llvm/test/Transforms/SLPVectorizer/X86/broadcast.ll

; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
; RUN: opt -passes=slp-vectorizer -S -mtriple=x86_64-unknown-linux -mcpu=corei7-avx -slp-threshold=-999 < %s | FileCheck %s


; S[0] = %v1 + %v2
; S[1] = %v2 + %v1
; S[2] = %v2 + %v1
; S[3] = %v1 + %v2
;
; We broadcast %v1 and %v2
;

define void @bcast_vals(ptr %A, ptr %B, ptr %S) {
; CHECK-LABEL: @bcast_vals(
; CHECK-NEXT:  entry:
; CHECK-NEXT:    [[A0:%.*]] = load i64, ptr [[A:%.*]], align 8
; CHECK-NEXT:    [[B0:%.*]] = load i64, ptr [[B:%.*]], align 8
; CHECK-NEXT:    [[V1:%.*]] = sub i64 [[A0]], 1
; CHECK-NEXT:    [[V2:%.*]] = sub i64 [[B0]], 1
; CHECK-NEXT:    [[TMP0:%.*]] = insertelement <4 x i64> poison, i64 [[V1]], i32 0
; CHECK-NEXT:    [[TMP1:%.*]] = shufflevector <4 x i64> [[TMP0]], <4 x i64> poison, <4 x i32> zeroinitializer
; CHECK-NEXT:    [[TMP2:%.*]] = insertelement <4 x i64> poison, i64 [[V2]], i32 0
; CHECK-NEXT:    [[TMP3:%.*]] = shufflevector <4 x i64> [[TMP2]], <4 x i64> poison, <4 x i32> zeroinitializer
; CHECK-NEXT:    [[TMP4:%.*]] = add <4 x i64> [[TMP1]], [[TMP3]]
; CHECK-NEXT:    store <4 x i64> [[TMP4]], ptr [[S:%.*]], align 8
; CHECK-NEXT:    ret void
;
entry:
  %A0 = load i64, ptr %A, align 8
  %B0 = load i64, ptr %B, align 8

  %v1 = sub i64 %A0, 1
  %v2 = sub i64 %B0, 1

  %Add0 = add i64 %v1, %v2
  %Add1 = add i64 %v2, %v1
  %Add2 = add i64 %v2, %v1
  %Add3 = add i64 %v1, %v2

  %idxS1 = getelementptr inbounds i64, ptr %S, i64 1
  %idxS2 = getelementptr inbounds i64, ptr %S, i64 2
  %idxS3 = getelementptr inbounds i64, ptr %S, i64 3

  store i64 %Add0, ptr %S, align 8
  store i64 %Add1, ptr %idxS1, align 8
  store i64 %Add2, ptr %idxS2, align 8
  store i64 %Add3, ptr %idxS3, align 8
  ret void
}

; S[0] = %v1 + %v2
; S[1] = %v3 + %v1
; S[2] = %v5 + %v1
; S[3] = %v1 + %v4
;
; We broadcast %v1.

;
define void @bcast_vals2(ptr %A, ptr %B, ptr %C, ptr %D, ptr %E, ptr %S) {
; CHECK-LABEL: @bcast_vals2(
; CHECK-NEXT:  entry:
; CHECK-NEXT:    [[A0:%.*]] = load i16, ptr [[A:%.*]], align 8
; CHECK-NEXT:    [[B0:%.*]] = load i16, ptr [[B:%.*]], align 8
; CHECK-NEXT:    [[C0:%.*]] = load i16, ptr [[C:%.*]], align 8
; CHECK-NEXT:    [[D0:%.*]] = load i16, ptr [[D:%.*]], align 8
; CHECK-NEXT:    [[E0:%.*]] = load i16, ptr [[E:%.*]], align 8
; CHECK-NEXT:    [[V1:%.*]] = sext i16 [[A0]] to i32
; CHECK-NEXT:    [[TMP0:%.*]] = insertelement <4 x i16> poison, i16 [[B0]], i32 0
; CHECK-NEXT:    [[TMP1:%.*]] = insertelement <4 x i16> [[TMP0]], i16 [[C0]], i32 1
; CHECK-NEXT:    [[TMP2:%.*]] = insertelement <4 x i16> [[TMP1]], i16 [[E0]], i32 2
; CHECK-NEXT:    [[TMP3:%.*]] = insertelement <4 x i16> [[TMP2]], i16 [[D0]], i32 3
; CHECK-NEXT:    [[TMP4:%.*]] = sext <4 x i16> [[TMP3]] to <4 x i32>
; CHECK-NEXT:    [[TMP5:%.*]] = insertelement <4 x i32> poison, i32 [[V1]], i32 0
; CHECK-NEXT:    [[TMP6:%.*]] = shufflevector <4 x i32> [[TMP5]], <4 x i32> poison, <4 x i32> zeroinitializer
; CHECK-NEXT:    [[TMP7:%.*]] = add <4 x i32> [[TMP6]], [[TMP4]]
; CHECK-NEXT:    store <4 x i32> [[TMP7]], ptr [[S:%.*]], align 8
; CHECK-NEXT:    ret void
;
entry:
  %A0 = load i16, ptr %A, align 8
  %B0 = load i16, ptr %B, align 8
  %C0 = load i16, ptr %C, align 8
  %D0 = load i16, ptr %D, align 8
  %E0 = load i16, ptr %E, align 8

  %v1 = sext i16 %A0 to i32
  %v2 = sext i16 %B0 to i32
  %v3 = sext i16 %C0 to i32
  %v4 = sext i16 %D0 to i32
  %v5 = sext i16 %E0 to i32

  %Add0 = add i32 %v1, %v2
  %Add1 = add i32 %v3, %v1
  %Add2 = add i32 %v5, %v1
  %Add3 = add i32 %v1, %v4

  %idxS1 = getelementptr inbounds i32, ptr %S, i64 1
  %idxS2 = getelementptr inbounds i32, ptr %S, i64 2
  %idxS3 = getelementptr inbounds i32, ptr %S, i64 3

  store i32 %Add0, ptr %S, align 8
  store i32 %Add1, ptr %idxS1, align 8
  store i32 %Add2, ptr %idxS2, align 8
  store i32 %Add3, ptr %idxS3, align 8
  ret void
}