I think I’ve spotted an undocumented error that can be generated when calling glTexStorage2D(). Namely:
GL_INVALID_VALUE is generated if target is GL_TEXTURE_CUBE_MAP and width does not equal height.
The error does not appear to be documented for glTexImage2D() either.
The docs that I’m using are these https://www.opengl.org/sdk/docs/man/
I’d like to point out that I am aware that this might not be the most appropriate forum for this thread, but as a newbie I cannot post in the suggestions forum and I haven’t found another way to report an issue like this.
I think this must be a bug. Every resource I’ve found says that those values must match the width and height of the image, and those can be different, of course.
I’m using an AMD Radeon HD 8600M. Has anyone encountered a similar issue?
I finally got my cube map to render loading each image separately and using glTexImage2D() with the dimensions of each one of them. I was trying to load them all into the same buffer and copy them to OpenGL moving the pointer, but I was failing. Some of the resources I’ve found around say that this is the way to do it, but I ran into the problem described above (be it an omission in the docs or a driver bug).
On the page you reference above, there is a section “Feedback” which includes instructions on how to report these sorts of things.
Thanks, I’ll do that. I think I missed the feedback section because I accessed the doc page via Google. If you skip the initial page you miss the feedback notice.