; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
; RUN: opt --arm-mve-gather-scatter-lowering -mtriple=thumbv8.1m.main-none-none-eabi -mattr=+mve.fp %s -S -o - | FileCheck %s

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

define arm_aapcs_vfpcc void @push_out_add_sub_block(i32* noalias nocapture readonly %data, i32* noalias nocapture %dst, i32 %n.vec) {
; CHECK-LABEL: @push_out_add_sub_block(
; CHECK-NEXT:  vector.ph:
; CHECK-NEXT:    [[PUSHEDOUTADD:%.*]] = add <4 x i32> <i32 0, i32 2, i32 4, i32 6>, <i32 6, i32 6, i32 6, i32 6>
; CHECK-NEXT:    [[SCALEDINDEX:%.*]] = shl <4 x i32> [[PUSHEDOUTADD]], <i32 2, i32 2, i32 2, i32 2>
; CHECK-NEXT:    [[TMP0:%.*]] = ptrtoint i32* [[DATA:%.*]] to i32
; CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <4 x i32> poison, i32 [[TMP0]], i32 0
; CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <4 x i32> [[DOTSPLATINSERT]], <4 x i32> poison, <4 x i32> zeroinitializer
; CHECK-NEXT:    [[STARTINDEX:%.*]] = add <4 x i32> [[SCALEDINDEX]], [[DOTSPLAT]]
; CHECK-NEXT:    [[PREINCREMENTSTARTINDEX:%.*]] = sub <4 x i32> [[STARTINDEX]], <i32 32, i32 32, i32 32, i32 32>
; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
; CHECK:       vector.body:
; CHECK-NEXT:    [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH:%.*]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY_END:%.*]] ]
; CHECK-NEXT:    [[VEC_IND:%.*]] = phi <4 x i32> [ [[PREINCREMENTSTARTINDEX]], [[VECTOR_PH]] ], [ [[TMP4:%.*]], [[VECTOR_BODY_END]] ]
; CHECK-NEXT:    [[TMP1:%.*]] = icmp eq i32 [[INDEX]], 48
; CHECK-NEXT:    br i1 [[TMP1]], label [[LOWER_BLOCK:%.*]], label [[END:%.*]]
; CHECK:       lower.block:
; CHECK-NEXT:    [[TMP2:%.*]] = call { <4 x i32>, <4 x i32> } @llvm.arm.mve.vldr.gather.base.wb.v4i32.v4i32(<4 x i32> [[VEC_IND]], i32 32)
; CHECK-NEXT:    [[TMP3:%.*]] = extractvalue { <4 x i32>, <4 x i32> } [[TMP2]], 0
; CHECK-NEXT:    [[TMP4]] = extractvalue { <4 x i32>, <4 x i32> } [[TMP2]], 1
; CHECK-NEXT:    [[TMP5:%.*]] = getelementptr inbounds i32, i32* [[DST:%.*]], i32 [[INDEX]]
; CHECK-NEXT:    [[TMP6:%.*]] = bitcast i32* [[TMP5]] to <4 x i32>*
; CHECK-NEXT:    store <4 x i32> [[TMP3]], <4 x i32>* [[TMP6]], align 4
; CHECK-NEXT:    [[INDEX_NEXT]] = add i32 [[INDEX]], 4
; CHECK-NEXT:    br label [[VECTOR_BODY_END]]
; CHECK:       vector.body.end:
; CHECK-NEXT:    [[TMP7:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC:%.*]]
; CHECK-NEXT:    br i1 [[TMP7]], label [[END]], label [[VECTOR_BODY]]
; CHECK:       end:
; CHECK-NEXT:    ret void
;

vector.ph:
  br label %vector.body

vector.body:                                      ; preds = %vector.body, %vector.ph
  %index = phi i32 [ 0, %vector.ph ], [ %index.next, %vector.body.end ]
  %vec.ind = phi <4 x i32> [ <i32 0, i32 2, i32 4, i32 6>, %vector.ph ], [ %vec.ind.next, %vector.body.end ]
  %0 = icmp eq i32 %index, 48
  br i1 %0, label %lower.block, label %end

lower.block:                             ; preds = %vector.body
  %1 = add <4 x i32> %vec.ind, <i32 6, i32 6, i32 6, i32 6>
  %2 = getelementptr inbounds i32, i32* %data, <4 x i32> %1
  %wide.masked.gather = call <4 x i32> @llvm.masked.gather.v4i32.v4p0i32(<4 x i32*> %2, i32 4, <4 x i1> <i1 true, i1 true, i1 true, i1 true>, <4 x i32> undef)
  %3 = getelementptr inbounds i32, i32* %dst, i32 %index
  %4 = bitcast i32* %3 to <4 x i32>*
  store <4 x i32> %wide.masked.gather, <4 x i32>* %4, align 4
  %index.next = add i32 %index, 4
  %vec.ind.next = add <4 x i32> %vec.ind, <i32 8, i32 8, i32 8, i32 8>
  br label %vector.body.end

vector.body.end:                             ; preds = %lower.block
  %5 = icmp eq i32 %index.next, %n.vec
  br i1 %5, label %end, label %vector.body

end:
  ret void;
}

define arm_aapcs_vfpcc void @push_out_mul_sub_block(i32* noalias nocapture readonly %data, i32* noalias nocapture %dst, i32 %n.vec) {
; CHECK-LABEL: @push_out_mul_sub_block(
; CHECK-NEXT:  vector.ph:
; CHECK-NEXT:    [[PUSHEDOUTMUL:%.*]] = mul <4 x i32> <i32 0, i32 2, i32 4, i32 6>, <i32 3, i32 3, i32 3, i32 3>
; CHECK-NEXT:    [[PRODUCT:%.*]] = mul <4 x i32> <i32 8, i32 8, i32 8, i32 8>, <i32 3, i32 3, i32 3, i32 3>
; CHECK-NEXT:    [[PUSHEDOUTADD:%.*]] = add <4 x i32> [[PUSHEDOUTMUL]], <i32 6, i32 6, i32 6, i32 6>
; CHECK-NEXT:    [[SCALEDINDEX:%.*]] = shl <4 x i32> [[PUSHEDOUTADD]], <i32 2, i32 2, i32 2, i32 2>
; CHECK-NEXT:    [[TMP0:%.*]] = ptrtoint i32* [[DATA:%.*]] to i32
; CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <4 x i32> poison, i32 [[TMP0]], i32 0
; CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <4 x i32> [[DOTSPLATINSERT]], <4 x i32> poison, <4 x i32> zeroinitializer
; CHECK-NEXT:    [[STARTINDEX:%.*]] = add <4 x i32> [[SCALEDINDEX]], [[DOTSPLAT]]
; CHECK-NEXT:    [[PREINCREMENTSTARTINDEX:%.*]] = sub <4 x i32> [[STARTINDEX]], <i32 96, i32 96, i32 96, i32 96>
; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
; CHECK:       vector.body:
; CHECK-NEXT:    [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH:%.*]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY_END:%.*]] ]
; CHECK-NEXT:    [[VEC_IND:%.*]] = phi <4 x i32> [ [[PREINCREMENTSTARTINDEX]], [[VECTOR_PH]] ], [ [[TMP4:%.*]], [[VECTOR_BODY_END]] ]
; CHECK-NEXT:    [[TMP1:%.*]] = icmp eq i32 [[INDEX]], 48
; CHECK-NEXT:    br i1 [[TMP1]], label [[LOWER_BLOCK:%.*]], label [[END:%.*]]
; CHECK:       lower.block:
; CHECK-NEXT:    [[TMP2:%.*]] = call { <4 x i32>, <4 x i32> } @llvm.arm.mve.vldr.gather.base.wb.v4i32.v4i32(<4 x i32> [[VEC_IND]], i32 96)
; CHECK-NEXT:    [[TMP3:%.*]] = extractvalue { <4 x i32>, <4 x i32> } [[TMP2]], 0
; CHECK-NEXT:    [[TMP4]] = extractvalue { <4 x i32>, <4 x i32> } [[TMP2]], 1
; CHECK-NEXT:    [[TMP5:%.*]] = getelementptr inbounds i32, i32* [[DST:%.*]], i32 [[INDEX]]
; CHECK-NEXT:    [[TMP6:%.*]] = bitcast i32* [[TMP5]] to <4 x i32>*
; CHECK-NEXT:    store <4 x i32> [[TMP3]], <4 x i32>* [[TMP6]], align 4
; CHECK-NEXT:    [[INDEX_NEXT]] = add i32 [[INDEX]], 4
; CHECK-NEXT:    br label [[VECTOR_BODY_END]]
; CHECK:       vector.body.end:
; CHECK-NEXT:    [[TMP7:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC:%.*]]
; CHECK-NEXT:    br i1 [[TMP7]], label [[END]], label [[VECTOR_BODY]]
; CHECK:       end:
; CHECK-NEXT:    ret void
;

vector.ph:
  br label %vector.body

vector.body:                                      ; preds = %vector.body, %vector.ph
  %index = phi i32 [ 0, %vector.ph ], [ %index.next, %vector.body.end ]
  %vec.ind = phi <4 x i32> [ <i32 0, i32 2, i32 4, i32 6>, %vector.ph ], [ %vec.ind.next, %vector.body.end ]
  %0 = icmp eq i32 %index, 48
  br i1 %0, label %lower.block, label %end

lower.block:                             ; preds = %vector.body
  %1 = mul <4 x i32> %vec.ind, <i32 3, i32 3, i32 3, i32 3>
  %2 = add <4 x i32> %1, <i32 6, i32 6, i32 6, i32 6>
  %3 = getelementptr inbounds i32, i32* %data, <4 x i32> %2
  %wide.masked.gather = call <4 x i32> @llvm.masked.gather.v4i32.v4p0i32(<4 x i32*> %3, i32 4, <4 x i1> <i1 true, i1 true, i1 true, i1 true>, <4 x i32> undef)
  %4 = getelementptr inbounds i32, i32* %dst, i32 %index
  %5 = bitcast i32* %4 to <4 x i32>*
  store <4 x i32> %wide.masked.gather, <4 x i32>* %5, align 4
  %index.next = add i32 %index, 4
  %vec.ind.next = add <4 x i32> %vec.ind, <i32 8, i32 8, i32 8, i32 8>
  br label %vector.body.end

vector.body.end:                             ; preds = %lower.block
  %6 = icmp eq i32 %index.next, %n.vec
  br i1 %6, label %end, label %vector.body

end:
  ret void;
}


define arm_aapcs_vfpcc void @push_out_mul_sub_loop(i32* noalias nocapture readonly %data, i32* noalias nocapture %dst, i32 %n.vec) {
; CHECK-LABEL: @push_out_mul_sub_loop(
; CHECK-NEXT:  vector.ph:
; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
; CHECK:       vector.body:
; CHECK-NEXT:    [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH:%.*]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY_END:%.*]] ]
; CHECK-NEXT:    [[VEC_IND:%.*]] = phi <4 x i32> [ <i32 0, i32 2, i32 4, i32 6>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY_END]] ]
; CHECK-NEXT:    br label [[VECTOR_2_PH:%.*]]
; CHECK:       vector.2.ph:
; CHECK-NEXT:    br label [[VECTOR_2_BODY:%.*]]
; CHECK:       vector.2.body:
; CHECK-NEXT:    [[TMP0:%.*]] = mul <4 x i32> [[VEC_IND]], <i32 3, i32 3, i32 3, i32 3>
; CHECK-NEXT:    [[SCALEDINDEX:%.*]] = shl <4 x i32> [[TMP0]], <i32 2, i32 2, i32 2, i32 2>
; CHECK-NEXT:    [[TMP1:%.*]] = ptrtoint i32* [[DATA:%.*]] to i32
; CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <4 x i32> poison, i32 [[TMP1]], i32 0
; CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <4 x i32> [[DOTSPLATINSERT]], <4 x i32> poison, <4 x i32> zeroinitializer
; CHECK-NEXT:    [[STARTINDEX:%.*]] = add <4 x i32> [[SCALEDINDEX]], [[DOTSPLAT]]
; CHECK-NEXT:    [[TMP2:%.*]] = call <4 x i32> @llvm.arm.mve.vldr.gather.base.v4i32.v4i32(<4 x i32> [[STARTINDEX]], i32 24)
; CHECK-NEXT:    [[TMP3:%.*]] = getelementptr inbounds i32, i32* [[DST:%.*]], i32 [[INDEX]]
; CHECK-NEXT:    [[TMP4:%.*]] = bitcast i32* [[TMP3]] to <4 x i32>*
; CHECK-NEXT:    store <4 x i32> [[TMP2]], <4 x i32>* [[TMP4]], align 4
; CHECK-NEXT:    br label [[VECTOR_2_BODY_END:%.*]]
; CHECK:       vector.2.body.end:
; CHECK-NEXT:    [[INDEX_2_NEXT:%.*]] = add i32 [[INDEX]], 4
; CHECK-NEXT:    [[TMP5:%.*]] = icmp eq i32 [[INDEX_2_NEXT]], 16
; CHECK-NEXT:    br i1 [[TMP5]], label [[VECTOR_BODY_END]], label [[VECTOR_2_BODY]]
; CHECK:       vector.body.end:
; CHECK-NEXT:    [[INDEX_NEXT]] = add i32 [[INDEX]], 4
; CHECK-NEXT:    [[VEC_IND_NEXT]] = add <4 x i32> [[VEC_IND]], <i32 8, i32 8, i32 8, i32 8>
; CHECK-NEXT:    [[TMP6:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC:%.*]]
; CHECK-NEXT:    br i1 [[TMP6]], label [[END:%.*]], label [[VECTOR_BODY]]
; CHECK:       end:
; CHECK-NEXT:    ret void
;

vector.ph:
  br label %vector.body

vector.body:                                      ; preds = %vector.body, %vector.ph
  %index = phi i32 [ 0, %vector.ph ], [ %index.next, %vector.body.end ]
  %vec.ind = phi <4 x i32> [ <i32 0, i32 2, i32 4, i32 6>, %vector.ph ], [ %vec.ind.next, %vector.body.end ]
  br label %vector.2.ph

vector.2.ph:
  br label %vector.2.body

vector.2.body:                             ; preds = %vector.body
  %0 = mul <4 x i32> %vec.ind, <i32 3, i32 3, i32 3, i32 3>
  %1 = add <4 x i32> %0, <i32 6, i32 6, i32 6, i32 6>
  %2 = getelementptr inbounds i32, i32* %data, <4 x i32> %1
  %wide.masked.gather = call <4 x i32> @llvm.masked.gather.v4i32.v4p0i32(<4 x i32*> %2, i32 4, <4 x i1> <i1 true, i1 true, i1 true, i1 true>, <4 x i32> undef)
  %3 = getelementptr inbounds i32, i32* %dst, i32 %index
  %4 = bitcast i32* %3 to <4 x i32>*
  store <4 x i32> %wide.masked.gather, <4 x i32>* %4, align 4
  br label %vector.2.body.end

vector.2.body.end:                             ; preds = %lower.block
  %index.2.next = add i32 %index, 4
  %5 = icmp eq i32 %index.2.next, 16
  br i1 %5, label %vector.body.end, label %vector.2.body

vector.body.end:                             ; preds = %lower.block
  %index.next = add i32 %index, 4
  %vec.ind.next = add <4 x i32> %vec.ind, <i32 8, i32 8, i32 8, i32 8>
  %6 = icmp eq i32 %index.next, %n.vec
  br i1 %6, label %end, label %vector.body

end:
  ret void;
}

define arm_aapcs_vfpcc void @invariant_add(i32* noalias nocapture readonly %data, i32* noalias nocapture %dst, i32 %n.vec) {
; CHECK-LABEL: @invariant_add(
; CHECK-NEXT:  vector.ph:
; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
; CHECK:       vector.body:
; CHECK-NEXT:    [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH:%.*]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[VEC_IND:%.*]] = phi <4 x i32> [ <i32 0, i32 2, i32 4, i32 6>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[L0:%.*]] = mul <4 x i32> [[VEC_IND]], <i32 3, i32 3, i32 3, i32 3>
; CHECK-NEXT:    [[L1:%.*]] = add <4 x i32> [[L0]], [[VEC_IND]]
; CHECK-NEXT:    [[TMP0:%.*]] = call <4 x i32> @llvm.arm.mve.vldr.gather.offset.v4i32.p0i32.v4i32(i32* [[DATA:%.*]], <4 x i32> [[L1]], i32 32, i32 2, i32 1)
; CHECK-NEXT:    [[L3:%.*]] = getelementptr inbounds i32, i32* [[DST:%.*]], i32 [[INDEX]]
; CHECK-NEXT:    [[L4:%.*]] = bitcast i32* [[L3]] to <4 x i32>*
; CHECK-NEXT:    store <4 x i32> [[TMP0]], <4 x i32>* [[L4]], align 4
; CHECK-NEXT:    [[INDEX_NEXT]] = add i32 [[INDEX]], 4
; CHECK-NEXT:    [[VEC_IND_NEXT]] = add <4 x i32> [[VEC_IND]], <i32 8, i32 8, i32 8, i32 8>
; CHECK-NEXT:    [[L5:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC:%.*]]
; CHECK-NEXT:    br i1 [[L5]], label [[END:%.*]], label [[VECTOR_BODY]]
; CHECK:       end:
; CHECK-NEXT:    ret void
;

vector.ph:
  br label %vector.body

vector.body:                                      ; preds = %vector.body, %vector.ph
  %index = phi i32 [ 0, %vector.ph ], [ %index.next, %vector.body ]
  %vec.ind = phi <4 x i32> [ <i32 0, i32 2, i32 4, i32 6>, %vector.ph ], [ %vec.ind.next, %vector.body ]
  %l0 = mul <4 x i32> %vec.ind, <i32 3, i32 3, i32 3, i32 3>
  %l1 = add <4 x i32> %l0, %vec.ind
  %l2 = getelementptr inbounds i32, i32* %data, <4 x i32> %l1
  %wide.masked.gather = call <4 x i32> @llvm.masked.gather.v4i32.v4p0i32(<4 x i32*> %l2, i32 4, <4 x i1> <i1 true, i1 true, i1 true, i1 true>, <4 x i32> undef)
  %l3 = getelementptr inbounds i32, i32* %dst, i32 %index
  %l4 = bitcast i32* %l3 to <4 x i32>*
  store <4 x i32> %wide.masked.gather, <4 x i32>* %l4, align 4
  %index.next = add i32 %index, 4
  %vec.ind.next = add <4 x i32> %vec.ind, <i32 8, i32 8, i32 8, i32 8>
  %l5 = icmp eq i32 %index.next, %n.vec
  br i1 %l5, label %end, label %vector.body

end:
  ret void;
}

define void @gatherload(i32 %n, i32 %m, i32* nocapture %a, i32* nocapture readonly %b, i32 %call.us.us) {
; CHECK-LABEL: @gatherload(
; CHECK-NEXT:  entry:
; CHECK-NEXT:    [[A57:%.*]] = bitcast i32* [[A:%.*]] to i8*
; CHECK-NEXT:    [[CMP38:%.*]] = icmp sgt i32 [[N:%.*]], 0
; CHECK-NEXT:    br i1 [[CMP38]], label [[FOR_BODY_LR_PH:%.*]], label [[FOR_END16:%.*]]
; CHECK:       for.body.lr.ph:
; CHECK-NEXT:    [[CMP636:%.*]] = icmp sgt i32 [[M:%.*]], 0
; CHECK-NEXT:    br i1 [[CMP636]], label [[FOR_BODY_US_US_PREHEADER:%.*]], label [[FOR_BODY:%.*]]
; CHECK:       for.body.us.us.preheader:
; CHECK-NEXT:    [[TMP0:%.*]] = shl nuw i32 [[M]], 2
; CHECK-NEXT:    [[SCEVGEP:%.*]] = getelementptr i32, i32* [[A]], i32 [[M]]
; CHECK-NEXT:    [[SCEVGEP64:%.*]] = getelementptr i32, i32* [[B:%.*]], i32 [[M]]
; CHECK-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i32 [[M]], 4
; CHECK-NEXT:    [[BOUND0:%.*]] = icmp ugt i32* [[SCEVGEP64]], [[A]]
; CHECK-NEXT:    [[BOUND1:%.*]] = icmp ugt i32* [[SCEVGEP]], [[B]]
; CHECK-NEXT:    [[FOUND_CONFLICT:%.*]] = and i1 [[BOUND0]], [[BOUND1]]
; CHECK-NEXT:    [[N_VEC:%.*]] = and i32 [[M]], -4
; CHECK-NEXT:    [[CMP_N:%.*]] = icmp eq i32 [[N_VEC]], [[M]]
; CHECK-NEXT:    br label [[FOR_BODY_US_US:%.*]]
; CHECK:       for.body.us.us:
; CHECK-NEXT:    [[I_039_US_US:%.*]] = phi i32 [ [[INC15_US_US:%.*]], [[FOR_COND5_FOR_END13_CRIT_EDGE_US_US:%.*]] ], [ 0, [[FOR_BODY_US_US_PREHEADER]] ]
; CHECK-NEXT:    [[VLA_US_US:%.*]] = alloca i32, i32 [[CALL_US_US:%.*]], align 4
; CHECK-NEXT:    [[VLA_US_US56:%.*]] = bitcast i32* [[VLA_US_US]] to i8*
; CHECK-NEXT:    call void @llvm.memcpy.p0i8.p0i8.i32(i8* nonnull align 4 [[VLA_US_US56]], i8* align 4 [[A57]], i32 [[TMP0]], i1 false)
; CHECK-NEXT:    [[BRMERGE:%.*]] = select i1 [[MIN_ITERS_CHECK]], i1 true, i1 [[FOUND_CONFLICT]]
; CHECK-NEXT:    br i1 [[BRMERGE]], label [[FOR_BODY7_US_US_PREHEADER:%.*]], label [[VECTOR_BODY:%.*]]
; CHECK:       vector.body:
; CHECK-NEXT:    [[INDEX:%.*]] = phi i32 [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ], [ 0, [[FOR_BODY_US_US]] ]
; CHECK-NEXT:    [[TMP1:%.*]] = getelementptr inbounds i32, i32* [[B]], i32 [[INDEX]]
; CHECK-NEXT:    [[TMP2:%.*]] = bitcast i32* [[TMP1]] to <4 x i32>*
; CHECK-NEXT:    [[WIDE_LOAD:%.*]] = load <4 x i32>, <4 x i32>* [[TMP2]], align 4
; CHECK-NEXT:    [[TMP3:%.*]] = call <4 x i32> @llvm.arm.mve.vldr.gather.offset.v4i32.p0i32.v4i32(i32* [[VLA_US_US]], <4 x i32> [[WIDE_LOAD]], i32 32, i32 2, i32 1)
; CHECK-NEXT:    [[TMP4:%.*]] = getelementptr inbounds i32, i32* [[A]], i32 [[INDEX]]
; CHECK-NEXT:    [[TMP5:%.*]] = bitcast i32* [[TMP4]] to <4 x i32>*
; CHECK-NEXT:    store <4 x i32> [[TMP3]], <4 x i32>* [[TMP5]], align 4
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i32 [[INDEX]], 4
; CHECK-NEXT:    [[TMP6:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC]]
; CHECK-NEXT:    br i1 [[TMP6]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]]
; CHECK:       middle.block:
; CHECK-NEXT:    br i1 [[CMP_N]], label [[FOR_COND5_FOR_END13_CRIT_EDGE_US_US]], label [[FOR_BODY7_US_US_PREHEADER]]
; CHECK:       for.body7.us.us.preheader:
; CHECK-NEXT:    [[J_137_US_US_PH:%.*]] = phi i32 [ 0, [[FOR_BODY_US_US]] ], [ [[N_VEC]], [[MIDDLE_BLOCK]] ]
; CHECK-NEXT:    br label [[FOR_BODY7_US_US:%.*]]
; CHECK:       for.body7.us.us:
; CHECK-NEXT:    [[J_137_US_US:%.*]] = phi i32 [ [[INC12_US_US:%.*]], [[FOR_BODY7_US_US]] ], [ [[J_137_US_US_PH]], [[FOR_BODY7_US_US_PREHEADER]] ]
; CHECK-NEXT:    [[ARRAYIDX8_US_US:%.*]] = getelementptr inbounds i32, i32* [[B]], i32 [[J_137_US_US]]
; CHECK-NEXT:    [[TMP7:%.*]] = load i32, i32* [[ARRAYIDX8_US_US]], align 4
; CHECK-NEXT:    [[ARRAYIDX9_US_US:%.*]] = getelementptr inbounds i32, i32* [[VLA_US_US]], i32 [[TMP7]]
; CHECK-NEXT:    [[TMP8:%.*]] = load i32, i32* [[ARRAYIDX9_US_US]], align 4
; CHECK-NEXT:    [[ARRAYIDX10_US_US:%.*]] = getelementptr inbounds i32, i32* [[A]], i32 [[J_137_US_US]]
; CHECK-NEXT:    store i32 [[TMP8]], i32* [[ARRAYIDX10_US_US]], align 4
; CHECK-NEXT:    [[INC12_US_US]] = add nuw nsw i32 [[J_137_US_US]], 1
; CHECK-NEXT:    [[EXITCOND58_NOT:%.*]] = icmp eq i32 [[INC12_US_US]], [[M]]
; CHECK-NEXT:    br i1 [[EXITCOND58_NOT]], label [[FOR_COND5_FOR_END13_CRIT_EDGE_US_US]], label [[FOR_BODY7_US_US]]
; CHECK:       for.cond5.for.end13_crit_edge.us.us:
; CHECK-NEXT:    [[INC15_US_US]] = add nuw nsw i32 [[I_039_US_US]], 1
; CHECK-NEXT:    [[EXITCOND59_NOT:%.*]] = icmp eq i32 [[INC15_US_US]], [[N]]
; CHECK-NEXT:    br i1 [[EXITCOND59_NOT]], label [[FOR_END16]], label [[FOR_BODY_US_US]]
; CHECK:       for.body:
; CHECK-NEXT:    [[I_039:%.*]] = phi i32 [ [[INC15:%.*]], [[FOR_BODY]] ], [ 0, [[FOR_BODY_LR_PH]] ]
; CHECK-NEXT:    [[INC15]] = add nuw nsw i32 [[I_039]], 1
; CHECK-NEXT:    [[EXITCOND_NOT:%.*]] = icmp eq i32 [[INC15]], [[N]]
; CHECK-NEXT:    br i1 [[EXITCOND_NOT]], label [[FOR_END16]], label [[FOR_BODY]]
; CHECK:       for.end16:
; CHECK-NEXT:    ret void
;
entry:
  %a57 = bitcast i32* %a to i8*
  %cmp38 = icmp sgt i32 %n, 0
  br i1 %cmp38, label %for.body.lr.ph, label %for.end16

for.body.lr.ph:                                   ; preds = %entry
  %cmp636 = icmp sgt i32 %m, 0
  br i1 %cmp636, label %for.body.us.us.preheader, label %for.body

for.body.us.us.preheader:                         ; preds = %for.body.lr.ph
  %0 = shl nuw i32 %m, 2
  %scevgep = getelementptr i32, i32* %a, i32 %m
  %scevgep64 = getelementptr i32, i32* %b, i32 %m
  %min.iters.check = icmp ult i32 %m, 4
  %bound0 = icmp ugt i32* %scevgep64, %a
  %bound1 = icmp ugt i32* %scevgep, %b
  %found.conflict = and i1 %bound0, %bound1
  %n.vec = and i32 %m, -4
  %cmp.n = icmp eq i32 %n.vec, %m
  br label %for.body.us.us

for.body.us.us:                                   ; preds = %for.body.us.us.preheader, %for.cond5.for.end13_crit_edge.us.us
  %i.039.us.us = phi i32 [ %inc15.us.us, %for.cond5.for.end13_crit_edge.us.us ], [ 0, %for.body.us.us.preheader ]
  %1 = add i32 0, 0
  %vla.us.us = alloca i32, i32 %call.us.us, align 4
  %vla.us.us56 = bitcast i32* %vla.us.us to i8*
  call void @llvm.memcpy.p0i8.p0i8.i32(i8* nonnull align 4 %vla.us.us56, i8* align 4 %a57, i32 %0, i1 false)
  %brmerge = select i1 %min.iters.check, i1 true, i1 %found.conflict
  br i1 %brmerge, label %for.body7.us.us.preheader, label %vector.body

vector.body:                                      ; preds = %for.body.us.us, %vector.body
  %index = phi i32 [ %index.next, %vector.body ], [ 0, %for.body.us.us ]
  %2 = getelementptr inbounds i32, i32* %b, i32 %index
  %3 = bitcast i32* %2 to <4 x i32>*
  %wide.load = load <4 x i32>, <4 x i32>* %3, align 4
  %4 = getelementptr inbounds i32, i32* %vla.us.us, <4 x i32> %wide.load
  %wide.masked.gather = call <4 x i32> @llvm.masked.gather.v4i32.v4p0i32(<4 x i32*> %4, i32 4, <4 x i1> <i1 true, i1 true, i1 true, i1 true>, <4 x i32> undef)
  %5 = getelementptr inbounds i32, i32* %a, i32 %index
  %6 = bitcast i32* %5 to <4 x i32>*
  store <4 x i32> %wide.masked.gather, <4 x i32>* %6, align 4
  %index.next = add nuw i32 %index, 4
  %7 = icmp eq i32 %index.next, %n.vec
  br i1 %7, label %middle.block, label %vector.body

middle.block:                                     ; preds = %vector.body
  br i1 %cmp.n, label %for.cond5.for.end13_crit_edge.us.us, label %for.body7.us.us.preheader

for.body7.us.us.preheader:                        ; preds = %for.body.us.us, %middle.block
  %j.137.us.us.ph = phi i32 [ 0, %for.body.us.us ], [ %n.vec, %middle.block ]
  br label %for.body7.us.us

for.body7.us.us:                                  ; preds = %for.body7.us.us.preheader, %for.body7.us.us
  %j.137.us.us = phi i32 [ %inc12.us.us, %for.body7.us.us ], [ %j.137.us.us.ph, %for.body7.us.us.preheader ]
  %arrayidx8.us.us = getelementptr inbounds i32, i32* %b, i32 %j.137.us.us
  %8 = load i32, i32* %arrayidx8.us.us, align 4
  %arrayidx9.us.us = getelementptr inbounds i32, i32* %vla.us.us, i32 %8
  %9 = load i32, i32* %arrayidx9.us.us, align 4
  %arrayidx10.us.us = getelementptr inbounds i32, i32* %a, i32 %j.137.us.us
  store i32 %9, i32* %arrayidx10.us.us, align 4
  %inc12.us.us = add nuw nsw i32 %j.137.us.us, 1
  %exitcond58.not = icmp eq i32 %inc12.us.us, %m
  br i1 %exitcond58.not, label %for.cond5.for.end13_crit_edge.us.us, label %for.body7.us.us

for.cond5.for.end13_crit_edge.us.us:              ; preds = %for.body7.us.us, %middle.block
  %inc15.us.us = add nuw nsw i32 %i.039.us.us, 1
  %exitcond59.not = icmp eq i32 %inc15.us.us, %n
  br i1 %exitcond59.not, label %for.end16, label %for.body.us.us

for.body:                                         ; preds = %for.body.lr.ph, %for.body
  %i.039 = phi i32 [ %inc15, %for.body ], [ 0, %for.body.lr.ph ]
  %inc15 = add nuw nsw i32 %i.039, 1
  %exitcond.not = icmp eq i32 %inc15, %n
  br i1 %exitcond.not, label %for.end16, label %for.body

for.end16:                                        ; preds = %for.body, %for.cond5.for.end13_crit_edge.us.us, %entry
  ret void
}

declare <4 x i32> @llvm.masked.gather.v4i32.v4p0i32(<4 x i32*>, i32, <4 x i1>, <4 x i32>)
declare void @llvm.memcpy.p0i8.p0i8.i32(i8*, i8*, i32, i1)