test/Transforms/LoopVectorize/alias-mask-data-tail-folding-style.ll

; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --check-globals none --version 6
; RUN: opt -S -passes=loop-vectorize -force-target-supports-masked-memory-ops -force-partial-aliasing-vectorization -tail-folding-policy=prefer-fold-tail -force-tail-folding-style=data -force-vector-width=4  %s | FileCheck %s

define void @test(ptr %src, ptr %dst, i32 %n) {
; CHECK-LABEL: define void @test(
; CHECK-SAME: ptr [[SRC:%.*]], ptr [[DST:%.*]], i32 [[N:%.*]]) {
; CHECK-NEXT:  [[ENTRY:.*:]]
; CHECK-NEXT:    [[SRC2:%.*]] = ptrtoaddr ptr [[SRC]] to i64
; CHECK-NEXT:    [[DST1:%.*]] = ptrtoaddr ptr [[DST]] to i64
; CHECK-NEXT:    [[UMAX3:%.*]] = call i32 @llvm.umax.i32(i32 [[N]], i32 1)
; CHECK-NEXT:    br label %[[VECTOR_SCEVCHECK:.*]]
; CHECK:       [[VECTOR_SCEVCHECK]]:
; CHECK-NEXT:    [[UMAX:%.*]] = call i32 @llvm.umax.i32(i32 [[N]], i32 1)
; CHECK-NEXT:    [[TMP0:%.*]] = add i32 [[UMAX]], -1
; CHECK-NEXT:    [[TMP1:%.*]] = icmp slt i32 [[TMP0]], 0
; CHECK-NEXT:    br i1 [[TMP1]], label %[[SCALAR_PH:.*]], label %[[VECTOR_CLAMPED_VF_CHECK:.*]]
; CHECK:       [[VECTOR_CLAMPED_VF_CHECK]]:
; CHECK-NEXT:    [[ALIAS_MASK:%.*]] = call <4 x i1> @llvm.loop.dependence.war.mask.v4i1.i64(i64 [[SRC2]], i64 [[DST1]], i64 4)
; CHECK-NEXT:    [[TMP3:%.*]] = zext <4 x i1> [[ALIAS_MASK]] to <4 x i64>
; CHECK-NEXT:    [[NUM_ACTIVE_LANES:%.*]] = call i64 @llvm.vector.reduce.add.v4i64(<4 x i64> [[TMP3]])
; CHECK-NEXT:    [[TMP7:%.*]] = trunc i64 [[NUM_ACTIVE_LANES]] to i32
; CHECK-NEXT:    [[VF_IS_SCALAR:%.*]] = icmp ule i32 [[TMP7]], 1
; CHECK-NEXT:    [[TMP8:%.*]] = sub i32 -1, [[UMAX3]]
; CHECK-NEXT:    [[VF_STEP_OVERFLOW:%.*]] = icmp ult i32 [[TMP8]], [[TMP7]]
; CHECK-NEXT:    [[TMP9:%.*]] = or i1 [[VF_IS_SCALAR]], [[VF_STEP_OVERFLOW]]
; CHECK-NEXT:    br i1 [[TMP9]], label %[[SCALAR_PH]], label %[[VECTOR_PH:.*]]
; CHECK:       [[VECTOR_PH]]:
; CHECK-NEXT:    [[TMP10:%.*]] = sub i32 [[TMP7]], 1
; CHECK-NEXT:    [[N_RND_UP:%.*]] = add i32 [[UMAX3]], [[TMP10]]
; CHECK-NEXT:    [[N_MOD_VF:%.*]] = urem i32 [[N_RND_UP]], [[TMP7]]
; CHECK-NEXT:    [[N_VEC:%.*]] = sub i32 [[N_RND_UP]], [[N_MOD_VF]]
; CHECK-NEXT:    br label %[[VECTOR_BODY:.*]]
; CHECK:       [[VECTOR_BODY]]:
; CHECK-NEXT:    [[INDEX1:%.*]] = phi i32 [ 0, %[[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], %[[VECTOR_BODY]] ]
; CHECK-NEXT:    [[ACTIVE_LANE_MASK:%.*]] = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i32(i32 [[INDEX1]], i32 [[UMAX3]])
; CHECK-NEXT:    [[MASK:%.*]] = and <4 x i1> [[ACTIVE_LANE_MASK]], [[ALIAS_MASK]]
; CHECK-NEXT:    [[TMP12:%.*]] = getelementptr i32, ptr [[SRC]], i32 [[INDEX1]]
; CHECK-NEXT:    [[WIDE_MASKED_LOAD:%.*]] = call <4 x i32> @llvm.masked.load.v4i32.p0(ptr align 4 [[TMP12]], <4 x i1> [[MASK]], <4 x i32> poison)
; CHECK-NEXT:    [[TMP13:%.*]] = getelementptr i32, ptr [[DST]], i32 [[INDEX1]]
; CHECK-NEXT:    call void @llvm.masked.store.v4i32.p0(<4 x i32> [[WIDE_MASKED_LOAD]], ptr align 4 [[TMP13]], <4 x i1> [[MASK]])
; CHECK-NEXT:    [[INDEX_NEXT]] = add i32 [[INDEX1]], [[TMP7]]
; CHECK-NEXT:    [[TMP15:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC]]
; CHECK-NEXT:    br i1 [[TMP15]], 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 i32 [ 0, %[[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], %[[LOOP]] ]
; CHECK-NEXT:    [[IV_NEXT]] = add i32 [[IV]], 1
; CHECK-NEXT:    [[GEP_SRC:%.*]] = getelementptr i32, ptr [[SRC]], i32 [[IV]]
; CHECK-NEXT:    [[VAL:%.*]] = load i32, ptr [[GEP_SRC]], align 4
; CHECK-NEXT:    [[GEP_DST:%.*]] = getelementptr i32, ptr [[DST]], i32 [[IV]]
; CHECK-NEXT:    store i32 [[VAL]], ptr [[GEP_DST]], align 4
; CHECK-NEXT:    [[COND:%.*]] = icmp ult i32 [[IV_NEXT]], [[N]]
; CHECK-NEXT:    br i1 [[COND]], label %[[LOOP]], label %[[EXIT]], !llvm.loop [[LOOP3:![0-9]+]]
; CHECK:       [[EXIT]]:
; CHECK-NEXT:    ret void
;
entry:
  br label %loop

loop:
  %iv = phi i32 [ 0, %entry ], [ %iv.next, %loop ]
  %iv.next = add i32 %iv, 1
  %gep.src = getelementptr i32, ptr %src, i32 %iv
  %val = load i32, ptr %gep.src, align 4
  %gep.dst = getelementptr i32, ptr %dst, i32 %iv
  store i32 %val, ptr %gep.dst, align 4
  %cond = icmp ult i32 %iv.next, %n
  br i1 %cond, label %loop, label %exit

exit:
  ret void
}