-/* -*- MODE: c; c-basic-offset: 8; indent-tabs-mode: nil; -*-
- * vim:expandtab:shiftwidth=8:tabstop=8:
+/*
+ * GPL HEADER START
*
- * fld/fld_internal.h
+ * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
*
- * Copyright (C) 2006 Cluster File Systems, Inc.
- * Author: Yury Umanets <umka@clusterfs.com>
- * Tom WangDi <wangdi@clusterfs.com>
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 only,
+ * as published by the Free Software Foundation.
*
- * This file is part of the Lustre file system, http://www.lustre.org
- * Lustre is a trademark of Cluster File Systems, Inc.
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License version 2 for more details (a copy is included
+ * in the LICENSE file that accompanied this code).
*
- * You may have signed or agreed to another license before downloading
- * this software. If so, you are bound by the terms and conditions
- * of that agreement, and the following does not apply to you. See the
- * LICENSE file included with this distribution for more information.
+ * You should have received a copy of the GNU General Public License
+ * version 2 along with this program; If not, see
+ * http://www.sun.com/software/products/lustre/docs/GPLv2.pdf
*
- * If you did not agree to a different license, then this copy of Lustre
- * is open source software; you can redistribute it and/or modify it
- * under the terms of version 2 of the GNU General Public License as
- * published by the Free Software Foundation.
+ * Please contact Sun Microsystems, Inc., 4150 Network Circle, Santa Clara,
+ * CA 95054 USA or visit www.sun.com if you need additional information or
+ * have any questions.
*
- * In either case, Lustre is distributed in the hope that it will be
- * useful, but WITHOUT ANY WARRANTY; without even the implied warranty
- * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * license text for more details.
+ * GPL HEADER END
+ */
+/*
+ * Copyright (c) 2007, 2010, Oracle and/or its affiliates. All rights reserved.
+ * Use is subject to license terms.
+ *
+ * Copyright (c) 2012, 2013, Intel Corporation.
+ */
+/*
+ * This file is part of Lustre, http://www.lustre.org/
+ * Lustre is a trademark of Sun Microsystems, Inc.
+ *
+ * lustre/fld/fld_internal.h
+ *
+ * Subsystem Description:
+ * FLD is FID Location Database, which stores where (IE, on which MDT)
+ * FIDs are located.
+ * The database is basically a record file, each record consists of a FID
+ * sequence range, MDT/OST index, and flags. The FLD for the whole FS
+ * is only stored on the sequence controller(MDT0) right now, but each target
+ * also has its local FLD, which only stores the local sequence.
+ *
+ * The FLD subsystem usually has two tasks:
+ * 1. maintain the database, i.e. when the sequence controller allocates
+ * new sequence ranges to some nodes, it will call the FLD API to insert the
+ * location information <sequence_range, node_index> in FLDB.
+ *
+ * 2. Handle requests from other nodes, i.e. if client needs to know where
+ * the FID is located, if it can not find the information in the local cache,
+ * it will send a FLD lookup RPC to the FLD service, and the FLD service will
+ * look up the FLDB entry and return the location information to client.
+ *
+ * Author: Yury Umanets <umka@clusterfs.com>
+ * Author: Tom WangDi <wangdi@clusterfs.com>
*/
#ifndef __FLD_INTERNAL_H
#define __FLD_INTERNAL_H
+#include <obd.h>
#include <lustre/lustre_idl.h>
-#include <dt_object.h>
+#include <libcfs/libcfs.h>
+#include <lustre_fld.h>
-#include <libcfs/list.h>
-#include <libcfs/kp30.h>
+enum {
+ LUSTRE_FLD_INIT = 1 << 0,
+ LUSTRE_FLD_RUN = 1 << 1
+};
-#include <linux/types.h>
-#include <lustre_req_layout.h>
-#include <lustre_fld.h>
+struct fld_stats {
+ __u64 fst_count;
+ __u64 fst_cache;
+};
+
+typedef int (*fld_hash_func_t) (struct lu_client_fld *, __u64);
+
+typedef struct lu_fld_target *
+(*fld_scan_func_t) (struct lu_client_fld *, __u64);
+
+struct lu_fld_hash {
+ const char *fh_name;
+ fld_hash_func_t fh_hash_func;
+ fld_scan_func_t fh_scan_func;
+};
+
+struct fld_cache_entry {
+ struct list_head fce_lru;
+ struct list_head fce_list;
+ /**
+ * fld cache entries are sorted on range->lsr_start field. */
+ struct lu_seq_range fce_range;
+};
-enum fld_op {
- FLD_CREATE = 0,
- FLD_DELETE = 1,
- FLD_LOOKUP = 2
+struct fld_cache {
+ /**
+ * Cache guard, protects fci_hash mostly because others immutable after
+ * init is finished.
+ */
+ rwlock_t fci_lock;
+
+ /**
+ * Cache shrink threshold */
+ int fci_threshold;
+
+ /**
+ * Prefered number of cached entries */
+ int fci_cache_size;
+
+ /**
+ * Current number of cached entries. Protected by \a fci_lock */
+ int fci_cache_count;
+
+ /**
+ * LRU list fld entries. */
+ struct list_head fci_lru;
+
+ /**
+ * sorted fld entries. */
+ struct list_head fci_entries_head;
+
+ /**
+ * Cache statistics. */
+ struct fld_stats fci_stat;
+
+ /**
+ * Cache name used for debug and messages. */
+ char fci_name[80];
+ unsigned int fci_no_shrink:1;
};
enum {
FLD_CLIENT_CACHE_THRESHOLD = 10
};
-enum {
- /*
- * One page is used for hashtable. That is sizeof(struct hlist_head) *
- * 1024.
- */
- FLD_CLIENT_HTABLE_SIZE = (1024 * 1),
-
- /*
- * Here 4 pages are used for hashtable of server cache. This is is
- * because cache it self is 4 times bugger.
- */
- FLD_SERVER_HTABLE_SIZE = (1024 * 4)
-};
-
extern struct lu_fld_hash fld_hash[];
-#ifdef __KERNEL__
+
+#ifdef LPROCFS
+extern struct proc_dir_entry *fld_type_proc_dir;
+extern struct lprocfs_seq_vars fld_client_proc_list[];
+#endif
+
+# ifdef HAVE_SERVER_SUPPORT
struct fld_thread_info {
- struct req_capsule fti_pill;
- int fti_rep_buf_size[REQ_MAX_FIELD_NR];
- __u64 fti_key;
- __u64 fti_rec;
- __u32 fti_flags;
+ struct lu_seq_range fti_rec;
+ struct lu_seq_range fti_lrange;
+ struct lu_seq_range fti_irange;
};
-int fld_index_init(struct lu_server_fld *fld,
- const struct lu_env *env,
- struct dt_device *dt);
+extern struct lu_context_key fld_thread_key;
+
+struct dt_device;
+int fld_index_init(const struct lu_env *env, struct lu_server_fld *fld,
+ struct dt_device *dt, int type);
-void fld_index_fini(struct lu_server_fld *fld,
- const struct lu_env *env);
+void fld_index_fini(const struct lu_env *env, struct lu_server_fld *fld);
-int fld_index_create(struct lu_server_fld *fld,
- const struct lu_env *env,
- seqno_t seq, mdsno_t mds);
+int fld_declare_index_create(const struct lu_env *env,
+ struct lu_server_fld *fld,
+ const struct lu_seq_range *new_range,
+ struct thandle *th);
-int fld_index_delete(struct lu_server_fld *fld,
- const struct lu_env *env,
- seqno_t seq);
+int fld_index_create(const struct lu_env *env, struct lu_server_fld *fld,
+ const struct lu_seq_range *new_range, struct thandle *th);
-int fld_index_lookup(struct lu_server_fld *fld,
- const struct lu_env *env,
- seqno_t seq, mdsno_t *mds);
+int fld_index_lookup(const struct lu_env *env, struct lu_server_fld *fld,
+ u64 seq, struct lu_seq_range *range);
+int fld_name_to_index(const char *name, __u32 *index);
+int fld_server_mod_init(void);
+
+void fld_server_mod_exit(void);
+
+int fld_server_read(const struct lu_env *env, struct lu_server_fld *fld,
+ struct lu_seq_range *range, void *data, int data_len);
#ifdef LPROCFS
-extern struct lprocfs_vars fld_server_proc_list[];
-extern struct lprocfs_vars fld_client_proc_list[];
+extern const struct file_operations fld_proc_seq_fops;
+extern struct lprocfs_seq_vars fld_server_proc_list[];
#endif
-#endif
+# endif /* HAVE_SERVER_SUPPORT */
+
+int fld_client_rpc(struct obd_export *exp,
+ struct lu_seq_range *range, __u32 fld_op,
+ struct ptlrpc_request **reqp);
+
+struct fld_cache *fld_cache_init(const char *name,
+ int cache_size, int cache_threshold);
+
+void fld_cache_fini(struct fld_cache *cache);
+
+void fld_cache_flush(struct fld_cache *cache);
+
+int fld_cache_insert(struct fld_cache *cache,
+ const struct lu_seq_range *range);
+
+struct fld_cache_entry
+*fld_cache_entry_create(const struct lu_seq_range *range);
+
+int fld_cache_insert_nolock(struct fld_cache *cache,
+ struct fld_cache_entry *f_new);
+void fld_cache_delete(struct fld_cache *cache,
+ const struct lu_seq_range *range);
+void fld_cache_delete_nolock(struct fld_cache *cache,
+ const struct lu_seq_range *range);
+int fld_cache_lookup(struct fld_cache *cache,
+ const u64 seq, struct lu_seq_range *range);
+
+struct fld_cache_entry *
+fld_cache_entry_lookup(struct fld_cache *cache,
+ const struct lu_seq_range *range);
+
+void fld_cache_entry_delete(struct fld_cache *cache,
+ struct fld_cache_entry *node);
+
+struct fld_cache_entry *
+fld_cache_entry_lookup_nolock(struct fld_cache *cache,
+ const struct lu_seq_range *range);
static inline const char *
-fld_target_name(struct lu_fld_target *tar)
+fld_target_name(const struct lu_fld_target *tar)
{
- if (tar->ft_srv != NULL)
- return tar->ft_srv->lsf_name;
+#ifdef HAVE_SERVER_SUPPORT
+ if (tar->ft_srv != NULL)
+ return tar->ft_srv->lsf_name;
+#endif
- return (const char *)tar->ft_exp->exp_obd->obd_name;
+ return tar->ft_exp->exp_obd->obd_name;
}
-extern cfs_proc_dir_entry_t *fld_type_proc_dir;
-
#endif /* __FLD_INTERNAL_H */