summaryrefslogtreecommitdiffstats
path: root/crypto/async_tx/async_tx.c
blob: 802a5ce437d97be97f5101060aa7928d7405c6e8 (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
/*
 * core routines for the asynchronous memory transfer/transform api
 *
 * Copyright © 2006, Intel Corporation.
 *
 *	Dan Williams <dan.j.williams@intel.com>
 *
 *	with architecture considerations by:
 *	Neil Brown <neilb@suse.de>
 *	Jeff Garzik <jeff@garzik.org>
 *
 * This program is free software; you can redistribute it and/or modify it
 * under the terms and conditions of the GNU General Public License,
 * version 2, as published by the Free Software Foundation.
 *
 * This program is distributed in the hope it will be useful, but WITHOUT
 * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
 * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
 * more details.
 *
 * You should have received a copy of the GNU General Public License along with
 * this program; if not, write to the Free Software Foundation, Inc.,
 * 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA.
 *
 */
#include <linux/rculist.h>
#include <linux/kernel.h>
#include <linux/async_tx.h>

#ifdef CONFIG_DMA_ENGINE
static int __init async_tx_init(void)
{
	async_dmaengine_get();

	printk(KERN_INFO "async_tx: api initialized (async)\n");

	return 0;
}

static void __exit async_tx_exit(void)
{
	async_dmaengine_put();
}

/**
 * __async_tx_find_channel - find a channel to carry out the operation or let
 *	the transaction execute synchronously
 * @submit: transaction dependency and submission modifiers
 * @tx_type: transaction type
 */
struct dma_chan *
__async_tx_find_channel(struct async_submit_ctl *submit,
			enum dma_transaction_type tx_type)
{
	struct dma_async_tx_descriptor *depend_tx = submit->depend_tx;

	/* see if we can keep the chain on one channel */
	if (depend_tx &&
	    dma_has_cap(tx_type, depend_tx->chan->device->cap_mask))
		return depend_tx->chan;
	return async_dma_find_channel(tx_type);
}
EXPORT_SYMBOL_GPL(__async_tx_find_channel);
#else
static int __init async_tx_init(void)
{
	printk(KERN_INFO "async_tx: api initialized (sync-only)\n");
	return 0;
}

static void __exit async_tx_exit(void)
{
	do { } while (0);
}
#endif


/**
 * async_tx_channel_switch - queue an interrupt descriptor with a dependency
 * 	pre-attached.
 * @depend_tx: the operation that must finish before the new operation runs
 * @tx: the new operation
 */
static void
async_tx_channel_switch(struct dma_async_tx_descriptor *depend_tx,
			struct dma_async_tx_descriptor *tx)
{
	struct dma_chan *chan;
	struct dma_device *device;
	struct dma_async_tx_descriptor *intr_tx = (void *) ~0;

	/* first check to see if we can still append to depend_tx */
	spin_lock_bh(&depend_tx->lock);
	if (depend_tx->parent && depend_tx->chan == tx->chan) {
		tx->parent = depend_tx;
		depend_tx->next = tx;
		intr_tx = NULL;
	}
	spin_unlock_bh(&depend_tx->lock);

	if (!intr_tx)
		return;

	chan = depend_tx->chan;
	device = chan->device;

	/* see if we can schedule an interrupt
	 * otherwise poll for completion
	 */
	if (dma_has_cap(DMA_INTERRUPT, device->cap_mask))
		intr_tx = device->device_prep_dma_interrupt(chan, 0);
	else
		intr_tx = NULL;

	if (intr_tx) {
		intr_tx->callback = NULL;
		intr_tx->callback_param = NULL;
		tx->parent = intr_tx;
		/* safe to set ->next outside the lock since we know we are
		 * not submitted yet
		 */
		intr_tx->next = tx;

		/* check if we need to append */
		spin_lock_bh(&depend_tx->lock);
		if (depend_tx->parent) {
			intr_tx->parent = depend_tx;
			depend_tx->next = intr_tx;
			async_tx_ack(intr_tx);
			intr_tx = NULL;
		}
		spin_unlock_bh(&depend_tx->lock);

		if (intr_tx) {
			intr_tx->parent = NULL;
			intr_tx->tx_submit(intr_tx);
			async_tx_ack(intr_tx);
		}
	} else {
		if (dma_wait_for_async_tx(depend_tx) == DMA_ERROR)
			panic("%s: DMA_ERROR waiting for depend_tx\n",
			      __func__);
		tx->tx_submit(tx);
	}
}


/**
 * submit_disposition - flags for routing an incoming operation
 * @ASYNC_TX_SUBMITTED: we were able to append the new operation under the lock
 * @ASYNC_TX_CHANNEL_SWITCH: when the lock is dropped schedule a channel switch
 * @ASYNC_TX_DIRECT_SUBMIT: when the lock is dropped submit directly
 *
 * while holding depend_tx->lock we must avoid submitting new operations
 * to prevent a circular locking dependency with drivers that already
 * hold a channel lock when calling async_tx_run_dependencies.
 */
enum submit_disposition {
	ASYNC_TX_SUBMITTED,
	ASYNC_TX_CHANNEL_SWITCH,
	ASYNC_TX_DIRECT_SUBMIT,
};

void
async_tx_submit(struct dma_chan *chan, struct dma_async_tx_descriptor *tx,
		struct async_submit_ctl *submit)
{
	struct dma_async_tx_descriptor *depend_tx = submit->depend_tx;

	tx->callback = submit->cb_fn;
	tx->callback_param = submit->cb_param;

	if (depend_tx) {
		enum submit_disposition s;

		/* sanity check the dependency chain:
		 * 1/ if ack is already set then we cannot be sure
		 * we are referring to the correct operation
		 * 2/ dependencies are 1:1 i.e. two transactions can
		 * not depend on the same parent
		 */
		BUG_ON(async_tx_test_ack(depend_tx) || depend_tx->next ||
		       tx->parent);

		/* the lock prevents async_tx_run_dependencies from missing
		 * the setting of ->next when ->parent != NULL
		 */
		spin_lock_bh(&depend_tx->lock);
		if (depend_tx->parent) {
			/* we have a parent so we can not submit directly
			 * if we are staying on the same channel: append
			 * else: channel switch
			 */
			if (depend_tx->chan == chan) {
				tx->parent = depend_tx;
				depend_tx->next = tx;
				s = ASYNC_TX_SUBMITTED;
			} else
				s = ASYNC_TX_CHANNEL_SWITCH;
		} else {
			/* we do not have a parent so we may be able to submit
			 * directly if we are staying on the same channel
			 */
			if (depend_tx->chan == chan)
				s = ASYNC_TX_DIRECT_SUBMIT;
			else
				s = ASYNC_TX_CHANNEL_SWITCH;
		}
		spin_unlock_bh(&depend_tx->lock);

		switch (s) {
		case ASYNC_TX_SUBMITTED:
			break;
		case ASYNC_TX_CHANNEL_SWITCH:
			async_tx_channel_switch(depend_tx, tx);
			break;
		case ASYNC_TX_DIRECT_SUBMIT:
			tx->parent = NULL;
			tx->tx_submit(tx);
			break;
		}
	} else {
		tx->parent = NULL;
		tx->tx_submit(tx);
	}

	if (submit->flags & ASYNC_TX_ACK)
		async_tx_ack(tx);

	if (depend_tx)
		async_tx_ack(depend_tx);
}
EXPORT_SYMBOL_GPL(async_tx_submit);

/**
 * async_trigger_callback - schedules the callback function to be run
 * @submit: submission and completion parameters
 *
 * honored flags: ASYNC_TX_ACK
 *
 * The callback is run after any dependent operations have completed.
 */
struct dma_async_tx_descriptor *
async_trigger_callback(struct async_submit_ctl *submit)
{
	struct dma_chan *chan;
	struct dma_device *device;
	struct dma_async_tx_descriptor *tx;
	struct dma_async_tx_descriptor *depend_tx = submit->depend_tx;

	if (depend_tx) {
		chan = depend_tx->chan;
		device = chan->device;

		/* see if we can schedule an interrupt
		 * otherwise poll for completion
		 */
		if (device && !dma_has_cap(DMA_INTERRUPT, device->cap_mask))
			device = NULL;

		tx = device ? device->device_prep_dma_interrupt(chan, 0) : NULL;
	} else
		tx = NULL;

	if (tx) {
		pr_debug("%s: (async)\n", __func__);

		async_tx_submit(chan, tx, submit);
	} else {
		pr_debug("%s: (sync)\n", __func__);

		/* wait for any prerequisite operations */
		async_tx_quiesce(&submit->depend_tx);

		async_tx_sync_epilog(submit);
	}

	return tx;
}
EXPORT_SYMBOL_GPL(async_trigger_callback);

/**
 * async_tx_quiesce - ensure tx is complete and freeable upon return
 * @tx - transaction to quiesce
 */
void async_tx_quiesce(struct dma_async_tx_descriptor **tx)
{
	if (*tx) {
		/* if ack is already set then we cannot be sure
		 * we are referring to the correct operation
		 */
		BUG_ON(async_tx_test_ack(*tx));
		if (dma_wait_for_async_tx(*tx) == DMA_ERROR)
			panic("DMA_ERROR waiting for transaction\n");
		async_tx_ack(*tx);
		*tx = NULL;
	}
}
EXPORT_SYMBOL_GPL(async_tx_quiesce);

module_init(async_tx_init);
module_exit(async_tx_exit);

MODULE_AUTHOR("Intel Corporation");
MODULE_DESCRIPTION("Asynchronous Bulk Memory Transactions API");
MODULE_LICENSE("GPL");