On this page, after performing a clean ubuntu install we are instructed to install the system76 repository and drivers via apt.

However my galago pro's wifi seems to be unavailable in this clean install and hence I can't yet use apt (at least, not directly without first enabling some other internet connection.)

I assume these lines in my dmesg are related. If my situation is typical of a clean install on this model (and if the solution is in fact found within the drivers we're trying to get) then it seems we may need new instructions to work around this chicken & egg issue.

$ sudo dmesg | grep iwlwifi
[  265.288061] iwlwifi 0000:00:14.3: TLV_FW_FSEQ_VERSION: FSEQ Version: 43.2.23.17
[  265.288064] iwlwifi 0000:00:14.3: Found debug destination: EXTERNAL_DRAM
[  265.288065] iwlwifi 0000:00:14.3: Found debug configuration: 0
[  265.288271] iwlwifi 0000:00:14.3: loaded firmware version 48.13675109.0 op_mode iwlmvm
[  265.362405] iwlwifi 0000:00:14.3: Detected Intel(R) Wireless-AC 9560, REV=0x354
[  265.369192] iwlwifi 0000:00:14.3: Applying debug destination EXTERNAL_DRAM
[  265.369589] iwlwifi 0000:00:14.3: Allocated 0x00400000 bytes for firmware monitor.
[  266.389990] iwlwifi 0000:00:14.3: Collecting data: trigger 15 fired.
[  266.390065] iwlwifi 0000:00:14.3: Start IWL Error Log Dump:
[  266.390067] iwlwifi 0000:00:14.3: Status: 0x00000000, count: -1987606070
[  266.390068] iwlwifi 0000:00:14.3: Loaded firmware version: 48.13675109.0
[  266.390069] iwlwifi 0000:00:14.3: 0x2F9C089A | ADVANCED_SYSASSERT          
[  266.390069] iwlwifi 0000:00:14.3: 0x256B5677 | trm_hw_status0
[  266.390070] iwlwifi 0000:00:14.3: 0xD27F3757 | trm_hw_status1
[  266.390071] iwlwifi 0000:00:14.3: 0x6ABDC029 | branchlink2
[  266.390072] iwlwifi 0000:00:14.3: 0xC096312C | interruptlink1
[  266.390072] iwlwifi 0000:00:14.3: 0x3312DB99 | interruptlink2
[  266.390073] iwlwifi 0000:00:14.3: 0x87DFD4E4 | data1
[  266.390074] iwlwifi 0000:00:14.3: 0x08FDA76F | data2
[  266.390074] iwlwifi 0000:00:14.3: 0x55A0507D | data3
[  266.390075] iwlwifi 0000:00:14.3: 0x4890E468 | beacon time
[  266.390076] iwlwifi 0000:00:14.3: 0x0387AE8D | tsf low
[  266.390077] iwlwifi 0000:00:14.3: 0xA3719EC9 | tsf hi
[  266.390077] iwlwifi 0000:00:14.3: 0x78635F59 | time gp1
[  266.390078] iwlwifi 0000:00:14.3: 0x61797522 | time gp2
[  266.390079] iwlwifi 0000:00:14.3: 0x4B20D912 | uCode revision type
[  266.390079] iwlwifi 0000:00:14.3: 0xDD03A702 | uCode version major
[  266.390080] iwlwifi 0000:00:14.3: 0x1A25C79A | uCode version minor
[  266.390081] iwlwifi 0000:00:14.3: 0xC861B4D3 | hw version
[  266.390081] iwlwifi 0000:00:14.3: 0x4C55C566 | board version
[  266.390082] iwlwifi 0000:00:14.3: 0xB3C1EA76 | hcmd
[  266.390083] iwlwifi 0000:00:14.3: 0xAF99C2F3 | isr0
[  266.390083] iwlwifi 0000:00:14.3: 0xB87BF372 | isr1
[  266.390085] iwlwifi 0000:00:14.3: 0x4EA41D3F | isr2
[  266.390085] iwlwifi 0000:00:14.3: 0x3A94D381 | isr3
[  266.390086] iwlwifi 0000:00:14.3: 0xE763FA53 | isr4
[  266.390087] iwlwifi 0000:00:14.3: 0x0CD884CE | last cmd Id
[  266.390087] iwlwifi 0000:00:14.3: 0xDA73871B | wait_event
[  266.390088] iwlwifi 0000:00:14.3: 0x2363DDCF | l2p_control
[  266.390089] iwlwifi 0000:00:14.3: 0x4CB7AE90 | l2p_duration
[  266.390089] iwlwifi 0000:00:14.3: 0x28C218B7 | l2p_mhvalid
[  266.390090] iwlwifi 0000:00:14.3: 0x89425401 | l2p_addr_match
[  266.390091] iwlwifi 0000:00:14.3: 0x9111B146 | lmpm_pmg_sel
[  266.390092] iwlwifi 0000:00:14.3: 0x6063D16E | timestamp
[  266.390092] iwlwifi 0000:00:14.3: 0xBAB682D7 | flow_handler
[  266.390122] iwlwifi 0000:00:14.3: Start IWL Error Log Dump:
[  266.390123] iwlwifi 0000:00:14.3: Status: 0x00000000, count: 7
[  266.390124] iwlwifi 0000:00:14.3: 0x201013F1 | ADVANCED_SYSASSERT
[  266.390125] iwlwifi 0000:00:14.3: 0x00000000 | umac branchlink1
[  266.390125] iwlwifi 0000:00:14.3: 0xC008CF5C | umac branchlink2
[  266.390126] iwlwifi 0000:00:14.3: 0x00000000 | umac interruptlink1
[  266.390127] iwlwifi 0000:00:14.3: 0x00000000 | umac interruptlink2
[  266.390128] iwlwifi 0000:00:14.3: 0x00000003 | umac data1
[  266.390128] iwlwifi 0000:00:14.3: 0x20000302 | umac data2
[  266.390129] iwlwifi 0000:00:14.3: 0x01300202 | umac data3
[  266.390130] iwlwifi 0000:00:14.3: 0x00000030 | umac major
[  266.390130] iwlwifi 0000:00:14.3: 0x13675109 | umac minor
[  266.390131] iwlwifi 0000:00:14.3: 0x00005CF2 | frame pointer
[  266.390132] iwlwifi 0000:00:14.3: 0xC0887F58 | stack pointer
[  266.390132] iwlwifi 0000:00:14.3: 0x00000000 | last host cmd
[  266.390133] iwlwifi 0000:00:14.3: 0x00000000 | isr status reg
[  266.390147] iwlwifi 0000:00:14.3: Fseq Registers:
[  266.390149] iwlwifi 0000:00:14.3: 0x00000003 | FSEQ_ERROR_CODE
[  266.390151] iwlwifi 0000:00:14.3: 0x00000000 | FSEQ_TOP_INIT_VERSION
[  266.390153] iwlwifi 0000:00:14.3: 0x993C007E | FSEQ_CNVIO_INIT_VERSION
[  266.390156] iwlwifi 0000:00:14.3: 0x0000A384 | FSEQ_OTP_VERSION
[  266.390158] iwlwifi 0000:00:14.3: 0xF7E26237 | FSEQ_TOP_CONTENT_VERSION
[  266.390160] iwlwifi 0000:00:14.3: 0x70E09BCC | FSEQ_ALIVE_TOKEN
[  266.390163] iwlwifi 0000:00:14.3: 0x3B1F3821 | FSEQ_CNVI_ID
[  266.390165] iwlwifi 0000:00:14.3: 0xEB1E049F | FSEQ_CNVR_ID
[  266.390167] iwlwifi 0000:00:14.3: 0x20000302 | CNVI_AUX_MISC_CHIP
[  266.390171] iwlwifi 0000:00:14.3: 0x01300202 | CNVR_AUX_MISC_CHIP
[  266.390176] iwlwifi 0000:00:14.3: 0x0000485B | CNVR_SCU_SD_REGS_SD_REG_DIG_DCDC_VTRIM
[  266.390210] iwlwifi 0000:00:14.3: 0xA5A5A5A2 | CNVR_SCU_SD_REGS_SD_REG_ACTIVE_VDIG_MIRROR
[  266.390270] iwlwifi 0000:00:14.3: SecBoot CPU1 Status: 0x5c63, CPU2 Status: 0x3
[  266.390271] iwlwifi 0000:00:14.3: Failed to start RT ucode: -110
[  266.390273] iwlwifi 0000:00:14.3: Firmware not running - cannot dump error
[  266.402044] iwlwifi 0000:00:14.3: Failed to run INIT ucode: -110

I have a 2 tb drive. I do not need 2tb for popos. I would like to set aside 120gig, and leave the rest of the drive alone.
PopOS does not offer a "use free space" option.
Currently I am doing a manual installation without using the recovery partition. No choice. I have 1 TB of other data on that 2TB SSD.
Please see what you can do to improve the disk space utilisation. A 120gig of free space should be more than adequate for POPOS.

And why is it mounting everyswap file on the system including the one created by your installer.

And by the way, I do like your version of Gnome over other versions around.

Thank you for a great concept and design.

From: https://support.system76.com/articles/roadmap/

Clicking on Pop!_OS Development Approach page leads to the following HTTP 404:

https://support.system76.com/docs/pop-os-development-approach/

Looks like version 4 was released. We should upgrade.

https://jekyllrb.com/docs/upgrading/3-to-4/

Nothing happening here.

"Press the| <Key Icon 1> / <Key Icon 2> | key then type 'disks' to launch the Disks application."

On my Darter Pro the "key icon 2" is not there. Key Icon 1 is right of the fn key. The other icon looks like the Ubuntu logo. I don't see it anywhere.

It may be obvious to people who use Ubuntu all of the time, but this is my first ubuntu laptop so I am just learning. You might either have an alternative way to open the disks app and/or make sure the people understand what model(s) on which the key is found.

https://support.system76.com/articles/install-pop/

The name of this article gives the impression that the tutorial is only valid for System76-manufactured computers.

WIP:
https://github.com/system76/docs/blob/energy-star/_articles/energy-star.md

Could not go to Desktop while anywhere on the screen.
Super + D doesn't work.

The manuals page does not list any instructions, maintenance, or repair information for the Serval WS.

I have this issue where I am reading through a tutorial and then click the System76 logo to go back, and that is not how to do it. Instead, you need to click the little support lifesaver icon. I don't like this. We need a big ole' back button or something like that in order to ensure folks who are reading through our support articles can easily go back to the main support page.

This is especially important if they read through the article and still need to open a support ticket. Let's make it super easy to navigate the support area.

Nothing happening here.

I'd like to put the troubleshooting steps into a matrix/table instead of a paragraph of text, along with 'common cause/try this.'

First, congrats on this awesome feature.
I posted a screenshot earlier of CUDA samples running on my oryxp4 after simply installing PoP clean and following the instructions on this page.
And that was a result of building all of the NVIDIA samples - a great way to assure that everything works.
Most people will not build those samples - they will load their CUDA + Tensorflow of choice using the alternative mechanism and start programming away in Python.
The thing I noticed is that if you try and build the samples in 9.1 or 9.0 instead of 9.2, the make / compile process will barf because most of the demo programs are written in C++, and there was a jump in gcc versions from 9.1 to 9.2 (which makes 9.2 compatible with gcc in PopOS/bionic 18.04). Of course your gcc toolchain can be controlled by the same alternatives mechanism described for CUDA / Tensorflow. If one wanted to be super anal about it, one could offer an option to switch the toolchain as well as the CUDA libs - but the toolchain swap could have side effects on other programs that depend on gcc.
My guess is that the person writing CUDA/Tensorflow in C++ is sophisticated enough to understand that error when they see it, and run the alternatives procedure for gcc separately. And they are comparatively rare among users.
Meanwhile, I am pretty sure the python-based programmer, comprising 98% of use cases, won't know the difference.
So, probably no action required - I think it likely that offering to swap gcc toolchain versions would cause more support incidents than it would prevent.
But I did at least want to point it out in case it ever comes up.

Hi,

I was trying to install Windows 10 on an existing Ubuntu 18.04 that shipped with Oryx Pro.

I follow the below instruction to repair Grub so that Ubuntu can be booted as an option with the newly installed Windows OS.

Git:
docs/_articles/grub.md

https://support.system76.com/articles/grub/

I found out that there is an additional command that needs to be executed to config the grub:

Before:

$...
$ apt install --reinstall grub-efi-amd64 linux-generic linux-headers-generic

After:

$...
$ apt install --reinstall grub-efi-amd64 linux-generic linux-headers-generic
$ update-grub

With that additional updating, GRUB is properly updated, and after a reboot, the system recognizes both OSes.

I have a AORUS GTX 1080 Gaming Box eGPU which attaches via Thunderbolt 3 to my Galago Pro with PopOS 18.04 and NVIDIA drivers.

In order to have the GPU driver available (as in show up with nvidia-smi) I have to plug in the device before boot and pull it out just after the hdd encryption password. Otherwise the laptop will be stuck and the PopOS login screen does not appear.
Once plugged out the login screen loads, I replug the camera, wait until the drivers have loaded (by waiting until the fancy LED lights on the eGPU turn on) and log into PopOS.
Then nvidia-smi works however and I can use the GPU in nvidia-docker successfully.
I am not sure if PopOS actually uses the GPU for rendering software though.

I can plug an external monitor into the graphics card and the monitor recognises being plugged in, but nothing is shown on the monitor. Keeping the monitor directly plugged into the laptop still works though.
Switching between the two is tricky and often requires a hard reset.

How can I
(a) get PopOS to launch without the need to replug the GPU during boot
(b) make sure PopOS uses the GPU for rendering software
(c) be able to plug in the GPU after PopOS is loaded

Thank you

History.
Installed the *.5 Nvidea version and it installed itself as 720x xxxx This was not workable. I could not install software to restore the regular gt218 Geforce 210 U 340.108.
Today I attempted to install the *.6 Nvidia version and this time the monitor came up at 1024 x 1080.
(ratio 4:3)
I am not able to use the POPOS with this resolution.
Fedora, Ubuntu, Suse, Arch all see the monitor as 1920x1080 with ration 16:9
My system ASUS X470-pro-prime, amd2700x, 16gigs ram, nvidea graphics card geforce 210
I do not mind trying to help you out to get the bug fix. Let me know when I have a new iso to try.

Leslie in Montreal

After a firmware update on my Thinkpad PopOS was gone from my boot menu and while I managed to fix it using the instructions at https://support.system76.com/articles/bootloader/, I struggled quite a bit with the first two commands under https://support.system76.com/articles/bootloader/#for-nvme-drives-2 as I had to figure out myself which partitions to use. I think for newbies like me who just follow the instructions without really knowing what to do a clarification on this would be quite helpful. For myself I figured out that the first one had to be the existing PopOS root and the second one the existing PopOS boot partition. I don't really know what this means though so I'll let someone else edit this in the docs.

https://support.system76.com/articles/pop-live-disk/ says

You can run a full version of Pop!_OS from a USB drive (often known as a thumb drive, flash drive, or USB stick) in what’s known as a live environment. Using a live environment (live disk) is useful for:

• Installing Pop!_OS
  [...]

we’ll use Popsicle.

Popsicle is an open source app for Linux that allows you to “burn images to USB drives safely and easily”

Then it continues / runs on ;) with:

, it is installed in Pop!_OS 18.04 and up. If you are running Pop!_OS 17.10 you can install it like so in a terminal:

It's great that Popsicle is installed in Pop!_OS, but if the readers of this page want to install Pop_OS!, these instructions won't help them.

Not sure what the target audience of this page is, but if it's prospective Pop!_OS users, Popsicle installation instructions for them would be useful.

https://support.system76.com/articles/cuda/

In the Not running Pop!_OS? section:

The previous instructions will work with Pop!_OS out of the box but for Ubuntu and other Debian derivatives the following commands will need to be ran first:

"ran" should be "run" in this instance

Overview: The Current State

• The ubuntu cuda 9.1 deb packages appear to be incomplete - missing libraries, headers, tools, and samples. On closer inspection, while some components - most critically, cuda samples - are indeed not present, most of the remaining issues are due to the fact that expected symbolic links are not being created, expected paths and environment variables are not being set. Much of this is attributable to not following nvidia's best-practices for configuration that are intended to allow for side-by-side installs of different CUDA versions. Some inspection of earlier packages (e.g. 8, 7) indicate that they are all made from the same template and subject to the same issues.
• About two weeks after the 18.04 release, NVidia pushed CUDA 9.2.88, which is overall more compatible with 18.04, but only explicitly states support through release 17.10.
• I tested CUDA 9.2, installed from nvidia deb packages, with an ubuntu/ppa-packaged nvidia graphics driver for 396.26 that was temporarily available in a maintainer's ppa. I have now set up a mirror of that installer in a separate ppa, detailed in a section below, so that others can reproduce it.
• The 396.26 install and subsequent CUDA install each required some minor tweaking by hand in synaptic, but once installed had excellent stability and performance. In particular, the extensive "CUDA samples" Makefile, which is a 10-15 minute build on and Intel i9 X-class processor with 128GB memory, ran to completion without error, and the compute-intensive examples performed well.
• It is notable that CUDA-based developers use the installation and make of these sample files as a de facto regression test for correct installation of CUDA and its interaction with other subsystems such as OpenGL.
• Moreover, these sample files are not included in the debian-based CUDA packages. They can be downloaded separately from other CUDA code, installed, and built individually or as a group via make: I downloaded the 9.1 version of CUDA samples sources after installing the 9.1 debian-based packages and found that the install process itself identified many apparently missing libraries and misconfigured paths. Running the makefile for CUDA samples against the debian-based CUDA install identified further misconfiguration, and ultimately crashed before completion.
• Again, this is the de facto regression test used by CUDA devs. It's good to have one.

The Login, Support, and Contact page icons don't show on any pages in docs. They continue to show on the main site.

article has instructions for using lightdm to manage desktop session, but I don't have that installed on 19.10 (looks like I am running gdm3)

I just attempted to install cuda using the instructions on this article: https://github.com/system76/docs/blob/gh-pages/_articles/cuda.md

Cuda was installed successfully but there were some issues:
1: When installed according to these instructions cuda is not found by default within cmake ( I had to set CUDA_TOOLKIT_ROOT_DIR to "/usr/lib/cuda")

2: Cuda executables are not runnable from Everywhere. for example running "nvcc -V" fails but "/usr/lib/cuda/bin/nvcc -V" runs fine

The file docs/_articles/dash-to-dock.md should say something about how it no longer works once you update to to Pop!_OS 20.04. Instead the Gnome Extensions page https://extensions.gnome.or just shows "ERROR" and the Extensions menu for Setup show a slider for Dash to Dock that does not respond when you click on it.

The following line:
https://github.com/system76/docs/blame/master/_articles/cuda.md#L98

does not work for me.
I get the following error:
cat: /usr/lib/cuda/version.txt: No such file or directory

when we have installed, for example, more than 10 window managers installed, the theme can’t list them all - this situation is quite annoying when we enjoy trying different window managers, and forces us to move files around /usr/share/xsessions/

Problem:

After hours of scouring the internet, and losing a few settings by rebuilding my user preferences from the advice under "Move old configuration files out of the way" here, I was able to reclaim my desktop without it constantly logging me out or suspending itself, by returning the universal access cursor size to its default setting. These issues were discovered oddly enough after installing and uninstalling Pithos, the Pandora radio client.

Initially, I thought this was an nvidia driver issue based on internal logs, but after reinstalling and rebooting drivers old and new / big and small, many times, no changes helped, even after following the steps detailed in "Move old configuration files out of the way", which at the very least let me log in for an extended period of time, and demonstrated that the issue was specific to my user profile. Finally, after resetting the cursor size, it stopped crashing.

Ways to replicate:

1. Set Universal Access Cursor Size to largest setting.
2. Install regular pop!_os updates.
3. Install and uninstall debian version of Pithos from pop store.
4. Open the terminal, and spin cursor around the screen rapidly until either the session ends and you are logged out, or the screen freezes and goes black and unresponsive.

This is a problematic issue on several levels, since it not only limits expected use of the OS, but it also causes additional issues for the visually impaired. Please let me know if there is anything else I can help you with. Thank you again for your time, patience, and such an excellent product - even with the bugs!

I checked out Switching to Ubuntu From Apple because I'm curious about how I can manage my photos if I switch. I currently use Apple Photos for this, and it syncs with my iPhone using iCloud.

However, under the heading Working With Images, photo management isn't mentioned. Instead, image manipulation is:

Ubuntu has several powerful image manipulation programs. Gimp, darktable, and Inkscape will do what Adobe products can. Plus, they are free, which is a huge bonus. Adobe products don’t currently run on Ubuntu, but there are plenty of replacement software options.

I understand that this is relevant for advanced users, but it would be interesting for me (as a less advanced user) to get some guidance on how I can replace Apple Photos.

Just to be clear, I don't know if any good options exist. iCloud.com works to some extent, but seems to lack a lot of features.

Copyright still says 2019. The footer is attempting to show the latest year with: {{ site.time | date: '%Y' }} but with no success.

it’s missing other window managers, like AwesomeWM, Fluxbox, WindowMaker, PekWM, JWM, etc. - i use them a lot

Update the screenshots here from 19.04 to 19.10 .

A time ago the version 10.2 of CUDA came out. I was wondering if there is going to be a possibility to update to that version from pop OS.

Best regards

I've been using the tensorflow-cuda-latest for quite some time now but its still in 9.2 which is depreciated for certain frameworks(NVIDIA RAPIDS, which clearly shows that). Will there be an update to 10.0 ? or do I have to do it manually?

Currently instructions mention install chrome-gnome-shell which I've never had to install to get dash to dock working.

Normally I just follow the instructions as guided on the extensions.gnome.org page: Click here to install browser extension

and then do the affirmative on any other popups that appear. May we want to suggest as the way to install GNOME shell integration? Also I believe installing gnome-tweak-tool is not necessary to install dash-to-dock (and there is no further mention of it on this page).

@ahoneybun @kylecorry31 Any thoughts/opinions?

won't work in Firefox since it is missing the DRM plugin. The instructions that they give do not work for me, I've gotten around that by using Chrome. Would it make sense to just add that to the 'chrome-flash' article and rename it to something like 'watching-drm-content'?

I don't think the are complete. In order to link my python 3.6 with the tensorflow-cuda-latest (which, at the time of writing, installs tensorflow 1.11 and cuda-10.0) I had to add the following lines to ~/.bashrc

export PATH=/usr/lib/tensorflow1.11-cuda10.0:/usr/lib/cuda-10.0/bin${PATH:+:${PATH}}
export LD_LIBRARY_PATH=/usr/lib/cuda-10.0/lib64${LD_LIBRARY_PATH:+:${LD_LIBRARY_PATH}}
export PYTHONPATH=/usr/lib/tensorflow1.11-cuda10.0/lib/python3.6:$PYTHONPATH

See Restore page.

Hi,
The download link for Nvidia return this error:

This XML file does not appear to have any style information associated with it. The document tree is shown below. 

<Error>
<Code>NoSuchKey</Code>
<BucketName>pop-iso</BucketName>
<RequestId>tx0000000000000e6f5c129-005ebea442-443fd0-sfo2a</RequestId>
<HostId>443fd0-sfo2a-sfo</HostId>
</Error>

Currently when reading a support doc, there is no button to press to get back to the main support page. We should include that on each page somewhere in the blue header (or if there is a better place I will rely on @cassidyjames design kung-fu to decide on that).

I think the Ambiance theme in GNOME just looks bad and takes away the Desktops beauty. This is more of a personal take on it then a issue but there is not really another way to bring it up. I would happily make the change myself and a PR but since you have all the screenshots on System76 hardware I don't want to remove that professional look.

My Darter Pro came from the factory with full-disk-encryption, and the docs for dual-booting Windows don't work. Specifically, I cannot shrink the partition.

https://support.system76.com/articles/dual-booting/

Can someone check to make sure you can still install Steam without issue from the normal repo? I believe I had to enable to multiverse last time (for 32-bit support).

Hello,
I try to install but boot stuck on this

Listening on load/save rf kill switch status /dev/rfkill watch

https://ibb.co/hyPjLgy

I have Ubuntu on my PC, he loading w/o any issues

While this is possible, the CPU does not support it so it seems unlikely. Does this system mainboard have a PCIe switch that aggregates the gen3 lanes from the cpu, or is that detail in the doc a typo?

the 'repairing package manager' link in the doc at
https://github.com/system76/docs/blob/master/_articles/upgrade-pop.md
to
https://support.system76.com/articles/package-manager/
results in this -

Hmm…
This article doesn’t exist here. Try the Support page.

This documentation seems to be not-hosted on Github, the reason I am posting is it does not disclose all logins and registrations being tracked by Google. In fact it is quite hard to contact system76 at all without registering and disclosing your usage to Google, which you don't disclose to your users. Writing directly to email addresses just bounces so it's unnecessarily difficult too.

https://system76.com/privacy

The Repair the bootloader page begins with the following paragraph.

GRUB is the bootloader. It takes care of getting the operating system started up. It is also responsible for allowing the user to select between multiple operating systems at boot. Sometimes, GRUB can break, and it may not let you boot into your computer to fix the problem.

On Pop!_OS machines, GRUB is not the default boot-loader (in most cases). This paragraph/document potentially should be reorganized to mention the bootloader that comes on our current machines first (systemd-boot). It potentially should also be done for the fact that it is linked to from Pop!_OS related documentation (https://pop.system76.com/docs/pop-live-disk/, https://support.system76.com/articles/pop-live-disk/)

Thoughts @ahoneybun ?