summaryrefslogtreecommitdiffstats
path: root/arch/arm/net/bpf_jit_32.c
blob: 6e8b7161303936908b3b2b7adfced5d17de379ce (plain)
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
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
1637
1638
1639
1640
1641
1642
1643
1644
1645
1646
1647
1648
1649
1650
1651
1652
1653
1654
1655
1656
1657
1658
1659
1660
1661
1662
1663
1664
1665
1666
1667
1668
1669
1670
1671
1672
1673
1674
1675
1676
1677
1678
1679
1680
1681
1682
1683
1684
1685
1686
1687
1688
1689
1690
1691
1692
1693
1694
1695
1696
1697
1698
1699
1700
1701
1702
1703
1704
1705
1706
1707
1708
1709
1710
1711
1712
1713
1714
1715
1716
1717
1718
1719
1720
1721
1722
1723
1724
1725
1726
1727
1728
1729
1730
1731
1732
1733
1734
1735
1736
1737
1738
1739
1740
1741
1742
1743
1744
1745
1746
1747
1748
1749
1750
1751
1752
1753
1754
1755
1756
1757
1758
1759
1760
1761
1762
1763
1764
1765
1766
1767
1768
1769
1770
1771
1772
1773
1774
1775
1776
1777
1778
1779
1780
1781
1782
1783
1784
1785
1786
1787
1788
1789
1790
1791
1792
1793
1794
1795
1796
1797
1798
1799
1800
1801
1802
1803
1804
1805
1806
1807
1808
1809
1810
1811
1812
1813
1814
1815
1816
1817
1818
1819
1820
1821
1822
1823
1824
1825
1826
1827
1828
1829
1830
1831
1832
1833
1834
1835
1836
1837
1838
1839
1840
1841
1842
1843
1844
1845
1846
1847
1848
1849
1850
1851
1852
1853
1854
1855
1856
1857
1858
1859
1860
1861
1862
1863
1864
1865
/*
 * Just-In-Time compiler for eBPF filters on 32bit ARM
 *
 * Copyright (c) 2017 Shubham Bansal <illusionist.neo@gmail.com>
 * Copyright (c) 2011 Mircea Gherzan <mgherzan@gmail.com>
 *
 * This program is free software; you can redistribute it and/or modify it
 * under the terms of the GNU General Public License as published by the
 * Free Software Foundation; version 2 of the License.
 */

#include <linux/bpf.h>
#include <linux/bitops.h>
#include <linux/compiler.h>
#include <linux/errno.h>
#include <linux/filter.h>
#include <linux/netdevice.h>
#include <linux/string.h>
#include <linux/slab.h>
#include <linux/if_vlan.h>

#include <asm/cacheflush.h>
#include <asm/hwcap.h>
#include <asm/opcodes.h>

#include "bpf_jit_32.h"

/*
 * eBPF prog stack layout:
 *
 *                         high
 * original ARM_SP =>     +-----+
 *                        |     | callee saved registers
 *                        +-----+ <= (BPF_FP + SCRATCH_SIZE)
 *                        | ... | eBPF JIT scratch space
 * eBPF fp register =>    +-----+
 *   (BPF_FP)             | ... | eBPF prog stack
 *                        +-----+
 *                        |RSVD | JIT scratchpad
 * current ARM_SP =>      +-----+ <= (BPF_FP - STACK_SIZE + SCRATCH_SIZE)
 *                        |     |
 *                        | ... | Function call stack
 *                        |     |
 *                        +-----+
 *                          low
 *
 * The callee saved registers depends on whether frame pointers are enabled.
 * With frame pointers (to be compliant with the ABI):
 *
 *                                high
 * original ARM_SP =>     +------------------+ \
 *                        |        pc        | |
 * current ARM_FP =>      +------------------+ } callee saved registers
 *                        |r4-r8,r10,fp,ip,lr| |
 *                        +------------------+ /
 *                                low
 *
 * Without frame pointers:
 *
 *                                high
 * original ARM_SP =>     +------------------+
 *                        | r4-r8,r10,fp,lr  | callee saved registers
 * current ARM_FP =>      +------------------+
 *                                low
 *
 * When popping registers off the stack at the end of a BPF function, we
 * reference them via the current ARM_FP register.
 */
#define CALLEE_MASK	(1 << ARM_R4 | 1 << ARM_R5 | 1 << ARM_R6 | \
			 1 << ARM_R7 | 1 << ARM_R8 | 1 << ARM_R10 | \
			 1 << ARM_FP)
#define CALLEE_PUSH_MASK (CALLEE_MASK | 1 << ARM_LR)
#define CALLEE_POP_MASK  (CALLEE_MASK | 1 << ARM_PC)

#define STACK_OFFSET(k)	(k)
#define TMP_REG_1	(MAX_BPF_JIT_REG + 0)	/* TEMP Register 1 */
#define TMP_REG_2	(MAX_BPF_JIT_REG + 1)	/* TEMP Register 2 */
#define TCALL_CNT	(MAX_BPF_JIT_REG + 2)	/* Tail Call Count */

#define FLAG_IMM_OVERFLOW	(1 << 0)

/*
 * Map eBPF registers to ARM 32bit registers or stack scratch space.
 *
 * 1. First argument is passed using the arm 32bit registers and rest of the
 * arguments are passed on stack scratch space.
 * 2. First callee-saved argument is mapped to arm 32 bit registers and rest
 * arguments are mapped to scratch space on stack.
 * 3. We need two 64 bit temp registers to do complex operations on eBPF
 * registers.
 *
 * As the eBPF registers are all 64 bit registers and arm has only 32 bit
 * registers, we have to map each eBPF registers with two arm 32 bit regs or
 * scratch memory space and we have to build eBPF 64 bit register from those.
 *
 */
static const u8 bpf2a32[][2] = {
	/* return value from in-kernel function, and exit value from eBPF */
	[BPF_REG_0] = {ARM_R1, ARM_R0},
	/* arguments from eBPF program to in-kernel function */
	[BPF_REG_1] = {ARM_R3, ARM_R2},
	/* Stored on stack scratch space */
	[BPF_REG_2] = {STACK_OFFSET(0), STACK_OFFSET(4)},
	[BPF_REG_3] = {STACK_OFFSET(8), STACK_OFFSET(12)},
	[BPF_REG_4] = {STACK_OFFSET(16), STACK_OFFSET(20)},
	[BPF_REG_5] = {STACK_OFFSET(24), STACK_OFFSET(28)},
	/* callee saved registers that in-kernel function will preserve */
	[BPF_REG_6] = {ARM_R5, ARM_R4},
	/* Stored on stack scratch space */
	[BPF_REG_7] = {STACK_OFFSET(32), STACK_OFFSET(36)},
	[BPF_REG_8] = {STACK_OFFSET(40), STACK_OFFSET(44)},
	[BPF_REG_9] = {STACK_OFFSET(48), STACK_OFFSET(52)},
	/* Read only Frame Pointer to access Stack */
	[BPF_REG_FP] = {STACK_OFFSET(56), STACK_OFFSET(60)},
	/* Temporary Register for internal BPF JIT, can be used
	 * for constant blindings and others.
	 */
	[TMP_REG_1] = {ARM_R7, ARM_R6},
	[TMP_REG_2] = {ARM_R10, ARM_R8},
	/* Tail call count. Stored on stack scratch space. */
	[TCALL_CNT] = {STACK_OFFSET(64), STACK_OFFSET(68)},
	/* temporary register for blinding constants.
	 * Stored on stack scratch space.
	 */
	[BPF_REG_AX] = {STACK_OFFSET(72), STACK_OFFSET(76)},
};

#define	dst_lo	dst[1]
#define dst_hi	dst[0]
#define src_lo	src[1]
#define src_hi	src[0]

/*
 * JIT Context:
 *
 * prog			:	bpf_prog
 * idx			:	index of current last JITed instruction.
 * prologue_bytes	:	bytes used in prologue.
 * epilogue_offset	:	offset of epilogue starting.
 * offsets		:	array of eBPF instruction offsets in
 *				JITed code.
 * target		:	final JITed code.
 * epilogue_bytes	:	no of bytes used in epilogue.
 * imm_count		:	no of immediate counts used for global
 *				variables.
 * imms			:	array of global variable addresses.
 */

struct jit_ctx {
	const struct bpf_prog *prog;
	unsigned int idx;
	unsigned int prologue_bytes;
	unsigned int epilogue_offset;
	u32 flags;
	u32 *offsets;
	u32 *target;
	u32 stack_size;
#if __LINUX_ARM_ARCH__ < 7
	u16 epilogue_bytes;
	u16 imm_count;
	u32 *imms;
#endif
};

/*
 * Wrappers which handle both OABI and EABI and assures Thumb2 interworking
 * (where the assembly routines like __aeabi_uidiv could cause problems).
 */
static u32 jit_udiv32(u32 dividend, u32 divisor)
{
	return dividend / divisor;
}

static u32 jit_mod32(u32 dividend, u32 divisor)
{
	return dividend % divisor;
}

static inline void _emit(int cond, u32 inst, struct jit_ctx *ctx)
{
	inst |= (cond << 28);
	inst = __opcode_to_mem_arm(inst);

	if (ctx->target != NULL)
		ctx->target[ctx->idx] = inst;

	ctx->idx++;
}

/*
 * Emit an instruction that will be executed unconditionally.
 */
static inline void emit(u32 inst, struct jit_ctx *ctx)
{
	_emit(ARM_COND_AL, inst, ctx);
}

/*
 * Checks if immediate value can be converted to imm12(12 bits) value.
 */
static int16_t imm8m(u32 x)
{
	u32 rot;

	for (rot = 0; rot < 16; rot++)
		if ((x & ~ror32(0xff, 2 * rot)) == 0)
			return rol32(x, 2 * rot) | (rot << 8);
	return -1;
}

/*
 * Initializes the JIT space with undefined instructions.
 */
static void jit_fill_hole(void *area, unsigned int size)
{
	u32 *ptr;
	/* We are guaranteed to have aligned memory. */
	for (ptr = area; size >= sizeof(u32); size -= sizeof(u32))
		*ptr++ = __opcode_to_mem_arm(ARM_INST_UDF);
}

#if defined(CONFIG_AEABI) && (__LINUX_ARM_ARCH__ >= 5)
/* EABI requires the stack to be aligned to 64-bit boundaries */
#define STACK_ALIGNMENT	8
#else
/* Stack must be aligned to 32-bit boundaries */
#define STACK_ALIGNMENT	4
#endif

/* Stack space for BPF_REG_2, BPF_REG_3, BPF_REG_4,
 * BPF_REG_5, BPF_REG_7, BPF_REG_8, BPF_REG_9,
 * BPF_REG_FP and Tail call counts.
 */
#define SCRATCH_SIZE 80

/* total stack size used in JITed code */
#define _STACK_SIZE	(ctx->prog->aux->stack_depth + SCRATCH_SIZE)
#define STACK_SIZE	ALIGN(_STACK_SIZE, STACK_ALIGNMENT)

/* Get the offset of eBPF REGISTERs stored on scratch space. */
#define STACK_VAR(off) (STACK_SIZE - off)

#if __LINUX_ARM_ARCH__ < 7

static u16 imm_offset(u32 k, struct jit_ctx *ctx)
{
	unsigned int i = 0, offset;
	u16 imm;

	/* on the "fake" run we just count them (duplicates included) */
	if (ctx->target == NULL) {
		ctx->imm_count++;
		return 0;
	}

	while ((i < ctx->imm_count) && ctx->imms[i]) {
		if (ctx->imms[i] == k)
			break;
		i++;
	}

	if (ctx->imms[i] == 0)
		ctx->imms[i] = k;

	/* constants go just after the epilogue */
	offset =  ctx->offsets[ctx->prog->len - 1] * 4;
	offset += ctx->prologue_bytes;
	offset += ctx->epilogue_bytes;
	offset += i * 4;

	ctx->target[offset / 4] = k;

	/* PC in ARM mode == address of the instruction + 8 */
	imm = offset - (8 + ctx->idx * 4);

	if (imm & ~0xfff) {
		/*
		 * literal pool is too far, signal it into flags. we
		 * can only detect it on the second pass unfortunately.
		 */
		ctx->flags |= FLAG_IMM_OVERFLOW;
		return 0;
	}

	return imm;
}

#endif /* __LINUX_ARM_ARCH__ */

static inline int bpf2a32_offset(int bpf_to, int bpf_from,
				 const struct jit_ctx *ctx) {
	int to, from;

	if (ctx->target == NULL)
		return 0;
	to = ctx->offsets[bpf_to];
	from = ctx->offsets[bpf_from];

	return to - from - 1;
}

/*
 * Move an immediate that's not an imm8m to a core register.
 */
static inline void emit_mov_i_no8m(const u8 rd, u32 val, struct jit_ctx *ctx)
{
#if __LINUX_ARM_ARCH__ < 7
	emit(ARM_LDR_I(rd, ARM_PC, imm_offset(val, ctx)), ctx);
#else
	emit(ARM_MOVW(rd, val & 0xffff), ctx);
	if (val > 0xffff)
		emit(ARM_MOVT(rd, val >> 16), ctx);
#endif
}

static inline void emit_mov_i(const u8 rd, u32 val, struct jit_ctx *ctx)
{
	int imm12 = imm8m(val);

	if (imm12 >= 0)
		emit(ARM_MOV_I(rd, imm12), ctx);
	else
		emit_mov_i_no8m(rd, val, ctx);
}

static void emit_bx_r(u8 tgt_reg, struct jit_ctx *ctx)
{
	if (elf_hwcap & HWCAP_THUMB)
		emit(ARM_BX(tgt_reg), ctx);
	else
		emit(ARM_MOV_R(ARM_PC, tgt_reg), ctx);
}

static inline void emit_blx_r(u8 tgt_reg, struct jit_ctx *ctx)
{
#if __LINUX_ARM_ARCH__ < 5
	emit(ARM_MOV_R(ARM_LR, ARM_PC), ctx);
	emit_bx_r(tgt_reg, ctx);
#else
	emit(ARM_BLX_R(tgt_reg), ctx);
#endif
}

static inline int epilogue_offset(const struct jit_ctx *ctx)
{
	int to, from;
	/* No need for 1st dummy run */
	if (ctx->target == NULL)
		return 0;
	to = ctx->epilogue_offset;
	from = ctx->idx;

	return to - from - 2;
}

static inline void emit_udivmod(u8 rd, u8 rm, u8 rn, struct jit_ctx *ctx, u8 op)
{
	const u8 *tmp = bpf2a32[TMP_REG_1];

#if __LINUX_ARM_ARCH__ == 7
	if (elf_hwcap & HWCAP_IDIVA) {
		if (op == BPF_DIV)
			emit(ARM_UDIV(rd, rm, rn), ctx);
		else {
			emit(ARM_UDIV(ARM_IP, rm, rn), ctx);
			emit(ARM_MLS(rd, rn, ARM_IP, rm), ctx);
		}
		return;
	}
#endif

	/*
	 * For BPF_ALU | BPF_DIV | BPF_K instructions
	 * As ARM_R1 and ARM_R0 contains 1st argument of bpf
	 * function, we need to save it on caller side to save
	 * it from getting destroyed within callee.
	 * After the return from the callee, we restore ARM_R0
	 * ARM_R1.
	 */
	if (rn != ARM_R1) {
		emit(ARM_MOV_R(tmp[0], ARM_R1), ctx);
		emit(ARM_MOV_R(ARM_R1, rn), ctx);
	}
	if (rm != ARM_R0) {
		emit(ARM_MOV_R(tmp[1], ARM_R0), ctx);
		emit(ARM_MOV_R(ARM_R0, rm), ctx);
	}

	/* Call appropriate function */
	emit_mov_i(ARM_IP, op == BPF_DIV ?
		   (u32)jit_udiv32 : (u32)jit_mod32, ctx);
	emit_blx_r(ARM_IP, ctx);

	/* Save return value */
	if (rd != ARM_R0)
		emit(ARM_MOV_R(rd, ARM_R0), ctx);

	/* Restore ARM_R0 and ARM_R1 */
	if (rn != ARM_R1)
		emit(ARM_MOV_R(ARM_R1, tmp[0]), ctx);
	if (rm != ARM_R0)
		emit(ARM_MOV_R(ARM_R0, tmp[1]), ctx);
}

/* Checks whether BPF register is on scratch stack space or not. */
static inline bool is_on_stack(u8 bpf_reg)
{
	static u8 stack_regs[] = {BPF_REG_AX, BPF_REG_3, BPF_REG_4, BPF_REG_5,
				BPF_REG_7, BPF_REG_8, BPF_REG_9, TCALL_CNT,
				BPF_REG_2, BPF_REG_FP};
	int i, reg_len = sizeof(stack_regs);

	for (i = 0 ; i < reg_len ; i++) {
		if (bpf_reg == stack_regs[i])
			return true;
	}
	return false;
}

static inline void emit_a32_mov_i(const u8 dst, const u32 val,
				  bool dstk, struct jit_ctx *ctx)
{
	const u8 *tmp = bpf2a32[TMP_REG_1];

	if (dstk) {
		emit_mov_i(tmp[1], val, ctx);
		emit(ARM_STR_I(tmp[1], ARM_SP, STACK_VAR(dst)), ctx);
	} else {
		emit_mov_i(dst, val, ctx);
	}
}

/* Sign extended move */
static inline void emit_a32_mov_i64(const bool is64, const u8 dst[],
				  const u32 val, bool dstk,
				  struct jit_ctx *ctx) {
	u32 hi = 0;

	if (is64 && (val & (1<<31)))
		hi = (u32)~0;
	emit_a32_mov_i(dst_lo, val, dstk, ctx);
	emit_a32_mov_i(dst_hi, hi, dstk, ctx);
}

static inline void emit_a32_add_r(const u8 dst, const u8 src,
			      const bool is64, const bool hi,
			      struct jit_ctx *ctx) {
	/* 64 bit :
	 *	adds dst_lo, dst_lo, src_lo
	 *	adc dst_hi, dst_hi, src_hi
	 * 32 bit :
	 *	add dst_lo, dst_lo, src_lo
	 */
	if (!hi && is64)
		emit(ARM_ADDS_R(dst, dst, src), ctx);
	else if (hi && is64)
		emit(ARM_ADC_R(dst, dst, src), ctx);
	else
		emit(ARM_ADD_R(dst, dst, src), ctx);
}

static inline void emit_a32_sub_r(const u8 dst, const u8 src,
				  const bool is64, const bool hi,
				  struct jit_ctx *ctx) {
	/* 64 bit :
	 *	subs dst_lo, dst_lo, src_lo
	 *	sbc dst_hi, dst_hi, src_hi
	 * 32 bit :
	 *	sub dst_lo, dst_lo, src_lo
	 */
	if (!hi && is64)
		emit(ARM_SUBS_R(dst, dst, src), ctx);
	else if (hi && is64)
		emit(ARM_SBC_R(dst, dst, src), ctx);
	else
		emit(ARM_SUB_R(dst, dst, src), ctx);
}

static inline void emit_alu_r(const u8 dst, const u8 src, const bool is64,
			      const bool hi, const u8 op, struct jit_ctx *ctx){
	switch (BPF_OP(op)) {
	/* dst = dst + src */
	case BPF_ADD:
		emit_a32_add_r(dst, src, is64, hi, ctx);
		break;
	/* dst = dst - src */
	case BPF_SUB:
		emit_a32_sub_r(dst, src, is64, hi, ctx);
		break;
	/* dst = dst | src */
	case BPF_OR:
		emit(ARM_ORR_R(dst, dst, src), ctx);
		break;
	/* dst = dst & src */
	case BPF_AND:
		emit(ARM_AND_R(dst, dst, src), ctx);
		break;
	/* dst = dst ^ src */
	case BPF_XOR:
		emit(ARM_EOR_R(dst, dst, src), ctx);
		break;
	/* dst = dst * src */
	case BPF_MUL:
		emit(ARM_MUL(dst, dst, src), ctx);
		break;
	/* dst = dst << src */
	case BPF_LSH:
		emit(ARM_LSL_R(dst, dst, src), ctx);
		break;
	/* dst = dst >> src */
	case BPF_RSH:
		emit(ARM_LSR_R(dst, dst, src), ctx);
		break;
	/* dst = dst >> src (signed)*/
	case BPF_ARSH:
		emit(ARM_MOV_SR(dst, dst, SRTYPE_ASR, src), ctx);
		break;
	}
}

/* ALU operation (32 bit)
 * dst = dst (op) src
 */
static inline void emit_a32_alu_r(const u8 dst, const u8 src,
				  bool dstk, bool sstk,
				  struct jit_ctx *ctx, const bool is64,
				  const bool hi, const u8 op) {
	const u8 *tmp = bpf2a32[TMP_REG_1];
	u8 rn = sstk ? tmp[1] : src;

	if (sstk)
		emit(ARM_LDR_I(rn, ARM_SP, STACK_VAR(src)), ctx);

	/* ALU operation */
	if (dstk) {
		emit(ARM_LDR_I(tmp[0], ARM_SP, STACK_VAR(dst)), ctx);
		emit_alu_r(tmp[0], rn, is64, hi, op, ctx);
		emit(ARM_STR_I(tmp[0], ARM_SP, STACK_VAR(dst)), ctx);
	} else {
		emit_alu_r(dst, rn, is64, hi, op, ctx);
	}
}

/* ALU operation (64 bit) */
static inline void emit_a32_alu_r64(const bool is64, const u8 dst[],
				  const u8 src[], bool dstk,
				  bool sstk, struct jit_ctx *ctx,
				  const u8 op) {
	emit_a32_alu_r(dst_lo, src_lo, dstk, sstk, ctx, is64, false, op);
	if (is64)
		emit_a32_alu_r(dst_hi, src_hi, dstk, sstk, ctx, is64, true, op);
	else
		emit_a32_mov_i(dst_hi, 0, dstk, ctx);
}

/* dst = imm (4 bytes)*/
static inline void emit_a32_mov_r(const u8 dst, const u8 src,
				  bool dstk, bool sstk,
				  struct jit_ctx *ctx) {
	const u8 *tmp = bpf2a32[TMP_REG_1];
	u8 rt = sstk ? tmp[0] : src;

	if (sstk)
		emit(ARM_LDR_I(tmp[0], ARM_SP, STACK_VAR(src)), ctx);
	if (dstk)
		emit(ARM_STR_I(rt, ARM_SP, STACK_VAR(dst)), ctx);
	else
		emit(ARM_MOV_R(dst, rt), ctx);
}

/* dst = src */
static inline void emit_a32_mov_r64(const bool is64, const u8 dst[],
				  const u8 src[], bool dstk,
				  bool sstk, struct jit_ctx *ctx) {
	emit_a32_mov_r(dst_lo, src_lo, dstk, sstk, ctx);
	if (is64) {
		/* complete 8 byte move */
		emit_a32_mov_r(dst_hi, src_hi, dstk, sstk, ctx);
	} else {
		/* Zero out high 4 bytes */
		emit_a32_mov_i(dst_hi, 0, dstk, ctx);
	}
}

/* Shift operations */
static inline void emit_a32_alu_i(const u8 dst, const u32 val, bool dstk,
				struct jit_ctx *ctx, const u8 op) {
	const u8 *tmp = bpf2a32[TMP_REG_1];
	u8 rd = dstk ? tmp[0] : dst;

	if (dstk)
		emit(ARM_LDR_I(rd, ARM_SP, STACK_VAR(dst)), ctx);

	/* Do shift operation */
	switch (op) {
	case BPF_LSH:
		emit(ARM_LSL_I(rd, rd, val), ctx);
		break;
	case BPF_RSH:
		emit(ARM_LSR_I(rd, rd, val), ctx);
		break;
	case BPF_NEG:
		emit(ARM_RSB_I(rd, rd, val), ctx);
		break;
	}

	if (dstk)
		emit(ARM_STR_I(rd, ARM_SP, STACK_VAR(dst)), ctx);
}

/* dst = ~dst (64 bit) */
static inline void emit_a32_neg64(const u8 dst[], bool dstk,
				struct jit_ctx *ctx){
	const u8 *tmp = bpf2a32[TMP_REG_1];
	u8 rd = dstk ? tmp[1] : dst[1];
	u8 rm = dstk ? tmp[0] : dst[0];

	/* Setup Operand */
	if (dstk) {
		emit(ARM_LDR_I(rd, ARM_SP, STACK_VAR(dst_lo)), ctx);
		emit(ARM_LDR_I(rm, ARM_SP, STACK_VAR(dst_hi)), ctx);
	}

	/* Do Negate Operation */
	emit(ARM_RSBS_I(rd, rd, 0), ctx);
	emit(ARM_RSC_I(rm, rm, 0), ctx);

	if (dstk) {
		emit(ARM_STR_I(rd, ARM_SP, STACK_VAR(dst_lo)), ctx);
		emit(ARM_STR_I(rm, ARM_SP, STACK_VAR(dst_hi)), ctx);
	}
}

/* dst = dst << src */
static inline void emit_a32_lsh_r64(const u8 dst[], const u8 src[], bool dstk,
				    bool sstk, struct jit_ctx *ctx) {
	const u8 *tmp = bpf2a32[TMP_REG_1];
	const u8 *tmp2 = bpf2a32[TMP_REG_2];

	/* Setup Operands */
	u8 rt = sstk ? tmp2[1] : src_lo;
	u8 rd = dstk ? tmp[1] : dst_lo;
	u8 rm = dstk ? tmp[0] : dst_hi;

	if (sstk)
		emit(ARM_LDR_I(rt, ARM_SP, STACK_VAR(src_lo)), ctx);
	if (dstk) {
		emit(ARM_LDR_I(rd, ARM_SP, STACK_VAR(dst_lo)), ctx);
		emit(ARM_LDR_I(rm, ARM_SP, STACK_VAR(dst_hi)), ctx);
	}

	/* Do LSH operation */
	emit(ARM_SUB_I(ARM_IP, rt, 32), ctx);
	emit(ARM_RSB_I(tmp2[0], rt, 32), ctx);
	emit(ARM_MOV_SR(ARM_LR, rm, SRTYPE_ASL, rt), ctx);
	emit(ARM_ORR_SR(ARM_LR, ARM_LR, rd, SRTYPE_ASL, ARM_IP), ctx);
	emit(ARM_ORR_SR(ARM_IP, ARM_LR, rd, SRTYPE_LSR, tmp2[0]), ctx);
	emit(ARM_MOV_SR(ARM_LR, rd, SRTYPE_ASL, rt), ctx);

	if (dstk) {
		emit(ARM_STR_I(ARM_LR, ARM_SP, STACK_VAR(dst_lo)), ctx);
		emit(ARM_STR_I(ARM_IP, ARM_SP, STACK_VAR(dst_hi)), ctx);
	} else {
		emit(ARM_MOV_R(rd, ARM_LR), ctx);
		emit(ARM_MOV_R(rm, ARM_IP), ctx);
	}
}

/* dst = dst >> src (signed)*/
static inline void emit_a32_arsh_r64(const u8 dst[], const u8 src[], bool dstk,
				    bool sstk, struct jit_ctx *ctx) {
	const u8 *tmp = bpf2a32[TMP_REG_1];
	const u8 *tmp2 = bpf2a32[TMP_REG_2];
	/* Setup Operands */
	u8 rt = sstk ? tmp2[1] : src_lo;
	u8 rd = dstk ? tmp[1] : dst_lo;
	u8 rm = dstk ? tmp[0] : dst_hi;

	if (sstk)
		emit(ARM_LDR_I(rt, ARM_SP, STACK_VAR(src_lo)), ctx);
	if (dstk) {
		emit(ARM_LDR_I(rd, ARM_SP, STACK_VAR(dst_lo)), ctx);
		emit(ARM_LDR_I(rm, ARM_SP, STACK_VAR(dst_hi)), ctx);
	}

	/* Do the ARSH operation */
	emit(ARM_RSB_I(ARM_IP, rt, 32), ctx);
	emit(ARM_SUBS_I(tmp2[0], rt, 32), ctx);
	emit(ARM_MOV_SR(ARM_LR, rd, SRTYPE_LSR, rt), ctx);
	emit(ARM_ORR_SR(ARM_LR, ARM_LR, rm, SRTYPE_ASL, ARM_IP), ctx);
	_emit(ARM_COND_MI, ARM_B(0), ctx);
	emit(ARM_ORR_SR(ARM_LR, ARM_LR, rm, SRTYPE_ASR, tmp2[0]), ctx);
	emit(ARM_MOV_SR(ARM_IP, rm, SRTYPE_ASR, rt), ctx);
	if (dstk) {
		emit(ARM_STR_I(ARM_LR, ARM_SP, STACK_VAR(dst_lo)), ctx);
		emit(ARM_STR_I(ARM_IP, ARM_SP, STACK_VAR(dst_hi)), ctx);
	} else {
		emit(ARM_MOV_R(rd, ARM_LR), ctx);
		emit(ARM_MOV_R(rm, ARM_IP), ctx);
	}
}

/* dst = dst >> src */
static inline void emit_a32_rsh_r64(const u8 dst[], const u8 src[], bool dstk,
				     bool sstk, struct jit_ctx *ctx) {
	const u8 *tmp = bpf2a32[TMP_REG_1];
	const u8 *tmp2 = bpf2a32[TMP_REG_2];
	/* Setup Operands */
	u8 rt = sstk ? tmp2[1] : src_lo;
	u8 rd = dstk ? tmp[1] : dst_lo;
	u8 rm = dstk ? tmp[0] : dst_hi;

	if (sstk)
		emit(ARM_LDR_I(rt, ARM_SP, STACK_VAR(src_lo)), ctx);
	if (dstk) {
		emit(ARM_LDR_I(rd, ARM_SP, STACK_VAR(dst_lo)), ctx);
		emit(ARM_LDR_I(rm, ARM_SP, STACK_VAR(dst_hi)), ctx);
	}

	/* Do RSH operation */
	emit(ARM_RSB_I(ARM_IP, rt, 32), ctx);
	emit(ARM_SUBS_I(tmp2[0], rt, 32), ctx);
	emit(ARM_MOV_SR(ARM_LR, rd, SRTYPE_LSR, rt), ctx);
	emit(ARM_ORR_SR(ARM_LR, ARM_LR, rm, SRTYPE_ASL, ARM_IP), ctx);
	emit(ARM_ORR_SR(ARM_LR, ARM_LR, rm, SRTYPE_LSR, tmp2[0]), ctx);
	emit(ARM_MOV_SR(ARM_IP, rm, SRTYPE_LSR, rt), ctx);
	if (dstk) {
		emit(ARM_STR_I(ARM_LR, ARM_SP, STACK_VAR(dst_lo)), ctx);
		emit(ARM_STR_I(ARM_IP, ARM_SP, STACK_VAR(dst_hi)), ctx);
	} else {
		emit(ARM_MOV_R(rd, ARM_LR), ctx);
		emit(ARM_MOV_R(rm, ARM_IP), ctx);
	}
}

/* dst = dst << val */
static inline void emit_a32_lsh_i64(const u8 dst[], bool dstk,
				     const u32 val, struct jit_ctx *ctx){
	const u8 *tmp = bpf2a32[TMP_REG_1];
	const u8 *tmp2 = bpf2a32[TMP_REG_2];
	/* Setup operands */
	u8 rd = dstk ? tmp[1] : dst_lo;
	u8 rm = dstk ? tmp[0] : dst_hi;

	if (dstk) {
		emit(ARM_LDR_I(rd, ARM_SP, STACK_VAR(dst_lo)), ctx);
		emit(ARM_LDR_I(rm, ARM_SP, STACK_VAR(dst_hi)), ctx);
	}

	/* Do LSH operation */
	if (val < 32) {
		emit(ARM_MOV_SI(tmp2[0], rm, SRTYPE_ASL, val), ctx);
		emit(ARM_ORR_SI(rm, tmp2[0], rd, SRTYPE_LSR, 32 - val), ctx);
		emit(ARM_MOV_SI(rd, rd, SRTYPE_ASL, val), ctx);
	} else {
		if (val == 32)
			emit(ARM_MOV_R(rm, rd), ctx);
		else
			emit(ARM_MOV_SI(rm, rd, SRTYPE_ASL, val - 32), ctx);
		emit(ARM_EOR_R(rd, rd, rd), ctx);
	}

	if (dstk) {
		emit(ARM_STR_I(rd, ARM_SP, STACK_VAR(dst_lo)), ctx);
		emit(ARM_STR_I(rm, ARM_SP, STACK_VAR(dst_hi)), ctx);
	}
}

/* dst = dst >> val */
static inline void emit_a32_rsh_i64(const u8 dst[], bool dstk,
				    const u32 val, struct jit_ctx *ctx) {
	const u8 *tmp = bpf2a32[TMP_REG_1];
	const u8 *tmp2 = bpf2a32[TMP_REG_2];
	/* Setup operands */
	u8 rd = dstk ? tmp[1] : dst_lo;
	u8 rm = dstk ? tmp[0] : dst_hi;

	if (dstk) {
		emit(ARM_LDR_I(rd, ARM_SP, STACK_VAR(dst_lo)), ctx);
		emit(ARM_LDR_I(rm, ARM_SP, STACK_VAR(dst_hi)), ctx);
	}

	/* Do LSR operation */
	if (val < 32) {
		emit(ARM_MOV_SI(tmp2[1], rd, SRTYPE_LSR, val), ctx);
		emit(ARM_ORR_SI(rd, tmp2[1], rm, SRTYPE_ASL, 32 - val), ctx);
		emit(ARM_MOV_SI(rm, rm, SRTYPE_LSR, val), ctx);
	} else if (val == 32) {
		emit(ARM_MOV_R(rd, rm), ctx);
		emit(ARM_MOV_I(rm, 0), ctx);
	} else {
		emit(ARM_MOV_SI(rd, rm, SRTYPE_LSR, val - 32), ctx);
		emit(ARM_MOV_I(rm, 0), ctx);
	}

	if (dstk) {
		emit(ARM_STR_I(rd, ARM_SP, STACK_VAR(dst_lo)), ctx);
		emit(ARM_STR_I(rm, ARM_SP, STACK_VAR(dst_hi)), ctx);
	}
}

/* dst = dst >> val (signed) */
static inline void emit_a32_arsh_i64(const u8 dst[], bool dstk,
				     const u32 val, struct jit_ctx *ctx){
	const u8 *tmp = bpf2a32[TMP_REG_1];
	const u8 *tmp2 = bpf2a32[TMP_REG_2];
	 /* Setup operands */
	u8 rd = dstk ? tmp[1] : dst_lo;
	u8 rm = dstk ? tmp[0] : dst_hi;

	if (dstk) {
		emit(ARM_LDR_I(rd, ARM_SP, STACK_VAR(dst_lo)), ctx);
		emit(ARM_LDR_I(rm, ARM_SP, STACK_VAR(dst_hi)), ctx);
	}

	/* Do ARSH operation */
	if (val < 32) {
		emit(ARM_MOV_SI(tmp2[1], rd, SRTYPE_LSR, val), ctx);
		emit(ARM_ORR_SI(rd, tmp2[1], rm, SRTYPE_ASL, 32 - val), ctx);
		emit(ARM_MOV_SI(rm, rm, SRTYPE_ASR, val), ctx);
	} else if (val == 32) {
		emit(ARM_MOV_R(rd, rm), ctx);
		emit(ARM_MOV_SI(rm, rm, SRTYPE_ASR, 31), ctx);
	} else {
		emit(ARM_MOV_SI(rd, rm, SRTYPE_ASR, val - 32), ctx);
		emit(ARM_MOV_SI(rm, rm, SRTYPE_ASR, 31), ctx);
	}

	if (dstk) {
		emit(ARM_STR_I(rd, ARM_SP, STACK_VAR(dst_lo)), ctx);
		emit(ARM_STR_I(rm, ARM_SP, STACK_VAR(dst_hi)), ctx);
	}
}

static inline void emit_a32_mul_r64(const u8 dst[], const u8 src[], bool dstk,
				    bool sstk, struct jit_ctx *ctx) {
	const u8 *tmp = bpf2a32[TMP_REG_1];
	const u8 *tmp2 = bpf2a32[TMP_REG_2];
	/* Setup operands for multiplication */
	u8 rd = dstk ? tmp[1] : dst_lo;
	u8 rm = dstk ? tmp[0] : dst_hi;
	u8 rt = sstk ? tmp2[1] : src_lo;
	u8 rn = sstk ? tmp2[0] : src_hi;

	if (dstk) {
		emit(ARM_LDR_I(rd, ARM_SP, STACK_VAR(dst_lo)), ctx);
		emit(ARM_LDR_I(rm, ARM_SP, STACK_VAR(dst_hi)), ctx);
	}
	if (sstk) {
		emit(ARM_LDR_I(rt, ARM_SP, STACK_VAR(src_lo)), ctx);
		emit(ARM_LDR_I(rn, ARM_SP, STACK_VAR(src_hi)), ctx);
	}

	/* Do Multiplication */
	emit(ARM_MUL(ARM_IP, rd, rn), ctx);
	emit(ARM_MUL(ARM_LR, rm, rt), ctx);
	emit(ARM_ADD_R(ARM_LR, ARM_IP, ARM_LR), ctx);

	emit(ARM_UMULL(ARM_IP, rm, rd, rt), ctx);
	emit(ARM_ADD_R(rm, ARM_LR, rm), ctx);
	if (dstk) {
		emit(ARM_STR_I(ARM_IP, ARM_SP, STACK_VAR(dst_lo)), ctx);
		emit(ARM_STR_I(rm, ARM_SP, STACK_VAR(dst_hi)), ctx);
	} else {
		emit(ARM_MOV_R(rd, ARM_IP), ctx);
	}
}

/* *(size *)(dst + off) = src */
static inline void emit_str_r(const u8 dst, const u8 src, bool dstk,
			      const s32 off, struct jit_ctx *ctx, const u8 sz){
	const u8 *tmp = bpf2a32[TMP_REG_1];
	u8 rd = dstk ? tmp[1] : dst;

	if (dstk)
		emit(ARM_LDR_I(rd, ARM_SP, STACK_VAR(dst)), ctx);
	if (off) {
		emit_a32_mov_i(tmp[0], off, false, ctx);
		emit(ARM_ADD_R(tmp[0], rd, tmp[0]), ctx);
		rd = tmp[0];
	}
	switch (sz) {
	case BPF_W:
		/* Store a Word */
		emit(ARM_STR_I(src, rd, 0), ctx);
		break;
	case BPF_H:
		/* Store a HalfWord */
		emit(ARM_STRH_I(src, rd, 0), ctx);
		break;
	case BPF_B:
		/* Store a Byte */
		emit(ARM_STRB_I(src, rd, 0), ctx);
		break;
	}
}

/* dst = *(size*)(src + off) */
static inline void emit_ldx_r(const u8 dst[], const u8 src, bool dstk,
			      s32 off, struct jit_ctx *ctx, const u8 sz){
	const u8 *tmp = bpf2a32[TMP_REG_1];
	const u8 *rd = dstk ? tmp : dst;
	u8 rm = src;
	s32 off_max;

	if (sz == BPF_H)
		off_max = 0xff;
	else
		off_max = 0xfff;

	if (off < 0 || off > off_max) {
		emit_a32_mov_i(tmp[0], off, false, ctx);
		emit(ARM_ADD_R(tmp[0], tmp[0], src), ctx);
		rm = tmp[0];
		off = 0;
	} else if (rd[1] == rm) {
		emit(ARM_MOV_R(tmp[0], rm), ctx);
		rm = tmp[0];
	}
	switch (sz) {
	case BPF_B:
		/* Load a Byte */
		emit(ARM_LDRB_I(rd[1], rm, off), ctx);
		emit_a32_mov_i(dst[0], 0, dstk, ctx);
		break;
	case BPF_H:
		/* Load a HalfWord */
		emit(ARM_LDRH_I(rd[1], rm, off), ctx);
		emit_a32_mov_i(dst[0], 0, dstk, ctx);
		break;
	case BPF_W:
		/* Load a Word */
		emit(ARM_LDR_I(rd[1], rm, off), ctx);
		emit_a32_mov_i(dst[0], 0, dstk, ctx);
		break;
	case BPF_DW:
		/* Load a Double Word */
		emit(ARM_LDR_I(rd[1], rm, off), ctx);
		emit(ARM_LDR_I(rd[0], rm, off + 4), ctx);
		break;
	}
	if (dstk)
		emit(ARM_STR_I(rd[1], ARM_SP, STACK_VAR(dst[1])), ctx);
	if (dstk && sz == BPF_DW)
		emit(ARM_STR_I(rd[0], ARM_SP, STACK_VAR(dst[0])), ctx);
}

/* Arithmatic Operation */
static inline void emit_ar_r(const u8 rd, const u8 rt, const u8 rm,
			     const u8 rn, struct jit_ctx *ctx, u8 op) {
	switch (op) {
	case BPF_JSET:
		emit(ARM_AND_R(ARM_IP, rt, rn), ctx);
		emit(ARM_AND_R(ARM_LR, rd, rm), ctx);
		emit(ARM_ORRS_R(ARM_IP, ARM_LR, ARM_IP), ctx);
		break;
	case BPF_JEQ:
	case BPF_JNE:
	case BPF_JGT:
	case BPF_JGE:
	case BPF_JLE:
	case BPF_JLT:
		emit(ARM_CMP_R(rd, rm), ctx);
		_emit(ARM_COND_EQ, ARM_CMP_R(rt, rn), ctx);
		break;
	case BPF_JSLE:
	case BPF_JSGT:
		emit(ARM_CMP_R(rn, rt), ctx);
		emit(ARM_SBCS_R(ARM_IP, rm, rd), ctx);
		break;
	case BPF_JSLT:
	case BPF_JSGE:
		emit(ARM_CMP_R(rt, rn), ctx);
		emit(ARM_SBCS_R(ARM_IP, rd, rm), ctx);
		break;
	}
}

static int out_offset = -1; /* initialized on the first pass of build_body() */
static int emit_bpf_tail_call(struct jit_ctx *ctx)
{

	/* bpf_tail_call(void *prog_ctx, struct bpf_array *array, u64 index) */
	const u8 *r2 = bpf2a32[BPF_REG_2];
	const u8 *r3 = bpf2a32[BPF_REG_3];
	const u8 *tmp = bpf2a32[TMP_REG_1];
	const u8 *tmp2 = bpf2a32[TMP_REG_2];
	const u8 *tcc = bpf2a32[TCALL_CNT];
	const int idx0 = ctx->idx;
#define cur_offset (ctx->idx - idx0)
#define jmp_offset (out_offset - (cur_offset) - 2)
	u32 off, lo, hi;

	/* if (index >= array->map.max_entries)
	 *	goto out;
	 */
	off = offsetof(struct bpf_array, map.max_entries);
	/* array->map.max_entries */
	emit_a32_mov_i(tmp[1], off, false, ctx);
	emit(ARM_LDR_I(tmp2[1], ARM_SP, STACK_VAR(r2[1])), ctx);
	emit(ARM_LDR_R(tmp[1], tmp2[1], tmp[1]), ctx);
	/* index is 32-bit for arrays */
	emit(ARM_LDR_I(tmp2[1], ARM_SP, STACK_VAR(r3[1])), ctx);
	/* index >= array->map.max_entries */
	emit(ARM_CMP_R(tmp2[1], tmp[1]), ctx);
	_emit(ARM_COND_CS, ARM_B(jmp_offset), ctx);

	/* if (tail_call_cnt > MAX_TAIL_CALL_CNT)
	 *	goto out;
	 * tail_call_cnt++;
	 */
	lo = (u32)MAX_TAIL_CALL_CNT;
	hi = (u32)((u64)MAX_TAIL_CALL_CNT >> 32);
	emit(ARM_LDR_I(tmp[1], ARM_SP, STACK_VAR(tcc[1])), ctx);
	emit(ARM_LDR_I(tmp[0], ARM_SP, STACK_VAR(tcc[0])), ctx);
	emit(ARM_CMP_I(tmp[0], hi), ctx);
	_emit(ARM_COND_EQ, ARM_CMP_I(tmp[1], lo), ctx);
	_emit(ARM_COND_HI, ARM_B(jmp_offset), ctx);
	emit(ARM_ADDS_I(tmp[1], tmp[1], 1), ctx);
	emit(ARM_ADC_I(tmp[0], tmp[0], 0), ctx);
	emit(ARM_STR_I(tmp[1], ARM_SP, STACK_VAR(tcc[1])), ctx);
	emit(ARM_STR_I(tmp[0], ARM_SP, STACK_VAR(tcc[0])), ctx);

	/* prog = array->ptrs[index]
	 * if (prog == NULL)
	 *	goto out;
	 */
	off = offsetof(struct bpf_array, ptrs);
	emit_a32_mov_i(tmp[1], off, false, ctx);
	emit(ARM_LDR_I(tmp2[1], ARM_SP, STACK_VAR(r2[1])), ctx);
	emit(ARM_ADD_R(tmp[1], tmp2[1], tmp[1]), ctx);
	emit(ARM_LDR_I(tmp2[1], ARM_SP, STACK_VAR(r3[1])), ctx);
	emit(ARM_MOV_SI(tmp[0], tmp2[1], SRTYPE_ASL, 2), ctx);
	emit(ARM_LDR_R(tmp[1], tmp[1], tmp[0]), ctx);
	emit(ARM_CMP_I(tmp[1], 0), ctx);
	_emit(ARM_COND_EQ, ARM_B(jmp_offset), ctx);

	/* goto *(prog->bpf_func + prologue_size); */
	off = offsetof(struct bpf_prog, bpf_func);
	emit_a32_mov_i(tmp2[1], off, false, ctx);
	emit(ARM_LDR_R(tmp[1], tmp[1], tmp2[1]), ctx);
	emit(ARM_ADD_I(tmp[1], tmp[1], ctx->prologue_bytes), ctx);
	emit_bx_r(tmp[1], ctx);

	/* out: */
	if (out_offset == -1)
		out_offset = cur_offset;
	if (cur_offset != out_offset) {
		pr_err_once("tail_call out_offset = %d, expected %d!\n",
			    cur_offset, out_offset);
		return -1;
	}
	return 0;
#undef cur_offset
#undef jmp_offset
}

/* 0xabcd => 0xcdab */
static inline void emit_rev16(const u8 rd, const u8 rn, struct jit_ctx *ctx)
{
#if __LINUX_ARM_ARCH__ < 6
	const u8 *tmp2 = bpf2a32[TMP_REG_2];

	emit(ARM_AND_I(tmp2[1], rn, 0xff), ctx);
	emit(ARM_MOV_SI(tmp2[0], rn, SRTYPE_LSR, 8), ctx);
	emit(ARM_AND_I(tmp2[0], tmp2[0], 0xff), ctx);
	emit(ARM_ORR_SI(rd, tmp2[0], tmp2[1], SRTYPE_LSL, 8), ctx);
#else /* ARMv6+ */
	emit(ARM_REV16(rd, rn), ctx);
#endif
}

/* 0xabcdefgh => 0xghefcdab */
static inline void emit_rev32(const u8 rd, const u8 rn, struct jit_ctx *ctx)
{
#if __LINUX_ARM_ARCH__ < 6
	const u8 *tmp2 = bpf2a32[TMP_REG_2];

	emit(ARM_AND_I(tmp2[1], rn, 0xff), ctx);
	emit(ARM_MOV_SI(tmp2[0], rn, SRTYPE_LSR, 24), ctx);
	emit(ARM_ORR_SI(ARM_IP, tmp2[0], tmp2[1], SRTYPE_LSL, 24), ctx);

	emit(ARM_MOV_SI(tmp2[1], rn, SRTYPE_LSR, 8), ctx);
	emit(ARM_AND_I(tmp2[1], tmp2[1], 0xff), ctx);
	emit(ARM_MOV_SI(tmp2[0], rn, SRTYPE_LSR, 16), ctx);
	emit(ARM_AND_I(tmp2[0], tmp2[0], 0xff), ctx);
	emit(ARM_MOV_SI(tmp2[0], tmp2[0], SRTYPE_LSL, 8), ctx);
	emit(ARM_ORR_SI(tmp2[0], tmp2[0], tmp2[1], SRTYPE_LSL, 16), ctx);
	emit(ARM_ORR_R(rd, ARM_IP, tmp2[0]), ctx);

#else /* ARMv6+ */
	emit(ARM_REV(rd, rn), ctx);
#endif
}

// push the scratch stack register on top of the stack
static inline void emit_push_r64(const u8 src[], const u8 shift,
		struct jit_ctx *ctx)
{
	const u8 *tmp2 = bpf2a32[TMP_REG_2];
	u16 reg_set = 0;

	emit(ARM_LDR_I(tmp2[1], ARM_SP, STACK_VAR(src[1]+shift)), ctx);
	emit(ARM_LDR_I(tmp2[0], ARM_SP, STACK_VAR(src[0]+shift)), ctx);

	reg_set = (1 << tmp2[1]) | (1 << tmp2[0]);
	emit(ARM_PUSH(reg_set), ctx);
}

static void build_prologue(struct jit_ctx *ctx)
{
	const u8 r0 = bpf2a32[BPF_REG_0][1];
	const u8 r2 = bpf2a32[BPF_REG_1][1];
	const u8 r3 = bpf2a32[BPF_REG_1][0];
	const u8 r4 = bpf2a32[BPF_REG_6][1];
	const u8 fplo = bpf2a32[BPF_REG_FP][1];
	const u8 fphi = bpf2a32[BPF_REG_FP][0];
	const u8 *tcc = bpf2a32[TCALL_CNT];

	/* Save callee saved registers. */
#ifdef CONFIG_FRAME_POINTER
	u16 reg_set = CALLEE_PUSH_MASK | 1 << ARM_IP | 1 << ARM_PC;
	emit(ARM_MOV_R(ARM_IP, ARM_SP), ctx);
	emit(ARM_PUSH(reg_set), ctx);
	emit(ARM_SUB_I(ARM_FP, ARM_IP, 4), ctx);
#else
	emit(ARM_PUSH(CALLEE_PUSH_MASK), ctx);
	emit(ARM_MOV_R(ARM_FP, ARM_SP), ctx);
#endif
	/* Save frame pointer for later */
	emit(ARM_SUB_I(ARM_IP, ARM_SP, SCRATCH_SIZE), ctx);

	ctx->stack_size = imm8m(STACK_SIZE);

	/* Set up function call stack */
	emit(ARM_SUB_I(ARM_SP, ARM_SP, ctx->stack_size), ctx);

	/* Set up BPF prog stack base register */
	emit_a32_mov_r(fplo, ARM_IP, true, false, ctx);
	emit_a32_mov_i(fphi, 0, true, ctx);

	/* mov r4, 0 */
	emit(ARM_MOV_I(r4, 0), ctx);

	/* Move BPF_CTX to BPF_R1 */
	emit(ARM_MOV_R(r3, r4), ctx);
	emit(ARM_MOV_R(r2, r0), ctx);
	/* Initialize Tail Count */
	emit(ARM_STR_I(r4, ARM_SP, STACK_VAR(tcc[0])), ctx);
	emit(ARM_STR_I(r4, ARM_SP, STACK_VAR(tcc[1])), ctx);
	/* end of prologue */
}

/* restore callee saved registers. */
static void build_epilogue(struct jit_ctx *ctx)
{
#ifdef CONFIG_FRAME_POINTER
	/* When using frame pointers, some additional registers need to
	 * be loaded. */
	u16 reg_set = CALLEE_POP_MASK | 1 << ARM_SP;
	emit(ARM_SUB_I(ARM_SP, ARM_FP, hweight16(reg_set) * 4), ctx);
	emit(ARM_LDM(ARM_SP, reg_set), ctx);
#else
	/* Restore callee saved registers. */
	emit(ARM_MOV_R(ARM_SP, ARM_FP), ctx);
	emit(ARM_POP(CALLEE_POP_MASK), ctx);
#endif
}

/*
 * Convert an eBPF instruction to native instruction, i.e
 * JITs an eBPF instruction.
 * Returns :
 *	0  - Successfully JITed an 8-byte eBPF instruction
 *	>0 - Successfully JITed a 16-byte eBPF instruction
 *	<0 - Failed to JIT.
 */
static int build_insn(const struct bpf_insn *insn, struct jit_ctx *ctx)
{
	const u8 code = insn->code;
	const u8 *dst = bpf2a32[insn->dst_reg];
	const u8 *src = bpf2a32[insn->src_reg];
	const u8 *tmp = bpf2a32[TMP_REG_1];
	const u8 *tmp2 = bpf2a32[TMP_REG_2];
	const s16 off = insn->off;
	const s32 imm = insn->imm;
	const int i = insn - ctx->prog->insnsi;
	const bool is64 = BPF_CLASS(code) == BPF_ALU64;
	const bool dstk = is_on_stack(insn->dst_reg);
	const bool sstk = is_on_stack(insn->src_reg);
	u8 rd, rt, rm, rn;
	s32 jmp_offset;

#define check_imm(bits, imm) do {				\
	if ((imm) >= (1 << ((bits) - 1)) ||			\
	    (imm) < -(1 << ((bits) - 1))) {			\
		pr_info("[%2d] imm=%d(0x%x) out of range\n",	\
			i, imm, imm);				\
		return -EINVAL;					\
	}							\
} while (0)
#define check_imm24(imm) check_imm(24, imm)

	switch (code) {
	/* ALU operations */

	/* dst = src */
	case BPF_ALU | BPF_MOV | BPF_K:
	case BPF_ALU | BPF_MOV | BPF_X:
	case BPF_ALU64 | BPF_MOV | BPF_K:
	case BPF_ALU64 | BPF_MOV | BPF_X:
		switch (BPF_SRC(code)) {
		case BPF_X:
			emit_a32_mov_r64(is64, dst, src, dstk, sstk, ctx);
			break;
		case BPF_K:
			/* Sign-extend immediate value to destination reg */
			emit_a32_mov_i64(is64, dst, imm, dstk, ctx);
			break;
		}
		break;
	/* dst = dst + src/imm */
	/* dst = dst - src/imm */
	/* dst = dst | src/imm */
	/* dst = dst & src/imm */
	/* dst = dst ^ src/imm */
	/* dst = dst * src/imm */
	/* dst = dst << src */
	/* dst = dst >> src */
	case BPF_ALU | BPF_ADD | BPF_K:
	case BPF_ALU | BPF_ADD | BPF_X:
	case BPF_ALU | BPF_SUB | BPF_K:
	case BPF_ALU | BPF_SUB | BPF_X:
	case BPF_ALU | BPF_OR | BPF_K:
	case BPF_ALU | BPF_OR | BPF_X:
	case BPF_ALU | BPF_AND | BPF_K:
	case BPF_ALU | BPF_AND | BPF_X:
	case BPF_ALU | BPF_XOR | BPF_K:
	case BPF_ALU | BPF_XOR | BPF_X:
	case BPF_ALU | BPF_MUL | BPF_K:
	case BPF_ALU | BPF_MUL | BPF_X:
	case BPF_ALU | BPF_LSH | BPF_X:
	case BPF_ALU | BPF_RSH | BPF_X:
	case BPF_ALU | BPF_ARSH | BPF_K:
	case BPF_ALU | BPF_ARSH | BPF_X:
	case BPF_ALU64 | BPF_ADD | BPF_K:
	case BPF_ALU64 | BPF_ADD | BPF_X:
	case BPF_ALU64 | BPF_SUB | BPF_K:
	case BPF_ALU64 | BPF_SUB | BPF_X:
	case BPF_ALU64 | BPF_OR | BPF_K:
	case BPF_ALU64 | BPF_OR | BPF_X:
	case BPF_ALU64 | BPF_AND | BPF_K:
	case BPF_ALU64 | BPF_AND | BPF_X:
	case BPF_ALU64 | BPF_XOR | BPF_K:
	case BPF_ALU64 | BPF_XOR | BPF_X:
		switch (BPF_SRC(code)) {
		case BPF_X:
			emit_a32_alu_r64(is64, dst, src, dstk, sstk,
					 ctx, BPF_OP(code));
			break;
		case BPF_K:
			/* Move immediate value to the temporary register
			 * and then do the ALU operation on the temporary
			 * register as this will sign-extend the immediate
			 * value into temporary reg and then it would be
			 * safe to do the operation on it.
			 */
			emit_a32_mov_i64(is64, tmp2, imm, false, ctx);
			emit_a32_alu_r64(is64, dst, tmp2, dstk, false,
					 ctx, BPF_OP(code));
			break;
		}
		break;
	/* dst = dst / src(imm) */
	/* dst = dst % src(imm) */
	case BPF_ALU | BPF_DIV | BPF_K:
	case BPF_ALU | BPF_DIV | BPF_X:
	case BPF_ALU | BPF_MOD | BPF_K:
	case BPF_ALU | BPF_MOD | BPF_X:
		rt = src_lo;
		rd = dstk ? tmp2[1] : dst_lo;
		if (dstk)
			emit(ARM_LDR_I(rd, ARM_SP, STACK_VAR(dst_lo)), ctx);
		switch (BPF_SRC(code)) {
		case BPF_X:
			rt = sstk ? tmp2[0] : rt;
			if (sstk)
				emit(ARM_LDR_I(rt, ARM_SP, STACK_VAR(src_lo)),
				     ctx);
			break;
		case BPF_K:
			rt = tmp2[0];
			emit_a32_mov_i(rt, imm, false, ctx);
			break;
		}
		emit_udivmod(rd, rd, rt, ctx, BPF_OP(code));
		if (dstk)
			emit(ARM_STR_I(rd, ARM_SP, STACK_VAR(dst_lo)), ctx);
		emit_a32_mov_i(dst_hi, 0, dstk, ctx);
		break;
	case BPF_ALU64 | BPF_DIV | BPF_K:
	case BPF_ALU64 | BPF_DIV | BPF_X:
	case BPF_ALU64 | BPF_MOD | BPF_K:
	case BPF_ALU64 | BPF_MOD | BPF_X:
		goto notyet;
	/* dst = dst >> imm */
	/* dst = dst << imm */
	case BPF_ALU | BPF_RSH | BPF_K:
	case BPF_ALU | BPF_LSH | BPF_K:
		if (unlikely(imm > 31))
			return -EINVAL;
		if (imm)
			emit_a32_alu_i(dst_lo, imm, dstk, ctx, BPF_OP(code));
		emit_a32_mov_i(dst_hi, 0, dstk, ctx);
		break;
	/* dst = dst << imm */
	case BPF_ALU64 | BPF_LSH | BPF_K:
		if (unlikely(imm > 63))
			return -EINVAL;
		emit_a32_lsh_i64(dst, dstk, imm, ctx);
		break;
	/* dst = dst >> imm */
	case BPF_ALU64 | BPF_RSH | BPF_K:
		if (unlikely(imm > 63))
			return -EINVAL;
		emit_a32_rsh_i64(dst, dstk, imm, ctx);
		break;
	/* dst = dst << src */
	case BPF_ALU64 | BPF_LSH | BPF_X:
		emit_a32_lsh_r64(dst, src, dstk, sstk, ctx);
		break;
	/* dst = dst >> src */
	case BPF_ALU64 | BPF_RSH | BPF_X:
		emit_a32_rsh_r64(dst, src, dstk, sstk, ctx);
		break;
	/* dst = dst >> src (signed) */
	case BPF_ALU64 | BPF_ARSH | BPF_X:
		emit_a32_arsh_r64(dst, src, dstk, sstk, ctx);
		break;
	/* dst = dst >> imm (signed) */
	case BPF_ALU64 | BPF_ARSH | BPF_K:
		if (unlikely(imm > 63))
			return -EINVAL;
		emit_a32_arsh_i64(dst, dstk, imm, ctx);
		break;
	/* dst = ~dst */
	case BPF_ALU | BPF_NEG:
		emit_a32_alu_i(dst_lo, 0, dstk, ctx, BPF_OP(code));
		emit_a32_mov_i(dst_hi, 0, dstk, ctx);
		break;
	/* dst = ~dst (64 bit) */
	case BPF_ALU64 | BPF_NEG:
		emit_a32_neg64(dst, dstk, ctx);
		break;
	/* dst = dst * src/imm */
	case BPF_ALU64 | BPF_MUL | BPF_X:
	case BPF_ALU64 | BPF_MUL | BPF_K:
		switch (BPF_SRC(code)) {
		case BPF_X:
			emit_a32_mul_r64(dst, src, dstk, sstk, ctx);
			break;
		case BPF_K:
			/* Move immediate value to the temporary register
			 * and then do the multiplication on it as this
			 * will sign-extend the immediate value into temp
			 * reg then it would be safe to do the operation
			 * on it.
			 */
			emit_a32_mov_i64(is64, tmp2, imm, false, ctx);
			emit_a32_mul_r64(dst, tmp2, dstk, false, ctx);
			break;
		}
		break;
	/* dst = htole(dst) */
	/* dst = htobe(dst) */
	case BPF_ALU | BPF_END | BPF_FROM_LE:
	case BPF_ALU | BPF_END | BPF_FROM_BE:
		rd = dstk ? tmp[0] : dst_hi;
		rt = dstk ? tmp[1] : dst_lo;
		if (dstk) {
			emit(ARM_LDR_I(rt, ARM_SP, STACK_VAR(dst_lo)), ctx);
			emit(ARM_LDR_I(rd, ARM_SP, STACK_VAR(dst_hi)), ctx);
		}
		if (BPF_SRC(code) == BPF_FROM_LE)
			goto emit_bswap_uxt;
		switch (imm) {
		case 16:
			emit_rev16(rt, rt, ctx);
			goto emit_bswap_uxt;
		case 32:
			emit_rev32(rt, rt, ctx);
			goto emit_bswap_uxt;
		case 64:
			emit_rev32(ARM_LR, rt, ctx);
			emit_rev32(rt, rd, ctx);
			emit(ARM_MOV_R(rd, ARM_LR), ctx);
			break;
		}
		goto exit;
emit_bswap_uxt:
		switch (imm) {
		case 16:
			/* zero-extend 16 bits into 64 bits */
#if __LINUX_ARM_ARCH__ < 6
			emit_a32_mov_i(tmp2[1], 0xffff, false, ctx);
			emit(ARM_AND_R(rt, rt, tmp2[1]), ctx);
#else /* ARMv6+ */
			emit(ARM_UXTH(rt, rt), ctx);
#endif
			emit(ARM_EOR_R(rd, rd, rd), ctx);
			break;
		case 32:
			/* zero-extend 32 bits into 64 bits */
			emit(ARM_EOR_R(rd, rd, rd), ctx);
			break;
		case 64:
			/* nop */
			break;
		}
exit:
		if (dstk) {
			emit(ARM_STR_I(rt, ARM_SP, STACK_VAR(dst_lo)), ctx);
			emit(ARM_STR_I(rd, ARM_SP, STACK_VAR(dst_hi)), ctx);
		}
		break;
	/* dst = imm64 */
	case BPF_LD | BPF_IMM | BPF_DW:
	{
		const struct bpf_insn insn1 = insn[1];
		u32 hi, lo = imm;

		hi = insn1.imm;
		emit_a32_mov_i(dst_lo, lo, dstk, ctx);
		emit_a32_mov_i(dst_hi, hi, dstk, ctx);

		return 1;
	}
	/* LDX: dst = *(size *)(src + off) */
	case BPF_LDX | BPF_MEM | BPF_W:
	case BPF_LDX | BPF_MEM | BPF_H:
	case BPF_LDX | BPF_MEM | BPF_B:
	case BPF_LDX | BPF_MEM | BPF_DW:
		rn = sstk ? tmp2[1] : src_lo;
		if (sstk)
			emit(ARM_LDR_I(rn, ARM_SP, STACK_VAR(src_lo)), ctx);
		emit_ldx_r(dst, rn, dstk, off, ctx, BPF_SIZE(code));
		break;
	/* ST: *(size *)(dst + off) = imm */
	case BPF_ST | BPF_MEM | BPF_W:
	case BPF_ST | BPF_MEM | BPF_H:
	case BPF_ST | BPF_MEM | BPF_B:
	case BPF_ST | BPF_MEM | BPF_DW:
		switch (BPF_SIZE(code)) {
		case BPF_DW:
			/* Sign-extend immediate value into temp reg */
			emit_a32_mov_i64(true, tmp2, imm, false, ctx);
			emit_str_r(dst_lo, tmp2[1], dstk, off, ctx, BPF_W);
			emit_str_r(dst_lo, tmp2[0], dstk, off+4, ctx, BPF_W);
			break;
		case BPF_W:
		case BPF_H:
		case BPF_B:
			emit_a32_mov_i(tmp2[1], imm, false, ctx);
			emit_str_r(dst_lo, tmp2[1], dstk, off, ctx,
				   BPF_SIZE(code));
			break;
		}
		break;
	/* STX XADD: lock *(u32 *)(dst + off) += src */
	case BPF_STX | BPF_XADD | BPF_W:
	/* STX XADD: lock *(u64 *)(dst + off) += src */
	case BPF_STX | BPF_XADD | BPF_DW:
		goto notyet;
	/* STX: *(size *)(dst + off) = src */
	case BPF_STX | BPF_MEM | BPF_W:
	case BPF_STX | BPF_MEM | BPF_H:
	case BPF_STX | BPF_MEM | BPF_B:
	case BPF_STX | BPF_MEM | BPF_DW:
	{
		u8 sz = BPF_SIZE(code);

		rn = sstk ? tmp2[1] : src_lo;
		rm = sstk ? tmp2[0] : src_hi;
		if (sstk) {
			emit(ARM_LDR_I(rn, ARM_SP, STACK_VAR(src_lo)), ctx);
			emit(ARM_LDR_I(rm, ARM_SP, STACK_VAR(src_hi)), ctx);
		}

		/* Store the value */
		if (BPF_SIZE(code) == BPF_DW) {
			emit_str_r(dst_lo, rn, dstk, off, ctx, BPF_W);
			emit_str_r(dst_lo, rm, dstk, off+4, ctx, BPF_W);
		} else {
			emit_str_r(dst_lo, rn, dstk, off, ctx, sz);
		}
		break;
	}
	/* PC += off if dst == src */
	/* PC += off if dst > src */
	/* PC += off if dst >= src */
	/* PC += off if dst < src */
	/* PC += off if dst <= src */
	/* PC += off if dst != src */
	/* PC += off if dst > src (signed) */
	/* PC += off if dst >= src (signed) */
	/* PC += off if dst < src (signed) */
	/* PC += off if dst <= src (signed) */
	/* PC += off if dst & src */
	case BPF_JMP | BPF_JEQ | BPF_X:
	case BPF_JMP | BPF_JGT | BPF_X:
	case BPF_JMP | BPF_JGE | BPF_X:
	case BPF_JMP | BPF_JNE | BPF_X:
	case BPF_JMP | BPF_JSGT | BPF_X:
	case BPF_JMP | BPF_JSGE | BPF_X:
	case BPF_JMP | BPF_JSET | BPF_X:
	case BPF_JMP | BPF_JLE | BPF_X:
	case BPF_JMP | BPF_JLT | BPF_X:
	case BPF_JMP | BPF_JSLT | BPF_X:
	case BPF_JMP | BPF_JSLE | BPF_X:
		/* Setup source registers */
		rm = sstk ? tmp2[0] : src_hi;
		rn = sstk ? tmp2[1] : src_lo;
		if (sstk) {
			emit(ARM_LDR_I(rn, ARM_SP, STACK_VAR(src_lo)), ctx);
			emit(ARM_LDR_I(rm, ARM_SP, STACK_VAR(src_hi)), ctx);
		}
		goto go_jmp;
	/* PC += off if dst == imm */
	/* PC += off if dst > imm */
	/* PC += off if dst >= imm */
	/* PC += off if dst < imm */
	/* PC += off if dst <= imm */
	/* PC += off if dst != imm */
	/* PC += off if dst > imm (signed) */
	/* PC += off if dst >= imm (signed) */
	/* PC += off if dst < imm (signed) */
	/* PC += off if dst <= imm (signed) */
	/* PC += off if dst & imm */
	case BPF_JMP | BPF_JEQ | BPF_K:
	case BPF_JMP | BPF_JGT | BPF_K:
	case BPF_JMP | BPF_JGE | BPF_K:
	case BPF_JMP | BPF_JNE | BPF_K:
	case BPF_JMP | BPF_JSGT | BPF_K:
	case BPF_JMP | BPF_JSGE | BPF_K:
	case BPF_JMP | BPF_JSET | BPF_K:
	case BPF_JMP | BPF_JLT | BPF_K:
	case BPF_JMP | BPF_JLE | BPF_K:
	case BPF_JMP | BPF_JSLT | BPF_K:
	case BPF_JMP | BPF_JSLE | BPF_K:
		if (off == 0)
			break;
		rm = tmp2[0];
		rn = tmp2[1];
		/* Sign-extend immediate value */
		emit_a32_mov_i64(true, tmp2, imm, false, ctx);
go_jmp:
		/* Setup destination register */
		rd = dstk ? tmp[0] : dst_hi;
		rt = dstk ? tmp[1] : dst_lo;
		if (dstk) {
			emit(ARM_LDR_I(rt, ARM_SP, STACK_VAR(dst_lo)), ctx);
			emit(ARM_LDR_I(rd, ARM_SP, STACK_VAR(dst_hi)), ctx);
		}

		/* Check for the condition */
		emit_ar_r(rd, rt, rm, rn, ctx, BPF_OP(code));

		/* Setup JUMP instruction */
		jmp_offset = bpf2a32_offset(i+off, i, ctx);
		switch (BPF_OP(code)) {
		case BPF_JNE:
		case BPF_JSET:
			_emit(ARM_COND_NE, ARM_B(jmp_offset), ctx);
			break;
		case BPF_JEQ:
			_emit(ARM_COND_EQ, ARM_B(jmp_offset), ctx);
			break;
		case BPF_JGT:
			_emit(ARM_COND_HI, ARM_B(jmp_offset), ctx);
			break;
		case BPF_JGE:
			_emit(ARM_COND_CS, ARM_B(jmp_offset), ctx);
			break;
		case BPF_JSGT:
			_emit(ARM_COND_LT, ARM_B(jmp_offset), ctx);
			break;
		case BPF_JSGE:
			_emit(ARM_COND_GE, ARM_B(jmp_offset), ctx);
			break;
		case BPF_JLE:
			_emit(ARM_COND_LS, ARM_B(jmp_offset), ctx);
			break;
		case BPF_JLT:
			_emit(ARM_COND_CC, ARM_B(jmp_offset), ctx);
			break;
		case BPF_JSLT:
			_emit(ARM_COND_LT, ARM_B(jmp_offset), ctx);
			break;
		case BPF_JSLE:
			_emit(ARM_COND_GE, ARM_B(jmp_offset), ctx);
			break;
		}
		break;
	/* JMP OFF */
	case BPF_JMP | BPF_JA:
	{
		if (off == 0)
			break;
		jmp_offset = bpf2a32_offset(i+off, i, ctx);
		check_imm24(jmp_offset);
		emit(ARM_B(jmp_offset), ctx);
		break;
	}
	/* tail call */
	case BPF_JMP | BPF_TAIL_CALL:
		if (emit_bpf_tail_call(ctx))
			return -EFAULT;
		break;
	/* function call */
	case BPF_JMP | BPF_CALL:
	{
		const u8 *r0 = bpf2a32[BPF_REG_0];
		const u8 *r1 = bpf2a32[BPF_REG_1];
		const u8 *r2 = bpf2a32[BPF_REG_2];
		const u8 *r3 = bpf2a32[BPF_REG_3];
		const u8 *r4 = bpf2a32[BPF_REG_4];
		const u8 *r5 = bpf2a32[BPF_REG_5];
		const u32 func = (u32)__bpf_call_base + (u32)imm;

		emit_a32_mov_r64(true, r0, r1, false, false, ctx);
		emit_a32_mov_r64(true, r1, r2, false, true, ctx);
		emit_push_r64(r5, 0, ctx);
		emit_push_r64(r4, 8, ctx);
		emit_push_r64(r3, 16, ctx);

		emit_a32_mov_i(tmp[1], func, false, ctx);
		emit_blx_r(tmp[1], ctx);

		emit(ARM_ADD_I(ARM_SP, ARM_SP, imm8m(24)), ctx); // callee clean
		break;
	}
	/* function return */
	case BPF_JMP | BPF_EXIT:
		/* Optimization: when last instruction is EXIT
		 * simply fallthrough to epilogue.
		 */
		if (i == ctx->prog->len - 1)
			break;
		jmp_offset = epilogue_offset(ctx);
		check_imm24(jmp_offset);
		emit(ARM_B(jmp_offset), ctx);
		break;
notyet:
		pr_info_once("*** NOT YET: opcode %02x ***\n", code);
		return -EFAULT;
	default:
		pr_err_once("unknown opcode %02x\n", code);
		return -EINVAL;
	}

	if (ctx->flags & FLAG_IMM_OVERFLOW)
		/*
		 * this instruction generated an overflow when
		 * trying to access the literal pool, so
		 * delegate this filter to the kernel interpreter.
		 */
		return -1;
	return 0;
}

static int build_body(struct jit_ctx *ctx)
{
	const struct bpf_prog *prog = ctx->prog;
	unsigned int i;

	for (i = 0; i < prog->len; i++) {
		const struct bpf_insn *insn = &(prog->insnsi[i]);
		int ret;

		ret = build_insn(insn, ctx);

		/* It's used with loading the 64 bit immediate value. */
		if (ret > 0) {
			i++;
			if (ctx->target == NULL)
				ctx->offsets[i] = ctx->idx;
			continue;
		}

		if (ctx->target == NULL)
			ctx->offsets[i] = ctx->idx;

		/* If unsuccesfull, return with error code */
		if (ret)
			return ret;
	}
	return 0;
}

static int validate_code(struct jit_ctx *ctx)
{
	int i;

	for (i = 0; i < ctx->idx; i++) {
		if (ctx->target[i] == __opcode_to_mem_arm(ARM_INST_UDF))
			return -1;
	}

	return 0;
}

void bpf_jit_compile(struct bpf_prog *prog)
{
	/* Nothing to do here. We support Internal BPF. */
}

struct bpf_prog *bpf_int_jit_compile(struct bpf_prog *prog)
{
	struct bpf_prog *tmp, *orig_prog = prog;
	struct bpf_binary_header *header;
	bool tmp_blinded = false;
	struct jit_ctx ctx;
	unsigned int tmp_idx;
	unsigned int image_size;
	u8 *image_ptr;

	/* If BPF JIT was not enabled then we must fall back to
	 * the interpreter.
	 */
	if (!prog->jit_requested)
		return orig_prog;

	/* If constant blinding was enabled and we failed during blinding
	 * then we must fall back to the interpreter. Otherwise, we save
	 * the new JITed code.
	 */
	tmp = bpf_jit_blind_constants(prog);

	if (IS_ERR(tmp))
		return orig_prog;
	if (tmp != prog) {
		tmp_blinded = true;
		prog = tmp;
	}

	memset(&ctx, 0, sizeof(ctx));
	ctx.prog = prog;

	/* Not able to allocate memory for offsets[] , then
	 * we must fall back to the interpreter
	 */
	ctx.offsets = kcalloc(prog->len, sizeof(int), GFP_KERNEL);
	if (ctx.offsets == NULL) {
		prog = orig_prog;
		goto out;
	}

	/* 1) fake pass to find in the length of the JITed code,
	 * to compute ctx->offsets and other context variables
	 * needed to compute final JITed code.
	 * Also, calculate random starting pointer/start of JITed code
	 * which is prefixed by random number of fault instructions.
	 *
	 * If the first pass fails then there is no chance of it
	 * being successful in the second pass, so just fall back
	 * to the interpreter.
	 */
	if (build_body(&ctx)) {
		prog = orig_prog;
		goto out_off;
	}

	tmp_idx = ctx.idx;
	build_prologue(&ctx);
	ctx.prologue_bytes = (ctx.idx - tmp_idx) * 4;

	ctx.epilogue_offset = ctx.idx;

#if __LINUX_ARM_ARCH__ < 7
	tmp_idx = ctx.idx;
	build_epilogue(&ctx);
	ctx.epilogue_bytes = (ctx.idx - tmp_idx) * 4;

	ctx.idx += ctx.imm_count;
	if (ctx.imm_count) {
		ctx.imms = kcalloc(ctx.imm_count, sizeof(u32), GFP_KERNEL);
		if (ctx.imms == NULL) {
			prog = orig_prog;
			goto out_off;
		}
	}
#else
	/* there's nothing about the epilogue on ARMv7 */
	build_epilogue(&ctx);
#endif
	/* Now we can get the actual image size of the JITed arm code.
	 * Currently, we are not considering the THUMB-2 instructions
	 * for jit, although it can decrease the size of the image.
	 *
	 * As each arm instruction is of length 32bit, we are translating
	 * number of JITed intructions into the size required to store these
	 * JITed code.
	 */
	image_size = sizeof(u32) * ctx.idx;

	/* Now we know the size of the structure to make */
	header = bpf_jit_binary_alloc(image_size, &image_ptr,
				      sizeof(u32), jit_fill_hole);
	/* Not able to allocate memory for the structure then
	 * we must fall back to the interpretation
	 */
	if (header == NULL) {
		prog = orig_prog;
		goto out_imms;
	}

	/* 2.) Actual pass to generate final JIT code */
	ctx.target = (u32 *) image_ptr;
	ctx.idx = 0;

	build_prologue(&ctx);

	/* If building the body of the JITed code fails somehow,
	 * we fall back to the interpretation.
	 */
	if (build_body(&ctx) < 0) {
		image_ptr = NULL;
		bpf_jit_binary_free(header);
		prog = orig_prog;
		goto out_imms;
	}
	build_epilogue(&ctx);

	/* 3.) Extra pass to validate JITed Code */
	if (validate_code(&ctx)) {
		image_ptr = NULL;
		bpf_jit_binary_free(header);
		prog = orig_prog;
		goto out_imms;
	}
	flush_icache_range((u32)header, (u32)(ctx.target + ctx.idx));

	if (bpf_jit_enable > 1)
		/* there are 2 passes here */
		bpf_jit_dump(prog->len, image_size, 2, ctx.target);

	set_memory_ro((unsigned long)header, header->pages);
	prog->bpf_func = (void *)ctx.target;
	prog->jited = 1;
	prog->jited_len = image_size;

out_imms:
#if __LINUX_ARM_ARCH__ < 7
	if (ctx.imm_count)
		kfree(ctx.imms);
#endif
out_off:
	kfree(ctx.offsets);
out:
	if (tmp_blinded)
		bpf_jit_prog_release_other(prog, prog == orig_prog ?
					   tmp : orig_prog);
	return prog;
}