summaryrefslogtreecommitdiff
path: root/drivers/gpu/drm/i915/gem/i915_gem_wait.c
blob: 26ec6579b7cdfa76c782664ba8bccf88c4ca58ff (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
/*
 * SPDX-License-Identifier: MIT
 *
 * Copyright © 2016 Intel Corporation
 */

#include <linux/dma-fence-array.h>
#include <linux/jiffies.h>

#include "gt/intel_engine.h"

#include "i915_gem_ioctls.h"
#include "i915_gem_object.h"

static long
i915_gem_object_wait_fence(struct dma_fence *fence,
			   unsigned int flags,
			   long timeout)
{
	BUILD_BUG_ON(I915_WAIT_INTERRUPTIBLE != 0x1);

	if (test_bit(DMA_FENCE_FLAG_SIGNALED_BIT, &fence->flags))
		return timeout;

	if (dma_fence_is_i915(fence))
		return i915_request_wait(to_request(fence), flags, timeout);

	return dma_fence_wait_timeout(fence,
				      flags & I915_WAIT_INTERRUPTIBLE,
				      timeout);
}

static long
i915_gem_object_wait_reservation(struct reservation_object *resv,
				 unsigned int flags,
				 long timeout)
{
	unsigned int seq = __read_seqcount_begin(&resv->seq);
	struct dma_fence *excl;
	bool prune_fences = false;

	if (flags & I915_WAIT_ALL) {
		struct dma_fence **shared;
		unsigned int count, i;
		int ret;

		ret = reservation_object_get_fences_rcu(resv,
							&excl, &count, &shared);
		if (ret)
			return ret;

		for (i = 0; i < count; i++) {
			timeout = i915_gem_object_wait_fence(shared[i],
							     flags, timeout);
			if (timeout < 0)
				break;

			dma_fence_put(shared[i]);
		}

		for (; i < count; i++)
			dma_fence_put(shared[i]);
		kfree(shared);

		/*
		 * If both shared fences and an exclusive fence exist,
		 * then by construction the shared fences must be later
		 * than the exclusive fence. If we successfully wait for
		 * all the shared fences, we know that the exclusive fence
		 * must all be signaled. If all the shared fences are
		 * signaled, we can prune the array and recover the
		 * floating references on the fences/requests.
		 */
		prune_fences = count && timeout >= 0;
	} else {
		excl = reservation_object_get_excl_rcu(resv);
	}

	if (excl && timeout >= 0)
		timeout = i915_gem_object_wait_fence(excl, flags, timeout);

	dma_fence_put(excl);

	/*
	 * Opportunistically prune the fences iff we know they have *all* been
	 * signaled and that the reservation object has not been changed (i.e.
	 * no new fences have been added).
	 */
	if (prune_fences && !__read_seqcount_retry(&resv->seq, seq)) {
		if (reservation_object_trylock(resv)) {
			if (!__read_seqcount_retry(&resv->seq, seq))
				reservation_object_add_excl_fence(resv, NULL);
			reservation_object_unlock(resv);
		}
	}

	return timeout;
}

static void __fence_set_priority(struct dma_fence *fence,
				 const struct i915_sched_attr *attr)
{
	struct i915_request *rq;
	struct intel_engine_cs *engine;

	if (dma_fence_is_signaled(fence) || !dma_fence_is_i915(fence))
		return;

	rq = to_request(fence);
	engine = rq->engine;

	local_bh_disable();
	rcu_read_lock(); /* RCU serialisation for set-wedged protection */
	if (engine->schedule)
		engine->schedule(rq, attr);
	rcu_read_unlock();
	local_bh_enable(); /* kick the tasklets if queues were reprioritised */
}

static void fence_set_priority(struct dma_fence *fence,
			       const struct i915_sched_attr *attr)
{
	/* Recurse once into a fence-array */
	if (dma_fence_is_array(fence)) {
		struct dma_fence_array *array = to_dma_fence_array(fence);
		int i;

		for (i = 0; i < array->num_fences; i++)
			__fence_set_priority(array->fences[i], attr);
	} else {
		__fence_set_priority(fence, attr);
	}
}

int
i915_gem_object_wait_priority(struct drm_i915_gem_object *obj,
			      unsigned int flags,
			      const struct i915_sched_attr *attr)
{
	struct dma_fence *excl;

	if (flags & I915_WAIT_ALL) {
		struct dma_fence **shared;
		unsigned int count, i;
		int ret;

		ret = reservation_object_get_fences_rcu(obj->base.resv,
							&excl, &count, &shared);
		if (ret)
			return ret;

		for (i = 0; i < count; i++) {
			fence_set_priority(shared[i], attr);
			dma_fence_put(shared[i]);
		}

		kfree(shared);
	} else {
		excl = reservation_object_get_excl_rcu(obj->base.resv);
	}

	if (excl) {
		fence_set_priority(excl, attr);
		dma_fence_put(excl);
	}
	return 0;
}

/**
 * Waits for rendering to the object to be completed
 * @obj: i915 gem object
 * @flags: how to wait (under a lock, for all rendering or just for writes etc)
 * @timeout: how long to wait
 */
int
i915_gem_object_wait(struct drm_i915_gem_object *obj,
		     unsigned int flags,
		     long timeout)
{
	might_sleep();
	GEM_BUG_ON(timeout < 0);

	timeout = i915_gem_object_wait_reservation(obj->base.resv,
						   flags, timeout);
	return timeout < 0 ? timeout : 0;
}

static inline unsigned long nsecs_to_jiffies_timeout(const u64 n)
{
	/* nsecs_to_jiffies64() does not guard against overflow */
	if (NSEC_PER_SEC % HZ &&
	    div_u64(n, NSEC_PER_SEC) >= MAX_JIFFY_OFFSET / HZ)
		return MAX_JIFFY_OFFSET;

	return min_t(u64, MAX_JIFFY_OFFSET, nsecs_to_jiffies64(n) + 1);
}

static unsigned long to_wait_timeout(s64 timeout_ns)
{
	if (timeout_ns < 0)
		return MAX_SCHEDULE_TIMEOUT;

	if (timeout_ns == 0)
		return 0;

	return nsecs_to_jiffies_timeout(timeout_ns);
}

/**
 * i915_gem_wait_ioctl - implements DRM_IOCTL_I915_GEM_WAIT
 * @dev: drm device pointer
 * @data: ioctl data blob
 * @file: drm file pointer
 *
 * Returns 0 if successful, else an error is returned with the remaining time in
 * the timeout parameter.
 *  -ETIME: object is still busy after timeout
 *  -ERESTARTSYS: signal interrupted the wait
 *  -ENONENT: object doesn't exist
 * Also possible, but rare:
 *  -EAGAIN: incomplete, restart syscall
 *  -ENOMEM: damn
 *  -ENODEV: Internal IRQ fail
 *  -E?: The add request failed
 *
 * The wait ioctl with a timeout of 0 reimplements the busy ioctl. With any
 * non-zero timeout parameter the wait ioctl will wait for the given number of
 * nanoseconds on an object becoming unbusy. Since the wait itself does so
 * without holding struct_mutex the object may become re-busied before this
 * function completes. A similar but shorter * race condition exists in the busy
 * ioctl
 */
int
i915_gem_wait_ioctl(struct drm_device *dev, void *data, struct drm_file *file)
{
	struct drm_i915_gem_wait *args = data;
	struct drm_i915_gem_object *obj;
	ktime_t start;
	long ret;

	if (args->flags != 0)
		return -EINVAL;

	obj = i915_gem_object_lookup(file, args->bo_handle);
	if (!obj)
		return -ENOENT;

	start = ktime_get();

	ret = i915_gem_object_wait(obj,
				   I915_WAIT_INTERRUPTIBLE |
				   I915_WAIT_PRIORITY |
				   I915_WAIT_ALL,
				   to_wait_timeout(args->timeout_ns));

	if (args->timeout_ns > 0) {
		args->timeout_ns -= ktime_to_ns(ktime_sub(ktime_get(), start));
		if (args->timeout_ns < 0)
			args->timeout_ns = 0;

		/*
		 * Apparently ktime isn't accurate enough and occasionally has a
		 * bit of mismatch in the jiffies<->nsecs<->ktime loop. So patch
		 * things up to make the test happy. We allow up to 1 jiffy.
		 *
		 * This is a regression from the timespec->ktime conversion.
		 */
		if (ret == -ETIME && !nsecs_to_jiffies(args->timeout_ns))
			args->timeout_ns = 0;

		/* Asked to wait beyond the jiffie/scheduler precision? */
		if (ret == -ETIME && args->timeout_ns)
			ret = -EAGAIN;
	}

	i915_gem_object_put(obj);
	return ret;
}