This article may be too technical for most readers to understand. Please help improve this article to make it understandable to non-experts, without removing the technical details. The talk page may contain suggestions. (December 2011)

File Allocation Table (FAT) is the name of a computer file system architecture and a family of industry standard file systems utilizing it.

The FAT file system is technically relatively simple yet robust. It offers reasonably good performance even in light-weight implementations and is therefore widely adopted and supported by virtually all existing operating systems for personal computers. This makes it a well-suited format for data exchange between computers and devices of almost any type and age from the early 1980s up to the present.

Originally designed in the late 1970s for use on floppy disks, it was soon adapted and used almost universally on hard disks throughout the DOS and Windows 9x eras for two decades. With the introduction of more powerful computers and operating systems, its use on hard drives has since started to decline, but it continues to be used on many computer systems.

Today, FAT file systems are still commonly found on floppy disks, solid-state memory cards, flash memory cards, and on many portable and embedded devices.

The name of the file system originates from the file system's prominent usage of an index table, the FAT, statically allocated at the time of formatting. The table contains entries for each cluster, a contiguous area of disk storage. Each entry contains either the number of the next cluster in the file, or else a marker indicating end of file, unused disk space, or special reserved areas of the disk. The root file directory of the disk contains the number of the first cluster; the operating system can then traverse the FAT table, looking up the cluster number of each successive part of the disk file as a cluster chain until the end of the file is reached.

As disk drives have evolved, the maximum number of clusters has significantly increased, and so the number of bits used to identify each cluster has grown. The successive major versions of the FAT format are named after the number of table element bits: 12 (FAT12), 16 (FAT16), and 32 (FAT32). Each of these variants is still in use. The FAT standard has also been expanded in other ways while generally preserving backward compatibility with existing software.

Overview[link]

Uses[link]

The FAT file system is simple yet robust. It offers reasonable performance even in light-weight implementations and is therefore widely adopted and supported by virtually all existing operating systems for personal computers as well as some home computers and embedded systems. This makes it a useful format for solid-state memory cards and a convenient way to share data between operating systems.

FAT file systems are the default file system for removable media (with the exception of CDs and DVDs) and as such are commonly found on floppy disks, super-floppies, memory and flash memory cards or USB flash drives and are supported by most portable devices such as PDAs, digital cameras, camcorders, media players, or mobile phones. While FAT12 is omnipresent on floppy disks, FAT16 and FAT32 are typically found on the larger media.

FAT was also commonly used on hard disks throughout the DOS and Windows 9x eras, but its use on hard drives has declined since the introduction of Windows XP, which primarily uses the newer NTFS. FAT is still used in hard drives expected to be used by multiple operating systems, such as in shared Windows and Linux environments.

Due to the widespread use of FAT-formatted media, many operating systems have provided support for FAT, and subsequently VFAT and FAT32, through official or third-party file system handlers. For example, Linux, FreeBSD, BeOS and JNode provide inbuilt support for FAT.

Mac OS 9 and Mac OS X also support FAT file systems on volumes other than the boot disk. AmigaOS supports FAT through the CrossDOS file system.

For many purposes, the NTFS file system is superior to FAT in terms of features and reliability; its main drawbacks are the size overhead for small volumes and the very limited support by anything other than the NT-based versions of Windows, since the exact specification is a trade secret of Microsoft. The availability of NTFS-3G since mid-2006 has led to much improved NTFS support in Unix-like operating systems, considerably alleviating this concern. It is still not possible to use NTFS in DOS-like operating systems without third-party drivers, which in turn makes it difficult to use a DOS floppy for recovery purposes. Microsoft provided a recovery console to work around this issue, but for security reasons it severely limited what could be done through the Recovery Console by default. The movement of recovery utilities to boot CDs based on BartPE or Linux (with NTFS-3G) is finally eroding this drawback.

FAT has a long history (over three decades) of usage on desktops and portable computers, and it is frequently used in embedded solutions. It continues to be the most widespread file system worldwide.

For floppy disks, FAT has been standardized as ECMA-107^[4] and ISO/IEC 9293:1994^[5] (superseding ISO 9293:1987^[6]). These standards cover FAT12 and FAT16 with only short 8.3 filename support; long filenames with VFAT are partially patented.^[7]

Historical evolution[link]

File system types[link]

Original 8-bit FAT[link]

Designed and coded by Marc McDonald, Microsoft Stand-alone Disk BASIC introduced the FAT in 1977 with 8-bit table elements,^[8] produced for NCR's 8-bit 8080 file system. The FAT, born during one of a series of discussions between McDonald and Bill Gates, was used in a stand-alone version of Microsoft BASIC for the 8086 chip in 1979 and eventually in the M-DOS operating system.^[8] The Microsoft Disk BASIC version supported three FATs.^[9]

FAT12[link]

In 1980, while borrowing Microsoft's FAT concept for Seattle Computer Products' 86-DOS, Tim Paterson extended the table elements to 12 bits for his implementation of FAT12, whereas the 8-bit file system precursor was not supported by 86-DOS. Later in 1981 86-DOS evolved into Microsoft's MS-DOS and IBM PC DOS.^[8][10][11]

Originally designed as a file system for floppy disks, FAT12 used 12-bit entries for the cluster addresses in the FAT, which not only limited the maximum generally possible count of data clusters to 4078 (for data clusters 0x002 to 0xFEF; in C-notation for hexadecimal numbers)^[12][13] or in some controlled scenarios even up to 4084 (for data clusters 0x002 to 0xFF5),^[4][14][15] but made FAT manipulation tricky with the PC's 8-bit and 16-bit registers. (NB. While MS-DOS and PC DOS support up to 4084 data clusters on FAT12 volumes in general, cluster value 0xFF0^{[nb 1]} is treated as additional end-of-chain marker on any FAT12 volume^[16] since MS-DOS/PC DOS 3.3, therefore restricting the maximum practical number of data clusters to 4078 for compatibility purposes with these operating systems.)

The disk's size was stored and calculated as a 16-bit count of sectors, which limited the size to 32 MB for a logical sector size of 512 bytes. FAT12 was used by several manufacturers with different physical formats, but a typical floppy disk at the time was 5.25-inch (130 mm), single-sided, 40 tracks, with 8 sectors per track, resulting in a capacity of 160 KB for both the system areas and files. The FAT12 limitations exceeded this capacity by a factor of ten or more. (NB. The 32 MB limit was later circumvented using logical sectored FATs with logical sector sizes larger than 512 bytes in some OEM versions of MS-DOS 3.x and it was further lifted with DOS 3.31, which supported 32-bit sector numbers, but seldomly used due to the availability of FAT16B.)

By convention, all the control structures were organized to fit inside the first track, thus avoiding head movement during read and write operations, although this varied depending on the manufacturer and physical format of the disk. A limitation which was not addressed until much later (with FAT32) was that any bad sector in the control structures area, track 0, could prevent the disk from being usable. The DOS formatting tool rejected such disks completely. Bad sectors were allowed only in the file data area and were marked with the reserved value 0xFF7 in the FAT. They made the entire containing cluster unusable.

While 86-DOS supported three disk formats (250.25 KB, 500.5 KB and 1232 KB with media descriptor bytes 0xFF and 0xFE) on 8-inch (200 mm) floppy drives, IBM PC DOS 1.0, released with the original IBM Personal Computer in 1981, supported only a 8-sector floppy format with a formatted capacity of 160 KB (media descriptor byte 0xFE) for single-sided 5.25-inch floppy drives, and PC DOS 1.1 added support for a double-sided format with 320 KB (media descriptor byte 0xFF). PC DOS 2.0 introduced support for 9-sector floppy formats with 180 KB (media descriptor byte 0xFC) and 360 KB (media descriptor byte 0xFD).

PC DOS 1.0 directory entries included only one date, the last modified date. PC DOS 1.1 added the last modified time. PC DOS 1.x file attributes included a hidden bit and system bit, with the remaining six bits undefined. At this time, DOS did not support a hierarchical file system, which was still acceptable, given that the number of files on a disk was typically not more than a few dozen.

The PC XT was the first PC with a hard drive from IBM, and PC DOS 2.0 supported that hard drive with FAT12 (media descriptor byte 0xF8). The fixed assumption of 8 sectors per clusters on hard disks practically limited the maximum partition size to 16 MB for 512 byte sectors and 4 KB clusters.

The BIOS Parameter Block (BPB) was introduced with PC DOS 2.0 as well, and this version also added read-only, archive, volume label, and directory attribute bits for hierarchical sub-directories.^[17]

MS-DOS 3.0 introduced support for high-density 1.2 MB 5.25-inch diskettes (media descriptor byte 0xF9), which notably had 15 sectors per track, hence more space for the FATs.

FAT12 remains in use on all common floppy disks, including 1.44 MB and later 2.88 MB disks (media descriptor byte 0xF0).

[edit] Initial FAT16

In 1984, IBM released the PC AT, which featured a 20 MB hard disk. Microsoft introduced MS-DOS 3.0 in parallel. Cluster addresses were increased to 16-bit, allowing for up to 65,524 clusters per volume, and consequently much greater file system sizes, at least in theory. However, the maximum possible number of sectors and the maximum (partition, rather than disk) size of 32 MB did not change. Therefore, although cluster addresses were 16 bits, this format was not what today is commonly understood as FAT16. A partition type 0x04 indicates this form of FAT16 with less than 65536 sectors (less than 32 MB for sector size 512).

With the initial implementation of FAT16 not actually providing for larger partition sizes than FAT12, the early benefit of FAT16 was to enable the use of smaller clusters, making disk usage more efficient, particularly for files several hundred bytes in size, which were far more common at the time.

MS-DOS 2.x hard disks larger than 15 MB are incompatible with later versions of MS-DOS.^[18] A 20 MB hard disk formatted under MS-DOS 3.0 was not accessible by the older MS-DOS 2.0 because MS-DOS 2.0 did not support version 3.0's FAT16. MS-DOS 3.0 could still access MS-DOS 2.0 style 8 KB-cluster partitions under 15 MB.

Logical sectored FAT[link]

When hard disks grew larger and the FAT12 and FAT16 file system implementation in MS-DOS / PC DOS did not provide means to take advantage of the extra storage, several manufacturers developed their own FAT variants to address the problem in their MS-DOS OEM issues.

Some vendors (AST and NEC) supported eight, instead of the standard four, primary partition entries in their custom extended Master Boot Record (MBR), and they adapted MS-DOS to use more than a single primary partition.

Other vendors worked around the volume size limits imposed by the 16-bit sector entries and arithmetics by increasing the size of the sectors the file system dealt with, thereby blowing up dimensions.

These so called logical sectors were larger (up to 8192 bytes) than the physical sector size (still typically 512 bytes) as expected by the ROM-BIOS INT 13h or the disk drive hardware. The DOS-BIOS or System BIOS would then combine multiple physical sectors into logical sectors for the file system to work with. These changes were transparent to the file system implementation in the DOS kernel, since on this abstraction level volumes are a series of logically addressable sectors (starting with logical sector 0) independent of the physical location of the volume on the physical medium and its geometry. The underlying DOS-BIOS translated these logical sectors into physical sectors according to partitioning information and the drive's physical geometry.

The drawback of this approach was a less memory-efficient sector buffering and deblocking in the DOS-BIOS, thereby causing an increased memory footprint for the DOS data structures. Since older DOS versions were not flexible enough to work with these logical geometries, the OEMs had to introduce new partition IDs for their FAT variants in order to hide them from off-the-shelf issues of MS-DOS and PC DOS. Known partition IDs for logical sectored FATs include: 0x08 (Commodore MS-DOS 3.x), 0x11 (Leading Edge MS-DOS 3.x), 0x14 (AST MS-DOS 3.x), 0x24 (NEC MS-DOS 3.30), 0x56 (AT&T MS-DOS 3.x), 0xE5 (Tandy MS-DOS), 0xF2 (Sperry IT MS-DOS 3.x, Unisys MS-DOS 3.3 — also used by Digital Research DOS Plus 2.1).^[19]

While non-standard and sub-optimal, these FAT variants are perfectly valid according to the specifications of the file system itself. Therefore, even if default issues of MS-DOS and PC DOS were not able to cope with them, most of these vendor-specific FAT12 and FAT16 variants can be mounted by more flexible file system implementations in operating systems such as DR-DOS, simply by changing the partition ID to one of the recognized types. Also, if they no longer need to be recognized by their original operating systems, existing partitions can be "converted" into FAT12 and FAT16 volumes more compliant with versions of MS-DOS/PC DOS like 5.0-6.3, which don't support sector sizes different from 512 bytes,^[20] by switching to a BPB with 32-bit entry for the number of sectors, as introduced since DOS 3.31 (see FAT16B below), keeping the cluster size and reducing the logical sector size in the BPB down to 512 bytes, while at the same time increasing the counts of logical sectors per cluster, reserved logical sectors, total logical sectors, and logical sectors per FAT by the same factor.

A parallel development in MS-DOS / PC DOS which allowed an increase in the maximum possible FAT size was the introduction of multiple FAT partitions on a hard disk. To allow the use of more FAT partitions in a compatible way, a new partition type was introduced in PC DOS 3.2 (1986), the extended partition (EBR),^[8] which is a container for an additional partition called logical drive. Since PC DOS 3.3 (April 1987), there is another, optional extended partition containing the next logical drive, and so on. The MBR of a hard disk can either define up to four primary partitions, or an extended partition in addition to up to three primary partitions.

[edit] Final FAT16

Finally in November 1987, Compaq MS-DOS 3.31 (a modified OEM version of MS-DOS 3.3 released by Compaq with their machines) introduced what today is simply known as the FAT16 format, with the expansion of the 16-bit disk sector count to 32 bits in the BPB. Although the on-disk changes were minor, the entire DOS disk driver had to be converted to use 32-bit sector numbers, a task complicated by the fact that it was written in 16-bit assembly language. The result was initially called the DOS 3.31 Large File System. Microsoft's DSKPROBE tool refers to type 0x06 as BigFAT,^[21] whereas some older versions of FDISK described it as BIGDOS. It is also known as FAT16B.

Since older versions of DOS could not cope with more than 65535 sectors, it was also necessary to introduce a new partition type for this format, to hide it from pre-DOS 3.31 issues of DOS. While the original form of FAT16 with less than 65536 sectors had a partition type 0x04, a type 0x06 indicates 65536 or more sectors. In addition to this changed partition ID and the expansion of the disk drivers format to cope with more than 65535, the only other difference between the original FAT16 and the FAT16B format is the usage of the newer BPB with 32-bit sector entry, so that newer operating systems can cope also with the original FAT16 format without any necessary changes. If partitions to be used by pre-DOS 3.31 issues of DOS need to be created by modern tools, the only criteria theoretically necessary to meet are a sector count of less than 65536 and the usage of the old partition ID. In practice, however, type 0x01 and 0x04 partitions should not be physically located outside the first 32 MB of the disk, due to other restrictions in MS-DOS 2.x, which could not cope with them otherwise.

In 1988, the FAT16B improvement became more generally available through DR DOS 3.31, MS-DOS 4.0 and OS/2 1.1. The limit on partition size was dictated by the 8-bit signed count of sectors per cluster, which originally had a maximum power-of-two value of 64. With the standard hard disk sector size of 512 bytes, this gives a maximum of 32 KB cluster size, thereby fixing the "definitive" limit for the FAT16 partition size at 2 GB for sector size 512. On magneto-optical media, which can have 1 or 2 KB sectors instead of 0.5 KB, this size limit is proportionally larger.

Much later, Windows NT increased the maximum cluster size to 64 KB, by considering the sectors-per-cluster count as unsigned. However, the resulting format was not compatible with any other FAT implementation of the time, and it generated greater internal fragmentation. Windows 98, SE and ME also supported reading and writing this variant, but its disk utilities did not work with it and some FCB services are not available for such volumes. This contributes to a confusing compatibility situation.

Prior to 1995, versions of DOS accessed the disk via CHS addressing only. When MS-DOS 7.0 / Windows 95 introduced LBA disk access, partitions could start being physically located outside the first ca. 8 GB of this disk and thereby out of the reach of the traditional CHS addressing scheme. Partitions partially or fully located beyond the CHS barrier therefore had to be hidden from non-LBA-enabled operating systems by using the new partion type 0x0E in the partition table instead. FAT16 partitions using this partition type are also named FAT16X.^[22] The only difference, compared to previous FAT16 partitions, is the fact that some CHS-related geometry entries in the BPB record, namely the number of sectors per track and the number of heads, may contain no or misleading values and should not be used.

The number of root directory entries available for FAT12 and FAT16 is determined when the volume is formatted, and is stored in a 16-bit field. For a given number RDE and sector size SS, the number RDS of root directory sectors is RDS=ceil((RDE×32)/SS), and RDE is normally chosen to fill these sectors, i.e., RDE*32=RDS*SS. FAT12 and FAT16 media typically use 512 root directory entries on non-floppy media. Some third-party tools, like mkdosfs, allow the user to set this parameter.^[23]

[edit] FAT32

In order to overcome the size limit of FAT16, while at the same time allowing DOS real mode code to handle the format, and without reducing available conventional memory unnecessarily, Microsoft expanded the cluster size yet again, calling the new revision FAT32. Cluster values are represented by 32-bit numbers, of which 28 bits are used to hold the cluster number. The boot sector uses a 32-bit field for the sector count, limiting the FAT32 volume size to 2 TB for sector size of 512 bytes and 16 TB for sector size of 4,096 bytes.^[24][25] FAT32 was introduced with MS-DOS 7.1 / Windows 95 OSR2 in 1996, although reformatting was needed to use it, and DriveSpace 3 (the version that came with Windows 95 OSR2 and Windows 98) never supported it. Windows 98 introduced a utility to convert existing hard disks from FAT16 to FAT32 without loss of data. In the Windows NT line, native support for FAT32 arrived in Windows 2000. A free FAT32 driver for Windows NT 4.0 was available from Winternals, a company later acquired by Microsoft. Since the acquisition the driver is no longer officially available. Since 1998, Caldera's dynamically loadable DRFAT32 driver could be used to enable FAT32 support in DR-DOS. The first version of DR-DOS to natively support FAT32 and LBA access was OEM DR-DOS 7.04 in 1999. That same year IMS introduced native FAT32 support with REAL/32 7.90, and IBM 4690 OS added FAT32 support with version 2.^[26] Ahead Software provided another dynamically loadable FAT32.EXE driver for DR-DOS 7.03 with Nero Burning ROM in 2004. IBM PC DOS introduced native FAT32 support with OEM PC DOS 7.10 in 2003.

The maximum possible size for a file on a FAT32 volume is 4 GB minus 1 byte or 4,294,967,295 (2³²−1) bytes. This limit is a consequence of the file length entry in the directory table and would also affect huge FAT16 partitions with a sufficient sector size.^[1] Video applications, large databases, and some other software easily exceed this limit.

The open FAT+^[2] specification proposes how to store larger files up to 256 GB minus 1 byte or 274,877,906,943 (2³⁸−1) bytes on slightly modified and otherwise backwards compatible FAT32 volumes, but imposes a risk that disk tools or FAT32 implementations not aware of this extension may truncate or delete files exceeding the normal FAT32 file size limit. Also, support for FAT32+ (and FAT16+) is limited to some versions of DR-DOS and FreeDOS and not available in mainstream operating systems so far. (This extension is critically incompatible with the /EAS option of the FAT32.IFS method to store OS/2 extended attributes on FAT32 volumes.)

As with previous file systems, the design of the FAT32 file system does not include direct built-in support for long filenames, but FAT32 volumes can optionally hold VFAT long filenames in addition to short filenames in exactly the same way as VFAT long filenames have been optionally implemented for FAT12 and FAT16 volumes.

Two partition types have been reserved for FAT32 partitions, 0x0B and 0x0C. The latter type is also named FAT32X in order to indicate usage of LBA disk access instead of CHS. On such partitions, some CHS-related geometry entries in the EBPB record, namely the number of sectors per track and the number of heads, may contain no or misleading values and should not be used.

A free Windows-based FAT32 formatter is available that overcomes many of the arbitrary limitations imposed by Microsoft.^[27]

Extensions[link]

[edit] Extended Attributes

OS/2 heavily depends on extended attributes (EAs) and stores them in a hidden file called "EA␠DATA.␠SF" in the root directory of the FAT12 or FAT16 volume. This file is indexed by two previously reserved bytes in the file's (or directory's) directory entry at offset 0x14.^[28] In the FAT32 format, these bytes hold the upper 16 bits of the starting cluster number of the file or directory, hence making it impossible to store OS/2 EAs on FAT32 using this method.

However, the third-party FAT32 installable file system (IFS) driver FAT32.IFS version 0.70 and higher by Henk Kelder & Netlabs for OS/2 and eComStation stores extended attributes in extra files with filenames having the string "␠EA.␠SF" appended to the regular filename of the file to which they belong. The driver also utilizes the byte at offset 0x0C in directory entries to store a special mark byte indicating the presence of extended attributes to help speed up things.^[29][30] (This extension is critically incompatible with the FAT32+ method to store files larger than 4 GB minus 1 on FAT32 volumes.^[2])

Extended attributes are accessible via the Workplace Shell desktop, through REXX scripts, and many system GUI and command-line utilities (such as 4OS2).^[31]

To accommodate its OS/2 subsystem, Windows NT supports the handling of extended attributes in HPFS, NTFS, FAT12 and FAT16. It stores EAs on FAT12, FAT16 and HPFS using exactly the same scheme as OS/2, but does not support any other kind of ADS as held on NTFS volumes. Trying to copy a file with any ADS other than EAs from an NTFS volume to a FAT or HPFS volume gives a warning message with the names of the ADSs that will be lost. It does not support the FAT32.IFS method to store EAs on FAT32 volumes.

Windows 2000 onward acts exactly as Windows NT, except that it ignores EAs when copying to FAT32 without any warning (but shows the warning for other ADSs, like "Macintosh Finder Info" and "Macintosh Resource Fork").

Cygwin uses "EA␠DATA.␠SF" files as well.

[edit] Long file names

One of the user experience goals for the designers of Windows 95 was the ability to use long filenames (LFNs—up to 255 UTF-16 code points long), in addition to classic 8.3 filenames (SFNs). For backward compatibility LFNs were implemented as an optional extension on top of the existing FAT file system structures using a workaround in the way directory entries are laid out.

This transparent method to store long file names in the existing FAT file systems without altering their data structures is usually known as VFAT (for "Virtual FAT") after the Windows 95 virtual device driver.^{[nb 2]}

In Windows NT, support for VFAT long filenames started from version 3.5.

Non VFAT-enabled operating systems can still access the files under their short file name alias without restrictions, however, the associated long file names may get lost, when files with long file names are copied under non VFAT-aware operating systems.

OS/2 added long filename support to FAT using extended attributes (EA) before the introduction of VFAT; thus, VFAT long filenames are invisible to OS/2, and EA long filenames are invisible to Windows.

In order to support Java applications, the FlexOS-based IBM 4690 OS version 2 introduced their own virtual file system (VFS) architecture to store long filenames in the FAT file system in a backwards compatible fashion. If enabled, the virtual filenames (VFN) are available under separate logical drive letters, whereas the real filenames (RFN) remain available under the original drive letters.^[32]

[edit] Forks and Alternate Data Streams

The FAT file system itself is not designed for supporting Alternate Data Streams (ADS), but some operating systems that heavily depend on them have devised various methods for handling them in FAT drives. Such methods either store the additional information in extra files and directories (Mac OS), or give new semantics to previously unused fields of the FAT on-disk data structures (OS/2 and Windows NT).

Mac OS using PC Exchange stores its various dates, file attributes and long filenames in a hidden file called "FINDER.DAT", and resource forks (a common Mac OS ADS) in a subdirectory called "RESOURCE.FRK", in every directory where they are used. From PC Exchange 2.1 onwards, they store the Mac OS long filenames as standard FAT long filenames and convert FAT filenames longer than 31 characters to unique 31-character filenames, which can then be made visible to Macintosh applications. Presumably, Star Trek used the same mechanisms on FAT12 and FAT16 volumes in 1992.

Mac OS X stores resource forks and metadata (file attributes, other ADS) in a hidden file with a name constructed from the owner filename prefixed with "._", and Finder stores some folder and file metadata in a hidden file called ".DS␠Store".

UMSDOS permissions and filenames[link]

Early Linux distributions also supported a format known as UMSDOS, a FAT variant with Unix file attributes (such as long file name and access permissions) stored in a separate file called "--linux-.---". UMSDOS fell into disuse after VFAT was released and it is not enabled by default in Linux kernels from version 2.5.7 onwards.^[33]

Derivatives[link]

FATX[link]

FATX is a family of file systems designed for Microsoft's Xbox video game console hard disk drives and memory cards,^[34][35] introduced in 2001.

While resembling the same basic design ideas as FAT16 and FAT32, the FATX16 and FATX32 on-disk structures are simplified but fundamentally incompatible with normal FAT16 and FAT32 file systems, making it impossible for normal FAT file system drivers to mount such volumes.

The non-bootable superblock sector is 4 KB in size and holds a 18 byte large BPB-like structure completely different from normal BPBs. Clusters are typically 16 KB in size and there is only one copy of the FAT on the Xbox. Directory entries are 64 bytes in size instead of the normal 32 bytes. Files can have filenames up to 42 characters long using the OEM character set and be up to 4 GB minus 1 byte in size. The on-disk timestamps hold creation, modification and access dates and times but differ from FAT: in FAT, the epoch is 1980; in FATX, the epoch is 2000.^{[citation needed]} On the Xbox 360, the epoch is 1980.^[36]

exFAT[link]

exFAT is an incompatible file system, that was introduced with Windows Embedded CE 6.0 in November 2006. It is loosely based on the File Allocation Table architecture, but proprietary and protected by patents.

The MBR partition type is 0x07 (the same as used for IFS, HPFS, NTFS, etc.). exFAT is intended for use on flash drives (such as SDXC and Memory Stick XC), where FAT32 is otherwise used.

It offers several benefits over FAT32 including breaking the 4 GB file size limit of FAT32 (only without the FAT+^[2] extension, which allows files larger than 4 GB also on FAT32 volumes), being more space-efficient for files smaller than 64 KB on large volumes and, compared to light-weight implementations of FAT32 in DOS and some embedded systems, it can offer faster seeks if more than a few thousand files are stored in a single sub-directory, whereas FAT32 is typically faster than exFAT for larger files as used on digital cameras, camcorders and media players or when flash cards are used mainly for archival purposes.

Storage devices formatted as exFAT cannot exchange data with equipment not supporting the format. Much equipment does not support exFAT, which requires acquisition of a license from Microsoft; this involves a cost, and is also against the policy of open source operating systems.

Technical design[link]

Layout[link]

A FAT file system is composed of four different sections:

• The Reserved sectors, located at the very beginning.

The first reserved sector (sector 0) is the Boot Sector (aka Volume Boot Record (VBR)). It includes an area called the BIOS Parameter Block (with some basic file system information, in particular its type, and pointers to the location of the other sections) and usually contains the operating system's boot loader code.

Important information from the Boot Sector is accessible through an operating system structure called the Drive Parameter Block (DPB) in DOS and OS/2.

The total count of reserved sectors is indicated by a field inside the Boot Sector.

For FAT32 file systems, the reserved sectors include a File System Information Sector at sector 1 and a Backup Boot Sector at sector 6.

• The FAT Region.

This typically contains two copies (may vary) of the File Allocation Table for the sake of redundancy checking, although rarely used, even by disk repair utilities.

These are maps of the Data Region, indicating which clusters are used by files and directories. In FAT12 and FAT16 they immediately follow the reserved sectors.

Typically the extra copies are kept in tight synchronization on writes, and on reads they are only used when errors occur in the first FAT. In FAT32, it is possible to switch from the default behaviour and select a single FAT out of the available ones to be used for diagnosis purposes.

• The Root Directory Region.

This is a Directory Table that stores information about the files and directories located in the root directory. It is only used with FAT12 and FAT16, and imposes on the root directory a fixed maximum size which is pre-allocated at creation of this volume. FAT32 stores the root directory in the Data Region, along with files and other directories, allowing it to grow without such a constraint. Thus, for FAT32, the Data Region starts here.

• The Data Region.

This is where the actual file and directory data is stored and takes up most of the partition. Traditionally, the unused parts of the data region are initialized with a filler value of 0xF6 during format on IBM compatible machines, but also used on the Atari Portfolio. 8-inch CP/M floppies typically came pre-formatted with a value of 0xE5;^[9] by way of Digital Research this value was also used on Atari ST formatted floppies.^{[nb 3]} Some modern formatters wipe hard disks with a value of 0x00, whereas a value of 0xFF is used on flash disks to reduce wear. The latter value is typically also used on ROM disks. (Some advanced formating tools allow to configure the format filler byte.^{[nb 4]})

The size of files and subdirectories can be increased arbitrarily (as long as there are free clusters) by simply adding more links to the file's chain in the FAT. Note however, that files are allocated in units of clusters, so if a 1 KB file resides in a 32 KB cluster, 31 KB are wasted.

FAT32 typically commences the Root Directory Table in cluster number 2: the first cluster of the Data Region.

FAT uses little-endian format for all entries in the header (except for, where explicitly mentioned, for some entries on Atari ST boot sectors) and the FAT(s). It is possible to allocate more FAT sectors than necessary for the number of clusters. The end of the last FAT sector can be unused if there are no corresponding clusters. The total number of sectors (as noted in the boot record) can be larger than the number of sectors used by data (clusters × sectors per cluster), FATs (number of FATs × sectors per FAT), and hidden sectors including the boot sector: this would result in unused sectors at the end of the volume. If a partition contains more sectors than the total number of sectors occupied by the file system it would also result in unused sectors at the end of the volume.

[edit] Boot Sector

On non-partitioned devices, e.g., floppy disks, the Boot Sector (VBR) is the first sector. For partitioned devices such as hard drives, the first sector is the Master Boot Record defining partitions, while the first sector of partitions formatted with a FAT file system is again the Boot Sector.

Common structure of the first 11 bytes used by most FAT versions for IBM compatible x86-machines since DOS 2.0 are:

FAT-formatted Atari ST floppies have a very similar boot sector layout:

FAT12-formatted MSX-DOS volumes have a very similar boot sector layout:

[edit] BIOS Parameter Block

Common structure of the first 25 bytes of the BIOS Parameter Block (BPB) used by FAT versions since DOS 2.0 are (bytes at sector offset 0x00B to 0x017 are stored since DOS 2.0, but not always used before DOS 3.2, values at 0x018 to 0x01B are used since DOS 3.0):

DOS 3.0 BPB:

DOS 3.2 BPB:

DOS 3.31 BPB:

A simple formula translates a given cluster number CN to a logical sector number LSN:^[4][5][6]

1. Determine (once) SSA=RSC+FN×SF+ceil((32×RDE)/SS), where the reserved sector count RSC is stored at offset 0x00E, the FAT number FN at offset 0x010, the sectors per FAT SF at offset 0x016 (FAT12/FAT16) or 0x024 (FAT32), the root directory entries RDE at offset 0x011, the sector size SS at offset 0x00B, and ceil(x) rounds up to a whole number.
2. Determine LSN=SSA+(CN-2)×SC, where the sectors per cluster SC are stored at offset 0x00D.

A translation of CHS to LSN is also simple: LSN=SPT×(HN+(NOS×TN))+SN-1, where the sectors per track SPT are stored at offset 0x018, and the number of sides NOS at offset 0x01A. Track number TN, head number HN, and sector number SN correspond to Cylinder-head-sector: the formula gives the known CHS to LBA translation.

[edit] Extended BIOS Parameter Block

Further structure used by FAT12 and FAT16 since OS/2 1.0 and DOS 4.0, also known as Extended BIOS Parameter Block (EBPB) (bytes below sector offset 0x024 are the same as for the DOS 3.31 BPB):

[edit] FAT32 Extended BIOS Parameter Block

In essence FAT32 inserts 28 bytes into the EBPB, followed by the remaining 26 EBPB bytes as shown above for FAT12 and FAT16:

Exceptions[link]

The implementation of FAT used in MS-DOS for the Apricot PC had a different boot sector layout, to accommodate that computer's non-IBM compatible BIOS. The jump instruction and OEM name were omitted, and the MS-DOS file system parameters (offsets 0x00B-0x017 in the standard sector) were located at offset 0x050. Later versions of Apricot MS-DOS gained the ability to read and write disks with the standard boot sector in addition to those with the Apricot one.

DOS Plus on the BBC Master 512 did not use conventional boot sectors at all. Data disks omitted the boot sector and began with a single copy of the FAT (the first byte of the FAT was used to determine disk capacity) while boot disks began with a miniature ADFS file system containing the boot loader, followed by a single FAT. It could also access standard PC disks formatted to 180 KB or 360 KB, again using the first byte of the FAT to determine the capacity.

FS Information Sector[link]

The "FS Information Sector" was introduced in FAT32^[50] for speeding up access times of certain operations (in particular, getting the amount of free space). It is located at a logical sector number specified in the FAT32 EBPB boot record at position 0x030 (usually logical sector 1, immediately after the boot record itself).

The sector's data may be outdated and not reflect the current media contents, because not all operating systems update or use this sector, and even if they do, the contents is not valid when the medium has been ejected without properly unmounting the volume or after a power-failure. Therefore, operating systems should first inspect a volume's optional shutdown status bitflags residing in the FAT entry of cluster 1 or the FAT32 EBPB at offset 0x041 and ignore the data stored in the FS information sector, if these bitflags indicate that the volume was not properly unmounted before. This does not cause any problems other than a possible speed penalty for the first free space query or data cluster allocation; see fragmentation.

If this sector is present on a FAT32 volume, the minimum allowed logical sector size is 512 bytes, whereas otherwise it would be 128 bytes. Some FAT32 implementations support a slight variation of Microsoft's specification by making the FS information sector optional by specifying a value of 0 in the entry at offset 0x030.

File Allocation Table[link]

A partition is divided up into identically sized clusters, small blocks of contiguous space. Cluster sizes vary depending on the type of FAT file system being used and the size of the partition, typically cluster sizes lie somewhere between 2 KB and 32 KB.

Each file may occupy one or more of these clusters depending on its size; thus, a file is represented by a chain of these clusters (referred to as a singly linked list). However these clusters are not necessarily stored adjacent to one another on the disk's surface but are often instead fragmented throughout the Data Region.

The File Allocation Table (FAT) is a list of entries that map to each cluster on the partition. Each entry records one of five things:

• the cluster number of the next cluster in a chain
• a special end of cluster-chain (EOC) entry that indicates the end of a chain
• a special entry to mark a bad cluster
• a zero to note that the cluster is unused

Each version of the FAT file system uses a different size for FAT entries. Smaller numbers result in a smaller FAT, but waste space in large partitions by needing to allocate in large clusters. The FAT12 file system uses 12 bits per FAT entry, thus two entries span 3 bytes. It is consistently little-endian: if those three bytes are considered as one little-endian 24-bit number, the 12 least significant bits represent the first entry (cluster 0) and the 12 most significant bits the second (cluster 1).

The first two entries in a FAT store special values:

The first entry (cluster 0 in the FAT) holds the media descriptor (allowed values 0xF0-0xFF with 0xF1-0xF7 reserved), which is also copied into the BPB of the boot sector, offset 0x015 since DOS 2.0. The remaining 4 bits (if FAT12), 8 bits (if FAT16) or 20 bits (if FAT32) of this entry are always 1. For media descriptors other than 0xFF (and 0x00) it is possible to determine the correct nibble and byte order (to be) used by the file system driver, however, the FAT file system officially uses a little-endian representation only and there are no known implementations of variants using big-endian values instead.

The second entry (cluster 1 in the FAT) nominally stores the end-of-cluster-chain marker as used by the formater, but typically always holds 0xFFF / 0xFFFF / 0x0FFFFFFF, that is, with the exception of bits 31-28 on FAT32 volumes these bits are normally always set. Some Microsoft operating systems, however, set these bits if the volume is not the volume holding the running operating system (that is, use 0xFFFFFFFF instead of 0x0FFFFFFF here).^[51] (In conjunction with alternative end-of-chain markers the lowest bits 2-0 can become zero for the lowest allowed end-of-chain marker 0xFF8 / 0xFFF8 / 0x?FFFFFF8; bit 3 should be reserved as well given that clusters 0xFF0 / 0xFFF0 / 0x?FFFFFF0 and higher are officially reserved.)

Since DOS 7.1 the two most-significant bits of the cluster entry may hold two optional bitflags representing the current volume status on FAT16 and FAT32, but not on FAT12 volumes. These bitflags are not supported by all operating systems, but operating systems supporting this feature would set these bits on shutdown and clear the most significant bit on startup:
If bit 15 (FAT16) respectively bit 27 (FAT32) is not set when mounting the volume, the volume was not properly unmounted before shutdown or ejection and thus is in an unknown and possibly "dirty" state.^[48] On FAT32 volumes, the FS Information Sector may hold outdated data and thus should not be used. The operating system would then typically run CHKDSK on the next startup (but not on insertion of removable media) to ensure and possibly reestablish the volume's integrity.
If bit 14 (FAT16) respectively bit 26 (FAT32) is cleared, the operating system has encountered disk I/O errors when it was last used, a possible indication for bad sectors. Operating systems aware of this extension will interpret this as a recommendation to carry out a surface scan (SCANDISK) on the next boot.^[48] (A similar set of bitflags exists in the FAT12/FAT16 EBPB at offset 0x1A or the FAT32 EBPB at offset 0x36. While the cluster 1 entry can be accessed by file system drivers once they have mounted the volume, the EBPB entry is available even when the volume is not mounted and thus easier to use by disk block device drivers or partitioning tools.)

If the number of FATs in the BPB is not set to 2, the second cluster entry in the first FAT (cluster 1) may also reflect the status of a TFAT volume for TFAT-aware operating systems. If the cluster 1 entry in that FAT holds the value 0, this may indicate that the second FAT represents the last known valid transaction state and should be copied over the first FAT, whereas the first FAT should be copied over the second FAT if all bits are set.

Because these first two FAT entries store special values, there is no data cluster 0 or 1. The first data cluster (after the root directory if FAT12/FAT16) is cluster 2.^[17]

FAT entry values:

Despite its name FAT32 uses only 28 bits of the 32 possible bits. The upper 4 bits are usually zero, but are reserved and should be left untouched. A standard conformant FAT32 file system driver or maintenance tool must not rely on the upper 4 bits to be zero and it must strip them off before evaluating the cluster number in order to cope with possible future expansions where these bits may be used for other purposes. They must not be cleared by the file system driver when allocating new clusters, but should be cleared during a reformat.

Directory table[link]

A directory table is a special type of file that represents a directory (also known as a folder). Each file or directory stored within it is represented by a 32-byte entry in the table. Each entry records the name, extension, attributes (archive, directory, hidden, read-only, system and volume), the date and time of last modification, the address of the first cluster of the file/directory's data and finally the size of the file/directory. Aside from the Root Directory Table in FAT12 and FAT16 file systems, which occupies the special Root Directory Region location, all Directory Tables are stored in the Data Region. The actual number of entries in a directory stored in the Data Region can grow by adding another cluster to the chain in the FAT.

The FAT file system itself does not impose any limits on the depth of a subdirectory tree for as long as there are free clusters available to allocate the subdirectories, however, the Current Directory Structure under MS-DOS/PC DOS limits the absolute path of a directory to 66 characters (including the drive letter, but excluding the zero delimiter),^[4][5][6] thereby limiting the maximum supported depth of subdirectories to 32, whatever occurs earlier. Concurrent DOS, Multiuser DOS and DR DOS 3.31 to 6.0 (up to including the 1992-11 updates) do not store absolute paths to working directories internally and therefore do not show this limitation.^[53] The same applies to Atari GEMDOS, but the Atari Desktop does not support more than 8 sub-directory levels. Most applications aware of this extension support paths up to at least 127 bytes. FlexOS, 4680 OS and 4690 OS support a length of up to 127 bytes as well, allowing depths down to 60 levels.^[26] PalmDOS, DR DOS 6.0 (since BDOS 7.1) and higher, Novell DOS, and OpenDOS sport a MS-DOS-compatible CDS and therefore have the same length limits as MS-DOS/PC DOS.

Note that before each entry there can be "fake entries" to support a long filename (LFN); see further below.

Legal characters for DOS short filenames include the following:

• Upper case letters A–Z
• Numbers 0–9
• Space (though trailing spaces in either the base name or the extension are considered to be padding and not a part of the file name, also filenames with space in them could not easily be used on the DOS command line prior to Windows 95 because of the lack of a suitable escaping system). Another exception are the internal commands MKDIR/MD and RMDIR/RD under DR-DOS which accept single arguments and therefore allow spaces to be entered.

• ! # $ % & ' ( ) - @ ^ _ ` { } ~
• Values 128–255

This excludes the following ASCII characters:

• " * / : < > ? \ |
  Windows/MS-DOS has no shell escape character
• + , . ; = [ ]
  They are allowed in long file names only.
• Lower case letters a–z
  Stored as A–Z. Allowed in long file names.
• Control characters 0–31
• Value 127 (DEL)

The following additional characters are allowed on Atari's GEMDOS, but should be avoided for compatibility with MS-DOS/PC DOS:

• " + , ; < = > [ ] |

The semicolon (;) should be avoided in filenames under DR DOS 3.31 and higher, PalmDOS, Novell DOS, OpenDOS, Concurrent DOS, Multiuser DOS, System Manager and REAL/32, because it may conflict with the syntax to specify file and directory passwords: "...\DIRSPEC.EXT;DIRPWD\FILESPEC.EXT;FILEPWD". The operating system will strip off one^[53] (and also two—since DR-DOS 7.02) semicolons and pending passwords from the filenames before storing them on disk. (The command processor 4DOS uses semicolons for include lists and requires the semicolon to be doubled for password protected files with any commands supporting wildcards.^[53])

The @-character is used for filelists by many DR-DOS, PalmDOS, Novell DOS, OpenDOS and Multiuser DOS, System Manager and REAL/32 commands as well as by 4DOS and may therefore sometimes be difficult to use in filenames.^[53]

Under Multiuser DOS and REAL/32, the exclamation mark (!) is not a valid filename character since it is used to separate multiple commands in a single command line.^[53]

Under IBM 4680 OS and 4690 OS, the following characters are not allowed in filenames:

• ? * : . ; , [ ] ! + = < > " - / \ |

Additionally, the following special characters are not allowed in the first, fourth, fifth and eight character of a filename, as they conflict with the host command processor (HCP) and input sequence table build file names:

• @ # ( ) { } $ &

The DOS file names are in the current OEM character set: this can have surprising effects if characters handled in one way for a given code page are interpreted differently for another code page (DOS command CHCP) with respect to lower and upper case, sorting, or validity as file name character.

[edit] Directory entry

Before Microsoft added support for long filenames and creation/access time stamps, bytes 0x0C–0x15 of the directory entry were used by other operating systems to store additional metadata, most notably the operating systems of the Digital Research family stored file passwords, access rights, owner IDs, and file deletion data there. While Microsoft's newer extensions are not fully compatible with these extensions by default, most of them can coexist in third-party FAT implementations (at least on FAT12 and FAT16 volumes).

Directory entries, both in the Root Directory Region and in subdirectories, are of the following format (see also 8.3 filename):