Ext2: Difference between revisions
[unchecked revision] | [unchecked revision] |
Content deleted Content added
→Base Superblock Fields: clarify the superblock block number field is also the starting block number |
|||
(29 intermediate revisions by 9 users not shown) | |||
Line 2:
The '''Second Extended Filesystem''' ('''ext2fs''') is a rewrite of the original ''Extended Filesystem'' and as such, is also based around the concept of "inodes." Ext2 served as the de facto filesystem of Linux for nearly a decade from the early 1990s to the early 2000s when it was superseded by the journaling file systems [[Ext3|ext3]] and [[ReiserFS]]. It has native support for UNIX ownership / access rights, symbolic- and hard-links, and other properties that are common among UNIX-like operating systems. Organizationally, it divides disk space up into groups called "block groups." Having these groups results in distribution of data across the disk which helps to minimize head movement as well as the impact of fragmentation. Further, some (if not all) groups are required to contain backups of important data that can be used to rebuild the file system in the event of disaster.
''Note: Most of the information here is based off of work done by Dave Poirier on the ext2-doc project (see the [[#Links|links section]]) which is graciously released under the [http://www.fsf.org/licenses/fdl.html GNU Free Documentation License]. Be sure to buy him a beer the next time you see him.''
== Basic Concepts ==
'''Important Note: All values are little-endian unless otherwise specified'''
=== What is a Block? ===
Line 11:
=== What is a Block Group? ===
Blocks, along with inodes, are
Each block group reserves a few of its blocks for special purposes such as:
Line 20:
=== What is an Inode? ===
An inode is a structure on the disk that represents a file, directory, symbolic link, etc. Inodes
== Superblock ==
Line 55:
| 16 || 19 || 4 || Total number of unallocated inodes
|-
| 20 || 23 || 4 || Block number of the block containing the superblock (also the starting block number, NOT always zero.)
|-
| 24 || 27 || 4 || ''log''<sub>2</sub> (block size) - 10. (In other words, the number to shift 1,024 to the left by to obtain the block size)
Line 132:
=== Extended Superblock Fields ===
These fields are only present if Major version (specified in the base superblock fields), is greater than or equal to 1.
{| {{Wikitable}}
! Starting
Line 209:
| 0x0002 || Directory entries contain a type field
|-
| 0x0004 || File system needs
|-
| 0x0008 || File system uses a journal device
Line 229:
== Block Group Descriptor Table ==
The Block Group Descriptor Table contains
=== Locating the Block Group Descriptor Table ===
The table is located in the block immediately following the Superblock.
=== Block Group Descriptor ===
A Block Group Descriptor contains information regarding where important data structures for that block group are located.
{| {{Wikitable}}
! Starting
Line 262 ⟶ 263:
Like blocks, each inode has a numerical address. It is extremely important to note that unlike block addresses, '''inode addresses start at 1'''.
With Ext2 versions prior to Major version 1, inodes 1 to 10 are reserved and should be in an allocated state. Starting with version 1, the first non-reserved inode is indicated via a field in the Superblock. Of the reserved inodes, number 2
Inodes have a fixed size of either 128 for version 0 Ext2 file systems, or as dictated by the field in the Superblock for version 1 file systems. All inodes reside in inode tables that belong to block groups. Therefore, looking up an inode is simply a matter of determining which block group it belongs to and indexing that block group's inode table.
=== Determining which Block Group contains an Inode ===
From an inode address (remember that they start at 1), we can determine which group the inode is in, by using the formula:
block group = (inode – 1) / INODES_PER_GROUP
where INODES_PER_GROUP is a field in the Superblock
=== Finding an inode inside of a Block Group ===
Once we know which group an inode resides in, we can look up the actual inode by first retrieving that block group's inode table's starting address (see [[#Block_Group_Descriptor|Block Group Descriptor]] above). The index of our inode in this block group's inode table can be determined by using the formula:
index = (inode – 1) % INODES_PER_GROUP
where % denotes the [http://en.wikipedia.org/wiki/Modulo_operation Modulo operation] and INODES_PER_GROUP is a field in the Superblock (the same field which was used to determine which block group the inode belongs to).
Next, we have to determine which block contains our inode. This is achieved from:
containing block = (index * INODE_SIZE) / BLOCK_SIZE
where INODE_SIZE is either fixed at 128 if VERSION < 1 or defined by a field in the Superblock if VERSION >= 1.0, and BLOCK_SIZE is defined by a field in the Superblock.
Finally, mask and shift as necessary to extract only the inode data from the containing block.
=== Reading the contents of an inode ===
Each inode contains 12 direct pointers, one singly indirect pointer, one doubly indirect block pointer, and one triply indirect pointer. The direct space "overflows" into the singly indirect space, which overflows into the doubly indirect space, which overflows into the triply indirect space.
'''Direct Block Pointers''': There are 12 direct block pointers. If valid, the value is non-zero. Each pointer is the block address of a block containing data for this inode.
'''Singly Indirect Block Pointer''': If a file needs more than 12 blocks, a separate block is allocated to store the block addresses of the remaining data blocks needed to store its contents. This separate block is called an indirect block because it adds an extra step (a level of indirection) between an inode and its data. The block addresses stored in the block are all 32-bit, and the capacity of stored addresses in this block is a function of the block size. The address of this indirect block is stored in the inode in the "Singly Indirect Block Pointer" field.
'''Doubly Indirect Block Pointer''': If a file has more blocks than can fit in the 12 direct pointers and the indirect block, a double indirect block is used. A double indirect block is an extension of the indirect block described above only now we have two intermediate blocks between the inode and data blocks. The inode structure has a "Doubly Indirect Block Pointer" field that points to this block if necessary.
'''Triply Indirect Block Pointer''': Lastly, if a file needs still more space, it can use a triple indirect block. Again, this is an extension of the double indirect block. So, a triple indirect block contains addresses of double indirect blocks, which contain addresses of single indirect blocks, which contain address of data blocks. The inode structure has a "Triply Indirect Block Pointer" field that points to this block if present.
[http://en.wikipedia.org/wiki/File:Ext2-inode.gif This image from Wikipedia] illustrates what is described above pretty well.
=== Inode Data Structure ===
{| {{Wikitable}}
! Starting
Byte
! Ending
Byte
! Size
in Bytes
! Field Description
|-
| 0 || 1 || 2 || Type and Permissions ([[#Inode_Type_and_Permissions|see below]])
|-
|
|-
|
|-
| 8 || 11 || 4 || Last Access Time (in [http://en.wikipedia.org/wiki/Unix_time POSIX time])
|-
| 12 || 15 || 4 || Creation Time (in [http://en.wikipedia.org/wiki/Unix_time POSIX time])
|-
| 16 || 19 || 4 || Last Modification time (in [http://en.wikipedia.org/wiki/Unix_time POSIX time])
|-
| 20 || 23 || 4 || Deletion time (in [http://en.wikipedia.org/wiki/Unix_time POSIX time])
|-
| 24 || 25 || 2 || Group ID
|-
| 26 || 27 || 2 || Count of hard links (directory entries) to this inode. When this reaches 0, the data blocks are marked as unallocated.
|-
| 28 || 31 || 4 || Count of disk sectors (not Ext2 blocks) in use by this inode, not counting the actual inode structure nor directory entries linking to the inode.
|-
| 32 || 35 || 4 || Flags ([[#Inode_Flags|see below]])
|-
| 36 || 39 || 4 || [[#OS_Specific_Value_1|Operating System Specific value #1]]
|-
| 40 || 43 || 4 || Direct Block Pointer 0
|-
| 44 || 47 || 4 || Direct Block Pointer 1
|-
| 48 || 51 || 4 || Direct Block Pointer 2
|-
| 52 || 55 || 4 || Direct Block Pointer 3
|-
| 56 || 59 || 4 || Direct Block Pointer 4
|-
| 60 || 63 || 4 || Direct Block Pointer 5
|-
| 64 || 67 || 4 || Direct Block Pointer 6
|-
|
|-
| 72 || 75 || 4 || Direct Block Pointer 8
|-
| 76 || 79 || 4 || Direct Block Pointer 9
|-
| 80 || 83 || 4 || Direct Block Pointer 10
|-
| 84 || 87 || 4 || Direct Block Pointer 11
|-
| 88 || 91 || 4 || Singly Indirect Block Pointer (Points to a block that is a list of block pointers to data)
|-
| 92 || 95 || 4 || Doubly Indirect Block Pointer (Points to a block that is a list of block pointers to Singly Indirect Blocks)
|-
| 96 || 99 || 4 || Triply Indirect Block Pointer (Points to a block that is a list of block pointers to Doubly Indirect Blocks)
|-
| 100 || 103 || 4 || Generation number (Primarily used for NFS)
|-
| 104 || 107 || 4 || In Ext2 version 0, this field is reserved. In version >= 1, Extended attribute block (File ACL).
|-
| 108 || 111 || 4 || In Ext2 version 0, this field is reserved. In version >= 1, Upper 32 bits of file size (if feature bit set) if it's a file, Directory ACL if it's a directory
|-
| 112 || 115 || 4 || Block address of fragment
|-
| 116 || 127 || 12 || [[#OS_Specific_Value_2|Operating System Specific Value #2]]
|}
==== Inode Type and Permissions ====
{| {{Wikitable}}
|+The type indicator occupies the top hex digit (bits 15 to 12) of this 16-bit field
! Type value
in hex
! Type Description
|-
| 0x1000 || FIFO
|-
| 0x2000 || Character device
|-
| 0x4000 || Directory
|-
| 0x6000 || Block device
|-
| 0x8000 || Regular file
|-
| 0xA000 || Symbolic link
|-
| 0xC000 || Unix socket
|}
{| {{Wikitable}}
|+Permissions occupy the bottom 12 bits of this 16-bit field
! Permission
value in hex
! Permission
value in octal
! Permission Description
|-
| 0x001 || 00001 || [http://en.wikipedia.org/wiki/Filesystem_permissions#Traditional_Unix_permissions Other—execute permission]
|-
| 0x002 || 00002 || [http://en.wikipedia.org/wiki/Filesystem_permissions#Traditional_Unix_permissions Other—write permission]
|-
| 0x004 || 00004 || [http://en.wikipedia.org/wiki/Filesystem_permissions#Traditional_Unix_permissions Other—read permission]
|-
| 0x008 || 00010 || [http://en.wikipedia.org/wiki/Filesystem_permissions#Traditional_Unix_permissions Group—execute permission]
|-
| 0x010 || 00020 || [http://en.wikipedia.org/wiki/Filesystem_permissions#Traditional_Unix_permissions Group—write permission]
|-
| 0x020 || 00040 || [http://en.wikipedia.org/wiki/Filesystem_permissions#Traditional_Unix_permissions Group—read permission]
|-
| 0x040 || 00100 || [http://en.wikipedia.org/wiki/Filesystem_permissions#Traditional_Unix_permissions User—execute permission]
|-
| 0x080 || 00200 || [http://en.wikipedia.org/wiki/Filesystem_permissions#Traditional_Unix_permissions User—write permission]
|-
| 0x100 || 00400 || [http://en.wikipedia.org/wiki/Filesystem_permissions#Traditional_Unix_permissions User—read permission]
|-
| 0x200 || 01000 || [http://en.wikipedia.org/wiki/Sticky_bit Sticky Bit]
|-
| 0x400 || 02000 || Set group ID
|-
|
|}
==== Inode
{| {{Wikitable}}
|-
Line 440 ⟶ 452:
| 0x00000040 || File is not included in 'dump' command
|-
| 0x00000080 ||
|-
| ... || (Reserved)
|-
| 0x00010000 || Hash indexed directory
|-
| 0x00020000 || AFS directory
|-
| 0x00040000 || Journal file data
|}
==== OS Specific Value 1 ====
{| {{Wikitable}}
! Operating
System
! How they use this field
|-
| Linux || (reserved)
|-
| HURD || "translator"?
|-
| MASIX || (reserved)
|}
==== OS Specific Value 2 ====
{| {{Wikitable}}
! Operating
System
! How they use this field
|-
| Linux ||
{| {{Wikitable}}
! Starting
Byte
! Ending
Byte
! Size
in Bytes
! Field Description
|-
| 116 || 116 || 1 || Fragment number
|-
| 117 || 117 || 1 || Fragment size
|-
| 118 || 119 || 2 || (reserved)
|-
| 120 || 121 || 2 || High 16 bits of 32-bit User ID
|-
| 122 || 123 || 2 || High 16 bits of 32-bit Group ID
|-
| 124 || 127 || 4 || (reserved)
|}
|-
| HURD ||
{| {{Wikitable}}
! Starting
Byte
! Ending
Byte
! Size
in Bytes
! Field Description
|-
| 116 || 116 || 1 || Fragment number
|-
| 117 || 117 || 1 || Fragment size
|-
| 118 || 119 || 2 || High 16 bits of 32-bit "Type and Permissions" field
|-
| 120 || 121 || 2 || High 16 bits of 32-bit User ID
|-
| 122 || 123 || 2 || High 16 bits of 32-bit Group ID
|-
| 124 || 127 || 4 || User ID of author (if == 0xFFFFFFFF, the normal User ID will be used)
|}
|-
| MASIX ||
{| {{Wikitable}}
! Starting
Byte
! Ending
Byte
! Size
in Bytes
! Field Description
|-
| 116 || 116 || 1 || Fragment number
|-
| 117 || 117 || 1 || Fragment size
|-
| 118 || 127 || X || (reserved)
|}
|}
=== Directories ===
Directories are inodes which contain some number of "entries" as their contents. These entries are nothing more than a name/inode pair. For instance the inode corresponding to the root directory might have an entry with the name of "etc" and an inode value of 50. A directory inode stores these entries in a linked-list fashion in its contents blocks.
The root directory is Inode 2.
The total size of a directory entry may be longer then the length of the name would imply (The name may not span to the end of the record), and records have to be aligned to 4-byte boundaries. Directory entries are also not allowed to span multiple blocks on the file-system, so there may be empty space in-between directory entries. Empty space is however not allowed in-between directory entries, so any possible empty space will be used as part of the preceding record by increasing its record length to include the empty space. Empty space may also be equivalently marked by a separate directory entry with an inode number of zero, indicating that directory entry should be skipped.
=== Directory Entry ===
{| {{Wikitable}}
! Starting
Byte
! Ending
Byte
! Size
in Bytes
! Field Description
|-
| 0 || 3 || 4 || Inode
|-
| 4 || 5 || 2 || Total size of this entry (Including all subfields)
|-
| 6 || 6 || 1 || Name Length least-significant 8 bits
|-
| 7 || 7 || 1 || [[#Directory_Entry_Type_Indicators|Type indicator]] (only if the feature bit for "directory entries have file type byte" is set, else this is the most-significant 8 bits of the Name Length)
|-
| 8 || 8+N-1 || N || Name characters
|}
====
{| {{Wikitable}}
|-
! Value
! Type Description
|-
| 0 || Unknown type
|-
| 1 || Regular file
|-
| 2 || Directory
|-
| 3 || Character device
|-
| 4 || Block device
|-
|
|-
| 6 || Socket
|-
| 7 || Symbolic link (soft link)
|}
==
=== How To Read
# Read the
# Determine which block group the inode belongs to.
# Read the Block Group Descriptor corresponding to the Block Group which contains the inode to be looked up.
# From the Block Group Descriptor, extract the location of the block group's inode table.
#
# Index the inode table (taking into account non-standard inode size).
Directory entry information and file contents are located within the data blocks that the Inode points to.
=== How To Read the Root Directory ===
The root directory's inode is defined to always be 2. Read/parse the contents of inode 2.
==See
===External Links===
* [http://www.nongnu.org/ext2-doc/ ext2-doc project: Second Extended File System] - implementation-oriented documentation, describes internal structure in human language.
* [http://web.mit.edu/tytso/www/linux/ext2intro.html Design and Implementation of the Second Extended Filesystem] (overview)
* [http://ext2.sourceforge.net/2005-ols/paper-html/ State of the Art: Where we are with the Ext3 filesystem] - Paper by Mingming Cao, Theodore Y. Ts'o, Badari Pulavarty, and Suparna Bhattacharya describing extended features for ext2
[[Category:Filesystems]]
[[de:Ext2]]
|