summaryrefslogtreecommitdiff
path: root/lib/Target/R600/SIInstructions.td
blob: e1eb95580ac04ce3e50939052ae10a768af0293e (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
//===-- SIInstructions.td - SI Instruction Defintions ---------------------===//
//
//                     The LLVM Compiler Infrastructure
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//
// This file was originally auto-generated from a GPU register header file and
// all the instruction definitions were originally commented out.  Instructions
// that are not yet supported remain commented out.
//===----------------------------------------------------------------------===//

class InterpSlots {
int P0 = 2;
int P10 = 0;
int P20 = 1;
}
def INTERP : InterpSlots;

def InterpSlot : Operand<i32> {
  let PrintMethod = "printInterpSlot";
}

def SendMsgImm : Operand<i32> {
  let PrintMethod = "printSendMsg";
}

def isSI : Predicate<"Subtarget.getGeneration() "
                      ">= AMDGPUSubtarget::SOUTHERN_ISLANDS">;

def isCI : Predicate<"Subtarget.getGeneration() "
                      ">= AMDGPUSubtarget::SEA_ISLANDS">;
def HasFlatAddressSpace : Predicate<"Subtarget.hasFlatAddressSpace()">;

def SWaitMatchClass : AsmOperandClass {
  let Name = "SWaitCnt";
  let RenderMethod = "addImmOperands";
  let ParserMethod = "parseSWaitCntOps";
}

def WAIT_FLAG : InstFlag<"printWaitFlag"> {
  let ParserMatchClass = SWaitMatchClass;
}

let SubtargetPredicate = isSI in {

//===----------------------------------------------------------------------===//
// EXP Instructions
//===----------------------------------------------------------------------===//

defm EXP : EXP_m;

//===----------------------------------------------------------------------===//
// SMRD Instructions
//===----------------------------------------------------------------------===//

let mayLoad = 1 in {

// We are using the SGPR_32 and not the SReg_32 register class for 32-bit
// SMRD instructions, because the SGPR_32 register class does not include M0
// and writing to M0 from an SMRD instruction will hang the GPU.
defm S_LOAD_DWORD : SMRD_Helper <0x00, "s_load_dword", SReg_64, SGPR_32>;
defm S_LOAD_DWORDX2 : SMRD_Helper <0x01, "s_load_dwordx2", SReg_64, SReg_64>;
defm S_LOAD_DWORDX4 : SMRD_Helper <0x02, "s_load_dwordx4", SReg_64, SReg_128>;
defm S_LOAD_DWORDX8 : SMRD_Helper <0x03, "s_load_dwordx8", SReg_64, SReg_256>;
defm S_LOAD_DWORDX16 : SMRD_Helper <0x04, "s_load_dwordx16", SReg_64, SReg_512>;

defm S_BUFFER_LOAD_DWORD : SMRD_Helper <
  0x08, "s_buffer_load_dword", SReg_128, SGPR_32
>;

defm S_BUFFER_LOAD_DWORDX2 : SMRD_Helper <
  0x09, "s_buffer_load_dwordx2", SReg_128, SReg_64
>;

defm S_BUFFER_LOAD_DWORDX4 : SMRD_Helper <
  0x0a, "s_buffer_load_dwordx4", SReg_128, SReg_128
>;

defm S_BUFFER_LOAD_DWORDX8 : SMRD_Helper <
  0x0b, "s_buffer_load_dwordx8", SReg_128, SReg_256
>;

defm S_BUFFER_LOAD_DWORDX16 : SMRD_Helper <
  0x0c, "s_buffer_load_dwordx16", SReg_128, SReg_512
>;

} // mayLoad = 1

//def S_MEMTIME : SMRD_ <0x0000001e, "s_memtime", []>;
//def S_DCACHE_INV : SMRD_ <0x0000001f, "s_dcache_inv", []>;

//===----------------------------------------------------------------------===//
// SOP1 Instructions
//===----------------------------------------------------------------------===//

let isMoveImm = 1 in {
let isReMaterializable = 1 in {
def S_MOV_B32 : SOP1_32 <0x00000003, "s_mov_b32", []>;
def S_MOV_B64 : SOP1_64 <0x00000004, "s_mov_b64", []>;
} // let isRematerializeable = 1
def S_CMOV_B32 : SOP1_32 <0x00000005, "s_cmov_b32", []>;
def S_CMOV_B64 : SOP1_64 <0x00000006, "s_cmov_b64", []>;
} // End isMoveImm = 1

def S_NOT_B32 : SOP1_32 <0x00000007, "s_not_b32",
  [(set i32:$dst, (not i32:$src0))]
>;

def S_NOT_B64 : SOP1_64 <0x00000008, "s_not_b64",
  [(set i64:$dst, (not i64:$src0))]
>;
def S_WQM_B32 : SOP1_32 <0x00000009, "s_wqm_b32", []>;
def S_WQM_B64 : SOP1_64 <0x0000000a, "s_wqm_b64", []>;
def S_BREV_B32 : SOP1_32 <0x0000000b, "s_brev_b32",
  [(set i32:$dst, (AMDGPUbrev i32:$src0))]
>;
def S_BREV_B64 : SOP1_64 <0x0000000c, "s_brev_b64", []>;

////def S_BCNT0_I32_B32 : SOP1_BCNT0 <0x0000000d, "s_bcnt0_i32_b32", []>;
////def S_BCNT0_I32_B64 : SOP1_BCNT0 <0x0000000e, "s_bcnt0_i32_b64", []>;
def S_BCNT1_I32_B32 : SOP1_32 <0x0000000f, "s_bcnt1_i32_b32",
  [(set i32:$dst, (ctpop i32:$src0))]
>;
def S_BCNT1_I32_B64 : SOP1_32_64 <0x00000010, "s_bcnt1_i32_b64", []>;

////def S_FF0_I32_B32 : SOP1_32 <0x00000011, "s_ff0_i32_b32", []>;
////def S_FF0_I32_B64 : SOP1_FF0 <0x00000012, "s_ff0_i32_b64", []>;
def S_FF1_I32_B32 : SOP1_32 <0x00000013, "s_ff1_i32_b32",
  [(set i32:$dst, (cttz_zero_undef i32:$src0))]
>;
////def S_FF1_I32_B64 : SOP1_FF1 <0x00000014, "s_ff1_i32_b64", []>;

def S_FLBIT_I32_B32 : SOP1_32 <0x00000015, "s_flbit_i32_b32",
  [(set i32:$dst, (ctlz_zero_undef i32:$src0))]
>;

//def S_FLBIT_I32_B64 : SOP1_32 <0x00000016, "s_flbit_i32_b64", []>;
def S_FLBIT_I32 : SOP1_32 <0x00000017, "s_flbit_i32", []>;
//def S_FLBIT_I32_I64 : SOP1_32 <0x00000018, "s_flbit_i32_i64", []>;
def S_SEXT_I32_I8 : SOP1_32 <0x00000019, "s_sext_i32_i8",
  [(set i32:$dst, (sext_inreg i32:$src0, i8))]
>;
def S_SEXT_I32_I16 : SOP1_32 <0x0000001a, "s_sext_i32_i16",
  [(set i32:$dst, (sext_inreg i32:$src0, i16))]
>;

////def S_BITSET0_B32 : SOP1_BITSET0 <0x0000001b, "s_bitset0_b32", []>;
////def S_BITSET0_B64 : SOP1_BITSET0 <0x0000001c, "s_bitset0_b64", []>;
////def S_BITSET1_B32 : SOP1_BITSET1 <0x0000001d, "s_bitset1_b32", []>;
////def S_BITSET1_B64 : SOP1_BITSET1 <0x0000001e, "s_bitset1_b64", []>;
def S_GETPC_B64 : SOP1 <
  0x0000001f, (outs SReg_64:$dst), (ins), "s_getpc_b64 $dst", []
> {
  let SSRC0 = 0;
}
def S_SETPC_B64 : SOP1_64 <0x00000020, "s_setpc_b64", []>;
def S_SWAPPC_B64 : SOP1_64 <0x00000021, "s_swappc_b64", []>;
def S_RFE_B64 : SOP1_64 <0x00000022, "s_rfe_b64", []>;

let hasSideEffects = 1, Uses = [EXEC], Defs = [EXEC] in {

def S_AND_SAVEEXEC_B64 : SOP1_64 <0x00000024, "s_and_saveexec_b64", []>;
def S_OR_SAVEEXEC_B64 : SOP1_64 <0x00000025, "s_or_saveexec_b64", []>;
def S_XOR_SAVEEXEC_B64 : SOP1_64 <0x00000026, "s_xor_saveexec_b64", []>;
def S_ANDN2_SAVEEXEC_B64 : SOP1_64 <0x00000027, "s_andn2_saveexec_b64", []>;
def S_ORN2_SAVEEXEC_B64 : SOP1_64 <0x00000028, "s_orn2_saveexec_b64", []>;
def S_NAND_SAVEEXEC_B64 : SOP1_64 <0x00000029, "s_nand_saveexec_b64", []>;
def S_NOR_SAVEEXEC_B64 : SOP1_64 <0x0000002a, "s_nor_saveexec_b64", []>;
def S_XNOR_SAVEEXEC_B64 : SOP1_64 <0x0000002b, "s_xnor_saveexec_b64", []>;

} // End hasSideEffects = 1

def S_QUADMASK_B32 : SOP1_32 <0x0000002c, "s_quadmask_b32", []>;
def S_QUADMASK_B64 : SOP1_64 <0x0000002d, "s_quadmask_b64", []>;
def S_MOVRELS_B32 : SOP1_32 <0x0000002e, "s_movrels_b32", []>;
def S_MOVRELS_B64 : SOP1_64 <0x0000002f, "s_movrels_b64", []>;
def S_MOVRELD_B32 : SOP1_32 <0x00000030, "s_movreld_b32", []>;
def S_MOVRELD_B64 : SOP1_64 <0x00000031, "s_movreld_b64", []>;
//def S_CBRANCH_JOIN : SOP1_ <0x00000032, "s_cbranch_join", []>;
def S_MOV_REGRD_B32 : SOP1_32 <0x00000033, "s_mov_regrd_b32", []>;
def S_ABS_I32 : SOP1_32 <0x00000034, "s_abs_i32", []>;
def S_MOV_FED_B32 : SOP1_32 <0x00000035, "s_mov_fed_b32", []>;

//===----------------------------------------------------------------------===//
// SOP2 Instructions
//===----------------------------------------------------------------------===//

let Defs = [SCC] in { // Carry out goes to SCC
let isCommutable = 1 in {
def S_ADD_U32 : SOP2_32 <0x00000000, "s_add_u32", []>;
def S_ADD_I32 : SOP2_32 <0x00000002, "s_add_i32",
  [(set i32:$dst, (add SSrc_32:$src0, SSrc_32:$src1))]
>;
} // End isCommutable = 1

def S_SUB_U32 : SOP2_32 <0x00000001, "s_sub_u32", []>;
def S_SUB_I32 : SOP2_32 <0x00000003, "s_sub_i32",
  [(set i32:$dst, (sub SSrc_32:$src0, SSrc_32:$src1))]
>;

let Uses = [SCC] in { // Carry in comes from SCC
let isCommutable = 1 in {
def S_ADDC_U32 : SOP2_32 <0x00000004, "s_addc_u32",
  [(set i32:$dst, (adde (i32 SSrc_32:$src0), (i32 SSrc_32:$src1)))]>;
} // End isCommutable = 1

def S_SUBB_U32 : SOP2_32 <0x00000005, "s_subb_u32",
  [(set i32:$dst, (sube (i32 SSrc_32:$src0), (i32 SSrc_32:$src1)))]>;
} // End Uses = [SCC]
} // End Defs = [SCC]

def S_MIN_I32 : SOP2_32 <0x00000006, "s_min_i32",
  [(set i32:$dst, (AMDGPUsmin i32:$src0, i32:$src1))]
>;
def S_MIN_U32 : SOP2_32 <0x00000007, "s_min_u32",
  [(set i32:$dst, (AMDGPUumin i32:$src0, i32:$src1))]
>;
def S_MAX_I32 : SOP2_32 <0x00000008, "s_max_i32",
  [(set i32:$dst, (AMDGPUsmax i32:$src0, i32:$src1))]
>;
def S_MAX_U32 : SOP2_32 <0x00000009, "s_max_u32",
  [(set i32:$dst, (AMDGPUumax i32:$src0, i32:$src1))]
>;

def S_CSELECT_B32 : SOP2_SELECT_32 <
  0x0000000a, "s_cselect_b32",
  []
>;

def S_CSELECT_B64 : SOP2_64 <0x0000000b, "s_cselect_b64", []>;

def S_AND_B32 : SOP2_32 <0x0000000e, "s_and_b32",
  [(set i32:$dst, (and i32:$src0, i32:$src1))]
>;

def S_AND_B64 : SOP2_64 <0x0000000f, "s_and_b64",
  [(set i64:$dst, (and i64:$src0, i64:$src1))]
>;

def S_OR_B32 : SOP2_32 <0x00000010, "s_or_b32",
  [(set i32:$dst, (or i32:$src0, i32:$src1))]
>;

def S_OR_B64 : SOP2_64 <0x00000011, "s_or_b64",
  [(set i64:$dst, (or i64:$src0, i64:$src1))]
>;

def S_XOR_B32 : SOP2_32 <0x00000012, "s_xor_b32",
  [(set i32:$dst, (xor i32:$src0, i32:$src1))]
>;

def S_XOR_B64 : SOP2_64 <0x00000013, "s_xor_b64",
  [(set i64:$dst, (xor i64:$src0, i64:$src1))]
>;
def S_ANDN2_B32 : SOP2_32 <0x00000014, "s_andn2_b32", []>;
def S_ANDN2_B64 : SOP2_64 <0x00000015, "s_andn2_b64", []>;
def S_ORN2_B32 : SOP2_32 <0x00000016, "s_orn2_b32", []>;
def S_ORN2_B64 : SOP2_64 <0x00000017, "s_orn2_b64", []>;
def S_NAND_B32 : SOP2_32 <0x00000018, "s_nand_b32", []>;
def S_NAND_B64 : SOP2_64 <0x00000019, "s_nand_b64", []>;
def S_NOR_B32 : SOP2_32 <0x0000001a, "s_nor_b32", []>;
def S_NOR_B64 : SOP2_64 <0x0000001b, "s_nor_b64", []>;
def S_XNOR_B32 : SOP2_32 <0x0000001c, "s_xnor_b32", []>;
def S_XNOR_B64 : SOP2_64 <0x0000001d, "s_xnor_b64", []>;

// Use added complexity so these patterns are preferred to the VALU patterns.
let AddedComplexity = 1 in {

def S_LSHL_B32 : SOP2_32 <0x0000001e, "s_lshl_b32",
  [(set i32:$dst, (shl i32:$src0, i32:$src1))]
>;
def S_LSHL_B64 : SOP2_SHIFT_64 <0x0000001f, "s_lshl_b64",
  [(set i64:$dst, (shl i64:$src0, i32:$src1))]
>;
def S_LSHR_B32 : SOP2_32 <0x00000020, "s_lshr_b32",
  [(set i32:$dst, (srl i32:$src0, i32:$src1))]
>;
def S_LSHR_B64 : SOP2_SHIFT_64 <0x00000021, "s_lshr_b64",
  [(set i64:$dst, (srl i64:$src0, i32:$src1))]
>;
def S_ASHR_I32 : SOP2_32 <0x00000022, "s_ashr_i32",
  [(set i32:$dst, (sra i32:$src0, i32:$src1))]
>;
def S_ASHR_I64 : SOP2_SHIFT_64 <0x00000023, "s_ashr_i64",
  [(set i64:$dst, (sra i64:$src0, i32:$src1))]
>;


def S_BFM_B32 : SOP2_32 <0x00000024, "s_bfm_b32", []>;
def S_BFM_B64 : SOP2_64 <0x00000025, "s_bfm_b64", []>;
def S_MUL_I32 : SOP2_32 <0x00000026, "s_mul_i32",
  [(set i32:$dst, (mul i32:$src0, i32:$src1))]
>;

} // End AddedComplexity = 1

def S_BFE_U32 : SOP2_32 <0x00000027, "s_bfe_u32", []>;
def S_BFE_I32 : SOP2_32 <0x00000028, "s_bfe_i32", []>;
def S_BFE_U64 : SOP2_64 <0x00000029, "s_bfe_u64", []>;
def S_BFE_I64 : SOP2_64_32 <0x0000002a, "s_bfe_i64", []>;
//def S_CBRANCH_G_FORK : SOP2_ <0x0000002b, "s_cbranch_g_fork", []>;
def S_ABSDIFF_I32 : SOP2_32 <0x0000002c, "s_absdiff_i32", []>;

//===----------------------------------------------------------------------===//
// SOPC Instructions
//===----------------------------------------------------------------------===//

def S_CMP_EQ_I32 : SOPC_32 <0x00000000, "s_cmp_eq_i32">;
def S_CMP_LG_I32 : SOPC_32 <0x00000001, "s_cmp_lg_i32">;
def S_CMP_GT_I32 : SOPC_32 <0x00000002, "s_cmp_gt_i32">;
def S_CMP_GE_I32 : SOPC_32 <0x00000003, "s_cmp_ge_i32">;
def S_CMP_LT_I32 : SOPC_32 <0x00000004, "s_cmp_lt_i32">;
def S_CMP_LE_I32 : SOPC_32 <0x00000005, "s_cmp_le_i32">;
def S_CMP_EQ_U32 : SOPC_32 <0x00000006, "s_cmp_eq_u32">;
def S_CMP_LG_U32 : SOPC_32 <0x00000007, "s_cmp_lg_u32">;
def S_CMP_GT_U32 : SOPC_32 <0x00000008, "s_cmp_gt_u32">;
def S_CMP_GE_U32 : SOPC_32 <0x00000009, "s_cmp_ge_u32">;
def S_CMP_LT_U32 : SOPC_32 <0x0000000a, "s_cmp_lt_u32">;
def S_CMP_LE_U32 : SOPC_32 <0x0000000b, "s_cmp_le_u32">;
////def S_BITCMP0_B32 : SOPC_BITCMP0 <0x0000000c, "s_bitcmp0_b32", []>;
////def S_BITCMP1_B32 : SOPC_BITCMP1 <0x0000000d, "s_bitcmp1_b32", []>;
////def S_BITCMP0_B64 : SOPC_BITCMP0 <0x0000000e, "s_bitcmp0_b64", []>;
////def S_BITCMP1_B64 : SOPC_BITCMP1 <0x0000000f, "s_bitcmp1_b64", []>;
//def S_SETVSKIP : SOPC_ <0x00000010, "s_setvskip", []>;

//===----------------------------------------------------------------------===//
// SOPK Instructions
//===----------------------------------------------------------------------===//

let isReMaterializable = 1 in {
def S_MOVK_I32 : SOPK_32 <0x00000000, "s_movk_i32", []>;
} // End isReMaterializable = 1
def S_CMOVK_I32 : SOPK_32 <0x00000002, "s_cmovk_i32", []>;

/*
This instruction is disabled for now until we can figure out how to teach
the instruction selector to correctly use the  S_CMP* vs V_CMP*
instructions.

When this instruction is enabled the code generator sometimes produces this
invalid sequence:

SCC = S_CMPK_EQ_I32 SGPR0, imm
VCC = COPY SCC
VGPR0 = V_CNDMASK VCC, VGPR0, VGPR1

def S_CMPK_EQ_I32 : SOPK <
  0x00000003, (outs SCCReg:$dst), (ins SReg_32:$src0, i32imm:$src1),
  "s_cmpk_eq_i32",
  [(set i1:$dst, (setcc i32:$src0, imm:$src1, SETEQ))]
>;
*/

let isCompare = 1, Defs = [SCC] in {
def S_CMPK_LG_I32 : SOPK_32 <0x00000004, "s_cmpk_lg_i32", []>;
def S_CMPK_GT_I32 : SOPK_32 <0x00000005, "s_cmpk_gt_i32", []>;
def S_CMPK_GE_I32 : SOPK_32 <0x00000006, "s_cmpk_ge_i32", []>;
def S_CMPK_LT_I32 : SOPK_32 <0x00000007, "s_cmpk_lt_i32", []>;
def S_CMPK_LE_I32 : SOPK_32 <0x00000008, "s_cmpk_le_i32", []>;
def S_CMPK_EQ_U32 : SOPK_32 <0x00000009, "s_cmpk_eq_u32", []>;
def S_CMPK_LG_U32 : SOPK_32 <0x0000000a, "s_cmpk_lg_u32", []>;
def S_CMPK_GT_U32 : SOPK_32 <0x0000000b, "s_cmpk_gt_u32", []>;
def S_CMPK_GE_U32 : SOPK_32 <0x0000000c, "s_cmpk_ge_u32", []>;
def S_CMPK_LT_U32 : SOPK_32 <0x0000000d, "s_cmpk_lt_u32", []>;
def S_CMPK_LE_U32 : SOPK_32 <0x0000000e, "s_cmpk_le_u32", []>;
} // End isCompare = 1, Defs = [SCC]

let Defs = [SCC], isCommutable = 1 in {
  def S_ADDK_I32 : SOPK_32 <0x0000000f, "s_addk_i32", []>;
  def S_MULK_I32 : SOPK_32 <0x00000010, "s_mulk_i32", []>;
}

//def S_CBRANCH_I_FORK : SOPK_ <0x00000011, "s_cbranch_i_fork", []>;
def S_GETREG_B32 : SOPK_32 <0x00000012, "s_getreg_b32", []>;
def S_SETREG_B32 : SOPK_32 <0x00000013, "s_setreg_b32", []>;
def S_GETREG_REGRD_B32 : SOPK_32 <0x00000014, "s_getreg_regrd_b32", []>;
//def S_SETREG_IMM32_B32 : SOPK_32 <0x00000015, "s_setreg_imm32_b32", []>;
//def EXP : EXP_ <0x00000000, "exp", []>;

//===----------------------------------------------------------------------===//
// SOPP Instructions
//===----------------------------------------------------------------------===//

def S_NOP : SOPP <0x00000000, (ins i16imm:$simm16), "s_nop $simm16">;

let isTerminator = 1 in {

def S_ENDPGM : SOPP <0x00000001, (ins), "s_endpgm",
  [(IL_retflag)]> {
  let simm16 = 0;
  let isBarrier = 1;
  let hasCtrlDep = 1;
}

let isBranch = 1 in {
def S_BRANCH : SOPP <
  0x00000002, (ins sopp_brtarget:$simm16), "s_branch $simm16",
  [(br bb:$simm16)]> {
  let isBarrier = 1;
}

let DisableEncoding = "$scc" in {
def S_CBRANCH_SCC0 : SOPP <
  0x00000004, (ins sopp_brtarget:$simm16, SCCReg:$scc),
  "s_cbranch_scc0 $simm16"
>;
def S_CBRANCH_SCC1 : SOPP <
  0x00000005, (ins sopp_brtarget:$simm16, SCCReg:$scc),
  "s_cbranch_scc1 $simm16"
>;
} // End DisableEncoding = "$scc"

def S_CBRANCH_VCCZ : SOPP <
  0x00000006, (ins sopp_brtarget:$simm16, VCCReg:$vcc),
  "s_cbranch_vccz $simm16"
>;
def S_CBRANCH_VCCNZ : SOPP <
  0x00000007, (ins sopp_brtarget:$simm16, VCCReg:$vcc),
  "s_cbranch_vccnz $simm16"
>;

let DisableEncoding = "$exec" in {
def S_CBRANCH_EXECZ : SOPP <
  0x00000008, (ins sopp_brtarget:$simm16, EXECReg:$exec),
  "s_cbranch_execz $simm16"
>;
def S_CBRANCH_EXECNZ : SOPP <
  0x00000009, (ins sopp_brtarget:$simm16, EXECReg:$exec),
  "s_cbranch_execnz $simm16"
>;
} // End DisableEncoding = "$exec"


} // End isBranch = 1
} // End isTerminator = 1

let hasSideEffects = 1 in {
def S_BARRIER : SOPP <0x0000000a, (ins), "s_barrier",
  [(int_AMDGPU_barrier_local)]
> {
  let simm16 = 0;
  let isBarrier = 1;
  let hasCtrlDep = 1;
  let mayLoad = 1;
  let mayStore = 1;
}

def S_WAITCNT : SOPP <0x0000000c, (ins WAIT_FLAG:$simm16), "s_waitcnt $simm16">;
def S_SETHALT : SOPP <0x0000000d, (ins i16imm:$simm16), "s_sethalt $simm16">;
def S_SLEEP : SOPP <0x0000000e, (ins i16imm:$simm16), "s_sleep $simm16">;
def S_SETPRIO : SOPP <0x0000000f, (ins i16imm:$sim16), "s_setprio $sim16">;

let Uses = [EXEC] in {
  def S_SENDMSG : SOPP <0x00000010, (ins SendMsgImm:$simm16, M0Reg:$m0), "s_sendmsg $simm16",
      [(int_SI_sendmsg imm:$simm16, M0Reg:$m0)]
  > {
    let DisableEncoding = "$m0";
  }
} // End Uses = [EXEC]

def S_SENDMSGHALT : SOPP <0x00000011, (ins i16imm:$simm16), "s_sendmsghalt $simm16">;
def S_TRAP : SOPP <0x00000012, (ins i16imm:$simm16), "s_trap $simm16">;
def S_ICACHE_INV : SOPP <0x00000013, (ins), "s_icache_inv"> {
	let simm16 = 0;
}
def S_INCPERFLEVEL : SOPP <0x00000014, (ins i16imm:$simm16), "s_incperflevel $simm16">;
def S_DECPERFLEVEL : SOPP <0x00000015, (ins i16imm:$simm16), "s_decperflevel $simm16">;
def S_TTRACEDATA : SOPP <0x00000016, (ins), "s_ttracedata"> {
  let simm16 = 0;
}
} // End hasSideEffects

//===----------------------------------------------------------------------===//
// VOPC Instructions
//===----------------------------------------------------------------------===//

let isCompare = 1 in {

defm V_CMP_F_F32 : VOPC_F32 <vopc<0x0>, "v_cmp_f_f32">;
defm V_CMP_LT_F32 : VOPC_F32 <vopc<0x1>, "v_cmp_lt_f32", COND_OLT>;
defm V_CMP_EQ_F32 : VOPC_F32 <vopc<0x2>, "v_cmp_eq_f32", COND_OEQ>;
defm V_CMP_LE_F32 : VOPC_F32 <vopc<0x3>, "v_cmp_le_f32", COND_OLE>;
defm V_CMP_GT_F32 : VOPC_F32 <vopc<0x4>, "v_cmp_gt_f32", COND_OGT>;
defm V_CMP_LG_F32 : VOPC_F32 <vopc<0x5>, "v_cmp_lg_f32">;
defm V_CMP_GE_F32 : VOPC_F32 <vopc<0x6>, "v_cmp_ge_f32", COND_OGE>;
defm V_CMP_O_F32 : VOPC_F32 <vopc<0x7>, "v_cmp_o_f32", COND_O>;
defm V_CMP_U_F32 : VOPC_F32 <vopc<0x8>, "v_cmp_u_f32", COND_UO>;
defm V_CMP_NGE_F32 : VOPC_F32 <vopc<0x9>, "v_cmp_nge_f32">;
defm V_CMP_NLG_F32 : VOPC_F32 <vopc<0xa>, "v_cmp_nlg_f32">;
defm V_CMP_NGT_F32 : VOPC_F32 <vopc<0xb>, "v_cmp_ngt_f32">;
defm V_CMP_NLE_F32 : VOPC_F32 <vopc<0xc>, "v_cmp_nle_f32">;
defm V_CMP_NEQ_F32 : VOPC_F32 <vopc<0xd>, "v_cmp_neq_f32", COND_UNE>;
defm V_CMP_NLT_F32 : VOPC_F32 <vopc<0xe>, "v_cmp_nlt_f32">;
defm V_CMP_TRU_F32 : VOPC_F32 <vopc<0xf>, "v_cmp_tru_f32">;

let hasSideEffects = 1 in {

defm V_CMPX_F_F32 : VOPCX_F32 <vopc<0x10>, "v_cmpx_f_f32">;
defm V_CMPX_LT_F32 : VOPCX_F32 <vopc<0x11>, "v_cmpx_lt_f32">;
defm V_CMPX_EQ_F32 : VOPCX_F32 <vopc<0x12>, "v_cmpx_eq_f32">;
defm V_CMPX_LE_F32 : VOPCX_F32 <vopc<0x13>, "v_cmpx_le_f32">;
defm V_CMPX_GT_F32 : VOPCX_F32 <vopc<0x14>, "v_cmpx_gt_f32">;
defm V_CMPX_LG_F32 : VOPCX_F32 <vopc<0x15>, "v_cmpx_lg_f32">;
defm V_CMPX_GE_F32 : VOPCX_F32 <vopc<0x16>, "v_cmpx_ge_f32">;
defm V_CMPX_O_F32 : VOPCX_F32 <vopc<0x17>, "v_cmpx_o_f32">;
defm V_CMPX_U_F32 : VOPCX_F32 <vopc<0x18>, "v_cmpx_u_f32">;
defm V_CMPX_NGE_F32 : VOPCX_F32 <vopc<0x19>, "v_cmpx_nge_f32">;
defm V_CMPX_NLG_F32 : VOPCX_F32 <vopc<0x1a>, "v_cmpx_nlg_f32">;
defm V_CMPX_NGT_F32 : VOPCX_F32 <vopc<0x1b>, "v_cmpx_ngt_f32">;
defm V_CMPX_NLE_F32 : VOPCX_F32 <vopc<0x1c>, "v_cmpx_nle_f32">;
defm V_CMPX_NEQ_F32 : VOPCX_F32 <vopc<0x1d>, "v_cmpx_neq_f32">;
defm V_CMPX_NLT_F32 : VOPCX_F32 <vopc<0x1e>, "v_cmpx_nlt_f32">;
defm V_CMPX_TRU_F32 : VOPCX_F32 <vopc<0x1f>, "v_cmpx_tru_f32">;

} // End hasSideEffects = 1

defm V_CMP_F_F64 : VOPC_F64 <vopc<0x20>, "v_cmp_f_f64">;
defm V_CMP_LT_F64 : VOPC_F64 <vopc<0x21>, "v_cmp_lt_f64", COND_OLT>;
defm V_CMP_EQ_F64 : VOPC_F64 <vopc<0x22>, "v_cmp_eq_f64", COND_OEQ>;
defm V_CMP_LE_F64 : VOPC_F64 <vopc<0x23>, "v_cmp_le_f64", COND_OLE>;
defm V_CMP_GT_F64 : VOPC_F64 <vopc<0x24>, "v_cmp_gt_f64", COND_OGT>;
defm V_CMP_LG_F64 : VOPC_F64 <vopc<0x25>, "v_cmp_lg_f64">;
defm V_CMP_GE_F64 : VOPC_F64 <vopc<0x26>, "v_cmp_ge_f64", COND_OGE>;
defm V_CMP_O_F64 : VOPC_F64 <vopc<0x27>, "v_cmp_o_f64", COND_O>;
defm V_CMP_U_F64 : VOPC_F64 <vopc<0x28>, "v_cmp_u_f64", COND_UO>;
defm V_CMP_NGE_F64 : VOPC_F64 <vopc<0x29>, "v_cmp_nge_f64">;
defm V_CMP_NLG_F64 : VOPC_F64 <vopc<0x2a>, "v_cmp_nlg_f64">;
defm V_CMP_NGT_F64 : VOPC_F64 <vopc<0x2b>, "v_cmp_ngt_f64">;
defm V_CMP_NLE_F64 : VOPC_F64 <vopc<0x2c>, "v_cmp_nle_f64">;
defm V_CMP_NEQ_F64 : VOPC_F64 <vopc<0x2d>, "v_cmp_neq_f64", COND_UNE>;
defm V_CMP_NLT_F64 : VOPC_F64 <vopc<0x2e>, "v_cmp_nlt_f64">;
defm V_CMP_TRU_F64 : VOPC_F64 <vopc<0x2f>, "v_cmp_tru_f64">;

let hasSideEffects = 1 in {

defm V_CMPX_F_F64 : VOPCX_F64 <vopc<0x30>, "v_cmpx_f_f64">;
defm V_CMPX_LT_F64 : VOPCX_F64 <vopc<0x31>, "v_cmpx_lt_f64">;
defm V_CMPX_EQ_F64 : VOPCX_F64 <vopc<0x32>, "v_cmpx_eq_f64">;
defm V_CMPX_LE_F64 : VOPCX_F64 <vopc<0x33>, "v_cmpx_le_f64">;
defm V_CMPX_GT_F64 : VOPCX_F64 <vopc<0x34>, "v_cmpx_gt_f64">;
defm V_CMPX_LG_F64 : VOPCX_F64 <vopc<0x35>, "v_cmpx_lg_f64">;
defm V_CMPX_GE_F64 : VOPCX_F64 <vopc<0x36>, "v_cmpx_ge_f64">;
defm V_CMPX_O_F64 : VOPCX_F64 <vopc<0x37>, "v_cmpx_o_f64">;
defm V_CMPX_U_F64 : VOPCX_F64 <vopc<0x38>, "v_cmpx_u_f64">;
defm V_CMPX_NGE_F64 : VOPCX_F64 <vopc<0x39>, "v_cmpx_nge_f64">;
defm V_CMPX_NLG_F64 : VOPCX_F64 <vopc<0x3a>, "v_cmpx_nlg_f64">;
defm V_CMPX_NGT_F64 : VOPCX_F64 <vopc<0x3b>, "v_cmpx_ngt_f64">;
defm V_CMPX_NLE_F64 : VOPCX_F64 <vopc<0x3c>, "v_cmpx_nle_f64">;
defm V_CMPX_NEQ_F64 : VOPCX_F64 <vopc<0x3d>, "v_cmpx_neq_f64">;
defm V_CMPX_NLT_F64 : VOPCX_F64 <vopc<0x3e>, "v_cmpx_nlt_f64">;
defm V_CMPX_TRU_F64 : VOPCX_F64 <vopc<0x3f>, "v_cmpx_tru_f64">;

} // End hasSideEffects = 1

defm V_CMPS_F_F32 : VOPC_F32 <vopc<0x40>, "v_cmps_f_f32">;
defm V_CMPS_LT_F32 : VOPC_F32 <vopc<0x41>, "v_cmps_lt_f32">;
defm V_CMPS_EQ_F32 : VOPC_F32 <vopc<0x42>, "v_cmps_eq_f32">;
defm V_CMPS_LE_F32 : VOPC_F32 <vopc<0x43>, "v_cmps_le_f32">;
defm V_CMPS_GT_F32 : VOPC_F32 <vopc<0x44>, "v_cmps_gt_f32">;
defm V_CMPS_LG_F32 : VOPC_F32 <vopc<0x45>, "v_cmps_lg_f32">;
defm V_CMPS_GE_F32 : VOPC_F32 <vopc<0x46>, "v_cmps_ge_f32">;
defm V_CMPS_O_F32 : VOPC_F32 <vopc<0x47>, "v_cmps_o_f32">;
defm V_CMPS_U_F32 : VOPC_F32 <vopc<0x48>, "v_cmps_u_f32">;
defm V_CMPS_NGE_F32 : VOPC_F32 <vopc<0x49>, "v_cmps_nge_f32">;
defm V_CMPS_NLG_F32 : VOPC_F32 <vopc<0x4a>, "v_cmps_nlg_f32">;
defm V_CMPS_NGT_F32 : VOPC_F32 <vopc<0x4b>, "v_cmps_ngt_f32">;
defm V_CMPS_NLE_F32 : VOPC_F32 <vopc<0x4c>, "v_cmps_nle_f32">;
defm V_CMPS_NEQ_F32 : VOPC_F32 <vopc<0x4d>, "v_cmps_neq_f32">;
defm V_CMPS_NLT_F32 : VOPC_F32 <vopc<0x4e>, "v_cmps_nlt_f32">;
defm V_CMPS_TRU_F32 : VOPC_F32 <vopc<0x4f>, "v_cmps_tru_f32">;

let hasSideEffects = 1 in {

defm V_CMPSX_F_F32 : VOPCX_F32 <vopc<0x50>, "v_cmpsx_f_f32">;
defm V_CMPSX_LT_F32 : VOPCX_F32 <vopc<0x51>, "v_cmpsx_lt_f32">;
defm V_CMPSX_EQ_F32 : VOPCX_F32 <vopc<0x52>, "v_cmpsx_eq_f32">;
defm V_CMPSX_LE_F32 : VOPCX_F32 <vopc<0x53>, "v_cmpsx_le_f32">;
defm V_CMPSX_GT_F32 : VOPCX_F32 <vopc<0x54>, "v_cmpsx_gt_f32">;
defm V_CMPSX_LG_F32 : VOPCX_F32 <vopc<0x55>, "v_cmpsx_lg_f32">;
defm V_CMPSX_GE_F32 : VOPCX_F32 <vopc<0x56>, "v_cmpsx_ge_f32">;
defm V_CMPSX_O_F32 : VOPCX_F32 <vopc<0x57>, "v_cmpsx_o_f32">;
defm V_CMPSX_U_F32 : VOPCX_F32 <vopc<0x58>, "v_cmpsx_u_f32">;
defm V_CMPSX_NGE_F32 : VOPCX_F32 <vopc<0x59>, "v_cmpsx_nge_f32">;
defm V_CMPSX_NLG_F32 : VOPCX_F32 <vopc<0x5a>, "v_cmpsx_nlg_f32">;
defm V_CMPSX_NGT_F32 : VOPCX_F32 <vopc<0x5b>, "v_cmpsx_ngt_f32">;
defm V_CMPSX_NLE_F32 : VOPCX_F32 <vopc<0x5c>, "v_cmpsx_nle_f32">;
defm V_CMPSX_NEQ_F32 : VOPCX_F32 <vopc<0x5d>, "v_cmpsx_neq_f32">;
defm V_CMPSX_NLT_F32 : VOPCX_F32 <vopc<0x5e>, "v_cmpsx_nlt_f32">;
defm V_CMPSX_TRU_F32 : VOPCX_F32 <vopc<0x5f>, "v_cmpsx_tru_f32">;

} // End hasSideEffects = 1

defm V_CMPS_F_F64 : VOPC_F64 <vopc<0x60>, "v_cmps_f_f64">;
defm V_CMPS_LT_F64 : VOPC_F64 <vopc<0x61>, "v_cmps_lt_f64">;
defm V_CMPS_EQ_F64 : VOPC_F64 <vopc<0x62>, "v_cmps_eq_f64">;
defm V_CMPS_LE_F64 : VOPC_F64 <vopc<0x63>, "v_cmps_le_f64">;
defm V_CMPS_GT_F64 : VOPC_F64 <vopc<0x64>, "v_cmps_gt_f64">;
defm V_CMPS_LG_F64 : VOPC_F64 <vopc<0x65>, "v_cmps_lg_f64">;
defm V_CMPS_GE_F64 : VOPC_F64 <vopc<0x66>, "v_cmps_ge_f64">;
defm V_CMPS_O_F64 : VOPC_F64 <vopc<0x67>, "v_cmps_o_f64">;
defm V_CMPS_U_F64 : VOPC_F64 <vopc<0x68>, "v_cmps_u_f64">;
defm V_CMPS_NGE_F64 : VOPC_F64 <vopc<0x69>, "v_cmps_nge_f64">;
defm V_CMPS_NLG_F64 : VOPC_F64 <vopc<0x6a>, "v_cmps_nlg_f64">;
defm V_CMPS_NGT_F64 : VOPC_F64 <vopc<0x6b>, "v_cmps_ngt_f64">;
defm V_CMPS_NLE_F64 : VOPC_F64 <vopc<0x6c>, "v_cmps_nle_f64">;
defm V_CMPS_NEQ_F64 : VOPC_F64 <vopc<0x6d>, "v_cmps_neq_f64">;
defm V_CMPS_NLT_F64 : VOPC_F64 <vopc<0x6e>, "v_cmps_nlt_f64">;
defm V_CMPS_TRU_F64 : VOPC_F64 <vopc<0x6f>, "v_cmps_tru_f64">;

let hasSideEffects = 1, Defs = [EXEC] in {

defm V_CMPSX_F_F64 : VOPC_F64 <vopc<0x70>, "v_cmpsx_f_f64">;
defm V_CMPSX_LT_F64 : VOPC_F64 <vopc<0x71>, "v_cmpsx_lt_f64">;
defm V_CMPSX_EQ_F64 : VOPC_F64 <vopc<0x72>, "v_cmpsx_eq_f64">;
defm V_CMPSX_LE_F64 : VOPC_F64 <vopc<0x73>, "v_cmpsx_le_f64">;
defm V_CMPSX_GT_F64 : VOPC_F64 <vopc<0x74>, "v_cmpsx_gt_f64">;
defm V_CMPSX_LG_F64 : VOPC_F64 <vopc<0x75>, "v_cmpsx_lg_f64">;
defm V_CMPSX_GE_F64 : VOPC_F64 <vopc<0x76>, "v_cmpsx_ge_f64">;
defm V_CMPSX_O_F64 : VOPC_F64 <vopc<0x77>, "v_cmpsx_o_f64">;
defm V_CMPSX_U_F64 : VOPC_F64 <vopc<0x78>, "v_cmpsx_u_f64">;
defm V_CMPSX_NGE_F64 : VOPC_F64 <vopc<0x79>, "v_cmpsx_nge_f64">;
defm V_CMPSX_NLG_F64 : VOPC_F64 <vopc<0x7a>, "v_cmpsx_nlg_f64">;
defm V_CMPSX_NGT_F64 : VOPC_F64 <vopc<0x7b>, "v_cmpsx_ngt_f64">;
defm V_CMPSX_NLE_F64 : VOPC_F64 <vopc<0x7c>, "v_cmpsx_nle_f64">;
defm V_CMPSX_NEQ_F64 : VOPC_F64 <vopc<0x7d>, "v_cmpsx_neq_f64">;
defm V_CMPSX_NLT_F64 : VOPC_F64 <vopc<0x7e>, "v_cmpsx_nlt_f64">;
defm V_CMPSX_TRU_F64 : VOPC_F64 <vopc<0x7f>, "v_cmpsx_tru_f64">;

} // End hasSideEffects = 1, Defs = [EXEC]

defm V_CMP_F_I32 : VOPC_I32 <vopc<0x80>, "v_cmp_f_i32">;
defm V_CMP_LT_I32 : VOPC_I32 <vopc<0x81>, "v_cmp_lt_i32", COND_SLT>;
defm V_CMP_EQ_I32 : VOPC_I32 <vopc<0x82>, "v_cmp_eq_i32", COND_EQ>;
defm V_CMP_LE_I32 : VOPC_I32 <vopc<0x83>, "v_cmp_le_i32", COND_SLE>;
defm V_CMP_GT_I32 : VOPC_I32 <vopc<0x84>, "v_cmp_gt_i32", COND_SGT>;
defm V_CMP_NE_I32 : VOPC_I32 <vopc<0x85>, "v_cmp_ne_i32", COND_NE>;
defm V_CMP_GE_I32 : VOPC_I32 <vopc<0x86>, "v_cmp_ge_i32", COND_SGE>;
defm V_CMP_T_I32 : VOPC_I32 <vopc<0x87>, "v_cmp_t_i32">;

let hasSideEffects = 1 in {

defm V_CMPX_F_I32 : VOPCX_I32 <vopc<0x90>, "v_cmpx_f_i32">;
defm V_CMPX_LT_I32 : VOPCX_I32 <vopc<0x91>, "v_cmpx_lt_i32">;
defm V_CMPX_EQ_I32 : VOPCX_I32 <vopc<0x92>, "v_cmpx_eq_i32">;
defm V_CMPX_LE_I32 : VOPCX_I32 <vopc<0x93>, "v_cmpx_le_i32">;
defm V_CMPX_GT_I32 : VOPCX_I32 <vopc<0x94>, "v_cmpx_gt_i32">;
defm V_CMPX_NE_I32 : VOPCX_I32 <vopc<0x95>, "v_cmpx_ne_i32">;
defm V_CMPX_GE_I32 : VOPCX_I32 <vopc<0x96>, "v_cmpx_ge_i32">;
defm V_CMPX_T_I32 : VOPCX_I32 <vopc<0x97>, "v_cmpx_t_i32">;

} // End hasSideEffects = 1

defm V_CMP_F_I64 : VOPC_I64 <vopc<0xa0>, "v_cmp_f_i64">;
defm V_CMP_LT_I64 : VOPC_I64 <vopc<0xa1>, "v_cmp_lt_i64", COND_SLT>;
defm V_CMP_EQ_I64 : VOPC_I64 <vopc<0xa2>, "v_cmp_eq_i64", COND_EQ>;
defm V_CMP_LE_I64 : VOPC_I64 <vopc<0xa3>, "v_cmp_le_i64", COND_SLE>;
defm V_CMP_GT_I64 : VOPC_I64 <vopc<0xa4>, "v_cmp_gt_i64", COND_SGT>;
defm V_CMP_NE_I64 : VOPC_I64 <vopc<0xa5>, "v_cmp_ne_i64", COND_NE>;
defm V_CMP_GE_I64 : VOPC_I64 <vopc<0xa6>, "v_cmp_ge_i64", COND_SGE>;
defm V_CMP_T_I64 : VOPC_I64 <vopc<0xa7>, "v_cmp_t_i64">;

let hasSideEffects = 1 in {

defm V_CMPX_F_I64 : VOPCX_I64 <vopc<0xb0>, "v_cmpx_f_i64">;
defm V_CMPX_LT_I64 : VOPCX_I64 <vopc<0xb1>, "v_cmpx_lt_i64">;
defm V_CMPX_EQ_I64 : VOPCX_I64 <vopc<0xb2>, "v_cmpx_eq_i64">;
defm V_CMPX_LE_I64 : VOPCX_I64 <vopc<0xb3>, "v_cmpx_le_i64">;
defm V_CMPX_GT_I64 : VOPCX_I64 <vopc<0xb4>, "v_cmpx_gt_i64">;
defm V_CMPX_NE_I64 : VOPCX_I64 <vopc<0xb5>, "v_cmpx_ne_i64">;
defm V_CMPX_GE_I64 : VOPCX_I64 <vopc<0xb6>, "v_cmpx_ge_i64">;
defm V_CMPX_T_I64 : VOPCX_I64 <vopc<0xb7>, "v_cmpx_t_i64">;

} // End hasSideEffects = 1

defm V_CMP_F_U32 : VOPC_I32 <vopc<0xc0>, "v_cmp_f_u32">;
defm V_CMP_LT_U32 : VOPC_I32 <vopc<0xc1>, "v_cmp_lt_u32", COND_ULT>;
defm V_CMP_EQ_U32 : VOPC_I32 <vopc<0xc2>, "v_cmp_eq_u32", COND_EQ>;
defm V_CMP_LE_U32 : VOPC_I32 <vopc<0xc3>, "v_cmp_le_u32", COND_ULE>;
defm V_CMP_GT_U32 : VOPC_I32 <vopc<0xc4>, "v_cmp_gt_u32", COND_UGT>;
defm V_CMP_NE_U32 : VOPC_I32 <vopc<0xc5>, "v_cmp_ne_u32", COND_NE>;
defm V_CMP_GE_U32 : VOPC_I32 <vopc<0xc6>, "v_cmp_ge_u32", COND_UGE>;
defm V_CMP_T_U32 : VOPC_I32 <vopc<0xc7>, "v_cmp_t_u32">;

let hasSideEffects = 1 in {

defm V_CMPX_F_U32 : VOPCX_I32 <vopc<0xd0>, "v_cmpx_f_u32">;
defm V_CMPX_LT_U32 : VOPCX_I32 <vopc<0xd1>, "v_cmpx_lt_u32">;
defm V_CMPX_EQ_U32 : VOPCX_I32 <vopc<0xd2>, "v_cmpx_eq_u32">;
defm V_CMPX_LE_U32 : VOPCX_I32 <vopc<0xd3>, "v_cmpx_le_u32">;
defm V_CMPX_GT_U32 : VOPCX_I32 <vopc<0xd4>, "v_cmpx_gt_u32">;
defm V_CMPX_NE_U32 : VOPCX_I32 <vopc<0xd5>, "v_cmpx_ne_u32">;
defm V_CMPX_GE_U32 : VOPCX_I32 <vopc<0xd6>, "v_cmpx_ge_u32">;
defm V_CMPX_T_U32 : VOPCX_I32 <vopc<0xd7>, "v_cmpx_t_u32">;

} // End hasSideEffects = 1

defm V_CMP_F_U64 : VOPC_I64 <vopc<0xe0>, "v_cmp_f_u64">;
defm V_CMP_LT_U64 : VOPC_I64 <vopc<0xe1>, "v_cmp_lt_u64", COND_ULT>;
defm V_CMP_EQ_U64 : VOPC_I64 <vopc<0xe2>, "v_cmp_eq_u64", COND_EQ>;
defm V_CMP_LE_U64 : VOPC_I64 <vopc<0xe3>, "v_cmp_le_u64", COND_ULE>;
defm V_CMP_GT_U64 : VOPC_I64 <vopc<0xe4>, "v_cmp_gt_u64", COND_UGT>;
defm V_CMP_NE_U64 : VOPC_I64 <vopc<0xe5>, "v_cmp_ne_u64", COND_NE>;
defm V_CMP_GE_U64 : VOPC_I64 <vopc<0xe6>, "v_cmp_ge_u64", COND_UGE>;
defm V_CMP_T_U64 : VOPC_I64 <vopc<0xe7>, "v_cmp_t_u64">;

let hasSideEffects = 1 in {

defm V_CMPX_F_U64 : VOPCX_I64 <vopc<0xf0>, "v_cmpx_f_u64">;
defm V_CMPX_LT_U64 : VOPCX_I64 <vopc<0xf1>, "v_cmpx_lt_u64">;
defm V_CMPX_EQ_U64 : VOPCX_I64 <vopc<0xf2>, "v_cmpx_eq_u64">;
defm V_CMPX_LE_U64 : VOPCX_I64 <vopc<0xf3>, "v_cmpx_le_u64">;
defm V_CMPX_GT_U64 : VOPCX_I64 <vopc<0xf4>, "v_cmpx_gt_u64">;
defm V_CMPX_NE_U64 : VOPCX_I64 <vopc<0xf5>, "v_cmpx_ne_u64">;
defm V_CMPX_GE_U64 : VOPCX_I64 <vopc<0xf6>, "v_cmpx_ge_u64">;
defm V_CMPX_T_U64 : VOPCX_I64 <vopc<0xf7>, "v_cmpx_t_u64">;

} // End hasSideEffects = 1

defm V_CMP_CLASS_F32 : VOPC_F32 <vopc<0x88>, "v_cmp_class_f32">;

let hasSideEffects = 1 in {
defm V_CMPX_CLASS_F32 : VOPCX_F32 <vopc<0x98>, "v_cmpx_class_f32">;
} // End hasSideEffects = 1

defm V_CMP_CLASS_F64 : VOPC_F64 <vopc<0xa8>, "v_cmp_class_f64">;

let hasSideEffects = 1 in {
defm V_CMPX_CLASS_F64 : VOPCX_F64 <vopc<0xb8>, "v_cmpx_class_f64">;
} // End hasSideEffects = 1

} // End isCompare = 1

//===----------------------------------------------------------------------===//
// DS Instructions
//===----------------------------------------------------------------------===//


def DS_ADD_U32 : DS_1A1D_NORET <0x0, "ds_add_u32", VReg_32>;
def DS_SUB_U32 : DS_1A1D_NORET <0x1, "ds_sub_u32", VReg_32>;
def DS_RSUB_U32 : DS_1A1D_NORET <0x2, "ds_rsub_u32", VReg_32>;
def DS_INC_U32 : DS_1A1D_NORET <0x3, "ds_inc_u32", VReg_32>;
def DS_DEC_U32 : DS_1A1D_NORET <0x4, "ds_dec_u32", VReg_32>;
def DS_MIN_I32 : DS_1A1D_NORET <0x5, "ds_min_i32", VReg_32>;
def DS_MAX_I32 : DS_1A1D_NORET <0x6, "ds_max_i32", VReg_32>;
def DS_MIN_U32 : DS_1A1D_NORET <0x7, "ds_min_u32", VReg_32>;
def DS_MAX_U32 : DS_1A1D_NORET <0x8, "ds_max_u32", VReg_32>;
def DS_AND_B32 : DS_1A1D_NORET <0x9, "ds_and_b32", VReg_32>;
def DS_OR_B32 : DS_1A1D_NORET <0xa, "ds_or_b32", VReg_32>;
def DS_XOR_B32 : DS_1A1D_NORET <0xb, "ds_xor_b32", VReg_32>;
def DS_MSKOR_B32 : DS_1A1D_NORET <0xc, "ds_mskor_b32", VReg_32>;
def DS_CMPST_B32 : DS_1A2D_NORET <0x10, "ds_cmpst_b32", VReg_32>;
def DS_CMPST_F32 : DS_1A2D_NORET <0x11, "ds_cmpst_f32", VReg_32>;
def DS_MIN_F32 : DS_1A1D_NORET <0x12, "ds_min_f32", VReg_32>;
def DS_MAX_F32 : DS_1A1D_NORET <0x13, "ds_max_f32", VReg_32>;

def DS_ADD_RTN_U32 : DS_1A1D_RET <0x20, "ds_add_rtn_u32", VReg_32, "ds_add_u32">;
def DS_SUB_RTN_U32 : DS_1A1D_RET <0x21, "ds_sub_rtn_u32", VReg_32, "ds_sub_u32">;
def DS_RSUB_RTN_U32 : DS_1A1D_RET <0x22, "ds_rsub_rtn_u32", VReg_32, "ds_rsub_u32">;
def DS_INC_RTN_U32 : DS_1A1D_RET <0x23, "ds_inc_rtn_u32", VReg_32, "ds_inc_u32">;
def DS_DEC_RTN_U32 : DS_1A1D_RET <0x24, "ds_dec_rtn_u32", VReg_32, "ds_dec_u32">;
def DS_MIN_RTN_I32 : DS_1A1D_RET <0x25, "ds_min_rtn_i32", VReg_32, "ds_min_i32">;
def DS_MAX_RTN_I32 : DS_1A1D_RET <0x26, "ds_max_rtn_i32", VReg_32, "ds_max_i32">;
def DS_MIN_RTN_U32 : DS_1A1D_RET <0x27, "ds_min_rtn_u32", VReg_32, "ds_min_u32">;
def DS_MAX_RTN_U32 : DS_1A1D_RET <0x28, "ds_max_rtn_u32", VReg_32, "ds_max_u32">;
def DS_AND_RTN_B32 : DS_1A1D_RET <0x29, "ds_and_rtn_b32", VReg_32, "ds_and_b32">;
def DS_OR_RTN_B32 : DS_1A1D_RET <0x2a, "ds_or_rtn_b32", VReg_32, "ds_or_b32">;
def DS_XOR_RTN_B32 : DS_1A1D_RET <0x2b, "ds_xor_rtn_b32", VReg_32, "ds_xor_b32">;
def DS_MSKOR_RTN_B32 : DS_1A1D_RET <0x2c, "ds_mskor_rtn_b32", VReg_32, "ds_mskor_b32">;
def DS_WRXCHG_RTN_B32 : DS_1A1D_RET <0x2d, "ds_wrxchg_rtn_b32", VReg_32>;
//def DS_WRXCHG2_RTN_B32 : DS_2A0D_RET <0x2e, "ds_wrxchg2_rtn_b32", VReg_32, "ds_wrxchg2_b32">;
//def DS_WRXCHG2ST64_RTN_B32 : DS_2A0D_RET <0x2f, "ds_wrxchg2_rtn_b32", VReg_32, "ds_wrxchg2st64_b32">;
def DS_CMPST_RTN_B32 : DS_1A2D_RET <0x30, "ds_cmpst_rtn_b32", VReg_32, "ds_cmpst_b32">;
def DS_CMPST_RTN_F32 : DS_1A2D_RET <0x31, "ds_cmpst_rtn_f32", VReg_32, "ds_cmpst_f32">;
def DS_MIN_RTN_F32 : DS_1A1D_RET <0x32, "ds_min_rtn_f32", VReg_32, "ds_min_f32">;
def DS_MAX_RTN_F32 : DS_1A1D_RET <0x33, "ds_max_rtn_f32", VReg_32, "ds_max_f32">;

let SubtargetPredicate = isCI in {
def DS_WRAP_RTN_F32 : DS_1A1D_RET <0x34, "ds_wrap_rtn_f32", VReg_32, "ds_wrap_f32">;
} // End isCI


def DS_ADD_U64 : DS_1A1D_NORET <0x40, "ds_add_u64", VReg_64>;
def DS_SUB_U64 : DS_1A1D_NORET <0x41, "ds_sub_u64", VReg_64>;
def DS_RSUB_U64 : DS_1A1D_NORET <0x42, "ds_rsub_u64", VReg_64>;
def DS_INC_U64 : DS_1A1D_NORET <0x43, "ds_inc_u64", VReg_64>;
def DS_DEC_U64 : DS_1A1D_NORET <0x44, "ds_dec_u64", VReg_64>;
def DS_MIN_I64 : DS_1A1D_NORET <0x45, "ds_min_i64", VReg_64>;
def DS_MAX_I64 : DS_1A1D_NORET <0x46, "ds_max_i64", VReg_64>;
def DS_MIN_U64 : DS_1A1D_NORET <0x47, "ds_min_u64", VReg_64>;
def DS_MAX_U64 : DS_1A1D_NORET <0x48, "ds_max_u64", VReg_64>;
def DS_AND_B64 : DS_1A1D_NORET <0x49, "ds_and_b64", VReg_64>;
def DS_OR_B64 : DS_1A1D_NORET <0x4a, "ds_or_b64", VReg_64>;
def DS_XOR_B64 : DS_1A1D_NORET <0x4b, "ds_xor_b64", VReg_64>;
def DS_MSKOR_B64 : DS_1A1D_NORET <0x4c, "ds_mskor_b64", VReg_64>;
def DS_CMPST_B64 : DS_1A2D_NORET <0x50, "ds_cmpst_b64", VReg_64>;
def DS_CMPST_F64 : DS_1A2D_NORET <0x51, "ds_cmpst_f64", VReg_64>;
def DS_MIN_F64 : DS_1A1D_NORET <0x52, "ds_min_f64", VReg_64>;
def DS_MAX_F64 : DS_1A1D_NORET <0x53, "ds_max_f64", VReg_64>;

def DS_ADD_RTN_U64 : DS_1A1D_RET <0x60, "ds_add_rtn_u64", VReg_64, "ds_add_u64">;
def DS_SUB_RTN_U64 : DS_1A1D_RET <0x61, "ds_sub_rtn_u64", VReg_64, "ds_sub_u64">;
def DS_RSUB_RTN_U64 : DS_1A1D_RET <0x62, "ds_rsub_rtn_u64", VReg_64, "ds_rsub_u64">;
def DS_INC_RTN_U64 : DS_1A1D_RET <0x63, "ds_inc_rtn_u64", VReg_64, "ds_inc_u64">;
def DS_DEC_RTN_U64 : DS_1A1D_RET <0x64, "ds_dec_rtn_u64", VReg_64, "ds_dec_u64">;
def DS_MIN_RTN_I64 : DS_1A1D_RET <0x65, "ds_min_rtn_i64", VReg_64, "ds_min_i64">;
def DS_MAX_RTN_I64 : DS_1A1D_RET <0x66, "ds_max_rtn_i64", VReg_64, "ds_max_i64">;
def DS_MIN_RTN_U64 : DS_1A1D_RET <0x67, "ds_min_rtn_u64", VReg_64, "ds_min_u64">;
def DS_MAX_RTN_U64 : DS_1A1D_RET <0x68, "ds_max_rtn_u64", VReg_64, "ds_max_u64">;
def DS_AND_RTN_B64 : DS_1A1D_RET <0x69, "ds_and_rtn_b64", VReg_64, "ds_and_b64">;
def DS_OR_RTN_B64 : DS_1A1D_RET <0x6a, "ds_or_rtn_b64", VReg_64, "ds_or_b64">;
def DS_XOR_RTN_B64 : DS_1A1D_RET <0x6b, "ds_xor_rtn_b64", VReg_64, "ds_xor_b64">;
def DS_MSKOR_RTN_B64 : DS_1A1D_RET <0x6c, "ds_mskor_rtn_b64", VReg_64, "ds_mskor_b64">;
def DS_WRXCHG_RTN_B64 : DS_1A1D_RET <0x6d, "ds_wrxchg_rtn_b64", VReg_64, "ds_wrxchg_b64">;
//def DS_WRXCHG2_RTN_B64 : DS_2A0D_RET <0x6e, "ds_wrxchg2_rtn_b64", VReg_64, "ds_wrxchg2_b64">;
//def DS_WRXCHG2ST64_RTN_B64 : DS_2A0D_RET <0x6f, "ds_wrxchg2_rtn_b64", VReg_64, "ds_wrxchg2st64_b64">;
def DS_CMPST_RTN_B64 : DS_1A2D_RET <0x70, "ds_cmpst_rtn_b64", VReg_64, "ds_cmpst_b64">;
def DS_CMPST_RTN_F64 : DS_1A2D_RET <0x71, "ds_cmpst_rtn_f64", VReg_64, "ds_cmpst_f64">;
def DS_MIN_RTN_F64 : DS_1A1D_RET <0x72, "ds_min_f64", VReg_64, "ds_min_f64">;
def DS_MAX_RTN_F64 : DS_1A1D_RET <0x73, "ds_max_f64", VReg_64, "ds_max_f64">;

//let SubtargetPredicate = isCI in {
// DS_CONDXCHG32_RTN_B64
// DS_CONDXCHG32_RTN_B128
//} // End isCI

// TODO: _SRC2_* forms

def DS_WRITE_B32 : DS_Store_Helper <0x0000000d, "ds_write_b32", VReg_32>;
def DS_WRITE_B8 : DS_Store_Helper <0x00000001e, "ds_write_b8", VReg_32>;
def DS_WRITE_B16 : DS_Store_Helper <0x00000001f, "ds_write_b16", VReg_32>;
def DS_WRITE_B64 : DS_Store_Helper <0x00000004d, "ds_write_b64", VReg_64>;

def DS_READ_B32 : DS_Load_Helper <0x00000036, "ds_read_b32", VReg_32>;
def DS_READ_I8 : DS_Load_Helper <0x00000039, "ds_read_i8", VReg_32>;
def DS_READ_U8 : DS_Load_Helper <0x0000003a, "ds_read_u8", VReg_32>;
def DS_READ_I16 : DS_Load_Helper <0x0000003b, "ds_read_i16", VReg_32>;
def DS_READ_U16 : DS_Load_Helper <0x0000003c, "ds_read_u16", VReg_32>;
def DS_READ_B64 : DS_Load_Helper <0x00000076, "ds_read_b64", VReg_64>;

// 2 forms.
def DS_WRITE2_B32 : DS_Store2_Helper <0x0000000E, "ds_write2_b32", VReg_32>;
def DS_WRITE2ST64_B32 : DS_Store2_Helper <0x0000000F, "ds_write2st64_b32", VReg_32>;
def DS_WRITE2_B64 : DS_Store2_Helper <0x0000004E, "ds_write2_b64", VReg_64>;
def DS_WRITE2ST64_B64 : DS_Store2_Helper <0x0000004F, "ds_write2st64_b64", VReg_64>;

def DS_READ2_B32 : DS_Load2_Helper <0x00000037, "ds_read2_b32", VReg_64>;
def DS_READ2ST64_B32 : DS_Load2_Helper <0x00000038, "ds_read2st64_b32", VReg_64>;
def DS_READ2_B64 : DS_Load2_Helper <0x00000075, "ds_read2_b64", VReg_128>;
def DS_READ2ST64_B64 : DS_Load2_Helper <0x00000076, "ds_read2st64_b64", VReg_128>;

//===----------------------------------------------------------------------===//
// MUBUF Instructions
//===----------------------------------------------------------------------===//

//def BUFFER_LOAD_FORMAT_X : MUBUF_ <0x00000000, "buffer_load_format_x", []>;
//def BUFFER_LOAD_FORMAT_XY : MUBUF_ <0x00000001, "buffer_load_format_xy", []>;
//def BUFFER_LOAD_FORMAT_XYZ : MUBUF_ <0x00000002, "buffer_load_format_xyz", []>;
defm BUFFER_LOAD_FORMAT_XYZW : MUBUF_Load_Helper <0x00000003, "buffer_load_format_xyzw", VReg_128>;
//def BUFFER_STORE_FORMAT_X : MUBUF_ <0x00000004, "buffer_store_format_x", []>;
//def BUFFER_STORE_FORMAT_XY : MUBUF_ <0x00000005, "buffer_store_format_xy", []>;
//def BUFFER_STORE_FORMAT_XYZ : MUBUF_ <0x00000006, "buffer_store_format_xyz", []>;
//def BUFFER_STORE_FORMAT_XYZW : MUBUF_ <0x00000007, "buffer_store_format_xyzw", []>;
defm BUFFER_LOAD_UBYTE : MUBUF_Load_Helper <
  0x00000008, "buffer_load_ubyte", VReg_32, i32, az_extloadi8_global
>;
defm BUFFER_LOAD_SBYTE : MUBUF_Load_Helper <
  0x00000009, "buffer_load_sbyte", VReg_32, i32, sextloadi8_global
>;
defm BUFFER_LOAD_USHORT : MUBUF_Load_Helper <
  0x0000000a, "buffer_load_ushort", VReg_32, i32, az_extloadi16_global
>;
defm BUFFER_LOAD_SSHORT : MUBUF_Load_Helper <
  0x0000000b, "buffer_load_sshort", VReg_32, i32, sextloadi16_global
>;
defm BUFFER_LOAD_DWORD : MUBUF_Load_Helper <
  0x0000000c, "buffer_load_dword", VReg_32, i32, global_load
>;
defm BUFFER_LOAD_DWORDX2 : MUBUF_Load_Helper <
  0x0000000d, "buffer_load_dwordx2", VReg_64, v2i32, global_load
>;
defm BUFFER_LOAD_DWORDX4 : MUBUF_Load_Helper <
  0x0000000e, "buffer_load_dwordx4", VReg_128, v4i32, global_load
>;

defm BUFFER_STORE_BYTE : MUBUF_Store_Helper <
  0x00000018, "buffer_store_byte", VReg_32, i32, truncstorei8_global
>;

defm BUFFER_STORE_SHORT : MUBUF_Store_Helper <
  0x0000001a, "buffer_store_short", VReg_32, i32, truncstorei16_global
>;

defm BUFFER_STORE_DWORD : MUBUF_Store_Helper <
  0x0000001c, "buffer_store_dword", VReg_32, i32, global_store
>;

defm BUFFER_STORE_DWORDX2 : MUBUF_Store_Helper <
  0x0000001d, "buffer_store_dwordx2", VReg_64, v2i32, global_store
>;

defm BUFFER_STORE_DWORDX4 : MUBUF_Store_Helper <
  0x0000001e, "buffer_store_dwordx4", VReg_128, v4i32, global_store
>;
//def BUFFER_ATOMIC_SWAP : MUBUF_ <0x00000030, "buffer_atomic_swap", []>;
defm BUFFER_ATOMIC_SWAP : MUBUF_Atomic <
  0x00000030, "buffer_atomic_swap", VReg_32, i32, atomic_swap_global
>;
//def BUFFER_ATOMIC_CMPSWAP : MUBUF_ <0x00000031, "buffer_atomic_cmpswap", []>;
defm BUFFER_ATOMIC_ADD : MUBUF_Atomic <
  0x00000032, "buffer_atomic_add", VReg_32, i32, atomic_add_global
>;
defm BUFFER_ATOMIC_SUB : MUBUF_Atomic <
  0x00000033, "buffer_atomic_sub", VReg_32, i32, atomic_sub_global
>;
//def BUFFER_ATOMIC_RSUB : MUBUF_ <0x00000034, "buffer_atomic_rsub", []>;
defm BUFFER_ATOMIC_SMIN : MUBUF_Atomic <
  0x00000035, "buffer_atomic_smin", VReg_32, i32, atomic_min_global
>;
defm BUFFER_ATOMIC_UMIN : MUBUF_Atomic <
  0x00000036, "buffer_atomic_umin", VReg_32, i32, atomic_umin_global
>;
defm BUFFER_ATOMIC_SMAX : MUBUF_Atomic <
  0x00000037, "buffer_atomic_smax", VReg_32, i32, atomic_max_global
>;
defm BUFFER_ATOMIC_UMAX : MUBUF_Atomic <
  0x00000038, "buffer_atomic_umax", VReg_32, i32, atomic_umax_global
>;
defm BUFFER_ATOMIC_AND : MUBUF_Atomic <
  0x00000039, "buffer_atomic_and", VReg_32, i32, atomic_and_global
>;
defm BUFFER_ATOMIC_OR : MUBUF_Atomic <
  0x0000003a, "buffer_atomic_or", VReg_32, i32, atomic_or_global
>;
defm BUFFER_ATOMIC_XOR : MUBUF_Atomic <
  0x0000003b, "buffer_atomic_xor", VReg_32, i32, atomic_xor_global
>;
//def BUFFER_ATOMIC_INC : MUBUF_ <0x0000003c, "buffer_atomic_inc", []>;
//def BUFFER_ATOMIC_DEC : MUBUF_ <0x0000003d, "buffer_atomic_dec", []>;
//def BUFFER_ATOMIC_FCMPSWAP : MUBUF_ <0x0000003e, "buffer_atomic_fcmpswap", []>;
//def BUFFER_ATOMIC_FMIN : MUBUF_ <0x0000003f, "buffer_atomic_fmin", []>;
//def BUFFER_ATOMIC_FMAX : MUBUF_ <0x00000040, "buffer_atomic_fmax", []>;
//def BUFFER_ATOMIC_SWAP_X2 : MUBUF_X2 <0x00000050, "buffer_atomic_swap_x2", []>;
//def BUFFER_ATOMIC_CMPSWAP_X2 : MUBUF_X2 <0x00000051, "buffer_atomic_cmpswap_x2", []>;
//def BUFFER_ATOMIC_ADD_X2 : MUBUF_X2 <0x00000052, "buffer_atomic_add_x2", []>;
//def BUFFER_ATOMIC_SUB_X2 : MUBUF_X2 <0x00000053, "buffer_atomic_sub_x2", []>;
//def BUFFER_ATOMIC_RSUB_X2 : MUBUF_X2 <0x00000054, "buffer_atomic_rsub_x2", []>;
//def BUFFER_ATOMIC_SMIN_X2 : MUBUF_X2 <0x00000055, "buffer_atomic_smin_x2", []>;
//def BUFFER_ATOMIC_UMIN_X2 : MUBUF_X2 <0x00000056, "buffer_atomic_umin_x2", []>;
//def BUFFER_ATOMIC_SMAX_X2 : MUBUF_X2 <0x00000057, "buffer_atomic_smax_x2", []>;
//def BUFFER_ATOMIC_UMAX_X2 : MUBUF_X2 <0x00000058, "buffer_atomic_umax_x2", []>;
//def BUFFER_ATOMIC_AND_X2 : MUBUF_X2 <0x00000059, "buffer_atomic_and_x2", []>;
//def BUFFER_ATOMIC_OR_X2 : MUBUF_X2 <0x0000005a, "buffer_atomic_or_x2", []>;
//def BUFFER_ATOMIC_XOR_X2 : MUBUF_X2 <0x0000005b, "buffer_atomic_xor_x2", []>;
//def BUFFER_ATOMIC_INC_X2 : MUBUF_X2 <0x0000005c, "buffer_atomic_inc_x2", []>;
//def BUFFER_ATOMIC_DEC_X2 : MUBUF_X2 <0x0000005d, "buffer_atomic_dec_x2", []>;
//def BUFFER_ATOMIC_FCMPSWAP_X2 : MUBUF_X2 <0x0000005e, "buffer_atomic_fcmpswap_x2", []>;
//def BUFFER_ATOMIC_FMIN_X2 : MUBUF_X2 <0x0000005f, "buffer_atomic_fmin_x2", []>;
//def BUFFER_ATOMIC_FMAX_X2 : MUBUF_X2 <0x00000060, "buffer_atomic_fmax_x2", []>;
//def BUFFER_WBINVL1_SC : MUBUF_WBINVL1 <0x00000070, "buffer_wbinvl1_sc", []>;
//def BUFFER_WBINVL1 : MUBUF_WBINVL1 <0x00000071, "buffer_wbinvl1", []>;

//===----------------------------------------------------------------------===//
// MTBUF Instructions
//===----------------------------------------------------------------------===//

//def TBUFFER_LOAD_FORMAT_X : MTBUF_ <0x00000000, "tbuffer_load_format_x", []>;
//def TBUFFER_LOAD_FORMAT_XY : MTBUF_ <0x00000001, "tbuffer_load_format_xy", []>;
//def TBUFFER_LOAD_FORMAT_XYZ : MTBUF_ <0x00000002, "tbuffer_load_format_xyz", []>;
defm TBUFFER_LOAD_FORMAT_XYZW : MTBUF_Load_Helper <0x00000003, "tbuffer_load_format_xyzw", VReg_128>;
defm TBUFFER_STORE_FORMAT_X : MTBUF_Store_Helper <0x00000004, "tbuffer_store_format_x", VReg_32>;
defm TBUFFER_STORE_FORMAT_XY : MTBUF_Store_Helper <0x00000005, "tbuffer_store_format_xy", VReg_64>;
defm TBUFFER_STORE_FORMAT_XYZ : MTBUF_Store_Helper <0x00000006, "tbuffer_store_format_xyz", VReg_128>;
defm TBUFFER_STORE_FORMAT_XYZW : MTBUF_Store_Helper <0x00000007, "tbuffer_store_format_xyzw", VReg_128>;

//===----------------------------------------------------------------------===//
// MIMG Instructions
//===----------------------------------------------------------------------===//

defm IMAGE_LOAD : MIMG_NoSampler <0x00000000, "image_load">;
defm IMAGE_LOAD_MIP : MIMG_NoSampler <0x00000001, "image_load_mip">;
//def IMAGE_LOAD_PCK : MIMG_NoPattern_ <"image_load_pck", 0x00000002>;
//def IMAGE_LOAD_PCK_SGN : MIMG_NoPattern_ <"image_load_pck_sgn", 0x00000003>;
//def IMAGE_LOAD_MIP_PCK : MIMG_NoPattern_ <"image_load_mip_pck", 0x00000004>;
//def IMAGE_LOAD_MIP_PCK_SGN : MIMG_NoPattern_ <"image_load_mip_pck_sgn", 0x00000005>;
//def IMAGE_STORE : MIMG_NoPattern_ <"image_store", 0x00000008>;
//def IMAGE_STORE_MIP : MIMG_NoPattern_ <"image_store_mip", 0x00000009>;
//def IMAGE_STORE_PCK : MIMG_NoPattern_ <"image_store_pck", 0x0000000a>;
//def IMAGE_STORE_MIP_PCK : MIMG_NoPattern_ <"image_store_mip_pck", 0x0000000b>;
defm IMAGE_GET_RESINFO : MIMG_NoSampler <0x0000000e, "image_get_resinfo">;
//def IMAGE_ATOMIC_SWAP : MIMG_NoPattern_ <"image_atomic_swap", 0x0000000f>;
//def IMAGE_ATOMIC_CMPSWAP : MIMG_NoPattern_ <"image_atomic_cmpswap", 0x00000010>;
//def IMAGE_ATOMIC_ADD : MIMG_NoPattern_ <"image_atomic_add", 0x00000011>;
//def IMAGE_ATOMIC_SUB : MIMG_NoPattern_ <"image_atomic_sub", 0x00000012>;
//def IMAGE_ATOMIC_RSUB : MIMG_NoPattern_ <"image_atomic_rsub", 0x00000013>;
//def IMAGE_ATOMIC_SMIN : MIMG_NoPattern_ <"image_atomic_smin", 0x00000014>;
//def IMAGE_ATOMIC_UMIN : MIMG_NoPattern_ <"image_atomic_umin", 0x00000015>;
//def IMAGE_ATOMIC_SMAX : MIMG_NoPattern_ <"image_atomic_smax", 0x00000016>;
//def IMAGE_ATOMIC_UMAX : MIMG_NoPattern_ <"image_atomic_umax", 0x00000017>;
//def IMAGE_ATOMIC_AND : MIMG_NoPattern_ <"image_atomic_and", 0x00000018>;
//def IMAGE_ATOMIC_OR : MIMG_NoPattern_ <"image_atomic_or", 0x00000019>;
//def IMAGE_ATOMIC_XOR : MIMG_NoPattern_ <"image_atomic_xor", 0x0000001a>;
//def IMAGE_ATOMIC_INC : MIMG_NoPattern_ <"image_atomic_inc", 0x0000001b>;
//def IMAGE_ATOMIC_DEC : MIMG_NoPattern_ <"image_atomic_dec", 0x0000001c>;
//def IMAGE_ATOMIC_FCMPSWAP : MIMG_NoPattern_ <"image_atomic_fcmpswap", 0x0000001d>;
//def IMAGE_ATOMIC_FMIN : MIMG_NoPattern_ <"image_atomic_fmin", 0x0000001e>;
//def IMAGE_ATOMIC_FMAX : MIMG_NoPattern_ <"image_atomic_fmax", 0x0000001f>;
defm IMAGE_SAMPLE           : MIMG_Sampler <0x00000020, "image_sample">;
defm IMAGE_SAMPLE_CL        : MIMG_Sampler <0x00000021, "image_sample_cl">;
defm IMAGE_SAMPLE_D         : MIMG_Sampler <0x00000022, "image_sample_d">;
defm IMAGE_SAMPLE_D_CL      : MIMG_Sampler <0x00000023, "image_sample_d_cl">;
defm IMAGE_SAMPLE_L         : MIMG_Sampler <0x00000024, "image_sample_l">;
defm IMAGE_SAMPLE_B         : MIMG_Sampler <0x00000025, "image_sample_b">;
defm IMAGE_SAMPLE_B_CL      : MIMG_Sampler <0x00000026, "image_sample_b_cl">;
defm IMAGE_SAMPLE_LZ        : MIMG_Sampler <0x00000027, "image_sample_lz">;
defm IMAGE_SAMPLE_C         : MIMG_Sampler <0x00000028, "image_sample_c">;
defm IMAGE_SAMPLE_C_CL      : MIMG_Sampler <0x00000029, "image_sample_c_cl">;
defm IMAGE_SAMPLE_C_D       : MIMG_Sampler <0x0000002a, "image_sample_c_d">;
defm IMAGE_SAMPLE_C_D_CL    : MIMG_Sampler <0x0000002b, "image_sample_c_d_cl">;
defm IMAGE_SAMPLE_C_L       : MIMG_Sampler <0x0000002c, "image_sample_c_l">;
defm IMAGE_SAMPLE_C_B       : MIMG_Sampler <0x0000002d, "image_sample_c_b">;
defm IMAGE_SAMPLE_C_B_CL    : MIMG_Sampler <0x0000002e, "image_sample_c_b_cl">;
defm IMAGE_SAMPLE_C_LZ      : MIMG_Sampler <0x0000002f, "image_sample_c_lz">;
defm IMAGE_SAMPLE_O         : MIMG_Sampler <0x00000030, "image_sample_o">;
defm IMAGE_SAMPLE_CL_O      : MIMG_Sampler <0x00000031, "image_sample_cl_o">;
defm IMAGE_SAMPLE_D_O       : MIMG_Sampler <0x00000032, "image_sample_d_o">;
defm IMAGE_SAMPLE_D_CL_O    : MIMG_Sampler <0x00000033, "image_sample_d_cl_o">;
defm IMAGE_SAMPLE_L_O       : MIMG_Sampler <0x00000034, "image_sample_l_o">;
defm IMAGE_SAMPLE_B_O       : MIMG_Sampler <0x00000035, "image_sample_b_o">;
defm IMAGE_SAMPLE_B_CL_O    : MIMG_Sampler <0x00000036, "image_sample_b_cl_o">;
defm IMAGE_SAMPLE_LZ_O      : MIMG_Sampler <0x00000037, "image_sample_lz_o">;
defm IMAGE_SAMPLE_C_O       : MIMG_Sampler <0x00000038, "image_sample_c_o">;
defm IMAGE_SAMPLE_C_CL_O    : MIMG_Sampler <0x00000039, "image_sample_c_cl_o">;
defm IMAGE_SAMPLE_C_D_O     : MIMG_Sampler <0x0000003a, "image_sample_c_d_o">;
defm IMAGE_SAMPLE_C_D_CL_O  : MIMG_Sampler <0x0000003b, "image_sample_c_d_cl_o">;
defm IMAGE_SAMPLE_C_L_O     : MIMG_Sampler <0x0000003c, "image_sample_c_l_o">;
defm IMAGE_SAMPLE_C_B_O     : MIMG_Sampler <0x0000003d, "image_sample_c_b_o">;
defm IMAGE_SAMPLE_C_B_CL_O  : MIMG_Sampler <0x0000003e, "image_sample_c_b_cl_o">;
defm IMAGE_SAMPLE_C_LZ_O    : MIMG_Sampler <0x0000003f, "image_sample_c_lz_o">;
defm IMAGE_GATHER4          : MIMG_Gather <0x00000040, "image_gather4">;
defm IMAGE_GATHER4_CL       : MIMG_Gather <0x00000041, "image_gather4_cl">;
defm IMAGE_GATHER4_L        : MIMG_Gather <0x00000044, "image_gather4_l">;
defm IMAGE_GATHER4_B        : MIMG_Gather <0x00000045, "image_gather4_b">;
defm IMAGE_GATHER4_B_CL     : MIMG_Gather <0x00000046, "image_gather4_b_cl">;
defm IMAGE_GATHER4_LZ       : MIMG_Gather <0x00000047, "image_gather4_lz">;
defm IMAGE_GATHER4_C        : MIMG_Gather <0x00000048, "image_gather4_c">;
defm IMAGE_GATHER4_C_CL     : MIMG_Gather <0x00000049, "image_gather4_c_cl">;
defm IMAGE_GATHER4_C_L      : MIMG_Gather <0x0000004c, "image_gather4_c_l">;
defm IMAGE_GATHER4_C_B      : MIMG_Gather <0x0000004d, "image_gather4_c_b">;
defm IMAGE_GATHER4_C_B_CL   : MIMG_Gather <0x0000004e, "image_gather4_c_b_cl">;
defm IMAGE_GATHER4_C_LZ     : MIMG_Gather <0x0000004f, "image_gather4_c_lz">;
defm IMAGE_GATHER4_O        : MIMG_Gather <0x00000050, "image_gather4_o">;
defm IMAGE_GATHER4_CL_O     : MIMG_Gather <0x00000051, "image_gather4_cl_o">;
defm IMAGE_GATHER4_L_O      : MIMG_Gather <0x00000054, "image_gather4_l_o">;
defm IMAGE_GATHER4_B_O      : MIMG_Gather <0x00000055, "image_gather4_b_o">;
defm IMAGE_GATHER4_B_CL_O   : MIMG_Gather <0x00000056, "image_gather4_b_cl_o">;
defm IMAGE_GATHER4_LZ_O     : MIMG_Gather <0x00000057, "image_gather4_lz_o">;
defm IMAGE_GATHER4_C_O      : MIMG_Gather <0x00000058, "image_gather4_c_o">;
defm IMAGE_GATHER4_C_CL_O   : MIMG_Gather <0x00000059, "image_gather4_c_cl_o">;
defm IMAGE_GATHER4_C_L_O    : MIMG_Gather <0x0000005c, "image_gather4_c_l_o">;
defm IMAGE_GATHER4_C_B_O    : MIMG_Gather <0x0000005d, "image_gather4_c_b_o">;
defm IMAGE_GATHER4_C_B_CL_O : MIMG_Gather <0x0000005e, "image_gather4_c_b_cl_o">;
defm IMAGE_GATHER4_C_LZ_O   : MIMG_Gather <0x0000005f, "image_gather4_c_lz_o">;
defm IMAGE_GET_LOD          : MIMG_Sampler <0x00000060, "image_get_lod">;
defm IMAGE_SAMPLE_CD        : MIMG_Sampler <0x00000068, "image_sample_cd">;
defm IMAGE_SAMPLE_CD_CL     : MIMG_Sampler <0x00000069, "image_sample_cd_cl">;
defm IMAGE_SAMPLE_C_CD      : MIMG_Sampler <0x0000006a, "image_sample_c_cd">;
defm IMAGE_SAMPLE_C_CD_CL   : MIMG_Sampler <0x0000006b, "image_sample_c_cd_cl">;
defm IMAGE_SAMPLE_CD_O      : MIMG_Sampler <0x0000006c, "image_sample_cd_o">;
defm IMAGE_SAMPLE_CD_CL_O   : MIMG_Sampler <0x0000006d, "image_sample_cd_cl_o">;
defm IMAGE_SAMPLE_C_CD_O    : MIMG_Sampler <0x0000006e, "image_sample_c_cd_o">;
defm IMAGE_SAMPLE_C_CD_CL_O : MIMG_Sampler <0x0000006f, "image_sample_c_cd_cl_o">;
//def IMAGE_RSRC256 : MIMG_NoPattern_RSRC256 <"image_rsrc256", 0x0000007e>;
//def IMAGE_SAMPLER : MIMG_NoPattern_ <"image_sampler", 0x0000007f>;

//===----------------------------------------------------------------------===//
// Flat Instructions
//===----------------------------------------------------------------------===//

let Predicates = [HasFlatAddressSpace] in {
def FLAT_LOAD_UBYTE : FLAT_Load_Helper <0x00000008, "flat_load_ubyte", VReg_32>;
def FLAT_LOAD_SBYTE : FLAT_Load_Helper <0x00000009, "flat_load_sbyte", VReg_32>;
def FLAT_LOAD_USHORT : FLAT_Load_Helper <0x0000000a, "flat_load_ushort", VReg_32>;
def FLAT_LOAD_SSHORT : FLAT_Load_Helper <0x0000000b, "flat_load_sshort", VReg_32>;
def FLAT_LOAD_DWORD : FLAT_Load_Helper <0x0000000c, "flat_load_dword", VReg_32>;
def FLAT_LOAD_DWORDX2 : FLAT_Load_Helper <0x0000000d, "flat_load_dwordx2", VReg_64>;
def FLAT_LOAD_DWORDX4 : FLAT_Load_Helper <0x0000000e, "flat_load_dwordx4", VReg_128>;
def FLAT_LOAD_DWORDX3 : FLAT_Load_Helper <0x00000010, "flat_load_dwordx3", VReg_96>;

def FLAT_STORE_BYTE : FLAT_Store_Helper <
  0x00000018, "flat_store_byte", VReg_32
>;

def FLAT_STORE_SHORT : FLAT_Store_Helper <
  0x0000001a, "flat_store_short", VReg_32
>;

def FLAT_STORE_DWORD : FLAT_Store_Helper <
  0x0000001c, "flat_store_dword", VReg_32
>;

def FLAT_STORE_DWORDX2 : FLAT_Store_Helper <
  0x0000001d, "flat_store_dwordx2", VReg_64
>;

def FLAT_STORE_DWORDX4 : FLAT_Store_Helper <
  0x0000001e, "flat_store_dwordx4", VReg_128
>;

def FLAT_STORE_DWORDX3 : FLAT_Store_Helper <
  0x0000001e, "flat_store_dwordx3", VReg_96
>;

//def FLAT_ATOMIC_SWAP : FLAT_ <0x00000030, "flat_atomic_swap", []>;
//def FLAT_ATOMIC_CMPSWAP : FLAT_ <0x00000031, "flat_atomic_cmpswap", []>;
//def FLAT_ATOMIC_ADD : FLAT_ <0x00000032, "flat_atomic_add", []>;
//def FLAT_ATOMIC_SUB : FLAT_ <0x00000033, "flat_atomic_sub", []>;
//def FLAT_ATOMIC_RSUB : FLAT_ <0x00000034, "flat_atomic_rsub", []>;
//def FLAT_ATOMIC_SMIN : FLAT_ <0x00000035, "flat_atomic_smin", []>;
//def FLAT_ATOMIC_UMIN : FLAT_ <0x00000036, "flat_atomic_umin", []>;
//def FLAT_ATOMIC_SMAX : FLAT_ <0x00000037, "flat_atomic_smax", []>;
//def FLAT_ATOMIC_UMAX : FLAT_ <0x00000038, "flat_atomic_umax", []>;
//def FLAT_ATOMIC_AND : FLAT_ <0x00000039, "flat_atomic_and", []>;
//def FLAT_ATOMIC_OR : FLAT_ <0x0000003a, "flat_atomic_or", []>;
//def FLAT_ATOMIC_XOR : FLAT_ <0x0000003b, "flat_atomic_xor", []>;
//def FLAT_ATOMIC_INC : FLAT_ <0x0000003c, "flat_atomic_inc", []>;
//def FLAT_ATOMIC_DEC : FLAT_ <0x0000003d, "flat_atomic_dec", []>;
//def FLAT_ATOMIC_FCMPSWAP : FLAT_ <0x0000003e, "flat_atomic_fcmpswap", []>;
//def FLAT_ATOMIC_FMIN : FLAT_ <0x0000003f, "flat_atomic_fmin", []>;
//def FLAT_ATOMIC_FMAX : FLAT_ <0x00000040, "flat_atomic_fmax", []>;
//def FLAT_ATOMIC_SWAP_X2 : FLAT_X2 <0x00000050, "flat_atomic_swap_x2", []>;
//def FLAT_ATOMIC_CMPSWAP_X2 : FLAT_X2 <0x00000051, "flat_atomic_cmpswap_x2", []>;
//def FLAT_ATOMIC_ADD_X2 : FLAT_X2 <0x00000052, "flat_atomic_add_x2", []>;
//def FLAT_ATOMIC_SUB_X2 : FLAT_X2 <0x00000053, "flat_atomic_sub_x2", []>;
//def FLAT_ATOMIC_RSUB_X2 : FLAT_X2 <0x00000054, "flat_atomic_rsub_x2", []>;
//def FLAT_ATOMIC_SMIN_X2 : FLAT_X2 <0x00000055, "flat_atomic_smin_x2", []>;
//def FLAT_ATOMIC_UMIN_X2 : FLAT_X2 <0x00000056, "flat_atomic_umin_x2", []>;
//def FLAT_ATOMIC_SMAX_X2 : FLAT_X2 <0x00000057, "flat_atomic_smax_x2", []>;
//def FLAT_ATOMIC_UMAX_X2 : FLAT_X2 <0x00000058, "flat_atomic_umax_x2", []>;
//def FLAT_ATOMIC_AND_X2 : FLAT_X2 <0x00000059, "flat_atomic_and_x2", []>;
//def FLAT_ATOMIC_OR_X2 : FLAT_X2 <0x0000005a, "flat_atomic_or_x2", []>;
//def FLAT_ATOMIC_XOR_X2 : FLAT_X2 <0x0000005b, "flat_atomic_xor_x2", []>;
//def FLAT_ATOMIC_INC_X2 : FLAT_X2 <0x0000005c, "flat_atomic_inc_x2", []>;
//def FLAT_ATOMIC_DEC_X2 : FLAT_X2 <0x0000005d, "flat_atomic_dec_x2", []>;
//def FLAT_ATOMIC_FCMPSWAP_X2 : FLAT_X2 <0x0000005e, "flat_atomic_fcmpswap_x2", []>;
//def FLAT_ATOMIC_FMIN_X2 : FLAT_X2 <0x0000005f, "flat_atomic_fmin_x2", []>;
//def FLAT_ATOMIC_FMAX_X2 : FLAT_X2 <0x00000060, "flat_atomic_fmax_x2", []>;

} // End HasFlatAddressSpace predicate
//===----------------------------------------------------------------------===//
// VOP1 Instructions
//===----------------------------------------------------------------------===//

//def V_NOP : VOP1_ <0x00000000, "v_nop", []>;

let isMoveImm = 1 in {
defm V_MOV_B32 : VOP1Inst <vop1<0x1>, "v_mov_b32", VOP_I32_I32>;
} // End isMoveImm = 1

let Uses = [EXEC] in {

def V_READFIRSTLANE_B32 : VOP1 <
  0x00000002,
  (outs SReg_32:$vdst),
  (ins VReg_32:$src0),
  "v_readfirstlane_b32 $vdst, $src0",
  []
>;

}

defm V_CVT_I32_F64 : VOP1Inst <vop1<0x3>, "v_cvt_i32_f64",
  VOP_I32_F64, fp_to_sint
>;
defm V_CVT_F64_I32 : VOP1Inst <vop1<0x4>, "v_cvt_f64_i32",
  VOP_F64_I32, sint_to_fp
>;
defm V_CVT_F32_I32 : VOP1Inst <vop1<0x5>, "v_cvt_f32_i32",
  VOP_F32_I32, sint_to_fp
>;
defm V_CVT_F32_U32 : VOP1Inst <vop1<0x6>, "v_cvt_f32_u32",
  VOP_F32_I32, uint_to_fp
>;
defm V_CVT_U32_F32 : VOP1Inst <vop1<0x7>, "v_cvt_u32_f32",
  VOP_I32_F32, fp_to_uint
>;
defm V_CVT_I32_F32 : VOP1Inst <vop1<0x8>, "v_cvt_i32_f32",
  VOP_I32_F32, fp_to_sint
>;
defm V_MOV_FED_B32 : VOP1Inst <vop1<0x9>, "v_mov_fed_b32", VOP_I32_I32>;
defm V_CVT_F16_F32 : VOP1Inst <vop1<0xa>, "v_cvt_f16_f32",
  VOP_I32_F32, fp_to_f16
>;
defm V_CVT_F32_F16 : VOP1Inst <vop1<0xb>, "v_cvt_f32_f16",
  VOP_F32_I32, f16_to_fp
>;
//defm V_CVT_RPI_I32_F32 : VOP1_32 <0x0000000c, "v_cvt_rpi_i32_f32", []>;
//defm V_CVT_FLR_I32_F32 : VOP1_32 <0x0000000d, "v_cvt_flr_i32_f32", []>;
//defm V_CVT_OFF_F32_I4 : VOP1_32 <0x0000000e, "v_cvt_off_f32_i4", []>;
defm V_CVT_F32_F64 : VOP1Inst <vop1<0xf>, "v_cvt_f32_f64",
  VOP_F32_F64, fround
>;
defm V_CVT_F64_F32 : VOP1Inst <vop1<0x10>, "v_cvt_f64_f32",
  VOP_F64_F32, fextend
>;
defm V_CVT_F32_UBYTE0 : VOP1Inst <vop1<0x11>, "v_cvt_f32_ubyte0",
  VOP_F32_I32, AMDGPUcvt_f32_ubyte0
>;
defm V_CVT_F32_UBYTE1 : VOP1Inst <vop1<0x12>, "v_cvt_f32_ubyte1",
  VOP_F32_I32, AMDGPUcvt_f32_ubyte1
>;
defm V_CVT_F32_UBYTE2 : VOP1Inst <vop1<0x13>, "v_cvt_f32_ubyte2",
  VOP_F32_I32, AMDGPUcvt_f32_ubyte2
>;
defm V_CVT_F32_UBYTE3 : VOP1Inst <vop1<0x14>, "v_cvt_f32_ubyte3",
  VOP_F32_I32, AMDGPUcvt_f32_ubyte3
>;
defm V_CVT_U32_F64 : VOP1Inst <vop1<0x15>, "v_cvt_u32_f64",
  VOP_I32_F64, fp_to_uint
>;
defm V_CVT_F64_U32 : VOP1Inst <vop1<0x16>, "v_cvt_f64_u32",
  VOP_F64_I32, uint_to_fp
>;

defm V_FRACT_F32 : VOP1Inst <vop1<0x20>, "v_fract_f32",
  VOP_F32_F32, AMDGPUfract
>;
defm V_TRUNC_F32 : VOP1Inst <vop1<0x21>, "v_trunc_f32",
  VOP_F32_F32, ftrunc
>;
defm V_CEIL_F32 : VOP1Inst <vop1<0x22>, "v_ceil_f32",
  VOP_F32_F32, fceil
>;
defm V_RNDNE_F32 : VOP1Inst <vop1<0x23>, "v_rndne_f32",
  VOP_F32_F32, frint
>;
defm V_FLOOR_F32 : VOP1Inst <vop1<0x24>, "v_floor_f32",
  VOP_F32_F32, ffloor
>;
defm V_EXP_F32 : VOP1Inst <vop1<0x25>, "v_exp_f32",
  VOP_F32_F32, fexp2
>;
defm V_LOG_CLAMP_F32 : VOP1Inst <vop1<0x26>, "v_log_clamp_f32", VOP_F32_F32>;
defm V_LOG_F32 : VOP1Inst <vop1<0x27>, "v_log_f32",
  VOP_F32_F32, flog2
>;

defm V_RCP_CLAMP_F32 : VOP1Inst <vop1<0x28>, "v_rcp_clamp_f32", VOP_F32_F32>;
defm V_RCP_LEGACY_F32 : VOP1Inst <vop1<0x29>, "v_rcp_legacy_f32", VOP_F32_F32>;
defm V_RCP_F32 : VOP1Inst <vop1<0x2a>, "v_rcp_f32",
  VOP_F32_F32, AMDGPUrcp
>;
defm V_RCP_IFLAG_F32 : VOP1Inst <vop1<0x2b>, "v_rcp_iflag_f32", VOP_F32_F32>;
defm V_RSQ_CLAMP_F32 : VOP1Inst <vop1<0x2c>, "v_rsq_clamp_f32",
  VOP_F32_F32, AMDGPUrsq_clamped
>;
defm V_RSQ_LEGACY_F32 : VOP1Inst <vop1<0x2d>, "v_rsq_legacy_f32",
  VOP_F32_F32, AMDGPUrsq_legacy
>;
defm V_RSQ_F32 : VOP1Inst <vop1<0x2e>, "v_rsq_f32",
  VOP_F32_F32, AMDGPUrsq
>;
defm V_RCP_F64 : VOP1Inst <vop1<0x2f>, "v_rcp_f64",
  VOP_F64_F64, AMDGPUrcp
>;
defm V_RCP_CLAMP_F64 : VOP1Inst <vop1<0x30>, "v_rcp_clamp_f64", VOP_F64_F64>;
defm V_RSQ_F64 : VOP1Inst <vop1<0x31>, "v_rsq_f64",
  VOP_F64_F64, AMDGPUrsq
>;
defm V_RSQ_CLAMP_F64 : VOP1Inst <vop1<0x32>, "v_rsq_clamp_f64",
  VOP_F64_F64, AMDGPUrsq_clamped
>;
defm V_SQRT_F32 : VOP1Inst <vop1<0x33>, "v_sqrt_f32",
  VOP_F32_F32, fsqrt
>;
defm V_SQRT_F64 : VOP1Inst <vop1<0x34>, "v_sqrt_f64",
  VOP_F64_F64, fsqrt
>;
defm V_SIN_F32 : VOP1Inst <vop1<0x35>, "v_sin_f32",
  VOP_F32_F32, AMDGPUsin
>;
defm V_COS_F32 : VOP1Inst <vop1<0x36>, "v_cos_f32",
  VOP_F32_F32, AMDGPUcos
>;
defm V_NOT_B32 : VOP1Inst <vop1<0x37>, "v_not_b32", VOP_I32_I32>;
defm V_BFREV_B32 : VOP1Inst <vop1<0x38>, "v_bfrev_b32", VOP_I32_I32>;
defm V_FFBH_U32 : VOP1Inst <vop1<0x39>, "v_ffbh_u32", VOP_I32_I32>;
defm V_FFBL_B32 : VOP1Inst <vop1<0x3a>, "v_ffbl_b32", VOP_I32_I32>;
defm V_FFBH_I32 : VOP1Inst <vop1<0x3b>, "v_ffbh_i32", VOP_I32_I32>;
//defm V_FREXP_EXP_I32_F64 : VOPInst <0x0000003c, "v_frexp_exp_i32_f64", VOP_I32_F32>;
defm V_FREXP_MANT_F64 : VOP1Inst <vop1<0x3d>, "v_frexp_mant_f64", VOP_F64_F64>;
defm V_FRACT_F64 : VOP1Inst <vop1<0x3e>, "v_fract_f64", VOP_F64_F64>;
//defm V_FREXP_EXP_I32_F32 : VOPInst <0x0000003f, "v_frexp_exp_i32_f32", VOP_I32_F32>;
defm V_FREXP_MANT_F32 : VOP1Inst <vop1<0x40>, "v_frexp_mant_f32", VOP_F32_F32>;
//def V_CLREXCP : VOP1_ <0x00000041, "v_clrexcp", []>;
defm V_MOVRELD_B32 : VOP1Inst <vop1<0x42>, "v_movreld_b32", VOP_I32_I32>;
defm V_MOVRELS_B32 : VOP1Inst <vop1<0x43>, "v_movrels_b32", VOP_I32_I32>;
defm V_MOVRELSD_B32 : VOP1Inst <vop1<0x44>, "v_movrelsd_b32", VOP_I32_I32>;


//===----------------------------------------------------------------------===//
// VINTRP Instructions
//===----------------------------------------------------------------------===//

def V_INTERP_P1_F32 : VINTRP <
  0x00000000,
  (outs VReg_32:$dst),
  (ins VReg_32:$i, i32imm:$attr_chan, i32imm:$attr, M0Reg:$m0),
  "v_interp_p1_f32 $dst, $i, $attr_chan, $attr, [$m0]",
  []> {
  let DisableEncoding = "$m0";
}

def V_INTERP_P2_F32 : VINTRP <
  0x00000001,
  (outs VReg_32:$dst),
  (ins VReg_32:$src0, VReg_32:$j, i32imm:$attr_chan, i32imm:$attr, M0Reg:$m0),
  "v_interp_p2_f32 $dst, [$src0], $j, $attr_chan, $attr, [$m0]",
  []> {

  let Constraints = "$src0 = $dst";
  let DisableEncoding = "$src0,$m0";

}

def V_INTERP_MOV_F32 : VINTRP <
  0x00000002,
  (outs VReg_32:$dst),
  (ins InterpSlot:$src0, i32imm:$attr_chan, i32imm:$attr, M0Reg:$m0),
  "v_interp_mov_f32 $dst, $src0, $attr_chan, $attr, [$m0]",
  []> {
  let DisableEncoding = "$m0";
}

//===----------------------------------------------------------------------===//
// VOP2 Instructions
//===----------------------------------------------------------------------===//

def V_CNDMASK_B32_e32 : VOP2 <0x00000000, (outs VReg_32:$dst),
  (ins VSrc_32:$src0, VReg_32:$src1, VCCReg:$vcc),
  "v_cndmask_b32_e32 $dst, $src0, $src1, [$vcc]",
  []
>{
  let DisableEncoding = "$vcc";
}

def V_CNDMASK_B32_e64 : VOP3 <0x00000100, (outs VReg_32:$dst),
  (ins VSrc_32:$src0, VSrc_32:$src1, SSrc_64:$src2),
  "v_cndmask_b32_e64 $dst, $src0, $src1, $src2",
  [(set i32:$dst, (select i1:$src2, i32:$src1, i32:$src0))]
> {
  let src0_modifiers = 0;
  let src1_modifiers = 0;
  let src2_modifiers = 0;
}

def V_READLANE_B32 : VOP2 <
  0x00000001,
  (outs SReg_32:$vdst),
  (ins VReg_32:$src0, SSrc_32:$vsrc1),
  "v_readlane_b32 $vdst, $src0, $vsrc1",
  []
>;

def V_WRITELANE_B32 : VOP2 <
  0x00000002,
  (outs VReg_32:$vdst),
  (ins SReg_32:$src0, SSrc_32:$vsrc1),
  "v_writelane_b32 $vdst, $src0, $vsrc1",
  []
>;

let isCommutable = 1 in {
defm V_ADD_F32 : VOP2Inst <vop2<0x3>, "v_add_f32",
  VOP_F32_F32_F32, fadd
>;

defm V_SUB_F32 : VOP2Inst <vop2<0x4>, "v_sub_f32", VOP_F32_F32_F32, fsub>;
defm V_SUBREV_F32 : VOP2Inst <vop2<0x5>, "v_subrev_f32",
  VOP_F32_F32_F32, null_frag, "v_sub_f32"
>;
} // End isCommutable = 1

let isCommutable = 1 in {

defm V_MAC_LEGACY_F32 : VOP2Inst <vop2<0x6>, "v_mac_legacy_f32",
  VOP_F32_F32_F32
>;

defm V_MUL_LEGACY_F32 : VOP2Inst <vop2<0x7>, "v_mul_legacy_f32",
  VOP_F32_F32_F32, int_AMDGPU_mul
>;

defm V_MUL_F32 : VOP2Inst <vop2<0x8>, "v_mul_f32",
  VOP_F32_F32_F32, fmul
>;

defm V_MUL_I32_I24 : VOP2Inst <vop2<0x9>, "v_mul_i32_i24",
  VOP_I32_I32_I32, AMDGPUmul_i24
>;
//defm V_MUL_HI_I32_I24 : VOP2_32 <0x0000000a, "v_mul_hi_i32_i24", []>;
defm V_MUL_U32_U24 : VOP2Inst <vop2<0xb>, "v_mul_u32_u24",
  VOP_I32_I32_I32, AMDGPUmul_u24
>;
//defm V_MUL_HI_U32_U24 : VOP2_32 <0x0000000c, "v_mul_hi_u32_u24", []>;


defm V_MIN_LEGACY_F32 : VOP2Inst <vop2<0xd>, "v_min_legacy_f32",
  VOP_F32_F32_F32, AMDGPUfmin_legacy
>;

defm V_MAX_LEGACY_F32 : VOP2Inst <vop2<0xe>, "v_max_legacy_f32",
  VOP_F32_F32_F32, AMDGPUfmax_legacy
>;

defm V_MIN_F32 : VOP2Inst <vop2<0xf>, "v_min_f32", VOP_F32_F32_F32, fminnum>;
defm V_MAX_F32 : VOP2Inst <vop2<0x10>, "v_max_f32", VOP_F32_F32_F32, fmaxnum>;
defm V_MIN_I32 : VOP2Inst <vop2<0x11>, "v_min_i32", VOP_I32_I32_I32, AMDGPUsmin>;
defm V_MAX_I32 : VOP2Inst <vop2<0x12>, "v_max_i32", VOP_I32_I32_I32, AMDGPUsmax>;
defm V_MIN_U32 : VOP2Inst <vop2<0x13>, "v_min_u32", VOP_I32_I32_I32, AMDGPUumin>;
defm V_MAX_U32 : VOP2Inst <vop2<0x14>, "v_max_u32", VOP_I32_I32_I32, AMDGPUumax>;

defm V_LSHR_B32 : VOP2Inst <vop2<0x15>, "v_lshr_b32", VOP_I32_I32_I32, srl>;

defm V_LSHRREV_B32 : VOP2Inst <
  vop2<0x16>, "v_lshrrev_b32", VOP_I32_I32_I32, null_frag, "v_lshr_b32"
>;

defm V_ASHR_I32 : VOP2Inst <vop2<0x17>, "v_ashr_i32",
  VOP_I32_I32_I32, sra
>;
defm V_ASHRREV_I32 : VOP2Inst <
  vop2<0x18>, "v_ashrrev_i32", VOP_I32_I32_I32, null_frag, "v_ashr_i32"
>;

let hasPostISelHook = 1 in {

defm V_LSHL_B32 : VOP2Inst <vop2<0x19>, "v_lshl_b32", VOP_I32_I32_I32, shl>;

}
defm V_LSHLREV_B32 : VOP2Inst <
  vop2<0x1a>, "v_lshlrev_b32", VOP_I32_I32_I32, null_frag, "v_lshl_b32"
>;

defm V_AND_B32 : VOP2Inst <vop2<0x1b>, "v_and_b32",
  VOP_I32_I32_I32, and>;
defm V_OR_B32 : VOP2Inst <vop2<0x1c>, "v_or_b32",
  VOP_I32_I32_I32, or
>;
defm V_XOR_B32 : VOP2Inst <vop2<0x1d>, "v_xor_b32",
  VOP_I32_I32_I32, xor
>;

} // End isCommutable = 1

defm V_BFM_B32 : VOP2Inst <vop2<0x1e>, "v_bfm_b32",
  VOP_I32_I32_I32, AMDGPUbfm>;

let isCommutable = 1 in {
defm V_MAC_F32 : VOP2Inst <vop2<0x1f>, "v_mac_f32", VOP_F32_F32_F32>;
} // End isCommutable = 1

defm V_MADMK_F32 : VOP2Inst <vop2<0x20>, "v_madmk_f32", VOP_F32_F32_F32>;

let isCommutable = 1 in {
defm V_MADAK_F32 : VOP2Inst <vop2<0x21>, "v_madak_f32", VOP_F32_F32_F32>;
} // End isCommutable = 1


defm V_BCNT_U32_B32 : VOP2Inst <vop2<0x22>, "v_bcnt_u32_b32", VOP_I32_I32_I32>;
defm V_MBCNT_LO_U32_B32 : VOP2Inst <vop2<0x23>, "v_mbcnt_lo_u32_b32",

  VOP_I32_I32_I32
>;
defm V_MBCNT_HI_U32_B32 : VOP2Inst <vop2<0x24>, "v_mbcnt_hi_u32_b32",
  VOP_I32_I32_I32
>;

let isCommutable = 1, Defs = [VCC] in { // Carry-out goes to VCC
// No patterns so that the scalar instructions are always selected.
// The scalar versions will be replaced with vector when needed later.
defm V_ADD_I32 : VOP2bInst <vop2<0x25>, "v_add_i32",
  VOP_I32_I32_I32, add
>;
defm V_SUB_I32 : VOP2bInst <vop2<0x26>, "v_sub_i32",
  VOP_I32_I32_I32, sub
>;
defm V_SUBREV_I32 : VOP2bInst <vop2<0x27>, "v_subrev_i32",
  VOP_I32_I32_I32, null_frag, "v_sub_i32"
>;

let Uses = [VCC] in { // Carry-in comes from VCC
defm V_ADDC_U32 : VOP2bInst <vop2<0x28>, "v_addc_u32",
  VOP_I32_I32_I32_VCC, adde
>;
defm V_SUBB_U32 : VOP2bInst <vop2<0x29>, "v_subb_u32",
  VOP_I32_I32_I32_VCC, sube
>;
defm V_SUBBREV_U32 : VOP2bInst <vop2<0x2a>, "v_subbrev_u32",
  VOP_I32_I32_I32_VCC, null_frag, "v_subb_u32"
>;

} // End Uses = [VCC]
} // End isCommutable = 1, Defs = [VCC]

defm V_LDEXP_F32 : VOP2Inst <vop2<0x2b>, "v_ldexp_f32",
  VOP_F32_F32_I32, AMDGPUldexp
>;
////def V_CVT_PKACCUM_U8_F32 : VOP2_U8 <0x0000002c, "v_cvt_pkaccum_u8_f32", []>;
////def V_CVT_PKNORM_I16_F32 : VOP2_I16 <0x0000002d, "v_cvt_pknorm_i16_f32", []>;
////def V_CVT_PKNORM_U16_F32 : VOP2_U16 <0x0000002e, "v_cvt_pknorm_u16_f32", []>;
defm V_CVT_PKRTZ_F16_F32 : VOP2Inst <vop2<0x2f>, "v_cvt_pkrtz_f16_f32",
 VOP_I32_F32_F32, int_SI_packf16
>;
////def V_CVT_PK_U16_U32 : VOP2_U16 <0x00000030, "v_cvt_pk_u16_u32", []>;
////def V_CVT_PK_I16_I32 : VOP2_I16 <0x00000031, "v_cvt_pk_i16_i32", []>;

//===----------------------------------------------------------------------===//
// VOP3 Instructions
//===----------------------------------------------------------------------===//

let isCommutable = 1 in {
defm V_MAD_LEGACY_F32 : VOP3Inst <vop3<0x140>, "v_mad_legacy_f32",
  VOP_F32_F32_F32_F32
>;

defm V_MAD_F32 : VOP3Inst <vop3<0x141>, "v_mad_f32",
  VOP_F32_F32_F32_F32, fmad
>;

defm V_MAD_I32_I24 : VOP3Inst <vop3<0x142>, "v_mad_i32_i24",
  VOP_I32_I32_I32_I32, AMDGPUmad_i24
>;
defm V_MAD_U32_U24 : VOP3Inst <vop3<0x143>, "v_mad_u32_u24",
  VOP_I32_I32_I32_I32, AMDGPUmad_u24
>;
} // End isCommutable = 1

defm V_CUBEID_F32 : VOP3Inst <vop3<0x144>, "v_cubeid_f32",
  VOP_F32_F32_F32_F32
>;
defm V_CUBESC_F32 : VOP3Inst <vop3<0x145>, "v_cubesc_f32",
  VOP_F32_F32_F32_F32
>;
defm V_CUBETC_F32 : VOP3Inst <vop3<0x146>, "v_cubetc_f32",
  VOP_F32_F32_F32_F32
>;
defm V_CUBEMA_F32 : VOP3Inst <vop3<0x147>, "v_cubema_f32",
  VOP_F32_F32_F32_F32
>;
defm V_BFE_U32 : VOP3Inst <vop3<0x148>, "v_bfe_u32",
  VOP_I32_I32_I32_I32, AMDGPUbfe_u32
>;
defm V_BFE_I32 : VOP3Inst <vop3<0x149>, "v_bfe_i32",
  VOP_I32_I32_I32_I32, AMDGPUbfe_i32
>;
defm V_BFI_B32 : VOP3Inst <vop3<0x14a>, "v_bfi_b32",
  VOP_I32_I32_I32_I32, AMDGPUbfi
>;

let isCommutable = 1 in {
defm V_FMA_F32 : VOP3Inst <vop3<0x14b>, "v_fma_f32",
  VOP_F32_F32_F32_F32, fma
>;
defm V_FMA_F64 : VOP3Inst <vop3<0x14c>, "v_fma_f64",
  VOP_F64_F64_F64_F64, fma
>;
} // End isCommutable = 1

//def V_LERP_U8 : VOP3_U8 <0x0000014d, "v_lerp_u8", []>;
defm V_ALIGNBIT_B32 : VOP3Inst <vop3<0x14e>, "v_alignbit_b32",
  VOP_I32_I32_I32_I32
>;
defm V_ALIGNBYTE_B32 : VOP3Inst <vop3<0x14f>, "v_alignbyte_b32",
  VOP_I32_I32_I32_I32
>;
defm V_MULLIT_F32 : VOP3Inst <vop3<0x150>, "v_mullit_f32",
  VOP_F32_F32_F32_F32>;
defm V_MIN3_F32 : VOP3Inst <vop3<0x151>, "v_min3_f32",
  VOP_F32_F32_F32_F32, AMDGPUfmin3>;

defm V_MIN3_I32 : VOP3Inst <vop3<0x152>, "v_min3_i32",
  VOP_I32_I32_I32_I32, AMDGPUsmin3
>;
defm V_MIN3_U32 : VOP3Inst <vop3<0x153>, "v_min3_u32",
  VOP_I32_I32_I32_I32, AMDGPUumin3
>;
defm V_MAX3_F32 : VOP3Inst <vop3<0x154>, "v_max3_f32",
  VOP_F32_F32_F32_F32, AMDGPUfmax3
>;
defm V_MAX3_I32 : VOP3Inst <vop3<0x155>, "v_max3_i32",
  VOP_I32_I32_I32_I32, AMDGPUsmax3
>;
defm V_MAX3_U32 : VOP3Inst <vop3<0x156>, "v_max3_u32",
  VOP_I32_I32_I32_I32, AMDGPUumax3
>;
//def V_MED3_F32 : VOP3_MED3 <0x00000157, "v_med3_f32", []>;
//def V_MED3_I32 : VOP3_MED3 <0x00000158, "v_med3_i32", []>;
//def V_MED3_U32 : VOP3_MED3 <0x00000159, "v_med3_u32", []>;
//def V_SAD_U8 : VOP3_U8 <0x0000015a, "v_sad_u8", []>;
//def V_SAD_HI_U8 : VOP3_U8 <0x0000015b, "v_sad_hi_u8", []>;
//def V_SAD_U16 : VOP3_U16 <0x0000015c, "v_sad_u16", []>;
defm V_SAD_U32 : VOP3Inst <vop3<0x15d>, "v_sad_u32",
  VOP_I32_I32_I32_I32
>;
////def V_CVT_PK_U8_F32 : VOP3_U8 <0x0000015e, "v_cvt_pk_u8_f32", []>;
defm V_DIV_FIXUP_F32 : VOP3Inst <
  vop3<0x15f>, "v_div_fixup_f32", VOP_F32_F32_F32_F32, AMDGPUdiv_fixup
>;
defm V_DIV_FIXUP_F64 : VOP3Inst <
  vop3<0x160>, "v_div_fixup_f64", VOP_F64_F64_F64_F64, AMDGPUdiv_fixup
>;

defm V_LSHL_B64 : VOP3Inst <vop3<0x161>, "v_lshl_b64",
  VOP_I64_I64_I32, shl
>;
defm V_LSHR_B64 : VOP3Inst <vop3<0x162>, "v_lshr_b64",
  VOP_I64_I64_I32, srl
>;
defm V_ASHR_I64 : VOP3Inst <vop3<0x163>, "v_ashr_i64",
  VOP_I64_I64_I32, sra
>;

let isCommutable = 1 in {

defm V_ADD_F64 : VOP3Inst <vop3<0x164>, "v_add_f64",
  VOP_F64_F64_F64, fadd
>;
defm V_MUL_F64 : VOP3Inst <vop3<0x165>, "v_mul_f64",
  VOP_F64_F64_F64, fmul
>;

defm V_MIN_F64 : VOP3Inst <vop3<0x166>, "v_min_f64",
  VOP_F64_F64_F64, fminnum
>;
defm V_MAX_F64 : VOP3Inst <vop3<0x167>, "v_max_f64",
  VOP_F64_F64_F64, fmaxnum
>;

} // isCommutable = 1

defm V_LDEXP_F64 : VOP3Inst <vop3<0x168>, "v_ldexp_f64",
  VOP_F64_F64_I32, AMDGPUldexp
>;

let isCommutable = 1 in {

defm V_MUL_LO_U32 : VOP3Inst <vop3<0x169>, "v_mul_lo_u32",
  VOP_I32_I32_I32
>;
defm V_MUL_HI_U32 : VOP3Inst <vop3<0x16a>, "v_mul_hi_u32",
  VOP_I32_I32_I32
>;
defm V_MUL_LO_I32 : VOP3Inst <vop3<0x16b>, "v_mul_lo_i32",
  VOP_I32_I32_I32
>;
defm V_MUL_HI_I32 : VOP3Inst <vop3<0x16c>, "v_mul_hi_i32",
  VOP_I32_I32_I32
>;

} // isCommutable = 1

defm V_DIV_SCALE_F32 : VOP3b_32 <vop3<0x16d>, "v_div_scale_f32", []>;

// Double precision division pre-scale.
defm V_DIV_SCALE_F64 : VOP3b_64 <vop3<0x16e>, "v_div_scale_f64", []>;

let isCommutable = 1 in {
defm V_DIV_FMAS_F32 : VOP3Inst <vop3<0x16f>, "v_div_fmas_f32",
  VOP_F32_F32_F32_F32, AMDGPUdiv_fmas
>;
defm V_DIV_FMAS_F64 : VOP3Inst <vop3<0x170>, "v_div_fmas_f64",
  VOP_F64_F64_F64_F64, AMDGPUdiv_fmas
>;
} // End isCommutable = 1

//def V_MSAD_U8 : VOP3_U8 <0x00000171, "v_msad_u8", []>;
//def V_QSAD_U8 : VOP3_U8 <0x00000172, "v_qsad_u8", []>;
//def V_MQSAD_U8 : VOP3_U8 <0x00000173, "v_mqsad_u8", []>;

defm V_TRIG_PREOP_F64 : VOP3Inst <
  vop3<0x174>, "v_trig_preop_f64", VOP_F64_F64_I32, AMDGPUtrig_preop
>;

//===----------------------------------------------------------------------===//
// Pseudo Instructions
//===----------------------------------------------------------------------===//

let isCodeGenOnly = 1, isPseudo = 1 in {

def V_MOV_I1 : InstSI <
  (outs VReg_1:$dst),
  (ins i1imm:$src),
  "", [(set i1:$dst, (imm:$src))]
>;

def V_AND_I1 : InstSI <
   (outs VReg_1:$dst), (ins VReg_1:$src0, VReg_1:$src1), "",
   [(set i1:$dst, (and i1:$src0, i1:$src1))]
>;

def V_OR_I1 : InstSI <
   (outs VReg_1:$dst), (ins VReg_1:$src0, VReg_1:$src1), "",
   [(set i1:$dst, (or i1:$src0, i1:$src1))]
>;

def V_XOR_I1 : InstSI <
  (outs VReg_1:$dst), (ins VReg_1:$src0, VReg_1:$src1), "",
  [(set i1:$dst, (xor i1:$src0, i1:$src1))]
>;

let hasSideEffects = 1 in {
def SGPR_USE : InstSI <(outs),(ins), "", []>;
}

// SI pseudo instructions. These are used by the CFG structurizer pass
// and should be lowered to ISA instructions prior to codegen.

let mayLoad = 1, mayStore = 1, hasSideEffects = 1,
    Uses = [EXEC], Defs = [EXEC] in {

let isBranch = 1, isTerminator = 1 in {

def SI_IF: InstSI <
  (outs SReg_64:$dst),
  (ins SReg_64:$vcc, brtarget:$target),
  "",
  [(set i64:$dst, (int_SI_if i1:$vcc, bb:$target))]
>;

def SI_ELSE : InstSI <
  (outs SReg_64:$dst),
  (ins SReg_64:$src, brtarget:$target),
  "",
  [(set i64:$dst, (int_SI_else i64:$src, bb:$target))]
> {
  let Constraints = "$src = $dst";
}

def SI_LOOP : InstSI <
  (outs),
  (ins SReg_64:$saved, brtarget:$target),
  "si_loop $saved, $target",
  [(int_SI_loop i64:$saved, bb:$target)]
>;

} // end isBranch = 1, isTerminator = 1

def SI_BREAK : InstSI <
  (outs SReg_64:$dst),
  (ins SReg_64:$src),
  "si_else $dst, $src",
  [(set i64:$dst, (int_SI_break i64:$src))]
>;

def SI_IF_BREAK : InstSI <
  (outs SReg_64:$dst),
  (ins SReg_64:$vcc, SReg_64:$src),
  "si_if_break $dst, $vcc, $src",
  [(set i64:$dst, (int_SI_if_break i1:$vcc, i64:$src))]
>;

def SI_ELSE_BREAK : InstSI <
  (outs SReg_64:$dst),
  (ins SReg_64:$src0, SReg_64:$src1),
  "si_else_break $dst, $src0, $src1",
  [(set i64:$dst, (int_SI_else_break i64:$src0, i64:$src1))]
>;

def SI_END_CF : InstSI <
  (outs),
  (ins SReg_64:$saved),
  "si_end_cf $saved",
  [(int_SI_end_cf i64:$saved)]
>;

def SI_KILL : InstSI <
  (outs),
  (ins VSrc_32:$src),
  "si_kill $src",
  [(int_AMDGPU_kill f32:$src)]
>;

} // end mayLoad = 1, mayStore = 1, hasSideEffects = 1
  // Uses = [EXEC], Defs = [EXEC]

let Uses = [EXEC], Defs = [EXEC,VCC,M0] in {

//defm SI_ : RegisterLoadStore <VReg_32, FRAMEri, ADDRIndirect>;

let UseNamedOperandTable = 1 in {

def SI_RegisterLoad : InstSI <
  (outs VReg_32:$dst, SReg_64:$temp),
  (ins FRAMEri32:$addr, i32imm:$chan),
  "", []
> {
  let isRegisterLoad = 1;
  let mayLoad = 1;
}

class SIRegStore<dag outs> : InstSI <
  outs,
  (ins VReg_32:$val, FRAMEri32:$addr, i32imm:$chan),
  "", []
> {
  let isRegisterStore = 1;
  let mayStore = 1;
}

let usesCustomInserter = 1 in {
def SI_RegisterStorePseudo : SIRegStore<(outs)>;
} // End usesCustomInserter = 1
def SI_RegisterStore : SIRegStore<(outs SReg_64:$temp)>;


} // End UseNamedOperandTable = 1

def SI_INDIRECT_SRC : InstSI <
  (outs VReg_32:$dst, SReg_64:$temp),
  (ins unknown:$src, VSrc_32:$idx, i32imm:$off),
  "si_indirect_src $dst, $temp, $src, $idx, $off",
  []
>;

class SI_INDIRECT_DST<RegisterClass rc> : InstSI <
  (outs rc:$dst, SReg_64:$temp),
  (ins unknown:$src, VSrc_32:$idx, i32imm:$off, VReg_32:$val),
  "si_indirect_dst $dst, $temp, $src, $idx, $off, $val",
  []
> {
  let Constraints = "$src = $dst";
}

def SI_INDIRECT_DST_V1 : SI_INDIRECT_DST<VReg_32>;
def SI_INDIRECT_DST_V2 : SI_INDIRECT_DST<VReg_64>;
def SI_INDIRECT_DST_V4 : SI_INDIRECT_DST<VReg_128>;
def SI_INDIRECT_DST_V8 : SI_INDIRECT_DST<VReg_256>;
def SI_INDIRECT_DST_V16 : SI_INDIRECT_DST<VReg_512>;

} // Uses = [EXEC,VCC,M0], Defs = [EXEC,VCC,M0]

let usesCustomInserter = 1 in {

def V_SUB_F64 : InstSI <
  (outs VReg_64:$dst),
  (ins VReg_64:$src0, VReg_64:$src1),
  "v_sub_f64 $dst, $src0, $src1",
  [(set f64:$dst, (fsub f64:$src0, f64:$src1))]
>;

} // end usesCustomInserter

multiclass SI_SPILL_SGPR <RegisterClass sgpr_class> {

  def _SAVE : InstSI <
    (outs),
    (ins sgpr_class:$src, i32imm:$frame_idx),
    "", []
  >;

  def _RESTORE : InstSI <
    (outs sgpr_class:$dst),
    (ins i32imm:$frame_idx),
    "", []
  >;

}

defm SI_SPILL_S32  : SI_SPILL_SGPR <SReg_32>;
defm SI_SPILL_S64  : SI_SPILL_SGPR <SReg_64>;
defm SI_SPILL_S128 : SI_SPILL_SGPR <SReg_128>;
defm SI_SPILL_S256 : SI_SPILL_SGPR <SReg_256>;
defm SI_SPILL_S512 : SI_SPILL_SGPR <SReg_512>;

multiclass SI_SPILL_VGPR <RegisterClass vgpr_class> {
  def _SAVE : InstSI <
    (outs),
    (ins vgpr_class:$src, i32imm:$frame_idx),
    "", []
  >;

  def _RESTORE : InstSI <
    (outs vgpr_class:$dst),
    (ins i32imm:$frame_idx),
    "", []
  >;
}

defm SI_SPILL_V32  : SI_SPILL_VGPR <VReg_32>;
defm SI_SPILL_V64  : SI_SPILL_VGPR <VReg_64>;
defm SI_SPILL_V96  : SI_SPILL_VGPR <VReg_96>;
defm SI_SPILL_V128 : SI_SPILL_VGPR <VReg_128>;
defm SI_SPILL_V256 : SI_SPILL_VGPR <VReg_256>;
defm SI_SPILL_V512 : SI_SPILL_VGPR <VReg_512>;

let Defs = [SCC] in {

def SI_CONSTDATA_PTR : InstSI <
  (outs SReg_64:$dst),
  (ins),
  "", [(set SReg_64:$dst, (i64 SIconstdata_ptr))]
>;

} // End Defs = [SCC]

} // end IsCodeGenOnly, isPseudo

} // end SubtargetPredicate = SI

let Predicates = [isSI] in {

def : Pat<
  (int_AMDGPU_cndlt f32:$src0, f32:$src1, f32:$src2),
  (V_CNDMASK_B32_e64 $src2, $src1,
                     (V_CMP_GT_F32_e64 SRCMODS.NONE, 0, SRCMODS.NONE, $src0,
                                       DSTCLAMP.NONE, DSTOMOD.NONE))
>;

def : Pat <
  (int_AMDGPU_kilp),
  (SI_KILL 0xbf800000)
>;

/* int_SI_vs_load_input */
def : Pat<
  (SIload_input v4i32:$tlst, imm:$attr_offset, i32:$buf_idx_vgpr),
  (BUFFER_LOAD_FORMAT_XYZW_IDXEN $tlst, $buf_idx_vgpr, imm:$attr_offset, 0, 0, 0, 0)
>;

/* int_SI_export */
def : Pat <
  (int_SI_export imm:$en, imm:$vm, imm:$done, imm:$tgt, imm:$compr,
                 f32:$src0, f32:$src1, f32:$src2, f32:$src3),
  (EXP imm:$en, imm:$tgt, imm:$compr, imm:$done, imm:$vm,
       $src0, $src1, $src2, $src3)
>;

//===----------------------------------------------------------------------===//
// SMRD Patterns
//===----------------------------------------------------------------------===//

multiclass SMRD_Pattern <SMRD Instr_IMM, SMRD Instr_SGPR, ValueType vt> {

  // 1. Offset as 8bit DWORD immediate
  def : Pat <
    (constant_load (add i64:$sbase, (i64 IMM8bitDWORD:$offset))),
    (vt (Instr_IMM $sbase, (as_dword_i32imm $offset)))
  >;

  // 2. Offset loaded in an 32bit SGPR
  def : Pat <
    (constant_load (add i64:$sbase, (i64 IMM32bit:$offset))),
    (vt (Instr_SGPR $sbase, (S_MOV_B32 (i32 (as_i32imm $offset)))))
  >;

  // 3. No offset at all
  def : Pat <
    (constant_load i64:$sbase),
    (vt (Instr_IMM $sbase, 0))
  >;
}

defm : SMRD_Pattern <S_LOAD_DWORD_IMM, S_LOAD_DWORD_SGPR, f32>;
defm : SMRD_Pattern <S_LOAD_DWORD_IMM, S_LOAD_DWORD_SGPR, i32>;
defm : SMRD_Pattern <S_LOAD_DWORDX2_IMM, S_LOAD_DWORDX2_SGPR, v2i32>;
defm : SMRD_Pattern <S_LOAD_DWORDX4_IMM, S_LOAD_DWORDX4_SGPR, v4i32>;
defm : SMRD_Pattern <S_LOAD_DWORDX8_IMM, S_LOAD_DWORDX8_SGPR, v32i8>;
defm : SMRD_Pattern <S_LOAD_DWORDX8_IMM, S_LOAD_DWORDX8_SGPR, v8i32>;
defm : SMRD_Pattern <S_LOAD_DWORDX16_IMM, S_LOAD_DWORDX16_SGPR, v16i32>;

// 1. Offset as 8bit DWORD immediate
def : Pat <
  (SIload_constant v4i32:$sbase, IMM8bitDWORD:$offset),
  (S_BUFFER_LOAD_DWORD_IMM $sbase, (as_dword_i32imm $offset))
>;

// 2. Offset loaded in an 32bit SGPR
def : Pat <
  (SIload_constant v4i32:$sbase, imm:$offset),
  (S_BUFFER_LOAD_DWORD_SGPR $sbase, (S_MOV_B32 imm:$offset))
>;

} // Predicates = [isSI] in {

//===----------------------------------------------------------------------===//
// SOP1 Patterns
//===----------------------------------------------------------------------===//

def : Pat <
  (i64 (ctpop i64:$src)),
    (i64 (REG_SEQUENCE SReg_64,
     (S_BCNT1_I32_B64 $src), sub0,
     (S_MOV_B32 0), sub1))
>;

//===----------------------------------------------------------------------===//
// SOP2 Patterns
//===----------------------------------------------------------------------===//

// V_ADD_I32_e32/S_ADD_U32 produces carry in VCC/SCC. For the vector
// case, the sgpr-copies pass will fix this to use the vector version.
def : Pat <
  (i32 (addc i32:$src0, i32:$src1)),
  (S_ADD_U32 $src0, $src1)
>;

let  Predicates = [isSI] in {

//===----------------------------------------------------------------------===//
// SOPP Patterns
//===----------------------------------------------------------------------===//

def : Pat <
  (int_AMDGPU_barrier_global),
  (S_BARRIER)
>;

//===----------------------------------------------------------------------===//
// VOP1 Patterns
//===----------------------------------------------------------------------===//

let Predicates = [UnsafeFPMath] in {
def : RcpPat<V_RCP_F64_e32, f64>;
defm : RsqPat<V_RSQ_F64_e32, f64>;
defm : RsqPat<V_RSQ_F32_e32, f32>;
}

//===----------------------------------------------------------------------===//
// VOP2 Patterns
//===----------------------------------------------------------------------===//

def : Pat <
  (i32 (add (i32 (ctpop i32:$popcnt)), i32:$val)),
  (V_BCNT_U32_B32_e64 $popcnt, $val)
>;

/********** ======================= **********/
/********** Image sampling patterns **********/
/********** ======================= **********/

// Image + sampler
class SampleRawPattern<SDPatternOperator name, MIMG opcode, ValueType vt> : Pat <
  (name vt:$addr, v8i32:$rsrc, v4i32:$sampler, i32:$dmask, i32:$unorm,
        i32:$r128, i32:$da, i32:$glc, i32:$slc, i32:$tfe, i32:$lwe),
  (opcode (as_i32imm $dmask), (as_i1imm $unorm), (as_i1imm $glc), (as_i1imm $da),
          (as_i1imm $r128), (as_i1imm $tfe), (as_i1imm $lwe), (as_i1imm $slc),
          $addr, $rsrc, $sampler)
>;

multiclass SampleRawPatterns<SDPatternOperator name, string opcode> {
  def : SampleRawPattern<name, !cast<MIMG>(opcode # _V4_V1), i32>;
  def : SampleRawPattern<name, !cast<MIMG>(opcode # _V4_V2), v2i32>;
  def : SampleRawPattern<name, !cast<MIMG>(opcode # _V4_V4), v4i32>;
  def : SampleRawPattern<name, !cast<MIMG>(opcode # _V4_V8), v8i32>;
  def : SampleRawPattern<name, !cast<MIMG>(opcode # _V4_V16), v16i32>;
}

// Image only
class ImagePattern<SDPatternOperator name, MIMG opcode, ValueType vt> : Pat <
  (name vt:$addr, v8i32:$rsrc, i32:$dmask, i32:$unorm,
        i32:$r128, i32:$da, i32:$glc, i32:$slc, i32:$tfe, i32:$lwe),
  (opcode (as_i32imm $dmask), (as_i1imm $unorm), (as_i1imm $glc), (as_i1imm $da),
          (as_i1imm $r128), (as_i1imm $tfe), (as_i1imm $lwe), (as_i1imm $slc),
          $addr, $rsrc)
>;

multiclass ImagePatterns<SDPatternOperator name, string opcode> {
  def : ImagePattern<name, !cast<MIMG>(opcode # _V4_V1), i32>;
  def : ImagePattern<name, !cast<MIMG>(opcode # _V4_V2), v2i32>;
  def : ImagePattern<name, !cast<MIMG>(opcode # _V4_V4), v4i32>;
}

// Basic sample
defm : SampleRawPatterns<int_SI_image_sample,           "IMAGE_SAMPLE">;
defm : SampleRawPatterns<int_SI_image_sample_cl,        "IMAGE_SAMPLE_CL">;
defm : SampleRawPatterns<int_SI_image_sample_d,         "IMAGE_SAMPLE_D">;
defm : SampleRawPatterns<int_SI_image_sample_d_cl,      "IMAGE_SAMPLE_D_CL">;
defm : SampleRawPatterns<int_SI_image_sample_l,         "IMAGE_SAMPLE_L">;
defm : SampleRawPatterns<int_SI_image_sample_b,         "IMAGE_SAMPLE_B">;
defm : SampleRawPatterns<int_SI_image_sample_b_cl,      "IMAGE_SAMPLE_B_CL">;
defm : SampleRawPatterns<int_SI_image_sample_lz,        "IMAGE_SAMPLE_LZ">;
defm : SampleRawPatterns<int_SI_image_sample_cd,        "IMAGE_SAMPLE_CD">;
defm : SampleRawPatterns<int_SI_image_sample_cd_cl,     "IMAGE_SAMPLE_CD_CL">;

// Sample with comparison
defm : SampleRawPatterns<int_SI_image_sample_c,         "IMAGE_SAMPLE_C">;
defm : SampleRawPatterns<int_SI_image_sample_c_cl,      "IMAGE_SAMPLE_C_CL">;
defm : SampleRawPatterns<int_SI_image_sample_c_d,       "IMAGE_SAMPLE_C_D">;
defm : SampleRawPatterns<int_SI_image_sample_c_d_cl,    "IMAGE_SAMPLE_C_D_CL">;
defm : SampleRawPatterns<int_SI_image_sample_c_l,       "IMAGE_SAMPLE_C_L">;
defm : SampleRawPatterns<int_SI_image_sample_c_b,       "IMAGE_SAMPLE_C_B">;
defm : SampleRawPatterns<int_SI_image_sample_c_b_cl,    "IMAGE_SAMPLE_C_B_CL">;
defm : SampleRawPatterns<int_SI_image_sample_c_lz,      "IMAGE_SAMPLE_C_LZ">;
defm : SampleRawPatterns<int_SI_image_sample_c_cd,      "IMAGE_SAMPLE_C_CD">;
defm : SampleRawPatterns<int_SI_image_sample_c_cd_cl,   "IMAGE_SAMPLE_C_CD_CL">;

// Sample with offsets
defm : SampleRawPatterns<int_SI_image_sample_o,         "IMAGE_SAMPLE_O">;
defm : SampleRawPatterns<int_SI_image_sample_cl_o,      "IMAGE_SAMPLE_CL_O">;
defm : SampleRawPatterns<int_SI_image_sample_d_o,       "IMAGE_SAMPLE_D_O">;
defm : SampleRawPatterns<int_SI_image_sample_d_cl_o,    "IMAGE_SAMPLE_D_CL_O">;
defm : SampleRawPatterns<int_SI_image_sample_l_o,       "IMAGE_SAMPLE_L_O">;
defm : SampleRawPatterns<int_SI_image_sample_b_o,       "IMAGE_SAMPLE_B_O">;
defm : SampleRawPatterns<int_SI_image_sample_b_cl_o,    "IMAGE_SAMPLE_B_CL_O">;
defm : SampleRawPatterns<int_SI_image_sample_lz_o,      "IMAGE_SAMPLE_LZ_O">;
defm : SampleRawPatterns<int_SI_image_sample_cd_o,      "IMAGE_SAMPLE_CD_O">;
defm : SampleRawPatterns<int_SI_image_sample_cd_cl_o,   "IMAGE_SAMPLE_CD_CL_O">;

// Sample with comparison and offsets
defm : SampleRawPatterns<int_SI_image_sample_c_o,       "IMAGE_SAMPLE_C_O">;
defm : SampleRawPatterns<int_SI_image_sample_c_cl_o,    "IMAGE_SAMPLE_C_CL_O">;
defm : SampleRawPatterns<int_SI_image_sample_c_d_o,     "IMAGE_SAMPLE_C_D_O">;
defm : SampleRawPatterns<int_SI_image_sample_c_d_cl_o,  "IMAGE_SAMPLE_C_D_CL_O">;
defm : SampleRawPatterns<int_SI_image_sample_c_l_o,     "IMAGE_SAMPLE_C_L_O">;
defm : SampleRawPatterns<int_SI_image_sample_c_b_o,     "IMAGE_SAMPLE_C_B_O">;
defm : SampleRawPatterns<int_SI_image_sample_c_b_cl_o,  "IMAGE_SAMPLE_C_B_CL_O">;
defm : SampleRawPatterns<int_SI_image_sample_c_lz_o,    "IMAGE_SAMPLE_C_LZ_O">;
defm : SampleRawPatterns<int_SI_image_sample_c_cd_o,    "IMAGE_SAMPLE_C_CD_O">;
defm : SampleRawPatterns<int_SI_image_sample_c_cd_cl_o, "IMAGE_SAMPLE_C_CD_CL_O">;

// Gather opcodes
// Only the variants which make sense are defined.
def : SampleRawPattern<int_SI_gather4,           IMAGE_GATHER4_V4_V2,        v2i32>;
def : SampleRawPattern<int_SI_gather4,           IMAGE_GATHER4_V4_V4,        v4i32>;
def : SampleRawPattern<int_SI_gather4_cl,        IMAGE_GATHER4_CL_V4_V4,     v4i32>;
def : SampleRawPattern<int_SI_gather4_l,         IMAGE_GATHER4_L_V4_V4,      v4i32>;
def : SampleRawPattern<int_SI_gather4_b,         IMAGE_GATHER4_B_V4_V4,      v4i32>;
def : SampleRawPattern<int_SI_gather4_b_cl,      IMAGE_GATHER4_B_CL_V4_V4,   v4i32>;
def : SampleRawPattern<int_SI_gather4_b_cl,      IMAGE_GATHER4_B_CL_V4_V8,   v8i32>;
def : SampleRawPattern<int_SI_gather4_lz,        IMAGE_GATHER4_LZ_V4_V2,     v2i32>;
def : SampleRawPattern<int_SI_gather4_lz,        IMAGE_GATHER4_LZ_V4_V4,     v4i32>;

def : SampleRawPattern<int_SI_gather4_c,         IMAGE_GATHER4_C_V4_V4,      v4i32>;
def : SampleRawPattern<int_SI_gather4_c_cl,      IMAGE_GATHER4_C_CL_V4_V4,   v4i32>;
def : SampleRawPattern<int_SI_gather4_c_cl,      IMAGE_GATHER4_C_CL_V4_V8,   v8i32>;
def : SampleRawPattern<int_SI_gather4_c_l,       IMAGE_GATHER4_C_L_V4_V4,    v4i32>;
def : SampleRawPattern<int_SI_gather4_c_l,       IMAGE_GATHER4_C_L_V4_V8,    v8i32>;
def : SampleRawPattern<int_SI_gather4_c_b,       IMAGE_GATHER4_C_B_V4_V4,    v4i32>;
def : SampleRawPattern<int_SI_gather4_c_b,       IMAGE_GATHER4_C_B_V4_V8,    v8i32>;
def : SampleRawPattern<int_SI_gather4_c_b_cl,    IMAGE_GATHER4_C_B_CL_V4_V8, v8i32>;
def : SampleRawPattern<int_SI_gather4_c_lz,      IMAGE_GATHER4_C_LZ_V4_V4,   v4i32>;

def : SampleRawPattern<int_SI_gather4_o,         IMAGE_GATHER4_O_V4_V4,      v4i32>;
def : SampleRawPattern<int_SI_gather4_cl_o,      IMAGE_GATHER4_CL_O_V4_V4,   v4i32>;
def : SampleRawPattern<int_SI_gather4_cl_o,      IMAGE_GATHER4_CL_O_V4_V8,   v8i32>;
def : SampleRawPattern<int_SI_gather4_l_o,       IMAGE_GATHER4_L_O_V4_V4,    v4i32>;
def : SampleRawPattern<int_SI_gather4_l_o,       IMAGE_GATHER4_L_O_V4_V8,    v8i32>;
def : SampleRawPattern<int_SI_gather4_b_o,       IMAGE_GATHER4_B_O_V4_V4,    v4i32>;
def : SampleRawPattern<int_SI_gather4_b_o,       IMAGE_GATHER4_B_O_V4_V8,    v8i32>;
def : SampleRawPattern<int_SI_gather4_b_cl_o,    IMAGE_GATHER4_B_CL_O_V4_V8, v8i32>;
def : SampleRawPattern<int_SI_gather4_lz_o,      IMAGE_GATHER4_LZ_O_V4_V4,   v4i32>;

def : SampleRawPattern<int_SI_gather4_c_o,       IMAGE_GATHER4_C_O_V4_V4,    v4i32>;
def : SampleRawPattern<int_SI_gather4_c_o,       IMAGE_GATHER4_C_O_V4_V8,    v8i32>;
def : SampleRawPattern<int_SI_gather4_c_cl_o,    IMAGE_GATHER4_C_CL_O_V4_V8, v8i32>;
def : SampleRawPattern<int_SI_gather4_c_l_o,     IMAGE_GATHER4_C_L_O_V4_V8,  v8i32>;
def : SampleRawPattern<int_SI_gather4_c_b_o,     IMAGE_GATHER4_C_B_O_V4_V8,  v8i32>;
def : SampleRawPattern<int_SI_gather4_c_b_cl_o,  IMAGE_GATHER4_C_B_CL_O_V4_V8, v8i32>;
def : SampleRawPattern<int_SI_gather4_c_lz_o,    IMAGE_GATHER4_C_LZ_O_V4_V4, v4i32>;
def : SampleRawPattern<int_SI_gather4_c_lz_o,    IMAGE_GATHER4_C_LZ_O_V4_V8, v8i32>;

def : SampleRawPattern<int_SI_getlod, IMAGE_GET_LOD_V4_V1, i32>;
def : SampleRawPattern<int_SI_getlod, IMAGE_GET_LOD_V4_V2, v2i32>;
def : SampleRawPattern<int_SI_getlod, IMAGE_GET_LOD_V4_V4, v4i32>;

def : ImagePattern<int_SI_getresinfo, IMAGE_GET_RESINFO_V4_V1, i32>;
defm : ImagePatterns<int_SI_image_load, "IMAGE_LOAD">;
defm : ImagePatterns<int_SI_image_load_mip, "IMAGE_LOAD_MIP">;

/* SIsample for simple 1D texture lookup */
def : Pat <
  (SIsample i32:$addr, v32i8:$rsrc, v4i32:$sampler, imm),
  (IMAGE_SAMPLE_V4_V1 0xf, 0, 0, 0, 0, 0, 0, 0, $addr, $rsrc, $sampler)
>;

class SamplePattern<SDNode name, MIMG opcode, ValueType vt> : Pat <
    (name vt:$addr, v32i8:$rsrc, v4i32:$sampler, imm),
    (opcode 0xf, 0, 0, 0, 0, 0, 0, 0, $addr, $rsrc, $sampler)
>;

class SampleRectPattern<SDNode name, MIMG opcode, ValueType vt> : Pat <
    (name vt:$addr, v32i8:$rsrc, v4i32:$sampler, TEX_RECT),
    (opcode 0xf, 1, 0, 0, 0, 0, 0, 0, $addr, $rsrc, $sampler)
>;

class SampleArrayPattern<SDNode name, MIMG opcode, ValueType vt> : Pat <
    (name vt:$addr, v32i8:$rsrc, v4i32:$sampler, TEX_ARRAY),
    (opcode 0xf, 0, 0, 1, 0, 0, 0, 0, $addr, $rsrc, $sampler)
>;

class SampleShadowPattern<SDNode name, MIMG opcode,
                          ValueType vt> : Pat <
    (name vt:$addr, v32i8:$rsrc, v4i32:$sampler, TEX_SHADOW),
    (opcode 0xf, 0, 0, 0, 0, 0, 0, 0, $addr, $rsrc, $sampler)
>;

class SampleShadowArrayPattern<SDNode name, MIMG opcode,
                               ValueType vt> : Pat <
    (name vt:$addr, v32i8:$rsrc, v4i32:$sampler, TEX_SHADOW_ARRAY),
    (opcode 0xf, 0, 0, 1, 0, 0, 0, 0, $addr, $rsrc, $sampler)
>;

/* SIsample* for texture lookups consuming more address parameters */
multiclass SamplePatterns<MIMG sample, MIMG sample_c, MIMG sample_l,
                          MIMG sample_c_l, MIMG sample_b, MIMG sample_c_b,
MIMG sample_d, MIMG sample_c_d, ValueType addr_type> {
  def : SamplePattern <SIsample, sample, addr_type>;
  def : SampleRectPattern <SIsample, sample, addr_type>;
  def : SampleArrayPattern <SIsample, sample, addr_type>;
  def : SampleShadowPattern <SIsample, sample_c, addr_type>;
  def : SampleShadowArrayPattern <SIsample, sample_c, addr_type>;

  def : SamplePattern <SIsamplel, sample_l, addr_type>;
  def : SampleArrayPattern <SIsamplel, sample_l, addr_type>;
  def : SampleShadowPattern <SIsamplel, sample_c_l, addr_type>;
  def : SampleShadowArrayPattern <SIsamplel, sample_c_l, addr_type>;

  def : SamplePattern <SIsampleb, sample_b, addr_type>;
  def : SampleArrayPattern <SIsampleb, sample_b, addr_type>;
  def : SampleShadowPattern <SIsampleb, sample_c_b, addr_type>;
  def : SampleShadowArrayPattern <SIsampleb, sample_c_b, addr_type>;

  def : SamplePattern <SIsampled, sample_d, addr_type>;
  def : SampleArrayPattern <SIsampled, sample_d, addr_type>;
  def : SampleShadowPattern <SIsampled, sample_c_d, addr_type>;
  def : SampleShadowArrayPattern <SIsampled, sample_c_d, addr_type>;
}

defm : SamplePatterns<IMAGE_SAMPLE_V4_V2, IMAGE_SAMPLE_C_V4_V2,
                      IMAGE_SAMPLE_L_V4_V2, IMAGE_SAMPLE_C_L_V4_V2,
                      IMAGE_SAMPLE_B_V4_V2, IMAGE_SAMPLE_C_B_V4_V2,
                      IMAGE_SAMPLE_D_V4_V2, IMAGE_SAMPLE_C_D_V4_V2,
                      v2i32>;
defm : SamplePatterns<IMAGE_SAMPLE_V4_V4, IMAGE_SAMPLE_C_V4_V4,
                      IMAGE_SAMPLE_L_V4_V4, IMAGE_SAMPLE_C_L_V4_V4,
                      IMAGE_SAMPLE_B_V4_V4, IMAGE_SAMPLE_C_B_V4_V4,
                      IMAGE_SAMPLE_D_V4_V4, IMAGE_SAMPLE_C_D_V4_V4,
                      v4i32>;
defm : SamplePatterns<IMAGE_SAMPLE_V4_V8, IMAGE_SAMPLE_C_V4_V8,
                      IMAGE_SAMPLE_L_V4_V8, IMAGE_SAMPLE_C_L_V4_V8,
                      IMAGE_SAMPLE_B_V4_V8, IMAGE_SAMPLE_C_B_V4_V8,
                      IMAGE_SAMPLE_D_V4_V8, IMAGE_SAMPLE_C_D_V4_V8,
                      v8i32>;
defm : SamplePatterns<IMAGE_SAMPLE_V4_V16, IMAGE_SAMPLE_C_V4_V16,
                      IMAGE_SAMPLE_L_V4_V16, IMAGE_SAMPLE_C_L_V4_V16,
                      IMAGE_SAMPLE_B_V4_V16, IMAGE_SAMPLE_C_B_V4_V16,
                      IMAGE_SAMPLE_D_V4_V16, IMAGE_SAMPLE_C_D_V4_V16,
                      v16i32>;

/* int_SI_imageload for texture fetches consuming varying address parameters */
class ImageLoadPattern<Intrinsic name, MIMG opcode, ValueType addr_type> : Pat <
    (name addr_type:$addr, v32i8:$rsrc, imm),
    (opcode 0xf, 0, 0, 0, 0, 0, 0, 0, $addr, $rsrc)
>;

class ImageLoadArrayPattern<Intrinsic name, MIMG opcode, ValueType addr_type> : Pat <
    (name addr_type:$addr, v32i8:$rsrc, TEX_ARRAY),
    (opcode 0xf, 0, 0, 1, 0, 0, 0, 0, $addr, $rsrc)
>;

class ImageLoadMSAAPattern<Intrinsic name, MIMG opcode, ValueType addr_type> : Pat <
    (name addr_type:$addr, v32i8:$rsrc, TEX_MSAA),
    (opcode 0xf, 0, 0, 0, 0, 0, 0, 0, $addr, $rsrc)
>;

class ImageLoadArrayMSAAPattern<Intrinsic name, MIMG opcode, ValueType addr_type> : Pat <
    (name addr_type:$addr, v32i8:$rsrc, TEX_ARRAY_MSAA),
    (opcode 0xf, 0, 0, 1, 0, 0, 0, 0, $addr, $rsrc)
>;

multiclass ImageLoadPatterns<MIMG opcode, ValueType addr_type> {
  def : ImageLoadPattern <int_SI_imageload, opcode, addr_type>;
  def : ImageLoadArrayPattern <int_SI_imageload, opcode, addr_type>;
}

multiclass ImageLoadMSAAPatterns<MIMG opcode, ValueType addr_type> {
  def : ImageLoadMSAAPattern <int_SI_imageload, opcode, addr_type>;
  def : ImageLoadArrayMSAAPattern <int_SI_imageload, opcode, addr_type>;
}

defm : ImageLoadPatterns<IMAGE_LOAD_MIP_V4_V2, v2i32>;
defm : ImageLoadPatterns<IMAGE_LOAD_MIP_V4_V4, v4i32>;

defm : ImageLoadMSAAPatterns<IMAGE_LOAD_V4_V2, v2i32>;
defm : ImageLoadMSAAPatterns<IMAGE_LOAD_V4_V4, v4i32>;

/* Image resource information */
def : Pat <
  (int_SI_resinfo i32:$mipid, v32i8:$rsrc, imm),
  (IMAGE_GET_RESINFO_V4_V1 0xf, 0, 0, 0, 0, 0, 0, 0, (V_MOV_B32_e32 $mipid), $rsrc)
>;

def : Pat <
  (int_SI_resinfo i32:$mipid, v32i8:$rsrc, TEX_ARRAY),
  (IMAGE_GET_RESINFO_V4_V1 0xf, 0, 0, 1, 0, 0, 0, 0, (V_MOV_B32_e32 $mipid), $rsrc)
>;

def : Pat <
  (int_SI_resinfo i32:$mipid, v32i8:$rsrc, TEX_ARRAY_MSAA),
  (IMAGE_GET_RESINFO_V4_V1 0xf, 0, 0, 1, 0, 0, 0, 0, (V_MOV_B32_e32 $mipid), $rsrc)
>;

/********** ============================================ **********/
/********** Extraction, Insertion, Building and Casting  **********/
/********** ============================================ **********/

foreach Index = 0-2 in {
  def Extract_Element_v2i32_#Index : Extract_Element <
    i32, v2i32, Index, !cast<SubRegIndex>(sub#Index)
  >;
  def Insert_Element_v2i32_#Index : Insert_Element <
    i32, v2i32, Index, !cast<SubRegIndex>(sub#Index)
  >;

  def Extract_Element_v2f32_#Index : Extract_Element <
    f32, v2f32, Index, !cast<SubRegIndex>(sub#Index)
  >;
  def Insert_Element_v2f32_#Index : Insert_Element <
    f32, v2f32, Index, !cast<SubRegIndex>(sub#Index)
  >;
}

foreach Index = 0-3 in {
  def Extract_Element_v4i32_#Index : Extract_Element <
    i32, v4i32, Index, !cast<SubRegIndex>(sub#Index)
  >;
  def Insert_Element_v4i32_#Index : Insert_Element <
    i32, v4i32, Index, !cast<SubRegIndex>(sub#Index)
  >;

  def Extract_Element_v4f32_#Index : Extract_Element <
    f32, v4f32, Index, !cast<SubRegIndex>(sub#Index)
  >;
  def Insert_Element_v4f32_#Index : Insert_Element <
    f32, v4f32, Index, !cast<SubRegIndex>(sub#Index)
  >;
}

foreach Index = 0-7 in {
  def Extract_Element_v8i32_#Index : Extract_Element <
    i32, v8i32, Index, !cast<SubRegIndex>(sub#Index)
  >;
  def Insert_Element_v8i32_#Index : Insert_Element <
    i32, v8i32, Index, !cast<SubRegIndex>(sub#Index)
  >;

  def Extract_Element_v8f32_#Index : Extract_Element <
    f32, v8f32, Index, !cast<SubRegIndex>(sub#Index)
  >;
  def Insert_Element_v8f32_#Index : Insert_Element <
    f32, v8f32, Index, !cast<SubRegIndex>(sub#Index)
  >;
}

foreach Index = 0-15 in {
  def Extract_Element_v16i32_#Index : Extract_Element <
    i32, v16i32, Index, !cast<SubRegIndex>(sub#Index)
  >;
  def Insert_Element_v16i32_#Index : Insert_Element <
    i32, v16i32, Index, !cast<SubRegIndex>(sub#Index)
  >;

  def Extract_Element_v16f32_#Index : Extract_Element <
    f32, v16f32, Index, !cast<SubRegIndex>(sub#Index)
  >;
  def Insert_Element_v16f32_#Index : Insert_Element <
    f32, v16f32, Index, !cast<SubRegIndex>(sub#Index)
  >;
}

def : BitConvert <i32, f32, SReg_32>;
def : BitConvert <i32, f32, VReg_32>;

def : BitConvert <f32, i32, SReg_32>;
def : BitConvert <f32, i32, VReg_32>;

def : BitConvert <i64, f64, VReg_64>;

def : BitConvert <f64, i64, VReg_64>;

def : BitConvert <v2f32, v2i32, VReg_64>;
def : BitConvert <v2i32, v2f32, VReg_64>;
def : BitConvert <v2i32, i64, VReg_64>;
def : BitConvert <i64, v2i32, VReg_64>;
def : BitConvert <v2f32, i64, VReg_64>;
def : BitConvert <i64, v2f32, VReg_64>;
def : BitConvert <v2i32, f64, VReg_64>;
def : BitConvert <f64, v2i32, VReg_64>;
def : BitConvert <v4f32, v4i32, VReg_128>;
def : BitConvert <v4i32, v4f32, VReg_128>;

def : BitConvert <v8f32, v8i32, SReg_256>;
def : BitConvert <v8i32, v8f32, SReg_256>;
def : BitConvert <v8i32, v32i8, SReg_256>;
def : BitConvert <v32i8, v8i32, SReg_256>;
def : BitConvert <v8i32, v32i8, VReg_256>;
def : BitConvert <v8i32, v8f32, VReg_256>;
def : BitConvert <v8f32, v8i32, VReg_256>;
def : BitConvert <v32i8, v8i32, VReg_256>;

def : BitConvert <v16i32, v16f32, VReg_512>;
def : BitConvert <v16f32, v16i32, VReg_512>;

/********** =================== **********/
/********** Src & Dst modifiers **********/
/********** =================== **********/

def : Pat <
  (AMDGPUclamp (VOP3Mods0Clamp f32:$src0, i32:$src0_modifiers, i32:$omod),
               (f32 FP_ZERO), (f32 FP_ONE)),
  (V_ADD_F32_e64 $src0_modifiers, $src0, 0, 0, 1, $omod)
>;

/********** ================================ **********/
/********** Floating point absolute/negative **********/
/********** ================================ **********/

// Prevent expanding both fneg and fabs.

// FIXME: Should use S_OR_B32
def : Pat <
  (fneg (fabs f32:$src)),
  (V_OR_B32_e32 $src, (V_MOV_B32_e32 0x80000000)) /* Set sign bit */
>;

// FIXME: Should use S_OR_B32
def : Pat <
  (fneg (fabs f64:$src)),
  (REG_SEQUENCE VReg_64,
    (i32 (EXTRACT_SUBREG f64:$src, sub0)),
    sub0,
    (V_OR_B32_e32 (EXTRACT_SUBREG f64:$src, sub1),
                  (V_MOV_B32_e32 0x80000000)), // Set sign bit.
    sub1)
>;

def : Pat <
  (fabs f32:$src),
  (V_AND_B32_e32 $src, (V_MOV_B32_e32 0x7fffffff))
>;

def : Pat <
  (fneg f32:$src),
  (V_XOR_B32_e32 $src, (V_MOV_B32_e32 0x80000000))
>;

def : Pat <
  (fabs f64:$src),
  (REG_SEQUENCE VReg_64,
    (i32 (EXTRACT_SUBREG f64:$src, sub0)),
    sub0,
    (V_AND_B32_e32 (EXTRACT_SUBREG f64:$src, sub1),
                   (V_MOV_B32_e32 0x7fffffff)), // Set sign bit.
     sub1)
>;

def : Pat <
  (fneg f64:$src),
  (REG_SEQUENCE VReg_64,
    (i32 (EXTRACT_SUBREG f64:$src, sub0)),
    sub0,
    (V_XOR_B32_e32 (EXTRACT_SUBREG f64:$src, sub1),
                   (V_MOV_B32_e32 0x80000000)),
    sub1)
>;

/********** ================== **********/
/********** Immediate Patterns **********/
/********** ================== **********/

def : Pat <
  (SGPRImm<(i32 imm)>:$imm),
  (S_MOV_B32 imm:$imm)
>;

def : Pat <
  (SGPRImm<(f32 fpimm)>:$imm),
  (S_MOV_B32 fpimm:$imm)
>;

def : Pat <
  (i32 imm:$imm),
  (V_MOV_B32_e32 imm:$imm)
>;

def : Pat <
  (f32 fpimm:$imm),
  (V_MOV_B32_e32 fpimm:$imm)
>;

def : Pat <
  (i64 InlineImm<i64>:$imm),
  (S_MOV_B64 InlineImm<i64>:$imm)
>;

/********** ===================== **********/
/********** Interpolation Paterns **********/
/********** ===================== **********/

def : Pat <
  (int_SI_fs_constant imm:$attr_chan, imm:$attr, i32:$params),
  (V_INTERP_MOV_F32 INTERP.P0, imm:$attr_chan, imm:$attr, $params)
>;

def : Pat <
  (int_SI_fs_interp imm:$attr_chan, imm:$attr, M0Reg:$params, v2i32:$ij),
  (V_INTERP_P2_F32 (V_INTERP_P1_F32 (EXTRACT_SUBREG v2i32:$ij, sub0),
                                    imm:$attr_chan, imm:$attr, i32:$params),
                   (EXTRACT_SUBREG $ij, sub1),
                   imm:$attr_chan, imm:$attr, $params)
>;

/********** ================== **********/
/********** Intrinsic Patterns **********/
/********** ================== **********/

/* llvm.AMDGPU.pow */
def : POW_Common <V_LOG_F32_e32, V_EXP_F32_e32, V_MUL_LEGACY_F32_e32>;

def : Pat <
  (int_AMDGPU_div f32:$src0, f32:$src1),
  (V_MUL_LEGACY_F32_e32 $src0, (V_RCP_LEGACY_F32_e32 $src1))
>;

def : Pat<
  (fdiv f64:$src0, f64:$src1),
  (V_MUL_F64 0 /* src0_modifiers */, $src0,
             0 /* src1_modifiers */, (V_RCP_F64_e32 $src1),
             0 /* clamp */, 0 /* omod */)
>;

def : Pat <
  (int_AMDGPU_cube v4f32:$src),
  (REG_SEQUENCE VReg_128,
    (V_CUBETC_F32 0 /* src0_modifiers */, (EXTRACT_SUBREG $src, sub0),
                  0 /* src1_modifiers */, (EXTRACT_SUBREG $src, sub1),
                  0 /* src2_modifiers */, (EXTRACT_SUBREG $src, sub2),
                  0 /* clamp */, 0 /* omod */), sub0,
    (V_CUBESC_F32 0 /* src0_modifiers */, (EXTRACT_SUBREG $src, sub0),
                  0 /* src1_modifiers */,(EXTRACT_SUBREG $src, sub1),
                  0 /* src2_modifiers */,(EXTRACT_SUBREG $src, sub2),
                  0 /* clamp */, 0 /* omod */), sub1,
    (V_CUBEMA_F32 0 /* src1_modifiers */,(EXTRACT_SUBREG $src, sub0),
                  0 /* src1_modifiers */,(EXTRACT_SUBREG $src, sub1),
                  0 /* src1_modifiers */,(EXTRACT_SUBREG $src, sub2),
                  0 /* clamp */, 0 /* omod */), sub2,
    (V_CUBEID_F32 0 /* src1_modifiers */,(EXTRACT_SUBREG $src, sub0),
                  0 /* src1_modifiers */,(EXTRACT_SUBREG $src, sub1),
                  0 /* src1_modifiers */,(EXTRACT_SUBREG $src, sub2),
                  0 /* clamp */, 0 /* omod */), sub3)
>;

def : Pat <
  (i32 (sext i1:$src0)),
  (V_CNDMASK_B32_e64 (i32 0), (i32 -1), $src0)
>;

class Ext32Pat <SDNode ext> : Pat <
  (i32 (ext i1:$src0)),
  (V_CNDMASK_B32_e64 (i32 0), (i32 1), $src0)
>;

def : Ext32Pat <zext>;
def : Ext32Pat <anyext>;

// Offset in an 32Bit VGPR
def : Pat <
  (SIload_constant v4i32:$sbase, i32:$voff),
  (BUFFER_LOAD_DWORD_OFFEN $sbase, $voff, 0, 0, 0, 0, 0)
>;

// The multiplication scales from [0,1] to the unsigned integer range
def : Pat <
  (AMDGPUurecip i32:$src0),
  (V_CVT_U32_F32_e32
    (V_MUL_F32_e32 CONST.FP_UINT_MAX_PLUS_1,
                   (V_RCP_IFLAG_F32_e32 (V_CVT_F32_U32_e32 $src0))))
>;

def : Pat <
  (int_SI_tid),
  (V_MBCNT_HI_U32_B32_e32 0xffffffff,
                          (V_MBCNT_LO_U32_B32_e64 0xffffffff, 0))
>;

//===----------------------------------------------------------------------===//
// VOP3 Patterns
//===----------------------------------------------------------------------===//

def : IMad24Pat<V_MAD_I32_I24>;
def : UMad24Pat<V_MAD_U32_U24>;

def : Pat <
  (mulhu i32:$src0, i32:$src1),
  (V_MUL_HI_U32 $src0, $src1)
>;

def : Pat <
  (mulhs i32:$src0, i32:$src1),
  (V_MUL_HI_I32 $src0, $src1)
>;

def : Vop3ModPat<V_MAD_F32, VOP_F32_F32_F32_F32, AMDGPUmad>;


defm : BFIPatterns <V_BFI_B32, S_MOV_B32, SReg_64>;
def : ROTRPattern <V_ALIGNBIT_B32>;

/********** ======================= **********/
/**********   Load/Store Patterns   **********/
/********** ======================= **********/

class DSReadPat <DS inst, ValueType vt, PatFrag frag> : Pat <
  (vt (frag (DS1Addr1Offset i32:$ptr, i32:$offset))),
  (inst (i1 0), $ptr, (as_i16imm $offset), (S_MOV_B32 -1))
>;

def : DSReadPat <DS_READ_I8,  i32, sextloadi8_local>;
def : DSReadPat <DS_READ_U8,  i32, az_extloadi8_local>;
def : DSReadPat <DS_READ_I16, i32, sextloadi16_local>;
def : DSReadPat <DS_READ_U16, i32, az_extloadi16_local>;
def : DSReadPat <DS_READ_B32, i32, local_load>;

let AddedComplexity = 100 in {

def : DSReadPat <DS_READ_B64, v2i32, local_load_aligned8bytes>;

} // End AddedComplexity = 100

def : Pat <
  (v2i32 (local_load (DS64Bit4ByteAligned i32:$ptr, i8:$offset0,
                                                    i8:$offset1))),
  (DS_READ2_B32 (i1 0), $ptr, $offset0, $offset1, (S_MOV_B32 -1))
>;

class DSWritePat <DS inst, ValueType vt, PatFrag frag> : Pat <
  (frag vt:$value, (DS1Addr1Offset i32:$ptr, i32:$offset)),
  (inst (i1 0), $ptr, $value, (as_i16imm $offset), (S_MOV_B32 -1))
>;

def : DSWritePat <DS_WRITE_B8, i32, truncstorei8_local>;
def : DSWritePat <DS_WRITE_B16, i32, truncstorei16_local>;
def : DSWritePat <DS_WRITE_B32, i32, local_store>;

let AddedComplexity = 100 in {

def : DSWritePat <DS_WRITE_B64, v2i32, local_store_aligned8bytes>;
} // End AddedComplexity = 100

def : Pat <
  (local_store v2i32:$value, (DS64Bit4ByteAligned i32:$ptr, i8:$offset0,
                                                            i8:$offset1)),
  (DS_WRITE2_B32 (i1 0), $ptr, (EXTRACT_SUBREG $value, sub0),
                        (EXTRACT_SUBREG $value, sub1), $offset0, $offset1,
                        (S_MOV_B32 -1))
>;

class DSAtomicRetPat<DS inst, ValueType vt, PatFrag frag> : Pat <
  (frag (DS1Addr1Offset i32:$ptr, i32:$offset), vt:$value),
  (inst (i1 0), $ptr, $value, (as_i16imm $offset), (S_MOV_B32 -1))
>;

// Special case of DSAtomicRetPat for add / sub 1 -> inc / dec
//
// We need to use something for the data0, so we set a register to
// -1. For the non-rtn variants, the manual says it does
// DS[A] = (DS[A] >= D0) ? 0 : DS[A] + 1, and setting D0 to uint_max
// will always do the increment so I'm assuming it's the same.
//
// We also load this -1 with s_mov_b32 / s_mov_b64 even though this
// needs to be a VGPR. The SGPR copy pass will fix this, and it's
// easier since there is no v_mov_b64.
class DSAtomicIncRetPat<DS inst, ValueType vt,
                        Instruction LoadImm, PatFrag frag> : Pat <
  (frag (DS1Addr1Offset i32:$ptr, i32:$offset), (vt 1)),
  (inst (i1 0), $ptr, (LoadImm (vt -1)), (as_i16imm $offset), (S_MOV_B32 -1))
>;


class DSAtomicCmpXChg <DS inst, ValueType vt, PatFrag frag> : Pat <
  (frag (DS1Addr1Offset i32:$ptr, i32:$offset), vt:$cmp, vt:$swap),
  (inst (i1 0), $ptr, $cmp, $swap, (as_i16imm $offset), (S_MOV_B32 -1))
>;


// 32-bit atomics.
def : DSAtomicIncRetPat<DS_INC_RTN_U32, i32,
                        S_MOV_B32, atomic_load_add_local>;
def : DSAtomicIncRetPat<DS_DEC_RTN_U32, i32,
                        S_MOV_B32, atomic_load_sub_local>;

def : DSAtomicRetPat<DS_WRXCHG_RTN_B32, i32, atomic_swap_local>;
def : DSAtomicRetPat<DS_ADD_RTN_U32, i32, atomic_load_add_local>;
def : DSAtomicRetPat<DS_SUB_RTN_U32, i32, atomic_load_sub_local>;
def : DSAtomicRetPat<DS_AND_RTN_B32, i32, atomic_load_and_local>;
def : DSAtomicRetPat<DS_OR_RTN_B32, i32, atomic_load_or_local>;
def : DSAtomicRetPat<DS_XOR_RTN_B32, i32, atomic_load_xor_local>;
def : DSAtomicRetPat<DS_MIN_RTN_I32, i32, atomic_load_min_local>;
def : DSAtomicRetPat<DS_MAX_RTN_I32, i32, atomic_load_max_local>;
def : DSAtomicRetPat<DS_MIN_RTN_U32, i32, atomic_load_umin_local>;
def : DSAtomicRetPat<DS_MAX_RTN_U32, i32, atomic_load_umax_local>;

def : DSAtomicCmpXChg<DS_CMPST_RTN_B32, i32, atomic_cmp_swap_32_local>;

// 64-bit atomics.
def : DSAtomicIncRetPat<DS_INC_RTN_U64, i64,
                        S_MOV_B64, atomic_load_add_local>;
def : DSAtomicIncRetPat<DS_DEC_RTN_U64, i64,
                        S_MOV_B64, atomic_load_sub_local>;

def : DSAtomicRetPat<DS_WRXCHG_RTN_B64, i64, atomic_swap_local>;
def : DSAtomicRetPat<DS_ADD_RTN_U64, i64, atomic_load_add_local>;
def : DSAtomicRetPat<DS_SUB_RTN_U64, i64, atomic_load_sub_local>;
def : DSAtomicRetPat<DS_AND_RTN_B64, i64, atomic_load_and_local>;
def : DSAtomicRetPat<DS_OR_RTN_B64, i64, atomic_load_or_local>;
def : DSAtomicRetPat<DS_XOR_RTN_B64, i64, atomic_load_xor_local>;
def : DSAtomicRetPat<DS_MIN_RTN_I64, i64, atomic_load_min_local>;
def : DSAtomicRetPat<DS_MAX_RTN_I64, i64, atomic_load_max_local>;
def : DSAtomicRetPat<DS_MIN_RTN_U64, i64, atomic_load_umin_local>;
def : DSAtomicRetPat<DS_MAX_RTN_U64, i64, atomic_load_umax_local>;

def : DSAtomicCmpXChg<DS_CMPST_RTN_B64, i64, atomic_cmp_swap_64_local>;


//===----------------------------------------------------------------------===//
// MUBUF Patterns
//===----------------------------------------------------------------------===//

multiclass MUBUFLoad_Pattern <MUBUF Instr_ADDR64, ValueType vt,
                              PatFrag constant_ld> {
  def : Pat <
     (vt (constant_ld (MUBUFAddr64 v4i32:$srsrc, i64:$vaddr, i16:$offset))),
     (Instr_ADDR64 $srsrc, $vaddr, $offset)
  >;
}

defm : MUBUFLoad_Pattern <BUFFER_LOAD_SBYTE_ADDR64, i32, sextloadi8_constant>;
defm : MUBUFLoad_Pattern <BUFFER_LOAD_UBYTE_ADDR64, i32, az_extloadi8_constant>;
defm : MUBUFLoad_Pattern <BUFFER_LOAD_SSHORT_ADDR64, i32, sextloadi16_constant>;
defm : MUBUFLoad_Pattern <BUFFER_LOAD_USHORT_ADDR64, i32, az_extloadi16_constant>;
defm : MUBUFLoad_Pattern <BUFFER_LOAD_DWORD_ADDR64, i32, constant_load>;
defm : MUBUFLoad_Pattern <BUFFER_LOAD_DWORDX2_ADDR64, v2i32, constant_load>;
defm : MUBUFLoad_Pattern <BUFFER_LOAD_DWORDX4_ADDR64, v4i32, constant_load>;

class MUBUFScratchLoadPat <MUBUF Instr, ValueType vt, PatFrag ld> : Pat <
  (vt (ld (MUBUFScratch v4i32:$srsrc, i32:$vaddr,
                        i32:$soffset, u16imm:$offset))),
  (Instr $srsrc, $vaddr, $soffset, $offset, 0, 0, 0)
>;

def : MUBUFScratchLoadPat <BUFFER_LOAD_SBYTE_OFFEN, i32, sextloadi8_private>;
def : MUBUFScratchLoadPat <BUFFER_LOAD_UBYTE_OFFEN, i32, extloadi8_private>;
def : MUBUFScratchLoadPat <BUFFER_LOAD_SSHORT_OFFEN, i32, sextloadi16_private>;
def : MUBUFScratchLoadPat <BUFFER_LOAD_USHORT_OFFEN, i32, extloadi16_private>;
def : MUBUFScratchLoadPat <BUFFER_LOAD_DWORD_OFFEN, i32, load_private>;
def : MUBUFScratchLoadPat <BUFFER_LOAD_DWORDX2_OFFEN, v2i32, load_private>;
def : MUBUFScratchLoadPat <BUFFER_LOAD_DWORDX4_OFFEN, v4i32, load_private>;

// BUFFER_LOAD_DWORD*, addr64=0
multiclass MUBUF_Load_Dword <ValueType vt, MUBUF offset, MUBUF offen, MUBUF idxen,
                             MUBUF bothen> {

  def : Pat <
    (vt (int_SI_buffer_load_dword v4i32:$rsrc, (i32 imm), i32:$soffset,
                                  imm:$offset, 0, 0, imm:$glc, imm:$slc,
                                  imm:$tfe)),
    (offset $rsrc, (as_i16imm $offset), $soffset, (as_i1imm $glc),
            (as_i1imm $slc), (as_i1imm $tfe))
  >;

  def : Pat <
    (vt (int_SI_buffer_load_dword v4i32:$rsrc, i32:$vaddr, i32:$soffset,
                                  imm:$offset, 1, 0, imm:$glc, imm:$slc,
                                  imm:$tfe)),
    (offen $rsrc, $vaddr, $soffset, (as_i16imm $offset), (as_i1imm $glc), (as_i1imm $slc),
           (as_i1imm $tfe))
  >;

  def : Pat <
    (vt (int_SI_buffer_load_dword v4i32:$rsrc, i32:$vaddr, i32:$soffset,
                                  imm:$offset, 0, 1, imm:$glc, imm:$slc,
                                  imm:$tfe)),
    (idxen $rsrc, $vaddr, (as_i16imm $offset), $soffset, (as_i1imm $glc),
           (as_i1imm $slc), (as_i1imm $tfe))
  >;

  def : Pat <
    (vt (int_SI_buffer_load_dword v4i32:$rsrc, v2i32:$vaddr, i32:$soffset,
                                  imm, 1, 1, imm:$glc, imm:$slc,
                                  imm:$tfe)),
    (bothen $rsrc, $vaddr, $soffset, (as_i1imm $glc), (as_i1imm $slc),
            (as_i1imm $tfe))
  >;
}

defm : MUBUF_Load_Dword <i32, BUFFER_LOAD_DWORD_OFFSET, BUFFER_LOAD_DWORD_OFFEN,
                         BUFFER_LOAD_DWORD_IDXEN, BUFFER_LOAD_DWORD_BOTHEN>;
defm : MUBUF_Load_Dword <v2i32, BUFFER_LOAD_DWORDX2_OFFSET, BUFFER_LOAD_DWORDX2_OFFEN,
                         BUFFER_LOAD_DWORDX2_IDXEN, BUFFER_LOAD_DWORDX2_BOTHEN>;
defm : MUBUF_Load_Dword <v4i32, BUFFER_LOAD_DWORDX4_OFFSET, BUFFER_LOAD_DWORDX4_OFFEN,
                         BUFFER_LOAD_DWORDX4_IDXEN, BUFFER_LOAD_DWORDX4_BOTHEN>;

class MUBUFScratchStorePat <MUBUF Instr, ValueType vt, PatFrag st> : Pat <
  (st vt:$value, (MUBUFScratch v4i32:$srsrc, i32:$vaddr, i32:$soffset,
                               u16imm:$offset)),
  (Instr $value, $srsrc, $vaddr, $soffset, $offset, 0, 0, 0)
>;

def : MUBUFScratchStorePat <BUFFER_STORE_BYTE_OFFEN, i32, truncstorei8_private>;
def : MUBUFScratchStorePat <BUFFER_STORE_SHORT_OFFEN, i32, truncstorei16_private>;
def : MUBUFScratchStorePat <BUFFER_STORE_DWORD_OFFEN, i32, store_private>;
def : MUBUFScratchStorePat <BUFFER_STORE_DWORDX2_OFFEN, v2i32, store_private>;
def : MUBUFScratchStorePat <BUFFER_STORE_DWORDX4_OFFEN, v4i32, store_private>;

/*
class MUBUFStore_Pattern <MUBUF Instr, ValueType vt, PatFrag st> : Pat <
  (st vt:$value, (MUBUFScratch v4i32:$srsrc, i64:$vaddr, u16imm:$offset)),
  (Instr $value, $srsrc, $vaddr, $offset)
>;

def : MUBUFStore_Pattern <BUFFER_STORE_BYTE_ADDR64, i32, truncstorei8_private>;
def : MUBUFStore_Pattern <BUFFER_STORE_SHORT_ADDR64, i32, truncstorei16_private>;
def : MUBUFStore_Pattern <BUFFER_STORE_DWORD_ADDR64, i32, store_private>;
def : MUBUFStore_Pattern <BUFFER_STORE_DWORDX2_ADDR64, v2i32, store_private>;
def : MUBUFStore_Pattern <BUFFER_STORE_DWORDX4_ADDR64, v4i32, store_private>;

*/

//===----------------------------------------------------------------------===//
// MTBUF Patterns
//===----------------------------------------------------------------------===//

// TBUFFER_STORE_FORMAT_*, addr64=0
class MTBUF_StoreResource <ValueType vt, int num_channels, MTBUF opcode> : Pat<
  (SItbuffer_store v4i32:$rsrc, vt:$vdata, num_channels, i32:$vaddr,
                   i32:$soffset, imm:$inst_offset, imm:$dfmt,
                   imm:$nfmt, imm:$offen, imm:$idxen,
                   imm:$glc, imm:$slc, imm:$tfe),
  (opcode
    $vdata, (as_i16imm $inst_offset), (as_i1imm $offen), (as_i1imm $idxen),
    (as_i1imm $glc), 0, (as_i8imm $dfmt), (as_i8imm $nfmt), $vaddr, $rsrc,
    (as_i1imm $slc), (as_i1imm $tfe), $soffset)
>;

def : MTBUF_StoreResource <i32, 1, TBUFFER_STORE_FORMAT_X>;
def : MTBUF_StoreResource <v2i32, 2, TBUFFER_STORE_FORMAT_XY>;
def : MTBUF_StoreResource <v4i32, 3, TBUFFER_STORE_FORMAT_XYZ>;
def : MTBUF_StoreResource <v4i32, 4, TBUFFER_STORE_FORMAT_XYZW>;

let SubtargetPredicate = isCI in {

// Sea island new arithmetic instructinos
defm V_TRUNC_F64 : VOP1Inst <vop1<0x17>, "v_trunc_f64",
  VOP_F64_F64, ftrunc
>;
defm V_CEIL_F64 : VOP1Inst <vop1<0x18>, "v_ceil_f64",
  VOP_F64_F64, fceil
>;
defm V_FLOOR_F64 : VOP1Inst <vop1<0x1A>, "v_floor_f64",
  VOP_F64_F64, ffloor
>;
defm V_RNDNE_F64 : VOP1Inst <vop1<0x19>, "v_rndne_f64",
  VOP_F64_F64, frint
>;

defm V_QSAD_PK_U16_U8 : VOP3Inst <vop3<0x173>, "v_qsad_pk_u16_u8",
  VOP_I32_I32_I32
>;
defm V_MQSAD_U16_U8 : VOP3Inst <vop3<0x172>, "v_mqsad_u16_u8",
  VOP_I32_I32_I32
>;
defm V_MQSAD_U32_U8 : VOP3Inst <vop3<0x175>, "v_mqsad_u32_u8",
  VOP_I32_I32_I32
>;

let isCommutable = 1 in {
defm V_MAD_U64_U32 : VOP3Inst <vop3<0x176>, "v_mad_u64_u32",
  VOP_I64_I32_I32_I64
>;

// XXX - Does this set VCC?
defm V_MAD_I64_I32 : VOP3Inst <vop3<0x177>, "v_mad_i64_i32",
  VOP_I64_I32_I32_I64
>;
} // End isCommutable = 1

// Remaining instructions:
// FLAT_*
// S_CBRANCH_CDBGUSER
// S_CBRANCH_CDBGSYS
// S_CBRANCH_CDBGSYS_OR_USER
// S_CBRANCH_CDBGSYS_AND_USER
// S_DCACHE_INV_VOL
// V_EXP_LEGACY_F32
// V_LOG_LEGACY_F32
// DS_NOP
// DS_GWS_SEMA_RELEASE_ALL
// DS_WRAP_RTN_B32
// DS_CNDXCHG32_RTN_B64
// DS_WRITE_B96
// DS_WRITE_B128
// DS_CONDXCHG32_RTN_B128
// DS_READ_B96
// DS_READ_B128
// BUFFER_LOAD_DWORDX3
// BUFFER_STORE_DWORDX3

} // End iSCI

//===----------------------------------------------------------------------===//
// Flat Patterns
//===----------------------------------------------------------------------===//

class FLATLoad_Pattern <FLAT Instr_ADDR64, ValueType vt,
                             PatFrag flat_ld> :
  Pat <(vt (flat_ld i64:$ptr)),
       (Instr_ADDR64 $ptr)
>;

def : FLATLoad_Pattern <FLAT_LOAD_SBYTE, i32, sextloadi8_flat>;
def : FLATLoad_Pattern <FLAT_LOAD_UBYTE, i32, az_extloadi8_flat>;
def : FLATLoad_Pattern <FLAT_LOAD_SSHORT, i32, sextloadi16_flat>;
def : FLATLoad_Pattern <FLAT_LOAD_USHORT, i32, az_extloadi16_flat>;
def : FLATLoad_Pattern <FLAT_LOAD_DWORD, i32, flat_load>;
def : FLATLoad_Pattern <FLAT_LOAD_DWORDX2, i64, flat_load>;
def : FLATLoad_Pattern <FLAT_LOAD_DWORDX2, i64, az_extloadi32_flat>;
def : FLATLoad_Pattern <FLAT_LOAD_DWORDX2, v2i32, flat_load>;
def : FLATLoad_Pattern <FLAT_LOAD_DWORDX4, v4i32, flat_load>;

class FLATStore_Pattern <FLAT Instr, ValueType vt, PatFrag st> :
  Pat <(st vt:$value, i64:$ptr),
        (Instr $value, $ptr)
  >;

def : FLATStore_Pattern <FLAT_STORE_BYTE, i32, truncstorei8_flat>;
def : FLATStore_Pattern <FLAT_STORE_SHORT, i32, truncstorei16_flat>;
def : FLATStore_Pattern <FLAT_STORE_DWORD, i32, flat_store>;
def : FLATStore_Pattern <FLAT_STORE_DWORDX2, i64, flat_store>;
def : FLATStore_Pattern <FLAT_STORE_DWORDX2, v2i32, flat_store>;
def : FLATStore_Pattern <FLAT_STORE_DWORDX4, v4i32, flat_store>;

/********** ====================== **********/
/**********   Indirect adressing   **********/
/********** ====================== **********/

multiclass SI_INDIRECT_Pattern <ValueType vt, ValueType eltvt, SI_INDIRECT_DST IndDst> {

  // 1. Extract with offset
  def : Pat<
    (vector_extract vt:$vec, (add i32:$idx, imm:$off)),
    (eltvt (SI_INDIRECT_SRC (IMPLICIT_DEF), $vec, $idx, imm:$off))
  >;

  // 2. Extract without offset
  def : Pat<
    (vector_extract vt:$vec, i32:$idx),
    (eltvt (SI_INDIRECT_SRC (IMPLICIT_DEF), $vec, $idx, 0))
  >;

  // 3. Insert with offset
  def : Pat<
    (vector_insert vt:$vec, eltvt:$val, (add i32:$idx, imm:$off)),
    (IndDst (IMPLICIT_DEF), $vec, $idx, imm:$off, $val)
  >;

  // 4. Insert without offset
  def : Pat<
    (vector_insert vt:$vec, eltvt:$val, i32:$idx),
    (IndDst (IMPLICIT_DEF), $vec, $idx, 0, $val)
  >;
}

defm : SI_INDIRECT_Pattern <v2f32, f32, SI_INDIRECT_DST_V2>;
defm : SI_INDIRECT_Pattern <v4f32, f32, SI_INDIRECT_DST_V4>;
defm : SI_INDIRECT_Pattern <v8f32, f32, SI_INDIRECT_DST_V8>;
defm : SI_INDIRECT_Pattern <v16f32, f32, SI_INDIRECT_DST_V16>;

defm : SI_INDIRECT_Pattern <v2i32, i32, SI_INDIRECT_DST_V2>;
defm : SI_INDIRECT_Pattern <v4i32, i32, SI_INDIRECT_DST_V4>;
defm : SI_INDIRECT_Pattern <v8i32, i32, SI_INDIRECT_DST_V8>;
defm : SI_INDIRECT_Pattern <v16i32, i32, SI_INDIRECT_DST_V16>;

//===----------------------------------------------------------------------===//
// Conversion Patterns
//===----------------------------------------------------------------------===//

def : Pat<(i32 (sext_inreg i32:$src, i1)),
  (S_BFE_I32 i32:$src, 65536)>; // 0 | 1 << 16

// Handle sext_inreg in i64
def : Pat <
  (i64 (sext_inreg i64:$src, i1)),
  (S_BFE_I64 i64:$src, 0x10000) // 0 | 1 << 16
>;

def : Pat <
  (i64 (sext_inreg i64:$src, i8)),
  (S_BFE_I64 i64:$src, 0x80000) // 0 | 8 << 16
>;

def : Pat <
  (i64 (sext_inreg i64:$src, i16)),
  (S_BFE_I64 i64:$src, 0x100000) // 0 | 16 << 16
>;

def : Pat <
  (i64 (sext_inreg i64:$src, i32)),
  (S_BFE_I64 i64:$src, 0x200000) // 0 | 32 << 16
>;

class ZExt_i64_i32_Pat <SDNode ext> : Pat <
  (i64 (ext i32:$src)),
  (REG_SEQUENCE SReg_64, $src, sub0, (S_MOV_B32 0), sub1)
>;

class ZExt_i64_i1_Pat <SDNode ext> : Pat <
  (i64 (ext i1:$src)),
    (REG_SEQUENCE VReg_64,
      (V_CNDMASK_B32_e64 (i32 0), (i32 1), $src), sub0,
      (S_MOV_B32 0), sub1)
>;


def : ZExt_i64_i32_Pat<zext>;
def : ZExt_i64_i32_Pat<anyext>;
def : ZExt_i64_i1_Pat<zext>;
def : ZExt_i64_i1_Pat<anyext>;

def : Pat <
  (i64 (sext i32:$src)),
    (REG_SEQUENCE SReg_64, $src, sub0,
    (S_ASHR_I32 $src, 31), sub1)
>;

def : Pat <
  (i64 (sext i1:$src)),
  (REG_SEQUENCE VReg_64,
    (V_CNDMASK_B32_e64 0, -1, $src), sub0,
    (V_CNDMASK_B32_e64 0, -1, $src), sub1)
>;

def : Pat <
  (f32 (sint_to_fp i1:$src)),
  (V_CNDMASK_B32_e64 (i32 0), CONST.FP32_NEG_ONE, $src)
>;

def : Pat <
  (f32 (uint_to_fp i1:$src)),
  (V_CNDMASK_B32_e64 (i32 0), CONST.FP32_ONE, $src)
>;

def : Pat <
  (f64 (sint_to_fp i1:$src)),
    (V_CVT_F64_I32_e32 (V_CNDMASK_B32_e64 (i32 0), (i32 -1), $src))
>;

def : Pat <
  (f64 (uint_to_fp i1:$src)),
  (V_CVT_F64_U32_e32 (V_CNDMASK_B32_e64 (i32 0), (i32 1), $src))
>;

//===----------------------------------------------------------------------===//
// Miscellaneous Patterns
//===----------------------------------------------------------------------===//

def : Pat <
  (i32 (trunc i64:$a)),
  (EXTRACT_SUBREG $a, sub0)
>;

def : Pat <
  (i1 (trunc i32:$a)),
  (V_CMP_EQ_I32_e64 (V_AND_B32_e64 (i32 1), $a), 1)
>;

def : Pat <
  (i32 (bswap i32:$a)),
  (V_BFI_B32 (S_MOV_B32 0x00ff00ff),
             (V_ALIGNBIT_B32 $a, $a, 24),
             (V_ALIGNBIT_B32 $a, $a, 8))
>;

//============================================================================//
// Miscellaneous Optimization Patterns
//============================================================================//

def : SHA256MaPattern <V_BFI_B32, V_XOR_B32_e64>;

} // End isSI predicate