I need to build fluidsynth as a standalone Universal dylib for use in my app distributions in and apart from the Mac App Store. As I understand it, this will require building fluidsynth from source on two different machines – one Apple silicon, the other Intel – then combining the libraries. Afterward, all the dependencies must be rerouted one by one internally, in order to make the dylib stand-alone distributable within an app. That last step is something I’ve done before for x86 dylibs, but I’ve never had to do it for two different architectures and I don’t know if something will be different at that step. I’ve never built fluidsynth from source, so I’m looking for help. Is anyone interested in helping?
You can usually cross compile on a Mac for both platforms.
See -arch parameter in clang/gcc.
That’s true, but there are a lot of potential problems because of Rosetta. I read about problems others were having trying to compile both on one machine, and I also had trouble builing a different dylib for both architectures. I had to do strange things like start terminal running in Rosetta. I don’t want to have to worry about that kind of tricky business, so I consider the best route to use separate machines having the right chip to compile.