summaryrefslogtreecommitdiff
path: root/drivers/gpu/drm/vc4/vc4_plane.c
blob: 84fd0a31b7017070f4d4b464584b6fb214086154 (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
// SPDX-License-Identifier: GPL-2.0-only
/*
 * Copyright (C) 2015 Broadcom
 */

/**
 * DOC: VC4 plane module
 *
 * Each DRM plane is a layer of pixels being scanned out by the HVS.
 *
 * At atomic modeset check time, we compute the HVS display element
 * state that would be necessary for displaying the plane (giving us a
 * chance to figure out if a plane configuration is invalid), then at
 * atomic flush time the CRTC will ask us to write our element state
 * into the region of the HVS that it has allocated for us.
 */

#include <drm/drm_atomic.h>
#include <drm/drm_atomic_helper.h>
#include <drm/drm_fb_cma_helper.h>
#include <drm/drm_plane_helper.h>
#include <drm/drm_atomic_uapi.h>

#include "uapi/drm/vc4_drm.h"
#include "vc4_drv.h"
#include "vc4_regs.h"

static const struct hvs_format {
	u32 drm; /* DRM_FORMAT_* */
	u32 hvs; /* HVS_FORMAT_* */
	u32 pixel_order;
} hvs_formats[] = {
	{
		.drm = DRM_FORMAT_XRGB8888, .hvs = HVS_PIXEL_FORMAT_RGBA8888,
		.pixel_order = HVS_PIXEL_ORDER_ABGR,
	},
	{
		.drm = DRM_FORMAT_ARGB8888, .hvs = HVS_PIXEL_FORMAT_RGBA8888,
		.pixel_order = HVS_PIXEL_ORDER_ABGR,
	},
	{
		.drm = DRM_FORMAT_ABGR8888, .hvs = HVS_PIXEL_FORMAT_RGBA8888,
		.pixel_order = HVS_PIXEL_ORDER_ARGB,
	},
	{
		.drm = DRM_FORMAT_XBGR8888, .hvs = HVS_PIXEL_FORMAT_RGBA8888,
		.pixel_order = HVS_PIXEL_ORDER_ARGB,
	},
	{
		.drm = DRM_FORMAT_RGB565, .hvs = HVS_PIXEL_FORMAT_RGB565,
		.pixel_order = HVS_PIXEL_ORDER_XRGB,
	},
	{
		.drm = DRM_FORMAT_BGR565, .hvs = HVS_PIXEL_FORMAT_RGB565,
		.pixel_order = HVS_PIXEL_ORDER_XBGR,
	},
	{
		.drm = DRM_FORMAT_ARGB1555, .hvs = HVS_PIXEL_FORMAT_RGBA5551,
		.pixel_order = HVS_PIXEL_ORDER_ABGR,
	},
	{
		.drm = DRM_FORMAT_XRGB1555, .hvs = HVS_PIXEL_FORMAT_RGBA5551,
		.pixel_order = HVS_PIXEL_ORDER_ABGR,
	},
	{
		.drm = DRM_FORMAT_RGB888, .hvs = HVS_PIXEL_FORMAT_RGB888,
		.pixel_order = HVS_PIXEL_ORDER_XRGB,
	},
	{
		.drm = DRM_FORMAT_BGR888, .hvs = HVS_PIXEL_FORMAT_RGB888,
		.pixel_order = HVS_PIXEL_ORDER_XBGR,
	},
	{
		.drm = DRM_FORMAT_YUV422,
		.hvs = HVS_PIXEL_FORMAT_YCBCR_YUV422_3PLANE,
		.pixel_order = HVS_PIXEL_ORDER_XYCBCR,
	},
	{
		.drm = DRM_FORMAT_YVU422,
		.hvs = HVS_PIXEL_FORMAT_YCBCR_YUV422_3PLANE,
		.pixel_order = HVS_PIXEL_ORDER_XYCRCB,
	},
	{
		.drm = DRM_FORMAT_YUV420,
		.hvs = HVS_PIXEL_FORMAT_YCBCR_YUV420_3PLANE,
		.pixel_order = HVS_PIXEL_ORDER_XYCBCR,
	},
	{
		.drm = DRM_FORMAT_YVU420,
		.hvs = HVS_PIXEL_FORMAT_YCBCR_YUV420_3PLANE,
		.pixel_order = HVS_PIXEL_ORDER_XYCRCB,
	},
	{
		.drm = DRM_FORMAT_NV12,
		.hvs = HVS_PIXEL_FORMAT_YCBCR_YUV420_2PLANE,
		.pixel_order = HVS_PIXEL_ORDER_XYCBCR,
	},
	{
		.drm = DRM_FORMAT_NV21,
		.hvs = HVS_PIXEL_FORMAT_YCBCR_YUV420_2PLANE,
		.pixel_order = HVS_PIXEL_ORDER_XYCRCB,
	},
	{
		.drm = DRM_FORMAT_NV16,
		.hvs = HVS_PIXEL_FORMAT_YCBCR_YUV422_2PLANE,
		.pixel_order = HVS_PIXEL_ORDER_XYCBCR,
	},
	{
		.drm = DRM_FORMAT_NV61,
		.hvs = HVS_PIXEL_FORMAT_YCBCR_YUV422_2PLANE,
		.pixel_order = HVS_PIXEL_ORDER_XYCRCB,
	},
};

static const struct hvs_format *vc4_get_hvs_format(u32 drm_format)
{
	unsigned i;

	for (i = 0; i < ARRAY_SIZE(hvs_formats); i++) {
		if (hvs_formats[i].drm == drm_format)
			return &hvs_formats[i];
	}

	return NULL;
}

static enum vc4_scaling_mode vc4_get_scaling_mode(u32 src, u32 dst)
{
	if (dst == src)
		return VC4_SCALING_NONE;
	if (3 * dst >= 2 * src)
		return VC4_SCALING_PPF;
	else
		return VC4_SCALING_TPZ;
}

static bool plane_enabled(struct drm_plane_state *state)
{
	return state->fb && state->crtc;
}

static struct drm_plane_state *vc4_plane_duplicate_state(struct drm_plane *plane)
{
	struct vc4_plane_state *vc4_state;

	if (WARN_ON(!plane->state))
		return NULL;

	vc4_state = kmemdup(plane->state, sizeof(*vc4_state), GFP_KERNEL);
	if (!vc4_state)
		return NULL;

	memset(&vc4_state->lbm, 0, sizeof(vc4_state->lbm));
	vc4_state->dlist_initialized = 0;

	__drm_atomic_helper_plane_duplicate_state(plane, &vc4_state->base);

	if (vc4_state->dlist) {
		vc4_state->dlist = kmemdup(vc4_state->dlist,
					   vc4_state->dlist_count * 4,
					   GFP_KERNEL);
		if (!vc4_state->dlist) {
			kfree(vc4_state);
			return NULL;
		}
		vc4_state->dlist_size = vc4_state->dlist_count;
	}

	return &vc4_state->base;
}

static void vc4_plane_destroy_state(struct drm_plane *plane,
				    struct drm_plane_state *state)
{
	struct vc4_dev *vc4 = to_vc4_dev(plane->dev);
	struct vc4_plane_state *vc4_state = to_vc4_plane_state(state);

	if (vc4_state->lbm.allocated) {
		unsigned long irqflags;

		spin_lock_irqsave(&vc4->hvs->mm_lock, irqflags);
		drm_mm_remove_node(&vc4_state->lbm);
		spin_unlock_irqrestore(&vc4->hvs->mm_lock, irqflags);
	}

	kfree(vc4_state->dlist);
	__drm_atomic_helper_plane_destroy_state(&vc4_state->base);
	kfree(state);
}

/* Called during init to allocate the plane's atomic state. */
static void vc4_plane_reset(struct drm_plane *plane)
{
	struct vc4_plane_state *vc4_state;

	WARN_ON(plane->state);

	vc4_state = kzalloc(sizeof(*vc4_state), GFP_KERNEL);
	if (!vc4_state)
		return;

	__drm_atomic_helper_plane_reset(plane, &vc4_state->base);
}

static void vc4_dlist_write(struct vc4_plane_state *vc4_state, u32 val)
{
	if (vc4_state->dlist_count == vc4_state->dlist_size) {
		u32 new_size = max(4u, vc4_state->dlist_count * 2);
		u32 *new_dlist = kmalloc_array(new_size, 4, GFP_KERNEL);

		if (!new_dlist)
			return;
		memcpy(new_dlist, vc4_state->dlist, vc4_state->dlist_count * 4);

		kfree(vc4_state->dlist);
		vc4_state->dlist = new_dlist;
		vc4_state->dlist_size = new_size;
	}

	vc4_state->dlist[vc4_state->dlist_count++] = val;
}

/* Returns the scl0/scl1 field based on whether the dimensions need to
 * be up/down/non-scaled.
 *
 * This is a replication of a table from the spec.
 */
static u32 vc4_get_scl_field(struct drm_plane_state *state, int plane)
{
	struct vc4_plane_state *vc4_state = to_vc4_plane_state(state);

	switch (vc4_state->x_scaling[plane] << 2 | vc4_state->y_scaling[plane]) {
	case VC4_SCALING_PPF << 2 | VC4_SCALING_PPF:
		return SCALER_CTL0_SCL_H_PPF_V_PPF;
	case VC4_SCALING_TPZ << 2 | VC4_SCALING_PPF:
		return SCALER_CTL0_SCL_H_TPZ_V_PPF;
	case VC4_SCALING_PPF << 2 | VC4_SCALING_TPZ:
		return SCALER_CTL0_SCL_H_PPF_V_TPZ;
	case VC4_SCALING_TPZ << 2 | VC4_SCALING_TPZ:
		return SCALER_CTL0_SCL_H_TPZ_V_TPZ;
	case VC4_SCALING_PPF << 2 | VC4_SCALING_NONE:
		return SCALER_CTL0_SCL_H_PPF_V_NONE;
	case VC4_SCALING_NONE << 2 | VC4_SCALING_PPF:
		return SCALER_CTL0_SCL_H_NONE_V_PPF;
	case VC4_SCALING_NONE << 2 | VC4_SCALING_TPZ:
		return SCALER_CTL0_SCL_H_NONE_V_TPZ;
	case VC4_SCALING_TPZ << 2 | VC4_SCALING_NONE:
		return SCALER_CTL0_SCL_H_TPZ_V_NONE;
	default:
	case VC4_SCALING_NONE << 2 | VC4_SCALING_NONE:
		/* The unity case is independently handled by
		 * SCALER_CTL0_UNITY.
		 */
		return 0;
	}
}

static int vc4_plane_margins_adj(struct drm_plane_state *pstate)
{
	struct vc4_plane_state *vc4_pstate = to_vc4_plane_state(pstate);
	unsigned int left, right, top, bottom, adjhdisplay, adjvdisplay;
	struct drm_crtc_state *crtc_state;

	crtc_state = drm_atomic_get_new_crtc_state(pstate->state,
						   pstate->crtc);

	vc4_crtc_get_margins(crtc_state, &left, &right, &top, &bottom);
	if (!left && !right && !top && !bottom)
		return 0;

	if (left + right >= crtc_state->mode.hdisplay ||
	    top + bottom >= crtc_state->mode.vdisplay)
		return -EINVAL;

	adjhdisplay = crtc_state->mode.hdisplay - (left + right);
	vc4_pstate->crtc_x = DIV_ROUND_CLOSEST(vc4_pstate->crtc_x *
					       adjhdisplay,
					       crtc_state->mode.hdisplay);
	vc4_pstate->crtc_x += left;
	if (vc4_pstate->crtc_x > crtc_state->mode.hdisplay - left)
		vc4_pstate->crtc_x = crtc_state->mode.hdisplay - left;

	adjvdisplay = crtc_state->mode.vdisplay - (top + bottom);
	vc4_pstate->crtc_y = DIV_ROUND_CLOSEST(vc4_pstate->crtc_y *
					       adjvdisplay,
					       crtc_state->mode.vdisplay);
	vc4_pstate->crtc_y += top;
	if (vc4_pstate->crtc_y > crtc_state->mode.vdisplay - top)
		vc4_pstate->crtc_y = crtc_state->mode.vdisplay - top;

	vc4_pstate->crtc_w = DIV_ROUND_CLOSEST(vc4_pstate->crtc_w *
					       adjhdisplay,
					       crtc_state->mode.hdisplay);
	vc4_pstate->crtc_h = DIV_ROUND_CLOSEST(vc4_pstate->crtc_h *
					       adjvdisplay,
					       crtc_state->mode.vdisplay);

	if (!vc4_pstate->crtc_w || !vc4_pstate->crtc_h)
		return -EINVAL;

	return 0;
}

static int vc4_plane_setup_clipping_and_scaling(struct drm_plane_state *state)
{
	struct vc4_plane_state *vc4_state = to_vc4_plane_state(state);
	struct drm_framebuffer *fb = state->fb;
	struct drm_gem_cma_object *bo = drm_fb_cma_get_gem_obj(fb, 0);
	u32 subpixel_src_mask = (1 << 16) - 1;
	u32 format = fb->format->format;
	int num_planes = fb->format->num_planes;
	struct drm_crtc_state *crtc_state;
	u32 h_subsample, v_subsample;
	int i, ret;

	crtc_state = drm_atomic_get_existing_crtc_state(state->state,
							state->crtc);
	if (!crtc_state) {
		DRM_DEBUG_KMS("Invalid crtc state\n");
		return -EINVAL;
	}

	ret = drm_atomic_helper_check_plane_state(state, crtc_state, 1,
						  INT_MAX, true, true);
	if (ret)
		return ret;

	h_subsample = drm_format_horz_chroma_subsampling(format);
	v_subsample = drm_format_vert_chroma_subsampling(format);

	for (i = 0; i < num_planes; i++)
		vc4_state->offsets[i] = bo->paddr + fb->offsets[i];

	/* We don't support subpixel source positioning for scaling. */
	if ((state->src.x1 & subpixel_src_mask) ||
	    (state->src.x2 & subpixel_src_mask) ||
	    (state->src.y1 & subpixel_src_mask) ||
	    (state->src.y2 & subpixel_src_mask)) {
		return -EINVAL;
	}

	vc4_state->src_x = state->src.x1 >> 16;
	vc4_state->src_y = state->src.y1 >> 16;
	vc4_state->src_w[0] = (state->src.x2 - state->src.x1) >> 16;
	vc4_state->src_h[0] = (state->src.y2 - state->src.y1) >> 16;

	vc4_state->crtc_x = state->dst.x1;
	vc4_state->crtc_y = state->dst.y1;
	vc4_state->crtc_w = state->dst.x2 - state->dst.x1;
	vc4_state->crtc_h = state->dst.y2 - state->dst.y1;

	ret = vc4_plane_margins_adj(state);
	if (ret)
		return ret;

	vc4_state->x_scaling[0] = vc4_get_scaling_mode(vc4_state->src_w[0],
						       vc4_state->crtc_w);
	vc4_state->y_scaling[0] = vc4_get_scaling_mode(vc4_state->src_h[0],
						       vc4_state->crtc_h);

	vc4_state->is_unity = (vc4_state->x_scaling[0] == VC4_SCALING_NONE &&
			       vc4_state->y_scaling[0] == VC4_SCALING_NONE);

	if (num_planes > 1) {
		vc4_state->is_yuv = true;

		vc4_state->src_w[1] = vc4_state->src_w[0] / h_subsample;
		vc4_state->src_h[1] = vc4_state->src_h[0] / v_subsample;

		vc4_state->x_scaling[1] =
			vc4_get_scaling_mode(vc4_state->src_w[1],
					     vc4_state->crtc_w);
		vc4_state->y_scaling[1] =
			vc4_get_scaling_mode(vc4_state->src_h[1],
					     vc4_state->crtc_h);

		/* YUV conversion requires that horizontal scaling be enabled
		 * on the UV plane even if vc4_get_scaling_mode() returned
		 * VC4_SCALING_NONE (which can happen when the down-scaling
		 * ratio is 0.5). Let's force it to VC4_SCALING_PPF in this
		 * case.
		 */
		if (vc4_state->x_scaling[1] == VC4_SCALING_NONE)
			vc4_state->x_scaling[1] = VC4_SCALING_PPF;
	} else {
		vc4_state->is_yuv = false;
		vc4_state->x_scaling[1] = VC4_SCALING_NONE;
		vc4_state->y_scaling[1] = VC4_SCALING_NONE;
	}

	return 0;
}

static void vc4_write_tpz(struct vc4_plane_state *vc4_state, u32 src, u32 dst)
{
	u32 scale, recip;

	scale = (1 << 16) * src / dst;

	/* The specs note that while the reciprocal would be defined
	 * as (1<<32)/scale, ~0 is close enough.
	 */
	recip = ~0 / scale;

	vc4_dlist_write(vc4_state,
			VC4_SET_FIELD(scale, SCALER_TPZ0_SCALE) |
			VC4_SET_FIELD(0, SCALER_TPZ0_IPHASE));
	vc4_dlist_write(vc4_state,
			VC4_SET_FIELD(recip, SCALER_TPZ1_RECIP));
}

static void vc4_write_ppf(struct vc4_plane_state *vc4_state, u32 src, u32 dst)
{
	u32 scale = (1 << 16) * src / dst;

	vc4_dlist_write(vc4_state,
			SCALER_PPF_AGC |
			VC4_SET_FIELD(scale, SCALER_PPF_SCALE) |
			VC4_SET_FIELD(0, SCALER_PPF_IPHASE));
}

static u32 vc4_lbm_size(struct drm_plane_state *state)
{
	struct vc4_plane_state *vc4_state = to_vc4_plane_state(state);
	/* This is the worst case number.  One of the two sizes will
	 * be used depending on the scaling configuration.
	 */
	u32 pix_per_line = max(vc4_state->src_w[0], (u32)vc4_state->crtc_w);
	u32 lbm;

	/* LBM is not needed when there's no vertical scaling. */
	if (vc4_state->y_scaling[0] == VC4_SCALING_NONE &&
	    vc4_state->y_scaling[1] == VC4_SCALING_NONE)
		return 0;

	if (!vc4_state->is_yuv) {
		if (vc4_state->y_scaling[0] == VC4_SCALING_TPZ)
			lbm = pix_per_line * 8;
		else {
			/* In special cases, this multiplier might be 12. */
			lbm = pix_per_line * 16;
		}
	} else {
		/* There are cases for this going down to a multiplier
		 * of 2, but according to the firmware source, the
		 * table in the docs is somewhat wrong.
		 */
		lbm = pix_per_line * 16;
	}

	lbm = roundup(lbm, 32);

	return lbm;
}

static void vc4_write_scaling_parameters(struct drm_plane_state *state,
					 int channel)
{
	struct vc4_plane_state *vc4_state = to_vc4_plane_state(state);

	/* Ch0 H-PPF Word 0: Scaling Parameters */
	if (vc4_state->x_scaling[channel] == VC4_SCALING_PPF) {
		vc4_write_ppf(vc4_state,
			      vc4_state->src_w[channel], vc4_state->crtc_w);
	}

	/* Ch0 V-PPF Words 0-1: Scaling Parameters, Context */
	if (vc4_state->y_scaling[channel] == VC4_SCALING_PPF) {
		vc4_write_ppf(vc4_state,
			      vc4_state->src_h[channel], vc4_state->crtc_h);
		vc4_dlist_write(vc4_state, 0xc0c0c0c0);
	}

	/* Ch0 H-TPZ Words 0-1: Scaling Parameters, Recip */
	if (vc4_state->x_scaling[channel] == VC4_SCALING_TPZ) {
		vc4_write_tpz(vc4_state,
			      vc4_state->src_w[channel], vc4_state->crtc_w);
	}

	/* Ch0 V-TPZ Words 0-2: Scaling Parameters, Recip, Context */
	if (vc4_state->y_scaling[channel] == VC4_SCALING_TPZ) {
		vc4_write_tpz(vc4_state,
			      vc4_state->src_h[channel], vc4_state->crtc_h);
		vc4_dlist_write(vc4_state, 0xc0c0c0c0);
	}
}

static void vc4_plane_calc_load(struct drm_plane_state *state)
{
	unsigned int hvs_load_shift, vrefresh, i;
	struct drm_framebuffer *fb = state->fb;
	struct vc4_plane_state *vc4_state;
	struct drm_crtc_state *crtc_state;
	unsigned int vscale_factor;

	vc4_state = to_vc4_plane_state(state);
	crtc_state = drm_atomic_get_existing_crtc_state(state->state,
							state->crtc);
	vrefresh = drm_mode_vrefresh(&crtc_state->adjusted_mode);

	/* The HVS is able to process 2 pixels/cycle when scaling the source,
	 * 4 pixels/cycle otherwise.
	 * Alpha blending step seems to be pipelined and it's always operating
	 * at 4 pixels/cycle, so the limiting aspect here seems to be the
	 * scaler block.
	 * HVS load is expressed in clk-cycles/sec (AKA Hz).
	 */
	if (vc4_state->x_scaling[0] != VC4_SCALING_NONE ||
	    vc4_state->x_scaling[1] != VC4_SCALING_NONE ||
	    vc4_state->y_scaling[0] != VC4_SCALING_NONE ||
	    vc4_state->y_scaling[1] != VC4_SCALING_NONE)
		hvs_load_shift = 1;
	else
		hvs_load_shift = 2;

	vc4_state->membus_load = 0;
	vc4_state->hvs_load = 0;
	for (i = 0; i < fb->format->num_planes; i++) {
		/* Even if the bandwidth/plane required for a single frame is
		 *
		 * vc4_state->src_w[i] * vc4_state->src_h[i] * cpp * vrefresh
		 *
		 * when downscaling, we have to read more pixels per line in
		 * the time frame reserved for a single line, so the bandwidth
		 * demand can be punctually higher. To account for that, we
		 * calculate the down-scaling factor and multiply the plane
		 * load by this number. We're likely over-estimating the read
		 * demand, but that's better than under-estimating it.
		 */
		vscale_factor = DIV_ROUND_UP(vc4_state->src_h[i],
					     vc4_state->crtc_h);
		vc4_state->membus_load += vc4_state->src_w[i] *
					  vc4_state->src_h[i] * vscale_factor *
					  fb->format->cpp[i];
		vc4_state->hvs_load += vc4_state->crtc_h * vc4_state->crtc_w;
	}

	vc4_state->hvs_load *= vrefresh;
	vc4_state->hvs_load >>= hvs_load_shift;
	vc4_state->membus_load *= vrefresh;
}

static int vc4_plane_allocate_lbm(struct drm_plane_state *state)
{
	struct vc4_dev *vc4 = to_vc4_dev(state->plane->dev);
	struct vc4_plane_state *vc4_state = to_vc4_plane_state(state);
	unsigned long irqflags;
	u32 lbm_size;

	lbm_size = vc4_lbm_size(state);
	if (!lbm_size)
		return 0;

	if (WARN_ON(!vc4_state->lbm_offset))
		return -EINVAL;

	/* Allocate the LBM memory that the HVS will use for temporary
	 * storage due to our scaling/format conversion.
	 */
	if (!vc4_state->lbm.allocated) {
		int ret;

		spin_lock_irqsave(&vc4->hvs->mm_lock, irqflags);
		ret = drm_mm_insert_node_generic(&vc4->hvs->lbm_mm,
						 &vc4_state->lbm,
						 lbm_size, 32, 0, 0);
		spin_unlock_irqrestore(&vc4->hvs->mm_lock, irqflags);

		if (ret)
			return ret;
	} else {
		WARN_ON_ONCE(lbm_size != vc4_state->lbm.size);
	}

	vc4_state->dlist[vc4_state->lbm_offset] = vc4_state->lbm.start;

	return 0;
}

/* Writes out a full display list for an active plane to the plane's
 * private dlist state.
 */
static int vc4_plane_mode_set(struct drm_plane *plane,
			      struct drm_plane_state *state)
{
	struct vc4_dev *vc4 = to_vc4_dev(plane->dev);
	struct vc4_plane_state *vc4_state = to_vc4_plane_state(state);
	struct drm_framebuffer *fb = state->fb;
	u32 ctl0_offset = vc4_state->dlist_count;
	const struct hvs_format *format = vc4_get_hvs_format(fb->format->format);
	u64 base_format_mod = fourcc_mod_broadcom_mod(fb->modifier);
	int num_planes = drm_format_num_planes(format->drm);
	u32 h_subsample, v_subsample;
	bool mix_plane_alpha;
	bool covers_screen;
	u32 scl0, scl1, pitch0;
	u32 tiling, src_y;
	u32 hvs_format = format->hvs;
	unsigned int rotation;
	int ret, i;

	if (vc4_state->dlist_initialized)
		return 0;

	ret = vc4_plane_setup_clipping_and_scaling(state);
	if (ret)
		return ret;

	/* SCL1 is used for Cb/Cr scaling of planar formats.  For RGB
	 * and 4:4:4, scl1 should be set to scl0 so both channels of
	 * the scaler do the same thing.  For YUV, the Y plane needs
	 * to be put in channel 1 and Cb/Cr in channel 0, so we swap
	 * the scl fields here.
	 */
	if (num_planes == 1) {
		scl0 = vc4_get_scl_field(state, 0);
		scl1 = scl0;
	} else {
		scl0 = vc4_get_scl_field(state, 1);
		scl1 = vc4_get_scl_field(state, 0);
	}

	h_subsample = drm_format_horz_chroma_subsampling(format->drm);
	v_subsample = drm_format_vert_chroma_subsampling(format->drm);

	rotation = drm_rotation_simplify(state->rotation,
					 DRM_MODE_ROTATE_0 |
					 DRM_MODE_REFLECT_X |
					 DRM_MODE_REFLECT_Y);

	/* We must point to the last line when Y reflection is enabled. */
	src_y = vc4_state->src_y;
	if (rotation & DRM_MODE_REFLECT_Y)
		src_y += vc4_state->src_h[0] - 1;

	switch (base_format_mod) {
	case DRM_FORMAT_MOD_LINEAR:
		tiling = SCALER_CTL0_TILING_LINEAR;
		pitch0 = VC4_SET_FIELD(fb->pitches[0], SCALER_SRC_PITCH);

		/* Adjust the base pointer to the first pixel to be scanned
		 * out.
		 */
		for (i = 0; i < num_planes; i++) {
			vc4_state->offsets[i] += src_y /
						 (i ? v_subsample : 1) *
						 fb->pitches[i];

			vc4_state->offsets[i] += vc4_state->src_x /
						 (i ? h_subsample : 1) *
						 fb->format->cpp[i];
		}

		break;

	case DRM_FORMAT_MOD_BROADCOM_VC4_T_TILED: {
		u32 tile_size_shift = 12; /* T tiles are 4kb */
		/* Whole-tile offsets, mostly for setting the pitch. */
		u32 tile_w_shift = fb->format->cpp[0] == 2 ? 6 : 5;
		u32 tile_h_shift = 5; /* 16 and 32bpp are 32 pixels high */
		u32 tile_w_mask = (1 << tile_w_shift) - 1;
		/* The height mask on 32-bit-per-pixel tiles is 63, i.e. twice
		 * the height (in pixels) of a 4k tile.
		 */
		u32 tile_h_mask = (2 << tile_h_shift) - 1;
		/* For T-tiled, the FB pitch is "how many bytes from one row to
		 * the next, such that
		 *
		 *	pitch * tile_h == tile_size * tiles_per_row
		 */
		u32 tiles_w = fb->pitches[0] >> (tile_size_shift - tile_h_shift);
		u32 tiles_l = vc4_state->src_x >> tile_w_shift;
		u32 tiles_r = tiles_w - tiles_l;
		u32 tiles_t = src_y >> tile_h_shift;
		/* Intra-tile offsets, which modify the base address (the
		 * SCALER_PITCH0_TILE_Y_OFFSET tells HVS how to walk from that
		 * base address).
		 */
		u32 tile_y = (src_y >> 4) & 1;
		u32 subtile_y = (src_y >> 2) & 3;
		u32 utile_y = src_y & 3;
		u32 x_off = vc4_state->src_x & tile_w_mask;
		u32 y_off = src_y & tile_h_mask;

		/* When Y reflection is requested we must set the
		 * SCALER_PITCH0_TILE_LINE_DIR flag to tell HVS that all lines
		 * after the initial one should be fetched in descending order,
		 * which makes sense since we start from the last line and go
		 * backward.
		 * Don't know why we need y_off = max_y_off - y_off, but it's
		 * definitely required (I guess it's also related to the "going
		 * backward" situation).
		 */
		if (rotation & DRM_MODE_REFLECT_Y) {
			y_off = tile_h_mask - y_off;
			pitch0 = SCALER_PITCH0_TILE_LINE_DIR;
		} else {
			pitch0 = 0;
		}

		tiling = SCALER_CTL0_TILING_256B_OR_T;
		pitch0 |= (VC4_SET_FIELD(x_off, SCALER_PITCH0_SINK_PIX) |
			   VC4_SET_FIELD(y_off, SCALER_PITCH0_TILE_Y_OFFSET) |
			   VC4_SET_FIELD(tiles_l, SCALER_PITCH0_TILE_WIDTH_L) |
			   VC4_SET_FIELD(tiles_r, SCALER_PITCH0_TILE_WIDTH_R));
		vc4_state->offsets[0] += tiles_t * (tiles_w << tile_size_shift);
		vc4_state->offsets[0] += subtile_y << 8;
		vc4_state->offsets[0] += utile_y << 4;

		/* Rows of tiles alternate left-to-right and right-to-left. */
		if (tiles_t & 1) {
			pitch0 |= SCALER_PITCH0_TILE_INITIAL_LINE_DIR;
			vc4_state->offsets[0] += (tiles_w - tiles_l) <<
						 tile_size_shift;
			vc4_state->offsets[0] -= (1 + !tile_y) << 10;
		} else {
			vc4_state->offsets[0] += tiles_l << tile_size_shift;
			vc4_state->offsets[0] += tile_y << 10;
		}

		break;
	}

	case DRM_FORMAT_MOD_BROADCOM_SAND64:
	case DRM_FORMAT_MOD_BROADCOM_SAND128:
	case DRM_FORMAT_MOD_BROADCOM_SAND256: {
		uint32_t param = fourcc_mod_broadcom_param(fb->modifier);
		u32 tile_w, tile, x_off, pix_per_tile;

		hvs_format = HVS_PIXEL_FORMAT_H264;

		switch (base_format_mod) {
		case DRM_FORMAT_MOD_BROADCOM_SAND64:
			tiling = SCALER_CTL0_TILING_64B;
			tile_w = 64;
			break;
		case DRM_FORMAT_MOD_BROADCOM_SAND128:
			tiling = SCALER_CTL0_TILING_128B;
			tile_w = 128;
			break;
		case DRM_FORMAT_MOD_BROADCOM_SAND256:
			tiling = SCALER_CTL0_TILING_256B_OR_T;
			tile_w = 256;
			break;
		default:
			break;
		}

		if (param > SCALER_TILE_HEIGHT_MASK) {
			DRM_DEBUG_KMS("SAND height too large (%d)\n", param);
			return -EINVAL;
		}

		pix_per_tile = tile_w / fb->format->cpp[0];
		tile = vc4_state->src_x / pix_per_tile;
		x_off = vc4_state->src_x % pix_per_tile;

		/* Adjust the base pointer to the first pixel to be scanned
		 * out.
		 */
		for (i = 0; i < num_planes; i++) {
			vc4_state->offsets[i] += param * tile_w * tile;
			vc4_state->offsets[i] += src_y /
						 (i ? v_subsample : 1) *
						 tile_w;
			vc4_state->offsets[i] += x_off /
						 (i ? h_subsample : 1) *
						 fb->format->cpp[i];
		}

		pitch0 = VC4_SET_FIELD(param, SCALER_TILE_HEIGHT);
		break;
	}

	default:
		DRM_DEBUG_KMS("Unsupported FB tiling flag 0x%16llx",
			      (long long)fb->modifier);
		return -EINVAL;
	}

	/* Control word */
	vc4_dlist_write(vc4_state,
			SCALER_CTL0_VALID |
			(rotation & DRM_MODE_REFLECT_X ? SCALER_CTL0_HFLIP : 0) |
			(rotation & DRM_MODE_REFLECT_Y ? SCALER_CTL0_VFLIP : 0) |
			VC4_SET_FIELD(SCALER_CTL0_RGBA_EXPAND_ROUND, SCALER_CTL0_RGBA_EXPAND) |
			(format->pixel_order << SCALER_CTL0_ORDER_SHIFT) |
			(hvs_format << SCALER_CTL0_PIXEL_FORMAT_SHIFT) |
			VC4_SET_FIELD(tiling, SCALER_CTL0_TILING) |
			(vc4_state->is_unity ? SCALER_CTL0_UNITY : 0) |
			VC4_SET_FIELD(scl0, SCALER_CTL0_SCL0) |
			VC4_SET_FIELD(scl1, SCALER_CTL0_SCL1));

	/* Position Word 0: Image Positions and Alpha Value */
	vc4_state->pos0_offset = vc4_state->dlist_count;
	vc4_dlist_write(vc4_state,
			VC4_SET_FIELD(state->alpha >> 8, SCALER_POS0_FIXED_ALPHA) |
			VC4_SET_FIELD(vc4_state->crtc_x, SCALER_POS0_START_X) |
			VC4_SET_FIELD(vc4_state->crtc_y, SCALER_POS0_START_Y));

	/* Position Word 1: Scaled Image Dimensions. */
	if (!vc4_state->is_unity) {
		vc4_dlist_write(vc4_state,
				VC4_SET_FIELD(vc4_state->crtc_w,
					      SCALER_POS1_SCL_WIDTH) |
				VC4_SET_FIELD(vc4_state->crtc_h,
					      SCALER_POS1_SCL_HEIGHT));
	}

	/* Don't waste cycles mixing with plane alpha if the set alpha
	 * is opaque or there is no per-pixel alpha information.
	 * In any case we use the alpha property value as the fixed alpha.
	 */
	mix_plane_alpha = state->alpha != DRM_BLEND_ALPHA_OPAQUE &&
			  fb->format->has_alpha;

	/* Position Word 2: Source Image Size, Alpha */
	vc4_state->pos2_offset = vc4_state->dlist_count;
	vc4_dlist_write(vc4_state,
			VC4_SET_FIELD(fb->format->has_alpha ?
				      SCALER_POS2_ALPHA_MODE_PIPELINE :
				      SCALER_POS2_ALPHA_MODE_FIXED,
				      SCALER_POS2_ALPHA_MODE) |
			(mix_plane_alpha ? SCALER_POS2_ALPHA_MIX : 0) |
			(fb->format->has_alpha ? SCALER_POS2_ALPHA_PREMULT : 0) |
			VC4_SET_FIELD(vc4_state->src_w[0], SCALER_POS2_WIDTH) |
			VC4_SET_FIELD(vc4_state->src_h[0], SCALER_POS2_HEIGHT));

	/* Position Word 3: Context.  Written by the HVS. */
	vc4_dlist_write(vc4_state, 0xc0c0c0c0);


	/* Pointer Word 0/1/2: RGB / Y / Cb / Cr Pointers
	 *
	 * The pointers may be any byte address.
	 */
	vc4_state->ptr0_offset = vc4_state->dlist_count;
	for (i = 0; i < num_planes; i++)
		vc4_dlist_write(vc4_state, vc4_state->offsets[i]);

	/* Pointer Context Word 0/1/2: Written by the HVS */
	for (i = 0; i < num_planes; i++)
		vc4_dlist_write(vc4_state, 0xc0c0c0c0);

	/* Pitch word 0 */
	vc4_dlist_write(vc4_state, pitch0);

	/* Pitch word 1/2 */
	for (i = 1; i < num_planes; i++) {
		if (hvs_format != HVS_PIXEL_FORMAT_H264) {
			vc4_dlist_write(vc4_state,
					VC4_SET_FIELD(fb->pitches[i],
						      SCALER_SRC_PITCH));
		} else {
			vc4_dlist_write(vc4_state, pitch0);
		}
	}

	/* Colorspace conversion words */
	if (vc4_state->is_yuv) {
		vc4_dlist_write(vc4_state, SCALER_CSC0_ITR_R_601_5);
		vc4_dlist_write(vc4_state, SCALER_CSC1_ITR_R_601_5);
		vc4_dlist_write(vc4_state, SCALER_CSC2_ITR_R_601_5);
	}

	vc4_state->lbm_offset = 0;

	if (vc4_state->x_scaling[0] != VC4_SCALING_NONE ||
	    vc4_state->x_scaling[1] != VC4_SCALING_NONE ||
	    vc4_state->y_scaling[0] != VC4_SCALING_NONE ||
	    vc4_state->y_scaling[1] != VC4_SCALING_NONE) {
		/* Reserve a slot for the LBM Base Address. The real value will
		 * be set when calling vc4_plane_allocate_lbm().
		 */
		if (vc4_state->y_scaling[0] != VC4_SCALING_NONE ||
		    vc4_state->y_scaling[1] != VC4_SCALING_NONE)
			vc4_state->lbm_offset = vc4_state->dlist_count++;

		if (num_planes > 1) {
			/* Emit Cb/Cr as channel 0 and Y as channel
			 * 1. This matches how we set up scl0/scl1
			 * above.
			 */
			vc4_write_scaling_parameters(state, 1);
		}
		vc4_write_scaling_parameters(state, 0);

		/* If any PPF setup was done, then all the kernel
		 * pointers get uploaded.
		 */
		if (vc4_state->x_scaling[0] == VC4_SCALING_PPF ||
		    vc4_state->y_scaling[0] == VC4_SCALING_PPF ||
		    vc4_state->x_scaling[1] == VC4_SCALING_PPF ||
		    vc4_state->y_scaling[1] == VC4_SCALING_PPF) {
			u32 kernel = VC4_SET_FIELD(vc4->hvs->mitchell_netravali_filter.start,
						   SCALER_PPF_KERNEL_OFFSET);

			/* HPPF plane 0 */
			vc4_dlist_write(vc4_state, kernel);
			/* VPPF plane 0 */
			vc4_dlist_write(vc4_state, kernel);
			/* HPPF plane 1 */
			vc4_dlist_write(vc4_state, kernel);
			/* VPPF plane 1 */
			vc4_dlist_write(vc4_state, kernel);
		}
	}

	vc4_state->dlist[ctl0_offset] |=
		VC4_SET_FIELD(vc4_state->dlist_count, SCALER_CTL0_SIZE);

	/* crtc_* are already clipped coordinates. */
	covers_screen = vc4_state->crtc_x == 0 && vc4_state->crtc_y == 0 &&
			vc4_state->crtc_w == state->crtc->mode.hdisplay &&
			vc4_state->crtc_h == state->crtc->mode.vdisplay;
	/* Background fill might be necessary when the plane has per-pixel
	 * alpha content or a non-opaque plane alpha and could blend from the
	 * background or does not cover the entire screen.
	 */
	vc4_state->needs_bg_fill = fb->format->has_alpha || !covers_screen ||
				   state->alpha != DRM_BLEND_ALPHA_OPAQUE;

	/* Flag the dlist as initialized to avoid checking it twice in case
	 * the async update check already called vc4_plane_mode_set() and
	 * decided to fallback to sync update because async update was not
	 * possible.
	 */
	vc4_state->dlist_initialized = 1;

	vc4_plane_calc_load(state);

	return 0;
}

/* If a modeset involves changing the setup of a plane, the atomic
 * infrastructure will call this to validate a proposed plane setup.
 * However, if a plane isn't getting updated, this (and the
 * corresponding vc4_plane_atomic_update) won't get called.  Thus, we
 * compute the dlist here and have all active plane dlists get updated
 * in the CRTC's flush.
 */
static int vc4_plane_atomic_check(struct drm_plane *plane,
				  struct drm_plane_state *state)
{
	struct vc4_plane_state *vc4_state = to_vc4_plane_state(state);
	int ret;

	vc4_state->dlist_count = 0;

	if (!plane_enabled(state))
		return 0;

	ret = vc4_plane_mode_set(plane, state);
	if (ret)
		return ret;

	return vc4_plane_allocate_lbm(state);
}

static void vc4_plane_atomic_update(struct drm_plane *plane,
				    struct drm_plane_state *old_state)
{
	/* No contents here.  Since we don't know where in the CRTC's
	 * dlist we should be stored, our dlist is uploaded to the
	 * hardware with vc4_plane_write_dlist() at CRTC atomic_flush
	 * time.
	 */
}

u32 vc4_plane_write_dlist(struct drm_plane *plane, u32 __iomem *dlist)
{
	struct vc4_plane_state *vc4_state = to_vc4_plane_state(plane->state);
	int i;

	vc4_state->hw_dlist = dlist;

	/* Can't memcpy_toio() because it needs to be 32-bit writes. */
	for (i = 0; i < vc4_state->dlist_count; i++)
		writel(vc4_state->dlist[i], &dlist[i]);

	return vc4_state->dlist_count;
}

u32 vc4_plane_dlist_size(const struct drm_plane_state *state)
{
	const struct vc4_plane_state *vc4_state =
		container_of(state, typeof(*vc4_state), base);

	return vc4_state->dlist_count;
}

/* Updates the plane to immediately (well, once the FIFO needs
 * refilling) scan out from at a new framebuffer.
 */
void vc4_plane_async_set_fb(struct drm_plane *plane, struct drm_framebuffer *fb)
{
	struct vc4_plane_state *vc4_state = to_vc4_plane_state(plane->state);
	struct drm_gem_cma_object *bo = drm_fb_cma_get_gem_obj(fb, 0);
	uint32_t addr;

	/* We're skipping the address adjustment for negative origin,
	 * because this is only called on the primary plane.
	 */
	WARN_ON_ONCE(plane->state->crtc_x < 0 || plane->state->crtc_y < 0);
	addr = bo->paddr + fb->offsets[0];

	/* Write the new address into the hardware immediately.  The
	 * scanout will start from this address as soon as the FIFO
	 * needs to refill with pixels.
	 */
	writel(addr, &vc4_state->hw_dlist[vc4_state->ptr0_offset]);

	/* Also update the CPU-side dlist copy, so that any later
	 * atomic updates that don't do a new modeset on our plane
	 * also use our updated address.
	 */
	vc4_state->dlist[vc4_state->ptr0_offset] = addr;
}

static void vc4_plane_atomic_async_update(struct drm_plane *plane,
					  struct drm_plane_state *state)
{
	struct vc4_plane_state *vc4_state, *new_vc4_state;

	swap(plane->state->fb, state->fb);
	plane->state->crtc_x = state->crtc_x;
	plane->state->crtc_y = state->crtc_y;
	plane->state->crtc_w = state->crtc_w;
	plane->state->crtc_h = state->crtc_h;
	plane->state->src_x = state->src_x;
	plane->state->src_y = state->src_y;
	plane->state->src_w = state->src_w;
	plane->state->src_h = state->src_h;
	plane->state->src_h = state->src_h;
	plane->state->alpha = state->alpha;
	plane->state->pixel_blend_mode = state->pixel_blend_mode;
	plane->state->rotation = state->rotation;
	plane->state->zpos = state->zpos;
	plane->state->normalized_zpos = state->normalized_zpos;
	plane->state->color_encoding = state->color_encoding;
	plane->state->color_range = state->color_range;
	plane->state->src = state->src;
	plane->state->dst = state->dst;
	plane->state->visible = state->visible;

	new_vc4_state = to_vc4_plane_state(state);
	vc4_state = to_vc4_plane_state(plane->state);

	vc4_state->crtc_x = new_vc4_state->crtc_x;
	vc4_state->crtc_y = new_vc4_state->crtc_y;
	vc4_state->crtc_h = new_vc4_state->crtc_h;
	vc4_state->crtc_w = new_vc4_state->crtc_w;
	vc4_state->src_x = new_vc4_state->src_x;
	vc4_state->src_y = new_vc4_state->src_y;
	memcpy(vc4_state->src_w, new_vc4_state->src_w,
	       sizeof(vc4_state->src_w));
	memcpy(vc4_state->src_h, new_vc4_state->src_h,
	       sizeof(vc4_state->src_h));
	memcpy(vc4_state->x_scaling, new_vc4_state->x_scaling,
	       sizeof(vc4_state->x_scaling));
	memcpy(vc4_state->y_scaling, new_vc4_state->y_scaling,
	       sizeof(vc4_state->y_scaling));
	vc4_state->is_unity = new_vc4_state->is_unity;
	vc4_state->is_yuv = new_vc4_state->is_yuv;
	memcpy(vc4_state->offsets, new_vc4_state->offsets,
	       sizeof(vc4_state->offsets));
	vc4_state->needs_bg_fill = new_vc4_state->needs_bg_fill;

	/* Update the current vc4_state pos0, pos2 and ptr0 dlist entries. */
	vc4_state->dlist[vc4_state->pos0_offset] =
		new_vc4_state->dlist[vc4_state->pos0_offset];
	vc4_state->dlist[vc4_state->pos2_offset] =
		new_vc4_state->dlist[vc4_state->pos2_offset];
	vc4_state->dlist[vc4_state->ptr0_offset] =
		new_vc4_state->dlist[vc4_state->ptr0_offset];

	/* Note that we can't just call vc4_plane_write_dlist()
	 * because that would smash the context data that the HVS is
	 * currently using.
	 */
	writel(vc4_state->dlist[vc4_state->pos0_offset],
	       &vc4_state->hw_dlist[vc4_state->pos0_offset]);
	writel(vc4_state->dlist[vc4_state->pos2_offset],
	       &vc4_state->hw_dlist[vc4_state->pos2_offset]);
	writel(vc4_state->dlist[vc4_state->ptr0_offset],
	       &vc4_state->hw_dlist[vc4_state->ptr0_offset]);
}

static int vc4_plane_atomic_async_check(struct drm_plane *plane,
					struct drm_plane_state *state)
{
	struct vc4_plane_state *old_vc4_state, *new_vc4_state;
	int ret;
	u32 i;

	ret = vc4_plane_mode_set(plane, state);
	if (ret)
		return ret;

	old_vc4_state = to_vc4_plane_state(plane->state);
	new_vc4_state = to_vc4_plane_state(state);
	if (old_vc4_state->dlist_count != new_vc4_state->dlist_count ||
	    old_vc4_state->pos0_offset != new_vc4_state->pos0_offset ||
	    old_vc4_state->pos2_offset != new_vc4_state->pos2_offset ||
	    old_vc4_state->ptr0_offset != new_vc4_state->ptr0_offset ||
	    vc4_lbm_size(plane->state) != vc4_lbm_size(state))
		return -EINVAL;

	/* Only pos0, pos2 and ptr0 DWORDS can be updated in an async update
	 * if anything else has changed, fallback to a sync update.
	 */
	for (i = 0; i < new_vc4_state->dlist_count; i++) {
		if (i == new_vc4_state->pos0_offset ||
		    i == new_vc4_state->pos2_offset ||
		    i == new_vc4_state->ptr0_offset ||
		    (new_vc4_state->lbm_offset &&
		     i == new_vc4_state->lbm_offset))
			continue;

		if (new_vc4_state->dlist[i] != old_vc4_state->dlist[i])
			return -EINVAL;
	}

	return 0;
}

static int vc4_prepare_fb(struct drm_plane *plane,
			  struct drm_plane_state *state)
{
	struct vc4_bo *bo;
	struct dma_fence *fence;
	int ret;

	if (!state->fb)
		return 0;

	bo = to_vc4_bo(&drm_fb_cma_get_gem_obj(state->fb, 0)->base);

	fence = reservation_object_get_excl_rcu(bo->base.base.resv);
	drm_atomic_set_fence_for_plane(state, fence);

	if (plane->state->fb == state->fb)
		return 0;

	ret = vc4_bo_inc_usecnt(bo);
	if (ret)
		return ret;

	return 0;
}

static void vc4_cleanup_fb(struct drm_plane *plane,
			   struct drm_plane_state *state)
{
	struct vc4_bo *bo;

	if (plane->state->fb == state->fb || !state->fb)
		return;

	bo = to_vc4_bo(&drm_fb_cma_get_gem_obj(state->fb, 0)->base);
	vc4_bo_dec_usecnt(bo);
}

static const struct drm_plane_helper_funcs vc4_plane_helper_funcs = {
	.atomic_check = vc4_plane_atomic_check,
	.atomic_update = vc4_plane_atomic_update,
	.prepare_fb = vc4_prepare_fb,
	.cleanup_fb = vc4_cleanup_fb,
	.atomic_async_check = vc4_plane_atomic_async_check,
	.atomic_async_update = vc4_plane_atomic_async_update,
};

static void vc4_plane_destroy(struct drm_plane *plane)
{
	drm_plane_cleanup(plane);
}

static bool vc4_format_mod_supported(struct drm_plane *plane,
				     uint32_t format,
				     uint64_t modifier)
{
	/* Support T_TILING for RGB formats only. */
	switch (format) {
	case DRM_FORMAT_XRGB8888:
	case DRM_FORMAT_ARGB8888:
	case DRM_FORMAT_ABGR8888:
	case DRM_FORMAT_XBGR8888:
	case DRM_FORMAT_RGB565:
	case DRM_FORMAT_BGR565:
	case DRM_FORMAT_ARGB1555:
	case DRM_FORMAT_XRGB1555:
		switch (fourcc_mod_broadcom_mod(modifier)) {
		case DRM_FORMAT_MOD_LINEAR:
		case DRM_FORMAT_MOD_BROADCOM_VC4_T_TILED:
			return true;
		default:
			return false;
		}
	case DRM_FORMAT_NV12:
	case DRM_FORMAT_NV21:
		switch (fourcc_mod_broadcom_mod(modifier)) {
		case DRM_FORMAT_MOD_LINEAR:
		case DRM_FORMAT_MOD_BROADCOM_SAND64:
		case DRM_FORMAT_MOD_BROADCOM_SAND128:
		case DRM_FORMAT_MOD_BROADCOM_SAND256:
			return true;
		default:
			return false;
		}
	case DRM_FORMAT_YUV422:
	case DRM_FORMAT_YVU422:
	case DRM_FORMAT_YUV420:
	case DRM_FORMAT_YVU420:
	case DRM_FORMAT_NV16:
	case DRM_FORMAT_NV61:
	default:
		return (modifier == DRM_FORMAT_MOD_LINEAR);
	}
}

static const struct drm_plane_funcs vc4_plane_funcs = {
	.update_plane = drm_atomic_helper_update_plane,
	.disable_plane = drm_atomic_helper_disable_plane,
	.destroy = vc4_plane_destroy,
	.set_property = NULL,
	.reset = vc4_plane_reset,
	.atomic_duplicate_state = vc4_plane_duplicate_state,
	.atomic_destroy_state = vc4_plane_destroy_state,
	.format_mod_supported = vc4_format_mod_supported,
};

struct drm_plane *vc4_plane_init(struct drm_device *dev,
				 enum drm_plane_type type)
{
	struct drm_plane *plane = NULL;
	struct vc4_plane *vc4_plane;
	u32 formats[ARRAY_SIZE(hvs_formats)];
	int ret = 0;
	unsigned i;
	static const uint64_t modifiers[] = {
		DRM_FORMAT_MOD_BROADCOM_VC4_T_TILED,
		DRM_FORMAT_MOD_BROADCOM_SAND128,
		DRM_FORMAT_MOD_BROADCOM_SAND64,
		DRM_FORMAT_MOD_BROADCOM_SAND256,
		DRM_FORMAT_MOD_LINEAR,
		DRM_FORMAT_MOD_INVALID
	};

	vc4_plane = devm_kzalloc(dev->dev, sizeof(*vc4_plane),
				 GFP_KERNEL);
	if (!vc4_plane)
		return ERR_PTR(-ENOMEM);

	for (i = 0; i < ARRAY_SIZE(hvs_formats); i++)
		formats[i] = hvs_formats[i].drm;

	plane = &vc4_plane->base;
	ret = drm_universal_plane_init(dev, plane, 0,
				       &vc4_plane_funcs,
				       formats, ARRAY_SIZE(formats),
				       modifiers, type, NULL);

	drm_plane_helper_add(plane, &vc4_plane_helper_funcs);

	drm_plane_create_alpha_property(plane);
	drm_plane_create_rotation_property(plane, DRM_MODE_ROTATE_0,
					   DRM_MODE_ROTATE_0 |
					   DRM_MODE_ROTATE_180 |
					   DRM_MODE_REFLECT_X |
					   DRM_MODE_REFLECT_Y);

	return plane;
}