#asahi-dev · 2025-08-19 14:47
svenoh well, guess they changed it at that point then #asahi-dev · 2025-08-19 14:47
nickchanand then rtkit crash#asahi-dev · 2025-08-19 14:47
nickchansven: [ 0.349736] nvme-apple 393cc0000.nvme: RTKit: Message (id=1): assert failed: [7442]:cmd fetch error for host Admin SQ, rresp code 2, status_reg: 0x1000000#asahi-dev · 2025-08-19 14:46
svenwhat happens on t6000?#asahi-dev · 2025-08-19 14:45
svenso I just tried t8103 and there it worked #asahi-dev · 2025-08-19 14:45
nickchansven: okay that does not work since you said you are going to try that#asahi-dev · 2025-08-19 14:40
nickchani am going to try t6000 nvme with the t8015 compatible#asahi-dev · 2025-08-18 19:17
svensure#asahi-dev · 2025-08-18 18:26
jannaushould we drop /var/log/Xorg.0.log from asahi-diagnose?#asahi-dev · 2025-08-18 11:21
svenlooks good otherwise, will take a closer look once it’s on the mailing list #asahi-dev · 2025-08-18 11:18
svenI would rearrange them some more, usually the dts changes are the very last commits #asahi-dev · 2025-08-18 11:00
chadmedif all is good i will write the cover letter and submit tomorrow #asahi-dev · 2025-08-18 10:59
chadmedpicked up the hid driver, added some #includes to devicetrees i missed, picked up R-b tags and rearranged commits #asahi-dev · 2025-08-18 10:59
chadmedhttps://github.com/chadmed/linux/commits/b4/macsmc-subdevs/#asahi-dev · 2025-08-18 09:06
sventhat’s what I would do #asahi-dev · 2025-08-18 09:06
chadmedsven: should i fold the ioft commit into the hwmon driver too? #asahi-dev · 2025-08-17 18:12
nickchani will mess with the rebased smc-hid driver on a11 later#asahi-dev · 2025-08-17 18:12
nickchanaddressed the review comments#asahi-dev · 2025-08-17 18:08
nickchant8015 nvme v2 will send in ~12 hours probably https://github.com/HoolockLinux/linux/tree/b4/t8015-nvme#asahi-dev · 2025-08-17 14:57
svenchadmed: https://github.com/AsahiLinux/linux/commit/24f118a8a983a2a17d69f74acafe5d9412a9225f that's the hid commit i was going to submit, feel free to include inside the hwmod+rtc series#asahi-dev · 2025-08-16 15:35
svenalright, time to rebase atcphy on -rc1 and write that cover letter i guess#asahi-dev · 2025-08-16 15:34
svenit's confusing how there's "t2 linux, running on the x86 chip and talking to apple's weird iOS clone on the t2" and "t2 linux, but actually runs on the t2" :D#asahi-dev · 2025-08-16 15:13
chaos_princessnope, 3 cases, pre-t2 x86, t2 and m-series. And idk about t1#asahi-dev · 2025-08-16 15:12
nickchanso there's only x86 and apple silicon cases#asahi-dev · 2025-08-16 15:11
nickchanT2 SMC from the point of the T2 has ~identical communication format as the M1 SMC. not sure from the intel side but i think that's about the same as non-t2 intel macs?#asahi-dev · 2025-08-16 14:26
jannauargh, why did the fastmail remove the line breaks#asahi-dev · 2025-08-16 14:25
jannau`b4 trailers` doesn't find anything after changing commit msg subject#asahi-dev · 2025-08-16 13:51
chadmedwe can get that from the las driver instead anyway #asahi-dev · 2025-08-16 13:51
jannauit's ok to handle that as a future problem (if it ever is)#asahi-dev · 2025-08-16 13:50
jannauthe only thing a can think of is the lid angle in hid#asahi-dev · 2025-08-16 13:49
jannauI was only saying that they bring their own float handling and do not use a common helper#asahi-dev · 2025-08-16 13:49
chadmedim mostly just concerned about the prospect of needing to read 'flt ' keys in other smc subdevices at some point in the future#asahi-dev · 2025-08-16 13:48
chadmedi dont think citing bcachefs as precedent is going to make us any friends right now :P #asahi-dev · 2025-08-16 13:47
jannauhid iio sensors match too, amdgpu and bcachefs#asahi-dev · 2025-08-16 13:45
jannaugrepping for exponent reveals a few drivers with custom float handling, best match is probably drivers/hwmon/pmbus/pmbus_core.c#asahi-dev · 2025-08-16 13:44
chadmedalright tomorrow night i will just yeet it into the hwmon driver and make a note of it in the cover letter#asahi-dev · 2025-08-16 13:25
chadmedif i throw this in the hwmon driver and we end up needing it for something else thats going to suck#asahi-dev · 2025-08-16 13:23
chadmedquick grep reveals no such helper#asahi-dev · 2025-08-16 13:23
chadmedour attempt at that went nowhere#asahi-dev · 2025-08-16 13:22
svenand the "get resources inside mfd/" and "put the code that actually talks to SMC into drivers/soc or driver/platform" actually makes sense if we (or hopefully someone else...) ever add support for t2 or x86 SMCs#asahi-dev · 2025-08-16 13:22
svenhave we ever checked if there's already a helper like that somewhere? :D#asahi-dev · 2025-08-16 13:21
chadmedi somewhat agree that a generic "turn 4 bytes into a float" helper probably shouldnt live in a driver. the issue is that im not terribly sure that this is a can of worms we want to open#asahi-dev · 2025-08-16 13:21
chadmedand i would like 100 billion trillion dollars#asahi-dev · 2025-08-16 13:17
sven*inside#asahi-dev · 2025-08-16 13:17
svenessentially what he'd really like is a small driver in mfd/ that just gets resources and then all the SMC api insider drivers/soc or whatever#asahi-dev · 2025-08-16 13:16
svenjust take a look at the other mfd drivers, smc is just weird#asahi-dev · 2025-08-16 13:15
chadmed"if your api isnt what i like then move it in to platform but i already told you not to use my api in platform" #asahi-dev · 2025-08-16 13:15
chadmed"no dont put this driver in platform/ i dont want my api in platform"#asahi-dev · 2025-08-16 13:15
chadmedthat whole message is a pisstake#asahi-dev · 2025-08-16 13:11
sveni think i just removed the function then instead of trying to solve it