#asahi-dev · 2025-07-12 08:57
chaos_princessAnd also asahi is almost impossible on m4 with macos 15.x due to apple bug that is fixed in 26.x#asahi-dev · 2025-07-12 08:57
chaos_princessdevices - well, you need the devices in question, so go to apple.com and you can easily get a quote that way. Time - we kinda got two "hard" generations in a row with m3 and m4. m3 has a new gpu ISA, which is the "hard" part for it, and m4 will need a lot of m1n1 work (m1 also did, ofc, but most of it was "hidden", as when that was happening the project was a lot more low profile).#asahi-dev · 2025-07-12 08:20
alethkit(napkin math is fine for this)#asahi-dev · 2025-07-12 08:20
alethkitOut of curiosity, how much funding would be needed in terms of developer time/devices to port the m4 series of devices?#asahi-dev · 2025-07-08 21:03
chaos_princesskettenis: mind tagging u-boot?#asahi-dev · 2025-07-08 19:50
jannau(fedora) matrix chat seems to be weirdly broken as in more than usual. there seems to be some kind of (net-)split#asahi-dev · 2025-07-07 20:57
chaos_princessnice#asahi-dev · 2025-07-07 20:55
kettenissince I managed to upstream a few more bits this round, I rebased the asahi branch on top of that#asahi-dev · 2025-07-07 20:55
kettenisu-boot 2025.07 was released#asahi-dev · 2025-07-07 11:05
leiocan't attach a second computer before wednesday#asahi-dev · 2025-07-07 11:04
leiook, I can try at a later date if useful, let me know#asahi-dev · 2025-07-07 11:01
fl0_idleio no worries, it is also not certain that this is the issue, I was just speculating#asahi-dev · 2025-07-07 11:00
leiook, I'm not fully sure it was the HV then anyhow, so don't take my word on it then#asahi-dev · 2025-07-07 10:57
fl0_idso yes, I assume that's what you mean by breakpointing, that you used the hv. (though I guess you could also do that from linux without hv)#asahi-dev · 2025-07-07 10:56
fl0_idleio it's only active if you start it. you can boot a linux kernel with the hv, but also without.#asahi-dev · 2025-07-07 10:53
leios/latter two/latter/#asahi-dev · 2025-07-07 10:53
leioif the latter two, then I might have somewhat accidentally did that on my 192G M2 ultra early on around Sept 2023; as it wasn't really intentional, I can't be 100% sure I remember right though#asahi-dev · 2025-07-07 10:52
leiofl0_id: HV is only active when running macos under it or also when able to essentially breakpoint the linux kernel?#asahi-dev · 2025-07-07 10:39
fl0_idmaz thx#asahi-dev · 2025-07-07 10:36
mazfl0_id: the IPA reported by the HW should tell you where this lands, assuming you have a 1:1 mapping at S2. You could then compare that with the HW memory map and work out what this is about.#asahi-dev · 2025-07-07 10:28
fl0_idregarding the HV on m3 - could it be related to memory size, that previous models just didn't support high amounts of ram or at least it was not tested well? (like f.e. did anybody test the HV with 96 GB ram, or 192 on a m2 ultra?) Otherwise it could also be related to differences in EL1 vs EL2, like some instructions seem to differ#asahi-dev · 2025-07-06 12:31
fl0_idso maybe somehow the memory management is different or somehow not mapped right?#asahi-dev · 2025-07-06 12:10
fl0_idand then bt was a lot of this https://www.pastery.net/ungxxm/ (- 0xfffffe00127d7890 (com.apple.kernel:fleh_dispatch64_common+0xa0))#asahi-dev · 2025-07-06 12:09
fl0_idhey, so I looked again into booting a macos kernel under hv on m3, and this time at least I got more debug output, with an exception at the end https://www.pastery.net/zzvzep/#asahi-dev · 2025-07-03 07:09
chadmeddaisy: it's really not a learn-on-the-job type of task#asahi-dev · 2025-07-03 04:21
sven*sigh*#asahi-dev · 2025-07-03 00:28
daisyhi!! idk if this is the right place to ask, but i have a Macbook w/ M4 Pro and i would really love to help asahi linux run on it at some point. is there any way i can help with the dev side of things? i personally don't have experience working at the OS kernel level before but i'm willing to try! i saw sven's mastodon post about the challenge in getting linux on M4 but didn't really understand it. is there absolutely #asahi-dev · 2025-07-02 07:25
chadmedonce wine outside of muvm works we should consider documenting this#asahi-dev · 2025-07-02 07:24
chadmedall of my classic 32 bit games work nicely on macos with gcenx's wow64 builds... but theyre all totally busted on linux even on my x86 machine #asahi-dev · 2025-07-02 07:24
chadmedi didnt realise how broken wow64 still is in wine on linux specifically #asahi-dev · 2025-07-01 14:05
floklinothing in my logs from the last 2 months#asahi-dev · 2025-07-01 13:50
kettenis(I see 0x7020000000000018 in a crashlog mailbox log that another OpenBSD developer sent me)#asahi-dev · 2025-07-01 13:49
kettenisHas anyone ever seen a notification for SMC event 0x7020?#asahi-dev · 2025-07-01 06:35
svennickchan: if you get a r-b I can merge these things quicker, otherwise I’ll take a look this or next weekend #asahi-dev · 2025-07-01 05:12
Auscyberunsure where to start#asahi-dev · 2025-07-01 05:12
AuscyberI have an m4 macbook pro and wanna contribute to asahi#asahi-dev · 2025-07-01 05:12
AuscyberHi#asahi-dev · 2025-06-30 15:49
svenyeah, will probably pick it up this weekend. ideally you could get a reviewed-by for it as well #asahi-dev · 2025-06-30 01:44
nickchansven: have you seen the missing framebuffer fix for j132 all it does is adding an include, should go into -rc https://lore.kernel.org/asahi/20250620-j132-fb-v2-1-65f100182085@gmail.com#asahi-dev · 2025-06-29 16:48
amarioguy2(in case it helps - i'm envisioning the flow here as: timer FIQ fires, traps to EL2, masks the FIQ for a bit to handle it, then HV kicks it to the guest as a virtual IRQ where it proceeds from there)#asahi-dev · 2025-06-29 16:44
amarioguy2my first guess would be the start exception lower proxy call?#asahi-dev · 2025-06-29 16:43
amarioguy2quick question - if we want to redirect the timer IRQ to the guest (as an vIRQ, not enabling the FIQ in EL1 or injecting a virtual FIQ) where would it be best to inject a virtual IRQ in the HV exception handler?#asahi-dev · 2025-06-29 16:26
martinrok I think that looks pretty good, I went back and fixed the strlen check too#asahi-dev · 2025-06-29 16:09
chaos_princessup to you#asahi-dev · 2025-06-29 16:08
martinrnone of the added code will be used without a write to sysfs#asahi-dev · 2025-06-29 16:07
martinri think it makes sense to keep the rest of tipd working if there's an issue with the VDM feature, especially because a device reset won't help with the sysfs creation error#asahi-dev · 2025-06-29 16:03
martinrfirst kernel driver attempt so I appreciate the hints#asahi-dev · 2025-06-29 16:03
chaos_princess*err_disconnect#asahi-dev · 2025-06-29 16:03
martinrthat makes sense #asahi-dev · 2025-06-29 16:02
chaos_princessdepending on if it is ok to operate in this "degraded" state