[llvm] f3a0ad2 - [LSR] Add test for #62660.

[Florian Hahn via llvm-commits]

Author: Florian Hahn
Date: 2023-06-17T17:37:25+01:00
New Revision: f3a0ad2d8bf65993851598cc9961c1868371abfd

URL: https://github.com/llvm/llvm-project/commit/f3a0ad2d8bf65993851598cc9961c1868371abfd
DIFF: https://github.com/llvm/llvm-project/commit/f3a0ad2d8bf65993851598cc9961c1868371abfd.diff

LOG: [LSR] Add test for #62660.

Add test for LSR miscompile.

Added: 
    llvm/test/Transforms/LoopStrengthReduce/X86/pr62660-normalization-failure.ll

Modified: 
    

Removed: 
    


################################################################################
diff  --git a/llvm/test/Transforms/LoopStrengthReduce/X86/pr62660-normalization-failure.ll b/llvm/test/Transforms/LoopStrengthReduce/X86/pr62660-normalization-failure.ll
new file mode 100644
index 0000000000000..3e59dcfdf9250
--- /dev/null
+++ b/llvm/test/Transforms/LoopStrengthReduce/X86/pr62660-normalization-failure.ll
@@ -0,0 +1,36 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 2
+; RUN: opt -loop-reduce -S %s | FileCheck %s
+
+target datalayout = "e-m:e-p270:32:32-p271:32:32-p272:64:64-i64:64-f80:128-n8:16:32:64-S128"
+target triple = "x86_64-apple-macos"
+
+define i64 @test_pr62660() {
+; CHECK-LABEL: define i64 @test_pr62660() {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    br label [[LOOP:%.*]]
+; CHECK:       loop:
+; CHECK-NEXT:    [[LSR_IV1:%.*]] = phi i32 [ [[LSR_IV_NEXT2:%.*]], [[LOOP]] ], [ 65533, [[ENTRY:%.*]] ]
+; CHECK-NEXT:    [[LSR_IV:%.*]] = phi i64 [ [[LSR_IV_NEXT:%.*]], [[LOOP]] ], [ -1, [[ENTRY]] ]
+; CHECK-NEXT:    [[LSR_IV_NEXT]] = add nsw i64 [[LSR_IV]], 1
+; CHECK-NEXT:    [[LSR_IV_NEXT2]] = add nuw nsw i32 [[LSR_IV1]], 2
+; CHECK-NEXT:    [[CMP:%.*]] = icmp sgt i32 [[LSR_IV_NEXT2]], 8
+; CHECK-NEXT:    br i1 [[CMP]], label [[LOOP]], label [[EXIT:%.*]]
+; CHECK:       exit:
+; CHECK-NEXT:    ret i64 [[LSR_IV_NEXT]]
+;
+entry:
+  br label %loop
+
+loop:
+  %iv = phi i32 [ 0, %entry ], [ %iv.next, %loop ]
+  %conv1 = and i32 %iv, 65535
+  %add = add nsw i32 %iv, -1
+  %sub = add i32 %add, %conv1
+  %cmp = icmp sgt i32 %sub, 8
+  %iv.next = add nuw nsw i32 %iv, 1
+  br i1 %cmp, label %loop, label %exit
+
+exit:
+  %conv5 = zext i32 %iv to i64
+  ret i64 %conv5
+}