make-kernel.md

install toolchain

Need to install cross-compiling toolchain since Ubunut is x86_64 arch.

# arm
sudo apt-get install gcc-aarch64-linux-gnu libncurses5-dev  build-essential git bison flex libssl-dev

# x86_64
sudo apt install -y build-essential libncurses-dev bison flex libssl-dev libelf-dev bc git

clone kernel

makdir -P /code && cd /code

# vpn doesn't support git:// protocol
nohup git clone https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git /code/linux &

# sched/core subsystem git repo
git remote add tip git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git
git remote add stable https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git

fetch tag

# fetch all tags
git fetch --tags

# fetch src(remote) tag as dst(local) tag
git fetch stable refs/tags/v6.1.120:refs/tags/v6.1.120

# list specific tag commit hash
git ls-remote stable refs/tags/v6.1.120

# filter tags
git tag -l | grep "v6.1.120*"

git show v6.1.120

make ramfs

Linux needs root fs to start, we can do this by make a ramfs by busybox.

make busybox

wget -P /code/ https://busybox.net/downloads/busybox-1.33.1.tar.bz2
tar -xjf busybox-1.33.1.tar.bz2 -C /code/
cd /code/busybox-1.33.1/

• specify compiler platform

  export ARCH=arm64 export CROSS_COMPILE=aarch64-linux-gnu-

• config busybox

  Remove networking/tc.c

  make menuconfig ARCH=arm64 ---> Settings --->  [*] Build static binary (no shared libs)

• make

  make -j$(nproc)
  make CONFIG_PREFIX=/code/ramfs install

_install dir under busybox is generated after successful make

config ramfs

We need additional config to start init process:

• mkdir mkdir etc dev lib

  cd /code/busybox-1.33.1/_install
  mkdir -p etc/init.d dev lib
  ls
  # bin  dev  etc  lib  linuxrc  sbin  usr

• vim etc/profile

  This is a shell configuration file executed when a user logs into a shell

  #!/bin/sh
  export HOSTNAME=jemmy
  export USER=root
  export HOME=/home
  export PS1="[$USER@$HOSTNAME \W]\# "
  PATH=/bin:/sbin:/usr/bin:/usr/sbin
  LD_LIBRARY_PATH=/lib:/usr/lib:$LD_LIBRARY_PATH
  export PATH LD_LIBRARY_PATH

• vim etc/inittab

  This is the configuration file for the init process (PID 1), which is the first process started by the Linux kernel during boot. It tells init how to manage system initialization, runlevels, and processes.

  ::sysinit:/etc/init.d/rcS   #Runs the /etc/init.d/rcS script at boot.
  ::respawn:-/bin/sh          # Starts a login shell
  ::askfirst:-/bin/sh
  ::ctrlaltdel:/bin/umount -a -r  # Defines behavior for Ctrl+Alt+Del
  # ::ctrlaltdel:/sbin/reboot

• vim etc/fstab

  This file lists filesystems to be mounted during boot, specifying their mount points, types, and options.

  #device  mount-point        type     options    dump fsck order
  proc    /proc               proc    defaults    0   0
  tmpfs   /tmp                tmpfs   defaults    0   0
  sysfs   /sys                sysfs   defaults    0   0
  tmpfs   /dev                tmpfs   defaults    0   0
  debugfs /sys/kernel/debug   debugfs defaults    0   0
  kmod_mount /mnt             9p      trans=virtio 0  0

• vim etc/init.d/rcS

  This is a shell script executed during system initialization (often triggered by inittab’s sysinit entry). It performs early setup tasks.

  #!/bin/sh
  
  mkdir -p /proc /sys /tmp /mnt /dev /etc /var
  
  # Create essential device nodes
  mknod -m 622 /dev/console c 5 1
  mknod -m 666 /dev/null c 1 3
  mknod -m 666 /dev/zero c 1 5
  mknod -m 666 /dev/tty c 5 0
  
  /bin/mount -a
  # Mount virtual filesystems
  mount -t proc proc /proc
  mount -t sysfs sysfs /sys
  mount -t devtmpfs none /dev
  mkdir -p /dev/pts
  mount -t devpts devpts /dev/pts
  
  # Optional: run mdev
  /sbin/mdev -s

  chmod 0777 etc/init.d/rcS

• Note:

  1. busybox 作为linuxrc启动后， 会读取/etc/profile, 这里面设置了一些环境变量和shell的属性
  2. 根据/etc/fstab提供的挂载信息， 进行文件系统的挂载
  3. busybox 会从 /etc/inittab中读取sysinit并执行， 这里sysinit指向了/etc/init.d/rcS
  4. /etc/init.d/rcS 中 ，mdev -s 这条命令很重要， 它会扫描/sys目录，查找字符设备和块设备，并在/dev下mknod

• dev dir

  cd busybox-1.33.1/_install/dev
  
  sudo mknod -m 622 console c 5 1
  sudo mknod -m 666 null c 1 3
  sudo mknod -m 666 zero c 1 5
  sudo mknod -m 666 tty c 5 0

  Without console, user output can't print

• lib dir

  copy system lib to lib

  cd /code/busybox-1.33.1/_install/lib
  cp /usr/lib/x86_64-linux-gnu/*.so* -a .
  cp /usr/aarch64-linux-gnu/lib/*.so*  -a .

copy the ramfs to /code/ramfs:

cp -r /code/busybox-1.33.1/_install/* /code/ramfs

make diskfs

1. Create an Empty Disk Image

   cd /code
   dd if=/dev/zero of=rootfs.ext4 bs=1M count=512

2. Format the Image

   mkfs.ext4 rootfs.ext4

3. Mount and Populate the Image

   # copy busybox
   mkdir /code/mnt
   sudo mount -o loop rootfs.ext4 /code/mnt
   
   sudo cp -r /path/to/busybox-1.36.1/_install/* /code/mnt/
   
   # copy LTP
   sudo mkdir -p mnt/opt/ltp
   sudo cp -rL /opt/ltp/* mnt/opt/ltp/  # LTP for aarch64

4. Verify Size

   du -sh mnt/  # Before unmounting, if still mounted
   ls -lh rootfs.ext4

5. Unmount

   sudo umount mnt

bash

有些内核测试脚本需要bash才能执行，busybox里面自带的是sh，只能支持posix的标准shell，因此需要自己下载编译bash，并安装到内核的root目录下

wget https://ftp.gnu.org/gnu/bash/bash-5.0.tar.gz
tar xf bash-5.0.tar.gz
cd bash-5.0
./configure --host=aarch64-linux-gnu --prefix=/home/xxx/linux/linux-stable/root
make
make install

fmt

wget http://ftp.gnu.org/gnu/coreutils/coreutils-9.0.tar.xz
tar xf coreutils-9.0.tar.xz
cd coreutils-9.0
./configure --host=aarch64-linux-gnu --disable-nls  --prefix=/home/xxx/linux/linux-stable/root/gnu
make
make install

ready-to-use-x86

# 1. download
wget https://cloud-images.ubuntu.com/jammy/current/jammy-server-cloudimg-amd64-root.tar.xz

sudo tar -xf jammy-server-cloudimg-amd64-root.tar.xz -C initramfs

# 2. Pack as initramfs
cd initramfs

vim init

#!/bin/sh

# Mount essential filesystems
mount -t devtmpfs devtmpfs /dev
mount -t proc proc /proc
mount -t sysfs sysfs /sys
mount -t tmpfs tmpfs /tmp

# Run the real init (systemd)
exec /sbin/init

chmod +x init

# enale password-less login
cd initramfs
# etc/shadow
change <root:*:20382:0:99999:7:::> to <root::20382:0:99999:7:::>

find . | cpio -H newc -o | gzip > ../initramfs.img

# 3. Boot QEMU
qemu-system-x86_64 \
    -m 65536 \
    -smp 112 \
    -kernel ~/code/linux/arch/x86/boot/bzImage \
    -initrd initramfs.img \
    -append "console=ttyS0" \
    -nographic

make kernel

• How to use the Microsoft Linux kernel v6 on Windows Subsystem for Linux version 2 (WSL2)

config

generate .config by arch/arm64/configs/defconfig

make defconfig ARCH=arm64

• .config for ramfs:

  CONFIG_DEBUG_INFO=y
  CONFIG_INITRAMFS_SOURCE="/code/ramfs" # the root dir of the ramfs
  CONFIG_INITRAMFS_ROOT_UID=0
  CONFIG_INITRAMFS_ROOT_GID=0
  CONFIG_DEBUG_SECTION_MISMATCH=y # prohibit inline while debuging

• .config for diskfs

  CONFIG_BLK_DEV_INITRD=n
  CONFIG_INITRAMFS_SOURCE=""
  CONFIG_EXT4_FS=y
  CONFIG_VIRTIO_BLK=y

# 1. General Perf Support
CONFIG_PERF_EVENTS=y
CONFIG_HW_PERF_EVENTS=y
CONFIG_FTRACE=y
CONFIG_FUNCTION_TRACER=y
CONFIG_STACKTRACE=y
CONFIG_KALLSYMS=y
CONFIG_KALLSYMS_ALL=y
CONFIG_DEBUG_FS=y

# Optional (recommended for more metrics/tools):
CONFIG_TRACING=y
CONFIG_SCHED_TRACER=y
CONFIG_FTRACE_SYSCALLS=y
CONFIG_FUNCTION_GRAPH_TRACER=y
CONFIG_TRACEPOINTS=y

# 2. eBPF Core Support, For loading and running BPF programs:
CONFIG_BPF=y
CONFIG_BPF_SYSCALL=y
CONFIG_BPF_JIT=y
CONFIG_HAVE_EBPF_JIT=y
CONFIG_CGROUP_BPF=y
CONFIG_BPF_EVENTS=y
CONFIG_BPF_LSM=y        # if you want LSM hooks
CONFIG_KPROBE_EVENTS=y
CONFIG_UPROBE_EVENTS=y
CONFIG_NET=y            # networking needed for tc/xdp

# 3. Networking + XDP (optional, if using eBPF for packet filtering)
CONFIG_BPF_JIT=y
CONFIG_NETFILTER_XT_MATCH_BPF=m
CONFIG_NETFILTER_XT_TARGET_BPF=m
CONFIG_XDP_SOCKETS=y
CONFIG_BPF_STREAM_PARSER=y

# 4. Helpers for bcc / bpftrace
CONFIG_IKHEADERS=m

# 5. Debugging / Development (optional but useful)
CONFIG_DEBUG_INFO=y
CONFIG_DEBUG_INFO_BTF=y
CONFIG_DEBUG_INFO_DWARF4=y
CONFIG_DEBUG_INFO_REDUCED=n
CONFIG_DEBUG_INFO_SPLIT=n

make

time make -j$(nproc) ARCH=arm64 Image  CROSS_COMPILE=aarch64-linux-gnu-

To speed up compile we only build Image, exclude modules.

Just press Enter to accept the default when encounter a choice.

start

# with ramfs as root fs
qemu-system-aarch64 \
    -m 4096M -smp 8 \
    -cpu cortex-a57 \
    -machine virt \
    -kernel ./arch/arm64/boot/Image \
    -append "rdinit=/linuxrc nokaslr console=ttyAMA0 loglevel=8" \
    -nographic

qemu-system-x86_64 \
    -machine q35 \
    -cpu max \
    -m 8196 \
    -kernel "/code/kern/arch/x86/boot/bzImage" \
    -nic user,model=virtio-net-pci \
    -append "nokaslr console=ttyS0 loglevel=8" \
    -nographic

# terminate qemu
ctl + a, x

# with diskfs as root fs
qemu-system-aarch64 \
    -m 4096M -smp 8 \
    -cpu cortex-a57 \
    -machine virt \
    -kernel arch/arm64/boot/Image \
    -hda /code/rootfs.ext4 \
    -append "root=/dev/vda rw nokaslr console=ttyAMA0 loglevel=8" \
    -nographic

gdb

vs code installs c/c++ plugin

qemu-system-aarch64 \
    -m 4096M -smp 8 -cpu cortex-a57 \
    -machine virt -kernel arch/arm64/boot/Image \
    -append "rdinit=/linuxrc nokaslr console=ttyAMA0 loglevel=8" \
    -nographic -S -gdb tcp::6688

./vscode/launch.json:

{
    // Use IntelliSense to learn about possible attributes.
    // Hover to view descriptions of existing attributes.
    // For more information, visit: https://go.microsoft.com/fwlink/?linkid=830387
    "version": "0.2.0",
    "configurations": [
        {
            "name": "kernel debug",
            "type": "cppdbg",
            "request": "launch",
            "program": "/code/linux/vmlinux",
            "cwd": "${workspaceFolder}",
            "MIMode": "gdb",
            "miDebuggerPath":"/usr/bin/gdb-multiarch",
            "miDebuggerServerAddress": "localhost:6688",
            "stopAtEntry": true,
            "setupCommands": [
                {
                    "description": "enable pretty print for gdb",
                    "text": "-enable-pretty-printing",
                    "ignoreFailures": true
                }
            ]
        }
    ]
}

or config "configurations" to xxx.code-workspace:

{
    "launch": {
        "version": "0.2.0",
        "configurations": [
            ...
        ]
    }
}

cleanup

# Removes most generated files (e.g., object files, binaries) but keeps the configuration (.config)
make clean

# Removes all generated files, including the .config file and backup files, returning the source tree to a pristine state
make mrproper

# Goes even further than mrproper by removing editor backup files, patch files, and other miscellaneous leftovers.
make distclean

make on macOS

• Kernel Programming Template for macOS + arm64 + VS Code
• Compiling Linux kernel on macOS
• Building Linux kernel on macOS natively

1. make a case-sensitive partiton

   Open Disk Utility.app and create an APFS (Case-sensitive) partition. You don't need to specify the size because it will scale automatically.

2. install tools and config environment

   Bee Headers provides a set of headers including byteswap.h, elf.h, and endian.h that are required to build the Linux kernel. This project allows easy installation of these headers via Homebrew.

   # brew tap bee-headers/bee-headers
   brew tap bee-headers/bee-headers
   brew install bee-headers/bee-headers/bee-headers
   
   brew install make llvm lld findutils gnu-sed libelf qemu
   

   # ~/.zshrc
   export OPENSSL_ROOT=$(brew --prefix openssl)
   # prefer llvm rather than default clang provided by xcode
   export PATH="$(brew --prefix llvm)/bin:$PATH"

3. make kernel

   make defconfig ARCH=arm64 LLVM=1
   
   # source with every new shell:
   source bee-init
   time make Image -j$(nproc) ARCH=arm64

4. start qemu

   Download the Arch Linux for UTM

   # -nic vmnet-shared needs root privilege
   sudo qemu-system-aarch64 \
       -machine virt \
       -cpu max \
       -m 8196 \
       -drive file=/Volumes/code/ArchLinux.utm/Data/BB208CBD-BFB4-4895-9542-48527C9E5473.qcow2,format=qcow2 \
       -kernel "/Volumes/code/linux/arch/arm64/boot/Image" \
       -nic vmnet-shared \
       -append "root=/dev/vda2" -nographic

hvp

• 在 MacOS 上做 Linux 内核开发的一些有趣经验和坑

brew install libvirt
sudo brew services start libvirt
sudo virtlogd -d

mkdir -p /Users/jemmy/.config/libvirt/qemu/nvram/
cp /opt/homebrew/Cellar/qemu/10.0.3/share/qemu/edk2-arm-vars.fd /Users/jemmy/.config/libvirt/qemu/nvram/Arch_VARS.fd

# arch-linux.xml is in the same path with make-kernel.md
sudo virsh define /Users/jemmy/.config/libvirt/arch-linux.xml
sudo virsh undefine Arch --nvram

sudo virsh list --all

sudo virsh start Arch
sudo virsh shutdown Arch
sudo virsh destroy Arch

sudo virsh console Arch

# mount shared directory
sudo mount -t 9p code -o trans=virtio /code
# -t 9p: Specifies the 9p filesystem.
# code: The target name from the XML.
# -o trans=virtio: Uses the virtio transport for performance.
# /code: The local mount point in the guest.

mount | grep 9p
df -h /code
ls /code

# persistent the mount, add the cfg to /etc/fstab
code /code 9p trans=virtio,nofail 0 0

# unmount
lsof /code
sudo fuser -km /code
sudo umount /code

ltp

git clone https://github.com/linux-test-project/ltp.git
cd ltp
make autotools
./configure
make -j$(nproc)
sudo make install # Installs to /opt/ltp on your workstation

# Copy LTP into Rootfs
sudo cp -r /opt/ltp/* /code/kern/root/ltp

sudo /opt/ltp/runltp -f sched
sudo /opt/ltp/bin/ltp -t rt_sched

mail config

Configuring Custom SMTP Settings for Gmail

1. In your Google/Gmail account, go to svgexport-10Settings - See all settings.
2. Select the Forwarding and POP/IMAP tab.
3. Under the IMAP access section, toggle on the option to Enable IMAP.

Enable 2-Step Verification

1. Navigate to your Google Account: App Passwords
2. Name your app Accredible, then click the Create button.
3. A new app password will be generated for you. Copy this password.

.gitconfig

# /etc/gitconfig  # system-wide config
[b4]
    midmask = https://lore.kernel.org/%s
    thread = true
    save-to = /var/cache/b4/%s.mbox

[user]
    name = xxx
    email = xxx@gmail.com

[sendemail]
	smtpServer = smtp.gmail.com
	smtpServerPort = 587
	smtpEncryption = tls
	smtpUser = xxx@gmail.com
	smtpPass = xxx
	suppresscc = self
	confirm = always
    # only enable if send mail by msmtp
    sendmailCmd = /usr/bin/msmtp
[https]
	proxy = http://localhost:7890
[http]
	proxy = http://localhost:7890

msmtprc

sudo apt-get install msmtp
sudo apt install git-email

chmod 600 ~/.msmtprc
# allow gmail 587 port
sudo iptables -A OUTPUT -p tcp --dport 587 -j ACCEPT

# ~/.msmtprc
account default
host smtp.gmail.com
port 587
auth on
user xxx@gmail.com
password xxx
from xxx@gmail.com
tls on
tls_starttls on
logfile ~/.msmtp.log

send patch

• https://blog.xzr.moe/archives/293/

• https://tinylab.org/mailing-list-intro/

• send at first time

  cd linux
  
  git add xxx.c
  git commit -s
  git format-patch -<n> -v<n> origin --cover-letter
  
  ./scripts/checkpatch.pl  vx-000x-name.patch
  ./scripts/get_maintainer.pl vx-000x-name.patch
  
  git send-email --to=xxx@gmail.com --cc=xxx@gmail.com vx-000x-name.patch

• send at subsequent time

  git add fs/namespace.c
  git commit -s --amend --no-edit
  git format-patch -<n> -v<n> origin
  
  ./scripts/checkpatch.pl  v2-0001-xxx.patch v2-0002-xxx.patch
  
  git send-email \
  --in-reply-to=<message-id> \
  --to=dnlplm@xxx.yyy \
  --cc=bjorn@xxx.yyy \
  --subject='Re: [PATCH net 1/1] xxx' \
  ./v2-0001-xxx.patch ./v2-0002-xxx.patch

• -<n>: The latest n commits are all patched

• --cover-latter: Generate a [PATCH 0/n] introduction for the latest n patches

• --annotate: Reply code comment

• --subject-prefix="PATCH <MOD NAME>"

• --in-reply-to: Message id can be found at https://lore.kernel.org/all/ according to the patch name

git cmd

git rebase -i HEAD~<n>
# amend code
git add .
git commit --amend
git rebase --continue

repo init -u https://android.googlesource.com/platform/manifest -b master
repo sync -j8
repo status
repo diff
repo start <branch-name> --all
repo branches

repo upload .
repo list
repo info
repo forall -c 'git clean -fdx'

vpn

download

• clash-linux-amd64-v1.18.0.gz - For x86 / x64 Linux
• clash-linux-arm64-v1.18.0.gz - For ARM Linux
• clash-darwin-amd64-v1.18.0.gz - For intel macOS
• clash-darwin-arm64-v1.18.0.gz - For Apple silicon macOS

gzip -dk clash-*-*-v1.18.0.gz
chmod +x clash-*-*-v1.18.0
cp clash-*-*-v1.18.0 /usr/local/bin/clash

mkdir /usr/local/etc/clash
cd /usr/local/etc/clash
# download proxy cfg
wget https://***.*/feeds/***/clash.yml
# rename
mv clash.yml ./config.yaml
# download Country.mmdb file
wget https://github.com/SukkaW/Koolshare-Clash/raw/refs/heads/master/koolclash/koolclash/config/Country.mmdb

# run clash
clash -d /usr/local/etc/clash # default 127.0.0.1:7890

Test

curl --proxy http://localhost:7890 https://kernel.org

curl --proxy http://localhost:7890 https://github.com

config

• config terminal

  # ~/.bashrc
  export socks5='socks5://127.0.0.1:7891'
  export http_proxy='http://127.0.0.1:7890'
  export https_proxy='http://127.0.0.1:7890'

• config github

  # ~/.ssh/config
  Host github.com
      Hostname github.com
      ProxyCommand nc -x localhost:7890 %h %p
      # git-for-windows can replace nc with connect
      # ProxyCommand connect -S localhost:1085 %h %p
  
  Host kernel.org
      Hostname kernel.org
      ProxyCommand nc -x localhost:7890 %h %p

• config apt

  # /etc/apt/apt.conf.d/95proxies
  Acquire::http::Proxy "http://127.0.0.1:7890/";
  Acquire::https::Proxy "http://127.0.0.1:7890/";
  
  # update apt cache
  sudo apt-get update

clash.service

# sudo vim /etc/systemd/system/clash.service
[Unit]
Description=Clash VPN Daemon
After=network.target

[Service]
Type=simple
ExecStart=/usr/local/bin/clash -d /usr/local/etc/clash
Restart=on-failure
RestartSec=20s
StartLimitIntervalSec=60s
StartLimitBurst=3
User=nobody
Group=nogroup

[Install]
WantedBy=multi-user.target

sudo systemctl daemon-reload
sudo systemctl enable clash.service
sudo systemctl start clash.service
sudo systemctl status clash.service
sudo systemctl stop clash.service

share macOS vpn to qemu

1. surge enables Enhanced Mode
2. macOS enables shareing network: setting -> general -> sharing -> internet sharing: from surge to thunderbolt bridge
3. add arg to qemu cmd: -nic vmnet-shared

ssh

sudo apt install openssh-server
sudo service ssh status
sudo service ssh start
sudo nano /etc/ssh/sshd_config
sudo systemctl restart sshd.service

sudo passwd root # set root user pwd

ssh [username]@[public_IP] -p[port_number]

Avoid SSH timeout from the server

# /etc/ssh/sshd_config
TCPKeepAlive no
ClientAliveInterval 300
ClientAliveCountMax 2400

PermitRootLogin yes # allow root user ssh login

Avoid SSH timeout from the client

# ~/.ssh/config
ServerAliveInterval 300

ubuntu on mac

• Linux support for Apple devices with the T2 security chip
• Getting Wi-Fi and Bluetooth to work
• Install drivers for the fan (if not working automatically or want to force a certain speed)

Remove GUI

sudo apt purge ubuntu-desktop gnome-shell gdm3
sudo apt autoremove

# Set the system to boot into multi-user (non-GUI) mode:
sudo systemctl set-default multi-user.target

Power Management:

Install tlp to optimize battery life:

sudo apt install tlp
sudo systemctl enable tlp

network

static ip:

# /etc/netplan/01-netcfg.yaml
# /etc/netplan/90-NM-e3f02c4a-7917-42a7-811c-c01f902752db.yaml
network:
  version: 2
  renderer: NetworkManager
  wifis:
    wlp3s0:
      addresses:
        - 192.168.1.100/24
      gateway4: 192.168.1.1
      nameservers:
        addresses: [192.168.1.1, 114.114.114.114, 8.8.8.8, 8.8.4.4]
      access-points:
        "Jemmy_5G":
          auth:
            key-management: "psk"
            password: "jm123456"
      networkmanager:
        uuid: "e3f02c4a-7917-42a7-811c-c01f902752db"
        name: "Jemmy_5G"

If using networkd, disable NetworkManager to avoid conflicts:

sudo systemctl disable NetworkManager
sudo systemctl stop NetworkManager # ssh will disconnect
sudo systemctl enable systemd-networkd
sudo systemctl start systemd-networkd

Apply the Configuration

sudo chmod 600 /etc/netplan/01-network-manager-all.yaml

sudo netplan generate
sudo netplan apply

Verify the Static IP

ip addr show wlp3s0
ping 8.8.8.8
ping google.com

Firewall (UFW):

sudo apt install ufw
sudo ufw allow OpenSSH
sudo ufw enable

screen off

gsettings get org.gnome.desktop.session idle-delay

# set screen turn off after 30s inactivity
gsettings set org.gnome.desktop.session idle-delay 30

sudo sh -c 'echo 30 > /sys/module/kernel/parameters/consoleblank'
# To make this persistent
sudo nano /etc/default/grub
# Modify or add the consoleblank parameter in the GRUB_CMDLINE_LINUX_DEFAULT line:
GRUB_CMDLINE_LINUX_DEFAULT="quiet splash consoleblank=30"

sudo update-grub

Troubleshooting

• The brightness of the screen is not fully dark, or screen shows nothing after reboot:

  # restart display-manager
  systemctl restart display-manager

• errors in systemctl status display-manager

  systemctl status display-manager
  
  gkr-pam: couldn't unlock the login keyring.
  gdm3[4388]: Gdm: on_display_added: assertion 'GDM_IS_REMOTE_DISPLAY (display)' failed
  gdm3[4388]: Gdm: on_display_removed: assertion 'GDM_IS_REMOTE_DISPLAY (display)' failed

  • Fix Keyring Issue

    rm -rf ~/.local/share/keyrings/*
    
    # /etc/pam.d/gdm-password and /etc/pam.d/gdm-launch-environment
    auth    optional    pam_gnome_keyring.so
    session optional    pam_gnome_keyring.so auto_start
    
    sudo systemctl restart display-manager

  • Address GDM Assertion Failures

    # /etc/gdm3/custom.conf
    [daemon]
    # Ensure Wayland or Xorg is correctly set
    # If you’re not using Wayland, set WaylandEnable=false and test.
    WaylandEnable=true
    
    sudo systemctl restart display-manager

  • Reinstall GDB

    If the above steps don’t resolve the issue, reinstall GDM:

    sudo apt install --reinstall gdm3
    sudo systemctl restart display-manager

  • Fallback to Another Display Manager

    sudo apt install lightdm
    sudo systemctl disable gdm3
    sudo systemctl enable lightdm
    sudo systemctl start lightdm

change boot menuentry

look up entry index in /boot/grub/grub.cfg

menuentry 'Ubuntu                                                       # 0
submenu 'Advanced options for Ubuntu' $menuentry_id_option
    menuentry 'Ubuntu, with Linux 6.15.0-rc2+'                          # 1>0
    menuentry 'Ubuntu, with Linux 6.15.0-rc2+ (recovery mode)'          # 1>1
    menuentry 'Ubuntu, with Linux 6.11.0-21-generic'                    # 1>2
    menuentry 'Ubuntu, with Linux 6.11.0-21-generic (recovery mode)'    # 1>3
menuentry "Memory test (memtest86+x64.efi)"                             # 2
menuentry 'Memory test (memtest86+x64.efi, serial console)'             # 3

set defualt entry in /etc/default/grub

GRUB_DEFAULT="1>2" # set entry to "menuentry 'Ubuntu, with Linux 6.11.0-21-generic'"

update-grub

reboot

prevent auto sleep

sudo vim /etc/systemd/logind.conf

# Controls the action when the lid is closed (on battery or AC).
# HandleLidSwitch=suspend
HandleLidSwitch=ignore

# Applies when the laptop is on AC power.
# HandleLidSwitchExternalPower=suspend
HandleLidSwitchExternalPower=ignore

# Applies when connected to a dock or external display (if applicable).
HandleLidSwitchDocked=ignore

# turn off screen, max 65535
echo 0 > /sys/class/backlight/gmux_backlight/brightness

cat /sys/class/backlight/gmux_backlight/max_brightness

sudo systemctl restart systemd-logind
sudo reboot

frp

• GitHub Doc

• download: https://github.com/fatedier/frp/releases

  cd /usr/local/bin
  wget xxxx
  gunzip xxx # tar -zvf xxx
  
  # mv frpc to client: /usr/local/bin
  # mv frps to server: /usr/local/bin

server

• config frps.toml

  mkdir -p /usr/local/etc/frps
  vim /usr/local/etc/frpc/frps.toml

  [common]
  # frp listen port
  bind_port = 7000
  token = 52010
  
  # frp management port
  dashboard_port = 7500
  dashboard_user = admin
  dashboard_pwd = admin
  enable_prometheus = true
  
  # frp log config
  log_file = /var/log/frps.log
  log_level = info
  log_max_days = 3

• create frps.service

  sudo vim /etc/systemd/system/frps.service

  [Unit]
  # service name
  Description = frp server
  After = network.target syslog.target
  Wants = network.target
  
  [Service]
  Type = simple
  ExecStart = /usr/local/bin/frps -c /usr/local/etc/frps/frps.toml
  Restart=on-failure
  RestartSec=20s
  StartLimitIntervalSec=60s
  StartLimitBurst=3
  
  [Install]
  WantedBy = multi-user.target

• enable service

  sudo systemctl daemon-reload
  sudo systemctl enable frps
  sudo systemctl start frps

client

• config frpc.toml

  mkdir -p /usr/local/etc/frpc
  vim /usr/local/etc/frpc/frpc.toml

  serverAddr = "139.196.161.94"
  # serverPort is used for communication between frps and frpc
  serverPort = 7000
  
  [[proxies]]
  name = "ssh"
  type = "tcp"
  localIP = "127.0.0.1"
  # localPort (listened on the client) and remotePort (exposed on the server)
  # are used for traffic going in and out of the frp system
  localPort = 22
  remotePort = 6000

• config frpc.service

  sudo vim /etc/systemd/system/frpc.service

  [Unit]
  # service name
  Description = frp client
  After = network.target syslog.target
  Wants = network.target
  
  [Service]
  Type = simple
  ExecStart = /usr/local/bin/frpc -c /usr/local/etc/frpc/frpc.toml
  Restart=on-failure
  RestartSec=20s
  StartLimitIntervalSec=60s
  StartLimitBurst=3
  
  [Install]
  WantedBy = multi-user.target

• enable service

  sudo systemctl daemon-reload
  sudo systemctl enable frpc
  sudo systemctl start frpc

test

ssh -o Port=6000 user@server-ip-addr

# ~/.ssh/config
Host mbp2018-remote
  HostName server-ip-addr
  User root
  Port 6000

vscode

gtags

# macOs
brew install global

# ubuntu
sudo apt install global

VS Code install C/C++ GNU Global plugin

VS code config:

# ubuntu
"gnuGlobal.gtagsExecutable": "/usr/bin/gtags",
"gnuGlobal.globalExecutable": "/usr/bin/global",

# macOS M chip
"gnuGlobal.gtagsExecutable": "/opt/homebrew/bin/gtags",
"gnuGlobal.globalExecutable": "/opt/homebrew/bin/global",

# macOs Intel chip
"gnuGlobal.gtagsExecutable": "/usr/local/bin/gtags",
"gnuGlobal.globalExecutable": "/usr/local/bin/global",

Usge:

# execute cmd in the root directory of the code
gtags -i

# enter F12 in vs code to goto the definition of symbol

gtags.conf

# ~/.globalrc
default:\
    :tc=native:

native:\
    :tc=gtags:

common:\
    :skip=out/,*.pid,*.sock:

gtags:\
    :tc=common:

# gtags --gtagsconf=~/.globalrc

update ubuntu kernel

# cp kernel config from current system
cp /boot/config-$(uname -r) /code/ubuntu/.config

# reset config in .config
CONFIG_SYSTEM_TRUSTED_KEYS=""
CONFIG_SYSTEM_REVOCATION_KEYS=""

make -j$(nproc) && make modules -j$(nproc)

sudo make modules_install && sudo make install && sudo update-grub && sudo reboot

uname -a

key	value
Linux	Kernel name: Indicates the operating system kernel is Linux, managing hardware and resources.
jemmy	Hostname: The unique name of the machine, "jemmy," used for network or local identification.
6.15.0-rc2+	Kernel version: Breaks down as 6.15.0 (major.minor.patch), rc2 (Release Candidate 2, a pre-release test version), + (custom patches or modifications applied).
#10	Build number: The 10th build of this kernel configuration on this system.
SMP	Symmetric Multiprocessing: Kernel supports multiple CPU cores for parallel processing.
PREEMPT_RT	Preempt Real-Time: Kernel is patched for real-time capabilities, prioritizing low-latency task scheduling for time-critical applications.
Sun Apr 20 14:59:36 CST 2025	Build timestamp: When the kernel was compiled (April 20, 2025, at 14:59:36 China Standard Time).
x86_64	CPU architecture (running): The kernel is running on a 64-bit x86 processor.
x86_64	Compiled architecture: The kernel was compiled for a 64-bit x86 architecture.
x86_64	User-space architecture: User programs interact with the system as 64-bit applications.
GNU/Linux	OS flavor: Indicates a Linux system using GNU userland tools (standard utilities and libraries).

new partition

# Identify the new partition:
sudo fdisk -l

# Format the partition as ext4:
sudo mkfs.ext4 /dev/nvme0n1p3

# Verify the filesystem:
sudo fsck /dev/nvme0n1p3

# Temporarily mount the new partition to test:
sudo mount /dev/nvme0n1p3 /code

# Verify the mount:
df -h /code

# Adjust permissions:
sudo chown root:root /code
sudo chmod u+rw /code

# Make the Mount Permanent
# Get the UUID of the new partition:
lsblk -f

# Edit /etc/fstab:
sudo nano /etc/fstab

# Add an entry:
UUID=4b363ad7-7792-4bdf-b2cb-069a8c278e4e /code ext4 defaults 0 2
# defaults: Standard mount options.
# 0 2: No fsck for non-root, check order 2.

# Test the fstab, If no errors, the configuration is correct.
sudo mount -a

# Verify:
df -h /code

build perf

cd tools/perf

make clean
make ARCH=arm64 CROSS_COMPILE=aarch64-linux-gnu- \
     CC="aarch64-linux-gnu-gcc" \
     CXX="aarch64-linux-gnu-g++" \
     NO_SDT=1 NO_SLANG=1 NO_PERL=1 NO_PYTHON=1 \
     LIBELF_FOUND=yes LIBTRACEEVENT_FOUND=yes LIBCRYPTO_FOUND=yes \
     -j$(nproc)

# copy perf to rootfs
cp perf usr/bin

# enable related kernel configs

Download any missing *.so dependencies from archlinuxarm.org, mv the *.so to usr/lib since perf is dynamic linked.