summaryrefslogtreecommitdiffstats
path: root/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
blob: 73f3b79ab1319f593b2b2baee715dea9b52ebfd4 (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
/*
 * Copyright 2016 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/firmware.h>
#include <linux/pci.h>

#include <drm/drm_cache.h>

#include "amdgpu.h"
#include "gmc_v9_0.h"
#include "amdgpu_atomfirmware.h"
#include "amdgpu_gem.h"

#include "hdp/hdp_4_0_offset.h"
#include "hdp/hdp_4_0_sh_mask.h"
#include "gc/gc_9_0_sh_mask.h"
#include "dce/dce_12_0_offset.h"
#include "dce/dce_12_0_sh_mask.h"
#include "vega10_enum.h"
#include "mmhub/mmhub_1_0_offset.h"
#include "athub/athub_1_0_offset.h"
#include "oss/osssys_4_0_offset.h"

#include "soc15.h"
#include "soc15_common.h"
#include "umc/umc_6_0_sh_mask.h"

#include "gfxhub_v1_0.h"
#include "mmhub_v1_0.h"
#include "gfxhub_v1_1.h"

#include "ivsrcid/vmc/irqsrcs_vmc_1_0.h"

#include "amdgpu_ras.h"

/* add these here since we already include dce12 headers and these are for DCN */
#define mmHUBP0_DCSURF_PRI_VIEWPORT_DIMENSION                                                          0x055d
#define mmHUBP0_DCSURF_PRI_VIEWPORT_DIMENSION_BASE_IDX                                                 2
#define HUBP0_DCSURF_PRI_VIEWPORT_DIMENSION__PRI_VIEWPORT_WIDTH__SHIFT                                        0x0
#define HUBP0_DCSURF_PRI_VIEWPORT_DIMENSION__PRI_VIEWPORT_HEIGHT__SHIFT                                       0x10
#define HUBP0_DCSURF_PRI_VIEWPORT_DIMENSION__PRI_VIEWPORT_WIDTH_MASK                                          0x00003FFFL
#define HUBP0_DCSURF_PRI_VIEWPORT_DIMENSION__PRI_VIEWPORT_HEIGHT_MASK                                         0x3FFF0000L

/* XXX Move this macro to VEGA10 header file, which is like vid.h for VI.*/
#define AMDGPU_NUM_OF_VMIDS			8

static const u32 golden_settings_vega10_hdp[] =
{
	0xf64, 0x0fffffff, 0x00000000,
	0xf65, 0x0fffffff, 0x00000000,
	0xf66, 0x0fffffff, 0x00000000,
	0xf67, 0x0fffffff, 0x00000000,
	0xf68, 0x0fffffff, 0x00000000,
	0xf6a, 0x0fffffff, 0x00000000,
	0xf6b, 0x0fffffff, 0x00000000,
	0xf6c, 0x0fffffff, 0x00000000,
	0xf6d, 0x0fffffff, 0x00000000,
	0xf6e, 0x0fffffff, 0x00000000,
};

static const struct soc15_reg_golden golden_settings_mmhub_1_0_0[] =
{
	SOC15_REG_GOLDEN_VALUE(MMHUB, 0, mmDAGB1_WRCLI2, 0x00000007, 0xfe5fe0fa),
	SOC15_REG_GOLDEN_VALUE(MMHUB, 0, mmMMEA1_DRAM_WR_CLI2GRP_MAP0, 0x00000030, 0x55555565)
};

static const struct soc15_reg_golden golden_settings_athub_1_0_0[] =
{
	SOC15_REG_GOLDEN_VALUE(ATHUB, 0, mmRPB_ARB_CNTL, 0x0000ff00, 0x00000800),
	SOC15_REG_GOLDEN_VALUE(ATHUB, 0, mmRPB_ARB_CNTL2, 0x00ff00ff, 0x00080008)
};

static const uint32_t ecc_umc_mcumc_ctrl_addrs[] = {
	(0x000143c0 + 0x00000000),
	(0x000143c0 + 0x00000800),
	(0x000143c0 + 0x00001000),
	(0x000143c0 + 0x00001800),
	(0x000543c0 + 0x00000000),
	(0x000543c0 + 0x00000800),
	(0x000543c0 + 0x00001000),
	(0x000543c0 + 0x00001800),
	(0x000943c0 + 0x00000000),
	(0x000943c0 + 0x00000800),
	(0x000943c0 + 0x00001000),
	(0x000943c0 + 0x00001800),
	(0x000d43c0 + 0x00000000),
	(0x000d43c0 + 0x00000800),
	(0x000d43c0 + 0x00001000),
	(0x000d43c0 + 0x00001800),
	(0x001143c0 + 0x00000000),
	(0x001143c0 + 0x00000800),
	(0x001143c0 + 0x00001000),
	(0x001143c0 + 0x00001800),
	(0x001543c0 + 0x00000000),
	(0x001543c0 + 0x00000800),
	(0x001543c0 + 0x00001000),
	(0x001543c0 + 0x00001800),
	(0x001943c0 + 0x00000000),
	(0x001943c0 + 0x00000800),
	(0x001943c0 + 0x00001000),
	(0x001943c0 + 0x00001800),
	(0x001d43c0 + 0x00000000),
	(0x001d43c0 + 0x00000800),
	(0x001d43c0 + 0x00001000),
	(0x001d43c0 + 0x00001800),
};

static const uint32_t ecc_umc_mcumc_ctrl_mask_addrs[] = {
	(0x000143e0 + 0x00000000),
	(0x000143e0 + 0x00000800),
	(0x000143e0 + 0x00001000),
	(0x000143e0 + 0x00001800),
	(0x000543e0 + 0x00000000),
	(0x000543e0 + 0x00000800),
	(0x000543e0 + 0x00001000),
	(0x000543e0 + 0x00001800),
	(0x000943e0 + 0x00000000),
	(0x000943e0 + 0x00000800),
	(0x000943e0 + 0x00001000),
	(0x000943e0 + 0x00001800),
	(0x000d43e0 + 0x00000000),
	(0x000d43e0 + 0x00000800),
	(0x000d43e0 + 0x00001000),
	(0x000d43e0 + 0x00001800),
	(0x001143e0 + 0x00000000),
	(0x001143e0 + 0x00000800),
	(0x001143e0 + 0x00001000),
	(0x001143e0 + 0x00001800),
	(0x001543e0 + 0x00000000),
	(0x001543e0 + 0x00000800),
	(0x001543e0 + 0x00001000),
	(0x001543e0 + 0x00001800),
	(0x001943e0 + 0x00000000),
	(0x001943e0 + 0x00000800),
	(0x001943e0 + 0x00001000),
	(0x001943e0 + 0x00001800),
	(0x001d43e0 + 0x00000000),
	(0x001d43e0 + 0x00000800),
	(0x001d43e0 + 0x00001000),
	(0x001d43e0 + 0x00001800),
};

static const uint32_t ecc_umc_mcumc_status_addrs[] = {
	(0x000143c2 + 0x00000000),
	(0x000143c2 + 0x00000800),
	(0x000143c2 + 0x00001000),
	(0x000143c2 + 0x00001800),
	(0x000543c2 + 0x00000000),
	(0x000543c2 + 0x00000800),
	(0x000543c2 + 0x00001000),
	(0x000543c2 + 0x00001800),
	(0x000943c2 + 0x00000000),
	(0x000943c2 + 0x00000800),
	(0x000943c2 + 0x00001000),
	(0x000943c2 + 0x00001800),
	(0x000d43c2 + 0x00000000),
	(0x000d43c2 + 0x00000800),
	(0x000d43c2 + 0x00001000),
	(0x000d43c2 + 0x00001800),
	(0x001143c2 + 0x00000000),
	(0x001143c2 + 0x00000800),
	(0x001143c2 + 0x00001000),
	(0x001143c2 + 0x00001800),
	(0x001543c2 + 0x00000000),
	(0x001543c2 + 0x00000800),
	(0x001543c2 + 0x00001000),
	(0x001543c2 + 0x00001800),
	(0x001943c2 + 0x00000000),
	(0x001943c2 + 0x00000800),
	(0x001943c2 + 0x00001000),
	(0x001943c2 + 0x00001800),
	(0x001d43c2 + 0x00000000),
	(0x001d43c2 + 0x00000800),
	(0x001d43c2 + 0x00001000),
	(0x001d43c2 + 0x00001800),
};

static int gmc_v9_0_ecc_interrupt_state(struct amdgpu_device *adev,
		struct amdgpu_irq_src *src,
		unsigned type,
		enum amdgpu_interrupt_state state)
{
	u32 bits, i, tmp, reg;

	bits = 0x7f;

	switch (state) {
	case AMDGPU_IRQ_STATE_DISABLE:
		for (i = 0; i < ARRAY_SIZE(ecc_umc_mcumc_ctrl_addrs); i++) {
			reg = ecc_umc_mcumc_ctrl_addrs[i];
			tmp = RREG32(reg);
			tmp &= ~bits;
			WREG32(reg, tmp);
		}
		for (i = 0; i < ARRAY_SIZE(ecc_umc_mcumc_ctrl_mask_addrs); i++) {
			reg = ecc_umc_mcumc_ctrl_mask_addrs[i];
			tmp = RREG32(reg);
			tmp &= ~bits;
			WREG32(reg, tmp);
		}
		break;
	case AMDGPU_IRQ_STATE_ENABLE:
		for (i = 0; i < ARRAY_SIZE(ecc_umc_mcumc_ctrl_addrs); i++) {
			reg = ecc_umc_mcumc_ctrl_addrs[i];
			tmp = RREG32(reg);
			tmp |= bits;
			WREG32(reg, tmp);
		}
		for (i = 0; i < ARRAY_SIZE(ecc_umc_mcumc_ctrl_mask_addrs); i++) {
			reg = ecc_umc_mcumc_ctrl_mask_addrs[i];
			tmp = RREG32(reg);
			tmp |= bits;
			WREG32(reg, tmp);
		}
		break;
	default:
		break;
	}

	return 0;
}

static int gmc_v9_0_process_ras_data_cb(struct amdgpu_device *adev,
		struct amdgpu_iv_entry *entry)
{
	kgd2kfd_set_sram_ecc_flag(adev->kfd.dev);
	amdgpu_ras_reset_gpu(adev, 0);
	return AMDGPU_RAS_UE;
}

static int gmc_v9_0_process_ecc_irq(struct amdgpu_device *adev,
		struct amdgpu_irq_src *source,
		struct amdgpu_iv_entry *entry)
{
	struct ras_common_if *ras_if = adev->gmc.ras_if;
	struct ras_dispatch_if ih_data = {
		.entry = entry,
	};

	if (!ras_if)
		return 0;

	ih_data.head = *ras_if;

	amdgpu_ras_interrupt_dispatch(adev, &ih_data);
	return 0;
}

static int gmc_v9_0_vm_fault_interrupt_state(struct amdgpu_device *adev,
					struct amdgpu_irq_src *src,
					unsigned type,
					enum amdgpu_interrupt_state state)
{
	struct amdgpu_vmhub *hub;
	u32 tmp, reg, bits, i, j;

	bits = VM_CONTEXT1_CNTL__RANGE_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
		VM_CONTEXT1_CNTL__DUMMY_PAGE_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
		VM_CONTEXT1_CNTL__PDE0_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
		VM_CONTEXT1_CNTL__VALID_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
		VM_CONTEXT1_CNTL__READ_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
		VM_CONTEXT1_CNTL__WRITE_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
		VM_CONTEXT1_CNTL__EXECUTE_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK;

	switch (state) {
	case AMDGPU_IRQ_STATE_DISABLE:
		for (j = 0; j < AMDGPU_MAX_VMHUBS; j++) {
			hub = &adev->vmhub[j];
			for (i = 0; i < 16; i++) {
				reg = hub->vm_context0_cntl + i;
				tmp = RREG32(reg);
				tmp &= ~bits;
				WREG32(reg, tmp);
			}
		}
		break;
	case AMDGPU_IRQ_STATE_ENABLE:
		for (j = 0; j < AMDGPU_MAX_VMHUBS; j++) {
			hub = &adev->vmhub[j];
			for (i = 0; i < 16; i++) {
				reg = hub->vm_context0_cntl + i;
				tmp = RREG32(reg);
				tmp |= bits;
				WREG32(reg, tmp);
			}
		}
	default:
		break;
	}

	return 0;
}

static int gmc_v9_0_process_interrupt(struct amdgpu_device *adev,
				struct amdgpu_irq_src *source,
				struct amdgpu_iv_entry *entry)
{
	struct amdgpu_vmhub *hub = &adev->vmhub[entry->vmid_src];
	bool retry_fault = !!(entry->src_data[1] & 0x80);
	uint32_t status = 0;
	u64 addr;

	addr = (u64)entry->src_data[0] << 12;
	addr |= ((u64)entry->src_data[1] & 0xf) << 44;

	if (retry_fault && amdgpu_gmc_filter_faults(adev, addr, entry->pasid,
						    entry->timestamp))
		return 1; /* This also prevents sending it to KFD */

	/* If it's the first fault for this address, process it normally */
	if (!amdgpu_sriov_vf(adev)) {
		status = RREG32(hub->vm_l2_pro_fault_status);
		WREG32_P(hub->vm_l2_pro_fault_cntl, 1, ~1);
	}

	if (printk_ratelimit()) {
		struct amdgpu_task_info task_info;

		memset(&task_info, 0, sizeof(struct amdgpu_task_info));
		amdgpu_vm_get_task_info(adev, entry->pasid, &task_info);

		dev_err(adev->dev,
			"[%s] %s page fault (src_id:%u ring:%u vmid:%u "
			"pasid:%u, for process %s pid %d thread %s pid %d)\n",
			entry->vmid_src ? "mmhub" : "gfxhub",
			retry_fault ? "retry" : "no-retry",
			entry->src_id, entry->ring_id, entry->vmid,
			entry->pasid, task_info.process_name, task_info.tgid,
			task_info.task_name, task_info.pid);
		dev_err(adev->dev, "  in page starting at address 0x%016llx from %d\n",
			addr, entry->client_id);
		if (!amdgpu_sriov_vf(adev))
			dev_err(adev->dev,
				"VM_L2_PROTECTION_FAULT_STATUS:0x%08X\n",
				status);
	}

	return 0;
}

static const struct amdgpu_irq_src_funcs gmc_v9_0_irq_funcs = {
	.set = gmc_v9_0_vm_fault_interrupt_state,
	.process = gmc_v9_0_process_interrupt,
};


static const struct amdgpu_irq_src_funcs gmc_v9_0_ecc_funcs = {
	.set = gmc_v9_0_ecc_interrupt_state,
	.process = gmc_v9_0_process_ecc_irq,
};

static void gmc_v9_0_set_irq_funcs(struct amdgpu_device *adev)
{
	adev->gmc.vm_fault.num_types = 1;
	adev->gmc.vm_fault.funcs = &gmc_v9_0_irq_funcs;

	adev->gmc.ecc_irq.num_types = 1;
	adev->gmc.ecc_irq.funcs = &gmc_v9_0_ecc_funcs;
}

static uint32_t gmc_v9_0_get_invalidate_req(unsigned int vmid,
					uint32_t flush_type)
{
	u32 req = 0;

	req = REG_SET_FIELD(req, VM_INVALIDATE_ENG0_REQ,
			    PER_VMID_INVALIDATE_REQ, 1 << vmid);
	req = REG_SET_FIELD(req, VM_INVALIDATE_ENG0_REQ, FLUSH_TYPE, flush_type);
	req = REG_SET_FIELD(req, VM_INVALIDATE_ENG0_REQ, INVALIDATE_L2_PTES, 1);
	req = REG_SET_FIELD(req, VM_INVALIDATE_ENG0_REQ, INVALIDATE_L2_PDE0, 1);
	req = REG_SET_FIELD(req, VM_INVALIDATE_ENG0_REQ, INVALIDATE_L2_PDE1, 1);
	req = REG_SET_FIELD(req, VM_INVALIDATE_ENG0_REQ, INVALIDATE_L2_PDE2, 1);
	req = REG_SET_FIELD(req, VM_INVALIDATE_ENG0_REQ, INVALIDATE_L1_PTES, 1);
	req = REG_SET_FIELD(req, VM_INVALIDATE_ENG0_REQ,
			    CLEAR_PROTECTION_FAULT_STATUS_ADDR,	0);

	return req;
}

/*
 * GART
 * VMID 0 is the physical GPU addresses as used by the kernel.
 * VMIDs 1-15 are used for userspace clients and are handled
 * by the amdgpu vm/hsa code.
 */

/**
 * gmc_v9_0_flush_gpu_tlb - tlb flush with certain type
 *
 * @adev: amdgpu_device pointer
 * @vmid: vm instance to flush
 * @flush_type: the flush type
 *
 * Flush the TLB for the requested page table using certain type.
 */
static void gmc_v9_0_flush_gpu_tlb(struct amdgpu_device *adev,
				uint32_t vmid, uint32_t flush_type)
{
	const unsigned eng = 17;
	unsigned i, j;

	for (i = 0; i < AMDGPU_MAX_VMHUBS; ++i) {
		struct amdgpu_vmhub *hub = &adev->vmhub[i];
		u32 tmp = gmc_v9_0_get_invalidate_req(vmid, flush_type);

		/* This is necessary for a HW workaround under SRIOV as well
		 * as GFXOFF under bare metal
		 */
		if (adev->gfx.kiq.ring.sched.ready &&
		    (amdgpu_sriov_runtime(adev) || !amdgpu_sriov_vf(adev)) &&
		    !adev->in_gpu_reset) {
			uint32_t req = hub->vm_inv_eng0_req + eng;
			uint32_t ack = hub->vm_inv_eng0_ack + eng;

			amdgpu_virt_kiq_reg_write_reg_wait(adev, req, ack, tmp,
							   1 << vmid);
			continue;
		}

		spin_lock(&adev->gmc.invalidate_lock);
		WREG32_NO_KIQ(hub->vm_inv_eng0_req + eng, tmp);
		for (j = 0; j < adev->usec_timeout; j++) {
			tmp = RREG32_NO_KIQ(hub->vm_inv_eng0_ack + eng);
			if (tmp & (1 << vmid))
				break;
			udelay(1);
		}
		spin_unlock(&adev->gmc.invalidate_lock);
		if (j < adev->usec_timeout)
			continue;

		DRM_ERROR("Timeout waiting for VM flush ACK!\n");
	}
}

static uint64_t gmc_v9_0_emit_flush_gpu_tlb(struct amdgpu_ring *ring,
					    unsigned vmid, uint64_t pd_addr)
{
	struct amdgpu_device *adev = ring->adev;
	struct amdgpu_vmhub *hub = &adev->vmhub[ring->funcs->vmhub];
	uint32_t req = gmc_v9_0_get_invalidate_req(vmid, 0);
	unsigned eng = ring->vm_inv_eng;

	amdgpu_ring_emit_wreg(ring, hub->ctx0_ptb_addr_lo32 + (2 * vmid),
			      lower_32_bits(pd_addr));

	amdgpu_ring_emit_wreg(ring, hub->ctx0_ptb_addr_hi32 + (2 * vmid),
			      upper_32_bits(pd_addr));

	amdgpu_ring_emit_reg_write_reg_wait(ring, hub->vm_inv_eng0_req + eng,
					    hub->vm_inv_eng0_ack + eng,
					    req, 1 << vmid);

	return pd_addr;
}

static void gmc_v9_0_emit_pasid_mapping(struct amdgpu_ring *ring, unsigned vmid,
					unsigned pasid)
{
	struct amdgpu_device *adev = ring->adev;
	uint32_t reg;

	if (ring->funcs->vmhub == AMDGPU_GFXHUB)
		reg = SOC15_REG_OFFSET(OSSSYS, 0, mmIH_VMID_0_LUT) + vmid;
	else
		reg = SOC15_REG_OFFSET(OSSSYS, 0, mmIH_VMID_0_LUT_MM) + vmid;

	amdgpu_ring_emit_wreg(ring, reg, pasid);
}

/*
 * PTE format on VEGA 10:
 * 63:59 reserved
 * 58:57 mtype
 * 56 F
 * 55 L
 * 54 P
 * 53 SW
 * 52 T
 * 50:48 reserved
 * 47:12 4k physical page base address
 * 11:7 fragment
 * 6 write
 * 5 read
 * 4 exe
 * 3 Z
 * 2 snooped
 * 1 system
 * 0 valid
 *
 * PDE format on VEGA 10:
 * 63:59 block fragment size
 * 58:55 reserved
 * 54 P
 * 53:48 reserved
 * 47:6 physical base address of PD or PTE
 * 5:3 reserved
 * 2 C
 * 1 system
 * 0 valid
 */

static uint64_t gmc_v9_0_get_vm_pte_flags(struct amdgpu_device *adev,
						uint32_t flags)

{
	uint64_t pte_flag = 0;

	if (flags & AMDGPU_VM_PAGE_EXECUTABLE)
		pte_flag |= AMDGPU_PTE_EXECUTABLE;
	if (flags & AMDGPU_VM_PAGE_READABLE)
		pte_flag |= AMDGPU_PTE_READABLE;
	if (flags & AMDGPU_VM_PAGE_WRITEABLE)
		pte_flag |= AMDGPU_PTE_WRITEABLE;

	switch (flags & AMDGPU_VM_MTYPE_MASK) {
	case AMDGPU_VM_MTYPE_DEFAULT:
		pte_flag |= AMDGPU_PTE_MTYPE_VG10(MTYPE_NC);
		break;
	case AMDGPU_VM_MTYPE_NC:
		pte_flag |= AMDGPU_PTE_MTYPE_VG10(MTYPE_NC);
		break;
	case AMDGPU_VM_MTYPE_WC:
		pte_flag |= AMDGPU_PTE_MTYPE_VG10(MTYPE_WC);
		break;
	case AMDGPU_VM_MTYPE_CC:
		pte_flag |= AMDGPU_PTE_MTYPE_VG10(MTYPE_CC);
		break;
	case AMDGPU_VM_MTYPE_UC:
		pte_flag |= AMDGPU_PTE_MTYPE_VG10(MTYPE_UC);
		break;
	default:
		pte_flag |= AMDGPU_PTE_MTYPE_VG10(MTYPE_NC);
		break;
	}

	if (flags & AMDGPU_VM_PAGE_PRT)
		pte_flag |= AMDGPU_PTE_PRT;

	return pte_flag;
}

static void gmc_v9_0_get_vm_pde(struct amdgpu_device *adev, int level,
				uint64_t *addr, uint64_t *flags)
{
	if (!(*flags & AMDGPU_PDE_PTE) && !(*flags & AMDGPU_PTE_SYSTEM))
		*addr = adev->vm_manager.vram_base_offset + *addr -
			adev->gmc.vram_start;
	BUG_ON(*addr & 0xFFFF00000000003FULL);

	if (!adev->gmc.translate_further)
		return;

	if (level == AMDGPU_VM_PDB1) {
		/* Set the block fragment size */
		if (!(*flags & AMDGPU_PDE_PTE))
			*flags |= AMDGPU_PDE_BFS(0x9);

	} else if (level == AMDGPU_VM_PDB0) {
		if (*flags & AMDGPU_PDE_PTE)
			*flags &= ~AMDGPU_PDE_PTE;
		else
			*flags |= AMDGPU_PTE_TF;
	}
}

static const struct amdgpu_gmc_funcs gmc_v9_0_gmc_funcs = {
	.flush_gpu_tlb = gmc_v9_0_flush_gpu_tlb,
	.emit_flush_gpu_tlb = gmc_v9_0_emit_flush_gpu_tlb,
	.emit_pasid_mapping = gmc_v9_0_emit_pasid_mapping,
	.get_vm_pte_flags = gmc_v9_0_get_vm_pte_flags,
	.get_vm_pde = gmc_v9_0_get_vm_pde
};

static void gmc_v9_0_set_gmc_funcs(struct amdgpu_device *adev)
{
	adev->gmc.gmc_funcs = &gmc_v9_0_gmc_funcs;
}

static int gmc_v9_0_early_init(void *handle)
{
	struct amdgpu_device *adev = (struct amdgpu_device *)handle;

	gmc_v9_0_set_gmc_funcs(adev);
	gmc_v9_0_set_irq_funcs(adev);

	adev->gmc.shared_aperture_start = 0x2000000000000000ULL;
	adev->gmc.shared_aperture_end =
		adev->gmc.shared_aperture_start + (4ULL << 30) - 1;
	adev->gmc.private_aperture_start = 0x1000000000000000ULL;
	adev->gmc.private_aperture_end =
		adev->gmc.private_aperture_start + (4ULL << 30) - 1;

	return 0;
}

static bool gmc_v9_0_keep_stolen_memory(struct amdgpu_device *adev)
{

	/*
	 * TODO:
	 * Currently there is a bug where some memory client outside
	 * of the driver writes to first 8M of VRAM on S3 resume,
	 * this overrides GART which by default gets placed in first 8M and
	 * causes VM_FAULTS once GTT is accessed.
	 * Keep the stolen memory reservation until the while this is not solved.
	 * Also check code in gmc_v9_0_get_vbios_fb_size and gmc_v9_0_late_init
	 */
	switch (adev->asic_type) {
	case CHIP_VEGA10:
	case CHIP_RAVEN:
		return true;
	case CHIP_VEGA12:
	case CHIP_VEGA20:
	default:
		return false;
	}
}

static int gmc_v9_0_allocate_vm_inv_eng(struct amdgpu_device *adev)
{
	struct amdgpu_ring *ring;
	unsigned vm_inv_engs[AMDGPU_MAX_VMHUBS] =
		{GFXHUB_FREE_VM_INV_ENGS_BITMAP, MMHUB_FREE_VM_INV_ENGS_BITMAP};
	unsigned i;
	unsigned vmhub, inv_eng;

	for (i = 0; i < adev->num_rings; ++i) {
		ring = adev->rings[i];
		vmhub = ring->funcs->vmhub;

		inv_eng = ffs(vm_inv_engs[vmhub]);
		if (!inv_eng) {
			dev_err(adev->dev, "no VM inv eng for ring %s\n",
				ring->name);
			return -EINVAL;
		}

		ring->vm_inv_eng = inv_eng - 1;
		vm_inv_engs[vmhub] &= ~(1 << ring->vm_inv_eng);

		dev_info(adev->dev, "ring %s uses VM inv eng %u on hub %u\n",
			 ring->name, ring->vm_inv_eng, ring->funcs->vmhub);
	}

	return 0;
}

static int gmc_v9_0_ecc_late_init(void *handle)
{
	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
	struct ras_common_if **ras_if = &adev->gmc.ras_if;
	struct ras_ih_if ih_info = {
		.cb = gmc_v9_0_process_ras_data_cb,
	};
	struct ras_fs_if fs_info = {
		.sysfs_name = "umc_err_count",
		.debugfs_name = "umc_err_inject",
	};
	struct ras_common_if ras_block = {
		.block = AMDGPU_RAS_BLOCK__UMC,
		.type = AMDGPU_RAS_ERROR__MULTI_UNCORRECTABLE,
		.sub_block_index = 0,
		.name = "umc",
	};
	int r;

	if (!amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__UMC)) {
		amdgpu_ras_feature_enable_on_boot(adev, &ras_block, 0);
		return 0;
	}
	/* handle resume path. */
	if (*ras_if) {
		/* resend ras TA enable cmd during resume.
		 * prepare to handle failure.
		 */
		ih_info.head = **ras_if;
		r = amdgpu_ras_feature_enable_on_boot(adev, *ras_if, 1);
		if (r) {
			if (r == -EAGAIN) {
				/* request a gpu reset. will run again. */
				amdgpu_ras_request_reset_on_boot(adev,
						AMDGPU_RAS_BLOCK__UMC);
				return 0;
			}
			/* fail to enable ras, cleanup all. */
			goto irq;
		}
		/* enable successfully. continue. */
		goto resume;
	}

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

	**ras_if = ras_block;

	r = amdgpu_ras_feature_enable_on_boot(adev, *ras_if, 1);
	if (r) {
		if (r == -EAGAIN) {
			amdgpu_ras_request_reset_on_boot(adev,
					AMDGPU_RAS_BLOCK__UMC);
			r = 0;
		}
		goto feature;
	}

	ih_info.head = **ras_if;
	fs_info.head = **ras_if;

	r = amdgpu_ras_interrupt_add_handler(adev, &ih_info);
	if (r)
		goto interrupt;

	amdgpu_ras_debugfs_create(adev, &fs_info);

	r = amdgpu_ras_sysfs_create(adev, &fs_info);
	if (r)
		goto sysfs;
resume:
	r = amdgpu_irq_get(adev, &adev->gmc.ecc_irq, 0);
	if (r)
		goto irq;

	return 0;
irq:
	amdgpu_ras_sysfs_remove(adev, *ras_if);
sysfs:
	amdgpu_ras_debugfs_remove(adev, *ras_if);
	amdgpu_ras_interrupt_remove_handler(adev, &ih_info);
interrupt:
	amdgpu_ras_feature_enable(adev, *ras_if, 0);
feature:
	kfree(*ras_if);
	*ras_if = NULL;
	return r;
}


static int gmc_v9_0_late_init(void *handle)
{
	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
	bool r;

	if (!gmc_v9_0_keep_stolen_memory(adev))
		amdgpu_bo_late_init(adev);

	r = gmc_v9_0_allocate_vm_inv_eng(adev);
	if (r)
		return r;
	/* Check if ecc is available */
	if (!amdgpu_sriov_vf(adev)) {
		switch (adev->asic_type) {
		case CHIP_VEGA10:
		case CHIP_VEGA20:
			r = amdgpu_atomfirmware_mem_ecc_supported(adev);
			if (!r) {
				DRM_INFO("ECC is not present.\n");
				if (adev->df_funcs->enable_ecc_force_par_wr_rmw)
					adev->df_funcs->enable_ecc_force_par_wr_rmw(adev, false);
			} else {
				DRM_INFO("ECC is active.\n");
			}

			r = amdgpu_atomfirmware_sram_ecc_supported(adev);
			if (!r) {
				DRM_INFO("SRAM ECC is not present.\n");
			} else {
				DRM_INFO("SRAM ECC is active.\n");
			}
			break;
		default:
			break;
		}
	}

	r = gmc_v9_0_ecc_late_init(handle);
	if (r)
		return r;

	return amdgpu_irq_get(adev, &adev->gmc.vm_fault, 0);
}

static void gmc_v9_0_vram_gtt_location(struct amdgpu_device *adev,
					struct amdgpu_gmc *mc)
{
	u64 base = 0;
	if (!amdgpu_sriov_vf(adev))
		base = mmhub_v1_0_get_fb_location(adev);
	/* add the xgmi offset of the physical node */
	base += adev->gmc.xgmi.physical_node_id * adev->gmc.xgmi.node_segment_size;
	amdgpu_gmc_vram_location(adev, mc, base);
	amdgpu_gmc_gart_location(adev, mc);
	if (!amdgpu_sriov_vf(adev))
		amdgpu_gmc_agp_location(adev, mc);
	/* base offset of vram pages */
	adev->vm_manager.vram_base_offset = gfxhub_v1_0_get_mc_fb_offset(adev);

	/* XXX: add the xgmi offset of the physical node? */
	adev->vm_manager.vram_base_offset +=
		adev->gmc.xgmi.physical_node_id * adev->gmc.xgmi.node_segment_size;
}

/**
 * gmc_v9_0_mc_init - initialize the memory controller driver params
 *
 * @adev: amdgpu_device pointer
 *
 * Look up the amount of vram, vram width, and decide how to place
 * vram and gart within the GPU's physical address space.
 * Returns 0 for success.
 */
static int gmc_v9_0_mc_init(struct amdgpu_device *adev)
{
	int chansize, numchan;
	int r;

	if (amdgpu_sriov_vf(adev)) {
		/* For Vega10 SR-IOV, vram_width can't be read from ATOM as RAVEN,
		 * and DF related registers is not readable, seems hardcord is the
		 * only way to set the correct vram_width
		 */
		adev->gmc.vram_width = 2048;
	} else if (amdgpu_emu_mode != 1) {
		adev->gmc.vram_width = amdgpu_atomfirmware_get_vram_width(adev);
	}

	if (!adev->gmc.vram_width) {
		/* hbm memory channel size */
		if (adev->flags & AMD_IS_APU)
			chansize = 64;
		else
			chansize = 128;

		numchan = adev->df_funcs->get_hbm_channel_number(adev);
		adev->gmc.vram_width = numchan * chansize;
	}

	/* size in MB on si */
	adev->gmc.mc_vram_size =
		adev->nbio_funcs->get_memsize(adev) * 1024ULL * 1024ULL;
	adev->gmc.real_vram_size = adev->gmc.mc_vram_size;

	if (!(adev->flags & AMD_IS_APU)) {
		r = amdgpu_device_resize_fb_bar(adev);
		if (r)
			return r;
	}
	adev->gmc.aper_base = pci_resource_start(adev->pdev, 0);
	adev->gmc.aper_size = pci_resource_len(adev->pdev, 0);

#ifdef CONFIG_X86_64
	if (adev->flags & AMD_IS_APU) {
		adev->gmc.aper_base = gfxhub_v1_0_get_mc_fb_offset(adev);
		adev->gmc.aper_size = adev->gmc.real_vram_size;
	}
#endif
	/* In case the PCI BAR is larger than the actual amount of vram */
	adev->gmc.visible_vram_size = adev->gmc.aper_size;
	if (adev->gmc.visible_vram_size > adev->gmc.real_vram_size)
		adev->gmc.visible_vram_size = adev->gmc.real_vram_size;

	/* set the gart size */
	if (amdgpu_gart_size == -1) {
		switch (adev->asic_type) {
		case CHIP_VEGA10:  /* all engines support GPUVM */
		case CHIP_VEGA12:  /* all engines support GPUVM */
		case CHIP_VEGA20:
		default:
			adev->gmc.gart_size = 512ULL << 20;
			break;
		case CHIP_RAVEN:   /* DCE SG support */
			adev->gmc.gart_size = 1024ULL << 20;
			break;
		}
	} else {
		adev->gmc.gart_size = (u64)amdgpu_gart_size << 20;
	}

	gmc_v9_0_vram_gtt_location(adev, &adev->gmc);

	return 0;
}

static int gmc_v9_0_gart_init(struct amdgpu_device *adev)
{
	int r;

	if (adev->gart.bo) {
		WARN(1, "VEGA10 PCIE GART already initialized\n");
		return 0;
	}
	/* Initialize common gart structure */
	r = amdgpu_gart_init(adev);
	if (r)
		return r;
	adev->gart.table_size = adev->gart.num_gpu_pages * 8;
	adev->gart.gart_pte_flags = AMDGPU_PTE_MTYPE_VG10(MTYPE_UC) |
				 AMDGPU_PTE_EXECUTABLE;
	return amdgpu_gart_table_vram_alloc(adev);
}

static unsigned gmc_v9_0_get_vbios_fb_size(struct amdgpu_device *adev)
{
	u32 d1vga_control = RREG32_SOC15(DCE, 0, mmD1VGA_CONTROL);
	unsigned size;

	/*
	 * TODO Remove once GART corruption is resolved
	 * Check related code in gmc_v9_0_sw_fini
	 * */
	if (gmc_v9_0_keep_stolen_memory(adev))
		return 9 * 1024 * 1024;

	if (REG_GET_FIELD(d1vga_control, D1VGA_CONTROL, D1VGA_MODE_ENABLE)) {
		size = 9 * 1024 * 1024; /* reserve 8MB for vga emulator and 1 MB for FB */
	} else {
		u32 viewport;

		switch (adev->asic_type) {
		case CHIP_RAVEN:
			viewport = RREG32_SOC15(DCE, 0, mmHUBP0_DCSURF_PRI_VIEWPORT_DIMENSION);
			size = (REG_GET_FIELD(viewport,
					      HUBP0_DCSURF_PRI_VIEWPORT_DIMENSION, PRI_VIEWPORT_HEIGHT) *
				REG_GET_FIELD(viewport,
					      HUBP0_DCSURF_PRI_VIEWPORT_DIMENSION, PRI_VIEWPORT_WIDTH) *
				4);
			break;
		case CHIP_VEGA10:
		case CHIP_VEGA12:
		case CHIP_VEGA20:
		default:
			viewport = RREG32_SOC15(DCE, 0, mmSCL0_VIEWPORT_SIZE);
			size = (REG_GET_FIELD(viewport, SCL0_VIEWPORT_SIZE, VIEWPORT_HEIGHT) *
				REG_GET_FIELD(viewport, SCL0_VIEWPORT_SIZE, VIEWPORT_WIDTH) *
				4);
			break;
		}
	}
	/* return 0 if the pre-OS buffer uses up most of vram */
	if ((adev->gmc.real_vram_size - size) < (8 * 1024 * 1024))
		return 0;

	return size;
}

static int gmc_v9_0_sw_init(void *handle)
{
	int r;
	int dma_bits;
	struct amdgpu_device *adev = (struct amdgpu_device *)handle;

	gfxhub_v1_0_init(adev);
	mmhub_v1_0_init(adev);

	spin_lock_init(&adev->gmc.invalidate_lock);

	adev->gmc.vram_type = amdgpu_atomfirmware_get_vram_type(adev);
	switch (adev->asic_type) {
	case CHIP_RAVEN:
		if (adev->rev_id == 0x0 || adev->rev_id == 0x1) {
			amdgpu_vm_adjust_size(adev, 256 * 1024, 9, 3, 48);
		} else {
			/* vm_size is 128TB + 512GB for legacy 3-level page support */
			amdgpu_vm_adjust_size(adev, 128 * 1024 + 512, 9, 2, 48);
			adev->gmc.translate_further =
				adev->vm_manager.num_level > 1;
		}
		break;
	case CHIP_VEGA10:
	case CHIP_VEGA12:
	case CHIP_VEGA20:
		/*
		 * To fulfill 4-level page support,
		 * vm size is 256TB (48bit), maximum size of Vega10,
		 * block size 512 (9bit)
		 */
		/* sriov restrict max_pfn below AMDGPU_GMC_HOLE */
		if (amdgpu_sriov_vf(adev))
			amdgpu_vm_adjust_size(adev, 256 * 1024, 9, 3, 47);
		else
			amdgpu_vm_adjust_size(adev, 256 * 1024, 9, 3, 48);
		break;
	default:
		break;
	}

	/* This interrupt is VMC page fault.*/
	r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_VMC, VMC_1_0__SRCID__VM_FAULT,
				&adev->gmc.vm_fault);
	if (r)
		return r;

	r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_UTCL2, UTCL2_1_0__SRCID__FAULT,
				&adev->gmc.vm_fault);

	if (r)
		return r;

	/* interrupt sent to DF. */
	r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_DF, 0,
			&adev->gmc.ecc_irq);
	if (r)
		return r;

	/* Set the internal MC address mask
	 * This is the max address of the GPU's
	 * internal address space.
	 */
	adev->gmc.mc_mask = 0xffffffffffffULL; /* 48 bit MC */

	/* set DMA mask + need_dma32 flags.
	 * PCIE - can handle 44-bits.
	 * IGP - can handle 44-bits
	 * PCI - dma32 for legacy pci gart, 44 bits on vega10
	 */
	adev->need_dma32 = false;
	dma_bits = adev->need_dma32 ? 32 : 44;
	r = pci_set_dma_mask(adev->pdev, DMA_BIT_MASK(dma_bits));
	if (r) {
		adev->need_dma32 = true;
		dma_bits = 32;
		printk(KERN_WARNING "amdgpu: No suitable DMA available.\n");
	}
	r = pci_set_consistent_dma_mask(adev->pdev, DMA_BIT_MASK(dma_bits));
	if (r) {
		pci_set_consistent_dma_mask(adev->pdev, DMA_BIT_MASK(32));
		printk(KERN_WARNING "amdgpu: No coherent DMA available.\n");
	}
	adev->need_swiotlb = drm_need_swiotlb(dma_bits);

	if (adev->gmc.xgmi.supported) {
		r = gfxhub_v1_1_get_xgmi_info(adev);
		if (r)
			return r;
	}

	r = gmc_v9_0_mc_init(adev);
	if (r)
		return r;

	adev->gmc.stolen_size = gmc_v9_0_get_vbios_fb_size(adev);

	/* Memory manager */
	r = amdgpu_bo_init(adev);
	if (r)
		return r;

	r = gmc_v9_0_gart_init(adev);
	if (r)
		return r;

	/*
	 * number of VMs
	 * VMID 0 is reserved for System
	 * amdgpu graphics/compute will use VMIDs 1-7
	 * amdkfd will use VMIDs 8-15
	 */
	adev->vm_manager.id_mgr[AMDGPU_GFXHUB].num_ids = AMDGPU_NUM_OF_VMIDS;
	adev->vm_manager.id_mgr[AMDGPU_MMHUB].num_ids = AMDGPU_NUM_OF_VMIDS;

	amdgpu_vm_manager_init(adev);

	return 0;
}

static int gmc_v9_0_sw_fini(void *handle)
{
	struct amdgpu_device *adev = (struct amdgpu_device *)handle;

	if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__UMC) &&
			adev->gmc.ras_if) {
		struct ras_common_if *ras_if = adev->gmc.ras_if;
		struct ras_ih_if ih_info = {
			.head = *ras_if,
		};

		/*remove fs first*/
		amdgpu_ras_debugfs_remove(adev, ras_if);
		amdgpu_ras_sysfs_remove(adev, ras_if);
		/*remove the IH*/
		amdgpu_ras_interrupt_remove_handler(adev, &ih_info);
		amdgpu_ras_feature_enable(adev, ras_if, 0);
		kfree(ras_if);
	}

	amdgpu_gem_force_release(adev);
	amdgpu_vm_manager_fini(adev);

	if (gmc_v9_0_keep_stolen_memory(adev))
		amdgpu_bo_free_kernel(&adev->stolen_vga_memory, NULL, NULL);

	amdgpu_gart_table_vram_free(adev);
	amdgpu_bo_fini(adev);
	amdgpu_gart_fini(adev);

	return 0;
}

static void gmc_v9_0_init_golden_registers(struct amdgpu_device *adev)
{

	switch (adev->asic_type) {
	case CHIP_VEGA10:
		if (amdgpu_virt_support_skip_setting(adev))
			break;
		/* fall through */
	case CHIP_VEGA20:
		soc15_program_register_sequence(adev,
						golden_settings_mmhub_1_0_0,
						ARRAY_SIZE(golden_settings_mmhub_1_0_0));
		soc15_program_register_sequence(adev,
						golden_settings_athub_1_0_0,
						ARRAY_SIZE(golden_settings_athub_1_0_0));
		break;
	case CHIP_VEGA12:
		break;
	case CHIP_RAVEN:
		soc15_program_register_sequence(adev,
						golden_settings_athub_1_0_0,
						ARRAY_SIZE(golden_settings_athub_1_0_0));
		break;
	default:
		break;
	}
}

/**
 * gmc_v9_0_gart_enable - gart enable
 *
 * @adev: amdgpu_device pointer
 */
static int gmc_v9_0_gart_enable(struct amdgpu_device *adev)
{
	int r;
	bool value;
	u32 tmp;

	amdgpu_device_program_register_sequence(adev,
						golden_settings_vega10_hdp,
						ARRAY_SIZE(golden_settings_vega10_hdp));

	if (adev->gart.bo == NULL) {
		dev_err(adev->dev, "No VRAM object for PCIE GART.\n");
		return -EINVAL;
	}
	r = amdgpu_gart_table_vram_pin(adev);
	if (r)
		return r;

	switch (adev->asic_type) {
	case CHIP_RAVEN:
		mmhub_v1_0_update_power_gating(adev, true);
		break;
	default:
		break;
	}

	r = gfxhub_v1_0_gart_enable(adev);
	if (r)
		return r;

	r = mmhub_v1_0_gart_enable(adev);
	if (r)
		return r;

	WREG32_FIELD15(HDP, 0, HDP_MISC_CNTL, FLUSH_INVALIDATE_CACHE, 1);

	tmp = RREG32_SOC15(HDP, 0, mmHDP_HOST_PATH_CNTL);
	WREG32_SOC15(HDP, 0, mmHDP_HOST_PATH_CNTL, tmp);

	WREG32_SOC15(HDP, 0, mmHDP_NONSURFACE_BASE, (adev->gmc.vram_start >> 8));
	WREG32_SOC15(HDP, 0, mmHDP_NONSURFACE_BASE_HI, (adev->gmc.vram_start >> 40));

	/* After HDP is initialized, flush HDP.*/
	adev->nbio_funcs->hdp_flush(adev, NULL);

	if (amdgpu_vm_fault_stop == AMDGPU_VM_FAULT_STOP_ALWAYS)
		value = false;
	else
		value = true;

	gfxhub_v1_0_set_fault_enable_default(adev, value);
	mmhub_v1_0_set_fault_enable_default(adev, value);
	gmc_v9_0_flush_gpu_tlb(adev, 0, 0);

	DRM_INFO("PCIE GART of %uM enabled (table at 0x%016llX).\n",
		 (unsigned)(adev->gmc.gart_size >> 20),
		 (unsigned long long)amdgpu_bo_gpu_offset(adev->gart.bo));
	adev->gart.ready = true;
	return 0;
}

static int gmc_v9_0_hw_init(void *handle)
{
	int r;
	struct amdgpu_device *adev = (struct amdgpu_device *)handle;

	/* The sequence of these two function calls matters.*/
	gmc_v9_0_init_golden_registers(adev);

	if (adev->mode_info.num_crtc) {
		/* Lockout access through VGA aperture*/
		WREG32_FIELD15(DCE, 0, VGA_HDP_CONTROL, VGA_MEMORY_DISABLE, 1);

		/* disable VGA render */
		WREG32_FIELD15(DCE, 0, VGA_RENDER_CONTROL, VGA_VSTATUS_CNTL, 0);
	}

	r = gmc_v9_0_gart_enable(adev);

	return r;
}

/**
 * gmc_v9_0_gart_disable - gart disable
 *
 * @adev: amdgpu_device pointer
 *
 * This disables all VM page table.
 */
static void gmc_v9_0_gart_disable(struct amdgpu_device *adev)
{
	gfxhub_v1_0_gart_disable(adev);
	mmhub_v1_0_gart_disable(adev);
	amdgpu_gart_table_vram_unpin(adev);
}

static int gmc_v9_0_hw_fini(void *handle)
{
	struct amdgpu_device *adev = (struct amdgpu_device *)handle;

	if (amdgpu_sriov_vf(adev)) {
		/* full access mode, so don't touch any GMC register */
		DRM_DEBUG("For SRIOV client, shouldn't do anything.\n");
		return 0;
	}

	amdgpu_irq_put(adev, &adev->gmc.ecc_irq, 0);
	amdgpu_irq_put(adev, &adev->gmc.vm_fault, 0);
	gmc_v9_0_gart_disable(adev);

	return 0;
}

static int gmc_v9_0_suspend(void *handle)
{
	struct amdgpu_device *adev = (struct amdgpu_device *)handle;

	return gmc_v9_0_hw_fini(adev);
}

static int gmc_v9_0_resume(void *handle)
{
	int r;
	struct amdgpu_device *adev = (struct amdgpu_device *)handle;

	r = gmc_v9_0_hw_init(adev);
	if (r)
		return r;

	amdgpu_vmid_reset_all(adev);

	return 0;
}

static bool gmc_v9_0_is_idle(void *handle)
{
	/* MC is always ready in GMC v9.*/
	return true;
}

static int gmc_v9_0_wait_for_idle(void *handle)
{
	/* There is no need to wait for MC idle in GMC v9.*/
	return 0;
}

static int gmc_v9_0_soft_reset(void *handle)
{
	/* XXX for emulation.*/
	return 0;
}

static int gmc_v9_0_set_clockgating_state(void *handle,
					enum amd_clockgating_state state)
{
	struct amdgpu_device *adev = (struct amdgpu_device *)handle;

	return mmhub_v1_0_set_clockgating(adev, state);
}

static void gmc_v9_0_get_clockgating_state(void *handle, u32 *flags)
{
	struct amdgpu_device *adev = (struct amdgpu_device *)handle;

	mmhub_v1_0_get_clockgating(adev, flags);
}

static int gmc_v9_0_set_powergating_state(void *handle,
					enum amd_powergating_state state)
{
	return 0;
}

const struct amd_ip_funcs gmc_v9_0_ip_funcs = {
	.name = "gmc_v9_0",
	.early_init = gmc_v9_0_early_init,
	.late_init = gmc_v9_0_late_init,
	.sw_init = gmc_v9_0_sw_init,
	.sw_fini = gmc_v9_0_sw_fini,
	.hw_init = gmc_v9_0_hw_init,
	.hw_fini = gmc_v9_0_hw_fini,
	.suspend = gmc_v9_0_suspend,
	.resume = gmc_v9_0_resume,
	.is_idle = gmc_v9_0_is_idle,
	.wait_for_idle = gmc_v9_0_wait_for_idle,
	.soft_reset = gmc_v9_0_soft_reset,
	.set_clockgating_state = gmc_v9_0_set_clockgating_state,
	.set_powergating_state = gmc_v9_0_set_powergating_state,
	.get_clockgating_state = gmc_v9_0_get_clockgating_state,
};

const struct amdgpu_ip_block_version gmc_v9_0_ip_block =
{
	.type = AMD_IP_BLOCK_TYPE_GMC,
	.major = 9,
	.minor = 0,
	.rev = 0,
	.funcs = &gmc_v9_0_ip_funcs,
};