ChanServ changed the topic of #wayland to: https://wayland.freedesktop.org | Discussion about the Wayland protocol and its implementations, plus libinput
Moprius has joined #wayland
<bluetail>
ty, I won't upgrade then
<bluetail>
I am on 6.11.5-arch-1-1 at the moment
<bluetail>
cause stuff is still working :)
<dviola>
looks like only QEMU virgl is affected
<dviola>
np
okt has joined #wayland
Moprius has quit [Quit: bye]
julio7359 has quit [Remote host closed the connection]
glennk has quit [Ping timeout: 480 seconds]
pramodvu has quit [Quit: Konversation terminated!]
pramodvu has joined #wayland
pramodvu has quit [Quit: Konversation terminated!]
pramodvu has joined #wayland
ke8cqc has joined #wayland
Brainium has joined #wayland
<ke8cqc>
I'm a long time Gentoo LInux user who has only used xorg. I would like to try Wayland. Through googling, I have found out that xorg and wayland can run concurrently on the same machine, so I've installed wayland and hyprland. Is there a way to disable xorg and use wayland without nuking my xorg config?
<whot1>
ke8cqc: if you're running wayland the X server you're running will be Xwayland which is a translation layer between the wayland protocol and the X protocol. You're not actually running /usr/bin/Xorg anymore. Xwayland doesn't read your config so you can leave it in place
<ke8cqc>
I guess I'm confused here. What's running on my machine is xorg using bspwm right now. I have installed wayland, xwayland, and hyprland. What I don't understand is how to set up my laptop to use wayland and hyprland.
<whot1>
ke8cqc: you tell whatever starts xorg on your machine to start your compositor instead
<whot1>
in gdm that's a button to change between GNOME on Xorg and GNOME on Wayland, other display managers will have their own configuration
<ke8cqc>
I'm currently using ligjhtdm. I notice that hyprland shows up in the menu of available window managers, but when I choose it, all I get is a blinking cursor in the upper left corner of my screen after loging in.
<ke8cqc>
*lightdm.
<whot1>
ke8cqc: check if there's anyhing in the journal to tell you what's going wrong
<ke8cqc>
I'm using OpenRC - not SystemD.
Moprius has joined #wayland
pramodvu has quit [Remote host closed the connection]
ke8cqc has quit [Quit: Page closed]
Brainium has quit [Quit: Konversation terminated!]
<jadahl>
vyivel: emersion: should be fixed now. thought I had already done it, sorry about that
<vyivel>
thanks!
<jadahl>
np
<emersion>
ty!
rasterman has joined #wayland
<pq>
mclasen, the question is more about the level of support. There is new protocol after v4 still. I also haven't looked carefully enough whether all implementations together cover all the features.
<jadahl>
pq: I think one missing thing is an implementation that has both icc support and hdr support that mixes the two (monitor in HDR mode + ICC profile client)
<pq>
OTOH, I cannot think how that might change the protocol design, but of course it's better to see than guess.
<jadahl>
right, we have an idea for how it will work with nothing new needed in the protocol
<jadahl>
roughly it's creating an icc profile for the intermediate buffer color state in standard range, and then do standard to high dynamic range after doing an icc->icc conversion
<pq>
That's a very good idea.
<pq>
I was slightly worried by the MR description which seems to say otherwise.
<jadahl>
right, it isn't doing that yet
<pq>
I've been thinking of interfacing with the ICC PCS in the "parametric" handling side, but crafting an ICC profile and letting a CMM handle it feels better.
<pq>
well... crafting that ICC profile is equivalent, and probably on par with difficulty.
<pq>
I think an important aspect would be to ensure tat
<pq>
*that when an ICC profile and a parametric image description are fully equal, they also result in equivalent color transformations for all kinds of image descriptions as the other side.
<pq>
*on the other side
<pq>
jadahl, is there a reason to start with explicit shaper-matrix ICC profile parsing when crafting the intermediate ICC profile would make that moot? Do you want a special case that works without a 3D LUT?
<jadahl>
I guess the downside of doing parametric via ICC is you only get luts from lcms2?
<pq>
ICC files can contains 3D LUTs which you cannot represent with parametric.
<jadahl>
I mean interfacing parametric via ICC profiles, but maybe I misunderstood you
<pq>
Crafting an ICC profile for the intermediate space works for everything. Transforming an ICC profile into parametric representation is limited to certain types of profiles only.
<jadahl>
ah, yes, I don't think we'd start with special casing ICC profiles when thats what the client hands us, but I'm not sure
<pq>
the commit messages certainly sound like they special-case shaper-matrix profiles
<pq>
An ICC profile can have different AToB/BToA/DToB/DToA tags for colorimetric, saturation, and perceptual rendering intents, so you also need to know the rendering intent when using a profile.
<pq>
jadahl, I can see the special-casing code already there.
<jadahl>
pq: the MR is from a state before the intended ICC->ICC(SDR)->intermediate(HDR) approach was described to Joan