summaryrefslogtreecommitdiff
path: root/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c
blob: a6c88f2fe6e5750ea42d153dcfb855046d1a5b25 (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
/*
 * 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/list.h>
#include "amdgpu.h"
#include "amdgpu_xgmi.h"
#include "amdgpu_ras.h"
#include "soc15.h"
#include "df/df_3_6_offset.h"
#include "xgmi/xgmi_4_0_0_smn.h"
#include "xgmi/xgmi_4_0_0_sh_mask.h"
#include "xgmi/xgmi_6_1_0_sh_mask.h"
#include "wafl/wafl2_4_0_0_smn.h"
#include "wafl/wafl2_4_0_0_sh_mask.h"

#include "amdgpu_reset.h"

#define smnPCS_XGMI3X16_PCS_ERROR_STATUS 0x11a0020c
#define smnPCS_XGMI3X16_PCS_ERROR_NONCORRECTABLE_MASK   0x11a00218
#define smnPCS_GOPX1_PCS_ERROR_STATUS    0x12200210
#define smnPCS_GOPX1_PCS_ERROR_NONCORRECTABLE_MASK      0x12200218

static DEFINE_MUTEX(xgmi_mutex);

#define AMDGPU_MAX_XGMI_DEVICE_PER_HIVE		4

static LIST_HEAD(xgmi_hive_list);

static const int xgmi_pcs_err_status_reg_vg20[] = {
	smnXGMI0_PCS_GOPX16_PCS_ERROR_STATUS,
	smnXGMI0_PCS_GOPX16_PCS_ERROR_STATUS + 0x100000,
};

static const int wafl_pcs_err_status_reg_vg20[] = {
	smnPCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS,
	smnPCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS + 0x100000,
};

static const int xgmi_pcs_err_status_reg_arct[] = {
	smnXGMI0_PCS_GOPX16_PCS_ERROR_STATUS,
	smnXGMI0_PCS_GOPX16_PCS_ERROR_STATUS + 0x100000,
	smnXGMI0_PCS_GOPX16_PCS_ERROR_STATUS + 0x500000,
	smnXGMI0_PCS_GOPX16_PCS_ERROR_STATUS + 0x600000,
	smnXGMI0_PCS_GOPX16_PCS_ERROR_STATUS + 0x700000,
	smnXGMI0_PCS_GOPX16_PCS_ERROR_STATUS + 0x800000,
};

/* same as vg20*/
static const int wafl_pcs_err_status_reg_arct[] = {
	smnPCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS,
	smnPCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS + 0x100000,
};

static const int xgmi3x16_pcs_err_status_reg_aldebaran[] = {
	smnPCS_XGMI3X16_PCS_ERROR_STATUS,
	smnPCS_XGMI3X16_PCS_ERROR_STATUS + 0x100000,
	smnPCS_XGMI3X16_PCS_ERROR_STATUS + 0x200000,
	smnPCS_XGMI3X16_PCS_ERROR_STATUS + 0x300000,
	smnPCS_XGMI3X16_PCS_ERROR_STATUS + 0x400000,
	smnPCS_XGMI3X16_PCS_ERROR_STATUS + 0x500000,
	smnPCS_XGMI3X16_PCS_ERROR_STATUS + 0x600000,
	smnPCS_XGMI3X16_PCS_ERROR_STATUS + 0x700000
};

static const int xgmi3x16_pcs_err_noncorrectable_mask_reg_aldebaran[] = {
	smnPCS_XGMI3X16_PCS_ERROR_NONCORRECTABLE_MASK,
	smnPCS_XGMI3X16_PCS_ERROR_NONCORRECTABLE_MASK + 0x100000,
	smnPCS_XGMI3X16_PCS_ERROR_NONCORRECTABLE_MASK + 0x200000,
	smnPCS_XGMI3X16_PCS_ERROR_NONCORRECTABLE_MASK + 0x300000,
	smnPCS_XGMI3X16_PCS_ERROR_NONCORRECTABLE_MASK + 0x400000,
	smnPCS_XGMI3X16_PCS_ERROR_NONCORRECTABLE_MASK + 0x500000,
	smnPCS_XGMI3X16_PCS_ERROR_NONCORRECTABLE_MASK + 0x600000,
	smnPCS_XGMI3X16_PCS_ERROR_NONCORRECTABLE_MASK + 0x700000
};

static const int walf_pcs_err_status_reg_aldebaran[] = {
	smnPCS_GOPX1_PCS_ERROR_STATUS,
	smnPCS_GOPX1_PCS_ERROR_STATUS + 0x100000
};

static const int walf_pcs_err_noncorrectable_mask_reg_aldebaran[] = {
	smnPCS_GOPX1_PCS_ERROR_NONCORRECTABLE_MASK,
	smnPCS_GOPX1_PCS_ERROR_NONCORRECTABLE_MASK + 0x100000
};

static const int xgmi3x16_pcs_err_status_reg_v6_4[] = {
	smnPCS_XGMI3X16_PCS_ERROR_STATUS,
	smnPCS_XGMI3X16_PCS_ERROR_STATUS + 0x100000
};

static const int xgmi3x16_pcs_err_noncorrectable_mask_reg_v6_4[] = {
	smnPCS_XGMI3X16_PCS_ERROR_NONCORRECTABLE_MASK,
	smnPCS_XGMI3X16_PCS_ERROR_NONCORRECTABLE_MASK + 0x100000
};

static const u64 xgmi_v6_4_0_mca_base_array[] = {
	0x11a09200,
	0x11b09200,
};

static const char *xgmi_v6_4_0_ras_error_code_ext[32] = {
	[0x00] = "XGMI PCS DataLossErr",
	[0x01] = "XGMI PCS TrainingErr",
	[0x02] = "XGMI PCS FlowCtrlAckErr",
	[0x03] = "XGMI PCS RxFifoUnderflowErr",
	[0x04] = "XGMI PCS RxFifoOverflowErr",
	[0x05] = "XGMI PCS CRCErr",
	[0x06] = "XGMI PCS BERExceededErr",
	[0x07] = "XGMI PCS TxMetaDataErr",
	[0x08] = "XGMI PCS ReplayBufParityErr",
	[0x09] = "XGMI PCS DataParityErr",
	[0x0a] = "XGMI PCS ReplayFifoOverflowErr",
	[0x0b] = "XGMI PCS ReplayFifoUnderflowErr",
	[0x0c] = "XGMI PCS ElasticFifoOverflowErr",
	[0x0d] = "XGMI PCS DeskewErr",
	[0x0e] = "XGMI PCS FlowCtrlCRCErr",
	[0x0f] = "XGMI PCS DataStartupLimitErr",
	[0x10] = "XGMI PCS FCInitTimeoutErr",
	[0x11] = "XGMI PCS RecoveryTimeoutErr",
	[0x12] = "XGMI PCS ReadySerialTimeoutErr",
	[0x13] = "XGMI PCS ReadySerialAttemptErr",
	[0x14] = "XGMI PCS RecoveryAttemptErr",
	[0x15] = "XGMI PCS RecoveryRelockAttemptErr",
	[0x16] = "XGMI PCS ReplayAttemptErr",
	[0x17] = "XGMI PCS SyncHdrErr",
	[0x18] = "XGMI PCS TxReplayTimeoutErr",
	[0x19] = "XGMI PCS RxReplayTimeoutErr",
	[0x1a] = "XGMI PCS LinkSubTxTimeoutErr",
	[0x1b] = "XGMI PCS LinkSubRxTimeoutErr",
	[0x1c] = "XGMI PCS RxCMDPktErr",
};

static const struct amdgpu_pcs_ras_field xgmi_pcs_ras_fields[] = {
	{"XGMI PCS DataLossErr",
	 SOC15_REG_FIELD(XGMI0_PCS_GOPX16_PCS_ERROR_STATUS, DataLossErr)},
	{"XGMI PCS TrainingErr",
	 SOC15_REG_FIELD(XGMI0_PCS_GOPX16_PCS_ERROR_STATUS, TrainingErr)},
	{"XGMI PCS CRCErr",
	 SOC15_REG_FIELD(XGMI0_PCS_GOPX16_PCS_ERROR_STATUS, CRCErr)},
	{"XGMI PCS BERExceededErr",
	 SOC15_REG_FIELD(XGMI0_PCS_GOPX16_PCS_ERROR_STATUS, BERExceededErr)},
	{"XGMI PCS TxMetaDataErr",
	 SOC15_REG_FIELD(XGMI0_PCS_GOPX16_PCS_ERROR_STATUS, TxMetaDataErr)},
	{"XGMI PCS ReplayBufParityErr",
	 SOC15_REG_FIELD(XGMI0_PCS_GOPX16_PCS_ERROR_STATUS, ReplayBufParityErr)},
	{"XGMI PCS DataParityErr",
	 SOC15_REG_FIELD(XGMI0_PCS_GOPX16_PCS_ERROR_STATUS, DataParityErr)},
	{"XGMI PCS ReplayFifoOverflowErr",
	 SOC15_REG_FIELD(XGMI0_PCS_GOPX16_PCS_ERROR_STATUS, ReplayFifoOverflowErr)},
	{"XGMI PCS ReplayFifoUnderflowErr",
	 SOC15_REG_FIELD(XGMI0_PCS_GOPX16_PCS_ERROR_STATUS, ReplayFifoUnderflowErr)},
	{"XGMI PCS ElasticFifoOverflowErr",
	 SOC15_REG_FIELD(XGMI0_PCS_GOPX16_PCS_ERROR_STATUS, ElasticFifoOverflowErr)},
	{"XGMI PCS DeskewErr",
	 SOC15_REG_FIELD(XGMI0_PCS_GOPX16_PCS_ERROR_STATUS, DeskewErr)},
	{"XGMI PCS DataStartupLimitErr",
	 SOC15_REG_FIELD(XGMI0_PCS_GOPX16_PCS_ERROR_STATUS, DataStartupLimitErr)},
	{"XGMI PCS FCInitTimeoutErr",
	 SOC15_REG_FIELD(XGMI0_PCS_GOPX16_PCS_ERROR_STATUS, FCInitTimeoutErr)},
	{"XGMI PCS RecoveryTimeoutErr",
	 SOC15_REG_FIELD(XGMI0_PCS_GOPX16_PCS_ERROR_STATUS, RecoveryTimeoutErr)},
	{"XGMI PCS ReadySerialTimeoutErr",
	 SOC15_REG_FIELD(XGMI0_PCS_GOPX16_PCS_ERROR_STATUS, ReadySerialTimeoutErr)},
	{"XGMI PCS ReadySerialAttemptErr",
	 SOC15_REG_FIELD(XGMI0_PCS_GOPX16_PCS_ERROR_STATUS, ReadySerialAttemptErr)},
	{"XGMI PCS RecoveryAttemptErr",
	 SOC15_REG_FIELD(XGMI0_PCS_GOPX16_PCS_ERROR_STATUS, RecoveryAttemptErr)},
	{"XGMI PCS RecoveryRelockAttemptErr",
	 SOC15_REG_FIELD(XGMI0_PCS_GOPX16_PCS_ERROR_STATUS, RecoveryRelockAttemptErr)},
};

static const struct amdgpu_pcs_ras_field wafl_pcs_ras_fields[] = {
	{"WAFL PCS DataLossErr",
	 SOC15_REG_FIELD(PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS, DataLossErr)},
	{"WAFL PCS TrainingErr",
	 SOC15_REG_FIELD(PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS, TrainingErr)},
	{"WAFL PCS CRCErr",
	 SOC15_REG_FIELD(PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS, CRCErr)},
	{"WAFL PCS BERExceededErr",
	 SOC15_REG_FIELD(PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS, BERExceededErr)},
	{"WAFL PCS TxMetaDataErr",
	 SOC15_REG_FIELD(PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS, TxMetaDataErr)},
	{"WAFL PCS ReplayBufParityErr",
	 SOC15_REG_FIELD(PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS, ReplayBufParityErr)},
	{"WAFL PCS DataParityErr",
	 SOC15_REG_FIELD(PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS, DataParityErr)},
	{"WAFL PCS ReplayFifoOverflowErr",
	 SOC15_REG_FIELD(PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS, ReplayFifoOverflowErr)},
	{"WAFL PCS ReplayFifoUnderflowErr",
	 SOC15_REG_FIELD(PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS, ReplayFifoUnderflowErr)},
	{"WAFL PCS ElasticFifoOverflowErr",
	 SOC15_REG_FIELD(PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS, ElasticFifoOverflowErr)},
	{"WAFL PCS DeskewErr",
	 SOC15_REG_FIELD(PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS, DeskewErr)},
	{"WAFL PCS DataStartupLimitErr",
	 SOC15_REG_FIELD(PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS, DataStartupLimitErr)},
	{"WAFL PCS FCInitTimeoutErr",
	 SOC15_REG_FIELD(PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS, FCInitTimeoutErr)},
	{"WAFL PCS RecoveryTimeoutErr",
	 SOC15_REG_FIELD(PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS, RecoveryTimeoutErr)},
	{"WAFL PCS ReadySerialTimeoutErr",
	 SOC15_REG_FIELD(PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS, ReadySerialTimeoutErr)},
	{"WAFL PCS ReadySerialAttemptErr",
	 SOC15_REG_FIELD(PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS, ReadySerialAttemptErr)},
	{"WAFL PCS RecoveryAttemptErr",
	 SOC15_REG_FIELD(PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS, RecoveryAttemptErr)},
	{"WAFL PCS RecoveryRelockAttemptErr",
	 SOC15_REG_FIELD(PCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS, RecoveryRelockAttemptErr)},
};

static const struct amdgpu_pcs_ras_field xgmi3x16_pcs_ras_fields[] = {
	{"XGMI3X16 PCS DataLossErr",
	 SOC15_REG_FIELD(PCS_XGMI3X16_PCS_ERROR_STATUS, DataLossErr)},
	{"XGMI3X16 PCS TrainingErr",
	 SOC15_REG_FIELD(PCS_XGMI3X16_PCS_ERROR_STATUS, TrainingErr)},
	{"XGMI3X16 PCS FlowCtrlAckErr",
	 SOC15_REG_FIELD(PCS_XGMI3X16_PCS_ERROR_STATUS, FlowCtrlAckErr)},
	{"XGMI3X16 PCS RxFifoUnderflowErr",
	 SOC15_REG_FIELD(PCS_XGMI3X16_PCS_ERROR_STATUS, RxFifoUnderflowErr)},
	{"XGMI3X16 PCS RxFifoOverflowErr",
	 SOC15_REG_FIELD(PCS_XGMI3X16_PCS_ERROR_STATUS, RxFifoOverflowErr)},
	{"XGMI3X16 PCS CRCErr",
	 SOC15_REG_FIELD(PCS_XGMI3X16_PCS_ERROR_STATUS, CRCErr)},
	{"XGMI3X16 PCS BERExceededErr",
	 SOC15_REG_FIELD(PCS_XGMI3X16_PCS_ERROR_STATUS, BERExceededErr)},
	{"XGMI3X16 PCS TxVcidDataErr",
	 SOC15_REG_FIELD(PCS_XGMI3X16_PCS_ERROR_STATUS, TxVcidDataErr)},
	{"XGMI3X16 PCS ReplayBufParityErr",
	 SOC15_REG_FIELD(PCS_XGMI3X16_PCS_ERROR_STATUS, ReplayBufParityErr)},
	{"XGMI3X16 PCS DataParityErr",
	 SOC15_REG_FIELD(PCS_XGMI3X16_PCS_ERROR_STATUS, DataParityErr)},
	{"XGMI3X16 PCS ReplayFifoOverflowErr",
	 SOC15_REG_FIELD(PCS_XGMI3X16_PCS_ERROR_STATUS, ReplayFifoOverflowErr)},
	{"XGMI3X16 PCS ReplayFifoUnderflowErr",
	 SOC15_REG_FIELD(PCS_XGMI3X16_PCS_ERROR_STATUS, ReplayFifoUnderflowErr)},
	{"XGMI3X16 PCS ElasticFifoOverflowErr",
	 SOC15_REG_FIELD(PCS_XGMI3X16_PCS_ERROR_STATUS, ElasticFifoOverflowErr)},
	{"XGMI3X16 PCS DeskewErr",
	 SOC15_REG_FIELD(PCS_XGMI3X16_PCS_ERROR_STATUS, DeskewErr)},
	{"XGMI3X16 PCS FlowCtrlCRCErr",
	 SOC15_REG_FIELD(PCS_XGMI3X16_PCS_ERROR_STATUS, FlowCtrlCRCErr)},
	{"XGMI3X16 PCS DataStartupLimitErr",
	 SOC15_REG_FIELD(PCS_XGMI3X16_PCS_ERROR_STATUS, DataStartupLimitErr)},
	{"XGMI3X16 PCS FCInitTimeoutErr",
	 SOC15_REG_FIELD(PCS_XGMI3X16_PCS_ERROR_STATUS, FCInitTimeoutErr)},
	{"XGMI3X16 PCS RecoveryTimeoutErr",
	 SOC15_REG_FIELD(PCS_XGMI3X16_PCS_ERROR_STATUS, RecoveryTimeoutErr)},
	{"XGMI3X16 PCS ReadySerialTimeoutErr",
	 SOC15_REG_FIELD(PCS_XGMI3X16_PCS_ERROR_STATUS, ReadySerialTimeoutErr)},
	{"XGMI3X16 PCS ReadySerialAttemptErr",
	 SOC15_REG_FIELD(PCS_XGMI3X16_PCS_ERROR_STATUS, ReadySerialAttemptErr)},
	{"XGMI3X16 PCS RecoveryAttemptErr",
	 SOC15_REG_FIELD(PCS_XGMI3X16_PCS_ERROR_STATUS, RecoveryAttemptErr)},
	{"XGMI3X16 PCS RecoveryRelockAttemptErr",
	 SOC15_REG_FIELD(PCS_XGMI3X16_PCS_ERROR_STATUS, RecoveryRelockAttemptErr)},
	{"XGMI3X16 PCS ReplayAttemptErr",
	 SOC15_REG_FIELD(PCS_XGMI3X16_PCS_ERROR_STATUS, ReplayAttemptErr)},
	{"XGMI3X16 PCS SyncHdrErr",
	 SOC15_REG_FIELD(PCS_XGMI3X16_PCS_ERROR_STATUS, SyncHdrErr)},
	{"XGMI3X16 PCS TxReplayTimeoutErr",
	 SOC15_REG_FIELD(PCS_XGMI3X16_PCS_ERROR_STATUS, TxReplayTimeoutErr)},
	{"XGMI3X16 PCS RxReplayTimeoutErr",
	 SOC15_REG_FIELD(PCS_XGMI3X16_PCS_ERROR_STATUS, RxReplayTimeoutErr)},
	{"XGMI3X16 PCS LinkSubTxTimeoutErr",
	 SOC15_REG_FIELD(PCS_XGMI3X16_PCS_ERROR_STATUS, LinkSubTxTimeoutErr)},
	{"XGMI3X16 PCS LinkSubRxTimeoutErr",
	 SOC15_REG_FIELD(PCS_XGMI3X16_PCS_ERROR_STATUS, LinkSubRxTimeoutErr)},
	{"XGMI3X16 PCS RxCMDPktErr",
	 SOC15_REG_FIELD(PCS_XGMI3X16_PCS_ERROR_STATUS, RxCMDPktErr)},
};

/**
 * DOC: AMDGPU XGMI Support
 *
 * XGMI is a high speed interconnect that joins multiple GPU cards
 * into a homogeneous memory space that is organized by a collective
 * hive ID and individual node IDs, both of which are 64-bit numbers.
 *
 * The file xgmi_device_id contains the unique per GPU device ID and
 * is stored in the /sys/class/drm/card${cardno}/device/ directory.
 *
 * Inside the device directory a sub-directory 'xgmi_hive_info' is
 * created which contains the hive ID and the list of nodes.
 *
 * The hive ID is stored in:
 *   /sys/class/drm/card${cardno}/device/xgmi_hive_info/xgmi_hive_id
 *
 * The node information is stored in numbered directories:
 *   /sys/class/drm/card${cardno}/device/xgmi_hive_info/node${nodeno}/xgmi_device_id
 *
 * Each device has their own xgmi_hive_info direction with a mirror
 * set of node sub-directories.
 *
 * The XGMI memory space is built by contiguously adding the power of
 * two padded VRAM space from each node to each other.
 *
 */

static struct attribute amdgpu_xgmi_hive_id = {
	.name = "xgmi_hive_id",
	.mode = S_IRUGO
};

static struct attribute *amdgpu_xgmi_hive_attrs[] = {
	&amdgpu_xgmi_hive_id,
	NULL
};
ATTRIBUTE_GROUPS(amdgpu_xgmi_hive);

static ssize_t amdgpu_xgmi_show_attrs(struct kobject *kobj,
	struct attribute *attr, char *buf)
{
	struct amdgpu_hive_info *hive = container_of(
		kobj, struct amdgpu_hive_info, kobj);

	if (attr == &amdgpu_xgmi_hive_id)
		return snprintf(buf, PAGE_SIZE, "%llu\n", hive->hive_id);

	return 0;
}

static void amdgpu_xgmi_hive_release(struct kobject *kobj)
{
	struct amdgpu_hive_info *hive = container_of(
		kobj, struct amdgpu_hive_info, kobj);

	amdgpu_reset_put_reset_domain(hive->reset_domain);
	hive->reset_domain = NULL;

	mutex_destroy(&hive->hive_lock);
	kfree(hive);
}

static const struct sysfs_ops amdgpu_xgmi_hive_ops = {
	.show = amdgpu_xgmi_show_attrs,
};

static const struct kobj_type amdgpu_xgmi_hive_type = {
	.release = amdgpu_xgmi_hive_release,
	.sysfs_ops = &amdgpu_xgmi_hive_ops,
	.default_groups = amdgpu_xgmi_hive_groups,
};

static ssize_t amdgpu_xgmi_show_device_id(struct device *dev,
				     struct device_attribute *attr,
				     char *buf)
{
	struct drm_device *ddev = dev_get_drvdata(dev);
	struct amdgpu_device *adev = drm_to_adev(ddev);

	return sysfs_emit(buf, "%llu\n", adev->gmc.xgmi.node_id);

}

static ssize_t amdgpu_xgmi_show_physical_id(struct device *dev,
				     struct device_attribute *attr,
				     char *buf)
{
	struct drm_device *ddev = dev_get_drvdata(dev);
	struct amdgpu_device *adev = drm_to_adev(ddev);

	return sysfs_emit(buf, "%u\n", adev->gmc.xgmi.physical_node_id);

}

static ssize_t amdgpu_xgmi_show_num_hops(struct device *dev,
					struct device_attribute *attr,
					char *buf)
{
	struct drm_device *ddev = dev_get_drvdata(dev);
	struct amdgpu_device *adev = drm_to_adev(ddev);
	struct psp_xgmi_topology_info *top = &adev->psp.xgmi_context.top_info;
	int i;

	for (i = 0; i < top->num_nodes; i++)
		sprintf(buf + 3 * i, "%02x ", top->nodes[i].num_hops);

	return sysfs_emit(buf, "%s\n", buf);
}

static ssize_t amdgpu_xgmi_show_num_links(struct device *dev,
					struct device_attribute *attr,
					char *buf)
{
	struct drm_device *ddev = dev_get_drvdata(dev);
	struct amdgpu_device *adev = drm_to_adev(ddev);
	struct psp_xgmi_topology_info *top = &adev->psp.xgmi_context.top_info;
	int i;

	for (i = 0; i < top->num_nodes; i++)
		sprintf(buf + 3 * i, "%02x ", top->nodes[i].num_links);

	return sysfs_emit(buf, "%s\n", buf);
}

static ssize_t amdgpu_xgmi_show_connected_port_num(struct device *dev,
					struct device_attribute *attr,
					char *buf)
{
	struct drm_device *ddev = dev_get_drvdata(dev);
	struct amdgpu_device *adev = drm_to_adev(ddev);
	struct psp_xgmi_topology_info *top = &adev->psp.xgmi_context.top_info;
	int i, j, size = 0;
	int current_node;
	/*
	 * get the node id in the sysfs for the current socket and show
	 * it in the port num info output in the sysfs for easy reading.
	 * it is NOT the one retrieved from xgmi ta.
	 */
	for (i = 0; i < top->num_nodes; i++) {
		if (top->nodes[i].node_id == adev->gmc.xgmi.node_id) {
			current_node = i;
			break;
		}
	}

	for (i = 0; i < top->num_nodes; i++) {
		for (j = 0; j < top->nodes[i].num_links; j++)
			/* node id in sysfs starts from 1 rather than 0 so +1 here */
			size += sysfs_emit_at(buf, size, "%02x:%02x ->  %02x:%02x\n", current_node + 1,
					      top->nodes[i].port_num[j].src_xgmi_port_num, i + 1,
					      top->nodes[i].port_num[j].dst_xgmi_port_num);
	}

	return size;
}

#define AMDGPU_XGMI_SET_FICAA(o)	((o) | 0x456801)
static ssize_t amdgpu_xgmi_show_error(struct device *dev,
				      struct device_attribute *attr,
				      char *buf)
{
	struct drm_device *ddev = dev_get_drvdata(dev);
	struct amdgpu_device *adev = drm_to_adev(ddev);
	uint32_t ficaa_pie_ctl_in, ficaa_pie_status_in;
	uint64_t fica_out;
	unsigned int error_count = 0;

	ficaa_pie_ctl_in = AMDGPU_XGMI_SET_FICAA(0x200);
	ficaa_pie_status_in = AMDGPU_XGMI_SET_FICAA(0x208);

	if ((!adev->df.funcs) ||
	    (!adev->df.funcs->get_fica) ||
	    (!adev->df.funcs->set_fica))
		return -EINVAL;

	fica_out = adev->df.funcs->get_fica(adev, ficaa_pie_ctl_in);
	if (fica_out != 0x1f)
		pr_err("xGMI error counters not enabled!\n");

	fica_out = adev->df.funcs->get_fica(adev, ficaa_pie_status_in);

	if ((fica_out & 0xffff) == 2)
		error_count = ((fica_out >> 62) & 0x1) + (fica_out >> 63);

	adev->df.funcs->set_fica(adev, ficaa_pie_status_in, 0, 0);

	return sysfs_emit(buf, "%u\n", error_count);
}


static DEVICE_ATTR(xgmi_device_id, S_IRUGO, amdgpu_xgmi_show_device_id, NULL);
static DEVICE_ATTR(xgmi_physical_id, 0444, amdgpu_xgmi_show_physical_id, NULL);
static DEVICE_ATTR(xgmi_error, S_IRUGO, amdgpu_xgmi_show_error, NULL);
static DEVICE_ATTR(xgmi_num_hops, S_IRUGO, amdgpu_xgmi_show_num_hops, NULL);
static DEVICE_ATTR(xgmi_num_links, S_IRUGO, amdgpu_xgmi_show_num_links, NULL);
static DEVICE_ATTR(xgmi_port_num, S_IRUGO, amdgpu_xgmi_show_connected_port_num, NULL);

static int amdgpu_xgmi_sysfs_add_dev_info(struct amdgpu_device *adev,
					 struct amdgpu_hive_info *hive)
{
	int ret = 0;
	char node[10] = { 0 };

	/* Create xgmi device id file */
	ret = device_create_file(adev->dev, &dev_attr_xgmi_device_id);
	if (ret) {
		dev_err(adev->dev, "XGMI: Failed to create device file xgmi_device_id\n");
		return ret;
	}

	ret = device_create_file(adev->dev, &dev_attr_xgmi_physical_id);
	if (ret) {
		dev_err(adev->dev, "XGMI: Failed to create device file xgmi_physical_id\n");
		return ret;
	}

	/* Create xgmi error file */
	ret = device_create_file(adev->dev, &dev_attr_xgmi_error);
	if (ret)
		pr_err("failed to create xgmi_error\n");

	/* Create xgmi num hops file */
	ret = device_create_file(adev->dev, &dev_attr_xgmi_num_hops);
	if (ret)
		pr_err("failed to create xgmi_num_hops\n");

	/* Create xgmi num links file */
	ret = device_create_file(adev->dev, &dev_attr_xgmi_num_links);
	if (ret)
		pr_err("failed to create xgmi_num_links\n");

	/* Create xgmi port num file if supported */
	if (adev->psp.xgmi_context.xgmi_ta_caps & EXTEND_PEER_LINK_INFO_CMD_FLAG) {
		ret = device_create_file(adev->dev, &dev_attr_xgmi_port_num);
		if (ret)
			dev_err(adev->dev, "failed to create xgmi_port_num\n");
	}

	/* Create sysfs link to hive info folder on the first device */
	if (hive->kobj.parent != (&adev->dev->kobj)) {
		ret = sysfs_create_link(&adev->dev->kobj, &hive->kobj,
					"xgmi_hive_info");
		if (ret) {
			dev_err(adev->dev, "XGMI: Failed to create link to hive info");
			goto remove_file;
		}
	}

	sprintf(node, "node%d", atomic_read(&hive->number_devices));
	/* Create sysfs link form the hive folder to yourself */
	ret = sysfs_create_link(&hive->kobj, &adev->dev->kobj, node);
	if (ret) {
		dev_err(adev->dev, "XGMI: Failed to create link from hive info");
		goto remove_link;
	}

	goto success;


remove_link:
	sysfs_remove_link(&adev->dev->kobj, adev_to_drm(adev)->unique);

remove_file:
	device_remove_file(adev->dev, &dev_attr_xgmi_device_id);
	device_remove_file(adev->dev, &dev_attr_xgmi_physical_id);
	device_remove_file(adev->dev, &dev_attr_xgmi_error);
	device_remove_file(adev->dev, &dev_attr_xgmi_num_hops);
	device_remove_file(adev->dev, &dev_attr_xgmi_num_links);
	if (adev->psp.xgmi_context.xgmi_ta_caps & EXTEND_PEER_LINK_INFO_CMD_FLAG)
		device_remove_file(adev->dev, &dev_attr_xgmi_port_num);

success:
	return ret;
}

static void amdgpu_xgmi_sysfs_rem_dev_info(struct amdgpu_device *adev,
					  struct amdgpu_hive_info *hive)
{
	char node[10];
	memset(node, 0, sizeof(node));

	device_remove_file(adev->dev, &dev_attr_xgmi_device_id);
	device_remove_file(adev->dev, &dev_attr_xgmi_physical_id);
	device_remove_file(adev->dev, &dev_attr_xgmi_error);
	device_remove_file(adev->dev, &dev_attr_xgmi_num_hops);
	device_remove_file(adev->dev, &dev_attr_xgmi_num_links);
	if (adev->psp.xgmi_context.xgmi_ta_caps & EXTEND_PEER_LINK_INFO_CMD_FLAG)
		device_remove_file(adev->dev, &dev_attr_xgmi_port_num);

	if (hive->kobj.parent != (&adev->dev->kobj))
		sysfs_remove_link(&adev->dev->kobj,"xgmi_hive_info");

	sprintf(node, "node%d", atomic_read(&hive->number_devices));
	sysfs_remove_link(&hive->kobj, node);

}



struct amdgpu_hive_info *amdgpu_get_xgmi_hive(struct amdgpu_device *adev)
{
	struct amdgpu_hive_info *hive = NULL;
	int ret;

	if (!adev->gmc.xgmi.hive_id)
		return NULL;

	if (adev->hive) {
		kobject_get(&adev->hive->kobj);
		return adev->hive;
	}

	mutex_lock(&xgmi_mutex);

	list_for_each_entry(hive, &xgmi_hive_list, node)  {
		if (hive->hive_id == adev->gmc.xgmi.hive_id)
			goto pro_end;
	}

	hive = kzalloc(sizeof(*hive), GFP_KERNEL);
	if (!hive) {
		dev_err(adev->dev, "XGMI: allocation failed\n");
		ret = -ENOMEM;
		hive = NULL;
		goto pro_end;
	}

	/* initialize new hive if not exist */
	ret = kobject_init_and_add(&hive->kobj,
			&amdgpu_xgmi_hive_type,
			&adev->dev->kobj,
			"%s", "xgmi_hive_info");
	if (ret) {
		dev_err(adev->dev, "XGMI: failed initializing kobject for xgmi hive\n");
		kobject_put(&hive->kobj);
		hive = NULL;
		goto pro_end;
	}

	/**
	 * Only init hive->reset_domain for none SRIOV configuration. For SRIOV,
	 * Host driver decide how to reset the GPU either through FLR or chain reset.
	 * Guest side will get individual notifications from the host for the FLR
	 * if necessary.
	 */
	if (!amdgpu_sriov_vf(adev)) {
	/**
	 * Avoid recreating reset domain when hive is reconstructed for the case
	 * of reset the devices in the XGMI hive during probe for passthrough GPU
	 * See https://www.spinics.net/lists/amd-gfx/msg58836.html
	 */
		if (adev->reset_domain->type != XGMI_HIVE) {
			hive->reset_domain =
				amdgpu_reset_create_reset_domain(XGMI_HIVE, "amdgpu-reset-hive");
			if (!hive->reset_domain) {
				dev_err(adev->dev, "XGMI: failed initializing reset domain for xgmi hive\n");
				ret = -ENOMEM;
				kobject_put(&hive->kobj);
				hive = NULL;
				goto pro_end;
			}
		} else {
			amdgpu_reset_get_reset_domain(adev->reset_domain);
			hive->reset_domain = adev->reset_domain;
		}
	}

	hive->hive_id = adev->gmc.xgmi.hive_id;
	INIT_LIST_HEAD(&hive->device_list);
	INIT_LIST_HEAD(&hive->node);
	mutex_init(&hive->hive_lock);
	atomic_set(&hive->number_devices, 0);
	task_barrier_init(&hive->tb);
	hive->pstate = AMDGPU_XGMI_PSTATE_UNKNOWN;
	hive->hi_req_gpu = NULL;

	/*
	 * hive pstate on boot is high in vega20 so we have to go to low
	 * pstate on after boot.
	 */
	hive->hi_req_count = AMDGPU_MAX_XGMI_DEVICE_PER_HIVE;
	list_add_tail(&hive->node, &xgmi_hive_list);

pro_end:
	if (hive)
		kobject_get(&hive->kobj);
	mutex_unlock(&xgmi_mutex);
	return hive;
}

void amdgpu_put_xgmi_hive(struct amdgpu_hive_info *hive)
{
	if (hive)
		kobject_put(&hive->kobj);
}

int amdgpu_xgmi_set_pstate(struct amdgpu_device *adev, int pstate)
{
	int ret = 0;
	struct amdgpu_hive_info *hive;
	struct amdgpu_device *request_adev;
	bool is_hi_req = pstate == AMDGPU_XGMI_PSTATE_MAX_VEGA20;
	bool init_low;

	hive = amdgpu_get_xgmi_hive(adev);
	if (!hive)
		return 0;

	request_adev = hive->hi_req_gpu ? hive->hi_req_gpu : adev;
	init_low = hive->pstate == AMDGPU_XGMI_PSTATE_UNKNOWN;
	amdgpu_put_xgmi_hive(hive);
	/* fw bug so temporarily disable pstate switching */
	return 0;

	if (!hive || adev->asic_type != CHIP_VEGA20)
		return 0;

	mutex_lock(&hive->hive_lock);

	if (is_hi_req)
		hive->hi_req_count++;
	else
		hive->hi_req_count--;

	/*
	 * Vega20 only needs single peer to request pstate high for the hive to
	 * go high but all peers must request pstate low for the hive to go low
	 */
	if (hive->pstate == pstate ||
			(!is_hi_req && hive->hi_req_count && !init_low))
		goto out;

	dev_dbg(request_adev->dev, "Set xgmi pstate %d.\n", pstate);

	ret = amdgpu_dpm_set_xgmi_pstate(request_adev, pstate);
	if (ret) {
		dev_err(request_adev->dev,
			"XGMI: Set pstate failure on device %llx, hive %llx, ret %d",
			request_adev->gmc.xgmi.node_id,
			request_adev->gmc.xgmi.hive_id, ret);
		goto out;
	}

	if (init_low)
		hive->pstate = hive->hi_req_count ?
					hive->pstate : AMDGPU_XGMI_PSTATE_MIN;
	else {
		hive->pstate = pstate;
		hive->hi_req_gpu = pstate != AMDGPU_XGMI_PSTATE_MIN ?
							adev : NULL;
	}
out:
	mutex_unlock(&hive->hive_lock);
	return ret;
}

int amdgpu_xgmi_update_topology(struct amdgpu_hive_info *hive, struct amdgpu_device *adev)
{
	int ret;

	if (amdgpu_sriov_vf(adev))
		return 0;

	/* Each psp need to set the latest topology */
	ret = psp_xgmi_set_topology_info(&adev->psp,
					 atomic_read(&hive->number_devices),
					 &adev->psp.xgmi_context.top_info);
	if (ret)
		dev_err(adev->dev,
			"XGMI: Set topology failure on device %llx, hive %llx, ret %d",
			adev->gmc.xgmi.node_id,
			adev->gmc.xgmi.hive_id, ret);

	return ret;
}


/*
 * NOTE psp_xgmi_node_info.num_hops layout is as follows:
 * num_hops[7:6] = link type (0 = xGMI2, 1 = xGMI3, 2/3 = reserved)
 * num_hops[5:3] = reserved
 * num_hops[2:0] = number of hops
 */
int amdgpu_xgmi_get_hops_count(struct amdgpu_device *adev,
		struct amdgpu_device *peer_adev)
{
	struct psp_xgmi_topology_info *top = &adev->psp.xgmi_context.top_info;
	uint8_t num_hops_mask = 0x7;
	int i;

	for (i = 0 ; i < top->num_nodes; ++i)
		if (top->nodes[i].node_id == peer_adev->gmc.xgmi.node_id)
			return top->nodes[i].num_hops & num_hops_mask;
	return	-EINVAL;
}

int amdgpu_xgmi_get_num_links(struct amdgpu_device *adev,
		struct amdgpu_device *peer_adev)
{
	struct psp_xgmi_topology_info *top = &adev->psp.xgmi_context.top_info;
	int i;

	for (i = 0 ; i < top->num_nodes; ++i)
		if (top->nodes[i].node_id == peer_adev->gmc.xgmi.node_id)
			return top->nodes[i].num_links;
	return	-EINVAL;
}

/*
 * Devices that support extended data require the entire hive to initialize with
 * the shared memory buffer flag set.
 *
 * Hive locks and conditions apply - see amdgpu_xgmi_add_device
 */
static int amdgpu_xgmi_initialize_hive_get_data_partition(struct amdgpu_hive_info *hive,
							bool set_extended_data)
{
	struct amdgpu_device *tmp_adev;
	int ret;

	list_for_each_entry(tmp_adev, &hive->device_list, gmc.xgmi.head) {
		ret = psp_xgmi_initialize(&tmp_adev->psp, set_extended_data, false);
		if (ret) {
			dev_err(tmp_adev->dev,
				"XGMI: Failed to initialize xgmi session for data partition %i\n",
				set_extended_data);
			return ret;
		}

	}

	return 0;
}

static void amdgpu_xgmi_fill_topology_info(struct amdgpu_device *adev,
	struct amdgpu_device *peer_adev)
{
	struct psp_xgmi_topology_info *top_info = &adev->psp.xgmi_context.top_info;
	struct psp_xgmi_topology_info *peer_info = &peer_adev->psp.xgmi_context.top_info;

	for (int i = 0; i < peer_info->num_nodes; i++) {
		if (peer_info->nodes[i].node_id == adev->gmc.xgmi.node_id) {
			for (int j = 0; j < top_info->num_nodes; j++) {
				if (top_info->nodes[j].node_id == peer_adev->gmc.xgmi.node_id) {
					peer_info->nodes[i].num_hops = top_info->nodes[j].num_hops;
					peer_info->nodes[i].is_sharing_enabled =
							top_info->nodes[j].is_sharing_enabled;
					peer_info->nodes[i].num_links =
							top_info->nodes[j].num_links;
					return;
				}
			}
		}
	}
}

int amdgpu_xgmi_add_device(struct amdgpu_device *adev)
{
	struct psp_xgmi_topology_info *top_info;
	struct amdgpu_hive_info *hive;
	struct amdgpu_xgmi	*entry;
	struct amdgpu_device *tmp_adev = NULL;

	int count = 0, ret = 0;

	if (!adev->gmc.xgmi.supported)
		return 0;

	if (!adev->gmc.xgmi.pending_reset &&
	    amdgpu_device_ip_get_ip_block(adev, AMD_IP_BLOCK_TYPE_PSP)) {
		ret = psp_xgmi_initialize(&adev->psp, false, true);
		if (ret) {
			dev_err(adev->dev,
				"XGMI: Failed to initialize xgmi session\n");
			return ret;
		}

		ret = psp_xgmi_get_hive_id(&adev->psp, &adev->gmc.xgmi.hive_id);
		if (ret) {
			dev_err(adev->dev,
				"XGMI: Failed to get hive id\n");
			return ret;
		}

		ret = psp_xgmi_get_node_id(&adev->psp, &adev->gmc.xgmi.node_id);
		if (ret) {
			dev_err(adev->dev,
				"XGMI: Failed to get node id\n");
			return ret;
		}
	} else {
		adev->gmc.xgmi.hive_id = 16;
		adev->gmc.xgmi.node_id = adev->gmc.xgmi.physical_node_id + 16;
	}

	hive = amdgpu_get_xgmi_hive(adev);
	if (!hive) {
		ret = -EINVAL;
		dev_err(adev->dev,
			"XGMI: node 0x%llx, can not match hive 0x%llx in the hive list.\n",
			adev->gmc.xgmi.node_id, adev->gmc.xgmi.hive_id);
		goto exit;
	}
	mutex_lock(&hive->hive_lock);

	top_info = &adev->psp.xgmi_context.top_info;

	list_add_tail(&adev->gmc.xgmi.head, &hive->device_list);
	list_for_each_entry(entry, &hive->device_list, head)
		top_info->nodes[count++].node_id = entry->node_id;
	top_info->num_nodes = count;
	atomic_set(&hive->number_devices, count);

	task_barrier_add_task(&hive->tb);

	if (!adev->gmc.xgmi.pending_reset &&
	    amdgpu_device_ip_get_ip_block(adev, AMD_IP_BLOCK_TYPE_PSP)) {
		list_for_each_entry(tmp_adev, &hive->device_list, gmc.xgmi.head) {
			/* update node list for other device in the hive */
			if (tmp_adev != adev) {
				top_info = &tmp_adev->psp.xgmi_context.top_info;
				top_info->nodes[count - 1].node_id =
					adev->gmc.xgmi.node_id;
				top_info->num_nodes = count;
			}
			ret = amdgpu_xgmi_update_topology(hive, tmp_adev);
			if (ret)
				goto exit_unlock;
		}

		if (amdgpu_sriov_vf(adev) &&
			adev->psp.xgmi_context.xgmi_ta_caps & EXTEND_PEER_LINK_INFO_CMD_FLAG) {
			/* only get topology for VF being init if it can support full duplex */
			ret = psp_xgmi_get_topology_info(&adev->psp, count,
						&adev->psp.xgmi_context.top_info, false);
			if (ret) {
				dev_err(adev->dev,
					"XGMI: Get topology failure on device %llx, hive %llx, ret %d",
					adev->gmc.xgmi.node_id,
					adev->gmc.xgmi.hive_id, ret);
				/* To do: continue with some node failed or disable the whole hive*/
				goto exit_unlock;
			}

			/* fill the topology info for peers instead of getting from PSP */
			list_for_each_entry(tmp_adev, &hive->device_list, gmc.xgmi.head) {
				amdgpu_xgmi_fill_topology_info(adev, tmp_adev);
			}
		} else {
			/* get latest topology info for each device from psp */
			list_for_each_entry(tmp_adev, &hive->device_list, gmc.xgmi.head) {
				ret = psp_xgmi_get_topology_info(&tmp_adev->psp, count,
					&tmp_adev->psp.xgmi_context.top_info, false);
				if (ret) {
					dev_err(tmp_adev->dev,
						"XGMI: Get topology failure on device %llx, hive %llx, ret %d",
						tmp_adev->gmc.xgmi.node_id,
						tmp_adev->gmc.xgmi.hive_id, ret);
					/* To do : continue with some node failed or disable the whole hive */
					goto exit_unlock;
				}
			}
		}

		/* get topology again for hives that support extended data */
		if (adev->psp.xgmi_context.supports_extended_data) {

			/* initialize the hive to get extended data.  */
			ret = amdgpu_xgmi_initialize_hive_get_data_partition(hive, true);
			if (ret)
				goto exit_unlock;

			/* get the extended data. */
			list_for_each_entry(tmp_adev, &hive->device_list, gmc.xgmi.head) {
				ret = psp_xgmi_get_topology_info(&tmp_adev->psp, count,
						&tmp_adev->psp.xgmi_context.top_info, true);
				if (ret) {
					dev_err(tmp_adev->dev,
						"XGMI: Get topology for extended data failure on device %llx, hive %llx, ret %d",
						tmp_adev->gmc.xgmi.node_id,
						tmp_adev->gmc.xgmi.hive_id, ret);
					goto exit_unlock;
				}
			}

			/* initialize the hive to get non-extended data for the next round. */
			ret = amdgpu_xgmi_initialize_hive_get_data_partition(hive, false);
			if (ret)
				goto exit_unlock;

		}
	}

	if (!ret && !adev->gmc.xgmi.pending_reset)
		ret = amdgpu_xgmi_sysfs_add_dev_info(adev, hive);

exit_unlock:
	mutex_unlock(&hive->hive_lock);
exit:
	if (!ret) {
		adev->hive = hive;
		dev_info(adev->dev, "XGMI: Add node %d, hive 0x%llx.\n",
			 adev->gmc.xgmi.physical_node_id, adev->gmc.xgmi.hive_id);
	} else {
		amdgpu_put_xgmi_hive(hive);
		dev_err(adev->dev, "XGMI: Failed to add node %d, hive 0x%llx ret: %d\n",
			adev->gmc.xgmi.physical_node_id, adev->gmc.xgmi.hive_id,
			ret);
	}

	return ret;
}

int amdgpu_xgmi_remove_device(struct amdgpu_device *adev)
{
	struct amdgpu_hive_info *hive = adev->hive;

	if (!adev->gmc.xgmi.supported)
		return -EINVAL;

	if (!hive)
		return -EINVAL;

	mutex_lock(&hive->hive_lock);
	task_barrier_rem_task(&hive->tb);
	amdgpu_xgmi_sysfs_rem_dev_info(adev, hive);
	if (hive->hi_req_gpu == adev)
		hive->hi_req_gpu = NULL;
	list_del(&adev->gmc.xgmi.head);
	mutex_unlock(&hive->hive_lock);

	amdgpu_put_xgmi_hive(hive);
	adev->hive = NULL;

	if (atomic_dec_return(&hive->number_devices) == 0) {
		/* Remove the hive from global hive list */
		mutex_lock(&xgmi_mutex);
		list_del(&hive->node);
		mutex_unlock(&xgmi_mutex);

		amdgpu_put_xgmi_hive(hive);
	}

	return 0;
}

static int amdgpu_xgmi_ras_late_init(struct amdgpu_device *adev, struct ras_common_if *ras_block)
{
	if (!adev->gmc.xgmi.supported ||
	    adev->gmc.xgmi.num_physical_nodes == 0)
		return 0;

	amdgpu_ras_reset_error_count(adev, AMDGPU_RAS_BLOCK__XGMI_WAFL);

	return amdgpu_ras_block_late_init(adev, ras_block);
}

uint64_t amdgpu_xgmi_get_relative_phy_addr(struct amdgpu_device *adev,
					   uint64_t addr)
{
	struct amdgpu_xgmi *xgmi = &adev->gmc.xgmi;
	return (addr + xgmi->physical_node_id * xgmi->node_segment_size);
}

static void pcs_clear_status(struct amdgpu_device *adev, uint32_t pcs_status_reg)
{
	WREG32_PCIE(pcs_status_reg, 0xFFFFFFFF);
	WREG32_PCIE(pcs_status_reg, 0);
}

static void amdgpu_xgmi_legacy_reset_ras_error_count(struct amdgpu_device *adev)
{
	uint32_t i;

	switch (adev->asic_type) {
	case CHIP_ARCTURUS:
		for (i = 0; i < ARRAY_SIZE(xgmi_pcs_err_status_reg_arct); i++)
			pcs_clear_status(adev,
					 xgmi_pcs_err_status_reg_arct[i]);
		break;
	case CHIP_VEGA20:
		for (i = 0; i < ARRAY_SIZE(xgmi_pcs_err_status_reg_vg20); i++)
			pcs_clear_status(adev,
					 xgmi_pcs_err_status_reg_vg20[i]);
		break;
	case CHIP_ALDEBARAN:
		for (i = 0; i < ARRAY_SIZE(xgmi3x16_pcs_err_status_reg_aldebaran); i++)
			pcs_clear_status(adev,
					 xgmi3x16_pcs_err_status_reg_aldebaran[i]);
		for (i = 0; i < ARRAY_SIZE(walf_pcs_err_status_reg_aldebaran); i++)
			pcs_clear_status(adev,
					 walf_pcs_err_status_reg_aldebaran[i]);
		break;
	default:
		break;
	}

	switch (amdgpu_ip_version(adev, XGMI_HWIP, 0)) {
	case IP_VERSION(6, 4, 0):
		for (i = 0; i < ARRAY_SIZE(xgmi3x16_pcs_err_status_reg_v6_4); i++)
			pcs_clear_status(adev,
					xgmi3x16_pcs_err_status_reg_v6_4[i]);
		break;
	default:
		break;
	}
}

static void __xgmi_v6_4_0_reset_error_count(struct amdgpu_device *adev, int xgmi_inst, u64 mca_base)
{
	WREG64_MCA(xgmi_inst, mca_base, MCA_REG_IDX_STATUS, 0ULL);
}

static void xgmi_v6_4_0_reset_error_count(struct amdgpu_device *adev, int xgmi_inst)
{
	int i;

	for (i = 0; i < ARRAY_SIZE(xgmi_v6_4_0_mca_base_array); i++)
		__xgmi_v6_4_0_reset_error_count(adev, xgmi_inst, xgmi_v6_4_0_mca_base_array[i]);
}

static void xgmi_v6_4_0_reset_ras_error_count(struct amdgpu_device *adev)
{
	int i;

	for_each_inst(i, adev->aid_mask)
		xgmi_v6_4_0_reset_error_count(adev, i);
}

static void amdgpu_xgmi_reset_ras_error_count(struct amdgpu_device *adev)
{
	switch (amdgpu_ip_version(adev, XGMI_HWIP, 0)) {
	case IP_VERSION(6, 4, 0):
		xgmi_v6_4_0_reset_ras_error_count(adev);
		break;
	default:
		amdgpu_xgmi_legacy_reset_ras_error_count(adev);
		break;
	}
}

static int amdgpu_xgmi_query_pcs_error_status(struct amdgpu_device *adev,
					      uint32_t value,
						  uint32_t mask_value,
					      uint32_t *ue_count,
					      uint32_t *ce_count,
					      bool is_xgmi_pcs,
						  bool check_mask)
{
	int i;
	int ue_cnt = 0;
	const struct amdgpu_pcs_ras_field *pcs_ras_fields = NULL;
	uint32_t field_array_size = 0;

	if (is_xgmi_pcs) {
		if (amdgpu_ip_version(adev, XGMI_HWIP, 0) ==
		    IP_VERSION(6, 1, 0) ||
		    amdgpu_ip_version(adev, XGMI_HWIP, 0) ==
		    IP_VERSION(6, 4, 0)) {
			pcs_ras_fields = &xgmi3x16_pcs_ras_fields[0];
			field_array_size = ARRAY_SIZE(xgmi3x16_pcs_ras_fields);
		} else {
			pcs_ras_fields = &xgmi_pcs_ras_fields[0];
			field_array_size = ARRAY_SIZE(xgmi_pcs_ras_fields);
		}
	} else {
		pcs_ras_fields = &wafl_pcs_ras_fields[0];
		field_array_size = ARRAY_SIZE(wafl_pcs_ras_fields);
	}

	if (check_mask)
		value = value & ~mask_value;

	/* query xgmi/walf pcs error status,
	 * only ue is supported */
	for (i = 0; value && i < field_array_size; i++) {
		ue_cnt = (value &
				pcs_ras_fields[i].pcs_err_mask) >>
				pcs_ras_fields[i].pcs_err_shift;
		if (ue_cnt) {
			dev_info(adev->dev, "%s detected\n",
				 pcs_ras_fields[i].err_name);
			*ue_count += ue_cnt;
		}

		/* reset bit value if the bit is checked */
		value &= ~(pcs_ras_fields[i].pcs_err_mask);
	}

	return 0;
}

static void amdgpu_xgmi_legacy_query_ras_error_count(struct amdgpu_device *adev,
						     void *ras_error_status)
{
	struct ras_err_data *err_data = (struct ras_err_data *)ras_error_status;
	int i, supported = 1;
	uint32_t data, mask_data = 0;
	uint32_t ue_cnt = 0, ce_cnt = 0;

	if (!amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__XGMI_WAFL))
		return ;

	err_data->ue_count = 0;
	err_data->ce_count = 0;

	switch (adev->asic_type) {
	case CHIP_ARCTURUS:
		/* check xgmi pcs error */
		for (i = 0; i < ARRAY_SIZE(xgmi_pcs_err_status_reg_arct); i++) {
			data = RREG32_PCIE(xgmi_pcs_err_status_reg_arct[i]);
			if (data)
				amdgpu_xgmi_query_pcs_error_status(adev, data,
						mask_data, &ue_cnt, &ce_cnt, true, false);
		}
		/* check wafl pcs error */
		for (i = 0; i < ARRAY_SIZE(wafl_pcs_err_status_reg_arct); i++) {
			data = RREG32_PCIE(wafl_pcs_err_status_reg_arct[i]);
			if (data)
				amdgpu_xgmi_query_pcs_error_status(adev, data,
						mask_data, &ue_cnt, &ce_cnt, false, false);
		}
		break;
	case CHIP_VEGA20:
		/* check xgmi pcs error */
		for (i = 0; i < ARRAY_SIZE(xgmi_pcs_err_status_reg_vg20); i++) {
			data = RREG32_PCIE(xgmi_pcs_err_status_reg_vg20[i]);
			if (data)
				amdgpu_xgmi_query_pcs_error_status(adev, data,
						mask_data, &ue_cnt, &ce_cnt, true, false);
		}
		/* check wafl pcs error */
		for (i = 0; i < ARRAY_SIZE(wafl_pcs_err_status_reg_vg20); i++) {
			data = RREG32_PCIE(wafl_pcs_err_status_reg_vg20[i]);
			if (data)
				amdgpu_xgmi_query_pcs_error_status(adev, data,
						mask_data, &ue_cnt, &ce_cnt, false, false);
		}
		break;
	case CHIP_ALDEBARAN:
		/* check xgmi3x16 pcs error */
		for (i = 0; i < ARRAY_SIZE(xgmi3x16_pcs_err_status_reg_aldebaran); i++) {
			data = RREG32_PCIE(xgmi3x16_pcs_err_status_reg_aldebaran[i]);
			mask_data =
				RREG32_PCIE(xgmi3x16_pcs_err_noncorrectable_mask_reg_aldebaran[i]);
			if (data)
				amdgpu_xgmi_query_pcs_error_status(adev, data,
						mask_data, &ue_cnt, &ce_cnt, true, true);
		}
		/* check wafl pcs error */
		for (i = 0; i < ARRAY_SIZE(walf_pcs_err_status_reg_aldebaran); i++) {
			data = RREG32_PCIE(walf_pcs_err_status_reg_aldebaran[i]);
			mask_data =
				RREG32_PCIE(walf_pcs_err_noncorrectable_mask_reg_aldebaran[i]);
			if (data)
				amdgpu_xgmi_query_pcs_error_status(adev, data,
						mask_data, &ue_cnt, &ce_cnt, false, true);
		}
		break;
	default:
		supported = 0;
		break;
	}

	switch (amdgpu_ip_version(adev, XGMI_HWIP, 0)) {
	case IP_VERSION(6, 4, 0):
		/* check xgmi3x16 pcs error */
		for (i = 0; i < ARRAY_SIZE(xgmi3x16_pcs_err_status_reg_v6_4); i++) {
			data = RREG32_PCIE(xgmi3x16_pcs_err_status_reg_v6_4[i]);
			mask_data =
				RREG32_PCIE(xgmi3x16_pcs_err_noncorrectable_mask_reg_v6_4[i]);
			if (data)
				amdgpu_xgmi_query_pcs_error_status(adev, data,
						mask_data, &ue_cnt, &ce_cnt, true, true);
		}
		break;
	default:
		if (!supported)
			dev_warn(adev->dev, "XGMI RAS error query not supported");
		break;
	}

	amdgpu_ras_reset_error_count(adev, AMDGPU_RAS_BLOCK__XGMI_WAFL);

	err_data->ue_count += ue_cnt;
	err_data->ce_count += ce_cnt;
}

static enum amdgpu_mca_error_type xgmi_v6_4_0_pcs_mca_get_error_type(struct amdgpu_device *adev, u64 status)
{
	const char *error_str;
	int ext_error_code;

	ext_error_code = MCA_REG__STATUS__ERRORCODEEXT(status);

	error_str = ext_error_code < ARRAY_SIZE(xgmi_v6_4_0_ras_error_code_ext) ?
		xgmi_v6_4_0_ras_error_code_ext[ext_error_code] : NULL;
	if (error_str)
		dev_info(adev->dev, "%s detected\n", error_str);

	switch (ext_error_code) {
	case 0:
		return AMDGPU_MCA_ERROR_TYPE_UE;
	case 6:
		return AMDGPU_MCA_ERROR_TYPE_CE;
	default:
		return -EINVAL;
	}

	return -EINVAL;
}

static void __xgmi_v6_4_0_query_error_count(struct amdgpu_device *adev, struct amdgpu_smuio_mcm_config_info *mcm_info,
					    u64 mca_base, struct ras_err_data *err_data)
{
	int xgmi_inst = mcm_info->die_id;
	u64 status = 0;

	status = RREG64_MCA(xgmi_inst, mca_base, MCA_REG_IDX_STATUS);
	if (!MCA_REG__STATUS__VAL(status))
		return;

	switch (xgmi_v6_4_0_pcs_mca_get_error_type(adev, status)) {
	case AMDGPU_MCA_ERROR_TYPE_UE:
		amdgpu_ras_error_statistic_ue_count(err_data, mcm_info, NULL, 1ULL);
		break;
	case AMDGPU_MCA_ERROR_TYPE_CE:
		amdgpu_ras_error_statistic_ce_count(err_data, mcm_info, NULL, 1ULL);
		break;
	default:
		break;
	}

	WREG64_MCA(xgmi_inst, mca_base, MCA_REG_IDX_STATUS, 0ULL);
}

static void xgmi_v6_4_0_query_error_count(struct amdgpu_device *adev, int xgmi_inst, struct ras_err_data *err_data)
{
	struct amdgpu_smuio_mcm_config_info mcm_info = {
		.socket_id = adev->smuio.funcs->get_socket_id(adev),
		.die_id = xgmi_inst,
	};
	int i;

	for (i = 0; i < ARRAY_SIZE(xgmi_v6_4_0_mca_base_array); i++)
		__xgmi_v6_4_0_query_error_count(adev, &mcm_info, xgmi_v6_4_0_mca_base_array[i], err_data);
}

static void xgmi_v6_4_0_query_ras_error_count(struct amdgpu_device *adev, void *ras_error_status)
{
	struct ras_err_data *err_data = (struct ras_err_data *)ras_error_status;
	int i;

	for_each_inst(i, adev->aid_mask)
		xgmi_v6_4_0_query_error_count(adev, i, err_data);
}

static void amdgpu_xgmi_query_ras_error_count(struct amdgpu_device *adev,
					      void *ras_error_status)
{
	switch (amdgpu_ip_version(adev, XGMI_HWIP, 0)) {
	case IP_VERSION(6, 4, 0):
		xgmi_v6_4_0_query_ras_error_count(adev, ras_error_status);
		break;
	default:
		amdgpu_xgmi_legacy_query_ras_error_count(adev, ras_error_status);
		break;
	}
}

/* Trigger XGMI/WAFL error */
static int amdgpu_ras_error_inject_xgmi(struct amdgpu_device *adev,
			void *inject_if, uint32_t instance_mask)
{
	int ret1, ret2;
	struct ta_ras_trigger_error_input *block_info =
				(struct ta_ras_trigger_error_input *)inject_if;

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

	ret1 = amdgpu_dpm_set_xgmi_plpd_mode(adev, XGMI_PLPD_DISALLOW);
	if (ret1 && ret1 != -EOPNOTSUPP)
		dev_warn(adev->dev, "Failed to disallow XGMI power down");

	ret2 = psp_ras_trigger_error(&adev->psp, block_info, instance_mask);

	if (amdgpu_ras_intr_triggered())
		return ret2;

	ret1 = amdgpu_dpm_set_xgmi_plpd_mode(adev, XGMI_PLPD_DEFAULT);
	if (ret1 && ret1 != -EOPNOTSUPP)
		dev_warn(adev->dev, "Failed to allow XGMI power down");

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

	return ret2;
}

struct amdgpu_ras_block_hw_ops  xgmi_ras_hw_ops = {
	.query_ras_error_count = amdgpu_xgmi_query_ras_error_count,
	.reset_ras_error_count = amdgpu_xgmi_reset_ras_error_count,
	.ras_error_inject = amdgpu_ras_error_inject_xgmi,
};

struct amdgpu_xgmi_ras xgmi_ras = {
	.ras_block = {
		.hw_ops = &xgmi_ras_hw_ops,
		.ras_late_init = amdgpu_xgmi_ras_late_init,
	},
};

int amdgpu_xgmi_ras_sw_init(struct amdgpu_device *adev)
{
	int err;
	struct amdgpu_xgmi_ras *ras;

	if (!adev->gmc.xgmi.ras)
		return 0;

	ras = adev->gmc.xgmi.ras;
	err = amdgpu_ras_register_ras_block(adev, &ras->ras_block);
	if (err) {
		dev_err(adev->dev, "Failed to register xgmi_wafl_pcs ras block!\n");
		return err;
	}

	strcpy(ras->ras_block.ras_comm.name, "xgmi_wafl");
	ras->ras_block.ras_comm.block = AMDGPU_RAS_BLOCK__XGMI_WAFL;
	ras->ras_block.ras_comm.type = AMDGPU_RAS_ERROR__MULTI_UNCORRECTABLE;
	adev->gmc.xgmi.ras_if = &ras->ras_block.ras_comm;

	return 0;
}