[deliverable/linux.git] / include / linux / ceph / osdmap.h

#ifndef _FS_CEPH_OSDMAP_H
#define _FS_CEPH_OSDMAP_H

#include <linux/rbtree.h>
#include <linux/ceph/types.h>
#include <linux/ceph/decode.h>
#include <linux/ceph/ceph_fs.h>
#include <linux/crush/crush.h>

/*
 * The osd map describes the current membership of the osd cluster and
 * specifies the mapping of objects to placement groups and placement
 * groups to (sets of) osds.  That is, it completely specifies the
 * (desired) distribution of all data objects in the system at some
 * point in time.
 *
 * Each map version is identified by an epoch, which increases monotonically.
 *
 * The map can be updated either via an incremental map (diff) describing
 * the change between two successive epochs, or as a fully encoded map.
 */
struct ceph_pg {
	uint64_t pool;
	uint32_t seed;
};

#define CEPH_POOL_FLAG_HASHPSPOOL  1

struct ceph_pg_pool_info {
	struct rb_node node;
	s64 id;
	u8 type;
	u8 size;
	u8 crush_ruleset;
	u8 object_hash;
	u32 pg_num, pgp_num;
	int pg_num_mask, pgp_num_mask;
	s64 read_tier;
	s64 write_tier; /* wins for read+write ops */
	u64 flags;
	char *name;
};

struct ceph_object_locator {
	s64 pool;
};

/*
 * Maximum supported by kernel client object name length
 *
 * (probably outdated: must be >= RBD_MAX_MD_NAME_LEN -- currently 100)
 */
#define CEPH_MAX_OID_NAME_LEN 100

struct ceph_object_id {
	char name[CEPH_MAX_OID_NAME_LEN];
	int name_len;
};

struct ceph_pg_mapping {
	struct rb_node node;
	struct ceph_pg pgid;
	int len;
	int osds[];
};

struct ceph_osdmap {
	struct ceph_fsid fsid;
	u32 epoch;
	u32 mkfs_epoch;
	struct ceph_timespec created, modified;

	u32 flags;         /* CEPH_OSDMAP_* */

	u32 max_osd;       /* size of osd_state, _offload, _addr arrays */
	u8 *osd_state;     /* CEPH_OSD_* */
	u32 *osd_weight;   /* 0 = failed, 0x10000 = 100% normal */
	struct ceph_entity_addr *osd_addr;

	struct rb_root pg_temp;
	struct rb_root pg_pools;
	u32 pool_max;

	/* the CRUSH map specifies the mapping of placement groups to
	 * the list of osds that store+replicate them. */
	struct crush_map *crush;

	struct mutex crush_scratch_mutex;
	int crush_scratch_ary[CEPH_PG_MAX_SIZE * 3];
};

static inline void ceph_oid_set_name(struct ceph_object_id *oid,
				     const char *name)
{
	int len;

	len = strlen(name);
	if (len > sizeof(oid->name)) {
		WARN(1, "ceph_oid_set_name '%s' len %d vs %zu, truncating\n",
		     name, len, sizeof(oid->name));
		len = sizeof(oid->name);
	}

	memcpy(oid->name, name, len);
	oid->name_len = len;
}

static inline void ceph_oid_copy(struct ceph_object_id *dest,
				 struct ceph_object_id *src)
{
	BUG_ON(src->name_len > sizeof(dest->name));
	memcpy(dest->name, src->name, src->name_len);
	dest->name_len = src->name_len;
}

static inline int ceph_osd_is_up(struct ceph_osdmap *map, int osd)
{
	return (osd < map->max_osd) && (map->osd_state[osd] & CEPH_OSD_UP);
}

static inline bool ceph_osdmap_flag(struct ceph_osdmap *map, int flag)
{
	return map && (map->flags & flag);
}

extern char *ceph_osdmap_state_str(char *str, int len, int state);

static inline struct ceph_entity_addr *ceph_osd_addr(struct ceph_osdmap *map,
						     int osd)
{
	if (osd >= map->max_osd)
		return NULL;
	return &map->osd_addr[osd];
}

static inline int ceph_decode_pgid(void **p, void *end, struct ceph_pg *pgid)
{
	__u8 version;

	if (!ceph_has_room(p, end, 1 + 8 + 4 + 4)) {
		pr_warning("incomplete pg encoding");

		return -EINVAL;
	}
	version = ceph_decode_8(p);
	if (version > 1) {
		pr_warning("do not understand pg encoding %d > 1",
			(int)version);
		return -EINVAL;
	}

	pgid->pool = ceph_decode_64(p);
	pgid->seed = ceph_decode_32(p);
	*p += 4;	/* skip deprecated preferred value */

	return 0;
}

extern struct ceph_osdmap *ceph_osdmap_decode(void **p, void *end);
extern struct ceph_osdmap *osdmap_apply_incremental(void **p, void *end,
					    struct ceph_osdmap *map,
					    struct ceph_messenger *msgr);
extern void ceph_osdmap_destroy(struct ceph_osdmap *map);

/* calculate mapping of a file extent to an object */
extern int ceph_calc_file_object_mapping(struct ceph_file_layout *layout,
					 u64 off, u64 len,
					 u64 *bno, u64 *oxoff, u64 *oxlen);

/* calculate mapping of object to a placement group */
extern int ceph_oloc_oid_to_pg(struct ceph_osdmap *osdmap,
			       struct ceph_object_locator *oloc,
			       struct ceph_object_id *oid,
			       struct ceph_pg *pg_out);

extern int ceph_calc_pg_acting(struct ceph_osdmap *osdmap,
			       struct ceph_pg pgid,
			       int *acting);
extern int ceph_calc_pg_primary(struct ceph_osdmap *osdmap,
				struct ceph_pg pgid);

extern struct ceph_pg_pool_info *ceph_pg_pool_by_id(struct ceph_osdmap *map,
						    u64 id);

extern const char *ceph_pg_pool_name_by_id(struct ceph_osdmap *map, u64 id);
extern int ceph_pg_poolid_by_name(struct ceph_osdmap *map, const char *name);

#endif
Commit	Line	Data
f24e9980 SW	1	#ifndef _FS_CEPH_OSDMAP_H
	2	#define _FS_CEPH_OSDMAP_H
	3
	4	#include <linux/rbtree.h>
a1ce3928	5	#include <linux/ceph/types.h>
ef4859d6	6	#include <linux/ceph/decode.h>
a1ce3928	7	#include <linux/ceph/ceph_fs.h>
3d14c5d2	8	#include <linux/crush/crush.h>
f24e9980 SW	9
	10	/*
	11	* The osd map describes the current membership of the osd cluster and
	12	* specifies the mapping of objects to placement groups and placement
	13	* groups to (sets of) osds. That is, it completely specifies the
	14	* (desired) distribution of all data objects in the system at some
	15	* point in time.
	16	*
	17	* Each map version is identified by an epoch, which increases monotonically.
	18	*
	19	* The map can be updated either via an incremental map (diff) describing
	20	* the change between two successive epochs, or as a fully encoded map.
	21	*/
5b191d99 SW	22	struct ceph_pg {
	23	uint64_t pool;
	24	uint32_t seed;
	25	};
	26
83ca14fd SW	27	#define CEPH_POOL_FLAG_HASHPSPOOL 1
83ca14fd SW	28
f24e9980	29	struct ceph_pg_pool_info {
4fc51be8	30	struct rb_node node;
4f6a7e5e SW	31	s64 id;
	32	u8 type;
	33	u8 size;
	34	u8 crush_ruleset;
	35	u8 object_hash;
	36	u32 pg_num, pgp_num;
	37	int pg_num_mask, pgp_num_mask;
17a13e40 ID	38	s64 read_tier;
17a13e40 ID	39	s64 write_tier; /* wins for read+write ops */
4f6a7e5e	40	u64 flags;
2844a76a	41	char *name;
f24e9980 SW	42	};
f24e9980 SW	43
4f6a7e5e	44	struct ceph_object_locator {
22116525	45	s64 pool;
4f6a7e5e SW	46	};
4f6a7e5e SW	47
4295f221 ID	48	/*
	49	* Maximum supported by kernel client object name length
	50	*
	51	* (probably outdated: must be >= RBD_MAX_MD_NAME_LEN -- currently 100)
	52	*/
	53	#define CEPH_MAX_OID_NAME_LEN 100
	54
	55	struct ceph_object_id {
	56	char name[CEPH_MAX_OID_NAME_LEN];
	57	int name_len;
	58	};
	59
f24e9980 SW	60	struct ceph_pg_mapping {
f24e9980 SW	61	struct rb_node node;
5b191d99	62	struct ceph_pg pgid;
f24e9980 SW	63	int len;
	64	int osds[];
	65	};
	66
	67	struct ceph_osdmap {
	68	struct ceph_fsid fsid;
	69	u32 epoch;
	70	u32 mkfs_epoch;
	71	struct ceph_timespec created, modified;
	72
	73	u32 flags; /* CEPH_OSDMAP_* */
	74
	75	u32 max_osd; /* size of osd_state, _offload, _addr arrays */
	76	u8 osd_state; / CEPH_OSD_* */
	77	u32 osd_weight; / 0 = failed, 0x10000 = 100% normal */
	78	struct ceph_entity_addr *osd_addr;
	79
	80	struct rb_root pg_temp;
4fc51be8 SW	81	struct rb_root pg_pools;
4fc51be8 SW	82	u32 pool_max;
f24e9980 SW	83
	84	/* the CRUSH map specifies the mapping of placement groups to
	85	* the list of osds that store+replicate them. */
	86	struct crush_map *crush;
9d521470 ID	87
	88	struct mutex crush_scratch_mutex;
	89	int crush_scratch_ary[CEPH_PG_MAX_SIZE * 3];
f24e9980 SW	90	};
f24e9980 SW	91
4295f221 ID	92	static inline void ceph_oid_set_name(struct ceph_object_id *oid,
	93	const char *name)
	94	{
	95	int len;
	96
	97	len = strlen(name);
	98	if (len > sizeof(oid->name)) {
	99	WARN(1, "ceph_oid_set_name '%s' len %d vs %zu, truncating\n",
	100	name, len, sizeof(oid->name));
	101	len = sizeof(oid->name);
	102	}
	103
	104	memcpy(oid->name, name, len);
	105	oid->name_len = len;
	106	}
	107
	108	static inline void ceph_oid_copy(struct ceph_object_id *dest,
	109	struct ceph_object_id *src)
	110	{
	111	BUG_ON(src->name_len > sizeof(dest->name));
	112	memcpy(dest->name, src->name, src->name_len);
	113	dest->name_len = src->name_len;
	114	}
	115
f24e9980 SW	116	static inline int ceph_osd_is_up(struct ceph_osdmap *map, int osd)
	117	{
	118	return (osd < map->max_osd) && (map->osd_state[osd] & CEPH_OSD_UP);
	119	}
	120
	121	static inline bool ceph_osdmap_flag(struct ceph_osdmap *map, int flag)
	122	{
	123	return map && (map->flags & flag);
	124	}
	125
	126	extern char ceph_osdmap_state_str(char str, int len, int state);
	127
	128	static inline struct ceph_entity_addr ceph_osd_addr(struct ceph_osdmap map,
	129	int osd)
	130	{
	131	if (osd >= map->max_osd)
	132	return NULL;
	133	return &map->osd_addr[osd];
	134	}
	135
ef4859d6 AE	136	static inline int ceph_decode_pgid(void *p, void end, struct ceph_pg *pgid)
	137	{
	138	__u8 version;
	139
	140	if (!ceph_has_room(p, end, 1 + 8 + 4 + 4)) {
	141	pr_warning("incomplete pg encoding");
	142
	143	return -EINVAL;
	144	}
	145	version = ceph_decode_8(p);
	146	if (version > 1) {
	147	pr_warning("do not understand pg encoding %d > 1",
	148	(int)version);
	149	return -EINVAL;
	150	}
	151
	152	pgid->pool = ceph_decode_64(p);
	153	pgid->seed = ceph_decode_32(p);
	154	p += 4; / skip deprecated preferred value */
	155
	156	return 0;
	157	}
	158
a2505d63	159	extern struct ceph_osdmap ceph_osdmap_decode(void p, void end);
f24e9980 SW	160	extern struct ceph_osdmap osdmap_apply_incremental(void p, void end,
	161	struct ceph_osdmap *map,
	162	struct ceph_messenger *msgr);
	163	extern void ceph_osdmap_destroy(struct ceph_osdmap *map);
	164
	165	/* calculate mapping of a file extent to an object */
d63b77f4	166	extern int ceph_calc_file_object_mapping(struct ceph_file_layout *layout,
e8afad65	167	u64 off, u64 len,
d63b77f4	168	u64 bno, u64 oxoff, u64 *oxlen);
f24e9980 SW	169
f24e9980 SW	170	/* calculate mapping of object to a placement group */
7c13cb64 ID	171	extern int ceph_oloc_oid_to_pg(struct ceph_osdmap *osdmap,
	172	struct ceph_object_locator *oloc,
	173	struct ceph_object_id *oid,
	174	struct ceph_pg *pg_out);
	175
12979354	176	extern int ceph_calc_pg_acting(struct ceph_osdmap *osdmap,
5b191d99	177	struct ceph_pg pgid,
d85b7056	178	int *acting);
51042122	179	extern int ceph_calc_pg_primary(struct ceph_osdmap *osdmap,
5b191d99	180	struct ceph_pg pgid);
f24e9980	181
ce7f6a27 ID	182	extern struct ceph_pg_pool_info ceph_pg_pool_by_id(struct ceph_osdmap map,
	183	u64 id);
	184
72afc71f	185	extern const char ceph_pg_pool_name_by_id(struct ceph_osdmap map, u64 id);
7669a2c9 YS	186	extern int ceph_pg_poolid_by_name(struct ceph_osdmap map, const char name);
7669a2c9 YS	187
f24e9980	188	#endif