summaryrefslogtreecommitdiff
path: root/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
blob: 50fe08bf2f727656d8680fa998795d490d6df218 (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
/*
 * Copyright 2018 Advanced Micro Devices, Inc.
 *
 * Permission is hereby granted, free of charge, to any person obtaining a
 * copy of this software and associated documentation files (the "Software"),
 * to deal in the Software without restriction, including without limitation
 * the rights to use, copy, modify, merge, publish, distribute, sublicense,
 * and/or sell copies of the Software, and to permit persons to whom the
 * Software is furnished to do so, subject to the following conditions:
 *
 * The above copyright notice and this permission notice shall be included in
 * all copies or substantial portions of the Software.
 *
 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
 * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
 * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
 * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
 * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
 * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
 * OTHER DEALINGS IN THE SOFTWARE.
 *
 *
 */
#include <linux/debugfs.h>
#include <linux/list.h>
#include <linux/module.h>
#include <linux/uaccess.h>
#include <linux/reboot.h>
#include <linux/syscalls.h>

#include "amdgpu.h"
#include "amdgpu_ras.h"
#include "amdgpu_atomfirmware.h"
#include "amdgpu_xgmi.h"
#include "ivsrcid/nbio/irqsrcs_nbif_7_4.h"

const char *ras_error_string[] = {
	"none",
	"parity",
	"single_correctable",
	"multi_uncorrectable",
	"poison",
};

const char *ras_block_string[] = {
	"umc",
	"sdma",
	"gfx",
	"mmhub",
	"athub",
	"pcie_bif",
	"hdp",
	"xgmi_wafl",
	"df",
	"smn",
	"sem",
	"mp0",
	"mp1",
	"fuse",
};

#define ras_err_str(i) (ras_error_string[ffs(i)])
#define ras_block_str(i) (ras_block_string[i])

#define AMDGPU_RAS_FLAG_INIT_BY_VBIOS		1
#define AMDGPU_RAS_FLAG_INIT_NEED_RESET		2
#define RAS_DEFAULT_FLAGS (AMDGPU_RAS_FLAG_INIT_BY_VBIOS)

/* inject address is 52 bits */
#define	RAS_UMC_INJECT_ADDR_LIMIT	(0x1ULL << 52)

enum amdgpu_ras_retire_page_reservation {
	AMDGPU_RAS_RETIRE_PAGE_RESERVED,
	AMDGPU_RAS_RETIRE_PAGE_PENDING,
	AMDGPU_RAS_RETIRE_PAGE_FAULT,
};

atomic_t amdgpu_ras_in_intr = ATOMIC_INIT(0);

static bool amdgpu_ras_check_bad_page(struct amdgpu_device *adev,
				uint64_t addr);

void amdgpu_ras_set_error_query_ready(struct amdgpu_device *adev, bool ready)
{
	if (adev && amdgpu_ras_get_context(adev))
		amdgpu_ras_get_context(adev)->error_query_ready = ready;
}

bool amdgpu_ras_get_error_query_ready(struct amdgpu_device *adev)
{
	if (adev && amdgpu_ras_get_context(adev))
		return amdgpu_ras_get_context(adev)->error_query_ready;

	return false;
}

static ssize_t amdgpu_ras_debugfs_read(struct file *f, char __user *buf,
					size_t size, loff_t *pos)
{
	struct ras_manager *obj = (struct ras_manager *)file_inode(f)->i_private;
	struct ras_query_if info = {
		.head = obj->head,
	};
	ssize_t s;
	char val[128];

	if (amdgpu_ras_error_query(obj->adev, &info))
		return -EINVAL;

	s = snprintf(val, sizeof(val), "%s: %lu\n%s: %lu\n",
			"ue", info.ue_count,
			"ce", info.ce_count);
	if (*pos >= s)
		return 0;

	s -= *pos;
	s = min_t(u64, s, size);


	if (copy_to_user(buf, &val[*pos], s))
		return -EINVAL;

	*pos += s;

	return s;
}

static const struct file_operations amdgpu_ras_debugfs_ops = {
	.owner = THIS_MODULE,
	.read = amdgpu_ras_debugfs_read,
	.write = NULL,
	.llseek = default_llseek
};

static int amdgpu_ras_find_block_id_by_name(const char *name, int *block_id)
{
	int i;

	for (i = 0; i < ARRAY_SIZE(ras_block_string); i++) {
		*block_id = i;
		if (strcmp(name, ras_block_str(i)) == 0)
			return 0;
	}
	return -EINVAL;
}

static int amdgpu_ras_debugfs_ctrl_parse_data(struct file *f,
		const char __user *buf, size_t size,
		loff_t *pos, struct ras_debug_if *data)
{
	ssize_t s = min_t(u64, 64, size);
	char str[65];
	char block_name[33];
	char err[9] = "ue";
	int op = -1;
	int block_id;
	uint32_t sub_block;
	u64 address, value;

	if (*pos)
		return -EINVAL;
	*pos = size;

	memset(str, 0, sizeof(str));
	memset(data, 0, sizeof(*data));

	if (copy_from_user(str, buf, s))
		return -EINVAL;

	if (sscanf(str, "disable %32s", block_name) == 1)
		op = 0;
	else if (sscanf(str, "enable %32s %8s", block_name, err) == 2)
		op = 1;
	else if (sscanf(str, "inject %32s %8s", block_name, err) == 2)
		op = 2;
	else if (str[0] && str[1] && str[2] && str[3])
		/* ascii string, but commands are not matched. */
		return -EINVAL;

	if (op != -1) {
		if (amdgpu_ras_find_block_id_by_name(block_name, &block_id))
			return -EINVAL;

		data->head.block = block_id;
		/* only ue and ce errors are supported */
		if (!memcmp("ue", err, 2))
			data->head.type = AMDGPU_RAS_ERROR__MULTI_UNCORRECTABLE;
		else if (!memcmp("ce", err, 2))
			data->head.type = AMDGPU_RAS_ERROR__SINGLE_CORRECTABLE;
		else
			return -EINVAL;

		data->op = op;

		if (op == 2) {
			if (sscanf(str, "%*s %*s %*s %u %llu %llu",
						&sub_block, &address, &value) != 3)
				if (sscanf(str, "%*s %*s %*s 0x%x 0x%llx 0x%llx",
							&sub_block, &address, &value) != 3)
					return -EINVAL;
			data->head.sub_block_index = sub_block;
			data->inject.address = address;
			data->inject.value = value;
		}
	} else {
		if (size < sizeof(*data))
			return -EINVAL;

		if (copy_from_user(data, buf, sizeof(*data)))
			return -EINVAL;
	}

	return 0;
}

/**
 * DOC: AMDGPU RAS debugfs control interface
 *
 * It accepts struct ras_debug_if who has two members.
 *
 * First member: ras_debug_if::head or ras_debug_if::inject.
 *
 * head is used to indicate which IP block will be under control.
 *
 * head has four members, they are block, type, sub_block_index, name.
 * block: which IP will be under control.
 * type: what kind of error will be enabled/disabled/injected.
 * sub_block_index: some IPs have subcomponets. say, GFX, sDMA.
 * name: the name of IP.
 *
 * inject has two more members than head, they are address, value.
 * As their names indicate, inject operation will write the
 * value to the address.
 *
 * The second member: struct ras_debug_if::op.
 * It has three kinds of operations.
 *
 * - 0: disable RAS on the block. Take ::head as its data.
 * - 1: enable RAS on the block. Take ::head as its data.
 * - 2: inject errors on the block. Take ::inject as its data.
 *
 * How to use the interface?
 *
 * Programs
 *
 * Copy the struct ras_debug_if in your codes and initialize it.
 * Write the struct to the control node.
 *
 * Shells
 *
 * .. code-block:: bash
 *
 *	echo op block [error [sub_block address value]] > .../ras/ras_ctrl
 *
 * Parameters:
 *
 * op: disable, enable, inject
 *	disable: only block is needed
 *	enable: block and error are needed
 *	inject: error, address, value are needed
 * block: umc, sdma, gfx, .........
 *	see ras_block_string[] for details
 * error: ue, ce
 *	ue: multi_uncorrectable
 *	ce: single_correctable
 * sub_block:
 *	sub block index, pass 0 if there is no sub block
 *
 * here are some examples for bash commands:
 *
 * .. code-block:: bash
 *
 *	echo inject umc ue 0x0 0x0 0x0 > /sys/kernel/debug/dri/0/ras/ras_ctrl
 *	echo inject umc ce 0 0 0 > /sys/kernel/debug/dri/0/ras/ras_ctrl
 *	echo disable umc > /sys/kernel/debug/dri/0/ras/ras_ctrl
 *
 * How to check the result?
 *
 * For disable/enable, please check ras features at
 * /sys/class/drm/card[0/1/2...]/device/ras/features
 *
 * For inject, please check corresponding err count at
 * /sys/class/drm/card[0/1/2...]/device/ras/[gfx/sdma/...]_err_count
 *
 * .. note::
 *	Operations are only allowed on blocks which are supported.
 *	Please check ras mask at /sys/module/amdgpu/parameters/ras_mask
 *	to see which blocks support RAS on a particular asic.
 *
 */
static ssize_t amdgpu_ras_debugfs_ctrl_write(struct file *f, const char __user *buf,
		size_t size, loff_t *pos)
{
	struct amdgpu_device *adev = (struct amdgpu_device *)file_inode(f)->i_private;
	struct ras_debug_if data;
	int ret = 0;

	if (!amdgpu_ras_get_error_query_ready(adev)) {
		dev_warn(adev->dev, "RAS WARN: error injection "
				"currently inaccessible\n");
		return size;
	}

	ret = amdgpu_ras_debugfs_ctrl_parse_data(f, buf, size, pos, &data);
	if (ret)
		return -EINVAL;

	if (!amdgpu_ras_is_supported(adev, data.head.block))
		return -EINVAL;

	switch (data.op) {
	case 0:
		ret = amdgpu_ras_feature_enable(adev, &data.head, 0);
		break;
	case 1:
		ret = amdgpu_ras_feature_enable(adev, &data.head, 1);
		break;
	case 2:
		if ((data.inject.address >= adev->gmc.mc_vram_size) ||
		    (data.inject.address >= RAS_UMC_INJECT_ADDR_LIMIT)) {
			ret = -EINVAL;
			break;
		}

		/* umc ce/ue error injection for a bad page is not allowed */
		if ((data.head.block == AMDGPU_RAS_BLOCK__UMC) &&
		    amdgpu_ras_check_bad_page(adev, data.inject.address)) {
			dev_warn(adev->dev, "RAS WARN: 0x%llx has been marked "
					"as bad before error injection!\n",
					data.inject.address);
			break;
		}

		/* data.inject.address is offset instead of absolute gpu address */
		ret = amdgpu_ras_error_inject(adev, &data.inject);
		break;
	default:
		ret = -EINVAL;
		break;
	}

	if (ret)
		return -EINVAL;

	return size;
}

/**
 * DOC: AMDGPU RAS debugfs EEPROM table reset interface
 *
 * Some boards contain an EEPROM which is used to persistently store a list of
 * bad pages which experiences ECC errors in vram.  This interface provides
 * a way to reset the EEPROM, e.g., after testing error injection.
 *
 * Usage:
 *
 * .. code-block:: bash
 *
 *	echo 1 > ../ras/ras_eeprom_reset
 *
 * will reset EEPROM table to 0 entries.
 *
 */
static ssize_t amdgpu_ras_debugfs_eeprom_write(struct file *f, const char __user *buf,
		size_t size, loff_t *pos)
{
	struct amdgpu_device *adev = (struct amdgpu_device *)file_inode(f)->i_private;
	int ret;

	ret = amdgpu_ras_eeprom_reset_table(&adev->psp.ras.ras->eeprom_control);

	return ret == 1 ? size : -EIO;
}

static const struct file_operations amdgpu_ras_debugfs_ctrl_ops = {
	.owner = THIS_MODULE,
	.read = NULL,
	.write = amdgpu_ras_debugfs_ctrl_write,
	.llseek = default_llseek
};

static const struct file_operations amdgpu_ras_debugfs_eeprom_ops = {
	.owner = THIS_MODULE,
	.read = NULL,
	.write = amdgpu_ras_debugfs_eeprom_write,
	.llseek = default_llseek
};

/**
 * DOC: AMDGPU RAS sysfs Error Count Interface
 *
 * It allows the user to read the error count for each IP block on the gpu through
 * /sys/class/drm/card[0/1/2...]/device/ras/[gfx/sdma/...]_err_count
 *
 * It outputs the multiple lines which report the uncorrected (ue) and corrected
 * (ce) error counts.
 *
 * The format of one line is below,
 *
 * [ce|ue]: count
 *
 * Example:
 *
 * .. code-block:: bash
 *
 *	ue: 0
 *	ce: 1
 *
 */
static ssize_t amdgpu_ras_sysfs_read(struct device *dev,
		struct device_attribute *attr, char *buf)
{
	struct ras_manager *obj = container_of(attr, struct ras_manager, sysfs_attr);
	struct ras_query_if info = {
		.head = obj->head,
	};

	if (!amdgpu_ras_get_error_query_ready(obj->adev))
		return snprintf(buf, PAGE_SIZE,
				"Query currently inaccessible\n");

	if (amdgpu_ras_error_query(obj->adev, &info))
		return -EINVAL;

	return snprintf(buf, PAGE_SIZE, "%s: %lu\n%s: %lu\n",
			"ue", info.ue_count,
			"ce", info.ce_count);
}

/* obj begin */

#define get_obj(obj) do { (obj)->use++; } while (0)
#define alive_obj(obj) ((obj)->use)

static inline void put_obj(struct ras_manager *obj)
{
	if (obj && --obj->use == 0)
		list_del(&obj->node);
	if (obj && obj->use < 0) {
		 DRM_ERROR("RAS ERROR: Unbalance obj(%s) use\n", obj->head.name);
	}
}

/* make one obj and return it. */
static struct ras_manager *amdgpu_ras_create_obj(struct amdgpu_device *adev,
		struct ras_common_if *head)
{
	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
	struct ras_manager *obj;

	if (!con)
		return NULL;

	if (head->block >= AMDGPU_RAS_BLOCK_COUNT)
		return NULL;

	obj = &con->objs[head->block];
	/* already exist. return obj? */
	if (alive_obj(obj))
		return NULL;

	obj->head = *head;
	obj->adev = adev;
	list_add(&obj->node, &con->head);
	get_obj(obj);

	return obj;
}

/* return an obj equal to head, or the first when head is NULL */
struct ras_manager *amdgpu_ras_find_obj(struct amdgpu_device *adev,
		struct ras_common_if *head)
{
	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
	struct ras_manager *obj;
	int i;

	if (!con)
		return NULL;

	if (head) {
		if (head->block >= AMDGPU_RAS_BLOCK_COUNT)
			return NULL;

		obj = &con->objs[head->block];

		if (alive_obj(obj)) {
			WARN_ON(head->block != obj->head.block);
			return obj;
		}
	} else {
		for (i = 0; i < AMDGPU_RAS_BLOCK_COUNT; i++) {
			obj = &con->objs[i];
			if (alive_obj(obj)) {
				WARN_ON(i != obj->head.block);
				return obj;
			}
		}
	}

	return NULL;
}
/* obj end */

void amdgpu_ras_parse_status_code(struct amdgpu_device* adev,
				  const char* 		invoke_type,
				  const char* 		block_name,
				  enum ta_ras_status 	ret)
{
	switch (ret) {
	case TA_RAS_STATUS__SUCCESS:
		return;
	case TA_RAS_STATUS__ERROR_RAS_NOT_AVAILABLE:
		dev_warn(adev->dev,
			"RAS WARN: %s %s currently unavailable\n",
			invoke_type,
			block_name);
		break;
	default:
		dev_err(adev->dev,
			"RAS ERROR: %s %s error failed ret 0x%X\n",
			invoke_type,
			block_name,
			ret);
	}
}

/* feature ctl begin */
static int amdgpu_ras_is_feature_allowed(struct amdgpu_device *adev,
		struct ras_common_if *head)
{
	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);

	return con->hw_supported & BIT(head->block);
}

static int amdgpu_ras_is_feature_enabled(struct amdgpu_device *adev,
		struct ras_common_if *head)
{
	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);

	return con->features & BIT(head->block);
}

/*
 * if obj is not created, then create one.
 * set feature enable flag.
 */
static int __amdgpu_ras_feature_enable(struct amdgpu_device *adev,
		struct ras_common_if *head, int enable)
{
	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
	struct ras_manager *obj = amdgpu_ras_find_obj(adev, head);

	/* If hardware does not support ras, then do not create obj.
	 * But if hardware support ras, we can create the obj.
	 * Ras framework checks con->hw_supported to see if it need do
	 * corresponding initialization.
	 * IP checks con->support to see if it need disable ras.
	 */
	if (!amdgpu_ras_is_feature_allowed(adev, head))
		return 0;
	if (!(!!enable ^ !!amdgpu_ras_is_feature_enabled(adev, head)))
		return 0;

	if (enable) {
		if (!obj) {
			obj = amdgpu_ras_create_obj(adev, head);
			if (!obj)
				return -EINVAL;
		} else {
			/* In case we create obj somewhere else */
			get_obj(obj);
		}
		con->features |= BIT(head->block);
	} else {
		if (obj && amdgpu_ras_is_feature_enabled(adev, head)) {
			con->features &= ~BIT(head->block);
			put_obj(obj);
		}
	}

	return 0;
}

/* wrapper of psp_ras_enable_features */
int amdgpu_ras_feature_enable(struct amdgpu_device *adev,
		struct ras_common_if *head, bool enable)
{
	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
	union ta_ras_cmd_input *info;
	int ret;

	if (!con)
		return -EINVAL;

        info = kzalloc(sizeof(union ta_ras_cmd_input), GFP_KERNEL);
	if (!info)
		return -ENOMEM;

	if (!enable) {
		info->disable_features = (struct ta_ras_disable_features_input) {
			.block_id =  amdgpu_ras_block_to_ta(head->block),
			.error_type = amdgpu_ras_error_to_ta(head->type),
		};
	} else {
		info->enable_features = (struct ta_ras_enable_features_input) {
			.block_id =  amdgpu_ras_block_to_ta(head->block),
			.error_type = amdgpu_ras_error_to_ta(head->type),
		};
	}

	/* Do not enable if it is not allowed. */
	WARN_ON(enable && !amdgpu_ras_is_feature_allowed(adev, head));
	/* Are we alerady in that state we are going to set? */
	if (!(!!enable ^ !!amdgpu_ras_is_feature_enabled(adev, head))) {
		ret = 0;
		goto out;
	}

	if (!amdgpu_ras_intr_triggered()) {
		ret = psp_ras_enable_features(&adev->psp, info, enable);
		if (ret) {
			amdgpu_ras_parse_status_code(adev,
						     enable ? "enable":"disable",
						     ras_block_str(head->block),
						    (enum ta_ras_status)ret);
			if (ret == TA_RAS_STATUS__RESET_NEEDED)
				ret = -EAGAIN;
			else
				ret = -EINVAL;

			goto out;
		}
	}

	/* setup the obj */
	__amdgpu_ras_feature_enable(adev, head, enable);
	ret = 0;
out:
	kfree(info);
	return ret;
}

/* Only used in device probe stage and called only once. */
int amdgpu_ras_feature_enable_on_boot(struct amdgpu_device *adev,
		struct ras_common_if *head, bool enable)
{
	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
	int ret;

	if (!con)
		return -EINVAL;

	if (con->flags & AMDGPU_RAS_FLAG_INIT_BY_VBIOS) {
		if (enable) {
			/* There is no harm to issue a ras TA cmd regardless of
			 * the currecnt ras state.
			 * If current state == target state, it will do nothing
			 * But sometimes it requests driver to reset and repost
			 * with error code -EAGAIN.
			 */
			ret = amdgpu_ras_feature_enable(adev, head, 1);
			/* With old ras TA, we might fail to enable ras.
			 * Log it and just setup the object.
			 * TODO need remove this WA in the future.
			 */
			if (ret == -EINVAL) {
				ret = __amdgpu_ras_feature_enable(adev, head, 1);
				if (!ret)
					dev_info(adev->dev,
						"RAS INFO: %s setup object\n",
						ras_block_str(head->block));
			}
		} else {
			/* setup the object then issue a ras TA disable cmd.*/
			ret = __amdgpu_ras_feature_enable(adev, head, 1);
			if (ret)
				return ret;

			ret = amdgpu_ras_feature_enable(adev, head, 0);
		}
	} else
		ret = amdgpu_ras_feature_enable(adev, head, enable);

	return ret;
}

static int amdgpu_ras_disable_all_features(struct amdgpu_device *adev,
		bool bypass)
{
	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
	struct ras_manager *obj, *tmp;

	list_for_each_entry_safe(obj, tmp, &con->head, node) {
		/* bypass psp.
		 * aka just release the obj and corresponding flags
		 */
		if (bypass) {
			if (__amdgpu_ras_feature_enable(adev, &obj->head, 0))
				break;
		} else {
			if (amdgpu_ras_feature_enable(adev, &obj->head, 0))
				break;
		}
	}

	return con->features;
}

static int amdgpu_ras_enable_all_features(struct amdgpu_device *adev,
		bool bypass)
{
	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
	int ras_block_count = AMDGPU_RAS_BLOCK_COUNT;
	int i;
	const enum amdgpu_ras_error_type default_ras_type =
		AMDGPU_RAS_ERROR__NONE;

	for (i = 0; i < ras_block_count; i++) {
		struct ras_common_if head = {
			.block = i,
			.type = default_ras_type,
			.sub_block_index = 0,
		};
		strcpy(head.name, ras_block_str(i));
		if (bypass) {
			/*
			 * bypass psp. vbios enable ras for us.
			 * so just create the obj
			 */
			if (__amdgpu_ras_feature_enable(adev, &head, 1))
				break;
		} else {
			if (amdgpu_ras_feature_enable(adev, &head, 1))
				break;
		}
	}

	return con->features;
}
/* feature ctl end */

/* query/inject/cure begin */
int amdgpu_ras_error_query(struct amdgpu_device *adev,
		struct ras_query_if *info)
{
	struct ras_manager *obj = amdgpu_ras_find_obj(adev, &info->head);
	struct ras_err_data err_data = {0, 0, 0, NULL};
	int i;

	if (!obj)
		return -EINVAL;

	switch (info->head.block) {
	case AMDGPU_RAS_BLOCK__UMC:
		if (adev->umc.funcs->query_ras_error_count)
			adev->umc.funcs->query_ras_error_count(adev, &err_data);
		/* umc query_ras_error_address is also responsible for clearing
		 * error status
		 */
		if (adev->umc.funcs->query_ras_error_address)
			adev->umc.funcs->query_ras_error_address(adev, &err_data);
		break;
	case AMDGPU_RAS_BLOCK__SDMA:
		if (adev->sdma.funcs->query_ras_error_count) {
			for (i = 0; i < adev->sdma.num_instances; i++)
				adev->sdma.funcs->query_ras_error_count(adev, i,
									&err_data);
		}
		break;
	case AMDGPU_RAS_BLOCK__GFX:
		if (adev->gfx.funcs->query_ras_error_count)
			adev->gfx.funcs->query_ras_error_count(adev, &err_data);
		break;
	case AMDGPU_RAS_BLOCK__MMHUB:
		if (adev->mmhub.funcs->query_ras_error_count)
			adev->mmhub.funcs->query_ras_error_count(adev, &err_data);
		break;
	case AMDGPU_RAS_BLOCK__PCIE_BIF:
		if (adev->nbio.funcs->query_ras_error_count)
			adev->nbio.funcs->query_ras_error_count(adev, &err_data);
		break;
	case AMDGPU_RAS_BLOCK__XGMI_WAFL:
		amdgpu_xgmi_query_ras_error_count(adev, &err_data);
		break;
	default:
		break;
	}

	obj->err_data.ue_count += err_data.ue_count;
	obj->err_data.ce_count += err_data.ce_count;

	info->ue_count = obj->err_data.ue_count;
	info->ce_count = obj->err_data.ce_count;

	if (err_data.ce_count) {
		dev_info(adev->dev, "%ld correctable hardware errors "
					"detected in %s block, no user "
					"action is needed.\n",
					obj->err_data.ce_count,
					ras_block_str(info->head.block));
	}
	if (err_data.ue_count) {
		dev_info(adev->dev, "%ld uncorrectable hardware errors "
					"detected in %s block\n",
					obj->err_data.ue_count,
					ras_block_str(info->head.block));
	}

	return 0;
}

/* Trigger XGMI/WAFL error */
int amdgpu_ras_error_inject_xgmi(struct amdgpu_device *adev,
				 struct ta_ras_trigger_error_input *block_info)
{
	int ret;

	if (amdgpu_dpm_set_df_cstate(adev, DF_CSTATE_DISALLOW))
		dev_warn(adev->dev, "Failed to disallow df cstate");

	if (amdgpu_dpm_allow_xgmi_power_down(adev, false))
		dev_warn(adev->dev, "Failed to disallow XGMI power down");

	ret = psp_ras_trigger_error(&adev->psp, block_info);

	if (amdgpu_ras_intr_triggered())
		return ret;

	if (amdgpu_dpm_allow_xgmi_power_down(adev, true))
		dev_warn(adev->dev, "Failed to allow XGMI power down");

	if (amdgpu_dpm_set_df_cstate(adev, DF_CSTATE_DISALLOW))
		dev_warn(adev->dev, "Failed to allow df cstate");

	return ret;
}

/* wrapper of psp_ras_trigger_error */
int amdgpu_ras_error_inject(struct amdgpu_device *adev,
		struct ras_inject_if *info)
{
	struct ras_manager *obj = amdgpu_ras_find_obj(adev, &info->head);
	struct ta_ras_trigger_error_input block_info = {
		.block_id =  amdgpu_ras_block_to_ta(info->head.block),
		.inject_error_type = amdgpu_ras_error_to_ta(info->head.type),
		.sub_block_index = info->head.sub_block_index,
		.address = info->address,
		.value = info->value,
	};
	int ret = 0;

	if (!obj)
		return -EINVAL;

	/* Calculate XGMI relative offset */
	if (adev->gmc.xgmi.num_physical_nodes > 1) {
		block_info.address =
			amdgpu_xgmi_get_relative_phy_addr(adev,
							  block_info.address);
	}

	switch (info->head.block) {
	case AMDGPU_RAS_BLOCK__GFX:
		if (adev->gfx.funcs->ras_error_inject)
			ret = adev->gfx.funcs->ras_error_inject(adev, info);
		else
			ret = -EINVAL;
		break;
	case AMDGPU_RAS_BLOCK__UMC:
	case AMDGPU_RAS_BLOCK__MMHUB:
	case AMDGPU_RAS_BLOCK__PCIE_BIF:
		ret = psp_ras_trigger_error(&adev->psp, &block_info);
		break;
	case AMDGPU_RAS_BLOCK__XGMI_WAFL:
		ret = amdgpu_ras_error_inject_xgmi(adev, &block_info);
		break;
	default:
		dev_info(adev->dev, "%s error injection is not supported yet\n",
			 ras_block_str(info->head.block));
		ret = -EINVAL;
	}

	amdgpu_ras_parse_status_code(adev,
				     "inject",
				     ras_block_str(info->head.block),
				     (enum ta_ras_status)ret);

	return ret;
}

int amdgpu_ras_error_cure(struct amdgpu_device *adev,
		struct ras_cure_if *info)
{
	/* psp fw has no cure interface for now. */
	return 0;
}

/* get the total error counts on all IPs */
unsigned long amdgpu_ras_query_error_count(struct amdgpu_device *adev,
		bool is_ce)
{
	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
	struct ras_manager *obj;
	struct ras_err_data data = {0, 0};

	if (!con)
		return 0;

	list_for_each_entry(obj, &con->head, node) {
		struct ras_query_if info = {
			.head = obj->head,
		};

		if (amdgpu_ras_error_query(adev, &info))
			return 0;

		data.ce_count += info.ce_count;
		data.ue_count += info.ue_count;
	}

	return is_ce ? data.ce_count : data.ue_count;
}
/* query/inject/cure end */


/* sysfs begin */

static int amdgpu_ras_badpages_read(struct amdgpu_device *adev,
		struct ras_badpage **bps, unsigned int *count);

static char *amdgpu_ras_badpage_flags_str(unsigned int flags)
{
	switch (flags) {
	case AMDGPU_RAS_RETIRE_PAGE_RESERVED:
		return "R";
	case AMDGPU_RAS_RETIRE_PAGE_PENDING:
		return "P";
	case AMDGPU_RAS_RETIRE_PAGE_FAULT:
	default:
		return "F";
	};
}

/**
 * DOC: AMDGPU RAS sysfs gpu_vram_bad_pages Interface
 *
 * It allows user to read the bad pages of vram on the gpu through
 * /sys/class/drm/card[0/1/2...]/device/ras/gpu_vram_bad_pages
 *
 * It outputs multiple lines, and each line stands for one gpu page.
 *
 * The format of one line is below,
 * gpu pfn : gpu page size : flags
 *
 * gpu pfn and gpu page size are printed in hex format.
 * flags can be one of below character,
 *
 * R: reserved, this gpu page is reserved and not able to use.
 *
 * P: pending for reserve, this gpu page is marked as bad, will be reserved
 * in next window of page_reserve.
 *
 * F: unable to reserve. this gpu page can't be reserved due to some reasons.
 *
 * Examples:
 *
 * .. code-block:: bash
 *
 *	0x00000001 : 0x00001000 : R
 *	0x00000002 : 0x00001000 : P
 *
 */

static ssize_t amdgpu_ras_sysfs_badpages_read(struct file *f,
		struct kobject *kobj, struct bin_attribute *attr,
		char *buf, loff_t ppos, size_t count)
{
	struct amdgpu_ras *con =
		container_of(attr, struct amdgpu_ras, badpages_attr);
	struct amdgpu_device *adev = con->adev;
	const unsigned int element_size =
		sizeof("0xabcdabcd : 0x12345678 : R\n") - 1;
	unsigned int start = div64_ul(ppos + element_size - 1, element_size);
	unsigned int end = div64_ul(ppos + count - 1, element_size);
	ssize_t s = 0;
	struct ras_badpage *bps = NULL;
	unsigned int bps_count = 0;

	memset(buf, 0, count);

	if (amdgpu_ras_badpages_read(adev, &bps, &bps_count))
		return 0;

	for (; start < end && start < bps_count; start++)
		s += scnprintf(&buf[s], element_size + 1,
				"0x%08x : 0x%08x : %1s\n",
				bps[start].bp,
				bps[start].size,
				amdgpu_ras_badpage_flags_str(bps[start].flags));

	kfree(bps);

	return s;
}

static ssize_t amdgpu_ras_sysfs_features_read(struct device *dev,
		struct device_attribute *attr, char *buf)
{
	struct amdgpu_ras *con =
		container_of(attr, struct amdgpu_ras, features_attr);

	return scnprintf(buf, PAGE_SIZE, "feature mask: 0x%x\n", con->features);
}

static int amdgpu_ras_sysfs_create_feature_node(struct amdgpu_device *adev)
{
	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
	struct attribute *attrs[] = {
		&con->features_attr.attr,
		NULL
	};
	struct bin_attribute *bin_attrs[] = {
		&con->badpages_attr,
		NULL
	};
	struct attribute_group group = {
		.name = "ras",
		.attrs = attrs,
		.bin_attrs = bin_attrs,
	};

	con->features_attr = (struct device_attribute) {
		.attr = {
			.name = "features",
			.mode = S_IRUGO,
		},
			.show = amdgpu_ras_sysfs_features_read,
	};

	con->badpages_attr = (struct bin_attribute) {
		.attr = {
			.name = "gpu_vram_bad_pages",
			.mode = S_IRUGO,
		},
		.size = 0,
		.private = NULL,
		.read = amdgpu_ras_sysfs_badpages_read,
	};

	sysfs_attr_init(attrs[0]);
	sysfs_bin_attr_init(bin_attrs[0]);

	return sysfs_create_group(&adev->dev->kobj, &group);
}

static int amdgpu_ras_sysfs_remove_feature_node(struct amdgpu_device *adev)
{
	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
	struct attribute *attrs[] = {
		&con->features_attr.attr,
		NULL
	};
	struct bin_attribute *bin_attrs[] = {
		&con->badpages_attr,
		NULL
	};
	struct attribute_group group = {
		.name = "ras",
		.attrs = attrs,
		.bin_attrs = bin_attrs,
	};

	sysfs_remove_group(&adev->dev->kobj, &group);

	return 0;
}

int amdgpu_ras_sysfs_create(struct amdgpu_device *adev,
		struct ras_fs_if *head)
{
	struct ras_manager *obj = amdgpu_ras_find_obj(adev, &head->head);

	if (!obj || obj->attr_inuse)
		return -EINVAL;

	get_obj(obj);

	memcpy(obj->fs_data.sysfs_name,
			head->sysfs_name,
			sizeof(obj->fs_data.sysfs_name));

	obj->sysfs_attr = (struct device_attribute){
		.attr = {
			.name = obj->fs_data.sysfs_name,
			.mode = S_IRUGO,
		},
			.show = amdgpu_ras_sysfs_read,
	};
	sysfs_attr_init(&obj->sysfs_attr.attr);

	if (sysfs_add_file_to_group(&adev->dev->kobj,
				&obj->sysfs_attr.attr,
				"ras")) {
		put_obj(obj);
		return -EINVAL;
	}

	obj->attr_inuse = 1;

	return 0;
}

int amdgpu_ras_sysfs_remove(struct amdgpu_device *adev,
		struct ras_common_if *head)
{
	struct ras_manager *obj = amdgpu_ras_find_obj(adev, head);

	if (!obj || !obj->attr_inuse)
		return -EINVAL;

	sysfs_remove_file_from_group(&adev->dev->kobj,
				&obj->sysfs_attr.attr,
				"ras");
	obj->attr_inuse = 0;
	put_obj(obj);

	return 0;
}

static int amdgpu_ras_sysfs_remove_all(struct amdgpu_device *adev)
{
	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
	struct ras_manager *obj, *tmp;

	list_for_each_entry_safe(obj, tmp, &con->head, node) {
		amdgpu_ras_sysfs_remove(adev, &obj->head);
	}

	amdgpu_ras_sysfs_remove_feature_node(adev);

	return 0;
}
/* sysfs end */

/**
 * DOC: AMDGPU RAS Reboot Behavior for Unrecoverable Errors
 *
 * Normally when there is an uncorrectable error, the driver will reset
 * the GPU to recover.  However, in the event of an unrecoverable error,
 * the driver provides an interface to reboot the system automatically
 * in that event.
 *
 * The following file in debugfs provides that interface:
 * /sys/kernel/debug/dri/[0/1/2...]/ras/auto_reboot
 *
 * Usage:
 *
 * .. code-block:: bash
 *
 *	echo true > .../ras/auto_reboot
 *
 */
/* debugfs begin */
static void amdgpu_ras_debugfs_create_ctrl_node(struct amdgpu_device *adev)
{
	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
	struct drm_minor *minor = adev->ddev->primary;

	con->dir = debugfs_create_dir("ras", minor->debugfs_root);
	debugfs_create_file("ras_ctrl", S_IWUGO | S_IRUGO, con->dir,
				adev, &amdgpu_ras_debugfs_ctrl_ops);
	debugfs_create_file("ras_eeprom_reset", S_IWUGO | S_IRUGO, con->dir,
				adev, &amdgpu_ras_debugfs_eeprom_ops);

	/*
	 * After one uncorrectable error happens, usually GPU recovery will
	 * be scheduled. But due to the known problem in GPU recovery failing
	 * to bring GPU back, below interface provides one direct way to
	 * user to reboot system automatically in such case within
	 * ERREVENT_ATHUB_INTERRUPT generated. Normal GPU recovery routine
	 * will never be called.
	 */
	debugfs_create_bool("auto_reboot", S_IWUGO | S_IRUGO, con->dir,
				&con->reboot);
}

void amdgpu_ras_debugfs_create(struct amdgpu_device *adev,
		struct ras_fs_if *head)
{
	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
	struct ras_manager *obj = amdgpu_ras_find_obj(adev, &head->head);

	if (!obj || obj->ent)
		return;

	get_obj(obj);

	memcpy(obj->fs_data.debugfs_name,
			head->debugfs_name,
			sizeof(obj->fs_data.debugfs_name));

	obj->ent = debugfs_create_file(obj->fs_data.debugfs_name,
				       S_IWUGO | S_IRUGO, con->dir, obj,
				       &amdgpu_ras_debugfs_ops);
}

void amdgpu_ras_debugfs_create_all(struct amdgpu_device *adev)
{
	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
	struct ras_manager *obj;
	struct ras_fs_if fs_info;

	/*
	 * it won't be called in resume path, no need to check
	 * suspend and gpu reset status
	 */
	if (!con)
		return;

	amdgpu_ras_debugfs_create_ctrl_node(adev);

	list_for_each_entry(obj, &con->head, node) {
		if (amdgpu_ras_is_supported(adev, obj->head.block) &&
			(obj->attr_inuse == 1)) {
			sprintf(fs_info.debugfs_name, "%s_err_inject",
					ras_block_str(obj->head.block));
			fs_info.head = obj->head;
			amdgpu_ras_debugfs_create(adev, &fs_info);
		}
	}
}

void amdgpu_ras_debugfs_remove(struct amdgpu_device *adev,
		struct ras_common_if *head)
{
	struct ras_manager *obj = amdgpu_ras_find_obj(adev, head);

	if (!obj || !obj->ent)
		return;

	debugfs_remove(obj->ent);
	obj->ent = NULL;
	put_obj(obj);
}

static void amdgpu_ras_debugfs_remove_all(struct amdgpu_device *adev)
{
	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
	struct ras_manager *obj, *tmp;

	list_for_each_entry_safe(obj, tmp, &con->head, node) {
		amdgpu_ras_debugfs_remove(adev, &obj->head);
	}

	debugfs_remove_recursive(con->dir);
	con->dir = NULL;
}
/* debugfs end */

/* ras fs */

static int amdgpu_ras_fs_init(struct amdgpu_device *adev)
{
	amdgpu_ras_sysfs_create_feature_node(adev);

	return 0;
}

static int amdgpu_ras_fs_fini(struct amdgpu_device *adev)
{
	amdgpu_ras_debugfs_remove_all(adev);
	amdgpu_ras_sysfs_remove_all(adev);
	return 0;
}
/* ras fs end */

/* ih begin */
static void amdgpu_ras_interrupt_handler(struct ras_manager *obj)
{
	struct ras_ih_data *data = &obj->ih_data;
	struct amdgpu_iv_entry entry;
	int ret;
	struct ras_err_data err_data = {0, 0, 0, NULL};

	while (data->rptr != data->wptr) {
		rmb();
		memcpy(&entry, &data->ring[data->rptr],
				data->element_size);

		wmb();
		data->rptr = (data->aligned_element_size +
				data->rptr) % data->ring_size;

		/* Let IP handle its data, maybe we need get the output
		 * from the callback to udpate the error type/count, etc
		 */
		if (data->cb) {
			ret = data->cb(obj->adev, &err_data, &entry);
			/* ue will trigger an interrupt, and in that case
			 * we need do a reset to recovery the whole system.
			 * But leave IP do that recovery, here we just dispatch
			 * the error.
			 */
			if (ret == AMDGPU_RAS_SUCCESS) {
				/* these counts could be left as 0 if
				 * some blocks do not count error number
				 */
				obj->err_data.ue_count += err_data.ue_count;
				obj->err_data.ce_count += err_data.ce_count;
			}
		}
	}
}

static void amdgpu_ras_interrupt_process_handler(struct work_struct *work)
{
	struct ras_ih_data *data =
		container_of(work, struct ras_ih_data, ih_work);
	struct ras_manager *obj =
		container_of(data, struct ras_manager, ih_data);

	amdgpu_ras_interrupt_handler(obj);
}

int amdgpu_ras_interrupt_dispatch(struct amdgpu_device *adev,
		struct ras_dispatch_if *info)
{
	struct ras_manager *obj = amdgpu_ras_find_obj(adev, &info->head);
	struct ras_ih_data *data = &obj->ih_data;

	if (!obj)
		return -EINVAL;

	if (data->inuse == 0)
		return 0;

	/* Might be overflow... */
	memcpy(&data->ring[data->wptr], info->entry,
			data->element_size);

	wmb();
	data->wptr = (data->aligned_element_size +
			data->wptr) % data->ring_size;

	schedule_work(&data->ih_work);

	return 0;
}

int amdgpu_ras_interrupt_remove_handler(struct amdgpu_device *adev,
		struct ras_ih_if *info)
{
	struct ras_manager *obj = amdgpu_ras_find_obj(adev, &info->head);
	struct ras_ih_data *data;

	if (!obj)
		return -EINVAL;

	data = &obj->ih_data;
	if (data->inuse == 0)
		return 0;

	cancel_work_sync(&data->ih_work);

	kfree(data->ring);
	memset(data, 0, sizeof(*data));
	put_obj(obj);

	return 0;
}

int amdgpu_ras_interrupt_add_handler(struct amdgpu_device *adev,
		struct ras_ih_if *info)
{
	struct ras_manager *obj = amdgpu_ras_find_obj(adev, &info->head);
	struct ras_ih_data *data;

	if (!obj) {
		/* in case we registe the IH before enable ras feature */
		obj = amdgpu_ras_create_obj(adev, &info->head);
		if (!obj)
			return -EINVAL;
	} else
		get_obj(obj);

	data = &obj->ih_data;
	/* add the callback.etc */
	*data = (struct ras_ih_data) {
		.inuse = 0,
		.cb = info->cb,
		.element_size = sizeof(struct amdgpu_iv_entry),
		.rptr = 0,
		.wptr = 0,
	};

	INIT_WORK(&data->ih_work, amdgpu_ras_interrupt_process_handler);

	data->aligned_element_size = ALIGN(data->element_size, 8);
	/* the ring can store 64 iv entries. */
	data->ring_size = 64 * data->aligned_element_size;
	data->ring = kmalloc(data->ring_size, GFP_KERNEL);
	if (!data->ring) {
		put_obj(obj);
		return -ENOMEM;
	}

	/* IH is ready */
	data->inuse = 1;

	return 0;
}

static int amdgpu_ras_interrupt_remove_all(struct amdgpu_device *adev)
{
	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
	struct ras_manager *obj, *tmp;

	list_for_each_entry_safe(obj, tmp, &con->head, node) {
		struct ras_ih_if info = {
			.head = obj->head,
		};
		amdgpu_ras_interrupt_remove_handler(adev, &info);
	}

	return 0;
}
/* ih end */

/* traversal all IPs except NBIO to query error counter */
static void amdgpu_ras_log_on_err_counter(struct amdgpu_device *adev)
{
	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
	struct ras_manager *obj;

	if (!con)
		return;

	list_for_each_entry(obj, &con->head, node) {
		struct ras_query_if info = {
			.head = obj->head,
		};

		/*
		 * PCIE_BIF IP has one different isr by ras controller
		 * interrupt, the specific ras counter query will be
		 * done in that isr. So skip such block from common
		 * sync flood interrupt isr calling.
		 */
		if (info.head.block == AMDGPU_RAS_BLOCK__PCIE_BIF)
			continue;

		amdgpu_ras_error_query(adev, &info);
	}
}

/* recovery begin */

/* return 0 on success.
 * caller need free bps.
 */
static int amdgpu_ras_badpages_read(struct amdgpu_device *adev,
		struct ras_badpage **bps, unsigned int *count)
{
	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
	struct ras_err_handler_data *data;
	int i = 0;
	int ret = 0;

	if (!con || !con->eh_data || !bps || !count)
		return -EINVAL;

	mutex_lock(&con->recovery_lock);
	data = con->eh_data;
	if (!data || data->count == 0) {
		*bps = NULL;
		ret = -EINVAL;
		goto out;
	}

	*bps = kmalloc(sizeof(struct ras_badpage) * data->count, GFP_KERNEL);
	if (!*bps) {
		ret = -ENOMEM;
		goto out;
	}

	for (; i < data->count; i++) {
		(*bps)[i] = (struct ras_badpage){
			.bp = data->bps[i].retired_page,
			.size = AMDGPU_GPU_PAGE_SIZE,
			.flags = AMDGPU_RAS_RETIRE_PAGE_RESERVED,
		};

		if (data->last_reserved <= i)
			(*bps)[i].flags = AMDGPU_RAS_RETIRE_PAGE_PENDING;
		else if (data->bps_bo[i] == NULL)
			(*bps)[i].flags = AMDGPU_RAS_RETIRE_PAGE_FAULT;
	}

	*count = data->count;
out:
	mutex_unlock(&con->recovery_lock);
	return ret;
}

static void amdgpu_ras_do_recovery(struct work_struct *work)
{
	struct amdgpu_ras *ras =
		container_of(work, struct amdgpu_ras, recovery_work);
	struct amdgpu_device *remote_adev = NULL;
	struct amdgpu_device *adev = ras->adev;
	struct list_head device_list, *device_list_handle =  NULL;
	struct amdgpu_hive_info *hive = amdgpu_get_xgmi_hive(adev, false);

	/* Build list of devices to query RAS related errors */
	if  (hive && adev->gmc.xgmi.num_physical_nodes > 1)
		device_list_handle = &hive->device_list;
	else {
		INIT_LIST_HEAD(&device_list);
		list_add_tail(&adev->gmc.xgmi.head, &device_list);
		device_list_handle = &device_list;
	}

	list_for_each_entry(remote_adev, device_list_handle, gmc.xgmi.head) {
		amdgpu_ras_log_on_err_counter(remote_adev);
	}

	if (amdgpu_device_should_recover_gpu(ras->adev))
		amdgpu_device_gpu_recover(ras->adev, 0);
	atomic_set(&ras->in_recovery, 0);
}

/* alloc/realloc bps array */
static int amdgpu_ras_realloc_eh_data_space(struct amdgpu_device *adev,
		struct ras_err_handler_data *data, int pages)
{
	unsigned int old_space = data->count + data->space_left;
	unsigned int new_space = old_space + pages;
	unsigned int align_space = ALIGN(new_space, 512);
	void *bps = kmalloc(align_space * sizeof(*data->bps), GFP_KERNEL);
	struct amdgpu_bo **bps_bo =
			kmalloc(align_space * sizeof(*data->bps_bo), GFP_KERNEL);

	if (!bps || !bps_bo) {
		kfree(bps);
		kfree(bps_bo);
		return -ENOMEM;
	}

	if (data->bps) {
		memcpy(bps, data->bps,
				data->count * sizeof(*data->bps));
		kfree(data->bps);
	}
	if (data->bps_bo) {
		memcpy(bps_bo, data->bps_bo,
				data->count * sizeof(*data->bps_bo));
		kfree(data->bps_bo);
	}

	data->bps = bps;
	data->bps_bo = bps_bo;
	data->space_left += align_space - old_space;
	return 0;
}

/* it deal with vram only. */
int amdgpu_ras_add_bad_pages(struct amdgpu_device *adev,
		struct eeprom_table_record *bps, int pages)
{
	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
	struct ras_err_handler_data *data;
	int ret = 0;

	if (!con || !con->eh_data || !bps || pages <= 0)
		return 0;

	mutex_lock(&con->recovery_lock);
	data = con->eh_data;
	if (!data)
		goto out;

	if (data->space_left <= pages)
		if (amdgpu_ras_realloc_eh_data_space(adev, data, pages)) {
			ret = -ENOMEM;
			goto out;
		}

	memcpy(&data->bps[data->count], bps, pages * sizeof(*data->bps));
	data->count += pages;
	data->space_left -= pages;

out:
	mutex_unlock(&con->recovery_lock);

	return ret;
}

/*
 * write error record array to eeprom, the function should be
 * protected by recovery_lock
 */
static int amdgpu_ras_save_bad_pages(struct amdgpu_device *adev)
{
	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
	struct ras_err_handler_data *data;
	struct amdgpu_ras_eeprom_control *control;
	int save_count;

	if (!con || !con->eh_data)
		return 0;

	control = &con->eeprom_control;
	data = con->eh_data;
	save_count = data->count - control->num_recs;
	/* only new entries are saved */
	if (save_count > 0)
		if (amdgpu_ras_eeprom_process_recods(control,
							&data->bps[control->num_recs],
							true,
							save_count)) {
			dev_err(adev->dev, "Failed to save EEPROM table data!");
			return -EIO;
		}

	return 0;
}

/*
 * read error record array in eeprom and reserve enough space for
 * storing new bad pages
 */
static int amdgpu_ras_load_bad_pages(struct amdgpu_device *adev)
{
	struct amdgpu_ras_eeprom_control *control =
					&adev->psp.ras.ras->eeprom_control;
	struct eeprom_table_record *bps = NULL;
	int ret = 0;

	/* no bad page record, skip eeprom access */
	if (!control->num_recs)
		return ret;

	bps = kcalloc(control->num_recs, sizeof(*bps), GFP_KERNEL);
	if (!bps)
		return -ENOMEM;

	if (amdgpu_ras_eeprom_process_recods(control, bps, false,
		control->num_recs)) {
		dev_err(adev->dev, "Failed to load EEPROM table records!");
		ret = -EIO;
		goto out;
	}

	ret = amdgpu_ras_add_bad_pages(adev, bps, control->num_recs);

out:
	kfree(bps);
	return ret;
}

/*
 * check if an address belongs to bad page
 *
 * Note: this check is only for umc block
 */
static bool amdgpu_ras_check_bad_page(struct amdgpu_device *adev,
				uint64_t addr)
{
	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
	struct ras_err_handler_data *data;
	int i;
	bool ret = false;

	if (!con || !con->eh_data)
		return ret;

	mutex_lock(&con->recovery_lock);
	data = con->eh_data;
	if (!data)
		goto out;

	addr >>= AMDGPU_GPU_PAGE_SHIFT;
	for (i = 0; i < data->count; i++)
		if (addr == data->bps[i].retired_page) {
			ret = true;
			goto out;
		}

out:
	mutex_unlock(&con->recovery_lock);
	return ret;
}

/* called in gpu recovery/init */
int amdgpu_ras_reserve_bad_pages(struct amdgpu_device *adev)
{
	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
	struct ras_err_handler_data *data;
	uint64_t bp;
	struct amdgpu_bo *bo = NULL;
	int i, ret = 0;

	if (!con || !con->eh_data)
		return 0;

	mutex_lock(&con->recovery_lock);
	data = con->eh_data;
	if (!data)
		goto out;
	/* reserve vram at driver post stage. */
	for (i = data->last_reserved; i < data->count; i++) {
		bp = data->bps[i].retired_page;

		/* There are two cases of reserve error should be ignored:
		 * 1) a ras bad page has been allocated (used by someone);
		 * 2) a ras bad page has been reserved (duplicate error injection
		 *    for one page);
		 */
		if (amdgpu_bo_create_kernel_at(adev, bp << AMDGPU_GPU_PAGE_SHIFT,
					       AMDGPU_GPU_PAGE_SIZE,
					       AMDGPU_GEM_DOMAIN_VRAM,
					       &bo, NULL))
			dev_warn(adev->dev, "RAS WARN: reserve vram for "
					"retired page %llx fail\n", bp);

		data->bps_bo[i] = bo;
		data->last_reserved = i + 1;
		bo = NULL;
	}

	/* continue to save bad pages to eeprom even reesrve_vram fails */
	ret = amdgpu_ras_save_bad_pages(adev);
out:
	mutex_unlock(&con->recovery_lock);
	return ret;
}

/* called when driver unload */
static int amdgpu_ras_release_bad_pages(struct amdgpu_device *adev)
{
	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
	struct ras_err_handler_data *data;
	struct amdgpu_bo *bo;
	int i;

	if (!con || !con->eh_data)
		return 0;

	mutex_lock(&con->recovery_lock);
	data = con->eh_data;
	if (!data)
		goto out;

	for (i = data->last_reserved - 1; i >= 0; i--) {
		bo = data->bps_bo[i];

		amdgpu_bo_free_kernel(&bo, NULL, NULL);

		data->bps_bo[i] = bo;
		data->last_reserved = i;
	}
out:
	mutex_unlock(&con->recovery_lock);
	return 0;
}

int amdgpu_ras_recovery_init(struct amdgpu_device *adev)
{
	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
	struct ras_err_handler_data **data;
	int ret;

	if (con)
		data = &con->eh_data;
	else
		return 0;

	*data = kmalloc(sizeof(**data), GFP_KERNEL | __GFP_ZERO);
	if (!*data) {
		ret = -ENOMEM;
		goto out;
	}

	mutex_init(&con->recovery_lock);
	INIT_WORK(&con->recovery_work, amdgpu_ras_do_recovery);
	atomic_set(&con->in_recovery, 0);
	con->adev = adev;

	ret = amdgpu_ras_eeprom_init(&con->eeprom_control);
	if (ret)
		goto free;

	if (con->eeprom_control.num_recs) {
		ret = amdgpu_ras_load_bad_pages(adev);
		if (ret)
			goto free;
		ret = amdgpu_ras_reserve_bad_pages(adev);
		if (ret)
			goto release;
	}

	return 0;

release:
	amdgpu_ras_release_bad_pages(adev);
free:
	kfree((*data)->bps);
	kfree((*data)->bps_bo);
	kfree(*data);
	con->eh_data = NULL;
out:
	dev_warn(adev->dev, "Failed to initialize ras recovery!\n");

	return ret;
}

static int amdgpu_ras_recovery_fini(struct amdgpu_device *adev)
{
	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
	struct ras_err_handler_data *data = con->eh_data;

	/* recovery_init failed to init it, fini is useless */
	if (!data)
		return 0;

	cancel_work_sync(&con->recovery_work);
	amdgpu_ras_release_bad_pages(adev);

	mutex_lock(&con->recovery_lock);
	con->eh_data = NULL;
	kfree(data->bps);
	kfree(data->bps_bo);
	kfree(data);
	mutex_unlock(&con->recovery_lock);

	return 0;
}
/* recovery end */

/* return 0 if ras will reset gpu and repost.*/
int amdgpu_ras_request_reset_on_boot(struct amdgpu_device *adev,
		unsigned int block)
{
	struct amdgpu_ras *ras = amdgpu_ras_get_context(adev);

	if (!ras)
		return -EINVAL;

	ras->flags |= AMDGPU_RAS_FLAG_INIT_NEED_RESET;
	return 0;
}

/*
 * check hardware's ras ability which will be saved in hw_supported.
 * if hardware does not support ras, we can skip some ras initializtion and
 * forbid some ras operations from IP.
 * if software itself, say boot parameter, limit the ras ability. We still
 * need allow IP do some limited operations, like disable. In such case,
 * we have to initialize ras as normal. but need check if operation is
 * allowed or not in each function.
 */
static void amdgpu_ras_check_supported(struct amdgpu_device *adev,
		uint32_t *hw_supported, uint32_t *supported)
{
	*hw_supported = 0;
	*supported = 0;

	if (amdgpu_sriov_vf(adev) || !adev->is_atom_fw ||
	    (adev->asic_type != CHIP_VEGA20 &&
	     adev->asic_type != CHIP_ARCTURUS))
		return;

	if (amdgpu_atomfirmware_mem_ecc_supported(adev)) {
		dev_info(adev->dev, "HBM ECC is active.\n");
		*hw_supported |= (1 << AMDGPU_RAS_BLOCK__UMC |
				1 << AMDGPU_RAS_BLOCK__DF);
	} else
		dev_info(adev->dev, "HBM ECC is not presented.\n");

	if (amdgpu_atomfirmware_sram_ecc_supported(adev)) {
		dev_info(adev->dev, "SRAM ECC is active.\n");
		*hw_supported |= ~(1 << AMDGPU_RAS_BLOCK__UMC |
				1 << AMDGPU_RAS_BLOCK__DF);
	} else
		dev_info(adev->dev, "SRAM ECC is not presented.\n");

	/* hw_supported needs to be aligned with RAS block mask. */
	*hw_supported &= AMDGPU_RAS_BLOCK_MASK;

	*supported = amdgpu_ras_enable == 0 ?
			0 : *hw_supported & amdgpu_ras_mask;
}

int amdgpu_ras_init(struct amdgpu_device *adev)
{
	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
	int r;

	if (con)
		return 0;

	con = kmalloc(sizeof(struct amdgpu_ras) +
			sizeof(struct ras_manager) * AMDGPU_RAS_BLOCK_COUNT,
			GFP_KERNEL|__GFP_ZERO);
	if (!con)
		return -ENOMEM;

	con->objs = (struct ras_manager *)(con + 1);

	amdgpu_ras_set_context(adev, con);

	amdgpu_ras_check_supported(adev, &con->hw_supported,
			&con->supported);
	if (!con->hw_supported) {
		amdgpu_ras_set_context(adev, NULL);
		kfree(con);
		return 0;
	}

	con->features = 0;
	INIT_LIST_HEAD(&con->head);
	/* Might need get this flag from vbios. */
	con->flags = RAS_DEFAULT_FLAGS;

	if (adev->nbio.funcs->init_ras_controller_interrupt) {
		r = adev->nbio.funcs->init_ras_controller_interrupt(adev);
		if (r)
			return r;
	}

	if (adev->nbio.funcs->init_ras_err_event_athub_interrupt) {
		r = adev->nbio.funcs->init_ras_err_event_athub_interrupt(adev);
		if (r)
			return r;
	}

	amdgpu_ras_mask &= AMDGPU_RAS_BLOCK_MASK;

	if (amdgpu_ras_fs_init(adev))
		goto fs_out;

	dev_info(adev->dev, "RAS INFO: ras initialized successfully, "
			"hardware ability[%x] ras_mask[%x]\n",
			con->hw_supported, con->supported);
	return 0;
fs_out:
	amdgpu_ras_set_context(adev, NULL);
	kfree(con);

	return -EINVAL;
}

/* helper function to handle common stuff in ip late init phase */
int amdgpu_ras_late_init(struct amdgpu_device *adev,
			 struct ras_common_if *ras_block,
			 struct ras_fs_if *fs_info,
			 struct ras_ih_if *ih_info)
{
	int r;

	/* disable RAS feature per IP block if it is not supported */
	if (!amdgpu_ras_is_supported(adev, ras_block->block)) {
		amdgpu_ras_feature_enable_on_boot(adev, ras_block, 0);
		return 0;
	}

	r = amdgpu_ras_feature_enable_on_boot(adev, ras_block, 1);
	if (r) {
		if (r == -EAGAIN) {
			/* request gpu reset. will run again */
			amdgpu_ras_request_reset_on_boot(adev,
					ras_block->block);
			return 0;
		} else if (adev->in_suspend || adev->in_gpu_reset) {
			/* in resume phase, if fail to enable ras,
			 * clean up all ras fs nodes, and disable ras */
			goto cleanup;
		} else
			return r;
	}

	/* in resume phase, no need to create ras fs node */
	if (adev->in_suspend || adev->in_gpu_reset)
		return 0;

	if (ih_info->cb) {
		r = amdgpu_ras_interrupt_add_handler(adev, ih_info);
		if (r)
			goto interrupt;
	}

	r = amdgpu_ras_sysfs_create(adev, fs_info);
	if (r)
		goto sysfs;

	return 0;
cleanup:
	amdgpu_ras_sysfs_remove(adev, ras_block);
sysfs:
	if (ih_info->cb)
		amdgpu_ras_interrupt_remove_handler(adev, ih_info);
interrupt:
	amdgpu_ras_feature_enable(adev, ras_block, 0);
	return r;
}

/* helper function to remove ras fs node and interrupt handler */
void amdgpu_ras_late_fini(struct amdgpu_device *adev,
			  struct ras_common_if *ras_block,
			  struct ras_ih_if *ih_info)
{
	if (!ras_block || !ih_info)
		return;

	amdgpu_ras_sysfs_remove(adev, ras_block);
	if (ih_info->cb)
                amdgpu_ras_interrupt_remove_handler(adev, ih_info);
	amdgpu_ras_feature_enable(adev, ras_block, 0);
}

/* do some init work after IP late init as dependence.
 * and it runs in resume/gpu reset/booting up cases.
 */
void amdgpu_ras_resume(struct amdgpu_device *adev)
{
	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
	struct ras_manager *obj, *tmp;

	if (!con)
		return;

	if (con->flags & AMDGPU_RAS_FLAG_INIT_BY_VBIOS) {
		/* Set up all other IPs which are not implemented. There is a
		 * tricky thing that IP's actual ras error type should be
		 * MULTI_UNCORRECTABLE, but as driver does not handle it, so
		 * ERROR_NONE make sense anyway.
		 */
		amdgpu_ras_enable_all_features(adev, 1);

		/* We enable ras on all hw_supported block, but as boot
		 * parameter might disable some of them and one or more IP has
		 * not implemented yet. So we disable them on behalf.
		 */
		list_for_each_entry_safe(obj, tmp, &con->head, node) {
			if (!amdgpu_ras_is_supported(adev, obj->head.block)) {
				amdgpu_ras_feature_enable(adev, &obj->head, 0);
				/* there should be no any reference. */
				WARN_ON(alive_obj(obj));
			}
		}
	}

	if (con->flags & AMDGPU_RAS_FLAG_INIT_NEED_RESET) {
		con->flags &= ~AMDGPU_RAS_FLAG_INIT_NEED_RESET;
		/* setup ras obj state as disabled.
		 * for init_by_vbios case.
		 * if we want to enable ras, just enable it in a normal way.
		 * If we want do disable it, need setup ras obj as enabled,
		 * then issue another TA disable cmd.
		 * See feature_enable_on_boot
		 */
		amdgpu_ras_disable_all_features(adev, 1);
		amdgpu_ras_reset_gpu(adev);
	}
}

void amdgpu_ras_suspend(struct amdgpu_device *adev)
{
	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);

	if (!con)
		return;

	amdgpu_ras_disable_all_features(adev, 0);
	/* Make sure all ras objects are disabled. */
	if (con->features)
		amdgpu_ras_disable_all_features(adev, 1);
}

/* do some fini work before IP fini as dependence */
int amdgpu_ras_pre_fini(struct amdgpu_device *adev)
{
	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);

	if (!con)
		return 0;

	/* Need disable ras on all IPs here before ip [hw/sw]fini */
	amdgpu_ras_disable_all_features(adev, 0);
	amdgpu_ras_recovery_fini(adev);
	return 0;
}

int amdgpu_ras_fini(struct amdgpu_device *adev)
{
	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);

	if (!con)
		return 0;

	amdgpu_ras_fs_fini(adev);
	amdgpu_ras_interrupt_remove_all(adev);

	WARN(con->features, "Feature mask is not cleared");

	if (con->features)
		amdgpu_ras_disable_all_features(adev, 1);

	amdgpu_ras_set_context(adev, NULL);
	kfree(con);

	return 0;
}

void amdgpu_ras_global_ras_isr(struct amdgpu_device *adev)
{
	uint32_t hw_supported, supported;

	amdgpu_ras_check_supported(adev, &hw_supported, &supported);
	if (!hw_supported)
		return;

	if (atomic_cmpxchg(&amdgpu_ras_in_intr, 0, 1) == 0) {
		dev_info(adev->dev, "uncorrectable hardware error"
			"(ERREVENT_ATHUB_INTERRUPT) detected!\n");

		amdgpu_ras_reset_gpu(adev);
	}
}