Broken Pipe Error Due to excessive spamming on Null Buffer attachment #11654
Unanswered
VimYoung
asked this question in
Bugs - Renderer
Replies: 2 comments
-
|
a WAYLAND_DEBUG log and a test client would help |
Beta Was this translation helpful? Give feedback.
0 replies
-
|
I am a bit busy so the work is stalled. I will provide a test case as soon as possible. |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
Hyprland version: 0.50.0
OS and Kernel: 6.16.4-artix1-1
Pre-context: I am building a client for wayland from scratch. It is a WIP widget system. It has a hide window feature. Internally the feature attaches Null to layer which unmaps the buffer and commits on calling
hidemethod and on calling backshow_againmethod, the previous buffer is reattached.Issue: On performance analysis, It was found that after calling hide and show a few times on a window. The event_queue is panicking with the following error.
The internal logs suggest that as soon as null is attached and commited, Hyprland spams a excessive stream of configure calls to the layer. On calling
show_again.The stream couldn't be processed at the same pace by the client and after a certain calls, the event_queue panics. Though I am not sure, there is a high possibility of it. I am open to corrections and suggestions.Expected Behavior: The recall on configure should be manageable enough by the client side of wayland to process.
Possible Solutions: I don't know the internal working of server side of wayland but following approaches seems more apt for the situation.
Here is the following video showing the same.
Config event is calledis printed on call to configuration of layer. You can observe the conclusion mentioned in the clip.screenrecord.mp4
Beta Was this translation helpful? Give feedback.
All reactions