test/Transforms/LoopVectorize/if-conversion.ll

; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --check-globals none --version 5
; RUN: opt < %s -passes=loop-vectorize,dce,instcombine -force-vector-interleave=1 -force-vector-width=4 -enable-if-conversion -S | FileCheck %s

target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64-S128"

; This is the loop in this example:
;
;int function0(int *a, int *b, int start, int end) {
;
;  for (int i=start; i<end; ++i) {
;    unsigned k = a[i];
;
;    if (a[i] > b[i])   <------ notice the IF inside the loop.
;      k = k * 5 + 3;
;
;    a[i] = k;  <---- K is a phi node that becomes vector-select.
;  }
;}

define i32 @function0(ptr nocapture %a, ptr nocapture %b, i32 %start, i32 %end) nounwind uwtable ssp {
; CHECK-LABEL: define i32 @function0(
; CHECK-SAME: ptr captures(none) [[A:%.*]], ptr captures(none) [[B:%.*]], i32 [[START:%.*]], i32 [[END:%.*]]) #[[ATTR0:[0-9]+]] {
; CHECK-NEXT:  [[ENTRY:.*:]]
; CHECK-NEXT:    [[CMP16:%.*]] = icmp slt i32 [[START]], [[END]]
; CHECK-NEXT:    br i1 [[CMP16]], label %[[FOR_BODY_LR_PH:.*]], label %[[FOR_END:.*]]
; CHECK:       [[FOR_BODY_LR_PH]]:
; CHECK-NEXT:    [[TMP0:%.*]] = sext i32 [[START]] to i64
; CHECK-NEXT:    [[TMP1:%.*]] = xor i32 [[START]], -1
; CHECK-NEXT:    [[TMP2:%.*]] = add i32 [[END]], [[TMP1]]
; CHECK-NEXT:    [[TMP3:%.*]] = zext i32 [[TMP2]] to i64
; CHECK-NEXT:    [[TMP4:%.*]] = add nuw nsw i64 [[TMP3]], 1
; CHECK-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i32 [[TMP2]], 3
; CHECK-NEXT:    br i1 [[MIN_ITERS_CHECK]], label %[[SCALAR_PH:.*]], label %[[VECTOR_MEMCHECK:.*]]
; CHECK:       [[VECTOR_MEMCHECK]]:
; CHECK-NEXT:    [[TMP5:%.*]] = shl nsw i64 [[TMP0]], 2
; CHECK-NEXT:    [[SCEVGEP:%.*]] = getelementptr i8, ptr [[A]], i64 [[TMP5]]
; CHECK-NEXT:    [[TMP6:%.*]] = xor i32 [[START]], -1
; CHECK-NEXT:    [[TMP7:%.*]] = add i32 [[END]], [[TMP6]]
; CHECK-NEXT:    [[TMP8:%.*]] = zext i32 [[TMP7]] to i64
; CHECK-NEXT:    [[TMP9:%.*]] = add nsw i64 [[TMP0]], [[TMP8]]
; CHECK-NEXT:    [[TMP10:%.*]] = shl nsw i64 [[TMP9]], 2
; CHECK-NEXT:    [[TMP11:%.*]] = add nsw i64 [[TMP10]], 4
; CHECK-NEXT:    [[SCEVGEP1:%.*]] = getelementptr i8, ptr [[A]], i64 [[TMP11]]
; CHECK-NEXT:    [[SCEVGEP2:%.*]] = getelementptr i8, ptr [[B]], i64 [[TMP5]]
; CHECK-NEXT:    [[SCEVGEP3:%.*]] = getelementptr i8, ptr [[B]], i64 [[TMP11]]
; CHECK-NEXT:    [[BOUND0:%.*]] = icmp ult ptr [[SCEVGEP]], [[SCEVGEP3]]
; CHECK-NEXT:    [[BOUND1:%.*]] = icmp ult ptr [[SCEVGEP2]], [[SCEVGEP1]]
; CHECK-NEXT:    [[FOUND_CONFLICT:%.*]] = and i1 [[BOUND0]], [[BOUND1]]
; CHECK-NEXT:    br i1 [[FOUND_CONFLICT]], label %[[SCALAR_PH]], label %[[VECTOR_PH:.*]]
; CHECK:       [[VECTOR_PH]]:
; CHECK-NEXT:    [[N_VEC:%.*]] = and i64 [[TMP4]], 8589934588
; CHECK-NEXT:    [[TMP12:%.*]] = add nsw i64 [[N_VEC]], [[TMP0]]
; CHECK-NEXT:    br label %[[VECTOR_BODY:.*]]
; CHECK:       [[VECTOR_BODY]]:
; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, %[[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], %[[VECTOR_BODY]] ]
; CHECK-NEXT:    [[OFFSET_IDX:%.*]] = add i64 [[INDEX]], [[TMP0]]
; CHECK-NEXT:    [[TMP13:%.*]] = getelementptr inbounds i32, ptr [[A]], i64 [[OFFSET_IDX]]
; CHECK-NEXT:    [[WIDE_LOAD:%.*]] = load <4 x i32>, ptr [[TMP13]], align 4, !alias.scope [[META0:![0-9]+]], !noalias [[META3:![0-9]+]]
; CHECK-NEXT:    [[TMP14:%.*]] = getelementptr inbounds i32, ptr [[B]], i64 [[OFFSET_IDX]]
; CHECK-NEXT:    [[WIDE_LOAD4:%.*]] = load <4 x i32>, ptr [[TMP14]], align 4, !alias.scope [[META3]]
; CHECK-NEXT:    [[DOTNOT:%.*]] = icmp sgt <4 x i32> [[WIDE_LOAD]], [[WIDE_LOAD4]]
; CHECK-NEXT:    [[TMP15:%.*]] = mul <4 x i32> [[WIDE_LOAD]], splat (i32 5)
; CHECK-NEXT:    [[TMP16:%.*]] = add <4 x i32> [[TMP15]], splat (i32 3)
; CHECK-NEXT:    [[PREDPHI:%.*]] = select <4 x i1> [[DOTNOT]], <4 x i32> [[TMP16]], <4 x i32> [[WIDE_LOAD]]
; CHECK-NEXT:    store <4 x i32> [[PREDPHI]], ptr [[TMP13]], align 4, !alias.scope [[META0]], !noalias [[META3]]
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 4
; CHECK-NEXT:    [[TMP17:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
; CHECK-NEXT:    br i1 [[TMP17]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP5:![0-9]+]]
; CHECK:       [[MIDDLE_BLOCK]]:
; CHECK-NEXT:    [[CMP_N:%.*]] = icmp eq i64 [[TMP4]], [[N_VEC]]
; CHECK-NEXT:    br i1 [[CMP_N]], label %[[FOR_END_LOOPEXIT:.*]], label %[[SCALAR_PH]]
; CHECK:       [[SCALAR_PH]]:
; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[TMP12]], %[[MIDDLE_BLOCK]] ], [ [[TMP0]], %[[FOR_BODY_LR_PH]] ], [ [[TMP0]], %[[VECTOR_MEMCHECK]] ]
; CHECK-NEXT:    br label %[[FOR_BODY:.*]]
; CHECK:       [[FOR_BODY]]:
; CHECK-NEXT:    [[INDVARS_IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], %[[SCALAR_PH]] ], [ [[INDVARS_IV_NEXT:%.*]], %[[IF_END:.*]] ]
; CHECK-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds i32, ptr [[A]], i64 [[INDVARS_IV]]
; CHECK-NEXT:    [[TMP18:%.*]] = load i32, ptr [[ARRAYIDX]], align 4
; CHECK-NEXT:    [[ARRAYIDX4:%.*]] = getelementptr inbounds i32, ptr [[B]], i64 [[INDVARS_IV]]
; CHECK-NEXT:    [[TMP19:%.*]] = load i32, ptr [[ARRAYIDX4]], align 4
; CHECK-NEXT:    [[CMP5:%.*]] = icmp sgt i32 [[TMP18]], [[TMP19]]
; CHECK-NEXT:    br i1 [[CMP5]], label %[[IF_THEN:.*]], label %[[IF_END]]
; CHECK:       [[IF_THEN]]:
; CHECK-NEXT:    [[MUL:%.*]] = mul i32 [[TMP18]], 5
; CHECK-NEXT:    [[ADD:%.*]] = add i32 [[MUL]], 3
; CHECK-NEXT:    br label %[[IF_END]]
; CHECK:       [[IF_END]]:
; CHECK-NEXT:    [[K_0:%.*]] = phi i32 [ [[ADD]], %[[IF_THEN]] ], [ [[TMP18]], %[[FOR_BODY]] ]
; CHECK-NEXT:    store i32 [[K_0]], ptr [[ARRAYIDX]], align 4
; CHECK-NEXT:    [[INDVARS_IV_NEXT]] = add i64 [[INDVARS_IV]], 1
; CHECK-NEXT:    [[TMP20:%.*]] = trunc i64 [[INDVARS_IV_NEXT]] to i32
; CHECK-NEXT:    [[CMP:%.*]] = icmp sgt i32 [[END]], [[TMP20]]
; CHECK-NEXT:    br i1 [[CMP]], label %[[FOR_BODY]], label %[[FOR_END_LOOPEXIT]], !llvm.loop [[LOOP8:![0-9]+]]
; CHECK:       [[FOR_END_LOOPEXIT]]:
; CHECK-NEXT:    br label %[[FOR_END]]
; CHECK:       [[FOR_END]]:
; CHECK-NEXT:    ret i32 undef
;
entry:
  %cmp16 = icmp slt i32 %start, %end
  br i1 %cmp16, label %for.body.lr.ph, label %for.end

for.body.lr.ph:
  %0 = sext i32 %start to i64
  br label %for.body

for.body:
  %indvars.iv = phi i64 [ %0, %for.body.lr.ph ], [ %indvars.iv.next, %if.end ]
  %arrayidx = getelementptr inbounds i32, ptr %a, i64 %indvars.iv
  %1 = load i32, ptr %arrayidx, align 4
  %arrayidx4 = getelementptr inbounds i32, ptr %b, i64 %indvars.iv
  %2 = load i32, ptr %arrayidx4, align 4
  %cmp5 = icmp sgt i32 %1, %2
  br i1 %cmp5, label %if.then, label %if.end

if.then:
  %mul = mul i32 %1, 5
  %add = add i32 %mul, 3
  br label %if.end

if.end:
  %k.0 = phi i32 [ %add, %if.then ], [ %1, %for.body ]
  store i32 %k.0, ptr %arrayidx, align 4
  %indvars.iv.next = add i64 %indvars.iv, 1
  %3 = trunc i64 %indvars.iv.next to i32
  %cmp = icmp slt i32 %3, %end
  br i1 %cmp, label %for.body, label %for.end

for.end:
  ret i32 undef
}



; int func(int *A, int n) {
;   unsigned sum = 0;
;   for (int i = 0; i < n; ++i)
;     if (A[i] > 30)
;       sum += A[i] + 2;
;
;   return sum;
; }

define i32 @reduction_func(ptr nocapture %A, i32 %n) nounwind uwtable readonly ssp {
; CHECK-LABEL: define i32 @reduction_func(
; CHECK-SAME: ptr captures(none) [[A:%.*]], i32 [[N:%.*]]) #[[ATTR1:[0-9]+]] {
; CHECK-NEXT:  [[ENTRY:.*]]:
; CHECK-NEXT:    [[CMP10:%.*]] = icmp sgt i32 [[N]], 0
; CHECK-NEXT:    br i1 [[CMP10]], label %[[FOR_BODY_PREHEADER:.*]], label %[[FOR_END:.*]]
; CHECK:       [[FOR_BODY_PREHEADER]]:
; CHECK-NEXT:    [[TMP0:%.*]] = zext nneg i32 [[N]] to i64
; CHECK-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i32 [[N]], 4
; CHECK-NEXT:    br i1 [[MIN_ITERS_CHECK]], label %[[SCALAR_PH:.*]], label %[[VECTOR_PH:.*]]
; CHECK:       [[VECTOR_PH]]:
; CHECK-NEXT:    [[N_VEC:%.*]] = and i64 [[TMP0]], 2147483644
; CHECK-NEXT:    br label %[[VECTOR_BODY:.*]]
; CHECK:       [[VECTOR_BODY]]:
; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, %[[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], %[[VECTOR_BODY]] ]
; CHECK-NEXT:    [[VEC_PHI:%.*]] = phi <4 x i32> [ zeroinitializer, %[[VECTOR_PH]] ], [ [[PREDPHI:%.*]], %[[VECTOR_BODY]] ]
; CHECK-NEXT:    [[TMP1:%.*]] = getelementptr inbounds i32, ptr [[A]], i64 [[INDEX]]
; CHECK-NEXT:    [[WIDE_LOAD:%.*]] = load <4 x i32>, ptr [[TMP1]], align 4
; CHECK-NEXT:    [[TMP2:%.*]] = icmp sgt <4 x i32> [[WIDE_LOAD]], splat (i32 30)
; CHECK-NEXT:    [[TMP3:%.*]] = add <4 x i32> [[VEC_PHI]], splat (i32 2)
; CHECK-NEXT:    [[TMP4:%.*]] = add <4 x i32> [[TMP3]], [[WIDE_LOAD]]
; CHECK-NEXT:    [[PREDPHI]] = select <4 x i1> [[TMP2]], <4 x i32> [[TMP4]], <4 x i32> [[VEC_PHI]]
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 4
; CHECK-NEXT:    [[TMP5:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
; CHECK-NEXT:    br i1 [[TMP5]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP9:![0-9]+]]
; CHECK:       [[MIDDLE_BLOCK]]:
; CHECK-NEXT:    [[TMP6:%.*]] = call i32 @llvm.vector.reduce.add.v4i32(<4 x i32> [[PREDPHI]])
; CHECK-NEXT:    [[CMP_N:%.*]] = icmp eq i64 [[N_VEC]], [[TMP0]]
; CHECK-NEXT:    br i1 [[CMP_N]], label %[[FOR_END_LOOPEXIT:.*]], label %[[SCALAR_PH]]
; CHECK:       [[SCALAR_PH]]:
; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], %[[MIDDLE_BLOCK]] ], [ 0, %[[FOR_BODY_PREHEADER]] ]
; CHECK-NEXT:    [[BC_MERGE_RDX:%.*]] = phi i32 [ [[TMP6]], %[[MIDDLE_BLOCK]] ], [ 0, %[[FOR_BODY_PREHEADER]] ]
; CHECK-NEXT:    br label %[[FOR_BODY:.*]]
; CHECK:       [[FOR_BODY]]:
; CHECK-NEXT:    [[INDVARS_IV:%.*]] = phi i64 [ [[INDVARS_IV_NEXT:%.*]], %[[FOR_INC:.*]] ], [ [[BC_RESUME_VAL]], %[[SCALAR_PH]] ]
; CHECK-NEXT:    [[SUM_011:%.*]] = phi i32 [ [[SUM_1:%.*]], %[[FOR_INC]] ], [ [[BC_MERGE_RDX]], %[[SCALAR_PH]] ]
; CHECK-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds i32, ptr [[A]], i64 [[INDVARS_IV]]
; CHECK-NEXT:    [[TMP7:%.*]] = load i32, ptr [[ARRAYIDX]], align 4
; CHECK-NEXT:    [[CMP1:%.*]] = icmp sgt i32 [[TMP7]], 30
; CHECK-NEXT:    br i1 [[CMP1]], label %[[IF_THEN:.*]], label %[[FOR_INC]]
; CHECK:       [[IF_THEN]]:
; CHECK-NEXT:    [[ADD:%.*]] = add i32 [[SUM_011]], 2
; CHECK-NEXT:    [[ADD4:%.*]] = add i32 [[ADD]], [[TMP7]]
; CHECK-NEXT:    br label %[[FOR_INC]]
; CHECK:       [[FOR_INC]]:
; CHECK-NEXT:    [[SUM_1]] = phi i32 [ [[ADD4]], %[[IF_THEN]] ], [ [[SUM_011]], %[[FOR_BODY]] ]
; CHECK-NEXT:    [[INDVARS_IV_NEXT]] = add i64 [[INDVARS_IV]], 1
; CHECK-NEXT:    [[LFTR_WIDEIV:%.*]] = trunc i64 [[INDVARS_IV_NEXT]] to i32
; CHECK-NEXT:    [[EXITCOND:%.*]] = icmp eq i32 [[N]], [[LFTR_WIDEIV]]
; CHECK-NEXT:    br i1 [[EXITCOND]], label %[[FOR_END_LOOPEXIT]], label %[[FOR_BODY]], !llvm.loop [[LOOP10:![0-9]+]]
; CHECK:       [[FOR_END_LOOPEXIT]]:
; CHECK-NEXT:    [[SUM_1_LCSSA:%.*]] = phi i32 [ [[SUM_1]], %[[FOR_INC]] ], [ [[TMP6]], %[[MIDDLE_BLOCK]] ]
; CHECK-NEXT:    br label %[[FOR_END]]
; CHECK:       [[FOR_END]]:
; CHECK-NEXT:    [[SUM_0_LCSSA:%.*]] = phi i32 [ 0, %[[ENTRY]] ], [ [[SUM_1_LCSSA]], %[[FOR_END_LOOPEXIT]] ]
; CHECK-NEXT:    ret i32 [[SUM_0_LCSSA]]
;
entry:
  %cmp10 = icmp sgt i32 %n, 0
  br i1 %cmp10, label %for.body, label %for.end

for.body:                                         ; preds = %entry, %for.inc
  %indvars.iv = phi i64 [ %indvars.iv.next, %for.inc ], [ 0, %entry ]
  %sum.011 = phi i32 [ %sum.1, %for.inc ], [ 0, %entry ]
  %arrayidx = getelementptr inbounds i32, ptr %A, i64 %indvars.iv
  %0 = load i32, ptr %arrayidx, align 4
  %cmp1 = icmp sgt i32 %0, 30
  br i1 %cmp1, label %if.then, label %for.inc

if.then:                                          ; preds = %for.body
  %add = add i32 %sum.011, 2
  %add4 = add i32 %add, %0
  br label %for.inc

for.inc:                                          ; preds = %for.body, %if.then
  %sum.1 = phi i32 [ %add4, %if.then ], [ %sum.011, %for.body ]
  %indvars.iv.next = add i64 %indvars.iv, 1
  %lftr.wideiv = trunc i64 %indvars.iv.next to i32
  %exitcond = icmp eq i32 %lftr.wideiv, %n
  br i1 %exitcond, label %for.end, label %for.body

for.end:                                          ; preds = %for.inc, %entry
  %sum.0.lcssa = phi i32 [ 0, %entry ], [ %sum.1, %for.inc ]
  ret i32 %sum.0.lcssa
}

@a = common global [1 x ptr] zeroinitializer, align 8
@c = common global ptr null, align 8

; We use to if convert this loop. This is not safe because there is a trapping
; constant expression.
; PR16729

; Handle PHI with single incoming value having a full mask.
; PR34523

define void @PR34523() {
; CHECK-LABEL: define void @PR34523() {
; CHECK-NEXT:  [[BB1:.*:]]
; CHECK-NEXT:    br i1 true, label %[[SCALAR_PH:.*]], label %[[VECTOR_PH:.*]]
; CHECK:       [[VECTOR_PH]]:
; CHECK-NEXT:    br label %[[VECTOR_BODY:.*]]
; CHECK:       [[VECTOR_BODY]]:
; CHECK-NEXT:    br i1 poison, label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP11:![0-9]+]]
; CHECK:       [[MIDDLE_BLOCK]]:
; CHECK-NEXT:    br i1 poison, label %[[BB5:.*]], label %[[SCALAR_PH]]
; CHECK:       [[SCALAR_PH]]:
; CHECK-NEXT:    br label %[[BB2:.*]]
; CHECK:       [[BB2]]:
; CHECK-NEXT:    [[I:%.*]] = phi i16 [ undef, %[[SCALAR_PH]] ], [ [[_TMP2:%.*]], %[[BB4:.*]] ]
; CHECK-NEXT:    br label %[[BB3:.*]]
; CHECK:       [[BB3]]:
; CHECK-NEXT:    br label %[[BB4]]
; CHECK:       [[BB4]]:
; CHECK-NEXT:    [[_TMP2]] = add i16 [[I]], 1
; CHECK-NEXT:    [[_TMP3:%.*]] = icmp slt i16 [[_TMP2]], 2
; CHECK-NEXT:    br i1 [[_TMP3]], label %[[BB2]], label %[[BB5]], !llvm.loop [[LOOP12:![0-9]+]]
; CHECK:       [[BB5]]:
; CHECK-NEXT:    unreachable
;
bb1:
  br label %bb2

bb2:                                             ; preds = %bb4, %bb1
  %i = phi i16 [ undef, %bb1 ], [ %_tmp2, %bb4 ]
  br label %bb3

bb3:                                             ; preds = %bb2
  %_tmp1 = phi ptr [ undef, %bb2 ]
  br label %bb4

bb4:                                             ; preds = %bb3
  %_tmp2 = add i16 %i, 1
  %_tmp3 = icmp slt i16 %_tmp2, 2
  br i1 %_tmp3, label %bb2, label %bb5

bb5:                                             ; preds = %bb4
  unreachable
}