grr... bugs

So as I'm trying to write out the details of how the flat environments work, I had an idea on how to optimize top-level references. While I was working on that, I discovered problems in the mapping tables that are carried around in the partial environments. Now I'm debugging those.

Kbob asks:

In your profiling, have you measured the distribution of frame sizes?

Indeed I have. Here is a histogram of frame sizes after a 10 billion evaluation run:

Frame      
Size       Count

[0]      99656436
[1]     817178031
[2]     219585322
[3]      45556970       
[4]       6140170       
[5]       2857104       
[6]        702372       
[7]        448574       
[8]          3080       
[9]             1       
[10]          568       
[11]          156       
[12]            3       
[13]            2       
[14]          177       
[15]            6