Ok, I installed gambit from master, ran the compile-all script with "gsi" but it has been running for an hour or so. Trying your gsi command above also seems t

Ok, I installed gambit from master, ran the compile-all script with "gsi" but it has been running for an hour or so. Trying your gsi command above also seems to take awfully long. I tried building rsc.scm with gsc and that made an executable that let me create repl-min.scm.c that was invalid C because of two definitions of "input". I commented out one of them and was able to compile a REPL that could do (+ 2 2) but segfaulted when I tried making a user-defined function and calling it. Something similar happened with the web demo on github. I see that the extra "input" definition in the .c file is in an #ifndef next to some clang pragmas: does it expect to be compiled with clang? I'm using gcc 8.3. I got rsc working with python after a few false starts, but that also crashes (with a python backtrace instead of a segfault) when I try to define and use a function. Also have similar crashes when the scheme output of rsc, under both gsi and guile.