#asahi · 2021-07-29 16:28
jannauthe generic arch linix arm rootfs works without problems#asahi · 2021-07-29 16:28
jannausven: https://gist.github.com/jannau/0af6697bb8b08a847ecc32a8a3d3a4f2 is based on the arch linux arm distro config#asahi · 2021-07-08 05:33
jannaubgb: when running linux? usb-c connection has to be manged by SW. data also only works for devices plugged in during boot#asahi · 2021-07-07 19:13
jannauthe hypervisor in m1n1 is implemented and marcan has started to trace various display subsystem parts#asahi · 2021-07-07 19:10
jannauX3r0: https://twitter.com/alyssarzg/status/1412600093505589251 GPU user space, kernel space GPU support is not that far since the plan is to RE that through observing Mac OS under a thin hypervisor#asahi · 2021-07-06 08:50
jannauhttps://gist.github.com/jannau/0af6697bb8b08a847ecc32a8a3d3a4f2 is the defconfig I use with the dart/dev kernel which boots arch linux arm's general arm64 rootfs#asahi · 2021-06-29 05:03
jannauamw: there is a clang-format file, `make format` reformats the code#asahi · 2021-06-18 05:22
jannauyes, 7 seems to be no group and the gpio chip has 7 AIC irqs in the ADT#asahi · 2021-06-18 05:22
jannauI just decided to duplicate code instead of calculating the group and pin from the index#asahi · 2021-06-18 05:21
jannauI think the irq groups could use just a single range. i.e. the space between them is unused and the design allows up to 512 pins#asahi · 2021-06-18 05:19
jannaufix pushed#asahi · 2021-06-18 05:14
jannauany thoughts of updating the cache last, i.e. after the handlers have run? I want to use to skip logging uninteresting events#asahi · 2021-06-18 05:13
jannaumarcan: RegMap range lookup is broken, see https://github.com/AsahiLinux/m1n1/pull/78/commits/cd84f67ce2865ce50f2a7e19b5c54fec0a5c75a0#asahi · 2021-06-15 18:38
jannaumarcan: looks nice#asahi · 2021-06-15 12:01
jannaumarcan: the tps irq PR is marked as draft since it doesn't work reliable. For (some) register writes it's aparently necessary to include the length byte in the length. that contradicts both tps6598x TRM and the linux driver#asahi · 2021-06-15 11:46
jannaumarcan: first impression is good. splitting makes sense. I can look at it toninght (CEST) in more detail#asahi · 2021-06-13 20:12
jannaumac os writes len+1 to data[0]#asahi · 2021-06-13 20:11
jannauthe kernel does what the documentation says. write 'len' to data[0] and then append len bytes of payload#asahi · 2021-06-13 19:58
jannauclearing IntMask1 works now#asahi · 2021-06-13 19:58
jannauthey broken the protocol for writing though. the length field is included in the length#asahi · 2021-06-13 19:22
jannauso the 0x0b was length byte#asahi · 2021-06-13 19:22
jannauso it requests 9 bytes from IntEvent1 but the controller reports it has 11 bytes#asahi · 2021-06-13 19:19
jannauor more like they have a different understanding of the read lengths than the controller#asahi · 2021-06-13 19:17
jannauit looks like apple broke the protocol for certain registers#asahi · 2021-06-13 19:16
jannausven: something does not make sense: https://gist.github.com/jannau/5334a649ad4e06e710c36ac18cc679db#asahi · 2021-06-13 18:00
jannauerr, the writes are fine. #asahi · 2021-06-13 17:59
jannaumaybe, it doesn't like the smbus writes#asahi · 2021-06-13 17:57
jannausigh, writing to IntEvent1 doesn't seem to stick#asahi · 2021-06-13 17:39
jannauTPS65987DDK might make sense since it it for USB4#asahi · 2021-06-13 17:35
jannaumy i2c tracer might be broken, in m1n1 I read 11 bytes from IntEvent 1/2#asahi · 2021-06-13 17:26
jannautps65987 trm has 11 byte wide interrupt registers https://www.ti.com/lit/pdf/slvubh2#asahi · 2021-06-13 17:24
jannauor it partially respects the smbus read length. The protocol in the TRM is not smbus#asahi · 2021-06-13 17:22
jannauI wonder why IntEvent1 is 10 byte and IntMask1 is only 9#asahi · 2021-06-13 17:20
jannauyes, I parse the length, so first byte is the first data byte. sorry#asahi · 2021-06-13 17:19
jannauI might be confused#asahi · 2021-06-13 17:13
jannauthe reads include the byte with the length#asahi · 2021-06-13 17:11
jannausven: https://gist.github.com/jannau/11bd30b329a9d5c19630e6814a8b587d, irq doesn't seem to be correctly disabled though although it should#asahi · 2021-06-13 16:58
jannausven: didn't look at the register values, let me capture some traces#asahi · 2021-06-12 09:29
jannaudo we want to support the hypervisor over uart? It's currently broken since the secondary CPUs are not removed. vuart might crash m1n1 too#asahi · 2021-06-12 09:23
jannauok#asahi · 2021-06-12 09:03
jannaumarcan: do you remember why 0x23d280090 is hooked in hv.py? it seems necessary anymore and the adress is used when an usb-c port is available to the guest#asahi · 2021-06-09 16:18
jannausven: did you see drivers/usb/typec/tipd/ in the kernel? that looks like it has already the glue code between the tps6598x and dwc3#asahi · 2021-06-09 16:17
jannauslowly, hadn't much time to work on it#asahi · 2021-06-09 16:16
jannauit reads register 0x5f during boot up#asahi · 2021-06-09 16:12
jannaubut dead after removing and replugging#asahi · 2021-06-09 16:11
jannaubooting with the keyboard connected works#asahi · 2021-06-09 16:11
jannaumarcan: feel free to report the bug to apple#asahi · 2021-06-09 16:08
jannauat least after replugging. nothing was plugged in during boot#asahi · 2021-06-09 16:07
jannauand the port is dead#asahi · 2021-06-09 16:07
jannauit does not reenable interrupts for the port not removed from the ADT