I was recently trying to test some micropython on my laptop that's targeted against an ESP8266. I used the framebuf module, but found it wasn't included in my default installation of micropython. I found that it was enabled in the "coverage" build, but I didn't want it significantly different from the standard build, I just wanted the framebuf module to be present. So I looked for configuration options and found mpconfigport.mk and decided to add MICROPY_PY_FRAMEBUF = 1 to it, but that didn't seem to change anything. Compiling it with V=1 showed that the Makefile wasn't passing it as a compilation flag, so whilst make knew it was defined, the C code did not.

It seems like have all configuration options being configurable in the same place (where a distribution can just add a line to the .mk file, rather than writing a define into the middle of a .h file) would make sense? As best I can tell, the Makefile needs to look for the variable, and if present add -DMICROPY_PY_FRAMEBUF=1. It doesn't feel like a big fix, but please let me know if you need more information or don't have time to handle it and I can try and mock up a pull request...