llvm/test/Transforms/LoopVectorize/vplan-printing-metadata.ll

; NOTE: Assertions have been autogenerated by utils/update_analyze_test_checks.py UTC_ARGS: --version 6
; REQUIRES: asserts

; RUN: opt -passes=loop-vectorize -debug-only=loop-vectorize -force-vector-interleave=1 -force-vector-width=4 -disable-output %s 2>&1 | FileCheck %s

define void @test_widen_metadata(ptr noalias %A, ptr noalias %B, i32 %n) {
; CHECK-LABEL: 'test_widen_metadata'
; CHECK:  VPlan 'Initial VPlan for VF={4},UF>=1' {
; CHECK-NEXT:  Live-in vp<[[VP0:%[0-9]+]]> = VF
; CHECK-NEXT:  Live-in vp<[[VP1:%[0-9]+]]> = VF * UF
; CHECK-NEXT:  Live-in vp<[[VP2:%[0-9]+]]> = vector-trip-count
; CHECK-NEXT:  Live-in ir<%n> = original trip-count
; CHECK-EMPTY:
; CHECK-NEXT:  ir-bb<entry>:
; CHECK-NEXT:  Successor(s): scalar.ph, vector.ph
; CHECK-EMPTY:
; CHECK-NEXT:  vector.ph:
; CHECK-NEXT:  Successor(s): vector loop
; CHECK-EMPTY:
; CHECK-NEXT:  <x1> vector loop: {
; CHECK-NEXT:    vector.body:
; CHECK-NEXT:      EMIT vp<[[VP3:%[0-9]+]]> = CANONICAL-INDUCTION ir<0>, vp<%index.next>
; CHECK-NEXT:      vp<[[VP4:%[0-9]+]]> = SCALAR-STEPS vp<[[VP3]]>, ir<1>, vp<[[VP0]]>
; CHECK-NEXT:      CLONE ir<%gep.A> = getelementptr inbounds ir<%A>, vp<[[VP4]]>
; CHECK-NEXT:      vp<[[VP5:%[0-9]+]]> = vector-pointer inbounds ir<%gep.A>
; CHECK-NEXT:      WIDEN ir<%lv> = load vp<[[VP5]]> (!tbaa !0)
; CHECK-NEXT:      WIDEN-CAST ir<%conv> = sitofp ir<%lv> to float (!fpmath !4)
; CHECK-NEXT:      WIDEN ir<%mul> = fmul ir<%conv>, ir<2.000000e+00> (!fpmath !4)
; CHECK-NEXT:      WIDEN-CAST ir<%conv.back> = fptosi ir<%mul> to i32
; CHECK-NEXT:      CLONE ir<%gep.B> = getelementptr inbounds ir<%B>, vp<[[VP4]]>
; CHECK-NEXT:      vp<[[VP6:%[0-9]+]]> = vector-pointer inbounds ir<%gep.B>
; CHECK-NEXT:      WIDEN store vp<[[VP6]]>, ir<%conv.back> (!tbaa !0)
; CHECK-NEXT:      EMIT vp<%index.next> = add nuw vp<[[VP3]]>, vp<[[VP1]]>
; CHECK-NEXT:      EMIT branch-on-count vp<%index.next>, vp<[[VP2]]>
; CHECK-NEXT:    No successors
; CHECK-NEXT:  }
; CHECK-NEXT:  Successor(s): middle.block
; CHECK-EMPTY:
; CHECK-NEXT:  middle.block:
; CHECK-NEXT:    EMIT vp<%cmp.n> = icmp eq ir<%n>, vp<[[VP2]]>
; CHECK-NEXT:    EMIT branch-on-cond vp<%cmp.n>
; CHECK-NEXT:  Successor(s): ir-bb<exit>, scalar.ph
; CHECK-EMPTY:
; CHECK-NEXT:  ir-bb<exit>:
; CHECK-NEXT:  No successors
; CHECK-EMPTY:
; CHECK-NEXT:  scalar.ph:
; CHECK-NEXT:    EMIT-SCALAR vp<%bc.resume.val> = phi [ vp<[[VP2]]>, middle.block ], [ ir<0>, ir-bb<entry> ]
; CHECK-NEXT:  Successor(s): ir-bb<loop>
; CHECK-EMPTY:
; CHECK-NEXT:  ir-bb<loop>:
; CHECK-NEXT:    IR   %i = phi i32 [ 0, %entry ], [ %i.next, %loop ] (extra operand: vp<%bc.resume.val> from scalar.ph)
; CHECK-NEXT:    IR   %gep.A = getelementptr inbounds i32, ptr %A, i32 %i
; CHECK-NEXT:    IR   %lv = load i32, ptr %gep.A, align 4, !tbaa !0, !range !3
; CHECK-NEXT:    IR   %conv = sitofp i32 %lv to float, !fpmath !4
; CHECK-NEXT:    IR   %mul = fmul float %conv, 2.000000e+00, !fpmath !4
; CHECK-NEXT:    IR   %conv.back = fptosi float %mul to i32
; CHECK-NEXT:    IR   %gep.B = getelementptr inbounds i32, ptr %B, i32 %i
; CHECK-NEXT:    IR   store i32 %conv.back, ptr %gep.B, align 4, !tbaa !0
; CHECK-NEXT:    IR   %i.next = add i32 %i, 1
; CHECK-NEXT:    IR   %cond = icmp eq i32 %i.next, %n
; CHECK-NEXT:  No successors
; CHECK-NEXT:  }
;
entry:
  br label %loop

loop:
  %i = phi i32 [ 0, %entry ], [ %i.next, %loop ]
  %gep.A = getelementptr inbounds i32, ptr %A, i32 %i
  %lv = load i32, ptr %gep.A, align 4, !tbaa !0, !range !6
  %conv = sitofp i32 %lv to float, !fpmath !5
  %mul = fmul float %conv, 2.0, !fpmath !5
  %conv.back = fptosi float %mul to i32
  %gep.B = getelementptr inbounds i32, ptr %B, i32 %i
  store i32 %conv.back, ptr %gep.B, align 4, !tbaa !0
  %i.next = add i32 %i, 1
  %cond = icmp eq i32 %i.next, %n
  br i1 %cond, label %exit, label %loop

exit:
  ret void
}

declare float @llvm.sqrt.f32(float)

define void @test_intrinsic_with_metadata(ptr noalias %A, ptr noalias %B, i32 %n) {
; CHECK-LABEL: 'test_intrinsic_with_metadata'
; CHECK:  VPlan 'Initial VPlan for VF={4},UF>=1' {
; CHECK-NEXT:  Live-in vp<[[VP0:%[0-9]+]]> = VF
; CHECK-NEXT:  Live-in vp<[[VP1:%[0-9]+]]> = VF * UF
; CHECK-NEXT:  Live-in vp<[[VP2:%[0-9]+]]> = vector-trip-count
; CHECK-NEXT:  Live-in ir<%n> = original trip-count
; CHECK-EMPTY:
; CHECK-NEXT:  ir-bb<entry>:
; CHECK-NEXT:  Successor(s): scalar.ph, vector.ph
; CHECK-EMPTY:
; CHECK-NEXT:  vector.ph:
; CHECK-NEXT:  Successor(s): vector loop
; CHECK-EMPTY:
; CHECK-NEXT:  <x1> vector loop: {
; CHECK-NEXT:    vector.body:
; CHECK-NEXT:      EMIT vp<[[VP3:%[0-9]+]]> = CANONICAL-INDUCTION ir<0>, vp<%index.next>
; CHECK-NEXT:      vp<[[VP4:%[0-9]+]]> = SCALAR-STEPS vp<[[VP3]]>, ir<1>, vp<[[VP0]]>
; CHECK-NEXT:      CLONE ir<%gep.A> = getelementptr inbounds ir<%A>, vp<[[VP4]]>
; CHECK-NEXT:      vp<[[VP5:%[0-9]+]]> = vector-pointer inbounds ir<%gep.A>
; CHECK-NEXT:      WIDEN ir<%lv> = load vp<[[VP5]]> (!tbaa !0)
; CHECK-NEXT:      WIDEN-INTRINSIC ir<%sqrt> = call llvm.sqrt(ir<%lv>) (!fpmath !3)
; CHECK-NEXT:      CLONE ir<%gep.B> = getelementptr inbounds ir<%B>, vp<[[VP4]]>
; CHECK-NEXT:      vp<[[VP6:%[0-9]+]]> = vector-pointer inbounds ir<%gep.B>
; CHECK-NEXT:      WIDEN store vp<[[VP6]]>, ir<%sqrt> (!tbaa !0)
; CHECK-NEXT:      EMIT vp<%index.next> = add nuw vp<[[VP3]]>, vp<[[VP1]]>
; CHECK-NEXT:      EMIT branch-on-count vp<%index.next>, vp<[[VP2]]>
; CHECK-NEXT:    No successors
; CHECK-NEXT:  }
; CHECK-NEXT:  Successor(s): middle.block
; CHECK-EMPTY:
; CHECK-NEXT:  middle.block:
; CHECK-NEXT:    EMIT vp<%cmp.n> = icmp eq ir<%n>, vp<[[VP2]]>
; CHECK-NEXT:    EMIT branch-on-cond vp<%cmp.n>
; CHECK-NEXT:  Successor(s): ir-bb<exit>, scalar.ph
; CHECK-EMPTY:
; CHECK-NEXT:  ir-bb<exit>:
; CHECK-NEXT:  No successors
; CHECK-EMPTY:
; CHECK-NEXT:  scalar.ph:
; CHECK-NEXT:    EMIT-SCALAR vp<%bc.resume.val> = phi [ vp<[[VP2]]>, middle.block ], [ ir<0>, ir-bb<entry> ]
; CHECK-NEXT:  Successor(s): ir-bb<loop>
; CHECK-EMPTY:
; CHECK-NEXT:  ir-bb<loop>:
; CHECK-NEXT:    IR   %i = phi i32 [ 0, %entry ], [ %i.next, %loop ] (extra operand: vp<%bc.resume.val> from scalar.ph)
; CHECK-NEXT:    IR   %gep.A = getelementptr inbounds float, ptr %A, i32 %i
; CHECK-NEXT:    IR   %lv = load float, ptr %gep.A, align 4, !tbaa !0
; CHECK-NEXT:    IR   %sqrt = call float @llvm.sqrt.f32(float %lv), !fpmath !3
; CHECK-NEXT:    IR   %gep.B = getelementptr inbounds float, ptr %B, i32 %i
; CHECK-NEXT:    IR   store float %sqrt, ptr %gep.B, align 4, !tbaa !0
; CHECK-NEXT:    IR   %i.next = add i32 %i, 1
; CHECK-NEXT:    IR   %cond = icmp eq i32 %i.next, %n
; CHECK-NEXT:  No successors
; CHECK-NEXT:  }
;
entry:
  br label %loop

loop:
  %i = phi i32 [ 0, %entry ], [ %i.next, %loop ]
  %gep.A = getelementptr inbounds float, ptr %A, i32 %i
  %lv = load float, ptr %gep.A, align 4, !tbaa !0
  %sqrt = call float @llvm.sqrt.f32(float %lv), !fpmath !5
  %gep.B = getelementptr inbounds float, ptr %B, i32 %i
  store float %sqrt, ptr %gep.B, align 4, !tbaa !0
  %i.next = add i32 %i, 1
  %cond = icmp eq i32 %i.next, %n
  br i1 %cond, label %exit, label %loop

exit:
  ret void
}

define void @test_widen_with_multiple_metadata(ptr noalias %A, ptr noalias %B, i32 %n) {
; CHECK-LABEL: 'test_widen_with_multiple_metadata'
; CHECK:  VPlan 'Initial VPlan for VF={4},UF>=1' {
; CHECK-NEXT:  Live-in vp<[[VP0:%[0-9]+]]> = VF
; CHECK-NEXT:  Live-in vp<[[VP1:%[0-9]+]]> = VF * UF
; CHECK-NEXT:  Live-in vp<[[VP2:%[0-9]+]]> = vector-trip-count
; CHECK-NEXT:  Live-in ir<%n> = original trip-count
; CHECK-EMPTY:
; CHECK-NEXT:  ir-bb<entry>:
; CHECK-NEXT:  Successor(s): scalar.ph, vector.ph
; CHECK-EMPTY:
; CHECK-NEXT:  vector.ph:
; CHECK-NEXT:  Successor(s): vector loop
; CHECK-EMPTY:
; CHECK-NEXT:  <x1> vector loop: {
; CHECK-NEXT:    vector.body:
; CHECK-NEXT:      EMIT vp<[[VP3:%[0-9]+]]> = CANONICAL-INDUCTION ir<0>, vp<%index.next>
; CHECK-NEXT:      vp<[[VP4:%[0-9]+]]> = SCALAR-STEPS vp<[[VP3]]>, ir<1>, vp<[[VP0]]>
; CHECK-NEXT:      CLONE ir<%gep.A> = getelementptr inbounds ir<%A>, vp<[[VP4]]>
; CHECK-NEXT:      vp<[[VP5:%[0-9]+]]> = vector-pointer inbounds ir<%gep.A>
; CHECK-NEXT:      WIDEN ir<%lv> = load vp<[[VP5]]> (!tbaa !0)
; CHECK-NEXT:      WIDEN-CAST ir<%conv> = sitofp ir<%lv> to float
; CHECK-NEXT:      WIDEN ir<%mul> = fmul ir<%conv>, ir<2.000000e+00>
; CHECK-NEXT:      WIDEN-CAST ir<%conv.back> = fptosi ir<%mul> to i32
; CHECK-NEXT:      CLONE ir<%gep.B> = getelementptr inbounds ir<%B>, vp<[[VP4]]>
; CHECK-NEXT:      vp<[[VP6:%[0-9]+]]> = vector-pointer inbounds ir<%gep.B>
; CHECK-NEXT:      WIDEN store vp<[[VP6]]>, ir<%conv.back> (!tbaa !0)
; CHECK-NEXT:      EMIT vp<%index.next> = add nuw vp<[[VP3]]>, vp<[[VP1]]>
; CHECK-NEXT:      EMIT branch-on-count vp<%index.next>, vp<[[VP2]]>
; CHECK-NEXT:    No successors
; CHECK-NEXT:  }
; CHECK-NEXT:  Successor(s): middle.block
; CHECK-EMPTY:
; CHECK-NEXT:  middle.block:
; CHECK-NEXT:    EMIT vp<%cmp.n> = icmp eq ir<%n>, vp<[[VP2]]>
; CHECK-NEXT:    EMIT branch-on-cond vp<%cmp.n>
; CHECK-NEXT:  Successor(s): ir-bb<exit>, scalar.ph
; CHECK-EMPTY:
; CHECK-NEXT:  ir-bb<exit>:
; CHECK-NEXT:  No successors
; CHECK-EMPTY:
; CHECK-NEXT:  scalar.ph:
; CHECK-NEXT:    EMIT-SCALAR vp<%bc.resume.val> = phi [ vp<[[VP2]]>, middle.block ], [ ir<0>, ir-bb<entry> ]
; CHECK-NEXT:  Successor(s): ir-bb<loop>
; CHECK-EMPTY:
; CHECK-NEXT:  ir-bb<loop>:
; CHECK-NEXT:    IR   %i = phi i32 [ 0, %entry ], [ %i.next, %loop ] (extra operand: vp<%bc.resume.val> from scalar.ph)
; CHECK-NEXT:    IR   %gep.A = getelementptr inbounds i32, ptr %A, i32 %i
; CHECK-NEXT:    IR   %lv = load i32, ptr %gep.A, align 4, !tbaa !0, !range !3
; CHECK-NEXT:    IR   %conv = sitofp i32 %lv to float
; CHECK-NEXT:    IR   %mul = fmul float %conv, 2.000000e+00
; CHECK-NEXT:    IR   %conv.back = fptosi float %mul to i32
; CHECK-NEXT:    IR   %gep.B = getelementptr inbounds i32, ptr %B, i32 %i
; CHECK-NEXT:    IR   store i32 %conv.back, ptr %gep.B, align 4, !tbaa !0
; CHECK-NEXT:    IR   %i.next = add i32 %i, 1
; CHECK-NEXT:    IR   %cond = icmp eq i32 %i.next, %n
; CHECK-NEXT:  No successors
; CHECK-NEXT:  }
;
entry:
  br label %loop

loop:
  %i = phi i32 [ 0, %entry ], [ %i.next, %loop ]
  %gep.A = getelementptr inbounds i32, ptr %A, i32 %i
  %lv = load i32, ptr %gep.A, align 4, !tbaa !0, !range !6
  %conv = sitofp i32 %lv to float
  %mul = fmul float %conv, 2.0
  %conv.back = fptosi float %mul to i32
  %gep.B = getelementptr inbounds i32, ptr %B, i32 %i
  store i32 %conv.back, ptr %gep.B, align 4, !tbaa !0
  %i.next = add i32 %i, 1
  %cond = icmp eq i32 %i.next, %n
  br i1 %cond, label %exit, label %loop

exit:
  ret void
}

!0 = !{!1, !1, i64 0}
!1 = !{!"float", !2}
!2 = !{!"root"}
!5 = !{float 2.500000e+00}
!6 = !{i32 0, i32 100}