summaryrefslogtreecommitdiff
path: root/block/blk-ioc.c
blob: dc5e69d335a070e2b74b7cf686231f96cabd1225 (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
/*
 * Functions related to io context handling
 */
#include <linux/kernel.h>
#include <linux/module.h>
#include <linux/init.h>
#include <linux/bio.h>
#include <linux/blkdev.h>
#include <linux/bootmem.h>	/* for max_pfn/max_low_pfn */
#include <linux/slab.h>

#include "blk.h"

/*
 * For io context allocations
 */
static struct kmem_cache *iocontext_cachep;

/**
 * get_io_context - increment reference count to io_context
 * @ioc: io_context to get
 *
 * Increment reference count to @ioc.
 */
void get_io_context(struct io_context *ioc)
{
	BUG_ON(atomic_long_read(&ioc->refcount) <= 0);
	atomic_long_inc(&ioc->refcount);
}
EXPORT_SYMBOL(get_io_context);

/*
 * Releasing ioc may nest into another put_io_context() leading to nested
 * fast path release.  As the ioc's can't be the same, this is okay but
 * makes lockdep whine.  Keep track of nesting and use it as subclass.
 */
#ifdef CONFIG_LOCKDEP
#define ioc_release_depth(q)		((q) ? (q)->ioc_release_depth : 0)
#define ioc_release_depth_inc(q)	(q)->ioc_release_depth++
#define ioc_release_depth_dec(q)	(q)->ioc_release_depth--
#else
#define ioc_release_depth(q)		0
#define ioc_release_depth_inc(q)	do { } while (0)
#define ioc_release_depth_dec(q)	do { } while (0)
#endif

/*
 * Slow path for ioc release in put_io_context().  Performs double-lock
 * dancing to unlink all icq's and then frees ioc.
 */
static void ioc_release_fn(struct work_struct *work)
{
	struct io_context *ioc = container_of(work, struct io_context,
					      release_work);
	struct request_queue *last_q = NULL;

	spin_lock_irq(&ioc->lock);

	while (!hlist_empty(&ioc->icq_list)) {
		struct io_cq *icq = hlist_entry(ioc->icq_list.first,
						struct io_cq, ioc_node);
		struct request_queue *this_q = icq->q;

		if (this_q != last_q) {
			/*
			 * Need to switch to @this_q.  Once we release
			 * @ioc->lock, it can go away along with @cic.
			 * Hold on to it.
			 */
			__blk_get_queue(this_q);

			/*
			 * blk_put_queue() might sleep thanks to kobject
			 * idiocy.  Always release both locks, put and
			 * restart.
			 */
			if (last_q) {
				spin_unlock(last_q->queue_lock);
				spin_unlock_irq(&ioc->lock);
				blk_put_queue(last_q);
			} else {
				spin_unlock_irq(&ioc->lock);
			}

			last_q = this_q;
			spin_lock_irq(this_q->queue_lock);
			spin_lock(&ioc->lock);
			continue;
		}
		ioc_release_depth_inc(this_q);
		icq->exit(icq);
		icq->release(icq);
		ioc_release_depth_dec(this_q);
	}

	if (last_q) {
		spin_unlock(last_q->queue_lock);
		spin_unlock_irq(&ioc->lock);
		blk_put_queue(last_q);
	} else {
		spin_unlock_irq(&ioc->lock);
	}

	kmem_cache_free(iocontext_cachep, ioc);
}

/**
 * put_io_context - put a reference of io_context
 * @ioc: io_context to put
 * @locked_q: request_queue the caller is holding queue_lock of (hint)
 *
 * Decrement reference count of @ioc and release it if the count reaches
 * zero.  If the caller is holding queue_lock of a queue, it can indicate
 * that with @locked_q.  This is an optimization hint and the caller is
 * allowed to pass in %NULL even when it's holding a queue_lock.
 */
void put_io_context(struct io_context *ioc, struct request_queue *locked_q)
{
	struct request_queue *last_q = locked_q;
	unsigned long flags;

	if (ioc == NULL)
		return;

	BUG_ON(atomic_long_read(&ioc->refcount) <= 0);
	if (locked_q)
		lockdep_assert_held(locked_q->queue_lock);

	if (!atomic_long_dec_and_test(&ioc->refcount))
		return;

	/*
	 * Destroy @ioc.  This is a bit messy because icq's are chained
	 * from both ioc and queue, and ioc->lock nests inside queue_lock.
	 * The inner ioc->lock should be held to walk our icq_list and then
	 * for each icq the outer matching queue_lock should be grabbed.
	 * ie. We need to do reverse-order double lock dancing.
	 *
	 * Another twist is that we are often called with one of the
	 * matching queue_locks held as indicated by @locked_q, which
	 * prevents performing double-lock dance for other queues.
	 *
	 * So, we do it in two stages.  The fast path uses the queue_lock
	 * the caller is holding and, if other queues need to be accessed,
	 * uses trylock to avoid introducing locking dependency.  This can
	 * handle most cases, especially if @ioc was performing IO on only
	 * single device.
	 *
	 * If trylock doesn't cut it, we defer to @ioc->release_work which
	 * can do all the double-locking dancing.
	 */
	spin_lock_irqsave_nested(&ioc->lock, flags,
				 ioc_release_depth(locked_q));

	while (!hlist_empty(&ioc->icq_list)) {
		struct io_cq *icq = hlist_entry(ioc->icq_list.first,
						struct io_cq, ioc_node);
		struct request_queue *this_q = icq->q;

		if (this_q != last_q) {
			if (last_q && last_q != locked_q)
				spin_unlock(last_q->queue_lock);
			last_q = NULL;

			if (!spin_trylock(this_q->queue_lock))
				break;
			last_q = this_q;
			continue;
		}
		ioc_release_depth_inc(this_q);
		icq->exit(icq);
		icq->release(icq);
		ioc_release_depth_dec(this_q);
	}

	if (last_q && last_q != locked_q)
		spin_unlock(last_q->queue_lock);

	spin_unlock_irqrestore(&ioc->lock, flags);

	/* if no icq is left, we're done; otherwise, kick release_work */
	if (hlist_empty(&ioc->icq_list))
		kmem_cache_free(iocontext_cachep, ioc);
	else
		schedule_work(&ioc->release_work);
}
EXPORT_SYMBOL(put_io_context);

/* Called by the exiting task */
void exit_io_context(struct task_struct *task)
{
	struct io_context *ioc;

	/* PF_EXITING prevents new io_context from being attached to @task */
	WARN_ON_ONCE(!(current->flags & PF_EXITING));

	task_lock(task);
	ioc = task->io_context;
	task->io_context = NULL;
	task_unlock(task);

	atomic_dec(&ioc->nr_tasks);
	put_io_context(ioc, NULL);
}

void create_io_context_slowpath(struct task_struct *task, gfp_t gfp_flags,
				int node)
{
	struct io_context *ioc;

	ioc = kmem_cache_alloc_node(iocontext_cachep, gfp_flags | __GFP_ZERO,
				    node);
	if (unlikely(!ioc))
		return;

	/* initialize */
	atomic_long_set(&ioc->refcount, 1);
	atomic_set(&ioc->nr_tasks, 1);
	spin_lock_init(&ioc->lock);
	INIT_RADIX_TREE(&ioc->icq_tree, GFP_ATOMIC | __GFP_HIGH);
	INIT_HLIST_HEAD(&ioc->icq_list);
	INIT_WORK(&ioc->release_work, ioc_release_fn);

	/* try to install, somebody might already have beaten us to it */
	task_lock(task);
	if (!task->io_context && !(task->flags & PF_EXITING))
		task->io_context = ioc;
	else
		kmem_cache_free(iocontext_cachep, ioc);
	task_unlock(task);
}
EXPORT_SYMBOL(create_io_context_slowpath);

/**
 * get_task_io_context - get io_context of a task
 * @task: task of interest
 * @gfp_flags: allocation flags, used if allocation is necessary
 * @node: allocation node, used if allocation is necessary
 *
 * Return io_context of @task.  If it doesn't exist, it is created with
 * @gfp_flags and @node.  The returned io_context has its reference count
 * incremented.
 *
 * This function always goes through task_lock() and it's better to use
 * %current->io_context + get_io_context() for %current.
 */
struct io_context *get_task_io_context(struct task_struct *task,
				       gfp_t gfp_flags, int node)
{
	struct io_context *ioc;

	might_sleep_if(gfp_flags & __GFP_WAIT);

	do {
		task_lock(task);
		ioc = task->io_context;
		if (likely(ioc)) {
			get_io_context(ioc);
			task_unlock(task);
			return ioc;
		}
		task_unlock(task);
	} while (create_io_context(task, gfp_flags, node));

	return NULL;
}
EXPORT_SYMBOL(get_task_io_context);

void ioc_set_changed(struct io_context *ioc, int which)
{
	struct io_cq *icq;
	struct hlist_node *n;

	hlist_for_each_entry(icq, n, &ioc->icq_list, ioc_node)
		set_bit(which, &icq->changed);
}

/**
 * ioc_ioprio_changed - notify ioprio change
 * @ioc: io_context of interest
 * @ioprio: new ioprio
 *
 * @ioc's ioprio has changed to @ioprio.  Set %ICQ_IOPRIO_CHANGED for all
 * icq's.  iosched is responsible for checking the bit and applying it on
 * request issue path.
 */
void ioc_ioprio_changed(struct io_context *ioc, int ioprio)
{
	unsigned long flags;

	spin_lock_irqsave(&ioc->lock, flags);
	ioc->ioprio = ioprio;
	ioc_set_changed(ioc, ICQ_IOPRIO_CHANGED);
	spin_unlock_irqrestore(&ioc->lock, flags);
}

/**
 * ioc_cgroup_changed - notify cgroup change
 * @ioc: io_context of interest
 *
 * @ioc's cgroup has changed.  Set %ICQ_CGROUP_CHANGED for all icq's.
 * iosched is responsible for checking the bit and applying it on request
 * issue path.
 */
void ioc_cgroup_changed(struct io_context *ioc)
{
	unsigned long flags;

	spin_lock_irqsave(&ioc->lock, flags);
	ioc_set_changed(ioc, ICQ_CGROUP_CHANGED);
	spin_unlock_irqrestore(&ioc->lock, flags);
}

static int __init blk_ioc_init(void)
{
	iocontext_cachep = kmem_cache_create("blkdev_ioc",
			sizeof(struct io_context), 0, SLAB_PANIC, NULL);
	return 0;
}
subsys_initcall(blk_ioc_init);