Training courses

Kernel and Embedded Linux

Bootlin training courses

Embedded Linux, kernel,
Yocto Project, Buildroot, real-time,
graphics, boot time, debugging...

Bootlin logo

Elixir Cross Referencer

  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
/*
 * SPDX-License-Identifier: MIT
 *
 * Copyright © 2014-2016 Intel Corporation
 */

#include <linux/pagevec.h>
#include <linux/swap.h>

#include "i915_drv.h"
#include "i915_gem_object.h"
#include "i915_scatterlist.h"
#include "i915_trace.h"

/*
 * Move pages to appropriate lru and release the pagevec, decrementing the
 * ref count of those pages.
 */
static void check_release_pagevec(struct pagevec *pvec)
{
	check_move_unevictable_pages(pvec);
	__pagevec_release(pvec);
	cond_resched();
}

static int shmem_get_pages(struct drm_i915_gem_object *obj)
{
	struct drm_i915_private *i915 = to_i915(obj->base.dev);
	const unsigned long page_count = obj->base.size / PAGE_SIZE;
	unsigned long i;
	struct address_space *mapping;
	struct sg_table *st;
	struct scatterlist *sg;
	struct sgt_iter sgt_iter;
	struct page *page;
	unsigned long last_pfn = 0;	/* suppress gcc warning */
	unsigned int max_segment = i915_sg_segment_size();
	unsigned int sg_page_sizes;
	struct pagevec pvec;
	gfp_t noreclaim;
	int ret;

	/*
	 * Assert that the object is not currently in any GPU domain. As it
	 * wasn't in the GTT, there shouldn't be any way it could have been in
	 * a GPU cache
	 */
	GEM_BUG_ON(obj->read_domains & I915_GEM_GPU_DOMAINS);
	GEM_BUG_ON(obj->write_domain & I915_GEM_GPU_DOMAINS);

	/*
	 * If there's no chance of allocating enough pages for the whole
	 * object, bail early.
	 */
	if (page_count > totalram_pages())
		return -ENOMEM;

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

rebuild_st:
	if (sg_alloc_table(st, page_count, GFP_KERNEL)) {
		kfree(st);
		return -ENOMEM;
	}

	/*
	 * Get the list of pages out of our struct file.  They'll be pinned
	 * at this point until we release them.
	 *
	 * Fail silently without starting the shrinker
	 */
	mapping = obj->base.filp->f_mapping;
	mapping_set_unevictable(mapping);
	noreclaim = mapping_gfp_constraint(mapping, ~__GFP_RECLAIM);
	noreclaim |= __GFP_NORETRY | __GFP_NOWARN;

	sg = st->sgl;
	st->nents = 0;
	sg_page_sizes = 0;
	for (i = 0; i < page_count; i++) {
		const unsigned int shrink[] = {
			I915_SHRINK_BOUND | I915_SHRINK_UNBOUND,
			0,
		}, *s = shrink;
		gfp_t gfp = noreclaim;

		do {
			cond_resched();
			page = shmem_read_mapping_page_gfp(mapping, i, gfp);
			if (!IS_ERR(page))
				break;

			if (!*s) {
				ret = PTR_ERR(page);
				goto err_sg;
			}

			i915_gem_shrink(i915, 2 * page_count, NULL, *s++);

			/*
			 * We've tried hard to allocate the memory by reaping
			 * our own buffer, now let the real VM do its job and
			 * go down in flames if truly OOM.
			 *
			 * However, since graphics tend to be disposable,
			 * defer the oom here by reporting the ENOMEM back
			 * to userspace.
			 */
			if (!*s) {
				/* reclaim and warn, but no oom */
				gfp = mapping_gfp_mask(mapping);

				/*
				 * Our bo are always dirty and so we require
				 * kswapd to reclaim our pages (direct reclaim
				 * does not effectively begin pageout of our
				 * buffers on its own). However, direct reclaim
				 * only waits for kswapd when under allocation
				 * congestion. So as a result __GFP_RECLAIM is
				 * unreliable and fails to actually reclaim our
				 * dirty pages -- unless you try over and over
				 * again with !__GFP_NORETRY. However, we still
				 * want to fail this allocation rather than
				 * trigger the out-of-memory killer and for
				 * this we want __GFP_RETRY_MAYFAIL.
				 */
				gfp |= __GFP_RETRY_MAYFAIL;
			}
		} while (1);

		if (!i ||
		    sg->length >= max_segment ||
		    page_to_pfn(page) != last_pfn + 1) {
			if (i) {
				sg_page_sizes |= sg->length;
				sg = sg_next(sg);
			}
			st->nents++;
			sg_set_page(sg, page, PAGE_SIZE, 0);
		} else {
			sg->length += PAGE_SIZE;
		}
		last_pfn = page_to_pfn(page);

		/* Check that the i965g/gm workaround works. */
		WARN_ON((gfp & __GFP_DMA32) && (last_pfn >= 0x00100000UL));
	}
	if (sg) { /* loop terminated early; short sg table */
		sg_page_sizes |= sg->length;
		sg_mark_end(sg);
	}

	/* Trim unused sg entries to avoid wasting memory. */
	i915_sg_trim(st);

	ret = i915_gem_gtt_prepare_pages(obj, st);
	if (ret) {
		/*
		 * DMA remapping failed? One possible cause is that
		 * it could not reserve enough large entries, asking
		 * for PAGE_SIZE chunks instead may be helpful.
		 */
		if (max_segment > PAGE_SIZE) {
			for_each_sgt_page(page, sgt_iter, st)
				put_page(page);
			sg_free_table(st);

			max_segment = PAGE_SIZE;
			goto rebuild_st;
		} else {
			dev_warn(&i915->drm.pdev->dev,
				 "Failed to DMA remap %lu pages\n",
				 page_count);
			goto err_pages;
		}
	}

	if (i915_gem_object_needs_bit17_swizzle(obj))
		i915_gem_object_do_bit_17_swizzle(obj, st);

	__i915_gem_object_set_pages(obj, st, sg_page_sizes);

	return 0;

err_sg:
	sg_mark_end(sg);
err_pages:
	mapping_clear_unevictable(mapping);
	pagevec_init(&pvec);
	for_each_sgt_page(page, sgt_iter, st) {
		if (!pagevec_add(&pvec, page))
			check_release_pagevec(&pvec);
	}
	if (pagevec_count(&pvec))
		check_release_pagevec(&pvec);
	sg_free_table(st);
	kfree(st);

	/*
	 * shmemfs first checks if there is enough memory to allocate the page
	 * and reports ENOSPC should there be insufficient, along with the usual
	 * ENOMEM for a genuine allocation failure.
	 *
	 * We use ENOSPC in our driver to mean that we have run out of aperture
	 * space and so want to translate the error from shmemfs back to our
	 * usual understanding of ENOMEM.
	 */
	if (ret == -ENOSPC)
		ret = -ENOMEM;

	return ret;
}

static void
shmem_truncate(struct drm_i915_gem_object *obj)
{
	/*
	 * Our goal here is to return as much of the memory as
	 * is possible back to the system as we are called from OOM.
	 * To do this we must instruct the shmfs to drop all of its
	 * backing pages, *now*.
	 */
	shmem_truncate_range(file_inode(obj->base.filp), 0, (loff_t)-1);
	obj->mm.madv = __I915_MADV_PURGED;
	obj->mm.pages = ERR_PTR(-EFAULT);
}

static void
shmem_writeback(struct drm_i915_gem_object *obj)
{
	struct address_space *mapping;
	struct writeback_control wbc = {
		.sync_mode = WB_SYNC_NONE,
		.nr_to_write = SWAP_CLUSTER_MAX,
		.range_start = 0,
		.range_end = LLONG_MAX,
		.for_reclaim = 1,
	};
	unsigned long i;

	/*
	 * Leave mmapings intact (GTT will have been revoked on unbinding,
	 * leaving only CPU mmapings around) and add those pages to the LRU
	 * instead of invoking writeback so they are aged and paged out
	 * as normal.
	 */
	mapping = obj->base.filp->f_mapping;

	/* Begin writeback on each dirty page */
	for (i = 0; i < obj->base.size >> PAGE_SHIFT; i++) {
		struct page *page;

		page = find_lock_entry(mapping, i);
		if (!page || xa_is_value(page))
			continue;

		if (!page_mapped(page) && clear_page_dirty_for_io(page)) {
			int ret;

			SetPageReclaim(page);
			ret = mapping->a_ops->writepage(page, &wbc);
			if (!PageWriteback(page))
				ClearPageReclaim(page);
			if (!ret)
				goto put;
		}
		unlock_page(page);
put:
		put_page(page);
	}
}

void
__i915_gem_object_release_shmem(struct drm_i915_gem_object *obj,
				struct sg_table *pages,
				bool needs_clflush)
{
	GEM_BUG_ON(obj->mm.madv == __I915_MADV_PURGED);

	if (obj->mm.madv == I915_MADV_DONTNEED)
		obj->mm.dirty = false;

	if (needs_clflush &&
	    (obj->read_domains & I915_GEM_DOMAIN_CPU) == 0 &&
	    !(obj->cache_coherent & I915_BO_CACHE_COHERENT_FOR_READ))
		drm_clflush_sg(pages);

	__start_cpu_write(obj);
}

static void
shmem_put_pages(struct drm_i915_gem_object *obj, struct sg_table *pages)
{
	struct sgt_iter sgt_iter;
	struct pagevec pvec;
	struct page *page;

	__i915_gem_object_release_shmem(obj, pages, true);

	i915_gem_gtt_finish_pages(obj, pages);

	if (i915_gem_object_needs_bit17_swizzle(obj))
		i915_gem_object_save_bit_17_swizzle(obj, pages);

	mapping_clear_unevictable(file_inode(obj->base.filp)->i_mapping);

	pagevec_init(&pvec);
	for_each_sgt_page(page, sgt_iter, pages) {
		if (obj->mm.dirty)
			set_page_dirty(page);

		if (obj->mm.madv == I915_MADV_WILLNEED)
			mark_page_accessed(page);

		if (!pagevec_add(&pvec, page))
			check_release_pagevec(&pvec);
	}
	if (pagevec_count(&pvec))
		check_release_pagevec(&pvec);
	obj->mm.dirty = false;

	sg_free_table(pages);
	kfree(pages);
}

static int
shmem_pwrite(struct drm_i915_gem_object *obj,
	     const struct drm_i915_gem_pwrite *arg)
{
	struct address_space *mapping = obj->base.filp->f_mapping;
	char __user *user_data = u64_to_user_ptr(arg->data_ptr);
	u64 remain, offset;
	unsigned int pg;

	/* Caller already validated user args */
	GEM_BUG_ON(!access_ok(user_data, arg->size));

	/*
	 * Before we instantiate/pin the backing store for our use, we
	 * can prepopulate the shmemfs filp efficiently using a write into
	 * the pagecache. We avoid the penalty of instantiating all the
	 * pages, important if the user is just writing to a few and never
	 * uses the object on the GPU, and using a direct write into shmemfs
	 * allows it to avoid the cost of retrieving a page (either swapin
	 * or clearing-before-use) before it is overwritten.
	 */
	if (i915_gem_object_has_pages(obj))
		return -ENODEV;

	if (obj->mm.madv != I915_MADV_WILLNEED)
		return -EFAULT;

	/*
	 * Before the pages are instantiated the object is treated as being
	 * in the CPU domain. The pages will be clflushed as required before
	 * use, and we can freely write into the pages directly. If userspace
	 * races pwrite with any other operation; corruption will ensue -
	 * that is userspace's prerogative!
	 */

	remain = arg->size;
	offset = arg->offset;
	pg = offset_in_page(offset);

	do {
		unsigned int len, unwritten;
		struct page *page;
		void *data, *vaddr;
		int err;
		char c;

		len = PAGE_SIZE - pg;
		if (len > remain)
			len = remain;

		/* Prefault the user page to reduce potential recursion */
		err = __get_user(c, user_data);
		if (err)
			return err;

		err = __get_user(c, user_data + len - 1);
		if (err)
			return err;

		err = pagecache_write_begin(obj->base.filp, mapping,
					    offset, len, 0,
					    &page, &data);
		if (err < 0)
			return err;

		vaddr = kmap_atomic(page);
		unwritten = __copy_from_user_inatomic(vaddr + pg,
						      user_data,
						      len);
		kunmap_atomic(vaddr);

		err = pagecache_write_end(obj->base.filp, mapping,
					  offset, len, len - unwritten,
					  page, data);
		if (err < 0)
			return err;

		/* We don't handle -EFAULT, leave it to the caller to check */
		if (unwritten)
			return -ENODEV;

		remain -= len;
		user_data += len;
		offset += len;
		pg = 0;
	} while (remain);

	return 0;
}

static void shmem_release(struct drm_i915_gem_object *obj)
{
	fput(obj->base.filp);
}

const struct drm_i915_gem_object_ops i915_gem_shmem_ops = {
	.flags = I915_GEM_OBJECT_HAS_STRUCT_PAGE |
		 I915_GEM_OBJECT_IS_SHRINKABLE,

	.get_pages = shmem_get_pages,
	.put_pages = shmem_put_pages,
	.truncate = shmem_truncate,
	.writeback = shmem_writeback,

	.pwrite = shmem_pwrite,

	.release = shmem_release,
};

static int create_shmem(struct drm_i915_private *i915,
			struct drm_gem_object *obj,
			size_t size)
{
	unsigned long flags = VM_NORESERVE;
	struct file *filp;

	drm_gem_private_object_init(&i915->drm, obj, size);

	if (i915->mm.gemfs)
		filp = shmem_file_setup_with_mnt(i915->mm.gemfs, "i915", size,
						 flags);
	else
		filp = shmem_file_setup("i915", size, flags);
	if (IS_ERR(filp))
		return PTR_ERR(filp);

	obj->filp = filp;
	return 0;
}

struct drm_i915_gem_object *
i915_gem_object_create_shmem(struct drm_i915_private *i915, u64 size)
{
	struct drm_i915_gem_object *obj;
	struct address_space *mapping;
	unsigned int cache_level;
	gfp_t mask;
	int ret;

	/* There is a prevalence of the assumption that we fit the object's
	 * page count inside a 32bit _signed_ variable. Let's document this and
	 * catch if we ever need to fix it. In the meantime, if you do spot
	 * such a local variable, please consider fixing!
	 */
	if (size >> PAGE_SHIFT > INT_MAX)
		return ERR_PTR(-E2BIG);

	if (overflows_type(size, obj->base.size))
		return ERR_PTR(-E2BIG);

	obj = i915_gem_object_alloc();
	if (!obj)
		return ERR_PTR(-ENOMEM);

	ret = create_shmem(i915, &obj->base, size);
	if (ret)
		goto fail;

	mask = GFP_HIGHUSER | __GFP_RECLAIMABLE;
	if (IS_I965GM(i915) || IS_I965G(i915)) {
		/* 965gm cannot relocate objects above 4GiB. */
		mask &= ~__GFP_HIGHMEM;
		mask |= __GFP_DMA32;
	}

	mapping = obj->base.filp->f_mapping;
	mapping_set_gfp_mask(mapping, mask);
	GEM_BUG_ON(!(mapping_gfp_mask(mapping) & __GFP_RECLAIM));

	i915_gem_object_init(obj, &i915_gem_shmem_ops);

	obj->write_domain = I915_GEM_DOMAIN_CPU;
	obj->read_domains = I915_GEM_DOMAIN_CPU;

	if (HAS_LLC(i915))
		/* On some devices, we can have the GPU use the LLC (the CPU
		 * cache) for about a 10% performance improvement
		 * compared to uncached.  Graphics requests other than
		 * display scanout are coherent with the CPU in
		 * accessing this cache.  This means in this mode we
		 * don't need to clflush on the CPU side, and on the
		 * GPU side we only need to flush internal caches to
		 * get data visible to the CPU.
		 *
		 * However, we maintain the display planes as UC, and so
		 * need to rebind when first used as such.
		 */
		cache_level = I915_CACHE_LLC;
	else
		cache_level = I915_CACHE_NONE;

	i915_gem_object_set_cache_coherency(obj, cache_level);

	trace_i915_gem_object_create(obj);

	return obj;

fail:
	i915_gem_object_free(obj);
	return ERR_PTR(ret);
}

/* Allocate a new GEM object and fill it with the supplied data */
struct drm_i915_gem_object *
i915_gem_object_create_shmem_from_data(struct drm_i915_private *dev_priv,
				       const void *data, size_t size)
{
	struct drm_i915_gem_object *obj;
	struct file *file;
	size_t offset;
	int err;

	obj = i915_gem_object_create_shmem(dev_priv, round_up(size, PAGE_SIZE));
	if (IS_ERR(obj))
		return obj;

	GEM_BUG_ON(obj->write_domain != I915_GEM_DOMAIN_CPU);

	file = obj->base.filp;
	offset = 0;
	do {
		unsigned int len = min_t(typeof(size), size, PAGE_SIZE);
		struct page *page;
		void *pgdata, *vaddr;

		err = pagecache_write_begin(file, file->f_mapping,
					    offset, len, 0,
					    &page, &pgdata);
		if (err < 0)
			goto fail;

		vaddr = kmap(page);
		memcpy(vaddr, data, len);
		kunmap(page);

		err = pagecache_write_end(file, file->f_mapping,
					  offset, len, len,
					  page, pgdata);
		if (err < 0)
			goto fail;

		size -= len;
		data += len;
		offset += len;
	} while (size);

	return obj;

fail:
	i915_gem_object_put(obj);
	return ERR_PTR(err);
}