llvm/test/Transforms/LoopVectorize/epilog-vectorization-fixed-order-recurrences.ll

; 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=8 -enable-epilogue-vectorization -epilogue-vectorization-force-VF=4 -S %s | FileCheck %s


define void @dead_for(ptr %a, i64 %N) {
; CHECK-LABEL: define void @dead_for(
; CHECK-SAME: ptr [[A:%.*]], i64 [[N:%.*]]) {
; CHECK-NEXT:  [[VECTOR_MAIN_LOOP_ITER_CHECK:.*]]:
; CHECK-NEXT:    [[MIN_ITERS_CHECK1:%.*]] = icmp ult i64 [[N]], 8
; CHECK-NEXT:    br i1 [[MIN_ITERS_CHECK1]], label %[[VEC_EPILOG_PH:.*]], label %[[VECTOR_PH:.*]]
; CHECK:       [[VECTOR_PH]]:
; CHECK-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[N]], 8
; CHECK-NEXT:    [[N_VEC:%.*]] = sub i64 [[N]], [[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:    [[TMP0:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[INDEX]]
; CHECK-NEXT:    [[WIDE_LOAD:%.*]] = load <8 x i64>, ptr [[TMP0]], align 4
; CHECK-NEXT:    [[TMP1:%.*]] = add <8 x i64> [[WIDE_LOAD]], splat (i64 10)
; CHECK-NEXT:    store <8 x i64> [[TMP1]], ptr [[TMP0]], align 4
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 8
; CHECK-NEXT:    [[TMP2:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
; CHECK-NEXT:    br i1 [[TMP2]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
; CHECK:       [[MIDDLE_BLOCK]]:
; CHECK-NEXT:    [[VECTOR_RECUR_EXTRACT:%.*]] = extractelement <8 x i64> [[WIDE_LOAD]], i64 7
; CHECK-NEXT:    [[CMP_N:%.*]] = icmp eq i64 [[N]], [[N_VEC]]
; CHECK-NEXT:    br i1 [[CMP_N]], label %[[EXIT:.*]], label %[[VEC_EPILOG_PH]]
; CHECK:       [[VEC_EPILOG_PH]]:
; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], %[[MIDDLE_BLOCK]] ], [ 0, %[[VECTOR_MAIN_LOOP_ITER_CHECK]] ]
; CHECK-NEXT:    [[SCALAR_RECUR_INIT:%.*]] = phi i64 [ [[VECTOR_RECUR_EXTRACT]], %[[MIDDLE_BLOCK]] ], [ 99, %[[VECTOR_MAIN_LOOP_ITER_CHECK]] ]
; CHECK-NEXT:    br label %[[LOOP:.*]]
; CHECK:       [[LOOP]]:
; CHECK-NEXT:    [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], %[[VEC_EPILOG_PH]] ], [ [[IV_NEXT:%.*]], %[[LOOP]] ]
; CHECK-NEXT:    [[FOR:%.*]] = phi i64 [ [[SCALAR_RECUR_INIT]], %[[VEC_EPILOG_PH]] ], [ [[L:%.*]], %[[LOOP]] ]
; CHECK-NEXT:    [[GEP:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[IV]]
; CHECK-NEXT:    [[L]] = load i64, ptr [[GEP]], align 4
; CHECK-NEXT:    [[ADD:%.*]] = add i64 [[L]], 10
; CHECK-NEXT:    store i64 [[ADD]], ptr [[GEP]], align 4
; CHECK-NEXT:    [[IV_NEXT]] = add nuw nsw i64 [[IV]], 1
; CHECK-NEXT:    [[EC:%.*]] = icmp eq i64 [[IV_NEXT]], [[N]]
; 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 [ 0, %entry ], [ %iv.next, %loop ]
  %for = phi i64 [ 99, %entry ], [ %l, %loop ]
  %gep = getelementptr inbounds i64, ptr %a, i64 %iv
  %l = load i64, ptr %gep
  %add = add i64 %l, 10
  store i64 %add, ptr %gep
  %iv.next = add nuw nsw i64 %iv, 1
  %ec = icmp eq i64 %iv.next, %N
  br i1 %ec, label %exit, label %loop

exit:
  ret void
}

define i64 @for_phi_used_in_loop_and_live_out(ptr %a, i64 %N) {
; CHECK-LABEL: define i64 @for_phi_used_in_loop_and_live_out(
; CHECK-SAME: ptr [[A:%.*]], i64 [[N:%.*]]) {
; CHECK-NEXT:  [[ENTRY:.*]]:
; CHECK-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 [[N]], 8
; CHECK-NEXT:    br i1 [[MIN_ITERS_CHECK]], label %[[SCALAR_PH:.*]], label %[[VECTOR_PH:.*]]
; CHECK:       [[VECTOR_PH]]:
; CHECK-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[N]], 8
; CHECK-NEXT:    [[N_VEC:%.*]] = sub i64 [[N]], [[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:    [[VECTOR_RECUR:%.*]] = phi <8 x i64> [ <i64 poison, i64 poison, i64 poison, i64 poison, i64 poison, i64 poison, i64 poison, i64 99>, %[[VECTOR_PH]] ], [ [[WIDE_LOAD:%.*]], %[[VECTOR_BODY]] ]
; CHECK-NEXT:    [[TMP0:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[INDEX]]
; CHECK-NEXT:    [[WIDE_LOAD]] = load <8 x i64>, ptr [[TMP0]], align 4
; CHECK-NEXT:    [[TMP1:%.*]] = shufflevector <8 x i64> [[VECTOR_RECUR]], <8 x i64> [[WIDE_LOAD]], <8 x i32> <i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14>
; CHECK-NEXT:    store <8 x i64> [[TMP1]], ptr [[TMP0]], align 4
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 8
; CHECK-NEXT:    [[TMP2:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
; CHECK-NEXT:    br i1 [[TMP2]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP4:![0-9]+]]
; CHECK:       [[MIDDLE_BLOCK]]:
; CHECK-NEXT:    [[VECTOR_RECUR_EXTRACT_FOR_PHI:%.*]] = extractelement <8 x i64> [[WIDE_LOAD]], i64 6
; CHECK-NEXT:    [[VECTOR_RECUR_EXTRACT:%.*]] = extractelement <8 x i64> [[WIDE_LOAD]], i64 7
; CHECK-NEXT:    [[CMP_N:%.*]] = icmp eq i64 [[N]], [[N_VEC]]
; CHECK-NEXT:    br i1 [[CMP_N]], label %[[EXIT:.*]], label %[[SCALAR_PH]]
; CHECK:       [[SCALAR_PH]]:
; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], %[[MIDDLE_BLOCK]] ], [ 0, %[[ENTRY]] ]
; CHECK-NEXT:    [[SCALAR_RECUR_INIT:%.*]] = phi i64 [ [[VECTOR_RECUR_EXTRACT]], %[[MIDDLE_BLOCK]] ], [ 99, %[[ENTRY]] ]
; CHECK-NEXT:    br label %[[LOOP:.*]]
; CHECK:       [[LOOP]]:
; CHECK-NEXT:    [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], %[[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], %[[LOOP]] ]
; CHECK-NEXT:    [[FOR:%.*]] = phi i64 [ [[SCALAR_RECUR_INIT]], %[[SCALAR_PH]] ], [ [[L:%.*]], %[[LOOP]] ]
; CHECK-NEXT:    [[GEP:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[IV]]
; CHECK-NEXT:    [[L]] = load i64, ptr [[GEP]], align 4
; CHECK-NEXT:    [[ADD:%.*]] = add i64 [[L]], 10
; CHECK-NEXT:    store i64 [[FOR]], ptr [[GEP]], align 4
; CHECK-NEXT:    [[IV_NEXT]] = add nuw nsw i64 [[IV]], 1
; CHECK-NEXT:    [[EC:%.*]] = icmp eq i64 [[IV_NEXT]], [[N]]
; CHECK-NEXT:    br i1 [[EC]], label %[[EXIT]], label %[[LOOP]], !llvm.loop [[LOOP5:![0-9]+]]
; CHECK:       [[EXIT]]:
; CHECK-NEXT:    [[FOR_LCSSA:%.*]] = phi i64 [ [[FOR]], %[[LOOP]] ], [ [[VECTOR_RECUR_EXTRACT_FOR_PHI]], %[[MIDDLE_BLOCK]] ]
; CHECK-NEXT:    [[L_LCSSA:%.*]] = phi i64 [ [[L]], %[[LOOP]] ], [ [[VECTOR_RECUR_EXTRACT]], %[[MIDDLE_BLOCK]] ]
; CHECK-NEXT:    [[RES:%.*]] = add i64 [[L_LCSSA]], [[FOR_LCSSA]]
; CHECK-NEXT:    ret i64 [[RES]]
;
entry:
  br label %loop

loop:
  %iv = phi i64 [ 0, %entry ], [ %iv.next, %loop ]
  %for = phi i64 [ 99, %entry ], [ %l, %loop ]
  %gep = getelementptr inbounds i64, ptr %a, i64 %iv
  %l = load i64, ptr %gep
  %add = add i64 %l, 10
  store i64 %for, ptr %gep
  %iv.next = add nuw nsw i64 %iv, 1
  %ec = icmp eq i64 %iv.next, %N
  br i1 %ec, label %exit, label %loop

exit:
  %res = add i64 %l, %for
  ret i64 %res
}

define i64 @for_phi_not_used_in_loop_and_live_out(ptr %a, i64 %N) {
; CHECK-LABEL: define i64 @for_phi_not_used_in_loop_and_live_out(
; CHECK-SAME: ptr [[A:%.*]], i64 [[N:%.*]]) {
; CHECK-NEXT:  [[VECTOR_MAIN_LOOP_ITER_CHECK:.*]]:
; CHECK-NEXT:    [[MIN_ITERS_CHECK1:%.*]] = icmp ult i64 [[N]], 8
; CHECK-NEXT:    br i1 [[MIN_ITERS_CHECK1]], label %[[VEC_EPILOG_PH:.*]], label %[[VECTOR_PH:.*]]
; CHECK:       [[VECTOR_PH]]:
; CHECK-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[N]], 8
; CHECK-NEXT:    [[N_VEC:%.*]] = sub i64 [[N]], [[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:    [[TMP0:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[INDEX]]
; CHECK-NEXT:    [[WIDE_LOAD:%.*]] = load <8 x i64>, ptr [[TMP0]], align 4
; CHECK-NEXT:    [[TMP1:%.*]] = add <8 x i64> [[WIDE_LOAD]], splat (i64 10)
; CHECK-NEXT:    store <8 x i64> [[TMP1]], ptr [[TMP0]], align 4
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 8
; CHECK-NEXT:    [[TMP2:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
; CHECK-NEXT:    br i1 [[TMP2]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP6:![0-9]+]]
; CHECK:       [[MIDDLE_BLOCK]]:
; CHECK-NEXT:    [[VECTOR_RECUR_EXTRACT_FOR_PHI:%.*]] = extractelement <8 x i64> [[WIDE_LOAD]], i64 6
; CHECK-NEXT:    [[VECTOR_RECUR_EXTRACT:%.*]] = extractelement <8 x i64> [[WIDE_LOAD]], i64 7
; CHECK-NEXT:    [[CMP_N:%.*]] = icmp eq i64 [[N]], [[N_VEC]]
; CHECK-NEXT:    br i1 [[CMP_N]], label %[[EXIT:.*]], label %[[VEC_EPILOG_PH]]
; CHECK:       [[VEC_EPILOG_PH]]:
; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], %[[MIDDLE_BLOCK]] ], [ 0, %[[VECTOR_MAIN_LOOP_ITER_CHECK]] ]
; CHECK-NEXT:    [[SCALAR_RECUR_INIT:%.*]] = phi i64 [ [[VECTOR_RECUR_EXTRACT]], %[[MIDDLE_BLOCK]] ], [ 99, %[[VECTOR_MAIN_LOOP_ITER_CHECK]] ]
; CHECK-NEXT:    br label %[[LOOP:.*]]
; CHECK:       [[LOOP]]:
; CHECK-NEXT:    [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], %[[VEC_EPILOG_PH]] ], [ [[IV_NEXT:%.*]], %[[LOOP]] ]
; CHECK-NEXT:    [[FOR:%.*]] = phi i64 [ [[SCALAR_RECUR_INIT]], %[[VEC_EPILOG_PH]] ], [ [[L:%.*]], %[[LOOP]] ]
; CHECK-NEXT:    [[GEP:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[IV]]
; CHECK-NEXT:    [[L]] = load i64, ptr [[GEP]], align 4
; CHECK-NEXT:    [[ADD:%.*]] = add i64 [[L]], 10
; CHECK-NEXT:    store i64 [[ADD]], ptr [[GEP]], align 4
; CHECK-NEXT:    [[IV_NEXT]] = add nuw nsw i64 [[IV]], 1
; CHECK-NEXT:    [[EC:%.*]] = icmp eq i64 [[IV_NEXT]], [[N]]
; CHECK-NEXT:    br i1 [[EC]], label %[[EXIT]], label %[[LOOP]], !llvm.loop [[LOOP7:![0-9]+]]
; CHECK:       [[EXIT]]:
; CHECK-NEXT:    [[FOR_LCSSA:%.*]] = phi i64 [ [[FOR]], %[[LOOP]] ], [ [[VECTOR_RECUR_EXTRACT_FOR_PHI]], %[[MIDDLE_BLOCK]] ]
; CHECK-NEXT:    [[L_LCSSA:%.*]] = phi i64 [ [[L]], %[[LOOP]] ], [ [[VECTOR_RECUR_EXTRACT]], %[[MIDDLE_BLOCK]] ]
; CHECK-NEXT:    [[RES:%.*]] = add i64 [[L_LCSSA]], [[FOR_LCSSA]]
; CHECK-NEXT:    ret i64 [[RES]]
;
entry:
  br label %loop

loop:
  %iv = phi i64 [ 0, %entry ], [ %iv.next, %loop ]
  %for = phi i64 [ 99, %entry ], [ %l, %loop ]
  %gep = getelementptr inbounds i64, ptr %a, i64 %iv
  %l = load i64, ptr %gep
  %add = add i64 %l, 10
  store i64 %add, ptr %gep
  %iv.next = add nuw nsw i64 %iv, 1
  %ec = icmp eq i64 %iv.next, %N
  br i1 %ec, label %exit, label %loop

exit:
  %res = add i64 %l, %for
  ret i64 %res
}