[mirror_ubuntu-zesty-kernel.git] / block / blk-cgroup.h

#ifndef _BLK_CGROUP_H
#define _BLK_CGROUP_H
/*
 * Common Block IO controller cgroup interface
 *
 * Based on ideas and code from CFQ, CFS and BFQ:
 * Copyright (C) 2003 Jens Axboe <axboe@kernel.dk>
 *
 * Copyright (C) 2008 Fabio Checconi <fabio@gandalf.sssup.it>
 *		      Paolo Valente <paolo.valente@unimore.it>
 *
 * Copyright (C) 2009 Vivek Goyal <vgoyal@redhat.com>
 * 	              Nauman Rafique <nauman@google.com>
 */

#include <linux/cgroup.h>
#include <linux/u64_stats_sync.h>

enum blkio_policy_id {
	BLKIO_POLICY_PROP = 0,		/* Proportional Bandwidth division */
	BLKIO_POLICY_THROTL,		/* Throttling */

	BLKIO_NR_POLICIES,
};

/* Max limits for throttle policy */
#define THROTL_IOPS_MAX		UINT_MAX

#ifdef CONFIG_BLK_CGROUP

enum stat_type {
	/* Number of IOs merged */
	BLKIO_STAT_MERGED,
	/* Total time spent (in ns) between request dispatch to the driver and
	 * request completion for IOs doen by this cgroup. This may not be
	 * accurate when NCQ is turned on. */
	BLKIO_STAT_SERVICE_TIME,
	/* Total time spent waiting in scheduler queue in ns */
	BLKIO_STAT_WAIT_TIME,
	/* Number of IOs queued up */
	BLKIO_STAT_QUEUED,

	/* All the single valued stats go below this */
	BLKIO_STAT_TIME,
#ifdef CONFIG_DEBUG_BLK_CGROUP
	/* Time not charged to this cgroup */
	BLKIO_STAT_UNACCOUNTED_TIME,
	BLKIO_STAT_AVG_QUEUE_SIZE,
	BLKIO_STAT_IDLE_TIME,
	BLKIO_STAT_EMPTY_TIME,
	BLKIO_STAT_GROUP_WAIT_TIME,
	BLKIO_STAT_DEQUEUE
#endif
};

/* Types lower than this live in stat_arr and have subtypes */
#define BLKIO_STAT_ARR_NR	(BLKIO_STAT_QUEUED + 1)

/* Per cpu stats */
enum stat_type_cpu {
	/* Total bytes transferred */
	BLKIO_STAT_CPU_SERVICE_BYTES,
	/* Total IOs serviced, post merge */
	BLKIO_STAT_CPU_SERVICED,

	/* All the single valued stats go below this */
	BLKIO_STAT_CPU_SECTORS,
};

#define BLKIO_STAT_CPU_ARR_NR	(BLKIO_STAT_CPU_SERVICED + 1)

enum stat_sub_type {
	BLKIO_STAT_READ = 0,
	BLKIO_STAT_WRITE,
	BLKIO_STAT_SYNC,
	BLKIO_STAT_ASYNC,
	BLKIO_STAT_TOTAL
};

/* blkg state flags */
enum blkg_state_flags {
	BLKG_waiting = 0,
	BLKG_idling,
	BLKG_empty,
};

/* cgroup files owned by proportional weight policy */
enum blkcg_file_name_prop {
	BLKIO_PROP_weight = 1,
	BLKIO_PROP_weight_device,
	BLKIO_PROP_io_service_bytes,
	BLKIO_PROP_io_serviced,
	BLKIO_PROP_time,
	BLKIO_PROP_sectors,
	BLKIO_PROP_unaccounted_time,
	BLKIO_PROP_io_service_time,
	BLKIO_PROP_io_wait_time,
	BLKIO_PROP_io_merged,
	BLKIO_PROP_io_queued,
	BLKIO_PROP_avg_queue_size,
	BLKIO_PROP_group_wait_time,
	BLKIO_PROP_idle_time,
	BLKIO_PROP_empty_time,
	BLKIO_PROP_dequeue,
};

/* cgroup files owned by throttle policy */
enum blkcg_file_name_throtl {
	BLKIO_THROTL_read_bps_device,
	BLKIO_THROTL_write_bps_device,
	BLKIO_THROTL_read_iops_device,
	BLKIO_THROTL_write_iops_device,
	BLKIO_THROTL_io_service_bytes,
	BLKIO_THROTL_io_serviced,
};

struct blkio_cgroup {
	struct cgroup_subsys_state css;
	unsigned int weight;
	spinlock_t lock;
	struct hlist_head blkg_list;

	/* for policies to test whether associated blkcg has changed */
	uint64_t id;
};

struct blkio_group_stats {
	struct u64_stats_sync syncp;
	/* total disk time and nr sectors dispatched by this group */
	uint64_t time;
	uint64_t stat_arr[BLKIO_STAT_ARR_NR][BLKIO_STAT_TOTAL];
#ifdef CONFIG_DEBUG_BLK_CGROUP
	/* Time not charged to this cgroup */
	uint64_t unaccounted_time;

	/* Sum of number of IOs queued across all samples */
	uint64_t avg_queue_size_sum;
	/* Count of samples taken for average */
	uint64_t avg_queue_size_samples;
	/* How many times this group has been removed from service tree */
	unsigned long dequeue;

	/* Total time spent waiting for it to be assigned a timeslice. */
	uint64_t group_wait_time;

	/* Time spent idling for this blkio_group */
	uint64_t idle_time;
	/*
	 * Total time when we have requests queued and do not contain the
	 * current active queue.
	 */
	uint64_t empty_time;

	/* fields after this shouldn't be cleared on stat reset */
	uint64_t start_group_wait_time;
	uint64_t start_idle_time;
	uint64_t start_empty_time;
	uint16_t flags;
#endif
};

#ifdef CONFIG_DEBUG_BLK_CGROUP
#define BLKG_STATS_DEBUG_CLEAR_START	\
	offsetof(struct blkio_group_stats, unaccounted_time)
#define BLKG_STATS_DEBUG_CLEAR_SIZE	\
	(offsetof(struct blkio_group_stats, start_group_wait_time) - \
	 BLKG_STATS_DEBUG_CLEAR_START)
#endif

/* Per cpu blkio group stats */
struct blkio_group_stats_cpu {
	uint64_t sectors;
	uint64_t stat_arr_cpu[BLKIO_STAT_CPU_ARR_NR][BLKIO_STAT_TOTAL];
	struct u64_stats_sync syncp;
};

struct blkio_group_conf {
	unsigned int weight;
	unsigned int iops[2];
	u64 bps[2];
};

/* per-blkg per-policy data */
struct blkg_policy_data {
	/* the blkg this per-policy data belongs to */
	struct blkio_group *blkg;

	/* Configuration */
	struct blkio_group_conf conf;

	struct blkio_group_stats stats;
	/* Per cpu stats pointer */
	struct blkio_group_stats_cpu __percpu *stats_cpu;

	/* pol->pdata_size bytes of private data used by policy impl */
	char pdata[] __aligned(__alignof__(unsigned long long));
};

struct blkio_group {
	/* Pointer to the associated request_queue */
	struct request_queue *q;
	struct list_head q_node;
	struct hlist_node blkcg_node;
	struct blkio_cgroup *blkcg;
	/* Store cgroup path */
	char path[128];
	/* reference count */
	int refcnt;

	struct blkg_policy_data *pd[BLKIO_NR_POLICIES];

	/* List of blkg waiting for per cpu stats memory to be allocated */
	struct list_head alloc_node;
	struct rcu_head rcu_head;
};

typedef void (blkio_init_group_fn)(struct blkio_group *blkg);
typedef void (blkio_update_group_weight_fn)(struct request_queue *q,
			struct blkio_group *blkg, unsigned int weight);
typedef void (blkio_update_group_read_bps_fn)(struct request_queue *q,
			struct blkio_group *blkg, u64 read_bps);
typedef void (blkio_update_group_write_bps_fn)(struct request_queue *q,
			struct blkio_group *blkg, u64 write_bps);
typedef void (blkio_update_group_read_iops_fn)(struct request_queue *q,
			struct blkio_group *blkg, unsigned int read_iops);
typedef void (blkio_update_group_write_iops_fn)(struct request_queue *q,
			struct blkio_group *blkg, unsigned int write_iops);

struct blkio_policy_ops {
	blkio_init_group_fn *blkio_init_group_fn;
	blkio_update_group_weight_fn *blkio_update_group_weight_fn;
	blkio_update_group_read_bps_fn *blkio_update_group_read_bps_fn;
	blkio_update_group_write_bps_fn *blkio_update_group_write_bps_fn;
	blkio_update_group_read_iops_fn *blkio_update_group_read_iops_fn;
	blkio_update_group_write_iops_fn *blkio_update_group_write_iops_fn;
};

struct blkio_policy_type {
	struct list_head list;
	struct blkio_policy_ops ops;
	enum blkio_policy_id plid;
	size_t pdata_size;		/* policy specific private data size */
};

extern int blkcg_init_queue(struct request_queue *q);
extern void blkcg_drain_queue(struct request_queue *q);
extern void blkcg_exit_queue(struct request_queue *q);

/* Blkio controller policy registration */
extern void blkio_policy_register(struct blkio_policy_type *);
extern void blkio_policy_unregister(struct blkio_policy_type *);
extern void blkg_destroy_all(struct request_queue *q, bool destroy_root);
extern void update_root_blkg_pd(struct request_queue *q,
				enum blkio_policy_id plid);

/**
 * blkg_to_pdata - get policy private data
 * @blkg: blkg of interest
 * @pol: policy of interest
 *
 * Return pointer to private data associated with the @blkg-@pol pair.
 */
static inline void *blkg_to_pdata(struct blkio_group *blkg,
			      struct blkio_policy_type *pol)
{
	return blkg ? blkg->pd[pol->plid]->pdata : NULL;
}

/**
 * pdata_to_blkg - get blkg associated with policy private data
 * @pdata: policy private data of interest
 *
 * @pdata is policy private data.  Determine the blkg it's associated with.
 */
static inline struct blkio_group *pdata_to_blkg(void *pdata)
{
	if (pdata) {
		struct blkg_policy_data *pd =
			container_of(pdata, struct blkg_policy_data, pdata);
		return pd->blkg;
	}
	return NULL;
}

static inline char *blkg_path(struct blkio_group *blkg)
{
	return blkg->path;
}

/**
 * blkg_get - get a blkg reference
 * @blkg: blkg to get
 *
 * The caller should be holding queue_lock and an existing reference.
 */
static inline void blkg_get(struct blkio_group *blkg)
{
	lockdep_assert_held(blkg->q->queue_lock);
	WARN_ON_ONCE(!blkg->refcnt);
	blkg->refcnt++;
}

void __blkg_release(struct blkio_group *blkg);

/**
 * blkg_put - put a blkg reference
 * @blkg: blkg to put
 *
 * The caller should be holding queue_lock.
 */
static inline void blkg_put(struct blkio_group *blkg)
{
	lockdep_assert_held(blkg->q->queue_lock);
	WARN_ON_ONCE(blkg->refcnt <= 0);
	if (!--blkg->refcnt)
		__blkg_release(blkg);
}

#else

struct blkio_group {
};

struct blkio_policy_type {
};

static inline int blkcg_init_queue(struct request_queue *q) { return 0; }
static inline void blkcg_drain_queue(struct request_queue *q) { }
static inline void blkcg_exit_queue(struct request_queue *q) { }
static inline void blkio_policy_register(struct blkio_policy_type *blkiop) { }
static inline void blkio_policy_unregister(struct blkio_policy_type *blkiop) { }
static inline void blkg_destroy_all(struct request_queue *q,
				    bool destory_root) { }
static inline void update_root_blkg_pd(struct request_queue *q,
				       enum blkio_policy_id plid) { }

static inline void *blkg_to_pdata(struct blkio_group *blkg,
				struct blkio_policy_type *pol) { return NULL; }
static inline struct blkio_group *pdata_to_blkg(void *pdata,
				struct blkio_policy_type *pol) { return NULL; }
static inline char *blkg_path(struct blkio_group *blkg) { return NULL; }
static inline void blkg_get(struct blkio_group *blkg) { }
static inline void blkg_put(struct blkio_group *blkg) { }

#endif

#define BLKIO_WEIGHT_MIN	10
#define BLKIO_WEIGHT_MAX	1000
#define BLKIO_WEIGHT_DEFAULT	500

#ifdef CONFIG_DEBUG_BLK_CGROUP
void blkiocg_update_avg_queue_size_stats(struct blkio_group *blkg,
					 struct blkio_policy_type *pol);
void blkiocg_update_dequeue_stats(struct blkio_group *blkg,
				  struct blkio_policy_type *pol,
				  unsigned long dequeue);
void blkiocg_update_set_idle_time_stats(struct blkio_group *blkg,
					struct blkio_policy_type *pol);
void blkiocg_update_idle_time_stats(struct blkio_group *blkg,
				    struct blkio_policy_type *pol);
void blkiocg_set_start_empty_time(struct blkio_group *blkg,
				  struct blkio_policy_type *pol);

#define BLKG_FLAG_FNS(name)						\
static inline void blkio_mark_blkg_##name(				\
		struct blkio_group_stats *stats)			\
{									\
	stats->flags |= (1 << BLKG_##name);				\
}									\
static inline void blkio_clear_blkg_##name(				\
		struct blkio_group_stats *stats)			\
{									\
	stats->flags &= ~(1 << BLKG_##name);				\
}									\
static inline int blkio_blkg_##name(struct blkio_group_stats *stats)	\
{									\
	return (stats->flags & (1 << BLKG_##name)) != 0;		\
}									\

BLKG_FLAG_FNS(waiting)
BLKG_FLAG_FNS(idling)
BLKG_FLAG_FNS(empty)
#undef BLKG_FLAG_FNS
#else
static inline void blkiocg_update_avg_queue_size_stats(struct blkio_group *blkg,
			struct blkio_policy_type *pol) { }
static inline void blkiocg_update_dequeue_stats(struct blkio_group *blkg,
			struct blkio_policy_type *pol, unsigned long dequeue) { }
static inline void blkiocg_update_set_idle_time_stats(struct blkio_group *blkg,
			struct blkio_policy_type *pol) { }
static inline void blkiocg_update_idle_time_stats(struct blkio_group *blkg,
			struct blkio_policy_type *pol) { }
static inline void blkiocg_set_start_empty_time(struct blkio_group *blkg,
			struct blkio_policy_type *pol) { }
#endif

#ifdef CONFIG_BLK_CGROUP
extern struct blkio_cgroup blkio_root_cgroup;
extern struct blkio_cgroup *cgroup_to_blkio_cgroup(struct cgroup *cgroup);
extern struct blkio_cgroup *bio_blkio_cgroup(struct bio *bio);
extern struct blkio_group *blkg_lookup(struct blkio_cgroup *blkcg,
				       struct request_queue *q);
struct blkio_group *blkg_lookup_create(struct blkio_cgroup *blkcg,
				       struct request_queue *q,
				       bool for_root);
void blkiocg_update_timeslice_used(struct blkio_group *blkg,
				   struct blkio_policy_type *pol,
				   unsigned long time,
				   unsigned long unaccounted_time);
void blkiocg_update_dispatch_stats(struct blkio_group *blkg,
				   struct blkio_policy_type *pol,
				   uint64_t bytes, bool direction, bool sync);
void blkiocg_update_completion_stats(struct blkio_group *blkg,
				     struct blkio_policy_type *pol,
				     uint64_t start_time,
				     uint64_t io_start_time, bool direction,
				     bool sync);
void blkiocg_update_io_merged_stats(struct blkio_group *blkg,
				    struct blkio_policy_type *pol,
				    bool direction, bool sync);
void blkiocg_update_io_add_stats(struct blkio_group *blkg,
				 struct blkio_policy_type *pol,
				 struct blkio_group *curr_blkg, bool direction,
				 bool sync);
void blkiocg_update_io_remove_stats(struct blkio_group *blkg,
				    struct blkio_policy_type *pol,
				    bool direction, bool sync);
#else
struct cgroup;
static inline struct blkio_cgroup *
cgroup_to_blkio_cgroup(struct cgroup *cgroup) { return NULL; }
static inline struct blkio_cgroup *
bio_blkio_cgroup(struct bio *bio) { return NULL; }

static inline struct blkio_group *blkg_lookup(struct blkio_cgroup *blkcg,
					      void *key) { return NULL; }
static inline void blkiocg_update_timeslice_used(struct blkio_group *blkg,
			struct blkio_policy_type *pol, unsigned long time,
			unsigned long unaccounted_time) { }
static inline void blkiocg_update_dispatch_stats(struct blkio_group *blkg,
			struct blkio_policy_type *pol, uint64_t bytes,
			bool direction, bool sync) { }
static inline void blkiocg_update_completion_stats(struct blkio_group *blkg,
			struct blkio_policy_type *pol, uint64_t start_time,
			uint64_t io_start_time, bool direction, bool sync) { }
static inline void blkiocg_update_io_merged_stats(struct blkio_group *blkg,
			struct blkio_policy_type *pol, bool direction,
			bool sync) { }
static inline void blkiocg_update_io_add_stats(struct blkio_group *blkg,
			struct blkio_policy_type *pol,
			struct blkio_group *curr_blkg, bool direction,
			bool sync) { }
static inline void blkiocg_update_io_remove_stats(struct blkio_group *blkg,
			struct blkio_policy_type *pol, bool direction,
			bool sync) { }
#endif
#endif /* _BLK_CGROUP_H */
Commit	Line	Data
31e4c28d VG	1	#ifndef _BLK_CGROUP_H
	2	#define _BLK_CGROUP_H
	3	/*
	4	* Common Block IO controller cgroup interface
	5	*
	6	* Based on ideas and code from CFQ, CFS and BFQ:
	7	* Copyright (C) 2003 Jens Axboe <axboe@kernel.dk>
	8	*
	9	* Copyright (C) 2008 Fabio Checconi <fabio@gandalf.sssup.it>
	10	* Paolo Valente <paolo.valente@unimore.it>
	11	*
	12	* Copyright (C) 2009 Vivek Goyal <vgoyal@redhat.com>
	13	* Nauman Rafique <nauman@google.com>
	14	*/
	15
	16	#include <linux/cgroup.h>
575969a0	17	#include <linux/u64_stats_sync.h>
31e4c28d	18
062a644d VG	19	enum blkio_policy_id {
062a644d VG	20	BLKIO_POLICY_PROP = 0, /* Proportional Bandwidth division */
4c9eefa1	21	BLKIO_POLICY_THROTL, /* Throttling */
035d10b2 TH	22
035d10b2 TH	23	BLKIO_NR_POLICIES,
062a644d VG	24	};
062a644d VG	25
9355aede VG	26	/* Max limits for throttle policy */
	27	#define THROTL_IOPS_MAX UINT_MAX
	28
32e380ae	29	#ifdef CONFIG_BLK_CGROUP
2f5ea477	30
84c124da	31	enum stat_type {
5fe224d2 TH	32	/* Number of IOs merged */
5fe224d2 TH	33	BLKIO_STAT_MERGED,
84c124da DS	34	/* Total time spent (in ns) between request dispatch to the driver and
	35	* request completion for IOs doen by this cgroup. This may not be
	36	* accurate when NCQ is turned on. */
5fe224d2	37	BLKIO_STAT_SERVICE_TIME,
84c124da DS	38	/* Total time spent waiting in scheduler queue in ns */
84c124da DS	39	BLKIO_STAT_WAIT_TIME,
cdc1184c DS	40	/* Number of IOs queued up */
cdc1184c DS	41	BLKIO_STAT_QUEUED,
c4c76a05	42
84c124da DS	43	/* All the single valued stats go below this */
84c124da DS	44	BLKIO_STAT_TIME,
a23e6869	45	#ifdef CONFIG_DEBUG_BLK_CGROUP
167400d3 JT	46	/* Time not charged to this cgroup */
167400d3 JT	47	BLKIO_STAT_UNACCOUNTED_TIME,
cdc1184c	48	BLKIO_STAT_AVG_QUEUE_SIZE,
812df48d DS	49	BLKIO_STAT_IDLE_TIME,
	50	BLKIO_STAT_EMPTY_TIME,
	51	BLKIO_STAT_GROUP_WAIT_TIME,
84c124da DS	52	BLKIO_STAT_DEQUEUE
	53	#endif
	54	};
	55
c4c76a05 TH	56	/* Types lower than this live in stat_arr and have subtypes */
	57	#define BLKIO_STAT_ARR_NR (BLKIO_STAT_QUEUED + 1)
	58
5624a4e4 VG	59	/* Per cpu stats */
5624a4e4 VG	60	enum stat_type_cpu {
5624a4e4 VG	61	/* Total bytes transferred */
	62	BLKIO_STAT_CPU_SERVICE_BYTES,
	63	/* Total IOs serviced, post merge */
	64	BLKIO_STAT_CPU_SERVICED,
2aa4a152 TH	65
	66	/* All the single valued stats go below this */
	67	BLKIO_STAT_CPU_SECTORS,
5624a4e4 VG	68	};
5624a4e4 VG	69
2aa4a152 TH	70	#define BLKIO_STAT_CPU_ARR_NR (BLKIO_STAT_CPU_SERVICED + 1)
2aa4a152 TH	71
84c124da DS	72	enum stat_sub_type {
	73	BLKIO_STAT_READ = 0,
	74	BLKIO_STAT_WRITE,
	75	BLKIO_STAT_SYNC,
	76	BLKIO_STAT_ASYNC,
	77	BLKIO_STAT_TOTAL
303a3acb DS	78	};
303a3acb DS	79
812df48d DS	80	/* blkg state flags */
	81	enum blkg_state_flags {
	82	BLKG_waiting = 0,
	83	BLKG_idling,
	84	BLKG_empty,
	85	};
	86
062a644d VG	87	/* cgroup files owned by proportional weight policy */
	88	enum blkcg_file_name_prop {
	89	BLKIO_PROP_weight = 1,
	90	BLKIO_PROP_weight_device,
	91	BLKIO_PROP_io_service_bytes,
	92	BLKIO_PROP_io_serviced,
	93	BLKIO_PROP_time,
	94	BLKIO_PROP_sectors,
167400d3	95	BLKIO_PROP_unaccounted_time,
062a644d VG	96	BLKIO_PROP_io_service_time,
	97	BLKIO_PROP_io_wait_time,
	98	BLKIO_PROP_io_merged,
	99	BLKIO_PROP_io_queued,
	100	BLKIO_PROP_avg_queue_size,
	101	BLKIO_PROP_group_wait_time,
	102	BLKIO_PROP_idle_time,
	103	BLKIO_PROP_empty_time,
	104	BLKIO_PROP_dequeue,
	105	};
	106
4c9eefa1 VG	107	/* cgroup files owned by throttle policy */
	108	enum blkcg_file_name_throtl {
	109	BLKIO_THROTL_read_bps_device,
	110	BLKIO_THROTL_write_bps_device,
7702e8f4 VG	111	BLKIO_THROTL_read_iops_device,
7702e8f4 VG	112	BLKIO_THROTL_write_iops_device,
4c9eefa1 VG	113	BLKIO_THROTL_io_service_bytes,
	114	BLKIO_THROTL_io_serviced,
	115	};
	116
31e4c28d VG	117	struct blkio_cgroup {
	118	struct cgroup_subsys_state css;
	119	unsigned int weight;
	120	spinlock_t lock;
	121	struct hlist_head blkg_list;
9a9e8a26 TH	122
	123	/* for policies to test whether associated blkcg has changed */
	124	uint64_t id;
31e4c28d VG	125	};
31e4c28d VG	126
303a3acb	127	struct blkio_group_stats {
edf1b879	128	struct u64_stats_sync syncp;
303a3acb DS	129	/* total disk time and nr sectors dispatched by this group */
303a3acb DS	130	uint64_t time;
c4c76a05	131	uint64_t stat_arr[BLKIO_STAT_ARR_NR][BLKIO_STAT_TOTAL];
303a3acb	132	#ifdef CONFIG_DEBUG_BLK_CGROUP
a23e6869 VG	133	/* Time not charged to this cgroup */
	134	uint64_t unaccounted_time;
	135
cdc1184c DS	136	/* Sum of number of IOs queued across all samples */
	137	uint64_t avg_queue_size_sum;
	138	/* Count of samples taken for average */
	139	uint64_t avg_queue_size_samples;
303a3acb DS	140	/* How many times this group has been removed from service tree */
303a3acb DS	141	unsigned long dequeue;
812df48d DS	142
	143	/* Total time spent waiting for it to be assigned a timeslice. */
	144	uint64_t group_wait_time;
812df48d DS	145
	146	/* Time spent idling for this blkio_group */
	147	uint64_t idle_time;
812df48d DS	148	/*
	149	* Total time when we have requests queued and do not contain the
	150	* current active queue.
	151	*/
	152	uint64_t empty_time;
997a026c TH	153
	154	/* fields after this shouldn't be cleared on stat reset */
	155	uint64_t start_group_wait_time;
	156	uint64_t start_idle_time;
812df48d DS	157	uint64_t start_empty_time;
812df48d DS	158	uint16_t flags;
303a3acb DS	159	#endif
	160	};
	161
997a026c TH	162	#ifdef CONFIG_DEBUG_BLK_CGROUP
	163	#define BLKG_STATS_DEBUG_CLEAR_START \
	164	offsetof(struct blkio_group_stats, unaccounted_time)
	165	#define BLKG_STATS_DEBUG_CLEAR_SIZE \
	166	(offsetof(struct blkio_group_stats, start_group_wait_time) - \
	167	BLKG_STATS_DEBUG_CLEAR_START)
	168	#endif
	169
5624a4e4 VG	170	/* Per cpu blkio group stats */
	171	struct blkio_group_stats_cpu {
	172	uint64_t sectors;
2aa4a152	173	uint64_t stat_arr_cpu[BLKIO_STAT_CPU_ARR_NR][BLKIO_STAT_TOTAL];
575969a0	174	struct u64_stats_sync syncp;
5624a4e4 VG	175	};
5624a4e4 VG	176
e56da7e2 TH	177	struct blkio_group_conf {
	178	unsigned int weight;
	179	unsigned int iops[2];
	180	u64 bps[2];
	181	};
	182
0381411e TH	183	/* per-blkg per-policy data */
	184	struct blkg_policy_data {
	185	/* the blkg this per-policy data belongs to */
	186	struct blkio_group *blkg;
	187
549d3aa8 TH	188	/* Configuration */
	189	struct blkio_group_conf conf;
	190
	191	struct blkio_group_stats stats;
	192	/* Per cpu stats pointer */
	193	struct blkio_group_stats_cpu __percpu *stats_cpu;
	194
0381411e TH	195	/* pol->pdata_size bytes of private data used by policy impl */
	196	char pdata[] __aligned(__alignof__(unsigned long long));
	197	};
	198
31e4c28d	199	struct blkio_group {
c875f4d0 TH	200	/* Pointer to the associated request_queue */
c875f4d0 TH	201	struct request_queue *q;
e8989fae	202	struct list_head q_node;
31e4c28d	203	struct hlist_node blkcg_node;
7ee9c562	204	struct blkio_cgroup *blkcg;
2868ef7b VG	205	/* Store cgroup path */
2868ef7b VG	206	char path[128];
1adaf3dd TH	207	/* reference count */
1adaf3dd TH	208	int refcnt;
22084190	209
549d3aa8	210	struct blkg_policy_data *pd[BLKIO_NR_POLICIES];
1adaf3dd	211
1cd9e039 VG	212	/* List of blkg waiting for per cpu stats memory to be allocated */
1cd9e039 VG	213	struct list_head alloc_node;
1adaf3dd	214	struct rcu_head rcu_head;
31e4c28d VG	215	};
31e4c28d VG	216
0381411e	217	typedef void (blkio_init_group_fn)(struct blkio_group *blkg);
ca32aefc	218	typedef void (blkio_update_group_weight_fn)(struct request_queue *q,
fe071437	219	struct blkio_group *blkg, unsigned int weight);
ca32aefc	220	typedef void (blkio_update_group_read_bps_fn)(struct request_queue *q,
fe071437	221	struct blkio_group *blkg, u64 read_bps);
ca32aefc	222	typedef void (blkio_update_group_write_bps_fn)(struct request_queue *q,
fe071437	223	struct blkio_group *blkg, u64 write_bps);
ca32aefc	224	typedef void (blkio_update_group_read_iops_fn)(struct request_queue *q,
fe071437	225	struct blkio_group *blkg, unsigned int read_iops);
ca32aefc	226	typedef void (blkio_update_group_write_iops_fn)(struct request_queue *q,
fe071437	227	struct blkio_group *blkg, unsigned int write_iops);
3e252066 VG	228
3e252066 VG	229	struct blkio_policy_ops {
0381411e	230	blkio_init_group_fn *blkio_init_group_fn;
3e252066	231	blkio_update_group_weight_fn *blkio_update_group_weight_fn;
4c9eefa1 VG	232	blkio_update_group_read_bps_fn *blkio_update_group_read_bps_fn;
4c9eefa1 VG	233	blkio_update_group_write_bps_fn *blkio_update_group_write_bps_fn;
7702e8f4 VG	234	blkio_update_group_read_iops_fn *blkio_update_group_read_iops_fn;
7702e8f4 VG	235	blkio_update_group_write_iops_fn *blkio_update_group_write_iops_fn;
3e252066 VG	236	};
	237
	238	struct blkio_policy_type {
	239	struct list_head list;
	240	struct blkio_policy_ops ops;
062a644d	241	enum blkio_policy_id plid;
0381411e	242	size_t pdata_size; /* policy specific private data size */
3e252066 VG	243	};
3e252066 VG	244
5efd6113 TH	245	extern int blkcg_init_queue(struct request_queue *q);
	246	extern void blkcg_drain_queue(struct request_queue *q);
	247	extern void blkcg_exit_queue(struct request_queue *q);
	248
3e252066 VG	249	/* Blkio controller policy registration */
	250	extern void blkio_policy_register(struct blkio_policy_type *);
	251	extern void blkio_policy_unregister(struct blkio_policy_type *);
e8989fae TH	252	extern void blkg_destroy_all(struct request_queue *q, bool destroy_root);
	253	extern void update_root_blkg_pd(struct request_queue *q,
	254	enum blkio_policy_id plid);
3e252066	255
0381411e TH	256	/**
	257	* blkg_to_pdata - get policy private data
	258	* @blkg: blkg of interest
	259	* @pol: policy of interest
	260	*
	261	* Return pointer to private data associated with the @blkg-@pol pair.
	262	*/
	263	static inline void blkg_to_pdata(struct blkio_group blkg,
	264	struct blkio_policy_type *pol)
	265	{
549d3aa8	266	return blkg ? blkg->pd[pol->plid]->pdata : NULL;
0381411e TH	267	}
	268
	269	/**
	270	* pdata_to_blkg - get blkg associated with policy private data
	271	* @pdata: policy private data of interest
0381411e	272	*
aaec55a0	273	* @pdata is policy private data. Determine the blkg it's associated with.
0381411e	274	*/
aaec55a0	275	static inline struct blkio_group pdata_to_blkg(void pdata)
0381411e TH	276	{
	277	if (pdata) {
	278	struct blkg_policy_data *pd =
	279	container_of(pdata, struct blkg_policy_data, pdata);
	280	return pd->blkg;
	281	}
	282	return NULL;
	283	}
	284
afc24d49 VG	285	static inline char blkg_path(struct blkio_group blkg)
	286	{
	287	return blkg->path;
	288	}
	289
1adaf3dd TH	290	/**
	291	* blkg_get - get a blkg reference
	292	* @blkg: blkg to get
	293	*
	294	* The caller should be holding queue_lock and an existing reference.
	295	*/
	296	static inline void blkg_get(struct blkio_group *blkg)
	297	{
	298	lockdep_assert_held(blkg->q->queue_lock);
	299	WARN_ON_ONCE(!blkg->refcnt);
	300	blkg->refcnt++;
	301	}
	302
	303	void __blkg_release(struct blkio_group *blkg);
	304
	305	/**
	306	* blkg_put - put a blkg reference
	307	* @blkg: blkg to put
	308	*
	309	* The caller should be holding queue_lock.
	310	*/
	311	static inline void blkg_put(struct blkio_group *blkg)
	312	{
	313	lockdep_assert_held(blkg->q->queue_lock);
	314	WARN_ON_ONCE(blkg->refcnt <= 0);
	315	if (!--blkg->refcnt)
	316	__blkg_release(blkg);
	317	}
	318
2f5ea477 JA	319	#else
	320
	321	struct blkio_group {
	322	};
	323
3e252066 VG	324	struct blkio_policy_type {
	325	};
	326
5efd6113 TH	327	static inline int blkcg_init_queue(struct request_queue *q) { return 0; }
	328	static inline void blkcg_drain_queue(struct request_queue *q) { }
	329	static inline void blkcg_exit_queue(struct request_queue *q) { }
3e252066 VG	330	static inline void blkio_policy_register(struct blkio_policy_type *blkiop) { }
3e252066 VG	331	static inline void blkio_policy_unregister(struct blkio_policy_type *blkiop) { }
03aa264a	332	static inline void blkg_destroy_all(struct request_queue *q,
03aa264a	333	bool destory_root) { }
e8989fae TH	334	static inline void update_root_blkg_pd(struct request_queue *q,
e8989fae TH	335	enum blkio_policy_id plid) { }
3e252066	336
0381411e TH	337	static inline void blkg_to_pdata(struct blkio_group blkg,
	338	struct blkio_policy_type *pol) { return NULL; }
	339	static inline struct blkio_group pdata_to_blkg(void pdata,
	340	struct blkio_policy_type *pol) { return NULL; }
afc24d49	341	static inline char blkg_path(struct blkio_group blkg) { return NULL; }
1adaf3dd TH	342	static inline void blkg_get(struct blkio_group *blkg) { }
1adaf3dd TH	343	static inline void blkg_put(struct blkio_group *blkg) { }
afc24d49	344
2f5ea477 JA	345	#endif
2f5ea477 JA	346
df457f84	347	#define BLKIO_WEIGHT_MIN 10
31e4c28d VG	348	#define BLKIO_WEIGHT_MAX 1000
	349	#define BLKIO_WEIGHT_DEFAULT 500
	350
2868ef7b	351	#ifdef CONFIG_DEBUG_BLK_CGROUP
c1768268 TH	352	void blkiocg_update_avg_queue_size_stats(struct blkio_group *blkg,
c1768268 TH	353	struct blkio_policy_type *pol);
9195291e	354	void blkiocg_update_dequeue_stats(struct blkio_group *blkg,
c1768268 TH	355	struct blkio_policy_type *pol,
	356	unsigned long dequeue);
	357	void blkiocg_update_set_idle_time_stats(struct blkio_group *blkg,
	358	struct blkio_policy_type *pol);
	359	void blkiocg_update_idle_time_stats(struct blkio_group *blkg,
	360	struct blkio_policy_type *pol);
	361	void blkiocg_set_start_empty_time(struct blkio_group *blkg,
	362	struct blkio_policy_type *pol);
812df48d DS	363
	364	#define BLKG_FLAG_FNS(name) \
	365	static inline void blkio_mark_blkg_##name( \
	366	struct blkio_group_stats *stats) \
	367	{ \
	368	stats->flags \|= (1 << BLKG_##name); \
	369	} \
	370	static inline void blkio_clear_blkg_##name( \
	371	struct blkio_group_stats *stats) \
	372	{ \
	373	stats->flags &= ~(1 << BLKG_##name); \
	374	} \
	375	static inline int blkio_blkg_##name(struct blkio_group_stats *stats) \
	376	{ \
	377	return (stats->flags & (1 << BLKG_##name)) != 0; \
	378	} \
	379
	380	BLKG_FLAG_FNS(waiting)
	381	BLKG_FLAG_FNS(idling)
	382	BLKG_FLAG_FNS(empty)
	383	#undef BLKG_FLAG_FNS
2868ef7b	384	#else
c1768268 TH	385	static inline void blkiocg_update_avg_queue_size_stats(struct blkio_group *blkg,
c1768268 TH	386	struct blkio_policy_type *pol) { }
9195291e	387	static inline void blkiocg_update_dequeue_stats(struct blkio_group *blkg,
c1768268 TH	388	struct blkio_policy_type *pol, unsigned long dequeue) { }
	389	static inline void blkiocg_update_set_idle_time_stats(struct blkio_group *blkg,
	390	struct blkio_policy_type *pol) { }
	391	static inline void blkiocg_update_idle_time_stats(struct blkio_group *blkg,
	392	struct blkio_policy_type *pol) { }
	393	static inline void blkiocg_set_start_empty_time(struct blkio_group *blkg,
	394	struct blkio_policy_type *pol) { }
2868ef7b VG	395	#endif
2868ef7b VG	396
32e380ae	397	#ifdef CONFIG_BLK_CGROUP
31e4c28d VG	398	extern struct blkio_cgroup blkio_root_cgroup;
31e4c28d VG	399	extern struct blkio_cgroup cgroup_to_blkio_cgroup(struct cgroup cgroup);
4f85cb96	400	extern struct blkio_cgroup bio_blkio_cgroup(struct bio bio);
cd1604fa	401	extern struct blkio_group blkg_lookup(struct blkio_cgroup blkcg,
e8989fae	402	struct request_queue *q);
cd1604fa TH	403	struct blkio_group blkg_lookup_create(struct blkio_cgroup blkcg,
cd1604fa TH	404	struct request_queue *q,
cd1604fa	405	bool for_root);
303a3acb	406	void blkiocg_update_timeslice_used(struct blkio_group *blkg,
c1768268 TH	407	struct blkio_policy_type *pol,
	408	unsigned long time,
	409	unsigned long unaccounted_time);
	410	void blkiocg_update_dispatch_stats(struct blkio_group *blkg,
	411	struct blkio_policy_type *pol,
	412	uint64_t bytes, bool direction, bool sync);
84c124da	413	void blkiocg_update_completion_stats(struct blkio_group *blkg,
c1768268 TH	414	struct blkio_policy_type *pol,
	415	uint64_t start_time,
	416	uint64_t io_start_time, bool direction,
	417	bool sync);
	418	void blkiocg_update_io_merged_stats(struct blkio_group *blkg,
	419	struct blkio_policy_type *pol,
	420	bool direction, bool sync);
a11cdaa7	421	void blkiocg_update_io_add_stats(struct blkio_group *blkg,
c1768268 TH	422	struct blkio_policy_type *pol,
	423	struct blkio_group *curr_blkg, bool direction,
	424	bool sync);
a11cdaa7	425	void blkiocg_update_io_remove_stats(struct blkio_group *blkg,
c1768268 TH	426	struct blkio_policy_type *pol,
c1768268 TH	427	bool direction, bool sync);
31e4c28d	428	#else
2f5ea477	429	struct cgroup;
31e4c28d VG	430	static inline struct blkio_cgroup *
31e4c28d VG	431	cgroup_to_blkio_cgroup(struct cgroup *cgroup) { return NULL; }
70087dc3	432	static inline struct blkio_cgroup *
4f85cb96	433	bio_blkio_cgroup(struct bio *bio) { return NULL; }
31e4c28d	434
cd1604fa TH	435	static inline struct blkio_group blkg_lookup(struct blkio_cgroup blkcg,
cd1604fa TH	436	void *key) { return NULL; }
303a3acb	437	static inline void blkiocg_update_timeslice_used(struct blkio_group *blkg,
c1768268 TH	438	struct blkio_policy_type *pol, unsigned long time,
c1768268 TH	439	unsigned long unaccounted_time) { }
84c124da	440	static inline void blkiocg_update_dispatch_stats(struct blkio_group *blkg,
c1768268 TH	441	struct blkio_policy_type *pol, uint64_t bytes,
c1768268 TH	442	bool direction, bool sync) { }
84c124da	443	static inline void blkiocg_update_completion_stats(struct blkio_group *blkg,
c1768268 TH	444	struct blkio_policy_type *pol, uint64_t start_time,
c1768268 TH	445	uint64_t io_start_time, bool direction, bool sync) { }
812d4026	446	static inline void blkiocg_update_io_merged_stats(struct blkio_group *blkg,
c1768268 TH	447	struct blkio_policy_type *pol, bool direction,
c1768268 TH	448	bool sync) { }
a11cdaa7	449	static inline void blkiocg_update_io_add_stats(struct blkio_group *blkg,
c1768268 TH	450	struct blkio_policy_type *pol,
	451	struct blkio_group *curr_blkg, bool direction,
	452	bool sync) { }
a11cdaa7	453	static inline void blkiocg_update_io_remove_stats(struct blkio_group *blkg,
c1768268 TH	454	struct blkio_policy_type *pol, bool direction,
c1768268 TH	455	bool sync) { }
31e4c28d VG	456	#endif
31e4c28d VG	457	#endif /* _BLK_CGROUP_H */