aboutsummaryrefslogtreecommitdiffstatshomepage
path: root/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c
blob: d2840c2f62865c23229d599fecfa530be2be7ce7 (plain) (blame)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
1637
1638
1639
1640
1641
1642
1643
1644
1645
1646
1647
1648
1649
1650
1651
1652
1653
1654
1655
1656
1657
1658
1659
1660
1661
1662
1663
1664
1665
1666
1667
1668
1669
1670
1671
1672
1673
1674
1675
1676
1677
1678
1679
1680
1681
1682
1683
1684
1685
1686
1687
1688
1689
1690
1691
1692
1693
1694
1695
1696
1697
1698
1699
1700
1701
1702
1703
1704
1705
1706
1707
1708
1709
1710
1711
1712
1713
1714
1715
1716
1717
1718
1719
1720
1721
1722
1723
1724
1725
1726
1727
1728
1729
1730
1731
1732
1733
1734
1735
1736
1737
1738
1739
1740
1741
1742
1743
1744
1745
1746
1747
1748
1749
/*
 * Copyright 2019 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/delay.h>
#include <linux/firmware.h>
#include <linux/module.h>
#include <linux/pci.h>

#include "amdgpu.h"
#include "amdgpu_ucode.h"
#include "amdgpu_trace.h"

#include "gc/gc_10_1_0_offset.h"
#include "gc/gc_10_1_0_sh_mask.h"
#include "hdp/hdp_5_0_0_offset.h"
#include "ivsrcid/sdma0/irqsrcs_sdma0_5_0.h"
#include "ivsrcid/sdma1/irqsrcs_sdma1_5_0.h"

#include "soc15_common.h"
#include "soc15.h"
#include "navi10_sdma_pkt_open.h"
#include "nbio_v2_3.h"
#include "sdma_v5_0.h"

MODULE_FIRMWARE("amdgpu/navi10_sdma.bin");
MODULE_FIRMWARE("amdgpu/navi10_sdma1.bin");

MODULE_FIRMWARE("amdgpu/navi14_sdma.bin");
MODULE_FIRMWARE("amdgpu/navi14_sdma1.bin");

MODULE_FIRMWARE("amdgpu/navi12_sdma.bin");
MODULE_FIRMWARE("amdgpu/navi12_sdma1.bin");

#define SDMA1_REG_OFFSET 0x600
#define SDMA0_HYP_DEC_REG_START 0x5880
#define SDMA0_HYP_DEC_REG_END 0x5893
#define SDMA1_HYP_DEC_REG_OFFSET 0x20

static void sdma_v5_0_set_ring_funcs(struct amdgpu_device *adev);
static void sdma_v5_0_set_buffer_funcs(struct amdgpu_device *adev);
static void sdma_v5_0_set_vm_pte_funcs(struct amdgpu_device *adev);
static void sdma_v5_0_set_irq_funcs(struct amdgpu_device *adev);

static const struct soc15_reg_golden golden_settings_sdma_5[] = {
	SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_CHICKEN_BITS, 0xffbf1f0f, 0x03ab0107),
	SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_GFX_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000),
	SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_PAGE_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000),
	SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_RLC0_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000),
	SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_RLC1_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000),
	SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_RLC2_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000),
	SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_RLC3_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000),
	SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_RLC4_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000),
	SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_RLC5_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000),
	SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_RLC6_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000),
	SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_RLC7_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000),
	SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_UTCL1_PAGE, 0x00ffffff, 0x000c5c00),
	SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_CHICKEN_BITS, 0xffbf1f0f, 0x03ab0107),
	SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_GFX_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000),
	SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_PAGE_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000),
	SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_RLC0_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000),
	SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_RLC1_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000),
	SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_RLC2_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000),
	SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_RLC3_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000),
	SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_RLC4_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000),
	SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_RLC5_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000),
	SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_RLC6_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000),
	SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_RLC7_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000),
	SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_UTCL1_PAGE, 0x00ffffff, 0x000c5c00)
};

static const struct soc15_reg_golden golden_settings_sdma_nv10[] = {
	SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_RLC3_RB_WPTR_POLL_CNTL, 0x0000fff0, 0x00403000),
	SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_RLC3_RB_WPTR_POLL_CNTL, 0x0000fff0, 0x00403000),
};

static const struct soc15_reg_golden golden_settings_sdma_nv14[] = {
	SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_RLC3_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000),
	SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_RLC3_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000),
};

static const struct soc15_reg_golden golden_settings_sdma_nv12[] = {
	SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_RLC3_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000),
	SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_RLC3_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000),
};

static u32 sdma_v5_0_get_reg_offset(struct amdgpu_device *adev, u32 instance, u32 internal_offset)
{
	u32 base;

	if (internal_offset >= SDMA0_HYP_DEC_REG_START &&
	    internal_offset <= SDMA0_HYP_DEC_REG_END) {
		base = adev->reg_offset[GC_HWIP][0][1];
		if (instance == 1)
			internal_offset += SDMA1_HYP_DEC_REG_OFFSET;
	} else {
		base = adev->reg_offset[GC_HWIP][0][0];
		if (instance == 1)
			internal_offset += SDMA1_REG_OFFSET;
	}

	return base + internal_offset;
}

static void sdma_v5_0_init_golden_registers(struct amdgpu_device *adev)
{
	switch (adev->asic_type) {
	case CHIP_NAVI10:
		soc15_program_register_sequence(adev,
						golden_settings_sdma_5,
						(const u32)ARRAY_SIZE(golden_settings_sdma_5));
		soc15_program_register_sequence(adev,
						golden_settings_sdma_nv10,
						(const u32)ARRAY_SIZE(golden_settings_sdma_nv10));
		break;
	case CHIP_NAVI14:
		soc15_program_register_sequence(adev,
						golden_settings_sdma_5,
						(const u32)ARRAY_SIZE(golden_settings_sdma_5));
		soc15_program_register_sequence(adev,
						golden_settings_sdma_nv14,
						(const u32)ARRAY_SIZE(golden_settings_sdma_nv14));
		break;
	case CHIP_NAVI12:
		soc15_program_register_sequence(adev,
						golden_settings_sdma_5,
						(const u32)ARRAY_SIZE(golden_settings_sdma_5));
		soc15_program_register_sequence(adev,
						golden_settings_sdma_nv12,
						(const u32)ARRAY_SIZE(golden_settings_sdma_nv12));
		break;
	default:
		break;
	}
}

/**
 * sdma_v5_0_init_microcode - load ucode images from disk
 *
 * @adev: amdgpu_device pointer
 *
 * Use the firmware interface to load the ucode images into
 * the driver (not loaded into hw).
 * Returns 0 on success, error on failure.
 */

// emulation only, won't work on real chip
// navi10 real chip need to use PSP to load firmware
static int sdma_v5_0_init_microcode(struct amdgpu_device *adev)
{
	const char *chip_name;
	char fw_name[30];
	int err = 0, i;
	struct amdgpu_firmware_info *info = NULL;
	const struct common_firmware_header *header = NULL;
	const struct sdma_firmware_header_v1_0 *hdr;

	DRM_DEBUG("\n");

	switch (adev->asic_type) {
	case CHIP_NAVI10:
		chip_name = "navi10";
		break;
	case CHIP_NAVI14:
		chip_name = "navi14";
		break;
	case CHIP_NAVI12:
		chip_name = "navi12";
		break;
	default:
		BUG();
	}

	for (i = 0; i < adev->sdma.num_instances; i++) {
		if (i == 0)
			snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_sdma.bin", chip_name);
		else
			snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_sdma1.bin", chip_name);
		err = request_firmware(&adev->sdma.instance[i].fw, fw_name, adev->dev);
		if (err)
			goto out;
		err = amdgpu_ucode_validate(adev->sdma.instance[i].fw);
		if (err)
			goto out;
		hdr = (const struct sdma_firmware_header_v1_0 *)adev->sdma.instance[i].fw->data;
		adev->sdma.instance[i].fw_version = le32_to_cpu(hdr->header.ucode_version);
		adev->sdma.instance[i].feature_version = le32_to_cpu(hdr->ucode_feature_version);
		if (adev->sdma.instance[i].feature_version >= 20)
			adev->sdma.instance[i].burst_nop = true;
		DRM_DEBUG("psp_load == '%s'\n",
				adev->firmware.load_type == AMDGPU_FW_LOAD_PSP ? "true" : "false");

		if (adev->firmware.load_type == AMDGPU_FW_LOAD_PSP) {
			info = &adev->firmware.ucode[AMDGPU_UCODE_ID_SDMA0 + i];
			info->ucode_id = AMDGPU_UCODE_ID_SDMA0 + i;
			info->fw = adev->sdma.instance[i].fw;
			header = (const struct common_firmware_header *)info->fw->data;
			adev->firmware.fw_size +=
				ALIGN(le32_to_cpu(header->ucode_size_bytes), PAGE_SIZE);
		}
	}
out:
	if (err) {
		DRM_ERROR("sdma_v5_0: Failed to load firmware \"%s\"\n", fw_name);
		for (i = 0; i < adev->sdma.num_instances; i++) {
			release_firmware(adev->sdma.instance[i].fw);
			adev->sdma.instance[i].fw = NULL;
		}
	}
	return err;
}

static unsigned sdma_v5_0_ring_init_cond_exec(struct amdgpu_ring *ring)
{
	unsigned ret;

	amdgpu_ring_write(ring, SDMA_PKT_HEADER_OP(SDMA_OP_COND_EXE));
	amdgpu_ring_write(ring, lower_32_bits(ring->cond_exe_gpu_addr));
	amdgpu_ring_write(ring, upper_32_bits(ring->cond_exe_gpu_addr));
	amdgpu_ring_write(ring, 1);
	ret = ring->wptr & ring->buf_mask;/* this is the offset we need patch later */
	amdgpu_ring_write(ring, 0x55aa55aa);/* insert dummy here and patch it later */

	return ret;
}

static void sdma_v5_0_ring_patch_cond_exec(struct amdgpu_ring *ring,
					   unsigned offset)
{
	unsigned cur;

	BUG_ON(offset > ring->buf_mask);
	BUG_ON(ring->ring[offset] != 0x55aa55aa);

	cur = (ring->wptr - 1) & ring->buf_mask;
	if (cur > offset)
		ring->ring[offset] = cur - offset;
	else
		ring->ring[offset] = (ring->buf_mask + 1) - offset + cur;
}

/**
 * sdma_v5_0_ring_get_rptr - get the current read pointer
 *
 * @ring: amdgpu ring pointer
 *
 * Get the current rptr from the hardware (NAVI10+).
 */
static uint64_t sdma_v5_0_ring_get_rptr(struct amdgpu_ring *ring)
{
	u64 *rptr;

	/* XXX check if swapping is necessary on BE */
	rptr = ((u64 *)&ring->adev->wb.wb[ring->rptr_offs]);

	DRM_DEBUG("rptr before shift == 0x%016llx\n", *rptr);
	return ((*rptr) >> 2);
}

/**
 * sdma_v5_0_ring_get_wptr - get the current write pointer
 *
 * @ring: amdgpu ring pointer
 *
 * Get the current wptr from the hardware (NAVI10+).
 */
static uint64_t sdma_v5_0_ring_get_wptr(struct amdgpu_ring *ring)
{
	struct amdgpu_device *adev = ring->adev;
	u64 *wptr = NULL;
	uint64_t local_wptr = 0;

	if (ring->use_doorbell) {
		/* XXX check if swapping is necessary on BE */
		wptr = ((u64 *)&adev->wb.wb[ring->wptr_offs]);
		DRM_DEBUG("wptr/doorbell before shift == 0x%016llx\n", *wptr);
		*wptr = (*wptr) >> 2;
		DRM_DEBUG("wptr/doorbell after shift == 0x%016llx\n", *wptr);
	} else {
		u32 lowbit, highbit;

		wptr = &local_wptr;
		lowbit = RREG32(sdma_v5_0_get_reg_offset(adev, ring->me, mmSDMA0_GFX_RB_WPTR)) >> 2;
		highbit = RREG32(sdma_v5_0_get_reg_offset(adev, ring->me, mmSDMA0_GFX_RB_WPTR_HI)) >> 2;

		DRM_DEBUG("wptr [%i]high== 0x%08x low==0x%08x\n",
				ring->me, highbit, lowbit);
		*wptr = highbit;
		*wptr = (*wptr) << 32;
		*wptr |= lowbit;
	}

	return *wptr;
}

/**
 * sdma_v5_0_ring_set_wptr - commit the write pointer
 *
 * @ring: amdgpu ring pointer
 *
 * Write the wptr back to the hardware (NAVI10+).
 */
static void sdma_v5_0_ring_set_wptr(struct amdgpu_ring *ring)
{
	struct amdgpu_device *adev = ring->adev;

	DRM_DEBUG("Setting write pointer\n");
	if (ring->use_doorbell) {
		DRM_DEBUG("Using doorbell -- "
				"wptr_offs == 0x%08x "
				"lower_32_bits(ring->wptr) << 2 == 0x%08x "
				"upper_32_bits(ring->wptr) << 2 == 0x%08x\n",
				ring->wptr_offs,
				lower_32_bits(ring->wptr << 2),
				upper_32_bits(ring->wptr << 2));
		/* XXX check if swapping is necessary on BE */
		adev->wb.wb[ring->wptr_offs] = lower_32_bits(ring->wptr << 2);
		adev->wb.wb[ring->wptr_offs + 1] = upper_32_bits(ring->wptr << 2);
		DRM_DEBUG("calling WDOORBELL64(0x%08x, 0x%016llx)\n",
				ring->doorbell_index, ring->wptr << 2);
		WDOORBELL64(ring->doorbell_index, ring->wptr << 2);
	} else {
		DRM_DEBUG("Not using doorbell -- "
				"mmSDMA%i_GFX_RB_WPTR == 0x%08x "
				"mmSDMA%i_GFX_RB_WPTR_HI == 0x%08x\n",
				ring->me,
				lower_32_bits(ring->wptr << 2),
				ring->me,
				upper_32_bits(ring->wptr << 2));
		WREG32(sdma_v5_0_get_reg_offset(adev, ring->me, mmSDMA0_GFX_RB_WPTR),
			lower_32_bits(ring->wptr << 2));
		WREG32(sdma_v5_0_get_reg_offset(adev, ring->me, mmSDMA0_GFX_RB_WPTR_HI),
			upper_32_bits(ring->wptr << 2));
	}
}

static void sdma_v5_0_ring_insert_nop(struct amdgpu_ring *ring, uint32_t count)
{
	struct amdgpu_sdma_instance *sdma = amdgpu_sdma_get_instance_from_ring(ring);
	int i;

	for (i = 0; i < count; i++)
		if (sdma && sdma->burst_nop && (i == 0))
			amdgpu_ring_write(ring, ring->funcs->nop |
				SDMA_PKT_NOP_HEADER_COUNT(count - 1));
		else
			amdgpu_ring_write(ring, ring->funcs->nop);
}

/**
 * sdma_v5_0_ring_emit_ib - Schedule an IB on the DMA engine
 *
 * @ring: amdgpu ring pointer
 * @ib: IB object to schedule
 *
 * Schedule an IB in the DMA ring (NAVI10).
 */
static void sdma_v5_0_ring_emit_ib(struct amdgpu_ring *ring,
				   struct amdgpu_job *job,
				   struct amdgpu_ib *ib,
				   uint32_t flags)
{
	unsigned vmid = AMDGPU_JOB_GET_VMID(job);
	uint64_t csa_mc_addr = amdgpu_sdma_get_csa_mc_addr(ring, vmid);

	/* Invalidate L2, because if we don't do it, we might get stale cache
	 * lines from previous IBs.
	 */
	amdgpu_ring_write(ring, SDMA_PKT_HEADER_OP(SDMA_OP_GCR_REQ));
	amdgpu_ring_write(ring, 0);
	amdgpu_ring_write(ring, (SDMA_GCR_GL2_INV |
				 SDMA_GCR_GL2_WB |
				 SDMA_GCR_GLM_INV |
				 SDMA_GCR_GLM_WB) << 16);
	amdgpu_ring_write(ring, 0xffffff80);
	amdgpu_ring_write(ring, 0xffff);

	/* An IB packet must end on a 8 DW boundary--the next dword
	 * must be on a 8-dword boundary. Our IB packet below is 6
	 * dwords long, thus add x number of NOPs, such that, in
	 * modular arithmetic,
	 * wptr + 6 + x = 8k, k >= 0, which in C is,
	 * (wptr + 6 + x) % 8 = 0.
	 * The expression below, is a solution of x.
	 */
	sdma_v5_0_ring_insert_nop(ring, (2 - lower_32_bits(ring->wptr)) & 7);

	amdgpu_ring_write(ring, SDMA_PKT_HEADER_OP(SDMA_OP_INDIRECT) |
			  SDMA_PKT_INDIRECT_HEADER_VMID(vmid & 0xf));
	/* base must be 32 byte aligned */
	amdgpu_ring_write(ring, lower_32_bits(ib->gpu_addr) & 0xffffffe0);
	amdgpu_ring_write(ring, upper_32_bits(ib->gpu_addr));
	amdgpu_ring_write(ring, ib->length_dw);
	amdgpu_ring_write(ring, lower_32_bits(csa_mc_addr));
	amdgpu_ring_write(ring, upper_32_bits(csa_mc_addr));
}

/**
 * sdma_v5_0_ring_emit_hdp_flush - emit an hdp flush on the DMA ring
 *
 * @ring: amdgpu ring pointer
 *
 * Emit an hdp flush packet on the requested DMA ring.
 */
static void sdma_v5_0_ring_emit_hdp_flush(struct amdgpu_ring *ring)
{
	struct amdgpu_device *adev = ring->adev;
	u32 ref_and_mask = 0;
	const struct nbio_hdp_flush_reg *nbio_hf_reg = adev->nbio.hdp_flush_reg;

	if (ring->me == 0)
		ref_and_mask = nbio_hf_reg->ref_and_mask_sdma0;
	else
		ref_and_mask = nbio_hf_reg->ref_and_mask_sdma1;

	amdgpu_ring_write(ring, SDMA_PKT_HEADER_OP(SDMA_OP_POLL_REGMEM) |
			  SDMA_PKT_POLL_REGMEM_HEADER_HDP_FLUSH(1) |
			  SDMA_PKT_POLL_REGMEM_HEADER_FUNC(3)); /* == */
	amdgpu_ring_write(ring, (adev->nbio.funcs->get_hdp_flush_done_offset(adev)) << 2);
	amdgpu_ring_write(ring, (adev->nbio.funcs->get_hdp_flush_req_offset(adev)) << 2);
	amdgpu_ring_write(ring, ref_and_mask); /* reference */
	amdgpu_ring_write(ring, ref_and_mask); /* mask */
	amdgpu_ring_write(ring, SDMA_PKT_POLL_REGMEM_DW5_RETRY_COUNT(0xfff) |
			  SDMA_PKT_POLL_REGMEM_DW5_INTERVAL(10)); /* retry count, poll interval */
}

/**
 * sdma_v5_0_ring_emit_fence - emit a fence on the DMA ring
 *
 * @ring: amdgpu ring pointer
 * @fence: amdgpu fence object
 *
 * Add a DMA fence packet to the ring to write
 * the fence seq number and DMA trap packet to generate
 * an interrupt if needed (NAVI10).
 */
static void sdma_v5_0_ring_emit_fence(struct amdgpu_ring *ring, u64 addr, u64 seq,
				      unsigned flags)
{
	struct amdgpu_device *adev = ring->adev;
	bool write64bit = flags & AMDGPU_FENCE_FLAG_64BIT;
	/* write the fence */
	amdgpu_ring_write(ring, SDMA_PKT_HEADER_OP(SDMA_OP_FENCE) |
			  SDMA_PKT_FENCE_HEADER_MTYPE(0x3)); /* Ucached(UC) */
	/* zero in first two bits */
	BUG_ON(addr & 0x3);
	amdgpu_ring_write(ring, lower_32_bits(addr));
	amdgpu_ring_write(ring, upper_32_bits(addr));
	amdgpu_ring_write(ring, lower_32_bits(seq));

	/* optionally write high bits as well */
	if (write64bit) {
		addr += 4;
		amdgpu_ring_write(ring, SDMA_PKT_HEADER_OP(SDMA_OP_FENCE) |
				  SDMA_PKT_FENCE_HEADER_MTYPE(0x3));
		/* zero in first two bits */
		BUG_ON(addr & 0x3);
		amdgpu_ring_write(ring, lower_32_bits(addr));
		amdgpu_ring_write(ring, upper_32_bits(addr));
		amdgpu_ring_write(ring, upper_32_bits(seq));
	}

	/* Interrupt not work fine on GFX10.1 model yet. Use fallback instead */
	if ((flags & AMDGPU_FENCE_FLAG_INT) && adev->pdev->device != 0x50) {
		/* generate an interrupt */
		amdgpu_ring_write(ring, SDMA_PKT_HEADER_OP(SDMA_OP_TRAP));
		amdgpu_ring_write(ring, SDMA_PKT_TRAP_INT_CONTEXT_INT_CONTEXT(0));
	}
}


/**
 * sdma_v5_0_gfx_stop - stop the gfx async dma engines
 *
 * @adev: amdgpu_device pointer
 *
 * Stop the gfx async dma ring buffers (NAVI10).
 */
static void sdma_v5_0_gfx_stop(struct amdgpu_device *adev)
{
	struct amdgpu_ring *sdma0 = &adev->sdma.instance[0].ring;
	struct amdgpu_ring *sdma1 = &adev->sdma.instance[1].ring;
	u32 rb_cntl, ib_cntl;
	int i;

	if ((adev->mman.buffer_funcs_ring == sdma0) ||
	    (adev->mman.buffer_funcs_ring == sdma1))
		amdgpu_ttm_set_buffer_funcs_status(adev, false);

	for (i = 0; i < adev->sdma.num_instances; i++) {
		rb_cntl = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_CNTL));
		rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_GFX_RB_CNTL, RB_ENABLE, 0);
		WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_CNTL), rb_cntl);
		ib_cntl = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_IB_CNTL));
		ib_cntl = REG_SET_FIELD(ib_cntl, SDMA0_GFX_IB_CNTL, IB_ENABLE, 0);
		WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_IB_CNTL), ib_cntl);
	}

	sdma0->sched.ready = false;
	sdma1->sched.ready = false;
}

/**
 * sdma_v5_0_rlc_stop - stop the compute async dma engines
 *
 * @adev: amdgpu_device pointer
 *
 * Stop the compute async dma queues (NAVI10).
 */
static void sdma_v5_0_rlc_stop(struct amdgpu_device *adev)
{
	/* XXX todo */
}

/**
 * sdma_v_0_ctx_switch_enable - stop the async dma engines context switch
 *
 * @adev: amdgpu_device pointer
 * @enable: enable/disable the DMA MEs context switch.
 *
 * Halt or unhalt the async dma engines context switch (NAVI10).
 */
static void sdma_v5_0_ctx_switch_enable(struct amdgpu_device *adev, bool enable)
{
	u32 f32_cntl, phase_quantum = 0;
	int i;

	if (amdgpu_sdma_phase_quantum) {
		unsigned value = amdgpu_sdma_phase_quantum;
		unsigned unit = 0;

		while (value > (SDMA0_PHASE0_QUANTUM__VALUE_MASK >>
				SDMA0_PHASE0_QUANTUM__VALUE__SHIFT)) {
			value = (value + 1) >> 1;
			unit++;
		}
		if (unit > (SDMA0_PHASE0_QUANTUM__UNIT_MASK >>
			    SDMA0_PHASE0_QUANTUM__UNIT__SHIFT)) {
			value = (SDMA0_PHASE0_QUANTUM__VALUE_MASK >>
				 SDMA0_PHASE0_QUANTUM__VALUE__SHIFT);
			unit = (SDMA0_PHASE0_QUANTUM__UNIT_MASK >>
				SDMA0_PHASE0_QUANTUM__UNIT__SHIFT);
			WARN_ONCE(1,
			"clamping sdma_phase_quantum to %uK clock cycles\n",
				  value << unit);
		}
		phase_quantum =
			value << SDMA0_PHASE0_QUANTUM__VALUE__SHIFT |
			unit  << SDMA0_PHASE0_QUANTUM__UNIT__SHIFT;
	}

	for (i = 0; i < adev->sdma.num_instances; i++) {
		f32_cntl = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_CNTL));
		f32_cntl = REG_SET_FIELD(f32_cntl, SDMA0_CNTL,
				AUTO_CTXSW_ENABLE, enable ? 1 : 0);
		if (enable && amdgpu_sdma_phase_quantum) {
			WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_PHASE0_QUANTUM),
			       phase_quantum);
			WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_PHASE1_QUANTUM),
			       phase_quantum);
			WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_PHASE2_QUANTUM),
			       phase_quantum);
		}
		WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_CNTL), f32_cntl);
	}

}

/**
 * sdma_v5_0_enable - stop the async dma engines
 *
 * @adev: amdgpu_device pointer
 * @enable: enable/disable the DMA MEs.
 *
 * Halt or unhalt the async dma engines (NAVI10).
 */
static void sdma_v5_0_enable(struct amdgpu_device *adev, bool enable)
{
	u32 f32_cntl;
	int i;

	if (enable == false) {
		sdma_v5_0_gfx_stop(adev);
		sdma_v5_0_rlc_stop(adev);
	}

	for (i = 0; i < adev->sdma.num_instances; i++) {
		f32_cntl = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_F32_CNTL));
		f32_cntl = REG_SET_FIELD(f32_cntl, SDMA0_F32_CNTL, HALT, enable ? 0 : 1);
		WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_F32_CNTL), f32_cntl);
	}
}

/**
 * sdma_v5_0_gfx_resume - setup and start the async dma engines
 *
 * @adev: amdgpu_device pointer
 *
 * Set up the gfx DMA ring buffers and enable them (NAVI10).
 * Returns 0 for success, error for failure.
 */
static int sdma_v5_0_gfx_resume(struct amdgpu_device *adev)
{
	struct amdgpu_ring *ring;
	u32 rb_cntl, ib_cntl;
	u32 rb_bufsz;
	u32 wb_offset;
	u32 doorbell;
	u32 doorbell_offset;
	u32 temp;
	u32 wptr_poll_cntl;
	u64 wptr_gpu_addr;
	int i, r;

	for (i = 0; i < adev->sdma.num_instances; i++) {
		ring = &adev->sdma.instance[i].ring;
		wb_offset = (ring->rptr_offs * 4);

		WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_SEM_WAIT_FAIL_TIMER_CNTL), 0);

		/* Set ring buffer size in dwords */
		rb_bufsz = order_base_2(ring->ring_size / 4);
		rb_cntl = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_CNTL));
		rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_GFX_RB_CNTL, RB_SIZE, rb_bufsz);
#ifdef __BIG_ENDIAN
		rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_GFX_RB_CNTL, RB_SWAP_ENABLE, 1);
		rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_GFX_RB_CNTL,
					RPTR_WRITEBACK_SWAP_ENABLE, 1);
#endif
		WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_CNTL), rb_cntl);

		/* Initialize the ring buffer's read and write pointers */
		WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_RPTR), 0);
		WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_RPTR_HI), 0);
		WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_WPTR), 0);
		WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_WPTR_HI), 0);

		/* setup the wptr shadow polling */
		wptr_gpu_addr = adev->wb.gpu_addr + (ring->wptr_offs * 4);
		WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_WPTR_POLL_ADDR_LO),
		       lower_32_bits(wptr_gpu_addr));
		WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_WPTR_POLL_ADDR_HI),
		       upper_32_bits(wptr_gpu_addr));
		wptr_poll_cntl = RREG32(sdma_v5_0_get_reg_offset(adev, i,
							 mmSDMA0_GFX_RB_WPTR_POLL_CNTL));
		wptr_poll_cntl = REG_SET_FIELD(wptr_poll_cntl,
					       SDMA0_GFX_RB_WPTR_POLL_CNTL,
					       F32_POLL_ENABLE, 1);
		WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_WPTR_POLL_CNTL),
		       wptr_poll_cntl);

		/* set the wb address whether it's enabled or not */
		WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_RPTR_ADDR_HI),
		       upper_32_bits(adev->wb.gpu_addr + wb_offset) & 0xFFFFFFFF);
		WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_RPTR_ADDR_LO),
		       lower_32_bits(adev->wb.gpu_addr + wb_offset) & 0xFFFFFFFC);

		rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_GFX_RB_CNTL, RPTR_WRITEBACK_ENABLE, 1);

		WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_BASE), ring->gpu_addr >> 8);
		WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_BASE_HI), ring->gpu_addr >> 40);

		ring->wptr = 0;

		/* before programing wptr to a less value, need set minor_ptr_update first */
		WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_MINOR_PTR_UPDATE), 1);

		if (!amdgpu_sriov_vf(adev)) { /* only bare-metal use register write for wptr */
			WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_WPTR), lower_32_bits(ring->wptr) << 2);
			WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_WPTR_HI), upper_32_bits(ring->wptr) << 2);
		}

		doorbell = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_DOORBELL));
		doorbell_offset = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_DOORBELL_OFFSET));

		if (ring->use_doorbell) {
			doorbell = REG_SET_FIELD(doorbell, SDMA0_GFX_DOORBELL, ENABLE, 1);
			doorbell_offset = REG_SET_FIELD(doorbell_offset, SDMA0_GFX_DOORBELL_OFFSET,
					OFFSET, ring->doorbell_index);
		} else {
			doorbell = REG_SET_FIELD(doorbell, SDMA0_GFX_DOORBELL, ENABLE, 0);
		}
		WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_DOORBELL), doorbell);
		WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_DOORBELL_OFFSET), doorbell_offset);

		adev->nbio.funcs->sdma_doorbell_range(adev, i, ring->use_doorbell,
						      ring->doorbell_index, 20);

		if (amdgpu_sriov_vf(adev))
			sdma_v5_0_ring_set_wptr(ring);

		/* set minor_ptr_update to 0 after wptr programed */
		WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_MINOR_PTR_UPDATE), 0);

		/* set utc l1 enable flag always to 1 */
		temp = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_CNTL));
		temp = REG_SET_FIELD(temp, SDMA0_CNTL, UTC_L1_ENABLE, 1);

		/* enable MCBP */
		temp = REG_SET_FIELD(temp, SDMA0_CNTL, MIDCMD_PREEMPT_ENABLE, 1);
		WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_CNTL), temp);

		/* Set up RESP_MODE to non-copy addresses */
		temp = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_UTCL1_CNTL));
		temp = REG_SET_FIELD(temp, SDMA0_UTCL1_CNTL, RESP_MODE, 3);
		temp = REG_SET_FIELD(temp, SDMA0_UTCL1_CNTL, REDO_DELAY, 9);
		WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_UTCL1_CNTL), temp);

		/* program default cache read and write policy */
		temp = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_UTCL1_PAGE));
		/* clean read policy and write policy bits */
		temp &= 0xFF0FFF;
		temp |= ((CACHE_READ_POLICY_L2__DEFAULT << 12) | (CACHE_WRITE_POLICY_L2__DEFAULT << 14));
		WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_UTCL1_PAGE), temp);

		if (!amdgpu_sriov_vf(adev)) {
			/* unhalt engine */
			temp = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_F32_CNTL));
			temp = REG_SET_FIELD(temp, SDMA0_F32_CNTL, HALT, 0);
			WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_F32_CNTL), temp);
		}

		/* enable DMA RB */
		rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_GFX_RB_CNTL, RB_ENABLE, 1);
		WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_CNTL), rb_cntl);

		ib_cntl = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_IB_CNTL));
		ib_cntl = REG_SET_FIELD(ib_cntl, SDMA0_GFX_IB_CNTL, IB_ENABLE, 1);
#ifdef __BIG_ENDIAN
		ib_cntl = REG_SET_FIELD(ib_cntl, SDMA0_GFX_IB_CNTL, IB_SWAP_ENABLE, 1);
#endif
		/* enable DMA IBs */
		WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_IB_CNTL), ib_cntl);

		ring->sched.ready = true;

		if (amdgpu_sriov_vf(adev)) { /* bare-metal sequence doesn't need below to lines */
			sdma_v5_0_ctx_switch_enable(adev, true);
			sdma_v5_0_enable(adev, true);
		}

		r = amdgpu_ring_test_helper(ring);
		if (r)
			return r;

		if (adev->mman.buffer_funcs_ring == ring)
			amdgpu_ttm_set_buffer_funcs_status(adev, true);
	}

	return 0;
}

/**
 * sdma_v5_0_rlc_resume - setup and start the async dma engines
 *
 * @adev: amdgpu_device pointer
 *
 * Set up the compute DMA queues and enable them (NAVI10).
 * Returns 0 for success, error for failure.
 */
static int sdma_v5_0_rlc_resume(struct amdgpu_device *adev)
{
	return 0;
}

/**
 * sdma_v5_0_load_microcode - load the sDMA ME ucode
 *
 * @adev: amdgpu_device pointer
 *
 * Loads the sDMA0/1 ucode.
 * Returns 0 for success, -EINVAL if the ucode is not available.
 */
static int sdma_v5_0_load_microcode(struct amdgpu_device *adev)
{
	const struct sdma_firmware_header_v1_0 *hdr;
	const __le32 *fw_data;
	u32 fw_size;
	int i, j;

	/* halt the MEs */
	sdma_v5_0_enable(adev, false);

	for (i = 0; i < adev->sdma.num_instances; i++) {
		if (!adev->sdma.instance[i].fw)
			return -EINVAL;

		hdr = (const struct sdma_firmware_header_v1_0 *)adev->sdma.instance[i].fw->data;
		amdgpu_ucode_print_sdma_hdr(&hdr->header);
		fw_size = le32_to_cpu(hdr->header.ucode_size_bytes) / 4;

		fw_data = (const __le32 *)
			(adev->sdma.instance[i].fw->data +
				le32_to_cpu(hdr->header.ucode_array_offset_bytes));

		WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_UCODE_ADDR), 0);

		for (j = 0; j < fw_size; j++) {
			if (amdgpu_emu_mode == 1 && j % 500 == 0)
				msleep(1);
			WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_UCODE_DATA), le32_to_cpup(fw_data++));
		}

		WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_UCODE_ADDR), adev->sdma.instance[i].fw_version);
	}

	return 0;
}

/**
 * sdma_v5_0_start - setup and start the async dma engines
 *
 * @adev: amdgpu_device pointer
 *
 * Set up the DMA engines and enable them (NAVI10).
 * Returns 0 for success, error for failure.
 */
static int sdma_v5_0_start(struct amdgpu_device *adev)
{
	int r = 0;

	if (amdgpu_sriov_vf(adev)) {
		sdma_v5_0_ctx_switch_enable(adev, false);
		sdma_v5_0_enable(adev, false);

		/* set RB registers */
		r = sdma_v5_0_gfx_resume(adev);
		return r;
	}

	if (adev->firmware.load_type == AMDGPU_FW_LOAD_DIRECT) {
		r = sdma_v5_0_load_microcode(adev);
		if (r)
			return r;

		/* The value of mmSDMA_F32_CNTL is invalid the moment after loading fw */
		if (amdgpu_emu_mode == 1 && adev->pdev->device == 0x4d)
			msleep(1000);
	}

	/* unhalt the MEs */
	sdma_v5_0_enable(adev, true);
	/* enable sdma ring preemption */
	sdma_v5_0_ctx_switch_enable(adev, true);

	/* start the gfx rings and rlc compute queues */
	r = sdma_v5_0_gfx_resume(adev);
	if (r)
		return r;
	r = sdma_v5_0_rlc_resume(adev);

	return r;
}

/**
 * sdma_v5_0_ring_test_ring - simple async dma engine test
 *
 * @ring: amdgpu_ring structure holding ring information
 *
 * Test the DMA engine by writing using it to write an
 * value to memory. (NAVI10).
 * Returns 0 for success, error for failure.
 */
static int sdma_v5_0_ring_test_ring(struct amdgpu_ring *ring)
{
	struct amdgpu_device *adev = ring->adev;
	unsigned i;
	unsigned index;
	int r;
	u32 tmp;
	u64 gpu_addr;

	r = amdgpu_device_wb_get(adev, &index);
	if (r) {
		dev_err(adev->dev, "(%d) failed to allocate wb slot\n", r);
		return r;
	}

	gpu_addr = adev->wb.gpu_addr + (index * 4);
	tmp = 0xCAFEDEAD;
	adev->wb.wb[index] = cpu_to_le32(tmp);

	r = amdgpu_ring_alloc(ring, 5);
	if (r) {
		DRM_ERROR("amdgpu: dma failed to lock ring %d (%d).\n", ring->idx, r);
		amdgpu_device_wb_free(adev, index);
		return r;
	}

	amdgpu_ring_write(ring, SDMA_PKT_HEADER_OP(SDMA_OP_WRITE) |
			  SDMA_PKT_HEADER_SUB_OP(SDMA_SUBOP_WRITE_LINEAR));
	amdgpu_ring_write(ring, lower_32_bits(gpu_addr));
	amdgpu_ring_write(ring, upper_32_bits(gpu_addr));
	amdgpu_ring_write(ring, SDMA_PKT_WRITE_UNTILED_DW_3_COUNT(0));
	amdgpu_ring_write(ring, 0xDEADBEEF);
	amdgpu_ring_commit(ring);

	for (i = 0; i < adev->usec_timeout; i++) {
		tmp = le32_to_cpu(adev->wb.wb[index]);
		if (tmp == 0xDEADBEEF)
			break;
		if (amdgpu_emu_mode == 1)
			msleep(1);
		else
			udelay(1);
	}

	if (i >= adev->usec_timeout)
		r = -ETIMEDOUT;

	amdgpu_device_wb_free(adev, index);

	return r;
}

/**
 * sdma_v5_0_ring_test_ib - test an IB on the DMA engine
 *
 * @ring: amdgpu_ring structure holding ring information
 *
 * Test a simple IB in the DMA ring (NAVI10).
 * Returns 0 on success, error on failure.
 */
static int sdma_v5_0_ring_test_ib(struct amdgpu_ring *ring, long timeout)
{
	struct amdgpu_device *adev = ring->adev;
	struct amdgpu_ib ib;
	struct dma_fence *f = NULL;
	unsigned index;
	long r;
	u32 tmp = 0;
	u64 gpu_addr;

	r = amdgpu_device_wb_get(adev, &index);
	if (r) {
		dev_err(adev->dev, "(%ld) failed to allocate wb slot\n", r);
		return r;
	}

	gpu_addr = adev->wb.gpu_addr + (index * 4);
	tmp = 0xCAFEDEAD;
	adev->wb.wb[index] = cpu_to_le32(tmp);
	memset(&ib, 0, sizeof(ib));
	r = amdgpu_ib_get(adev, NULL, 256, &ib);
	if (r) {
		DRM_ERROR("amdgpu: failed to get ib (%ld).\n", r);
		goto err0;
	}

	ib.ptr[0] = SDMA_PKT_HEADER_OP(SDMA_OP_WRITE) |
		SDMA_PKT_HEADER_SUB_OP(SDMA_SUBOP_WRITE_LINEAR);
	ib.ptr[1] = lower_32_bits(gpu_addr);
	ib.ptr[2] = upper_32_bits(gpu_addr);
	ib.ptr[3] = SDMA_PKT_WRITE_UNTILED_DW_3_COUNT(0);
	ib.ptr[4] = 0xDEADBEEF;
	ib.ptr[5] = SDMA_PKT_NOP_HEADER_OP(SDMA_OP_NOP);
	ib.ptr[6] = SDMA_PKT_NOP_HEADER_OP(SDMA_OP_NOP);
	ib.ptr[7] = SDMA_PKT_NOP_HEADER_OP(SDMA_OP_NOP);
	ib.length_dw = 8;

	r = amdgpu_ib_schedule(ring, 1, &ib, NULL, &f);
	if (r)
		goto err1;

	r = dma_fence_wait_timeout(f, false, timeout);
	if (r == 0) {
		DRM_ERROR("amdgpu: IB test timed out\n");
		r = -ETIMEDOUT;
		goto err1;
	} else if (r < 0) {
		DRM_ERROR("amdgpu: fence wait failed (%ld).\n", r);
		goto err1;
	}
	tmp = le32_to_cpu(adev->wb.wb[index]);
	if (tmp == 0xDEADBEEF)
		r = 0;
	else
		r = -EINVAL;

err1:
	amdgpu_ib_free(adev, &ib, NULL);
	dma_fence_put(f);
err0:
	amdgpu_device_wb_free(adev, index);
	return r;
}


/**
 * sdma_v5_0_vm_copy_pte - update PTEs by copying them from the GART
 *
 * @ib: indirect buffer to fill with commands
 * @pe: addr of the page entry
 * @src: src addr to copy from
 * @count: number of page entries to update
 *
 * Update PTEs by copying them from the GART using sDMA (NAVI10).
 */
static void sdma_v5_0_vm_copy_pte(struct amdgpu_ib *ib,
				  uint64_t pe, uint64_t src,
				  unsigned count)
{
	unsigned bytes = count * 8;

	ib->ptr[ib->length_dw++] = SDMA_PKT_HEADER_OP(SDMA_OP_COPY) |
		SDMA_PKT_HEADER_SUB_OP(SDMA_SUBOP_COPY_LINEAR);
	ib->ptr[ib->length_dw++] = bytes - 1;
	ib->ptr[ib->length_dw++] = 0; /* src/dst endian swap */
	ib->ptr[ib->length_dw++] = lower_32_bits(src);
	ib->ptr[ib->length_dw++] = upper_32_bits(src);
	ib->ptr[ib->length_dw++] = lower_32_bits(pe);
	ib->ptr[ib->length_dw++] = upper_32_bits(pe);

}

/**
 * sdma_v5_0_vm_write_pte - update PTEs by writing them manually
 *
 * @ib: indirect buffer to fill with commands
 * @pe: addr of the page entry
 * @addr: dst addr to write into pe
 * @count: number of page entries to update
 * @incr: increase next addr by incr bytes
 * @flags: access flags
 *
 * Update PTEs by writing them manually using sDMA (NAVI10).
 */
static void sdma_v5_0_vm_write_pte(struct amdgpu_ib *ib, uint64_t pe,
				   uint64_t value, unsigned count,
				   uint32_t incr)
{
	unsigned ndw = count * 2;

	ib->ptr[ib->length_dw++] = SDMA_PKT_HEADER_OP(SDMA_OP_WRITE) |
		SDMA_PKT_HEADER_SUB_OP(SDMA_SUBOP_WRITE_LINEAR);
	ib->ptr[ib->length_dw++] = lower_32_bits(pe);
	ib->ptr[ib->length_dw++] = upper_32_bits(pe);
	ib->ptr[ib->length_dw++] = ndw - 1;
	for (; ndw > 0; ndw -= 2) {
		ib->ptr[ib->length_dw++] = lower_32_bits(value);
		ib->ptr[ib->length_dw++] = upper_32_bits(value);
		value += incr;
	}
}

/**
 * sdma_v5_0_vm_set_pte_pde - update the page tables using sDMA
 *
 * @ib: indirect buffer to fill with commands
 * @pe: addr of the page entry
 * @addr: dst addr to write into pe
 * @count: number of page entries to update
 * @incr: increase next addr by incr bytes
 * @flags: access flags
 *
 * Update the page tables using sDMA (NAVI10).
 */
static void sdma_v5_0_vm_set_pte_pde(struct amdgpu_ib *ib,
				     uint64_t pe,
				     uint64_t addr, unsigned count,
				     uint32_t incr, uint64_t flags)
{
	/* for physically contiguous pages (vram) */
	ib->ptr[ib->length_dw++] = SDMA_PKT_HEADER_OP(SDMA_OP_PTEPDE);
	ib->ptr[ib->length_dw++] = lower_32_bits(pe); /* dst addr */
	ib->ptr[ib->length_dw++] = upper_32_bits(pe);
	ib->ptr[ib->length_dw++] = lower_32_bits(flags); /* mask */
	ib->ptr[ib->length_dw++] = upper_32_bits(flags);
	ib->ptr[ib->length_dw++] = lower_32_bits(addr); /* value */
	ib->ptr[ib->length_dw++] = upper_32_bits(addr);
	ib->ptr[ib->length_dw++] = incr; /* increment size */
	ib->ptr[ib->length_dw++] = 0;
	ib->ptr[ib->length_dw++] = count - 1; /* number of entries */
}

/**
 * sdma_v5_0_ring_pad_ib - pad the IB
 * @ib: indirect buffer to fill with padding
 *
 * Pad the IB with NOPs to a boundary multiple of 8.
 */
static void sdma_v5_0_ring_pad_ib(struct amdgpu_ring *ring, struct amdgpu_ib *ib)
{
	struct amdgpu_sdma_instance *sdma = amdgpu_sdma_get_instance_from_ring(ring);
	u32 pad_count;
	int i;

	pad_count = (-ib->length_dw) & 0x7;
	for (i = 0; i < pad_count; i++)
		if (sdma && sdma->burst_nop && (i == 0))
			ib->ptr[ib->length_dw++] =
				SDMA_PKT_HEADER_OP(SDMA_OP_NOP) |
				SDMA_PKT_NOP_HEADER_COUNT(pad_count - 1);
		else
			ib->ptr[ib->length_dw++] =
				SDMA_PKT_HEADER_OP(SDMA_OP_NOP);
}


/**
 * sdma_v5_0_ring_emit_pipeline_sync - sync the pipeline
 *
 * @ring: amdgpu_ring pointer
 *
 * Make sure all previous operations are completed (CIK).
 */
static void sdma_v5_0_ring_emit_pipeline_sync(struct amdgpu_ring *ring)
{
	uint32_t seq = ring->fence_drv.sync_seq;
	uint64_t addr = ring->fence_drv.gpu_addr;

	/* wait for idle */
	amdgpu_ring_write(ring, SDMA_PKT_HEADER_OP(SDMA_OP_POLL_REGMEM) |
			  SDMA_PKT_POLL_REGMEM_HEADER_HDP_FLUSH(0) |
			  SDMA_PKT_POLL_REGMEM_HEADER_FUNC(3) | /* equal */
			  SDMA_PKT_POLL_REGMEM_HEADER_MEM_POLL(1));
	amdgpu_ring_write(ring, addr & 0xfffffffc);
	amdgpu_ring_write(ring, upper_32_bits(addr) & 0xffffffff);
	amdgpu_ring_write(ring, seq); /* reference */
	amdgpu_ring_write(ring, 0xffffffff); /* mask */
	amdgpu_ring_write(ring, SDMA_PKT_POLL_REGMEM_DW5_RETRY_COUNT(0xfff) |
			  SDMA_PKT_POLL_REGMEM_DW5_INTERVAL(4)); /* retry count, poll interval */
}


/**
 * sdma_v5_0_ring_emit_vm_flush - vm flush using sDMA
 *
 * @ring: amdgpu_ring pointer
 * @vm: amdgpu_vm pointer
 *
 * Update the page table base and flush the VM TLB
 * using sDMA (NAVI10).
 */
static void sdma_v5_0_ring_emit_vm_flush(struct amdgpu_ring *ring,
					 unsigned vmid, uint64_t pd_addr)
{
	amdgpu_gmc_emit_flush_gpu_tlb(ring, vmid, pd_addr);
}

static void sdma_v5_0_ring_emit_wreg(struct amdgpu_ring *ring,
				     uint32_t reg, uint32_t val)
{
	amdgpu_ring_write(ring, SDMA_PKT_HEADER_OP(SDMA_OP_SRBM_WRITE) |
			  SDMA_PKT_SRBM_WRITE_HEADER_BYTE_EN(0xf));
	amdgpu_ring_write(ring, reg);
	amdgpu_ring_write(ring, val);
}

static void sdma_v5_0_ring_emit_reg_wait(struct amdgpu_ring *ring, uint32_t reg,
					 uint32_t val, uint32_t mask)
{
	amdgpu_ring_write(ring, SDMA_PKT_HEADER_OP(SDMA_OP_POLL_REGMEM) |
			  SDMA_PKT_POLL_REGMEM_HEADER_HDP_FLUSH(0) |
			  SDMA_PKT_POLL_REGMEM_HEADER_FUNC(3)); /* equal */
	amdgpu_ring_write(ring, reg << 2);
	amdgpu_ring_write(ring, 0);
	amdgpu_ring_write(ring, val); /* reference */
	amdgpu_ring_write(ring, mask); /* mask */
	amdgpu_ring_write(ring, SDMA_PKT_POLL_REGMEM_DW5_RETRY_COUNT(0xfff) |
			  SDMA_PKT_POLL_REGMEM_DW5_INTERVAL(10));
}

static void sdma_v5_0_ring_emit_reg_write_reg_wait(struct amdgpu_ring *ring,
						   uint32_t reg0, uint32_t reg1,
						   uint32_t ref, uint32_t mask)
{
	amdgpu_ring_emit_wreg(ring, reg0, ref);
	/* wait for a cycle to reset vm_inv_eng*_ack */
	amdgpu_ring_emit_reg_wait(ring, reg0, 0, 0);
	amdgpu_ring_emit_reg_wait(ring, reg1, mask, mask);
}

static int sdma_v5_0_early_init(void *handle)
{
	struct amdgpu_device *adev = (struct amdgpu_device *)handle;

	adev->sdma.num_instances = 2;

	sdma_v5_0_set_ring_funcs(adev);
	sdma_v5_0_set_buffer_funcs(adev);
	sdma_v5_0_set_vm_pte_funcs(adev);
	sdma_v5_0_set_irq_funcs(adev);

	return 0;
}


static int sdma_v5_0_sw_init(void *handle)
{
	struct amdgpu_ring *ring;
	int r, i;
	struct amdgpu_device *adev = (struct amdgpu_device *)handle;

	/* SDMA trap event */
	r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_SDMA0,
			      SDMA0_5_0__SRCID__SDMA_TRAP,
			      &adev->sdma.trap_irq);
	if (r)
		return r;

	/* SDMA trap event */
	r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_SDMA1,
			      SDMA1_5_0__SRCID__SDMA_TRAP,
			      &adev->sdma.trap_irq);
	if (r)
		return r;

	r = sdma_v5_0_init_microcode(adev);
	if (r) {
		DRM_ERROR("Failed to load sdma firmware!\n");
		return r;
	}

	for (i = 0; i < adev->sdma.num_instances; i++) {
		ring = &adev->sdma.instance[i].ring;
		ring->ring_obj = NULL;
		ring->use_doorbell = true;

		DRM_INFO("use_doorbell being set to: [%s]\n",
				ring->use_doorbell?"true":"false");

		ring->doorbell_index = (i == 0) ?
			(adev->doorbell_index.sdma_engine[0] << 1) //get DWORD offset
			: (adev->doorbell_index.sdma_engine[1] << 1); // get DWORD offset

		sprintf(ring->name, "sdma%d", i);
		r = amdgpu_ring_init(adev, ring, 1024,
				     &adev->sdma.trap_irq,
				     (i == 0) ?
				     AMDGPU_SDMA_IRQ_INSTANCE0 :
				     AMDGPU_SDMA_IRQ_INSTANCE1);
		if (r)
			return r;
	}

	return r;
}

static int sdma_v5_0_sw_fini(void *handle)
{
	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
	int i;

	for (i = 0; i < adev->sdma.num_instances; i++)
		amdgpu_ring_fini(&adev->sdma.instance[i].ring);

	return 0;
}

static int sdma_v5_0_hw_init(void *handle)
{
	int r;
	struct amdgpu_device *adev = (struct amdgpu_device *)handle;

	sdma_v5_0_init_golden_registers(adev);

	r = sdma_v5_0_start(adev);

	return r;
}

static int sdma_v5_0_hw_fini(void *handle)
{
	struct amdgpu_device *adev = (struct amdgpu_device *)handle;

	if (amdgpu_sriov_vf(adev))
		return 0;

	sdma_v5_0_ctx_switch_enable(adev, false);
	sdma_v5_0_enable(adev, false);

	return 0;
}

static int sdma_v5_0_suspend(void *handle)
{
	struct amdgpu_device *adev = (struct amdgpu_device *)handle;

	return sdma_v5_0_hw_fini(adev);
}

static int sdma_v5_0_resume(void *handle)
{
	struct amdgpu_device *adev = (struct amdgpu_device *)handle;

	return sdma_v5_0_hw_init(adev);
}

static bool sdma_v5_0_is_idle(void *handle)
{
	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
	u32 i;

	for (i = 0; i < adev->sdma.num_instances; i++) {
		u32 tmp = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_STATUS_REG));

		if (!(tmp & SDMA0_STATUS_REG__IDLE_MASK))
			return false;
	}

	return true;
}

static int sdma_v5_0_wait_for_idle(void *handle)
{
	unsigned i;
	u32 sdma0, sdma1;
	struct amdgpu_device *adev = (struct amdgpu_device *)handle;

	for (i = 0; i < adev->usec_timeout; i++) {
		sdma0 = RREG32(sdma_v5_0_get_reg_offset(adev, 0, mmSDMA0_STATUS_REG));
		sdma1 = RREG32(sdma_v5_0_get_reg_offset(adev, 1, mmSDMA0_STATUS_REG));

		if (sdma0 & sdma1 & SDMA0_STATUS_REG__IDLE_MASK)
			return 0;
		udelay(1);
	}
	return -ETIMEDOUT;
}

static int sdma_v5_0_soft_reset(void *handle)
{
	/* todo */

	return 0;
}

static int sdma_v5_0_ring_preempt_ib(struct amdgpu_ring *ring)
{
	int i, r = 0;
	struct amdgpu_device *adev = ring->adev;
	u32 index = 0;
	u64 sdma_gfx_preempt;

	amdgpu_sdma_get_index_from_ring(ring, &index);
	if (index == 0)
		sdma_gfx_preempt = mmSDMA0_GFX_PREEMPT;
	else
		sdma_gfx_preempt = mmSDMA1_GFX_PREEMPT;

	/* assert preemption condition */
	amdgpu_ring_set_preempt_cond_exec(ring, false);

	/* emit the trailing fence */
	ring->trail_seq += 1;
	amdgpu_ring_alloc(ring, 10);
	sdma_v5_0_ring_emit_fence(ring, ring->trail_fence_gpu_addr,
				  ring->trail_seq, 0);
	amdgpu_ring_commit(ring);

	/* assert IB preemption */
	WREG32(sdma_gfx_preempt, 1);

	/* poll the trailing fence */
	for (i = 0; i < adev->usec_timeout; i++) {
		if (ring->trail_seq ==
		    le32_to_cpu(*(ring->trail_fence_cpu_addr)))
			break;
		udelay(1);
	}

	if (i >= adev->usec_timeout) {
		r = -EINVAL;
		DRM_ERROR("ring %d failed to be preempted\n", ring->idx);
	}

	/* deassert IB preemption */
	WREG32(sdma_gfx_preempt, 0);

	/* deassert the preemption condition */
	amdgpu_ring_set_preempt_cond_exec(ring, true);
	return r;
}

static int sdma_v5_0_set_trap_irq_state(struct amdgpu_device *adev,
					struct amdgpu_irq_src *source,
					unsigned type,
					enum amdgpu_interrupt_state state)
{
	u32 sdma_cntl;

	u32 reg_offset = (type == AMDGPU_SDMA_IRQ_INSTANCE0) ?
		sdma_v5_0_get_reg_offset(adev, 0, mmSDMA0_CNTL) :
		sdma_v5_0_get_reg_offset(adev, 1, mmSDMA0_CNTL);

	sdma_cntl = RREG32(reg_offset);
	sdma_cntl = REG_SET_FIELD(sdma_cntl, SDMA0_CNTL, TRAP_ENABLE,
		       state == AMDGPU_IRQ_STATE_ENABLE ? 1 : 0);
	WREG32(reg_offset, sdma_cntl);

	return 0;
}

static int sdma_v5_0_process_trap_irq(struct amdgpu_device *adev,
				      struct amdgpu_irq_src *source,
				      struct amdgpu_iv_entry *entry)
{
	DRM_DEBUG("IH: SDMA trap\n");
	switch (entry->client_id) {
	case SOC15_IH_CLIENTID_SDMA0:
		switch (entry->ring_id) {
		case 0:
			amdgpu_fence_process(&adev->sdma.instance[0].ring);
			break;
		case 1:
			/* XXX compute */
			break;
		case 2:
			/* XXX compute */
			break;
		case 3:
			/* XXX page queue*/
			break;
		}
		break;
	case SOC15_IH_CLIENTID_SDMA1:
		switch (entry->ring_id) {
		case 0:
			amdgpu_fence_process(&adev->sdma.instance[1].ring);
			break;
		case 1:
			/* XXX compute */
			break;
		case 2:
			/* XXX compute */
			break;
		case 3:
			/* XXX page queue*/
			break;
		}
		break;
	}
	return 0;
}

static int sdma_v5_0_process_illegal_inst_irq(struct amdgpu_device *adev,
					      struct amdgpu_irq_src *source,
					      struct amdgpu_iv_entry *entry)
{
	return 0;
}

static void sdma_v5_0_update_medium_grain_clock_gating(struct amdgpu_device *adev,
						       bool enable)
{
	uint32_t data, def;
	int i;

	for (i = 0; i < adev->sdma.num_instances; i++) {
		if (enable && (adev->cg_flags & AMD_CG_SUPPORT_SDMA_MGCG)) {
			/* Enable sdma clock gating */
			def = data = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_CLK_CTRL));
			data &= ~(SDMA0_CLK_CTRL__SOFT_OVERRIDE7_MASK |
				  SDMA0_CLK_CTRL__SOFT_OVERRIDE6_MASK |
				  SDMA0_CLK_CTRL__SOFT_OVERRIDE5_MASK |
				  SDMA0_CLK_CTRL__SOFT_OVERRIDE4_MASK |
				  SDMA0_CLK_CTRL__SOFT_OVERRIDE3_MASK |
				  SDMA0_CLK_CTRL__SOFT_OVERRIDE2_MASK |
				  SDMA0_CLK_CTRL__SOFT_OVERRIDE1_MASK |
				  SDMA0_CLK_CTRL__SOFT_OVERRIDE0_MASK);
			if (def != data)
				WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_CLK_CTRL), data);
		} else {
			/* Disable sdma clock gating */
			def = data = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_CLK_CTRL));
			data |= (SDMA0_CLK_CTRL__SOFT_OVERRIDE7_MASK |
				 SDMA0_CLK_CTRL__SOFT_OVERRIDE6_MASK |
				 SDMA0_CLK_CTRL__SOFT_OVERRIDE5_MASK |
				 SDMA0_CLK_CTRL__SOFT_OVERRIDE4_MASK |
				 SDMA0_CLK_CTRL__SOFT_OVERRIDE3_MASK |
				 SDMA0_CLK_CTRL__SOFT_OVERRIDE2_MASK |
				 SDMA0_CLK_CTRL__SOFT_OVERRIDE1_MASK |
				 SDMA0_CLK_CTRL__SOFT_OVERRIDE0_MASK);
			if (def != data)
				WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_CLK_CTRL), data);
		}
	}
}

static void sdma_v5_0_update_medium_grain_light_sleep(struct amdgpu_device *adev,
						      bool enable)
{
	uint32_t data, def;
	int i;

	for (i = 0; i < adev->sdma.num_instances; i++) {
		if (enable && (adev->cg_flags & AMD_CG_SUPPORT_SDMA_LS)) {
			/* Enable sdma mem light sleep */
			def = data = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_POWER_CNTL));
			data |= SDMA0_POWER_CNTL__MEM_POWER_OVERRIDE_MASK;
			if (def != data)
				WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_POWER_CNTL), data);

		} else {
			/* Disable sdma mem light sleep */
			def = data = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_POWER_CNTL));
			data &= ~SDMA0_POWER_CNTL__MEM_POWER_OVERRIDE_MASK;
			if (def != data)
				WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_POWER_CNTL), data);

		}
	}
}

static int sdma_v5_0_set_clockgating_state(void *handle,
					   enum amd_clockgating_state state)
{
	struct amdgpu_device *adev = (struct amdgpu_device *)handle;

	if (amdgpu_sriov_vf(adev))
		return 0;

	switch (adev->asic_type) {
	case CHIP_NAVI10:
	case CHIP_NAVI14:
	case CHIP_NAVI12:
		sdma_v5_0_update_medium_grain_clock_gating(adev,
				state == AMD_CG_STATE_GATE);
		sdma_v5_0_update_medium_grain_light_sleep(adev,
				state == AMD_CG_STATE_GATE);
		break;
	default:
		break;
	}

	return 0;
}

static int sdma_v5_0_set_powergating_state(void *handle,
					  enum amd_powergating_state state)
{
	return 0;
}

static void sdma_v5_0_get_clockgating_state(void *handle, u32 *flags)
{
	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
	int data;

	if (amdgpu_sriov_vf(adev))
		*flags = 0;

	/* AMD_CG_SUPPORT_SDMA_MGCG */
	data = RREG32(sdma_v5_0_get_reg_offset(adev, 0, mmSDMA0_CLK_CTRL));
	if (!(data & SDMA0_CLK_CTRL__SOFT_OVERRIDE7_MASK))
		*flags |= AMD_CG_SUPPORT_SDMA_MGCG;

	/* AMD_CG_SUPPORT_SDMA_LS */
	data = RREG32(sdma_v5_0_get_reg_offset(adev, 0, mmSDMA0_POWER_CNTL));
	if (data & SDMA0_POWER_CNTL__MEM_POWER_OVERRIDE_MASK)
		*flags |= AMD_CG_SUPPORT_SDMA_LS;
}

const struct amd_ip_funcs sdma_v5_0_ip_funcs = {
	.name = "sdma_v5_0",
	.early_init = sdma_v5_0_early_init,
	.late_init = NULL,
	.sw_init = sdma_v5_0_sw_init,
	.sw_fini = sdma_v5_0_sw_fini,
	.hw_init = sdma_v5_0_hw_init,
	.hw_fini = sdma_v5_0_hw_fini,
	.suspend = sdma_v5_0_suspend,
	.resume = sdma_v5_0_resume,
	.is_idle = sdma_v5_0_is_idle,
	.wait_for_idle = sdma_v5_0_wait_for_idle,
	.soft_reset = sdma_v5_0_soft_reset,
	.set_clockgating_state = sdma_v5_0_set_clockgating_state,
	.set_powergating_state = sdma_v5_0_set_powergating_state,
	.get_clockgating_state = sdma_v5_0_get_clockgating_state,
};

static const struct amdgpu_ring_funcs sdma_v5_0_ring_funcs = {
	.type = AMDGPU_RING_TYPE_SDMA,
	.align_mask = 0xf,
	.nop = SDMA_PKT_NOP_HEADER_OP(SDMA_OP_NOP),
	.support_64bit_ptrs = true,
	.vmhub = AMDGPU_GFXHUB_0,
	.get_rptr = sdma_v5_0_ring_get_rptr,
	.get_wptr = sdma_v5_0_ring_get_wptr,
	.set_wptr = sdma_v5_0_ring_set_wptr,
	.emit_frame_size =
		5 + /* sdma_v5_0_ring_init_cond_exec */
		6 + /* sdma_v5_0_ring_emit_hdp_flush */
		3 + /* hdp_invalidate */
		6 + /* sdma_v5_0_ring_emit_pipeline_sync */
		/* sdma_v5_0_ring_emit_vm_flush */
		SOC15_FLUSH_GPU_TLB_NUM_WREG * 3 +
		SOC15_FLUSH_GPU_TLB_NUM_REG_WAIT * 6 * 2 +
		10 + 10 + 10, /* sdma_v5_0_ring_emit_fence x3 for user fence, vm fence */
	.emit_ib_size = 5 + 7 + 6, /* sdma_v5_0_ring_emit_ib */
	.emit_ib = sdma_v5_0_ring_emit_ib,
	.emit_fence = sdma_v5_0_ring_emit_fence,
	.emit_pipeline_sync = sdma_v5_0_ring_emit_pipeline_sync,
	.emit_vm_flush = sdma_v5_0_ring_emit_vm_flush,
	.emit_hdp_flush = sdma_v5_0_ring_emit_hdp_flush,
	.test_ring = sdma_v5_0_ring_test_ring,
	.test_ib = sdma_v5_0_ring_test_ib,
	.insert_nop = sdma_v5_0_ring_insert_nop,
	.pad_ib = sdma_v5_0_ring_pad_ib,
	.emit_wreg = sdma_v5_0_ring_emit_wreg,
	.emit_reg_wait = sdma_v5_0_ring_emit_reg_wait,
	.emit_reg_write_reg_wait = sdma_v5_0_ring_emit_reg_write_reg_wait,
	.init_cond_exec = sdma_v5_0_ring_init_cond_exec,
	.patch_cond_exec = sdma_v5_0_ring_patch_cond_exec,
	.preempt_ib = sdma_v5_0_ring_preempt_ib,
};

static void sdma_v5_0_set_ring_funcs(struct amdgpu_device *adev)
{
	int i;

	for (i = 0; i < adev->sdma.num_instances; i++) {
		adev->sdma.instance[i].ring.funcs = &sdma_v5_0_ring_funcs;
		adev->sdma.instance[i].ring.me = i;
	}
}

static const struct amdgpu_irq_src_funcs sdma_v5_0_trap_irq_funcs = {
	.set = sdma_v5_0_set_trap_irq_state,
	.process = sdma_v5_0_process_trap_irq,
};

static const struct amdgpu_irq_src_funcs sdma_v5_0_illegal_inst_irq_funcs = {
	.process = sdma_v5_0_process_illegal_inst_irq,
};

static void sdma_v5_0_set_irq_funcs(struct amdgpu_device *adev)
{
	adev->sdma.trap_irq.num_types = AMDGPU_SDMA_IRQ_INSTANCE0 +
					adev->sdma.num_instances;
	adev->sdma.trap_irq.funcs = &sdma_v5_0_trap_irq_funcs;
	adev->sdma.illegal_inst_irq.funcs = &sdma_v5_0_illegal_inst_irq_funcs;
}

/**
 * sdma_v5_0_emit_copy_buffer - copy buffer using the sDMA engine
 *
 * @ring: amdgpu_ring structure holding ring information
 * @src_offset: src GPU address
 * @dst_offset: dst GPU address
 * @byte_count: number of bytes to xfer
 *
 * Copy GPU buffers using the DMA engine (NAVI10).
 * Used by the amdgpu ttm implementation to move pages if
 * registered as the asic copy callback.
 */
static void sdma_v5_0_emit_copy_buffer(struct amdgpu_ib *ib,
				       uint64_t src_offset,
				       uint64_t dst_offset,
				       uint32_t byte_count)
{
	ib->ptr[ib->length_dw++] = SDMA_PKT_HEADER_OP(SDMA_OP_COPY) |
		SDMA_PKT_HEADER_SUB_OP(SDMA_SUBOP_COPY_LINEAR);
	ib->ptr[ib->length_dw++] = byte_count - 1;
	ib->ptr[ib->length_dw++] = 0; /* src/dst endian swap */
	ib->ptr[ib->length_dw++] = lower_32_bits(src_offset);
	ib->ptr[ib->length_dw++] = upper_32_bits(src_offset);
	ib->ptr[ib->length_dw++] = lower_32_bits(dst_offset);
	ib->ptr[ib->length_dw++] = upper_32_bits(dst_offset);
}

/**
 * sdma_v5_0_emit_fill_buffer - fill buffer using the sDMA engine
 *
 * @ring: amdgpu_ring structure holding ring information
 * @src_data: value to write to buffer
 * @dst_offset: dst GPU address
 * @byte_count: number of bytes to xfer
 *
 * Fill GPU buffers using the DMA engine (NAVI10).
 */
static void sdma_v5_0_emit_fill_buffer(struct amdgpu_ib *ib,
				       uint32_t src_data,
				       uint64_t dst_offset,
				       uint32_t byte_count)
{
	ib->ptr[ib->length_dw++] = SDMA_PKT_HEADER_OP(SDMA_OP_CONST_FILL);
	ib->ptr[ib->length_dw++] = lower_32_bits(dst_offset);
	ib->ptr[ib->length_dw++] = upper_32_bits(dst_offset);
	ib->ptr[ib->length_dw++] = src_data;
	ib->ptr[ib->length_dw++] = byte_count - 1;
}

static const struct amdgpu_buffer_funcs sdma_v5_0_buffer_funcs = {
	.copy_max_bytes = 0x400000,
	.copy_num_dw = 7,
	.emit_copy_buffer = sdma_v5_0_emit_copy_buffer,

	.fill_max_bytes = 0x400000,
	.fill_num_dw = 5,
	.emit_fill_buffer = sdma_v5_0_emit_fill_buffer,
};

static void sdma_v5_0_set_buffer_funcs(struct amdgpu_device *adev)
{
	if (adev->mman.buffer_funcs == NULL) {
		adev->mman.buffer_funcs = &sdma_v5_0_buffer_funcs;
		adev->mman.buffer_funcs_ring = &adev->sdma.instance[0].ring;
	}
}

static const struct amdgpu_vm_pte_funcs sdma_v5_0_vm_pte_funcs = {
	.copy_pte_num_dw = 7,
	.copy_pte = sdma_v5_0_vm_copy_pte,
	.write_pte = sdma_v5_0_vm_write_pte,
	.set_pte_pde = sdma_v5_0_vm_set_pte_pde,
};

static void sdma_v5_0_set_vm_pte_funcs(struct amdgpu_device *adev)
{
	unsigned i;

	if (adev->vm_manager.vm_pte_funcs == NULL) {
		adev->vm_manager.vm_pte_funcs = &sdma_v5_0_vm_pte_funcs;
		for (i = 0; i < adev->sdma.num_instances; i++) {
			adev->vm_manager.vm_pte_scheds[i] =
				&adev->sdma.instance[i].ring.sched;
		}
		adev->vm_manager.vm_pte_num_scheds = adev->sdma.num_instances;
	}
}

const struct amdgpu_ip_block_version sdma_v5_0_ip_block = {
	.type = AMD_IP_BLOCK_TYPE_SDMA,
	.major = 5,
	.minor = 0,
	.rev = 0,
	.funcs = &sdma_v5_0_ip_funcs,
};