[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;
};

static inline bool ceph_can_shift_osds(struct ceph_pg_pool_info *pool)
{
	switch (pool->type) {
	case CEPH_POOL_TYPE_REP:
		return true;
	case CEPH_POOL_TYPE_EC:
		return false;
	default:
		BUG_ON(1);
	}
}

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 * (probably outdated: must be 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;

	union {
		struct {
			int len;
			int osds[];
		} pg_temp;
		struct {
			int osd;
		} primary_temp;
	};
};

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 primary_temp;

	u32 *osd_primary_affinity;

	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_exists(struct ceph_osdmap *map, int osd)
{
	return osd >= 0 && osd < map->max_osd &&
	       (map->osd_state[osd] & CEPH_OSD_EXISTS);
}

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

static inline int ceph_osd_is_down(struct ceph_osdmap *map, int osd)
{
	return !ceph_osd_is_up(map, osd);
}

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);
extern u32 ceph_get_primary_affinity(struct ceph_osdmap *map, int osd);

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_warn("incomplete pg encoding\n");
		return -EINVAL;
	}
	version = ceph_decode_8(p);
	if (version > 1) {
		pr_warn("do not understand pg encoding %d > 1\n",
			(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 *osds, int *primary);
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
2abebdbc ID	44	static inline bool ceph_can_shift_osds(struct ceph_pg_pool_info *pool)
	45	{
	46	switch (pool->type) {
	47	case CEPH_POOL_TYPE_REP:
	48	return true;
	49	case CEPH_POOL_TYPE_EC:
	50	return false;
	51	default:
	52	BUG_ON(1);
	53	}
	54	}
	55
4f6a7e5e	56	struct ceph_object_locator {
22116525	57	s64 pool;
4f6a7e5e SW	58	};
4f6a7e5e SW	59
4295f221 ID	60	/*
	61	* Maximum supported by kernel client object name length
	62	*
	63	* (probably outdated: must be >= RBD_MAX_MD_NAME_LEN -- currently 100)
	64	*/
	65	#define CEPH_MAX_OID_NAME_LEN 100
	66
	67	struct ceph_object_id {
	68	char name[CEPH_MAX_OID_NAME_LEN];
	69	int name_len;
	70	};
	71
f24e9980 SW	72	struct ceph_pg_mapping {
f24e9980 SW	73	struct rb_node node;
5b191d99	74	struct ceph_pg pgid;
35a935d7 ID	75
	76	union {
	77	struct {
	78	int len;
	79	int osds[];
	80	} pg_temp;
9686f94c ID	81	struct {
	82	int osd;
	83	} primary_temp;
35a935d7	84	};
f24e9980 SW	85	};
	86
	87	struct ceph_osdmap {
	88	struct ceph_fsid fsid;
	89	u32 epoch;
	90	u32 mkfs_epoch;
	91	struct ceph_timespec created, modified;
	92
	93	u32 flags; /* CEPH_OSDMAP_* */
	94
	95	u32 max_osd; /* size of osd_state, _offload, _addr arrays */
	96	u8 osd_state; / CEPH_OSD_* */
	97	u32 osd_weight; / 0 = failed, 0x10000 = 100% normal */
	98	struct ceph_entity_addr *osd_addr;
	99
	100	struct rb_root pg_temp;
9686f94c ID	101	struct rb_root primary_temp;
9686f94c ID	102
2cfa34f2 ID	103	u32 *osd_primary_affinity;
2cfa34f2 ID	104
4fc51be8 SW	105	struct rb_root pg_pools;
4fc51be8 SW	106	u32 pool_max;
f24e9980 SW	107
	108	/* the CRUSH map specifies the mapping of placement groups to
	109	* the list of osds that store+replicate them. */
	110	struct crush_map *crush;
9d521470 ID	111
	112	struct mutex crush_scratch_mutex;
	113	int crush_scratch_ary[CEPH_PG_MAX_SIZE * 3];
f24e9980 SW	114	};
f24e9980 SW	115
4295f221 ID	116	static inline void ceph_oid_set_name(struct ceph_object_id *oid,
	117	const char *name)
	118	{
	119	int len;
	120
	121	len = strlen(name);
	122	if (len > sizeof(oid->name)) {
	123	WARN(1, "ceph_oid_set_name '%s' len %d vs %zu, truncating\n",
	124	name, len, sizeof(oid->name));
	125	len = sizeof(oid->name);
	126	}
	127
	128	memcpy(oid->name, name, len);
	129	oid->name_len = len;
	130	}
	131
	132	static inline void ceph_oid_copy(struct ceph_object_id *dest,
	133	struct ceph_object_id *src)
	134	{
	135	BUG_ON(src->name_len > sizeof(dest->name));
	136	memcpy(dest->name, src->name, src->name_len);
	137	dest->name_len = src->name_len;
	138	}
	139
246138fa ID	140	static inline int ceph_osd_exists(struct ceph_osdmap *map, int osd)
	141	{
	142	return osd >= 0 && osd < map->max_osd &&
	143	(map->osd_state[osd] & CEPH_OSD_EXISTS);
	144	}
	145
f24e9980 SW	146	static inline int ceph_osd_is_up(struct ceph_osdmap *map, int osd)
f24e9980 SW	147	{
246138fa ID	148	return ceph_osd_exists(map, osd) &&
	149	(map->osd_state[osd] & CEPH_OSD_UP);
	150	}
	151
	152	static inline int ceph_osd_is_down(struct ceph_osdmap *map, int osd)
	153	{
	154	return !ceph_osd_is_up(map, osd);
f24e9980 SW	155	}
	156
	157	static inline bool ceph_osdmap_flag(struct ceph_osdmap *map, int flag)
	158	{
	159	return map && (map->flags & flag);
	160	}
	161
	162	extern char ceph_osdmap_state_str(char str, int len, int state);
2cfa34f2	163	extern u32 ceph_get_primary_affinity(struct ceph_osdmap *map, int osd);
f24e9980 SW	164
	165	static inline struct ceph_entity_addr ceph_osd_addr(struct ceph_osdmap map,
	166	int osd)
	167	{
	168	if (osd >= map->max_osd)
	169	return NULL;
	170	return &map->osd_addr[osd];
	171	}
	172
ef4859d6 AE	173	static inline int ceph_decode_pgid(void *p, void end, struct ceph_pg *pgid)
	174	{
	175	__u8 version;
	176
	177	if (!ceph_has_room(p, end, 1 + 8 + 4 + 4)) {
3ef650d3	178	pr_warn("incomplete pg encoding\n");
ef4859d6 AE	179	return -EINVAL;
	180	}
	181	version = ceph_decode_8(p);
	182	if (version > 1) {
3ef650d3	183	pr_warn("do not understand pg encoding %d > 1\n",
ef4859d6 AE	184	(int)version);
	185	return -EINVAL;
	186	}
	187
	188	pgid->pool = ceph_decode_64(p);
	189	pgid->seed = ceph_decode_32(p);
	190	p += 4; / skip deprecated preferred value */
	191
	192	return 0;
	193	}
	194
a2505d63	195	extern struct ceph_osdmap ceph_osdmap_decode(void p, void end);
f24e9980 SW	196	extern struct ceph_osdmap osdmap_apply_incremental(void p, void end,
	197	struct ceph_osdmap *map,
	198	struct ceph_messenger *msgr);
	199	extern void ceph_osdmap_destroy(struct ceph_osdmap *map);
	200
	201	/* calculate mapping of a file extent to an object */
d63b77f4	202	extern int ceph_calc_file_object_mapping(struct ceph_file_layout *layout,
e8afad65	203	u64 off, u64 len,
d63b77f4	204	u64 bno, u64 oxoff, u64 *oxlen);
f24e9980 SW	205
f24e9980 SW	206	/* calculate mapping of object to a placement group */
7c13cb64 ID	207	extern int ceph_oloc_oid_to_pg(struct ceph_osdmap *osdmap,
	208	struct ceph_object_locator *oloc,
	209	struct ceph_object_id *oid,
	210	struct ceph_pg *pg_out);
	211
12979354	212	extern int ceph_calc_pg_acting(struct ceph_osdmap *osdmap,
5b191d99	213	struct ceph_pg pgid,
8008ab10	214	int osds, int primary);
51042122	215	extern int ceph_calc_pg_primary(struct ceph_osdmap *osdmap,
5b191d99	216	struct ceph_pg pgid);
f24e9980	217
ce7f6a27 ID	218	extern struct ceph_pg_pool_info ceph_pg_pool_by_id(struct ceph_osdmap map,
	219	u64 id);
	220
72afc71f	221	extern const char ceph_pg_pool_name_by_id(struct ceph_osdmap map, u64 id);
7669a2c9 YS	222	extern int ceph_pg_poolid_by_name(struct ceph_osdmap map, const char name);
7669a2c9 YS	223
f24e9980	224	#endif