So far the state of IuCS aka 3G voice is such that we will not merge the branch to master unless we have a proper A-interface.

But come to think of it, it would technically be possible to split the code not by git branches, but rather by the --enable-iu configure switch. In that case openbsc master would contain all IuCS code, and if compiled with --enable-iu, compile time switches would disable osmo-nitb, enable osmo-cscn and (currently) destroy 2G due to lack of an A-interface implementation.

The benefit would be simply to not have a separate branch, making for easier 3G build instructions and possibly reducing rebase merge conflicts; developers could adjust the 3G code paths along with their 2G patches.

It's just an idea that came to mind... In the end it's just a kind of "fake" merge, so not sure whether it's worth the effort.

Any thoughts?

~N