summaryrefslogtreecommitdiffstats
path: root/drivers/gpu/drm/i915/i915_gem.c
blob: b3cedd20f3653c17a446dc5c656ee35e711c3ba6 (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
/*
 * Copyright © 2008-2015 Intel Corporation
 *
 * 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 (including the next
 * paragraph) 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 AUTHORS OR COPYRIGHT HOLDERS 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.
 *
 * Authors:
 *    Eric Anholt <eric@anholt.net>
 *
 */

#include <drm/drm_vma_manager.h>
#include <linux/dma-fence-array.h>
#include <linux/kthread.h>
#include <linux/dma-resv.h>
#include <linux/shmem_fs.h>
#include <linux/slab.h>
#include <linux/stop_machine.h>
#include <linux/swap.h>
#include <linux/pci.h>
#include <linux/dma-buf.h>
#include <linux/mman.h>

#include "display/intel_display.h"
#include "display/intel_frontbuffer.h"

#include "gem/i915_gem_clflush.h"
#include "gem/i915_gem_context.h"
#include "gem/i915_gem_ioctls.h"
#include "gem/i915_gem_mman.h"
#include "gem/i915_gem_region.h"
#include "gt/intel_engine_user.h"
#include "gt/intel_gt.h"
#include "gt/intel_gt_pm.h"
#include "gt/intel_workarounds.h"

#include "i915_drv.h"
#include "i915_trace.h"
#include "i915_vgpu.h"

#include "intel_pm.h"

static int
insert_mappable_node(struct i915_ggtt *ggtt, struct drm_mm_node *node, u32 size)
{
	int err;

	err = mutex_lock_interruptible(&ggtt->vm.mutex);
	if (err)
		return err;

	memset(node, 0, sizeof(*node));
	err = drm_mm_insert_node_in_range(&ggtt->vm.mm, node,
					  size, 0, I915_COLOR_UNEVICTABLE,
					  0, ggtt->mappable_end,
					  DRM_MM_INSERT_LOW);

	mutex_unlock(&ggtt->vm.mutex);

	return err;
}

static void
remove_mappable_node(struct i915_ggtt *ggtt, struct drm_mm_node *node)
{
	mutex_lock(&ggtt->vm.mutex);
	drm_mm_remove_node(node);
	mutex_unlock(&ggtt->vm.mutex);
}

int
i915_gem_get_aperture_ioctl(struct drm_device *dev, void *data,
			    struct drm_file *file)
{
	struct i915_ggtt *ggtt = &to_i915(dev)->ggtt;
	struct drm_i915_gem_get_aperture *args = data;
	struct i915_vma *vma;
	u64 pinned;

	if (mutex_lock_interruptible(&ggtt->vm.mutex))
		return -EINTR;

	pinned = ggtt->vm.reserved;
	list_for_each_entry(vma, &ggtt->vm.bound_list, vm_link)
		if (i915_vma_is_pinned(vma))
			pinned += vma->node.size;

	mutex_unlock(&ggtt->vm.mutex);

	args->aper_size = ggtt->vm.total;
	args->aper_available_size = args->aper_size - pinned;

	return 0;
}

int i915_gem_object_unbind(struct drm_i915_gem_object *obj,
			   unsigned long flags)
{
	struct intel_runtime_pm *rpm = &to_i915(obj->base.dev)->runtime_pm;
	LIST_HEAD(still_in_list);
	intel_wakeref_t wakeref;
	struct i915_vma *vma;
	int ret;

	if (list_empty(&obj->vma.list))
		return 0;

	/*
	 * As some machines use ACPI to handle runtime-resume callbacks, and
	 * ACPI is quite kmalloc happy, we cannot resume beneath the vm->mutex
	 * as they are required by the shrinker. Ergo, we wake the device up
	 * first just in case.
	 */
	wakeref = intel_runtime_pm_get(rpm);

try_again:
	ret = 0;
	spin_lock(&obj->vma.lock);
	while (!ret && (vma = list_first_entry_or_null(&obj->vma.list,
						       struct i915_vma,
						       obj_link))) {
		struct i915_address_space *vm = vma->vm;

		list_move_tail(&vma->obj_link, &still_in_list);
		if (!i915_vma_is_bound(vma, I915_VMA_BIND_MASK))
			continue;

		if (flags & I915_GEM_OBJECT_UNBIND_TEST) {
			ret = -EBUSY;
			break;
		}

		ret = -EAGAIN;
		if (!i915_vm_tryopen(vm))
			break;

		/* Prevent vma being freed by i915_vma_parked as we unbind */
		vma = __i915_vma_get(vma);
		spin_unlock(&obj->vma.lock);

		if (vma) {
			ret = -EBUSY;
			if (flags & I915_GEM_OBJECT_UNBIND_ACTIVE ||
			    !i915_vma_is_active(vma))
				ret = i915_vma_unbind(vma);

			__i915_vma_put(vma);
		}

		i915_vm_close(vm);
		spin_lock(&obj->vma.lock);
	}
	list_splice_init(&still_in_list, &obj->vma.list);
	spin_unlock(&obj->vma.lock);

	if (ret == -EAGAIN && flags & I915_GEM_OBJECT_UNBIND_BARRIER) {
		rcu_barrier(); /* flush the i915_vm_release() */
		goto try_again;
	}

	intel_runtime_pm_put(rpm, wakeref);

	return ret;
}

static int
shmem_pread(struct page *page, int offset, int len, char __user *user_data,
	    bool needs_clflush)
{
	char *vaddr;
	int ret;

	vaddr = kmap(page);

	if (needs_clflush)
		drm_clflush_virt_range(vaddr + offset, len);

	ret = __copy_to_user(user_data, vaddr + offset, len);

	kunmap(page);

	return ret ? -EFAULT : 0;
}

static int
i915_gem_shmem_pread(struct drm_i915_gem_object *obj,
		     struct drm_i915_gem_pread *args)
{
	unsigned int needs_clflush;
	unsigned int idx, offset;
	char __user *user_data;
	u64 remain;
	int ret;

	ret = i915_gem_object_lock_interruptible(obj, NULL);
	if (ret)
		return ret;

	ret = i915_gem_object_pin_pages(obj);
	if (ret)
		goto err_unlock;

	ret = i915_gem_object_prepare_read(obj, &needs_clflush);
	if (ret)
		goto err_unpin;

	i915_gem_object_finish_access(obj);
	i915_gem_object_unlock(obj);

	remain = args->size;
	user_data = u64_to_user_ptr(args->data_ptr);
	offset = offset_in_page(args->offset);
	for (idx = args->offset >> PAGE_SHIFT; remain; idx++) {
		struct page *page = i915_gem_object_get_page(obj, idx);
		unsigned int length = min_t(u64, remain, PAGE_SIZE - offset);

		ret = shmem_pread(page, offset, length, user_data,
				  needs_clflush);
		if (ret)
			break;

		remain -= length;
		user_data += length;
		offset = 0;
	}

	i915_gem_object_unpin_pages(obj);
	return ret;

err_unpin:
	i915_gem_object_unpin_pages(obj);
err_unlock:
	i915_gem_object_unlock(obj);
	return ret;
}

static inline bool
gtt_user_read(struct io_mapping *mapping,
	      loff_t base, int offset,
	      char __user *user_data, int length)
{
	void __iomem *vaddr;
	unsigned long unwritten;

	/* We can use the cpu mem copy function because this is X86. */
	vaddr = io_mapping_map_atomic_wc(mapping, base);
	unwritten = __copy_to_user_inatomic(user_data,
					    (void __force *)vaddr + offset,
					    length);
	io_mapping_unmap_atomic(vaddr);
	if (unwritten) {
		vaddr = io_mapping_map_wc(mapping, base, PAGE_SIZE);
		unwritten = copy_to_user(user_data,
					 (void __force *)vaddr + offset,
					 length);
		io_mapping_unmap(vaddr);
	}
	return unwritten;
}

static struct i915_vma *i915_gem_gtt_prepare(struct drm_i915_gem_object *obj,
					     struct drm_mm_node *node,
					     bool write)
{
	struct drm_i915_private *i915 = to_i915(obj->base.dev);
	struct i915_ggtt *ggtt = &i915->ggtt;
	struct i915_vma *vma;
	struct i915_gem_ww_ctx ww;
	int ret;

	i915_gem_ww_ctx_init(&ww, true);
retry:
	vma = ERR_PTR(-ENODEV);
	ret = i915_gem_object_lock(obj, &ww);
	if (ret)
		goto err_ww;

	ret = i915_gem_object_set_to_gtt_domain(obj, write);
	if (ret)
		goto err_ww;

	if (!i915_gem_object_is_tiled(obj))
		vma = i915_gem_object_ggtt_pin_ww(obj, &ww, NULL, 0, 0,
						  PIN_MAPPABLE |
						  PIN_NONBLOCK /* NOWARN */ |
						  PIN_NOEVICT);
	if (vma == ERR_PTR(-EDEADLK)) {
		ret = -EDEADLK;
		goto err_ww;
	} else if (!IS_ERR(vma)) {
		node->start = i915_ggtt_offset(vma);
		node->flags = 0;
	} else {
		ret = insert_mappable_node(ggtt, node, PAGE_SIZE);
		if (ret)
			goto err_ww;
		GEM_BUG_ON(!drm_mm_node_allocated(node));
		vma = NULL;
	}

	ret = i915_gem_object_pin_pages(obj);
	if (ret) {
		if (drm_mm_node_allocated(node)) {
			ggtt->vm.clear_range(&ggtt->vm, node->start, node->size);
			remove_mappable_node(ggtt, node);
		} else {
			i915_vma_unpin(vma);
		}
	}

err_ww:
	if (ret == -EDEADLK) {
		ret = i915_gem_ww_ctx_backoff(&ww);
		if (!ret)
			goto retry;
	}
	i915_gem_ww_ctx_fini(&ww);

	return ret ? ERR_PTR(ret) : vma;
}

static void i915_gem_gtt_cleanup(struct drm_i915_gem_object *obj,
				 struct drm_mm_node *node,
				 struct i915_vma *vma)
{
	struct drm_i915_private *i915 = to_i915(obj->base.dev);
	struct i915_ggtt *ggtt = &i915->ggtt;

	i915_gem_object_unpin_pages(obj);
	if (drm_mm_node_allocated(node)) {
		ggtt->vm.clear_range(&ggtt->vm, node->start, node->size);
		remove_mappable_node(ggtt, node);
	} else {
		i915_vma_unpin(vma);
	}
}

static int
i915_gem_gtt_pread(struct drm_i915_gem_object *obj,
		   const struct drm_i915_gem_pread *args)
{
	struct drm_i915_private *i915 = to_i915(obj->base.dev);
	struct i915_ggtt *ggtt = &i915->ggtt;
	intel_wakeref_t wakeref;
	struct drm_mm_node node;
	void __user *user_data;
	struct i915_vma *vma;
	u64 remain, offset;
	int ret = 0;

	wakeref = intel_runtime_pm_get(&i915->runtime_pm);

	vma = i915_gem_gtt_prepare(obj, &node, false);
	if (IS_ERR(vma)) {
		ret = PTR_ERR(vma);
		goto out_rpm;
	}

	user_data = u64_to_user_ptr(args->data_ptr);
	remain = args->size;
	offset = args->offset;

	while (remain > 0) {
		/* Operation in this page
		 *
		 * page_base = page offset within aperture
		 * page_offset = offset within page
		 * page_length = bytes to copy for this page
		 */
		u32 page_base = node.start;
		unsigned page_offset = offset_in_page(offset);
		unsigned page_length = PAGE_SIZE - page_offset;
		page_length = remain < page_length ? remain : page_length;
		if (drm_mm_node_allocated(&node)) {
			ggtt->vm.insert_page(&ggtt->vm,
					     i915_gem_object_get_dma_address(obj, offset >> PAGE_SHIFT),
					     node.start, I915_CACHE_NONE, 0);
		} else {
			page_base += offset & PAGE_MASK;
		}

		if (gtt_user_read(&ggtt->iomap, page_base, page_offset,
				  user_data, page_length)) {
			ret = -EFAULT;
			break;
		}

		remain -= page_length;
		user_data += page_length;
		offset += page_length;
	}

	i915_gem_gtt_cleanup(obj, &node, vma);
out_rpm:
	intel_runtime_pm_put(&i915->runtime_pm, wakeref);
	return ret;
}

/**
 * Reads data from the object referenced by handle.
 * @dev: drm device pointer
 * @data: ioctl data blob
 * @file: drm file pointer
 *
 * On error, the contents of *data are undefined.
 */
int
i915_gem_pread_ioctl(struct drm_device *dev, void *data,
		     struct drm_file *file)
{
	struct drm_i915_private *i915 = to_i915(dev);
	struct drm_i915_gem_pread *args = data;
	struct drm_i915_gem_object *obj;
	int ret;

	/* PREAD is disallowed for all platforms after TGL-LP.  This also
	 * covers all platforms with local memory.
	 */
	if (INTEL_GEN(i915) >= 12 && !IS_TIGERLAKE(i915))
		return -EOPNOTSUPP;

	if (args->size == 0)
		return 0;

	if (!access_ok(u64_to_user_ptr(args->data_ptr),
		       args->size))
		return -EFAULT;

	obj = i915_gem_object_lookup(file, args->handle);
	if (!obj)
		return -ENOENT;

	/* Bounds check source.  */
	if (range_overflows_t(u64, args->offset, args->size, obj->base.size)) {
		ret = -EINVAL;
		goto out;
	}

	trace_i915_gem_object_pread(obj, args->offset, args->size);
	ret = -ENODEV;
	if (obj->ops->pread)
		ret = obj->ops->pread(obj, args);
	if (ret != -ENODEV)
		goto out;

	ret = -ENODEV;
	if (obj->ops->pread)
		ret = obj->ops->pread(obj, args);
	if (ret != -ENODEV)
		goto out;

	ret = i915_gem_object_wait(obj,
				   I915_WAIT_INTERRUPTIBLE,
				   MAX_SCHEDULE_TIMEOUT);
	if (ret)
		goto out;

	ret = i915_gem_shmem_pread(obj, args);
	if (ret == -EFAULT || ret == -ENODEV)
		ret = i915_gem_gtt_pread(obj, args);

out:
	i915_gem_object_put(obj);
	return ret;
}

/* This is the fast write path which cannot handle
 * page faults in the source data
 */

static inline bool
ggtt_write(struct io_mapping *mapping,
	   loff_t base, int offset,
	   char __user *user_data, int length)
{
	void __iomem *vaddr;
	unsigned long unwritten;

	/* We can use the cpu mem copy function because this is X86. */
	vaddr = io_mapping_map_atomic_wc(mapping, base);
	unwritten = __copy_from_user_inatomic_nocache((void __force *)vaddr + offset,
						      user_data, length);
	io_mapping_unmap_atomic(vaddr);
	if (unwritten) {
		vaddr = io_mapping_map_wc(mapping, base, PAGE_SIZE);
		unwritten = copy_from_user((void __force *)vaddr + offset,
					   user_data, length);
		io_mapping_unmap(vaddr);
	}

	return unwritten;
}

/**
 * This is the fast pwrite path, where we copy the data directly from the
 * user into the GTT, uncached.
 * @obj: i915 GEM object
 * @args: pwrite arguments structure
 */
static int
i915_gem_gtt_pwrite_fast(struct drm_i915_gem_object *obj,
			 const struct drm_i915_gem_pwrite *args)
{
	struct drm_i915_private *i915 = to_i915(obj->base.dev);
	struct i915_ggtt *ggtt = &i915->ggtt;
	struct intel_runtime_pm *rpm = &i915->runtime_pm;
	intel_wakeref_t wakeref;
	struct drm_mm_node node;
	struct i915_vma *vma;
	u64 remain, offset;
	void __user *user_data;
	int ret = 0;

	if (i915_gem_object_has_struct_page(obj)) {
		/*
		 * Avoid waking the device up if we can fallback, as
		 * waking/resuming is very slow (worst-case 10-100 ms
		 * depending on PCI sleeps and our own resume time).
		 * This easily dwarfs any performance advantage from
		 * using the cache bypass of indirect GGTT access.
		 */
		wakeref = intel_runtime_pm_get_if_in_use(rpm);
		if (!wakeref)
			return -EFAULT;
	} else {
		/* No backing pages, no fallback, we must force GGTT access */
		wakeref = intel_runtime_pm_get(rpm);
	}

	vma = i915_gem_gtt_prepare(obj, &node, true);
	if (IS_ERR(vma)) {
		ret = PTR_ERR(vma);
		goto out_rpm;
	}

	i915_gem_object_invalidate_frontbuffer(obj, ORIGIN_CPU);

	user_data = u64_to_user_ptr(args->data_ptr);
	offset = args->offset;
	remain = args->size;
	while (remain) {
		/* Operation in this page
		 *
		 * page_base = page offset within aperture
		 * page_offset = offset within page
		 * page_length = bytes to copy for this page
		 */
		u32 page_base = node.start;
		unsigned int page_offset = offset_in_page(offset);
		unsigned int page_length = PAGE_SIZE - page_offset;
		page_length = remain < page_length ? remain : page_length;
		if (drm_mm_node_allocated(&node)) {
			/* flush the write before we modify the GGTT */
			intel_gt_flush_ggtt_writes(ggtt->vm.gt);
			ggtt->vm.insert_page(&ggtt->vm,
					     i915_gem_object_get_dma_address(obj, offset >> PAGE_SHIFT),
					     node.start, I915_CACHE_NONE, 0);
			wmb(); /* flush modifications to the GGTT (insert_page) */
		} else {
			page_base += offset & PAGE_MASK;
		}
		/* If we get a fault while copying data, then (presumably) our
		 * source page isn't available.  Return the error and we'll
		 * retry in the slow path.
		 * If the object is non-shmem backed, we retry again with the
		 * path that handles page fault.
		 */
		if (ggtt_write(&ggtt->iomap, page_base, page_offset,
			       user_data, page_length)) {
			ret = -EFAULT;
			break;
		}

		remain -= page_length;
		user_data += page_length;
		offset += page_length;
	}

	intel_gt_flush_ggtt_writes(ggtt->vm.gt);
	i915_gem_object_flush_frontbuffer(obj, ORIGIN_CPU);

	i915_gem_gtt_cleanup(obj, &node, vma);
out_rpm:
	intel_runtime_pm_put(rpm, wakeref);
	return ret;
}

/* Per-page copy function for the shmem pwrite fastpath.
 * Flushes invalid cachelines before writing to the target if
 * needs_clflush_before is set and flushes out any written cachelines after
 * writing if needs_clflush is set.
 */
static int
shmem_pwrite(struct page *page, int offset, int len, char __user *user_data,
	     bool needs_clflush_before,
	     bool needs_clflush_after)
{
	char *vaddr;
	int ret;

	vaddr = kmap(page);

	if (needs_clflush_before)
		drm_clflush_virt_range(vaddr + offset, len);

	ret = __copy_from_user(vaddr + offset, user_data, len);
	if (!ret && needs_clflush_after)
		drm_clflush_virt_range(vaddr + offset, len);

	kunmap(page);

	return ret ? -EFAULT : 0;
}

static int
i915_gem_shmem_pwrite(struct drm_i915_gem_object *obj,
		      const struct drm_i915_gem_pwrite *args)
{
	unsigned int partial_cacheline_write;
	unsigned int needs_clflush;
	unsigned int offset, idx;
	void __user *user_data;
	u64 remain;
	int ret;

	ret = i915_gem_object_lock_interruptible(obj, NULL);
	if (ret)
		return ret;

	ret = i915_gem_object_pin_pages(obj);
	if (ret)
		goto err_unlock;

	ret = i915_gem_object_prepare_write(obj, &needs_clflush);
	if (ret)
		goto err_unpin;

	i915_gem_object_finish_access(obj);
	i915_gem_object_unlock(obj);

	/* If we don't overwrite a cacheline completely we need to be
	 * careful to have up-to-date data by first clflushing. Don't
	 * overcomplicate things and flush the entire patch.
	 */
	partial_cacheline_write = 0;
	if (needs_clflush & CLFLUSH_BEFORE)
		partial_cacheline_write = boot_cpu_data.x86_clflush_size - 1;

	user_data = u64_to_user_ptr(args->data_ptr);
	remain = args->size;
	offset = offset_in_page(args->offset);
	for (idx = args->offset >> PAGE_SHIFT; remain; idx++) {
		struct page *page = i915_gem_object_get_page(obj, idx);
		unsigned int length = min_t(u64, remain, PAGE_SIZE - offset);

		ret = shmem_pwrite(page, offset, length, user_data,
				   (offset | length) & partial_cacheline_write,
				   needs_clflush & CLFLUSH_AFTER);
		if (ret)
			break;

		remain -= length;
		user_data += length;
		offset = 0;
	}

	i915_gem_object_flush_frontbuffer(obj, ORIGIN_CPU);

	i915_gem_object_unpin_pages(obj);
	return ret;

err_unpin:
	i915_gem_object_unpin_pages(obj);
err_unlock:
	i915_gem_object_unlock(obj);
	return ret;
}

/**
 * Writes data to the object referenced by handle.
 * @dev: drm device
 * @data: ioctl data blob
 * @file: drm file
 *
 * On error, the contents of the buffer that were to be modified are undefined.
 */
int
i915_gem_pwrite_ioctl(struct drm_device *dev, void *data,
		      struct drm_file *file)
{
	struct drm_i915_private *i915 = to_i915(dev);
	struct drm_i915_gem_pwrite *args = data;
	struct drm_i915_gem_object *obj;
	int ret;

	/* PWRITE is disallowed for all platforms after TGL-LP.  This also
	 * covers all platforms with local memory.
	 */
	if (INTEL_GEN(i915) >= 12 && !IS_TIGERLAKE(i915))
		return -EOPNOTSUPP;

	if (args->size == 0)
		return 0;

	if (!access_ok(u64_to_user_ptr(args->data_ptr), args->size))
		return -EFAULT;

	obj = i915_gem_object_lookup(file, args->handle);
	if (!obj)
		return -ENOENT;

	/* Bounds check destination. */
	if (range_overflows_t(u64, args->offset, args->size, obj->base.size)) {
		ret = -EINVAL;
		goto err;
	}

	/* Writes not allowed into this read-only object */
	if (i915_gem_object_is_readonly(obj)) {
		ret = -EINVAL;
		goto err;
	}

	trace_i915_gem_object_pwrite(obj, args->offset, args->size);

	ret = -ENODEV;
	if (obj->ops->pwrite)
		ret = obj->ops->pwrite(obj, args);
	if (ret != -ENODEV)
		goto err;

	ret = i915_gem_object_wait(obj,
				   I915_WAIT_INTERRUPTIBLE |
				   I915_WAIT_ALL,
				   MAX_SCHEDULE_TIMEOUT);
	if (ret)
		goto err;

	ret = -EFAULT;
	/* We can only do the GTT pwrite on untiled buffers, as otherwise
	 * it would end up going through the fenced access, and we'll get
	 * different detiling behavior between reading and writing.
	 * pread/pwrite currently are reading and writing from the CPU
	 * perspective, requiring manual detiling by the client.
	 */
	if (!i915_gem_object_has_struct_page(obj) ||
	    cpu_write_needs_clflush(obj))
		/* Note that the gtt paths might fail with non-page-backed user
		 * pointers (e.g. gtt mappings when moving data between
		 * textures). Fallback to the shmem path in that case.
		 */
		ret = i915_gem_gtt_pwrite_fast(obj, args);

	if (ret == -EFAULT || ret == -ENOSPC) {
		if (i915_gem_object_has_struct_page(obj))
			ret = i915_gem_shmem_pwrite(obj, args);
	}

err:
	i915_gem_object_put(obj);
	return ret;
}

/**
 * Called when user space has done writes to this buffer
 * @dev: drm device
 * @data: ioctl data blob
 * @file: drm file
 */
int
i915_gem_sw_finish_ioctl(struct drm_device *dev, void *data,
			 struct drm_file *file)
{
	struct drm_i915_gem_sw_finish *args = data;
	struct drm_i915_gem_object *obj;

	obj = i915_gem_object_lookup(file, args->handle);
	if (!obj)
		return -ENOENT;

	/*
	 * Proxy objects are barred from CPU access, so there is no
	 * need to ban sw_finish as it is a nop.
	 */

	/* Pinned buffers may be scanout, so flush the cache */
	i915_gem_object_flush_if_display(obj);
	i915_gem_object_put(obj);

	return 0;
}

void i915_gem_runtime_suspend(struct drm_i915_private *i915)
{
	struct drm_i915_gem_object *obj, *on;
	int i;

	/*
	 * Only called during RPM suspend. All users of the userfault_list
	 * must be holding an RPM wakeref to ensure that this can not
	 * run concurrently with themselves (and use the struct_mutex for
	 * protection between themselves).
	 */

	list_for_each_entry_safe(obj, on,
				 &i915->ggtt.userfault_list, userfault_link)
		__i915_gem_object_release_mmap_gtt(obj);

	/*
	 * The fence will be lost when the device powers down. If any were
	 * in use by hardware (i.e. they are pinned), we should not be powering
	 * down! All other fences will be reacquired by the user upon waking.
	 */
	for (i = 0; i < i915->ggtt.num_fences; i++) {
		struct i915_fence_reg *reg = &i915->ggtt.fence_regs[i];

		/*
		 * Ideally we want to assert that the fence register is not
		 * live at this point (i.e. that no piece of code will be
		 * trying to write through fence + GTT, as that both violates
		 * our tracking of activity and associated locking/barriers,
		 * but also is illegal given that the hw is powered down).
		 *
		 * Previously we used reg->pin_count as a "liveness" indicator.
		 * That is not sufficient, and we need a more fine-grained
		 * tool if we want to have a sanity check here.
		 */

		if (!reg->vma)
			continue;

		GEM_BUG_ON(i915_vma_has_userfault(reg->vma));
		reg->dirty = true;
	}
}

static void discard_ggtt_vma(struct i915_vma *vma)
{
	struct drm_i915_gem_object *obj = vma->obj;

	spin_lock(&obj->vma.lock);
	if (!RB_EMPTY_NODE(&vma->obj_node)) {
		rb_erase(&vma->obj_node, &obj->vma.tree);
		RB_CLEAR_NODE(&vma->obj_node);
	}
	spin_unlock(&obj->vma.lock);
}

struct i915_vma *
i915_gem_object_ggtt_pin_ww(struct drm_i915_gem_object *obj,
			    struct i915_gem_ww_ctx *ww,
			    const struct i915_ggtt_view *view,
			    u64 size, u64 alignment, u64 flags)
{
	struct drm_i915_private *i915 = to_i915(obj->base.dev);
	struct i915_ggtt *ggtt = &i915->ggtt;
	struct i915_vma *vma;
	int ret;

	if (flags & PIN_MAPPABLE &&
	    (!view || view->type == I915_GGTT_VIEW_NORMAL)) {
		/*
		 * If the required space is larger than the available
		 * aperture, we will not able to find a slot for the
		 * object and unbinding the object now will be in
		 * vain. Worse, doing so may cause us to ping-pong
		 * the object in and out of the Global GTT and
		 * waste a lot of cycles under the mutex.
		 */
		if (obj->base.size > ggtt->mappable_end)
			return ERR_PTR(-E2BIG);

		/*
		 * If NONBLOCK is set the caller is optimistically
		 * trying to cache the full object within the mappable
		 * aperture, and *must* have a fallback in place for
		 * situations where we cannot bind the object. We
		 * can be a little more lax here and use the fallback
		 * more often to avoid costly migrations of ourselves
		 * and other objects within the aperture.
		 *
		 * Half-the-aperture is used as a simple heuristic.
		 * More interesting would to do search for a free
		 * block prior to making the commitment to unbind.
		 * That caters for the self-harm case, and with a
		 * little more heuristics (e.g. NOFAULT, NOEVICT)
		 * we could try to minimise harm to others.
		 */
		if (flags & PIN_NONBLOCK &&
		    obj->base.size > ggtt->mappable_end / 2)
			return ERR_PTR(-ENOSPC);
	}

new_vma:
	vma = i915_vma_instance(obj, &ggtt->vm, view);
	if (IS_ERR(vma))
		return vma;

	if (i915_vma_misplaced(vma, size, alignment, flags)) {
		if (flags & PIN_NONBLOCK) {
			if (i915_vma_is_pinned(vma) || i915_vma_is_active(vma))
				return ERR_PTR(-ENOSPC);

			if (flags & PIN_MAPPABLE &&
			    vma->fence_size > ggtt->mappable_end / 2)
				return ERR_PTR(-ENOSPC);
		}

		if (i915_vma_is_pinned(vma) || i915_vma_is_active(vma)) {
			discard_ggtt_vma(vma);
			goto new_vma;
		}

		ret = i915_vma_unbind(vma);
		if (ret)
			return ERR_PTR(ret);
	}

	if (ww)
		ret = i915_vma_pin_ww(vma, ww, size, alignment, flags | PIN_GLOBAL);
	else
		ret = i915_vma_pin(vma, size, alignment, flags | PIN_GLOBAL);

	if (ret)
		return ERR_PTR(ret);

	if (vma->fence && !i915_gem_object_is_tiled(obj)) {
		mutex_lock(&ggtt->vm.mutex);
		i915_vma_revoke_fence(vma);
		mutex_unlock(&ggtt->vm.mutex);
	}

	ret = i915_vma_wait_for_bind(vma);
	if (ret) {
		i915_vma_unpin(vma);
		return ERR_PTR(ret);
	}

	return vma;
}

int
i915_gem_madvise_ioctl(struct drm_device *dev, void *data,
		       struct drm_file *file_priv)
{
	struct drm_i915_private *i915 = to_i915(dev);
	struct drm_i915_gem_madvise *args = data;
	struct drm_i915_gem_object *obj;
	int err;

	switch (args->madv) {
	case I915_MADV_DONTNEED:
	case I915_MADV_WILLNEED:
	    break;
	default:
	    return -EINVAL;
	}

	obj = i915_gem_object_lookup(file_priv, args->handle);
	if (!obj)
		return -ENOENT;

	err = i915_gem_object_lock_interruptible(obj, NULL);
	if (err)
		goto out;

	if (i915_gem_object_has_pages(obj) &&
	    i915_gem_object_is_tiled(obj) &&
	    i915->quirks & QUIRK_PIN_SWIZZLED_PAGES) {
		if (obj->mm.madv == I915_MADV_WILLNEED) {
			GEM_BUG_ON(!i915_gem_object_has_tiling_quirk(obj));
			i915_gem_object_clear_tiling_quirk(obj);
			i915_gem_object_make_shrinkable(obj);
		}
		if (args->madv == I915_MADV_WILLNEED) {
			GEM_BUG_ON(i915_gem_object_has_tiling_quirk(obj));
			i915_gem_object_make_unshrinkable(obj);
			i915_gem_object_set_tiling_quirk(obj);
		}
	}

	if (obj->mm.madv != __I915_MADV_PURGED)
		obj->mm.madv = args->madv;

	if (i915_gem_object_has_pages(obj)) {
		unsigned long flags;

		spin_lock_irqsave(&i915->mm.obj_lock, flags);
		if (!list_empty(&obj->mm.link)) {
			struct list_head *list;

			if (obj->mm.madv != I915_MADV_WILLNEED)
				list = &i915->mm.purge_list;
			else
				list = &i915->mm.shrink_list;
			list_move_tail(&obj->mm.link, list);

		}
		spin_unlock_irqrestore(&i915->mm.obj_lock, flags);
	}

	/* if the object is no longer attached, discard its backing storage */
	if (obj->mm.madv == I915_MADV_DONTNEED &&
	    !i915_gem_object_has_pages(obj))
		i915_gem_object_truncate(obj);

	args->retained = obj->mm.madv != __I915_MADV_PURGED;

	i915_gem_object_unlock(obj);
out:
	i915_gem_object_put(obj);
	return err;
}

int i915_gem_init(struct drm_i915_private *dev_priv)
{
	int ret;

	/* We need to fallback to 4K pages if host doesn't support huge gtt. */
	if (intel_vgpu_active(dev_priv) && !intel_vgpu_has_huge_gtt(dev_priv))
		mkwrite_device_info(dev_priv)->page_sizes =
			I915_GTT_PAGE_SIZE_4K;

	ret = i915_gem_init_userptr(dev_priv);
	if (ret)
		return ret;

	intel_uc_fetch_firmwares(&dev_priv->gt.uc);
	intel_wopcm_init(&dev_priv->wopcm);

	ret = i915_init_ggtt(dev_priv);
	if (ret) {
		GEM_BUG_ON(ret == -EIO);
		goto err_unlock;
	}

	/*
	 * Despite its name intel_init_clock_gating applies both display
	 * clock gating workarounds; GT mmio workarounds and the occasional
	 * GT power context workaround. Worse, sometimes it includes a context
	 * register workaround which we need to apply before we record the
	 * default HW state for all contexts.
	 *
	 * FIXME: break up the workarounds and apply them at the right time!
	 */
	intel_init_clock_gating(dev_priv);

	ret = intel_gt_init(&dev_priv->gt);
	if (ret)
		goto err_unlock;

	return 0;

	/*
	 * Unwinding is complicated by that we want to handle -EIO to mean
	 * disable GPU submission but keep KMS alive. We want to mark the
	 * HW as irrevisibly wedged, but keep enough state around that the
	 * driver doesn't explode during runtime.
	 */
err_unlock:
	i915_gem_drain_workqueue(dev_priv);

	if (ret != -EIO)
		intel_uc_cleanup_firmwares(&dev_priv->gt.uc);

	if (ret == -EIO) {
		/*
		 * Allow engines or uC initialisation to fail by marking the GPU
		 * as wedged. But we only want to do this when the GPU is angry,
		 * for all other failure, such as an allocation failure, bail.
		 */
		if (!intel_gt_is_wedged(&dev_priv->gt)) {
			i915_probe_error(dev_priv,
					 "Failed to initialize GPU, declaring it wedged!\n");
			intel_gt_set_wedged(&dev_priv->gt);
		}

		/* Minimal basic recovery for KMS */
		ret = i915_ggtt_enable_hw(dev_priv);
		i915_ggtt_resume(&dev_priv->ggtt);
		intel_init_clock_gating(dev_priv);
	}

	i915_gem_drain_freed_objects(dev_priv);
	return ret;
}

void i915_gem_driver_register(struct drm_i915_private *i915)
{
	i915_gem_driver_register__shrinker(i915);

	intel_engines_driver_register(i915);
}

void i915_gem_driver_unregister(struct drm_i915_private *i915)
{
	i915_gem_driver_unregister__shrinker(i915);
}

void i915_gem_driver_remove(struct drm_i915_private *dev_priv)
{
	intel_wakeref_auto_fini(&dev_priv->ggtt.userfault_wakeref);

	i915_gem_suspend_late(dev_priv);
	intel_gt_driver_remove(&dev_priv->gt);
	dev_priv->uabi_engines = RB_ROOT;

	/* Flush any outstanding unpin_work. */
	i915_gem_drain_workqueue(dev_priv);

	i915_gem_drain_freed_objects(dev_priv);
}

void i915_gem_driver_release(struct drm_i915_private *dev_priv)
{
	intel_gt_driver_release(&dev_priv->gt);

	intel_wa_list_free(&dev_priv->gt_wa_list);

	intel_uc_cleanup_firmwares(&dev_priv->gt.uc);

	i915_gem_drain_freed_objects(dev_priv);

	drm_WARN_ON(&dev_priv->drm, !list_empty(&dev_priv->gem.contexts.list));
}

static void i915_gem_init__mm(struct drm_i915_private *i915)
{
	spin_lock_init(&i915->mm.obj_lock);

	init_llist_head(&i915->mm.free_list);

	INIT_LIST_HEAD(&i915->mm.purge_list);
	INIT_LIST_HEAD(&i915->mm.shrink_list);

	i915_gem_init__objects(i915);
}

void i915_gem_init_early(struct drm_i915_private *dev_priv)
{
	i915_gem_init__mm(dev_priv);
	i915_gem_init__contexts(dev_priv);

	spin_lock_init(&dev_priv->fb_tracking.lock);
}

void i915_gem_cleanup_early(struct drm_i915_private *dev_priv)
{
	i915_gem_drain_freed_objects(dev_priv);
	GEM_BUG_ON(!llist_empty(&dev_priv->mm.free_list));
	GEM_BUG_ON(atomic_read(&dev_priv->mm.free_count));
	drm_WARN_ON(&dev_priv->drm, dev_priv->mm.shrink_count);
}

int i915_gem_open(struct drm_i915_private *i915, struct drm_file *file)
{
	struct drm_i915_file_private *file_priv;
	int ret;

	DRM_DEBUG("\n");

	file_priv = kzalloc(sizeof(*file_priv), GFP_KERNEL);
	if (!file_priv)
		return -ENOMEM;

	file->driver_priv = file_priv;
	file_priv->dev_priv = i915;
	file_priv->file = file;

	file_priv->bsd_engine = -1;
	file_priv->hang_timestamp = jiffies;

	ret = i915_gem_context_open(i915, file);
	if (ret)
		kfree(file_priv);

	return ret;
}

void i915_gem_ww_ctx_init(struct i915_gem_ww_ctx *ww, bool intr)
{
	ww_acquire_init(&ww->ctx, &reservation_ww_class);
	INIT_LIST_HEAD(&ww->obj_list);
	ww->intr = intr;
	ww->contended = NULL;
}

static void i915_gem_ww_ctx_unlock_all(struct i915_gem_ww_ctx *ww)
{
	struct drm_i915_gem_object *obj;

	while ((obj = list_first_entry_or_null(&ww->obj_list, struct drm_i915_gem_object, obj_link))) {
		list_del(&obj->obj_link);
		i915_gem_object_unlock(obj);
	}
}

void i915_gem_ww_unlock_single(struct drm_i915_gem_object *obj)
{
	list_del(&obj->obj_link);
	i915_gem_object_unlock(obj);
}

void i915_gem_ww_ctx_fini(struct i915_gem_ww_ctx *ww)
{
	i915_gem_ww_ctx_unlock_all(ww);
	WARN_ON(ww->contended);
	ww_acquire_fini(&ww->ctx);
}

int __must_check i915_gem_ww_ctx_backoff(struct i915_gem_ww_ctx *ww)
{
	int ret = 0;

	if (WARN_ON(!ww->contended))
		return -EINVAL;

	i915_gem_ww_ctx_unlock_all(ww);
	if (ww->intr)
		ret = dma_resv_lock_slow_interruptible(ww->contended->base.resv, &ww->ctx);
	else
		dma_resv_lock_slow(ww->contended->base.resv, &ww->ctx);

	if (!ret)
		list_add_tail(&ww->contended->obj_link, &ww->obj_list);

	ww->contended = NULL;

	return ret;
}

#if IS_ENABLED(CONFIG_DRM_I915_SELFTEST)
#include "selftests/mock_gem_device.c"
#include "selftests/i915_gem.c"
#endif