File: regalloc.py

package info (click to toggle)
pypy3 7.3.19%2Bdfsg-2
  • links: PTS, VCS
  • area: main
  • in suites: forky, sid, trixie
  • size: 212,236 kB
  • sloc: python: 2,098,316; ansic: 540,565; sh: 21,462; asm: 14,419; cpp: 4,451; makefile: 4,209; objc: 761; xml: 530; exp: 499; javascript: 314; pascal: 244; lisp: 45; csh: 12; awk: 4
file content (1420 lines) | stat: -rw-r--r-- 54,408 bytes parent folder | download | duplicates (2)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
#!/usr/bin/env python

from rpython.jit.backend.llsupport import rewrite
from rpython.jit.backend.llsupport.descr import CallDescr
from rpython.jit.backend.llsupport.gcmap import allocate_gcmap
from rpython.jit.backend.llsupport.jump import remap_frame_layout_mixed
from rpython.jit.backend.llsupport.regalloc import (
    BaseRegalloc, FrameManager, RegisterManager, TempVar,
    compute_vars_longevity)
from rpython.jit.backend.riscv import registers as r
from rpython.jit.backend.riscv.arch import (
    JITFRAME_FIXED_SIZE, SHAMT_MAX, SINT12_IMM_MAX, SINT12_IMM_MIN, XLEN)
from rpython.jit.backend.riscv.instruction_util import (
    COND_BEQ, COND_BGE, COND_BGEU, COND_BLT, COND_BLTU, COND_BNE, COND_INVALID,
    check_imm_arg, get_negated_branch_inst)
from rpython.jit.backend.riscv.locations import (
    FloatImmLocation, ImmLocation, StackLocation, get_fp_offset)
from rpython.jit.backend.riscv.opassembler import asm_comp_operations
from rpython.jit.codewriter import longlong
from rpython.jit.codewriter.effectinfo import EffectInfo
from rpython.jit.metainterp.history import (
    Const, ConstFloat, ConstInt, ConstPtr, FLOAT, INT, REF, TargetToken)
from rpython.jit.metainterp.resoperation import rop
from rpython.rlib.rarithmetic import r_uint
from rpython.rtyper.lltypesystem import lltype, rffi
from rpython.rtyper.lltypesystem.lloperation import llop


class TempInt(TempVar):
    type = INT

    def __repr__(self):
        return "<TempInt at %s>" % (id(self),)


class TempPtr(TempVar):
    type = REF

    def __repr__(self):
        return "<TempPtr at %s>" % (id(self),)


class TempFloat(TempVar):
    type = FLOAT

    def __repr__(self):
        return "<TempFloat at %s>" % (id(self),)


class RISCVFrameManager(FrameManager):
    def __init__(self, base_ofs):
        FrameManager.__init__(self)
        self.base_ofs = base_ofs

    def frame_pos(self, i, box_type):
        return StackLocation(i, get_fp_offset(self.base_ofs, i), box_type)

    @staticmethod
    def frame_size(type):
        # Returns the number of stack slots that are required to hold the
        # variable of `type`.
        #
        # In RV64IMAFD, both integer, float, and pointer types are 64-bit.  The
        # stack slot size is defined as 64-bit.  Thus, we return 1 regardless
        # the `type`.
        #
        # TODO: In RV32IMAFD, the integer and pointer types are 32-bit but the
        # float type is 64-bit (i.e. represent Python `float` value with double
        # precision floating point).  The stack slot size should be defined as
        # 32-bit.  We should return 2 when the `type` is `FLOAT` and return 1
        # otherwise.
        #
        # Note: `FrameManager.get_new_loc()` guarantees that the position is
        # always a multiple of size.
        return 1

    @staticmethod
    def get_loc_index(loc):
        assert loc.is_stack()
        return loc.position


class RISCVRegisterManager(RegisterManager):
    # Registers assigned to temp boxes must not be spilled before
    # `free_temp_vars()` is called.
    #
    # This ensures that the constant register allocated by `return_constant`
    # is not spilled by subsequent `make_sure_var_in_reg` (Note: passing
    # `op.getarglist()` as the `forbidden_vars` argument is not sufficient
    # because `get_scratch_reg` associates the register with `TempInt`,
    # `TempPtr`, or `TempFloat` instead of `ConstInt`, `ConstPtr`, or
    # `ConstFloat`.)
    FORBID_TEMP_BOXES = True


class CoreRegisterManager(RISCVRegisterManager):
    all_regs = r.allocatable_registers
    box_types = None
    no_lower_byte_regs = all_regs
    save_around_call_regs = r.caller_saved_registers
    frame_reg = r.jfp

    def __init__(self, longevity, frame_manager=None, assembler=None):
        RISCVRegisterManager.__init__(self, longevity, frame_manager,
                                      assembler)

    def call_result_location(self, v):
        return r.x10

    def convert_to_imm(self, c):
        # Note: `c.value` below have different types.
        if isinstance(c, ConstInt):
            val = rffi.cast(lltype.Signed, c.value)
        else:
            assert isinstance(c, ConstPtr)
            val = rffi.cast(lltype.Signed, c.value)
        return ImmLocation(val)

    def return_constant(self, v, forbidden_vars=[], selected_reg=None):
        self._check_type(v)
        if isinstance(v, Const):
            if isinstance(v, ConstInt):
                reg_type = INT
            else:
                assert isinstance(v, ConstPtr)
                reg_type = REF

            if not v.nonnull() and selected_reg is None:
                return r.zero

            loc = self.get_scratch_reg(reg_type, forbidden_vars,
                                       selected_reg=selected_reg)
            immvalue = self.convert_to_imm(v)
            self.assembler.load_imm(loc, immvalue)
            return loc
        return RISCVRegisterManager.return_constant(self, v, forbidden_vars,
                                                    selected_reg)

    def get_scratch_reg(self, type=INT, forbidden_vars=[], selected_reg=None):
        assert type == INT or type == REF
        box = None
        if type == INT:
            box = TempInt()
        else:
            box = TempPtr()
        reg = self.force_allocate_reg(box, forbidden_vars=forbidden_vars,
                                      selected_reg=selected_reg)
        self.temp_boxes.append(box)
        return reg


class FloatRegisterManager(RISCVRegisterManager):
    all_regs = r.allocatable_fp_registers
    box_types = [FLOAT]
    save_around_call_regs = r.caller_saved_fp_registers

    def __init__(self, longevity, frame_manager=None, assembler=None):
        RISCVRegisterManager.__init__(self, longevity, frame_manager,
                                      assembler)

    def call_result_location(self, v):
        return r.f10

    def convert_to_imm(self, c):
        imm_bits = longlong.extract_bits(c.getfloatstorage())
        return FloatImmLocation(imm_bits)

    def return_constant(self, v, forbidden_vars=[], selected_reg=None):
        self._check_type(v)
        if isinstance(v, Const):
            assert isinstance(v, ConstFloat)
            loc = self.get_scratch_reg(FLOAT, forbidden_vars,
                                       selected_reg=selected_reg)
            immvalue = self.convert_to_imm(v)
            self.assembler.load_imm(loc, immvalue)
            return loc
        return RISCVRegisterManager.return_constant(self, v, forbidden_vars,
                                                    selected_reg)

    def get_scratch_reg(self, type=FLOAT, forbidden_vars=[], selected_reg=None):
        assert type == FLOAT
        box = TempFloat()
        reg = self.force_allocate_reg(box, forbidden_vars=forbidden_vars,
                                      selected_reg=selected_reg)
        self.temp_boxes.append(box)
        return reg


def check_imm_box(arg):
    if isinstance(arg, ConstInt):
        return check_imm_arg(arg.getint())
    return False

def check_negative_imm_box(arg):
    if isinstance(arg, ConstInt):
        return check_imm_arg(-arg.getint())
    return False

def check_plus_one_imm_box(arg):
    if isinstance(arg, ConstInt):
        return check_imm_arg(arg.getint() + 1)
    return False

def check_uint_le_rhs_imm_arg(imm):
    return imm >= SINT12_IMM_MIN - 1 and imm <= SINT12_IMM_MAX - 1

def check_uint_le_rhs_imm_box(arg):
    if isinstance(arg, ConstInt):
        return check_uint_le_rhs_imm_arg(arg.getint())
    return False

def check_shamt_imm_arg(imm):
    return imm >= 0 and imm <= SHAMT_MAX

def check_shamt_imm_box(arg):
    if isinstance(arg, ConstInt):
        return check_shamt_imm_arg(arg.getint())
    return False


class Regalloc(BaseRegalloc):
    def __init__(self, assembler):
        self.cpu = assembler.cpu
        self.assembler = assembler
        self.frame_manager = None
        self.rm = None
        self.fprm = None
        self.jump_target_descr = None
        self.final_jump_op = None

    def _prepare(self, inputargs, operations, allgcrefs):
        cpu = self.cpu

        self.fm = RISCVFrameManager(cpu.get_baseofs_of_frame_field())
        self.frame_manager = self.fm
        operations = cpu.gc_ll_descr.rewrite_assembler(cpu, operations,
                                                       allgcrefs)

        longevity = compute_vars_longevity(inputargs, operations)
        self.longevity = longevity

        self.rm = CoreRegisterManager(longevity, self.fm, self.assembler)
        self.fprm = FloatRegisterManager(longevity, self.fm, self.assembler)
        return operations

    def prepare_loop(self, inputargs, operations, looptoken, allgcrefs):
        operations = self._prepare(inputargs, operations, allgcrefs)
        self._set_initial_bindings(inputargs, looptoken)
        self.possibly_free_vars(list(inputargs))
        return operations

    def prepare_bridge(self, inputargs, arglocs, operations, allgcrefs,
                       frame_info):
        operations = self._prepare(inputargs, operations, allgcrefs)
        self._update_bindings(arglocs, inputargs)
        return operations

    def _update_bindings(self, locs, inputargs):
        # Bind the boxes to locations (e.g. registers or stack slots).
        used = {}
        i = 0
        for loc in locs:
            if loc is None:
                loc = r.jfp
            arg = inputargs[i]
            i += 1
            if loc.is_core_reg():
                self.rm.reg_bindings[arg] = loc
                used[loc] = None
            elif loc.is_fp_reg():
                self.fprm.reg_bindings[arg] = loc
                used[loc] = None
            else:
                assert loc.is_stack()
                self.frame_manager.bind(arg, loc)

        # Collect the free registers.
        self.rm.free_regs = []
        for reg in self.rm.all_regs:
            if reg not in used:
                self.rm.free_regs.append(reg)
        self.fprm.free_regs = []
        for reg in self.fprm.all_regs:
            if reg not in used:
                self.fprm.free_regs.append(reg)

        # Note: we need to make a copy of inputargs because possibly_free_vars
        # is also used on op args, which is a non-resizable list.
        self.possibly_free_vars(list(inputargs))

        self.frame_manager.finish_binding()
        self._check_invariants()

    def _prepare_op_int_commutative_binary_op(self, op):
        boxes = op.getarglist()
        a0, a1 = boxes
        imm_a0 = check_imm_box(a0)
        imm_a1 = check_imm_box(a1)
        if not imm_a0 and imm_a1:
            l0 = self.make_sure_var_in_reg(a0, boxes)
            l1 = self.convert_to_imm(a1)
        elif imm_a0 and not imm_a1:
            l1 = self.convert_to_imm(a0)
            l0 = self.make_sure_var_in_reg(a1, boxes)
        else:
            l0 = self.make_sure_var_in_reg(a0, boxes)
            l1 = self.make_sure_var_in_reg(a1, boxes)
        self.possibly_free_vars_for_op(op)
        self.free_temp_vars()
        res = self.force_allocate_reg(op)
        return [l0, l1, res]

    prepare_op_int_add = _prepare_op_int_commutative_binary_op
    prepare_op_int_and = _prepare_op_int_commutative_binary_op
    prepare_op_int_or  = _prepare_op_int_commutative_binary_op
    prepare_op_int_xor = _prepare_op_int_commutative_binary_op
    prepare_op_nursery_ptr_increment = _prepare_op_int_commutative_binary_op

    def prepare_op_int_sub(self, op):
        boxes = op.getarglist()
        a0, a1 = boxes
        # int_sub immediate is lowered into addi with negative immediate, thus
        # we have to check whether the negative immediate can be kept in
        # simm12.
        if check_negative_imm_box(a1):
            l0 = self.make_sure_var_in_reg(a0, boxes)
            l1 = self.convert_to_imm(a1)
        else:
            l0 = self.make_sure_var_in_reg(a0, boxes)
            l1 = self.make_sure_var_in_reg(a1, boxes)
        self.possibly_free_vars_for_op(op)
        self.free_temp_vars()
        res = self.force_allocate_reg(op)
        return [l0, l1, res]

    def _prepare_op_int_mul_op(self, op):
        boxes = op.getarglist()
        a0, a1 = boxes
        l0 = self.make_sure_var_in_reg(a0, boxes)
        l1 = self.make_sure_var_in_reg(a1, boxes)
        self.possibly_free_vars_for_op(op)
        self.free_temp_vars()
        res = self.force_allocate_reg(op)
        return [l0, l1, res]

    prepare_op_int_mul = _prepare_op_int_mul_op
    prepare_op_uint_mul_high = _prepare_op_int_mul_op

    def _prepare_op_int_shift_op(self, op):
        boxes = op.getarglist()
        a0, a1 = boxes
        if check_shamt_imm_box(a1):
            l0 = self.make_sure_var_in_reg(a0, boxes)
            l1 = self.convert_to_imm(a1)
        else:
            l0 = self.make_sure_var_in_reg(a0, boxes)
            l1 = self.make_sure_var_in_reg(a1, boxes)
        self.possibly_free_vars_for_op(op)
        self.free_temp_vars()
        res = self.force_allocate_reg(op)
        return [l0, l1, res]

    prepare_op_int_lshift = _prepare_op_int_shift_op
    prepare_op_int_rshift = _prepare_op_int_shift_op
    prepare_op_uint_rshift = _prepare_op_int_shift_op

    def _gen_prepare_op_int_lt(swap_operands):
        def _prepare_op_int_lt(self, op):
            boxes = op.getarglist()

            if swap_operands:
                a1, a0 = boxes
            else:
                a0, a1 = boxes

            if check_imm_box(a1):
                l0 = self.make_sure_var_in_reg(a0, boxes)
                l1 = self.convert_to_imm(a1)
            else:
                l0 = self.make_sure_var_in_reg(a0, boxes)
                l1 = self.make_sure_var_in_reg(a1, boxes)

            self.possibly_free_vars_for_op(op)
            self.free_temp_vars()
            res = self.force_allocate_reg(op)
            return [l0, l1, res]
        return _prepare_op_int_lt

    prepare_op_int_lt = _gen_prepare_op_int_lt(swap_operands=False)
    prepare_op_int_gt = _gen_prepare_op_int_lt(swap_operands=True)

    prepare_op_uint_lt = prepare_op_int_lt
    prepare_op_uint_gt = prepare_op_int_gt

    prepare_comp_op_int_lt = prepare_op_int_lt
    prepare_comp_op_int_gt = prepare_op_int_gt

    prepare_comp_op_uint_lt = prepare_op_uint_lt
    prepare_comp_op_uint_gt = prepare_op_uint_gt

    def _gen_prepare_op_int_le(swap_operands):
        def _prepare_op_int_le(self, op):
            boxes = op.getarglist()

            if swap_operands:
                a1, a0 = boxes
            else:
                a0, a1 = boxes

            if check_plus_one_imm_box(a1):
                l0 = self.make_sure_var_in_reg(a0, boxes)
                l1 = self.convert_to_imm(a1)
            elif check_imm_box(a0):
                l0 = self.convert_to_imm(a0)
                l1 = self.make_sure_var_in_reg(a1, boxes)
            else:
                l0 = self.make_sure_var_in_reg(a0, boxes)
                l1 = self.make_sure_var_in_reg(a1, boxes)

            self.possibly_free_vars_for_op(op)
            self.free_temp_vars()
            res = self.force_allocate_reg(op)
            return [l0, l1, res]
        return _prepare_op_int_le

    prepare_op_int_le = _gen_prepare_op_int_le(swap_operands=False)
    prepare_op_int_ge = _gen_prepare_op_int_le(swap_operands=True)

    prepare_comp_op_int_le = prepare_op_int_le
    prepare_comp_op_int_ge = prepare_op_int_ge

    def _gen_prepare_op_uint_le(swap_operands):
        def _prepare_op_uint_le(self, op):
            boxes = op.getarglist()

            if swap_operands:
                a1, a0 = boxes
            else:
                a0, a1 = boxes

            if check_uint_le_rhs_imm_box(a1):
                l0 = self.make_sure_var_in_reg(a0, boxes)
                l1 = self.convert_to_imm(a1)
            elif check_imm_box(a0):
                l0 = self.convert_to_imm(a0)
                l1 = self.make_sure_var_in_reg(a1, boxes)
            else:
                l0 = self.make_sure_var_in_reg(a0, boxes)
                l1 = self.make_sure_var_in_reg(a1, boxes)

            self.possibly_free_vars_for_op(op)
            self.free_temp_vars()
            res = self.force_allocate_reg(op)
            return [l0, l1, res]
        return _prepare_op_uint_le

    prepare_op_uint_le = _gen_prepare_op_uint_le(swap_operands=False)
    prepare_op_uint_ge = _gen_prepare_op_uint_le(swap_operands=True)

    prepare_comp_op_uint_le = prepare_op_uint_le
    prepare_comp_op_uint_ge = prepare_op_uint_ge

    def _prepare_op_int_commutative_compare_op(self, op):
        boxes = op.getarglist()
        a0, a1 = boxes
        imm_a0 = check_imm_box(a0)
        imm_a1 = check_imm_box(a1)
        if not imm_a0 and imm_a1:
            l0 = self.make_sure_var_in_reg(a0, boxes)
            l1 = self.convert_to_imm(a1)
        elif imm_a0 and not imm_a1:
            l1 = self.convert_to_imm(a0)
            l0 = self.make_sure_var_in_reg(a1, boxes)
        else:
            l0 = self.make_sure_var_in_reg(a0, boxes)
            l1 = self.make_sure_var_in_reg(a1, boxes)
        self.possibly_free_vars_for_op(op)
        self.free_temp_vars()
        res = self.force_allocate_reg(op)
        return [l0, l1, res]

    prepare_op_int_eq = _prepare_op_int_commutative_compare_op
    prepare_op_int_ne = _prepare_op_int_commutative_compare_op

    prepare_comp_op_int_eq = prepare_op_int_eq
    prepare_comp_op_int_ne = prepare_op_int_ne

    prepare_op_ptr_eq = prepare_op_instance_ptr_eq = prepare_op_int_eq
    prepare_op_ptr_ne = prepare_op_instance_ptr_ne = prepare_op_int_ne

    prepare_comp_op_ptr_eq = prepare_op_int_eq
    prepare_comp_op_ptr_ne = prepare_op_int_ne

    def _prepare_op_unary_op(self, op):
        boxes = op.getarglist()
        a0 = boxes[0]
        l0 = self.make_sure_var_in_reg(a0, boxes)
        self.possibly_free_vars_for_op(op)
        self.free_temp_vars()
        res = self.force_allocate_reg(op)
        return [l0, res]

    prepare_op_int_is_true = _prepare_op_unary_op
    prepare_op_int_neg = _prepare_op_unary_op
    prepare_op_int_invert = _prepare_op_unary_op
    prepare_op_int_is_zero = _prepare_op_unary_op
    prepare_op_int_force_ge_zero = _prepare_op_unary_op

    prepare_comp_op_int_is_true = prepare_op_int_is_true
    prepare_comp_op_int_is_zero = prepare_op_int_is_zero

    def prepare_op_int_signext(self, op):
        a0, a1 = op.getarglist()
        l0 = self.loc(a0)
        if l0.is_stack():
            # If a0 is on the stack, we can perform sign extension directly
            # with LB/LH/LW.
            pass
        else:
            # If a0 is not on the stack, we make sure it is in the register and
            # perform sign extension with SLLI/SRAI.
            l0 = self.make_sure_var_in_reg(a0)
        l1 = self.convert_to_imm(a1)  # num_bytes
        self.possibly_free_vars_for_op(op)
        self.free_temp_vars()
        res = self.force_allocate_reg(op)
        return [l0, l1, res]

    def _gen_prepare_op_float_binary_op(swap_operands):
        def _prepare_op_float_binary_op(self, op):
            boxes = op.getarglist()
            if swap_operands:
                a1, a0 = boxes
            else:
                a0, a1 = boxes
            l0 = self.make_sure_var_in_reg(a0, boxes)
            l1 = self.make_sure_var_in_reg(a1, boxes)
            self.possibly_free_vars_for_op(op)
            self.free_temp_vars()
            res = self.force_allocate_reg(op)
            return [l0, l1, res]
        return _prepare_op_float_binary_op

    _prepare_op_float_binary_op = _gen_prepare_op_float_binary_op(False)
    _prepare_op_float_binary_op_swapped = _gen_prepare_op_float_binary_op(True)

    prepare_op_float_add = _prepare_op_float_binary_op
    prepare_op_float_sub = _prepare_op_float_binary_op
    prepare_op_float_mul = _prepare_op_float_binary_op
    prepare_op_float_truediv = _prepare_op_float_binary_op

    prepare_op_float_lt = _prepare_op_float_binary_op
    prepare_op_float_le = _prepare_op_float_binary_op
    prepare_op_float_gt = _prepare_op_float_binary_op_swapped
    prepare_op_float_ge = _prepare_op_float_binary_op_swapped
    prepare_op_float_eq = _prepare_op_float_binary_op
    prepare_op_float_ne = _prepare_op_float_binary_op

    prepare_comp_op_float_lt = prepare_op_float_lt
    prepare_comp_op_float_le = prepare_op_float_le
    prepare_comp_op_float_gt = prepare_op_float_gt
    prepare_comp_op_float_ge = prepare_op_float_ge
    prepare_comp_op_float_eq = prepare_op_float_eq
    prepare_comp_op_float_ne = prepare_op_float_ne

    prepare_op_float_neg = _prepare_op_unary_op
    prepare_op_float_abs = _prepare_op_unary_op
    prepare_op_cast_float_to_int = _prepare_op_unary_op
    prepare_op_cast_int_to_float = _prepare_op_unary_op
    prepare_op_convert_float_bytes_to_longlong = _prepare_op_unary_op
    prepare_op_convert_longlong_bytes_to_float= _prepare_op_unary_op

    def prepare_op_gc_store(self, op):
        boxes = op.getarglist()
        base_loc = self.make_sure_var_in_reg(boxes[0], boxes)
        ofs = boxes[1]
        if check_imm_box(ofs):
            ofs_loc = self.convert_to_imm(ofs)
        else:
            ofs_loc = self.make_sure_var_in_reg(ofs, boxes)
        value_loc = self.make_sure_var_in_reg(boxes[2], boxes)
        size_loc = self.convert_to_imm(boxes[3])
        return [value_loc, base_loc, ofs_loc, size_loc]

    def _prepare_op_gc_load(self, op):
        boxes = op.getarglist()
        base_loc = self.make_sure_var_in_reg(boxes[0], boxes)
        ofs = boxes[1]
        if check_imm_box(ofs):
            ofs_loc = self.convert_to_imm(ofs)
        else:
            ofs_loc = self.make_sure_var_in_reg(ofs, boxes)
        nsize_loc = self.convert_to_imm(boxes[2])  # Negative for "signed"
        self.possibly_free_vars_for_op(op)
        res_loc = self.force_allocate_reg(op)
        return [base_loc, ofs_loc, res_loc, nsize_loc]

    prepare_op_gc_load_i = _prepare_op_gc_load
    prepare_op_gc_load_r = _prepare_op_gc_load
    prepare_op_gc_load_f = _prepare_op_gc_load

    def prepare_op_gc_store_indexed(self, op):
        boxes = op.getarglist()
        base_loc = self.make_sure_var_in_reg(boxes[0], boxes)
        index_loc = self.make_sure_var_in_reg(boxes[1], boxes)
        value_loc = self.make_sure_var_in_reg(boxes[2], boxes)
        assert boxes[3].getint() == 1  # scale
        ofs = boxes[4]
        if check_imm_box(ofs):
            ofs_loc = self.convert_to_imm(ofs)
        else:
            ofs_loc = self.make_sure_var_in_reg(ofs, boxes)
        size_loc = self.convert_to_imm(boxes[5])
        return [value_loc, base_loc, index_loc, ofs_loc, size_loc]

    def _prepare_op_gc_load_indexed(self, op):
        boxes = op.getarglist()
        base_loc = self.make_sure_var_in_reg(boxes[0], boxes)
        index_loc = self.make_sure_var_in_reg(boxes[1], boxes)
        assert boxes[2].getint() == 1  # scale
        ofs = boxes[3]
        if check_imm_box(ofs):
            ofs_loc = self.convert_to_imm(ofs)
        else:
            ofs_loc = self.make_sure_var_in_reg(ofs, boxes)
        nsize_loc = self.convert_to_imm(boxes[4])  # Negative for "signed"
        self.possibly_free_vars_for_op(op)
        res_loc = self.force_allocate_reg(op)
        return [base_loc, index_loc, ofs_loc, res_loc, nsize_loc]

    prepare_op_gc_load_indexed_i = _prepare_op_gc_load_indexed
    prepare_op_gc_load_indexed_r = _prepare_op_gc_load_indexed
    prepare_op_gc_load_indexed_f = _prepare_op_gc_load_indexed

    def _prepare_guard_arglocs(self, op):
        arglocs = [None] * (len(op.getfailargs()) + 1)
        arglocs[0] = ImmLocation(self.frame_manager.get_frame_depth())
        failargs = op.getfailargs()
        for i in range(len(failargs)):
            if failargs[i]:
                arglocs[i + 1] = self.loc(failargs[i])
        return arglocs

    def _prepare_op_guard_unary_op(self, op):
        boxes = op.getarglist()
        a0 = boxes[0]
        l0 = self.make_sure_var_in_reg(a0, boxes)

        # Note[#dont_free_vars]: Do not call `possibly_free_vars_for_op` or
        # `free_temp_vars` here because `_prepare_guard_arglocs` still need the
        # mapping between boxes and locations for `op.getfailargs()`.
        #
        # For example:
        #
        #    [i0, i1, i2]
        #    i3 = int_add(i0, i1)
        #    guard_value(i3, i2, descr=...) [i3]
        #    finish(descr=...)
        #
        # If we call `possibly_free_vars_for_op` here, `i3` will be freed, but
        # we still need it as failargs and the `self.loc(failargs[i])` in
        # `_prepare_guard_arglocs` will associate a random free frame slot,
        # which will result in incorrect result.
        #
        # Just don't call these two functions. The callsite in `assembler.py`
        # will call these two functions for us.

        guard_arglocs = self._prepare_guard_arglocs(op)
        return [l0] + guard_arglocs

    prepare_op_guard_true    = _prepare_op_guard_unary_op
    prepare_op_guard_false   = _prepare_op_guard_unary_op
    prepare_op_guard_nonnull = _prepare_op_guard_unary_op
    prepare_op_guard_isnull  = _prepare_op_guard_unary_op

    def prepare_op_guard_value(self, op):
        boxes = op.getarglist()
        a0, a1 = boxes
        l0 = self.make_sure_var_in_reg(a0, boxes)
        l1 = self.make_sure_var_in_reg(a1, boxes)
        # Note[#dont_free_vars]: Do not call `possibly_free_vars_for_op` or
        # `free_temp_vars`.
        return [l0, l1] + self._prepare_guard_arglocs(op)

    def _gen_prepare_guard_op_guard_bool_op(guard_true):
        def _prepare_guard_op_guard_bool_op(self, op, guard_op):
            boxes = op.getarglist()

            opnum = op.getopnum()
            if (opnum == rop.INT_EQ or opnum == rop.PTR_EQ or
                opnum == rop.INSTANCE_PTR_EQ):
                guard_branch_inst = COND_BEQ
            elif (opnum == rop.INT_NE or opnum == rop.PTR_NE or
                  opnum == rop.INSTANCE_PTR_NE):
                guard_branch_inst = COND_BNE
            elif opnum == rop.INT_LT:
                guard_branch_inst = COND_BLT
            elif opnum == rop.INT_LE:
                boxes = [boxes[1], boxes[0]]
                guard_branch_inst = COND_BGE
            elif opnum == rop.INT_GT:
                boxes = [boxes[1], boxes[0]]
                guard_branch_inst = COND_BLT
            elif opnum == rop.INT_GE:
                guard_branch_inst = COND_BGE
            elif opnum == rop.UINT_LT:
                guard_branch_inst = COND_BLTU
            elif opnum == rop.UINT_LE:
                if check_plus_one_imm_box(boxes[1]):
                    boxes = [boxes[0], ConstInt(boxes[1].getint() + 1)]
                    guard_branch_inst = COND_BLTU
                else:
                    boxes = [boxes[1], boxes[0]]
                    guard_branch_inst = COND_BGEU
            elif opnum == rop.UINT_GT:
                boxes = [boxes[1], boxes[0]]
                guard_branch_inst = COND_BLTU
            elif opnum == rop.UINT_GE:
                if check_plus_one_imm_box(boxes[0]):
                    boxes = [boxes[1], ConstInt(boxes[0].getint() + 1)]
                    guard_branch_inst = COND_BLTU
                else:
                    guard_branch_inst = COND_BGEU
            elif opnum == rop.INT_IS_ZERO:
                boxes = [boxes[0], ConstInt(0)]
                guard_branch_inst = COND_BEQ
            elif opnum == rop.INT_IS_TRUE:
                boxes = [boxes[0], ConstInt(0)]
                guard_branch_inst = COND_BNE
            else:
                assert False, 'unexpected case'

            locs = [self.make_sure_var_in_reg(v, boxes) for v in boxes]

            if not guard_true:
                guard_branch_inst = get_negated_branch_inst(guard_branch_inst)

            # Note[#dont_free_vars]: Do not call `possibly_free_vars_for_op` or
            # `free_temp_vars`.
            guard_arglocs = self._prepare_guard_arglocs(guard_op)

            arglocs = locs + guard_arglocs
            return arglocs, guard_branch_inst
        return _prepare_guard_op_guard_bool_op

    prepare_guard_op_guard_true  = _gen_prepare_guard_op_guard_bool_op(True)
    prepare_guard_op_guard_false = _gen_prepare_guard_op_guard_bool_op(False)

    def _gen_prepare_guard_op_guard_overflow_op(guard_overflow):
        def _prepare_guard_op_guard_overflow_op(self, op, guard_op):
            boxes = op.getarglist()

            # TODO: Optimize ADD/SUB with constant operands.

            # INT_ADD_OVF:
            #   add a2, a0, a1
            #   slt t0, a2, a0
            #   slti t1, a1, 0
            #   beq t0, t1, no_overflow
            # overflow:
            #   j ...
            # no_overflow:

            # INT_SUB_OVF:
            #   sub a2, a0, a1
            #   slt t0, a2, a0
            #   sgtz t1, a1
            #   beq t0, t1, no_overflow
            # overflow:
            #   j ...
            # no_overflow:

            # INT_MUL_OVF:
            #   mul a2, a0, a1
            #   mulh t0, a0, a1
            #   srai t1, a2, 63
            #   beq t0, t1, no_overflow
            # overflow:
            #   j ...
            # no_overflow:

            locs = [self.make_sure_var_in_reg(v, boxes) for v in boxes]
            # Note: Do not call `possibly_free_vars_for_op` or
            # `free_temp_vars`. The result register (a2) overlaps with the
            # lifetime of the arguments (a0/a1) and the lifetime of temporaries
            # (t0/t1).
            locs.append(self.force_allocate_reg(op, boxes))
            locs.append(self.rm.get_scratch_reg(INT, boxes + [op]))  # t0
            locs.append(self.rm.get_scratch_reg(INT, boxes + [op]))  # t1

            guard_branch_inst = COND_BNE if guard_overflow else COND_BEQ

            # Note[#dont_free_vars]: Do not call `possibly_free_vars_for_op` or
            # `free_temp_vars`.
            arglocs = locs + self._prepare_guard_arglocs(guard_op)
            return arglocs, guard_branch_inst
        return _prepare_guard_op_guard_overflow_op

    prepare_guard_op_guard_overflow = \
        _gen_prepare_guard_op_guard_overflow_op(True)
    prepare_guard_op_guard_no_overflow = \
        _gen_prepare_guard_op_guard_overflow_op(False)

    def prepare_op_guard_class(self, op):
        obj_loc = self.make_sure_var_in_reg(op.getarg(0))
        cls_loc = ImmLocation(rffi.cast(lltype.Signed, op.getarg(1).getint()))
        # Note[#dont_free_vars]: Do not call `possibly_free_vars_for_op` or
        # `free_temp_vars`.
        return [obj_loc, cls_loc] + self._prepare_guard_arglocs(op)

    prepare_op_guard_nonnull_class = prepare_op_guard_class

    prepare_op_guard_gc_type = prepare_op_guard_class
    prepare_op_guard_subclass = prepare_op_guard_class

    def prepare_op_guard_is_object(self, op):
        obj_loc = self.make_sure_var_in_reg(op.getarg(0))
        # Note[#dont_free_vars]: Do not call `possibly_free_vars_for_op` or
        # `free_temp_vars`.
        return [obj_loc] + self._prepare_guard_arglocs(op)

    def prepare_op_guard_not_invalidated(self, op):
        return self._prepare_guard_arglocs(op)

    def prepare_op_guard_exception(self, op):
        boxes = op.getarglist()

        expected_exc_tp_loc = self.make_sure_var_in_reg(boxes[0], boxes)
        if op in self.longevity and self.rm.stays_alive(op):
            res_exc_val_loc = self.force_allocate_reg(op, boxes)
        else:
            res_exc_val_loc = None

        arglocs = [expected_exc_tp_loc, res_exc_val_loc]
        return arglocs + self._prepare_guard_arglocs(op)

    def prepare_op_guard_no_exception(self, op):
        return self._prepare_guard_arglocs(op)  # Only failargs

    def prepare_op_save_exception(self, op):
        res = self.force_allocate_reg(op)
        return [res]

    prepare_op_save_exc_class = prepare_op_save_exception

    def prepare_op_restore_exception(self, op):
        boxes = op.getarglist()
        exc_tp_loc = self.make_sure_var_in_reg(boxes[0], boxes)
        exc_val_loc = self.make_sure_var_in_reg(boxes[1], boxes)
        return [exc_tp_loc, exc_val_loc]

    def prepare_op_check_memory_error(self, op):
        l0 = self.make_sure_var_in_reg(op.getarg(0))
        return [l0]

    def _prepare_op_same_as(self, op):
        boxes = op.getarglist()
        a0 = boxes[0]
        if check_imm_box(a0):
            l0 = self.convert_to_imm(a0)
        else:
            l0 = self.make_sure_var_in_reg(a0)
        self.possibly_free_vars_for_op(op)
        self.free_temp_vars()
        res = self.force_allocate_reg(op)
        return [l0, res]

    prepare_op_same_as_i = _prepare_op_same_as
    prepare_op_same_as_r = _prepare_op_same_as
    prepare_op_same_as_f = _prepare_op_same_as
    prepare_op_cast_ptr_to_int = _prepare_op_same_as
    prepare_op_cast_int_to_ptr = _prepare_op_same_as

    def _prepare_op_math_sqrt(self, op):
        # res = call_f(math_sqrt, arg1)
        l1 = self.make_sure_var_in_reg(op.getarg(1))
        self.possibly_free_vars_for_op(op)
        self.free_temp_vars()
        res = self.force_allocate_reg(op)
        return [l1, res]

    def _prepare_op_threadlocalref_get(self, op):
        # res = call_r(threadlocalref_get)
        res = self.force_allocate_reg(op)
        return [res]

    def before_call(self, force_store=[], save_all_regs=False):
        # Spill caller save registers.
        self.rm.before_call(force_store=force_store,
                            save_all_regs=save_all_regs)
        self.fprm.before_call(force_store=force_store,
                              save_all_regs=save_all_regs)

    def after_call(self, v):
        # Bind the return register to the variable `v`, which should hold the
        # returned value.
        if v.type == 'v':
            return
        if v.type == FLOAT:
            return self.fprm.after_call(v)
        else:
            return self.rm.after_call(v)

    def _call(self, op, gc_level):
        # Spill variables that need to be saved around calls:
        # gc_level == 0: callee cannot invoke the GC
        # gc_level == 1: can invoke GC, save all regs that contain pointers
        # gc_level == 2: can force, save all regs
        save_all_regs = gc_level == 2
        self.fprm.before_call(save_all_regs=save_all_regs)
        if gc_level == 1 and self.cpu.gc_ll_descr.gcrootmap:
            save_all_regs = 2
        self.rm.before_call(save_all_regs=save_all_regs)
        resloc = self.after_call(op)
        return resloc

    def _prepare_call(self, op, save_all_regs=False, first_arg_index=1):
        locs = [None] * (op.numargs() + 3)
        calldescr = op.getdescr()
        assert isinstance(calldescr, CallDescr)
        assert len(calldescr.arg_classes) == op.numargs() - first_arg_index

        for i in range(op.numargs()):
            locs[i + 3] = self.loc(op.getarg(i))

        size = calldescr.get_result_size()
        sign = calldescr.is_result_signed()
        sign_loc = ImmLocation(sign)
        locs[1] = ImmLocation(size)
        locs[2] = sign_loc

        effectinfo = calldescr.get_extra_info()
        if save_all_regs:
            gc_level = 2
        elif effectinfo is None or effectinfo.check_can_collect():
            gc_level = 1
        else:
            gc_level = 0

        locs[0] = self._call(op, gc_level)
        return locs

    def _prepare_op_call(self, op):
        calldescr = op.getdescr()
        assert calldescr is not None
        effectinfo = calldescr.get_extra_info()
        if effectinfo is not None:
            oopspecindex = effectinfo.oopspecindex
            if oopspecindex == EffectInfo.OS_MATH_SQRT:
                args = self._prepare_op_math_sqrt(op)
                self.assembler.math_sqrt(op, args)
                return
            elif oopspecindex == EffectInfo.OS_THREADLOCALREF_GET:
                args = self._prepare_op_threadlocalref_get(op)
                self.assembler.threadlocalref_get(op, args)
                return
        return self._prepare_call(op)

    prepare_op_call_i = _prepare_op_call
    prepare_op_call_f = _prepare_op_call
    prepare_op_call_r = _prepare_op_call
    prepare_op_call_n = _prepare_op_call

    def _prepare_op_cond_call(self, op):
        assert 2 <= op.numargs() <= 4 + 2

        func_addr = op.getarg(1)
        assert isinstance(func_addr, Const)

        # Allocate `r.x30` as a scratch reg because `cond_call_slowpath`
        # returns the result with `r.x30` and uses `r.x31` as a scratch
        # register internally.
        self.rm.get_scratch_reg(INT, selected_reg=r.x30)

        # Move function arguments to argument registers.
        _FUNC_ARGS = [r.x10, r.x11, r.x12, r.x13, r.x14, r.x15, r.x16, r.x17]
        allocated_arg_vars = []
        for i in range(2, op.numargs()):
            reg = _FUNC_ARGS[i - 2]
            arg = op.getarg(i)
            assert arg.type != FLOAT
            self.make_sure_var_in_reg(arg, allocated_arg_vars, selected_reg=reg)
            allocated_arg_vars.append(arg)

        # Move the `cond` variable to a register.
        argloc = self.make_sure_var_in_reg(op.getarg(0), allocated_arg_vars)

        if op.type == 'v':
            # A plain COND_CALL.  Calls the function when args[0] is true.
            # Often used just after a comparison operation.
            return [argloc]
        else:
            # COND_CALL_VALUE_I/R.  Calls the function when args[0] is equal to
            # 0 or NULL.  Returns the result from the function call if done, or
            # args[0] if it was not 0/NULL.  Implemented by forcing the result
            # to live in the same register as args[0], and overwriting it if we
            # really do the call.

            # Load the register for the result.  Possibly reuse 'args[0]'.  But
            # the old value of args[0], if it survives, is first spilled away.
            # We can't overwrite any of op.args[2:] here.
            args = op.getarglist()
            resloc = self.rm.force_result_in_reg(op, args[0],
                                                 forbidden_vars=args[2:])
            return [argloc, resloc]

    prepare_op_cond_call = _prepare_op_cond_call
    prepare_op_cond_call_value_i = _prepare_op_cond_call
    prepare_op_cond_call_value_r = _prepare_op_cond_call

    def prepare_guard_op_cond_call(self, cond_op, cond_call_op):
        # Allocate and emit `cond_op`.
        #
        # Note: We do this before `_prepare_op_cond_call` because
        # `_prepare_op_cond_call` may generate code to move register around.
        prepare_cond_op = regalloc_comp_operations[cond_op.getopnum()]
        cond_op_arglocs = prepare_cond_op(self, cond_op)
        asm_comp_operations[cond_op.getopnum()](self.assembler, cond_op,
                                                cond_op_arglocs)

        # Allocate registers for `cond_call_op`.
        op_arglocs = self._prepare_op_cond_call(cond_call_op)
        return (op_arglocs, COND_INVALID)

    def _prepare_op_cond_call_gc_wb(self, op):
        # Allocate an extra scratch register for `_write_barrier_fastpath`.
        tmplocs = [self.rm.get_scratch_reg(INT)]

        # We force all arguments in a reg because it will be needed anyway by
        # the following gc_store. It avoids loading it twice from the memory.
        args = op.getarglist()
        arglocs = [self.make_sure_var_in_reg(op.getarg(i), args)
                   for i in range(op.numargs())]
        return tmplocs + arglocs

    prepare_op_cond_call_gc_wb = _prepare_op_cond_call_gc_wb
    prepare_op_cond_call_gc_wb_array = _prepare_op_cond_call_gc_wb

    def prepare_guard_op_guard_not_forced(self, op, guard_op):
        if rop.is_call_release_gil(op.getopnum()):
            arglocs = self._prepare_call(op, save_all_regs=True,
                                         first_arg_index=2)
        elif rop.is_call_assembler(op.getopnum()):
            locs = self.locs_for_call_assembler(op)
            resloc = self._call(op, gc_level=2)
            arglocs = locs + [resloc]
        else:
            assert rop.is_call_may_force(op.getopnum())
            arglocs = self._prepare_call(op, save_all_regs=True)
        guard_arglocs = self._prepare_guard_arglocs(guard_op)
        return arglocs + guard_arglocs, len(arglocs)

    def prepare_op_guard_not_forced_2(self, op):
        self.rm.before_call(force_store=op.getfailargs(), save_all_regs=True)
        self.fprm.before_call(force_store=op.getfailargs(), save_all_regs=True)
        arglocs = self._prepare_guard_arglocs(op)
        return arglocs

    def prepare_op_call_malloc_nursery(self, op):
        size_box = op.getarg(0)
        assert isinstance(size_box, ConstInt)

        # Allocate registers for `malloc_slowpath`
        self.rm.get_scratch_reg(INT, selected_reg=r.x10)
        self.rm.get_scratch_reg(INT, selected_reg=r.x11)

        gcmap = self.get_gcmap([r.x10, r.x11])

        self.rm.free_temp_vars()
        self.rm.force_allocate_reg(op, selected_reg=r.x10)

        # Emit code for `malloc_nursery`.
        gc_ll_descr = self.cpu.gc_ll_descr
        self.assembler.malloc_cond(
            gc_ll_descr.get_nursery_free_addr(),
            gc_ll_descr.get_nursery_top_addr(), size_box.getint(), gcmap)

    def prepare_op_call_malloc_nursery_varsize_frame(self, op):
        size_box = op.getarg(0)
        assert not isinstance(size_box, ConstInt)  # We can't have a const here

        # Allocate registers for `malloc_slowpath`
        self.rm.get_scratch_reg(INT, selected_reg=r.x10)
        self.rm.get_scratch_reg(INT, selected_reg=r.x11)
        size_loc = self.rm.make_sure_var_in_reg(size_box)

        gcmap = self.get_gcmap([r.x10, r.x11])

        self.rm.possibly_free_vars_for_op(op)
        self.rm.free_temp_vars()
        self.rm.force_allocate_reg(op, selected_reg=r.x10)

        # Emit code for `malloc_nursery_varsize_frame`.
        gc_ll_descr = self.cpu.gc_ll_descr
        self.assembler.malloc_cond_varsize_frame(
            gc_ll_descr.get_nursery_free_addr(),
            gc_ll_descr.get_nursery_top_addr(), size_loc, gcmap)

    def prepare_op_call_malloc_nursery_varsize(self, op):
        # malloc_nursery_varsize(kind, itemsize, length)

        gc_ll_descr = self.cpu.gc_ll_descr
        if not hasattr(gc_ll_descr, 'max_size_of_young_obj'):
            raise Exception('unreachable code')
            # for boehm, this function should never be called

        arraydescr = op.getdescr()
        kind = op.getarg(0).getint()
        itemsize = op.getarg(1).getint()
        length_box = op.getarg(2)
        assert not isinstance(length_box, Const)

        # Allocate registers for `malloc_slowpath`
        self.rm.get_scratch_reg(INT, selected_reg=r.x10)
        self.rm.get_scratch_reg(INT, selected_reg=r.x11)
        if kind == rewrite.FLAG_ARRAY:
            self.rm.get_scratch_reg(INT, selected_reg=r.x12)

        length_loc = self.rm.make_sure_var_in_reg(length_box)

        if kind == rewrite.FLAG_ARRAY:
            gcmap = self.get_gcmap([r.x10, r.x11, r.x12])
        else:
            gcmap = self.get_gcmap([r.x10, r.x11])

        self.rm.possibly_free_vars_for_op(op)
        self.rm.free_temp_vars()
        self.rm.force_allocate_reg(op, selected_reg=r.x10)

        # Emit code for `malloc_nursery_varsize`.
        maxlength = (gc_ll_descr.max_size_of_young_obj - XLEN * 2) / itemsize
        self.assembler.malloc_cond_varsize(
            kind, gc_ll_descr.get_nursery_free_addr(),
            gc_ll_descr.get_nursery_top_addr(), length_loc, itemsize,
            maxlength, gcmap, arraydescr)

    def prepare_op_zero_array(self, op):
        # There are multiple implementations for `zero_array`. Leave the
        # register allocation to `opassembler.py`.
        return []

    def prepare_op_force_token(self, op):
        res = self.force_allocate_reg(op)
        return [res]

    def prepare_op_load_from_gc_table(self, op):
        res = self.force_allocate_reg(op)
        return [res]

    def prepare_op_enter_portal_frame(self, op):
        return []

    def prepare_op_leave_portal_frame(self, op):
        return []

    def prepare_op_keepalive(self, op):
        return []

    def prepare_op_jit_debug(self, op):
        return []

    def prepare_op_increment_debug_counter(self, op):
        boxes = op.getarglist()
        a0 = boxes[0]
        base_loc = self.make_sure_var_in_reg(a0, boxes)
        return [base_loc]

    def compute_hint_frame_locations(self, operations):
        # Fill in the `hint_frame_pos` dictionary of `frame_manager` based on
        # the JUMP at the end of the loop, by looking at where we would like
        # the boxes to be after the jump.
        #
        # Note: This is only an optimization.

        op = operations[-1]
        if op.getopnum() != rop.JUMP:
            return

        self.final_jump_op = op
        descr = op.getdescr()
        assert isinstance(descr, TargetToken)
        if descr._ll_loop_code != 0:
            # Set up `hint_frame_pos` if the target LABEL had been compiled
            # already, i.e. if it belongs to some already-compiled piece of
            # code.
            self._compute_hint_frame_locations_from_descr(descr)
        #else:
        #   The loop ends in a JUMP going back to a LABEL in the same loop.
        #   We cannot fill 'hint_frame_locations' immediately, but we can
        #   wait until the corresponding prepare_op_label() to know where the
        #   we would like the boxes to be after the jump.

    def _compute_hint_frame_locations_from_descr(self, descr):
        arglocs = descr._riscv_arglocs
        jump_op = self.final_jump_op
        assert len(arglocs) == jump_op.numargs()
        for i in range(jump_op.numargs()):
            box = jump_op.getarg(i)
            if not isinstance(box, Const):
                loc = arglocs[i]
                if loc is not None and loc.is_stack():
                    self.frame_manager.hint_frame_pos[box] = (
                        self.frame_manager.get_loc_index(loc))

    def prepare_op_label(self, op):
        descr = op.getdescr()
        assert isinstance(descr, TargetToken)
        inputargs = op.getarglist()
        arglocs = [None] * len(inputargs)

        # We use force_spill() on the boxes that are not going to be really
        # used any more in the loop, but they are kept alive anyway by being in
        # a next LABEL's or a JUMP's argument or fail_args of some guard
        position = self.rm.position
        for arg in inputargs:
            assert not isinstance(arg, Const)
            if self.longevity[arg].is_last_real_use_before(position):
                self.force_spill_var(arg)

        # Dissociate the register and stack slot because the jump op will copy
        # the updated value to the register and such value may be unrelated to
        # the stack slot. If the register is spilled by upcoming ops, we must
        # ensure the register is spilled to other stack slots instead of
        # overwriting to the old stack slot.
        for i in range(len(inputargs)):
            arg = inputargs[i]
            assert not isinstance(arg, Const)
            loc = self.loc(arg)
            arglocs[i] = loc
            if loc.is_core_reg() or loc.is_fp_reg():
                self.frame_manager.mark_as_free(arg)

        descr._riscv_arglocs = arglocs
        descr._riscv_clt = self.assembler.current_clt
        descr._ll_loop_code = self.assembler.mc.get_relative_pos()
        self.assembler.target_tokens_currently_compiling[descr] = None
        self.possibly_free_vars_for_op(op)

        # If the LABEL's descr is precisely the target of the JUMP at the end
        # of the same loop, i.e. if what we are compiling is a single loop that
        # ends up jumping to this LABEL, then we can now provide the hints
        # about the expected position of the spilled variables.
        jump_op = self.final_jump_op
        if jump_op is not None and jump_op.getdescr() is descr:
            self._compute_hint_frame_locations_from_descr(descr)
        return []

    def prepare_op_jump(self, op):
        assert self.jump_target_descr is None
        descr = op.getdescr()
        assert isinstance(descr, TargetToken)
        self.jump_target_descr = descr
        arglocs = descr._riscv_arglocs

        # If we are jumping to another loop/bridge, check the frame
        # depth before we jump.
        self.assembler.check_frame_depth_before_jump(self.jump_target_descr)

        # Core registers
        src_core_locs = []
        dst_core_locs = []
        scratch_core_reg = r.x31

        # Floating point registers
        src_fp_locs = []
        dst_fp_locs = []
        scratch_fp_reg = r.f31

        # Build the four lists
        for i in range(op.numargs()):
            box = op.getarg(i)
            src_loc = self.loc(box)
            dst_loc = arglocs[i]
            if box.type != FLOAT:
                src_core_locs.append(src_loc)
                dst_core_locs.append(dst_loc)
            else:
                src_fp_locs.append(src_loc)
                dst_fp_locs.append(dst_loc)

        remap_frame_layout_mixed(self.assembler, src_core_locs, dst_core_locs,
                                 scratch_core_reg, src_fp_locs, dst_fp_locs,
                                 scratch_fp_reg, XLEN)
        return []

    def prepare_op_finish(self, op):
        if op.numargs() == 1:
            loc = self.make_sure_var_in_reg(op.getarg(0))
            locs = [loc]
        else:
            locs = []
        return locs

    def loc(self, var):
        if var.type == FLOAT:
            return self.fprm.loc(var)
        else:
            return self.rm.loc(var)

    def make_sure_var_in_reg(self, var, forbidden_vars=[],
                             selected_reg=None, need_lower_byte=False):
        if var.type == FLOAT:
            return self.fprm.make_sure_var_in_reg(var, forbidden_vars,
                                                  selected_reg, need_lower_byte)
        else:
            return self.rm.make_sure_var_in_reg(var, forbidden_vars,
                                                selected_reg, need_lower_byte)

    def force_allocate_reg(self, var, forbidden_vars=[], selected_reg=None):
        if var.type == FLOAT:
            return self.fprm.force_allocate_reg(var, forbidden_vars,
                                                selected_reg)
        else:
            return self.rm.force_allocate_reg(var, forbidden_vars, selected_reg)

    def possibly_free_var(self, var):
        if var.type == FLOAT:
            self.fprm.possibly_free_var(var)
        else:
            self.rm.possibly_free_var(var)

    def possibly_free_vars_for_op(self, op):
        for i in range(op.numargs()):
            var = op.getarg(i)
            if var is not None:
                self.possibly_free_var(var)
        if op.is_guard():
            self.possibly_free_vars(op.getfailargs())

    def possibly_free_vars(self, vars):
        for var in vars:
            if var is not None:
                self.possibly_free_var(var)

    def force_spill_var(self, var):
        if var.type == FLOAT:
            self.fprm.force_spill_var(var)
        else:
            self.rm.force_spill_var(var)

    def free_temp_vars(self):
        self.rm.free_temp_vars()
        self.fprm.free_temp_vars()

    def _check_invariants(self):
        self.rm._check_invariants()
        self.fprm._check_invariants()

    def convert_to_imm(self, value):
        if isinstance(value, ConstInt):
            return self.rm.convert_to_imm(value)
        else:
            assert isinstance(value, ConstFloat)
            return self.fprm.convert_to_imm(value)

    def position(self):
        return self.rm.position

    def next_instruction(self):
        self.rm.next_instruction()
        self.fprm.next_instruction()

    def get_gcmap(self, forbidden_regs=[], noregs=False):
        frame_depth = self.frame_manager.get_frame_depth()
        gcmap = allocate_gcmap(self.assembler,
                               frame_depth, JITFRAME_FIXED_SIZE)
        for box, loc in self.rm.reg_bindings.iteritems():
            if loc in forbidden_regs:
                continue
            if box.type == REF and self.rm.is_still_alive(box):
                assert not noregs
                assert loc.is_core_reg()
                val = self.cpu.all_reg_indexes[loc.value]
                gcmap[val // XLEN // 8] |= r_uint(1) << (val % (XLEN * 8))
        for box, loc in self.frame_manager.bindings.iteritems():
            if box.type == REF and self.rm.is_still_alive(box):
                assert loc.is_stack()
                val = loc.position + JITFRAME_FIXED_SIZE
                gcmap[val // XLEN // 8] |= r_uint(1) << (val % (XLEN * 8))
        return gcmap

    def get_final_frame_depth(self):
        return self.frame_manager.get_frame_depth()


def not_implemented(self, op):
    llop.debug_print(lltype.Void,
                     '[riscv/regalloc] %s not implemented' % op.getopname())
    raise NotImplementedError(op)

def not_implemented_guard_op(self, op, prevop):
    llop.debug_print(lltype.Void,
                     '[riscv/regalloc] %s not implemented' % op.getopname())
    raise NotImplementedError(op)

def not_implemented_comp_op(self, op):
    llop.debug_print(lltype.Void,
                     '[riscv/regalloc] %s not implemented' % op.getopname())
    raise NotImplementedError(op)

regalloc_operations = [not_implemented] * (rop._LAST + 1)
regalloc_guard_operations = [not_implemented_guard_op] * (rop._LAST + 1)
regalloc_comp_operations = [not_implemented_comp_op] * (rop._LAST + 1)

for key, value in rop.__dict__.items():
    key = key.lower()
    if key.startswith('_'):
        continue
    method_name = 'prepare_op_%s' % key
    if hasattr(Regalloc, method_name):
        func = getattr(Regalloc, method_name).im_func
        regalloc_operations[value] = func
    method_name = 'prepare_guard_op_%s' % key
    if hasattr(Regalloc, method_name):
        func = getattr(Regalloc, method_name).im_func
        regalloc_guard_operations[value] = func
    method_name = 'prepare_comp_op_%s' % key
    if hasattr(Regalloc, method_name):
        func = getattr(Regalloc, method_name).im_func
        regalloc_comp_operations[value] = func