00:10 <enyalios> qcvss8280.mbn goes into <firmware_dir>/qcom/sc8280xp/LENOVO/21BX/ , correct?

01:04 <enyalios> looks like it `/usr/src/linux-6.5.3+/arch/arm64/boot/dts/qcom/sc8280xp-lenovo-thinkpad-x13s.dts: firmware-name = "qcom/sc8280xp/LENOVO/21BX/qcvss8280.mbn";`

01:20 <xnox> ubuntu isos were pulled, but will be back again once we have fixed translations strings

05:04 <cenunix[m]> Sorry if this is a bit unrelated but I’m trying to get Nixos to boot into an installer iso using the mainline kernel

05:05 <cenunix[m]> Am wondering if I’m missing some kernel modules or something? I’m using the kernel params that jhovold listed on the x13s wiki page.

05:07 <cenunix[m]> An error occurred in stage 1 of the boot process, which must mount the root filesystem on mnt-root' and then start stage 2.

05:25 <Guest2683> cenunix[m]: not able to help, but you have a WIP branch where you make this installer? I have tried the same, but gave up after a few weeks of not understanding why the iso wasn't mounted, so stage 1 wasn't even started. So interested in seeing if my setup can get further and I might be able to help out (or if I'll just get stuck at the same place)

05:28 <cenunix[m]> For this I’m using Nixos-generators, will link the config I use in a sec

07:18 <Jasper[m]> <cenunix[m]> "An error occurred in stage 1..." <- There are some issues currently with getting usb to work on boot

07:19 <Jasper[m]> Dit you try replugging the usb stick while it's searching for the rootfs?

07:22 <Jasper[m]> <xnox> "ubuntu isos were pulled, but..." <- https://fxtwitter.com/Tick_Hack_bot/status/1712544420644663475 oh uhh, you weren't kidding

09:33 <cenunix[m]> <Guest2683> "cenunix: not able to help, but..." <- https://gist.github.com/cenunix/b332ecb4026ccf3dda62bf564d86b529

09:33 <cenunix[m]> command i use to build the iso: nix run github:nix-community/nixos-generators -- -f install-iso --system aarch64-linux -c x13s.nix

09:35 <cenunix[m]> <Jasper[m]> "Dit you try replugging the usb..." <- yeah just tried that with no luck, i'll try it a couple more times though

09:36 <cenunix[m]> im gonna try the old flip the usb upside down in a sec too

09:36 <Jasper[m]> cenunix[m]: If it doesn't you should try following the instructions on Fedora's wiki about the X13s https://fedoraproject.org/wiki/Thinkpad_X13s#Rawhide_hacking

09:36 <Jasper[m]> You just need to do those initfs changes for the boot iso

09:39 <cenunix[m]> that should be doable thanks so much for the link

09:40 <cenunix[m]> just for clarification i assume you mean this part especially :... (full message at <https://matrix.org/_matrix/media/v3/download/matrix.org/CaIxoGjElgHwoiFmQdIAeFFt>)

09:40 <Jasper[m]> No problem! I haven't tested this personally and @clover hasn't yet either for archiso. But I'm assuming Fedora know what they're doing

09:40 <Jasper[m]> cenunix[m]: > <@selmer443:matrix.org> just for clarification i assume you mean this part especially :... (full message at <https://matrix.org/_matrix/media/v3/download/matrix.org/dGcQlOmkZRpPOcbpsuWwIwYQ>)

09:41 <cenunix[m]> sweet ill give it a try, hopefully I can get a reasonably stable iso built so it can be added to community support ARM devices on the NixOS wiki

09:41 <Jasper[m]> By the way, audio is almost perfect for me on this thing. Holy crap

09:42 <Jasper[m]> Headphones are well amplified and the noise is solved as said in the talk

09:43 <Jasper[m]> Speakers are loud enough (not entirely on Windows stock level, but there could be some software magic involved there)

09:43 <Jasper[m]> Only thing I've noticed with the headphones is that there's some noise/distortion on lower volume parts of songs

09:45 <Jasper[m]> A good example is the intro for J Dilla, Guilty Simpson and Madlib - Baby (proceed with caution if you don't like hiphop haha). Some of the noise pulls through the entirety of the song. Not sure if that's because of some frequency range being present or not.

09:47 <Jasper[m]> Okay nvm, it's mostly the louder music where it gets overblown hahaha, music that is mastered less loudly tend to still have the noise and distortion (Yussef Dayes - Love is the message Live at abby road)

10:26 <cenunix[m]> good to hear sound is working so well :)

10:26 <cenunix[m]> we've come so far from the early days haha

11:34 <Bioxvirizm-x13s[m]> <Jasper[m]> "https://fxtwitter.com/Tick_Hack_..."; <- lol

11:52 <qzed> lun: Error 22 is invalid argument... I'll try to see if I can make some sense of it this evening

15:08 <konradybcio> https://www.phoronix.com/news/Blender-Vulkan-Experimental

15:08 <konradybcio> blender on x13s hmm

15:32 <HdkR> Is it just using Vulkan Compute?

15:33 <HdkR> Oh, for the renderer?

15:33 <HdkR> That's not quite exciting as I thought. X13s already supports GL 4.6 so it should just work?

17:39 <steev> hm

17:39 <steev> there's definitely something up with usb on 6.6

17:39 <steev> Bus 004 Device 002: ID 0bda:9210 Realtek Semiconductor Corp. RTL9210 M.2 NVME Adapter

17:42 <steev> on 6.5 i can do something like grep through all 655 git repo checkouts of kali packages, but with 6.6, doing so causes it to 2023-10-11T21:00:01.366768-05:00 wintermute kernel: [ 2315.389963] sd 0:0:0:0: [sda] tag#23 uas_eh_abort_handler 0 uas-tag 5 inflight: CMD

17:42 <steev> 2023-10-11T21:00:01.510596-05:00 wintermute kernel: [ 2315.534723] usb 6-1: reset SuperSpeed Plus Gen 2x1 USB device number 2 using xhci-hcd

19:26 <DanaG> The one ARM desktop-ish thing I have is a Honeycomb LX2160A, but man, getting an ideal GPU to work with that thing has been a struggle. Old AMD can't do 4k 120Hz (even if I can't game on it, 120hz is great for just plain desktop use). New AMD can't load the firmware. New NVIDIA falls off the bus, and doesn't have the binary drivers for 32-bit stuff.

19:26 <DanaG> Can any of the Qualcomm laptops do 4k 120hz over DisplayPort?

19:40 <Jasper[m]> <DanaG> "The one ARM desktop-ish thing..." <- Didn't someone make a x86 emulator EFI exactly for that firmware issue you mentioned?

19:41 <Jasper[m]> https://github.com/ardbiesheuvel/X86EmulatorPkg

19:47 <DanaG> I don't mean the UEFI GOP driver, I mean the driver-load-time PSP (and such) firmware.

19:48 <DanaG> Oh man, imagine if we could get an Exynos laptop with the AMD-based GPU, that could be cool. Though, how good even is Samsung at upstreaming their stuff?

19:49 <DanaG> Speaking of upstreaming, some of the issues with LX2160A have fixes that the upstream has rejected or roadblocked for ages. That's why I'm considering getting something else for desktop use.

19:52 <steev> they've definitely gotten better about upstreaming post... what were those things... s5p?

19:54 <Jasper[m]> <DanaG> "I don't mean the UEFI GOP driver..." <- Ahh yeah, didn't know that was architecture dependent.

19:55 <Jasper[m]> steev: The early iphone socs?

19:55 <steev> was the early iphone a samsung s5p? i was more thinking the nanopi devices using the s5p6818 or some such, which never made it to mainline

20:01 <Jasper[m]> <steev> "was the early iphone a samsung s..." <- They didn't have the typical Apple Asomething nomenclature yet so they let the fab name them

20:02 <Jasper[m]> S5L8900

20:24 <steev> ah, i was thinking the older nanopi boards, but then i remembered that i got rid of most of them

22:06 <DanaG> And actually, the reason the GOP isn't a problem is because AMD released aarch64 GOP .efi drivers.