2006-10-11 15:20:50 +07:00
|
|
|
/*
|
2008-04-30 05:13:32 +07:00
|
|
|
* ext4.h
|
2006-10-11 15:20:50 +07:00
|
|
|
*
|
|
|
|
* Copyright (C) 1992, 1993, 1994, 1995
|
|
|
|
* Remy Card (card@masi.ibp.fr)
|
|
|
|
* Laboratoire MASI - Institut Blaise Pascal
|
|
|
|
* Universite Pierre et Marie Curie (Paris VI)
|
|
|
|
*
|
|
|
|
* from
|
|
|
|
*
|
|
|
|
* linux/include/linux/minix_fs.h
|
|
|
|
*
|
|
|
|
* Copyright (C) 1991, 1992 Linus Torvalds
|
|
|
|
*/
|
|
|
|
|
2008-04-30 05:13:32 +07:00
|
|
|
#ifndef _EXT4_H
|
|
|
|
#define _EXT4_H
|
2006-10-11 15:20:50 +07:00
|
|
|
|
|
|
|
#include <linux/types.h>
|
2006-10-11 15:21:05 +07:00
|
|
|
#include <linux/blkdev.h>
|
2006-10-11 15:20:50 +07:00
|
|
|
#include <linux/magic.h>
|
2009-01-06 09:49:55 +07:00
|
|
|
#include <linux/jbd2.h>
|
2009-01-23 00:13:05 +07:00
|
|
|
#include <linux/quota.h>
|
2009-05-02 00:44:33 +07:00
|
|
|
#include <linux/rwsem.h>
|
|
|
|
#include <linux/rbtree.h>
|
|
|
|
#include <linux/seqlock.h>
|
|
|
|
#include <linux/mutex.h>
|
2009-05-04 03:33:44 +07:00
|
|
|
#include <linux/timer.h>
|
|
|
|
#include <linux/wait.h>
|
|
|
|
#include <linux/blockgroup_lock.h>
|
|
|
|
#include <linux/percpu_counter.h>
|
2008-01-29 12:19:52 +07:00
|
|
|
|
2006-10-11 15:20:50 +07:00
|
|
|
/*
|
2008-07-12 06:27:31 +07:00
|
|
|
* The fourth extended filesystem constants/structures
|
2006-10-11 15:20:50 +07:00
|
|
|
*/
|
|
|
|
|
|
|
|
/*
|
2006-10-11 15:20:53 +07:00
|
|
|
* Define EXT4FS_DEBUG to produce debug messages
|
2006-10-11 15:20:50 +07:00
|
|
|
*/
|
2006-10-11 15:20:53 +07:00
|
|
|
#undef EXT4FS_DEBUG
|
2006-10-11 15:20:50 +07:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Debug code
|
|
|
|
*/
|
2006-10-11 15:20:53 +07:00
|
|
|
#ifdef EXT4FS_DEBUG
|
|
|
|
#define ext4_debug(f, a...) \
|
2006-10-11 15:20:50 +07:00
|
|
|
do { \
|
2008-09-09 09:25:24 +07:00
|
|
|
printk(KERN_DEBUG "EXT4-fs DEBUG (%s, %d): %s:", \
|
2008-07-14 08:03:29 +07:00
|
|
|
__FILE__, __LINE__, __func__); \
|
2008-09-09 09:25:24 +07:00
|
|
|
printk(KERN_DEBUG f, ## a); \
|
2006-10-11 15:20:50 +07:00
|
|
|
} while (0)
|
|
|
|
#else
|
2006-10-11 15:20:53 +07:00
|
|
|
#define ext4_debug(f, a...) do {} while (0)
|
2006-10-11 15:20:50 +07:00
|
|
|
#endif
|
|
|
|
|
2009-05-02 00:44:33 +07:00
|
|
|
/* data type for block offset of block group */
|
|
|
|
typedef int ext4_grpblk_t;
|
|
|
|
|
|
|
|
/* data type for filesystem-wide blocks number */
|
|
|
|
typedef unsigned long long ext4_fsblk_t;
|
|
|
|
|
|
|
|
/* data type for file logical block number */
|
|
|
|
typedef __u32 ext4_lblk_t;
|
|
|
|
|
|
|
|
/* data type for block group number */
|
|
|
|
typedef unsigned int ext4_group_t;
|
|
|
|
|
|
|
|
|
2008-01-29 12:19:52 +07:00
|
|
|
/* prefer goal again. length */
|
2009-08-10 03:46:13 +07:00
|
|
|
#define EXT4_MB_HINT_MERGE 0x0001
|
2008-01-29 12:19:52 +07:00
|
|
|
/* blocks already reserved */
|
2009-08-10 03:46:13 +07:00
|
|
|
#define EXT4_MB_HINT_RESERVED 0x0002
|
2008-01-29 12:19:52 +07:00
|
|
|
/* metadata is being allocated */
|
2009-08-10 03:46:13 +07:00
|
|
|
#define EXT4_MB_HINT_METADATA 0x0004
|
2008-01-29 12:19:52 +07:00
|
|
|
/* first blocks in the file */
|
2009-08-10 03:46:13 +07:00
|
|
|
#define EXT4_MB_HINT_FIRST 0x0008
|
2008-01-29 12:19:52 +07:00
|
|
|
/* search for the best chunk */
|
2009-08-10 03:46:13 +07:00
|
|
|
#define EXT4_MB_HINT_BEST 0x0010
|
2008-01-29 12:19:52 +07:00
|
|
|
/* data is being allocated */
|
2009-08-10 03:46:13 +07:00
|
|
|
#define EXT4_MB_HINT_DATA 0x0020
|
2008-01-29 12:19:52 +07:00
|
|
|
/* don't preallocate (for tails) */
|
2009-08-10 03:46:13 +07:00
|
|
|
#define EXT4_MB_HINT_NOPREALLOC 0x0040
|
2008-01-29 12:19:52 +07:00
|
|
|
/* allocate for locality group */
|
2009-08-10 03:46:13 +07:00
|
|
|
#define EXT4_MB_HINT_GROUP_ALLOC 0x0080
|
2008-01-29 12:19:52 +07:00
|
|
|
/* allocate goal blocks or none */
|
2009-08-10 03:46:13 +07:00
|
|
|
#define EXT4_MB_HINT_GOAL_ONLY 0x0100
|
2008-01-29 12:19:52 +07:00
|
|
|
/* goal is meaningful */
|
2009-08-10 03:46:13 +07:00
|
|
|
#define EXT4_MB_HINT_TRY_GOAL 0x0200
|
2008-07-15 04:52:37 +07:00
|
|
|
/* blocks already pre-reserved by delayed allocation */
|
2009-08-10 03:46:13 +07:00
|
|
|
#define EXT4_MB_DELALLOC_RESERVED 0x0400
|
2009-08-10 09:01:13 +07:00
|
|
|
/* We are doing stream allocation */
|
|
|
|
#define EXT4_MB_STREAM_ALLOC 0x0800
|
2008-07-15 04:52:37 +07:00
|
|
|
|
2008-01-29 12:19:52 +07:00
|
|
|
|
|
|
|
struct ext4_allocation_request {
|
|
|
|
/* target inode for block we're allocating */
|
|
|
|
struct inode *inode;
|
2009-07-13 21:24:17 +07:00
|
|
|
/* how many blocks we want to allocate */
|
|
|
|
unsigned int len;
|
2008-01-29 12:19:52 +07:00
|
|
|
/* logical block in target inode */
|
|
|
|
ext4_lblk_t logical;
|
|
|
|
/* the closest logical allocated block to the left */
|
|
|
|
ext4_lblk_t lleft;
|
|
|
|
/* the closest logical allocated block to the right */
|
|
|
|
ext4_lblk_t lright;
|
2009-07-13 21:24:17 +07:00
|
|
|
/* phys. target (a hint) */
|
|
|
|
ext4_fsblk_t goal;
|
|
|
|
/* phys. block for the closest logical allocated block to the left */
|
|
|
|
ext4_fsblk_t pleft;
|
|
|
|
/* phys. block for the closest logical allocated block to the right */
|
2008-01-29 12:19:52 +07:00
|
|
|
ext4_fsblk_t pright;
|
|
|
|
/* flags. see above EXT4_MB_HINT_* */
|
2008-11-05 12:14:04 +07:00
|
|
|
unsigned int flags;
|
2008-01-29 12:19:52 +07:00
|
|
|
};
|
|
|
|
|
2009-09-01 10:13:11 +07:00
|
|
|
/*
|
|
|
|
* For delayed allocation tracking
|
|
|
|
*/
|
|
|
|
struct mpage_da_data {
|
|
|
|
struct inode *inode;
|
|
|
|
sector_t b_blocknr; /* start block number of extent */
|
|
|
|
size_t b_size; /* size of extent */
|
|
|
|
unsigned long b_state; /* state of the extent */
|
|
|
|
unsigned long first_page, next_page; /* extent of pages */
|
|
|
|
struct writeback_control *wbc;
|
|
|
|
int io_done;
|
|
|
|
int pages_written;
|
|
|
|
int retval;
|
|
|
|
};
|
|
|
|
|
2009-09-29 02:49:08 +07:00
|
|
|
typedef struct ext4_io_end {
|
|
|
|
struct inode *inode; /* file being written to */
|
|
|
|
unsigned int flag; /* sync IO or AIO */
|
|
|
|
int error; /* I/O error code */
|
|
|
|
ext4_lblk_t offset; /* offset in the file */
|
|
|
|
size_t size; /* size of the extent */
|
|
|
|
struct work_struct work; /* data work queue */
|
|
|
|
} ext4_io_end_t;
|
|
|
|
|
2006-10-11 15:20:50 +07:00
|
|
|
/*
|
|
|
|
* Special inodes numbers
|
|
|
|
*/
|
2006-10-11 15:20:53 +07:00
|
|
|
#define EXT4_BAD_INO 1 /* Bad blocks inode */
|
|
|
|
#define EXT4_ROOT_INO 2 /* Root inode */
|
|
|
|
#define EXT4_BOOT_LOADER_INO 5 /* Boot loader inode */
|
|
|
|
#define EXT4_UNDEL_DIR_INO 6 /* Undelete directory inode */
|
|
|
|
#define EXT4_RESIZE_INO 7 /* Reserved group descriptors inode */
|
|
|
|
#define EXT4_JOURNAL_INO 8 /* Journal inode */
|
2006-10-11 15:20:50 +07:00
|
|
|
|
2006-10-11 15:20:53 +07:00
|
|
|
/* First non-reserved inode for old ext4 filesystems */
|
|
|
|
#define EXT4_GOOD_OLD_FIRST_INO 11
|
2006-10-11 15:20:50 +07:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Maximal count of links to a file
|
|
|
|
*/
|
2007-07-18 19:38:01 +07:00
|
|
|
#define EXT4_LINK_MAX 65000
|
2006-10-11 15:20:50 +07:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Macro-instructions used to manage several block sizes
|
|
|
|
*/
|
2006-10-11 15:20:53 +07:00
|
|
|
#define EXT4_MIN_BLOCK_SIZE 1024
|
2008-01-29 11:58:27 +07:00
|
|
|
#define EXT4_MAX_BLOCK_SIZE 65536
|
|
|
|
#define EXT4_MIN_BLOCK_LOG_SIZE 10
|
2006-10-11 15:20:50 +07:00
|
|
|
#ifdef __KERNEL__
|
2006-10-11 15:20:53 +07:00
|
|
|
# define EXT4_BLOCK_SIZE(s) ((s)->s_blocksize)
|
2006-10-11 15:20:50 +07:00
|
|
|
#else
|
2006-10-11 15:20:53 +07:00
|
|
|
# define EXT4_BLOCK_SIZE(s) (EXT4_MIN_BLOCK_SIZE << (s)->s_log_block_size)
|
2006-10-11 15:20:50 +07:00
|
|
|
#endif
|
2008-09-09 09:25:24 +07:00
|
|
|
#define EXT4_ADDR_PER_BLOCK(s) (EXT4_BLOCK_SIZE(s) / sizeof(__u32))
|
2006-10-11 15:20:50 +07:00
|
|
|
#ifdef __KERNEL__
|
2006-10-11 15:20:53 +07:00
|
|
|
# define EXT4_BLOCK_SIZE_BITS(s) ((s)->s_blocksize_bits)
|
2006-10-11 15:20:50 +07:00
|
|
|
#else
|
2006-10-11 15:20:53 +07:00
|
|
|
# define EXT4_BLOCK_SIZE_BITS(s) ((s)->s_log_block_size + 10)
|
2006-10-11 15:20:50 +07:00
|
|
|
#endif
|
|
|
|
#ifdef __KERNEL__
|
2006-10-11 15:20:53 +07:00
|
|
|
#define EXT4_ADDR_PER_BLOCK_BITS(s) (EXT4_SB(s)->s_addr_per_block_bits)
|
|
|
|
#define EXT4_INODE_SIZE(s) (EXT4_SB(s)->s_inode_size)
|
|
|
|
#define EXT4_FIRST_INO(s) (EXT4_SB(s)->s_first_ino)
|
2006-10-11 15:20:50 +07:00
|
|
|
#else
|
2006-10-11 15:20:53 +07:00
|
|
|
#define EXT4_INODE_SIZE(s) (((s)->s_rev_level == EXT4_GOOD_OLD_REV) ? \
|
|
|
|
EXT4_GOOD_OLD_INODE_SIZE : \
|
2006-10-11 15:20:50 +07:00
|
|
|
(s)->s_inode_size)
|
2006-10-11 15:20:53 +07:00
|
|
|
#define EXT4_FIRST_INO(s) (((s)->s_rev_level == EXT4_GOOD_OLD_REV) ? \
|
|
|
|
EXT4_GOOD_OLD_FIRST_INO : \
|
2006-10-11 15:20:50 +07:00
|
|
|
(s)->s_first_ino)
|
|
|
|
#endif
|
2007-07-18 08:42:41 +07:00
|
|
|
#define EXT4_BLOCK_ALIGN(size, blkbits) ALIGN((size), (1 << (blkbits)))
|
2006-10-11 15:20:50 +07:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Structure of a blocks group descriptor
|
|
|
|
*/
|
2006-10-11 15:20:53 +07:00
|
|
|
struct ext4_group_desc
|
2006-10-11 15:20:50 +07:00
|
|
|
{
|
2007-10-17 05:38:25 +07:00
|
|
|
__le32 bg_block_bitmap_lo; /* Blocks bitmap block */
|
2007-10-17 05:38:25 +07:00
|
|
|
__le32 bg_inode_bitmap_lo; /* Inodes bitmap block */
|
|
|
|
__le32 bg_inode_table_lo; /* Inodes table block */
|
2009-01-06 10:20:24 +07:00
|
|
|
__le16 bg_free_blocks_count_lo;/* Free blocks count */
|
|
|
|
__le16 bg_free_inodes_count_lo;/* Free inodes count */
|
|
|
|
__le16 bg_used_dirs_count_lo; /* Directories count */
|
Ext4: Uninitialized Block Groups
In pass1 of e2fsck, every inode table in the fileystem is scanned and checked,
regardless of whether it is in use. This is this the most time consuming part
of the filesystem check. The unintialized block group feature can greatly
reduce e2fsck time by eliminating checking of uninitialized inodes.
With this feature, there is a a high water mark of used inodes for each block
group. Block and inode bitmaps can be uninitialized on disk via a flag in the
group descriptor to avoid reading or scanning them at e2fsck time. A checksum
of each group descriptor is used to ensure that corruption in the group
descriptor's bit flags does not cause incorrect operation.
The feature is enabled through a mkfs option
mke2fs /dev/ -O uninit_groups
A patch adding support for uninitialized block groups to e2fsprogs tools has
been posted to the linux-ext4 mailing list.
The patches have been stress tested with fsstress and fsx. In performance
tests testing e2fsck time, we have seen that e2fsck time on ext3 grows
linearly with the total number of inodes in the filesytem. In ext4 with the
uninitialized block groups feature, the e2fsck time is constant, based
solely on the number of used inodes rather than the total inode count.
Since typical ext4 filesystems only use 1-10% of their inodes, this feature can
greatly reduce e2fsck time for users. With performance improvement of 2-20
times, depending on how full the filesystem is.
The attached graph shows the major improvements in e2fsck times in filesystems
with a large total inode count, but few inodes in use.
In each group descriptor if we have
EXT4_BG_INODE_UNINIT set in bg_flags:
Inode table is not initialized/used in this group. So we can skip
the consistency check during fsck.
EXT4_BG_BLOCK_UNINIT set in bg_flags:
No block in the group is used. So we can skip the block bitmap
verification for this group.
We also add two new fields to group descriptor as a part of
uninitialized group patch.
__le16 bg_itable_unused; /* Unused inodes count */
__le16 bg_checksum; /* crc16(sb_uuid+group+desc) */
bg_itable_unused:
If we have EXT4_BG_INODE_UNINIT not set in bg_flags
then bg_itable_unused will give the offset within
the inode table till the inodes are used. This can be
used by fsck to skip list of inodes that are marked unused.
bg_checksum:
Now that we depend on bg_flags and bg_itable_unused to determine
the block and inode usage, we need to make sure group descriptor
is not corrupt. We add checksum to group descriptor to
detect corruption. If the descriptor is found to be corrupt, we
mark all the blocks and inodes in the group used.
Signed-off-by: Avantika Mathur <mathur@us.ibm.com>
Signed-off-by: Andreas Dilger <adilger@clusterfs.com>
Signed-off-by: Mingming Cao <cmm@us.ibm.com>
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
2007-10-17 05:38:25 +07:00
|
|
|
__le16 bg_flags; /* EXT4_BG_flags (INODE_UNINIT, etc) */
|
|
|
|
__u32 bg_reserved[2]; /* Likely block/inode bitmap checksum */
|
2009-01-06 10:20:24 +07:00
|
|
|
__le16 bg_itable_unused_lo; /* Unused inodes count */
|
Ext4: Uninitialized Block Groups
In pass1 of e2fsck, every inode table in the fileystem is scanned and checked,
regardless of whether it is in use. This is this the most time consuming part
of the filesystem check. The unintialized block group feature can greatly
reduce e2fsck time by eliminating checking of uninitialized inodes.
With this feature, there is a a high water mark of used inodes for each block
group. Block and inode bitmaps can be uninitialized on disk via a flag in the
group descriptor to avoid reading or scanning them at e2fsck time. A checksum
of each group descriptor is used to ensure that corruption in the group
descriptor's bit flags does not cause incorrect operation.
The feature is enabled through a mkfs option
mke2fs /dev/ -O uninit_groups
A patch adding support for uninitialized block groups to e2fsprogs tools has
been posted to the linux-ext4 mailing list.
The patches have been stress tested with fsstress and fsx. In performance
tests testing e2fsck time, we have seen that e2fsck time on ext3 grows
linearly with the total number of inodes in the filesytem. In ext4 with the
uninitialized block groups feature, the e2fsck time is constant, based
solely on the number of used inodes rather than the total inode count.
Since typical ext4 filesystems only use 1-10% of their inodes, this feature can
greatly reduce e2fsck time for users. With performance improvement of 2-20
times, depending on how full the filesystem is.
The attached graph shows the major improvements in e2fsck times in filesystems
with a large total inode count, but few inodes in use.
In each group descriptor if we have
EXT4_BG_INODE_UNINIT set in bg_flags:
Inode table is not initialized/used in this group. So we can skip
the consistency check during fsck.
EXT4_BG_BLOCK_UNINIT set in bg_flags:
No block in the group is used. So we can skip the block bitmap
verification for this group.
We also add two new fields to group descriptor as a part of
uninitialized group patch.
__le16 bg_itable_unused; /* Unused inodes count */
__le16 bg_checksum; /* crc16(sb_uuid+group+desc) */
bg_itable_unused:
If we have EXT4_BG_INODE_UNINIT not set in bg_flags
then bg_itable_unused will give the offset within
the inode table till the inodes are used. This can be
used by fsck to skip list of inodes that are marked unused.
bg_checksum:
Now that we depend on bg_flags and bg_itable_unused to determine
the block and inode usage, we need to make sure group descriptor
is not corrupt. We add checksum to group descriptor to
detect corruption. If the descriptor is found to be corrupt, we
mark all the blocks and inodes in the group used.
Signed-off-by: Avantika Mathur <mathur@us.ibm.com>
Signed-off-by: Andreas Dilger <adilger@clusterfs.com>
Signed-off-by: Mingming Cao <cmm@us.ibm.com>
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
2007-10-17 05:38:25 +07:00
|
|
|
__le16 bg_checksum; /* crc16(sb_uuid+group+desc) */
|
2006-10-11 15:21:15 +07:00
|
|
|
__le32 bg_block_bitmap_hi; /* Blocks bitmap block MSB */
|
|
|
|
__le32 bg_inode_bitmap_hi; /* Inodes bitmap block MSB */
|
|
|
|
__le32 bg_inode_table_hi; /* Inodes table block MSB */
|
2008-01-29 11:58:27 +07:00
|
|
|
__le16 bg_free_blocks_count_hi;/* Free blocks count MSB */
|
|
|
|
__le16 bg_free_inodes_count_hi;/* Free inodes count MSB */
|
|
|
|
__le16 bg_used_dirs_count_hi; /* Directories count MSB */
|
2009-01-06 10:20:24 +07:00
|
|
|
__le16 bg_itable_unused_hi; /* Unused inodes count MSB */
|
2008-01-29 11:58:27 +07:00
|
|
|
__u32 bg_reserved2[3];
|
2006-10-11 15:20:50 +07:00
|
|
|
};
|
|
|
|
|
2008-07-12 06:27:31 +07:00
|
|
|
/*
|
|
|
|
* Structure of a flex block group info
|
|
|
|
*/
|
|
|
|
|
|
|
|
struct flex_groups {
|
2009-03-05 07:09:10 +07:00
|
|
|
atomic_t free_inodes;
|
|
|
|
atomic_t free_blocks;
|
2009-03-05 07:31:53 +07:00
|
|
|
atomic_t used_dirs;
|
2008-07-12 06:27:31 +07:00
|
|
|
};
|
|
|
|
|
Ext4: Uninitialized Block Groups
In pass1 of e2fsck, every inode table in the fileystem is scanned and checked,
regardless of whether it is in use. This is this the most time consuming part
of the filesystem check. The unintialized block group feature can greatly
reduce e2fsck time by eliminating checking of uninitialized inodes.
With this feature, there is a a high water mark of used inodes for each block
group. Block and inode bitmaps can be uninitialized on disk via a flag in the
group descriptor to avoid reading or scanning them at e2fsck time. A checksum
of each group descriptor is used to ensure that corruption in the group
descriptor's bit flags does not cause incorrect operation.
The feature is enabled through a mkfs option
mke2fs /dev/ -O uninit_groups
A patch adding support for uninitialized block groups to e2fsprogs tools has
been posted to the linux-ext4 mailing list.
The patches have been stress tested with fsstress and fsx. In performance
tests testing e2fsck time, we have seen that e2fsck time on ext3 grows
linearly with the total number of inodes in the filesytem. In ext4 with the
uninitialized block groups feature, the e2fsck time is constant, based
solely on the number of used inodes rather than the total inode count.
Since typical ext4 filesystems only use 1-10% of their inodes, this feature can
greatly reduce e2fsck time for users. With performance improvement of 2-20
times, depending on how full the filesystem is.
The attached graph shows the major improvements in e2fsck times in filesystems
with a large total inode count, but few inodes in use.
In each group descriptor if we have
EXT4_BG_INODE_UNINIT set in bg_flags:
Inode table is not initialized/used in this group. So we can skip
the consistency check during fsck.
EXT4_BG_BLOCK_UNINIT set in bg_flags:
No block in the group is used. So we can skip the block bitmap
verification for this group.
We also add two new fields to group descriptor as a part of
uninitialized group patch.
__le16 bg_itable_unused; /* Unused inodes count */
__le16 bg_checksum; /* crc16(sb_uuid+group+desc) */
bg_itable_unused:
If we have EXT4_BG_INODE_UNINIT not set in bg_flags
then bg_itable_unused will give the offset within
the inode table till the inodes are used. This can be
used by fsck to skip list of inodes that are marked unused.
bg_checksum:
Now that we depend on bg_flags and bg_itable_unused to determine
the block and inode usage, we need to make sure group descriptor
is not corrupt. We add checksum to group descriptor to
detect corruption. If the descriptor is found to be corrupt, we
mark all the blocks and inodes in the group used.
Signed-off-by: Avantika Mathur <mathur@us.ibm.com>
Signed-off-by: Andreas Dilger <adilger@clusterfs.com>
Signed-off-by: Mingming Cao <cmm@us.ibm.com>
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
2007-10-17 05:38:25 +07:00
|
|
|
#define EXT4_BG_INODE_UNINIT 0x0001 /* Inode table/bitmap not in use */
|
|
|
|
#define EXT4_BG_BLOCK_UNINIT 0x0002 /* Block bitmap not in use */
|
|
|
|
#define EXT4_BG_INODE_ZEROED 0x0004 /* On-disk itable initialized to zero */
|
|
|
|
|
2006-10-11 15:20:50 +07:00
|
|
|
/*
|
|
|
|
* Macro-instructions used to manage group descriptors
|
|
|
|
*/
|
2006-10-11 15:21:14 +07:00
|
|
|
#define EXT4_MIN_DESC_SIZE 32
|
2006-10-11 15:21:15 +07:00
|
|
|
#define EXT4_MIN_DESC_SIZE_64BIT 64
|
2006-10-11 15:21:14 +07:00
|
|
|
#define EXT4_MAX_DESC_SIZE EXT4_MIN_BLOCK_SIZE
|
|
|
|
#define EXT4_DESC_SIZE(s) (EXT4_SB(s)->s_desc_size)
|
2006-10-11 15:20:50 +07:00
|
|
|
#ifdef __KERNEL__
|
2006-10-11 15:20:53 +07:00
|
|
|
# define EXT4_BLOCKS_PER_GROUP(s) (EXT4_SB(s)->s_blocks_per_group)
|
|
|
|
# define EXT4_DESC_PER_BLOCK(s) (EXT4_SB(s)->s_desc_per_block)
|
|
|
|
# define EXT4_INODES_PER_GROUP(s) (EXT4_SB(s)->s_inodes_per_group)
|
|
|
|
# define EXT4_DESC_PER_BLOCK_BITS(s) (EXT4_SB(s)->s_desc_per_block_bits)
|
2006-10-11 15:20:50 +07:00
|
|
|
#else
|
2006-10-11 15:20:53 +07:00
|
|
|
# define EXT4_BLOCKS_PER_GROUP(s) ((s)->s_blocks_per_group)
|
2006-10-11 15:21:14 +07:00
|
|
|
# define EXT4_DESC_PER_BLOCK(s) (EXT4_BLOCK_SIZE(s) / EXT4_DESC_SIZE(s))
|
2006-10-11 15:20:53 +07:00
|
|
|
# define EXT4_INODES_PER_GROUP(s) ((s)->s_inodes_per_group)
|
2006-10-11 15:20:50 +07:00
|
|
|
#endif
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Constants relative to the data blocks
|
|
|
|
*/
|
2006-10-11 15:20:53 +07:00
|
|
|
#define EXT4_NDIR_BLOCKS 12
|
|
|
|
#define EXT4_IND_BLOCK EXT4_NDIR_BLOCKS
|
|
|
|
#define EXT4_DIND_BLOCK (EXT4_IND_BLOCK + 1)
|
|
|
|
#define EXT4_TIND_BLOCK (EXT4_DIND_BLOCK + 1)
|
|
|
|
#define EXT4_N_BLOCKS (EXT4_TIND_BLOCK + 1)
|
2006-10-11 15:20:50 +07:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Inode flags
|
|
|
|
*/
|
2006-10-11 15:20:53 +07:00
|
|
|
#define EXT4_SECRM_FL 0x00000001 /* Secure deletion */
|
|
|
|
#define EXT4_UNRM_FL 0x00000002 /* Undelete */
|
|
|
|
#define EXT4_COMPR_FL 0x00000004 /* Compress file */
|
|
|
|
#define EXT4_SYNC_FL 0x00000008 /* Synchronous updates */
|
|
|
|
#define EXT4_IMMUTABLE_FL 0x00000010 /* Immutable file */
|
|
|
|
#define EXT4_APPEND_FL 0x00000020 /* writes to file may only append */
|
|
|
|
#define EXT4_NODUMP_FL 0x00000040 /* do not dump file */
|
|
|
|
#define EXT4_NOATIME_FL 0x00000080 /* do not update atime */
|
2006-10-11 15:20:50 +07:00
|
|
|
/* Reserved for compression usage... */
|
2006-10-11 15:20:53 +07:00
|
|
|
#define EXT4_DIRTY_FL 0x00000100
|
|
|
|
#define EXT4_COMPRBLK_FL 0x00000200 /* One or more compressed clusters */
|
|
|
|
#define EXT4_NOCOMPR_FL 0x00000400 /* Don't compress */
|
|
|
|
#define EXT4_ECOMPR_FL 0x00000800 /* Compression error */
|
2006-10-11 15:20:50 +07:00
|
|
|
/* End compression flags --- maybe not all used */
|
2006-10-11 15:20:53 +07:00
|
|
|
#define EXT4_INDEX_FL 0x00001000 /* hash-indexed directory */
|
|
|
|
#define EXT4_IMAGIC_FL 0x00002000 /* AFS directory */
|
|
|
|
#define EXT4_JOURNAL_DATA_FL 0x00004000 /* file data should be journaled */
|
|
|
|
#define EXT4_NOTAIL_FL 0x00008000 /* file tail should not be merged */
|
|
|
|
#define EXT4_DIRSYNC_FL 0x00010000 /* dirsync behaviour (directories only) */
|
|
|
|
#define EXT4_TOPDIR_FL 0x00020000 /* Top of directory hierarchies*/
|
2008-01-29 11:58:27 +07:00
|
|
|
#define EXT4_HUGE_FILE_FL 0x00040000 /* Set to each huge file */
|
2006-10-11 15:21:03 +07:00
|
|
|
#define EXT4_EXTENTS_FL 0x00080000 /* Inode uses extents */
|
2008-01-29 11:58:27 +07:00
|
|
|
#define EXT4_RESERVED_FL 0x80000000 /* reserved for ext4 lib */
|
2006-10-11 15:20:50 +07:00
|
|
|
|
2006-10-11 15:21:03 +07:00
|
|
|
#define EXT4_FL_USER_VISIBLE 0x000BDFFF /* User visible flags */
|
2008-10-09 10:34:06 +07:00
|
|
|
#define EXT4_FL_USER_MODIFIABLE 0x000B80FF /* User modifiable flags */
|
2006-10-11 15:20:50 +07:00
|
|
|
|
2009-02-16 06:57:26 +07:00
|
|
|
/* Flags that should be inherited by new inodes from their parent. */
|
|
|
|
#define EXT4_FL_INHERITED (EXT4_SECRM_FL | EXT4_UNRM_FL | EXT4_COMPR_FL |\
|
|
|
|
EXT4_SYNC_FL | EXT4_IMMUTABLE_FL | EXT4_APPEND_FL |\
|
|
|
|
EXT4_NODUMP_FL | EXT4_NOATIME_FL |\
|
|
|
|
EXT4_NOCOMPR_FL | EXT4_JOURNAL_DATA_FL |\
|
|
|
|
EXT4_NOTAIL_FL | EXT4_DIRSYNC_FL)
|
|
|
|
|
2009-02-16 06:09:20 +07:00
|
|
|
/* Flags that are appropriate for regular files (all but dir-specific ones). */
|
|
|
|
#define EXT4_REG_FLMASK (~(EXT4_DIRSYNC_FL | EXT4_TOPDIR_FL))
|
|
|
|
|
|
|
|
/* Flags that are appropriate for non-directories/regular files. */
|
|
|
|
#define EXT4_OTHER_FLMASK (EXT4_NODUMP_FL | EXT4_NOATIME_FL)
|
|
|
|
|
|
|
|
/* Mask out flags that are inappropriate for the given type of inode. */
|
|
|
|
static inline __u32 ext4_mask_flags(umode_t mode, __u32 flags)
|
|
|
|
{
|
|
|
|
if (S_ISDIR(mode))
|
|
|
|
return flags;
|
|
|
|
else if (S_ISREG(mode))
|
|
|
|
return flags & EXT4_REG_FLMASK;
|
|
|
|
else
|
|
|
|
return flags & EXT4_OTHER_FLMASK;
|
|
|
|
}
|
|
|
|
|
2006-10-11 15:20:50 +07:00
|
|
|
/*
|
|
|
|
* Inode dynamic state flags
|
|
|
|
*/
|
2006-10-11 15:20:53 +07:00
|
|
|
#define EXT4_STATE_JDATA 0x00000001 /* journaled data exists */
|
|
|
|
#define EXT4_STATE_NEW 0x00000002 /* inode is newly created */
|
|
|
|
#define EXT4_STATE_XATTR 0x00000004 /* has in-inode xattrs */
|
2007-07-18 20:19:57 +07:00
|
|
|
#define EXT4_STATE_NO_EXPAND 0x00000008 /* No space for expansion */
|
2009-02-24 20:21:14 +07:00
|
|
|
#define EXT4_STATE_DA_ALLOC_CLOSE 0x00000010 /* Alloc DA blks on close */
|
2009-09-17 19:32:22 +07:00
|
|
|
#define EXT4_STATE_EXT_MIGRATE 0x00000020 /* Inode is migrating */
|
2006-10-11 15:20:50 +07:00
|
|
|
|
|
|
|
/* Used to pass group descriptor data when online resize is done */
|
2006-10-11 15:20:53 +07:00
|
|
|
struct ext4_new_group_input {
|
2007-05-25 00:04:54 +07:00
|
|
|
__u32 group; /* Group number for this data */
|
|
|
|
__u64 block_bitmap; /* Absolute block number of block bitmap */
|
|
|
|
__u64 inode_bitmap; /* Absolute block number of inode bitmap */
|
|
|
|
__u64 inode_table; /* Absolute block number of inode table start */
|
|
|
|
__u32 blocks_count; /* Total number of blocks in this group */
|
|
|
|
__u16 reserved_blocks; /* Number of reserved blocks in this group */
|
2006-10-11 15:20:50 +07:00
|
|
|
__u16 unused;
|
|
|
|
};
|
|
|
|
|
2006-10-11 15:20:53 +07:00
|
|
|
/* The struct ext4_new_group_input in kernel space, with free_blocks_count */
|
|
|
|
struct ext4_new_group_data {
|
2006-10-11 15:20:50 +07:00
|
|
|
__u32 group;
|
2006-10-11 15:21:10 +07:00
|
|
|
__u64 block_bitmap;
|
|
|
|
__u64 inode_bitmap;
|
|
|
|
__u64 inode_table;
|
2006-10-11 15:20:50 +07:00
|
|
|
__u32 blocks_count;
|
|
|
|
__u16 reserved_blocks;
|
|
|
|
__u16 unused;
|
|
|
|
__u32 free_blocks_count;
|
|
|
|
};
|
|
|
|
|
2007-07-18 08:42:41 +07:00
|
|
|
/*
|
2009-05-14 11:58:52 +07:00
|
|
|
* Flags used by ext4_get_blocks()
|
2007-07-18 08:42:41 +07:00
|
|
|
*/
|
2009-05-14 11:58:52 +07:00
|
|
|
/* Allocate any needed blocks and/or convert an unitialized
|
|
|
|
extent to be an initialized ext4 */
|
2009-05-15 00:57:08 +07:00
|
|
|
#define EXT4_GET_BLOCKS_CREATE 0x0001
|
2009-05-14 11:58:52 +07:00
|
|
|
/* Request the creation of an unitialized extent */
|
2009-05-15 00:57:08 +07:00
|
|
|
#define EXT4_GET_BLOCKS_UNINIT_EXT 0x0002
|
2009-05-14 11:58:52 +07:00
|
|
|
#define EXT4_GET_BLOCKS_CREATE_UNINIT_EXT (EXT4_GET_BLOCKS_UNINIT_EXT|\
|
|
|
|
EXT4_GET_BLOCKS_CREATE)
|
|
|
|
/* Caller is from the delayed allocation writeout path,
|
2009-05-15 00:57:08 +07:00
|
|
|
so set the magic i_delalloc_reserve_flag after taking the
|
|
|
|
inode allocation semaphore for */
|
2009-06-09 11:17:05 +07:00
|
|
|
#define EXT4_GET_BLOCKS_DELALLOC_RESERVE 0x0004
|
2009-05-15 00:57:08 +07:00
|
|
|
/* Call ext4_da_update_reserve_space() after successfully
|
|
|
|
allocating the blocks */
|
2009-06-09 11:17:05 +07:00
|
|
|
#define EXT4_GET_BLOCKS_UPDATE_RESERVE_SPACE 0x0008
|
2009-09-29 02:49:08 +07:00
|
|
|
/* caller is from the direct IO path, request to creation of an
|
|
|
|
unitialized extents if not allocated, split the uninitialized
|
|
|
|
extent if blocks has been preallocated already*/
|
|
|
|
#define EXT4_GET_BLOCKS_DIO 0x0010
|
|
|
|
#define EXT4_GET_BLOCKS_CONVERT 0x0020
|
|
|
|
#define EXT4_GET_BLOCKS_DIO_CREATE_EXT (EXT4_GET_BLOCKS_DIO|\
|
|
|
|
EXT4_GET_BLOCKS_CREATE_UNINIT_EXT)
|
|
|
|
/* Convert extent to initialized after direct IO complete */
|
|
|
|
#define EXT4_GET_BLOCKS_DIO_CONVERT_EXT (EXT4_GET_BLOCKS_CONVERT|\
|
|
|
|
EXT4_GET_BLOCKS_DIO_CREATE_EXT)
|
2006-10-11 15:20:50 +07:00
|
|
|
|
|
|
|
/*
|
|
|
|
* ioctl commands
|
|
|
|
*/
|
2006-10-11 15:20:53 +07:00
|
|
|
#define EXT4_IOC_GETFLAGS FS_IOC_GETFLAGS
|
|
|
|
#define EXT4_IOC_SETFLAGS FS_IOC_SETFLAGS
|
|
|
|
#define EXT4_IOC_GETVERSION _IOR('f', 3, long)
|
|
|
|
#define EXT4_IOC_SETVERSION _IOW('f', 4, long)
|
|
|
|
#define EXT4_IOC_GETVERSION_OLD FS_IOC_GETVERSION
|
|
|
|
#define EXT4_IOC_SETVERSION_OLD FS_IOC_SETVERSION
|
2007-07-18 19:57:06 +07:00
|
|
|
#ifdef CONFIG_JBD2_DEBUG
|
2006-10-11 15:20:53 +07:00
|
|
|
#define EXT4_IOC_WAIT_FOR_READONLY _IOR('f', 99, long)
|
2006-10-11 15:20:50 +07:00
|
|
|
#endif
|
2006-10-11 15:20:53 +07:00
|
|
|
#define EXT4_IOC_GETRSVSZ _IOR('f', 5, long)
|
|
|
|
#define EXT4_IOC_SETRSVSZ _IOW('f', 6, long)
|
2008-09-14 06:54:35 +07:00
|
|
|
#define EXT4_IOC_GROUP_EXTEND _IOW('f', 7, unsigned long)
|
|
|
|
#define EXT4_IOC_GROUP_ADD _IOW('f', 8, struct ext4_new_group_input)
|
|
|
|
#define EXT4_IOC_MIGRATE _IO('f', 9)
|
2009-02-26 13:04:07 +07:00
|
|
|
/* note ioctl 10 reserved for an early version of the FIEMAP ioctl */
|
2008-09-14 06:54:35 +07:00
|
|
|
/* note ioctl 11 reserved for filesystem-independent FIEMAP ioctl */
|
2009-02-26 13:04:07 +07:00
|
|
|
#define EXT4_IOC_ALLOC_DA_BLKS _IO('f', 12)
|
2009-06-18 06:24:03 +07:00
|
|
|
#define EXT4_IOC_MOVE_EXT _IOWR('f', 15, struct move_extent)
|
2006-10-11 15:20:50 +07:00
|
|
|
|
|
|
|
/*
|
|
|
|
* ioctl commands in 32 bit emulation
|
|
|
|
*/
|
2006-10-11 15:20:53 +07:00
|
|
|
#define EXT4_IOC32_GETFLAGS FS_IOC32_GETFLAGS
|
|
|
|
#define EXT4_IOC32_SETFLAGS FS_IOC32_SETFLAGS
|
|
|
|
#define EXT4_IOC32_GETVERSION _IOR('f', 3, int)
|
|
|
|
#define EXT4_IOC32_SETVERSION _IOW('f', 4, int)
|
|
|
|
#define EXT4_IOC32_GETRSVSZ _IOR('f', 5, int)
|
|
|
|
#define EXT4_IOC32_SETRSVSZ _IOW('f', 6, int)
|
|
|
|
#define EXT4_IOC32_GROUP_EXTEND _IOW('f', 7, unsigned int)
|
2007-07-18 19:57:06 +07:00
|
|
|
#ifdef CONFIG_JBD2_DEBUG
|
2006-10-11 15:20:53 +07:00
|
|
|
#define EXT4_IOC32_WAIT_FOR_READONLY _IOR('f', 99, int)
|
2006-10-11 15:20:50 +07:00
|
|
|
#endif
|
2006-10-11 15:20:53 +07:00
|
|
|
#define EXT4_IOC32_GETVERSION_OLD FS_IOC32_GETVERSION
|
|
|
|
#define EXT4_IOC32_SETVERSION_OLD FS_IOC32_SETVERSION
|
2006-10-11 15:20:50 +07:00
|
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Mount options
|
|
|
|
*/
|
2006-10-11 15:20:53 +07:00
|
|
|
struct ext4_mount_options {
|
2006-10-11 15:20:50 +07:00
|
|
|
unsigned long s_mount_opt;
|
|
|
|
uid_t s_resuid;
|
|
|
|
gid_t s_resgid;
|
|
|
|
unsigned long s_commit_interval;
|
2009-01-04 08:27:38 +07:00
|
|
|
u32 s_min_batch_time, s_max_batch_time;
|
2006-10-11 15:20:50 +07:00
|
|
|
#ifdef CONFIG_QUOTA
|
|
|
|
int s_jquota_fmt;
|
|
|
|
char *s_qf_names[MAXQUOTAS];
|
|
|
|
#endif
|
|
|
|
};
|
|
|
|
|
ext4: limit block allocations for indirect-block files to < 2^32
Today, the ext4 allocator will happily allocate blocks past
2^32 for indirect-block files, which results in the block
numbers getting truncated, and corruption ensues.
This patch limits such allocations to < 2^32, and adds
BUG_ONs if we do get blocks larger than that.
This should address RH Bug 519471, ext4 bitmap allocator
must limit blocks to < 2^32
* ext4_find_goal() is modified to choose a goal < UINT_MAX,
so that our starting point is in an acceptable range.
* ext4_xattr_block_set() is modified such that the goal block
is < UINT_MAX, as above.
* ext4_mb_regular_allocator() is modified so that the group
search does not continue into groups which are too high
* ext4_mb_use_preallocated() has a check that we don't use
preallocated space which is too far out
* ext4_alloc_blocks() and ext4_xattr_block_set() add some BUG_ONs
No attempt has been made to limit inode locations to < 2^32,
so we may wind up with blocks far from their inodes. Doing
this much already will lead to some odd ENOSPC issues when the
"lower 32" gets full, and further restricting inodes could
make that even weirder.
For high inodes, choosing a goal of the original, % UINT_MAX,
may be a bit odd, but then we're in an odd situation anyway,
and I don't know of a better heuristic.
Signed-off-by: Eric Sandeen <sandeen@redhat.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2009-09-17 01:45:10 +07:00
|
|
|
/* Max physical block we can addres w/o extents */
|
|
|
|
#define EXT4_MAX_BLOCK_FILE_PHYS 0xFFFFFFFF
|
|
|
|
|
2006-10-11 15:20:50 +07:00
|
|
|
/*
|
|
|
|
* Structure of an inode on the disk
|
|
|
|
*/
|
2006-10-11 15:20:53 +07:00
|
|
|
struct ext4_inode {
|
2006-10-11 15:20:50 +07:00
|
|
|
__le16 i_mode; /* File mode */
|
|
|
|
__le16 i_uid; /* Low 16 bits of Owner Uid */
|
2008-01-29 11:58:27 +07:00
|
|
|
__le32 i_size_lo; /* Size in bytes */
|
2006-10-11 15:20:50 +07:00
|
|
|
__le32 i_atime; /* Access time */
|
2007-07-18 20:15:20 +07:00
|
|
|
__le32 i_ctime; /* Inode Change time */
|
2006-10-11 15:20:50 +07:00
|
|
|
__le32 i_mtime; /* Modification time */
|
|
|
|
__le32 i_dtime; /* Deletion Time */
|
|
|
|
__le16 i_gid; /* Low 16 bits of Group Id */
|
|
|
|
__le16 i_links_count; /* Links count */
|
2008-01-29 11:58:26 +07:00
|
|
|
__le32 i_blocks_lo; /* Blocks count */
|
2006-10-11 15:20:50 +07:00
|
|
|
__le32 i_flags; /* File flags */
|
|
|
|
union {
|
|
|
|
struct {
|
2008-01-29 11:58:27 +07:00
|
|
|
__le32 l_i_version;
|
2006-10-11 15:20:50 +07:00
|
|
|
} linux1;
|
|
|
|
struct {
|
|
|
|
__u32 h_i_translator;
|
|
|
|
} hurd1;
|
|
|
|
struct {
|
|
|
|
__u32 m_i_reserved1;
|
|
|
|
} masix1;
|
|
|
|
} osd1; /* OS dependent 1 */
|
2006-10-11 15:20:53 +07:00
|
|
|
__le32 i_block[EXT4_N_BLOCKS];/* Pointers to blocks */
|
2006-10-11 15:20:50 +07:00
|
|
|
__le32 i_generation; /* File version (for NFS) */
|
2008-01-29 11:58:27 +07:00
|
|
|
__le32 i_file_acl_lo; /* File ACL */
|
2008-01-29 11:58:27 +07:00
|
|
|
__le32 i_size_high;
|
2007-10-17 05:38:25 +07:00
|
|
|
__le32 i_obso_faddr; /* Obsoleted fragment address */
|
2006-10-11 15:20:50 +07:00
|
|
|
union {
|
|
|
|
struct {
|
2008-01-29 11:58:26 +07:00
|
|
|
__le16 l_i_blocks_high; /* were l_i_reserved1 */
|
2006-10-11 15:21:09 +07:00
|
|
|
__le16 l_i_file_acl_high;
|
2007-05-25 00:04:54 +07:00
|
|
|
__le16 l_i_uid_high; /* these 2 fields */
|
2006-10-11 15:20:50 +07:00
|
|
|
__le16 l_i_gid_high; /* were reserved2[0] */
|
|
|
|
__u32 l_i_reserved2;
|
|
|
|
} linux2;
|
|
|
|
struct {
|
2007-10-17 05:38:25 +07:00
|
|
|
__le16 h_i_reserved1; /* Obsoleted fragment number/size which are removed in ext4 */
|
2006-10-11 15:20:50 +07:00
|
|
|
__u16 h_i_mode_high;
|
|
|
|
__u16 h_i_uid_high;
|
|
|
|
__u16 h_i_gid_high;
|
|
|
|
__u32 h_i_author;
|
|
|
|
} hurd2;
|
|
|
|
struct {
|
2007-10-17 05:38:25 +07:00
|
|
|
__le16 h_i_reserved1; /* Obsoleted fragment number/size which are removed in ext4 */
|
2006-10-11 15:21:09 +07:00
|
|
|
__le16 m_i_file_acl_high;
|
2006-10-11 15:20:50 +07:00
|
|
|
__u32 m_i_reserved2[2];
|
|
|
|
} masix2;
|
|
|
|
} osd2; /* OS dependent 2 */
|
|
|
|
__le16 i_extra_isize;
|
|
|
|
__le16 i_pad1;
|
2007-07-18 20:15:20 +07:00
|
|
|
__le32 i_ctime_extra; /* extra Change time (nsec << 2 | epoch) */
|
|
|
|
__le32 i_mtime_extra; /* extra Modification time(nsec << 2 | epoch) */
|
|
|
|
__le32 i_atime_extra; /* extra Access time (nsec << 2 | epoch) */
|
|
|
|
__le32 i_crtime; /* File Creation time */
|
|
|
|
__le32 i_crtime_extra; /* extra FileCreationtime (nsec << 2 | epoch) */
|
2008-01-29 11:58:27 +07:00
|
|
|
__le32 i_version_hi; /* high 32 bits for 64-bit version */
|
2006-10-11 15:20:50 +07:00
|
|
|
};
|
|
|
|
|
2009-06-18 06:24:03 +07:00
|
|
|
struct move_extent {
|
|
|
|
__u32 reserved; /* should be zero */
|
|
|
|
__u32 donor_fd; /* donor file descriptor */
|
|
|
|
__u64 orig_start; /* logical start offset in block for orig */
|
|
|
|
__u64 donor_start; /* logical start offset in block for donor */
|
|
|
|
__u64 len; /* block length to be moved */
|
|
|
|
__u64 moved_len; /* moved block length */
|
|
|
|
};
|
2006-10-11 15:20:50 +07:00
|
|
|
|
2007-07-18 20:15:20 +07:00
|
|
|
#define EXT4_EPOCH_BITS 2
|
|
|
|
#define EXT4_EPOCH_MASK ((1 << EXT4_EPOCH_BITS) - 1)
|
|
|
|
#define EXT4_NSEC_MASK (~0UL << EXT4_EPOCH_BITS)
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Extended fields will fit into an inode if the filesystem was formatted
|
|
|
|
* with large inodes (-I 256 or larger) and there are not currently any EAs
|
|
|
|
* consuming all of the available space. For new inodes we always reserve
|
|
|
|
* enough space for the kernel's known extended fields, but for inodes
|
|
|
|
* created with an old kernel this might not have been the case. None of
|
|
|
|
* the extended inode fields is critical for correct filesystem operation.
|
|
|
|
* This macro checks if a certain field fits in the inode. Note that
|
|
|
|
* inode-size = GOOD_OLD_INODE_SIZE + i_extra_isize
|
|
|
|
*/
|
|
|
|
#define EXT4_FITS_IN_INODE(ext4_inode, einode, field) \
|
|
|
|
((offsetof(typeof(*ext4_inode), field) + \
|
|
|
|
sizeof((ext4_inode)->field)) \
|
|
|
|
<= (EXT4_GOOD_OLD_INODE_SIZE + \
|
|
|
|
(einode)->i_extra_isize)) \
|
|
|
|
|
|
|
|
static inline __le32 ext4_encode_extra_time(struct timespec *time)
|
|
|
|
{
|
|
|
|
return cpu_to_le32((sizeof(time->tv_sec) > 4 ?
|
|
|
|
time->tv_sec >> 32 : 0) |
|
|
|
|
((time->tv_nsec << 2) & EXT4_NSEC_MASK));
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline void ext4_decode_extra_time(struct timespec *time, __le32 extra)
|
|
|
|
{
|
|
|
|
if (sizeof(time->tv_sec) > 4)
|
|
|
|
time->tv_sec |= (__u64)(le32_to_cpu(extra) & EXT4_EPOCH_MASK)
|
|
|
|
<< 32;
|
|
|
|
time->tv_nsec = (le32_to_cpu(extra) & EXT4_NSEC_MASK) >> 2;
|
|
|
|
}
|
|
|
|
|
|
|
|
#define EXT4_INODE_SET_XTIME(xtime, inode, raw_inode) \
|
|
|
|
do { \
|
|
|
|
(raw_inode)->xtime = cpu_to_le32((inode)->xtime.tv_sec); \
|
|
|
|
if (EXT4_FITS_IN_INODE(raw_inode, EXT4_I(inode), xtime ## _extra)) \
|
|
|
|
(raw_inode)->xtime ## _extra = \
|
|
|
|
ext4_encode_extra_time(&(inode)->xtime); \
|
|
|
|
} while (0)
|
|
|
|
|
|
|
|
#define EXT4_EINODE_SET_XTIME(xtime, einode, raw_inode) \
|
|
|
|
do { \
|
|
|
|
if (EXT4_FITS_IN_INODE(raw_inode, einode, xtime)) \
|
|
|
|
(raw_inode)->xtime = cpu_to_le32((einode)->xtime.tv_sec); \
|
|
|
|
if (EXT4_FITS_IN_INODE(raw_inode, einode, xtime ## _extra)) \
|
|
|
|
(raw_inode)->xtime ## _extra = \
|
|
|
|
ext4_encode_extra_time(&(einode)->xtime); \
|
|
|
|
} while (0)
|
|
|
|
|
|
|
|
#define EXT4_INODE_GET_XTIME(xtime, inode, raw_inode) \
|
|
|
|
do { \
|
|
|
|
(inode)->xtime.tv_sec = (signed)le32_to_cpu((raw_inode)->xtime); \
|
|
|
|
if (EXT4_FITS_IN_INODE(raw_inode, EXT4_I(inode), xtime ## _extra)) \
|
|
|
|
ext4_decode_extra_time(&(inode)->xtime, \
|
|
|
|
raw_inode->xtime ## _extra); \
|
|
|
|
} while (0)
|
|
|
|
|
|
|
|
#define EXT4_EINODE_GET_XTIME(xtime, einode, raw_inode) \
|
|
|
|
do { \
|
|
|
|
if (EXT4_FITS_IN_INODE(raw_inode, einode, xtime)) \
|
|
|
|
(einode)->xtime.tv_sec = \
|
|
|
|
(signed)le32_to_cpu((raw_inode)->xtime); \
|
|
|
|
if (EXT4_FITS_IN_INODE(raw_inode, einode, xtime ## _extra)) \
|
|
|
|
ext4_decode_extra_time(&(einode)->xtime, \
|
|
|
|
raw_inode->xtime ## _extra); \
|
|
|
|
} while (0)
|
|
|
|
|
2008-01-29 11:58:27 +07:00
|
|
|
#define i_disk_version osd1.linux1.l_i_version
|
|
|
|
|
2006-10-11 15:20:50 +07:00
|
|
|
#if defined(__KERNEL__) || defined(__linux__)
|
|
|
|
#define i_reserved1 osd1.linux1.l_i_reserved1
|
2006-10-11 15:21:09 +07:00
|
|
|
#define i_file_acl_high osd2.linux2.l_i_file_acl_high
|
2008-01-29 11:58:26 +07:00
|
|
|
#define i_blocks_high osd2.linux2.l_i_blocks_high
|
2006-10-11 15:20:50 +07:00
|
|
|
#define i_uid_low i_uid
|
|
|
|
#define i_gid_low i_gid
|
|
|
|
#define i_uid_high osd2.linux2.l_i_uid_high
|
|
|
|
#define i_gid_high osd2.linux2.l_i_gid_high
|
|
|
|
#define i_reserved2 osd2.linux2.l_i_reserved2
|
|
|
|
|
|
|
|
#elif defined(__GNU__)
|
|
|
|
|
|
|
|
#define i_translator osd1.hurd1.h_i_translator
|
|
|
|
#define i_uid_high osd2.hurd2.h_i_uid_high
|
|
|
|
#define i_gid_high osd2.hurd2.h_i_gid_high
|
|
|
|
#define i_author osd2.hurd2.h_i_author
|
|
|
|
|
|
|
|
#elif defined(__masix__)
|
|
|
|
|
|
|
|
#define i_reserved1 osd1.masix1.m_i_reserved1
|
2006-10-11 15:21:09 +07:00
|
|
|
#define i_file_acl_high osd2.masix2.m_i_file_acl_high
|
2006-10-11 15:20:50 +07:00
|
|
|
#define i_reserved2 osd2.masix2.m_i_reserved2
|
|
|
|
|
|
|
|
#endif /* defined(__KERNEL__) || defined(__linux__) */
|
|
|
|
|
2009-05-02 00:44:33 +07:00
|
|
|
/*
|
|
|
|
* storage for cached extent
|
|
|
|
*/
|
|
|
|
struct ext4_ext_cache {
|
|
|
|
ext4_fsblk_t ec_start;
|
|
|
|
ext4_lblk_t ec_block;
|
|
|
|
__u32 ec_len; /* must be 32bit to return holes */
|
|
|
|
__u32 ec_type;
|
|
|
|
};
|
|
|
|
|
|
|
|
/*
|
|
|
|
* fourth extended file system inode data in memory
|
|
|
|
*/
|
|
|
|
struct ext4_inode_info {
|
|
|
|
__le32 i_data[15]; /* unconverted */
|
|
|
|
__u32 i_flags;
|
|
|
|
ext4_fsblk_t i_file_acl;
|
|
|
|
__u32 i_dtime;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* i_block_group is the number of the block group which contains
|
|
|
|
* this file's inode. Constant across the lifetime of the inode,
|
|
|
|
* it is ued for making block allocation decisions - we try to
|
|
|
|
* place a file's data blocks near its inode block, and new inodes
|
|
|
|
* near to their parent directory's inode.
|
|
|
|
*/
|
|
|
|
ext4_group_t i_block_group;
|
|
|
|
__u32 i_state; /* Dynamic state flags for ext4 */
|
|
|
|
|
|
|
|
ext4_lblk_t i_dir_start_lookup;
|
|
|
|
#ifdef CONFIG_EXT4_FS_XATTR
|
|
|
|
/*
|
|
|
|
* Extended attributes can be read independently of the main file
|
|
|
|
* data. Taking i_mutex even when reading would cause contention
|
|
|
|
* between readers of EAs and writers of regular file data, so
|
|
|
|
* instead we synchronize on xattr_sem when reading or changing
|
|
|
|
* EAs.
|
|
|
|
*/
|
|
|
|
struct rw_semaphore xattr_sem;
|
|
|
|
#endif
|
|
|
|
|
|
|
|
struct list_head i_orphan; /* unlinked but open inodes */
|
|
|
|
|
|
|
|
/*
|
|
|
|
* i_disksize keeps track of what the inode size is ON DISK, not
|
|
|
|
* in memory. During truncate, i_size is set to the new size by
|
|
|
|
* the VFS prior to calling ext4_truncate(), but the filesystem won't
|
|
|
|
* set i_disksize to 0 until the truncate is actually under way.
|
|
|
|
*
|
|
|
|
* The intent is that i_disksize always represents the blocks which
|
|
|
|
* are used by this file. This allows recovery to restart truncate
|
|
|
|
* on orphans if we crash during truncate. We actually write i_disksize
|
|
|
|
* into the on-disk inode when writing inodes out, instead of i_size.
|
|
|
|
*
|
|
|
|
* The only time when i_disksize and i_size may be different is when
|
|
|
|
* a truncate is in progress. The only things which change i_disksize
|
|
|
|
* are ext4_get_block (growth) and ext4_truncate (shrinkth).
|
|
|
|
*/
|
|
|
|
loff_t i_disksize;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* i_data_sem is for serialising ext4_truncate() against
|
|
|
|
* ext4_getblock(). In the 2.4 ext2 design, great chunks of inode's
|
|
|
|
* data tree are chopped off during truncate. We can't do that in
|
|
|
|
* ext4 because whenever we perform intermediate commits during
|
|
|
|
* truncate, the inode and all the metadata blocks *must* be in a
|
|
|
|
* consistent state which allows truncation of the orphans to restart
|
|
|
|
* during recovery. Hence we must fix the get_block-vs-truncate race
|
|
|
|
* by other means, so we have i_data_sem.
|
|
|
|
*/
|
|
|
|
struct rw_semaphore i_data_sem;
|
|
|
|
struct inode vfs_inode;
|
|
|
|
struct jbd2_inode jinode;
|
|
|
|
|
|
|
|
struct ext4_ext_cache i_cached_extent;
|
|
|
|
/*
|
|
|
|
* File creation time. Its function is same as that of
|
|
|
|
* struct timespec i_{a,c,m}time in the generic inode.
|
|
|
|
*/
|
|
|
|
struct timespec i_crtime;
|
|
|
|
|
|
|
|
/* mballoc */
|
|
|
|
struct list_head i_prealloc_list;
|
|
|
|
spinlock_t i_prealloc_lock;
|
|
|
|
|
|
|
|
/* ialloc */
|
|
|
|
ext4_group_t i_last_alloc_group;
|
|
|
|
|
|
|
|
/* allocation reservation info for delalloc */
|
|
|
|
unsigned int i_reserved_data_blocks;
|
|
|
|
unsigned int i_reserved_meta_blocks;
|
|
|
|
unsigned int i_allocated_meta_blocks;
|
|
|
|
unsigned short i_delalloc_reserved_flag;
|
|
|
|
|
|
|
|
/* on-disk additional length */
|
|
|
|
__u16 i_extra_isize;
|
|
|
|
|
|
|
|
spinlock_t i_block_reservation_lock;
|
|
|
|
};
|
|
|
|
|
2006-10-11 15:20:50 +07:00
|
|
|
/*
|
|
|
|
* File system states
|
|
|
|
*/
|
2006-10-11 15:20:53 +07:00
|
|
|
#define EXT4_VALID_FS 0x0001 /* Unmounted cleanly */
|
|
|
|
#define EXT4_ERROR_FS 0x0002 /* Errors detected */
|
|
|
|
#define EXT4_ORPHAN_FS 0x0004 /* Orphans being recovered */
|
2006-10-11 15:20:50 +07:00
|
|
|
|
2008-02-10 13:11:44 +07:00
|
|
|
/*
|
|
|
|
* Misc. filesystem flags
|
|
|
|
*/
|
|
|
|
#define EXT2_FLAGS_SIGNED_HASH 0x0001 /* Signed dirhash in use */
|
|
|
|
#define EXT2_FLAGS_UNSIGNED_HASH 0x0002 /* Unsigned dirhash in use */
|
|
|
|
#define EXT2_FLAGS_TEST_FILESYS 0x0004 /* to test development code */
|
|
|
|
|
2006-10-11 15:20:50 +07:00
|
|
|
/*
|
|
|
|
* Mount flags
|
|
|
|
*/
|
2006-10-11 15:20:53 +07:00
|
|
|
#define EXT4_MOUNT_OLDALLOC 0x00002 /* Don't use the new Orlov allocator */
|
|
|
|
#define EXT4_MOUNT_GRPID 0x00004 /* Create files with directory's group */
|
|
|
|
#define EXT4_MOUNT_DEBUG 0x00008 /* Some debugging messages */
|
|
|
|
#define EXT4_MOUNT_ERRORS_CONT 0x00010 /* Continue on errors */
|
|
|
|
#define EXT4_MOUNT_ERRORS_RO 0x00020 /* Remount fs ro on errors */
|
|
|
|
#define EXT4_MOUNT_ERRORS_PANIC 0x00040 /* Panic on errors */
|
|
|
|
#define EXT4_MOUNT_MINIX_DF 0x00080 /* Mimics the Minix statfs */
|
|
|
|
#define EXT4_MOUNT_NOLOAD 0x00100 /* Don't use existing journal*/
|
|
|
|
#define EXT4_MOUNT_DATA_FLAGS 0x00C00 /* Mode for data writes: */
|
|
|
|
#define EXT4_MOUNT_JOURNAL_DATA 0x00400 /* Write data to journal */
|
|
|
|
#define EXT4_MOUNT_ORDERED_DATA 0x00800 /* Flush data before commit */
|
|
|
|
#define EXT4_MOUNT_WRITEBACK_DATA 0x00C00 /* No data ordering */
|
|
|
|
#define EXT4_MOUNT_UPDATE_JOURNAL 0x01000 /* Update the journal format */
|
|
|
|
#define EXT4_MOUNT_NO_UID32 0x02000 /* Disable 32-bit UIDs */
|
|
|
|
#define EXT4_MOUNT_XATTR_USER 0x04000 /* Extended user attributes */
|
|
|
|
#define EXT4_MOUNT_POSIX_ACL 0x08000 /* POSIX Access Control Lists */
|
2009-03-17 10:12:23 +07:00
|
|
|
#define EXT4_MOUNT_NO_AUTO_DA_ALLOC 0x10000 /* No auto delalloc mapping */
|
2006-10-11 15:20:53 +07:00
|
|
|
#define EXT4_MOUNT_BARRIER 0x20000 /* Use block barriers */
|
|
|
|
#define EXT4_MOUNT_NOBH 0x40000 /* No bufferheads */
|
|
|
|
#define EXT4_MOUNT_QUOTA 0x80000 /* Some quota option set */
|
|
|
|
#define EXT4_MOUNT_USRQUOTA 0x100000 /* "old" user quota */
|
|
|
|
#define EXT4_MOUNT_GRPQUOTA 0x200000 /* "old" group quota */
|
2008-01-29 11:58:27 +07:00
|
|
|
#define EXT4_MOUNT_JOURNAL_ASYNC_COMMIT 0x1000000 /* Journal Async Commit */
|
2008-01-29 11:58:27 +07:00
|
|
|
#define EXT4_MOUNT_I_VERSION 0x2000000 /* i_version support */
|
2008-07-12 06:27:31 +07:00
|
|
|
#define EXT4_MOUNT_DELALLOC 0x8000000 /* Delalloc support */
|
2008-10-11 09:12:43 +07:00
|
|
|
#define EXT4_MOUNT_DATA_ERR_ABORT 0x10000000 /* Abort on file data write */
|
2009-05-18 02:38:01 +07:00
|
|
|
#define EXT4_MOUNT_BLOCK_VALIDITY 0x20000000 /* Block validity checking */
|
2008-10-11 09:12:43 +07:00
|
|
|
|
2006-10-11 15:20:53 +07:00
|
|
|
#define clear_opt(o, opt) o &= ~EXT4_MOUNT_##opt
|
|
|
|
#define set_opt(o, opt) o |= EXT4_MOUNT_##opt
|
|
|
|
#define test_opt(sb, opt) (EXT4_SB(sb)->s_mount_opt & \
|
|
|
|
EXT4_MOUNT_##opt)
|
2006-10-11 15:20:50 +07:00
|
|
|
|
2006-10-11 15:20:53 +07:00
|
|
|
#define ext4_set_bit ext2_set_bit
|
|
|
|
#define ext4_set_bit_atomic ext2_set_bit_atomic
|
|
|
|
#define ext4_clear_bit ext2_clear_bit
|
|
|
|
#define ext4_clear_bit_atomic ext2_clear_bit_atomic
|
|
|
|
#define ext4_test_bit ext2_test_bit
|
|
|
|
#define ext4_find_first_zero_bit ext2_find_first_zero_bit
|
|
|
|
#define ext4_find_next_zero_bit ext2_find_next_zero_bit
|
2008-01-29 11:58:27 +07:00
|
|
|
#define ext4_find_next_bit ext2_find_next_bit
|
2006-10-11 15:20:50 +07:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Maximal mount counts between two filesystem checks
|
|
|
|
*/
|
2006-10-11 15:20:53 +07:00
|
|
|
#define EXT4_DFL_MAX_MNT_COUNT 20 /* Allow 20 mounts */
|
|
|
|
#define EXT4_DFL_CHECKINTERVAL 0 /* Don't use interval check */
|
2006-10-11 15:20:50 +07:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Behaviour when detecting errors
|
|
|
|
*/
|
2006-10-11 15:20:53 +07:00
|
|
|
#define EXT4_ERRORS_CONTINUE 1 /* Continue execution */
|
|
|
|
#define EXT4_ERRORS_RO 2 /* Remount fs read-only */
|
|
|
|
#define EXT4_ERRORS_PANIC 3 /* Panic */
|
|
|
|
#define EXT4_ERRORS_DEFAULT EXT4_ERRORS_CONTINUE
|
2006-10-11 15:20:50 +07:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Structure of the super block
|
|
|
|
*/
|
2006-10-11 15:20:53 +07:00
|
|
|
struct ext4_super_block {
|
2006-10-11 15:20:50 +07:00
|
|
|
/*00*/ __le32 s_inodes_count; /* Inodes count */
|
2007-10-17 05:38:25 +07:00
|
|
|
__le32 s_blocks_count_lo; /* Blocks count */
|
2007-10-17 05:38:25 +07:00
|
|
|
__le32 s_r_blocks_count_lo; /* Reserved blocks count */
|
|
|
|
__le32 s_free_blocks_count_lo; /* Free blocks count */
|
2006-10-11 15:20:50 +07:00
|
|
|
/*10*/ __le32 s_free_inodes_count; /* Free inodes count */
|
|
|
|
__le32 s_first_data_block; /* First Data Block */
|
|
|
|
__le32 s_log_block_size; /* Block size */
|
2007-10-17 05:38:25 +07:00
|
|
|
__le32 s_obso_log_frag_size; /* Obsoleted fragment size */
|
2006-10-11 15:20:50 +07:00
|
|
|
/*20*/ __le32 s_blocks_per_group; /* # Blocks per group */
|
2007-10-17 05:38:25 +07:00
|
|
|
__le32 s_obso_frags_per_group; /* Obsoleted fragments per group */
|
2006-10-11 15:20:50 +07:00
|
|
|
__le32 s_inodes_per_group; /* # Inodes per group */
|
|
|
|
__le32 s_mtime; /* Mount time */
|
|
|
|
/*30*/ __le32 s_wtime; /* Write time */
|
|
|
|
__le16 s_mnt_count; /* Mount count */
|
|
|
|
__le16 s_max_mnt_count; /* Maximal mount count */
|
|
|
|
__le16 s_magic; /* Magic signature */
|
|
|
|
__le16 s_state; /* File system state */
|
|
|
|
__le16 s_errors; /* Behaviour when detecting errors */
|
|
|
|
__le16 s_minor_rev_level; /* minor revision level */
|
|
|
|
/*40*/ __le32 s_lastcheck; /* time of last check */
|
|
|
|
__le32 s_checkinterval; /* max. time between checks */
|
|
|
|
__le32 s_creator_os; /* OS */
|
|
|
|
__le32 s_rev_level; /* Revision level */
|
|
|
|
/*50*/ __le16 s_def_resuid; /* Default uid for reserved blocks */
|
|
|
|
__le16 s_def_resgid; /* Default gid for reserved blocks */
|
|
|
|
/*
|
2006-10-11 15:20:53 +07:00
|
|
|
* These fields are for EXT4_DYNAMIC_REV superblocks only.
|
2006-10-11 15:20:50 +07:00
|
|
|
*
|
|
|
|
* Note: the difference between the compatible feature set and
|
|
|
|
* the incompatible feature set is that if there is a bit set
|
|
|
|
* in the incompatible feature set that the kernel doesn't
|
|
|
|
* know about, it should refuse to mount the filesystem.
|
|
|
|
*
|
|
|
|
* e2fsck's requirements are more strict; if it doesn't know
|
|
|
|
* about a feature in either the compatible or incompatible
|
|
|
|
* feature set, it must abort and not try to meddle with
|
|
|
|
* things it doesn't understand...
|
|
|
|
*/
|
|
|
|
__le32 s_first_ino; /* First non-reserved inode */
|
2006-10-11 15:21:14 +07:00
|
|
|
__le16 s_inode_size; /* size of inode structure */
|
2006-10-11 15:20:50 +07:00
|
|
|
__le16 s_block_group_nr; /* block group # of this superblock */
|
|
|
|
__le32 s_feature_compat; /* compatible feature set */
|
|
|
|
/*60*/ __le32 s_feature_incompat; /* incompatible feature set */
|
|
|
|
__le32 s_feature_ro_compat; /* readonly-compatible feature set */
|
|
|
|
/*68*/ __u8 s_uuid[16]; /* 128-bit uuid for volume */
|
|
|
|
/*78*/ char s_volume_name[16]; /* volume name */
|
|
|
|
/*88*/ char s_last_mounted[64]; /* directory where last mounted */
|
|
|
|
/*C8*/ __le32 s_algorithm_usage_bitmap; /* For compression */
|
|
|
|
/*
|
|
|
|
* Performance hints. Directory preallocation should only
|
2006-10-11 15:20:53 +07:00
|
|
|
* happen if the EXT4_FEATURE_COMPAT_DIR_PREALLOC flag is on.
|
2006-10-11 15:20:50 +07:00
|
|
|
*/
|
|
|
|
__u8 s_prealloc_blocks; /* Nr of blocks to try to preallocate*/
|
|
|
|
__u8 s_prealloc_dir_blocks; /* Nr to preallocate for dirs */
|
|
|
|
__le16 s_reserved_gdt_blocks; /* Per group desc for online growth */
|
|
|
|
/*
|
2006-10-11 15:20:53 +07:00
|
|
|
* Journaling support valid if EXT4_FEATURE_COMPAT_HAS_JOURNAL set.
|
2006-10-11 15:20:50 +07:00
|
|
|
*/
|
|
|
|
/*D0*/ __u8 s_journal_uuid[16]; /* uuid of journal superblock */
|
|
|
|
/*E0*/ __le32 s_journal_inum; /* inode number of journal file */
|
|
|
|
__le32 s_journal_dev; /* device number of journal file */
|
|
|
|
__le32 s_last_orphan; /* start of list of inodes to delete */
|
|
|
|
__le32 s_hash_seed[4]; /* HTREE hash seed */
|
|
|
|
__u8 s_def_hash_version; /* Default hash version to use */
|
|
|
|
__u8 s_reserved_char_pad;
|
2006-10-11 15:21:14 +07:00
|
|
|
__le16 s_desc_size; /* size of group descriptor */
|
2006-10-11 15:21:10 +07:00
|
|
|
/*100*/ __le32 s_default_mount_opts;
|
2006-10-11 15:20:50 +07:00
|
|
|
__le32 s_first_meta_bg; /* First metablock block group */
|
2006-10-11 15:21:10 +07:00
|
|
|
__le32 s_mkfs_time; /* When the filesystem was created */
|
|
|
|
__le32 s_jnl_blocks[17]; /* Backup of the journal inode */
|
|
|
|
/* 64bit support valid if EXT4_FEATURE_COMPAT_64BIT */
|
|
|
|
/*150*/ __le32 s_blocks_count_hi; /* Blocks count */
|
|
|
|
__le32 s_r_blocks_count_hi; /* Reserved blocks count */
|
|
|
|
__le32 s_free_blocks_count_hi; /* Free blocks count */
|
2007-10-17 05:38:25 +07:00
|
|
|
__le16 s_min_extra_isize; /* All inodes have at least # bytes */
|
|
|
|
__le16 s_want_extra_isize; /* New inodes should reserve # bytes */
|
|
|
|
__le32 s_flags; /* Miscellaneous flags */
|
|
|
|
__le16 s_raid_stride; /* RAID stride */
|
|
|
|
__le16 s_mmp_interval; /* # seconds to wait in MMP checking */
|
|
|
|
__le64 s_mmp_block; /* Block for multi-mount protection */
|
|
|
|
__le32 s_raid_stripe_width; /* blocks on all data disks (N*stride)*/
|
2008-07-12 06:27:31 +07:00
|
|
|
__u8 s_log_groups_per_flex; /* FLEX_BG group size */
|
|
|
|
__u8 s_reserved_char_pad2;
|
|
|
|
__le16 s_reserved_pad;
|
2009-03-01 07:39:58 +07:00
|
|
|
__le64 s_kbytes_written; /* nr of lifetime kilobytes written */
|
|
|
|
__u32 s_reserved[160]; /* Padding to the end of the block */
|
2006-10-11 15:20:50 +07:00
|
|
|
};
|
|
|
|
|
|
|
|
#ifdef __KERNEL__
|
2009-06-13 21:09:48 +07:00
|
|
|
|
|
|
|
/*
|
2009-06-13 21:09:36 +07:00
|
|
|
* run-time mount flags
|
2009-06-13 21:09:48 +07:00
|
|
|
*/
|
|
|
|
#define EXT4_MF_MNTDIR_SAMPLED 0x0001
|
2009-06-13 21:09:36 +07:00
|
|
|
#define EXT4_MF_FS_ABORTED 0x0002 /* Fatal error detected */
|
2009-06-13 21:09:48 +07:00
|
|
|
|
2009-05-04 03:33:44 +07:00
|
|
|
/*
|
|
|
|
* fourth extended-fs super-block data in memory
|
|
|
|
*/
|
|
|
|
struct ext4_sb_info {
|
|
|
|
unsigned long s_desc_size; /* Size of a group descriptor in bytes */
|
|
|
|
unsigned long s_inodes_per_block;/* Number of inodes per block */
|
|
|
|
unsigned long s_blocks_per_group;/* Number of blocks in a group */
|
|
|
|
unsigned long s_inodes_per_group;/* Number of inodes in a group */
|
|
|
|
unsigned long s_itb_per_group; /* Number of inode table blocks per group */
|
|
|
|
unsigned long s_gdb_count; /* Number of group descriptor blocks */
|
|
|
|
unsigned long s_desc_per_block; /* Number of group descriptors per block */
|
|
|
|
ext4_group_t s_groups_count; /* Number of groups in the fs */
|
ext4: limit block allocations for indirect-block files to < 2^32
Today, the ext4 allocator will happily allocate blocks past
2^32 for indirect-block files, which results in the block
numbers getting truncated, and corruption ensues.
This patch limits such allocations to < 2^32, and adds
BUG_ONs if we do get blocks larger than that.
This should address RH Bug 519471, ext4 bitmap allocator
must limit blocks to < 2^32
* ext4_find_goal() is modified to choose a goal < UINT_MAX,
so that our starting point is in an acceptable range.
* ext4_xattr_block_set() is modified such that the goal block
is < UINT_MAX, as above.
* ext4_mb_regular_allocator() is modified so that the group
search does not continue into groups which are too high
* ext4_mb_use_preallocated() has a check that we don't use
preallocated space which is too far out
* ext4_alloc_blocks() and ext4_xattr_block_set() add some BUG_ONs
No attempt has been made to limit inode locations to < 2^32,
so we may wind up with blocks far from their inodes. Doing
this much already will lead to some odd ENOSPC issues when the
"lower 32" gets full, and further restricting inodes could
make that even weirder.
For high inodes, choosing a goal of the original, % UINT_MAX,
may be a bit odd, but then we're in an odd situation anyway,
and I don't know of a better heuristic.
Signed-off-by: Eric Sandeen <sandeen@redhat.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2009-09-17 01:45:10 +07:00
|
|
|
ext4_group_t s_blockfile_groups;/* Groups acceptable for non-extent files */
|
2009-05-04 03:33:44 +07:00
|
|
|
unsigned long s_overhead_last; /* Last calculated overhead */
|
|
|
|
unsigned long s_blocks_last; /* Last seen block count */
|
|
|
|
loff_t s_bitmap_maxbytes; /* max bytes for bitmap files */
|
|
|
|
struct buffer_head * s_sbh; /* Buffer containing the super block */
|
|
|
|
struct ext4_super_block *s_es; /* Pointer to the super block in the buffer */
|
|
|
|
struct buffer_head **s_group_desc;
|
2009-06-13 21:09:41 +07:00
|
|
|
unsigned int s_mount_opt;
|
2009-06-13 21:09:48 +07:00
|
|
|
unsigned int s_mount_flags;
|
2009-05-04 03:33:44 +07:00
|
|
|
ext4_fsblk_t s_sb_block;
|
|
|
|
uid_t s_resuid;
|
|
|
|
gid_t s_resgid;
|
|
|
|
unsigned short s_mount_state;
|
|
|
|
unsigned short s_pad;
|
|
|
|
int s_addr_per_block_bits;
|
|
|
|
int s_desc_per_block_bits;
|
|
|
|
int s_inode_size;
|
|
|
|
int s_first_ino;
|
|
|
|
unsigned int s_inode_readahead_blks;
|
2009-06-13 22:45:35 +07:00
|
|
|
unsigned int s_inode_goal;
|
2009-05-04 03:33:44 +07:00
|
|
|
spinlock_t s_next_gen_lock;
|
|
|
|
u32 s_next_generation;
|
|
|
|
u32 s_hash_seed[4];
|
|
|
|
int s_def_hash_version;
|
|
|
|
int s_hash_unsigned; /* 3 if hash should be signed, 0 if not */
|
|
|
|
struct percpu_counter s_freeblocks_counter;
|
|
|
|
struct percpu_counter s_freeinodes_counter;
|
|
|
|
struct percpu_counter s_dirs_counter;
|
|
|
|
struct percpu_counter s_dirtyblocks_counter;
|
|
|
|
struct blockgroup_lock *s_blockgroup_lock;
|
|
|
|
struct proc_dir_entry *s_proc;
|
|
|
|
struct kobject s_kobj;
|
|
|
|
struct completion s_kobj_unregister;
|
|
|
|
|
|
|
|
/* Journaling */
|
|
|
|
struct inode *s_journal_inode;
|
|
|
|
struct journal_s *s_journal;
|
|
|
|
struct list_head s_orphan;
|
|
|
|
struct mutex s_orphan_lock;
|
|
|
|
struct mutex s_resize_lock;
|
|
|
|
unsigned long s_commit_interval;
|
|
|
|
u32 s_max_batch_time;
|
|
|
|
u32 s_min_batch_time;
|
|
|
|
struct block_device *journal_bdev;
|
|
|
|
#ifdef CONFIG_JBD2_DEBUG
|
|
|
|
struct timer_list turn_ro_timer; /* For turning read-only (crash simulation) */
|
|
|
|
wait_queue_head_t ro_wait_queue; /* For people waiting for the fs to go read-only */
|
|
|
|
#endif
|
|
|
|
#ifdef CONFIG_QUOTA
|
|
|
|
char *s_qf_names[MAXQUOTAS]; /* Names of quota files with journalled quota */
|
|
|
|
int s_jquota_fmt; /* Format of quota to use */
|
|
|
|
#endif
|
|
|
|
unsigned int s_want_extra_isize; /* New inodes should reserve # bytes */
|
2009-05-18 02:38:01 +07:00
|
|
|
struct rb_root system_blks;
|
2009-05-04 03:33:44 +07:00
|
|
|
|
|
|
|
#ifdef EXTENTS_STATS
|
|
|
|
/* ext4 extents stats */
|
|
|
|
unsigned long s_ext_min;
|
|
|
|
unsigned long s_ext_max;
|
|
|
|
unsigned long s_depth_max;
|
|
|
|
spinlock_t s_ext_stats_lock;
|
|
|
|
unsigned long s_ext_blocks;
|
|
|
|
unsigned long s_ext_extents;
|
|
|
|
#endif
|
|
|
|
|
|
|
|
/* for buddy allocator */
|
|
|
|
struct ext4_group_info ***s_group_info;
|
|
|
|
struct inode *s_buddy_cache;
|
|
|
|
long s_blocks_reserved;
|
|
|
|
spinlock_t s_reserve_lock;
|
|
|
|
spinlock_t s_md_lock;
|
|
|
|
tid_t s_last_transaction;
|
|
|
|
unsigned short *s_mb_offsets;
|
|
|
|
unsigned int *s_mb_maxs;
|
|
|
|
|
|
|
|
/* tunables */
|
|
|
|
unsigned long s_stripe;
|
|
|
|
unsigned int s_mb_stream_request;
|
|
|
|
unsigned int s_mb_max_to_scan;
|
|
|
|
unsigned int s_mb_min_to_scan;
|
|
|
|
unsigned int s_mb_stats;
|
|
|
|
unsigned int s_mb_order2_reqs;
|
|
|
|
unsigned int s_mb_group_prealloc;
|
2009-09-30 00:31:31 +07:00
|
|
|
unsigned int s_max_writeback_mb_bump;
|
2009-05-04 03:33:44 +07:00
|
|
|
/* where last allocation was done - for stream allocation */
|
|
|
|
unsigned long s_mb_last_group;
|
|
|
|
unsigned long s_mb_last_start;
|
|
|
|
|
|
|
|
/* history to debug policy */
|
|
|
|
struct ext4_mb_history *s_mb_history;
|
|
|
|
int s_mb_history_cur;
|
|
|
|
int s_mb_history_max;
|
|
|
|
int s_mb_history_num;
|
|
|
|
spinlock_t s_mb_history_lock;
|
|
|
|
int s_mb_history_filter;
|
|
|
|
|
|
|
|
/* stats for buddy allocator */
|
|
|
|
spinlock_t s_mb_pa_lock;
|
|
|
|
atomic_t s_bal_reqs; /* number of reqs with len > 1 */
|
|
|
|
atomic_t s_bal_success; /* we found long enough chunks */
|
|
|
|
atomic_t s_bal_allocated; /* in blocks */
|
|
|
|
atomic_t s_bal_ex_scanned; /* total extents scanned */
|
|
|
|
atomic_t s_bal_goals; /* goal hits */
|
|
|
|
atomic_t s_bal_breaks; /* too long searches */
|
|
|
|
atomic_t s_bal_2orders; /* 2^order hits */
|
|
|
|
spinlock_t s_bal_lock;
|
|
|
|
unsigned long s_mb_buddies_generated;
|
|
|
|
unsigned long long s_mb_generation_time;
|
|
|
|
atomic_t s_mb_lost_chunks;
|
|
|
|
atomic_t s_mb_preallocated;
|
|
|
|
atomic_t s_mb_discarded;
|
ext4: Avoid group preallocation for closed files
Currently the group preallocation code tries to find a large (512)
free block from which to do per-cpu group allocation for small files.
The problem with this scheme is that it leaves the filesystem horribly
fragmented. In the worst case, if the filesystem is unmounted and
remounted (after a system shutdown, for example) we forget the fact
that wee were using a particular (now-partially filled) 512 block
extent. So the next time we try to allocate space for a small file,
we will find *another* completely free 512 block chunk to allocate
small files. Given that there are 32,768 blocks in a block group,
after 64 iterations of "mount, write one 4k file in a directory,
unmount", the block group will have 64 files, each separated by 511
blocks, and the block group will no longer have any free 512
completely free chunks of blocks for group preallocation space.
So if we try to allocate blocks for a file that has been closed, such
that we know the final size of the file, and the filesystem is not
busy, avoid using group preallocation.
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2009-09-19 00:34:02 +07:00
|
|
|
atomic_t s_lock_busy;
|
2009-05-04 03:33:44 +07:00
|
|
|
|
|
|
|
/* locality groups */
|
|
|
|
struct ext4_locality_group *s_locality_groups;
|
|
|
|
|
|
|
|
/* for write statistics */
|
|
|
|
unsigned long s_sectors_written_start;
|
|
|
|
u64 s_kbytes_written;
|
|
|
|
|
|
|
|
unsigned int s_log_groups_per_flex;
|
|
|
|
struct flex_groups *s_flex_groups;
|
2009-09-29 02:48:41 +07:00
|
|
|
|
|
|
|
/* workqueue for dio unwritten */
|
|
|
|
struct workqueue_struct *dio_unwritten_wq;
|
2009-05-04 03:33:44 +07:00
|
|
|
};
|
|
|
|
|
2008-09-09 09:25:24 +07:00
|
|
|
static inline struct ext4_sb_info *EXT4_SB(struct super_block *sb)
|
2006-10-11 15:20:50 +07:00
|
|
|
{
|
|
|
|
return sb->s_fs_info;
|
|
|
|
}
|
2006-10-11 15:20:53 +07:00
|
|
|
static inline struct ext4_inode_info *EXT4_I(struct inode *inode)
|
2006-10-11 15:20:50 +07:00
|
|
|
{
|
2006-10-11 15:20:53 +07:00
|
|
|
return container_of(inode, struct ext4_inode_info, vfs_inode);
|
2006-10-11 15:20:50 +07:00
|
|
|
}
|
|
|
|
|
2007-07-18 20:15:20 +07:00
|
|
|
static inline struct timespec ext4_current_time(struct inode *inode)
|
|
|
|
{
|
|
|
|
return (inode->i_sb->s_time_gran < NSEC_PER_SEC) ?
|
|
|
|
current_fs_time(inode->i_sb) : CURRENT_TIME_SEC;
|
|
|
|
}
|
|
|
|
|
2006-10-11 15:20:53 +07:00
|
|
|
static inline int ext4_valid_inum(struct super_block *sb, unsigned long ino)
|
2006-10-11 15:20:50 +07:00
|
|
|
{
|
2006-10-11 15:20:53 +07:00
|
|
|
return ino == EXT4_ROOT_INO ||
|
|
|
|
ino == EXT4_JOURNAL_INO ||
|
|
|
|
ino == EXT4_RESIZE_INO ||
|
|
|
|
(ino >= EXT4_FIRST_INO(sb) &&
|
|
|
|
ino <= le32_to_cpu(EXT4_SB(sb)->s_es->s_inodes_count));
|
2006-10-11 15:20:50 +07:00
|
|
|
}
|
|
|
|
#else
|
2006-10-11 15:20:53 +07:00
|
|
|
/* Assume that user mode programs are passing in an ext4fs superblock, not
|
2006-10-11 15:20:50 +07:00
|
|
|
* a kernel struct super_block. This will allow us to call the feature-test
|
|
|
|
* macros from user land. */
|
2006-10-11 15:20:53 +07:00
|
|
|
#define EXT4_SB(sb) (sb)
|
2006-10-11 15:20:50 +07:00
|
|
|
#endif
|
|
|
|
|
2006-10-11 15:20:53 +07:00
|
|
|
#define NEXT_ORPHAN(inode) EXT4_I(inode)->i_dtime
|
2006-10-11 15:20:50 +07:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Codes for operating systems
|
|
|
|
*/
|
2006-10-11 15:20:53 +07:00
|
|
|
#define EXT4_OS_LINUX 0
|
|
|
|
#define EXT4_OS_HURD 1
|
|
|
|
#define EXT4_OS_MASIX 2
|
|
|
|
#define EXT4_OS_FREEBSD 3
|
|
|
|
#define EXT4_OS_LITES 4
|
2006-10-11 15:20:50 +07:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Revision levels
|
|
|
|
*/
|
2006-10-11 15:20:53 +07:00
|
|
|
#define EXT4_GOOD_OLD_REV 0 /* The good old (original) format */
|
|
|
|
#define EXT4_DYNAMIC_REV 1 /* V2 format w/ dynamic inode sizes */
|
2006-10-11 15:20:50 +07:00
|
|
|
|
2006-10-11 15:20:53 +07:00
|
|
|
#define EXT4_CURRENT_REV EXT4_GOOD_OLD_REV
|
|
|
|
#define EXT4_MAX_SUPP_REV EXT4_DYNAMIC_REV
|
2006-10-11 15:20:50 +07:00
|
|
|
|
2006-10-11 15:20:53 +07:00
|
|
|
#define EXT4_GOOD_OLD_INODE_SIZE 128
|
2006-10-11 15:20:50 +07:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Feature set definitions
|
|
|
|
*/
|
|
|
|
|
2006-10-11 15:20:53 +07:00
|
|
|
#define EXT4_HAS_COMPAT_FEATURE(sb,mask) \
|
2008-11-23 03:04:59 +07:00
|
|
|
((EXT4_SB(sb)->s_es->s_feature_compat & cpu_to_le32(mask)) != 0)
|
2006-10-11 15:20:53 +07:00
|
|
|
#define EXT4_HAS_RO_COMPAT_FEATURE(sb,mask) \
|
2008-11-23 03:04:59 +07:00
|
|
|
((EXT4_SB(sb)->s_es->s_feature_ro_compat & cpu_to_le32(mask)) != 0)
|
2006-10-11 15:20:53 +07:00
|
|
|
#define EXT4_HAS_INCOMPAT_FEATURE(sb,mask) \
|
2008-11-23 03:04:59 +07:00
|
|
|
((EXT4_SB(sb)->s_es->s_feature_incompat & cpu_to_le32(mask)) != 0)
|
2006-10-11 15:20:53 +07:00
|
|
|
#define EXT4_SET_COMPAT_FEATURE(sb,mask) \
|
|
|
|
EXT4_SB(sb)->s_es->s_feature_compat |= cpu_to_le32(mask)
|
|
|
|
#define EXT4_SET_RO_COMPAT_FEATURE(sb,mask) \
|
|
|
|
EXT4_SB(sb)->s_es->s_feature_ro_compat |= cpu_to_le32(mask)
|
|
|
|
#define EXT4_SET_INCOMPAT_FEATURE(sb,mask) \
|
|
|
|
EXT4_SB(sb)->s_es->s_feature_incompat |= cpu_to_le32(mask)
|
|
|
|
#define EXT4_CLEAR_COMPAT_FEATURE(sb,mask) \
|
|
|
|
EXT4_SB(sb)->s_es->s_feature_compat &= ~cpu_to_le32(mask)
|
|
|
|
#define EXT4_CLEAR_RO_COMPAT_FEATURE(sb,mask) \
|
|
|
|
EXT4_SB(sb)->s_es->s_feature_ro_compat &= ~cpu_to_le32(mask)
|
|
|
|
#define EXT4_CLEAR_INCOMPAT_FEATURE(sb,mask) \
|
|
|
|
EXT4_SB(sb)->s_es->s_feature_incompat &= ~cpu_to_le32(mask)
|
|
|
|
|
|
|
|
#define EXT4_FEATURE_COMPAT_DIR_PREALLOC 0x0001
|
|
|
|
#define EXT4_FEATURE_COMPAT_IMAGIC_INODES 0x0002
|
|
|
|
#define EXT4_FEATURE_COMPAT_HAS_JOURNAL 0x0004
|
|
|
|
#define EXT4_FEATURE_COMPAT_EXT_ATTR 0x0008
|
|
|
|
#define EXT4_FEATURE_COMPAT_RESIZE_INODE 0x0010
|
|
|
|
#define EXT4_FEATURE_COMPAT_DIR_INDEX 0x0020
|
|
|
|
|
|
|
|
#define EXT4_FEATURE_RO_COMPAT_SPARSE_SUPER 0x0001
|
|
|
|
#define EXT4_FEATURE_RO_COMPAT_LARGE_FILE 0x0002
|
|
|
|
#define EXT4_FEATURE_RO_COMPAT_BTREE_DIR 0x0004
|
2008-01-29 11:58:26 +07:00
|
|
|
#define EXT4_FEATURE_RO_COMPAT_HUGE_FILE 0x0008
|
Ext4: Uninitialized Block Groups
In pass1 of e2fsck, every inode table in the fileystem is scanned and checked,
regardless of whether it is in use. This is this the most time consuming part
of the filesystem check. The unintialized block group feature can greatly
reduce e2fsck time by eliminating checking of uninitialized inodes.
With this feature, there is a a high water mark of used inodes for each block
group. Block and inode bitmaps can be uninitialized on disk via a flag in the
group descriptor to avoid reading or scanning them at e2fsck time. A checksum
of each group descriptor is used to ensure that corruption in the group
descriptor's bit flags does not cause incorrect operation.
The feature is enabled through a mkfs option
mke2fs /dev/ -O uninit_groups
A patch adding support for uninitialized block groups to e2fsprogs tools has
been posted to the linux-ext4 mailing list.
The patches have been stress tested with fsstress and fsx. In performance
tests testing e2fsck time, we have seen that e2fsck time on ext3 grows
linearly with the total number of inodes in the filesytem. In ext4 with the
uninitialized block groups feature, the e2fsck time is constant, based
solely on the number of used inodes rather than the total inode count.
Since typical ext4 filesystems only use 1-10% of their inodes, this feature can
greatly reduce e2fsck time for users. With performance improvement of 2-20
times, depending on how full the filesystem is.
The attached graph shows the major improvements in e2fsck times in filesystems
with a large total inode count, but few inodes in use.
In each group descriptor if we have
EXT4_BG_INODE_UNINIT set in bg_flags:
Inode table is not initialized/used in this group. So we can skip
the consistency check during fsck.
EXT4_BG_BLOCK_UNINIT set in bg_flags:
No block in the group is used. So we can skip the block bitmap
verification for this group.
We also add two new fields to group descriptor as a part of
uninitialized group patch.
__le16 bg_itable_unused; /* Unused inodes count */
__le16 bg_checksum; /* crc16(sb_uuid+group+desc) */
bg_itable_unused:
If we have EXT4_BG_INODE_UNINIT not set in bg_flags
then bg_itable_unused will give the offset within
the inode table till the inodes are used. This can be
used by fsck to skip list of inodes that are marked unused.
bg_checksum:
Now that we depend on bg_flags and bg_itable_unused to determine
the block and inode usage, we need to make sure group descriptor
is not corrupt. We add checksum to group descriptor to
detect corruption. If the descriptor is found to be corrupt, we
mark all the blocks and inodes in the group used.
Signed-off-by: Avantika Mathur <mathur@us.ibm.com>
Signed-off-by: Andreas Dilger <adilger@clusterfs.com>
Signed-off-by: Mingming Cao <cmm@us.ibm.com>
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
2007-10-17 05:38:25 +07:00
|
|
|
#define EXT4_FEATURE_RO_COMPAT_GDT_CSUM 0x0010
|
2007-07-18 19:38:01 +07:00
|
|
|
#define EXT4_FEATURE_RO_COMPAT_DIR_NLINK 0x0020
|
2007-07-18 20:15:20 +07:00
|
|
|
#define EXT4_FEATURE_RO_COMPAT_EXTRA_ISIZE 0x0040
|
2006-10-11 15:20:53 +07:00
|
|
|
|
|
|
|
#define EXT4_FEATURE_INCOMPAT_COMPRESSION 0x0001
|
|
|
|
#define EXT4_FEATURE_INCOMPAT_FILETYPE 0x0002
|
|
|
|
#define EXT4_FEATURE_INCOMPAT_RECOVER 0x0004 /* Needs recovery */
|
|
|
|
#define EXT4_FEATURE_INCOMPAT_JOURNAL_DEV 0x0008 /* Journal device */
|
|
|
|
#define EXT4_FEATURE_INCOMPAT_META_BG 0x0010
|
2006-10-11 15:21:03 +07:00
|
|
|
#define EXT4_FEATURE_INCOMPAT_EXTENTS 0x0040 /* extents support */
|
2006-10-11 15:21:10 +07:00
|
|
|
#define EXT4_FEATURE_INCOMPAT_64BIT 0x0080
|
2008-01-29 11:58:26 +07:00
|
|
|
#define EXT4_FEATURE_INCOMPAT_MMP 0x0100
|
2007-10-17 05:38:25 +07:00
|
|
|
#define EXT4_FEATURE_INCOMPAT_FLEX_BG 0x0200
|
2006-10-11 15:20:53 +07:00
|
|
|
|
|
|
|
#define EXT4_FEATURE_COMPAT_SUPP EXT2_FEATURE_COMPAT_EXT_ATTR
|
|
|
|
#define EXT4_FEATURE_INCOMPAT_SUPP (EXT4_FEATURE_INCOMPAT_FILETYPE| \
|
|
|
|
EXT4_FEATURE_INCOMPAT_RECOVER| \
|
2006-10-11 15:21:03 +07:00
|
|
|
EXT4_FEATURE_INCOMPAT_META_BG| \
|
2006-10-11 15:21:10 +07:00
|
|
|
EXT4_FEATURE_INCOMPAT_EXTENTS| \
|
2007-10-17 05:38:25 +07:00
|
|
|
EXT4_FEATURE_INCOMPAT_64BIT| \
|
|
|
|
EXT4_FEATURE_INCOMPAT_FLEX_BG)
|
2006-10-11 15:20:53 +07:00
|
|
|
#define EXT4_FEATURE_RO_COMPAT_SUPP (EXT4_FEATURE_RO_COMPAT_SPARSE_SUPER| \
|
|
|
|
EXT4_FEATURE_RO_COMPAT_LARGE_FILE| \
|
Ext4: Uninitialized Block Groups
In pass1 of e2fsck, every inode table in the fileystem is scanned and checked,
regardless of whether it is in use. This is this the most time consuming part
of the filesystem check. The unintialized block group feature can greatly
reduce e2fsck time by eliminating checking of uninitialized inodes.
With this feature, there is a a high water mark of used inodes for each block
group. Block and inode bitmaps can be uninitialized on disk via a flag in the
group descriptor to avoid reading or scanning them at e2fsck time. A checksum
of each group descriptor is used to ensure that corruption in the group
descriptor's bit flags does not cause incorrect operation.
The feature is enabled through a mkfs option
mke2fs /dev/ -O uninit_groups
A patch adding support for uninitialized block groups to e2fsprogs tools has
been posted to the linux-ext4 mailing list.
The patches have been stress tested with fsstress and fsx. In performance
tests testing e2fsck time, we have seen that e2fsck time on ext3 grows
linearly with the total number of inodes in the filesytem. In ext4 with the
uninitialized block groups feature, the e2fsck time is constant, based
solely on the number of used inodes rather than the total inode count.
Since typical ext4 filesystems only use 1-10% of their inodes, this feature can
greatly reduce e2fsck time for users. With performance improvement of 2-20
times, depending on how full the filesystem is.
The attached graph shows the major improvements in e2fsck times in filesystems
with a large total inode count, but few inodes in use.
In each group descriptor if we have
EXT4_BG_INODE_UNINIT set in bg_flags:
Inode table is not initialized/used in this group. So we can skip
the consistency check during fsck.
EXT4_BG_BLOCK_UNINIT set in bg_flags:
No block in the group is used. So we can skip the block bitmap
verification for this group.
We also add two new fields to group descriptor as a part of
uninitialized group patch.
__le16 bg_itable_unused; /* Unused inodes count */
__le16 bg_checksum; /* crc16(sb_uuid+group+desc) */
bg_itable_unused:
If we have EXT4_BG_INODE_UNINIT not set in bg_flags
then bg_itable_unused will give the offset within
the inode table till the inodes are used. This can be
used by fsck to skip list of inodes that are marked unused.
bg_checksum:
Now that we depend on bg_flags and bg_itable_unused to determine
the block and inode usage, we need to make sure group descriptor
is not corrupt. We add checksum to group descriptor to
detect corruption. If the descriptor is found to be corrupt, we
mark all the blocks and inodes in the group used.
Signed-off-by: Avantika Mathur <mathur@us.ibm.com>
Signed-off-by: Andreas Dilger <adilger@clusterfs.com>
Signed-off-by: Mingming Cao <cmm@us.ibm.com>
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
2007-10-17 05:38:25 +07:00
|
|
|
EXT4_FEATURE_RO_COMPAT_GDT_CSUM| \
|
2007-07-18 19:38:01 +07:00
|
|
|
EXT4_FEATURE_RO_COMPAT_DIR_NLINK | \
|
2007-07-18 20:15:20 +07:00
|
|
|
EXT4_FEATURE_RO_COMPAT_EXTRA_ISIZE | \
|
2008-01-29 11:58:26 +07:00
|
|
|
EXT4_FEATURE_RO_COMPAT_BTREE_DIR |\
|
|
|
|
EXT4_FEATURE_RO_COMPAT_HUGE_FILE)
|
2006-10-11 15:20:50 +07:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Default values for user and/or group using reserved blocks
|
|
|
|
*/
|
2006-10-11 15:20:53 +07:00
|
|
|
#define EXT4_DEF_RESUID 0
|
|
|
|
#define EXT4_DEF_RESGID 0
|
2006-10-11 15:20:50 +07:00
|
|
|
|
2008-10-10 10:53:47 +07:00
|
|
|
#define EXT4_DEF_INODE_READAHEAD_BLKS 32
|
|
|
|
|
2006-10-11 15:20:50 +07:00
|
|
|
/*
|
|
|
|
* Default mount options
|
|
|
|
*/
|
2006-10-11 15:20:53 +07:00
|
|
|
#define EXT4_DEFM_DEBUG 0x0001
|
|
|
|
#define EXT4_DEFM_BSDGROUPS 0x0002
|
|
|
|
#define EXT4_DEFM_XATTR_USER 0x0004
|
|
|
|
#define EXT4_DEFM_ACL 0x0008
|
|
|
|
#define EXT4_DEFM_UID16 0x0010
|
|
|
|
#define EXT4_DEFM_JMODE 0x0060
|
|
|
|
#define EXT4_DEFM_JMODE_DATA 0x0020
|
|
|
|
#define EXT4_DEFM_JMODE_ORDERED 0x0040
|
|
|
|
#define EXT4_DEFM_JMODE_WBACK 0x0060
|
2006-10-11 15:20:50 +07:00
|
|
|
|
2009-01-04 08:27:38 +07:00
|
|
|
/*
|
|
|
|
* Default journal batch times
|
|
|
|
*/
|
|
|
|
#define EXT4_DEF_MIN_BATCH_TIME 0
|
|
|
|
#define EXT4_DEF_MAX_BATCH_TIME 15000 /* 15ms */
|
|
|
|
|
2009-03-12 23:18:34 +07:00
|
|
|
/*
|
|
|
|
* Minimum number of groups in a flexgroup before we separate out
|
|
|
|
* directories into the first block group of a flexgroup
|
|
|
|
*/
|
|
|
|
#define EXT4_FLEX_SIZE_DIR_ALLOC_SCHEME 4
|
|
|
|
|
2006-10-11 15:20:50 +07:00
|
|
|
/*
|
|
|
|
* Structure of a directory entry
|
|
|
|
*/
|
2006-10-11 15:20:53 +07:00
|
|
|
#define EXT4_NAME_LEN 255
|
2006-10-11 15:20:50 +07:00
|
|
|
|
2006-10-11 15:20:53 +07:00
|
|
|
struct ext4_dir_entry {
|
2006-10-11 15:20:50 +07:00
|
|
|
__le32 inode; /* Inode number */
|
|
|
|
__le16 rec_len; /* Directory entry length */
|
|
|
|
__le16 name_len; /* Name length */
|
2006-10-11 15:20:53 +07:00
|
|
|
char name[EXT4_NAME_LEN]; /* File name */
|
2006-10-11 15:20:50 +07:00
|
|
|
};
|
|
|
|
|
|
|
|
/*
|
2006-10-11 15:20:53 +07:00
|
|
|
* The new version of the directory entry. Since EXT4 structures are
|
2006-10-11 15:20:50 +07:00
|
|
|
* stored in intel byte order, and the name_len field could never be
|
|
|
|
* bigger than 255 chars, it's safe to reclaim the extra byte for the
|
|
|
|
* file_type field.
|
|
|
|
*/
|
2006-10-11 15:20:53 +07:00
|
|
|
struct ext4_dir_entry_2 {
|
2006-10-11 15:20:50 +07:00
|
|
|
__le32 inode; /* Inode number */
|
|
|
|
__le16 rec_len; /* Directory entry length */
|
|
|
|
__u8 name_len; /* Name length */
|
|
|
|
__u8 file_type;
|
2006-10-11 15:20:53 +07:00
|
|
|
char name[EXT4_NAME_LEN]; /* File name */
|
2006-10-11 15:20:50 +07:00
|
|
|
};
|
|
|
|
|
|
|
|
/*
|
2006-10-11 15:20:53 +07:00
|
|
|
* Ext4 directory file types. Only the low 3 bits are used. The
|
2006-10-11 15:20:50 +07:00
|
|
|
* other bits are reserved for now.
|
|
|
|
*/
|
2006-10-11 15:20:53 +07:00
|
|
|
#define EXT4_FT_UNKNOWN 0
|
|
|
|
#define EXT4_FT_REG_FILE 1
|
|
|
|
#define EXT4_FT_DIR 2
|
|
|
|
#define EXT4_FT_CHRDEV 3
|
|
|
|
#define EXT4_FT_BLKDEV 4
|
|
|
|
#define EXT4_FT_FIFO 5
|
|
|
|
#define EXT4_FT_SOCK 6
|
|
|
|
#define EXT4_FT_SYMLINK 7
|
2006-10-11 15:20:50 +07:00
|
|
|
|
2006-10-11 15:20:53 +07:00
|
|
|
#define EXT4_FT_MAX 8
|
2006-10-11 15:20:50 +07:00
|
|
|
|
|
|
|
/*
|
2006-10-11 15:20:53 +07:00
|
|
|
* EXT4_DIR_PAD defines the directory entries boundaries
|
2006-10-11 15:20:50 +07:00
|
|
|
*
|
|
|
|
* NOTE: It must be a multiple of 4
|
|
|
|
*/
|
2006-10-11 15:20:53 +07:00
|
|
|
#define EXT4_DIR_PAD 4
|
|
|
|
#define EXT4_DIR_ROUND (EXT4_DIR_PAD - 1)
|
|
|
|
#define EXT4_DIR_REC_LEN(name_len) (((name_len) + 8 + EXT4_DIR_ROUND) & \
|
|
|
|
~EXT4_DIR_ROUND)
|
2008-01-29 11:58:27 +07:00
|
|
|
#define EXT4_MAX_REC_LEN ((1<<16)-1)
|
|
|
|
|
2006-10-11 15:20:50 +07:00
|
|
|
/*
|
|
|
|
* Hash Tree Directory indexing
|
|
|
|
* (c) Daniel Phillips, 2001
|
|
|
|
*/
|
|
|
|
|
2007-10-17 05:38:25 +07:00
|
|
|
#define is_dx(dir) (EXT4_HAS_COMPAT_FEATURE(dir->i_sb, \
|
|
|
|
EXT4_FEATURE_COMPAT_DIR_INDEX) && \
|
2006-10-11 15:20:53 +07:00
|
|
|
(EXT4_I(dir)->i_flags & EXT4_INDEX_FL))
|
|
|
|
#define EXT4_DIR_LINK_MAX(dir) (!is_dx(dir) && (dir)->i_nlink >= EXT4_LINK_MAX)
|
|
|
|
#define EXT4_DIR_LINK_EMPTY(dir) ((dir)->i_nlink == 2 || (dir)->i_nlink == 1)
|
2006-10-11 15:20:50 +07:00
|
|
|
|
|
|
|
/* Legal values for the dx_root hash_version field: */
|
|
|
|
|
|
|
|
#define DX_HASH_LEGACY 0
|
|
|
|
#define DX_HASH_HALF_MD4 1
|
|
|
|
#define DX_HASH_TEA 2
|
2008-10-29 00:21:44 +07:00
|
|
|
#define DX_HASH_LEGACY_UNSIGNED 3
|
|
|
|
#define DX_HASH_HALF_MD4_UNSIGNED 4
|
|
|
|
#define DX_HASH_TEA_UNSIGNED 5
|
2006-10-11 15:20:50 +07:00
|
|
|
|
|
|
|
#ifdef __KERNEL__
|
|
|
|
|
|
|
|
/* hash info structure used by the directory hash */
|
|
|
|
struct dx_hash_info
|
|
|
|
{
|
|
|
|
u32 hash;
|
|
|
|
u32 minor_hash;
|
|
|
|
int hash_version;
|
|
|
|
u32 *seed;
|
|
|
|
};
|
|
|
|
|
2006-10-11 15:20:53 +07:00
|
|
|
#define EXT4_HTREE_EOF 0x7fffffff
|
2006-10-11 15:20:50 +07:00
|
|
|
|
|
|
|
/*
|
2006-10-11 15:20:53 +07:00
|
|
|
* Control parameters used by ext4_htree_next_block
|
2006-10-11 15:20:50 +07:00
|
|
|
*/
|
|
|
|
#define HASH_NB_ALWAYS 1
|
|
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Describe an inode's exact location on disk and in memory
|
|
|
|
*/
|
2006-10-11 15:20:53 +07:00
|
|
|
struct ext4_iloc
|
2006-10-11 15:20:50 +07:00
|
|
|
{
|
|
|
|
struct buffer_head *bh;
|
|
|
|
unsigned long offset;
|
2008-01-29 11:58:27 +07:00
|
|
|
ext4_group_t block_group;
|
2006-10-11 15:20:50 +07:00
|
|
|
};
|
|
|
|
|
2006-10-11 15:20:53 +07:00
|
|
|
static inline struct ext4_inode *ext4_raw_inode(struct ext4_iloc *iloc)
|
2006-10-11 15:20:50 +07:00
|
|
|
{
|
2006-10-11 15:20:53 +07:00
|
|
|
return (struct ext4_inode *) (iloc->bh->b_data + iloc->offset);
|
2006-10-11 15:20:50 +07:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* This structure is stuffed into the struct file's private_data field
|
|
|
|
* for directories. It is where we put information so that we can do
|
|
|
|
* readdir operations in hash tree order.
|
|
|
|
*/
|
|
|
|
struct dir_private_info {
|
|
|
|
struct rb_root root;
|
|
|
|
struct rb_node *curr_node;
|
|
|
|
struct fname *extra_fname;
|
|
|
|
loff_t last_pos;
|
|
|
|
__u32 curr_hash;
|
|
|
|
__u32 curr_minor_hash;
|
|
|
|
__u32 next_hash;
|
|
|
|
};
|
|
|
|
|
|
|
|
/* calculate the first block number of the group */
|
2006-10-11 15:20:53 +07:00
|
|
|
static inline ext4_fsblk_t
|
2008-01-29 11:58:27 +07:00
|
|
|
ext4_group_first_block_no(struct super_block *sb, ext4_group_t group_no)
|
2006-10-11 15:20:50 +07:00
|
|
|
{
|
2006-10-11 15:20:53 +07:00
|
|
|
return group_no * (ext4_fsblk_t)EXT4_BLOCKS_PER_GROUP(sb) +
|
|
|
|
le32_to_cpu(EXT4_SB(sb)->s_es->s_first_data_block);
|
2006-10-11 15:20:50 +07:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Special error return code only used by dx_probe() and its callers.
|
|
|
|
*/
|
|
|
|
#define ERR_BAD_DX_DIR -75000
|
|
|
|
|
2006-10-11 15:21:18 +07:00
|
|
|
void ext4_get_group_no_and_offset(struct super_block *sb, ext4_fsblk_t blocknr,
|
2009-01-06 10:18:16 +07:00
|
|
|
ext4_group_t *blockgrpp, ext4_grpblk_t *offsetp);
|
2006-10-11 15:21:05 +07:00
|
|
|
|
2008-09-23 20:18:24 +07:00
|
|
|
extern struct proc_dir_entry *ext4_proc_root;
|
|
|
|
|
2006-10-11 15:20:50 +07:00
|
|
|
/*
|
|
|
|
* Function prototypes
|
|
|
|
*/
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Ok, these declarations are also in <linux/kernel.h> but none of the
|
2006-10-11 15:20:53 +07:00
|
|
|
* ext4 source programs needs to include it so they are duplicated here.
|
2006-10-11 15:20:50 +07:00
|
|
|
*/
|
2007-05-25 00:04:54 +07:00
|
|
|
# define NORET_TYPE /**/
|
|
|
|
# define ATTRIB_NORET __attribute__((noreturn))
|
|
|
|
# define NORET_AND noreturn,
|
2006-10-11 15:20:50 +07:00
|
|
|
|
2008-11-05 12:14:04 +07:00
|
|
|
/* bitmap.c */
|
|
|
|
extern unsigned int ext4_count_free(struct buffer_head *, unsigned);
|
|
|
|
|
2006-10-11 15:20:50 +07:00
|
|
|
/* balloc.c */
|
2006-10-11 15:21:05 +07:00
|
|
|
extern unsigned int ext4_block_group(struct super_block *sb,
|
|
|
|
ext4_fsblk_t blocknr);
|
|
|
|
extern ext4_grpblk_t ext4_block_group_offset(struct super_block *sb,
|
|
|
|
ext4_fsblk_t blocknr);
|
2008-01-29 11:58:27 +07:00
|
|
|
extern int ext4_bg_has_super(struct super_block *sb, ext4_group_t group);
|
|
|
|
extern unsigned long ext4_bg_num_gdb(struct super_block *sb,
|
|
|
|
ext4_group_t group);
|
2008-07-12 06:27:31 +07:00
|
|
|
extern ext4_fsblk_t ext4_new_meta_blocks(handle_t *handle, struct inode *inode,
|
2006-10-11 15:20:53 +07:00
|
|
|
ext4_fsblk_t goal, unsigned long *count, int *errp);
|
2008-10-09 10:12:24 +07:00
|
|
|
extern int ext4_claim_free_blocks(struct ext4_sb_info *sbi, s64 nblocks);
|
2008-10-28 11:08:12 +07:00
|
|
|
extern int ext4_has_free_blocks(struct ext4_sb_info *sbi, s64 nblocks);
|
2008-09-09 09:25:24 +07:00
|
|
|
extern void ext4_free_blocks(handle_t *handle, struct inode *inode,
|
2008-01-29 12:19:52 +07:00
|
|
|
ext4_fsblk_t block, unsigned long count, int metadata);
|
2009-01-06 09:36:02 +07:00
|
|
|
extern void ext4_add_groupblocks(handle_t *handle, struct super_block *sb,
|
|
|
|
ext4_fsblk_t block, unsigned long count);
|
2008-09-09 09:25:24 +07:00
|
|
|
extern ext4_fsblk_t ext4_count_free_blocks(struct super_block *);
|
|
|
|
extern void ext4_check_blocks_bitmap(struct super_block *);
|
2006-10-11 15:20:53 +07:00
|
|
|
extern struct ext4_group_desc * ext4_get_group_desc(struct super_block * sb,
|
2008-01-29 11:58:27 +07:00
|
|
|
ext4_group_t block_group,
|
2006-10-11 15:20:50 +07:00
|
|
|
struct buffer_head ** bh);
|
2006-10-11 15:20:53 +07:00
|
|
|
extern int ext4_should_retry_alloc(struct super_block *sb, int *retries);
|
2009-05-02 06:44:44 +07:00
|
|
|
struct buffer_head *ext4_read_block_bitmap(struct super_block *sb,
|
|
|
|
ext4_group_t block_group);
|
|
|
|
extern unsigned ext4_init_block_bitmap(struct super_block *sb,
|
|
|
|
struct buffer_head *bh,
|
|
|
|
ext4_group_t group,
|
|
|
|
struct ext4_group_desc *desc);
|
|
|
|
#define ext4_free_blocks_after_init(sb, group, desc) \
|
|
|
|
ext4_init_block_bitmap(sb, NULL, group, desc)
|
2006-10-11 15:20:50 +07:00
|
|
|
|
|
|
|
/* dir.c */
|
2006-10-11 15:20:53 +07:00
|
|
|
extern int ext4_check_dir_entry(const char *, struct inode *,
|
|
|
|
struct ext4_dir_entry_2 *,
|
2008-11-05 12:14:04 +07:00
|
|
|
struct buffer_head *, unsigned int);
|
2006-10-11 15:20:53 +07:00
|
|
|
extern int ext4_htree_store_dirent(struct file *dir_file, __u32 hash,
|
2006-10-11 15:20:50 +07:00
|
|
|
__u32 minor_hash,
|
2006-10-11 15:20:53 +07:00
|
|
|
struct ext4_dir_entry_2 *dirent);
|
|
|
|
extern void ext4_htree_free_dir_info(struct dir_private_info *p);
|
2006-10-11 15:20:50 +07:00
|
|
|
|
|
|
|
/* fsync.c */
|
2008-09-09 09:25:24 +07:00
|
|
|
extern int ext4_sync_file(struct file *, struct dentry *, int);
|
2006-10-11 15:20:50 +07:00
|
|
|
|
|
|
|
/* hash.c */
|
2006-10-11 15:20:53 +07:00
|
|
|
extern int ext4fs_dirhash(const char *name, int len, struct
|
2006-10-11 15:20:50 +07:00
|
|
|
dx_hash_info *hinfo);
|
|
|
|
|
|
|
|
/* ialloc.c */
|
2009-06-13 22:09:42 +07:00
|
|
|
extern struct inode *ext4_new_inode(handle_t *, struct inode *, int,
|
2009-06-13 22:45:35 +07:00
|
|
|
const struct qstr *qstr, __u32 goal);
|
2008-09-09 09:25:24 +07:00
|
|
|
extern void ext4_free_inode(handle_t *, struct inode *);
|
|
|
|
extern struct inode * ext4_orphan_get(struct super_block *, unsigned long);
|
|
|
|
extern unsigned long ext4_count_free_inodes(struct super_block *);
|
|
|
|
extern unsigned long ext4_count_dirs(struct super_block *);
|
|
|
|
extern void ext4_check_inodes_bitmap(struct super_block *);
|
2009-05-02 06:44:44 +07:00
|
|
|
extern unsigned ext4_init_inode_bitmap(struct super_block *sb,
|
|
|
|
struct buffer_head *bh,
|
|
|
|
ext4_group_t group,
|
|
|
|
struct ext4_group_desc *desc);
|
|
|
|
extern void mark_bitmap_end(int start_bit, int end_bit, char *bitmap);
|
2006-10-11 15:20:50 +07:00
|
|
|
|
2008-01-29 12:19:52 +07:00
|
|
|
/* mballoc.c */
|
|
|
|
extern long ext4_mb_stats;
|
|
|
|
extern long ext4_mb_max_to_scan;
|
|
|
|
extern int ext4_mb_init(struct super_block *, int);
|
|
|
|
extern int ext4_mb_release(struct super_block *);
|
|
|
|
extern ext4_fsblk_t ext4_mb_new_blocks(handle_t *,
|
|
|
|
struct ext4_allocation_request *, int *);
|
|
|
|
extern int ext4_mb_reserve_blocks(struct super_block *, int);
|
2008-10-10 20:40:52 +07:00
|
|
|
extern void ext4_discard_preallocations(struct inode *);
|
2008-01-29 12:19:52 +07:00
|
|
|
extern int __init init_ext4_mballoc(void);
|
|
|
|
extern void exit_ext4_mballoc(void);
|
|
|
|
extern void ext4_mb_free_blocks(handle_t *, struct inode *,
|
2009-06-15 14:45:05 +07:00
|
|
|
ext4_fsblk_t, unsigned long, int, unsigned long *);
|
2009-01-06 09:36:19 +07:00
|
|
|
extern int ext4_mb_add_groupinfo(struct super_block *sb,
|
2008-07-12 06:27:31 +07:00
|
|
|
ext4_group_t i, struct ext4_group_desc *desc);
|
2009-01-06 09:36:19 +07:00
|
|
|
extern int ext4_mb_get_buddy_cache_lock(struct super_block *, ext4_group_t);
|
|
|
|
extern void ext4_mb_put_buddy_cache_lock(struct super_block *,
|
|
|
|
ext4_group_t, int);
|
2006-10-11 15:20:50 +07:00
|
|
|
/* inode.c */
|
2006-10-11 15:20:53 +07:00
|
|
|
int ext4_forget(handle_t *handle, int is_metadata, struct inode *inode,
|
|
|
|
struct buffer_head *bh, ext4_fsblk_t blocknr);
|
2008-01-29 11:58:27 +07:00
|
|
|
struct buffer_head *ext4_getblk(handle_t *, struct inode *,
|
|
|
|
ext4_lblk_t, int, int *);
|
|
|
|
struct buffer_head *ext4_bread(handle_t *, struct inode *,
|
|
|
|
ext4_lblk_t, int, int *);
|
2008-10-07 11:46:36 +07:00
|
|
|
int ext4_get_block(struct inode *inode, sector_t iblock,
|
|
|
|
struct buffer_head *bh_result, int create);
|
2006-10-11 15:20:50 +07:00
|
|
|
|
2008-02-07 15:15:37 +07:00
|
|
|
extern struct inode *ext4_iget(struct super_block *, unsigned long);
|
2008-09-09 09:25:24 +07:00
|
|
|
extern int ext4_write_inode(struct inode *, int);
|
|
|
|
extern int ext4_setattr(struct dentry *, struct iattr *);
|
2008-07-12 06:27:31 +07:00
|
|
|
extern int ext4_getattr(struct vfsmount *mnt, struct dentry *dentry,
|
|
|
|
struct kstat *stat);
|
2008-09-09 09:25:24 +07:00
|
|
|
extern void ext4_delete_inode(struct inode *);
|
|
|
|
extern int ext4_sync_inode(handle_t *, struct inode *);
|
2006-10-11 15:20:53 +07:00
|
|
|
extern void ext4_dirty_inode(struct inode *);
|
|
|
|
extern int ext4_change_inode_journal_flag(struct inode *, int);
|
|
|
|
extern int ext4_get_inode_loc(struct inode *, struct ext4_iloc *);
|
2008-07-12 06:27:31 +07:00
|
|
|
extern int ext4_can_truncate(struct inode *inode);
|
2008-09-09 09:25:24 +07:00
|
|
|
extern void ext4_truncate(struct inode *);
|
2009-08-18 09:17:20 +07:00
|
|
|
extern int ext4_truncate_restart_trans(handle_t *, struct inode *, int nblocks);
|
2006-10-11 15:20:53 +07:00
|
|
|
extern void ext4_set_inode_flags(struct inode *);
|
2007-07-18 20:24:20 +07:00
|
|
|
extern void ext4_get_inode_flags(struct ext4_inode_info *);
|
2009-02-26 13:04:07 +07:00
|
|
|
extern int ext4_alloc_da_blocks(struct inode *inode);
|
2006-10-11 15:20:53 +07:00
|
|
|
extern void ext4_set_aops(struct inode *inode);
|
2006-10-11 15:21:03 +07:00
|
|
|
extern int ext4_writepage_trans_blocks(struct inode *);
|
ext4: journal credits calulation cleanup and fix for non-extent writepage
When considering how many journal credits are needed for modifying a
chunk of data, we need to account for the super block, inode block,
quota blocks and xattr block, indirect/index blocks, also, group bitmap
and group descriptor blocks for new allocation (including data and
indirect/index blocks). There are many places in ext4 do the calculation
on their own and often missed one or two meta blocks, and often they
assume single block allocation, and did not considering the multile
chunk of allocation case.
This patch is trying to cleanup current journal credit code, provides
some common helper funtion to calculate the journal credits, to be used
for writepage, writepages, DIO, fallocate, migration, defrag, and for
both nonextent and extent files.
This patch modified the writepage/write_begin credit caculation for
nonextent files, to use the new helper function. It also fixed the
problem that writepage on nonextent files did not consider the case
blocksize <pagesize, thus could possibelly need multiple block
allocation in a single transaction.
Signed-off-by: Mingming Cao <cmm@us.ibm.com>
Reviewed-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2008-08-20 09:16:07 +07:00
|
|
|
extern int ext4_meta_trans_blocks(struct inode *, int nrblocks, int idxblocks);
|
2008-08-20 09:16:03 +07:00
|
|
|
extern int ext4_chunk_trans_blocks(struct inode *, int nrblocks);
|
2008-07-12 06:27:31 +07:00
|
|
|
extern int ext4_block_truncate_page(handle_t *handle,
|
2006-10-11 15:21:03 +07:00
|
|
|
struct address_space *mapping, loff_t from);
|
2009-04-01 05:23:21 +07:00
|
|
|
extern int ext4_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf);
|
2009-01-23 00:13:05 +07:00
|
|
|
extern qsize_t ext4_get_reserved_space(struct inode *inode);
|
2006-10-11 15:20:50 +07:00
|
|
|
|
|
|
|
/* ioctl.c */
|
2008-04-30 09:03:54 +07:00
|
|
|
extern long ext4_ioctl(struct file *, unsigned int, unsigned long);
|
2008-09-09 09:25:24 +07:00
|
|
|
extern long ext4_compat_ioctl(struct file *, unsigned int, unsigned long);
|
2006-10-11 15:20:50 +07:00
|
|
|
|
2008-01-29 11:58:26 +07:00
|
|
|
/* migrate.c */
|
2008-09-13 23:52:26 +07:00
|
|
|
extern int ext4_ext_migrate(struct inode *);
|
2009-02-15 11:01:36 +07:00
|
|
|
|
2006-10-11 15:20:50 +07:00
|
|
|
/* namei.c */
|
2009-02-15 11:01:36 +07:00
|
|
|
extern unsigned int ext4_rec_len_from_disk(__le16 dlen, unsigned blocksize);
|
|
|
|
extern __le16 ext4_rec_len_to_disk(unsigned len, unsigned blocksize);
|
2006-10-11 15:20:53 +07:00
|
|
|
extern int ext4_orphan_add(handle_t *, struct inode *);
|
|
|
|
extern int ext4_orphan_del(handle_t *, struct inode *);
|
|
|
|
extern int ext4_htree_fill_tree(struct file *dir_file, __u32 start_hash,
|
2006-10-11 15:20:50 +07:00
|
|
|
__u32 start_minor_hash, __u32 *next_hash);
|
|
|
|
|
|
|
|
/* resize.c */
|
2006-10-11 15:20:53 +07:00
|
|
|
extern int ext4_group_add(struct super_block *sb,
|
|
|
|
struct ext4_new_group_data *input);
|
|
|
|
extern int ext4_group_extend(struct super_block *sb,
|
|
|
|
struct ext4_super_block *es,
|
|
|
|
ext4_fsblk_t n_blocks_count);
|
2006-10-11 15:20:50 +07:00
|
|
|
|
|
|
|
/* super.c */
|
2008-09-09 09:25:24 +07:00
|
|
|
extern void ext4_error(struct super_block *, const char *, const char *, ...)
|
2006-10-11 15:20:50 +07:00
|
|
|
__attribute__ ((format (printf, 3, 4)));
|
2008-09-09 09:25:24 +07:00
|
|
|
extern void __ext4_std_error(struct super_block *, const char *, int);
|
|
|
|
extern void ext4_abort(struct super_block *, const char *, const char *, ...)
|
2006-10-11 15:20:50 +07:00
|
|
|
__attribute__ ((format (printf, 3, 4)));
|
2008-09-09 09:25:24 +07:00
|
|
|
extern void ext4_warning(struct super_block *, const char *, const char *, ...)
|
2006-10-11 15:20:50 +07:00
|
|
|
__attribute__ ((format (printf, 3, 4)));
|
2009-06-05 04:36:36 +07:00
|
|
|
extern void ext4_msg(struct super_block *, const char *, const char *, ...)
|
|
|
|
__attribute__ ((format (printf, 3, 4)));
|
2009-01-06 10:19:52 +07:00
|
|
|
extern void ext4_grp_locked_error(struct super_block *, ext4_group_t,
|
|
|
|
const char *, const char *, ...)
|
|
|
|
__attribute__ ((format (printf, 4, 5)));
|
2008-09-09 09:25:24 +07:00
|
|
|
extern void ext4_update_dynamic_rev(struct super_block *sb);
|
2008-01-29 11:58:27 +07:00
|
|
|
extern int ext4_update_compat_feature(handle_t *handle, struct super_block *sb,
|
|
|
|
__u32 compat);
|
|
|
|
extern int ext4_update_rocompat_feature(handle_t *handle,
|
|
|
|
struct super_block *sb, __u32 rocompat);
|
|
|
|
extern int ext4_update_incompat_feature(handle_t *handle,
|
|
|
|
struct super_block *sb, __u32 incompat);
|
2006-10-11 15:21:15 +07:00
|
|
|
extern ext4_fsblk_t ext4_block_bitmap(struct super_block *sb,
|
|
|
|
struct ext4_group_desc *bg);
|
|
|
|
extern ext4_fsblk_t ext4_inode_bitmap(struct super_block *sb,
|
|
|
|
struct ext4_group_desc *bg);
|
|
|
|
extern ext4_fsblk_t ext4_inode_table(struct super_block *sb,
|
|
|
|
struct ext4_group_desc *bg);
|
2009-01-06 10:20:24 +07:00
|
|
|
extern __u32 ext4_free_blks_count(struct super_block *sb,
|
|
|
|
struct ext4_group_desc *bg);
|
|
|
|
extern __u32 ext4_free_inodes_count(struct super_block *sb,
|
|
|
|
struct ext4_group_desc *bg);
|
|
|
|
extern __u32 ext4_used_dirs_count(struct super_block *sb,
|
|
|
|
struct ext4_group_desc *bg);
|
|
|
|
extern __u32 ext4_itable_unused_count(struct super_block *sb,
|
|
|
|
struct ext4_group_desc *bg);
|
2006-10-11 15:21:15 +07:00
|
|
|
extern void ext4_block_bitmap_set(struct super_block *sb,
|
|
|
|
struct ext4_group_desc *bg, ext4_fsblk_t blk);
|
|
|
|
extern void ext4_inode_bitmap_set(struct super_block *sb,
|
|
|
|
struct ext4_group_desc *bg, ext4_fsblk_t blk);
|
|
|
|
extern void ext4_inode_table_set(struct super_block *sb,
|
|
|
|
struct ext4_group_desc *bg, ext4_fsblk_t blk);
|
2009-01-06 10:20:24 +07:00
|
|
|
extern void ext4_free_blks_set(struct super_block *sb,
|
|
|
|
struct ext4_group_desc *bg, __u32 count);
|
|
|
|
extern void ext4_free_inodes_set(struct super_block *sb,
|
|
|
|
struct ext4_group_desc *bg, __u32 count);
|
|
|
|
extern void ext4_used_dirs_set(struct super_block *sb,
|
|
|
|
struct ext4_group_desc *bg, __u32 count);
|
|
|
|
extern void ext4_itable_unused_set(struct super_block *sb,
|
|
|
|
struct ext4_group_desc *bg, __u32 count);
|
2009-05-02 06:44:44 +07:00
|
|
|
extern __le16 ext4_group_desc_csum(struct ext4_sb_info *sbi, __u32 group,
|
|
|
|
struct ext4_group_desc *gdp);
|
|
|
|
extern int ext4_group_desc_csum_verify(struct ext4_sb_info *sbi, __u32 group,
|
|
|
|
struct ext4_group_desc *gdp);
|
2006-10-11 15:21:10 +07:00
|
|
|
|
|
|
|
static inline ext4_fsblk_t ext4_blocks_count(struct ext4_super_block *es)
|
|
|
|
{
|
|
|
|
return ((ext4_fsblk_t)le32_to_cpu(es->s_blocks_count_hi) << 32) |
|
2007-10-17 05:38:25 +07:00
|
|
|
le32_to_cpu(es->s_blocks_count_lo);
|
2006-10-11 15:21:10 +07:00
|
|
|
}
|
|
|
|
|
|
|
|
static inline ext4_fsblk_t ext4_r_blocks_count(struct ext4_super_block *es)
|
|
|
|
{
|
|
|
|
return ((ext4_fsblk_t)le32_to_cpu(es->s_r_blocks_count_hi) << 32) |
|
2007-10-17 05:38:25 +07:00
|
|
|
le32_to_cpu(es->s_r_blocks_count_lo);
|
2006-10-11 15:21:10 +07:00
|
|
|
}
|
|
|
|
|
|
|
|
static inline ext4_fsblk_t ext4_free_blocks_count(struct ext4_super_block *es)
|
|
|
|
{
|
|
|
|
return ((ext4_fsblk_t)le32_to_cpu(es->s_free_blocks_count_hi) << 32) |
|
2007-10-17 05:38:25 +07:00
|
|
|
le32_to_cpu(es->s_free_blocks_count_lo);
|
2006-10-11 15:21:10 +07:00
|
|
|
}
|
|
|
|
|
|
|
|
static inline void ext4_blocks_count_set(struct ext4_super_block *es,
|
|
|
|
ext4_fsblk_t blk)
|
|
|
|
{
|
2007-10-17 05:38:25 +07:00
|
|
|
es->s_blocks_count_lo = cpu_to_le32((u32)blk);
|
2006-10-11 15:21:10 +07:00
|
|
|
es->s_blocks_count_hi = cpu_to_le32(blk >> 32);
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline void ext4_free_blocks_count_set(struct ext4_super_block *es,
|
|
|
|
ext4_fsblk_t blk)
|
|
|
|
{
|
2007-10-17 05:38:25 +07:00
|
|
|
es->s_free_blocks_count_lo = cpu_to_le32((u32)blk);
|
2006-10-11 15:21:10 +07:00
|
|
|
es->s_free_blocks_count_hi = cpu_to_le32(blk >> 32);
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline void ext4_r_blocks_count_set(struct ext4_super_block *es,
|
|
|
|
ext4_fsblk_t blk)
|
|
|
|
{
|
2007-10-17 05:38:25 +07:00
|
|
|
es->s_r_blocks_count_lo = cpu_to_le32((u32)blk);
|
2006-10-11 15:21:10 +07:00
|
|
|
es->s_r_blocks_count_hi = cpu_to_le32(blk >> 32);
|
|
|
|
}
|
|
|
|
|
2008-01-29 11:58:27 +07:00
|
|
|
static inline loff_t ext4_isize(struct ext4_inode *raw_inode)
|
|
|
|
{
|
2009-01-18 06:41:37 +07:00
|
|
|
if (S_ISREG(le16_to_cpu(raw_inode->i_mode)))
|
|
|
|
return ((loff_t)le32_to_cpu(raw_inode->i_size_high) << 32) |
|
|
|
|
le32_to_cpu(raw_inode->i_size_lo);
|
|
|
|
else
|
|
|
|
return (loff_t) le32_to_cpu(raw_inode->i_size_lo);
|
2008-01-29 11:58:27 +07:00
|
|
|
}
|
2006-10-11 15:21:10 +07:00
|
|
|
|
2008-01-29 11:58:27 +07:00
|
|
|
static inline void ext4_isize_set(struct ext4_inode *raw_inode, loff_t i_size)
|
|
|
|
{
|
|
|
|
raw_inode->i_size_lo = cpu_to_le32(i_size);
|
|
|
|
raw_inode->i_size_high = cpu_to_le32(i_size >> 32);
|
|
|
|
}
|
2006-10-11 15:20:50 +07:00
|
|
|
|
2008-01-29 12:19:52 +07:00
|
|
|
static inline
|
|
|
|
struct ext4_group_info *ext4_get_group_info(struct super_block *sb,
|
2008-09-09 09:25:24 +07:00
|
|
|
ext4_group_t group)
|
2008-01-29 12:19:52 +07:00
|
|
|
{
|
|
|
|
struct ext4_group_info ***grp_info;
|
|
|
|
long indexv, indexh;
|
|
|
|
grp_info = EXT4_SB(sb)->s_group_info;
|
|
|
|
indexv = group >> (EXT4_DESC_PER_BLOCK_BITS(sb));
|
|
|
|
indexh = group & ((EXT4_DESC_PER_BLOCK(sb)) - 1);
|
|
|
|
return grp_info[indexv][indexh];
|
|
|
|
}
|
|
|
|
|
2009-05-01 19:50:38 +07:00
|
|
|
/*
|
|
|
|
* Reading s_groups_count requires using smp_rmb() afterwards. See
|
|
|
|
* the locking protocol documented in the comments of ext4_group_add()
|
|
|
|
* in resize.c
|
|
|
|
*/
|
|
|
|
static inline ext4_group_t ext4_get_groups_count(struct super_block *sb)
|
|
|
|
{
|
|
|
|
ext4_group_t ngroups = EXT4_SB(sb)->s_groups_count;
|
|
|
|
|
|
|
|
smp_rmb();
|
|
|
|
return ngroups;
|
|
|
|
}
|
2008-01-29 12:19:52 +07:00
|
|
|
|
2008-07-12 06:27:31 +07:00
|
|
|
static inline ext4_group_t ext4_flex_group(struct ext4_sb_info *sbi,
|
|
|
|
ext4_group_t block_group)
|
|
|
|
{
|
|
|
|
return block_group >> sbi->s_log_groups_per_flex;
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline unsigned int ext4_flex_bg_size(struct ext4_sb_info *sbi)
|
|
|
|
{
|
|
|
|
return 1 << sbi->s_log_groups_per_flex;
|
|
|
|
}
|
|
|
|
|
2006-10-11 15:20:53 +07:00
|
|
|
#define ext4_std_error(sb, errno) \
|
2006-10-11 15:20:50 +07:00
|
|
|
do { \
|
|
|
|
if ((errno)) \
|
2008-07-14 08:03:29 +07:00
|
|
|
__ext4_std_error((sb), __func__, (errno)); \
|
2006-10-11 15:20:50 +07:00
|
|
|
} while (0)
|
|
|
|
|
2008-10-09 21:56:23 +07:00
|
|
|
#ifdef CONFIG_SMP
|
2009-01-07 05:41:04 +07:00
|
|
|
/* Each CPU can accumulate percpu_counter_batch blocks in their local
|
2008-10-09 21:56:23 +07:00
|
|
|
* counters. So we need to make sure we have free blocks more
|
2009-01-07 05:41:04 +07:00
|
|
|
* than percpu_counter_batch * nr_cpu_ids. Also add a window of 4 times.
|
2008-10-09 21:56:23 +07:00
|
|
|
*/
|
2009-01-07 05:41:04 +07:00
|
|
|
#define EXT4_FREEBLOCKS_WATERMARK (4 * (percpu_counter_batch * nr_cpu_ids))
|
2008-10-09 21:56:23 +07:00
|
|
|
#else
|
|
|
|
#define EXT4_FREEBLOCKS_WATERMARK 0
|
|
|
|
#endif
|
|
|
|
|
2008-09-14 00:06:18 +07:00
|
|
|
static inline void ext4_update_i_disksize(struct inode *inode, loff_t newsize)
|
|
|
|
{
|
|
|
|
/*
|
|
|
|
* XXX: replace with spinlock if seen contended -bzzz
|
|
|
|
*/
|
|
|
|
down_write(&EXT4_I(inode)->i_data_sem);
|
|
|
|
if (newsize > EXT4_I(inode)->i_disksize)
|
|
|
|
EXT4_I(inode)->i_disksize = newsize;
|
|
|
|
up_write(&EXT4_I(inode)->i_data_sem);
|
|
|
|
return ;
|
|
|
|
}
|
2008-10-09 21:56:23 +07:00
|
|
|
|
2009-01-06 10:19:52 +07:00
|
|
|
struct ext4_group_info {
|
|
|
|
unsigned long bb_state;
|
|
|
|
struct rb_root bb_free_root;
|
2009-08-26 09:36:45 +07:00
|
|
|
ext4_grpblk_t bb_first_free; /* first free block */
|
|
|
|
ext4_grpblk_t bb_free; /* total free blocks */
|
|
|
|
ext4_grpblk_t bb_fragments; /* nr of freespace fragments */
|
2009-01-06 10:19:52 +07:00
|
|
|
struct list_head bb_prealloc_list;
|
|
|
|
#ifdef DOUBLE_CHECK
|
|
|
|
void *bb_bitmap;
|
|
|
|
#endif
|
|
|
|
struct rw_semaphore alloc_sem;
|
2009-08-26 09:36:45 +07:00
|
|
|
ext4_grpblk_t bb_counters[]; /* Nr of free power-of-two-block
|
|
|
|
* regions, index is order.
|
|
|
|
* bb_counters[3] = 5 means
|
|
|
|
* 5 free 8-block regions. */
|
2009-01-06 10:19:52 +07:00
|
|
|
};
|
|
|
|
|
|
|
|
#define EXT4_GROUP_INFO_NEED_INIT_BIT 0
|
|
|
|
|
|
|
|
#define EXT4_MB_GRP_NEED_INIT(grp) \
|
|
|
|
(test_bit(EXT4_GROUP_INFO_NEED_INIT_BIT, &((grp)->bb_state)))
|
|
|
|
|
ext4: Avoid group preallocation for closed files
Currently the group preallocation code tries to find a large (512)
free block from which to do per-cpu group allocation for small files.
The problem with this scheme is that it leaves the filesystem horribly
fragmented. In the worst case, if the filesystem is unmounted and
remounted (after a system shutdown, for example) we forget the fact
that wee were using a particular (now-partially filled) 512 block
extent. So the next time we try to allocate space for a small file,
we will find *another* completely free 512 block chunk to allocate
small files. Given that there are 32,768 blocks in a block group,
after 64 iterations of "mount, write one 4k file in a directory,
unmount", the block group will have 64 files, each separated by 511
blocks, and the block group will no longer have any free 512
completely free chunks of blocks for group preallocation space.
So if we try to allocate blocks for a file that has been closed, such
that we know the final size of the file, and the filesystem is not
busy, avoid using group preallocation.
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2009-09-19 00:34:02 +07:00
|
|
|
#define EXT4_MAX_CONTENTION 8
|
|
|
|
#define EXT4_CONTENTION_THRESHOLD 2
|
|
|
|
|
2009-05-03 07:35:09 +07:00
|
|
|
static inline spinlock_t *ext4_group_lock_ptr(struct super_block *sb,
|
|
|
|
ext4_group_t group)
|
2009-01-06 10:19:52 +07:00
|
|
|
{
|
2009-05-03 07:35:09 +07:00
|
|
|
return bgl_lock_ptr(EXT4_SB(sb)->s_blockgroup_lock, group);
|
|
|
|
}
|
2009-01-06 10:19:52 +07:00
|
|
|
|
ext4: Avoid group preallocation for closed files
Currently the group preallocation code tries to find a large (512)
free block from which to do per-cpu group allocation for small files.
The problem with this scheme is that it leaves the filesystem horribly
fragmented. In the worst case, if the filesystem is unmounted and
remounted (after a system shutdown, for example) we forget the fact
that wee were using a particular (now-partially filled) 512 block
extent. So the next time we try to allocate space for a small file,
we will find *another* completely free 512 block chunk to allocate
small files. Given that there are 32,768 blocks in a block group,
after 64 iterations of "mount, write one 4k file in a directory,
unmount", the block group will have 64 files, each separated by 511
blocks, and the block group will no longer have any free 512
completely free chunks of blocks for group preallocation space.
So if we try to allocate blocks for a file that has been closed, such
that we know the final size of the file, and the filesystem is not
busy, avoid using group preallocation.
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2009-09-19 00:34:02 +07:00
|
|
|
/*
|
|
|
|
* Returns true if the filesystem is busy enough that attempts to
|
|
|
|
* access the block group locks has run into contention.
|
|
|
|
*/
|
|
|
|
static inline int ext4_fs_is_busy(struct ext4_sb_info *sbi)
|
|
|
|
{
|
|
|
|
return (atomic_read(&sbi->s_lock_busy) > EXT4_CONTENTION_THRESHOLD);
|
|
|
|
}
|
|
|
|
|
2009-05-03 07:35:09 +07:00
|
|
|
static inline void ext4_lock_group(struct super_block *sb, ext4_group_t group)
|
|
|
|
{
|
ext4: Avoid group preallocation for closed files
Currently the group preallocation code tries to find a large (512)
free block from which to do per-cpu group allocation for small files.
The problem with this scheme is that it leaves the filesystem horribly
fragmented. In the worst case, if the filesystem is unmounted and
remounted (after a system shutdown, for example) we forget the fact
that wee were using a particular (now-partially filled) 512 block
extent. So the next time we try to allocate space for a small file,
we will find *another* completely free 512 block chunk to allocate
small files. Given that there are 32,768 blocks in a block group,
after 64 iterations of "mount, write one 4k file in a directory,
unmount", the block group will have 64 files, each separated by 511
blocks, and the block group will no longer have any free 512
completely free chunks of blocks for group preallocation space.
So if we try to allocate blocks for a file that has been closed, such
that we know the final size of the file, and the filesystem is not
busy, avoid using group preallocation.
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2009-09-19 00:34:02 +07:00
|
|
|
spinlock_t *lock = ext4_group_lock_ptr(sb, group);
|
|
|
|
if (spin_trylock(lock))
|
|
|
|
/*
|
|
|
|
* We're able to grab the lock right away, so drop the
|
|
|
|
* lock contention counter.
|
|
|
|
*/
|
|
|
|
atomic_add_unless(&EXT4_SB(sb)->s_lock_busy, -1, 0);
|
|
|
|
else {
|
|
|
|
/*
|
|
|
|
* The lock is busy, so bump the contention counter,
|
|
|
|
* and then wait on the spin lock.
|
|
|
|
*/
|
|
|
|
atomic_add_unless(&EXT4_SB(sb)->s_lock_busy, 1,
|
|
|
|
EXT4_MAX_CONTENTION);
|
|
|
|
spin_lock(lock);
|
|
|
|
}
|
2009-01-06 10:19:52 +07:00
|
|
|
}
|
|
|
|
|
|
|
|
static inline void ext4_unlock_group(struct super_block *sb,
|
|
|
|
ext4_group_t group)
|
|
|
|
{
|
2009-05-03 07:35:09 +07:00
|
|
|
spin_unlock(ext4_group_lock_ptr(sb, group));
|
2009-01-06 10:19:52 +07:00
|
|
|
}
|
|
|
|
|
2006-10-11 15:20:50 +07:00
|
|
|
/*
|
|
|
|
* Inodes and files operations
|
|
|
|
*/
|
|
|
|
|
|
|
|
/* dir.c */
|
2006-10-11 15:20:53 +07:00
|
|
|
extern const struct file_operations ext4_dir_operations;
|
2006-10-11 15:20:50 +07:00
|
|
|
|
|
|
|
/* file.c */
|
2007-02-12 15:55:40 +07:00
|
|
|
extern const struct inode_operations ext4_file_inode_operations;
|
2006-10-11 15:20:53 +07:00
|
|
|
extern const struct file_operations ext4_file_operations;
|
2006-10-11 15:20:50 +07:00
|
|
|
|
|
|
|
/* namei.c */
|
2007-02-12 15:55:40 +07:00
|
|
|
extern const struct inode_operations ext4_dir_inode_operations;
|
|
|
|
extern const struct inode_operations ext4_special_inode_operations;
|
2009-05-02 00:49:15 +07:00
|
|
|
extern struct dentry *ext4_get_parent(struct dentry *child);
|
2006-10-11 15:20:50 +07:00
|
|
|
|
|
|
|
/* symlink.c */
|
2007-02-12 15:55:40 +07:00
|
|
|
extern const struct inode_operations ext4_symlink_inode_operations;
|
|
|
|
extern const struct inode_operations ext4_fast_symlink_inode_operations;
|
2006-10-11 15:20:50 +07:00
|
|
|
|
2009-05-18 02:38:01 +07:00
|
|
|
/* block_validity */
|
|
|
|
extern void ext4_release_system_zone(struct super_block *sb);
|
|
|
|
extern int ext4_setup_system_zone(struct super_block *sb);
|
|
|
|
extern int __init init_ext4_system_zone(void);
|
|
|
|
extern void exit_ext4_system_zone(void);
|
|
|
|
extern int ext4_data_block_valid(struct ext4_sb_info *sbi,
|
|
|
|
ext4_fsblk_t start_blk,
|
|
|
|
unsigned int count);
|
|
|
|
|
2006-10-11 15:21:03 +07:00
|
|
|
/* extents.c */
|
|
|
|
extern int ext4_ext_tree_init(handle_t *handle, struct inode *);
|
|
|
|
extern int ext4_ext_writepage_trans_blocks(struct inode *, int);
|
ext4: journal credits calulation cleanup and fix for non-extent writepage
When considering how many journal credits are needed for modifying a
chunk of data, we need to account for the super block, inode block,
quota blocks and xattr block, indirect/index blocks, also, group bitmap
and group descriptor blocks for new allocation (including data and
indirect/index blocks). There are many places in ext4 do the calculation
on their own and often missed one or two meta blocks, and often they
assume single block allocation, and did not considering the multile
chunk of allocation case.
This patch is trying to cleanup current journal credit code, provides
some common helper funtion to calculate the journal credits, to be used
for writepage, writepages, DIO, fallocate, migration, defrag, and for
both nonextent and extent files.
This patch modified the writepage/write_begin credit caculation for
nonextent files, to use the new helper function. It also fixed the
problem that writepage on nonextent files did not consider the case
blocksize <pagesize, thus could possibelly need multiple block
allocation in a single transaction.
Signed-off-by: Mingming Cao <cmm@us.ibm.com>
Reviewed-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2008-08-20 09:16:07 +07:00
|
|
|
extern int ext4_ext_index_trans_blocks(struct inode *inode, int nrblocks,
|
|
|
|
int chunk);
|
2006-10-11 15:21:03 +07:00
|
|
|
extern int ext4_ext_get_blocks(handle_t *handle, struct inode *inode,
|
2008-11-05 12:14:04 +07:00
|
|
|
ext4_lblk_t iblock, unsigned int max_blocks,
|
2009-05-14 11:58:52 +07:00
|
|
|
struct buffer_head *bh_result, int flags);
|
2008-07-12 06:27:31 +07:00
|
|
|
extern void ext4_ext_truncate(struct inode *);
|
2006-10-11 15:21:03 +07:00
|
|
|
extern void ext4_ext_init(struct super_block *);
|
|
|
|
extern void ext4_ext_release(struct super_block *);
|
2007-07-18 08:42:41 +07:00
|
|
|
extern long ext4_fallocate(struct inode *inode, int mode, loff_t offset,
|
|
|
|
loff_t len);
|
2009-09-29 02:49:08 +07:00
|
|
|
extern int ext4_convert_unwritten_extents(struct inode *inode, loff_t offset,
|
|
|
|
loff_t len);
|
2009-05-14 11:57:44 +07:00
|
|
|
extern int ext4_get_blocks(handle_t *handle, struct inode *inode,
|
|
|
|
sector_t block, unsigned int max_blocks,
|
2009-05-14 11:58:52 +07:00
|
|
|
struct buffer_head *bh, int flags);
|
2008-11-23 03:04:59 +07:00
|
|
|
extern int ext4_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
|
|
|
|
__u64 start, __u64 len);
|
2009-06-18 06:24:03 +07:00
|
|
|
/* move_extent.c */
|
|
|
|
extern int ext4_move_extents(struct file *o_filp, struct file *d_filp,
|
|
|
|
__u64 start_orig, __u64 start_donor,
|
|
|
|
__u64 len, __u64 *moved_len);
|
|
|
|
|
2008-11-23 03:04:59 +07:00
|
|
|
|
2009-01-06 09:49:55 +07:00
|
|
|
/*
|
|
|
|
* Add new method to test wether block and inode bitmaps are properly
|
|
|
|
* initialized. With uninit_bg reading the block from disk is not enough
|
|
|
|
* to mark the bitmap uptodate. We need to also zero-out the bitmap
|
|
|
|
*/
|
|
|
|
#define BH_BITMAP_UPTODATE BH_JBDPrivateStart
|
|
|
|
|
|
|
|
static inline int bitmap_uptodate(struct buffer_head *bh)
|
|
|
|
{
|
|
|
|
return (buffer_uptodate(bh) &&
|
|
|
|
test_bit(BH_BITMAP_UPTODATE, &(bh)->b_state));
|
|
|
|
}
|
|
|
|
static inline void set_bitmap_uptodate(struct buffer_head *bh)
|
|
|
|
{
|
|
|
|
set_bit(BH_BITMAP_UPTODATE, &(bh)->b_state);
|
|
|
|
}
|
|
|
|
|
2006-10-11 15:20:50 +07:00
|
|
|
#endif /* __KERNEL__ */
|
|
|
|
|
2008-04-30 05:13:32 +07:00
|
|
|
#endif /* _EXT4_H */
|