8288983: broken link in com.sun.net.httpserver.SimpleFileServer
• 20f55ab: 8289044: ARM32: missing LIR_Assembler::cmove metadata type support
• bdf9902: 8288120: VerifyError with JEP 405 pattern match
• b0db333: 8288528: broken links in java.desktop
• 651cbeb: 8288080: (fc) FileChannel::map for MemorySegments should state it always throws UOE
• 1f9521e: 8287076: Document.normalizeDocument() produces different results
• a716f79: 8288589: Files.readString ignores encoding errors for UTF-16
• 4c9ea7e: 8286580: serviceability/jvmti/vthread/GetSetLocalTest failed with assert: Not supported for heap frames
• 3f5e48a: 8288781: C1: LIR_OpVisitState::maxNumberOfOperands too small
• ... and 28 more: https://git.openjdk.org/jdk19/compare/ff3db52044f69754b1ccf54961f453d4afbfba3a...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.

[shqking]

Instead of introducing immI_positive, I wonder if we can generate orr dst src for zero shift count, just as the SVE part does

I considered using left-shift by 0, but I don't think it's worth it to optimize this case in the back-end. If we really want to optimize shift by 0, I think it should be done in the front-end.

Okay.

I think we may want to make the same update to the corresponding SVE part, e.g. https://github.com/openjdk/jdk/blob/master/src/hotspot/cpu/aarch64/aarch64_sve.ad#L3662, as it's better to align with the NEON and SVE implementations.

It may deserve one separate patch, not in this bugfix one.

[dean-long]

Thanks Tobias.

[dean-long]

@shqking Yes, that's a good idea for a separate RFE targeted for jdk20.

[shqking]

LGTM (I'm not a Reviewer).

[nsjian]

Looks good to me.

[dean-long]

Thanks @nsjian, @shqking, and @theRealAph.

[dean-long]

/integrate

[openjdk]

Going to push as commit b449038.
Since your change was applied there have been 47 commits pushed to the master branch:

• 2efa89a: 8289251: ProblemList java/lang/ref/OOMEInReferenceHandler.java
• 17ef8ca: 8288524: Allow @systemProperty to appear in overview documentation
• caa6b74: 8289240: ProblemList java/lang/reflect/callerCache/ReflectionCallerCacheTest.java in -Xcomp mode
• 28913f4: 8289235: ProblemList vmTestbase/nsk/jdi/ClassType/invokeMethod/invokemethod011/TestDescription.java when run with vthread wrapper
• 7746664: 8280826: Document set of strings javac recognizes for SuppressWarnings
• 2c8ada6: 8289188: SegmentAllocator:allocateArray(*) default behavior mismatch to spec
• 699ad45: 8289093: BlockLocationPrinter fails to decode addresses with G1
• 784a0f0: 8288683: C2: And node gets wrong type due to not adding it back to the worklist in CCP
• 7e13cdb: 8289079: java/lang/Thread/jni/AttachCurrentThread/AttachTest.java#id1 failed with "RuntimeException: Test failed"
• 9c92da5: 8247407: tools/jlink/plugins/CompressorPluginTest.java test failing
• ... and 37 more: https://git.openjdk.org/jdk19/compare/ff3db52044f69754b1ccf54961f453d4afbfba3a...master

Your commit was automatically rebased without conflicts.

[openjdk]

@dean-long Pushed as commit b449038.

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