[deliverable/linux.git] / block / blk.h

#ifndef BLK_INTERNAL_H
#define BLK_INTERNAL_H

#include <linux/idr.h>

/* Amount of time in which a process may batch requests */
#define BLK_BATCH_TIME	(HZ/50UL)

/* Number of requests a "batching" process may submit */
#define BLK_BATCH_REQ	32

/* Max future timer expiry for timeouts */
#define BLK_MAX_TIMEOUT		(5 * HZ)

extern struct kmem_cache *blk_requestq_cachep;
extern struct kmem_cache *request_cachep;
extern struct kobj_type blk_queue_ktype;
extern struct ida blk_queue_ida;

static inline void __blk_get_queue(struct request_queue *q)
{
	kobject_get(&q->kobj);
}

int blk_init_rl(struct request_list *rl, struct request_queue *q,
		gfp_t gfp_mask);
void blk_exit_rl(struct request_list *rl);
void init_request_from_bio(struct request *req, struct bio *bio);
void blk_rq_bio_prep(struct request_queue *q, struct request *rq,
			struct bio *bio);
int blk_rq_append_bio(struct request_queue *q, struct request *rq,
		      struct bio *bio);
void blk_queue_bypass_start(struct request_queue *q);
void blk_queue_bypass_end(struct request_queue *q);
void blk_dequeue_request(struct request *rq);
void __blk_queue_free_tags(struct request_queue *q);
bool __blk_end_bidi_request(struct request *rq, int error,
			    unsigned int nr_bytes, unsigned int bidi_bytes);

void blk_rq_timed_out_timer(unsigned long data);
void blk_rq_check_expired(struct request *rq, unsigned long *next_timeout,
			  unsigned int *next_set);
unsigned long blk_rq_timeout(unsigned long timeout);
void blk_add_timer(struct request *req);
void blk_delete_timer(struct request *);


bool bio_attempt_front_merge(struct request_queue *q, struct request *req,
			     struct bio *bio);
bool bio_attempt_back_merge(struct request_queue *q, struct request *req,
			    struct bio *bio);
bool blk_attempt_plug_merge(struct request_queue *q, struct bio *bio,
			    unsigned int *request_count);

void blk_account_io_start(struct request *req, bool new_io);
void blk_account_io_completion(struct request *req, unsigned int bytes);
void blk_account_io_done(struct request *req);

/*
 * Internal atomic flags for request handling
 */
enum rq_atomic_flags {
	REQ_ATOM_COMPLETE = 0,
	REQ_ATOM_STARTED,
};

/*
 * EH timer and IO completion will both attempt to 'grab' the request, make
 * sure that only one of them succeeds
 */
static inline int blk_mark_rq_complete(struct request *rq)
{
	return test_and_set_bit(REQ_ATOM_COMPLETE, &rq->atomic_flags);
}

static inline void blk_clear_rq_complete(struct request *rq)
{
	clear_bit(REQ_ATOM_COMPLETE, &rq->atomic_flags);
}

/*
 * Internal elevator interface
 */
#define ELV_ON_HASH(rq) ((rq)->cmd_flags & REQ_HASHED)

void blk_insert_flush(struct request *rq);
void blk_abort_flushes(struct request_queue *q);

static inline struct request *__elv_next_request(struct request_queue *q)
{
	struct request *rq;

	while (1) {
		if (!list_empty(&q->queue_head)) {
			rq = list_entry_rq(q->queue_head.next);
			return rq;
		}

		/*
		 * Flush request is running and flush request isn't queueable
		 * in the drive, we can hold the queue till flush request is
		 * finished. Even we don't do this, driver can't dispatch next
		 * requests and will requeue them. And this can improve
		 * throughput too. For example, we have request flush1, write1,
		 * flush 2. flush1 is dispatched, then queue is hold, write1
		 * isn't inserted to queue. After flush1 is finished, flush2
		 * will be dispatched. Since disk cache is already clean,
		 * flush2 will be finished very soon, so looks like flush2 is
		 * folded to flush1.
		 * Since the queue is hold, a flag is set to indicate the queue
		 * should be restarted later. Please see flush_end_io() for
		 * details.
		 */
		if (q->flush_pending_idx != q->flush_running_idx &&
				!queue_flush_queueable(q)) {
			q->flush_queue_delayed = 1;
			return NULL;
		}
		if (unlikely(blk_queue_bypass(q)) ||
		    !q->elevator->type->ops.elevator_dispatch_fn(q, 0))
			return NULL;
	}
}

static inline void elv_activate_rq(struct request_queue *q, struct request *rq)
{
	struct elevator_queue *e = q->elevator;

	if (e->type->ops.elevator_activate_req_fn)
		e->type->ops.elevator_activate_req_fn(q, rq);
}

static inline void elv_deactivate_rq(struct request_queue *q, struct request *rq)
{
	struct elevator_queue *e = q->elevator;

	if (e->type->ops.elevator_deactivate_req_fn)
		e->type->ops.elevator_deactivate_req_fn(q, rq);
}

#ifdef CONFIG_FAIL_IO_TIMEOUT
int blk_should_fake_timeout(struct request_queue *);
ssize_t part_timeout_show(struct device *, struct device_attribute *, char *);
ssize_t part_timeout_store(struct device *, struct device_attribute *,
				const char *, size_t);
#else
static inline int blk_should_fake_timeout(struct request_queue *q)
{
	return 0;
}
#endif

int ll_back_merge_fn(struct request_queue *q, struct request *req,
		     struct bio *bio);
int ll_front_merge_fn(struct request_queue *q, struct request *req, 
		      struct bio *bio);
int attempt_back_merge(struct request_queue *q, struct request *rq);
int attempt_front_merge(struct request_queue *q, struct request *rq);
int blk_attempt_req_merge(struct request_queue *q, struct request *rq,
				struct request *next);
void blk_recalc_rq_segments(struct request *rq);
void blk_rq_set_mixed_merge(struct request *rq);
bool blk_rq_merge_ok(struct request *rq, struct bio *bio);
int blk_try_merge(struct request *rq, struct bio *bio);

void blk_queue_congestion_threshold(struct request_queue *q);

void __blk_run_queue_uncond(struct request_queue *q);

int blk_dev_init(void);


/*
 * Return the threshold (number of used requests) at which the queue is
 * considered to be congested.  It include a little hysteresis to keep the
 * context switch rate down.
 */
static inline int queue_congestion_on_threshold(struct request_queue *q)
{
	return q->nr_congestion_on;
}

/*
 * The threshold at which a queue is considered to be uncongested
 */
static inline int queue_congestion_off_threshold(struct request_queue *q)
{
	return q->nr_congestion_off;
}

extern int blk_update_nr_requests(struct request_queue *, unsigned int);

/*
 * Contribute to IO statistics IFF:
 *
 *	a) it's attached to a gendisk, and
 *	b) the queue had IO stats enabled when this request was started, and
 *	c) it's a file system request
 */
static inline int blk_do_io_stat(struct request *rq)
{
	return rq->rq_disk &&
	       (rq->cmd_flags & REQ_IO_STAT) &&
		(rq->cmd_type == REQ_TYPE_FS);
}

/*
 * Internal io_context interface
 */
void get_io_context(struct io_context *ioc);
struct io_cq *ioc_lookup_icq(struct io_context *ioc, struct request_queue *q);
struct io_cq *ioc_create_icq(struct io_context *ioc, struct request_queue *q,
			     gfp_t gfp_mask);
void ioc_clear_queue(struct request_queue *q);

int create_task_io_context(struct task_struct *task, gfp_t gfp_mask, int node);

/**
 * create_io_context - try to create task->io_context
 * @gfp_mask: allocation mask
 * @node: allocation node
 *
 * If %current->io_context is %NULL, allocate a new io_context and install
 * it.  Returns the current %current->io_context which may be %NULL if
 * allocation failed.
 *
 * Note that this function can't be called with IRQ disabled because
 * task_lock which protects %current->io_context is IRQ-unsafe.
 */
static inline struct io_context *create_io_context(gfp_t gfp_mask, int node)
{
	WARN_ON_ONCE(irqs_disabled());
	if (unlikely(!current->io_context))
		create_task_io_context(current, gfp_mask, node);
	return current->io_context;
}

/*
 * Internal throttling interface
 */
#ifdef CONFIG_BLK_DEV_THROTTLING
extern bool blk_throtl_bio(struct request_queue *q, struct bio *bio);
extern void blk_throtl_drain(struct request_queue *q);
extern int blk_throtl_init(struct request_queue *q);
extern void blk_throtl_exit(struct request_queue *q);
#else /* CONFIG_BLK_DEV_THROTTLING */
static inline bool blk_throtl_bio(struct request_queue *q, struct bio *bio)
{
	return false;
}
static inline void blk_throtl_drain(struct request_queue *q) { }
static inline int blk_throtl_init(struct request_queue *q) { return 0; }
static inline void blk_throtl_exit(struct request_queue *q) { }
#endif /* CONFIG_BLK_DEV_THROTTLING */

#endif /* BLK_INTERNAL_H */
Commit	Line	Data
	1	#ifndef BLK_INTERNAL_H
	2	#define BLK_INTERNAL_H
	3
	4	#include <linux/idr.h>
	5
	6	/* Amount of time in which a process may batch requests */
	7	#define BLK_BATCH_TIME (HZ/50UL)
	8
	9	/* Number of requests a "batching" process may submit */
	10	#define BLK_BATCH_REQ 32
	11
	12	/* Max future timer expiry for timeouts */
	13	#define BLK_MAX_TIMEOUT (5 * HZ)
	14
	15	extern struct kmem_cache *blk_requestq_cachep;
	16	extern struct kmem_cache *request_cachep;
	17	extern struct kobj_type blk_queue_ktype;
	18	extern struct ida blk_queue_ida;
	19
	20	static inline void __blk_get_queue(struct request_queue *q)
	21	{
	22	kobject_get(&q->kobj);
	23	}
	24
	25	int blk_init_rl(struct request_list rl, struct request_queue q,
	26	gfp_t gfp_mask);
	27	void blk_exit_rl(struct request_list *rl);
	28	void init_request_from_bio(struct request req, struct bio bio);
	29	void blk_rq_bio_prep(struct request_queue q, struct request rq,
	30	struct bio *bio);
	31	int blk_rq_append_bio(struct request_queue q, struct request rq,
	32	struct bio *bio);
	33	void blk_queue_bypass_start(struct request_queue *q);
	34	void blk_queue_bypass_end(struct request_queue *q);
	35	void blk_dequeue_request(struct request *rq);
	36	void __blk_queue_free_tags(struct request_queue *q);
	37	bool __blk_end_bidi_request(struct request *rq, int error,
	38	unsigned int nr_bytes, unsigned int bidi_bytes);
	39
	40	void blk_rq_timed_out_timer(unsigned long data);
	41	void blk_rq_check_expired(struct request rq, unsigned long next_timeout,
	42	unsigned int *next_set);
	43	unsigned long blk_rq_timeout(unsigned long timeout);
	44	void blk_add_timer(struct request *req);
	45	void blk_delete_timer(struct request *);
	46
	47
	48	bool bio_attempt_front_merge(struct request_queue q, struct request req,
	49	struct bio *bio);
	50	bool bio_attempt_back_merge(struct request_queue q, struct request req,
	51	struct bio *bio);
	52	bool blk_attempt_plug_merge(struct request_queue q, struct bio bio,
	53	unsigned int *request_count);
	54
	55	void blk_account_io_start(struct request *req, bool new_io);
	56	void blk_account_io_completion(struct request *req, unsigned int bytes);
	57	void blk_account_io_done(struct request *req);
	58
	59	/*
	60	* Internal atomic flags for request handling
	61	*/
	62	enum rq_atomic_flags {
	63	REQ_ATOM_COMPLETE = 0,
	64	REQ_ATOM_STARTED,
	65	};
	66
	67	/*
	68	* EH timer and IO completion will both attempt to 'grab' the request, make
	69	* sure that only one of them succeeds
	70	*/
	71	static inline int blk_mark_rq_complete(struct request *rq)
	72	{
	73	return test_and_set_bit(REQ_ATOM_COMPLETE, &rq->atomic_flags);
	74	}
	75
	76	static inline void blk_clear_rq_complete(struct request *rq)
	77	{
	78	clear_bit(REQ_ATOM_COMPLETE, &rq->atomic_flags);
	79	}
	80
	81	/*
	82	* Internal elevator interface
	83	*/
	84	#define ELV_ON_HASH(rq) ((rq)->cmd_flags & REQ_HASHED)
	85
	86	void blk_insert_flush(struct request *rq);
	87	void blk_abort_flushes(struct request_queue *q);
	88
	89	static inline struct request __elv_next_request(struct request_queue q)
	90	{
	91	struct request *rq;
	92
	93	while (1) {
	94	if (!list_empty(&q->queue_head)) {
	95	rq = list_entry_rq(q->queue_head.next);
	96	return rq;
	97	}
	98
	99	/*
	100	* Flush request is running and flush request isn't queueable
	101	* in the drive, we can hold the queue till flush request is
	102	* finished. Even we don't do this, driver can't dispatch next
	103	* requests and will requeue them. And this can improve
	104	* throughput too. For example, we have request flush1, write1,
	105	* flush 2. flush1 is dispatched, then queue is hold, write1
	106	* isn't inserted to queue. After flush1 is finished, flush2
	107	* will be dispatched. Since disk cache is already clean,
	108	* flush2 will be finished very soon, so looks like flush2 is
	109	* folded to flush1.
	110	* Since the queue is hold, a flag is set to indicate the queue
	111	* should be restarted later. Please see flush_end_io() for
	112	* details.
	113	*/
	114	if (q->flush_pending_idx != q->flush_running_idx &&
	115	!queue_flush_queueable(q)) {
	116	q->flush_queue_delayed = 1;
	117	return NULL;
	118	}
	119	if (unlikely(blk_queue_bypass(q)) \|\|
	120	!q->elevator->type->ops.elevator_dispatch_fn(q, 0))
	121	return NULL;
	122	}
	123	}
	124
	125	static inline void elv_activate_rq(struct request_queue q, struct request rq)
	126	{
	127	struct elevator_queue *e = q->elevator;
	128
	129	if (e->type->ops.elevator_activate_req_fn)
	130	e->type->ops.elevator_activate_req_fn(q, rq);
	131	}
	132
	133	static inline void elv_deactivate_rq(struct request_queue q, struct request rq)
	134	{
	135	struct elevator_queue *e = q->elevator;
	136
	137	if (e->type->ops.elevator_deactivate_req_fn)
	138	e->type->ops.elevator_deactivate_req_fn(q, rq);
	139	}
	140
	141	#ifdef CONFIG_FAIL_IO_TIMEOUT
	142	int blk_should_fake_timeout(struct request_queue *);
	143	ssize_t part_timeout_show(struct device , struct device_attribute , char *);
	144	ssize_t part_timeout_store(struct device , struct device_attribute ,
	145	const char *, size_t);
	146	#else
	147	static inline int blk_should_fake_timeout(struct request_queue *q)
	148	{
	149	return 0;
	150	}
	151	#endif
	152
	153	int ll_back_merge_fn(struct request_queue q, struct request req,
	154	struct bio *bio);
	155	int ll_front_merge_fn(struct request_queue q, struct request req,
	156	struct bio *bio);
	157	int attempt_back_merge(struct request_queue q, struct request rq);
	158	int attempt_front_merge(struct request_queue q, struct request rq);
	159	int blk_attempt_req_merge(struct request_queue q, struct request rq,
	160	struct request *next);
	161	void blk_recalc_rq_segments(struct request *rq);
	162	void blk_rq_set_mixed_merge(struct request *rq);
	163	bool blk_rq_merge_ok(struct request rq, struct bio bio);
	164	int blk_try_merge(struct request rq, struct bio bio);
	165
	166	void blk_queue_congestion_threshold(struct request_queue *q);
	167
	168	void __blk_run_queue_uncond(struct request_queue *q);
	169
	170	int blk_dev_init(void);
	171
	172
	173	/*
	174	* Return the threshold (number of used requests) at which the queue is
	175	* considered to be congested. It include a little hysteresis to keep the
	176	* context switch rate down.
	177	*/
	178	static inline int queue_congestion_on_threshold(struct request_queue *q)
	179	{
	180	return q->nr_congestion_on;
	181	}
	182
	183	/*
	184	* The threshold at which a queue is considered to be uncongested
	185	*/
	186	static inline int queue_congestion_off_threshold(struct request_queue *q)
	187	{
	188	return q->nr_congestion_off;
	189	}
	190
	191	extern int blk_update_nr_requests(struct request_queue *, unsigned int);
	192
	193	/*
	194	* Contribute to IO statistics IFF:
	195	*
	196	* a) it's attached to a gendisk, and
	197	* b) the queue had IO stats enabled when this request was started, and
	198	* c) it's a file system request
	199	*/
	200	static inline int blk_do_io_stat(struct request *rq)
	201	{
	202	return rq->rq_disk &&
	203	(rq->cmd_flags & REQ_IO_STAT) &&
	204	(rq->cmd_type == REQ_TYPE_FS);
	205	}
	206
	207	/*
	208	* Internal io_context interface
	209	*/
	210	void get_io_context(struct io_context *ioc);
	211	struct io_cq ioc_lookup_icq(struct io_context ioc, struct request_queue *q);
	212	struct io_cq ioc_create_icq(struct io_context ioc, struct request_queue *q,
	213	gfp_t gfp_mask);
	214	void ioc_clear_queue(struct request_queue *q);
	215
	216	int create_task_io_context(struct task_struct *task, gfp_t gfp_mask, int node);
	217
	218	/**
	219	* create_io_context - try to create task->io_context
	220	* @gfp_mask: allocation mask
	221	* @node: allocation node
	222	*
	223	* If %current->io_context is %NULL, allocate a new io_context and install
	224	* it. Returns the current %current->io_context which may be %NULL if
	225	* allocation failed.
	226	*
	227	* Note that this function can't be called with IRQ disabled because
	228	* task_lock which protects %current->io_context is IRQ-unsafe.
	229	*/
	230	static inline struct io_context *create_io_context(gfp_t gfp_mask, int node)
	231	{
	232	WARN_ON_ONCE(irqs_disabled());
	233	if (unlikely(!current->io_context))
	234	create_task_io_context(current, gfp_mask, node);
	235	return current->io_context;
	236	}
	237
	238	/*
	239	* Internal throttling interface
	240	*/
	241	#ifdef CONFIG_BLK_DEV_THROTTLING
	242	extern bool blk_throtl_bio(struct request_queue q, struct bio bio);
	243	extern void blk_throtl_drain(struct request_queue *q);
	244	extern int blk_throtl_init(struct request_queue *q);
	245	extern void blk_throtl_exit(struct request_queue *q);
	246	#else /* CONFIG_BLK_DEV_THROTTLING */
	247	static inline bool blk_throtl_bio(struct request_queue q, struct bio bio)
	248	{
	249	return false;
	250	}
	251	static inline void blk_throtl_drain(struct request_queue *q) { }
	252	static inline int blk_throtl_init(struct request_queue *q) { return 0; }
	253	static inline void blk_throtl_exit(struct request_queue *q) { }
	254	#endif /* CONFIG_BLK_DEV_THROTTLING */
	255
	256	#endif /* BLK_INTERNAL_H */