path: root/fs/logfs/logfs.h

/*
 * fs/logfs/logfs.h
 *
 * As should be obvious for Linux kernel code, license is GPLv2
 *
 * Copyright (c) 2005-2008 Joern Engel <joern@logfs.org>
 *
 * Private header for logfs.
 */
#ifndef FS_LOGFS_LOGFS_H
#define FS_LOGFS_LOGFS_H

#undef __CHECK_ENDIAN__
#define __CHECK_ENDIAN__

#include <linux/btree.h>
#include <linux/crc32.h>
#include <linux/fs.h>
#include <linux/kernel.h>
#include <linux/mempool.h>
#include <linux/pagemap.h>
#include <linux/mtd/mtd.h>
#include "logfs_abi.h"

#define LOGFS_DEBUG_SUPER	(0x0001)
#define LOGFS_DEBUG_SEGMENT	(0x0002)
#define LOGFS_DEBUG_JOURNAL	(0x0004)
#define LOGFS_DEBUG_DIR		(0x0008)
#define LOGFS_DEBUG_FILE	(0x0010)
#define LOGFS_DEBUG_INODE	(0x0020)
#define LOGFS_DEBUG_READWRITE	(0x0040)
#define LOGFS_DEBUG_GC		(0x0080)
#define LOGFS_DEBUG_GC_NOISY	(0x0100)
#define LOGFS_DEBUG_ALIASES	(0x0200)
#define LOGFS_DEBUG_BLOCKMOVE	(0x0400)
#define LOGFS_DEBUG_ALL		(0xffffffff)

#define LOGFS_DEBUG		(0x01)
/*
 * To enable specific log messages, simply define LOGFS_DEBUG to match any
 * or all of the above.
 */
#ifndef LOGFS_DEBUG
#define LOGFS_DEBUG		(0)
#endif

#define log_cond(cond, fmt, arg...) do {	\
	if (cond)				\
		printk(KERN_DEBUG fmt, ##arg);	\
} while (0)

#define log_super(fmt, arg...) \
	log_cond(LOGFS_DEBUG & LOGFS_DEBUG_SUPER, fmt, ##arg)
#define log_segment(fmt, arg...) \
	log_cond(LOGFS_DEBUG & LOGFS_DEBUG_SEGMENT, fmt, ##arg)
#define log_journal(fmt, arg...) \
	log_cond(LOGFS_DEBUG & LOGFS_DEBUG_JOURNAL, fmt, ##arg)
#define log_dir(fmt, arg...) \
	log_cond(LOGFS_DEBUG & LOGFS_DEBUG_DIR, fmt, ##arg)
#define log_file(fmt, arg...) \
	log_cond(LOGFS_DEBUG & LOGFS_DEBUG_FILE, fmt, ##arg)
#define log_inode(fmt, arg...) \
	log_cond(LOGFS_DEBUG & LOGFS_DEBUG_INODE, fmt, ##arg)
#define log_readwrite(fmt, arg...) \
	log_cond(LOGFS_DEBUG & LOGFS_DEBUG_READWRITE, fmt, ##arg)
#define log_gc(fmt, arg...) \
	log_cond(LOGFS_DEBUG & LOGFS_DEBUG_GC, fmt, ##arg)
#define log_gc_noisy(fmt, arg...) \
	log_cond(LOGFS_DEBUG & LOGFS_DEBUG_GC_NOISY, fmt, ##arg)
#define log_aliases(fmt, arg...) \
	log_cond(LOGFS_DEBUG & LOGFS_DEBUG_ALIASES, fmt, ##arg)
#define log_blockmove(fmt, arg...) \
	log_cond(LOGFS_DEBUG & LOGFS_DEBUG_BLOCKMOVE, fmt, ##arg)

#define PG_pre_locked		PG_owner_priv_1
#define PagePreLocked(page)	test_bit(PG_pre_locked, &(page)->flags)
#define SetPagePreLocked(page)	set_bit(PG_pre_locked, &(page)->flags)
#define ClearPagePreLocked(page) clear_bit(PG_pre_locked, &(page)->flags)

/* FIXME: This should really be somewhere in the 64bit area. */
#define LOGFS_LINK_MAX		(1<<30)

/* Read-only filesystem */
#define LOGFS_SB_FLAG_RO	0x0001
#define LOGFS_SB_FLAG_DIRTY	0x0002
#define LOGFS_SB_FLAG_OBJ_ALIAS	0x0004
#define LOGFS_SB_FLAG_SHUTDOWN	0x0008

/* Write Control Flags */
#define WF_LOCK			0x01 /* take write lock */
#define WF_WRITE		0x02 /* write block */
#define WF_DELETE		0x04 /* delete old block */

typedef u8 __bitwise level_t;
typedef u8 __bitwise gc_level_t;

#define LEVEL(level) ((__force level_t)(level))
#define GC_LEVEL(gc_level) ((__force gc_level_t)(gc_level))

#define SUBLEVEL(level) ( (void)((level) == LEVEL(1)),	\
		(__force level_t)((__force u8)(level) - 1) )

/**
 * struct logfs_area - area management information
 *
 * @a_sb:			the superblock this area belongs to
 * @a_is_open:			1 if the area is currently open, else 0
 * @a_segno:			segment number of area
 * @a_written_bytes:		number of bytes already written back
 * @a_used_bytes:		number of used bytes
 * @a_ops:			area operations (either journal or ostore)
 * @a_erase_count:		erase count
 * @a_level:			GC level
 */
struct logfs_area { /* a segment open for writing */
	struct super_block *a_sb;
	int	a_is_open;
	u32	a_segno;
	u32	a_written_bytes;
	u32	a_used_bytes;
	const struct logfs_area_ops *a_ops;
	u32	a_erase_count;
	gc_level_t a_level;
};

/**
 * struct logfs_area_ops - area operations
 *
 * @get_free_segment:		fill area->ofs with the offset of a free segment
 * @get_erase_count:		fill area->erase_count (needs area->ofs)
 * @erase_segment:		erase and setup segment
 */
struct logfs_area_ops {
	void	(*get_free_segment)(struct logfs_area *area);
	void	(*get_erase_count)(struct logfs_area *area);
	int	(*erase_segment)(struct logfs_area *area);
};

struct logfs_super;	/* forward */
/**
 * struct logfs_device_ops - device access operations
 *
 * @readpage:			read one page (mm page)
 * @writeseg:			write one segment.  may be a partial segment
 * @erase:			erase one segment
 * @read:			read from the device
 * @erase:			erase part of the device
 * @can_write_buf:		decide whether wbuf can be written to ofs
 */
struct logfs_device_ops {
	struct page *(*find_first_sb)(struct super_block *sb, u64 *ofs);
	struct page *(*find_last_sb)(struct super_block *sb, u64 *ofs);
	int (*write_sb)(struct super_block *sb, struct page *page);
	int (*readpage)(void *_sb, struct page *page);
	void (*writeseg)(struct super_block *sb, u64 ofs, size_t len);
	int (*erase)(struct super_block *sb, loff_t ofs, size_t len,
			int ensure_write);
	int (*can_write_buf)(struct super_block *sb, u64 ofs);
	void (*sync)(struct super_block *sb);
	void (*put_device)(struct logfs_super *s);
};

/**
 * struct candidate_list - list of similar candidates
 */
struct candidate_list {
	struct rb_root rb_tree;
	int count;
	int maxcount;
	int sort_by_ec;
};

/**
 * struct gc_candidate - "candidate" segment to be garbage collected next
 *
 * @list:			list (either free of low)
 * @segno:			segment number
 * @valid:			number of valid bytes
 * @erase_count:		erase count of segment
 * @dist:			distance from tree root
 *
 * Candidates can be on two lists.  The free list contains electees rather
 * than candidates - segments that no longer contain any valid data.  The
 * low list contains candidates to be picked for GC.  It should be kept
 * short.  It is not required to always pick a perfect candidate.  In the
 * worst case GC will have to move more data than absolutely necessary.
 */
struct gc_candidate {
	struct rb_node rb_node;
	struct candidate_list *list;
	u32	segno;
	u32	valid;
	u32	erase_count;
	u8	dist;
};

/**
 * struct logfs_journal_entry - temporary structure used during journal scan
 *
 * @used:
 * @version:			normalized version
 * @len:			length
 * @offset:			offset
 */
struct logfs_journal_entry {
	int used;
	s16 version;
	u16 len;
	u16 datalen;
	u64 offset;
};

enum transaction_state {
	CREATE_1 = 1,
	CREATE_2,
	UNLINK_1,
	UNLINK_2,
	CROSS_RENAME_1,
	CROSS_RENAME_2,
	TARGET_RENAME_1,
	TARGET_RENAME_2,
	TARGET_RENAME_3
};

/**
 * struct logfs_transaction - essential fields to support atomic dirops
 *
 * @ino:			target inode
 * @dir:			inode of directory containing dentry
 * @pos:			pos of dentry in directory
 */
struct logfs_transaction {
	enum transaction_state state;
	u64	 ino;
	u64	 dir;
	u64	 pos;
};

/**
 * struct logfs_shadow - old block in the shadow of a not-yet-committed new one
 * @old_ofs:			offset of old block on medium
 * @new_ofs:			offset of new block on medium
 * @ino:			inode number
 * @bix:			block index
 * @old_len:			size of old block, including header
 * @new_len:			size of new block, including header
 * @level:			block level
 */
struct logfs_shadow {
	u64 old_ofs;
	u64 new_ofs;
	u64 ino;
	u64 bix;
	int old_len;
	int new_len;
	gc_level_t gc_level;
};

/**
 * struct shadow_tree
 * @new:			shadows where old_ofs==0, indexed by new_ofs
 * @old:			shadows where old_ofs!=0, indexed by old_ofs
 * @segment_map:		bitfield of segments containing shadows
 * @no_shadowed_segment:	number of segments containing shadows
 */
struct shadow_tree {
	struct btree_head64 new;
	struct btree_head64 old;
	struct btree_head32 segment_map;
	int no_shadowed_segments;
};

struct object_alias_item {
	struct list_head list;
	__be64 val;
	int child_no;
};

/**
 * struct logfs_block - contains any block state
 * @type:			indirect block or inode
 * @full:			number of fully populated children
 * @partial:			number of partially populated children
 *
 * Most blocks are directly represented by page cache pages.  But when a block
 * becomes dirty, is part of a transaction, contains aliases or is otherwise
 * special, a struct logfs_block is allocated to track the additional state.
 * Inodes are very similar to indirect blocks, so they can also get one of
 * these structures added when appropriate.
 */
#define BLOCK_INDIRECT	1	/* Indirect block */
#define BLOCK_INODE	2	/* Inode */
struct logfs_block_ops;
struct logfs_block {
	struct list_head alias_list;
	struct list_head item_list;
	struct super_block *sb;
	u64 ino;
	u64 bix;
	level_t level;
	struct page *page;
	struct inode *inode;
	struct logfs_transaction *ta;
	unsigned long alias_map[LOGFS_BLOCK_FACTOR / BITS_PER_LONG];
	const struct logfs_block_ops *ops;
	int full;
	int partial;
	int reserved_bytes;
};

typedef int write_alias_t(struct super_block *sb, u64 ino, u64 bix,
		level_t level, int child_no, __be64 val);
struct logfs_block_ops {
	void	(*write_block)(struct logfs_block *block);
	void	(*free_block)(struct super_block *sb, struct logfs_block*block);
	int	(*write_alias)(struct super_block *sb,
			struct logfs_bloc