#
#asahi-dev · 2026-03-10 08:48
chadmedyeah
#asahi-dev · 2026-03-10 08:48
svenif we need to re-apply chickens we need a bit more of m1n1 to stay resident once linux runs but we'll need that anyway if we ever get to psci-via-efi or whatever
#asahi-dev · 2026-03-10 08:48
nickchanbehavior on a11 should be close enough if wfi retention bit is unset but it's probably best to verify on m1
#asahi-dev · 2026-03-10 08:48
chadmedshould we have printfs in that hot path
#asahi-dev · 2026-03-10 08:47
nickchanit tries to CPU idle in linux then on uart i immediately see m1n1 output until some spinlock exploded and rebooted
#asahi-dev · 2026-03-10 08:47
svenwhich doesn't matter for us, what matters is what happens on M1+
#asahi-dev · 2026-03-10 08:46
nickchani think that's what happened when the cpuidle driver tried to run on A7-A11
#asahi-dev · 2026-03-10 08:46
svenin theory, yes. but has anyone tried that?
#asahi-dev · 2026-03-10 08:46
nickchanif IRQs are not masked during shutdown in CYC_OVRD then the core would come up again when IRQ's received
#asahi-dev · 2026-03-10 08:44
svenwe also don't know if we need to re-apply chickens, etc. or if that persists across full shutdown
#asahi-dev · 2026-03-10 08:44
svenchadmed: and yeah, something like that hack would work downstream I think
#asahi-dev · 2026-03-10 08:44
sventhat's why I said we don't fully understand it
#asahi-dev · 2026-03-10 08:43
svenbut i don't think anyone ever tried it for e.g. _all_ cores at the same time
#asahi-dev · 2026-03-10 08:43
svenyeah
#asahi-dev · 2026-03-10 08:43
nickchanisn't that just the smp_stop_secondary code
#asahi-dev · 2026-03-10 08:43
sveni think there's code in m1n1 that does it (the shutdown secondary core stuff)
#asahi-dev · 2026-03-10 08:38
chadmedas in, coming out of the state seems standard enough but do we know how to actually get into it properly
#asahi-dev · 2026-03-10 08:36
chadmedbut if we dont actually understand how it works fully theres no point
#asahi-dev · 2026-03-10 08:36
chadmedyeah i was looking into what would have to change in m1n1, i dont think its _that_ much (put a trampoline at the entry point, reserve the range, resume table for the cpuidle driver to write to?)
#asahi-dev · 2026-03-10 08:34
svenyeah, that one
#asahi-dev · 2026-03-10 08:34
sventhere’s a deeper state that we don’t fully understand
#asahi-dev · 2026-03-10 08:34
yukaI think completely shutting the core so that it would boot back up at RVBAR
#asahi-dev · 2026-03-10 08:29
chadmedcpuidle-apple already does deep wfi it seems
#asahi-dev · 2026-03-10 08:29
chadmedhuh i thought the busted sleep state was deep wfi, is there a deeper state we're missing?
#asahi-dev · 2026-03-09 18:11
chaos_princessyes. they can comment (or not) if they wish
#asahi-dev · 2026-03-09 18:10
cyberheadchaos_princess: is someone still working on sptm?
#asahi-dev · 2026-03-09 18:02
chaos_princessyes-ish, you can boot something on m4, but nobody is really looking into it, since the main blocker for m4 is sptm
#asahi-dev · 2026-03-09 17:55
cyberheadhas there any progress been done with booting on M4?
#asahi-dev · 2026-03-07 21:13
StevenDay[m]Cybersecurity student getting his master degree wanting to help any way I can. Have a macbook air 15in M3 2024, 24Gb of ram.
#asahi-dev · 2026-03-07 21:11
noopwafel(unfortunately it also needs the hacks which we were using to boot it on M3..)
#asahi-dev · 2026-03-07 21:10
noopwafelhm, virtualized 14.8.3 on M1 is ending up with all CPUs in arm64_retention_wfi (or if I boot with wfi=0, in an idle loop) if I give it more than one CPU
#asahi-dev · 2026-03-07 20:45
kitlithunderstandable!
#asahi-dev · 2026-03-07 20:40
sven*downstream
#asahi-dev · 2026-03-07 20:40
sveni'd much rather focus on getting whatever is still downstreamed reviewed so that we don't have this problem anymore rather than adding yet another bandaid
#asahi-dev · 2026-03-07 20:37
kitlithnods that would be the piece that i'm missing.
#asahi-dev · 2026-03-07 20:36
chaos_princessi don't think overlays can express all the device tree surgery that we do
#asahi-dev · 2026-03-07 20:35
kitlithchadmed: from my perspective the problem is currently in m1n1, and I'd like to move it out of m1n1. specifically, the fact that m1n1 needs a copy of the device tree that it adds a bunch of details to, and then passes on to uboot, kernel, etc. nothing else can make those changes, so once we're in our boot menu and trying to select between different kernels, we can't select a different devicetree.
#asahi-dev · 2026-03-07 19:41
fl0_idFiniteCode[m] sure, whatever works for you and ofc that's just my opinion
#asahi-dev · 2026-03-07 19:40
FiniteCode[m]fl0_id: mhmm sure, I'm about to complete my highschool exams on 23rd March, so after that surely yes!
#asahi-dev · 2026-03-07 19:36
fl0_idFiniteCode[m] well, integralpilot is more busy with school and stuff now I think, so could take a while I assume. In the meantime, I would recommend just trying this PR manually, at least then you will have a working dev setup and get a feel for things
#asahi-dev · 2026-03-07 17:57
FiniteCode[m]my bad for messing up
#asahi-dev · 2026-03-07 17:57
FiniteCode[m]I meant the same PR XD
#asahi-dev · 2026-03-07 17:57
FiniteCode[m]jannau: ohh ok
#asahi-dev · 2026-03-07 17:57
jannauFiniteCode[m]: you mean https://github.com/AsahiLinux/linux/pull/445 ? Not my PR, I just reviewed and waiting for IntegralPilot to address the review comments
#asahi-dev · 2026-03-07 17:52
FiniteCode[m]fl0_id: I do, I saw a PR from jannau that hasn't been merged yet, i was planning to write new dts but since, jannau has already submitted a PR, it would be very helpful if it were merged soon, so I could contribute and make sure that everyone's on the same page
#asahi-dev · 2026-03-07 17:15
jannauwe want to support 14.x firmware on all devices once we bring up DCP for M3 devices
#asahi-dev · 2026-03-07 17:13
jannauah, the one in main. I was thinking of the one in hv_init(): `display_shutdown(DCP_QUIESCED)`
#asahi-dev · 2026-03-07 17:04
noopwafelthere's a comment which says that DCP is kicked to sleep to avoid losing the FB on monitors which cause reconnect cycles. but okay, it sounds like this isn't worth spending time on, it is working great for me and 14.x isn't a supported config on M1 :)
#asahi-dev · 2026-03-07 16:56
jannauI think the issue wasn't the sleep state itself
#asahi-dev · 2026-03-07 16:55
jannauwhich comment? I'm trying to remember what macos does to bring dcp up

plus ›