-/* -*- mode: c; c-basic-offset: 8; indent-tabs-mode: nil; -*-
- * vim:expandtab:shiftwidth=8:tabstop=8:
- *
+/*
* GPL HEADER START
*
* DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
* GPL HEADER END
*/
/*
- * Copyright 2008 Sun Microsystems, Inc. All rights reserved
+ * Copyright (c) 2007, 2010, Oracle and/or its affiliates. All rights reserved.
* Use is subject to license terms.
+ *
+ * Copyright (c) 2011, 2012, Intel Corporation.
*/
/*
* This file is part of Lustre, http://www.lustre.org/
};
/**
+ * For lu_object_conf flags
+ */
+typedef enum {
+ /* This is a new object to be allocated, or the file
+ * corresponding to the object does not exists. */
+ LOC_F_NEW = 0x00000001,
+} loc_flags_t;
+
+/**
* Object configuration, describing particulars of object being created. On
* server this is not used, as server objects are full identified by fid. On
* client configuration contains struct lustre_md.
*/
struct lu_object_conf {
+ /**
+ * Some hints for obj find and alloc.
+ */
+ loc_flags_t loc_flags;
};
/**
* A list of references to this object, for debugging.
*/
struct lu_ref ld_reference;
+ /**
+ * Link the device to the site.
+ **/
+ cfs_list_t ld_linkage;
};
struct lu_device_type_operations;
/** size in bytes */
__u64 la_size;
/** modification time in seconds since Epoch */
- __u64 la_mtime;
+ obd_time la_mtime;
/** access time in seconds since Epoch */
- __u64 la_atime;
+ obd_time la_atime;
/** change time in seconds since Epoch */
- __u64 la_ctime;
+ obd_time la_ctime;
/** 512-byte blocks allocated to object */
__u64 la_blocks;
/** permission bits and file type */
LA_NLINK = 1 << 10,
LA_RDEV = 1 << 11,
LA_BLKSIZE = 1 << 12,
+ LA_KILL_SUID = 1 << 13,
+ LA_KILL_SGID = 1 << 14,
};
/**
struct fld;
+struct lu_site_bkt_data {
+ /**
+ * number of busy object on this bucket
+ */
+ long lsb_busy;
+ /**
+ * LRU list, updated on each access to object. Protected by
+ * bucket lock of lu_site::ls_obj_hash.
+ *
+ * "Cold" end of LRU is lu_site::ls_lru.next. Accessed object are
+ * moved to the lu_site::ls_lru.prev (this is due to the non-existence
+ * of list_for_each_entry_safe_reverse()).
+ */
+ cfs_list_t lsb_lru;
+ /**
+ * Wait-queue signaled when an object in this site is ultimately
+ * destroyed (lu_object_free()). It is used by lu_object_find() to
+ * wait before re-trying when object in the process of destruction is
+ * found in the hash table.
+ *
+ * \see htable_lookup().
+ */
+ cfs_waitq_t lsb_marche_funebre;
+};
+
+enum {
+ LU_SS_CREATED = 0,
+ LU_SS_CACHE_HIT,
+ LU_SS_CACHE_MISS,
+ LU_SS_CACHE_RACE,
+ LU_SS_CACHE_DEATH_RACE,
+ LU_SS_LRU_PURGED,
+ LU_SS_LAST_STAT
+};
+
/**
* lu_site is a "compartment" within which objects are unique, and LRU
* discipline is maintained.
*/
struct lu_site {
/**
- * Site-wide lock.
- *
- * lock protecting:
- *
- * - lu_site::ls_hash hash table (and its linkages in objects);
- *
- * - lu_site::ls_lru list (and its linkages in objects);
- *
- * - 0/1 transitions of object lu_object_header::loh_ref
- * reference count;
- *
- * yes, it's heavy.
- */
- cfs_rwlock_t ls_guard;
- /**
- * Hash-table where objects are indexed by fid.
- */
- cfs_hlist_head_t *ls_hash;
- /**
- * Bit-mask for hash-table size.
- */
- int ls_hash_mask;
- /**
- * Order of hash-table.
- */
- int ls_hash_bits;
- /**
- * Number of buckets in the hash-table.
- */
- int ls_hash_size;
-
- /**
- * LRU list, updated on each access to object. Protected by
- * lu_site::ls_guard.
- *
- * "Cold" end of LRU is lu_site::ls_lru.next. Accessed object are
- * moved to the lu_site::ls_lru.prev (this is due to the non-existence
- * of list_for_each_entry_safe_reverse()).
- */
- cfs_list_t ls_lru;
- /**
- * Total number of objects in this site. Protected by
- * lu_site::ls_guard.
+ * objects hash table
*/
- unsigned ls_total;
+ cfs_hash_t *ls_obj_hash;
/**
- * Total number of objects in this site with reference counter greater
- * than 0. Protected by lu_site::ls_guard.
+ * index of bucket on hash table while purging
*/
- unsigned ls_busy;
-
+ int ls_purge_start;
/**
* Top-level device for this stack.
*/
struct lu_device *ls_top_dev;
-
- /**
- * Wait-queue signaled when an object in this site is ultimately
- * destroyed (lu_object_free()). It is used by lu_object_find() to
- * wait before re-trying when object in the process of destruction is
- * found in the hash table.
- *
- * If having a single wait-queue turns out to be a problem, a
- * wait-queue per hash-table bucket can be easily implemented.
- *
- * \see htable_lookup().
- */
- cfs_waitq_t ls_marche_funebre;
-
- /** statistical counters. Protected by nothing, races are accepted. */
- struct {
- __u32 s_created;
- __u32 s_cache_hit;
- __u32 s_cache_miss;
- /**
- * Number of hash-table entry checks made.
- *
- * ->s_cache_check / (->s_cache_miss + ->s_cache_hit)
- *
- * is an average number of hash slots inspected during single
- * lookup.
- */
- __u32 s_cache_check;
- /** Races with cache insertions. */
- __u32 s_cache_race;
- /**
- * Races with object destruction.
- *
- * \see lu_site::ls_marche_funebre.
- */
- __u32 s_cache_death_race;
- __u32 s_lru_purged;
- } ls_stats;
-
+ /**
+ * Bottom-level device for this stack
+ */
+ struct lu_device *ls_bottom_dev;
/**
* Linkage into global list of sites.
*/
cfs_list_t ls_linkage;
- struct lprocfs_stats *ls_time_stats;
+ /**
+ * List for lu device for this site, protected
+ * by ls_ld_lock.
+ **/
+ cfs_list_t ls_ld_linkage;
+ spinlock_t ls_ld_lock;
+
+ /**
+ * lu_site stats
+ */
+ struct lprocfs_stats *ls_stats;
+ /**
+ * XXX: a hack! fld has to find md_site via site, remove when possible
+ */
+ struct seq_server_site *ld_seq_site;
};
+static inline struct lu_site_bkt_data *
+lu_site_bkt_from_fid(struct lu_site *site, struct lu_fid *fid)
+{
+ cfs_hash_bd_t bd;
+
+ cfs_hash_bd_get(site->ls_obj_hash, fid, &bd);
+ return cfs_hash_bd_extra_get(site->ls_obj_hash, &bd);
+}
+
/** \name ctors
* Constructors/destructors.
* @{
void lu_object_add_top (struct lu_object_header *h, struct lu_object *o);
void lu_object_add (struct lu_object *before, struct lu_object *o);
+void lu_dev_add_linkage(struct lu_site *s, struct lu_device *d);
+void lu_dev_del_linkage(struct lu_site *s, struct lu_device *d);
+
/**
* Helpers to initialize and finalize device types.
*/
*/
static inline int lu_object_is_dying(const struct lu_object_header *h)
{
- return cfs_test_bit(LU_OBJECT_HEARD_BANSHEE, &h->loh_flags);
+ return test_bit(LU_OBJECT_HEARD_BANSHEE, &h->loh_flags);
}
void lu_object_put(const struct lu_env *env, struct lu_object *o);
+void lu_object_put_nocache(const struct lu_env *env, struct lu_object *o);
int lu_site_purge(const struct lu_env *env, struct lu_site *s, int nr);
struct lu_object *lu_object_locate(struct lu_object_header *h,
const struct lu_device_type *dtype);
-struct lu_cdebug_print_info {
- int lpi_subsys;
- int lpi_mask;
- const char *lpi_file;
- const char *lpi_fn;
- int lpi_line;
-};
-
/**
* Printer function emitting messages through libcfs_debug_msg().
*/
int lu_cdebug_printer(const struct lu_env *env,
void *cookie, const char *format, ...);
-#define DECLARE_LU_CDEBUG_PRINT_INFO(var, mask) \
- struct lu_cdebug_print_info var = { \
- .lpi_subsys = DEBUG_SUBSYSTEM, \
- .lpi_mask = (mask), \
- .lpi_file = __FILE__, \
- .lpi_fn = __FUNCTION__, \
- .lpi_line = __LINE__ \
- }
-
/**
* Print object description followed by a user-supplied message.
*/
-#define LU_OBJECT_DEBUG(mask, env, object, format, ...) \
-do { \
- static DECLARE_LU_CDEBUG_PRINT_INFO(__info, mask); \
- \
- if (cfs_cdebug_show(mask, DEBUG_SUBSYSTEM)) { \
- lu_object_print(env, &__info, lu_cdebug_printer, object); \
- CDEBUG(mask, format , ## __VA_ARGS__); \
- } \
+#define LU_OBJECT_DEBUG(mask, env, object, format, ...) \
+do { \
+ LIBCFS_DEBUG_MSG_DATA_DECL(msgdata, mask, NULL); \
+ \
+ if (cfs_cdebug_show(mask, DEBUG_SUBSYSTEM)) { \
+ lu_object_print(env, &msgdata, lu_cdebug_printer, object);\
+ CDEBUG(mask, format , ## __VA_ARGS__); \
+ } \
} while (0)
/**
*/
#define LU_OBJECT_HEADER(mask, env, object, format, ...) \
do { \
- static DECLARE_LU_CDEBUG_PRINT_INFO(__info, mask); \
+ LIBCFS_DEBUG_MSG_DATA_DECL(msgdata, mask, NULL); \
\
if (cfs_cdebug_show(mask, DEBUG_SUBSYSTEM)) { \
- lu_object_header_print(env, &__info, lu_cdebug_printer, \
+ lu_object_header_print(env, &msgdata, lu_cdebug_printer,\
(object)->lo_header); \
- lu_cdebug_printer(env, &__info, "\n"); \
+ lu_cdebug_printer(env, &msgdata, "\n"); \
CDEBUG(mask, format , ## __VA_ARGS__); \
} \
} while (0)
/**
* \retval 1 iff object \a o exists on stable storage,
+ * \retval 0 iff object \a o not exists on stable storage.
* \retval -1 iff object \a o is on remote server.
*/
static inline int lu_object_exists(const struct lu_object *o)
* from enum lu_context_tag.
*/
__u32 lc_tags;
+ enum lu_context_state lc_state;
/**
* Pointer to the home service thread. NULL for other execution
* contexts.
* detail.
*/
void **lc_value;
- enum lu_context_state lc_state;
/**
* Linkage into a list of all remembered contexts. Only
* `non-transient' contexts, i.e., ones created for service threads
* a client.
*/
LCT_SESSION = 1 << 4,
-
+ /**
+ * A per-request data on OSP device
+ */
+ LCT_OSP_THREAD = 1 << 5,
+ /**
+ * MGS device thread
+ */
+ LCT_MG_THREAD = 1 << 6,
+ /**
+ * Context for local operations
+ */
+ LCT_LOCAL = 1 << 7,
/**
* Set when at least one of keys, having values in this context has
* non-NULL lu_context_key::lct_exit() method. This is used to
void lu_context_key_revive_many (struct lu_context_key *k, ...);
void lu_context_key_quiesce_many (struct lu_context_key *k, ...);
+/*
+ * update/clear ctx/ses tags.
+ */
+void lu_context_tags_update(__u32 tags);
+void lu_context_tags_clear(__u32 tags);
+void lu_session_tags_update(__u32 tags);
+void lu_session_tags_clear(__u32 tags);
+
/**
* Environment.
*/
int lu_env_init (struct lu_env *env, __u32 tags);
void lu_env_fini (struct lu_env *env);
int lu_env_refill(struct lu_env *env);
+int lu_env_refill_by_tags(struct lu_env *env, __u32 ctags, __u32 stags);
/** @} lu_context */
+struct lu_ucred {
+ __u32 uc_valid;
+ __u32 uc_o_uid;
+ __u32 uc_o_gid;
+ __u32 uc_o_fsuid;
+ __u32 uc_o_fsgid;
+ __u32 uc_uid;
+ __u32 uc_gid;
+ __u32 uc_fsuid;
+ __u32 uc_fsgid;
+ __u32 uc_suppgids[2];
+ cfs_cap_t uc_cap;
+ __u32 uc_umask;
+ cfs_group_info_t *uc_ginfo;
+ struct md_identity *uc_identity;
+};
+struct lu_ucred *lu_ucred(const struct lu_env *env);
+struct lu_ucred *lu_ucred_check(const struct lu_env *env);
+struct lu_ucred *lu_ucred_assert(const struct lu_env *env);
+
/**
* Output site statistical counters into a buffer. Suitable for
* ll_rd_*()-style functions.
struct lu_buf {
void *lb_buf;
ssize_t lb_len;
- int lb_vmalloc:1;
};
/** null buffer */
extern struct lu_buf LU_BUF_NULL;
-#define DLUBUF "(%p %z)"
+#define DLUBUF "(%p %zu)"
#define PLUBUF(buf) (buf)->lb_buf, (buf)->lb_len
/**
* One-time initializers, called at obdclass module initialization, not
*/
void lu_global_fini(void);
-enum {
- LU_TIME_FIND_LOOKUP,
- LU_TIME_FIND_ALLOC,
- LU_TIME_FIND_INSERT,
- LU_TIME_NR
-};
-
-extern const char *lu_time_names[LU_TIME_NR];
-
struct lu_kmem_descr {
cfs_mem_cache_t **ckd_cache;
const char *ckd_name;
int lu_kmem_init(struct lu_kmem_descr *caches);
void lu_kmem_fini(struct lu_kmem_descr *caches);
+void lu_object_assign_fid(const struct lu_env *env, struct lu_object *o,
+ const struct lu_fid *fid);
+struct lu_object *lu_object_anon(const struct lu_env *env,
+ struct lu_device *dev,
+ const struct lu_object_conf *conf);
+
/** @} lu */
#endif /* __LUSTRE_LU_OBJECT_H */