[patch, libgfortran] Some more transformational intrinsic speedups

[FX]

>  Without looking at the details of Thomas' patch I think intptr_t is the
> right type, i.e. an integer type that is defined to always be wide enough to
> contain a pointer.

uintptr_t (the unsigned type) seems more appropriate here. But are we
guaranteed it is defined? When I used intptr_t in the past (in
intrinsics/signal.c and runtime/backtrace.c) I've defined a macro
depending on HAVE_INTPTR_T.

FX

-- 
FX Coudert
http://www.homepages.ucl.ac.uk/~uccafco/