Intel® Fortran Compiler 17.0 Developer Guide and Reference

Qsfalign

Specifies stack alignment for functions.

Architecture Restrictions

Only available on IA-32 architecture

Syntax

Linux and macOS:

None

Windows:

/Qsfalign[n]

Arguments

n

Is the byte size of aligned variables. Possible values are:

8

Specifies that alignment should occur for functions with 8-byte aligned variables. At this setting the compiler aligns the stack to 16 bytes if there is any 16-byte or 8-byte data on the stack. For 8-byte data, the compiler only aligns the stack if the alignment will produce a performance advantage.

16

Specifies that alignment should occur for functions with 16-byte aligned variables. At this setting, the compiler only aligns the stack for 16-byte data. No attempt is made to align for 8-byte data.

Default

/Qsfalign8

Alignment occurs for functions with 8-byte aligned variables.

Description

This option specifies stack alignment for functions. It lets you disable the normal optimization that aligns a stack for 8-byte data.

If you do not specify n, stack alignment occurs for all functions. If you specify /Qsfalign-, no stack alignment occurs for any function.

IDE Equivalent

None

Alternate Options

None