Hi, * Chow Loong Jin <[email protected]> [2026-08-25 10:40]:
From my testing it looks like there are just two variants of hyprland binaries that gets produced, rather than randomness that does not resolve -- clicking rebuild once or twice makes it succeed.
I did two test builds locally and got different results in: build/CMakeFiles/hyprland_lib.dir/src/managers/fullscreen/FullscreenController.cpp.o build/CMakeFiles/hyprland_lib.dir/src/plugins/PluginAPI.cpp.o build/CMakeFiles/hyprland_lib.dir/src/protocols/XDGShell.cpp.oFor all of them the start of section headers, as reported by readelf -a was different, as well as all addresses based on that. I tried rerunning the gcc commands for those but all produced the same result and building the package again also produced the same result. I also did not find anything relevant that could explain the different offset so currently I am stuck here.
Do you have any idea how to trigger different results or why the offset would be different?
Cheers Jochen
signature.asc
Description: PGP signature
