Closed Ekdohibs closed 2 weeks ago
(The precondition for this bug cannot be satisfied with the current compiler.)
Does this need a fix for runtime5?
I think this doesn't need a fix for runtime5, but I just noticed extern.c
might need a fix as well; I'll look.
Compaction in runtime5 has already been merged, and works differently to runtime4 so doesn't need to be changed.
If function slots were not consecutive, the GC would previously incorrectly scan them during compaction while trying to revert infix pointers. Note: due to an invariant in slot_offsets, this can't happen with the existing compiler.