summaryrefslogtreecommitdiffstats
path: root/drivers/gpu/drm/amd/amdgpu/amdgpu_benchmark.c
blob: edc6377ec5ff2c7c534aa7b77da93ac55bbeab56 (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
/*
 * Copyright 2009 Jerome Glisse.
 *
 * 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: Jerome Glisse
 */

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

#define AMDGPU_BENCHMARK_ITERATIONS 1024
#define AMDGPU_BENCHMARK_COMMON_MODES_N 17

static int amdgpu_benchmark_do_move(struct amdgpu_device *adev, unsigned size,
				    uint64_t saddr, uint64_t daddr, int n, s64 *time_ms)
{
	ktime_t stime, etime;
	struct dma_fence *fence;
	int i, r;

	stime = ktime_get();
	for (i = 0; i < n; i++) {
		struct amdgpu_ring *ring = adev->mman.buffer_funcs_ring;
		r = amdgpu_copy_buffer(ring, saddr, daddr, size, NULL, &fence,
				       false, false, false);
		if (r)
			goto exit_do_move;
		r = dma_fence_wait(fence, false);
		dma_fence_put(fence);
		if (r)
			goto exit_do_move;
	}

exit_do_move:
	etime = ktime_get();
	*time_ms = ktime_ms_delta(etime, stime);

	return r;
}


static void amdgpu_benchmark_log_results(struct amdgpu_device *adev,
					 int n, unsigned size,
					 s64 time_ms,
					 unsigned sdomain, unsigned ddomain,
					 char *kind)
{
	s64 throughput = (n * (size >> 10));

	throughput = div64_s64(throughput, time_ms);

	dev_info(adev->dev, "amdgpu: %s %u bo moves of %u kB from"
		 " %d to %d in %lld ms, throughput: %lld Mb/s or %lld MB/s\n",
		 kind, n, size >> 10, sdomain, ddomain, time_ms,
		 throughput * 8, throughput);
}

static int amdgpu_benchmark_move(struct amdgpu_device *adev, unsigned size,
				 unsigned sdomain, unsigned ddomain)
{
	struct amdgpu_bo *dobj = NULL;
	struct amdgpu_bo *sobj = NULL;
	uint64_t saddr, daddr;
	s64 time_ms;
	int r, n;

	n = AMDGPU_BENCHMARK_ITERATIONS;

	r = amdgpu_bo_create_kernel(adev, size,
				    PAGE_SIZE, sdomain,
				    &sobj,
				    &saddr,
				    NULL);
	if (r)
		goto out_cleanup;
	r = amdgpu_bo_create_kernel(adev, size,
				    PAGE_SIZE, ddomain,
				    &dobj,
				    &daddr,
				    NULL);
	if (r)
		goto out_cleanup;

	if (adev->mman.buffer_funcs) {
		r = amdgpu_benchmark_do_move(adev, size, saddr, daddr, n, &time_ms);
		if (r)
			goto out_cleanup;
		else
			amdgpu_benchmark_log_results(adev, n, size, time_ms,
						     sdomain, ddomain, "dma");
	}

out_cleanup:
	/* Check error value now. The value can be overwritten when clean up.*/
	if (r < 0)
		dev_info(adev->dev, "Error while benchmarking BO move.\n");

	if (sobj)
		amdgpu_bo_free_kernel(&sobj, &saddr, NULL);
	if (dobj)
		amdgpu_bo_free_kernel(&dobj, &daddr, NULL);
	return r;
}

int amdgpu_benchmark(struct amdgpu_device *adev, int test_number)
{
	int i, r;
	static const int common_modes[AMDGPU_BENCHMARK_COMMON_MODES_N] = {
		640 * 480 * 4,
		720 * 480 * 4,
		800 * 600 * 4,
		848 * 480 * 4,
		1024 * 768 * 4,
		1152 * 768 * 4,
		1280 * 720 * 4,
		1280 * 800 * 4,
		1280 * 854 * 4,
		1280 * 960 * 4,
		1280 * 1024 * 4,
		1440 * 900 * 4,
		1400 * 1050 * 4,
		1680 * 1050 * 4,
		1600 * 1200 * 4,
		1920 * 1080 * 4,
		1920 * 1200 * 4
	};

	mutex_lock(&adev->benchmark_mutex);
	switch (test_number) {
	case 1:
		dev_info(adev->dev,
			 "benchmark test: %d (simple test, VRAM to GTT and GTT to VRAM)\n",
			 test_number);
		/* simple test, VRAM to GTT and GTT to VRAM */
		r = amdgpu_benchmark_move(adev, 1024*1024, AMDGPU_GEM_DOMAIN_GTT,
					  AMDGPU_GEM_DOMAIN_VRAM);
		if (r)
			goto done;
		r = amdgpu_benchmark_move(adev, 1024*1024, AMDGPU_GEM_DOMAIN_VRAM,
					  AMDGPU_GEM_DOMAIN_GTT);
		if (r)
			goto done;
		break;
	case 2:
		dev_info(adev->dev,
			 "benchmark test: %d (simple test, VRAM to VRAM)\n",
			 test_number);
		/* simple test, VRAM to VRAM */
		r = amdgpu_benchmark_move(adev, 1024*1024, AMDGPU_GEM_DOMAIN_VRAM,
					  AMDGPU_GEM_DOMAIN_VRAM);
		if (r)
			goto done;
		break;
	case 3:
		dev_info(adev->dev,
			 "benchmark test: %d (GTT to VRAM, buffer size sweep, powers of 2)\n",
			 test_number);
		/* GTT to VRAM, buffer size sweep, powers of 2 */
		for (i = 1; i <= 16384; i <<= 1) {
			r = amdgpu_benchmark_move(adev, i * AMDGPU_GPU_PAGE_SIZE,
						  AMDGPU_GEM_DOMAIN_GTT,
						  AMDGPU_GEM_DOMAIN_VRAM);
			if (r)
				goto done;
		}
		break;
	case 4:
		dev_info(adev->dev,
			 "benchmark test: %d (VRAM to GTT, buffer size sweep, powers of 2)\n",
			 test_number);
		/* VRAM to GTT, buffer size sweep, powers of 2 */
		for (i = 1; i <= 16384; i <<= 1) {
			r = amdgpu_benchmark_move(adev, i * AMDGPU_GPU_PAGE_SIZE,
						  AMDGPU_GEM_DOMAIN_VRAM,
						  AMDGPU_GEM_DOMAIN_GTT);
			if (r)
				goto done;
		}
		break;
	case 5:
		dev_info(adev->dev,
			 "benchmark test: %d (VRAM to VRAM, buffer size sweep, powers of 2)\n",
			 test_number);
		/* VRAM to VRAM, buffer size sweep, powers of 2 */
		for (i = 1; i <= 16384; i <<= 1) {
			r = amdgpu_benchmark_move(adev, i * AMDGPU_GPU_PAGE_SIZE,
						  AMDGPU_GEM_DOMAIN_VRAM,
						  AMDGPU_GEM_DOMAIN_VRAM);
			if (r)
				goto done;
		}
		break;
	case 6:
		dev_info(adev->dev,
			 "benchmark test: %d (GTT to VRAM, buffer size sweep, common modes)\n",
			 test_number);
		/* GTT to VRAM, buffer size sweep, common modes */
		for (i = 0; i < AMDGPU_BENCHMARK_COMMON_MODES_N; i++) {
			r = amdgpu_benchmark_move(adev, common_modes[i],
						  AMDGPU_GEM_DOMAIN_GTT,
						  AMDGPU_GEM_DOMAIN_VRAM);
			if (r)
				goto done;
		}
		break;
	case 7:
		dev_info(adev->dev,
			 "benchmark test: %d (VRAM to GTT, buffer size sweep, common modes)\n",
			 test_number);
		/* VRAM to GTT, buffer size sweep, common modes */
		for (i = 0; i < AMDGPU_BENCHMARK_COMMON_MODES_N; i++) {
			r = amdgpu_benchmark_move(adev, common_modes[i],
						  AMDGPU_GEM_DOMAIN_VRAM,
						  AMDGPU_GEM_DOMAIN_GTT);
			if (r)
				goto done;
		}
		break;
	case 8:
		dev_info(adev->dev,
			 "benchmark test: %d (VRAM to VRAM, buffer size sweep, common modes)\n",
			 test_number);
		/* VRAM to VRAM, buffer size sweep, common modes */
		for (i = 0; i < AMDGPU_BENCHMARK_COMMON_MODES_N; i++) {
			r = amdgpu_benchmark_move(adev, common_modes[i],
					      AMDGPU_GEM_DOMAIN_VRAM,
					      AMDGPU_GEM_DOMAIN_VRAM);
			if (r)
				goto done;
		}
		break;

	default:
		dev_info(adev->dev, "Unknown benchmark %d\n", test_number);
		r = -EINVAL;
		break;
	}

done:
	mutex_unlock(&adev->benchmark_mutex);

	return r;
}