Is there a way to check at runtime whether the GPU supports #include?
It looks like some AMD GPUs may not support it, but recent-ish NVIDIA ones seem to.
My RayTK library currently depends on using #include when it constructs its shaders. Ideally I’d like to continue using it when it’s supported. When it isn’t supported I can inline those other DATs, but if possible I’d prefer to only do that when it’s necessary.
Or should we not rely on #include support for widely used libraries?
Nevermind, I’ve confirmed it’s not supported yet. I stand corrected. Seems like the support they said was added only made it into their linux drivers. I would say for now you need to avoid #include statements for anything you want to release publicly. Working on getting this fixed.
Any updates on this? I’ve had quite a few of GP users over the last year who report this on similar gpu hardware. Do you know if there is a ln expectation this will eventually be supported?
Well, the good news is that it’s fully supported on the 2021.30000 series since I’m now in control of the GLSL compiler. No more vendor specific bugs there.
I’ve reached out to AMD to see if they fixed it in their OpenGL drivers
! Wow what another awesome piece of good news for .30000
I noticed some odd stuff with the render top btw, is it too early to start posting bugs? I assume a lot of them are known.