Standard C math.h compatibility

• Use numerical functions from math.h whenever possible. When there is a corresponding function in Numerics.h, the functions are computationally identical on CommonPoint platforms.
• Use types float and double for long-term storage of floating-point data. Don't depend on the type long double (which in many cases is the same as double). Use of the types float_t and double_t is recommended, but you'll have to provide the definitions for each platform; they're not defined in math.h
• For CommonPoint capabilities beyond the Standard C libraries, use the definition from Numerics.h for the CommonPoint implementation and build a substitute implementation for other environments.
• Don't manipulate the environment, whose features as discussed in this documentation are tied to the IEEE floating-point standards. And don't use the Standard C errno mechanism to test for floating-point errors; it is not supported.
• Avoid FPCE constructs such as hexadecimal floating constants.
• Don't assume that the transcendental functions are as accurate as the CommonPoint implementation. But you can assume that they're monotonic where appropriate, that they remain within their mathematical range, and that they satisfy the usual mathematical identities to within roundoff.
• Avoid dependence on the byte order of numbers in memory. (The problem applies to all data, not just floating-point values.)
• Don't assume the platform subnormalizes underflowed values. It may set underflowed values to zero.