[net-next-2.6.git] / fs / ceph / mds_client.h

#ifndef _FS_CEPH_MDS_CLIENT_H
#define _FS_CEPH_MDS_CLIENT_H

#include <linux/completion.h>
#include <linux/kref.h>
#include <linux/list.h>
#include <linux/mutex.h>
#include <linux/radix-tree.h>
#include <linux/spinlock.h>

#include "types.h"
#include "messenger.h"
#include "mdsmap.h"

/*
 * Some lock dependencies:
 *
 * session->s_mutex
 *         mdsc->mutex
 *
 *         mdsc->snap_rwsem
 *
 *         inode->i_lock
 *                 mdsc->snap_flush_lock
 *                 mdsc->cap_delay_lock
 *
 */

struct ceph_client;
struct ceph_cap;

/*
 * parsed info about a single inode.  pointers are into the encoded
 * on-wire structures within the mds reply message payload.
 */
struct ceph_mds_reply_info_in {
	struct ceph_mds_reply_inode *in;
	u32 symlink_len;
	char *symlink;
	u32 xattr_len;
	char *xattr_data;
};

/*
 * parsed info about an mds reply, including information about the
 * target inode and/or its parent directory and dentry, and directory
 * contents (for readdir results).
 */
struct ceph_mds_reply_info_parsed {
	struct ceph_mds_reply_head    *head;

	struct ceph_mds_reply_info_in diri, targeti;
	struct ceph_mds_reply_dirfrag *dirfrag;
	char                          *dname;
	u32                           dname_len;
	struct ceph_mds_reply_lease   *dlease;

	struct ceph_mds_reply_dirfrag *dir_dir;
	int                           dir_nr;
	char                          **dir_dname;
	u32                           *dir_dname_len;
	struct ceph_mds_reply_lease   **dir_dlease;
	struct ceph_mds_reply_info_in *dir_in;
	u8                            dir_complete, dir_end;

	/* encoded blob describing snapshot contexts for certain
	   operations (e.g., open) */
	void *snapblob;
	int snapblob_len;
};


/*
 * cap releases are batched and sent to the MDS en masse.
 */
#define CEPH_CAPS_PER_RELEASE ((PAGE_CACHE_SIZE -			\
				sizeof(struct ceph_mds_cap_release)) /	\
			       sizeof(struct ceph_mds_cap_item))


/*
 * state associated with each MDS<->client session
 */
enum {
	CEPH_MDS_SESSION_NEW = 1,
	CEPH_MDS_SESSION_OPENING = 2,
	CEPH_MDS_SESSION_OPEN = 3,
	CEPH_MDS_SESSION_HUNG = 4,
	CEPH_MDS_SESSION_CLOSING = 5,
	CEPH_MDS_SESSION_RESTARTING = 6,
	CEPH_MDS_SESSION_RECONNECTING = 7,
};

struct ceph_mds_session {
	struct ceph_mds_client *s_mdsc;
	int               s_mds;
	int               s_state;
	unsigned long     s_ttl;      /* time until mds kills us */
	u64               s_seq;      /* incoming msg seq # */
	struct mutex      s_mutex;    /* serialize session messages */

	struct ceph_connection s_con;

	struct ceph_authorizer *s_authorizer;
	void             *s_authorizer_buf, *s_authorizer_reply_buf;
	size_t            s_authorizer_buf_len, s_authorizer_reply_buf_len;

	/* protected by s_cap_lock */
	spinlock_t        s_cap_lock;
	u32               s_cap_gen;  /* inc each time we get mds stale msg */
	unsigned long     s_cap_ttl;  /* when session caps expire */
	struct list_head  s_caps;     /* all caps issued by this session */
	int               s_nr_caps, s_trim_caps;
	int               s_num_cap_releases;
	struct list_head  s_cap_releases; /* waiting cap_release messages */
	struct list_head  s_cap_releases_done; /* ready to send */
	bool              s_iterating_caps;

	/* protected by mutex */
	struct list_head  s_cap_flushing;     /* inodes w/ flushing caps */
	struct list_head  s_cap_snaps_flushing;
	unsigned long     s_renew_requested; /* last time we sent a renew req */
	u64               s_renew_seq;

	atomic_t          s_ref;
	struct list_head  s_waiting;  /* waiting requests */
	struct list_head  s_unsafe;   /* unsafe requests */
};

/*
 * modes of choosing which MDS to send a request to
 */
enum {
	USE_ANY_MDS,
	USE_RANDOM_MDS,
	USE_AUTH_MDS,   /* prefer authoritative mds for this metadata item */
};

struct ceph_mds_request;
struct ceph_mds_client;

/*
 * request completion callback
 */
typedef void (*ceph_mds_request_callback_t) (struct ceph_mds_client *mdsc,
					     struct ceph_mds_request *req);

/*
 * an in-flight mds request
 */
struct ceph_mds_request {
	u64 r_tid;                   /* transaction id */

	int r_op;                    /* mds op code */
	int r_mds;

	/* operation on what? */
	struct inode *r_inode;              /* arg1 */
	struct dentry *r_dentry;            /* arg1 */
	struct dentry *r_old_dentry;        /* arg2: rename from or link from */
	char *r_path1, *r_path2;
	struct ceph_vino r_ino1, r_ino2;

	struct inode *r_locked_dir; /* dir (if any) i_mutex locked by vfs */
	struct inode *r_target_inode;       /* resulting inode */

	union ceph_mds_request_args r_args;
	int r_fmode;        /* file mode, if expecting cap */

	/* for choosing which mds to send this request to */
	int r_direct_mode;
	u32 r_direct_hash;      /* choose dir frag based on this dentry hash */
	bool r_direct_is_hash;  /* true if r_direct_hash is valid */

	/* data payload is used for xattr ops */
	struct page **r_pages;
	int r_num_pages;
	int r_data_len;

	/* what caps shall we drop? */
	int r_inode_drop, r_inode_unless;
	int r_dentry_drop, r_dentry_unless;
	int r_old_dentry_drop, r_old_dentry_unless;
	struct inode *r_old_inode;
	int r_old_inode_drop, r_old_inode_unless;

	struct ceph_msg  *r_request;  /* original request */
	struct ceph_msg  *r_reply;
	struct ceph_mds_reply_info_parsed r_reply_info;
	int r_err;

	unsigned long r_timeout;  /* optional.  jiffies */
	unsigned long r_started;  /* start time to measure timeout against */
	unsigned long r_request_started; /* start time for mds request only,
					    used to measure lease durations */

	/* link unsafe requests to parent directory, for fsync */
	struct inode	*r_unsafe_dir;
	struct list_head r_unsafe_dir_item;

	struct ceph_mds_session *r_session;

	int               r_attempts;   /* resend attempts */
	int               r_num_fwd;    /* number of forward attempts */
	int               r_num_stale;
	int               r_resend_mds; /* mds to resend to next, if any*/

	struct kref       r_kref;
	struct list_head  r_wait;
	struct completion r_completion;
	struct completion r_safe_completion;
	ceph_mds_request_callback_t r_callback;
	struct list_head  r_unsafe_item;  /* per-session unsafe list item */
	bool		  r_got_unsafe, r_got_safe;

	bool              r_did_prepopulate;
	u32               r_readdir_offset;

	struct ceph_cap_reservation r_caps_reservation;
	int r_num_caps;
};

/*
 * mds client state
 */
struct ceph_mds_client {
	struct ceph_client      *client;
	struct mutex            mutex;         /* all nested structures */

	struct ceph_mdsmap      *mdsmap;
	struct completion       safe_umount_waiters, session_close_waiters;
	struct list_head        waiting_for_map;

	struct ceph_mds_session **sessions;    /* NULL for mds if no session */
	int                     max_sessions;  /* len of s_mds_sessions */
	int                     stopping;      /* true if shutting down */

	/*
	 * snap_rwsem will cover cap linkage into snaprealms, and
	 * realm snap contexts.  (later, we can do per-realm snap
	 * contexts locks..)  the empty list contains realms with no
	 * references (implying they contain no inodes with caps) that
	 * should be destroyed.
	 */
	struct rw_semaphore     snap_rwsem;
	struct radix_tree_root  snap_realms;
	struct list_head        snap_empty;
	spinlock_t              snap_empty_lock;  /* protect snap_empty */

	u64                    last_tid;      /* most recent mds request */
	struct radix_tree_root request_tree;  /* pending mds requests */
	struct delayed_work    delayed_work;  /* delayed work */
	unsigned long    last_renew_caps;  /* last time we renewed our caps */
	struct list_head cap_delay_list;   /* caps with delayed release */
	spinlock_t       cap_delay_lock;   /* protects cap_delay_list */
	struct list_head snap_flush_list;  /* cap_snaps ready to flush */
	spinlock_t       snap_flush_lock;

	u64               cap_flush_seq;
	struct list_head  cap_dirty;        /* inodes with dirty caps */
	int               num_cap_flushing; /* # caps we are flushing */
	spinlock_t        cap_dirty_lock;   /* protects above items */
	wait_queue_head_t cap_flushing_wq;

#ifdef CONFIG_DEBUG_FS
	struct dentry 	  *debugfs_file;
#endif

	spinlock_t	  dentry_lru_lock;
	struct list_head  dentry_lru;
	int		  num_dentry;
};

extern const char *ceph_mds_op_name(int op);

extern struct ceph_mds_session *
__ceph_lookup_mds_session(struct ceph_mds_client *, int mds);

static inline struct ceph_mds_session *
ceph_get_mds_session(struct ceph_mds_session *s)
{
	atomic_inc(&s->s_ref);
	return s;
}

extern void ceph_put_mds_session(struct ceph_mds_session *s);

extern int ceph_send_msg_mds(struct ceph_mds_client *mdsc,
			     struct ceph_msg *msg, int mds);

extern int ceph_mdsc_init(struct ceph_mds_client *mdsc,
			   struct ceph_client *client);
extern void ceph_mdsc_close_sessions(struct ceph_mds_client *mdsc);
extern void ceph_mdsc_stop(struct ceph_mds_client *mdsc);

extern void ceph_mdsc_sync(struct ceph_mds_client *mdsc);

extern void ceph_mdsc_lease_release(struct ceph_mds_client *mdsc,
				    struct inode *inode,
				    struct dentry *dn, int mask);

extern struct ceph_mds_request *
ceph_mdsc_create_request(struct ceph_mds_client *mdsc, int op, int mode);
extern void ceph_mdsc_submit_request(struct ceph_mds_client *mdsc,
				     struct ceph_mds_request *req);
extern int ceph_mdsc_do_request(struct ceph_mds_client *mdsc,
				struct inode *dir,
				struct ceph_mds_request *req);
static inline void ceph_mdsc_get_request(struct ceph_mds_request *req)
{
	kref_get(&req->r_kref);
}
extern void ceph_mdsc_release_request(struct kref *kref);
static inline void ceph_mdsc_put_request(struct ceph_mds_request *req)
{
	kref_put(&req->r_kref, ceph_mdsc_release_request);
}

extern void ceph_mdsc_pre_umount(struct ceph_mds_client *mdsc);

extern char *ceph_mdsc_build_path(struct dentry *dentry, int *plen, u64 *base,
				  int stop_on_nosnap);

extern void __ceph_mdsc_drop_dentry_lease(struct dentry *dentry);
extern void ceph_mdsc_lease_send_msg(struct ceph_mds_session *session,
				     struct inode *inode,
				     struct dentry *dentry, char action,
				     u32 seq);

extern void ceph_mdsc_handle_map(struct ceph_mds_client *mdsc,
				 struct ceph_msg *msg);

#endif
Commit	Line	Data
2f2dc053 SW	1	#ifndef _FS_CEPH_MDS_CLIENT_H
	2	#define _FS_CEPH_MDS_CLIENT_H
	3
	4	#include <linux/completion.h>
153c8e6b	5	#include <linux/kref.h>
2f2dc053 SW	6	#include <linux/list.h>
	7	#include <linux/mutex.h>
	8	#include <linux/radix-tree.h>
	9	#include <linux/spinlock.h>
	10
	11	#include "types.h"
	12	#include "messenger.h"
	13	#include "mdsmap.h"
	14
	15	/*
	16	* Some lock dependencies:
	17	*
	18	* session->s_mutex
	19	* mdsc->mutex
	20	*
	21	* mdsc->snap_rwsem
	22	*
	23	* inode->i_lock
	24	* mdsc->snap_flush_lock
	25	* mdsc->cap_delay_lock
	26	*
	27	*/
	28
	29	struct ceph_client;
	30	struct ceph_cap;
	31
	32	/*
	33	* parsed info about a single inode. pointers are into the encoded
	34	* on-wire structures within the mds reply message payload.
	35	*/
	36	struct ceph_mds_reply_info_in {
	37	struct ceph_mds_reply_inode *in;
	38	u32 symlink_len;
	39	char *symlink;
	40	u32 xattr_len;
	41	char *xattr_data;
	42	};
	43
	44	/*
	45	* parsed info about an mds reply, including information about the
	46	* target inode and/or its parent directory and dentry, and directory
	47	* contents (for readdir results).
	48	*/
	49	struct ceph_mds_reply_info_parsed {
	50	struct ceph_mds_reply_head *head;
	51
	52	struct ceph_mds_reply_info_in diri, targeti;
	53	struct ceph_mds_reply_dirfrag *dirfrag;
	54	char *dname;
	55	u32 dname_len;
	56	struct ceph_mds_reply_lease *dlease;
	57
	58	struct ceph_mds_reply_dirfrag *dir_dir;
	59	int dir_nr;
	60	char **dir_dname;
	61	u32 *dir_dname_len;
	62	struct ceph_mds_reply_lease **dir_dlease;
	63	struct ceph_mds_reply_info_in *dir_in;
	64	u8 dir_complete, dir_end;
	65
	66	/* encoded blob describing snapshot contexts for certain
	67	operations (e.g., open) */
	68	void *snapblob;
	69	int snapblob_len;
70	};
71
72
73	/*
74	* cap releases are batched and sent to the MDS en masse.
75	*/
76	#define CEPH_CAPS_PER_RELEASE ((PAGE_CACHE_SIZE - \
77	sizeof(struct ceph_mds_cap_release)) / \
78	sizeof(struct ceph_mds_cap_item))
79
80
81	/*
82	* state associated with each MDS<->client session
83	*/
84	enum {
85	CEPH_MDS_SESSION_NEW = 1,
86	CEPH_MDS_SESSION_OPENING = 2,
87	CEPH_MDS_SESSION_OPEN = 3,
88	CEPH_MDS_SESSION_HUNG = 4,
89	CEPH_MDS_SESSION_CLOSING = 5,
90	CEPH_MDS_SESSION_RESTARTING = 6,
91	CEPH_MDS_SESSION_RECONNECTING = 7,
92	};
93
94	struct ceph_mds_session {
95	struct ceph_mds_client *s_mdsc;
96	int s_mds;
97	int s_state;
98	unsigned long s_ttl; /* time until mds kills us */
99	u64 s_seq; /* incoming msg seq # */
100	struct mutex s_mutex; /* serialize session messages */
101
102	struct ceph_connection s_con;
103
4e7a5dcd SW	104	struct ceph_authorizer *s_authorizer;
	105	void s_authorizer_buf, s_authorizer_reply_buf;
	106	size_t s_authorizer_buf_len, s_authorizer_reply_buf_len;
	107
2f2dc053 SW	108	/* protected by s_cap_lock */
	109	spinlock_t s_cap_lock;
	110	u32 s_cap_gen; /* inc each time we get mds stale msg */
	111	unsigned long s_cap_ttl; /* when session caps expire */
	112	struct list_head s_caps; /* all caps issued by this session */
	113	int s_nr_caps, s_trim_caps;
	114	int s_num_cap_releases;
	115	struct list_head s_cap_releases; /* waiting cap_release messages */
	116	struct list_head s_cap_releases_done; /* ready to send */
5dacf091	117	bool s_iterating_caps;
2f2dc053 SW	118
	119	/* protected by mutex */
	120	struct list_head s_cap_flushing; /* inodes w/ flushing caps */
	121	struct list_head s_cap_snaps_flushing;
	122	unsigned long s_renew_requested; /* last time we sent a renew req */
	123	u64 s_renew_seq;
	124
	125	atomic_t s_ref;
	126	struct list_head s_waiting; /* waiting requests */
	127	struct list_head s_unsafe; /* unsafe requests */
	128	};
	129
	130	/*
	131	* modes of choosing which MDS to send a request to
	132	*/
	133	enum {
	134	USE_ANY_MDS,
	135	USE_RANDOM_MDS,
	136	USE_AUTH_MDS, /* prefer authoritative mds for this metadata item */
	137	};
	138
	139	struct ceph_mds_request;
	140	struct ceph_mds_client;
	141
	142	/*
	143	* request completion callback
	144	*/
	145	typedef void (ceph_mds_request_callback_t) (struct ceph_mds_client mdsc,
	146	struct ceph_mds_request *req);
	147
	148	/*
	149	* an in-flight mds request
	150	*/
	151	struct ceph_mds_request {
	152	u64 r_tid; /* transaction id */
	153
	154	int r_op; /* mds op code */
	155	int r_mds;
	156
	157	/* operation on what? */
	158	struct inode r_inode; / arg1 */
	159	struct dentry r_dentry; / arg1 */
	160	struct dentry r_old_dentry; / arg2: rename from or link from */
	161	char r_path1, r_path2;
	162	struct ceph_vino r_ino1, r_ino2;
	163
	164	struct inode r_locked_dir; / dir (if any) i_mutex locked by vfs */
	165	struct inode r_target_inode; / resulting inode */
	166
	167	union ceph_mds_request_args r_args;
	168	int r_fmode; /* file mode, if expecting cap */
	169
	170	/* for choosing which mds to send this request to */
	171	int r_direct_mode;
	172	u32 r_direct_hash; /* choose dir frag based on this dentry hash */
	173	bool r_direct_is_hash; /* true if r_direct_hash is valid */
	174
	175	/* data payload is used for xattr ops */
	176	struct page **r_pages;
	177	int r_num_pages;
	178	int r_data_len;
	179
	180	/* what caps shall we drop? */
	181	int r_inode_drop, r_inode_unless;
182	int r_dentry_drop, r_dentry_unless;
183	int r_old_dentry_drop, r_old_dentry_unless;
184	struct inode *r_old_inode;
185	int r_old_inode_drop, r_old_inode_unless;
186
187	struct ceph_msg r_request; / original request */
188	struct ceph_msg *r_reply;
189	struct ceph_mds_reply_info_parsed r_reply_info;
190	int r_err;
191
192	unsigned long r_timeout; /* optional. jiffies */
193	unsigned long r_started; /* start time to measure timeout against */
194	unsigned long r_request_started; /* start time for mds request only,
195	used to measure lease durations */
196
197	/* link unsafe requests to parent directory, for fsync */
198	struct inode *r_unsafe_dir;
199	struct list_head r_unsafe_dir_item;
200
201	struct ceph_mds_session *r_session;
202
203	int r_attempts; /* resend attempts */
204	int r_num_fwd; /* number of forward attempts */
205	int r_num_stale;
206	int r_resend_mds; /* mds to resend to next, if any*/
207
153c8e6b	208	struct kref r_kref;
2f2dc053 SW	209	struct list_head r_wait;
	210	struct completion r_completion;
	211	struct completion r_safe_completion;
	212	ceph_mds_request_callback_t r_callback;
	213	struct list_head r_unsafe_item; /* per-session unsafe list item */
	214	bool r_got_unsafe, r_got_safe;
	215
	216	bool r_did_prepopulate;
	217	u32 r_readdir_offset;
	218
	219	struct ceph_cap_reservation r_caps_reservation;
	220	int r_num_caps;
	221	};
	222
	223	/*
	224	* mds client state
	225	*/
	226	struct ceph_mds_client {
	227	struct ceph_client *client;
	228	struct mutex mutex; /* all nested structures */
	229
	230	struct ceph_mdsmap *mdsmap;
	231	struct completion safe_umount_waiters, session_close_waiters;
	232	struct list_head waiting_for_map;
	233
	234	struct ceph_mds_session *sessions; / NULL for mds if no session */
	235	int max_sessions; /* len of s_mds_sessions */
	236	int stopping; /* true if shutting down */
	237
	238	/*
	239	* snap_rwsem will cover cap linkage into snaprealms, and
	240	* realm snap contexts. (later, we can do per-realm snap
	241	* contexts locks..) the empty list contains realms with no
	242	* references (implying they contain no inodes with caps) that
	243	* should be destroyed.
	244	*/
	245	struct rw_semaphore snap_rwsem;
	246	struct radix_tree_root snap_realms;
	247	struct list_head snap_empty;
	248	spinlock_t snap_empty_lock; /* protect snap_empty */
	249
	250	u64 last_tid; /* most recent mds request */
	251	struct radix_tree_root request_tree; /* pending mds requests */
	252	struct delayed_work delayed_work; /* delayed work */
	253	unsigned long last_renew_caps; /* last time we renewed our caps */
	254	struct list_head cap_delay_list; /* caps with delayed release */
	255	spinlock_t cap_delay_lock; /* protects cap_delay_list */
	256	struct list_head snap_flush_list; /* cap_snaps ready to flush */
	257	spinlock_t snap_flush_lock;
	258
	259	u64 cap_flush_seq;
	260	struct list_head cap_dirty; /* inodes with dirty caps */
	261	int num_cap_flushing; /* # caps we are flushing */
	262	spinlock_t cap_dirty_lock; /* protects above items */
	263	wait_queue_head_t cap_flushing_wq;
	264
039934b8	265	#ifdef CONFIG_DEBUG_FS
2f2dc053	266	struct dentry *debugfs_file;
039934b8	267	#endif
2f2dc053 SW	268
	269	spinlock_t dentry_lru_lock;
	270	struct list_head dentry_lru;
	271	int num_dentry;
	272	};
	273
	274	extern const char *ceph_mds_op_name(int op);
	275
	276	extern struct ceph_mds_session *
	277	__ceph_lookup_mds_session(struct ceph_mds_client *, int mds);
	278
	279	static inline struct ceph_mds_session *
	280	ceph_get_mds_session(struct ceph_mds_session *s)
	281	{
	282	atomic_inc(&s->s_ref);
	283	return s;
	284	}
	285
	286	extern void ceph_put_mds_session(struct ceph_mds_session *s);
	287
	288	extern int ceph_send_msg_mds(struct ceph_mds_client *mdsc,
	289	struct ceph_msg *msg, int mds);
	290
5f44f142	291	extern int ceph_mdsc_init(struct ceph_mds_client *mdsc,
2f2dc053 SW	292	struct ceph_client *client);
	293	extern void ceph_mdsc_close_sessions(struct ceph_mds_client *mdsc);
	294	extern void ceph_mdsc_stop(struct ceph_mds_client *mdsc);
	295
	296	extern void ceph_mdsc_sync(struct ceph_mds_client *mdsc);
	297
	298	extern void ceph_mdsc_lease_release(struct ceph_mds_client *mdsc,
	299	struct inode *inode,
	300	struct dentry *dn, int mask);
	301
	302	extern struct ceph_mds_request *
	303	ceph_mdsc_create_request(struct ceph_mds_client *mdsc, int op, int mode);
	304	extern void ceph_mdsc_submit_request(struct ceph_mds_client *mdsc,
	305	struct ceph_mds_request *req);
	306	extern int ceph_mdsc_do_request(struct ceph_mds_client *mdsc,
	307	struct inode *dir,
	308	struct ceph_mds_request *req);
	309	static inline void ceph_mdsc_get_request(struct ceph_mds_request *req)
	310	{
153c8e6b SW	311	kref_get(&req->r_kref);
	312	}
	313	extern void ceph_mdsc_release_request(struct kref *kref);
	314	static inline void ceph_mdsc_put_request(struct ceph_mds_request *req)
	315	{
	316	kref_put(&req->r_kref, ceph_mdsc_release_request);
2f2dc053	317	}
2f2dc053 SW	318
	319	extern void ceph_mdsc_pre_umount(struct ceph_mds_client *mdsc);
	320
	321	extern char ceph_mdsc_build_path(struct dentry dentry, int plen, u64 base,
	322	int stop_on_nosnap);
	323
	324	extern void __ceph_mdsc_drop_dentry_lease(struct dentry *dentry);
	325	extern void ceph_mdsc_lease_send_msg(struct ceph_mds_session *session,
	326	struct inode *inode,
	327	struct dentry *dentry, char action,
	328	u32 seq);
	329
	330	extern void ceph_mdsc_handle_map(struct ceph_mds_client *mdsc,
	331	struct ceph_msg *msg);
	332
	333	#endif