marcan changed the topic of #asahi to: Asahi Linux: porting Linux to Apple Silicon macs | Not ready for end users / self contained install yet. Soon. | General project discussion | GitHub: https://alx.sh/g | Wiki: https://alx.sh/w | Topics: #asahi-dev #asahi-re #asahi-gpu #asahi-stream #asahi-offtopic | Keep things on topic | Logs: https://alx.sh/l/asahi
marvin24 has quit [Ping timeout: 480 seconds]
aleasto has quit [Remote host closed the connection]
aleasto has joined #asahi
aleasto has quit []
user982492 has joined #asahi
thevar1able has joined #asahi
off^ has quit [Remote host closed the connection]
artemist has joined #asahi
Retr0id has joined #asahi
Retr0id7 has joined #asahi
Retr0id has quit [Read error: Connection reset by peer]
Retr0id7 is now known as Retr0id
thevar1able has quit [Quit: My MacBook Air has gone to sleep. ZZZzzz…]
thevar1able has joined #asahi
thevar1able has quit [Quit: My MacBook Air has gone to sleep. ZZZzzz…]
user982492 has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
user982492 has joined #asahi
yuyichao has quit [Ping timeout: 480 seconds]
thevar1able has joined #asahi
PhilippvK has joined #asahi
dominic_ has joined #asahi
kov has quit [Quit: Coyote finally caught me]
phiologe has quit [Ping timeout: 480 seconds]
kov has joined #asahi
<dominic_>
Hi! I am a soft. eng. student in Canada and I really want to help on the project. Where can I start?
dominic_ has quit []
user982492 has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
user982492 has joined #asahi
akemin_dayo has quit [Ping timeout: 480 seconds]
marvin24 has joined #asahi
marvin24_ has quit [Ping timeout: 480 seconds]
thevar1able has quit [Quit: My MacBook Air has gone to sleep. ZZZzzz…]
<Glanzmann>
Missing is the initrd. But it is a deboostrap + kernel + firmware (no propiertary apple stuff) + tmux + vim + tcpdump + vlan + bridge-utils, an empty root passowrd, and a symlink that starts systemd.
<marcan>
streaming shortly, #asahi-stream as usual
marvin24_ has joined #asahi
MajorBiscuit has joined #asahi
marvin24 has quit [Ping timeout: 480 seconds]
Major_Biscuit has joined #asahi
MajorBiscuit has quit [Ping timeout: 480 seconds]
<landscape15[m]>
Glanzmann: thanks. But I was wondering how you can bundle all dtbs, because I usually cat m1n1 > dtb > Image, and with u-boot it should be m1n1 > dtb > u-boot
<landscape15[m]>
btw thanks for the guide, I will try u-boot when I have some time
<j`ey>
landscape15[m]: you can cat m1n1 > *.dtb > Image
<j`ey>
m1n1 will pick the right dtb
<_jannau_>
landscape15[m]: m1n1 only uses a matching dtb in the payload
<landscape15[m]>
jannau: oh that makes sense
<Glanzmann>
landscape15[m]: I wondered about that as well when kettenis told me first about this, but this is of course very convenient. Because you can use the same binary for multiple types of systems.
<Glanzmann>
Btw. the Debian initrd is very nice. You can do everything, including desktop, install arbitrary software and use it to install also other distributions, if you have a root tar ball.
<Glanzmann>
I like it that well, that I'm going to do a amd64 version of this as well.
<landscape15[m]>
Yeah but I don’t think we will ever reach the point the end user needs only an img. Simply because you need to set m1n1 > u-boot first. Have you installed the kernel in the Debian initrd?
<landscape15[m]>
I mean, you installed the kernel directly with debootstrap right?
Namidairo has quit [coherence.oftc.net nucleus.oftc.net]
hendry has quit [coherence.oftc.net nucleus.oftc.net]
Namidairo has joined #asahi
hendry has joined #asahi
eroux has quit [Ping timeout: 480 seconds]
eroux_ has joined #asahi
<mps>
I want to package m1n1 for alpine linux, what should I put as description? does 'experimentation playground for Apple Silicon' sounds good?
<j`ey>
"A bootloader and experimentation playground for Apple Silicon"
<mps>
j`ey: thanks
user982492 has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
aleasto has joined #asahi
<mps>
hmm, I installed 'llvm bash imagemagick dtc' but got error `make: *** No rule to make target 'data/bootlogo_128.png', needed by 'build/bootlogo_128.bin'. Stop.` seems something else is needed
<kettenis>
you forgot --recursive
<j`ey>
did you update the git submodules?
<j`ey>
^
<mps>
no, I downloaded tar.gz using last commit id
<_jannau_>
you need then another tar.gz for the artwork submodule
<mps>
hmm, yes, --recursive is missing but that is not good idea for distro builders
<Glanzmann>
landscape15[m]: No I did not, I just copied the modules over, but in the future I'll do that, because that allows for 'copy initrd to disk' installations. But honestly I was thinking more of users of the image could untar a rootfs or run debootstrap. But yes I should install the kernel in the image.
<Glanzmann>
landscape15[m]: The user will run the asahi installer. That leaves him with m1n1 + u-boot.
<Glanzmann>
Than the user can you any efi enabled installation media to boot the installer and install the OS, easy peasy.
<Glanzmann>
And with the debian live you can now install debian and any linux distribution which has a rootfs lying around somewhere.
aleasto has quit [Remote host closed the connection]
aleasto has joined #asahi
aleasto has quit [Remote host closed the connection]
aleasto has joined #asahi
aleasto has quit [Remote host closed the connection]
aleasto has joined #asahi
aleasto has quit [Remote host closed the connection]
aleasto has joined #asahi
<landscape15[m]>
Glanzmann: ok thanks. For now I’m gonna try jannau’s repo with asahi_defconfig, and see if the keyboard works. Maybe I should add the patch for Fn keys.
<mps>
waiting for patches kettenis posted to u-boot to be merged and then will make also u-boot
<marcan>
that just builds it, right?
<mps>
marcan: yes, and only bin and macho are in final package
boardwalk has quit [Quit: Ping timeout (120 seconds)]
boardwalk has joined #asahi
<mps>
I'll look later if I can make proxyclient as subpackage, if this will be needed for end users
<j`ey>
it shouldnt be
<mps>
I think so, but who know what would people ask.
<mps>
I packaged m1n1.bin and m1n1.macho as they will be needed for u-boot to chainload all them properly
gladiac has quit [Quit: k thx bye]
gladiac has joined #asahi
<NightRaven[m]>
Can you use wifi without a second machine ?
<j`ey>
with uboot
<DmitrySboychakov[m]>
afaik - not yet
<DmitrySboychakov[m]>
but I assume soon
<DmitrySboychakov[m]>
uboot enables pci?
<j`ey>
yeah
<DmitrySboychakov[m]>
good to know
<NightRaven[m]>
Oh ok thanks
<j`ey>
at least kettenis current branch, but they just said in -dev, that they wont submit the smc driver upstream
<mps>
NightRaven[m]: you can use wifi now without second machine, if I understood your question
<NightRaven[m]>
Ohh ok
<mps>
you need to prepare firmware first
<NightRaven[m]>
So the same procedure build the kernel and the .bin file and the payload
<NightRaven[m]>
Ohh do I just copy the files ?
<mps>
NightRaven[m]: yes
<mps>
NightRaven[m]: no to last question
<NightRaven[m]>
Oh ok
<mps>
asahi-installer contains script to extract firmware from macos and then to untar it to linux
<mps>
NightRaven[m]: I think Glanzmann created a guide for this
<NightRaven[m]>
Oh ok I will get started later on that but thanks
<mps>
btw, is audio card fixed
<j`ey>
no, audio hasnt had any changes
<mps>
would be nice to have even on headphone only
akemin_dayo has joined #asahi
<povik>
slowly working on it
<povik>
didn't see anyting in the pmgr dump you sent the other day, mps, that would explain the issue you see
<povik>
(no differences from the pmgr state on mac mini that linux boots from)
<mps>
povik: iiuc it works on mini?
<povik>
yes, on mine at least
<mps>
ah, interesting
<mps>
I thought it doesn't work on all
<mps>
if I can help please to not hesitate to give me instructions what to test or try to change
<mps>
s/to not/do not/
<povik>
which exact branch are you running? i should maybe build the same and remove the speaker from mac mini's devicetree, see if that doesn't reproduce it
<povik>
have you tried yesterday's asahi branch?
<mps>
asahi with latest commit id 43f1e1f83569b9ffee1076d413fc8621f08666c7
<mps>
and added jannau spi fix plus my patch for terminus 12x24 console font
<mps>
but these two do not 'touch' sound subsystem I think
<kettenis>
jannau: u-boot dt should be mostly up to date with asahi
<Glanzmann>
povik: I rebooted, had the headphones in the ear, I heard two clicks when x started.
<Glanzmann>
But now sound.
<Glanzmann>
no*
<povik>
see the MIXER again
<Glanzmann>
povik: Already checked the mixer.
<kettenis>
(I haven't looked at the merges marcan did yesterday yet)
yuyichao has quit [Ping timeout: 480 seconds]
<povik>
hmm, to summarize: seems like at some point the codec stops replying over i2c
<_jannau_>
nothing new yesterday just a rebase onto a newer linux-next tree since the old one had broken build system commits
<povik>
because it reads the device id and gets through probe, then later it breaks
<Glanzmann>
povik: I see. Btw. on the mini I was hearing sounce a few weeks ago.
<Glanzmann>
povik: I also disabled the iboot startup sound because I boot to much and it get on the nerves of the family. So maybe iboot only initializes something if it is going to use it.
<j`ey>
Glanzmann: time to reboot into macOS then :P
<Glanzmann>
j`ey: I can try it.
<Glanzmann>
give me a second
<mps>
so iiuc asahi branch is updated in last days?
<j`ey>
mps: on top of next-20220118
<Glanzmann>
j`ey: Could you give me a quick recap of the two smc streams today and yesterday. Anything interesting? I only read the chat.
<j`ey>
Glanzmann: smc core support yesterday, and smc gpio today
<j`ey>
Glanzmann: next step is to make smc gpio available from the DT, so that it can be enabled automatically
<mps>
j`ey: ah, good, will look later to pull
<j`ey>
(so it doesnt enable PCIe automatically yet)
<povik>
Glanzmann: iboot initializing or not should not be related to the i2c breakage, as we are taking the jack codec out of reset ourselves, starting with fresh slate
<Glanzmann>
povik: I see. I hear two clicks when x starts same as before, but no sound. Also played with the mixer.
<Glanzmann>
j`ey: I see. Is there any userland necessary to get all the goodies?
<Glanzmann>
j`ey: What features are in smc? I heard about enable pcie, powering off, lid close events, lcd brightness.
<j`ey>
Glanzmann: battery, fans
<Glanzmann>
I see, nice.
<j`ey>
and they should hopefully just go through normal kernel interfaces, so userland Just Works?
<Glanzmann>
I see.
<Glanzmann>
mps: Did you already build an asahi installer for alpine?
<Glanzmann>
mps: Or an alpine installer for m1/uboot/m1n1?
<mps>
Glanzmann: no, not yet, I'm not in hurry for this
<mps>
just added m1n1 in alpine, waiting for patches kettenis posted to u-boot to be merged and then will add u-boot-m1 to alpine
<Glanzmann>
mps: Nice. The thing i posted yesterday night, can be used for all sorts of stuff. Should have done that earlier and it was really easy once I found out about the symlink for init.
<povik>
j`ey: ad trying m1n1 scripts. maybe later it will come to that, but i would need to brush one off for the jack. so far i think this breakage is something linux-specific
<mps>
Glanzmann: would you mind to share what you posted, I missed that
<Glanzmann>
mps: https://tg.st/u/asahi-debian-live.txt - This allows you to create usb stick with efi/grub on it. So you can tell u-boot to boot from the usb stick and than you have a full blown Debian testing in ram. from there you can install gnome and use it as a desktop or use it to install debian or any other linux distrinution that has a rootfs on the internal nvme.
<Glanzmann>
mps: I'll also install iwd on it an soemone could extract the wifi firmware and than you can use it with wifi ...
<mps>
Glanzmann: thank you
<mps>
am I blind, don't see anything added today in asahi branch
<j`ey>
mps: nothing added just rebased
<mps>
j`ey: so not yet SMC merged
<j`ey>
no
<mps>
j`ey: thanks
<mps>
anyway will try rebuild with yesterdays merges to see will it help with sound
<povik>
unlikely
sikkiladho[m] is now known as sikkileo[m]
yuyichao has joined #asahi
yuyichao_ has quit [Ping timeout: 480 seconds]
<povik>
checked macos doesn't do anything special with gpio configuration of the i2c pins
<povik>
i should probably focus on the regcache issue, which is something i can tinker with, which may have the same root cause as the breakage on Glanzmann's air
<povik>
aaaaand... the regcache issue disappeared
<sven>
:D
<Glanzmann>
povik: If it helps, I can give you next week remote access to an air hooked up to a mini and a usb sound card that you can query remotely ...
Major_Biscuit has quit [Ping timeout: 480 seconds]
gor[m] has left #asahi [#asahi]
yuyichao_ has joined #asahi
joske has joined #asahi
<joske>
povik: I have exactly the same issue as Glanzmann on macbook air (asahi branch of yesterday)
<joske>
nothing if compiled as module, but these strange errors when linked in
Z751 is now known as Z750
yuyichao has quit [Ping timeout: 480 seconds]
<povik>
apple-pmgr-pwrstate 23b700000.power-management:power-controller@2c0: PS mca1: Failed to reach power state 0xf (now: 0x24f)
<povik>
joske: do you maybe see ^ when compiled as module? (if you still have the dmesg)
<joske>
povik, I don't have it anymore but can easily do
<joske>
both the DMA and the sound card driver as module?
<povik>
yes, that's how mps had it compiled i think
<Glanzmann>
povik: I have it.
joske has quit [Remote host closed the connection]
<Glanzmann>
povik: But I don't see these messages, but I'm sure that I have seen them before.
<Glanzmann>
povik: Wait, I was on the wrong laptop.
<povik>
:)
<Glanzmann>
povik: This is from yesterday (when I had the kernel with the modules). It is full of it: https://tg.st/u/kern.log.1
<povik>
joske: do you have CONFIG_COMMON_CLK_APPLE_NCO?
<povik>
Glanzmann: great, that seems to confirm that pwrstate issue is related to drivers being modules
<sven>
how strange
<joske>
povik, yes
<povik>
shrug, guess i am not doing something right in the driver
<povik>
at least now i can reproduce
<sven>
I don’t really see how that could happen, especially because the pwrstate change seems to be triggered
<sven>
sounds like the pmgr just refuses to transition
<sven>
oh… maybe some auto powermangement issue?
<povik>
the mca driver is so far oblivious to power management
<povik>
joske: as to your case, i don't know what happens there
<sven>
right, I’m talking about the PMGR auto power state magic ;)
<sven>
maybe it’s enabled and just want to keep the device in state 0x4 instead of going all the way to 0xf
<joske>
povik, what can I do to help?
<Glanzmann>
joske: Do you have Debian, if that is the case I can give you my 'deb' where I had the messages and you can try it, if you want.
<joske>
no manjaro
<Glanzmann>
I see.
<joske>
povik, wait, I only seem to have loaded the admac
<joske>
what is the module name for the sound driver?
<povik>
hm, module name, let's see how do i find out
<sven>
hrm, no, looks like auto pm is disabled according to that message :/
<sven>
no idea what’s going on then
<povik>
i assume platform drivers are modprobed on demand autoamtically for platform device drivers
<povik>
sven: true?
<sven>
uh, no idea :D
<povik>
:D
<povik>
*drivers for platform devices
<povik>
you got the gist
<sven>
but I’m not sure how that would work
<joske>
never mind, got it
<povik>
joske: snd-soc-apple-mca?
<joske>
povik, then yes: [ 736.414344] apple-pmgr-pwrstate 23b700000.power-management:power-controller@2c0: PS mca1: Failed to reach power state 0xf (now: 0x24f)
<povik>
there you go
<sven>
how strange
<sven>
it wants to remain clock gated for some reason
joske has quit [Remote host closed the connection]
user982492 has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<rkjnsn[m]>
Glanzmann, I just noticed in your alsamixer screenshot, the "card" is set to the PulseAudio compatibility device. povik can correct me if I'm wrong, but I'm guessing you want to be looking for a MIXER knob on the actual audio device. You can press F6 to switch which card alsamixer is controlling.