aboutsummaryrefslogtreecommitdiffstats
blob: bed8494a8b666fbda58ebf9d1da98b404d97aa06 (plain) (blame)
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
/*
 * Copyright 2017 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.
 *
*/

#ifdef HAVE_CONFIG_H
#include "config.h"
#endif

#include <stdio.h>
#include <inttypes.h>

#include "CUnit/Basic.h"

#include "util_math.h"

#include "amdgpu_test.h"
#include "amdgpu_drm.h"
#include "amdgpu_internal.h"
#include "frame.h"
#include "uve_ib.h"

#define IB_SIZE		4096
#define MAX_RESOURCES	16

struct amdgpu_uvd_enc_bo {
	amdgpu_bo_handle handle;
	amdgpu_va_handle va_handle;
	uint64_t addr;
	uint64_t size;
	uint8_t *ptr;
};

struct amdgpu_uvd_enc {
	unsigned width;
	unsigned height;
	struct amdgpu_uvd_enc_bo session;
	struct amdgpu_uvd_enc_bo vbuf;
	struct amdgpu_uvd_enc_bo bs;
	struct amdgpu_uvd_enc_bo fb;
	struct amdgpu_uvd_enc_bo cpb;
};

static amdgpu_device_handle device_handle;
static uint32_t major_version;
static uint32_t minor_version;
static uint32_t family_id;

static amdgpu_context_handle context_handle;
static amdgpu_bo_handle ib_handle;
static amdgpu_va_handle ib_va_handle;
static uint64_t ib_mc_address;
static uint32_t *ib_cpu;

static struct amdgpu_uvd_enc enc;
static amdgpu_bo_handle resources[MAX_RESOURCES];
static unsigned num_resources;

static void amdgpu_cs_uvd_enc_create(void);
static void amdgpu_cs_uvd_enc_session_init(void);
static void amdgpu_cs_uvd_enc_encode(void);
static void amdgpu_cs_uvd_enc_destroy(void);


CU_TestInfo uvd_enc_tests[] = {
	{ "UVD ENC create",  amdgpu_cs_uvd_enc_create },
	{ "UVD ENC session init",  amdgpu_cs_uvd_enc_session_init },
	{ "UVD ENC encode",  amdgpu_cs_uvd_enc_encode },
	{ "UVD ENC destroy",  amdgpu_cs_uvd_enc_destroy },
	CU_TEST_INFO_NULL,
};

CU_BOOL suite_uvd_enc_tests_enable(void)
{
	int r;
	struct drm_amdgpu_info_hw_ip info;

	if (amdgpu_device_initialize(drm_amdgpu[0], &major_version,
					     &minor_version, &device_handle))
		return CU_FALSE;

	r = amdgpu_query_hw_ip_info(device_handle, AMDGPU_HW_IP_UVD_ENC, 0, &info);

	if (amdgpu_device_deinitialize(device_handle))
		return CU_FALSE;

	if (!info.available_rings)
		printf("\n\nThe ASIC NOT support UVD ENC, suite disabled.\n");

	return (r == 0 && (info.available_rings ? CU_TRUE : CU_FALSE));
}


int suite_uvd_enc_tests_init(void)
{
	int r;

	r = amdgpu_device_initialize(drm_amdgpu[0], &major_version,
				     &minor_version, &device_handle);
	if (r)
		return CUE_SINIT_FAILED;

	family_id = device_handle->info.family_id;

	r = amdgpu_cs_ctx_create(device_handle, &context_handle);
	if (r)
		return CUE_SINIT_FAILED;

	r = amdgpu_bo_alloc_and_map(device_handle, IB_SIZE, 4096,
				    AMDGPU_GEM_DOMAIN_GTT, 0,
				    &ib_handle, (void**)&ib_cpu,
				    &ib_mc_address, &ib_va_handle);
	if (r)
		return CUE_SINIT_FAILED;

	return CUE_SUCCESS;
}

int suite_uvd_enc_tests_clean(void)
{
	int r;

	r = amdgpu_bo_unmap_and_free(ib_handle, ib_va_handle,
				     ib_mc_address, IB_SIZE);
	if (r)
		return CUE_SCLEAN_FAILED;

	r = amdgpu_cs_ctx_free(context_handle);
	if (r)
		return CUE_SCLEAN_FAILED;

	r = amdgpu_device_deinitialize(device_handle);
	if (r)
		return CUE_SCLEAN_FAILED;

	return CUE_SUCCESS;
}

static int submit(unsigned ndw, unsigned ip)
{
	struct amdgpu_cs_request ibs_request = {0};
	struct amdgpu_cs_ib_info ib_info = {0};
	struct amdgpu_cs_fence fence_status = {0};
	uint32_t expired;
	int r;

	ib_info.ib_mc_address = ib_mc_address;
	ib_info.size = ndw;

	ibs_request.ip_type = ip;

	r = amdgpu_bo_list_create(device_handle, num_resources, resources,
				  NULL, &ibs_request.resources);
	if (r)
		return r;

	ibs_request.number_of_ibs = 1;
	ibs_request.ibs = &ib_info;
	ibs_request.fence_info.handle = NULL;

	r = amdgpu_cs_submit(context_handle, 0, &ibs_request, 1);
	if (r)
		return r;

	r = amdgpu_bo_list_destroy(ibs_request.resources);
	if (r)
		return r;

	fence_status.context = context_handle;
	fence_status.ip_type = ip;
	fence_status.fence = ibs_request.seq_no;

	r = amdgpu_cs_query_fence_status(&fence_status,
					 AMDGPU_TIMEOUT_INFINITE,
					 0, &expired);
	if (r)
		return r;

	return 0;
}

static void alloc_resource(struct amdgpu_uvd_enc_bo *uvd_enc_bo,
			unsigned size, unsigned domain)
{
	struct amdgpu_bo_alloc_request req = {0};
	amdgpu_bo_handle buf_handle;
	amdgpu_va_handle va_handle;
	uint64_t va = 0;
	int r;

	req.alloc_size = ALIGN(size, 4096);
	req.preferred_heap = domain;
	r = amdgpu_bo_alloc(device_handle, &req, &buf_handle);
	CU_ASSERT_EQUAL(r, 0);
	r = amdgpu_va_range_alloc(device_handle,
				  amdgpu_gpu_va_range_general,
				  req.alloc_size, 1, 0, &va,
				  &va_handle, 0);
	CU_ASSERT_EQUAL(r, 0);
	r = amdgpu_bo_va_op(buf_handle, 0, req.alloc_size, va, 0,
			    AMDGPU_VA_OP_MAP);
	CU_ASSERT_EQUAL(r, 0);
	uvd_enc_bo->addr = va;
	uvd_enc_bo->handle = buf_handle;
	uvd_enc_bo->size = req.alloc_size;
	uvd_enc_bo->va_handle = va_handle;
	r = amdgpu_bo_cpu_map(uvd_enc_bo->handle, (void **)&uvd_enc_bo->ptr);
	CU_ASSERT_EQUAL(r, 0);
	memset(uvd_enc_bo->ptr, 0, size);
	r = amdgpu_bo_cpu_unmap(uvd_enc_bo->handle);
	CU_ASSERT_EQUAL(r, 0);
}

static void free_resource(struct amdgpu_uvd_enc_bo *uvd_enc_bo)
{
	int r;

	r = amdgpu_bo_va_op(uvd_enc_bo->handle, 0, uvd_enc_bo->size,
			    uvd_enc_bo->addr, 0, AMDGPU_VA_OP_UNMAP);
	CU_ASSERT_EQUAL(r, 0);

	r = amdgpu_va_range_free(uvd_enc_bo->va_handle);
	CU_ASSERT_EQUAL(r, 0);

	r = amdgpu_bo_free(uvd_enc_bo->handle);
	CU_ASSERT_EQUAL(r, 0);
	memset(uvd_enc_bo, 0, sizeof(*uvd_enc_bo));
}

static void amdgpu_cs_uvd_enc_create(void)
{
	int len, r;

	enc.width = 160;
	enc.height = 128;

	num_resources  = 0;
	alloc_resource(&enc.session, 128 * 1024, AMDGPU_GEM_DOMAIN_GTT);
	resources[num_resources++] = enc.session.handle;
	resources[num_resources++] = ib_handle;
}

static void check_result(struct amdgpu_uvd_enc *enc)
{
	uint64_t sum;
	uint32_t s = 175602;
	uint32_t *ptr, size;
	int i, j, r;

	r = amdgpu_bo_cpu_map(enc->fb.handle, (void **)&enc->fb.ptr);
	CU_ASSERT_EQUAL(r, 0);
	ptr = (uint32_t *)enc->fb.ptr;
	size = ptr[6];
	r = amdgpu_bo_cpu_unmap(enc->fb.handle);
	CU_ASSERT_EQUAL(r, 0);
	r = amdgpu_bo_cpu_map(enc->bs.handle, (void **)&enc->bs.ptr);
	CU_ASSERT_EQUAL(r, 0);
	for (j = 0, sum = 0; j < size; ++j)
		sum += enc->bs.ptr[j];
	CU_ASSERT_EQUAL(sum, s);
	r = amdgpu_bo_cpu_unmap(enc->bs.handle);
	CU_ASSERT_EQUAL(r, 0);

}

static void amdgpu_cs_uvd_enc_session_init(void)
{
	int len, r;

	len = 0;
	memcpy((ib_cpu + len), uve_session_info, sizeof(uve_session_info));
	len += sizeof(uve_session_info) / 4;
	ib_cpu[len++] = enc.session.addr >> 32;
	ib_cpu[len++] = enc.session.addr;

	memcpy((ib_cpu + len), uve_task_info, sizeof(uve_task_info));
	len += sizeof(uve_task_info) / 4;
	ib_cpu[len++] = 0x000000d8;
	ib_cpu[len++] = 0x00000000;
	ib_cpu[len++] = 0x00000000;

	memcpy((ib_cpu + len), uve_op_init, sizeof(uve_op_init));
	len += sizeof(uve_op_init) / 4;

	memcpy((ib_cpu + len), uve_session_init, sizeof(uve_session_init));
	len += sizeof(uve_session_init) / 4;

	memcpy((ib_cpu + len), uve_layer_ctrl, sizeof(uve_layer_ctrl));
	len += sizeof(uve_layer_ctrl) / 4;

	memcpy((ib_cpu + len), uve_slice_ctrl, sizeof(uve_slice_ctrl));
	len += sizeof(uve_slice_ctrl) / 4;

	memcpy((ib_cpu + len), uve_spec_misc, sizeof(uve_spec_misc));
	len += sizeof(uve_spec_misc) / 4;

	memcpy((ib_cpu + len), uve_rc_session_init, sizeof(uve_rc_session_init));
	len += sizeof(uve_rc_session_init) / 4;

	memcpy((ib_cpu + len), uve_deblocking_filter, sizeof(uve_deblocking_filter));
	len += sizeof(uve_deblocking_filter) / 4;

	memcpy((ib_cpu + len), uve_quality_params, sizeof(uve_quality_params));
	len += sizeof(uve_quality_params) / 4;

	memcpy((ib_cpu + len), uve_op_init_rc, sizeof(uve_op_init_rc));
	len += sizeof(uve_op_init_rc) / 4;

	memcpy((ib_cpu + len), uve_op_init_rc_vbv_level, sizeof(uve_op_init_rc_vbv_level));
	len += sizeof(uve_op_init_rc_vbv_level) / 4;

	r = submit(len, AMDGPU_HW_IP_UVD_ENC);
	CU_ASSERT_EQUAL(r, 0);
}

static void amdgpu_cs_uvd_enc_encode(void)
{
	int len, r, i;
	uint64_t luma_offset, chroma_offset;
	uint32_t vbuf_size, bs_size = 0x003f4800, cpb_size;
	unsigned align = (family_id >= AMDGPU_FAMILY_AI) ? 256 : 16;
	vbuf_size = ALIGN(enc.width, align) * ALIGN(enc.height, 16) * 1.5;
	cpb_size = vbuf_size * 10;


	num_resources  = 0;
	alloc_resource(&enc.fb, 4096, AMDGPU_GEM_DOMAIN_VRAM);
	resources[num_resources++] = enc.fb.handle;
	alloc_resource(&enc.bs, bs_size, AMDGPU_GEM_DOMAIN_VRAM);
	resources[num_resources++] = enc.bs.handle;
	alloc_resource(&enc.vbuf, vbuf_size, AMDGPU_GEM_DOMAIN_VRAM);
	resources[num_resources++] = enc.vbuf.handle;
	alloc_resource(&enc.cpb, cpb_size, AMDGPU_GEM_DOMAIN_VRAM);
	resources[num_resources++] = enc.cpb.handle;
	resources[num_resources++] = ib_handle;

	r = amdgpu_bo_cpu_map(enc.vbuf.handle, (void **)&enc.vbuf.ptr);
	CU_ASSERT_EQUAL(r, 0);

	memset(enc.vbuf.ptr, 0, vbuf_size);
	for (i = 0; i < enc.height; ++i) {
		memcpy(enc.vbuf.ptr, (frame + i * enc.width), enc.width);
		enc.vbuf.ptr += ALIGN(enc.width, align);
	}
	for (i = 0; i < enc.height / 2; ++i) {
		memcpy(enc.vbuf.ptr, ((frame + enc.height * enc.width) + i * enc.width), enc.width);
		enc.vbuf.ptr += ALIGN(enc.width, align);
	}

	r = amdgpu_bo_cpu_unmap(enc.vbuf.handle);
	CU_ASSERT_EQUAL(r, 0);

	len = 0;
	memcpy((ib_cpu + len), uve_session_info, sizeof(uve_session_info));
	len += sizeof(uve_session_info) / 4;
	ib_cpu[len++] = enc.session.addr >> 32;
	ib_cpu[len++] = enc.session.addr;

	memcpy((ib_cpu + len), uve_task_info, sizeof(uve_task_info));
	len += sizeof(uve_task_info) / 4;
	ib_cpu[len++] = 0x000005e0;
	ib_cpu[len++] = 0x00000001;
	ib_cpu[len++] = 0x00000001;

	memcpy((ib_cpu + len), uve_nalu_buffer_1, sizeof(uve_nalu_buffer_1));
	len += sizeof(uve_nalu_buffer_1) / 4;

	memcpy((ib_cpu + len), uve_nalu_buffer_2, sizeof(uve_nalu_buffer_2));
	len += sizeof(uve_nalu_buffer_2) / 4;

	memcpy((ib_cpu + len), uve_nalu_buffer_3, sizeof(uve_nalu_buffer_3));
	len += sizeof(uve_nalu_buffer_3) / 4;

	memcpy((ib_cpu + len), uve_nalu_buffer_4, sizeof(uve_nalu_buffer_4));
	len += sizeof(uve_nalu_buffer_4) / 4;

	memcpy((ib_cpu + len), uve_slice_header, sizeof(uve_slice_header));
	len += sizeof(uve_slice_header) / 4;

	ib_cpu[len++] = 0x00000254;
	ib_cpu[len++] = 0x00000010;
	ib_cpu[len++] = enc.cpb.addr >> 32;
	ib_cpu[len++] = enc.cpb.addr;
	memcpy((ib_cpu + len), uve_ctx_buffer, sizeof(uve_ctx_buffer));
	len += sizeof(uve_ctx_buffer) / 4;

	memcpy((ib_cpu + len), uve_bitstream_buffer, sizeof(uve_bitstream_buffer));
	len += sizeof(uve_bitstream_buffer) / 4;
	ib_cpu[len++] = 0x00000000;
	ib_cpu[len++] = enc.bs.addr >> 32;
	ib_cpu[len++] = enc.bs.addr;
	ib_cpu[len++] = 0x003f4800;
	ib_cpu[len++] = 0x00000000;

	memcpy((ib_cpu + len), uve_feedback_buffer, sizeof(uve_feedback_buffer));
	len += sizeof(uve_feedback_buffer) / 4;
	ib_cpu[len++] = enc.fb.addr >> 32;
	ib_cpu[len++] = enc.fb.addr;
	ib_cpu[len++] = 0x00000010;
	ib_cpu[len++] = 0x00000028;

	memcpy((ib_cpu + len), uve_feedback_buffer_additional, sizeof(uve_feedback_buffer_additional));
	len += sizeof(uve_feedback_buffer_additional) / 4;

	memcpy((ib_cpu + len), uve_intra_refresh, sizeof(uve_intra_refresh));
	len += sizeof(uve_intra_refresh) / 4;

	memcpy((ib_cpu + len), uve_layer_select, sizeof(uve_layer_select));
	len += sizeof(uve_layer_select) / 4;

	memcpy((ib_cpu + len), uve_rc_layer_init, sizeof(uve_rc_layer_init));
	len += sizeof(uve_rc_layer_init) / 4;

	memcpy((ib_cpu + len), uve_layer_select, sizeof(uve_layer_select));
	len += sizeof(uve_layer_select) / 4;

	memcpy((ib_cpu + len), uve_rc_per_pic, sizeof(uve_rc_per_pic));
	len += sizeof(uve_rc_per_pic) / 4;

	unsigned luma_size = ALIGN(enc.width, align) * ALIGN(enc.height, 16);
	luma_offset = enc.vbuf.addr;
	chroma_offset = luma_offset + luma_size;
	ib_cpu[len++] = 0x00000054;
	ib_cpu[len++] = 0x0000000c;
	ib_cpu[len++] = 0x00000002;
	ib_cpu[len++] = 0x003f4800;
	ib_cpu[len++] = luma_offset >> 32;
	ib_cpu[len++] = luma_offset;
	ib_cpu[len++] = chroma_offset >> 32;
	ib_cpu[len++] = chroma_offset;
	memcpy((ib_cpu + len), uve_encode_param, sizeof(uve_encode_param));
	ib_cpu[len] = ALIGN(enc.width, align);
	ib_cpu[len + 1] = ALIGN(enc.width, align);
	len += sizeof(uve_encode_param) / 4;

	memcpy((ib_cpu + len), uve_op_speed_enc_mode, sizeof(uve_op_speed_enc_mode));
	len += sizeof(uve_op_speed_enc_mode) / 4;

	memcpy((ib_cpu + len), uve_op_encode, sizeof(uve_op_encode));
	len += sizeof(uve_op_encode) / 4;

	r = submit(len, AMDGPU_HW_IP_UVD_ENC);
	CU_ASSERT_EQUAL(r, 0);

	check_result(&enc);

	free_resource(&enc.fb);
	free_resource(&enc.bs);
	free_resource(&enc.vbuf);
	free_resource(&enc.cpb);
}

static void amdgpu_cs_uvd_enc_destroy(void)
{
	struct amdgpu_uvd_enc_bo sw_ctx;
	int len, r;

	num_resources  = 0;
	resources[num_resources++] = ib_handle;

	len = 0;
	memcpy((ib_cpu + len), uve_session_info, sizeof(uve_session_info));
	len += sizeof(uve_session_info) / 4;
	ib_cpu[len++] = enc.session.addr >> 32;
	ib_cpu[len++] = enc.session.addr;

	memcpy((ib_cpu + len), uve_task_info, sizeof(uve_task_info));
	len += sizeof(uve_task_info) / 4;
	ib_cpu[len++] = 0xffffffff;
	ib_cpu[len++] = 0x00000002;
	ib_cpu[len++] = 0x00000000;

	memcpy((ib_cpu + len), uve_op_close, sizeof(uve_op_close));
	len += sizeof(uve_op_close) / 4;

	r = submit(len, AMDGPU_HW_IP_UVD_ENC);
	CU_ASSERT_EQUAL(r, 0);

	free_resource(&enc.session);
}