#asahi-dev · 2026-04-17 19:22
yukaNow I'm somewhat consistently running into the following: https://termbin.com/cdky#asahi-dev · 2026-04-17 19:21
yukaI could not identify any other relevant wfis in inline assembly#asahi-dev · 2026-04-17 19:21
yukaSo I replaced the wfi in arch/arm64/include/asm/barrier.h with a nop#asahi-dev · 2026-04-17 11:17
JamesCalligeros[m]unless anyone has seen any of these values ever use more than 24 bits, i think it's either 8.16 or maaaaybe 16.16 but i've no >255 hz monitors to test if it is#asahi-dev · 2026-04-17 11:15
JamesCalligeros[m]TimingElements says that the refresh rates are 64-bit integers but it doesnt look like they actually are#asahi-dev · 2026-04-17 11:15
JamesCalligeros[m]ahahaha theres a reason we dont support anything above 240 hz#asahi-dev · 2026-04-17 09:45
sventrue, find_regs should’ve found everything I think #asahi-dev · 2026-04-17 08:36
chaos_princess libT6000.os.DEVELOPMENT.a under KernelSupport (but for the m4 soc)#asahi-dev · 2026-04-17 08:35
yukasven: doesn't the find_regs already check if I can write to the regs, and the result was that all are readonly? or is there possibly still some registers missed by find_regs#asahi-dev · 2026-04-17 08:35
chaos_princessshould be less stuff to look through#asahi-dev · 2026-04-17 08:34
chaos_princessget the kdk, there is a libT6040.a or something like that in there#asahi-dev · 2026-04-17 08:34
svenit may not even try to reconfigure that behavior #asahi-dev · 2026-04-17 08:34
chaos_princessactually, you probably don't even need to do the full xnu#asahi-dev · 2026-04-17 08:34
sventhough I guess XNU always guards wfi with reg save/restore anyway #asahi-dev · 2026-04-17 08:34
svenand then check if we can write to any of them from raw mode as well #asahi-dev · 2026-04-17 08:33
nickchanthe hypervisor section would save restore all the funny registers#asahi-dev · 2026-04-17 08:33
sventhat’s what I’d do as well #asahi-dev · 2026-04-17 08:33
chaos_princessimo, throw m4 xnu into a disassembler and find a list of all msrs that it writes to that we don't know the purpose#asahi-dev · 2026-04-17 08:32
nickchanokay I missed that#asahi-dev · 2026-04-17 08:32
yukayep#asahi-dev · 2026-04-17 08:32
chaos_princessthe one that makes non-boot cores lose state on wfi#asahi-dev · 2026-04-17 08:31
chaos_princesswait, but we are talking about a power management(ish) register#asahi-dev · 2026-04-17 08:31
nickchanresearch guest should have most secret registers that aren't power-management related#asahi-dev · 2026-04-17 08:30
nickchansptm probably would have its own (different) bootargs struct though#asahi-dev · 2026-04-17 08:30
nickchanyuka: under sptm is a relatively cursed environment I think for starters you should restore a pcc research VM and then set m1n1 as the custom sptm (all documented)#asahi-dev · 2026-04-17 08:28
chadmedi might clean up the feature support pages in the docs tonight, theyre in need of some love#asahi-dev · 2026-04-17 08:27
chadmedwhat a scary and ominous statement ;) #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 08:25
nickchanchaos_princess: best bet is to make Linux work and then make windows under kvm :)#asahi-dev · 2026-04-17 08:23
nickchannamely any hvcs that is part of the vmapple pac calls#asahi-dev · 2026-04-17 08:22
nickchannot only is the env cursed i have reasons to suspect that the host hv would eat some hvcs so the guest hypervisor can't receive them#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 · 2026-04-17 08:20
nickchanand nested virtualization itself is not exposed to guest so no recursive nesting either#asahi-dev · 2026-04-17 08:20
nickchanthat is the nested virtualization is nVHE (this actually breaks it with windows 11 since it requires VHE for Hyper-V)#asahi-dev · 2026-04-17 08:20
nickchanthe biggest one being E2H=0#asahi-dev · 2026-04-17 08:19
nickchanbut it's quite a bit different environment, I am not actually sure if NV is supported for internal ISA level in hypervisor.framework#asahi-dev · 2026-04-17 08:19
nickchannested virtualization so like vGL2#asahi-dev · 2026-04-17 08:19
yukawhat is NV?#asahi-dev · 2026-04-17 08:18
nickchanwhich even has NV on M4!#asahi-dev · 2026-04-17 08:18
yukaI guess I could start m1n1 as macho boot object and experiment in there?#asahi-dev · 2026-04-17 08:18
nickchananother place where those secret registers could be used is inside a internal/research ISA level VM#asahi-dev · 2026-04-17 08:17
nickchanso yeah a macho boot object or a macos kext can use the registers but there's GL2 only registers#asahi-dev · 2026-04-17 06:22
yukadear fruit overlord, please let me use the computer I bought from you#asahi-dev · 2026-04-17 06:22
JamesCalligeros[m]sees locked register is it for me 🥺👉👈#asahi-dev · 2026-04-17 06:21
yukaOr for 👉️👈️ at 🍏 for letting us use them#asahi-dev · 2026-04-17 06:20
yukaOh for example for detecting easily whether saving and restoring registers is necessary#asahi-dev · 2026-04-17 01:08
nickchani don't see how finding out meaning of impl regs that can't be nomially used could be useful#asahi-dev · 2026-04-16 19:09
sven*commits