How do the library functions handle invalid (infinite, indefinite, etc.) floating-point values in input? There is an error code,ippStsPointAtInfinity, but I only find it referenced in the Error Codes section (no function seems to have it listed as a possible return value).
