summaryrefslogtreecommitdiffstats
path: root/drivers/gpu/drm/msm/msm_gem.c
blob: 29eacfa29cfb6952c38b89d8e51883890d9125f2 (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
/*
 * Copyright (C) 2013 Red Hat
 * Author: Rob Clark <robdclark@gmail.com>
 *
 * This program is free software; you can redistribute it and/or modify it
 * under the terms of the GNU General Public License version 2 as published by
 * the Free Software Foundation.
 *
 * This program is distributed in the hope that it will be useful, but WITHOUT
 * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
 * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
 * more details.
 *
 * You should have received a copy of the GNU General Public License along with
 * this program.  If not, see <http://www.gnu.org/licenses/>.
 */

#include <linux/spinlock.h>
#include <linux/shmem_fs.h>

#include "msm_drv.h"
#include "msm_gem.h"
#include "msm_gpu.h"


/* called with dev->struct_mutex held */
static struct page **get_pages(struct drm_gem_object *obj)
{
	struct msm_gem_object *msm_obj = to_msm_bo(obj);

	if (!msm_obj->pages) {
		struct drm_device *dev = obj->dev;
		struct page **p = drm_gem_get_pages(obj, 0);
		int npages = obj->size >> PAGE_SHIFT;

		if (IS_ERR(p)) {
			dev_err(dev->dev, "could not get pages: %ld\n",
					PTR_ERR(p));
			return p;
		}

		msm_obj->sgt = drm_prime_pages_to_sg(p, npages);
		if (IS_ERR(msm_obj->sgt)) {
			dev_err(dev->dev, "failed to allocate sgt\n");
			return ERR_CAST(msm_obj->sgt);
		}

		msm_obj->pages = p;

		/* For non-cached buffers, ensure the new pages are clean
		 * because display controller, GPU, etc. are not coherent:
		 */
		if (msm_obj->flags & (MSM_BO_WC|MSM_BO_UNCACHED))
			dma_map_sg(dev->dev, msm_obj->sgt->sgl,
					msm_obj->sgt->nents, DMA_BIDIRECTIONAL);
	}

	return msm_obj->pages;
}

static void put_pages(struct drm_gem_object *obj)
{
	struct msm_gem_object *msm_obj = to_msm_bo(obj);

	if (msm_obj->pages) {
		/* For non-cached buffers, ensure the new pages are clean
		 * because display controller, GPU, etc. are not coherent:
		 */
		if (msm_obj->flags & (MSM_BO_WC|MSM_BO_UNCACHED))
			dma_unmap_sg(obj->dev->dev, msm_obj->sgt->sgl,
					msm_obj->sgt->nents, DMA_BIDIRECTIONAL);
		sg_free_table(msm_obj->sgt);
		kfree(msm_obj->sgt);

		drm_gem_put_pages(obj, msm_obj->pages, true, false);
		msm_obj->pages = NULL;
	}
}

int msm_gem_mmap_obj(struct drm_gem_object *obj,
		struct vm_area_struct *vma)
{
	struct msm_gem_object *msm_obj = to_msm_bo(obj);

	vma->vm_flags &= ~VM_PFNMAP;
	vma->vm_flags |= VM_MIXEDMAP;

	if (msm_obj->flags & MSM_BO_WC) {
		vma->vm_page_prot = pgprot_writecombine(vm_get_page_prot(vma->vm_flags));
	} else if (msm_obj->flags & MSM_BO_UNCACHED) {
		vma->vm_page_prot = pgprot_noncached(vm_get_page_prot(vma->vm_flags));
	} else {
		/*
		 * Shunt off cached objs to shmem file so they have their own
		 * address_space (so unmap_mapping_range does what we want,
		 * in particular in the case of mmap'd dmabufs)
		 */
		fput(vma->vm_file);
		get_file(obj->filp);
		vma->vm_pgoff = 0;
		vma->vm_file  = obj->filp;

		vma->vm_page_prot = vm_get_page_prot(vma->vm_flags);
	}

	return 0;
}

int msm_gem_mmap(struct file *filp, struct vm_area_struct *vma)
{
	int ret;

	ret = drm_gem_mmap(filp, vma);
	if (ret) {
		DBG("mmap failed: %d", ret);
		return ret;
	}

	return msm_gem_mmap_obj(vma->vm_private_data, vma);
}

int msm_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
{
	struct drm_gem_object *obj = vma->vm_private_data;
	struct msm_gem_object *msm_obj = to_msm_bo(obj);
	struct drm_device *dev = obj->dev;
	struct page **pages;
	unsigned long pfn;
	pgoff_t pgoff;
	int ret;

	/* Make sure we don't parallel update on a fault, nor move or remove
	 * something from beneath our feet
	 */
	ret = mutex_lock_interruptible(&dev->struct_mutex);
	if (ret)
		goto out;

	/* make sure we have pages attached now */
	pages = get_pages(obj);
	if (IS_ERR(pages)) {
		ret = PTR_ERR(pages);
		goto out_unlock;
	}

	/* We don't use vmf->pgoff since that has the fake offset: */
	pgoff = ((unsigned long)vmf->virtual_address -
			vma->vm_start) >> PAGE_SHIFT;

	pfn = page_to_pfn(msm_obj->pages[pgoff]);

	VERB("Inserting %p pfn %lx, pa %lx", vmf->virtual_address,
			pfn, pfn << PAGE_SHIFT);

	ret = vm_insert_mixed(vma, (unsigned long)vmf->virtual_address, pfn);

out_unlock:
	mutex_unlock(&dev->struct_mutex);
out:
	switch (ret) {
	case -EAGAIN:
	case 0:
	case -ERESTARTSYS:
	case -EINTR:
		return VM_FAULT_NOPAGE;
	case -ENOMEM:
		return VM_FAULT_OOM;
	default:
		return VM_FAULT_SIGBUS;
	}
}

/** get mmap offset */
static uint64_t mmap_offset(struct drm_gem_object *obj)
{
	struct drm_device *dev = obj->dev;
	int ret;

	WARN_ON(!mutex_is_locked(&dev->struct_mutex));

	/* Make it mmapable */
	ret = drm_gem_create_mmap_offset(obj);

	if (ret) {
		dev_err(dev->dev, "could not allocate mmap offset\n");
		return 0;
	}

	return drm_vma_node_offset_addr(&obj->vma_node);
}

uint64_t msm_gem_mmap_offset(struct drm_gem_object *obj)
{
	uint64_t offset;
	mutex_lock(&obj->dev->struct_mutex);
	offset = mmap_offset(obj);
	mutex_unlock(&obj->dev->struct_mutex);
	return offset;
}

/* helpers for dealing w/ iommu: */
static int map_range(struct iommu_domain *domain, unsigned int iova,
		struct sg_table *sgt, unsigned int len, int prot)
{
	struct scatterlist *sg;
	unsigned int da = iova;
	unsigned int i, j;
	int ret;

	if (!domain || !sgt)
		return -EINVAL;

	for_each_sg(sgt->sgl, sg, sgt->nents, i) {
		u32 pa = sg_phys(sg) - sg->offset;
		size_t bytes = sg->length + sg->offset;

		VERB("map[%d]: %08x %08x(%x)", i, iova, pa, bytes);

		ret = iommu_map(domain, da, pa, bytes, prot);
		if (ret)
			goto fail;

		da += bytes;
	}

	return 0;

fail:
	da = iova;

	for_each_sg(sgt->sgl, sg, i, j) {
		size_t bytes = sg->length + sg->offset;
		iommu_unmap(domain, da, bytes);
		da += bytes;
	}
	return ret;
}

static void unmap_range(struct iommu_domain *domain, unsigned int iova,
		struct sg_table *sgt, unsigned int len)
{
	struct scatterlist *sg;
	unsigned int da = iova;
	int i;

	for_each_sg(sgt->sgl, sg, sgt->nents, i) {
		size_t bytes = sg->length + sg->offset;
		size_t unmapped;

		unmapped = iommu_unmap(domain, da, bytes);
		if (unmapped < bytes)
			break;

		VERB("unmap[%d]: %08x(%x)", i, iova, bytes);

		BUG_ON(!IS_ALIGNED(bytes, PAGE_SIZE));

		da += bytes;
	}
}

/* should be called under struct_mutex.. although it can be called
 * from atomic context without struct_mutex to acquire an extra
 * iova ref if you know one is already held.
 *
 * That means when I do eventually need to add support for unpinning
 * the refcnt counter needs to be atomic_t.
 */
int msm_gem_get_iova_locked(struct drm_gem_object *obj, int id,
		uint32_t *iova)
{
	struct msm_gem_object *msm_obj = to_msm_bo(obj);
	int ret = 0;

	if (!msm_obj->domain[id].iova) {
		struct msm_drm_private *priv = obj->dev->dev_private;
		uint32_t offset = (uint32_t)mmap_offset(obj);
		struct page **pages;
		pages = get_pages(obj);
		if (IS_ERR(pages))
			return PTR_ERR(pages);
		// XXX ideally we would not map buffers writable when not needed...
		ret = map_range(priv->iommus[id], offset, msm_obj->sgt,
				obj->size, IOMMU_READ | IOMMU_WRITE);
		msm_obj->domain[id].iova = offset;
	}

	if (!ret)
		*iova = msm_obj->domain[id].iova;

	return ret;
}

int msm_gem_get_iova(struct drm_gem_object *obj, int id, uint32_t *iova)
{
	int ret;
	mutex_lock(&obj->dev->struct_mutex);
	ret = msm_gem_get_iova_locked(obj, id, iova);
	mutex_unlock(&obj->dev->struct_mutex);
	return ret;
}

void msm_gem_put_iova(struct drm_gem_object *obj, int id)
{
	// XXX TODO ..
	// NOTE: probably don't need a _locked() version.. we wouldn't
	// normally unmap here, but instead just mark that it could be
	// unmapped (if the iova refcnt drops to zero), but then later
	// if another _get_iova_locked() fails we can start unmapping
	// things that are no longer needed..
}

int msm_gem_dumb_create(struct drm_file *file, struct drm_device *dev,
		struct drm_mode_create_dumb *args)
{
	args->pitch = align_pitch(args->width, args->bpp);
	args->size  = PAGE_ALIGN(args->pitch * args->height);
	return msm_gem_new_handle(dev, file, args->size,
			MSM_BO_SCANOUT | MSM_BO_WC, &args->handle);
}

int msm_gem_dumb_destroy(struct drm_file *file, struct drm_device *dev,
		uint32_t handle)
{
	/* No special work needed, drop the reference and see what falls out */
	return drm_gem_handle_delete(file, handle);
}

int msm_gem_dumb_map_offset(struct drm_file *file, struct drm_device *dev,
		uint32_t handle, uint64_t *offset)
{
	struct drm_gem_object *obj;
	int ret = 0;

	/* GEM does all our handle to object mapping */
	obj = drm_gem_object_lookup(dev, file, handle);
	if (obj == NULL) {
		ret = -ENOENT;
		goto fail;
	}

	*offset = msm_gem_mmap_offset(obj);

	drm_gem_object_unreference_unlocked(obj);

fail:
	return ret;
}

void *msm_gem_vaddr_locked(struct drm_gem_object *obj)
{
	struct msm_gem_object *msm_obj = to_msm_bo(obj);
	WARN_ON(!mutex_is_locked(&obj->dev->struct_mutex));
	if (!msm_obj->vaddr) {
		struct page **pages = get_pages(obj);
		if (IS_ERR(pages))
			return ERR_CAST(pages);
		msm_obj->vaddr = vmap(pages, obj->size >> PAGE_SHIFT,
				VM_MAP, pgprot_writecombine(PAGE_KERNEL));
	}
	return msm_obj->vaddr;
}

void *msm_gem_vaddr(struct drm_gem_object *obj)
{
	void *ret;
	mutex_lock(&obj->dev->struct_mutex);
	ret = msm_gem_vaddr_locked(obj);
	mutex_unlock(&obj->dev->struct_mutex);
	return ret;
}

int msm_gem_queue_inactive_work(struct drm_gem_object *obj,
		struct work_struct *work)
{
	struct drm_device *dev = obj->dev;
	struct msm_drm_private *priv = dev->dev_private;
	struct msm_gem_object *msm_obj = to_msm_bo(obj);
	int ret = 0;

	mutex_lock(&dev->struct_mutex);
	if (!list_empty(&work->entry)) {
		ret = -EINVAL;
	} else if (is_active(msm_obj)) {
		list_add_tail(&work->entry, &msm_obj->inactive_work);
	} else {
		queue_work(priv->wq, work);
	}
	mutex_unlock(&dev->struct_mutex);

	return ret;
}

void msm_gem_move_to_active(struct drm_gem_object *obj,
		struct msm_gpu *gpu, bool write, uint32_t fence)
{
	struct msm_gem_object *msm_obj = to_msm_bo(obj);
	msm_obj->gpu = gpu;
	if (write)
		msm_obj->write_fence = fence;
	else
		msm_obj->read_fence = fence;
	list_del_init(&msm_obj->mm_list);
	list_add_tail(&msm_obj->mm_list, &gpu->active_list);
}

void msm_gem_move_to_inactive(struct drm_gem_object *obj)
{
	struct drm_device *dev = obj->dev;
	struct msm_drm_private *priv = dev->dev_private;
	struct msm_gem_object *msm_obj = to_msm_bo(obj);

	WARN_ON(!mutex_is_locked(&dev->struct_mutex));

	msm_obj->gpu = NULL;
	msm_obj->read_fence = 0;
	msm_obj->write_fence = 0;
	list_del_init(&msm_obj->mm_list);
	list_add_tail(&msm_obj->mm_list, &priv->inactive_list);

	while (!list_empty(&msm_obj->inactive_work)) {
		struct work_struct *work;

		work = list_first_entry(&msm_obj->inactive_work,
				struct work_struct, entry);

		list_del_init(&work->entry);
		queue_work(priv->wq, work);
	}
}

int msm_gem_cpu_prep(struct drm_gem_object *obj, uint32_t op,
		struct timespec *timeout)
{
	struct drm_device *dev = obj->dev;
	struct msm_gem_object *msm_obj = to_msm_bo(obj);
	int ret = 0;

	if (is_active(msm_obj)) {
		uint32_t fence = 0;

		if (op & MSM_PREP_READ)
			fence = msm_obj->write_fence;
		if (op & MSM_PREP_WRITE)
			fence = max(fence, msm_obj->read_fence);
		if (op & MSM_PREP_NOSYNC)
			timeout = NULL;

		ret = msm_wait_fence_interruptable(dev, fence, timeout);
	}

	/* TODO cache maintenance */

	return ret;
}

int msm_gem_cpu_fini(struct drm_gem_object *obj)
{
	/* TODO cache maintenance */
	return 0;
}

#ifdef CONFIG_DEBUG_FS
void msm_gem_describe(struct drm_gem_object *obj, struct seq_file *m)
{
	struct drm_device *dev = obj->dev;
	struct msm_gem_object *msm_obj = to_msm_bo(obj);
	uint64_t off = drm_vma_node_start(&obj->vma_node);

	WARN_ON(!mutex_is_locked(&dev->struct_mutex));
	seq_printf(m, "%08x: %c(r=%u,w=%u) %2d (%2d) %08llx %p %d\n",
			msm_obj->flags, is_active(msm_obj) ? 'A' : 'I',
			msm_obj->read_fence, msm_obj->write_fence,
			obj->name, obj->refcount.refcount.counter,
			off, msm_obj->vaddr, obj->size);
}

void msm_gem_describe_objects(struct list_head *list, struct seq_file *m)
{
	struct msm_gem_object *msm_obj;
	int count = 0;
	size_t size = 0;

	list_for_each_entry(msm_obj, list, mm_list) {
		struct drm_gem_object *obj = &msm_obj->base;
		seq_printf(m, "   ");
		msm_gem_describe(obj, m);
		count++;
		size += obj->size;
	}

	seq_printf(m, "Total %d objects, %zu bytes\n", count, size);
}
#endif

void msm_gem_free_object(struct drm_gem_object *obj)
{
	struct drm_device *dev = obj->dev;
	struct msm_gem_object *msm_obj = to_msm_bo(obj);
	int id;

	WARN_ON(!mutex_is_locked(&dev->struct_mutex));

	/* object should not be on active list: */
	WARN_ON(is_active(msm_obj));

	list_del(&msm_obj->mm_list);

	for (id = 0; id < ARRAY_SIZE(msm_obj->domain); id++) {
		if (msm_obj->domain[id].iova) {
			struct msm_drm_private *priv = obj->dev->dev_private;
			uint32_t offset = (uint32_t)mmap_offset(obj);
			unmap_range(priv->iommus[id], offset,
					msm_obj->sgt, obj->size);
		}
	}

	drm_gem_free_mmap_offset(obj);

	if (msm_obj->vaddr)
		vunmap(msm_obj->vaddr);

	put_pages(obj);

	if (msm_obj->resv == &msm_obj->_resv)
		reservation_object_fini(msm_obj->resv);

	drm_gem_object_release(obj);

	kfree(msm_obj);
}

/* convenience method to construct a GEM buffer object, and userspace handle */
int msm_gem_new_handle(struct drm_device *dev, struct drm_file *file,
		uint32_t size, uint32_t flags, uint32_t *handle)
{
	struct drm_gem_object *obj;
	int ret;

	ret = mutex_lock_interruptible(&dev->struct_mutex);
	if (ret)
		return ret;

	obj = msm_gem_new(dev, size, flags);

	mutex_unlock(&dev->struct_mutex);

	if (IS_ERR(obj))
		return PTR_ERR(obj);

	ret = drm_gem_handle_create(file, obj, handle);

	/* drop reference from allocate - handle holds it now */
	drm_gem_object_unreference_unlocked(obj);

	return ret;
}

struct drm_gem_object *msm_gem_new(struct drm_device *dev,
		uint32_t size, uint32_t flags)
{
	struct msm_drm_private *priv = dev->dev_private;
	struct msm_gem_object *msm_obj;
	struct drm_gem_object *obj = NULL;
	int ret;

	WARN_ON(!mutex_is_locked(&dev->struct_mutex));

	size = PAGE_ALIGN(size);

	switch (flags & MSM_BO_CACHE_MASK) {
	case MSM_BO_UNCACHED:
	case MSM_BO_CACHED:
	case MSM_BO_WC:
		break;
	default:
		dev_err(dev->dev, "invalid cache flag: %x\n",
				(flags & MSM_BO_CACHE_MASK));
		ret = -EINVAL;
		goto fail;
	}

	msm_obj = kzalloc(sizeof(*msm_obj), GFP_KERNEL);
	if (!msm_obj) {
		ret = -ENOMEM;
		goto fail;
	}

	obj = &msm_obj->base;

	ret = drm_gem_object_init(dev, obj, size);
	if (ret)
		goto fail;

	msm_obj->flags = flags;

	msm_obj->resv = &msm_obj->_resv;
	reservation_object_init(msm_obj->resv);

	INIT_LIST_HEAD(&msm_obj->submit_entry);
	INIT_LIST_HEAD(&msm_obj->inactive_work);
	list_add_tail(&msm_obj->mm_list, &priv->inactive_list);

	return obj;

fail:
	if (obj)
		drm_gem_object_unreference_unlocked(obj);

	return ERR_PTR(ret);
}
OpenPOWER on IntegriCloud