为啥Linux的fdisk分区时First Sector是63或2048？

Posted 2023-03-28

To understand why not 1, I need to describe some PC history. A long time ago, when the PC was first provided with hard drives, the first block on the drive was used by the Master Boot Record. Thats still true today, the first partition always began on Cylinder 0, Head 1, Sector 0, under the Cylinder/Head/Sector addressing scheme, whic is all there was than. To add to the interest, hard drives actually had differing numbers of sectors per track then and you had to describe the real geometry of the drive to the Bios. There was no auto dectect or Logical Block Addressing as used today. The Cylinder/Head/Sector addressing scheme broke permently when hard drives reached 4Gb but its legacy lives on. The largest numbers that fit in the CHS addressing scheme is 255 heads and 63 sectors per track. Hence until recently, the first partition always started at sector 63. Thats head 1, sector 0. //CHS方式寻址时，扇区号是从1开始的。但是实际的物理扇区是从0开始的，对于LBA寻址方式来说扇区号也是从0开始。这里出现的secotr 0是指物理扇区号。Over the years, boot loaders have used some of the 'unallocated' space between the MBR and the start of the first partition. With the death of the legacy BIOS (ok, its not quite dead yet) and its replacement with EFI BIOS, a special boot partitionis needed to allow EFI systems to boot in EFI mode. Starting the first partition at sector 2048 leaves 1Mb for the EFI boot code. Modern partitioning tools do this anyway and fdisk has been updated to follow suit. You can force some tools to give you back most of this 1Mb but its usually a bad idea. Your 2Tb drive will use 4k physical sectors. Its very important that you create partitions that are aligned on 4kb boundaries or your read/write speeds will be very poor. That is, partition starts must be a multiple of eight sectors. Notice that 63 is not a multiple of 8 but that 2048 is. As you really need to leave some space for grub, you should not start your first partition before sector 64. If you know how big grub is, you can make that smaller but its not worth the effort. 参考技术A To understand why not 1, I need to describe some PC history. A long time ago, when the PC was first provided with hard drives, the first block on the drive was used by the Master Boot Record. Thats still true today, the first partition always began on Cylinder 0, Head 1, Sector 0, under the Cylinder/Head/Sector addressing scheme, whic is all there was than. To add to the interest, hard drives actually had differing numbers of sectors per track then and you had to describe the real geometry of the drive to the BIOS. There was no auto dectect or Logical Block Addressing as used today. The Cylinder/Head/Sector addressing scheme broke permently when hard drives reached 4Gb but its legacy lives on. The largest numbers that fit in the CHS addressing scheme is 255 heads and 63 sectors per track. Hence until recently, the first partition always started at sector 63. Thats head 1, sector 0. //CHS方式寻址时，扇区号是从1开始的。但是实际的物理扇区是从0开始的，对于LBA寻址方式来说扇区号也是从0开始。这里出现的secotr 0是指物理扇区号。Over the years, boot loaders have used some of the 'unallocated' space between the MBR and the start of the first partition. With the death of the legacy BIOS (ok, its not quite dead yet) and its replacement with EFI BIOS, a special boot partitionis needed to allow EFI systems to boot in EFI mode. Starting the first partition at sector 2048 leaves 1Mb for the EFI boot code. Modern partitioning tools do this anyway and fdisk has been updated to follow suit. You can force some tools to give you back most of this 1Mb but its usually a bad idea. Your 2Tb drive will use 4k physical sectors. Its very important that you create partitions that are aligned on 4kb boundaries or your read/write speeds will be very poor. That is, partition starts must be a multiple of eight sectors. Notice that 63 is not a multiple of 8 but that 2048 is. As you really need to leave some space for grub, you should not start your first partition before sector 64. If you know how big grub is, you can make that smaller but its not worth the effort.

Linux文件系统管理 fdisk分区命令

概述

我们在安装操作系统的过程中已经对系统硬盘进行了分区，但是如果我新添加了一块硬盘，想要正常使用时，在Linux中有专门的分区命令 fdisk 和 parted。其中 fdisk 命令较为常用，但不支持大于 2TB 的分区；如果需要支持大于 2TB 的分区，则需要使用 parted 命令，当然 parted 命令也能分配较小的分区。我们先来看看如何使用 fdisk 命令进行分区。

fdisk命令

fdisk 命令的格式如下：

#列出系统分区
[[email protected] ~]# fdisk -l
#给硬盘分区
[[email protected] ~]# fdisk 设备文件名

通过fdisk -l 命令可以得到以下信息

#查询本机可以识别的硬盘和分区
[[email protected] ~]# fdisk -l
#硬盘文件名和硬盘大小
Disk /dev/sda: 21.5 GB, 21474836480 bytes, 41943040 sectors
#每个柱面的大小
Units = sectors of 1 * 512 = 512 bytes
#每个扇区的大小
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disk label type: dos
Disk identifier: 0x0007be9f
Device Boot Start End Blocks ld System
/dev/sda1   *        2048      411647      204800   83  Linux
/dev/sda2          411648    41943039    20765696   8e  Linux LVM

使用"fdisk -l"査看分区信息，能够看到我们添加的硬盘（/dev/sda）的信息。信息的上半部分态是硬盘的整体状态，/dev/sda 硬盘的总大小是 21.5 GB
信息的下半部分是分区的信息，共 7 列，含义如下：

Device	Boot	Start	End	Blocks	id	System
分区的设备文件名	是否为启动引导分区，在这里/dev/sda1为启动引导分区	起始柱面，代表分区从哪里开始	终止柱面，代表分区到哪里结束	分区的大小，单位是 KB	分区内文件系统的 ID。在fdisk 命令中，可以使用"i"查看	分区内安装的系统

如果这个分区并没有占满整块硬盘，就会提示 "Partition 1 does not end on cyl inder boundary"，表示第一个分区没有到硬盘的结束柱面。

假如有一个新的硬盘 /dev/sdb，使用命令给硬盘 /dev/sdb分区，命令如下:

#给/dev/sdb分区
[[email protected] ~]# fdisk /dev/sdb
Device contains neither a valid DOS partition table, nor Sun, SGI or OSF disklabel
Building a new DOS disklabel with disk identifier 0xed7e8bc7.
Changes will remain in memory only, until you decide to write them.
After that, of course, the previous content won‘t be recoverable.
Warning: invalid flag 0x0000 of partition table 4 will be corrected by w(rite)
WARNING: DOS-compatible mode is deprecated.it‘s strongly recommended to switch off the mode (command ‘c‘) and change display units to sectors (command ‘u‘).
Command (m for help):m
#交互界面的等待输入指令的位置，输入 m 得到帮助
Command action
#可用指令
a toggle a bootable flag
b edit bsd disklabel
c toggle the dos compatibility flag
d delete a partition
I list known partition types m print this menu
n add a new partition
o create a new empty DOS partition table
p print the partition table
q quit without saving changes
s create a new empty Sun disklabel
t change a partition‘s system id
u change display/entry units
v verity the partition table
w write table to disk and exit
x extra functionality (experts only)

注意这里的分区命令是"fdisk/dev/sdb"，这是因为我们的硬盘现在并没有分区，使用 fdisk 命令的目的就是建立分区，所以"1~59"这些数字还不存在。
在 fdisk 交互界面中输入m可以得到帮助，帮助里列出了 fdisk 可以识别的交互命令，我们来解释一下这些命令，如下所示。

命令	说 明
a	设置可引导标记
b	编辑 bsd 磁盘标签
c	设置 DOS 操作系统兼容标记
d	删除一个分区
1	显示已知的文件系统类型。82 为 Linux swap 分区，83 为 Linux 分区
m	显示帮助菜单
n	新建分区
0	建立空白 DOS 分区表
P	显示分区列表
q	不保存退出
s	新建空白 SUN 磁盘标签
t	改变一个分区的系统 ID
u	改变显示记录单位
V	验证分区表
w	保存退出
X	附加功能（仅专家）

fdisk命令创建主分区

建立主分区的过程是："fdisk 硬盘名 -> n(新建)->p(建立主分区) -> 1(指定分区号) -> 回车（默认从 1 柱面开始建立分区）-> +5G(指定分区大小)"。此时我们的分区还没有格式化和挂载，所以还不能使用。

[[email protected] ~]# fdisk /dev/sdb
…省略部分输出…
Command (m for help): p
#显示当前硬盘的分区列表
Disk /dev/sdb: 21.5 GB, 21474836480 bytes 255 heads, 63 sectors/track, 2610 cylinders
Units = cylinders of 16065 *512 = 8225280 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disk identifier: 0xb4b0720c
Device Boot Start End Blocks id System
#目前一个分区都没有
Command (m for help): n
#那么我们新建一个分区
Command action
#指定分区类型
e extended
#扩展分区
p primary partition (1-4)
#主分区
p
#这里选择p，建立一个主分区
Partition number (1-4): 1
#选择分区号，范围为1~4，这里选择1
First cylinder (1 -2610, default 1):
#分区的起始柱面，默认从1开始。因为要从硬盘头开始分区，所以直接回车
Using default value 1
#提示使用的是默认值1
Last cylinder, +cylinders or +size{K, M, G}(1-2610, default 2610): +5G
#指定硬盘大小。可以按照柱面指定(1-2610)。我们对柱面不熟悉，那么可以使用size{K, M, G}的方式指定硬盘大小。这里指定+5G，建立一个5GB大小的分区
Command (m for help):
#主分区就建立了，又回到了fdisk交互界面的提示符
Command (m for help): p
#查询一下新建立的分区
Disk /dev/sdb: 21.5GB, 21474836480 bytes
255 heads,63 sectors/track, 2610 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Sector size (logical/physical): 512 bytes 1512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disk identifier: 0xb4b0720c
Device Boot Start End Blocks id System
/dev/sdb1 1 654 5253223+ 83 Linux
#dev/sdb1已经建立

fdisk命令创建扩展分区

把 /dev/sdb 硬盘的所有剩余空间都建立为扩展分区，也就是建立一个主分区，剩余空间都建立成扩展分区，再由扩展分区中建立逻辑分区。

 扩展分区的建立命令如下：

[[email protected] ~]# fdisk /dev/sdb
…省略部分输出…
Command (m for help): n
#新建立分区
Command action
e extended
p primary partition (1-4)
e
#这次建立扩展分区
Partition number (1-4): 2
#给扩展分区指定分区号2
First cylinder (655-2610, default 655):
#扩展分区的起始柱面。上节建立的主分区1已经占用了1~654个柱面，所以我们从655开始建立，注意：如果没有特殊要求，则不要跳开柱面建立分区，应该紧挨着建立分区
Using default value 655
提示使用的是默认值655
Last cylinder, +cylinders or +size{K, M, G} (655-2610, default 2610):
#这里把整块硬盘的剩余空间都建立为扩展分区
Using default value 2610
#提示使用的是默认值2610

fdisk命令创建逻辑分区

扩展分区是不能被格式化和直接使用的，所以还要在扩展分区内部再建立逻辑分区。
逻辑分区的建立过程，命令如下：

[[email protected] ~]# fdisk /dev/sdb
…省略部分输出…
Command (m for help): n
#建立新分区
Command action
l logical (5 or over)
#由于在前面章节中，扩展分区已经建立，所以这里变成了l(logic)
p primary partition (1-4)
l
#建立逻辑分区
First cylinder (655-2610, default 655):
#不用指定分区号，默认会从5开始分配，所以直接选择起始柱面
#注意：逻辑分区是在扩展分区内部再划分的，所以柱面是和扩展分区重叠的
Using default value 655
Last cylinder, +cylinders or +size{K, M, G} (655-2610, default 2610):+2G
#分配2GB大小
Command (m for help): n
#再建立一个逻辑分区
Command action
l logical (5 or over)
p primary partition (1-4)
l
First cylinder (917-2610, default 917):
Using default value 917
Last cylinder, +cylinders or +size{K, M, G} (917-2610, default 2610):+2G
Command (m for help): p
#查看一下已经建立的分区
Disk /dev/sdb: 21.5GB, 21474836480 bytes
255 heads, 63 sectors/track, 2610 cylinders
Units = cylinders of 16065 *512 = 8225280 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes Disk identifier: 0xb4b0720c
Device Boot Start End Blocks id System
/dev/sdb1 1 654
5253223+ 83 Linux
#主分区
/dev/sdb2 655 2610 15711570
5 Extended
#扩展分区
/dev/sdb5 655 916
2104483+ 83 Linux
#逻辑分区 1
/dev/sdb6 917 1178
2104483+ 83 Linux
#逻辑分区2
Command (m for help): w
#保存并退出
The partition table has been altered!
Calling ioctl。to re-read partition table.
Syncing disks.
[[email protected] -]#
#退回到提示符界面

所有的分区立过程中如果不保存并退出是不会生效的，所以建立错了也没有关系，使用 q 命令不保存退出即可。如果使用了 w 命令，就会保存退出。有时因为系统的分区表正忙，所以需要重新启动系统才能使新的分区表生效。命令如下：

Command (m for help): w
#保存并退出
The partition table has been altered!
Calling ioctl() to re-read partition table.
WARNING: Re-reading the partition table failed with error 16:
Device or resource busy.
The kernel still uses the old table.
The new table will be used at the next reboot.
#要求重新启动，才能格式化
Syncing disks.

必须重新启动！

mkfs命令格式化分区

分区完成后，如果不格式化写入文件系统，则是不能正常使用的。所以我们需要使用 mkfs 命令进行格式化。
mkfs 命令格式如下：

[[email protected] ~]# mkfs [选项] 分区设备文件名

选项：-t 文件系统：指定格式化的文件系统，如ext3、ext4；

在Linux分区中  /dev/sdb1（主分区）、/dev/sdb2（扩展分区）、/dev/sdb5（逻辑分区）和 /dev/sdb6（逻辑分区）这几个分区，其中 /dev/sdb2 不能被格式化。剩余的三个分区都需要格式化之后使用。命令如下：

[[email protected] ~]# mkfs -t ext4 /dev/sdb6
mke2fs 1.41.12 (17-May-2010)
文件系统标签=
这里指的是卷标名，我们没有设置卷标
操作系统：Linux
块大小=4096 (log=2)
分块大小=4096 (log=2)
Stride=0 blocks, Stripe width=0 blocks
131648 inodes, 526120 blocks
26306 blocks (5.00%) reserved for the super user
第一个数据块=0
Maximum filesystem blocks=541065216 17 block groups
32768 blocks per group, 32768 fragments per group
7744 inodes per group
Superblock backups stored on blocks:
32768, 98304, 163840, 229376, 294912
正在写入inodes表：完成
Creating journal (16384 blocks):完成
Writing superblocks and filesystem accounting information:完成
This filesystem will be automatically checked every 39 mounts or 180 days, whichever comes first. Use tune2fs -c
or -i to override.
[[email protected] ~]# mkfs -t ext4 /dev/sdb5
#把dev/sdb5也格式化

mkfs 命令非常简单易用，不过是不能调整分区的默认参数的（比如块大小是 4096 Bytes），这些默认参数除非特殊清况，否则不需要调整。如果想要调整，就需要使用 mke2fs 命令重新格式化。命令格式如下：

[[email protected] ~]# mke2fs [选项] 分区设备文件名

选项：

选项	选项说明
-t	文件系统：指定格式化成哪个文件系统， 如 ext2、ext3、ext4
-b	字节：指定 block 的大小
-i	字节：指定"字节 inode "的比例，也就是多少字节分配一个 inode
-j	建立带有 ext3 日志功能的文件系统
-L	卷标名：给文件系统设置卷标名，就不使用 e2label 命令设定了

格式化过程如下：

[[email protected] ~]# mke2fs -t ext4 -b 2048 /dev/sdb6
#格式化分区，并指定block的大小为2048 Bytes
mke2fe 1.41.12 (17-May-2010)
文件系统标签=
操作系统：Linux
块大小=2048 (log=1)
#block的大小就不再是4096Bytes 了
分块大小=2048 (log=1)
Stride=0 blocks, Stripe width=0 blocks 131560
inodes,1052240 blocks 52612 blocks (5.00%) reserved for the super user
第一个数据块=0
Maximum filesystem blocks=538968064
65 block groups
16384 blocks per group, 16384 fragments per group
2024 inodes per group
Superblock backups stored on blocks:
16384, 49152, 81920, 114688, 147456, 409600, 442368, 802816
正在写入inode表：完成
Creating journal (32768 blocks):完成
Writing superblocks and filesystem accounting information:完成
This filesystem will be automatically checked every 38 mounts or
180 days, whichever comes first. Use tune2fs -c or-i to override.