#
#asahi-dev · 2026-01-30 01:14
jskIs there anything that needs work on t8122 (except maybe the whole gpu stuff haha) that one could help with, without lots of re experience?
#asahi-dev · 2026-01-29 15:02
noopwafelwould be great, no rush; just right now the pcie stuff is very 'works on t8122' which is no basis for stability
#asahi-dev · 2026-01-29 14:56
fl0_idnoopwafel would be happy to provide that. but probably can only do this on the weekend, as I'm traveling and teaching all day tomorrow
#asahi-dev · 2026-01-29 14:18
noopwafelactually I would really like a pcie trace from t6031 :)
#asahi-dev · 2026-01-29 14:17
noopwafel(we can also provide traces if needed btw)
#asahi-dev · 2026-01-29 14:16
Shizlet's clean it up in a bit :p
#asahi-dev · 2026-01-29 14:13
noopwafelmaybe we can extract some pieces at least..
#asahi-dev · 2026-01-29 14:10
noopwafeljust at least it is a way to work on traces.
#asahi-dev · 2026-01-29 14:10
noopwafelmy nonsense to a branch but it should indeed not be inspiration for anything
#asahi-dev · 2026-01-29 14:09
noopwafelI am pushing all
#asahi-dev · 2026-01-29 14:06
noopwafelunfortunately they are all hacky/bad
#asahi-dev · 2026-01-29 13:44
fl0_idShiz might try them on t6031 too if possible.. can you dm me as well?
#asahi-dev · 2026-01-29 13:35
ShizArtingl:
#asahi-dev · 2026-01-29 13:35
Shizwill send you a pm because it's better if random people not get any ideas from these patches :p
#asahi-dev · 2026-01-29 13:34
Shizok yeah, 14.8.3 should be fine
#asahi-dev · 2026-01-29 13:34
ArtinglShiz: yeah i know they are kernel specific. as far as i understand, they just patch out some msr operations. i tried to do it manually, couldn't get it. i am using kernel 14.8.3. is there a way to access these dirty patched? would like to look into them :)
#asahi-dev · 2026-01-29 13:27
Shizalso the patches are very kernel-specific :p so what macOS version are you trying to boot, Artingl?
#asahi-dev · 2026-01-29 13:26
Shiz(afaik t6030 has the same issues, so it's not *that* specific)
#asahi-dev · 2026-01-29 13:25
Shizwe have an extremely quick-n-dirty hacky branch of m1n1 that makes it boot, not sure if noopwafel fully cleaned itu p
#asahi-dev · 2026-01-29 13:25
Shizthere's a number of issues yes
#asahi-dev · 2026-01-29 13:12
fl0_idArtingl look at this for example https://github.com/AsahiLinux/m1n1/pull/518
#asahi-dev · 2026-01-29 13:11
fl0_idArtingl they are different, afaik all the chickens are for sure not the same. You can as integralpilot when they come on, or shiz / noopwaffel maybe. Not sure if there work is online yet. like t6030 chickens also do not work on t6031
#asahi-dev · 2026-01-29 13:01
Artingli was trying to run hypervisor on t8122, and had some issues. it always hangs on reading/writing to msr s3_0_c15_c2_3 and s3_0_c15_c2_4 for some reason. i've seen that it is possible to run hv on t6030 with some patches. i tried to apply them on my system and it didnt help, crashes later during process on some ldr instruction. is t8122 that different from it? or maybe i should use different kernel version?
#asahi-dev · 2026-01-29 00:38
fl0_id___prob should start fresh tomorrow
#asahi-dev · 2026-01-29 00:35
fl0_id___pursuant to git I shouldn't have any other changes, but sth must be wrong...
#asahi-dev · 2026-01-29 00:34
fl0_id___argh. I thought I had only committed my changes, but now nothing works atm, it doesn't even find the aic
#asahi-dev · 2026-01-28 22:01
jannaulook like we can avoid dmi if we set the long_name from the driver
#asahi-dev · 2026-01-28 21:54
jannauit's the kernel and not alsa userspace. The second line in /proc/cards has this new format
#asahi-dev · 2026-01-28 21:20
jannauI'd say it would make sense to at least fall back to that
#asahi-dev · 2026-01-28 21:19
jannauI suppose the card names are not vendor specific on PCs and so it won't be possible to add device specific use cases/configs
#asahi-dev · 2026-01-28 21:18
JamesCalligeros[m]Because it seems really really silly
#asahi-dev · 2026-01-28 21:18
JamesCalligeros[m]I think this is worth asking the alsa folks about
#asahi-dev · 2026-01-28 21:17
jannau"fixed" by symlinking the correct config
#asahi-dev · 2026-01-28 21:15
JamesCalligeros[m]Thats stupid, why does it not just use the card names provided by the driver like it does without smbios
#asahi-dev · 2026-01-28 21:12
jannauthe issue is that the information does not match. it's trying to open macaudio/AppleInc.-MacBookAir101--J313.conf
#asahi-dev · 2026-01-28 21:08
JamesCalligeros[m](Or we could just not use it and let all the little sysinfo utilities figure out devicetree like kfsolid did)
#asahi-dev · 2026-01-28 21:07
JamesCalligeros[m]So the smbios tables need to be very complete :p
#asahi-dev · 2026-01-28 21:07
JamesCalligeros[m]Yeah ok thatd be it then
#asahi-dev · 2026-01-28 21:06
jannaubut same error
#asahi-dev · 2026-01-28 21:06
jannautext though
#asahi-dev · 2026-01-28 21:06
jannauno
#asahi-dev · 2026-01-28 21:05
JamesCalligeros[m]Does alsaucm dump json still work?
#asahi-dev · 2026-01-28 21:05
jannaumy best guess is alsaucm, pipewire or wireplumber using SMBIOS for matching instead of devicetree
#asahi-dev · 2026-01-28 21:03
jannauany good theories how SMBIOS can break speaker and headphone playback? everything seems to work as expected but I don't here a thing. KDE volume apple shows palyback in the meter
#asahi-dev · 2026-01-28 20:55
jannaunever ;) fix pushed to asahi-wip, dropping 30-bit formats instead
#asahi-dev · 2026-01-28 20:48
chadmedcan we drop support for it now
#asahi-dev · 2026-01-28 18:12
jannaupossibly a 12.3/12.4 DCP firmware issue
#asahi-dev · 2026-01-28 17:56
jannausigh: "UPPipeDCP_H13P.cpp:3302: IOMFB verify_surfaces: No support for format l10r"
#asahi-dev · 2026-01-28 13:05
nickchanit seem that compressed dtbs generate too much spam when decompressing, and compressing dtb as a whole don't work
#asahi-dev · 2026-01-28 12:12
integralpiloti've decoded mov & basic arithmatic now, it's not changed too much and i think i've got the format fully now. but vertex shaders have changed a lot, looking into it now

plus ›