ChanServ changed the topic of #wayland to: https://wayland.freedesktop.org | Discussion about the Wayland protocol and its implementations, plus libinput | register your nick to speak
columbarius has joined #wayland
co1umbarius has quit [Ping timeout: 480 seconds]
CodeSpelunker has joined #wayland
duxsco has quit [Remote host closed the connection]
duxsco has joined #wayland
lxsameer3 has quit [Ping timeout: 480 seconds]
nerdopolis has joined #wayland
CodeSpelunker has quit [Ping timeout: 480 seconds]
CodeSpelunker has joined #wayland
ofourdan_ has joined #wayland
ofourdan has quit [Ping timeout: 480 seconds]
duxsco has quit [Quit: duxsco]
nerdopolis_ has joined #wayland
nerdopolis has quit [Read error: Connection reset by peer]
duxsco has joined #wayland
duxsco has quit [Quit: duxsco]
nerdopolis_ has quit [Ping timeout: 480 seconds]
CodeSpelunker has quit [Quit: CodeSpelunker]
shankaru has joined #wayland
dcz has joined #wayland
zebrag has quit [Quit: Konversation terminated!]
naveenk2 has joined #wayland
ecloud_ has quit [Ping timeout: 480 seconds]
rv1sr has joined #wayland
ecloud_ has joined #wayland
hardening has joined #wayland
mvlad has joined #wayland
maxzor has quit [Remote host closed the connection]
maxzor has joined #wayland
neonking has quit []
neonking has joined #wayland
___nick___ has joined #wayland
slattann has joined #wayland
<slattann>
Test MEsg
___nick___ has quit []
___nick___ has joined #wayland
___nick___ has quit []
___nick___ has joined #wayland
lxsameer3 has joined #wayland
danieldg has quit [Ping timeout: 480 seconds]
danieldg has joined #wayland
dcz has quit [Ping timeout: 480 seconds]
shankaru has quit [Quit: Leaving.]
shankaru has joined #wayland
<mceier>
is it possible for compositors to handle unresponsive clients gracefully (don't close the wayland connection unless user says so) or would that require fundamental wayland libraries redesign ?
<kennylevinsen>
mceier: unresponsive means that they're not reading from their socket, at which point killing them is the only option. Discussion has been made to increase the buffer size and/or stop sending certain messages when the buffer fills past a certain threshold but nothing has materialized.
<slattann>
anyone tried compiling weston recently on Ubuntu21.04 ? when I tried hitting this dependency and could not find the package:
<slattann>
Run-time dependency libseat found: NO (tried pkgconfig and cmake)
<slattann>
libweston/meson.build:179:1: ERROR: Dependency "libseat" not found, tried pkgconfig and cmake
<kennylevinsen>
In those cases the point was also not to give the user a choice as much as just making it less likely that a well-behaved application ends up unresponsive.
<mceier>
kennylevinsen: yeah, unresponsive also means they can't read their socket because another process is saturating I/O
<kennylevinsen>
slattann: you either need libseat installed or need to disable the libseat launcher. See the list of meson options.
<mceier>
kennylevinsen: I was wondering if compositors could do anything about it with current design
<kennylevinsen>
mceier: how would "saturating I/O" matter for a Unix domain socket?
<slattann>
kennylevinsen: how can I disable it since I am not going to use libseat launcher ?
<kennylevinsen>
They can't really do anything right now, and the priority is not a user prompt but rather sorting out unfair kills where a client is bombed by high-speed input events and the likes.
<kennylevinsen>
slattann: check the meson options file for available options
<mceier>
kennylevinsen: well, it somehow matters... when firefox or alacritty (terminal) is dying with message "Gdk-Message: 09:08:12.409: Lost connection to Wayland compositor." due to mouse movement while steam is starting up ;)
Guest2190 has joined #wayland
<mceier>
while steam or during llvm checkout
<kennylevinsen>
The only time I/O would matter is if you're out of memory and swapping, or if a client is doing blocking I/O on its main thread (which it should not do)
<mceier>
so compositors can't do anything about it ?
<kennylevinsen>
But maybe you're hitting a combination of a bad client + high-speed input (if you have a high resolution mouse)
<kennylevinsen>
Not until something comes out of the discussion about this exact issue
<kennylevinsen>
I think we have an issue open about it?