summaryrefslogtreecommitdiffstats
path: root/tests/amdgpu/hotunplug_tests.c
blob: 2b265673e73a91def57bf1190c662668f9cad515 (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
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
/*
 * Copyright 2021 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.
 *
*/

#include <stdlib.h>
#include <unistd.h>
#include <sys/types.h>
#include <sys/stat.h>
#include <fcntl.h>
#if HAVE_ALLOCA_H
# include <alloca.h>
#endif

#include "CUnit/Basic.h"

#include "amdgpu_test.h"
#include "amdgpu_drm.h"
#include "amdgpu_internal.h"
#include "xf86drm.h"
#include <pthread.h>

#define GFX_COMPUTE_NOP  0xffff1000

static  amdgpu_device_handle device_handle;
static  uint32_t  major_version;
static  uint32_t  minor_version;
static char *sysfs_remove = NULL;
static bool do_cs;

CU_BOOL suite_hotunplug_tests_enable(void)
{
	CU_BOOL enable = CU_TRUE;
	drmDevicePtr device;

	if (drmGetDevice2(drm_amdgpu[0], DRM_DEVICE_GET_PCI_REVISION, &device)) {
		printf("\n\nGPU Failed to get DRM device PCI info!\n");
		return CU_FALSE;
	}

	if (device->bustype != DRM_BUS_PCI) {
		printf("\n\nGPU device is not on PCI bus!\n");
		amdgpu_device_deinitialize(device_handle);
		return CU_FALSE;
	}

	if (amdgpu_device_initialize(drm_amdgpu[0], &major_version,
					     &minor_version, &device_handle))
		return CU_FALSE;
	
	/* Latest tested amdgpu version to work with all the tests */
        if (minor_version < 46)
                enable = false;

	if (amdgpu_device_deinitialize(device_handle))
		return CU_FALSE;

	return enable;
}

int suite_hotunplug_tests_init(void)
{
	/* We need to open/close device at each test manually */
	amdgpu_close_devices();

	return CUE_SUCCESS;
}

int suite_hotunplug_tests_clean(void)
{


	return CUE_SUCCESS;
}

static int amdgpu_hotunplug_trigger(const char *pathname)
{
	int fd, len;

	fd = open(pathname, O_WRONLY);
	if (fd < 0)
		return -errno;

	len = write(fd, "1", 1);
	close(fd);

	return len;
}

static int amdgpu_hotunplug_setup_test()
{
	int r;
	char *tmp_str;

	if (amdgpu_open_device_on_test_index(open_render_node) < 0) {
		printf("\n\n Failed to reopen device file!\n");
		return CUE_SINIT_FAILED;



	}

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

	if (r) {
		if ((r == -EACCES) && (errno == EACCES))
			printf("\n\nError:%s. "
				"Hint:Try to run this test program as root.",
				strerror(errno));
		return CUE_SINIT_FAILED;
	}

	tmp_str = amdgpu_get_device_from_fd(drm_amdgpu[0]);
	if (!tmp_str){
		printf("\n\n Device path not found!\n");
		return  CUE_SINIT_FAILED;
	}

	sysfs_remove = realloc(tmp_str, strlen(tmp_str) * 2);
	strcat(sysfs_remove, "/remove");

	return 0;
}

static int amdgpu_hotunplug_teardown_test()
{
	if (amdgpu_device_deinitialize(device_handle))
		return CUE_SCLEAN_FAILED;

	amdgpu_close_devices();

	if (sysfs_remove)
		free(sysfs_remove);

	return 0;
}

static inline int amdgpu_hotunplug_remove()
{
	return amdgpu_hotunplug_trigger(sysfs_remove);
}

static inline int amdgpu_hotunplug_rescan()
{
	return amdgpu_hotunplug_trigger("/sys/bus/pci/rescan");
}

static int amdgpu_cs_sync(amdgpu_context_handle context,
			   unsigned int ip_type,
			   int ring,
			   unsigned int seqno)
{
	struct amdgpu_cs_fence fence = {
		.context = context,
		.ip_type = ip_type,
		.ring = ring,
		.fence = seqno,
	};
	uint32_t expired;

	return  amdgpu_cs_query_fence_status(&fence,
					   AMDGPU_TIMEOUT_INFINITE,
					   0, &expired);
}

static void *amdgpu_nop_cs()
{
	amdgpu_bo_handle ib_result_handle;
	void *ib_result_cpu;
	uint64_t ib_result_mc_address;
	uint32_t *ptr;
	int i, r;
	amdgpu_bo_list_handle bo_list;
	amdgpu_va_handle va_handle;
	amdgpu_context_handle context;
	struct amdgpu_cs_request ibs_request;
	struct amdgpu_cs_ib_info ib_info;

	r = amdgpu_cs_ctx_create(device_handle, &context);
	CU_ASSERT_EQUAL(r, 0);

	r = amdgpu_bo_alloc_and_map(device_handle, 4096, 4096,
				    AMDGPU_GEM_DOMAIN_GTT, 0,
				    &ib_result_handle, &ib_result_cpu,
				    &ib_result_mc_address, &va_handle);
	CU_ASSERT_EQUAL(r, 0);

	ptr = ib_result_cpu;
	for (i = 0; i < 16; ++i)
		ptr[i] = GFX_COMPUTE_NOP;

	r = amdgpu_bo_list_create(device_handle, 1, &ib_result_handle, NULL, &bo_list);
	CU_ASSERT_EQUAL(r, 0);

	memset(&ib_info, 0, sizeof(struct amdgpu_cs_ib_info));
	ib_info.ib_mc_address = ib_result_mc_address;
	ib_info.size = 16;

	memset(&ibs_request, 0, sizeof(struct amdgpu_cs_request));
	ibs_request.ip_type = AMDGPU_HW_IP_GFX;
	ibs_request.ring = 0;
	ibs_request.number_of_ibs = 1;
	ibs_request.ibs = &ib_info;
	ibs_request.resources = bo_list;

	while (do_cs)
		amdgpu_cs_submit(context, 0, &ibs_request, 1);

	amdgpu_cs_sync(context, AMDGPU_HW_IP_GFX, 0, ibs_request.seq_no);
	amdgpu_bo_list_destroy(bo_list);
	amdgpu_bo_unmap_and_free(ib_result_handle, va_handle,
				 ib_result_mc_address, 4096);

	amdgpu_cs_ctx_free(context);

	return (void *)0;
}

static pthread_t* amdgpu_create_cs_thread()
{
	int r;
	pthread_t *thread = malloc(sizeof(*thread));
	if (!thread)
		return NULL;

	do_cs = true;

	r = pthread_create(thread, NULL, amdgpu_nop_cs, NULL);
	CU_ASSERT_EQUAL(r, 0);

	/* Give thread enough time to start*/
	usleep(100000);
	return thread;
}

static void amdgpu_destroy_cs_thread(pthread_t *thread)
{
	void *status;

	do_cs = false;

	pthread_join(*thread, &status);
	CU_ASSERT_EQUAL(status, 0);

	free(thread);
}


static void amdgpu_hotunplug_test(bool with_cs)
{
	int r;
	pthread_t *thread = NULL;

	r = amdgpu_hotunplug_setup_test();
	CU_ASSERT_EQUAL(r , 0);

	if (with_cs) {
		thread = amdgpu_create_cs_thread();
		CU_ASSERT_NOT_EQUAL(thread, NULL);
	}

	r = amdgpu_hotunplug_remove();
	CU_ASSERT_EQUAL(r > 0, 1);

	if (with_cs)
		amdgpu_destroy_cs_thread(thread);

	r = amdgpu_hotunplug_teardown_test();
	CU_ASSERT_EQUAL(r , 0);

	r = amdgpu_hotunplug_rescan();
	CU_ASSERT_EQUAL(r > 0, 1);
}

static void amdgpu_hotunplug_simple(void)
{
	amdgpu_hotunplug_test(false);
}

static void amdgpu_hotunplug_with_cs(void)
{
	amdgpu_hotunplug_test(true);
}

static void amdgpu_hotunplug_with_exported_bo(void)
{
	int r;
	uint32_t dma_buf_fd;
	unsigned int *ptr;
	amdgpu_bo_handle bo_handle;

	struct amdgpu_bo_alloc_request request = {
		.alloc_size = 4096,
		.phys_alignment = 4096,
		.preferred_heap = AMDGPU_GEM_DOMAIN_GTT,
		.flags = 0,
	};

	r = amdgpu_hotunplug_setup_test();
	CU_ASSERT_EQUAL(r , 0);

	amdgpu_bo_alloc(device_handle, &request, &bo_handle);
	CU_ASSERT_EQUAL(r, 0);

	r = amdgpu_bo_export(bo_handle, amdgpu_bo_handle_type_dma_buf_fd, &dma_buf_fd);
	CU_ASSERT_EQUAL(r, 0);

	ptr = mmap(NULL, 4096, PROT_READ | PROT_WRITE, MAP_SHARED, dma_buf_fd, 0);
	CU_ASSERT_NOT_EQUAL(ptr,  MAP_FAILED);

	r = amdgpu_hotunplug_remove();
	CU_ASSERT_EQUAL(r > 0, 1);

	amdgpu_bo_free(bo_handle);

	r = amdgpu_hotunplug_teardown_test();
	CU_ASSERT_EQUAL(r , 0);

	*ptr = 0xdeafbeef;

	munmap(ptr, 4096);
	close (dma_buf_fd);

	r = amdgpu_hotunplug_rescan();
	CU_ASSERT_EQUAL(r > 0, 1);
}

static void amdgpu_hotunplug_with_exported_fence(void)
{
	amdgpu_bo_handle ib_result_handle;
	void *ib_result_cpu;
	uint64_t ib_result_mc_address;
	uint32_t *ptr, sync_obj_handle, sync_obj_handle2;
	int i, r;
	amdgpu_bo_list_handle bo_list;
	amdgpu_va_handle va_handle;
	uint32_t major2, minor2;
	amdgpu_device_handle device2;
	amdgpu_context_handle context;
	struct amdgpu_cs_request ibs_request;
	struct amdgpu_cs_ib_info ib_info;
	struct amdgpu_cs_fence fence_status = {0};
	int shared_fd;

	r = amdgpu_hotunplug_setup_test();
	CU_ASSERT_EQUAL(r , 0);

	r = amdgpu_device_initialize(drm_amdgpu[1], &major2, &minor2, &device2);
	CU_ASSERT_EQUAL(r, 0);

	r = amdgpu_cs_ctx_create(device_handle, &context);
	CU_ASSERT_EQUAL(r, 0);

	r = amdgpu_bo_alloc_and_map(device_handle, 4096, 4096,
				    AMDGPU_GEM_DOMAIN_GTT, 0,
				    &ib_result_handle, &ib_result_cpu,
				    &ib_result_mc_address, &va_handle);
	CU_ASSERT_EQUAL(r, 0);

	ptr = ib_result_cpu;
	for (i = 0; i < 16; ++i)
		ptr[i] = GFX_COMPUTE_NOP;

	r = amdgpu_bo_list_create(device_handle, 1, &ib_result_handle, NULL, &bo_list);
	CU_ASSERT_EQUAL(r, 0);

	memset(&ib_info, 0, sizeof(struct amdgpu_cs_ib_info));
	ib_info.ib_mc_address = ib_result_mc_address;
	ib_info.size = 16;

	memset(&ibs_request, 0, sizeof(struct amdgpu_cs_request));
	ibs_request.ip_type = AMDGPU_HW_IP_GFX;
	ibs_request.ring = 0;
	ibs_request.number_of_ibs = 1;
	ibs_request.ibs = &ib_info;
	ibs_request.resources = bo_list;

	CU_ASSERT_EQUAL(amdgpu_cs_submit(context, 0, &ibs_request, 1), 0);

	fence_status.context = context;
	fence_status.ip_type = AMDGPU_HW_IP_GFX;
	fence_status.ip_instance = 0;
	fence_status.fence = ibs_request.seq_no;

	CU_ASSERT_EQUAL(amdgpu_cs_fence_to_handle(device_handle, &fence_status,
						AMDGPU_FENCE_TO_HANDLE_GET_SYNCOBJ,
						&sync_obj_handle),
						0);

	CU_ASSERT_EQUAL(amdgpu_cs_export_syncobj(device_handle, sync_obj_handle, &shared_fd), 0);

	CU_ASSERT_EQUAL(amdgpu_cs_import_syncobj(device2, shared_fd, &sync_obj_handle2), 0);

	CU_ASSERT_EQUAL(amdgpu_cs_destroy_syncobj(device_handle, sync_obj_handle), 0);

	CU_ASSERT_EQUAL(amdgpu_bo_list_destroy(bo_list), 0);
	CU_ASSERT_EQUAL(amdgpu_bo_unmap_and_free(ib_result_handle, va_handle,
				 ib_result_mc_address, 4096), 0);
	CU_ASSERT_EQUAL(amdgpu_cs_ctx_free(context), 0);

	r = amdgpu_hotunplug_remove();
	CU_ASSERT_EQUAL(r > 0, 1);

	CU_ASSERT_EQUAL(amdgpu_cs_syncobj_wait(device2, &sync_obj_handle2, 1, 100000000, 0, NULL), 0);

	CU_ASSERT_EQUAL(amdgpu_cs_destroy_syncobj(device2, sync_obj_handle2), 0);

	amdgpu_device_deinitialize(device2);

	r = amdgpu_hotunplug_teardown_test();
	CU_ASSERT_EQUAL(r , 0);

	r = amdgpu_hotunplug_rescan();
	CU_ASSERT_EQUAL(r > 0, 1);
}


CU_TestInfo hotunplug_tests[] = {
	{ "Unplug card and rescan the bus to plug it back", amdgpu_hotunplug_simple },
	{ "Same as first test but with command submission", amdgpu_hotunplug_with_cs },
	{ "Unplug with exported bo", amdgpu_hotunplug_with_exported_bo },
	{ "Unplug with exported fence", amdgpu_hotunplug_with_exported_fence },
	CU_TEST_INFO_NULL,
};