Overloaded functions

Older libraries--without function overloading--have different versions of logarithm, depending on their type of expression evaluation:

• logf takes an argument converted to float and returns a float result. This function has appeared on platforms that, like PA-RISC, have full support for float operations. It would not appear on platforms like PowerPC and X86 that evaluate to double or long double only.
• log takes an argument converted to double (or long double, in the case of platforms like X86) and returns a double (or long double) result.
• logl takes a long double argument and returns a long double result. It would not appear on platforms like X86, where log suffices.

NOTE Overloaded numerical functions simplify porting code between CommonPoint platforms of diverse underlying architectures. Porting to or from other systems might require name changes beyond cosmetic differences in capitalization.

---