|
| 1 | +; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 5 |
| 2 | +; RUN: opt -S --passes=slp-vectorizer -mtriple=x86_64-unknown-linux-gnu < %s | FileCheck %s |
| 3 | + |
| 4 | +define i32 @test() { |
| 5 | +; CHECK-LABEL: define i32 @test() { |
| 6 | +; CHECK-NEXT: [[ENTRY:.*:]] |
| 7 | +; CHECK-NEXT: [[TMP0:%.*]] = load <4 x i64>, ptr null, align 16 |
| 8 | +; CHECK-NEXT: [[TMP1:%.*]] = or i64 poison, 0 |
| 9 | +; CHECK-NEXT: [[TMP2:%.*]] = shufflevector <4 x i64> [[TMP0]], <4 x i64> poison, <8 x i32> <i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 1, i32 poison, i32 poison> |
| 10 | +; CHECK-NEXT: [[TMP3:%.*]] = shufflevector <4 x i64> [[TMP0]], <4 x i64> poison, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 poison, i32 poison, i32 poison, i32 poison> |
| 11 | +; CHECK-NEXT: [[TMP4:%.*]] = shufflevector <8 x i64> [[TMP3]], <8 x i64> <i64 poison, i64 poison, i64 poison, i64 poison, i64 0, i64 poison, i64 poison, i64 poison>, <8 x i32> <i32 poison, i32 poison, i32 poison, i32 poison, i32 12, i32 1, i32 poison, i32 poison> |
| 12 | +; CHECK-NEXT: [[TMP5:%.*]] = call <8 x i64> @llvm.vector.insert.v8i64.v4i64(<8 x i64> [[TMP4]], <4 x i64> [[TMP0]], i64 0) |
| 13 | +; CHECK-NEXT: [[TMP6:%.*]] = trunc <8 x i64> [[TMP5]] to <8 x i32> |
| 14 | +; CHECK-NEXT: [[TMP7:%.*]] = shufflevector <8 x i32> [[TMP6]], <8 x i32> poison, <16 x i32> <i32 0, i32 0, i32 0, i32 1, i32 1, i32 1, i32 2, i32 2, i32 2, i32 3, i32 3, i32 3, i32 4, i32 4, i32 5, i32 5> |
| 15 | +; CHECK-NEXT: [[TMP8:%.*]] = add <16 x i32> [[TMP7]], zeroinitializer |
| 16 | +; CHECK-NEXT: [[TMP9:%.*]] = extractelement <4 x i64> [[TMP0]], i32 0 |
| 17 | +; CHECK-NEXT: [[INC_3_3_I_1:%.*]] = or i64 [[TMP9]], 0 |
| 18 | +; CHECK-NEXT: [[TMP10:%.*]] = call i32 @llvm.vector.reduce.or.v16i32(<16 x i32> [[TMP8]]) |
| 19 | +; CHECK-NEXT: [[TMP11:%.*]] = call i32 @llvm.vector.reduce.or.v8i32(<8 x i32> poison) |
| 20 | +; CHECK-NEXT: [[OP_RDX:%.*]] = or i32 [[TMP10]], [[TMP11]] |
| 21 | +; CHECK-NEXT: ret i32 [[OP_RDX]] |
| 22 | +; |
| 23 | +entry: |
| 24 | + %.pre.i = load i64, ptr getelementptr inbounds nuw (i8, ptr null, i64 24), align 8 |
| 25 | + %.pre50.i = load i64, ptr getelementptr inbounds nuw (i8, ptr null, i64 16), align 16 |
| 26 | + %.pre51.i = load i64, ptr getelementptr inbounds nuw (i8, ptr null, i64 8), align 8 |
| 27 | + %.pre52.i = load i64, ptr null, align 16 |
| 28 | + %0 = or i64 %.pre51.i, 0 |
| 29 | + %1 = trunc i64 %.pre.i to i32 |
| 30 | + %2 = add i32 %1, 0 |
| 31 | + %3 = trunc i64 %.pre50.i to i32 |
| 32 | + %4 = add i32 %3, 0 |
| 33 | + %5 = trunc i64 %.pre51.i to i32 |
| 34 | + %6 = add i32 %5, 0 |
| 35 | + %7 = trunc i64 0 to i32 |
| 36 | + %8 = add i32 %5, 0 |
| 37 | + %9 = add i32 %7, 0 |
| 38 | + %10 = add i32 %1, 0 |
| 39 | + %11 = add i32 %3, 0 |
| 40 | + %12 = add i32 %5, 0 |
| 41 | + %13 = add i32 %7, 0 |
| 42 | + %14 = trunc i64 %.pre.i to i32 |
| 43 | + %15 = add i32 %14, 0 |
| 44 | + %16 = trunc i64 %.pre50.i to i32 |
| 45 | + %17 = add i32 %16, 0 |
| 46 | + %18 = trunc i64 %.pre51.i to i32 |
| 47 | + %19 = add i32 %18, 0 |
| 48 | + %20 = trunc i64 %.pre52.i to i32 |
| 49 | + %conv14.1.i = or i32 %9, %13 |
| 50 | + %21 = or i32 %conv14.1.i, %6 |
| 51 | + %22 = or i32 %21, %8 |
| 52 | + %23 = or i32 %22, %12 |
| 53 | + %24 = or i32 %23, %4 |
| 54 | + %25 = or i32 %24, %11 |
| 55 | + %26 = or i32 %25, %2 |
| 56 | + %27 = or i32 %26, %10 |
| 57 | + %28 = or i32 %27, %15 |
| 58 | + %29 = or i32 %28, %17 |
| 59 | + %30 = or i32 %29, %19 |
| 60 | + %31 = add i32 %14, 0 |
| 61 | + %32 = add i32 %16, 0 |
| 62 | + %33 = add i32 %18, 0 |
| 63 | + %34 = add i32 %20, 0 |
| 64 | + %35 = add i32 %14, 0 |
| 65 | + %36 = add i32 %16, 0 |
| 66 | + %37 = add i32 %18, 0 |
| 67 | + %38 = add i32 %20, 0 |
| 68 | + %39 = add i32 %14, 0 |
| 69 | + %40 = add i32 %16, 0 |
| 70 | + %41 = add i32 %18, 0 |
| 71 | + %42 = add i32 %20, 0 |
| 72 | + %inc.3.3.i.1 = or i64 %.pre52.i, 0 |
| 73 | + %conv14.i.1 = or i32 %38, %34 |
| 74 | + %conv14.1.i.1 = or i32 %conv14.i.1, %42 |
| 75 | + %conv14.3.i.1 = or i32 %conv14.1.i.1, %33 |
| 76 | + %conv14.145.i.1 = or i32 %conv14.3.i.1, %37 |
| 77 | + %conv14.1.1.i.1 = or i32 %conv14.145.i.1, %41 |
| 78 | + %conv14.3.1.i.1 = or i32 %conv14.1.1.i.1, %32 |
| 79 | + %conv14.247.i.1 = or i32 %conv14.3.1.i.1, %36 |
| 80 | + %conv14.1.2.i.1 = or i32 %conv14.247.i.1, %40 |
| 81 | + %conv14.3.2.i.1 = or i32 %conv14.1.2.i.1, %31 |
| 82 | + %conv14.349.i.1 = or i32 %conv14.3.2.i.1, %35 |
| 83 | + %conv14.1.3.i.1 = or i32 %conv14.349.i.1, %39 |
| 84 | + %conv14.3.3.i.1 = or i32 %conv14.1.3.i.1, %30 |
| 85 | + ret i32 %conv14.3.3.i.1 |
| 86 | +} |
0 commit comments