summaryrefslogtreecommitdiff
path: root/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c
blob: 67f66f2f1809cb8597697163c050f382832c557f (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
/*
 * Copyright 2015 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 <linux/kthread.h>
#include <linux/wait.h>
#include <linux/sched.h>

#include <drm/drm_drv.h>

#include "amdgpu.h"
#include "amdgpu_trace.h"

static enum drm_gpu_sched_stat amdgpu_job_timedout(struct drm_sched_job *s_job)
{
	struct amdgpu_ring *ring = to_amdgpu_ring(s_job->sched);
	struct amdgpu_job *job = to_amdgpu_job(s_job);
	struct amdgpu_task_info ti;
	struct amdgpu_device *adev = ring->adev;
	int idx;
	int r;

	if (!drm_dev_enter(adev_to_drm(adev), &idx)) {
		DRM_INFO("%s - device unplugged skipping recovery on scheduler:%s",
			 __func__, s_job->sched->name);

		/* Effectively the job is aborted as the device is gone */
		return DRM_GPU_SCHED_STAT_ENODEV;
	}

	memset(&ti, 0, sizeof(struct amdgpu_task_info));

	if (amdgpu_gpu_recovery &&
	    amdgpu_ring_soft_recovery(ring, job->vmid, s_job->s_fence->parent)) {
		DRM_ERROR("ring %s timeout, but soft recovered\n",
			  s_job->sched->name);
		goto exit;
	}

	amdgpu_vm_get_task_info(ring->adev, job->pasid, &ti);
	DRM_ERROR("ring %s timeout, signaled seq=%u, emitted seq=%u\n",
		  job->base.sched->name, atomic_read(&ring->fence_drv.last_seq),
		  ring->fence_drv.sync_seq);
	DRM_ERROR("Process information: process %s pid %d thread %s pid %d\n",
		  ti.process_name, ti.tgid, ti.task_name, ti.pid);

	if (amdgpu_device_should_recover_gpu(ring->adev)) {
		r = amdgpu_device_gpu_recover_imp(ring->adev, job);
		if (r)
			DRM_ERROR("GPU Recovery Failed: %d\n", r);
	} else {
		drm_sched_suspend_timeout(&ring->sched);
		if (amdgpu_sriov_vf(adev))
			adev->virt.tdr_debug = true;
	}

exit:
	drm_dev_exit(idx);
	return DRM_GPU_SCHED_STAT_NOMINAL;
}

int amdgpu_job_alloc(struct amdgpu_device *adev, unsigned num_ibs,
		     struct amdgpu_job **job, struct amdgpu_vm *vm)
{
	if (num_ibs == 0)
		return -EINVAL;

	*job = kzalloc(struct_size(*job, ibs, num_ibs), GFP_KERNEL);
	if (!*job)
		return -ENOMEM;

	/*
	 * Initialize the scheduler to at least some ring so that we always
	 * have a pointer to adev.
	 */
	(*job)->base.sched = &adev->rings[0]->sched;
	(*job)->vm = vm;
	(*job)->num_ibs = num_ibs;

	amdgpu_sync_create(&(*job)->sync);
	amdgpu_sync_create(&(*job)->sched_sync);
	(*job)->vram_lost_counter = atomic_read(&adev->vram_lost_counter);
	(*job)->vm_pd_addr = AMDGPU_BO_INVALID_OFFSET;

	return 0;
}

int amdgpu_job_alloc_with_ib(struct amdgpu_device *adev, unsigned size,
		enum amdgpu_ib_pool_type pool_type,
		struct amdgpu_job **job)
{
	int r;

	r = amdgpu_job_alloc(adev, 1, job, NULL);
	if (r)
		return r;

	r = amdgpu_ib_get(adev, NULL, size, pool_type, &(*job)->ibs[0]);
	if (r)
		kfree(*job);

	return r;
}

void amdgpu_job_free_resources(struct amdgpu_job *job)
{
	struct amdgpu_ring *ring = to_amdgpu_ring(job->base.sched);
	struct dma_fence *f;
	struct dma_fence *hw_fence;
	unsigned i;

	if (job->hw_fence.ops == NULL)
		hw_fence = job->external_hw_fence;
	else
		hw_fence = &job->hw_fence;

	/* use sched fence if available */
	f = job->base.s_fence ? &job->base.s_fence->finished : hw_fence;
	for (i = 0; i < job->num_ibs; ++i)
		amdgpu_ib_free(ring->adev, &job->ibs[i], f);
}

static void amdgpu_job_free_cb(struct drm_sched_job *s_job)
{
	struct amdgpu_job *job = to_amdgpu_job(s_job);

	drm_sched_job_cleanup(s_job);

	amdgpu_sync_free(&job->sync);
	amdgpu_sync_free(&job->sched_sync);

    /* only put the hw fence if has embedded fence */
	if (job->hw_fence.ops != NULL)
		dma_fence_put(&job->hw_fence);
	else
		kfree(job);
}

void amdgpu_job_free(struct amdgpu_job *job)
{
	amdgpu_job_free_resources(job);
	amdgpu_sync_free(&job->sync);
	amdgpu_sync_free(&job->sched_sync);

	/* only put the hw fence if has embedded fence */
	if (job->hw_fence.ops != NULL)
		dma_fence_put(&job->hw_fence);
	else
		kfree(job);
}

int amdgpu_job_submit(struct amdgpu_job *job, struct drm_sched_entity *entity,
		      void *owner, struct dma_fence **f)
{
	int r;

	if (!f)
		return -EINVAL;

	r = drm_sched_job_init(&job->base, entity, owner);
	if (r)
		return r;

	drm_sched_job_arm(&job->base);

	*f = dma_fence_get(&job->base.s_fence->finished);
	amdgpu_job_free_resources(job);
	drm_sched_entity_push_job(&job->base);

	return 0;
}

int amdgpu_job_submit_direct(struct amdgpu_job *job, struct amdgpu_ring *ring,
			     struct dma_fence **fence)
{
	int r;

	job->base.sched = &ring->sched;
	r = amdgpu_ib_schedule(ring, job->num_ibs, job->ibs, NULL, fence);
	/* record external_hw_fence for direct submit */
	job->external_hw_fence = dma_fence_get(*fence);
	if (r)
		return r;

	amdgpu_job_free(job);
	dma_fence_put(*fence);

	return 0;
}

static struct dma_fence *amdgpu_job_dependency(struct drm_sched_job *sched_job,
					       struct drm_sched_entity *s_entity)
{
	struct amdgpu_ring *ring = to_amdgpu_ring(s_entity->rq->sched);
	struct amdgpu_job *job = to_amdgpu_job(sched_job);
	struct amdgpu_vm *vm = job->vm;
	struct dma_fence *fence;
	int r;

	fence = amdgpu_sync_get_fence(&job->sync);
	if (fence && drm_sched_dependency_optimized(fence, s_entity)) {
		r = amdgpu_sync_fence(&job->sched_sync, fence);
		if (r)
			DRM_ERROR("Error adding fence (%d)\n", r);
	}

	while (fence == NULL && vm && !job->vmid) {
		r = amdgpu_vmid_grab(vm, ring, &job->sync,
				     &job->base.s_fence->finished,
				     job);
		if (r)
			DRM_ERROR("Error getting VM ID (%d)\n", r);

		fence = amdgpu_sync_get_fence(&job->sync);
	}

	return fence;
}

static struct dma_fence *amdgpu_job_run(struct drm_sched_job *sched_job)
{
	struct amdgpu_ring *ring = to_amdgpu_ring(sched_job->sched);
	struct dma_fence *fence = NULL, *finished;
	struct amdgpu_job *job;
	int r = 0;

	job = to_amdgpu_job(sched_job);
	finished = &job->base.s_fence->finished;

	BUG_ON(amdgpu_sync_peek_fence(&job->sync, NULL));

	trace_amdgpu_sched_run_job(job);

	if (job->vram_lost_counter != atomic_read(&ring->adev->vram_lost_counter))
		dma_fence_set_error(finished, -ECANCELED);/* skip IB as well if VRAM lost */

	if (finished->error < 0) {
		DRM_INFO("Skip scheduling IBs!\n");
	} else {
		r = amdgpu_ib_schedule(ring, job->num_ibs, job->ibs, job,
				       &fence);
		if (r)
			DRM_ERROR("Error scheduling IBs (%d)\n", r);
	}

	if (!job->job_run_counter)
		dma_fence_get(fence);
	else if (finished->error < 0)
		dma_fence_put(&job->hw_fence);
	job->job_run_counter++;
	amdgpu_job_free_resources(job);

	fence = r ? ERR_PTR(r) : fence;
	return fence;
}

#define to_drm_sched_job(sched_job)		\
		container_of((sched_job), struct drm_sched_job, queue_node)

void amdgpu_job_stop_all_jobs_on_sched(struct drm_gpu_scheduler *sched)
{
	struct drm_sched_job *s_job;
	struct drm_sched_entity *s_entity = NULL;
	int i;

	/* Signal all jobs not yet scheduled */
	for (i = DRM_SCHED_PRIORITY_COUNT - 1; i >= DRM_SCHED_PRIORITY_MIN; i--) {
		struct drm_sched_rq *rq = &sched->sched_rq[i];

		if (!rq)
			continue;

		spin_lock(&rq->lock);
		list_for_each_entry(s_entity, &rq->entities, list) {
			while ((s_job = to_drm_sched_job(spsc_queue_pop(&s_entity->job_queue)))) {
				struct drm_sched_fence *s_fence = s_job->s_fence;

				dma_fence_signal(&s_fence->scheduled);
				dma_fence_set_error(&s_fence->finished, -EHWPOISON);
				dma_fence_signal(&s_fence->finished);
			}
		}
		spin_unlock(&rq->lock);
	}

	/* Signal all jobs already scheduled to HW */
	list_for_each_entry(s_job, &sched->pending_list, list) {
		struct drm_sched_fence *s_fence = s_job->s_fence;

		dma_fence_set_error(&s_fence->finished, -EHWPOISON);
		dma_fence_signal(&s_fence->finished);
	}
}

const struct drm_sched_backend_ops amdgpu_sched_ops = {
	.dependency = amdgpu_job_dependency,
	.run_job = amdgpu_job_run,
	.timedout_job = amdgpu_job_timedout,
	.free_job = amdgpu_job_free_cb
};