[deliverable/linux.git] / include / linux / backing-dev-defs.h

#ifndef __LINUX_BACKING_DEV_DEFS_H
#define __LINUX_BACKING_DEV_DEFS_H

#include <linux/list.h>
#include <linux/radix-tree.h>
#include <linux/rbtree.h>
#include <linux/spinlock.h>
#include <linux/percpu_counter.h>
#include <linux/percpu-refcount.h>
#include <linux/flex_proportions.h>
#include <linux/timer.h>
#include <linux/workqueue.h>

struct page;
struct device;
struct dentry;

/*
 * Bits in bdi_writeback.state
 */
enum wb_state {
	WB_registered,		/* bdi_register() was done */
	WB_writeback_running,	/* Writeback is in progress */
	WB_has_dirty_io,	/* Dirty inodes on ->b_{dirty|io|more_io} */
};

enum wb_congested_state {
	WB_async_congested,	/* The async (write) queue is getting full */
	WB_sync_congested,	/* The sync queue is getting full */
};

typedef int (congested_fn)(void *, int);

enum wb_stat_item {
	WB_RECLAIMABLE,
	WB_WRITEBACK,
	WB_DIRTIED,
	WB_WRITTEN,
	NR_WB_STAT_ITEMS
};

#define WB_STAT_BATCH (8*(1+ilog2(nr_cpu_ids)))

/*
 * For cgroup writeback, multiple wb's may map to the same blkcg.  Those
 * wb's can operate mostly independently but should share the congested
 * state.  To facilitate such sharing, the congested state is tracked using
 * the following struct which is created on demand, indexed by blkcg ID on
 * its bdi, and refcounted.
 */
struct bdi_writeback_congested {
	unsigned long state;		/* WB_[a]sync_congested flags */

#ifdef CONFIG_CGROUP_WRITEBACK
	struct backing_dev_info *bdi;	/* the associated bdi */
	atomic_t refcnt;		/* nr of attached wb's and blkg */
	int blkcg_id;			/* ID of the associated blkcg */
	struct rb_node rb_node;		/* on bdi->cgwb_congestion_tree */
#endif
};

/*
 * Each wb (bdi_writeback) can perform writeback operations, is measured
 * and throttled, independently.  Without cgroup writeback, each bdi
 * (bdi_writeback) is served by its embedded bdi->wb.
 *
 * On the default hierarchy, blkcg implicitly enables memcg.  This allows
 * using memcg's page ownership for attributing writeback IOs, and every
 * memcg - blkcg combination can be served by its own wb by assigning a
 * dedicated wb to each memcg, which enables isolation across different
 * cgroups and propagation of IO back pressure down from the IO layer upto
 * the tasks which are generating the dirty pages to be written back.
 *
 * A cgroup wb is indexed on its bdi by the ID of the associated memcg,
 * refcounted with the number of inodes attached to it, and pins the memcg
 * and the corresponding blkcg.  As the corresponding blkcg for a memcg may
 * change as blkcg is disabled and enabled higher up in the hierarchy, a wb
 * is tested for blkcg after lookup and removed from index on mismatch so
 * that a new wb for the combination can be created.
 */
struct bdi_writeback {
	struct backing_dev_info *bdi;	/* our parent bdi */

	unsigned long state;		/* Always use atomic bitops on this */
	unsigned long last_old_flush;	/* last old data flush */

	struct list_head b_dirty;	/* dirty inodes */
	struct list_head b_io;		/* parked for writeback */
	struct list_head b_more_io;	/* parked for more writeback */
	struct list_head b_dirty_time;	/* time stamps are dirty */
	spinlock_t list_lock;		/* protects the b_* lists */

	struct percpu_counter stat[NR_WB_STAT_ITEMS];

	struct bdi_writeback_congested *congested;

	unsigned long bw_time_stamp;	/* last time write bw is updated */
	unsigned long dirtied_stamp;
	unsigned long written_stamp;	/* pages written at bw_time_stamp */
	unsigned long write_bandwidth;	/* the estimated write bandwidth */
	unsigned long avg_write_bandwidth; /* further smoothed write bw, > 0 */

	/*
	 * The base dirty throttle rate, re-calculated on every 200ms.
	 * All the bdi tasks' dirty rate will be curbed under it.
	 * @dirty_ratelimit tracks the estimated @balanced_dirty_ratelimit
	 * in small steps and is much more smooth/stable than the latter.
	 */
	unsigned long dirty_ratelimit;
	unsigned long balanced_dirty_ratelimit;

	struct fprop_local_percpu completions;
	int dirty_exceeded;

	spinlock_t work_lock;		/* protects work_list & dwork scheduling */
	struct list_head work_list;
	struct delayed_work dwork;	/* work item used for writeback */

#ifdef CONFIG_CGROUP_WRITEBACK
	struct percpu_ref refcnt;	/* used only for !root wb's */
	struct fprop_local_percpu memcg_completions;
	struct cgroup_subsys_state *memcg_css; /* the associated memcg */
	struct cgroup_subsys_state *blkcg_css; /* and blkcg */
	struct list_head memcg_node;	/* anchored at memcg->cgwb_list */
	struct list_head blkcg_node;	/* anchored at blkcg->cgwb_list */

	union {
		struct work_struct release_work;
		struct rcu_head rcu;
	};
#endif
};

struct backing_dev_info {
	struct list_head bdi_list;
	unsigned long ra_pages;	/* max readahead in PAGE_CACHE_SIZE units */
	unsigned int capabilities; /* Device capabilities */
	congested_fn *congested_fn; /* Function pointer if device is md/dm */
	void *congested_data;	/* Pointer to aux data for congested func */

	char *name;

	unsigned int min_ratio;
	unsigned int max_ratio, max_prop_frac;

	/*
	 * Sum of avg_write_bw of wbs with dirty inodes.  > 0 if there are
	 * any dirty wbs, which is depended upon by bdi_has_dirty().
	 */
	atomic_long_t tot_write_bandwidth;

	struct bdi_writeback wb;  /* the root writeback info for this bdi */
	struct bdi_writeback_congested wb_congested; /* its congested state */
#ifdef CONFIG_CGROUP_WRITEBACK
	struct radix_tree_root cgwb_tree; /* radix tree of active cgroup wbs */
	struct rb_root cgwb_congested_tree; /* their congested states */
	atomic_t usage_cnt; /* counts both cgwbs and cgwb_contested's */
#endif
	wait_queue_head_t wb_waitq;

	struct device *dev;

	struct timer_list laptop_mode_wb_timer;

#ifdef CONFIG_DEBUG_FS
	struct dentry *debug_dir;
	struct dentry *debug_stats;
#endif
};

enum {
	BLK_RW_ASYNC	= 0,
	BLK_RW_SYNC	= 1,
};

void clear_wb_congested(struct bdi_writeback_congested *congested, int sync);
void set_wb_congested(struct bdi_writeback_congested *congested, int sync);

static inline void clear_bdi_congested(struct backing_dev_info *bdi, int sync)
{
	clear_wb_congested(bdi->wb.congested, sync);
}

static inline void set_bdi_congested(struct backing_dev_info *bdi, int sync)
{
	set_wb_congested(bdi->wb.congested, sync);
}

#ifdef CONFIG_CGROUP_WRITEBACK

/**
 * wb_tryget - try to increment a wb's refcount
 * @wb: bdi_writeback to get
 */
static inline bool wb_tryget(struct bdi_writeback *wb)
{
	if (wb != &wb->bdi->wb)
		return percpu_ref_tryget(&wb->refcnt);
	return true;
}

/**
 * wb_get - increment a wb's refcount
 * @wb: bdi_writeback to get
 */
static inline void wb_get(struct bdi_writeback *wb)
{
	if (wb != &wb->bdi->wb)
		percpu_ref_get(&wb->refcnt);
}

/**
 * wb_put - decrement a wb's refcount
 * @wb: bdi_writeback to put
 */
static inline void wb_put(struct bdi_writeback *wb)
{
	if (wb != &wb->bdi->wb)
		percpu_ref_put(&wb->refcnt);
}

/**
 * wb_dying - is a wb dying?
 * @wb: bdi_writeback of interest
 *
 * Returns whether @wb is unlinked and being drained.
 */
static inline bool wb_dying(struct bdi_writeback *wb)
{
	return percpu_ref_is_dying(&wb->refcnt);
}

#else	/* CONFIG_CGROUP_WRITEBACK */

static inline bool wb_tryget(struct bdi_writeback *wb)
{
	return true;
}

static inline void wb_get(struct bdi_writeback *wb)
{
}

static inline void wb_put(struct bdi_writeback *wb)
{
}

static inline bool wb_dying(struct bdi_writeback *wb)
{
	return false;
}

#endif	/* CONFIG_CGROUP_WRITEBACK */

#endif	/* __LINUX_BACKING_DEV_DEFS_H */
Commit	Line	Data
66114cad TH	1	#ifndef __LINUX_BACKING_DEV_DEFS_H
	2	#define __LINUX_BACKING_DEV_DEFS_H
	3
	4	#include <linux/list.h>
52ebea74 TH	5	#include <linux/radix-tree.h>
52ebea74 TH	6	#include <linux/rbtree.h>
66114cad TH	7	#include <linux/spinlock.h>
66114cad TH	8	#include <linux/percpu_counter.h>
52ebea74	9	#include <linux/percpu-refcount.h>
66114cad TH	10	#include <linux/flex_proportions.h>
	11	#include <linux/timer.h>
	12	#include <linux/workqueue.h>
	13
	14	struct page;
	15	struct device;
	16	struct dentry;
	17
	18	/*
	19	* Bits in bdi_writeback.state
	20	*/
	21	enum wb_state {
66114cad TH	22	WB_registered, /* bdi_register() was done */
66114cad TH	23	WB_writeback_running, /* Writeback is in progress */
d6c10f1f	24	WB_has_dirty_io, /* Dirty inodes on ->b_{dirty\|io\|more_io} */
66114cad TH	25	};
66114cad TH	26
4aa9c692 TH	27	enum wb_congested_state {
	28	WB_async_congested, /* The async (write) queue is getting full */
	29	WB_sync_congested, /* The sync queue is getting full */
	30	};
	31
66114cad TH	32	typedef int (congested_fn)(void *, int);
	33
	34	enum wb_stat_item {
	35	WB_RECLAIMABLE,
	36	WB_WRITEBACK,
	37	WB_DIRTIED,
	38	WB_WRITTEN,
	39	NR_WB_STAT_ITEMS
	40	};
	41
	42	#define WB_STAT_BATCH (8*(1+ilog2(nr_cpu_ids)))
	43
52ebea74 TH	44	/*
	45	* For cgroup writeback, multiple wb's may map to the same blkcg. Those
	46	* wb's can operate mostly independently but should share the congested
	47	* state. To facilitate such sharing, the congested state is tracked using
	48	* the following struct which is created on demand, indexed by blkcg ID on
	49	* its bdi, and refcounted.
	50	*/
4aa9c692 TH	51	struct bdi_writeback_congested {
4aa9c692 TH	52	unsigned long state; /* WB_[a]sync_congested flags */
52ebea74 TH	53
	54	#ifdef CONFIG_CGROUP_WRITEBACK
	55	struct backing_dev_info bdi; / the associated bdi */
	56	atomic_t refcnt; /* nr of attached wb's and blkg */
	57	int blkcg_id; /* ID of the associated blkcg */
	58	struct rb_node rb_node; /* on bdi->cgwb_congestion_tree */
	59	#endif
4aa9c692 TH	60	};
4aa9c692 TH	61
52ebea74 TH	62	/*
	63	* Each wb (bdi_writeback) can perform writeback operations, is measured
	64	* and throttled, independently. Without cgroup writeback, each bdi
	65	* (bdi_writeback) is served by its embedded bdi->wb.
	66	*
	67	* On the default hierarchy, blkcg implicitly enables memcg. This allows
	68	* using memcg's page ownership for attributing writeback IOs, and every
	69	* memcg - blkcg combination can be served by its own wb by assigning a
	70	* dedicated wb to each memcg, which enables isolation across different
	71	* cgroups and propagation of IO back pressure down from the IO layer upto
	72	* the tasks which are generating the dirty pages to be written back.
	73	*
	74	* A cgroup wb is indexed on its bdi by the ID of the associated memcg,
	75	* refcounted with the number of inodes attached to it, and pins the memcg
	76	* and the corresponding blkcg. As the corresponding blkcg for a memcg may
	77	* change as blkcg is disabled and enabled higher up in the hierarchy, a wb
	78	* is tested for blkcg after lookup and removed from index on mismatch so
	79	* that a new wb for the combination can be created.
	80	*/
66114cad TH	81	struct bdi_writeback {
	82	struct backing_dev_info bdi; / our parent bdi */
	83
	84	unsigned long state; /* Always use atomic bitops on this */
	85	unsigned long last_old_flush; /* last old data flush */
	86
	87	struct list_head b_dirty; /* dirty inodes */
	88	struct list_head b_io; /* parked for writeback */
	89	struct list_head b_more_io; /* parked for more writeback */
	90	struct list_head b_dirty_time; /* time stamps are dirty */
	91	spinlock_t list_lock; /* protects the b_* lists */
	92
	93	struct percpu_counter stat[NR_WB_STAT_ITEMS];
	94
4aa9c692 TH	95	struct bdi_writeback_congested *congested;
4aa9c692 TH	96
66114cad TH	97	unsigned long bw_time_stamp; /* last time write bw is updated */
	98	unsigned long dirtied_stamp;
	99	unsigned long written_stamp; /* pages written at bw_time_stamp */
	100	unsigned long write_bandwidth; /* the estimated write bandwidth */
95a46c65	101	unsigned long avg_write_bandwidth; /* further smoothed write bw, > 0 */
66114cad TH	102
	103	/*
	104	* The base dirty throttle rate, re-calculated on every 200ms.
	105	* All the bdi tasks' dirty rate will be curbed under it.
	106	* @dirty_ratelimit tracks the estimated @balanced_dirty_ratelimit
	107	* in small steps and is much more smooth/stable than the latter.
	108	*/
	109	unsigned long dirty_ratelimit;
	110	unsigned long balanced_dirty_ratelimit;
	111
	112	struct fprop_local_percpu completions;
	113	int dirty_exceeded;
	114
	115	spinlock_t work_lock; /* protects work_list & dwork scheduling */
	116	struct list_head work_list;
	117	struct delayed_work dwork; /* work item used for writeback */
52ebea74 TH	118
	119	#ifdef CONFIG_CGROUP_WRITEBACK
	120	struct percpu_ref refcnt; /* used only for !root wb's */
841710aa	121	struct fprop_local_percpu memcg_completions;
52ebea74 TH	122	struct cgroup_subsys_state memcg_css; / the associated memcg */
	123	struct cgroup_subsys_state blkcg_css; / and blkcg */
	124	struct list_head memcg_node; /* anchored at memcg->cgwb_list */
	125	struct list_head blkcg_node; /* anchored at blkcg->cgwb_list */
	126
	127	union {
	128	struct work_struct release_work;
	129	struct rcu_head rcu;
	130	};
	131	#endif
66114cad TH	132	};
	133
	134	struct backing_dev_info {
	135	struct list_head bdi_list;
	136	unsigned long ra_pages; /* max readahead in PAGE_CACHE_SIZE units */
	137	unsigned int capabilities; /* Device capabilities */
	138	congested_fn congested_fn; / Function pointer if device is md/dm */
	139	void congested_data; / Pointer to aux data for congested func */
	140
	141	char *name;
	142
	143	unsigned int min_ratio;
	144	unsigned int max_ratio, max_prop_frac;
	145
95a46c65 TH	146	/*
	147	* Sum of avg_write_bw of wbs with dirty inodes. > 0 if there are
	148	* any dirty wbs, which is depended upon by bdi_has_dirty().
	149	*/
	150	atomic_long_t tot_write_bandwidth;
766a9d6e	151
52ebea74 TH	152	struct bdi_writeback wb; /* the root writeback info for this bdi */
	153	struct bdi_writeback_congested wb_congested; /* its congested state */
	154	#ifdef CONFIG_CGROUP_WRITEBACK
	155	struct radix_tree_root cgwb_tree; /* radix tree of active cgroup wbs */
	156	struct rb_root cgwb_congested_tree; /* their congested states */
	157	atomic_t usage_cnt; /* counts both cgwbs and cgwb_contested's */
	158	#endif
cc395d7f TH	159	wait_queue_head_t wb_waitq;
cc395d7f TH	160
66114cad TH	161	struct device *dev;
	162
	163	struct timer_list laptop_mode_wb_timer;
	164
	165	#ifdef CONFIG_DEBUG_FS
	166	struct dentry *debug_dir;
	167	struct dentry *debug_stats;
	168	#endif
	169	};
	170
	171	enum {
	172	BLK_RW_ASYNC = 0,
	173	BLK_RW_SYNC = 1,
	174	};
	175
ec8a6f26 TH	176	void clear_wb_congested(struct bdi_writeback_congested *congested, int sync);
	177	void set_wb_congested(struct bdi_writeback_congested *congested, int sync);
	178
	179	static inline void clear_bdi_congested(struct backing_dev_info *bdi, int sync)
	180	{
	181	clear_wb_congested(bdi->wb.congested, sync);
	182	}
	183
	184	static inline void set_bdi_congested(struct backing_dev_info *bdi, int sync)
	185	{
	186	set_wb_congested(bdi->wb.congested, sync);
	187	}
66114cad	188
21c6321f TH	189	#ifdef CONFIG_CGROUP_WRITEBACK
	190
	191	/**
	192	* wb_tryget - try to increment a wb's refcount
	193	* @wb: bdi_writeback to get
	194	*/
	195	static inline bool wb_tryget(struct bdi_writeback *wb)
	196	{
	197	if (wb != &wb->bdi->wb)
	198	return percpu_ref_tryget(&wb->refcnt);
	199	return true;
	200	}
	201
	202	/**
	203	* wb_get - increment a wb's refcount
	204	* @wb: bdi_writeback to get
	205	*/
	206	static inline void wb_get(struct bdi_writeback *wb)
	207	{
	208	if (wb != &wb->bdi->wb)
	209	percpu_ref_get(&wb->refcnt);
	210	}
	211
	212	/**
	213	* wb_put - decrement a wb's refcount
	214	* @wb: bdi_writeback to put
	215	*/
	216	static inline void wb_put(struct bdi_writeback *wb)
	217	{
	218	if (wb != &wb->bdi->wb)
	219	percpu_ref_put(&wb->refcnt);
	220	}
	221
e8a7abf5 TH	222	/**
	223	* wb_dying - is a wb dying?
	224	* @wb: bdi_writeback of interest
	225	*
	226	* Returns whether @wb is unlinked and being drained.
	227	*/
	228	static inline bool wb_dying(struct bdi_writeback *wb)
	229	{
	230	return percpu_ref_is_dying(&wb->refcnt);
	231	}
	232
21c6321f TH	233	#else /* CONFIG_CGROUP_WRITEBACK */
	234
	235	static inline bool wb_tryget(struct bdi_writeback *wb)
	236	{
	237	return true;
	238	}
	239
	240	static inline void wb_get(struct bdi_writeback *wb)
	241	{
	242	}
	243
	244	static inline void wb_put(struct bdi_writeback *wb)
	245	{
	246	}
	247
e8a7abf5 TH	248	static inline bool wb_dying(struct bdi_writeback *wb)
	249	{
	250	return false;
	251	}
	252
21c6321f TH	253	#endif /* CONFIG_CGROUP_WRITEBACK */
21c6321f TH	254
66114cad	255	#endif /* __LINUX_BACKING_DEV_DEFS_H */