llvm/test/Transforms/LoopVectorize/VPlan/find-last.ll

; NOTE: Assertions have been autogenerated by utils/update_analyze_test_checks.py UTC_ARGS: --version 6
; RUN: opt -disable-output %s -p loop-vectorize -force-vector-width=4 -vplan-print-after=optimizeFindIVReductions 2>&1 | FileCheck %s

define i32 @find_last_with_select(ptr noalias %a, ptr noalias %b) {
; CHECK-LABEL: VPlan for loop in 'find_last_with_select'
; CHECK:  VPlan ' for 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<500> = 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:  vp<[[VP3:%[0-9]+]]> = CANONICAL-IV
; CHECK-EMPTY:
; CHECK-NEXT:    vector.body:
; CHECK-NEXT:      ir<%iv> = WIDEN-INDUCTION nuw nsw ir<0>, ir<1>, vp<[[VP0]]>
; CHECK-NEXT:      WIDEN-REDUCTION-PHI ir<%rdx> = phi (find-last) ir<1>, ir<%sel>
; CHECK-NEXT:      CLONE ir<%gep.a> = getelementptr inbounds ir<%a>, ir<%iv>
; CHECK-NEXT:      vp<[[VP4:%[0-9]+]]> = vector-pointer inbounds i32, ir<%gep.a>, ir<1>
; CHECK-NEXT:      WIDEN ir<%load.a> = load vp<[[VP4]]>
; CHECK-NEXT:      CLONE ir<%gep.b> = getelementptr inbounds ir<%b>, ir<%iv>
; CHECK-NEXT:      vp<[[VP5:%[0-9]+]]> = vector-pointer inbounds i32, ir<%gep.b>, ir<1>
; CHECK-NEXT:      WIDEN ir<%load.b> = load vp<[[VP5]]>
; CHECK-NEXT:      WIDEN ir<%cmp> = icmp slt ir<%load.a>, ir<%load.b>
; CHECK-NEXT:      WIDEN ir<%sel> = select ir<%cmp>, ir<%rdx>, ir<%load.a>
; CHECK-NEXT:      CLONE ir<%iv.next> = add nuw nsw ir<%iv>, ir<1>
; CHECK-NEXT:      CLONE ir<%exitcond> = icmp eq ir<%iv.next>, ir<500>
; 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<[[VP7:%[0-9]+]]> = compute-reduction-result (find-last) ir<%sel>
; CHECK-NEXT:    EMIT vp<[[VP8:%[0-9]+]]> = exiting-iv-value ir<%iv>
; CHECK-NEXT:    EMIT vp<[[VP9:%[0-9]+]]> = extract-last-part vp<[[VP7]]>
; CHECK-NEXT:    EMIT vp<[[VP10:%[0-9]+]]> = extract-last-lane vp<[[VP9]]>
; CHECK-NEXT:    EMIT vp<[[VP11:%[0-9]+]]> = extract-last-part vp<[[VP7]]>
; CHECK-NEXT:    EMIT vp<[[VP12:%[0-9]+]]> = extract-last-lane vp<[[VP11]]>
; CHECK-NEXT:    EMIT vp<%cmp.n> = icmp eq ir<500>, 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:    IR   %sel.lcssa = phi i32 [ %sel, %loop ] (extra operand: vp<[[VP7]]> from middle.block)
; CHECK-NEXT:  No successors
; CHECK-EMPTY:
; CHECK-NEXT:  scalar.ph:
; CHECK-NEXT:    EMIT-SCALAR vp<%bc.resume.val> = phi [ vp<[[VP8]]>, middle.block ], [ ir<0>, ir-bb<entry> ]
; CHECK-NEXT:    EMIT-SCALAR vp<%bc.merge.rdx> = phi [ vp<[[VP7]]>, middle.block ], [ ir<1>, ir-bb<entry> ]
; CHECK-NEXT:  Successor(s): ir-bb<loop>
; CHECK-EMPTY:
; CHECK-NEXT:  ir-bb<loop>:
; CHECK-NEXT:    IR   %iv = phi i64 [ 0, %entry ], [ %iv.next, %loop ] (extra operand: vp<%bc.resume.val> from scalar.ph)
; CHECK-NEXT:    IR   %rdx = phi i32 [ 1, %entry ], [ %sel, %loop ] (extra operand: vp<%bc.merge.rdx> from scalar.ph)
; CHECK-NEXT:    IR   %gep.a = getelementptr inbounds i32, ptr %a, i64 %iv
; CHECK-NEXT:    IR   %load.a = load i32, ptr %gep.a, align 4
; CHECK-NEXT:    IR   %gep.b = getelementptr inbounds i32, ptr %b, i64 %iv
; CHECK-NEXT:    IR   %load.b = load i32, ptr %gep.b, align 4
; CHECK-NEXT:    IR   %cmp = icmp slt i32 %load.a, %load.b
; CHECK-NEXT:    IR   %sel = select i1 %cmp, i32 %rdx, i32 %load.a
; CHECK-NEXT:    IR   %iv.next = add nuw nsw i64 %iv, 1
; CHECK-NEXT:    IR   %exitcond = icmp eq i64 %iv.next, 500
; CHECK-NEXT:  No successors
; CHECK-NEXT:  }
;
entry:
  br label %loop

loop:
  %iv = phi i64 [ 0, %entry ], [ %iv.next, %loop ]
  %rdx = phi i32 [ 1, %entry ], [ %sel, %loop ]
  %gep.a = getelementptr inbounds i32, ptr %a, i64 %iv
  %load.a = load i32, ptr %gep.a, align 4
  %gep.b = getelementptr inbounds i32, ptr %b, i64 %iv
  %load.b = load i32, ptr %gep.b, align 4
  %cmp = icmp slt i32 %load.a, %load.b
  %sel = select i1 %cmp, i32 %rdx, i32 %load.a
  %iv.next = add nuw nsw i64 %iv, 1
  %exitcond = icmp eq i64 %iv.next, 500
  br i1 %exitcond, label %exit, label %loop

exit:
  ret i32 %sel
}