summaryrefslogtreecommitdiff
path: root/drivers/dma-buf/dma-fence-array.c
blob: cb1bacb5a42b543599b35a670f023e7b777c7e41 (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
// SPDX-License-Identifier: GPL-2.0-only
/*
 * dma-fence-array: aggregate fences to be waited together
 *
 * Copyright (C) 2016 Collabora Ltd
 * Copyright (C) 2016 Advanced Micro Devices, Inc.
 * Authors:
 *	Gustavo Padovan <gustavo@padovan.org>
 *	Christian König <christian.koenig@amd.com>
 */

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

#define PENDING_ERROR 1

static const char *dma_fence_array_get_driver_name(struct dma_fence *fence)
{
	return "dma_fence_array";
}

static const char *dma_fence_array_get_timeline_name(struct dma_fence *fence)
{
	return "unbound";
}

static void dma_fence_array_set_pending_error(struct dma_fence_array *array,
					      int error)
{
	/*
	 * Propagate the first error reported by any of our fences, but only
	 * before we ourselves are signaled.
	 */
	if (error)
		cmpxchg(&array->base.error, PENDING_ERROR, error);
}

static void dma_fence_array_clear_pending_error(struct dma_fence_array *array)
{
	/* Clear the error flag if not actually set. */
	cmpxchg(&array->base.error, PENDING_ERROR, 0);
}

static void irq_dma_fence_array_work(struct irq_work *wrk)
{
	struct dma_fence_array *array = container_of(wrk, typeof(*array), work);

	dma_fence_array_clear_pending_error(array);

	dma_fence_signal(&array->base);
	dma_fence_put(&array->base);
}

static void dma_fence_array_cb_func(struct dma_fence *f,
				    struct dma_fence_cb *cb)
{
	struct dma_fence_array_cb *array_cb =
		container_of(cb, struct dma_fence_array_cb, cb);
	struct dma_fence_array *array = array_cb->array;

	dma_fence_array_set_pending_error(array, f->error);

	if (atomic_dec_and_test(&array->num_pending))
		irq_work_queue(&array->work);
	else
		dma_fence_put(&array->base);
}

static bool dma_fence_array_enable_signaling(struct dma_fence *fence)
{
	struct dma_fence_array *array = to_dma_fence_array(fence);
	struct dma_fence_array_cb *cb = (void *)(&array[1]);
	unsigned i;

	for (i = 0; i < array->num_fences; ++i) {
		cb[i].array = array;
		/*
		 * As we may report that the fence is signaled before all
		 * callbacks are complete, we need to take an additional
		 * reference count on the array so that we do not free it too
		 * early. The core fence handling will only hold the reference
		 * until we signal the array as complete (but that is now
		 * insufficient).
		 */
		dma_fence_get(&array->base);
		if (dma_fence_add_callback(array->fences[i], &cb[i].cb,
					   dma_fence_array_cb_func)) {
			int error = array->fences[i]->error;

			dma_fence_array_set_pending_error(array, error);
			dma_fence_put(&array->base);
			if (atomic_dec_and_test(&array->num_pending)) {
				dma_fence_array_clear_pending_error(array);
				return false;
			}
		}
	}

	return true;
}

static bool dma_fence_array_signaled(struct dma_fence *fence)
{
	struct dma_fence_array *array = to_dma_fence_array(fence);

	if (atomic_read(&array->num_pending) > 0)
		return false;

	dma_fence_array_clear_pending_error(array);
	return true;
}

static void dma_fence_array_release(struct dma_fence *fence)
{
	struct dma_fence_array *array = to_dma_fence_array(fence);
	unsigned i;

	for (i = 0; i < array->num_fences; ++i)
		dma_fence_put(array->fences[i]);

	kfree(array->fences);
	dma_fence_free(fence);
}

const struct dma_fence_ops dma_fence_array_ops = {
	.get_driver_name = dma_fence_array_get_driver_name,
	.get_timeline_name = dma_fence_array_get_timeline_name,
	.enable_signaling = dma_fence_array_enable_signaling,
	.signaled = dma_fence_array_signaled,
	.release = dma_fence_array_release,
};
EXPORT_SYMBOL(dma_fence_array_ops);

/**
 * dma_fence_array_create - Create a custom fence array
 * @num_fences:		[in]	number of fences to add in the array
 * @fences:		[in]	array containing the fences
 * @context:		[in]	fence context to use
 * @seqno:		[in]	sequence number to use
 * @signal_on_any:	[in]	signal on any fence in the array
 *
 * Allocate a dma_fence_array object and initialize the base fence with
 * dma_fence_init().
 * In case of error it returns NULL.
 *
 * The caller should allocate the fences array with num_fences size
 * and fill it with the fences it wants to add to the object. Ownership of this
 * array is taken and dma_fence_put() is used on each fence on release.
 *
 * If @signal_on_any is true the fence array signals if any fence in the array
 * signals, otherwise it signals when all fences in the array signal.
 */
struct dma_fence_array *dma_fence_array_create(int num_fences,
					       struct dma_fence **fences,
					       u64 context, unsigned seqno,
					       bool signal_on_any)
{
	struct dma_fence_array *array;
	size_t size = sizeof(*array);

	/* Allocate the callback structures behind the array. */
	size += num_fences * sizeof(struct dma_fence_array_cb);
	array = kzalloc(size, GFP_KERNEL);
	if (!array)
		return NULL;

	spin_lock_init(&array->lock);
	dma_fence_init(&array->base, &dma_fence_array_ops, &array->lock,
		       context, seqno);
	init_irq_work(&array->work, irq_dma_fence_array_work);

	array->num_fences = num_fences;
	atomic_set(&array->num_pending, signal_on_any ? 1 : num_fences);
	array->fences = fences;

	array->base.error = PENDING_ERROR;

	/*
	 * dma_fence_array objects should never contain any other fence
	 * containers or otherwise we run into recursion and potential kernel
	 * stack overflow on operations on the dma_fence_array.
	 *
	 * The correct way of handling this is to flatten out the array by the
	 * caller instead.
	 *
	 * Enforce this here by checking that we don't create a dma_fence_array
	 * with any container inside.
	 */
	while (num_fences--)
		WARN_ON(dma_fence_is_container(fences[num_fences]));

	return array;
}
EXPORT_SYMBOL(dma_fence_array_create);

/**
 * dma_fence_match_context - Check if all fences are from the given context
 * @fence:		[in]	fence or fence array
 * @context:		[in]	fence context to check all fences against
 *
 * Checks the provided fence or, for a fence array, all fences in the array
 * against the given context. Returns false if any fence is from a different
 * context.
 */
bool dma_fence_match_context(struct dma_fence *fence, u64 context)
{
	struct dma_fence_array *array = to_dma_fence_array(fence);
	unsigned i;

	if (!dma_fence_is_array(fence))
		return fence->context == context;

	for (i = 0; i < array->num_fences; i++) {
		if (array->fences[i]->context != context)
			return false;
	}

	return true;
}
EXPORT_SYMBOL(dma_fence_match_context);