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
tertu2 has quit [Read error: Connection reset by peer]
tertu has joined #asahi
linearcannon has quit [Read error: Connection reset by peer]
linearcannon has joined #asahi
chadmed has joined #asahi
chadmed has quit [Quit: Konversation terminated!]
chadmed has joined #asahi
eta has quit [Remote host closed the connection]
eta has joined #asahi
chadmed has quit [Quit: Konversation terminated!]
chadmed has joined #asahi
aleasto has quit [Quit: Konversation terminated!]
yuyichao_ has quit [Ping timeout: 480 seconds]
kit_ty_kate has quit [Quit: WeeChat 2.9]
nobodynada has joined #asahi
nobodynada has quit [Ping timeout: 480 seconds]
nobodynada has joined #asahi
nobodynada has quit [Ping timeout: 480 seconds]
<tpw_rules>
so i tried to build chromium with some patches for 16k pages just for fun (took 166 minutes on a mac mini) but it appears its IPC mechanisms are kind of broken
nobodynada has joined #asahi
caef^ has quit [Ping timeout: 480 seconds]
caef^ has joined #asahi
nobodynada has quit [Ping timeout: 480 seconds]
kit_ty_kate has joined #asahi
yuyichao_ has joined #asahi
jmr2 has joined #asahi
<jmr2>
I'm also unable to install the 12.0 beta8 stub, from either MacOS or 1TR. Similar symptoms in both cases - the fast loop never detects that the boot disk has changed.
<jmr2>
Issue not present when installing the 12.0.1 stub, at least when installing from MacOS. I haven't tried beta5.
<jmr2>
This is on a recently nuked MacBook Air, reinstalled with 12.1. So being "stuck" on 12.0.1 doesn't matter.
<jmr2>
Also, just tried the wifi/take1 branch: works perfectly on both bands, nothing to report. Great work!
jmr2 has quit [Remote host closed the connection]
phiologe has quit [Ping timeout: 480 seconds]
<marcan>
that means they stopped signing that version
<marcan>
before the reboot, when you get the boot picker in the installer, run bputil -g in another terminal
<marcan>
that will fix it
<marcan>
I've been meaning to change the installer flow to do that for this exact reason
jmr2 has joined #asahi
<jmr2>
Thanks. I'll let Glanzmann try it.
jmr2 has quit []
marvin24 has joined #asahi
marvin24_ has quit [Ping timeout: 480 seconds]
<Glanzmann>
Interesting that wifi works for jmr2. So maybe I have the old boot stub on the macbook air. I'll upgrade to 12.0.1 and try again.
nobodynada has joined #asahi
nobodynada has quit [Ping timeout: 480 seconds]
<Glanzmann>
marcan: Still no luck with the wifi. I can give you ssh access to the macbook air, if that helps. I tried the latest stub (12.1) - same symptom: interface is there, associated, no traffic.
<Glanzmann>
mps: About the wifi, the log entries in dmesg about not finding the firmware is normal. It only prints the files it is not finding and than it finds the files. You can tell if you have an additional wlan interface in (ip a s)
caef^ has quit [Read error: Connection reset by peer]
caef^ has joined #asahi
Dcow has quit [Ping timeout: 480 seconds]
Dcow has joined #asahi
tertu has quit [Ping timeout: 480 seconds]
sailorek1234 has joined #asahi
nobodynada has joined #asahi
joske has joined #asahi
MajorBiscuit has joined #asahi
chengsun has quit [Quit: Quit]
nobodynada has quit [Ping timeout: 480 seconds]
chengsun has joined #asahi
chengsun has quit []
chengsun has joined #asahi
<mps>
Glanzmann: interesting observation, though I think it is not 'normal'. and I don't have interface instantiated in kernel. Do you have any patch on top of wifi/take1 branch
<Glanzmann>
nope.
<Glanzmann>
but if you want I can send you my kernel, config and modules tree.
<marcan>
Glanzmann: I have an Air, I will test it later
<marcan>
but I find it unlikely that this is a driver problem considering it works on the mini and the air literally uses the same chip and firmware
<mps>
Glanzmann: thanks, but maybe only your config would be good to see diffs between it and mine
<marcan>
currently working on T2 support but I finally cracked the OTP problem
Dcow_ has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
Dcow_ has joined #asahi
<mps>
Glanzmann: though I guess I have pmgr problem and that could be reason interface is not instated
<mps>
Glanzmann: does audio works for you
<Glanzmann>
mps: I have currently no internal audio only via usb.
<Glanzmann>
marcan: I get it, but it does not work for me and mps, for jmr2 it seems to work, and I think there was another guy who also got it running on air. So maybe it is missing kernel option or problem with the userland?
<mps>
Glanzmann: I have this line few times in dmesg out: apple-pmgr-pwrstate 23b700000.power-management:power-controller@2c0: PS mca1: Failed to reach power state 0xf (now: 0x24f)
<marcan>
and I think that is just because that mca is disabled on the air
<marcan>
and we're currently not handling that properly
Major_Biscuit has joined #asahi
<Glanzmann>
mps: I'm currently using the macbook air on macos, so can't test right now, but i pasted my dmesg output using pastebot so it has to be in he backlog
<cth451_desktop>
but otherwise the system starts fine
<j`ey>
sorry 12.0.1
<cth451_desktop>
Ah.
<FireFox317>
Glanzmann, I can imagine that bputil -g only works in 1TR
<Glanzmann>
firefox317: I thought so, too. But marcan asked me to: 05:11 <@marcan> before the reboot, when you get the boot picker in the installer, run bputil -g in another terminal
<Glanzmann>
So i gave it a shot.
<joske>
hello everyone, tried several distros/desktops but none recognize the touchpad (clicking works, but no tapping, 2finger taps etc)
<joske>
so thinking it's a kernel issue?
<joske>
I used the config of Glanzmann
<joske>
can someone who has touchpad working on air post their config pretty please?
<mps>
https://tpaste.us/6Wdl this is my /etc/X11/xorg.conf.d/30-libinput.conf and it works
<mps>
joske: ^
<joske>
thx
<mps>
"Touchscreen" section is not needed, I copied this from my chromebook which have touchscreen
<Glanzmann>
joske: For me worked with debian testing and jannaus patch and config. At the moment I don't get it to work with the wifi branch which according to jannau is on par with the asahi branch regarding keyboard and touchpad.
<mps>
Glanzmann: I'm running wifi/take1 branch but it worked also with asahi branch
<Glanzmann>
I see, perfect. Thank you.
<Glanzmann>
mps: I also nuked my working userland from both system because I wanted to do a quickstart video, but now I'm unable to get the mini with hdmi up and running again ...
<joske>
mps, not working for me unfortunately. How can you check Xorg is using that file from xorg.conf.d?
<Glanzmann>
joske: Check xorg log and compare it to mps.
<Glanzmann>
mps: Could you post your xorg lorg, please?
<mps>
joske: /var/log/Xorg.0.log
<mps>
sure
<joske>
mps, checked that, didn't see anything there
<Glanzmann>
It will be a good reference for the future.
<joske>
mps, thx, which input packages do you have installed?
<mps>
and here is my POC of installer for alpine linux https://tpaste.us/mqDJ which is simple shell script
<mps>
joske: xf86-input-libinput
<joske>
also evdev?
<mps>
though I tested with xf86-input-synaptics
<mps>
joske: no, don't have xf86-input-evdev installed
<joske>
removed evdev but still same, will compare kernel config
sailorek1234 has quit []
sailorek1234 has joined #asahi
<Glanzmann>
mps: Thank you for your installer script. :-)
<mps>
Glanzmann: it is not finished, just made it to testing install
<joske>
mps, with your kernel config it works! thx!
<mps>
joske: nice to hear
Major_Biscuit has quit []
MajorBiscuit has joined #asahi
BastienSaidi[m] has joined #asahi
cth451_desktop has quit [Remote host closed the connection]
cth451_desktop has joined #asahi
Major_Biscuit has joined #asahi
<joske>
mps, but, with your config, no network via my ethernet via usb C dock
___nick___ has quit []
<joske>
this works with the other config :-/
MajorBiscuit has quit [Ping timeout: 480 seconds]
<FireFox317>
I think we should just build kernels with the same config that distros like arch linux arm are using
<FireFox317>
That way we wont have these annoying config issues
___nick___ has joined #asahi
YJSNPI has joined #asahi
YJSNPI has quit []
YJSNPI has joined #asahi
<marcan>
Glanzmann: the installer never tells you to go to macos, what is that about?
<marcan>
the recommended installation method is from 1TR, and the only one that will work for stubs not currently signed by Apple
<marcan>
I'm not even going to document doing installs from macOS once this is public, in fact I might add a warning to the installer if you're not in 1TR
<FireFox317>
marcan, there is some t2 linux discord that im in, should i ask them to test those patches?
<mps>
joske: this depends on ethernet device, I enabled one I have in my usb dongle and it works. you have to enable driver for your one and probably install firmware for it
<marcan>
please do
<marcan>
wifi/take2
YJSNPI is now known as RedL0tus
<marcan>
remember you need to use our firmware munger/copier
<marcan>
it won't work with the old manual method or whatever random firmware-du-jour people have been using until now
<FireFox317>
yes, will tell them
<marcan>
(in fact I explicitly broke it for bcm4364, which was added upstream merging two revisions into one... which was just completely broken)
<marcan>
(so those people using vanilla upstream with 4364 and manually copied firmware will have to at least rename it to add the right rev to the filenames)
RedL0tus is now known as YJSNPI
YJSNPI is now known as RedL0tus
RedL0tus is now known as YajuuSenpai
YajuuSenpai is now known as RedL0tus
RedL0tus is now known as Gur3n
Gur3n has left #asahi [ARRRRRRRRRRRGH]
<marcan>
oh also, I think the macaddr isn't set properly on T2s
<marcan>
I'll look into that
<marcan>
I think it's supposed to come from SROM
YJSNPI has joined #asahi
Dcow__ has joined #asahi
Dcow has quit [Ping timeout: 480 seconds]
Dcow has joined #asahi
Dcow__ has quit [Ping timeout: 480 seconds]
<marcan>
ahh,
<marcan>
so the MAC address shouldn't be in NVRAM if it's in SROM. I just need to change the firmware copier to not add it except for necessary platforms
DragoonAethis has quit [Remote host closed the connection]
DragoonAethis has joined #asahi
<mps>
is that reason for 'brcmfmac 0000:01:00.0: brcmf_pcie_download_fw_nvram: FW failed to initialize' in dmesg out
<joske>
marcan, this take2 branch does not compile?
<ChaosPrincess>
kernel folks now set Werror by default
<joske>
it's a warning treated as error
<mps>
iirc previously this was warning
<mps>
ChaosPrincess: ah yes
<mps>
I disabled this
<ChaosPrincess>
there is a config feature to disable it
<mps>
joske: if you use 'my' config this should be disabled already
<joske>
mps, I merged yours and Glanzmann's
<joske>
in any case, it points to some problems with that file
<joske>
that's why it's now enabled ;-)
<mps>
there are more of these, that's why disabled -Werror
<joske>
yes disabled it too for now
<mps>
anyway build with latest commit passed now
<mps>
marcan: should we test wifi/take2 branch?
<marcan>
wait for take3, I'm going to change the MAC handling
<marcan>
in a way that affects M1
<mps>
ok, though I already rebooted with take2
Gaspare has joined #asahi
<landscape15[m]>
mps: thanks a lot for your builds. They work perfectly.
<landscape15[m]>
I was able to run my custom rootFS
Dcow_ has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<mps>
landscape15[m]: you are welcome
chadmed has quit [Remote host closed the connection]
Dcow_ has joined #asahi
Gaspare has quit [Quit: Gaspare]
<marcan>
ok, pushed wifi/take3 with a change to how MAC address handling is done
<marcan>
you also need to update the asahi-installer repo and re-extract the firmware
<marcan>
this should make the MAC work properly on both T2 and M1
<landscape15[m]>
marcan: why did you need to fake MAC address before?
yuyichao_ has quit [Ping timeout: 480 seconds]
sailorek1234 has quit []
Gaspare has joined #asahi
psykose has quit [Remote host closed the connection]
psykose has joined #asahi
<marcan>
landscape15[m]: you need to have *some* MAC address in NVRAM, otherwise the M1 machine firmwares won't start without a MAC at all
<marcan>
previously I was just having a fake one in the nvram file and then setting the real one using firmware APIs on startup
<marcan>
but I just now realize that doing that on dongles which *do* have a MAC in SROM like on T2s overrides it
<marcan>
(and I don't want to go having the firmware extractor decide whether to add the var or not based on platform)
<landscape15[m]>
Ok thanks. So does Asahi-installer also copy all the needed firmwares?
<marcan>
the module in there yes
<marcan>
the installer does not, and besides that needs distro support
<marcan>
that will happen once the installer is a real installer
<marcan>
right now it just has that module that isn't otherwise used that you need to invoke by hand
Gaspare has quit [Quit: Gaspare]
joske has quit [Ping timeout: 480 seconds]
<mps>
got this: brcmfmac 0000:01:00.0: brcmf_pcie_init_ringbuffers: Allocating ring buffers failed
<mps>
lets try to upgrade firmware
nobodynada has joined #asahi
<mps>
same with updated firmware
nobodynada has quit [Ping timeout: 480 seconds]
yuyichao_ has joined #asahi
jmr2 has joined #asahi
<jmr2>
Nothing to report on wifi/take3. Works great on Air M1.
jmr2 has quit []
<mps>
anyone with MBP tested take3
<landscape15[m]>
Yeah I know the installer is not meant for end users, but I hope you’ll keep a dev version like that for engineers.
<landscape15[m]>
Especially because manually creating partitions can be annoying
aleasto has quit [Quit: Konversation terminated!]
aleasto has joined #asahi
aleasto has quit [Remote host closed the connection]
aleasto has joined #asahi
caef^ has quit [Ping timeout: 480 seconds]
caef^ has joined #asahi
<marcan>
landscape15[m]: dev version of what? the final installer will do all the repartitioning for you including resizing macos and creating the EFI partition and such
<marcan>
of course it will allow you to just create the stub+EFI partition if that's all you want, it's not going to force you to install a rootfs
<marcan>
(dunno if I will have a mode to do it without the EFI partition, seems kinda pointless since most people will want that anyway, but I will have a mode to re-create an existing stub with another version)
Dcow_ has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<mps>
Glanzmann: did you tested wifi/take3 on air?
jonnymac[m] has joined #asahi
aleasto has quit [Quit: Konversation terminated!]
aleasto has joined #asahi
aleasto has quit []
aleasto has joined #asahi
<zimsneexh>
Should the Firmware be re-extracted from macOS for wifi/take3? I saw the commit in the installer repo
<kettenis>
yes
<tpw_rules>
marcan: friendly reminder to add that sort pass to the firmware extractor
Gaspare has joined #asahi
tertu has joined #asahi
skipwich has joined #asahi
<landscape15[m]>
marcan: I mean, as there should be a fully-automated installer (for end user), a completely customizable installer is a good option.
<mps>
anyone have idea why I got this in dmesg when loading brcmfmac module 'brcmfmac 0000:01:00.0: brcmf_pcie_init_ringbuffers: Allocating ring buffers failed'
<marcan>
landscape15[m]: it's not just going to be a one-click installer; there will definitely be at least basic flexibility, possibly behind a toggle, if not an outright expert menu at some point
<marcan>
mps: I did increase the ringbuffer size, but I find it hard to see how it could behave differently on the Air and on the Mini, unless it's a firmware difference
<mps>
marcan: this is on mb pro
<mps>
j293 model
<marcan>
still the same chip and firmware
<mps>
aha
<marcan>
I'll test this later on more machines though
<mps>
I will try to find where this in source and look at if I could understand it
caef^ has quit [Ping timeout: 480 seconds]
caef^ has joined #asahi
<marcan>
you could try reverting d795c6f4901
<mps>
aha, ok
nobodynada has joined #asahi
<mps>
marcan: didn't helped, same message
nobodynada has quit [Ping timeout: 480 seconds]
martinstewart has joined #asahi
<martinstewart>
hello
<Major_Biscuit>
martinstewart: 👋
Gaspare has quit [Quit: Gaspare]
<martinstewart>
how is everyone
nobodynada has joined #asahi
<landscape15[m]>
One question, which config should I use to compile the kernel (I tried with `make menuconfig` but I found nothing Apple related)?
nobodynada has quit [Ping timeout: 480 seconds]
martinstewart has quit [Remote host closed the connection]
nobodynada has joined #asahi
<zimsneexh>
landscape15[m]: Did you specify ARCH=arm64 and the crosscompiler flag? It's documented on the wiki
joske has joined #asahi
leo60228 has joined #asahi
leo60228- has quit [Ping timeout: 480 seconds]
<landscape15[m]>
zimsneexh: yes I did, and I compiled with defconfig, which doesn’t contain any Apple specific config
darkapex has joined #asahi
<SamAdams[m]>
pcie
<j`ey>
landscape15[m]: defconfig has ARCH_APPLE
<j`ey>
thats enough to boot, but not all the fancy new features
<landscape15[m]>
j`ey: In fact, also USB doesn’t work having only ARCH_APPLE enabled
darkapex4 has quit [Ping timeout: 480 seconds]
<jannau>
landscape15[m]: which kernel repo/branch/version?
<j`ey>
arnd: should all m1 drivers be `default ARCH_APPLE`, or whats the "rule" about this?
<NightRaven[m]>
So u still need a second machine to get the wifi working
<j`ey>
clover[m]: and it mentions installing them with homebrew too
<j`ey>
NightRaven[m]: yes
martinstewart has joined #asahi
<NightRaven[m]>
Hmm ok
<clover[m]>
also needed `brew install p7zip` FYI
<j`ey>
hm, what failed without that?
<arnd>
j`ey it depends a bit on the subsystem maintainer. I generally like the 'tristate "some driver" if ARCH_APPLE ; depends on ARCH_APPLE || COMPILE_TEST ; default ARCH_APPLE' logic, but it's a bit wordy
<landscape15[m]>
jannau: latest linux-asahi
<arnd>
j`ey: actually I just got that wrong myself, I mixed up the case where the driver should be hidden but enabled and the case where it should be visible
<j`ey>
hidden but enabled is weird
<arnd>
a) is 'tristate "some driver" if COMPILE_TEST && !ARCH_APPLE; depends on THIS_SUBSYSTEM; depends on ARCH_APPLE || COMPILE_TEST ; default ARCH_APPLE'
nobodynada has quit [Remote host closed the connection]
nobodynada has joined #asahi
<j`ey>
that is quite wordy
<arnd>
j`ey: you'd do this for drivers that are absolutely required to boot the platform, but can be compile-tested
martinstewart has quit [Ping timeout: 480 seconds]
<j`ey>
AIC for example
<mps>
NightRaven[m]: what machine you have?
<NightRaven[m]>
<mps> "NightRaven: what machine you..." <- MacBook pro 2020 13 inch
<NightRaven[m]>
I would really like to get the wifi working
<j`ey>
NightRaven[m]: you'll have to wait a little longer!
<j`ey>
or get your wifi dongle working
___nick___ has quit [Ping timeout: 480 seconds]
<mps>
NightRaven[m]: does wifi works on it when you boot with hypervisor
<NightRaven[m]>
Boot with hypervisor ?
<NightRaven[m]>
I just use kmutil
<NightRaven[m]>
And that's it
<mps>
chainload boot from another machine with usb cable
<j`ey>
mps: NightRaven[m] wants to do it without a 2nd machine
<NightRaven[m]>
Yes
<j`ey>
therefore they need to wait
<mps>
ah, I see
<NightRaven[m]>
K
<NightRaven[m]>
I will wait
<mps>
someone reported today that wifi works fine on mb air
<mps>
ofc with 'hypervisor boot'
<joske>
does anyone have a guide on how to setup u-boot?
MajorBiscuit has joined #asahi
<Glanzmann>
mps: Not yet had very long day today.
Major_Biscuit has quit [Ping timeout: 480 seconds]
martinstewart has joined #asahi
<Glanzmann>
marcan: When I tried asahi-installer from 1tr it told me that it only allows me to change the security policy of the active boot volume. Since it was not possible for me to change that from 1tr because the boot picker endless looped me, this was not working out for me.
<joske>
mps, yes, works very wel on air here, without hypervisor
<joske>
mps, it does take a few minutes before the driver comes through
<joske>
but then works
<joske>
had to symlink the firmware, as the driver is looking for another filename :-/
<mps>
joske: but it doesn't have internal keyboard yet
<joske>
mps, thx, so not very useful yet, I'll wait
<mps>
if you are ok with usb keyboard then it is fine
<FireFox317>
joske: you dont have to create those symlinks. The dmesg only showed error messages for the ones it tried, but IIRC it doesnt print a message of the one that actually succeeds
<joske>
mps, so then in 1TR you use that m1n1 and then it always boots to uefi?
<j`ey>
joske: yeah are you sure it didnt work without those symlinks?
<joske>
firefox317, ah, didn't know
<joske>
j`ey, no, maybe I didn't wait long enough
<joske>
it seems to take several minutes before wireless networks (or even wlan0) show up in network manager
<FireFox317>
joske: probably the kernel loads the bcrmfmac driver before your rootfs is mounted, and thus it will fail to find the firmware on initial boot. I think it then tries to load the driver again after a minute or so, so the best solution here is put the firmware in a initramfs
<joske>
firefox317, aha, maybe something like that. And then you pass the initrd as argument and not the actual rootfs?
<joske>
did not try to use an initrd yet
martinstewart has quit [Ping timeout: 480 seconds]
<FireFox317>
joske: i guess you are booting with a second pc right? you dont to change the linux boot-args (root=etc), but you can just concatenate a initramfs after the device tree
<joske>
firefox317, yes, booting from another pc. So creating the initramfs on the other PC?
<FireFox317>
joske: jep, basically you need a directory with the same structure as the linux fs. So a new directory with /lib/firmware/brcm/<etc> and then put that into a cpio archive and gzip it
<joske>
firefox317, aha, seems simple enough
<jannau>
joske: check if FW_LOADER_USER_HELPER_FALLBACK I think that adds a timeout for every tried firmware load
<joske>
jannau, thx, will check that
martinstewart has joined #asahi
<jannau>
a timeout of ~60 seconds
<mps>
yes cat /sys/class/firmware/timeout
<joske>
jannau, I don't see any firmware errors, but still takes some time for wifi to appear
<tpw_rules>
oh christ that's what that was
<tpw_rules>
is there a way to change the default delay
<joske>
don't find any parameter for this in .config
martinstewart has quit [Ping timeout: 480 seconds]
<mps>
only in source iirc
<mps>
or write to this file on boot
<joske>
it takes 430 seconds for wifi to appear
<mps>
echo 30 > /sys/class/firmware/timeout
<mps>
joske: you can check dmesg if there is something about that
<joske>
mps, did, nothing useful
<jannau>
joske: disable CONFIG_FW_LOADER_USER_HELPER CONFIG_FW_LOADER_USER_HELPER_FALLBACK in the kernel config
<mps>
hmm
<jannau>
disabling CONFIG_FW_LOADER_USER_HELPER_FALLBACK is probably enough
<joske>
jannau, indeed, now wifi instantaneous
<joske>
thx!
<joske>
goint to sleep, night all
<joske>
*going
<joske>
very good progress :-)
joske has quit [Ping timeout: 480 seconds]
nobodynada has quit [Ping timeout: 480 seconds]
hckr has joined #asahi
hckr has quit []
yuyichao has joined #asahi
hckr has joined #asahi
yuyichao_ has quit [Ping timeout: 480 seconds]
yuyichao_ has joined #asahi
MajorBiscuit has quit [Ping timeout: 480 seconds]
yuyichao has quit [Ping timeout: 480 seconds]
nobodynada has joined #asahi
yuyichao_ has quit [Ping timeout: 480 seconds]
nobodynada has quit [Ping timeout: 480 seconds]
nobodynada has joined #asahi
nobodynada has quit [Ping timeout: 480 seconds]
hckr has quit [Quit: hckr]
manawyrm has quit [Quit: Read error: 2.99792458 x 10^8 meters/second (Excessive speed of light)]