Scalar Operations

Overview

Intrinsics to operate on floating point/integer scalar values, configuration, conversions, addressing, locks and events.

Note

For floating point scalar operations like add, subtract, multiply and compare, these can either be mapped on 'vector floating point' or on 'softfloat lib'

- by Default, softfloat lib implementation will be chosen which takes quite a few cycles since its emulated.

- AIE_OPTION_SCALAR_FLOAT_ON_VECTOR flag can be used in application (pre-processor macro) to map these operation on vector floating point datapath to reduce number of cycles.

For floating point non-linear scalar operations like sine/cosine, sqrt and inv, these can either be mapped on 'scalar non-linear unction' or on 'runtime lib (math.c)'

- by Default, runtime lib implementation will be chosen which takes quite a few cycles since its emulated.

- AIE_OPTION_SINF_COSF_FAST flag can be used in application (pre-processor macro) to map these operations on scalar non-linear functions to reduce number of cycles.

Modules
	AIE ID
	Addressing Operations
	Cycle Counter
	Datatype Conversions
	Support for converting floating-point numbers to fixed-point and fixed-point numbers to floating-point.
	Delayed Scalar Moves
	Intrinsic that returns its input after 6 clock cycles. Used for scheduling optimization.
	Elementary Functions
	Events
	Integer Operations
	Locks