opt-subscript-in-range, Qopt-subscript-in-range

Determines whether the compiler assumes that there are no "large" integers being used or being computed inside loops.

Syntax

Linux OS and OS X:

-opt-subscript-in-range

-no-opt-subscript-in-range

Windows OS:

/Qopt-subscript-in-range

/Qopt-subscript-in-range-

Arguments

None

Default

-no-opt-subscript-in-range
or/Qopt-subscript-in-range-

The compiler assumes there are "large" integers being used or being computed within loops.

Description

This option determines whether the compiler assumes that there are no "large" integers being used or being computed inside loops.

If you specify [Q]opt-subscript-in-range, the compiler assumes that there are no "large" integers being used or being computed inside loops. A "large" integer is typically > 2³¹. This feature can enable more loop transformations.

IDE Equivalent

None

Alternate Options

None

Example

The following shows an example where these options can be useful. m is declared as type integer(kind=8) (64-bits) and all other variables inside the subscript are declared as type integer(kind=4) (32-bits):

A[ i  + j  + ( n  + k) * m ]