#asahi-dev · 2025-06-21 14:57
fl0_idthis is a dev kernel#asahi-dev · 2025-06-21 14:56
jannaudevelopment kernel with symbols might be helpful as long it's not panicking#asahi-dev · 2025-06-21 14:56
fl0_idbased on the output, is this still just m1n1 output or already xnu? It doesn't look like xnu output that I am used to from just turning on -v mode#asahi-dev · 2025-06-21 14:55
fl0_idmmh k. as long as it's not necessary to solder a uart or sth#asahi-dev · 2025-06-21 14:55
svenat this point it comes down to figuring out what xnu is doing and why it gets stuck#asahi-dev · 2025-06-21 14:55
svenflags aren't going to help here probably#asahi-dev · 2025-06-21 14:54
fl0_idsven yeah. is there a way to turn more debugging/verbosity on? I just used the flags/command from the docs, only adapted for t6031 kernel#asahi-dev · 2025-06-21 14:49
svenprobably need to figure out where xnu is stuck now and see what's missing/broken#asahi-dev · 2025-06-21 14:45
fl0_idjust would have expected some kind of error or sth#asahi-dev · 2025-06-21 14:40
fl0_idmmh yeah that would make sense. if that is necessary for bootup :) anyway, if you have any idea on boot options to try, let me know. though I assume it needs support in m1n1#asahi-dev · 2025-06-21 14:24
chaos_princessfl0_id: usb power gating is probably not working properly#asahi-dev · 2025-06-21 14:21
chadmedyeah display on my studio never powers off. seems to be okay on both my laptops#asahi-dev · 2025-06-21 14:21
fl0_idsecond serial device didn't show anythin#asahi-dev · 2025-06-21 14:21
fl0_iddoes anybody have an idea why this seemingly stops at bootup? https://www.pastery.net/exwxde/#asahi-dev · 2025-06-21 14:15
jannaubreaks how? display does never power off? what happens when you press esc on the lock screen?#asahi-dev · 2025-06-21 14:14
jannauworks for me on studio and I think on j493 (both fedora)#asahi-dev · 2025-06-21 13:46
chadmedplasma 6.4 seems to break dpms#asahi-dev · 2025-06-20 21:05
amarioguy2might have a timer interrupt and MADT soon#asahi-dev · 2025-06-20 21:04
fl0_id*help#asahi-dev · 2025-06-20 21:04
fl0_idglad to hr#asahi-dev · 2025-06-20 21:03
amarioguy2fl0_id: i see, thanks! looks like the spec of the vGIC remains unchanged on M3 - which makes implementation at least consistent across M1-M3 platforms#asahi-dev · 2025-06-20 20:50
fl0_idoh and screen of the m3 turned off, (as it said fb turned off) and I waited quite a while, but seemed to be stuck#asahi-dev · 2025-06-20 20:49
fl0_idI also tried with less boot arg, but same results, and from the output maybe it didn't even get to that. I also had a picocom open at the same time and that didn't show anything#asahi-dev · 2025-06-20 20:48
fl0_idnew paste link for m3 hv bootup attempt https://www.pastery.net/exwxde/#asahi-dev · 2025-06-20 20:42
fl0_idurgh why did they remove my paste#asahi-dev · 2025-06-20 19:57
fl0_idfor extracting kernel / kernel cache I adjusted the path to kernel.development.t6031 b/c otherwise it said it didn't find it (gave the full path) but I assume that should be fine#asahi-dev · 2025-06-20 19:54
fl0_idIt seems to start but doesn't pass to xnu I think. log output: https://pastebin.com/qDYbR1pz#asahi-dev · 2025-06-20 19:54
fl0_idcouldn't get macos to boot on HV yet to check out the panic reset counter thing#asahi-dev · 2025-06-20 19:52
fl0_idamarioguy2 the other mrs works In [1]: mrs(ICH_VTR_EL2) Out[1]: 2421686279#asahi-dev · 2025-06-20 13:24
nickchanconfirmed t2 smc can do rtc... but only 32 bit#asahi-dev · 2025-06-19 08:26
maznickchan: I don't think Will has any issue with that (we speak f2f often enough).#asahi-dev · 2025-06-18 18:35
jannauthere are hole in what arm64 defconfig enables. simpledrm is surprisingly missing#asahi-dev · 2025-06-18 10:31
chadmedoh right#asahi-dev · 2025-06-18 10:31
chaos_princessfat pointer, and then you can copy it yourself#asahi-dev · 2025-06-18 10:31
chadmeddoes slice::from_raw_parts actually copy though? i thought it just gives you a fat pointer to the slice at the address of its first argument#asahi-dev · 2025-06-18 10:21
chadmedthe unsized member was throwing me but it doesnt matter if we just have a pointer to the top byte. we know the total size of the thing #asahi-dev · 2025-06-18 10:20
chadmedhuh i actually didnt consider treating the whole struct as a slice of bytes#asahi-dev · 2025-06-18 10:12
chaos_princessslice::from_raw_parts(thing as *const Whatever as *const u8, mem::size_of::<Whatever>()) ?#asahi-dev · 2025-06-18 10:08
chadmedtrying to reimplement adt_getprop_copy and i just cannot figure out a way to do it that isnt total dogshit#asahi-dev · 2025-06-18 10:08
chadmedi really like rust but man i wish i could just memcpy some bytes when i need to#asahi-dev · 2025-06-17 17:29
nickchanmaz: well... Will have issues with that approach... so should express these thoughts on the ML!#asahi-dev · 2025-06-17 16:56
mazI don't have a lot of bandwidth left for it though ($work is keeping me busy enough).#asahi-dev · 2025-06-17 16:55
maznickchan: there is no need for a particular maintainer entry, but I'm happy to look after the m1/m2 parts of it.#asahi-dev · 2025-06-17 16:55
nickchan(not to mention m2 has different counter sizes)#asahi-dev · 2025-06-17 16:54
nickchanunfortunately I cannot really maintaine the m1/el2 part despite having a t6000 since that machine has become a critical single point of failure not to be messed with in my entire workflow as my other machines broke down#asahi-dev · 2025-06-17 16:52
nickchanmaybe something about the maintainership can be worked out and then respin that series with a maintainer addition patch#asahi-dev · 2025-06-17 16:49
nickchanwhen trying to come up with some capacity-dmips-mhz for the cores#asahi-dev · 2025-06-17 16:48
nickchanerm maybe not for testing energy aware scheduling but definitely the cpufreq part#asahi-dev · 2025-06-17 16:47
nickchanit's quite useful for cpufreq testing and EAS though#asahi-dev · 2025-06-17 16:46
nickchanthis driver is from maz and didn't have any maintainer entry