#asahi-dev · 2025-11-29 08:06
jannauthat looks wrong as well unless you're using the debug uart#asahi-dev · 2025-11-29 08:04
jannauintegralpilot: I think it's caused by the missing spmi USB-PD support. uartproxy_run() probes down from IODEV_MAX. try starting with "iodev_id_t iodev = IODEV_USB0;"#asahi-dev · 2025-11-29 08:04
integralpilotInitializing hypervisor over iodev 0#asahi-dev · 2025-11-29 08:04
integralpilotjust checked#asahi-dev · 2025-11-29 07:57
integralpiloti'll check#asahi-dev · 2025-11-29 07:55
jannaudo you have a line "Initializing hypervisor over iodev N"?#asahi-dev · 2025-11-29 07:55
integralpilotwait actually self.iodev must be reporting as 6 then#asahi-dev · 2025-11-29 07:53
integralpilotyes, the one next to magsafe. weird it's coming up as index 3 then.#asahi-dev · 2025-11-29 07:53
jannauthe one next to the magsafe port? apple uses "left back" to describe it. yes, that should be the dfu port#asahi-dev · 2025-11-29 07:51
integralpiloti'm plugged into what i think is the DFU port, leftmost port on the left side (which has 2 ports)#asahi-dev · 2025-11-29 07:50
jannausee enum _iodev_id_t in src/iodev.h#asahi-dev · 2025-11-29 07:49
jannauindex 3 is usb1, should be the left front port#asahi-dev · 2025-11-29 07:46
jannauisn't the uart index 0?#asahi-dev · 2025-11-29 07:45
integralpilotor is it most likely it's deciding the IODEV index wrong#asahi-dev · 2025-11-29 07:45
integralpilotmight it be that the IODEV index is actually 3 but that isn't a 1:1 mapping to the PMGR index like the code assumes?#asahi-dev · 2025-11-29 07:45
jannauyeah, no doubt that only the dp-phy of port3 is used to drive the hdmi port#asahi-dev · 2025-11-29 07:45
integralpilotyeah#asahi-dev · 2025-11-29 07:44
jannauthe main issue to solve is why the python code thinks that usb3 is used#asahi-dev · 2025-11-29 07:44
integralpilotbut there's definetly no USB-related items in the device tree so i think it's safe to conclude it's the HDMI port#asahi-dev · 2025-11-29 07:43
integralpilotoh, i guess that makes sense#asahi-dev · 2025-11-29 07:43
jannauyou can't deduce "usb-drd3 = no ATC3_USB". don't expect logic or internal consistency from the ADT. It's not at the same level as linux devicetrees#asahi-dev · 2025-11-29 07:39
integralpilotcause no usb-drd3 = no ATC3_USB = the error. it shouldn't be trying to map the HDMI controller as the controller it's talking through#asahi-dev · 2025-11-29 07:38
jannaumissing usb-drd3 and dart-usb3 are expected. that's was always the case#asahi-dev · 2025-11-29 07:38
integralpilotatc3 definetly seems like the HDMI port. the python logic seems to try to register the contorller it thinks it is talking over, so need to figure out why it thinks this is the HDMI controller#asahi-dev · 2025-11-29 07:37
jannaumaybe related to the missing spmi usb-pd controller support? although I don't see how#asahi-dev · 2025-11-29 07:36
integralpilotbut atc-phy3, atc3-dpxbar, atc3-dpphy exist#asahi-dev · 2025-11-29 07:35
integralpilotthere's no usb-drd3 (likely causing the issue) and no dart-usb3 (whereas 0, 1 and 2 have these)#asahi-dev · 2025-11-29 07:35
integralpilotokay just dumped my device tree#asahi-dev · 2025-11-29 07:33
integralpilotit seems that the python code thinks that the USB port host is talking to target over is controlled by ATC3_USB which can't be right#asahi-dev · 2025-11-29 07:30
jannaupossible that it is solved differently on m3 pro or that they removed unneeded parts from the adt#asahi-dev · 2025-11-29 07:28
jannauthere is/was kinda: the 4th controller is used to drive the hdmi port on m1/m2 pro/max macbooks. "ATC3_USB" exists on those machines#asahi-dev · 2025-11-29 07:23
integralpiloti'm trying to find out rn#asahi-dev · 2025-11-29 07:23
integralpilotwhich raises the question - why there an IODEV index for a non-existent USB controller?#asahi-dev · 2025-11-29 07:23
integralpilotlooks like for each IODEV index it assumes a matching PMGR exists in the ADT#asahi-dev · 2025-11-29 07:20
integralpiloti'm taking a look at the python code now to see what's causing it to assume the existance of atc3#asahi-dev · 2025-11-29 07:10
integralpilotI think the problem is that I my mac only has 3 thunderbolt ports, not 4, so ATC-3_USB isn't defined in the device tree, only 0, 1 and 2#asahi-dev · 2025-11-29 06:58
integralpilotcan provide more debugging info if needed#asahi-dev · 2025-11-29 06:57
integralpilotany idea what this is & how to fix? thanks so much!#asahi-dev · 2025-11-29 06:57
integralpilotin map_essential.py#asahi-dev · 2025-11-29 06:57
integralpiloton the line 'dev = dev_by_name[name]#asahi-dev · 2025-11-29 06:57
integralpilothowever, now when i try to run proxyclient/tools/run_guest.py to load XNU, i get KeyError: 'ATC-3_USB' during "Doing essential MMIO remaps..."#asahi-dev · 2025-11-29 06:56
integralpilotgot the right cable, and did the kext setup and now macdvmtool is working for me#asahi-dev · 2025-11-28 09:30
integralpilotthanks so much everyone for trying to help me get set up for M3 dev! :)#asahi-dev · 2025-11-28 09:30
integralpilotthanks, maz. unfortunetly looked around and i don't actually have any of the right cables (none display-port capable either). i'll get one tomorrow.#asahi-dev · 2025-11-28 09:17
mazintegralpilot: if your cable is display-port capable, you should be good.#asahi-dev · 2025-11-28 09:11
integralpiloti'm not sure if i have a USB 3.x / thunderbolt 3/4 cable, i'll go look for one, thanks#asahi-dev · 2025-11-28 08:57
chaos_princesshttps://github.com/AsahiLinux/macvdmtool#asahi-dev · 2025-11-28 08:57
chaos_princesstry using the real serial port and turning off the virtual one in m1n1#asahi-dev · 2025-11-28 08:56
integralpilothost is macbook air m3, target is macbook pro m3 pro#asahi-dev · 2025-11-28 08:56
chaos_princesswhat is your other device?