llvm/test/Transforms/LoopVectorize/VPlan/for-phi-ordering.ll

; NOTE: Assertions have been autogenerated by utils/update_analyze_test_checks.py UTC_ARGS: --filter-out-after "ir-bb<loop.header>:" --version 6
; RUN: opt %s -passes=loop-vectorize -force-vector-width=4 -force-vector-interleave=1 -vplan-print-after=printOptimizedVPlan -disable-output 2>&1 | FileCheck %s

define i32 @test(i32 %limit, i1 %guard, i32 %divisor) {
; CHECK-LABEL: VPlan for loop in 'test'
; 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:  vp<[[VP3:%[0-9]+]]> = original trip-count
; CHECK-EMPTY:
; CHECK-NEXT:  ir-bb<entry>:
; CHECK-NEXT:    EMIT vp<[[VP3]]> = EXPAND SCEV (1 + (0 smax %limit))<nuw>
; 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<[[VP4:%[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:      EMIT vp<%index.next> = add nuw vp<[[VP4]]>, 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:    WIDEN ir<%prev.iv> = add nsw ir<%iv>, ir<-1>
; CHECK-NEXT:    EMIT vp<[[VP6:%[0-9]+]]> = extract-last-part ir<%prev.iv>
; CHECK-NEXT:    EMIT vp<[[VP7:%[0-9]+]]> = extract-last-lane vp<[[VP6]]>
; CHECK-NEXT:    EMIT vp<[[VP8:%[0-9]+]]> = extract-last-part ir<0>
; CHECK-NEXT:    EMIT vp<[[VP9:%[0-9]+]]> = extract-last-lane vp<[[VP8]]>
; CHECK-NEXT:    EMIT vp<%cmp.n> = icmp eq vp<[[VP3]]>, 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   %sum.next.lcssa = phi i32 [ %sum.next, %loop.latch ] (extra operand: vp<[[VP9]]> from middle.block)
; CHECK-NEXT:  No successors
; CHECK-EMPTY:
; CHECK-NEXT:  scalar.ph:
; CHECK-NEXT:    EMIT-SCALAR vp<[[VP11:%[0-9]+]]> = phi [ vp<[[VP2]]>, middle.block ], [ ir<0>, ir-bb<entry> ]
; CHECK-NEXT:    EMIT-SCALAR vp<[[VP12:%[0-9]+]]> = phi [ vp<[[VP7]]>, middle.block ], [ ir<0>, ir-bb<entry> ]
; CHECK-NEXT:    EMIT-SCALAR vp<[[VP13:%[0-9]+]]> = phi [ vp<[[VP9]]>, middle.block ], [ ir<0>, ir-bb<entry> ]
; CHECK-NEXT:  Successor(s): ir-bb<loop.header>
; CHECK-EMPTY:
; CHECK-NEXT:  ir-bb<loop.header>:
;
entry:
  br label %loop.header

loop.header:
  %iv = phi i32 [ 0, %entry ], [ %iv.next, %loop.latch ]
  %rec = phi i32 [ 0, %entry ], [ %prev.iv, %loop.latch ]
  %sum = phi i32 [ 0, %entry ], [ %sum.next, %loop.latch ]

  ; move
  %conv = zext i32 %rec to i64

  %iv.nonzero = icmp ne i32 %iv, 0
  %or.cond = select i1 %iv.nonzero, i1 %guard, i1 false
  br label %loop.latch

loop.latch:
  %cond = phi i32 [ 0, %loop.header ]

  ; after
  %prev.iv = add nsw i32 %iv, -1
  %sum.next = add i32 %sum, %cond

  %iv.next = add nuw nsw i32 %iv, 1
  %cmp = icmp slt i32 %iv, %limit
  br i1 %cmp, label %loop.header, label %exit

exit:
  ret i32 %sum.next
}