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
|
// SPDX-License-Identifier: GPL-2.0 OR MIT
/*
* Copyright 2015-2022 Advanced Micro Devices, Inc.
*
* Permission is hereby granted, free of charge, to any person obtaining a
* copy of this software and associated documentation files (the "Software"),
* to deal in the Software without restriction, including without limitation
* the rights to use, copy, modify, merge, publish, distribute, sublicense,
* and/or sell copies of the Software, and to permit persons to whom the
* Software is furnished to do so, subject to the following conditions:
*
* The above copyright notice and this permission notice shall be included in
* all copies or substantial portions of the Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
* THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
* OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
* ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
* OTHER DEALINGS IN THE SOFTWARE.
*/
#include <linux/pci.h>
#include <linux/acpi.h>
#include "kfd_crat.h"
#include "kfd_priv.h"
#include "kfd_topology.h"
#include "amdgpu.h"
#include "amdgpu_amdkfd.h"
#include "amdgpu_xgmi.h"
/* GPU Processor ID base for dGPUs for which VCRAT needs to be created.
* GPU processor ID are expressed with Bit[31]=1.
* The base is set to 0x8000_0000 + 0x1000 to avoid collision with GPU IDs
* used in the CRAT.
*/
static uint32_t gpu_processor_id_low = 0x80001000;
/* Return the next available gpu_processor_id and increment it for next GPU
* @total_cu_count - Total CUs present in the GPU including ones
* masked off
*/
static inline unsigned int get_and_inc_gpu_processor_id(
unsigned int total_cu_count)
{
int current_id = gpu_processor_id_low;
gpu_processor_id_low += total_cu_count;
return current_id;
}
static struct kfd_gpu_cache_info kaveri_cache_info[] = {
{
/* TCP L1 Cache per CU */
.cache_size = 16,
.cache_level = 1,
.cache_line_size = 64,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 1,
},
{
/* Scalar L1 Instruction Cache (in SQC module) per bank */
.cache_size = 16,
.cache_level = 1,
.cache_line_size = 64,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_INST_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 2,
},
{
/* Scalar L1 Data Cache (in SQC module) per bank */
.cache_size = 8,
.cache_level = 1,
.cache_line_size = 64,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 2,
},
/* TODO: Add L2 Cache information */
};
static struct kfd_gpu_cache_info carrizo_cache_info[] = {
{
/* TCP L1 Cache per CU */
.cache_size = 16,
.cache_level = 1,
.cache_line_size = 64,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 1,
},
{
/* Scalar L1 Instruction Cache (in SQC module) per bank */
.cache_size = 32,
.cache_level = 1,
.cache_line_size = 64,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_INST_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 4,
},
{
/* Scalar L1 Data Cache (in SQC module) per bank. */
.cache_size = 16,
.cache_level = 1,
.cache_line_size = 64,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 4,
},
/* TODO: Add L2 Cache information */
};
#define hawaii_cache_info kaveri_cache_info
#define tonga_cache_info carrizo_cache_info
#define fiji_cache_info carrizo_cache_info
#define polaris10_cache_info carrizo_cache_info
#define polaris11_cache_info carrizo_cache_info
#define polaris12_cache_info carrizo_cache_info
#define vegam_cache_info carrizo_cache_info
/* NOTE: L1 cache information has been updated and L2/L3
* cache information has been added for Vega10 and
* newer ASICs. The unit for cache_size is KiB.
* In future, check & update cache details
* for every new ASIC is required.
*/
static struct kfd_gpu_cache_info vega10_cache_info[] = {
{
/* TCP L1 Cache per CU */
.cache_size = 16,
.cache_level = 1,
.cache_line_size = 64,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 1,
},
{
/* Scalar L1 Instruction Cache per SQC */
.cache_size = 32,
.cache_level = 1,
.cache_line_size = 64,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_INST_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 3,
},
{
/* Scalar L1 Data Cache per SQC */
.cache_size = 16,
.cache_level = 1,
.cache_line_size = 64,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 3,
},
{
/* L2 Data Cache per GPU (Total Tex Cache) */
.cache_size = 4096,
.cache_level = 2,
.cache_line_size = 64,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 16,
},
};
static struct kfd_gpu_cache_info raven_cache_info[] = {
{
/* TCP L1 Cache per CU */
.cache_size = 16,
.cache_level = 1,
.cache_line_size = 64,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 1,
},
{
/* Scalar L1 Instruction Cache per SQC */
.cache_size = 32,
.cache_level = 1,
.cache_line_size = 64,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_INST_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 3,
},
{
/* Scalar L1 Data Cache per SQC */
.cache_size = 16,
.cache_level = 1,
.cache_line_size = 64,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 3,
},
{
/* L2 Data Cache per GPU (Total Tex Cache) */
.cache_size = 1024,
.cache_level = 2,
.cache_line_size = 64,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 11,
},
};
static struct kfd_gpu_cache_info renoir_cache_info[] = {
{
/* TCP L1 Cache per CU */
.cache_size = 16,
.cache_level = 1,
.cache_line_size = 64,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 1,
},
{
/* Scalar L1 Instruction Cache per SQC */
.cache_size = 32,
.cache_level = 1,
.cache_line_size = 64,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_INST_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 3,
},
{
/* Scalar L1 Data Cache per SQC */
.cache_size = 16,
.cache_level = 1,
.cache_line_size = 64,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 3,
},
{
/* L2 Data Cache per GPU (Total Tex Cache) */
.cache_size = 1024,
.cache_level = 2,
.cache_line_size = 64,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 8,
},
};
static struct kfd_gpu_cache_info vega12_cache_info[] = {
{
/* TCP L1 Cache per CU */
.cache_size = 16,
.cache_level = 1,
.cache_line_size = 64,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 1,
},
{
/* Scalar L1 Instruction Cache per SQC */
.cache_size = 32,
.cache_level = 1,
.cache_line_size = 64,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_INST_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 3,
},
{
/* Scalar L1 Data Cache per SQC */
.cache_size = 16,
.cache_level = 1,
.cache_line_size = 64,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 3,
},
{
/* L2 Data Cache per GPU (Total Tex Cache) */
.cache_size = 2048,
.cache_level = 2,
.cache_line_size = 64,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 5,
},
};
static struct kfd_gpu_cache_info vega20_cache_info[] = {
{
/* TCP L1 Cache per CU */
.cache_size = 16,
.cache_level = 1,
.cache_line_size = 64,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 1,
},
{
/* Scalar L1 Instruction Cache per SQC */
.cache_size = 32,
.cache_level = 1,
.cache_line_size = 64,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_INST_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 3,
},
{
/* Scalar L1 Data Cache per SQC */
.cache_size = 16,
.cache_level = 1,
.cache_line_size = 64,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 3,
},
{
/* L2 Data Cache per GPU (Total Tex Cache) */
.cache_size = 8192,
.cache_level = 2,
.cache_line_size = 64,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 16,
},
};
static struct kfd_gpu_cache_info aldebaran_cache_info[] = {
{
/* TCP L1 Cache per CU */
.cache_size = 16,
.cache_level = 1,
.cache_line_size = 64,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 1,
},
{
/* Scalar L1 Instruction Cache per SQC */
.cache_size = 32,
.cache_level = 1,
.cache_line_size = 64,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_INST_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 2,
},
{
/* Scalar L1 Data Cache per SQC */
.cache_size = 16,
.cache_level = 1,
.cache_line_size = 64,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 2,
},
{
/* L2 Data Cache per GPU (Total Tex Cache) */
.cache_size = 8192,
.cache_level = 2,
.cache_line_size = 128,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 14,
},
};
static struct kfd_gpu_cache_info navi10_cache_info[] = {
{
/* TCP L1 Cache per CU */
.cache_size = 16,
.cache_level = 1,
.cache_line_size = 128,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 1,
},
{
/* Scalar L1 Instruction Cache per SQC */
.cache_size = 32,
.cache_level = 1,
.cache_line_size = 64,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_INST_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 2,
},
{
/* Scalar L1 Data Cache per SQC */
.cache_size = 16,
.cache_level = 1,
.cache_line_size = 64,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 2,
},
{
/* GL1 Data Cache per SA */
.cache_size = 128,
.cache_level = 1,
.cache_line_size = 128,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 10,
},
{
/* L2 Data Cache per GPU (Total Tex Cache) */
.cache_size = 4096,
.cache_level = 2,
.cache_line_size = 128,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 10,
},
};
static struct kfd_gpu_cache_info vangogh_cache_info[] = {
{
/* TCP L1 Cache per CU */
.cache_size = 16,
.cache_level = 1,
.cache_line_size = 128,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 1,
},
{
/* Scalar L1 Instruction Cache per SQC */
.cache_size = 32,
.cache_level = 1,
.cache_line_size = 64,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_INST_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 2,
},
{
/* Scalar L1 Data Cache per SQC */
.cache_size = 16,
.cache_level = 1,
.cache_line_size = 64,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 2,
},
{
/* GL1 Data Cache per SA */
.cache_size = 128,
.cache_level = 1,
.cache_line_size = 128,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 8,
},
{
/* L2 Data Cache per GPU (Total Tex Cache) */
.cache_size = 1024,
.cache_level = 2,
.cache_line_size = 128,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 8,
},
};
static struct kfd_gpu_cache_info navi14_cache_info[] = {
{
/* TCP L1 Cache per CU */
.cache_size = 16,
.cache_level = 1,
.cache_line_size = 128,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 1,
},
{
/* Scalar L1 Instruction Cache per SQC */
.cache_size = 32,
.cache_level = 1,
.cache_line_size = 64,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_INST_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 2,
},
{
/* Scalar L1 Data Cache per SQC */
.cache_size = 16,
.cache_level = 1,
.cache_line_size = 64,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 2,
},
{
/* GL1 Data Cache per SA */
.cache_size = 128,
.cache_level = 1,
.cache_line_size = 128,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 12,
},
{
/* L2 Data Cache per GPU (Total Tex Cache) */
.cache_size = 2048,
.cache_level = 2,
.cache_line_size = 128,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 12,
},
};
static struct kfd_gpu_cache_info sienna_cichlid_cache_info[] = {
{
/* TCP L1 Cache per CU */
.cache_size = 16,
.cache_level = 1,
.cache_line_size = 128,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 1,
},
{
/* Scalar L1 Instruction Cache per SQC */
.cache_size = 32,
.cache_level = 1,
.cache_line_size = 64,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_INST_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 2,
},
{
/* Scalar L1 Data Cache per SQC */
.cache_size = 16,
.cache_level = 1,
.cache_line_size = 64,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 2,
},
{
/* GL1 Data Cache per SA */
.cache_size = 128,
.cache_level = 1,
.cache_line_size = 128,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 10,
},
{
/* L2 Data Cache per GPU (Total Tex Cache) */
.cache_size = 4096,
.cache_level = 2,
.cache_line_size = 128,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 10,
},
{
/* L3 Data Cache per GPU */
.cache_size = 128*1024,
.cache_level = 3,
.cache_line_size = 64,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 10,
},
};
static struct kfd_gpu_cache_info navy_flounder_cache_info[] = {
{
/* TCP L1 Cache per CU */
.cache_size = 16,
.cache_level = 1,
.cache_line_size = 128,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 1,
},
{
/* Scalar L1 Instruction Cache per SQC */
.cache_size = 32,
.cache_level = 1,
.cache_line_size = 64,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_INST_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 2,
},
{
/* Scalar L1 Data Cache per SQC */
.cache_size = 16,
.cache_level = 1,
.cache_line_size = 64,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 2,
},
{
/* GL1 Data Cache per SA */
.cache_size = 128,
.cache_level = 1,
.cache_line_size = 128,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 10,
},
{
/* L2 Data Cache per GPU (Total Tex Cache) */
.cache_size = 3072,
.cache_level = 2,
.cache_line_size = 128,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 10,
},
{
/* L3 Data Cache per GPU */
.cache_size = 96*1024,
.cache_level = 3,
.cache_line_size = 64,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 10,
},
};
static struct kfd_gpu_cache_info dimgrey_cavefish_cache_info[] = {
{
/* TCP L1 Cache per CU */
.cache_size = 16,
.cache_level = 1,
.cache_line_size = 128,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 1,
},
{
/* Scalar L1 Instruction Cache per SQC */
.cache_size = 32,
.cache_level = 1,
.cache_line_size = 64,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_INST_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 2,
},
{
/* Scalar L1 Data Cache per SQC */
.cache_size = 16,
.cache_level = 1,
.cache_line_size = 64,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 2,
},
{
/* GL1 Data Cache per SA */
.cache_size = 128,
.cache_level = 1,
.cache_line_size = 128,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 8,
},
{
/* L2 Data Cache per GPU (Total Tex Cache) */
.cache_size = 2048,
.cache_level = 2,
.cache_line_size = 128,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 8,
},
{
/* L3 Data Cache per GPU */
.cache_size = 32*1024,
.cache_level = 3,
.cache_line_size = 64,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 8,
},
};
static struct kfd_gpu_cache_info beige_goby_cache_info[] = {
{
/* TCP L1 Cache per CU */
.cache_size = 16,
.cache_level = 1,
.cache_line_size = 128,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 1,
},
{
/* Scalar L1 Instruction Cache per SQC */
.cache_size = 32,
.cache_level = 1,
.cache_line_size = 64,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_INST_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 2,
},
{
/* Scalar L1 Data Cache per SQC */
.cache_size = 16,
.cache_level = 1,
.cache_line_size = 64,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 2,
},
{
/* GL1 Data Cache per SA */
.cache_size = 128,
.cache_level = 1,
.cache_line_size = 128,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 8,
},
{
/* L2 Data Cache per GPU (Total Tex Cache) */
.cache_size = 1024,
.cache_level = 2,
.cache_line_size = 128,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 8,
},
{
/* L3 Data Cache per GPU */
.cache_size = 16*1024,
.cache_level = 3,
.cache_line_size = 64,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 8,
},
};
static struct kfd_gpu_cache_info yellow_carp_cache_info[] = {
{
/* TCP L1 Cache per CU */
.cache_size = 16,
.cache_level = 1,
.cache_line_size = 128,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 1,
},
{
/* Scalar L1 Instruction Cache per SQC */
.cache_size = 32,
.cache_level = 1,
.cache_line_size = 64,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_INST_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 2,
},
{
/* Scalar L1 Data Cache per SQC */
.cache_size = 16,
.cache_level = 1,
.cache_line_size = 64,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 2,
},
{
/* GL1 Data Cache per SA */
.cache_size = 128,
.cache_level = 1,
.cache_line_size = 128,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 6,
},
{
/* L2 Data Cache per GPU (Total Tex Cache) */
.cache_size = 2048,
.cache_level = 2,
.cache_line_size = 128,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 6,
},
};
static struct kfd_gpu_cache_info gfx1037_cache_info[] = {
{
/* TCP L1 Cache per CU */
.cache_size = 16,
.cache_level = 1,
.cache_line_size = 128,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 1,
},
{
/* Scalar L1 Instruction Cache per SQC */
.cache_size = 32,
.cache_level = 1,
.cache_line_size = 64,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_INST_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 2,
},
{
/* Scalar L1 Data Cache per SQC */
.cache_size = 16,
.cache_level = 1,
.cache_line_size = 64,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 2,
},
{
/* GL1 Data Cache per SA */
.cache_size = 128,
.cache_level = 1,
.cache_line_size = 128,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 2,
},
{
/* L2 Data Cache per GPU (Total Tex Cache) */
.cache_size = 256,
.cache_level = 2,
.cache_line_size = 128,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 2,
},
};
static struct kfd_gpu_cache_info gc_10_3_6_cache_info[] = {
{
/* TCP L1 Cache per CU */
.cache_size = 16,
.cache_level = 1,
.cache_line_size = 128,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 1,
},
{
/* Scalar L1 Instruction Cache per SQC */
.cache_size = 32,
.cache_level = 1,
.cache_line_size = 64,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_INST_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 2,
},
{
/* Scalar L1 Data Cache per SQC */
.cache_size = 16,
.cache_level = 1,
.cache_line_size = 64,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 2,
},
{
/* GL1 Data Cache per SA */
.cache_size = 128,
.cache_level = 1,
.cache_line_size = 128,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 2,
},
{
/* L2 Data Cache per GPU (Total Tex Cache) */
.cache_size = 256,
.cache_level = 2,
.cache_line_size = 128,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 2,
},
};
static struct kfd_gpu_cache_info dummy_cache_info[] = {
{
/* TCP L1 Cache per CU */
.cache_size = 16,
.cache_level = 1,
.cache_line_size = 64,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 1,
},
{
/* Scalar L1 Instruction Cache per SQC */
.cache_size = 32,
.cache_level = 1,
.cache_line_size = 64,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_INST_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 2,
},
{
/* Scalar L1 Data Cache per SQC */
.cache_size = 16,
.cache_level = 1,
.cache_line_size = 64,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 2,
},
{
/* GL1 Data Cache per SA */
.cache_size = 128,
.cache_level = 1,
.cache_line_size = 64,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 6,
},
{
/* L2 Data Cache per GPU (Total Tex Cache) */
.cache_size = 2048,
.cache_level = 2,
.cache_line_size = 64,
.flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE),
.num_cu_shared = 6,
},
};
static void kfd_populated_cu_info_cpu(struct kfd_topology_device *dev,
struct crat_subtype_computeunit *cu)
{
dev->node_props.cpu_cores_count = cu->num_cpu_cores;
dev->node_props.cpu_core_id_base = cu->processor_id_low;
if (cu->hsa_capability & CRAT_CU_FLAGS_IOMMU_PRESENT)
dev->node_props.capability |= HSA_CAP_ATS_PRESENT;
pr_debug("CU CPU: cores=%d id_base=%d\n", cu->num_cpu_cores,
cu->processor_id_low);
}
static void kfd_populated_cu_info_gpu(struct kfd_topology_device *dev,
struct crat_subtype_computeunit *cu)
{
dev->node_props.simd_id_base = cu->processor_id_low;
dev->node_props.simd_count = cu->num_simd_cores;
dev->node_props.lds_size_in_kb = cu->lds_size_in_kb;
dev->node_props.max_waves_per_simd = cu->max_waves_simd;
dev->node_props.wave_front_size = cu->wave_front_size;
dev->node_props.array_count = cu->array_count;
dev->node_props.cu_per_simd_array = cu->num_cu_per_array;
dev->node_props.simd_per_cu = cu->num_simd_per_cu;
dev->node_props.max_slots_scratch_cu = cu->max_slots_scatch_cu;
if (cu->hsa_capability & CRAT_CU_FLAGS_HOT_PLUGGABLE)
dev->node_props.capability |= HSA_CAP_HOT_PLUGGABLE;
pr_debug("CU GPU: id_base=%d\n", cu->processor_id_low);
}
/* kfd_parse_subtype_cu - parse compute unit subtypes and attach it to correct
* topology device present in the device_list
*/
static int kfd_parse_subtype_cu(struct crat_subtype_computeunit *cu,
struct list_head *device_list)
{
struct kfd_topology_device *dev;
pr_debug("Found CU entry in CRAT table with proximity_domain=%d caps=%x\n",
cu->proximity_domain, cu->hsa_capability);
list_for_each_entry(dev, device_list, list) {
if (cu->proximity_domain == dev->proximity_domain) {
if (cu->flags & CRAT_CU_FLAGS_CPU_PRESENT)
kfd_populated_cu_info_cpu(dev, cu);
if (cu->flags & CRAT_CU_FLAGS_GPU_PRESENT)
kfd_populated_cu_info_gpu(dev, cu);
break;
}
}
return 0;
}
static struct kfd_mem_properties *
find_subtype_mem(uint32_t heap_type, uint32_t flags, uint32_t width,
struct kfd_topology_device *dev)
{
struct kfd_mem_properties *props;
list_for_each_entry(props, &dev->mem_props, list) {
if (props->heap_type == heap_type
&& props->flags == flags
&& props->width == width)
return props;
}
return NULL;
}
/* kfd_parse_subtype_mem - parse memory subtypes and attach it to correct
* topology device present in the device_list
*/
static int kfd_parse_subtype_mem(struct crat_subtype_memory *mem,
struct list_head *device_list)
{
struct kfd_mem_properties *props;
struct kfd_topology_device *dev;
uint32_t heap_type;
uint64_t size_in_bytes;
uint32_t flags = 0;
uint32_t width;
pr_debug("Found memory entry in CRAT table with proximity_domain=%d\n",
mem->proximity_domain);
list_for_each_entry(dev, device_list, list) {
if (mem->proximity_domain == dev->proximity_domain) {
/* We're on GPU node */
if (dev->node_props.cpu_cores_count == 0) {
/* APU */
if (mem->visibility_type == 0)
heap_type =
HSA_MEM_HEAP_TYPE_FB_PRIVATE;
/* dGPU */
else
heap_type = mem->visibility_type;
} else
heap_type = HSA_MEM_HEAP_TYPE_SYSTEM;
if (mem->flags & CRAT_MEM_FLAGS_HOT_PLUGGABLE)
flags |= HSA_MEM_FLAGS_HOT_PLUGGABLE;
if (mem->flags & CRAT_MEM_FLAGS_NON_VOLATILE)
flags |= HSA_MEM_FLAGS_NON_VOLATILE;
size_in_bytes =
((uint64_t)mem->length_high << 32) +
mem->length_low;
width = mem->width;
/* Multiple banks of the same type are aggregated into
* one. User mode doesn't care about multiple physical
* memory segments. It's managed as a single virtual
* heap for user mode.
*/
props = find_subtype_mem(heap_type, flags, width, dev);
if (props) {
props->size_in_bytes += size_in_bytes;
break;
}
props = kfd_alloc_struct(props);
if (!props)
return -ENOMEM;
props->heap_type = heap_type;
props->flags = flags;
props->size_in_bytes = size_in_bytes;
props->width = width;
dev->node_props.mem_banks_count++;
list_add_tail(&props->list, &dev->mem_props);
break;
}
}
return 0;
}
/* kfd_parse_subtype_cache - parse cache subtypes and attach it to correct
* topology device present in the device_list
*/
static int kfd_parse_subtype_cache(struct crat_subtype_cache *cache,
struct list_head *device_list)
{
struct kfd_cache_properties *props;
struct kfd_topology_device *dev;
uint32_t id;
uint32_t total_num_of_cu;
id = cache->processor_id_low;
pr_debug("Found cache entry in CRAT table with processor_id=%d\n", id);
list_for_each_entry(dev, device_list, list) {
total_num_of_cu = (dev->node_props.array_count *
dev->node_props.cu_per_simd_array);
/* Cache infomration in CRAT doesn't have proximity_domain
* information as it is associated with a CPU core or GPU
* Compute Unit. So map the cache using CPU core Id or SIMD
* (GPU) ID.
* TODO: This works because currently we can safely assume that
* Compute Units are parsed before caches are parsed. In
* future, remove this dependency
*/
if ((id >= dev->node_props.cpu_core_id_base &&
id <= dev->node_props.cpu_core_id_base +
dev->node_props.cpu_cores_count) ||
(id >= dev->node_props.simd_id_base &&
id < dev->node_props.simd_id_base +
total_num_of_cu)) {
props = kfd_alloc_struct(props);
if (!props)
return -ENOMEM;
props->processor_id_low = id;
props->cache_level = cache->cache_level;
props->cache_size = cache->cache_size;
props->cacheline_size = cache->cache_line_size;
props->cachelines_per_tag = cache->lines_per_tag;
props->cache_assoc = cache->associativity;
props->cache_latency = cache->cache_latency;
memcpy(props->sibling_map, cache->sibling_map,
CRAT_SIBLINGMAP_SIZE);
/* set the sibling_map_size as 32 for CRAT from ACPI */
props->sibling_map_size = CRAT_SIBLINGMAP_SIZE;
if (cache->flags & CRAT_CACHE_FLAGS_DATA_CACHE)
props->cache_type |= HSA_CACHE_TYPE_DATA;
if (cache->flags & CRAT_CACHE_FLAGS_INST_CACHE)
props->cache_type |= HSA_CACHE_TYPE_INSTRUCTION;
if (cache->flags & CRAT_CACHE_FLAGS_CPU_CACHE)
props->cache_type |= HSA_CACHE_TYPE_CPU;
if (cache->flags & CRAT_CACHE_FLAGS_SIMD_CACHE)
props->cache_type |= HSA_CACHE_TYPE_HSACU;
dev->node_props.caches_count++;
list_add_tail(&props->list, &dev->cache_props);
break;
}
}
return 0;
}
/* kfd_parse_subtype_iolink - parse iolink subtypes and attach it to correct
* topology device present in the device_list
*/
static int kfd_parse_subtype_iolink(struct crat_subtype_iolink *iolink,
struct list_head *device_list)
{
struct kfd_iolink_properties *props = NULL, *props2;
struct kfd_topology_device *dev, *to_dev;
uint32_t id_from;
uint32_t id_to;
id_from = iolink->proximity_domain_from;
id_to = iolink->proximity_domain_to;
pr_debug("Found IO link entry in CRAT table with id_from=%d, id_to %d\n",
id_from, id_to);
list_for_each_entry(dev, device_list, list) {
if (id_from == dev->proximity_domain) {
props = kfd_alloc_struct(props);
if (!props)
return -ENOMEM;
props->node_from = id_from;
props->node_to = id_to;
props->ver_maj = iolink->version_major;
props->ver_min = iolink->version_minor;
props->iolink_type = iolink->io_interface_type;
if (props->iolink_type == CRAT_IOLINK_TYPE_PCIEXPRESS)
props->weight = 20;
else if (props->iolink_type == CRAT_IOLINK_TYPE_XGMI)
props->weight = iolink->weight_xgmi;
else
props->weight = node_distance(id_from, id_to);
props->min_latency = iolink->minimum_latency;
props->max_latency = iolink->maximum_latency;
props->min_bandwidth = iolink->minimum_bandwidth_mbs;
props->max_bandwidth = iolink->maximum_bandwidth_mbs;
props->rec_transfer_size =
iolink->recommended_transfer_size;
dev->node_props.io_links_count++;
list_add_tail(&props->list, &dev->io_link_props);
break;
}
}
/* CPU topology is created before GPUs are detected, so CPU->GPU
* links are not built at that time. If a PCIe type is discovered, it
* means a GPU is detected and we are adding GPU->CPU to the topology.
* At this time, also add the corresponded CPU->GPU link if GPU
* is large bar.
* For xGMI, we only added the link with one direction in the crat
* table, add corresponded reversed direction link now.
*/
if (props && (iolink->flags & CRAT_IOLINK_FLAGS_BI_DIRECTIONAL)) {
to_dev = kfd_topology_device_by_proximity_domain_no_lock(id_to);
if (!to_dev)
return -ENODEV;
/* same everything but the other direction */
props2 = kmemdup(props, sizeof(*props2), GFP_KERNEL);
if (!props2)
return -ENOMEM;
props2->node_from = id_to;
props2->node_to = id_from;
props2->kobj = NULL;
to_dev->node_props.io_links_count++;
list_add_tail(&props2->list, &to_dev->io_link_props);
}
return 0;
}
/* kfd_parse_subtype - parse subtypes and attach it to correct topology device
* present in the device_list
* @sub_type_hdr - subtype section of crat_image
* @device_list - list of topology devices present in this crat_image
*/
static int kfd_parse_subtype(struct crat_subtype_generic *sub_type_hdr,
struct list_head *device_list)
{
struct crat_subtype_computeunit *cu;
struct crat_subtype_memory *mem;
struct crat_subtype_cache *cache;
struct crat_subtype_iolink *iolink;
int ret = 0;
switch (sub_type_hdr->type) {
case CRAT_SUBTYPE_COMPUTEUNIT_AFFINITY:
cu = (struct crat_subtype_computeunit *)sub_type_hdr;
ret = kfd_parse_subtype_cu(cu, device_list);
break;
case CRAT_SUBTYPE_MEMORY_AFFINITY:
mem = (struct crat_subtype_memory *)sub_type_hdr;
ret = kfd_parse_subtype_mem(mem, device_list);
break;
case CRAT_SUBTYPE_CACHE_AFFINITY:
cache = (struct crat_subtype_cache *)sub_type_hdr;
ret = kfd_parse_subtype_cache(cache, device_list);
break;
case CRAT_SUBTYPE_TLB_AFFINITY:
/*
* For now, nothing to do here
*/
pr_debug("Found TLB entry in CRAT table (not processing)\n");
break;
case CRAT_SUBTYPE_CCOMPUTE_AFFINITY:
/*
* For now, nothing to do here
*/
pr_debug("Found CCOMPUTE entry in CRAT table (not processing)\n");
break;
case CRAT_SUBTYPE_IOLINK_AFFINITY:
iolink = (struct crat_subtype_iolink *)sub_type_hdr;
ret = kfd_parse_subtype_iolink(iolink, device_list);
break;
default:
pr_warn("Unknown subtype %d in CRAT\n",
sub_type_hdr->type);
}
return ret;
}
/* kfd_parse_crat_table - parse CRAT table. For each node present in CRAT
* create a kfd_topology_device and add in to device_list. Also parse
* CRAT subtypes and attach it to appropriate kfd_topology_device
* @crat_image - input image containing CRAT
* @device_list - [OUT] list of kfd_topology_device generated after
* parsing crat_image
* @proximity_domain - Proximity domain of the first device in the table
*
* Return - 0 if successful else -ve value
*/
int kfd_parse_crat_table(void *crat_image, struct list_head *device_list,
uint32_t proximity_domain)
{
struct kfd_topology_device *top_dev = NULL;
struct crat_subtype_generic *sub_type_hdr;
uint16_t node_id;
int ret = 0;
struct crat_header *crat_table = (struct crat_header *)crat_image;
uint16_t num_nodes;
uint32_t image_len;
if (!crat_image)
return -EINVAL;
if (!list_empty(device_list)) {
pr_warn("Error device list should be empty\n");
return -EINVAL;
}
num_nodes = crat_table->num_domains;
image_len = crat_table->length;
pr_debug("Parsing CRAT table with %d nodes\n", num_nodes);
for (node_id = 0; node_id < num_nodes; node_id++) {
top_dev = kfd_create_topology_device(device_list);
if (!top_dev)
break;
top_dev->proximity_domain = proximity_domain++;
}
if (!top_dev) {
ret = -ENOMEM;
goto err;
}
memcpy(top_dev->oem_id, crat_table->oem_id, CRAT_OEMID_LENGTH);
memcpy(top_dev->oem_table_id, crat_table->oem_table_id,
CRAT_OEMTABLEID_LENGTH);
top_dev->oem_revision = crat_table->oem_revision;
sub_type_hdr = (struct crat_subtype_generic *)(crat_table+1);
while ((char *)sub_type_hdr + sizeof(struct crat_subtype_generic) <
((char *)crat_image) + image_len) {
if (sub_type_hdr->flags & CRAT_SUBTYPE_FLAGS_ENABLED) {
ret = kfd_parse_subtype(sub_type_hdr, device_list);
if (ret)
break;
}
sub_type_hdr = (typeof(sub_type_hdr))((char *)sub_type_hdr +
sub_type_hdr->length);
}
err:
if (ret)
kfd_release_topology_device_list(device_list);
return ret;
}
static int kfd_fill_gpu_cache_info_from_gfx_config(struct kfd_dev *kdev,
struct kfd_gpu_cache_info *pcache_info)
{
struct amdgpu_device *adev = kdev->adev;
int i = 0;
/* TCP L1 Cache per CU */
if (adev->gfx.config.gc_tcp_l1_size) {
pcache_info[i].cache_size = adev->gfx.config.gc_tcp_l1_size;
pcache_info[i].cache_level = 1;
pcache_info[i].flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE);
pcache_info[i].num_cu_shared = adev->gfx.config.gc_num_tcp_per_wpg / 2;
pcache_info[i].cache_line_size = adev->gfx.config.gc_tcp_cache_line_size;
i++;
}
/* Scalar L1 Instruction Cache per SQC */
if (adev->gfx.config.gc_l1_instruction_cache_size_per_sqc) {
pcache_info[i].cache_size =
adev->gfx.config.gc_l1_instruction_cache_size_per_sqc;
pcache_info[i].cache_level = 1;
pcache_info[i].flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_INST_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE);
pcache_info[i].num_cu_shared = adev->gfx.config.gc_num_sqc_per_wgp * 2;
pcache_info[i].cache_line_size = adev->gfx.config.gc_instruction_cache_line_size;
i++;
}
/* Scalar L1 Data Cache per SQC */
if (adev->gfx.config.gc_l1_data_cache_size_per_sqc) {
pcache_info[i].cache_size = adev->gfx.config.gc_l1_data_cache_size_per_sqc;
pcache_info[i].cache_level = 1;
pcache_info[i].flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE);
pcache_info[i].num_cu_shared = adev->gfx.config.gc_num_sqc_per_wgp * 2;
pcache_info[i].cache_line_size = adev->gfx.config.gc_scalar_data_cache_line_size;
i++;
}
/* GL1 Data Cache per SA */
if (adev->gfx.config.gc_gl1c_per_sa &&
adev->gfx.config.gc_gl1c_size_per_instance) {
pcache_info[i].cache_size = adev->gfx.config.gc_gl1c_per_sa *
adev->gfx.config.gc_gl1c_size_per_instance;
pcache_info[i].cache_level = 1;
pcache_info[i].flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE);
pcache_info[i].num_cu_shared = adev->gfx.config.max_cu_per_sh;
pcache_info[i].cache_line_size = 0;
i++;
}
/* L2 Data Cache per GPU (Total Tex Cache) */
if (adev->gfx.config.gc_gl2c_per_gpu) {
pcache_info[i].cache_size = adev->gfx.config.gc_gl2c_per_gpu;
pcache_info[i].cache_level = 2;
pcache_info[i].flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE);
pcache_info[i].num_cu_shared = adev->gfx.config.max_cu_per_sh;
pcache_info[i].cache_line_size = adev->gfx.config.gc_tcc_cache_line_size;
i++;
}
/* L3 Data Cache per GPU */
if (adev->gmc.mall_size) {
pcache_info[i].cache_size = adev->gmc.mall_size / 1024;
pcache_info[i].cache_level = 3;
pcache_info[i].flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE);
pcache_info[i].num_cu_shared = adev->gfx.config.max_cu_per_sh;
pcache_info[i].cache_line_size = 0;
i++;
}
return i;
}
static int kfd_fill_gpu_cache_info_from_gfx_config_v2(struct kfd_dev *kdev,
struct kfd_gpu_cache_info *pcache_info)
{
struct amdgpu_device *adev = kdev->adev;
int i = 0;
/* TCP L1 Cache per CU */
if (adev->gfx.config.gc_tcp_size_per_cu) {
pcache_info[i].cache_size = adev->gfx.config.gc_tcp_size_per_cu;
pcache_info[i].cache_level = 1;
/* Cacheline size not available in IP discovery for gc943,gc944 */
pcache_info[i].cache_line_size = 128;
pcache_info[i].flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE);
pcache_info[i].num_cu_shared = 1;
i++;
}
/* Scalar L1 Instruction Cache per SQC */
if (adev->gfx.config.gc_l1_instruction_cache_size_per_sqc) {
pcache_info[i].cache_size =
adev->gfx.config.gc_l1_instruction_cache_size_per_sqc;
pcache_info[i].cache_level = 1;
pcache_info[i].cache_line_size = 64;
pcache_info[i].flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_INST_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE);
pcache_info[i].num_cu_shared = adev->gfx.config.gc_num_cu_per_sqc;
i++;
}
/* Scalar L1 Data Cache per SQC */
if (adev->gfx.config.gc_l1_data_cache_size_per_sqc) {
pcache_info[i].cache_size = adev->gfx.config.gc_l1_data_cache_size_per_sqc;
pcache_info[i].cache_level = 1;
pcache_info[i].cache_line_size = 64;
pcache_info[i].flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE);
pcache_info[i].num_cu_shared = adev->gfx.config.gc_num_cu_per_sqc;
i++;
}
/* L2 Data Cache per GPU (Total Tex Cache) */
if (adev->gfx.config.gc_tcc_size) {
pcache_info[i].cache_size = adev->gfx.config.gc_tcc_size;
pcache_info[i].cache_level = 2;
pcache_info[i].cache_line_size = 128;
pcache_info[i].flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE);
pcache_info[i].num_cu_shared = adev->gfx.config.max_cu_per_sh;
i++;
}
/* L3 Data Cache per GPU */
if (adev->gmc.mall_size) {
pcache_info[i].cache_size = adev->gmc.mall_size / 1024;
pcache_info[i].cache_level = 3;
pcache_info[i].cache_line_size = 64;
pcache_info[i].flags = (CRAT_CACHE_FLAGS_ENABLED |
CRAT_CACHE_FLAGS_DATA_CACHE |
CRAT_CACHE_FLAGS_SIMD_CACHE);
pcache_info[i].num_cu_shared = adev->gfx.config.max_cu_per_sh;
i++;
}
return i;
}
int kfd_get_gpu_cache_info(struct kfd_node *kdev, struct kfd_gpu_cache_info **pcache_info)
{
int num_of_cache_types = 0;
switch (kdev->adev->asic_type) {
case CHIP_KAVERI:
*pcache_info = kaveri_cache_info;
num_of_cache_types = ARRAY_SIZE(kaveri_cache_info);
break;
case CHIP_HAWAII:
*pcache_info = hawaii_cache_info;
num_of_cache_types = ARRAY_SIZE(hawaii_cache_info);
break;
case CHIP_CARRIZO:
*pcache_info = carrizo_cache_info;
num_of_cache_types = ARRAY_SIZE(carrizo_cache_info);
break;
case CHIP_TONGA:
*pcache_info = tonga_cache_info;
num_of_cache_types = ARRAY_SIZE(tonga_cache_info);
break;
case CHIP_FIJI:
*pcache_info = fiji_cache_info;
num_of_cache_types = ARRAY_SIZE(fiji_cache_info);
break;
case CHIP_POLARIS10:
*pcache_info = polaris10_cache_info;
num_of_cache_types = ARRAY_SIZE(polaris10_cache_info);
break;
case CHIP_POLARIS11:
*pcache_info = polaris11_cache_info;
num_of_cache_types = ARRAY_SIZE(polaris11_cache_info);
break;
case CHIP_POLARIS12:
*pcache_info = polaris12_cache_info;
num_of_cache_types = ARRAY_SIZE(polaris12_cache_info);
break;
case CHIP_VEGAM:
*pcache_info = vegam_cache_info;
num_of_cache_types = ARRAY_SIZE(vegam_cache_info);
break;
default:
switch (KFD_GC_VERSION(kdev)) {
case IP_VERSION(9, 0, 1):
*pcache_info = vega10_cache_info;
num_of_cache_types = ARRAY_SIZE(vega10_cache_info);
break;
case IP_VERSION(9, 2, 1):
*pcache_info = vega12_cache_info;
num_of_cache_types = ARRAY_SIZE(vega12_cache_info);
break;
case IP_VERSION(9, 4, 0):
case IP_VERSION(9, 4, 1):
*pcache_info = vega20_cache_info;
num_of_cache_types = ARRAY_SIZE(vega20_cache_info);
break;
case IP_VERSION(9, 4, 2):
*pcache_info = aldebaran_cache_info;
num_of_cache_types = ARRAY_SIZE(aldebaran_cache_info);
break;
case IP_VERSION(9, 4, 3):
case IP_VERSION(9, 4, 4):
num_of_cache_types =
kfd_fill_gpu_cache_info_from_gfx_config_v2(kdev->kfd,
*pcache_info);
break;
case IP_VERSION(9, 1, 0):
case IP_VERSION(9, 2, 2):
*pcache_info = raven_cache_info;
num_of_cache_types = ARRAY_SIZE(raven_cache_info);
break;
case IP_VERSION(9, 3, 0):
*pcache_info = renoir_cache_info;
num_of_cache_types = ARRAY_SIZE(renoir_cache_info);
break;
case IP_VERSION(10, 1, 10):
case IP_VERSION(10, 1, 2):
case IP_VERSION(10, 1, 3):
case IP_VERSION(10, 1, 4):
*pcache_info = navi10_cache_info;
num_of_cache_types = ARRAY_SIZE(navi10_cache_info);
break;
case IP_VERSION(10, 1, 1):
*pcache_info = navi14_cache_info;
num_of_cache_types = ARRAY_SIZE(navi14_cache_info);
break;
case IP_VERSION(10, 3, 0):
*pcache_info = sienna_cichlid_cache_info;
num_of_cache_types = ARRAY_SIZE(sienna_cichlid_cache_info);
break;
case IP_VERSION(10, 3, 2):
*pcache_info = navy_flounder_cache_info;
num_of_cache_types = ARRAY_SIZE(navy_flounder_cache_info);
break;
case IP_VERSION(10, 3, 4):
*pcache_info = dimgrey_cavefish_cache_info;
num_of_cache_types = ARRAY_SIZE(dimgrey_cavefish_cache_info);
break;
case IP_VERSION(10, 3, 1):
*pcache_info = vangogh_cache_info;
num_of_cache_types = ARRAY_SIZE(vangogh_cache_info);
break;
case IP_VERSION(10, 3, 5):
*pcache_info = beige_goby_cache_info;
num_of_cache_types = ARRAY_SIZE(beige_goby_cache_info);
break;
case IP_VERSION(10, 3, 3):
*pcache_info = yellow_carp_cache_info;
num_of_cache_types = ARRAY_SIZE(yellow_carp_cache_info);
break;
case IP_VERSION(10, 3, 6):
*pcache_info = gc_10_3_6_cache_info;
num_of_cache_types = ARRAY_SIZE(gc_10_3_6_cache_info);
break;
case IP_VERSION(10, 3, 7):
*pcache_info = gfx1037_cache_info;
num_of_cache_types = ARRAY_SIZE(gfx1037_cache_info);
break;
case IP_VERSION(11, 0, 0):
case IP_VERSION(11, 0, 1):
case IP_VERSION(11, 0, 2):
case IP_VERSION(11, 0, 3):
case IP_VERSION(11, 0, 4):
case IP_VERSION(11, 5, 0):
case IP_VERSION(11, 5, 1):
case IP_VERSION(11, 5, 2):
case IP_VERSION(12, 0, 0):
case IP_VERSION(12, 0, 1):
num_of_cache_types =
kfd_fill_gpu_cache_info_from_gfx_config(kdev->kfd, *pcache_info);
break;
default:
*pcache_info = dummy_cache_info;
num_of_cache_types = ARRAY_SIZE(dummy_cache_info);
pr_warn("dummy cache info is used temporarily and real cache info need update later.\n");
break;
}
}
return num_of_cache_types;
}
/* Memory required to create Virtual CRAT.
* Since there is no easy way to predict the amount of memory required, the
* following amount is allocated for GPU Virtual CRAT. This is
* expected to cover all known conditions. But to be safe additional check
* is put in the code to ensure we don't overwrite.
*/
#define VCRAT_SIZE_FOR_GPU (4 * PAGE_SIZE)
/* kfd_fill_cu_for_cpu - Fill in Compute info for the given CPU NUMA node
*
* @numa_node_id: CPU NUMA node id
* @avail_size: Available size in the memory
* @sub_type_hdr: Memory into which compute info will be filled in
*
* Return 0 if successful else return -ve value
*/
static int kfd_fill_cu_for_cpu(int numa_node_id, int *avail_size,
int proximity_domain,
struct crat_subtype_computeunit *sub_type_hdr)
{
const struct cpumask *cpumask;
*avail_size -= sizeof(struct crat_subtype_computeunit);
if (*avail_size < 0)
return -ENOMEM;
memset(sub_type_hdr, 0, sizeof(struct crat_subtype_computeunit));
/* Fill in subtype header data */
sub_type_hdr->type = CRAT_SUBTYPE_COMPUTEUNIT_AFFINITY;
sub_type_hdr->length = sizeof(struct crat_subtype_computeunit);
sub_type_hdr->flags = CRAT_SUBTYPE_FLAGS_ENABLED;
cpumask = cpumask_of_node(numa_node_id);
/* Fill in CU data */
sub_type_hdr->flags |= CRAT_CU_FLAGS_CPU_PRESENT;
sub_type_hdr->proximity_domain = proximity_domain;
sub_type_hdr->processor_id_low = kfd_numa_node_to_apic_id(numa_node_id);
if (sub_type_hdr->processor_id_low == -1)
return -EINVAL;
sub_type_hdr->num_cpu_cores = cpumask_weight(cpumask);
return 0;
}
/* kfd_fill_mem_info_for_cpu - Fill in Memory info for the given CPU NUMA node
*
* @numa_node_id: CPU NUMA node id
* @avail_size: Available size in the memory
* @sub_type_hdr: Memory into which compute info will be filled in
*
* Return 0 if successful else return -ve value
*/
static int kfd_fill_mem_info_for_cpu(int numa_node_id, int *avail_size,
int proximity_domain,
struct crat_subtype_memory *sub_type_hdr)
{
uint64_t mem_in_bytes = 0;
pg_data_t *pgdat;
int zone_type;
*avail_size -= sizeof(struct crat_subtype_memory);
if (*avail_size < 0)
return -ENOMEM;
memset(sub_type_hdr, 0, sizeof(struct crat_subtype_memory));
/* Fill in subtype header data */
sub_type_hdr->type = CRAT_SUBTYPE_MEMORY_AFFINITY;
sub_type_hdr->length = sizeof(struct crat_subtype_memory);
sub_type_hdr->flags = CRAT_SUBTYPE_FLAGS_ENABLED;
/* Fill in Memory Subunit data */
/* Unlike si_meminfo, si_meminfo_node is not exported. So
* the following lines are duplicated from si_meminfo_node
* function
*/
pgdat = NODE_DATA(numa_node_id);
for (zone_type = 0; zone_type < MAX_NR_ZONES; zone_type++)
mem_in_bytes += zone_managed_pages(&pgdat->node_zones[zone_type]);
mem_in_bytes <<= PAGE_SHIFT;
sub_type_hdr->length_low = lower_32_bits(mem_in_bytes);
sub_type_hdr->length_high = upper_32_bits(mem_in_bytes);
sub_type_hdr->proximity_domain = proximity_domain;
return 0;
}
#ifdef CONFIG_X86_64
static int kfd_fill_iolink_info_for_cpu(int numa_node_id, int *avail_size,
uint32_t *num_entries,
struct crat_subtype_iolink *sub_type_hdr)
{
int nid;
struct cpuinfo_x86 *c = &cpu_data(0);
uint8_t link_type;
if (c->x86_vendor == X86_VENDOR_AMD)
link_type = CRAT_IOLINK_TYPE_HYPERTRANSPORT;
else
link_type = CRAT_IOLINK_TYPE_QPI_1_1;
*num_entries = 0;
/* Create IO links from this node to other CPU nodes */
for_each_online_node(nid) {
if (nid == numa_node_id) /* node itself */
continue;
*avail_size -= sizeof(struct crat_subtype_iolink);
if (*avail_size < 0)
return -ENOMEM;
memset(sub_type_hdr, 0, sizeof(struct crat_subtype_iolink));
/* Fill in subtype header data */
sub_type_hdr->type = CRAT_SUBTYPE_IOLINK_AFFINITY;
sub_type_hdr->length = sizeof(struct crat_subtype_iolink);
sub_type_hdr->flags = CRAT_SUBTYPE_FLAGS_ENABLED;
/* Fill in IO link data */
sub_type_hdr->proximity_domain_from = numa_node_id;
sub_type_hdr->proximity_domain_to = nid;
sub_type_hdr->io_interface_type = link_type;
(*num_entries)++;
sub_type_hdr++;
}
return 0;
}
#endif
/* kfd_create_vcrat_image_cpu - Create Virtual CRAT for CPU
*
* @pcrat_image: Fill in VCRAT for CPU
* @size: [IN] allocated size of crat_image.
* [OUT] actual size of data filled in crat_image
*/
static int kfd_create_vcrat_image_cpu(void *pcrat_image, size_t *size)
{
struct crat_header *crat_table = (struct crat_header *)pcrat_image;
struct acpi_table_header *acpi_table;
acpi_status status;
struct crat_subtype_generic *sub_type_hdr;
int avail_size = *size;
int numa_node_id;
#ifdef CONFIG_X86_64
uint32_t entries = 0;
#endif
int ret = 0;
if (!pcrat_image)
return -EINVAL;
/* Fill in CRAT Header.
* Modify length and total_entries as subunits are added.
*/
avail_size -= sizeof(struct crat_header);
if (avail_size < 0)
return -ENOMEM;
memset(crat_table, 0, sizeof(struct crat_header));
memcpy(&crat_table->signature, CRAT_SIGNATURE,
sizeof(crat_table->signature));
crat_table->length = sizeof(struct crat_header);
status = acpi_get_table("DSDT", 0, &acpi_table);
if (status != AE_OK)
pr_warn("DSDT table not found for OEM information\n");
else {
crat_table->oem_revision = acpi_table->revision;
memcpy(crat_table->oem_id, acpi_table->oem_id,
CRAT_OEMID_LENGTH);
memcpy(crat_table->oem_table_id, acpi_table->oem_table_id,
CRAT_OEMTABLEID_LENGTH);
acpi_put_table(acpi_table);
}
crat_table->total_entries = 0;
crat_table->num_domains = 0;
sub_type_hdr = (struct crat_subtype_generic *)(crat_table+1);
for_each_online_node(numa_node_id) {
if (kfd_numa_node_to_apic_id(numa_node_id) == -1)
continue;
/* Fill in Subtype: Compute Unit */
ret = kfd_fill_cu_for_cpu(numa_node_id, &avail_size,
crat_table->num_domains,
(struct crat_subtype_computeunit *)sub_type_hdr);
if (ret < 0)
return ret;
crat_table->length += sub_type_hdr->length;
crat_table->total_entries++;
sub_type_hdr = (typeof(sub_type_hdr))((char *)sub_type_hdr +
sub_type_hdr->length);
/* Fill in Subtype: Memory */
ret = kfd_fill_mem_info_for_cpu(numa_node_id, &avail_size,
crat_table->num_domains,
(struct crat_subtype_memory *)sub_type_hdr);
if (ret < 0)
return ret;
crat_table->length += sub_type_hdr->length;
crat_table->total_entries++;
sub_type_hdr = (typeof(sub_type_hdr))((char *)sub_type_hdr +
sub_type_hdr->length);
/* Fill in Subtype: IO Link */
#ifdef CONFIG_X86_64
ret = kfd_fill_iolink_info_for_cpu(numa_node_id, &avail_size,
&entries,
(struct crat_subtype_iolink *)sub_type_hdr);
if (ret < 0)
return ret;
if (entries) {
crat_table->length += (sub_type_hdr->length * entries);
crat_table->total_entries += entries;
sub_type_hdr = (typeof(sub_type_hdr))((char *)sub_type_hdr +
sub_type_hdr->length * entries);
}
#else
pr_info("IO link not available for non x86 platforms\n");
#endif
crat_table->num_domains++;
}
/* TODO: Add cache Subtype for CPU.
* Currently, CPU cache information is available in function
* detect_cache_attributes(cpu) defined in the file
* ./arch/x86/kernel/cpu/intel_cacheinfo.c. This function is not
* exported and to get the same information the code needs to be
* duplicated.
*/
*size = crat_table->length;
pr_info("Virtual CRAT table created for CPU\n");
return 0;
}
static int kfd_fill_gpu_memory_affinity(int *avail_size,
struct kfd_node *kdev, uint8_t type, uint64_t size,
struct crat_subtype_memory *sub_type_hdr,
uint32_t proximity_domain,
const struct kfd_local_mem_info *local_mem_info)
{
*avail_size -= sizeof(struct crat_subtype_memory);
if (*avail_size < 0)
return -ENOMEM;
memset((void *)sub_type_hdr, 0, sizeof(struct crat_subtype_memory));
sub_type_hdr->type = CRAT_SUBTYPE_MEMORY_AFFINITY;
sub_type_hdr->length = sizeof(struct crat_subtype_memory);
sub_type_hdr->flags |= CRAT_SUBTYPE_FLAGS_ENABLED;
sub_type_hdr->proximity_domain = proximity_domain;
pr_debug("Fill gpu memory affinity - type 0x%x size 0x%llx\n",
type, size);
sub_type_hdr->length_low = lower_32_bits(size);
sub_type_hdr->length_high = upper_32_bits(size);
sub_type_hdr->width = local_mem_info->vram_width;
sub_type_hdr->visibility_type = type;
return 0;
}
#ifdef CONFIG_ACPI_NUMA
static void kfd_find_numa_node_in_srat(struct kfd_node *kdev)
{
struct acpi_table_header *table_header = NULL;
struct acpi_subtable_header *sub_header = NULL;
unsigned long table_end, subtable_len;
u32 pci_id = pci_domain_nr(kdev->adev->pdev->bus) << 16 |
pci_dev_id(kdev->adev->pdev);
u32 bdf;
acpi_status status;
struct acpi_srat_cpu_affinity *cpu;
struct acpi_srat_generic_affinity *gpu;
int pxm = 0, max_pxm = 0;
int numa_node = NUMA_NO_NODE;
bool found = false;
/* Fetch the SRAT table from ACPI */
status = acpi_get_table(ACPI_SIG_SRAT, 0, &table_header);
if (status == AE_NOT_FOUND) {
pr_warn("SRAT table not found\n");
return;
} else if (ACPI_FAILURE(status)) {
const char *err = acpi_format_exception(status);
pr_err("SRAT table error: %s\n", err);
return;
}
table_end = (unsigned long)table_header + table_header->length;
/* Parse all entries looking for a match. */
sub_header = (struct acpi_subtable_header *)
((unsigned long)table_header +
sizeof(struct acpi_table_srat));
subtable_len = sub_header->length;
while (((unsigned long)sub_header) + subtable_len < table_end) {
/*
* If length is 0, break from this loop to avoid
* infinite loop.
*/
if (subtable_len == 0) {
pr_err("SRAT invalid zero length\n");
break;
}
switch (sub_header->type) {
case ACPI_SRAT_TYPE_CPU_AFFINITY:
cpu = (struct acpi_srat_cpu_affinity *)sub_header;
pxm = *((u32 *)cpu->proximity_domain_hi) << 8 |
cpu->proximity_domain_lo;
if (pxm > max_pxm)
max_pxm = pxm;
break;
case ACPI_SRAT_TYPE_GENERIC_AFFINITY:
gpu = (struct acpi_srat_generic_affinity *)sub_header;
bdf = *((u16 *)(&gpu->device_handle[0])) << 16 |
*((u16 *)(&gpu->device_handle[2]));
if (bdf == pci_id) {
found = true;
numa_node = pxm_to_node(gpu->proximity_domain);
}
break;
default:
break;
}
if (found)
break;
sub_header = (struct acpi_subtable_header *)
((unsigned long)sub_header + subtable_len);
subtable_len = sub_header->length;
}
acpi_put_table(table_header);
/* Workaround bad cpu-gpu binding case */
if (found && (numa_node < 0 ||
numa_node > pxm_to_node(max_pxm)))
numa_node = 0;
if (numa_node != NUMA_NO_NODE)
set_dev_node(&kdev->adev->pdev->dev, numa_node);
}
#endif
#define KFD_CRAT_INTRA_SOCKET_WEIGHT 13
#define KFD_CRAT_XGMI_WEIGHT 15
/* kfd_fill_gpu_direct_io_link - Fill in direct io link from GPU
* to its NUMA node
* @avail_size: Available size in the memory
* @kdev - [IN] GPU device
* @sub_type_hdr: Memory into which io link info will be filled in
* @proximity_domain - proximity domain of the GPU node
*
* Return 0 if successful else return -ve value
*/
static int kfd_fill_gpu_direct_io_link_to_cpu(int *avail_size,
struct kfd_node *kdev,
struct crat_subtype_iolink *sub_type_hdr,
uint32_t proximity_domain)
{
*avail_size -= sizeof(struct crat_subtype_iolink);
if (*avail_size < 0)
return -ENOMEM;
memset((void *)sub_type_hdr, 0, sizeof(struct crat_subtype_iolink));
/* Fill in subtype header data */
sub_type_hdr->type = CRAT_SUBTYPE_IOLINK_AFFINITY;
sub_type_hdr->length = sizeof(struct crat_subtype_iolink);
sub_type_hdr->flags |= CRAT_SUBTYPE_FLAGS_ENABLED;
if (kfd_dev_is_large_bar(kdev))
sub_type_hdr->flags |= CRAT_IOLINK_FLAGS_BI_DIRECTIONAL;
/* Fill in IOLINK subtype.
* TODO: Fill-in other fields of iolink subtype
*/
if (kdev->adev->gmc.xgmi.connected_to_cpu ||
(KFD_GC_VERSION(kdev) == IP_VERSION(9, 4, 3) &&
kdev->adev->smuio.funcs->get_pkg_type(kdev->adev) ==
AMDGPU_PKG_TYPE_APU)) {
bool ext_cpu = KFD_GC_VERSION(kdev) != IP_VERSION(9, 4, 3);
int mem_bw = 819200, weight = ext_cpu ? KFD_CRAT_XGMI_WEIGHT :
KFD_CRAT_INTRA_SOCKET_WEIGHT;
uint32_t bandwidth = ext_cpu ? amdgpu_amdkfd_get_xgmi_bandwidth_mbytes(
kdev->adev, NULL, true) : mem_bw;
/*
* with host gpu xgmi link, host can access gpu memory whether
* or not pcie bar type is large, so always create bidirectional
* io link.
*/
sub_type_hdr->flags |= CRAT_IOLINK_FLAGS_BI_DIRECTIONAL;
sub_type_hdr->io_interface_type = CRAT_IOLINK_TYPE_XGMI;
sub_type_hdr->weight_xgmi = weight;
sub_type_hdr->minimum_bandwidth_mbs = bandwidth;
sub_type_hdr->maximum_bandwidth_mbs = bandwidth;
} else {
sub_type_hdr->io_interface_type = CRAT_IOLINK_TYPE_PCIEXPRESS;
sub_type_hdr->minimum_bandwidth_mbs =
amdgpu_amdkfd_get_pcie_bandwidth_mbytes(kdev->adev, true);
sub_type_hdr->maximum_bandwidth_mbs =
amdgpu_amdkfd_get_pcie_bandwidth_mbytes(kdev->adev, false);
}
sub_type_hdr->proximity_domain_from = proximity_domain;
#ifdef CONFIG_ACPI_NUMA
if (kdev->adev->pdev->dev.numa_node == NUMA_NO_NODE &&
num_possible_nodes() > 1)
kfd_find_numa_node_in_srat(kdev);
#endif
#ifdef CONFIG_NUMA
if (kdev->adev->pdev->dev.numa_node == NUMA_NO_NODE)
sub_type_hdr->proximity_domain_to = 0;
else
sub_type_hdr->proximity_domain_to = kdev->adev->pdev->dev.numa_node;
#else
sub_type_hdr->proximity_domain_to = 0;
#endif
return 0;
}
static int kfd_fill_gpu_xgmi_link_to_gpu(int *avail_size,
struct kfd_node *kdev,
struct kfd_node *peer_kdev,
struct crat_subtype_iolink *sub_type_hdr,
uint32_t proximity_domain_from,
uint32_t proximity_domain_to)
{
bool use_ta_info = kdev->kfd->num_nodes == 1;
*avail_size -= sizeof(struct crat_subtype_iolink);
if (*avail_size < 0)
return -ENOMEM;
memset((void *)sub_type_hdr, 0, sizeof(struct crat_subtype_iolink));
sub_type_hdr->type = CRAT_SUBTYPE_IOLINK_AFFINITY;
sub_type_hdr->length = sizeof(struct crat_subtype_iolink);
sub_type_hdr->flags |= CRAT_SUBTYPE_FLAGS_ENABLED |
CRAT_IOLINK_FLAGS_BI_DIRECTIONAL;
sub_type_hdr->io_interface_type = CRAT_IOLINK_TYPE_XGMI;
sub_type_hdr->proximity_domain_from = proximity_domain_from;
sub_type_hdr->proximity_domain_to = proximity_domain_to;
if (use_ta_info) {
sub_type_hdr->weight_xgmi = KFD_CRAT_XGMI_WEIGHT *
amdgpu_amdkfd_get_xgmi_hops_count(kdev->adev, peer_kdev->adev);
sub_type_hdr->maximum_bandwidth_mbs =
amdgpu_amdkfd_get_xgmi_bandwidth_mbytes(kdev->adev,
peer_kdev->adev, false);
sub_type_hdr->minimum_bandwidth_mbs = sub_type_hdr->maximum_bandwidth_mbs ?
amdgpu_amdkfd_get_xgmi_bandwidth_mbytes(kdev->adev, NULL, true) : 0;
} else {
bool is_single_hop = kdev->kfd == peer_kdev->kfd;
int weight = is_single_hop ? KFD_CRAT_INTRA_SOCKET_WEIGHT :
(2 * KFD_CRAT_INTRA_SOCKET_WEIGHT) + KFD_CRAT_XGMI_WEIGHT;
int mem_bw = 819200;
sub_type_hdr->weight_xgmi = weight;
sub_type_hdr->maximum_bandwidth_mbs = is_single_hop ? mem_bw : 0;
sub_type_hdr->minimum_bandwidth_mbs = is_single_hop ? mem_bw : 0;
}
return 0;
}
/* kfd_create_vcrat_image_gpu - Create Virtual CRAT for CPU
*
* @pcrat_image: Fill in VCRAT for GPU
* @size: [IN] allocated size of crat_image.
* [OUT] actual size of data filled in crat_image
*/
static int kfd_create_vcrat_image_gpu(void *pcrat_image,
size_t *size, struct kfd_node *kdev,
uint32_t proximity_domain)
{
struct crat_header *crat_table = (struct crat_header *)pcrat_image;
struct amdgpu_gfx_config *gfx_info = &kdev->adev->gfx.config;
struct amdgpu_cu_info *cu_info = &kdev->adev->gfx.cu_info;
struct crat_subtype_generic *sub_type_hdr;
struct kfd_local_mem_info local_mem_info;
struct kfd_topology_device *peer_dev;
struct crat_subtype_computeunit *cu;
int avail_size = *size;
uint32_t total_num_of_cu;
uint32_t nid = 0;
int ret = 0;
if (!pcrat_image || avail_size < VCRAT_SIZE_FOR_GPU)
return -EINVAL;
/* Fill the CRAT Header.
* Modify length and total_entries as subunits are added.
*/
avail_size -= sizeof(struct crat_header);
memset(crat_table, 0, sizeof(struct crat_header));
memcpy(&crat_table->signature, CRAT_SIGNATURE,
sizeof(crat_table->signature));
/* Change length as we add more subtypes*/
crat_table->length = sizeof(struct crat_header);
crat_table->num_domains = 1;
crat_table->total_entries = 0;
/* Fill in Subtype: Compute Unit
* First fill in the sub type header and then sub type data
*/
avail_size -= sizeof(struct crat_subtype_computeunit);
sub_type_hdr = (struct crat_subtype_generic *)(crat_table + 1);
memset(sub_type_hdr, 0, sizeof(struct crat_subtype_computeunit));
sub_type_hdr->type = CRAT_SUBTYPE_COMPUTEUNIT_AFFINITY;
sub_type_hdr->length = sizeof(struct crat_subtype_computeunit);
sub_type_hdr->flags = CRAT_SUBTYPE_FLAGS_ENABLED;
/* Fill CU subtype data */
cu = (struct crat_subtype_computeunit *)sub_type_hdr;
cu->flags |= CRAT_CU_FLAGS_GPU_PRESENT;
cu->proximity_domain = proximity_domain;
cu->num_simd_per_cu = cu_info->simd_per_cu;
cu->num_simd_cores = cu_info->simd_per_cu *
(cu_info->number / kdev->kfd->num_nodes);
cu->max_waves_simd = cu_info->max_waves_per_simd;
cu->wave_front_size = cu_info->wave_front_size;
cu->array_count = gfx_info->max_sh_per_se *
gfx_info->max_shader_engines;
total_num_of_cu = (cu->array_count * gfx_info->max_cu_per_sh);
cu->processor_id_low = get_and_inc_gpu_processor_id(total_num_of_cu);
cu->num_cu_per_array = gfx_info->max_cu_per_sh;
cu->max_slots_scatch_cu = cu_info->max_scratch_slots_per_cu;
cu->num_banks = gfx_info->max_shader_engines;
cu->lds_size_in_kb = cu_info->lds_size;
cu->hsa_capability = 0;
crat_table->length += sub_type_hdr->length;
crat_table->total_entries++;
/* Fill in Subtype: Memory. Only on systems with large BAR (no
* private FB), report memory as public. On other systems
* report the total FB size (public+private) as a single
* private heap.
*/
local_mem_info = kdev->local_mem_info;
sub_type_hdr = (typeof(sub_type_hdr))((char *)sub_type_hdr +
sub_type_hdr->length);
if (kdev->adev->debug_largebar)
local_mem_info.local_mem_size_private = 0;
if (local_mem_info.local_mem_size_private == 0)
ret = kfd_fill_gpu_memory_affinity(&avail_size,
kdev, HSA_MEM_HEAP_TYPE_FB_PUBLIC,
local_mem_info.local_mem_size_public,
(struct crat_subtype_memory *)sub_type_hdr,
proximity_domain,
&local_mem_info);
else
ret = kfd_fill_gpu_memory_affinity(&avail_size,
kdev, HSA_MEM_HEAP_TYPE_FB_PRIVATE,
local_mem_info.local_mem_size_public +
local_mem_info.local_mem_size_private,
(struct crat_subtype_memory *)sub_type_hdr,
proximity_domain,
&local_mem_info);
if (ret < 0)
return ret;
crat_table->length += sizeof(struct crat_subtype_memory);
crat_table->total_entries++;
/* Fill in Subtype: IO_LINKS
* Only direct links are added here which is Link from GPU to
* its NUMA node. Indirect links are added by userspace.
*/
sub_type_hdr = (typeof(sub_type_hdr))((char *)sub_type_hdr +
sub_type_hdr->length);
ret = kfd_fill_gpu_direct_io_link_to_cpu(&avail_size, kdev,
(struct crat_subtype_iolink *)sub_type_hdr, proximity_domain);
if (ret < 0)
return ret;
crat_table->length += sub_type_hdr->length;
crat_table->total_entries++;
/* Fill in Subtype: IO_LINKS
* Direct links from GPU to other GPUs through xGMI.
* We will loop GPUs that already be processed (with lower value
* of proximity_domain), add the link for the GPUs with same
* hive id (from this GPU to other GPU) . The reversed iolink
* (from other GPU to this GPU) will be added
* in kfd_parse_subtype_iolink.
*/
if (kdev->kfd->hive_id) {
for (nid = 0; nid < proximity_domain; ++nid) {
peer_dev = kfd_topology_device_by_proximity_domain_no_lock(nid);
if (!peer_dev->gpu)
continue;
if (peer_dev->gpu->kfd->hive_id != kdev->kfd->hive_id)
continue;
if (!amdgpu_xgmi_get_is_sharing_enabled(kdev->adev, peer_dev->gpu->adev))
continue;
sub_type_hdr = (typeof(sub_type_hdr))(
(char *)sub_type_hdr +
sizeof(struct crat_subtype_iolink));
ret = kfd_fill_gpu_xgmi_link_to_gpu(
&avail_size, kdev, peer_dev->gpu,
(struct crat_subtype_iolink *)sub_type_hdr,
proximity_domain, nid);
if (ret < 0)
return ret;
crat_table->length += sub_type_hdr->length;
crat_table->total_entries++;
}
}
*size = crat_table->length;
pr_info("Virtual CRAT table created for GPU\n");
return ret;
}
/* kfd_create_crat_image_virtual - Allocates memory for CRAT image and
* creates a Virtual CRAT (VCRAT) image
*
* NOTE: Call kfd_destroy_crat_image to free CRAT image memory
*
* @crat_image: VCRAT image created because ACPI does not have a
* CRAT for this device
* @size: [OUT] size of virtual crat_image
* @flags: COMPUTE_UNIT_CPU - Create VCRAT for CPU device
* COMPUTE_UNIT_GPU - Create VCRAT for GPU
* (COMPUTE_UNIT_CPU | COMPUTE_UNIT_GPU) - Create VCRAT for APU
* -- this option is not currently implemented.
* The assumption is that all AMD APUs will have CRAT
* @kdev: Valid kfd_node required if flags contain COMPUTE_UNIT_GPU
*
* Return 0 if successful else return -ve value
*/
int kfd_create_crat_image_virtual(void **crat_image, size_t *size,
int flags, struct kfd_node *kdev,
uint32_t proximity_domain)
{
void *pcrat_image = NULL;
int ret = 0, num_nodes;
size_t dyn_size;
if (!crat_image)
return -EINVAL;
*crat_image = NULL;
/* Allocate the CPU Virtual CRAT size based on the number of online
* nodes. Allocate VCRAT_SIZE_FOR_GPU for GPU virtual CRAT image.
* This should cover all the current conditions. A check is put not
* to overwrite beyond allocated size for GPUs
*/
switch (flags) {
case COMPUTE_UNIT_CPU:
num_nodes = num_online_nodes();
dyn_size = sizeof(struct crat_header) +
num_nodes * (sizeof(struct crat_subtype_computeunit) +
sizeof(struct crat_subtype_memory) +
(num_nodes - 1) * sizeof(struct crat_subtype_iolink));
pcrat_image = kvmalloc(dyn_size, GFP_KERNEL);
if (!pcrat_image)
return -ENOMEM;
*size = dyn_size;
pr_debug("CRAT size is %ld", dyn_size);
ret = kfd_create_vcrat_image_cpu(pcrat_image, size);
break;
case COMPUTE_UNIT_GPU:
if (!kdev)
return -EINVAL;
pcrat_image = kvmalloc(VCRAT_SIZE_FOR_GPU, GFP_KERNEL);
if (!pcrat_image)
return -ENOMEM;
*size = VCRAT_SIZE_FOR_GPU;
ret = kfd_create_vcrat_image_gpu(pcrat_image, size, kdev,
proximity_domain);
break;
case (COMPUTE_UNIT_CPU | COMPUTE_UNIT_GPU):
/* TODO: */
ret = -EINVAL;
pr_err("VCRAT not implemented for APU\n");
break;
default:
ret = -EINVAL;
}
if (!ret)
*crat_image = pcrat_image;
else
kvfree(pcrat_image);
return ret;
}
/* kfd_destroy_crat_image
*
* @crat_image: [IN] - crat_image from kfd_create_crat_image_xxx(..)
*
*/
void kfd_destroy_crat_image(void *crat_image)
{
kvfree(crat_image);
}
|