> If it's something that depends on the order of static initialisation across files, > then really it's a legitimate bug that needs to be fixed. But I don't quite > understand the issue as you described, the text meandered, so I'm not sure what > you're saying.
The stock colors are initialized after global statics are constructed, so this:
static const rgb_t colors[] = { rgb_t::white };
will yield a random (usually black) color, unless it's inside a function's scope. This has been a known thing for at least 2 years now. The stock colors really should be #defines or something to prevent this. I had to change the Apple II drivers to just use hex RGB values instead.
Edited by R. Belmont (09/07/16 03:19 PM)
|