#asahi-dev · 2025-12-27 21:49
jannauexpose the full 1500 (or 1000) nits range as backlight values. Values over the max SDR brightness are clipped to the current max if only SDR content is shown#asahi-dev · 2025-12-27 21:46
oliverNot sure I understand. Did you mean: clip the brightness in hdr mode when no hdr content is visible? #asahi-dev · 2025-12-27 21:46
jannauyou seem to be right that mutter doesn't modulate the backlight for HDR: https://blog.sebastianwick.net/posts/gnome-49-backlight-changes/#asahi-dev · 2025-12-27 21:41
jannaualso not quite sure how to integrate dcp's brightness control. Most practical solution might be to extend the range and clip the brightness if no HDR content is shown#asahi-dev · 2025-12-27 21:39
oliverI think mutter is currently ignoring that. But sure what kwin does with that. It looks like the whole space on Linux isn't quite there yet #asahi-dev · 2025-12-27 21:33
jannauDisplayID standard is available free of charge after registration from https://vesa.org/vesa-standards/#asahi-dev · 2025-12-27 21:33
jannauoliver: DisplayID Display Parameters Data block contains "Native Maximum Luminance (Full Coverage)", "Native Maximum Luminance (10% Rectangular Coverage)" and "Native Minimum Luminance" fields. So that would be solved by providing an EDID as well#asahi-dev · 2025-12-27 21:29
sven being ambushed happened quite a few times today to me as well :D#asahi-dev · 2025-12-27 21:27
ellyqwsleep well, sven:D#asahi-dev · 2025-12-27 21:27
ellyqwsorry, i was supposed to come around but i've been ambushed by multiple people#asahi-dev · 2025-12-27 21:23
fl0_idsven oh wow sure. get some sleep :)#asahi-dev · 2025-12-27 21:23
svenyeah, same, just went back to the hotel since I’ve been up since 5am#asahi-dev · 2025-12-27 21:21
fl0_idsven I tried booting kernel and device tree - fails at cpufreq / core initialization I think. Next thing could be to comment out some cores like other testing trees did. but I'll leave that for tomorrow, I need some sleep soon#asahi-dev · 2025-12-27 20:48
jannauhopefully the new drm/kms backlight api arrives soon: https://zamundaaa.github.io/wayland/2025/07/22/display-next-hackfest-2025.html#new-backlight-api#asahi-dev · 2025-12-27 20:47
jannauI suppose that would be normally signalled through edid#asahi-dev · 2025-12-27 20:46
jannauoliver: my understanding for backlight control is that kwin still wants to control the backlight for HDR. For that to work you have to set maxBrightnessOverride, maxAverageBrightnessOverride and minBrightnessOverride via kscreen-doctor: https://paste.debian.net/hidden/99798351#asahi-dev · 2025-12-27 20:25
oliverThanks#asahi-dev · 2025-12-27 20:22
jannauI haven't seen anything (but I haven't searched extensively) and I suspect they just hard-code their screen properties directly without EDID so building it from know properties and guesses is the only solution. I have a photo meter so we can at least to some degreee test how good our guesses are#asahi-dev · 2025-12-27 20:16
oliverSounds reasonable. Do you know if I get it from macos or do I need to build one up from the ground? #asahi-dev · 2025-12-27 20:13
jannauoliver: plan for internal display edid is/was to let m1n1 it embed in the device tree (or provide it as firmware)#asahi-dev · 2025-12-27 19:34
oliver* and totally fine with kde too#asahi-dev · 2025-12-27 19:17
oliverI put something out today. It works pretty fine with vanilla gnome 49 and okayish with kde for now. would be nice if someone could try this out. I have the feeling, that the whole hdr space on linux is still not really there yet... but at least it works: https://github.com/AsahiLinux/linux/pull/431#asahi-dev · 2025-12-27 17:54
svenintegralpilot: do you use gcc or clang to build m1n1? maybe you're running into a variant of https://github.com/AsahiLinux/m1n1/issues/463 w.r.t your corruption#asahi-dev · 2025-12-27 17:32
ellyqwack, will do after we get some cash from ATM for drinks on venue and bring some spezi (i crave spezi)#asahi-dev · 2025-12-27 17:25
sveni'm not nearly anxious enough to start working on them :D#asahi-dev · 2025-12-27 17:25
svenfeel free to head over#asahi-dev · 2025-12-27 17:25
sveneh... realistically ill do the slides on day 3#asahi-dev · 2025-12-27 17:24
ellyqwsven: should i head to your assembbly later then, or do you wanna focus on the talk? :D#asahi-dev · 2025-12-27 16:34
jannauI haven't tested extensively but I've encoutered the broken raw custom boot objects only on m4*. I don't remember if I tested m3#asahi-dev · 2025-12-27 16:30
chaos_princessdon't remember post-m2 all that well :P#asahi-dev · 2025-12-27 16:30
chaos_princessor, actually nvm, i confused the soc identifiers#asahi-dev · 2025-12-27 16:29
chaos_princessare you sure? i checked the ipsw and the sptm img4 was there#asahi-dev · 2025-12-27 16:29
nickchanlike raw custom boot object broken on 15.2+...#asahi-dev · 2025-12-27 16:28
nickchan15.x doesn't have sptm for m3 but still don't use it for other reasons#asahi-dev · 2025-12-27 12:41
chaos_princessintegralpilot: do not go for 15+ on m3, as it has sptm#asahi-dev · 2025-12-27 12:36
chadmedi dont want to give up on zero copy scanout from the gpu but i cannot figure out where the heck we're supposed to tell iomfb that the format is tiled/twiddled even after spending all evening in ghidra :/#asahi-dev · 2025-12-27 12:35
chadmedbut once it looks more like a real kms atomic driver we should port it #asahi-dev · 2025-12-27 12:35
chadmedim happy to keep hacking on the C version because my kernel rust is even worse than my kernel C #asahi-dev · 2025-12-27 12:35
chadmedyeah#asahi-dev · 2025-12-27 12:34
jannauI kinda hope that we don't add another FW version to the C dcp driver#asahi-dev · 2025-12-27 12:33
jannauI think it's useful for bringup to use major macOS versions as stages as long as nobody sinks time into analysing complex FW interfaces like dcp or agx#asahi-dev · 2025-12-27 12:29
oliverthank you!#asahi-dev · 2025-12-27 12:28
jannauoliver: easiest option to spot this would be experiments/dcp.py I don't think we explictily documented this outside of code. consider m1n1's python implementations and experiments as documentation#asahi-dev · 2025-12-27 12:26
chadmedsimilarly it would be good to drop 12.3 out of the dcp driver since we already know iomfb changed again #asahi-dev · 2025-12-27 12:25
chadmedobv doesnt matter if its just for a bit of fun at 39c3 but for prod i would want to bring everything up to a new lowest common denominator (drm uapi stability guarantees notwithstanding) #asahi-dev · 2025-12-27 12:24
chadmedyeah but do we want to support 14.x and then yet another for m4#asahi-dev · 2025-12-27 12:22
jannaulatest 14.x would be the easiest option for M3 devices. nvme there might jsut work#asahi-dev · 2025-12-27 12:15
chadmednot sure if 26.x firmware expects SPTM though#asahi-dev · 2025-12-27 12:15
chadmedintegralpilot: 26.x is going to be your best bet as there were some nasty bugs in 15.x that _should_ all be fixed now#asahi-dev · 2025-12-27 12:14
chadmedoliver: the devicetrees have them, but i think for the dcp/dispX thing its not obvious and probably should be documented somewhere