dorkbutt has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
dorkbutt has joined #asahi
Null_A_ has quit [Remote host closed the connection]
dorkbutt has quit [Max SendQ exceeded]
Null_A has joined #asahi
<refi64>
hello! so I'm trying to get the new GPU driver working, but, err, it seems that my Asahi install is old enough that it was using firmware 12.1. How to I update the firmware data to 12.3 like the installer uses nowadays?
Null_A has quit [Ping timeout: 480 seconds]
Null_A has joined #asahi
Null_A has quit [Ping timeout: 480 seconds]
Fah has joined #asahi
Fah has quit [Remote host closed the connection]
dorkbutt has joined #asahi
axboe_ is now known as axboe
Null_A has joined #asahi
riker77_ has joined #asahi
riker77 has quit [Ping timeout: 480 seconds]
riker77_ is now known as riker77
Null_A has quit [Read error: Connection timed out]
Null_A has joined #asahi
Null_A has quit [Ping timeout: 480 seconds]
Null_A has joined #asahi
Null_A has quit [Ping timeout: 480 seconds]
dorkbutt has quit [Read error: Connection reset by peer]
kettenis has quit [Remote host closed the connection]
<refi64>
Well I created a Frankenstein version of parts of the installer just to get out the 12.3 kernelcache file, replaced it, did an asahi-fwextract, and...it still didn't work? Not sure where else the GPU driver would be loading firmware from in the first place
Null_A has quit [Ping timeout: 480 seconds]
das_j has quit [Remote host closed the connection]
Guest1730 has quit [Quit: G-line: User has been permanently banned from this network.]
vx^ has joined #asahi
<refi64>
Well I have been informed that the GPU driver doesn't rely on firmware loaded by the kernel at all, so I've been looking in the wrong place the whole time, oops
marvin24_ has joined #asahi
marvin24 has quit [Ping timeout: 480 seconds]
mxz has quit [Quit: cya]
mxz has joined #asahi
vx^ has quit [Quit: G-line: User has been permanently banned from this network.]
user982492 has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<bcrumb>
my kernel is complaining that "Unknown kernel command line parameters "tp_printk tps6598x:* splash", will be passed to user space.", should I have entered the someone else and not kernel args?
bcrumb has quit [Quit: WeeChat 3.7.1]
tired- has joined #asahi
tired has quit [Read error: Connection reset by peer]
jacksonchen666 has joined #asahi
rvalue has quit [Remote host closed the connection]
rvalue has joined #asahi
jacksonchen666 has quit [Remote host closed the connection]
<jannau>
bcrumb: that sounds like the kernel misses tracing/ftrace. also 'tps6598x:*' should be 'trace_event=tps6598x:*'
jacksonchen666 has joined #asahi
Chainfire has quit [Remote host closed the connection]
eroc1990 has quit [Ping timeout: 480 seconds]
jamespmorgan has joined #asahi
jamespmo_ has quit [Ping timeout: 480 seconds]
jluthra has quit [Remote host closed the connection]
jluthra has joined #asahi
yuka has quit [Remote host closed the connection]
yuka has joined #asahi
LinuxM1 has quit [Quit: Leaving]
<refi64>
So I'm still stuck on this firmware issue 😅 I've updated macOS to 12.6, but the asahi driver still says it's on firmware 12.1.0, which isn't compatible with the "G13" GPU
<sven>
the firmware of the stub must be 12.3
<sven>
each installation has its own firmware
<refi64>
sven: ah, is it possible to update that?
<refi64>
I already updated my m1n1 version if that's what I'd need, so the installer doesn't let me anymore (but...really I could just patch it to let me 😅)
<sven>
did you install before the first alpha release?
<refi64>
I believe it was afterwards, but I accidentally was still on macOS 12.1 (I guess the installer didn't check yet?)
<marcan>
the installer always checked, but had a 12.1 option available with the UEFI boot option I think?
<marcan>
the actual distro images always required 12.3
<refi64>
Ah yeah that's what I did because I used Fedora instead of ALARM
<marcan>
then yeah, you'll have to reinstall the stub. probably back up the EFI partition, delete it and the stub partition, reinstall the stub, and restore the relevant contents (EFI subdirectory and maybe m1n1/boot.bin if you updated m1n1 more recently than the latest images)
<marcan>
and you'll have to fix the UUID for the EFI partition in your /etc/fstab most likely
<marcan>
(I hope this actually works, I'm not sure I ever tried an in-place "tight" reinstall like that... there could be issues with sizing, I should test it)
tinstew has joined #asahi
<refi64>
Oof so the m1n1 updater wouldn't even cover this? I guess presumably once I delete the ESP the installer would see the empty disk space
<marcan>
yes, it requires a full stub reinstall and the installer does not have a way of doing that without also doing a full install
<refi64>
Well this should be fun
<marcan>
it will see the empty space, I just hope it doesn't think it isn't enough or ends up with rounding issues
<refi64>
Would the risk there be "install fails" or "nukes my Linux partition"
<marcan>
install fails
<marcan>
I'm mostly worried that we fixed/tweaked the rounding since you installed, and what used to fit won't any more
<refi64>
🤞 hopefully it all works out, thanks for the help!
<marcan>
worst case I can just edit the EFI partition size to 490MB or something in the dev installer momentarily and you can use that
<marcan>
but try it as-is first
tinstew has quit [Ping timeout: 480 seconds]
bcrumb has joined #asahi
<jannau>
or if your macOS partition is directly before the stub shrink it by a small amount to make additional space
<bcrumb>
jannau: thanks, I will try it with that setting, I'll also check my kernel settings
<j`ey>
bcrumb: is it the asahi kernel? or your own build
<bcrumb>
j`ey: asahi-edge kernel
<bcrumb>
6.1rc8.asahi3-1
<j`ey>
that doesnt have any trace config options enabled :/
<bcrumb>
yuuup
<bcrumb>
How come we don't auto enable some settings like this for the kernels? Impacts performance?
<kettenis>
marcan: I helped upgrading another OpenBSD developer who was an early adopter
<kettenis>
there the partitions sizes changed by small amounts, but things still fit
<j`ey>
bcrumb: maybe they could be enabled on the edge kernel for that, up to markan
<kettenis>
the size changes meant that the BSD disklabel no longer had the correct bounds for the ESP though
<kettenis>
so OpenBSD users really should just reinstall
<bcrumb>
j`ey: 👍
bcrumb_ has joined #asahi
bcrumb has quit [Ping timeout: 480 seconds]
eroc1990 has joined #asahi
ah- has quit [Ping timeout: 480 seconds]
jamespmo_ has joined #asahi
tinstew has joined #asahi
jamespmorgan has quit [Ping timeout: 480 seconds]
<mamba>
2/6
leitao has joined #asahi
leitao has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
leitao has joined #asahi
Stary has quit [Read error: Connection reset by peer]
leitao has quit [Ping timeout: 480 seconds]
Stary has joined #asahi
zalyx has quit [Quit: later alligator]
zalyx has joined #asahi
zalyx has quit []
zalyx has joined #asahi
malte has quit [Remote host closed the connection]
malte has joined #asahi
Glanzmann has joined #asahi
<Glanzmann>
Btw. I rember once when the stub was increased by half GB very early on, you can also just reclaim a little bit of space from macos. Or you can use the debian live iso to resize the root partition using gparted.
Fah has joined #asahi
tired- has quit [Quit: /]
tired has joined #asahi
Fah has quit [Remote host closed the connection]
landscape15 has joined #asahi
Glanzmann has left #asahi [#asahi]
<landscape15>
Why is the installer in alpha state even though it's quite mature?
<landscape15>
Many people use it daily
<sven>
imho everything that isn't upstream is still alpha
<sven>
and we still rely on quite a few downstream patches
landscape15 has quit [Remote host closed the connection]
Fah has joined #asahi
landscape15 has joined #asahi
<landscape15>
Yeah I was thinking about that. In that case how "useful" is the mainstream linux kernel (using m1n1 of course)?
<Fah>
Anyone know the roadmap for hdmi support on the macbooks?
<Fah>
If i could get my external display working i'd spend more time in asahi since most of the stuff I care about works reasonably well
<j`ey>
I think 2038 was the last estimate
<sven>
I'll count that as asking "when is it done" though so i'll have to double that again!
<j`ey>
but in all seriousness, sv3n has a working prototype, and a bunch of the DCP code is in the main tree now, so progress!
<sven>
you still think i'm joking about the ETA? :P
* eta
doubles herself
<sven>
:>
<Fah>
Oh nice. which tree is this? I'm relatively late to the scene but i'm comfortable compiling kernels to test once i know where the code lives.
<sven>
it's not ready for testing
<Fah>
Fair nuff, i'm not trying to be a pest i'm just generally ignorant of the state of things.
<Fah>
Well, either way the progress on this is impressive. So thanks
user982492 has joined #asahi
cylm has quit [Ping timeout: 480 seconds]
mini0n has joined #asahi
dorkbutt has joined #asahi
bcrumb has joined #asahi
bcrumb_ has quit [Ping timeout: 480 seconds]
giskard has quit [Quit: My iMac has gone to sleep. ZZZzzz…]
hidden has joined #asahi
hidden has quit []
bcrumb has quit [Quit: WeeChat 3.7.1]
jacksonchen666 has quit [Ping timeout: 480 seconds]
jacksonchen666 has joined #asahi
jamespmorgan has joined #asahi
julio7359 has joined #asahi
bcrumb has joined #asahi
tinstew has quit [Ping timeout: 480 seconds]
bcrumb has quit []
bcrumb has joined #asahi
julio7359 has quit [Remote host closed the connection]
bcrumb has quit []
julio7359 has joined #asahi
jamespmo_ has quit [Ping timeout: 480 seconds]
bcrumb has joined #asahi
julio7359 has quit [Remote host closed the connection]
julio7359 has joined #asahi
bcrumb has quit [Quit: WeeChat 3.7.1]
bcrumb has joined #asahi
bcrumb has quit []
bcrumb has joined #asahi
bcrumb has quit []
bcrumb has joined #asahi
julio7359 has quit [Ping timeout: 480 seconds]
<refi64>
so I'm going to do the ESP recreation now, but first I noticed something...weird that I'm not sure how to feel about, diskutil seems to think I have 4TB of free space on an 8TB disk with two existing 4TB partitions: https://paste.sr.ht/~refi64/f58bcfa8822409829673c859fc6c99f5b9b26f48 right after disk0 partition 4
<refi64>
I'm not sure if this is going to screw up the installer's calculations or not? it's certainly strange to me, given that you can see the main 4TB macOS partition and 4TB Linux partition, but I'm also not super familiar with any "quirks" diskutil might have
<Glanzmann>
refi64: Make backups first. I have done such things in the following way: https://tg.st/u/upgrade_stub.txt Of course that is totally unsuported. I wanted to do a video so let me do a video and post it here in an hour or so.
<refi64>
Glanzmann: looks like my plan is similar, just no Debian & without using parted (because I've heard terrible things about it)
<refi64>
also using diskutil over mount directly
<refi64>
...out of curiosity, do you also have the suspiciously large empty space there?
<refi64>
like, a size showing as free equivalent to the size of your macOS or Linux partitions
<Glanzmann>
refi64: I would prefer parted over diskutil. But godspeed.
<marcan>
refi64: re the weird "free space", I think that usually happens if your partitions are out of order, which only happens if you used other partitioning tools
<marcan>
make sure you always put your partition table back in order if you use fdisk or something
<marcan>
(there was an expert option in fdisk for that, can't remember off the top of my head)
<marcan>
diskutil gets really confused by out of order partitions
<Glanzmann>
I have that only on the mini. Not on the air.
<refi64>
marcan: interesting, I only ever did things from within diskutil iirc
<refi64>
unless I used the buggy GUI at some point and forgot
mini is now known as mini_
<marcan>
(could also just be another diskutil bug...)
<Glanzmann>
Anyway, I need to get some sleep. See you guys tomorrow.
<refi64>
Fah: oh I missed this before, as-is you can use displaylink adapters w/ the out-of-tree evdi module, but you need the version from git which includes b884877267f11edaeb2a0f05201943e4252e22f2 (negativo's repo on Fedora ships it afaik, not sure about anywhere else), and also it appears to be massively with the new asahi GPU driver (not sure why yet)
<marcan>
Glanzmann: the wifi firmware goes with the stub firmware, and *this* is why we keep them in sync, because upgrades there can also break things
<marcan>
if you used an unsupported wifi firmware, shrug
<marcan>
we'll deal with it with the next official firmware/stub upgrade
<Glanzmann>
marcan: I deleted my stub partition and recreated it with the offical asahi installer (m1n1 + u-boot) only. That firmware seems to trigger the problem for me with the latest kernel m1n1, u-boot.
<Glanzmann>
Than I copied the firmware from the air to the mini and the oops is gone.
user982492_ has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<Glanzmann>
marcan: Must be something I'm missing. Because I took checksums of the firmware files and they're the same.
<Glanzmann>
Maybe I managed to corrupt my firmware files somehow. That would at least explain why the issue was gone after removed them and put them back.