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
|
/*
* Copyright 2009 Jerome Glisse.
* All Rights Reserved.
*
* 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, sub license, 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 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 NON-INFRINGEMENT. IN NO EVENT SHALL
* THE COPYRIGHT HOLDERS, AUTHORS AND/OR ITS SUPPLIERS 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.
*
* The above copyright notice and this permission notice (including the
* next paragraph) shall be included in all copies or substantial portions
* of the Software.
*
*/
/*
* Authors:
* Jerome Glisse <glisse@freedesktop.org>
* Thomas Hellstrom <thomas-at-tungstengraphics-dot-com>
* Dave Airlie
*/
#include <linux/dma-mapping.h>
#include <linux/iommu.h>
#include <linux/hmm.h>
#include <linux/pagemap.h>
#include <linux/sched/task.h>
#include <linux/sched/mm.h>
#include <linux/seq_file.h>
#include <linux/slab.h>
#include <linux/swap.h>
#include <linux/swiotlb.h>
#include <linux/dma-buf.h>
#include <drm/ttm/ttm_bo_api.h>
#include <drm/ttm/ttm_bo_driver.h>
#include <drm/ttm/ttm_placement.h>
#include <drm/ttm/ttm_module.h>
#include <drm/ttm/ttm_page_alloc.h>
#include <drm/drm_debugfs.h>
#include <drm/amdgpu_drm.h>
#include "amdgpu.h"
#include "amdgpu_object.h"
#include "amdgpu_trace.h"
#include "amdgpu_amdkfd.h"
#include "amdgpu_sdma.h"
#include "amdgpu_ras.h"
#include "bif/bif_4_1_d.h"
static int amdgpu_map_buffer(struct ttm_buffer_object *bo,
struct ttm_mem_reg *mem, unsigned num_pages,
uint64_t offset, unsigned window,
struct amdgpu_ring *ring,
uint64_t *addr);
static int amdgpu_ttm_debugfs_init(struct amdgpu_device *adev);
static void amdgpu_ttm_debugfs_fini(struct amdgpu_device *adev);
static int amdgpu_invalidate_caches(struct ttm_bo_device *bdev, uint32_t flags)
{
return 0;
}
/**
* amdgpu_init_mem_type - Initialize a memory manager for a specific type of
* memory request.
*
* @bdev: The TTM BO device object (contains a reference to amdgpu_device)
* @type: The type of memory requested
* @man: The memory type manager for each domain
*
* This is called by ttm_bo_init_mm() when a buffer object is being
* initialized.
*/
static int amdgpu_init_mem_type(struct ttm_bo_device *bdev, uint32_t type,
struct ttm_mem_type_manager *man)
{
struct amdgpu_device *adev;
adev = amdgpu_ttm_adev(bdev);
switch (type) {
case TTM_PL_SYSTEM:
/* System memory */
man->flags = TTM_MEMTYPE_FLAG_MAPPABLE;
man->available_caching = TTM_PL_MASK_CACHING;
man->default_caching = TTM_PL_FLAG_CACHED;
break;
case TTM_PL_TT:
/* GTT memory */
man->func = &amdgpu_gtt_mgr_func;
man->gpu_offset = adev->gmc.gart_start;
man->available_caching = TTM_PL_MASK_CACHING;
man->default_caching = TTM_PL_FLAG_CACHED;
man->flags = TTM_MEMTYPE_FLAG_MAPPABLE | TTM_MEMTYPE_FLAG_CMA;
break;
case TTM_PL_VRAM:
/* "On-card" video ram */
man->func = &amdgpu_vram_mgr_func;
man->gpu_offset = adev->gmc.vram_start;
man->flags = TTM_MEMTYPE_FLAG_FIXED |
TTM_MEMTYPE_FLAG_MAPPABLE;
man->available_caching = TTM_PL_FLAG_UNCACHED | TTM_PL_FLAG_WC;
man->default_caching = TTM_PL_FLAG_WC;
break;
case AMDGPU_PL_GDS:
case AMDGPU_PL_GWS:
case AMDGPU_PL_OA:
/* On-chip GDS memory*/
man->func = &ttm_bo_manager_func;
man->gpu_offset = 0;
man->flags = TTM_MEMTYPE_FLAG_FIXED | TTM_MEMTYPE_FLAG_CMA;
man->available_caching = TTM_PL_FLAG_UNCACHED;
man->default_caching = TTM_PL_FLAG_UNCACHED;
break;
default:
DRM_ERROR("Unsupported memory type %u\n", (unsigned)type);
return -EINVAL;
}
return 0;
}
/**
* amdgpu_evict_flags - Compute placement flags
*
* @bo: The buffer object to evict
* @placement: Possible destination(s) for evicted BO
*
* Fill in placement data when ttm_bo_evict() is called
*/
static void amdgpu_evict_flags(struct ttm_buffer_object *bo,
struct ttm_placement *placement)
{
struct amdgpu_device *adev = amdgpu_ttm_adev(bo->bdev);
struct amdgpu_bo *abo;
static const struct ttm_place placements = {
.fpfn = 0,
.lpfn = 0,
.flags = TTM_PL_MASK_CACHING | TTM_PL_FLAG_SYSTEM
};
/* Don't handle scatter gather BOs */
if (bo->type == ttm_bo_type_sg) {
placement->num_placement = 0;
placement->num_busy_placement = 0;
return;
}
/* Object isn't an AMDGPU object so ignore */
if (!amdgpu_bo_is_amdgpu_bo(bo)) {
placement->placement = &placements;
placement->busy_placement = &placements;
placement->num_placement = 1;
placement->num_busy_placement = 1;
return;
}
abo = ttm_to_amdgpu_bo(bo);
switch (bo->mem.mem_type) {
case AMDGPU_PL_GDS:
case AMDGPU_PL_GWS:
case AMDGPU_PL_OA:
placement->num_placement = 0;
placement->num_busy_placement = 0;
return;
case TTM_PL_VRAM:
if (!adev->mman.buffer_funcs_enabled) {
/* Move to system memory */
amdgpu_bo_placement_from_domain(abo, AMDGPU_GEM_DOMAIN_CPU);
} else if (!amdgpu_gmc_vram_full_visible(&adev->gmc) &&
!(abo->flags & AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED) &&
amdgpu_bo_in_cpu_visible_vram(abo)) {
/* Try evicting to the CPU inaccessible part of VRAM
* first, but only set GTT as busy placement, so this
* BO will be evicted to GTT rather than causing other
* BOs to be evicted from VRAM
*/
amdgpu_bo_placement_from_domain(abo, AMDGPU_GEM_DOMAIN_VRAM |
AMDGPU_GEM_DOMAIN_GTT);
abo->placements[0].fpfn = adev->gmc.visible_vram_size >> PAGE_SHIFT;
abo->placements[0].lpfn = 0;
abo->placement.busy_placement = &abo->placements[1];
abo->placement.num_busy_placement = 1;
} else {
/* Move to GTT memory */
amdgpu_bo_placement_from_domain(abo, AMDGPU_GEM_DOMAIN_GTT);
}
break;
case TTM_PL_TT:
default:
amdgpu_bo_placement_from_domain(abo, AMDGPU_GEM_DOMAIN_CPU);
break;
}
*placement = abo->placement;
}
/**
* amdgpu_verify_access - Verify access for a mmap call
*
* @bo: The buffer object to map
* @filp: The file pointer from the process performing the mmap
*
* This is called by ttm_bo_mmap() to verify whether a process
* has the right to mmap a BO to their process space.
*/
static int amdgpu_verify_access(struct ttm_buffer_object *bo, struct file *filp)
{
struct amdgpu_bo *abo = ttm_to_amdgpu_bo(bo);
/*
* Don't verify access for KFD BOs. They don't have a GEM
* object associated with them.
*/
if (abo->kfd_bo)
return 0;
if (amdgpu_ttm_tt_get_usermm(bo->ttm))
return -EPERM;
return drm_vma_node_verify_access(&abo->tbo.base.vma_node,
filp->private_data);
}
/**
* amdgpu_move_null - Register memory for a buffer object
*
* @bo: The bo to assign the memory to
* @new_mem: The memory to be assigned.
*
* Assign the memory from new_mem to the memory of the buffer object bo.
*/
static void amdgpu_move_null(struct ttm_buffer_object *bo,
struct ttm_mem_reg *new_mem)
{
struct ttm_mem_reg *old_mem = &bo->mem;
BUG_ON(old_mem->mm_node != NULL);
*old_mem = *new_mem;
new_mem->mm_node = NULL;
}
/**
* amdgpu_mm_node_addr - Compute the GPU relative offset of a GTT buffer.
*
* @bo: The bo to assign the memory to.
* @mm_node: Memory manager node for drm allocator.
* @mem: The region where the bo resides.
*
*/
static uint64_t amdgpu_mm_node_addr(struct ttm_buffer_object *bo,
struct drm_mm_node *mm_node,
struct ttm_mem_reg *mem)
{
uint64_t addr = 0;
if (mm_node->start != AMDGPU_BO_INVALID_OFFSET) {
addr = mm_node->start << PAGE_SHIFT;
addr += bo->bdev->man[mem->mem_type].gpu_offset;
}
return addr;
}
/**
* amdgpu_find_mm_node - Helper function finds the drm_mm_node corresponding to
* @offset. It also modifies the offset to be within the drm_mm_node returned
*
* @mem: The region where the bo resides.
* @offset: The offset that drm_mm_node is used for finding.
*
*/
static struct drm_mm_node *amdgpu_find_mm_node(struct ttm_mem_reg *mem,
unsigned long *offset)
{
struct drm_mm_node *mm_node = mem->mm_node;
while (*offset >= (mm_node->size << PAGE_SHIFT)) {
*offset -= (mm_node->size << PAGE_SHIFT);
++mm_node;
}
return mm_node;
}
/**
* amdgpu_copy_ttm_mem_to_mem - Helper function for copy
*
* The function copies @size bytes from {src->mem + src->offset} to
* {dst->mem + dst->offset}. src->bo and dst->bo could be same BO for a
* move and different for a BO to BO copy.
*
* @f: Returns the last fence if multiple jobs are submitted.
*/
int amdgpu_ttm_copy_mem_to_mem(struct amdgpu_device *adev,
struct amdgpu_copy_mem *src,
struct amdgpu_copy_mem *dst,
uint64_t size,
struct dma_resv *resv,
struct dma_fence **f)
{
struct amdgpu_ring *ring = adev->mman.buffer_funcs_ring;
struct drm_mm_node *src_mm, *dst_mm;
uint64_t src_node_start, dst_node_start, src_node_size,
dst_node_size, src_page_offset, dst_page_offset;
struct dma_fence *fence = NULL;
int r = 0;
const uint64_t GTT_MAX_BYTES = (AMDGPU_GTT_MAX_TRANSFER_SIZE *
AMDGPU_GPU_PAGE_SIZE);
if (!adev->mman.buffer_funcs_enabled) {
DRM_ERROR("Trying to move memory with ring turned off.\n");
return -EINVAL;
}
src_mm = amdgpu_find_mm_node(src->mem, &src->offset);
src_node_start = amdgpu_mm_node_addr(src->bo, src_mm, src->mem) +
src->offset;
src_node_size = (src_mm->size << PAGE_SHIFT) - src->offset;
src_page_offset = src_node_start & (PAGE_SIZE - 1);
dst_mm = amdgpu_find_mm_node(dst->mem, &dst->offset);
dst_node_start = amdgpu_mm_node_addr(dst->bo, dst_mm, dst->mem) +
dst->offset;
dst_node_size = (dst_mm->size << PAGE_SHIFT) - dst->offset;
dst_page_offset = dst_node_start & (PAGE_SIZE - 1);
mutex_lock(&adev->mman.gtt_window_lock);
while (size) {
unsigned long cur_size;
uint64_t from = src_node_start, to = dst_node_start;
struct dma_fence *next;
/* Copy size cannot exceed GTT_MAX_BYTES. So if src or dst
* begins at an offset, then adjust the size accordingly
*/
cur_size = min3(min(src_node_size, dst_node_size), size,
GTT_MAX_BYTES);
if (cur_size + src_page_offset > GTT_MAX_BYTES ||
cur_size + dst_page_offset > GTT_MAX_BYTES)
cur_size -= max(src_page_offset, dst_page_offset);
/* Map only what needs to be accessed. Map src to window 0 and
* dst to window 1
*/
if (src->mem->start == AMDGPU_BO_INVALID_OFFSET) {
r = amdgpu_map_buffer(src->bo, src->mem,
PFN_UP(cur_size + src_page_offset),
src_node_start, 0, ring,
&from);
if (r)
goto error;
/* Adjust the offset because amdgpu_map_buffer returns
* start of mapped page
*/
from += src_page_offset;
}
if (dst->mem->start == AMDGPU_BO_INVALID_OFFSET) {
r = amdgpu_map_buffer(dst->bo, dst->mem,
PFN_UP(cur_size + dst_page_offset),
dst_node_start, 1, ring,
&to);
if (r)
goto error;
to += dst_page_offset;
}
r = amdgpu_copy_buffer(ring, from, to, cur_size,
resv, &next, false, true);
if (r)
goto error;
dma_fence_put(fence);
fence = next;
size -= cur_size;
if (!size)
break;
src_node_size -= cur_size;
if (!src_node_size) {
src_node_start = amdgpu_mm_node_addr(src->bo, ++src_mm,
src->mem);
src_node_size = (src_mm->size << PAGE_SHIFT);
src_page_offset = 0;
} else {
src_node_start += cur_size;
src_page_offset = src_node_start & (PAGE_SIZE - 1);
}
dst_node_size -= cur_size;
if (!dst_node_size) {
dst_node_start = amdgpu_mm_node_addr(dst->bo, ++dst_mm,
dst->mem);
dst_node_size = (dst_mm->size << PAGE_SHIFT);
dst_page_offset = 0;
} else {
dst_node_start += cur_size;
dst_page_offset = dst_node_start & (PAGE_SIZE - 1);
}
}
error:
mutex_unlock(&adev->mman.gtt_window_lock);
if (f)
*f = dma_fence_get(fence);
dma_fence_put(fence);
return r;
}
/**
* amdgpu_move_blit - Copy an entire buffer to another buffer
*
* This is a helper called by amdgpu_bo_move() and amdgpu_move_vram_ram() to
* help move buffers to and from VRAM.
*/
static int amdgpu_move_blit(struct ttm_buffer_object *bo,
bool evict, bool no_wait_gpu,
struct ttm_mem_reg *new_mem,
struct ttm_mem_reg *old_mem)
{
struct amdgpu_device *adev = amdgpu_ttm_adev(bo->bdev);
struct amdgpu_copy_mem src, dst;
struct dma_fence *fence = NULL;
int r;
src.bo = bo;
dst.bo = bo;
src.mem = old_mem;
dst.mem = new_mem;
src.offset = 0;
dst.offset = 0;
r = amdgpu_ttm_copy_mem_to_mem(adev, &src, &dst,
new_mem->num_pages << PAGE_SHIFT,
bo->base.resv, &fence);
if (r)
goto error;
/* clear the space being freed */
if (old_mem->mem_type == TTM_PL_VRAM &&
(ttm_to_amdgpu_bo(bo)->flags &
AMDGPU_GEM_CREATE_VRAM_WIPE_ON_RELEASE)) {
struct dma_fence *wipe_fence = NULL;
r = amdgpu_fill_buffer(ttm_to_amdgpu_bo(bo), AMDGPU_POISON,
NULL, &wipe_fence);
if (r) {
goto error;
} else if (wipe_fence) {
dma_fence_put(fence);
fence = wipe_fence;
}
}
/* Always block for VM page tables before committing the new location */
if (bo->type == ttm_bo_type_kernel)
r = ttm_bo_move_accel_cleanup(bo, fence, true, new_mem);
else
r = ttm_bo_pipeline_move(bo, fence, evict, new_mem);
dma_fence_put(fence);
return r;
error:
if (fence)
dma_fence_wait(fence, false);
dma_fence_put(fence);
return r;
}
/**
* amdgpu_move_vram_ram - Copy VRAM buffer to RAM buffer
*
* Called by amdgpu_bo_move().
*/
static int amdgpu_move_vram_ram(struct ttm_buffer_object *bo, bool evict,
struct ttm_operation_ctx *ctx,
struct ttm_mem_reg *new_mem)
{
struct ttm_mem_reg *old_mem = &bo->mem;
struct ttm_mem_reg tmp_mem;
struct ttm_place placements;
struct ttm_placement placement;
int r;
/* create space/pages for new_mem in GTT space */
tmp_mem = *new_mem;
tmp_mem.mm_node = NULL;
placement.num_placement = 1;
placement.placement = &placements;
placement.num_busy_placement = 1;
placement.busy_placement = &placements;
placements.fpfn = 0;
placements.lpfn = 0;
placements.flags = TTM_PL_MASK_CACHING | TTM_PL_FLAG_TT;
r = ttm_bo_mem_space(bo, &placement, &tmp_mem, ctx);
if (unlikely(r)) {
pr_err("Failed to find GTT space for blit from VRAM\n");
return r;
}
/* set caching flags */
r = ttm_tt_set_placement_caching(bo->ttm, tmp_mem.placement);
if (unlikely(r)) {
goto out_cleanup;
}
/* Bind the memory to the GTT space */
r = ttm_tt_bind(bo->ttm, &tmp_mem, ctx);
if (unlikely(r)) {
goto out_cleanup;
}
/* blit VRAM to GTT */
r = amdgpu_move_blit(bo, evict, ctx->no_wait_gpu, &tmp_mem, old_mem);
if (unlikely(r)) {
goto out_cleanup;
}
/* move BO (in tmp_mem) to new_mem */
r = ttm_bo_move_ttm(bo, ctx, new_mem);
out_cleanup:
ttm_bo_mem_put(bo, &tmp_mem);
return r;
}
/**
* amdgpu_move_ram_vram - Copy buffer from RAM to VRAM
*
* Called by amdgpu_bo_move().
*/
static int amdgpu_move_ram_vram(struct ttm_buffer_object *bo, bool evict,
struct ttm_operation_ctx *ctx,
struct ttm_mem_reg *new_mem)
{
struct ttm_mem_reg *old_mem = &bo->mem;
struct ttm_mem_reg tmp_mem;
struct ttm_placement placement;
struct ttm_place placements;
int r;
/* make space in GTT for old_mem buffer */
tmp_mem = *new_mem;
tmp_mem.mm_node = NULL;
placement.num_placement = 1;
placement.placement = &placements;
placement.num_busy_placement = 1;
placement.busy_placement = &placements;
placements.fpfn = 0;
placements.lpfn = 0;
placements.flags = TTM_PL_MASK_CACHING | TTM_PL_FLAG_TT;
r = ttm_bo_mem_space(bo, &placement, &tmp_mem, ctx);
if (unlikely(r)) {
pr_err("Failed to find GTT space for blit to VRAM\n");
return r;
}
/* move/bind old memory to GTT space */
r = ttm_bo_move_ttm(bo, ctx, &tmp_mem);
if (unlikely(r)) {
goto out_cleanup;
}
/* copy to VRAM */
r = amdgpu_move_blit(bo, evict, ctx->no_wait_gpu, new_mem, old_mem);
if (unlikely(r)) {
goto out_cleanup;
}
out_cleanup:
ttm_bo_mem_put(bo, &tmp_mem);
return r;
}
/**
* amdgpu_mem_visible - Check that memory can be accessed by ttm_bo_move_memcpy
*
* Called by amdgpu_bo_move()
*/
static bool amdgpu_mem_visible(struct amdgpu_device *adev,
struct ttm_mem_reg *mem)
{
struct drm_mm_node *nodes = mem->mm_node;
if (mem->mem_type == TTM_PL_SYSTEM ||
mem->mem_type == TTM_PL_TT)
return true;
if (mem->mem_type != TTM_PL_VRAM)
return false;
/* ttm_mem_reg_ioremap only supports contiguous memory */
if (nodes->size != mem->num_pages)
return false;
return ((nodes->start + nodes->size) << PAGE_SHIFT)
<= adev->gmc.visible_vram_size;
}
/**
* amdgpu_bo_move - Move a buffer object to a new memory location
*
* Called by ttm_bo_handle_move_mem()
*/
static int amdgpu_bo_move(struct ttm_buffer_object *bo, bool evict,
struct ttm_operation_ctx *ctx,
struct ttm_mem_reg *new_mem)
{
struct amdgpu_device *adev;
struct amdgpu_bo *abo;
struct ttm_mem_reg *old_mem = &bo->mem;
int r;
/* Can't move a pinned BO */
abo = ttm_to_amdgpu_bo(bo);
if (WARN_ON_ONCE(abo->pin_count > 0))
return -EINVAL;
adev = amdgpu_ttm_adev(bo->bdev);
if (old_mem->mem_type == TTM_PL_SYSTEM && bo->ttm == NULL) {
amdgpu_move_null(bo, new_mem);
return 0;
}
if ((old_mem->mem_type == TTM_PL_TT &&
new_mem->mem_type == TTM_PL_SYSTEM) ||
(old_mem->mem_type == TTM_PL_SYSTEM &&
new_mem->mem_type == TTM_PL_TT)) {
/* bind is enough */
amdgpu_move_null(bo, new_mem);
return 0;
}
if (old_mem->mem_type == AMDGPU_PL_GDS ||
old_mem->mem_type == AMDGPU_PL_GWS ||
old_mem->mem_type == AMDGPU_PL_OA ||
new_mem->mem_type == AMDGPU_PL_GDS ||
new_mem->mem_type == AMDGPU_PL_GWS ||
new_mem->mem_type == AMDGPU_PL_OA) {
/* Nothing to save here */
amdgpu_move_null(bo, new_mem);
return 0;
}
if (!adev->mman.buffer_funcs_enabled) {
r = -ENODEV;
goto memcpy;
}
if (old_mem->mem_type == TTM_PL_VRAM &&
new_mem->mem_type == TTM_PL_SYSTEM) {
r = amdgpu_move_vram_ram(bo, evict, ctx, new_mem);
} else if (old_mem->mem_type == TTM_PL_SYSTEM &&
new_mem->mem_type == TTM_PL_VRAM) {
r = amdgpu_move_ram_vram(bo, evict, ctx, new_mem);
} else {
r = amdgpu_move_blit(bo, evict, ctx->no_wait_gpu,
new_mem, old_mem);
}
if (r) {
memcpy:
/* Check that all memory is CPU accessible */
if (!amdgpu_mem_visible(adev, old_mem) ||
!amdgpu_mem_visible(adev, new_mem)) {
pr_err("Move buffer fallback to memcpy unavailable\n");
return r;
}
r = ttm_bo_move_memcpy(bo, ctx, new_mem);
if (r)
return r;
}
if (bo->type == ttm_bo_type_device &&
new_mem->mem_type == TTM_PL_VRAM &&
old_mem->mem_type != TTM_PL_VRAM) {
/* amdgpu_bo_fault_reserve_notify will re-set this if the CPU
* accesses the BO after it's moved.
*/
abo->flags &= ~AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED;
}
/* update statistics */
atomic64_add((u64)bo->num_pages << PAGE_SHIFT, &adev->num_bytes_moved);
return 0;
}
/**
* amdgpu_ttm_io_mem_reserve - Reserve a block of memory during a fault
*
* Called by ttm_mem_io_reserve() ultimately via ttm_bo_vm_fault()
*/
static int amdgpu_ttm_io_mem_reserve(struct ttm_bo_device *bdev, struct ttm_mem_reg *mem)
{
struct ttm_mem_type_manager *man = &bdev->man[mem->mem_type];
struct amdgpu_device *adev = amdgpu_ttm_adev(bdev);
struct drm_mm_node *mm_node = mem->mm_node;
mem->bus.addr = NULL;
mem->bus.offset = 0;
mem->bus.size = mem->num_pages << PAGE_SHIFT;
mem->bus.base = 0;
mem->bus.is_iomem = false;
if (!(man->flags & TTM_MEMTYPE_FLAG_MAPPABLE))
return -EINVAL;
switch (mem->mem_type) {
case TTM_PL_SYSTEM:
/* system memory */
return 0;
case TTM_PL_TT:
break;
case TTM_PL_VRAM:
mem->bus.offset = mem->start << PAGE_SHIFT;
/* check if it's visible */
if ((mem->bus.offset + mem->bus.size) > adev->gmc.visible_vram_size)
return -EINVAL;
/* Only physically contiguous buffers apply. In a contiguous
* buffer, size of the first mm_node would match the number of
* pages in ttm_mem_reg.
*/
if (adev->mman.aper_base_kaddr &&
(mm_node->size == mem->num_pages))
mem->bus.addr = (u8 *)adev->mman.aper_base_kaddr +
mem->bus.offset;
mem->bus.base = adev->gmc.aper_base;
mem->bus.is_iomem = true;
break;
default:
return -EINVAL;
}
return 0;
}
static void amdgpu_ttm_io_mem_free(struct ttm_bo_device *bdev, struct ttm_mem_reg *mem)
{
}
static unsigned long amdgpu_ttm_io_mem_pfn(struct ttm_buffer_object *bo,
unsigned long page_offset)
{
struct drm_mm_node *mm;
unsigned long offset = (page_offset << PAGE_SHIFT);
mm = amdgpu_find_mm_node(&bo->mem, &offset);
return (bo->mem.bus.base >> PAGE_SHIFT) + mm->start +
(offset >> PAGE_SHIFT);
}
/*
* TTM backend functions.
*/
struct amdgpu_ttm_tt {
struct ttm_dma_tt ttm;
struct drm_gem_object *gobj;
u64 offset;
uint64_t userptr;
struct task_struct *usertask;
uint32_t userflags;
#if IS_ENABLED(CONFIG_DRM_AMDGPU_USERPTR)
struct hmm_range *range;
#endif
};
#ifdef CONFIG_DRM_AMDGPU_USERPTR
/* flags used by HMM internal, not related to CPU/GPU PTE flags */
static const uint64_t hmm_range_flags[HMM_PFN_FLAG_MAX] = {
(1 << 0), /* HMM_PFN_VALID */
(1 << 1), /* HMM_PFN_WRITE */
0 /* HMM_PFN_DEVICE_PRIVATE */
};
static const uint64_t hmm_range_values[HMM_PFN_VALUE_MAX] = {
0xfffffffffffffffeUL, /* HMM_PFN_ERROR */
0, /* HMM_PFN_NONE */
0xfffffffffffffffcUL /* HMM_PFN_SPECIAL */
};
/**
* amdgpu_ttm_tt_get_user_pages - get device accessible pages that back user
* memory and start HMM tracking CPU page table update
*
* Calling function must call amdgpu_ttm_tt_userptr_range_done() once and only
* once afterwards to stop HMM tracking
*/
int amdgpu_ttm_tt_get_user_pages(struct amdgpu_bo *bo, struct page **pages)
{
struct ttm_tt *ttm = bo->tbo.ttm;
struct amdgpu_ttm_tt *gtt = (void *)ttm;
unsigned long start = gtt->userptr;
struct vm_area_struct *vma;
struct hmm_range *range;
unsigned long timeout;
struct mm_struct *mm;
unsigned long i;
int r = 0;
mm = bo->notifier.mm;
if (unlikely(!mm)) {
DRM_DEBUG_DRIVER("BO is not registered?\n");
return -EFAULT;
}
/* Another get_user_pages is running at the same time?? */
if (WARN_ON(gtt->range))
return -EFAULT;
if (!mmget_not_zero(mm)) /* Happens during process shutdown */
return -ESRCH;
range = kzalloc(sizeof(*range), GFP_KERNEL);
if (unlikely(!range)) {
r = -ENOMEM;
goto out;
}
range->notifier = &bo->notifier;
range->flags = hmm_range_flags;
range->values = hmm_range_values;
range->pfn_shift = PAGE_SHIFT;
range->start = bo->notifier.interval_tree.start;
range->end = bo->notifier.interval_tree.last + 1;
range->default_flags = hmm_range_flags[HMM_PFN_VALID];
if (!amdgpu_ttm_tt_is_readonly(ttm))
range->default_flags |= range->flags[HMM_PFN_WRITE];
range->pfns = kvmalloc_array(ttm->num_pages, sizeof(*range->pfns),
GFP_KERNEL);
if (unlikely(!range->pfns)) {
r = -ENOMEM;
goto out_free_ranges;
}
down_read(&mm->mmap_sem);
vma = find_vma(mm, start);
if (unlikely(!vma || start < vma->vm_start)) {
r = -EFAULT;
goto out_unlock;
}
if (unlikely((gtt->userflags & AMDGPU_GEM_USERPTR_ANONONLY) &&
vma->vm_file)) {
r = -EPERM;
goto out_unlock;
}
up_read(&mm->mmap_sem);
timeout = jiffies + msecs_to_jiffies(HMM_RANGE_DEFAULT_TIMEOUT);
retry:
range->notifier_seq = mmu_interval_read_begin(&bo->notifier);
down_read(&mm->mmap_sem);
r = hmm_range_fault(range, 0);
up_read(&mm->mmap_sem);
if (unlikely(r <= 0)) {
/*
* FIXME: This timeout should encompass the retry from
* mmu_interval_read_retry() as well.
*/
if ((r == 0 || r == -EBUSY) && !time_after(jiffies, timeout))
goto retry;
goto out_free_pfns;
}
for (i = 0; i < ttm->num_pages; i++) {
/* FIXME: The pages cannot be touched outside the notifier_lock */
pages[i] = hmm_device_entry_to_page(range, range->pfns[i]);
if (unlikely(!pages[i])) {
pr_err("Page fault failed for pfn[%lu] = 0x%llx\n",
i, range->pfns[i]);
r = -ENOMEM;
goto out_free_pfns;
}
}
gtt->range = range;
mmput(mm);
return 0;
out_unlock:
up_read(&mm->mmap_sem);
out_free_pfns:
kvfree(range->pfns);
out_free_ranges:
kfree(range);
out:
mmput(mm);
return r;
}
/**
* amdgpu_ttm_tt_userptr_range_done - stop HMM track the CPU page table change
* Check if the pages backing this ttm range have been invalidated
*
* Returns: true if pages are still valid
*/
bool amdgpu_ttm_tt_get_user_pages_done(struct ttm_tt *ttm)
{
struct amdgpu_ttm_tt *gtt = (void *)ttm;
bool r = false;
if (!gtt || !gtt->userptr)
return false;
DRM_DEBUG_DRIVER("user_pages_done 0x%llx pages 0x%lx\n",
gtt->userptr, ttm->num_pages);
WARN_ONCE(!gtt->range || !gtt->range->pfns,
"No user pages to check\n");
if (gtt->range) {
/*
* FIXME: Must always hold notifier_lock for this, and must
* not ignore the return code.
*/
r = mmu_interval_read_retry(gtt->range->notifier,
gtt->range->notifier_seq);
kvfree(gtt->range->pfns);
kfree(gtt->range);
gtt->range = NULL;
}
return !r;
}
#endif
/**
* amdgpu_ttm_tt_set_user_pages - Copy pages in, putting old pages as necessary.
*
* Called by amdgpu_cs_list_validate(). This creates the page list
* that backs user memory and will ultimately be mapped into the device
* address space.
*/
void amdgpu_ttm_tt_set_user_pages(struct ttm_tt *ttm, struct page **pages)
{
unsigned long i;
for (i = 0; i < ttm->num_pages; ++i)
ttm->pages[i] = pages ? pages[i] : NULL;
}
/**
* amdgpu_ttm_tt_pin_userptr - prepare the sg table with the user pages
*
* Called by amdgpu_ttm_backend_bind()
**/
static int amdgpu_ttm_tt_pin_userptr(struct ttm_tt *ttm)
{
struct amdgpu_device *adev = amdgpu_ttm_adev(ttm->bdev);
struct amdgpu_ttm_tt *gtt = (void *)ttm;
unsigned nents;
int r;
int write = !(gtt->userflags & AMDGPU_GEM_USERPTR_READONLY);
enum dma_data_direction direction = write ?
DMA_BIDIRECTIONAL : DMA_TO_DEVICE;
/* Allocate an SG array and squash pages into it */
r = sg_alloc_table_from_pages(ttm->sg, ttm->pages, ttm->num_pages, 0,
ttm->num_pages << PAGE_SHIFT,
GFP_KERNEL);
if (r)
goto release_sg;
/* Map SG to device */
r = -ENOMEM;
nents = dma_map_sg(adev->dev, ttm->sg->sgl, ttm->sg->nents, direction);
if (nents != ttm->sg->nents)
goto release_sg;
/* convert SG to linear array of pages and dma addresses */
drm_prime_sg_to_page_addr_arrays(ttm->sg, ttm->pages,
gtt->ttm.dma_address, ttm->num_pages);
return 0;
release_sg:
kfree(ttm->sg);
return r;
}
/**
* amdgpu_ttm_tt_unpin_userptr - Unpin and unmap userptr pages
*/
static void amdgpu_ttm_tt_unpin_userptr(struct ttm_tt *ttm)
{
struct amdgpu_device *adev = amdgpu_ttm_adev(ttm->bdev);
struct amdgpu_ttm_tt *gtt = (void *)ttm;
int write = !(gtt->userflags & AMDGPU_GEM_USERPTR_READONLY);
enum dma_data_direction direction = write ?
DMA_BIDIRECTIONAL : DMA_TO_DEVICE;
/* double check that we don't free the table twice */
if (!ttm->sg->sgl)
return;
/* unmap the pages mapped to the device */
dma_unmap_sg(adev->dev, ttm->sg->sgl, ttm->sg->nents, direction);
sg_free_table(ttm->sg);
#if IS_ENABLED(CONFIG_DRM_AMDGPU_USERPTR)
if (gtt->range) {
unsigned long i;
for (i = 0; i < ttm->num_pages; i++) {
if (ttm->pages[i] !=
hmm_device_entry_to_page(gtt->range,
gtt->range->pfns[i]))
break;
}
WARN((i == ttm->num_pages), "Missing get_user_page_done\n");
}
#endif
}
int amdgpu_ttm_gart_bind(struct amdgpu_device *adev,
struct ttm_buffer_object *tbo,
uint64_t flags)
{
struct amdgpu_bo *abo = ttm_to_amdgpu_bo(tbo);
struct ttm_tt *ttm = tbo->ttm;
struct amdgpu_ttm_tt *gtt = (void *)ttm;
int r;
if (abo->flags & AMDGPU_GEM_CREATE_MQD_GFX9) {
uint64_t page_idx = 1;
r = amdgpu_gart_bind(adev, gtt->offset, page_idx,
ttm->pages, gtt->ttm.dma_address, flags);
if (r)
goto gart_bind_fail;
/* Patch mtype of the second part BO */
flags &= ~AMDGPU_PTE_MTYPE_VG10_MASK;
flags |= AMDGPU_PTE_MTYPE_VG10(AMDGPU_MTYPE_NC);
r = amdgpu_gart_bind(adev,
gtt->offset + (page_idx << PAGE_SHIFT),
ttm->num_pages - page_idx,
&ttm->pages[page_idx],
&(gtt->ttm.dma_address[page_idx]), flags);
} else {
r = amdgpu_gart_bind(adev, gtt->offset, ttm->num_pages,
ttm->pages, gtt->ttm.dma_address, flags);
}
gart_bind_fail:
if (r)
DRM_ERROR("failed to bind %lu pages at 0x%08llX\n",
ttm->num_pages, gtt->offset);
return r;
}
/**
* amdgpu_ttm_backend_bind - Bind GTT memory
*
* Called by ttm_tt_bind() on behalf of ttm_bo_handle_move_mem().
* This handles binding GTT memory to the device address space.
*/
static int amdgpu_ttm_backend_bind(struct ttm_tt *ttm,
struct ttm_mem_reg *bo_mem)
{
struct amdgpu_device *adev = amdgpu_ttm_adev(ttm->bdev);
struct amdgpu_ttm_tt *gtt = (void*)ttm;
uint64_t flags;
int r = 0;
if (gtt->userptr) {
r = amdgpu_ttm_tt_pin_userptr(ttm);
if (r) {
DRM_ERROR("failed to pin userptr\n");
return r;
}
}
if (!ttm->num_pages) {
WARN(1, "nothing to bind %lu pages for mreg %p back %p!\n",
ttm->num_pages, bo_mem, ttm);
}
if (bo_mem->mem_type == AMDGPU_PL_GDS ||
bo_mem->mem_type == AMDGPU_PL_GWS ||
bo_mem->mem_type == AMDGPU_PL_OA)
return -EINVAL;
if (!amdgpu_gtt_mgr_has_gart_addr(bo_mem)) {
gtt->offset = AMDGPU_BO_INVALID_OFFSET;
return 0;
}
/* compute PTE flags relevant to this BO memory */
flags = amdgpu_ttm_tt_pte_flags(adev, ttm, bo_mem);
/* bind pages into GART page tables */
gtt->offset = (u64)bo_mem->start << PAGE_SHIFT;
r = amdgpu_gart_bind(adev, gtt->offset, ttm->num_pages,
ttm->pages, gtt->ttm.dma_address, flags);
if (r)
DRM_ERROR("failed to bind %lu pages at 0x%08llX\n",
ttm->num_pages, gtt->offset);
return r;
}
/**
* amdgpu_ttm_alloc_gart - Allocate GART memory for buffer object
*/
int amdgpu_ttm_alloc_gart(struct ttm_buffer_object *bo)
{
struct amdgpu_device *adev = amdgpu_ttm_adev(bo->bdev);
struct ttm_operation_ctx ctx = { false, false };
struct amdgpu_ttm_tt *gtt = (void*)bo->ttm;
struct ttm_mem_reg tmp;
struct ttm_placement placement;
struct ttm_place placements;
uint64_t addr, flags;
int r;
if (bo->mem.start != AMDGPU_BO_INVALID_OFFSET)
return 0;
addr = amdgpu_gmc_agp_addr(bo);
if (addr != AMDGPU_BO_INVALID_OFFSET) {
bo->mem.start = addr >> PAGE_SHIFT;
} else {
/* allocate GART space */
tmp = bo->mem;
tmp.mm_node = NULL;
placement.num_placement = 1;
placement.placement = &placements;
placement.num_busy_placement = 1;
placement.busy_placement = &placements;
placements.fpfn = 0;
placements.lpfn = adev->gmc.gart_size >> PAGE_SHIFT;
placements.flags = (bo->mem.placement & ~TTM_PL_MASK_MEM) |
TTM_PL_FLAG_TT;
r = ttm_bo_mem_space(bo, &placement, &tmp, &ctx);
if (unlikely(r))
return r;
/* compute PTE flags for this buffer object */
flags = amdgpu_ttm_tt_pte_flags(adev, bo->ttm, &tmp);
/* Bind pages */
gtt->offset = (u64)tmp.start << PAGE_SHIFT;
r = amdgpu_ttm_gart_bind(adev, bo, flags);
if (unlikely(r)) {
ttm_bo_mem_put(bo, &tmp);
return r;
}
ttm_bo_mem_put(bo, &bo->mem);
bo->mem = tmp;
}
bo->offset = (bo->mem.start << PAGE_SHIFT) +
bo->bdev->man[bo->mem.mem_type].gpu_offset;
return 0;
}
/**
* amdgpu_ttm_recover_gart - Rebind GTT pages
*
* Called by amdgpu_gtt_mgr_recover() from amdgpu_device_reset() to
* rebind GTT pages during a GPU reset.
*/
int amdgpu_ttm_recover_gart(struct ttm_buffer_object *tbo)
{
struct amdgpu_device *adev = amdgpu_ttm_adev(tbo->bdev);
uint64_t flags;
int r;
if (!tbo->ttm)
return 0;
flags = amdgpu_ttm_tt_pte_flags(adev, tbo->ttm, &tbo->mem);
r = amdgpu_ttm_gart_bind(adev, tbo, flags);
return r;
}
/**
* amdgpu_ttm_backend_unbind - Unbind GTT mapped pages
*
* Called by ttm_tt_unbind() on behalf of ttm_bo_move_ttm() and
* ttm_tt_destroy().
*/
static int amdgpu_ttm_backend_unbind(struct ttm_tt *ttm)
{
struct amdgpu_device *adev = amdgpu_ttm_adev(ttm->bdev);
struct amdgpu_ttm_tt *gtt = (void *)ttm;
int r;
/* if the pages have userptr pinning then clear that first */
if (gtt->userptr)
amdgpu_ttm_tt_unpin_userptr(ttm);
if (gtt->offset == AMDGPU_BO_INVALID_OFFSET)
return 0;
/* unbind shouldn't be done for GDS/GWS/OA in ttm_bo_clean_mm */
r = amdgpu_gart_unbind(adev, gtt->offset, ttm->num_pages);
if (r)
DRM_ERROR("failed to unbind %lu pages at 0x%08llX\n",
gtt->ttm.ttm.num_pages, gtt->offset);
return r;
}
static void amdgpu_ttm_backend_destroy(struct ttm_tt *ttm)
{
struct amdgpu_ttm_tt *gtt = (void *)ttm;
if (gtt->usertask)
put_task_struct(gtt->usertask);
ttm_dma_tt_fini(>t->ttm);
kfree(gtt);
}
static struct ttm_backend_func amdgpu_backend_func = {
.bind = &amdgpu_ttm_backend_bind,
.unbind = &amdgpu_ttm_backend_unbind,
.destroy = &amdgpu_ttm_backend_destroy,
};
/**
* amdgpu_ttm_tt_create - Create a ttm_tt object for a given BO
*
* @bo: The buffer object to create a GTT ttm_tt object around
*
* Called by ttm_tt_create().
*/
static struct ttm_tt *amdgpu_ttm_tt_create(struct ttm_buffer_object *bo,
uint32_t page_flags)
{
struct amdgpu_ttm_tt *gtt;
gtt = kzalloc(sizeof(struct amdgpu_ttm_tt), GFP_KERNEL);
if (gtt == NULL) {
return NULL;
}
gtt->ttm.ttm.func = &amdgpu_backend_func;
gtt->gobj = &bo->base;
/* allocate space for the uninitialized page entries */
if (ttm_sg_tt_init(>t->ttm, bo, page_flags)) {
kfree(gtt);
return NULL;
}
return >t->ttm.ttm;
}
/**
* amdgpu_ttm_tt_populate - Map GTT pages visible to the device
*
* Map the pages of a ttm_tt object to an address space visible
* to the underlying device.
*/
static int amdgpu_ttm_tt_populate(struct ttm_tt *ttm,
struct ttm_operation_ctx *ctx)
{
struct amdgpu_device *adev = amdgpu_ttm_adev(ttm->bdev);
struct amdgpu_ttm_tt *gtt = (void *)ttm;
/* user pages are bound by amdgpu_ttm_tt_pin_userptr() */
if (gtt && gtt->userptr) {
ttm->sg = kzalloc(sizeof(struct sg_table), GFP_KERNEL);
if (!ttm->sg)
return -ENOMEM;
ttm->page_flags |= TTM_PAGE_FLAG_SG;
ttm->state = tt_unbound;
return 0;
}
if (ttm->page_flags & TTM_PAGE_FLAG_SG) {
if (!ttm->sg) {
struct dma_buf_attachment *attach;
struct sg_table *sgt;
attach = gtt->gobj->import_attach;
sgt = dma_buf_map_attachment(attach, DMA_BIDIRECTIONAL);
if (IS_ERR(sgt))
return PTR_ERR(sgt);
ttm->sg = sgt;
}
drm_prime_sg_to_page_addr_arrays(ttm->sg, ttm->pages,
gtt->ttm.dma_address,
ttm->num_pages);
ttm->state = tt_unbound;
return 0;
}
#ifdef CONFIG_SWIOTLB
if (adev->need_swiotlb && swiotlb_nr_tbl()) {
return ttm_dma_populate(>t->ttm, adev->dev, ctx);
}
#endif
/* fall back to generic helper to populate the page array
* and map them to the device */
return ttm_populate_and_map_pages(adev->dev, >t->ttm, ctx);
}
/**
* amdgpu_ttm_tt_unpopulate - unmap GTT pages and unpopulate page arrays
*
* Unmaps pages of a ttm_tt object from the device address space and
* unpopulates the page array backing it.
*/
static void amdgpu_ttm_tt_unpopulate(struct ttm_tt *ttm)
{
struct amdgpu_ttm_tt *gtt = (void *)ttm;
struct amdgpu_device *adev;
if (gtt && gtt->userptr) {
amdgpu_ttm_tt_set_user_pages(ttm, NULL);
kfree(ttm->sg);
ttm->page_flags &= ~TTM_PAGE_FLAG_SG;
return;
}
if (ttm->sg && gtt->gobj->import_attach) {
struct dma_buf_attachment *attach;
attach = gtt->gobj->import_attach;
dma_buf_unmap_attachment(attach, ttm->sg, DMA_BIDIRECTIONAL);
ttm->sg = NULL;
return;
}
if (ttm->page_flags & TTM_PAGE_FLAG_SG)
return;
adev = amdgpu_ttm_adev(ttm->bdev);
#ifdef CONFIG_SWIOTLB
if (adev->need_swiotlb && swiotlb_nr_tbl()) {
ttm_dma_unpopulate(>t->ttm, adev->dev);
return;
}
#endif
/* fall back to generic helper to unmap and unpopulate array */
ttm_unmap_and_unpopulate_pages(adev->dev, >t->ttm);
}
/**
* amdgpu_ttm_tt_set_userptr - Initialize userptr GTT ttm_tt for the current
* task
*
* @ttm: The ttm_tt object to bind this userptr object to
* @addr: The address in the current tasks VM space to use
* @flags: Requirements of userptr object.
*
* Called by amdgpu_gem_userptr_ioctl() to bind userptr pages
* to current task
*/
int amdgpu_ttm_tt_set_userptr(struct ttm_tt *ttm, uint64_t addr,
uint32_t flags)
{
struct amdgpu_ttm_tt *gtt = (void *)ttm;
if (gtt == NULL)
return -EINVAL;
gtt->userptr = addr;
gtt->userflags = flags;
if (gtt->usertask)
put_task_struct(gtt->usertask);
gtt->usertask = current->group_leader;
get_task_struct(gtt->usertask);
return 0;
}
/**
* amdgpu_ttm_tt_get_usermm - Return memory manager for ttm_tt object
*/
struct mm_struct *amdgpu_ttm_tt_get_usermm(struct ttm_tt *ttm)
{
struct amdgpu_ttm_tt *gtt = (void *)ttm;
if (gtt == NULL)
return NULL;
if (gtt->usertask == NULL)
return NULL;
return gtt->usertask->mm;
}
/**
* amdgpu_ttm_tt_affect_userptr - Determine if a ttm_tt object lays inside an
* address range for the current task.
*
*/
bool amdgpu_ttm_tt_affect_userptr(struct ttm_tt *ttm, unsigned long start,
unsigned long end)
{
struct amdgpu_ttm_tt *gtt = (void *)ttm;
unsigned long size;
if (gtt == NULL || !gtt->userptr)
return false;
/* Return false if no part of the ttm_tt object lies within
* the range
*/
size = (unsigned long)gtt->ttm.ttm.num_pages * PAGE_SIZE;
if (gtt->userptr > end || gtt->userptr + size <= start)
return false;
return true;
}
/**
* amdgpu_ttm_tt_is_userptr - Have the pages backing by userptr?
*/
bool amdgpu_ttm_tt_is_userptr(struct ttm_tt *ttm)
{
struct amdgpu_ttm_tt *gtt = (void *)ttm;
if (gtt == NULL || !gtt->userptr)
return false;
return true;
}
/**
* amdgpu_ttm_tt_is_readonly - Is the ttm_tt object read only?
*/
bool amdgpu_ttm_tt_is_readonly(struct ttm_tt *ttm)
{
struct amdgpu_ttm_tt *gtt = (void *)ttm;
if (gtt == NULL)
return false;
return !!(gtt->userflags & AMDGPU_GEM_USERPTR_READONLY);
}
/**
* amdgpu_ttm_tt_pde_flags - Compute PDE flags for ttm_tt object
*
* @ttm: The ttm_tt object to compute the flags for
* @mem: The memory registry backing this ttm_tt object
*
* Figure out the flags to use for a VM PDE (Page Directory Entry).
*/
uint64_t amdgpu_ttm_tt_pde_flags(struct ttm_tt *ttm, struct ttm_mem_reg *mem)
{
uint64_t flags = 0;
if (mem && mem->mem_type != TTM_PL_SYSTEM)
flags |= AMDGPU_PTE_VALID;
if (mem && mem->mem_type == TTM_PL_TT) {
flags |= AMDGPU_PTE_SYSTEM;
if (ttm->caching_state == tt_cached)
flags |= AMDGPU_PTE_SNOOPED;
}
return flags;
}
/**
* amdgpu_ttm_tt_pte_flags - Compute PTE flags for ttm_tt object
*
* @ttm: The ttm_tt object to compute the flags for
* @mem: The memory registry backing this ttm_tt object
* Figure out the flags to use for a VM PTE (Page Table Entry).
*/
uint64_t amdgpu_ttm_tt_pte_flags(struct amdgpu_device *adev, struct ttm_tt *ttm,
struct ttm_mem_reg *mem)
{
uint64_t flags = amdgpu_ttm_tt_pde_flags(ttm, mem);
flags |= adev->gart.gart_pte_flags;
flags |= AMDGPU_PTE_READABLE;
if (!amdgpu_ttm_tt_is_readonly(ttm))
flags |= AMDGPU_PTE_WRITEABLE;
return flags;
}
/**
* amdgpu_ttm_bo_eviction_valuable - Check to see if we can evict a buffer
* object.
*
* Return true if eviction is sensible. Called by ttm_mem_evict_first() on
* behalf of ttm_bo_mem_force_space() which tries to evict buffer objects until
* it can find space for a new object and by ttm_bo_force_list_clean() which is
* used to clean out a memory space.
*/
static bool amdgpu_ttm_bo_eviction_valuable(struct ttm_buffer_object *bo,
const struct ttm_place *place)
{
unsigned long num_pages = bo->mem.num_pages;
struct drm_mm_node *node = bo->mem.mm_node;
struct dma_resv_list *flist;
struct dma_fence *f;
int i;
if (bo->type == ttm_bo_type_kernel &&
!amdgpu_vm_evictable(ttm_to_amdgpu_bo(bo)))
return false;
/* If bo is a KFD BO, check if the bo belongs to the current process.
* If true, then return false as any KFD process needs all its BOs to
* be resident to run successfully
*/
flist = dma_resv_get_list(bo->base.resv);
if (flist) {
for (i = 0; i < flist->shared_count; ++i) {
f = rcu_dereference_protected(flist->shared[i],
dma_resv_held(bo->base.resv));
if (amdkfd_fence_check_mm(f, current->mm))
return false;
}
}
switch (bo->mem.mem_type) {
case TTM_PL_TT:
return true;
case TTM_PL_VRAM:
/* Check each drm MM node individually */
while (num_pages) {
if (place->fpfn < (node->start + node->size) &&
!(place->lpfn && place->lpfn <= node->start))
return true;
num_pages -= node->size;
++node;
}
return false;
default:
break;
}
return ttm_bo_eviction_valuable(bo, place);
}
/**
* amdgpu_ttm_access_memory - Read or Write memory that backs a buffer object.
*
* @bo: The buffer object to read/write
* @offset: Offset into buffer object
* @buf: Secondary buffer to write/read from
* @len: Length in bytes of access
* @write: true if writing
*
* This is used to access VRAM that backs a buffer object via MMIO
* access for debugging purposes.
*/
static int amdgpu_ttm_access_memory(struct ttm_buffer_object *bo,
unsigned long offset,
void *buf, int len, int write)
{
struct amdgpu_bo *abo = ttm_to_amdgpu_bo(bo);
struct amdgpu_device *adev = amdgpu_ttm_adev(abo->tbo.bdev);
struct drm_mm_node *nodes;
uint32_t value = 0;
int ret = 0;
uint64_t pos;
unsigned long flags;
if (bo->mem.mem_type != TTM_PL_VRAM)
return -EIO;
nodes = amdgpu_find_mm_node(&abo->tbo.mem, &offset);
pos = (nodes->start << PAGE_SHIFT) + offset;
while (len && pos < adev->gmc.mc_vram_size) {
uint64_t aligned_pos = pos & ~(uint64_t)3;
uint32_t bytes = 4 - (pos & 3);
uint32_t shift = (pos & 3) * 8;
uint32_t mask = 0xffffffff << shift;
if (len < bytes) {
mask &= 0xffffffff >> (bytes - len) * 8;
bytes = len;
}
spin_lock_irqsave(&adev->mmio_idx_lock, flags);
WREG32_NO_KIQ(mmMM_INDEX, ((uint32_t)aligned_pos) | 0x80000000);
WREG32_NO_KIQ(mmMM_INDEX_HI, aligned_pos >> 31);
if (!write || mask != 0xffffffff)
value = RREG32_NO_KIQ(mmMM_DATA);
if (write) {
value &= ~mask;
value |= (*(uint32_t *)buf << shift) & mask;
WREG32_NO_KIQ(mmMM_DATA, value);
}
spin_unlock_irqrestore(&adev->mmio_idx_lock, flags);
if (!write) {
value = (value & mask) >> shift;
memcpy(buf, &value, bytes);
}
ret += bytes;
buf = (uint8_t *)buf + bytes;
pos += bytes;
len -= bytes;
if (pos >= (nodes->start + nodes->size) << PAGE_SHIFT) {
++nodes;
pos = (nodes->start << PAGE_SHIFT);
}
}
return ret;
}
static struct ttm_bo_driver amdgpu_bo_driver = {
.ttm_tt_create = &amdgpu_ttm_tt_create,
.ttm_tt_populate = &amdgpu_ttm_tt_populate,
.ttm_tt_unpopulate = &amdgpu_ttm_tt_unpopulate,
.invalidate_caches = &amdgpu_invalidate_caches,
.init_mem_type = &amdgpu_init_mem_type,
.eviction_valuable = amdgpu_ttm_bo_eviction_valuable,
.evict_flags = &amdgpu_evict_flags,
.move = &amdgpu_bo_move,
.verify_access = &amdgpu_verify_access,
.move_notify = &amdgpu_bo_move_notify,
.release_notify = &amdgpu_bo_release_notify,
.fault_reserve_notify = &amdgpu_bo_fault_reserve_notify,
.io_mem_reserve = &amdgpu_ttm_io_mem_reserve,
.io_mem_free = &amdgpu_ttm_io_mem_free,
.io_mem_pfn = amdgpu_ttm_io_mem_pfn,
.access_memory = &amdgpu_ttm_access_memory,
.del_from_lru_notify = &amdgpu_vm_del_from_lru_notify
};
/*
* Firmware Reservation functions
*/
/**
* amdgpu_ttm_fw_reserve_vram_fini - free fw reserved vram
*
* @adev: amdgpu_device pointer
*
* free fw reserved vram if it has been reserved.
*/
static void amdgpu_ttm_fw_reserve_vram_fini(struct amdgpu_device *adev)
{
amdgpu_bo_free_kernel(&adev->fw_vram_usage.reserved_bo,
NULL, &adev->fw_vram_usage.va);
}
/**
* amdgpu_ttm_fw_reserve_vram_init - create bo vram reservation from fw
*
* @adev: amdgpu_device pointer
*
* create bo vram reservation from fw.
*/
static int amdgpu_ttm_fw_reserve_vram_init(struct amdgpu_device *adev)
{
uint64_t vram_size = adev->gmc.visible_vram_size;
adev->fw_vram_usage.va = NULL;
adev->fw_vram_usage.reserved_bo = NULL;
if (adev->fw_vram_usage.size == 0 ||
adev->fw_vram_usage.size > vram_size)
return 0;
return amdgpu_bo_create_kernel_at(adev,
adev->fw_vram_usage.start_offset,
adev->fw_vram_usage.size,
AMDGPU_GEM_DOMAIN_VRAM,
&adev->fw_vram_usage.reserved_bo,
&adev->fw_vram_usage.va);
}
/*
* Memoy training reservation functions
*/
/**
* amdgpu_ttm_training_reserve_vram_fini - free memory training reserved vram
*
* @adev: amdgpu_device pointer
*
* free memory training reserved vram if it has been reserved.
*/
static int amdgpu_ttm_training_reserve_vram_fini(struct amdgpu_device *adev)
{
struct psp_memory_training_context *ctx = &adev->psp.mem_train_ctx;
ctx->init = PSP_MEM_TRAIN_NOT_SUPPORT;
amdgpu_bo_free_kernel(&ctx->c2p_bo, NULL, NULL);
ctx->c2p_bo = NULL;
amdgpu_bo_free_kernel(&ctx->p2c_bo, NULL, NULL);
ctx->p2c_bo = NULL;
return 0;
}
/**
* amdgpu_ttm_training_reserve_vram_init - create bo vram reservation from memory training
*
* @adev: amdgpu_device pointer
*
* create bo vram reservation from memory training.
*/
static int amdgpu_ttm_training_reserve_vram_init(struct amdgpu_device *adev)
{
int ret;
struct psp_memory_training_context *ctx = &adev->psp.mem_train_ctx;
memset(ctx, 0, sizeof(*ctx));
if (!adev->fw_vram_usage.mem_train_support) {
DRM_DEBUG("memory training does not support!\n");
return 0;
}
ctx->c2p_train_data_offset = adev->fw_vram_usage.mem_train_fb_loc;
ctx->p2c_train_data_offset = (adev->gmc.mc_vram_size - GDDR6_MEM_TRAINING_OFFSET);
ctx->train_data_size = GDDR6_MEM_TRAINING_DATA_SIZE_IN_BYTES;
DRM_DEBUG("train_data_size:%llx,p2c_train_data_offset:%llx,c2p_train_data_offset:%llx.\n",
ctx->train_data_size,
ctx->p2c_train_data_offset,
ctx->c2p_train_data_offset);
ret = amdgpu_bo_create_kernel_at(adev,
ctx->p2c_train_data_offset,
ctx->train_data_size,
AMDGPU_GEM_DOMAIN_VRAM,
&ctx->p2c_bo,
NULL);
if (ret) {
DRM_ERROR("alloc p2c_bo failed(%d)!\n", ret);
goto Err_out;
}
ret = amdgpu_bo_create_kernel_at(adev,
ctx->c2p_train_data_offset,
ctx->train_data_size,
AMDGPU_GEM_DOMAIN_VRAM,
&ctx->c2p_bo,
NULL);
if (ret) {
DRM_ERROR("alloc c2p_bo failed(%d)!\n", ret);
goto Err_out;
}
ctx->init = PSP_MEM_TRAIN_RESERVE_SUCCESS;
return 0;
Err_out:
amdgpu_ttm_training_reserve_vram_fini(adev);
return ret;
}
/**
* amdgpu_ttm_init - Init the memory management (ttm) as well as various
* gtt/vram related fields.
*
* This initializes all of the memory space pools that the TTM layer
* will need such as the GTT space (system memory mapped to the device),
* VRAM (on-board memory), and on-chip memories (GDS, GWS, OA) which
* can be mapped per VMID.
*/
int amdgpu_ttm_init(struct amdgpu_device *adev)
{
uint64_t gtt_size;
int r;
u64 vis_vram_limit;
void *stolen_vga_buf;
mutex_init(&adev->mman.gtt_window_lock);
/* No others user of address space so set it to 0 */
r = ttm_bo_device_init(&adev->mman.bdev,
&amdgpu_bo_driver,
adev->ddev->anon_inode->i_mapping,
adev->ddev->vma_offset_manager,
dma_addressing_limited(adev->dev));
if (r) {
DRM_ERROR("failed initializing buffer object driver(%d).\n", r);
return r;
}
adev->mman.initialized = true;
/* We opt to avoid OOM on system pages allocations */
adev->mman.bdev.no_retry = true;
/* Initialize VRAM pool with all of VRAM divided into pages */
r = ttm_bo_init_mm(&adev->mman.bdev, TTM_PL_VRAM,
adev->gmc.real_vram_size >> PAGE_SHIFT);
if (r) {
DRM_ERROR("Failed initializing VRAM heap.\n");
return r;
}
/* Reduce size of CPU-visible VRAM if requested */
vis_vram_limit = (u64)amdgpu_vis_vram_limit * 1024 * 1024;
if (amdgpu_vis_vram_limit > 0 &&
vis_vram_limit <= adev->gmc.visible_vram_size)
adev->gmc.visible_vram_size = vis_vram_limit;
/* Change the size here instead of the init above so only lpfn is affected */
amdgpu_ttm_set_buffer_funcs_status(adev, false);
#ifdef CONFIG_64BIT
adev->mman.aper_base_kaddr = ioremap_wc(adev->gmc.aper_base,
adev->gmc.visible_vram_size);
#endif
/*
*The reserved vram for firmware must be pinned to the specified
*place on the VRAM, so reserve it early.
*/
r = amdgpu_ttm_fw_reserve_vram_init(adev);
if (r) {
return r;
}
/*
*The reserved vram for memory training must be pinned to the specified
*place on the VRAM, so reserve it early.
*/
r = amdgpu_ttm_training_reserve_vram_init(adev);
if (r)
return r;
/* allocate memory as required for VGA
* This is used for VGA emulation and pre-OS scanout buffers to
* avoid display artifacts while transitioning between pre-OS
* and driver. */
r = amdgpu_bo_create_kernel(adev, adev->gmc.stolen_size, PAGE_SIZE,
AMDGPU_GEM_DOMAIN_VRAM,
&adev->stolen_vga_memory,
NULL, &stolen_vga_buf);
if (r)
return r;
/*
* reserve one TMR (64K) memory at the top of VRAM which holds
* IP Discovery data and is protected by PSP.
*/
r = amdgpu_bo_create_kernel_at(adev,
adev->gmc.real_vram_size - DISCOVERY_TMR_SIZE,
DISCOVERY_TMR_SIZE,
AMDGPU_GEM_DOMAIN_VRAM,
&adev->discovery_memory,
NULL);
if (r)
return r;
DRM_INFO("amdgpu: %uM of VRAM memory ready\n",
(unsigned) (adev->gmc.real_vram_size / (1024 * 1024)));
/* Compute GTT size, either bsaed on 3/4th the size of RAM size
* or whatever the user passed on module init */
if (amdgpu_gtt_size == -1) {
struct sysinfo si;
si_meminfo(&si);
gtt_size = min(max((AMDGPU_DEFAULT_GTT_SIZE_MB << 20),
adev->gmc.mc_vram_size),
((uint64_t)si.totalram * si.mem_unit * 3/4));
}
else
gtt_size = (uint64_t)amdgpu_gtt_size << 20;
/* Initialize GTT memory pool */
r = ttm_bo_init_mm(&adev->mman.bdev, TTM_PL_TT, gtt_size >> PAGE_SHIFT);
if (r) {
DRM_ERROR("Failed initializing GTT heap.\n");
return r;
}
DRM_INFO("amdgpu: %uM of GTT memory ready.\n",
(unsigned)(gtt_size / (1024 * 1024)));
/* Initialize various on-chip memory pools */
r = ttm_bo_init_mm(&adev->mman.bdev, AMDGPU_PL_GDS,
adev->gds.gds_size);
if (r) {
DRM_ERROR("Failed initializing GDS heap.\n");
return r;
}
r = ttm_bo_init_mm(&adev->mman.bdev, AMDGPU_PL_GWS,
adev->gds.gws_size);
if (r) {
DRM_ERROR("Failed initializing gws heap.\n");
return r;
}
r = ttm_bo_init_mm(&adev->mman.bdev, AMDGPU_PL_OA,
adev->gds.oa_size);
if (r) {
DRM_ERROR("Failed initializing oa heap.\n");
return r;
}
/* Register debugfs entries for amdgpu_ttm */
r = amdgpu_ttm_debugfs_init(adev);
if (r) {
DRM_ERROR("Failed to init debugfs\n");
return r;
}
return 0;
}
/**
* amdgpu_ttm_late_init - Handle any late initialization for amdgpu_ttm
*/
void amdgpu_ttm_late_init(struct amdgpu_device *adev)
{
void *stolen_vga_buf;
/* return the VGA stolen memory (if any) back to VRAM */
amdgpu_bo_free_kernel(&adev->stolen_vga_memory, NULL, &stolen_vga_buf);
}
/**
* amdgpu_ttm_fini - De-initialize the TTM memory pools
*/
void amdgpu_ttm_fini(struct amdgpu_device *adev)
{
if (!adev->mman.initialized)
return;
amdgpu_ttm_debugfs_fini(adev);
amdgpu_ttm_training_reserve_vram_fini(adev);
/* return the IP Discovery TMR memory back to VRAM */
amdgpu_bo_free_kernel(&adev->discovery_memory, NULL, NULL);
amdgpu_ttm_fw_reserve_vram_fini(adev);
if (adev->mman.aper_base_kaddr)
iounmap(adev->mman.aper_base_kaddr);
adev->mman.aper_base_kaddr = NULL;
ttm_bo_clean_mm(&adev->mman.bdev, TTM_PL_VRAM);
ttm_bo_clean_mm(&adev->mman.bdev, TTM_PL_TT);
ttm_bo_clean_mm(&adev->mman.bdev, AMDGPU_PL_GDS);
ttm_bo_clean_mm(&adev->mman.bdev, AMDGPU_PL_GWS);
ttm_bo_clean_mm(&adev->mman.bdev, AMDGPU_PL_OA);
ttm_bo_device_release(&adev->mman.bdev);
adev->mman.initialized = false;
DRM_INFO("amdgpu: ttm finalized\n");
}
/**
* amdgpu_ttm_set_buffer_funcs_status - enable/disable use of buffer functions
*
* @adev: amdgpu_device pointer
* @enable: true when we can use buffer functions.
*
* Enable/disable use of buffer functions during suspend/resume. This should
* only be called at bootup or when userspace isn't running.
*/
void amdgpu_ttm_set_buffer_funcs_status(struct amdgpu_device *adev, bool enable)
{
struct ttm_mem_type_manager *man = &adev->mman.bdev.man[TTM_PL_VRAM];
uint64_t size;
int r;
if (!adev->mman.initialized || adev->in_gpu_reset ||
adev->mman.buffer_funcs_enabled == enable)
return;
if (enable) {
struct amdgpu_ring *ring;
struct drm_gpu_scheduler *sched;
ring = adev->mman.buffer_funcs_ring;
sched = &ring->sched;
r = drm_sched_entity_init(&adev->mman.entity,
DRM_SCHED_PRIORITY_KERNEL, &sched,
1, NULL);
if (r) {
DRM_ERROR("Failed setting up TTM BO move entity (%d)\n",
r);
return;
}
} else {
drm_sched_entity_destroy(&adev->mman.entity);
dma_fence_put(man->move);
man->move = NULL;
}
/* this just adjusts TTM size idea, which sets lpfn to the correct value */
if (enable)
size = adev->gmc.real_vram_size;
else
size = adev->gmc.visible_vram_size;
man->size = size >> PAGE_SHIFT;
adev->mman.buffer_funcs_enabled = enable;
}
int amdgpu_mmap(struct file *filp, struct vm_area_struct *vma)
{
struct drm_file *file_priv = filp->private_data;
struct amdgpu_device *adev = file_priv->minor->dev->dev_private;
if (adev == NULL)
return -EINVAL;
return ttm_bo_mmap(filp, vma, &adev->mman.bdev);
}
static int amdgpu_map_buffer(struct ttm_buffer_object *bo,
struct ttm_mem_reg *mem, unsigned num_pages,
uint64_t offset, unsigned window,
struct amdgpu_ring *ring,
uint64_t *addr)
{
struct amdgpu_ttm_tt *gtt = (void *)bo->ttm;
struct amdgpu_device *adev = ring->adev;
struct ttm_tt *ttm = bo->ttm;
struct amdgpu_job *job;
unsigned num_dw, num_bytes;
dma_addr_t *dma_address;
struct dma_fence *fence;
uint64_t src_addr, dst_addr;
uint64_t flags;
int r;
BUG_ON(adev->mman.buffer_funcs->copy_max_bytes <
AMDGPU_GTT_MAX_TRANSFER_SIZE * 8);
*addr = adev->gmc.gart_start;
*addr += (u64)window * AMDGPU_GTT_MAX_TRANSFER_SIZE *
AMDGPU_GPU_PAGE_SIZE;
num_dw = ALIGN(adev->mman.buffer_funcs->copy_num_dw, 8);
num_bytes = num_pages * 8;
r = amdgpu_job_alloc_with_ib(adev, num_dw * 4 + num_bytes, &job);
if (r)
return r;
src_addr = num_dw * 4;
src_addr += job->ibs[0].gpu_addr;
dst_addr = amdgpu_bo_gpu_offset(adev->gart.bo);
dst_addr += window * AMDGPU_GTT_MAX_TRANSFER_SIZE * 8;
amdgpu_emit_copy_buffer(adev, &job->ibs[0], src_addr,
dst_addr, num_bytes);
amdgpu_ring_pad_ib(ring, &job->ibs[0]);
WARN_ON(job->ibs[0].length_dw > num_dw);
dma_address = >t->ttm.dma_address[offset >> PAGE_SHIFT];
flags = amdgpu_ttm_tt_pte_flags(adev, ttm, mem);
r = amdgpu_gart_map(adev, 0, num_pages, dma_address, flags,
&job->ibs[0].ptr[num_dw]);
if (r)
goto error_free;
r = amdgpu_job_submit(job, &adev->mman.entity,
AMDGPU_FENCE_OWNER_UNDEFINED, &fence);
if (r)
goto error_free;
dma_fence_put(fence);
return r;
error_free:
amdgpu_job_free(job);
return r;
}
int amdgpu_copy_buffer(struct amdgpu_ring *ring, uint64_t src_offset,
uint64_t dst_offset, uint32_t byte_count,
struct dma_resv *resv,
struct dma_fence **fence, bool direct_submit,
bool vm_needs_flush)
{
struct amdgpu_device *adev = ring->adev;
struct amdgpu_job *job;
uint32_t max_bytes;
unsigned num_loops, num_dw;
unsigned i;
int r;
if (direct_submit && !ring->sched.ready) {
DRM_ERROR("Trying to move memory with ring turned off.\n");
return -EINVAL;
}
max_bytes = adev->mman.buffer_funcs->copy_max_bytes;
num_loops = DIV_ROUND_UP(byte_count, max_bytes);
num_dw = ALIGN(num_loops * adev->mman.buffer_funcs->copy_num_dw, 8);
r = amdgpu_job_alloc_with_ib(adev, num_dw * 4, &job);
if (r)
return r;
if (vm_needs_flush) {
job->vm_pd_addr = amdgpu_gmc_pd_addr(adev->gart.bo);
job->vm_needs_flush = true;
}
if (resv) {
r = amdgpu_sync_resv(adev, &job->sync, resv,
AMDGPU_FENCE_OWNER_UNDEFINED,
false);
if (r) {
DRM_ERROR("sync failed (%d).\n", r);
goto error_free;
}
}
for (i = 0; i < num_loops; i++) {
uint32_t cur_size_in_bytes = min(byte_count, max_bytes);
amdgpu_emit_copy_buffer(adev, &job->ibs[0], src_offset,
dst_offset, cur_size_in_bytes);
src_offset += cur_size_in_bytes;
dst_offset += cur_size_in_bytes;
byte_count -= cur_size_in_bytes;
}
amdgpu_ring_pad_ib(ring, &job->ibs[0]);
WARN_ON(job->ibs[0].length_dw > num_dw);
if (direct_submit)
r = amdgpu_job_submit_direct(job, ring, fence);
else
r = amdgpu_job_submit(job, &adev->mman.entity,
AMDGPU_FENCE_OWNER_UNDEFINED, fence);
if (r)
goto error_free;
return r;
error_free:
amdgpu_job_free(job);
DRM_ERROR("Error scheduling IBs (%d)\n", r);
return r;
}
int amdgpu_fill_buffer(struct amdgpu_bo *bo,
uint32_t src_data,
struct dma_resv *resv,
struct dma_fence **fence)
{
struct amdgpu_device *adev = amdgpu_ttm_adev(bo->tbo.bdev);
uint32_t max_bytes = adev->mman.buffer_funcs->fill_max_bytes;
struct amdgpu_ring *ring = adev->mman.buffer_funcs_ring;
struct drm_mm_node *mm_node;
unsigned long num_pages;
unsigned int num_loops, num_dw;
struct amdgpu_job *job;
int r;
if (!adev->mman.buffer_funcs_enabled) {
DRM_ERROR("Trying to clear memory with ring turned off.\n");
return -EINVAL;
}
if (bo->tbo.mem.mem_type == TTM_PL_TT) {
r = amdgpu_ttm_alloc_gart(&bo->tbo);
if (r)
return r;
}
num_pages = bo->tbo.num_pages;
mm_node = bo->tbo.mem.mm_node;
num_loops = 0;
while (num_pages) {
uint64_t byte_count = mm_node->size << PAGE_SHIFT;
num_loops += DIV_ROUND_UP_ULL(byte_count, max_bytes);
num_pages -= mm_node->size;
++mm_node;
}
num_dw = num_loops * adev->mman.buffer_funcs->fill_num_dw;
/* for IB padding */
num_dw += 64;
r = amdgpu_job_alloc_with_ib(adev, num_dw * 4, &job);
if (r)
return r;
if (resv) {
r = amdgpu_sync_resv(adev, &job->sync, resv,
AMDGPU_FENCE_OWNER_UNDEFINED, false);
if (r) {
DRM_ERROR("sync failed (%d).\n", r);
goto error_free;
}
}
num_pages = bo->tbo.num_pages;
mm_node = bo->tbo.mem.mm_node;
while (num_pages) {
uint64_t byte_count = mm_node->size << PAGE_SHIFT;
uint64_t dst_addr;
dst_addr = amdgpu_mm_node_addr(&bo->tbo, mm_node, &bo->tbo.mem);
while (byte_count) {
uint32_t cur_size_in_bytes = min_t(uint64_t, byte_count,
max_bytes);
amdgpu_emit_fill_buffer(adev, &job->ibs[0], src_data,
dst_addr, cur_size_in_bytes);
dst_addr += cur_size_in_bytes;
byte_count -= cur_size_in_bytes;
}
num_pages -= mm_node->size;
++mm_node;
}
amdgpu_ring_pad_ib(ring, &job->ibs[0]);
WARN_ON(job->ibs[0].length_dw > num_dw);
r = amdgpu_job_submit(job, &adev->mman.entity,
AMDGPU_FENCE_OWNER_UNDEFINED, fence);
if (r)
goto error_free;
return 0;
error_free:
amdgpu_job_free(job);
return r;
}
#if defined(CONFIG_DEBUG_FS)
static int amdgpu_mm_dump_table(struct seq_file *m, void *data)
{
struct drm_info_node *node = (struct drm_info_node *)m->private;
unsigned ttm_pl = (uintptr_t)node->info_ent->data;
struct drm_device *dev = node->minor->dev;
struct amdgpu_device *adev = dev->dev_private;
struct ttm_mem_type_manager *man = &adev->mman.bdev.man[ttm_pl];
struct drm_printer p = drm_seq_file_printer(m);
man->func->debug(man, &p);
return 0;
}
static const struct drm_info_list amdgpu_ttm_debugfs_list[] = {
{"amdgpu_vram_mm", amdgpu_mm_dump_table, 0, (void *)TTM_PL_VRAM},
{"amdgpu_gtt_mm", amdgpu_mm_dump_table, 0, (void *)TTM_PL_TT},
{"amdgpu_gds_mm", amdgpu_mm_dump_table, 0, (void *)AMDGPU_PL_GDS},
{"amdgpu_gws_mm", amdgpu_mm_dump_table, 0, (void *)AMDGPU_PL_GWS},
{"amdgpu_oa_mm", amdgpu_mm_dump_table, 0, (void *)AMDGPU_PL_OA},
{"ttm_page_pool", ttm_page_alloc_debugfs, 0, NULL},
#ifdef CONFIG_SWIOTLB
{"ttm_dma_page_pool", ttm_dma_page_alloc_debugfs, 0, NULL}
#endif
};
/**
* amdgpu_ttm_vram_read - Linear read access to VRAM
*
* Accesses VRAM via MMIO for debugging purposes.
*/
static ssize_t amdgpu_ttm_vram_read(struct file *f, char __user *buf,
size_t size, loff_t *pos)
{
struct amdgpu_device *adev = file_inode(f)->i_private;
ssize_t result = 0;
int r;
if (size & 0x3 || *pos & 0x3)
return -EINVAL;
if (*pos >= adev->gmc.mc_vram_size)
return -ENXIO;
while (size) {
unsigned long flags;
uint32_t value;
if (*pos >= adev->gmc.mc_vram_size)
return result;
spin_lock_irqsave(&adev->mmio_idx_lock, flags);
WREG32_NO_KIQ(mmMM_INDEX, ((uint32_t)*pos) | 0x80000000);
WREG32_NO_KIQ(mmMM_INDEX_HI, *pos >> 31);
value = RREG32_NO_KIQ(mmMM_DATA);
spin_unlock_irqrestore(&adev->mmio_idx_lock, flags);
r = put_user(value, (uint32_t *)buf);
if (r)
return r;
result += 4;
buf += 4;
*pos += 4;
size -= 4;
}
return result;
}
/**
* amdgpu_ttm_vram_write - Linear write access to VRAM
*
* Accesses VRAM via MMIO for debugging purposes.
*/
static ssize_t amdgpu_ttm_vram_write(struct file *f, const char __user *buf,
size_t size, loff_t *pos)
{
struct amdgpu_device *adev = file_inode(f)->i_private;
ssize_t result = 0;
int r;
if (size & 0x3 || *pos & 0x3)
return -EINVAL;
if (*pos >= adev->gmc.mc_vram_size)
return -ENXIO;
while (size) {
unsigned long flags;
uint32_t value;
if (*pos >= adev->gmc.mc_vram_size)
return result;
r = get_user(value, (uint32_t *)buf);
if (r)
return r;
spin_lock_irqsave(&adev->mmio_idx_lock, flags);
WREG32_NO_KIQ(mmMM_INDEX, ((uint32_t)*pos) | 0x80000000);
WREG32_NO_KIQ(mmMM_INDEX_HI, *pos >> 31);
WREG32_NO_KIQ(mmMM_DATA, value);
spin_unlock_irqrestore(&adev->mmio_idx_lock, flags);
result += 4;
buf += 4;
*pos += 4;
size -= 4;
}
return result;
}
static const struct file_operations amdgpu_ttm_vram_fops = {
.owner = THIS_MODULE,
.read = amdgpu_ttm_vram_read,
.write = amdgpu_ttm_vram_write,
.llseek = default_llseek,
};
#ifdef CONFIG_DRM_AMDGPU_GART_DEBUGFS
/**
* amdgpu_ttm_gtt_read - Linear read access to GTT memory
*/
static ssize_t amdgpu_ttm_gtt_read(struct file *f, char __user *buf,
size_t size, loff_t *pos)
{
struct amdgpu_device *adev = file_inode(f)->i_private;
ssize_t result = 0;
int r;
while (size) {
loff_t p = *pos / PAGE_SIZE;
unsigned off = *pos & ~PAGE_MASK;
size_t cur_size = min_t(size_t, size, PAGE_SIZE - off);
struct page *page;
void *ptr;
if (p >= adev->gart.num_cpu_pages)
return result;
page = adev->gart.pages[p];
if (page) {
ptr = kmap(page);
ptr += off;
r = copy_to_user(buf, ptr, cur_size);
kunmap(adev->gart.pages[p]);
} else
r = clear_user(buf, cur_size);
if (r)
return -EFAULT;
result += cur_size;
buf += cur_size;
*pos += cur_size;
size -= cur_size;
}
return result;
}
static const struct file_operations amdgpu_ttm_gtt_fops = {
.owner = THIS_MODULE,
.read = amdgpu_ttm_gtt_read,
.llseek = default_llseek
};
#endif
/**
* amdgpu_iomem_read - Virtual read access to GPU mapped memory
*
* This function is used to read memory that has been mapped to the
* GPU and the known addresses are not physical addresses but instead
* bus addresses (e.g., what you'd put in an IB or ring buffer).
*/
static ssize_t amdgpu_iomem_read(struct file *f, char __user *buf,
size_t size, loff_t *pos)
{
struct amdgpu_device *adev = file_inode(f)->i_private;
struct iommu_domain *dom;
ssize_t result = 0;
int r;
/* retrieve the IOMMU domain if any for this device */
dom = iommu_get_domain_for_dev(adev->dev);
while (size) {
phys_addr_t addr = *pos & PAGE_MASK;
loff_t off = *pos & ~PAGE_MASK;
size_t bytes = PAGE_SIZE - off;
unsigned long pfn;
struct page *p;
void *ptr;
bytes = bytes < size ? bytes : size;
/* Translate the bus address to a physical address. If
* the domain is NULL it means there is no IOMMU active
* and the address translation is the identity
*/
addr = dom ? iommu_iova_to_phys(dom, addr) : addr;
pfn = addr >> PAGE_SHIFT;
if (!pfn_valid(pfn))
return -EPERM;
p = pfn_to_page(pfn);
if (p->mapping != adev->mman.bdev.dev_mapping)
return -EPERM;
ptr = kmap(p);
r = copy_to_user(buf, ptr + off, bytes);
kunmap(p);
if (r)
return -EFAULT;
size -= bytes;
*pos += bytes;
result += bytes;
}
return result;
}
/**
* amdgpu_iomem_write - Virtual write access to GPU mapped memory
*
* This function is used to write memory that has been mapped to the
* GPU and the known addresses are not physical addresses but instead
* bus addresses (e.g., what you'd put in an IB or ring buffer).
*/
static ssize_t amdgpu_iomem_write(struct file *f, const char __user *buf,
size_t size, loff_t *pos)
{
struct amdgpu_device *adev = file_inode(f)->i_private;
struct iommu_domain *dom;
ssize_t result = 0;
int r;
dom = iommu_get_domain_for_dev(adev->dev);
while (size) {
phys_addr_t addr = *pos & PAGE_MASK;
loff_t off = *pos & ~PAGE_MASK;
size_t bytes = PAGE_SIZE - off;
unsigned long pfn;
struct page *p;
void *ptr;
bytes = bytes < size ? bytes : size;
addr = dom ? iommu_iova_to_phys(dom, addr) : addr;
pfn = addr >> PAGE_SHIFT;
if (!pfn_valid(pfn))
return -EPERM;
p = pfn_to_page(pfn);
if (p->mapping != adev->mman.bdev.dev_mapping)
return -EPERM;
ptr = kmap(p);
r = copy_from_user(ptr + off, buf, bytes);
kunmap(p);
if (r)
return -EFAULT;
size -= bytes;
*pos += bytes;
result += bytes;
}
return result;
}
static const struct file_operations amdgpu_ttm_iomem_fops = {
.owner = THIS_MODULE,
.read = amdgpu_iomem_read,
.write = amdgpu_iomem_write,
.llseek = default_llseek
};
static const struct {
char *name;
const struct file_operations *fops;
int domain;
} ttm_debugfs_entries[] = {
{ "amdgpu_vram", &amdgpu_ttm_vram_fops, TTM_PL_VRAM },
#ifdef CONFIG_DRM_AMDGPU_GART_DEBUGFS
{ "amdgpu_gtt", &amdgpu_ttm_gtt_fops, TTM_PL_TT },
#endif
{ "amdgpu_iomem", &amdgpu_ttm_iomem_fops, TTM_PL_SYSTEM },
};
#endif
static int amdgpu_ttm_debugfs_init(struct amdgpu_device *adev)
{
#if defined(CONFIG_DEBUG_FS)
unsigned count;
struct drm_minor *minor = adev->ddev->primary;
struct dentry *ent, *root = minor->debugfs_root;
for (count = 0; count < ARRAY_SIZE(ttm_debugfs_entries); count++) {
ent = debugfs_create_file(
ttm_debugfs_entries[count].name,
S_IFREG | S_IRUGO, root,
adev,
ttm_debugfs_entries[count].fops);
if (IS_ERR(ent))
return PTR_ERR(ent);
if (ttm_debugfs_entries[count].domain == TTM_PL_VRAM)
i_size_write(ent->d_inode, adev->gmc.mc_vram_size);
else if (ttm_debugfs_entries[count].domain == TTM_PL_TT)
i_size_write(ent->d_inode, adev->gmc.gart_size);
adev->mman.debugfs_entries[count] = ent;
}
count = ARRAY_SIZE(amdgpu_ttm_debugfs_list);
#ifdef CONFIG_SWIOTLB
if (!(adev->need_swiotlb && swiotlb_nr_tbl()))
--count;
#endif
return amdgpu_debugfs_add_files(adev, amdgpu_ttm_debugfs_list, count);
#else
return 0;
#endif
}
static void amdgpu_ttm_debugfs_fini(struct amdgpu_device *adev)
{
#if defined(CONFIG_DEBUG_FS)
unsigned i;
for (i = 0; i < ARRAY_SIZE(ttm_debugfs_entries); i++)
debugfs_remove(adev->mman.debugfs_entries[i]);
#endif
}
|