; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --check-globals none --version 6
; RUN: opt -passes=loop-vectorize -force-vector-width=4 -S %s | FileCheck %s

; B[i] = A[i % 4]
define void @bounded_load_with_store(ptr %A, ptr %B) {
; CHECK-LABEL: define void @bounded_load_with_store(
; CHECK-SAME: ptr [[A:%.*]], ptr [[B:%.*]]) {
; CHECK-NEXT:  [[ENTRY:.*]]:
; CHECK-NEXT:    br label %[[LOOP:.*]]
; CHECK:       [[LOOP]]:
; CHECK-NEXT:    [[IV:%.*]] = phi i32 [ 0, %[[ENTRY]] ], [ [[IV_NEXT:%.*]], %[[LOOP]] ]
; CHECK-NEXT:    [[BOUNDED:%.*]] = urem i32 [[IV]], 4
; CHECK-NEXT:    [[GEP_A:%.*]] = getelementptr inbounds i32, ptr [[A]], i32 [[BOUNDED]]
; CHECK-NEXT:    [[LV:%.*]] = load i32, ptr [[GEP_A]], align 4
; CHECK-NEXT:    [[GEP_B:%.*]] = getelementptr inbounds i32, ptr [[B]], i32 [[IV]]
; CHECK-NEXT:    store i32 [[LV]], ptr [[GEP_B]], align 4
; CHECK-NEXT:    [[IV_NEXT]] = add nuw nsw i32 [[IV]], 1
; CHECK-NEXT:    [[COND:%.*]] = icmp eq i32 [[IV_NEXT]], 128
; CHECK-NEXT:    br i1 [[COND]], label %[[EXIT:.*]], label %[[LOOP]]
; CHECK:       [[EXIT]]:
; CHECK-NEXT:    ret void
;
entry:
  br label %loop

loop:
  %iv = phi i32 [ 0, %entry ], [ %iv.next, %loop ]
  %bounded = urem i32 %iv, 4
  %gep.A = getelementptr inbounds i32, ptr %A, i32 %bounded
  %lv = load i32, ptr %gep.A
  %gep.B = getelementptr inbounds i32, ptr %B, i32 %iv
  store i32 %lv, ptr %gep.B
  %iv.next = add nuw nsw i32 %iv, 1
  %cond = icmp eq i32 %iv.next, 128
  br i1 %cond, label %exit, label %loop

exit:
  ret void
}

; B[i % 4] = A[i] + 10
define void @store_bounded_index(ptr %A, ptr %B) {
; CHECK-LABEL: define void @store_bounded_index(
; CHECK-SAME: ptr [[A:%.*]], ptr [[B:%.*]]) {
; CHECK-NEXT:  [[ENTRY:.*]]:
; CHECK-NEXT:    br label %[[LOOP:.*]]
; CHECK:       [[LOOP]]:
; CHECK-NEXT:    [[IV:%.*]] = phi i32 [ 0, %[[ENTRY]] ], [ [[IV_NEXT:%.*]], %[[LOOP]] ]
; CHECK-NEXT:    [[GEP_A:%.*]] = getelementptr inbounds i32, ptr [[A]], i32 [[IV]]
; CHECK-NEXT:    [[LV:%.*]] = load i32, ptr [[GEP_A]], align 4
; CHECK-NEXT:    [[ADD:%.*]] = add i32 [[LV]], 10
; CHECK-NEXT:    [[BOUNDED:%.*]] = urem i32 [[IV]], 4
; CHECK-NEXT:    [[GEP_B:%.*]] = getelementptr inbounds i32, ptr [[B]], i32 [[BOUNDED]]
; CHECK-NEXT:    store i32 [[ADD]], ptr [[GEP_B]], align 4
; CHECK-NEXT:    [[IV_NEXT]] = add nuw nsw i32 [[IV]], 1
; CHECK-NEXT:    [[COND:%.*]] = icmp eq i32 [[IV_NEXT]], 128
; CHECK-NEXT:    br i1 [[COND]], label %[[EXIT:.*]], label %[[LOOP]]
; CHECK:       [[EXIT]]:
; CHECK-NEXT:    ret void
;
entry:
  br label %loop

loop:
  %iv = phi i32 [ 0, %entry ], [ %iv.next, %loop ]
  %gep.A = getelementptr inbounds i32, ptr %A, i32 %iv
  %lv = load i32, ptr %gep.A
  %add = add i32 %lv, 10
  %bounded = urem i32 %iv, 4
  %gep.B = getelementptr inbounds i32, ptr %B, i32 %bounded
  store i32 %add, ptr %gep.B
  %iv.next = add nuw nsw i32 %iv, 1
  %cond = icmp eq i32 %iv.next, 128
  br i1 %cond, label %exit, label %loop

exit:
  ret void
}

define void @bounded_load_with_store_optsize(ptr noalias %A, ptr noalias %B) optsize {
; CHECK-LABEL: define void @bounded_load_with_store_optsize(
; CHECK-SAME: ptr noalias [[A:%.*]], ptr noalias [[B:%.*]]) #[[ATTR0:[0-9]+]] {
; CHECK-NEXT:  [[ENTRY:.*:]]
; CHECK-NEXT:    br label %[[VECTOR_PH:.*]]
; CHECK:       [[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 1, i32 2, i32 3>, %[[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], %[[VECTOR_BODY]] ]
; CHECK-NEXT:    [[TMP0:%.*]] = urem <4 x i32> [[VEC_IND]], splat (i32 4)
; CHECK-NEXT:    [[TMP1:%.*]] = extractelement <4 x i32> [[TMP0]], i64 0
; CHECK-NEXT:    [[TMP5:%.*]] = getelementptr inbounds i32, ptr [[A]], i32 [[TMP1]]
; CHECK-NEXT:    [[TMP2:%.*]] = extractelement <4 x i32> [[TMP0]], i64 1
; CHECK-NEXT:    [[TMP6:%.*]] = getelementptr inbounds i32, ptr [[A]], i32 [[TMP2]]
; CHECK-NEXT:    [[TMP3:%.*]] = extractelement <4 x i32> [[TMP0]], i64 2
; CHECK-NEXT:    [[TMP7:%.*]] = getelementptr inbounds i32, ptr [[A]], i32 [[TMP3]]
; CHECK-NEXT:    [[TMP4:%.*]] = extractelement <4 x i32> [[TMP0]], i64 3
; CHECK-NEXT:    [[TMP8:%.*]] = getelementptr inbounds i32, ptr [[A]], i32 [[TMP4]]
; CHECK-NEXT:    [[TMP9:%.*]] = load i32, ptr [[TMP5]], align 4
; CHECK-NEXT:    [[TMP10:%.*]] = load i32, ptr [[TMP6]], align 4
; CHECK-NEXT:    [[TMP11:%.*]] = load i32, ptr [[TMP7]], align 4
; CHECK-NEXT:    [[TMP12:%.*]] = load i32, ptr [[TMP8]], align 4
; CHECK-NEXT:    [[TMP13:%.*]] = insertelement <4 x i32> poison, i32 [[TMP9]], i32 0
; CHECK-NEXT:    [[TMP14:%.*]] = insertelement <4 x i32> [[TMP13]], i32 [[TMP10]], i32 1
; CHECK-NEXT:    [[TMP15:%.*]] = insertelement <4 x i32> [[TMP14]], i32 [[TMP11]], i32 2
; CHECK-NEXT:    [[TMP16:%.*]] = insertelement <4 x i32> [[TMP15]], i32 [[TMP12]], i32 3
; CHECK-NEXT:    [[TMP17:%.*]] = getelementptr inbounds i32, ptr [[B]], i32 [[INDEX]]
; CHECK-NEXT:    store <4 x i32> [[TMP16]], ptr [[TMP17]], align 4
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i32 [[INDEX]], 4
; CHECK-NEXT:    [[VEC_IND_NEXT]] = add nuw nsw <4 x i32> [[VEC_IND]], splat (i32 4)
; CHECK-NEXT:    [[TMP18:%.*]] = icmp eq i32 [[INDEX_NEXT]], 128
; CHECK-NEXT:    br i1 [[TMP18]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
; CHECK:       [[MIDDLE_BLOCK]]:
; CHECK-NEXT:    br label %[[EXIT:.*]]
; CHECK:       [[EXIT]]:
; CHECK-NEXT:    ret void
;
entry:
  br label %loop

loop:
  %iv = phi i32 [ 0, %entry ], [ %iv.next, %loop ]
  %bounded = urem i32 %iv, 4
  %gep.A = getelementptr inbounds i32, ptr %A, i32 %bounded
  %lv = load i32, ptr %gep.A
  %gep.B = getelementptr inbounds i32, ptr %B, i32 %iv
  store i32 %lv, ptr %gep.B
  %iv.next = add nuw nsw i32 %iv, 1
  %cond = icmp eq i32 %iv.next, 128
  br i1 %cond, label %exit, label %loop

exit:
  ret void
}

; As above but A and B may alias. Under optsize the runtime alias check cannot
; be emitted, so the loop must stay scalar.
define void @bounded_load_with_store_optsize_may_alias(ptr %A, ptr %B) optsize {
; CHECK-LABEL: define void @bounded_load_with_store_optsize_may_alias(
; CHECK-SAME: ptr [[A:%.*]], ptr [[B:%.*]]) #[[ATTR0]] {
; CHECK-NEXT:  [[ENTRY:.*]]:
; CHECK-NEXT:    br label %[[LOOP:.*]]
; CHECK:       [[LOOP]]:
; CHECK-NEXT:    [[IV:%.*]] = phi i32 [ 0, %[[ENTRY]] ], [ [[IV_NEXT:%.*]], %[[LOOP]] ]
; CHECK-NEXT:    [[BOUNDED:%.*]] = urem i32 [[IV]], 4
; CHECK-NEXT:    [[GEP_A:%.*]] = getelementptr inbounds i32, ptr [[A]], i32 [[BOUNDED]]
; CHECK-NEXT:    [[LV:%.*]] = load i32, ptr [[GEP_A]], align 4
; CHECK-NEXT:    [[GEP_B:%.*]] = getelementptr inbounds i32, ptr [[B]], i32 [[IV]]
; CHECK-NEXT:    store i32 [[LV]], ptr [[GEP_B]], align 4
; CHECK-NEXT:    [[IV_NEXT]] = add nuw nsw i32 [[IV]], 1
; CHECK-NEXT:    [[COND:%.*]] = icmp eq i32 [[IV_NEXT]], 128
; CHECK-NEXT:    br i1 [[COND]], label %[[EXIT:.*]], label %[[LOOP]]
; CHECK:       [[EXIT]]:
; CHECK-NEXT:    ret void
;
entry:
  br label %loop

loop:
  %iv = phi i32 [ 0, %entry ], [ %iv.next, %loop ]
  %bounded = urem i32 %iv, 4
  %gep.A = getelementptr inbounds i32, ptr %A, i32 %bounded
  %lv = load i32, ptr %gep.A
  %gep.B = getelementptr inbounds i32, ptr %B, i32 %iv
  store i32 %lv, ptr %gep.B
  %iv.next = add nuw nsw i32 %iv, 1
  %cond = icmp eq i32 %iv.next, 128
  br i1 %cond, label %exit, label %loop

exit:
  ret void
}