; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
; RUN: opt < %s -passes=loop-vectorize,instcombine -force-vector-width=4 -force-vector-interleave=1 -tail-folding-policy=prefer-fold-tail -prefer-predicated-reduction-select -force-target-supports-masked-memory-ops -S | FileCheck %s

target datalayout = "e-m:e-p:32:32-Fi8-i64:64-v128:64:128-a:0:32-n32-S64"

define i32 @reduction_sum_single(ptr noalias nocapture %A) {
; CHECK-LABEL: @reduction_sum_single(
; CHECK-NEXT:  entry:
; CHECK-NEXT:    br label [[VECTOR_PH:%.*]]
; CHECK:       vector.ph:
; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
; CHECK:       vector.body:
; CHECK-NEXT:    [[TMP3:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[VEC_PHI:%.*]] = phi <4 x i32> [ zeroinitializer, [[VECTOR_PH]] ], [ [[TMP25:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[VEC_IND:%.*]] = phi <4 x i16> [ <i16 0, i16 1, i16 2, i16 3>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[TMP0:%.*]] = icmp ult <4 x i16> [[VEC_IND]], splat (i16 257)
; CHECK-NEXT:    [[TMP1:%.*]] = getelementptr inbounds [4 x i8], ptr [[A:%.*]], i32 [[TMP3]]
; CHECK-NEXT:    [[TMP24:%.*]] = call <4 x i32> @llvm.masked.load.v4i32.p0(ptr align 4 [[TMP1]], <4 x i1> [[TMP0]], <4 x i32> zeroinitializer)
; CHECK-NEXT:    [[TMP25]] = add <4 x i32> [[VEC_PHI]], [[TMP24]]
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i32 [[TMP3]], 4
; CHECK-NEXT:    [[VEC_IND_NEXT]] = add nuw <4 x i16> [[VEC_IND]], splat (i16 4)
; CHECK-NEXT:    [[TMP26:%.*]] = icmp eq i32 [[INDEX_NEXT]], 260
; CHECK-NEXT:    br i1 [[TMP26]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
; CHECK:       middle.block:
; CHECK-NEXT:    br label [[DOTLR_PH:%.*]]
; CHECK:       ._crit_edge:
; CHECK-NEXT:    [[SUM_0_LCSSA:%.*]] = call i32 @llvm.vector.reduce.add.v4i32(<4 x i32> [[TMP25]])
; CHECK-NEXT:    ret i32 [[SUM_0_LCSSA]]
;
entry:
  br label %.lr.ph

.lr.ph:
  %indvars.iv = phi i32 [ %indvars.iv.next, %.lr.ph ], [ 0, %entry ]
  %sum.02 = phi i32 [ %l7, %.lr.ph ], [ 0, %entry ]
  %l2 = getelementptr inbounds i32, ptr %A, i32 %indvars.iv
  %l3 = load i32, ptr %l2, align 4
  %l7 = add i32 %sum.02, %l3
  %indvars.iv.next = add i32 %indvars.iv, 1
  %exitcond = icmp eq i32 %indvars.iv.next, 257
  br i1 %exitcond, label %._crit_edge, label %.lr.ph

._crit_edge:
  %sum.0.lcssa = phi i32 [ %l7, %.lr.ph ]
  ret i32 %sum.0.lcssa
}

define i32 @reduction_sum(ptr noalias nocapture %A, ptr noalias nocapture %B) {
; CHECK-LABEL: @reduction_sum(
; CHECK-NEXT:  entry:
; CHECK-NEXT:    br label [[VECTOR_PH:%.*]]
; CHECK:       vector.ph:
; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
; CHECK:       vector.body:
; CHECK-NEXT:    [[TMP31:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[VEC_IND:%.*]] = phi <4 x i32> [ <i32 0, i32 1, i32 2, i32 3>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[VEC_PHI:%.*]] = phi <4 x i32> [ zeroinitializer, [[VECTOR_PH]] ], [ [[TMP43:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[TMP0:%.*]] = icmp ult <4 x i32> [[VEC_IND]], splat (i32 257)
; CHECK-NEXT:    [[TMP32:%.*]] = getelementptr inbounds [4 x i8], ptr [[A:%.*]], i32 [[TMP31]]
; CHECK-NEXT:    [[TMP38:%.*]] = call <4 x i32> @llvm.masked.load.v4i32.p0(ptr align 4 [[TMP32]], <4 x i1> [[TMP0]], <4 x i32> poison)
; CHECK-NEXT:    [[TMP35:%.*]] = getelementptr inbounds [4 x i8], ptr [[B:%.*]], i32 [[TMP31]]
; CHECK-NEXT:    [[TMP39:%.*]] = call <4 x i32> @llvm.masked.load.v4i32.p0(ptr align 4 [[TMP35]], <4 x i1> [[TMP0]], <4 x i32> poison)
; CHECK-NEXT:    [[TMP40:%.*]] = add <4 x i32> [[VEC_PHI]], [[VEC_IND]]
; CHECK-NEXT:    [[TMP41:%.*]] = add <4 x i32> [[TMP40]], [[TMP38]]
; CHECK-NEXT:    [[TMP42:%.*]] = add <4 x i32> [[TMP41]], [[TMP39]]
; CHECK-NEXT:    [[TMP43]] = select <4 x i1> [[TMP0]], <4 x i32> [[TMP42]], <4 x i32> [[VEC_PHI]]
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i32 [[TMP31]], 4
; CHECK-NEXT:    [[VEC_IND_NEXT]] = add <4 x i32> [[VEC_IND]], splat (i32 4)
; CHECK-NEXT:    [[TMP44:%.*]] = icmp eq i32 [[INDEX_NEXT]], 260
; CHECK-NEXT:    br i1 [[TMP44]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP3:![0-9]+]]
; CHECK:       middle.block:
; CHECK-NEXT:    br label [[DOTLR_PH:%.*]]
; CHECK:       ._crit_edge:
; CHECK-NEXT:    [[SUM_0_LCSSA:%.*]] = call i32 @llvm.vector.reduce.add.v4i32(<4 x i32> [[TMP43]])
; CHECK-NEXT:    ret i32 [[SUM_0_LCSSA]]
;
entry:
  br label %.lr.ph

.lr.ph:
  %indvars.iv = phi i32 [ %indvars.iv.next, %.lr.ph ], [ 0, %entry ]
  %sum.02 = phi i32 [ %l9, %.lr.ph ], [ 0, %entry ]
  %l2 = getelementptr inbounds i32, ptr %A, i32 %indvars.iv
  %l3 = load i32, ptr %l2, align 4
  %l4 = getelementptr inbounds i32, ptr %B, i32 %indvars.iv
  %l5 = load i32, ptr %l4, align 4
  %l7 = add i32 %sum.02, %indvars.iv
  %l8 = add i32 %l7, %l3
  %l9 = add i32 %l8, %l5
  %indvars.iv.next = add i32 %indvars.iv, 1
  %exitcond = icmp eq i32 %indvars.iv.next, 257
  br i1 %exitcond, label %._crit_edge, label %.lr.ph

._crit_edge:
  %sum.0.lcssa = phi i32 [ %l9, %.lr.ph ]
  ret i32 %sum.0.lcssa
}

define i32 @reduction_prod(ptr noalias nocapture %A, ptr noalias nocapture %B) {
; CHECK-LABEL: @reduction_prod(
; CHECK-NEXT:  entry:
; CHECK-NEXT:    br label [[VECTOR_PH:%.*]]
; CHECK:       vector.ph:
; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
; CHECK:       vector.body:
; CHECK-NEXT:    [[TMP6:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[VEC_PHI:%.*]] = phi <4 x i32> [ splat (i32 1), [[VECTOR_PH]] ], [ [[TMP42:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[VEC_IND:%.*]] = phi <4 x i16> [ <i16 0, i16 1, i16 2, i16 3>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[TMP0:%.*]] = icmp ult <4 x i16> [[VEC_IND]], splat (i16 257)
; CHECK-NEXT:    [[TMP1:%.*]] = getelementptr inbounds [4 x i8], ptr [[A:%.*]], i32 [[TMP6]]
; CHECK-NEXT:    [[TMP38:%.*]] = call <4 x i32> @llvm.masked.load.v4i32.p0(ptr align 4 [[TMP1]], <4 x i1> [[TMP0]], <4 x i32> poison)
; CHECK-NEXT:    [[TMP2:%.*]] = getelementptr inbounds [4 x i8], ptr [[B:%.*]], i32 [[TMP6]]
; CHECK-NEXT:    [[TMP39:%.*]] = call <4 x i32> @llvm.masked.load.v4i32.p0(ptr align 4 [[TMP2]], <4 x i1> [[TMP0]], <4 x i32> poison)
; CHECK-NEXT:    [[TMP40:%.*]] = mul <4 x i32> [[VEC_PHI]], [[TMP38]]
; CHECK-NEXT:    [[TMP41:%.*]] = mul <4 x i32> [[TMP40]], [[TMP39]]
; CHECK-NEXT:    [[TMP42]] = select <4 x i1> [[TMP0]], <4 x i32> [[TMP41]], <4 x i32> [[VEC_PHI]]
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i32 [[TMP6]], 4
; CHECK-NEXT:    [[VEC_IND_NEXT]] = add nuw <4 x i16> [[VEC_IND]], splat (i16 4)
; CHECK-NEXT:    [[TMP43:%.*]] = icmp eq i32 [[INDEX_NEXT]], 260
; CHECK-NEXT:    br i1 [[TMP43]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP4:![0-9]+]]
; CHECK:       middle.block:
; CHECK-NEXT:    br label [[DOTLR_PH:%.*]]
; CHECK:       ._crit_edge:
; CHECK-NEXT:    [[PROD_0_LCSSA:%.*]] = call i32 @llvm.vector.reduce.mul.v4i32(<4 x i32> [[TMP42]])
; CHECK-NEXT:    ret i32 [[PROD_0_LCSSA]]
;
entry:
  br label %.lr.ph

.lr.ph:
  %indvars.iv = phi i32 [ %indvars.iv.next, %.lr.ph ], [ 0, %entry ]
  %prod.02 = phi i32 [ %l9, %.lr.ph ], [ 1, %entry ]
  %l2 = getelementptr inbounds i32, ptr %A, i32 %indvars.iv
  %l3 = load i32, ptr %l2, align 4
  %l4 = getelementptr inbounds i32, ptr %B, i32 %indvars.iv
  %l5 = load i32, ptr %l4, align 4
  %l8 = mul i32 %prod.02, %l3
  %l9 = mul i32 %l8, %l5
  %indvars.iv.next = add i32 %indvars.iv, 1
  %exitcond = icmp eq i32 %indvars.iv.next, 257
  br i1 %exitcond, label %._crit_edge, label %.lr.ph

._crit_edge:
  %prod.0.lcssa = phi i32 [ %l9, %.lr.ph ]
  ret i32 %prod.0.lcssa
}

define i32 @reduction_and(ptr nocapture %A, ptr nocapture %B) {
; CHECK-LABEL: @reduction_and(
; CHECK-NEXT:  entry:
; CHECK-NEXT:    br label [[VECTOR_PH:%.*]]
; CHECK:       vector.ph:
; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
; CHECK:       vector.body:
; CHECK-NEXT:    [[TMP6:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[VEC_PHI:%.*]] = phi <4 x i32> [ splat (i32 -1), [[VECTOR_PH]] ], [ [[TMP42:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[VEC_IND:%.*]] = phi <4 x i16> [ <i16 0, i16 1, i16 2, i16 3>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[TMP0:%.*]] = icmp ult <4 x i16> [[VEC_IND]], splat (i16 257)
; CHECK-NEXT:    [[TMP1:%.*]] = getelementptr inbounds [4 x i8], ptr [[A:%.*]], i32 [[TMP6]]
; CHECK-NEXT:    [[TMP38:%.*]] = call <4 x i32> @llvm.masked.load.v4i32.p0(ptr align 4 [[TMP1]], <4 x i1> [[TMP0]], <4 x i32> poison)
; CHECK-NEXT:    [[TMP2:%.*]] = getelementptr inbounds [4 x i8], ptr [[B:%.*]], i32 [[TMP6]]
; CHECK-NEXT:    [[TMP39:%.*]] = call <4 x i32> @llvm.masked.load.v4i32.p0(ptr align 4 [[TMP2]], <4 x i1> [[TMP0]], <4 x i32> poison)
; CHECK-NEXT:    [[TMP40:%.*]] = and <4 x i32> [[TMP38]], [[TMP39]]
; CHECK-NEXT:    [[TMP41:%.*]] = select <4 x i1> [[TMP0]], <4 x i32> [[TMP40]], <4 x i32> splat (i32 -1)
; CHECK-NEXT:    [[TMP42]] = and <4 x i32> [[VEC_PHI]], [[TMP41]]
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i32 [[TMP6]], 4
; CHECK-NEXT:    [[VEC_IND_NEXT]] = add nuw <4 x i16> [[VEC_IND]], splat (i16 4)
; CHECK-NEXT:    [[TMP43:%.*]] = icmp eq i32 [[INDEX_NEXT]], 260
; CHECK-NEXT:    br i1 [[TMP43]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP5:![0-9]+]]
; CHECK:       middle.block:
; CHECK-NEXT:    br label [[FOR_BODY:%.*]]
; CHECK:       for.end:
; CHECK-NEXT:    [[RESULT_0_LCSSA:%.*]] = call i32 @llvm.vector.reduce.and.v4i32(<4 x i32> [[TMP42]])
; CHECK-NEXT:    ret i32 [[RESULT_0_LCSSA]]
;
entry:
  br label %for.body

for.body:
  %indvars.iv = phi i32 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
  %result.08 = phi i32 [ %and, %for.body ], [ -1, %entry ]
  %arrayidx = getelementptr inbounds i32, ptr %A, i32 %indvars.iv
  %l0 = load i32, ptr %arrayidx, align 4
  %arrayidx2 = getelementptr inbounds i32, ptr %B, i32 %indvars.iv
  %l1 = load i32, ptr %arrayidx2, align 4
  %add = and i32 %result.08, %l0
  %and = and i32 %add, %l1
  %indvars.iv.next = add i32 %indvars.iv, 1
  %exitcond = icmp eq i32 %indvars.iv.next, 257
  br i1 %exitcond, label %for.end, label %for.body

for.end:
  %result.0.lcssa = phi i32 [ %and, %for.body ]
  ret i32 %result.0.lcssa
}

define i32 @reduction_or(ptr nocapture %A, ptr nocapture %B) {
; CHECK-LABEL: @reduction_or(
; CHECK-NEXT:  entry:
; CHECK-NEXT:    br label [[VECTOR_PH:%.*]]
; CHECK:       vector.ph:
; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
; CHECK:       vector.body:
; CHECK-NEXT:    [[TMP6:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[VEC_PHI:%.*]] = phi <4 x i32> [ zeroinitializer, [[VECTOR_PH]] ], [ [[TMP42:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[VEC_IND:%.*]] = phi <4 x i16> [ <i16 0, i16 1, i16 2, i16 3>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[TMP0:%.*]] = icmp ult <4 x i16> [[VEC_IND]], splat (i16 257)
; CHECK-NEXT:    [[TMP1:%.*]] = getelementptr inbounds [4 x i8], ptr [[A:%.*]], i32 [[TMP6]]
; CHECK-NEXT:    [[TMP38:%.*]] = call <4 x i32> @llvm.masked.load.v4i32.p0(ptr align 4 [[TMP1]], <4 x i1> [[TMP0]], <4 x i32> poison)
; CHECK-NEXT:    [[TMP2:%.*]] = getelementptr inbounds [4 x i8], ptr [[B:%.*]], i32 [[TMP6]]
; CHECK-NEXT:    [[TMP39:%.*]] = call <4 x i32> @llvm.masked.load.v4i32.p0(ptr align 4 [[TMP2]], <4 x i1> [[TMP0]], <4 x i32> poison)
; CHECK-NEXT:    [[TMP40:%.*]] = add nsw <4 x i32> [[TMP39]], [[TMP38]]
; CHECK-NEXT:    [[TMP41:%.*]] = select <4 x i1> [[TMP0]], <4 x i32> [[TMP40]], <4 x i32> zeroinitializer
; CHECK-NEXT:    [[TMP42]] = or <4 x i32> [[VEC_PHI]], [[TMP41]]
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i32 [[TMP6]], 4
; CHECK-NEXT:    [[VEC_IND_NEXT]] = add nuw <4 x i16> [[VEC_IND]], splat (i16 4)
; CHECK-NEXT:    [[TMP43:%.*]] = icmp eq i32 [[INDEX_NEXT]], 260
; CHECK-NEXT:    br i1 [[TMP43]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP6:![0-9]+]]
; CHECK:       middle.block:
; CHECK-NEXT:    br label [[FOR_BODY:%.*]]
; CHECK:       for.end:
; CHECK-NEXT:    [[RESULT_0_LCSSA:%.*]] = call i32 @llvm.vector.reduce.or.v4i32(<4 x i32> [[TMP42]])
; CHECK-NEXT:    ret i32 [[RESULT_0_LCSSA]]
;
entry:
  br label %for.body

for.body:
  %indvars.iv = phi i32 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
  %result.08 = phi i32 [ %or, %for.body ], [ 0, %entry ]
  %arrayidx = getelementptr inbounds i32, ptr %A, i32 %indvars.iv
  %l0 = load i32, ptr %arrayidx, align 4
  %arrayidx2 = getelementptr inbounds i32, ptr %B, i32 %indvars.iv
  %l1 = load i32, ptr %arrayidx2, align 4
  %add = add nsw i32 %l1, %l0
  %or = or i32 %add, %result.08
  %indvars.iv.next = add i32 %indvars.iv, 1
  %exitcond = icmp eq i32 %indvars.iv.next, 257
  br i1 %exitcond, label %for.end, label %for.body

for.end:
  %result.0.lcssa = phi i32 [ %or, %for.body ]
  ret i32 %result.0.lcssa
}

define i32 @reduction_xor(ptr nocapture %A, ptr nocapture %B) {
; CHECK-LABEL: @reduction_xor(
; CHECK-NEXT:  entry:
; CHECK-NEXT:    br label [[VECTOR_PH:%.*]]
; CHECK:       vector.ph:
; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
; CHECK:       vector.body:
; CHECK-NEXT:    [[TMP6:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[VEC_PHI:%.*]] = phi <4 x i32> [ zeroinitializer, [[VECTOR_PH]] ], [ [[TMP42:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[VEC_IND:%.*]] = phi <4 x i16> [ <i16 0, i16 1, i16 2, i16 3>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[TMP0:%.*]] = icmp ult <4 x i16> [[VEC_IND]], splat (i16 257)
; CHECK-NEXT:    [[TMP1:%.*]] = getelementptr inbounds [4 x i8], ptr [[A:%.*]], i32 [[TMP6]]
; CHECK-NEXT:    [[TMP38:%.*]] = call <4 x i32> @llvm.masked.load.v4i32.p0(ptr align 4 [[TMP1]], <4 x i1> [[TMP0]], <4 x i32> poison)
; CHECK-NEXT:    [[TMP2:%.*]] = getelementptr inbounds [4 x i8], ptr [[B:%.*]], i32 [[TMP6]]
; CHECK-NEXT:    [[TMP39:%.*]] = call <4 x i32> @llvm.masked.load.v4i32.p0(ptr align 4 [[TMP2]], <4 x i1> [[TMP0]], <4 x i32> poison)
; CHECK-NEXT:    [[TMP40:%.*]] = add nsw <4 x i32> [[TMP39]], [[TMP38]]
; CHECK-NEXT:    [[TMP41:%.*]] = select <4 x i1> [[TMP0]], <4 x i32> [[TMP40]], <4 x i32> zeroinitializer
; CHECK-NEXT:    [[TMP42]] = xor <4 x i32> [[VEC_PHI]], [[TMP41]]
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i32 [[TMP6]], 4
; CHECK-NEXT:    [[VEC_IND_NEXT]] = add nuw <4 x i16> [[VEC_IND]], splat (i16 4)
; CHECK-NEXT:    [[TMP43:%.*]] = icmp eq i32 [[INDEX_NEXT]], 260
; CHECK-NEXT:    br i1 [[TMP43]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP7:![0-9]+]]
; CHECK:       middle.block:
; CHECK-NEXT:    br label [[FOR_BODY:%.*]]
; CHECK:       for.end:
; CHECK-NEXT:    [[RESULT_0_LCSSA:%.*]] = call i32 @llvm.vector.reduce.xor.v4i32(<4 x i32> [[TMP42]])
; CHECK-NEXT:    ret i32 [[RESULT_0_LCSSA]]
;
entry:
  br label %for.body

for.body:
  %indvars.iv = phi i32 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
  %result.08 = phi i32 [ %xor, %for.body ], [ 0, %entry ]
  %arrayidx = getelementptr inbounds i32, ptr %A, i32 %indvars.iv
  %l0 = load i32, ptr %arrayidx, align 4
  %arrayidx2 = getelementptr inbounds i32, ptr %B, i32 %indvars.iv
  %l1 = load i32, ptr %arrayidx2, align 4
  %add = add nsw i32 %l1, %l0
  %xor = xor i32 %add, %result.08
  %indvars.iv.next = add i32 %indvars.iv, 1
  %exitcond = icmp eq i32 %indvars.iv.next, 257
  br i1 %exitcond, label %for.end, label %for.body

for.end:
  %result.0.lcssa = phi i32 [ %xor, %for.body ]
  ret i32 %result.0.lcssa
}

define float @reduction_fadd(ptr nocapture %A, ptr nocapture %B) {
; CHECK-LABEL: @reduction_fadd(
; CHECK-NEXT:  entry:
; CHECK-NEXT:    br label [[VECTOR_PH:%.*]]
; CHECK:       vector.ph:
; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
; CHECK:       vector.body:
; CHECK-NEXT:    [[TMP31:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[VEC_PHI:%.*]] = phi <4 x float> [ zeroinitializer, [[VECTOR_PH]] ], [ [[TMP42:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[VEC_IND:%.*]] = phi <4 x i16> [ <i16 0, i16 1, i16 2, i16 3>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[TMP0:%.*]] = icmp ult <4 x i16> [[VEC_IND]], splat (i16 257)
; CHECK-NEXT:    [[TMP32:%.*]] = getelementptr inbounds [4 x i8], ptr [[A:%.*]], i32 [[TMP31]]
; CHECK-NEXT:    [[TMP38:%.*]] = call <4 x float> @llvm.masked.load.v4f32.p0(ptr align 4 [[TMP32]], <4 x i1> [[TMP0]], <4 x float> poison)
; CHECK-NEXT:    [[TMP35:%.*]] = getelementptr inbounds [4 x i8], ptr [[B:%.*]], i32 [[TMP31]]
; CHECK-NEXT:    [[TMP39:%.*]] = call <4 x float> @llvm.masked.load.v4f32.p0(ptr align 4 [[TMP35]], <4 x i1> [[TMP0]], <4 x float> poison)
; CHECK-NEXT:    [[TMP40:%.*]] = fadd fast <4 x float> [[VEC_PHI]], [[TMP38]]
; CHECK-NEXT:    [[TMP41:%.*]] = fadd fast <4 x float> [[TMP40]], [[TMP39]]
; CHECK-NEXT:    [[TMP42]] = select fast <4 x i1> [[TMP0]], <4 x float> [[TMP41]], <4 x float> [[VEC_PHI]]
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i32 [[TMP31]], 4
; CHECK-NEXT:    [[VEC_IND_NEXT]] = add nuw <4 x i16> [[VEC_IND]], splat (i16 4)
; CHECK-NEXT:    [[TMP43:%.*]] = icmp eq i32 [[INDEX_NEXT]], 260
; CHECK-NEXT:    br i1 [[TMP43]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP8:![0-9]+]]
; CHECK:       middle.block:
; CHECK-NEXT:    br label [[FOR_BODY:%.*]]
; CHECK:       for.end:
; CHECK-NEXT:    [[RESULT_0_LCSSA:%.*]] = call fast float @llvm.vector.reduce.fadd.v4f32(float 0.000000e+00, <4 x float> [[TMP42]])
; CHECK-NEXT:    ret float [[RESULT_0_LCSSA]]
;
entry:
  br label %for.body

for.body:
  %indvars.iv = phi i32 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
  %result.08 = phi float [ %fadd, %for.body ], [ 0.0, %entry ]
  %arrayidx = getelementptr inbounds float, ptr %A, i32 %indvars.iv
  %l0 = load float, ptr %arrayidx, align 4
  %arrayidx2 = getelementptr inbounds float, ptr %B, i32 %indvars.iv
  %l1 = load float, ptr %arrayidx2, align 4
  %add = fadd fast float %result.08, %l0
  %fadd = fadd fast float %add, %l1
  %indvars.iv.next = add i32 %indvars.iv, 1
  %exitcond = icmp eq i32 %indvars.iv.next, 257
  br i1 %exitcond, label %for.end, label %for.body

for.end:
  %result.0.lcssa = phi float [ %fadd, %for.body ]
  ret float %result.0.lcssa
}

define float @reduction_fmul(ptr nocapture %A, ptr nocapture %B) {
; CHECK-LABEL: @reduction_fmul(
; CHECK-NEXT:  entry:
; CHECK-NEXT:    br label [[VECTOR_PH:%.*]]
; CHECK:       vector.ph:
; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
; CHECK:       vector.body:
; CHECK-NEXT:    [[TMP31:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[VEC_PHI:%.*]] = phi <4 x float> [ <float 0.000000e+00, float 1.000000e+00, float 1.000000e+00, float 1.000000e+00>, [[VECTOR_PH]] ], [ [[TMP42:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[VEC_IND:%.*]] = phi <4 x i16> [ <i16 0, i16 1, i16 2, i16 3>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[TMP0:%.*]] = icmp ult <4 x i16> [[VEC_IND]], splat (i16 257)
; CHECK-NEXT:    [[TMP32:%.*]] = getelementptr inbounds [4 x i8], ptr [[A:%.*]], i32 [[TMP31]]
; CHECK-NEXT:    [[TMP38:%.*]] = call <4 x float> @llvm.masked.load.v4f32.p0(ptr align 4 [[TMP32]], <4 x i1> [[TMP0]], <4 x float> poison)
; CHECK-NEXT:    [[TMP35:%.*]] = getelementptr inbounds [4 x i8], ptr [[B:%.*]], i32 [[TMP31]]
; CHECK-NEXT:    [[TMP39:%.*]] = call <4 x float> @llvm.masked.load.v4f32.p0(ptr align 4 [[TMP35]], <4 x i1> [[TMP0]], <4 x float> poison)
; CHECK-NEXT:    [[TMP40:%.*]] = fmul fast <4 x float> [[VEC_PHI]], [[TMP38]]
; CHECK-NEXT:    [[TMP41:%.*]] = fmul fast <4 x float> [[TMP40]], [[TMP39]]
; CHECK-NEXT:    [[TMP42]] = select fast <4 x i1> [[TMP0]], <4 x float> [[TMP41]], <4 x float> [[VEC_PHI]]
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i32 [[TMP31]], 4
; CHECK-NEXT:    [[VEC_IND_NEXT]] = add nuw <4 x i16> [[VEC_IND]], splat (i16 4)
; CHECK-NEXT:    [[TMP43:%.*]] = icmp eq i32 [[INDEX_NEXT]], 260
; CHECK-NEXT:    br i1 [[TMP43]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP9:![0-9]+]]
; CHECK:       middle.block:
; CHECK-NEXT:    br label [[FOR_BODY:%.*]]
; CHECK:       for.end:
; CHECK-NEXT:    [[RESULT_0_LCSSA:%.*]] = call fast float @llvm.vector.reduce.fmul.v4f32(float 1.000000e+00, <4 x float> [[TMP42]])
; CHECK-NEXT:    ret float [[RESULT_0_LCSSA]]
;
entry:
  br label %for.body

for.body:
  %indvars.iv = phi i32 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
  %result.08 = phi float [ %fmul, %for.body ], [ 0.0, %entry ]
  %arrayidx = getelementptr inbounds float, ptr %A, i32 %indvars.iv
  %l0 = load float, ptr %arrayidx, align 4
  %arrayidx2 = getelementptr inbounds float, ptr %B, i32 %indvars.iv
  %l1 = load float, ptr %arrayidx2, align 4
  %add = fmul fast float %result.08, %l0
  %fmul = fmul fast float %add, %l1
  %indvars.iv.next = add i32 %indvars.iv, 1
  %exitcond = icmp eq i32 %indvars.iv.next, 257
  br i1 %exitcond, label %for.end, label %for.body

for.end:
  %result.0.lcssa = phi float [ %fmul, %for.body ]
  ret float %result.0.lcssa
}

define i32 @reduction_min(ptr nocapture %A, ptr nocapture %B) {
; CHECK-LABEL: @reduction_min(
; CHECK-NEXT:  entry:
; CHECK-NEXT:    br label [[VECTOR_PH:%.*]]
; CHECK:       vector.ph:
; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
; CHECK:       vector.body:
; CHECK-NEXT:    [[TMP3:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[VEC_PHI:%.*]] = phi <4 x i32> [ splat (i32 1000), [[VECTOR_PH]] ], [ [[TMP25:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[VEC_IND:%.*]] = phi <4 x i16> [ <i16 0, i16 1, i16 2, i16 3>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[TMP0:%.*]] = icmp ult <4 x i16> [[VEC_IND]], splat (i16 257)
; CHECK-NEXT:    [[TMP1:%.*]] = getelementptr inbounds [4 x i8], ptr [[A:%.*]], i32 [[TMP3]]
; CHECK-NEXT:    [[TMP23:%.*]] = call <4 x i32> @llvm.masked.load.v4i32.p0(ptr align 4 [[TMP1]], <4 x i1> [[TMP0]], <4 x i32> poison)
; CHECK-NEXT:    [[TMP24:%.*]] = call <4 x i32> @llvm.smin.v4i32(<4 x i32> [[VEC_PHI]], <4 x i32> [[TMP23]])
; CHECK-NEXT:    [[TMP25]] = select <4 x i1> [[TMP0]], <4 x i32> [[TMP24]], <4 x i32> [[VEC_PHI]]
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i32 [[TMP3]], 4
; CHECK-NEXT:    [[VEC_IND_NEXT]] = add nuw <4 x i16> [[VEC_IND]], splat (i16 4)
; CHECK-NEXT:    [[TMP26:%.*]] = icmp eq i32 [[INDEX_NEXT]], 260
; CHECK-NEXT:    br i1 [[TMP26]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP10:![0-9]+]]
; CHECK:       middle.block:
; CHECK-NEXT:    br label [[FOR_BODY:%.*]]
; CHECK:       for.end:
; CHECK-NEXT:    [[RESULT_0_LCSSA:%.*]] = call i32 @llvm.vector.reduce.smin.v4i32(<4 x i32> [[TMP25]])
; CHECK-NEXT:    ret i32 [[RESULT_0_LCSSA]]
;
entry:
  br label %for.body

for.body:
  %indvars.iv = phi i32 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
  %result.08 = phi i32 [ %v0, %for.body ], [ 1000, %entry ]
  %arrayidx = getelementptr inbounds i32, ptr %A, i32 %indvars.iv
  %l0 = load i32, ptr %arrayidx, align 4
  %c0 = icmp slt i32 %result.08, %l0
  %v0 = select i1 %c0, i32 %result.08, i32 %l0
  %indvars.iv.next = add i32 %indvars.iv, 1
  %exitcond = icmp eq i32 %indvars.iv.next, 257
  br i1 %exitcond, label %for.end, label %for.body

for.end:
  %result.0.lcssa = phi i32 [ %v0, %for.body ]
  ret i32 %result.0.lcssa
}

define i32 @reduction_max(ptr nocapture %A, ptr nocapture %B) {
; CHECK-LABEL: @reduction_max(
; CHECK-NEXT:  entry:
; CHECK-NEXT:    br label [[VECTOR_PH:%.*]]
; CHECK:       vector.ph:
; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
; CHECK:       vector.body:
; CHECK-NEXT:    [[TMP3:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[VEC_PHI:%.*]] = phi <4 x i32> [ splat (i32 1000), [[VECTOR_PH]] ], [ [[TMP25:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[VEC_IND:%.*]] = phi <4 x i16> [ <i16 0, i16 1, i16 2, i16 3>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[TMP0:%.*]] = icmp ult <4 x i16> [[VEC_IND]], splat (i16 257)
; CHECK-NEXT:    [[TMP1:%.*]] = getelementptr inbounds [4 x i8], ptr [[A:%.*]], i32 [[TMP3]]
; CHECK-NEXT:    [[TMP23:%.*]] = call <4 x i32> @llvm.masked.load.v4i32.p0(ptr align 4 [[TMP1]], <4 x i1> [[TMP0]], <4 x i32> poison)
; CHECK-NEXT:    [[TMP24:%.*]] = call <4 x i32> @llvm.umax.v4i32(<4 x i32> [[VEC_PHI]], <4 x i32> [[TMP23]])
; CHECK-NEXT:    [[TMP25]] = select <4 x i1> [[TMP0]], <4 x i32> [[TMP24]], <4 x i32> [[VEC_PHI]]
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i32 [[TMP3]], 4
; CHECK-NEXT:    [[VEC_IND_NEXT]] = add nuw <4 x i16> [[VEC_IND]], splat (i16 4)
; CHECK-NEXT:    [[TMP26:%.*]] = icmp eq i32 [[INDEX_NEXT]], 260
; CHECK-NEXT:    br i1 [[TMP26]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP11:![0-9]+]]
; CHECK:       middle.block:
; CHECK-NEXT:    br label [[FOR_BODY:%.*]]
; CHECK:       for.end:
; CHECK-NEXT:    [[RESULT_0_LCSSA:%.*]] = call i32 @llvm.vector.reduce.umax.v4i32(<4 x i32> [[TMP25]])
; CHECK-NEXT:    ret i32 [[RESULT_0_LCSSA]]
;
entry:
  br label %for.body

for.body:
  %indvars.iv = phi i32 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
  %result.08 = phi i32 [ %v0, %for.body ], [ 1000, %entry ]
  %arrayidx = getelementptr inbounds i32, ptr %A, i32 %indvars.iv
  %l0 = load i32, ptr %arrayidx, align 4
  %c0 = icmp ugt i32 %result.08, %l0
  %v0 = select i1 %c0, i32 %result.08, i32 %l0
  %indvars.iv.next = add i32 %indvars.iv, 1
  %exitcond = icmp eq i32 %indvars.iv.next, 257
  br i1 %exitcond, label %for.end, label %for.body

for.end:
  %result.0.lcssa = phi i32 [ %v0, %for.body ]
  ret i32 %result.0.lcssa
}