In my case, it’s the LCD model, too (and X1 Plus 10-core, because no Elite with 5G). Maybe I’ll try your ArchLinux image and just not outright install it, to see if the LCD works.
I was actually primarily thinking: I hope Canonical staff see your work and talk to you about it!
I would want to try Arch from your repo, but I am afraid to do something wrong. If you are willing to explain a little further, I would genuinely try it. I don’t feel a noob, but neither an expert.
Honestly, I wouldn’t recommend my repo if you aren’t prepared to deal with a bricked machine. Arch Linux requires a reasonable amount of understanding of how Linux works at the system level; nothing is automated for you, there’s no hand-holding. Installation is manual, one command at a time.
If you want to learn Arch Linux, maybe try in a virtual machine and become familiar with it before messing with your expensive Surface. That said, the USB image should be safe to try - it’ll get you to a login prompt and you can connect to the Internet and install a desktop if you want.
The goal of the repo is to reduce friction for developers who could help get the rest of the machine working. I wouldn’t recommend it for end-users.
Happy to add your patches to our kernel build! I have been a bit distracted by release preparations and launchpad issues this week but I really appreciate your work and I can see that a lot of people seem to be interested in getting the Surface 11 up and running.
I gave up trying to find any information on how to build the ARM Ubuntu ISO, so I’ll just focus on distros I’m more familiar with until the kernel is working well enough.
I agree that there is a unfortunately a lack of documentation and good tooling in that area. We have built this repo to make it a little easier to locally reproduce image builds. I was planning to polish it a bit and remove some more weirdness but that turned out to be harder than expected. Maybe it is time to give it another try.
That’s great to hear, thank you! I appreciate you have a lot of X1 machine variants to take care of and your work is definitely cut out for you!
I think the most important parts are:
Please note that most of these patches are WIP or just complete hacks (e.g. Wi-Fi rfkill) until better solutions can be identified.
I’ll have another look at the ubuntu-iso-build repo, thank you! If you have some spare time to write me a TL;DR of how to use it I’d love to try and get things building locally so it can be tested.
Cheers 
Bluetooth is working on my t14 with 6.14.0-13-qcom-x1e, thanks!!
I’ve been building my own DTB (this patch) to get the display backlight control working on my t14s, but since I enabled full disk encryption, I have to type the unlock password into a dark screen.
Is it a matter of adding some module to initramfs to have backlight working? If yes, which one? Thanks
try bwm_bl, this should load it. in /etc/initramfs-tools/modules
Thanks! It was pwm_bl, but that worked like a charm.
Thanks for figuring it out, I’ll add it to the meta package to have it included by default!
Can you push your dts somewhere? Working on resolving my brightness issue on the ThinkBook 16
it was literally just this:
git show dce657f91da932d9b2f4069b1c866c80cbd7503b -- arch/arm64/boot/dts/qcom/x1e78100-lenovo-thinkpad-t14s.dts b/arch/arm64/boot/dts/qcom/x1e78100-lenovo-thinkpad-t14s.dts | git apply -R --reject
I know that sound is not working on Yoga 7 slim
[ 39.039601] snd-x1e80100 sound: probe with driver snd-x1e80100 failed with error -22
But is this error solvable somehow ?
[ 84.309120] qcom-iris aa00000.video-codec: core init failed
[ 1479.508442] qcom-iris aa00000.video-codec: error -22 initializing firmware qcom/vpu/vpu30_p4.mbn
[ 1479.517547] qcom-iris aa00000.video-codec: firmware download failed
no 
The reason it started showing the screen during FDE prompt is that update_initramfs reverted my custom DTB. Not because of pwm_bl, which seems to not do anything. @glathe
So the backlight won’t adjust now. Sorry for the false excitement, I only noticed it now.
But is this error solvable somehow ?
This shouldn’t really cause any problems. This is the experimentally enabled iris video acceleration driver, we don’t ship or extract any firmware for it yet and if we would we’d probably have to adjust the path.
You can either ignore it or block the qcom-iris module from loading if it bothers you.
Hello, did anybody manage to get this running on an X1P42100? I’ve made a bug report, but there doesn’t seem to be any active development as of right now. Pretty interesting considering how many of these CPUs are being sold as we speak.
Hi, well this is an asymmetric issue. Enabling or bringing up new devices or variants is mainly a voluntary effort. We have Purwa (X1P42100, X126100) support now thanks to QCom/Linaro, although no GPU acceleration yet. Someone needs to do the dtbs for them - even if the base device is already supported. This is harder if you don’t have the device and / or documentation (and time) to do it. I’ve done this for ThinkBook 16, and VivoBook S15. I’m somewhat stuck now with 6.14 - many changes and improvements that show errors in previously working combinations. This is normal, I guess. But it takes a lot of time, too, making sure that the things already published work.
Back to the topic - there aren’t enough developers to do the stuff the vendors don’t. This will only change over time. The Acer Swift Go is supported now, shouldn’t be too hard to do the refactoring for x1p42100 and x1e80100. I’m not sure if the dtbloader can distinguish them, though, since they need separate dtbs. As a template on how to do this, I have this commit on what needs to be done.
Hi, I really hope I didn’t come across as rude there; I appreciate all the work you guys have done to no extent! I just found it curious is all.
I’ll be glad to help you however I can if you (or anybody else) gets around to making the relevant .dtb. I tried taking a shot at it myself but got lost in seconds, I am amazed by how you guys can write this stuff.
You said it’s supported, where? Should I just redownload the latest image? Even without gpu acceleration, having ubuntu installed would be massive!
What do you recommend to all the hardware owners out there to throw the Vendors into their face that accelerate things? Is it lack of documentation, lack of ready-to-use drivers or simple license issues with the existing windows drivers? I’m on the Lenovo (Yoga Slim 7X) front and were somewhat unamused that lenovo doesn’t provide native linux support since Day 0. I guess, a somewhat orchestrated support request offensive won’t result in instant happiness on any side, but may nudge them into a direction we want to have?!!
What else can be done if one is uncapable of developing kernel drivers?
Again, your work is very appreciated and valued in a personal and professional manner.
The Acer Swift 14 AI should be in the newer ISO’s I guess, the dtb and efivars enabling entry is in my repo for sure. @bilbao12 Is using it AFAIK. But that’s the x1e variant. It is the basis for commonizing the dtb to support both SoCs. Hamoa (12 core) and Purwa (8 core) are pin compatible and can be used on the same board.
For Purwa to boot, you also need patches in the kernel to support some newer PCIe configurations to bring up the SoC. Otherwise it will just reset. From what I’ve read, they seem to be in the newer ISOs from Canonical, so that should be sorted. Booting with the ISO is still a pain, since you need the deviating dtb. I have no good solution for this problem, my workaround is using a raw image that can boot and bootstrap to the preinstalled desktop. Since it’s one image for every dedicated dtb, its sorta unworkable at scale. But for special cases its the only way I know to bring it up.
As I said, I’m currently wrestling with 6.14-rc4 to work with my HP Omnibook X14, there are changes in there for the USB4 / DP retimers which kill the display for me. Looks like my guessing of i2c buses was wrong, and I have to re-research until they work as expected.