[llvm] [SLP] Skip MMRA metadata if instruction can't have it (PR #173319)
via llvm-commits
llvm-commits at lists.llvm.org
Mon Dec 22 15:53:41 PST 2025
github-actions[bot] wrote:
<!--PREMERGE ADVISOR COMMENT: Linux-->
# :penguin: Linux x64 Test Results
* 167411 tests passed
* 2973 tests skipped
* 1 test failed
## Failed Tests
(click on a test name to see its output)
### LLVM
<details>
<summary>LLVM.Transforms/SLPVectorizer/X86/propagate-mmra.ll</summary>
```
Exit Code: 1
Command Output (stdout):
--
# RUN: at line 2
/home/gha/actions-runner/_work/llvm-project/llvm-project/build/bin/opt -passes=slp-vectorizer -S -mtriple=x86_64-unknown-linux-gnu -slp-vectorize-hor < /home/gha/actions-runner/_work/llvm-project/llvm-project/llvm/test/Transforms/SLPVectorizer/X86/propagate-mmra.ll | /home/gha/actions-runner/_work/llvm-project/llvm-project/build/bin/FileCheck /home/gha/actions-runner/_work/llvm-project/llvm-project/llvm/test/Transforms/SLPVectorizer/X86/propagate-mmra.ll
# executed command: /home/gha/actions-runner/_work/llvm-project/llvm-project/build/bin/opt -passes=slp-vectorizer -S -mtriple=x86_64-unknown-linux-gnu -slp-vectorize-hor
# note: command had no output on stdout or stderr
# executed command: /home/gha/actions-runner/_work/llvm-project/llvm-project/build/bin/FileCheck /home/gha/actions-runner/_work/llvm-project/llvm-project/llvm/test/Transforms/SLPVectorizer/X86/propagate-mmra.ll
# .---command stderr------------
# | /home/gha/actions-runner/_work/llvm-project/llvm-project/llvm/test/Transforms/SLPVectorizer/X86/propagate-mmra.ll:8:15: error: CHECK-NEXT: is not on the line after the previous match
# | ; CHECK-NEXT: [[TMP1:%.*]] = insertelement <2 x float> poison, float [[TMP0]], i32 0
# | ^
# | <stdin>:10:2: note: 'next' match was here
# | %2 = insertelement <2 x float> poison, float %0, i32 0
# | ^
# | <stdin>:8:46: note: previous match ended here
# | %0 = load float, ptr null, align 4, !mmra !0
# | ^
# | <stdin>:9:1: note: non-matching line after previous match is here
# | %1 = fmul float %0, 0.000000e+00
# | ^
# |
# | Input file: <stdin>
# | Check file: /home/gha/actions-runner/_work/llvm-project/llvm-project/llvm/test/Transforms/SLPVectorizer/X86/propagate-mmra.ll
# |
# | -dump-input=help explains the following input dump.
# |
# | Input was:
# | <<<<<<
# | .
# | .
# | .
# | 5:
# | 6: define void @foo() {
# | 7: .preheader16.preheader:
# | 8: %0 = load float, ptr null, align 4, !mmra !0
# | 9: %1 = fmul float %0, 0.000000e+00
# | 10: %2 = insertelement <2 x float> poison, float %0, i32 0
# | next:8 !~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ error: match on wrong line
# | 11: %3 = insertelement <2 x float> %2, float %1, i32 1
# | 12: %4 = fadd <2 x float> %3, zeroinitializer
# | 13: %5 = select <2 x i1> zeroinitializer, <2 x float> %4, <2 x float> zeroinitializer
# | 14: store <2 x float> %5, ptr null, align 16
# | 15: ret void
# | 16: }
# | 17:
# | 18: !0 = !{!"scope", !"device"}
# | >>>>>>
# `-----------------------------
# error: command failed with exit status: 1
--
```
</details>
If these failures are unrelated to your changes (for example tests are broken or flaky at HEAD), please open an issue at https://github.com/llvm/llvm-project/issues and add the `infrastructure` label.
https://github.com/llvm/llvm-project/pull/173319
More information about the llvm-commits
mailing list