Yeah the TCO is not complete, it allocates a call frame for each call regardless which is why it ran out of memory. It doesn’t use up the C call stack though, s

Yeah the TCO is not complete, it allocates a call frame for each call regardless which is why it ran out of memory. It doesn’t use up the C call stack though, so you could in theory increase the GC heap size to make it survive longer. Due to the primitive environment implementation it’ll get slower and slower as it goes. Thanks for the note on Mac OS, I’ll fix that.