Start a minimal file system from nfs

1. Introduction

The development environment is Ubuntu server 18.04. The nand used in the information I found is 64M, and the board I have is 256M, so the image bin file in the information cannot be used directly. There will be an ecc problem, because he uses yaffs on the 64M nand and can only use yaffs2 on the 256M nand. Due to my insufficient level, in fact, I am just lazy, so I don’t want to modify the relevant code and choose to start the root file system from nfs. After all, it is convenient to port drivers or write applications in the future and directly use nfs to mount. After several days of tossing, I finally filled in all the holes and successfully started to the shell, but self-study still takes too many detours.

2. Network environment configuration

To start the development board from nfs, you must first establish a network connection between the virtual machine and the development board. Because my virtual machine is connected to the external network through vmnet8, I choose to bridge vmnet0 to the USB network card to connect to the development board.

Run the virtual network editor as an administrator and configure vmnet0 as follows:

Then add a network card to the virtual machine and specify the connection to vmnet0:

Enter the virtual machine and assign an address to the virtual network card:

sudo vim /etc/netplan/00-installer-config.yaml #The specific file name varies for different virtual machines, just complete it in tab

My configuration is as follows:

network:

version: 2

bonds: {}

bridges: {}

ethernets:

ens33:

addresses:

- 192.168.112.3/24

dhcp4: false

dhcp6: false

gateway4: 192.168.112.2

nameservers:

addresses:

- 192.168.112.2

ens35:

addresses:

- 192.168.8.3/24

dhcp4: false

dhcp6: false

routes:

- to: 192.168.8.0/24

via: 192.168.8.2

vlans: {}

wifi: {}

Set the gateway and other parameters according to the ip summarized above. ens35 is vmnet0, which is bridged to the USB network card. Be careful not to use gateway4 to configure the gateway, otherwise a double default gateway will be generated when the machine is turned on, and there will be problems connecting the virtual machine to the external network.

If there is no ens35, you can use ifconfig -a to check first and then use ( ifconfig network card name up ) to enable the corresponding network card.

route -n

Use the above command to view the routing table as follows:

streleizia@ubuntu:~$ route -n

Kernel IP routing table

Destination     Gateway         Genmask         Flags Metric Ref    Use Iface

0.0.0.0 192.168.112.2 0.0.0.0 AND 0 0 0 ens33

192.168.8.0 0.0.0.0 255.255.255.0 U 0 0 0 ens35

192.168.8.0 192.168.8.2 255.255.255.0 AND 0 0 0 ens35

192.168.112.0 0.0.0.0 255.255.255.0 U 0 0 0 ens33

streleizia@ubuntu:~$

You can see that there is only one default route (destination is 0.0.0.0) whose gateway is the external network card (192.168.112.2). In this way, traffic accessing the external network will not be mistakenly sent to the internal network gateway.

In addition, dual network cards may cause network problems on the host (Windows). You can view the host routing table (enter route print in cmd to view it) and then delete the route pointing to the USB network card, and configure the default route of the external network card as follows:

IPv4 routing table

===========================================================================

Active Routing:

Network Destination Netmask Gateway Interface Metric

0.0.0.0          0.0.0.0      192.168.1.1    192.168.1.107     35

127.0.0.0 255.0.0.0 On link 127.0.0.1 331

127.0.0.1 255.255.255.255 On link 127.0.0.1 331

127.255.255.255 255.255.255.255 on link 127.0.0.1 331

192.168.1.0 255.255.255.0 on link 192.168.1.107 291

192.168.1.107 255.255.255.255 On link 192.168.1.107 291

192.168.1.255 255.255.255.255 on link 192.168.1.107 291

192.168.112.0 255.255.255.0 on link 192.168.112.1 291

192.168.112.1 255.255.255.255 On link 192.168.112.1 291

192.168.112.255 255.255.255.255 on link 192.168.112.1 291

224.0.0.0 240.0.0.0 on link 127.0.0.1 331

224.0.0.0 240.0.0.0 on link 192.168.112.1 291

224.0.0.0 240.0.0.0 on link 192.168.1.107 291

255.255.255.255 255.255.255.255 on link 127.0.0.1 331

255.255.255.255 255.255.255.255 on link 192.168.112.1 291

255.255.255.255 255.255.255.255 on link 192.168.1.107 291

===========================================================================

Permanent Routing:

Network AddressNetwork MaskGateway AddressMetrics

0.0.0.0  0.0.0.0  192.168.1.1   1

===========================================================================

192.168.1.1 is the address of the home router

Finally, make sure to configure the host's USB network card to the same network segment as the virtual machine, as shown in the figure:

Summarize:

The USB network card IP is manually configured to: 192.168.8.2

The virtual machine IP is manually configured as: 192.168.8.3 The gateway is 192.168.8.2

The development board IP is manually configured as: 192.168.8.100 The gateway is 192.168.8.2 (will be configured later)

Ping each other. If the ping is successful, it is considered successful. If there is no response when pinging the Windows host, it is because the host firewall is not turned off (it only affects ping, not the subsequent nfs and tftp transmission). In addition, the Linux virtual machine firewall may also be turned on, just turn it off.

3. Configure the software environment

The network is set up, and all that remains is to install various software. Use the following command to install the software package in Ubuntu:

sudo apt-get install tftpd tftp xinetd nfs-kernel-server vsftpd

Then configure them one by one:

First configure ftp:

sudo you /etc/vsftpd.conf

Remove the # sign before the following configuration

#local_enable=YES

#write_enable=YES

After saving, restart the service using the following command:

sudo /etc/init.d/vsftpd restart

Next, configure tftp:

First create a folder in a preferred location. I chose to create the nfs_root directory in the home directory. Don’t forget to add permissions for all users with sudo chmod 777 nfs_root -R

streleizia@ubuntu:~/nfs_root$ pwd

/home/streleizia/nfs_root #You can see the path: /home/streleizia/nfs_root

streleizia@ubuntu:~/nfs_root$ tree -d

.

├── fs_mini #busybox is installed in this folder

│   ├── bin

│   ├── dev

│   ├── etc

│   │   └── init.d

│ ├── lib

│   ├── mnt

│   ├── proc

│   ├── root

│   ├── sbin

│   ├── sys

│   ├── tmp

│   └── usr

│   ├── bin

│   └── sbin

└── fs_mini_mdev

├── bin

├── dev

├── etc

│   └── init.d

├── lib

├── mnt

├── proc

├── root

├── sbin

├── sys

├── tmp

└── usr

├── bin

└── sbin

30 directories

streleizia@ubuntu:~/nfs_root$

sudo you /etc/xinetd.d/tftp

After opening the configuration file, enter the following:

service tftp

{

socket_type     =dgram

protocol        =udp

wait            =yes

user            =root

server          =/usr/sbin/in.tftpd

server_args =-s /home/streleizia/tftp_root #Note that the path here is consistent with the one created above

disable         =no

per_source      =11

cps             =100 2

flags           =IPv4

}

sudo /etc/init.d/xinetd restart

Restart the service and it will be ok if there is no error.

sudo vi /etc/exports

Finally configure nfs:

Open the nfs configuration file and add the following configuration:

# /etc/exports: the access control list for filesystems which may be exported

#               to NFS clients.  See exports(5).

#

# Example for NFSv2 and NFSv3:

# /srv/homes       hostname1(rw,sync,no_subtree_check) hostname2(ro,sync,no_subtree_check)

#

# Example for NFSv4:

# /srv/nfs4        gss/krb5i(rw,sync,fsid=0,crossmnt,no_subtree_check)

# /srv/nfs4/homes  gss/krb5i(rw,sync,no_subtree_check)

/home/streleizia/workspace/yaffs2_tool/nfs_root *(rw,sync,no_root_squash) #Here I created three nfs directories, which are root file systems built in different ways.

/home/streleizia/nfs_root/fs_mini *(rw,sync,no_root_squash) #Note that it is consistent with the directory path created above

/home/streleizia/nfs_root/fs_mini_mdev 　　　　　　　　　　*(rw,sync,no_root_squash)

sudo /etc/init.d/nfs-kernel-server restart

Restart the service and it will be ok if there is no error.

Here comes the point. The nfs version automatically installed by Ubuntu 18.04 is too high and does not match the 2.6.22 kernel version, which causes nfs to fail to mount and the kernel is stuck at:

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

Solution reference article:

https://blog.csdn.net/Qiuoooooo/article/details/112601940

That is: add RPCNFSDOPTS="--nfs-version 2,3,4 --debug --syslog" in /etc/default/nfs-kernel-server and restart NFS

4. Test nfs

Use tftp to burn the kernel image with uboot. The detailed method will not be repeated here. Pay attention to using the matching root file system and kernel image. It is best to be able to start normally on nand. Set the startup parameters in uboot as follows:

bootargs=noinitrd root=/dev/nfs nfsroot=192.168.8.3:/home/streleizia/nfs_root/fs_mini ip=192.168.8.100:192.168.8.3:192.168.8.2:255.255.255.0:linux:eth0:off init=/linuxrc console=ttySAC0

The boot kernel outputs information and successfully boots into the busybox command line!

drivers/hid/usbhid/hid-core.c: v2.6:USB HID core driver

TCP cubic registered

NET: Registered protocol family 1

s3c2440-sdi s3c2440-sdi: CMD[TIMEOUT] #5 op:APP_CMD(55) arg:0x00000000 flags:0x0875 retries:0 Status:nothing to complete

s3c2440-sdi s3c2440-sdi: CMD[TIMEOUT] #6 op:APP_CMD(55) arg:0x00000000 flags:0x0875 retries:0 Status:nothing to complete