Intel® Fortran Compiler 16.0 User and Reference Guide

qopt-prefetch, Qopt-prefetch

Enables or disables prefetch insertion optimization. Option -qopt-prefetch is the replacement option for -opt-prefetch, which is deprecated.

Syntax

Linux and OS X:

-qopt-prefetch[=n]

-qno-opt-prefetch

Windows:

/Qopt-prefetch[:n]

/Qopt-prefetch-

Arguments

n

Is the level of software prefetching optimization desired. Possible values are:

0

Disables software prefetching. This is the same as specifying -qno-opt-prefetch (Linux* OS and OS X*) or /Qopt-prefetch- (Windows* OS).

1 to 4

Enables different levels of software prefetching. If you do not specify a value for n, the default is 2. Use lower values to reduce the amount of prefetching.

Default

-qno-opt-prefetch
or/Qopt-prefetch-

Prefetch insertion optimization is disabled.

Description

This option enables or disables prefetch insertion optimization. The goal of prefetching is to reduce cache misses by providing hints to the processor about when data should be loaded into the cache.

On Intel® 64 architecture targeting the Intel® Xeon Phi™ coprocessor x100 product family (formerly code name Knights Corner), option -qopt-prefetch=3 (Linux* OS and OS X*) and /Qopt-prefetch:3 (Windows* OS) are enabled by default if option O2 or higher is set. To disable prefetching at these optimization levels, specify option -qno-opt-prefetch or /Qopt-prefetch-.

This option enables prefetching when higher optimization levels are specified.

IDE Equivalent

Visual Studio: Optimization > Prefetch Insertion

Eclipse: None

Xcode: Optimization > Enable Prefetch Insertion

Alternate Options

None

See Also