#asahi-dev · 2026-10-18 17:15
svenm4pro_kernelcache_new.b: Mach-O 64-bit arm64e (caps: PAK00) filetype=12#asahi-dev · 2026-10-18 17:15
svenm4pro_sptm.b: Mach-O 64-bit arm64e (caps: PAK00) executable, flags:<NOUNDEFS|PIE>#asahi-dev · 2026-10-18 17:15
svenspeter@sven-macbookair:~/projects/asahi/m1n1$ file m4pro_kernelcache_new.b #asahi-dev · 2026-10-18 17:15
svenspeter@sven-macbookair:~/projects/asahi/m1n1$ file m4pro_sptm.b #asahi-dev · 2026-10-18 17:15
svenm4pro_txm.b: Mach-O 64-bit arm64e (caps: PAK00) executable, flags:<NOUNDEFS|PIE>#asahi-dev · 2026-10-18 17:15
svenspeter@sven-macbookair:~/projects/asahi/m1n1$ file m4pro_txm.b #asahi-dev · 2026-10-18 17:15
sveni thought that's what ipsw outputs, but maybe i ran an extra step and forgot#asahi-dev · 2026-10-18 17:14
sventhe extraced macho, "Mach-O 64-bit arm64e (caps: PAK00) executable, flags:<NOUNDEFS|PIE>"#asahi-dev · 2026-10-18 17:13
chaos_princessfor clarification, that is the img4s for sptm/txm, right?#asahi-dev · 2026-10-18 16:39
yukaNice, I'll try it soon#asahi-dev · 2026-10-18 16:38
yukaSART4 patches sent#asahi-dev · 2026-10-18 16:38
svenah, tl;dr for how to use the HV: grab sptm, txm and a kernelcache. they're all in the preboot volume or just use ipsw extract --sptm and --kernel and then python3 proxyclient/tools/run_guest.py -l m4_`date +%Y%m%d-%H%M%S`.log --sptm-mode=emulation --sptm m4pro_sptm.b --txm m4pro_txm.b m4pro_kernelcache.b -S debug=0x14e serial=3 wdt=-1#asahi-dev · 2026-10-18 16:21
sveni hope it won't be too much of a mess now that I can finally trace it :D#asahi-dev · 2026-10-18 16:20
yuka(prepare the patch series)#asahi-dev · 2026-10-18 16:20
yukaI'll prepare sartv4 for linux now that I confirmed it works#asahi-dev · 2026-10-18 16:20
yukaatcphy would be amazing#asahi-dev · 2026-10-18 16:19
sveni'm probably gonna work on tbt (again) and on the atcphy on the m4 pro now#asahi-dev · 2026-10-18 16:19
yukat6050 additionally has ps_aphy_st{0,1}_aus5 which I needed to add as dependency to ps_apcie_sys_st{0,1} or else ANS would crash. t8142 has the same just with only one of each.#asahi-dev · 2026-10-18 16:18
svennice!#asahi-dev · 2026-10-18 16:18
yukapreviously I used _phy_sw as apcie0 power domain for the nvme node on t8132, but I now changed that to always-on and instead use sys_st like on M3.#asahi-dev · 2026-10-18 16:13
yukaGot t6050 nvme working. I'm somewhat confident t8142 will now work too.#asahi-dev · 2026-10-18 16:09
svenbut that should be good enough to trace most hardware for now#asahi-dev · 2026-10-18 16:09
svensomething odd is going on though since it seems to freeze a minute or so after i login now #asahi-dev · 2026-10-18 16:08
svenand https://github.com/AsahiLinux/m1n1/pull/732 will hide that AMCC interrupt from the guest and prevent it from panicking :D#asahi-dev · 2026-10-18 14:28
sventhere’s a bunch of more weird hacks in my tree but I don’t think those are required #asahi-dev · 2026-10-18 14:22
svenpython3 proxyclient/tools/run_guest.py -l m4_`date +%Y%m%d-%H%M%S`.log --sptm-mode=emulation --sptm m4pro_sptm.b --txm m4pro_txm.b m4pro_kernelcache.b -S debug=0x14e serial=3 wdt=-1#asahi-dev · 2026-10-18 14:22
svenalright, so I think the two PRs should be enough to get a unstable boot to userspace if you're lucky enough. all that's missing now is to ignore those weird AMCC panics that sometimes happen#asahi-dev · 2026-10-18 11:46
jannaubtw asahi "leaks" kernel objects (probably keeps them alive by holding refenrences). I'd recommend rebooting often to avoid OOM#asahi-dev · 2026-10-18 11:44
sofusm3 pro, i had the same deferred probe and cio pmgr things#asahi-dev · 2026-10-18 11:43
jannausofus: on the m3 pro macbook pro? on m2 pro it does not work with USB4=y#asahi-dev · 2026-10-18 11:42
chadmedinterchange works on the mac studio with 6.7.5#asahi-dev · 2026-10-18 11:41
chadmedyeah its the same on 7.1.13 which is what im booted on now#asahi-dev · 2026-10-18 11:41
jannaucan you try without interchange support#asahi-dev · 2026-10-18 11:40
sofuschadmed: re usb issue. I had the exact same symptoms, iirc CONFIG_USB4=y fixed it#asahi-dev · 2026-10-18 11:37
jannauI'm not surprised if it broke. kmsro isn't quite multi-gpu#asahi-dev · 2026-10-18 11:35
chadmedgiven that was supposed to Just Work im also surprised there are no tranches with dcp as the target#asahi-dev · 2026-10-18 11:34
chadmedyeah, im asking now#asahi-dev · 2026-10-18 11:34
jannaupossibly related to multi-gpu work in kwin 6.8#asahi-dev · 2026-10-18 11:31
chadmedgiven all the ycc formats are gone im a little worried this is not a bug and is now expected behaviour#asahi-dev · 2026-10-18 11:30
chadmedmachine with the same kernel and mesa still on 6.7.x has the linear options#asahi-dev · 2026-10-18 11:30
chadmedkwin 6.8 on this machine#asahi-dev · 2026-10-18 11:30
chadmedwhich is why i said kwin is not reporting NV12 LINEAR in the tranches#asahi-dev · 2026-10-18 11:29
chadmedwayland-info in the tranches#asahi-dev · 2026-10-18 11:28
jannauwhere? the planes report NV12,... with MOD_LINEAR#asahi-dev · 2026-10-18 11:27
chadmedoh none of the YCbCr formats have a linear option now#asahi-dev · 2026-10-18 11:25
chadmedthe NV12 thing is not mesa or the kernel hmm #asahi-dev · 2026-10-18 11:15
chadmedso dmabuf-wayland is broken for mpv #asahi-dev · 2026-10-18 11:15
chadmedsomethings up with kwin too, no tranches report NV12 LINEAR for some reason#asahi-dev · 2026-10-18 11:14
chadmedproblem for tomorrow now though i think#asahi-dev · 2026-10-18 11:13
chadmedoh yeah here we go