#
#asahi-dev · 2026-05-30 11:41
jannauworking wlan on t6030 (m1n1 init seems to similar enough to t8122)
#asahi-dev · 2026-05-30 11:36
yukaIf we figure it out properly, even better. Otherwise I implemented the trial and error guessing: https://github.com/AsahiLinux/kisd/pull/8
#asahi-dev · 2026-05-30 11:36
yukaI couldn't find any such packets in my dumps of m1 pro/max, m3 max, m4 pro.
#asahi-dev · 2026-05-30 11:09
jannauyuka: it was before sending data to the target. I'm unsure wether it was a packet containing data from the target. it has the full address we need for sending data
#asahi-dev · 2026-05-30 11:06
jannauchadmed: I don't remember but probably on the hdmi port
#asahi-dev · 2026-05-30 10:58
yukaif the device only responds with this address after we send a request with the correct base, it's of no use
#asahi-dev · 2026-05-30 10:58
yukajannau: the correct address as in what we listed in the readme? or with which offset? I primarily searched in the initial messages which happen before mac os starts sending requests to the device using this base.
#asahi-dev · 2026-05-30 10:50
chadmedthe avservice ddc functions abstract away the low level transport which is nice, but idk about trying to shove that into a fake linux i2c device
#asahi-dev · 2026-05-30 10:49
chadmedover displayport too? im curious how that would look with dp aux packets
#asahi-dev · 2026-05-30 10:48
jannauchadmed: iirc there was an interface that looked like an i2c bridge which worked at least for reading the edid
#asahi-dev · 2026-05-30 10:46
jannauI did not save them but I searched the packet data for the high bytes of the address (little endian)
#asahi-dev · 2026-05-30 10:45
yukaif you happen to have the pcaps of that, I would be interested
#asahi-dev · 2026-05-30 10:42
jannauyuka: I've seen a packet from the device which has the correct address on some pro/max devices
#asahi-dev · 2026-05-30 10:42
yukaotherwise the only path I see is binary reversing of the deviceinterfaced
#asahi-dev · 2026-05-30 10:42
yukait's obviously not what mac os does, but it would work for all devices I tested so far
#asahi-dev · 2026-05-30 10:41
yukado you think this is too hacky?
#asahi-dev · 2026-05-30 10:41
yukaI thought about base guessing for kisd a bit more, and now that we have much more data it seems doable to simply read from +0x0, +0x400000 and +0x700000 respecitvely and check if the thing there looks like a dockchannel uart
#asahi-dev · 2026-05-30 10:29
chadmedthe ddc api takes standard mccs/vcp commands and data but obviously the api looks nothing like an i2c device so thats going to be fun
#asahi-dev · 2026-05-30 10:25
sofusdruz: i dont think so. h264 is _mostly_ complete
#asahi-dev · 2026-05-30 10:11
yukaAnd it seemed to work fine using the existing macsmc-gpio instance and pin 8
#asahi-dev · 2026-05-30 10:11
yukafunction-reg_on = 137:pKW8('pcIO', 0x800000, 0x80000)
#asahi-dev · 2026-05-30 10:11
yukaThe macbook neo has this for wifi enable
#asahi-dev · 2026-05-30 10:10
jannauthis seems to be the least annoying solution
#asahi-dev · 2026-05-30 10:10
yukanice, thank you
#asahi-dev · 2026-05-30 10:10
jannauI settled for adding additional macsmc-gpio instances to handle this. already running locally and writing commit messages right now
#asahi-dev · 2026-05-30 10:07
yukathough these work differently, they don't have the number encoded in the hex name
#asahi-dev · 2026-05-30 10:05
yukaquick search through the device trees also gives me bp, pm
#asahi-dev · 2026-05-30 10:05
yukaThere are possibly more smc-gpio 'banks' than gP and gp
#asahi-dev · 2026-05-30 09:32
jannauack, those are the missing memory channels
#asahi-dev · 2026-05-30 09:30
yukahttps://github.com/AsahiLinux/linux/pull/506
#asahi-dev · 2026-05-30 09:23
yukaamcc{6,7}, dcs_{24-31}
#asahi-dev · 2026-05-30 09:22
yukawhich probably means those nodes should be disabled in the linux device tree
#asahi-dev · 2026-05-30 09:22
yukaThe t6034 pmgr has more nodes with no_ps flags
#asahi-dev · 2026-05-30 09:03
jannausven: replied to the v3 mail. do you want to pick it up from the v2 series or shall I resend it
#asahi-dev · 2026-05-30 08:18
jannauno, I don't see apple,pmgr being picked up. no idea why I thought it was picked up and dropped it from v3
#asahi-dev · 2026-05-30 08:10
svenI manually cherry picked it when running the check
#asahi-dev · 2026-05-30 08:10
svenhuh, wasn’t that one picked up as well?
#asahi-dev · 2026-05-30 08:06
jannauat least dtbs_check complains about it in next-20260529
#asahi-dev · 2026-05-30 08:05
jannauthe apple,pmgr dt-binding change appears to be missing from next
#asahi-dev · 2026-05-30 08:04
jannauhmm, it's testing commit "[7324/9137] arm64: dts: apple: Initial t8122 (M3) device trees" so it's probably just missing the dt-binding changes that went through other trees
#asahi-dev · 2026-05-30 08:03
jannausven: I've seen it but I'm confused. the b4 branch passed with "apple,pmgr" and "apple,i2c" changes which ere already picked
#asahi-dev · 2026-05-30 07:13
sveni thought i ran the dt checks and everything was fine
#asahi-dev · 2026-05-30 07:13
svenjannau: huh, did you see https://lore.kernel.org/oe-kbuild-all/202605292354.YWmq9NMc-lkp@intel.com/ ?
#asahi-dev · 2026-05-30 03:03
chadmedreads work and are accurate, and the monitor acks writes but it just doesnt actually do anything. i know the monitor does actually support it though so maybe the tracer is throwing off timings
#asahi-dev · 2026-05-30 03:02
chadmedunfortunately.... it does not seem to work :P
#asahi-dev · 2026-05-30 03:02
chadmedfound ddc/ci, as suspected it's just a pair of commands in dcpavserv to read/write
#asahi-dev · 2026-05-30 01:15
yukabluetooth fails: "no brcm,taurus-bf-cal-blob property", "probe with driver hci_bcm4377 failed with error -2"
#asahi-dev · 2026-05-30 01:13
yukabut hey, wifi works, this is quite usable already :-)
#asahi-dev · 2026-05-30 01:12
yukaAlso getting lots of deferred probe from power-management (AFNC0_LS does not want to reach its target state)
#asahi-dev · 2026-05-30 01:11
yukaI'm now also running into the gpio issue, specifically that I can't get wifi and touchpad to work at the same time

plus ›