#
#asahi-re · 2026-03-22 10:08
chaos_princessunexpected
#asahi-re · 2026-03-22 10:06
noopwafelthat .. actually boots fine :) race conditions \o/ https://noopwafel.net/asahi/m3-pmgr-log-good.txt if it's interesting/useful despite this
#asahi-re · 2026-03-22 09:46
chaos_princesscan you run on m3 with this kernel command line "debug=0x14e serial=3 -enable-kprintf-spam wdt=-1 clpc=0 pmgr-debug=0xffffffff pmgr-debug-devices=0xffffffff pmgr-debug-always-on=0xffffffff pmgr-debug-energy-model=0xffffffff pmgr-debug-restore-counter=1 pmgr-trace=1 -disable_mtx_chk" and with -d flag to run_guest, then pastebin xnu console output
#asahi-re · 2026-03-22 09:36
noopwafel"ApplePMGR: void ApplePMGR::_syncDeviceStatusChange(DeviceStatusChangeData *, UInt32):11702 " "_syncDeviceStatusChange takes too long execTime 45418 us threshold 30000 us PsLevelChangeTime 45406 us turningOn 1 threadSlept 0" @ApplePMGR.cpp:11702
#asahi-re · 2026-03-22 09:36
noopwafel[cpu0] #### MacOS paniced
#asahi-re · 2026-03-22 09:36
noopwafelon the M1 that is perfect but on the M3 it is not enough..
#asahi-re · 2026-03-22 09:34
jannauI've been using clpc=0
#asahi-re · 2026-03-22 09:24
noopwafeldoes anyone know if there's any kind of disable/hack for the ApplePMGR timing asserts? from irc logs, marcan mentioned pmgr-disable-energy-panic for the energy panics
#asahi-re · 2026-03-22 04:11
zakiryaanyone who wants to look at the ADT mapping
#asahi-re · 2026-03-22 04:11
zakiryaYeah, the missing IOKit stuff in the public git is a pain. That’s why I'm targeting the vmapple machine model specifically. Since it's built for VirtIO and standard GICv3, I'm hoping it's less of a 'brain in a jar' than the bare-metal kernel. Right now I'm just trying to get a serial output through m1n1. If I can get the Mach-O to execute the first few instructions and hit a sync exception I can trace, I'll put the repo up for
#asahi-re · 2026-03-22 04:03
zakiryaI'm focusing on the XNU kernel research first. Specifically, I'm trying to use the vmapple variant as a skeleton to reach a bootable state on generic ARM64 using m1n1 as the shim. The goal is to prove we can dummy the PAC (HVC 0xC1000002) and map GICv3 interrupts to bypass the AIC dependency. Once we get a serial console from a raw Mach-O boot, then we can worry about the rest of the OS stack."
#asahi-re · 2026-03-21 22:38
chaos_princessall the fun stuff is in kexts
#asahi-re · 2026-03-21 22:38
chaos_princessxnu.git has next to nothing tbh
#asahi-re · 2026-03-21 22:38
nicolas17(I'm sure the EULA doesn't apply to what you do with apple-oss-distributions/xnu.git)
#asahi-re · 2026-03-21 22:34
nicolas17zakirya: are you trying to eventually run all of macOS or are you building the XNU kernel from source code?
#asahi-re · 2026-03-21 12:47
fl0_idzakirya at least normal-xnu boots without pmgr
#asahi-re · 2026-03-21 12:06
zakiryaAlso I am I'm researching the vmapple kernel variant's compatibility with generic ARM GICv3 distributors. Has anyone here mapped the minimum required Device Tree nodes to get a serial log on a non-Apple host, or does it hard-panic without the Apple-specific power management (PMGR)?"
#asahi-re · 2026-03-21 11:27
zakiryaSorry if I am asking too many questions but I am researching yeah I need this info
#asahi-re · 2026-03-21 11:27
chaos_princesshttps://github.com/AsahiLinux/m1n1/blob/main/proxyclient/tools/chainload.py - start here and follow the "chainload xnu" code path
#asahi-re · 2026-03-21 11:26
zakiryaI’m aware of the EULA restrictions. This is purely for personal educational research into XNU’s boot-up sequence and GIC v3 implementation. I'm more interested in the technical 'how' of the kernel's initialization than the 'legal' end result."
#asahi-re · 2026-03-21 11:23
chaos_princessalso, again, jfyi, running macos on non-apple hw is a eula violation
#asahi-re · 2026-03-21 11:23
chaos_princessyou must have a hypervisor for it, it does hypercalls, but there aren't that many
#asahi-re · 2026-03-21 11:22
zakiryaUnderstood. So if XNU is SoC-specific, my focus is now strictly on the VMAPPLE build. Does anyone know if that variant is restricted to Apple's hypervisor (signatures/entitlements) or if it can be bootstrap-loaded by a third-party bootloader like OpenCore-AA64 on a standard GIC v3 system?"
#asahi-re · 2026-03-21 11:18
chaos_princessthey literally ship separate builds for every soc
#asahi-re · 2026-03-21 11:18
chaos_princessthis is not how xnu works
#asahi-re · 2026-03-21 11:17
zakiryaThanks @chaos_princess, that's a massive tip. If vmapple uses a GIC, has anyone successfully extracted the vmapple device tree (DTB) or the specific boot-args needed to force the XNU kernel into vmapple mode on non-Apple hardware? I'm looking at the AVPBooter in the Virtualization framework to see how it initializes the GIC v3."
#asahi-re · 2026-03-21 11:15
chaos_princessalso, kind of out of scope
#asahi-re · 2026-03-21 11:15
chaos_princessvmapple kernel uses a gic
#asahi-re · 2026-03-21 11:13
zakirya I'm researching booting the XNU arm64 kernel on non-Apple hardware (Qualcomm/Snapdragon). I'm currently looking into the differences between the Apple Interrupt Controller (AIC) and standard GIC. Does anyone know of any existing documentation or shims that map AIC-style interrupts to a standard GIC for the Darwin kernel?"
#asahi-re · 2026-03-21 11:13
zakiryaAnyone here?
#asahi-re · 2026-03-14 21:26
amarioguy(not the person who was testing on m2 btw, rather was the one looking into "m1n1 as mach-o")
#asahi-re · 2026-03-14 21:06
amarioguytbh based on this i don't think the payload properties are the determining factor for whether or not something is going down the mach-o path or raw binary path
#asahi-re · 2026-03-14 21:06
amarioguysome cross-referencing...seems lpol related
#asahi-re · 2026-03-14 20:39
amarioguy(the logic to decide whether or not to boot via the fuOS path or not)
#asahi-re · 2026-03-14 20:39
amarioguythere's some tag it's checking somehow called `hsec`
#asahi-re · 2026-03-14 20:38
amarioguyre the SPTM things, decided to peek at iBoot again
#asahi-re · 2026-03-14 03:43
UnknownArtistBotHi all! check this out, An Unknown Artist - https://www.anunknownartist.com/
#asahi-re · 2026-03-12 04:29
FiniteCode[m]hmm, let's check!
#asahi-re · 2026-03-12 04:29
JamesCalligeros[m]agx maybe?
#asahi-re · 2026-03-12 04:29
FiniteCode[m]chaos_princess: any similar word for graphics/gpu?
#asahi-re · 2026-03-08 10:50
chaos_princessuse the irc logs to search for the keyword "sep"
#asahi-re · 2026-03-08 10:50
Robert[m]1Hi, I'm completely new to asahi. I'm really interested in contributing in reverse engineering the secure enclave. Does somebody have resources to start from and get up to date on current state of re?
#asahi-re · 2026-03-05 02:13
chadmedthe scheduler's gotten pretty good :P
#asahi-re · 2026-03-05 02:13
chadmedhopefully we dont have to do much beyond benching and normalising the performance numbers
#asahi-re · 2026-03-05 01:10
nickchanStandard Efficiency Super Performance
#asahi-re · 2026-03-05 01:10
nickchanxnu cluster types:
#asahi-re · 2026-03-05 01:10
nickchankernelcache.release.mac17j
#asahi-re · 2026-03-04 22:59
nicolas17CPU_DIE / GFX_DIE
#asahi-re · 2026-03-04 22:59
nicolas17jannau: https://github.com/blacktop/ipsw-diffs/blob/main/26_3_23D127__vs_26_3_1_23D8133/DYLIBS/libhwtrace.dylib.md
#asahi-re · 2026-03-04 20:31
jannauI whink something went wrong with the template devicetrees in the IPSW. I have trouble to imagine that all those `:sotra_h17[cds]*` conditionals are intended to be include

plus ›