[llvm] deecdb9 - [SLP][NFC]Add extra test with alternate add/sub vectorization, NFC

via llvm-commits llvm-commits at lists.llvm.org
Mon Aug 3 07:51:46 PDT 2026


Author: Alexey Bataev
Date: 2026-08-03T10:51:42-04:00
New Revision: deecdb92e1600c47b0dad4a13730440c7dbe7c17

URL: https://github.com/llvm/llvm-project/commit/deecdb92e1600c47b0dad4a13730440c7dbe7c17
DIFF: https://github.com/llvm/llvm-project/commit/deecdb92e1600c47b0dad4a13730440c7dbe7c17.diff

LOG: [SLP][NFC]Add extra test with alternate add/sub vectorization, NFC



Reviewers: 

Pull Request: https://github.com/llvm/llvm-project/pull/213693

Added: 
    llvm/test/Transforms/SLPVectorizer/AArch64/alt-addsub-signflip.ll

Modified: 
    

Removed: 
    


################################################################################
diff  --git a/llvm/test/Transforms/SLPVectorizer/AArch64/alt-addsub-signflip.ll b/llvm/test/Transforms/SLPVectorizer/AArch64/alt-addsub-signflip.ll
new file mode 100644
index 0000000000000..d99f0a3925fcf
--- /dev/null
+++ b/llvm/test/Transforms/SLPVectorizer/AArch64/alt-addsub-signflip.ll
@@ -0,0 +1,58 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
+; RUN: opt < %s -passes=slp-vectorizer -S -mtriple=aarch64-unknown-linux-gnu | FileCheck %s
+
+; The sign-flip form of an add/sub alternate node negates the second
+; operand in the subtracted lanes, so the nsw/nuw flags of the original
+; scalar adds and subtracts do not carry over to the combined add.
+
+define <4 x i32> @addsub_nsw(<4 x i32> %a, <4 x i32> %b) {
+; CHECK-LABEL: @addsub_nsw(
+; CHECK-NEXT:    [[TMP1:%.*]] = add nsw <4 x i32> [[A:%.*]], [[B:%.*]]
+; CHECK-NEXT:    [[TMP3:%.*]] = sub nsw <4 x i32> [[A]], [[B]]
+; CHECK-NEXT:    [[TMP2:%.*]] = shufflevector <4 x i32> [[TMP1]], <4 x i32> [[TMP3]], <4 x i32> <i32 0, i32 5, i32 2, i32 7>
+; CHECK-NEXT:    ret <4 x i32> [[TMP2]]
+;
+  %a0 = extractelement <4 x i32> %a, i32 0
+  %a1 = extractelement <4 x i32> %a, i32 1
+  %a2 = extractelement <4 x i32> %a, i32 2
+  %a3 = extractelement <4 x i32> %a, i32 3
+  %b0 = extractelement <4 x i32> %b, i32 0
+  %b1 = extractelement <4 x i32> %b, i32 1
+  %b2 = extractelement <4 x i32> %b, i32 2
+  %b3 = extractelement <4 x i32> %b, i32 3
+  %s0 = add nsw i32 %a0, %b0
+  %s1 = sub nsw i32 %a1, %b1
+  %s2 = add nsw i32 %a2, %b2
+  %s3 = sub nsw i32 %a3, %b3
+  %r0 = insertelement <4 x i32> poison, i32 %s0, i32 0
+  %r1 = insertelement <4 x i32> %r0, i32 %s1, i32 1
+  %r2 = insertelement <4 x i32> %r1, i32 %s2, i32 2
+  %r3 = insertelement <4 x i32> %r2, i32 %s3, i32 3
+  ret <4 x i32> %r3
+}
+
+define <4 x i32> @addsub_nuw(<4 x i32> %a, <4 x i32> %b) {
+; CHECK-LABEL: @addsub_nuw(
+; CHECK-NEXT:    [[TMP1:%.*]] = add nuw <4 x i32> [[A:%.*]], [[B:%.*]]
+; CHECK-NEXT:    [[TMP3:%.*]] = sub nuw <4 x i32> [[A]], [[B]]
+; CHECK-NEXT:    [[TMP2:%.*]] = shufflevector <4 x i32> [[TMP1]], <4 x i32> [[TMP3]], <4 x i32> <i32 0, i32 5, i32 2, i32 7>
+; CHECK-NEXT:    ret <4 x i32> [[TMP2]]
+;
+  %a0 = extractelement <4 x i32> %a, i32 0
+  %a1 = extractelement <4 x i32> %a, i32 1
+  %a2 = extractelement <4 x i32> %a, i32 2
+  %a3 = extractelement <4 x i32> %a, i32 3
+  %b0 = extractelement <4 x i32> %b, i32 0
+  %b1 = extractelement <4 x i32> %b, i32 1
+  %b2 = extractelement <4 x i32> %b, i32 2
+  %b3 = extractelement <4 x i32> %b, i32 3
+  %s0 = add nuw i32 %a0, %b0
+  %s1 = sub nuw i32 %a1, %b1
+  %s2 = add nuw i32 %a2, %b2
+  %s3 = sub nuw i32 %a3, %b3
+  %r0 = insertelement <4 x i32> poison, i32 %s0, i32 0
+  %r1 = insertelement <4 x i32> %r0, i32 %s1, i32 1
+  %r2 = insertelement <4 x i32> %r1, i32 %s2, i32 2
+  %r3 = insertelement <4 x i32> %r2, i32 %s3, i32 3
+  ret <4 x i32> %r3
+}


        


More information about the llvm-commits mailing list