Problems accumulated during the startup process of embedded Linux

1.Bad Magic Number

## Booting image at 33000000 ...

Bad Magic Number

OMAP5912 OSK #  (After tftp downloads the kernel uImage, it stops here and cannot start the kernel)

Cause of the problem: The startup parameter setting is wrong, and 0x30000000 cannot be executed.

Some development boards have sdram not at 0x33000000, so the kernel uImage cannot be downloaded to 0x33000000 and run. For example, my previous bootcmd parameter was: setenv bootcmd tftpboot 33000000 uImage/; bootm 33000000. But the sdram address of the board Omap5912 is at 0x100000000. After changing the parameter to setenv bootcmd tftpboot 10000000 uImage/; bootm 10000000, the kernel can be started.

2. The startup stops at "Starting kernel ..."

TFTP from server 192.168.167.170; our IP address is 192.168.167.15

Filename 'uImage'

Load address: 0x10000000

Loading: #################################################################

         #################################################################

done

Bytes transferred = 2025908 (1ee9b4 hex)

## Booting image at 10000000 ...

   Image Name:   Linux-2.6.18-mh8_pro500-versatil

   Image Type:   ARM Linux Kernel Image (uncompressed)

   Data Size:    2025844 Bytes =  1.9 MB

   Load Address: 30008000

   Entry Point:  30008000

   Verifying Checksum ... OK

OK

Starting kernel ...    (stuck here)

Cause of the problem: Most likely the kernel was not compiled successfully.

Confirm whether the configure parameters are configured correctly and whether the correct target compilation platform is selected, such as smdk2410.

3. Unable to start kernel

Starting kernel ...

Uncompressing Linux.........................................................

.................... done, booting the kernel.

Cause: The Bootargs parameter may be set incorrectly. Check whether the bootargs setting is correct.

4. Unable to mount nfs

eth0: link up

IP-Config: Complete:

      device=eth0, addr=192.168.167.15, mask=255.255.255.0, gw=192.168.167.254,

     host=192.168.167.15, domain=, nis-domain=(none),

     bootserver=192.168.167.170, rootserver=192.168.167.170, rootpath=

Looking up port of RPC 100003/2 on 192.168.167.170

Root-NFS: Unable to get nfsd port number from server, using default

Looking up port of RPC 100005/1 on 192.168.167.170

Root-NFS: Unable to get mountd port number from server, using default

mount: server 192.168.167.170 not responding, timed out

Root-NFS: Server returned error -5 while mounting /work/nfs/rootfs_bluetooth_omap

VFS: Unable to mount root fs via NFS, trying floppy.

VFS: Cannot open root device "nfs" or unknown-block(2,0)

Please append a correct "root=" boot option

Kernel panic - not syncing: VFS: Unable to mount root fs on unknown-block(2,0)

Cause: This is usually an NFS configuration problem.

Confirm whether the nfs-related ip address information in the uboot bootargs parameter is set correctly, and whether the host machine /etc/exports configuration is correct, restart the nfs service, and try to connect again. In addition, you need to pay attention to the settings of the console and mem parameters in bootargs. After kernel2.6, it is best to set the console to ttySAC0, and mem should also be set correctly according to the actual situation of the development board.

5. The file system cannot be started

eth0: link up

IP-Config: Complete:

      device=eth0, addr=192.168.167.15, mask=255.255.255.0, gw=192.168.167.254,

     host=192.168.167.15, domain=, nis-domain=(none),

     bootserver=192.168.167.170, rootserver=192.168.167.170, rootpath=

Looking up port of RPC 100003/2 on 192.168.167.170

Looking up port of RPC 100005/1 on 192.168.167.170

VFS: Mounted root (nfs filesystem).

Freeing init memory: 128K

/sbin/initKernel panic - not syncing: Attempted to kill init!

Cause of the problem: The created file system lacks the libcrypt.so library required to run busybox. The new version will have a missing library prompt, but the old version (1.60) does not.

Note: Running a busybox file system requires at least the following libraries:

ld-linux.so.x

libc.so.6

libcrypt.so.x

Newer versions of busybox may also require

libm.so.6

libgcc_s.so.x

(x is the version number)

6. File system cannot be started problem 2

eth0: link up

IP-Config: Complete:

      device=eth0, addr=192.168.167.15, mask=255.255.255.0, gw=192.168.167.254,

     host=192.168.167.15, domain=, nis-domain=(none),

     bootserver=192.168.167.170, rootserver=192.168.167.170, rootpath=

Looking up port of RPC 100003/2 on 192.168.167.170

Looking up port of RPC 100005/1 on 192.168.167.170

VFS: Mounted root (nfs filesystem).

Freeing init memory: 128K

Kernel panic - not syncing: No init found.  Try passing init= option to kernel.

Cause of the problem: After comparing with an available file system, it was found that the ld-linux.so.x library was missing. There was only the link file of ld-linux.so.x in the file system, and the library file was not copied.

8. Unable to obtain account UID information

Could not get password database information for UID of current process: User "???" unknown or no memory to allocate password entry

Unknown username "root" in message bus configuration file

Could not get password database information for UID of current process: User "???" unknown or no memory to allocate password entry

Failed to start message bus: Could not get UID and GID for username "root"

problem causes:

Case 1: System account verification has problems. It is suspected that the correct value is not returned when calling getuid and getguid. It may be that the account verification related library is missing. After actual investigation, it is found that the libnss_files library is missing. After copying the cross-compiler's libnss_files library to the file system, the file system is started successfully.

Case 2: The system does not have a root account. This can be seen by the whoami command.

Create an account manually.

#vi /etc/passwd

root:x:0:0:root:/root:/bin/sh

kyo:x:500:500:kyo:/home/kyo:/bin/bash

Add Group

#vi group

root:x:0:root

9.

Freeing init memory: 128K

init started: BusyBox v1.6.1 (2007-08-27 14:33:15 CST) multi-call binary

starting pid 834, tty '': '/etc/init.d/rcS'

Cannot run '/etc/init.d/rcS': No such file or directory

Please press Enter to activate this console.

I found that it can be started normally without the /etc/init.d/rcS file system. It seems that rcS is only used to set some random startup parameters and has little to do with whether the file can run normally.

Note: This is not a mistake, it was discovered by accident! :)