<ifreund>
LarstiQ: no, it's mainly me finishing some work on my website and writing a blog post
<ifreund>
I don't think I'll solve 999 before 0.3.0 unless someone motivates me to work on it with a non-proprietary reproducer
<dnkl>
ifreund: I got a crash today, caused by wlfreerdp 3.4. It's not deterministic, but I could probably get you a backtrace, if you're interested. wlfreerdp is deprecated, so it's not worth the effort for its sake. But a crash is a crash... so might be something you want to fix anyway
<ifreund>
dnkl: yes, I'm always happy to get river stacktraces, river should ideally never crash no matter what clients do
<dnkl>
thought so :)
<ifreund>
if you haven't updated in a while there have been a couple of recent fixes too fwiw
<dnkl>
I'm almost always on HEAD, or just a few commits away
<ifreund>
cool
<dnkl>
got it. It's an assert in wlroots, so guessing it's not really an issue with river
<dnkl>
the client is using a bad viewport. Most of the times, this just results in a protocol error. But occasionally we crash instead
<ifreund>
hmm, what wlroots version do you have?
<dnkl>
0.17.2
<ifreund>
I ask because there were some similar sounding wlroots bugs that caused assertion failures fixed in 0.17.1, but it seem you already have those
<dnkl>
can't post the backtrace right now, but its in wlr_render_pass_add_texture
<dnkl>
could be. I'll see if I can trigger it again, with WAYLAND_DEBUG
<dnkl>
I'll post all logs later, but I suspect river is the one causing the crash, in a way. The client appears to have been properly terminated with a protocol error, and *then* river crashes
<ifreund>
ah, yeah just open an issue with the logs then when you have a chance and I'll take a look :)
<LarstiQ>
ifreund: check. I'm eagerly looking forward to it. 0.2.6 was taking 80% cpu this morning when I tried displaylink
<ifreund>
I mean, if you don't mind compiling master branch it is pretty much exactly what 0.3.0 will be now
<ifreund>
0.2.6 taking 80% cpu sounds like you were using software rendering though
<ifreund>
river log messages would say for sure
<LarstiQ>
ifreund: I'm still on Asahi aarch64-linux with some master-building issues
<ifreund>
I'm confused, why would 0.2.6 be easier to build?
<ifreund>
the standard library has always had a static, comptime known page size
<ifreund>
also, you could build zig from source and set that page size to 16k for linux as a workaround, it would be a 1 line patch
novakane has quit [Read error: Connection reset by peer]
andrea has quit [Read error: Connection reset by peer]
geemili has quit [Read error: Connection reset by peer]
dzoidberg has quit [Write error: Connection reset by peer]
kennylevinsen has quit [Read error: Connection reset by peer]
leon-p has quit [Read error: Connection reset by peer]
mainiomano has quit [Read error: Connection reset by peer]
dnkl has quit [Read error: Connection reset by peer]
pepe has quit [Read error: Connection reset by peer]
kindablue has quit [Write error: Connection reset by peer]
voroskoi has quit [Read error: Connection reset by peer]
Ankhers has quit [Read error: Connection reset by peer]
Ronan-Dplq_ has quit [Read error: Connection reset by peer]
anjan has quit [Read error: Connection reset by peer]
ane has quit [Read error: Connection reset by peer]
bfiedler has quit [Read error: Connection reset by peer]
whereswaldon has quit [Write error: Connection reset by peer]
szgy has quit [Write error: Connection reset by peer]
gbrlsnchs has quit [Write error: Connection reset by peer]
lizog has quit [Read error: Connection reset by peer]
selenebun has quit [Read error: Connection reset by peer]
pvsr has quit [Read error: Connection reset by peer]
rodrgz has quit [Read error: Connection reset by peer]
p00f has quit [Write error: Connection reset by peer]
ptrckd has quit [Remote host closed the connection]
coder_kalyan has quit [Read error: Connection reset by peer]
raiaq has quit [Write error: Connection reset by peer]
ifreund has quit [Write error: Connection reset by peer]
greenfork has quit [Read error: Connection reset by peer]
psnszsn has quit [Write error: Connection reset by peer]
maringuu has quit [Write error: Connection reset by peer]
bfiedler has joined #river
lizog has joined #river
Ronan-Dplq has joined #river
geemili has joined #river
coder_kalyan has joined #river
szgy has joined #river
gbrlsnchs has joined #river
ifreund has joined #river
kindablue has joined #river
Ankhers has joined #river
p00f has joined #river
maringuu has joined #river
whereswaldon has joined #river
greenfork has joined #river
anjan has joined #river
dnkl has joined #river
ptrckd has joined #river
kennylevinsen has joined #river
andrea has joined #river
selenebun has joined #river
mainiomano has joined #river
pvsr has joined #river
leon-p has joined #river
rodrgz has joined #river
novakane has joined #river
ane has joined #river
dzoidberg has joined #river
voroskoi has joined #river
raiaq has joined #river
psnszsn has joined #river
pepe has joined #river
anjan has quit [Remote host closed the connection]
greenfork has quit [Remote host closed the connection]
ptrckd has quit [Remote host closed the connection]
kindablue has quit [Remote host closed the connection]
ifreund has quit [Remote host closed the connection]
mainiomano has quit [Remote host closed the connection]
ane has quit [Remote host closed the connection]
maringuu has quit [Remote host closed the connection]
kennylevinsen has quit [Remote host closed the connection]
szgy has quit [Remote host closed the connection]
geemili has quit [Remote host closed the connection]
Ronan-Dplq has quit [Remote host closed the connection]
lizog has quit [Remote host closed the connection]
gbrlsnchs has quit [Remote host closed the connection]
andrea has quit [Remote host closed the connection]
pepe has quit [Remote host closed the connection]
raiaq has quit [Remote host closed the connection]
p00f has quit [Remote host closed the connection]
Ankhers has quit [Remote host closed the connection]
coder_kalyan has quit [Remote host closed the connection]
bfiedler has quit [Remote host closed the connection]
leon-p has quit [Remote host closed the connection]
psnszsn has quit [Remote host closed the connection]
rodrgz has quit [Remote host closed the connection]
voroskoi has quit [Remote host closed the connection]
dnkl has quit [Remote host closed the connection]
selenebun has quit [Remote host closed the connection]
whereswaldon has quit [Remote host closed the connection]
pvsr has quit [Remote host closed the connection]
dzoidberg has quit [Remote host closed the connection]
novakane has quit [Remote host closed the connection]
lizog has joined #river
Ronan-Dplq has joined #river
bfiedler has joined #river
geemili has joined #river
voroskoi has joined #river
pvsr has joined #river
Ankhers has joined #river
szgy has joined #river
novakane has joined #river
ifreund has joined #river
andrea has joined #river
p00f has joined #river
kennylevinsen has joined #river
dnkl has joined #river
kindablue has joined #river
ane has joined #river
mainiomano has joined #river
kennylevinsen has quit [Changing host]
rodrgz has joined #river
kennylevinsen has joined #river
leon-p has joined #river
psnszsn has joined #river
selenebun has joined #river
greenfork has joined #river
maringuu has joined #river
dzoidberg has joined #river
whereswaldon has joined #river
gbrlsnchs has joined #river
pepe has joined #river
raiaq has joined #river
anjan has joined #river
coder_kalyan has joined #river
ptrckd has joined #river
<LarstiQ>
ifreund: I don't build it, presumably the NixOS build machinery uses a non-Asahai aarch64 builder that produces output that happens to work
<LarstiQ>
ifreund: I've tried the page size patch it does produce a working binary but then can't find Mesa or something, haven't had the patience to dig into it
<ifreund>
if you're referring to that PR someone opened for runtime page size, I really don't trust their code quality
<ifreund>
applying a single-line patch to std/mem.zig to change the hardcoded aarch64 linux page to what ashai uses should be fine
<ifreund>
and much simpler, though obviously not upstreamable
<LarstiQ>
no I mean the single-line patch
<adamcstephens>
LarstiQ: the nixos build machines for aarch64-Linux are definitely not apple silicon
<ifreund>
LarstiQ: not being able to find mesa sounds like a nixos-induced problem tbh
pfr has joined #river
<LarstiQ>
ifreund: yes I know
<LarstiQ>
adamcstephens: cheers! The package they produce works fine enough luckily
<LarstiQ>
(the mesa issue comes from the asahi mesa overlay being different than the nixpkgs one, I'm sure it's solvable if I sit down for it)
<adamcstephens>
Using overlay or replace for mesa?
<LarstiQ>
replace atm
waleee has joined #river
Guest76 has joined #river
<Guest76>
Hello, does anyone know how I can enable dark mode for libadwaita apps? I've already set `org.gnome.desktop.interface color-scheme` to `prefer-dark` using gsettings, tried setting `ADW_DISABLE_PORTAL=1` and tried using xdg-desktop-portal-gtk instead of xdg-desktop-portal-wlr. GTK-4 apps are using dark mode by default but not those using libadwaita!
<ifreund>
sorry, I don't know anything about how libadwaita works personally
<FireFly>
I export GTK_THEME=Adwaita:dark and have $XDG_CONFIG_HOME/gtk-3.0/settings.ini as [Settings] \n gtk-application-prefer-dark-theme = true
<FireFly>
IIRC there's some fiddling around since they seem to change conventions between major versions, so Gtk 2, 3 and 4 want slightly different ways of telling it to use a dark colourscheme
<ifreund>
lovely
<ifreund>
Guest76: ^
<Guest76>
Well dark mode for "pure GTK" apps work but not for GTK apps using libadwaita :/
<FireFly>
do you have an example of such an app using libadwaita?
<FireFly>
I don't know much about it eithe, but I could at least test if it's light or dark for me
<FireFly>
at least gnome-chess and gnome-characters show up in dark for me
<Guest76>
Huh? `GTK_THEME=Adwaita:dark` seems to work?
<Guest76>
I was sure I tried that before.
<FireFly>
this came up when Dino made some change, I thought it was moving to Gtk4 but maybe it was incorporating libadwaita? either way one day it was suddenly Very Bright and it took some digging around until I got it dark again with GTK_THEME
<Guest76>
Maybe I also set `org.gnome.desktop.wm.preferences theme` to Adwaita:dark and it just ... broke and then I didn't try GTK_THEME. I can't really remember.
<Guest76>
The pains of using a tiling WM instead of GNOME ;(
<FireFly>
yeah I don't know either, I don't know gnome/gtk well
<Guest76>
I honestly like the libadwaita look and feel though. It's just that I also like dark mode.
<Guest76>
But it's great that it works now. Thanks for pointing out GTK_THEME.
Guest76 has quit [Quit: Client closed]
<leon-p>
I happen to know that coupling your applications dark/light setting with the corresponding dconf entry is something you have to do manually in GTK / libadwaita
<leon-p>
probably because by default, both are used (f.e. the image viewer is dark by default but all other applications are light on gnome)
<leon-p>
so that applications can implement their own preference logic
<adamcstephens>
LarstiQ: you may want to try using overlay then if you can afford all the rebuilds.