; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --check-globals none --version 5
; RUN: opt -p loop-vectorize -force-vector-width=2 -force-vector-interleave=1 -S %s | FileCheck %s

define i32 @loop_with_at_least_2_iterations_via_guards_order_1(ptr %dst, i32 %n) {
; CHECK-LABEL: define i32 @loop_with_at_least_2_iterations_via_guards_order_1(
; CHECK-SAME: ptr [[DST:%.*]], i32 [[N:%.*]]) {
; CHECK-NEXT:  [[ENTRY:.*:]]
; CHECK-NEXT:    [[PRE_0:%.*]] = icmp ne i32 [[N]], 0
; CHECK-NEXT:    [[PRE_1:%.*]] = icmp slt i32 [[N]], 5
; CHECK-NEXT:    [[AND_PRE:%.*]] = and i1 [[PRE_0]], [[PRE_1]]
; CHECK-NEXT:    br i1 [[AND_PRE]], label %[[PH:.*]], label %[[EXIT:.*]]
; CHECK:       [[PH]]:
; CHECK-NEXT:    [[N_EXT:%.*]] = zext i32 [[N]] to i64
; CHECK-NEXT:    [[TMP0:%.*]] = add nuw nsw i64 [[N_EXT]], 1
; CHECK-NEXT:    br label %[[VECTOR_PH:.*]]
; CHECK:       [[VECTOR_PH]]:
; CHECK-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[TMP0]], 2
; CHECK-NEXT:    [[N_VEC:%.*]] = sub i64 [[TMP0]], [[N_MOD_VF]]
; CHECK-NEXT:    br label %[[VECTOR_BODY:.*]]
; CHECK:       [[VECTOR_BODY]]:
; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, %[[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], %[[VECTOR_BODY]] ]
; CHECK-NEXT:    [[TMP1:%.*]] = getelementptr inbounds i32, ptr [[DST]], i64 [[INDEX]]
; CHECK-NEXT:    store <2 x i32> splat (i32 1), ptr [[TMP1]], align 4
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 2
; CHECK-NEXT:    [[TMP3:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
; CHECK-NEXT:    br i1 [[TMP3]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
; CHECK:       [[MIDDLE_BLOCK]]:
; CHECK-NEXT:    [[CMP_N:%.*]] = icmp eq i64 [[TMP0]], [[N_VEC]]
; CHECK-NEXT:    br i1 [[CMP_N]], label %[[EXIT_LOOPEXIT:.*]], label %[[SCALAR_PH:.*]]
; CHECK:       [[SCALAR_PH]]:
; CHECK-NEXT:    br label %[[LOOP:.*]]
; CHECK:       [[LOOP]]:
; CHECK-NEXT:    [[IV:%.*]] = phi i64 [ [[N_VEC]], %[[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], %[[LOOP]] ]
; CHECK-NEXT:    [[GEP:%.*]] = getelementptr inbounds i32, ptr [[DST]], i64 [[IV]]
; CHECK-NEXT:    store i32 1, ptr [[GEP]], align 4
; CHECK-NEXT:    [[IV_NEXT]] = add i64 [[IV]], 1
; CHECK-NEXT:    [[EC:%.*]] = icmp eq i64 [[IV]], [[N_EXT]]
; CHECK-NEXT:    br i1 [[EC]], label %[[EXIT_LOOPEXIT]], label %[[LOOP]], !llvm.loop [[LOOP3:![0-9]+]]
; CHECK:       [[EXIT_LOOPEXIT]]:
; CHECK-NEXT:    br label %[[EXIT]]
; CHECK:       [[EXIT]]:
; CHECK-NEXT:    ret i32 0
;
entry:
  %pre.0 = icmp ne i32 %n, 0
  %pre.1 = icmp slt i32 %n, 5
  %and.pre = and i1 %pre.0, %pre.1
  br i1 %and.pre, label %ph, label %exit

ph:                               ; preds = %entry
  %n.ext = zext i32 %n to i64
  br label %loop

loop:                                         ; preds = %loop, %ph
  %iv = phi i64 [ 0, %ph ], [ %iv.next, %loop ]
  %gep = getelementptr inbounds i32, ptr %dst, i64 %iv
  store i32 1, ptr %gep
  %iv.next = add i64 %iv, 1
  %ec = icmp eq i64 %iv, %n.ext
  br i1 %ec, label %exit, label %loop

exit:                                       ; preds = %loop, %entry
  ret i32 0
}

; Same as loop_with_at_least_2_iterations_via_guards_order_1 but with operands
; of the AND swapped.
; TODO: Should be able to prove that %min.iters.check is false.
define i32 @loop_with_at_least_2_iterations_via_guards_order_2(ptr %dst, i32 %n) {
; CHECK-LABEL: define i32 @loop_with_at_least_2_iterations_via_guards_order_2(
; CHECK-SAME: ptr [[DST:%.*]], i32 [[N:%.*]]) {
; CHECK-NEXT:  [[ENTRY:.*:]]
; CHECK-NEXT:    [[PRE_0:%.*]] = icmp ne i32 [[N]], 0
; CHECK-NEXT:    [[PRE_1:%.*]] = icmp slt i32 [[N]], 5
; CHECK-NEXT:    [[AND_PRE:%.*]] = and i1 [[PRE_1]], [[PRE_0]]
; CHECK-NEXT:    br i1 [[AND_PRE]], label %[[PH:.*]], label %[[EXIT:.*]]
; CHECK:       [[PH]]:
; CHECK-NEXT:    [[N_EXT:%.*]] = zext i32 [[N]] to i64
; CHECK-NEXT:    [[TMP0:%.*]] = add nuw nsw i64 [[N_EXT]], 1
; CHECK-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 [[TMP0]], 2
; CHECK-NEXT:    br i1 [[MIN_ITERS_CHECK]], label %[[SCALAR_PH:.*]], label %[[VECTOR_PH:.*]]
; CHECK:       [[VECTOR_PH]]:
; CHECK-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[TMP0]], 2
; CHECK-NEXT:    [[N_VEC:%.*]] = sub i64 [[TMP0]], [[N_MOD_VF]]
; CHECK-NEXT:    br label %[[VECTOR_BODY:.*]]
; CHECK:       [[VECTOR_BODY]]:
; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, %[[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], %[[VECTOR_BODY]] ]
; CHECK-NEXT:    [[TMP1:%.*]] = getelementptr inbounds i32, ptr [[DST]], i64 [[INDEX]]
; CHECK-NEXT:    store <2 x i32> splat (i32 1), ptr [[TMP1]], align 4
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 2
; CHECK-NEXT:    [[TMP3:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
; CHECK-NEXT:    br i1 [[TMP3]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP4:![0-9]+]]
; CHECK:       [[MIDDLE_BLOCK]]:
; CHECK-NEXT:    [[CMP_N:%.*]] = icmp eq i64 [[TMP0]], [[N_VEC]]
; CHECK-NEXT:    br i1 [[CMP_N]], label %[[EXIT_LOOPEXIT:.*]], label %[[SCALAR_PH]]
; CHECK:       [[SCALAR_PH]]:
; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], %[[MIDDLE_BLOCK]] ], [ 0, %[[PH]] ]
; CHECK-NEXT:    br label %[[LOOP:.*]]
; CHECK:       [[LOOP]]:
; CHECK-NEXT:    [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], %[[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], %[[LOOP]] ]
; CHECK-NEXT:    [[GEP:%.*]] = getelementptr inbounds i32, ptr [[DST]], i64 [[IV]]
; CHECK-NEXT:    store i32 1, ptr [[GEP]], align 4
; CHECK-NEXT:    [[IV_NEXT]] = add i64 [[IV]], 1
; CHECK-NEXT:    [[EC:%.*]] = icmp eq i64 [[IV]], [[N_EXT]]
; CHECK-NEXT:    br i1 [[EC]], label %[[EXIT_LOOPEXIT]], label %[[LOOP]], !llvm.loop [[LOOP5:![0-9]+]]
; CHECK:       [[EXIT_LOOPEXIT]]:
; CHECK-NEXT:    br label %[[EXIT]]
; CHECK:       [[EXIT]]:
; CHECK-NEXT:    ret i32 0
;
entry:
  %pre.0 = icmp ne i32 %n, 0
  %pre.1 = icmp slt i32 %n, 5
  %and.pre = and i1 %pre.1, %pre.0
  br i1 %and.pre, label %ph, label %exit

ph:
  %n.ext = zext i32 %n to i64
  br label %loop

loop:
  %iv = phi i64 [ 0, %ph ], [ %iv.next, %loop ]
  %gep = getelementptr inbounds i32, ptr %dst, i64 %iv
  store i32 1, ptr %gep
  %iv.next = add i64 %iv, 1
  %ec = icmp eq i64 %iv, %n.ext
  br i1 %ec, label %exit, label %loop

exit:
  ret i32 0
}

; TODO: Should be able to prove min.iters.check is false, same as
; @loop_never_executes_precondition_order_1_predicates_flipped.
define void @loop_never_executes_precondition_order_1(i64 %start, ptr %dst) {
; CHECK-LABEL: define void @loop_never_executes_precondition_order_1(
; CHECK-SAME: i64 [[START:%.*]], ptr [[DST:%.*]]) {
; CHECK-NEXT:  [[ENTRY:.*:]]
; CHECK-NEXT:    [[PRE_0:%.*]] = icmp ugt i64 [[START]], 1
; CHECK-NEXT:    br i1 [[PRE_0]], label %[[EXIT:.*]], label %[[PH:.*]]
; CHECK:       [[PH]]:
; CHECK-NEXT:    [[MUL:%.*]] = shl i64 [[START]], 1
; CHECK-NEXT:    [[PRE_1:%.*]] = icmp slt i64 [[MUL]], [[START]]
; CHECK-NEXT:    br i1 [[PRE_1]], label %[[LOOP_PREHEADER:.*]], label %[[EXIT]]
; CHECK:       [[LOOP_PREHEADER]]:
; CHECK-NEXT:    [[TMP0:%.*]] = sub i64 1, [[START]]
; CHECK-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 [[TMP0]], 2
; CHECK-NEXT:    br i1 [[MIN_ITERS_CHECK]], label %[[SCALAR_PH:.*]], label %[[VECTOR_PH:.*]]
; CHECK:       [[VECTOR_PH]]:
; CHECK-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[TMP0]], 2
; CHECK-NEXT:    [[N_VEC:%.*]] = sub i64 [[TMP0]], [[N_MOD_VF]]
; CHECK-NEXT:    [[TMP1:%.*]] = add i64 [[START]], [[N_VEC]]
; CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <2 x i64> poison, i64 [[START]], i64 0
; CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <2 x i64> [[DOTSPLATINSERT]], <2 x i64> poison, <2 x i32> zeroinitializer
; CHECK-NEXT:    [[INDUCTION:%.*]] = add <2 x i64> [[DOTSPLAT]], <i64 0, i64 1>
; CHECK-NEXT:    br label %[[VECTOR_BODY:.*]]
; CHECK:       [[VECTOR_BODY]]:
; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, %[[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], %[[VECTOR_BODY]] ]
; CHECK-NEXT:    [[VEC_IND:%.*]] = phi <2 x i64> [ [[INDUCTION]], %[[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], %[[VECTOR_BODY]] ]
; CHECK-NEXT:    [[OFFSET_IDX:%.*]] = add i64 [[START]], [[INDEX]]
; CHECK-NEXT:    [[TMP2:%.*]] = getelementptr inbounds i64, ptr [[DST]], i64 [[OFFSET_IDX]]
; CHECK-NEXT:    store <2 x i64> [[VEC_IND]], ptr [[TMP2]], align 4
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 2
; CHECK-NEXT:    [[VEC_IND_NEXT]] = add <2 x i64> [[VEC_IND]], splat (i64 2)
; CHECK-NEXT:    [[TMP4:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
; CHECK-NEXT:    br i1 [[TMP4]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP6:![0-9]+]]
; CHECK:       [[MIDDLE_BLOCK]]:
; CHECK-NEXT:    [[CMP_N:%.*]] = icmp eq i64 [[TMP0]], [[N_VEC]]
; CHECK-NEXT:    br i1 [[CMP_N]], label %[[EXIT_LOOPEXIT:.*]], label %[[SCALAR_PH]]
; CHECK:       [[SCALAR_PH]]:
; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[TMP1]], %[[MIDDLE_BLOCK]] ], [ [[START]], %[[LOOP_PREHEADER]] ]
; CHECK-NEXT:    br label %[[LOOP:.*]]
; CHECK:       [[LOOP]]:
; CHECK-NEXT:    [[IV:%.*]] = phi i64 [ [[IV_NEXT:%.*]], %[[LOOP]] ], [ [[BC_RESUME_VAL]], %[[SCALAR_PH]] ]
; CHECK-NEXT:    [[GEP:%.*]] = getelementptr inbounds i64, ptr [[DST]], i64 [[IV]]
; CHECK-NEXT:    store i64 [[IV]], ptr [[GEP]], align 4
; CHECK-NEXT:    [[IV_NEXT]] = add i64 [[IV]], 1
; CHECK-NEXT:    [[EC:%.*]] = icmp eq i64 [[IV]], 0
; CHECK-NEXT:    br i1 [[EC]], label %[[EXIT_LOOPEXIT]], label %[[LOOP]], !llvm.loop [[LOOP7:![0-9]+]]
; CHECK:       [[EXIT_LOOPEXIT]]:
; CHECK-NEXT:    br label %[[EXIT]]
; CHECK:       [[EXIT]]:
; CHECK-NEXT:    ret void
;
entry:
  %pre.0 = icmp ugt i64 %start, 1
  br i1 %pre.0, label %exit, label %ph

ph:
  %mul = shl i64 %start, 1
  %pre.1 = icmp slt i64 %mul, %start
  br i1 %pre.1, label %loop, label %exit

loop:
  %iv = phi i64 [ %start, %ph ], [ %iv.next, %loop ]
  %gep = getelementptr inbounds i64, ptr %dst, i64 %iv
  store i64 %iv, ptr %gep
  %iv.next = add i64 %iv, 1
  %ec = icmp eq i64 %iv, 0
  br i1 %ec, label %exit, label %loop

exit:
  ret void
}

; TODO: Should be able to prove min.iters.check is false, same as
; @loop_never_executes_precondition_order_1_predicates_flipped.
define void @loop_never_executes_precondition_order_1_predicates_flipped(i64 %start, ptr %dst) {
; CHECK-LABEL: define void @loop_never_executes_precondition_order_1_predicates_flipped(
; CHECK-SAME: i64 [[START:%.*]], ptr [[DST:%.*]]) {
; CHECK-NEXT:  [[ENTRY:.*:]]
; CHECK-NEXT:    [[PRE_0:%.*]] = icmp ule i64 [[START]], 1
; CHECK-NEXT:    br i1 [[PRE_0]], label %[[PH:.*]], label %[[EXIT:.*]]
; CHECK:       [[PH]]:
; CHECK-NEXT:    [[MUL:%.*]] = shl i64 [[START]], 1
; CHECK-NEXT:    [[PRE_1:%.*]] = icmp slt i64 [[MUL]], [[START]]
; CHECK-NEXT:    br i1 [[PRE_1]], label %[[LOOP_PREHEADER:.*]], label %[[EXIT]]
; CHECK:       [[LOOP_PREHEADER]]:
; CHECK-NEXT:    [[TMP0:%.*]] = sub i64 1, [[START]]
; CHECK-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 [[TMP0]], 2
; CHECK-NEXT:    br i1 [[MIN_ITERS_CHECK]], label %[[SCALAR_PH:.*]], label %[[VECTOR_PH:.*]]
; CHECK:       [[VECTOR_PH]]:
; CHECK-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[TMP0]], 2
; CHECK-NEXT:    [[N_VEC:%.*]] = sub i64 [[TMP0]], [[N_MOD_VF]]
; CHECK-NEXT:    [[TMP1:%.*]] = add i64 [[START]], [[N_VEC]]
; CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <2 x i64> poison, i64 [[START]], i64 0
; CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <2 x i64> [[DOTSPLATINSERT]], <2 x i64> poison, <2 x i32> zeroinitializer
; CHECK-NEXT:    [[INDUCTION:%.*]] = add <2 x i64> [[DOTSPLAT]], <i64 0, i64 1>
; CHECK-NEXT:    br label %[[VECTOR_BODY:.*]]
; CHECK:       [[VECTOR_BODY]]:
; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, %[[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], %[[VECTOR_BODY]] ]
; CHECK-NEXT:    [[VEC_IND:%.*]] = phi <2 x i64> [ [[INDUCTION]], %[[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], %[[VECTOR_BODY]] ]
; CHECK-NEXT:    [[OFFSET_IDX:%.*]] = add i64 [[START]], [[INDEX]]
; CHECK-NEXT:    [[TMP2:%.*]] = getelementptr inbounds i64, ptr [[DST]], i64 [[OFFSET_IDX]]
; CHECK-NEXT:    store <2 x i64> [[VEC_IND]], ptr [[TMP2]], align 4
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 2
; CHECK-NEXT:    [[VEC_IND_NEXT]] = add <2 x i64> [[VEC_IND]], splat (i64 2)
; CHECK-NEXT:    [[TMP4:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
; CHECK-NEXT:    br i1 [[TMP4]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP8:![0-9]+]]
; CHECK:       [[MIDDLE_BLOCK]]:
; CHECK-NEXT:    [[CMP_N:%.*]] = icmp eq i64 [[TMP0]], [[N_VEC]]
; CHECK-NEXT:    br i1 [[CMP_N]], label %[[EXIT_LOOPEXIT:.*]], label %[[SCALAR_PH]]
; CHECK:       [[SCALAR_PH]]:
; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[TMP1]], %[[MIDDLE_BLOCK]] ], [ [[START]], %[[LOOP_PREHEADER]] ]
; CHECK-NEXT:    br label %[[LOOP:.*]]
; CHECK:       [[LOOP]]:
; CHECK-NEXT:    [[IV:%.*]] = phi i64 [ [[IV_NEXT:%.*]], %[[LOOP]] ], [ [[BC_RESUME_VAL]], %[[SCALAR_PH]] ]
; CHECK-NEXT:    [[GEP:%.*]] = getelementptr inbounds i64, ptr [[DST]], i64 [[IV]]
; CHECK-NEXT:    store i64 [[IV]], ptr [[GEP]], align 4
; CHECK-NEXT:    [[IV_NEXT]] = add i64 [[IV]], 1
; CHECK-NEXT:    [[EC:%.*]] = icmp eq i64 [[IV]], 0
; CHECK-NEXT:    br i1 [[EC]], label %[[EXIT_LOOPEXIT]], label %[[LOOP]], !llvm.loop [[LOOP9:![0-9]+]]
; CHECK:       [[EXIT_LOOPEXIT]]:
; CHECK-NEXT:    br label %[[EXIT]]
; CHECK:       [[EXIT]]:
; CHECK-NEXT:    ret void
;
entry:
  %pre.0 = icmp ule i64 %start, 1
  br i1 %pre.0, label %ph, label %exit

ph:
  %mul = shl i64 %start, 1
  %pre.1 = icmp slt i64 %mul, %start
  br i1 %pre.1, label %loop, label %exit

loop:
  %iv = phi i64 [ %start, %ph ], [ %iv.next, %loop ]
  %gep = getelementptr inbounds i64, ptr %dst, i64 %iv
  store i64 %iv, ptr %gep
  %iv.next = add i64 %iv, 1
  %ec = icmp eq i64 %iv, 0
  br i1 %ec, label %exit, label %loop

exit:
  ret void
}

define void @loop_never_executes_precondition_order_2_predicates_flipped(i64 %start, ptr %dst) {
; CHECK-LABEL: define void @loop_never_executes_precondition_order_2_predicates_flipped(
; CHECK-SAME: i64 [[START:%.*]], ptr [[DST:%.*]]) {
; CHECK-NEXT:  [[ENTRY:.*:]]
; CHECK-NEXT:    [[MUL:%.*]] = shl i64 [[START]], 1
; CHECK-NEXT:    [[PRE_1:%.*]] = icmp slt i64 [[MUL]], [[START]]
; CHECK-NEXT:    br i1 [[PRE_1]], label %[[PH:.*]], label %[[EXIT:.*]]
; CHECK:       [[PH]]:
; CHECK-NEXT:    [[PRE_0:%.*]] = icmp ule i64 [[START]], 1
; CHECK-NEXT:    br i1 [[PRE_0]], label %[[LOOP_PREHEADER:.*]], label %[[EXIT]]
; CHECK:       [[LOOP_PREHEADER]]:
; CHECK-NEXT:    br label %[[LOOP:.*]]
; CHECK:       [[LOOP]]:
; CHECK-NEXT:    [[IV:%.*]] = phi i64 [ [[IV_NEXT:%.*]], %[[LOOP]] ], [ [[START]], %[[LOOP_PREHEADER]] ]
; CHECK-NEXT:    [[GEP:%.*]] = getelementptr inbounds i64, ptr [[DST]], i64 [[IV]]
; CHECK-NEXT:    store i64 [[IV]], ptr [[GEP]], align 4
; CHECK-NEXT:    [[IV_NEXT]] = add i64 [[IV]], 1
; CHECK-NEXT:    [[EC:%.*]] = icmp eq i64 [[IV]], 0
; CHECK-NEXT:    br i1 [[EC]], label %[[EXIT_LOOPEXIT:.*]], label %[[LOOP]]
; CHECK:       [[EXIT_LOOPEXIT]]:
; CHECK-NEXT:    br label %[[EXIT]]
; CHECK:       [[EXIT]]:
; CHECK-NEXT:    ret void
;
entry:
  %mul = shl i64 %start, 1
  %pre.1 = icmp slt i64 %mul, %start
  br i1 %pre.1, label %ph, label %exit

ph:
  %pre.0 = icmp ule i64 %start, 1
  br i1 %pre.0, label %loop, label %exit

loop:
  %iv = phi i64 [ %start, %ph ], [ %iv.next, %loop ]
  %gep = getelementptr inbounds i64, ptr %dst, i64 %iv
  store i64 %iv, ptr %gep
  %iv.next = add i64 %iv, 1
  %ec = icmp eq i64 %iv, 0
  br i1 %ec, label %exit, label %loop

exit:
  ret void
}