Tips for optimizing the system

System boot

To find out which process is slowing down your system startup the most, run the systemd-analyze blame command

Let me guess, NetworkManager-wait-online has turned out to be a traitor and is taking ~4 seconds off your system startup? No problem. You can just disable it)

sudo systemctl mask NetworkManager-wait-online.service

Speeding up initramfs unpacking

This is the initial boot environment that completes the Linux kernel image. To save space, it is presented as a self-extracting archive. In Arch Linux, the mkinitcpio utility creates initramfs and by default uses the zstd compression algorithm. To speed up loading, it is better to use the lz4 algorithm.

1. Change compression algorithm

   Edit the /etc/mkinitcpio.conf file, adding the following lines:

   COMPRESSION="lz4"
   COMPRESSION_OPTIONS=(-9)

2. Replace hooks with systemd (optional)

   For additional acceleration, replace base and udev hooks with systemd:

   HOOKS=(systemd autodetect microcode modconf kms keyboard sd-vconsole block filesystems fsck)

   Attention

   For systems with an encrypted root partition, add sd-encrypt after the sd-vconsole hook.

3. Update initramfs images

   After making changes, run the command:

   sudo mkinitcpio -P

Compilation flags

Compilation flags are pointers for the compiler on what instructions and optimizations to use when building programs.

To optimize the build process, we can change them by creating a custom ~/.makepkg.conf config in the home directory to override system settings:

CFLAGS="-march=native -mtune=native -O2 -pipe -fno-plt -fexceptions \
      -Wp,-D_FORTIFY_SOURCE=3 -Wformat -Werror=format-security \
      -fstack-clash-protection -fcf-protection"
CXXFLAGS="$CFLAGS -Wp,-D_GLIBCXX_ASSERTIONS"
RUSTFLAGS="-C opt-level=3 -C target-cpu=native -C link-arg=-z -C link-arg=pack-relative-relocs"
MAKEFLAGS="-j$(nproc) -l$(nproc)"

Important

These compiler flags maximize compilation performance, but may cause build errors in very rare applications. If this happens, disable the ‘lto’ parameter in the options line by adding an exclamation mark character in front of it ! (“!lto”).

Ccache

Linux has programs that can take more than two hours to build, and ccache can be used to speed up the recompilation of applications such as Wine or Proton-GE.

Ccache is a cache for C/C++ compilers, compatible with GCC and Clang, that speeds up recompilation of the same code. If identical blocks of source code are found when building a new version of a program, ccache uses already compiled code from the cache, which greatly speeds up the compilation process.

To install it, run the following commands:

sudo pacman -S ccache

After installation, it still needs to be activated in your makepkg settings. To do this, edit the ~/.makepkg.conf configuration file by adding the following configurations:

BUILDENV=(!distcc color ccache check !sign)

Warning

ccache can break the build of some programs, so be careful with its use.

TRIM

This is a command used in solid state drives (SSDs) that allows the operating system to tell the drive which data blocks are no longer needed and can be cleared. This helps maintain SSD performance by preventing the SSD from slowing down when writing data.

To enable it, run the following commands:

sudo systemctl enable fstrim.timer
sudo fstrim -v /

Btrfs Users

If you use the Btrfs file system and have kernel version 6.2 or higher, you do not need to enable the service to periodically execute the TRIM command, because Btrfs itself executes it asynchronously.

OOM

Out-Of-Memory Killer (OOM) is a Linux process that terminates an application to save the kernel from crashing. It sacrifices the application to keep the OS running.

To enable it, run the command:

sudo systemctl enable --now systemd-oomd

Ananicy CPP

Ananicy is a task prioritization daemon that greatly improves system responsiveness. We will install Ananicy CPP, rewritten in C++ for increased performance.

1. Install Ananicy CPP

   Clone and build the package from AUR:

   git clone https://aur.archlinux.org/ananicy-cpp.git
   cd ananicy-cpp
   makepkg -sric

2. Enable the service

   Enable and start the service:

   sudo systemctl enable --now ananicy-cpp

3. Install additional rules (recommended)

   Install extended rules for better optimization:

   git clone https://aur.archlinux.org/cachyos-ananicy-rules-git.git
   cd cachyos-ananicy-rules-git
   makepkg -sric

4. Restart the service

   Restart the service to apply new rules:

   sudo systemctl restart ananicy-cpp

Haveged

I can also recommend the package Haveged. This is a daemon that monitors the entropy of the system. It is needed to speed up system startup at high systemd-analyze blame (More than 1 second).

To install it, run the following commands:

sudo pacman -S haveged
sudo systemctl enable haveged

Rng-tools

Rng-tools - a daemon that also monitors system entropy, but, unlike haveged, via a hardware timer. It is needed to speed up system startup at high systemd-analyze blame (More than 1 second).

To install it, run the following commands:

sudo pacman -S rng-tools
sudo systemctl enable --now rngd

Warning

Not to be used in conjunction with Ananicy CPP

Cpu Power

By default, the processor dynamically changes its frequency, which provides a balance between power saving and performance. However, for maximum performance, you can lock the maximum frequency mode.

1. Install CPUPower utility

   sudo pacman -S cpupower

2. Set performance mode

   sudo cpupower frequency-set -g performance

3. Configure settings

   Open the configuration file:

   sudo nano /etc/default/cpupower

   Find and edit the line:

   governor='performance'

4. Enable the service

   sudo systemctl enable cpupower

Auto CpuFreq

This is a utility for automatically optimizing CPU speed and power consumption. It monitors battery status, CPU utilization, temperature and system load, dynamically switching between power saving and high performance modes.

Recommendation

This is the best option for most users.

1. Install from AUR

   yay -S auto-cpufreq

2. Enable the service

   sudo systemctl enable --now auto-cpufreq

Optimizing kernel loading

You can change the kernel boot parameters in GRUB a bit, change the following line in /etc/default/grub adding the following parameters to your own:

GRUB_CMDLINE_LINUX_DEFAULT="... loglevel=2 nowatchdog split_lock_detect=off processor.ignore_ppc=1 migrations=off msr.allow_writes=on pcie_aspm=force module.sig_unenforce cryptomgr.notests initcall_debug no_timer_check noreplace-smp page_alloc.shuffle=1 rcupdate.rcu_expedited=1 tsc=reliable ..."

And finally update grub

sudo grub-mkconfig -o /boot/grub/grub.cfg

Careful!

Changes to kernel boot parameters in GRUB can affect security and overall system behavior. Do so at your own risk.

Impact on Safety:

• Disabling safety mechanisms: nowatchdog and split_lock_detect=off reduce protection against hangs
• Ignoring checks: module.sig_unenforce allows loading unsigned modules

Impact on Performance:

• System optimization: pcie_aspm=force and page_alloc.shuffle=1 improve performance
• Stability: tsc=reliable provides better time synchronization

You want more?

I suggest you visit this repo. It contains all the ways to optimize Arch linux.