#asahi-dev · 2026-08-23 18:18
svenfeel free to pull it in #asahi-dev · 2026-08-23 18:18
jannausven: are you comfortable with pulling tbt-reset-wip into the downstream kernel or do you want me to wait until you sent it / the first feedback round?#asahi-dev · 2026-08-23 17:26
chaos_princessno, that is one place where you can't do that for #reasons#asahi-dev · 2026-08-23 17:26
tryhard[m]makes sense, i thought something like asahi linux contributors or something#asahi-dev · 2026-08-23 17:26
chaos_princessyou#asahi-dev · 2026-08-23 17:25
tryhard[m]who is the maintainer?#asahi-dev · 2026-08-23 17:23
tryhard[m]dam i hoped you would say you will do them#asahi-dev · 2026-08-23 17:23
sveni don't wanna do them, i just wanted to suggest to take a look at v2 and copy/paste from there :D#asahi-dev · 2026-08-23 17:22
tryhard[m]if you wanna do that then sure#asahi-dev · 2026-08-23 17:22
tryhard[m]i'm just tryna look at the other ones#asahi-dev · 2026-08-23 17:22
chaos_princessno, but it shouldn't be that tricky for that#asahi-dev · 2026-08-23 17:22
tryhard[m]no#asahi-dev · 2026-08-23 17:21
svendt-bindings are a bit tricky if you haven't done them before, do we have some for pmpv2 already?#asahi-dev · 2026-08-23 17:20
tryhard[m]ok im adding something#asahi-dev · 2026-08-23 17:17
chaos_princesswe will care when upstreaming#asahi-dev · 2026-08-23 17:17
tryhard[m]sent a code block: https://matrix.org/oftc/media/v1/media/download/AXz_m2l4brmHH27DpZhHL8OXQgv0ag05AnCD70tckGCFjvukDwwbEQBpJUhRidCsdqUJocDTmB4r1GXKudysY0RCegd7lGBAAG1hdHJpeC5vcmcvV2pjSk5raFJMTEl1a2lzSGlMTmtzU1hD#asahi-dev · 2026-08-23 17:02
tryhard[m]think i figured it out#asahi-dev · 2026-08-23 17:01
chaos_princessno clue how to operate vim :P#asahi-dev · 2026-08-23 17:01
tryhard[m]and in vim language? lol#asahi-dev · 2026-08-23 16:59
chaos_princessM-< C-SPC M-> M-x tabify RET#asahi-dev · 2026-08-23 16:58
tryhard[m]i have like a thousand errors of these#asahi-dev · 2026-08-23 16:57
tryhard[m]sent a code block: https://matrix.org/oftc/media/v1/media/download/AfjcII2V_1p2mjeJzOIUnCHDHImkleTpMYH1kWXTky7YgWLO8xKxrk1VIfFMWbKN8av8EOH9vl7469OW7sb37WJCegd6evWgAG1hdHJpeC5vcmcvS1JxQURxVXdnbnZVdk9mTmJyVWdkS1JJ#asahi-dev · 2026-08-23 16:08
chaos_princessi guess i can switch to asahi branch for a bit#asahi-dev · 2026-08-23 16:07
jannauwhat's the pain point in pushing work in progress work to asahi-wip? automated builds of asahi-wip?#asahi-dev · 2026-08-23 16:07
chaos_princessbits/* is fine#asahi-dev · 2026-08-23 16:05
jannaumain pain point for me are the bits/* branches. I could skip pushing asahi-wip but asahi is currently at the same commit.#asahi-dev · 2026-08-23 15:58
chaos_princesswould it cause problems to not push it to -wip, but to, -wip2 or something?#asahi-dev · 2026-08-23 15:52
jannauI plan to push v7.2 based branches to asahi-wip and bits/* soon with a broken gpu driver (marked in Kconfig as such so it won't build). It might take a few days until the gpu driver is fixed but there a a few things I don't want top do twice for v7.1 and v7.2#asahi-dev · 2026-08-23 12:45
ffirefoxAre there any relevant guidelines or channels that could help me get started?#asahi-dev · 2026-08-23 12:11
chadmedi dont think anyone's actively working on that#asahi-dev · 2026-08-23 12:03
ffirefoxIs anyone currently working on this, and is there a discussion or branch I could follow to learn more?#asahi-dev · 2026-08-23 12:03
ffirefox Hi! I'm interested in the XKB/hid_apple keyboard cleanup, especially Korean Mac keyboard layouts. I have an M1 Mac with a built-in Korean-layout keyboard. #asahi-dev · 2026-08-23 07:23
tryhard[m]anyway, seems like pmp v1 driver is ready chaos_princess#asahi-dev · 2026-08-23 07:23
tryhard[m]apparently i accidentally used a stack variable for every genpd... rust doesnt always save you from memory issues!#asahi-dev · 2026-08-22 21:23
tryhard[m]seems like something when probing the gpu is failing, because now it fails even before the gpu logs anything#asahi-dev · 2026-08-22 21:16
tryhard[m]during#asahi-dev · 2026-08-22 21:16
tryhard[m]only [ 2.099022] KASAN: probably user-memory-access in range [0x0000000426c71500-0x0000000426c71507] before the crash#asahi-dev · 2026-08-22 21:16
tryhard[m]took some time to build. kasan doesnt much really#asahi-dev · 2026-08-22 20:06
chaos_princessrun with kasan#asahi-dev · 2026-08-22 20:06
chaos_princessyou are most likely smashing memory somehow#asahi-dev · 2026-08-22 20:06
tryhard[m]first line is showing it fails at the gpu bus address#asahi-dev · 2026-08-22 20:05
tryhard[m]tho i can see that the permission fault here is from the gpu, which is weird because how pmp is even related to gpu's memory?#asahi-dev · 2026-08-22 20:05
tryhard[m]so if i change to pmp_report_gfx, my driver wont crash, still not sure why. https://pastebin.com/6RfACjeM#asahi-dev · 2026-08-22 19:31
tryhard[m]in pmp v2 i can see dtsi does do the same change#asahi-dev · 2026-08-22 19:30
tryhard[m]ps_gfx should keep my driver up#asahi-dev · 2026-08-22 19:30
tryhard[m]so it should be the other way#asahi-dev · 2026-08-22 19:29
chaos_princessgood question, i think we don't manage gpu with pmp on pmpv2 devices either#asahi-dev · 2026-08-22 19:28
tryhard[m]in t8103 dtsi when changing gpu power domain ps_gfx to my pmp_report_gfx doesn't crash my pmp driver, tho if i keep ps_gfx it makes my pmp driver crash at rtkit wake with error code -62 (etime). any idea what could be the issue here?#asahi-dev · 2026-08-22 05:28
NickChan[m]chaos_princess: it also applies to a12, so maybe 5 users?#asahi-dev · 2026-08-22 00:05
JamesCalligeros[m]I dont consider limiting future hardware to a fundamentally broken lowest common denominator acceptable