http://baike.baidu.com/link?url=xkck9gR5bzOx0oBKP1qNJwGGq3IO56V4i8cg9zTSpSDMVBMA0F7jr0AdkQTGyk7F0FGjUv431XSlKf24beUo6a

UUID
编辑
 
UUID含义是通用唯一识别码 (Universally Unique Identifier),这 是一个软件建构的标准,也是被开源软件基金会 (Open Software Foundation, OSF) 的组织应用在分布式计算环境 (Distributed Computing Environment, DCE) 领域的一部分。

目 录

1作用

2组成

3应用

 
 

1作用

UUID 的目的,是让分布式系统中的所有元素,都能有唯一的辨识资讯,而不需要透过中央控制端来做辨识资讯的指定。如此一来,每个人都可以建立不与其它人冲突的 UUID。在这样的情况下,就不需考虑数据库建立时的名称重复问题。目前最广泛应用的 UUID,即是微软的 Microsoft's Globally Unique Identifiers (GUIDs),而其他重要的应用,则有 Linux ext2/ext3 档案系统、LUKS 加密分割区、GNOME、KDE、Mac OS X 等等。

2组成

UUID是指在一台机器上生成的数字,它保证对在同一时空中的所有机器都是唯一的。通常平台会提供生成的API。按照开放软件基金会(OSF)制定的标准计算,用到了以太网卡地址、纳秒级时间、芯片ID码和许多可能的数字
UUID由以下几部分的组合:
(1)当前日期和时间,UUID的第一个部分与时间有关,如果你在生成一个UUID之后,过几秒又生成一个UUID,则第一个部分不同,其余相同。
(2)时钟序列。
(3)全局唯一的IEEE机器识别号,如果有网卡,从网卡MAC地址获得,没有网卡以其他方式获得。
UUID的唯一缺陷在于生成的结果串会比较长。关于UUID这个标准使用最普遍的是微软的GUID(Globals Unique Identifiers)。在ColdFusion中可以用CreateUUID()函数很简单地生成UUID,其格式为:xxxxxxxx-xxxx- xxxx-xxxxxxxxxxxxxxxx(8-4-4-16),其中每个 x 是 0-9 或 a-f 范围内的一个十六进制的数字。而标准的UUID格式为:xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxx (8-4-4-4-12),可以从cflib 下载CreateGUID() UDF进行转换。

3应用

使用UUID的好处在分布式的软件系统中(比如:DCE/RPC, COM+,CORBA)就能体现出来,它能保证每个节点所生成的标识都不会重复,并且随着WEB服务等整合技术的发展,UUID的优势将更加明显。根据使用的特定机制,UUID不仅需要保证是彼此不相同的,或者最少也是与公元3400年之前其他任何生成的通用唯一标识符有非常大的区别。
通用唯一标识符还可以用来指向大多数的可能的物体。微软和其他一些软件公司都倾向使用全球唯一标识符(GUID),这也是通用唯一标识符的一种类型,可用来指向组建对象模块对象和其他的软件组件。第一个通用唯一标识符是在网络计算机系统(NCS)中创建,并且随后成为开放软件基金会(OSF)的分布式计算环境(DCE)的组件。
 
 
 
 
GUID
编辑
 
1.全局唯一标识符(GUID,Globally Unique Identifier)也称作 UUID(Universally Unique IDentifier) 。GUID是一种由算法生成的二进制长度为128位的数字标识符。GUID主要用于在拥有多个节点、多台计算机的网络或系统中。在理想情况下,任何计算机和计算机集群都不会生成两个相同的GUID。GUID 的总数达到了2^128(3.4×10^38)个,所以随机生成两个相同GUID的可能性非常小,但并不为0。GUID一词有时也专指微软对UUID标准的实现。
2.磁盘分区表方案
全局唯一标识分区表是一个实体硬盘分区表的结构布局的标准。

目 录

1简介

2格式

3特点

4争议

5程序

  1. 5.1 VB
  2. 5.2 C++
  3. 5.3 C#
  4. 5.4 Java
  5. 5.5 OC
 
 

1简介

全局唯一标识符,简称GUID,是一种由算法生成的唯一标识。GUID的主要目的是产生完全唯一的数字。在理想情况下,任何计算机计算机集群都不会生成两个相同的GUID。随机生成两个相同GUID的可能性是非常小的,但并不为0。所以,用于生成GUID的算法通常都加入了非随机的参数(如时间),以保证这种重复的情况不会发生。
在 Windows 平台上,GUID 广泛应用于微软的产品中,用于标识如注册表项、类及接口标识、数据库、系统目录等对象。

2格式

GUID 的格式为“xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx”,其中每个 x 是 0-9 或 a-f 范围内的一个32位十六进制数。例如:6F9619FF-8B86-D011-B42D-00C04FC964FF 即为有效的 GUID 值。

3特点

★需要GUID的时候,可以完全由算法自动生成,不需要一个权威机构来管理。
★GUID的长度固定,并且相对而言较短小,非常适合于排序、标识和存储。

4争议

由于GUID值产生的潜在弊端已经引起了保密性提倡者的关注。1999年3月,美国联邦商务委员会接到要求,对微软的GUID值使用进行调查。争议主要涉及Office 97和Office 2000文档对GUID值的使用。Office文档,如Word文件或Excel电子数据表,所使用的GUID值对用户是不可见的。但有很多报道宣称,文档的作者是可以通过GUID值的跟踪查到的,即使作者已经采用特殊方法,他们还是可以被追踪到。 为了回应上述问题,微软已经发布了一个Office 97修补版SR2,它禁止了GUID功能的使用,并且还可以将现存文档的GUID去除。
潜在的对Intel处理器序列号滥用的问题与上述GUID值的问题本质是一样的。
硬盘分区表类型
 
先说说目前广泛使用的磁盘分区表方案。传统的分区方案(称为MBR分区方案)是将分区信息保存到磁盘的第一个扇区(MBR扇区)中的64个字节中,每个分区项占用16个字节,这16个字节中存有活动状态标志、文件系统标识、起止柱面号、磁头号、扇区号、隐含扇区数目(4个字节)、分区总扇区数目(4个字节)等内容。由于MBR扇区只有64个字节用于分区表,所以只能记录4个分区的信息。这就是硬盘主分区数目不能超过4个的原因。后来为了支持更多的分区,引入了扩展分区及逻辑分区的概念。但每个分区项仍用16个字节存储。
MBR分区方案不是用得好好的吗?为什么要提出新的方案呢?那就让我们看看MBR分区方案有什么问题。前面已经提到了主分区数目不能超过4个的限制,这是其一,很多时候,4个主分区并不能满足需要。另外最关键的是MBR分区方案无法支持超过2TB容量的磁盘。因为这一方案用4个字节存储分区的总扇区数,最大能表示2的32次方的扇区个数,按每扇区512字节计算,每个分区最大不能超过2TB。磁盘容量超过2TB以后,分区的起始位置也就无法表示了。在硬盘容量突飞猛进的今天,2TB的限制将很快被突破。由此可见,MBR分区方案已经无法满足需要了。下面介绍GUID分区表方案。
GUID分区表(简称GPT。使用GUID分区表的磁盘称为GPT磁盘)是源自EFI标准的一种较新的磁盘分区表结构的标准。与目前普遍使用的主引导记录(MBR)分区方案相比,GPT提供了更加灵活的磁盘分区机制。它具有如下优点:
  1、支持2TB以上的大硬盘。
  2、每个磁盘的分区个数几乎没有限制。为什么说“几乎”呢?是因为Windows系统最多只允许划分128个分区。不过也完全够用了。
  3、分区大小几乎没有限制。又是一个“几乎”。因为它用64位的整数表示扇区号。夸张一点说,一个64位整数能代表的分区大小已经是个“天文数字”了,若干年内你都无法见到这样大小的硬盘,更不用说分区了。
  4、分区表自带备份。在磁盘的首尾部分分别保存了一份相同的分区表。其中一份被破坏后,可以通过另一份恢复。
  5、每个分区可以有一个名称(不同于卷标)。
既然GUID分区方案具有如此多的优点,在分区时是不是可以全部采用这种方案呢?不是的。并不是所有的Windows系统都支持这种分区方案。 请看下表:
Windows种类
能否读写GPT磁盘
能否从GPT磁盘启动
32位 Windows XP
不能。只能看到一个Protective MBR分区
不支持
Windows 2000/NT/9x
不能。只能看到一个Protective MBR分区
不支持
64位 Windows XP
只有基于Itanium的系统才能从GPT磁盘启动
Windows Server 2003 SP1及以上版本
只有基于Itanium的系统才能从GPT磁盘启动
Windows Vista
只有基于 EFI 的系统支持从GPT磁盘启动
Windows Server 2008
只有基于 EFI 的系统支持从GPT磁盘启动
Windows 7
只有基于 EFI 的系统支持从GPT磁盘启动
Windows 8/8.1 只有基于 EFI 的系统支持从GPT磁盘启动
是不是很失望?目前多数的个人电脑系统还无法完美支持GPT磁盘。但是这并不意味着我们不需要了解GUID分区方案。别忘了,硬件的发展速度总是令人吃惊的。1.5TB的硬盘已经大量上市,2TB以上容量的硬盘很快就会普及,基于EFI的主板也正在销售。GUID分区方案终将成为主流 。
做为一款分区软件,DiskGenius从3.1版本开始支持GUID分区表。这是国内第一款支持GUID分区表的分区软件。DiskGenius提供了GUID分区的建立、删除、格式化、已丢失分区恢复、文件恢复、分区表备份、GUID分区表格式与MBR分区表格式之间的相互转换(无误转换)等功能。
 
 

注:UUID-Universally Unique IDentifiers全局唯一标识符

一、Linux磁盘分区UUID的获取方法

1、[san@localhost ~]$ ls -l /dev/disk/by-uuid/ 
total 0
lrwxrwxrwx 1 root root 10 2010-01-18 02:18 0733f5c1-cb85-4f98-9d4f-122cfcee9806 -> ../../sdc1

lrwxrwxrwx 1 root root 10 2010-01-18 01:13 3754-1BDB -> ../../sda5 
lrwxrwxrwx 1 root root 10 2010-01-18 01:13 41a18221-6b1f-4ca2-9bc3-dc353c87d932 -> ../../sda9 
lrwxrwxrwx 1 root root 10 2010-01-18 01:13 57183ff9-d4a5-4623-a47f-f8f17339be03 -> ../../sda7 
lrwxrwxrwx 1 root root 10 2010-01-18 01:13 6bdf487f-cad7-4197-b0d9-4ddc6df1de2d -> ../../sda8 
lrwxrwxrwx 1 root root 10 2010-01-18 01:13 ae6dcc02-3f7f-47cc-8a6e-e29218b4d345 -> ../../sda6 
lrwxrwxrwx 1 root root 10 2010-01-18 01:13 CC47-2A04 -> ../../sda1

lrwxrwxrwx 1 root root 10 2010-01-18 02:18 d2154d3e-3006-4a05-a134-f721145f1670 -> ../../sdc2 
lrwxrwxrwx 1 root root 10 2010-01-18 02:18 df974270-dbba-4f87-8121-427636dab396 -> ../../sdc3

lrwxrwxrwx 1 root root 10 2010-01-18 01:52 f535fef8-f392-4c84-8e7a-85915d9179fb -> ../../sdb1

 

VS:

lrwxrwxrwx 1 root root 10 2010-01-18 02:25 0733f5c1-cb85-4f98-9d4f-122cfcee9806 -> ../../sdb1
lrwxrwxrwx 1 root root 10 2010-01-18 01:13 3754-1BDB -> ../../sda5
lrwxrwxrwx 1 root root 10 2010-01-18 01:13 41a18221-6b1f-4ca2-9bc3-dc353c87d932 -> ../../sda9
lrwxrwxrwx 1 root root 10 2010-01-18 01:13 57183ff9-d4a5-4623-a47f-f8f17339be03 -> ../../sda7
lrwxrwxrwx 1 root root 10 2010-01-18 01:13 6bdf487f-cad7-4197-b0d9-4ddc6df1de2d -> ../../sda8
lrwxrwxrwx 1 root root 10 2010-01-18 01:13 ae6dcc02-3f7f-47cc-8a6e-e29218b4d345 -> ../../sda6
lrwxrwxrwx 1 root root 10 2010-01-18 01:13 CC47-2A04 -> ../../sda1
lrwxrwxrwx 1 root root 10 2010-01-18 02:25 d2154d3e-3006-4a05-a134-f721145f1670 -> ../../sdb2
lrwxrwxrwx 1 root root 10 2010-01-18 02:25 df974270-dbba-4f87-8121-427636dab396 -> ../../sdb3

 

2、通过blkid命令

 

[san@localhost ~]$ blkid /dev/sdb1 
/dev/sdb1: LABEL="SAN " UUID="f535fef8-f392-4c84-8e7a-85915d9179fb " TYPE="ext3 "

[san@localhost ~]$ blkid /dev/sda6

/dev/sda6: LABEL="/" UUID="ae6dcc02-3f7f-47cc-8a6e-e29218b4d345 " TYPE="ext3" SEC_TYPE="ext2 "

二、Linux UUID的作用及意义

 原因1:它是真正的唯一标志符

UUID为系统中的存储设备提供唯一的标识字符串,不管这个设备是什么类型的。如果你在系统中添加了新的存储设备如硬盘,很可能会造成一些麻烦,比如说启动的时候因为找不到设备而失败,而使用UUID则不会有这样的问题。

原因2:设备名并非总是不变的

自动分配的设备名称并非总是一致的,它们依赖于启动时内核加载模块的顺序。如果你在插入了USB盘时启动了系统,而下次启动时又把它拔掉了,就有可能导致设备名分配不一致。

使用UUID对于挂载移动设备也非常有好处──例如我有一个24合一的读卡器,它支持各种各样的卡,而使用UUID总可以使同一块卡挂载在同一个地方。

原因3:ubuntu中的许多关键功能现在开始依赖于UUID

例如grub──系统引导程序,现在可以识别UUID,打开你的/boot/grub/menu.lst,你可以看到类似如下的语句:
title Ubuntu hardy (development branch), kernel 2.6.24-16-generic
root (hd2,0)
kernel /boot/vmlinuz-2.6.24-16-generic root=UUID=c73a37c8-ef7f-40e4-b9de-8b2f81038441 ro quiet splash
initrd /boot/initrd.img-2.6.24-16-generic
quiet

http://en.wikipedia.org/wiki/GUID_Partition_Table

GUID Partition Table

From Wikipedia, the free encyclopedia
 
 

GUID Partition Table (GPT) is a standard for the layout of the partition table on a physical hard disk, using globally unique identifiers (GUID). Although it forms a part of the Unified Extensible Firmware Interface (UEFI) standard (Unified EFI Forum proposed replacement for the PC BIOS), it is also used on some BIOS systems because of the limitations of MBR partition tables, which use 32 bits for storinglogical block addresses and size information.

As of 2010, most current operating systems support GPT. Some, including OS X and Microsoft Windows, only support booting from GPT partitions on systems with EFI firmware, but FreeBSD and most Linux distributions can boot from GPT partitions on systems with either legacy BIOS firmware interface or EFI.

Diagram illustrating the layout of GPT scheme. In this example, each logical block (LBA) is 512 bytes in size, and each partition entry is 128 bytes, and the corresponding partition entries are assumed to be located in LBA 2-33, here. LBA addresses that are negative indicate position from the end of the volume, with −1 as the last addressable block.

History[edit]

Main articles: INT 13H and Enhanced BIOS

The widespread MBR partitioning scheme, dating from the early 1980s, imposed limitations which affect the use of modern hardware. One of the main limitations is the usage of 32 bits for storing logical block addresses and size information.

For hard disks with 512-byte sectors, the MBR partition table entries allow up to a maximum of 2 TiB (232×512 Bytes)).[1] GPT allocates 64 bits for logical block addresses and therefore allows a maximum partition size of 264−1 sectors. For disks with 512-byte sectors, that would be 9.4 ZB (9.4 × 1021 bytes) or 8 ZiB−512 bytes (9,444,732,965,739,290,426,880 bytes or 18,446,744,073,709,551,615 (264−1) sectors × 512 (29) bytes per sector).[1][2]

Intel therefore developed a new partition-table format in the late 1990s as part of what eventually became UEFI. The GPT as of 2010 forms a subset of the UEFI specification.[3]

Features[edit]

This section needs additional citations for verification. Please help improve this article by adding citations to reliable sources. Unsourced material may be challenged and removed. (July 2013)

MBR-based partition table schemes insert the partitioning information for (usually) four "primary" partitions in the master boot record (MBR) (which on a BIOS system is also the container for code that begins the process of booting the system). In a GPT, the first sector of the disk is reserved for a "protective MBR" such that booting a BIOS-based computer from a GPT disk is supported, but the boot loader and O/S must both be GPT aware. Regardless of the sector size, the GPT header begins on the second logical block of the device.

Like modern MBRs, GPTs use logical block addressing (LBA) in place of the historical cylinder-head-sector (CHS) addressing. The protective MBR is contained in LBA 0, the GPT header is in LBA 1, and the GPT header has a pointer to the partition table, or Partition Entry Array, typically LBA 2. The UEFI specification[4] stipulates that a minimum of 16,384 bytes, regardless of sector size, be allocated for the Partition Entry Array. On a disk having 512-byte sectors, a partition entry array size of 16,384 bytes and the minimum size of 128 bytes for each partition entry, LBA 34 is the first usable sector on the disk.

Hard disk manufacturers are transitioning to 4,096-byte sectors. As of 2010, the first such drives continue to present 512-byte physical sectors to the OS, so degraded performance can result when the drive's (hidden) internal 4 KiB sector boundaries do not coincide with the 4 KiB logical blocks, clusters and virtual memory pages common in many operating systems and file systems. This is a particular problem on writes when the drive is forced to perform two read-modify-write operations to satisfy a single misaligned 4 KiB write operation.[5] Such a misalignment occurs by default if the first partition is placed immediately after the GPT, as the next block is LBA 34, whereas the next 4 KiB boundary begins with LBA 40.

For backward compatibility with most operating systems before Windows Vista, including DOS, OS/2 and Windows, MBR partitions must always start on track boundaries according to the traditional CHS addressing scheme and end on a cylinder boundary. This even holds true for partitions with emulated CHS geometries (as reflected by the BIOS and the CHS sectors entries in the MBR partition table) or partitions accessed only via LBA. Extended partitions always start on cylinder boundaries as well.

This typically causes the first primary partition to start at LBA 63 on disks accessed via LBA, leaving a gap of 62 sectors with MBR-based disks, sometimes called "MBR gap", "boot track", or "embedding area". That otherwise unused disk space is commonly used by boot loaders such as GRUB for storing their second stages.[6] On older computers using alternative LBA/CHS translation schemes or different extended CHS mappings, with smaller LBA-accessed disks, or on disks accessed via CHS only, the value could be even smaller, although not normally less than LBA 16 on normal hard disks.

Since Windows Vista, the first partition usually starts after a gap of 2,047 sectors at LBA 2,048 as part of its new 1 MiB partition alignment policy, so no large-sector misalignment occurs by default, but serious compatibility problems with older operating systems and disk tools exist.

Drives which boot Intel-based Macs are typically formatted with a GPT, rather than with the Apple Partition Map (APM).

GPT also provides redundancy, writing the GPT header and partition table both at the beginning and at the end of the disk.

If the minimum size of 16,384 bytes is allocated for the partition entry array, and the default size of 128 bytes is used for each partition entry, then the maximum number of partitions is limited to 128.

Legacy MBR (LBA 0)[edit]

Traditionally, in IBM PC compatible systems the first sector of the disk holds the Master Boot Record (MBR), containing the drive's partitioning information and the code of the first stage boot loader for BIOS-based systems. For limited backward compatibility, this sector is still reserved for a MBR in the GPT specification, but it is now used in a way that prevents MBR-based disk utilities from mis-recognizing, and possibly over-writing, GPT disks. This is referred to as a "protective MBR".

A single partition type of EEh, encompassing the entire GPT drive (where "entire" actually means as much of the drive as can be represented in an MBR), is indicated and identifies it as GPT. Operating systems and tools which cannot read GPT disks will generally recognize the disk as containing one partition of unknown type and no empty space, and will typically refuse to modify the disk unless the user explicitly requests and confirms the deletion of this partition. This minimizes accidental erasures. Furthermore, GPT-aware OSes will check the protective MBR and if the enclosed partition type is not of typeEEh or if there are multiple partitions defined on the target device, the device should not be manipulated[dubious – discuss].

While the MBR layout (and also the protective MBR layout) was defined around a sector size of 512 bytes per sector, the actual sector size can be larger on various media such as MO disks or hard disks with Advanced Format. Extra space in the MBR typically remains unused.

If the actual size of the disk exceeds the maximum partition size representable using the legacy 32-bit LBA entries in the MBR partition table, the recorded size of this partition is clipped at the maximum, thereby ignoring the rest of disk. This amounts to a maximum reported size of 2 TiB, assuming a disk with 512 bytes per sector (see 512e). It would result in 16 TiB with 4 KB sectors (4Kn), but since many older operating systems and tools are hard-wired for a sector size of 512 bytes or are limited to 32-bit calculations, exceeding the 2 TiB limit would cause serious compatibility problems.

In operating systems that support GPT-based boot through BIOS services rather than EFI, the first sector is also still used to store the first stage of the bootloader code, but modified to recognize GPT partitions. The boot loader in the MBR must not assume a fixed sector size of 512 bytes / sector.

Apple's Boot Camp Intel based Apple macs software creates a hybrid partition table to allow the booting of Windows (which at the time of Boot Camp's creation did not support GPT or EFI). In this system the protective partition is reduced in size to cover from sector 1 to the sector before the first regular partition included in the hybrid MBR. Additional MBR partitions are then defined to correspond to the next three GPT[citation needed] partitions.

Partition table header (LBA 1)[edit]

The partition table header defines the usable blocks on the disk. It also defines the number and size of the partition entries that make up the partition table. The EFI stipulates a minimum of 16,384 bytes be reserved for the partition table array, so there are 128 partition entries reserved, each 128 bytes long.

The header contains the disk globally unique identifier (GUID). It records its own size and location (always LBA 1!) and the size and location of the secondary GPT header and table (always the last sectors on the disk). Importantly, it also contains a CRC32 checksum for itself and for the partition table, which may be verified by the firmware, bootloader and/or operating system on boot. Because of this, hex editors should not be used to modify the contents of the GPT. Such modification would render the checksum invalid. In this case, the primary GPT may be overwritten with the secondary one by disk recovery software. If both GPTs contain invalid checksums, the disk would be unusable, by software that checks the checksum.

GPT header format
Offset Length Contents
+0 8 bytes Signature ("EFI PART", 45h 46h 49h 20h 50h 41h 52h 54h)
+8 4 bytes Revision (for GPT version 1.0 (through at least UEFI version 2.3.1), the value is 00h 00h 01h 00h)
+12 4 bytes Header size in little endian (in bytes, usually 5Ch 00h 00h 00h meaning 92 bytes)
+16 4 bytes CRC32 of header (offset +0 up to header size), with this field zeroed during calculation
+20 4 bytes Reserved; must be zero
+24 8 bytes Current LBA (location of this header copy)
+32 8 bytes Backup LBA (location of the other header copy)
+40 8 bytes First usable LBA for partitions (primary partition table last LBA + 1)
+48 8 bytes Last usable LBA (secondary partition table first LBA - 1)
+56 16 bytes Disk GUID (also referred as UUID on UNIXes)
+72 8 bytes Starting LBA of array of partition entries (always 2 in primary copy)
+80 4 bytes Number of partition entries in array
+84 4 bytes Size of a single partition entry (usually 128)
+88 4 bytes CRC32 of partition array
+92 * Reserved; must be zeroes for the rest of the block (420 bytes for a sector size of 512 bytes; but can be more with larger sector sizes)
LBA size Total

The values for current and backup LBAs of the primary header should be the second sector of the disk (LBA 1) and the last sector of the disk, respectively. The secondary header at the end of the disk identifies its own table of partition entries, which is located directly before that header.

This table must be referenced relative to LBA 1. This means, that on disks with 4Kn sectors, it does not follow the 512 bytes of the MBR physically (stored in LBA 0) and thereby become part of LBA 0 on disks with larger sector sizes. While the described arrangement happens to occur on disks with 512 bytes per sector, there may be "gaps" of unused space between them on disks with larger sector sizes. If multi-sector reads are performed, the actual sector size must be included in the calculation when referencing this table.

Partition entries[edit]

GUID partition entry format
Offset Length Contents
0 16 bytes Partition type GUID
16 16 bytes Unique partition GUID
32 8 bytes First LBA (little endian)
40 8 bytes Last LBA (inclusive, usually odd)
48 8 bytes Attribute flags (e.g. bit 60 denotes read-only)
56 72 bytes Partition name (36 UTF-16LE code units)
  128 bytes total

The GPT uses simple and straightforward entries to describe partitions. The first 16 bytes designate the partition type globally unique identifier (GUID). For example, the GUID for an EFI System partition is {C12A7328-F81F-11D2-BA4B-00A0C93EC93B}. The second 16 bytes contain a GUID unique to the partition. Then follow the starting and ending 64-bit LBAs, partition attributes and partition names. As is the nature and purpose of GUIDs, no central registry is needed to ensure the uniqueness of the GUID partition type designators. The location of the partition entries array on disk is defined in the GPT header.

The GPT header contains a field that specifies the size of a partition table entry. The minimum required is 128 bytes, but implementations must allow for other values (see this warning).

Also, the sector size must not be assumed to be hard-wired to 512 bytes per sector in calculations (see Advanced Format), that is, there can be more than four partition entries in a single sector, and (with possible future much larger partition table entries) it is possible to have a sector hold only a fraction of a partition entry. Except for the first two sectors (LBA 0 and LBA 1), the GPT specification just describes the size and organization of a data structure, not in how many sectors it is stored on disk.

The 64 bits partition table attributes are shared between 48 bits common attributes for all partition types, and 16 bits type-specific attributes.

Partition attributes
Bit Content
0 System partition (disk partitioning utilities must preserve the partition as is)
1 EFI firmware should ignore the content of the partition and not try to read from it
2 Legacy BIOS bootable (equivalent to active flag (typically bit 7 set) at offset +0h in partition entries of the MBR partition table)[7]
3-47 Reserved for future use
48-63 Defined and used by the individual partition type

Microsoft defines the type-specific attributes for Basic data partition according to this TechNet article as:

Basic data partition attributes
Bit Content
60 Read-only
62 Hidden
63 Do not automount (i.e., do not assign drive letter)

Operating System support of GPT[edit]

Hybrid MBRs are non-standard and can be interpreted in different ways by different OSes.[8] Unless otherwise noted, OSes provide precedence to the GPT data when a hybrid MBR configuration is encountered.

The term No native support on this arch and version. should be understood this way:

Not supported as data disk,[9] only known legacy partitions found in protective MBR are accessible via the OS. Detachable disks: only support for MBR partitioning; No access with end user applications. GPT contained raw data is accessible with third-party administrator tools for low level disk access. True file system level support in read or read-write form might be subject of software from a third-party vendor.

UNIX and UNIX-like operating systems[edit]

Details of GPT support on UNIX and Unix-like operating systems
OS family Version or edition Platform Read and write support Boot support Note
FreeBSD Since 7.0 IA-32x86-64 Yes Yes In a hybrid configuration, both GPT and MBR partition identifiers may be used.
Linux Most of the x86 Linux distributions

Fedora 8+ and Ubuntu 8.04+[10]

IA-32x86-64 Yes Yes New tools such as gdisk,[11] GNU Parted,[12][13] util-linux v2.23+ fdisk,[14][15] SyslinuxGRUB 0.96 + patchesand GRUB 2 have been GPT-enabled.
Mac OS X Since 10.4.0 (some features since 10.4.6)[16] IA-32x86-64 Yes Yes Only Intel Macintosh computers can boot from GPT.
MidnightBSD Since 0.4-CURRENT IA-32x86-64 Yes Requires BIOS In a hybrid configuration, both GPT and MBR partition identifiers may be used.
Solaris Since Solaris 10 IA-32x86-64,SPARC Yes Yes [17]
HP-UX Since HP-UX 11.20 IA-64 Yes Yes [18]

Windows: 32-bit versions[edit]

Microsoft does not support EFI on 32-bit platforms, and therefore does not allow booting from GPT partitions.

Details of GPT support on 32-bit editions of Microsoft Windows[9]
OS version Release date Platform Read or write support Boot support Note
Windows XP 2001-10-25 IA-32 No No  
Windows Server 2003 2003-04-24 IA-32 No No  
Windows Server 2003 SP1 2005-03-30 IA-32 Yes No MBR takes precedence in hybrid configuration[8]
Windows Vista 2006-07-22 IA-32 Yes No MBR takes precedence in hybrid configuration[8]
Windows Server 2008 2008-02-27 IA-32 Yes No MBR takes precedence in hybrid configuration[8]
Windows 7 2009-10-22 IA-32 Yes No MBR takes precedence in hybrid configuration[8]
Windows 8 2012-08-01 IA-32 Yes No MBR takes precedence in hybrid configuration[8]

Windows: 64-bit versions[edit]

Details of GPT support on 64-bit editions of Microsoft Windows[9]
OS version Release date Platform Read and write support Boot support Note
Windows XP Professional x64 Edition
Windows Server 2003
2005-04-25[19] x64 Yes No MBR takes precedence in hybrid MBR configuration[8]
Windows Server 2003 2005-04-25 IA-64 Yes Yes MBR takes precedence in hybrid MBR configuration[8]
Windows Vista 2006-07-22 x64 Yes Requires UEFI MBR takes precedence in hybrid configuration[8]
Windows Server 2008 2008-02-27 x64 Yes Requires UEFI MBR takes precedence in hybrid configuration[8]
Windows Server 2008 2008-02-27 IA-64 Yes Yes MBR takes precedence in hybrid configuration[8]
Windows 7
Windows Server 2008 R2
2009-10-22 x64 Yes Requires UEFI MBR takes precedence in hybrid configuration.[8]
Windows Server 2008 R2 2009-10-22 IA-64 Yes Yes MBR takes precedence in hybrid configuration[8]
Windows 8
Windows Server 2012
2012-08-01 x64 Yes Requires UEFI MBR takes precedence in hybrid configuration.[8]

Partition type GUIDs[edit]

Assoc. OS Partition type Globally unique identifier (GUID)[A]
(None) Unused entry 00000000-0000-0000-0000-000000000000
MBR partition scheme 024DEE41-33E7-11D3-9D69-0008C781F39F
EFI System partition C12A7328-F81F-11D2-BA4B-00A0C93EC93B
BIOS Boot partition[B] 21686148-6449-6E6F-744E-656564454649
Intel Fast Flash (iFFS) partition (for Intel Rapid Start technology)[20][21] D3BFE2DE-3DAF-11DF-BA40-E3A556D89593
Windows Microsoft Reserved Partition (MSR) E3C9E316-0B5C-4DB8-817D-F92DF00215AE
Basic data partition[C] EBD0A0A2-B9E5-4433-87C0-68B6B72699C7
Logical Disk Manager (LDM) metadata partition 5808C8AA-7E8F-42E0-85D2-E1E90434CFB3
Logical Disk Manager data partition AF9B60A0-1431-4F62-BC68-3311714A69AD
Windows Recovery Environment DE94BBA4-06D1-4D40-A16A-BFD50179D6AC
IBM General Parallel File System (GPFS) partition 37AFFC90-EF7D-4E96-91C3-2D7AE055B174
HP-UX Data partition 75894C1E-3AEB-11D3-B7C1-7B03A0000000
Service Partition E2A1E728-32E3-11D6-A682-7B03A0000000
Linux Linux filesystem data[C] 0FC63DAF-8483-4772-8E79-3D69D8477DE4
RAID partition A19D880F-05FC-4D3B-A006-743F0F84911E
Swap partition 0657FD6D-A4AB-43C4-84E5-0933C84B4F4F
Logical Volume Manager (LVM) partition E6D6D379-F507-44C2-A23C-238F2A3DF928
/home partition 933AC7E1-2EB4-4F13-B844-0E14E2AEF915
Reserved 8DA63339-0007-60C0-C436-083AC8230908
FreeBSD Boot partition 83BD6B9D-7F41-11DC-BE0B-001560B84F0F
Data partition 516E7CB4-6ECF-11D6-8FF8-00022D09712B
Swap partition 516E7CB5-6ECF-11D6-8FF8-00022D09712B
Unix File System (UFS) partition 516E7CB6-6ECF-11D6-8FF8-00022D09712B
Vinum volume manager partition 516E7CB8-6ECF-11D6-8FF8-00022D09712B
ZFS partition 516E7CBA-6ECF-11D6-8FF8-00022D09712B
Mac OS X Hierarchical File System Plus (HFS+) partition 48465300-0000-11AA-AA11-00306543ECAC
Apple UFS 55465300-0000-11AA-AA11-00306543ECAC
ZFS[D] 6A898CC3-1DD2-11B2-99A6-080020736631
Apple RAID partition 52414944-0000-11AA-AA11-00306543ECAC
Apple RAID partition, offline 52414944-5F4F-11AA-AA11-00306543ECAC
Apple Boot partition 426F6F74-0000-11AA-AA11-00306543ECAC
Apple Label 4C616265-6C00-11AA-AA11-00306543ECAC
Apple TV Recovery partition 5265636F-7665-11AA-AA11-00306543ECAC
Apple Core Storage (i.e. Lion FileVault) partition 53746F72-6167-11AA-AA11-00306543ECAC
Solaris Boot partition 6A82CB45-1DD2-11B2-99A6-080020736631
Root partition 6A85CF4D-1DD2-11B2-99A6-080020736631
Swap partition 6A87C46F-1DD2-11B2-99A6-080020736631
Backup partition 6A8B642B-1DD2-11B2-99A6-080020736631
/usr partition[D] 6A898CC3-1DD2-11B2-99A6-080020736631
/var partition 6A8EF2E9-1DD2-11B2-99A6-080020736631
/home partition 6A90BA39-1DD2-11B2-99A6-080020736631
Alternate sector 6A9283A5-1DD2-11B2-99A6-080020736631
Reserved partition 6A945A3B-1DD2-11B2-99A6-080020736631
6A9630D1-1DD2-11B2-99A6-080020736631
6A980767-1DD2-11B2-99A6-080020736631
6A96237F-1DD2-11B2-99A6-080020736631
6A8D2AC7-1DD2-11B2-99A6-080020736631
NetBSD[E][22] Swap partition 49F48D32-B10E-11DC-B99B-0019D1879648
FFS partition 49F48D5A-B10E-11DC-B99B-0019D1879648
LFS partition 49F48D82-B10E-11DC-B99B-0019D1879648
RAID partition 49F48DAA-B10E-11DC-B99B-0019D1879648
Concatenated partition 2DB519C4-B10F-11DC-B99B-0019D1879648
Encrypted partition 2DB519EC-B10F-11DC-B99B-0019D1879648
ChromeOS[23] ChromeOS kernel FE3A2A5D-4F32-41A7-B725-ACCC3285A309
ChromeOS rootfs 3CB8E202-3B7E-47DD-8A3C-7FF2A13CFCEC
ChromeOS future use 2E0A753D-9E48-43B0-8337-B15192CB1B5E
Haiku[24] Haiku BFS 42465331-3BA3-10F1-802A-4861696B7521
MidnightBSD[E][25] Boot partition 85D5E45E-237C-11E1-B4B3-E89A8F7FC3A7
Data partition 85D5E45A-237C-11E1-B4B3-E89A8F7FC3A7
Swap partition 85D5E45B-237C-11E1-B4B3-E89A8F7FC3A7
Unix File System (UFS) partition 0394EF8B-237E-11E1-B4B3-E89A8F7FC3A7
Vinum volume manager partition 85D5E45C-237C-11E1-B4B3-E89A8F7FC3A7
ZFS partition 85D5E45D-237C-11E1-B4B3-E89A8F7FC3A7
A. ^ The GUIDs in this table are written assuming a little-endian byte order. For example, the GUID for an EFI System partition is written as {C12A7328-F81F-11D2-BA4B-00A0C93EC93B} here, which corresponds to the 16 byte sequence 28h 73h 2Ah C1h 1Fh F8h D2h 11h BAh 4Bh 00h A0h C9h 3Eh C9h 3Bh — only the first three blocks are byte-swapped.
B. ^ The formation of this GUID does not follow the GUID definition; it is formed by using the ASCII codes for the string "Hah!IdontNeedEFI". Such formation of "GUID" value breaks down the guaranteed uniqueness of GUID.
C. a b Previously Linux used the same GUID for the data partitions as Windows (Basic data partition: {EBD0A0A2-B9E5-4433-87C0-68B6B72699C7}). Linux never had a separate unique partition type GUID defined for its data partitions. This created problems when dual-booting Linux and Windows in UEFI-GPT setup. The new GUID (Linux filesystem data: {0FC63DAF-8483-4772-8E79-3D69D8477DE4}) was defined jointly by GPT fdisk and GNU Parted developers. It is identified as type code 0x8300 in GPT fdisk. (See definitions in gdisk's parttypes.cc)
D. a b The GUID for /usr on Solaris is used as a generic GUID for ZFS by Mac OS X.
E. a b NetBSD and MidnightBSD had used the FreeBSD GUIDs before their unique GUIDs were created.
 
 
 

UUID GUID的更多相关文章

  1. PostgreSQL 下生成 UUID(Guid)

    最近在Windows 10 下安装了 PostgreSQL(postgresql-9.6.3-1-windows.exe),在学习过程中,发现PostgreSQL 支持UUID(Guid)类型,但是却 ...

  2. Windows 10 下 PostgreSQL 生成 UUID(Guid)

    最近在Windows 10 下安装了 PostgreSQL(postgresql-9.6.3-1-windows.exe),在学习过程中,发现PostgreSQL 支持UUID(Guid)类型,但是却 ...

  3. 随机生成UUID(GUID)的方法

    - (NSString *)UUID { CFUUIDRef uuid_ref = CFUUIDCreate(NULL); CFStringRef uuid_string_ref= CFUUIDCre ...

  4. c++ 创建 uuid guid

    如果没安装,先安装: [root@localhost]# yum install libuuid-devel #include "uuid/uuid.h" 引用 libuuid.s ...

  5. javascript guid(uuid)

    http://stackoverflow.com/questions/105034/how-to-create-a-guid-uuid-in-javascript function generateU ...

  6. UUID(即GUID)

    UUID(GUID)是Cocoa Touch提供的一种生成唯一标识的机制.类型为CFUUIDRef的对象可以代表UUID,UUID是基于当前时间.计数器和硬件标识(通常是以太网卡的MAC地址)等数据计 ...

  7. python使用uuid库生成唯一id

    概述: UUID是128位的全局唯一标识符,通常由32字节的字符串表示. 它可以保证时间和空间的唯一性,也称为GUID,全称为: UUID -- Universally Unique IDentifi ...

  8. C# 中GUID生成格式的四种格式

    var uuid = Guid.NewGuid().ToString(); // 9af7f46a-ea52-4aa3-b8c3-9fd484c2af12 var uuidN = Guid.NewGu ...

  9. Guid算法与标识列(自动增长字段)在表中的应用

    <<1>>int(bigint)+标识列(自动增长字段) 用标识列实现字段自增可以避免并发等问题.不需开发人员自己控制自增,用标识列的字段在Insert的时候不用指定主键的值. ...

随机推荐

  1. Repo command reference

    Repo command reference In this document init sync upload diff download forall prune start status Rep ...

  2. Java-字符串大小写转换

    package com.tj; public class MyClass implements Cloneable { public static void main(String[] args) { ...

  3. 九度oj 题目1130:日志排序

    题目描述: 有一个网络日志,记录了网络中计算任务的执行情况,每个计算任务对应一条如下形式的日志记录:“hs_10000_p”是计算任务的名称,“2007-01-17 19:22:53,315”是计算任 ...

  4. [UOJ#130][BZOJ4198][Noi2015]荷马史诗

    [UOJ#130][BZOJ4198][Noi2015]荷马史诗 试题描述 追逐影子的人,自己就是影子. ——荷马 Allison 最近迷上了文学.她喜欢在一个慵懒的午后,细细地品上一杯卡布奇诺,静静 ...

  5. [luoguP3413] SAC#1 - 萌数(数位DP)

    传送门 gtm的数位dp! 看到好多题解,都是记忆化搜索,好像非常方便啊,但是我还是用递推好了,毕竟还是有些类似数位dp的题用递推的思路,记忆化做不了,现在多培养一下思路 首先这道题, 只看长度大于等 ...

  6. 【bzoj2393】Cirno的完美算数教室 数论容斥

    Description ~Cirno发现了一种baka数,这种数呢~只含有2和⑨两种数字~~ 现在Cirno想知道~一个区间中~~有多少个数能被baka数整除~ 但是Cirno这么天才的妖精才不屑去数 ...

  7. 软件包管理器(bzoj 4196)

    Description Linux用户和OSX用户一定对软件包管理器不会陌生.通过软件包管理器,你可以通过一行命令安装某一个软件包,然后软件包管理器会帮助你从软件源下载软件包,同时自动解决所有的依赖( ...

  8. Vxlan简介

    1.为什么需要Vxlan 1.什么是VXLAN VXLAN(Virtual eXtensible LAN可扩展虚拟局域网),是一种mac in UDP技术.传统的二层帧被封装到了UDP的报文中,通过U ...

  9. ElasticSearch索引自定义类型

    ES可以自动检测字段并设置映射类型.如果设置的索引类型不是我们所需要的,我们可以自行定义. Rest API设置自定义索引 首先通过ES自动映射一个IP地址的字段的类型: <pre name=& ...

  10. hihocoder #1290 : Demo Day

    传送门 #1290 : Demo Day 时间限制:10000ms 单点时限:1000ms 内存限制:256MB 描述 You work as an intern at a robotics star ...