linux 编译system.img,android生成sysytem.img的命令过程

From

http://xianghuaclm.blog.163.com/blog/static/6668783201110221486240/

使用#file system.img

输出：system.img: Linux rev 0.0 ext2 filesystem data,

UUID=4eb0489c-647d-49d8-9111-94657b5fd342, volume name

"system"。

然而本人使用命令#mkyaffs2image system/ system.img，

再使用#file system.img，却输出:system.img: VMS Alpha executable

前一个system.img文件时可用的，而使用mkyaffs2image命令生成的system.img却不能使用(将ramdisk.img,system.img,install.img等打包成iso文件，用virtualbox运行)。

谁 清楚android编译系统是用什么命令生成system.img的？本人使用#make

systemimage命令，只显示出如下信息: Install system fs image:

out/target/product/eeepc/system.img

out/host/linux-x86/bin/acp -fpt

out/target/product/eeepc/obj/PACKAGING /systemimage_unopt_intermediates/system.img

out/target/product/eeepc/system.img

true

没能查到是哪个脚本生成的system.img

根据这个网址http://blog.csdn.net/liaoshengji

... /03/04/3957749.aspx，

1.执行#make showcommands snod

2.得出android制作system.img命令为:

#!/bin/bash

export PATH="/home/tony/tmp/e2fsprogs-1.41.3/prefix/sbin:$PATH"

num_blocks=`du -sk out/target/product/eeepc/system | tail -n1 |

awk '{print $1;}'`

if [ $num_blocks -lt 20480 ]; then

extra_blocks=3072;

else

extra_blocks=20480;

fi

num_blocks=`expr $num_blocks + $extra_blocks`

num_inodes=`find out/target/product/eeepc/system | wc -l`

num_inodes=`expr $num_inodes + 500`

out/host/linux-x86/bin/genext2fs -a -d

out/target/product/eeepc/system -b $num_blocks -N $num_inodes -m 0

out/target/product/eeepc/system.img

tune2fs -L system out/target/product/eeepc/system.img

tune2fs -C 1 out/target/product/eeepc/system.img

e2fsck -fy out/target/product/eeepc/system.img

一.android2.3系统的generic文件系统生成命令：

out/host/linux-x86/bin/mkyaffs2image -f out/target/product/generic/system out/target/product/generic/obj/PACKAGING/systemimage_intermediates/system.img

chmod a+r

out/target/product/generic/obj/PACKAGING/systemimage_intermediates/system.img

Install system fs image:

out/target/product/generic/system.img

out/host/linux-x86/bin/acp -fpt

out/target/product/generic/obj/PACKAGING/systemimage_intermediates/system.img

out/target/product/generic/system.img

true

out/host/linux-x86/bin/mkbootfs

out/target/product/generic/root | out/host/linux-x86/bin/minigzip

> out/target/product/generic/ramdisk.img

out/host/linux-x86/bin/mkyaffs2image -f out/target/product/generic/data

out/target/product/generic/userdata.img

chmod a+r

out/target/product/generic/userdata.imgtrue

二.android2.2系统的MID文件系统生成命令：

file out/target/product/mid/system.img

out/target/product/mid/system.img: u-boot legacy uImage, Android,

Linux/ARM, Filesystem Image (any type) (Not compressed), 143233024

bytes, Tue Nov 22 10:48:07 2011, Load Address: 0x40008000, Entry

Point: 0x40008000, Header CRC: 0x6D462E1C, Data CRC: 0xE1876EFF

file sys.img

sys.img: Linux rev 1.0 ext3 filesystem data,

UUID=2c363e3e-f731-4800-9553-a0613a1264eb, volume name "system

dd if=system.img of=sys.img bs=64 skip=1

sudo mount -t ext3 -o loop sys.img mnt

修改

sudo umount mnt

./rcmkimage sys.img system.img(./mkimage -A arm -O linux -C none

-T filesystem -a 1 -e 1 -n "system.img" -d $1 $2)

make showcommands

Target system fs image:

out/target/product/mid/obj/PACKAGING/systemimage_unopt_intermediates/system.img

Target system fs image: ext2image

out/target/product/mid/system/framework

if [ -f

out/target/product/mid/system/framework/framework-res-TG.apk ];

then mv

out/target/product/mid/system/framework/framework-res-TG.apk

out/target/product/mid/system/framework/framework-res.apk; fi

num_blocks=`du -sk out/target/product/mid/system | tail -n1 | awk

'{print $1;}'`; if [ $num_blocks -lt 20480 ]; then

extra_blocks=3072; else extra_blocks=20480; fi ; num_blocks=`expr

$num_blocks + $extra_blocks` ; num_inodes=`find

out/target/product/mid/system | wc -l` ; num_inodes=`expr

$num_inodes + 500`;

out/host/linux-x86/bin/genext2fs -a

-d out/target/product/mid/system -b $num_blocks -N $num_inodes -m 0

out/target/product/mid/obj/PACKAGING/systemimage_unopt_intermediates/system.img

tune2fs -L system

out/target/product/mid/obj/PACKAGING/systemimage_unopt_intermediates/system.img

tune2fs 1.41.14 (22-Dec-2010)

tune2fs -C 1

out/target/product/mid/obj/PACKAGING/systemimage_unopt_intermediates/system.img

tune2fs 1.41.14 (22-Dec-2010)

Setting current mount count to 1

e2fsck -fy

out/target/product/mid/obj/PACKAGING/systemimage_unopt_intermediates/system.img

; [ $? -lt 4 ]e2fsck 1.41.14

(22-Dec-2010)

文件系统 did not have a UUID; generating one.

第一步: 检查inode,块,和大小

第二步: 检查目录结构

第3步: 检查目录连接性

/lost+found未找到.创建? 是

Pass 4: Checking reference counts

第5步: 检查簇概要信息

system: ***** 文件系统已修改 *****

system: 648/1152 files (1.7% non-contiguous), 119292/139876

blocks

Install system fs image: out/target/product/mid/system.img

out/host/linux-x86/bin/acp -fpt

out/target/product/mid/obj/PACKAGING/systemimage_unopt_intermediates/system.img

out/target/product/mid/system.img

true

Generate system.img for iNAND

mv out/target/product/mid/system.img

out/target/product/mid/system.img.ext3

tune2fs -j

out/target/product/mid/system.img.ext3

tune2fs 1.41.14 (22-Dec-2010)

Creating journal inode: 完成

This filesystem will be automatically checked every 20 mounts

or

0 days, whichever comes first. Use tune2fs -c or

-i to override.

out/host/linux-x86/bin/mkimage -A arm

-O linux -T filesystem -C

none \

-a 40008000 -n "Android" -d out/target/product/mid/system.img.ext3

out/target/product/mid/system.imgImage

Name: Android

Created: Tue Nov 22 11:05:21 2011

Image Type: ARM Linux

Filesystem Image (uncompressed)

Data Size: 143233024 Bytes = 139876.00 kB = 136.60 MB

Load Address: 40008000

Entry Point: 40008000

gzip -9

out/target/product/mid/system.img.ext3

out/host/linux-x86/bin/mkimage -A arm

-O linux -T filesystem -C

none \

-a 40008000 -n "zAndroid" -d

out/target/product/mid/system.img.ext3.gz

out/target/product/mid/zSYS.img

Image Name: zAndroid

Created: Tue Nov 22 11:05:34 2011

Image Type: ARM Linux

Filesystem Image (uncompressed)

Data Size: 80767125 Bytes = 78874.15 kB = 77.03 MB

Load Address: 40008000

Entry Point: 40008000

rm -f out/target/product/mid/system.img.ext3.gz

out/host/linux-x86/bin/mkbootfs out/target/product/mid/root |

out/host/linux-x86/bin/minigzip >

out/target/product/mid/ramdisk.img

mv out/target/product/mid/ramdisk.img

out/target/product/mid/ramdisk.img.ori

out/host/linux-x86/bin/mkimage -A arm -O linux -T ramdisk \

-C none -a 0x41000000 -n "hRamdisk" -d

out/target/product/mid/ramdisk.img.ori

out/target/product/mid/ramdisk.img

Image Name: hRamdisk

Created: Tue Nov 22 11:05:35 2011

Image Type: ARM Linux RAMDisk

Image (uncompressed)

Data Size: 170001 Bytes = 166.02 kB = 0.16 MB

Load Address: 41000000

Entry Point: 41000000

Installed file list:

out/target/product/mid/installed-files.txt

build/tools/fileslist.py out/target/product/mid/system

out/target/product/mid/data >

out/target/product/mid/installed-files.txt

三.相关命令使用说明：

genext2fs：

out/host/linux-x86/bin/genext2fs --help

Usage: out/host/linux-x86/bin/genext2fs [options] image

Create an ext2 filesystem image from directories/files

-x, --starting-image

-d, --root

-D, --devtable

-b, --size-in-blocks

-i, --bytes-per-inode

-N, --number-of-inodes

-m, --reserved-percentage

-g, --block-map

Generate a block map file for this path.

-e, --fill-value

Fill unallocated blocks with

value.

-z,

--allow-holes Allow files with holes.

-f,

--faketime Set filesystem timestamps to 0 (for testing).

-q,

--squash Same as "-U -P".

-U,

--squash-uids Squash owners making all files be owned by root.

-P,

--squash-perms Squash permissions on all files.

-a,

--fix-android-stats Fix-up file stats (user, perms, ...)

-h, --help

-V, --version

-v, --verbose

tune2fs ：

tune2fs

tune2fs 1.41.14 (22-Dec-2010)

Usage: tune2fs [-c max_mounts_count] [-e errors_behavior] [-g

group]

[-i interval[d|m|w]] [-j] [-J journal_options] [-l]

[-m reserved_blocks_percent] [-o [^]mount_options[,...]]

[-r reserved_blocks_count] [-u user] [-C mount_count] [-L

volume_label]

[-M last_mounted_dir] [-O [^]feature[,...]]

[-E extended-option[,...]] [-T last_check_time] [-U UUID]

[ -I new_inode_size ] device

一.简介：

调整/查看ext2/ext3文件系统的文件系统参数，Windows下面如果出现意外断电死机情况，下次开机一般都会出现系统自检。Linux系统下面也有文件系统自检，而且是可以通过tune2fs命令，自行定义自检周期及方式。

二.用法：

tune2fs [ -l ] [ -c max-mount-counts ] [ -e errors-behavior ] [

-f ] [ -i interval-between-checks ] [ -j ] [ -J journal-options ] [

-m

reserved-blocks-percentage ] [ -o [^]mount-options[,...] ] [ -r

reserved-blocks-count ] [ -s sparse-super-flag ] [ -u user ] [

-g

group ] [ -C mount-count ] [ -L volume-name ] [ -M

last-mounted-directory ] [ -O [^]feature[,...] ] [ -T

time-last-checked ] [ -U

UUID ] device

常用选项说明：

-l 查看文件系统信息

-c max-mount-counts 设置强制自检的挂载次数，如果开启，每挂载一次mount

conut就会加1，超过次数就会强制自检

-i interval-between-checks[d|m|w] 设置强制自检的时间间隔[d天m月w周]

-m reserved-blocks-percentage 保留块的百分比

-j 将ext2文件系统转换为ext3类型的文件系统

-L volume-label 类似e2label的功能，可以修改文件系统的标签

-r reserved-blocks-count 调整系统保留空间

-o [^]mount-option[,...] Set or clear the indicated default mount

options in the filesystem. 设置或清除默认挂载的文件系统选项

三.示例：

tune2fs -c 30 /dev/hda1 设置强制检查前文件系统可以挂载的次数

tune2fs -c -l /dev/hda1 关闭强制检查挂载次数限制。

tune2fs -i 10 /dev/hda1 10天后检查

tune2fs -i 1d /dev/hda1 1天后检查

tune2fs -i 3w /dev/hda1 3周后检查

tune2fs -i 6m /dev/hda1 半年后检查

tune2fs -i 0 /dev/hda1 禁用时间检查

tune2fs -j /dev/hda1 添加日志功能，将ext2转换成ext3文件系统

tune2fs -r 40000 /dev/hda1 调整/dev/hda1分区的保留空间为40000个磁盘块

tune2fs -o acl,user_xattr /dev/hda1 设置/dev/hda1挂载选项，启用Posix

Access Control Lists和用户指定的扩展属性

e2fsck ：

e2fsck

Usage: e2fsck [-panyrcdfvtDFV] [-b superblock] [-B blocksize]

[-I inode_buffer_blocks] [-P process_inode_size]

[-l|-L bad_blocks_file] [-C fd] [-j external_journal]

[-E extended-options] device

Emergency help:

-p Automatic repair (no questions)

-n Make no changes to the filesystem

-y Assume "yes" to all questions

-c Check for bad blocks and add them to the badblock list

-f Force checking even if filesystem is marked clean

-v Be verbose

-b

superblock Use alternative superblock

-B

blocksize Force blocksize when looking for superblock

-j external_journal Set

location of the external journal

-l

bad_blocks_file Add to

badblocks list

-L

bad_blocks_file Set badblocks

list

e2fsck(ext2

file system

check)

功能说明：检查ext2文件系统的正确性。

语法：e2fsck [-acCdfFnprsStvVy][-b

<

href="http://www.linuxso.com/command/su.html">superblock>][-B

][-l ][-L ][设备名称]

补充说明：e2fsck执行后的传回值及代表意义如下。

0 没有任何错误发生。

1 文件系统发生错误，并且已经修正。

2 文件系统发生错误，并且已经修正。

4 文件系统发生错误，但没有修正。

8 运作时发生错误。

16 使用的语法发生错误。

128 共享的函数库发生错误。

参数：

-a 不询问使用者意见，便自动修复文件系统。

-b 指定superblock，而不使用预设的superblock。

-B 指定区块的大小，单位为字节。

-c 一并执行badblocks，以标示损坏的区块。

-C 将检查过程的信息完整记录在file

descriptor中，使得整个检查过程都能完整监控。

-d 显示排错信息。

-f 即使文件系统没有错误迹象，仍强制地检查正确性。

-F 执行前先清除设备的缓冲区。

-l 将文件中指定的区块加到损坏区块列表。

-L 先清除损坏区块列表，再将文件中指定的区块加到损坏区块列表。因此损坏区块列表的区块跟文件中指定的区块是一样的。

-n 以只读模式开启文件系统，并采取非互动方式执行，所有的问题对话均设置以"no"回答。

-p 不询问使用者意见，便自动修复文件系统。

-r 此参数只为了兼容性而存在，并无实际作用。

-s 如果文件系统的字节顺序不适当，就交换字节顺序，否则不做任何动作。

-S 不管文件系统的字节顺序，一律交换字节顺序。

-t 显示时间信息。

-v 执行时显示详细的信息。

-V 显示版本信息。

-y 采取非互动方式执行，所有的问题均设置以"yes"回答。

acp ：

out/host/linux-x86/bin/acp -?

out/host/linux-x86/bin/acp: invalid option -- '?'

Unexpected arg -?

Usage: acp [OPTION]... SOURCE DEST

or: acp [OPTION]... SOURCE...

DIRECTORY

Options:

-d never follow (dereference)

symbolic links

-e if source file doesn't

exist, try adding '.exe' [Win32 only]

-f use force, removing

existing file if it's not writeable

-p preserve mode,

ownership

-r recursive copy

-t preserve timestamps

-u update only: don't copy if

dest is newer

-v verbose output (-vv is more

verbose)

mkimage --help

Usage: mkimage -l image

-l ==> list image header information

mkimage [-x] -A arch -O os -T type -C comp -a addr -e ep -n name -d

data_file[:data_file...] image

-A ==> set architecture to 'arch'

-O ==> set operating system to 'os'

-T ==> set image type to 'type'

-C ==> set compression type 'comp'

-a ==> set load address to 'addr' (hex)

-e ==> set entry point to 'ep' (hex)

-n ==> set image name to 'name'

-d ==> use image data from 'datafile'

-x ==> set XIP (execute in place)

mkimage [-D dtc_options] -f fit-image.its fit-image