On Sun, Jun 2, 2013 at 11:28 PM, Eric Marsden <eric.marsden@free.fr> wrote:
Thanks. I'll note that it doesn't do this rewrite for interpreted code, which leads to different results for interpreted and compiled code (probably is allowed by CLtS 3.2.2.3).

The interpreter cannot do the optimizations that the compiler does. Also, the same function may or may not return the same value under different optimizations or even TYPEP need not return the same value on every invocation.

Juanjo

--
Instituto de Física Fundamental, CSIC
c/ Serrano, 113b, Madrid 28006 (Spain) 
http://juanjose.garciaripoll.googlepages.com