IEEE Std 754:2019 pdf free download – IEEE Standard for Floating-PointArithmetic
2.1 Definitions
For the purposcs of this standard, the following terms and definitions apply.
applicable attribute: The value of an attribute governing a particular instance of execution of acomputational operation of this standard. Languages specify how the applicable attribute is determined.arithmetic format: A floating-point format that can be used to represcnt floating-point operands or resultsfor the operations of this standard.
attribute: An implicit parameter to operations of this standard，which a user might statically set in aprogramming language by specifying a constant value.The term attribute might refer to the parameter (asin””rounding-direction attribute”) or its value (as in“roundTowardZero attribute””).
basic format:One of five floating-point representations, three binary and two decimal, whose encodingsare specified by this standard, and which can be used for arithmetic.One or more of the basic formats isimplemented in any conforming implementation.
biased exponent: The sum of the exponent and a constant (bias) chosen to make the biased exponent’srange non-negative.
binary floating-point number: A floating-point number with radix two.
block: A language-defined syntactic unit for which a user can specify attributes. Language standards mightprovide means for users to specify attributes for blocks of varying scopes, even as large as an entireprogram and as small as a single operation.
canonical encoding: A preferred encoding of a floating-point representation in a format.“Canonicalencoding” also applies to declets, significands of finite numbers, infinities, and NaNs, especially in decimalformats.
cohort: The set of all floating-point representations that represent a given floating-point number in a givenfloating-point format. In this context—O and +0 are considered distinct and are in different cohorts.
computational operation: An operation that produces floating-point results or that might signal floating-point exceptions.Computational operations produce results in floating-point or other destination formatsby rounding them to fit if necessary.
correct rounding: This standard’s method of converting an infinitely precise result to a floating-pointnumber, as determined by the applicable rounding direction. A floating-point number so obtained is said tobe correctly rounded.
decimal floating-point number: A floating-point number with radix ten.
declet: An encoding of three decimal digits into ten bits using the densely packed decimal encodingscheme. Computational operations accept all 1024 possible declets in operands.Most computationaloperations produce only the 1000 canonical declets.
denormalized number: See: subnormal number.
destination: The location for the result of an operation upon one or more operands. A destination might beeither explicitly designated by the user or implicitly supplied by the system (for example,intermediateresults in subexpressions or arguments for procedures). Even though some languages place the results ofintermediate calculations in destinations beyond the user’s control, this standard defines the result of anoperation in terms of that destination’s format and the operands’values.
dynamic mode: An optional method of dynamically setting attributes by means of operations of thisstandard to set, test, save, and restore them.
exception: An event that occurs when an operation on some particular operands has no outcome suitablefor every reasonable application. That operation might signal an exception by invoking default exceptionhandling or alternate exception handling.Exception handling might signal further exceptions.Recognizethat event, exception , and signal are defined in diverse ways in diferent programming environments.