summaryrefslogtreecommitdiff
path: root/drivers/gpu/drm/xe/xe_device.c
blob: 70d4e4d46c3c86592f0e984d725778090af99222 (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
// SPDX-License-Identifier: MIT
/*
 * Copyright © 2021 Intel Corporation
 */

#include "xe_device.h"

#include <linux/delay.h>
#include <linux/units.h>

#include <drm/drm_aperture.h>
#include <drm/drm_atomic_helper.h>
#include <drm/drm_client.h>
#include <drm/drm_gem_ttm_helper.h>
#include <drm/drm_ioctl.h>
#include <drm/drm_managed.h>
#include <drm/drm_print.h>
#include <uapi/drm/xe_drm.h>

#include "display/xe_display.h"
#include "instructions/xe_gpu_commands.h"
#include "regs/xe_gt_regs.h"
#include "regs/xe_regs.h"
#include "xe_bo.h"
#include "xe_debugfs.h"
#include "xe_devcoredump.h"
#include "xe_dma_buf.h"
#include "xe_drm_client.h"
#include "xe_drv.h"
#include "xe_exec.h"
#include "xe_exec_queue.h"
#include "xe_force_wake.h"
#include "xe_ggtt.h"
#include "xe_gsc_proxy.h"
#include "xe_gt.h"
#include "xe_gt_mcr.h"
#include "xe_gt_printk.h"
#include "xe_gt_sriov_vf.h"
#include "xe_guc.h"
#include "xe_hw_engine_group.h"
#include "xe_hwmon.h"
#include "xe_irq.h"
#include "xe_memirq.h"
#include "xe_mmio.h"
#include "xe_module.h"
#include "xe_observation.h"
#include "xe_pat.h"
#include "xe_pcode.h"
#include "xe_pm.h"
#include "xe_query.h"
#include "xe_sriov.h"
#include "xe_tile.h"
#include "xe_ttm_stolen_mgr.h"
#include "xe_ttm_sys_mgr.h"
#include "xe_vm.h"
#include "xe_vram.h"
#include "xe_wait_user_fence.h"
#include "xe_wa.h"

#include <generated/xe_wa_oob.h>

static int xe_file_open(struct drm_device *dev, struct drm_file *file)
{
	struct xe_device *xe = to_xe_device(dev);
	struct xe_drm_client *client;
	struct xe_file *xef;
	int ret = -ENOMEM;
	struct task_struct *task = NULL;

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

	client = xe_drm_client_alloc();
	if (!client) {
		kfree(xef);
		return ret;
	}

	xef->drm = file;
	xef->client = client;
	xef->xe = xe;

	mutex_init(&xef->vm.lock);
	xa_init_flags(&xef->vm.xa, XA_FLAGS_ALLOC1);

	mutex_init(&xef->exec_queue.lock);
	xa_init_flags(&xef->exec_queue.xa, XA_FLAGS_ALLOC1);

	spin_lock(&xe->clients.lock);
	xe->clients.count++;
	spin_unlock(&xe->clients.lock);

	file->driver_priv = xef;
	kref_init(&xef->refcount);

	task = get_pid_task(rcu_access_pointer(file->pid), PIDTYPE_PID);
	if (task) {
		xef->process_name = kstrdup(task->comm, GFP_KERNEL);
		xef->pid = task->pid;
		put_task_struct(task);
	}

	return 0;
}

static void xe_file_destroy(struct kref *ref)
{
	struct xe_file *xef = container_of(ref, struct xe_file, refcount);
	struct xe_device *xe = xef->xe;

	xa_destroy(&xef->exec_queue.xa);
	mutex_destroy(&xef->exec_queue.lock);
	xa_destroy(&xef->vm.xa);
	mutex_destroy(&xef->vm.lock);

	spin_lock(&xe->clients.lock);
	xe->clients.count--;
	spin_unlock(&xe->clients.lock);

	xe_drm_client_put(xef->client);
	kfree(xef->process_name);
	kfree(xef);
}

/**
 * xe_file_get() - Take a reference to the xe file object
 * @xef: Pointer to the xe file
 *
 * Anyone with a pointer to xef must take a reference to the xe file
 * object using this call.
 *
 * Return: xe file pointer
 */
struct xe_file *xe_file_get(struct xe_file *xef)
{
	kref_get(&xef->refcount);
	return xef;
}

/**
 * xe_file_put() - Drop a reference to the xe file object
 * @xef: Pointer to the xe file
 *
 * Used to drop reference to the xef object
 */
void xe_file_put(struct xe_file *xef)
{
	kref_put(&xef->refcount, xe_file_destroy);
}

static void xe_file_close(struct drm_device *dev, struct drm_file *file)
{
	struct xe_device *xe = to_xe_device(dev);
	struct xe_file *xef = file->driver_priv;
	struct xe_vm *vm;
	struct xe_exec_queue *q;
	unsigned long idx;

	xe_pm_runtime_get(xe);

	/*
	 * No need for exec_queue.lock here as there is no contention for it
	 * when FD is closing as IOCTLs presumably can't be modifying the
	 * xarray. Taking exec_queue.lock here causes undue dependency on
	 * vm->lock taken during xe_exec_queue_kill().
	 */
	xa_for_each(&xef->exec_queue.xa, idx, q) {
		if (q->vm && q->hwe->hw_engine_group)
			xe_hw_engine_group_del_exec_queue(q->hwe->hw_engine_group, q);
		xe_exec_queue_kill(q);
		xe_exec_queue_put(q);
	}
	mutex_lock(&xef->vm.lock);
	xa_for_each(&xef->vm.xa, idx, vm)
		xe_vm_close_and_put(vm);
	mutex_unlock(&xef->vm.lock);

	xe_file_put(xef);

	xe_pm_runtime_put(xe);
}

static const struct drm_ioctl_desc xe_ioctls[] = {
	DRM_IOCTL_DEF_DRV(XE_DEVICE_QUERY, xe_query_ioctl, DRM_RENDER_ALLOW),
	DRM_IOCTL_DEF_DRV(XE_GEM_CREATE, xe_gem_create_ioctl, DRM_RENDER_ALLOW),
	DRM_IOCTL_DEF_DRV(XE_GEM_MMAP_OFFSET, xe_gem_mmap_offset_ioctl,
			  DRM_RENDER_ALLOW),
	DRM_IOCTL_DEF_DRV(XE_VM_CREATE, xe_vm_create_ioctl, DRM_RENDER_ALLOW),
	DRM_IOCTL_DEF_DRV(XE_VM_DESTROY, xe_vm_destroy_ioctl, DRM_RENDER_ALLOW),
	DRM_IOCTL_DEF_DRV(XE_VM_BIND, xe_vm_bind_ioctl, DRM_RENDER_ALLOW),
	DRM_IOCTL_DEF_DRV(XE_EXEC, xe_exec_ioctl, DRM_RENDER_ALLOW),
	DRM_IOCTL_DEF_DRV(XE_EXEC_QUEUE_CREATE, xe_exec_queue_create_ioctl,
			  DRM_RENDER_ALLOW),
	DRM_IOCTL_DEF_DRV(XE_EXEC_QUEUE_DESTROY, xe_exec_queue_destroy_ioctl,
			  DRM_RENDER_ALLOW),
	DRM_IOCTL_DEF_DRV(XE_EXEC_QUEUE_GET_PROPERTY, xe_exec_queue_get_property_ioctl,
			  DRM_RENDER_ALLOW),
	DRM_IOCTL_DEF_DRV(XE_WAIT_USER_FENCE, xe_wait_user_fence_ioctl,
			  DRM_RENDER_ALLOW),
	DRM_IOCTL_DEF_DRV(XE_OBSERVATION, xe_observation_ioctl, DRM_RENDER_ALLOW),
};

static long xe_drm_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
{
	struct drm_file *file_priv = file->private_data;
	struct xe_device *xe = to_xe_device(file_priv->minor->dev);
	long ret;

	if (xe_device_wedged(xe))
		return -ECANCELED;

	ret = xe_pm_runtime_get_ioctl(xe);
	if (ret >= 0)
		ret = drm_ioctl(file, cmd, arg);
	xe_pm_runtime_put(xe);

	return ret;
}

#ifdef CONFIG_COMPAT
static long xe_drm_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
{
	struct drm_file *file_priv = file->private_data;
	struct xe_device *xe = to_xe_device(file_priv->minor->dev);
	long ret;

	if (xe_device_wedged(xe))
		return -ECANCELED;

	ret = xe_pm_runtime_get_ioctl(xe);
	if (ret >= 0)
		ret = drm_compat_ioctl(file, cmd, arg);
	xe_pm_runtime_put(xe);

	return ret;
}
#else
/* similarly to drm_compat_ioctl, let's it be assigned to .compat_ioct unconditionally */
#define xe_drm_compat_ioctl NULL
#endif

static const struct file_operations xe_driver_fops = {
	.owner = THIS_MODULE,
	.open = drm_open,
	.release = drm_release_noglobal,
	.unlocked_ioctl = xe_drm_ioctl,
	.mmap = drm_gem_mmap,
	.poll = drm_poll,
	.read = drm_read,
	.compat_ioctl = xe_drm_compat_ioctl,
	.llseek = noop_llseek,
#ifdef CONFIG_PROC_FS
	.show_fdinfo = drm_show_fdinfo,
#endif
	.fop_flags = FOP_UNSIGNED_OFFSET,
};

static struct drm_driver driver = {
	/* Don't use MTRRs here; the Xserver or userspace app should
	 * deal with them for Intel hardware.
	 */
	.driver_features =
	    DRIVER_GEM |
	    DRIVER_RENDER | DRIVER_SYNCOBJ |
	    DRIVER_SYNCOBJ_TIMELINE | DRIVER_GEM_GPUVA,
	.open = xe_file_open,
	.postclose = xe_file_close,

	.gem_prime_import = xe_gem_prime_import,

	.dumb_create = xe_bo_dumb_create,
	.dumb_map_offset = drm_gem_ttm_dumb_map_offset,
#ifdef CONFIG_PROC_FS
	.show_fdinfo = xe_drm_client_fdinfo,
#endif
	.ioctls = xe_ioctls,
	.num_ioctls = ARRAY_SIZE(xe_ioctls),
	.fops = &xe_driver_fops,
	.name = DRIVER_NAME,
	.desc = DRIVER_DESC,
	.date = DRIVER_DATE,
	.major = DRIVER_MAJOR,
	.minor = DRIVER_MINOR,
	.patchlevel = DRIVER_PATCHLEVEL,
};

static void xe_device_destroy(struct drm_device *dev, void *dummy)
{
	struct xe_device *xe = to_xe_device(dev);

	if (xe->preempt_fence_wq)
		destroy_workqueue(xe->preempt_fence_wq);

	if (xe->ordered_wq)
		destroy_workqueue(xe->ordered_wq);

	if (xe->unordered_wq)
		destroy_workqueue(xe->unordered_wq);

	ttm_device_fini(&xe->ttm);
}

struct xe_device *xe_device_create(struct pci_dev *pdev,
				   const struct pci_device_id *ent)
{
	struct xe_device *xe;
	int err;

	xe_display_driver_set_hooks(&driver);

	err = drm_aperture_remove_conflicting_pci_framebuffers(pdev, &driver);
	if (err)
		return ERR_PTR(err);

	xe = devm_drm_dev_alloc(&pdev->dev, &driver, struct xe_device, drm);
	if (IS_ERR(xe))
		return xe;

	err = ttm_device_init(&xe->ttm, &xe_ttm_funcs, xe->drm.dev,
			      xe->drm.anon_inode->i_mapping,
			      xe->drm.vma_offset_manager, false, false);
	if (WARN_ON(err))
		goto err;

	err = drmm_add_action_or_reset(&xe->drm, xe_device_destroy, NULL);
	if (err)
		goto err;

	xe->info.devid = pdev->device;
	xe->info.revid = pdev->revision;
	xe->info.force_execlist = xe_modparam.force_execlist;

	spin_lock_init(&xe->irq.lock);
	spin_lock_init(&xe->clients.lock);

	init_waitqueue_head(&xe->ufence_wq);

	err = drmm_mutex_init(&xe->drm, &xe->usm.lock);
	if (err)
		goto err;

	xa_init_flags(&xe->usm.asid_to_vm, XA_FLAGS_ALLOC);

	if (IS_ENABLED(CONFIG_DRM_XE_DEBUG)) {
		/* Trigger a large asid and an early asid wrap. */
		u32 asid;

		BUILD_BUG_ON(XE_MAX_ASID < 2);
		err = xa_alloc_cyclic(&xe->usm.asid_to_vm, &asid, NULL,
				      XA_LIMIT(XE_MAX_ASID - 2, XE_MAX_ASID - 1),
				      &xe->usm.next_asid, GFP_KERNEL);
		drm_WARN_ON(&xe->drm, err);
		if (err >= 0)
			xa_erase(&xe->usm.asid_to_vm, asid);
	}

	spin_lock_init(&xe->pinned.lock);
	INIT_LIST_HEAD(&xe->pinned.kernel_bo_present);
	INIT_LIST_HEAD(&xe->pinned.external_vram);
	INIT_LIST_HEAD(&xe->pinned.evicted);

	xe->preempt_fence_wq = alloc_ordered_workqueue("xe-preempt-fence-wq", 0);
	xe->ordered_wq = alloc_ordered_workqueue("xe-ordered-wq", 0);
	xe->unordered_wq = alloc_workqueue("xe-unordered-wq", 0, 0);
	if (!xe->ordered_wq || !xe->unordered_wq ||
	    !xe->preempt_fence_wq) {
		/*
		 * Cleanup done in xe_device_destroy via
		 * drmm_add_action_or_reset register above
		 */
		drm_err(&xe->drm, "Failed to allocate xe workqueues\n");
		err = -ENOMEM;
		goto err;
	}

	err = xe_display_create(xe);
	if (WARN_ON(err))
		goto err;

	return xe;

err:
	return ERR_PTR(err);
}

/*
 * The driver-initiated FLR is the highest level of reset that we can trigger
 * from within the driver. It is different from the PCI FLR in that it doesn't
 * fully reset the SGUnit and doesn't modify the PCI config space and therefore
 * it doesn't require a re-enumeration of the PCI BARs. However, the
 * driver-initiated FLR does still cause a reset of both GT and display and a
 * memory wipe of local and stolen memory, so recovery would require a full HW
 * re-init and saving/restoring (or re-populating) the wiped memory. Since we
 * perform the FLR as the very last action before releasing access to the HW
 * during the driver release flow, we don't attempt recovery at all, because
 * if/when a new instance of i915 is bound to the device it will do a full
 * re-init anyway.
 */
static void xe_driver_flr(struct xe_device *xe)
{
	const unsigned int flr_timeout = 3 * MICRO; /* specs recommend a 3s wait */
	struct xe_gt *gt = xe_root_mmio_gt(xe);
	int ret;

	if (xe_mmio_read32(gt, GU_CNTL_PROTECTED) & DRIVERINT_FLR_DIS) {
		drm_info_once(&xe->drm, "BIOS Disabled Driver-FLR\n");
		return;
	}

	drm_dbg(&xe->drm, "Triggering Driver-FLR\n");

	/*
	 * Make sure any pending FLR requests have cleared by waiting for the
	 * FLR trigger bit to go to zero. Also clear GU_DEBUG's DRIVERFLR_STATUS
	 * to make sure it's not still set from a prior attempt (it's a write to
	 * clear bit).
	 * Note that we should never be in a situation where a previous attempt
	 * is still pending (unless the HW is totally dead), but better to be
	 * safe in case something unexpected happens
	 */
	ret = xe_mmio_wait32(gt, GU_CNTL, DRIVERFLR, 0, flr_timeout, NULL, false);
	if (ret) {
		drm_err(&xe->drm, "Driver-FLR-prepare wait for ready failed! %d\n", ret);
		return;
	}
	xe_mmio_write32(gt, GU_DEBUG, DRIVERFLR_STATUS);

	/* Trigger the actual Driver-FLR */
	xe_mmio_rmw32(gt, GU_CNTL, 0, DRIVERFLR);

	/* Wait for hardware teardown to complete */
	ret = xe_mmio_wait32(gt, GU_CNTL, DRIVERFLR, 0, flr_timeout, NULL, false);
	if (ret) {
		drm_err(&xe->drm, "Driver-FLR-teardown wait completion failed! %d\n", ret);
		return;
	}

	/* Wait for hardware/firmware re-init to complete */
	ret = xe_mmio_wait32(gt, GU_DEBUG, DRIVERFLR_STATUS, DRIVERFLR_STATUS,
			     flr_timeout, NULL, false);
	if (ret) {
		drm_err(&xe->drm, "Driver-FLR-reinit wait completion failed! %d\n", ret);
		return;
	}

	/* Clear sticky completion status */
	xe_mmio_write32(gt, GU_DEBUG, DRIVERFLR_STATUS);
}

static void xe_driver_flr_fini(void *arg)
{
	struct xe_device *xe = arg;

	if (xe->needs_flr_on_fini)
		xe_driver_flr(xe);
}

static void xe_device_sanitize(void *arg)
{
	struct xe_device *xe = arg;
	struct xe_gt *gt;
	u8 id;

	for_each_gt(gt, xe, id)
		xe_gt_sanitize(gt);
}

static int xe_set_dma_info(struct xe_device *xe)
{
	unsigned int mask_size = xe->info.dma_mask_size;
	int err;

	dma_set_max_seg_size(xe->drm.dev, xe_sg_segment_size(xe->drm.dev));

	err = dma_set_mask(xe->drm.dev, DMA_BIT_MASK(mask_size));
	if (err)
		goto mask_err;

	err = dma_set_coherent_mask(xe->drm.dev, DMA_BIT_MASK(mask_size));
	if (err)
		goto mask_err;

	return 0;

mask_err:
	drm_err(&xe->drm, "Can't set DMA mask/consistent mask (%d)\n", err);
	return err;
}

static bool verify_lmem_ready(struct xe_gt *gt)
{
	u32 val = xe_mmio_read32(gt, GU_CNTL) & LMEM_INIT;

	return !!val;
}

static int wait_for_lmem_ready(struct xe_device *xe)
{
	struct xe_gt *gt = xe_root_mmio_gt(xe);
	unsigned long timeout, start;

	if (!IS_DGFX(xe))
		return 0;

	if (IS_SRIOV_VF(xe))
		return 0;

	if (verify_lmem_ready(gt))
		return 0;

	drm_dbg(&xe->drm, "Waiting for lmem initialization\n");

	start = jiffies;
	timeout = start + msecs_to_jiffies(60 * 1000); /* 60 sec! */

	do {
		if (signal_pending(current))
			return -EINTR;

		/*
		 * The boot firmware initializes local memory and
		 * assesses its health. If memory training fails,
		 * the punit will have been instructed to keep the GT powered
		 * down.we won't be able to communicate with it
		 *
		 * If the status check is done before punit updates the register,
		 * it can lead to the system being unusable.
		 * use a timeout and defer the probe to prevent this.
		 */
		if (time_after(jiffies, timeout)) {
			drm_dbg(&xe->drm, "lmem not initialized by firmware\n");
			return -EPROBE_DEFER;
		}

		msleep(20);

	} while (!verify_lmem_ready(gt));

	drm_dbg(&xe->drm, "lmem ready after %ums",
		jiffies_to_msecs(jiffies - start));

	return 0;
}

static void update_device_info(struct xe_device *xe)
{
	/* disable features that are not available/applicable to VFs */
	if (IS_SRIOV_VF(xe)) {
		xe->info.probe_display = 0;
		xe->info.has_heci_gscfi = 0;
		xe->info.skip_guc_pc = 1;
		xe->info.skip_pcode = 1;
	}
}

/**
 * xe_device_probe_early: Device early probe
 * @xe: xe device instance
 *
 * Initialize MMIO resources that don't require any
 * knowledge about tile count. Also initialize pcode and
 * check vram initialization on root tile.
 *
 * Return: 0 on success, error code on failure
 */
int xe_device_probe_early(struct xe_device *xe)
{
	int err;

	err = xe_mmio_init(xe);
	if (err)
		return err;

	xe_sriov_probe_early(xe);

	update_device_info(xe);

	err = xe_pcode_probe_early(xe);
	if (err)
		return err;

	err = wait_for_lmem_ready(xe);
	if (err)
		return err;

	xe->wedged.mode = xe_modparam.wedged_mode;

	return 0;
}

static int xe_device_set_has_flat_ccs(struct  xe_device *xe)
{
	u32 reg;
	int err;

	if (GRAPHICS_VER(xe) < 20 || !xe->info.has_flat_ccs)
		return 0;

	struct xe_gt *gt = xe_root_mmio_gt(xe);

	err = xe_force_wake_get(gt_to_fw(gt), XE_FW_GT);
	if (err)
		return err;

	reg = xe_gt_mcr_unicast_read_any(gt, XE2_FLAT_CCS_BASE_RANGE_LOWER);
	xe->info.has_flat_ccs = (reg & XE2_FLAT_CCS_ENABLE);

	if (!xe->info.has_flat_ccs)
		drm_dbg(&xe->drm,
			"Flat CCS has been disabled in bios, May lead to performance impact");

	return xe_force_wake_put(gt_to_fw(gt), XE_FW_GT);
}

int xe_device_probe(struct xe_device *xe)
{
	struct xe_tile *tile;
	struct xe_gt *gt;
	int err;
	u8 last_gt;
	u8 id;

	xe_pat_init_early(xe);

	err = xe_sriov_init(xe);
	if (err)
		return err;

	xe->info.mem_region_mask = 1;
	err = xe_display_init_nommio(xe);
	if (err)
		return err;

	err = xe_set_dma_info(xe);
	if (err)
		return err;

	err = xe_mmio_probe_tiles(xe);
	if (err)
		return err;

	xe_ttm_sys_mgr_init(xe);

	for_each_gt(gt, xe, id) {
		err = xe_gt_init_early(gt);
		if (err)
			return err;
	}

	for_each_tile(tile, xe, id) {
		if (IS_SRIOV_VF(xe)) {
			xe_guc_comm_init_early(&tile->primary_gt->uc.guc);
			err = xe_gt_sriov_vf_bootstrap(tile->primary_gt);
			if (err)
				return err;
			err = xe_gt_sriov_vf_query_config(tile->primary_gt);
			if (err)
				return err;
		}
		err = xe_ggtt_init_early(tile->mem.ggtt);
		if (err)
			return err;
		if (IS_SRIOV_VF(xe)) {
			err = xe_memirq_init(&tile->sriov.vf.memirq);
			if (err)
				return err;
		}
	}

	for_each_gt(gt, xe, id) {
		err = xe_gt_init_hwconfig(gt);
		if (err)
			return err;
	}

	err = xe_devcoredump_init(xe);
	if (err)
		return err;
	err = devm_add_action_or_reset(xe->drm.dev, xe_driver_flr_fini, xe);
	if (err)
		return err;

	err = xe_display_init_noirq(xe);
	if (err)
		return err;

	err = xe_irq_install(xe);
	if (err)
		goto err;

	err = xe_device_set_has_flat_ccs(xe);
	if (err)
		goto err;

	err = xe_vram_probe(xe);
	if (err)
		goto err;

	for_each_tile(tile, xe, id) {
		err = xe_tile_init_noalloc(tile);
		if (err)
			goto err;
	}

	/* Allocate and map stolen after potential VRAM resize */
	xe_ttm_stolen_mgr_init(xe);

	/*
	 * Now that GT is initialized (TTM in particular),
	 * we can try to init display, and inherit the initial fb.
	 * This is the reason the first allocation needs to be done
	 * inside display.
	 */
	err = xe_display_init_noaccel(xe);
	if (err)
		goto err;

	for_each_gt(gt, xe, id) {
		last_gt = id;

		err = xe_gt_init(gt);
		if (err)
			goto err_fini_gt;
	}

	xe_heci_gsc_init(xe);

	err = xe_oa_init(xe);
	if (err)
		goto err_fini_gt;

	err = xe_display_init(xe);
	if (err)
		goto err_fini_oa;

	err = drm_dev_register(&xe->drm, 0);
	if (err)
		goto err_fini_display;

	xe_display_register(xe);

	xe_oa_register(xe);

	xe_debugfs_register(xe);

	xe_hwmon_register(xe);

	for_each_gt(gt, xe, id)
		xe_gt_sanitize_freq(gt);

	return devm_add_action_or_reset(xe->drm.dev, xe_device_sanitize, xe);

err_fini_display:
	xe_display_driver_remove(xe);

err_fini_oa:
	xe_oa_fini(xe);

err_fini_gt:
	for_each_gt(gt, xe, id) {
		if (id < last_gt)
			xe_gt_remove(gt);
		else
			break;
	}

err:
	xe_display_fini(xe);
	return err;
}

static void xe_device_remove_display(struct xe_device *xe)
{
	xe_display_unregister(xe);

	drm_dev_unplug(&xe->drm);
	xe_display_driver_remove(xe);
}

void xe_device_remove(struct xe_device *xe)
{
	struct xe_gt *gt;
	u8 id;

	xe_oa_unregister(xe);

	xe_device_remove_display(xe);

	xe_display_fini(xe);

	xe_oa_fini(xe);

	xe_heci_gsc_fini(xe);

	for_each_gt(gt, xe, id)
		xe_gt_remove(gt);
}

void xe_device_shutdown(struct xe_device *xe)
{
}

/**
 * xe_device_wmb() - Device specific write memory barrier
 * @xe: the &xe_device
 *
 * While wmb() is sufficient for a barrier if we use system memory, on discrete
 * platforms with device memory we additionally need to issue a register write.
 * Since it doesn't matter which register we write to, use the read-only VF_CAP
 * register that is also marked as accessible by the VFs.
 */
void xe_device_wmb(struct xe_device *xe)
{
	struct xe_gt *gt = xe_root_mmio_gt(xe);

	wmb();
	if (IS_DGFX(xe))
		xe_mmio_write32(gt, VF_CAP_REG, 0);
}

/**
 * xe_device_td_flush() - Flush transient L3 cache entries
 * @xe: The device
 *
 * Display engine has direct access to memory and is never coherent with L3/L4
 * caches (or CPU caches), however KMD is responsible for specifically flushing
 * transient L3 GPU cache entries prior to the flip sequence to ensure scanout
 * can happen from such a surface without seeing corruption.
 *
 * Display surfaces can be tagged as transient by mapping it using one of the
 * various L3:XD PAT index modes on Xe2.
 *
 * Note: On non-discrete xe2 platforms, like LNL, the entire L3 cache is flushed
 * at the end of each submission via PIPE_CONTROL for compute/render, since SA
 * Media is not coherent with L3 and we want to support render-vs-media
 * usescases. For other engines like copy/blt the HW internally forces uncached
 * behaviour, hence why we can skip the TDF on such platforms.
 */
void xe_device_td_flush(struct xe_device *xe)
{
	struct xe_gt *gt;
	u8 id;

	if (!IS_DGFX(xe) || GRAPHICS_VER(xe) < 20)
		return;

	if (XE_WA(xe_root_mmio_gt(xe), 16023588340)) {
		xe_device_l2_flush(xe);
		return;
	}

	for_each_gt(gt, xe, id) {
		if (xe_gt_is_media_type(gt))
			continue;

		if (xe_force_wake_get(gt_to_fw(gt), XE_FW_GT))
			return;

		xe_mmio_write32(gt, XE2_TDF_CTRL, TRANSIENT_FLUSH_REQUEST);
		/*
		 * FIXME: We can likely do better here with our choice of
		 * timeout. Currently we just assume the worst case, i.e. 150us,
		 * which is believed to be sufficient to cover the worst case
		 * scenario on current platforms if all cache entries are
		 * transient and need to be flushed..
		 */
		if (xe_mmio_wait32(gt, XE2_TDF_CTRL, TRANSIENT_FLUSH_REQUEST, 0,
				   150, NULL, false))
			xe_gt_err_once(gt, "TD flush timeout\n");

		xe_force_wake_put(gt_to_fw(gt), XE_FW_GT);
	}
}

void xe_device_l2_flush(struct xe_device *xe)
{
	struct xe_gt *gt;
	int err;

	gt = xe_root_mmio_gt(xe);

	if (!XE_WA(gt, 16023588340))
		return;

	err = xe_force_wake_get(gt_to_fw(gt), XE_FW_GT);
	if (err)
		return;

	spin_lock(&gt->global_invl_lock);
	xe_mmio_write32(gt, XE2_GLOBAL_INVAL, 0x1);

	if (xe_mmio_wait32(gt, XE2_GLOBAL_INVAL, 0x1, 0x0, 150, NULL, true))
		xe_gt_err_once(gt, "Global invalidation timeout\n");
	spin_unlock(&gt->global_invl_lock);

	xe_force_wake_put(gt_to_fw(gt), XE_FW_GT);
}

u32 xe_device_ccs_bytes(struct xe_device *xe, u64 size)
{
	return xe_device_has_flat_ccs(xe) ?
		DIV_ROUND_UP_ULL(size, NUM_BYTES_PER_CCS_BYTE(xe)) : 0;
}

/**
 * xe_device_assert_mem_access - Inspect the current runtime_pm state.
 * @xe: xe device instance
 *
 * To be used before any kind of memory access. It will splat a debug warning
 * if the device is currently sleeping. But it doesn't guarantee in any way
 * that the device is going to remain awake. Xe PM runtime get and put
 * functions might be added to the outer bound of the memory access, while
 * this check is intended for inner usage to splat some warning if the worst
 * case has just happened.
 */
void xe_device_assert_mem_access(struct xe_device *xe)
{
	xe_assert(xe, !xe_pm_runtime_suspended(xe));
}

void xe_device_snapshot_print(struct xe_device *xe, struct drm_printer *p)
{
	struct xe_gt *gt;
	u8 id;

	drm_printf(p, "PCI ID: 0x%04x\n", xe->info.devid);
	drm_printf(p, "PCI revision: 0x%02x\n", xe->info.revid);

	for_each_gt(gt, xe, id) {
		drm_printf(p, "GT id: %u\n", id);
		drm_printf(p, "\tType: %s\n",
			   gt->info.type == XE_GT_TYPE_MAIN ? "main" : "media");
		drm_printf(p, "\tIP ver: %u.%u.%u\n",
			   REG_FIELD_GET(GMD_ID_ARCH_MASK, gt->info.gmdid),
			   REG_FIELD_GET(GMD_ID_RELEASE_MASK, gt->info.gmdid),
			   REG_FIELD_GET(GMD_ID_REVID, gt->info.gmdid));
		drm_printf(p, "\tCS reference clock: %u\n", gt->info.reference_clock);
	}
}

u64 xe_device_canonicalize_addr(struct xe_device *xe, u64 address)
{
	return sign_extend64(address, xe->info.va_bits - 1);
}

u64 xe_device_uncanonicalize_addr(struct xe_device *xe, u64 address)
{
	return address & GENMASK_ULL(xe->info.va_bits - 1, 0);
}

static void xe_device_wedged_fini(struct drm_device *drm, void *arg)
{
	struct xe_device *xe = arg;

	xe_pm_runtime_put(xe);
}

/**
 * xe_device_declare_wedged - Declare device wedged
 * @xe: xe device instance
 *
 * This is a final state that can only be cleared with a mudule
 * re-probe (unbind + bind).
 * In this state every IOCTL will be blocked so the GT cannot be used.
 * In general it will be called upon any critical error such as gt reset
 * failure or guc loading failure.
 * If xe.wedged module parameter is set to 2, this function will be called
 * on every single execution timeout (a.k.a. GPU hang) right after devcoredump
 * snapshot capture. In this mode, GT reset won't be attempted so the state of
 * the issue is preserved for further debugging.
 */
void xe_device_declare_wedged(struct xe_device *xe)
{
	struct xe_gt *gt;
	u8 id;

	if (xe->wedged.mode == 0) {
		drm_dbg(&xe->drm, "Wedged mode is forcibly disabled\n");
		return;
	}

	if (drmm_add_action_or_reset(&xe->drm, xe_device_wedged_fini, xe)) {
		drm_err(&xe->drm, "Failed to register xe_device_wedged_fini clean-up. Although device is wedged.\n");
		return;
	}

	xe_pm_runtime_get_noresume(xe);

	if (!atomic_xchg(&xe->wedged.flag, 1)) {
		xe->needs_flr_on_fini = true;
		drm_err(&xe->drm,
			"CRITICAL: Xe has declared device %s as wedged.\n"
			"IOCTLs and executions are blocked. Only a rebind may clear the failure\n"
			"Please file a _new_ bug report at https://gitlab.freedesktop.org/drm/xe/kernel/issues/new\n",
			dev_name(xe->drm.dev));
	}

	for_each_gt(gt, xe, id)
		xe_gt_declare_wedged(gt);
}