ChanServ changed the topic of #wayland to: https://wayland.freedesktop.org | Discussion about the Wayland protocol and its implementations, plus libinput
feaneron has quit [Ping timeout: 480 seconds]
<whot>
emersion: recommendation is that if a user enables drag-locking it should be enabled as sticky because that's (probably) the more intuitive behavior
<whot>
suggested default for drag lock: off, suggested default behavior if drag lock is on: sticky
iomari891 has quit [Ping timeout: 480 seconds]
Lyude has quit [Read error: Connection reset by peer]
Lyude has joined #wayland
garnacho has quit [Ping timeout: 480 seconds]
Moprius has quit [Remote host closed the connection]
feaneron has joined #wayland
sally has joined #wayland
sally has quit []
the_sea_peoples has quit [Ping timeout: 480 seconds]
sally has joined #wayland
nerdopolis has quit [Ping timeout: 480 seconds]
coldfeet has joined #wayland
karenw has quit [Ping timeout: 480 seconds]
glennk has joined #wayland
coldfeet has quit [Quit: Lost terminal]
feaneron has quit [Ping timeout: 480 seconds]
eluks has quit [Remote host closed the connection]
chiku has quit [Read error: Connection reset by peer]
eluks has joined #wayland
Sid127 has joined #wayland
kts has joined #wayland
kts has quit [Ping timeout: 480 seconds]
bim9262_ has joined #wayland
AJ_Z0 has quit [Ping timeout: 480 seconds]
bim9262 has quit [Ping timeout: 480 seconds]
bim9262_ is now known as bim9262
kts has joined #wayland
sima has joined #wayland
kts has quit [Ping timeout: 480 seconds]
cptaffe has quit [Read error: Connection reset by peer]
cptaffe has joined #wayland
tzimmermann has joined #wayland
<emersion>
whot: thanks!
austriancoder has quit [Read error: Connection reset by peer]
Ariadne has quit [Read error: Connection reset by peer]
snek has quit [Write error: connection closed]
snek has joined #wayland
austriancoder has joined #wayland
Ariadne has joined #wayland
garnacho has joined #wayland
Ariadne has quit [Read error: Connection reset by peer]
Ariadne has joined #wayland
<MrCooper>
zamundaaa[m]: with VRR on, does KWin still trigger a scanout cycle ASAP on cursor movement?
<TimRex>
Been trying to diagnose an issue i'm having with a wayland app I'm building... and stumbling up against the following
<TimRex>
```Display Queue} wl_display#1.error(wp_linux_drm_syncobj_surface_v1#93, 3, "Release or Acquire point set but no buffer attached")```
<TimRex>
From a userspace application perspective... is it possible this is programmer error? Or is this more likely to be an issue with the compositor and/or drivers?
<TimRex>
mutter/gnome FWIW... Arch Linux... Can not repro on Fedora at all.. Just not sure how confident I can be that this isn't an issue of my own making
<kennylevinsen>
TimRex: the Wayland server is telling the client that it violated the syncobj protocol by calling `set_acquire_point` or `set_release_point` in a surface commit that did *not* also attach a buffer, which is invalid.
<kennylevinsen>
I'll by default assume that the server is in the right to send that message and didn't make a mistake. You can read the rest of the WAYLAND_DEBUG=1 output of the client to see if indeed, a `wl_surface_commit` call was preceded by set_release_point/set_acquire_point but *not* wl_surface_attach
<MrCooper>
usually this happens due to multiple entities fighting over setting surface state / calling wl_surface_commit
<TimRex>
Ok, Thanks Kenny.. I don't have any explicit calls that match `set_acquire_point` or `set_release_point` but I can see mention of those in the output with WAYLAND_DEBUG enabled..
<MrCooper>
is this using the nvidia driver by any chance?
<TimRex>
For once, no :)
<TimRex>
amdgpu
<TimRex>
I don't encounter the issue with llvmpipe FWIW
<jadahl>
TimRex: is your app multi threaded?
<kennylevinsen>
The protocol that your app trips over is for synchronizing access to GPU buffers, so it makes sense that CPU rendering with llvmpipe would not reproduce :)
<TimRex>
It *is* multithreaded, though I _believe_ all rendering and window management occurs from a single thread
<TimRex>
Thanks all, I think that gives me some avenues to chase... the issue only occurs when i'm resizing the app (and even then, sporadic), and absolutely only when using my own client-side decorators.. it's clear the issue is stemming from (or triggered by) that code somewhere..
<MrCooper>
this might indicate otherwise; IIRC current libwayland can print thread IDs in the debug logs, don't remember how to enable that offhand though
<TimRex>
I don't have the same issue when using CSD's provided by `libdecor` so that's something
<davidre>
So it's happening with SSD a from libdecor? Maybe libdecors calls commit when it shouldnt
<TimRex>
No, sorry.. does *not* occur with libdecor or with SSD
<TimRex>
Only with my own custom CSD
<TimRex>
So clearly, that's where the fault lies :)
<TimRex>
AFK for a moment.. dog needs to go out
<davidre>
I would test the standard CSD from libdecor and see what happens
<davidre>
And if it doesnt have any problem stare at both to find the difference
<TimRex>
Standard CSD from libdecor seems fine.. I'll see what I can glean from it.. I am still mildly curious that I can't repro it in Fedora (arch only).. but that isn't to say this isn't still my issue
<MrCooper>
same Mesa version in both cases?
<jadahl>
libdecor never ever touches the apps surface itself
fmuellner has joined #wayland
<zamundaaa[m]>
[@_oftc_MrCooper:matrix.org](https://matrix.to/#/@_oftc_MrCooper:matrix.org): not necessarily. If the focused window is updating fast enough, cursor updates are delayed until the next primary plane update
<MrCooper>
right, but if no new buffer for the primary plane is ready when the previous commit completes, a cursor-only commit is submitted ASAP?
Sid127- has quit [Read error: Connection reset by peer]
chiku has joined #wayland
<zamundaaa[m]>
No
<zamundaaa[m]>
"fast enough" = 30fps or more
<zamundaaa[m]>
Or whatever the minimum vrr refresh rate is, if we know it
<MrCooper>
I see, cool, thanks
<MrCooper>
that's a neat idea
Sid127- has joined #wayland
chiku has quit [Read error: Connection reset by peer]
leonanavi has joined #wayland
leon-anavi has quit [Remote host closed the connection]
Calandracas has quit [Remote host closed the connection]
bitblt has quit [Remote host closed the connection]
Calandracas has joined #wayland
bitblt has joined #wayland
bitblt_ has joined #wayland
Feijstag has joined #wayland
AJ_Z0 has joined #wayland
bitblt has quit [Ping timeout: 480 seconds]
fatal has quit [Remote host closed the connection]
fatal has joined #wayland
bitblt_ is now known as bitblt
narodnik has quit [Quit: WeeChat 4.6.3]
narodnik has joined #wayland
fatal has quit [Remote host closed the connection]