[net-next-2.6.git] / fs / ext4 / ext4_extents.h

/*
 * Copyright (c) 2003-2006, Cluster File Systems, Inc, info@clusterfs.com
 * Written by Alex Tomas <alex@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 as
 * published by the Free Software Foundation.
 *
 * 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 Licens
 * along with this program; if not, write to the Free Software
 * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-
 */

#ifndef _EXT4_EXTENTS
#define _EXT4_EXTENTS

#include "ext4.h"

/*
 * With AGGRESSIVE_TEST defined, the capacity of index/leaf blocks
 * becomes very small, so index split, in-depth growing and
 * other hard changes happen much more often.
 * This is for debug purposes only.
 */
#define AGGRESSIVE_TEST_

/*
 * With EXTENTS_STATS defined, the number of blocks and extents
 * are collected in the truncate path. They'll be shown at
 * umount time.
 */
#define EXTENTS_STATS__

/*
 * If CHECK_BINSEARCH is defined, then the results of the binary search
 * will also be checked by linear search.
 */
#define CHECK_BINSEARCH__

/*
 * If EXT_DEBUG is defined you can use the 'extdebug' mount option
 * to get lots of info about what's going on.
 */
#define EXT_DEBUG__
#ifdef EXT_DEBUG
#define ext_debug(a...)		printk(a)
#else
#define ext_debug(a...)
#endif

/*
 * If EXT_STATS is defined then stats numbers are collected.
 * These number will be displayed at umount time.
 */
#define EXT_STATS_


/*
 * ext4_inode has i_block array (60 bytes total).
 * The first 12 bytes store ext4_extent_header;
 * the remainder stores an array of ext4_extent.
 */

/*
 * This is the extent on-disk structure.
 * It's used at the bottom of the tree.
 */
struct ext4_extent {
	__le32	ee_block;	/* first logical block extent covers */
	__le16	ee_len;		/* number of blocks covered by extent */
	__le16	ee_start_hi;	/* high 16 bits of physical block */
	__le32	ee_start_lo;	/* low 32 bits of physical block */
};

/*
 * This is index on-disk structure.
 * It's used at all the levels except the bottom.
 */
struct ext4_extent_idx {
	__le32	ei_block;	/* index covers logical blocks from 'block' */
	__le32	ei_leaf_lo;	/* pointer to the physical block of the next *
				 * level. leaf or next index could be there */
	__le16	ei_leaf_hi;	/* high 16 bits of physical block */
	__u16	ei_unused;
};

/*
 * Each block (leaves and indexes), even inode-stored has header.
 */
struct ext4_extent_header {
	__le16	eh_magic;	/* probably will support different formats */
	__le16	eh_entries;	/* number of valid entries */
	__le16	eh_max;		/* capacity of store in entries */
	__le16	eh_depth;	/* has tree real underlying blocks? */
	__le32	eh_generation;	/* generation of the tree */
};

#define EXT4_EXT_MAGIC		cpu_to_le16(0xf30a)

/*
 * Array of ext4_ext_path contains path to some extent.
 * Creation/lookup routines use it for traversal/splitting/etc.
 * Truncate uses it to simulate recursive walking.
 */
struct ext4_ext_path {
	ext4_fsblk_t			p_block;
	__u16				p_depth;
	struct ext4_extent		*p_ext;
	struct ext4_extent_idx		*p_idx;
	struct ext4_extent_header	*p_hdr;
	struct buffer_head		*p_bh;
};

/*
 * structure for external API
 */

#define EXT4_EXT_CACHE_NO	0
#define EXT4_EXT_CACHE_GAP	1
#define EXT4_EXT_CACHE_EXTENT	2

/*
 * to be called by ext4_ext_walk_space()
 * negative retcode - error
 * positive retcode - signal for ext4_ext_walk_space(), see below
 * callback must return valid extent (passed or newly created)
 */
typedef int (*ext_prepare_callback)(struct inode *, struct ext4_ext_path *,
					struct ext4_ext_cache *,
					struct ext4_extent *, void *);

#define EXT_CONTINUE   0
#define EXT_BREAK      1
#define EXT_REPEAT     2

#define EXT_MAX_BLOCK	0xffffffff

/*
 * EXT_INIT_MAX_LEN is the maximum number of blocks we can have in an
 * initialized extent. This is 2^15 and not (2^16 - 1), since we use the
 * MSB of ee_len field in the extent datastructure to signify if this
 * particular extent is an initialized extent or an uninitialized (i.e.
 * preallocated).
 * EXT_UNINIT_MAX_LEN is the maximum number of blocks we can have in an
 * uninitialized extent.
 * If ee_len is <= 0x8000, it is an initialized extent. Otherwise, it is an
 * uninitialized one. In other words, if MSB of ee_len is set, it is an
 * uninitialized extent with only one special scenario when ee_len = 0x8000.
 * In this case we can not have an uninitialized extent of zero length and
 * thus we make it as a special case of initialized extent with 0x8000 length.
 * This way we get better extent-to-group alignment for initialized extents.
 * Hence, the maximum number of blocks we can have in an *initialized*
 * extent is 2^15 (32768) and in an *uninitialized* extent is 2^15-1 (32767).
 */
#define EXT_INIT_MAX_LEN	(1UL << 15)
#define EXT_UNINIT_MAX_LEN	(EXT_INIT_MAX_LEN - 1)


#define EXT_FIRST_EXTENT(__hdr__) \
	((struct ext4_extent *) (((char *) (__hdr__)) +		\
				 sizeof(struct ext4_extent_header)))
#define EXT_FIRST_INDEX(__hdr__) \
	((struct ext4_extent_idx *) (((char *) (__hdr__)) +	\
				     sizeof(struct ext4_extent_header)))
#define EXT_HAS_FREE_INDEX(__path__) \
	(le16_to_cpu((__path__)->p_hdr->eh_entries) \
				     < le16_to_cpu((__path__)->p_hdr->eh_max))
#define EXT_LAST_EXTENT(__hdr__) \
	(EXT_FIRST_EXTENT((__hdr__)) + le16_to_cpu((__hdr__)->eh_entries) - 1)
#define EXT_LAST_INDEX(__hdr__) \
	(EXT_FIRST_INDEX((__hdr__)) + le16_to_cpu((__hdr__)->eh_entries) - 1)
#define EXT_MAX_EXTENT(__hdr__) \
	(EXT_FIRST_EXTENT((__hdr__)) + le16_to_cpu((__hdr__)->eh_max) - 1)
#define EXT_MAX_INDEX(__hdr__) \
	(EXT_FIRST_INDEX((__hdr__)) + le16_to_cpu((__hdr__)->eh_max) - 1)

static inline struct ext4_extent_header *ext_inode_hdr(struct inode *inode)
{
	return (struct ext4_extent_header *) EXT4_I(inode)->i_data;
}

static inline struct ext4_extent_header *ext_block_hdr(struct buffer_head *bh)
{
	return (struct ext4_extent_header *) bh->b_data;
}

static inline unsigned short ext_depth(struct inode *inode)
{
	return le16_to_cpu(ext_inode_hdr(inode)->eh_depth);
}

static inline void
ext4_ext_invalidate_cache(struct inode *inode)
{
	EXT4_I(inode)->i_cached_extent.ec_type = EXT4_EXT_CACHE_NO;
}

static inline void ext4_ext_mark_uninitialized(struct ext4_extent *ext)
{
	/* We can not have an uninitialized extent of zero length! */
	BUG_ON((le16_to_cpu(ext->ee_len) & ~EXT_INIT_MAX_LEN) == 0);
	ext->ee_len |= cpu_to_le16(EXT_INIT_MAX_LEN);
}

static inline int ext4_ext_is_uninitialized(struct ext4_extent *ext)
{
	/* Extent with ee_len of 0x8000 is treated as an initialized extent */
	return (le16_to_cpu(ext->ee_len) > EXT_INIT_MAX_LEN);
}

static inline int ext4_ext_get_actual_len(struct ext4_extent *ext)
{
	return (le16_to_cpu(ext->ee_len) <= EXT_INIT_MAX_LEN ?
		le16_to_cpu(ext->ee_len) :
		(le16_to_cpu(ext->ee_len) - EXT_INIT_MAX_LEN));
}

extern int ext4_ext_calc_metadata_amount(struct inode *inode, int blocks);
extern ext4_fsblk_t idx_pblock(struct ext4_extent_idx *);
extern void ext4_ext_store_pblock(struct ext4_extent *, ext4_fsblk_t);
extern int ext4_extent_tree_init(handle_t *, struct inode *);
extern int ext4_ext_calc_credits_for_single_extent(struct inode *inode,
						   int num,
						   struct ext4_ext_path *path);
extern int ext4_ext_try_to_merge(struct inode *inode,
				 struct ext4_ext_path *path,
				 struct ext4_extent *);
extern unsigned int ext4_ext_check_overlap(struct inode *, struct ext4_extent *, struct ext4_ext_path *);
extern int ext4_ext_insert_extent(handle_t *, struct inode *, struct ext4_ext_path *, struct ext4_extent *);
extern int ext4_ext_walk_space(struct inode *, ext4_lblk_t, ext4_lblk_t,
							ext_prepare_callback, void *);
extern struct ext4_ext_path *ext4_ext_find_extent(struct inode *, ext4_lblk_t,
							struct ext4_ext_path *);
extern int ext4_ext_search_left(struct inode *, struct ext4_ext_path *,
						ext4_lblk_t *, ext4_fsblk_t *);
extern int ext4_ext_search_right(struct inode *, struct ext4_ext_path *,
						ext4_lblk_t *, ext4_fsblk_t *);
extern void ext4_ext_drop_refs(struct ext4_ext_path *);
extern int ext4_ext_check_inode(struct inode *inode);
#endif /* _EXT4_EXTENTS */
Commit	Line	Data
a86c6181 AT	1	/*
	2	* Copyright (c) 2003-2006, Cluster File Systems, Inc, info@clusterfs.com
	3	* Written by Alex Tomas <alex@clusterfs.com>
	4	*
	5	* This program is free software; you can redistribute it and/or modify
	6	* it under the terms of the GNU General Public License version 2 as
	7	* published by the Free Software Foundation.
	8	*
	9	* This program is distributed in the hope that it will be useful,
	10	* but WITHOUT ANY WARRANTY; without even the implied warranty of
	11	* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
	12	* GNU General Public License for more details.
	13	*
	14	* You should have received a copy of the GNU General Public Licens
	15	* along with this program; if not, write to the Free Software
	16	* Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-
	17	*/
	18
3dcf5451 CH	19	#ifndef _EXT4_EXTENTS
3dcf5451 CH	20	#define _EXT4_EXTENTS
a86c6181	21
3dcf5451	22	#include "ext4.h"
a86c6181 AT	23
a86c6181 AT	24	/*
bbf2f9fb	25	* With AGGRESSIVE_TEST defined, the capacity of index/leaf blocks
d0d856e8 RD	26	* becomes very small, so index split, in-depth growing and
	27	* other hard changes happen much more often.
	28	* This is for debug purposes only.
a86c6181	29	*/
bbf2f9fb	30	#define AGGRESSIVE_TEST_
a86c6181 AT	31
a86c6181 AT	32	/*
d0d856e8 RD	33	* With EXTENTS_STATS defined, the number of blocks and extents
	34	* are collected in the truncate path. They'll be shown at
	35	* umount time.
a86c6181 AT	36	*/
	37	#define EXTENTS_STATS__
	38
	39	/*
d0d856e8 RD	40	* If CHECK_BINSEARCH is defined, then the results of the binary search
d0d856e8 RD	41	* will also be checked by linear search.
a86c6181 AT	42	*/
	43	#define CHECK_BINSEARCH__
	44
	45	/*
d0d856e8 RD	46	* If EXT_DEBUG is defined you can use the 'extdebug' mount option
d0d856e8 RD	47	* to get lots of info about what's going on.
a86c6181 AT	48	*/
	49	#define EXT_DEBUG__
	50	#ifdef EXT_DEBUG
	51	#define ext_debug(a...) printk(a)
	52	#else
	53	#define ext_debug(a...)
	54	#endif
	55
	56	/*
d0d856e8 RD	57	* If EXT_STATS is defined then stats numbers are collected.
d0d856e8 RD	58	* These number will be displayed at umount time.
a86c6181 AT	59	*/
	60	#define EXT_STATS_
	61
	62
	63	/*
d0d856e8 RD	64	* ext4_inode has i_block array (60 bytes total).
	65	* The first 12 bytes store ext4_extent_header;
	66	* the remainder stores an array of ext4_extent.
a86c6181 AT	67	*/
	68
	69	/*
d0d856e8 RD	70	* This is the extent on-disk structure.
d0d856e8 RD	71	* It's used at the bottom of the tree.
a86c6181 AT	72	*/
	73	struct ext4_extent {
	74	__le32 ee_block; /* first logical block extent covers */
	75	__le16 ee_len; /* number of blocks covered by extent */
	76	__le16 ee_start_hi; /* high 16 bits of physical block */
b377611d	77	__le32 ee_start_lo; /* low 32 bits of physical block */
a86c6181 AT	78	};
	79
	80	/*
d0d856e8 RD	81	* This is index on-disk structure.
d0d856e8 RD	82	* It's used at all the levels except the bottom.
a86c6181 AT	83	*/
	84	struct ext4_extent_idx {
	85	__le32 ei_block; /* index covers logical blocks from 'block' */
d8dd0b45	86	__le32 ei_leaf_lo; /* pointer to the physical block of the next *
d0d856e8	87	* level. leaf or next index could be there */
a86c6181 AT	88	__le16 ei_leaf_hi; /* high 16 bits of physical block */
	89	__u16 ei_unused;
	90	};
	91
	92	/*
d0d856e8	93	* Each block (leaves and indexes), even inode-stored has header.
a86c6181 AT	94	*/
	95	struct ext4_extent_header {
	96	__le16 eh_magic; /* probably will support different formats */
	97	__le16 eh_entries; /* number of valid entries */
	98	__le16 eh_max; /* capacity of store in entries */
d0d856e8	99	__le16 eh_depth; /* has tree real underlying blocks? */
a86c6181 AT	100	__le32 eh_generation; /* generation of the tree */
	101	};
	102
	103	#define EXT4_EXT_MAGIC cpu_to_le16(0xf30a)
	104
	105	/*
d0d856e8 RD	106	* Array of ext4_ext_path contains path to some extent.
	107	* Creation/lookup routines use it for traversal/splitting/etc.
	108	* Truncate uses it to simulate recursive walking.
a86c6181 AT	109	*/
a86c6181 AT	110	struct ext4_ext_path {
f65e6fba	111	ext4_fsblk_t p_block;
a86c6181 AT	112	__u16 p_depth;
	113	struct ext4_extent *p_ext;
	114	struct ext4_extent_idx *p_idx;
	115	struct ext4_extent_header *p_hdr;
	116	struct buffer_head *p_bh;
	117	};
	118
	119	/*
	120	* structure for external API
	121	*/
	122
	123	#define EXT4_EXT_CACHE_NO 0
	124	#define EXT4_EXT_CACHE_GAP 1
	125	#define EXT4_EXT_CACHE_EXTENT 2
	126
6873fa0d ES	127	/*
	128	* to be called by ext4_ext_walk_space()
	129	* negative retcode - error
	130	* positive retcode - signal for ext4_ext_walk_space(), see below
	131	* callback must return valid extent (passed or newly created)
	132	*/
	133	typedef int (ext_prepare_callback)(struct inode , struct ext4_ext_path *,
	134	struct ext4_ext_cache *,
	135	struct ext4_extent , void );
	136
	137	#define EXT_CONTINUE 0
	138	#define EXT_BREAK 1
	139	#define EXT_REPEAT 2
a86c6181 AT	140
	141	#define EXT_MAX_BLOCK 0xffffffff
	142
749269fa AA	143	/*
	144	* EXT_INIT_MAX_LEN is the maximum number of blocks we can have in an
	145	* initialized extent. This is 2^15 and not (2^16 - 1), since we use the
	146	* MSB of ee_len field in the extent datastructure to signify if this
	147	* particular extent is an initialized extent or an uninitialized (i.e.
	148	* preallocated).
	149	* EXT_UNINIT_MAX_LEN is the maximum number of blocks we can have in an
	150	* uninitialized extent.
	151	* If ee_len is <= 0x8000, it is an initialized extent. Otherwise, it is an
	152	* uninitialized one. In other words, if MSB of ee_len is set, it is an
	153	* uninitialized extent with only one special scenario when ee_len = 0x8000.
	154	* In this case we can not have an uninitialized extent of zero length and
	155	* thus we make it as a special case of initialized extent with 0x8000 length.
	156	* This way we get better extent-to-group alignment for initialized extents.
	157	* Hence, the maximum number of blocks we can have in an initialized
	158	* extent is 2^15 (32768) and in an uninitialized extent is 2^15-1 (32767).
	159	*/
	160	#define EXT_INIT_MAX_LEN (1UL << 15)
	161	#define EXT_UNINIT_MAX_LEN (EXT_INIT_MAX_LEN - 1)
471d4011	162
a86c6181 AT	163
	164	#define EXT_FIRST_EXTENT(__hdr__) \
	165	((struct ext4_extent ) (((char ) (__hdr__)) + \
	166	sizeof(struct ext4_extent_header)))
	167	#define EXT_FIRST_INDEX(__hdr__) \
	168	((struct ext4_extent_idx ) (((char ) (__hdr__)) + \
	169	sizeof(struct ext4_extent_header)))
	170	#define EXT_HAS_FREE_INDEX(__path__) \
8c55e204 DK	171	(le16_to_cpu((__path__)->p_hdr->eh_entries) \
8c55e204 DK	172	< le16_to_cpu((__path__)->p_hdr->eh_max))
a86c6181 AT	173	#define EXT_LAST_EXTENT(__hdr__) \
	174	(EXT_FIRST_EXTENT((__hdr__)) + le16_to_cpu((__hdr__)->eh_entries) - 1)
	175	#define EXT_LAST_INDEX(__hdr__) \
	176	(EXT_FIRST_INDEX((__hdr__)) + le16_to_cpu((__hdr__)->eh_entries) - 1)
	177	#define EXT_MAX_EXTENT(__hdr__) \
	178	(EXT_FIRST_EXTENT((__hdr__)) + le16_to_cpu((__hdr__)->eh_max) - 1)
	179	#define EXT_MAX_INDEX(__hdr__) \
	180	(EXT_FIRST_INDEX((__hdr__)) + le16_to_cpu((__hdr__)->eh_max) - 1)
	181
	182	static inline struct ext4_extent_header ext_inode_hdr(struct inode inode)
	183	{
	184	return (struct ext4_extent_header *) EXT4_I(inode)->i_data;
	185	}
	186
	187	static inline struct ext4_extent_header ext_block_hdr(struct buffer_head bh)
	188	{
	189	return (struct ext4_extent_header *) bh->b_data;
	190	}
	191
	192	static inline unsigned short ext_depth(struct inode *inode)
	193	{
	194	return le16_to_cpu(ext_inode_hdr(inode)->eh_depth);
	195	}
	196
a86c6181 AT	197	static inline void
	198	ext4_ext_invalidate_cache(struct inode *inode)
	199	{
	200	EXT4_I(inode)->i_cached_extent.ec_type = EXT4_EXT_CACHE_NO;
	201	}
	202
a2df2a63 AA	203	static inline void ext4_ext_mark_uninitialized(struct ext4_extent *ext)
a2df2a63 AA	204	{
749269fa AA	205	/* We can not have an uninitialized extent of zero length! */
	206	BUG_ON((le16_to_cpu(ext->ee_len) & ~EXT_INIT_MAX_LEN) == 0);
	207	ext->ee_len \|= cpu_to_le16(EXT_INIT_MAX_LEN);
a2df2a63 AA	208	}
	209
	210	static inline int ext4_ext_is_uninitialized(struct ext4_extent *ext)
	211	{
749269fa AA	212	/* Extent with ee_len of 0x8000 is treated as an initialized extent */
749269fa AA	213	return (le16_to_cpu(ext->ee_len) > EXT_INIT_MAX_LEN);
a2df2a63 AA	214	}
	215
	216	static inline int ext4_ext_get_actual_len(struct ext4_extent *ext)
	217	{
749269fa AA	218	return (le16_to_cpu(ext->ee_len) <= EXT_INIT_MAX_LEN ?
	219	le16_to_cpu(ext->ee_len) :
	220	(le16_to_cpu(ext->ee_len) - EXT_INIT_MAX_LEN));
a2df2a63 AA	221	}
a2df2a63 AA	222
d2a17637	223	extern int ext4_ext_calc_metadata_amount(struct inode *inode, int blocks);
c14c6fd5 AK	224	extern ext4_fsblk_t idx_pblock(struct ext4_extent_idx *);
c14c6fd5 AK	225	extern void ext4_ext_store_pblock(struct ext4_extent *, ext4_fsblk_t);
a86c6181	226	extern int ext4_extent_tree_init(handle_t , struct inode );
ee12b630 MC	227	extern int ext4_ext_calc_credits_for_single_extent(struct inode *inode,
	228	int num,
	229	struct ext4_ext_path *path);
56055d3a AA	230	extern int ext4_ext_try_to_merge(struct inode *inode,
	231	struct ext4_ext_path *path,
	232	struct ext4_extent *);
25d14f98	233	extern unsigned int ext4_ext_check_overlap(struct inode , struct ext4_extent , struct ext4_ext_path *);
a86c6181	234	extern int ext4_ext_insert_extent(handle_t , struct inode , struct ext4_ext_path , struct ext4_extent );
6873fa0d ES	235	extern int ext4_ext_walk_space(struct inode *, ext4_lblk_t, ext4_lblk_t,
6873fa0d ES	236	ext_prepare_callback, void *);
725d26d3 AK	237	extern struct ext4_ext_path ext4_ext_find_extent(struct inode , ext4_lblk_t,
725d26d3 AK	238	struct ext4_ext_path *);
1988b51e AT	239	extern int ext4_ext_search_left(struct inode , struct ext4_ext_path ,
	240	ext4_lblk_t , ext4_fsblk_t );
	241	extern int ext4_ext_search_right(struct inode , struct ext4_ext_path ,
	242	ext4_lblk_t , ext4_fsblk_t );
b35905c1	243	extern void ext4_ext_drop_refs(struct ext4_ext_path *);
7a262f7c	244	extern int ext4_ext_check_inode(struct inode *inode);
3dcf5451	245	#endif /* _EXT4_EXTENTS */
a86c6181	246