summaryrefslogtreecommitdiffstats
path: root/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
blob: 07ff3b1514f129edc23875c1f42053f7ef1aaa72 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
1637
1638
1639
1640
1641
1642
1643
1644
1645
1646
1647
1648
1649
1650
1651
1652
1653
1654
1655
1656
1657
1658
1659
1660
1661
1662
1663
1664
1665
1666
1667
1668
1669
1670
1671
1672
1673
1674
1675
1676
1677
1678
1679
1680
1681
1682
1683
1684
1685
1686
1687
1688
1689
1690
1691
1692
1693
1694
1695
1696
1697
1698
1699
1700
1701
1702
1703
1704
1705
1706
1707
1708
1709
1710
1711
1712
1713
1714
1715
1716
1717
1718
1719
1720
1721
1722
1723
1724
1725
1726
1727
1728
1729
1730
1731
1732
1733
1734
1735
1736
1737
1738
1739
1740
1741
1742
1743
1744
1745
1746
1747
1748
1749
1750
1751
1752
1753
1754
1755
1756
1757
1758
1759
1760
1761
1762
1763
1764
1765
1766
1767
1768
1769
1770
1771
1772
1773
1774
1775
1776
1777
1778
1779
1780
1781
1782
1783
1784
1785
1786
1787
1788
1789
1790
1791
1792
1793
1794
1795
1796
1797
1798
1799
1800
1801
1802
1803
1804
1805
1806
1807
1808
1809
1810
1811
1812
1813
1814
1815
1816
1817
1818
1819
1820
1821
1822
1823
1824
1825
1826
1827
1828
1829
1830
1831
1832
1833
1834
1835
1836
1837
1838
1839
1840
1841
1842
1843
1844
1845
1846
1847
1848
1849
1850
1851
1852
1853
1854
1855
1856
1857
1858
1859
1860
1861
1862
1863
1864
1865
1866
1867
1868
1869
1870
1871
1872
1873
1874
1875
1876
1877
1878
1879
1880
1881
1882
1883
1884
1885
1886
1887
1888
1889
1890
1891
1892
1893
1894
1895
1896
1897
1898
1899
1900
1901
1902
1903
1904
1905
1906
1907
1908
1909
1910
1911
1912
1913
1914
1915
1916
1917
1918
1919
1920
1921
1922
1923
1924
1925
1926
1927
1928
1929
1930
1931
1932
1933
1934
1935
1936
1937
1938
1939
1940
1941
1942
1943
1944
1945
1946
1947
1948
1949
1950
1951
1952
1953
1954
1955
1956
1957
1958
1959
1960
1961
1962
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
2036
2037
2038
2039
2040
2041
2042
2043
2044
2045
2046
2047
2048
2049
2050
2051
2052
2053
2054
2055
2056
2057
2058
2059
2060
2061
2062
2063
2064
2065
2066
2067
2068
2069
2070
2071
2072
2073
2074
2075
2076
2077
2078
2079
2080
2081
2082
2083
2084
2085
2086
2087
2088
2089
2090
2091
2092
2093
2094
2095
2096
2097
2098
2099
2100
2101
2102
2103
2104
2105
2106
2107
2108
2109
2110
2111
2112
2113
2114
2115
2116
2117
2118
2119
2120
2121
2122
2123
2124
2125
2126
2127
2128
2129
2130
2131
2132
2133
2134
2135
2136
2137
2138
2139
2140
2141
2142
2143
2144
2145
2146
2147
2148
2149
2150
2151
2152
2153
2154
2155
2156
2157
2158
2159
2160
2161
2162
2163
2164
2165
2166
2167
2168
2169
2170
2171
2172
2173
2174
2175
2176
2177
2178
2179
2180
2181
2182
2183
2184
2185
2186
2187
2188
2189
2190
2191
2192
2193
2194
2195
2196
2197
2198
2199
2200
2201
2202
2203
2204
2205
2206
2207
2208
2209
2210
2211
2212
2213
2214
2215
2216
2217
2218
2219
2220
2221
2222
2223
2224
2225
2226
2227
2228
2229
2230
2231
2232
2233
2234
2235
2236
2237
2238
2239
2240
2241
2242
2243
2244
2245
2246
2247
2248
2249
2250
2251
2252
2253
2254
2255
2256
2257
2258
2259
2260
2261
2262
2263
2264
2265
2266
2267
2268
2269
2270
2271
2272
2273
2274
2275
2276
2277
2278
2279
2280
2281
2282
2283
2284
2285
2286
2287
2288
2289
2290
2291
2292
2293
2294
2295
2296
2297
2298
2299
2300
2301
2302
2303
2304
/*
 * Copyright 2008 Advanced Micro Devices, Inc.
 * Copyright 2008 Red Hat Inc.
 * Copyright 2009 Jerome Glisse.
 *
 * 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.
 *
 * Authors: Dave Airlie
 *          Alex Deucher
 *          Jerome Glisse
 */
#include <linux/dma-fence-array.h>
#include <linux/interval_tree_generic.h>
#include <drm/drmP.h>
#include <drm/amdgpu_drm.h>
#include "amdgpu.h"
#include "amdgpu_trace.h"

/*
 * GPUVM
 * GPUVM is similar to the legacy gart on older asics, however
 * rather than there being a single global gart table
 * for the entire GPU, there are multiple VM page tables active
 * at any given time.  The VM page tables can contain a mix
 * vram pages and system memory pages and system memory pages
 * can be mapped as snooped (cached system pages) or unsnooped
 * (uncached system pages).
 * Each VM has an ID associated with it and there is a page table
 * associated with each VMID.  When execting a command buffer,
 * the kernel tells the the ring what VMID to use for that command
 * buffer.  VMIDs are allocated dynamically as commands are submitted.
 * The userspace drivers maintain their own address space and the kernel
 * sets up their pages tables accordingly when they submit their
 * command buffers and a VMID is assigned.
 * Cayman/Trinity support up to 8 active VMs at any given time;
 * SI supports 16.
 */

#define START(node) ((node)->start)
#define LAST(node) ((node)->last)

INTERVAL_TREE_DEFINE(struct amdgpu_bo_va_mapping, rb, uint64_t, __subtree_last,
		     START, LAST, static, amdgpu_vm_it)

#undef START
#undef LAST

/* Local structure. Encapsulate some VM table update parameters to reduce
 * the number of function parameters
 */
struct amdgpu_pte_update_params {
	/* amdgpu device we do this update for */
	struct amdgpu_device *adev;
	/* optional amdgpu_vm we do this update for */
	struct amdgpu_vm *vm;
	/* address where to copy page table entries from */
	uint64_t src;
	/* indirect buffer to fill with commands */
	struct amdgpu_ib *ib;
	/* Function which actually does the update */
	void (*func)(struct amdgpu_pte_update_params *params, uint64_t pe,
		     uint64_t addr, unsigned count, uint32_t incr,
		     uint64_t flags);
	/* indicate update pt or its shadow */
	bool shadow;
};

/* Helper to disable partial resident texture feature from a fence callback */
struct amdgpu_prt_cb {
	struct amdgpu_device *adev;
	struct dma_fence_cb cb;
};

/**
 * amdgpu_vm_num_entries - return the number of entries in a PD/PT
 *
 * @adev: amdgpu_device pointer
 *
 * Calculate the number of entries in a page directory or page table.
 */
static unsigned amdgpu_vm_num_entries(struct amdgpu_device *adev,
				      unsigned level)
{
	if (level == 0)
		/* For the root directory */
		return adev->vm_manager.max_pfn >>
			(adev->vm_manager.block_size *
			 adev->vm_manager.num_level);
	else if (level == adev->vm_manager.num_level)
		/* For the page tables on the leaves */
		return AMDGPU_VM_PTE_COUNT(adev);
	else
		/* Everything in between */
		return 1 << adev->vm_manager.block_size;
}

/**
 * amdgpu_vm_bo_size - returns the size of the BOs in bytes
 *
 * @adev: amdgpu_device pointer
 *
 * Calculate the size of the BO for a page directory or page table in bytes.
 */
static unsigned amdgpu_vm_bo_size(struct amdgpu_device *adev, unsigned level)
{
	return AMDGPU_GPU_PAGE_ALIGN(amdgpu_vm_num_entries(adev, level) * 8);
}

/**
 * amdgpu_vm_get_pd_bo - add the VM PD to a validation list
 *
 * @vm: vm providing the BOs
 * @validated: head of validation list
 * @entry: entry to add
 *
 * Add the page directory to the list of BOs to
 * validate for command submission.
 */
void amdgpu_vm_get_pd_bo(struct amdgpu_vm *vm,
			 struct list_head *validated,
			 struct amdgpu_bo_list_entry *entry)
{
	entry->robj = vm->root.bo;
	entry->priority = 0;
	entry->tv.bo = &entry->robj->tbo;
	entry->tv.shared = true;
	entry->user_pages = NULL;
	list_add(&entry->tv.head, validated);
}

/**
 * amdgpu_vm_validate_layer - validate a single page table level
 *
 * @parent: parent page table level
 * @validate: callback to do the validation
 * @param: parameter for the validation callback
 *
 * Validate the page table BOs on command submission if neccessary.
 */
static int amdgpu_vm_validate_level(struct amdgpu_vm_pt *parent,
				    int (*validate)(void *, struct amdgpu_bo *),
				    void *param)
{
	unsigned i;
	int r;

	if (!parent->entries)
		return 0;

	for (i = 0; i <= parent->last_entry_used; ++i) {
		struct amdgpu_vm_pt *entry = &parent->entries[i];

		if (!entry->bo)
			continue;

		r = validate(param, entry->bo);
		if (r)
			return r;

		/*
		 * Recurse into the sub directory. This is harmless because we
		 * have only a maximum of 5 layers.
		 */
		r = amdgpu_vm_validate_level(entry, validate, param);
		if (r)
			return r;
	}

	return r;
}

/**
 * amdgpu_vm_validate_pt_bos - validate the page table BOs
 *
 * @adev: amdgpu device pointer
 * @vm: vm providing the BOs
 * @validate: callback to do the validation
 * @param: parameter for the validation callback
 *
 * Validate the page table BOs on command submission if neccessary.
 */
int amdgpu_vm_validate_pt_bos(struct amdgpu_device *adev, struct amdgpu_vm *vm,
			      int (*validate)(void *p, struct amdgpu_bo *bo),
			      void *param)
{
	uint64_t num_evictions;

	/* We only need to validate the page tables
	 * if they aren't already valid.
	 */
	num_evictions = atomic64_read(&adev->num_evictions);
	if (num_evictions == vm->last_eviction_counter)
		return 0;

	return amdgpu_vm_validate_level(&vm->root, validate, param);
}

/**
 * amdgpu_vm_move_level_in_lru - move one level of PT BOs to the LRU tail
 *
 * @adev: amdgpu device instance
 * @vm: vm providing the BOs
 *
 * Move the PT BOs to the tail of the LRU.
 */
static void amdgpu_vm_move_level_in_lru(struct amdgpu_vm_pt *parent)
{
	unsigned i;

	if (!parent->entries)
		return;

	for (i = 0; i <= parent->last_entry_used; ++i) {
		struct amdgpu_vm_pt *entry = &parent->entries[i];

		if (!entry->bo)
			continue;

		ttm_bo_move_to_lru_tail(&entry->bo->tbo);
		amdgpu_vm_move_level_in_lru(entry);
	}
}

/**
 * amdgpu_vm_move_pt_bos_in_lru - move the PT BOs to the LRU tail
 *
 * @adev: amdgpu device instance
 * @vm: vm providing the BOs
 *
 * Move the PT BOs to the tail of the LRU.
 */
void amdgpu_vm_move_pt_bos_in_lru(struct amdgpu_device *adev,
				  struct amdgpu_vm *vm)
{
	struct ttm_bo_global *glob = adev->mman.bdev.glob;

	spin_lock(&glob->lru_lock);
	amdgpu_vm_move_level_in_lru(&vm->root);
	spin_unlock(&glob->lru_lock);
}

 /**
 * amdgpu_vm_alloc_levels - allocate the PD/PT levels
 *
 * @adev: amdgpu_device pointer
 * @vm: requested vm
 * @saddr: start of the address range
 * @eaddr: end of the address range
 *
 * Make sure the page directories and page tables are allocated
 */
static int amdgpu_vm_alloc_levels(struct amdgpu_device *adev,
				  struct amdgpu_vm *vm,
				  struct amdgpu_vm_pt *parent,
				  uint64_t saddr, uint64_t eaddr,
				  unsigned level)
{
	unsigned shift = (adev->vm_manager.num_level - level) *
		adev->vm_manager.block_size;
	unsigned pt_idx, from, to;
	int r;

	if (!parent->entries) {
		unsigned num_entries = amdgpu_vm_num_entries(adev, level);

		parent->entries = drm_calloc_large(num_entries,
						   sizeof(struct amdgpu_vm_pt));
		if (!parent->entries)
			return -ENOMEM;
		memset(parent->entries, 0 , sizeof(struct amdgpu_vm_pt));
	}

	from = saddr >> shift;
	to = eaddr >> shift;
	if (from >= amdgpu_vm_num_entries(adev, level) ||
	    to >= amdgpu_vm_num_entries(adev, level))
		return -EINVAL;

	if (to > parent->last_entry_used)
		parent->last_entry_used = to;

	++level;
	saddr = saddr & ((1 << shift) - 1);
	eaddr = eaddr & ((1 << shift) - 1);

	/* walk over the address space and allocate the page tables */
	for (pt_idx = from; pt_idx <= to; ++pt_idx) {
		struct reservation_object *resv = vm->root.bo->tbo.resv;
		struct amdgpu_vm_pt *entry = &parent->entries[pt_idx];
		struct amdgpu_bo *pt;

		if (!entry->bo) {
			r = amdgpu_bo_create(adev,
					     amdgpu_vm_bo_size(adev, level),
					     AMDGPU_GPU_PAGE_SIZE, true,
					     AMDGPU_GEM_DOMAIN_VRAM,
					     AMDGPU_GEM_CREATE_NO_CPU_ACCESS |
					     AMDGPU_GEM_CREATE_SHADOW |
					     AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS |
					     AMDGPU_GEM_CREATE_VRAM_CLEARED,
					     NULL, resv, &pt);
			if (r)
				return r;

			/* Keep a reference to the root directory to avoid
			* freeing them up in the wrong order.
			*/
			pt->parent = amdgpu_bo_ref(vm->root.bo);

			entry->bo = pt;
			entry->addr = 0;
		}

		if (level < adev->vm_manager.num_level) {
			uint64_t sub_saddr = (pt_idx == from) ? saddr : 0;
			uint64_t sub_eaddr = (pt_idx == to) ? eaddr :
				((1 << shift) - 1);
			r = amdgpu_vm_alloc_levels(adev, vm, entry, sub_saddr,
						   sub_eaddr, level);
			if (r)
				return r;
		}
	}

	return 0;
}

/**
 * amdgpu_vm_alloc_pts - Allocate page tables.
 *
 * @adev: amdgpu_device pointer
 * @vm: VM to allocate page tables for
 * @saddr: Start address which needs to be allocated
 * @size: Size from start address we need.
 *
 * Make sure the page tables are allocated.
 */
int amdgpu_vm_alloc_pts(struct amdgpu_device *adev,
			struct amdgpu_vm *vm,
			uint64_t saddr, uint64_t size)
{
	uint64_t last_pfn;
	uint64_t eaddr;

	/* validate the parameters */
	if (saddr & AMDGPU_GPU_PAGE_MASK || size & AMDGPU_GPU_PAGE_MASK)
		return -EINVAL;

	eaddr = saddr + size - 1;
	last_pfn = eaddr / AMDGPU_GPU_PAGE_SIZE;
	if (last_pfn >= adev->vm_manager.max_pfn) {
		dev_err(adev->dev, "va above limit (0x%08llX >= 0x%08llX)\n",
			last_pfn, adev->vm_manager.max_pfn);
		return -EINVAL;
	}

	saddr /= AMDGPU_GPU_PAGE_SIZE;
	eaddr /= AMDGPU_GPU_PAGE_SIZE;

	return amdgpu_vm_alloc_levels(adev, vm, &vm->root, saddr, eaddr, 0);
}

/**
 * amdgpu_vm_had_gpu_reset - check if reset occured since last use
 *
 * @adev: amdgpu_device pointer
 * @id: VMID structure
 *
 * Check if GPU reset occured since last use of the VMID.
 */
static bool amdgpu_vm_had_gpu_reset(struct amdgpu_device *adev,
				    struct amdgpu_vm_id *id)
{
	return id->current_gpu_reset_count !=
		atomic_read(&adev->gpu_reset_counter);
}

/**
 * amdgpu_vm_grab_id - allocate the next free VMID
 *
 * @vm: vm to allocate id for
 * @ring: ring we want to submit job to
 * @sync: sync object where we add dependencies
 * @fence: fence protecting ID from reuse
 *
 * Allocate an id for the vm, adding fences to the sync obj as necessary.
 */
int amdgpu_vm_grab_id(struct amdgpu_vm *vm, struct amdgpu_ring *ring,
		      struct amdgpu_sync *sync, struct dma_fence *fence,
		      struct amdgpu_job *job)
{
	struct amdgpu_device *adev = ring->adev;
	unsigned vmhub = ring->funcs->vmhub;
	struct amdgpu_vm_id_manager *id_mgr = &adev->vm_manager.id_mgr[vmhub];
	uint64_t fence_context = adev->fence_context + ring->idx;
	struct dma_fence *updates = sync->last_vm_update;
	struct amdgpu_vm_id *id, *idle;
	struct dma_fence **fences;
	unsigned i;
	int r = 0;

	fences = kmalloc_array(sizeof(void *), id_mgr->num_ids, GFP_KERNEL);
	if (!fences)
		return -ENOMEM;

	mutex_lock(&id_mgr->lock);

	/* Check if we have an idle VMID */
	i = 0;
	list_for_each_entry(idle, &id_mgr->ids_lru, list) {
		fences[i] = amdgpu_sync_peek_fence(&idle->active, ring);
		if (!fences[i])
			break;
		++i;
	}

	/* If we can't find a idle VMID to use, wait till one becomes available */
	if (&idle->list == &id_mgr->ids_lru) {
		u64 fence_context = adev->vm_manager.fence_context + ring->idx;
		unsigned seqno = ++adev->vm_manager.seqno[ring->idx];
		struct dma_fence_array *array;
		unsigned j;

		for (j = 0; j < i; ++j)
			dma_fence_get(fences[j]);

		array = dma_fence_array_create(i, fences, fence_context,
					   seqno, true);
		if (!array) {
			for (j = 0; j < i; ++j)
				dma_fence_put(fences[j]);
			kfree(fences);
			r = -ENOMEM;
			goto error;
		}


		r = amdgpu_sync_fence(ring->adev, sync, &array->base);
		dma_fence_put(&array->base);
		if (r)
			goto error;

		mutex_unlock(&id_mgr->lock);
		return 0;

	}
	kfree(fences);

	job->vm_needs_flush = false;
	/* Check if we can use a VMID already assigned to this VM */
	list_for_each_entry_reverse(id, &id_mgr->ids_lru, list) {
		struct dma_fence *flushed;
		bool needs_flush = false;

		/* Check all the prerequisites to using this VMID */
		if (amdgpu_vm_had_gpu_reset(adev, id))
			continue;

		if (atomic64_read(&id->owner) != vm->client_id)
			continue;

		if (job->vm_pd_addr != id->pd_gpu_addr)
			continue;

		if (!id->last_flush ||
		    (id->last_flush->context != fence_context &&
		     !dma_fence_is_signaled(id->last_flush)))
			needs_flush = true;

		flushed  = id->flushed_updates;
		if (updates && (!flushed || dma_fence_is_later(updates, flushed)))
			needs_flush = true;

		/* Concurrent flushes are only possible starting with Vega10 */
		if (adev->asic_type < CHIP_VEGA10 && needs_flush)
			continue;

		/* Good we can use this VMID. Remember this submission as
		 * user of the VMID.
		 */
		r = amdgpu_sync_fence(ring->adev, &id->active, fence);
		if (r)
			goto error;

		if (updates && (!flushed || dma_fence_is_later(updates, flushed))) {
			dma_fence_put(id->flushed_updates);
			id->flushed_updates = dma_fence_get(updates);
		}

		if (needs_flush)
			goto needs_flush;
		else
			goto no_flush_needed;

	};

	/* Still no ID to use? Then use the idle one found earlier */
	id = idle;

	/* Remember this submission as user of the VMID */
	r = amdgpu_sync_fence(ring->adev, &id->active, fence);
	if (r)
		goto error;

	id->pd_gpu_addr = job->vm_pd_addr;
	dma_fence_put(id->flushed_updates);
	id->flushed_updates = dma_fence_get(updates);
	id->current_gpu_reset_count = atomic_read(&adev->gpu_reset_counter);
	atomic64_set(&id->owner, vm->client_id);

needs_flush:
	job->vm_needs_flush = true;
	dma_fence_put(id->last_flush);
	id->last_flush = NULL;

no_flush_needed:
	list_move_tail(&id->list, &id_mgr->ids_lru);

	job->vm_id = id - id_mgr->ids;
	trace_amdgpu_vm_grab_id(vm, ring, job);

error:
	mutex_unlock(&id_mgr->lock);
	return r;
}

static bool amdgpu_vm_ring_has_compute_vm_bug(struct amdgpu_ring *ring)
{
	struct amdgpu_device *adev = ring->adev;
	const struct amdgpu_ip_block *ip_block;

	if (ring->funcs->type != AMDGPU_RING_TYPE_COMPUTE)
		/* only compute rings */
		return false;

	ip_block = amdgpu_get_ip_block(adev, AMD_IP_BLOCK_TYPE_GFX);
	if (!ip_block)
		return false;

	if (ip_block->version->major <= 7) {
		/* gfx7 has no workaround */
		return true;
	} else if (ip_block->version->major == 8) {
		if (adev->gfx.mec_fw_version >= 673)
			/* gfx8 is fixed in MEC firmware 673 */
			return false;
		else
			return true;
	}
	return false;
}

static u64 amdgpu_vm_adjust_mc_addr(struct amdgpu_device *adev, u64 mc_addr)
{
	u64 addr = mc_addr;

	if (adev->gart.gart_funcs->adjust_mc_addr)
		addr = adev->gart.gart_funcs->adjust_mc_addr(adev, addr);

	return addr;
}

/**
 * amdgpu_vm_flush - hardware flush the vm
 *
 * @ring: ring to use for flush
 * @vm_id: vmid number to use
 * @pd_addr: address of the page directory
 *
 * Emit a VM flush when it is necessary.
 */
int amdgpu_vm_flush(struct amdgpu_ring *ring, struct amdgpu_job *job)
{
	struct amdgpu_device *adev = ring->adev;
	unsigned vmhub = ring->funcs->vmhub;
	struct amdgpu_vm_id_manager *id_mgr = &adev->vm_manager.id_mgr[vmhub];
	struct amdgpu_vm_id *id = &id_mgr->ids[job->vm_id];
	bool gds_switch_needed = ring->funcs->emit_gds_switch && (
		id->gds_base != job->gds_base ||
		id->gds_size != job->gds_size ||
		id->gws_base != job->gws_base ||
		id->gws_size != job->gws_size ||
		id->oa_base != job->oa_base ||
		id->oa_size != job->oa_size);
	bool vm_flush_needed = job->vm_needs_flush ||
		amdgpu_vm_ring_has_compute_vm_bug(ring);
	unsigned patch_offset = 0;
	int r;

	if (amdgpu_vm_had_gpu_reset(adev, id)) {
		gds_switch_needed = true;
		vm_flush_needed = true;
	}

	if (!vm_flush_needed && !gds_switch_needed)
		return 0;

	if (ring->funcs->init_cond_exec)
		patch_offset = amdgpu_ring_init_cond_exec(ring);

	if (ring->funcs->emit_pipeline_sync && !job->need_pipeline_sync)
		amdgpu_ring_emit_pipeline_sync(ring);

	if (ring->funcs->emit_vm_flush && vm_flush_needed) {
		u64 pd_addr = amdgpu_vm_adjust_mc_addr(adev, job->vm_pd_addr);
		struct dma_fence *fence;

		trace_amdgpu_vm_flush(ring, job->vm_id, pd_addr);
		amdgpu_ring_emit_vm_flush(ring, job->vm_id, pd_addr);

		r = amdgpu_fence_emit(ring, &fence);
		if (r)
			return r;

		mutex_lock(&id_mgr->lock);
		dma_fence_put(id->last_flush);
		id->last_flush = fence;
		mutex_unlock(&id_mgr->lock);
	}

	if (gds_switch_needed) {
		id->gds_base = job->gds_base;
		id->gds_size = job->gds_size;
		id->gws_base = job->gws_base;
		id->gws_size = job->gws_size;
		id->oa_base = job->oa_base;
		id->oa_size = job->oa_size;
		amdgpu_ring_emit_gds_switch(ring, job->vm_id, job->gds_base,
					    job->gds_size, job->gws_base,
					    job->gws_size, job->oa_base,
					    job->oa_size);
	}

	if (ring->funcs->patch_cond_exec)
		amdgpu_ring_patch_cond_exec(ring, patch_offset);

	/* the double SWITCH_BUFFER here *cannot* be skipped by COND_EXEC */
	if (ring->funcs->emit_switch_buffer) {
		amdgpu_ring_emit_switch_buffer(ring);
		amdgpu_ring_emit_switch_buffer(ring);
	}
	return 0;
}

/**
 * amdgpu_vm_reset_id - reset VMID to zero
 *
 * @adev: amdgpu device structure
 * @vm_id: vmid number to use
 *
 * Reset saved GDW, GWS and OA to force switch on next flush.
 */
void amdgpu_vm_reset_id(struct amdgpu_device *adev, unsigned vmhub,
			unsigned vmid)
{
	struct amdgpu_vm_id_manager *id_mgr = &adev->vm_manager.id_mgr[vmhub];
	struct amdgpu_vm_id *id = &id_mgr->ids[vmid];

	id->gds_base = 0;
	id->gds_size = 0;
	id->gws_base = 0;
	id->gws_size = 0;
	id->oa_base = 0;
	id->oa_size = 0;
}

/**
 * amdgpu_vm_bo_find - find the bo_va for a specific vm & bo
 *
 * @vm: requested vm
 * @bo: requested buffer object
 *
 * Find @bo inside the requested vm.
 * Search inside the @bos vm list for the requested vm
 * Returns the found bo_va or NULL if none is found
 *
 * Object has to be reserved!
 */
struct amdgpu_bo_va *amdgpu_vm_bo_find(struct amdgpu_vm *vm,
				       struct amdgpu_bo *bo)
{
	struct amdgpu_bo_va *bo_va;

	list_for_each_entry(bo_va, &bo->va, bo_list) {
		if (bo_va->vm == vm) {
			return bo_va;
		}
	}
	return NULL;
}

/**
 * amdgpu_vm_do_set_ptes - helper to call the right asic function
 *
 * @params: see amdgpu_pte_update_params definition
 * @pe: addr of the page entry
 * @addr: dst addr to write into pe
 * @count: number of page entries to update
 * @incr: increase next addr by incr bytes
 * @flags: hw access flags
 *
 * Traces the parameters and calls the right asic functions
 * to setup the page table using the DMA.
 */
static void amdgpu_vm_do_set_ptes(struct amdgpu_pte_update_params *params,
				  uint64_t pe, uint64_t addr,
				  unsigned count, uint32_t incr,
				  uint64_t flags)
{
	trace_amdgpu_vm_set_ptes(pe, addr, count, incr, flags);

	if (count < 3) {
		amdgpu_vm_write_pte(params->adev, params->ib, pe,
				    addr | flags, count, incr);

	} else {
		amdgpu_vm_set_pte_pde(params->adev, params->ib, pe, addr,
				      count, incr, flags);
	}
}

/**
 * amdgpu_vm_do_copy_ptes - copy the PTEs from the GART
 *
 * @params: see amdgpu_pte_update_params definition
 * @pe: addr of the page entry
 * @addr: dst addr to write into pe
 * @count: number of page entries to update
 * @incr: increase next addr by incr bytes
 * @flags: hw access flags
 *
 * Traces the parameters and calls the DMA function to copy the PTEs.
 */
static void amdgpu_vm_do_copy_ptes(struct amdgpu_pte_update_params *params,
				   uint64_t pe, uint64_t addr,
				   unsigned count, uint32_t incr,
				   uint64_t flags)
{
	uint64_t src = (params->src + (addr >> 12) * 8);


	trace_amdgpu_vm_copy_ptes(pe, src, count);

	amdgpu_vm_copy_pte(params->adev, params->ib, pe, src, count);
}

/**
 * amdgpu_vm_map_gart - Resolve gart mapping of addr
 *
 * @pages_addr: optional DMA address to use for lookup
 * @addr: the unmapped addr
 *
 * Look up the physical address of the page that the pte resolves
 * to and return the pointer for the page table entry.
 */
static uint64_t amdgpu_vm_map_gart(const dma_addr_t *pages_addr, uint64_t addr)
{
	uint64_t result;

	/* page table offset */
	result = pages_addr[addr >> PAGE_SHIFT];

	/* in case cpu page size != gpu page size*/
	result |= addr & (~PAGE_MASK);

	result &= 0xFFFFFFFFFFFFF000ULL;

	return result;
}

/*
 * amdgpu_vm_update_level - update a single level in the hierarchy
 *
 * @adev: amdgpu_device pointer
 * @vm: requested vm
 * @parent: parent directory
 *
 * Makes sure all entries in @parent are up to date.
 * Returns 0 for success, error for failure.
 */
static int amdgpu_vm_update_level(struct amdgpu_device *adev,
				  struct amdgpu_vm *vm,
				  struct amdgpu_vm_pt *parent,
				  unsigned level)
{
	struct amdgpu_bo *shadow;
	struct amdgpu_ring *ring;
	uint64_t pd_addr, shadow_addr;
	uint32_t incr = amdgpu_vm_bo_size(adev, level + 1);
	uint64_t last_pde = ~0, last_pt = ~0, last_shadow = ~0;
	unsigned count = 0, pt_idx, ndw;
	struct amdgpu_job *job;
	struct amdgpu_pte_update_params params;
	struct dma_fence *fence = NULL;

	int r;

	if (!parent->entries)
		return 0;
	ring = container_of(vm->entity.sched, struct amdgpu_ring, sched);

	/* padding, etc. */
	ndw = 64;

	/* assume the worst case */
	ndw += parent->last_entry_used * 6;

	pd_addr = amdgpu_bo_gpu_offset(parent->bo);

	shadow = parent->bo->shadow;
	if (shadow) {
		r = amdgpu_ttm_bind(&shadow->tbo, &shadow->tbo.mem);
		if (r)
			return r;
		shadow_addr = amdgpu_bo_gpu_offset(shadow);
		ndw *= 2;
	} else {
		shadow_addr = 0;
	}

	r = amdgpu_job_alloc_with_ib(adev, ndw * 4, &job);
	if (r)
		return r;

	memset(&params, 0, sizeof(params));
	params.adev = adev;
	params.ib = &job->ibs[0];

	/* walk over the address space and update the directory */
	for (pt_idx = 0; pt_idx <= parent->last_entry_used; ++pt_idx) {
		struct amdgpu_bo *bo = parent->entries[pt_idx].bo;
		uint64_t pde, pt;

		if (bo == NULL)
			continue;

		if (bo->shadow) {
			struct amdgpu_bo *pt_shadow = bo->shadow;

			r = amdgpu_ttm_bind(&pt_shadow->tbo,
					    &pt_shadow->tbo.mem);
			if (r)
				return r;
		}

		pt = amdgpu_bo_gpu_offset(bo);
		if (parent->entries[pt_idx].addr == pt)
			continue;

		parent->entries[pt_idx].addr = pt;

		pde = pd_addr + pt_idx * 8;
		if (((last_pde + 8 * count) != pde) ||
		    ((last_pt + incr * count) != pt) ||
		    (count == AMDGPU_VM_MAX_UPDATE_SIZE)) {

			if (count) {
				uint64_t pt_addr =
					amdgpu_vm_adjust_mc_addr(adev, last_pt);

				if (shadow)
					amdgpu_vm_do_set_ptes(&params,
							      last_shadow,
							      pt_addr, count,
							      incr,
							      AMDGPU_PTE_VALID);

				amdgpu_vm_do_set_ptes(&params, last_pde,
						      pt_addr, count, incr,
						      AMDGPU_PTE_VALID);
			}

			count = 1;
			last_pde = pde;
			last_shadow = shadow_addr + pt_idx * 8;
			last_pt = pt;
		} else {
			++count;
		}
	}

	if (count) {
		uint64_t pt_addr = amdgpu_vm_adjust_mc_addr(adev, last_pt);

		if (vm->root.bo->shadow)
			amdgpu_vm_do_set_ptes(&params, last_shadow, pt_addr,
					      count, incr, AMDGPU_PTE_VALID);

		amdgpu_vm_do_set_ptes(&params, last_pde, pt_addr,
				      count, incr, AMDGPU_PTE_VALID);
	}

	if (params.ib->length_dw == 0) {
		amdgpu_job_free(job);
	} else {
		amdgpu_ring_pad_ib(ring, params.ib);
		amdgpu_sync_resv(adev, &job->sync, parent->bo->tbo.resv,
				 AMDGPU_FENCE_OWNER_VM);
		if (shadow)
			amdgpu_sync_resv(adev, &job->sync, shadow->tbo.resv,
					 AMDGPU_FENCE_OWNER_VM);

		WARN_ON(params.ib->length_dw > ndw);
		r = amdgpu_job_submit(job, ring, &vm->entity,
				AMDGPU_FENCE_OWNER_VM, &fence);
		if (r)
			goto error_free;

		amdgpu_bo_fence(parent->bo, fence, true);
		dma_fence_put(vm->last_dir_update);
		vm->last_dir_update = dma_fence_get(fence);
		dma_fence_put(fence);
	}
	/*
	 * Recurse into the subdirectories. This recursion is harmless because
	 * we only have a maximum of 5 layers.
	 */
	for (pt_idx = 0; pt_idx <= parent->last_entry_used; ++pt_idx) {
		struct amdgpu_vm_pt *entry = &parent->entries[pt_idx];

		if (!entry->bo)
			continue;

		r = amdgpu_vm_update_level(adev, vm, entry, level + 1);
		if (r)
			return r;
	}

	return 0;

error_free:
	amdgpu_job_free(job);
	return r;
}

/*
 * amdgpu_vm_update_directories - make sure that all directories are valid
 *
 * @adev: amdgpu_device pointer
 * @vm: requested vm
 *
 * Makes sure all directories are up to date.
 * Returns 0 for success, error for failure.
 */
int amdgpu_vm_update_directories(struct amdgpu_device *adev,
				 struct amdgpu_vm *vm)
{
	return amdgpu_vm_update_level(adev, vm, &vm->root, 0);
}

/**
 * amdgpu_vm_find_pt - find the page table for an address
 *
 * @p: see amdgpu_pte_update_params definition
 * @addr: virtual address in question
 *
 * Find the page table BO for a virtual address, return NULL when none found.
 */
static struct amdgpu_bo *amdgpu_vm_get_pt(struct amdgpu_pte_update_params *p,
					  uint64_t addr)
{
	struct amdgpu_vm_pt *entry = &p->vm->root;
	unsigned idx, level = p->adev->vm_manager.num_level;

	while (entry->entries) {
		idx = addr >> (p->adev->vm_manager.block_size * level--);
		idx %= amdgpu_bo_size(entry->bo) / 8;
		entry = &entry->entries[idx];
	}

	if (level)
		return NULL;

	return entry->bo;
}

/**
 * amdgpu_vm_update_ptes - make sure that page tables are valid
 *
 * @params: see amdgpu_pte_update_params definition
 * @vm: requested vm
 * @start: start of GPU address range
 * @end: end of GPU address range
 * @dst: destination address to map to, the next dst inside the function
 * @flags: mapping flags
 *
 * Update the page tables in the range @start - @end.
 */
static void amdgpu_vm_update_ptes(struct amdgpu_pte_update_params *params,
				  uint64_t start, uint64_t end,
				  uint64_t dst, uint64_t flags)
{
	struct amdgpu_device *adev = params->adev;
	const uint64_t mask = AMDGPU_VM_PTE_COUNT(adev) - 1;

	uint64_t cur_pe_start, cur_nptes, cur_dst;
	uint64_t addr; /* next GPU address to be updated */
	struct amdgpu_bo *pt;
	unsigned nptes; /* next number of ptes to be updated */
	uint64_t next_pe_start;

	/* initialize the variables */
	addr = start;
	pt = amdgpu_vm_get_pt(params, addr);
	if (!pt) {
		pr_err("PT not found, aborting update_ptes\n");
		return;
	}

	if (params->shadow) {
		if (!pt->shadow)
			return;
		pt = pt->shadow;
	}
	if ((addr & ~mask) == (end & ~mask))
		nptes = end - addr;
	else
		nptes = AMDGPU_VM_PTE_COUNT(adev) - (addr & mask);

	cur_pe_start = amdgpu_bo_gpu_offset(pt);
	cur_pe_start += (addr & mask) * 8;
	cur_nptes = nptes;
	cur_dst = dst;

	/* for next ptb*/
	addr += nptes;
	dst += nptes * AMDGPU_GPU_PAGE_SIZE;

	/* walk over the address space and update the page tables */
	while (addr < end) {
		pt = amdgpu_vm_get_pt(params, addr);
		if (!pt) {
			pr_err("PT not found, aborting update_ptes\n");
			return;
		}

		if (params->shadow) {
			if (!pt->shadow)
				return;
			pt = pt->shadow;
		}

		if ((addr & ~mask) == (end & ~mask))
			nptes = end - addr;
		else
			nptes = AMDGPU_VM_PTE_COUNT(adev) - (addr & mask);

		next_pe_start = amdgpu_bo_gpu_offset(pt);
		next_pe_start += (addr & mask) * 8;

		if ((cur_pe_start + 8 * cur_nptes) == next_pe_start &&
		    ((cur_nptes + nptes) <= AMDGPU_VM_MAX_UPDATE_SIZE)) {
			/* The next ptb is consecutive to current ptb.
			 * Don't call the update function now.
			 * Will update two ptbs together in future.
			*/
			cur_nptes += nptes;
		} else {
			params->func(params, cur_pe_start, cur_dst, cur_nptes,
				     AMDGPU_GPU_PAGE_SIZE, flags);

			cur_pe_start = next_pe_start;
			cur_nptes = nptes;
			cur_dst = dst;
		}

		/* for next ptb*/
		addr += nptes;
		dst += nptes * AMDGPU_GPU_PAGE_SIZE;
	}

	params->func(params, cur_pe_start, cur_dst, cur_nptes,
		     AMDGPU_GPU_PAGE_SIZE, flags);
}

/*
 * amdgpu_vm_frag_ptes - add fragment information to PTEs
 *
 * @params: see amdgpu_pte_update_params definition
 * @vm: requested vm
 * @start: first PTE to handle
 * @end: last PTE to handle
 * @dst: addr those PTEs should point to
 * @flags: hw mapping flags
 */
static void amdgpu_vm_frag_ptes(struct amdgpu_pte_update_params	*params,
				uint64_t start, uint64_t end,
				uint64_t dst, uint64_t flags)
{
	/**
	 * The MC L1 TLB supports variable sized pages, based on a fragment
	 * field in the PTE. When this field is set to a non-zero value, page
	 * granularity is increased from 4KB to (1 << (12 + frag)). The PTE
	 * flags are considered valid for all PTEs within the fragment range
	 * and corresponding mappings are assumed to be physically contiguous.
	 *
	 * The L1 TLB can store a single PTE for the whole fragment,
	 * significantly increasing the space available for translation
	 * caching. This leads to large improvements in throughput when the
	 * TLB is under pressure.
	 *
	 * The L2 TLB distributes small and large fragments into two
	 * asymmetric partitions. The large fragment cache is significantly
	 * larger. Thus, we try to use large fragments wherever possible.
	 * Userspace can support this by aligning virtual base address and
	 * allocation size to the fragment size.
	 */

	/* SI and newer are optimized for 64KB */
	uint64_t frag_flags = AMDGPU_PTE_FRAG(AMDGPU_LOG2_PAGES_PER_FRAG);
	uint64_t frag_align = 1 << AMDGPU_LOG2_PAGES_PER_FRAG;

	uint64_t frag_start = ALIGN(start, frag_align);
	uint64_t frag_end = end & ~(frag_align - 1);

	/* system pages are non continuously */
	if (params->src || !(flags & AMDGPU_PTE_VALID) ||
	    (frag_start >= frag_end)) {

		amdgpu_vm_update_ptes(params, start, end, dst, flags);
		return;
	}

	/* handle the 4K area at the beginning */
	if (start != frag_start) {
		amdgpu_vm_update_ptes(params, start, frag_start,
				      dst, flags);
		dst += (frag_start - start) * AMDGPU_GPU_PAGE_SIZE;
	}

	/* handle the area in the middle */
	amdgpu_vm_update_ptes(params, frag_start, frag_end, dst,
			      flags | frag_flags);

	/* handle the 4K area at the end */
	if (frag_end != end) {
		dst += (frag_end - frag_start) * AMDGPU_GPU_PAGE_SIZE;
		amdgpu_vm_update_ptes(params, frag_end, end, dst, flags);
	}
}

/**
 * amdgpu_vm_bo_update_mapping - update a mapping in the vm page table
 *
 * @adev: amdgpu_device pointer
 * @exclusive: fence we need to sync to
 * @src: address where to copy page table entries from
 * @pages_addr: DMA addresses to use for mapping
 * @vm: requested vm
 * @start: start of mapped range
 * @last: last mapped entry
 * @flags: flags for the entries
 * @addr: addr to set the area to
 * @fence: optional resulting fence
 *
 * Fill in the page table entries between @start and @last.
 * Returns 0 for success, -EINVAL for failure.
 */
static int amdgpu_vm_bo_update_mapping(struct amdgpu_device *adev,
				       struct dma_fence *exclusive,
				       uint64_t src,
				       dma_addr_t *pages_addr,
				       struct amdgpu_vm *vm,
				       uint64_t start, uint64_t last,
				       uint64_t flags, uint64_t addr,
				       struct dma_fence **fence)
{
	struct amdgpu_ring *ring;
	void *owner = AMDGPU_FENCE_OWNER_VM;
	unsigned nptes, ncmds, ndw;
	struct amdgpu_job *job;
	struct amdgpu_pte_update_params params;
	struct dma_fence *f = NULL;
	int r;

	memset(&params, 0, sizeof(params));
	params.adev = adev;
	params.vm = vm;
	params.src = src;

	ring = container_of(vm->entity.sched, struct amdgpu_ring, sched);

	/* sync to everything on unmapping */
	if (!(flags & AMDGPU_PTE_VALID))
		owner = AMDGPU_FENCE_OWNER_UNDEFINED;

	nptes = last - start + 1;

	/*
	 * reserve space for one command every (1 << BLOCK_SIZE)
	 *  entries or 2k dwords (whatever is smaller)
	 */
	ncmds = (nptes >> min(adev->vm_manager.block_size, 11u)) + 1;

	/* padding, etc. */
	ndw = 64;

	if (src) {
		/* only copy commands needed */
		ndw += ncmds * 7;

		params.func = amdgpu_vm_do_copy_ptes;

	} else if (pages_addr) {
		/* copy commands needed */
		ndw += ncmds * 7;

		/* and also PTEs */
		ndw += nptes * 2;

		params.func = amdgpu_vm_do_copy_ptes;

	} else {
		/* set page commands needed */
		ndw += ncmds * 10;

		/* two extra commands for begin/end of fragment */
		ndw += 2 * 10;

		params.func = amdgpu_vm_do_set_ptes;
	}

	r = amdgpu_job_alloc_with_ib(adev, ndw * 4, &job);
	if (r)
		return r;

	params.ib = &job->ibs[0];

	if (!src && pages_addr) {
		uint64_t *pte;
		unsigned i;

		/* Put the PTEs at the end of the IB. */
		i = ndw - nptes * 2;
		pte= (uint64_t *)&(job->ibs->ptr[i]);
		params.src = job->ibs->gpu_addr + i * 4;

		for (i = 0; i < nptes; ++i) {
			pte[i] = amdgpu_vm_map_gart(pages_addr, addr + i *
						    AMDGPU_GPU_PAGE_SIZE);
			pte[i] |= flags;
		}
		addr = 0;
	}

	r = amdgpu_sync_fence(adev, &job->sync, exclusive);
	if (r)
		goto error_free;

	r = amdgpu_sync_resv(adev, &job->sync, vm->root.bo->tbo.resv,
			     owner);
	if (r)
		goto error_free;

	r = reservation_object_reserve_shared(vm->root.bo->tbo.resv);
	if (r)
		goto error_free;

	params.shadow = true;
	amdgpu_vm_frag_ptes(&params, start, last + 1, addr, flags);
	params.shadow = false;
	amdgpu_vm_frag_ptes(&params, start, last + 1, addr, flags);

	amdgpu_ring_pad_ib(ring, params.ib);
	WARN_ON(params.ib->length_dw > ndw);
	r = amdgpu_job_submit(job, ring, &vm->entity,
			      AMDGPU_FENCE_OWNER_VM, &f);
	if (r)
		goto error_free;

	amdgpu_bo_fence(vm->root.bo, f, true);
	dma_fence_put(*fence);
	*fence = f;
	return 0;

error_free:
	amdgpu_job_free(job);
	return r;
}

/**
 * amdgpu_vm_bo_split_mapping - split a mapping into smaller chunks
 *
 * @adev: amdgpu_device pointer
 * @exclusive: fence we need to sync to
 * @gtt_flags: flags as they are used for GTT
 * @pages_addr: DMA addresses to use for mapping
 * @vm: requested vm
 * @mapping: mapped range and flags to use for the update
 * @flags: HW flags for the mapping
 * @nodes: array of drm_mm_nodes with the MC addresses
 * @fence: optional resulting fence
 *
 * Split the mapping into smaller chunks so that each update fits
 * into a SDMA IB.
 * Returns 0 for success, -EINVAL for failure.
 */
static int amdgpu_vm_bo_split_mapping(struct amdgpu_device *adev,
				      struct dma_fence *exclusive,
				      uint64_t gtt_flags,
				      dma_addr_t *pages_addr,
				      struct amdgpu_vm *vm,
				      struct amdgpu_bo_va_mapping *mapping,
				      uint64_t flags,
				      struct drm_mm_node *nodes,
				      struct dma_fence **fence)
{
	uint64_t pfn, src = 0, start = mapping->start;
	int r;

	/* normally,bo_va->flags only contians READABLE and WIRTEABLE bit go here
	 * but in case of something, we filter the flags in first place
	 */
	if (!(mapping->flags & AMDGPU_PTE_READABLE))
		flags &= ~AMDGPU_PTE_READABLE;
	if (!(mapping->flags & AMDGPU_PTE_WRITEABLE))
		flags &= ~AMDGPU_PTE_WRITEABLE;

	flags &= ~AMDGPU_PTE_EXECUTABLE;
	flags |= mapping->flags & AMDGPU_PTE_EXECUTABLE;

	flags &= ~AMDGPU_PTE_MTYPE_MASK;
	flags |= (mapping->flags & AMDGPU_PTE_MTYPE_MASK);

	if ((mapping->flags & AMDGPU_PTE_PRT) &&
	    (adev->asic_type >= CHIP_VEGA10)) {
		flags |= AMDGPU_PTE_PRT;
		flags &= ~AMDGPU_PTE_VALID;
	}

	trace_amdgpu_vm_bo_update(mapping);

	pfn = mapping->offset >> PAGE_SHIFT;
	if (nodes) {
		while (pfn >= nodes->size) {
			pfn -= nodes->size;
			++nodes;
		}
	}

	do {
		uint64_t max_entries;
		uint64_t addr, last;

		if (nodes) {
			addr = nodes->start << PAGE_SHIFT;
			max_entries = (nodes->size - pfn) *
				(PAGE_SIZE / AMDGPU_GPU_PAGE_SIZE);
		} else {
			addr = 0;
			max_entries = S64_MAX;
		}

		if (pages_addr) {
			if (flags == gtt_flags)
				src = adev->gart.table_addr +
					(addr >> AMDGPU_GPU_PAGE_SHIFT) * 8;
			else
				max_entries = min(max_entries, 16ull * 1024ull);
			addr = 0;
		} else if (flags & AMDGPU_PTE_VALID) {
			addr += adev->vm_manager.vram_base_offset;
		}
		addr += pfn << PAGE_SHIFT;

		last = min((uint64_t)mapping->last, start + max_entries - 1);
		r = amdgpu_vm_bo_update_mapping(adev, exclusive,
						src, pages_addr, vm,
						start, last, flags, addr,
						fence);
		if (r)
			return r;

		pfn += last - start + 1;
		if (nodes && nodes->size == pfn) {
			pfn = 0;
			++nodes;
		}
		start = last + 1;

	} while (unlikely(start != mapping->last + 1));

	return 0;
}

/**
 * amdgpu_vm_bo_update - update all BO mappings in the vm page table
 *
 * @adev: amdgpu_device pointer
 * @bo_va: requested BO and VM object
 * @clear: if true clear the entries
 *
 * Fill in the page table entries for @bo_va.
 * Returns 0 for success, -EINVAL for failure.
 */
int amdgpu_vm_bo_update(struct amdgpu_device *adev,
			struct amdgpu_bo_va *bo_va,
			bool clear)
{
	struct amdgpu_vm *vm = bo_va->vm;
	struct amdgpu_bo_va_mapping *mapping;
	dma_addr_t *pages_addr = NULL;
	uint64_t gtt_flags, flags;
	struct ttm_mem_reg *mem;
	struct drm_mm_node *nodes;
	struct dma_fence *exclusive;
	int r;

	if (clear || !bo_va->bo) {
		mem = NULL;
		nodes = NULL;
		exclusive = NULL;
	} else {
		struct ttm_dma_tt *ttm;

		mem = &bo_va->bo->tbo.mem;
		nodes = mem->mm_node;
		if (mem->mem_type == TTM_PL_TT) {
			ttm = container_of(bo_va->bo->tbo.ttm, struct
					   ttm_dma_tt, ttm);
			pages_addr = ttm->dma_address;
		}
		exclusive = reservation_object_get_excl(bo_va->bo->tbo.resv);
	}

	if (bo_va->bo) {
		flags = amdgpu_ttm_tt_pte_flags(adev, bo_va->bo->tbo.ttm, mem);
		gtt_flags = (amdgpu_ttm_is_bound(bo_va->bo->tbo.ttm) &&
			adev == amdgpu_ttm_adev(bo_va->bo->tbo.bdev)) ?
			flags : 0;
	} else {
		flags = 0x0;
		gtt_flags = ~0x0;
	}

	spin_lock(&vm->status_lock);
	if (!list_empty(&bo_va->vm_status))
		list_splice_init(&bo_va->valids, &bo_va->invalids);
	spin_unlock(&vm->status_lock);

	list_for_each_entry(mapping, &bo_va->invalids, list) {
		r = amdgpu_vm_bo_split_mapping(adev, exclusive,
					       gtt_flags, pages_addr, vm,
					       mapping, flags, nodes,
					       &bo_va->last_pt_update);
		if (r)
			return r;
	}

	if (trace_amdgpu_vm_bo_mapping_enabled()) {
		list_for_each_entry(mapping, &bo_va->valids, list)
			trace_amdgpu_vm_bo_mapping(mapping);

		list_for_each_entry(mapping, &bo_va->invalids, list)
			trace_amdgpu_vm_bo_mapping(mapping);
	}

	spin_lock(&vm->status_lock);
	list_splice_init(&bo_va->invalids, &bo_va->valids);
	list_del_init(&bo_va->vm_status);
	if (clear)
		list_add(&bo_va->vm_status, &vm->cleared);
	spin_unlock(&vm->status_lock);

	return 0;
}

/**
 * amdgpu_vm_update_prt_state - update the global PRT state
 */
static void amdgpu_vm_update_prt_state(struct amdgpu_device *adev)
{
	unsigned long flags;
	bool enable;

	spin_lock_irqsave(&adev->vm_manager.prt_lock, flags);
	enable = !!atomic_read(&adev->vm_manager.num_prt_users);
	adev->gart.gart_funcs->set_prt(adev, enable);
	spin_unlock_irqrestore(&adev->vm_manager.prt_lock, flags);
}

/**
 * amdgpu_vm_prt_get - add a PRT user
 */
static void amdgpu_vm_prt_get(struct amdgpu_device *adev)
{
	if (!adev->gart.gart_funcs->set_prt)
		return;

	if (atomic_inc_return(&adev->vm_manager.num_prt_users) == 1)
		amdgpu_vm_update_prt_state(adev);
}

/**
 * amdgpu_vm_prt_put - drop a PRT user
 */
static void amdgpu_vm_prt_put(struct amdgpu_device *adev)
{
	if (atomic_dec_return(&adev->vm_manager.num_prt_users) == 0)
		amdgpu_vm_update_prt_state(adev);
}

/**
 * amdgpu_vm_prt_cb - callback for updating the PRT status
 */
static void amdgpu_vm_prt_cb(struct dma_fence *fence, struct dma_fence_cb *_cb)
{
	struct amdgpu_prt_cb *cb = container_of(_cb, struct amdgpu_prt_cb, cb);

	amdgpu_vm_prt_put(cb->adev);
	kfree(cb);
}

/**
 * amdgpu_vm_add_prt_cb - add callback for updating the PRT status
 */
static void amdgpu_vm_add_prt_cb(struct amdgpu_device *adev,
				 struct dma_fence *fence)
{
	struct amdgpu_prt_cb *cb;

	if (!adev->gart.gart_funcs->set_prt)
		return;

	cb = kmalloc(sizeof(struct amdgpu_prt_cb), GFP_KERNEL);
	if (!cb) {
		/* Last resort when we are OOM */
		if (fence)
			dma_fence_wait(fence, false);

		amdgpu_vm_prt_put(adev);
	} else {
		cb->adev = adev;
		if (!fence || dma_fence_add_callback(fence, &cb->cb,
						     amdgpu_vm_prt_cb))
			amdgpu_vm_prt_cb(fence, &cb->cb);
	}
}

/**
 * amdgpu_vm_free_mapping - free a mapping
 *
 * @adev: amdgpu_device pointer
 * @vm: requested vm
 * @mapping: mapping to be freed
 * @fence: fence of the unmap operation
 *
 * Free a mapping and make sure we decrease the PRT usage count if applicable.
 */
static void amdgpu_vm_free_mapping(struct amdgpu_device *adev,
				   struct amdgpu_vm *vm,
				   struct amdgpu_bo_va_mapping *mapping,
				   struct dma_fence *fence)
{
	if (mapping->flags & AMDGPU_PTE_PRT)
		amdgpu_vm_add_prt_cb(adev, fence);
	kfree(mapping);
}

/**
 * amdgpu_vm_prt_fini - finish all prt mappings
 *
 * @adev: amdgpu_device pointer
 * @vm: requested vm
 *
 * Register a cleanup callback to disable PRT support after VM dies.
 */
static void amdgpu_vm_prt_fini(struct amdgpu_device *adev, struct amdgpu_vm *vm)
{
	struct reservation_object *resv = vm->root.bo->tbo.resv;
	struct dma_fence *excl, **shared;
	unsigned i, shared_count;
	int r;

	r = reservation_object_get_fences_rcu(resv, &excl,
					      &shared_count, &shared);
	if (r) {
		/* Not enough memory to grab the fence list, as last resort
		 * block for all the fences to complete.
		 */
		reservation_object_wait_timeout_rcu(resv, true, false,
						    MAX_SCHEDULE_TIMEOUT);
		return;
	}

	/* Add a callback for each fence in the reservation object */
	amdgpu_vm_prt_get(adev);
	amdgpu_vm_add_prt_cb(adev, excl);

	for (i = 0; i < shared_count; ++i) {
		amdgpu_vm_prt_get(adev);
		amdgpu_vm_add_prt_cb(adev, shared[i]);
	}

	kfree(shared);
}

/**
 * amdgpu_vm_clear_freed - clear freed BOs in the PT
 *
 * @adev: amdgpu_device pointer
 * @vm: requested vm
 * @fence: optional resulting fence (unchanged if no work needed to be done
 * or if an error occurred)
 *
 * Make sure all freed BOs are cleared in the PT.
 * Returns 0 for success.
 *
 * PTs have to be reserved and mutex must be locked!
 */
int amdgpu_vm_clear_freed(struct amdgpu_device *adev,
			  struct amdgpu_vm *vm,
			  struct dma_fence **fence)
{
	struct amdgpu_bo_va_mapping *mapping;
	struct dma_fence *f = NULL;
	int r;

	while (!list_empty(&vm->freed)) {
		mapping = list_first_entry(&vm->freed,
			struct amdgpu_bo_va_mapping, list);
		list_del(&mapping->list);

		r = amdgpu_vm_bo_update_mapping(adev, NULL, 0, NULL, vm,
						mapping->start, mapping->last,
						0, 0, &f);
		amdgpu_vm_free_mapping(adev, vm, mapping, f);
		if (r) {
			dma_fence_put(f);
			return r;
		}
	}

	if (fence && f) {
		dma_fence_put(*fence);
		*fence = f;
	} else {
		dma_fence_put(f);
	}

	return 0;

}

/**
 * amdgpu_vm_clear_invalids - clear invalidated BOs in the PT
 *
 * @adev: amdgpu_device pointer
 * @vm: requested vm
 *
 * Make sure all invalidated BOs are cleared in the PT.
 * Returns 0 for success.
 *
 * PTs have to be reserved and mutex must be locked!
 */
int amdgpu_vm_clear_invalids(struct amdgpu_device *adev,
			     struct amdgpu_vm *vm, struct amdgpu_sync *sync)
{
	struct amdgpu_bo_va *bo_va = NULL;
	int r = 0;

	spin_lock(&vm->status_lock);
	while (!list_empty(&vm->invalidated)) {
		bo_va = list_first_entry(&vm->invalidated,
			struct amdgpu_bo_va, vm_status);
		spin_unlock(&vm->status_lock);

		r = amdgpu_vm_bo_update(adev, bo_va, true);
		if (r)
			return r;

		spin_lock(&vm->status_lock);
	}
	spin_unlock(&vm->status_lock);

	if (bo_va)
		r = amdgpu_sync_fence(adev, sync, bo_va->last_pt_update);

	return r;
}

/**
 * amdgpu_vm_bo_add - add a bo to a specific vm
 *
 * @adev: amdgpu_device pointer
 * @vm: requested vm
 * @bo: amdgpu buffer object
 *
 * Add @bo into the requested vm.
 * Add @bo to the list of bos associated with the vm
 * Returns newly added bo_va or NULL for failure
 *
 * Object has to be reserved!
 */
struct amdgpu_bo_va *amdgpu_vm_bo_add(struct amdgpu_device *adev,
				      struct amdgpu_vm *vm,
				      struct amdgpu_bo *bo)
{
	struct amdgpu_bo_va *bo_va;

	bo_va = kzalloc(sizeof(struct amdgpu_bo_va), GFP_KERNEL);
	if (bo_va == NULL) {
		return NULL;
	}
	bo_va->vm = vm;
	bo_va->bo = bo;
	bo_va->ref_count = 1;
	INIT_LIST_HEAD(&bo_va->bo_list);
	INIT_LIST_HEAD(&bo_va->valids);
	INIT_LIST_HEAD(&bo_va->invalids);
	INIT_LIST_HEAD(&bo_va->vm_status);

	if (bo)
		list_add_tail(&bo_va->bo_list, &bo->va);

	return bo_va;
}

/**
 * amdgpu_vm_bo_map - map bo inside a vm
 *
 * @adev: amdgpu_device pointer
 * @bo_va: bo_va to store the address
 * @saddr: where to map the BO
 * @offset: requested offset in the BO
 * @flags: attributes of pages (read/write/valid/etc.)
 *
 * Add a mapping of the BO at the specefied addr into the VM.
 * Returns 0 for success, error for failure.
 *
 * Object has to be reserved and unreserved outside!
 */
int amdgpu_vm_bo_map(struct amdgpu_device *adev,
		     struct amdgpu_bo_va *bo_va,
		     uint64_t saddr, uint64_t offset,
		     uint64_t size, uint64_t flags)
{
	struct amdgpu_bo_va_mapping *mapping, *tmp;
	struct amdgpu_vm *vm = bo_va->vm;
	uint64_t eaddr;

	/* validate the parameters */
	if (saddr & AMDGPU_GPU_PAGE_MASK || offset & AMDGPU_GPU_PAGE_MASK ||
	    size == 0 || size & AMDGPU_GPU_PAGE_MASK)
		return -EINVAL;

	/* make sure object fit at this offset */
	eaddr = saddr + size - 1;
	if (saddr >= eaddr ||
	    (bo_va->bo && offset + size > amdgpu_bo_size(bo_va->bo)))
		return -EINVAL;

	saddr /= AMDGPU_GPU_PAGE_SIZE;
	eaddr /= AMDGPU_GPU_PAGE_SIZE;

	tmp = amdgpu_vm_it_iter_first(&vm->va, saddr, eaddr);
	if (tmp) {
		/* bo and tmp overlap, invalid addr */
		dev_err(adev->dev, "bo %p va 0x%010Lx-0x%010Lx conflict with "
			"0x%010Lx-0x%010Lx\n", bo_va->bo, saddr, eaddr,
			tmp->start, tmp->last + 1);
		return -EINVAL;
	}

	mapping = kmalloc(sizeof(*mapping), GFP_KERNEL);
	if (!mapping)
		return -ENOMEM;

	INIT_LIST_HEAD(&mapping->list);
	mapping->start = saddr;
	mapping->last = eaddr;
	mapping->offset = offset;
	mapping->flags = flags;

	list_add(&mapping->list, &bo_va->invalids);
	amdgpu_vm_it_insert(mapping, &vm->va);

	if (flags & AMDGPU_PTE_PRT)
		amdgpu_vm_prt_get(adev);

	return 0;
}

/**
 * amdgpu_vm_bo_replace_map - map bo inside a vm, replacing existing mappings
 *
 * @adev: amdgpu_device pointer
 * @bo_va: bo_va to store the address
 * @saddr: where to map the BO
 * @offset: requested offset in the BO
 * @flags: attributes of pages (read/write/valid/etc.)
 *
 * Add a mapping of the BO at the specefied addr into the VM. Replace existing
 * mappings as we do so.
 * Returns 0 for success, error for failure.
 *
 * Object has to be reserved and unreserved outside!
 */
int amdgpu_vm_bo_replace_map(struct amdgpu_device *adev,
			     struct amdgpu_bo_va *bo_va,
			     uint64_t saddr, uint64_t offset,
			     uint64_t size, uint64_t flags)
{
	struct amdgpu_bo_va_mapping *mapping;
	struct amdgpu_vm *vm = bo_va->vm;
	uint64_t eaddr;
	int r;

	/* validate the parameters */
	if (saddr & AMDGPU_GPU_PAGE_MASK || offset & AMDGPU_GPU_PAGE_MASK ||
	    size == 0 || size & AMDGPU_GPU_PAGE_MASK)
		return -EINVAL;

	/* make sure object fit at this offset */
	eaddr = saddr + size - 1;
	if (saddr >= eaddr ||
	    (bo_va->bo && offset + size > amdgpu_bo_size(bo_va->bo)))
		return -EINVAL;

	/* Allocate all the needed memory */
	mapping = kmalloc(sizeof(*mapping), GFP_KERNEL);
	if (!mapping)
		return -ENOMEM;

	r = amdgpu_vm_bo_clear_mappings(adev, bo_va->vm, saddr, size);
	if (r) {
		kfree(mapping);
		return r;
	}

	saddr /= AMDGPU_GPU_PAGE_SIZE;
	eaddr /= AMDGPU_GPU_PAGE_SIZE;

	mapping->start = saddr;
	mapping->last = eaddr;
	mapping->offset = offset;
	mapping->flags = flags;

	list_add(&mapping->list, &bo_va->invalids);
	amdgpu_vm_it_insert(mapping, &vm->va);

	if (flags & AMDGPU_PTE_PRT)
		amdgpu_vm_prt_get(adev);

	return 0;
}

/**
 * amdgpu_vm_bo_unmap - remove bo mapping from vm
 *
 * @adev: amdgpu_device pointer
 * @bo_va: bo_va to remove the address from
 * @saddr: where to the BO is mapped
 *
 * Remove a mapping of the BO at the specefied addr from the VM.
 * Returns 0 for success, error for failure.
 *
 * Object has to be reserved and unreserved outside!
 */
int amdgpu_vm_bo_unmap(struct amdgpu_device *adev,
		       struct amdgpu_bo_va *bo_va,
		       uint64_t saddr)
{
	struct amdgpu_bo_va_mapping *mapping;
	struct amdgpu_vm *vm = bo_va->vm;
	bool valid = true;

	saddr /= AMDGPU_GPU_PAGE_SIZE;

	list_for_each_entry(mapping, &bo_va->valids, list) {
		if (mapping->start == saddr)
			break;
	}

	if (&mapping->list == &bo_va->valids) {
		valid = false;

		list_for_each_entry(mapping, &bo_va->invalids, list) {
			if (mapping->start == saddr)
				break;
		}

		if (&mapping->list == &bo_va->invalids)
			return -ENOENT;
	}

	list_del(&mapping->list);
	amdgpu_vm_it_remove(mapping, &vm->va);
	trace_amdgpu_vm_bo_unmap(bo_va, mapping);

	if (valid)
		list_add(&mapping->list, &vm->freed);
	else
		amdgpu_vm_free_mapping(adev, vm, mapping,
				       bo_va->last_pt_update);

	return 0;
}

/**
 * amdgpu_vm_bo_clear_mappings - remove all mappings in a specific range
 *
 * @adev: amdgpu_device pointer
 * @vm: VM structure to use
 * @saddr: start of the range
 * @size: size of the range
 *
 * Remove all mappings in a range, split them as appropriate.
 * Returns 0 for success, error for failure.
 */
int amdgpu_vm_bo_clear_mappings(struct amdgpu_device *adev,
				struct amdgpu_vm *vm,
				uint64_t saddr, uint64_t size)
{
	struct amdgpu_bo_va_mapping *before, *after, *tmp, *next;
	LIST_HEAD(removed);
	uint64_t eaddr;

	eaddr = saddr + size - 1;
	saddr /= AMDGPU_GPU_PAGE_SIZE;
	eaddr /= AMDGPU_GPU_PAGE_SIZE;

	/* Allocate all the needed memory */
	before = kzalloc(sizeof(*before), GFP_KERNEL);
	if (!before)
		return -ENOMEM;
	INIT_LIST_HEAD(&before->list);

	after = kzalloc(sizeof(*after), GFP_KERNEL);
	if (!after) {
		kfree(before);
		return -ENOMEM;
	}
	INIT_LIST_HEAD(&after->list);

	/* Now gather all removed mappings */
	tmp = amdgpu_vm_it_iter_first(&vm->va, saddr, eaddr);
	while (tmp) {
		/* Remember mapping split at the start */
		if (tmp->start < saddr) {
			before->start = tmp->start;
			before->last = saddr - 1;
			before->offset = tmp->offset;
			before->flags = tmp->flags;
			list_add(&before->list, &tmp->list);
		}

		/* Remember mapping split at the end */
		if (tmp->last > eaddr) {
			after->start = eaddr + 1;
			after->last = tmp->last;
			after->offset = tmp->offset;
			after->offset += after->start - tmp->start;
			after->flags = tmp->flags;
			list_add(&after->list, &tmp->list);
		}

		list_del(&tmp->list);
		list_add(&tmp->list, &removed);

		tmp = amdgpu_vm_it_iter_next(tmp, saddr, eaddr);
	}

	/* And free them up */
	list_for_each_entry_safe(tmp, next, &removed, list) {
		amdgpu_vm_it_remove(tmp, &vm->va);
		list_del(&tmp->list);

		if (tmp->start < saddr)
		    tmp->start = saddr;
		if (tmp->last > eaddr)
		    tmp->last = eaddr;

		list_add(&tmp->list, &vm->freed);
		trace_amdgpu_vm_bo_unmap(NULL, tmp);
	}

	/* Insert partial mapping before the range */
	if (!list_empty(&before->list)) {
		amdgpu_vm_it_insert(before, &vm->va);
		if (before->flags & AMDGPU_PTE_PRT)
			amdgpu_vm_prt_get(adev);
	} else {
		kfree(before);
	}

	/* Insert partial mapping after the range */
	if (!list_empty(&after->list)) {
		amdgpu_vm_it_insert(after, &vm->va);
		if (after->flags & AMDGPU_PTE_PRT)
			amdgpu_vm_prt_get(adev);
	} else {
		kfree(after);
	}

	return 0;
}

/**
 * amdgpu_vm_bo_rmv - remove a bo to a specific vm
 *
 * @adev: amdgpu_device pointer
 * @bo_va: requested bo_va
 *
 * Remove @bo_va->bo from the requested vm.
 *
 * Object have to be reserved!
 */
void amdgpu_vm_bo_rmv(struct amdgpu_device *adev,
		      struct amdgpu_bo_va *bo_va)
{
	struct amdgpu_bo_va_mapping *mapping, *next;
	struct amdgpu_vm *vm = bo_va->vm;

	list_del(&bo_va->bo_list);

	spin_lock(&vm->status_lock);
	list_del(&bo_va->vm_status);
	spin_unlock(&vm->status_lock);

	list_for_each_entry_safe(mapping, next, &bo_va->valids, list) {
		list_del(&mapping->list);
		amdgpu_vm_it_remove(mapping, &vm->va);
		trace_amdgpu_vm_bo_unmap(bo_va, mapping);
		list_add(&mapping->list, &vm->freed);
	}
	list_for_each_entry_safe(mapping, next, &bo_va->invalids, list) {
		list_del(&mapping->list);
		amdgpu_vm_it_remove(mapping, &vm->va);
		amdgpu_vm_free_mapping(adev, vm, mapping,
				       bo_va->last_pt_update);
	}

	dma_fence_put(bo_va->last_pt_update);
	kfree(bo_va);
}

/**
 * amdgpu_vm_bo_invalidate - mark the bo as invalid
 *
 * @adev: amdgpu_device pointer
 * @vm: requested vm
 * @bo: amdgpu buffer object
 *
 * Mark @bo as invalid.
 */
void amdgpu_vm_bo_invalidate(struct amdgpu_device *adev,
			     struct amdgpu_bo *bo)
{
	struct amdgpu_bo_va *bo_va;

	list_for_each_entry(bo_va, &bo->va, bo_list) {
		spin_lock(&bo_va->vm->status_lock);
		if (list_empty(&bo_va->vm_status))
			list_add(&bo_va->vm_status, &bo_va->vm->invalidated);
		spin_unlock(&bo_va->vm->status_lock);
	}
}

static uint32_t amdgpu_vm_get_block_size(uint64_t vm_size)
{
	/* Total bits covered by PD + PTs */
	unsigned bits = ilog2(vm_size) + 18;

	/* Make sure the PD is 4K in size up to 8GB address space.
	   Above that split equal between PD and PTs */
	if (vm_size <= 8)
		return (bits - 9);
	else
		return ((bits + 3) / 2);
}

/**
 * amdgpu_vm_adjust_size - adjust vm size and block size
 *
 * @adev: amdgpu_device pointer
 * @vm_size: the default vm size if it's set auto
 */
void amdgpu_vm_adjust_size(struct amdgpu_device *adev, uint64_t vm_size)
{
	/* adjust vm size firstly */
	if (amdgpu_vm_size == -1)
		adev->vm_manager.vm_size = vm_size;
	else
		adev->vm_manager.vm_size = amdgpu_vm_size;

	/* block size depends on vm size */
	if (amdgpu_vm_block_size == -1)
		adev->vm_manager.block_size =
			amdgpu_vm_get_block_size(adev->vm_manager.vm_size);
	else
		adev->vm_manager.block_size = amdgpu_vm_block_size;

	DRM_INFO("vm size is %llu GB, block size is %u-bit\n",
		adev->vm_manager.vm_size, adev->vm_manager.block_size);
}

/**
 * amdgpu_vm_init - initialize a vm instance
 *
 * @adev: amdgpu_device pointer
 * @vm: requested vm
 *
 * Init @vm fields.
 */
int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm)
{
	const unsigned align = min(AMDGPU_VM_PTB_ALIGN_SIZE,
		AMDGPU_VM_PTE_COUNT(adev) * 8);
	unsigned ring_instance;
	struct amdgpu_ring *ring;
	struct amd_sched_rq *rq;
	int r;

	vm->va = RB_ROOT;
	vm->client_id = atomic64_inc_return(&adev->vm_manager.client_counter);
	spin_lock_init(&vm->status_lock);
	INIT_LIST_HEAD(&vm->invalidated);
	INIT_LIST_HEAD(&vm->cleared);
	INIT_LIST_HEAD(&vm->freed);

	/* create scheduler entity for page table updates */

	ring_instance = atomic_inc_return(&adev->vm_manager.vm_pte_next_ring);
	ring_instance %= adev->vm_manager.vm_pte_num_rings;
	ring = adev->vm_manager.vm_pte_rings[ring_instance];
	rq = &ring->sched.sched_rq[AMD_SCHED_PRIORITY_KERNEL];
	r = amd_sched_entity_init(&ring->sched, &vm->entity,
				  rq, amdgpu_sched_jobs);
	if (r)
		return r;

	vm->last_dir_update = NULL;

	r = amdgpu_bo_create(adev, amdgpu_vm_bo_size(adev, 0), align, true,
			     AMDGPU_GEM_DOMAIN_VRAM,
			     AMDGPU_GEM_CREATE_NO_CPU_ACCESS |
			     AMDGPU_GEM_CREATE_SHADOW |
			     AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS |
			     AMDGPU_GEM_CREATE_VRAM_CLEARED,
			     NULL, NULL, &vm->root.bo);
	if (r)
		goto error_free_sched_entity;

	r = amdgpu_bo_reserve(vm->root.bo, false);
	if (r)
		goto error_free_root;

	vm->last_eviction_counter = atomic64_read(&adev->num_evictions);
	amdgpu_bo_unreserve(vm->root.bo);

	return 0;

error_free_root:
	amdgpu_bo_unref(&vm->root.bo->shadow);
	amdgpu_bo_unref(&vm->root.bo);
	vm->root.bo = NULL;

error_free_sched_entity:
	amd_sched_entity_fini(&ring->sched, &vm->entity);

	return r;
}

/**
 * amdgpu_vm_free_levels - free PD/PT levels
 *
 * @level: PD/PT starting level to free
 *
 * Free the page directory or page table level and all sub levels.
 */
static void amdgpu_vm_free_levels(struct amdgpu_vm_pt *level)
{
	unsigned i;

	if (level->bo) {
		amdgpu_bo_unref(&level->bo->shadow);
		amdgpu_bo_unref(&level->bo);
	}

	if (level->entries)
		for (i = 0; i <= level->last_entry_used; i++)
			amdgpu_vm_free_levels(&level->entries[i]);

	drm_free_large(level->entries);
}

/**
 * amdgpu_vm_fini - tear down a vm instance
 *
 * @adev: amdgpu_device pointer
 * @vm: requested vm
 *
 * Tear down @vm.
 * Unbind the VM and remove all bos from the vm bo list
 */
void amdgpu_vm_fini(struct amdgpu_device *adev, struct amdgpu_vm *vm)
{
	struct amdgpu_bo_va_mapping *mapping, *tmp;
	bool prt_fini_needed = !!adev->gart.gart_funcs->set_prt;

	amd_sched_entity_fini(vm->entity.sched, &vm->entity);

	if (!RB_EMPTY_ROOT(&vm->va)) {
		dev_err(adev->dev, "still active bo inside vm\n");
	}
	rbtree_postorder_for_each_entry_safe(mapping, tmp, &vm->va, rb) {
		list_del(&mapping->list);
		amdgpu_vm_it_remove(mapping, &vm->va);
		kfree(mapping);
	}
	list_for_each_entry_safe(mapping, tmp, &vm->freed, list) {
		if (mapping->flags & AMDGPU_PTE_PRT && prt_fini_needed) {
			amdgpu_vm_prt_fini(adev, vm);
			prt_fini_needed = false;
		}

		list_del(&mapping->list);
		amdgpu_vm_free_mapping(adev, vm, mapping, NULL);
	}

	amdgpu_vm_free_levels(&vm->root);
	dma_fence_put(vm->last_dir_update);
}

/**
 * amdgpu_vm_manager_init - init the VM manager
 *
 * @adev: amdgpu_device pointer
 *
 * Initialize the VM manager structures
 */
void amdgpu_vm_manager_init(struct amdgpu_device *adev)
{
	unsigned i, j;

	for (i = 0; i < AMDGPU_MAX_VMHUBS; ++i) {
		struct amdgpu_vm_id_manager *id_mgr =
			&adev->vm_manager.id_mgr[i];

		mutex_init(&id_mgr->lock);
		INIT_LIST_HEAD(&id_mgr->ids_lru);

		/* skip over VMID 0, since it is the system VM */
		for (j = 1; j < id_mgr->num_ids; ++j) {
			amdgpu_vm_reset_id(adev, i, j);
			amdgpu_sync_create(&id_mgr->ids[i].active);
			list_add_tail(&id_mgr->ids[j].list, &id_mgr->ids_lru);
		}
	}

	adev->vm_manager.fence_context =
		dma_fence_context_alloc(AMDGPU_MAX_RINGS);
	for (i = 0; i < AMDGPU_MAX_RINGS; ++i)
		adev->vm_manager.seqno[i] = 0;


	atomic_set(&adev->vm_manager.vm_pte_next_ring, 0);
	atomic64_set(&adev->vm_manager.client_counter, 0);
	spin_lock_init(&adev->vm_manager.prt_lock);
	atomic_set(&adev->vm_manager.num_prt_users, 0);
}

/**
 * amdgpu_vm_manager_fini - cleanup VM manager
 *
 * @adev: amdgpu_device pointer
 *
 * Cleanup the VM manager and free resources.
 */
void amdgpu_vm_manager_fini(struct amdgpu_device *adev)
{
	unsigned i, j;

	for (i = 0; i < AMDGPU_MAX_VMHUBS; ++i) {
		struct amdgpu_vm_id_manager *id_mgr =
			&adev->vm_manager.id_mgr[i];

		mutex_destroy(&id_mgr->lock);
		for (j = 0; j < AMDGPU_NUM_VM; ++j) {
			struct amdgpu_vm_id *id = &id_mgr->ids[j];

			amdgpu_sync_free(&id->active);
			dma_fence_put(id->flushed_updates);
			dma_fence_put(id->last_flush);
		}
	}
}