Developer Guide

FPGA Optimization Guide for Intel® oneAPI Toolkits

ID 767853
Date 7/13/2023
Public

A newer version of this document is available. Customers should click here to go to the newest version.

Document Table of Contents

Control Semantics of Floating-Point Operations (-fp-model=<value>)

Include the -fp-model=<value> option in your icpx command to direct the Intel® oneAPI DPC++/C++ Compiler to control the semantics of floating-point operations.

NOTE:

You can also use the Floating-Point Pragmas to influence the floating-point operations.

Example

icpx -fsycl -fintelfpga -Xshardware -fp-model=<value> <source_file>.cpp

where, the <value> can be one of the following values:

Value Description
fast[=1|2] Default value. Enables more aggressive optimizations on floating-point operations.
NOTE:

There is currently no difference between fast=1 and fast=2.

precise Disables optimizations that are not value-safe on floating-point data.

For additional information about this flag, refer to fp-model, fp topic in the Intel® oneAPI DPC++/C++ Compiler Developer Guide and Reference.