Archlinux installation process on Dell XPS 15 7590

2023-11-27T00:00:00+00:00
I am once more installing Archlinux after my SSD fried. I had documented the process last time I did this in a list of bullet points, this article meant to formalize it a bit more.
This is mostly targeted at my future self, so all the software, config, and procedure will be tailored to my needs. Other people might find some value in it though, so I'm putting it out there.
A lot of the information in here is also available in the Archlinux installation guide</a>.
Create a live USB</h2>
We first need to prepare a bootable USB with the latest iso image, which you can find here</a>.
Do make sure to verify the checksum of the image after downloading it. While it's rare, I have been burned beforeby a corrupted image, and it's not fun to debug issues for hours before realizing that the error was in the very first step, and was easily avoidable. Check the output of this command against the checksum on the download page (search for SHA256):
sha256sum</span></span> archlinux-2023.11.01-x86_64.iso</span>
</span></code></pre>
Then plug in your USB stick, make sure it doesn't contain important data, and figure out
its device name somehow.
I used sudo dmesg</code>, which should show the device name around the time you plugged it in:</p>
</p>
In my case, my device is named /dev/sda</code>.</p>




Note the distinction between the device name and the partition names.
In my case, the device name is /dev/sda</code>, and the partition names are /dev/sda1</code> and
/dev/sda2</code>.</p>
</div>

Then, use dd</code> to write the iso image to the USB stick:</p>
#</span></span> make sure the USB stick is not mounted</span>
</span></span>sudo</span></span> umount /dev/sda*</span></span>
</span>#</span></span> note that the dd command targets the device, not one of its partitions</span>
</span></span>sudo</span></span> dd if=archlinux-2023.11.01-x86_64.iso of=/dev/sda status=progress</span>
</span>#</span></span> make sure all the data is properly written</span>
</span></span>sync</span></span>
</span></code></pre>
Boot from the USB stick</h2>
You can then extract the USB stick and plug it in the computer where you want to install
Archlinux.
Restart the computer, and adjust your BIOS settings if necessary:</p>

disable secure boot</li>
configure the SATA controller to use AHCI mode</li>
</ul>
You can then boot the USB stick.
Set your keyboard layout and time zone:</p>
loadkeys</span></span> fr-latin1</span>
</span>timedatectl</span></span> set-timezone Europe/Oslo</span>
</span></code></pre>
Connect to the internet:</p>
#</span></span> check whether your network interface is up</span>
</span></span>rfkill</span></span>
</span>ip</span></span> link</span>
</span>#</span></span> turn it on if it's not</span>
</span></span>ip</span></span> link set dev wlan0 up</span>
</span>#</span></span> connect to your wifi</span>
</span></span>iwctl</span></span>
</span>
</span>#</span></span> iwctl commands:</span>
</span></span>#</span></span> device list</span>
</span></span>#</span></span> station wlan0 scan</span>
</span></span>#</span></span> station wlan0 get-networks</span>
</span></span>#</span></span> station wlan0 connect <SSID></span>
</span></span>#</span></span> station wlan0 show</span>
</span></span>
</span>#</span></span> check that you're connected</span>
</span></span>ping</span></span> archlinux.org</span>
</span></code></pre>




Pro tip: tab completion works on the SSID, you don't need to type it all out.</p>
</div>

Get a GUI</h2>
I don't like using fdisk or parted to manage my partitions, I'd much rather use gparted,
which is a graphical interface for parted.
I also don't want to type out all of the install commands, I prefer to copy-paste them.
For these reasons, we'll get a GUI up and running as soon as possible.</p>
#</span></span> resize the root partition on the fly to make room for packages</span>
</span></span>mount</span></span> -</span>o</span> remount,size=4G /run/archiso/cowspace</span>
</span>#</span></span> refresh the package list</span>
</span></span>pacman</span></span> -</span>Syy</span></span>
</span>#</span></span> install a display server, a light weight desktop environment and a browser</span>
</span></span>pacman</span></span> -</span>S</span> xorg-server xorg-xinit xfce4 firefox</span>
</span>#</span></span> configure Xorg to start xfce</span>
</span></span>echo</span></span> "</span>exec startxfce4"</span></span> >></span> ~</span></span>/.xinitrc</span>
</span>#</span></span> start the graphical session</span>
</span></span>startx</span></span>
</span></code></pre>
From there, we can use the browser to copy commands into the terminal.</p>
If, for some reason, you need to get back to the tty, you can use Ctrl+Alt+F2</code>.</p>
#</span></span> set the keyboard layout in the X session</span>
</span></span>setxkbmap</span></span> fr</span>
</span></code></pre>
Partition the disk</h2>
Next, we'll create all of our partitions.
I use Windows on a dual boot (already installed), I'll try and not break it.</p>
Here are the partitions I want:</p>

/boot</code> to contain the linux kernels and initramfs, 1GB</li>
/</code> for the system, 100GB. Should be more than enough, the main reason I'm giving it
that much is for docker stuff that can quickly take up a lot of space</li>
/home</code> for my personal files, 150GB. Separate from /</code> so I don't lose it all if I
ever need to reinstall the system</li>
a swap partition with the same size as my RAM, so I can see things slowing down before
they get oom-killed if I use up all my RAM</li>
/media/data</code> for any big files, formatted in NTFS and shared by Windows and Linux</li>
</ul>
I don't like using fdisk or parted as I find it to easy to break things and hard to keep
track of what I'm doing, so I'll use gparted, which is the main reason why I wanted a
graphical session:</p>
#</span></span> install gparted</span>
</span></span>pacman</span></span> -</span>S</span> gparted</span>
</span>#</span></span> run it</span>
</span></span>gparted</span></span>
</span></code></pre>
</p>
As you can see, Windows already created a few partitions of its own.
We will be adding an entry in the EFI partition to boot Archlinux, and reducing the size
of the main windows partition.
I don't know what the other two partitions are for, but they're tiny so I'll leave them
be.</p>
The first step is to resize the Windows partition.</p>




You should do this from Windows instead, if you don't want to run into the same issues I did.</p>
</div>

I reduced it to 150GB, which is more than it currently needs but leaves some room for it
to grow.
I like to reboot here into Windows to make sure it is still working fine, even if it
means redoing all of the previous preparation steps.
It is usually easier to fix issues as early as possible, rather than do a bunch of
changes at once and not know which one caused the issue.
That was a good shout, because Windows got stuck in a loop of trying to repair disk
issues and rebooting. I solved this by running bootrec /FixMbr</code> in the command prompt
available in Windows' recovery tool, but it is probably better to resize the partition
from the official Windows utility instead (not sure how it would handle resizing the
partition it is currently running on though).
If that doesn't work for you, you could try one of the other solutions suggested
here</a>.</p>
Here is the list of partitions I created, with names and labels always with the same
value:</p>

arch_boot</code>: 2GB, fat32, for the kernels and initramfs. Once created, right click >
manage flags > tick bls_boot</code> to make it an</li>
arch_root</code>: 100GB, ext4, for the system</li>
arch_home</code>: 150GB, ext4, for my personal files</li>
swap</code>: 16GB, same size as RAM</li>
data</code>: everything else, NTFS so Windows can also use it</li>
</ul>
Once the changes are applied, we can reboot to make sure Windows can see and use the new
data partition.</p>
Install the system</h2>
We'll start with mounting all the partitions.</p>
#</span></span> set variables so we don't have to remember which partition is which all the time</span>
</span></span>export</span> PARTITION_EFI</span>=</span>"</span>/dev/nvme0n1p1"</span></span></span></span>
</span>export</span> PARTITION_BOOT</span>=</span>"</span>/dev/nvme0n1p5"</span></span></span></span>
</span>export</span> PARTITION_ROOT</span>=</span>"</span>/dev/nvme0n1p6"</span></span></span></span>
</span>export</span> PARTITION_HOME</span>=</span>"</span>/dev/nvme0n1p7"</span></span></span></span>
</span>export</span> PARTITION_SWAP</span>=</span>"</span>/dev/nvme0n1p8"</span></span></span></span>
</span>export</span> PARTITION_DATA</span>=</span>"</span>/dev/nvme0n1p9"</span></span></span></span>
</span>
</span>#</span></span> mount the partitions</span>
</span></span>mount</span></span> $</span>PARTITION_ROOT</span></span> /mnt</span>
</span>mount</span></span> --</span>mkdir</span> $</span>PARTITION_EFI</span></span> /mnt/efi</span>
</span>mount</span></span> --</span>mkdir</span> $</span>PARTITION_BOOT</span></span> /mnt/boot</span>
</span>mount</span></span> --</span>mkdir</span> $</span>PARTITION_HOME</span></span> /mnt/home</span>
</span></code></pre>
The Archlinux wiki</a> explains in
great detail the various ways to handle the EFI partition.
I'll go for the second typical mount point described there: mounting the EFI partition
to /efi</code>.</p>
Install the base system, after making sure you will be using a mirror close to you:</p>
reflector</span></span> --</span>country</span> '</span>Norway,Sweden'</span></span> ></span> /etc/pacman.d/mirrorlist</span>
</span>pacstrap</span></span> -</span>K</span> /mnt base linux linux-firmware</span>
</span></code></pre>
We'll then use genfstab</code> to make sure all the partitions we just mounted are always
mounted automatically in the new system.</p>
genfstab</span></span> -</span>U</span> /mnt >></span> /mnt/etc/fstab</span>
</span></code></pre>
We can now chroot into the system and configure it more.</p>
#</span></span> chroot into the new system</span>
</span></span>arch-chroot</span></span> /mnt</span>
</span>#</span></span> enable colors in pacman</span>
</span></span>sed</span></span> -</span>i</span> '</span>s/#Color/Color/g'</span></span> /etc/pacman.conf</span>
</span>#</span></span> upgrade everything</span>
</span></span>pacman</span></span> -</span>Syyu</span></span>
</span>#</span></span> update your processor microcode if necessary</span>
</span></span>pacman</span></span> -</span>S</span> intel-ucode</span>
</span>#</span></span> configure the timezone</span>
</span></span>ln</span></span> -</span>sf</span> /usr/share/zoneinfo/Europe/Oslo /etc/localtime</span>
</span>#</span></span> sync your system time with the internet</span>
</span></span>timedatectl</span></span> set-ntp true</span>
</span>#</span></span> set the hardware clock to UTC</span>
</span></span>hwclock</span></span> --</span>systohc</span></span>
</span>#</span></span> uncomment the locale you want to use in this file, and install it</span>
</span></span>#</span></span> en_DK.UTF-8 uses ISO date format (yyyy-mm-dd)</span>
</span></span>sed</span></span> -</span>i</span> '</span>s/#en_DK.UTF-8 UTF-8/en_DK.UTF-8 UTF-8/'</span></span> /etc/locale.gen</span>
</span>sed</span></span> -</span>i</span> '</span>s/#en_US.UTF-8 UTF-8/en_US.UTF-8 UTF-8/'</span></span> /etc/locale.gen</span>
</span>locale-gen</span></span>
</span>echo</span></span> "</span>LANG=en_US.UTF-8"</span></span> ></span> /etc/locale.conf</span>
</span>#</span></span> make your keyboard layout in the console persistent</span>
</span></span>echo</span></span> "</span>KEYMAP=fr-latin1"</span></span> ></span> /etc/vconsole.conf</span>
</span>#</span></span> set your hostname</span>
</span></span>echo</span></span> "</span>maahl-laptop"</span></span> ></span> /etc/hostname</span>
</span>#</span></span> set your root password</span>
</span></span>passwd</span></span>
</span></code></pre>
You might need to set Windows to use
UTC</a> as well.</p>
We'll configure the bootloader with
systemd-boot</a>.
If you would prefer to use something else, the
wiki</a> might have some
documentation.</p>
#</span></span> create your initial ramdisk</span>
</span></span>mkinitcpio</span></span> -</span>P</span></span>
</span>#</span></span> install systemd-boot on your EFI partition</span>
</span></span>bootctl</span></span> install</span>
</span>#</span></span> configure the archlinux boot entry</span>
</span></span>#</span></span> don't forget the microcode line if you installed it</span>
</span></span>echo</span></span> "</span>title Arch Linux
</span></span></span>linux /vmlinuz-linux
</span></span></span>initrd /intel-ucode.img
</span></span></span>initrd /initramfs-linux.img
</span></span></span>options root=LABEL=arch_root"</span></span> ></span> /boot/loader/entries/arch.conf</span>
</span>#</span></span> configure the loader</span>
</span></span>echo</span></span> "</span>timeout 2"</span></span> ></span> /efi/loader/loader.conf</span>
</span></code></pre>




Note that there are two config folders for systemd-boot: /boot/loader</code> and
/efi/loader</code>.
I found this very confusing, but what clarified it for me is this excerpt from the
wiki</a> (with paths adjusted
to the current situation):</p>

systemd-boot will search for boot menu items in /efi/loader/entries/*.conf</code> and
additionally in /boot/loader/entries/*.conf</code> if using XBOOTLDR</code>. Note that entries
in /efi</code> can only use files (e.g. kernels, initramfs, images, etc.) in /efi</code>.
Similarly, entries in /boot</code> can only use files in /boot</code>.</p>
</blockquote>
We are using XBOOTLDR</code> (this is what setting the bts_boot</code> flag on the /boot partition
did at the gparted step), and the kernels will be installed to /boot</code>, so we need to
put our bootloader entry in /boot</code> as well.</p>
</div>

We can now make sure bootctl doesn't report any error:</p>
bootctl</span></span> status</span>
</span></code></pre>
We'll now configure our main user.</p>
#</span></span> install sudo for privilege escalation, the shell that will be used by the user, and</span>
</span></span>#</span></span> vim for visudo</span>
</span></span>pacman</span></span> -</span>S</span> sudo zsh vim</span>
</span>#</span></span> create a group for sudoers</span>
</span></span>groupadd</span></span> sudo</span>
</span>#</span></span> create the user and add it to the sudoers group</span>
</span></span>useradd</span></span> -</span>m</span> -</span>G</span> sudo,wheel -</span>s</span> /usr/bin/zsh maahl</span>
</span>#</span></span> set the user's password</span>
</span></span>passwd</span></span> maahl</span>
</span>#</span></span> make members of the sudo group part of the sudoers</span>
</span></span>#</span></span> uncomment this line:</span>
</span></span>#</span></span> %sudo ALL=(ALL:ALL) ALL</span>
</span></span>visudo</span></span>
</span></code></pre>
Now on to installing some packages.</p>
For my window manager, I use i3, mainly for the capability to assign workspaces to
monitors, and being able to switch workspace on one monitor without anything changing on
the other monitors.</p>
I would have liked to have something with a few more bells and whistles, but I didn't
manage to adjust the alternatives I investigated to my needs (Gnome Shell with the
gnome-shell-extension-dash-to-panel extension, and Hyprland).</p>
#</span></span> install a backup text editor</span>
</span></span>pacman</span></span> -</span>S</span> nano</span>
</span>#</span></span> install i3 and related tools</span>
</span></span>pacman</span></span> -</span>S</span> \
</span></span></span>  i3 i3lock polybar rofi picom dunst \
</span></span></span>  arandr autorandr rofi-autorandr \
</span></span></span>  lightdm lightdm-slick-greeter lightdm-settings</span>
</span>sudo</span></span> sed -</span>i</span> '</span>s/#greeter-session=example-gtk-gnome/greeter-session=lightdm-slick-greeter/'</span></span> /etc/lightdm/lightdm.conf</span>
</span>systemctl</span></span> enable lightdm</span>
</span>#</span></span> install some graphical utilities</span>
</span></span>pacman</span></span> -</span>S</span> firefox wezterm nemo gparted</span>
</span>#</span></span> install network utilities</span>
</span></span>pacman</span></span> -</span>S</span> networkmanager networkmanager-openvpn network-manager-applet</span>
</span>systemctl</span></span> enable NetworkManager</span>
</span>#</span></span> install some command line tools</span>
</span></span>pacman</span></span> -</span>S</span> base-devel git inetutils wget efibootmgr</span>
</span></code></pre>
When given a choice about audio stuff, I went with those:</p>

pipewire-jack</li>
wireplumber</li>
</ul>
I also like to disable the locking system</a>
that prevents auth after 3 failed attempts, especially since I tend to juggle between
keyboard layouts.</p>
sudo</span></span> sed -</span>i</span> '</span>s/# deny = 3/deny = 0/g'</span></span> /etc/security/faillock.conf</span>
</span></code></pre>
We can now reboot into Windows first to make sure we didn't break anything, then to our
new system to finish the configuration.</p>
We haven't configured a display manager yet, so after rebooting into Linux, you'll be
greeted with a tty.
After login, you can just run Hyprland</code> to start it.</p>
Install dotfiles</h2>
I'll first start by fetching my dotfiles, to prevent software I install from creating
them before I created the right symlinks.</p>
#</span></span> set your keyboard layout</span>
</span></span>sudo</span></span> localectl set-keymap fr</span>
</span>sudo</span></span> localectl set-x11-keymap fr</span>
</span>#</span></span> install oh-my-zsh and a plugin for it</span>
</span></span>sh</span></span> -</span>c</span> "</span>$</span>(</span>curl</span></span> -</span>fsSL</span> https://raw.github.com/ohmyzsh/ohmyzsh/master/tools/install.sh</span>)</span></span>"</span></span></span> &&</span> rm</span></span> ~</span></span>/.zshrc</span>
</span>git</span></span> clone https://github.com/zsh-users/zsh-autosuggestions $</span>{</span></span>ZSH_CUSTOM</span></span>:-</span></span>~</span></span>/.oh-my-zsh/custom</span>}</span></span>/plugins/zsh-autosuggestions</span>
</span>#</span></span> generate an ssh key</span>
</span></span>ssh-keygen</span></span> -</span>t</span> ed25519 -</span>C</span> "</span>[your@email]"</span></span></span>
</span>#</span></span> add the public key to your git platform</span>
</span></span>cat</span></span> ~</span></span>/.ssh/id_ed25519.pub</span>
</span>#</span></span> clone your config files repo</span>
</span></span>git</span></span> clone [</span>ssh://your-</span>repo]</span></span>
</span>#</span></span> create symlinks to the config files</span>
</span></span>cd</span></span> config-files</span> &&</span> ./init-config.sh</span></span>
</span>#</span></span> install a vim plugin manager, and install the plugins</span>
</span></span>git</span></span> clone https://github.com/VundleVim/Vundle.vim.git ~</span></span>/.vim/bundle/Vundle.vim</span>
</span>vim</span></span> +PluginInstall +qall</span>
</span></code></pre>
Install paru</h2>
In order to install packages from the AUR, I use paru</a>.</p>
sudo</span></span> pacman -</span>S</span> rustup</span>
</span>rustup</span></span> default stable</span>
</span>mkdir</span></span> ~</span></span>/tmp</span> &&</span> cd</span></span> ~</span></span>/tmp</span>
</span>git</span></span> clone https://aur.archlinux.org/paru.git</span> &&</span> cd</span></span> paru</span> &&</span> makepkg</span></span> -</span>si</span></span>
</span>#</span></span> get the most relevant results at the bottom</span>
</span></span>sudo</span></span> sed -</span>i</span> '</span>s/#BottomUp/BottomUp/g'</span></span> /etc/paru.conf</span>
</span></code></pre>
Install other packages</h2>
#</span></span> address some of the "missing firmware" warnings during mkinitcpio</span>
</span></span>paru</span></span> -</span>S</span> mkinitcpio-firmware</span>
</span>#</span></span> power management tools</span>
</span></span>paru</span></span> -</span>S</span> powertop tlp tlp-rdw</span>
</span>sudo</span></span> systemctl enable tlp</span>
</span>sudo</span></span> systemctl enable NetworkManager-dispatcher</span>
</span>sudo</span></span> systemctl mask systemd-rfkill.service</span>
</span>sudo</span></span> systemctl mask systemd-rfkill.socket</span>
</span>#</span></span> some useful command line tools</span>
</span></span>paru</span></span> -</span>S</span> autojump-rs bat diff-so-fancy exa htop man-db man-pages pspg ripgrep starship sysfsutils tldr</span>
</span>#</span></span> vscode, the MS release as I need some extensions that are only on their marketplace</span>
</span></span>paru</span></span> -</span>S</span> visual-studio-code-bin</span>
</span>#</span></span> some development tools</span>
</span></span>paru</span></span> -</span>S</span> aws-session-manager-plugin cmake docker docker-compose jq python-pynvim tfenv tk valgrind</span>
</span>sudo</span></span> usermod -</span>aG</span> docker $</span>USER</span></span></span>
</span>sudo</span></span> usermod -</span>aG</span> tfenv $</span>USER</span></span></span>
</span>#</span></span> python environments manager, so as not to mess with system files</span>
</span></span>curl</span></span> https://pyenv.run</span> |</span> bash</span></span>
</span>pyenv</span></span> install 3.12</span>
</span>pyenv</span></span> global 3.12</span>
</span>#</span></span> some other tools</span>
</span></span>paru</span></span> -</span>S</span> android-tools flameshot peek zola</span>
</span></code></pre>
Set up a backup kernel</h2>
If some kernel update were to break something, it's nice to have a backup option rather
than having to fix the issue from the rescue mode.
To that end, we'll install the long-term support (LTS) version of the kernel, and add a
boot entry for it.</p>
#</span></span> install the lts kernel</span>
</span></span>paru</span></span> -</span>S</span> linux-lts</span>
</span>#</span></span> create the boot entry for it</span>
</span></span>echo</span></span> "</span>title Arch Linux with LTS kernel
</span></span></span>linux /vmlinuz-linux-lts
</span></span></span>initrd /intel-ucode.img
</span></span></span>initrd /initramfs-linux-lts.img
</span></span></span>options root=LABEL=arch_root"</span></span></span> |</span> sudo</span></span> tee /boot/loader/entries/arch-lts.conf</span>
</span></code></pre>
We can then reboot with the new boot entry to confirm that the kernel can be used.</p>
Configure audio</h2>
I'll use Pipewire, which should have been installed earlier as a dependency of another
package.</p>
#</span></span> enable sound, approve removing pulseaudio packages when asked</span>
</span></span>paru</span></span> -</span>S</span> pipewire-pulse</span>
</span>#</span></span> enable bluetooth</span>
</span></span>paru</span></span> -</span>S</span> bluez bluez-utils blueman</span>
</span>sudo</span></span> systemctl enable bluetooth</span>
</span>sudo</span></span> systemctl start bluetooth</span>
</span>sudo</span></span> systemctl enable blueman-mechanism</span>
</span>sudo</span></span> systemctl start blueman-mechanism</span>
</span>#</span></span> to control audio input and output</span>
</span></span>paru</span></span> -</span>S</span> pavucontrol pulseaudio-control</span>
</span></code></pre>
Configure hibernation</h3>
We'll use suspend-then-hibernate, where the system saves its state both to RAM and disk.
This allows us to have the system immediately available if we wake it up before the
hibernate delay, but to preserve more battery if the laptop remains asleep for longer.</p>
First, let's set up hibernation.
We'll use a swap file, so that if we do end up using the swap partition, we can still
hibernate.</p>
#</span></span> create a swap file the size of RAM</span>
</span></span>sudo</span></span> dd if=/dev/zero of=/swapfile bs=1M count=16k</span>
</span>sudo</span></span> chmod 0600 /swapfile</span>
</span>#</span></span> format it</span>
</span></span>sudo</span></span> mkswap /swapfile</span>
</span>#</span></span> enable swap on it</span>
</span></span>sudo</span></span> swapon /swapfile</span>
</span>#</span></span> add it to fstab</span>
</span></span>echo</span></span> "</span># swap file for hibernation
</span></span></span>/swapfile none swap defaults 0 0"</span></span></span> |</span> sudo</span></span> tee -</span>a</span> /etc/fstab</span>
</span>#</span></span> this should now list your swap partition and the newly created file</span>
</span></span>sudo</span></span> swapon -</span>s</span></span>
</span></code></pre>
The kernel needs to know where to check for resume info when waking up from hibernation.
We do so by adding the resume</code> hook to the initramfs and pointing the kernel to the
file.</p>
#</span></span> add `resume` to the HOOKS array, somewhere after udev, eg:</span>
</span></span>#</span></span> HOOKS=(base udev autodetect modconf keyboard keymap consolefont block filesystems resume fsck)</span>
</span></span>sudo</span></span> vim /etc/mkinitcpio.conf</span>
</span>#</span></span> regenerate the initramfs</span>
</span></span>sudo</span></span> mkinitcpio -</span>P</span></span>
</span>#</span></span> remember which partition is your root partition</span>
</span></span>export</span> PARTITION_ROOT</span>=</span>"</span>/dev/nvme0n1p6"</span></span></span></span>
</span>#</span></span> get the UUID of your root partition</span>
</span></span>export</span> PARTITION_ROOT_UUID</span>=</span>$</span>(</span>sudo</span></span> blkid $</span>PARTITION_ROOT</span></span> -</span>s</span> UUID -</span>o</span> value</span>)</span></span></span></span>
</span>#</span></span> get the offset of your swap file</span>
</span></span>export</span> SWAP_FILE_OFFSET</span>=</span>$</span>(</span>sudo</span></span> filefrag -</span>v</span> /swapfile</span> |</span> awk</span></span> '</span>$1=="0:" {print substr($4, 1, length($4)-2)}'</span></span></span>)</span></span></span></span>
</span>#</span></span> output the kernel options you will need</span>
</span></span>echo</span></span> "</span>resume=UUID=$</span>{</span></span>PARTITION_ROOT_UUID</span></span>}</span></span> resume_offset=$</span>{</span></span>SWAP_FILE_OFFSET</span></span>}</span></span>"</span></span></span>
</span>#</span></span> copy the output of the previous command at the end of the options line in this file</span>
</span></span>#</span></span> this will tell the kernel where to resume from</span>
</span></span>sudo</span></span> vim /boot/loader/entries/arch.conf</span>
</span></code></pre>
After a reboot, you should be able to hibernate and resume.</p>
#</span></span> set up a short hibernate delay for testing</span>
</span></span>sudo</span></span> sed -</span>i</span> '</span>s/#HibernateDelaySec=/HibernateDelaySec=60/g'</span></span> /etc/systemd/sleep.conf</span>
</span>#</span></span> put the computer to sleep, wake it up immediately and note how long it gets to wake up</span>
</span></span>systemctl</span></span> suspend</span>
</span>#</span></span> set it to suspend then hibernate, wait a bit longer than the delay you set earlier,</span>
</span></span>#</span></span> and wake the laptop again</span>
</span></span>systemctl</span></span> suspend-then-hibernate</span>
</span></code></pre>
Don't forget to set the HibernateDelaySec</code> parameter to a more sensible value once
you're done with tests:</p>
sudo</span></span> sed -</span>i</span> '</span>s/HibernateDelaySec=60/HibernateDelaySec=3600/g'</span></span> /etc/systemd/sleep.conf</span>
</span></code></pre>
Now I want my computer to automatically suspend-then-hibernate when I close it:</p>
#</span></span> change all three "LidSwitch" actions</span>
</span></span>sudo</span></span> sed -</span>i</span> '</span>s/#HandleLidSwitch=suspend/HandleLidSwitch=suspend-then-hibernate/g'</span></span> /etc/systemd/logind.conf</span>
</span>sudo</span></span> sed -</span>i</span> '</span>s/#HandleLidSwitchExternalPower=suspend/HandleLidSwitchExternalPower=suspend-then-hibernate/g'</span></span> /etc/systemd/logind.conf</span>
</span>sudo</span></span> sed -</span>i</span> '</span>s/#HandleLidSwitchDocked=ignore/HandleLidSwitchDocked=suspend-then-hibernate/g'</span></span> /etc/systemd/logind.conf</span>
</span>#</span></span> make sure clicking on the power button by mistake does turn everything off</span>
</span></span>sudo</span></span> sed -</span>i</span> '</span>s/#HandlePowerKey=poweroff/HandlePowerKey=suspend-then-hibernate/g'</span></span> /etc/systemd/logind.conf</span>
</span></code></pre>
I also want my system to be locked whenever I suspend my laptop.
I achieved this with a systemd unit file:</p>
echo</span></span> "</span>Description=User suspend actions
</span></span></span>Before=sleep.target
</span></span></span>
</span></span></span>[Service]
</span></span></span>User=%I
</span></span></span>Type=forking
</span></span></span>Environment=DISPLAY=:0
</span></span></span>ExecStart=/usr/bin/i3lock -f -c 444444
</span></span></span>ExecStartPost=/usr/bin/sleep 1
</span></span></span>
</span></span></span>[Install]
</span></span></span>WantedBy=sleep.target"</span></span></span> |</span> sudo</span></span> tee /etc/systemd/system/suspend@.service</span>
</span>sudo</span></span> systemctl enable suspend@$</span>USER</span></span>.service</span>
</span></code></pre>
Configure the GPU</h2>
The GPU on this laptop is an Nvidia GeForce GTX 1650 Mobile.</p>
To save battery, I want the GPU to be enabled only when I explicitly require it, and to
use the integrated graphics the rest of the time.
This can be done with Prime</a>, and we'll be
using the nvidia proprietary driver, to avoid some potential issues with nouveau.</p>
Configure the driver</h3>
#</span></span> install some utils</span>
</span></span>paru</span></span> -</span>S</span> mesa-utils</span>
</span>#</span></span> check what gpu is currently in use, intel graphics in my case</span>
</span></span>glxinfo</span></span> |</span> grep</span></span> "</span>OpenGL renderer"</span></span></span>
</span>#</span></span> install the drivers for the usual kernel and the lts one</span>
</span></span>paru</span></span> -</span>S</span> nvidia nvidia-lts</span>
</span>#</span></span> remove "kms" from the HOOKS array in that file, so as not to load nouveau</span>
</span></span>sudo</span></span> vim /etc/mkinitcpio.conf</span>
</span>#</span></span> regenerate the initramfs</span>
</span></span>sudo</span></span> mkinitcpio -</span>P</span></span>
</span>#</span></span> configure the driver</span>
</span></span>echo</span></span> "</span>options nvidia_drm modeset=1 fbdev=1"</span></span></span> |</span> sudo</span></span> tee /etc/modprobe.d/nvidia-drm.conf</span>
</span>echo</span></span> "</span>options nvidia NVreg_DynamicPowerManagement=0x02
</span></span></span>blacklist nouveau"</span></span></span> |</span> sudo</span></span> tee /etc/modprobe.d/nvidia.conf</span>
</span>#</span></span> create a udev rule to turn off the GPU when not in use</span>
</span></span>echo</span></span> '</span># Enable runtime PM for NVIDIA VGA/3D controller devices on driver bind
</span></span></span>ACTION=="bind", SUBSYSTEM=="pci", ATTR{vendor}=="0x10de", ATTR{class}=="0x030000", TEST=="power/control", ATTR{power/control}="auto"
</span></span></span>ACTION=="bind", SUBSYSTEM=="pci", ATTR{vendor}=="0x10de", ATTR{class}=="0x030200", TEST=="power/control", ATTR{power/control}="auto"
</span></span></span>
</span></span></span># Disable runtime PM for NVIDIA VGA/3D controller devices on driver unbind
</span></span></span>ACTION=="unbind", SUBSYSTEM=="pci", ATTR{vendor}=="0x10de", ATTR{class}=="0x030000", TEST=="power/control", ATTR{power/control}="on"
</span></span></span>ACTION=="unbind", SUBSYSTEM=="pci", ATTR{vendor}=="0x10de", ATTR{class}=="0x030200", TEST=="power/control", ATTR{power/control}="on"
</span></span></span>'</span></span></span> |</span> sudo</span></span> tee /etc/udev/rules.d/80-nvidia-pm.rules</span>
</span></code></pre>
Reboot and make sure both the usual kernel and the lts one allow you to boot your system
properly.</p>
Use the GPU when desired</h3>
We can then setup GPU switching with prime:</p>
paru</span></span> -</span>S</span> nvidia-prime</span>
</span>#</span></span> make sure you can now use the discrete graphics card on demand</span>
</span></span>#</span></span> this should say intel</span>
</span></span>glxinfo</span></span> |</span> grep</span></span> "</span>OpenGL renderer"</span></span></span>
</span>#</span></span> this should say nvidia</span>
</span></span>prime-run</span></span> glxinfo</span> |</span> grep</span></span> "</span>OpenGL renderer"</span></span></span>
</span></code></pre>
You can now check that the GPU gets properly suspended when nothing uses it:</p>
#</span></span> check that this says suspended</span>
</span></span>cat</span></span> /sys/bus/pci/devices/0000:01:00.0/power/runtime_status</span>
</span>#</span></span> use the GPU, this should say "nvidia"</span>
</span></span>prime-run</span></span> glxinfo</span> |</span> grep</span></span> "</span>OpenGL renderer"</span></span></span>
</span>#</span></span> this should say "active" for about 10 seconds after the end of the previous command,</span>
</span></span>#</span></span> then "suspended" when the GPU is turned back off</span>
</span></span>cat</span></span> /sys/bus/pci/devices/0000:01:00.0/power/runtime_status</span>
</span></code></pre>
Keep in mind that utilities such as nvidia-smi</code> will power the card on even if it is
off, for about 10 seconds.</p>
If you use an app launcher that reads .desktop entries, you can run a program on the GPU
by default (if run with xdg-open or anything that reads the .desktop entry) by editing
the corresponding .desktop entry to use prime-run.</p>
For instance, here is how I did it for Factorio:</p>
sed</span></span> -</span>i</span> '</span>s#Exec="/home/maahl/games/factorio/start.sh" ""#Exec="prime-run" "/home/maahl/games/factorio/start.sh"#'</span></span> ~</span></span>/.local/share/applications/gog_com-Factorio_1.desktop</span>
</span></code></pre>
The change will be taken into account next time the desktop entry is reloaded, for
instance after logging out and back in.</p>
At any given time, you can run nvidia-smi</code> to see what processes are using the GPU.</p>
There is technically a PrefersNonDefaultGPU</code> field in the .desktop
specification</a>,
but the only program I've seen that respects it is Gnome Shell with switcherooctl (not
sure which program is responsible for what there).
It's also limited in that if your default GPU is the discrete GPU, this won't behave the
same as setups that use the integrated GPU by default.</p>





    
        </i>
        Preserving video memory allocations
    </div>
    
        With the configuration described above, there is an issue where if a program is using
the GPU when the laptop goes to sleep, it will have glitches when it resumes. This is
because the nvidia driver only saves essential video memory allocations, and has to
ditch the rest.</p>
</p>
This
page</a>
describes two methods that are supposed to fix the issue:
NVreg_PreserveVideoMemoryAllocations</code> and NVreg_EnableS0ixPowerManagement</code>.
I'll describe both here in case you want to try them out, but neither fixed the issue
for me, so I resigned myself to have to restart any program running on the GPU if I need
to hibernate.</p>
The first method using the NVreg_PreserveVideoMemoryAllocations</code> parameter:</p>
#</span></span> allow turning off the GPU entirely when it's not in use</span>
</span></span>echo</span></span> "</span>options nvidia NVreg_DynamicPowerManagement=0x02 NVreg_PreserveVideoMemoryAllocations=1 NVreg_TemporaryFilePath=/var/tmp
</span></span></span>blacklist nouveau"</span></span></span> |</span> sudo</span></span> tee /etc/modprobe.d/nvidia.conf</span>
</span>#</span></span> enable nvidia services</span>
</span></span>sudo</span></span> systemctl enable nvidia-suspend</span>
</span>sudo</span></span> systemctl enable nvidia-hibernate</span>
</span>sudo</span></span> systemctl enable nvidia-resume</span>
</span></code></pre>
This works for suspend and hibernate, however it doesn't for suspend-then-hibernate, and
fails as if the services weren't enabled:</p>
#</span></span> after a reboot so the kernel module param is taken into account</span>
</span></span>systemctl</span></span> suspend-then-hibernate</span>
</span></code></pre>
NVRM: GPU 0000:01:00.0: PreserveVideoMemoryAllocations module parameter is set. System Power Management attempted without driver procfs suspend interface. Please refer to the 'Configuring Power Management Support' section in the driver README.
</span></code></pre>
There is a dirty workaround described
here</a>,
that consists in replicating the nvidia services' functionality manually.</p>
The solution described in that post requires overriding a script that is managed by the
nvidia package, so I'll instead make a separate script that only handles the pre</code> case.</p>
#</span></span> disable the nvidia services</span>
</span></span>sudo</span></span> systemctl disable nvidia-suspend</span>
</span>sudo</span></span> systemctl disable nvidia-hibernate</span>
</span>sudo</span></span> systemctl disable nvidia-resume</span>
</span>#</span></span> replicate their functionality</span>
</span></span>echo</span></span> '</span>#!/bin/sh
</span></span></span>
</span></span></span>case "$1" in
</span></span></span>    pre)
</span></span></span>        sleep 4
</span></span></span>        case "$SYSTEMD_SLEEP_ACTION" in
</span></span></span>            suspend|hibernate)
</span></span></span>                /usr/bin/nvidia-sleep.sh "$SYSTEMD_SLEEP_ACTION"
</span></span></span>                ;;
</span></span></span>            suspend-after-failed-hibernate)
</span></span></span>                /usr/bin/nvidia-sleep.sh "suspend"
</span></span></span>                ;;
</span></span></span>        esac
</span></span></span>        ;;
</span></span></span>esac
</span></span></span>'</span></span></span> |</span> sudo</span></span> tee /usr/lib/systemd/system-sleep/nvidia-suspend-then-hibernate-fix</span>
</span>sudo</span></span> chmod +x /usr/lib/systemd/system-sleep/nvidia-suspend-then-hibernate-fix</span>
</span></code></pre>
I still had artifacts upon resuming in programs running on the GPU, so
I tried the S0ix method instead.
Note that this method would only work if your GPU supports it and your system's suspend state is
s2idle.</p>
#</span></span> remove the script we just created</span>
</span></span>sudo</span></span> rm /usr/lib/systemd/system-sleep/nvidia-suspend-then-hibernate-fix</span>
</span>#</span></span> use another method of preserving video memory</span>
</span></span>echo</span></span> "</span>options nvidia NVreg_DynamicPowerManagement=0x02 NVreg_EnableS0ixPowerManagement=1
</span></span></span>blacklist nouveau"</span></span></span> |</span> sudo</span></span> tee /etc/modprobe.d/nvidia.conf</span>
</span></code></pre>
While this fixed the issue for suspend, this didn't solve the glitches after hibernation either, so I just
scrapped the idea and will just restart any program that needs the GPU after I
hibernate.</p>
I'm not sure if that makes any sense, but just in case I tried combining both methods:</p>
#</span></span> configure both video memory preservation methods:</span>
</span></span>echo</span></span> "</span>options nvidia NVreg_DynamicPowerManagement=0x02 NVreg_EnableS0ixPowerManagement=1 NVreg_PreserveVideoMemoryAllocations=1 NVreg_TemporaryFilePath=/var/tmp
</span></span></span>blacklist nouveau"</span></span></span> |</span> sudo</span></span> tee /etc/modprobe.d/nvidia.conf</span>
</span>#</span></span> enable nvidia services</span>
</span></span>sudo</span></span> systemctl enable nvidia-suspend</span>
</span>sudo</span></span> systemctl enable nvidia-hibernate</span>
</span>sudo</span></span> systemctl enable nvidia-resume</span>
</span></code></pre>
Sadly I still got glitches, so I abandoned altogether the idea of recovering GPU
programs from hibernation.</p>
I ran this to return to the state before this section:</p>
#</span></span> same as the first time</span>
</span></span>echo</span></span> "</span>options nvidia NVreg_DynamicPowerManagement=0x02
</span></span></span>blacklist nouveau"</span></span></span> |</span> sudo</span></span> tee /etc/modprobe.d/nvidia.conf</span>
</span>#</span></span> disable the nvidia services</span>
</span></span>sudo</span></span> systemctl disable nvidia-suspend</span>
</span>sudo</span></span> systemctl disable nvidia-hibernate</span>
</span>sudo</span></span> systemctl disable nvidia-resume</span>
</span></code></pre>
</div>
    </div>
</div>






    
        </i>
        Early loading nvidia drivers
    </div>
    
        I had initially set the nvidia driver and modules to early load, by adding them to
mkinitcpio.conf</code>. However, this caused a number of issues with hibernation, that are
described here</a>.</p>
I'll copy here the steps I initially took to make that work, but keep in mind this won't
play nicely with hibernation.</p>
#</span></span> add "nvidia nvidia_modeset nvidia_uvm nvidia_drm" to the MODULES array, and add</span>
</span></span>#</span></span> the udev rule to the initramfs, by adding /etc/udev/rules.d/80-nvidia-pm.rules</span>
</span></span>#</span></span> to the FILES array</span>
</span></span>sudo</span></span> vim /etc/mkinitcpio.conf</span>
</span>#</span></span> regenerate the initramfs</span>
</span></span>sudo</span></span> mkinitcpio -</span>P</span></span>
</span>#</span></span> make sure the initramfs is always regenerated after a driver update</span>
</span></span>echo</span></span> "</span>[Trigger]
</span></span></span>Operation=Install
</span></span></span>Operation=Upgrade
</span></span></span>Operation=Remove
</span></span></span>Type=Package
</span></span></span>Target=nvidia
</span></span></span>
</span></span></span>[Action]
</span></span></span>Description=Update NVIDIA module in initcpio
</span></span></span>Depends=mkinitcpio
</span></span></span>When=PostTransaction
</span></span></span>NeedsTargets
</span></span></span>Exec=/usr/bin/mkinitcpio -P"</span></span></span> |</span> sudo</span></span> tee /etc/pacman.d/hooks/nvidia.hook</span>
</span></code></pre>
</div>
    </div>
</div>

Update firmware</h2>
We'll install firmware updates if any are available.
Note that you must be connected to AC power for many of them.</p>
#</span></span> install a firmware updater and a GUI for it</span>
</span></span>paru</span></span> -</span>S</span> fwupd gnome-firmware</span>
</span>#</span></span> download metadata</span>
</span></span>fwupdmgr</span></span> refresh</span>
</span>#</span></span> check for available updates</span>
</span></span>fwupdmgr</span></span> get-updates</span>
</span>#</span></span> install updates, if any</span>
</span></span>fwupdmgr</span></span> update</span>
</span></code></pre>
Other misc configuration</h2>
Enable a thunderbolt device:</p>
#</span></span> authorize a thunderbolt device</span>
</span></span>paru</span></span> -</span>S</span> bolt</span>
</span>boltctl</span></span> list</span>
</span>boltctl</span></span> enroll [</span>uuid-</span>of-</span>your-</span>device]</span></span>
</span></code></pre>
Themes, fonts, icons:</p>
#</span></span> install more fonts</span>
</span></span>paru</span></span> -</span>S</span> \
</span></span></span>    adobe-source-code-pro-fonts \
</span></span></span>    adobe-source-han-sans-otc-fonts \
</span></span></span>    adobe-source-sans-fonts \
</span></span></span>    adobe-source-serif-fonts \
</span></span></span>    awesome-terminal-fonts \
</span></span></span>    noto-fonts-emoji \
</span></span></span>    ttf-material-design-icons-git \
</span></span></span>    nerd-fonts \
</span></span></span>    ttf-liberation</span>
</span>#</span></span> install more gtk themes</span>
</span></span>paru</span></span> -</span>S</span> gnome-themes-extra</span>
</span>#</span></span> install nicer-looking cursors</span>
</span></span>paru</span></span> -</span>S</span> vimix-cursors</span>
</span>#</span></span> install nicer-looking icons</span>
</span></span>paru</span></span> -</span>S</span> obsidian-icon-theme</span>
</span>#</span></span> install a utility to switch gtk themes</span>
</span></span>paru</span></span> -</span>S</span> lxappearance-gtk3</span>
</span>lxappearance</span></span>
</span>#</span></span> make sure autorandr is run when lightdm starts</span>
</span></span>sudo</span></span> ln -</span>s</span> ~</span></span>/.config/autorandr /etc/xdg/autorandr</span>
</span>sudo</span></span> sed -</span>i</span> '</span>s/#display-setup-script=/display-setup-script=autorandr --change/'</span></span> /etc/lightdm/lightdm.conf</span>
</span></code></pre>
Wallpaper stuff:</p>
#</span></span> install a wallpaper utility</span>
</span></span>paru</span></span> -</span>S</span> feh</span>
</span>#</span></span> set a wallpaper on login, and change it once in a while:</span>
</span></span>echo</span></span> '</span>
</span></span></span># random wallpaper every once in a while
</span></span></span>set $set_wallpaper "/usr/bin/feh --bg-fill --randomize --recursive ~/wallpapers"
</span></span></span>exec --no-startup-id systemd-run --on-calendar="*-*-* *:00:00" --user -- $set_wallpaper
</span></span></span>exec --no-startup-id $set_wallpaper
</span></span></span>'</span></span> >></span> ~</span></span>/.config/i3/config</span>
</span>#</span></span> customize the login screen, note that wallpapers should not be in /home</span>
</span></span>sudo</span></span> mkdir -</span>p</span> /usr/share/backgrounds</span>
</span>sudo</span></span> chmod go+w /usr/share/backgrounds</span>
</span>cp</span></span> my-wallpaper.png /usr/share/backgrounds</span>
</span>sudo</span></span> lightdm-settings</span>
</span>slick-greeter</span></span> --</span>test-mode</span></span>
</span></code></pre>
Install Steam:</p>
sudo</span></span> sed -</span>z</span> -</span>i</span> '</span>s|#\[multilib\]\n#Include|\[multilib\]\nInclude|g'</span></span> /etc/pacman.conf</span>
</span>paru</span></span> -</span>Syyu</span></span>
</span>paru</span></span> -</span>S</span> lib32-nvidia-utils steam</span>
</span></code></pre>
Conclusion</h2>
I'm a mere beginner in Rust and AWS, and I've sometimes struggled to find documentation on how to piece some components together. I hope this blog post will fill in some of the gaps for other people, and I'm very open to feedback and suggestions for improvement.</p>

maahl.net

Grafana stack part 0: intro

Archlinux installation process on Dell XPS 15 7590

Create a Lambda in Rust using Terraform

pg_explain_lexer: `EXPLAIN ANALYZE` syntax highlighting

pg_dropbuffers extension

PostgreSQL cardinality estimation quality

PostgreSQL awesome resources

maahl.net

Grafana stack part 0: intro

Archlinux installation process on Dell XPS 15 7590

Boot from the USB stick</h2> You can then extract the USB stick and plug it in the computer where you want to install Archlinux. Restart the computer, and adjust your BIOS settings if necessary:</p>

Install the system</h2> We'll start with mounting all the partitions.</p>

Set up a backup kernel</h2> If some kernel update were to break something, it's nice to have a backup option rather than having to fix the issue from the rescue mode. To that end, we'll install the long-term support (LTS) version of the kernel, and add a boot entry for it.</p>

Configure audio</h2> I'll use Pipewire, which should have been installed earlier as a dependency of another package.</p>

Use the GPU when desired</h3> We can then setup GPU switching with prime:</p>

Create a Lambda in Rust using Terraform

Conclusion</h2> I'm a mere beginner in Rust and AWS, and I've sometimes struggled to find documentation on how to piece some components together. I hope this blog post will fill in some of the gaps for other people, and I'm very open to feedback and suggestions for improvement.</p>

pg_explain_lexer: `EXPLAIN ANALYZE` syntax highlighting

pg_dropbuffers extension

PostgreSQL cardinality estimation quality

PostgreSQL awesome resources

Conclusion</h2>
I'm a mere beginner in Rust and AWS, and I've sometimes struggled to find documentation on how to piece some components together. I hope this blog post will fill in some of the gaps for other people, and I'm very open to feedback and suggestions for improvement.</p>
