2 * Copyright 2009 Jerome Glisse.
4 * Permission is hereby granted, free of charge, to any person obtaining a
5 * copy of this software and associated documentation files (the "Software"),
6 * to deal in the Software without restriction, including without limitation
7 * the rights to use, copy, modify, merge, publish, distribute, sublicense,
8 * and/or sell copies of the Software, and to permit persons to whom the
9 * Software is furnished to do so, subject to the following conditions:
11 * The above copyright notice and this permission notice shall be included in
12 * all copies or substantial portions of the Software.
14 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
15 * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
16 * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
17 * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
18 * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
19 * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
20 * OTHER DEALINGS IN THE SOFTWARE.
22 * Authors: Jerome Glisse
25 #include <drm/amdgpu_drm.h>
28 #define AMDGPU_BENCHMARK_ITERATIONS 1024
29 #define AMDGPU_BENCHMARK_COMMON_MODES_N 17
31 static int amdgpu_benchmark_do_move(struct amdgpu_device *adev, unsigned size,
32 uint64_t saddr, uint64_t daddr, int n, s64 *time_ms)
35 struct dma_fence *fence;
39 for (i = 0; i < n; i++) {
40 struct amdgpu_ring *ring = adev->mman.buffer_funcs_ring;
41 r = amdgpu_copy_buffer(ring, saddr, daddr, size, NULL, &fence,
45 r = dma_fence_wait(fence, false);
53 *time_ms = ktime_ms_delta(etime, stime);
59 static void amdgpu_benchmark_log_results(struct amdgpu_device *adev,
62 unsigned sdomain, unsigned ddomain,
65 s64 throughput = (n * (size >> 10));
67 throughput = div64_s64(throughput, time_ms);
69 dev_info(adev->dev, "amdgpu: %s %u bo moves of %u kB from"
70 " %d to %d in %lld ms, throughput: %lld Mb/s or %lld MB/s\n",
71 kind, n, size >> 10, sdomain, ddomain, time_ms,
72 throughput * 8, throughput);
75 static int amdgpu_benchmark_move(struct amdgpu_device *adev, unsigned size,
76 unsigned sdomain, unsigned ddomain)
78 struct amdgpu_bo *dobj = NULL;
79 struct amdgpu_bo *sobj = NULL;
80 uint64_t saddr, daddr;
84 n = AMDGPU_BENCHMARK_ITERATIONS;
86 r = amdgpu_bo_create_kernel(adev, size,
93 r = amdgpu_bo_create_kernel(adev, size,
101 if (adev->mman.buffer_funcs) {
102 r = amdgpu_benchmark_do_move(adev, size, saddr, daddr, n, &time_ms);
106 amdgpu_benchmark_log_results(adev, n, size, time_ms,
107 sdomain, ddomain, "dma");
111 /* Check error value now. The value can be overwritten when clean up.*/
113 dev_info(adev->dev, "Error while benchmarking BO move.\n");
116 amdgpu_bo_free_kernel(&sobj, &saddr, NULL);
118 amdgpu_bo_free_kernel(&dobj, &daddr, NULL);
122 int amdgpu_benchmark(struct amdgpu_device *adev, int test_number)
125 static const int common_modes[AMDGPU_BENCHMARK_COMMON_MODES_N] = {
145 mutex_lock(&adev->benchmark_mutex);
146 switch (test_number) {
149 "benchmark test: %d (simple test, VRAM to GTT and GTT to VRAM)\n",
151 /* simple test, VRAM to GTT and GTT to VRAM */
152 r = amdgpu_benchmark_move(adev, 1024*1024, AMDGPU_GEM_DOMAIN_GTT,
153 AMDGPU_GEM_DOMAIN_VRAM);
156 r = amdgpu_benchmark_move(adev, 1024*1024, AMDGPU_GEM_DOMAIN_VRAM,
157 AMDGPU_GEM_DOMAIN_GTT);
163 "benchmark test: %d (simple test, VRAM to VRAM)\n",
165 /* simple test, VRAM to VRAM */
166 r = amdgpu_benchmark_move(adev, 1024*1024, AMDGPU_GEM_DOMAIN_VRAM,
167 AMDGPU_GEM_DOMAIN_VRAM);
173 "benchmark test: %d (GTT to VRAM, buffer size sweep, powers of 2)\n",
175 /* GTT to VRAM, buffer size sweep, powers of 2 */
176 for (i = 1; i <= 16384; i <<= 1) {
177 r = amdgpu_benchmark_move(adev, i * AMDGPU_GPU_PAGE_SIZE,
178 AMDGPU_GEM_DOMAIN_GTT,
179 AMDGPU_GEM_DOMAIN_VRAM);
186 "benchmark test: %d (VRAM to GTT, buffer size sweep, powers of 2)\n",
188 /* VRAM to GTT, buffer size sweep, powers of 2 */
189 for (i = 1; i <= 16384; i <<= 1) {
190 r = amdgpu_benchmark_move(adev, i * AMDGPU_GPU_PAGE_SIZE,
191 AMDGPU_GEM_DOMAIN_VRAM,
192 AMDGPU_GEM_DOMAIN_GTT);
199 "benchmark test: %d (VRAM to VRAM, buffer size sweep, powers of 2)\n",
201 /* VRAM to VRAM, buffer size sweep, powers of 2 */
202 for (i = 1; i <= 16384; i <<= 1) {
203 r = amdgpu_benchmark_move(adev, i * AMDGPU_GPU_PAGE_SIZE,
204 AMDGPU_GEM_DOMAIN_VRAM,
205 AMDGPU_GEM_DOMAIN_VRAM);
212 "benchmark test: %d (GTT to VRAM, buffer size sweep, common modes)\n",
214 /* GTT to VRAM, buffer size sweep, common modes */
215 for (i = 0; i < AMDGPU_BENCHMARK_COMMON_MODES_N; i++) {
216 r = amdgpu_benchmark_move(adev, common_modes[i],
217 AMDGPU_GEM_DOMAIN_GTT,
218 AMDGPU_GEM_DOMAIN_VRAM);
225 "benchmark test: %d (VRAM to GTT, buffer size sweep, common modes)\n",
227 /* VRAM to GTT, buffer size sweep, common modes */
228 for (i = 0; i < AMDGPU_BENCHMARK_COMMON_MODES_N; i++) {
229 r = amdgpu_benchmark_move(adev, common_modes[i],
230 AMDGPU_GEM_DOMAIN_VRAM,
231 AMDGPU_GEM_DOMAIN_GTT);
238 "benchmark test: %d (VRAM to VRAM, buffer size sweep, common modes)\n",
240 /* VRAM to VRAM, buffer size sweep, common modes */
241 for (i = 0; i < AMDGPU_BENCHMARK_COMMON_MODES_N; i++) {
242 r = amdgpu_benchmark_move(adev, common_modes[i],
243 AMDGPU_GEM_DOMAIN_VRAM,
244 AMDGPU_GEM_DOMAIN_VRAM);
251 dev_info(adev->dev, "Unknown benchmark %d\n", test_number);
257 mutex_unlock(&adev->benchmark_mutex);