Preface

This guide is part of a set of manuals that describe how to use the PGI Fortran compilers and program development tools integrated with Microsoft Visual Studio. These tools, combined with Visual Studio and assorted libraries, are collectively known as PGI Visual Fortran®, or PVF®. You can use PVF to edit, compile, debug, optimize, and profile serial and parallel applications for x64 processor-based systems.

The PGI Visual Fortran Reference Manual is the reference companion to the PGI Visual Fortran User’s Guide which provides operating instructions for both the Visual Studio integrated development environment as well as command-level compilation and general information about PGI’s compilers. Neither guide teaches the Fortran programming language.

Audience Description

This manual is intended for scientists and engineers using PGI Visual Fortran. To fully understand this guide, you should be aware of the role of high-level languages, such as Fortran, in the software development process; and you should have some level of understanding of programming. PGI Visual Fortran is available on a variety of x86-64/x64 hardware platforms and variants of the Windows operating system. You need to be familiar with the basic commands available on your system.

Compatibility and Conformance to Standards

Your system needs to be running a properly installed and configured version of this PGI product. For information on installing PVF, refer to the Release Notes and Installation Guide included with your software.

For further information, refer to the following:

  • American National Standard Programming Language FORTRAN, ANSI X3. -1978 (1978).
  • ISO/IEC 1539-1 : 1991, Information technology – Programming Languages – Fortran, Geneva, 1991 (Fortran 90).
  • ISO/IEC 1539-1 : 1997, Information technology – Programming Languages – Fortran, Geneva, 1997 (Fortran 95).
  • ISO/IEC 1539-1 : 2004, Information technology – Programming Languages – Fortran, Geneva, 2004 (Fortran 2003).
  • ISO/IEC 1539-1 : 2010, Information technology – Programming Languages – Fortran, Geneva, 2010 (Fortran 2008).
  • Fortran 95 Handbook Complete ISO/ANSI Reference, Adams et al, The MIT Press, Cambridge, Mass, 1997.
  • The Fortran 2003 Handbook, Adams et al, Springer, 2009.
  • OpenMP Application Program Interface, Version 3.1, July 2011, http://www.openmp.org.
  • Programming in VAX Fortran, Version 4.0, Digital Equipment Corporation (September, 1984).
  • IBM VS Fortran, IBM Corporation, Rev. GC26-4119.
  • Military Standard, Fortran, DOD Supplement to American National Standard Programming Language Fortran, ANSI x.3-1978, MIL-STD-1753 (November 9, 1978).
  • ISO/IEC 9899:2011, Information Technology – Programming Languages – C, Geneva, 2011 (C11).
  • ISO/IEC 14882:2011, Information Technology – Programming Languages – C++, Geneva, 2011 (C++11).

Organization

Users typically begin by wanting to know how to use a product and often then find that they need more information and facts about specific areas of the product. Knowing how as well as why you might use certain options or perform certain tasks is key to using the PGI compilers and tools effectively and efficiently. However, once you have this knowledge and understanding, you very likely might find yourself wanting to know much more about specific areas or specific topics.

To facilitate ease of use, this manual contains detailed reference information about specific aspects of the compiler, such as the details of compiler options, directives, and more. This guide contains these sections:

Fortran Data Types describes the data types that are supported by the PGI Fortran compilers.

Command-Line Options Reference provides a detailed description of each command-line option.

Directives Reference contains detailed descriptions of PGI’s proprietary directives.

Runtime Environment describes the programming model supported for compiler code generation, including register conventions and calling conventions for x64 processor-based systems running a Windows operating system.

PVF Properties provides a description of Property Pages that PGI supports.

PVF Build Macros provides a description of the build macros that PVF supports.

Fortran Module/Library Interfaces for Windows provides a description of the Fortran module library interfaces that PVF supports.

Messages provides a list of compiler error messages.

Hardware and Software Constraints

This guide describes versions of the PGI Visual Fortran that are intended for use onx64 processor-based systems. Details concerning environment-specific values and defaults and system-specific features or limitations are presented in the release notes delivered with the PGI Visual Fortran.

Conventions

This guide uses the following conventions:

italic
is used for emphasis.
Constant Width
is used for filenames, directories, arguments, options, examples, and for language statements in the text, including assembly language statements.
Bold
is used for commands.
[ item1 ]
in general, square brackets indicate optional items. In this case item1 is optional. In the context of p/t-sets, square brackets are required to specify a p/t-set.
{ item2 | item 3 }
braces indicate that a selection is required. In this case, you must select either item2 or item3.
filename ...
ellipsis indicate a repetition. Zero or more of the preceding item may occur. In this example, multiple filenames are allowed.
FORTRAN
Fortran language statements are shown in the text of this guide using a reduced fixed point size.
C/C++
C/C++ language statements are shown in the test of this guide using a reduced fixed point size.

The PGI compilers and tools are supported on a wide variety of Linux, macOS and Windows operating systems running on 64-bit x86-compatible processors, and on Linux running on OpenPOWER processors. (Currently, the PGI debugger is supported on x86-64/x64 only.) See the Compatibility and Installation section on the PGI website for a comprehensive listing of supported platforms.

Note: Support for 32-bit development was deprecated in PGI 2016 and is no longer available as of the PGI 2017 release. PGI 2017 is only available for 64-bit operating systems and does not include the ability to compile 32-bit applications for execution on either 32- or 64-bit operating systems.

Terms

A number of terms related to systems, processors, compilers and tools are used throughout this guide. For example:

accelerator FMA -mcmodel=medium static linking
AVX host -mcmodel=small Win32
CUDA hyperthreading (HT) MPI Win64
device large arrays multicore x64
DLL license keys NUMA s86
driver LLVM SIMD x87
DWARF manycore SSE  

For a complete definition of these terms and other terms in this guide with which you may be unfamiliar, please refer to the PGI online glossary.

The following table lists the PGI compilers and tools and their corresponding commands:

Table 1. PGI Compilers and Commands
Compiler or Tool Language or Function Command
PGF77 ANSI FORTRAN 77 pgf77
PGFORTRAN ISO/ANSI Fortran 2003 pgfortran
PGI Debugger Source code debugger pgdbg
PGI Profiler Performance profiler pgprof

In general, the designation PGI Fortran is used to refer to the PGI Fortran 2003 compiler, and pgfortran is used to refer to the command that invokes the compiler. A similar convention is used for each of the PGI compilers and tools.

For simplicity, examples of command-line invocation of the compilers generally reference the pgfortran command, and most source code examples are written in Fortran. Usage of the PGF77 compiler, whose features are a subset of PGFORTRAN, is similar.

There are a wide variety of 64-bit x86-compatible processors in use. All are supported by the PGI compilers and tools. Most of these processors are forward-compatible, but not backward-compatible, meaning that code compiled to target a given processor will not necessarily execute correctly on a previous-generation processor.

A table listing the processor options that PGI supports is available in the Release Notes. The table also includes the features utilized by the PGI compilers that distinguish them from a compatibility standpoint.

In this manual, the convention is to use "x86" to specify the group of processors that are "32-bit" but not "64-bit". The convention is to use "x64" to specify the group of processors that are both "32-bit" and "64-bit". x86 processor-based systems can run only 32-bit operating systems. x64 processor-based systems can run either 32-bit or 64-bit operating systems, and can execute all 32-bit x86 binaries in either case. x64 processors have additional registers and 64-bit addressing capabilities that are utilized by the PGI compilers and tools when running on a 64-bit operating system. The prefetch, SSE1, SSE2, SSE3, and AVX processor features further distinguish the various processors. Where such distinctions are important with respect to a given compiler option or feature, it is explicitly noted in this manual.

Note: The default for performing scalar floating-point arithmetic is to use SSE instructions on targets that support SSE1 and SSE2.
Note: Support for 32-bit development was deprecated in PGI 2016 and is no longer available as of the PGI 2017 release. PGI 2017 is only available for 64-bit operating systems and does not include the ability to compile 32-bit applications for execution on either 32-bit or 64-bit operating systems.

1. Fortran Data Types

This section describes the scalar and aggregate data types recognized by the PGI Fortran compilers, the format and alignment of each type in memory, and the range of values each type can have on 64-bit operating systems.

1.1. Fortran Data Types

1.1.1. Fortran Scalars

A scalar data type holds a single value, such as the integer value 42 or the real value 112.6. The next table lists scalar data types, their size, format and range. Table 3 shows the range and approximate precision for Fortran real data types. Table 4 shows the alignment for different scalar data types. The alignments apply to all scalars, whether they are independent or contained in an array, a structure or a union.

Table 2. Representation of Fortran Data Types
Fortran Data Type Format Range
INTEGER 2's complement integer -231 to 231-1
INTEGER*2 2's complement integer -32768 to 32767
INTEGER*4 2's complement integer -231 to 231-1
INTEGER*8 2's complement integer -263 to 263-1
LOGICAL 32-bit value true or false
LOGICAL*1 8-bit value true or false
LOGICAL*2 16-bit value true or false
LOGICAL*4 32-bit value true or false
LOGICAL*8 64-bit value true or false
BYTE 2's complement -128 to 127
REAL Single-precision floating point 10-37 to 1038(1)
REAL*4 Single-precision floating point 10-37 to 10 38(1)
REAL*8 Double-precision floating point 10-307 to 10 308(1)
DOUBLE PRECISION Double-precision floating point 10-307 to 10308(1)
COMPLEX Single-precision floating point 10-37 to 1038(1)
DOUBLE COMPLEX Double-precision floating point 10-307 to 10308(1)
COMPLEX*16 Double-precision floating point 10-307 to 10308(1)
CHARACTER*n Sequence of n bytes  

(1) Approximate value

The logical constants .TRUE. and .FALSE. are all ones and all zeroes, respectively. Internally, the value of a logical variable is true if the least significant bit is one and false otherwise. When the option -⁠Munixlogical is set, a logical variable with a non-zero value is true and with a zero value is false.

Note: A variable of logical type may appear in an arithmetic context, and the logical type is then treated as an integer of the same size.
Table 3. Real Data Type Ranges
Data Type Binary Range Decimal Range Digits of Precision
REAL -2-126 to 2128 10-37 to 1038(1) 7–8
REAL*8 -2-1022 to 21024 10-307 to 10308(1) 15–16
Table 4. Scalar Type Alignment
This Type... ...Is aligned on this size boundary
LOGICAL*1 1-byte
LOGICAL*2 2-byte
LOGICAL*4 4-byte
LOGICAL*8 8-byte
BYTE 1-byte
INTEGER*2 2-byte
INTEGER*4 4-byte
INTEGER*8 8-byte
REAL*4 4-byte
REAL*8 8-byte
COMPLEX*8 4-byte
COMPLEX*16 8-byte

1.1.2. FORTRAN 77 Aggregate Data Type Extensions

The PGFORTRAN compiler supports de facto standard extensions to FORTRAN 77 that allow for aggregate data types. An aggregate data type consists of one or more scalar data type objects. You can declare the following aggregate data types:

  • An array consists of one or more elements of a single data type placed in contiguous locations from first to last.
  • A structure can contain different data types. The members are allocated in the order they appear in the definition but may not occupy contiguous locations.
  • A union is a single location that can contain any of a specified set of scalar or aggregate data types. A union can have only one value at a time. The data type of the union member to which data is assigned determines the data type of the union after that assignment.

The alignment of an array, a structure or union (an aggregate) affects how much space the object occupies and how efficiently the processor can address members. Arrays use the alignment of their members.

Array types
align according to the alignment of the array elements. For example, an array of INTEGER*2 data aligns on a 2-byte boundary.
Structures and Unions
align according to the alignment of the most restricted data type of the structure or union. In the next example, the union aligns on a 4-byte boundary since the alignment of c, the most restrictive element, is four.
STRUCTURE /astr/
UNION
 MAP
 INTEGER*2 a ! 2 bytes
 END MAP
 MAP
 BYTE b ! 1 byte
 END MAP
 MAP
 INTEGER*4 c ! 4 bytes
 END MAP
END UNION
END STRUCTURE

Structure alignment can result in unused space called padding. Padding between members of the structure is called internal padding. Padding between the last member and the end of the space is called tail padding.

The offset of a structure member from the beginning of the structure is a multiple of the member's alignment. For example, since an INTEGER*2 aligns on a 2-byte boundary, the offset of an INTEGER*2 member from the beginning of a structure is a multiple of two bytes.

1.1.3. Fortran 90 Aggregate Data Types (Derived Types)

The Fortran 90 standard added formal support for aggregate data types. The TYPE statement begins a derived type data specification or declares variables of a specified user-defined type. For example, the following would define a derived type ATTENDEE:

TYPE ATTENDEE
 CHARACTER(LEN=30) NAME
 CHARACTER(LEN=30) ORGANIZATION
 CHARACTER (LEN=30) EMAIL
END TYPE ATTENDEE

In order to declare a variable of type ATTENDEE and access the contents of such a variable, code such as the following would be used:

TYPE (ATTENDEE) ATTLIST(100)
. . .
ATTLIST(1)%NAME = ‘JOHN DOE’

2. Command-Line Options Reference

A command-line option allows you to specify specific behavior when a program is compiled and linked. Compiler options perform a variety of functions, such as setting compiler characteristics, describing the object code to be produced, controlling the diagnostic messages emitted, and performing some preprocessor functions. Most options that are not explicitly set take the default settings. This reference section describes the syntax and operation of each compiler option. For easy reference, the options are arranged in alphabetical order.

For an overview and tips on options usage and which options are best for which tasks, refer to the ‘Using Command-line Options’ section of the PVF User's Guide, which also provides summary tables of the different options.

This section uses the following notation:

[item]
Square brackets indicate that the enclosed item is optional.
{item | item}
Braces indicate that you must select one and only one of the enclosed items. A vertical bar (|) separates the choices.
...
Horizontal ellipses indicate that zero or more instances of the preceding item are valid.

2.1. PGI Compiler Option Summary

The following tables include all the PGI compiler options that are not language-specific. The options are separated by category for easier reference.

For a complete description of each option, refer to the detailed information later in this section.

2.2. Generic PGI Compiler Options

The following descriptions are for all the PGI options. For easy reference, the options are arranged in alphabetical order. For a list of options by tasks, refer to the tables in the beginning of this section.

2.2.1. -#

Displays the invocations of the compiler, assembler and linker.

Default

The compiler does not display individual phase invocations.

Usage

The following command-line requests verbose invocation information.

$ pgfortran -# prog.f

Description

The -⁠# option displays the invocations of the compiler, assembler and linker. These invocations are command-lines created by the driver from your command-line input and the default value.

2.2.2. -###

Displays the invocations of the compiler, assembler and linker, but does not execute them.

Default

The compiler does not display individual phase invocations.

Usage

The following command-line requests verbose invocation information.

$ pgfortran -### myprog.f

Description

Use the -⁠### option to display the invocations of the compiler, assembler and linker but not to execute them. These invocations are command lines created by the compiler driver from the rc files and the command-line options.

2.2.3. -acc

Enables OpenACC directives.

Default

The compiler enables OpenACC directives.

Syntax

-acc[=[no]autopar|[no]required|strict|verystrict]
[no]autopar
Enable [default] loop autoparallelization within acc parallel. The default is to autopar, that is, to enable loop autoparallelization.
[no]required
Instructs the compiler to issue a compiler error if the compute regions fail to accelerate. The default is required.
strict
Instructs the compiler to issue warnings for non-OpenACC accelerator directives.
verystrict
Instructs the compiler to fail with an error for any non-OpenACC accelerator directive.

Usage

The following command-line requests that OpenACC directives be enabled and that an error be issued for any non-OpenACC accelerator directive.

$ pgfortran -acc=verystrict -g prog.f

Description

The -⁠acc option enables OpenACC directives. You can use the suboptions to specify loop autoparallelization, how the compiler reports compute regions failures to accelerate, and whether to issue a warning or an error for non-OpenACC accelerator directives.

Starting in PGI 14.1, you control the OpenACC compiler behavior related to accelerator code generation failures with the required suboption. The OpenACC compilers now issue a compile-time error if accelerator code generation fails. In previous releases, the compiler would issue a warning, then generate code to run the compute kernel on the host. This previous behavior generates incorrect results if the compute kernels are inside a data region and the host and device memory values are inconsistent. You can enable the old behavior by using the -acc norequired switch.

2.2.4. -Bdynamic

Compiles for and links to the shared object version of the PGI runtime libraries.

Default

The compiler uses static libraries.

Usage

On Windows, you can create the DLL obj1.dll and its import library obj1.lib using the following series of commands:

% pgfortran -Bdynamic -c object1.f
% pgfortran -Mmakedll object1.obj -o obj1.dll

Then compile the main program using this command:

$ pgfortran -# prog.f

For a complete example in Windows, refer to the example: ‘Build a DLL: Fortran’ in the ‘Creating and Using Libraries’ section of the PGI Compiler User’s Guide.

Description

Use this option to compile for and link to the shared object version of the PGI runtime libraries. This flag is required when linking with any DLL built by the PGI compilers. For Windows, this flag corresponds to the /MD flag used by Microsoft’s cl compilers.

Note:

On Windows, -⁠Bdynamic must be used for both compiling and linking.

When you use the PGI compiler flag -⁠Bdynamic to create an executable that links to the shared object form of the runtime, the executable built is smaller than one built without -⁠Bdynamic. The PGI runtime shared object(s), however, must be available on the system where the executable is run. The -⁠Bdynamic flag must be used when an executable is linked against a shared object built by the PGI compilers.

2.2.5. -Bstatic

Compiles for and links to the static version of the PGI runtime libraries.

Default

The compiler uses static libraries.

Usage

The following command line explicitly compiles for and links to the static version of the PGI runtime libraries:
% pgfortran -Bstatic -c object1.f

Description

You can use this option to explicitly compile for and link to the static version of the PGI runtime libraries.

Note:

On Windows, -⁠Bstatic must be used for both compiling and linking.

For more information on using static libraries on Windows, refer to ‘Creating and Using Static Libraries on Windows’ in the ‘Creating and Using Libraries’ section of the PGI Compiler User’s Guide.

2.2.6. -Bstatic_pgi

Linux only. Compiles for and links to the static version of the PGI runtime libraries. Implies -⁠Mnorpath.

Default

The compiler uses static libraries.

Usage

The following command line explicitly compiles for and links to the static version of the PGI runtime libraries:

% pgfortran -Bstatic -c object1.f

Description

You can use this option to explicitly compile for and link to the static version of the PGI runtime libraries.

Note: On Linux, -⁠Bstatic_pgi results in code that runs on most Linux systems without requiring a Portability package.

For more information on using static libraries on Windows, refer to ‘Creating and Using Static Libraries on Windows’ in the ‘Creating and Using Libraries’ section of the PVF User's Guide.

2.2.7. -byteswapio

Swaps the byte-order of data in unformatted Fortran data files on input/output.

Default

The compiler does not byte-swap data on input/output.

Usage

The following command-line requests that byte-swapping be performed on input/output.

$ pgfortran -byteswapio myprog.f

Description

Use the -⁠byteswapio option to swap the byte-order of data in unformatted Fortran data files on input/output. When this option is used, the order of bytes is swapped in both the data and record control words; the latter occurs in unformatted sequential files.

You can use this option to convert big-endian format data files produced by most legacy RISC workstations to the little-endian format used on x86-64/x64 or OpenPOWER systems on the fly during file reads/writes.

This option assumes that the record layouts of unformatted sequential access and direct access files are the same on the systems. It further assumes that the IEEE representation is used for floating-point numbers. In particular, the format of unformatted data files produced by PGI Fortran compilers is identical to the format used on Sun and SGI workstations; this format allows you to read and write unformatted Fortran data files produced on those platforms from a program compiled for an x86-64/x64 or OpenPOWER platform using the -⁠byteswapio option.

2.2.8. -C

(Fortran only) Generates code to check array bounds.

Default

The compiler does not enable array bounds checking.

Usage

In this example, the compiler instruments the executable produced from myprog.f to perform array bounds checking at runtime:

$ pgfortran -C myprog.f

Description

Use this option to enable array bounds checking. If an array is an assumed size array, the bounds checking only applies to the lower bound. If an array bounds violation occurs during execution, an error message describing the error is printed and the program terminates. The text of the error message includes the name of the array, the location where the error occurred (the source file and the line number in the source), and information about the out of bounds subscript (its value, its lower and upper bounds, and its dimension).

2.2.9. -c

Halts the compilation process after the assembling phase and writes the object code to a file.

Default

The compiler produces an executable file and does not use the -⁠c option.

Usage

In this example, the compiler produces the object file myprog.obj in the current directory.

$ pgfortran -c myprog.f

Description

Use the -⁠c option to halt the compilation process after the assembling phase and write the object code to a file. If the input file is filename.f, the output file is .

2.2.10. -D

Creates a preprocessor macro with a given value.

Note:

You can use the -⁠D option more than once on a compiler command line. The number of active macro definitions is limited only by available memory.

Syntax

-Dname[=value]

Where name is the symbolic name and value is either an integer value or a character string.

Default

If you define a macro name without specifying a value, the preprocessor assigns the string 1 to the macro name.

Usage

In the following example, the macro PATHLENGTH has the value 256 until a subsequent compilation. If the -⁠D option is not used, PATHLENGTH is set to 128.

$ pgfortran -DPATHLENGTH=256 myprog.F

The source text in myprog.F is this:

	#ifndef PATHLENGTH
#define PATHLENGTH 128 	
#endif 	SUBROUTINE SUB 	CHARACTER*PATHLENGTH path 
	... 	
END

Description

Use the -⁠D option to create a preprocessor macro with a given value. The value must be either an integer or a character string.

You can use macros with conditional compilation to select source text during preprocessing. A macro defined in the compiler invocation remains in effect for each module on the command line, unless you remove the macro with an #undef preprocessor directive or with the -⁠U option. The compiler processes all of the -⁠U options in a command line after processing the -⁠D options.

To set this option in PVF, use the Fortran | Preprocessor | Preprocessor Definitions property, described in ‘Preprocessor Definitions’.

2.2.11. -dryrun

Displays the invocations of the compiler, assembler, and linker but does not execute them.

Default

The compiler does not display individual phase invocations.

Usage

The following command line requests verbose invocation information.

$ pgfortran -dryrun myprog.f

Description

Use the -⁠dryrun option to display the invocations of the compiler, assembler, and linker but not have them executed. These invocations are command lines created by the compiler driver from the rc files and the command-line supplied with -⁠dryrun.

2.2.12. -drystdinc

Displays the standard include directories and then exits the compiler.

Default

The compiler does not display standard include directories.

Usage

The following command line requests a display for the standard include directories.

$ pgfortran -drystdinc myprog.f

Description

Use the -⁠drystdinc option to display the standard include directories and then exit the compiler.

2.2.13. -E

Halts the compilation process after the preprocessing phase and displays the preprocessed output on the standard output.

Default

The compiler produces an executable file.

Usage

In the following example the compiler displays the preprocessed myprog.f on the standard output.

$ pgfortran -E myprog.f

Description

Use the -⁠E option to halt the compilation process after the preprocessing phase and display the preprocessed output on the standard output.

2.2.14. -F

Stops compilation after the preprocessing phase.

Default

The compiler produces an executable file.

Usage

In the following example the compiler produces the preprocessed file myprog.f in the current directory.

$ pgfortran -F myprog.F

Description

Use the -⁠F option to halt the compilation process after preprocessing and write the preprocessed output to a file. If the input file is filename.F, then the output file is filename.f.

2.2.15. -fast

Enables vectorization with SIMD instructions, cache alignment, and flushz for 64-bit targets.

Default

The compiler enables vectorization with SIMD instructions, cache alignment, and flushz.

Usage

In the following example the compiler produces vector SIMD code when targeting a 64-bit machine.

$ pgfortran -fast vadd.f95

Description

When you use this option, a generally optimal set of options is chosen for targets that support SIMD capability. In addition, the appropriate -⁠tp option is automatically included to enable generation of code optimized for the type of system on which compilation is performed. This option enables vectorization with SIMD instructions, cache alignment, and flushz.

Note: Auto-selection of the appropriate -⁠tp option means that programs built using the -⁠fastsse option on a given system are not necessarily backward-compatible with older systems.
Note: C/C++ compilers enable -⁠Mautoinline with -⁠fast.

To set this option in PVF, use the Fortran | General | Optimization property, described in ‘Optimization’.

2.2.16. -fastsse

Synonymous with -⁠fast.

2.2.17. --flagcheck

Causes the compiler to check that flags are correct and then exit without any compilation occuring.

Default

The compiler begins a compile without the additional step to first validate that flags are correct.

Usage

In the following example the compiler checks that flags are correct, and then exits.

$ pgfortran --flagcheck myprog.f

Description

Use this option to make the compiler check that flags are correct and then exit. If flags are all correct then the compiler returns a zero status. No compilation occurs.

2.2.18. -flags

Displays valid driver options on the standard output.

Default

The compiler does not display the driver options.

Usage

In the following example the user requests information about the known switches.

$ pgfortran -flags

Description

Use this option to display driver options on the standard output. When you use this option with -⁠v, in addition to the valid options, the compiler lists options that are recognized and ignored.

2.2.19. -g

Instructs the compiler to include symbolic debugging information in the object module; sets the optimization level to zero unless a -⁠O option is present on the command line.

Default

The compiler does not put debugging information into the object module.

Usage

In the following example, the object file myprog.obj contains symbolic debugging information.

$ pgfortran -c -g myprog.f

Description

Use the -⁠g option to instruct the compiler to include symbolic debugging information in the object module. Debuggers, including the PGI debugger, require symbolic debugging information in the object module to display and manipulate program variables and source code.

If you specify the -⁠g option on the command-line, the compiler sets the optimization level to -⁠O0 (zero), unless you specify the -⁠O option. For more information on the interaction between the -⁠g and -⁠O options, refer to the -⁠O entry. Symbolic debugging may give confusing results if an optimization level other than zero is selected.

Note:

Including symbolic debugging information increases the size of the object module.

To set this option in PVF, use the Fortran | General | Debug Information Format property, described in ‘Debug Information Format’ on page 377.

2.2.20. -gopt

Instructs the compiler to include symbolic debugging information in the object file, and to generate optimized code identical to that generated when -⁠g is not specified.

Default

The compiler does not put debugging information into the object module.

Usage

In the following example, the object file myprog.obj contains symbolic debugging information.

$ pgfortran -c -gopt myprog.f

Description

Using -⁠g alters how optimized code is generated in ways that are intended to enable or improve debugging of optimized code. The -⁠gopt option instructs the compiler to include symbolic debugging information in the object file, and to generate optimized code identical to that generated when -⁠g is not specified.

To set this option in PVF, use the Fortran | General | Debug Information Format property described in ‘Debug Information Format’.

2.2.21. -help

Used with no other options, -⁠help displays options recognized by the driver on the standard output. When used in combination with one or more additional options, usage information for those options is displayed to standard output.

Default

The compiler does not display usage information.

Usage

In the following example, usage information for -⁠Minline is printed to standard output.

$ pgcc -⁠help -⁠Minline   
-Minline[=lib:<inlib>|<maxsize>|<func>|except:<func>|name:<func>|maxsize:<n>|
totalsize:<n>|smallsize:<n>|reshape]
                    Enable function inlining
    lib:<inlib>     Use extracted functions from inlib
    <maxsize>       Set maximum function size to inline
    <func>          Inline function func
    except:<func>   Do not inline function func
    name:<func>     Inline function func
    maxsize:<n>     Inline only functions smaller than n
    totalsize:<n>   Limit inlining to total size of n
    smallsize:<n>   Always inline functions smaller than n
    reshape         Allow inlining in Fortran even when array shapes do not
                    match
    -Minline        Inline all functions that were extracted

In the following example, usage information for -⁠help shows how groups of options can be listed or examined according to function.

$ pgcc -help -help 
-help[=groups|asm|debug|language|linker|opt|other|
overall|phase|prepro|suffix|switch|target|variable]

Description

Use the -⁠help option to obtain information about available options and their syntax. You can use -⁠help in one of three ways:

  • Use -⁠help with no parameters to obtain a list of all the available options with a brief one-line description of each.
  • Add a parameter to -⁠help to restrict the output to information about a specific option. The syntax for this usage is this:
    -help <command line option>
  • Add a parameter to -⁠help to restrict the output to a specific set of options or to a building process. The syntax for this usage is this:
    -help=<subgroup>

The following table lists and describes the subgroups available with -⁠help.

Table 9. Subgroups for -⁠help Option
Use this -⁠help option To get this information...
-help=asm A list of options specific to the assembly phase.
-help=debug A list of options related to debug information generation.
-help=groups A list of available switch classifications.
-help=language A list of language-specific options.
-help=linker A list of options specific to link phase.
-help=opt A list of options specific to optimization phase.
-help=other A list of other options, such as ANSI conformance pointer aliasing for C.
-help=overall A list of options generic to any PGI compiler.
-help=phase A list of build process phases and to which compiler they apply.
-help=prepro A list of options specific to the preprocessing phase.
-help=suffix A list of known file suffixes and to which phases they apply.
-help=switch A list of all known options; this is equivalent to usage of -⁠help without any parameter.
-help=target A list of options specific to target processor.
-help=variable A list of all variables and their current value. They can be redefined on the command line using syntax VAR=VALUE.

2.2.22. -I

Adds a directory to the search path for files that are included using either the INCLUDE statement or the preprocessor directive #include.

Default

The compiler searches only certain directories for included files.

Syntax

-Idirectory

Where directory is the name of the directory added to the standard search path for include files.

Usage

In the following example, the compiler first searches the directory mydir and then searches the default directories for include files.

$ pgfortran -Imydir

Description

Adds a directory to the search path for files that are included using the INCLUDE statement or the preprocessor directive #include. Use the -⁠I option to add a directory to the list of where to search for the included files. The compiler searches the directory specified by the -⁠I option before the default directories.

The Fortran INCLUDE statement directs the compiler to begin reading from another file. The compiler uses two rules to locate the file:

  • If the file name specified in the INCLUDE statement includes a path name, the compiler begins reading from the file it specifies.
  • If no path name is provided in the INCLUDE statement, the compiler searches (in order):
    1. Any directories specified using the -⁠I option (in the order specified)
    2. The directory containing the source file
    3. The current directory

    For example, the compiler applies rule (1) to the following statements:

    INCLUDE '/bob/include/file1' (absolute path name)
    INCLUDE '../../file1' (relative path name)

    and rule (2) to this statement:

    INCLUDE 'file1'

To set this option in PVF, use the Fortran | General | Additional Include Directories property, described in ‘Additional Include Directories’, or the Fortran | Preprocessor | Additional Include Directories property, described in ‘Additional Include Directories’.

2.2.23. -i2, -⁠i4, -⁠i8

Treat INTEGER and LOGICAL variables as either two, four, or eight bytes.

Default

The compiler treats INTERGER and LOGICAL variables as four bytes.

Usage

In the following example, using the -⁠i8 switch causes the integer variables to be treated as 64 bits.

$ pgfortran -i8 int.f

int.f is a function similar to this:

int.f
     print *, "Integer size:", bit_size(i)
     end

Description

Use this option to treat INTEGER and LOGICAL variables as either two, four, or eight bytes. INTEGER*8 values not only occupy 8 bytes of storage, but operations use 64 bits, instead of 32 bits.

  • -i2: Treat INTEGER variables as 2 bytes.
  • -i4: Treat INTEGER variables as 4 bytes.
  • -i8: Treat INTEGER and LOGICAL variables as 8 bytes and use 64-bits for INTEGER*8 operations.

2.2.24. -K<flag>

Requests that the compiler provide special compilation semantics with regard to conformance to IEEE 754.

Default

The default is -⁠Knoieee and the compiler does not provide special compilation semantics.

Syntax

-K<flag>

Where flag is one of the following:

ieee Perform floating-point operations in strict conformance with the IEEE 754 standard. Some optimizations are disabled, and on some systems a more accurate math library is linked if -⁠Kieee is used during the link step.

To set this option in PVF, use the Fortran | Floating Point Options | IEEE Arithmetic property, described in ‘IEEE Arithmetic’.

noieee Default flag. Use the fastest available means to perform floating-point operations, link in faster non-IEEE libraries if available, and disable underflow traps.
trap=option

[,option]...

Controls the behavior of the processor when floating-point exceptions occur.

Possible options include:

  • fp
  • align (ignored)
  • inv
  • denorm
  • divz
  • ovf
  • unf
  • inexact

Usage

In the following example, the compiler performs floating-point operations in strict conformance with the IEEE 754 standard

$ pgfortran -Kieee myprog.f

Description

Use -⁠K to instruct the compiler to provide special compilation semantics.

-⁠Ktrap is only processed by the compilers when compiling main functions or programs. The options inv, denorm, divz, ovf, unf, and inexact correspond to the processor’s exception mask bits: invalid operation, denormalized operand, divide-by-zero, overflow, underflow, and precision, respectively.

Normally, the processor’s exception mask bits are on, meaning that floating-point exceptions are masked – the processor recovers from the exceptions and continues. If a floating-point exception occurs and its corresponding mask bit is off, or "unmasked", execution terminates with an arithmetic exception (C's SIGFPE signal).

-⁠Ktrap=fp is equivalent to -⁠Ktrap=inv,divz,ovf.

To set this option in PVF, use the Fortran | Floating Point Options | Floating Point Exception Handling property, described in ‘Floating Point Exception Handling’.

Note: The PGI compilers do not support exception-free execution for -⁠Ktrap=inexact. The purpose of this hardware support is for those who have specific uses for its execution, along with the appropriate signal handlers for handling exceptions it produces. It is not designed for normal floating point operation code support.

2.2.25. --keeplnk

(Windows only.) Preserves the temporary file when the compiler generates a temporary indirect file for a long linker command.

Usage

In the following example the compiler preserves each temporary file rather than deleting it.

$ pgfortran --keeplnk myprog.f

Description

If the compiler generates a temporary indirect file for a long linker command, use this option to instruct the compiler to preserve the temporary file instead of deleting it.

2.2.26. -L

Specifies a directory to search for libraries.

Note: Multiple -⁠L options are valid. However, the position of multiple -⁠L options is important relative to -⁠l options supplied.

Default

The compiler searches the standard library directory.

Syntax

-Ldirectory

Where directory is the name of the library directory.

Usage

In the following example, the library directory is /lib and the linker links in the standard libraries required by PGFORTRAN from this directory.

$ pgfortran -L/lib myprog.f

In the following example, the library directory /lib is searched for the library file libx.a and both the directories /lib and /libz are searched for liby.a.

$ pgfortran -L/lib -lx -L/libz -ly myprog.f

Description

Use the -⁠L option to specify a directory to search for libraries. Using -⁠L allows you to add directories to the search path for library files.

2.2.27. -l<library>

Instructs the linker to load the specified library. The linker searches <library>in addition to the standard libraries.

Note: The linker searches the libraries specified with -⁠l in order of appearance before searching the standard libraries.

Syntax

-llibrary

Where library is the name of the library to search.

Usage: In the following example, if the standard library directory is /lib the linker loads the library /lib/libmylib.a, in addition to the standard libraries.

$ pgfortran myprog.f -lmylib

Description

Use this option to instruct the linker to load the specified library. The compiler prepends the characters lib to the library name and adds the .a extension following the library name. The linker searches each library specified before searching the standard libraries.

2.2.28. -M

Generate make dependence lists. You can use -⁠MD,filename (pgc++ only) to generate make dependence lists and print them to the specified file.

2.2.29. -m

Displays a link map on the standard output.

Default

The compiler does display the link map and does not use the -⁠m option.

Usage

When the following example is executed on Windows, pgfortran creates a link map in the file myprog.map.

$ pgfortran -m myprog.f

Description

Use this option to display a link map.

  • On Linux, the map is written to stdout.
  • On Windows, the map is written to a .map file whose name depends on the executable. If the executable is myprog.f, the map file is in myprog.map.

2.2.30. -m64

Use the 64-bit compiler for the default processor type.

Usage

When the following example is executed, pgfortran uses the 64-bit compiler for the default processor type.

$ pgfortran -m64 myprog.f

Description

Use this option to specify the 64-bit compiler as the default processor type.

2.2.31. -M<pgflag>

Selects options for code generation. The options are divided into the following categories:

Code generation Fortran Language Controls Optimization
Environment C/C++ Language Controls Miscellaneous
Inlining    

The following table lists and briefly describes the options alphabetically and includes a field showing the category. For more details about the options as they relate to these categories, refer to ‘-⁠M Options by Category’ on page 113.

Table 10. -M Options Summary
pgflag Description Category
allocatable=95|03 Controls whether to use Fortran 95 or Fortran 2003 semantics in allocatable array assignments. Fortran Language
anno Annotate the assembly code with source code. Miscellaneous
[no]autoinline When a C/C++ function is declared with the inline keyword, inline it at -⁠O2. Inlining
[no]backslash Determines how the backslash character is treated in quoted strings. Fortran Language
[no]bounds Specifies whether array bounds checking is enabled or disabled. Miscellaneous
byteswapio Swap byte-order (big-endian to little-endian or vice versa) during I/O of Fortran unformatted data. Miscellaneous
cache_align Where possible, align data objects of size greater than or equal to 16 bytes on cache-line boundaries. Optimization
chkfpstk Check for internal consistency of the x87 FP stack in the prologue of a function and after returning from a function or subroutine call (-⁠tp px/p5/p6/piii targets only). Miscellaneous
chkptr Check for NULL pointers (pgf95, pgfortran only). Miscellaneous
chkstk Check the stack for available space upon entry to and before the start of a parallel region. Useful when many private variables are declared. Miscellaneous
concur Enable auto-concurrentization of loops. Multiple processors or cores will be used to execute parallelizable loops. Optimization
cpp Run the PGI cpp-like preprocessor without performing subsequent compilation steps. Miscellaneous
cray Force Cray Fortran (CF77) compatibility. Optimization
cuda Enables CUDA Fortran. Fortran Language
[no]daz Do/don’t treat denormalized numbers as zero. Code Generation
[no]dclchk Determines whether all program variables must be declared. Fortran Language
[no]defaultunit Determines how the asterisk character ("*") is treated in relation to standard input and standard output, regardless of the status of I/O units 5 and 6.. Fortran Language
[no]depchk Checks for potential data dependencies. Optimization
[no]dse Enables [disables] dead store elimination phase for programs making extensive use of function inlining. Optimization
[no]dlines Determines whether the compiler treats lines containing the letter "D" in column one as executable statements. Fortran Language
dll Link with the DLL version of the runtime libraries (Windows only). Miscellaneous
dollar,char Specifies the character to which the compiler maps the dollar sign code. Fortran Language
[no]dwarf Specifies not to add DWARF debug information. Code Generation
dwarf1 When used with -⁠g, generate DWARF1 format debug information. Code Generation
dwarf2 When used with -⁠g, generate DWARF2 format debug information. Code Generation
dwarf3 When used with -⁠g, generate DWARF3 format debug information. Code Generation
extend Instructs the compiler to accept 132-column source code; otherwise it accepts 72-column code. Fortran Language
extract invokes the function extractor. Inlining
[no]f[=option] Perform certain floating point intrinsic functions using relaxed precision. Optimization
fixed Instructs the compiler to assume F77-style fixed format source code (pgf95, pgfortran only). Fortran Language
[no]flushz Do/don't set SSE flush-to-zero mode Code Generation
[no]fpapprox Specifies not to use low-precision fp approximation operations. Optimization
free Instructs the compiler to assume F90-style free format source code. Fortran Language
func32 The compiler aligns all functions to 32-byte boundaries. Code Generation
gccbug[s] Matches behavior of certain gcc bugs Miscellaneous
info Prints informational messages regarding optimization and code generation to standard output as compilation proceeds. Miscellaneous
inform Specifies the minimum level of error severity that the compiler displays. Miscellaneous
inline Invokes the function inliner. Inlining
[no]iomutex Determines whether critical sections are generated around Fortran I/O calls. Fortran Language
[no]ipa Invokes interprocedural analysis and optimization. Optimization
keepasm Instructs the compiler to keep the assembly file. Miscellaneous
largeaddressaware [Win64 only] Generates code that allows for addresses greater than 2GB, using RIP-relative addressing. Code Generation
[no]large_arrays Enables support for 64-bit indexing and single static data objects of size larger than 2GB. Code Generation
list Specifies whether the compiler creates a listing file. Miscellaneous
[no]loop32 Aligns [does not align] innermost loops on 32 byte boundaries with -⁠tp barcelona Code Generation
[no]lre Enable [disable] loop-carried redundancy elimination. Optimization
makedll Generate a dynamic link library (DLL).. Miscellaneous
makeimplib Passes the -def switch to the librarian without a deffile, when used without -⁠def:deffile. Miscellaneous
mpi=option Link to MPI libraries: MPICH, SGI, or Microsoft MPI libraries Code Generation
neginfo Instructs the compiler to produce information on why certain optimizations are not performed. Miscellaneous
noframe Eliminates operations that set up a true stack frame pointer for functions. Optimization
noi4 Determines how the compiler treats INTEGER variables. Optimization
nomain When the link step is called, don’t include the object file that calls the Fortran main program.. Code Generation
noopenmp When used in combination with the -⁠mp option, the compiler ignores OpenMP parallelization directives , but still processes SGI-style parallelization directives. Miscellaneous
nopgdllmain Do not link the module containing the default DllMain() into the DLL. Miscellaneous
nosgimp When used in combination with the -⁠mp option, the compiler ignores SGI-style parallelization directives, but still processes OpenMP directives. Miscellaneous
nostdinc Instructs the compiler to not search the standard location for include files. To set this option in PVF, use the Fortran | Preprocessor | Ignore Standard Include Path property. Environment
nostdlib Instructs the linker to not link in the standard libraries. Environment
[no]onetrip Determines whether each DO loop executes at least once. Language
novintr Disable idiom recognition and generation of calls to optimized vector functions. Optimization
pfi Instrument the generated code and link in libraries for dynamic collection of profile and data information at runtime. Optimization
pre Read a pgfi.out trace file and use the information to enable or guide optimizations. Optimization
[no]pre Force [disable] generation of non-temporal moves and prefetching. Code Generation
[no]prefetch Enable [disable] generation of prefetch instructions. Optimization
preprocess Perform cpp-like preprocessing on assembly language and Fortran input source files. Miscellaneous
prof Enable Compiler feedback and modify DWARF sections. Code Generation
[no]r8 Determines whether the compiler promotes REAL variables and constants to DOUBLE PRECISION. Optimization
[no]r8intrinsics Determines how the compiler treats the intrinsics CMPLX and REAL. Optimization
[no]recursive Allocate [do not allocate] local variables on the stack; this allows recursion. SAVEd, data-initialized, or namelist members are always allocated statically, regardless of the setting of this switch. Code Generation
[no]reentrant Specifies whether the compiler avoids optimizations that can prevent code from being reentrant. Code Generation
[no]ref_externals Do [do not] force references to names appearing in EXTERNAL statements. Code Generation
safe_lastval In the case where a scalar is used after a loop, but is not defined on every iteration of the loop, the compiler does not by default parallelize the loop. However, this option tells the compiler it is safe to parallelize the loop. For a given loop, the last value computed for all scalars make it safe to parallelize the loop. Code Generation
[no]save Determines whether the compiler assumes that all local variables are subject to the SAVE statement. Fortran Language
[no]scalarsse Do [do not] use SSE/SSE2 instructions to perform scalar floating-point arithmetic. Optimization
[no]second_underscore Do [do not] add the second underscore to the name of a Fortran global if its name already contains an underscore. Code Generation
[no]signextend Do [do not] extend the sign bit, if it is set. Code Generation
[no]smart Do [do not] enable optional post-pass assembly optimizer. Optimization
[no]smartalloc[=huge| huge:<n>|hugebss] Add a call to the routine mallopt in the main routine. Supports large TLBs on Linux and Windows.
Tip: To be effective, this switch must be specified when compiling the file containing the Fortran, C, or C++ main program.
Environment
standard Causes the compiler to flag source code that does not conform to the ANSI standard. Fortran Language
[no]stride0 Do [do not] generate alternate code for a loop that contains an induction variable whose increment may be zero. Code Generation
[no]unixlogical Determines how the compiler treats logical values.. Fortran Language
[no]unroll Controls loop unrolling. Optimization
[no]upcase Determines whether the compiler preserves uppercase letters in identifiers.. Fortran Language
varargs Forces Fortran program units to assume calls are to C functions with a varargs type interface . Code Generation
[no]vect Do [do not] invoke the code vectorizer. Optimization

2.2.32. -module <moduledir>

Allows you to specify a particular directory in which generated intermediate .mod files should be placed.

Default

The compiler places .mod files in the current working directory, and searches only in the current working directory for pre-compiled intermediate .mod files.

Usage

The following command line requests that any intermediate module file produced during compilation of myprog.f be placed in the directory mymods; specifically, the file ./mymods/myprog.mod is used.

$ pgfortran -module mymods myprog.f

Description

Use the -⁠module option to specify a particular directory in which generated intermediate .mod files should be placed. If the -⁠module <moduledir> option is present, and USE statements are present in a compiled program unit, then <moduledir> is searched for .mod intermediate files prior to a search in the default local directory.

To set this option in PVF, use the Fortran | Output | Module Path property, described in ‘Module Path’.

2.2.33. -mp

Instructs the compiler to interpret user-inserted OpenMP shared-memory parallel programming directives, and to generate an executable file which will utilize multiple processors in a shared-memory parallel system.

Default

The compiler interprets user-inserted shared-memory parallel programming directives when linking. To disable this option, use the -⁠nomp option when linking.

Usage

The following command line requests processing of any shared-memory directives present in myprog.f:

$ pgfortran -mp myprog.f

Description

Use the -⁠mp option to instruct the compiler to interpret user-inserted OpenMP shared-memory parallel programming directives and to generate an executable file which utilizes multiple processors in a shared-memory parallel system.

The suboptions are one or more of the following:

align
Forces loop iterations to be allocated to OpenMP processes using an algorithm that maximizes alignment of vector sub-sections in loops that are both parallelized and vectorized for SSE. This allocation can improve performance in program units that include many such loops. It can also result in load-balancing problems that significantly decrease performance in program units with relatively short loops that contain a large amount of work in each iteration. The numa suboption uses libnuma on systems where it is available.
allcores
Instructs the compiler to target all available cores. You specify this suboption at link time.
bind
Instructs the compiler to bind threads to cores. You specify this suboption at link time.
[no]numa
Uses [does not use] libnuma on systems where it is available.

For a detailed description of this programming model and the associated directives, refer to Section 9, ‘Using OpenMP’ of the PGI Compiler User's Guide.

To set this option in PVF, use the Fortran | Language | Enable OpenMP Directives property, described in ‘Enable OpenMP Directives’.

2.2.34. -noswitcherror

Issues warnings instead of errors for unknown switches. Ignores unknown command line switches after printing a warning message.

Default

The compiler prints an error message and then halts.

Usage

In the following example, the compiler ignores unknown command line switches after printing a warning message.

$ pgfortran -noswitcherror myprog.f

Description

Use this option to instruct the compiler to ignore unknown command line switches after printing an warning message.

Tip: You can configure this behavior in the siterc file by adding: set NOSWITCHERROR=1.

2.2.35. -O<level>

Invokes code optimization at the specified level.

Default

The compiler optimizes at level 2.

Syntax

-O [level]

Where level is an integer from 0 to 4.

Usage

In the following example, since no -⁠O option is specified, the compiler sets the optimization to level 1.

$ pgfortran myprog.f

In the following example, since no optimization level is specified and a -⁠O option is specified, the compiler sets the optimization to level 2.

$ pgfortran -O myprog.f

Description

Use this option to invoke code optimization.Using the PGI compiler commands with the -⁠Olevel option (the capital O is for Optimize), you can specify any of the following optimization levels:

-O0
Level zero specifies no optimization. A basic block is generated for each language statement.
-O1
Level one specifies local optimization. Scheduling of basic blocks is performed. Register allocation is performed.
-O
When no level is specified, level two global optimizations are performed, including traditional scalar optimizations, induction recognition, and loop invariant motion. No SIMD vectorization is enabled.
-O2
Level two specifies global optimization. This level performs all level-one local optimization as well as level-two global optimization described in -⁠O. In addition, this level enables more advanced optimizations such as SIMD code generation, cache alignment, and partial redundancy elimination.
-O3
Level three specifies aggressive global optimization. This level performs all level-one and level-two optimizations and enables more aggressive hoisting and scalar replacement optimizations that may or may not be profitable.
-O4
Level four performs all level-one, level-two, and level-three optimizations and enables hoisting of guarded invariant floating point expressions.

To set this option (-⁠O2 or -⁠O3) in PVF, use the Fortran | Optimization | Global Optimizations property, described in ‘Global Optimizations’.

The following table shows the interaction between the -⁠O option, -⁠g option, -⁠Mvect, and -⁠Mconcur options.

Table 11. Optimization and -⁠O, -⁠g, -⁠Mvect, and -⁠Mconcur Options
Optimize Option Debug Option -M Option Optimization Level
none none none 1
none none -Mvect 2
none none -Mconcur 2
none -g none 0
-O none or -⁠g none 2
-Olevel none or -⁠g none level
-Olevel < 2 none or -⁠g -Mvect 2
-Olevel < 2 none or -⁠g -Mconcur 2

Unoptimized code compiled using the option -⁠O0 can be significantly slower than code generated at other optimization levels. Like the -⁠Mvect option, the -⁠Munroll option sets the optimization level to level-2 if no -⁠O or -⁠g options are supplied. The -⁠gopt option is recommended for generation of debug information with optimized code. For more information on optimization, refer to the ‘Optimizing and Parallelizing’ section of the PVF User's Guide.

2.2.36. -o

Names the executable file. Use the -⁠o option to specify the filename of the compiler object file. The final output is the result of linking.

Default

The compiler creates executable filenames as needed. If you do not specify the -⁠o option, the default filename is the linker output file with a name comprised of the base file name, such as myprog, plus the extension .exe, for example: myprog.exe.

Syntax

-o filename

Where filename is the name of the file for the compilation output. The filename should not have a .f extension.

Usage

In the following example, the executable file is myp.exe instead of the default a.outmyprog.exe.

$ pgfortran myprog.f -o myp

To set this option in PVF, use the Fortran | Output | Object File Name property, described in ‘Object File Name’ on page 377.

2.2.37. -pc

Note: This option is available only for -⁠tp px/p5/p6/piii targets.

Allows you to control the precision of operations performed using the x87 floating point unit, and their representation on the x87 floating point stack.

Syntax

-pc { 32 | 64 | 80 }

Usage

$ pgfortran -pc 64 myprog.f

Description

The x87 architecture implements a floating-point stack using eight 80-bit registers. Each register uses bits 0–63 as the significant, bits 64–78 for the exponent, and bit 79 is the sign bit. This 80-bit real format is the default format, called the extended format. When values are loaded into the floating point stack they are automatically converted into extended real format. The precision of the floating point stack can be controlled, however, by setting the precision control bits (bits 8 and 9) of the floating control word appropriately. In this way, you can explicitly set the precision to standard IEEE double-precision using 64 bits, or to single precision using 32 bits.

According to Intel documentation, this only affects the x87 operations of add, subtract, multiply, divide, and square root. In particular, it does not appear to affect the x87 transcendental instructions.

The default precision is system dependent. To alter the precision in a given program unit, the main program must be compiled with the same -pc option. The command line option -⁠pc val lets the programmer set the compiler’s precision preference.

Valid values for val are:

32 single precision 64 double precision 80 extended precision

Extended Precision Option – Operations performed exclusively on the floating-point stack using extended precision, without storing into or loading from memory, can cause problems with accumulated values within the extra 16 bits of extended precision values. This can lead to answers, when rounded, that do not match expected results.

For example, if the argument to sin is the result of previous calculations performed on the floating-point stack, then an 80-bit value used instead of a 64-bit value can result in slight discrepancies. Results can even change sign due to the sin curve being too close to an x-intercept value when evaluated. To maintain consistency in this case, you can assure that the compiler generates code that calls a function. According to the x86 ABI, a function call must push its arguments on the stack (in this way memory is guaranteed to be accessed, even if the argument is an actual constant). Thus, even if the called function simply performs the inline expansion, using the function call as a wrapper to sin has the effect of trimming the argument precision down to the expected size. Using the -⁠Mnobuiltin option on the command line for C accomplishes this task by resolving all math routines in the library libm, performing a function call of necessity. The other method of generating a function call for math routines, but one that may still produce the inline instructions, is by using the -⁠Kieee switch.

A second example illustrates the precision control problem using a section of code to determine machine precision:

program find_precision 
			 
	 w = 1.0
	100 w=w+w
	 y=w+1
	 z=y-w
	 if (z .gt. 0) goto 100
	C now w is just big enough that |((w+1)-w)-1| >= 1
	...
	 print*,w
	 end

In this case, where the variables are implicitly real*4, operations are performed on the floating-point stack where optimization removes unnecessary loads and stores from memory. The general case of copy propagation being performed follows this pattern:

a = x
  y = 2.0 + a

Instead of storing x into a, then loading a to perform the addition, the value of x can be left on the floating-point stack and added to 2.0. Thus, memory accesses in some cases can be avoided, leaving answers in the extended real format. If copy propagation is disabled, stores of all left-hand sides will be performed automatically and reloaded when needed. This will have the effect of rounding any results to their declared sizes.

The find_precision program has a value of 1.8446744E+19 when executed using default (extended) precision. If, however, -⁠Kieee is set, the value becomes 1.6777216E+07 (single precision.) This difference is due to the fact that -⁠Kieee disables copy propagation, so all intermediate results are stored into memory, then reloaded when needed. Copy propagation is only disabled for floating-point operations, not integer. With this particular example, setting the -⁠pc switch will also adjust the result.

The -⁠Kieee switch also has the effect of making function calls to perform all transcendental operations. Except when the -⁠Mnobuiltin switch is set in C, the function still produces the x86 machine instruction for computation, and arguments are passed on the stack, which results in a memory store and load.

Finally, -⁠Kieee also disables reciprocal division for constant divisors. That is, for a/b with unknown a and constant b, the expression is usually converted at compile time to a*(1/b), thus turning an expensive divide into a relatively fast scalar multiplication. However, numerical discrepancies can occur when this optimization is used.

Understanding and correctly using the -⁠pc, -⁠Mnobuiltin, and -⁠Kieee switches should enable you to produce the desired and expected precision for calculations which utilize floating-point operations.

2.2.38. --pedantic

Prints warnings from included <system header files>.

Default

The compiler prints the warnings from the included system header files.

Usage

In the following example, the compiler prints the warnings from the included system header files.

$ pgc++ --power myprog.cc

2.2.39. -pgc++libs

Instructs the compiler to append C⁠+⁠+ runtime libraries to the link line for programs built using either PGF77 or PGF90 .

Default

The C/C++ compilers do not append the C++ runtime libraries to the link line.

Usage

In the following example the C⁠+⁠+ runtime libraries are linked with an object file compiled with pgf77 .

$ pgf90 main.f90 mycpp.o -pgc++libs

Description

Use this option to instruct the compiler to append C⁠+⁠+ runtime libraries to the link line for programs built using either PGF77 or PGF90 .

2.2.40. -pgf77libs

Instructs the compiler to append PGF77 runtime libraries to the link line.

Default

The C/C++ compilers do not append the PGF77 runtime libraries to the link line.

Usage

In the following example a .c main program is linked with an object file compiled with pgf77.

$ pgcc main.c myf77.o -pgf77libs

Description

Use this option to instruct the compiler to append PGF77 runtime libraries to the link line.

2.2.41. -pgf90libs

Instructs the compiler to append PGF90/PGF95/PGFORTRAN runtime libraries to the link line.

Default

The C/C++ compilers do not append the PGF90/PGF95/PGFORTRAN runtime libraries to the link line.

Usage

In the following example a .c main program is linked with an object file compiled with pgfortran.

$ pgcc main.c myf95.o -pgf90libs

Description

Use this option to instruct the compiler to append PGF90/PGF95/PGFORTRAN runtime libraries to the link line.

2.2.42. -r4 and -⁠r8

Interprets DOUBLE PRECISION variables as REAL (-⁠r4), or interprets REAL variables as DOUBLE PRECISION (-⁠r8).

Usage

In this example, the double precision variables are interpreted as REAL.

$ pgfortran -r4 myprog.f

Description

Interpret DOUBLE PRECISION variables as REAL (-⁠r4) or REAL variables as DOUBLE PRECISION (-⁠r8).

2.2.43. -rc

Specifies the name of the driver startup configuration file. If the file or pathname supplied is not a full pathname, the path for the configuration file loaded is relative to the $DRIVER path (the path of the currently executing driver). If a full pathname is supplied, that file is used for the driver configuration file.

Syntax

-rc [path] filename

Where path is either a relative pathname, relative to the value of $DRIVER, or a full pathname beginning with "/". Filename is the driver configuration file.

Usage

In the following example, the file .pgfortranrctest, relative to /usr/pgi/linux86-64/bin , the value of $DRIVER, is the driver configuration file.

$ pgfortran -rc .pgfortranrctest myprog.f

Description

Use this option to specify the name of the driver startup configuration file. If the file or pathname supplied is not a full pathname, the path for the configuration file loaded is relative to the $DRIVER path – the path of the currently executing driver. If a full pathname is supplied, that file is used for the driver configuration file.

2.2.44. -S

Stops compilation after the compiling phase and writes the assembly-language output to a file.

Default

The compiler does not retain a .s file.

Usage

In this example, pgfortran produces the file myprog.s in the current directory.

$ pgfortran -S myprog.f

Description

Use this option to stop compilation after the compiling phase and then write the assembly-language output to a file. If the input file is filename.f, then the output file is filename.s.

2.2.45. -show

Produces driver help information describing the current driver configuration.

Default

The compiler does not show driver help information.

Usage

In the following example, the driver displays configuration information to the standard output after processing the driver configuration file.

$ pgfortran -show myprog.f

Description

Use this option to produce driver help information describing the current driver configuration.

2.2.46. -silent

Do not print warning messages.

Default

The compiler prints warning messages.

Usage

In the following example, the driver does not display warning messages.

$ pgfortran -silent myprog.f

Description

Use this option to suppress warning messages.

2.2.47. -stack

(Windows only) Allows you to explicitly set stack properties for your program.

Default

If -⁠stack is not specified, then the defaults are as followed:

Win64
No default setting

Syntax

-stack={ (reserved bytes)[,(committed bytes)] }{, [no]check }

Usage

The following example demonstrates how to reserve 524,288 stack bytes (512KB), commit 262,144 stack bytes for each routine (256KB), and disable the stack initialization code with the nocheck argument.

$ pgfortran -stack=524288,262144,nocheck myprog.f

Description

Use this option to explicitly set stack properties for your program. The -⁠stack option takes one or more arguments: (reserved bytes), (committed bytes), [no]check.

reserved bytes
Specifies the total stack bytes required in your program.
committed bytes
Specifies the number of stack bytes that the Operating System will allocate for each routine in your program. This value must be less than or equal to the stack reserved bytes value.

Default for this argument is 4096 bytes.

[no]check
Instructs the compiler to generate or not to generate stack initialization code upon entry of each routine. Check is the default, so stack initialization code is generated.

Stack initialization code is required when a routine's stack exceeds the committed bytes size. When your committed bytes is equal to the reserved bytes or equal to the stack bytes required for each routine, then you can turn off the stack initialization code using the -stack=nocheck compiler option. If you do this, the compiler assumes that you are specifying enough committed stack space; and therefore, your program does not have to manage its own stack size.

For more information on determining the amount of stack required by your program, refer to -⁠Mchkstk compiler option, described in ‘Miscellaneous Controls’.

Note:-stack=(reserved bytes),(committed bytes) are linker options.

-stack=[no]check is a compiler option.

If you specify -stack=(reserved bytes),(committed bytes) on your compile line, it is only used during the link step of your build. Similarly, -stack=[no]check can be specified on your link line, but it's only used during the compile step of your build.

2.2.48. -ta=tesla(tesla_suboptions),host

Defines the target accelerator and the type of code to generate. This flag is valid for Fortran, C, and C++ on supported platforms.

Note: There are three major suboptions:
  • tesla(:tesla_suboptions)
  • host
  • multicore

Default

The compiler uses -ta=tesla,host.

Usage

In the following example, tesla is the accelerator target architecture and the accelerator generates code for compute capability 3.0.

$ pgfortran -ta=tesla,cc30

Description

Use this option to select the accelerator target and, optionally, to define the type of code to genertate.

The -ta flag has the following options:

tesla
Select the tesla accelerator target. This option has the following tesla-suboptions:
cc30
Generate code for compute capability 3.0.
cc35
Generate code for compute capability 3.5.
cc3x
Generate code for the lowest 3.x compute capability possible.
cc3+
Is equivalent to cc3x.
[no]debug
Enable[disable] debug information generation in device code.
fastmath
Use routines from the fast math library.
[no]flushz
Enable[disable] flush-to-zero mode for floating point computations in the GPU code generated forPGI Accelerator model compute regions.
keep
Keep the kernel files.
kepler
is equivalent to cc3x.
kepler+
is equivalent to cc3+.
llvm
Generate code using the llvm-based back-end.
[no]debug
Enable[disable] GPU debug information generation.
deepcopy
Enable full deep copy of aggregate data structions in OpenACC; Fortran only.
[no]lineinfo
Enable[disable] GPU line information generation.
maxregcount:n
Specify the maximum number of registers to use on the GPU. Leaving this blank indicates no limit.
[no]fma
Do not generate fused multiply-add instructions.
noL1
Prevents the use of L1 hardware data cache to cache global variables.
pin+
is equivalent to cc3+.
[no]rdc
Generate [do not generate] relocatable device code.
[no]required
Generate [do not generate] a compiler error if accelerator device code cannot be generated.
tesla
is equivalent to -ta=tesla,cc2+
host
Use the host option to generate code to execute OpenACC regions on the host.

The -ta=host flag has no suboptions.

multicore
Use the multicore option to generate OpenACC parallel regions to execute in parallel on individual host cores.

The -ta=multicore flag has no suboptions.

Multiple Targets

When host is one of the multiple targets, such as -⁠ta=tesla,host, the result is generated code that can be run with or without an attached accelerator.

Relocatable Device Code

A rdc option is available for the -⁠ta and -⁠Mcuda flags that specifies to generate relocatable device code. Starting in PGI 14.1, the default code generation and linking mode for NVIDIA-target OpenACC and CUDA Fortran is rdc, relocatable device code.

You can disable the default and enable the old behavior and non-relocatable code by specifying any of the following: -⁠ta=tesla:nordc, -⁠Mcuda=nordc.

LLVM and Native GPU Code Generation

For accelerator code generation, PGI 2017 has two options.

  • The compilers generate an LLVM-based intermediate representation by default.
  • In legacy mode, the compilers generate low-level CUDA C code. To enable this code generation, use -⁠ta=tesla:nollvm on NVIDIA Tesla hardware.

DWARF Debugging Formats

PGI's debugging capability for Tesla uses the LLVM back-end. Use the compiler's -⁠g option to enable the generation of full dwarf information on both the host and device; in the absence of other optimization flags, -⁠g sets the optimization level to zero. If a -⁠O option raises the optimization level to one or higher, only GPU line information is generated on the device even when -⁠g is specified. To enforce full dwarf generation for device code at optimization levels above zero, use the debug suboption to -⁠ta=tesla. Conversely, to prevent the generation of dwarf information for device code, use the nodebug suboption to -⁠ta=tesla. Both debug and nodebug can be used independently of -⁠g.

2.2.49. -time

Print execution times for various compilation steps.

Default

The compiler does not print execution times for compilation steps.

Usage

In the following example, pgfortran prints the execution times for the various compilation steps.

$ pgfortran -time myprog.f

Description

Use this option to print execution times for various compilation steps.

2.2.50. -tp <target>[,target...]

Sets the target processor.

Default

The PGI compilers produce code specifically targeted to the type of processor on which the compilation is performed. In particular, the default is to use all supported instructions wherever possible when compiling on a given system.

The default target processor is auto-selected depending on the processor on which the compilation is performed. You can specify a target processor to compile for a different processor type, such as to select a more generic processor, allowing the code to run on more system types. Specifying two or more target processors enables unified binary code generation, where two or more versions of each function may be generated, each version optimized for the specific instruction set available in each target processor.

Executables created on a given system without the -tp flag may not be usable on previous generation systems. For example, executables created on an Intel Sandybridge processor may use instructions that are not available on earlier Intel Nehalem or Intel P7 systems.

Syntax

Syntax for 64-bit targets:

-tp {k8-64 | k8-64e | p7-64 | core2-64 | x64}

Usage

In the following example, pgfortran sets the target processor to a 64-bit Intel Nehalem processor:

$ pgfortran -tp=nehalem-64 myprog.f

Description

Use this option to set the target architecture. By default, the PGI compiler uses all supported instructions wherever possible when compiling on a given system.

Processor-specific optimizations can be specified or limited explicitly by using the -⁠tp option. Thus, it is possible to create executables that are usable on previous generation systems.

To set this option in PVF, use the Fortran | Target Processors | Unified Binary Information property, described in ‘Unified Binary Information’.

The following list contains the possible suboptions for -⁠tp and the processors that each suboption is intended to target. Options without a bit-length suffix use the current width associated with the driver on your path.

barcelona
generate code for AMD Opteron/Quadcore and compatible processors.
bulldozer
generate code for AMD Bulldozer and compatible processors.
core2
generate code for Intel Core 2 Duo and compatible processors.
haswell
generate code that is usable on any Haswell processor-based system.
istanbul
generate code that is usable on any Istanbul processor-based system.
k8
generate code hat is usable on any AMD64 and compatible processor.
k8-64e
generate 64-bit code for AMD Opteron Revision E, AMD Turion, and compatible processors.
nehalem
generate code that is usable on any Nehalem processor-based system.
p7
generate code for Pentium 4 and compatible processors.
penryn
generate code for Intel Penryn Architecture and compatible processors.
piledriver
generate code that is usable on any Piledriver processor-based system.
px
generate code that is usable on any x86-64 processor-based system.
sandybridge
generate code for Intel Sandy Bridge and compatible processors.
shanghai
generate code that is usable on any AMD Shanghai processor-based system.
x64
generate 64-bit unified binary code including full optimizations and support for both AMD and Intel x86-64 processors.

Refer to the PGI Release Notes for a concise list of the features of these processors that distinguish them as separate targets when using the PGI compilers and tools.

Different processors have differences, some subtle, in hardware features such as instruction sets and cache size. The compilers make architecture-specific decisions about such things as instruction selection, instruction scheduling, and vectorization. Any of these decisions can have significant effects on performance and compatibility. PGI unified binaries provide a low-overhead means for a single program to run well on a number of hardware platforms.

You can use the -⁠tp option to produce PGI Unified Binary programs. The compilers generate, and combine into one executable, multiple binary code streams, each optimized for a specific platform. At runtime, this one executable senses the environment and dynamically selects the appropriate code stream.

The target processor switch, -⁠tp , accepts a comma-separated list of 64-bit targets and will generate code optimized for each listed target. For example, the following switch generates optimized code for three targets: k8-64, p7-64, and core2-64.

Syntax for optimizing for multiple targets:

-tp k8-64,p7-64,core2-64

The -⁠tp k8-64 and -⁠tp k8-64e options result in generation of code supported on and optimized for AMD x64 processors, while the -⁠tp p7-64 option results in generation of code that is supported on and optimized for Intel x86-64 processors. Performance of k8-64 or k8-64e code executed on Intel x86-64 processors, or of p7-64 code executed on AMD x86-64 processors, can often be significantly less than that obtained with a native binary.

The special -⁠tp x64 option is equivalent to -⁠tp k8-64,p7-64 . This switch produces PGI Unified Binary programs containing code streams fully optimized and supported for bothAMD64 and Intel 64 processors.

For more information on unified binaries, refer to the section ’Processor-Specific Optimization and the Unified Binary’ in the PGI Compiler User’s Guide.

2.2.51. -[no]traceback

Adds debug information for runtime traceback for use with the environment variable PGI_TERM.

Default

The compiler enables traceback for FORTRAN and disables traceback for C and C++.

Syntax

-traceback

Usage

In this example, pgfortran enables traceback for the program myprog.f.

$ pgfortran -traceback myprog.f

Description

Use this option to enable or disable runtime traceback information for use with the environment variable PGI_TERM.

Setting setTRACEBACK=OFF; in siterc or .mypg*rc also disables default traceback.

Using ON instead of OFF enables default traceback.

2.2.52. -u

Initializes the symbol-table with <symbol>, which is undefined for the linker. An undefined symbol triggers loading of the first member of an archive library.

Default

The compiler does not use the -⁠u option.

Syntax

-usymbol

Where symbol is a symbolic name.

Usage

In this example, pgfortran initializes symbol-table with test.

$ pgfortran -utest myprog.f

Description

Use this option to initialize the symbol-table with <symbol>, which is undefined for the linker. An undefined symbol triggers loading of the first member of an archive library.

2.2.53. -U

Undefines a preprocessor macro.

Syntax

-Usymbol

Where symbol is a symbolic name.

Usage

The following examples undefine the macro test.

$ pgfortran -Utest myprog.F
$ pgfortran -Dtest -Utest myprog.F

Description

Use this option to undefine a preprocessor macro. You can also use the #undef pre-processor directive to undefine macros.

To set this option in PVF, use the Fortran | Preprocessor | Undefine Preprocessor Definitions property, described in ‘Undefine Preprocessor Definitions’.

2.2.54. -V[release_number]

Displays additional information, including version messages. Further, if a release_number is appended, the compiler driver attempts to compile using the specified release instead of the default release.

Note: There can be no space between -V and release_number.

Default

The compiler does not display version information and uses the release specified by your path to compile.

Usage

The following command-line shows the output using the -⁠V option.

% pgfortran -V myprog.f

The following command-line causes pgcc to compile using the 5.2 release instead of the default release.

% pgcc -V5.2 myprog.c

Description

Use this option to display additional information, including version messages or, if a release_number is appended, to instruct the compiler driver to attempt to compile using the specified release instead of the default release.

The specified release must be co-installed with the default release, and must have a release number greater than or equal to 4.1, which was the first release that supported this functionality.

To set this option in PVF, use the Fortran | General | Display Startup Banner property, described in ‘Display Startup Banner’.

2.2.55. -v

Displays the invocations of the compiler, assembler, and linker.

Default

The compiler does not display individual phase invocations.

Usage

In the following example you use -⁠v to see the commands sent to compiler tools, assembler, and linker.

$ pgfortran -v myprog.f90

Description

Use the -⁠v option to display the invocations of the compiler, assembler, and linker. These invocations are command lines created by the compiler driver from the files and the -⁠W options you specify on the compiler command-line.

2.2.56. -W

Passes arguments to a specific phase.

Syntax

-W{0 | a | l },option[,option...]
Note: You cannot have a space between the -⁠W and the single-letter pass identifier, between the identifier and the comma, or between the comma and the option.
0
(the number zero) specifies the compiler.
a
specifies the assembler.
l
(lowercase letter l) specifies the linker.
option
is a string that is passed to and interpreted by the compiler, assembler or linker. Options separated by commas are passed as separate command line arguments.

Usage

In the following example the linker loads the text segment at address 0xffc00000 and the data segment at address 0xffe00000.

$ pgfortran -Wl,-k,-t,0xffc00000,-d,0xffe00000 myprog.f

Description

Use this option to pass arguments to a specific phase. You can use the -⁠W option to specify options for the assembler, compiler, or linker.

A given PGI compiler command invokes the compiler driver, which parses the command-line, and generates the appropriate commands for the compiler, assembler, and linker.

2.2.57. -w

Do not print warning messages.

Default

The compiler prints warning messages.

Usage

In the following example no warning messages are printed.

$ pgfortran -w myprog.f

Description

Use the -⁠w option to not print warning messages. Sometimes the compiler issues many warning in which you may have no interest. You can use this option to not issue those warnings.

2.3. -M Options by Category

This section describes each of the options available with -⁠M by the categories:

Code Generation Fortran Language Controls Optimization Environment
C/C++ Language Controls Inlining Miscellaneous  

The following sections provide detailed descriptions of several, but not all, of the -⁠M<pgflag> options. For a complete alphabetical list of all the options, refer to Table 10. These options are grouped according to categories and are listed with exact syntax, defaults, and notes concerning similar or related options.

2.3.1. Code Generation Controls

This section describes the -⁠M<pgflag> options that control code generation.

Default: For arguments that you do not specify, the default code generation controls are these:

nodaz norecursive nosecond_underscore
noflushz noreentrant nostride0
largeaddressaware noref_externals signextend

Related options:-⁠D, -⁠I, -⁠L, -⁠l, -⁠U.

The following list provides the syntax for each -⁠M<pgflag> option that controls code generation. Each option has a description and, if appropriate, any related options.

-Mdaz
Set IEEE denormalized input values to zero; there is a performance benefit but misleading results can occur, such as when dividing a small normalized number by a denormalized number.
To take effect, this option must be set for the main program. To set this option in PVF, use the Fortran |
Floating Point Options | Treat Denormalized Values as Zero property, described in ‘Treat Denormalized Values as Zero’
-Mnodaz
Do not treat denormalized numbers as zero.
To take effect, this option must be set for the main program.
-Mnodwarf
Specifies not to add DWARF debug information.
To take effect, this option must be used in combination with -⁠g.
-Mdwarf1
Generate DWARF1 format debug information.
To take effect, this option must be used in combination with -⁠g.
-Mdwarf2
Generate DWARF2 format debug information.
To take effect, this option must be used in combination with -⁠g.
-Mdwarf3
Generate DWARF3 format debug information.
To take effect, this option must be used in combination with -⁠g.
-Mflushz
Set SSE flush-to-zero mode; if a floating-point underflow occurs, the value is set to zero.
To take effect, this option must be set for the main program.
To set this option in PVF, use the Fortran | Floating Point Options | Flush Denormalized Results to Zero property, described in ‘Flush Denormalized Results to Zero’ on page 391.
-Mnoflushz
Do not set SSE flush-to-zero mode; generate underflows.
To take effect, this option must be set for the main program.
-Mfunc32
Align functions on 32-byte boundaries.
-Minstrument[=functions] (linux86-64 only)
Generate additional code to enable instrumentation of functions. The option -⁠Minstrument=functions is the same as -⁠Minstrument.
Implies -⁠Minfo=ccff and -⁠Mframe.
-Mlargeaddressaware=[no]
[Win64 only] Generates code that allows for addresses greater than 2 GB, using RIP-relative addressing.
Use-⁠Mlargeaddressaware=no for a direct addressing mechanism that restricts the total addressable memory.
Note: Do not use -⁠Mlargeaddressaware=no if the object file will be placed in a DLL.

If -⁠Mlargeaddressaware=no is used to compile any object file, it must also be used when linking.
-Mlarge_arrays
Enable support for 64-bit indexing and single static data objects larger than 2 GB in size. This option is the default in the presence of -⁠mcmodel=medium. It can be used separately together with the default small memory model for certain 64-bit applications that manage their own memory space.
For more information, refer to the ‘Programming Considerations for 64-Bit Environments’ section of the PVF User's Guide.
-Mnolarge_arrays
Disable support for 64-bit indexing and single static data objects larger than 2 GB in size. When this option is placed after -⁠mcmodel=medium on the command line, it disables use of 64-bit indexing for applications that have no single data object larger than 2 GB.
For more information, refer to the ‘Programming Considerations for 64-Bit Environments’ section of the PVF User's Guide.
-Mnomain
Instructs the compiler not to include the object file that calls the Fortran main program as part of the link step. This option is useful for linking programs in which the main program is written in C/C++ and one or more subroutines are written in Fortran.
-Mmpi=option
-⁠Mmpi adds the include and library options to the compile and link commands necessary to build an MPI application using MPI header files and libraries.
To use -⁠Mmpi, you must have a version of MPI installed on your system.
This option tells the compiler to use the headers and libraries for the specified version of MPI.
-Mmpi=msmpi – Select the default Microsoft MPI libraries on Windows.

For more information, refer to the ‘Programming Considerations for 64-Bit Environments’ section of the PVF User's Guide.
-M[no]movnt
Instructs the compiler to generate nontemporal move and prefetch instructions even in cases where the compiler cannot determine statically at compile-time that these instructions will be beneficial.
-M[no]pre
enables [disables] partial redundancy elimination.
-Mprof[=option[,option,...]]
Set performance profiling options. Use of these options changes which sections are included in the binary. These sections can be read by the PGI profiler.
The option argument can be any of the following:
[no]ccff
Enable [disable] common compiler feedback format, CCFF, information.
dwarf
Add limited DWARF symbol information sufficient for most performance profilers.
-Mrecursive
instructs the compiler to allow Fortran subprograms to be called recursively.
-Mnorecursive
Fortran subprograms may not be called recursively.
-Mref_externals
force references to names appearing in EXTERNAL statements.
-Mnoref_externals
do not force references to names appearing in EXTERNAL statements.
-Mreentrant
instructs the compiler to avoid optimizations that can prevent code from being reentrant.
-Mnoreentrant
instructs the compiler not to avoid optimizations that can prevent code from being reentrant.
-Msecond_underscore
instructs the compiler to add a second underscore to the name of a Fortran global symbol if its name already contains an underscore. This option is useful for maintaining compatibility with object code compiled using g77, which uses this convention by default.
-Mnosecond_underscore
instructs the compiler not to add a second underscore to the name of a Fortran global symbol if its name already contains an underscore.
-Msafe_lastval
When a scalar is used after a loop, but is not defined on every iteration of the loop, the compiler does not by default parallelize the loop. However, this option tells the compiler it’s safe to parallelize the loop. For a given loop, the last value computed for all scalars makes it safe to parallelize the loop.
-Msignextend
instructs the compiler to extend the sign bit that is set as a result of converting an object of one data type to an object of a larger signed data type.
-Mnosignextend
instructs the compiler not to extend the sign bit that is set as the result of converting an object of one data type to an object of a larger data type.
-Mstack_arrays
places automatic arrays on the stack.
-Mnostack_arrays
allocates automatic arrays on the heap. -Mnostack_arrays is the default and what traditionally has been the approach used.
-Mstride0
instructs the compiler to inhibit certain optimizations and to allow for stride 0 array references. This option may degrade performance and should only be used if zero-stride induction variables are possible.
-Mnostride0
instructs the compiler to perform certain optimizations and to disallow for stride 0 array references.
-Mvarargs
force Fortran program units to assume procedure calls are to C functions with a varargs-type interface.

2.3.2. Environment Controls

This section describes the -⁠M<pgflag> options that control environments.

Default: For arguments that you do not specify, the default environment option depends on your configuration.

The following list provides the syntax for each -⁠M<pgflag> option that controls environments. Each option has a description and, if appropriate, a list of any related options.

-Mnostartup
instructs the linker not to link in the standard startup routine that contains the entry point (_start) for the program.
Note: If you use the -⁠Mnostartup option and do not supply an entry point, the linker issues the following error message: Warning: cannot find entry symbol _start
-M[no]smartalloc[=huge|huge:<n>|hugebss|nohuge]
adds a call to the routine mallopt in the main routine. This option supports large TLBs on Linux and Windows. This option must be used to compile the main routine to enable optimized malloc routines.
The option arguments can be any of the following:
huge
Link in the huge page runtime library.
Enables large 2-megabyte pages to be allocated. The effect is to reduce the number of TLB entries required to execute a program. This option is most effective on Barcelona and Core 2 systems; older architectures do not have enough TLB entries for this option to be beneficial. By itself, the huge suboption tries to allocate as many huge pages as required.
huge:<n>
Link the huge page runtime library and allocate n huge pages. Use this suboption to limit the number of huge pages allocated to n.
You can also limit the pages allocated by using the environment variable PGI_HUGE_PAGES.
hugebss
(64-bit only) Puts the BSS section in huge pages; attempts to put a program's uninitialized data section into huge pages.
Note: This flag dynamically links the library libhugetlbfs_pgi even if -⁠Bstatic is used.
nohuge
Overrides a previous -⁠Msmartalloc=huge setting.
Tip: To be effective, this switch must be specified when compiling the file containing the Fortran, C, or C++ main program.
-Mnostdinc
instructs the compiler to not search the standard location for include files. To set this option in PVF, use the Fortran | Preprocessor | Ignore Standard Include Path property, described in ‘Ignore Standard Include Path’ on page 381.
-Mnostdlib
instructs the linker not to link in the standard libraries in the library directory lib within the standard directory. You can link in your own library with the -⁠l option or specify a library directory with the -⁠L option.

2.3.3. Fortran Language Controls

This section describes the -⁠M<pgflag> options that affect Fortran language interpretations by the PGI Fortran compilers. These options are valid only for the Fortran compiler drivers.

Default: Before looking at all the options, let's look at the defaults. For arguments that you do not specify, the defaults are as follows:

backslash nodefaultunit dollar,_ noonetrip nounixlogical
nodclchk nodlines noiomutex nosave noupcase

The following list provides the syntax for each -⁠M<pgflag> option that affect Fortran language interpretations. Each option has a description and, if appropriate, a list of any related options.

-Mallocatable=95|03
controls whether Fortran 95 or Fortran 2003 semantics are used in allocatable array assignments. The default behavior is to use Fortran 95 semantics; the 03 option instructs the compiler to use Fortran 2003 semantics.
-Mbackslash
instructs the compiler to treat the backslash as a normal character, and not as an escape character in quoted strings.
-Mnobackslash
instructs the compiler to recognize a backslash as an escape character in quoted strings (in accordance with standard C usage).
-Mcuda
instructs the compiler to enable CUDA Fortran.

The following suboptions exist:

Note: If more than one option is on the command line, all the specified options occur.
cc30
Generate code for compute capability 3.0.
cc35
Generate code for compute capability 3.5.
cc3x
Generate code for the lowest 3.x compute capability possible.
cc3+
Is equivalent to cc3x.
cc50
Generate code for compute capability 5.0.
cc60
Generate code for compute capability 6.0.
cuda7.5 or 7.5
Specify the NVIDIA CUDA 7.5 version of the toolkit. This is the default.
cuda8.0 or 8.0
Specify the NVIDIA CUDA 8.0 version of the toolkit.
Note: Compile with the CUDA 7.5 or CUDA 8.0 toolkit either by using the -Mcuda=7.5 or -Mcuda=8.0 option, or by adding set DEFCUDAVERSION=7.5 or set DEFCUDAVERSION=8.0 to the siterc file. This action generates binaries that may not work on machines with an earlier CUDA driver.
pgaccelinfo prints the driver version as the first line of output.
  • For a 7.5 driver: CUDA Driver Version 7050
  • For an 8.0 driver: CUDA Driver Version 8000
emu
Enable CUDA Fortran emulation mode.
fastmath
Use routines from the fast math library.
fermi
is equivalent to -Mcuda,cc2x
[no]flushz
Enable[disable] flush-to-zero mode for floating point computations in the GPU code generated for CUDA Fortran kernels.
generate rdc
Generate relocatable device code
keepbin
Keep the generated binary (.bin) file for CUDA Fortran.
keepgpu
Keep the generated GPU code for CUDA Fortran.
keepptx
Keep the portable assembly (.ptx) file for the GPU code.
kepler
is equivalent to -Mcuda,cc3x
llvm
Generate code using the llvm-based back-end.
[no]debug
Enable[disable] GPU debug information generation.
[no]lineinfo
Enable[disable] GPU line information generation.
maxregcount:n
Specify the maximum number of registers to use on the GPU. Leaving this blank indicates no limit.
nofma
Do not generate fused multiply-add instructions.
noL1
Prevent the use of L1 hardware data cache to cache global variables.
ptxinfo
Show PTXAS informational messages during compilation.
rdc
Enable CUDA Fortran separate compilation and linking of device routines, including device routines in Fortran modules.
To enable separate compilation and linking, include the command line option -Mcuda=rdc on both the compile and the link steps.
-Mdclchk
instructs the compiler to require that all program variables be declared.
-Mnodclchk
instructs the compiler not to require that all program variables be declared.
-Mdefaultunit
instructs the compiler to treat "*" as a synonym for standard input for reading and standard output for writing.
-Mnodefaultunit
instructs the compiler to treat "*" as a synonym for unit 5 on input and unit 6 on output.
-Mdlines
instructs the compiler to treat lines containing "D" in column 1 as executable statements (ignoring the "D").
-Mnodlines
instructs the compiler not to treat lines containing "D" in column 1 as executable statements. The compiler does not ignore the "D".
-Mdollar,char
char specifies the character to which the compiler maps the dollar sign. The compiler allows the dollar sign in names.
-Mextend
instructs the compiler to accept 132-column source code; otherwise it accepts 72-column code.
-Mfixed
instructs the compiler to assume input source files are in FORTRAN 77-style fixed form format.
-Mfree
instructs the compiler to assume input source files are in Fortran 90/95 freeform format.
-Miomutex
instructs the compiler to generate critical section calls around Fortran I/O statements.
-Mnoiomutex
instructs the compiler not to generate critical section calls around Fortran I/O statements.
-Monetrip
instructs the compiler to force each DO loop to execute at least once. This option is useful for programs written for earlier versions of Fortran.
-Mnoonetrip
instructs the compiler not to force each DO loop to execute at least once.
-Msave
instructs the compiler to assume that all local variables are subject to the SAVE statement.
This may allow older Fortran programs to run, but it can greatly reduce performance.
-Mnosave
instructs the compiler not to assume that all local variables are subject to the SAVE statement.
-Mstandard
instructs the compiler to flag non-ANSI-conforming source code.
-Munixlogical
directs the compiler to treat logical values as true if the value is non-zero and false if the value is zero (UNIX F77 convention). When -⁠Munixlogical is enabled, a logical value or test that is non-zero is .TRUE., and a value or test that is zero is .FALSE.. In addition, the value of a logical expression is guaranteed to be one (1) when the result is .TRUE..
-Mnounixlogical
directs the compiler to use the VMS convention for logical values for true and false. Even values are true and odd values are false.
-Mupcase
instructs the compiler to preserve uppercase letters in identifiers.
With -⁠Mupcase, the identifiers "X" and "x" are different. Keywords must be in lower case.
This selection affects the linking process. If you compile and link the same source code using -⁠Mupcase on one occasion and -⁠Mnoupcase on another, you may get two different executables – depending on whether the source contains uppercase letters. The standard libraries are compiled using the default -⁠Mnoupcase .
-Mnoupcase
instructs the compiler to convert all identifiers to lower case.
This selection affects the linking process. If you compile and link the same source code using -⁠Mupcase on one occasion and -⁠Mnoupcase on another, you may get two different executables, depending on whether the source contains uppercase letters. The standard libraries are compiled using -⁠Mnoupcase.

2.3.4. Inlining Controls

This section describes the -⁠M<pgflag> options that control function inlining.

Usage:Before looking at all the options, let’s look at a couple examples. In the following example, the compiler extracts functions that have 500 or fewer statements from the source file myprog.f and saves them in the file extract.il.

$ pgfortran -Mextract=500 -o extract.il myprog.f

In the following example, the compiler inlines functions with fewer than approximately 100 statements in the source file myprog.f.

$ pgfortran -Minline=maxsize:100 myprog.f

Related options: -⁠o, -⁠Mextract

The following list provides the syntax for each -⁠M<pgflag> option that controls function inlining. Each option has a description and, if appropriate, a list of any related options.

- M[no]autoinline[=option[,option,...]]
instructs the compiler to inline [not to inline] a C/C++ function at -⁠O2, where the option can be any of these:
maxsize:n
instructs the compiler not to inline functions of size > n. The default size is 100.
totalsize:n
instructs the compiler to stop inlining when the size equals n. The default size is 800.
-Mextract[=option[,option,...]]
Extracts functions from the file indicated on the command line and creates or appends to the specified extract directory where option can be any of the following:
name:func
instructs the extractor to extract function func from the file.
size:number
instructs the extractor to extract functions with number or fewer statements from the file.
lib:filename.ext
instructs the extractor to use directory filename.ext as the extract directory, which is required to save and re-use inline libraries.

If you specify both name and size, the compiler extracts functions that match func, or that have number or fewer statements. For examples of extracting functions, refer to the ‘Using Function Inlining’ section of the PVF User's Guide.

-Minline[=option[,option,...]]
instructs the compiler to pass options to the function inliner, where the option can be any of the following:
except:func
Inlines all eligible functions except func, a function in the source text. You can use a comma-separated list to specify multiple functions.
[name:]func
Inlines all functions in the source text whose name matches func. You can use a comma-separated list to specify multiple functions.

The function name should be a non-numeric string that does not contain a period. You can also use a name: prefix followed by the function name. If name: is specified, what follows is always the name of a function.

[maxsize:]number
A numeric option is assumed to be a size. Functions of size number or less are inlined. If both number and function are specified, then functions matching the given name(s) or meeting the size requirements are inlined.

The size number need not exactly equal the number of statements in a selected function; the size parameter is merely a rough guage.

[no]reshape
instructs the inliner to allow [disallow] inlining in Fortran even when array shapes do not match. The default is -⁠Minline=noreshape, except with -⁠Mconcur or -⁠mp, where the default is -⁠Minline=reshape,=reshape.
smallsize:number
Always inline functions of size smaller than number regardless of other size limits.
totalsize:number
Stop inlining in a function when the function's total inlined size reaches the number specified.
[lib:]filename.ext
instructs the inliner to inline the functions within the library file filename.ext. The compiler assumes that a filename.ext option containing a period is a library file.
Tip: Create the library file using the -⁠Mextract option. You can also use a lib: prefix followed by the library name.
  • If lib: is specified, no period is necessary in the library name. Functions from the specified library are inlined.
  • If no library is specified, functions are extracted from a temporary library created during an extract prepass.

If you specify both func and number, the compiler inlines functions that match the function name or have number or fewer statements.

Inlining can be disabled with -⁠Mnoinline.

To set this option in PVF, use the Fortran | Optimization | Inlining property, described in ‘Inlining’

For examples of inlining functions, refer to ‘Using Function Inlining’ in the PGI Compiler User’s Guide.

2.3.5. Optimization Controls

This section describes the -⁠M<pgflag> options that control optimization.

Default: Before looking at all the options, let's look at the defaults. For arguments that you do not specify, the default optimization control options are as follows:

depchk noipa nounroll nor8
i4 nolre novect nor8intrinsics
nofprelaxed noprefetch    
Note: If you do not supply an option to -⁠Mvect, the compiler uses defaults that are dependent upon the target system.

Usage: In this example, the compiler invokes the vectorizer with use of packed SSE instructions enabled.

>$ pgfortran -Mvect=sse -Mcache_align myprog.f

Related options:-⁠g, -⁠O

The following list provides the syntax for each -⁠M<pgflag> option that controls optimization. Each option has a description and, if appropriate, a list of any related options.

-Mcache_align
Align unconstrained objects of length greater than or equal to 16 bytes on cache-line boundaries. An unconstrained object is a data object that is not a member of an aggregate structure or common block. This option does not affect the alignment of allocatable or automatic arrays.
To effect cache-line alignment of stack-based local variables, the main program or function must be compiled with -⁠Mcache_align.
-Mconcur[=option [,option,...]]
Instructs the compiler to enable auto-concurrentization of loops. If -⁠Mconcur is specified, multiple processors will be used to execute loops that the compiler determines to be parallelizable.
option is one of the following:
allcores
Instructs the compiler to use all available cores. Use this option at link time.
[no]altcode:n
Instructs the parallelizer to generate alternate serial code for parallelized loops.
  • If altcode is specified without arguments, the parallelizer determines an appropriate cutoff length and generates serial code to be executed whenever the loop count is less than or equal to that length.
  • If altcode:n is specified, the serial altcode is executed whenever the loop count is less than or equal to n.
  • If noaltcode is specified, the parallelized version of the loop is always executed regardless of the loop count.
cncall
Indicates that calls in parallel loops are safe to parallelize.
Loops containing calls are candidates for parallelization. Also, no minimum loop count threshold must be satisfied before parallelization will occur, and last values of scalars are assumed to be safe.
[no]innermost
Instructs the parallelizer to enable parallelization of innermost loops. The default is to not parallelize innermost loops, since it is usually not profitable on dual-core processors.
noassoc
Instructs the parallelizer to disable parallelization of loops with reductions.
When linking, the -⁠Mconcur switch must be specified or unresolved references result. The NCPUS environment variable controls how many processors or cores are used to execute parallelized loops.
To set this option in PVF, use the Fortran | Optimization | Auto-Parallelization property, described in ‘Auto-Parallelization’.
Note: This option applies only on shared-memory multi-processor (SMP) or multicore processor-based systems.
-Mcray[=option[,option,...]]
Force Cray Fortran (CF77) compatibility with respect to the listed options. Possible values of option include:
pointer
for purposes of optimization, it is assumed that pointer-based variables do not overlay the storage of any other variable.
-Mdepchk
instructs the compiler to assume unresolved data dependencies actually conflict.
-Mnodepchk
Instructs the compiler to assume potential data dependencies do not conflict. However, if data dependencies exist, this option can produce incorrect code.
-Mdse
Enables a dead store elimination phase that is useful for programs that rely on extensive use of inline function calls for performance. This is disabled by default.
-Mnodse
Disables the dead store elimination phase. This is the default.
-M[no]fpapprox[=option]
Perform certain floating point operations using low-precision approximation.
-⁠Mnofpapprox specifies not to use low-precision fp approximation operations.
By default -⁠Mfpapprox is not used.
If -⁠Mfpapprox is used without suboptions, it defaults to use approximate div, sqrt, and rsqrt. The available suboptions are these:
div
Approximate floating point division
sqrt
Approximate floating point square root
rsqrt
Approximate floating point reciprocal square root
-M[no]fpmisalign
Instructs the compiler to allow (not allow) vector arithmetic instructions with memory operands that are not aligned on 16-byte boundaries. The default is -⁠Mnofpmisalign on all processors.
Note: Applicable only with one of these options: -⁠tp barcelona or -⁠tp barcelona-64 or newer processors.
-M[no]fprelaxed[=option]
Instructs the compiler to use [not use] relaxed precision in the calculation of some intrinsic functions. Can result in improved performance at the expense of numerical accuracy.
To set this option in PVF, use the Fortran | Floating Point Options | Floating Point Consistency property. For more information on this property, refer to ‘Floating Point Consistency’.

The possible values for option are:
div
Perform divide using relaxed precision.
intrinsic
Enables use of relaxed precision intrinsics.
noorder
Do not allow expression reordering or factoring.
order
Allow expression reordering, including factoring.
recip
Perform reciprocal using relaxed precision.
rsqrt
Perform reciprocal square root (1/sqrt) using relaxed precision.
sqrt
Perform square root with relaxed precision.
With no options, -⁠Mfprelaxed generates relaxed precision code for those operations that generate a significant performance improvement, depending on the target processor.
The default is -⁠Mnofprelaxed which instructs the compiler to not use relaxed precision in the calculation of intrinsic functions.
-Mi4
instructs the compiler to treat INTEGER variables as INTEGER*4.
-Mlre[=array | assoc | noassoc]
Enables loop-carried redundancy elimination, an optimization that can reduce the number of arithmetic operations and memory references in loops. The available suboptions are:
array
treat individual array element references as candidates for possible loop-carried redundancy elimination. The default is to eliminate only redundant expressions involving two or more operands.
assoc
allow expression re-association. Specifying this suboption can increase opportunities for loop-carried redundancy elimination but may alter numerical results.
noassoc
disallow expression re-association.
-Mnolre
Disable loop-carried redundancy elimination.
-Mnoframe
Eliminate operations that set up a true stack frame pointer for every function. With this option enabled, you cannot perform a traceback on the generated code and you cannot access local variables.
To set this option in PVF, use the Fortran | Optimization | Use Frame Pointer property, described in ‘Use Frame Pointer’
-Mnoi4
instructs the compiler to treat INTEGER variables as INTEGER*2.
-Mpre
Enables partial redundancy elimination.
-Mprefetch[=option [,option...]]
enables generation of prefetch instructions on processors where they are supported. Possible values for option include:
d:m
set the fetch-ahead distance for prefetch instructions to m cache lines.
n:p
set the maximum number of prefetch instructions to generate for a given loop to p.
nta
use the prefetch instruction.
plain
use the prefetch instruction (default).
t0
use the prefetcht0 instruction.
w
use the AMD-specific prefetchw instruction.
-Mnoprefetch
Disables generation of prefetch instructions.
-M[no]propcond
Enables or disables constant propagation from assertions derived from equality conditionals.
The default is enabled.
-Mr8
The compiler promotes REAL variables and constants to DOUBLE PRECISION variables and constants, respectively. DOUBLE PRECISION elements are 8 bytes in length.
-Mnor8
The compiler does not promote REAL variables and constants to DOUBLE PRECISION. REAL variables will be single precision (4 bytes in length).
-Mr8intrinsics
The compiler treats the intrinsics CMPLX and REAL as DCMPLX and DBLE, respectively.
-Mnor8intrinsics
The compiler does not promote the intrinsics CMPLX and REAL to DCMPLX and DBLE, respectively.
-Mscalarsse
Use SSE/SSE2 instructions to perform scalar floating-point arithmetic. This option is valid only on option -⁠tp [p7 | k8-32 | k8-64] targets.
-Mnoscalarsse
Do not use SSE/SSE2 instructions to perform scalar floating-point arithmetic; use x87 instructions instead. This option is not valid in combination with the -⁠tp k8-64 option.
-Msmart
instructs the compiler driver to invoke a post-pass assembly optimization utility.
-Mnosmart
instructs the compiler not to invoke an AMD64-specific post-pass assembly optimization utility.
-Munroll[=option [,option...]]
invokes the loop unroller to execute multiple instances of the loop during each iteration. This also sets the optimization level to 2 if the level is set to less than 2, or if no -⁠O or -⁠g options are supplied. The option is one of the following:
c:m
instructs the compiler to completely unroll loops with a constant loop count less than or equal to m, a supplied constant. If this value is not supplied, the m count is set to 4.
m:<n>
instructs the compiler to unroll multi-block loops n times. This option is useful for loops that have conditional statements. If n is not supplied, then the default value is 4. The default setting is not to enable -⁠Munroll=m.
n:<n>
instructs the compiler to unroll single-block loops n times, a loop that is not completely unrolled, or has a non-constant loop count. If n is not supplied, the unroller computes the number of times a candidate loop is unrolled.

To set this option in PVF, use the Fortran | Optimization | Loop Unroll Count property, described in ‘Loop Unroll Count’
-Mnounroll
instructs the compiler not to unroll loops.
-M[no]vect[=option [,option,...]]
enable [disable] the code vectorizer, where option is one of the following:
altcode
Instructs the vectorizer to generate alternate code (altcode) for vectorized loops when appropriate. For each vectorized loop the compiler decides whether to generate altcode and what type or types to generate, which may be any or all of: altcode without iteration peeling, altcode with non-temporal stores and other data cache optimizations, and altcode based on array alignments calculated dynamically at runtime. The compiler also determines suitable loop count and array alignment conditionals for executing the altcode. This option is enabled by default.
noaltcode
Instructs the vectorizer to disable alternate code generation for vectorized loops.
assoc
Instructs the vectorizer to enable certain associativity conversions that can change the results of a computation due to roundoff error. A typical optimization is to change an arithmetic operation to an arithmetic operation that is mathematically correct, but can be computationally different, due to round-off error.
noassoc
Instructs the vectorizer to disable associativity conversions.
cachesize:n
Instructs the vectorizer, when performing cache tiling optimizations, to assume a cache size of n. The default is set per processor type, either using the -⁠tp switch or auto-detected from the host computer.
[no]gather
Instructs the vectorizer to vectorize loops containing indirect array references, such as this one:
sum = 0.d0
do k=d(j),d(j+1)-1
     sum = sum + a(k)*b(c(k))
enddo
The default is gather.
partial
Instructs the vectorizer to enable partial loop vectorization through innermost loop distribution.
prefetch
Instructs the vectorizer to search for vectorizable loops and, wherever possible, make use of prefetch instructions.
[no]short
Instructs the vectorizer to enable [disable] short vector operations. -Mvect=short enables generation of packed SIMD instructions for short vector operations that arise from scalar code outside of loops or within the body of a loop iteration.
[no]sizelimit
Instructs the vectorizer to generate vector code for all loops where possible regardless of the number of statements in the loop. This overrides a heuristic in the vectorizer that ordinarily prevents vectorization of loops with a number of statements that exceeds a certain threshold. The default is nosizelimit.
smallvect[:n]
Instructs the vectorizer to assume that the maximum vector length is less than or equal to n. The vectorizer uses this information to eliminate generation of the stripmine loop for vectorized loops wherever possible. If the size n is omitted, the default is 100.
Note: No space is allowed on either side of the colon (:).
[no]sse
Instructs the vectorizer to search for vectorizable loops and, wherever possible, make use of SSE, SSE2, and prefetch instructions. The default is nosse.
[no]uniform
Instructs the vectorizer to perform the same optimizations in the vectorized and residual loops.
Note: This option may affect the performance of the residual loop.

To set this option in PVF, use the Fortran | Optimization Vectorization property, described in ‘Vectorization’
-Mnovect
instructs the compiler not to perform vectorization. You can use this option to override a previous instance of -⁠Mvect on the command-line, in particular for cases in which -⁠Mvect is included in an aggregate option such as -⁠fastsse.
-Mvect=[option]
instructs the compiler to enable loop vectorization, where option is one of the following:
partial
Enable partial loop vectorization through innermost loop distribution.
[no]short
Enable [disable] short vector operations. Enables [disables] generation of packed SIMD instructions for short vector operations that arise from scalar code outside of loops or within the body of a loop iteration.
simd[:{128|256}]
Specifies to vectorize using SIMD instructions and data, either 128 bits or 256 bits wide, on processors where there is a choice.
tile
Enable tiling/blocking over multiple nested loops for more efficient cache utilization.
-Mnovintr
instructs the compiler not to perform idiom recognition or introduce calls to hand-optimized vector functions.

2.3.6. Miscellaneous Controls

This section describes the -⁠M<pgflag> options that do not easily fit into one of the other categories of -⁠M<pgflag> options.

Default: Before looking at all the options, let’s look at the defaults. For arguments that you do not specify, the default miscellaneous options are as follows:

inform nobounds nolist warn

Related options: -⁠m, -⁠S, -⁠V, -⁠v

Usage: In the following example, the compiler includes Fortran source code with the assembly code.

 $ pgfortran -Manno -S myprog.f

In the following example, the assembler does not delete the assembly file myprog.s after the assembly pass.

 $ pgfortran -Mkeepasm myprog.f

In the following example, the compiler displays information about inlined functions with fewer than approximately 20 source lines in the source file myprog.f.

 $ pgfortran -Minfo=inline -Minline=20 myprog.f

In the following example, the compiler creates the listing file myprog.lst.

 $ pgfortran -Mlist myprog.f

In the following example, array bounds checking is enabled.

 $ pgfortran -Mbounds myprog.f

The following list provides the syntax for each miscellaneous -⁠M<pgflag> option. Each option has a description and, if appropriate, a list of any related options.

-Manno
annotate the generated assembly code with source code. Implies -⁠Mkeepasm.
To set this option in PVF, use the Fortran | Output | Annotated ASM Listing property, described in ‘Annotate Assembly’.
-Mbounds
enables array bounds checking.
  • If an array is an assumed size array, the bounds checking only applies to the lower bound.
  • If an array bounds violation occurs during execution, an error message describing the error is printed and the program terminates. The text of the error message includes the name of the array, the location where the error occurred (the source file and the line number in the source), and information about the out of bounds subscript (its value, its lower and upper bounds, and its dimension).
The following is a sample error message:
PGFTN-F-Subscript out of range for array a (a.f: 2) 
subscript=3, lower bound=1, upper bound=2, dimension=2
-Mnobounds
disables array bounds checking.
-Mbyteswapio
swap byte-order from big-endian to little-endian or vice versa upon input/output of Fortran unformatted data files.
-Mchkptr
instructs the compiler to check for pointers that are dereferenced while initialized to NULL.
-Mchkstk
instructs the compiler to check the stack for available space in the prologue of a function and before the start of a parallel region. Prints a warning message and aborts the program gracefully if stack space is insufficient.
This option is useful when many local and private variables are declared in an OpenMP program.
If the user also sets the PGI_STACK_USAGE environment variable to any value, then the program displays the stack space allocated and used after the program exits. For example, you might see something similar to the following message:
thread 0 stack: max 8180KB, used 48KB
This message indicates that the program used 48KB of a 8180KB allocated stack. This information is useful when you want to explicitly set a reserved and committed stack size for your programs, such as using the -⁠stack option on Windows.
For more information on the PGI_STACK_USAGE, refer to ‘PGI_STACK_USAGE’ in the PGI Compiler User’s Guide.
-Mcpp[=option [,option,...]]
run the PGI cpp-like preprocessor without execution of any subsequent compilation steps. This option is useful for generating dependence information to be included in makefiles.
Note: Only one of the m, md, mm or mmd options can be present; if multiple of these options are listed, the last one listed is accepted and the others are ignored.
The option is one or more of the following:
m
print makefile dependencies to stdout.
md
print makefile dependencies to filename.d, where filename is the root name of the input file being processed, ignoring system include files.
mm
print makefile dependencies to stdout, ignoring system include files.
mmd
print makefile dependencies to filename.d, where filename is the root name of the input file being processed, ignoring system include files.
[no]comment
do [do not] retain comments in output.
[suffix:]<suff>
use <suff> as the suffix of the output file containing makefile dependencies.
-Mdll
This Windows-only flag has been deprecated. Refer to -⁠Bdynamic. This flag was used to link with the DLL versions of the runtime libraries, and it was required when linking with any DLL built by any PGI compilers. This option implied -⁠D_DLL, which defines the preprocessor symbol _DLL.
-Mgccbug[s]
instructs the compiler to match the behavior of certain gcc bugs.
-Miface[=option]
adjusts the calling conventions for Fortran, where option is one of the following:
cref
uses CREF calling conventions, no trailing underscores.
mixed_str_len_arg
places the lengths of character arguments immediately after their corresponding argument. Has affect only with the CREF calling convention.
nomixed_str_len_arg
places the lengths of character arguments at the end of the argument list. Has affect only with the CREF calling convention.
-Minfo[=option [,option,...]]
instructs the compiler to produce information on standard error, where option is one of the following:
all
instructs the compiler to produce all available -⁠Minfo information. Implies a number of suboptions:
-Mneginfo=accel,inline,ipa,loop,lre,mp,opt,par,vect 
accel
instructs the compiler to enable accelerator information.
ccff
instructs the compiler to append common compiler feedback format information, such as optimization information, to the object file.
ftn
instructs the compiler to enable Fortran-specific information.
inline
instructs the compiler to display information about extracted or inlined functions. This option is not useful without either the -⁠Mextract or -⁠Minline option.
intensity
instructs the compiler to provide informational messages about the intensity of the loop. Specify <n> to get messages on nested loops.
  • For floating point loops, intensity is defined as the number of floating point operations divided by the number of floating point loads and stores.
  • For integer loops, the loop intensity is defined as the total number of integer arithmetic operations, which may include updates of loop counts and addresses, divided by the total number of integer loads and stores.
  • By default, the messages just apply to innermost loops.
ipa
instructs the compiler to display information about interprocedural optimizations.
loop
instructs the compiler to display information about loops, such as information on vectorization.
lre
instructs the compiler to enable LRE, loop-carried redundancy elimination, information.
mp
instructs the compiler to display information about parallelization.
opt
instructs the compiler to display information about optimization.
par
instructs the compiler to enable parallelizer information.
pfo
instructs the compiler to enable profile feedback information.
time
instructs the compiler to display compilation statistics.
unroll
instructs the compiler to display information about loop unrolling.
vect
instructs the compiler to enable vectorizer information.
-Minform=level
instructs the compiler to display error messages at the specified and higher levels, where level is one of the following:
fatal
instructs the compiler to display fatal error messages.
[no]file
instructs the compiler to print or not print source file names as they are compiled. The default is to print the names: -⁠Minform=file.
inform
instructs the compiler to display all error messages (inform, warn, severe and fatal).
severe
instructs the compiler to display severe and fatal error messages.
warn
instructs the compiler to display warning, severe and fatal error messages.

To set this option in PVF, use the Fortran | Diagnostics | Warning Level property, described in ‘Warning Level’.
-Minstrumentation=option
specifies the level of instrumentation calls generated. This option implies -Minfo=ccff, -Mframe.
option is one of the following:
level
specifies the level of instrumentation calls generated.
function (default)
generates instrumentation calls for entry and exit to functions.
Just after function entry and just before function exit, the following profiling functions are called with the address of the current function and its call site. (linux86-64 only).
void __cyg_profile_func_enter (void *this_fn, void *call_site);
void __cyg_profile_func_exit (void *this_fn, void *call_site);
In these calls, the first argument is the address of the start of the current function.

To set this option in PVF, use the Fortran | Diagnostics | Warning Level property, described in ‘Warning Level’.
-Mkeepasm
instructs the compiler to keep the assembly file as compilation continues. Normally, the assembler deletes this file when it is finished. The assembly file has the same filename as the source file, but with a .s extension.
To set this option in PVF, use the Fortran | Output | Assembler Output property, described in ‘Generate Assembly’.
-M list
instructs the compiler to create a listing file. The listing file is filename.lst, where the name of the source file is filename.f.
-Mmakedll
generate a dynamic link library (DLL).
-Mmakeimplib
generate an import library for a DLL without creating the DLL. When used without -def:deffile, passes the switch -def to the librarian without a deffile.
-Mnames=lowercase|uppercase
specifies the case for the names of Fortran externals.
  • lowercase - Use lowercase for Fortran externals.
  • uppercase - Use uppercase for Fortran externals.
-Mneginfo[=option [,option,...]]
instructs the compiler to produce information on standard error, where option is one of the following:
all
instructs the compiler to produce all available information on why various optimizations are not performed.
accel
instructs the compiler to enable accelerator information.
ccff
instructs the compiler to append information, such as optimization information, to the object file.
concur
instructs the compiler to produce all available information on why loops are not automatically parallelized. In particular, if a loop is not parallelized due to potential data dependence, the variable(s) that cause the potential dependence are listed in the messages that you see when using the option -⁠Mneginfo.
ftn
instructs the compiler to enable Fortran-specific information.
inline
instructs the compiler to display information about extracted or inlined functions. This option is not useful without either the -⁠Mextract or -⁠Minline option.
ipa
instructs the compiler to display information about interprocedural optimizations.
loop
instructs the compiler to display information about loops, such as information on vectorization.
lre
instructs the compiler to enable LRE, loop-carried redundancy elimination, information.
mp
instructs the compiler to display information about parallelization.
opt
instructs the compiler to display information about optimization.
par
instructs the compiler to enable parallelizer information.
pfo
instructs the compiler to enable profile feedback information.
vect
instructs the compiler to enable vectorizer information.
-Mnolist
the compiler does not create a listing file. This is the default.
-Mnoopenmp
when used in combination with the -⁠mp option, the compiler ignores OpenMP parallelization directives or pragmas, but still processes SGI-style parallelization directives or pragmas.
-Mnosgimp
when used in combination with the -⁠mp option, the compiler ignores SGI-style parallelization directives, but still processes OpenMP parallelization directives or pragmas.
-Mnopgdllmain
(Windows only) do not link the module containing the default DllMain() into the DLL. This flag applies to building DLLs with the PGFORTRAN compilers. If you want to replace the default DllMain() routine with a custom DllMain(), use this flag and add the object containing the custom DllMain() to the link line. The latest version of the default DllMain() used by PGFORTRAN is included in the Release Notes for each release. The PGFORTRAN-specific code in this routine must be incorporated into the custom version of DllMain() to ensure the appropriate function of your DLL.
-Mpreprocess
instruct the compiler to perform cpp-like preprocessing on assembly and Fortran input source files.
To set this option in PVF, use the Fortran | Preprocessor | Preprocess Source File property, described in ‘Preprocessor Definitions’.
-Mwritable_strings
stores string constants in the writable data segment.
Note: Options -⁠Xs and -⁠Xst include -⁠Mwritable_strings.

3. Directives Reference

PGI Fortran compilers support proprietary directives. These directives override corresponding command-line options. For usage information such as the scope and related command-line options, refer to the PGI Compiler User’s Guide.

This section contains detailed descriptions of PGI’s proprietary directives.

3.1. PGI Proprietary Fortran Directive Summary

Directives (Fortran comments) may be supplied by the user in a source file to provide information to the compiler. Directives alter the effects of certain command line options or default behavior of the compiler. They provide pragmatic information that control the actions of the compiler in a particular portion of a program without affecting the program as a whole. That is, while a command line option affects the entire source file that is being compiled, directives apply, or disable, the effects of a command line option to selected subprograms or to selected loops in the source file, for example, to optimize a specific area of code. Use directives to tune selected routines or loops.

The Fortran directives may have any of the following forms:

!pgi$g directive
!pgi$r directive
!pgi$l directive
!pgi$ directive

where the scope indicator follows the $ and is either g (global), r (routine), or l (loop). This indicator controls the scope of the directive, though some directives ignore the scope indicator.

Note: If the input is in fixed format, the comment character, !, * or C, must begin in column 1.

Directives override corresponding command-line options. For usage information such as the scope and related command-line options, refer to the the ‘Using Directives and Pragmas’ section of the PVF User's Guide.

3.1.1. altcode (noaltcode)

The altcode directive instructs the compiler to generate alternate code for vectorized or parallelized loops.

The noaltcode directive disables generation of alternate code.

Scope: This directive affects the compiler only when -⁠Mvect=sse or -⁠Mconcur is enabled on the command line.

!pgi$ altcode
Enables alternate code (altcode) generation for vectorized loops. For each loop the compiler decides whether to generate altcode and what type(s) to generate, which may be any or all of: altcode without iteration peeling, altcode with non-temporal stores and other data cache optimizations, and altcode based on array alignments calculated dynamically at runtime. The compiler also determines suitable loop count and array alignment conditions for executing the alternate code.
!pgi$ altcode alignment
For a vectorized loop, if possible, generates an alternate vectorized loop containing additional aligned moves which is executed if a runtime array alignment test is passed.
!pgi$ altcode [(n)] concur
For each auto-parallelized loop, generates an alternate serial loop to be executed if the loop count is less than or equal to n. If n is omitted or n is 0, the compiler determines a suitable value of n for each loop.
!pgi$ altcode [(n)] concurreduction
Sets the loop count threshold for parallelization of reduction loops to n. For each auto-parallelized reduction loop, generate an alternate serial loop to be executed if the loop count is less than or equal to n. If n is omitted or n is 0, the compiler determines a suitable value of n for each loop.
!pgi$ altcode [(n)] nontemporal
For a vectorized loop, if possible, generates an alternate vectorized loop containing non-temporal stores and other cache optimizations to be executed if the loop count is greater than n. If n is omitted or n is 1, the compiler determines a suitable value of n for each loop. The alternate code is optimized for the case when the data referenced in the loop does not all fit in level 2 cache.
!pgi$ altcode [(n)] nopeel
For a vectorized loop where iteration peeling is performed by default, if possible, generates an alternate vectorized loop without iteration peeling to be executed if the loop count is less than or equal to n. If n is omitted or n is 1, the compiler determines a suitable value of n for each loop, and in some cases it may decide not to generate an alternate unpeeled loop.
!pgi$ altcode [(n)] vector
For each vectorized loop, generates an alternate scalar loop to be executed if the loop count is less than or equal to n. If n is omitted or n is 1, the compiler determines a suitable value of n for each loop.
!pgi$ noaltcode
Sets the loop count thresholds for parallelization of all innermost loops to 0, and disables alternate code generation for vectorized loops.

3.1.2. assoc (noassoc)

This directive toggles the effects of the -⁠Mvect=noassoc command-line option, an optimization -⁠M control.

Scope: This directive affects the compiler only when -⁠Mvect=sse is enabled on the command line.

By default, when scalar reductions are present the vectorizer may change the order of operations, such as dot product, so that it can generate better code. Such transformations may change the result of the computation due to roundoff error. The noassoc directive disables these transformations.

3.1.3. bounds (nobounds)

This directive alters the effects of the -⁠Mbounds command line option. This directive enables the checking of array bounds when subscripted array references are performed. By default, array bounds checking is not performed.

3.1.4. cncall (nocncall)

This directive indicates that loops within the specified scope are considered for parallelization, even if they contain calls to user-defined subroutines or functions. A nocncall directive cancels the effect of a previous cncall.

3.1.5. concur (noconcur)

This directive alters the effects of the -⁠Mconcur command-line option. The directive instructs the auto-parallelizer to enable auto-concurrentization of loops.

Scope: This directive affects the compiler only when -⁠Mconcur is enabled on the command line.

If concur is specified, the compiler uses multiple processors to execute loops which the auto-parallelizer determines to be parallelizable. The noconcur directive disables these transformations; however, use of concur overrides previous noconcur statements.

3.1.6. depchk (nodepchk)

This directive alters the effects of the -⁠Mdepchk command line option. When potential data dependencies exist, the compiler, by default, assumes that there is a data dependence that in turn may inhibit certain optimizations or vectorizations. nodepchk directs the compiler to ignore unknown data dependencies.

3.1.7. eqvchk (noeqvchk)

The eqvchk directive specifies to check dependencies between EQUIVALENCE associated elements. When examining data dependencies, noeqvchk directs the compiler to ignore any dependencies between variables appearing in EQUIVALENCE statements.

3.1.8. invarif (noinvarif)

This directive has no corresponding command-line option. Normally, the compiler removes certain invariant if constructs from within a loop and places them outside of the loop. The directive noinvarif directs the compiler not to move such constructs. The directive invarif toggles a previous noinvarif.

3.1.9. ivdep

The ivdep directive assists the compiler's dependence analysis and is equivalent to the directive nodepchk.

3.1.10. lstval (nolstval)

This directive has no corresponding command-line option. The compiler determines whether the last values for loop iteration control variables and promoted scalars need to be computed. In certain cases, the compiler must assume that the last values of these variables are needed and therefore computes their last values. The directive nolstval directs the compiler not to compute the last values for those cases.

3.1.11. opt

The opt directive overrides the value specified by the -⁠On command line option.

The syntax of this directive is:

!pgi$<scope> opt=<level>

where the optional <scope> is r or g and <level> is an integer constant representing the optimization level to be used when compiling a subprogram (routine scope) or all subprograms in a file (global scope).

3.1.12. prefetch

The prefetch directive the compiler emits prefetch instructions whereby elements are fetched into the data cache prior to first use. By varying the prefetch distance, it is sometimes possible to reduce the effects of main memory latency and improve performance.

The syntax of this directive is:

!$mem prefetch <var1>[,<var2>[,...]]

where <varn> is any valid variable, member, or array element reference.

safe_lastval

During parallelization, scalars within loops need to be privatized. Problems are possible if a scalar is accessed outside the loop. If you know that a scalar is assigned on the last iteration of the loop, making it safe to parallelize the loop, you use the safe_lastval directive to let the compiler know the loop is safe to parallelize.

For example, use the following pragma to tell the compiler that for a given loop the last value computed for all scalars make it safe to parallelize the loop:

cpgi$l safe_lastval

The command-line option-Msafe_lastval provides the same information for all loops within the routines being compiled, essentially providing global scope.

In the following example, the value of t may not be computed on the last iteration of the loop.

do i = 1, N
     if( f(x(i)) > 5.0) then
          t = x(i)
     endif
enddo
v = t

If a scalar assigned within a loop is used outside the loop, we normally save the last value of the scalar. Essentially the value of the scalar on the "last iteration" is saved, in this case when i=N.

If the loop is parallelized and the scalar is not assigned on every iteration, it may be difficult to determine on what iteration t is last assigned, without resorting to costly critical sections. Analysis allows the compiler to determine if a scalar is assigned on every iteration, thus the loop is safe to parallelize if the scalar is used later. An example loop is:

do i = 1, N
     if( x(i) > 0.0 ) then
         t = 2.0
     else
         t = 3.0
     endif
     ...
     y(i) = t
     ...
enddo
v = t

where t is assigned on every iteration of the loop. However, there are cases where a scalar may be privatizable. If it is used after the loop, it is unsafe to parallelize. Examine this loop:

do i = 1,N
     if( x(i) > 0.0 ) then
         t = x(i)
         ...
         y(i) = t
         ...
     endif
enddo
v = t

where each use of t within the loop is reached by a definition from the same iteration. Here t is privatizable, but the use of t outside the loop may yield incorrect results since the compiler may not be able to detect on which iteration of the parallelized loop t is assigned last.

The compiler detects these cases. When a scalar is used after the loop, but is not defined on every iteration of the loop, parallelization does not occur.

3.1.14. tp

You use the directive tp to specify one or more processor targets for which to generate code.

!pgi$ tp [target]... 
Note: The tp directive can only be applied at the routine or global level. For more information about these levels, refer to the PVF User's Guide.

Refer to -tp <target>[,target...] for a list of targets that can be used as parameters to the tp directive.

3.1.15. unroll (nounroll)

The unroll directive enables loop unrolling while nounroll disables loop unrolling.

Note:

The unroll directive has no effect on vectorized loops.

The unroll directive takes arguments c, n and m.

  • c specifies that c complete unrolling should be turned on or off.
  • n specifies single block loop unrolling.
  • m specifies multi-block loop unrolling.

In addition, a constant may be specified for the c, n and m arguments.

  • c:v sets the threshold to which c unrolling applies. v is a constant; and a loop whose constant loop count is less than or equal to (<=) v is completely unrolled.
    !pgi$ unroll = c:v
  • n:v unrolls single block loops v times.
    !pgi$ unroll = n:v
  • m:v unrolls single block loops v times.
    !pgi$ unroll = m:v

The directives unroll and nounroll only apply if-⁠Munroll is selected on the command line.

3.1.16. vector (novector)

The directive novector disables vectorization. The directive vector re-enables vectorization after a previous novector directive. The directives vector and novector only apply if -⁠Mvect has been selected on the command line.

3.1.17. vintr (novintr)

The directive novintr directs the vectorizer to disable recognition of vector intrinsics. The directive vintr is re-enables recognition of vector intrinsics after a previous novintr directive. The directives vintr and novintr only apply if -⁠Mvect has been selected on the command line.

3.2. Prefetch Directives and Pragmas

Prefetch instructions can increase the speed of an application substantially by bringing data into cache so that it is available when the processor needs it. The PGI prefetch directive takes the form:

The syntax of a prefetch directive in Fortran is as follows:

!$mem prefetch <var1>[,<var2>[,...]]

where <varn> is any valid variable, member, or array element reference.

For examples on how to use the prefetch directive, refer to the Prefetch Directives section of the PVF User's Guide.

3.3. IGNORE_TKR Directive

This directive indicates to the compiler to ignore the type, kind, and/or rank (/TKR/) of the specified dummy arguments in an interface of a procedure. The compiler also ignores the type, kind, and/or rank of the actual arguments when checking all the specifics in a generic call for ambiguities.

3.3.1. IGNORE_TKR Directive Syntax

The syntax for the IGNORE_TKR directive is this:

!DIR$ IGNORE_TKR [ [(<letter>) <dummy_arg>] ... ]
<letter>
is one or any combination of the following:
T – type K – kind R – rank

For example, KR indicates to ignore both kind and rank rules and TKR indicates to ignore the type, kind, and rank arguments.

<dummy_arg>
if specified, indicates the dummy argument for which TKR rules should be ignored. If not specified, TKR rules are ignored for all dummy arguments in the procedure that contains the directive.

3.3.2. IGNORE_TKR Directive Format Requirements

The following rules apply to this directive:

  • IGNORE_TKR must not specify dummy arguments that are allocatable, Fortran 90 pointers, or assumed-shape arrays.
  • IGNORE_TKR may appear in the body of an interface block or in the body of a module procedure, and may specify dummy argument names only.
  • IGNORE_TKR may appear before or after the declarations of the dummy arguments it specifies.
  • If dummy argument names are specified, IGNORE_TKR applies only to those particular dummy arguments.
  • If no dummy argument names are specified, IGNORE_TKR applies to all dummy arguments except those that are allocatable objects, Fortran 90 pointers, or assumed-shape arrays.

3.3.3. Sample Usage of IGNORE_TKR Directive

Consider this subroutine fragment:

subroutine example(A,B,C,D)
!DIR$ IGNORE_TKR A, (R) B, (TK) C, (K) D

Table 12 indicates which rules are ignored for which dummy arguments in the preceding sample subroutine fragment:

Table 12. IGNORE_TKR Example
Dummy Argument Ignored Rules
A Type, Kind and Rank
B Only rank
C Type and Kind
D Only Kind

Notice that no letters were specified for A, so all type, kind, and rank rules are ignored.

3.4. !DEC\$ Directives

PGI Fortran compilers for Microsoft Windows support directives that help with inter-language calling and importing and exporting routines to and from DLLs. These directives all take the form:

!DEC$ directive

For specific format requirements, refer to the section ‘!DEC$ Directives’ in the PGI Compiler User's Guide.

3.4.1. ALIAS Directive

This directive specifies an alternative name with which to resolve a routine.

The syntax for the ALIAS directive is either of the following:

!DEC$ ALIAS routine_name , external_name
!DEC$ ALIAS routine_name : external_name

In this syntax, external_name is used as the external name for the specified routine_name.

If external_name is an identifier name, the name (in uppercase) is used as the external name for the specified routine_name. If external_name is a character constant, it is used as-is; the string is not changed to uppercase, nor are blanks removed.

You can also supply an alias for a routine using the ATTRIBUTES directive, described in the next section:

!DEC$ ATTIRIBUTES ALIAS : 'alias_name' :: routine_name

This directive specifies an alternative name with which to resolve a routine, as illustrated in the following code fragment that provides external names for three routines. In this fragment, the external name for sub1 is name1, for sub2 is name2, and for sub3 is name3.

subroutine sub
!DEC$ alias sub1 , 'name1'
!DEC$ alias sub2 : 'name2'
!DEC$ attributes alias : 'name3' :: sub3 

3.4.2. ATTRIBUTES Directive

This directive lets you specify properties for data objects and procedures.

The syntax for the ATTRIBUTES directive is this:

!DEC$ ATTRIBUTES <list>

where <list> is one of the following:

ALIAS : 'alias_name' :: routine_name
Specifies an alternative name with which to resolve routine_name.
C :: routine_name
Specifies that the routine routine_name will have its arguments passed by value. When a routine marked C is called, arguments, except arrays, are sent by value. For characters, only the first character is passed. The standard Fortran calling convention is pass by reference.
DLLEXPORT :: name
Specifies that name is being exported from a DLL.
DLLIMPORT :: name
Specifies that name is being imported from a DLL.
NOMIXED_STR_LEN_ARG
Specifies that hidden lengths are placed in sequential order at the end of the list.
Note: This attribute only applies to routines that are compiled with -Miface=cref or that use the default Windows calling conventions.
REFERENCE :: name
Specifies that the argument name is being passed by reference. Often this attribute is used in conjunction with STDCALL, where STDCALL refers to an entire routine; then individual arguments are modified with REFERENCE.
STDCALL :: routine_name
Specifies that routine routine_name will have its arguments passed by value. When a routine marked STDCALL is called, arguments (except arrays and characters) will be sent by value. The standard Fortran calling convention is pass by reference.
VALUE :: name
Specifies that the argument 'name' is being passed by value.

3.4.3. DECORATE Directive

The DECORATE directive specifies that the name specified in the ALIAS directive should have the prefix and postfix decorations performed on it that are associated with the calling conventions that are in effect. These declarations are the same ones performed on the name when ALIAS is not specified.

The syntax for the DECORATE directive is this:

!DEC$ DECORATE
Note: When ALIAS is not specified, this directive has no effect.

3.4.4. DISTRIBUTE Directive

This directive is front-end based, and tells the compiler at what point within a loop to split into two loops.

The syntax for the DISTRIBUTE directive is either of the following:

!DEC$ DISTRIBUTE POINT
!DEC$ DISTRIBUTEPOINT

Example:

subroutine dist(a,b,n)
    integer i
    integer n
    integer a(*)
    integer b(*)
    do i = 1,n
        a(i) = a(i)+2
!DEC$ DISTRIBUTE POINT
        b(i) = b(i)*4
    enddo
end subroutine 

4. Runtime Environment

This section describes the programming model supported for compiler code generation, including register conventions and calling conventions for x64 processor-based systems running a Windows operating system.

Note: In this section we sometimes refer to word, halfword, and double word. The equivalent byte information is word (4 byte), halfword (2 byte), and double word (8 byte).

4.1. Win64 Programming Model

This section defines compiler and assembly language conventions for the use of certain aspects of an x64 processor running a Win64 operating system. These standards must be followed to guarantee that compilers, application programs, and operating systems written by different people and organizations will work together. The conventions supported by the Fortran compiler implement the application binary interface (ABI) as defined in the AMD64 Software Conventions document.

4.1.1. Function Calling Sequence

This section describes the standard function calling sequence, including the stack frame, register usage, and parameter passing.

Register Usage Conventions

Table 13 defines the standard for register allocation. The 64-bit AMD64 and Intel 64 architectures provide a number of registers. All the general purpose registers, XMM registers, and x87 registers are global to all procedures in a running program.

Table 13. Register Allocation
Type Name Purpose
General %rax return value register
  %rbx callee-saved
  %rcx pass 1st argument to functions
  %rdx pass 2nd argument to functions
  %rsp stack pointer
  %rbp callee-saved; optional stack frame pointer
  %rsi callee-saved
  %rdi callee-saved
  %r8 pass 3rd argument to functions
  %r9 pass 4th argument to functions
  %r10-%r11 temporary registers; used in syscall/sysret instructions
  %r12-r15 callee-saved registers
XMM %xmm0 pass 1st floating point argument; return value register
  %xmm1 pass 2nd floating point argument
  %xmm2 pass 3rd floating point argument
  %xmm3 pass 4th floating point argument
  %xmm4-%xmm5 temporary registers
  %xmm6-%xmm15 callee-saved registers

In addition to the registers, each function has a frame on the run-time stack. This stack grows downward from high addresses. Table 14 shows the stack frame organization.

Table 14. Standard Stack Frame
Position Contents Frame
8n-120 (%rbp) argument eightbyte n previous
  . . .  
-80 (%rbp) argument eightbyte 5  
-88 (%rbp) %r9 home  
-96 (%rbp) %r8 home  
-104 (%rbp) %rdx home  
-112 (%rbp) %rcx home  
-120 (%rbp) return address current
-128 (%rbp) caller's %rbp  
  . . .  
0 (%rsp) variable size  

Key points concerning the stack frame:

  • The parameter area at the bottom of the stack must contain enough space to hold all the parameters needed by any function call. Space must be set aside for the four register parameters to be "homed" to the stack even if there are less than four register parameters used in a given call.
  • Sixteen-byte alignment of the stack is required except within a function’s prolog and within leaf functions.

All registers on an x64 system are global and thus visible to both a calling and a called function. Registers %rbx, %rsp, %rbp, %rsi, %rdi, %r12, %r13, %r14, and %r15 are non-volatile. Therefore, a called function must preserve these registers’ values for its caller. Remaining registers are scratch. If a calling function wants to preserve such a register value across a function call, it must save a value in its local stack frame.

Registers are used in the standard calling sequence. The first four arguments are passed in registers. Integral and pointer arguments are passed in these general purpose registers (listed in order): %rcx, %rdx, %r8, %r9. Floating point arguments are passed in the first four XMM registers: %xmm0, %xmm1, %xmm2, %xmm3. Registers are assigned using the argument’s ordinal position in the argument list. For example, if a function’s first argument is an integral type and its second argument is a floating-point type, the first argument will be passed in the first general purpose register (%rcx) and the second argument will be passed in the second XMM register (%xmm1); the first XMM register and second general purpose register are ignored. Arguments after the first four are passed on the stack.

Integral and pointer type return values are returned in %rax. Floating point return values are returned in %xmm0.

Additional registers with assigned roles in the standard calling sequence:

%rsp
The stack pointer holds the limit of the current stack frame, which is the address of the stack’s bottom-most, valid word. The stack pointer should point to a 16-byte aligned area unless in the prolog or a leaf function.
%rbp
The frame pointer, if used, can provide a way to reference the previous frames on the stack. Details are implementation dependent. A function must preserve this register value for its caller.
MXCSR
The flags register MXCSR contains the system flags, such as the direction flag and the carry flag. The six status flags (MXCSR[0:5]) are volatile; the remainder of the register is nonvolatile.
x87 - Floating Point Control Word (FPCSR)
The control word contains the floating-point flags, such as the rounding mode and exception masking. This register is initialized at process initialization time and its value must be preserved.

Signals can interrupt processes. Functions called during signal handling have no unusual restriction on their use of registers. Moreover, if a signal handling function returns, the process resumes its original execution path with registers restored to their original values. Thus, programs and compilers may freely use all registers without danger of signal handlers changing their values.

4.1.2. Function Return Values

Functions Returning Scalars or No Value

  • A function that returns an integral or pointer value that fits in 64 bits places its result in %rax.
  • A function that returns a floating point value that fits in the XMM registers returns this value in %xmm0.
  • A function that returns a value in memory via the stack places the address of this memory (passed to the function as a "hidden" first argument in %rcx) in %rax.
  • Functions that return no value (also called procedures or void functions) put no particular value in any register.
  • A call instruction pushes the address of the next instruction (the return address) onto the stack. The return instruction pops the address off the stack and effectively continues execution at the next instruction after the call instruction. A function that returns a scalar or no value must preserve the caller's registers as previously described. Further, the called function must remove the return address from the stack, leaving the stack pointer (%rsp) with the value it had before the call instruction was executed.

Functions Returning Structures or Unions

A function can use either registers or the stack to return a structure or union. The size and type of the structure or union determine how it is returned. A structure or union is returned in memory if it is larger than 8 bytes or if its size is 3, 5, 6, or 7 bytes. A structure or union is returned in %rax if its size is 1, 2, 4, or 8 bytes.

If a structure or union is to be returned in memory, the caller provides space for the return value and passes its address to the function as a "hidden" first argument in %rcx. This address will also be returned in %rax.

4.1.3. Argument Passing

Integral and Pointer Arguments

Integral and pointer arguments are passed to a function using the next available register of the sequence %rcx, %rdx, %r8, %r9. After this list of registers has been exhausted, all remaining integral and pointer arguments are passed to the function via the stack.

Floating-Point Arguments

Float and double arguments are passed to a function using the next available XMM register of the sequence %xmm0, %xmm1, %xmm2, %xmm3. After this list of registers has been exhausted, all remaining XMM floating-point arguments are passed to the function via the stack.

Array, Structure, and Union Arguments

Arrays and strings are passed to functions using a pointer to caller-allocated memory.

Structure and union arguments of size 1, 2, 4, or 8 bytes will be passed as if they were integers of the same size. Structures and unions of other sizes will be passed as a pointer to a temporary, allocated by the caller, and whose value contains the value of the argument. The caller-allocated temporary memory used for arguments of aggregate type must be 16-byte aligned.

Passing Arguments on the Stack

Registers are assigned using the argument’s ordinal position in the argument list. For example, if a function’s first argument is an integral type and its second argument is a floating-point type, the first argument will be passed in the first general purpose register (%rcx) and the second argument will be passed in the second XMM register (%xmm1); the first XMM register and second general purpose register are ignored. Arguments after the first four are passed on the stack; they are pushed on the stack in reverse order, with the last argument pushed first.

Parameter Passing

Table 15 shows the register allocation and stack frame offsets for the function declaration and call shown in the following example.

typedef struct {  
    int i;   float f; 
    } 
    struct1;  int i; float f;  double d; long l; long long ll; struct1 s1; 
    extern void 
    func (int i, float f, struct1 s1, double d, long long ll, long l);  
    func (i, f, s1, d, ll, l);
Table 15. Register Allocation for Example A-4
General Purpose Registers Floating Point Registers Stack Frame Offset
%rcx: i %xmm0: <ignored> 32: ll
%rdx: <ignored> %xmm1: f 40: l
%r8: s1.i, s1.f %xmm2: <ignored>  
%r9: <ignored> %xmm3: d  

Implementing a Stack

In general, compilers and programmers must maintain a software stack. The stack pointer, register %rsp, is set by the operating system for the application when the program is started. The stack must grow downwards from high addresses.

A separate frame pointer enables calls to routines that change the stack pointer to allocate space on the stack at run-time (e.g. alloca). Some languages can also return values from a routine allocated on stack space below the original top-of-stack pointer. Such a routine prevents the calling function from using %rsp-relative addressing to get at values on the stack. If the compiler does not call routines that leave %rsp in an altered state when they return, a frame pointer is not needed and is not used if the compiler option -⁠Mnoframe is specified.

The stack must always be 16-byte aligned except within the prolog and within leaf functions.

Variable Length Parameter Lists

Parameter passing in registers can handle a variable number of parameters. The C language uses a special method to access variable-count parameters. The stdarg.h and varargs.h files define several functions to access these parameters. A C routine with variable parameters must use the va_start macro to set up a data structure before the parameters can be used. The va_arg macro must be used to access the successive parameters.

For unprototyped functions or functions that use varargs, floating-point arguments passed in registers must be passed in both an XMM register and its corresponding general purpose register.

C Parameter Conversion

In C, for a called prototyped function, the parameter type in the called function must match the argument type in the calling function.

  • If the called function is not prototyped, the calling convention uses the types of the arguments but promotes char or short to int, and unsigned char or unsigned short to unsigned int and promotes float to double, unless you use the -⁠Msingle option.

    For more information on the -⁠Msingle option, refer to -M Options by Category.

  • If the called function is prototyped, the unused bits of a register containing a char or short parameter are undefined and the called function must extend the sign of the unused bits when needed.

Calling Assembly Language Programs

C Program Calling an Assembly-language Routine

/* File: testmain.c */
main() {
 long l_para1 = 0x3f800000;
 float f_para2 = 1.0;
 double d_para3 = 0.5;
 float f_return;
 extern float sum_3 (long para1, float para2, double para3);
 f_return = sum_3(l_para1,f_para2, d_para3);
 printf("Parameter one, type long = %08x\n",l_para1);
 printf("Parameter two, type float = %f\n",f_para2);
 printf("Parameter three, type double = %g\n",d_para3);
 printf("The sum after conversion = %f\n",f_return);
}
# File: sum_3.s
# Computes ( para1 + para2 ) + para3
 .text
 .align 16
 .globl sum_3
sum_3:
 pushq %rbp
 leaq 128(%rsp), %rbp
 cvtsi2ss %ecx, %xmm0 
 addss %xmm1, %xmm0 
 cvtss2sd %xmm0, %xmm0 
 addsd %xmm2, %xmm0
 cvtsd2ss %xmm0, %xmm0
 popq %rbp
 ret
 .type sum_3,@function
 .size sum_3,.-sum_3

4.1.4. Win64 Fortran Supplement

Sections A3.4.1 through A3.4.4 of the AMD64 Software Conventions for Win64 define the Fortran supplement. The register usage conventions set forth in that document remain the same for Fortran.

Fortran Fundamental Types

Table 16. Win64 Fortran Fundamental Types
Fortran Type Size (bytes) Alignment (bytes)
INTEGER 4 4
INTEGER*1 1 1
INTEGER*2 2 2
INTEGER*4 4 4
INTEGER*8 8 8
LOGICAL 4 4
LOGICAL*1 1 1
LOGICAL*2 2 2
LOGICAL*4 4 4
LOGICAL*8 8 8
BYTE 1 1
CHARACTER*n n 1
REAL 4 4
REAL*4 4 4
REAL*8 8 8
DOUBLE PRECISION 8 8
COMPLEX 8 4
COMPLEX*8 8 4
COMPLEX*16 16 8
DOUBLE COMPLEX 16 8

A logical constant is one of:

  • .TRUE.
  • .FALSE.

The logical constants .TRUE. and .FALSE. are defined to be the four-byte value 1 and 0 respectively. A logical expression is defined to be .TRUE. if its least significant bit is 1 and .FALSE. otherwise.

Note that the value of a character is not automatically NULL-terminated.

Fortran Naming Conventions

By default, all globally visible Fortran symbol names (subroutines, functions, common blocks) are converted to lower-case. In addition, an underscore is appended to Fortran global names to distinguish the Fortran name space from the C/C++ name space.

Fortran Argument Passing and Return Conventions

Arguments are passed by reference, meaning the address of the argument is passed rather than the argument itself. In contrast, C/C++ arguments are passed by value.

When passing an argument declared as Fortran type CHARACTER, an argument representing the length of the CHARACTER argument is also passed to the function. This length argument is a four-byte integer passed by value, and is passed at the end of the parameter list following the other formal arguments. A length argument is passed for each CHARACTER argument; the length arguments are passed in the same order as their respective CHARACTER arguments.

A Fortran function, returning a value of type CHARACTER, adds two arguments to the beginning of its argument list. The first additional argument is the address of the area created by the caller for the return value; the second additional argument is the length of the return value. If a Fortran function is declared to return a character value of constant length, for example CHARACTER*4 FUNCTION CHF(), the second extra parameter representing the length of the return value must still be supplied.

A Fortran complex function returns its value in memory. The caller provides space for the return value and passes the address of this storage as if it were the first argument to the function.

Alternate return specifiers of a Fortran function are not passed as arguments by the caller. The alternate return function passes the appropriate return value back to the caller in %rax.

The handling of the following Fortran 90 features is implementation-defined: internal procedures, pointer arguments, assumed-shape arguments, functions returning arrays, and functions returning derived types.

Inter-language Calling

Inter-language calling between Fortran and C/C++ is possible if function/subroutine parameters and return values match types. If a C/C++ function returns a value, call it from Fortran as a function, otherwise, call it as a subroutine. If a Fortran function has type CHARACTER or COMPLEX, call it from C/C++ as a void function. If a Fortran subroutine has alternate returns, call it from C/C++ as a function returning int; the value of such a subroutine is the value of the integer expression specified in the alternate RETURN statement. If a Fortran subroutine does not contain alternate returns, call it from C/C++ as a void function.

Table 17 provides the C/C++ data type corresponding to each Fortran data type.

Table 17. Fortran and C/C++ Data Type Compatibility
Fortran Type C/C++ Type Size (bytes)
CHARACTER*n x char x[n] n
REAL x float x 4
REAL*4 x float x 4
REAL*8 x double x 8
DOUBLE PRECISION x double x 8
INTEGER x int x 4
INTEGER*1 x signed char x 1
INTEGER*2 x short x 2
INTEGER*4 x int x 4
INTEGER*8 x long long x 8
LOGICAL x int x 4
LOGICAL*1 x char x 1
LOGICAL*2 x short x 2
LOGICAL*4 x int x 4
LOGICAL*8 x long long x 8

The PGI Compiler User’s Guide contains a table that provides the Fortran and C/C++ representation of the COMPLEX type.

Table 18. Fortran and C/C++ Representation of the COMPLEX Type
Fortran Type (lower case) C/C++ Type Size (bytes)
complex x struct {float r,i;} x; 8
  float complex x; 8
complex*8 x struct {float r,i;} x; 8
  float complex x; 8
double complex x struct {double dr,di;} x; 16
  double complex x; 16
complex *16 x struct {double dr,di;} x; 16
  double complex x; 16
Note: For C/C++, the complex type implies C99 or later.

Arrays

For a number of reasons inter-language function mixing is not recommended for arrays other than single dimensional arrays and square two-dimensional arrays.

  • C/C++ arrays and Fortran arrays use different default initial array index values. By default, C/C++ arrays start at 0 and Fortran arrays start at 1. However, a Fortran array can be declared to start at zero.
  • Fortran and C/C++ arrays use different storage methods. Fortran uses column-major order and C/C++ use row-major order. For one-dimensional arrays, this poses no problems. For two-dimensional arrays, where there are an equal number of rows and columns, row and column indexes can simply be reversed.

Structures, Unions, Maps, and Derived Types.

Fields within Fortran structures and derived types, and multiple map declarations within a Fortran union, conform to the same alignment requirements used by C structures.

Common Blocks

A named Fortran common block can be represented in C/C++ by a structure whose members correspond to the members of the common block. The name of the structure in C/C++ must have the added underscore. Here is an example.

Fortran common block:

	INTEGER I, J
	COMPLEX C 
	DOUBLE COMPLEX CD
	DOUBLE PRECISION D
	COMMON /COM/ i, j, c, cd, d

C equivalent:

	extern struct {
 int i;
 int j;
 struct {float real, imag;} c;
 struct {double real, imag;} cd;
 double d; 
	} com_; 

C++ equivalent:

	extern "C" struct {
 int i;
 int j;
 struct {float real, imag;} c;
 struct {double real, imag;} cd;
 double d;
	} com_;
Note: The compiler-provided name of the BLANK COMMON block is implementation-specific.

Calling Fortran COMPLEX and CHARACTER functions from C/C++ is not as straightforward as calling other types of Fortran functions. Additional arguments must be passed to the Fortran function by the C/C++ caller. A Fortran COMPLEX function returns its value in memory; the first argument passed to the function must contain the address of the storage for this value. A Fortran CHARACTER function adds two arguments to the beginning of its argument list. The following example of calling a Fortran CHARACTER function from C/C++ illustrates these caller-provided extra parameters:

CHARACTER*(*) FUNCTION CHF(C1, I)
CHARACTER*(*) C1
INTEGER I
	END
	extern void chf_();
char tmp[10];
char c1[9];
int i;
chf_(tmp, 10, c1, &i, 9);

The extra parameters tmp and 10 are supplied for the return value, while 9 is supplied as the length of c1.

5. PVF Properties

There are a number of property pages that are available in a PVF project. These property pages are grouped into categories that you can access from the Property Page dialog. Further, each of PVF’s property pages contains one or more properties, or configuration options. The set of categories and property pages available vary, depending on the type of project.

The properties in a PVF project are divided into the following categories:

  • General
  • Debugging
  • Fortran
  • Linker
  • Librarian
  • Resources
  • Build Events
  • Custom Build Step

This section contains descriptions of each of PVF’s property pages, and detailed descriptions of the properties, organized as you would see them in the Property Page dialog: by category and property page.

Tip: The Fortran, Linker, and Librarian categories contain a Command Line property page where you can see the command line derived from the properties in that category. Options that are not supported by the PVF property pages can be added to the command line from this property page by entering them in the Additional Options field.

5.1. General Property Page

This section contains the properties that are included on the General property page.

5.1.2. Output Directory

Use this property to specify a relative path to the output file directory. This directory is where the project’s output files are built.

5.1.3. Intermediate Directory

Use this property to specify a relative path to the intermediate file directory. This directory is where the intermediate files (i.e., object files) are created when the project is built.

5.1.4. Extensions to Delete on Clean

Use this property to specify which files in the intermediate directory should be deleted when the project is cleaned or before it is rebuilt. This property uses a semi-colon-delimited wildcard specification for the files.

5.1.5. Configuration Type

Use this property to change the output type that the project produces.

When you create a project, you specify the type of output that the project produces: executable, static library, or dynamic library. If you want to change the output type, use this property to do so.

5.1.6. Build Log File

Use this property to specify the build log file that is produced when the project is built.

5.1.7. Build Log Level

Use this property to specify the level of detail to be included in the build log file.

Note: Any setting above Default can produce large amounts of output and may potentially slow down the building of your project.

5.2. Debugging Property Page

This section contains the properties that are included on the Debugging property page.

5.2.2. Application Command

Use this property to specify the application to execute when you select Start Debugging or Start Without Debugging from the Debug menu.

  • If the Startup Project in your solution is a PVF project that builds an executable, there is probably no need to change this property.
  • If the Startup Project in your solution is a PVF project that builds a DLL or static library, you must use the Command property to specify an application to execute when you run (with or without debugging).
Note: To use the PVF debug engine, the Startup Project must be a PVF project. If, for example, your main executable is built by a Visual C++ project that links against a PVF project, you would designate the PVF project as the Startup Project; and in its Debugging | Application Command property, you would specify the path to the executable built by the Visual C++ project.
Tip: The Startup Project is the project listed in boldface in the solution explorer. You can change the Startup Project by right-clicking on any project in the solution explorer and selecting Set as Startup Project from the context menu.

5.2.3. Application Arguments

Use this property to pass command line arguments to the application when it is run or debugged.

5.2.4. Environment

Use this property to specify any environment variables to set for the application when it runs. One common use of this property is to augment the PATH environment variable. For example, if the application requires DLLs to run but the general environment is not set to find these, the path to these DLLs could be added to the PATH environment variable.

For more information on PATH, refer to the PVF User's Guide.

If the Merge Environment property is set to Yes, then the contents of the Environment property are merged with the existing environment when the application is run or debugged.

5.2.5. Merge Environment

Use this property to merge the environment variables in the Environment property with the existing environment when the application is run or debugged. To do this, set the Merge Environment property to Yes.

5.2.6. Accelerator Profiling

Use this property to generate accelerator profiling information at runtime. To do this, set the Accelerator Profiling property to Yes.

Setting this property to Yes sets the PGI_ACC_TIME environment variable to 1.

5.2.7. MPI Debugging

Use this property to enable MPI debugging and select local MPI debugging.

The value selected for this property determines which properties are displayed following it on the Debugging property page.

Important: If you change the value of this property and the displayed properties do not change, be sure to click Apply in the property page dialog box.
  • When MPI Debugging is set to Disabled, the application is run or debugged in serial mode.
  • When MPI Debugging is set to Local, the application is run or debugged using mpiexec. All processes launched are local to the system on which the application is run.

5.2.8. Working Directory

[Serial]

Use this property to specify the application's working directory when it is run or debugged serially. By default, the working directory is set to the solution directory.

This property is displayed when the MPI Debugging property is set to Disabled.

5.2.9. Number of Processes

[Local MPI]

Use this property to specify the number of MPI processes to use when the application is run or debugged. The number of processes is passed to mpiexec using the -n option.

This property is displayed when the MPI Debugging property is set to Local.

5.2.10. Working Directory

[Local MPI]

Use this property to specify the application's working directory when it is run or debugged using local MPI. By default, the working directory is set to the solution directory.

This property is displayed when the MPI Debugging property is set to Local.

5.2.11. Additional Arguments: mpiexec

[Local MPI]

Use this property to specify additional arguments to be passed to mpiexec when the application is run or debugged.

This property is displayed when the MPI Debugging property is set to Local.

5.2.12. Location of mpiexec

[Local MPI]

Use this property to override the default path to mpiexec as specified in the system PATH variable.

This property is displayed when the MPI Debugging property is set to Local.

5.3. Fortran Property Pages

This section contains the property pages that are included in the Fortran category. This category is further divided into the following property pages, displayed in the following order:

  • General
  • Optimization
  • Preprocessing
  • Code Generation
  • Language
  • Floating Point Options
  • External Procedures
  • Target Processors
  • Target Accelerators
  • Diagnostics
  • Profiling
  • Command Line

The following sections describe the properties available on each property page.

5.4. Fortran | General

The following properties are available from the Fortran | General property page.

5.4.1. Display Startup Banner

Use this property to determine whether to display the compiler’s startup banner during compilation.

Changing the property to Yes adds the -⁠V switch to the compilation line, which causes the compiler to display the startup banner during compilation.

For more information on -⁠V, refer to -V[release_number].

5.4.2. Additional Include Directories

Use this property to add one or more directories to the compiler’s include path.

For every path that is added to this property, PVF adds -⁠I<path> to the compilation line.

There are two ways to add directories to this property:

  • Type the information directly into the property page box.

    Use a semi-colon (‘;’) to separate each directory.

  • Click the ellipsis (‘...’) button in the property page box to open the Additional Include Directories dialog box.

    Enter each directory on its own line in this box. Do not use semi-colons to separate directories; the semi-colons are added automatically when the box is closed.

Note: This property is also available from the Fortran | Preprocessing Property page.

5.4.3. Module Path

Use this property to specify the location of module (.mod) files.

For every directory that is added to this property, PVF adds -⁠module <dir> to the compilation line, causing the compiler to search each listed directory for modules during compilation.

Note: The first directory in the list is also the module output directory, which is where PVF puts all module files created when the project is built.

There are two ways to add directories to this property:

  • Type the information directly into the property page box.

    Use a semi-colon (‘;’) to separate each path.

  • Click the ellipsis (‘...’) button in the property page box to open the Module Path dialog box.

    Enter each directory on its own line in this box. Do not use semi-colons to separate entries; the semi-colons are added automatically when the box is closed.

5.4.4. Object File Name

Use of this property depends on whether it is being applied to a file or a project:

  • File level: Use this property to set the name of the object file. Setting the name adds the -⁠o switch to the compilation line.

    For more information on -⁠o, refer to -o.

  • Project level: Use this property to set the location of the object files created by a build.

    To change the default location for the object files, specify a different directory name for this property.

    Note: You must append a backslash (\) to the directory path or the value of this property will be interpreted as a file.

5.4.5. Debug Information Format

Use this property to specify whether the compiler should generate debug information and if so, in what format.

  • The richest debugging experience is obtained when this option is set to "Full Debug Information (-g).’
  • If you are debugging a project built with optimizations, you may want to select "Full Debug Information with Full Optimization (-gopt)." This selection prevents the generation of debug information from affecting optimizations.

For more information on -⁠g, refer to -g. For more information on -⁠gopt, refer to -gopt.

5.4.6. Optimization

Use this property to select the overall code optimization.

This property can be set to one of the following values:

  • No Optimization - the default value for Debug configurations.
  • Maximize Speed - the default value for Release Configurations.
  • Maximize Speed Across the Whole Program
Note: This property is also available from the Fortran | Optimization Property page.

5.5. Fortran | Optimization

The following properties are available from the Fortran | Optimization property page.

5.5.1. Optimization

Use this property to select the overall code optimization.

This property can be set to one of the following values:

  • No Optimization - the default value for Debug configurations.
  • Maximize Speed - the default value for Release Configurations.
  • Maximize Speed Across the Whole Program
Note: This property is also available from the Fortran | General Property page.

5.5.2. Global Optimizations

Use this property to set the compiler’s global optimization level.

Setting this property adds one of the -⁠O options to the compilation line.

For more information on -⁠O, refer to -O<level>.

5.5.3. Vectorization

Use this property to specify the type of vectorization to perform.

The PVF compilers use the -⁠Mvect options to vectorize code that is vectorizable. Select the appropriate vectorization from these options:

  • Default: Accepts the default vectorization.
  • Enable Vectorization: Enables vectorization by adding the -⁠Mvect switch to the PVF compilation and link lines.
  • Vectorize using SSE instructions: Enables vectorization using SSE instructions by adding the -⁠Mvect=sse switch to the PVF compilation line.
  • Vectorize using SIMD instructions: Enables vectorization using SIMD instructions and data, by adding the -⁠Mvect=simd switch to the PVF compilation line.
  • Vectorize using 128-bit SIMD instructions: Enables vectorization using SIMD 128-bit instructions and data, by adding the -⁠Mvect=simd:128 switch to the PVF compilation line.
  • Vectorize using 256-bit SIMD instructions: Enables vectorization using SIMD 256-bit instructions and data, by adding the -⁠Mvect=simd:256 switch to the PVF compilation line.

For more information on -⁠Mvect, refer to Optimization Controls.

5.5.4. Inlining

Use this property to enable inlining of certain subprograms.

Setting this property to Yes adds the -⁠Minline switch to the compilation command line.

For more information on -⁠Minline, refer to -Minline[=option[,option,...]].

5.5.5. Use Frame Pointer

Use this property to specify whether to generate code that uses a frame pointer.

Setting this property to Yes adds the -⁠Mframe switch to the compilation command line and PVF compilers generate code that uses a frame pointer.

Setting this property to No, the default, adds the -⁠Mnoframe switch to the compilation command line and PVF compilers generate code that does not use frame pointers.

For more information on -⁠Mframe, refer to Optimization Controls.

5.5.6. Loop Unroll Count

Use this property to select the appropriate value for unrolling.

Loop unrolling is a common optimization. This property allows you to specify unrolling by two or four. Using this option adds the -⁠Munroll option to the compilation line.

For more information on -⁠Munroll, refer to Optimization Controls.

5.5.7. Auto-Parallelization

Use this property to auto-parallelize code that is parallelizable. Using this option adds the -⁠Mconcur option to the compilation line.

For more information on -⁠Mconcur, refer to Optimization Controls.

5.6. Fortran | Preprocessing

The following properties are available from the Fortran | Preprocessing Property page.

5.6.1. Preprocess Source File

Use this property to specify whether the compiler should preprocess source files.

Setting this property to Yes adds the -⁠Mpreprocess switch to the compilation command line.

For more information on -⁠Mpreprocess, refer to Miscellaneous Controls.

5.6.2. Additional Include Directories

Use this property to add one or more directories to the compiler’s include path.

For every path that is added to this property, PVF adds -⁠I<path> to the compilation line.

There are two ways to add directories to this property:

  • Type the information directly into the property page box.

    Use a semi-colon (‘;’) to separate each directory.

  • Click the ellipsis (‘...’) button in the property page box to open the Additional Include Directories dialog box.

    Enter each directory on its own line in this box. Do not use semi-colons to separate directories; the semi-colons are added automatically when the box is closed.

For more information on -⁠I<path>, refer to -I.

Note: This property is also available from the Fortran | General Property page.

5.6.3. Ignore Standard Include Path

Use this property to specify whether the preprocessor should ignore the standard include path.

Setting this property to Yes adds the -⁠Mnostdinc switch to the compilation command line.

For more information on -⁠Mnostdinc, refer to Environment Controls.

5.6.4. Preprocessor Definitions

Use this property to add one or more preprocessor definitions to compilation.

For every definition that is added to this property, PVF adds -⁠D<definition> to the compilation line.

There are two ways to add definitions to this property:

  • Type the information directly into the property page box.

    Use a semi-colon (‘;’) to separate each definition.

    For example, DEF1;DEF2=2 defines DEF1, and defines DEF2 and initializes it to 2.

  • Click the ellipsis (‘...’) button in the property page box to open the Preprocessor Definitions dialog box.

    Enter each definition on its own line in this box. Do not use semi-colons to separate definitions; the semi-colons are added automatically when the box is closed.

For more information on -⁠D<definition>, refer to -D.

5.6.5. Undefine Preprocessor Definitions

Use this property to undefine one or more preprocessor definitions.

For every definition that is added to this property, PVF adds -⁠U<definition> to the compilation line.

There are two ways to add definitions to this property:

  • Type the information directly into the property page box.

    Use a semi-colon (‘;’) to separate each definition.

    For example, DEF1;DEF2 undefines DEF1 and DEF2.

  • Click the ellipsis (‘...’) button in the property page box to open the Undefine Preprocessor Definitions dialog box.

    Enter each definition on its own line in this box. Do not use semi-colons to separate definitions; the semi-colons are added automatically when the box is closed.

For more information on -⁠U<definition>, refer to -U.

5.7. Fortran | Code Generation

The following properties are available from the Fortran | Code Generation property page.

5.7.1. Runtime Library

Use this property to specify the type of runtime libraries to use during linking.

Default: Depends on the project:

  • For executable and static library projects: multi-threaded static libraries.

    Using this option adds the -⁠Bstatic option to the compilation line. This choice corresponds to Microsoft’s /MT compilation option.

    For more information on -⁠Bstatic, refer to -Bstatic.

  • For dynamic-link library projects: multi-threaded DLL libraries.

    Using this option adds the -⁠Bdynamic option to the compilation line. This choice corresponds to Microsoft’s /MD compilation option.

    For more information on -⁠Bdynamic, refer to -Bdynamic.

Note: It is important to keep the type of runtime libraries consistent within a solution. PVF projects that build DLLs should link to the multi-threaded DLL runtime, and projects that link to these PVF DLLs should also use the multi-threaded DLL runtime.

5.8. Fortran | Language

The following properties are available from the Fortran | Language property page.

5.8.1. Fortran Dialect

Use this property to select the Fortran dialect to use during compilation.

PVF supports two Fortran language dialects: Fortran 95 and FORTRAN 77. The dialect determines which PGI compiler driver is used during compilation.

  • Default: The dialect is set to Fortran 95, even for fixed-format .f files, and the pgfortran driver is used.
  • Fortran 77: Use the pgf77 driver. You can select the FORTRAN 77 dialect at the project or file level.

5.8.2. Treat Backslash as Character

Use this property to specify how the compilers should treat the backslash (\) character.

Default: PVF treats the backslash (\) as a regular character.

This default action is equivalent to adding the -⁠Mbackslash switch to compilation.

If you want the backslash character to be treated as an escape character, which is how C and C++ compilers handle backslashes, set this property to No.

For more information on -⁠Mbackslash, refer to Fortran Language Controls.

5.8.3. Extend Line Length

Use this property to extend the line length for fixed-format Fortran files to 132 characters.

Fixed-format Fortran files limit the accepted line length to 72 characters. To extend the line length for these types of files to 132 characters, set this property to Yes, which adds the -⁠Mextend switch to the PVF compilation line.

For more information on -⁠Mextend, refer to Fortran Language Controls.

5.8.4. Enable OpenMP Directives

Use this property to enable OpenMP 3.0 directives.

Setting this property to Yes adds the -⁠mp switch to the PVF compilation and link lines.

For more information on -⁠mp, refer to -mp.

5.8.5. Enable OpenACC Directives

Use this property to enable OpenACC directives.

Setting this property to Yes adds the -⁠acc switch to the PVF compilation and link lines and activates access to these additional properties:

For more information on -⁠acc, refer to -acc.

5.8.6. OpenACC Autoparallelization

When Enable OpenACC Directives is set to Yes, use this property to control loop autoparallelization within acc parallel.

  • Default: Allows the compiler to control loop autoparallelization within acc parallel. This selection adds no additional sub-options to -acc.
  • Yes: Directs the compiler to enable loop autoparallelization within an OpenACC parallel region (-acc=autopar).
  • No: Directs the compiler to disable loop autoparallelization within an OpenAcc parallel region (-acc=noautopar).

5.8.7. OpenACC Required

When Enable OpenACC Directives is set to Yes, use this property to control the compiler’s behavior when it is unable to accelerate a compute region.

  • Default: Use the compiler defaults for handling instances where compute regions cannot be accelerated. This selection adds no additional sub-options to -acc.
  • Yes: Directs the compiler to stop compilation with an error when it cannot accelerate a compute region (-acc=required).
  • No: Directs the compiler to issue warnings when it cannot accelerate a compute region; compilation does not stop but accelerator kernels are not generated (-acc=norequired).

5.8.8. OpenACC Routineseq

When Enable OpenACC Directives is set to Yes, use this property to compile every routine for the device.

  • Default: Uses compiler defaults handling compile every routine for the device. This selection adds no additional sub-options to -acc.
  • Yes: Enables compiling every routine for the device by adding -acc=routineseq switch to the PVF compilation and link lines.
  • No: Disables compiling every routine for the device by adding -acc=noroutineseq switch to the PVF compilation and link lines.

5.8.9. OpenACC Wait

When Enable OpenACC Directives is set to Yes, use this property to wait for each device kernel to finish.

  • Default: Uses compiler defaults handling wait for each device kernel to finish. This selection adds no additional sub-options to -acc.
  • Yes: Enables wait for each device kernel to finish by adding -acc=wait switch to the PVF compilation and link lines.
  • No: Disables wait for each device kernel to finish by adding -acc=nowait switch to the PVF compilation and link lines.

5.8.10. OpenACC Conformance Level

When Enable OpenACC Directives is set to Yes, use this property to leverage the compiler’s detection of extensions to standard OpenACC directives.

  • Default: When non-OpenACC accelerator directives are found, they are ignored..
  • Strict: Add -acc=strict to the PVF compilation and link lines. The compiler emits a warning when a non-OpenACC accelerator directive is found.
  • Very Strict: Add -acc=strict to the PVF compilation and link lines. The compiler stops with an error when a non-OpenACC accelerator directive is found.

5.8.11. OpenACC Sync

When Enable OpenACC Directives is set to Yes, use this property to ignore async clauses.

Setting this property to Yes adds the -acc=sync switch to the PVF compilation and link lines.

5.8.12. MPI

Use this property to enable compilation and linking using the Microsoft MPI headers and libraries.

Setting this property to Microsoft MPI adds the -Mmpi=msmpi switch to the PVF compilation and link lines.

5.8.13. Enable CUDA Fortran

Use this property to enable CUDA Fortran.

Setting this property to Yes adds the -⁠Mcuda switch to the PVF compilation and link lines and activates access to these additional properties:

Important: If you select Yes and the additional properties do not appear, click Apply in the Property page dialog.

For more information on -⁠Mcuda, refer to Optimization Controls.

5.8.14. CUDA Fortran Register Limit

When Enable CUDA Fortran is set to Yes, use this property to specify the number of registers to use on the GPU.

Setting this property to an integer value, n, adds the -⁠Mcuda=maxregcount:n switch to the PVF compilation and link lines.

Leaving this property blank indicates no limit to the number of registers to use on the GPU.

For more information on -⁠Mcuda, refer to Optimization Controls.

5.8.15. CUDA Fortran Use Fused Multiply-Adds

When Enable CUDA Fortran is set to Yes, use this property to control the generation of fused multiply-add (FMA) instructions.

  • Default: Allows the compiler to control generation of FMA instructions. This selection adds no additional sub-options to -Mcuda.
  • Yes: Enables generation of FMA instructions by adding -Mcuda=fma switch to the PVF compilation and link lines.
  • No: Disables generation of FMA instructions by adding -Mcuda=nofma switch to the PVF compilation and link lines.

For more information on -⁠Mcuda, refer to Optimization Controls.

5.8.16. CUDA Fortran Use Fast Math Library

When Enable CUDA Fortran is set to Yes, use this property to use routines from the fast math library.

Setting this property to Yes adds the -⁠Mcuda=fastmath switch to compilation and linking.

For more information on -⁠Mcuda, refer to Optimization Controls.

5.8.17. CUDA Fortran Debug

When Enable CUDA Fortran is set to Yes, use this property to control generatation of GPU debug information.

  • Default: Allows the compiler to control generatation of GPU debug information. This selection adds no additional sub-options to -⁠Mcuda.
  • Yes: Enables generatation of GPU debug information by adding the -⁠Mcuda=debug switch to the PVF compilation and link lines.
  • No: Disables generatation of GPU debug information by adding the -⁠Mcuda=nodebug switch to the PVF compilation and link lines.

For more information on -⁠Mcuda, refer to Optimization Controls.

5.8.18. CUDA Fortran Line Information

When Enable CUDA Fortran is set to Yes, use this property to control generatation of GPU line information.

  • Default: Allows the compiler to control generatation of GPU line information. This selection adds no additional sub-options to -⁠Mcuda.
  • Yes: Enables generatation of GPU line information by adding the -⁠Mcuda=lineinfo switch to the PVF compilation and link lines.
  • No: Disables generatation of GPU line information by adding the -⁠Mcuda=nolineinfo switch to the PVF compilation and link lines.

For more information on -⁠Mcuda, refer to Optimization Controls.

5.8.19. CUDA Fortran Use LLVM Back End

When Enable CUDA Fortran is set to Yes, use this property to control using LLVM back end.

  • Default: Allows the compiler to control using LLVM back end. This selection adds no additional sub-options to -⁠Mcuda.
  • Yes: Use LLVM back end by adding the -⁠Mcuda=llvm switch to the PVF compilation and link lines.
  • No: Use CUDA C back end by adding the -⁠Mcuda=nollvm switch to the PVF compilation and link lines.

For more information on -⁠Mcuda, refer to Optimization Controls.

5.8.20. CUDA Fortran Unroll

When Enable CUDA Fortran is set to Yes, use this property to control automatic inner loop unrolling.

  • Default: Allows the compiler to control automatic inner loop unrolling. This selection adds no additional sub-options to -⁠Mcuda.
  • Yes: Enables automatic inner loop unrolling by adding the -⁠Mcuda=unroll switch to the PVF compilation and link lines.
  • No: Disables automatic inner loop unrolling by adding the -⁠Mcuda=nounroll switch to the PVF compilation and link lines.

For more information on -⁠Mcuda, refer to Optimization Controls.

5.8.21. CUDA Fortran Flush to Zero

When Enable CUDA Fortran is set to Yes, use this property to control flush-to-zero mode for floating point computations on in GPU code generated for CUDA Fortran kernels.

  • Default: Accepts the default handling of floating point computations in the GPU code generated for CUDA Fortran kernels.
  • Yes: Enables flush-to-zero mode by adding the -⁠Mcuda=flushz switch to the PVF compilation and link lines.
  • No: Disables flush-to-zero mode by adding the -⁠Mcuda=noflushz switch to the PVF compilation and link lines.

For more information on -⁠Mcuda, refer to Optimization Controls.

5.8.22. CUDA Fortran Toolkit

When Enable CUDA Fortran is set to Yes, use this property to specify the version of the CUDA toolkit that is targeted by the compilers.

  • Default: The compiler selects the default CUDA toolkit version.
  • 7.5: Use the default version 7.5 of the CUDA toolkit. This selection adds the -⁠Mcuda=cuda7.5 switch to the PVF compilation and link lines.
  • 8.0: Use version 8.0 of the CUDA toolkit. This selection adds the -⁠Mcuda=cuda8.0 switch to the PVF compilation and link lines.
Note:pgaccelinfo prints the driver version as the first line of output.
  • For a 7.5 driver: CUDA Driver Version 7050
  • For a 8.0 driver: CUDA Driver Version 8000

For more information on -⁠Mcuda, refer to Fortran Language Controls.

5.8.23. CUDA Fortran Compute Capability

When Enable CUDA Fortran is set to Yes, use this property to either automatically generate code compatible with all applicable compute capabilities, or to direct the compiler to use a manually-selected set.

Select either Automatic or Manual.

  • Automatic: Let the compiler generate code for all applicable compute capabilities. This is the default.
  • Manual: Choose one or more compute capabilities to target. The compiler generates code for each capability specified.

    If you select Manual, then you can select any or all of the following compute capabilities that are described in the next sections.

Important: If you select Manual and the additional properties do not appear, click Apply in the Property page dialog.

For more information on -⁠Mcuda, refer to Optimization Controls.

5.8.24. CUDA Fortran Fermi

When Enable CUDA Fortran is set to Yes and CUDA Fortran Compute Capability is set to Manual, use this property to generate code for the Fermi architecture.

Setting this property to Yes adds the -⁠Mcuda=fermi switch to the PVF compilation and link lines.

5.8.25. CUDA Fortran Fermi+

When Enable CUDA Fortran is set to Yes and CUDA Fortran Compute Capability is set to Manual, use this property to generate code for Fermi architecture and above.

Setting this property to Yes adds the -⁠Mcuda=fermi+ switch to the PVF compilation and link lines.

5.8.26. CUDA Fortran Kepler

When Enable CUDA Fortran is set to Yes and CUDA Fortran Compute Capability is set to Manual, use this property to generate code for the Kepler architecture.

Setting this property to Yes adds the -⁠Mcuda=kepler switch to the PVF compilation and link lines.

5.8.27. CUDA Fortran Kepler+

When Enable CUDA Fortran is set to Yes and CUDA Fortran Compute Capability is set to Manual, use this property to generate code for Kepler architecture and above.

Setting this property to Yes adds the -⁠Mcuda=kepler+ switch to the PVF compilation and link lines.

5.8.28. CUDA Fortran Keep Binary

Use this property to keep the CUDA binary (.bin) file.

Setting this property to Yes adds the -⁠Mcuda=keepbin switch to the PVF compilation and link lines.

For more information on -⁠Mcuda, refer to Optimization Controls.

5.8.29. CUDA Fortran Keep Kernel Source

When Enable CUDA Fortran is set to Yes, use this property to keep the kernel source files.

Setting this property to Yes adds the -⁠Mcuda=keepgpu switch to the PVF compilation and link lines.

For more information on -⁠Mcuda, refer to Optimization Controls.

5.8.30. CUDA Fortran Keep PTX

When Enable CUDA Fortran is set to Yes, use this property to keep the portable assembly (.ptx) file for the GPU code.

Setting this property to Yes adds the -⁠Mcuda=keepptx switch to the PVF compilation and link lines.

For more information on -⁠Mcuda, refer to Optimization Controls.

5.8.31. CUDA Fortran Keep PTXAS

Use this property to show PTXAS informational messages during compilation.

Setting this property to Yes adds the -⁠Mcuda=ptxinfo switch to the PVF compilation and link lines.

For more information on -⁠Mcuda, refer to Optimization Controls.

5.8.32. CUDA Fortran Generate RDC

Use this property to generate relocatable device code (-Mcuda=rdc).

Setting this property to Yes adds the -⁠Mcuda=rdc switch to the PVF compilation and link lines.

For more information on -⁠Mcuda, refer to Optimization Controls.

5.8.33. CUDA Fortran Emulation

When Enable CUDA Fortran is set to Yes, use this property to enable CUDA Fortran emulation mode.

Setting this property to Yes adds the -⁠Mcuda=emu switch to the PVF compilation and link lines.

For more information on -⁠Mcuda, refer to Optimization Controls.

5.8.34. CUDA Fortran Madconst

When Enable CUDA Fortran is set to Yes, use this property to control putting module array descriptors in CUDA constant memory.

Setting this property to Yes adds the -⁠Mcuda=madconst switch to the PVF compilation and link lines.

For more information on -⁠Mcuda, refer to Optimization Controls.

5.9. Fortran | Floating Point Options

The following properties are available from the Fortran | Floating Point Options property page.

5.9.1. Floating Point Exception Handling

Use this property to enable floating point exceptions.

Setting this property to Yes adds the -⁠Ktrap=fp option to compilation.

For more information on -⁠Ktrap, refer to -K<flag>.

5.9.2. Floating Point Consistency

Use this property to enable relaxed floating point accuracy in favor of speed.

Setting this property to Yes adds the -⁠Mfprelaxed option to compilation.

For more information on -⁠Mfprelaxed, refer to Optimization Controls.

5.9.3. Flush Denormalized Results to Zero

Use this property to specify how to handle denormalized floating point results.

  • Default: Accepts the default handling of denormalized floating point results.
  • Yes: Enables SSE flush-to-zero mode using the -⁠Mflushz compilation option.
  • No: Disables SSE flush-to-zero mode using the -⁠Mnoflushz compilation option.

For more information on -⁠M[no]flushz, refer to Code Generation Controls.

5.9.4. Treat Denormalized Values as Zero

Use this property to specify how to treat denormalized floating point values.

  • Default: Accept the default treatment of denormalized floating point values.
  • Yes: Enable the treatment of denormalized floating point values as zero using the -⁠Mdaz compilation option.
  • No: Disable the treatment of denormalized floating point values as zero using the -⁠Mnodaz compilation option.

For more information on -⁠M[no]daz, refer to Code Generation Controls.

5.9.5. IEEE Arithmetic

Use this option to specify IEEE floating point arithmetic.

  • Default: Accept the default floating point arithmetic.
  • Yes: Enable IEEE floating point arithmetic using the -⁠Kieee compilation option.
  • No: Disable IEEE floating point arithmetic using the -⁠Knoieee compilation option.

For more information on -⁠K[no]ieee, refer to -K<flag>.

5.10. Fortran | External Procedures

The following properties are available from the Fortran | External Procedures property page.

5.10.1. Calling Convention

Use this property to specify an alternate Fortran calling convention.

  • Default: Accept the default calling convention.
  • C By Reference: Use the CREF calling convention. Adds -⁠Miface=cref to compilation. On x64 platforms, no trailing underscores are used with this option and this option also causes Fortran externals to be uppercase and lengths of character arguments to be put at the end of the argument list.

For more information on -⁠Miface, refer to Miscellaneous Controls.

5.10.2. String Length Arguments

Use this property to change where string length arguments are placed in the argument list.

  • Default: Use the calling convention's default placement for passing string length arguments.
  • After Every String Argument: Lengths of character arguments are placed immediately after their corresponding argument. This option adds -⁠Miface=mixed_str_len_arg to compilation.
  • After All Arguments: Places lengths of character arguments at the end of the argument list. This option adds -⁠Miface=nomixed_str_len_arg to the compilation.
Note: The After Every String Argument and After All Arguments options only have an effect when using the C By Reference calling convention.

For more information on -⁠Miface, refer to Miscellaneous Controls.

5.10.3. Case of External Names

Use this property to specify the case used for Fortran external names.

  • Default: Use the calling convention's default case for external names.
  • Lower Case: Make Fortran external names lower case. This option adds -⁠Mnames=lowercase to the compilation.
  • Upper Case: Make Fortran external names upper case. This option adds -⁠Mnames=uppercase to the compilation.
Note: The Lower Case and Upper Case options only have an effect when using the C By Reference calling convention.

5.11. Fortran | Libraries

The Fortran | Libraries property page contains properties that make it easier to use third-party libraries. To use these libraries, however, the appropriate binaries, such as .lib and .dll files, must be installed on your system.

5.11.1. Use MKL

Use this property to build for and link against the Intel Math Kernel Library (MKL), which is available from Intel.

  • Yes: Use the Intel Math Kernel Library when building and linking programs.
  • No: Do not use the Intel Math Kernel Library when building and linking programs.

5.12. Fortran | Target Processors

The properties that are available from the Fortran | Target Processors property page depend on the platform you are using. The platform selection box in the center of the Property Pages dialog indicates the platform: x64.

Note:
x64 Platform
You can target multiple processors for optimization on the x64 platform.

The Target Processors properties add the -⁠tp=<target> option to compilation. For more information on the -⁠tp switch referenced throughout the following descriptions, refer to -⁠tp [k8-64].

5.12.1. AMD Athlon

Use this property to optimize for AMD Athlon64, AMD Opteron and compatible processors.

x64: Setting this property to Yes adds the -⁠tp=k8-64 switch to compilation.

5.12.2. AMD Barcelona

Use this property to optimize for AMD Opteron/Quadcore and compatible processors.

x64: Setting this property to Yes adds the -⁠tp=barcelona-64 switch to compilation.

5.12.3. AMD Bulldozer

Use this property to optimize for AMD Bulldozer and compatible processors.

x64: Setting this property to Yes adds the -⁠tp=bulldozer-64 switch to compilation.

5.12.4. AMD Istanbul

Use this property to optimize for AMD Istanbul processor-based systems.

x64: Setting this property to Yes adds the -⁠tp=istanbul-64 switch to compilation.

5.12.5. AMD Piledriver

Use this property to optimize for AMD Piledriver processor-based systems.

x64: Setting this property to Yes adds the -⁠tp=piledriver-64 switch to compilation.

5.12.6. AMD Shanghai

Use this property to optimize for AMD Shanghai processor-based systems.

x64: Setting this property to Yes adds the -⁠tp=shanghai-64 switch to compilation.

5.12.7. Intel Core 2

Use this property to optimize for Intel Core 2 Duo and compatible processors.

x64: Setting this property to Yes adds the -⁠tp=core2-64 switch to compilation.

5.12.8. Intel Core i7

Use this property to optimize for Intel Core i7 (Nehalem) processor-based systems.

x64: Setting this property to Yes adds the -⁠tp=nehalem-64 switch to compilation.

5.12.9. Intel Penryn

Use this property to optimize for Intel Penryn architecture and compatible processors.

x64: Setting this property to Yes adds the -⁠tp=penryn-64 switch to compilation.

5.12.10. Intel Pentium 4

Use this property to optimize for Intel Pentium 4 and compatible processors.

5.12.11. Intel Sandy Bridge

Use this property to optimize for Intel Sandy Bridge architecture and compatible processors.

x64: Setting this property to Yes adds the -⁠tp=sandybridge-64 switch to compilation.

5.12.12. Generic x86-64 [x64 only]

Use this property to optimize for any x86-64 processor-based system.

x64: Setting this property to Yes adds the -⁠tp=px-64 switch to compilation.

5.13. Fortran | Target Accelerators

The following properties are available from the Fortran | Target Accelerators property page.

For more information about the PGI’s accelerator compilers or on the options in this section, refer to the PVF User's Guide.

5.13.1. Target NVIDIA Tesla

Use this property to select NVIDIA Tesla targets.

Setting this property to Yes adds the -⁠ta=tesla switch to the PVF compilation and link lines and activates access to these additional properties:

Important: If you change the value of this property and the displayed properties do not change, be sure to click Apply in the property page dialog box.

5.13.2. Tesla Register Limit

Use this property to specify the number of registers to use on the GPU.

Setting this property to an integer value, n, adds the -⁠ta=tesla:maxregcount:n switch to the PVF compilation and link lines.

Leaving this property blank indicates no limit to the number of registers to use on the GPU.

5.13.3. Tesla Use Fused Multiply-Adds

When Target NVIDIA Tesla is set to Yes, use this property to control the generation of fused multiply-add (FMA) instructions.

  • Default: Allows the compiler to control generation of FMA instructions. This selection adds no additional sub-options to -⁠ta=tesla.
  • Yes: Enables generation of FMA instructions by adding -⁠ta=tesla:fma to the PVF compilation and link lines.
  • No: Disables generation of FMA instructions by adding -⁠ta=tesla:nofma to the PVF compilation and link lines.

5.13.4. Tesla Use Fast Math Library

When Target NVIDIA Tesla is set to Yes, use this property to use routines from the fast math library.

Setting this property to Yes adds the -⁠ta=tesla:fastmath switch to the PVF compilation and link lines.

5.13.5. Tesla LLVM

When Target NVIDIA Tesla is set to Yes, use this property to control using of LLVM back end.

Setting this property to Yes adds the -⁠ta=tesla:llvm switch to the PVF compilation and link lines.

5.13.6. Tesla Noattach

When Target NVIDIA Tesla is set to Yes, use this property to prevent attaching to existing CUDA context.

Setting this property to Yes adds the -⁠ta=tesla:noattach switch to the PVF compilation and link lines.

5.13.7. Tesla Pin Host Memory

When Target NVIDIA Tesla is set to Yes, use this property to set default to pin host memory.

Setting this property to Yes adds the -⁠ta=tesla:pin switch to the PVF compilation and link lines.

5.13.8. Tesla Autocollapse

When Target NVIDIA Tesla is set to Yes, use this property to automatically collapse tightly nested loops.

  • Default: Allows the compiler to control automatic collapse of tightly nested loops. This select adds no additional sub-options to -⁠ta=tesla.
  • Yes: Enables automatic collapse of tightly nested loops by adding the -⁠ta=tesla:autocollapse switch to the PVF compilation and link lines.
  • No: Disables automatic collapse of tightly nested loops by adding the -⁠ta=tesla:noautocollapse switch to the PVF compilation and link lines.

5.13.9. Tesla Debug

When Target NVIDIA Tesla is set to Yes, use this property to control generation of GPU debug information.

  • Default: Allows the compiler to control generation of GPU debug information. This select adds no additional sub-options to -⁠ta=tesla.
  • Yes: Enables generation of GPU debug information by adding the -⁠ta=tesla:debug switch to the PVF compilation and link lines.
  • No: Disables generation of GPU debug information by adding the -⁠ta=tesla:nodebug switch to the PVF compilation and link lines.

5.13.10. Tesla Lineinfo

When Target NVIDIA Tesla is set to Yes, use this property to control generation of GPU line information.

  • Default: Allows the compiler to control generation of GPU line information. This select adds no additional sub-options to -⁠ta=tesla.
  • Yes: Enables generation of GPU line information by adding the -⁠ta=tesla:lineinfo switch to the PVF compilation and link lines.
  • No: Disables generation of GPU line information by adding the -⁠ta=tesla:nolineinfo switch to the PVF compilation and link lines.

5.13.11. Tesla Unroll

When Target NVIDIA Tesla is set to Yes, use this property to control automatic inner loop unrolling.

  • Default: Allows the compiler to control automatic inner loop unrolling. This select adds no additional sub-options to -⁠ta=tesla.
  • Yes: Enables automatic inner loop unrolling by adding the -⁠ta=tesla:unroll switch to the PVF compilation and link lines.
  • No: Disables automatic inner loop unrolling by adding the -⁠ta=tesla:nounroll switch to the PVF compilation and link lines.

5.13.12. Tesla Required

When Target NVIDIA Tesla is set to Yes, use this property to direct the compiler to issue error if the compute regions fail to accelerate.

  • Default: Uses the compiler defaults for handling instances where compute regions cannot be accelerated. This select adds no additional sub-options to -⁠ta=tesla.
  • Yes: Directs the compiler to stop compilation with an error when it cannot accelerate a compute region by adding the -⁠ta=tesla:required switch to the PVF compilation and link lines.
  • No: Directs the compiler to issue warnings when it cannot accelerate a compute region by adding the -⁠ta=tesla:norequired switch to the PVF compilation and link lines. Compilation does not stop but accelerator kernels are not generated.

5.13.13. Tesla Flush to Zero

When Target NVIDIA Tesla is set to Yes, use this property to control flush-to-zero mode for floating point computations in the GPU code generated for PGI Accelerator model compute regions.

  • Default: Accepts the default handling of floating point computations in the GPU code generated for CUDA Fortran kernels.
  • Yes: Enables flush-to-zero mode by adding the -⁠ta=tesla:flushz switch to the PVF compilation and link lines.
  • No: Disables flush-to-zero mode by adding the -⁠ta=tesla:noflushz switch to the PVF compilation and link lines.

5.13.14. Tesla Generate RDC

When Target NVIDIA Tesla is set to Yes, use this property to control generation of relocatable device code.

  • Default: Accepts the compiler’s default generation of relocatable device code.
  • Yes: Directs the compiler to generate relocatable device code by adding -⁠ta=tesla:rdc switch to the PVF compilation and link lines.
  • No: Prevents the compiler from generating relocatable device code by adding -⁠ta=tesla:nordc switch to the PVF compilation and link lines.

5.13.15. Tesla CUDA Toolkit

When Target NVIDIA Tesla is set to Yes, use this property to specify the version of the NVIDIA CUDA toolkit that is targeted by the compilers:

  • Default: The compiler selects the default CUDA toolkit version.
  • 7.5: Use the default version 7.5 of the CUDA toolkit. This selection adds the -⁠ta=tesla:cuda7.5 switch to the PVF compilation and link lines.
  • 8.0: Use version 8.0 of the CUDA toolkit. This selection adds the -⁠ta=tesla:cuda8.0 switch to the PVF compilation and link lines.
Note:pgaccelinfo prints the driver version as the first line of output.
  • For a 7.5 driver: CUDA Driver Version 7050
  • For an 8.0 driver: CUDA Driver Version 8000

5.13.16. Tesla Compute Capability

When Target NVIDIA Tesla is set to Yes, use this property to either automatically generate code compatible with all applicable compute capabilities, or to direct the compiler to use a manually-selected set.

Select either Automatic or Manual.

  • Automatic: Let the compiler generate code for all applicable compute capabilities. This is the default.
  • Manual: Choose one or more compute capabilities to target. The compiler generates code for each capability specified.

    If you select Manual, then you can select any or all of the following compute capabilities that are described in the next sections.

Important: If you select Manual and the additional properties do not appear, click Apply in the Property page dialog.

5.13.17. Tesla CC Fermi

When Target NVIDIA Tesla is set to Yes and Tesla Compute Capability is set to Manual, use this property to generate code for the Fermi Architecture.

Setting this property to Yes adds the -⁠ta=tesla:fermi switch to the PVF compilation and link lines.

5.13.18. Tesla CC Fermi+

When Target NVIDIA Tesla is set to Yes and Tesla Compute Capability is set to Manual, use this property to generate code for Fermi Architecture and above.

Setting this property to Yes adds the -⁠ta=tesla:fermi+ switch to the PVF compilation and link lines.

5.13.19. Tesla CC Kepler

When Target NVIDIA Tesla is set to Yes and Tesla Compute Capability is set to Manual, use this property to generate code for Kepler Architecture.

Setting this property to Yes adds the -⁠ta=tesla:kepler switch to the PVF compilation and link lines.

5.13.20. Tesla CC Kepler+

When Target NVIDIA Tesla is set to Yes and Tesla Compute Capability is set to Manual, use this property to generate code for Kepler Architecture and above

Setting this property to Yes adds the -⁠ta=tesla:cc30 switch to the PVF compilation and link lines.

5.13.21. Tesla: Keep Kernel Files

When Target NVIDIA Tesla is set to Yes, use this property to keep kernel files.

Setting this property to Yes adds the -⁠ta=tesla:keep switch to the PVF compilation and link lines.

5.14. Fortran | Diagnostics

The following properties are available from the Fortran | Diagnostics property page. These properties allow you to add switches to the compilation line that control the amount and type of information that the compiler provides.

For more information on the options referenced in these pages, refer to Miscellaneous Controls and specifically to the -⁠Minfo option.

5.14.1. Warning Level

Use this property to select the level of diagnostic reporting you want the compiler to use.

There are several levels of the -⁠Minform option available through this property. For more information on this option, refer to Miscellaneous Controls.

5.14.2. Generate Assembly

Use this property to generate an assembly file for each compiled source file.

Setting this property to Yes adds the -⁠Mkeepasm switch to the compilation line.

For more information on -⁠Mkeepasm, refer to Miscellaneous Controls.

5.14.3. Annotate Assembly

Use this property to generate assembly files and to annotate the assembly with source code.

Setting this property to Yes adds the -⁠Manno switch to the compilation line.

For more information on -⁠Manno, refer to Miscellaneous Controls.

5.14.4. Accelerator Information

Use this property to generate information about accelerator regions.

Setting this property to Yes adds the -⁠Minfo=accel switch to the compilation line.

5.14.5. CCFF Information

Use this property to append common compiler feedback format (CCFF) information to object files.

Setting this property to Yes adds the -⁠Minfo=ccff switch to the compilation line.

5.14.6. Fortran Language Information

Use this property to generate information about Fortran language features.

Setting this property to Yes adds the -⁠Minfo=ftn switch to the compilation line.

5.14.7. Inlining Information

Use this property to generate information about inlining.

Setting this property to Yes adds the -⁠Minfo=inline switch to the compilation line.

5.14.8. IPA Information

Use this property to generate information about interprocedural analysis (IPA) optimizations.

Setting this property to Yes adds the -⁠Minfo=ipa switch to the compilation line.

5.14.9. Loop Intensity Information

Use this property to generate compute intensity information about loops.

Setting this property to Yes adds the -⁠Minfo=intensity switch to the compilation line.

5.14.10. Loop Optimization Information

Use this property to generate information about loop optimizations.

Setting this property to Yes adds the -⁠Minfo=loop switch to the compilation line.

5.14.11. LRE Information

Use this property to generate information about loop-carried redundancy (LRE) elimination.

Setting this property to Yes adds the -⁠Minfo=lre switch to the compilation line.

5.14.12. OpenMP Information

Use this property to generate information about OpenMP.

Setting this property to Yes adds the -⁠Minfo=mp switch to the compilation line.

5.14.13. Optimization Information

Use this property to generate information about general optimizations.

Setting this property to Yes adds the -⁠Minfo=opt switch to the compilation line.

5.14.14. Parallelization Information

Use this property to generate information about parallel optimizations.

Setting this property to Yes adds the -⁠Minfo=par switch to the compilation line.

5.14.15. Unified Binary Information

Use this property to generate information specific to the PGI Unified Binary.

Setting this property to Yes adds the -⁠Minfo=unified switch to the compilation line.

5.14.16. Vectorization Information

Use this property to generate vectorization information.

Setting this property to Yes adds the -⁠Minfo=vect switch to the compilation line.

5.15. Fortran | Profiling

The following properties are available from the Fortran | Profiling property page. These properties allow you to add switches to the compilation line that control the information that the compiler provides.

For more information on the options referenced in these pages, refer to Miscellaneous Controls and specifically to the -⁠Mprof option.

5.15.1. Suppress CCFF Information

Use this property to suppress profiling's default generation of CCFF information.

Setting this property to Yes adds the -⁠Mprof=noccff switch to the compiling and linking lines.

5.15.2. Enable Limited DWARF

Use this property to generate limited DWARF information which can be used with performance profilers.

Setting this property to Yes adds the -⁠Mprof=dwarf switch to the compiling and linking lines.

5.16. Fortran | Runtime

The following properties are available from the Fortran | Runtime property page to allow the application to make additional checks at runtime.

5.16.1. Check Array Bounds

Use this property to enable array bounds checking at runtime.

Setting this property to Yes adds the -⁠Mbounds switch to the compilation line.

Setting this property to No adds no option to the compilation line, and there is no array bounds checking at runtime. No is the default.

5.16.2. Check Pointers

Use this property to perform runtime checks for pointers that are dereferenced while initialized to null.

Setting this property to Yes adds the -⁠Mchkptr switch to the compilation line.

Setting this property to No adds no option to the compilation line, and there is no runtime check for pointers that are dereferenced while initialized to null. No is the default.

5.16.3. Check Stack

Use this property to perform runtime stack checks for available space in the prologue of a function and before the start of a parallel region.

Setting this property to Yes adds the -⁠Mchkstk switch to the compilation line.

Setting this property to No adds no option to the compilation line, and there are no runtime stack checks. No is the default.

5.16.4. Command Line

This property page contains two boxes.

  • The first box, titled All options, is a read-only description of what the compilation line will be. This description is based on the values of the properties set in the Fortran property pages.
  • The second box, titled Additional options, allows you to specify any other options that you want the compiler to use. Use this box when the option you need is not available through any of the Fortran property pages.

    For more information on all the compiler options that are available, refer to Command-Line Options Reference.

5.17. Fortran | Command Line

The following properties are available from the Fortran | Command Line property page.

5.17.1. Command Line

This property page contains two boxes.

  • The first box, titled All options, is a read-only description of what the compilation line will be. This description is based on the values of the properties set in the Fortran property pages.
  • The second box, titled Additional options, allows you to specify any other options that you want the compiler to use. Use this box when the option you need is not available through any of the Fortran property pages.

    For more information on all the compiler options that are available, refer to Command-Line Options Reference.

5.18. Linker Property Pages

This section contains the property pages that are included in the Linker category. The Linker property page category is available for projects that build an executable or a dynamically linked library (DLL).

5.19. Linker | General

The following properties are available from the Linker | General property page.

5.19.1. Output File

Use this property to override the default output file name.

Providing the file name and the file’s extension is equivalent to using the -⁠o switch.

Note: You must provide the file’s extension.

For more information on -⁠o, refer to -o.

5.19.2. Additional Library Directories

Use this property to add one or more directories to the library search path.

For every directory path that is added to this property, PVF adds /LIBPATH:[dir] to the link line.

There are two ways to add directories to this property:

  • Type the information directly into the property page box.

    Use a semi-colon (‘;’) to separate each directory.

  • Click the ellipsis (‘...’) button in the property page box to open the Additional Library Directories dialog box.

    Enter each directory on its own line in this box. Do not use semi-colons to separate directories; the semi-colons are added automatically when the box is closed.

Tip: To add directories, use this property. To add libraries, use the Additional Dependencies property on the Linker | Input Property page.

5.19.3. Stack Reserve Size

Use this property to specify the total number of bytes for stack allocation in virtual memory. Use decimal notation. This property is equivalent to the -stack=reserve option. Leave this property blank to direct the linker to choose a default size for the stack.

5.19.4. Stack Commit Size

Use this property to specify the total number of bytes for stack allocation in physical memory. Use decimal notation. This property is equivalent to the -⁠stack=reserve,commit option. Commit Size is used only if a size is also specified for Stack Reserve.

5.19.5. Export Symbols

Use this property to specify whether the DLL will export symbols. This property is only visible for DLL project types.

5.20. Linker | Input

The following properties are available from the Linker | Input property page.

5.20.1. Additional Dependencies

Use this property to specify additional dependencies, such as libraries, to the link line.

There are two ways to add libraries to this property:

  • Type the information directly into the property page box.
    Note: Use spaces, not semi-colons, to separate multiple libraries. If the name of a library contains a space, use double quotes around that library name.