aboutsummaryrefslogtreecommitdiffstats
path: root/lib/Target/ARM/ARMInstrNEON.td
blob: d3e83f42b1ca3134c40b6b11a13760c2a9dca2b0 (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
1866
1867
1868
1869
1870
1871
1872
1873
1874
1875
1876
1877
1878
1879
1880
1881
1882
1883
1884
1885
1886
1887
1888
1889
1890
1891
1892
1893
1894
1895
1896
1897
1898
1899
1900
1901
1902
1903
1904
1905
1906
1907
1908
1909
1910
1911
1912
1913
1914
1915
1916
1917
1918
1919
1920
1921
1922
1923
1924
1925
1926
1927
1928
1929
1930
1931
1932
1933
1934
1935
1936
1937
1938
1939
1940
1941
1942
1943
1944
1945
1946
1947
1948
1949
1950
1951
1952
1953
1954
1955
1956
1957
1958
1959
1960
1961
1962
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
2036
2037
2038
2039
2040
2041
2042
2043
2044
2045
2046
2047
2048
2049
2050
2051
2052
2053
2054
2055
2056
2057
2058
2059
2060
2061
2062
2063
2064
2065
2066
2067
2068
2069
2070
2071
2072
2073
2074
2075
2076
2077
2078
2079
2080
2081
2082
2083
2084
2085
2086
2087
2088
2089
2090
2091
2092
2093
2094
2095
2096
2097
2098
2099
2100
2101
2102
2103
2104
2105
2106
2107
2108
2109
2110
2111
2112
2113
2114
2115
2116
2117
2118
2119
2120
2121
2122
2123
2124
2125
2126
2127
2128
2129
2130
2131
2132
2133
2134
2135
2136
2137
2138
2139
2140
2141
2142
2143
2144
2145
2146
2147
2148
2149
2150
2151
2152
2153
2154
2155
2156
2157
2158
2159
2160
2161
2162
2163
2164
2165
2166
2167
2168
2169
2170
2171
2172
2173
2174
2175
2176
2177
2178
2179
2180
2181
2182
2183
2184
2185
2186
2187
2188
2189
2190
2191
2192
2193
2194
2195
2196
2197
2198
2199
2200
2201
2202
2203
2204
2205
2206
2207
2208
2209
2210
2211
2212
2213
2214
2215
2216
2217
2218
2219
2220
2221
2222
2223
2224
2225
2226
2227
2228
2229
2230
2231
2232
2233
2234
2235
2236
2237
2238
2239
2240
2241
2242
2243
2244
2245
2246
2247
2248
2249
2250
2251
2252
2253
2254
2255
2256
2257
2258
2259
2260
2261
2262
2263
2264
2265
2266
2267
2268
2269
2270
2271
2272
2273
2274
2275
2276
2277
2278
2279
2280
2281
2282
2283
2284
2285
2286
2287
2288
2289
2290
2291
2292
2293
2294
2295
2296
2297
2298
2299
2300
2301
2302
2303
2304
2305
2306
2307
2308
2309
2310
2311
2312
2313
2314
2315
2316
2317
2318
2319
2320
2321
2322
2323
2324
2325
2326
2327
2328
2329
2330
2331
2332
2333
2334
2335
2336
2337
2338
2339
2340
2341
2342
2343
2344
2345
2346
2347
2348
2349
2350
2351
2352
2353
2354
2355
2356
2357
2358
2359
2360
2361
2362
2363
2364
2365
2366
2367
2368
2369
2370
2371
2372
2373
2374
2375
2376
2377
2378
2379
2380
2381
2382
2383
2384
2385
2386
2387
2388
2389
2390
2391
2392
2393
2394
2395
2396
2397
2398
2399
2400
2401
2402
2403
2404
2405
2406
2407
2408
2409
2410
2411
2412
2413
2414
2415
2416
2417
2418
2419
2420
2421
2422
2423
2424
2425
2426
2427
2428
2429
2430
2431
2432
2433
2434
2435
2436
2437
2438
2439
2440
2441
2442
2443
2444
2445
2446
2447
2448
2449
2450
2451
2452
2453
2454
2455
2456
2457
2458
2459
2460
2461
2462
2463
2464
2465
2466
2467
2468
2469
2470
2471
2472
2473
2474
2475
2476
2477
2478
2479
2480
2481
2482
2483
2484
2485
2486
2487
2488
2489
2490
2491
2492
2493
2494
2495
2496
2497
2498
2499
2500
2501
2502
2503
2504
2505
2506
2507
2508
2509
2510
2511
2512
2513
2514
2515
2516
2517
2518
2519
2520
2521
2522
2523
2524
2525
2526
2527
2528
2529
2530
2531
2532
2533
2534
2535
2536
2537
2538
2539
2540
2541
2542
2543
2544
2545
2546
2547
2548
2549
2550
2551
2552
2553
2554
2555
2556
2557
2558
2559
2560
2561
2562
2563
2564
2565
2566
2567
2568
2569
2570
2571
2572
2573
2574
2575
2576
2577
2578
2579
2580
2581
2582
2583
2584
2585
2586
2587
2588
2589
2590
2591
2592
2593
2594
2595
2596
2597
2598
2599
2600
2601
2602
2603
2604
2605
2606
2607
2608
2609
2610
2611
2612
2613
2614
2615
2616
2617
2618
2619
2620
2621
2622
2623
2624
2625
2626
2627
2628
2629
2630
2631
2632
2633
2634
2635
2636
2637
2638
2639
2640
2641
2642
2643
2644
2645
2646
2647
2648
2649
2650
2651
2652
2653
2654
2655
2656
2657
2658
2659
2660
2661
2662
2663
2664
2665
2666
2667
2668
2669
2670
2671
2672
2673
2674
2675
2676
2677
2678
2679
2680
2681
2682
2683
2684
2685
2686
2687
2688
2689
2690
2691
2692
2693
2694
2695
2696
2697
2698
2699
2700
2701
2702
2703
2704
2705
2706
2707
2708
2709
2710
2711
2712
2713
2714
2715
2716
2717
2718
2719
2720
2721
2722
2723
2724
2725
2726
2727
2728
2729
2730
2731
2732
2733
2734
2735
2736
2737
2738
2739
2740
2741
2742
2743
2744
2745
2746
2747
2748
2749
2750
2751
2752
2753
2754
2755
2756
2757
2758
2759
2760
2761
2762
2763
2764
2765
2766
2767
2768
2769
2770
2771
2772
2773
2774
2775
2776
2777
2778
2779
2780
2781
2782
2783
2784
2785
2786
2787
2788
2789
2790
2791
2792
2793
2794
2795
2796
2797
2798
2799
2800
2801
2802
2803
2804
2805
2806
2807
2808
2809
2810
2811
2812
2813
2814
2815
2816
2817
2818
2819
2820
2821
2822
2823
2824
2825
2826
2827
2828
2829
2830
2831
2832
2833
2834
2835
2836
2837
2838
2839
2840
2841
2842
2843
2844
2845
2846
2847
2848
2849
2850
2851
2852
2853
2854
2855
2856
2857
2858
2859
2860
2861
2862
2863
2864
2865
2866
2867
2868
2869
2870
2871
2872
2873
2874
2875
2876
2877
2878
2879
2880
2881
2882
2883
2884
2885
2886
2887
2888
2889
2890
2891
2892
2893
2894
2895
2896
2897
2898
2899
2900
2901
2902
2903
2904
2905
2906
2907
2908
2909
2910
2911
2912
2913
2914
2915
2916
2917
2918
2919
2920
2921
2922
2923
2924
2925
2926
2927
2928
2929
2930
2931
2932
2933
2934
2935
2936
2937
2938
2939
2940
2941
2942
2943
2944
2945
2946
2947
2948
2949
2950
2951
2952
2953
2954
2955
2956
2957
2958
2959
2960
2961
2962
2963
2964
2965
2966
2967
2968
2969
2970
2971
2972
2973
2974
2975
2976
2977
2978
2979
2980
2981
2982
2983
2984
2985
2986
2987
2988
2989
2990
2991
2992
2993
2994
2995
2996
2997
2998
2999
3000
3001
3002
3003
3004
3005
3006
3007
3008
3009
3010
3011
3012
3013
3014
3015
3016
3017
3018
3019
3020
3021
3022
3023
3024
3025
3026
3027
3028
3029
3030
3031
3032
3033
3034
3035
3036
3037
3038
3039
3040
3041
3042
3043
3044
3045
3046
3047
3048
3049
3050
3051
3052
3053
3054
3055
3056
3057
3058
3059
3060
3061
3062
3063
3064
3065
3066
3067
3068
3069
3070
3071
3072
3073
3074
3075
3076
3077
3078
3079
3080
3081
3082
3083
3084
3085
3086
3087
3088
3089
3090
3091
3092
3093
3094
3095
3096
3097
3098
3099
3100
3101
3102
3103
3104
3105
3106
3107
3108
3109
3110
3111
3112
3113
3114
3115
3116
3117
3118
3119
3120
3121
3122
3123
3124
3125
3126
3127
3128
3129
3130
3131
3132
3133
3134
3135
3136
3137
3138
3139
3140
3141
3142
3143
3144
3145
3146
3147
3148
3149
3150
3151
3152
3153
3154
3155
3156
3157
3158
3159
3160
3161
3162
3163
3164
3165
3166
3167
3168
3169
3170
3171
3172
3173
3174
3175
3176
3177
3178
3179
3180
3181
3182
3183
3184
3185
3186
3187
3188
3189
3190
3191
3192
3193
3194
3195
3196
3197
3198
3199
3200
3201
3202
3203
3204
3205
3206
3207
3208
3209
3210
3211
3212
3213
3214
3215
3216
3217
3218
3219
3220
3221
3222
3223
3224
3225
3226
3227
3228
3229
3230
3231
3232
3233
3234
3235
3236
3237
3238
3239
3240
3241
3242
3243
3244
3245
3246
3247
3248
3249
3250
3251
3252
3253
3254
3255
3256
3257
3258
3259
3260
3261
3262
3263
3264
3265
3266
3267
3268
3269
3270
3271
3272
3273
3274
3275
3276
3277
3278
3279
3280
3281
3282
3283
3284
3285
3286
3287
3288
3289
3290
3291
3292
3293
3294
3295
3296
3297
3298
3299
3300
3301
3302
3303
3304
3305
3306
3307
3308
3309
3310
3311
3312
3313
3314
3315
3316
3317
3318
3319
3320
3321
3322
3323
3324
3325
3326
3327
3328
3329
3330
3331
3332
3333
3334
3335
3336
3337
3338
3339
3340
3341
3342
3343
3344
3345
3346
3347
3348
3349
3350
3351
3352
3353
3354
3355
3356
3357
3358
3359
3360
3361
3362
3363
3364
3365
3366
3367
3368
3369
3370
3371
3372
3373
3374
3375
3376
3377
3378
3379
3380
3381
3382
3383
3384
3385
3386
3387
3388
3389
3390
3391
3392
3393
3394
3395
3396
3397
3398
3399
3400
3401
3402
3403
3404
3405
3406
3407
3408
3409
3410
3411
3412
3413
3414
3415
3416
3417
3418
3419
3420
3421
3422
3423
3424
3425
3426
3427
3428
3429
3430
3431
3432
3433
3434
3435
3436
3437
3438
3439
3440
3441
3442
3443
3444
3445
3446
3447
3448
3449
3450
3451
3452
3453
3454
3455
3456
3457
3458
3459
3460
3461
3462
3463
3464
3465
3466
3467
3468
3469
3470
3471
3472
3473
3474
3475
3476
3477
3478
3479
3480
3481
3482
3483
3484
3485
3486
3487
3488
3489
3490
3491
3492
3493
3494
3495
3496
3497
3498
3499
3500
3501
3502
3503
3504
3505
3506
3507
3508
3509
3510
3511
3512
3513
3514
3515
3516
3517
3518
3519
3520
3521
3522
3523
3524
3525
3526
3527
3528
3529
3530
3531
3532
3533
3534
3535
3536
3537
3538
3539
3540
3541
3542
3543
3544
3545
3546
3547
3548
3549
3550
3551
3552
3553
3554
3555
3556
3557
3558
3559
3560
3561
3562
3563
3564
3565
3566
3567
3568
3569
3570
3571
3572
3573
3574
3575
3576
3577
3578
3579
3580
3581
3582
3583
3584
3585
3586
3587
3588
3589
3590
3591
3592
3593
3594
3595
3596
3597
3598
3599
3600
3601
3602
3603
3604
3605
3606
3607
3608
3609
3610
3611
3612
3613
3614
3615
3616
3617
3618
3619
3620
3621
3622
3623
3624
3625
3626
3627
3628
3629
3630
3631
3632
3633
3634
3635
3636
3637
3638
3639
3640
3641
3642
3643
3644
3645
3646
3647
3648
3649
3650
3651
3652
3653
3654
3655
3656
3657
3658
3659
3660
3661
3662
3663
3664
3665
3666
3667
3668
3669
3670
3671
3672
3673
3674
3675
3676
3677
3678
3679
3680
3681
3682
3683
3684
3685
3686
3687
3688
3689
3690
3691
3692
3693
3694
3695
3696
3697
3698
3699
3700
3701
3702
3703
3704
3705
3706
3707
3708
3709
3710
3711
3712
3713
3714
3715
3716
3717
3718
3719
3720
3721
3722
3723
3724
3725
3726
3727
3728
3729
3730
3731
3732
3733
3734
3735
3736
3737
3738
3739
3740
3741
3742
3743
3744
3745
3746
3747
3748
3749
3750
3751
3752
3753
3754
3755
3756
3757
3758
3759
3760
3761
3762
3763
3764
3765
3766
3767
3768
3769
3770
3771
3772
3773
3774
3775
3776
3777
3778
3779
3780
3781
3782
3783
3784
3785
3786
3787
3788
3789
3790
3791
3792
3793
3794
3795
3796
3797
3798
3799
3800
3801
3802
3803
3804
3805
3806
3807
3808
3809
3810
3811
3812
3813
3814
3815
3816
3817
3818
3819
3820
3821
3822
3823
3824
3825
3826
3827
3828
3829
3830
3831
3832
3833
3834
3835
3836
3837
3838
3839
3840
3841
3842
3843
3844
3845
3846
3847
3848
3849
3850
3851
3852
3853
3854
3855
3856
3857
3858
3859
3860
3861
3862
3863
3864
3865
3866
3867
3868
3869
3870
3871
3872
3873
3874
3875
3876
3877
3878
3879
3880
3881
3882
3883
3884
3885
3886
3887
3888
3889
3890
3891
3892
3893
3894
3895
3896
3897
3898
3899
3900
3901
3902
3903
3904
3905
3906
3907
3908
3909
3910
3911
3912
3913
3914
3915
3916
3917
3918
3919
3920
3921
3922
3923
3924
3925
3926
3927
3928
3929
3930
3931
3932
3933
3934
3935
3936
3937
3938
3939
3940
3941
3942
3943
3944
3945
3946
3947
3948
3949
3950
3951
3952
3953
3954
3955
3956
3957
3958
3959
3960
3961
3962
3963
3964
3965
3966
3967
3968
3969
3970
3971
3972
3973
3974
3975
3976
3977
3978
3979
3980
3981
3982
3983
3984
3985
3986
3987
3988
3989
3990
3991
3992
3993
3994
3995
3996
3997
3998
3999
4000
4001
4002
4003
4004
4005
4006
4007
4008
4009
4010
4011
4012
4013
4014
4015
4016
4017
4018
4019
4020
4021
4022
4023
4024
4025
4026
4027
4028
4029
4030
4031
4032
4033
4034
4035
4036
4037
4038
4039
4040
4041
4042
4043
4044
4045
4046
4047
4048
4049
4050
4051
4052
4053
4054
4055
4056
4057
4058
4059
4060
4061
4062
4063
4064
4065
4066
4067
4068
4069
4070
4071
4072
4073
4074
4075
4076
4077
4078
4079
4080
4081
4082
4083
4084
4085
4086
4087
4088
4089
4090
4091
4092
4093
4094
4095
4096
4097
4098
4099
4100
4101
4102
4103
4104
4105
4106
4107
4108
4109
4110
4111
4112
4113
4114
4115
4116
4117
4118
4119
4120
4121
4122
4123
4124
4125
4126
4127
4128
4129
4130
4131
4132
4133
4134
4135
4136
4137
4138
4139
4140
4141
4142
4143
4144
4145
4146
4147
4148
4149
4150
4151
4152
4153
4154
4155
4156
4157
4158
4159
4160
4161
4162
4163
4164
4165
4166
4167
4168
4169
4170
4171
4172
4173
4174
4175
4176
4177
4178
4179
4180
4181
4182
4183
4184
4185
4186
4187
4188
4189
4190
4191
4192
4193
4194
4195
4196
4197
4198
4199
4200
4201
4202
4203
4204
4205
4206
4207
4208
4209
4210
4211
4212
4213
4214
4215
4216
4217
4218
4219
4220
4221
4222
4223
4224
4225
4226
4227
4228
4229
4230
4231
4232
4233
4234
4235
4236
4237
4238
4239
4240
4241
4242
4243
4244
4245
4246
4247
4248
4249
4250
4251
4252
4253
4254
4255
4256
4257
4258
4259
4260
4261
4262
4263
4264
4265
4266
4267
4268
4269
4270
4271
4272
4273
4274
4275
4276
4277
4278
4279
4280
4281
4282
4283
4284
4285
4286
4287
4288
4289
4290
4291
4292
4293
4294
4295
4296
4297
4298
4299
4300
4301
4302
4303
4304
4305
4306
4307
4308
4309
4310
4311
4312
4313
4314
4315
4316
4317
4318
4319
4320
4321
4322
4323
4324
4325
4326
4327
4328
4329
4330
4331
4332
4333
4334
4335
4336
4337
4338
4339
4340
4341
4342
4343
4344
4345
4346
4347
4348
4349
4350
4351
4352
4353
4354
4355
4356
4357
4358
4359
4360
4361
4362
4363
4364
4365
4366
4367
4368
4369
4370
4371
4372
4373
4374
4375
4376
4377
4378
4379
4380
4381
4382
4383
4384
4385
4386
4387
4388
4389
4390
4391
4392
4393
4394
4395
4396
4397
4398
4399
4400
4401
4402
4403
4404
4405
4406
4407
4408
4409
4410
4411
4412
4413
4414
4415
4416
4417
4418
4419
4420
4421
4422
4423
4424
4425
4426
4427
4428
4429
4430
4431
4432
4433
4434
4435
4436
4437
4438
4439
4440
4441
4442
4443
4444
4445
4446
4447
4448
4449
4450
4451
4452
4453
4454
4455
4456
4457
4458
4459
4460
4461
4462
4463
4464
4465
4466
4467
4468
4469
4470
4471
4472
4473
4474
4475
4476
4477
4478
4479
4480
4481
4482
4483
4484
4485
4486
4487
4488
4489
4490
4491
4492
4493
4494
4495
4496
4497
4498
4499
4500
4501
4502
4503
4504
4505
4506
4507
4508
4509
4510
4511
4512
4513
4514
4515
4516
4517
4518
4519
4520
4521
4522
4523
4524
4525
4526
4527
4528
4529
4530
4531
4532
4533
4534
4535
4536
4537
4538
4539
4540
4541
4542
4543
4544
4545
4546
4547
4548
4549
4550
4551
4552
4553
4554
4555
4556
4557
4558
4559
4560
4561
4562
4563
4564
4565
4566
4567
4568
4569
4570
4571
4572
4573
4574
4575
4576
4577
4578
4579
4580
4581
4582
4583
4584
4585
4586
4587
4588
4589
4590
4591
4592
4593
4594
4595
4596
4597
4598
4599
4600
4601
4602
4603
4604
4605
4606
4607
4608
4609
4610
4611
4612
4613
4614
4615
4616
4617
4618
4619
4620
4621
4622
4623
4624
4625
4626
4627
4628
4629
4630
4631
4632
4633
//===- ARMInstrNEON.td - NEON support for ARM -----------------------------===//
//
//                     The LLVM Compiler Infrastructure
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//
//
// This file describes the ARM NEON instruction set.
//
//===----------------------------------------------------------------------===//

//===----------------------------------------------------------------------===//
// NEON-specific DAG Nodes.
//===----------------------------------------------------------------------===//

def SDTARMVCMP    : SDTypeProfile<1, 2, [SDTCisInt<0>, SDTCisSameAs<1, 2>]>;
def SDTARMVCMPZ   : SDTypeProfile<1, 1, []>;

def NEONvceq      : SDNode<"ARMISD::VCEQ", SDTARMVCMP>;
def NEONvceqz     : SDNode<"ARMISD::VCEQZ", SDTARMVCMPZ>;
def NEONvcge      : SDNode<"ARMISD::VCGE", SDTARMVCMP>;
def NEONvcgez     : SDNode<"ARMISD::VCGEZ", SDTARMVCMPZ>;
def NEONvclez     : SDNode<"ARMISD::VCLEZ", SDTARMVCMPZ>;
def NEONvcgeu     : SDNode<"ARMISD::VCGEU", SDTARMVCMP>;
def NEONvcgt      : SDNode<"ARMISD::VCGT", SDTARMVCMP>;
def NEONvcgtz     : SDNode<"ARMISD::VCGTZ", SDTARMVCMPZ>;
def NEONvcltz     : SDNode<"ARMISD::VCLTZ", SDTARMVCMPZ>;
def NEONvcgtu     : SDNode<"ARMISD::VCGTU", SDTARMVCMP>;
def NEONvtst      : SDNode<"ARMISD::VTST", SDTARMVCMP>;

// Types for vector shift by immediates.  The "SHX" version is for long and
// narrow operations where the source and destination vectors have different
// types.  The "SHINS" version is for shift and insert operations.
def SDTARMVSH     : SDTypeProfile<1, 2, [SDTCisInt<0>, SDTCisSameAs<0, 1>,
                                         SDTCisVT<2, i32>]>;
def SDTARMVSHX    : SDTypeProfile<1, 2, [SDTCisInt<0>, SDTCisInt<1>,
                                         SDTCisVT<2, i32>]>;
def SDTARMVSHINS  : SDTypeProfile<1, 3, [SDTCisInt<0>, SDTCisSameAs<0, 1>,
                                         SDTCisSameAs<0, 2>, SDTCisVT<3, i32>]>;

def NEONvshl      : SDNode<"ARMISD::VSHL", SDTARMVSH>;
def NEONvshrs     : SDNode<"ARMISD::VSHRs", SDTARMVSH>;
def NEONvshru     : SDNode<"ARMISD::VSHRu", SDTARMVSH>;
def NEONvshlls    : SDNode<"ARMISD::VSHLLs", SDTARMVSHX>;
def NEONvshllu    : SDNode<"ARMISD::VSHLLu", SDTARMVSHX>;
def NEONvshlli    : SDNode<"ARMISD::VSHLLi", SDTARMVSHX>;
def NEONvshrn     : SDNode<"ARMISD::VSHRN", SDTARMVSHX>;

def NEONvrshrs    : SDNode<"ARMISD::VRSHRs", SDTARMVSH>;
def NEONvrshru    : SDNode<"ARMISD::VRSHRu", SDTARMVSH>;
def NEONvrshrn    : SDNode<"ARMISD::VRSHRN", SDTARMVSHX>;

def NEONvqshls    : SDNode<"ARMISD::VQSHLs", SDTARMVSH>;
def NEONvqshlu    : SDNode<"ARMISD::VQSHLu", SDTARMVSH>;
def NEONvqshlsu   : SDNode<"ARMISD::VQSHLsu", SDTARMVSH>;
def NEONvqshrns   : SDNode<"ARMISD::VQSHRNs", SDTARMVSHX>;
def NEONvqshrnu   : SDNode<"ARMISD::VQSHRNu", SDTARMVSHX>;
def NEONvqshrnsu  : SDNode<"ARMISD::VQSHRNsu", SDTARMVSHX>;

def NEONvqrshrns  : SDNode<"ARMISD::VQRSHRNs", SDTARMVSHX>;
def NEONvqrshrnu  : SDNode<"ARMISD::VQRSHRNu", SDTARMVSHX>;
def NEONvqrshrnsu : SDNode<"ARMISD::VQRSHRNsu", SDTARMVSHX>;

def NEONvsli      : SDNode<"ARMISD::VSLI", SDTARMVSHINS>;
def NEONvsri      : SDNode<"ARMISD::VSRI", SDTARMVSHINS>;

def SDTARMVGETLN  : SDTypeProfile<1, 2, [SDTCisVT<0, i32>, SDTCisInt<1>,
                                         SDTCisVT<2, i32>]>;
def NEONvgetlaneu : SDNode<"ARMISD::VGETLANEu", SDTARMVGETLN>;
def NEONvgetlanes : SDNode<"ARMISD::VGETLANEs", SDTARMVGETLN>;

def SDTARMVMOVIMM : SDTypeProfile<1, 1, [SDTCisVec<0>, SDTCisVT<1, i32>]>;
def NEONvmovImm   : SDNode<"ARMISD::VMOVIMM", SDTARMVMOVIMM>;
def NEONvmvnImm   : SDNode<"ARMISD::VMVNIMM", SDTARMVMOVIMM>;

def SDTARMVORRIMM : SDTypeProfile<1, 2, [SDTCisVec<0>, SDTCisSameAs<0, 1>,
                                           SDTCisVT<2, i32>]>;
def NEONvorrImm   : SDNode<"ARMISD::VORRIMM", SDTARMVORRIMM>;
def NEONvbicImm   : SDNode<"ARMISD::VBICIMM", SDTARMVORRIMM>;

def NEONvdup      : SDNode<"ARMISD::VDUP", SDTypeProfile<1, 1, [SDTCisVec<0>]>>;

// VDUPLANE can produce a quad-register result from a double-register source,
// so the result is not constrained to match the source.
def NEONvduplane  : SDNode<"ARMISD::VDUPLANE",
                           SDTypeProfile<1, 2, [SDTCisVec<0>, SDTCisVec<1>,
                                                SDTCisVT<2, i32>]>>;

def SDTARMVEXT    : SDTypeProfile<1, 3, [SDTCisVec<0>, SDTCisSameAs<0, 1>,
                                         SDTCisSameAs<0, 2>, SDTCisVT<3, i32>]>;
def NEONvext      : SDNode<"ARMISD::VEXT", SDTARMVEXT>;

def SDTARMVSHUF   : SDTypeProfile<1, 1, [SDTCisVec<0>, SDTCisSameAs<0, 1>]>;
def NEONvrev64    : SDNode<"ARMISD::VREV64", SDTARMVSHUF>;
def NEONvrev32    : SDNode<"ARMISD::VREV32", SDTARMVSHUF>;
def NEONvrev16    : SDNode<"ARMISD::VREV16", SDTARMVSHUF>;

def SDTARMVSHUF2  : SDTypeProfile<2, 2, [SDTCisVec<0>, SDTCisSameAs<0, 1>,
                                         SDTCisSameAs<0, 2>,
                                         SDTCisSameAs<0, 3>]>;
def NEONzip       : SDNode<"ARMISD::VZIP", SDTARMVSHUF2>;
def NEONuzp       : SDNode<"ARMISD::VUZP", SDTARMVSHUF2>;
def NEONtrn       : SDNode<"ARMISD::VTRN", SDTARMVSHUF2>;

def SDTARMVMULL   : SDTypeProfile<1, 2, [SDTCisInt<0>, SDTCisInt<1>,
                                         SDTCisSameAs<1, 2>]>;
def NEONvmulls    : SDNode<"ARMISD::VMULLs", SDTARMVMULL>;
def NEONvmullu    : SDNode<"ARMISD::VMULLu", SDTARMVMULL>;

def SDTARMFMAX    : SDTypeProfile<1, 2, [SDTCisVT<0, f32>, SDTCisSameAs<0, 1>,
                                         SDTCisSameAs<0, 2>]>;
def NEONfmax      : SDNode<"ARMISD::FMAX", SDTARMFMAX>;
def NEONfmin      : SDNode<"ARMISD::FMIN", SDTARMFMAX>;

def NEONimmAllZerosV: PatLeaf<(NEONvmovImm (i32 timm)), [{
  ConstantSDNode *ConstVal = cast<ConstantSDNode>(N->getOperand(0));
  unsigned EltBits = 0;
  uint64_t EltVal = ARM_AM::decodeNEONModImm(ConstVal->getZExtValue(), EltBits);
  return (EltBits == 32 && EltVal == 0);
}]>;

def NEONimmAllOnesV: PatLeaf<(NEONvmovImm (i32 timm)), [{
  ConstantSDNode *ConstVal = cast<ConstantSDNode>(N->getOperand(0));
  unsigned EltBits = 0;
  uint64_t EltVal = ARM_AM::decodeNEONModImm(ConstVal->getZExtValue(), EltBits);
  return (EltBits == 8 && EltVal == 0xff);
}]>;

//===----------------------------------------------------------------------===//
// NEON operand definitions
//===----------------------------------------------------------------------===//

def nModImm : Operand<i32> {
  let PrintMethod = "printNEONModImmOperand";
}

//===----------------------------------------------------------------------===//
// NEON load / store instructions
//===----------------------------------------------------------------------===//

// Use VLDM to load a Q register as a D register pair.
// This is a pseudo instruction that is expanded to VLDMD after reg alloc.
def VLDMQ
  : PseudoVFPLdStM<(outs QPR:$dst), (ins GPR:$Rn, ldstm_mode:$mode),
                    IIC_fpLoad_m, "",
                   [(set QPR:$dst, (v2f64 (load GPR:$Rn)))]>;

// Use VSTM to store a Q register as a D register pair.
// This is a pseudo instruction that is expanded to VSTMD after reg alloc.
def VSTMQ
  : PseudoVFPLdStM<(outs), (ins QPR:$src, GPR:$Rn, ldstm_mode:$mode),
                    IIC_fpStore_m, "",
                   [(store (v2f64 QPR:$src), GPR:$Rn)]>;

let mayLoad = 1, neverHasSideEffects = 1, hasExtraDefRegAllocReq = 1 in {

// Classes for VLD* pseudo-instructions with multi-register operands.
// These are expanded to real instructions after register allocation.
class VLDQPseudo<InstrItinClass itin>
  : PseudoNLdSt<(outs QPR:$dst), (ins addrmode6:$addr), itin, "">;
class VLDQWBPseudo<InstrItinClass itin>
  : PseudoNLdSt<(outs QPR:$dst, GPR:$wb),
                (ins addrmode6:$addr, am6offset:$offset), itin,
                "$addr.addr = $wb">;
class VLDQQPseudo<InstrItinClass itin>
  : PseudoNLdSt<(outs QQPR:$dst), (ins addrmode6:$addr), itin, "">;
class VLDQQWBPseudo<InstrItinClass itin>
  : PseudoNLdSt<(outs QQPR:$dst, GPR:$wb),
                (ins addrmode6:$addr, am6offset:$offset), itin,
                "$addr.addr = $wb">;
class VLDQQQQWBPseudo<InstrItinClass itin>
  : PseudoNLdSt<(outs QQQQPR:$dst, GPR:$wb),
                (ins addrmode6:$addr, am6offset:$offset, QQQQPR:$src), itin,
                "$addr.addr = $wb, $src = $dst">;

//   VLD1     : Vector Load (multiple single elements)
class VLD1D<bits<4> op7_4, string Dt>
  : NLdSt<0,0b10,0b0111,op7_4, (outs DPR:$Vd),
          (ins addrmode6:$Rn), IIC_VLD1,
          "vld1", Dt, "\\{$Vd\\}, $Rn", "", []> {
  let Rm = 0b1111;
  let Inst{4} = Rn{4};
}
class VLD1Q<bits<4> op7_4, string Dt>
  : NLdSt<0,0b10,0b1010,op7_4, (outs DPR:$Vd, DPR:$dst2),
          (ins addrmode6:$Rn), IIC_VLD1x2,
          "vld1", Dt, "\\{$Vd, $dst2\\}, $Rn", "", []> {
  let Rm = 0b1111;
  let Inst{5-4} = Rn{5-4};
}

def  VLD1d8   : VLD1D<{0,0,0,?}, "8">;
def  VLD1d16  : VLD1D<{0,1,0,?}, "16">;
def  VLD1d32  : VLD1D<{1,0,0,?}, "32">;
def  VLD1d64  : VLD1D<{1,1,0,?}, "64">;

def  VLD1q8   : VLD1Q<{0,0,?,?}, "8">;
def  VLD1q16  : VLD1Q<{0,1,?,?}, "16">;
def  VLD1q32  : VLD1Q<{1,0,?,?}, "32">;
def  VLD1q64  : VLD1Q<{1,1,?,?}, "64">;

def  VLD1q8Pseudo  : VLDQPseudo<IIC_VLD1x2>;
def  VLD1q16Pseudo : VLDQPseudo<IIC_VLD1x2>;
def  VLD1q32Pseudo : VLDQPseudo<IIC_VLD1x2>;
def  VLD1q64Pseudo : VLDQPseudo<IIC_VLD1x2>;

// ...with address register writeback:
class VLD1DWB<bits<4> op7_4, string Dt>
  : NLdSt<0,0b10,0b0111,op7_4, (outs DPR:$Vd, GPR:$wb),
          (ins addrmode6:$Rn, am6offset:$Rm), IIC_VLD1u,
          "vld1", Dt, "\\{$Vd\\}, $Rn$Rm",
          "$Rn.addr = $wb", []> {
  let Inst{4} = Rn{4}; 
}
class VLD1QWB<bits<4> op7_4, string Dt>
  : NLdSt<0,0b10,0b1010,op7_4, (outs DPR:$Vd, DPR:$dst2, GPR:$wb),
          (ins addrmode6:$Rn, am6offset:$Rm), IIC_VLD1x2u,
          "vld1", Dt, "\\{$Vd, $dst2\\}, $Rn$Rm",
          "$Rn.addr = $wb", []> {
  let Inst{5-4} = Rn{5-4};
}

def VLD1d8_UPD  : VLD1DWB<{0,0,0,?}, "8">;
def VLD1d16_UPD : VLD1DWB<{0,1,0,?}, "16">;
def VLD1d32_UPD : VLD1DWB<{1,0,0,?}, "32">;
def VLD1d64_UPD : VLD1DWB<{1,1,0,?}, "64">;

def VLD1q8_UPD  : VLD1QWB<{0,0,?,?}, "8">;
def VLD1q16_UPD : VLD1QWB<{0,1,?,?}, "16">;
def VLD1q32_UPD : VLD1QWB<{1,0,?,?}, "32">;
def VLD1q64_UPD : VLD1QWB<{1,1,?,?}, "64">;

def VLD1q8Pseudo_UPD  : VLDQWBPseudo<IIC_VLD1x2u>;
def VLD1q16Pseudo_UPD : VLDQWBPseudo<IIC_VLD1x2u>;
def VLD1q32Pseudo_UPD : VLDQWBPseudo<IIC_VLD1x2u>;
def VLD1q64Pseudo_UPD : VLDQWBPseudo<IIC_VLD1x2u>;

// ...with 3 registers (some of these are only for the disassembler):
class VLD1D3<bits<4> op7_4, string Dt>
  : NLdSt<0,0b10,0b0110,op7_4, (outs DPR:$Vd, DPR:$dst2, DPR:$dst3),
          (ins addrmode6:$Rn), IIC_VLD1x3, "vld1", Dt,
          "\\{$Vd, $dst2, $dst3\\}, $Rn", "", []> {
  let Rm = 0b1111;
  let Inst{4} = Rn{4};
}
class VLD1D3WB<bits<4> op7_4, string Dt>
  : NLdSt<0,0b10,0b0110,op7_4, (outs DPR:$Vd, DPR:$dst2, DPR:$dst3, GPR:$wb),
          (ins addrmode6:$Rn, am6offset:$Rm), IIC_VLD1x3u, "vld1", Dt,
          "\\{$Vd, $dst2, $dst3\\}, $Rn$Rm", "$Rn.addr = $wb", []> {
  let Inst{4} = Rn{4};
}

def VLD1d8T      : VLD1D3<{0,0,0,?}, "8">;
def VLD1d16T     : VLD1D3<{0,1,0,?}, "16">;
def VLD1d32T     : VLD1D3<{1,0,0,?}, "32">;
def VLD1d64T     : VLD1D3<{1,1,0,?}, "64">;

def VLD1d8T_UPD  : VLD1D3WB<{0,0,0,?}, "8">;
def VLD1d16T_UPD : VLD1D3WB<{0,1,0,?}, "16">;
def VLD1d32T_UPD : VLD1D3WB<{1,0,0,?}, "32">;
def VLD1d64T_UPD : VLD1D3WB<{1,1,0,?}, "64">;

def VLD1d64TPseudo     : VLDQQPseudo<IIC_VLD1x3>;
def VLD1d64TPseudo_UPD : VLDQQWBPseudo<IIC_VLD1x3u>;

// ...with 4 registers (some of these are only for the disassembler):
class VLD1D4<bits<4> op7_4, string Dt>
  : NLdSt<0,0b10,0b0010,op7_4,(outs DPR:$Vd, DPR:$dst2, DPR:$dst3, DPR:$dst4),
          (ins addrmode6:$Rn), IIC_VLD1x4, "vld1", Dt,
          "\\{$Vd, $dst2, $dst3, $dst4\\}, $Rn", "", []> {
  let Rm = 0b1111;
  let Inst{5-4} = Rn{5-4};
}
class VLD1D4WB<bits<4> op7_4, string Dt>
  : NLdSt<0,0b10,0b0010,op7_4,
          (outs DPR:$Vd, DPR:$dst2, DPR:$dst3, DPR:$dst4, GPR:$wb),
          (ins addrmode6:$Rn, am6offset:$Rm), IIC_VLD4, "vld1", Dt,
          "\\{$Vd, $dst2, $dst3, $dst4\\}, $Rn$Rm", "$Rn.addr = $wb",
          []> {
  let Inst{5-4} = Rn{5-4};
}

def VLD1d8Q      : VLD1D4<{0,0,?,?}, "8">;
def VLD1d16Q     : VLD1D4<{0,1,?,?}, "16">;
def VLD1d32Q     : VLD1D4<{1,0,?,?}, "32">;
def VLD1d64Q     : VLD1D4<{1,1,?,?}, "64">;

def VLD1d8Q_UPD  : VLD1D4WB<{0,0,?,?}, "8">;
def VLD1d16Q_UPD : VLD1D4WB<{0,1,?,?}, "16">;
def VLD1d32Q_UPD : VLD1D4WB<{1,0,?,?}, "32">;
def VLD1d64Q_UPD : VLD1D4WB<{1,1,?,?}, "64">;

def VLD1d64QPseudo     : VLDQQPseudo<IIC_VLD1x4>;
def VLD1d64QPseudo_UPD : VLDQQWBPseudo<IIC_VLD1x4u>;

//   VLD2     : Vector Load (multiple 2-element structures)
class VLD2D<bits<4> op11_8, bits<4> op7_4, string Dt>
  : NLdSt<0, 0b10, op11_8, op7_4, (outs DPR:$Vd, DPR:$dst2),
          (ins addrmode6:$Rn), IIC_VLD2,
          "vld2", Dt, "\\{$Vd, $dst2\\}, $Rn", "", []> {
  let Rm = 0b1111;
  let Inst{5-4} = Rn{5-4};
}
class VLD2Q<bits<4> op7_4, string Dt>
  : NLdSt<0, 0b10, 0b0011, op7_4,
          (outs DPR:$Vd, DPR:$dst2, DPR:$dst3, DPR:$dst4),
          (ins addrmode6:$Rn), IIC_VLD2x2,
          "vld2", Dt, "\\{$Vd, $dst2, $dst3, $dst4\\}, $Rn", "", []> {
  let Rm = 0b1111;
  let Inst{5-4} = Rn{5-4};
}

def  VLD2d8   : VLD2D<0b1000, {0,0,?,?}, "8">;
def  VLD2d16  : VLD2D<0b1000, {0,1,?,?}, "16">;
def  VLD2d32  : VLD2D<0b1000, {1,0,?,?}, "32">;

def  VLD2q8   : VLD2Q<{0,0,?,?}, "8">;
def  VLD2q16  : VLD2Q<{0,1,?,?}, "16">;
def  VLD2q32  : VLD2Q<{1,0,?,?}, "32">;

def  VLD2d8Pseudo  : VLDQPseudo<IIC_VLD2>;
def  VLD2d16Pseudo : VLDQPseudo<IIC_VLD2>;
def  VLD2d32Pseudo : VLDQPseudo<IIC_VLD2>;

def  VLD2q8Pseudo  : VLDQQPseudo<IIC_VLD2x2>;
def  VLD2q16Pseudo : VLDQQPseudo<IIC_VLD2x2>;
def  VLD2q32Pseudo : VLDQQPseudo<IIC_VLD2x2>;

// ...with address register writeback:
class VLD2DWB<bits<4> op11_8, bits<4> op7_4, string Dt>
  : NLdSt<0, 0b10, op11_8, op7_4, (outs DPR:$Vd, DPR:$dst2, GPR:$wb),
          (ins addrmode6:$Rn, am6offset:$Rm), IIC_VLD2u,
          "vld2", Dt, "\\{$Vd, $dst2\\}, $Rn$Rm",
          "$Rn.addr = $wb", []> {
  let Inst{5-4} = Rn{5-4};
}
class VLD2QWB<bits<4> op7_4, string Dt>
  : NLdSt<0, 0b10, 0b0011, op7_4,
          (outs DPR:$Vd, DPR:$dst2, DPR:$dst3, DPR:$dst4, GPR:$wb),
          (ins addrmode6:$Rn, am6offset:$Rm), IIC_VLD2x2u,
          "vld2", Dt, "\\{$Vd, $dst2, $dst3, $dst4\\}, $Rn$Rm",
          "$Rn.addr = $wb", []> {
  let Inst{5-4} = Rn{5-4};
}

def VLD2d8_UPD  : VLD2DWB<0b1000, {0,0,?,?}, "8">;
def VLD2d16_UPD : VLD2DWB<0b1000, {0,1,?,?}, "16">;
def VLD2d32_UPD : VLD2DWB<0b1000, {1,0,?,?}, "32">;

def VLD2q8_UPD  : VLD2QWB<{0,0,?,?}, "8">;
def VLD2q16_UPD : VLD2QWB<{0,1,?,?}, "16">;
def VLD2q32_UPD : VLD2QWB<{1,0,?,?}, "32">;

def VLD2d8Pseudo_UPD  : VLDQWBPseudo<IIC_VLD2u>;
def VLD2d16Pseudo_UPD : VLDQWBPseudo<IIC_VLD2u>;
def VLD2d32Pseudo_UPD : VLDQWBPseudo<IIC_VLD2u>;

def VLD2q8Pseudo_UPD  : VLDQQWBPseudo<IIC_VLD2x2u>;
def VLD2q16Pseudo_UPD : VLDQQWBPseudo<IIC_VLD2x2u>;
def VLD2q32Pseudo_UPD : VLDQQWBPseudo<IIC_VLD2x2u>;

// ...with double-spaced registers (for disassembly only):
def VLD2b8      : VLD2D<0b1001, {0,0,?,?}, "8">;
def VLD2b16     : VLD2D<0b1001, {0,1,?,?}, "16">;
def VLD2b32     : VLD2D<0b1001, {1,0,?,?}, "32">;
def VLD2b8_UPD  : VLD2DWB<0b1001, {0,0,?,?}, "8">;
def VLD2b16_UPD : VLD2DWB<0b1001, {0,1,?,?}, "16">;
def VLD2b32_UPD : VLD2DWB<0b1001, {1,0,?,?}, "32">;

//   VLD3     : Vector Load (multiple 3-element structures)
class VLD3D<bits<4> op11_8, bits<4> op7_4, string Dt>
  : NLdSt<0, 0b10, op11_8, op7_4, (outs DPR:$Vd, DPR:$dst2, DPR:$dst3),
          (ins addrmode6:$Rn), IIC_VLD3,
          "vld3", Dt, "\\{$Vd, $dst2, $dst3\\}, $Rn", "", []> {
  let Rm = 0b1111;
  let Inst{4} = Rn{4};
}

def  VLD3d8   : VLD3D<0b0100, {0,0,0,?}, "8">;
def  VLD3d16  : VLD3D<0b0100, {0,1,0,?}, "16">;
def  VLD3d32  : VLD3D<0b0100, {1,0,0,?}, "32">;

def  VLD3d8Pseudo  : VLDQQPseudo<IIC_VLD3>;
def  VLD3d16Pseudo : VLDQQPseudo<IIC_VLD3>;
def  VLD3d32Pseudo : VLDQQPseudo<IIC_VLD3>;

// ...with address register writeback:
class VLD3DWB<bits<4> op11_8, bits<4> op7_4, string Dt>
  : NLdSt<0, 0b10, op11_8, op7_4,
          (outs DPR:$Vd, DPR:$dst2, DPR:$dst3, GPR:$wb),
          (ins addrmode6:$Rn, am6offset:$Rm), IIC_VLD3u,
          "vld3", Dt, "\\{$Vd, $dst2, $dst3\\}, $Rn$Rm",
          "$Rn.addr = $wb", []> {
  let Inst{4} = Rn{4};
}

def VLD3d8_UPD  : VLD3DWB<0b0100, {0,0,0,?}, "8">;
def VLD3d16_UPD : VLD3DWB<0b0100, {0,1,0,?}, "16">;
def VLD3d32_UPD : VLD3DWB<0b0100, {1,0,0,?}, "32">;

def VLD3d8Pseudo_UPD  : VLDQQWBPseudo<IIC_VLD3u>;
def VLD3d16Pseudo_UPD : VLDQQWBPseudo<IIC_VLD3u>;
def VLD3d32Pseudo_UPD : VLDQQWBPseudo<IIC_VLD3u>;

// ...with double-spaced registers (non-updating versions for disassembly only):
def VLD3q8      : VLD3D<0b0101, {0,0,0,?}, "8">;
def VLD3q16     : VLD3D<0b0101, {0,1,0,?}, "16">;
def VLD3q32     : VLD3D<0b0101, {1,0,0,?}, "32">;
def VLD3q8_UPD  : VLD3DWB<0b0101, {0,0,0,?}, "8">;
def VLD3q16_UPD : VLD3DWB<0b0101, {0,1,0,?}, "16">;
def VLD3q32_UPD : VLD3DWB<0b0101, {1,0,0,?}, "32">;

def VLD3q8Pseudo_UPD  : VLDQQQQWBPseudo<IIC_VLD3u>;
def VLD3q16Pseudo_UPD : VLDQQQQWBPseudo<IIC_VLD3u>;
def VLD3q32Pseudo_UPD : VLDQQQQWBPseudo<IIC_VLD3u>;

// ...alternate versions to be allocated odd register numbers:
def VLD3q8oddPseudo_UPD  : VLDQQQQWBPseudo<IIC_VLD3u>;
def VLD3q16oddPseudo_UPD : VLDQQQQWBPseudo<IIC_VLD3u>;
def VLD3q32oddPseudo_UPD : VLDQQQQWBPseudo<IIC_VLD3u>;

//   VLD4     : Vector Load (multiple 4-element structures)
class VLD4D<bits<4> op11_8, bits<4> op7_4, string Dt>
  : NLdSt<0, 0b10, op11_8, op7_4,
          (outs DPR:$Vd, DPR:$dst2, DPR:$dst3, DPR:$dst4),
          (ins addrmode6:$Rn), IIC_VLD4,
          "vld4", Dt, "\\{$Vd, $dst2, $dst3, $dst4\\}, $Rn", "", []> {
  let Rm = 0b1111;
  let Inst{5-4} = Rn{5-4};
}

def  VLD4d8   : VLD4D<0b0000, {0,0,?,?}, "8">;
def  VLD4d16  : VLD4D<0b0000, {0,1,?,?}, "16">;
def  VLD4d32  : VLD4D<0b0000, {1,0,?,?}, "32">;

def  VLD4d8Pseudo  : VLDQQPseudo<IIC_VLD4>;
def  VLD4d16Pseudo : VLDQQPseudo<IIC_VLD4>;
def  VLD4d32Pseudo : VLDQQPseudo<IIC_VLD4>;

// ...with address register writeback:
class VLD4DWB<bits<4> op11_8, bits<4> op7_4, string Dt>
  : NLdSt<0, 0b10, op11_8, op7_4,
          (outs DPR:$Vd, DPR:$dst2, DPR:$dst3, DPR:$dst4, GPR:$wb),
          (ins addrmode6:$Rn, am6offset:$Rm), IIC_VLD4,
          "vld4", Dt, "\\{$Vd, $dst2, $dst3, $dst4\\}, $Rn$Rm",
          "$Rn.addr = $wb", []> {
  let Inst{5-4} = Rn{5-4};
}

def VLD4d8_UPD  : VLD4DWB<0b0000, {0,0,?,?}, "8">;
def VLD4d16_UPD : VLD4DWB<0b0000, {0,1,?,?}, "16">;
def VLD4d32_UPD : VLD4DWB<0b0000, {1,0,?,?}, "32">;

def VLD4d8Pseudo_UPD  : VLDQQWBPseudo<IIC_VLD4>;
def VLD4d16Pseudo_UPD : VLDQQWBPseudo<IIC_VLD4>;
def VLD4d32Pseudo_UPD : VLDQQWBPseudo<IIC_VLD4>;

// ...with double-spaced registers (non-updating versions for disassembly only):
def VLD4q8      : VLD4D<0b0001, {0,0,?,?}, "8">;
def VLD4q16     : VLD4D<0b0001, {0,1,?,?}, "16">;
def VLD4q32     : VLD4D<0b0001, {1,0,?,?}, "32">;
def VLD4q8_UPD  : VLD4DWB<0b0001, {0,0,?,?}, "8">;
def VLD4q16_UPD : VLD4DWB<0b0001, {0,1,?,?}, "16">;
def VLD4q32_UPD : VLD4DWB<0b0001, {1,0,?,?}, "32">;

def VLD4q8Pseudo_UPD  : VLDQQQQWBPseudo<IIC_VLD4>;
def VLD4q16Pseudo_UPD : VLDQQQQWBPseudo<IIC_VLD4>;
def VLD4q32Pseudo_UPD : VLDQQQQWBPseudo<IIC_VLD4>;

// ...alternate versions to be allocated odd register numbers:
def VLD4q8oddPseudo_UPD  : VLDQQQQWBPseudo<IIC_VLD4>;
def VLD4q16oddPseudo_UPD : VLDQQQQWBPseudo<IIC_VLD4>;
def VLD4q32oddPseudo_UPD : VLDQQQQWBPseudo<IIC_VLD4>;

} // mayLoad = 1, neverHasSideEffects = 1, hasExtraDefRegAllocReq = 1

// Classes for VLD*LN pseudo-instructions with multi-register operands.
// These are expanded to real instructions after register allocation.
class VLDQLNPseudo<InstrItinClass itin>
  : PseudoNLdSt<(outs QPR:$dst),
                (ins addrmode6:$addr, QPR:$src, nohash_imm:$lane),
                itin, "$src = $dst">;
class VLDQLNWBPseudo<InstrItinClass itin>
  : PseudoNLdSt<(outs QPR:$dst, GPR:$wb),
                (ins addrmode6:$addr, am6offset:$offset, QPR:$src,
                 nohash_imm:$lane), itin, "$addr.addr = $wb, $src = $dst">;
class VLDQQLNPseudo<InstrItinClass itin>
  : PseudoNLdSt<(outs QQPR:$dst),
                (ins addrmode6:$addr, QQPR:$src, nohash_imm:$lane),
                itin, "$src = $dst">;
class VLDQQLNWBPseudo<InstrItinClass itin>
  : PseudoNLdSt<(outs QQPR:$dst, GPR:$wb),
                (ins addrmode6:$addr, am6offset:$offset, QQPR:$src,
                 nohash_imm:$lane), itin, "$addr.addr = $wb, $src = $dst">;
class VLDQQQQLNPseudo<InstrItinClass itin>
  : PseudoNLdSt<(outs QQQQPR:$dst),
                (ins addrmode6:$addr, QQQQPR:$src, nohash_imm:$lane),
                itin, "$src = $dst">;
class VLDQQQQLNWBPseudo<InstrItinClass itin>
  : PseudoNLdSt<(outs QQQQPR:$dst, GPR:$wb),
                (ins addrmode6:$addr, am6offset:$offset, QQQQPR:$src,
                 nohash_imm:$lane), itin, "$addr.addr = $wb, $src = $dst">;

//   VLD1LN   : Vector Load (single element to one lane)
class VLD1LN<bits<4> op11_8, bits<4> op7_4, string Dt, ValueType Ty,
             PatFrag LoadOp>
  : NLdStLn<1, 0b10, op11_8, op7_4, (outs DPR:$Vd),
          (ins addrmode6:$Rn, DPR:$src, nohash_imm:$lane),
          IIC_VLD1ln, "vld1", Dt, "\\{$Vd[$lane]\\}, $Rn",
          "$src = $Vd",
          [(set DPR:$Vd, (vector_insert (Ty DPR:$src),
                                         (i32 (LoadOp addrmode6:$Rn)),
                                         imm:$lane))]> {
  let Rm = 0b1111;
}
class VLD1QLNPseudo<ValueType Ty, PatFrag LoadOp> : VLDQLNPseudo<IIC_VLD1ln> {
  let Pattern = [(set QPR:$dst, (vector_insert (Ty QPR:$src),
                                               (i32 (LoadOp addrmode6:$addr)),
                                               imm:$lane))];
}

def VLD1LNd8  : VLD1LN<0b0000, {?,?,?,0}, "8", v8i8, extloadi8> {
  let Inst{7-5} = lane{2-0};
}
def VLD1LNd16 : VLD1LN<0b0100, {?,?,0,?}, "16", v4i16, extloadi16> {
  let Inst{7-6} = lane{1-0};
  let Inst{4}   = Rn{4};
}
def VLD1LNd32 : VLD1LN<0b1000, {?,0,?,?}, "32", v2i32, load> {
  let Inst{7} = lane{0};
  let Inst{5} = Rn{4};
  let Inst{4} = Rn{4};
}

def VLD1LNq8Pseudo  : VLD1QLNPseudo<v16i8, extloadi8>;
def VLD1LNq16Pseudo : VLD1QLNPseudo<v8i16, extloadi16>;
def VLD1LNq32Pseudo : VLD1QLNPseudo<v4i32, load>;

let mayLoad = 1, neverHasSideEffects = 1, hasExtraDefRegAllocReq = 1 in {

// ...with address register writeback:
class VLD1LNWB<bits<4> op11_8, bits<4> op7_4, string Dt>
  : NLdStLn<1, 0b10, op11_8, op7_4, (outs DPR:$Vd, GPR:$wb),
          (ins addrmode6:$Rn, am6offset:$Rm,
           DPR:$src, nohash_imm:$lane), IIC_VLD1lnu, "vld1", Dt,
          "\\{$Vd[$lane]\\}, $Rn$Rm",
          "$src = $Vd, $Rn.addr = $wb", []>;

def VLD1LNd8_UPD  : VLD1LNWB<0b0000, {?,?,?,0}, "8"> {
  let Inst{7-5} = lane{2-0};
}
def VLD1LNd16_UPD : VLD1LNWB<0b0100, {?,?,0,?}, "16"> {
  let Inst{7-6} = lane{1-0};
  let Inst{4}   = Rn{4};
}
def VLD1LNd32_UPD : VLD1LNWB<0b1000, {?,0,?,?}, "32"> {
  let Inst{7} = lane{0};
  let Inst{5} = Rn{4};
  let Inst{4} = Rn{4};
}

def VLD1LNq8Pseudo_UPD  : VLDQLNWBPseudo<IIC_VLD1lnu>;
def VLD1LNq16Pseudo_UPD : VLDQLNWBPseudo<IIC_VLD1lnu>;
def VLD1LNq32Pseudo_UPD : VLDQLNWBPseudo<IIC_VLD1lnu>;

//   VLD2LN   : Vector Load (single 2-element structure to one lane)
class VLD2LN<bits<4> op11_8, bits<4> op7_4, string Dt>
  : NLdStLn<1, 0b10, op11_8, op7_4, (outs DPR:$Vd, DPR:$dst2),
          (ins addrmode6:$Rn, DPR:$src1, DPR:$src2, nohash_imm:$lane),
          IIC_VLD2ln, "vld2", Dt, "\\{$Vd[$lane], $dst2[$lane]\\}, $Rn",
          "$src1 = $Vd, $src2 = $dst2", []> {
  let Rm = 0b1111;
  let Inst{4}   = Rn{4};
}

def VLD2LNd8  : VLD2LN<0b0001, {?,?,?,?}, "8"> {
  let Inst{7-5} = lane{2-0};
}
def VLD2LNd16 : VLD2LN<0b0101, {?,?,0,?}, "16"> {
  let Inst{7-6} = lane{1-0};
}
def VLD2LNd32 : VLD2LN<0b1001, {?,0,0,?}, "32"> {
  let Inst{7} = lane{0};
}

def VLD2LNd8Pseudo  : VLDQLNPseudo<IIC_VLD2ln>;
def VLD2LNd16Pseudo : VLDQLNPseudo<IIC_VLD2ln>;
def VLD2LNd32Pseudo : VLDQLNPseudo<IIC_VLD2ln>;

// ...with double-spaced registers:
def VLD2LNq16 : VLD2LN<0b0101, {?,?,1,?}, "16"> {
  let Inst{7-6} = lane{1-0};
}
def VLD2LNq32 : VLD2LN<0b1001, {?,1,0,?}, "32"> {
  let Inst{7} = lane{0};
}

def VLD2LNq16Pseudo : VLDQQLNPseudo<IIC_VLD2ln>;
def VLD2LNq32Pseudo : VLDQQLNPseudo<IIC_VLD2ln>;

// ...with address register writeback:
class VLD2LNWB<bits<4> op11_8, bits<4> op7_4, string Dt>
  : NLdStLn<1, 0b10, op11_8, op7_4, (outs DPR:$Vd, DPR:$dst2, GPR:$wb),
          (ins addrmode6:$Rn, am6offset:$Rm,
           DPR:$src1, DPR:$src2, nohash_imm:$lane), IIC_VLD2lnu, "vld2", Dt,
          "\\{$Vd[$lane], $dst2[$lane]\\}, $Rn$Rm",
          "$src1 = $Vd, $src2 = $dst2, $Rn.addr = $wb", []> {
  let Inst{4}   = Rn{4};
}

def VLD2LNd8_UPD  : VLD2LNWB<0b0001, {?,?,?,?}, "8"> {
  let Inst{7-5} = lane{2-0};
}
def VLD2LNd16_UPD : VLD2LNWB<0b0101, {?,?,0,?}, "16"> {
  let Inst{7-6} = lane{1-0};
}
def VLD2LNd32_UPD : VLD2LNWB<0b1001, {?,0,0,?}, "32"> {
  let Inst{7} = lane{0};
}

def VLD2LNd8Pseudo_UPD  : VLDQLNWBPseudo<IIC_VLD2lnu>;
def VLD2LNd16Pseudo_UPD : VLDQLNWBPseudo<IIC_VLD2lnu>;
def VLD2LNd32Pseudo_UPD : VLDQLNWBPseudo<IIC_VLD2lnu>;

def VLD2LNq16_UPD : VLD2LNWB<0b0101, {?,?,1,?}, "16"> {
  let Inst{7-6} = lane{1-0};
}
def VLD2LNq32_UPD : VLD2LNWB<0b1001, {?,1,0,?}, "32"> {
  let Inst{7} = lane{0};
}

def VLD2LNq16Pseudo_UPD : VLDQQLNWBPseudo<IIC_VLD2lnu>;
def VLD2LNq32Pseudo_UPD : VLDQQLNWBPseudo<IIC_VLD2lnu>;

//   VLD3LN   : Vector Load (single 3-element structure to one lane)
class VLD3LN<bits<4> op11_8, bits<4> op7_4, string Dt>
  : NLdStLn<1, 0b10, op11_8, op7_4, (outs DPR:$Vd, DPR:$dst2, DPR:$dst3),
          (ins addrmode6:$Rn, DPR:$src1, DPR:$src2, DPR:$src3,
          nohash_imm:$lane), IIC_VLD3ln, "vld3", Dt,
          "\\{$Vd[$lane], $dst2[$lane], $dst3[$lane]\\}, $Rn",
          "$src1 = $Vd, $src2 = $dst2, $src3 = $dst3", []> {
  let Rm = 0b1111;
}

def VLD3LNd8  : VLD3LN<0b0010, {?,?,?,0}, "8"> {
  let Inst{7-5} = lane{2-0};
}
def VLD3LNd16 : VLD3LN<0b0110, {?,?,0,0}, "16"> {
  let Inst{7-6} = lane{1-0};
}
def VLD3LNd32 : VLD3LN<0b1010, {?,0,0,0}, "32"> {
  let Inst{7}   = lane{0};
}

def VLD3LNd8Pseudo  : VLDQQLNPseudo<IIC_VLD3ln>;
def VLD3LNd16Pseudo : VLDQQLNPseudo<IIC_VLD3ln>;
def VLD3LNd32Pseudo : VLDQQLNPseudo<IIC_VLD3ln>;

// ...with double-spaced registers:
def VLD3LNq16 : VLD3LN<0b0110, {?,?,1,0}, "16"> {
  let Inst{7-6} = lane{1-0};
}
def VLD3LNq32 : VLD3LN<0b1010, {?,1,0,0}, "32"> {
  let Inst{7}   = lane{0};
}

def VLD3LNq16Pseudo : VLDQQQQLNPseudo<IIC_VLD3ln>;
def VLD3LNq32Pseudo : VLDQQQQLNPseudo<IIC_VLD3ln>;

// ...with address register writeback:
class VLD3LNWB<bits<4> op11_8, bits<4> op7_4, string Dt>
  : NLdStLn<1, 0b10, op11_8, op7_4,
          (outs DPR:$Vd, DPR:$dst2, DPR:$dst3, GPR:$wb),
          (ins addrmode6:$Rn, am6offset:$Rm,
           DPR:$src1, DPR:$src2, DPR:$src3, nohash_imm:$lane),
          IIC_VLD3lnu, "vld3", Dt,
          "\\{$Vd[$lane], $dst2[$lane], $dst3[$lane]\\}, $Rn$Rm",
          "$src1 = $Vd, $src2 = $dst2, $src3 = $dst3, $Rn.addr = $wb",
          []>;

def VLD3LNd8_UPD  : VLD3LNWB<0b0010, {?,?,?,0}, "8"> {
  let Inst{7-5} = lane{2-0};
}
def VLD3LNd16_UPD : VLD3LNWB<0b0110, {?,?,0,0}, "16"> {
  let Inst{7-6} = lane{1-0};
}
def VLD3LNd32_UPD : VLD3LNWB<0b1010, {?,0,0,0}, "32"> {
  let Inst{7}   = lane{0};
}

def VLD3LNd8Pseudo_UPD  : VLDQQLNWBPseudo<IIC_VLD3lnu>;
def VLD3LNd16Pseudo_UPD : VLDQQLNWBPseudo<IIC_VLD3lnu>;
def VLD3LNd32Pseudo_UPD : VLDQQLNWBPseudo<IIC_VLD3lnu>;

def VLD3LNq16_UPD : VLD3LNWB<0b0110, {?,?,1,0}, "16"> {
  let Inst{7-6} = lane{1-0};
}
def VLD3LNq32_UPD : VLD3LNWB<0b1010, {?,1,0,0}, "32"> {
  let Inst{7}   = lane{0};
}

def VLD3LNq16Pseudo_UPD : VLDQQQQLNWBPseudo<IIC_VLD3lnu>;
def VLD3LNq32Pseudo_UPD : VLDQQQQLNWBPseudo<IIC_VLD3lnu>;

//   VLD4LN   : Vector Load (single 4-element structure to one lane)
class VLD4LN<bits<4> op11_8, bits<4> op7_4, string Dt>
  : NLdStLn<1, 0b10, op11_8, op7_4,
          (outs DPR:$Vd, DPR:$dst2, DPR:$dst3, DPR:$dst4),
          (ins addrmode6:$Rn, DPR:$src1, DPR:$src2, DPR:$src3, DPR:$src4,
          nohash_imm:$lane), IIC_VLD4ln, "vld4", Dt,
          "\\{$Vd[$lane], $dst2[$lane], $dst3[$lane], $dst4[$lane]\\}, $Rn",
          "$src1 = $Vd, $src2 = $dst2, $src3 = $dst3, $src4 = $dst4", []> {
  let Rm = 0b1111;
  let Inst{4}   = Rn{4};
}

def VLD4LNd8  : VLD4LN<0b0011, {?,?,?,?}, "8"> {
  let Inst{7-5} = lane{2-0};
}
def VLD4LNd16 : VLD4LN<0b0111, {?,?,0,?}, "16"> {
  let Inst{7-6} = lane{1-0};
}
def VLD4LNd32 : VLD4LN<0b1011, {?,0,?,?}, "32"> {
  let Inst{7}   = lane{0};
  let Inst{5} = Rn{5};
}

def VLD4LNd8Pseudo  : VLDQQLNPseudo<IIC_VLD4ln>;
def VLD4LNd16Pseudo : VLDQQLNPseudo<IIC_VLD4ln>;
def VLD4LNd32Pseudo : VLDQQLNPseudo<IIC_VLD4ln>;

// ...with double-spaced registers:
def VLD4LNq16 : VLD4LN<0b0111, {?,?,1,?}, "16"> {
  let Inst{7-6} = lane{1-0};
}
def VLD4LNq32 : VLD4LN<0b1011, {?,1,?,?}, "32"> {
  let Inst{7}   = lane{0};
  let Inst{5} = Rn{5};
}

def VLD4LNq16Pseudo : VLDQQQQLNPseudo<IIC_VLD4ln>;
def VLD4LNq32Pseudo : VLDQQQQLNPseudo<IIC_VLD4ln>;

// ...with address register writeback:
class VLD4LNWB<bits<4> op11_8, bits<4> op7_4, string Dt>
  : NLdStLn<1, 0b10, op11_8, op7_4,
          (outs DPR:$Vd, DPR:$dst2, DPR:$dst3, DPR:$dst4, GPR:$wb),
          (ins addrmode6:$Rn, am6offset:$Rm,
           DPR:$src1, DPR:$src2, DPR:$src3, DPR:$src4, nohash_imm:$lane),
          IIC_VLD4ln, "vld4", Dt,
"\\{$Vd[$lane], $dst2[$lane], $dst3[$lane], $dst4[$lane]\\}, $Rn$Rm",
"$src1 = $Vd, $src2 = $dst2, $src3 = $dst3, $src4 = $dst4, $Rn.addr = $wb",
          []> {
  let Inst{4}   = Rn{4};
}

def VLD4LNd8_UPD  : VLD4LNWB<0b0011, {?,?,?,?}, "8"> {
  let Inst{7-5} = lane{2-0};
}
def VLD4LNd16_UPD : VLD4LNWB<0b0111, {?,?,0,?}, "16"> {
  let Inst{7-6} = lane{1-0};
}
def VLD4LNd32_UPD : VLD4LNWB<0b1011, {?,0,?,?}, "32"> {
  let Inst{7}   = lane{0};
  let Inst{5} = Rn{5};
}

def VLD4LNd8Pseudo_UPD  : VLDQQLNWBPseudo<IIC_VLD4lnu>;
def VLD4LNd16Pseudo_UPD : VLDQQLNWBPseudo<IIC_VLD4lnu>;
def VLD4LNd32Pseudo_UPD : VLDQQLNWBPseudo<IIC_VLD4lnu>;

def VLD4LNq16_UPD : VLD4LNWB<0b0111, {?,?,1,?}, "16"> {
  let Inst{7-6} = lane{1-0};
}
def VLD4LNq32_UPD : VLD4LNWB<0b1011, {?,1,?,?}, "32"> {
  let Inst{7}   = lane{0};
  let Inst{5} = Rn{5};
}

def VLD4LNq16Pseudo_UPD : VLDQQQQLNWBPseudo<IIC_VLD4lnu>;
def VLD4LNq32Pseudo_UPD : VLDQQQQLNWBPseudo<IIC_VLD4lnu>;

//   VLD1DUP  : Vector Load (single element to all lanes)
//   VLD2DUP  : Vector Load (single 2-element structure to all lanes)
//   VLD3DUP  : Vector Load (single 3-element structure to all lanes)
//   VLD4DUP  : Vector Load (single 4-element structure to all lanes)
//   FIXME: Not yet implemented.
} // mayLoad = 1, neverHasSideEffects = 1, hasExtraDefRegAllocReq = 1

let mayStore = 1, neverHasSideEffects = 1, hasExtraSrcRegAllocReq = 1 in {

// Classes for VST* pseudo-instructions with multi-register operands.
// These are expanded to real instructions after register allocation.
class VSTQPseudo<InstrItinClass itin>
  : PseudoNLdSt<(outs), (ins addrmode6:$addr, QPR:$src), itin, "">;
class VSTQWBPseudo<InstrItinClass itin>
  : PseudoNLdSt<(outs GPR:$wb),
                (ins addrmode6:$addr, am6offset:$offset, QPR:$src), itin,
                "$addr.addr = $wb">;
class VSTQQPseudo<InstrItinClass itin>
  : PseudoNLdSt<(outs), (ins addrmode6:$addr, QQPR:$src), itin, "">;
class VSTQQWBPseudo<InstrItinClass itin>
  : PseudoNLdSt<(outs GPR:$wb),
                (ins addrmode6:$addr, am6offset:$offset, QQPR:$src), itin,
                "$addr.addr = $wb">;
class VSTQQQQWBPseudo<InstrItinClass itin>
  : PseudoNLdSt<(outs GPR:$wb),
                (ins addrmode6:$addr, am6offset:$offset, QQQQPR:$src), itin,
                "$addr.addr = $wb">;

//   VST1     : Vector Store (multiple single elements)
class VST1D<bits<4> op7_4, string Dt>
  : NLdSt<0,0b00,0b0111,op7_4, (outs), (ins addrmode6:$Rn, DPR:$Vd),
          IIC_VST1, "vst1", Dt, "\\{$Vd\\}, $Rn", "", []> {
  let Rm = 0b1111;
  let Inst{4} = Rn{4};
}
class VST1Q<bits<4> op7_4, string Dt>
  : NLdSt<0,0b00,0b1010,op7_4, (outs),
          (ins addrmode6:$Rn, DPR:$Vd, DPR:$src2), IIC_VST1x2,
          "vst1", Dt, "\\{$Vd, $src2\\}, $Rn", "", []> {
  let Rm = 0b1111;
  let Inst{5-4} = Rn{5-4};
}

def  VST1d8   : VST1D<{0,0,0,?}, "8">;
def  VST1d16  : VST1D<{0,1,0,?}, "16">;
def  VST1d32  : VST1D<{1,0,0,?}, "32">;
def  VST1d64  : VST1D<{1,1,0,?}, "64">;

def  VST1q8   : VST1Q<{0,0,?,?}, "8">;
def  VST1q16  : VST1Q<{0,1,?,?}, "16">;
def  VST1q32  : VST1Q<{1,0,?,?}, "32">;
def  VST1q64  : VST1Q<{1,1,?,?}, "64">;

def  VST1q8Pseudo  : VSTQPseudo<IIC_VST1x2>;
def  VST1q16Pseudo : VSTQPseudo<IIC_VST1x2>;
def  VST1q32Pseudo : VSTQPseudo<IIC_VST1x2>;
def  VST1q64Pseudo : VSTQPseudo<IIC_VST1x2>;

// ...with address register writeback:
class VST1DWB<bits<4> op7_4, string Dt>
  : NLdSt<0, 0b00, 0b0111, op7_4, (outs GPR:$wb),
          (ins addrmode6:$Rn, am6offset:$Rm, DPR:$Vd), IIC_VST1u,
          "vst1", Dt, "\\{$Vd\\}, $Rn$Rm", "$Rn.addr = $wb", []> {
  let Inst{4} = Rn{4};
}
class VST1QWB<bits<4> op7_4, string Dt>
  : NLdSt<0, 0b00, 0b1010, op7_4, (outs GPR:$wb),
          (ins addrmode6:$Rn, am6offset:$Rm, DPR:$Vd, DPR:$src2),
          IIC_VST1x2u, "vst1", Dt, "\\{$Vd, $src2\\}, $Rn$Rm",
          "$Rn.addr = $wb", []> {
  let Inst{5-4} = Rn{5-4};
}

def VST1d8_UPD  : VST1DWB<{0,0,0,?}, "8">;
def VST1d16_UPD : VST1DWB<{0,1,0,?}, "16">;
def VST1d32_UPD : VST1DWB<{1,0,0,?}, "32">;
def VST1d64_UPD : VST1DWB<{1,1,0,?}, "64">;

def VST1q8_UPD  : VST1QWB<{0,0,?,?}, "8">;
def VST1q16_UPD : VST1QWB<{0,1,?,?}, "16">;
def VST1q32_UPD : VST1QWB<{1,0,?,?}, "32">;
def VST1q64_UPD : VST1QWB<{1,1,?,?}, "64">;

def VST1q8Pseudo_UPD  : VSTQWBPseudo<IIC_VST1x2u>;
def VST1q16Pseudo_UPD : VSTQWBPseudo<IIC_VST1x2u>;
def VST1q32Pseudo_UPD : VSTQWBPseudo<IIC_VST1x2u>;
def VST1q64Pseudo_UPD : VSTQWBPseudo<IIC_VST1x2u>;

// ...with 3 registers (some of these are only for the disassembler):
class VST1D3<bits<4> op7_4, string Dt>
  : NLdSt<0, 0b00, 0b0110, op7_4, (outs),
          (ins addrmode6:$Rn, DPR:$Vd, DPR:$src2, DPR:$src3),
          IIC_VST1x3, "vst1", Dt, "\\{$Vd, $src2, $src3\\}, $Rn", "", []> {
  let Rm = 0b1111;
  let Inst{4} = Rn{4};
}
class VST1D3WB<bits<4> op7_4, string Dt>
  : NLdSt<0, 0b00, 0b0110, op7_4, (outs GPR:$wb),
          (ins addrmode6:$Rn, am6offset:$Rm,
           DPR:$Vd, DPR:$src2, DPR:$src3),
          IIC_VST1x3u, "vst1", Dt, "\\{$Vd, $src2, $src3\\}, $Rn$Rm",
          "$Rn.addr = $wb", []> {
  let Inst{4} = Rn{4};
}

def VST1d8T      : VST1D3<{0,0,0,?}, "8">;
def VST1d16T     : VST1D3<{0,1,0,?}, "16">;
def VST1d32T     : VST1D3<{1,0,0,?}, "32">;
def VST1d64T     : VST1D3<{1,1,0,?}, "64">;

def VST1d8T_UPD  : VST1D3WB<{0,0,0,?}, "8">;
def VST1d16T_UPD : VST1D3WB<{0,1,0,?}, "16">;
def VST1d32T_UPD : VST1D3WB<{1,0,0,?}, "32">;
def VST1d64T_UPD : VST1D3WB<{1,1,0,?}, "64">;

def VST1d64TPseudo     : VSTQQPseudo<IIC_VST1x3>;
def VST1d64TPseudo_UPD : VSTQQWBPseudo<IIC_VST1x3u>;

// ...with 4 registers (some of these are only for the disassembler):
class VST1D4<bits<4> op7_4, string Dt>
  : NLdSt<0, 0b00, 0b0010, op7_4, (outs),
          (ins addrmode6:$Rn, DPR:$Vd, DPR:$src2, DPR:$src3, DPR:$src4),
          IIC_VST1x4, "vst1", Dt, "\\{$Vd, $src2, $src3, $src4\\}, $Rn", "",
          []> {
  let Rm = 0b1111;
  let Inst{5-4} = Rn{5-4};
}
class VST1D4WB<bits<4> op7_4, string Dt>
  : NLdSt<0, 0b00, 0b0010, op7_4, (outs GPR:$wb),
          (ins addrmode6:$Rn, am6offset:$Rm,
           DPR:$Vd, DPR:$src2, DPR:$src3, DPR:$src4), IIC_VST1x4u,
          "vst1", Dt, "\\{$Vd, $src2, $src3, $src4\\}, $Rn$Rm",
          "$Rn.addr = $wb", []> {
  let Inst{5-4} = Rn{5-4};
}

def VST1d8Q      : VST1D4<{0,0,?,?}, "8">;
def VST1d16Q     : VST1D4<{0,1,?,?}, "16">;
def VST1d32Q     : VST1D4<{1,0,?,?}, "32">;
def VST1d64Q     : VST1D4<{1,1,?,?}, "64">;

def VST1d8Q_UPD  : VST1D4WB<{0,0,?,?}, "8">;
def VST1d16Q_UPD : VST1D4WB<{0,1,?,?}, "16">;
def VST1d32Q_UPD : VST1D4WB<{1,0,?,?}, "32">;
def VST1d64Q_UPD : VST1D4WB<{1,1,?,?}, "64">;

def VST1d64QPseudo     : VSTQQPseudo<IIC_VST1x4>;
def VST1d64QPseudo_UPD : VSTQQWBPseudo<IIC_VST1x4u>;

//   VST2     : Vector Store (multiple 2-element structures)
class VST2D<bits<4> op11_8, bits<4> op7_4, string Dt>
  : NLdSt<0, 0b00, op11_8, op7_4, (outs),
          (ins addrmode6:$Rn, DPR:$Vd, DPR:$src2),
          IIC_VST2, "vst2", Dt, "\\{$Vd, $src2\\}, $Rn", "", []> {
  let Rm = 0b1111;
  let Inst{5-4} = Rn{5-4};
}
class VST2Q<bits<4> op7_4, string Dt>
  : NLdSt<0, 0b00, 0b0011, op7_4, (outs),
          (ins addrmode6:$Rn, DPR:$Vd, DPR:$src2, DPR:$src3, DPR:$src4),
          IIC_VST2x2, "vst2", Dt, "\\{$Vd, $src2, $src3, $src4\\}, $Rn",
          "", []> {
  let Rm = 0b1111;
  let Inst{5-4} = Rn{5-4};
}

def  VST2d8   : VST2D<0b1000, {0,0,?,?}, "8">;
def  VST2d16  : VST2D<0b1000, {0,1,?,?}, "16">;
def  VST2d32  : VST2D<0b1000, {1,0,?,?}, "32">;

def  VST2q8   : VST2Q<{0,0,?,?}, "8">;
def  VST2q16  : VST2Q<{0,1,?,?}, "16">;
def  VST2q32  : VST2Q<{1,0,?,?}, "32">;

def  VST2d8Pseudo  : VSTQPseudo<IIC_VST2>;
def  VST2d16Pseudo : VSTQPseudo<IIC_VST2>;
def  VST2d32Pseudo : VSTQPseudo<IIC_VST2>;

def  VST2q8Pseudo  : VSTQQPseudo<IIC_VST2x2>;
def  VST2q16Pseudo : VSTQQPseudo<IIC_VST2x2>;
def  VST2q32Pseudo : VSTQQPseudo<IIC_VST2x2>;

// ...with address register writeback:
class VST2DWB<bits<4> op11_8, bits<4> op7_4, string Dt>
  : NLdSt<0, 0b00, op11_8, op7_4, (outs GPR:$wb),
          (ins addrmode6:$Rn, am6offset:$Rm, DPR:$Vd, DPR:$src2),
          IIC_VST2u, "vst2", Dt, "\\{$Vd, $src2\\}, $Rn$Rm",
          "$Rn.addr = $wb", []> {
  let Inst{5-4} = Rn{5-4};
}
class VST2QWB<bits<4> op7_4, string Dt>
  : NLdSt<0, 0b00, 0b0011, op7_4, (outs GPR:$wb),
          (ins addrmode6:$Rn, am6offset:$Rm,
           DPR:$Vd, DPR:$src2, DPR:$src3, DPR:$src4), IIC_VST2x2u,
          "vst2", Dt, "\\{$Vd, $src2, $src3, $src4\\}, $Rn$Rm",
          "$Rn.addr = $wb", []> {
  let Inst{5-4} = Rn{5-4};
}

def VST2d8_UPD  : VST2DWB<0b1000, {0,0,?,?}, "8">;
def VST2d16_UPD : VST2DWB<0b1000, {0,1,?,?}, "16">;
def VST2d32_UPD : VST2DWB<0b1000, {1,0,?,?}, "32">;

def VST2q8_UPD  : VST2QWB<{0,0,?,?}, "8">;
def VST2q16_UPD : VST2QWB<{0,1,?,?}, "16">;
def VST2q32_UPD : VST2QWB<{1,0,?,?}, "32">;

def VST2d8Pseudo_UPD  : VSTQWBPseudo<IIC_VST2u>;
def VST2d16Pseudo_UPD : VSTQWBPseudo<IIC_VST2u>;
def VST2d32Pseudo_UPD : VSTQWBPseudo<IIC_VST2u>;

def VST2q8Pseudo_UPD  : VSTQQWBPseudo<IIC_VST2x2u>;
def VST2q16Pseudo_UPD : VSTQQWBPseudo<IIC_VST2x2u>;
def VST2q32Pseudo_UPD : VSTQQWBPseudo<IIC_VST2x2u>;

// ...with double-spaced registers (for disassembly only):
def VST2b8      : VST2D<0b1001, {0,0,?,?}, "8">;
def VST2b16     : VST2D<0b1001, {0,1,?,?}, "16">;
def VST2b32     : VST2D<0b1001, {1,0,?,?}, "32">;
def VST2b8_UPD  : VST2DWB<0b1001, {0,0,?,?}, "8">;
def VST2b16_UPD : VST2DWB<0b1001, {0,1,?,?}, "16">;
def VST2b32_UPD : VST2DWB<0b1001, {1,0,?,?}, "32">;

//   VST3     : Vector Store (multiple 3-element structures)
class VST3D<bits<4> op11_8, bits<4> op7_4, string Dt>
  : NLdSt<0, 0b00, op11_8, op7_4, (outs),
          (ins addrmode6:$Rn, DPR:$Vd, DPR:$src2, DPR:$src3), IIC_VST3,
          "vst3", Dt, "\\{$Vd, $src2, $src3\\}, $Rn", "", []> {
  let Rm = 0b1111;
  let Inst{4} = Rn{4};
}

def  VST3d8   : VST3D<0b0100, {0,0,0,?}, "8">;
def  VST3d16  : VST3D<0b0100, {0,1,0,?}, "16">;
def  VST3d32  : VST3D<0b0100, {1,0,0,?}, "32">;

def  VST3d8Pseudo  : VSTQQPseudo<IIC_VST3>;
def  VST3d16Pseudo : VSTQQPseudo<IIC_VST3>;
def  VST3d32Pseudo : VSTQQPseudo<IIC_VST3>;

// ...with address register writeback:
class VST3DWB<bits<4> op11_8, bits<4> op7_4, string Dt>
  : NLdSt<0, 0b00, op11_8, op7_4, (outs GPR:$wb),
          (ins addrmode6:$Rn, am6offset:$Rm,
           DPR:$Vd, DPR:$src2, DPR:$src3), IIC_VST3u,
          "vst3", Dt, "\\{$Vd, $src2, $src3\\}, $Rn$Rm",
          "$Rn.addr = $wb", []> {
  let Inst{4} = Rn{4};
}

def VST3d8_UPD  : VST3DWB<0b0100, {0,0,0,?}, "8">;
def VST3d16_UPD : VST3DWB<0b0100, {0,1,0,?}, "16">;
def VST3d32_UPD : VST3DWB<0b0100, {1,0,0,?}, "32">;

def VST3d8Pseudo_UPD  : VSTQQWBPseudo<IIC_VST3u>;
def VST3d16Pseudo_UPD : VSTQQWBPseudo<IIC_VST3u>;
def VST3d32Pseudo_UPD : VSTQQWBPseudo<IIC_VST3u>;

// ...with double-spaced registers (non-updating versions for disassembly only):
def VST3q8      : VST3D<0b0101, {0,0,0,?}, "8">;
def VST3q16     : VST3D<0b0101, {0,1,0,?}, "16">;
def VST3q32     : VST3D<0b0101, {1,0,0,?}, "32">;
def VST3q8_UPD  : VST3DWB<0b0101, {0,0,0,?}, "8">;
def VST3q16_UPD : VST3DWB<0b0101, {0,1,0,?}, "16">;
def VST3q32_UPD : VST3DWB<0b0101, {1,0,0,?}, "32">;

def VST3q8Pseudo_UPD  : VSTQQQQWBPseudo<IIC_VST3u>;
def VST3q16Pseudo_UPD : VSTQQQQWBPseudo<IIC_VST3u>;
def VST3q32Pseudo_UPD : VSTQQQQWBPseudo<IIC_VST3u>;

// ...alternate versions to be allocated odd register numbers:
def VST3q8oddPseudo_UPD  : VSTQQQQWBPseudo<IIC_VST3u>;
def VST3q16oddPseudo_UPD : VSTQQQQWBPseudo<IIC_VST3u>;
def VST3q32oddPseudo_UPD : VSTQQQQWBPseudo<IIC_VST3u>;

//   VST4     : Vector Store (multiple 4-element structures)
class VST4D<bits<4> op11_8, bits<4> op7_4, string Dt>
  : NLdSt<0, 0b00, op11_8, op7_4, (outs),
          (ins addrmode6:$Rn, DPR:$Vd, DPR:$src2, DPR:$src3, DPR:$src4),
          IIC_VST4, "vst4", Dt, "\\{$Vd, $src2, $src3, $src4\\}, $Rn",
          "", []> {
  let Rm = 0b1111;
  let Inst{5-4} = Rn{5-4};
}

def  VST4d8   : VST4D<0b0000, {0,0,?,?}, "8">;
def  VST4d16  : VST4D<0b0000, {0,1,?,?}, "16">;
def  VST4d32  : VST4D<0b0000, {1,0,?,?}, "32">;

def  VST4d8Pseudo  : VSTQQPseudo<IIC_VST4>;
def  VST4d16Pseudo : VSTQQPseudo<IIC_VST4>;
def  VST4d32Pseudo : VSTQQPseudo<IIC_VST4>;

// ...with address register writeback:
class VST4DWB<bits<4> op11_8, bits<4> op7_4, string Dt>
  : NLdSt<0, 0b00, op11_8, op7_4, (outs GPR:$wb),
          (ins addrmode6:$Rn, am6offset:$Rm,
           DPR:$Vd, DPR:$src2, DPR:$src3, DPR:$src4), IIC_VST4u,
           "vst4", Dt, "\\{$Vd, $src2, $src3, $src4\\}, $Rn$Rm",
          "$Rn.addr = $wb", []> {
  let Inst{5-4} = Rn{5-4};
}

def VST4d8_UPD  : VST4DWB<0b0000, {0,0,?,?}, "8">;
def VST4d16_UPD : VST4DWB<0b0000, {0,1,?,?}, "16">;
def VST4d32_UPD : VST4DWB<0b0000, {1,0,?,?}, "32">;

def VST4d8Pseudo_UPD  : VSTQQWBPseudo<IIC_VST4u>;
def VST4d16Pseudo_UPD : VSTQQWBPseudo<IIC_VST4u>;
def VST4d32Pseudo_UPD : VSTQQWBPseudo<IIC_VST4u>;

// ...with double-spaced registers (non-updating versions for disassembly only):
def VST4q8      : VST4D<0b0001, {0,0,?,?}, "8">;
def VST4q16     : VST4D<0b0001, {0,1,?,?}, "16">;
def VST4q32     : VST4D<0b0001, {1,0,?,?}, "32">;
def VST4q8_UPD  : VST4DWB<0b0001, {0,0,?,?}, "8">;
def VST4q16_UPD : VST4DWB<0b0001, {0,1,?,?}, "16">;
def VST4q32_UPD : VST4DWB<0b0001, {1,0,?,?}, "32">;

def VST4q8Pseudo_UPD  : VSTQQQQWBPseudo<IIC_VST4u>;
def VST4q16Pseudo_UPD : VSTQQQQWBPseudo<IIC_VST4u>;
def VST4q32Pseudo_UPD : VSTQQQQWBPseudo<IIC_VST4u>;

// ...alternate versions to be allocated odd register numbers:
def VST4q8oddPseudo_UPD  : VSTQQQQWBPseudo<IIC_VST4u>;
def VST4q16oddPseudo_UPD : VSTQQQQWBPseudo<IIC_VST4u>;
def VST4q32oddPseudo_UPD : VSTQQQQWBPseudo<IIC_VST4u>;

} // mayStore = 1, neverHasSideEffects = 1, hasExtraSrcRegAllocReq = 1

// Classes for VST*LN pseudo-instructions with multi-register operands.
// These are expanded to real instructions after register allocation.
class VSTQLNPseudo<InstrItinClass itin>
  : PseudoNLdSt<(outs), (ins addrmode6:$addr, QPR:$src, nohash_imm:$lane),
                itin, "">;
class VSTQLNWBPseudo<InstrItinClass itin>
  : PseudoNLdSt<(outs GPR:$wb),
                (ins addrmode6:$addr, am6offset:$offset, QPR:$src,
                 nohash_imm:$lane), itin, "$addr.addr = $wb">;
class VSTQQLNPseudo<InstrItinClass itin>
  : PseudoNLdSt<(outs), (ins addrmode6:$addr, QQPR:$src, nohash_imm:$lane),
                itin, "">;
class VSTQQLNWBPseudo<InstrItinClass itin>
  : PseudoNLdSt<(outs GPR:$wb),
                (ins addrmode6:$addr, am6offset:$offset, QQPR:$src,
                 nohash_imm:$lane), itin, "$addr.addr = $wb">;
class VSTQQQQLNPseudo<InstrItinClass itin>
  : PseudoNLdSt<(outs), (ins addrmode6:$addr, QQQQPR:$src, nohash_imm:$lane),
                itin, "">;
class VSTQQQQLNWBPseudo<InstrItinClass itin>
  : PseudoNLdSt<(outs GPR:$wb),
                (ins addrmode6:$addr, am6offset:$offset, QQQQPR:$src,
                 nohash_imm:$lane), itin, "$addr.addr = $wb">;

//   VST1LN   : Vector Store (single element from one lane)
class VST1LN<bits<4> op11_8, bits<4> op7_4, string Dt, ValueType Ty,
             PatFrag StoreOp, SDNode ExtractOp>
  : NLdStLn<1, 0b00, op11_8, op7_4, (outs),
          (ins addrmode6:$Rn, DPR:$Vd, nohash_imm:$lane),
          IIC_VST1ln, "vst1", Dt, "\\{$Vd[$lane]\\}, $Rn", "",
          [(StoreOp (ExtractOp (Ty DPR:$Vd), imm:$lane), addrmode6:$Rn)]> {
  let Rm = 0b1111;
}
class VST1QLNPseudo<ValueType Ty, PatFrag StoreOp, SDNode ExtractOp>
  : VSTQLNPseudo<IIC_VST1ln> {
  let Pattern = [(StoreOp (ExtractOp (Ty QPR:$src), imm:$lane),
                          addrmode6:$addr)];
}

def VST1LNd8  : VST1LN<0b0000, {?,?,?,0}, "8", v8i8, truncstorei8,
                       NEONvgetlaneu> {
  let Inst{7-5} = lane{2-0};
}
def VST1LNd16 : VST1LN<0b0100, {?,?,0,?}, "16", v4i16, truncstorei16,
                       NEONvgetlaneu> {
  let Inst{7-6} = lane{1-0};
  let Inst{4}   = Rn{5};
}
def VST1LNd32 : VST1LN<0b1000, {?,0,?,?}, "32", v2i32, store, extractelt> {
  let Inst{7}   = lane{0};
  let Inst{5-4} = Rn{5-4};
}

def VST1LNq8Pseudo  : VST1QLNPseudo<v16i8, truncstorei8, NEONvgetlaneu>;
def VST1LNq16Pseudo : VST1QLNPseudo<v8i16, truncstorei16, NEONvgetlaneu>;
def VST1LNq32Pseudo : VST1QLNPseudo<v4i32, store, extractelt>;

let mayStore = 1, neverHasSideEffects = 1, hasExtraSrcRegAllocReq = 1 in {

// ...with address register writeback:
class VST1LNWB<bits<4> op11_8, bits<4> op7_4, string Dt>
  : NLdStLn<1, 0b00, op11_8, op7_4, (outs GPR:$wb),
          (ins addrmode6:$Rn, am6offset:$Rm,
           DPR:$Vd, nohash_imm:$lane), IIC_VST1lnu, "vst1", Dt,
          "\\{$Vd[$lane]\\}, $Rn$Rm",
          "$Rn.addr = $wb", []>;

def VST1LNd8_UPD  : VST1LNWB<0b0000, {?,?,?,0}, "8"> {
  let Inst{7-5} = lane{2-0};
}
def VST1LNd16_UPD : VST1LNWB<0b0100, {?,?,0,?}, "16"> {
  let Inst{7-6} = lane{1-0};
  let Inst{4}   = Rn{5};
}
def VST1LNd32_UPD : VST1LNWB<0b1000, {?,0,?,?}, "32"> {
  let Inst{7}   = lane{0};
  let Inst{5-4} = Rn{5-4};
}

def VST1LNq8Pseudo_UPD  : VSTQLNWBPseudo<IIC_VST1lnu>;
def VST1LNq16Pseudo_UPD : VSTQLNWBPseudo<IIC_VST1lnu>;
def VST1LNq32Pseudo_UPD : VSTQLNWBPseudo<IIC_VST1lnu>;

//   VST2LN   : Vector Store (single 2-element structure from one lane)
class VST2LN<bits<4> op11_8, bits<4> op7_4, string Dt>
  : NLdStLn<1, 0b00, op11_8, op7_4, (outs),
          (ins addrmode6:$Rn, DPR:$Vd, DPR:$src2, nohash_imm:$lane),
          IIC_VST2ln, "vst2", Dt, "\\{$Vd[$lane], $src2[$lane]\\}, $Rn",
          "", []> {
  let Rm = 0b1111;
  let Inst{4}   = Rn{4};
}

def VST2LNd8  : VST2LN<0b0001, {?,?,?,?}, "8"> {
  let Inst{7-5} = lane{2-0};
}
def VST2LNd16 : VST2LN<0b0101, {?,?,0,?}, "16"> {
  let Inst{7-6} = lane{1-0};
}
def VST2LNd32 : VST2LN<0b1001, {?,0,0,?}, "32"> {
  let Inst{7}   = lane{0};
}

def VST2LNd8Pseudo  : VSTQLNPseudo<IIC_VST2ln>;
def VST2LNd16Pseudo : VSTQLNPseudo<IIC_VST2ln>;
def VST2LNd32Pseudo : VSTQLNPseudo<IIC_VST2ln>;

// ...with double-spaced registers:
def VST2LNq16 : VST2LN<0b0101, {?,?,1,?}, "16"> {
  let Inst{7-6} = lane{1-0};
  let Inst{4}   = Rn{4};
}
def VST2LNq32 : VST2LN<0b1001, {?,1,0,?}, "32"> {
  let Inst{7}   = lane{0};
  let Inst{4}   = Rn{4};
}

def VST2LNq16Pseudo : VSTQQLNPseudo<IIC_VST2ln>;
def VST2LNq32Pseudo : VSTQQLNPseudo<IIC_VST2ln>;

// ...with address register writeback:
class VST2LNWB<bits<4> op11_8, bits<4> op7_4, string Dt>
  : NLdStLn<1, 0b00, op11_8, op7_4, (outs GPR:$wb),
          (ins addrmode6:$addr, am6offset:$offset,
           DPR:$src1, DPR:$src2, nohash_imm:$lane), IIC_VST2lnu, "vst2", Dt,
          "\\{$src1[$lane], $src2[$lane]\\}, $addr$offset",
          "$addr.addr = $wb", []> {
  let Inst{4}   = Rn{4};
}

def VST2LNd8_UPD  : VST2LNWB<0b0001, {?,?,?,?}, "8"> {
  let Inst{7-5} = lane{2-0};
}
def VST2LNd16_UPD : VST2LNWB<0b0101, {?,?,0,?}, "16"> {
  let Inst{7-6} = lane{1-0};
}
def VST2LNd32_UPD : VST2LNWB<0b1001, {?,0,0,?}, "32"> {
  let Inst{7}   = lane{0};
}

def VST2LNd8Pseudo_UPD  : VSTQLNWBPseudo<IIC_VST2lnu>;
def VST2LNd16Pseudo_UPD : VSTQLNWBPseudo<IIC_VST2lnu>;
def VST2LNd32Pseudo_UPD : VSTQLNWBPseudo<IIC_VST2lnu>;

def VST2LNq16_UPD : VST2LNWB<0b0101, {?,?,1,?}, "16"> {
  let Inst{7-6} = lane{1-0};
}
def VST2LNq32_UPD : VST2LNWB<0b1001, {?,1,0,?}, "32"> {
  let Inst{7}   = lane{0};
}

def VST2LNq16Pseudo_UPD : VSTQQLNWBPseudo<IIC_VST2lnu>;
def VST2LNq32Pseudo_UPD : VSTQQLNWBPseudo<IIC_VST2lnu>;

//   VST3LN   : Vector Store (single 3-element structure from one lane)
class VST3LN<bits<4> op11_8, bits<4> op7_4, string Dt>
  : NLdStLn<1, 0b00, op11_8, op7_4, (outs),
          (ins addrmode6:$Rn, DPR:$Vd, DPR:$src2, DPR:$src3,
           nohash_imm:$lane), IIC_VST3ln, "vst3", Dt,
          "\\{$Vd[$lane], $src2[$lane], $src3[$lane]\\}, $Rn", "", []> {
  let Rm = 0b1111;
}

def VST3LNd8  : VST3LN<0b0010, {?,?,?,0}, "8"> {
  let Inst{7-5} = lane{2-0};
}
def VST3LNd16 : VST3LN<0b0110, {?,?,0,0}, "16"> {
  let Inst{7-6} = lane{1-0};
}
def VST3LNd32 : VST3LN<0b1010, {?,0,0,0}, "32"> {
  let Inst{7}   = lane{0};
}

def VST3LNd8Pseudo  : VSTQQLNPseudo<IIC_VST3ln>;
def VST3LNd16Pseudo : VSTQQLNPseudo<IIC_VST3ln>;
def VST3LNd32Pseudo : VSTQQLNPseudo<IIC_VST3ln>;

// ...with double-spaced registers:
def VST3LNq16 : VST3LN<0b0110, {?,?,1,0}, "16"> {
  let Inst{7-6} = lane{1-0};
}
def VST3LNq32 : VST3LN<0b1010, {?,1,0,0}, "32"> {
  let Inst{7}   = lane{0};
}

def VST3LNq16Pseudo : VSTQQQQLNPseudo<IIC_VST3ln>;
def VST3LNq32Pseudo : VSTQQQQLNPseudo<IIC_VST3ln>;

// ...with address register writeback:
class VST3LNWB<bits<4> op11_8, bits<4> op7_4, string Dt>
  : NLdStLn<1, 0b00, op11_8, op7_4, (outs GPR:$wb),
          (ins addrmode6:$Rn, am6offset:$Rm,
           DPR:$Vd, DPR:$src2, DPR:$src3, nohash_imm:$lane),
          IIC_VST3lnu, "vst3", Dt,
          "\\{$Vd[$lane], $src2[$lane], $src3[$lane]\\}, $Rn$Rm",
          "$Rn.addr = $wb", []>;

def VST3LNd8_UPD  : VST3LNWB<0b0010, {?,?,?,0}, "8"> {
  let Inst{7-5} = lane{2-0};
}
def VST3LNd16_UPD : VST3LNWB<0b0110, {?,?,0,0}, "16"> {
  let Inst{7-6} = lane{1-0};
}
def VST3LNd32_UPD : VST3LNWB<0b1010, {?,0,0,0}, "32"> {
  let Inst{7}   = lane{0};
}

def VST3LNd8Pseudo_UPD  : VSTQQLNWBPseudo<IIC_VST3lnu>;
def VST3LNd16Pseudo_UPD : VSTQQLNWBPseudo<IIC_VST3lnu>;
def VST3LNd32Pseudo_UPD : VSTQQLNWBPseudo<IIC_VST3lnu>;

def VST3LNq16_UPD : VST3LNWB<0b0110, {?,?,1,0}, "16"> {
  let Inst{7-6} = lane{1-0};
}
def VST3LNq32_UPD : VST3LNWB<0b1010, {?,1,0,0}, "32"> {
  let Inst{7}   = lane{0};
}

def VST3LNq16Pseudo_UPD : VSTQQQQLNWBPseudo<IIC_VST3lnu>;
def VST3LNq32Pseudo_UPD : VSTQQQQLNWBPseudo<IIC_VST3lnu>;

//   VST4LN   : Vector Store (single 4-element structure from one lane)
class VST4LN<bits<4> op11_8, bits<4> op7_4, string Dt>
  : NLdStLn<1, 0b00, op11_8, op7_4, (outs),
          (ins addrmode6:$Rn, DPR:$Vd, DPR:$src2, DPR:$src3, DPR:$src4,
           nohash_imm:$lane), IIC_VST4ln, "vst4", Dt,
          "\\{$Vd[$lane], $src2[$lane], $src3[$lane], $src4[$lane]\\}, $Rn",
          "", []> {
  let Rm = 0b1111;
  let Inst{4} = Rn{4};
}

def VST4LNd8  : VST4LN<0b0011, {?,?,?,?}, "8"> {
  let Inst{7-5} = lane{2-0};
}
def VST4LNd16 : VST4LN<0b0111, {?,?,0,?}, "16"> {
  let Inst{7-6} = lane{1-0};
}
def VST4LNd32 : VST4LN<0b1011, {?,0,?,?}, "32"> {
  let Inst{7}   = lane{0};
  let Inst{5} = Rn{5};
}

def VST4LNd8Pseudo  : VSTQQLNPseudo<IIC_VST4ln>;
def VST4LNd16Pseudo : VSTQQLNPseudo<IIC_VST4ln>;
def VST4LNd32Pseudo : VSTQQLNPseudo<IIC_VST4ln>;

// ...with double-spaced registers:
def VST4LNq16 : VST4LN<0b0111, {?,?,1,?}, "16"> {
  let Inst{7-6} = lane{1-0};
}
def VST4LNq32 : VST4LN<0b1011, {?,1,?,?}, "32"> {
  let Inst{7}   = lane{0};
  let Inst{5} = Rn{5};
}

def VST4LNq16Pseudo : VSTQQQQLNPseudo<IIC_VST4ln>;
def VST4LNq32Pseudo : VSTQQQQLNPseudo<IIC_VST4ln>;

// ...with address register writeback:
class VST4LNWB<bits<4> op11_8, bits<4> op7_4, string Dt>
  : NLdStLn<1, 0b00, op11_8, op7_4, (outs GPR:$wb),
          (ins addrmode6:$Rn, am6offset:$Rm,
           DPR:$Vd, DPR:$src2, DPR:$src3, DPR:$src4, nohash_imm:$lane),
          IIC_VST4lnu, "vst4", Dt,
  "\\{$Vd[$lane], $src2[$lane], $src3[$lane], $src4[$lane]\\}, $Rn$Rm",
          "$Rn.addr = $wb", []> {
  let Inst{4} = Rn{4};
}

def VST4LNd8_UPD  : VST4LNWB<0b0011, {?,?,?,?}, "8"> {
  let Inst{7-5} = lane{2-0};
}
def VST4LNd16_UPD : VST4LNWB<0b0111, {?,?,0,?}, "16"> {
  let Inst{7-6} = lane{1-0};
}
def VST4LNd32_UPD : VST4LNWB<0b1011, {?,0,?,?}, "32"> {
  let Inst{7}   = lane{0};
  let Inst{5} = Rn{5};
}

def VST4LNd8Pseudo_UPD  : VSTQQLNWBPseudo<IIC_VST4lnu>;
def VST4LNd16Pseudo_UPD : VSTQQLNWBPseudo<IIC_VST4lnu>;
def VST4LNd32Pseudo_UPD : VSTQQLNWBPseudo<IIC_VST4lnu>;

def VST4LNq16_UPD : VST4LNWB<0b0111, {?,?,1,?}, "16"> {
  let Inst{7-6} = lane{1-0};
}
def VST4LNq32_UPD : VST4LNWB<0b1011, {?,1,?,?}, "32"> {
  let Inst{7}   = lane{0};
  let Inst{5} = Rn{5};
}

def VST4LNq16Pseudo_UPD : VSTQQQQLNWBPseudo<IIC_VST4lnu>;
def VST4LNq32Pseudo_UPD : VSTQQQQLNWBPseudo<IIC_VST4lnu>;

} // mayStore = 1, neverHasSideEffects = 1, hasExtraSrcRegAllocReq = 1


//===----------------------------------------------------------------------===//
// NEON pattern fragments
//===----------------------------------------------------------------------===//

// Extract D sub-registers of Q registers.
def DSubReg_i8_reg  : SDNodeXForm<imm, [{
  assert(ARM::dsub_7 == ARM::dsub_0+7 && "Unexpected subreg numbering");
  return CurDAG->getTargetConstant(ARM::dsub_0 + N->getZExtValue()/8, MVT::i32);
}]>;
def DSubReg_i16_reg : SDNodeXForm<imm, [{
  assert(ARM::dsub_7 == ARM::dsub_0+7 && "Unexpected subreg numbering");
  return CurDAG->getTargetConstant(ARM::dsub_0 + N->getZExtValue()/4, MVT::i32);
}]>;
def DSubReg_i32_reg : SDNodeXForm<imm, [{
  assert(ARM::dsub_7 == ARM::dsub_0+7 && "Unexpected subreg numbering");
  return CurDAG->getTargetConstant(ARM::dsub_0 + N->getZExtValue()/2, MVT::i32);
}]>;
def DSubReg_f64_reg : SDNodeXForm<imm, [{
  assert(ARM::dsub_7 == ARM::dsub_0+7 && "Unexpected subreg numbering");
  return CurDAG->getTargetConstant(ARM::dsub_0 + N->getZExtValue(), MVT::i32);
}]>;

// Extract S sub-registers of Q/D registers.
def SSubReg_f32_reg : SDNodeXForm<imm, [{
  assert(ARM::ssub_3 == ARM::ssub_0+3 && "Unexpected subreg numbering");
  return CurDAG->getTargetConstant(ARM::ssub_0 + N->getZExtValue(), MVT::i32);
}]>;

// Translate lane numbers from Q registers to D subregs.
def SubReg_i8_lane  : SDNodeXForm<imm, [{
  return CurDAG->getTargetConstant(N->getZExtValue() & 7, MVT::i32);
}]>;
def SubReg_i16_lane : SDNodeXForm<imm, [{
  return CurDAG->getTargetConstant(N->getZExtValue() & 3, MVT::i32);
}]>;
def SubReg_i32_lane : SDNodeXForm<imm, [{
  return CurDAG->getTargetConstant(N->getZExtValue() & 1, MVT::i32);
}]>;

//===----------------------------------------------------------------------===//
// Instruction Classes
//===----------------------------------------------------------------------===//

// Basic 2-register operations: single-, double- and quad-register.
class N2VS<bits<2> op24_23, bits<2> op21_20, bits<2> op19_18,
           bits<2> op17_16, bits<5> op11_7, bit op4, string OpcodeStr,
           string Dt, ValueType ResTy, ValueType OpTy, SDNode OpNode>
  : N2V<op24_23, op21_20, op19_18, op17_16, op11_7, 0, op4,
        (outs DPR_VFP2:$dst), (ins DPR_VFP2:$src),
        IIC_VUNAD, OpcodeStr, Dt, "$dst, $src", "", []>;
class N2VD<bits<2> op24_23, bits<2> op21_20, bits<2> op19_18,
           bits<2> op17_16, bits<5> op11_7, bit op4, string OpcodeStr,
           string Dt, ValueType ResTy, ValueType OpTy, SDNode OpNode>
  : N2V<op24_23, op21_20, op19_18, op17_16, op11_7, 0, op4, (outs DPR:$dst),
        (ins DPR:$src), IIC_VUNAD, OpcodeStr, Dt,"$dst, $src", "",
        [(set DPR:$dst, (ResTy (OpNode (OpTy DPR:$src))))]>;
class N2VQ<bits<2> op24_23, bits<2> op21_20, bits<2> op19_18,
           bits<2> op17_16, bits<5> op11_7, bit op4, string OpcodeStr,
           string Dt, ValueType ResTy, ValueType OpTy, SDNode OpNode>
  : N2V<op24_23, op21_20, op19_18, op17_16, op11_7, 1, op4, (outs QPR:$dst),
        (ins QPR:$src), IIC_VUNAQ, OpcodeStr, Dt,"$dst, $src", "",
        [(set QPR:$dst, (ResTy (OpNode (OpTy QPR:$src))))]>;

// Basic 2-register intrinsics, both double- and quad-register.
class N2VDInt<bits<2> op24_23, bits<2> op21_20, bits<2> op19_18,
              bits<2> op17_16, bits<5> op11_7, bit op4,
              InstrItinClass itin, string OpcodeStr, string Dt,
              ValueType ResTy, ValueType OpTy, Intrinsic IntOp>
  : N2V<op24_23, op21_20, op19_18, op17_16, op11_7, 0, op4, (outs DPR:$dst),
        (ins DPR:$src), itin, OpcodeStr, Dt, "$dst, $src", "",
        [(set DPR:$dst, (ResTy (IntOp (OpTy DPR:$src))))]>;
class N2VQInt<bits<2> op24_23, bits<2> op21_20, bits<2> op19_18,
              bits<2> op17_16, bits<5> op11_7, bit op4,
              InstrItinClass itin, string OpcodeStr, string Dt,
              ValueType ResTy, ValueType OpTy, Intrinsic IntOp>
  : N2V<op24_23, op21_20, op19_18, op17_16, op11_7, 1, op4, (outs QPR:$dst),
        (ins QPR:$src), itin, OpcodeStr, Dt, "$dst, $src", "",
        [(set QPR:$dst, (ResTy (IntOp (OpTy QPR:$src))))]>;

// Narrow 2-register operations.
class N2VN<bits<2> op24_23, bits<2> op21_20, bits<2> op19_18,
           bits<2> op17_16, bits<5> op11_7, bit op6, bit op4,
           InstrItinClass itin, string OpcodeStr, string Dt,
           ValueType TyD, ValueType TyQ, SDNode OpNode>
  : N2V<op24_23, op21_20, op19_18, op17_16, op11_7, op6, op4, (outs DPR:$dst),
        (ins QPR:$src), itin, OpcodeStr, Dt, "$dst, $src", "",
        [(set DPR:$dst, (TyD (OpNode (TyQ QPR:$src))))]>;

// Narrow 2-register intrinsics.
class N2VNInt<bits<2> op24_23, bits<2> op21_20, bits<2> op19_18,
              bits<2> op17_16, bits<5> op11_7, bit op6, bit op4,
              InstrItinClass itin, string OpcodeStr, string Dt,
              ValueType TyD, ValueType TyQ, Intrinsic IntOp>
  : N2V<op24_23, op21_20, op19_18, op17_16, op11_7, op6, op4, (outs DPR:$dst),
        (ins QPR:$src), itin, OpcodeStr, Dt, "$dst, $src", "",
        [(set DPR:$dst, (TyD (IntOp (TyQ QPR:$src))))]>;

// Long 2-register operations (currently only used for VMOVL).
class N2VL<bits<2> op24_23, bits<2> op21_20, bits<2> op19_18,
           bits<2> op17_16, bits<5> op11_7, bit op6, bit op4,
           InstrItinClass itin, string OpcodeStr, string Dt,
           ValueType TyQ, ValueType TyD, SDNode OpNode>
  : N2V<op24_23, op21_20, op19_18, op17_16, op11_7, op6, op4, (outs QPR:$dst),
        (ins DPR:$src), itin, OpcodeStr, Dt, "$dst, $src", "",
        [(set QPR:$dst, (TyQ (OpNode (TyD DPR:$src))))]>;

// 2-register shuffles (VTRN/VZIP/VUZP), both double- and quad-register.
class N2VDShuffle<bits<2> op19_18, bits<5> op11_7, string OpcodeStr, string Dt>
  : N2V<0b11, 0b11, op19_18, 0b10, op11_7, 0, 0, (outs DPR:$dst1, DPR:$dst2),
        (ins DPR:$src1, DPR:$src2), IIC_VPERMD, 
        OpcodeStr, Dt, "$dst1, $dst2",
        "$src1 = $dst1, $src2 = $dst2", []>;
class N2VQShuffle<bits<2> op19_18, bits<5> op11_7,
                  InstrItinClass itin, string OpcodeStr, string Dt>
  : N2V<0b11, 0b11, op19_18, 0b10, op11_7, 1, 0, (outs QPR:$dst1, QPR:$dst2),
        (ins QPR:$src1, QPR:$src2), itin, OpcodeStr, Dt, "$dst1, $dst2",
        "$src1 = $dst1, $src2 = $dst2", []>;

// Basic 3-register operations: single-, double- and quad-register.
class N3VS<bit op24, bit op23, bits<2> op21_20, bits<4> op11_8, bit op4,
           string OpcodeStr, string Dt, ValueType ResTy, ValueType OpTy,
           SDNode OpNode, bit Commutable>
  : N3V<op24, op23, op21_20, op11_8, 0, op4,
        (outs DPR_VFP2:$dst), (ins DPR_VFP2:$src1, DPR_VFP2:$src2), N3RegFrm,
        IIC_VBIND, OpcodeStr, Dt, "$dst, $src1, $src2", "", []> {
  let isCommutable = Commutable;
}

class N3VD<bit op24, bit op23, bits<2> op21_20, bits<4> op11_8, bit op4,
           InstrItinClass itin, string OpcodeStr, string Dt,
           ValueType ResTy, ValueType OpTy, SDNode OpNode, bit Commutable>
  : N3V<op24, op23, op21_20, op11_8, 0, op4,
        (outs DPR:$Vd), (ins DPR:$Vn, DPR:$Vm), N3RegFrm, itin,
        OpcodeStr, Dt, "$Vd, $Vn, $Vm", "",
        [(set DPR:$Vd, (ResTy (OpNode (OpTy DPR:$Vn), (OpTy DPR:$Vm))))]> {
  let isCommutable = Commutable;
}
// Same as N3VD but no data type.
class N3VDX<bit op24, bit op23, bits<2> op21_20, bits<4> op11_8, bit op4,
           InstrItinClass itin, string OpcodeStr,
           ValueType ResTy, ValueType OpTy,
           SDNode OpNode, bit Commutable>
  : N3VX<op24, op23, op21_20, op11_8, 0, op4,
         (outs DPR:$dst), (ins DPR:$src1, DPR:$src2), N3RegFrm, itin, 
         OpcodeStr, "$dst, $src1, $src2", "",
         [(set DPR:$dst, (ResTy (OpNode (OpTy DPR:$src1), (OpTy DPR:$src2))))]>{
  let isCommutable = Commutable;
}

class N3VDSL<bits<2> op21_20, bits<4> op11_8, 
             InstrItinClass itin, string OpcodeStr, string Dt,
             ValueType Ty, SDNode ShOp>
  : N3V<0, 1, op21_20, op11_8, 1, 0,
        (outs DPR:$dst), (ins DPR:$src1, DPR_VFP2:$src2, nohash_imm:$lane),
        NVMulSLFrm, itin, OpcodeStr, Dt, "$dst, $src1, $src2[$lane]", "",
        [(set (Ty DPR:$dst),
              (Ty (ShOp (Ty DPR:$src1),
                        (Ty (NEONvduplane (Ty DPR_VFP2:$src2),imm:$lane)))))]> {
  let isCommutable = 0;
}
class N3VDSL16<bits<2> op21_20, bits<4> op11_8, 
               string OpcodeStr, string Dt, ValueType Ty, SDNode ShOp>
  : N3V<0, 1, op21_20, op11_8, 1, 0,
        (outs DPR:$dst), (ins DPR:$src1, DPR_8:$src2, nohash_imm:$lane),
        NVMulSLFrm, IIC_VMULi16D, OpcodeStr, Dt,"$dst, $src1, $src2[$lane]","",
        [(set (Ty DPR:$dst),
              (Ty (ShOp (Ty DPR:$src1),
                        (Ty (NEONvduplane (Ty DPR_8:$src2), imm:$lane)))))]> {
  let isCommutable = 0;
}

class N3VQ<bit op24, bit op23, bits<2> op21_20, bits<4> op11_8, bit op4,
           InstrItinClass itin, string OpcodeStr, string Dt,
           ValueType ResTy, ValueType OpTy, SDNode OpNode, bit Commutable>
  : N3V<op24, op23, op21_20, op11_8, 1, op4,
        (outs QPR:$Qd), (ins QPR:$Qn, QPR:$Qm), N3RegFrm, itin, 
        OpcodeStr, Dt, "$Qd, $Qn, $Qm", "",
        [(set QPR:$Qd, (ResTy (OpNode (OpTy QPR:$Qn), (OpTy QPR:$Qm))))]> {
  let isCommutable = Commutable;
}
class N3VQX<bit op24, bit op23, bits<2> op21_20, bits<4> op11_8, bit op4,
           InstrItinClass itin, string OpcodeStr,
           ValueType ResTy, ValueType OpTy, SDNode OpNode, bit Commutable>
  : N3VX<op24, op23, op21_20, op11_8, 1, op4,
         (outs QPR:$dst), (ins QPR:$src1, QPR:$src2), N3RegFrm, itin, 
         OpcodeStr, "$dst, $src1, $src2", "",
         [(set QPR:$dst, (ResTy (OpNode (OpTy QPR:$src1), (OpTy QPR:$src2))))]>{
  let isCommutable = Commutable;
}
class N3VQSL<bits<2> op21_20, bits<4> op11_8, 
             InstrItinClass itin, string OpcodeStr, string Dt,
             ValueType ResTy, ValueType OpTy, SDNode ShOp>
  : N3V<1, 1, op21_20, op11_8, 1, 0,
        (outs QPR:$dst), (ins QPR:$src1, DPR_VFP2:$src2, nohash_imm:$lane),
        NVMulSLFrm, itin, OpcodeStr, Dt, "$dst, $src1, $src2[$lane]", "",
        [(set (ResTy QPR:$dst),
              (ResTy (ShOp (ResTy QPR:$src1),
                           (ResTy (NEONvduplane (OpTy DPR_VFP2:$src2),
                                                imm:$lane)))))]> {
  let isCommutable = 0;
}
class N3VQSL16<bits<2> op21_20, bits<4> op11_8, string OpcodeStr, string Dt,
               ValueType ResTy, ValueType OpTy, SDNode ShOp>
  : N3V<1, 1, op21_20, op11_8, 1, 0,
        (outs QPR:$dst), (ins QPR:$src1, DPR_8:$src2, nohash_imm:$lane),
        NVMulSLFrm, IIC_VMULi16Q, OpcodeStr, Dt,"$dst, $src1, $src2[$lane]","",
        [(set (ResTy QPR:$dst),
              (ResTy (ShOp (ResTy QPR:$src1),
                           (ResTy (NEONvduplane (OpTy DPR_8:$src2),
                                                imm:$lane)))))]> {
  let isCommutable = 0;
}

// Basic 3-register intrinsics, both double- and quad-register.
class N3VDInt<bit op24, bit op23, bits<2> op21_20, bits<4> op11_8, bit op4,
              Format f, InstrItinClass itin, string OpcodeStr, string Dt,
              ValueType ResTy, ValueType OpTy, Intrinsic IntOp, bit Commutable>
  : N3V<op24, op23, op21_20, op11_8, 0, op4,
        (outs DPR:$Vd), (ins DPR:$Vn, DPR:$Vm), f, itin,
        OpcodeStr, Dt, "$Vd, $Vn, $Vm", "",
        [(set DPR:$Vd, (ResTy (IntOp (OpTy DPR:$Vn), (OpTy DPR:$Vm))))]> {
  let isCommutable = Commutable;
}
class N3VDIntSL<bits<2> op21_20, bits<4> op11_8, InstrItinClass itin, 
                string OpcodeStr, string Dt, ValueType Ty, Intrinsic IntOp>
  : N3V<0, 1, op21_20, op11_8, 1, 0,
        (outs DPR:$dst), (ins DPR:$src1, DPR_VFP2:$src2, nohash_imm:$lane),
        NVMulSLFrm, itin, OpcodeStr, Dt, "$dst, $src1, $src2[$lane]", "",
        [(set (Ty DPR:$dst),
              (Ty (IntOp (Ty DPR:$src1),
                         (Ty (NEONvduplane (Ty DPR_VFP2:$src2),
                                           imm:$lane)))))]> {
  let isCommutable = 0;
}
class N3VDIntSL16<bits<2> op21_20, bits<4> op11_8, InstrItinClass itin,
                  string OpcodeStr, string Dt, ValueType Ty, Intrinsic IntOp>
  : N3V<0, 1, op21_20, op11_8, 1, 0,
        (outs DPR:$dst), (ins DPR:$src1, DPR_8:$src2, nohash_imm:$lane),
        NVMulSLFrm, itin, OpcodeStr, Dt, "$dst, $src1, $src2[$lane]", "",
        [(set (Ty DPR:$dst),
              (Ty (IntOp (Ty DPR:$src1),
                         (Ty (NEONvduplane (Ty DPR_8:$src2), imm:$lane)))))]> {
  let isCommutable = 0;
}
class N3VDIntSh<bit op24, bit op23, bits<2> op21_20, bits<4> op11_8, bit op4,
              Format f, InstrItinClass itin, string OpcodeStr, string Dt,
              ValueType ResTy, ValueType OpTy, Intrinsic IntOp>
  : N3V<op24, op23, op21_20, op11_8, 0, op4,
        (outs DPR:$Vd), (ins DPR:$Vm, DPR:$Vn), f, itin,
        OpcodeStr, Dt, "$Vd, $Vm, $Vn", "",
        [(set DPR:$Vd, (ResTy (IntOp (OpTy DPR:$Vm), (OpTy DPR:$Vn))))]> {
  let isCommutable = 0;
}

class N3VQInt<bit op24, bit op23, bits<2> op21_20, bits<4> op11_8, bit op4,
              Format f, InstrItinClass itin, string OpcodeStr, string Dt,
              ValueType ResTy, ValueType OpTy, Intrinsic IntOp, bit Commutable>
  : N3V<op24, op23, op21_20, op11_8, 1, op4,
        (outs QPR:$Vd), (ins QPR:$Vn, QPR:$Vm), f, itin,
        OpcodeStr, Dt, "$Vd, $Vn, $Vm", "",
        [(set QPR:$Vd, (ResTy (IntOp (OpTy QPR:$Vn), (OpTy QPR:$Vm))))]> {
  let isCommutable = Commutable;
}
class N3VQIntSL<bits<2> op21_20, bits<4> op11_8, InstrItinClass itin, 
                string OpcodeStr, string Dt,
                ValueType ResTy, ValueType OpTy, Intrinsic IntOp>
  : N3V<1, 1, op21_20, op11_8, 1, 0,
        (outs QPR:$dst), (ins QPR:$src1, DPR_VFP2:$src2, nohash_imm:$lane),
        NVMulSLFrm, itin, OpcodeStr, Dt, "$dst, $src1, $src2[$lane]", "",
        [(set (ResTy QPR:$dst),
              (ResTy (IntOp (ResTy QPR:$src1),
                            (ResTy (NEONvduplane (OpTy DPR_VFP2:$src2),
                                                 imm:$lane)))))]> {
  let isCommutable = 0;
}
class N3VQIntSL16<bits<2> op21_20, bits<4> op11_8, InstrItinClass itin,
                  string OpcodeStr, string Dt,
                  ValueType ResTy, ValueType OpTy, Intrinsic IntOp>
  : N3V<1, 1, op21_20, op11_8, 1, 0,
        (outs QPR:$dst), (ins QPR:$src1, DPR_8:$src2, nohash_imm:$lane),
        NVMulSLFrm, itin, OpcodeStr, Dt, "$dst, $src1, $src2[$lane]", "",
        [(set (ResTy QPR:$dst),
              (ResTy (IntOp (ResTy QPR:$src1),
                            (ResTy (NEONvduplane (OpTy DPR_8:$src2),
                                                 imm:$lane)))))]> {
  let isCommutable = 0;
}
class N3VQIntSh<bit op24, bit op23, bits<2> op21_20, bits<4> op11_8, bit op4,
              Format f, InstrItinClass itin, string OpcodeStr, string Dt,
              ValueType ResTy, ValueType OpTy, Intrinsic IntOp>
  : N3V<op24, op23, op21_20, op11_8, 1, op4,
        (outs QPR:$Vd), (ins QPR:$Vm, QPR:$Vn), f, itin,
        OpcodeStr, Dt, "$Vd, $Vm, $Vn", "",
        [(set QPR:$Vd, (ResTy (IntOp (OpTy QPR:$Vm), (OpTy QPR:$Vn))))]> {
  let isCommutable = 0;
}

// Multiply-Add/Sub operations: single-, double- and quad-register.
class N3VSMulOp<bit op24, bit op23, bits<2> op21_20, bits<4> op11_8, bit op4,
                InstrItinClass itin, string OpcodeStr, string Dt,
                ValueType Ty, SDNode MulOp, SDNode OpNode>
  : N3V<op24, op23, op21_20, op11_8, 0, op4,
        (outs DPR_VFP2:$dst),
        (ins DPR_VFP2:$src1, DPR_VFP2:$src2, DPR_VFP2:$src3), N3RegFrm, itin,
        OpcodeStr, Dt, "$dst, $src2, $src3", "$src1 = $dst", []>;

class N3VDMulOp<bit op24, bit op23, bits<2> op21_20, bits<4> op11_8, bit op4,
                InstrItinClass itin, string OpcodeStr, string Dt,
                ValueType Ty, SDNode MulOp, SDNode OpNode>
  : N3V<op24, op23, op21_20, op11_8, 0, op4,
        (outs DPR:$Vd), (ins DPR:$src1, DPR:$Vn, DPR:$Vm), N3RegFrm, itin,
        OpcodeStr, Dt, "$Vd, $Vn, $Vm", "$src1 = $Vd",
        [(set DPR:$Vd, (Ty (OpNode DPR:$src1,
                             (Ty (MulOp DPR:$Vn, DPR:$Vm)))))]>;

class N3VDMulOpSL<bits<2> op21_20, bits<4> op11_8, InstrItinClass itin,
                  string OpcodeStr, string Dt,
                  ValueType Ty, SDNode MulOp, SDNode ShOp>
  : N3V<0, 1, op21_20, op11_8, 1, 0,
        (outs DPR:$dst),
        (ins DPR:$src1, DPR:$src2, DPR_VFP2:$src3, nohash_imm:$lane),
        NVMulSLFrm, itin,
        OpcodeStr, Dt, "$dst, $src2, $src3[$lane]", "$src1 = $dst",
        [(set (Ty DPR:$dst),
              (Ty (ShOp (Ty DPR:$src1),
                        (Ty (MulOp DPR:$src2,
                                   (Ty (NEONvduplane (Ty DPR_VFP2:$src3),
                                                     imm:$lane)))))))]>;
class N3VDMulOpSL16<bits<2> op21_20, bits<4> op11_8, InstrItinClass itin,
                    string OpcodeStr, string Dt,
                    ValueType Ty, SDNode MulOp, SDNode ShOp>
  : N3V<0, 1, op21_20, op11_8, 1, 0,
        (outs DPR:$Vd),
        (ins DPR:$src1, DPR:$Vn, DPR_8:$Vm, nohash_imm:$lane),
        NVMulSLFrm, itin,
        OpcodeStr, Dt, "$Vd, $Vn, $Vm[$lane]", "$src1 = $Vd",
        [(set (Ty DPR:$Vd),
              (Ty (ShOp (Ty DPR:$src1),
                        (Ty (MulOp DPR:$Vn,
                                   (Ty (NEONvduplane (Ty DPR_8:$Vm),
                                                     imm:$lane)))))))]>;

class N3VQMulOp<bit op24, bit op23, bits<2> op21_20, bits<4> op11_8, bit op4,
                InstrItinClass itin, string OpcodeStr, string Dt, ValueType Ty,
                SDNode MulOp, SDNode OpNode>
  : N3V<op24, op23, op21_20, op11_8, 1, op4,
        (outs QPR:$Vd), (ins QPR:$src1, QPR:$Vn, QPR:$Vm), N3RegFrm, itin,
        OpcodeStr, Dt, "$Vd, $Vn, $Vm", "$src1 = $Vd",
        [(set QPR:$Vd, (Ty (OpNode QPR:$src1,
                             (Ty (MulOp QPR:$Vn, QPR:$Vm)))))]>;
class N3VQMulOpSL<bits<2> op21_20, bits<4> op11_8, InstrItinClass itin,
                  string OpcodeStr, string Dt, ValueType ResTy, ValueType OpTy,
                  SDNode MulOp, SDNode ShOp>
  : N3V<1, 1, op21_20, op11_8, 1, 0,
        (outs QPR:$dst),
        (ins QPR:$src1, QPR:$src2, DPR_VFP2:$src3, nohash_imm:$lane),
        NVMulSLFrm, itin,
        OpcodeStr, Dt, "$dst, $src2, $src3[$lane]", "$src1 = $dst",
        [(set (ResTy QPR:$dst),
              (ResTy (ShOp (ResTy QPR:$src1),
                           (ResTy (MulOp QPR:$src2,
                                   (ResTy (NEONvduplane (OpTy DPR_VFP2:$src3),
                                                        imm:$lane)))))))]>;
class N3VQMulOpSL16<bits<2> op21_20, bits<4> op11_8, InstrItinClass itin,
                    string OpcodeStr, string Dt,
                    ValueType ResTy, ValueType OpTy,
                    SDNode MulOp, SDNode ShOp>
  : N3V<1, 1, op21_20, op11_8, 1, 0,
        (outs QPR:$dst),
        (ins QPR:$src1, QPR:$src2, DPR_8:$src3, nohash_imm:$lane),
        NVMulSLFrm, itin,
        OpcodeStr, Dt, "$dst, $src2, $src3[$lane]", "$src1 = $dst",
        [(set (ResTy QPR:$dst),
              (ResTy (ShOp (ResTy QPR:$src1),
                           (ResTy (MulOp QPR:$src2,
                                   (ResTy (NEONvduplane (OpTy DPR_8:$src3),
                                                        imm:$lane)))))))]>;

// Neon Intrinsic-Op instructions (VABA): double- and quad-register.
class N3VDIntOp<bit op24, bit op23, bits<2> op21_20, bits<4> op11_8, bit op4,
                InstrItinClass itin, string OpcodeStr, string Dt,
                ValueType Ty, Intrinsic IntOp, SDNode OpNode>
  : N3V<op24, op23, op21_20, op11_8, 0, op4,
        (outs DPR:$Vd), (ins DPR:$src1, DPR:$Vn, DPR:$Vm), N3RegFrm, itin,
        OpcodeStr, Dt, "$Vd, $Vn, $Vm", "$src1 = $Vd",
        [(set DPR:$Vd, (Ty (OpNode DPR:$src1,
                             (Ty (IntOp (Ty DPR:$Vn), (Ty DPR:$Vm))))))]>;
class N3VQIntOp<bit op24, bit op23, bits<2> op21_20, bits<4> op11_8, bit op4,
                InstrItinClass itin, string OpcodeStr, string Dt,
                ValueType Ty, Intrinsic IntOp, SDNode OpNode>
  : N3V<op24, op23, op21_20, op11_8, 1, op4,
        (outs QPR:$Vd), (ins QPR:$src1, QPR:$Vn, QPR:$Vm), N3RegFrm, itin,
        OpcodeStr, Dt, "$Vd, $Vn, $Vm", "$src1 = $Vd",
        [(set QPR:$Vd, (Ty (OpNode QPR:$src1,
                             (Ty (IntOp (Ty QPR:$Vn), (Ty QPR:$Vm))))))]>;

// Neon 3-argument intrinsics, both double- and quad-register.
// The destination register is also used as the first source operand register.
class N3VDInt3<bit op24, bit op23, bits<2> op21_20, bits<4> op11_8, bit op4,
               InstrItinClass itin, string OpcodeStr, string Dt,
               ValueType ResTy, ValueType OpTy, Intrinsic IntOp>
  : N3V<op24, op23, op21_20, op11_8, 0, op4,
        (outs DPR:$dst), (ins DPR:$src1, DPR:$src2, DPR:$src3), N3RegFrm, itin,
        OpcodeStr, Dt, "$dst, $src2, $src3", "$src1 = $dst",
        [(set DPR:$dst, (ResTy (IntOp (OpTy DPR:$src1),
                                      (OpTy DPR:$src2), (OpTy DPR:$src3))))]>;
class N3VQInt3<bit op24, bit op23, bits<2> op21_20, bits<4> op11_8, bit op4,
               InstrItinClass itin, string OpcodeStr, string Dt,
               ValueType ResTy, ValueType OpTy, Intrinsic IntOp>
  : N3V<op24, op23, op21_20, op11_8, 1, op4,
        (outs QPR:$dst), (ins QPR:$src1, QPR:$src2, QPR:$src3), N3RegFrm, itin,
        OpcodeStr, Dt, "$dst, $src2, $src3", "$src1 = $dst",
        [(set QPR:$dst, (ResTy (IntOp (OpTy QPR:$src1),
                                      (OpTy QPR:$src2), (OpTy QPR:$src3))))]>;

// Long Multiply-Add/Sub operations.
class N3VLMulOp<bit op24, bit op23, bits<2> op21_20, bits<4> op11_8, bit op4,
                InstrItinClass itin, string OpcodeStr, string Dt,
                ValueType TyQ, ValueType TyD, SDNode MulOp, SDNode OpNode>
  : N3V<op24, op23, op21_20, op11_8, 0, op4,
        (outs QPR:$Vd), (ins QPR:$src1, DPR:$Vn, DPR:$Vm), N3RegFrm, itin,
        OpcodeStr, Dt, "$Vd, $Vn, $Vm", "$src1 = $Vd",
        [(set QPR:$Vd, (OpNode (TyQ QPR:$src1),
                                (TyQ (MulOp (TyD DPR:$Vn),
                                            (TyD DPR:$Vm)))))]>;
class N3VLMulOpSL<bit op24, bits<2> op21_20, bits<4> op11_8,
                  InstrItinClass itin, string OpcodeStr, string Dt,
                  ValueType TyQ, ValueType TyD, SDNode MulOp, SDNode OpNode>
  : N3V<op24, 1, op21_20, op11_8, 1, 0, (outs QPR:$dst),
        (ins QPR:$src1, DPR:$src2, DPR_VFP2:$src3, nohash_imm:$lane),
        NVMulSLFrm, itin,
        OpcodeStr, Dt, "$dst, $src2, $src3[$lane]", "$src1 = $dst",
        [(set QPR:$dst,
          (OpNode (TyQ QPR:$src1),
                  (TyQ (MulOp (TyD DPR:$src2),
                              (TyD (NEONvduplane (TyD DPR_VFP2:$src3),
                                                 imm:$lane))))))]>;
class N3VLMulOpSL16<bit op24, bits<2> op21_20, bits<4> op11_8,
                    InstrItinClass itin, string OpcodeStr, string Dt,
                    ValueType TyQ, ValueType TyD, SDNode MulOp, SDNode OpNode>
  : N3V<op24, 1, op21_20, op11_8, 1, 0, (outs QPR:$dst),
        (ins QPR:$src1, DPR:$src2, DPR_8:$src3, nohash_imm:$lane),
        NVMulSLFrm, itin,
        OpcodeStr, Dt, "$dst, $src2, $src3[$lane]", "$src1 = $dst",
        [(set QPR:$dst,
          (OpNode (TyQ QPR:$src1),
                  (TyQ (MulOp (TyD DPR:$src2),
                              (TyD (NEONvduplane (TyD DPR_8:$src3),
                                                 imm:$lane))))))]>;

// Long Intrinsic-Op vector operations with explicit extend (VABAL).
class N3VLIntExtOp<bit op24, bit op23, bits<2> op21_20, bits<4> op11_8, bit op4,
                   InstrItinClass itin, string OpcodeStr, string Dt,
                   ValueType TyQ, ValueType TyD, Intrinsic IntOp, SDNode ExtOp,
                   SDNode OpNode>
  : N3V<op24, op23, op21_20, op11_8, 0, op4,
        (outs QPR:$Vd), (ins QPR:$src1, DPR:$Vn, DPR:$Vm), N3RegFrm, itin,
        OpcodeStr, Dt, "$Vd, $Vn, $Vm", "$src1 = $Vd",
        [(set QPR:$Vd, (OpNode (TyQ QPR:$src1),
                                (TyQ (ExtOp (TyD (IntOp (TyD DPR:$Vn),
                                                        (TyD DPR:$Vm)))))))]>;

// Neon Long 3-argument intrinsic.  The destination register is
// a quad-register and is also used as the first source operand register.
class N3VLInt3<bit op24, bit op23, bits<2> op21_20, bits<4> op11_8, bit op4,
               InstrItinClass itin, string OpcodeStr, string Dt,
               ValueType TyQ, ValueType TyD, Intrinsic IntOp>
  : N3V<op24, op23, op21_20, op11_8, 0, op4,
        (outs QPR:$Vd), (ins QPR:$src1, DPR:$Vn, DPR:$Vm), N3RegFrm, itin,
        OpcodeStr, Dt, "$Vd, $Vn, $Vm", "$src1 = $Vd",
        [(set QPR:$Vd,
          (TyQ (IntOp (TyQ QPR:$src1), (TyD DPR:$Vn), (TyD DPR:$Vm))))]>;
class N3VLInt3SL<bit op24, bits<2> op21_20, bits<4> op11_8, InstrItinClass itin,
                 string OpcodeStr, string Dt,
                 ValueType ResTy, ValueType OpTy, Intrinsic IntOp>
  : N3V<op24, 1, op21_20, op11_8, 1, 0,
        (outs QPR:$dst),
        (ins QPR:$src1, DPR:$src2, DPR_VFP2:$src3, nohash_imm:$lane),
        NVMulSLFrm, itin,
        OpcodeStr, Dt, "$dst, $src2, $src3[$lane]", "$src1 = $dst",
        [(set (ResTy QPR:$dst),
              (ResTy (IntOp (ResTy QPR:$src1),
                            (OpTy DPR:$src2),
                            (OpTy (NEONvduplane (OpTy DPR_VFP2:$src3),
                                                imm:$lane)))))]>;
class N3VLInt3SL16<bit op24, bits<2> op21_20, bits<4> op11_8,
                   InstrItinClass itin, string OpcodeStr, string Dt,
                   ValueType ResTy, ValueType OpTy, Intrinsic IntOp>
  : N3V<op24, 1, op21_20, op11_8, 1, 0,
        (outs QPR:$dst),
        (ins QPR:$src1, DPR:$src2, DPR_8:$src3, nohash_imm:$lane),
        NVMulSLFrm, itin,
        OpcodeStr, Dt, "$dst, $src2, $src3[$lane]", "$src1 = $dst",
        [(set (ResTy QPR:$dst),
              (ResTy (IntOp (ResTy QPR:$src1),
                            (OpTy DPR:$src2),
                            (OpTy (NEONvduplane (OpTy DPR_8:$src3),
                                                imm:$lane)))))]>;

// Narrowing 3-register intrinsics.
class N3VNInt<bit op24, bit op23, bits<2> op21_20, bits<4> op11_8, bit op4,
              string OpcodeStr, string Dt, ValueType TyD, ValueType TyQ,
              Intrinsic IntOp, bit Commutable>
  : N3V<op24, op23, op21_20, op11_8, 0, op4,
        (outs DPR:$dst), (ins QPR:$src1, QPR:$src2), N3RegFrm, IIC_VBINi4D,
        OpcodeStr, Dt, "$dst, $src1, $src2", "",
        [(set DPR:$dst, (TyD (IntOp (TyQ QPR:$src1), (TyQ QPR:$src2))))]> {
  let isCommutable = Commutable;
}

// Long 3-register operations.
class N3VL<bit op24, bit op23, bits<2> op21_20, bits<4> op11_8, bit op4,
           InstrItinClass itin, string OpcodeStr, string Dt,
           ValueType TyQ, ValueType TyD, SDNode OpNode, bit Commutable>
  : N3V<op24, op23, op21_20, op11_8, 0, op4,
        (outs QPR:$dst), (ins DPR:$src1, DPR:$src2), N3RegFrm, itin,
        OpcodeStr, Dt, "$dst, $src1, $src2", "",
        [(set QPR:$dst, (TyQ (OpNode (TyD DPR:$src1), (TyD DPR:$src2))))]> {
  let isCommutable = Commutable;
}
class N3VLSL<bit op24, bits<2> op21_20, bits<4> op11_8,
             InstrItinClass itin, string OpcodeStr, string Dt,
             ValueType TyQ, ValueType TyD, SDNode OpNode>
  : N3V<op24, 1, op21_20, op11_8, 1, 0,
        (outs QPR:$dst), (ins DPR:$src1, DPR_VFP2:$src2, nohash_imm:$lane),
        NVMulSLFrm, itin, OpcodeStr, Dt, "$dst, $src1, $src2[$lane]", "",
        [(set QPR:$dst,
          (TyQ (OpNode (TyD DPR:$src1),
                       (TyD (NEONvduplane (TyD DPR_VFP2:$src2),imm:$lane)))))]>;
class N3VLSL16<bit op24, bits<2> op21_20, bits<4> op11_8,
               InstrItinClass itin, string OpcodeStr, string Dt,
               ValueType TyQ, ValueType TyD, SDNode OpNode>
  : N3V<op24, 1, op21_20, op11_8, 1, 0,
        (outs QPR:$dst), (ins DPR:$src1, DPR_8:$src2, nohash_imm:$lane), 
        NVMulSLFrm, itin, OpcodeStr, Dt, "$dst, $src1, $src2[$lane]", "",
        [(set QPR:$dst,
          (TyQ (OpNode (TyD DPR:$src1),
                       (TyD (NEONvduplane (TyD DPR_8:$src2), imm:$lane)))))]>;

// Long 3-register operations with explicitly extended operands.
class N3VLExt<bit op24, bit op23, bits<2> op21_20, bits<4> op11_8, bit op4,
              InstrItinClass itin, string OpcodeStr, string Dt,
              ValueType TyQ, ValueType TyD, SDNode OpNode, SDNode ExtOp,
              bit Commutable>
  : N3V<op24, op23, op21_20, op11_8, 0, op4,
        (outs QPR:$Qd), (ins DPR:$Dn, DPR:$Dm), N3RegFrm, itin,
        OpcodeStr, Dt, "$Qd, $Dn, $Dm", "",
        [(set QPR:$Qd, (OpNode (TyQ (ExtOp (TyD DPR:$Dn))),
                                (TyQ (ExtOp (TyD DPR:$Dm)))))]> {
  let isCommutable = Commutable;
}

// Long 3-register intrinsics with explicit extend (VABDL).
class N3VLIntExt<bit op24, bit op23, bits<2> op21_20, bits<4> op11_8, bit op4,
                 InstrItinClass itin, string OpcodeStr, string Dt,
                 ValueType TyQ, ValueType TyD, Intrinsic IntOp, SDNode ExtOp,
                 bit Commutable>
  : N3V<op24, op23, op21_20, op11_8, 0, op4,
        (outs QPR:$dst), (ins DPR:$src1, DPR:$src2), N3RegFrm, itin,
        OpcodeStr, Dt, "$dst, $src1, $src2", "",
        [(set QPR:$dst, (TyQ (ExtOp (TyD (IntOp (TyD DPR:$src1),
                                                (TyD DPR:$src2))))))]> {
  let isCommutable = Commutable;
}

// Long 3-register intrinsics.
class N3VLInt<bit op24, bit op23, bits<2> op21_20, bits<4> op11_8, bit op4,
              InstrItinClass itin, string OpcodeStr, string Dt,
              ValueType TyQ, ValueType TyD, Intrinsic IntOp, bit Commutable>
  : N3V<op24, op23, op21_20, op11_8, 0, op4,
        (outs QPR:$dst), (ins DPR:$src1, DPR:$src2), N3RegFrm, itin,
        OpcodeStr, Dt, "$dst, $src1, $src2", "",
        [(set QPR:$dst, (TyQ (IntOp (TyD DPR:$src1), (TyD DPR:$src2))))]> {
  let isCommutable = Commutable;
}
class N3VLIntSL<bit op24, bits<2> op21_20, bits<4> op11_8, InstrItinClass itin,
                string OpcodeStr, string Dt,
                ValueType ResTy, ValueType OpTy, Intrinsic IntOp>
  : N3V<op24, 1, op21_20, op11_8, 1, 0,
        (outs QPR:$dst), (ins DPR:$src1, DPR_VFP2:$src2, nohash_imm:$lane),
        NVMulSLFrm, itin, OpcodeStr, Dt, "$dst, $src1, $src2[$lane]", "",
        [(set (ResTy QPR:$dst),
              (ResTy (IntOp (OpTy DPR:$src1),
                            (OpTy (NEONvduplane (OpTy DPR_VFP2:$src2),
                                                imm:$lane)))))]>;
class N3VLIntSL16<bit op24, bits<2> op21_20, bits<4> op11_8,
                  InstrItinClass itin, string OpcodeStr, string Dt,
                  ValueType ResTy, ValueType OpTy, Intrinsic IntOp>
  : N3V<op24, 1, op21_20, op11_8, 1, 0,
        (outs QPR:$dst), (ins DPR:$src1, DPR_8:$src2, nohash_imm:$lane), 
        NVMulSLFrm, itin, OpcodeStr, Dt, "$dst, $src1, $src2[$lane]", "",
        [(set (ResTy QPR:$dst),
              (ResTy (IntOp (OpTy DPR:$src1),
                            (OpTy (NEONvduplane (OpTy DPR_8:$src2),
                                                imm:$lane)))))]>;

// Wide 3-register operations.
class N3VW<bit op24, bit op23, bits<2> op21_20, bits<4> op11_8, bit op4,
           string OpcodeStr, string Dt, ValueType TyQ, ValueType TyD,
           SDNode OpNode, SDNode ExtOp, bit Commutable>
  : N3V<op24, op23, op21_20, op11_8, 0, op4,
        (outs QPR:$Qd), (ins QPR:$Qn, DPR:$Dm), N3RegFrm, IIC_VSUBiD,
        OpcodeStr, Dt, "$Qd, $Qn, $Dm", "",
        [(set QPR:$Qd, (OpNode (TyQ QPR:$Qn),
                                (TyQ (ExtOp (TyD DPR:$Dm)))))]> {
  let isCommutable = Commutable;
}

// Pairwise long 2-register intrinsics, both double- and quad-register.
class N2VDPLInt<bits<2> op24_23, bits<2> op21_20, bits<2> op19_18,
                bits<2> op17_16, bits<5> op11_7, bit op4,
                string OpcodeStr, string Dt,
                ValueType ResTy, ValueType OpTy, Intrinsic IntOp>
  : N2V<op24_23, op21_20, op19_18, op17_16, op11_7, 0, op4, (outs DPR:$dst),
        (ins DPR:$src), IIC_VSHLiD, OpcodeStr, Dt, "$dst, $src", "",
        [(set DPR:$dst, (ResTy (IntOp (OpTy DPR:$src))))]>;
class N2VQPLInt<bits<2> op24_23, bits<2> op21_20, bits<2> op19_18,
                bits<2> op17_16, bits<5> op11_7, bit op4,
                string OpcodeStr, string Dt,
                ValueType ResTy, ValueType OpTy, Intrinsic IntOp>
  : N2V<op24_23, op21_20, op19_18, op17_16, op11_7, 1, op4, (outs QPR:$dst),
        (ins QPR:$src), IIC_VSHLiD, OpcodeStr, Dt, "$dst, $src", "",
        [(set QPR:$dst, (ResTy (IntOp (OpTy QPR:$src))))]>;

// Pairwise long 2-register accumulate intrinsics,
// both double- and quad-register.
// The destination register is also used as the first source operand register.
class N2VDPLInt2<bits<2> op24_23, bits<2> op21_20, bits<2> op19_18,
                 bits<2> op17_16, bits<5> op11_7, bit op4,
                 string OpcodeStr, string Dt,
                 ValueType ResTy, ValueType OpTy, Intrinsic IntOp>
  : N2V<op24_23, op21_20, op19_18, op17_16, op11_7, 0, op4,
        (outs DPR:$Vd), (ins DPR:$src1, DPR:$Vm), IIC_VPALiD,
        OpcodeStr, Dt, "$Vd, $Vm", "$src1 = $Vd",
        [(set DPR:$Vd, (ResTy (IntOp (ResTy DPR:$src1), (OpTy DPR:$Vm))))]>;
class N2VQPLInt2<bits<2> op24_23, bits<2> op21_20, bits<2> op19_18,
                 bits<2> op17_16, bits<5> op11_7, bit op4,
                 string OpcodeStr, string Dt,
                 ValueType ResTy, ValueType OpTy, Intrinsic IntOp>
  : N2V<op24_23, op21_20, op19_18, op17_16, op11_7, 1, op4,
        (outs QPR:$Vd), (ins QPR:$src1, QPR:$Vm), IIC_VPALiQ,
        OpcodeStr, Dt, "$Vd, $Vm", "$src1 = $Vd",
        [(set QPR:$Vd, (ResTy (IntOp (ResTy QPR:$src1), (OpTy QPR:$Vm))))]>;

// Shift by immediate,
// both double- and quad-register.
class N2VDSh<bit op24, bit op23, bits<4> op11_8, bit op7, bit op4,
             Format f, InstrItinClass itin, string OpcodeStr, string Dt,
             ValueType Ty, SDNode OpNode>
  : N2VImm<op24, op23, op11_8, op7, 0, op4,
           (outs DPR:$dst), (ins DPR:$src, i32imm:$SIMM), f, itin,
           OpcodeStr, Dt, "$dst, $src, $SIMM", "",
           [(set DPR:$dst, (Ty (OpNode (Ty DPR:$src), (i32 imm:$SIMM))))]>;
class N2VQSh<bit op24, bit op23, bits<4> op11_8, bit op7, bit op4,
             Format f, InstrItinClass itin, string OpcodeStr, string Dt,
             ValueType Ty, SDNode OpNode>
  : N2VImm<op24, op23, op11_8, op7, 1, op4,
           (outs QPR:$dst), (ins QPR:$src, i32imm:$SIMM), f, itin,
           OpcodeStr, Dt, "$dst, $src, $SIMM", "",
           [(set QPR:$dst, (Ty (OpNode (Ty QPR:$src), (i32 imm:$SIMM))))]>;

// Long shift by immediate.
class N2VLSh<bit op24, bit op23, bits<4> op11_8, bit op7, bit op6, bit op4,
             string OpcodeStr, string Dt,
             ValueType ResTy, ValueType OpTy, SDNode OpNode>
  : N2VImm<op24, op23, op11_8, op7, op6, op4,
           (outs QPR:$dst), (ins DPR:$src, i32imm:$SIMM), N2RegVShLFrm,
           IIC_VSHLiD, OpcodeStr, Dt, "$dst, $src, $SIMM", "",
           [(set QPR:$dst, (ResTy (OpNode (OpTy DPR:$src),
                                          (i32 imm:$SIMM))))]>;

// Narrow shift by immediate.
class N2VNSh<bit op24, bit op23, bits<4> op11_8, bit op7, bit op6, bit op4,
             InstrItinClass itin, string OpcodeStr, string Dt,
             ValueType ResTy, ValueType OpTy, SDNode OpNode>
  : N2VImm<op24, op23, op11_8, op7, op6, op4,
           (outs DPR:$dst), (ins QPR:$src, i32imm:$SIMM), N2RegVShRFrm, itin,
           OpcodeStr, Dt, "$dst, $src, $SIMM", "",
           [(set DPR:$dst, (ResTy (OpNode (OpTy QPR:$src),
                                          (i32 imm:$SIMM))))]>;

// Shift right by immediate and accumulate,
// both double- and quad-register.
class N2VDShAdd<bit op24, bit op23, bits<4> op11_8, bit op7, bit op4,
                string OpcodeStr, string Dt, ValueType Ty, SDNode ShOp>
  : N2VImm<op24, op23, op11_8, op7, 0, op4, (outs DPR:$Vd),
           (ins DPR:$src1, DPR:$Vm, i32imm:$SIMM), N2RegVShRFrm, IIC_VPALiD,
           OpcodeStr, Dt, "$Vd, $Vm, $SIMM", "$src1 = $Vd",
           [(set DPR:$Vd, (Ty (add DPR:$src1,
                                (Ty (ShOp DPR:$Vm, (i32 imm:$SIMM))))))]>;
class N2VQShAdd<bit op24, bit op23, bits<4> op11_8, bit op7, bit op4,
                string OpcodeStr, string Dt, ValueType Ty, SDNode ShOp>
  : N2VImm<op24, op23, op11_8, op7, 1, op4, (outs QPR:$Vd),
           (ins QPR:$src1, QPR:$Vm, i32imm:$SIMM), N2RegVShRFrm, IIC_VPALiD,
           OpcodeStr, Dt, "$Vd, $Vm, $SIMM", "$src1 = $Vd",
           [(set QPR:$Vd, (Ty (add QPR:$src1,
                                (Ty (ShOp QPR:$Vm, (i32 imm:$SIMM))))))]>;

// Shift by immediate and insert,
// both double- and quad-register.
class N2VDShIns<bit op24, bit op23, bits<4> op11_8, bit op7, bit op4,
                Format f, string OpcodeStr, string Dt, ValueType Ty,SDNode ShOp>
  : N2VImm<op24, op23, op11_8, op7, 0, op4, (outs DPR:$Vd),
           (ins DPR:$src1, DPR:$Vm, i32imm:$SIMM), f, IIC_VSHLiD,
           OpcodeStr, Dt, "$Vd, $Vm, $SIMM", "$src1 = $Vd",
           [(set DPR:$Vd, (Ty (ShOp DPR:$src1, DPR:$Vm, (i32 imm:$SIMM))))]>;
class N2VQShIns<bit op24, bit op23, bits<4> op11_8, bit op7, bit op4,
                Format f, string OpcodeStr, string Dt, ValueType Ty,SDNode ShOp>
  : N2VImm<op24, op23, op11_8, op7, 1, op4, (outs QPR:$Vd),
           (ins QPR:$src1, QPR:$Vm, i32imm:$SIMM), f, IIC_VSHLiQ,
           OpcodeStr, Dt, "$Vd, $Vm, $SIMM", "$src1 = $Vd",
           [(set QPR:$Vd, (Ty (ShOp QPR:$src1, QPR:$Vm, (i32 imm:$SIMM))))]>;

// Convert, with fractional bits immediate,
// both double- and quad-register.
class N2VCvtD<bit op24, bit op23, bits<4> op11_8, bit op7, bit op4,
              string OpcodeStr, string Dt, ValueType ResTy, ValueType OpTy,
              Intrinsic IntOp>
  : N2VImm<op24, op23, op11_8, op7, 0, op4,
           (outs DPR:$Vd), (ins DPR:$Vm, neon_vcvt_imm32:$SIMM), NVCVTFrm,
           IIC_VUNAD, OpcodeStr, Dt, "$Vd, $Vm, $SIMM", "",
           [(set DPR:$Vd, (ResTy (IntOp (OpTy DPR:$Vm), (i32 imm:$SIMM))))]>;
class N2VCvtQ<bit op24, bit op23, bits<4> op11_8, bit op7, bit op4,
              string OpcodeStr, string Dt, ValueType ResTy, ValueType OpTy,
              Intrinsic IntOp>
  : N2VImm<op24, op23, op11_8, op7, 1, op4,
           (outs QPR:$Vd), (ins QPR:$Vm, neon_vcvt_imm32:$SIMM), NVCVTFrm,
           IIC_VUNAQ, OpcodeStr, Dt, "$Vd, $Vm, $SIMM", "",
           [(set QPR:$Vd, (ResTy (IntOp (OpTy QPR:$Vm), (i32 imm:$SIMM))))]>;

//===----------------------------------------------------------------------===//
// Multiclasses
//===----------------------------------------------------------------------===//

// Abbreviations used in multiclass suffixes:
//   Q = quarter int (8 bit) elements
//   H = half int (16 bit) elements
//   S = single int (32 bit) elements
//   D = double int (64 bit) elements

// Neon 2-register vector operations -- for disassembly only.

// First with only element sizes of 8, 16 and 32 bits:
multiclass N2V_QHS_cmp<bits<2> op24_23, bits<2> op21_20, bits<2> op17_16,
                       bits<5> op11_7, bit op4, string opc, string Dt,
                       string asm, SDNode OpNode> {
  // 64-bit vector types.
  def v8i8  : N2V<op24_23, op21_20, 0b00, op17_16, op11_7, 0, op4,
                  (outs DPR:$dst), (ins DPR:$src), NoItinerary,
                  opc, !strconcat(Dt, "8"), asm, "",
                  [(set DPR:$dst, (v8i8 (OpNode (v8i8 DPR:$src))))]>;
  def v4i16 : N2V<op24_23, op21_20, 0b01, op17_16, op11_7, 0, op4,
                  (outs DPR:$dst), (ins DPR:$src), NoItinerary,
                  opc, !strconcat(Dt, "16"), asm, "",
                  [(set DPR:$dst, (v4i16 (OpNode (v4i16 DPR:$src))))]>;
  def v2i32 : N2V<op24_23, op21_20, 0b10, op17_16, op11_7, 0, op4,
                  (outs DPR:$dst), (ins DPR:$src), NoItinerary,
                  opc, !strconcat(Dt, "32"), asm, "",
                  [(set DPR:$dst, (v2i32 (OpNode (v2i32 DPR:$src))))]>;
  def v2f32 : N2V<op24_23, op21_20, 0b10, op17_16, op11_7, 0, op4,
                  (outs DPR:$dst), (ins DPR:$src), NoItinerary,
                  opc, "f32", asm, "",
                  [(set DPR:$dst, (v2f32 (OpNode (v2f32 DPR:$src))))]> {
    let Inst{10} = 1; // overwrite F = 1
  }

  // 128-bit vector types.
  def v16i8 : N2V<op24_23, op21_20, 0b00, op17_16, op11_7, 1, op4,
                  (outs QPR:$dst), (ins QPR:$src), NoItinerary,
                  opc, !strconcat(Dt, "8"), asm, "",
                  [(set QPR:$dst, (v16i8 (OpNode (v16i8 QPR:$src))))]>;
  def v8i16 : N2V<op24_23, op21_20, 0b01, op17_16, op11_7, 1, op4,
                  (outs QPR:$dst), (ins QPR:$src), NoItinerary,
                  opc, !strconcat(Dt, "16"), asm, "",
                  [(set QPR:$dst, (v8i16 (OpNode (v8i16 QPR:$src))))]>;
  def v4i32 : N2V<op24_23, op21_20, 0b10, op17_16, op11_7, 1, op4,
                  (outs QPR:$dst), (ins QPR:$src), NoItinerary,
                  opc, !strconcat(Dt, "32"), asm, "",
                  [(set QPR:$dst, (v4i32 (OpNode (v4i32 QPR:$src))))]>;
  def v4f32 : N2V<op24_23, op21_20, 0b10, op17_16, op11_7, 1, op4,
                  (outs QPR:$dst), (ins QPR:$src), NoItinerary,
                  opc, "f32", asm, "",
                  [(set QPR:$dst, (v4f32 (OpNode (v4f32 QPR:$src))))]> {
    let Inst{10} = 1; // overwrite F = 1
  }
}

// Neon 3-register vector operations.

// First with only element sizes of 8, 16 and 32 bits:
multiclass N3V_QHS<bit op24, bit op23, bits<4> op11_8, bit op4,
                   InstrItinClass itinD16, InstrItinClass itinD32,
                   InstrItinClass itinQ16, InstrItinClass itinQ32,
                   string OpcodeStr, string Dt,
                   SDNode OpNode, bit Commutable = 0> {
  // 64-bit vector types.
  def v8i8  : N3VD<op24, op23, 0b00, op11_8, op4, itinD16, 
                   OpcodeStr, !strconcat(Dt, "8"),
                   v8i8, v8i8, OpNode, Commutable>;
  def v4i16 : N3VD<op24, op23, 0b01, op11_8, op4, itinD16,
                   OpcodeStr, !strconcat(Dt, "16"),
                   v4i16, v4i16, OpNode, Commutable>;
  def v2i32 : N3VD<op24, op23, 0b10, op11_8, op4, itinD32,
                   OpcodeStr, !strconcat(Dt, "32"),
                   v2i32, v2i32, OpNode, Commutable>;

  // 128-bit vector types.
  def v16i8 : N3VQ<op24, op23, 0b00, op11_8, op4, itinQ16,
                   OpcodeStr, !strconcat(Dt, "8"),
                   v16i8, v16i8, OpNode, Commutable>;
  def v8i16 : N3VQ<op24, op23, 0b01, op11_8, op4, itinQ16,
                   OpcodeStr, !strconcat(Dt, "16"),
                   v8i16, v8i16, OpNode, Commutable>;
  def v4i32 : N3VQ<op24, op23, 0b10, op11_8, op4, itinQ32,
                   OpcodeStr, !strconcat(Dt, "32"),
                   v4i32, v4i32, OpNode, Commutable>;
}

multiclass N3VSL_HS<bits<4> op11_8, string OpcodeStr, string Dt, SDNode ShOp> {
  def v4i16 : N3VDSL16<0b01, op11_8, OpcodeStr, !strconcat(Dt, "16"),
                       v4i16, ShOp>;
  def v2i32 : N3VDSL<0b10, op11_8, IIC_VMULi32D, OpcodeStr, !strconcat(Dt,"32"),
                     v2i32, ShOp>;
  def v8i16 : N3VQSL16<0b01, op11_8, OpcodeStr, !strconcat(Dt, "16"),
                       v8i16, v4i16, ShOp>;
  def v4i32 : N3VQSL<0b10, op11_8, IIC_VMULi32Q, OpcodeStr, !strconcat(Dt,"32"),
                     v4i32, v2i32, ShOp>;
}

// ....then also with element size 64 bits:
multiclass N3V_QHSD<bit op24, bit op23, bits<4> op11_8, bit op4,
                    InstrItinClass itinD, InstrItinClass itinQ,
                    string OpcodeStr, string Dt,
                    SDNode OpNode, bit Commutable = 0>
  : N3V_QHS<op24, op23, op11_8, op4, itinD, itinD, itinQ, itinQ,
            OpcodeStr, Dt, OpNode, Commutable> {
  def v1i64 : N3VD<op24, op23, 0b11, op11_8, op4, itinD,
                   OpcodeStr, !strconcat(Dt, "64"),
                   v1i64, v1i64, OpNode, Commutable>;
  def v2i64 : N3VQ<op24, op23, 0b11, op11_8, op4, itinQ,
                   OpcodeStr, !strconcat(Dt, "64"),
                   v2i64, v2i64, OpNode, Commutable>;
}


// Neon Narrowing 2-register vector operations,
//   source operand element sizes of 16, 32 and 64 bits:
multiclass N2VN_HSD<bits<2> op24_23, bits<2> op21_20, bits<2> op17_16,
                    bits<5> op11_7, bit op6, bit op4, 
                    InstrItinClass itin, string OpcodeStr, string Dt,
                    SDNode OpNode> {
  def v8i8  : N2VN<op24_23, op21_20, 0b00, op17_16, op11_7, op6, op4,
                   itin, OpcodeStr, !strconcat(Dt, "16"),
                   v8i8, v8i16, OpNode>;
  def v4i16 : N2VN<op24_23, op21_20, 0b01, op17_16, op11_7, op6, op4,
                   itin, OpcodeStr, !strconcat(Dt, "32"),
                   v4i16, v4i32, OpNode>;
  def v2i32 : N2VN<op24_23, op21_20, 0b10, op17_16, op11_7, op6, op4,
                   itin, OpcodeStr, !strconcat(Dt, "64"),
                   v2i32, v2i64, OpNode>;
}

// Neon Narrowing 2-register vector intrinsics,
//   source operand element sizes of 16, 32 and 64 bits:
multiclass N2VNInt_HSD<bits<2> op24_23, bits<2> op21_20, bits<2> op17_16,
                       bits<5> op11_7, bit op6, bit op4, 
                       InstrItinClass itin, string OpcodeStr, string Dt,
                       Intrinsic IntOp> {
  def v8i8  : N2VNInt<op24_23, op21_20, 0b00, op17_16, op11_7, op6, op4,
                      itin, OpcodeStr, !strconcat(Dt, "16"),
                      v8i8, v8i16, IntOp>;
  def v4i16 : N2VNInt<op24_23, op21_20, 0b01, op17_16, op11_7, op6, op4,
                      itin, OpcodeStr, !strconcat(Dt, "32"),
                      v4i16, v4i32, IntOp>;
  def v2i32 : N2VNInt<op24_23, op21_20, 0b10, op17_16, op11_7, op6, op4,
                      itin, OpcodeStr, !strconcat(Dt, "64"),
                      v2i32, v2i64, IntOp>;
}


// Neon Lengthening 2-register vector intrinsic (currently specific to VMOVL).
//   source operand element sizes of 16, 32 and 64 bits:
multiclass N2VL_QHS<bits<2> op24_23, bits<5> op11_7, bit op6, bit op4,
                    string OpcodeStr, string Dt, SDNode OpNode> {
  def v8i16 : N2VL<op24_23, 0b00, 0b10, 0b00, op11_7, op6, op4, IIC_VQUNAiD,
                   OpcodeStr, !strconcat(Dt, "8"), v8i16, v8i8, OpNode>;
  def v4i32 : N2VL<op24_23, 0b01, 0b00, 0b00, op11_7, op6, op4, IIC_VQUNAiD,
                   OpcodeStr, !strconcat(Dt, "16"), v4i32, v4i16, OpNode>;
  def v2i64 : N2VL<op24_23, 0b10, 0b00, 0b00, op11_7, op6, op4, IIC_VQUNAiD,
                   OpcodeStr, !strconcat(Dt, "32"), v2i64, v2i32, OpNode>;
}


// Neon 3-register vector intrinsics.

// First with only element sizes of 16 and 32 bits:
multiclass N3VInt_HS<bit op24, bit op23, bits<4> op11_8, bit op4, Format f,
                     InstrItinClass itinD16, InstrItinClass itinD32,
                     InstrItinClass itinQ16, InstrItinClass itinQ32,
                     string OpcodeStr, string Dt,
                     Intrinsic IntOp, bit Commutable = 0> {
  // 64-bit vector types.
  def v4i16 : N3VDInt<op24, op23, 0b01, op11_8, op4, f, itinD16,
                      OpcodeStr, !strconcat(Dt, "16"),
                      v4i16, v4i16, IntOp, Commutable>;
  def v2i32 : N3VDInt<op24, op23, 0b10, op11_8, op4, f, itinD32,
                      OpcodeStr, !strconcat(Dt, "32"),
                      v2i32, v2i32, IntOp, Commutable>;

  // 128-bit vector types.
  def v8i16 : N3VQInt<op24, op23, 0b01, op11_8, op4, f, itinQ16,
                      OpcodeStr, !strconcat(Dt, "16"),
                      v8i16, v8i16, IntOp, Commutable>;
  def v4i32 : N3VQInt<op24, op23, 0b10, op11_8, op4, f, itinQ32,
                      OpcodeStr, !strconcat(Dt, "32"),
                      v4i32, v4i32, IntOp, Commutable>;
}
multiclass N3VInt_HSSh<bit op24, bit op23, bits<4> op11_8, bit op4, Format f,
                     InstrItinClass itinD16, InstrItinClass itinD32,
                     InstrItinClass itinQ16, InstrItinClass itinQ32,
                     string OpcodeStr, string Dt,
                     Intrinsic IntOp> {
  // 64-bit vector types.
  def v4i16 : N3VDIntSh<op24, op23, 0b01, op11_8, op4, f, itinD16,
                      OpcodeStr, !strconcat(Dt, "16"),
                      v4i16, v4i16, IntOp>;
  def v2i32 : N3VDIntSh<op24, op23, 0b10, op11_8, op4, f, itinD32,
                      OpcodeStr, !strconcat(Dt, "32"),
                      v2i32, v2i32, IntOp>;

  // 128-bit vector types.
  def v8i16 : N3VQIntSh<op24, op23, 0b01, op11_8, op4, f, itinQ16,
                      OpcodeStr, !strconcat(Dt, "16"),
                      v8i16, v8i16, IntOp>;
  def v4i32 : N3VQIntSh<op24, op23, 0b10, op11_8, op4, f, itinQ32,
                      OpcodeStr, !strconcat(Dt, "32"),
                      v4i32, v4i32, IntOp>;
}

multiclass N3VIntSL_HS<bits<4> op11_8, 
                       InstrItinClass itinD16, InstrItinClass itinD32,
                       InstrItinClass itinQ16, InstrItinClass itinQ32,
                       string OpcodeStr, string Dt, Intrinsic IntOp> {
  def v4i16 : N3VDIntSL16<0b01, op11_8, itinD16,
                          OpcodeStr, !strconcat(Dt, "16"), v4i16, IntOp>;
  def v2i32 : N3VDIntSL<0b10, op11_8, itinD32,
                        OpcodeStr, !strconcat(Dt, "32"), v2i32, IntOp>;
  def v8i16 : N3VQIntSL16<0b01, op11_8, itinQ16,
                          OpcodeStr, !strconcat(Dt, "16"), v8i16, v4i16, IntOp>;
  def v4i32 : N3VQIntSL<0b10, op11_8, itinQ32,
                        OpcodeStr, !strconcat(Dt, "32"), v4i32, v2i32, IntOp>;
}

// ....then also with element size of 8 bits:
multiclass N3VInt_QHS<bit op24, bit op23, bits<4> op11_8, bit op4, Format f,
                      InstrItinClass itinD16, InstrItinClass itinD32,
                      InstrItinClass itinQ16, InstrItinClass itinQ32,
                      string OpcodeStr, string Dt,
                      Intrinsic IntOp, bit Commutable = 0>
  : N3VInt_HS<op24, op23, op11_8, op4, f, itinD16, itinD32, itinQ16, itinQ32,
              OpcodeStr, Dt, IntOp, Commutable> {
  def v8i8  : N3VDInt<op24, op23, 0b00, op11_8, op4, f, itinD16,
                      OpcodeStr, !strconcat(Dt, "8"),
                      v8i8, v8i8, IntOp, Commutable>;
  def v16i8 : N3VQInt<op24, op23, 0b00, op11_8, op4, f, itinQ16,
                      OpcodeStr, !strconcat(Dt, "8"),
                      v16i8, v16i8, IntOp, Commutable>;
}
multiclass N3VInt_QHSSh<bit op24, bit op23, bits<4> op11_8, bit op4, Format f,
                      InstrItinClass itinD16, InstrItinClass itinD32,
                      InstrItinClass itinQ16, InstrItinClass itinQ32,
                      string OpcodeStr, string Dt,
                      Intrinsic IntOp>
  : N3VInt_HSSh<op24, op23, op11_8, op4, f, itinD16, itinD32, itinQ16, itinQ32,
              OpcodeStr, Dt, IntOp> {
  def v8i8  : N3VDIntSh<op24, op23, 0b00, op11_8, op4, f, itinD16,
                      OpcodeStr, !strconcat(Dt, "8"),
                      v8i8, v8i8, IntOp>;
  def v16i8 : N3VQIntSh<op24, op23, 0b00, op11_8, op4, f, itinQ16,
                      OpcodeStr, !strconcat(Dt, "8"),
                      v16i8, v16i8, IntOp>;
}


// ....then also with element size of 64 bits:
multiclass N3VInt_QHSD<bit op24, bit op23, bits<4> op11_8, bit op4, Format f,
                       InstrItinClass itinD16, InstrItinClass itinD32,
                       InstrItinClass itinQ16, InstrItinClass itinQ32,
                       string OpcodeStr, string Dt,
                       Intrinsic IntOp, bit Commutable = 0>
  : N3VInt_QHS<op24, op23, op11_8, op4, f, itinD16, itinD32, itinQ16, itinQ32,
               OpcodeStr, Dt, IntOp, Commutable> {
  def v1i64 : N3VDInt<op24, op23, 0b11, op11_8, op4, f, itinD32,
                      OpcodeStr, !strconcat(Dt, "64"),
                      v1i64, v1i64, IntOp, Commutable>;
  def v2i64 : N3VQInt<op24, op23, 0b11, op11_8, op4, f, itinQ32,
                      OpcodeStr, !strconcat(Dt, "64"),
                      v2i64, v2i64, IntOp, Commutable>;
}
multiclass N3VInt_QHSDSh<bit op24, bit op23, bits<4> op11_8, bit op4, Format f,
                       InstrItinClass itinD16, InstrItinClass itinD32,
                       InstrItinClass itinQ16, InstrItinClass itinQ32,
                       string OpcodeStr, string Dt,
                       Intrinsic IntOp>
  : N3VInt_QHSSh<op24, op23, op11_8, op4, f, itinD16, itinD32, itinQ16, itinQ32,
               OpcodeStr, Dt, IntOp> {
  def v1i64 : N3VDIntSh<op24, op23, 0b11, op11_8, op4, f, itinD32,
                      OpcodeStr, !strconcat(Dt, "64"),
                      v1i64, v1i64, IntOp>;
  def v2i64 : N3VQIntSh<op24, op23, 0b11, op11_8, op4, f, itinQ32,
                      OpcodeStr, !strconcat(Dt, "64"),
                      v2i64, v2i64, IntOp>;
}

// Neon Narrowing 3-register vector intrinsics,
//   source operand element sizes of 16, 32 and 64 bits:
multiclass N3VNInt_HSD<bit op24, bit op23, bits<4> op11_8, bit op4,
                       string OpcodeStr, string Dt,
                       Intrinsic IntOp, bit Commutable = 0> {
  def v8i8  : N3VNInt<op24, op23, 0b00, op11_8, op4,
                      OpcodeStr, !strconcat(Dt, "16"),
                      v8i8, v8i16, IntOp, Commutable>;
  def v4i16 : N3VNInt<op24, op23, 0b01, op11_8, op4,
                      OpcodeStr, !strconcat(Dt, "32"),
                      v4i16, v4i32, IntOp, Commutable>;
  def v2i32 : N3VNInt<op24, op23, 0b10, op11_8, op4,
                      OpcodeStr, !strconcat(Dt, "64"),
                      v2i32, v2i64, IntOp, Commutable>;
}


// Neon Long 3-register vector operations.

multiclass N3VL_QHS<bit op24, bit op23, bits<4> op11_8, bit op4,
                    InstrItinClass itin16, InstrItinClass itin32,
                    string OpcodeStr, string Dt,
                    SDNode OpNode, bit Commutable = 0> {
  def v8i16 : N3VL<op24, op23, 0b00, op11_8, op4, itin16,
                   OpcodeStr, !strconcat(Dt, "8"),
                   v8i16, v8i8, OpNode, Commutable>;
  def v4i32 : N3VL<op24, op23, 0b01, op11_8, op4, itin16, 
                   OpcodeStr, !strconcat(Dt, "16"),
                   v4i32, v4i16, OpNode, Commutable>;
  def v2i64 : N3VL<op24, op23, 0b10, op11_8, op4, itin32,
                   OpcodeStr, !strconcat(Dt, "32"),
                   v2i64, v2i32, OpNode, Commutable>;
}

multiclass N3VLSL_HS<bit op24, bits<4> op11_8,
                     InstrItinClass itin, string OpcodeStr, string Dt,
                     SDNode OpNode> {
  def v4i16 : N3VLSL16<op24, 0b01, op11_8, itin, OpcodeStr,
                       !strconcat(Dt, "16"), v4i32, v4i16, OpNode>;
  def v2i32 : N3VLSL<op24, 0b10, op11_8, itin, OpcodeStr,
                     !strconcat(Dt, "32"), v2i64, v2i32, OpNode>;
}

multiclass N3VLExt_QHS<bit op24, bit op23, bits<4> op11_8, bit op4,
                       InstrItinClass itin16, InstrItinClass itin32,
                       string OpcodeStr, string Dt,
                       SDNode OpNode, SDNode ExtOp, bit Commutable = 0> {
  def v8i16 : N3VLExt<op24, op23, 0b00, op11_8, op4, itin16,
                      OpcodeStr, !strconcat(Dt, "8"),
                      v8i16, v8i8, OpNode, ExtOp, Commutable>;
  def v4i32 : N3VLExt<op24, op23, 0b01, op11_8, op4, itin16, 
                      OpcodeStr, !strconcat(Dt, "16"),
                      v4i32, v4i16, OpNode, ExtOp, Commutable>;
  def v2i64 : N3VLExt<op24, op23, 0b10, op11_8, op4, itin32,
                      OpcodeStr, !strconcat(Dt, "32"),
                      v2i64, v2i32, OpNode, ExtOp, Commutable>;
}

// Neon Long 3-register vector intrinsics.

// First with only element sizes of 16 and 32 bits:
multiclass N3VLInt_HS<bit op24, bit op23, bits<4> op11_8, bit op4,
                      InstrItinClass itin16, InstrItinClass itin32,
                      string OpcodeStr, string Dt,
                      Intrinsic IntOp, bit Commutable = 0> {
  def v4i32 : N3VLInt<op24, op23, 0b01, op11_8, op4, itin16, 
                      OpcodeStr, !strconcat(Dt, "16"),
                      v4i32, v4i16, IntOp, Commutable>;
  def v2i64 : N3VLInt<op24, op23, 0b10, op11_8, op4, itin32,
                      OpcodeStr, !strconcat(Dt, "32"),
                      v2i64, v2i32, IntOp, Commutable>;
}

multiclass N3VLIntSL_HS<bit op24, bits<4> op11_8,
                        InstrItinClass itin, string OpcodeStr, string Dt,
                        Intrinsic IntOp> {
  def v4i16 : N3VLIntSL16<op24, 0b01, op11_8, itin, 
                          OpcodeStr, !strconcat(Dt, "16"), v4i32, v4i16, IntOp>;
  def v2i32 : N3VLIntSL<op24, 0b10, op11_8, itin,
                        OpcodeStr, !strconcat(Dt, "32"), v2i64, v2i32, IntOp>;
}

// ....then also with element size of 8 bits:
multiclass N3VLInt_QHS<bit op24, bit op23, bits<4> op11_8, bit op4,
                       InstrItinClass itin16, InstrItinClass itin32,
                       string OpcodeStr, string Dt,
                       Intrinsic IntOp, bit Commutable = 0>
  : N3VLInt_HS<op24, op23, op11_8, op4, itin16, itin32, OpcodeStr, Dt,
               IntOp, Commutable> {
  def v8i16 : N3VLInt<op24, op23, 0b00, op11_8, op4, itin16,
                      OpcodeStr, !strconcat(Dt, "8"),
                      v8i16, v8i8, IntOp, Commutable>;
}

// ....with explicit extend (VABDL).
multiclass N3VLIntExt_QHS<bit op24, bit op23, bits<4> op11_8, bit op4,
                       InstrItinClass itin, string OpcodeStr, string Dt,
                       Intrinsic IntOp, SDNode ExtOp, bit Commutable = 0> {
  def v8i16 : N3VLIntExt<op24, op23, 0b00, op11_8, op4, itin,
                         OpcodeStr, !strconcat(Dt, "8"),
                         v8i16, v8i8, IntOp, ExtOp, Commutable>;
  def v4i32 : N3VLIntExt<op24, op23, 0b01, op11_8, op4, itin, 
                         OpcodeStr, !strconcat(Dt, "16"),
                         v4i32, v4i16, IntOp, ExtOp, Commutable>;
  def v2i64 : N3VLIntExt<op24, op23, 0b10, op11_8, op4, itin,
                         OpcodeStr, !strconcat(Dt, "32"),
                         v2i64, v2i32, IntOp, ExtOp, Commutable>;
}


// Neon Wide 3-register vector intrinsics,
//   source operand element sizes of 8, 16 and 32 bits:
multiclass N3VW_QHS<bit op24, bit op23, bits<4> op11_8, bit op4,
                    string OpcodeStr, string Dt,
                    SDNode OpNode, SDNode ExtOp, bit Commutable = 0> {
  def v8i16 : N3VW<op24, op23, 0b00, op11_8, op4,
                   OpcodeStr, !strconcat(Dt, "8"),
                   v8i16, v8i8, OpNode, ExtOp, Commutable>;
  def v4i32 : N3VW<op24, op23, 0b01, op11_8, op4,
                   OpcodeStr, !strconcat(Dt, "16"),
                   v4i32, v4i16, OpNode, ExtOp, Commutable>;
  def v2i64 : N3VW<op24, op23, 0b10, op11_8, op4,
                   OpcodeStr, !strconcat(Dt, "32"),
                   v2i64, v2i32, OpNode, ExtOp, Commutable>;
}


// Neon Multiply-Op vector operations,
//   element sizes of 8, 16 and 32 bits:
multiclass N3VMulOp_QHS<bit op24, bit op23, bits<4> op11_8, bit op4,
                        InstrItinClass itinD16, InstrItinClass itinD32,
                        InstrItinClass itinQ16, InstrItinClass itinQ32,
                        string OpcodeStr, string Dt, SDNode OpNode> {
  // 64-bit vector types.
  def v8i8  : N3VDMulOp<op24, op23, 0b00, op11_8, op4, itinD16,
                        OpcodeStr, !strconcat(Dt, "8"), v8i8, mul, OpNode>;
  def v4i16 : N3VDMulOp<op24, op23, 0b01, op11_8, op4, itinD16,
                        OpcodeStr, !strconcat(Dt, "16"), v4i16, mul, OpNode>;
  def v2i32 : N3VDMulOp<op24, op23, 0b10, op11_8, op4, itinD32,
                        OpcodeStr, !strconcat(Dt, "32"), v2i32, mul, OpNode>;

  // 128-bit vector types.
  def v16i8 : N3VQMulOp<op24, op23, 0b00, op11_8, op4, itinQ16,
                        OpcodeStr, !strconcat(Dt, "8"), v16i8, mul, OpNode>;
  def v8i16 : N3VQMulOp<op24, op23, 0b01, op11_8, op4, itinQ16,
                        OpcodeStr, !strconcat(Dt, "16"), v8i16, mul, OpNode>;
  def v4i32 : N3VQMulOp<op24, op23, 0b10, op11_8, op4, itinQ32,
                        OpcodeStr, !strconcat(Dt, "32"), v4i32, mul, OpNode>;
}

multiclass N3VMulOpSL_HS<bits<4> op11_8, 
                         InstrItinClass itinD16, InstrItinClass itinD32,
                         InstrItinClass itinQ16, InstrItinClass itinQ32,
                         string OpcodeStr, string Dt, SDNode ShOp> {
  def v4i16 : N3VDMulOpSL16<0b01, op11_8, itinD16,
                            OpcodeStr, !strconcat(Dt, "16"), v4i16, mul, ShOp>;
  def v2i32 : N3VDMulOpSL<0b10, op11_8, itinD32,
                          OpcodeStr, !strconcat(Dt, "32"), v2i32, mul, ShOp>;
  def v8i16 : N3VQMulOpSL16<0b01, op11_8, itinQ16,
                            OpcodeStr, !strconcat(Dt, "16"), v8i16, v4i16,
                            mul, ShOp>;
  def v4i32 : N3VQMulOpSL<0b10, op11_8, itinQ32,
                          OpcodeStr, !strconcat(Dt, "32"), v4i32, v2i32,
                          mul, ShOp>;
}

// Neon Intrinsic-Op vector operations,
//   element sizes of 8, 16 and 32 bits:
multiclass N3VIntOp_QHS<bit op24, bit op23, bits<4> op11_8, bit op4,
                        InstrItinClass itinD, InstrItinClass itinQ,
                        string OpcodeStr, string Dt, Intrinsic IntOp,
                        SDNode OpNode> {
  // 64-bit vector types.
  def v8i8  : N3VDIntOp<op24, op23, 0b00, op11_8, op4, itinD,
                        OpcodeStr, !strconcat(Dt, "8"), v8i8, IntOp, OpNode>;
  def v4i16 : N3VDIntOp<op24, op23, 0b01, op11_8, op4, itinD,
                        OpcodeStr, !strconcat(Dt, "16"), v4i16, IntOp, OpNode>;
  def v2i32 : N3VDIntOp<op24, op23, 0b10, op11_8, op4, itinD,
                        OpcodeStr, !strconcat(Dt, "32"), v2i32, IntOp, OpNode>;

  // 128-bit vector types.
  def v16i8 : N3VQIntOp<op24, op23, 0b00, op11_8, op4, itinQ,
                        OpcodeStr, !strconcat(Dt, "8"), v16i8, IntOp, OpNode>;
  def v8i16 : N3VQIntOp<op24, op23, 0b01, op11_8, op4, itinQ,
                        OpcodeStr, !strconcat(Dt, "16"), v8i16, IntOp, OpNode>;
  def v4i32 : N3VQIntOp<op24, op23, 0b10, op11_8, op4, itinQ,
                        OpcodeStr, !strconcat(Dt, "32"), v4i32, IntOp, OpNode>;
}

// Neon 3-argument intrinsics,
//   element sizes of 8, 16 and 32 bits:
multiclass N3VInt3_QHS<bit op24, bit op23, bits<4> op11_8, bit op4,
                       InstrItinClass itinD, InstrItinClass itinQ,
                       string OpcodeStr, string Dt, Intrinsic IntOp> {
  // 64-bit vector types.
  def v8i8  : N3VDInt3<op24, op23, 0b00, op11_8, op4, itinD,
                       OpcodeStr, !strconcat(Dt, "8"), v8i8, v8i8, IntOp>;
  def v4i16 : N3VDInt3<op24, op23, 0b01, op11_8, op4, itinD,
                       OpcodeStr, !strconcat(Dt, "16"), v4i16, v4i16, IntOp>;
  def v2i32 : N3VDInt3<op24, op23, 0b10, op11_8, op4, itinD,
                       OpcodeStr, !strconcat(Dt, "32"), v2i32, v2i32, IntOp>;

  // 128-bit vector types.
  def v16i8 : N3VQInt3<op24, op23, 0b00, op11_8, op4, itinQ,
                       OpcodeStr, !strconcat(Dt, "8"), v16i8, v16i8, IntOp>;
  def v8i16 : N3VQInt3<op24, op23, 0b01, op11_8, op4, itinQ,
                       OpcodeStr, !strconcat(Dt, "16"), v8i16, v8i16, IntOp>;
  def v4i32 : N3VQInt3<op24, op23, 0b10, op11_8, op4, itinQ,
                       OpcodeStr, !strconcat(Dt, "32"), v4i32, v4i32, IntOp>;
}


// Neon Long Multiply-Op vector operations,
//   element sizes of 8, 16 and 32 bits:
multiclass N3VLMulOp_QHS<bit op24, bit op23, bits<4> op11_8, bit op4,
                         InstrItinClass itin16, InstrItinClass itin32,
                         string OpcodeStr, string Dt, SDNode MulOp,
                         SDNode OpNode> {
  def v8i16 : N3VLMulOp<op24, op23, 0b00, op11_8, op4, itin16, OpcodeStr,
                        !strconcat(Dt, "8"), v8i16, v8i8, MulOp, OpNode>;
  def v4i32 : N3VLMulOp<op24, op23, 0b01, op11_8, op4, itin16, OpcodeStr,
                        !strconcat(Dt, "16"), v4i32, v4i16, MulOp, OpNode>;
  def v2i64 : N3VLMulOp<op24, op23, 0b10, op11_8, op4, itin32, OpcodeStr,
                        !strconcat(Dt, "32"), v2i64, v2i32, MulOp, OpNode>;
}

multiclass N3VLMulOpSL_HS<bit op24, bits<4> op11_8, string OpcodeStr,
                          string Dt, SDNode MulOp, SDNode OpNode> {
  def v4i16 : N3VLMulOpSL16<op24, 0b01, op11_8, IIC_VMACi16D, OpcodeStr,
                            !strconcat(Dt,"16"), v4i32, v4i16, MulOp, OpNode>;
  def v2i32 : N3VLMulOpSL<op24, 0b10, op11_8, IIC_VMACi32D, OpcodeStr,
                          !strconcat(Dt, "32"), v2i64, v2i32, MulOp, OpNode>;
}


// Neon Long 3-argument intrinsics.

// First with only element sizes of 16 and 32 bits:
multiclass N3VLInt3_HS<bit op24, bit op23, bits<4> op11_8, bit op4,
                       InstrItinClass itin16, InstrItinClass itin32,
                       string OpcodeStr, string Dt, Intrinsic IntOp> {
  def v4i32 : N3VLInt3<op24, op23, 0b01, op11_8, op4, itin16,
                       OpcodeStr, !strconcat(Dt, "16"), v4i32, v4i16, IntOp>;
  def v2i64 : N3VLInt3<op24, op23, 0b10, op11_8, op4, itin32,
                       OpcodeStr, !strconcat(Dt, "32"), v2i64, v2i32, IntOp>;
}

multiclass N3VLInt3SL_HS<bit op24, bits<4> op11_8,
                         string OpcodeStr, string Dt, Intrinsic IntOp> {
  def v4i16 : N3VLInt3SL16<op24, 0b01, op11_8, IIC_VMACi16D,
                           OpcodeStr, !strconcat(Dt,"16"), v4i32, v4i16, IntOp>;
  def v2i32 : N3VLInt3SL<op24, 0b10, op11_8, IIC_VMACi32D,
                         OpcodeStr, !strconcat(Dt, "32"), v2i64, v2i32, IntOp>;
}

// ....then also with element size of 8 bits:
multiclass N3VLInt3_QHS<bit op24, bit op23, bits<4> op11_8, bit op4,
                        InstrItinClass itin16, InstrItinClass itin32,
                        string OpcodeStr, string Dt, Intrinsic IntOp>
  : N3VLInt3_HS<op24, op23, op11_8, op4, itin16, itin32, OpcodeStr, Dt, IntOp> {
  def v8i16 : N3VLInt3<op24, op23, 0b00, op11_8, op4, itin16,
                       OpcodeStr, !strconcat(Dt, "8"), v8i16, v8i8, IntOp>;
}

// ....with explicit extend (VABAL).
multiclass N3VLIntExtOp_QHS<bit op24, bit op23, bits<4> op11_8, bit op4,
                            InstrItinClass itin, string OpcodeStr, string Dt,
                            Intrinsic IntOp, SDNode ExtOp, SDNode OpNode> {
  def v8i16 : N3VLIntExtOp<op24, op23, 0b00, op11_8, op4, itin,
                           OpcodeStr, !strconcat(Dt, "8"), v8i16, v8i8,
                           IntOp, ExtOp, OpNode>;
  def v4i32 : N3VLIntExtOp<op24, op23, 0b01, op11_8, op4, itin,
                           OpcodeStr, !strconcat(Dt, "16"), v4i32, v4i16,
                           IntOp, ExtOp, OpNode>;
  def v2i64 : N3VLIntExtOp<op24, op23, 0b10, op11_8, op4, itin,
                           OpcodeStr, !strconcat(Dt, "32"), v2i64, v2i32,
                           IntOp, ExtOp, OpNode>;
}


// Neon 2-register vector intrinsics,
//   element sizes of 8, 16 and 32 bits:
multiclass N2VInt_QHS<bits<2> op24_23, bits<2> op21_20, bits<2> op17_16,
                      bits<5> op11_7, bit op4,
                      InstrItinClass itinD, InstrItinClass itinQ,
                      string OpcodeStr, string Dt, Intrinsic IntOp> {
  // 64-bit vector types.
  def v8i8  : N2VDInt<op24_23, op21_20, 0b00, op17_16, op11_7, op4,
                      itinD, OpcodeStr, !strconcat(Dt, "8"), v8i8, v8i8, IntOp>;
  def v4i16 : N2VDInt<op24_23, op21_20, 0b01, op17_16, op11_7, op4,
                      itinD, OpcodeStr, !strconcat(Dt, "16"),v4i16,v4i16,IntOp>;
  def v2i32 : N2VDInt<op24_23, op21_20, 0b10, op17_16, op11_7, op4,
                      itinD, OpcodeStr, !strconcat(Dt, "32"),v2i32,v2i32,IntOp>;

  // 128-bit vector types.
  def v16i8 : N2VQInt<op24_23, op21_20, 0b00, op17_16, op11_7, op4,
                      itinQ, OpcodeStr, !strconcat(Dt, "8"), v16i8,v16i8,IntOp>;
  def v8i16 : N2VQInt<op24_23, op21_20, 0b01, op17_16, op11_7, op4,
                      itinQ, OpcodeStr, !strconcat(Dt, "16"),v8i16,v8i16,IntOp>;
  def v4i32 : N2VQInt<op24_23, op21_20, 0b10, op17_16, op11_7, op4,
                      itinQ, OpcodeStr, !strconcat(Dt, "32"),v4i32,v4i32,IntOp>;
}


// Neon Pairwise long 2-register intrinsics,
//   element sizes of 8, 16 and 32 bits:
multiclass N2VPLInt_QHS<bits<2> op24_23, bits<2> op21_20, bits<2> op17_16,
                        bits<5> op11_7, bit op4,
                        string OpcodeStr, string Dt, Intrinsic IntOp> {
  // 64-bit vector types.
  def v8i8  : N2VDPLInt<op24_23, op21_20, 0b00, op17_16, op11_7, op4,
                        OpcodeStr, !strconcat(Dt, "8"), v4i16, v8i8, IntOp>;
  def v4i16 : N2VDPLInt<op24_23, op21_20, 0b01, op17_16, op11_7, op4,
                        OpcodeStr, !strconcat(Dt, "16"), v2i32, v4i16, IntOp>;
  def v2i32 : N2VDPLInt<op24_23, op21_20, 0b10, op17_16, op11_7, op4,
                        OpcodeStr, !strconcat(Dt, "32"), v1i64, v2i32, IntOp>;

  // 128-bit vector types.
  def v16i8 : N2VQPLInt<op24_23, op21_20, 0b00, op17_16, op11_7, op4,
                        OpcodeStr, !strconcat(Dt, "8"), v8i16, v16i8, IntOp>;
  def v8i16 : N2VQPLInt<op24_23, op21_20, 0b01, op17_16, op11_7, op4,
                        OpcodeStr, !strconcat(Dt, "16"), v4i32, v8i16, IntOp>;
  def v4i32 : N2VQPLInt<op24_23, op21_20, 0b10, op17_16, op11_7, op4,
                        OpcodeStr, !strconcat(Dt, "32"), v2i64, v4i32, IntOp>;
}


// Neon Pairwise long 2-register accumulate intrinsics,
//   element sizes of 8, 16 and 32 bits:
multiclass N2VPLInt2_QHS<bits<2> op24_23, bits<2> op21_20, bits<2> op17_16,
                         bits<5> op11_7, bit op4,
                         string OpcodeStr, string Dt, Intrinsic IntOp> {
  // 64-bit vector types.
  def v8i8  : N2VDPLInt2<op24_23, op21_20, 0b00, op17_16, op11_7, op4,
                         OpcodeStr, !strconcat(Dt, "8"), v4i16, v8i8, IntOp>;
  def v4i16 : N2VDPLInt2<op24_23, op21_20, 0b01, op17_16, op11_7, op4,
                         OpcodeStr, !strconcat(Dt, "16"), v2i32, v4i16, IntOp>;
  def v2i32 : N2VDPLInt2<op24_23, op21_20, 0b10, op17_16, op11_7, op4,
                         OpcodeStr, !strconcat(Dt, "32"), v1i64, v2i32, IntOp>;

  // 128-bit vector types.
  def v16i8 : N2VQPLInt2<op24_23, op21_20, 0b00, op17_16, op11_7, op4,
                         OpcodeStr, !strconcat(Dt, "8"), v8i16, v16i8, IntOp>;
  def v8i16 : N2VQPLInt2<op24_23, op21_20, 0b01, op17_16, op11_7, op4,
                         OpcodeStr, !strconcat(Dt, "16"), v4i32, v8i16, IntOp>;
  def v4i32 : N2VQPLInt2<op24_23, op21_20, 0b10, op17_16, op11_7, op4,
                         OpcodeStr, !strconcat(Dt, "32"), v2i64, v4i32, IntOp>;
}


// Neon 2-register vector shift by immediate,
//   with f of either N2RegVShLFrm or N2RegVShRFrm
//   element sizes of 8, 16, 32 and 64 bits:
multiclass N2VSh_QHSD<bit op24, bit op23, bits<4> op11_8, bit op4,
                     InstrItinClass itin, string OpcodeStr, string Dt,
                     SDNode OpNode, Format f> {
  // 64-bit vector types.
  def v8i8  : N2VDSh<op24, op23, op11_8, 0, op4, f, itin,
                     OpcodeStr, !strconcat(Dt, "8"), v8i8, OpNode> {
    let Inst{21-19} = 0b001; // imm6 = 001xxx
  }
  def v4i16 : N2VDSh<op24, op23, op11_8, 0, op4, f, itin,
                     OpcodeStr, !strconcat(Dt, "16"), v4i16, OpNode> {
    let Inst{21-20} = 0b01;  // imm6 = 01xxxx
  }
  def v2i32 : N2VDSh<op24, op23, op11_8, 0, op4, f, itin,
                     OpcodeStr, !strconcat(Dt, "32"), v2i32, OpNode> {
    let Inst{21} = 0b1;      // imm6 = 1xxxxx
  }
  def v1i64 : N2VDSh<op24, op23, op11_8, 1, op4, f, itin,
                     OpcodeStr, !strconcat(Dt, "64"), v1i64, OpNode>;
                             // imm6 = xxxxxx

  // 128-bit vector types.
  def v16i8 : N2VQSh<op24, op23, op11_8, 0, op4, f, itin,
                     OpcodeStr, !strconcat(Dt, "8"), v16i8, OpNode> {
    let Inst{21-19} = 0b001; // imm6 = 001xxx
  }
  def v8i16 : N2VQSh<op24, op23, op11_8, 0, op4, f, itin,
                     OpcodeStr, !strconcat(Dt, "16"), v8i16, OpNode> {
    let Inst{21-20} = 0b01;  // imm6 = 01xxxx
  }
  def v4i32 : N2VQSh<op24, op23, op11_8, 0, op4, f, itin,
                     OpcodeStr, !strconcat(Dt, "32"), v4i32, OpNode> {
    let Inst{21} = 0b1;      // imm6 = 1xxxxx
  }
  def v2i64 : N2VQSh<op24, op23, op11_8, 1, op4, f, itin,
                     OpcodeStr, !strconcat(Dt, "64"), v2i64, OpNode>;
                             // imm6 = xxxxxx
}

// Neon Shift-Accumulate vector operations,
//   element sizes of 8, 16, 32 and 64 bits:
multiclass N2VShAdd_QHSD<bit op24, bit op23, bits<4> op11_8, bit op4,
                         string OpcodeStr, string Dt, SDNode ShOp> {
  // 64-bit vector types.
  def v8i8  : N2VDShAdd<op24, op23, op11_8, 0, op4,
                        OpcodeStr, !strconcat(Dt, "8"), v8i8, ShOp> {
    let Inst{21-19} = 0b001; // imm6 = 001xxx
  }
  def v4i16 : N2VDShAdd<op24, op23, op11_8, 0, op4,
                        OpcodeStr, !strconcat(Dt, "16"), v4i16, ShOp> {
    let Inst{21-20} = 0b01;  // imm6 = 01xxxx
  }
  def v2i32 : N2VDShAdd<op24, op23, op11_8, 0, op4,
                        OpcodeStr, !strconcat(Dt, "32"), v2i32, ShOp> {
    let Inst{21} = 0b1;      // imm6 = 1xxxxx
  }
  def v1i64 : N2VDShAdd<op24, op23, op11_8, 1, op4,
                        OpcodeStr, !strconcat(Dt, "64"), v1i64, ShOp>;
                             // imm6 = xxxxxx

  // 128-bit vector types.
  def v16i8 : N2VQShAdd<op24, op23, op11_8, 0, op4,
                        OpcodeStr, !strconcat(Dt, "8"), v16i8, ShOp> {
    let Inst{21-19} = 0b001; // imm6 = 001xxx
  }
  def v8i16 : N2VQShAdd<op24, op23, op11_8, 0, op4,
                        OpcodeStr, !strconcat(Dt, "16"), v8i16, ShOp> {
    let Inst{21-20} = 0b01;  // imm6 = 01xxxx
  }
  def v4i32 : N2VQShAdd<op24, op23, op11_8, 0, op4,
                        OpcodeStr, !strconcat(Dt, "32"), v4i32, ShOp> {
    let Inst{21} = 0b1;      // imm6 = 1xxxxx
  }
  def v2i64 : N2VQShAdd<op24, op23, op11_8, 1, op4,
                        OpcodeStr, !strconcat(Dt, "64"), v2i64, ShOp>;
                             // imm6 = xxxxxx
}


// Neon Shift-Insert vector operations,
//   with f of either N2RegVShLFrm or N2RegVShRFrm
//   element sizes of 8, 16, 32 and 64 bits:
multiclass N2VShIns_QHSD<bit op24, bit op23, bits<4> op11_8, bit op4,
                         string OpcodeStr, SDNode ShOp,
                         Format f> {
  // 64-bit vector types.
  def v8i8  : N2VDShIns<op24, op23, op11_8, 0, op4,
                        f, OpcodeStr, "8", v8i8, ShOp> {
    let Inst{21-19} = 0b001; // imm6 = 001xxx
  }
  def v4i16 : N2VDShIns<op24, op23, op11_8, 0, op4,
                        f, OpcodeStr, "16", v4i16, ShOp> {
    let Inst{21-20} = 0b01;  // imm6 = 01xxxx
  }
  def v2i32 : N2VDShIns<op24, op23, op11_8, 0, op4,
                        f, OpcodeStr, "32", v2i32, ShOp> {
    let Inst{21} = 0b1;      // imm6 = 1xxxxx
  }
  def v1i64 : N2VDShIns<op24, op23, op11_8, 1, op4,
                        f, OpcodeStr, "64", v1i64, ShOp>;
                             // imm6 = xxxxxx

  // 128-bit vector types.
  def v16i8 : N2VQShIns<op24, op23, op11_8, 0, op4,
                        f, OpcodeStr, "8", v16i8, ShOp> {
    let Inst{21-19} = 0b001; // imm6 = 001xxx
  }
  def v8i16 : N2VQShIns<op24, op23, op11_8, 0, op4,
                        f, OpcodeStr, "16", v8i16, ShOp> {
    let Inst{21-20} = 0b01;  // imm6 = 01xxxx
  }
  def v4i32 : N2VQShIns<op24, op23, op11_8, 0, op4,
                        f, OpcodeStr, "32", v4i32, ShOp> {
    let Inst{21} = 0b1;      // imm6 = 1xxxxx
  }
  def v2i64 : N2VQShIns<op24, op23, op11_8, 1, op4,
                        f, OpcodeStr, "64", v2i64, ShOp>;
                             // imm6 = xxxxxx
}

// Neon Shift Long operations,
//   element sizes of 8, 16, 32 bits:
multiclass N2VLSh_QHS<bit op24, bit op23, bits<4> op11_8, bit op7, bit op6,
                      bit op4, string OpcodeStr, string Dt, SDNode OpNode> {
  def v8i16 : N2VLSh<op24, op23, op11_8, op7, op6, op4,
                 OpcodeStr, !strconcat(Dt, "8"), v8i16, v8i8, OpNode> {
    let Inst{21-19} = 0b001; // imm6 = 001xxx
  }
  def v4i32 : N2VLSh<op24, op23, op11_8, op7, op6, op4,
                  OpcodeStr, !strconcat(Dt, "16"), v4i32, v4i16, OpNode> {
    let Inst{21-20} = 0b01;  // imm6 = 01xxxx
  }
  def v2i64 : N2VLSh<op24, op23, op11_8, op7, op6, op4,
                  OpcodeStr, !strconcat(Dt, "32"), v2i64, v2i32, OpNode> {
    let Inst{21} = 0b1;      // imm6 = 1xxxxx
  }
}

// Neon Shift Narrow operations,
//   element sizes of 16, 32, 64 bits:
multiclass N2VNSh_HSD<bit op24, bit op23, bits<4> op11_8, bit op7, bit op6,
                      bit op4, InstrItinClass itin, string OpcodeStr, string Dt,
                      SDNode OpNode> {
  def v8i8 : N2VNSh<op24, op23, op11_8, op7, op6, op4, itin,
                    OpcodeStr, !strconcat(Dt, "16"), v8i8, v8i16, OpNode> {
    let Inst{21-19} = 0b001; // imm6 = 001xxx
  }
  def v4i16 : N2VNSh<op24, op23, op11_8, op7, op6, op4, itin,
                     OpcodeStr, !strconcat(Dt, "32"), v4i16, v4i32, OpNode> {
    let Inst{21-20} = 0b01;  // imm6 = 01xxxx
  }
  def v2i32 : N2VNSh<op24, op23, op11_8, op7, op6, op4, itin,
                     OpcodeStr, !strconcat(Dt, "64"), v2i32, v2i64, OpNode> {
    let Inst{21} = 0b1;      // imm6 = 1xxxxx
  }
}

//===----------------------------------------------------------------------===//
// Instruction Definitions.
//===----------------------------------------------------------------------===//

// Vector Add Operations.

//   VADD     : Vector Add (integer and floating-point)
defm VADD     : N3V_QHSD<0, 0, 0b1000, 0, IIC_VBINiD, IIC_VBINiQ, "vadd", "i",
                         add, 1>;
def  VADDfd   : N3VD<0, 0, 0b00, 0b1101, 0, IIC_VBIND, "vadd", "f32",
                     v2f32, v2f32, fadd, 1>;
def  VADDfq   : N3VQ<0, 0, 0b00, 0b1101, 0, IIC_VBINQ, "vadd", "f32",
                     v4f32, v4f32, fadd, 1>;
//   VADDL    : Vector Add Long (Q = D + D)
defm VADDLs   : N3VLExt_QHS<0,1,0b0000,0, IIC_VSHLiD, IIC_VSHLiD,
                            "vaddl", "s", add, sext, 1>;
defm VADDLu   : N3VLExt_QHS<1,1,0b0000,0, IIC_VSHLiD, IIC_VSHLiD,
                            "vaddl", "u", add, zext, 1>;
//   VADDW    : Vector Add Wide (Q = Q + D)
defm VADDWs   : N3VW_QHS<0,1,0b0001,0, "vaddw", "s", add, sext, 0>;
defm VADDWu   : N3VW_QHS<1,1,0b0001,0, "vaddw", "u", add, zext, 0>;
//   VHADD    : Vector Halving Add
defm VHADDs   : N3VInt_QHS<0, 0, 0b0000, 0, N3RegFrm,
                           IIC_VBINi4D, IIC_VBINi4D, IIC_VBINi4Q, IIC_VBINi4Q,
                           "vhadd", "s", int_arm_neon_vhadds, 1>;
defm VHADDu   : N3VInt_QHS<1, 0, 0b0000, 0, N3RegFrm,
                           IIC_VBINi4D, IIC_VBINi4D, IIC_VBINi4Q, IIC_VBINi4Q,
                           "vhadd", "u", int_arm_neon_vhaddu, 1>;
//   VRHADD   : Vector Rounding Halving Add
defm VRHADDs  : N3VInt_QHS<0, 0, 0b0001, 0, N3RegFrm,
                           IIC_VBINi4D, IIC_VBINi4D, IIC_VBINi4Q, IIC_VBINi4Q,
                           "vrhadd", "s", int_arm_neon_vrhadds, 1>;
defm VRHADDu  : N3VInt_QHS<1, 0, 0b0001, 0, N3RegFrm,
                           IIC_VBINi4D, IIC_VBINi4D, IIC_VBINi4Q, IIC_VBINi4Q,
                           "vrhadd", "u", int_arm_neon_vrhaddu, 1>;
//   VQADD    : Vector Saturating Add
defm VQADDs   : N3VInt_QHSD<0, 0, 0b0000, 1, N3RegFrm,
                            IIC_VBINi4D, IIC_VBINi4D, IIC_VBINi4Q, IIC_VBINi4Q,
                            "vqadd", "s", int_arm_neon_vqadds, 1>;
defm VQADDu   : N3VInt_QHSD<1, 0, 0b0000, 1, N3RegFrm,
                            IIC_VBINi4D, IIC_VBINi4D, IIC_VBINi4Q, IIC_VBINi4Q,
                            "vqadd", "u", int_arm_neon_vqaddu, 1>;
//   VADDHN   : Vector Add and Narrow Returning High Half (D = Q + Q)
defm VADDHN   : N3VNInt_HSD<0,1,0b0100,0, "vaddhn", "i",
                            int_arm_neon_vaddhn, 1>;
//   VRADDHN  : Vector Rounding Add and Narrow Returning High Half (D = Q + Q)
defm VRADDHN  : N3VNInt_HSD<1,1,0b0100,0, "vraddhn", "i",
                            int_arm_neon_vraddhn, 1>;

// Vector Multiply Operations.

//   VMUL     : Vector Multiply (integer, polynomial and floating-point)
defm VMUL     : N3V_QHS<0, 0, 0b1001, 1, IIC_VMULi16D, IIC_VMULi32D,
                        IIC_VMULi16Q, IIC_VMULi32Q, "vmul", "i", mul, 1>;
def  VMULpd   : N3VDInt<1, 0, 0b00, 0b1001, 1, N3RegFrm, IIC_VMULi16D, "vmul",
                        "p8", v8i8, v8i8, int_arm_neon_vmulp, 1>;
def  VMULpq   : N3VQInt<1, 0, 0b00, 0b1001, 1, N3RegFrm, IIC_VMULi16Q, "vmul",
                        "p8", v16i8, v16i8, int_arm_neon_vmulp, 1>;
def  VMULfd   : N3VD<1, 0, 0b00, 0b1101, 1, IIC_VFMULD, "vmul", "f32",
                     v2f32, v2f32, fmul, 1>;
def  VMULfq   : N3VQ<1, 0, 0b00, 0b1101, 1, IIC_VFMULQ, "vmul", "f32",
                     v4f32, v4f32, fmul, 1>;
defm VMULsl   : N3VSL_HS<0b1000, "vmul", "i", mul>;
def  VMULslfd : N3VDSL<0b10, 0b1001, IIC_VBIND, "vmul", "f32", v2f32, fmul>;
def  VMULslfq : N3VQSL<0b10, 0b1001, IIC_VBINQ, "vmul", "f32", v4f32,
                       v2f32, fmul>;

def : Pat<(v8i16 (mul (v8i16 QPR:$src1),
                      (v8i16 (NEONvduplane (v8i16 QPR:$src2), imm:$lane)))),
          (v8i16 (VMULslv8i16 (v8i16 QPR:$src1),
                              (v4i16 (EXTRACT_SUBREG QPR:$src2,
                                      (DSubReg_i16_reg imm:$lane))),
                              (SubReg_i16_lane imm:$lane)))>;
def : Pat<(v4i32 (mul (v4i32 QPR:$src1),
                      (v4i32 (NEONvduplane (v4i32 QPR:$src2), imm:$lane)))),
          (v4i32 (VMULslv4i32 (v4i32 QPR:$src1),
                              (v2i32 (EXTRACT_SUBREG QPR:$src2,
                                      (DSubReg_i32_reg imm:$lane))),
                              (SubReg_i32_lane imm:$lane)))>;
def : Pat<(v4f32 (fmul (v4f32 QPR:$src1),
                       (v4f32 (NEONvduplane (v4f32 QPR:$src2), imm:$lane)))),
          (v4f32 (VMULslfq (v4f32 QPR:$src1),
                           (v2f32 (EXTRACT_SUBREG QPR:$src2,
                                   (DSubReg_i32_reg imm:$lane))),
                           (SubReg_i32_lane imm:$lane)))>;

//   VQDMULH  : Vector Saturating Doubling Multiply Returning High Half
defm VQDMULH  : N3VInt_HS<0, 0, 0b1011, 0, N3RegFrm, IIC_VMULi16D, IIC_VMULi32D,
                          IIC_VMULi16Q, IIC_VMULi32Q, 
                          "vqdmulh", "s", int_arm_neon_vqdmulh, 1>;
defm VQDMULHsl: N3VIntSL_HS<0b1100, IIC_VMULi16D, IIC_VMULi32D,
                            IIC_VMULi16Q, IIC_VMULi32Q,
                            "vqdmulh", "s",  int_arm_neon_vqdmulh>;
def : Pat<(v8i16 (int_arm_neon_vqdmulh (v8i16 QPR:$src1),
                                       (v8i16 (NEONvduplane (v8i16 QPR:$src2),
                                                            imm:$lane)))),
          (v8i16 (VQDMULHslv8i16 (v8i16 QPR:$src1),
                                 (v4i16 (EXTRACT_SUBREG QPR:$src2,
                                         (DSubReg_i16_reg imm:$lane))),
                                 (SubReg_i16_lane imm:$lane)))>;
def : Pat<(v4i32 (int_arm_neon_vqdmulh (v4i32 QPR:$src1),
                                       (v4i32 (NEONvduplane (v4i32 QPR:$src2),
                                                            imm:$lane)))),
          (v4i32 (VQDMULHslv4i32 (v4i32 QPR:$src1),
                                 (v2i32 (EXTRACT_SUBREG QPR:$src2,
                                         (DSubReg_i32_reg imm:$lane))),
                                 (SubReg_i32_lane imm:$lane)))>;

//   VQRDMULH : Vector Rounding Saturating Doubling Multiply Returning High Half
defm VQRDMULH   : N3VInt_HS<1, 0, 0b1011, 0, N3RegFrm,
                            IIC_VMULi16D,IIC_VMULi32D,IIC_VMULi16Q,IIC_VMULi32Q,
                            "vqrdmulh", "s", int_arm_neon_vqrdmulh, 1>;
defm VQRDMULHsl : N3VIntSL_HS<0b1101, IIC_VMULi16D, IIC_VMULi32D,
                              IIC_VMULi16Q, IIC_VMULi32Q,
                              "vqrdmulh", "s",  int_arm_neon_vqrdmulh>;
def : Pat<(v8i16 (int_arm_neon_vqrdmulh (v8i16 QPR:$src1),
                                        (v8i16 (NEONvduplane (v8i16 QPR:$src2),
                                                             imm:$lane)))),
          (v8i16 (VQRDMULHslv8i16 (v8i16 QPR:$src1),
                                  (v4i16 (EXTRACT_SUBREG QPR:$src2,
                                          (DSubReg_i16_reg imm:$lane))),
                                  (SubReg_i16_lane imm:$lane)))>;
def : Pat<(v4i32 (int_arm_neon_vqrdmulh (v4i32 QPR:$src1),
                                        (v4i32 (NEONvduplane (v4i32 QPR:$src2),
                                                             imm:$lane)))),
          (v4i32 (VQRDMULHslv4i32 (v4i32 QPR:$src1),
                                  (v2i32 (EXTRACT_SUBREG QPR:$src2,
                                          (DSubReg_i32_reg imm:$lane))),
                                  (SubReg_i32_lane imm:$lane)))>;

//   VMULL    : Vector Multiply Long (integer and polynomial) (Q = D * D)
defm VMULLs   : N3VL_QHS<0,1,0b1100,0, IIC_VMULi16D, IIC_VMULi32D,
                         "vmull", "s", NEONvmulls, 1>;
defm VMULLu   : N3VL_QHS<1,1,0b1100,0, IIC_VMULi16D, IIC_VMULi32D,
                         "vmull", "u", NEONvmullu, 1>;
def  VMULLp   : N3VLInt<0, 1, 0b00, 0b1110, 0, IIC_VMULi16D, "vmull", "p8",
                        v8i16, v8i8, int_arm_neon_vmullp, 1>;
defm VMULLsls : N3VLSL_HS<0, 0b1010, IIC_VMULi16D, "vmull", "s", NEONvmulls>;
defm VMULLslu : N3VLSL_HS<1, 0b1010, IIC_VMULi16D, "vmull", "u", NEONvmullu>;

//   VQDMULL  : Vector Saturating Doubling Multiply Long (Q = D * D)
defm VQDMULL  : N3VLInt_HS<0,1,0b1101,0, IIC_VMULi16D, IIC_VMULi32D,
                           "vqdmull", "s", int_arm_neon_vqdmull, 1>;
defm VQDMULLsl: N3VLIntSL_HS<0, 0b1011, IIC_VMULi16D,
                             "vqdmull", "s", int_arm_neon_vqdmull>;

// Vector Multiply-Accumulate and Multiply-Subtract Operations.

//   VMLA     : Vector Multiply Accumulate (integer and floating-point)
defm VMLA     : N3VMulOp_QHS<0, 0, 0b1001, 0, IIC_VMACi16D, IIC_VMACi32D,
                             IIC_VMACi16Q, IIC_VMACi32Q, "vmla", "i", add>;
def  VMLAfd   : N3VDMulOp<0, 0, 0b00, 0b1101, 1, IIC_VMACD, "vmla", "f32",
                          v2f32, fmul, fadd>;
def  VMLAfq   : N3VQMulOp<0, 0, 0b00, 0b1101, 1, IIC_VMACQ, "vmla", "f32",
                          v4f32, fmul, fadd>;
defm VMLAsl   : N3VMulOpSL_HS<0b0000, IIC_VMACi16D, IIC_VMACi32D,
                              IIC_VMACi16Q, IIC_VMACi32Q, "vmla", "i", add>;
def  VMLAslfd : N3VDMulOpSL<0b10, 0b0001, IIC_VMACD, "vmla", "f32",
                            v2f32, fmul, fadd>;
def  VMLAslfq : N3VQMulOpSL<0b10, 0b0001, IIC_VMACQ, "vmla", "f32",
                            v4f32, v2f32, fmul, fadd>;

def : Pat<(v8i16 (add (v8i16 QPR:$src1),
                  (mul (v8i16 QPR:$src2),
                       (v8i16 (NEONvduplane (v8i16 QPR:$src3), imm:$lane))))),
          (v8i16 (VMLAslv8i16 (v8i16 QPR:$src1), (v8i16 QPR:$src2),
                              (v4i16 (EXTRACT_SUBREG QPR:$src3,
                                      (DSubReg_i16_reg imm:$lane))),
                              (SubReg_i16_lane imm:$lane)))>;

def : Pat<(v4i32 (add (v4i32 QPR:$src1),
                  (mul (v4i32 QPR:$src2),
                       (v4i32 (NEONvduplane (v4i32 QPR:$src3), imm:$lane))))),
          (v4i32 (VMLAslv4i32 (v4i32 QPR:$src1), (v4i32 QPR:$src2),
                              (v2i32 (EXTRACT_SUBREG QPR:$src3,
                                      (DSubReg_i32_reg imm:$lane))),
                              (SubReg_i32_lane imm:$lane)))>;

def : Pat<(v4f32 (fadd (v4f32 QPR:$src1),
                  (fmul (v4f32 QPR:$src2),
                        (v4f32 (NEONvduplane (v4f32 QPR:$src3), imm:$lane))))),
          (v4f32 (VMLAslfq (v4f32 QPR:$src1),
                           (v4f32 QPR:$src2),
                           (v2f32 (EXTRACT_SUBREG QPR:$src3,
                                   (DSubReg_i32_reg imm:$lane))),
                           (SubReg_i32_lane imm:$lane)))>;

//   VMLAL    : Vector Multiply Accumulate Long (Q += D * D)
defm VMLALs   : N3VLMulOp_QHS<0,1,0b1000,0, IIC_VMACi16D, IIC_VMACi32D,
                              "vmlal", "s", NEONvmulls, add>;
defm VMLALu   : N3VLMulOp_QHS<1,1,0b1000,0, IIC_VMACi16D, IIC_VMACi32D,
                              "vmlal", "u", NEONvmullu, add>;

defm VMLALsls : N3VLMulOpSL_HS<0, 0b0010, "vmlal", "s", NEONvmulls, add>;
defm VMLALslu : N3VLMulOpSL_HS<1, 0b0010, "vmlal", "u", NEONvmullu, add>;

//   VQDMLAL  : Vector Saturating Doubling Multiply Accumulate Long (Q += D * D)
defm VQDMLAL  : N3VLInt3_HS<0, 1, 0b1001, 0, IIC_VMACi16D, IIC_VMACi32D,
                            "vqdmlal", "s", int_arm_neon_vqdmlal>;
defm VQDMLALsl: N3VLInt3SL_HS<0, 0b0011, "vqdmlal", "s", int_arm_neon_vqdmlal>;

//   VMLS     : Vector Multiply Subtract (integer and floating-point)
defm VMLS     : N3VMulOp_QHS<1, 0, 0b1001, 0, IIC_VMACi16D, IIC_VMACi32D,
                             IIC_VMACi16Q, IIC_VMACi32Q, "vmls", "i", sub>;
def  VMLSfd   : N3VDMulOp<0, 0, 0b10, 0b1101, 1, IIC_VMACD, "vmls", "f32",
                          v2f32, fmul, fsub>;
def  VMLSfq   : N3VQMulOp<0, 0, 0b10, 0b1101, 1, IIC_VMACQ, "vmls", "f32",
                          v4f32, fmul, fsub>;
defm VMLSsl   : N3VMulOpSL_HS<0b0100, IIC_VMACi16D, IIC_VMACi32D,
                              IIC_VMACi16Q, IIC_VMACi32Q, "vmls", "i", sub>;
def  VMLSslfd : N3VDMulOpSL<0b10, 0b0101, IIC_VMACD, "vmls", "f32",
                            v2f32, fmul, fsub>;
def  VMLSslfq : N3VQMulOpSL<0b10, 0b0101, IIC_VMACQ, "vmls", "f32",
                            v4f32, v2f32, fmul, fsub>;

def : Pat<(v8i16 (sub (v8i16 QPR:$src1),
                  (mul (v8i16 QPR:$src2),
                       (v8i16 (NEONvduplane (v8i16 QPR:$src3), imm:$lane))))),
          (v8i16 (VMLSslv8i16 (v8i16 QPR:$src1), (v8i16 QPR:$src2),
                              (v4i16 (EXTRACT_SUBREG QPR:$src3,
                                      (DSubReg_i16_reg imm:$lane))),
                              (SubReg_i16_lane imm:$lane)))>;

def : Pat<(v4i32 (sub (v4i32 QPR:$src1),
                  (mul (v4i32 QPR:$src2),
                     (v4i32 (NEONvduplane (v4i32 QPR:$src3), imm:$lane))))),
          (v4i32 (VMLSslv4i32 (v4i32 QPR:$src1), (v4i32 QPR:$src2),
                              (v2i32 (EXTRACT_SUBREG QPR:$src3,
                                      (DSubReg_i32_reg imm:$lane))),
                              (SubReg_i32_lane imm:$lane)))>;

def : Pat<(v4f32 (fsub (v4f32 QPR:$src1),
                  (fmul (v4f32 QPR:$src2),
                        (v4f32 (NEONvduplane (v4f32 QPR:$src3), imm:$lane))))),
          (v4f32 (VMLSslfq (v4f32 QPR:$src1), (v4f32 QPR:$src2),
                           (v2f32 (EXTRACT_SUBREG QPR:$src3,
                                   (DSubReg_i32_reg imm:$lane))),
                           (SubReg_i32_lane imm:$lane)))>;

//   VMLSL    : Vector Multiply Subtract Long (Q -= D * D)
defm VMLSLs   : N3VLMulOp_QHS<0,1,0b1010,0, IIC_VMACi16D, IIC_VMACi32D,
                              "vmlsl", "s", NEONvmulls, sub>;
defm VMLSLu   : N3VLMulOp_QHS<1,1,0b1010,0, IIC_VMACi16D, IIC_VMACi32D,
                              "vmlsl", "u", NEONvmullu, sub>;

defm VMLSLsls : N3VLMulOpSL_HS<0, 0b0110, "vmlsl", "s", NEONvmulls, sub>;
defm VMLSLslu : N3VLMulOpSL_HS<1, 0b0110, "vmlsl", "u", NEONvmullu, sub>;

//   VQDMLSL  : Vector Saturating Doubling Multiply Subtract Long (Q -= D * D)
defm VQDMLSL  : N3VLInt3_HS<0, 1, 0b1011, 0, IIC_VMACi16D, IIC_VMACi32D,
                            "vqdmlsl", "s", int_arm_neon_vqdmlsl>;
defm VQDMLSLsl: N3VLInt3SL_HS<0, 0b111, "vqdmlsl", "s", int_arm_neon_vqdmlsl>;

// Vector Subtract Operations.

//   VSUB     : Vector Subtract (integer and floating-point)
defm VSUB     : N3V_QHSD<1, 0, 0b1000, 0, IIC_VSUBiD, IIC_VSUBiQ,
                         "vsub", "i", sub, 0>;
def  VSUBfd   : N3VD<0, 0, 0b10, 0b1101, 0, IIC_VBIND, "vsub", "f32",
                     v2f32, v2f32, fsub, 0>;
def  VSUBfq   : N3VQ<0, 0, 0b10, 0b1101, 0, IIC_VBINQ, "vsub", "f32",
                     v4f32, v4f32, fsub, 0>;
//   VSUBL    : Vector Subtract Long (Q = D - D)
defm VSUBLs   : N3VLExt_QHS<0,1,0b0010,0, IIC_VSHLiD, IIC_VSHLiD,
                            "vsubl", "s", sub, sext, 0>;
defm VSUBLu   : N3VLExt_QHS<1,1,0b0010,0, IIC_VSHLiD, IIC_VSHLiD,
                            "vsubl", "u", sub, zext, 0>;
//   VSUBW    : Vector Subtract Wide (Q = Q - D)
defm VSUBWs   : N3VW_QHS<0,1,0b0011,0, "vsubw", "s", sub, sext, 0>;
defm VSUBWu   : N3VW_QHS<1,1,0b0011,0, "vsubw", "u", sub, zext, 0>;
//   VHSUB    : Vector Halving Subtract
defm VHSUBs   : N3VInt_QHS<0, 0, 0b0010, 0, N3RegFrm,
                           IIC_VSUBi4D, IIC_VSUBi4D, IIC_VSUBi4Q, IIC_VSUBi4Q,
                           "vhsub", "s", int_arm_neon_vhsubs, 0>;
defm VHSUBu   : N3VInt_QHS<1, 0, 0b0010, 0, N3RegFrm,
                           IIC_VSUBi4D, IIC_VSUBi4D, IIC_VSUBi4Q, IIC_VSUBi4Q,
                           "vhsub", "u", int_arm_neon_vhsubu, 0>;
//   VQSUB    : Vector Saturing Subtract
defm VQSUBs   : N3VInt_QHSD<0, 0, 0b0010, 1, N3RegFrm,
                            IIC_VSUBi4D, IIC_VSUBi4D, IIC_VSUBi4Q, IIC_VSUBi4Q,
                            "vqsub", "s", int_arm_neon_vqsubs, 0>;
defm VQSUBu   : N3VInt_QHSD<1, 0, 0b0010, 1, N3RegFrm,
                            IIC_VSUBi4D, IIC_VSUBi4D, IIC_VSUBi4Q, IIC_VSUBi4Q,
                            "vqsub", "u", int_arm_neon_vqsubu, 0>;
//   VSUBHN   : Vector Subtract and Narrow Returning High Half (D = Q - Q)
defm VSUBHN   : N3VNInt_HSD<0,1,0b0110,0, "vsubhn", "i",
                            int_arm_neon_vsubhn, 0>;
//   VRSUBHN  : Vector Rounding Subtract and Narrow Returning High Half (D=Q-Q)
defm VRSUBHN  : N3VNInt_HSD<1,1,0b0110,0, "vrsubhn", "i",
                            int_arm_neon_vrsubhn, 0>;

// Vector Comparisons.

//   VCEQ     : Vector Compare Equal
defm VCEQ     : N3V_QHS<1, 0, 0b1000, 1, IIC_VSUBi4D, IIC_VSUBi4D, IIC_VSUBi4Q,
                        IIC_VSUBi4Q, "vceq", "i", NEONvceq, 1>;
def  VCEQfd   : N3VD<0,0,0b00,0b1110,0, IIC_VBIND, "vceq", "f32", v2i32, v2f32,
                     NEONvceq, 1>;
def  VCEQfq   : N3VQ<0,0,0b00,0b1110,0, IIC_VBINQ, "vceq", "f32", v4i32, v4f32,
                     NEONvceq, 1>;

defm VCEQz    : N2V_QHS_cmp<0b11, 0b11, 0b01, 0b00010, 0, "vceq", "i",
                            "$dst, $src, #0", NEONvceqz>;

//   VCGE     : Vector Compare Greater Than or Equal
defm VCGEs    : N3V_QHS<0, 0, 0b0011, 1, IIC_VSUBi4D, IIC_VSUBi4D, IIC_VSUBi4Q,
                        IIC_VSUBi4Q, "vcge", "s", NEONvcge, 0>;
defm VCGEu    : N3V_QHS<1, 0, 0b0011, 1, IIC_VSUBi4D, IIC_VSUBi4D, IIC_VSUBi4Q, 
                        IIC_VSUBi4Q, "vcge", "u", NEONvcgeu, 0>;
def  VCGEfd   : N3VD<1,0,0b00,0b1110,0, IIC_VBIND, "vcge", "f32", v2i32, v2f32,
                     NEONvcge, 0>;
def  VCGEfq   : N3VQ<1,0,0b00,0b1110,0, IIC_VBINQ, "vcge", "f32", v4i32, v4f32,
                     NEONvcge, 0>;

defm VCGEz    : N2V_QHS_cmp<0b11, 0b11, 0b01, 0b00001, 0, "vcge", "s",
                            "$dst, $src, #0", NEONvcgez>;
defm VCLEz    : N2V_QHS_cmp<0b11, 0b11, 0b01, 0b00011, 0, "vcle", "s",
                            "$dst, $src, #0", NEONvclez>;

//   VCGT     : Vector Compare Greater Than
defm VCGTs    : N3V_QHS<0, 0, 0b0011, 0, IIC_VSUBi4D, IIC_VSUBi4D, IIC_VSUBi4Q,
                        IIC_VSUBi4Q, "vcgt", "s", NEONvcgt, 0>;
defm VCGTu    : N3V_QHS<1, 0, 0b0011, 0, IIC_VSUBi4D, IIC_VSUBi4D, IIC_VSUBi4Q,
                        IIC_VSUBi4Q, "vcgt", "u", NEONvcgtu, 0>;
def  VCGTfd   : N3VD<1,0,0b10,0b1110,0, IIC_VBIND, "vcgt", "f32", v2i32, v2f32,
                     NEONvcgt, 0>;
def  VCGTfq   : N3VQ<1,0,0b10,0b1110,0, IIC_VBINQ, "vcgt", "f32", v4i32, v4f32,
                     NEONvcgt, 0>;

defm VCGTz    : N2V_QHS_cmp<0b11, 0b11, 0b01, 0b00000, 0, "vcgt", "s",
                            "$dst, $src, #0", NEONvcgtz>;
defm VCLTz    : N2V_QHS_cmp<0b11, 0b11, 0b01, 0b00100, 0, "vclt", "s",
                            "$dst, $src, #0", NEONvcltz>;

//   VACGE    : Vector Absolute Compare Greater Than or Equal (aka VCAGE)
def  VACGEd   : N3VDInt<1, 0, 0b00, 0b1110, 1, N3RegFrm, IIC_VBIND, "vacge",
                        "f32", v2i32, v2f32, int_arm_neon_vacged, 0>;
def  VACGEq   : N3VQInt<1, 0, 0b00, 0b1110, 1, N3RegFrm, IIC_VBINQ, "vacge",
                        "f32", v4i32, v4f32, int_arm_neon_vacgeq, 0>;
//   VACGT    : Vector Absolute Compare Greater Than (aka VCAGT)
def  VACGTd   : N3VDInt<1, 0, 0b10, 0b1110, 1, N3RegFrm, IIC_VBIND, "vacgt",
                        "f32", v2i32, v2f32, int_arm_neon_vacgtd, 0>;
def  VACGTq   : N3VQInt<1, 0, 0b10, 0b1110, 1, N3RegFrm, IIC_VBINQ, "vacgt",
                        "f32", v4i32, v4f32, int_arm_neon_vacgtq, 0>;
//   VTST     : Vector Test Bits
defm VTST     : N3V_QHS<0, 0, 0b1000, 1, IIC_VBINi4D, IIC_VBINi4D, IIC_VBINi4Q, 
                        IIC_VBINi4Q, "vtst", "", NEONvtst, 1>;

// Vector Bitwise Operations.

def vnotd : PatFrag<(ops node:$in),
                    (xor node:$in, (bitconvert (v8i8 NEONimmAllOnesV)))>;
def vnotq : PatFrag<(ops node:$in),
                    (xor node:$in, (bitconvert (v16i8 NEONimmAllOnesV)))>;


//   VAND     : Vector Bitwise AND
def  VANDd    : N3VDX<0, 0, 0b00, 0b0001, 1, IIC_VBINiD, "vand",
                      v2i32, v2i32, and, 1>;
def  VANDq    : N3VQX<0, 0, 0b00, 0b0001, 1, IIC_VBINiQ, "vand",
                      v4i32, v4i32, and, 1>;

//   VEOR     : Vector Bitwise Exclusive OR
def  VEORd    : N3VDX<1, 0, 0b00, 0b0001, 1, IIC_VBINiD, "veor",
                      v2i32, v2i32, xor, 1>;
def  VEORq    : N3VQX<1, 0, 0b00, 0b0001, 1, IIC_VBINiQ, "veor",
                      v4i32, v4i32, xor, 1>;

//   VORR     : Vector Bitwise OR
def  VORRd    : N3VDX<0, 0, 0b10, 0b0001, 1, IIC_VBINiD, "vorr",
                      v2i32, v2i32, or, 1>;
def  VORRq    : N3VQX<0, 0, 0b10, 0b0001, 1, IIC_VBINiQ, "vorr",
                      v4i32, v4i32, or, 1>;

def VORRiv4i16 : N1ModImm<1, 0b000, {1,0,?,1}, 0, 0, 0, 1,
                          (outs DPR:$Vd), (ins nModImm:$SIMM, DPR:$src),
                          IIC_VMOVImm,
                          "vorr", "i16", "$Vd, $SIMM", "$src = $Vd",
                          [(set DPR:$Vd,
                            (v4i16 (NEONvorrImm DPR:$src, timm:$SIMM)))]> {
  let Inst{9} = SIMM{9};
}

def VORRiv2i32 : N1ModImm<1, 0b000, {0,?,?,1}, 0, 0, 0, 1,
                          (outs DPR:$Vd), (ins nModImm:$SIMM, DPR:$src),
                          IIC_VMOVImm,
                          "vorr", "i32", "$Vd, $SIMM", "$src = $Vd",
                          [(set DPR:$Vd,
                            (v2i32 (NEONvorrImm DPR:$src, timm:$SIMM)))]> {
  let Inst{10-9} = SIMM{10-9};
}

def VORRiv8i16 : N1ModImm<1, 0b000, {1,0,?,1}, 0, 1, 0, 1,
                          (outs QPR:$Vd), (ins nModImm:$SIMM, QPR:$src),
                          IIC_VMOVImm,
                          "vorr", "i16", "$Vd, $SIMM", "$src = $Vd",
                          [(set QPR:$Vd,
                            (v8i16 (NEONvorrImm QPR:$src, timm:$SIMM)))]> {
  let Inst{9} = SIMM{9};
}

def VORRiv4i32 : N1ModImm<1, 0b000, {0,?,?,1}, 0, 1, 0, 1,
                          (outs QPR:$Vd), (ins nModImm:$SIMM, QPR:$src),
                          IIC_VMOVImm,
                          "vorr", "i32", "$Vd, $SIMM", "$src = $Vd",
                          [(set QPR:$Vd,
                            (v4i32 (NEONvorrImm QPR:$src, timm:$SIMM)))]> {
  let Inst{10-9} = SIMM{10-9};
}


//   VBIC     : Vector Bitwise Bit Clear (AND NOT)
def  VBICd    : N3VX<0, 0, 0b01, 0b0001, 0, 1, (outs DPR:$dst),
                     (ins DPR:$src1, DPR:$src2), N3RegFrm, IIC_VBINiD,
                     "vbic", "$dst, $src1, $src2", "",
                     [(set DPR:$dst, (v2i32 (and DPR:$src1,
                                                 (vnotd DPR:$src2))))]>;
def  VBICq    : N3VX<0, 0, 0b01, 0b0001, 1, 1, (outs QPR:$dst),
                     (ins QPR:$src1, QPR:$src2), N3RegFrm, IIC_VBINiQ,
                     "vbic", "$dst, $src1, $src2", "",
                     [(set QPR:$dst, (v4i32 (and QPR:$src1,
                                                 (vnotq QPR:$src2))))]>;

def VBICiv4i16 : N1ModImm<1, 0b000, {1,0,?,1}, 0, 0, 1, 1,
                          (outs DPR:$Vd), (ins nModImm:$SIMM, DPR:$src),
                          IIC_VMOVImm,
                          "vbic", "i16", "$Vd, $SIMM", "$src = $Vd",
                          [(set DPR:$Vd,
                            (v4i16 (NEONvbicImm DPR:$src, timm:$SIMM)))]> {
  let Inst{9} = SIMM{9};
}

def VBICiv2i32 : N1ModImm<1, 0b000, {0,?,?,1}, 0, 0, 1, 1,
                          (outs DPR:$Vd), (ins nModImm:$SIMM, DPR:$src),
                          IIC_VMOVImm,
                          "vbic", "i32", "$Vd, $SIMM", "$src = $Vd",
                          [(set DPR:$Vd,
                            (v2i32 (NEONvbicImm DPR:$src, timm:$SIMM)))]> {
  let Inst{10-9} = SIMM{10-9};
}

def VBICiv8i16 : N1ModImm<1, 0b000, {1,0,?,1}, 0, 1, 1, 1,
                          (outs QPR:$Vd), (ins nModImm:$SIMM, QPR:$src),
                          IIC_VMOVImm,
                          "vbic", "i16", "$Vd, $SIMM", "$src = $Vd",
                          [(set QPR:$Vd,
                            (v8i16 (NEONvbicImm QPR:$src, timm:$SIMM)))]> {
  let Inst{9} = SIMM{9};
}

def VBICiv4i32 : N1ModImm<1, 0b000, {0,?,?,1}, 0, 1, 1, 1,
                          (outs QPR:$Vd), (ins nModImm:$SIMM, QPR:$src),
                          IIC_VMOVImm,
                          "vbic", "i32", "$Vd, $SIMM", "$src = $Vd",
                          [(set QPR:$Vd,
                            (v4i32 (NEONvbicImm QPR:$src, timm:$SIMM)))]> {
  let Inst{10-9} = SIMM{10-9};
}

//   VORN     : Vector Bitwise OR NOT
def  VORNd    : N3VX<0, 0, 0b11, 0b0001, 0, 1, (outs DPR:$dst),
                     (ins DPR:$src1, DPR:$src2), N3RegFrm, IIC_VBINiD,
                     "vorn", "$dst, $src1, $src2", "",
                     [(set DPR:$dst, (v2i32 (or DPR:$src1,
                                                (vnotd DPR:$src2))))]>;
def  VORNq    : N3VX<0, 0, 0b11, 0b0001, 1, 1, (outs QPR:$dst),
                     (ins QPR:$src1, QPR:$src2), N3RegFrm, IIC_VBINiQ,
                     "vorn", "$dst, $src1, $src2", "",
                     [(set QPR:$dst, (v4i32 (or QPR:$src1,
                                                (vnotq QPR:$src2))))]>;

//   VMVN     : Vector Bitwise NOT (Immediate)

let isReMaterializable = 1 in {

def VMVNv4i16 : N1ModImm<1, 0b000, {1,0,?,0}, 0, 0, 1, 1, (outs DPR:$dst),
                         (ins nModImm:$SIMM), IIC_VMOVImm,
                         "vmvn", "i16", "$dst, $SIMM", "",
                         [(set DPR:$dst, (v4i16 (NEONvmvnImm timm:$SIMM)))]> {
  let Inst{9} = SIMM{9};
}

def VMVNv8i16 : N1ModImm<1, 0b000, {1,0,?,0}, 0, 1, 1, 1, (outs QPR:$dst),
                         (ins nModImm:$SIMM), IIC_VMOVImm,
                         "vmvn", "i16", "$dst, $SIMM", "",
                         [(set QPR:$dst, (v8i16 (NEONvmvnImm timm:$SIMM)))]> {
  let Inst{9} = SIMM{9};
}

def VMVNv2i32 : N1ModImm<1, 0b000, {?,?,?,?}, 0, 0, 1, 1, (outs DPR:$dst),
                         (ins nModImm:$SIMM), IIC_VMOVImm,
                         "vmvn", "i32", "$dst, $SIMM", "",
                         [(set DPR:$dst, (v2i32 (NEONvmvnImm timm:$SIMM)))]> {
  let Inst{11-8} = SIMM{11-8};
}

def VMVNv4i32 : N1ModImm<1, 0b000, {?,?,?,?}, 0, 1, 1, 1, (outs QPR:$dst),
                         (ins nModImm:$SIMM), IIC_VMOVImm,
                         "vmvn", "i32", "$dst, $SIMM", "",
                         [(set QPR:$dst, (v4i32 (NEONvmvnImm timm:$SIMM)))]> {
  let Inst{11-8} = SIMM{11-8};
}
}

//   VMVN     : Vector Bitwise NOT
def  VMVNd    : N2VX<0b11, 0b11, 0b00, 0b00, 0b01011, 0, 0,
                     (outs DPR:$dst), (ins DPR:$src), IIC_VSUBiD,
                     "vmvn", "$dst, $src", "",
                     [(set DPR:$dst, (v2i32 (vnotd DPR:$src)))]>;
def  VMVNq    : N2VX<0b11, 0b11, 0b00, 0b00, 0b01011, 1, 0,
                     (outs QPR:$dst), (ins QPR:$src), IIC_VSUBiD,
                     "vmvn", "$dst, $src", "",
                     [(set QPR:$dst, (v4i32 (vnotq QPR:$src)))]>;
def : Pat<(v2i32 (vnotd DPR:$src)), (VMVNd DPR:$src)>;
def : Pat<(v4i32 (vnotq QPR:$src)), (VMVNq QPR:$src)>;

//   VBSL     : Vector Bitwise Select
def  VBSLd    : N3VX<1, 0, 0b01, 0b0001, 0, 1, (outs DPR:$Vd),
                     (ins DPR:$src1, DPR:$Vn, DPR:$Vm),
                     N3RegFrm, IIC_VCNTiD,
                     "vbsl", "$Vd, $Vn, $Vm", "$src1 = $Vd",
                     [(set DPR:$Vd,
                       (v2i32 (or (and DPR:$Vn, DPR:$src1),
                                  (and DPR:$Vm, (vnotd DPR:$src1)))))]>;
def  VBSLq    : N3VX<1, 0, 0b01, 0b0001, 1, 1, (outs QPR:$Vd),
                     (ins QPR:$src1, QPR:$Vn, QPR:$Vm),
                     N3RegFrm, IIC_VCNTiQ,
                     "vbsl", "$Vd, $Vn, $Vm", "$src1 = $Vd",
                     [(set QPR:$Vd,
                       (v4i32 (or (and QPR:$Vn, QPR:$src1),
                                  (and QPR:$Vm, (vnotq QPR:$src1)))))]>;

//   VBIF     : Vector Bitwise Insert if False
//              like VBSL but with: "vbif $dst, $src3, $src1", "$src2 = $dst",
// FIXME: This instruction's encoding MAY NOT BE correct.
def  VBIFd    : N3VX<1, 0, 0b11, 0b0001, 0, 1,
                     (outs DPR:$Vd), (ins DPR:$src1, DPR:$Vn, DPR:$Vm),
                     N3RegFrm, IIC_VBINiD,
                     "vbif", "$Vd, $Vn, $Vm", "$src1 = $Vd",
                     [/* For disassembly only; pattern left blank */]>;
def  VBIFq    : N3VX<1, 0, 0b11, 0b0001, 1, 1,
                     (outs QPR:$Vd), (ins QPR:$src1, QPR:$Vn, QPR:$Vm),
                     N3RegFrm, IIC_VBINiQ,
                     "vbif", "$Vd, $Vn, $Vm", "$src1 = $Vd",
                     [/* For disassembly only; pattern left blank */]>;

//   VBIT     : Vector Bitwise Insert if True
//              like VBSL but with: "vbit $dst, $src2, $src1", "$src3 = $dst",
// FIXME: This instruction's encoding MAY NOT BE correct.
def  VBITd    : N3VX<1, 0, 0b10, 0b0001, 0, 1,
                     (outs DPR:$Vd), (ins DPR:$src1, DPR:$Vn, DPR:$Vm),
                     N3RegFrm, IIC_VBINiD,
                     "vbit", "$Vd, $Vn, $Vm", "$src1 = $Vd",
                     [/* For disassembly only; pattern left blank */]>;
def  VBITq    : N3VX<1, 0, 0b10, 0b0001, 1, 1,
                     (outs QPR:$Vd), (ins QPR:$src1, QPR:$Vn, QPR:$Vm),
                     N3RegFrm, IIC_VBINiQ,
                     "vbit", "$Vd, $Vn, $Vm", "$src1 = $Vd",
                     [/* For disassembly only; pattern left blank */]>;

// VBIT/VBIF are not yet implemented.  The TwoAddress pass will not go looking
// for equivalent operations with different register constraints; it just
// inserts copies.

// Vector Absolute Differences.

//   VABD     : Vector Absolute Difference
defm VABDs    : N3VInt_QHS<0, 0, 0b0111, 0, N3RegFrm,
                           IIC_VSUBi4D, IIC_VSUBi4D, IIC_VSUBi4Q, IIC_VSUBi4Q,
                           "vabd", "s", int_arm_neon_vabds, 1>;
defm VABDu    : N3VInt_QHS<1, 0, 0b0111, 0, N3RegFrm,
                           IIC_VSUBi4D, IIC_VSUBi4D, IIC_VSUBi4Q, IIC_VSUBi4Q,
                           "vabd", "u", int_arm_neon_vabdu, 1>;
def  VABDfd   : N3VDInt<1, 0, 0b10, 0b1101, 0, N3RegFrm, IIC_VBIND,
                        "vabd", "f32", v2f32, v2f32, int_arm_neon_vabds, 1>;
def  VABDfq   : N3VQInt<1, 0, 0b10, 0b1101, 0, N3RegFrm, IIC_VBINQ,
                        "vabd", "f32", v4f32, v4f32, int_arm_neon_vabds, 1>;

//   VABDL    : Vector Absolute Difference Long (Q = | D - D |)
defm VABDLs   : N3VLIntExt_QHS<0,1,0b0111,0, IIC_VSUBi4Q,
                               "vabdl", "s", int_arm_neon_vabds, zext, 1>;
defm VABDLu   : N3VLIntExt_QHS<1,1,0b0111,0, IIC_VSUBi4Q,
                               "vabdl", "u", int_arm_neon_vabdu, zext, 1>;

//   VABA     : Vector Absolute Difference and Accumulate
defm VABAs    : N3VIntOp_QHS<0,0,0b0111,1, IIC_VABAD, IIC_VABAQ,
                             "vaba", "s", int_arm_neon_vabds, add>;
defm VABAu    : N3VIntOp_QHS<1,0,0b0111,1, IIC_VABAD, IIC_VABAQ,
                             "vaba", "u", int_arm_neon_vabdu, add>;

//   VABAL    : Vector Absolute Difference and Accumulate Long (Q += | D - D |)
defm VABALs   : N3VLIntExtOp_QHS<0,1,0b0101,0, IIC_VABAD,
                                 "vabal", "s", int_arm_neon_vabds, zext, add>;
defm VABALu   : N3VLIntExtOp_QHS<1,1,0b0101,0, IIC_VABAD,
                                 "vabal", "u", int_arm_neon_vabdu, zext, add>;

// Vector Maximum and Minimum.

//   VMAX     : Vector Maximum
defm VMAXs    : N3VInt_QHS<0, 0, 0b0110, 0, N3RegFrm,
                           IIC_VSUBi4D, IIC_VSUBi4D, IIC_VSUBi4Q, IIC_VSUBi4Q,
                           "vmax", "s", int_arm_neon_vmaxs, 1>;
defm VMAXu    : N3VInt_QHS<1, 0, 0b0110, 0, N3RegFrm,
                           IIC_VSUBi4D, IIC_VSUBi4D, IIC_VSUBi4Q, IIC_VSUBi4Q,
                           "vmax", "u", int_arm_neon_vmaxu, 1>;
def  VMAXfd   : N3VDInt<0, 0, 0b00, 0b1111, 0, N3RegFrm, IIC_VBIND,
                        "vmax", "f32",
                        v2f32, v2f32, int_arm_neon_vmaxs, 1>;
def  VMAXfq   : N3VQInt<0, 0, 0b00, 0b1111, 0, N3RegFrm, IIC_VBINQ,
                        "vmax", "f32",
                        v4f32, v4f32, int_arm_neon_vmaxs, 1>;

//   VMIN     : Vector Minimum
defm VMINs    : N3VInt_QHS<0, 0, 0b0110, 1, N3RegFrm,
                           IIC_VSUBi4D, IIC_VSUBi4D, IIC_VSUBi4Q, IIC_VSUBi4Q,
                           "vmin", "s", int_arm_neon_vmins, 1>;
defm VMINu    : N3VInt_QHS<1, 0, 0b0110, 1, N3RegFrm,
                           IIC_VSUBi4D, IIC_VSUBi4D, IIC_VSUBi4Q, IIC_VSUBi4Q,
                           "vmin", "u", int_arm_neon_vminu, 1>;
def  VMINfd   : N3VDInt<0, 0, 0b10, 0b1111, 0, N3RegFrm, IIC_VBIND,
                        "vmin", "f32",
                        v2f32, v2f32, int_arm_neon_vmins, 1>;
def  VMINfq   : N3VQInt<0, 0, 0b10, 0b1111, 0, N3RegFrm, IIC_VBINQ,
                        "vmin", "f32",
                        v4f32, v4f32, int_arm_neon_vmins, 1>;

// Vector Pairwise Operations.

//   VPADD    : Vector Pairwise Add
def  VPADDi8  : N3VDInt<0, 0, 0b00, 0b1011, 1, N3RegFrm, IIC_VSHLiD,
                        "vpadd", "i8",
                        v8i8, v8i8, int_arm_neon_vpadd, 0>;
def  VPADDi16 : N3VDInt<0, 0, 0b01, 0b1011, 1, N3RegFrm, IIC_VSHLiD,
                        "vpadd", "i16",
                        v4i16, v4i16, int_arm_neon_vpadd, 0>;
def  VPADDi32 : N3VDInt<0, 0, 0b10, 0b1011, 1, N3RegFrm, IIC_VSHLiD,
                        "vpadd", "i32",
                        v2i32, v2i32, int_arm_neon_vpadd, 0>;
def  VPADDf   : N3VDInt<1, 0, 0b00, 0b1101, 0, N3RegFrm, 
                        IIC_VPBIND, "vpadd", "f32",
                        v2f32, v2f32, int_arm_neon_vpadd, 0>;

//   VPADDL   : Vector Pairwise Add Long
defm VPADDLs  : N2VPLInt_QHS<0b11, 0b11, 0b00, 0b00100, 0, "vpaddl", "s",
                             int_arm_neon_vpaddls>;
defm VPADDLu  : N2VPLInt_QHS<0b11, 0b11, 0b00, 0b00101, 0, "vpaddl", "u",
                             int_arm_neon_vpaddlu>;

//   VPADAL   : Vector Pairwise Add and Accumulate Long
defm VPADALs  : N2VPLInt2_QHS<0b11, 0b11, 0b00, 0b01100, 0, "vpadal", "s",
                              int_arm_neon_vpadals>;
defm VPADALu  : N2VPLInt2_QHS<0b11, 0b11, 0b00, 0b01101, 0, "vpadal", "u",
                              int_arm_neon_vpadalu>;

//   VPMAX    : Vector Pairwise Maximum
def  VPMAXs8  : N3VDInt<0, 0, 0b00, 0b1010, 0, N3RegFrm, IIC_VSUBi4D, "vpmax",
                        "s8", v8i8, v8i8, int_arm_neon_vpmaxs, 0>;
def  VPMAXs16 : N3VDInt<0, 0, 0b01, 0b1010, 0, N3RegFrm, IIC_VSUBi4D, "vpmax",
                        "s16", v4i16, v4i16, int_arm_neon_vpmaxs, 0>;
def  VPMAXs32 : N3VDInt<0, 0, 0b10, 0b1010, 0, N3RegFrm, IIC_VSUBi4D, "vpmax",
                        "s32", v2i32, v2i32, int_arm_neon_vpmaxs, 0>;
def  VPMAXu8  : N3VDInt<1, 0, 0b00, 0b1010, 0, N3RegFrm, IIC_VSUBi4D, "vpmax",
                        "u8", v8i8, v8i8, int_arm_neon_vpmaxu, 0>;
def  VPMAXu16 : N3VDInt<1, 0, 0b01, 0b1010, 0, N3RegFrm, IIC_VSUBi4D, "vpmax",
                        "u16", v4i16, v4i16, int_arm_neon_vpmaxu, 0>;
def  VPMAXu32 : N3VDInt<1, 0, 0b10, 0b1010, 0, N3RegFrm, IIC_VSUBi4D, "vpmax",
                        "u32", v2i32, v2i32, int_arm_neon_vpmaxu, 0>;
def  VPMAXf   : N3VDInt<1, 0, 0b00, 0b1111, 0, N3RegFrm, IIC_VPBIND, "vpmax",
                        "f32", v2f32, v2f32, int_arm_neon_vpmaxs, 0>;

//   VPMIN    : Vector Pairwise Minimum
def  VPMINs8  : N3VDInt<0, 0, 0b00, 0b1010, 1, N3RegFrm, IIC_VSUBi4D, "vpmin",
                        "s8", v8i8, v8i8, int_arm_neon_vpmins, 0>;
def  VPMINs16 : N3VDInt<0, 0, 0b01, 0b1010, 1, N3RegFrm, IIC_VSUBi4D, "vpmin",
                        "s16", v4i16, v4i16, int_arm_neon_vpmins, 0>;
def  VPMINs32 : N3VDInt<0, 0, 0b10, 0b1010, 1, N3RegFrm, IIC_VSUBi4D, "vpmin",
                        "s32", v2i32, v2i32, int_arm_neon_vpmins, 0>;
def  VPMINu8  : N3VDInt<1, 0, 0b00, 0b1010, 1, N3RegFrm, IIC_VSUBi4D, "vpmin",
                        "u8", v8i8, v8i8, int_arm_neon_vpminu, 0>;
def  VPMINu16 : N3VDInt<1, 0, 0b01, 0b1010, 1, N3RegFrm, IIC_VSUBi4D, "vpmin",
                        "u16", v4i16, v4i16, int_arm_neon_vpminu, 0>;
def  VPMINu32 : N3VDInt<1, 0, 0b10, 0b1010, 1, N3RegFrm, IIC_VSUBi4D, "vpmin",
                        "u32", v2i32, v2i32, int_arm_neon_vpminu, 0>;
def  VPMINf   : N3VDInt<1, 0, 0b10, 0b1111, 0, N3RegFrm, IIC_VPBIND, "vpmin",
                        "f32", v2f32, v2f32, int_arm_neon_vpmins, 0>;

// Vector Reciprocal and Reciprocal Square Root Estimate and Step.

//   VRECPE   : Vector Reciprocal Estimate
def  VRECPEd  : N2VDInt<0b11, 0b11, 0b10, 0b11, 0b01000, 0, 
                        IIC_VUNAD, "vrecpe", "u32",
                        v2i32, v2i32, int_arm_neon_vrecpe>;
def  VRECPEq  : N2VQInt<0b11, 0b11, 0b10, 0b11, 0b01000, 0, 
                        IIC_VUNAQ, "vrecpe", "u32",
                        v4i32, v4i32, int_arm_neon_vrecpe>;
def  VRECPEfd : N2VDInt<0b11, 0b11, 0b10, 0b11, 0b01010, 0,
                        IIC_VUNAD, "vrecpe", "f32",
                        v2f32, v2f32, int_arm_neon_vrecpe>;
def  VRECPEfq : N2VQInt<0b11, 0b11, 0b10, 0b11, 0b01010, 0,
                        IIC_VUNAQ, "vrecpe", "f32",
                        v4f32, v4f32, int_arm_neon_vrecpe>;

//   VRECPS   : Vector Reciprocal Step
def  VRECPSfd : N3VDInt<0, 0, 0b00, 0b1111, 1, N3RegFrm,
                        IIC_VRECSD, "vrecps", "f32",
                        v2f32, v2f32, int_arm_neon_vrecps, 1>;
def  VRECPSfq : N3VQInt<0, 0, 0b00, 0b1111, 1, N3RegFrm,
                        IIC_VRECSQ, "vrecps", "f32",
                        v4f32, v4f32, int_arm_neon_vrecps, 1>;

//   VRSQRTE  : Vector Reciprocal Square Root Estimate
def  VRSQRTEd  : N2VDInt<0b11, 0b11, 0b10, 0b11, 0b01001, 0,
                         IIC_VUNAD, "vrsqrte", "u32",
                         v2i32, v2i32, int_arm_neon_vrsqrte>;
def  VRSQRTEq  : N2VQInt<0b11, 0b11, 0b10, 0b11, 0b01001, 0,
                         IIC_VUNAQ, "vrsqrte", "u32",
                         v4i32, v4i32, int_arm_neon_vrsqrte>;
def  VRSQRTEfd : N2VDInt<0b11, 0b11, 0b10, 0b11, 0b01011, 0,
                         IIC_VUNAD, "vrsqrte", "f32",
                         v2f32, v2f32, int_arm_neon_vrsqrte>;
def  VRSQRTEfq : N2VQInt<0b11, 0b11, 0b10, 0b11, 0b01011, 0, 
                         IIC_VUNAQ, "vrsqrte", "f32",
                         v4f32, v4f32, int_arm_neon_vrsqrte>;

//   VRSQRTS  : Vector Reciprocal Square Root Step
def VRSQRTSfd : N3VDInt<0, 0, 0b10, 0b1111, 1, N3RegFrm,
                        IIC_VRECSD, "vrsqrts", "f32",
                        v2f32, v2f32, int_arm_neon_vrsqrts, 1>;
def VRSQRTSfq : N3VQInt<0, 0, 0b10, 0b1111, 1, N3RegFrm,
                        IIC_VRECSQ, "vrsqrts", "f32",
                        v4f32, v4f32, int_arm_neon_vrsqrts, 1>;

// Vector Shifts.

//   VSHL     : Vector Shift
defm VSHLs    : N3VInt_QHSDSh<0, 0, 0b0100, 0, N3RegVShFrm,
                            IIC_VSHLiD, IIC_VSHLiD, IIC_VSHLiQ, IIC_VSHLiQ,
                            "vshl", "s", int_arm_neon_vshifts>;
defm VSHLu    : N3VInt_QHSDSh<1, 0, 0b0100, 0, N3RegVShFrm,
                            IIC_VSHLiD, IIC_VSHLiD, IIC_VSHLiQ, IIC_VSHLiQ,
                            "vshl", "u", int_arm_neon_vshiftu>;
//   VSHL     : Vector Shift Left (Immediate)
defm VSHLi    : N2VSh_QHSD<0, 1, 0b0101, 1, IIC_VSHLiD, "vshl", "i", NEONvshl,
                           N2RegVShLFrm>;
//   VSHR     : Vector Shift Right (Immediate)
defm VSHRs    : N2VSh_QHSD<0, 1, 0b0000, 1, IIC_VSHLiD, "vshr", "s", NEONvshrs,
                           N2RegVShRFrm>;
defm VSHRu    : N2VSh_QHSD<1, 1, 0b0000, 1, IIC_VSHLiD, "vshr", "u", NEONvshru,
                           N2RegVShRFrm>;

//   VSHLL    : Vector Shift Left Long
defm VSHLLs   : N2VLSh_QHS<0, 1, 0b1010, 0, 0, 1, "vshll", "s", NEONvshlls>;
defm VSHLLu   : N2VLSh_QHS<1, 1, 0b1010, 0, 0, 1, "vshll", "u", NEONvshllu>;

//   VSHLL    : Vector Shift Left Long (with maximum shift count)
class N2VLShMax<bit op24, bit op23, bits<6> op21_16, bits<4> op11_8, bit op7,
                bit op6, bit op4, string OpcodeStr, string Dt, ValueType ResTy,
                ValueType OpTy, SDNode OpNode>
  : N2VLSh<op24, op23, op11_8, op7, op6, op4, OpcodeStr, Dt,
           ResTy, OpTy, OpNode> {
  let Inst{21-16} = op21_16;
}
def  VSHLLi8  : N2VLShMax<1, 1, 0b110010, 0b0011, 0, 0, 0, "vshll", "i8",
                          v8i16, v8i8, NEONvshlli>;
def  VSHLLi16 : N2VLShMax<1, 1, 0b110110, 0b0011, 0, 0, 0, "vshll", "i16",
                          v4i32, v4i16, NEONvshlli>;
def  VSHLLi32 : N2VLShMax<1, 1, 0b111010, 0b0011, 0, 0, 0, "vshll", "i32",
                          v2i64, v2i32, NEONvshlli>;

//   VSHRN    : Vector Shift Right and Narrow
defm VSHRN    : N2VNSh_HSD<0,1,0b1000,0,0,1, IIC_VSHLiD, "vshrn", "i",
                           NEONvshrn>;

//   VRSHL    : Vector Rounding Shift
defm VRSHLs   : N3VInt_QHSDSh<0, 0, 0b0101, 0, N3RegVShFrm,
                            IIC_VSHLi4D, IIC_VSHLi4D, IIC_VSHLi4Q, IIC_VSHLi4Q,
                            "vrshl", "s", int_arm_neon_vrshifts>;
defm VRSHLu   : N3VInt_QHSDSh<1, 0, 0b0101, 0, N3RegVShFrm,
                            IIC_VSHLi4D, IIC_VSHLi4D, IIC_VSHLi4Q, IIC_VSHLi4Q,
                            "vrshl", "u", int_arm_neon_vrshiftu>;
//   VRSHR    : Vector Rounding Shift Right
defm VRSHRs   : N2VSh_QHSD<0,1,0b0010,1, IIC_VSHLi4D, "vrshr", "s", NEONvrshrs,
                           N2RegVShRFrm>;
defm VRSHRu   : N2VSh_QHSD<1,1,0b0010,1, IIC_VSHLi4D, "vrshr", "u", NEONvrshru,
                           N2RegVShRFrm>;

//   VRSHRN   : Vector Rounding Shift Right and Narrow
defm VRSHRN   : N2VNSh_HSD<0, 1, 0b1000, 0, 1, 1, IIC_VSHLi4D, "vrshrn", "i",
                           NEONvrshrn>;

//   VQSHL    : Vector Saturating Shift
defm VQSHLs   : N3VInt_QHSDSh<0, 0, 0b0100, 1, N3RegVShFrm,
                            IIC_VSHLi4D, IIC_VSHLi4D, IIC_VSHLi4Q, IIC_VSHLi4Q,
                            "vqshl", "s", int_arm_neon_vqshifts>;
defm VQSHLu   : N3VInt_QHSDSh<1, 0, 0b0100, 1, N3RegVShFrm,
                            IIC_VSHLi4D, IIC_VSHLi4D, IIC_VSHLi4Q, IIC_VSHLi4Q,
                            "vqshl", "u", int_arm_neon_vqshiftu>;
//   VQSHL    : Vector Saturating Shift Left (Immediate)
defm VQSHLsi  : N2VSh_QHSD<0,1,0b0111,1, IIC_VSHLi4D, "vqshl", "s",NEONvqshls,
                           N2RegVShLFrm>;
defm VQSHLui  : N2VSh_QHSD<1,1,0b0111,1, IIC_VSHLi4D, "vqshl", "u",NEONvqshlu,
                           N2RegVShLFrm>;
//   VQSHLU   : Vector Saturating Shift Left (Immediate, Unsigned)
defm VQSHLsu  : N2VSh_QHSD<1,1,0b0110,1, IIC_VSHLi4D,"vqshlu","s",NEONvqshlsu,
                           N2RegVShLFrm>;

//   VQSHRN   : Vector Saturating Shift Right and Narrow
defm VQSHRNs  : N2VNSh_HSD<0, 1, 0b1001, 0, 0, 1, IIC_VSHLi4D, "vqshrn", "s",
                           NEONvqshrns>;
defm VQSHRNu  : N2VNSh_HSD<1, 1, 0b1001, 0, 0, 1, IIC_VSHLi4D, "vqshrn", "u",
                           NEONvqshrnu>;

//   VQSHRUN  : Vector Saturating Shift Right and Narrow (Unsigned)
defm VQSHRUN  : N2VNSh_HSD<1, 1, 0b1000, 0, 0, 1, IIC_VSHLi4D, "vqshrun", "s",
                           NEONvqshrnsu>;

//   VQRSHL   : Vector Saturating Rounding Shift
defm VQRSHLs  : N3VInt_QHSDSh<0, 0, 0b0101, 1, N3RegVShFrm,
                            IIC_VSHLi4D, IIC_VSHLi4D, IIC_VSHLi4Q, IIC_VSHLi4Q,
                            "vqrshl", "s", int_arm_neon_vqrshifts>;
defm VQRSHLu  : N3VInt_QHSDSh<1, 0, 0b0101, 1, N3RegVShFrm,
                            IIC_VSHLi4D, IIC_VSHLi4D, IIC_VSHLi4Q, IIC_VSHLi4Q,
                            "vqrshl", "u", int_arm_neon_vqrshiftu>;

//   VQRSHRN  : Vector Saturating Rounding Shift Right and Narrow
defm VQRSHRNs : N2VNSh_HSD<0, 1, 0b1001, 0, 1, 1, IIC_VSHLi4D, "vqrshrn", "s",
                           NEONvqrshrns>;
defm VQRSHRNu : N2VNSh_HSD<1, 1, 0b1001, 0, 1, 1, IIC_VSHLi4D, "vqrshrn", "u",
                           NEONvqrshrnu>;

//   VQRSHRUN : Vector Saturating Rounding Shift Right and Narrow (Unsigned)
defm VQRSHRUN : N2VNSh_HSD<1, 1, 0b1000, 0, 1, 1, IIC_VSHLi4D, "vqrshrun", "s",
                           NEONvqrshrnsu>;

//   VSRA     : Vector Shift Right and Accumulate
defm VSRAs    : N2VShAdd_QHSD<0, 1, 0b0001, 1, "vsra", "s", NEONvshrs>;
defm VSRAu    : N2VShAdd_QHSD<1, 1, 0b0001, 1, "vsra", "u", NEONvshru>;
//   VRSRA    : Vector Rounding Shift Right and Accumulate
defm VRSRAs   : N2VShAdd_QHSD<0, 1, 0b0011, 1, "vrsra", "s", NEONvrshrs>;
defm VRSRAu   : N2VShAdd_QHSD<1, 1, 0b0011, 1, "vrsra", "u", NEONvrshru>;

//   VSLI     : Vector Shift Left and Insert
defm VSLI     : N2VShIns_QHSD<1, 1, 0b0101, 1, "vsli", NEONvsli, N2RegVShLFrm>;
//   VSRI     : Vector Shift Right and Insert
defm VSRI     : N2VShIns_QHSD<1, 1, 0b0100, 1, "vsri", NEONvsri, N2RegVShRFrm>;

// Vector Absolute and Saturating Absolute.

//   VABS     : Vector Absolute Value
defm VABS     : N2VInt_QHS<0b11, 0b11, 0b01, 0b00110, 0, 
                           IIC_VUNAiD, IIC_VUNAiQ, "vabs", "s",
                           int_arm_neon_vabs>;
def  VABSfd   : N2VDInt<0b11, 0b11, 0b10, 0b01, 0b01110, 0,
                        IIC_VUNAD, "vabs", "f32",
                        v2f32, v2f32, int_arm_neon_vabs>;
def  VABSfq   : N2VQInt<0b11, 0b11, 0b10, 0b01, 0b01110, 0,
                        IIC_VUNAQ, "vabs", "f32",
                        v4f32, v4f32, int_arm_neon_vabs>;

//   VQABS    : Vector Saturating Absolute Value
defm VQABS    : N2VInt_QHS<0b11, 0b11, 0b00, 0b01110, 0, 
                           IIC_VQUNAiD, IIC_VQUNAiQ, "vqabs", "s",
                           int_arm_neon_vqabs>;

// Vector Negate.

def vnegd  : PatFrag<(ops node:$in),
                     (sub (bitconvert (v2i32 NEONimmAllZerosV)), node:$in)>;
def vnegq  : PatFrag<(ops node:$in),
                     (sub (bitconvert (v4i32 NEONimmAllZerosV)), node:$in)>;

class VNEGD<bits<2> size, string OpcodeStr, string Dt, ValueType Ty>
  : N2V<0b11, 0b11, size, 0b01, 0b00111, 0, 0, (outs DPR:$dst), (ins DPR:$src),
        IIC_VSHLiD, OpcodeStr, Dt, "$dst, $src", "",
        [(set DPR:$dst, (Ty (vnegd DPR:$src)))]>;
class VNEGQ<bits<2> size, string OpcodeStr, string Dt, ValueType Ty>
  : N2V<0b11, 0b11, size, 0b01, 0b00111, 1, 0, (outs QPR:$dst), (ins QPR:$src),
        IIC_VSHLiQ, OpcodeStr, Dt, "$dst, $src", "",
        [(set QPR:$dst, (Ty (vnegq QPR:$src)))]>;

//   VNEG     : Vector Negate (integer)
def  VNEGs8d  : VNEGD<0b00, "vneg", "s8", v8i8>;
def  VNEGs16d : VNEGD<0b01, "vneg", "s16", v4i16>;
def  VNEGs32d : VNEGD<0b10, "vneg", "s32", v2i32>;
def  VNEGs8q  : VNEGQ<0b00, "vneg", "s8", v16i8>;
def  VNEGs16q : VNEGQ<0b01, "vneg", "s16", v8i16>;
def  VNEGs32q : VNEGQ<0b10, "vneg", "s32", v4i32>;

//   VNEG     : Vector Negate (floating-point)
def  VNEGfd   : N2V<0b11, 0b11, 0b10, 0b01, 0b01111, 0, 0,
                    (outs DPR:$dst), (ins DPR:$src), IIC_VUNAD,
                    "vneg", "f32", "$dst, $src", "",
                    [(set DPR:$dst, (v2f32 (fneg DPR:$src)))]>;
def  VNEGf32q : N2V<0b11, 0b11, 0b10, 0b01, 0b01111, 1, 0,
                    (outs QPR:$dst), (ins QPR:$src), IIC_VUNAQ,
                    "vneg", "f32", "$dst, $src", "",
                    [(set QPR:$dst, (v4f32 (fneg QPR:$src)))]>;

def : Pat<(v8i8  (vnegd  DPR:$src)), (VNEGs8d DPR:$src)>;
def : Pat<(v4i16 (vnegd  DPR:$src)), (VNEGs16d DPR:$src)>;
def : Pat<(v2i32 (vnegd  DPR:$src)), (VNEGs32d DPR:$src)>;
def : Pat<(v16i8 (vnegq QPR:$src)), (VNEGs8q QPR:$src)>;
def : Pat<(v8i16 (vnegq QPR:$src)), (VNEGs16q QPR:$src)>;
def : Pat<(v4i32 (vnegq QPR:$src)), (VNEGs32q QPR:$src)>;

//   VQNEG    : Vector Saturating Negate
defm VQNEG    : N2VInt_QHS<0b11, 0b11, 0b00, 0b01111, 0, 
                           IIC_VQUNAiD, IIC_VQUNAiQ, "vqneg", "s",
                           int_arm_neon_vqneg>;

// Vector Bit Counting Operations.

//   VCLS     : Vector Count Leading Sign Bits
defm VCLS     : N2VInt_QHS<0b11, 0b11, 0b00, 0b01000, 0, 
                           IIC_VCNTiD, IIC_VCNTiQ, "vcls", "s",
                           int_arm_neon_vcls>;
//   VCLZ     : Vector Count Leading Zeros
defm VCLZ     : N2VInt_QHS<0b11, 0b11, 0b00, 0b01001, 0, 
                           IIC_VCNTiD, IIC_VCNTiQ, "vclz", "i",
                           int_arm_neon_vclz>;
//   VCNT     : Vector Count One Bits
def  VCNTd    : N2VDInt<0b11, 0b11, 0b00, 0b00, 0b01010, 0, 
                        IIC_VCNTiD, "vcnt", "8",
                        v8i8, v8i8, int_arm_neon_vcnt>;
def  VCNTq    : N2VQInt<0b11, 0b11, 0b00, 0b00, 0b01010, 0,
                        IIC_VCNTiQ, "vcnt", "8",
                        v16i8, v16i8, int_arm_neon_vcnt>;

// Vector Swap -- for disassembly only.
def  VSWPd    : N2VX<0b11, 0b11, 0b00, 0b10, 0b00000, 0, 0,
                     (outs DPR:$dst), (ins DPR:$src), NoItinerary,
                     "vswp", "$dst, $src", "", []>;
def  VSWPq    : N2VX<0b11, 0b11, 0b00, 0b10, 0b00000, 1, 0,
                     (outs QPR:$dst), (ins QPR:$src), NoItinerary,
                     "vswp", "$dst, $src", "", []>;

// Vector Move Operations.

//   VMOV     : Vector Move (Register)

let neverHasSideEffects = 1 in {
def  VMOVDneon: N3VX<0, 0, 0b10, 0b0001, 0, 1, (outs DPR:$dst), (ins DPR:$src),
                     N3RegFrm, IIC_VMOV, "vmov", "$dst, $src", "", []>;
def  VMOVQ    : N3VX<0, 0, 0b10, 0b0001, 1, 1, (outs QPR:$dst), (ins QPR:$src),
                     N3RegFrm, IIC_VMOV, "vmov", "$dst, $src", "", []>;

// Pseudo vector move instructions for QQ and QQQQ registers. This should
// be expanded after register allocation is completed.
def  VMOVQQ   : PseudoInst<(outs QQPR:$dst), (ins QQPR:$src),
                NoItinerary, "", []>;

def  VMOVQQQQ : PseudoInst<(outs QQQQPR:$dst), (ins QQQQPR:$src),
                NoItinerary, "", []>;
} // neverHasSideEffects

//   VMOV     : Vector Move (Immediate)

let isReMaterializable = 1 in {
def VMOVv8i8  : N1ModImm<1, 0b000, 0b1110, 0, 0, 0, 1, (outs DPR:$dst),
                         (ins nModImm:$SIMM), IIC_VMOVImm,
                         "vmov", "i8", "$dst, $SIMM", "",
                         [(set DPR:$dst, (v8i8 (NEONvmovImm timm:$SIMM)))]>;
def VMOVv16i8 : N1ModImm<1, 0b000, 0b1110, 0, 1, 0, 1, (outs QPR:$dst),
                         (ins nModImm:$SIMM), IIC_VMOVImm,
                         "vmov", "i8", "$dst, $SIMM", "",
                         [(set QPR:$dst, (v16i8 (NEONvmovImm timm:$SIMM)))]>;

def VMOVv4i16 : N1ModImm<1, 0b000, {1,0,?,0}, 0, 0, 0, 1, (outs DPR:$dst),
                         (ins nModImm:$SIMM), IIC_VMOVImm,
                         "vmov", "i16", "$dst, $SIMM", "",
                         [(set DPR:$dst, (v4i16 (NEONvmovImm timm:$SIMM)))]> {
  let Inst{9} = SIMM{9};                         
}

def VMOVv8i16 : N1ModImm<1, 0b000, {1,0,?,0}, 0, 1, 0, 1, (outs QPR:$dst),
                         (ins nModImm:$SIMM), IIC_VMOVImm,
                         "vmov", "i16", "$dst, $SIMM", "",
                         [(set QPR:$dst, (v8i16 (NEONvmovImm timm:$SIMM)))]> {
 let Inst{9} = SIMM{9};
}

def VMOVv2i32 : N1ModImm<1, 0b000, {?,?,?,?}, 0, 0, 0, 1, (outs DPR:$dst),
                         (ins nModImm:$SIMM), IIC_VMOVImm,
                         "vmov", "i32", "$dst, $SIMM", "",
                         [(set DPR:$dst, (v2i32 (NEONvmovImm timm:$SIMM)))]> {
  let Inst{11-8} = SIMM{11-8};
}

def VMOVv4i32 : N1ModImm<1, 0b000, {?,?,?,?}, 0, 1, 0, 1, (outs QPR:$dst),
                         (ins nModImm:$SIMM), IIC_VMOVImm,
                         "vmov", "i32", "$dst, $SIMM", "",
                         [(set QPR:$dst, (v4i32 (NEONvmovImm timm:$SIMM)))]> {
  let Inst{11-8} = SIMM{11-8};
}

def VMOVv1i64 : N1ModImm<1, 0b000, 0b1110, 0, 0, 1, 1, (outs DPR:$dst),
                         (ins nModImm:$SIMM), IIC_VMOVImm,
                         "vmov", "i64", "$dst, $SIMM", "",
                         [(set DPR:$dst, (v1i64 (NEONvmovImm timm:$SIMM)))]>;
def VMOVv2i64 : N1ModImm<1, 0b000, 0b1110, 0, 1, 1, 1, (outs QPR:$dst),
                         (ins nModImm:$SIMM), IIC_VMOVImm,
                         "vmov", "i64", "$dst, $SIMM", "",
                         [(set QPR:$dst, (v2i64 (NEONvmovImm timm:$SIMM)))]>;
} // isReMaterializable

//   VMOV     : Vector Get Lane (move scalar to ARM core register)

def VGETLNs8  : NVGetLane<{1,1,1,0,0,1,?,1}, 0b1011, {?,?},
                          (outs GPR:$R), (ins DPR:$V, nohash_imm:$lane),
                          IIC_VMOVSI, "vmov", "s8", "$R, $V[$lane]",
                          [(set GPR:$R, (NEONvgetlanes (v8i8 DPR:$V),
                                           imm:$lane))]> {
  let Inst{21}  = lane{2};
  let Inst{6-5} = lane{1-0};
}
def VGETLNs16 : NVGetLane<{1,1,1,0,0,0,?,1}, 0b1011, {?,1},
                          (outs GPR:$R), (ins DPR:$V, nohash_imm:$lane),
                          IIC_VMOVSI, "vmov", "s16", "$R, $V[$lane]",
                          [(set GPR:$R, (NEONvgetlanes (v4i16 DPR:$V),
                                           imm:$lane))]> {
  let Inst{21} = lane{1};
  let Inst{6}  = lane{0};
}
def VGETLNu8  : NVGetLane<{1,1,1,0,1,1,?,1}, 0b1011, {?,?},
                          (outs GPR:$R), (ins DPR:$V, nohash_imm:$lane),
                          IIC_VMOVSI, "vmov", "u8", "$R, $V[$lane]",
                          [(set GPR:$R, (NEONvgetlaneu (v8i8 DPR:$V),
                                           imm:$lane))]> {
  let Inst{21}  = lane{2};
  let Inst{6-5} = lane{1-0};
}
def VGETLNu16 : NVGetLane<{1,1,1,0,1,0,?,1}, 0b1011, {?,1},
                          (outs GPR:$R), (ins DPR:$V, nohash_imm:$lane),
                          IIC_VMOVSI, "vmov", "u16", "$R, $V[$lane]",
                          [(set GPR:$R, (NEONvgetlaneu (v4i16 DPR:$V),
                                           imm:$lane))]> {
  let Inst{21} = lane{1};
  let Inst{6}  = lane{0};
}
def VGETLNi32 : NVGetLane<{1,1,1,0,0,0,?,1}, 0b1011, 0b00,
                          (outs GPR:$R), (ins DPR:$V, nohash_imm:$lane),
                          IIC_VMOVSI, "vmov", "32", "$R, $V[$lane]",
                          [(set GPR:$R, (extractelt (v2i32 DPR:$V),
                                           imm:$lane))]> {
  let Inst{21} = lane{0};
}
// def VGETLNf32: see FMRDH and FMRDL in ARMInstrVFP.td
def : Pat<(NEONvgetlanes (v16i8 QPR:$src), imm:$lane),
          (VGETLNs8 (v8i8 (EXTRACT_SUBREG QPR:$src,
                           (DSubReg_i8_reg imm:$lane))),
                     (SubReg_i8_lane imm:$lane))>;
def : Pat<(NEONvgetlanes (v8i16 QPR:$src), imm:$lane),
          (VGETLNs16 (v4i16 (EXTRACT_SUBREG QPR:$src,
                             (DSubReg_i16_reg imm:$lane))),
                     (SubReg_i16_lane imm:$lane))>;
def : Pat<(NEONvgetlaneu (v16i8 QPR:$src), imm:$lane),
          (VGETLNu8 (v8i8 (EXTRACT_SUBREG QPR:$src,
                           (DSubReg_i8_reg imm:$lane))),
                     (SubReg_i8_lane imm:$lane))>;
def : Pat<(NEONvgetlaneu (v8i16 QPR:$src), imm:$lane),
          (VGETLNu16 (v4i16 (EXTRACT_SUBREG QPR:$src,
                             (DSubReg_i16_reg imm:$lane))),
                     (SubReg_i16_lane imm:$lane))>;
def : Pat<(extractelt (v4i32 QPR:$src), imm:$lane),
          (VGETLNi32 (v2i32 (EXTRACT_SUBREG QPR:$src,
                             (DSubReg_i32_reg imm:$lane))),
                     (SubReg_i32_lane imm:$lane))>;
def : Pat<(extractelt (v2f32 DPR:$src1), imm:$src2),
          (EXTRACT_SUBREG (v2f32 (COPY_TO_REGCLASS (v2f32 DPR:$src1),DPR_VFP2)),
                          (SSubReg_f32_reg imm:$src2))>;
def : Pat<(extractelt (v4f32 QPR:$src1), imm:$src2),
          (EXTRACT_SUBREG (v4f32 (COPY_TO_REGCLASS (v4f32 QPR:$src1),QPR_VFP2)),
                          (SSubReg_f32_reg imm:$src2))>;
//def : Pat<(extractelt (v2i64 QPR:$src1), imm:$src2),
//          (EXTRACT_SUBREG QPR:$src1, (DSubReg_f64_reg imm:$src2))>;
def : Pat<(extractelt (v2f64 QPR:$src1), imm:$src2),
          (EXTRACT_SUBREG QPR:$src1, (DSubReg_f64_reg imm:$src2))>;


//   VMOV     : Vector Set Lane (move ARM core register to scalar)

let Constraints = "$src1 = $V" in {
def VSETLNi8  : NVSetLane<{1,1,1,0,0,1,?,0}, 0b1011, {?,?}, (outs DPR:$V),
                          (ins DPR:$src1, GPR:$R, nohash_imm:$lane),
                          IIC_VMOVISL, "vmov", "8", "$V[$lane], $R",
                          [(set DPR:$V, (vector_insert (v8i8 DPR:$src1),
                                           GPR:$R, imm:$lane))]> {
  let Inst{21}  = lane{2};
  let Inst{6-5} = lane{1-0};
}
def VSETLNi16 : NVSetLane<{1,1,1,0,0,0,?,0}, 0b1011, {?,1}, (outs DPR:$V),
                          (ins DPR:$src1, GPR:$R, nohash_imm:$lane),
                          IIC_VMOVISL, "vmov", "16", "$V[$lane], $R",
                          [(set DPR:$V, (vector_insert (v4i16 DPR:$src1),
                                           GPR:$R, imm:$lane))]> {
  let Inst{21} = lane{1};
  let Inst{6}  = lane{0};
}
def VSETLNi32 : NVSetLane<{1,1,1,0,0,0,?,0}, 0b1011, 0b00, (outs DPR:$V),
                          (ins DPR:$src1, GPR:$R, nohash_imm:$lane),
                          IIC_VMOVISL, "vmov", "32", "$V[$lane], $R",
                          [(set DPR:$V, (insertelt (v2i32 DPR:$src1),
                                           GPR:$R, imm:$lane))]> {
  let Inst{21} = lane{0};
}
}
def : Pat<(vector_insert (v16i8 QPR:$src1), GPR:$src2, imm:$lane),
          (v16i8 (INSERT_SUBREG QPR:$src1, 
                  (v8i8 (VSETLNi8 (v8i8 (EXTRACT_SUBREG QPR:$src1,
                                   (DSubReg_i8_reg imm:$lane))),
                            GPR:$src2, (SubReg_i8_lane imm:$lane))),
                  (DSubReg_i8_reg imm:$lane)))>;
def : Pat<(vector_insert (v8i16 QPR:$src1), GPR:$src2, imm:$lane),
          (v8i16 (INSERT_SUBREG QPR:$src1, 
                  (v4i16 (VSETLNi16 (v4i16 (EXTRACT_SUBREG QPR:$src1,
                                     (DSubReg_i16_reg imm:$lane))),
                             GPR:$src2, (SubReg_i16_lane imm:$lane))),
                  (DSubReg_i16_reg imm:$lane)))>;
def : Pat<(insertelt (v4i32 QPR:$src1), GPR:$src2, imm:$lane),
          (v4i32 (INSERT_SUBREG QPR:$src1, 
                  (v2i32 (VSETLNi32 (v2i32 (EXTRACT_SUBREG QPR:$src1,
                                     (DSubReg_i32_reg imm:$lane))),
                             GPR:$src2, (SubReg_i32_lane imm:$lane))),
                  (DSubReg_i32_reg imm:$lane)))>;

def : Pat<(v2f32 (insertelt DPR:$src1, SPR:$src2, imm:$src3)),
          (INSERT_SUBREG (v2f32 (COPY_TO_REGCLASS DPR:$src1, DPR_VFP2)),
                                SPR:$src2, (SSubReg_f32_reg imm:$src3))>;
def : Pat<(v4f32 (insertelt QPR:$src1, SPR:$src2, imm:$src3)),
          (INSERT_SUBREG (v4f32 (COPY_TO_REGCLASS QPR:$src1, QPR_VFP2)),
                                SPR:$src2, (SSubReg_f32_reg imm:$src3))>;

//def : Pat<(v2i64 (insertelt QPR:$src1, DPR:$src2, imm:$src3)),
//          (INSERT_SUBREG QPR:$src1, DPR:$src2, (DSubReg_f64_reg imm:$src3))>;
def : Pat<(v2f64 (insertelt QPR:$src1, DPR:$src2, imm:$src3)),
          (INSERT_SUBREG QPR:$src1, DPR:$src2, (DSubReg_f64_reg imm:$src3))>;

def : Pat<(v2f32 (scalar_to_vector SPR:$src)),
          (INSERT_SUBREG (v2f32 (IMPLICIT_DEF)), SPR:$src, ssub_0)>;
def : Pat<(v2f64 (scalar_to_vector (f64 DPR:$src))),
          (INSERT_SUBREG (v2f64 (IMPLICIT_DEF)), DPR:$src, dsub_0)>;
def : Pat<(v4f32 (scalar_to_vector SPR:$src)),
          (INSERT_SUBREG (v4f32 (IMPLICIT_DEF)), SPR:$src, ssub_0)>;

def : Pat<(v8i8 (scalar_to_vector GPR:$src)),
          (VSETLNi8  (v8i8  (IMPLICIT_DEF)), GPR:$src, (i32 0))>;
def : Pat<(v4i16 (scalar_to_vector GPR:$src)),
          (VSETLNi16 (v4i16 (IMPLICIT_DEF)), GPR:$src, (i32 0))>;
def : Pat<(v2i32 (scalar_to_vector GPR:$src)),
          (VSETLNi32 (v2i32 (IMPLICIT_DEF)), GPR:$src, (i32 0))>;

def : Pat<(v16i8 (scalar_to_vector GPR:$src)),
          (INSERT_SUBREG (v16i8 (IMPLICIT_DEF)),
                         (VSETLNi8 (v8i8 (IMPLICIT_DEF)), GPR:$src, (i32 0)),
                         dsub_0)>;
def : Pat<(v8i16 (scalar_to_vector GPR:$src)),
          (INSERT_SUBREG (v8i16 (IMPLICIT_DEF)),
                         (VSETLNi16 (v4i16 (IMPLICIT_DEF)), GPR:$src, (i32 0)),
                         dsub_0)>;
def : Pat<(v4i32 (scalar_to_vector GPR:$src)),
          (INSERT_SUBREG (v4i32 (IMPLICIT_DEF)),
                         (VSETLNi32 (v2i32 (IMPLICIT_DEF)), GPR:$src, (i32 0)),
                         dsub_0)>;

//   VDUP     : Vector Duplicate (from ARM core register to all elements)

class VDUPD<bits<8> opcod1, bits<2> opcod3, string Dt, ValueType Ty>
  : NVDup<opcod1, 0b1011, opcod3, (outs DPR:$dst), (ins GPR:$src),
          IIC_VMOVIS, "vdup", Dt, "$dst, $src",
          [(set DPR:$dst, (Ty (NEONvdup (i32 GPR:$src))))]>;
class VDUPQ<bits<8> opcod1, bits<2> opcod3, string Dt, ValueType Ty>
  : NVDup<opcod1, 0b1011, opcod3, (outs QPR:$dst), (ins GPR:$src),
          IIC_VMOVIS, "vdup", Dt, "$dst, $src",
          [(set QPR:$dst, (Ty (NEONvdup (i32 GPR:$src))))]>;

def  VDUP8d   : VDUPD<0b11101100, 0b00, "8", v8i8>;
def  VDUP16d  : VDUPD<0b11101000, 0b01, "16", v4i16>;
def  VDUP32d  : VDUPD<0b11101000, 0b00, "32", v2i32>;
def  VDUP8q   : VDUPQ<0b11101110, 0b00, "8", v16i8>;
def  VDUP16q  : VDUPQ<0b11101010, 0b01, "16", v8i16>;
def  VDUP32q  : VDUPQ<0b11101010, 0b00, "32", v4i32>;

def  VDUPfd   : NVDup<0b11101000, 0b1011, 0b00, (outs DPR:$dst), (ins GPR:$src),
                      IIC_VMOVIS, "vdup", "32", "$dst, $src",
                      [(set DPR:$dst, (v2f32 (NEONvdup
                                              (f32 (bitconvert GPR:$src)))))]>;
def  VDUPfq   : NVDup<0b11101010, 0b1011, 0b00, (outs QPR:$dst), (ins GPR:$src),
                      IIC_VMOVIS, "vdup", "32", "$dst, $src",
                      [(set QPR:$dst, (v4f32 (NEONvdup
                                              (f32 (bitconvert GPR:$src)))))]>;

//   VDUP     : Vector Duplicate Lane (from scalar to all elements)

class VDUPLND<bits<4> op19_16, string OpcodeStr, string Dt,
              ValueType Ty>
  : NVDupLane<op19_16, 0, (outs DPR:$dst), (ins DPR:$src, nohash_imm:$lane),
              IIC_VMOVD, OpcodeStr, Dt, "$dst, $src[$lane]",
              [(set DPR:$dst, (Ty (NEONvduplane (Ty DPR:$src), imm:$lane)))]>;

class VDUPLNQ<bits<4> op19_16, string OpcodeStr, string Dt,
              ValueType ResTy, ValueType OpTy>
  : NVDupLane<op19_16, 1, (outs QPR:$dst), (ins DPR:$src, nohash_imm:$lane),
              IIC_VMOVQ, OpcodeStr, Dt, "$dst, $src[$lane]",
              [(set QPR:$dst, (ResTy (NEONvduplane (OpTy DPR:$src),
                                      imm:$lane)))]>;

// Inst{19-16} is partially specified depending on the element size.

def VDUPLN8d  : VDUPLND<{?,?,?,1}, "vdup", "8", v8i8> {
  let Inst{19-17} = lane{2-0};
}
def VDUPLN16d : VDUPLND<{?,?,1,0}, "vdup", "16", v4i16> {
  let Inst{19-18} = lane{1-0};
}
def VDUPLN32d : VDUPLND<{?,1,0,0}, "vdup", "32", v2i32> {
  let Inst{19} = lane{0};
}
def VDUPLNfd  : VDUPLND<{?,1,0,0}, "vdup", "32", v2f32> {
  let Inst{19} = lane{0};
}
def VDUPLN8q  : VDUPLNQ<{?,?,?,1}, "vdup", "8", v16i8, v8i8> {
  let Inst{19-17} = lane{2-0};
}
def VDUPLN16q : VDUPLNQ<{?,?,1,0}, "vdup", "16", v8i16, v4i16> {
  let Inst{19-18} = lane{1-0};
}
def VDUPLN32q : VDUPLNQ<{?,1,0,0}, "vdup", "32", v4i32, v2i32> {
  let Inst{19} = lane{0};
}
def VDUPLNfq  : VDUPLNQ<{?,1,0,0}, "vdup", "32", v4f32, v2f32> {
  let Inst{19} = lane{0};
}

def : Pat<(v16i8 (NEONvduplane (v16i8 QPR:$src), imm:$lane)),
          (v16i8 (VDUPLN8q (v8i8 (EXTRACT_SUBREG QPR:$src,
                                  (DSubReg_i8_reg imm:$lane))),
                           (SubReg_i8_lane imm:$lane)))>;
def : Pat<(v8i16 (NEONvduplane (v8i16 QPR:$src), imm:$lane)),
          (v8i16 (VDUPLN16q (v4i16 (EXTRACT_SUBREG QPR:$src,
                                    (DSubReg_i16_reg imm:$lane))),
                            (SubReg_i16_lane imm:$lane)))>;
def : Pat<(v4i32 (NEONvduplane (v4i32 QPR:$src), imm:$lane)),
          (v4i32 (VDUPLN32q (v2i32 (EXTRACT_SUBREG QPR:$src,
                                    (DSubReg_i32_reg imm:$lane))),
                            (SubReg_i32_lane imm:$lane)))>;
def : Pat<(v4f32 (NEONvduplane (v4f32 QPR:$src), imm:$lane)),
          (v4f32 (VDUPLNfq (v2f32 (EXTRACT_SUBREG QPR:$src,
                                   (DSubReg_i32_reg imm:$lane))),
                           (SubReg_i32_lane imm:$lane)))>;

def  VDUPfdf : PseudoNeonI<(outs DPR:$dst), (ins SPR:$src), IIC_VMOVD, "",
                    [(set DPR:$dst, (v2f32 (NEONvdup (f32 SPR:$src))))]>;
def  VDUPfqf : PseudoNeonI<(outs QPR:$dst), (ins SPR:$src), IIC_VMOVD, "",
                    [(set QPR:$dst, (v4f32 (NEONvdup (f32 SPR:$src))))]>;

//   VMOVN    : Vector Narrowing Move
defm VMOVN    : N2VN_HSD<0b11,0b11,0b10,0b00100,0,0, IIC_VMOVN,
                         "vmovn", "i", trunc>;
//   VQMOVN   : Vector Saturating Narrowing Move
defm VQMOVNs  : N2VNInt_HSD<0b11,0b11,0b10,0b00101,0,0, IIC_VQUNAiD,
                            "vqmovn", "s", int_arm_neon_vqmovns>;
defm VQMOVNu  : N2VNInt_HSD<0b11,0b11,0b10,0b00101,1,0, IIC_VQUNAiD,
                            "vqmovn", "u", int_arm_neon_vqmovnu>;
defm VQMOVNsu : N2VNInt_HSD<0b11,0b11,0b10,0b00100,1,0, IIC_VQUNAiD,
                            "vqmovun", "s", int_arm_neon_vqmovnsu>;
//   VMOVL    : Vector Lengthening Move
defm VMOVLs   : N2VL_QHS<0b01,0b10100,0,1, "vmovl", "s", sext>;
defm VMOVLu   : N2VL_QHS<0b11,0b10100,0,1, "vmovl", "u", zext>;

// Vector Conversions.

//   VCVT     : Vector Convert Between Floating-Point and Integers
def  VCVTf2sd : N2VD<0b11, 0b11, 0b10, 0b11, 0b01110, 0, "vcvt", "s32.f32",
                     v2i32, v2f32, fp_to_sint>;
def  VCVTf2ud : N2VD<0b11, 0b11, 0b10, 0b11, 0b01111, 0, "vcvt", "u32.f32",
                     v2i32, v2f32, fp_to_uint>;
def  VCVTs2fd : N2VD<0b11, 0b11, 0b10, 0b11, 0b01100, 0, "vcvt", "f32.s32",
                     v2f32, v2i32, sint_to_fp>;
def  VCVTu2fd : N2VD<0b11, 0b11, 0b10, 0b11, 0b01101, 0, "vcvt", "f32.u32",
                     v2f32, v2i32, uint_to_fp>;

def  VCVTf2sq : N2VQ<0b11, 0b11, 0b10, 0b11, 0b01110, 0, "vcvt", "s32.f32",
                     v4i32, v4f32, fp_to_sint>;
def  VCVTf2uq : N2VQ<0b11, 0b11, 0b10, 0b11, 0b01111, 0, "vcvt", "u32.f32",
                     v4i32, v4f32, fp_to_uint>;
def  VCVTs2fq : N2VQ<0b11, 0b11, 0b10, 0b11, 0b01100, 0, "vcvt", "f32.s32",
                     v4f32, v4i32, sint_to_fp>;
def  VCVTu2fq : N2VQ<0b11, 0b11, 0b10, 0b11, 0b01101, 0, "vcvt", "f32.u32",
                     v4f32, v4i32, uint_to_fp>;

//   VCVT     : Vector Convert Between Floating-Point and Fixed-Point.
def VCVTf2xsd : N2VCvtD<0, 1, 0b1111, 0, 1, "vcvt", "s32.f32",
                        v2i32, v2f32, int_arm_neon_vcvtfp2fxs>;
def VCVTf2xud : N2VCvtD<1, 1, 0b1111, 0, 1, "vcvt", "u32.f32",
                        v2i32, v2f32, int_arm_neon_vcvtfp2fxu>;
def VCVTxs2fd : N2VCvtD<0, 1, 0b1110, 0, 1, "vcvt", "f32.s32",
                        v2f32, v2i32, int_arm_neon_vcvtfxs2fp>;
def VCVTxu2fd : N2VCvtD<1, 1, 0b1110, 0, 1, "vcvt", "f32.u32",
                        v2f32, v2i32, int_arm_neon_vcvtfxu2fp>;

def VCVTf2xsq : N2VCvtQ<0, 1, 0b1111, 0, 1, "vcvt", "s32.f32",
                        v4i32, v4f32, int_arm_neon_vcvtfp2fxs>;
def VCVTf2xuq : N2VCvtQ<1, 1, 0b1111, 0, 1, "vcvt", "u32.f32",
                        v4i32, v4f32, int_arm_neon_vcvtfp2fxu>;
def VCVTxs2fq : N2VCvtQ<0, 1, 0b1110, 0, 1, "vcvt", "f32.s32",
                        v4f32, v4i32, int_arm_neon_vcvtfxs2fp>;
def VCVTxu2fq : N2VCvtQ<1, 1, 0b1110, 0, 1, "vcvt", "f32.u32",
                        v4f32, v4i32, int_arm_neon_vcvtfxu2fp>;

// Vector Reverse.

//   VREV64   : Vector Reverse elements within 64-bit doublewords

class VREV64D<bits<2> op19_18, string OpcodeStr, string Dt, ValueType Ty>
  : N2V<0b11, 0b11, op19_18, 0b00, 0b00000, 0, 0, (outs DPR:$dst),
        (ins DPR:$src), IIC_VMOVD, 
        OpcodeStr, Dt, "$dst, $src", "",
        [(set DPR:$dst, (Ty (NEONvrev64 (Ty DPR:$src))))]>;
class VREV64Q<bits<2> op19_18, string OpcodeStr, string Dt, ValueType Ty>
  : N2V<0b11, 0b11, op19_18, 0b00, 0b00000, 1, 0, (outs QPR:$dst),
        (ins QPR:$src), IIC_VMOVQ, 
        OpcodeStr, Dt, "$dst, $src", "",
        [(set QPR:$dst, (Ty (NEONvrev64 (Ty QPR:$src))))]>;

def VREV64d8  : VREV64D<0b00, "vrev64", "8", v8i8>;
def VREV64d16 : VREV64D<0b01, "vrev64", "16", v4i16>;
def VREV64d32 : VREV64D<0b10, "vrev64", "32", v2i32>;
def VREV64df  : VREV64D<0b10, "vrev64", "32", v2f32>;

def VREV64q8  : VREV64Q<0b00, "vrev64", "8", v16i8>;
def VREV64q16 : VREV64Q<0b01, "vrev64", "16", v8i16>;
def VREV64q32 : VREV64Q<0b10, "vrev64", "32", v4i32>;
def VREV64qf  : VREV64Q<0b10, "vrev64", "32", v4f32>;

//   VREV32   : Vector Reverse elements within 32-bit words

class VREV32D<bits<2> op19_18, string OpcodeStr, string Dt, ValueType Ty>
  : N2V<0b11, 0b11, op19_18, 0b00, 0b00001, 0, 0, (outs DPR:$dst),
        (ins DPR:$src), IIC_VMOVD, 
        OpcodeStr, Dt, "$dst, $src", "",
        [(set DPR:$dst, (Ty (NEONvrev32 (Ty DPR:$src))))]>;
class VREV32Q<bits<2> op19_18, string OpcodeStr, string Dt, ValueType Ty>
  : N2V<0b11, 0b11, op19_18, 0b00, 0b00001, 1, 0, (outs QPR:$dst),
        (ins QPR:$src), IIC_VMOVQ, 
        OpcodeStr, Dt, "$dst, $src", "",
        [(set QPR:$dst, (Ty (NEONvrev32 (Ty QPR:$src))))]>;

def VREV32d8  : VREV32D<0b00, "vrev32", "8", v8i8>;
def VREV32d16 : VREV32D<0b01, "vrev32", "16", v4i16>;

def VREV32q8  : VREV32Q<0b00, "vrev32", "8", v16i8>;
def VREV32q16 : VREV32Q<0b01, "vrev32", "16", v8i16>;

//   VREV16   : Vector Reverse elements within 16-bit halfwords

class VREV16D<bits<2> op19_18, string OpcodeStr, string Dt, ValueType Ty>
  : N2V<0b11, 0b11, op19_18, 0b00, 0b00010, 0, 0, (outs DPR:$dst),
        (ins DPR:$src), IIC_VMOVD, 
        OpcodeStr, Dt, "$dst, $src", "",
        [(set DPR:$dst, (Ty (NEONvrev16 (Ty DPR:$src))))]>;
class VREV16Q<bits<2> op19_18, string OpcodeStr, string Dt, ValueType Ty>
  : N2V<0b11, 0b11, op19_18, 0b00, 0b00010, 1, 0, (outs QPR:$dst),
        (ins QPR:$src), IIC_VMOVQ, 
        OpcodeStr, Dt, "$dst, $src", "",
        [(set QPR:$dst, (Ty (NEONvrev16 (Ty QPR:$src))))]>;

def VREV16d8  : VREV16D<0b00, "vrev16", "8", v8i8>;
def VREV16q8  : VREV16Q<0b00, "vrev16", "8", v16i8>;

// Other Vector Shuffles.

//   VEXT     : Vector Extract

class VEXTd<string OpcodeStr, string Dt, ValueType Ty>
  : N3V<0,1,0b11,{?,?,?,?},0,0, (outs DPR:$dst),
        (ins DPR:$lhs, DPR:$rhs, i32imm:$index), NVExtFrm,
        IIC_VEXTD, OpcodeStr, Dt, "$dst, $lhs, $rhs, $index", "",
        [(set DPR:$dst, (Ty (NEONvext (Ty DPR:$lhs),
                                      (Ty DPR:$rhs), imm:$index)))]> {
  bits<4> index;
  let Inst{11-8} = index{3-0};
}

class VEXTq<string OpcodeStr, string Dt, ValueType Ty>
  : N3V<0,1,0b11,{?,?,?,?},1,0, (outs QPR:$dst),
        (ins QPR:$lhs, QPR:$rhs, i32imm:$index), NVExtFrm,
        IIC_VEXTQ, OpcodeStr, Dt, "$dst, $lhs, $rhs, $index", "",
        [(set QPR:$dst, (Ty (NEONvext (Ty QPR:$lhs),
                                      (Ty QPR:$rhs), imm:$index)))]> {
  bits<4> index;
  let Inst{11-8} = index{3-0};
}

def VEXTd8  : VEXTd<"vext", "8",  v8i8> {
  let Inst{11-8} = index{3-0};
}
def VEXTd16 : VEXTd<"vext", "16", v4i16> {
  let Inst{11-9} = index{2-0};
  let Inst{8}    = 0b0;
}
def VEXTd32 : VEXTd<"vext", "32", v2i32> {
  let Inst{11-10} = index{1-0};
  let Inst{9-8}    = 0b00;
}
def VEXTdf  : VEXTd<"vext", "32", v2f32> {
  let Inst{11}    = index{0};
  let Inst{10-8}  = 0b000;
}

def VEXTq8  : VEXTq<"vext", "8",  v16i8> {
  let Inst{11-8} = index{3-0};
}
def VEXTq16 : VEXTq<"vext", "16", v8i16> {
  let Inst{11-9} = index{2-0};
  let Inst{8}    = 0b0;
}
def VEXTq32 : VEXTq<"vext", "32", v4i32> {
  let Inst{11-10} = index{1-0};
  let Inst{9-8}    = 0b00;
}
def VEXTqf  : VEXTq<"vext", "32", v4f32> {
  let Inst{11}    = index{0};
  let Inst{10-8}  = 0b000;
}

//   VTRN     : Vector Transpose

def  VTRNd8   : N2VDShuffle<0b00, 0b00001, "vtrn", "8">;
def  VTRNd16  : N2VDShuffle<0b01, 0b00001, "vtrn", "16">;
def  VTRNd32  : N2VDShuffle<0b10, 0b00001, "vtrn", "32">;

def  VTRNq8   : N2VQShuffle<0b00, 0b00001, IIC_VPERMQ, "vtrn", "8">;
def  VTRNq16  : N2VQShuffle<0b01, 0b00001, IIC_VPERMQ, "vtrn", "16">;
def  VTRNq32  : N2VQShuffle<0b10, 0b00001, IIC_VPERMQ, "vtrn", "32">;

//   VUZP     : Vector Unzip (Deinterleave)

def  VUZPd8   : N2VDShuffle<0b00, 0b00010, "vuzp", "8">;
def  VUZPd16  : N2VDShuffle<0b01, 0b00010, "vuzp", "16">;
def  VUZPd32  : N2VDShuffle<0b10, 0b00010, "vuzp", "32">;

def  VUZPq8   : N2VQShuffle<0b00, 0b00010, IIC_VPERMQ3, "vuzp", "8">;
def  VUZPq16  : N2VQShuffle<0b01, 0b00010, IIC_VPERMQ3, "vuzp", "16">;
def  VUZPq32  : N2VQShuffle<0b10, 0b00010, IIC_VPERMQ3, "vuzp", "32">;

//   VZIP     : Vector Zip (Interleave)

def  VZIPd8   : N2VDShuffle<0b00, 0b00011, "vzip", "8">;
def  VZIPd16  : N2VDShuffle<0b01, 0b00011, "vzip", "16">;
def  VZIPd32  : N2VDShuffle<0b10, 0b00011, "vzip", "32">;

def  VZIPq8   : N2VQShuffle<0b00, 0b00011, IIC_VPERMQ3, "vzip", "8">;
def  VZIPq16  : N2VQShuffle<0b01, 0b00011, IIC_VPERMQ3, "vzip", "16">;
def  VZIPq32  : N2VQShuffle<0b10, 0b00011, IIC_VPERMQ3, "vzip", "32">;

// Vector Table Lookup and Table Extension.

//   VTBL     : Vector Table Lookup
def  VTBL1
  : N3V<1,1,0b11,0b1000,0,0, (outs DPR:$Vd),
        (ins DPR:$Vn, DPR:$Vm), NVTBLFrm, IIC_VTB1,
        "vtbl", "8", "$Vd, \\{$Vn\\}, $Vm", "",
        [(set DPR:$Vd, (v8i8 (int_arm_neon_vtbl1 DPR:$Vn, DPR:$Vm)))]>;
let hasExtraSrcRegAllocReq = 1 in {
def  VTBL2
  : N3V<1,1,0b11,0b1001,0,0, (outs DPR:$Vd),
        (ins DPR:$Vn, DPR:$tbl2, DPR:$Vm), NVTBLFrm, IIC_VTB2,
        "vtbl", "8", "$Vd, \\{$Vn, $tbl2\\}, $Vm", "", []>;
def  VTBL3
  : N3V<1,1,0b11,0b1010,0,0, (outs DPR:$Vd),
        (ins DPR:$Vn, DPR:$tbl2, DPR:$tbl3, DPR:$Vm), NVTBLFrm, IIC_VTB3,
        "vtbl", "8", "$Vd, \\{$Vn, $tbl2, $tbl3\\}, $Vm", "", []>;
def  VTBL4
  : N3V<1,1,0b11,0b1011,0,0, (outs DPR:$Vd),
        (ins DPR:$Vn, DPR:$tbl2, DPR:$tbl3, DPR:$tbl4, DPR:$Vm),
        NVTBLFrm, IIC_VTB4,
        "vtbl", "8", "$Vd, \\{$Vn, $tbl2, $tbl3, $tbl4\\}, $Vm", "", []>;
} // hasExtraSrcRegAllocReq = 1

def  VTBL2Pseudo
  : PseudoNeonI<(outs DPR:$dst), (ins QPR:$tbl, DPR:$src), IIC_VTB2, "", []>;
def  VTBL3Pseudo
  : PseudoNeonI<(outs DPR:$dst), (ins QQPR:$tbl, DPR:$src), IIC_VTB3, "", []>;
def  VTBL4Pseudo
  : PseudoNeonI<(outs DPR:$dst), (ins QQPR:$tbl, DPR:$src), IIC_VTB4, "", []>;

//   VTBX     : Vector Table Extension
def  VTBX1
  : N3V<1,1,0b11,0b1000,1,0, (outs DPR:$Vd),
        (ins DPR:$orig, DPR:$Vn, DPR:$Vm), NVTBLFrm, IIC_VTBX1,
        "vtbx", "8", "$Vd, \\{$Vn\\}, $Vm", "$orig = $Vd",
        [(set DPR:$Vd, (v8i8 (int_arm_neon_vtbx1
                               DPR:$orig, DPR:$Vn, DPR:$Vm)))]>;
let hasExtraSrcRegAllocReq = 1 in {
def  VTBX2
  : N3V<1,1,0b11,0b1001,1,0, (outs DPR:$Vd),
        (ins DPR:$orig, DPR:$Vn, DPR:$tbl2, DPR:$Vm), NVTBLFrm, IIC_VTBX2,
        "vtbx", "8", "$Vd, \\{$Vn, $tbl2\\}, $Vm", "$orig = $Vd", []>;
def  VTBX3
  : N3V<1,1,0b11,0b1010,1,0, (outs DPR:$Vd),
        (ins DPR:$orig, DPR:$Vn, DPR:$tbl2, DPR:$tbl3, DPR:$Vm),
        NVTBLFrm, IIC_VTBX3,
        "vtbx", "8", "$Vd, \\{$Vn, $tbl2, $tbl3\\}, $Vm",
        "$orig = $Vd", []>;
def  VTBX4
  : N3V<1,1,0b11,0b1011,1,0, (outs DPR:$Vd), (ins DPR:$orig, DPR:$Vn,
        DPR:$tbl2, DPR:$tbl3, DPR:$tbl4, DPR:$Vm), NVTBLFrm, IIC_VTBX4,
        "vtbx", "8", "$Vd, \\{$Vn, $tbl2, $tbl3, $tbl4\\}, $Vm",
        "$orig = $Vd", []>;
} // hasExtraSrcRegAllocReq = 1

def  VTBX2Pseudo
  : PseudoNeonI<(outs DPR:$dst), (ins DPR:$orig, QPR:$tbl, DPR:$src),
                IIC_VTBX2, "$orig = $dst", []>;
def  VTBX3Pseudo
  : PseudoNeonI<(outs DPR:$dst), (ins DPR:$orig, QQPR:$tbl, DPR:$src),
                IIC_VTBX3, "$orig = $dst", []>;
def  VTBX4Pseudo
  : PseudoNeonI<(outs DPR:$dst), (ins DPR:$orig, QQPR:$tbl, DPR:$src),
                IIC_VTBX4, "$orig = $dst", []>;

//===----------------------------------------------------------------------===//
// NEON instructions for single-precision FP math
//===----------------------------------------------------------------------===//

class N2VSPat<SDNode OpNode, ValueType ResTy, ValueType OpTy, NeonI Inst>
  : NEONFPPat<(ResTy (OpNode SPR:$a)),
              (EXTRACT_SUBREG (OpTy (Inst (INSERT_SUBREG (OpTy (IMPLICIT_DEF)),
                                                       SPR:$a, ssub_0))),
                              ssub_0)>;

class N3VSPat<SDNode OpNode, NeonI Inst>
  : NEONFPPat<(f32 (OpNode SPR:$a, SPR:$b)),
              (EXTRACT_SUBREG (v2f32
                                 (Inst (INSERT_SUBREG (v2f32 (IMPLICIT_DEF)),
                                                      SPR:$a, ssub_0),
                                       (INSERT_SUBREG (v2f32 (IMPLICIT_DEF)),
                                                      SPR:$b, ssub_0))),
                              ssub_0)>;

class N3VSMulOpPat<SDNode MulNode, SDNode OpNode, NeonI Inst>
  : NEONFPPat<(f32 (OpNode SPR:$acc, (f32 (MulNode SPR:$a, SPR:$b)))),
              (EXTRACT_SUBREG (Inst (INSERT_SUBREG (v2f32 (IMPLICIT_DEF)),
                                                   SPR:$acc, ssub_0),
                                    (INSERT_SUBREG (v2f32 (IMPLICIT_DEF)),
                                                   SPR:$a, ssub_0),
                                    (INSERT_SUBREG (v2f32 (IMPLICIT_DEF)),
                                                   SPR:$b, ssub_0)),
                              ssub_0)>;

// These need separate instructions because they must use DPR_VFP2 register
// class which have SPR sub-registers.

// Vector Add Operations used for single-precision FP
let neverHasSideEffects = 1 in
def VADDfd_sfp : N3VS<0,0,0b00,0b1101,0, "vadd", "f32", v2f32, v2f32, fadd, 1>;
def : N3VSPat<fadd, VADDfd_sfp>;

// Vector Sub Operations used for single-precision FP
let neverHasSideEffects = 1 in
def VSUBfd_sfp : N3VS<0,0,0b10,0b1101,0, "vsub", "f32", v2f32, v2f32, fsub, 0>;
def : N3VSPat<fsub, VSUBfd_sfp>;

// Vector Multiply Operations used for single-precision FP
let neverHasSideEffects = 1 in
def VMULfd_sfp : N3VS<1,0,0b00,0b1101,1, "vmul", "f32", v2f32, v2f32, fmul, 1>;
def : N3VSPat<fmul, VMULfd_sfp>;

// Vector Multiply-Accumulate/Subtract used for single-precision FP
// vml[as].f32 can cause 4-8 cycle stalls in following ASIMD instructions, so
// we want to avoid them for now. e.g., alternating vmla/vadd instructions.

//let neverHasSideEffects = 1 in
//def VMLAfd_sfp : N3VSMulOp<0,0,0b00,0b1101,1, IIC_VMACD, "vmla", "f32",
//                           v2f32, fmul, fadd>;
//def : N3VSMulOpPat<fmul, fadd, VMLAfd_sfp>;

//let neverHasSideEffects = 1 in
//def VMLSfd_sfp : N3VSMulOp<0,0,0b10,0b1101,1, IIC_VMACD, "vmls", "f32",
//                           v2f32, fmul, fsub>;
//def : N3VSMulOpPat<fmul, fsub, VMLSfd_sfp>;

// Vector Absolute used for single-precision FP
let neverHasSideEffects = 1 in
def  VABSfd_sfp : N2V<0b11, 0b11, 0b10, 0b01, 0b01110, 0, 0,
                      (outs DPR_VFP2:$dst), (ins DPR_VFP2:$src), IIC_VUNAD,
                      "vabs", "f32", "$dst, $src", "", []>;
def : N2VSPat<fabs, f32, v2f32, VABSfd_sfp>;

// Vector Negate used for single-precision FP
let neverHasSideEffects = 1 in
def  VNEGfd_sfp : N2V<0b11, 0b11, 0b10, 0b01, 0b01111, 0, 0,
                      (outs DPR_VFP2:$dst), (ins DPR_VFP2:$src), IIC_VUNAD,
                      "vneg", "f32", "$dst, $src", "", []>;
def : N2VSPat<fneg, f32, v2f32, VNEGfd_sfp>;

// Vector Maximum used for single-precision FP
let neverHasSideEffects = 1 in
def VMAXfd_sfp : N3V<0, 0, 0b00, 0b1111, 0, 0, (outs DPR_VFP2:$dst),
                     (ins DPR_VFP2:$src1, DPR_VFP2:$src2), N3RegFrm, IIC_VBIND,
                     "vmax", "f32", "$dst, $src1, $src2", "", []>;
def : N3VSPat<NEONfmax, VMAXfd_sfp>;

// Vector Minimum used for single-precision FP
let neverHasSideEffects = 1 in
def VMINfd_sfp : N3V<0, 0, 0b00, 0b1111, 0, 0, (outs DPR_VFP2:$dst),
                     (ins DPR_VFP2:$src1, DPR_VFP2:$src2), N3RegFrm, IIC_VBIND,
                     "vmin", "f32", "$dst, $src1, $src2", "", []>;
def : N3VSPat<NEONfmin, VMINfd_sfp>;

// Vector Convert between single-precision FP and integer
let neverHasSideEffects = 1 in
def  VCVTf2sd_sfp : N2VS<0b11, 0b11, 0b10, 0b11, 0b01110, 0, "vcvt", "s32.f32",
                         v2i32, v2f32, fp_to_sint>;
def : N2VSPat<arm_ftosi, f32, v2f32, VCVTf2sd_sfp>;

let neverHasSideEffects = 1 in
def  VCVTf2ud_sfp : N2VS<0b11, 0b11, 0b10, 0b11, 0b01111, 0, "vcvt", "u32.f32",
                         v2i32, v2f32, fp_to_uint>;
def : N2VSPat<arm_ftoui, f32, v2f32, VCVTf2ud_sfp>;

let neverHasSideEffects = 1 in
def  VCVTs2fd_sfp : N2VS<0b11, 0b11, 0b10, 0b11, 0b01100, 0, "vcvt", "f32.s32",
                         v2f32, v2i32, sint_to_fp>;
def : N2VSPat<arm_sitof, f32, v2i32, VCVTs2fd_sfp>;

let neverHasSideEffects = 1 in
def  VCVTu2fd_sfp : N2VS<0b11, 0b11, 0b10, 0b11, 0b01101, 0, "vcvt", "f32.u32",
                         v2f32, v2i32, uint_to_fp>;
def : N2VSPat<arm_uitof, f32, v2i32, VCVTu2fd_sfp>;

//===----------------------------------------------------------------------===//
// Non-Instruction Patterns
//===----------------------------------------------------------------------===//

// bit_convert
def : Pat<(v1i64 (bitconvert (v2i32 DPR:$src))), (v1i64 DPR:$src)>;
def : Pat<(v1i64 (bitconvert (v4i16 DPR:$src))), (v1i64 DPR:$src)>;
def : Pat<(v1i64 (bitconvert (v8i8  DPR:$src))), (v1i64 DPR:$src)>;
def : Pat<(v1i64 (bitconvert (f64   DPR:$src))), (v1i64 DPR:$src)>;
def : Pat<(v1i64 (bitconvert (v2f32 DPR:$src))), (v1i64 DPR:$src)>;
def : Pat<(v2i32 (bitconvert (v1i64 DPR:$src))), (v2i32 DPR:$src)>;
def : Pat<(v2i32 (bitconvert (v4i16 DPR:$src))), (v2i32 DPR:$src)>;
def : Pat<(v2i32 (bitconvert (v8i8  DPR:$src))), (v2i32 DPR:$src)>;
def : Pat<(v2i32 (bitconvert (f64   DPR:$src))), (v2i32 DPR:$src)>;
def : Pat<(v2i32 (bitconvert (v2f32 DPR:$src))), (v2i32 DPR:$src)>;
def : Pat<(v4i16 (bitconvert (v1i64 DPR:$src))), (v4i16 DPR:$src)>;
def : Pat<(v4i16 (bitconvert (v2i32 DPR:$src))), (v4i16 DPR:$src)>;
def : Pat<(v4i16 (bitconvert (v8i8  DPR:$src))), (v4i16 DPR:$src)>;
def : Pat<(v4i16 (bitconvert (f64   DPR:$src))), (v4i16 DPR:$src)>;
def : Pat<(v4i16 (bitconvert (v2f32 DPR:$src))), (v4i16 DPR:$src)>;
def : Pat<(v8i8  (bitconvert (v1i64 DPR:$src))), (v8i8  DPR:$src)>;
def : Pat<(v8i8  (bitconvert (v2i32 DPR:$src))), (v8i8  DPR:$src)>;
def : Pat<(v8i8  (bitconvert (v4i16 DPR:$src))), (v8i8  DPR:$src)>;
def : Pat<(v8i8  (bitconvert (f64   DPR:$src))), (v8i8  DPR:$src)>;
def : Pat<(v8i8  (bitconvert (v2f32 DPR:$src))), (v8i8  DPR:$src)>;
def : Pat<(f64   (bitconvert (v1i64 DPR:$src))), (f64   DPR:$src)>;
def : Pat<(f64   (bitconvert (v2i32 DPR:$src))), (f64   DPR:$src)>;
def : Pat<(f64   (bitconvert (v4i16 DPR:$src))), (f64   DPR:$src)>;
def : Pat<(f64   (bitconvert (v8i8  DPR:$src))), (f64   DPR:$src)>;
def : Pat<(f64   (bitconvert (v2f32 DPR:$src))), (f64   DPR:$src)>;
def : Pat<(v2f32 (bitconvert (f64   DPR:$src))), (v2f32 DPR:$src)>;
def : Pat<(v2f32 (bitconvert (v1i64 DPR:$src))), (v2f32 DPR:$src)>;
def : Pat<(v2f32 (bitconvert (v2i32 DPR:$src))), (v2f32 DPR:$src)>;
def : Pat<(v2f32 (bitconvert (v4i16 DPR:$src))), (v2f32 DPR:$src)>;
def : Pat<(v2f32 (bitconvert (v8i8  DPR:$src))), (v2f32 DPR:$src)>;

def : Pat<(v2i64 (bitconvert (v4i32 QPR:$src))), (v2i64 QPR:$src)>;
def : Pat<(v2i64 (bitconvert (v8i16 QPR:$src))), (v2i64 QPR:$src)>;
def : Pat<(v2i64 (bitconvert (v16i8 QPR:$src))), (v2i64 QPR:$src)>;
def : Pat<(v2i64 (bitconvert (v2f64 QPR:$src))), (v2i64 QPR:$src)>;
def : Pat<(v2i64 (bitconvert (v4f32 QPR:$src))), (v2i64 QPR:$src)>;
def : Pat<(v4i32 (bitconvert (v2i64 QPR:$src))), (v4i32 QPR:$src)>;
def : Pat<(v4i32 (bitconvert (v8i16 QPR:$src))), (v4i32 QPR:$src)>;
def : Pat<(v4i32 (bitconvert (v16i8 QPR:$src))), (v4i32 QPR:$src)>;
def : Pat<(v4i32 (bitconvert (v2f64 QPR:$src))), (v4i32 QPR:$src)>;
def : Pat<(v4i32 (bitconvert (v4f32 QPR:$src))), (v4i32 QPR:$src)>;
def : Pat<(v8i16 (bitconvert (v2i64 QPR:$src))), (v8i16 QPR:$src)>;
def : Pat<(v8i16 (bitconvert (v4i32 QPR:$src))), (v8i16 QPR:$src)>;
def : Pat<(v8i16 (bitconvert (v16i8 QPR:$src))), (v8i16 QPR:$src)>;
def : Pat<(v8i16 (bitconvert (v2f64 QPR:$src))), (v8i16 QPR:$src)>;
def : Pat<(v8i16 (bitconvert (v4f32 QPR:$src))), (v8i16 QPR:$src)>;
def : Pat<(v16i8 (bitconvert (v2i64 QPR:$src))), (v16i8 QPR:$src)>;
def : Pat<(v16i8 (bitconvert (v4i32 QPR:$src))), (v16i8 QPR:$src)>;
def : Pat<(v16i8 (bitconvert (v8i16 QPR:$src))), (v16i8 QPR:$src)>;
def : Pat<(v16i8 (bitconvert (v2f64 QPR:$src))), (v16i8 QPR:$src)>;
def : Pat<(v16i8 (bitconvert (v4f32 QPR:$src))), (v16i8 QPR:$src)>;
def : Pat<(v4f32 (bitconvert (v2i64 QPR:$src))), (v4f32 QPR:$src)>;
def : Pat<(v4f32 (bitconvert (v4i32 QPR:$src))), (v4f32 QPR:$src)>;
def : Pat<(v4f32 (bitconvert (v8i16 QPR:$src))), (v4f32 QPR:$src)>;
def : Pat<(v4f32 (bitconvert (v16i8 QPR:$src))), (v4f32 QPR:$src)>;
def : Pat<(v4f32 (bitconvert (v2f64 QPR:$src))), (v4f32 QPR:$src)>;
def : Pat<(v2f64 (bitconvert (v2i64 QPR:$src))), (v2f64 QPR:$src)>;
def : Pat<(v2f64 (bitconvert (v4i32 QPR:$src))), (v2f64 QPR:$src)>;
def : Pat<(v2f64 (bitconvert (v8i16 QPR:$src))), (v2f64 QPR:$src)>;
def : Pat<(v2f64 (bitconvert (v16i8 QPR:$src))), (v2f64 QPR:$src)>;
def : Pat<(v2f64 (bitconvert (v4f32 QPR:$src))), (v2f64 QPR:$src)>;