summaryrefslogtreecommitdiff
path: root/drivers/gpu/drm/i915/gt/intel_timeline.c
blob: 46d20f5f3ddcc93ff898d23cd3f779cff093d23d (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
/*
 * SPDX-License-Identifier: MIT
 *
 * Copyright © 2016-2018 Intel Corporation
 */

#include "i915_drv.h"

#include "i915_active.h"
#include "i915_syncmap.h"
#include "intel_gt.h"
#include "intel_ring.h"
#include "intel_timeline.h"

#define ptr_set_bit(ptr, bit) ((typeof(ptr))((unsigned long)(ptr) | BIT(bit)))
#define ptr_test_bit(ptr, bit) ((unsigned long)(ptr) & BIT(bit))

#define CACHELINE_BITS 6
#define CACHELINE_FREE CACHELINE_BITS

struct intel_timeline_hwsp {
	struct intel_gt *gt;
	struct intel_gt_timelines *gt_timelines;
	struct list_head free_link;
	struct i915_vma *vma;
	u64 free_bitmap;
};

static struct i915_vma *__hwsp_alloc(struct intel_gt *gt)
{
	struct drm_i915_private *i915 = gt->i915;
	struct drm_i915_gem_object *obj;
	struct i915_vma *vma;

	obj = i915_gem_object_create_internal(i915, PAGE_SIZE);
	if (IS_ERR(obj))
		return ERR_CAST(obj);

	i915_gem_object_set_cache_coherency(obj, I915_CACHE_LLC);

	vma = i915_vma_instance(obj, &gt->ggtt->vm, NULL);
	if (IS_ERR(vma))
		i915_gem_object_put(obj);

	return vma;
}

static struct i915_vma *
hwsp_alloc(struct intel_timeline *timeline, unsigned int *cacheline)
{
	struct intel_gt_timelines *gt = &timeline->gt->timelines;
	struct intel_timeline_hwsp *hwsp;

	BUILD_BUG_ON(BITS_PER_TYPE(u64) * CACHELINE_BYTES > PAGE_SIZE);

	spin_lock_irq(&gt->hwsp_lock);

	/* hwsp_free_list only contains HWSP that have available cachelines */
	hwsp = list_first_entry_or_null(&gt->hwsp_free_list,
					typeof(*hwsp), free_link);
	if (!hwsp) {
		struct i915_vma *vma;

		spin_unlock_irq(&gt->hwsp_lock);

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

		vma = __hwsp_alloc(timeline->gt);
		if (IS_ERR(vma)) {
			kfree(hwsp);
			return vma;
		}

		GT_TRACE(timeline->gt, "new HWSP allocated\n");

		vma->private = hwsp;
		hwsp->gt = timeline->gt;
		hwsp->vma = vma;
		hwsp->free_bitmap = ~0ull;
		hwsp->gt_timelines = gt;

		spin_lock_irq(&gt->hwsp_lock);
		list_add(&hwsp->free_link, &gt->hwsp_free_list);
	}

	GEM_BUG_ON(!hwsp->free_bitmap);
	*cacheline = __ffs64(hwsp->free_bitmap);
	hwsp->free_bitmap &= ~BIT_ULL(*cacheline);
	if (!hwsp->free_bitmap)
		list_del(&hwsp->free_link);

	spin_unlock_irq(&gt->hwsp_lock);

	GEM_BUG_ON(hwsp->vma->private != hwsp);
	return hwsp->vma;
}

static void __idle_hwsp_free(struct intel_timeline_hwsp *hwsp, int cacheline)
{
	struct intel_gt_timelines *gt = hwsp->gt_timelines;
	unsigned long flags;

	spin_lock_irqsave(&gt->hwsp_lock, flags);

	/* As a cacheline becomes available, publish the HWSP on the freelist */
	if (!hwsp->free_bitmap)
		list_add_tail(&hwsp->free_link, &gt->hwsp_free_list);

	GEM_BUG_ON(cacheline >= BITS_PER_TYPE(hwsp->free_bitmap));
	hwsp->free_bitmap |= BIT_ULL(cacheline);

	/* And if no one is left using it, give the page back to the system */
	if (hwsp->free_bitmap == ~0ull) {
		i915_vma_put(hwsp->vma);
		list_del(&hwsp->free_link);
		kfree(hwsp);
	}

	spin_unlock_irqrestore(&gt->hwsp_lock, flags);
}

static void __rcu_cacheline_free(struct rcu_head *rcu)
{
	struct intel_timeline_cacheline *cl =
		container_of(rcu, typeof(*cl), rcu);

	i915_active_fini(&cl->active);
	kfree(cl);
}

static void __idle_cacheline_free(struct intel_timeline_cacheline *cl)
{
	GEM_BUG_ON(!i915_active_is_idle(&cl->active));

	i915_gem_object_unpin_map(cl->hwsp->vma->obj);
	i915_vma_put(cl->hwsp->vma);
	__idle_hwsp_free(cl->hwsp, ptr_unmask_bits(cl->vaddr, CACHELINE_BITS));

	call_rcu(&cl->rcu, __rcu_cacheline_free);
}

__i915_active_call
static void __cacheline_retire(struct i915_active *active)
{
	struct intel_timeline_cacheline *cl =
		container_of(active, typeof(*cl), active);

	i915_vma_unpin(cl->hwsp->vma);
	if (ptr_test_bit(cl->vaddr, CACHELINE_FREE))
		__idle_cacheline_free(cl);
}

static int __cacheline_active(struct i915_active *active)
{
	struct intel_timeline_cacheline *cl =
		container_of(active, typeof(*cl), active);

	__i915_vma_pin(cl->hwsp->vma);
	return 0;
}

static struct intel_timeline_cacheline *
cacheline_alloc(struct intel_timeline_hwsp *hwsp, unsigned int cacheline)
{
	struct intel_timeline_cacheline *cl;
	void *vaddr;

	GEM_BUG_ON(cacheline >= BIT(CACHELINE_BITS));

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

	vaddr = i915_gem_object_pin_map(hwsp->vma->obj, I915_MAP_WB);
	if (IS_ERR(vaddr)) {
		kfree(cl);
		return ERR_CAST(vaddr);
	}

	i915_vma_get(hwsp->vma);
	cl->hwsp = hwsp;
	cl->vaddr = page_pack_bits(vaddr, cacheline);

	i915_active_init(&cl->active, __cacheline_active, __cacheline_retire);

	return cl;
}

static void cacheline_acquire(struct intel_timeline_cacheline *cl)
{
	if (cl)
		i915_active_acquire(&cl->active);
}

static void cacheline_release(struct intel_timeline_cacheline *cl)
{
	if (cl)
		i915_active_release(&cl->active);
}

static void cacheline_free(struct intel_timeline_cacheline *cl)
{
	if (!i915_active_acquire_if_busy(&cl->active)) {
		__idle_cacheline_free(cl);
		return;
	}

	GEM_BUG_ON(ptr_test_bit(cl->vaddr, CACHELINE_FREE));
	cl->vaddr = ptr_set_bit(cl->vaddr, CACHELINE_FREE);

	i915_active_release(&cl->active);
}

static int intel_timeline_init(struct intel_timeline *timeline,
			       struct intel_gt *gt,
			       struct i915_vma *hwsp)
{
	void *vaddr;

	kref_init(&timeline->kref);
	atomic_set(&timeline->pin_count, 0);

	timeline->gt = gt;

	timeline->has_initial_breadcrumb = !hwsp;
	timeline->hwsp_cacheline = NULL;

	if (!hwsp) {
		struct intel_timeline_cacheline *cl;
		unsigned int cacheline;

		hwsp = hwsp_alloc(timeline, &cacheline);
		if (IS_ERR(hwsp))
			return PTR_ERR(hwsp);

		cl = cacheline_alloc(hwsp->private, cacheline);
		if (IS_ERR(cl)) {
			__idle_hwsp_free(hwsp->private, cacheline);
			return PTR_ERR(cl);
		}

		timeline->hwsp_cacheline = cl;
		timeline->hwsp_offset = cacheline * CACHELINE_BYTES;

		vaddr = page_mask_bits(cl->vaddr);
	} else {
		timeline->hwsp_offset = I915_GEM_HWS_SEQNO_ADDR;

		vaddr = i915_gem_object_pin_map(hwsp->obj, I915_MAP_WB);
		if (IS_ERR(vaddr))
			return PTR_ERR(vaddr);
	}

	timeline->hwsp_seqno =
		memset(vaddr + timeline->hwsp_offset, 0, CACHELINE_BYTES);

	timeline->hwsp_ggtt = i915_vma_get(hwsp);
	GEM_BUG_ON(timeline->hwsp_offset >= hwsp->size);

	timeline->fence_context = dma_fence_context_alloc(1);

	mutex_init(&timeline->mutex);

	INIT_ACTIVE_FENCE(&timeline->last_request);
	INIT_LIST_HEAD(&timeline->requests);

	i915_syncmap_init(&timeline->sync);

	return 0;
}

void intel_gt_init_timelines(struct intel_gt *gt)
{
	struct intel_gt_timelines *timelines = &gt->timelines;

	spin_lock_init(&timelines->lock);
	INIT_LIST_HEAD(&timelines->active_list);

	spin_lock_init(&timelines->hwsp_lock);
	INIT_LIST_HEAD(&timelines->hwsp_free_list);
}

static void intel_timeline_fini(struct intel_timeline *timeline)
{
	GEM_BUG_ON(atomic_read(&timeline->pin_count));
	GEM_BUG_ON(!list_empty(&timeline->requests));
	GEM_BUG_ON(timeline->retire);

	if (timeline->hwsp_cacheline)
		cacheline_free(timeline->hwsp_cacheline);
	else
		i915_gem_object_unpin_map(timeline->hwsp_ggtt->obj);

	i915_vma_put(timeline->hwsp_ggtt);
}

struct intel_timeline *
intel_timeline_create(struct intel_gt *gt, struct i915_vma *global_hwsp)
{
	struct intel_timeline *timeline;
	int err;

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

	err = intel_timeline_init(timeline, gt, global_hwsp);
	if (err) {
		kfree(timeline);
		return ERR_PTR(err);
	}

	return timeline;
}

int intel_timeline_pin(struct intel_timeline *tl)
{
	int err;

	if (atomic_add_unless(&tl->pin_count, 1, 0))
		return 0;

	err = i915_ggtt_pin(tl->hwsp_ggtt, 0, PIN_HIGH);
	if (err)
		return err;

	tl->hwsp_offset =
		i915_ggtt_offset(tl->hwsp_ggtt) +
		offset_in_page(tl->hwsp_offset);
	GT_TRACE(tl->gt, "timeline:%llx using HWSP offset:%x\n",
		 tl->fence_context, tl->hwsp_offset);

	cacheline_acquire(tl->hwsp_cacheline);
	if (atomic_fetch_inc(&tl->pin_count)) {
		cacheline_release(tl->hwsp_cacheline);
		__i915_vma_unpin(tl->hwsp_ggtt);
	}

	return 0;
}

void intel_timeline_reset_seqno(const struct intel_timeline *tl)
{
	/* Must be pinned to be writable, and no requests in flight. */
	GEM_BUG_ON(!atomic_read(&tl->pin_count));
	WRITE_ONCE(*(u32 *)tl->hwsp_seqno, tl->seqno);
}

void intel_timeline_enter(struct intel_timeline *tl)
{
	struct intel_gt_timelines *timelines = &tl->gt->timelines;

	/*
	 * Pretend we are serialised by the timeline->mutex.
	 *
	 * While generally true, there are a few exceptions to the rule
	 * for the engine->kernel_context being used to manage power
	 * transitions. As the engine_park may be called from under any
	 * timeline, it uses the power mutex as a global serialisation
	 * lock to prevent any other request entering its timeline.
	 *
	 * The rule is generally tl->mutex, otherwise engine->wakeref.mutex.
	 *
	 * However, intel_gt_retire_request() does not know which engine
	 * it is retiring along and so cannot partake in the engine-pm
	 * barrier, and there we use the tl->active_count as a means to
	 * pin the timeline in the active_list while the locks are dropped.
	 * Ergo, as that is outside of the engine-pm barrier, we need to
	 * use atomic to manipulate tl->active_count.
	 */
	lockdep_assert_held(&tl->mutex);

	if (atomic_add_unless(&tl->active_count, 1, 0))
		return;

	spin_lock(&timelines->lock);
	if (!atomic_fetch_inc(&tl->active_count)) {
		/*
		 * The HWSP is volatile, and may have been lost while inactive,
		 * e.g. across suspend/resume. Be paranoid, and ensure that
		 * the HWSP value matches our seqno so we don't proclaim
		 * the next request as already complete.
		 */
		intel_timeline_reset_seqno(tl);
		list_add_tail(&tl->link, &timelines->active_list);
	}
	spin_unlock(&timelines->lock);
}

void intel_timeline_exit(struct intel_timeline *tl)
{
	struct intel_gt_timelines *timelines = &tl->gt->timelines;

	/* See intel_timeline_enter() */
	lockdep_assert_held(&tl->mutex);

	GEM_BUG_ON(!atomic_read(&tl->active_count));
	if (atomic_add_unless(&tl->active_count, -1, 1))
		return;

	spin_lock(&timelines->lock);
	if (atomic_dec_and_test(&tl->active_count))
		list_del(&tl->link);
	spin_unlock(&timelines->lock);

	/*
	 * Since this timeline is idle, all bariers upon which we were waiting
	 * must also be complete and so we can discard the last used barriers
	 * without loss of information.
	 */
	i915_syncmap_free(&tl->sync);
}

static u32 timeline_advance(struct intel_timeline *tl)
{
	GEM_BUG_ON(!atomic_read(&tl->pin_count));
	GEM_BUG_ON(tl->seqno & tl->has_initial_breadcrumb);

	return tl->seqno += 1 + tl->has_initial_breadcrumb;
}

static void timeline_rollback(struct intel_timeline *tl)
{
	tl->seqno -= 1 + tl->has_initial_breadcrumb;
}

static noinline int
__intel_timeline_get_seqno(struct intel_timeline *tl,
			   struct i915_request *rq,
			   u32 *seqno)
{
	struct intel_timeline_cacheline *cl;
	unsigned int cacheline;
	struct i915_vma *vma;
	void *vaddr;
	int err;

	might_lock(&tl->gt->ggtt->vm.mutex);
	GT_TRACE(tl->gt, "timeline:%llx wrapped\n", tl->fence_context);

	/*
	 * If there is an outstanding GPU reference to this cacheline,
	 * such as it being sampled by a HW semaphore on another timeline,
	 * we cannot wraparound our seqno value (the HW semaphore does
	 * a strict greater-than-or-equals compare, not i915_seqno_passed).
	 * So if the cacheline is still busy, we must detach ourselves
	 * from it and leave it inflight alongside its users.
	 *
	 * However, if nobody is watching and we can guarantee that nobody
	 * will, we could simply reuse the same cacheline.
	 *
	 * if (i915_active_request_is_signaled(&tl->last_request) &&
	 *     i915_active_is_signaled(&tl->hwsp_cacheline->active))
	 *	return 0;
	 *
	 * That seems unlikely for a busy timeline that needed to wrap in
	 * the first place, so just replace the cacheline.
	 */

	vma = hwsp_alloc(tl, &cacheline);
	if (IS_ERR(vma)) {
		err = PTR_ERR(vma);
		goto err_rollback;
	}

	err = i915_ggtt_pin(vma, 0, PIN_HIGH);
	if (err) {
		__idle_hwsp_free(vma->private, cacheline);
		goto err_rollback;
	}

	cl = cacheline_alloc(vma->private, cacheline);
	if (IS_ERR(cl)) {
		err = PTR_ERR(cl);
		__idle_hwsp_free(vma->private, cacheline);
		goto err_unpin;
	}
	GEM_BUG_ON(cl->hwsp->vma != vma);

	/*
	 * Attach the old cacheline to the current request, so that we only
	 * free it after the current request is retired, which ensures that
	 * all writes into the cacheline from previous requests are complete.
	 */
	err = i915_active_ref(&tl->hwsp_cacheline->active, tl, &rq->fence);
	if (err)
		goto err_cacheline;

	cacheline_release(tl->hwsp_cacheline); /* ownership now xfered to rq */
	cacheline_free(tl->hwsp_cacheline);

	i915_vma_unpin(tl->hwsp_ggtt); /* binding kept alive by old cacheline */
	i915_vma_put(tl->hwsp_ggtt);

	tl->hwsp_ggtt = i915_vma_get(vma);

	vaddr = page_mask_bits(cl->vaddr);
	tl->hwsp_offset = cacheline * CACHELINE_BYTES;
	tl->hwsp_seqno =
		memset(vaddr + tl->hwsp_offset, 0, CACHELINE_BYTES);

	tl->hwsp_offset += i915_ggtt_offset(vma);
	GT_TRACE(tl->gt, "timeline:%llx using HWSP offset:%x\n",
		 tl->fence_context, tl->hwsp_offset);

	cacheline_acquire(cl);
	tl->hwsp_cacheline = cl;

	*seqno = timeline_advance(tl);
	GEM_BUG_ON(i915_seqno_passed(*tl->hwsp_seqno, *seqno));
	return 0;

err_cacheline:
	cacheline_free(cl);
err_unpin:
	i915_vma_unpin(vma);
err_rollback:
	timeline_rollback(tl);
	return err;
}

int intel_timeline_get_seqno(struct intel_timeline *tl,
			     struct i915_request *rq,
			     u32 *seqno)
{
	*seqno = timeline_advance(tl);

	/* Replace the HWSP on wraparound for HW semaphores */
	if (unlikely(!*seqno && tl->hwsp_cacheline))
		return __intel_timeline_get_seqno(tl, rq, seqno);

	return 0;
}

static int cacheline_ref(struct intel_timeline_cacheline *cl,
			 struct i915_request *rq)
{
	return i915_active_add_request(&cl->active, rq);
}

int intel_timeline_read_hwsp(struct i915_request *from,
			     struct i915_request *to,
			     u32 *hwsp)
{
	struct intel_timeline_cacheline *cl;
	int err;

	GEM_BUG_ON(!rcu_access_pointer(from->hwsp_cacheline));

	rcu_read_lock();
	cl = rcu_dereference(from->hwsp_cacheline);
	if (i915_request_completed(from)) /* confirm cacheline is valid */
		goto unlock;
	if (unlikely(!i915_active_acquire_if_busy(&cl->active)))
		goto unlock; /* seqno wrapped and completed! */
	if (unlikely(i915_request_completed(from)))
		goto release;
	rcu_read_unlock();

	err = cacheline_ref(cl, to);
	if (err)
		goto out;

	*hwsp = i915_ggtt_offset(cl->hwsp->vma) +
		ptr_unmask_bits(cl->vaddr, CACHELINE_BITS) * CACHELINE_BYTES;

out:
	i915_active_release(&cl->active);
	return err;

release:
	i915_active_release(&cl->active);
unlock:
	rcu_read_unlock();
	return 1;
}

void intel_timeline_unpin(struct intel_timeline *tl)
{
	GEM_BUG_ON(!atomic_read(&tl->pin_count));
	if (!atomic_dec_and_test(&tl->pin_count))
		return;

	cacheline_release(tl->hwsp_cacheline);

	__i915_vma_unpin(tl->hwsp_ggtt);
}

void __intel_timeline_free(struct kref *kref)
{
	struct intel_timeline *timeline =
		container_of(kref, typeof(*timeline), kref);

	intel_timeline_fini(timeline);
	kfree_rcu(timeline, rcu);
}

void intel_gt_fini_timelines(struct intel_gt *gt)
{
	struct intel_gt_timelines *timelines = &gt->timelines;

	GEM_BUG_ON(!list_empty(&timelines->active_list));
	GEM_BUG_ON(!list_empty(&timelines->hwsp_free_list));
}

#if IS_ENABLED(CONFIG_DRM_I915_SELFTEST)
#include "gt/selftests/mock_timeline.c"
#include "gt/selftest_timeline.c"
#endif