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
|
//===-- SIInstrInfo.td -----------------------------------------------------===//
//
// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
// See https://llvm.org/LICENSE.txt for license information.
// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
//
//===----------------------------------------------------------------------===//
def isWave32 : Predicate<"Subtarget->getWavefrontSize() == 32">,
AssemblerPredicate <(all_of FeatureWavefrontSize32)>;
def isWave64 : Predicate<"Subtarget->getWavefrontSize() == 64">,
AssemblerPredicate <(all_of FeatureWavefrontSize64)>;
class GCNPredicateControl : PredicateControl {
Predicate SIAssemblerPredicate = isGFX6GFX7;
Predicate VIAssemblerPredicate = isGFX8GFX9;
}
// Except for the NONE field, this must be kept in sync with the
// SIEncodingFamily enum in SIInstrInfo.cpp and the columns of the
// getMCOpcodeGen table.
def SIEncodingFamily {
int NONE = -1;
int SI = 0;
int VI = 1;
int SDWA = 2;
int SDWA9 = 3;
int GFX80 = 4;
int GFX9 = 5;
int GFX10 = 6;
int SDWA10 = 7;
int GFX90A = 8;
int GFX940 = 9;
int GFX11 = 10;
int GFX12 = 11;
}
//===----------------------------------------------------------------------===//
// SI DAG Nodes
//===----------------------------------------------------------------------===//
def AMDGPUclamp : SDNode<"AMDGPUISD::CLAMP", SDTFPUnaryOp>;
def SDTSBufferLoad : SDTypeProfile<1, 3,
[ // vdata
SDTCisVT<1, v4i32>, // rsrc
SDTCisVT<2, i32>, // offset(imm)
SDTCisVT<3, i32>]>; // cachepolicy
def SIsbuffer_load : SDNode<"AMDGPUISD::SBUFFER_LOAD", SDTSBufferLoad,
[SDNPMayLoad, SDNPMemOperand]>;
def SIsbuffer_load_byte : SDNode<"AMDGPUISD::SBUFFER_LOAD_BYTE", SDTSBufferLoad,
[SDNPMayLoad, SDNPMemOperand]>;
def SIsbuffer_load_ubyte
: SDNode<"AMDGPUISD::SBUFFER_LOAD_UBYTE", SDTSBufferLoad,
[SDNPMayLoad, SDNPMemOperand]>;
def SIsbuffer_load_short
: SDNode<"AMDGPUISD::SBUFFER_LOAD_SHORT", SDTSBufferLoad,
[SDNPMayLoad, SDNPMemOperand]>;
def SIsbuffer_load_ushort
: SDNode<"AMDGPUISD::SBUFFER_LOAD_USHORT", SDTSBufferLoad,
[SDNPMayLoad, SDNPMemOperand]>;
def SIds_ordered_count : SDNode<"AMDGPUISD::DS_ORDERED_COUNT",
SDTypeProfile<1, 2, [SDTCisVT<0, i32>, SDTCisVT<1, i32>, SDTCisVT<2, i16>]>,
[SDNPMayLoad, SDNPMayStore, SDNPMemOperand, SDNPHasChain, SDNPInGlue]
>;
def SDTAtomic2_f32 : SDTypeProfile<1, 2, [
SDTCisSameAs<0,2>, SDTCisFP<0>, SDTCisPtrTy<1>
]>;
def SIatomic_fmin : SDNode<"AMDGPUISD::ATOMIC_LOAD_FMIN", SDTAtomic2_f32,
[SDNPMayLoad, SDNPMayStore, SDNPMemOperand, SDNPHasChain]
>;
def SIatomic_fmax : SDNode<"AMDGPUISD::ATOMIC_LOAD_FMAX", SDTAtomic2_f32,
[SDNPMayLoad, SDNPMayStore, SDNPMemOperand, SDNPHasChain]
>;
// load_d16_{lo|hi} ptr, tied_input
def SIload_d16 : SDTypeProfile<1, 2, [
SDTCisPtrTy<1>,
SDTCisSameAs<0, 2>
]>;
def SDTtbuffer_load : SDTypeProfile<1, 8,
[ // vdata
SDTCisVT<1, v4i32>, // rsrc
SDTCisVT<2, i32>, // vindex(VGPR)
SDTCisVT<3, i32>, // voffset(VGPR)
SDTCisVT<4, i32>, // soffset(SGPR)
SDTCisVT<5, i32>, // offset(imm)
SDTCisVT<6, i32>, // format(imm)
SDTCisVT<7, i32>, // cachepolicy, swizzled buffer(imm)
SDTCisVT<8, i1> // idxen(imm)
]>;
def SItbuffer_load : SDNode<"AMDGPUISD::TBUFFER_LOAD_FORMAT", SDTtbuffer_load,
[SDNPMayLoad, SDNPMemOperand, SDNPHasChain]>;
def SItbuffer_load_d16 : SDNode<"AMDGPUISD::TBUFFER_LOAD_FORMAT_D16",
SDTtbuffer_load,
[SDNPMayLoad, SDNPMemOperand, SDNPHasChain]>;
def SDTtbuffer_store : SDTypeProfile<0, 9,
[ // vdata
SDTCisVT<1, v4i32>, // rsrc
SDTCisVT<2, i32>, // vindex(VGPR)
SDTCisVT<3, i32>, // voffset(VGPR)
SDTCisVT<4, i32>, // soffset(SGPR)
SDTCisVT<5, i32>, // offset(imm)
SDTCisVT<6, i32>, // format(imm)
SDTCisVT<7, i32>, // cachepolicy, swizzled buffer(imm)
SDTCisVT<8, i1> // idxen(imm)
]>;
def SItbuffer_store : SDNode<"AMDGPUISD::TBUFFER_STORE_FORMAT", SDTtbuffer_store,
[SDNPMayStore, SDNPMemOperand, SDNPHasChain]>;
def SItbuffer_store_d16 : SDNode<"AMDGPUISD::TBUFFER_STORE_FORMAT_D16",
SDTtbuffer_store,
[SDNPMayStore, SDNPMemOperand, SDNPHasChain]>;
def SDTBufferLoad : SDTypeProfile<1, 7,
[ // vdata
SDTCisVT<1, v4i32>, // rsrc
SDTCisVT<2, i32>, // vindex(VGPR)
SDTCisVT<3, i32>, // voffset(VGPR)
SDTCisVT<4, i32>, // soffset(SGPR)
SDTCisVT<5, i32>, // offset(imm)
SDTCisVT<6, i32>, // cachepolicy, swizzled buffer(imm)
SDTCisVT<7, i1>]>; // idxen(imm)
def SIbuffer_load : SDNode <"AMDGPUISD::BUFFER_LOAD", SDTBufferLoad,
[SDNPMemOperand, SDNPHasChain, SDNPMayLoad]>;
def SIbuffer_load_ubyte : SDNode <"AMDGPUISD::BUFFER_LOAD_UBYTE", SDTBufferLoad,
[SDNPMemOperand, SDNPHasChain, SDNPMayLoad]>;
def SIbuffer_load_ushort : SDNode <"AMDGPUISD::BUFFER_LOAD_USHORT", SDTBufferLoad,
[SDNPMemOperand, SDNPHasChain, SDNPMayLoad]>;
def SIbuffer_load_byte : SDNode <"AMDGPUISD::BUFFER_LOAD_BYTE", SDTBufferLoad,
[SDNPMemOperand, SDNPHasChain, SDNPMayLoad]>;
def SIbuffer_load_short: SDNode <"AMDGPUISD::BUFFER_LOAD_SHORT", SDTBufferLoad,
[SDNPMemOperand, SDNPHasChain, SDNPMayLoad]>;
def SIbuffer_load_format : SDNode <"AMDGPUISD::BUFFER_LOAD_FORMAT", SDTBufferLoad,
[SDNPMemOperand, SDNPHasChain, SDNPMayLoad]>;
def SIbuffer_load_format_tfe : SDNode <"AMDGPUISD::BUFFER_LOAD_FORMAT_TFE", SDTBufferLoad,
[SDNPMemOperand, SDNPHasChain, SDNPMayLoad]>;
def SIbuffer_load_format_d16 : SDNode <"AMDGPUISD::BUFFER_LOAD_FORMAT_D16",
SDTBufferLoad,
[SDNPMemOperand, SDNPHasChain, SDNPMayLoad]>;
def SDTBufferStore : SDTypeProfile<0, 8,
[ // vdata
SDTCisVT<1, v4i32>, // rsrc
SDTCisVT<2, i32>, // vindex(VGPR)
SDTCisVT<3, i32>, // voffset(VGPR)
SDTCisVT<4, i32>, // soffset(SGPR)
SDTCisVT<5, i32>, // offset(imm)
SDTCisVT<6, i32>, // cachepolicy, swizzled buffer(imm)
SDTCisVT<7, i1>]>; // idxen(imm)
def SIbuffer_store : SDNode <"AMDGPUISD::BUFFER_STORE", SDTBufferStore,
[SDNPMayStore, SDNPMemOperand, SDNPHasChain]>;
def SIbuffer_store_byte: SDNode <"AMDGPUISD::BUFFER_STORE_BYTE",
SDTBufferStore,
[SDNPMayStore, SDNPMemOperand, SDNPHasChain]>;
def SIbuffer_store_short : SDNode <"AMDGPUISD::BUFFER_STORE_SHORT",
SDTBufferStore,
[SDNPMayStore, SDNPMemOperand, SDNPHasChain]>;
def SIbuffer_store_format : SDNode <"AMDGPUISD::BUFFER_STORE_FORMAT",
SDTBufferStore,
[SDNPMayStore, SDNPMemOperand, SDNPHasChain]>;
def SIbuffer_store_format_d16 : SDNode <"AMDGPUISD::BUFFER_STORE_FORMAT_D16",
SDTBufferStore,
[SDNPMayStore, SDNPMemOperand, SDNPHasChain]>;
multiclass SDBufferAtomic<string opcode> {
def "" : SDNode <opcode,
SDTypeProfile<1, 8,
[SDTCisVT<2, v4i32>, // rsrc
SDTCisVT<3, i32>, // vindex(VGPR)
SDTCisVT<4, i32>, // voffset(VGPR)
SDTCisVT<5, i32>, // soffset(SGPR)
SDTCisVT<6, i32>, // offset(imm)
SDTCisVT<7, i32>, // cachepolicy(imm)
SDTCisVT<8, i1>]>, // idxen(imm)
[SDNPMemOperand, SDNPHasChain, SDNPMayLoad, SDNPMayStore]
>;
def "_noret" : PatFrag<
(ops node:$vdata_in, node:$rsrc, node:$vindex, node:$voffset, node:$soffset,
node:$offset, node:$cachepolicy, node:$idxen),
(!cast<SDNode>(NAME) node:$vdata_in, node:$rsrc, node:$vindex,
node:$voffset, node:$soffset, node:$offset, node:$cachepolicy,
node:$idxen)> {
let HasNoUse = true;
}
}
defm SIbuffer_atomic_swap : SDBufferAtomic <"AMDGPUISD::BUFFER_ATOMIC_SWAP">;
defm SIbuffer_atomic_add : SDBufferAtomic <"AMDGPUISD::BUFFER_ATOMIC_ADD">;
defm SIbuffer_atomic_sub : SDBufferAtomic <"AMDGPUISD::BUFFER_ATOMIC_SUB">;
defm SIbuffer_atomic_smin : SDBufferAtomic <"AMDGPUISD::BUFFER_ATOMIC_SMIN">;
defm SIbuffer_atomic_umin : SDBufferAtomic <"AMDGPUISD::BUFFER_ATOMIC_UMIN">;
defm SIbuffer_atomic_smax : SDBufferAtomic <"AMDGPUISD::BUFFER_ATOMIC_SMAX">;
defm SIbuffer_atomic_umax : SDBufferAtomic <"AMDGPUISD::BUFFER_ATOMIC_UMAX">;
defm SIbuffer_atomic_and : SDBufferAtomic <"AMDGPUISD::BUFFER_ATOMIC_AND">;
defm SIbuffer_atomic_or : SDBufferAtomic <"AMDGPUISD::BUFFER_ATOMIC_OR">;
defm SIbuffer_atomic_xor : SDBufferAtomic <"AMDGPUISD::BUFFER_ATOMIC_XOR">;
defm SIbuffer_atomic_inc : SDBufferAtomic <"AMDGPUISD::BUFFER_ATOMIC_INC">;
defm SIbuffer_atomic_dec : SDBufferAtomic <"AMDGPUISD::BUFFER_ATOMIC_DEC">;
defm SIbuffer_atomic_csub : SDBufferAtomic <"AMDGPUISD::BUFFER_ATOMIC_CSUB">;
defm SIbuffer_atomic_fadd : SDBufferAtomic <"AMDGPUISD::BUFFER_ATOMIC_FADD">;
defm SIbuffer_atomic_fadd_bf16 : SDBufferAtomic <"AMDGPUISD::BUFFER_ATOMIC_FADD_BF16">;
defm SIbuffer_atomic_fmin : SDBufferAtomic <"AMDGPUISD::BUFFER_ATOMIC_FMIN">;
defm SIbuffer_atomic_fmax : SDBufferAtomic <"AMDGPUISD::BUFFER_ATOMIC_FMAX">;
defm SIbuffer_atomic_cond_sub_u32 : SDBufferAtomic <"AMDGPUISD::BUFFER_ATOMIC_COND_SUB_U32">;
def SIbuffer_atomic_cmpswap : SDNode <"AMDGPUISD::BUFFER_ATOMIC_CMPSWAP",
SDTypeProfile<1, 9,
[SDTCisVT<3, v4i32>, // rsrc
SDTCisVT<4, i32>, // vindex(VGPR)
SDTCisVT<5, i32>, // voffset(VGPR)
SDTCisVT<6, i32>, // soffset(SGPR)
SDTCisVT<7, i32>, // offset(imm)
SDTCisVT<8, i32>, // cachepolicy(imm)
SDTCisVT<9, i1>]>, // idxen(imm)
[SDNPMemOperand, SDNPHasChain, SDNPMayLoad, SDNPMayStore]
>;
def SIbuffer_atomic_cmpswap_noret : PatFrag<
(ops node:$src, node:$cmp, node:$rsrc, node:$vindex, node:$voffset,
node:$soffset, node:$offset, node:$cachepolicy, node:$idxen),
(SIbuffer_atomic_cmpswap node:$src, node:$cmp, node:$rsrc, node:$vindex,
node:$voffset, node:$soffset, node:$offset, node:$cachepolicy,
node:$idxen)> {
let HasNoUse = true;
}
class SDGlobalAtomicNoRtn<string opcode, ValueType ty> : SDNode <opcode,
SDTypeProfile<0, 2,
[SDTCisPtrTy<0>, // vaddr
SDTCisVT<1, ty>]>, // vdata
[SDNPMemOperand, SDNPHasChain, SDNPMayLoad, SDNPMayStore]
>;
def SIpc_add_rel_offset : SDNode<"AMDGPUISD::PC_ADD_REL_OFFSET",
SDTypeProfile<1, 2, [SDTCisVT<0, iPTR>, SDTCisSameAs<0,1>, SDTCisSameAs<0,2>]>
>;
def SIlds : SDNode<"AMDGPUISD::LDS",
SDTypeProfile<1, 1, [SDTCisVT<0, iPTR>, SDTCisSameAs<0,1>]>
>;
def SIload_d16_lo : SDNode<"AMDGPUISD::LOAD_D16_LO",
SIload_d16,
[SDNPMayLoad, SDNPMemOperand, SDNPHasChain]
>;
def SIload_d16_lo_u8 : SDNode<"AMDGPUISD::LOAD_D16_LO_U8",
SIload_d16,
[SDNPMayLoad, SDNPMemOperand, SDNPHasChain]
>;
def SIload_d16_lo_i8 : SDNode<"AMDGPUISD::LOAD_D16_LO_I8",
SIload_d16,
[SDNPMayLoad, SDNPMemOperand, SDNPHasChain]
>;
def SIload_d16_hi : SDNode<"AMDGPUISD::LOAD_D16_HI",
SIload_d16,
[SDNPMayLoad, SDNPMemOperand, SDNPHasChain]
>;
def SIload_d16_hi_u8 : SDNode<"AMDGPUISD::LOAD_D16_HI_U8",
SIload_d16,
[SDNPMayLoad, SDNPMemOperand, SDNPHasChain]
>;
def SIload_d16_hi_i8 : SDNode<"AMDGPUISD::LOAD_D16_HI_I8",
SIload_d16,
[SDNPMayLoad, SDNPMemOperand, SDNPHasChain]
>;
def SIdenorm_mode : SDNode<"AMDGPUISD::DENORM_MODE",
SDTypeProfile<0 ,1, [SDTCisInt<0>]>,
[SDNPHasChain, SDNPOptInGlue, SDNPOutGlue]
>;
def SIfptrunc_round_upward : SDNode<"AMDGPUISD::FPTRUNC_ROUND_UPWARD",
SDTFPRoundOp
>;
def SIfptrunc_round_downward : SDNode<"AMDGPUISD::FPTRUNC_ROUND_DOWNWARD",
SDTFPRoundOp
>;
//===----------------------------------------------------------------------===//
// ValueType helpers
//===----------------------------------------------------------------------===//
class isIntType<ValueType SrcVT> {
bit ret = !and(SrcVT.isInteger, !ne(SrcVT.Value, i1.Value));
}
//===----------------------------------------------------------------------===//
// PatFrags for global memory operations
//===----------------------------------------------------------------------===//
defm atomic_load_fmin : binary_atomic_op_all_as<SIatomic_fmin, 0>;
defm atomic_load_fmax : binary_atomic_op_all_as<SIatomic_fmax, 0>;
//===----------------------------------------------------------------------===//
// SDNodes PatFrags for loads/stores with a glue input.
// This is for SDNodes and PatFrag for local loads and stores to
// enable s_mov_b32 m0, -1 to be glued to the memory instructions.
//
// These mirror the regular load/store PatFrags and rely on special
// processing during Select() to add the glued copy.
//
//===----------------------------------------------------------------------===//
def AMDGPUld_glue : SDNode <"ISD::LOAD", SDTLoad,
[SDNPHasChain, SDNPMayLoad, SDNPMemOperand, SDNPInGlue]
>;
def AMDGPUatomic_ld_glue : SDNode <"ISD::ATOMIC_LOAD", SDTAtomicLoad,
[SDNPHasChain, SDNPMayLoad, SDNPMemOperand, SDNPInGlue]
>;
def unindexedload_glue : PatFrag <(ops node:$ptr), (AMDGPUld_glue node:$ptr)> {
let IsLoad = 1;
let IsUnindexed = 1;
}
def load_glue : PatFrag <(ops node:$ptr), (unindexedload_glue node:$ptr)> {
let IsLoad = 1;
let IsNonExtLoad = 1;
}
def atomic_load_8_glue : PatFrag<(ops node:$ptr),
(AMDGPUatomic_ld_glue node:$ptr)> {
let IsAtomic = 1;
let MemoryVT = i8;
}
def atomic_load_16_glue : PatFrag<(ops node:$ptr),
(AMDGPUatomic_ld_glue node:$ptr)> {
let IsAtomic = 1;
let MemoryVT = i16;
}
def atomic_load_32_glue : PatFrag<(ops node:$ptr),
(AMDGPUatomic_ld_glue node:$ptr)> {
let IsAtomic = 1;
let MemoryVT = i32;
}
def atomic_load_64_glue : PatFrag<(ops node:$ptr),
(AMDGPUatomic_ld_glue node:$ptr)> {
let IsAtomic = 1;
let MemoryVT = i64;
}
def extload_glue : PatFrag<(ops node:$ptr), (unindexedload_glue node:$ptr)> {
let IsLoad = 1;
let IsAnyExtLoad = 1;
}
def sextload_glue : PatFrag<(ops node:$ptr), (unindexedload_glue node:$ptr)> {
let IsLoad = 1;
let IsSignExtLoad = 1;
}
def zextload_glue : PatFrag<(ops node:$ptr), (unindexedload_glue node:$ptr)> {
let IsLoad = 1;
let IsZeroExtLoad = 1;
}
def extloadi8_glue : PatFrag<(ops node:$ptr), (extload_glue node:$ptr)> {
let IsLoad = 1;
let MemoryVT = i8;
}
def zextloadi8_glue : PatFrag<(ops node:$ptr), (zextload_glue node:$ptr)> {
let IsLoad = 1;
let MemoryVT = i8;
}
def extloadi16_glue : PatFrag<(ops node:$ptr), (extload_glue node:$ptr)> {
let IsLoad = 1;
let MemoryVT = i16;
}
def zextloadi16_glue : PatFrag<(ops node:$ptr), (zextload_glue node:$ptr)> {
let IsLoad = 1;
let MemoryVT = i16;
}
def sextloadi8_glue : PatFrag<(ops node:$ptr), (sextload_glue node:$ptr)> {
let IsLoad = 1;
let MemoryVT = i8;
}
def sextloadi16_glue : PatFrag<(ops node:$ptr), (sextload_glue node:$ptr)> {
let IsLoad = 1;
let MemoryVT = i16;
}
let IsLoad = 1, AddressSpaces = LoadAddress_local.AddrSpaces in {
def load_local_m0 : PatFrag<(ops node:$ptr), (load_glue node:$ptr)> {
let IsNonExtLoad = 1;
}
def extloadi8_local_m0 : PatFrag<(ops node:$ptr), (extloadi8_glue node:$ptr)>;
def sextloadi8_local_m0 : PatFrag<(ops node:$ptr), (sextloadi8_glue node:$ptr)>;
def zextloadi8_local_m0 : PatFrag<(ops node:$ptr), (zextloadi8_glue node:$ptr)>;
def extloadi16_local_m0 : PatFrag<(ops node:$ptr), (extloadi16_glue node:$ptr)>;
def sextloadi16_local_m0 : PatFrag<(ops node:$ptr), (sextloadi16_glue node:$ptr)>;
def zextloadi16_local_m0 : PatFrag<(ops node:$ptr), (zextloadi16_glue node:$ptr)>;
} // End IsLoad = 1, , AddressSpaces = LoadAddress_local.AddrSpaces
def load_align8_local_m0 : PatFrag<(ops node:$ptr),
(load_local_m0 node:$ptr)> {
let IsLoad = 1;
int MinAlignment = 8;
}
def load_align16_local_m0 : PatFrag<(ops node:$ptr),
(load_local_m0 node:$ptr)> {
let IsLoad = 1;
int MinAlignment = 16;
}
let IsAtomic = 1, AddressSpaces = LoadAddress_local.AddrSpaces in {
def atomic_load_8_local_m0 : PatFrag<(ops node:$ptr),
(atomic_load_8_glue node:$ptr)>;
def atomic_load_16_local_m0 : PatFrag<(ops node:$ptr),
(atomic_load_16_glue node:$ptr)>;
def atomic_load_32_local_m0 : PatFrag<(ops node:$ptr),
(atomic_load_32_glue node:$ptr)>;
def atomic_load_64_local_m0 : PatFrag<(ops node:$ptr),
(atomic_load_64_glue node:$ptr)>;
} // End let AddressSpaces = LoadAddress_local.AddrSpaces
def AMDGPUst_glue : SDNode <"ISD::STORE", SDTStore,
[SDNPHasChain, SDNPMayStore, SDNPMemOperand, SDNPInGlue]
>;
def AMDGPUatomic_st_glue : SDNode <"ISD::ATOMIC_STORE", SDTAtomicStore,
[SDNPHasChain, SDNPMayStore, SDNPMemOperand, SDNPInGlue]
>;
def unindexedstore_glue : PatFrag<(ops node:$val, node:$ptr),
(AMDGPUst_glue node:$val, node:$ptr)> {
let IsStore = 1;
let IsUnindexed = 1;
}
def store_glue : PatFrag<(ops node:$val, node:$ptr),
(unindexedstore_glue node:$val, node:$ptr)> {
let IsStore = 1;
let IsTruncStore = 0;
}
def truncstore_glue : PatFrag<(ops node:$val, node:$ptr),
(unindexedstore_glue node:$val, node:$ptr)> {
let IsStore = 1;
let IsTruncStore = 1;
}
def truncstorei8_glue : PatFrag<(ops node:$val, node:$ptr),
(truncstore_glue node:$val, node:$ptr)> {
let IsStore = 1;
let MemoryVT = i8;
let IsTruncStore = 1;
}
def truncstorei16_glue : PatFrag<(ops node:$val, node:$ptr),
(truncstore_glue node:$val, node:$ptr)> {
let IsStore = 1;
let MemoryVT = i16;
let IsTruncStore = 1;
}
let IsStore = 1, AddressSpaces = StoreAddress_local.AddrSpaces in {
def store_local_m0 : PatFrag<(ops node:$val, node:$ptr),
(store_glue node:$val, node:$ptr)>;
def truncstorei8_local_m0 : PatFrag<(ops node:$val, node:$ptr),
(truncstorei8_glue node:$val, node:$ptr)>;
def truncstorei16_local_m0 : PatFrag<(ops node:$val, node:$ptr),
(truncstorei16_glue node:$val, node:$ptr)>;
}
def store_align8_local_m0 : PatFrag <(ops node:$value, node:$ptr),
(store_local_m0 node:$value, node:$ptr)>,
Aligned<8> {
let IsStore = 1;
}
def store_align16_local_m0 : PatFrag <(ops node:$value, node:$ptr),
(store_local_m0 node:$value, node:$ptr)>,
Aligned<16> {
let IsStore = 1;
}
let PredicateCode = [{return cast<MemSDNode>(N)->getAlign() < 4;}],
GISelPredicateCode = [{return (*MI.memoperands_begin())->getAlign() < 4;}],
AddressSpaces = [ AddrSpaces.Local ] in {
def load_align_less_than_4_local : PatFrag<(ops node:$ptr),
(load_local node:$ptr)> {
let IsLoad = 1;
let IsNonExtLoad = 1;
}
def load_align_less_than_4_local_m0 : PatFrag<(ops node:$ptr),
(load_local_m0 node:$ptr)> {
let IsLoad = 1;
let IsNonExtLoad = 1;
}
def store_align_less_than_4_local : PatFrag <(ops node:$value, node:$ptr),
(store_local node:$value, node:$ptr)> {
let IsStore = 1;
let IsTruncStore = 0;
}
def store_align_less_than_4_local_m0 : PatFrag <(ops node:$value, node:$ptr),
(store_local_m0 node:$value, node:$ptr)> {
let IsStore = 1;
let IsTruncStore = 0;
}
}
def atomic_store_8_glue : PatFrag <
(ops node:$ptr, node:$value),
(AMDGPUatomic_st_glue node:$ptr, node:$value)> {
let IsAtomic = 1;
let MemoryVT = i8;
}
def atomic_store_16_glue : PatFrag <
(ops node:$ptr, node:$value),
(AMDGPUatomic_st_glue node:$ptr, node:$value)> {
let IsAtomic = 1;
let MemoryVT = i16;
}
def atomic_store_32_glue : PatFrag <
(ops node:$ptr, node:$value),
(AMDGPUatomic_st_glue node:$ptr, node:$value)> {
let IsAtomic = 1;
let MemoryVT = i32;
}
def atomic_store_64_glue : PatFrag <
(ops node:$ptr, node:$value),
(AMDGPUatomic_st_glue node:$ptr, node:$value)> {
let IsAtomic = 1;
let MemoryVT = i64;
}
let IsAtomic = 1, AddressSpaces = StoreAddress_local.AddrSpaces in {
def atomic_store_8_local_m0 : PatFrag<(ops node:$val, node:$ptr),
(atomic_store_8_glue node:$val, node:$ptr)>;
def atomic_store_16_local_m0 : PatFrag<(ops node:$val, node:$ptr),
(atomic_store_16_glue node:$val, node:$ptr)>;
def atomic_store_32_local_m0 : PatFrag<(ops node:$val, node:$ptr),
(atomic_store_32_glue node:$val, node:$ptr)>;
def atomic_store_64_local_m0 : PatFrag<(ops node:$val, node:$ptr),
(atomic_store_64_glue node:$val, node:$ptr)>;
} // End let IsAtomic = 1, AddressSpaces = StoreAddress_local.AddrSpaces
//===----------------------------------------------------------------------===//
// SDNodes PatFrags for a16 loads and stores with 3 components.
// v3f16/v3i16 is widened to v4f16/v4i16, so we need to match on the memory
// load/store size.
//===----------------------------------------------------------------------===//
class mubuf_intrinsic_load<SDPatternOperator name, ValueType vt> : PatFrag <
(ops node:$rsrc, node:$vindex, node:$voffset, node:$soffset, node:$offset,
node:$auxiliary, node:$idxen),
(name node:$rsrc, node:$vindex, node:$voffset, node:$soffset, node:$offset,
node:$auxiliary, node:$idxen)> {
let IsLoad = 1;
let MemoryVT = vt;
}
class mubuf_intrinsic_store<SDPatternOperator name, ValueType vt> : PatFrag <
(ops node:$vdata, node:$rsrc, node:$vindex, node:$voffset, node:$soffset, node:$offset,
node:$auxiliary, node:$idxen),
(name node:$vdata, node:$rsrc, node:$vindex, node:$voffset, node:$soffset, node:$offset,
node:$auxiliary, node:$idxen)> {
let IsStore = 1;
let MemoryVT = vt;
}
class mtbuf_intrinsic_load<SDPatternOperator name, ValueType vt> : PatFrag <
(ops node:$rsrc, node:$vindex, node:$voffset, node:$soffset, node:$offset,
node:$format, node:$auxiliary, node:$idxen),
(name node:$rsrc, node:$vindex, node:$voffset, node:$soffset, node:$offset,
node:$format, node:$auxiliary, node:$idxen)> {
let IsLoad = 1;
let MemoryVT = vt;
}
class mtbuf_intrinsic_store<SDPatternOperator name, ValueType vt> : PatFrag <
(ops node:$vdata, node:$rsrc, node:$vindex, node:$voffset, node:$soffset, node:$offset,
node:$format, node:$auxiliary, node:$idxen),
(name node:$vdata, node:$rsrc, node:$vindex, node:$voffset, node:$soffset, node:$offset,
node:$format, node:$auxiliary, node:$idxen)> {
let IsStore = 1;
let MemoryVT = vt;
}
//===----------------------------------------------------------------------===//
// SDNodes PatFrags for d16 loads
//===----------------------------------------------------------------------===//
class LoadD16Frag <SDPatternOperator op> : PatFrag<
(ops node:$ptr, node:$tied_in),
(op node:$ptr, node:$tied_in)> {
let IsLoad = 1;
}
foreach as = [ "global", "flat", "constant", "local", "private", "region" ] in {
let AddressSpaces = !cast<AddressSpaceList>("LoadAddress_"#as).AddrSpaces in {
def load_d16_hi_#as : LoadD16Frag <SIload_d16_hi>;
def az_extloadi8_d16_hi_#as : LoadD16Frag <SIload_d16_hi_u8> {
let MemoryVT = i8;
}
def sextloadi8_d16_hi_#as : LoadD16Frag <SIload_d16_hi_i8> {
let MemoryVT = i8;
}
def load_d16_lo_#as : LoadD16Frag <SIload_d16_lo>;
def az_extloadi8_d16_lo_#as : LoadD16Frag <SIload_d16_lo_u8> {
let MemoryVT = i8;
}
def sextloadi8_d16_lo_#as : LoadD16Frag <SIload_d16_lo_i8> {
let MemoryVT = i8;
}
} // End let AddressSpaces = ...
} // End foreach AddrSpace
def lshr_rev : PatFrag <
(ops node:$src1, node:$src0),
(srl $src0, $src1)
>;
def ashr_rev : PatFrag <
(ops node:$src1, node:$src0),
(sra $src0, $src1)
>;
def lshl_rev : PatFrag <
(ops node:$src1, node:$src0),
(shl $src0, $src1)
>;
def add_ctpop : PatFrag <
(ops node:$src0, node:$src1),
(add (ctpop $src0), $src1)
>;
def xnor : PatFrag <
(ops node:$src0, node:$src1),
(not (xor $src0, $src1))
>;
foreach I = 1-4 in {
def shl#I#_add : PatFrag <
(ops node:$src0, node:$src1),
(add (shl_oneuse $src0, (i32 I)), $src1)> {
// FIXME: Poor substitute for disabling pattern in SelectionDAG
let PredicateCode = [{return false;}];
let GISelPredicateCode = [{return true;}];
}
}
multiclass SIAtomicM0Glue2 <string op_name, bit is_amdgpu = 0,
SDTypeProfile tc = SDTAtomic2,
bit IsInt = 1> {
def _glue : SDNode <
!if(is_amdgpu, "AMDGPUISD", "ISD")#"::ATOMIC_"#op_name, tc,
[SDNPHasChain, SDNPMayStore, SDNPMayLoad, SDNPMemOperand, SDNPInGlue]
>;
let AddressSpaces = StoreAddress_local.AddrSpaces in {
defm _local_m0 : binary_atomic_op <!cast<SDNode>(NAME#"_glue"), IsInt>;
defm _local_m0 : noret_binary_atomic_op <!cast<SDNode>(NAME#"_glue"),
IsInt>;
}
let AddressSpaces = StoreAddress_region.AddrSpaces in {
defm _region_m0 : binary_atomic_op <!cast<SDNode>(NAME#"_glue"), IsInt>;
defm _region_m0 : noret_binary_atomic_op <!cast<SDNode>(NAME#"_glue"),
IsInt>;
}
}
defm atomic_load_add : SIAtomicM0Glue2 <"LOAD_ADD">;
defm atomic_load_sub : SIAtomicM0Glue2 <"LOAD_SUB">;
defm atomic_load_uinc_wrap : SIAtomicM0Glue2 <"LOAD_UINC_WRAP">;
defm atomic_load_udec_wrap : SIAtomicM0Glue2 <"LOAD_UDEC_WRAP">;
defm atomic_load_and : SIAtomicM0Glue2 <"LOAD_AND">;
defm atomic_load_min : SIAtomicM0Glue2 <"LOAD_MIN">;
defm atomic_load_max : SIAtomicM0Glue2 <"LOAD_MAX">;
defm atomic_load_or : SIAtomicM0Glue2 <"LOAD_OR">;
defm atomic_load_xor : SIAtomicM0Glue2 <"LOAD_XOR">;
defm atomic_load_umin : SIAtomicM0Glue2 <"LOAD_UMIN">;
defm atomic_load_umax : SIAtomicM0Glue2 <"LOAD_UMAX">;
defm atomic_swap : SIAtomicM0Glue2 <"SWAP">;
defm atomic_load_fadd : SIAtomicM0Glue2 <"LOAD_FADD", 0, SDTAtomic2_f32, 0>;
defm atomic_load_fmin : SIAtomicM0Glue2 <"LOAD_FMIN", 1, SDTAtomic2_f32, 0>;
defm atomic_load_fmax : SIAtomicM0Glue2 <"LOAD_FMAX", 1, SDTAtomic2_f32, 0>;
def as_i1timm : SDNodeXForm<timm, [{
return CurDAG->getTargetConstant(N->getZExtValue(), SDLoc(N), MVT::i1);
}]>;
def as_i8imm : SDNodeXForm<imm, [{
return CurDAG->getTargetConstant(N->getZExtValue(), SDLoc(N), MVT::i8);
}]>;
def as_i8timm : SDNodeXForm<timm, [{
return CurDAG->getTargetConstant(N->getSExtValue(), SDLoc(N), MVT::i16);
}]>;
def as_i16imm : SDNodeXForm<imm, [{
return CurDAG->getTargetConstant(N->getSExtValue(), SDLoc(N), MVT::i16);
}]>;
def as_i16timm : SDNodeXForm<timm, [{
return CurDAG->getTargetConstant(N->getSExtValue(), SDLoc(N), MVT::i16);
}]>;
def as_i32imm: SDNodeXForm<imm, [{
return CurDAG->getTargetConstant(N->getSExtValue(), SDLoc(N), MVT::i32);
}]>;
def as_i32timm: SDNodeXForm<timm, [{
return CurDAG->getTargetConstant(N->getSExtValue(), SDLoc(N), MVT::i32);
}]>;
def as_i64imm: SDNodeXForm<imm, [{
return CurDAG->getTargetConstant(N->getSExtValue(), SDLoc(N), MVT::i64);
}]>;
def cond_as_i32imm: SDNodeXForm<cond, [{
return CurDAG->getTargetConstant(N->get(), SDLoc(N), MVT::i32);
}]>;
// Copied from the AArch64 backend:
def bitcast_fpimm_to_i32 : SDNodeXForm<fpimm, [{
return CurDAG->getTargetConstant(
N->getValueAPF().bitcastToAPInt().getZExtValue(), SDLoc(N), MVT::i32);
}]>;
def frameindex_to_targetframeindex : SDNodeXForm<frameindex, [{
auto FI = cast<FrameIndexSDNode>(N);
return CurDAG->getTargetFrameIndex(FI->getIndex(), MVT::i32);
}]>;
// Copied from the AArch64 backend:
def bitcast_fpimm_to_i64 : SDNodeXForm<fpimm, [{
return CurDAG->getTargetConstant(
N->getValueAPF().bitcastToAPInt().getZExtValue(), SDLoc(N), MVT::i64);
}]>;
class bitextract_imm<int bitnum> : SDNodeXForm<imm, [{
uint64_t Imm = N->getZExtValue();
unsigned Bit = (Imm >> }] # bitnum # [{ ) & 1;
return CurDAG->getTargetConstant(Bit, SDLoc(N), MVT::i1);
}]>;
def SIMM16bit : TImmLeaf <i32,
[{return isInt<16>(Imm) || isUInt<16>(Imm);}],
as_i16timm
>;
def i64imm_32bit : ImmLeaf<i64, [{
return (Imm & 0xffffffffULL) == static_cast<uint64_t>(Imm);
}]>;
def InlineImm64 : IntImmLeaf<i64, [{
return isInlineImmediate(Imm);
}]>;
def InlineImmFP32 : FPImmLeaf<f32, [{
return isInlineImmediate(Imm);
}]>;
def InlineImmFP64 : FPImmLeaf<f64, [{
return isInlineImmediate(Imm);
}]>;
class VGPRImm <dag frag> : PatLeaf<frag, [{
return isVGPRImm(N);
}]>;
def NegateImm : SDNodeXForm<imm, [{
return CurDAG->getConstant(-N->getSExtValue(), SDLoc(N), MVT::i32);
}]>;
// TODO: When FP inline imm values work?
def NegSubInlineConst32 : ImmLeaf<i32, [{
return Imm < -16 && Imm >= -64;
}], NegateImm>;
def NegSubInlineIntConst16 : ImmLeaf<i16, [{
return Imm < -16 && Imm >= -64;
}], NegateImm>;
def ShiftAmt32Imm : ImmLeaf <i32, [{
return Imm < 32;
}]>;
def fp16_zeros_high_16bits : PatLeaf<(f16 VGPR_32:$src), [{
return fp16SrcZerosHighBits(N->getOpcode());
}]>;
//===----------------------------------------------------------------------===//
// MUBUF/SMEM Patterns
//===----------------------------------------------------------------------===//
def extract_cpol : SDNodeXForm<timm, [{
return CurDAG->getTargetConstant(
N->getZExtValue() & (Subtarget->getGeneration() >= AMDGPUSubtarget::GFX12
? AMDGPU::CPol::ALL
: AMDGPU::CPol::ALL_pregfx12),
SDLoc(N), MVT::i8);
}]>;
def extract_swz : SDNodeXForm<timm, [{
const bool Swizzle =
N->getZExtValue() & (Subtarget->getGeneration() >= AMDGPUSubtarget::GFX12
? AMDGPU::CPol::SWZ
: AMDGPU::CPol::SWZ_pregfx12);
return CurDAG->getTargetConstant(Swizzle, SDLoc(N), MVT::i8);
}]>;
def extract_cpol_set_glc : SDNodeXForm<timm, [{
const uint32_t cpol = N->getZExtValue() & (Subtarget->getGeneration() >= AMDGPUSubtarget::GFX12
? AMDGPU::CPol::ALL
: AMDGPU::CPol::ALL_pregfx12);
return CurDAG->getTargetConstant(cpol | AMDGPU::CPol::GLC, SDLoc(N), MVT::i8);
}]>;
//===----------------------------------------------------------------------===//
// Custom Operands
//===----------------------------------------------------------------------===//
def SOPPBrTarget : CustomOperand<OtherVT> {
let PrintMethod = "printOperand";
let EncoderMethod = "getSOPPBrEncoding";
let DecoderMethod = "decodeSOPPBrTarget";
let OperandType = "OPERAND_PCREL";
}
def si_ga : Operand<iPTR>;
def InterpSlot : CustomOperand<i32>;
// It appears to be necessary to create a separate operand for this to
// be able to parse attr<num> with no space.
def InterpAttr : CustomOperand<i32>;
def InterpAttrChan : ImmOperand<i32>;
def SplitBarrier : ImmOperand<i32> {
let OperandNamespace = "AMDGPU";
let OperandType = "OPERAND_INLINE_SPLIT_BARRIER_INT32";
let DecoderMethod = "decodeSplitBarrier";
let PrintMethod = "printOperand";
}
def VReg32OrOffClass : AsmOperandClass {
let Name = "VReg32OrOff";
let ParserMethod = "parseVReg32OrOff";
}
def SendMsg : CustomOperand<i32>;
def Swizzle : CustomOperand<i16, 1>;
def Endpgm : CustomOperand<i16, 1>;
def SWaitCnt : CustomOperand<i32>;
def DepCtr : CustomOperand<i32>;
def SDelayALU : CustomOperand<i32>;
include "SIInstrFormats.td"
include "VIInstrFormats.td"
def BoolReg : AsmOperandClass {
let Name = "BoolReg";
let ParserMethod = "parseBoolReg";
let RenderMethod = "addRegOperands";
}
class BoolRC : RegisterOperand<SReg_1> {
let ParserMatchClass = BoolReg;
let DecoderMethod = "decodeBoolReg";
}
def SSrc_i1 : RegisterOperand<SReg_1_XEXEC> {
let ParserMatchClass = BoolReg;
let DecoderMethod = "decodeBoolReg";
}
def VOPDstS64orS32 : BoolRC {
let PrintMethod = "printVOPDst";
}
// SCSrc_i1 is the operand for pseudo instructions only.
// Boolean immediates shall not be exposed to codegen instructions.
def SCSrc_i1 : RegisterOperand<SReg_1_XEXEC> {
let OperandNamespace = "AMDGPU";
let OperandType = "OPERAND_REG_IMM_INT32";
let ParserMatchClass = BoolReg;
let DecoderMethod = "decodeBoolReg";
}
// ===----------------------------------------------------------------------===//
// ExpSrc* Special cases for exp src operands which are printed as
// "off" depending on en operand.
// ===----------------------------------------------------------------------===//
def ExpSrc0 : RegisterOperand<VGPR_32> {
let PrintMethod = "printExpSrc0";
let ParserMatchClass = VReg32OrOffClass;
}
def ExpSrc1 : RegisterOperand<VGPR_32> {
let PrintMethod = "printExpSrc1";
let ParserMatchClass = VReg32OrOffClass;
}
def ExpSrc2 : RegisterOperand<VGPR_32> {
let PrintMethod = "printExpSrc2";
let ParserMatchClass = VReg32OrOffClass;
}
def ExpSrc3 : RegisterOperand<VGPR_32> {
let PrintMethod = "printExpSrc3";
let ParserMatchClass = VReg32OrOffClass;
}
class SDWASrc<ValueType vt> : RegisterOperand<VS_32> {
let OperandNamespace = "AMDGPU";
string Type = !if(vt.isFP, "FP", "INT");
let OperandType = "OPERAND_REG_INLINE_C_"#Type#vt.Size;
let DecoderMethod = "decodeSDWASrc"#vt.Size;
let EncoderMethod = "getSDWASrcEncoding";
}
def SDWASrc_i32 : SDWASrc<i32>;
def SDWASrc_i16 : SDWASrc<i16>;
def SDWASrc_f32 : SDWASrc<f32>;
def SDWASrc_f16 : SDWASrc<f16>;
def SDWAVopcDst : BoolRC {
let OperandNamespace = "AMDGPU";
let OperandType = "OPERAND_SDWA_VOPC_DST";
let EncoderMethod = "getSDWAVopcDstEncoding";
let DecoderMethod = "decodeSDWAVopcDst";
let PrintMethod = "printVOPDst";
}
class NamedIntOperand<ValueType Type, string Prefix, bit Optional = 1,
string ConvertMethod = "nullptr">
: CustomOperand<Type, Optional, NAME> {
let ParserMethod =
"[this](OperandVector &Operands) -> ParseStatus { "#
"return parseIntWithPrefix(\""#Prefix#"\", Operands, "#
"AMDGPUOperand::"#ImmTy#", "#ConvertMethod#"); }";
}
class NamedBitOperand<string Id, string Name = NAME>
: CustomOperand<i1, 1, Name> {
let PredicateMethod = "isImmTy<AMDGPUOperand::"#ImmTy#">";
let ParserMethod =
"[this](OperandVector &Operands) -> ParseStatus { "#
"return parseNamedBit(\""#Id#"\", Operands, AMDGPUOperand::"#ImmTy#"); }";
let PrintMethod = "[this](const MCInst *MI, unsigned OpNo, "#
"const MCSubtargetInfo &STI, raw_ostream &O) { "#
"printNamedBit(MI, OpNo, O, \""#Id#"\"); }";
}
class DefaultOperand<CustomOperand Op, int Value>
: OperandWithDefaultOps<Op.Type, (ops (Op.Type Value))>,
CustomOperandProps<1> {
let ParserMatchClass = Op.ParserMatchClass;
let PrintMethod = Op.PrintMethod;
}
class SDWAOperand<string Id, string Name = NAME>
: CustomOperand<i32, 1, Name> {
let ParserMethod =
"[this](OperandVector &Operands) -> ParseStatus { "#
"return parseSDWASel(Operands, \""#Id#"\", AMDGPUOperand::"#ImmTy#"); }";
}
class ArrayOperand0<string Id, string Name = NAME>
: OperandWithDefaultOps<i32, (ops (i32 0))>,
CustomOperandProps<1, Name> {
let ParserMethod =
"[this](OperandVector &Operands) -> ParseStatus { "#
"return parseOperandArrayWithPrefix(\""#Id#"\", Operands, "#
"AMDGPUOperand::"#ImmTy#"); }";
}
let ImmTy = "ImmTyOffset" in
def flat_offset : CustomOperand<i32, 1, "FlatOffset">;
def Offset : NamedIntOperand<i32, "offset">;
def Offset0 : NamedIntOperand<i8, "offset0">;
def Offset1 : NamedIntOperand<i8, "offset1">;
def gds : NamedBitOperand<"gds", "GDS">;
def omod : CustomOperand<i32, 1, "OModSI">;
def omod0 : DefaultOperand<omod, 0>;
// We need to make the cases with a default of 0 distinct from no
// default to help deal with some cases where the operand appears
// before a mandatory operand.
def clampmod : NamedBitOperand<"clamp", "ClampSI">;
def clampmod0 : DefaultOperand<clampmod, 0>;
def highmod : NamedBitOperand<"high", "High">;
def CPol : CustomOperand<i32, 1>;
def CPol_0 : DefaultOperand<CPol, 0>;
def CPol_GLC1 : DefaultOperand<CPol, 1>;
def CPol_GLC : ValuePredicatedOperand<CPol, "Op.getImm() & CPol::GLC">;
def CPol_NonGLC : ValuePredicatedOperand<CPol, "!(Op.getImm() & CPol::GLC)", 1>;
def CPol_GLC_WithDefault : DefaultOperand<CPol_GLC, !shl(1, CPolBit.GLC)>;
def CPol_NonGLC_WithDefault : DefaultOperand<CPol_NonGLC, 0>;
def TFE : NamedBitOperand<"tfe">;
def UNorm : NamedBitOperand<"unorm">;
def DA : NamedBitOperand<"da">;
def R128A16 : CustomOperand<i1, 1>;
def A16 : NamedBitOperand<"a16">;
def D16 : NamedBitOperand<"d16">;
def LWE : NamedBitOperand<"lwe">;
def exp_compr : NamedBitOperand<"compr", "ExpCompr">;
def exp_vm : NamedBitOperand<"vm", "ExpVM">;
def FORMAT : CustomOperand<i8>;
def DMask : NamedIntOperand<i16, "dmask">;
def Dim : CustomOperand<i8>;
def dst_sel : SDWAOperand<"dst_sel", "SDWADstSel">;
def src0_sel : SDWAOperand<"src0_sel", "SDWASrc0Sel">;
def src1_sel : SDWAOperand<"src1_sel", "SDWASrc1Sel">;
def dst_unused : CustomOperand<i32, 1, "SDWADstUnused">;
def op_sel0 : ArrayOperand0<"op_sel", "OpSel">;
def op_sel_hi0 : ArrayOperand0<"op_sel_hi", "OpSelHi">;
def neg_lo0 : ArrayOperand0<"neg_lo", "NegLo">;
def neg_hi0 : ArrayOperand0<"neg_hi", "NegHi">;
def IndexKey16bit : CustomOperand<i32, 1>;
def IndexKey8bit : CustomOperand<i32, 1>;
def dpp8 : CustomOperand<i32, 0, "DPP8">;
def dpp_ctrl : CustomOperand<i32, 0, "DPPCtrl">;
let DefaultValue = "0xf" in {
def DppRowMask : NamedIntOperand<i32, "row_mask">;
def DppBankMask : NamedIntOperand<i32, "bank_mask">;
}
def DppBoundCtrl : NamedIntOperand<i1, "bound_ctrl", 1,
"[this] (int64_t &BC) -> bool { return convertDppBoundCtrl(BC); }">;
def DppFI : NamedIntOperand<i32, "fi">;
def blgp : CustomOperand<i32, 1, "BLGP">;
def CBSZ : NamedIntOperand<i32, "cbsz">;
def ABID : NamedIntOperand<i32, "abid">;
def hwreg : CustomOperand<i32, 0, "Hwreg">;
def exp_tgt : CustomOperand<i32, 0, "ExpTgt">;
def WaitVDST : NamedIntOperand<i8, "wait_vdst">;
def WaitEXP : NamedIntOperand<i8, "wait_exp">;
def WaitVAVDst : NamedIntOperand<i8, "wait_va_vdst">;
def WaitVMVSrc : NamedIntOperand<i8, "wait_vm_vsrc">;
class KImmFPOperand<ValueType vt> : ImmOperand<vt> {
let OperandNamespace = "AMDGPU";
let OperandType = "OPERAND_KIMM"#vt.Size;
let PrintMethod = "printU"#vt.Size#"ImmOperand";
let DecoderMethod = "decodeOperand_KImmFP";
}
// 32-bit VALU immediate operand that uses the constant bus.
def KImmFP32 : KImmFPOperand<i32>;
// 32-bit VALU immediate operand with a 16-bit value that uses the
// constant bus.
def KImmFP16 : KImmFPOperand<i16>;
class FPInputModsMatchClass <int opSize> : AsmOperandClass {
let Name = "RegOrImmWithFP"#opSize#"InputMods";
let ParserMethod = "parseRegOrImmWithFPInputMods";
let PredicateMethod = "isRegOrImmWithFP"#opSize#"InputMods";
}
class FPVCSrcInputModsMatchClass <int opSize> : FPInputModsMatchClass <opSize> {
let Name = "RegOrInlineImmWithFP"#opSize#"InputMods";
let PredicateMethod = "isRegOrInlineImmWithFP"#opSize#"InputMods";
}
def FP16InputModsMatchClass : FPInputModsMatchClass<16>;
def FPT16InputModsMatchClass : FPInputModsMatchClass<16> {
let Name = "RegOrImmWithFPT16InputMods";
let PredicateMethod = "isRegOrImmWithFPT16InputMods";
}
def FP32InputModsMatchClass : FPInputModsMatchClass<32>;
def FP64InputModsMatchClass : FPInputModsMatchClass<64>;
class FP16VCSrcInputModsMatchClass<bit IsFake16>
: FPVCSrcInputModsMatchClass<16> {
let Name = !if(IsFake16, "RegOrInlineImmWithFPFake16InputMods",
"RegOrInlineImmWithFPT16InputMods");
let PredicateMethod = "isRegOrInlineImmWithFP16InputMods<" #
!if(IsFake16, "true", "false") # ">";
}
def FP32VCSrcInputModsMatchClass : FPVCSrcInputModsMatchClass<32>;
class InputMods <AsmOperandClass matchClass> : Operand <i32> {
let OperandNamespace = "AMDGPU";
let OperandType = "OPERAND_INPUT_MODS";
let ParserMatchClass = matchClass;
}
class FPInputMods <FPInputModsMatchClass matchClass> : InputMods <matchClass> {
let PrintMethod = "printOperandAndFPInputMods";
}
def FP16InputMods : FPInputMods<FP16InputModsMatchClass>;
def FPT16InputMods : FPInputMods<FPT16InputModsMatchClass>;
def FP32InputMods : FPInputMods<FP32InputModsMatchClass>;
def FP64InputMods : FPInputMods<FP64InputModsMatchClass>;
class FP16VCSrcInputMods<bit IsFake16>
: FPInputMods<FP16VCSrcInputModsMatchClass<IsFake16>>;
def FP32VCSrcInputMods : FPInputMods<FP32VCSrcInputModsMatchClass>;
class IntInputModsMatchClass <int opSize> : AsmOperandClass {
let Name = "RegOrImmWithInt"#opSize#"InputMods";
let ParserMethod = "parseRegOrImmWithIntInputMods";
let PredicateMethod = "isRegOrImmWithInt"#opSize#"InputMods";
}
class IntVCSrcInputModsMatchClass <int opSize> : IntInputModsMatchClass <opSize> {
let Name = "RegOrInlineImmWithInt"#opSize#"InputMods";
let PredicateMethod = "isRegOrInlineImmWithInt"#opSize#"InputMods";
}
def IntT16InputModsMatchClass : IntInputModsMatchClass<16> {
let Name = "RegOrImmWithIntT16InputMods";
let PredicateMethod = "isRegOrImmWithIntT16InputMods";
}
def Int32InputModsMatchClass : IntInputModsMatchClass<32>;
def Int64InputModsMatchClass : IntInputModsMatchClass<64>;
def Int32VCSrcInputModsMatchClass : IntVCSrcInputModsMatchClass<32>;
class IntInputMods <IntInputModsMatchClass matchClass> : InputMods <matchClass> {
let PrintMethod = "printOperandAndIntInputMods";
}
def IntT16InputMods : IntInputMods<IntT16InputModsMatchClass>;
def Int32InputMods : IntInputMods<Int32InputModsMatchClass>;
def Int64InputMods : IntInputMods<Int64InputModsMatchClass>;
def Int32VCSrcInputMods : IntInputMods<Int32VCSrcInputModsMatchClass>;
class OpSelModsMatchClass : AsmOperandClass {
let Name = "OpSelMods";
let ParserMethod = "parseRegOrImm";
let PredicateMethod = "isRegOrImm";
}
def IntOpSelModsMatchClass : OpSelModsMatchClass;
def IntOpSelMods : InputMods<IntOpSelModsMatchClass>;
class FPSDWAInputModsMatchClass <int opSize> : AsmOperandClass {
let Name = "SDWAWithFP"#opSize#"InputMods";
let ParserMethod = "parseRegOrImmWithFPInputMods";
let PredicateMethod = "isSDWAFP"#opSize#"Operand";
}
def FP16SDWAInputModsMatchClass : FPSDWAInputModsMatchClass<16>;
def FP32SDWAInputModsMatchClass : FPSDWAInputModsMatchClass<32>;
class FPSDWAInputMods <FPSDWAInputModsMatchClass matchClass> :
InputMods <matchClass> {
let PrintMethod = "printOperandAndFPInputMods";
}
def FP16SDWAInputMods : FPSDWAInputMods<FP16SDWAInputModsMatchClass>;
def FP32SDWAInputMods : FPSDWAInputMods<FP32SDWAInputModsMatchClass>;
def FPVRegInputModsMatchClass : AsmOperandClass {
let Name = "VRegWithFPInputMods";
let ParserMethod = "parseRegWithFPInputMods";
let PredicateMethod = "isVRegWithInputMods";
}
class FPT16VRegInputModsMatchClass<bit IsFake16> : AsmOperandClass {
let Name = !if(IsFake16, "Fake16VRegWithFPInputMods",
"T16VRegWithFPInputMods");
let ParserMethod = "parseRegWithFPInputMods";
let PredicateMethod = "isT16VRegWithInputMods<" #
!if(IsFake16, "true", "false") # ">";
}
def FPVRegInputMods : InputMods <FPVRegInputModsMatchClass> {
let PrintMethod = "printOperandAndFPInputMods";
}
class FPT16VRegInputMods<bit IsFake16>
: InputMods <FPT16VRegInputModsMatchClass<IsFake16>> {
let PrintMethod = "printOperandAndFPInputMods";
}
class IntSDWAInputModsMatchClass <int opSize> : AsmOperandClass {
let Name = "SDWAWithInt"#opSize#"InputMods";
let ParserMethod = "parseRegOrImmWithIntInputMods";
let PredicateMethod = "isSDWAInt"#opSize#"Operand";
}
def Int16SDWAInputModsMatchClass : IntSDWAInputModsMatchClass<16>;
def Int32SDWAInputModsMatchClass : IntSDWAInputModsMatchClass<32>;
def Bin32SDWAInputModsMatchClass : IntSDWAInputModsMatchClass<32> {
let Name = "SDWAWithBin32InputMods";
let ParserMethod = "parseRegOrImm";
}
class IntSDWAInputMods <IntSDWAInputModsMatchClass matchClass> :
InputMods <matchClass> {
let PrintMethod = "printOperandAndIntInputMods";
}
def Int16SDWAInputMods : IntSDWAInputMods<Int16SDWAInputModsMatchClass>;
def Int32SDWAInputMods : IntSDWAInputMods<Int32SDWAInputModsMatchClass>;
def Bin32SDWAInputMods : IntSDWAInputMods<Bin32SDWAInputModsMatchClass>;
def IntVRegInputModsMatchClass : AsmOperandClass {
let Name = "VRegWithIntInputMods";
let ParserMethod = "parseRegWithIntInputMods";
let PredicateMethod = "isVRegWithInputMods";
}
class IntT16VRegInputModsMatchClass<bit IsFake16> : AsmOperandClass {
let Name = !if(IsFake16, "Fake16VRegWithIntInputMods",
"T16VRegWithIntInputMods");
let ParserMethod = "parseRegWithIntInputMods";
let PredicateMethod = "isT16VRegWithInputMods<" #
!if(IsFake16, "true", "false") # ">";
}
class IntT16VRegInputMods<bit IsFake16>
: InputMods <IntT16VRegInputModsMatchClass<IsFake16>> {
let PrintMethod = "printOperandAndIntInputMods";
}
def IntVRegInputMods : InputMods <IntVRegInputModsMatchClass> {
let PrintMethod = "printOperandAndIntInputMods";
}
class PackedFPInputModsMatchClass <int opSize> : AsmOperandClass {
let Name = "PackedFP"#opSize#"InputMods";
let ParserMethod = "parseRegOrImmWithFPInputMods";
let PredicateMethod = "isPackedFP"#opSize#"InputMods";
}
class PackedIntInputModsMatchClass <int opSize> : AsmOperandClass {
let Name = "PackedInt"#opSize#"InputMods";
let ParserMethod = "parseRegOrImm";
let PredicateMethod = "isRegOrImm";
// let PredicateMethod = "isPackedInt"#opSize#"InputMods";
}
def PackedF16InputModsMatchClass : PackedFPInputModsMatchClass<16>;
def PackedI16InputModsMatchClass : PackedIntInputModsMatchClass<16>;
class PackedFPInputMods <PackedFPInputModsMatchClass matchClass> : InputMods <matchClass> {
let PrintMethod = "printOperandAndFPInputMods";
}
class PackedIntInputMods <PackedIntInputModsMatchClass matchClass> : InputMods <matchClass> {
//let PrintMethod = "printPackedIntInputMods";
}
def PackedF16InputMods : PackedFPInputMods<PackedF16InputModsMatchClass>;
def PackedI16InputMods : PackedIntInputMods<PackedI16InputModsMatchClass>;
//===----------------------------------------------------------------------===//
// Complex patterns
//===----------------------------------------------------------------------===//
def DS1Addr1Offset : ComplexPattern<iPTR, 2, "SelectDS1Addr1Offset">;
def DS64Bit4ByteAligned : ComplexPattern<iPTR, 3, "SelectDS64Bit4ByteAligned">;
def DS128Bit8ByteAligned : ComplexPattern<iPTR, 3, "SelectDS128Bit8ByteAligned">;
def MOVRELOffset : ComplexPattern<iPTR, 2, "SelectMOVRELOffset">;
def VOP3Mods0 : ComplexPattern<untyped, 4, "SelectVOP3Mods0">;
// Modifiers for floating point instructions.
def VOP3Mods : ComplexPattern<untyped, 2, "SelectVOP3Mods">;
// VOP3 modifiers used for instructions that do not read canonicalized
// floating point values (i.e. integer operations with FP source
// modifiers)
def VOP3ModsNonCanonicalizing : ComplexPattern<untyped, 2,
"SelectVOP3ModsNonCanonicalizing">;
def VOP3NoMods : ComplexPattern<untyped, 1, "SelectVOP3NoMods">;
def VOP3OMods : ComplexPattern<untyped, 3, "SelectVOP3OMods">;
def VOP3PMods : ComplexPattern<untyped, 2, "SelectVOP3PMods">;
def VOP3PModsDOT : ComplexPattern<untyped, 2, "SelectVOP3PModsDOT">;
def VOP3PModsNeg : ComplexPattern<untyped, 1, "SelectVOP3PModsNeg">;
def WMMAOpSelVOP3PMods : ComplexPattern<untyped, 1, "SelectWMMAOpSelVOP3PMods">;
def WMMAModsF32NegAbs : ComplexPattern<untyped, 2, "SelectWMMAModsF32NegAbs">;
def WMMAModsF16Neg : ComplexPattern<untyped, 2, "SelectWMMAModsF16Neg">;
def WMMAModsF16NegAbs : ComplexPattern<untyped, 2, "SelectWMMAModsF16NegAbs">;
def WMMAVISrc : ComplexPattern<untyped, 1, "SelectWMMAVISrc">;
def SWMMACIndex8 : ComplexPattern<untyped, 2, "SelectSWMMACIndex8">;
def SWMMACIndex16 : ComplexPattern<untyped, 2, "SelectSWMMACIndex16">;
def VOP3OpSel : ComplexPattern<untyped, 2, "SelectVOP3OpSel">;
def VOP3OpSelMods : ComplexPattern<untyped, 2, "SelectVOP3OpSelMods">;
def VOP3PMadMixModsExt : ComplexPattern<untyped, 2, "SelectVOP3PMadMixModsExt">;
def VOP3PMadMixMods : ComplexPattern<untyped, 2, "SelectVOP3PMadMixMods">;
def VINTERPMods : ComplexPattern<untyped, 2, "SelectVINTERPMods">;
def VINTERPModsHi : ComplexPattern<untyped, 2, "SelectVINTERPModsHi">;
//===----------------------------------------------------------------------===//
// SI assembler operands
//===----------------------------------------------------------------------===//
def SIOperand {
int ZERO = 0x80;
int VCC = 0x6A;
int FLAT_SCR = 0x68;
}
// This should be kept in sync with SISrcMods enum
def SRCMODS {
int NONE = 0;
int NEG = 1;
int ABS = 2;
int NEG_ABS = 3;
int NEG_HI = ABS;
int OP_SEL_0 = 4;
int OP_SEL_1 = 8;
int DST_OP_SEL = 8;
}
def DSTCLAMP {
int NONE = 0;
int ENABLE = 1;
}
def DSTOMOD {
int NONE = 0;
}
def HWREG {
int MODE = 1;
int STATUS = 2;
int TRAPSTS = 3;
int HW_ID = 4;
int GPR_ALLOC = 5;
int LDS_ALLOC = 6;
int IB_STS = 7;
int MEM_BASES = 15;
int TBA_LO = 16;
int TBA_HI = 17;
int TMA_LO = 18;
int TMA_HI = 19;
int FLAT_SCR_LO = 20;
int FLAT_SCR_HI = 21;
int XNACK_MASK = 22;
int POPS_PACKER = 25;
int SHADER_CYCLES = 29;
}
class getHwRegImm<int Reg, int Offset = 0, int Size = 32> {
int ret = !and(!or(Reg,
!shl(Offset, 6),
!shl(!add(Size, -1), 11)), 65535);
}
//===----------------------------------------------------------------------===//
//
// SI Instruction multiclass helpers.
//
// Instructions with _32 take 32-bit operands.
// Instructions with _64 take 64-bit operands.
//
// VOP_* instructions can use either a 32-bit or 64-bit encoding. The 32-bit
// encoding is the standard encoding, but instruction that make use of
// any of the instruction modifiers must use the 64-bit encoding.
//
// Instructions with _e32 use the 32-bit encoding.
// Instructions with _e64 use the 64-bit encoding.
//
//===----------------------------------------------------------------------===//
class SIMCInstr <string pseudo, int subtarget> {
string PseudoInstr = pseudo;
int Subtarget = subtarget;
}
//===----------------------------------------------------------------------===//
// Vector ALU classes
//===----------------------------------------------------------------------===//
class getNumSrcArgs<ValueType Src0, ValueType Src1, ValueType Src2> {
int ret =
!if (!eq(Src0.Value, untyped.Value), 0,
!if (!eq(Src1.Value, untyped.Value), 1, // VOP1
!if (!eq(Src2.Value, untyped.Value), 2, // VOP2
3))); // VOP3
}
// Returns the register class to use for the destination of VOP[123C]
// instructions for the given VT.
class getVALUDstForVT<ValueType VT, bit IsTrue16 = 0, bit IsVOP3Encoding = 0> {
defvar op16 = !if(IsTrue16, !if (IsVOP3Encoding, VOPDstOperand_t16,
VOPDstOperand_t16Lo128),
VOPDstOperand<VGPR_32>);
RegisterOperand ret = !cond(!eq(VT.Size, 256) : VOPDstOperand<VReg_256>,
!eq(VT.Size, 128) : VOPDstOperand<VReg_128>,
!eq(VT.Size, 64) : VOPDstOperand<VReg_64>,
!eq(VT.Size, 32) : VOPDstOperand<VGPR_32>,
!eq(VT.Size, 16) : op16,
1 : VOPDstS64orS32); // else VT == i1
}
class getVALUDstForVT_fake16<ValueType VT> {
RegisterOperand ret = !if(!eq(VT.Size, 32), VOPDstOperand<VGPR_32>,
!if(!eq(VT.Size, 128), VOPDstOperand<VReg_128>,
!if(!eq(VT.Size, 64), VOPDstOperand<VReg_64>,
!if(!eq(VT.Size, 16), VOPDstOperand<VGPR_32_Lo128>,
VOPDstS64orS32)))); // else VT == i1
}
// Returns the register class to use for the destination of VOP[12C]
// instructions with SDWA extension
class getSDWADstForVT<ValueType VT> {
RegisterOperand ret = !if(!eq(VT.Size, 1),
SDWAVopcDst, // VOPC
VOPDstOperand<VGPR_32>); // VOP1/2 32-bit dst
}
// Returns the register class to use for source 0 of VOP[12C]
// instructions for the given VT.
class getVOPSrc0ForVT<ValueType VT, bit IsTrue16, bit IsFake16 = 1> {
RegisterOperand ret =
!cond(!eq(VT, i64) : VSrc_b64,
!eq(VT, f64) : VSrc_f64,
!eq(VT, i32) : VSrc_b32,
!eq(VT, f32) : VSrc_f32,
!eq(VT, i16) : !if(IsTrue16,
!if(IsFake16, VSrcFake16_b16_Lo128, VSrcT_b16_Lo128),
VSrc_b16),
!eq(VT, f16) : !if(IsTrue16,
!if(IsFake16, VSrcFake16_f16_Lo128, VSrcT_f16_Lo128),
VSrc_f16),
!eq(VT, bf16) : !if(IsTrue16,
!if(IsFake16, VSrcFake16_bf16_Lo128, VSrcT_bf16_Lo128),
VSrc_bf16),
!eq(VT, v2i16) : VSrc_v2b16,
!eq(VT, v2f16) : VSrc_v2f16,
!eq(VT, v2bf16) : VSrc_v2bf16,
!eq(VT, v4f16) : AVSrc_64,
!eq(VT, v4bf16) : AVSrc_64,
1 : VSrc_b32);
}
class getSOPSrcForVT<ValueType VT> {
RegisterOperand ret = !if(!eq(VT.Size, 64), SSrc_b64, SSrc_b32);
}
// Returns the vreg register class to use for source operand given VT
class getVregSrcForVT<ValueType VT, bit IsTrue16 = 0, bit IsFake16 = 0> {
RegisterOperand ret =
!cond(!eq(VT.Size, 128) : RegisterOperand<VReg_128>,
!eq(VT.Size, 96) : RegisterOperand<VReg_96>,
!eq(VT.Size, 64) : RegisterOperand<VReg_64>,
!eq(VT.Size, 48) : RegisterOperand<VReg_64>,
!eq(VT.Size, 16) : !if(IsTrue16,
!if(IsFake16, VGPRSrc_32_Lo128, VGPRSrc_16_Lo128),
RegisterOperand<VGPR_32>),
1 : RegisterOperand<VGPR_32>);
}
class getSDWASrcForVT <ValueType VT> {
RegisterOperand retFlt = !if(!eq(VT.Size, 16), SDWASrc_f16, SDWASrc_f32);
RegisterOperand retInt = !if(!eq(VT.Size, 16), SDWASrc_i16, SDWASrc_i32);
RegisterOperand ret = !if(VT.isFP, retFlt, retInt);
}
// Returns the register class to use for sources of VOP3 instructions for the
// given VT.
class getVOP3SrcForVT<ValueType VT, bit IsTrue16 = 0> {
RegisterOperand ret =
!cond(!eq(VT, f64) : VSrc_f64,
!eq(VT, f32) : VSrc_f32,
!eq(VT, f16) : !if(IsTrue16, VSrcT_f16, VSrc_f16),
!eq(VT, bf16) : !if(IsTrue16, VSrcT_bf16, VSrc_bf16),
!eq(VT, i16) : !if(IsTrue16, VSrcT_b16, VSrc_b16),
!eq(VT, i1) : SSrc_i1,
!eq(VT, v2f32) : VSrc_v2f32,
!eq(VT, v2i32) : VSrc_v2b32,
!eq(VT, v2f16) : VSrc_v2f16,
!eq(VT, v2bf16) : VSrc_v2bf16,
!eq(VT, v2i16) : VSrc_v2b16,
!eq(VT, v4f16) : AVSrc_64,
!eq(VT, v4bf16) : AVSrc_64,
!eq(VT.Size, 128) : VRegSrc_128,
!eq(VT.Size, 96) : VRegSrc_96,
!eq(VT.Size, 64) : VSrc_b64,
1 : VSrc_b32);
}
// Src2 of VOP3 DPP instructions cannot be a literal
class getVOP3DPPSrcForVT<ValueType VT> {
RegisterOperand ret =
!cond(!eq(VT, i1) : SSrc_i1,
!eq(VT, i16) : VCSrc_b16,
!eq(VT, f16) : VCSrc_f16,
!eq(VT, bf16) : VCSrc_bf16,
!eq(VT, v2i16) : VCSrc_v2b16,
!eq(VT, v2f16) : VCSrc_v2f16,
!eq(VT, v2bf16) : VCSrc_v2bf16,
!eq(VT, f32) : VCSrc_f32,
1 : VCSrc_b32);
}
// Float or packed int
class isModifierType<ValueType SrcVT> {
bit ret = !or(!eq(SrcVT.Value, f16.Value),
!eq(SrcVT.Value, bf16.Value),
!eq(SrcVT.Value, f32.Value),
!eq(SrcVT.Value, f64.Value),
!eq(SrcVT.Value, v2f16.Value),
!eq(SrcVT.Value, v2i16.Value),
!eq(SrcVT.Value, v2bf16.Value),
!eq(SrcVT.Value, v2f32.Value),
!eq(SrcVT.Value, v2i32.Value),
!eq(SrcVT.Value, v4f16.Value),
!eq(SrcVT.Value, v4i16.Value),
!eq(SrcVT.Value, v4bf16.Value),
!eq(SrcVT.Value, v4f32.Value),
!eq(SrcVT.Value, v4i32.Value),
!eq(SrcVT.Value, v8f16.Value),
!eq(SrcVT.Value, v8i16.Value),
!eq(SrcVT.Value, v8bf16.Value),
!eq(SrcVT.Value, v8f32.Value),
!eq(SrcVT.Value, v8i32.Value),
!eq(SrcVT.Value, v16f16.Value),
!eq(SrcVT.Value, v16i16.Value),
!eq(SrcVT.Value, v16bf16.Value));
}
// Return type of input modifiers operand for specified input operand
class getSrcMod <ValueType VT, bit IsTrue16 = 0> {
Operand ret = !if(!eq(VT.Size, 64),
!if(VT.isFP, FP64InputMods, Int64InputMods),
!if(!eq(VT.Size, 16),
!if(VT.isFP, !if(IsTrue16, FPT16InputMods, FP16InputMods),
!if(IsTrue16, IntT16InputMods, IntOpSelMods)),
!if(VT.isFP, FP32InputMods, Int32InputMods)));
}
class getOpSelMod <ValueType VT> {
Operand ret = !cond(!eq(VT, f16) : FP16InputMods,
!eq(VT, bf16) : FP16InputMods,
!eq(VT, v2f16) : PackedF16InputMods,
!eq(VT, v2bf16) : PackedF16InputMods,
1 : IntOpSelMods);
}
// Return type of input modifiers operand specified input operand for DPP
class getSrcModDPP <ValueType VT> {
Operand ret = !if(VT.isFP, FPVRegInputMods, IntVRegInputMods);
}
class getSrcModDPP_t16 <ValueType VT, bit IsFake16 = 1> {
Operand ret =
!if (VT.isFP,
!if (!or(!eq(VT.Value, f16.Value), !eq(VT.Value, bf16.Value)),
FPT16VRegInputMods<IsFake16>, FPVRegInputMods),
!if (!eq(VT.Value, i16.Value),
IntT16VRegInputMods<IsFake16>, IntVRegInputMods));
}
// Return type of input modifiers operand for specified input operand for DPP
class getSrcModVOP3DPP <ValueType VT, bit IsFake16 = 1> {
Operand ret =
!if (VT.isFP,
!if (!or(!eq(VT.Value, f16.Value), !eq(VT.Value, bf16.Value)),
FP16VCSrcInputMods<IsFake16>, FP32VCSrcInputMods),
Int32VCSrcInputMods);
}
// Return type of input modifiers operand specified input operand for SDWA
class getSrcModSDWA <ValueType VT> {
Operand ret = !if(!eq(VT.Value, f16.Value), FP16SDWAInputMods,
!if(!eq(VT.Value, f32.Value), FP32SDWAInputMods,
!if(!eq(VT.Value, i16.Value), Int16SDWAInputMods,
!if(!eq(VT.Value, bf16.Value), FP16SDWAInputMods,
Int32SDWAInputMods))));
}
// Returns the input arguments for VOP[12C] instructions for the given SrcVT.
class getIns32 <RegisterOperand Src0RC, RegisterOperand Src1RC, int NumSrcArgs> {
dag ret = !if(!eq(NumSrcArgs, 1), (ins Src0RC:$src0), // VOP1
!if(!eq(NumSrcArgs, 2), (ins Src0RC:$src0, Src1RC:$src1), // VOP2
(ins)));
}
// Returns the input arguments for VOP3 instructions for the given SrcVT.
class getIns64 <RegisterOperand Src0RC, RegisterOperand Src1RC,
RegisterOperand Src2RC, int NumSrcArgs,
bit HasClamp, bit HasModifiers, bit HasSrc2Mods, bit HasOMod,
Operand Src0Mod, Operand Src1Mod, Operand Src2Mod> {
dag ret =
!if (!eq(NumSrcArgs, 0),
// VOP1 without input operands (V_NOP, V_CLREXCP)
(ins),
/* else */
!if (!eq(NumSrcArgs, 1),
!if (HasModifiers,
// VOP1 with modifiers
!if(HasOMod,
(ins Src0Mod:$src0_modifiers, Src0RC:$src0,
clampmod0:$clamp, omod0:$omod),
!if (HasClamp,
(ins Src0Mod:$src0_modifiers, Src0RC:$src0, clampmod0:$clamp),
(ins Src0Mod:$src0_modifiers, Src0RC:$src0)))
/* else */,
// VOP1 without modifiers
!if (HasClamp,
(ins Src0RC:$src0, clampmod0:$clamp),
(ins Src0RC:$src0))
/* endif */ ),
!if (!eq(NumSrcArgs, 2),
!if (HasModifiers,
// VOP 2 with modifiers
!if(HasOMod,
(ins Src0Mod:$src0_modifiers, Src0RC:$src0,
Src1Mod:$src1_modifiers, Src1RC:$src1,
clampmod0:$clamp, omod0:$omod),
(ins Src0Mod:$src0_modifiers, Src0RC:$src0,
Src1Mod:$src1_modifiers, Src1RC:$src1,
clampmod0:$clamp))
/* else */,
// VOP2 without modifiers
!if (HasClamp,
(ins Src0RC:$src0, Src1RC:$src1, clampmod0:$clamp),
(ins Src0RC:$src0, Src1RC:$src1))
/* endif */ )
/* NumSrcArgs == 3 */,
!if (HasModifiers,
!if (HasSrc2Mods,
// VOP3 with modifiers
!if (HasOMod,
(ins Src0Mod:$src0_modifiers, Src0RC:$src0,
Src1Mod:$src1_modifiers, Src1RC:$src1,
Src2Mod:$src2_modifiers, Src2RC:$src2,
clampmod0:$clamp, omod0:$omod),
!if (HasClamp,
(ins Src0Mod:$src0_modifiers, Src0RC:$src0,
Src1Mod:$src1_modifiers, Src1RC:$src1,
Src2Mod:$src2_modifiers, Src2RC:$src2,
clampmod0:$clamp),
(ins Src0Mod:$src0_modifiers, Src0RC:$src0,
Src1Mod:$src1_modifiers, Src1RC:$src1,
Src2Mod:$src2_modifiers, Src2RC:$src2))),
// VOP3 with modifiers except src2
!if (HasOMod,
(ins Src0Mod:$src0_modifiers, Src0RC:$src0,
Src1Mod:$src1_modifiers, Src1RC:$src1,
Src2RC:$src2, clampmod0:$clamp, omod0:$omod),
!if (HasClamp,
(ins Src0Mod:$src0_modifiers, Src0RC:$src0,
Src1Mod:$src1_modifiers, Src1RC:$src1,
Src2RC:$src2, clampmod0:$clamp),
(ins Src0Mod:$src0_modifiers, Src0RC:$src0,
Src1Mod:$src1_modifiers, Src1RC:$src1,
Src2RC:$src2))))
/* else */,
// VOP3 without modifiers
!if (HasClamp,
(ins Src0RC:$src0, Src1RC:$src1, Src2RC:$src2, clampmod0:$clamp),
(ins Src0RC:$src0, Src1RC:$src1, Src2RC:$src2))
/* endif */ ))));
}
class getInsVOP3Base<RegisterOperand Src0RC, RegisterOperand Src1RC,
RegisterOperand Src2RC, int NumSrcArgs,
bit HasClamp, bit HasModifiers, bit HasSrc2Mods, bit HasOMod,
Operand Src0Mod, Operand Src1Mod, Operand Src2Mod, bit HasOpSel> {
// getInst64 handles clamp and omod. implicit mutex between vop3p and omod
dag base = getIns64 <Src0RC, Src1RC, Src2RC, NumSrcArgs,
HasClamp, HasModifiers, HasSrc2Mods, HasOMod,
Src0Mod, Src1Mod, Src2Mod>.ret;
dag opsel = (ins op_sel0:$op_sel);
dag ret = !con(base, !if(HasOpSel, opsel, (ins)));
}
class getInsVOP3P <RegisterOperand Src0RC, RegisterOperand Src1RC,
RegisterOperand Src2RC, int NumSrcArgs, bit HasClamp, bit HasOpSel,
Operand Src0Mod, Operand Src1Mod, Operand Src2Mod> {
dag base = getInsVOP3Base<Src0RC, Src1RC, Src2RC, NumSrcArgs,
HasClamp, 1/*HasModifiers*/, 1/*HasSrc2Mods*/,
0/*HasOMod*/, Src0Mod, Src1Mod, Src2Mod, HasOpSel>.ret;
dag vop3pOpsel = (ins op_sel_hi0:$op_sel_hi);
dag vop3p_neg = (ins neg_lo0:$neg_lo, neg_hi0:$neg_hi);
dag vop3pFields = !con(!if(HasOpSel, vop3pOpsel, (ins)), vop3p_neg);
dag ret = !con(base, vop3pFields);
}
class getInsVOP3OpSel <RegisterOperand Src0RC, RegisterOperand Src1RC,
RegisterOperand Src2RC, int NumSrcArgs,
bit HasClamp, bit HasOMod,
Operand Src0Mod, Operand Src1Mod, Operand Src2Mod> {
dag ret = getInsVOP3Base<Src0RC, Src1RC,
Src2RC, NumSrcArgs,
HasClamp, 1/*HasModifiers*/, 1/*HasSrc2Mods*/, HasOMod,
Src0Mod, Src1Mod, Src2Mod, /*HasOpSel=*/1>.ret;
}
class getInsDPPBase <RegisterOperand OldRC, RegisterOperand Src0RC, RegisterOperand Src1RC,
RegisterOperand Src2RC, int NumSrcArgs, bit HasModifiers,
Operand Src0Mod, Operand Src1Mod, Operand Src2Mod, bit HasOld> {
dag ret = !if(!eq(NumSrcArgs, 0),
// VOP1 without input operands (V_NOP)
(ins ),
!con(
!if(HasOld ,(ins OldRC:$old), (ins)),
!if (!eq(NumSrcArgs, 1),
!if (HasModifiers,
// VOP1_DPP with modifiers
(ins Src0Mod:$src0_modifiers, Src0RC:$src0)
/* else */,
// VOP1_DPP without modifiers
(ins Src0RC:$src0)
/* endif */),
!if (!eq(NumSrcArgs, 2),
!if (HasModifiers,
// VOP2_DPP with modifiers
(ins Src0Mod:$src0_modifiers, Src0RC:$src0,
Src1Mod:$src1_modifiers, Src1RC:$src1)
/* else */,
// VOP2_DPP without modifiers
(ins Src0RC:$src0, Src1RC:$src1)
)
/* NumSrcArgs == 3, VOP3 */,
!if (HasModifiers,
// VOP3_DPP with modifiers
(ins Src0Mod:$src0_modifiers, Src0RC:$src0,
Src1Mod:$src1_modifiers, Src1RC:$src1,
Src2Mod:$src2_modifiers, Src2RC:$src2)
/* else */,
// VOP3_DPP without modifiers
(ins Src0RC:$src0, Src1RC:$src1,
Src2RC:$src2)
)
)
)
)
);
}
class getInsDPP <RegisterOperand OldRC, RegisterOperand Src0RC, RegisterOperand Src1RC,
RegisterOperand Src2RC, int NumSrcArgs, bit HasModifiers,
Operand Src0Mod, Operand Src1Mod, Operand Src2Mod, bit HasOld = 1> {
dag ret = !con(getInsDPPBase<OldRC, Src0RC, Src1RC, Src2RC, NumSrcArgs,
HasModifiers, Src0Mod, Src1Mod, Src2Mod, HasOld>.ret,
(ins dpp_ctrl:$dpp_ctrl, DppRowMask:$row_mask,
DppBankMask:$bank_mask, DppBoundCtrl:$bound_ctrl));
}
class getInsDPP16 <RegisterOperand OldRC, RegisterOperand Src0RC, RegisterOperand Src1RC,
RegisterOperand Src2RC, int NumSrcArgs, bit HasModifiers,
Operand Src0Mod, Operand Src1Mod, Operand Src2Mod, bit HasOld = 1> {
dag ret = !con(getInsDPP<OldRC, Src0RC, Src1RC, Src2RC, NumSrcArgs,
HasModifiers, Src0Mod, Src1Mod, Src2Mod, HasOld>.ret,
(ins DppFI:$fi));
}
class getInsDPP8 <RegisterOperand OldRC, RegisterOperand Src0RC, RegisterOperand Src1RC,
RegisterOperand Src2RC, int NumSrcArgs, bit HasModifiers,
Operand Src0Mod, Operand Src1Mod, Operand Src2Mod, bit HasOld = 1> {
dag ret = !con(getInsDPPBase<OldRC, Src0RC, Src1RC, Src2RC, NumSrcArgs,
HasModifiers, Src0Mod, Src1Mod, Src2Mod, HasOld>.ret,
(ins dpp8:$dpp8, DppFI:$fi));
}
class getInsVOP3DPPBase<dag VOP3Base, RegisterOperand OldRC, int NumSrcArgs, bit HasOld> {
dag old = ( ins OldRC:$old );
dag base = VOP3Base;
dag ret = !con(
!if(!and(HasOld,!ne(NumSrcArgs, 0)), old, (ins)),
base
);
}
class getInsVOP3DPP<dag VOP3Base, RegisterOperand OldRC, int NumSrcArgs, bit HasOld = 1> {
dag ret = !con(getInsVOP3DPPBase<VOP3Base,OldRC,NumSrcArgs,HasOld>.ret,
(ins dpp_ctrl:$dpp_ctrl, DppRowMask:$row_mask,
DppBankMask:$bank_mask, DppBoundCtrl:$bound_ctrl));
}
class getInsVOP3DPP16<dag VOP3Base, RegisterOperand OldRC, int NumSrcArgs, bit HasOld = 1> {
dag ret = !con(getInsVOP3DPP<VOP3Base,OldRC,NumSrcArgs,HasOld>.ret,
(ins DppFI:$fi));
}
class getInsVOP3DPP8<dag VOP3Base, RegisterOperand OldRC, int NumSrcArgs, bit HasOld = 1> {
dag ret = !con(getInsVOP3DPPBase<VOP3Base,OldRC,NumSrcArgs,HasOld>.ret,
(ins dpp8:$dpp8, DppFI:$fi));
}
// Ins for SDWA
class getInsSDWA <RegisterOperand Src0RC, RegisterOperand Src1RC, int NumSrcArgs,
bit HasSDWAOMod, Operand Src0Mod, Operand Src1Mod,
ValueType DstVT> {
dag ret = !if(!eq(NumSrcArgs, 0),
// VOP1 without input operands (V_NOP)
(ins),
!if(!eq(NumSrcArgs, 1),
// VOP1
!if(!not(HasSDWAOMod),
// VOP1_SDWA without omod
(ins Src0Mod:$src0_modifiers, Src0RC:$src0,
clampmod:$clamp,
dst_sel:$dst_sel, dst_unused:$dst_unused,
src0_sel:$src0_sel),
// VOP1_SDWA with omod
(ins Src0Mod:$src0_modifiers, Src0RC:$src0,
clampmod:$clamp, omod:$omod,
dst_sel:$dst_sel, dst_unused:$dst_unused,
src0_sel:$src0_sel)),
!if(!eq(NumSrcArgs, 2),
!if(!eq(DstVT.Size, 1),
// VOPC_SDWA
(ins Src0Mod:$src0_modifiers, Src0RC:$src0,
Src1Mod:$src1_modifiers, Src1RC:$src1,
clampmod:$clamp, src0_sel:$src0_sel, src1_sel:$src1_sel),
// VOP2_SDWA
!if(!not(HasSDWAOMod),
// VOP2_SDWA without omod
(ins Src0Mod:$src0_modifiers, Src0RC:$src0,
Src1Mod:$src1_modifiers, Src1RC:$src1,
clampmod:$clamp,
dst_sel:$dst_sel, dst_unused:$dst_unused,
src0_sel:$src0_sel, src1_sel:$src1_sel),
// VOP2_SDWA with omod
(ins Src0Mod:$src0_modifiers, Src0RC:$src0,
Src1Mod:$src1_modifiers, Src1RC:$src1,
clampmod:$clamp, omod:$omod,
dst_sel:$dst_sel, dst_unused:$dst_unused,
src0_sel:$src0_sel, src1_sel:$src1_sel))),
(ins)/* endif */)));
}
// Outs for DPP
class getOutsDPP <bit HasDst, ValueType DstVT, RegisterOperand DstRCDPP> {
dag ret = !if(HasDst,
!if(!eq(DstVT.Size, 1),
(outs), // no dst for VOPC, we use "vcc"-token as dst in SDWA VOPC instructions
(outs DstRCDPP:$vdst)),
(outs)); // V_NOP
}
// Outs for SDWA
class getOutsSDWA <bit HasDst, ValueType DstVT, RegisterOperand DstRCSDWA> {
dag ret = !if(HasDst,
!if(!eq(DstVT.Size, 1),
(outs DstRCSDWA:$sdst),
(outs DstRCSDWA:$vdst)),
(outs)); // V_NOP
}
// Returns the assembly string for the inputs and outputs of a VOP[12C]
// instruction.
class getAsm32 <bit HasDst, int NumSrcArgs, ValueType DstVT = i32> {
string dst = !if(!eq(DstVT.Size, 1), "$sdst", "$vdst"); // use $sdst for VOPC
string src0 = ", $src0";
string src1 = ", $src1";
string src2 = ", $src2";
string ret = !if(HasDst, dst, "") #
!if(!eq(NumSrcArgs, 1), src0, "") #
!if(!eq(NumSrcArgs, 2), src0#src1, "") #
!if(!eq(NumSrcArgs, 3), src0#src1#src2, "");
}
class getAsmVOPDPart <int NumSrcArgs, string XorY> {
string dst = "$vdst" # XorY;
string src0 = ", $src0" # XorY;
string src1 = ", $vsrc1" # XorY;
string ret = dst #
!if(!ge(NumSrcArgs, 1), src0, "") #
!if(!ge(NumSrcArgs, 2), src1, "");
}
// Returns the assembly string for the inputs and outputs of a VOP3P
// instruction.
class getAsmVOP3P <int NumSrcArgs, bit HasModifiers,
bit HasClamp, bit HasOpSel> {
string dst = "$vdst";
string src0 = !if(!eq(NumSrcArgs, 1), "$src0", "$src0,");
string src1 = !if(!eq(NumSrcArgs, 1), "",
!if(!eq(NumSrcArgs, 2), " $src1",
" $src1,"));
string src2 = !if(!eq(NumSrcArgs, 3), " $src2", "");
string mods = !if(HasModifiers, "$neg_lo$neg_hi", "");
string clamp = !if(HasClamp, "$clamp", "");
string opsel = !if(HasOpSel, "$op_sel$op_sel_hi", "");
// Each modifier is printed as an array of bits for each operand, so
// all operands are printed as part of src0_modifiers.
string ret = dst#", "#src0#src1#src2#opsel#mods#clamp;
}
class getAsmVOP3OpSel <int NumSrcArgs,
bit HasClamp,
bit HasOMod,
bit Src0HasMods,
bit Src1HasMods,
bit Src2HasMods> {
string dst = "$vdst";
string isrc0 = !if(!eq(NumSrcArgs, 1), "$src0", "$src0,");
string isrc1 = !if(!eq(NumSrcArgs, 1), "",
!if(!eq(NumSrcArgs, 2), " $src1",
" $src1,"));
string isrc2 = !if(!eq(NumSrcArgs, 3), " $src2", "");
string fsrc0 = !if(!eq(NumSrcArgs, 1), "$src0_modifiers", "$src0_modifiers,");
string fsrc1 = !if(!eq(NumSrcArgs, 1), "",
!if(!eq(NumSrcArgs, 2), " $src1_modifiers",
" $src1_modifiers,"));
string fsrc2 = !if(!eq(NumSrcArgs, 3), " $src2_modifiers", "");
string src0 = !if(Src0HasMods, fsrc0, isrc0);
string src1 = !if(Src1HasMods, fsrc1, isrc1);
string src2 = !if(Src2HasMods, fsrc2, isrc2);
string clamp = !if(HasClamp, "$clamp", "");
string omod = !if(HasOMod, "$omod", "");
string ret = dst#", "#src0#src1#src2#"$op_sel"#clamp#omod;
}
class getAsmDPP <bit HasDst, int NumSrcArgs, bit HasModifiers, ValueType DstVT = i32> {
string dst = !if(HasDst,
!if(!eq(DstVT.Size, 1),
"$sdst",
"$vdst"),
""); // use $sdst for VOPC
string src0 = !if(!eq(NumSrcArgs, 1), "$src0_modifiers", "$src0_modifiers,");
string src1 = !if(!eq(NumSrcArgs, 1), "",
!if(!eq(NumSrcArgs, 2), " $src1_modifiers",
" $src1_modifiers,"));
string args = !if(!not(HasModifiers),
getAsm32<0, NumSrcArgs, DstVT>.ret,
", "#src0#src1);
string ret = dst#args#" $dpp_ctrl$row_mask$bank_mask$bound_ctrl";
}
class getAsmDPP16 <bit HasDst, int NumSrcArgs, bit HasModifiers, ValueType DstVT = i32> {
string ret = getAsmDPP<HasDst, NumSrcArgs, HasModifiers, DstVT>.ret#"$fi";
}
class getAsmDPP8 <bit HasDst, int NumSrcArgs, bit HasModifiers, ValueType DstVT = i32>
: getAsmDPP<HasDst, NumSrcArgs, HasModifiers, DstVT>{
let ret = dst#args#" $dpp8$fi";
}
class getAsmVOP3Base <int NumSrcArgs, bit HasDst, bit HasClamp,
bit HasOpSel, bit HasOMod, bit IsVOP3P,
bit HasModifiers, bit Src0HasMods,
bit Src1HasMods, bit Src2HasMods, ValueType DstVT = i32> {
string dst = !if(HasDst,
!if(!eq(DstVT.Size, 1),
"$sdst",
"$vdst"),
""); // use $sdst for VOPC
string src0nomods = !if(!eq(NumSrcArgs, 1), "$src0", "$src0,");
string src1nomods = !if(!eq(NumSrcArgs, 1), "",
!if(!eq(NumSrcArgs, 2), " $src1",
" $src1,"));
string src2nomods = !if(!eq(NumSrcArgs, 3), " $src2", "");
string src0mods = !if(!eq(NumSrcArgs, 1), "$src0_modifiers", "$src0_modifiers,");
string src1mods = !if(!eq(NumSrcArgs, 1), "",
!if(!eq(NumSrcArgs, 2), " $src1_modifiers",
" $src1_modifiers,"));
string src2mods = !if(!eq(NumSrcArgs, 3), " $src2_modifiers", "");
string src0 = !if(Src0HasMods, src0mods, src0nomods);
string src1 = !if(Src1HasMods, src1mods, src1nomods);
string src2 = !if(Src2HasMods, src2mods, src2nomods);
string opsel = !if(HasOpSel, "$op_sel", "");
string 3PMods = !if(IsVOP3P,
!if(HasOpSel, "$op_sel_hi", "")
#!if(HasModifiers, "$neg_lo$neg_hi", ""),
"");
string clamp = !if(HasClamp, "$clamp", "");
string omod = !if(HasOMod, "$omod", "");
string ret = dst#!if(!gt(NumSrcArgs,0),", "#src0#src1#src2#opsel#3PMods#clamp#omod, "");
}
class getAsmVOP3DPP<string base> {
string ret = base # " $dpp_ctrl$row_mask$bank_mask$bound_ctrl";
}
class getAsmVOP3DPP16<string base> {
string ret = getAsmVOP3DPP<base>.ret # "$fi";
}
class getAsmVOP3DPP8<string base> {
string ret = base # " $dpp8$fi";
}
class getAsmSDWA <bit HasDst, int NumSrcArgs, ValueType DstVT = i32> {
string dst = !if(HasDst,
!if(!eq(DstVT.Size, 1),
" vcc", // use vcc token as dst for VOPC instructions
"$vdst"),
"");
string src0 = "$src0_modifiers";
string src1 = "$src1_modifiers";
string args = !if(!eq(NumSrcArgs, 0),
"",
!if(!eq(NumSrcArgs, 1),
", "#src0#"$clamp",
", "#src0#", "#src1#"$clamp"
)
);
string sdwa = !if(!eq(NumSrcArgs, 0),
"",
!if(!eq(NumSrcArgs, 1),
" $dst_sel $dst_unused $src0_sel",
!if(!eq(DstVT.Size, 1),
" $src0_sel $src1_sel", // No dst_sel and dst_unused for VOPC
" $dst_sel $dst_unused $src0_sel $src1_sel"
)
)
);
string ret = dst#args#sdwa;
}
class getAsmSDWA9 <bit HasDst, bit HasOMod, int NumSrcArgs,
ValueType DstVT = i32> {
string dst = !if(HasDst,
!if(!eq(DstVT.Size, 1),
"$sdst", // VOPC
"$vdst"), // VOP1/2
"");
string src0 = "$src0_modifiers";
string src1 = "$src1_modifiers";
string out_mods = !if(!not(HasOMod), "$clamp", "$clamp$omod");
string args = !if(!eq(NumSrcArgs, 0), "",
!if(!eq(NumSrcArgs, 1),
", "#src0,
", "#src0#", "#src1
)
);
string sdwa = !if(!eq(NumSrcArgs, 0), "",
!if(!eq(NumSrcArgs, 1),
out_mods#" $dst_sel $dst_unused $src0_sel",
!if(!eq(DstVT.Size, 1),
" $src0_sel $src1_sel", // No dst_sel, dst_unused and output modifiers for VOPC
out_mods#" $dst_sel $dst_unused $src0_sel $src1_sel"
)
)
);
string ret = dst#args#sdwa;
}
class getHas64BitOps <int NumSrcArgs, ValueType DstVT, ValueType Src0VT,
ValueType Src1VT> {
bit ret = !if(!eq(NumSrcArgs, 3),
0,
!if(!eq(DstVT.Size, 64),
1,
!if(!eq(Src0VT.Size, 64),
1,
!if(!eq(Src1VT.Size, 64),
1,
0
)
)
)
);
}
class getHasSDWA <int NumSrcArgs, ValueType DstVT = i32, ValueType Src0VT = i32,
ValueType Src1VT = i32> {
bit ret = !if(!eq(NumSrcArgs, 3),
0, // NumSrcArgs == 3 - No SDWA for VOP3
!if(!eq(DstVT.Size, 64),
0, // 64-bit dst - No SDWA for 64-bit operands
!if(!eq(Src0VT.Size, 64),
0, // 64-bit src0
!if(!eq(Src1VT.Size, 64),
0, // 64-bit src2
1
)
)
)
);
}
class getHasDPP <int NumSrcArgs> {
bit ret = !if(!eq(NumSrcArgs, 3),
0, // NumSrcArgs == 3 - No DPP for VOP3
1);
}
class getHasExt32BitDPP <int NumSrcArgs, ValueType DstVT = i32, ValueType Src0VT = i32,
ValueType Src1VT = i32> {
bit ret = !and(getHasDPP<NumSrcArgs>.ret,
!not(getHas64BitOps<NumSrcArgs, DstVT, Src0VT, Src1VT>.ret));
}
class getHasExt64BitDPP <int NumSrcArgs, ValueType DstVT = i32, ValueType Src0VT = i32,
ValueType Src1VT = i32> {
bit ret = !and(getHasDPP<NumSrcArgs>.ret,
getHas64BitOps<NumSrcArgs, DstVT, Src0VT, Src1VT>.ret);
}
// Function that checks if instruction supports DPP and SDWA
class getHasExt <int NumSrcArgs, ValueType DstVT = i32, ValueType Src0VT = i32,
ValueType Src1VT = i32> {
bit ret = !or(getHasDPP<NumSrcArgs>.ret,
getHasSDWA<NumSrcArgs, DstVT, Src0VT, Src1VT>.ret);
}
// Return an AGPR+VGPR operand class for the given VGPR register class.
class getLdStRegisterOperand<RegisterClass RC> {
RegisterOperand ret =
!if(!eq(RC.Size, 32), AVLdSt_32,
!if(!eq(RC.Size, 64), AVLdSt_64,
!if(!eq(RC.Size, 96), AVLdSt_96,
!if(!eq(RC.Size, 128), AVLdSt_128,
!if(!eq(RC.Size, 160), AVLdSt_160,
RegisterOperand<VReg_1> // invalid register
)))));
}
class getHasVOP3DPP <ValueType DstVT = i32, ValueType Src0VT = i32,
ValueType Src1VT = i32, ValueType Src2VT = i32> {
bit ret = !if(!eq(DstVT.Size, 64),
0, // 64-bit dst No DPP for 64-bit operands
!if(!eq(Src0VT.Size, 64),
0, // 64-bit src0
!if(!eq(Src1VT.Size, 64),
0, // 64-bit src1
!if(!eq(Src2VT.Size, 64),
0, // 64-bit src2
1
)
)
)
);
}
def PatGenMode {
int NoPattern = 0;
int Pattern = 1;
}
class VOPProfile <list<ValueType> _ArgVT, bit _EnableClamp = 0> {
field list<ValueType> ArgVT = _ArgVT;
field bit EnableClamp = _EnableClamp;
field bit IsTrue16 = 0;
field bit IsRealTrue16 = 0;
field ValueType DstVT = ArgVT[0];
field ValueType Src0VT = ArgVT[1];
field ValueType Src1VT = ArgVT[2];
field ValueType Src2VT = ArgVT[3];
field RegisterOperand DstRC = getVALUDstForVT<DstVT>.ret;
field RegisterOperand DstRCDPP = DstRC;
field RegisterOperand DstRC64 = DstRC;
field RegisterOperand DstRCVOP3DPP = DstRC64;
field RegisterOperand DstRCSDWA = getSDWADstForVT<DstVT>.ret;
field RegisterOperand Src0RC32 = getVOPSrc0ForVT<Src0VT, IsTrue16>.ret;
field RegisterOperand Src1RC32 = getVregSrcForVT<Src1VT>.ret;
field RegisterOperand Src0RC64 = getVOP3SrcForVT<Src0VT>.ret;
field RegisterOperand Src1RC64 = getVOP3SrcForVT<Src1VT>.ret;
field RegisterOperand Src2RC64 = getVOP3SrcForVT<Src2VT>.ret;
field RegisterOperand Src0DPP = getVregSrcForVT<Src0VT>.ret;
field RegisterOperand Src1DPP = getVregSrcForVT<Src1VT>.ret;
field RegisterOperand Src2DPP = getVregSrcForVT<Src2VT>.ret;
field RegisterOperand Src0VOP3DPP = VGPRSrc_32;
field RegisterOperand Src1VOP3DPP = getVOP3DPPSrcForVT<Src1VT>.ret;
field RegisterOperand Src2VOP3DPP = getVOP3DPPSrcForVT<Src2VT>.ret;
field RegisterOperand Src0SDWA = getSDWASrcForVT<Src0VT>.ret;
field RegisterOperand Src1SDWA = getSDWASrcForVT<Src0VT>.ret;
field Operand Src0Mod = getSrcMod<Src0VT>.ret;
field Operand Src1Mod = getSrcMod<Src1VT>.ret;
field Operand Src2Mod = getSrcMod<Src2VT>.ret;
field Operand Src0ModDPP = getSrcModDPP<Src0VT>.ret;
field Operand Src1ModDPP = getSrcModDPP<Src1VT>.ret;
field Operand Src2ModDPP = getSrcModDPP<Src2VT>.ret;
field Operand Src0ModVOP3DPP = getSrcModDPP<Src0VT>.ret;
field Operand Src1ModVOP3DPP = getSrcModDPP<Src1VT>.ret;
field Operand Src2ModVOP3DPP = getSrcModVOP3DPP<Src2VT>.ret;
field Operand Src0ModSDWA = getSrcModSDWA<Src0VT>.ret;
field Operand Src1ModSDWA = getSrcModSDWA<Src1VT>.ret;
field bit IsMAI = 0;
field bit IsVOP3P = 0;
field bit IsDOT = 0;
field bit IsSingle = 0;
field bit IsWMMA = 0;
field bit IsSWMMAC = 0;
field bit IsFP8 = 0;
field bit HasDst = !ne(DstVT.Value, untyped.Value);
field bit HasDst32 = HasDst;
field bit EmitDst = HasDst; // force dst encoding, see v_movreld_b32 special case
field bit EmitDstSel = EmitDst;
field int NumSrcArgs = getNumSrcArgs<Src0VT, Src1VT, Src2VT>.ret;
field bit HasSrc0 = !ne(Src0VT.Value, untyped.Value);
field bit HasSrc1 = !ne(Src1VT.Value, untyped.Value);
field bit HasSrc2 = !ne(Src2VT.Value, untyped.Value);
field bit HasSrc0FloatMods = Src0VT.isFP;
field bit HasSrc1FloatMods = Src1VT.isFP;
field bit HasSrc2FloatMods = Src2VT.isFP;
field bit HasSrc0IntMods = isIntType<Src0VT>.ret;
field bit HasSrc1IntMods = isIntType<Src1VT>.ret;
field bit HasSrc2IntMods = isIntType<Src2VT>.ret;
field bit HasClamp = !or(isModifierType<Src0VT>.ret, EnableClamp);
field bit HasSDWAClamp = EmitDst;
field bit HasFPClamp = !and(DstVT.isFP, HasClamp);
field bit HasIntClamp = !if(DstVT.isFP, 0, HasClamp);
field bit HasClampLo = HasClamp;
field bit HasClampHi = !and(DstVT.isVector, HasClamp);
field bit HasHigh = 0;
field bit IsPacked = Src0VT.isVector;
field bit HasOpSel = IsPacked;
field bit HasOMod = !if(IsVOP3P, 0, DstVT.isFP);
field bit HasSDWAOMod = DstVT.isFP;
field bit HasModifiers = !or(isModifierType<Src0VT>.ret,
isModifierType<Src1VT>.ret,
isModifierType<Src2VT>.ret,
HasOMod);
field bit HasSrc0Mods = HasModifiers;
field bit HasSrc1Mods = !if(HasModifiers, !or(HasSrc1FloatMods, HasSrc1IntMods), 0);
field bit HasSrc2Mods = !if(HasModifiers, !or(HasSrc2FloatMods, HasSrc2IntMods), 0);
field bit HasExt = getHasExt<NumSrcArgs, DstVT, Src0VT, Src1VT>.ret;
field bit HasExtVOP3DPP = getHasVOP3DPP<DstVT, Src0VT, Src1VT, Src2VT>.ret;
field bit HasExtDPP = !or(getHasDPP<NumSrcArgs>.ret, HasExtVOP3DPP);
field bit HasExt32BitDPP = getHasExt32BitDPP<NumSrcArgs, DstVT, Src0VT, Src1VT>.ret;
field bit HasExt64BitDPP = getHasExt64BitDPP<NumSrcArgs, DstVT, Src0VT, Src1VT>.ret;
field bit HasExtSDWA = getHasSDWA<NumSrcArgs, DstVT, Src0VT, Src1VT>.ret;
field bit HasExtSDWA9 = HasExtSDWA;
field int NeedPatGen = PatGenMode.NoPattern;
field Operand Src0PackedMod = !if(HasSrc0FloatMods, PackedF16InputMods, PackedI16InputMods);
field Operand Src1PackedMod = !if(HasSrc1FloatMods, PackedF16InputMods, PackedI16InputMods);
field Operand Src2PackedMod = !if(HasSrc2FloatMods, PackedF16InputMods, PackedI16InputMods);
field dag Outs = !if(HasDst,(outs DstRC:$vdst),(outs));
// VOP3b instructions are a special case with a second explicit
// output. This is manually overridden for them.
field dag Outs32 = Outs;
field dag Outs64 = !if(HasDst,(outs DstRC64:$vdst),(outs));
field dag OutsDPP = getOutsDPP<HasDst, DstVT, DstRCDPP>.ret;
field dag OutsDPP8 = OutsDPP;
field dag OutsVOP3DPP = getOutsDPP<HasDst, DstVT, DstRCVOP3DPP>.ret;
field dag OutsVOP3DPP8 = OutsVOP3DPP;
field dag OutsSDWA = getOutsSDWA<HasDst, DstVT, DstRCSDWA>.ret;
field dag Ins32 = getIns32<Src0RC32, Src1RC32, NumSrcArgs>.ret;
field dag Ins64 = getIns64<Src0RC64, Src1RC64, Src2RC64, NumSrcArgs,
HasClamp, HasModifiers, HasSrc2Mods,
HasOMod, Src0Mod, Src1Mod, Src2Mod>.ret;
field dag InsVOP3P = getInsVOP3P<Src0RC64, Src1RC64, Src2RC64,
NumSrcArgs, HasClamp, HasOpSel,
Src0PackedMod, Src1PackedMod, Src2PackedMod>.ret;
field dag InsVOP3OpSel = getInsVOP3OpSel<Src0RC64, Src1RC64, Src2RC64,
NumSrcArgs, HasClamp, HasOMod,
getOpSelMod<Src0VT>.ret,
getOpSelMod<Src1VT>.ret,
getOpSelMod<Src2VT>.ret>.ret;
field dag InsDPP = !if(HasExtDPP,
getInsDPP<DstRCDPP, Src0DPP, Src1DPP, Src2DPP, NumSrcArgs,
HasModifiers, Src0ModDPP, Src1ModDPP, Src2ModDPP>.ret,
(ins));
field dag InsDPP16 = getInsDPP16<DstRCDPP, Src0DPP, Src1DPP, Src2DPP, NumSrcArgs,
HasModifiers, Src0ModDPP, Src1ModDPP, Src2ModDPP>.ret;
field dag InsDPP8 = getInsDPP8<DstRCDPP, Src0DPP, Src1DPP, Src2DPP,
NumSrcArgs, HasModifiers,
Src0ModDPP, Src1ModDPP, Src2ModDPP>.ret;
defvar InsVOP3DPPBase = getInsVOP3Base<Src0VOP3DPP, Src1VOP3DPP,
Src2VOP3DPP, NumSrcArgs, HasClamp, HasModifiers, HasSrc2Mods, HasOMod,
Src0ModVOP3DPP, Src1ModVOP3DPP, Src2ModVOP3DPP, HasOpSel>.ret;
defvar InsVOP3PDPPBase = getInsVOP3P<Src0VOP3DPP, Src1VOP3DPP,
Src2VOP3DPP, NumSrcArgs, HasClamp, HasOpSel,
Src0ModVOP3DPP, Src1ModVOP3DPP, Src2ModVOP3DPP>.ret;
field dag InsVOP3Base = !if(IsVOP3P, InsVOP3PDPPBase, InsVOP3DPPBase);
field dag InsVOP3DPP = getInsVOP3DPP<InsVOP3Base, DstRCVOP3DPP, NumSrcArgs>.ret;
field dag InsVOP3DPP16 = getInsVOP3DPP16<InsVOP3Base, DstRCVOP3DPP, NumSrcArgs>.ret;
field dag InsVOP3DPP8 = getInsVOP3DPP8<InsVOP3Base, DstRCVOP3DPP, NumSrcArgs>.ret;
field dag InsSDWA = getInsSDWA<Src0SDWA, Src1SDWA, NumSrcArgs,
HasSDWAOMod, Src0ModSDWA, Src1ModSDWA,
DstVT>.ret;
field dag InsVOPDX = (ins Src0RC32:$src0X, Src1RC32:$vsrc1X);
// It is a slight misnomer to use the deferred f32 operand type for non-float
// operands, but this operand type will only be used if the other dual
// component is FMAAK or FMAMK
field dag InsVOPDXDeferred = (ins !if(!eq(Src0VT.Size, 32), VSrc_f32_Deferred, VSrc_f16_Deferred):$src0X, VGPR_32:$vsrc1X);
field dag InsVOPDY = (ins Src0RC32:$src0Y, Src1RC32:$vsrc1Y);
field dag InsVOPDYDeferred = (ins !if(!eq(Src1VT.Size, 32), VSrc_f32_Deferred, VSrc_f16_Deferred):$src0Y, VGPR_32:$vsrc1Y);
field string Asm32 = getAsm32<HasDst, NumSrcArgs, DstVT>.ret;
field string AsmDPP = !if(HasExtDPP,
getAsmDPP<HasDst, NumSrcArgs, HasModifiers, DstVT>.ret, "");
field string AsmDPP16 = getAsmDPP16<HasDst, NumSrcArgs, HasModifiers, DstVT>.ret;
// DPP8 encoding has no fields for modifiers, and it is enforced by setting
// the asm operand name via this HasModifiers flag
field string AsmDPP8 = getAsmDPP8<HasDst, NumSrcArgs, 0 /*HasModifiers*/, DstVT>.ret;
field string AsmVOP3Base = getAsmVOP3Base<NumSrcArgs, HasDst, HasClamp,
HasOpSel, HasOMod, IsVOP3P, HasModifiers, HasModifiers, HasModifiers,
HasModifiers, DstVT>.ret;
field string Asm64 = AsmVOP3Base;
field string AsmVOP3P = getAsmVOP3P<NumSrcArgs, HasModifiers, HasClamp, HasOpSel>.ret;
field string AsmVOP3OpSel = getAsmVOP3OpSel<NumSrcArgs,
HasClamp,
HasOMod,
HasSrc0FloatMods,
HasSrc1FloatMods,
HasSrc2FloatMods>.ret;
field string AsmVOP3DPP = getAsmVOP3DPP<AsmVOP3Base>.ret;
field string AsmVOP3DPP16 = getAsmVOP3DPP16<AsmVOP3Base>.ret;
field string AsmVOP3DPP8 = getAsmVOP3DPP8<AsmVOP3Base>.ret;
field string AsmSDWA = getAsmSDWA<HasDst, NumSrcArgs, DstVT>.ret;
field string AsmSDWA9 = getAsmSDWA9<HasDst, HasSDWAOMod, NumSrcArgs, DstVT>.ret;
field string AsmVOPDX = getAsmVOPDPart<NumSrcArgs, "X">.ret;
field string AsmVOPDY = getAsmVOPDPart<NumSrcArgs, "Y">.ret;
field string TieRegDPP = "$old";
}
class VOP_NO_EXT <VOPProfile p> : VOPProfile <p.ArgVT> {
let HasExt = 0;
let HasExtDPP = 0;
let HasExtVOP3DPP = 0;
let HasExt32BitDPP = 0;
let HasExt64BitDPP = 0;
let HasExtSDWA = 0;
let HasExtSDWA9 = 0;
}
class VOP_PAT_GEN <VOPProfile p, int mode=PatGenMode.NoPattern> : VOPProfile <p.ArgVT> {
let NeedPatGen = mode;
}
// VOPC_Profile_t16, VOPC_NoSdst_Profile_t16, VOPC_Class_Profile_t16,
// VOPC_Class_NoSdst_Profile_t16, and VOP_MAC_F16_t16 do not inherit from this
// class, so copy changes to this class in those profiles
class VOPProfile_True16<VOPProfile P> : VOPProfile<P.ArgVT> {
let IsTrue16 = 1;
let IsRealTrue16 = 1;
let HasOpSel = 1;
let HasModifiers = 1; // All instructions at least have OpSel.
// Most DstVT are 16-bit, but not all.
let DstRC = getVALUDstForVT<DstVT, 1 /*IsTrue16*/, 0 /*IsVOP3Encoding*/>.ret;
let DstRC64 = getVALUDstForVT<DstVT>.ret;
let Src0RC32 = getVOPSrc0ForVT<Src0VT, 1 /*IsTrue16*/, 0 /*IsFake16*/>.ret;
let Src1RC32 = getVregSrcForVT<Src1VT, 1 /*IsTrue16*/, 0 /*IsFake16*/>.ret;
let Src0DPP = getVregSrcForVT<Src0VT, 1 /*IsTrue16*/, 0 /*IsFake16*/>.ret;
let Src1DPP = getVregSrcForVT<Src1VT, 1 /*IsTrue16*/, 0 /*IsFake16*/>.ret;
let Src2DPP = getVregSrcForVT<Src2VT, 1 /*IsTrue16*/, 0 /*IsFake16*/>.ret;
let Src0ModDPP = getSrcModDPP_t16<Src0VT, 0 /*IsFake16*/>.ret;
let Src1ModDPP = getSrcModDPP_t16<Src1VT, 0 /*IsFake16*/>.ret;
let Src2ModDPP = getSrcModDPP_t16<Src2VT, 0 /*IsFake16*/>.ret;
let Src0VOP3DPP = VGPRSrc_16;
let Src0ModVOP3DPP = getSrcModVOP3DPP<Src0VT, 0 /*IsFake16*/>.ret;
let Src1ModVOP3DPP = getSrcModVOP3DPP<Src1VT, 0 /*IsFake16*/>.ret;
let Src2ModVOP3DPP = getSrcModVOP3DPP<Src2VT, 0 /*IsFake16*/>.ret;
let DstRC64 = getVALUDstForVT<DstVT, 1 /*IsTrue16*/, 1 /*IsVOP3Encoding*/>.ret;
let Src0RC64 = getVOP3SrcForVT<Src0VT, 1 /*IsTrue16*/>.ret;
let Src1RC64 = getVOP3SrcForVT<Src1VT, 1 /*IsTrue16*/>.ret;
let Src2RC64 = getVOP3SrcForVT<Src2VT, 1 /*IsTrue16*/>.ret;
let Src0Mod = getSrcMod<Src0VT, 1 /*IsTrue16*/>.ret;
let Src1Mod = getSrcMod<Src1VT, 1 /*IsTrue16*/>.ret;
let Src2Mod = getSrcMod<Src2VT, 1 /*IsTrue16*/>.ret;
}
class VOPProfile_Fake16<VOPProfile P> : VOPProfile<P.ArgVT> {
let IsTrue16 = 1;
// Most DstVT are 16-bit, but not all
let DstRC = getVALUDstForVT_fake16<DstVT>.ret;
let DstRC64 = getVALUDstForVT<DstVT>.ret;
let Src1RC32 = getVregSrcForVT<Src1VT, 1/*IsTrue16*/, 1/*IsFake16*/>.ret;
let Src0DPP = getVregSrcForVT<Src0VT, 1/*IsTrue16*/, 1/*IsFake16*/>.ret;
let Src1DPP = getVregSrcForVT<Src1VT, 1/*IsTrue16*/, 1/*IsFake16*/>.ret;
let Src2DPP = getVregSrcForVT<Src2VT, 1/*IsTrue16*/, 1/*IsFake16*/>.ret;
let Src0ModDPP = getSrcModDPP_t16<Src0VT>.ret;
let Src1ModDPP = getSrcModDPP_t16<Src1VT>.ret;
let Src2ModDPP = getSrcModDPP_t16<Src2VT>.ret;
}
def VOP_F16_F16 : VOPProfile<[f16, f16, untyped, untyped]>;
def VOP_F16_I16 : VOPProfile <[f16, i16, untyped, untyped]>;
def VOP_I16_F16 : VOPProfile <[i16, f16, untyped, untyped]>;
def VOP_I16_I16 : VOPProfile <[i16, i16, untyped, untyped]>;
def VOP_F16_F16_F16 : VOPProfile <[f16, f16, f16, untyped]>;
def VOP_F16_F16_I16 : VOPProfile <[f16, f16, i16, untyped]>;
def VOP_F16_F16_I32 : VOPProfile <[f16, f16, i32, untyped]>;
def VOP_I16_I16_I16 : VOPProfile <[i16, i16, i16, untyped]>;
def VOP_I16_I16_I16_ARITH : VOPProfile <[i16, i16, i16, untyped], /*EnableClamp=*/1>;
def VOP_I16_I16_I16_I16 : VOPProfile <[i16, i16, i16, i16, untyped]>;
def VOP_F16_F16_F16_F16 : VOPProfile <[f16, f16, f16, f16, untyped]>;
def VOP_I32_I16_I16_I32 : VOPProfile <[i32, i16, i16, i32, untyped]>;
def VOP_I32_I16 : VOPProfile <[i32, i16, untyped, untyped]>;
def VOP_I16_I32 : VOPProfile <[i16, i32, untyped, untyped]>;
def VOP_V2F16_V2F16_V2F16 : VOPProfile <[v2f16, v2f16, v2f16, untyped]>;
def VOP_V2I16_V2I16_V2I16 : VOPProfile <[v2i16, v2i16, v2i16, untyped]>;
def VOP_B32_F16_F16 : VOPProfile <[i32, f16, f16, untyped]>;
def VOP_V2F16_V2F16_V2F16_V2F16 : VOPProfile <[v2f16, v2f16, v2f16, v2f16]>;
def VOP_V2I16_V2I16_V2I16_V2I16 : VOPProfile <[v2i16, v2i16, v2i16, v2i16]>;
def VOP_V2I16_F32_F32 : VOPProfile <[v2i16, f32, f32, untyped]>;
def VOP_V2I16_I32_I32 : VOPProfile <[v2i16, i32, i32, untyped]>;
def VOP_F16_V2F16_V2F16_F16 : VOPProfile <[f16, v2f16, v2f16, f16]>;
def VOP_BF16_V2BF16_V2BF16_BF16: VOPProfile <[bf16, v2bf16, v2bf16, bf16]>;
def VOP_F32_V2BF16_V2BF16_F32 : VOPProfile <[f32, v2bf16, v2bf16, f32]>;
def VOP_F32_V2F16_V2F16_V2F16 : VOPProfile <[f32, v2f16, v2f16, v2f16]>;
def VOP_NONE : VOPProfile <[untyped, untyped, untyped, untyped]>;
def VOP_F32_F32 : VOPProfile <[f32, f32, untyped, untyped]>;
def VOP_F32_F64 : VOPProfile <[f32, f64, untyped, untyped]>;
def VOP_F32_I32 : VOPProfile <[f32, i32, untyped, untyped]>;
def VOP_F64_F32 : VOPProfile <[f64, f32, untyped, untyped]>;
def VOP_F64_F64 : VOPProfile <[f64, f64, untyped, untyped]>;
def VOP_F64_I32 : VOPProfile <[f64, i32, untyped, untyped]>;
def VOP_I32_F32 : VOPProfile <[i32, f32, untyped, untyped]>;
def VOP_I32_F64 : VOPProfile <[i32, f64, untyped, untyped]>;
def VOP_I32_I32 : VOPProfile <[i32, i32, untyped, untyped]>;
def VOP_F16_F32 : VOPProfile <[f16, f32, untyped, untyped]>;
def VOP_F32_F16 : VOPProfile <[f32, f16, untyped, untyped]>;
def VOP_I64_I64 : VOPProfile <[i64, i64, untyped, untyped]>;
def VOP_F32_F32_F16 : VOPProfile <[f32, f32, f16, untyped]>;
def VOP_F32_F32_F32 : VOPProfile <[f32, f32, f32, untyped]>;
def VOP_F32_F32_I32 : VOPProfile <[f32, f32, i32, untyped]>;
def VOP_F64_F64_F64 : VOPProfile <[f64, f64, f64, untyped]>;
def VOP_F64_F64_I32 : VOPProfile <[f64, f64, i32, untyped]>;
def VOP_I32_F32_F32 : VOPProfile <[i32, f32, f32, untyped]>;
def VOP_I32_F32_I32 : VOPProfile <[i32, f32, i32, untyped]>;
def VOP_I32_I32_I32 : VOPProfile <[i32, i32, i32, untyped]>;
def VOP_I32_I32_I32_ARITH : VOPProfile <[i32, i32, i32, untyped], /*EnableClamp=*/1>;
def VOP_V2F16_F32_F32 : VOPProfile <[v2f16, f32, f32, untyped]>;
def VOP_F32_F16_F16_F16 : VOPProfile <[f32, f16, f16, f16]>;
def VOP_I64_I64_I32 : VOPProfile <[i64, i64, i32, untyped]>;
def VOP_I64_I32_I64 : VOPProfile <[i64, i32, i64, untyped]>;
def VOP_I64_I64_I64 : VOPProfile <[i64, i64, i64, untyped]>;
def VOP_F16_F32_F16_F32 : VOPProfile <[f16, f32, f16, f32]>;
def VOP_F32_F32_F16_F16 : VOPProfile <[f32, f32, f16, f16]>;
def VOP_F32_F32_F32_F32 : VOPProfile <[f32, f32, f32, f32]>;
def VOP_F64_F64_F64_F64 : VOPProfile <[f64, f64, f64, f64]>;
def VOP_I32_I32_I32_I32 : VOPProfile <[i32, i32, i32, i32]>;
def VOP_I64_I32_I32_I64 : VOPProfile <[i64, i32, i32, i64]>;
def VOP_I32_F32_I32_I32 : VOPProfile <[i32, f32, i32, i32]>;
def VOP_I64_I64_I32_I64 : VOPProfile <[i64, i64, i32, i64]>;
def VOP_V4I32_I64_I32_V4I32 : VOPProfile <[v4i32, i64, i32, v4i32]>;
def VOP_F32_V2F16_V2F16_F32 : VOPProfile <[f32, v2f16, v2f16, f32]>;
def VOP_I32_V2I16_V2I16_I32 : VOPProfile <[i32, v2i16, v2i16, i32]>;
def VOP_V4F32_F32_F32_V4F32 : VOPProfile <[v4f32, f32, f32, v4f32]>;
def VOP_V16F32_F32_F32_V16F32 : VOPProfile <[v16f32, f32, f32, v16f32]>;
def VOP_V32F32_F32_F32_V32F32 : VOPProfile <[v32f32, f32, f32, v32f32]>;
def VOP_V4F32_V4F16_V4F16_V4F32 : VOPProfile <[v4f32, v4f16, v4f16, v4f32]>;
def VOP_V16F32_V4F16_V4F16_V16F32 : VOPProfile <[v16f32, v4f16, v4f16, v16f32]>;
def VOP_V32F32_V4F16_V4F16_V32F32 : VOPProfile <[v32f32, v4f16, v4f16, v32f32]>;
def VOP_V4F32_V2I16_V2I16_V4F32 : VOPProfile <[v4f32, v2i16, v2i16, v4f32]>;
def VOP_V16F32_V2I16_V2I16_V16F32 : VOPProfile <[v16f32, v2i16, v2i16, v16f32]>;
def VOP_V32F32_V2I16_V2I16_V32F32 : VOPProfile <[v32f32, v2i16, v2i16, v32f32]>;
def VOP_V4I32_I32_I32_V4I32 : VOPProfile <[v4i32, i32, i32, v4i32]>;
def VOP_V16I32_I32_I32_V16I32 : VOPProfile <[v16i32, i32, i32, v16i32]>;
def VOP_V32I32_I32_I32_V32I32 : VOPProfile <[v32i32, i32, i32, v32i32]>;
def VOP_V4F64_F64_F64_V4F64 : VOPProfile <[v4f64, f64, f64, v4f64]>;
def VOP_V1F64_F64_F64_V1F64 : VOPProfile <[v1f64, f64, f64, v1f64]>;
def VOP_V2F32_V2F32_V2F32_V2F32 : VOPProfile <[v2f32, v2f32, v2f32, v2f32]>;
def VOP_V2F32_V2F32_V2F32 : VOPProfile <[v2f32, v2f32, v2f32, untyped]>;
def VOP_V2I32_V2I32_V2I32 : VOPProfile <[v2i32, v2i32, v2i32, untyped]>;
def VOP_V4F32_V4I16_V4I16_V4F32 : VOPProfile <[v4f32, v4i16, v4i16, v4f32]>;
def VOP_V16F32_V4I16_V4I16_V16F32 : VOPProfile <[v16f32, v4i16, v4i16, v16f32]>;
def VOP_V32F32_V4I16_V4I16_V32F32 : VOPProfile <[v32f32, v4i16, v4i16, v32f32]>;
def VOP_V4I32_I64_I64_V4I32 : VOPProfile <[v4i32, i64, i64, v4i32]>;
def VOP_V16I32_I64_I64_V16I32 : VOPProfile <[v16i32, i64, i64, v16i32]>;
def VOP_V4F32_V2F32_V2F32_V4F32 : VOPProfile <[v4f32, v2f32, v2f32, v4f32]>;
def VOP_V16F32_V2F32_V2F32_V16F32 : VOPProfile <[v16f32, v2f32, v2f32, v16f32]>;
def VOP_V4F32_I64_I64_V4F32 : VOPProfile <[v4f32, i64, i64, v4f32]>;
def VOP_V16F32_I64_I64_V16F32 : VOPProfile <[v16f32, i64, i64, v16f32]>;
def VOP_V4F32_V4F16_V8F16_I32 : VOPProfile <[v4f32, v4f16, v8f16, i32]>;
def VOP_V16F32_V4F16_V8F16_I32 : VOPProfile <[v16f32, v4f16, v8f16, i32]>;
def VOP_V4F32_V4I16_V8I16_I32 : VOPProfile <[v4f32, v4i16, v8i16, i32]>;
def VOP_V16F32_V4I16_V8I16_I32 : VOPProfile <[v16f32, v4i16, v8i16, i32]>;
def VOP_V4I32_V2I32_V4I32_I32 : VOPProfile <[v4i32, v2i32, v4i32, i32]>;
def VOP_V16I32_V2I32_V4I32_I32 : VOPProfile <[v16i32, v2i32, v4i32, i32]>;
def VOP_V4F32_V2I32_V4I32_I32 : VOPProfile <[v4f32, v2i32, v4i32, i32]>;
def VOP_V16F32_V2I32_V4I32_I32 : VOPProfile <[v16f32, v2i32, v4i32, i32]>;
class Commutable_REV <string revOp, bit isOrig> {
string RevOp = revOp;
bit IsOrig = isOrig;
}
class AtomicNoRet <string noRetOp, bit isRet> {
string NoRetOp = noRetOp;
bit IsRet = isRet;
}
//===----------------------------------------------------------------------===//
// Interpolation opcodes
//===----------------------------------------------------------------------===//
class VINTRPDstOperand <RegisterClass rc> : RegisterOperand <rc, "printVINTRPDst">;
class VINTRP_Pseudo <string opName, dag outs, dag ins, list<dag> pattern> :
VINTRPCommon <outs, ins, "", pattern>,
SIMCInstr<opName, SIEncodingFamily.NONE> {
let isPseudo = 1;
let isCodeGenOnly = 1;
}
// FIXME-GFX10: WIP.
class VINTRP_Real_si <bits <2> op, string opName, dag outs, dag ins,
string asm, int encodingFamily> :
VINTRPCommon <outs, ins, asm, []>,
VINTRPe <op>,
SIMCInstr<opName, encodingFamily> {
}
class VINTRP_Real_vi <bits <2> op, string opName, dag outs, dag ins,
string asm> :
VINTRPCommon <outs, ins, asm, []>,
VINTRPe_vi <op>,
SIMCInstr<opName, SIEncodingFamily.VI> {
let AssemblerPredicate = VIAssemblerPredicate;
let DecoderNamespace = "GFX8";
}
// FIXME-GFX10: WIP.
multiclass VINTRP_m <bits <2> op, dag outs, dag ins, string asm,
list<dag> pattern = []> {
def "" : VINTRP_Pseudo <NAME, outs, ins, pattern>;
let AssemblerPredicate = isGFX6GFX7, DecoderNamespace = "GFX6GFX7" in {
def _si : VINTRP_Real_si <op, NAME, outs, ins, asm, SIEncodingFamily.SI>;
} // End AssemblerPredicate = isGFX6GFX7, DecoderNamespace = "GFX6GFX7"
def _vi : VINTRP_Real_vi <op, NAME, outs, ins, asm>;
let AssemblerPredicate = isGFX10Only, DecoderNamespace = "GFX10" in {
def _gfx10 : VINTRP_Real_si<op, NAME, outs, ins, asm, SIEncodingFamily.GFX10>;
} // End AssemblerPredicate = isGFX10Only, DecoderNamespace = "GFX10"
}
//===----------------------------------------------------------------------===//
// Vector instruction mappings
//===----------------------------------------------------------------------===//
// Maps an opcode in e32 form to its e64 equivalent
def getVOPe64 : InstrMapping {
let FilterClass = "VOP";
let RowFields = ["OpName"];
let ColFields = ["Size", "VOP3"];
let KeyCol = ["4", "0"];
let ValueCols = [["8", "1"]];
}
// Maps an opcode in e64 form to its e32 equivalent
def getVOPe32 : InstrMapping {
let FilterClass = "VOP";
let RowFields = ["OpName"];
let ColFields = ["Size", "VOP3"];
let KeyCol = ["8", "1"];
let ValueCols = [["4", "0"]];
}
// Maps ordinary instructions to their SDWA counterparts
def getSDWAOp : InstrMapping {
let FilterClass = "VOP";
let RowFields = ["OpName"];
let ColFields = ["AsmVariantName"];
let KeyCol = ["Default"];
let ValueCols = [["SDWA"]];
}
// Maps SDWA instructions to their ordinary counterparts
def getBasicFromSDWAOp : InstrMapping {
let FilterClass = "VOP";
let RowFields = ["OpName"];
let ColFields = ["AsmVariantName"];
let KeyCol = ["SDWA"];
let ValueCols = [["Default"]];
}
// Maps ordinary instructions to their DPP counterparts
def getDPPOp32 : InstrMapping {
let FilterClass = "VOP";
let RowFields = ["OpName"];
let ColFields = ["AsmVariantName"];
let KeyCol = ["Default"];
let ValueCols = [["DPP"]];
}
def getDPPOp64 : InstrMapping {
let FilterClass = "VOP";
let RowFields = ["OpName"];
let ColFields = ["AsmVariantName"];
let KeyCol = ["VOP3"];
let ValueCols = [["VOP3_DPP"]];
}
// Maps an commuted opcode to its original version
def getCommuteOrig : InstrMapping {
let FilterClass = "Commutable_REV";
let RowFields = ["RevOp"];
let ColFields = ["IsOrig"];
let KeyCol = ["0"];
let ValueCols = [["1"]];
}
// Maps an original opcode to its commuted version
def getCommuteRev : InstrMapping {
let FilterClass = "Commutable_REV";
let RowFields = ["RevOp"];
let ColFields = ["IsOrig"];
let KeyCol = ["1"];
let ValueCols = [["0"]];
}
def getMCOpcodeGen : InstrMapping {
let FilterClass = "SIMCInstr";
let RowFields = ["PseudoInstr"];
let ColFields = ["Subtarget"];
let KeyCol = [!cast<string>(SIEncodingFamily.NONE)];
// These columns must be kept in sync with the SIEncodingFamily enumeration.
let ValueCols = [[!cast<string>(SIEncodingFamily.SI)],
[!cast<string>(SIEncodingFamily.VI)],
[!cast<string>(SIEncodingFamily.SDWA)],
[!cast<string>(SIEncodingFamily.SDWA9)],
// GFX80 encoding is added to work around a multiple matching
// issue for buffer instructions with unpacked d16 data. This
// does not actually change the encoding, and thus may be
// removed later.
[!cast<string>(SIEncodingFamily.GFX80)],
[!cast<string>(SIEncodingFamily.GFX9)],
[!cast<string>(SIEncodingFamily.GFX10)],
[!cast<string>(SIEncodingFamily.SDWA10)],
[!cast<string>(SIEncodingFamily.GFX90A)],
[!cast<string>(SIEncodingFamily.GFX940)],
[!cast<string>(SIEncodingFamily.GFX11)],
[!cast<string>(SIEncodingFamily.GFX12)]];
}
// Get equivalent SOPK instruction.
def getSOPKOp : InstrMapping {
let FilterClass = "SOPKInstTable";
let RowFields = ["BaseCmpOp"];
let ColFields = ["IsSOPK"];
let KeyCol = ["0"];
let ValueCols = [["1"]];
}
def getAddr64Inst : InstrMapping {
let FilterClass = "MUBUFAddr64Table";
let RowFields = ["OpName"];
let ColFields = ["IsAddr64"];
let KeyCol = ["0"];
let ValueCols = [["1"]];
}
def getIfAddr64Inst : InstrMapping {
let FilterClass = "MUBUFAddr64Table";
let RowFields = ["OpName"];
let ColFields = ["IsAddr64"];
let KeyCol = ["1"];
let ValueCols = [["1"]];
}
// Maps an atomic opcode to its returnless version.
def getAtomicNoRetOp : InstrMapping {
let FilterClass = "AtomicNoRet";
let RowFields = ["NoRetOp"];
let ColFields = ["IsRet"];
let KeyCol = ["1"];
let ValueCols = [["0"]];
}
// Maps a GLOBAL to its SADDR form.
def getGlobalSaddrOp : InstrMapping {
let FilterClass = "GlobalSaddrTable";
let RowFields = ["SaddrOp"];
let ColFields = ["IsSaddr"];
let KeyCol = ["0"];
let ValueCols = [["1"]];
}
// Maps a GLOBAL SADDR to its VADDR form.
def getGlobalVaddrOp : InstrMapping {
let FilterClass = "GlobalSaddrTable";
let RowFields = ["SaddrOp"];
let ColFields = ["IsSaddr"];
let KeyCol = ["1"];
let ValueCols = [["0"]];
}
// Maps a v_cmpx opcode with sdst to opcode without sdst.
def getVCMPXNoSDstOp : InstrMapping {
let FilterClass = "VCMPXNoSDstTable";
let RowFields = ["NoSDstOp"];
let ColFields = ["HasSDst"];
let KeyCol = ["1"];
let ValueCols = [["0"]];
}
// Maps a SOPP to a SOPP with S_NOP
def getSOPPWithRelaxation : InstrMapping {
let FilterClass = "SOPPRelaxTable";
let RowFields = ["KeyName"];
let ColFields = ["IsRelaxed"];
let KeyCol = ["0"];
let ValueCols = [["1"]];
}
// Maps flat scratch opcodes by addressing modes
def getFlatScratchInstSTfromSS : InstrMapping {
let FilterClass = "FlatScratchInst";
let RowFields = ["SVOp"];
let ColFields = ["Mode"];
let KeyCol = ["SS"];
let ValueCols = [["ST"]];
}
def getFlatScratchInstSSfromSV : InstrMapping {
let FilterClass = "FlatScratchInst";
let RowFields = ["SVOp"];
let ColFields = ["Mode"];
let KeyCol = ["SV"];
let ValueCols = [["SS"]];
}
def getFlatScratchInstSVfromSVS : InstrMapping {
let FilterClass = "FlatScratchInst";
let RowFields = ["SVOp"];
let ColFields = ["Mode"];
let KeyCol = ["SVS"];
let ValueCols = [["SV"]];
}
def getFlatScratchInstSVfromSS : InstrMapping {
let FilterClass = "FlatScratchInst";
let RowFields = ["SVOp"];
let ColFields = ["Mode"];
let KeyCol = ["SS"];
let ValueCols = [["SV"]];
}
def getMFMAEarlyClobberOp : InstrMapping {
let FilterClass = "MFMATable";
let RowFields = ["FMAOp"];
let ColFields = ["IsMac"];
let KeyCol = ["1"];
let ValueCols = [["0"]];
}
// Maps an v_cmp instruction to its v_cmpx equivalent.
def getVCMPXOpFromVCMP : InstrMapping {
let FilterClass = "VCMPVCMPXTable";
let RowFields = ["VCMPOp"];
let ColFields = ["IsVCMPX"];
let KeyCol = ["0"];
let ValueCols = [["1"]];
}
def VOPDComponentTable : GenericTable {
let FilterClass = "VOPD_Component";
let CppTypeName = "VOPDComponentInfo";
let Fields = ["BaseVOP", "VOPDOp", "CanBeVOPDX"];
let PrimaryKey = ["BaseVOP"];
let PrimaryKeyName = "getVOPDComponentHelper";
}
def getVOPDBaseFromComponent : SearchIndex {
let Table = VOPDComponentTable;
let Key = ["VOPDOp"];
}
def VOPDPairs : GenericTable {
let FilterClass = "VOPD_Base";
let CppTypeName = "VOPDInfo";
let Fields = ["Opcode", "OpX", "OpY", "SubTgt"];
let PrimaryKey = ["Opcode"];
let PrimaryKeyName = "getVOPDOpcodeHelper";
}
def getVOPDInfoFromComponentOpcodes : SearchIndex {
let Table = VOPDPairs;
let Key = ["OpX", "OpY", "SubTgt"];
}
include "SIInstructions.td"
include "DSInstructions.td"
include "MIMGInstructions.td"
|