#asahi-dev · 2026-02-11 16:23
sven"public sptm bypass" just doesn't exist and is extremely unlikely#asahi-dev · 2026-02-11 16:23
freetolgasven: considering i silenced much worse ideas if it existed it would work though right#asahi-dev · 2026-02-11 16:23
nickchan:-)#asahi-dev · 2026-02-11 16:23
sven:-)#asahi-dev · 2026-02-11 16:22
chaos_princessi too want a million usd and a pony#asahi-dev · 2026-02-11 16:22
murdakahnickchan: hvc? hvf?#asahi-dev · 2026-02-11 16:22
chaos_princess"public sptm bypass"#asahi-dev · 2026-02-11 16:22
svenfreetolga: yeah, no#asahi-dev · 2026-02-11 16:22
chaos_princesshahahaha#asahi-dev · 2026-02-11 16:22
freetolgathe only logical idea i have is to find some public sptm bypass for ios some kernel version and use that for reverse engineering on macos#asahi-dev · 2026-02-11 16:22
svenyeah#asahi-dev · 2026-02-11 16:22
chaos_princessmurdakah: in this case read-only means that the kernel does not change itself so we only need to patch it once#asahi-dev · 2026-02-11 16:21
svenmurdakah: what? i don't understand that question#asahi-dev · 2026-02-11 16:21
murdakahsven: I'm struggling to tunderstand how anything readonly from a memory perspective is useful if EL1 is read-only? Unless you want to repeatedly perform the same operation\#asahi-dev · 2026-02-11 16:21
chaos_princessseems that way#asahi-dev · 2026-02-11 16:21
svenmeh, at least dart and sart are simple enough#asahi-dev · 2026-02-11 16:21
nickchannow that you talk about it i realized ans2 sart must also be handled in the same way#asahi-dev · 2026-02-11 16:20
nickchanthere's a bigger problem than gpu uat#asahi-dev · 2026-02-11 16:20
svenit's m4/m5 where we lose the apple isa extensions in raw mode and can't use apple-entitled mode because we only get control after SPTM is already up and running#asahi-dev · 2026-02-11 16:20
svenon m1/m2/m3 raw mode vs. apple mode doesn't really make a difference fwiw, raw mode is just more convenient #asahi-dev · 2026-02-11 16:18
chaos_princess-d+s#asahi-dev · 2026-02-11 16:18
chaos_princessyed#asahi-dev · 2026-02-11 16:18
freetolgadoes asahi normally use raw mode?#asahi-dev · 2026-02-11 16:15
chaos_princessthat is in libT6000.os.DEVELOPMENT.a#asahi-dev · 2026-02-11 16:15
svenhas to be inside SPTM and used to be in PPL#asahi-dev · 2026-02-11 16:14
nickchanchaos_princess: is that not stuffed into a kext#asahi-dev · 2026-02-11 16:13
chaos_princessexcept it is not dart#asahi-dev · 2026-02-11 16:12
chaos_princessgpu dart#asahi-dev · 2026-02-11 16:12
nickchan...Userspace AT?#asahi-dev · 2026-02-11 16:12
chaos_princessi am more concerned about missing uat in new vmapple kernels#asahi-dev · 2026-02-11 16:12
nickchanmacos 12 doesn't use pv apple pac so it can run there just fine#asahi-dev · 2026-02-11 16:11
nickchangui#asahi-dev · 2026-02-11 16:11
chaos_princessgui macos?#asahi-dev · 2026-02-11 16:11
nickchanimplies ACTLR_EL1 zeroed at that time (not true anymore for the TSO bit)#asahi-dev · 2026-02-11 16:10
nickchanchaos_princess: qemu on hvf with public isa level#asahi-dev · 2026-02-11 16:10
chaos_princesshow exactly was "vmapple macos 12 can run with basically no isa extension" done?#asahi-dev · 2026-02-11 16:10
chaos_princessnickchan: so, about that. #asahi-dev · 2026-02-11 16:10
nickchanof course the only way is know for sure is to test but because of that fiq mask register that doesn't work#asahi-dev · 2026-02-11 16:09
nickchanthat isn't trapped all the way to the vmm#asahi-dev · 2026-02-11 16:08
nickchanlater versions has issues but it's due to some hvc #asahi-dev · 2026-02-11 16:07
nickchanof course rosetta doesn't work in this state#asahi-dev · 2026-02-11 16:07
nickchanit's known that vmapple macos 12 can run with basically no isa extension#asahi-dev · 2026-02-11 15:55
svenbut we have no idea if any of those approaches work or what their limitations are#asahi-dev · 2026-02-11 15:53
svendoesn't solve the "EL0 uses apple isa extensions" problem though#asahi-dev · 2026-02-11 15:53
svenand the vmapple idea is to run that kernel in non-apple-entitled mode and inject hardware kexts to trace them that way#asahi-dev · 2026-02-11 15:52
svenEL0 is a bit more tricky, but i guess we could patch EL1 exception handlers and redirect anything undef to EL2 that way as well#asahi-dev · 2026-02-11 15:52
svenEL1 code is read-only anyway so could just static patch genter/exit and other stuff to a hvc#asahi-dev · 2026-02-11 15:48
nickchanas for lineage; it's piles on top of piles of extensions on the Apple A7 design :-)