#asahi-dev · 2026-04-17 08:26
chaos_princessi guess that's the plan, but i still have "unfinished business" with m1 and adjacent#asahi-dev · 2026-04-17 08:21
chaos_princessi've been trying to make hyper-v work on my m3max :(#asahi-dev · 2026-04-17 08:21
chaos_princessnickchan: aaaaa. thank you for pointing this out, but still aaaaa.#asahi-dev · 2026-04-16 14:37
chaos_princessbtw, can we interact with aic while in non-hv m1n1? there seems to be no code to operate it, beyond a few bits in smp setup#asahi-dev · 2026-04-16 14:35
chaos_princessthey did get pwned once in x86 days where they forgot to lock efi flash on s3 resume, so i do not expect them to make the same mistake twice#asahi-dev · 2026-04-16 14:22
chaos_princesswe don't know. the register sequence i use is the same as xnu does.#asahi-dev · 2026-04-16 14:18
chaos_princesswon't be surprised if we don't need to do anything and iboot does it#asahi-dev · 2026-04-16 14:17
chaos_princessmore likely during early boot and then locked down#asahi-dev · 2026-04-16 14:17
chaos_princessthough, i do not expect it to be programmed around sleep#asahi-dev · 2026-04-16 14:17
chaos_princessjust beware that there is a lot of them, since as i've said, pmgr driver does a lot of random stuff#asahi-dev · 2026-04-16 14:16
chaos_princessprobably trace pmgr mmios around sleep i guess#asahi-dev · 2026-04-16 14:16
chaos_princessthe "kitchen sink" driver is pmgr#asahi-dev · 2026-04-16 14:16
chaos_princessprobably not#asahi-dev · 2026-04-16 14:07
chaos_princesshttps://github.com/apple-oss-distributions/xnu/blob/rel/xnu-8792/osfmk/arm64/machine_routines_asm.s#asahi-dev · 2026-04-16 14:05
chaos_princessit is mentioned once or twice in the opensource xnu code. #asahi-dev · 2026-04-16 13:59
chaos_princessthat is x86 behavior, on arm, there is something called a "reconfig engine" that should dump you directly to rvbar#asahi-dev · 2026-04-16 13:58
chaos_princessit is not signalling iboot#asahi-dev · 2026-04-16 13:55
chaos_princessofc they are, the memory layout is completely different#asahi-dev · 2026-04-16 13:54
chaos_princesswell, yes#asahi-dev · 2026-04-16 13:53
chaos_princessso, that is not a sleep, you just reset the machine#asahi-dev · 2026-04-15 19:29
chaos_princessand the fact that these days a gpu is basically a second computer that is flying in close formation, and has a completely proprietary instruction set that changes with every generation#asahi-dev · 2026-04-15 19:27
chaos_princessefi and architecture are irrelevant, the hard bits are the lack of documentation and the fact that computers do a lot more these days#asahi-dev · 2026-04-15 14:00
chaos_princesswhatever#asahi-dev · 2026-04-15 13:59
chaos_princessit's own field where?#asahi-dev · 2026-04-15 13:58
chaos_princesswell, more like it needs to store your driver struct somewhere and it picked drvdata#asahi-dev · 2026-04-15 13:58
chaos_princessno, it needs the place to store it's own stuff#asahi-dev · 2026-04-15 13:57
chaos_princessthose bus drivers do not touch drvdata (i think?)#asahi-dev · 2026-04-15 13:57
chaos_princessthat place ain't yours :P#asahi-dev · 2026-04-15 13:56
chaos_princesscause bus driver also sets it.#asahi-dev · 2026-04-15 13:53
chaos_princessthe same thing as aop does, it returns the thing from the probe() fn, and the bus driver writes it to drvdata#asahi-dev · 2026-04-15 13:48
chaos_princessapparently you can't use set_drvdata from drivers, it is only for the bus code#asahi-dev · 2026-04-15 07:50
chaos_princessdoes xnu write to those anywhere?#asahi-dev · 2026-04-14 21:09
chaos_princesshttps://github.com/AsahiLinux/m1n1/blob/main/proxyclient/m1n1/find_regs.py#asahi-dev · 2026-04-14 21:05
chaos_princessok#asahi-dev · 2026-04-14 21:04
chaos_princess26.whatever?#asahi-dev · 2026-04-14 21:04
chaos_princessoh, btw, quick check, what macos are you on?#asahi-dev · 2026-04-14 21:02
chaos_princessi think that is not the only place where iboot configures the boot core and then expects xnu to perform the same config on non-boot ones#asahi-dev · 2026-04-14 21:00
chaos_princesswhy is it different on boot core then?#asahi-dev · 2026-04-14 20:59
chaos_princessi really think that a msr scan is more promising#asahi-dev · 2026-04-14 20:47
chaos_princessand we do vhe with kernel running at el2#asahi-dev · 2026-04-14 20:47
chaos_princessthat is for nvhe#asahi-dev · 2026-04-14 20:39
chaos_princessdid msr scan not find anything?#asahi-dev · 2026-04-14 19:17
chaos_princessand yes, you were already told to cut it out with farming out stuff to llms#asahi-dev · 2026-04-14 19:17
chaos_princesswilfredor: the reason i asked what HW you have is because then you can collect this data yourself and not order others around#asahi-dev · 2026-04-14 17:48
chaos_princessdid you just write to some register in pmgr that survives a reboot and concluded that this is a successful sleep entry/exit?#asahi-dev · 2026-04-14 17:47
chaos_princess"writes to gated MMIO stall the bus on Apple Silicon" - no, they SError#asahi-dev · 2026-04-14 14:48
chaos_princessi personally build with LLVM=1, maybe that is relevant?#asahi-dev · 2026-04-14 14:45
chaos_princessdoes rust depend on it or sth?#asahi-dev · 2026-04-14 14:43
chaos_princessi just no longer remember what exactly#asahi-dev · 2026-04-14 14:43
chaos_princessif you turn some of those off, it should build