alloc.h 8.7 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28
/* -*- mode: c; c-basic-offset: 8; -*-
 * vim: noexpandtab sw=8 ts=8 sts=0:
 *
 * alloc.h
 *
 * Function prototypes
 *
 * Copyright (C) 2002, 2004 Oracle.  All rights reserved.
 *
 * This program is free software; you can redistribute it and/or
 * modify it under the terms of the GNU General Public
 * License as published by the Free Software Foundation; either
 * version 2 of the License, or (at your option) any later version.
 *
 * 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 for more details.
 *
 * You should have received a copy of the GNU General Public
 * License along with this program; if not, write to the
 * Free Software Foundation, Inc., 59 Temple Place - Suite 330,
 * Boston, MA 021110-1307, USA.
 */

#ifndef OCFS2_ALLOC_H
#define OCFS2_ALLOC_H

29

30 31 32 33 34
/*
 * For xattr tree leaf, we limit the leaf byte size to be 64K.
 */
#define OCFS2_MAX_XATTR_TREE_LEAF_SIZE 65536

35 36 37 38 39 40 41 42 43
/*
 * ocfs2_extent_tree and ocfs2_extent_tree_operations are used to abstract
 * the b-tree operations in ocfs2. Now all the b-tree operations are not
 * limited to ocfs2_dinode only. Any data which need to allocate clusters
 * to store can use b-tree. And it only needs to implement its ocfs2_extent_tree
 * and operation.
 *
 * ocfs2_extent_tree becomes the first-class object for extent tree
 * manipulation.  Callers of the alloc.c code need to fill it via one of
44
 * the ocfs2_init_*_extent_tree() operations below.
45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61
 *
 * ocfs2_extent_tree contains info for the root of the b-tree, it must have a
 * root ocfs2_extent_list and a root_bh so that they can be used in the b-tree
 * functions.
 * ocfs2_extent_tree_operations abstract the normal operations we do for
 * the root of extent b-tree.
 */
struct ocfs2_extent_tree_operations;
struct ocfs2_extent_tree {
	struct ocfs2_extent_tree_operations	*et_ops;
	struct buffer_head			*et_root_bh;
	struct ocfs2_extent_list		*et_root_el;
	void					*et_object;
	unsigned int				et_max_leaf_clusters;
};

/*
62 63
 * ocfs2_init_*_extent_tree() will fill an ocfs2_extent_tree from the
 * specified object buffer.
64
 */
65 66 67 68
void ocfs2_init_dinode_extent_tree(struct ocfs2_extent_tree *et,
				   struct inode *inode,
				   struct buffer_head *bh);
void ocfs2_init_xattr_tree_extent_tree(struct ocfs2_extent_tree *et,
69
				       struct inode *inode,
70 71 72 73 74
				       struct buffer_head *bh);
void ocfs2_init_xattr_value_extent_tree(struct ocfs2_extent_tree *et,
					struct inode *inode,
					struct buffer_head *bh,
					struct ocfs2_xattr_value_root *xv);
75

76 77 78 79 80 81 82 83
/*
 * Read an extent block into *bh.  If *bh is NULL, a bh will be
 * allocated.  This is a cached read.  The extent block will be validated
 * with ocfs2_validate_extent_block().
 */
int ocfs2_read_extent_block(struct inode *inode, u64 eb_blkno,
			    struct buffer_head **bh);

84
struct ocfs2_alloc_context;
85 86 87 88 89 90 91 92 93 94
int ocfs2_insert_extent(struct ocfs2_super *osb,
			handle_t *handle,
			struct inode *inode,
			struct ocfs2_extent_tree *et,
			u32 cpos,
			u64 start_blk,
			u32 new_clusters,
			u8 flags,
			struct ocfs2_alloc_context *meta_ac);

95 96 97 98 99 100 101 102 103 104
enum ocfs2_alloc_restarted {
	RESTART_NONE = 0,
	RESTART_TRANS,
	RESTART_META
};
int ocfs2_add_clusters_in_btree(struct ocfs2_super *osb,
				struct inode *inode,
				u32 *logical_offset,
				u32 clusters_to_add,
				int mark_unwritten,
105
				struct ocfs2_extent_tree *et,
106 107 108
				handle_t *handle,
				struct ocfs2_alloc_context *data_ac,
				struct ocfs2_alloc_context *meta_ac,
109
				enum ocfs2_alloc_restarted *reason_ret);
110
struct ocfs2_cached_dealloc_ctxt;
111 112
int ocfs2_mark_extent_written(struct inode *inode,
			      struct ocfs2_extent_tree *et,
113 114
			      handle_t *handle, u32 cpos, u32 len, u32 phys,
			      struct ocfs2_alloc_context *meta_ac,
115 116 117
			      struct ocfs2_cached_dealloc_ctxt *dealloc);
int ocfs2_remove_extent(struct inode *inode,
			struct ocfs2_extent_tree *et,
118 119
			u32 cpos, u32 len, handle_t *handle,
			struct ocfs2_alloc_context *meta_ac,
120
			struct ocfs2_cached_dealloc_ctxt *dealloc);
121 122 123 124 125
int ocfs2_remove_btree_range(struct inode *inode,
			     struct ocfs2_extent_tree *et,
			     u32 cpos, u32 phys_cpos, u32 len,
			     struct ocfs2_cached_dealloc_ctxt *dealloc);

126 127
int ocfs2_num_free_extents(struct ocfs2_super *osb,
			   struct inode *inode,
128
			   struct ocfs2_extent_tree *et);
129

130 131 132 133 134 135 136 137
/*
 * how many new metadata chunks would an allocation need at maximum?
 *
 * Please note that the caller must make sure that root_el is the root
 * of extent tree. So for an inode, it should be &fe->id2.i_list. Otherwise
 * the result may be wrong.
 */
static inline int ocfs2_extend_meta_needed(struct ocfs2_extent_list *root_el)
138 139 140 141 142 143 144 145 146
{
	/*
	 * Rather than do all the work of determining how much we need
	 * (involves a ton of reads and locks), just ask for the
	 * maximal limit.  That's a tree depth shift.  So, one block for
	 * level of the tree (current l_tree_depth), one block for the
	 * new tree_depth==0 extent_block, and one block at the new
	 * top-of-the tree.
	 */
147
	return le16_to_cpu(root_el->l_tree_depth) + 2;
148 149
}

150
void ocfs2_dinode_new_extent_list(struct inode *inode, struct ocfs2_dinode *di);
M
Mark Fasheh 已提交
151 152 153 154
void ocfs2_set_inode_data_inline(struct inode *inode, struct ocfs2_dinode *di);
int ocfs2_convert_inline_data_to_extents(struct inode *inode,
					 struct buffer_head *di_bh);

155 156 157 158 159 160 161 162 163 164
int ocfs2_truncate_log_init(struct ocfs2_super *osb);
void ocfs2_truncate_log_shutdown(struct ocfs2_super *osb);
void ocfs2_schedule_truncate_log_flush(struct ocfs2_super *osb,
				       int cancel);
int ocfs2_flush_truncate_log(struct ocfs2_super *osb);
int ocfs2_begin_truncate_log_recovery(struct ocfs2_super *osb,
				      int slot_num,
				      struct ocfs2_dinode **tl_copy);
int ocfs2_complete_truncate_log_recovery(struct ocfs2_super *osb,
					 struct ocfs2_dinode *tl_copy);
165 166 167 168 169 170
int ocfs2_truncate_log_needs_flush(struct ocfs2_super *osb);
int ocfs2_truncate_log_append(struct ocfs2_super *osb,
			      handle_t *handle,
			      u64 start_blk,
			      unsigned int num_clusters);
int __ocfs2_flush_truncate_log(struct ocfs2_super *osb);
171

172 173 174 175 176 177 178 179 180 181 182
/*
 * Process local structure which describes the block unlinks done
 * during an operation. This is populated via
 * ocfs2_cache_block_dealloc().
 *
 * ocfs2_run_deallocs() should be called after the potentially
 * de-allocating routines. No journal handles should be open, and most
 * locks should have been dropped.
 */
struct ocfs2_cached_dealloc_ctxt {
	struct ocfs2_per_slot_free_list		*c_first_suballocator;
183
	struct ocfs2_cached_block_free 		*c_global_allocator;
184 185 186 187
};
static inline void ocfs2_init_dealloc_ctxt(struct ocfs2_cached_dealloc_ctxt *c)
{
	c->c_first_suballocator = NULL;
188
	c->c_global_allocator = NULL;
189
}
190 191
int ocfs2_cache_cluster_dealloc(struct ocfs2_cached_dealloc_ctxt *ctxt,
				u64 blkno, unsigned int bit);
192 193 194 195
static inline int ocfs2_dealloc_has_cluster(struct ocfs2_cached_dealloc_ctxt *c)
{
	return c->c_global_allocator != NULL;
}
196 197 198
int ocfs2_run_deallocs(struct ocfs2_super *osb,
		       struct ocfs2_cached_dealloc_ctxt *ctxt);

199
struct ocfs2_truncate_context {
200
	struct ocfs2_cached_dealloc_ctxt tc_dealloc;
201 202 203 204 205
	int tc_ext_alloc_locked; /* is it cluster locked? */
	/* these get destroyed once it's passed to ocfs2_commit_truncate. */
	struct buffer_head *tc_last_eb_bh;
};

206 207
int ocfs2_zero_range_for_truncate(struct inode *inode, handle_t *handle,
				  u64 range_start, u64 range_end);
208 209 210 211 212 213 214 215
int ocfs2_prepare_truncate(struct ocfs2_super *osb,
			   struct inode *inode,
			   struct buffer_head *fe_bh,
			   struct ocfs2_truncate_context **tc);
int ocfs2_commit_truncate(struct ocfs2_super *osb,
			  struct inode *inode,
			  struct buffer_head *fe_bh,
			  struct ocfs2_truncate_context *tc);
M
Mark Fasheh 已提交
216 217
int ocfs2_truncate_inline(struct inode *inode, struct buffer_head *di_bh,
			  unsigned int start, unsigned int end, int trunc);
218

219 220
int ocfs2_find_leaf(struct inode *inode, struct ocfs2_extent_list *root_el,
		    u32 cpos, struct buffer_head **leaf_bh);
221
int ocfs2_search_extent_list(struct ocfs2_extent_list *el, u32 v_cluster);
222

223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241
/*
 * Helper function to look at the # of clusters in an extent record.
 */
static inline unsigned int ocfs2_rec_clusters(struct ocfs2_extent_list *el,
					      struct ocfs2_extent_rec *rec)
{
	/*
	 * Cluster count in extent records is slightly different
	 * between interior nodes and leaf nodes. This is to support
	 * unwritten extents which need a flags field in leaf node
	 * records, thus shrinking the available space for a clusters
	 * field.
	 */
	if (el->l_tree_depth)
		return le32_to_cpu(rec->e_int_clusters);
	else
		return le16_to_cpu(rec->e_leaf_clusters);
}

M
Mark Fasheh 已提交
242 243 244 245 246 247 248 249 250
/*
 * This is only valid for leaf nodes, which are the only ones that can
 * have empty extents anyway.
 */
static inline int ocfs2_is_empty_extent(struct ocfs2_extent_rec *rec)
{
	return !rec->e_leaf_clusters;
}

251
#endif /* OCFS2_ALLOC_H */