#asahi · 2026-01-03 17:48
Glanzmannsven / jannau: The kernel oops the other day was caused by an outdated m1n1. As a result the calibration values where not loaded correctly and that caused the 0 pointer derefence.#asahi · 2026-01-03 00:31
lukaskelThanks a lot for your work of looking into it. :) Most people might be quiet about this stuff, but for everyone using linux in production this will be huge and contribute tons.#asahi · 2026-01-03 00:29
lukaskelFingers crossed. Wish I could help but still to early in my coding life to contribute aside testing and reporting potential bugs. If any tests are ever needed for color ranges and hdr, I‘ll be there.#asahi · 2026-01-03 00:26
chadmedwe also need to be able to force a modeset to the higher colour modes for hdr and i havent worked out how to do that reliably yet#asahi · 2026-01-03 00:25
chadmedi had a patch to work around this and it worked, but it was useless while we cant set up hdr properly in the first place#asahi · 2026-01-03 00:25
chadmedone other but possible to work around limitation is that the diplay controller heuristic for "best" colour mode to put a display into prioritises bit depth over everything else, so if one exists it will pick an ugyl 12-bit limited range YCbCr mode over a 10-bit RGB mode#asahi · 2026-01-03 00:24
chadmedthe display controller expects each plane to accurately tell it its colourspace, transfer function and framebuffer format. we cannot do the former two for rgb planes until the colour pipeline patches are present, and even then the colourspace property is on the "to do" list#asahi · 2026-01-03 00:23
chadmedhdr probably wont be _reliably_ possible until at least 6.19, perhaps later#asahi · 2026-01-03 00:22
lukaskelAhh thats amazing. Already super happy that DP is now working. Will experiment more and report any bugs, but first impression is super smooth. Hats off to everyone#asahi · 2026-01-03 00:21
jannauit's not supported right now but James and Oliver are working on it#asahi · 2026-01-03 00:20
jannauno HDR is the display driver#asahi · 2026-01-03 00:19
lukaskelThanks everyone!#asahi · 2026-01-03 00:19
lukaskelMy external screen is now working, which is amazing! 4k, 120hz. Super smooth and hot plugging seems smooth as well. Sadly no HDR showing, but that could also be KDE or my cable. Or in general linux being semi hdr friendly still#asahi · 2026-01-03 00:18
lukaskelNvm, I got it to work with setting new default with grubby command. :)#asahi · 2026-01-02 23:51
lukaskelThe kernel now compiled without issues and but after make modules_install and make install, as well as grub2-mkconfig it still boots the old kernel which is now with the new dtb laggy as hell.#asahi · 2026-01-02 21:18
lukaskelThank you so much so far!#asahi · 2026-01-02 21:17
lukaskelGotcha. Will double check, was sure I checked in there but maybe I just searched for secure boot#asahi · 2026-01-02 21:16
chaos_princess`make menuconfig`, search for "sbat", empty that field#asahi · 2026-01-02 21:16
lukaskelconfig from existing install always throws me an error during compile that "no rule to make target 'kernel.sbat' needed by 'arch/arm64/boot/zboot-header.o', so I went from defconfig#asahi · 2026-01-02 21:12
chaos_princesscopy config from existing install and use that#asahi · 2026-01-02 21:12
chaos_princesswait, do not use defconfig#asahi · 2026-01-02 21:11
lukaskeldoes "make defconf" and afterwards "make asahi.config" need any additional configuration if that module ends up in my kernel? Was also suprised to see it connected to mostly broadcom cards from my lil research and no apple information at all#asahi · 2026-01-02 21:10
jannauor to be more precise. p2p wlan was always broken. I think there is a quirk to disable it#asahi · 2026-01-02 21:09
jannauthere is no p2p wlan with the apple firmware#asahi · 2026-01-02 21:09
chaos_princessidk on that one#asahi · 2026-01-02 21:09
lukaskelAny idea where the p2p-dev error comes from when booting? Or is that also a cause of the missing rustavailable part potentially? Sometimes needing more than 10 reboots to get past that error again and into fedora#asahi · 2026-01-02 21:07
chaos_princessrest sounds alright#asahi · 2026-01-02 21:07
lukaskelGoing crazy if that was my only issue now, lol#asahi · 2026-01-02 21:07
lukaskelThanks. Aside of that my whole pipeline seems correct though, right? Stage 2 boot.bin with the right .dtb is mandatory at that point, right?#asahi · 2026-01-02 21:03
lukaskelEnding up with still laggy display performance, no usb c dp compability and very often boot errors that say tat p2p-dev-wlp1s0f type 10 failed#asahi · 2026-01-02 21:03
chaos_princessrun `make rustavailable` inside the kernel source directory#asahi · 2026-01-02 21:02
lukaskelCloning fairydust branch and setting it up with its own defconf and adding asahi.config. Compiling is with usual make command. Afterwards I update grub with mkconfig and end up the new kernel. Weirdly the whole macbook (m1 pro) alr feels laggy at that point. Someone on reddit pointed out that the dtb is not correct, so I follow the asahi documentation and build stage 2 boot.bin with newly compiled m1n1, u-boot and the correct .dtb for my macboo#asahi · 2026-01-02 20:59
lukaskelCurrently experimenting with the fairydust kernel to see if usb c -> hdmi adapters for 4k 120hz displays like the LG C2 work. Sadly I keep stumbling upon issues. Maybe someone sees a major issue in my logic of fairydust branch setup. Would appreciate any help! Will just give a rough outline, so if anyone might have a clue, that'd be great! #asahi · 2026-01-02 20:57
lukaskelHey everone :)#asahi · 2026-01-02 18:46
Moderation[m][@_oftc_hdbngr:matrix.org](https://matrix.to/#/@_oftc_hdbngr:matrix.org) Hi, you are changing your room membership too frequently, and may be temporarily banned as an automated precaution if you continue.#asahi · 2026-01-02 11:24
monthebiffevening, did i see some news stuff about the m3 being discontinued by apple?#asahi · 2026-01-01 21:38
oddlinehttps://media.ccc.de/v/39c3-asahi-linux-porting-linux-to-apple-silicon if anyone else is interested :)#asahi · 2026-01-01 21:32
cxoThanks#asahi · 2026-01-01 21:29
fl0_idcxo should be on media.ccc.de#asahi · 2026-01-01 21:24
cxoAnyone have a link to a recording of this session? https://fahrplan.events.ccc.de/congress/2025/fahrplan/event/asahi-linux-porting-linux-to-apple-silicon#asahi · 2026-01-01 16:24
chaos_princessbut the further you go on the vanilla/pro/max axis, the worse it is#asahi · 2026-01-01 16:23
chaos_princessthe most power hungry part of your machine is the display backlight, which means that the chip being less power efficient under linux does not matter that much#asahi · 2026-01-01 16:23
StrataByteThat's good to know then. Do other devices struggle? I'm assuming the newer M4's are still being worked on and whatnot?#asahi · 2026-01-01 16:22
chaos_princessaside from sleep, which is way worse than macos#asahi · 2026-01-01 16:22
chaos_princessthe power use is fine there#asahi · 2026-01-01 16:22
StrataByteMacBook Air A2941, M2 chip#asahi · 2026-01-01 16:21
chaos_princesswhat machine is it#asahi · 2026-01-01 16:20
StrataBytefl0_id luckily it doesn't leave the house (too expensive to risk a drop) so it'll always be close to a charger 🤣 as long as it works, it's stable, and she can keep doing her work, art, and 3D modelling, I don't think she will mind have to grab the cable every now and again 🤞🏻#asahi · 2026-01-01 16:17
fl0_idStrataByte battery life could be an issue#asahi · 2026-01-01 15:56
StrataByteAgreed, though it's minor at most. The majority of devices I use or manage are business machines anyways, no fancy fingerprint sensors in sight ;D