summaryrefslogtreecommitdiffstats
path: root/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
blob: b6333f92ba4565e9b5949f48e643f2d47daaa2a0 (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
/*
 * Copyright 2016 Advanced Micro Devices, Inc.
 *
 * 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 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 COPYRIGHT HOLDER(S) OR AUTHOR(S) 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: Christian König
 */

#include <drm/drmP.h>
#include "amdgpu.h"

struct amdgpu_vram_mgr {
	struct drm_mm mm;
	spinlock_t lock;
	atomic64_t usage;
	atomic64_t vis_usage;
};

/**
 * amdgpu_vram_mgr_init - init VRAM manager and DRM MM
 *
 * @man: TTM memory type manager
 * @p_size: maximum size of VRAM
 *
 * Allocate and initialize the VRAM manager.
 */
static int amdgpu_vram_mgr_init(struct ttm_mem_type_manager *man,
				unsigned long p_size)
{
	struct amdgpu_vram_mgr *mgr;

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

	drm_mm_init(&mgr->mm, 0, p_size);
	spin_lock_init(&mgr->lock);
	man->priv = mgr;
	return 0;
}

/**
 * amdgpu_vram_mgr_fini - free and destroy VRAM manager
 *
 * @man: TTM memory type manager
 *
 * Destroy and free the VRAM manager, returns -EBUSY if ranges are still
 * allocated inside it.
 */
static int amdgpu_vram_mgr_fini(struct ttm_mem_type_manager *man)
{
	struct amdgpu_vram_mgr *mgr = man->priv;

	spin_lock(&mgr->lock);
	drm_mm_takedown(&mgr->mm);
	spin_unlock(&mgr->lock);
	kfree(mgr);
	man->priv = NULL;
	return 0;
}

/**
 * amdgpu_vram_mgr_vis_size - Calculate visible node size
 *
 * @adev: amdgpu device structure
 * @node: MM node structure
 *
 * Calculate how many bytes of the MM node are inside visible VRAM
 */
static u64 amdgpu_vram_mgr_vis_size(struct amdgpu_device *adev,
				    struct drm_mm_node *node)
{
	uint64_t start = node->start << PAGE_SHIFT;
	uint64_t end = (node->size + node->start) << PAGE_SHIFT;

	if (start >= adev->gmc.visible_vram_size)
		return 0;

	return (end > adev->gmc.visible_vram_size ?
		adev->gmc.visible_vram_size : end) - start;
}

/**
 * amdgpu_vram_mgr_bo_invisible_size - CPU invisible BO size
 *
 * @bo: &amdgpu_bo buffer object (must be in VRAM)
 *
 * Returns:
 * How much of the given &amdgpu_bo buffer object lies in CPU invisible VRAM.
 */
u64 amdgpu_vram_mgr_bo_invisible_size(struct amdgpu_bo *bo)
{
	struct amdgpu_device *adev = amdgpu_ttm_adev(bo->tbo.bdev);
	struct ttm_mem_reg *mem = &bo->tbo.mem;
	struct drm_mm_node *nodes = mem->mm_node;
	unsigned pages = mem->num_pages;
	u64 usage = 0;

	if (adev->gmc.visible_vram_size == adev->gmc.real_vram_size)
		return 0;

	if (mem->start >= adev->gmc.visible_vram_size >> PAGE_SHIFT)
		return amdgpu_bo_size(bo);

	while (nodes && pages) {
		usage += nodes->size << PAGE_SHIFT;
		usage -= amdgpu_vram_mgr_vis_size(adev, nodes);
		pages -= nodes->size;
		++nodes;
	}

	return usage;
}

/**
 * amdgpu_vram_mgr_new - allocate new ranges
 *
 * @man: TTM memory type manager
 * @tbo: TTM BO we need this range for
 * @place: placement flags and restrictions
 * @mem: the resulting mem object
 *
 * Allocate VRAM for the given BO.
 */
static int amdgpu_vram_mgr_new(struct ttm_mem_type_manager *man,
			       struct ttm_buffer_object *tbo,
			       const struct ttm_place *place,
			       struct ttm_mem_reg *mem)
{
	struct amdgpu_device *adev = amdgpu_ttm_adev(man->bdev);
	struct amdgpu_vram_mgr *mgr = man->priv;
	struct drm_mm *mm = &mgr->mm;
	struct drm_mm_node *nodes;
	enum drm_mm_insert_mode mode;
	unsigned long lpfn, num_nodes, pages_per_node, pages_left;
	uint64_t usage = 0, vis_usage = 0;
	unsigned i;
	int r;

	lpfn = place->lpfn;
	if (!lpfn)
		lpfn = man->size;

	if (place->flags & TTM_PL_FLAG_CONTIGUOUS ||
	    amdgpu_vram_page_split == -1) {
		pages_per_node = ~0ul;
		num_nodes = 1;
	} else {
		pages_per_node = max((uint32_t)amdgpu_vram_page_split,
				     mem->page_alignment);
		num_nodes = DIV_ROUND_UP(mem->num_pages, pages_per_node);
	}

	nodes = kvmalloc_array(num_nodes, sizeof(*nodes),
			       GFP_KERNEL | __GFP_ZERO);
	if (!nodes)
		return -ENOMEM;

	mode = DRM_MM_INSERT_BEST;
	if (place->flags & TTM_PL_FLAG_TOPDOWN)
		mode = DRM_MM_INSERT_HIGH;

	mem->start = 0;
	pages_left = mem->num_pages;

	spin_lock(&mgr->lock);
	for (i = 0; i < num_nodes; ++i) {
		unsigned long pages = min(pages_left, pages_per_node);
		uint32_t alignment = mem->page_alignment;
		unsigned long start;

		if (pages == pages_per_node)
			alignment = pages_per_node;

		r = drm_mm_insert_node_in_range(mm, &nodes[i],
						pages, alignment, 0,
						place->fpfn, lpfn,
						mode);
		if (unlikely(r))
			goto error;

		usage += nodes[i].size << PAGE_SHIFT;
		vis_usage += amdgpu_vram_mgr_vis_size(adev, &nodes[i]);

		/* Calculate a virtual BO start address to easily check if
		 * everything is CPU accessible.
		 */
		start = nodes[i].start + nodes[i].size;
		if (start > mem->num_pages)
			start -= mem->num_pages;
		else
			start = 0;
		mem->start = max(mem->start, start);
		pages_left -= pages;
	}
	spin_unlock(&mgr->lock);

	atomic64_add(usage, &mgr->usage);
	atomic64_add(vis_usage, &mgr->vis_usage);

	mem->mm_node = nodes;

	return 0;

error:
	while (i--)
		drm_mm_remove_node(&nodes[i]);
	spin_unlock(&mgr->lock);

	kvfree(nodes);
	return r == -ENOSPC ? 0 : r;
}

/**
 * amdgpu_vram_mgr_del - free ranges
 *
 * @man: TTM memory type manager
 * @tbo: TTM BO we need this range for
 * @place: placement flags and restrictions
 * @mem: TTM memory object
 *
 * Free the allocated VRAM again.
 */
static void amdgpu_vram_mgr_del(struct ttm_mem_type_manager *man,
				struct ttm_mem_reg *mem)
{
	struct amdgpu_device *adev = amdgpu_ttm_adev(man->bdev);
	struct amdgpu_vram_mgr *mgr = man->priv;
	struct drm_mm_node *nodes = mem->mm_node;
	uint64_t usage = 0, vis_usage = 0;
	unsigned pages = mem->num_pages;

	if (!mem->mm_node)
		return;

	spin_lock(&mgr->lock);
	while (pages) {
		pages -= nodes->size;
		drm_mm_remove_node(nodes);
		usage += nodes->size << PAGE_SHIFT;
		vis_usage += amdgpu_vram_mgr_vis_size(adev, nodes);
		++nodes;
	}
	spin_unlock(&mgr->lock);

	atomic64_sub(usage, &mgr->usage);
	atomic64_sub(vis_usage, &mgr->vis_usage);

	kvfree(mem->mm_node);
	mem->mm_node = NULL;
}

/**
 * amdgpu_vram_mgr_usage - how many bytes are used in this domain
 *
 * @man: TTM memory type manager
 *
 * Returns how many bytes are used in this domain.
 */
uint64_t amdgpu_vram_mgr_usage(struct ttm_mem_type_manager *man)
{
	struct amdgpu_vram_mgr *mgr = man->priv;

	return atomic64_read(&mgr->usage);
}

/**
 * amdgpu_vram_mgr_vis_usage - how many bytes are used in the visible part
 *
 * @man: TTM memory type manager
 *
 * Returns how many bytes are used in the visible part of VRAM
 */
uint64_t amdgpu_vram_mgr_vis_usage(struct ttm_mem_type_manager *man)
{
	struct amdgpu_vram_mgr *mgr = man->priv;

	return atomic64_read(&mgr->vis_usage);
}

/**
 * amdgpu_vram_mgr_debug - dump VRAM table
 *
 * @man: TTM memory type manager
 * @printer: DRM printer to use
 *
 * Dump the table content using printk.
 */
static void amdgpu_vram_mgr_debug(struct ttm_mem_type_manager *man,
				  struct drm_printer *printer)
{
	struct amdgpu_vram_mgr *mgr = man->priv;

	spin_lock(&mgr->lock);
	drm_mm_print(&mgr->mm, printer);
	spin_unlock(&mgr->lock);

	drm_printf(printer, "man size:%llu pages, ram usage:%lluMB, vis usage:%lluMB\n",
		   man->size, amdgpu_vram_mgr_usage(man) >> 20,
		   amdgpu_vram_mgr_vis_usage(man) >> 20);
}

const struct ttm_mem_type_manager_func amdgpu_vram_mgr_func = {
	.init		= amdgpu_vram_mgr_init,
	.takedown	= amdgpu_vram_mgr_fini,
	.get_node	= amdgpu_vram_mgr_new,
	.put_node	= amdgpu_vram_mgr_del,
	.debug		= amdgpu_vram_mgr_debug
};
OpenPOWER on IntegriCloud