#
#asahi-dev · 2026-02-20 21:27
chaos_princessi wouldn't describe it using those specific words, but something like that, yes
#asahi-dev · 2026-02-20 21:26
abrenderRight, drama about architecture-specfic drivers
#asahi-dev · 2026-02-20 21:26
chaos_princessyea, upstream kernel will not have apple cpuidle any time soon
#asahi-dev · 2026-02-20 21:25
fl0_idno I'm not, but if you read about it being added, then it was added to that upstream, not asahi
#asahi-dev · 2026-02-20 21:25
abrenderfl0_id - are you sure? I don't see apple silicon here: https://github.com/torvalds/linux/tree/master/drivers/cpuidle - maybe it's somewhere else?
#asahi-dev · 2026-02-20 21:23
abrenderThanks for pointing out that the commit I linked to was a rebase. I should have looked closer.
#asahi-dev · 2026-02-20 21:22
abrenderI said "yeah" because I thought you were referring to the official kernel released by Asahi rather than me custom compiling a kernel from a branch in the repo. Thank you very much for clarifying
#asahi-dev · 2026-02-20 21:22
fl0_idso upstream "official" linux now also has cpuidle for apple silicon
#asahi-dev · 2026-02-20 21:22
abrenderokay, gotcha. thanks.
#asahi-dev · 2026-02-20 21:21
fl0_idand it is slowly changing
#asahi-dev · 2026-02-20 21:21
chaos_princessit is possible, just mostly useless on not-m1-mini
#asahi-dev · 2026-02-20 21:21
abrenderI thought it was impossible to run linux on Apple Silicon using the upstream linus kernel without all of the asahi patches?
#asahi-dev · 2026-02-20 21:21
chaos_princessas in the main one that is _the_ linux kernel
#asahi-dev · 2026-02-20 21:20
chaos_princessthe linus tree
#asahi-dev · 2026-02-20 21:19
abrender@chaos_princess: can you help me to understand what you mean by upstream kernel? Sorry for the ignorance - just trying to learn the lingo
#asahi-dev · 2026-02-20 21:16
jannauthe author date of that commit is "Date: Mon Apr 17 20:41:13 2023 +0900"
#asahi-dev · 2026-02-20 21:16
chaos_princesshttps://asahilinux.org/2022/11/november-2022-report/ ctrl+f "Idle Ideations"
#asahi-dev · 2026-02-20 21:16
j`eyfor example https://github.com/AsahiLinux/linux/commit/cbf1c41a00095679a216896cc0bbf8a1d3db0d8a
#asahi-dev · 2026-02-20 21:15
chaos_princess"commited" means "last changed"
#asahi-dev · 2026-02-20 21:15
j`eythat's rebased code though
#asahi-dev · 2026-02-20 21:14
abrenderIt looked to me like cpuidle wasn't added until Dec 2025: https://github.com/AsahiLinux/linux/commit/4c135ef5c4c15479593b86a6da1b7abf87ffa6e7
#asahi-dev · 2026-02-20 21:14
abrenderoh sorry then maybe I misunderstood by upstream? It's just regular Asahi Fedora and I ran the `dnf upgrade` command
#asahi-dev · 2026-02-20 21:14
chaos_princessand we had cpuidle since forever
#asahi-dev · 2026-02-20 21:14
chaos_princessthat is not the upstream kernel
#asahi-dev · 2026-02-20 21:13
abrenderYeah - 6.18.10-402.asahi.fc42.aarch64
#asahi-dev · 2026-02-20 21:13
chaos_princessare you using upstream kernel?
#asahi-dev · 2026-02-20 21:13
abrender(I basically put the laptop into suspend, then closed the lid, and opened it a few hours later and checked the loss of battery %)
#asahi-dev · 2026-02-20 21:12
abrenderQuick data point: M2 Macbook Air 14 inch (J413) - in January I tested and it lost 1.5% battery per hour on suspend. Now that the cpuidle driver was added in 6.18 I measured a 33% reduction in battery use - 1% per hour instead. I notice there was no mention of cpuidle in the progress reports - maybe it slipped through, or maybe it wasn't expected to make any difference?
#asahi-dev · 2026-02-20 19:40
jannausiirko[m]: yes, u-boot needs explicit m3 support
#asahi-dev · 2026-02-20 15:04
fl0_idin some cases for specific devices, if the older compatible still works, it would still work. but if it only checks the machine, yes. you could just try adding t8122
#asahi-dev · 2026-02-20 14:53
siirko[m]Seeing this might explain why it get stucked, there just no implementation yet for m3
#asahi-dev · 2026-02-20 14:53
siirko[m]https://github.com/AsahiLinux/u-boot/blob/asahi/arch/arm/mach-apple/board.c#L713
#asahi-dev · 2026-02-20 14:53
siirko[m]* https://github.com/AsahiLinux/u-boot/blob/asahi/arch/arm/mach-apple/board.c#L21
#asahi-dev · 2026-02-20 14:26
siirko[m]https://github.com/AsahiLinux/u-boot/blob/asahi/arch/arm/mach-apple/board.c#L21 Seeing this might explain why it get stucked, there just no implementation yet for m3
#asahi-dev · 2026-02-20 13:56
siirko[m]fl0_id: yup, just hangs on asahi logo
#asahi-dev · 2026-02-20 13:55
fl0_idmaybe someone else can weigh in, haven't used uboot much. so it hangs after loading kernel, but before you can give it commands?
#asahi-dev · 2026-02-20 13:53
siirko[m]edit : I've looked at m1n1 code and I think this is closely related to u-boot or something else that I've not done correctly on u-boot code
#asahi-dev · 2026-02-20 13:53
siirko[m]* I compiled it from source like explained here https://asahilinux.org/docs/sw/u-boot/#prerequisites I put the right dtb from https://github.com/AsahiLinux/linux/pull/445 and applied https://github.com/AsahiLinux/m1n1/pull/538
#asahi-dev · 2026-02-20 13:38
siirko[m]fl0_id: I compiled it from source like explained here https://asahilinux.org/docs/sw/u-boot/#prerequisites I put the right dtb from https://github.com/AsahiLinux/linux/pull/445 and applied https://github.com/AsahiLinux/m1n1/pull/538
#asahi-dev · 2026-02-20 13:35
fl0_idah. yeah not sure about u-boot. did you install that manually or via (modified) installer?
#asahi-dev · 2026-02-20 13:32
siirko[m]fl0_id: nevermind about that it was bad english , I meaned that while I tried to boot a kernel directly through m1n1 stage 1 I tried with m1n1 -> uboot
#asahi-dev · 2026-02-20 13:29
fl0_idsiirko[m] what do you mean by other way round?
#asahi-dev · 2026-02-20 13:16
siirko[m]I tried to go the other way around and do the m1n1 -> u-boot, but it seems that I'm stuck on asahi logo. This happens juste before it prepare to load kernel. Is anyone have a clue of what I might did wrong ? I see within the log that it loads correct device tree blob and pcie initialisation goes alright
#asahi-dev · 2026-02-20 09:53
chaos_princessthat assumes that nvme works. it probably does, but having an initramfs will help with figuring out if everything is ok, or something went wrong
#asahi-dev · 2026-02-20 09:53
siirko[m]I choose the path of booting a linux kernel directly like explained in https://asahilinux.org/docs/sw/m1n1-user-guide/#configuring-to-boot-linux-directly
#asahi-dev · 2026-02-20 09:53
siirko[m]jannau
#asahi-dev · 2026-02-20 09:53
siirko[m]chaos_princess: Thank you !
#asahi-dev · 2026-02-20 09:50
jannauinitramfs shouldn't be necessary when everything is buit-in (apple nvme and sart for nvme
#asahi-dev · 2026-02-20 09:47
chaos_princessyou are correct. give it an initramfs
#asahi-dev · 2026-02-20 09:46
siirko[m]Hi !, I've managed to boot up partially linux kernel on my m3, but it is blocked with "waiting for root device /dev/nvme....", I put this string into the boot bin 'chosen.bootargs=earlycon debug root=/dev/nvme0n1p6 rootwait rw' with other information. The only thing that came to my mind is that it might be because I didn't gave it a initramfs and at this stage it can't load nvme

plus ›