#asahi-dev · 2022-06-27 14:38
svenurhg... broadcom... why do they make the device2host queue index configurable but only for one packet type (HCI) but just hardcode it and ignore whatever I configure for another one?! (ACL)#asahi-dev · 2022-06-27 08:35
svenit's probably somewhere in the ADT#asahi-dev · 2022-06-27 08:09
svenand macvdmtool first switches the local mux to UART and then instructs that CD321x chip to talk to the remote chip and switch it to UART as well#asahi-dev · 2022-06-27 08:08
svenhttps://github.com/AsahiLinux/macvdmtool/blob/main/main.cpp#L318 readRegister just sends a read register command over i2c and command just sends a command to that CD321x chip (https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/drivers/usb/typec/tipd/core.c#n288)#asahi-dev · 2022-06-27 08:06
svenyeah, i'm not sure how long that mode will survive#asahi-dev · 2022-06-27 08:04
svenmacvdmtool does just that. iirc the source code even contains the i2c commands you have to send#asahi-dev · 2022-06-27 08:04
svenyou can talk to that CD321x chip via i2c from the AP and mux out UART#asahi-dev · 2022-06-26 14:20
sven\o/#asahi-dev · 2022-06-26 14:20
sven[NEW] Device XX:XX:XX:XX:XX:XX photon#asahi-dev · 2022-06-26 14:20
sven[bluetooth]# devices#asahi-dev · 2022-06-26 13:56
svenstill broken, but that looks like progress \o/#asahi-dev · 2022-06-26 13:56
sven[ 3.234773] Bluetooth: hci0: command tx timeout#asahi-dev · 2022-06-26 13:56
sven[ 3.234754] Bluetooth: hci0: Opcode 0x2002 failed: -110#asahi-dev · 2022-06-26 10:47
svenhmmm… like like I don’t quite understand how device to host IPC works yet #asahi-dev · 2022-06-26 10:31
sven[ 3.233935] Bluetooth: hci0: Opcode 0x c03 failed: -110 :-(#asahi-dev · 2022-06-23 19:26
svennext step is to figure out the mess that is the other direction and then finally how to hook this all up to the bluetooth subsystem#asahi-dev · 2022-06-23 19:26
svenhost -> device IPC works :)#asahi-dev · 2022-06-23 19:26
sven[ 1.170766] btbcmpci 0000:01:00.1: entry in d2h queue #0 with msg_id 1#asahi-dev · 2022-06-23 19:26
sven[ 1.170754] btbcmpci 0000:01:00.1: d2h queue #0: head: 2, tail: 1#asahi-dev · 2022-06-23 19:26
sven[ 1.170742] btbcmpci 0000:01:00.1: d2h interrupt#asahi-dev · 2022-06-22 14:47
svenI’d certainly have use for a mbp that’s “broken” as well :D#asahi-dev · 2022-06-22 14:42
sveni don't think there'll be any issues :)#asahi-dev · 2022-06-22 14:41
sveni have no screen connected to my mac mini which I use as a target for development and which also is my only m1 running linux#asahi-dev · 2022-06-21 19:34
svennice! And it even comes with documentation :)#asahi-dev · 2022-06-21 18:52
svenpretty sure I just had a custom reload function that really just ran my script again at one point #asahi-dev · 2022-06-21 18:52
svenI think that does something else :D#asahi-dev · 2022-06-21 18:50
svenhrm, I think I actually hacked reload in some script myself. run_script to start the same thing again should work though #asahi-dev · 2022-06-21 18:47
svenOr yeah, just run the script again #asahi-dev · 2022-06-21 18:47
svenjust reload iirc #asahi-dev · 2022-06-21 18:36
svenso it fixes a bug that’s rarely hit? I’m just confused how that commit is related to a asahi rebase #asahi-dev · 2022-06-21 18:29
svenwhat’s so special about that commit?#asahi-dev · 2022-06-21 14:40
svenyeah :/#asahi-dev · 2022-06-21 14:38
svenwe're installing the 12.3 DCP firmware right now anyway. it'll only be required for M2 so we can just ignore it for now ;)#asahi-dev · 2022-06-19 20:42
svennow let's see about these complex IPC structures. so far I could drop most of the mystery init pokes as usual#asahi-dev · 2022-06-19 20:41
svenwith almost-sane code now :D#asahi-dev · 2022-06-19 20:41
sven[ 1.174772] applebt 0000:01:00.1: firmware has booted (2)#asahi-dev · 2022-06-19 20:41
sven[ 0.735752] applebt 0000:01:00.1: firmware name: brcmbt4378b1-apple,atlantisb-m.bin; ptb name: brcmbt4378b1-apple,atlantisb-m.ptb#asahi-dev · 2022-06-19 20:29
svenyuni: yup :-(#asahi-dev · 2022-06-19 19:55
sveni got a bit sidetracked with what I assumed to be a quick and fun driver (Bluetooth) to take a break from that atcphy mess#asahi-dev · 2022-06-19 19:55
svenbriefly#asahi-dev · 2022-06-19 18:53
svenat least that confirms that this abomination actually requires the msi doorbell addressed to be configured inside some register in the BAR though#asahi-dev · 2022-06-19 18:32
svenit booted almost immediately but it hated me because I didn't setup irqs correctly#asahi-dev · 2022-06-19 18:32
sven*sigh*#asahi-dev · 2022-06-19 14:36
svenOTP parsing works just the same. let's see if I can actually get it to boot that firmware now#asahi-dev · 2022-06-19 14:35
sven[ 0.727909] applebt 0000:01:00.1: firmware name: brcmbt4378b1-apple,atlantisb-m.bin; ptb name: brcmbt4378b1-apple,atlantisb-m.ptb#asahi-dev · 2022-06-19 14:35
sven[ 0.727905] applebt 0000:01:00.1: OTP: revision=b1, vendor=m#asahi-dev · 2022-06-19 12:11
svenyeah#asahi-dev · 2022-06-19 12:11
svenohhhh…. duh#asahi-dev · 2022-06-19 10:35
svenokay, looks like that's the bluetooth OTP then. unfortunately it doesn't seem to contain that "USI" or "MUR" string that's part of the bt firmware name#asahi-dev · 2022-06-19 10:33
svenmarcan: ^--