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

#ifndef _FS_CEPH_OSD_CLIENT_H
#define _FS_CEPH_OSD_CLIENT_H

#include <linux/completion.h>
#include <linux/kref.h>
#include <linux/mempool.h>
#include <linux/rbtree.h>

#include <linux/ceph/types.h>
#include <linux/ceph/osdmap.h>
#include <linux/ceph/messenger.h>
#include <linux/ceph/auth.h>
#include <linux/ceph/pagelist.h>

/* 
 * Maximum object name size 
 * (must be at least as big as RBD_MAX_MD_NAME_LEN -- currently 100) 
 */
#define MAX_OBJ_NAME_SIZE 100

struct ceph_msg;
struct ceph_snap_context;
struct ceph_osd_request;
struct ceph_osd_client;
struct ceph_authorizer;

/*
 * completion callback for async writepages
 */
typedef void (*ceph_osdc_callback_t)(struct ceph_osd_request *,
				     struct ceph_msg *);

/* a given osd we're communicating with */
struct ceph_osd {
	atomic_t o_ref;
	struct ceph_osd_client *o_osdc;
	int o_osd;
	int o_incarnation;
	struct rb_node o_node;
	struct ceph_connection o_con;
	struct list_head o_requests;
	struct list_head o_linger_requests;
	struct list_head o_osd_lru;
	struct ceph_auth_handshake o_auth;
	unsigned long lru_ttl;
	int o_marked_for_keepalive;
	struct list_head o_keepalive_item;
};


#define CEPH_OSD_MAX_OP 10

enum ceph_osd_data_type {
	CEPH_OSD_DATA_TYPE_NONE,
	CEPH_OSD_DATA_TYPE_PAGES,
#ifdef CONFIG_BLOCK
	CEPH_OSD_DATA_TYPE_BIO,
#endif /* CONFIG_BLOCK */
};

struct ceph_osd_data {
	enum ceph_osd_data_type	type;
	union {
		struct {
			struct page	**pages;
			u32		num_pages;
			u32		alignment;
			bool		pages_from_pool;
			bool		own_pages;
		};
#ifdef CONFIG_BLOCK
		struct bio       *bio;
#endif /* CONFIG_BLOCK */
	};
};

/* an in-flight request */
struct ceph_osd_request {
	u64             r_tid;              /* unique for this client */
	struct rb_node  r_node;
	struct list_head r_req_lru_item;
	struct list_head r_osd_item;
	struct list_head r_linger_item;
	struct list_head r_linger_osd;
	struct ceph_osd *r_osd;
	struct ceph_pg   r_pgid;
	int              r_pg_osds[CEPH_PG_MAX_SIZE];
	int              r_num_pg_osds;

	struct ceph_connection *r_con_filling_msg;

	struct ceph_msg  *r_request, *r_reply;
	int               r_flags;     /* any additional flags for the osd */
	u32               r_sent;      /* >0 if r_request is sending/sent */
	int               r_num_ops;

	/* encoded message content */
	struct ceph_osd_op *r_request_ops;
	/* these are updated on each send */
	__le32           *r_request_osdmap_epoch;
	__le32           *r_request_flags;
	__le64           *r_request_pool;
	void             *r_request_pgid;
	__le32           *r_request_attempts;
	struct ceph_eversion *r_request_reassert_version;

	int               r_result;
	int               r_reply_op_len[CEPH_OSD_MAX_OP];
	s32               r_reply_op_result[CEPH_OSD_MAX_OP];
	int               r_got_reply;
	int		  r_linger;
	int		  r_completed;

	struct ceph_osd_client *r_osdc;
	struct kref       r_kref;
	bool              r_mempool;
	struct completion r_completion, r_safe_completion;
	ceph_osdc_callback_t r_callback, r_safe_callback;
	struct ceph_eversion r_reassert_version;
	struct list_head  r_unsafe_item;

	struct inode *r_inode;         	      /* for use by callbacks */
	void *r_priv;			      /* ditto */

	char              r_oid[MAX_OBJ_NAME_SIZE];          /* object name */
	int               r_oid_len;
	u64               r_snapid;
	unsigned long     r_stamp;            /* send OR check time */

	struct ceph_file_layout r_file_layout;
	struct ceph_snap_context *r_snapc;    /* snap context for writes */

	struct ceph_osd_data r_data;
	struct ceph_pagelist r_trail;	      /* trailing part of the data */
};

struct ceph_osd_event {
	u64 cookie;
	int one_shot;
	struct ceph_osd_client *osdc;
	void (*cb)(u64, u64, u8, void *);
	void *data;
	struct rb_node node;
	struct list_head osd_node;
	struct kref kref;
};

struct ceph_osd_event_work {
	struct work_struct work;
	struct ceph_osd_event *event;
        u64 ver;
        u64 notify_id;
        u8 opcode;
};

struct ceph_osd_client {
	struct ceph_client     *client;

	struct ceph_osdmap     *osdmap;       /* current map */
	struct rw_semaphore    map_sem;
	struct completion      map_waiters;
	u64                    last_requested_map;

	struct mutex           request_mutex;
	struct rb_root         osds;          /* osds */
	struct list_head       osd_lru;       /* idle osds */
	u64                    timeout_tid;   /* tid of timeout triggering rq */
	u64                    last_tid;      /* tid of last request */
	struct rb_root         requests;      /* pending requests */
	struct list_head       req_lru;	      /* in-flight lru */
	struct list_head       req_unsent;    /* unsent/need-resend queue */
	struct list_head       req_notarget;  /* map to no osd */
	struct list_head       req_linger;    /* lingering requests */
	int                    num_requests;
	struct delayed_work    timeout_work;
	struct delayed_work    osds_timeout_work;
#ifdef CONFIG_DEBUG_FS
	struct dentry 	       *debugfs_file;
#endif

	mempool_t              *req_mempool;

	struct ceph_msgpool	msgpool_op;
	struct ceph_msgpool	msgpool_op_reply;

	spinlock_t		event_lock;
	struct rb_root		event_tree;
	u64			event_count;

	struct workqueue_struct	*notify_wq;
};

struct ceph_osd_req_op {
	u16 op;           /* CEPH_OSD_OP_* */
	u32 payload_len;
	union {
		struct {
			u64 offset, length;
			u64 truncate_size;
			u32 truncate_seq;
		} extent;
		struct {
			const char *name;
			const void *val;
			u32 name_len;
			u32 value_len;
			__u8 cmp_op;       /* CEPH_OSD_CMPXATTR_OP_* */
			__u8 cmp_mode;     /* CEPH_OSD_CMPXATTR_MODE_* */
		} xattr;
		struct {
			const char *class_name;
			const char *method_name;
			const void *indata;
			u32 indata_len;
			__u8 class_len;
			__u8 method_len;
			__u8 argc;
		} cls;
		struct {
			u64 cookie;
			u64 count;
		} pgls;
	        struct {
		        u64 snapid;
	        } snap;
		struct {
			u64 cookie;
			u64 ver;
			u32 prot_ver;
			u32 timeout;
			__u8 flag;
		} watch;
	};
};

extern int ceph_osdc_init(struct ceph_osd_client *osdc,
			  struct ceph_client *client);
extern void ceph_osdc_stop(struct ceph_osd_client *osdc);

extern void ceph_osdc_handle_reply(struct ceph_osd_client *osdc,
				   struct ceph_msg *msg);
extern void ceph_osdc_handle_map(struct ceph_osd_client *osdc,
				 struct ceph_msg *msg);

extern struct ceph_osd_request *ceph_osdc_alloc_request(struct ceph_osd_client *osdc,
					       struct ceph_snap_context *snapc,
					       unsigned int num_op,
					       bool use_mempool,
					       gfp_t gfp_flags);

extern void ceph_osdc_build_request(struct ceph_osd_request *req,
				    u64 off, u64 len,
				    unsigned int num_op,
				    struct ceph_osd_req_op *src_ops,
				    struct ceph_snap_context *snapc,
				    u64 snap_id,
				    struct timespec *mtime);

extern struct ceph_osd_request *ceph_osdc_new_request(struct ceph_osd_client *,
				      struct ceph_file_layout *layout,
				      struct ceph_vino vino,
				      u64 offset, u64 *len, int op, int flags,
				      struct ceph_snap_context *snapc,
				      int do_sync, u32 truncate_seq,
				      u64 truncate_size,
				      struct timespec *mtime,
				      bool use_mempool);

extern void ceph_osdc_set_request_linger(struct ceph_osd_client *osdc,
					 struct ceph_osd_request *req);
extern void ceph_osdc_unregister_linger_request(struct ceph_osd_client *osdc,
						struct ceph_osd_request *req);

static inline void ceph_osdc_get_request(struct ceph_osd_request *req)
{
	kref_get(&req->r_kref);
}
extern void ceph_osdc_release_request(struct kref *kref);
static inline void ceph_osdc_put_request(struct ceph_osd_request *req)
{
	kref_put(&req->r_kref, ceph_osdc_release_request);
}

extern int ceph_osdc_start_request(struct ceph_osd_client *osdc,
				   struct ceph_osd_request *req,
				   bool nofail);
extern int ceph_osdc_wait_request(struct ceph_osd_client *osdc,
				  struct ceph_osd_request *req);
extern void ceph_osdc_sync(struct ceph_osd_client *osdc);

extern int ceph_osdc_readpages(struct ceph_osd_client *osdc,
			       struct ceph_vino vino,
			       struct ceph_file_layout *layout,
			       u64 off, u64 *plen,
			       u32 truncate_seq, u64 truncate_size,
			       struct page **pages, int nr_pages,
			       int page_align);

extern int ceph_osdc_writepages(struct ceph_osd_client *osdc,
				struct ceph_vino vino,
				struct ceph_file_layout *layout,
				struct ceph_snap_context *sc,
				u64 off, u64 len,
				u32 truncate_seq, u64 truncate_size,
				struct timespec *mtime,
				struct page **pages, int nr_pages);

/* watch/notify events */
extern int ceph_osdc_create_event(struct ceph_osd_client *osdc,
				  void (*event_cb)(u64, u64, u8, void *),
				  void *data, struct ceph_osd_event **pevent);
extern void ceph_osdc_cancel_event(struct ceph_osd_event *event);
extern void ceph_osdc_put_event(struct ceph_osd_event *event);
#endif
Commit	Line	Data
f24e9980 SW	1	#ifndef _FS_CEPH_OSD_CLIENT_H
	2	#define _FS_CEPH_OSD_CLIENT_H
	3
	4	#include <linux/completion.h>
415e49a9	5	#include <linux/kref.h>
f24e9980 SW	6	#include <linux/mempool.h>
	7	#include <linux/rbtree.h>
	8
6c4a1915 AE	9	#include <linux/ceph/types.h>
	10	#include <linux/ceph/osdmap.h>
	11	#include <linux/ceph/messenger.h>
	12	#include <linux/ceph/auth.h>
c885837f	13	#include <linux/ceph/pagelist.h>
f24e9980	14
224736d9 SP	15	/*
	16	* Maximum object name size
	17	* (must be at least as big as RBD_MAX_MD_NAME_LEN -- currently 100)
	18	*/
	19	#define MAX_OBJ_NAME_SIZE 100
	20
f24e9980 SW	21	struct ceph_msg;
	22	struct ceph_snap_context;
	23	struct ceph_osd_request;
	24	struct ceph_osd_client;
4e7a5dcd	25	struct ceph_authorizer;
f24e9980 SW	26
	27	/*
	28	* completion callback for async writepages
	29	*/
	30	typedef void (ceph_osdc_callback_t)(struct ceph_osd_request ,
	31	struct ceph_msg *);
	32
	33	/* a given osd we're communicating with */
	34	struct ceph_osd {
	35	atomic_t o_ref;
	36	struct ceph_osd_client *o_osdc;
	37	int o_osd;
	38	int o_incarnation;
	39	struct rb_node o_node;
	40	struct ceph_connection o_con;
	41	struct list_head o_requests;
a40c4f10	42	struct list_head o_linger_requests;
f5a2041b	43	struct list_head o_osd_lru;
6c4a1915	44	struct ceph_auth_handshake o_auth;
f5a2041b	45	unsigned long lru_ttl;
422d2cb8 YS	46	int o_marked_for_keepalive;
422d2cb8 YS	47	struct list_head o_keepalive_item;
f24e9980 SW	48	};
f24e9980 SW	49
1b83bef2 SW	50
	51	#define CEPH_OSD_MAX_OP 10
	52
2ac2b7a6 AE	53	enum ceph_osd_data_type {
	54	CEPH_OSD_DATA_TYPE_NONE,
	55	CEPH_OSD_DATA_TYPE_PAGES,
	56	#ifdef CONFIG_BLOCK
	57	CEPH_OSD_DATA_TYPE_BIO,
	58	#endif /* CONFIG_BLOCK */
	59	};
	60
2794a82a	61	struct ceph_osd_data {
2ac2b7a6 AE	62	enum ceph_osd_data_type type;
2ac2b7a6 AE	63	union {
2794a82a AE	64	struct {
	65	struct page **pages;
	66	u32 num_pages;
	67	u32 alignment;
	68	bool pages_from_pool;
	69	bool own_pages;
	70	};
	71	#ifdef CONFIG_BLOCK
	72	struct bio *bio;
	73	#endif /* CONFIG_BLOCK */
	74	};
	75	};
	76
f24e9980 SW	77	/* an in-flight request */
	78	struct ceph_osd_request {
	79	u64 r_tid; /* unique for this client */
	80	struct rb_node r_node;
422d2cb8	81	struct list_head r_req_lru_item;
f24e9980	82	struct list_head r_osd_item;
a40c4f10 YS	83	struct list_head r_linger_item;
a40c4f10 YS	84	struct list_head r_linger_osd;
f24e9980	85	struct ceph_osd *r_osd;
5b191d99	86	struct ceph_pg r_pgid;
d85b7056 SW	87	int r_pg_osds[CEPH_PG_MAX_SIZE];
d85b7056 SW	88	int r_num_pg_osds;
f24e9980	89
0d59ab81	90	struct ceph_connection *r_con_filling_msg;
350b1c32	91
f24e9980	92	struct ceph_msg r_request, r_reply;
f24e9980 SW	93	int r_flags; /* any additional flags for the osd */
f24e9980 SW	94	u32 r_sent; /* >0 if r_request is sending/sent */
1b83bef2 SW	95	int r_num_ops;
	96
	97	/* encoded message content */
	98	struct ceph_osd_op *r_request_ops;
	99	/* these are updated on each send */
	100	__le32 *r_request_osdmap_epoch;
	101	__le32 *r_request_flags;
	102	__le64 *r_request_pool;
	103	void *r_request_pgid;
	104	__le32 *r_request_attempts;
	105	struct ceph_eversion *r_request_reassert_version;
	106
	107	int r_result;
	108	int r_reply_op_len[CEPH_OSD_MAX_OP];
	109	s32 r_reply_op_result[CEPH_OSD_MAX_OP];
350b1c32	110	int r_got_reply;
a40c4f10	111	int r_linger;
0d5af164	112	int r_completed;
f24e9980 SW	113
f24e9980 SW	114	struct ceph_osd_client *r_osdc;
415e49a9	115	struct kref r_kref;
f24e9980 SW	116	bool r_mempool;
	117	struct completion r_completion, r_safe_completion;
	118	ceph_osdc_callback_t r_callback, r_safe_callback;
	119	struct ceph_eversion r_reassert_version;
	120	struct list_head r_unsafe_item;
	121
	122	struct inode r_inode; / for use by callbacks */
3d14c5d2	123	void r_priv; / ditto */
f24e9980	124
224736d9	125	char r_oid[MAX_OBJ_NAME_SIZE]; /* object name */
f24e9980	126	int r_oid_len;
2169aea6	127	u64 r_snapid;
3dd72fc0	128	unsigned long r_stamp; /* send OR check time */
f24e9980 SW	129
	130	struct ceph_file_layout r_file_layout;
	131	struct ceph_snap_context r_snapc; / snap context for writes */
68b4476b	132
2794a82a	133	struct ceph_osd_data r_data;
c885837f	134	struct ceph_pagelist r_trail; /* trailing part of the data */
f24e9980 SW	135	};
f24e9980 SW	136
a40c4f10 YS	137	struct ceph_osd_event {
	138	u64 cookie;
	139	int one_shot;
	140	struct ceph_osd_client *osdc;
	141	void (cb)(u64, u64, u8, void );
	142	void *data;
	143	struct rb_node node;
	144	struct list_head osd_node;
	145	struct kref kref;
a40c4f10 YS	146	};
	147
	148	struct ceph_osd_event_work {
	149	struct work_struct work;
	150	struct ceph_osd_event *event;
	151	u64 ver;
	152	u64 notify_id;
	153	u8 opcode;
	154	};
	155
f24e9980 SW	156	struct ceph_osd_client {
	157	struct ceph_client *client;
	158
	159	struct ceph_osdmap osdmap; / current map */
	160	struct rw_semaphore map_sem;
	161	struct completion map_waiters;
	162	u64 last_requested_map;
	163
	164	struct mutex request_mutex;
	165	struct rb_root osds; /* osds */
f5a2041b	166	struct list_head osd_lru; /* idle osds */
f24e9980 SW	167	u64 timeout_tid; /* tid of timeout triggering rq */
	168	u64 last_tid; /* tid of last request */
	169	struct rb_root requests; /* pending requests */
6f6c7006 SW	170	struct list_head req_lru; /* in-flight lru */
	171	struct list_head req_unsent; /* unsent/need-resend queue */
	172	struct list_head req_notarget; /* map to no osd */
a40c4f10	173	struct list_head req_linger; /* lingering requests */
f24e9980 SW	174	int num_requests;
f24e9980 SW	175	struct delayed_work timeout_work;
f5a2041b	176	struct delayed_work osds_timeout_work;
039934b8	177	#ifdef CONFIG_DEBUG_FS
f24e9980	178	struct dentry *debugfs_file;
039934b8	179	#endif
f24e9980 SW	180
	181	mempool_t *req_mempool;
	182
0d59ab81	183	struct ceph_msgpool msgpool_op;
c16e7869	184	struct ceph_msgpool msgpool_op_reply;
a40c4f10 YS	185
	186	spinlock_t event_lock;
	187	struct rb_root event_tree;
	188	u64 event_count;
	189
	190	struct workqueue_struct *notify_wq;
f24e9980 SW	191	};
f24e9980 SW	192
68b4476b YS	193	struct ceph_osd_req_op {
68b4476b YS	194	u16 op; /* CEPH_OSD_OP_* */
e7e319a9	195	u32 payload_len;
68b4476b YS	196	union {
	197	struct {
	198	u64 offset, length;
	199	u64 truncate_size;
	200	u32 truncate_seq;
	201	} extent;
	202	struct {
	203	const char *name;
2a24d1f4	204	const void *val;
e7e319a9	205	u32 name_len;
68b4476b YS	206	u32 value_len;
	207	__u8 cmp_op; /* CEPH_OSD_CMPXATTR_OP_* */
	208	__u8 cmp_mode; /* CEPH_OSD_CMPXATTR_MODE_* */
	209	} xattr;
	210	struct {
ae1533b6	211	const char *class_name;
ae1533b6	212	const char *method_name;
2a24d1f4	213	const void *indata;
68b4476b	214	u32 indata_len;
e7e319a9 AE	215	__u8 class_len;
	216	__u8 method_len;
	217	__u8 argc;
68b4476b YS	218	} cls;
68b4476b YS	219	struct {
e7e319a9 AE	220	u64 cookie;
e7e319a9 AE	221	u64 count;
68b4476b	222	} pgls;
ae1533b6 YS	223	struct {
	224	u64 snapid;
	225	} snap;
a40c4f10 YS	226	struct {
	227	u64 cookie;
	228	u64 ver;
a40c4f10 YS	229	u32 prot_ver;
a40c4f10 YS	230	u32 timeout;
e7e319a9	231	__u8 flag;
a40c4f10	232	} watch;
68b4476b	233	};
68b4476b YS	234	};
68b4476b YS	235
f24e9980 SW	236	extern int ceph_osdc_init(struct ceph_osd_client *osdc,
	237	struct ceph_client *client);
	238	extern void ceph_osdc_stop(struct ceph_osd_client *osdc);
	239
	240	extern void ceph_osdc_handle_reply(struct ceph_osd_client *osdc,
	241	struct ceph_msg *msg);
	242	extern void ceph_osdc_handle_map(struct ceph_osd_client *osdc,
	243	struct ceph_msg *msg);
	244
3499e8a5	245	extern struct ceph_osd_request ceph_osdc_alloc_request(struct ceph_osd_client osdc,
3499e8a5	246	struct ceph_snap_context *snapc,
ae7ca4a3	247	unsigned int num_op,
3499e8a5	248	bool use_mempool,
54a54007	249	gfp_t gfp_flags);
3499e8a5 YS	250
3499e8a5 YS	251	extern void ceph_osdc_build_request(struct ceph_osd_request *req,
0120be3c	252	u64 off, u64 len,
ae7ca4a3	253	unsigned int num_op,
68b4476b YS	254	struct ceph_osd_req_op *src_ops,
68b4476b YS	255	struct ceph_snap_context *snapc,
4d6b250b	256	u64 snap_id,
af77f26c	257	struct timespec *mtime);
3499e8a5	258
f24e9980 SW	259	extern struct ceph_osd_request ceph_osdc_new_request(struct ceph_osd_client ,
	260	struct ceph_file_layout *layout,
	261	struct ceph_vino vino,
	262	u64 offset, u64 *len, int op, int flags,
	263	struct ceph_snap_context *snapc,
	264	int do_sync, u32 truncate_seq,
	265	u64 truncate_size,
	266	struct timespec *mtime,
153e5167	267	bool use_mempool);
f24e9980	268
a40c4f10 YS	269	extern void ceph_osdc_set_request_linger(struct ceph_osd_client *osdc,
	270	struct ceph_osd_request *req);
	271	extern void ceph_osdc_unregister_linger_request(struct ceph_osd_client *osdc,
	272	struct ceph_osd_request *req);
	273
f24e9980 SW	274	static inline void ceph_osdc_get_request(struct ceph_osd_request *req)
f24e9980 SW	275	{
415e49a9 SW	276	kref_get(&req->r_kref);
	277	}
	278	extern void ceph_osdc_release_request(struct kref *kref);
	279	static inline void ceph_osdc_put_request(struct ceph_osd_request *req)
	280	{
	281	kref_put(&req->r_kref, ceph_osdc_release_request);
f24e9980	282	}
f24e9980 SW	283
	284	extern int ceph_osdc_start_request(struct ceph_osd_client *osdc,
	285	struct ceph_osd_request *req,
	286	bool nofail);
	287	extern int ceph_osdc_wait_request(struct ceph_osd_client *osdc,
	288	struct ceph_osd_request *req);
	289	extern void ceph_osdc_sync(struct ceph_osd_client *osdc);
	290
	291	extern int ceph_osdc_readpages(struct ceph_osd_client *osdc,
	292	struct ceph_vino vino,
	293	struct ceph_file_layout *layout,
	294	u64 off, u64 *plen,
	295	u32 truncate_seq, u64 truncate_size,
b7495fc2 SW	296	struct page **pages, int nr_pages,
b7495fc2 SW	297	int page_align);
f24e9980 SW	298
	299	extern int ceph_osdc_writepages(struct ceph_osd_client *osdc,
	300	struct ceph_vino vino,
	301	struct ceph_file_layout *layout,
	302	struct ceph_snap_context *sc,
	303	u64 off, u64 len,
	304	u32 truncate_seq, u64 truncate_size,
	305	struct timespec *mtime,
24808826	306	struct page **pages, int nr_pages);
f24e9980	307
a40c4f10 YS	308	/* watch/notify events */
	309	extern int ceph_osdc_create_event(struct ceph_osd_client *osdc,
	310	void (event_cb)(u64, u64, u8, void ),
3c663bbd	311	void data, struct ceph_osd_event *pevent);
a40c4f10	312	extern void ceph_osdc_cancel_event(struct ceph_osd_event *event);
a40c4f10	313	extern void ceph_osdc_put_event(struct ceph_osd_event *event);
f24e9980 SW	314	#endif
f24e9980 SW	315