#asahi-dev · 2025-12-29 09:42
integralpilotcurrently downloading at 1.2MB/s :(#asahi-dev · 2025-12-29 09:41
integralpilotplug/unplug events patch almost ready - just have to unshallow my clone and i have slow wifi so it might take a while sorry#asahi-dev · 2025-12-29 09:19
integralpilotthough i do also see log spam (every minute or so) about "Emergency notification: battery status is okay?" is that what you mean by critical charge reporting?#asahi-dev · 2025-12-29 09:19
integralpilotjannau: the issue in the screenshot was because apple changed the SMC event codes for plugging/unplugging charging cables to differentiate between MagSafe and USB-C, i've actually finished the code fix for that now and am writing up the PR rn#asahi-dev · 2025-12-29 09:18
integralpilotperfect, sounds great! :)#asahi-dev · 2025-12-29 09:07
integralpilotoh just noticed the M3 bootup one has some SMC power management errors (i'm actually working on fixing them rn) oops#asahi-dev · 2025-12-29 09:04
integralpilotgood luck with your talk and thanks for including me/the M3 work! :)#asahi-dev · 2025-12-29 09:04
integralpilotthat's it, not really sure which ones are better thought i'd leave it up to you.#asahi-dev · 2025-12-29 09:04
integralpilotM3 (t8122) bootup, neofetch and busybox (though accidentally didn't change the init script and if you look close it says "Welcome to M3 Pro Linux"): https://imgur.com/66mDBhi#asahi-dev · 2025-12-29 09:03
integralpilotM3 Pro (t6030) bootup, neofetch and busybox: https://imgur.com/Hwug7HN#asahi-dev · 2025-12-29 09:02
integralpilotM3 SMC - looking at power & temp on linux through this nice TUI program i found (CPUs/SMP and processes also shown): https://imgur.com/LdKLqiA#asahi-dev · 2025-12-29 09:00
integralpilotcontrolling keyboard backlight brightness through busybox 'echo' (also MagSafe LED shown): https://imgur.com/NQW0eqw#asahi-dev · 2025-12-29 08:59
integralpilotplaying doom under linux on M3 base: https://imgur.com/3YmmNmt#asahi-dev · 2025-12-29 08:56
integralpilotsven: got the photos now. took a couple, pick whichever you like#asahi-dev · 2025-12-29 06:48
integralpilotwhich devices do you have?#asahi-dev · 2025-12-29 05:48
integralpilotoh interesting, thanks!#asahi-dev · 2025-12-29 05:46
integralpilotI think low level operations like turning on/off coprocessors or loading firmware has changed with the asc v6 now, but processors that are already running and we don’t need to change like SMC still work#asahi-dev · 2025-12-29 05:44
integralpilotim mainly working on v6 coprocessor communication rn#asahi-dev · 2025-12-29 05:43
integralpilotI’ll give it a go#asahi-dev · 2025-12-29 05:41
integralpilotthe smc works fully which I was really surprised about, it sends us messages, we can query it etc! GPIO, SPI and I2C are all also working, can also do some stuff like turn on keyboard backlight. power management and cpufreq work too. haven’t tried USB but I don’t think it would work either#asahi-dev · 2025-12-29 05:36
integralpilotM3 update: now have alpine Linux working!#asahi-dev · 2025-12-29 05:08
integralpilotoh and forgot to say but run_guest_kernel.sh is rly simple, the args are just [path to the root of your Linux kernel clone] “[boot args or empty]” [path to initramfs]#asahi-dev · 2025-12-29 05:06
integralpilotand if it works we can add more devices to make Max support more things like I’ve got for base and Pro currently#asahi-dev · 2025-12-29 05:04
integralpilothopefully it can work on M3 Max as well and you might he able to do a live demo, lmk if there’s any issues as I don’t have the actual device to test I might have made a mistake somewhere#asahi-dev · 2025-12-29 05:03
integralpilotOh and there’s no AIC driver modification needed, I got confused with the layout of some things#asahi-dev · 2025-12-29 05:01
integralpilot(with run_guest_kernel.sh, m1n1 runs in both EL2 and then loads itself into EL1 so will error)#asahi-dev · 2025-12-29 05:01
integralpilotbtw, if you enabled the new chickens I found for XNU boot, comment these out and recompile as when there’s a hypervisor in EL2, you can’t touch them from EL1. From my testing they look sha512/crypto related so we don’t need them for this early bring up anyway#asahi-dev · 2025-12-29 04:59
integralpilotcan you use run_guest_kernel.sh instead of linux.py, it’s easier to see what’s going on when m1n1 is hypervising?#asahi-dev · 2025-12-29 04:58
integralpilothttps://gist.github.com/IntegralPilot/74d1c364e70382ddcad3ccf6a333e13e#asahi-dev · 2025-12-29 04:57
integralpilotfl0_id: I’ve written minimal t6031 devicetrees (just framebuffer & AIC) that should be enough based on the data you gave to boot a single core all the way (and show 1 tux, logs etc.)#asahi-dev · 2025-12-28 23:35
integralpilotthanks f10_id, i've got it. i'm out right now but when i get home i'll take a look and write a basic devicetree#asahi-dev · 2025-12-28 22:20
integralpiloti have the pro devicetree but that's probably not going to be helpful as max is different, but i can send it if you would like an example just lmk#asahi-dev · 2025-12-28 22:17
integralpilotif you send me the output of `ioreg -l` on macOS and what m1n1 prints to host stdout during a chainload (has some important things to check) i might be able to write you a minimal devicetree#asahi-dev · 2025-12-28 22:15
integralpilotthe Pro (and I'm guessing Max) AIC3 has a slightly different layout to the base so the linux driver needed a tiny change#asahi-dev · 2025-12-28 22:15
integralpilotf10_ld: it's probably the AIC3 issue, i will write a PR for my fix now#asahi-dev · 2025-12-28 22:14
integralpilotsven: thank you, that would be so cool! i'll get some pictures of busybox and doing some cool stuff (i.e. querying battery %, turning on keyboard backlight). my github would be good, it's just integralpilot as well, https://github.com/IntegralPilot#asahi-dev · 2025-12-28 11:14
integralpilotwith the AIC3 change and my device tree, M3 Pro now starts up to busybox yay!#asahi-dev · 2025-12-28 11:14
integralpilotthanks for the confirmation though, just wanted to make sure I wasn’t mixed up with what I thought the structure was it had already been tested#asahi-dev · 2025-12-28 11:05
integralpilothaven’t tried charging while using it as a target yet, just wanted it to charge as quickly as possible so left it alone and switched devices (needed a fresh challenge anyway)#asahi-dev · 2025-12-28 10:44
integralpilot(My M3 I was doing NVMe/RTKit work on got to 0% so doing M3 Pro early bring up now while it charges)#asahi-dev · 2025-12-28 10:38
integralpilothas the Linux code for AIC3 been tested on M3 Pro? I think the layout is a bit different and I needed to make a change and it seems to be working properly now but want to make sure I’m not confused#asahi-dev · 2025-12-28 09:10
integralpilotoh i guess it's better to not look then, thank you for letting me know before it finished downloading and i diffed it#asahi-dev · 2025-12-28 09:06
integralpilotwell time to diff 13.5 RTBuddy.kext vs 14.x and hope not too much is changed in v6 i guess, i'll take a look#asahi-dev · 2025-12-28 09:05
integralpilot:(#asahi-dev · 2025-12-28 08:49
integralpiloti guess the M3 RTKit running on the ANS uses a different message format for shutdown, will have to RE it#asahi-dev · 2025-12-28 08:49
integralpilotand now we get the fun of an RTKit crash when we try to shut it down for the soft-reset :/#asahi-dev · 2025-12-28 08:46
integralpilotoh makes sense, thank you!#asahi-dev · 2025-12-28 08:16
integralpilothow do we determine if Apple made a new major version, or if it's just a hardware revision that patched this?#asahi-dev · 2025-12-28 08:13
integralpilothave to make a compadible tag to avoid applying that chicken on M3+, should i call it ANS v2 rev 1 or ANS v3?#asahi-dev · 2025-12-28 08:11
integralpilotlooks like apple patched the chicken at https://github.com/AsahiLinux/linux/blob/asahi/drivers/nvme/host/apple.c line 1134 and the register we use there isn't in the memory map anymore