Darío
Darío
> Oof some recent merges just conflicted this PR. I'm still testing this, but this already needs a rebase xD You mean like in the middle of the hour I...
@Riteo Alright, conflicts should be fixed now.
> I tried this branch briefly with the Wayland backend on my crappy laptop and I couldn't notice any particular artifacts on the [Garage 42 demo](https://godotengine.org/asset-library/asset/2379) (I'll test properly on...
> There's a segfault on exit though: > > ``` > Thread 1 "godot.linuxbsd." received signal SIGILL, Illegal instruction. > 0x000055555f098bae in LocalVector::operator[] (this=0x7fffef3aac68, p_index=1) at ./core/templates/local_vector.h:161 > 161 CRASH_BAD_UNSIGNED_INDEX(p_index,...
Rebased to fix conflicts.
I'm not sure I can add much here, this is a very typical issue that crops up with seams if you do not map the UV2 to be a contiguous...
> * `driver_clear_resources_with_views` and the other bool member above could become bitfields. Admittedly the purpose of these is just to store one flag and be consistent throughout the execution of...
> I believe anyone hoping to understand this area of the engine is aware of bitfields. It's not a great saving, but why missing the opportunity of packing a few...
> I'm not sure my feedback (the directly actionable part) has been addressed yet. Would you say this covers the copy engine suggestion and the bitfields? I feel the arena...
> I believe anyone hoping to understand this area of the engine is aware of bitfields. It's not a great saving, but why missing the opportunity of packing a few...