llvm/test/Transforms/SLPVectorizer/X86/multi-node-user-with-copyable-ops.ll

; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 6
; RUN: opt -S --passes=slp-vectorizer -mtriple=x86_64-unknown-linux-gnu < %s | FileCheck %s

define i1 @test(double %circ_radius, ptr %x, double %0) {
; CHECK-LABEL: define i1 @test(
; CHECK-SAME: double [[CIRC_RADIUS:%.*]], ptr [[X:%.*]], double [[TMP0:%.*]]) {
; CHECK-NEXT:  [[ENTRY:.*:]]
; CHECK-NEXT:    [[TMP1:%.*]] = load double, ptr [[X]], align 8
; CHECK-NEXT:    [[TMP2:%.*]] = insertelement <4 x double> poison, double [[TMP0]], i64 1
; CHECK-NEXT:    [[TMP3:%.*]] = insertelement <4 x double> [[TMP2]], double [[TMP1]], i64 0
; CHECK-NEXT:    [[TMP4:%.*]] = shufflevector <4 x double> [[TMP3]], <4 x double> poison, <4 x i32> <i32 0, i32 0, i32 0, i32 1>
; CHECK-NEXT:    [[TMP5:%.*]] = fadd <4 x double> [[TMP4]], <double -0.000000e+00, double -0.000000e+00, double 0.000000e+00, double -0.000000e+00>
; CHECK-NEXT:    [[TMP6:%.*]] = shufflevector <4 x double> [[TMP3]], <4 x double> <double 0.000000e+00, double poison, double 0.000000e+00, double 0.000000e+00>, <4 x i32> <i32 4, i32 0, i32 6, i32 7>
; CHECK-NEXT:    [[TMP7:%.*]] = fmul <4 x double> [[TMP5]], [[TMP6]]
; CHECK-NEXT:    [[TMP8:%.*]] = shufflevector <4 x double> [[TMP7]], <4 x double> poison, <4 x i32> <i32 1, i32 2, i32 poison, i32 0>
; CHECK-NEXT:    [[TMP9:%.*]] = insertelement <4 x double> poison, double [[CIRC_RADIUS]], i64 0
; CHECK-NEXT:    [[TMP10:%.*]] = shufflevector <4 x double> [[TMP9]], <4 x double> poison, <4 x i32> <i32 poison, i32 poison, i32 0, i32 poison>
; CHECK-NEXT:    [[TMP11:%.*]] = shufflevector <4 x double> [[TMP8]], <4 x double> [[TMP10]], <4 x i32> <i32 0, i32 1, i32 6, i32 3>
; CHECK-NEXT:    [[TMP12:%.*]] = fadd <4 x double> [[TMP7]], [[TMP11]]
; CHECK-NEXT:    [[TMP13:%.*]] = call <4 x double> @llvm.sqrt.v4f64(<4 x double> [[TMP12]])
; CHECK-NEXT:    [[TMP14:%.*]] = fcmp olt <4 x double> [[TMP13]], splat (double 1.000000e+00)
; CHECK-NEXT:    [[TMP15:%.*]] = call i1 @llvm.vector.reduce.or.v4i1(<4 x i1> [[TMP14]])
; CHECK-NEXT:    ret i1 [[TMP15]]
;
entry:
  %1 = load double, ptr %x, align 8
  %square = fmul double %1, 0.000000e+00
  %square105 = fmul double %1, %1
  %add = fadd double %square, %square105
  %sqrt116 = call double @llvm.sqrt.f64(double %add)
  %add20 = fadd double %1, 0.000000e+00
  %square106 = fmul double %add20, 0.000000e+00
  %add25 = fadd double %square105, %square106
  %sqrt115 = call double @llvm.sqrt.f64(double %add25)
  %square109 = fmul double %0, 0.000000e+00
  %add39 = fadd double %square106, %circ_radius
  %sqrt114 = call double @llvm.sqrt.f64(double %add39)
  %add50 = fadd double %square, %square109
  %sqrt = call double @llvm.sqrt.f64(double %add50)
  %cmp52 = fcmp olt double %sqrt116, 1.000000e+00
  %cmp54 = fcmp olt double %sqrt115, 1.000000e+00
  %or.cond = or i1 %cmp52, %cmp54
  %cmp57 = fcmp olt double %sqrt114, 1.000000e+00
  %or.cond112 = or i1 %or.cond, %cmp57
  %cmp60 = fcmp olt double %sqrt, 1.000000e+00
  %or.cond113 = or i1 %or.cond112, %cmp60
  ret i1 %or.cond113
}

declare double @llvm.sqrt.f64(double) #0