ChanServ changed the topic of #asahi to: Asahi Linux: porting Linux to Apple Silicon macs | "Does XXX work yet?": https://alx.sh/fs | GitHub: https://alx.sh/g | Wiki: https://alx.sh/w | Topics: #asahi-dev #asahi-re #asahi-gpu #asahi-alt #asahi-stream #asahi-offtopic | Keep things on topic | Logs: https://alx.sh/l/asahi
ddxtanx_ has quit [Ping timeout: 480 seconds]
zumi_ has joined #asahi
zumi has quit [Remote host closed the connection]
zumi_ is now known as zumi
sand4rt has joined #asahi
sand4rt has quit [Ping timeout: 480 seconds]
bashiordache has joined #asahi
Calandracas has quit [Remote host closed the connection]
<bashiordache>
aleksrutins: thanks for the link but I'm a little more attached to my GUI than I'd like to admit (even using vscode with vim bindings rather than full on vim)
<bashiordache>
I think I'm most of the way there to having Assetto Corsa working, I feel like I forgot something fundamental tho
Calandracas has joined #asahi
<bashiordache>
so, I build wine and wine-common to enable me to run protontricks, when running protontricks I get these messages in the logs: (incoming long communication)
<bashiordache>
<jemalloc>: Unsupported system page size
<bashiordache>
<jemalloc>: Unsupported system page size
<bashiordache>
<jemalloc>: Unsupported system page size
<bashiordache>
<jemalloc>: Unsupported system page size
<bashiordache>
Traceback (most recent call last):
<bashiordache>
File "/home/iordachesebastian/.local/bin/protontricks", line 8, in <module>
<bashiordache>
sys.exit(cli())
<bashiordache>
File "/home/iordachesebastian/.local/lib/python3.13/site-packages/protontricks/cli/main.py", line 32, in cli
<bashiordache>
especially the unsupported system page size, I feel like I had a solution for that some days ago but can't recall, what am I missing?
<bashiordache>
thanks
<nicolas17>
pasting 28 lines of text into IRC is considered annoying, use a pastebin website next time
<nicolas17>
it looks like you're running that directly in asahi?
<nicolas17>
you should probably be running it inside the muvm, which has a 4k pagesize
pb17 has quit [Ping timeout: 480 seconds]
kujeger has joined #asahi
shiggitay has joined #asahi
Calandracas has quit [Remote host closed the connection]
Calandracas has joined #asahi
kujeger- has quit [Ping timeout: 480 seconds]
pb17 has joined #asahi
bashiordache has quit [Remote host closed the connection]
Calandracas has quit [Remote host closed the connection]
Calandracas has joined #asahi
Brainium has quit [Quit: Konversation terminated!]
Moprius has quit [Quit: bye]
gabuscus has quit []
dostoevsky has quit [Remote host closed the connection]
ddxtanx has quit [Ping timeout: 480 seconds]
pb17 has quit [Ping timeout: 480 seconds]
tobhe has joined #asahi
pb17 has joined #asahi
ddxtanx has joined #asahi
tobhe_ has quit [Ping timeout: 480 seconds]
johey has quit [Ping timeout: 480 seconds]
gabuscus has joined #asahi
johey has joined #asahi
ddxtanx has quit [Remote host closed the connection]
pb17 has quit [Ping timeout: 480 seconds]
<tsujp>
fl0_id: yeah I do use libkrun on macOS, been following that project for a while. `libkrun-efi` makes it a lot easier to use now that you don't have to compile a custom Linux kernel yourself (setting up a VM for that etc as a bootstrap). I actually just finished packaging krunkit, libkrun, and slp's virglrenderer for MacPorts.
<tsujp>
fl0_id: I was just curious about muvm because of the pagefile size mention; there's also krunvm which is slp's older (now deprecated via our Matrix chats) vm runner.
pb17 has joined #asahi
luna__ has joined #asahi
zumi_ has joined #asahi
zumi has quit [Remote host closed the connection]
luna__ has left #asahi [#asahi]
zumi_ is now known as zumi
<chadmed>
muvm grew out of krunvm
ddxtanx has joined #asahi
pb17 has quit [Ping timeout: 480 seconds]
pb17 has joined #asahi
KindOne has quit [Remote host closed the connection]
KindOne has joined #asahi
shiggitay has quit [Ping timeout: 480 seconds]
chadmed has quit [Quit: Konversation terminated!]
chadmed has joined #asahi
Sebastian2 has joined #asahi
Sebastian2 has quit []
pb17 has quit [Ping timeout: 480 seconds]
aliasrush has joined #asahi
pb17 has joined #asahi
millefy0 has joined #asahi
millefy has quit [Ping timeout: 480 seconds]
millefy0 is now known as millefy
riker77 has quit [Quit: Quitting IRC - gone for good...]
JayBeeFOSS has quit [Ping timeout: 480 seconds]
sand4rt has joined #asahi
JayBeeFOSS has joined #asahi
riker77 has joined #asahi
dcavalca8569 has quit [Ping timeout: 480 seconds]
dcavalca8569 has joined #asahi
Moprius has joined #asahi
pb17 has quit [Ping timeout: 480 seconds]
pb17 has joined #asahi
aliasrush_ has joined #asahi
aliasrush has quit [Ping timeout: 480 seconds]
aliasrush_ has quit [Quit: Konversation terminated!]
zumi_ has joined #asahi
zumi has quit [Remote host closed the connection]
zumi_ is now known as zumi
monthebiff has joined #asahi
<monthebiff>
hey hey, just had a glance through chat logs, saw some messages with people surprised at gaming being an in demand feature... guess im one the people in the camp surprised thats a surprise :P
<monthebiff>
with only the macbook on me while travelling, gaming opportunites are limited on the macos front... but the idea of dragging around a windows system saddens me. WSL is very good, but i just can't enjoy that desktop environment.
<monthebiff>
honestly i've been wishing for more linux gaming on deskop x86 machines for years, just to get away from windows. but the apple machine, linux, and video games? that's like a holy grail pipedream to me.
<Fridtjof>
does anyone else at congress have issues with the eap-tls wifi and OWE? All i can connect to is 38C3-open in unencrypted mode, which is obviously way less than ideal
___nick___ has joined #asahi
<Fridtjof>
Enterprise wifi just doesn't connect, while trying OWE seems to crash the wifi firmware, judging by the eternal timeouts logged by the kernel after trying it
<monthebiff>
i dunno if its the same thing but i've had issues connecting to the hostel wifi with webauth, seems to freeze the entire network interface/service. get a notification about a lack of secrets provided, but no dialogue / browser for web auth.
<monthebiff>
made me sad
pb17 has quit [Ping timeout: 480 seconds]
<PaulFertser>
Fridtjof: if you can do a quick check for the same in macOS on that device probably that would be a good datapoint. Because if firmware is buggy then it would be visible on all OSes, and if it's a driver limitation it can be fixed eventually.
<Fridtjof>
oh this is absolutely just an asahi issue
<Fridtjof>
one thing i'm not 100% about but seems annoying to check is whether the firmware that's ended up on the asahi side is somehow older than on the macOS side
<PaulFertser>
You mean you actually checked both on macOS already, ok.
<Fridtjof>
yeah
<PaulFertser>
The firmware version is visible in the Linux, the kernel, log, but hm, how to check what it is on macOS, no real idea, probably in some log too, or probably you can find the file and "strings" on it.
<Fridtjof>
so before i go investigating if i "maybe possibly" have an outdated firmware and whether it's worth trying to update, i'd make sure this works for anyone at all right now
<PaulFertser>
Should be in /usr/share/firmware/wifi on macOS.
<chaos_princess>
you do not have an "outdated firmware", except for old, pre-fedora installs asahi always uses the specific version from macos 13.5
<PaulFertser>
Extracted from IPSW?
pb17 has joined #asahi
Brainium has joined #asahi
<PaulFertser>
So comparison of wifi features between macOS and Linux is almost guaranteed to compare different firmware for the PCIe bcmfmac chip?
<Fridtjof>
ahh okay, so it wouldn't be a firmware problem but rather a linux driver problem, then?
<PaulFertser>
Fridtjof: is that OWE working for you on macOS 13.5 specifically?
<Fridtjof>
no, 14.7.2. Again, I'm mostly wondering if this is somehow reproducible for anyone else (=> something's broken in general), or if it's just my machine
<Fridtjof>
(well, machine and/or asahi install. asahi fedora 41 in this case)
<monthebiff>
is there a way i can check if its the same thing on my hostel wifi network without tanking the entire network when i try and connect?
<PaulFertser>
I'm not sure I understand why chaos_princess says it's not "outdated firmware" because if it's taken from 13.5 then it's guaranteed to be outdated if Apple updated it since then.
<Fridtjof>
it's not outdated in the sense that you're not supposed to have any newer firmware on the linux side - i'm guessing for compatibility reasons?
<chaos_princess>
^
<PaulFertser>
Another question is whether brcmfmac in Linux supports OWE at all on any hardware.
___nick___ has quit []
<PaulFertser>
Is it known that brcmfmac firmware that Apple now ships is indeed incompatible with current Linux brcmfmac or is it just in the vein "we only support people who stick to versions we tested ourselves"?
___nick___ has joined #asahi
<PaulFertser>
Because brcmfmac is not really the same deal as all the other tricky firmware which Asahi Linux has to support for the special Apple hardware.
___nick___ has quit []
kdc has joined #asahi
___nick___ has joined #asahi
xsk8060 has quit [Quit: ZNC 1.8.2+deb2ubuntu0.1 - https://znc.in]
monthebiff has quit [Quit: leaving]
dextaa60 has joined #asahi
dextaa6 has quit [Ping timeout: 480 seconds]
dextaa60 is now known as dextaa6
Halian has quit [Quit: I'll be back.]
Moprius has quit [Quit: bye]
pb17 has quit [Ping timeout: 480 seconds]
Moprius has joined #asahi
pb17 has joined #asahi
Calandracas has quit [Remote host closed the connection]
Moprius has quit [Remote host closed the connection]
<cy8aer>
yes - it is reproducable (I hate pink...)
<cy8aer>
when changing the frame size.
<cy8aer>
I called it the same way like nautilus
<cy8aer>
(loupe)
darkapex has quit [Remote host closed the connection]
darkapex has joined #asahi
<jannau>
LIBGL_ALWAYS_SOFTWARE=1 doesn't work for vulkan, use VK_DRIVER_FILES=/usr/share/vulkan/icd.d/lvp_icd.aarch64.json (path for fedora asahi remix) instead. that fixes the loupe issue here
pb17 has joined #asahi
ceph__ has joined #asahi
bn_ has quit [Quit: Connection closed for inactivity]
<cy8aer>
yes I can confirm this.
<leio>
oops, I forgot in my command history I was benchmarking only llvmpipe and combined with GSK_RENDERED=ngl, sorry
<cy8aer>
but it does not fix the problem with mouse over images in nautilus
<cy8aer>
ngl works (of course)
<cy8aer>
Ah, I had a typo in my nautilus call (K_DRIVER_FILES): this does not make any artifacts:
ipatch has quit [Read error: Connection reset by peer]
ipatch has joined #asahi
Calandracas has joined #asahi
pb17 has quit [Ping timeout: 480 seconds]
roxfan2 is now known as roxfan
<leio>
jannau: cy8aer uses debian trixie
<leio>
cy8aer: do you use fractional scaling?
<cy8aer>
yes 125%
<leio>
can you check if the problem goes away with 100%
<cy8aer>
No, the same problem with 100%
<cy8aer>
According to Issue 7146: for me the artifacts occur on images (and with loupe too). 7146 looks like some "normal" part of the window.
<leio>
Yeah, I too think it's a different problem
<jannau>
reproducable here with mutter and without fractional scaling on fedora
<leio>
7146 is also affecting GL
<jannau>
and RGBA8888 rendering so a different problem
pb17 has joined #asahi
<leio>
note that gtk before the version used with GNOME 47 did not default to Vulkan renderer yet
<leio>
i.e, GNOME 47 associated GTK 4.16 started defaulting to Vulkan, while before (4.14) you would need GSK_RENDERER=vulkan to get it to use it, also 4.14 was a completely different old implementation, so distros didn't really enable it yet
<jannau>
is GDK_DEBUG=vulkan-validate new? seems to be unknown in fedora 41
<leio>
sorry, 4.14 had new vulkan renderer already, just wasn't default. 4.12 had an old one