; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 5
; RUN: opt -passes=loop-vectorize -force-vector-width=4 -S %s | FileCheck %s

define void @test(ptr %A, i32 %x) {
; CHECK-LABEL: define void @test(
; CHECK-SAME: ptr [[A:%.*]], i32 [[X:%.*]]) {
; CHECK-NEXT:  [[ENTRY:.*:]]
; CHECK-NEXT:    br label %[[VECTOR_SCEVCHECK:.*]]
; CHECK:       [[VECTOR_SCEVCHECK]]:
; CHECK-NEXT:    [[IDENT_CHECK:%.*]] = icmp ne i32 [[X]], 1
; CHECK-NEXT:    br i1 [[IDENT_CHECK]], label %[[SCALAR_PH:.*]], label %[[VECTOR_PH:.*]]
; CHECK:       [[VECTOR_PH]]:
; CHECK-NEXT:    br label %[[VECTOR_BODY:.*]]
; CHECK:       [[VECTOR_BODY]]:
; CHECK-NEXT:    [[OFFSET_IDX:%.*]] = phi i64 [ 0, %[[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], %[[VECTOR_BODY]] ]
; CHECK-NEXT:    [[TMP0:%.*]] = trunc i64 [[OFFSET_IDX]] to i32
; CHECK-NEXT:    [[TMP3:%.*]] = add nuw nsw i64 [[OFFSET_IDX]], 1
; CHECK-NEXT:    [[TMP4:%.*]] = trunc i64 [[TMP3]] to i32
; CHECK-NEXT:    [[TMP5:%.*]] = zext i32 [[TMP4]] to i64
; CHECK-NEXT:    [[TMP6:%.*]] = getelementptr inbounds float, ptr [[A]], i64 [[TMP5]]
; CHECK-NEXT:    [[WIDE_LOAD:%.*]] = load <4 x float>, ptr [[TMP6]], align 4
; CHECK-NEXT:    [[TMP8:%.*]] = zext i32 [[TMP0]] to i64
; CHECK-NEXT:    [[TMP9:%.*]] = getelementptr inbounds float, ptr [[A]], i64 [[TMP8]]
; CHECK-NEXT:    store <4 x float> [[WIDE_LOAD]], ptr [[TMP9]], align 4
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[OFFSET_IDX]], 4
; CHECK-NEXT:    [[TMP7:%.*]] = icmp eq i64 [[INDEX_NEXT]], 1000
; CHECK-NEXT:    br i1 [[TMP7]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
; CHECK:       [[MIDDLE_BLOCK]]:
; CHECK-NEXT:    br label %[[EXIT:.*]]
; CHECK:       [[SCALAR_PH]]:
; CHECK-NEXT:    br label %[[LOOP:.*]]
; CHECK:       [[LOOP]]:
; CHECK-NEXT:    [[IV:%.*]] = phi i64 [ [[IV_NEXT:%.*]], %[[LOOP]] ], [ 0, %[[SCALAR_PH]] ]
; CHECK-NEXT:    [[IV_NEXT]] = add nuw nsw i64 [[IV]], 1
; CHECK-NEXT:    [[T_IV_NEXT:%.*]] = trunc i64 [[IV_NEXT]] to i32
; CHECK-NEXT:    [[MUL_IV_NEXT:%.*]] = mul i32 [[T_IV_NEXT]], [[X]]
; CHECK-NEXT:    [[IDX_1:%.*]] = zext i32 [[MUL_IV_NEXT]] to i64
; CHECK-NEXT:    [[ARRAYIDX1215:%.*]] = getelementptr inbounds float, ptr [[A]], i64 [[IDX_1]]
; CHECK-NEXT:    [[LV:%.*]] = load float, ptr [[ARRAYIDX1215]], align 4
; CHECK-NEXT:    [[T_IV:%.*]] = trunc i64 [[IV]] to i32
; CHECK-NEXT:    [[MUL_IV:%.*]] = mul i32 [[T_IV]], [[X]]
; CHECK-NEXT:    [[IDX_2:%.*]] = zext i32 [[MUL_IV]] to i64
; CHECK-NEXT:    [[ARRAYIDX1209:%.*]] = getelementptr inbounds float, ptr [[A]], i64 [[IDX_2]]
; CHECK-NEXT:    store float [[LV]], ptr [[ARRAYIDX1209]], align 4
; CHECK-NEXT:    [[EC:%.*]] = icmp eq i64 [[IV_NEXT]], 1000
; CHECK-NEXT:    br i1 [[EC]], label %[[EXIT]], label %[[LOOP]], !llvm.loop [[LOOP3:![0-9]+]]
; CHECK:       [[EXIT]]:
; CHECK-NEXT:    ret void
;
entry:
  br label %loop

loop:
  %iv = phi i64 [ %iv.next, %loop ], [ 0, %entry ]
  %iv.next = add nuw nsw i64 %iv, 1
  %t.iv.next = trunc i64 %iv.next to i32
  %mul.iv.next = mul i32 %t.iv.next, %x
  %idx.1 = zext i32 %mul.iv.next to i64
  %arrayidx1215 = getelementptr inbounds float, ptr %A, i64 %idx.1
  %lv = load float, ptr %arrayidx1215, align 4
  %t.iv = trunc i64 %iv to i32
  %mul.iv = mul i32 %t.iv, %x
  %idx.2 = zext i32 %mul.iv to i64
  %arrayidx1209 = getelementptr inbounds float, ptr %A, i64 %idx.2
  store float %lv, ptr %arrayidx1209, align 4
  %ec = icmp eq i64 %iv.next, 1000
  br i1 %ec, label %exit, label %loop

exit:
  ret void
}

; !llvm.loop.unroll.runtime.disable metadata should be added to the
; scalar loop, as there are no runtime checks needed (or they can be proven
; false).
define void @diff_memcheck_known_false_for_vf_4(ptr %B, ptr %A, ptr %end) {
; CHECK-LABEL: define void @diff_memcheck_known_false_for_vf_4(
; CHECK-SAME: ptr [[B:%.*]], ptr [[A:%.*]], ptr [[END:%.*]]) {
; CHECK-NEXT:  [[ENTRY:.*]]:
; CHECK-NEXT:    [[END1:%.*]] = ptrtoint ptr [[END]] to i64
; CHECK-NEXT:    [[A_INT:%.*]] = ptrtoint ptr [[A]] to i64
; CHECK-NEXT:    [[B_CAST:%.*]] = ptrtoint ptr [[B]] to i64
; CHECK-NEXT:    [[PTR_SUB:%.*]] = sub i64 [[A_INT]], [[B_CAST]]
; CHECK-NEXT:    [[ADD_PTR11:%.*]] = getelementptr i8, ptr [[B]], i64 [[PTR_SUB]]
; CHECK-NEXT:    [[TMP0:%.*]] = sub i64 [[A_INT]], [[END1]]
; CHECK-NEXT:    [[TMP1:%.*]] = lshr i64 [[TMP0]], 3
; CHECK-NEXT:    [[TMP2:%.*]] = add nuw nsw i64 [[TMP1]], 1
; CHECK-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 [[TMP2]], 4
; CHECK-NEXT:    br i1 [[MIN_ITERS_CHECK]], label %[[SCALAR_PH:.*]], label %[[VECTOR_PH:.*]]
; CHECK:       [[VECTOR_PH]]:
; CHECK-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[TMP2]], 4
; CHECK-NEXT:    [[N_VEC:%.*]] = sub i64 [[TMP2]], [[N_MOD_VF]]
; CHECK-NEXT:    [[TMP3:%.*]] = mul i64 [[N_VEC]], -8
; CHECK-NEXT:    [[TMP4:%.*]] = getelementptr i8, ptr [[A]], i64 [[TMP3]]
; CHECK-NEXT:    [[TMP5:%.*]] = mul i64 [[N_VEC]], -8
; CHECK-NEXT:    [[TMP6:%.*]] = getelementptr i8, ptr [[ADD_PTR11]], i64 [[TMP5]]
; 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:%.*]] = mul i64 [[INDEX]], -8
; CHECK-NEXT:    [[NEXT_GEP:%.*]] = getelementptr i8, ptr [[A]], i64 [[OFFSET_IDX]]
; CHECK-NEXT:    [[TMP7:%.*]] = getelementptr i64, ptr [[NEXT_GEP]], i32 0
; CHECK-NEXT:    [[TMP8:%.*]] = getelementptr i64, ptr [[TMP7]], i32 -3
; CHECK-NEXT:    store <4 x i64> zeroinitializer, ptr [[TMP8]], align 8
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 4
; CHECK-NEXT:    [[TMP9:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
; CHECK-NEXT:    br i1 [[TMP9]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP4:![0-9]+]]
; CHECK:       [[MIDDLE_BLOCK]]:
; CHECK-NEXT:    [[CMP_N:%.*]] = icmp eq i64 [[TMP2]], [[N_VEC]]
; CHECK-NEXT:    br i1 [[CMP_N]], label %[[EXIT:.*]], label %[[SCALAR_PH]]
; CHECK:       [[SCALAR_PH]]:
; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi ptr [ [[TMP4]], %[[MIDDLE_BLOCK]] ], [ [[A]], %[[ENTRY]] ]
; CHECK-NEXT:    [[BC_RESUME_VAL2:%.*]] = phi ptr [ [[TMP6]], %[[MIDDLE_BLOCK]] ], [ [[ADD_PTR11]], %[[ENTRY]] ]
; CHECK-NEXT:    br label %[[LOOP:.*]]
; CHECK:       [[LOOP]]:
; CHECK-NEXT:    [[IV_1:%.*]] = phi ptr [ [[IV_1_NEXT:%.*]], %[[LOOP]] ], [ [[BC_RESUME_VAL]], %[[SCALAR_PH]] ]
; CHECK-NEXT:    [[IV_2:%.*]] = phi ptr [ [[IV_2_NEXT:%.*]], %[[LOOP]] ], [ [[BC_RESUME_VAL2]], %[[SCALAR_PH]] ]
; CHECK-NEXT:    [[IV_2_NEXT]] = getelementptr nusw i8, ptr [[IV_2]], i64 -8
; CHECK-NEXT:    [[IV_1_NEXT]] = getelementptr i8, ptr [[IV_1]], i64 -8
; CHECK-NEXT:    [[TMP10:%.*]] = load i64, ptr [[IV_2_NEXT]], align 8
; CHECK-NEXT:    store i64 0, ptr [[IV_1]], align 8
; CHECK-NEXT:    [[CMP_NOT_I_I_I_I:%.*]] = icmp eq ptr [[END]], [[IV_2]]
; CHECK-NEXT:    br i1 [[CMP_NOT_I_I_I_I]], label %[[EXIT]], label %[[LOOP]], !llvm.loop [[LOOP5:![0-9]+]]
; CHECK:       [[EXIT]]:
; CHECK-NEXT:    ret void
;
entry:
  %A.int = ptrtoint ptr %A to i64
  %B.cast = ptrtoint ptr %B to i64
  %ptr.sub = sub i64 %A.int, %B.cast
  %add.ptr11 = getelementptr i8, ptr %B, i64 %ptr.sub
  br label %loop

loop:
  %iv.1 = phi ptr [ %iv.1.next, %loop ], [ %A, %entry ]
  %iv.2 = phi ptr [ %iv.2.next, %loop ], [ %add.ptr11, %entry ]
  %iv.2.next = getelementptr nusw i8, ptr %iv.2, i64 -8
  %iv.1.next = getelementptr i8, ptr %iv.1, i64 -8
  %2 = load i64, ptr %iv.2.next, align 8
  store i64 0, ptr %iv.1, align 8
  %cmp.not.i.i.i.i = icmp eq ptr %end, %iv.2
  br i1 %cmp.not.i.i.i.i, label %exit, label %loop

exit:
  ret void
}
;.
; CHECK: [[LOOP0]] = distinct !{[[LOOP0]], [[META1:![0-9]+]], [[META2:![0-9]+]]}
; CHECK: [[META1]] = !{!"llvm.loop.isvectorized", i32 1}
; CHECK: [[META2]] = !{!"llvm.loop.unroll.runtime.disable"}
; CHECK: [[LOOP3]] = distinct !{[[LOOP3]], [[META1]]}
; CHECK: [[LOOP4]] = distinct !{[[LOOP4]], [[META1]], [[META2]]}
; CHECK: [[LOOP5]] = distinct !{[[LOOP5]], [[META2]], [[META1]]}
;.