#
✕ nick:integralpilot
#asahi-dev · 2026-06-24 08:41
integralpilotit is also used for a weird serial protocol we call "dockchannel uart", though its not really full uart
#asahi-dev · 2026-06-24 08:41
integralpiloton M2+ it is used to talk to the inbuilt keyboard and trackpad on macbooks
#asahi-dev · 2026-06-24 08:40
integralpilotSobek[m]: its an apple custom FIFO
#asahi-dev · 2026-06-24 08:26
integralpilot*dockchannel-uart tty driver
#asahi-dev · 2026-06-24 08:26
integralpilotjust finished the first draft
#asahi-dev · 2026-06-24 08:26
integralpilotokay i am testing my dockchannel tty driver (that uses the mailbox) now. will report back how it goes
#asahi-dev · 2026-06-24 08:10
integralpilotis that not ideal?
#asahi-dev · 2026-06-24 08:07
integralpilotand with tty the faking and implementing meaningless functions isnt needed
#asahi-dev · 2026-06-24 08:06
integralpilotand i think faking etc. is not really ideal for upstream
#asahi-dev · 2026-06-24 08:06
integralpiloti might be remembering wrong though!
#asahi-dev · 2026-06-24 08:05
integralpiloti think with uart drivers, implementing (which you would have to fake) baud, line control, a `struct uart_port` is required, not optional (unlike tty) and you would have to implement a lot of meaningless functions like set_termios, tx_empty, get_mctrl etc.
#asahi-dev · 2026-06-24 07:59
integralpilotyes, i believe so
#asahi-dev · 2026-06-24 07:57
integralpilota plain TTY driver is probably better
#asahi-dev · 2026-06-24 07:57
integralpilotyuka: are you sure you want a uart driver? there's no baud/termios/modem state etc. for dockchannel (at least to my knowledge), which i think uart_port needs
#asahi-dev · 2026-06-24 06:12
integralpiloti'll try it out later tonight
#asahi-dev · 2026-06-24 06:05
integralpiloti am honestly fine with either but probably leaning towards #1 just as its cleaner and gives us more options later when we do serial, hb you guys?
#asahi-dev · 2026-06-24 06:05
integralpilot2. have a TTY driver for DockChannel. and then the HID driver accesses that through serdev. more complex though.
#asahi-dev · 2026-06-24 06:04
integralpilot1. mailbox for HID. for serial: either coalesced mailbox, or just access raw hardware regs directly in the TTY driver (upstream might not like, but if justifiable for speed might be ok)
#asahi-dev · 2026-06-24 06:04
integralpilotso i guess we got 2 options:
#asahi-dev · 2026-06-24 06:03
integralpilotthough that would make things a lot more complex
#asahi-dev · 2026-06-24 06:02
integralpilotthat might actually be the neatest solution but maybe i'm missing something
#asahi-dev · 2026-06-24 06:02
integralpiloti guess the other option is we make a DockChannel TTY/UART controller driver, and then the HID driver consumes it through serdev
#asahi-dev · 2026-06-24 06:01
integralpilotwould it be fine upstream to have the DockChannel mailbox driver for HID, and then a seperate TTY/serial driver that will access hardware regs directly for serial?
#asahi-dev · 2026-06-24 05:56
integralpilotif we do upstream HID now with mailbox, and then it doesn't work for serial with whatever reason, when serial is downstream we can kinda just do whatever. but if serial wanted to go upstream, would they accept a) moving the already existing driver or b) having a seperate lightweight tty driver for dockchannel serial uses vs mailbox for other stuff
#asahi-dev · 2026-06-24 05:55
integralpiloti still keep thinking about it and i can't decide if mailbox is right
#asahi-dev · 2026-06-24 05:55
integralpilotoh doesnt matter then lol!
#asahi-dev · 2026-06-24 05:47
integralpilotoh wait * isn't italtics, its _ - i forgot sorry!
#asahi-dev · 2026-06-24 05:44
integralpiloti think mailbox *should* be fine now for serial too, but not 100% sure
#asahi-dev · 2026-06-24 05:34
integralpilotand thank you for those ideas on how to improve it so uart would (in future) work better without as much modifications!
#asahi-dev · 2026-06-24 05:34
integralpilotmy thoughts are that for now, where HID is the only one we really know fully about and have working, its okay to get it working through the mailbox system. when we get around to uart, we can try out the mailbox system for that, but if it doesn't work, then we can look into alternatives (but, i think it should work for uart with the TX workqueue gone and no more per-RX kmalloc etc. which I've done)
#asahi-dev · 2026-06-24 05:26
integralpiloti think with the fixes i just worked on, it should be okay at dealing with lots of tiny uart messages. if it coalesces writes into chunks that will be fine, but if every byte is a mailbox message that wouldn't be ideal
#asahi-dev · 2026-06-24 04:50
integralpilotjust pushed new version of MTP addressing the feedback, thanks again for it :)
#asahi-dev · 2026-06-24 02:10
integralpilotbtw, thank you for your feedback, i'll get around to addressing that soon! :)
#asahi-dev · 2026-06-24 02:10
integralpilotbut if you think it would be better in another subsystem happy to do that, i don't really mind, i think it's a bit of a weird thing and could fit in many places
#asahi-dev · 2026-06-24 02:09
integralpilotjannau: looking at the logs, i think we (me, you, sven) decided to do it as a mailbox on 2026-01-25
#asahi-dev · 2026-06-23 13:23
integralpilotdone: https://github.com/IntegralPilot/linux2/pull/1
#asahi-dev · 2026-06-23 13:15
integralpilotit was handy because last time i worked on this i already mostly completed the re-write, just left it locally!
#asahi-dev · 2026-06-23 13:14
integralpilotimo its ready for upstream (i ran all the dt-binding and formatting tests etc) but probably good to see if anyone has any thoughts or sees something i miss beforehand
#asahi-dev · 2026-06-23 13:14
integralpiloti will push my current status to a branch and then drop the link here, if anyone has time to review / give thoughts would really appreciate it
#asahi-dev · 2026-06-23 13:13
integralpilotIT WORKS! yay
#asahi-dev · 2026-06-23 12:52
integralpilotbtw finished DockChannel mailbox re-write, i'm just testing it for the first time now (fingers crossed!)
#asahi-dev · 2026-06-23 12:52
integralpilotperfect, thanks!
#asahi-dev · 2026-06-23 12:35
integralpilotmultiple posts now from people on the sub complaining they updated to 27 beta and now battery doesn't work. i replied to them with the explanation, but arghhh when will people learn to stop installing beta software and expecting everything to work (and then complaining we don't support it yet!)
#asahi-dev · 2026-06-23 05:15
integralpilottracks for other sizes (64, 256, 1024, 8192, 16384) that core 4 is always about 3x slower than the other p-clores (5,6,7)
#asahi-dev · 2026-06-23 05:15
integralpilotfor AES-256-GCM, core 4 is 58941.07k bytes/sec with 16 bytes size while 5,6,7 are ~150k
#asahi-dev · 2026-06-23 05:13
integralpilotjannau: yes, i am on base m3. tested with openssl speed and can confirm core 4 is much slower than the other perf cores for me as well
#asahi-dev · 2026-06-21 10:02
integralpilotjannau: thanks! yes, i do, i will check now
#asahi-dev · 2026-06-21 09:38
integralpiloti think i remember before i was working on MTP/DockChannel upstreaming, if that's still needed (i think we needed to rewrite dockchannel as a mailbox?) happy to help with that :)
#asahi-dev · 2026-06-21 09:37
integralpilothi everyone, long time no see, hope everything's been well! i've been really busy unfortunently but its school holidays now so i've got a bit of free time. any problems i could help with?
#asahi-dev · 2026-03-03 06:24
integralpilotgot a bit more free time now (was very busy last couple weeks), anything else that could benefit from upstreaming?

plus ›