llvm/test/Transforms/LoopVectorize/AArch64/low-trip-count-runtime-check.ll

; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --check-globals none --version 6
; RUN: opt -S < %s -passes=loop-vectorize -mtriple=aarch64-unknown-linux-gnu | FileCheck %s

define void @low_trip_count_12_needs_runtime_check(ptr %dst, ptr %a, ptr %b, i32 %m) {
; CHECK-LABEL: define void @low_trip_count_12_needs_runtime_check(
; CHECK-SAME: ptr [[DST:%.*]], ptr [[A:%.*]], ptr [[B:%.*]], i32 [[M:%.*]]) {
; CHECK-NEXT:  [[ENTRY:.*:]]
; CHECK-NEXT:    [[B3:%.*]] = ptrtoaddr ptr [[B]] to i64
; CHECK-NEXT:    [[A2:%.*]] = ptrtoaddr ptr [[A]] to i64
; CHECK-NEXT:    [[DST1:%.*]] = ptrtoaddr ptr [[DST]] to i64
; CHECK-NEXT:    [[N:%.*]] = mul nuw i32 [[M]], 4
; CHECK-NEXT:    [[GUARD:%.*]] = icmp eq i32 [[N]], 0
; CHECK-NEXT:    br i1 [[GUARD]], label %[[EXIT:.*]], label %[[LOOP_PH:.*]]
; CHECK:       [[LOOP_PH]]:
; CHECK-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i32 [[N]], 8
; CHECK-NEXT:    br i1 [[MIN_ITERS_CHECK]], label %[[SCALAR_PH:.*]], label %[[VECTOR_MEMCHECK:.*]], !prof [[PROF0:![0-9]+]]
; CHECK:       [[VECTOR_MEMCHECK]]:
; CHECK-NEXT:    [[TMP0:%.*]] = sub i64 [[DST1]], [[A2]]
; CHECK-NEXT:    [[TMP1:%.*]] = sub i64 [[TMP0]], 1
; CHECK-NEXT:    [[DIFF_CHECK:%.*]] = icmp ult i64 [[TMP1]], 15
; CHECK-NEXT:    [[TMP2:%.*]] = sub i64 [[DST1]], [[B3]]
; CHECK-NEXT:    [[TMP3:%.*]] = sub i64 [[TMP2]], 1
; CHECK-NEXT:    [[DIFF_CHECK4:%.*]] = icmp ult i64 [[TMP3]], 15
; CHECK-NEXT:    [[CONFLICT_RDX:%.*]] = or i1 [[DIFF_CHECK]], [[DIFF_CHECK4]]
; CHECK-NEXT:    br i1 [[CONFLICT_RDX]], label %[[SCALAR_PH]], label %[[VECTOR_PH:.*]], !prof [[PROF0]]
; 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:    [[TMP4:%.*]] = getelementptr inbounds i32, ptr [[A]], i32 [[INDEX]]
; CHECK-NEXT:    [[WIDE_LOAD:%.*]] = load <4 x i32>, ptr [[TMP4]], align 4
; CHECK-NEXT:    [[TMP5:%.*]] = getelementptr inbounds i32, ptr [[B]], i32 [[INDEX]]
; CHECK-NEXT:    [[WIDE_LOAD5:%.*]] = load <4 x i32>, ptr [[TMP5]], align 4
; CHECK-NEXT:    [[TMP6:%.*]] = add <4 x i32> [[WIDE_LOAD]], [[WIDE_LOAD5]]
; CHECK-NEXT:    [[TMP7:%.*]] = getelementptr inbounds i32, ptr [[DST]], i32 [[INDEX]]
; CHECK-NEXT:    store <4 x i32> [[TMP6]], ptr [[TMP7]], align 4
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i32 [[INDEX]], 4
; CHECK-NEXT:    [[TMP8:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N]]
; CHECK-NEXT:    br i1 [[TMP8]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !prof [[PROF1:![0-9]+]], !llvm.loop [[LOOP2:![0-9]+]]
; CHECK:       [[MIDDLE_BLOCK]]:
; CHECK-NEXT:    br label %[[EXIT_LOOPEXIT:.*]]
; CHECK:       [[SCALAR_PH]]:
; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i32 [ 0, %[[LOOP_PH]] ], [ 0, %[[VECTOR_MEMCHECK]] ]
; CHECK-NEXT:    br label %[[LOOP:.*]]
; CHECK:       [[LOOP]]:
; CHECK-NEXT:    [[IV:%.*]] = phi i32 [ [[BC_RESUME_VAL]], %[[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], %[[LOOP]] ]
; CHECK-NEXT:    [[A_GEP:%.*]] = getelementptr inbounds i32, ptr [[A]], i32 [[IV]]
; CHECK-NEXT:    [[AV:%.*]] = load i32, ptr [[A_GEP]], align 4
; CHECK-NEXT:    [[B_GEP:%.*]] = getelementptr inbounds i32, ptr [[B]], i32 [[IV]]
; CHECK-NEXT:    [[BV:%.*]] = load i32, ptr [[B_GEP]], align 4
; CHECK-NEXT:    [[SUM:%.*]] = add i32 [[AV]], [[BV]]
; CHECK-NEXT:    [[D_GEP:%.*]] = getelementptr inbounds i32, ptr [[DST]], i32 [[IV]]
; CHECK-NEXT:    store i32 [[SUM]], ptr [[D_GEP]], align 4
; CHECK-NEXT:    [[IV_NEXT]] = add nuw nsw i32 [[IV]], 1
; CHECK-NEXT:    [[CMP:%.*]] = icmp ult i32 [[IV_NEXT]], [[N]]
; CHECK-NEXT:    br i1 [[CMP]], label %[[LOOP]], label %[[EXIT_LOOPEXIT]], !prof [[PROF6:![0-9]+]], !llvm.loop [[LOOP7:![0-9]+]]
; CHECK:       [[EXIT_LOOPEXIT]]:
; CHECK-NEXT:    br label %[[EXIT]]
; CHECK:       [[EXIT]]:
; CHECK-NEXT:    ret void
;
entry:
  %n = mul nuw i32 %m, 4
  %guard = icmp eq i32 %n, 0
  br i1 %guard, label %exit, label %loop.ph

loop.ph:
  br label %loop

loop:
  %iv = phi i32 [ 0, %loop.ph ], [ %iv.next, %loop ]
  %a.gep = getelementptr inbounds i32, ptr %a, i32 %iv
  %av = load i32, ptr %a.gep, align 4
  %b.gep = getelementptr inbounds i32, ptr %b, i32 %iv
  %bv = load i32, ptr %b.gep, align 4
  %sum = add i32 %av, %bv
  %d.gep = getelementptr inbounds i32, ptr %dst, i32 %iv
  store i32 %sum, ptr %d.gep, align 4
  %iv.next = add nuw nsw i32 %iv, 1
  %cmp = icmp ult i32 %iv.next, %n
  br i1 %cmp, label %loop, label %exit, !prof !0

exit:
  ret void
}

!0 = !{!"branch_weights", i32 110, i32 10}