16:21:36Krystofoh, look, I left myself a todo comment about it
16:21:54KrystofI also cunningly left in commented out ancient implementations of traced methods to confuse myself mightily
16:22:09mfianoHello. I think I found a bug, or at least an inconsistency with previous SBCL versions
16:24:18mfianoThe release notes say that CONSTANTLY has been made more efficient when given a constant argument, and indeed, it no longer prints as a closure object. However, inspecting it still shows it is closing over the value NIL. The problem is, ALL closures are now printed as #<FUNCTION ...> rather than #<CLOSURE ...>. I just wasted 20 minutes because something I thought was not a closure was not obvious