#
✕ nick:chaos_princess
#asahi-dev · 2025-11-16 16:12
chaos_princessguest os is at el1
#asahi-dev · 2025-11-16 16:10
chaos_princessthere is no el3 support on m-series
#asahi-dev · 2025-11-16 16:10
chaos_princessfor m1n1? no, both are el2, stage 2 replaces stage 1
#asahi-dev · 2025-11-16 15:40
chaos_princessthen port gunyah to it? arguably would be a better choice than m1n1
#asahi-dev · 2025-11-16 15:39
chaos_princessuse linux kvm as hv
#asahi-dev · 2025-11-16 15:39
chaos_princesswhy do you care about hv being m1n1 then
#asahi-dev · 2025-11-16 15:39
chaos_princesssame
#asahi-dev · 2025-11-16 15:38
chaos_princessbut on m2+ we have nested virt, so why bother
#asahi-dev · 2025-11-16 15:38
chaos_princesstheoretically, and again, very theoretically, you can have all hw on vm0, while vm1/vm2/etc is doing something muvm like and using virtualized hw
#asahi-dev · 2025-11-16 15:35
chaos_princesswhile all other hardware is "draw the rest of the owl"
#asahi-dev · 2025-11-16 15:34
chaos_princessthe cpu is kinda the easy part
#asahi-dev · 2025-11-16 15:33
chaos_princesssince it is not asahi-specific
#asahi-dev · 2025-11-16 15:33
chaos_princessfirst was more appropriate, and you may have more luck asking in generic fedora/generic kde channels
#asahi-dev · 2025-11-16 15:31
chaos_princessdeadf00d: effectively impossible, m1n1 passes through all hardware except what it needs for itself. if you want a real VMM, use a real VMM
#asahi-dev · 2025-11-12 19:08
chaos_princessmaybe, but the firmware tarball is still there, and the scripts to re-extract if necessary too
#asahi-dev · 2025-11-12 19:06
chaos_princesseffectively yes, i think
#asahi-dev · 2025-11-02 18:55
chaos_princessseems to be working alright
#asahi-dev · 2025-11-02 12:58
chaos_princessack, will try later today
#asahi-dev · 2025-11-02 12:55
chaos_princessany specific things to try?
#asahi-dev · 2025-10-28 09:27
chaos_princesswhich is probably a good thing to poke at if someone wants to help out, the RE is all done for you already :P
#asahi-dev · 2025-10-28 09:26
chaos_princessmodulo spmi i guess
#asahi-dev · 2025-10-28 09:24
chaos_princesson more serious note, do we know if tbt/usb-c/that entire mess is mostly same on newer SoCs?
#asahi-dev · 2025-10-28 09:22
chaos_princesssve n: m5 wen :P
#asahi-dev · 2025-10-27 15:48
chaos_princessi mean in general, there is no document per se, just read the code that works with rtkit and do by example
#asahi-dev · 2025-10-27 15:34
chaos_princessrtkit.c
#asahi-dev · 2025-10-25 21:14
chaos_princessthat mostly does power management, but is also a kitchen sink of stuff that apple couldn't find a better place for
#asahi-dev · 2025-10-25 21:13
chaos_princessin this case pmgr refers to a specific device in the ADT
#asahi-dev · 2025-10-25 21:09
chaos_princesswhat you want is to set up m1n1, trace macos writes to pmgr mmios, replicate them on linux and see which ones help.
#asahi-dev · 2025-10-25 21:09
chaos_princessno, that thing would just tell you what is happening during entry to sleep, not what happens during sleep
#asahi-dev · 2025-10-25 21:04
chaos_princessyou want to remove "fixed costs" (disable the screen, etc), and then measure system power. Then figure out how to reduce it.
#asahi-dev · 2025-10-25 21:04
chaos_princess^
#asahi-dev · 2025-10-25 21:01
chaos_princessanyway, if you want to measure power, look at /sys/class/power_supply/macsmc-battery/power_now
#asahi-dev · 2025-10-25 20:58
chaos_princessi do not think it matters as we are doing s2idle, not "s3 sleep" (not quite correct, but whatever). So, the timer does not need to be capable of waking the machine, only waking the cpu
#asahi-dev · 2025-10-24 21:24
chaos_princessor, jump into the deep end of the pool and fix nvme too, the choice is yours :P
#asahi-dev · 2025-10-24 21:24
chaos_princessget a 14 pkg from here https://mrmacintosh.com/macos-sonoma-full-installer-database-download-directly-from-apple/ use it to create a usb-bootable installer, create a second apfs parition, and install it there
#asahi-dev · 2025-10-24 21:23
chaos_princessyes, something changed in iirc 15.x that breaks our nvme driver... somehow
#asahi-dev · 2025-10-24 21:22
chaos_princessright now there is a spmi driver, but it is extremely barebones, enough for m1/2, but will be insufficient for m3+, as they moved the usb pd controller to spmi from i2c
#asahi-dev · 2025-10-24 21:21
chaos_princesssomeone else already did a lot of RE work and made some pretty good docs
#asahi-dev · 2025-10-24 21:21
chaos_princessso, my idea is spmi
#asahi-dev · 2025-10-24 21:16
chaos_princessanyway, m1n1 is a good start
#asahi-dev · 2025-10-24 21:16
chaos_princesswell, we all started somewhere.
#asahi-dev · 2025-10-24 21:14
chaos_princessdo you have experience in specifically RE-d drivers or ones where you had the hardware reference?
#asahi-dev · 2025-10-24 21:13
chaos_princessyes, start out with m1n1, and probably pick a 14.x macos to install it
#asahi-dev · 2025-10-24 15:49
chaos_princessconvince your friend to allow you to borrow the m3 for a very extended periods of time :P
#asahi-dev · 2025-10-24 15:46
chaos_princesswould be pretty hard to write drivers for hw that you do not have, no?
#asahi-dev · 2025-10-18 22:28
chaos_princess*kb backlight
#asahi-dev · 2025-10-18 22:28
chaos_princesskb is not a smc key and is pretty easy
#asahi-dev · 2025-10-18 21:10
chaos_princessat least i do not see anything backtrace-related in core
#asahi-dev · 2025-10-18 21:09
chaos_princessprobably not possible as it is no_std
#asahi-dev · 2025-10-13 17:50
chaos_princessand by "better option" i mean that someone will complain that the mixed commit is unreviewable

plus ›