#
#asahi-dev · 2026-10-27 09:17
yukapeople came up with very cursed ways of tracing this
#asahi-dev · 2026-10-27 09:16
svenugh..
#asahi-dev · 2026-10-27 09:16
jannauit doesn't even have thunderbolt/usb4
#asahi-dev · 2026-10-27 09:16
chaos_princessneo has a different kext
#asahi-dev · 2026-10-27 09:16
yukabut not sufficiently similar for usb3 to work
#asahi-dev · 2026-10-27 09:16
yukait's very similar to M3
#asahi-dev · 2026-10-27 09:16
yukano
#asahi-dev · 2026-10-27 09:16
svenoh, does the Neo use the "new" atcphy?
#asahi-dev · 2026-10-27 09:16
yukasince there is no other usb controller to connect the proxy to
#asahi-dev · 2026-10-27 09:15
yukaThe annoying part is actually knowing the needed mmio writes
#asahi-dev · 2026-10-27 09:15
sven*take care
#asahi-dev · 2026-10-27 09:15
svenand you can skip all the DP mess. and I guess you don't need to take of synchronization between dwc3 and atcphy since we control all of that anyway
#asahi-dev · 2026-10-27 09:14
svenhrm, true. you probably only need to get cable orientation from the pd controller and that's just a single mmio write inside atcphy iirc
#asahi-dev · 2026-10-27 09:14
yukaso basically can be a static sequence of mmio writes
#asahi-dev · 2026-10-27 09:13
yukamaybe it's not as annoying as you think because it requires only one specific configuration
#asahi-dev · 2026-10-27 09:13
svenbut the annoying part is atcphy bringup
#asahi-dev · 2026-10-27 09:13
svenI think you also need to adjust the dwc3 acm cdc stuff a bit to make use of usb3 packet sizes
#asahi-dev · 2026-10-27 09:13
yukanope
#asahi-dev · 2026-10-27 09:12
svenI guess we still don't know how to bypass that usb hub on the Neo? :/
#asahi-dev · 2026-10-27 09:12
snooze6214[m]hmm okay, thanks, I'll take a look
#asahi-dev · 2026-10-27 09:11
svenbut the code for that should be inside the linux atcphy driver already
#asahi-dev · 2026-10-27 09:11
svensnooze6214[m]: it would be very annoying and tedious since it requires atcphy bringup inside m1n1
#asahi-dev · 2026-10-27 09:10
myriahttps://docs.google.com/spreadsheets/d/14wlj1sbkqplcxlUPqWSno1v4XaCQdzX9XCfqX76ytuQ/edit?usp=sharing
#asahi-dev · 2026-10-27 09:10
yukamyria: do you have the link again?
#asahi-dev · 2026-10-27 09:08
snooze6214[m]yuka: you mentioned device mode would work on the mb neo if it's using usb 3 right? how hard is it to implement this for someone who's new to this?
#asahi-dev · 2026-10-27 09:07
myriayuka: added! Anything else I've missed?
#asahi-dev · 2026-10-27 09:03
yukathe previous ones are aligned
#asahi-dev · 2026-10-27 09:03
yuka0x5ff8 bytes?
#asahi-dev · 2026-10-27 09:03
yuka> RTKit: buffer request for 0x5ff8 bytes at 0x000001008025d000
#asahi-dev · 2026-10-27 08:35
yukabut setting it to 0x20001 does not fix it
#asahi-dev · 2026-10-27 08:35
yukaoutbox_ctrl = = 2ff01
#asahi-dev · 2026-10-27 08:33
chaos_princessbut please re-check
#asahi-dev · 2026-10-27 08:33
chaos_princessiirc outbox ctrl is already set to the correct value
#asahi-dev · 2026-10-27 08:31
jannauI think the OUTBOX_CTRL difference exists in 12.x and 13.x as well and for other co-procs as well. I don't think it's the problem if you can start endpoints and get messages back from dcp
#asahi-dev · 2026-10-27 08:26
yukano, this seems to be i2a_control in the linux driver. it's never written
#asahi-dev · 2026-10-27 08:25
yukathe OUTBOX_CTRL difference looks most sus to me, but maybe linux reads it and finds it already has the correct value?
#asahi-dev · 2026-10-27 08:22
yukaInvalidate Stream: 0/5/4 (linux) vs Invalidate all (macOS)
#asahi-dev · 2026-10-27 08:21
yukaOUTBOX_CTRL = 0x20001 (FIFOCNT=0x0, OVERFLOW=0, EMPTY=1, FULL=0, RPTR=0x0, WPTR=0x0, ENABLE=1) (macOS only)
#asahi-dev · 2026-10-27 08:21
yukaCPU_CONTROL = 0x10 (RUN=1) (linux only)
#asahi-dev · 2026-10-27 08:21
yukadifferences in M3 Max dcp init (macOS vs https://github.com/AsahiLinux/linux/pull/608):
#asahi-dev · 2026-10-27 08:09
yukaMichaelReeves[m] myria
#asahi-dev · 2026-10-27 08:07
yukacan someone who has a google account add my last submissions there?
#asahi-dev · 2026-10-27 08:07
yukawe had a google docs patch tracker somewhere
#asahi-dev · 2026-10-27 08:03
yukaInitial M5 device trees sent
#asahi-dev · 2026-10-27 07:29
svenhah, awesome
#asahi-dev · 2026-10-27 07:29
jannauit == power button
#asahi-dev · 2026-10-27 07:29
jannaudoesn't look like m4 mac mini would need soldering. you can remove it and have two nice contacts: https://support.apple.com/en-us/120996
#asahi-dev · 2026-10-27 07:22
svenso no soldering but just a nvram update + "smart" external power switch then
#asahi-dev · 2026-10-27 07:22
jannauand since 26.x auto-boot no longer requires that it lost power while powered-on
#asahi-dev · 2026-10-27 07:22
svenno idea. it just always did that and then stopped at some point and I haven't bothered to figure out why

plus ›