[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

/* 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 */
	unsigned          r_num_pages;        /* size of page array (follows) */
	unsigned          r_page_alignment;   /* io offset in first page */
	struct page     **r_pages;            /* pages for data payload */
	int               r_pages_from_pool;
	int               r_own_pages;        /* if true, i own page list */
#ifdef CONFIG_BLOCK
	struct bio       *r_bio;	      /* instead of pages */
#endif

	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, int page_align);

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
f24e9980 SW	53	/* an in-flight request */
	54	struct ceph_osd_request {
	55	u64 r_tid; /* unique for this client */
	56	struct rb_node r_node;
422d2cb8	57	struct list_head r_req_lru_item;
f24e9980	58	struct list_head r_osd_item;
a40c4f10 YS	59	struct list_head r_linger_item;
a40c4f10 YS	60	struct list_head r_linger_osd;
f24e9980	61	struct ceph_osd *r_osd;
5b191d99	62	struct ceph_pg r_pgid;
d85b7056 SW	63	int r_pg_osds[CEPH_PG_MAX_SIZE];
d85b7056 SW	64	int r_num_pg_osds;
f24e9980	65
0d59ab81	66	struct ceph_connection *r_con_filling_msg;
350b1c32	67
f24e9980	68	struct ceph_msg r_request, r_reply;
f24e9980 SW	69	int r_flags; /* any additional flags for the osd */
f24e9980 SW	70	u32 r_sent; /* >0 if r_request is sending/sent */
1b83bef2 SW	71	int r_num_ops;
	72
	73	/* encoded message content */
	74	struct ceph_osd_op *r_request_ops;
	75	/* these are updated on each send */
	76	__le32 *r_request_osdmap_epoch;
	77	__le32 *r_request_flags;
	78	__le64 *r_request_pool;
	79	void *r_request_pgid;
	80	__le32 *r_request_attempts;
	81	struct ceph_eversion *r_request_reassert_version;
	82
	83	int r_result;
	84	int r_reply_op_len[CEPH_OSD_MAX_OP];
	85	s32 r_reply_op_result[CEPH_OSD_MAX_OP];
350b1c32	86	int r_got_reply;
a40c4f10	87	int r_linger;
0d5af164	88	int r_completed;
f24e9980 SW	89
f24e9980 SW	90	struct ceph_osd_client *r_osdc;
415e49a9	91	struct kref r_kref;
f24e9980 SW	92	bool r_mempool;
	93	struct completion r_completion, r_safe_completion;
	94	ceph_osdc_callback_t r_callback, r_safe_callback;
	95	struct ceph_eversion r_reassert_version;
	96	struct list_head r_unsafe_item;
	97
	98	struct inode r_inode; / for use by callbacks */
3d14c5d2	99	void r_priv; / ditto */
f24e9980	100
224736d9	101	char r_oid[MAX_OBJ_NAME_SIZE]; /* object name */
f24e9980	102	int r_oid_len;
2169aea6	103	u64 r_snapid;
3dd72fc0	104	unsigned long r_stamp; /* send OR check time */
f24e9980 SW	105
	106	struct ceph_file_layout r_file_layout;
	107	struct ceph_snap_context r_snapc; / snap context for writes */
	108	unsigned r_num_pages; /* size of page array (follows) */
b7495fc2	109	unsigned r_page_alignment; /* io offset in first page */
f24e9980 SW	110	struct page *r_pages; / pages for data payload */
	111	int r_pages_from_pool;
	112	int r_own_pages; /* if true, i own page list */
68b4476b YS	113	#ifdef CONFIG_BLOCK
	114	struct bio r_bio; / instead of pages */
	115	#endif
	116
c885837f	117	struct ceph_pagelist r_trail; /* trailing part of the data */
f24e9980 SW	118	};
f24e9980 SW	119
a40c4f10 YS	120	struct ceph_osd_event {
	121	u64 cookie;
	122	int one_shot;
	123	struct ceph_osd_client *osdc;
	124	void (cb)(u64, u64, u8, void );
	125	void *data;
	126	struct rb_node node;
	127	struct list_head osd_node;
	128	struct kref kref;
a40c4f10 YS	129	};
	130
	131	struct ceph_osd_event_work {
	132	struct work_struct work;
	133	struct ceph_osd_event *event;
	134	u64 ver;
	135	u64 notify_id;
	136	u8 opcode;
	137	};
	138
f24e9980 SW	139	struct ceph_osd_client {
	140	struct ceph_client *client;
	141
	142	struct ceph_osdmap osdmap; / current map */
	143	struct rw_semaphore map_sem;
	144	struct completion map_waiters;
	145	u64 last_requested_map;
	146
	147	struct mutex request_mutex;
	148	struct rb_root osds; /* osds */
f5a2041b	149	struct list_head osd_lru; /* idle osds */
f24e9980 SW	150	u64 timeout_tid; /* tid of timeout triggering rq */
	151	u64 last_tid; /* tid of last request */
	152	struct rb_root requests; /* pending requests */
6f6c7006 SW	153	struct list_head req_lru; /* in-flight lru */
	154	struct list_head req_unsent; /* unsent/need-resend queue */
	155	struct list_head req_notarget; /* map to no osd */
a40c4f10	156	struct list_head req_linger; /* lingering requests */
f24e9980 SW	157	int num_requests;
f24e9980 SW	158	struct delayed_work timeout_work;
f5a2041b	159	struct delayed_work osds_timeout_work;
039934b8	160	#ifdef CONFIG_DEBUG_FS
f24e9980	161	struct dentry *debugfs_file;
039934b8	162	#endif
f24e9980 SW	163
	164	mempool_t *req_mempool;
	165
0d59ab81	166	struct ceph_msgpool msgpool_op;
c16e7869	167	struct ceph_msgpool msgpool_op_reply;
a40c4f10 YS	168
	169	spinlock_t event_lock;
	170	struct rb_root event_tree;
	171	u64 event_count;
	172
	173	struct workqueue_struct *notify_wq;
f24e9980 SW	174	};
f24e9980 SW	175
68b4476b YS	176	struct ceph_osd_req_op {
68b4476b YS	177	u16 op; /* CEPH_OSD_OP_* */
e7e319a9	178	u32 payload_len;
68b4476b YS	179	union {
	180	struct {
	181	u64 offset, length;
	182	u64 truncate_size;
	183	u32 truncate_seq;
	184	} extent;
	185	struct {
	186	const char *name;
2a24d1f4	187	const void *val;
e7e319a9	188	u32 name_len;
68b4476b YS	189	u32 value_len;
	190	__u8 cmp_op; /* CEPH_OSD_CMPXATTR_OP_* */
	191	__u8 cmp_mode; /* CEPH_OSD_CMPXATTR_MODE_* */
	192	} xattr;
	193	struct {
ae1533b6	194	const char *class_name;
ae1533b6	195	const char *method_name;
2a24d1f4	196	const void *indata;
68b4476b	197	u32 indata_len;
e7e319a9 AE	198	__u8 class_len;
	199	__u8 method_len;
	200	__u8 argc;
68b4476b YS	201	} cls;
68b4476b YS	202	struct {
e7e319a9 AE	203	u64 cookie;
e7e319a9 AE	204	u64 count;
68b4476b	205	} pgls;
ae1533b6 YS	206	struct {
	207	u64 snapid;
	208	} snap;
a40c4f10 YS	209	struct {
	210	u64 cookie;
	211	u64 ver;
a40c4f10 YS	212	u32 prot_ver;
a40c4f10 YS	213	u32 timeout;
e7e319a9	214	__u8 flag;
a40c4f10	215	} watch;
68b4476b	216	};
68b4476b YS	217	};
68b4476b YS	218
f24e9980 SW	219	extern int ceph_osdc_init(struct ceph_osd_client *osdc,
	220	struct ceph_client *client);
	221	extern void ceph_osdc_stop(struct ceph_osd_client *osdc);
	222
	223	extern void ceph_osdc_handle_reply(struct ceph_osd_client *osdc,
	224	struct ceph_msg *msg);
	225	extern void ceph_osdc_handle_map(struct ceph_osd_client *osdc,
	226	struct ceph_msg *msg);
	227
3499e8a5	228	extern struct ceph_osd_request ceph_osdc_alloc_request(struct ceph_osd_client osdc,
3499e8a5	229	struct ceph_snap_context *snapc,
ae7ca4a3	230	unsigned int num_op,
3499e8a5	231	bool use_mempool,
54a54007	232	gfp_t gfp_flags);
3499e8a5 YS	233
3499e8a5 YS	234	extern void ceph_osdc_build_request(struct ceph_osd_request *req,
0120be3c	235	u64 off, u64 len,
ae7ca4a3	236	unsigned int num_op,
68b4476b YS	237	struct ceph_osd_req_op *src_ops,
68b4476b YS	238	struct ceph_snap_context *snapc,
4d6b250b	239	u64 snap_id,
af77f26c	240	struct timespec *mtime);
3499e8a5	241
f24e9980 SW	242	extern struct ceph_osd_request ceph_osdc_new_request(struct ceph_osd_client ,
	243	struct ceph_file_layout *layout,
	244	struct ceph_vino vino,
	245	u64 offset, u64 *len, int op, int flags,
	246	struct ceph_snap_context *snapc,
	247	int do_sync, u32 truncate_seq,
	248	u64 truncate_size,
	249	struct timespec *mtime,
a3bea47e	250	bool use_mempool, int page_align);
f24e9980	251
a40c4f10 YS	252	extern void ceph_osdc_set_request_linger(struct ceph_osd_client *osdc,
	253	struct ceph_osd_request *req);
	254	extern void ceph_osdc_unregister_linger_request(struct ceph_osd_client *osdc,
	255	struct ceph_osd_request *req);
	256
f24e9980 SW	257	static inline void ceph_osdc_get_request(struct ceph_osd_request *req)
f24e9980 SW	258	{
415e49a9 SW	259	kref_get(&req->r_kref);
	260	}
	261	extern void ceph_osdc_release_request(struct kref *kref);
	262	static inline void ceph_osdc_put_request(struct ceph_osd_request *req)
	263	{
	264	kref_put(&req->r_kref, ceph_osdc_release_request);
f24e9980	265	}
f24e9980 SW	266
	267	extern int ceph_osdc_start_request(struct ceph_osd_client *osdc,
	268	struct ceph_osd_request *req,
	269	bool nofail);
	270	extern int ceph_osdc_wait_request(struct ceph_osd_client *osdc,
	271	struct ceph_osd_request *req);
	272	extern void ceph_osdc_sync(struct ceph_osd_client *osdc);
	273
	274	extern int ceph_osdc_readpages(struct ceph_osd_client *osdc,
	275	struct ceph_vino vino,
	276	struct ceph_file_layout *layout,
	277	u64 off, u64 *plen,
	278	u32 truncate_seq, u64 truncate_size,
b7495fc2 SW	279	struct page **pages, int nr_pages,
b7495fc2 SW	280	int page_align);
f24e9980 SW	281
	282	extern int ceph_osdc_writepages(struct ceph_osd_client *osdc,
	283	struct ceph_vino vino,
	284	struct ceph_file_layout *layout,
	285	struct ceph_snap_context *sc,
	286	u64 off, u64 len,
	287	u32 truncate_seq, u64 truncate_size,
	288	struct timespec *mtime,
24808826	289	struct page **pages, int nr_pages);
f24e9980	290
a40c4f10 YS	291	/* watch/notify events */
	292	extern int ceph_osdc_create_event(struct ceph_osd_client *osdc,
	293	void (event_cb)(u64, u64, u8, void ),
3c663bbd	294	void data, struct ceph_osd_event *pevent);
a40c4f10	295	extern void ceph_osdc_cancel_event(struct ceph_osd_event *event);
a40c4f10	296	extern void ceph_osdc_put_event(struct ceph_osd_event *event);
f24e9980 SW	297	#endif
f24e9980 SW	298