PGI compilers & tools software User's Guide

USER'S GUIDE FOR X86-64 CPUS
Version 2017
TABLE OF CONTENTS
Preface............................................................................................................ xii
Audience Description......................................................................................... xii
Compatibility and Conformance to Standards............................................................xii
Organization................................................................................................... xiii
Hardware and Software Constraints.......................................................................xiv
Conventions.................................................................................................... xiv
Terms............................................................................................................ xv
Related Publications.........................................................................................xvii
Chapter 1. Getting Started..................................................................................... 1
1.1. Overview................................................................................................... 1
1.2. Creating an Example..................................................................................... 2
1.3. Invoking the Command-level PGI Compilers......................................................... 2
1.3.1. Command-line Syntax...............................................................................2
1.3.2. Command-line Options............................................................................. 3
1.3.3. Fortran Directives and C/C++ Pragmas.......................................................... 3
1.4. Filename Conventions....................................................................................4
1.4.1. Input Files............................................................................................ 4
1.4.2. Output Files.......................................................................................... 6
1.5. Fortran, C, and C++ Data Types........................................................................7
1.6. Parallel Programming Using the PGI Compilers...................................................... 7
1.6.1. Run SMP Parallel Programs.........................................................................8
1.7. Platform-specific considerations....................................................................... 8
1.7.1. Using the PGI Compilers on Linux................................................................ 8
1.7.2. Using the PGI Compilers on Windows............................................................ 9
1.7.3. PGI on the Windows Desktop.................................................................... 10
1.7.4. Using the PGI Compilers on macOS............................................................. 12
1.8. Site-Specific Customization of the Compilers...................................................... 13
1.8.1. Use siterc Files..................................................................................... 13
1.8.2. Using User rc Files.................................................................................13
1.9. Common Development Tasks.......................................................................... 14
Chapter 2. Use Command-line Options.................................................................... 16
2.1. Command-line Option Overview...................................................................... 16
2.1.1. Command-line Options Syntax................................................................... 16
2.1.2. Command-line Suboptions........................................................................ 17
2.1.3. Command-line Conflicting Options.............................................................. 17
2.2. Help with Command-line Options.................................................................... 17
2.3. Getting Started with Performance................................................................... 19
2.3.1. Using -fast...........................................................................................19
2.3.2. Other Performance-Related Options............................................................ 20
2.4. Targeting Multiple Systems — Using the -tp Option............................................... 20
User's Guide for x86-64 CPUs
Version 2017 | ii
2.5. Frequently-used Options............................................................................... 21
Chapter 3. Optimizing and Parallelizing................................................................... 23
3.1. Overview of Optimization..............................................................................24
3.1.1. Local Optimization.................................................................................24
3.1.2. Global Optimization............................................................................... 24
3.1.3. Loop Optimization: Unrolling, Vectorization and Parallelization........................... 24
3.1.4. Interprocedural Analysis (IPA) and Optimization..............................................25
3.1.5. Function Inlining................................................................................... 25
3.1.6. Profile-Feedback Optimization (PFO)........................................................... 25
3.2. Getting Started with Optimization................................................................... 25
3.2.1. -help..................................................................................................26
3.2.2. -Minfo................................................................................................ 27
3.2.3. -Mneginfo............................................................................................ 27
3.2.4. -dryrun............................................................................................... 27
3.2.5. -v......................................................................................................27
3.2.6. PGI Profiler..........................................................................................27
3.3. Common Compiler Feedback Format (CCFF)....................................................... 28
3.4. Local and Global Optimization........................................................................28
3.4.1. -Msafeptr............................................................................................ 28
3.4.2. -O..................................................................................................... 28
3.5. Loop Unrolling using -Munroll......................................................................... 30
3.6. Vectorization using -Mvect.............................................................................32
3.6.1. Vectorization Sub-options.........................................................................32
3.6.2. Vectorization Example Using SIMD Instructions............................................... 34
3.7. Auto-Parallelization using -Mconcur.................................................................. 36
3.7.1. Auto-Parallelization Sub-options.................................................................37
3.7.2. Loops That Fail to Parallelize................................................................... 38
3.8. Processor-Specific Optimization and the Unified Binary.......................................... 42
3.9. Interprocedural Analysis and Optimization using -Mipa........................................... 43
3.9.1. Building a Program Without IPA – Single Step................................................. 43
3.9.2. Building a Program Without IPA – Several Steps.............................................. 44
3.9.3. Building a Program Without IPA Using Make................................................... 44
3.9.4. Building a Program with IPA......................................................................44
3.9.5. Building a Program with IPA – Single Step..................................................... 45
3.9.6. Building a Program with IPA – Several Steps.................................................. 46
3.9.7. Building a Program with IPA Using Make....................................................... 46
3.9.8. Questions about IPA............................................................................... 46
3.10. Profile-Feedback Optimization using -Mpfi/-Mpfo................................................ 48
3.11. Default Optimization Levels......................................................................... 48
3.12. Local Optimization Using Directives and Pragmas................................................ 49
3.13. Execution Timing and Instruction Counting........................................................49
3.14. Portability of Multi-Threaded Programs on Linux.................................................50
3.14.1. libnuma.............................................................................................51
User's Guide for x86-64 CPUs
Version 2017 | iii
Chapter 4. Using Function Inlining..........................................................................52
4.1. Automatic function inlining in C/C++................................................................52
4.2. Invoking Function Inlining..............................................................................53
4.3. Using an Inline Library................................................................................. 54
4.4. Creating an Inline Library............................................................................. 54
4.4.1. Working with Inline Libraries.................................................................... 55
4.4.2. Dependencies....................................................................................... 56
4.4.3. Updating Inline Libraries – Makefiles........................................................... 56
4.5. Error Detection during Inlining........................................................................56
4.6. Examples.................................................................................................. 57
4.7. Restrictions on Inlining................................................................................. 57
Chapter 5. Using OpenMP.....................................................................................59
5.1. OpenMP Overview....................................................................................... 59
5.1.1. OpenMP Shared-Memory Parallel Programming Model....................................... 60
5.1.2. Terminology......................................................................................... 60
5.1.3. OpenMP Example...................................................................................61
5.2. Task Overview............................................................................................62
5.3. Fortran Parallelization Directives.....................................................................63
5.4. C/C++ Parallelization Pragmas........................................................................ 64
5.5. Directive and Pragma Recognition....................................................................64
5.6. Directive and Pragma Summary Table............................................................... 64
5.6.1. Directive and Pragma Summary Table.......................................................... 65
5.7. Directive and Pragma Clauses.........................................................................66
5.8. Runtime Library Routines.............................................................................. 69
5.9. Environment Variables.................................................................................. 74
Chapter 6. Using MPI...........................................................................................76
6.1. MPI Overview.............................................................................................76
6.2. Compiling and Linking MPI Applications............................................................. 76
6.3. Debugging MPI Applications............................................................................77
6.4. Using MPICH on Linux and Apple macOS............................................................ 78
6.5. Using MPICH1, MPICH2, and MVAPICH1 on Linux................................................... 78
6.6. Using MVAPICH2 on Linux.............................................................................. 78
6.7. Using MS-MPI on Windows............................................................................. 79
6.8. Using Open MPI on Linux.............................................................................. 79
6.9. Using SGI MPI on Linux................................................................................. 79
6.10. Using MPI Compiler Wrappers........................................................................79
6.11. Limitations.............................................................................................. 80
6.12. Testing and Benchmarking............................................................................80
Chapter 7. Using an Accelerator............................................................................ 81
7.1. Overview.................................................................................................. 81
7.1.1. Components......................................................................................... 81
7.1.2. Availability.......................................................................................... 82
7.1.3. User-directed Accelerator Programming....................................................... 82
User's Guide for x86-64 CPUs
Version 2017 | iv
7.1.4. Features Not Covered or Implemented........................................................ 82
7.2. Terminology...............................................................................................82
7.3. System Requirements................................................................................... 84
7.4. Supported Processors and GPUs...................................................................... 84
7.5. Installation and Licensing.............................................................................. 85
7.5.1. Required Files...................................................................................... 85
7.5.2. Command-line Flag................................................................................ 85
7.6. Execution Model......................................................................................... 85
7.6.1. Host Functions......................................................................................86
7.6.2. Levels of Parallelism.............................................................................. 86
7.7. Memory Model........................................................................................... 86
7.7.1. Separate Host and Accelerator Memory Considerations..................................... 87
7.7.2. Accelerator Memory............................................................................... 87
7.7.3. Cache Management................................................................................ 87
7.8. Running an Accelerator Program..................................................................... 87
7.9. Accelerator Directives.................................................................................. 88
7.9.1. Enable Accelerator Directives................................................................... 88
7.9.2. Format............................................................................................... 89
7.9.3. C Directives......................................................................................... 89
7.9.4. Free-Form Fortran Directives.................................................................... 90
7.9.5. Fixed-Form Fortran Directives................................................................... 90
7.9.6. OpenACC Directive Summary.....................................................................91
7.10. Accelerator Directive Clauses........................................................................96
7.11. OpenAcc Runtime Libraries.......................................................................... 99
7.11.1. Runtime Library Definitions.....................................................................99
7.11.2. Runtime Library Routines....................................................................... 99
7.12. Environment Variables............................................................................... 100
7.13. Applicable Command-line Options................................................................. 101
7.14. Profiling Accelerator Kernels....................................................................... 103
7.15. Related Accelerator Programming Tools.......................................................... 103
7.15.1. NVIDIA CUDA Profile............................................................................ 104
7.15.2. TAU – Tuning and Analysis Utility.............................................................104
7.16. Supported Intrinsics.................................................................................. 104
7.16.1. Supported Fortran Intrinsics Summary Table............................................... 104
7.16.2. Supported C Intrinsics Summary Table...................................................... 106
7.17. References related to Accelerators............................................................... 107
Chapter 8. Eclipse............................................................................................ 108
8.1. Install Eclipse CDT..................................................................................... 108
8.2. Use Eclipse CDT........................................................................................ 109
Chapter 9. Using Directives and Pragmas................................................................110
9.1. PGI Proprietary Fortran Directives.................................................................. 110
9.2. PGI Proprietary C and C++ Pragmas................................................................ 111
9.3. PGI Proprietary Optimization Directive and Pragma Summary................................. 111
User's Guide for x86-64 CPUs
Version 2017 | v
9.4. Scope of Fortran Directives and Command-Line Options........................................113
9.5. Scope of C/C++ Pragmas and Command-Line Options........................................... 114
9.6. Prefetch Directives and Pragmas.................................................................... 117
9.6.1. Prefetch Directive Syntax in Fortran.......................................................... 117
9.6.2. Prefetch Directive Format Requirements..................................................... 117
9.6.3. Sample Usage of Prefetch Directive...........................................................118
9.6.4. Prefetch Pragma Syntax in C/C++............................................................. 118
9.6.5. Sample Usage of Prefetch Pragma.............................................................118
9.7. !$PRAGMA C.............................................................................................118
9.8. IGNORE_TKR Directive................................................................................ 119
9.8.1. IGNORE_TKR Directive Syntax.................................................................. 119
9.8.2. IGNORE_TKR Directive Format Requirements................................................ 119
9.8.3. Sample Usage of IGNORE_TKR Directive......................................................120
9.9. !DEC$ Directives....................................................................................... 120
9.9.1. !DEC$ Directive Syntax.......................................................................... 120
9.9.2. Format Requirements............................................................................ 120
9.9.3. Summary Table....................................................................................121
Chapter 10. Creating and Using Libraries................................................................122
10.1. Using builtin Math Functions in C/C++............................................................122
10.2. Using System Library Routines..................................................................... 123
10.3. Creating and Using Shared Object Files on Linux............................................... 123
10.3.1. Procedure to create a use a shared object file............................................123
10.3.2. ldd Command.................................................................................... 124
10.4. Creating and Using Dynamic Libraries on Apple macOX........................................125
10.5. PGI Runtime Libraries on Windows................................................................ 125
10.6. Creating and Using Static Libraries on Windows................................................ 125
10.6.1. ar command......................................................................................125
Syntax.................................................................................................... 126
Options...................................................................................................126
10.6.2. ranlib command................................................................................. 126
Syntax.................................................................................................... 126
Options...................................................................................................127
10.7. Creating and Using Dynamic-Link Libraries on Windows....................................... 127
10.7.1. Build a DLL: Fortran............................................................................ 129
10.7.2. Build a DLL: C................................................................................... 130
10.7.3. Build DLLs Containing Circular Mutual Imports: C......................................... 131
10.7.4. Build DLLs Containing Mutual Imports: Fortran............................................ 132
10.7.5. Import a Fortran module from a DLL........................................................ 133
10.8. Using LIB3F............................................................................................ 134
10.9. LAPACK, BLAS and FFTs..............................................................................134
10.10. Linking with ScaLAPACK............................................................................ 135
10.11. The C++ Standard Template Library............................................................. 135
Chapter 11. Using Environment Variables............................................................... 136
User's Guide for x86-64 CPUs
Version 2017 | vi
11.1. Setting Environment Variables..................................................................... 136
11.1.1. Setting Environment Variables on Linux..................................................... 136
11.1.2. Setting Environment Variables on Windows.................................................137
11.1.3. Setting Environment Variables on Apple macOS........................................... 137
11.2. PGI-Related Environment Variables................................................................138
11.3. PGI Environment Variables.......................................................................... 139
11.3.1. FLEXLM_BATCH...................................................................................140
11.3.2. FORTRANOPT..................................................................................... 140
11.3.3. GMON_OUT_PREFIX............................................................................. 140
11.3.4. LD_LIBRARY_PATH............................................................................... 140
11.3.5. LM_LICENSE_FILE................................................................................ 141
11.3.6. MANPATH.......................................................................................... 141
11.3.7. MPSTKZ............................................................................................141
11.3.8. MP_BIND...........................................................................................142
11.3.9. MP_BLIST..........................................................................................142
11.3.10. MP_SPIN......................................................................................... 142
11.3.11. MP_WARN........................................................................................ 142
11.3.12. NCPUS............................................................................................143
11.3.13. NCPUS_MAX..................................................................................... 143
11.3.14. NO_STOP_MESSAGE............................................................................ 143
11.3.15. PATH..............................................................................................143
11.3.16. PGI................................................................................................144
11.3.17. PGI_CONTINUE..................................................................................144
11.3.18. PGI_OBJSUFFIX................................................................................. 144
11.3.19. PGI_STACK_USAGE............................................................................. 144
11.3.20. PGI_TERM........................................................................................145
11.3.21. PGI_TERM_DEBUG.............................................................................. 146
11.3.22. PGROUPD_LICENSE_FILE...................................................................... 147
11.3.23. PWD.............................................................................................. 147
11.3.24. STATIC_RANDOM_SEED.........................................................................147
11.3.25. TMP...............................................................................................147
11.3.26. TMPDIR...........................................................................................148
11.4. Using Environment Modules on Linux............................................................. 148
11.5. Stack Traceback and JIT Debugging............................................................... 149
Chapter 12. Distributing Files – Deployment............................................................ 150
12.1. Deploying Applications on Linux................................................................... 150
12.1.1. Runtime Library Considerations...............................................................150
12.1.2. 64-bit Linux Considerations................................................................... 151
12.1.3. Linux Redistributable Files.................................................................... 151
12.1.4. Restrictions on Linux Portability..............................................................151
12.1.5. Licensing for Redistributable Files........................................................... 152
12.2. Deploying Applications on Windows............................................................... 152
12.2.1. PGI Redistributables............................................................................ 152
User's Guide for x86-64 CPUs
Version 2017 | vii
12.2.2. Microsoft Redistributables..................................................................... 152
12.3. Code Generation and Processor Architecture....................................................153
12.3.1. Generating Generic x86-64 Code............................................................. 153
12.3.2. Generating Code for a Specific Processor...................................................153
12.4. Generating One Executable for Multiple Types of Processors................................. 153
12.4.1. PGI Unified Binary Command-line Switches................................................ 154
12.4.2. PGI Unified Binary Directives and Pragmas................................................. 154
Chapter 13. Inter-language Calling........................................................................ 155
13.1. Overview of Calling Conventions.................................................................. 155
13.2. Inter-language Calling Considerations.............................................................155
13.3. Functions and Subroutines.......................................................................... 156
13.4. Upper and Lower Case Conventions, Underscores.............................................. 157
13.5. Compatible Data Types.............................................................................. 157
13.5.1. Fortran Named Common Blocks.............................................................. 158
13.6. Argument Passing and Return Values..............................................................159
13.6.1. Passing by Value (%VAL)........................................................................ 159
13.6.2. Character Return Values....................................................................... 159
13.6.3. Complex Return Values.........................................................................160
13.7. Array Indices.......................................................................................... 160
13.8. Examples............................................................................................... 161
13.8.1. Example – Fortran Calling C...................................................................161
13.8.2. Example – C Calling Fortran...................................................................162
13.8.3. Example – C++ Calling C....................................................................... 163
13.8.4. Example – C Calling C ++...................................................................... 163
13.8.5. Example – Fortran Calling C++................................................................ 164
13.8.6. Example – C++ Calling Fortran................................................................ 165
Chapter 14. Programming Considerations for 64-Bit Environments................................ 167
14.1. Data Types in the 64-Bit Environment............................................................ 167
14.1.1. C/C++ Data Types............................................................................... 167
14.1.2. Fortran Data Types..............................................................................168
14.2. Large Static Data in Linux.......................................................................... 168
14.3. Large Dynamically Allocated Data................................................................. 168
14.4. 64-Bit Array Indexing................................................................................ 169
14.5. Compiler Options for 64-bit Programming........................................................169
14.6. Practical Limitations of Large Array Programming.............................................. 170
14.7. Medium Memory Model and Large Array in C.................................................... 171
14.8. Medium Memory Model and Large Array in Fortran.............................................172
14.9. Large Array and Small Memory Model in Fortran............................................... 173
Chapter 15. C/C++ Inline Assembly and Intrinsics..................................................... 175
15.1. Inline Assembly....................................................................................... 175
15.2. Extended Inline Assembly........................................................................... 176
15.2.1. Output Operands................................................................................ 177
15.2.2. Input Operands.................................................................................. 179
User's Guide for x86-64 CPUs
Version 2017 | viii
15.2.3. Clobber List...................................................................................... 180
15.2.4. Additional Constraints.......................................................................... 181
15.2.5. Simple Constraints.............................................................................. 181
15.2.6. Machine Constraints............................................................................ 183
15.2.7. Multiple Alternative Constraints.............................................................. 184
15.2.8. Constraint Modifiers............................................................................ 185
15.3. Operand Aliases.......................................................................................187
15.4. Assembly String Modifiers........................................................................... 187
15.5. Extended Asm Macros................................................................................188
15.6. Intrinsics............................................................................................... 189
Chapter 16. Contact Information.......................................................................... 191
User's Guide for x86-64 CPUs
Version 2017 | ix
LIST OF TABLES
Table 1 PGI Compilers and Commands ..................................................................... xvi
Table 2 Option Descriptions .................................................................................... 6
Table 3 Examples of Usine siterc and User rc Files .......................................................13
Table 4 Typical -fast Options ................................................................................. 19
Table 5 Additional -fast Options ............................................................................. 19
Table 6 Commonly Used Command-Line Options ..........................................................21
Table 7 Example of Effect of Code Unrolling .............................................................. 31
Table 8 -Mvect Suboptions .................................................................................... 33
Table 9 -Mconcur Suboptions ................................................................................. 37
Table 10 Optimization and -O, -g and -M<opt> Options ................................................. 48
Table 11 Directive and Pragma Summary Table ........................................................... 65
Table 12 Directive and Pragma Summary Table ........................................................... 66
Table 13 Runtime Library Routines Summary .............................................................. 70
Table 14 OpenMP-related Environment Variable Summary Table ....................................... 75
Table 15 MPI Distribution Options ........................................................................... 77
Table 16 PGI Accelerator Directive Summary Table ...................................................... 91
Table 17 Directive Clauses Summary ........................................................................ 96
Table 18 Accelerator Runtime Library Routines ..........................................................100
Table 19 Accelerator Environment Variables ............................................................. 101
Table 20 Supported Fortran Intrinsics ..................................................................... 105
Table 21 Supported C Intrinsic Double Functions ........................................................ 106
Table 22 Supported C Intrinsic Float Functions .......................................................... 106
Table 23 Proprietary Optimization-Related Fortran Directive and C/C++ Pragma Summary ...... 112
Table 24 IGNORE_TKR Example ............................................................................. 120
User's Guide for x86-64 CPUs
Version 2017 | x
Table 25 !DEC$ Directives Summary Table ................................................................ 121
Table 26 PGI-Related Environment Variable Summary .................................................. 138
Table 27 Supported PGI_TERM Values ......................................................................145
Table 28 Fortran and C/C++ Data Type Compatibility .................................................. 157
Table 29 Fortran and C/C++ Representation of the COMPLEX Type ...................................158
Table 30 64-bit Compiler Options .......................................................................... 169
Table 31 Effects of Options on Memory and Array Sizes ............................................... 170
Table 32 64-Bit Limitations .................................................................................. 170
Table 33 Simple Constraints ................................................................................. 182
Table 34 x86_64 Machine Constraints ......................................................................183
Table 35 Multiple Alternative Constraints .................................................................185
Table 36 Constraint Modifier Characters .................................................................. 185
Table 37 Assembly String Modifier Characters ........................................................... 187
Table 38 Intrinsic Header File Organization .............................................................. 189
User's Guide for x86-64 CPUs
Version 2017 | xi
PREFACE
This guide is part of a set of manuals that describe how to use The Portland Group (PGI)
Fortran, C, and C++ compilers and program development tools. These compilers and
tools include the PGF77, PGFORTRAN, PGC++, and PGCC compilers, the PGI profiler,
and the PGI debugger. They work in conjunction with an x86-64 assembler and linker.
You can use the PGI compilers and tools to compile, debug, optimize, and profile serial
and parallel applications for x86 processor-based systems.
The PGI Compiler User’s Guide provides operating instructions for the PGI commandlevel development environment. The PGI Compiler Reference Manual contains details
concerning the PGI compilers' interpretation of the Fortran language, implementation
of Fortran language extensions, and command-level compilation. Users are expected
to have previous experience with or knowledge of the Fortran programming language.
Neither guide teaches the Fortran programming language.
Audience Description
This manual is intended for scientists and engineers using the PGI compilers. To use
these compilers, you should be aware of the role of high-level languages, such as
Fortran, C, and C++, as well as assembly-language in the software development process;
and you should have some level of understanding of programming. The PGI compilers
are available on a variety of x86-64/x64 hardware platforms and operating systems. 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 PGI compilers and tools, 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).
User's Guide for x86-64 CPUs
Version 2017 | xii
Preface
‣
‣
‣
‣
‣
‣
‣
‣
‣
‣
‣
‣
‣
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).
American National Standard Programming Language C, ANSI X3.159-1989.
ISO/IEC 9899:1999, Information technology – Programming Languages – C, Geneva,
1999 (C99).
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.
This book contains the essential information on how to use the compiler and is divided
into these sections:
Getting Started provides an introduction to the PGI compilers and describes their use
and overall features.
Use Command-line Options provides an overview of the command-line options as well
as task-related lists of options.
Optimizing and Parallelizing describes standard optimization techniques that, with little
effort, allow users to significantly improve the performance of programs.
Using Function Inlining describes how to use function inlining and shows how to create
an inline library.
User's Guide for x86-64 CPUs
Version 2017 | xiii
Preface
Using OpenMP provides a description of the OpenMP Fortran parallelization directives
and of the OpenMP C and C++ parallelization pragmas, and shows examples of their
use.
Using MPI describes how to use MPI with PGI products.
Using an Accelerator describes how to use the PGI Accelerator compilers.
Eclipse describes how to use the PGI C/C++ compilers from within Eclipse, a free, opensource, integrated software development environment.
Using Directives and Pragmas provides a description of each Fortran optimization
directive and C/C++ optimization pragma, and shows examples of their use.
Creating and Using Libraries discusses PGI support libraries, shared object files, and
environment variables that affect the behavior of the PGI compilers.
Using Environment Variables describes the environment variables that affect the
behavior of the PGI compilers.
Distributing Files – Deployment describes the deployment of your files once you have
built, debugged and compiled them successfully.
Inter-language Calling provides examples showing how to place C language calls in a
Fortran program and Fortran language calls in a C program.
Programming Considerations for 64-Bit Environments discusses issues of which
programmers should be aware when targeting 64-bit processors.
C/C++ Inline Assembly and Intrinsics describes how to use inline assembly code in C
and C++ programs, as well as how to use intrinsic functions that map directly to x86-64
machine instructions.
Hardware and Software Constraints
This guide describes versions of the PGI compilers that produce assembly code for x64
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 compilers.
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.
User's Guide for x86-64 CPUs
Version 2017 | xiv
Preface
[ 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.
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 32or 64-bit operating systems.
Terms
A number of terms related to systems, processors, compilers and tools are used
throughout this guide. For example:
AMD64
static linking
AVX
linux86-64
osx86-64
Win32
DLL
macOS
shared library
Win64
driver
-mcmodel=small
SSE
Windows
dynamic library
-mcmodel=medium
SSE1
x64
hyperthreading (HT)
MPI
SSE2
x86
Intel 64
MPICH
SSE3
x87
large arrays
multicore
SSE4A and ABM
license keys
NUMA
SSSE3
For a complete definition of these terms and other terms in this guide with which you
may be unfamiliar, PGI provides a glossary of terms which you can access at http://
www.pgroup.com/support/definitions.htm.
User's Guide for x86-64 CPUs
Version 2017 | xv
Preface
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
PGCC
ISO/ANSI C11 and K&R C
pgcc
PGC++
ISO/ANSI C++14 with GNU
compatibility
pgc++ on Linux and macOS
PGI Debugger
Source code debugger
pgdbg
PGI Profiler
Performance profiler
pgprof
In general, the designation PGI Fortran is used to refer to The Portland Group's 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. Usage of PGC++ and PGCC is consistent with PGFORTRAN and PGF77, though
there are command-line options and features of these compilers that do not apply to
PGFORTRAN and PGF77, and vice versa.
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.
User's Guide for x86-64 CPUs
Version 2017 | xvi
Preface
Where such distinctions are important with respect to a given compiler option or
feature, it is explicitly noted in this manual.
The default for performing scalar floating-point arithmetic is to use SSE instructions
on targets that support SSE1 and SSE2.
Support for 32-bit development was deprecated in PGI 2016 and is no longer be
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.
Related Publications
The following documents contain additional information related to the x86-64 and x64
architectures, and the compilers and tools available from The Portland Group.
‣
‣
‣
‣
‣
‣
‣
‣
‣
PGI Fortran Reference manual describes the FORTRAN 77, Fortran 90/95, Fortran
2003 statements, data types, input/output format specifiers, and additional reference
material related to use of the PGI Fortran compilers.
System V Application Binary Interface Processor Supplement by AT&T UNIX System
Laboratories, Inc. (Prentice Hall, Inc.).
System V Application Binary Interface X86-64 Architecture Processor Supplement, http://
www.x86-64.org/documentation_folder/abi.pdf.
Fortran 95 Handbook Complete ISO/ANSI Reference, Adams et al, The MIT Press,
Cambridge, Mass, 1997.
Programming in VAX Fortran, Version 4.0, Digital Equipment Corporation (September,
1984).
IBM VS Fortran, IBM Corporation, Rev. GC26-4119.
The C Programming Language by Kernighan and Ritchie (Prentice Hall).
C: A Reference Manual by Samuel P. Harbison and Guy L. Steele Jr. (Prentice Hall,
1987).
The Annotated C++ Reference Manual by Margaret Ellis and Bjarne Stroustrup, AT&T
Bell Laboratories, Inc. (Addison-Wesley Publishing Co., 1990).
User's Guide for x86-64 CPUs
Version 2017 | xvii
Preface
User's Guide for x86-64 CPUs
Version 2017 | xviii
Chapter 1.
GETTING STARTED
This section describes how to use the PGI compilers.
1.1. Overview
The command used to invoke a compiler, such as the pgfortran command, is called
a compiler driver. The compiler driver controls the following phases of compilation:
preprocessing, compiling, assembling, and linking. Once a file is compiled and an
executable file is produced, you can execute, debug, or profile the program on your
system. Executables produced by the PGI compilers are unconstrained, meaning
they can be executed on any compatible x86-64 processor-based system, regardless of
whether the PGI compilers are installed on that system.
In general, using a PGI compiler involves three steps:
1. Produce program source code in a file containing a .f extension or another
appropriate extension, as described in Input Files. This program may be one that
you have written or one that you are modifying.
2. Compile the program using the appropriate compiler command.
3. Execute, debug, or profile the executable file on your system.
You might also want to deploy your application, though this is not a required step.
The PGI compilers allow many variations on these general program development steps.
These variations include the following:
‣
‣
‣
Stop the compilation after preprocessing, compiling or assembling to save and
examine intermediate results.
Provide options to the driver that control compiler optimization or that specify
various features or limitations.
Include as input intermediate files such as preprocessor output, compiler output, or
assembler output.
User's Guide for x86-64 CPUs
Version 2017 | 1
Getting Started
1.2. Creating an Example
Let's look at a simple example of using the PGI compiler to create, compile, and execute
a program that prints:
hello
1. Create your program.
For this example, suppose you enter the following simple Fortran program in the file
hello.f:
print *, "hello"
end
2. Compile the program.
When you created your program, you called it hello.f. In this example, we compile
it from a shell command prompt using the default pgfortran driver option. Use the
following syntax:
$ pgfortran hello.f
By default, the executable output is placed in the file a.out, or, on Windows
platforms, in a filename based on the name of the first source or object file on the
command line. However, you can specify an output file name by using the -o option.
To place the executable output in the file hello, use this command:
$ pgfortran -o hello hello.f
3. Execute the program.
To execute the resulting hello program, simply type the filename at the command
prompt and press the Return or Enter key on your keyboard:
$ hello
Below is the expected output:
hello
1.3. Invoking the Command-level PGI Compilers
To translate and link a Fortran, C, or C++ program, the pgf77, pgf95, pgfortran,
pgcc and pgc++ commands do the following:
1.
2.
3.
4.
Preprocess the source text file.
Check the syntax of the source text.
Generate an assembly language file.
Pass control to the subsequent assembly and linking steps.
1.3.1. Command-line Syntax
The compiler command-line syntax, using pgfortran as an example, is:
pgfortran [options] [path]filename [...]
User's Guide for x86-64 CPUs
Version 2017 | 2
Getting Started
Where:
options
is one or more command-line options, all of which are described in detail in Use
Command-line Options.
path
is the pathname to the directory containing the file named by filename. If you do not
specify the path for a filename, the compiler uses the current directory. You must
specify the path separately for each filename not in the current directory.
filename
is the name of a source file, preprocessed source file, assembly-language file, object
file, or library to be processed by the compilation system. You can specify more than
one [path]filename.
1.3.2. Command-line Options
The command-line options control various aspects of the compilation process. For a
complete alphabetical listing and a description of all the command-line options, refer to
Use Command-Line Options.
The following list provides important information about proper use of command-line
options.
‣
‣
‣
Command-line options and their arguments are case sensitive.
The compiler drivers recognize characters preceded by a hyphen (-) as commandline options. For example, the -Mlist option specifies that the compiler creates a
listing file.
The convention for the text of this manual is to show command-line options using
a dash instead of a hyphen; for example, you see -Mlist.
The order of options and the filename is flexible. That is, you can place options
before and after the filename argument on the command line. However, the
placement of some options is significant, such as the -l option, in which the order of
the filenames determines the search order.
If two or more options contradict each other, the last one in the command line
takes precedence.
1.3.3. Fortran Directives and C/C++ Pragmas
You can insert Fortran directives and C/C++ pragmas in program source code to alter the
effects of certain command-line options and to control various aspects of the compilation
process for a specific routine or a specific program loop. For more information on
Fortran directives and C/C++ pragmas, refer to Using OpenMP and Using Directives and
Pragmas.
User's Guide for x86-64 CPUs
Version 2017 | 3
Getting Started
1.4. Filename Conventions
The PGI compilers use the filenames that you specify on the command line to find and
to create input and output files. This section describes the input and output filename
conventions for the phases of the compilation process.
1.4.1. Input Files
You can specify assembly-language files, preprocessed source files, Fortran/C/C++
source files, object files, and libraries as inputs on the command line. The compiler
driver determines the type of each input file by examining the filename extensions.
For systems with a case-insensitive file system, use the -Mpreprocess option,
described in ‘Command-Line Options Reference’ section of the PGI Compiler’s
Reference Manual, under the commands for Fortran preprocessing.
The drivers use the following conventions:
filename.f
indicates a Fortran source file.
filename.F
indicates a Fortran source file that can contain macros and preprocessor directives (to
be preprocessed).
filename.FOR
indicates a Fortran source file that can contain macros and preprocessor directives (to
be preprocessed).
filename.F90
indicates a Fortran 90/95 source file that can contain macros and preprocessor
directives (to be preprocessed).
filename.F95
indicates a Fortran 90/95 source file that can contain macros and preprocessor
directives (to be preprocessed).
filename.f90
indicates a Fortran 90/95 source file that is in freeform format.
filename.f95
indicates a Fortran 90/95 source file that is in freeform format.
filename.cuf
indicates a Fortran 90/95 source file in free format with CUDA Fortran extensions.
filename.CUF
indicates a Fortran 90/95 source file in free format with CUDA Fortran extensions and
that can contain macros and preprocessor directives (to be preprocessed).
filename.c
indicates a C source file that can contain macros and preprocessor directives (to be
preprocessed).
filename.C
indicates a C++ source file that can contain macros and preprocessor directives (to be
preprocessed).
User's Guide for x86-64 CPUs
Version 2017 | 4
Getting Started
filename.i
indicates a preprocessed C or C++ source file.
filename.cc
indicates a C++ source file that can contain macros and preprocessor directives (to be
preprocessed).
filename.cpp
indicates a C++ source file that can contain macros and preprocessor directives (to be
preprocessed).
filename.s
indicates an assembly-language file.
filename.o
(Linux and macOS) indicates an object file.
filename.obj
(Windows systems only) indicates an object file.
filename.a
(Linux and macOS) indicates a library of object files.
filename.lib
(Windows systems only) indicates a statically-linked library of object files or an
import library.
filename.so
(Linux only) indicates a library of shared object files.
filename.dll
(Windows systems only) indicates a dynamically-linked library.
filename.dylib
(macOS systems only) indicates a dynamically-linked library.
The driver passes files with .s extensions to the assembler and files with .o, .obj, .so,
.dll, .a, and .lib extensions to the linker. Input files with unrecognized extensions,
or no extension, are also passed to the linker.
Files with a .F (Capital F) or .FOR suffix are first preprocessed by the Fortran
compilers and the output is passed to the compilation phase. The Fortran preprocessor
functions like cpp for C programs, but is built in to the Fortran compilers rather than
implemented through an invocation of cpp. This design ensures consistency in the
preprocessing step regardless of the type or revision of operating system under which
you are compiling.
Any input files not needed for a particular phase of processing are not processed. For
example, if on the command line you specify an assembly-language file (filename.s)
and the -S option to stop before the assembly phase, the compiler takes no action on
the assembly language file. Processing stops after compilation and the assembler does
not run. In this scenario, the compilation must have been completed in a previous pass
which created the .s file. For a complete description of the -S option, refer to Output
Files.
In addition to specifying primary input files on the command line, code within other
files can be compiled as part of include files using the INCLUDE statement in a Fortran
source file or the preprocessor #include directive in Fortran source files that use a .F
extension or C and C++ source files.
User's Guide for x86-64 CPUs
Version 2017 | 5
Getting Started
When linking a program with a library, the linker extracts only those library components
that the program needs. The compiler drivers link in several libraries by default. For
more information about libraries, refer to Create and Use Libraries.
1.4.2. Output Files
By default, an executable output file produced by one of the PGI compilers is placed in
the file a.out, or, on Windows, in a filename based on the name of the first source or
object file on the command line. As the Hello example shows, you can use the -o option
to specify the output file name.
If you use option -F (Fortran only), -P (C/C++ only), -S or -c, the compiler produces a file
containing the output of the last completed phase for each input file, as specified by the
option supplied.
The output file is a preprocessed source file, an assembly-language file, or an unlinked
object file respectively. Similarly, the -E option does not produce a file, but displays
the preprocessed source file on the standard output. Using any of these options,
the -o option is valid only if you specify a single input file. If no errors occur during
processing, you can use the files created by these options as input to a future invocation
of any of the PGI compiler drivers.
The following table lists the stop-after options and the output files that the compilers
create when you use these options. It also indicates the accepted input files.
Table 2 Option Descriptions
Option
Stop After
Input
Output
-E
preprocessing
Source files
preprocessed file to standard out
-F
preprocessing
Source files. This option is not
valid for pgcc or pgc++.
preprocessed file (.f)
-P
preprocessing
Source files. This option is not
valid for pgf77, pgf95, or
pgfortran.
-S
compilation
Source files or preprocessed files
-c
assembly
Source files, or preprocessed
files, or assembly-language files
none
linking
Source files, or preprocessed
files, assembly-language files,
object files, or libraries
preprocessed file (.i)
assembly-language file (.s)
unlinked object file (.o or
.obj)
executable file (a.out or
.exe)
If you specify multiple input files or do not specify an object filename, the compiler uses
the input filenames to derive corresponding default output filenames of the following
form, where filename is the input filename without its extension:
filename.f
indicates a preprocessed file, if you compiled a Fortran file using the -F option.
filename.i
indicates a preprocessed file, if you compiled using the -P option.
User's Guide for x86-64 CPUs
Version 2017 | 6
Getting Started
filename.lst
indicates a listing file from the -Mlist option.
filename.o or filename.obj
indicates a object file from the -c option.
filename.s
indicates an assembly-language file from the -S option.
Unless you specify otherwise, the destination directory for any output file is the
current working directory. If the file exists in the destination directory, the compiler
overwrites it.
The following example demonstrates the use of output filename extensions.
$ pgfortran -c proto.f proto1.F
This produces the output files proto.o and proto1.o, or, on Windows, proto.obj
and proto1.obj, all of which are binary object files. Prior to compilation, the file
proto1.F is preprocessed because it has a .F filename extension.
1.5. Fortran, C, and C++ Data Types
The PGI Fortran, C, and C++ compilers recognize scalar and aggregate data types. A
scalar data type holds a single value, such as the integer value 42 or the real value 112.6.
An aggregate data type consists of one or more scalar data type objects, such as an array
of integer values.
For information about the format and alignment of each data type in memory, and the
range of values each type can have on x64 processor-based systems, refer to ‘Fortran, C,
and C++ Data Types’ section of the PGI Compiler’s Reference Guide.
For more information on x86-64-specific data representation, refer to the System V
Application Binary Interface Processor Supplement by AT&T UNIX System Laboratories, Inc.
(Prentice Hall, Inc.) listed in the Related Publications section in the Preface.
For more information on x64 processor-based systems and the application binary
interface (ABI) for those systems, see http://www.x86-64.org/documentation/abi.pdf.
1.6. Parallel Programming Using the PGI Compilers
The PGI compilers support many styles of parallel programming:
‣
‣
Automatic shared-memory parallel programs compiled using the -Mconcur option
to pgf77, pgf95, pgfortran, pgcc or pgc++. Parallel programs of this variety
can be run on shared-memory parallel (SMP) systems such as dual-core or multiprocessor workstations.
OpenMP shared-memory parallel programs compiled using the -mp option to
pgf77, pgf95, pgfortran, pgcc or pgc++. Parallel programs of this variety
can be run on SMP systems. Carefully coded user-directed parallel programs
using OpenMP directives can often achieve significant speed-ups on dual-core
User's Guide for x86-64 CPUs
Version 2017 | 7
Getting Started
‣
‣
workstations or large numbers of processors on SMP server systems. Using OpenMP
contains complete descriptions of user-directed parallel programming.
Distributed computing using an MPI message-passing library for communication
between distributed processes.
Accelerated computing using either a low-level model such as CUDA Fortran or a
high-level model such as the PGI Accelerator model or OpenACC to target a manycore GPU or other attached accelerator.
The first two types of parallel programs are collectively referred to as SMP parallel
programs.
On a single silicon die, today's CPUs incorporate two or more complete processor cores
– functional units, registers, level 1 cache, level 2 cache, and so on. These CPUs are
known as multicore processors. For purposes of threads or OpenMP parallelism, these
cores function as two or more distinct processors. However, the processing cores are on
a single chip occupying a single socket on a system motherboard. For purposes of PGI
software licensing, a multicore processor is treated as a single CPU.
1.6.1. Run SMP Parallel Programs
When you execute an SMP parallel program, by default it uses only one processor. To
run on more than one processor, set the NCPUS environment variable to the desired
number of processors. For information on how to set environment variables, refer to
Setting Environment Variables.
If you set NCPUS to a number larger than the number of physical processors, your
program may execute very slowly.
1.7. Platform-specific considerations
The following list are the platforms supported by the PGI compilers and tools:
‣
‣
‣
64-bit Linux – supported on 64-bit Linux operating systems running on an x86-64
compatible processor.
64-bit Windows – supported on 64-bit Windows operating systems running an x64compatible processor.
64-bit macOS – supported on 64-bit macOS operating systems running on a 64-bit
Intel-based Mac system.
The following sections describe the specific considerations required to use the PGI
compilers on the various platforms: Linux, Windows, and macOS.
1.7.1. Using the PGI Compilers on Linux
Linux Header Files
The Linux system header files contain many GNU gcc extensions. PGI supports many of
these extensions, thus allowing the PGI C and C++ compilers to compile most programs
User's Guide for x86-64 CPUs
Version 2017 | 8
Getting Started
that the GNU compilers can compile. A few header files not interoperable with the
PGI compilers have been rewritten. These files are included in $PGI/linux86-64/
include and in $PGI/linux86-64/include --gcc*, such as sigset.h, asm/
byteorder.h, stddef.h, asm/posix_types.h and others. Also, PGI’s version of
stdarg.h supports changes in newer versions of Linux.
If you are using the PGI C or C++ compilers, please make sure that the supplied versions
of these include files are found before the system versions. This hierarchy happens by
default unless you explicitly add a -I option that references one of the system include
directories.
Running Parallel Programs on Linux
You may encounter difficulties running auto-parallel or OpenMP programs on
Linux systems when the per-thread stack size is set to the default (2MB). If you have
unexplained failures, please try setting the environment variable OMP_STACKSIZE to
a larger value, such as 8MB. For information on setting environment variables, refer to
Setting Environment Variables.
If your program is still failing, you may be encountering the hard 8 MB limit on main
process stack sizes in Linux. You can work around the problem by issuing the following
command:
In csh:
% limit stacksize unlimited
In bash, sh, zsh, or ksh, use:
$ ulimit -s unlimited
1.7.2. Using the PGI Compilers on Windows
PGI on the Windows Start Menu
PGI provides a Start menu entry that provides access to different versions of PGI
command shells as well as easy access to the PGI Debugger, the PGI Profiler,
documentation, and licensing. The following sections provide a quick overview of the
menu selections.
To access the main PGI menu, from the Start menu, select Start | All Programs | PGI.
Command Shell Submenus
From the PGI menu, you have access to PGI command shells for each version of PGI
installed on your system. For example, if you have both PGI 14.1 and PGI 13.9 installed,
then you have a submenu for each of these versions.
The PGI submenus for each version include the following:
User's Guide for x86-64 CPUs
Version 2017 | 9
Getting Started
‣
‣
PGI Bash (64) – Select this option to launch a Cygwin bash shell in which the
environment is pre-initialized to use the 64-bit PGI compilers and tools. The default
environment variables are already set and available. (Available only on x64 systems
with Cygwin installed.)
PGI Cmd (64) – Select this option to launch a Microsoft command shell in which the
environment is pre-initialized to use the 64-bit PGI compilers and tools. The default
environment variables are already set and available. (Available only on x64 systems.)
The command window launched by PGI menu command or desktop icon can be
customized using the "Properties" selection on the menu accessible by right-clicking the
window's title bar.
Debugger & Profiler Submenu
From the Debugger & Profiler menu, you have access to the PGI debugging and
profiling tools. The PGI debugger is a symbolic debugger for Fortran, C and assembly
language programs. It provides debugger features, such as execution control using
breakpoints, single-stepping, and examination and modification of application variables,
memory locations, and registers.
‣
‣
PGI Debugger – Select this option to launch the PGI debugger.
PGI Profiler – Select this option to launch the PGI performance profiler. The profiler
provides a way to visualize and diagnose the performance of the components of
your program, and provides features for helping you to understand why certain
parts of your program have high execution times.
Documentation Submenu
All PGI documentation is available online. The Documentation submenu contains a link
to the online location.
Licensing Submenu
From the Licensing menu, you have access to the PGI License Agreement and an
automated license generating tool:
‣
‣
Generate License – Select this option to display the PGI License Setup dialog that
walks you through the steps required to download and install a PGI license. To
complete this process you need an Internet connection.
License Agreement – Select this option to display the license agreement that is
associated with use of PGI software.
1.7.3. PGI on the Windows Desktop
By default, a PGI installation creates a shortcut on the Windows desktop. This shortcut
launches a Cygwin bash shell if Cygwin is installed; otherwise it launches a Microsoft
User's Guide for x86-64 CPUs
Version 2017 | 10
Getting Started
command shell. The environment for this shell is pre-configured to use PGI compilers
and tools.
BASH Shell Environment (Cygwin)
A UNIX-like shell environment, Cygwin, is bundled with PGI compilers and tools for
Windows to provide a familiar development environment for Linux or UNIX users.
After installation you'll have a PGI icon on your Windows desktop. Double-left-click on
this icon to launch an instance of the Cygwin bash command shell window. Working
within BASH is very much like working within the sh or ksh shells on a Linux system;
yet BASH has a command history feature similar to csh and several other unique
features. Shell programming is fully supported.
The BASH shell window is pre-initialized for usage of the PGI compilers and tools, so
there is no need to set environment variables or modify your command path when the
command window comes up. In addition to the PGI compiler commands, within BASH
you have access to over 100 common commands and utilities, including but not limited
to the following:
vi
gzip / gunzip
ftp
tar / untar
grep / egrep / fgrep
awk
sed
cksum
cp
cat
diff
du
date
kill
ls
find
mv
printenv / env
more / less
touch
wc
rm / rmdir
make
If you are familiar with program development in a Linux environment, editing,
compiling, and executing programs within bash will be very comfortable. If you have
not previously used such an environment, you might want to familiarize yourself
with vi or other editors and with makefiles. The Web has an extensive online
tutorial available for the vi editor as well as a number of thorough introductions to the
construction and use of makefiles.
ar or ranlib
For library compatibility, PGI provides versions of ar and ranlib that are compatible
with native Windows object-file formats. For more information on these commands,
refer to Creating and Using Static Libraries on Windows.
User's Guide for x86-64 CPUs
Version 2017 | 11
Getting Started
1.7.4. Using the PGI Compilers on macOS
PGI products for macOS support most of the features of the 32-and 64-bit versions
for Linux environments. Typically the PGI compilers and tools on macOS function
identically to their Linux counterparts.
macOS Header Files
The macOS header files contain numerous non-standard extensions. PGI supports
many of these extensions, thus allowing the PGI C and C++ compilers to compile most
programs that the GNU compilers can compile. A few header files not interoperable
with the PGI compilers have been rewritten. These files are included in $PGI/osx86/17.5/
include or $PGI/osx86-64/17.5/include. These files are: stdarg.h, stddef.h, and others.
If you are using the PGI C or C++ compilers, please make sure that the supplied versions
of these include files are found before the system versions. This will happen by default
unless you explicitly add a -I option that references one of the system include directories.
Apple macOS Debugging Requirements
Both the -g and -Mkeepobj switches play important roles when compiling a program
on macOS for debugging.
‣
‣
To debug a program with symbol information on macOS, files must be compiled
with the -g switch to keep the program's object files, the files with a ".o" extension.
Further, these object files must remain in the same directory in which they were
created.
If a program is built with separate compile and link steps, by compiling with the -c
switch which generates the ".o" object files, then using the -g switch guarantees the
required object files are available for debugging.
Use the following command sequence to compile and then link your code.
To compile the programs, use these commands:
pgcc -c -g main.c
pgcc -c -g foo.c
pgcc -c -g bar.c
To link, use this command:
pgcc -g main.o foo.o bar.o
Linking on macOS
On macOS, the PGI compilers do not support static linking of user binaries. For
compatibility with future Apple updates, the compilers support dynamic linking of
user binaries. For more information on dynamic linking, refer to Creating and Using
Dynamic Libraries on Apple macOX.
User's Guide for x86-64 CPUs
Version 2017 | 12
Getting Started
Running Parallel Programs on macOS
You may encounter difficulties running auto-parallel or OpenMP programs on
macOS systems when the per-thread stack size is set to the default (8MB). If you have
unexplained failures, please try setting the environment variable OMP_STACKSIZE to a
larger value, such as 16MB. For information on how to set environment variables, refer
to Setting Environment Variables.
1.8. Site-Specific Customization of the Compilers
If you are using the PGI compilers and want all your users to have access to specific
libraries or other files, there are special files that allow you to customize the compilers
for your site.
1.8.1. Use siterc Files
The PGI compiler drivers utilize a file named siterc to enable site-specific
customization of the behavior of the PGI compilers. The siterc file is located in the
bin subdirectory of the PGI installation directory. Using siterc, you can control how
the compiler drivers invoke the various components in the compilation tool chain.
1.8.2. Using User rc Files
In addition to the siterc file, user rc files can reside in a given user's home directory, as
specified by the user's HOME environment variable. You can use these files to control
the respective PGI compilers. All of these files are optional.
On Linux and macOS, these files are named .mypgf77rc, .mypgf90rc, .mypgccrc,
and .mypgc++rc.
On Windows, these files are named mypgf77rc, mypgf90rc, mypgf95rc,
mypgfortranrc and mypgccrc.
The following examples show how you can use these rc files to tailor a given installation
for a particular purpose.
Table 3 Examples of Usine siterc and User rc Files
To do this...
Add the line shown to the indicated file(s)
Make available to all linux86-64
set SITELIB=/opt/newlibs/64;
compilations the libraries found
in
/opt/newlibs/64
to
/opt/pgi/linux86-64/17.5/bin/siterc
Add to all linux86-64
append SITELIB=/opt/local/fast;
compilations a new library path:
/opt/local/fast
to
User's Guide for x86-64 CPUs
/opt/pgi/linux86-64/17.5/bin/siterc
Version 2017 | 13
Getting Started
To do this...
Add the line shown to the indicated file(s)
With linux86-64 compilations,
set MPILIBDIR=/opt/mympi/64;
/opt/
mympi/64/libmpix.a
change -Mmpi to link in
set MPILIBNAME=mpix;
to
With linux86-64 compilations,
always add
-DAMD
-DIS64BIT
/opt/pgi/linux86-64/17.5/bin/siterc
set SITEDEF=IS64BIT AMD;
to
/opt/pgi/linux86-64/17.5/bin/siterc
Build an F90 or F95 executable
set set RPATH=./REDIST;
for linux86-64 or linux86 that
resolves PGI shared objects
to
in the relative directory
REDIST
./
~/.mypgfortranrc
Note.This only affects the behavior of PGFORTRAN for the given
user.
1.9. Common Development Tasks
Now that you have a brief introduction to the compiler, let's look at some common
development tasks that you might wish to perform.
‣
‣
‣
‣
When you compile code you can specify a number of options on the command
line that define specific characteristics related to how the program is compiled and
linked, typically enhancing or overriding the default behavior of the compiler. For a
list of the most common command line options and information on all the command
line options, refer to Use Command-line Options.
Code optimization and parallelization allows the compiler to organize your code
for efficient execution. While possibly increasing compilation time and making the
code more difficult to debug, these techniques typically produce code that runs
significantly faster than code that does not use them. For more information on
optimization and parallelization, refer to Optimizing and Parallelizing.
Function inlining, a special type of optimization, replaces a call to a function or a
subroutine with the body of the function or subroutine. This process can speed up
execution by eliminating parameter passing and the function or subroutine call and
return overhead. In addition, function inlining allows the compiler to optimize the
function with the rest of the code. However, function inlining may also result in
much larger code size with no increase in execution speed. For more information on
function inlining, refer to Using Function Inlining.
Directives and pragmas allow users to place hints in the source code to help the
compiler generate better assembly code. You typically use directives and pragmas
to control the actions of the compiler in a particular portion of a program without
affecting the program as a whole. You place them in your source code where you
want them to take effect. A directive or pragma typically stays in effect from the
User's Guide for x86-64 CPUs
Version 2017 | 14
Getting Started
‣
‣
‣
‣
point where included until the end of the compilation unit or until another directive
or pragma changes its status. For more information on directives and pragmas, refer
to Using OpenMP and Using Directives and Pragmas.
A library is a collection of functions or subprograms used to develop software.
Libraries contain "helper" code and data, which provide services to independent
programs, allowing code and data to be shared and changed in a modular fashion.
The functions and programs in a library are grouped for ease of use and linking.
When creating your programs, it is often useful to incorporate standard libraries or
proprietary ones. For more information on this topic, refer to Creating and Using
Libraries.
Environment variables define a set of dynamic values that can affect the way
running processes behave on a computer. It is often useful to use these variables to
set and pass information that alters the default behavior of the PGI compilers and
the executables which they generate. For more information on these variables, refer
to Using Environment Variables.
Deployment, though possibly an infrequent task, can present some unique issues
related to concerns of porting the code to other systems. Deployment, in this context,
involves distribution of a specific file or set of files that are already compiled and
configured. The distribution must occur in such a way that the application executes
accurately on another system which may not be configured exactly the same as
the system on which the code was created. For more information on what you
might need to know to successfully deploy your code, refer to Distributing Files –
Deployment.
An intrinsic is a function available in a given language whose implementation
is handled specially by the compiler. Intrinsics make using processor-specific
enhancements easier because they provide a C/C++ language interface to assembly
instructions. In doing so, the compiler manages details that the user would normally
have to be concerned with, such as register names, register allocations, and memory
locations of data. For C/C++ programs, PGI provides support for MMX SSE, SSE2,
SSE3, SSSE3, SSE4A, ABM, and AVX intrinsics. For more information on these
intrinsics, refer to the C/C++ MMX/SSE Inline Intrinsics section of the PGI Compiler’s
Reference Manual.
User's Guide for x86-64 CPUs
Version 2017 | 15
Chapter 2.
USE COMMAND-LINE OPTIONS
A command line option allows you to control specific behavior when a program is
compiled and linked. This section describes the syntax for properly using command-line
options and provides a brief overview of a few of the more common options.
For a complete list of command-line options, their descriptions and use, refer to the
Command-Line Options Reference section of the PGI Compiler’s Reference Guide.
2.1. Command-line Option Overview
Before looking at all the command-line options, first become familiar with the syntax for
these options. There are a large number of options available to you, yet most users only
use a few of them. So, start simple and progress into using the more advanced options.
By default, the PGI compilers generate code that is optimized for the type of processor
on which compilation is performed, the compilation host. Before adding options to your
command-line, review Help with Command-line Options and Frequently-used Options.
2.1.1. Command-line Options Syntax
On a command-line, options need to be preceded by a hyphen (-). If the compiler
does not recognize an option, you get an unknown switch error. The error can be
downgraded to a warning by adding the -noswitcherror option.
This document uses the following notation when describing options:
[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.
User's Guide for x86-64 CPUs
Version 2017 | 16
Use Command-line Options
...
Horizontal ellipses indicate that zero or more instances of the preceding item are
valid.
Some options do not allow a space between the option and its argument or within
an argument. When applicable, the syntax section of the option description in the
Command-Line Options Reference section of the PGI Compiler’s Reference Guide .
2.1.2. Command-line Suboptions
Some options accept several suboptions. You can specify these suboptions either by
using the full option statement multiple times or by using a comma-separated list for the
suboptions.
The following two command lines are equivalent:
pgfortran -Mvect=simd -Mvect=noaltcode
pgfortran -Mvect=simd,noaltcode
2.1.3. Command-line Conflicting Options
Some options have an opposite or negated counterpart. For example, both -Mvect and
-Mnovect are available. -Mvect enables vectorization and -Mnovect disables it. If you
used both of these commands on a command line, they would conflict.
Rule: When you use conflicting options on a command line, the last encountered
option takes precedence over any previous one.
The conflicting options rule is important for a number of reasons.
‣
‣
Some options, such as -fast, include other options. Therefore, it is possible for you
to be unaware that you have conflicting options.
You can use this rule to create makefiles that apply specific flags to a set of files, as
shown in the following example.
Example: Makefiles with Options
In this makefile fragment, CCFLAGS uses vectorization. CCNOVECTFLAGS uses the
flags defined for CCFLAGS but disables vectorization.
CCFLAGS=c -Mvect=simd
CCNOVECTFLAGS=$(CCFLAGS) -Mnovect
2.2. Help with Command-line Options
If you are just getting started with the PGI compilers and tools, it is helpful to know
which options are available, when to use them, and which options most users find
effective.
User's Guide for x86-64 CPUs
Version 2017 | 17
Use Command-line Options
Using -help
The -help option is useful because it provides information about all options supported
by a given compiler.
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:
-help <command line option>
Suppose you use the following command to restrict the output to information about
the -fast option:
$ pgfortran -help -fast
The output you see is similar to:
-fast Common optimizations; includes -O2 -Munroll=c:1 -Mnoframe -Mlre
In the following example, we add the -help parameter to restrict the output to
information about the help command. The usage information for -help shows how
groups of options can be listed or examined according to function.
‣
$ pgfortran -help -help
-help[=groups|asm|debug|language|linker|opt|other|overall|phase|prepro|
suffix|switch|target|variable]
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>
By using the command pgfortran -help -help, as previously shown, we can see
output that shows the available subgroups. You can use the following command to
restrict the output on the -help command to information about only the options
related to only one group, such as debug information generation.
$ pgfortran -help=debug
The output you see is similar to this:
Debugging switches:
-M[no]bounds Generate code to check array bounds
-Mchkstk Check for sufficient stack space upon subprogram entry
-Mcoff Generate COFF format object
-Mdwarf1 Generate DWARF1 debug information with -g
-Mdwarf2 Generate DWARF2 debug information with -g
-Mdwarf3 Generate DWARF3 debug information with -g
-Melf Generate ELF format object
-g Generate information for debugger
-gopt Generate information for debugger without disabling
optimizations
For a complete description of subgroups, refer to the -help description in the
Command-line Options Reference section of the PGI Compiler Reference Manual.
User's Guide for x86-64 CPUs
Version 2017 | 18
Use Command-line Options
2.3. Getting Started with Performance
One of the top priorities of most users is performance and optimization. This section
provides a quick overview of a few of the command-line options that are useful in
improving performance.
2.3.1. Using -fast
PGI compilers implement a wide range of options that allow users a fine degree of
control on each optimization phase. When it comes to optimization of code, the quickest
way to start is to use the option -fast. These options create a generally optimal set of
flags. They incorporate optimization options to enable use of vector streaming SIMD
instructions for 64-bit targets. They enable vectorization with SIMD instructions, cache
alignment, and flush to zero mode.
The contents of the -fast option are host-dependent. Further, you should use these
options on both compile and link command lines.
The following table shows the typical -fast options.
Table 4 Typical -fast Options
Use this option...
To do this...
-O2
Specifies a code optimization level of 2.
-Munroll=c:1
Unrolls loops, executing multiple instances of the original loop during each
iteration.
-Mnoframe
Indicates to not generate code to set up a stack frame.
Note. With this option, a stack trace does not work.
-Mlre
Indicates loop-carried redundancy elimination.
-Mpre
Indicates partial redundancy elimination
-fast typically includes the options shown in this table:
Table 5 Additional -fast Options
Use this option...
To do this...
-Mvect=simd
Generates packed SIMD instructions.
-Mcache_align
Aligns long objects on cache-line boundaries.
-Mflushz
Sets flush-to-zero mode.
User's Guide for x86-64 CPUs
Version 2017 | 19
Use Command-line Options
Use this option...
To do this...
-M[no]vect
Controls automatic vector pipelining.
For best performance on processors that support SIMD instructions, use the
PGFORTRAN compiler, even for FORTRAN 77 code, and the -fast option.
To see the specific behavior of -fast for your target, use the following command:
$ pgfortran -help -fast
2.3.2. Other Performance-Related Options
While -fast is designed to be the quickest route to best performance, it is limited to
routine boundaries. Depending on the nature and writing style of the source code,
the compiler often can perform further optimization by knowing the global context of
usage of a given routine. For instance, determining the possible value range of actual
parameters of a routine could enable a loop to be vectorized; similarly, determining
static occurrence of calls helps to decide which routine is beneficial to inline.
These types of global optimizations are under control of Interprocedural Analysis (IPA)
in PGI compilers. Option -Mipa enables Interprocedural Analysis. -Mipa=fast is the
recommended option to get best performances for global optimization. You can also add
the suboption inline to enable automatic global inlining across files. You might consider
using -Mipa=fast,inline. This option for interprocedural analysis and global optimization
can improve performance.
As of the PGI 16.3 release, -Mipa has been disabled on Windows.
You may also obtain further performance improvements by experimenting with the
-M<pgflag> options described in the section ‘-M Options by Category’ section of the
PGI Compiler’s Reference Guide. These options include, but are not limited to, -Mvect,
-Munroll, -Minline, -Mconcur, -Mpfi and -Mpfo. However, performance improvements
using these options are typically application- and system-dependent. It is important
to time your application carefully when using these options to ensure no performance
degradations occur.
For more information on optimization, refer to Optimizing and Parallelizing. For specific
information about these options, refer to the ‘Optimization Controls’ section of the PGI
Compiler’s Reference Guide.
2.4. Targeting Multiple Systems — Using the -tp
Option
The -tp option allows you to set the target architecture. By default, the PGI compiler
uses all supported instructions wherever possible when compiling on a given system.
As a result, executables created on a given system may not be usable on previous
User's Guide for x86-64 CPUs
Version 2017 | 20
Use Command-line Options
generation systems. For example, executables created on a Pentium 4 may fail to execute
on a Pentium III or Pentium II.
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. Using a -tp flag option of k8 or p7 produces an executable that runs on most
x86-64 hardware in use today.
For more information about the -tp option, refer to the -tp <target>
[,target...] description in the Command-Line Options Reference section of the PGI
Compiler’s Reference Guide.
2.5. Frequently-used Options
In addition to overall performance, there are a number of other options that many users
find useful when getting started. The following table provides a brief summary of these
options.
For more information on these options, refer to the complete description of each option
available in the Command-Line Options Reference section of the PGI Compiler’s Reference
Guide. Also, there are a number of suboptions available with each of the -M options
listed. For more information on those options, refer to the specific section on M Options
by Category.
Table 6 Commonly Used Command-Line Options
Use this option...
To do this...
-fast
These options create a generally optimal set of flags for targets that
support SIMD capability. They incorporate optimization options to
enable use of vector streaming SIMD instructions (64-bit targets)
and enable vectorization with SIMD instructions, cache aligned and
flushz.
-fastsse
-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.
--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.
-help
Provides information about available options.
-mcmodel=medium
Enables medium=model core generation for 64-bit targets, which is
useful when the data space of the program exceeds 4GB.
-Mconcur
Instructs the compiler to enable auto-concurrentization of loops. If
specified, the compiler uses multiple processors to execute loops
that it determines to be parallelizable; thus, loop iterations are split
to execute optimally in a multithreaded execution context.
-Minfo
Instructs the compiler to produce information on standard error.
-Minline
User's Guide for x86-64 CPUs
Enables function inlining.
Version 2017 | 21
Use Command-line Options
Use this option...
To do this...
-Mipa=fast,inline
Enables interprocedural analysis and optimization. Also enables
automatic procedure inlining.
-Mpfi or -Mpfo
Enable profile feedback driven optimizations
-Mkeepasm
Keeps the generated assembly files.
-Munroll
Invokes the loop unroller to unroll loops, executing 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.
-M[no]vect
Enables/Disables the code vectorizer.
--[no_]exceptions
Removes exception handling from user code. For C++, declares that
the functions in this file generate no C++ exceptions, allowing more
optimal code generation.
-o
Names the output file.
-O <level>
Specifies code optimization level where <level> is 0, 1, 2, 3, or 4.
-tp <target>
[,target...]
Specify the target processor(s); for the 64-bit compilers, more than
one target is allowed, and enables generation of PGI Unified Binary
executables.
-Wl, <option>
Compiler driver passes the specified options to the linker.
User's Guide for x86-64 CPUs
Version 2017 | 22
Chapter 3.
OPTIMIZING AND PARALLELIZING
Source code that is readable, maintainable, and produces correct results is not always
organized for efficient execution. Normally, the first step in the program development
process involves producing code that executes and produces the correct results. This first
step usually involves compiling without much worry about optimization. After code is
compiled and debugged, code optimization and parallelization become an issue.
Invoking one of the PGI compiler commands with certain options instructs the compiler
to generate optimized code. Optimization is not always performed since it increases
compilation time and may make debugging difficult. However, optimization produces
more efficient code that usually runs significantly faster than code that is not optimized.
The compilers optimize code according to the specified optimization level. You can use
a number of options to specify the optimization levels, including -O, -Mvect, -Mipa, and
-Mconcur. In addition, you can use several of the -M<pgflag> switches to control specific
types of optimization and parallelization.
As of the PGI 16.3 release, -Mipa has been disabled on Windows.
This chapter describes these optimization options:
-fast
-Minline
-O
-Munroll
-Mconcur
-Mipa=fast
-Mpfi
-Mvect
-Minfo
-Mneginfo
-Mpfo
-Msafeptr
-fast
-Minline
-O
-Munroll
-Mconcur
-Mpfi
-Mvect
-Minfo
-Mneginfo
-Mpfo
-Mipa=fast,inline
This chapter also describes how to choose optimization options to use with the PGI
compilers. This overview is helpful if you are just getting started with one of the PGI
compilers, or wish to experiment with individual optimizations.
User's Guide for x86-64 CPUs
Version 2017 | 23
Optimizing and Parallelizing
Complete specifications of each of these options is available in the Command-Line Options
Reference section of the PGI Compiler Reference Manual.
3.1. Overview of Optimization
In general, optimization involves using transformations and replacements that generate
more efficient code. This is done by the compiler and involves replacements that are
independent of the particular target processor's architecture as well as replacements that
take advantage of the x86-64 architecture, instruction set and registers.
For discussion purposes, we categorize optimization:
Local Optimization
Global Optimization
Loop Optimization
Interprocedural Analysis (IPA) and Optimization
Optimization Through Function Inlining
Profile Feedback Optimization (PFO)
3.1.1. Local Optimization
A basic block is a sequence of statements in which the flow of control enters at the
beginning and leaves at the end without the possibility of branching, except at the end.
Local optimization is performed on a block-by-block basis within a program’s basic
blocks.
The PGI compilers perform many types of local optimization including: algebraic
identity removal, constant folding, common sub-expression elimination, redundant load
and store elimination, scheduling, strength reduction, and peephole optimizations.
3.1.2. Global Optimization
This optimization is performed on a subprogram/function over all its basic blocks. The
optimizer performs control-flow and data-flow analysis for an entire program unit. All
loops, including those formed by ad hoc branches such as IFs or GOTOs, are detected
and optimized.
Global optimization includes: constant propagation, copy propagation, dead store
elimination, global register allocation, invariant code motion, and induction variable
elimination.
3.1.3. Loop Optimization: Unrolling, Vectorization and
Parallelization
The performance of certain classes of loops may be improved through vectorization
or unrolling options. Vectorization transforms loops to improve memory access
performance and make use of packed SSE instructions which perform the same
operation on multiple data items concurrently. Unrolling replicates the body of loops to
reduce loop branching overhead and provide better opportunities for local optimization,
User's Guide for x86-64 CPUs
Version 2017 | 24
Optimizing and Parallelizing
vectorization and scheduling of instructions. Performance for loops on systems with
multiple processors may also improve using the parallelization features of the PGI
compilers.
3.1.4. Interprocedural Analysis (IPA) and Optimization
Interprocedural analysis (IPA) allows use of information across function call boundaries
to perform optimizations that would otherwise be unavailable. For example, if the
actual argument to a function is in fact a constant in the caller, it may be possible to
propagate that constant into the callee and perform optimizations that are not valid if
the dummy argument is treated as a variable. A wide range of optimizations are enabled
or improved by using IPA, including but not limited to data alignment optimizations,
argument removal, constant propagation, pointer disambiguation, pure function
detection, F90/F95 array shape propagation, data placement, empty function removal,
automatic function inlining, inlining of functions from pre-compiled libraries, and
interprocedural optimization of functions from pre-compiled libraries.
3.1.5. Function Inlining
This optimization allows a call to a function to be replaced by a copy of the body of
that function. This optimization will sometimes speed up execution by eliminating
the function call and return overhead. Function inlining may also create opportunities
for other types of optimization. Function inlining is not always beneficial. When used
improperly it may increase code size and generate less efficient code.
3.1.6. Profile-Feedback Optimization (PFO)
Profile-feedback optimization (PFO) makes use of information from a trace file
produced by specially instrumented executables which capture and save information on
branch frequency, function and subroutine call frequency, semi-invariant values, loop
index ranges, and other input data dependent information that can only be collected
dynamically during execution of a program.
By definition, use of profile-feedback optimization is a two-phase process: compilation
and execution of a specially-instrumented executable, followed by a subsequent
compilation which reads a trace file generated during the first phase and uses the
information in that trace file to guide compiler optimizations.
3.2. Getting Started with Optimization
The first concern should be getting the program to execute and produce correct results.
To get the program running, start by compiling and linking without optimization. Add
-O0 to the compile line to select no optimization; or add -g to debug the program easily
and isolate any coding errors exposed during porting to x86-64 platforms. For more
information on debugging, refer to the PGI Debugger Guide.
User's Guide for x86-64 CPUs
Version 2017 | 25
Optimizing and Parallelizing
To get started quickly with optimization, a good set of options to use with any of the PGI
compilers is -fast . For example:
$ pgfortran -fast -Mipa=fast,inline prog.f
As of the PGI 16.3 release,
-Mipa
has been disabled on Windows.
For all of the PGI Fortran, C, and C++ compilers, the -fast -Mipa=fast,inline options
generally produce code that is well-optimized without the possibility of significant
slowdowns due to pathological cases.
‣
‣
The-fast option is an aggregate option that includes a number of individual PGI
compiler options; which PGI compiler options are included depends on the target
for which compilation is performed.
The -Mipa=fast,inline option invokes interprocedural analysis (IPA), including
several IPA suboptions. The inline suboption enables automatic inlining with IPA. If
you do not wish to use automatic inlining, you can compile with -Mipa=fast and use
several IPA suboptions without inlining.
By experimenting with individual compiler options on a file-by-file basis, further
significant performance gains can sometimes be realized. However, depending on the
coding style, individual optimizations can sometimes cause slowdowns, and must be
used carefully to ensure performance improvements.
There are other useful command line options related to optimization and parallelization,
such as -help, -Minfo, -Mneginfo, -dryrun, and -v.
3.2.1. -help
As described in Help with Command-Line Options, you can see a specification of any
command-line option by invoking any of the PGI compilers with -help in combination
with the option in question, without specifying any input files.
For example, you might want information on -O:
$ pgfortran
-help -O
The resulting output is similar to this:
-O Set opt level. All -O1 optimizations plus traditional scheduling and
global scalar optimizations performed
Or you can see the full functionality of -help itself, which can return information on
either an individual option or groups of options:
$ pgfortran
-help -help
The resulting output is similar to this:
-help[=groups|asm|debug|language|linker|opt|other|overall|
phase|prepro|suffix|switch|target|variable]
Show compiler switches
User's Guide for x86-64 CPUs
Version 2017 | 26
Optimizing and Parallelizing
3.2.2. -Minfo
You can use the -Minfo option to display compile-time optimization listings. When
this option is used, the PGI compilers issue informational messages to standard error
(stderr) as compilation proceeds. From these messages, you can determine which
loops are optimized using unrolling, SSE instructions , vectorization, parallelization,
interprocedural optimizations and various miscellaneous optimizations. You can also see
where and whether functions are inlined.
For more information on -Minfo, refer to Optimization Controls section of the PGI
Compiler Reference Manual.
3.2.3. -Mneginfo
You can use the -Mneginfo option to display informational messages to standard error
(stderr) that explain why certain optimizations are inhibited.
For more information on -Mneginfo, refer to Optimization Controls section of the PGI
Compiler Reference Manual.
3.2.4. -dryrun
The -dryrun option can be useful as a diagnostic tool if you need to see the steps used
by the compiler driver to preprocess, compile, assemble and link in the presence of a
given set of command line inputs. When you specify the -dryrun option, these steps
are printed to standard error (stderr) but are not actually performed. For example, you
can use this option to inspect the default and user-specified libraries that are searched
during the link phase, and the order in which they are searched by the linker.
3.2.5. -v
The -v option is similar to -dryrun, except each compilation step is performed and not
simply printed.
3.2.6. PGI Profiler
The PGI profiler is a profiling tool that provides a way to visualize the performance
of the components of your program. Using tables and graphs, the profiler associates
execution time and resource utilization data with the source code and instructions of
your program. This association allows you to see where a program’s execution time is
spent. Through resource utilization data and compiler analysis information, the profiler
helps you to understand why certain parts of your program have high execution times.
This information may help you with selecting which optimization options to use with
your program.
The profiler also allows you to correlate the messages produced by -Minfo and
-Mneginfo, described above, to your program’s source code. This feature is known as
the Common Compiler Feedback Format (CCFF).
For more information on the profiler, refer to the PGI Profiler Guide.
User's Guide for x86-64 CPUs
Version 2017 | 27
Optimizing and Parallelizing
3.3. Common Compiler Feedback Format (CCFF)
Using the Common Compiler Feedback Format (CCFF), PGI compilers save information
about how your program was optimized, or why a particular optimization was not
made, in the executable file. To append this information to the object file, use the
compiler option -Minfo=ccff.
If you choose to use the PGI profiler to aid with your optimization, it can extract this
information and associate it with source code and other performance data, allowing you
to view all of this information simultaneously in one of the available profiler panels.
3.4. Local and Global Optimization
This section describes local and global optimization.
3.4.1. -Msafeptr
The -Msafeptr option can significantly improve performance of C/C++ programs in
which there is known to be no pointer aliasing. For obvious reasons, this command-line
option must be used carefully. There are a number of suboptions for -Msafeptr:
‣
‣
‣
‣
‣
-Msafeptr=all – All pointers are safe. Equivalent to the default setting:
-Msafeptr.
-Msafeptr=arg – Function formal argument pointers are safe. Equivalent to
-Msafeptr=dummy.
-Msafeptr=global – Global pointers are safe.
-Msafeptr=local – Local pointers are safe. Equivalent to -Msafeptr=auto.
-Msafeptr=static – Static local pointers are safe.
If your C/C++ program has pointer aliasing and you also want automating inlining,
then compiling with -Mipa=fast or -Mipa=fast,inline includes pointer aliasing
optimizations. IPA may be able to optimize some of the alias references in your program
and leave intact those that cannot be safely optimizied.
3.4.2. -O
Using the PGI compiler commands with the -O<level> option (the capital O is for
Optimize), you can specify any integer level from 0 to 4.
-O0
Level zero specifies no optimization. A basic block is generated for each language
statement. At this level, the compiler generates a basic block for each statement.
Performance will almost always be slowest using this optimization level. This level
is useful for the initial execution of a program. It is also useful for debugging, since
User's Guide for x86-64 CPUs
Version 2017 | 28
Optimizing and Parallelizing
there is a direct correlation between the program text and the code generated. To enable
debugging, include -g on your compile line.
-O1
Level one specifies local optimization. Scheduling of basic blocks is performed. Register
allocation is performed.
Local optimization is a good choice when the code is very irregular, such as code that
contains many short statements containing IF statements and does not contain loops
(DO or DO WHILE statements ). Although this case rarely occurs, for certain types of
code, this optimization level may perform better than level-two (-O2).
-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, more
advanced optimizations such as SIMD code generation, cache alignment, and partial
redundancy elimination are enabled.
-O3
Level three specifies aggressive global optimization. This level performs all levelone 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.
Types of Optimizations
The PGI compilers perform many different types of local optimizations, including but
not limited to:
Algebraic identity removal
Constant folding
Common subexpression elimination
Local register optimization
Peephole optimizations
User's Guide for x86-64 CPUs
Version 2017 | 29
Optimizing and Parallelizing
Redundant load and store elimination
Strength reductions
Level-two optimization (-O2 or -O) specifies global optimization. The -fast option
generally specifies global optimization; however, the -fast switch varies from release
to release, depending on a reasonable selection of switches for any one particular
release. The -O or -O2 level performs all level-one local optimizations as well as global
optimizations. Control flow analysis is applied and global registers are allocated for
all functions and subroutines. Loop regions are given special consideration. This
optimization level is a good choice when the program contains loops, the loops are
short, and the structure of the code is regular.
The PGI compilers perform many different types of global optimizations, including but
not limited to:
Branch to branch elimination
Constant propagation
Copy propagation
Dead store elimination
Global register allocation
Induction variable elimination
Invariant code motion
You can explicitly select the optimization level on the command line. For example,
the following command line specifies level-two optimization which results in global
optimization:
$ pgfortran
-O2 prog.f
The default optimization level changes depending on which options you select on the
command line. For example, when you select the -g debugging option, the default
optimization level is set to level-zero (-O0). However, if you need to debug optimized
code, you can use the -gopt option to generate debug information without perturbing
optimization. For a description of the default levels, refer to Default Optimization
Levels.
The -fast option includes -O2 on all targets. If you want to override the default for
-fast with -O3 while maintaining all other elements of -fast, simply compile as
follows:
$ pgfortran
-fast -O3 prog.f
3.5. Loop Unrolling using -Munroll
This optimization unrolls loops, which reduces branch overhead, and can improve
execution speed by creating better opportunities for instruction scheduling. A loop
with a constant count may be completely unrolled or partially unrolled. A loop with a
User's Guide for x86-64 CPUs
Version 2017 | 30
Optimizing and Parallelizing
non-constant count may also be unrolled. A candidate loop must be an innermost loop
containing one to four blocks of code.
The following example shows the use of the -Munroll option:
$ pgfortran -Munroll prog.f
The -Munroll option is included as part of -fast on all targets. The loop unroller
expands the contents of a loop and reduces the number of times a loop is executed.
Branching overhead is reduced when a loop is unrolled two or more times, since each
iteration of the unrolled loop corresponds to two or more iterations of the original loop;
the number of branch instructions executed is proportionately reduced. When a loop is
unrolled completely, the loop’s branch overhead is eliminated altogether.
Loop unrolling may be beneficial for the instruction scheduler. When a loop is
completely unrolled or unrolled two or more times, opportunities for improved
scheduling may be presented. The code generator can take advantage of more
possibilities for instruction grouping or filling instruction delays found within the loop.
Examples Showing Effect of Unrolling
The following side-by-side examples show the effect of code unrolling on a segment that
computes a dot product.
This example is only meant to represent how the compiler can transform the loop;
it is not meant to imply that the programmer needs to manually change code. In
fact, manually unrolling your code can sometimes inhibit the compiler’s analysis and
optimization.
Table 7 Example of Effect of Code Unrolling
Dot Product Code
REAL*4 A(100), B(100), Z
INTEGER I
DO I=1, 100
Z = Z + A(i) * B(i)
END DO
END
Unrolled Dot Product Code
REAL*4 A(100), B(100), Z
INTEGER I
DO I=1, 100, 2
Z = Z + A(i) * B(i)
Z = Z + A(i+1) * B(i+1)
END DO
END
Using the -Minfo option, the compiler informs you when a loop is being unrolled. For
example, a message similar to the following, indicating the line number, and the number
of times the code is unrolled, displays when a loop is unrolled:
dot:
5, Loop unrolled 5 times
Using the c:<m> and n:<m> sub-options to -Munroll, or using -Mnounroll, you can
control whether and how loops are unrolled on a file-by-file basis. Using directives or
pragmas , you can precisely control whether and how a given loop is unrolled. For more
information on -Munroll, refer to Use Command-line Options.
User's Guide for x86-64 CPUs
Version 2017 | 31
Optimizing and Parallelizing
3.6. Vectorization using -Mvect
The -Mvect option is included as part of -fast on all targets. If your program contains
computationally-intensive loops, the -Mvect option may be helpful. If in addition you
specify -Minfo, and your code contains loops that can be vectorized, the compiler reports
relevant information on the optimizations applied.
When a PGI compiler command is invoked with the -Mvect option, the vectorizer scans
code searching for loops that are candidates for high-level transformations such as
loop distribution, loop interchange, cache tiling, and idiom recognition (replacement
of a recognizable code sequence, such as a reduction loop, with optimized code
sequences or function calls). When the vectorizer finds vectorization opportunities,
it internally rearranges or replaces sections of loops (the vectorizer changes the code
generated; your source code’s loops are not altered). In addition to performing these
loop transformations, the vectorizer produces extensive data dependence information
for use by other phases of compilation and detects opportunities to use vector or
packed Streaming SIMD Extensions (SSE) instructions on x86 processors where these are
supported.
The -Mvect option can speed up code which contains well-behaved countable loops
which operate on large floating point arrays in Fortran and their C/C++ counterparts.
However, it is possible that some codes will show a decrease in performance when
compiled with the -Mvect option due to the generation of conditionally executed code
segments, inability to determine data alignment, and other code generation factors. For
this reason, it is recommended that you check carefully whether particular program
units or loops show improved performance when compiled with this option enabled.
3.6.1. Vectorization Sub-options
The vectorizer performs high-level loop transformations on countable loops. A loop
is countable if the number of iterations is set only before loop execution and cannot
be modified during loop execution. Some of the vectorizer transformations can be
controlled by arguments to the -Mvect command line option. The following sections
describe the arguments that affect the operation of the vectorizer. In addition, some of
these vectorizer operations can be controlled from within code using directives and
pragmas. For details on the use of directives and pragmas, refer to Using Directives and
Pragmas.
The vectorizer performs the following operations:
‣
‣
‣
‣
‣
‣
‣
‣
Loop interchange
Loop splitting
Loop fusion
Memory-hierarchy (cache tiling) optimizations
Generation of SSE instructions on processors where these are supported
Generation of prefetch instructions on processors where these are supported
Loop iteration peeling to maximize vector alignment
Alternate code generation
User's Guide for x86-64 CPUs
Version 2017 | 32
Optimizing and Parallelizing
By default, -Mvect without any sub-options is equivalent to:
-Mvect=assoc,cachesize=c
where c is the actual cache size of the machine.
This enables the options for nested loop transformation and various other vectorizer
options. These defaults may vary depending on the target system. The following table
lists and briefly describes some of the -Mvect suboptions.
Table 8 -Mvect Suboptions
Use this option ...
To instruct the vectorizer to do this ...
-Mvect=altcode
Generate appropriate code for vectorized loops.
-Mvect=[no]assoc
Perform[disable] associativity conversions that
can change the results of a computation due
to a round-off error. For example, a typical
optimization is to change one arithmetic
operation to another arithmetic operation
that is mathematically correct, but can be
computationally different and generate faster
code. This option is provided to enable or disable
this transformation, since a round-off error
for such associativity conversions may produce
unacceptable results.
-Mvect=cachesize:n
Tiles nested loop operations, assuming a data
cache size of n bytes. By default, the vectorizer
attempts to tile nested loop operations, such as
matrix multiply, using multi-dimensional stripmining techniques to maximize re-use of items in
the data cache.
-Mvect=fuse
Enable loop fusion.
-Mvect=gather
Enable vectorization of indirect array references.
-Mvect=idiom
Enable idiom recognition.
-Mvect=levels:<n>
Set the maximum next level of loops to optimize.
-Mvect=nocond
Disable vectorization of loops with conditions.
-Mvect=partial
Enable partial loop vectorization via inner loop
distribution.
-Mvect=prefetch
Automatically generate prefetch instructions when
vectorizable loops are encountered, even in cases
where SSE instructions are not generated.
-Mvect=short
Enable short vector operations.
-Mvect=simd
Automatically generate packed SSE (Streaming
SIMD Extensions), and prefetch instructions
when vectorizable loops are encountered. SIMD
instructions, first introduced on Pentium III and
AthlonXP processors, operate on single-precision
floating-point data.
-Mvect=sizelimit:n
Limit the size of vectorized loops.
-Mvect=sse
Equivalent to -Mvect=simd.
User's Guide for x86-64 CPUs
Version 2017 | 33
Optimizing and Parallelizing
Use this option ...
To instruct the vectorizer to do this ...
-Mvect=tile
Enable loop tiling.
-Mvect=uniform
Perform consistent optimizations in both
vectorized and residual loops. Be aware that this
may affect the performance of the residual loop.
Inserting no in front of the option disables the option. For example, to disable the
generation of SSE (or SIMD) instructions on x86, compile with -Mvect=nosimd.
3.6.2. Vectorization Example Using SIMD Instructions
One of the most important vectorization options is -Mvect=simd. When you use this
option, the compiler automatically generates SSE instructions, where possible, when
targeting x86 processors on which these instructions are supported. This process can
improve performance by several factors compared with the equivalent scalar code. All
of the PGI Fortran, C and C++ compilers support this capability. The PGI Release Notes
show which x86 and x64 processors PGI supports.
In the program in Vector operation using SIMD instructions, the vectorizer recognizes
the vector operation in subroutine 'loop' when either the compiler switch -Mvect=simd
or -fast is used. This example shows the compilation, informational messages, and
runtime results using the SSE instructions on a x86 4 Core Intel Sandybridge 2.5 GHz
system, along with issues that affect SSE performance.
Loops vectorized using x86 SSE instructions operate much more efficiently when
processing vectors that are aligned to a cache-line boundary. You can cause
unconstrained data objects of size 16 bytes or greater to be cache-aligned by compiling
with the -Mcache_align switch. An unconstrained data object is a data object that is not a
common block member and not a member of an aggregate data structure.
For stack-based local variables to be properly aligned, the main program or function
must be compiled with -Mcache_align.
The -Mcache_align switch has no effect on the alignment of Fortran allocatable or
automatic arrays. If you have arrays that are constrained, such as vectors that are
members of Fortran common blocks, you must specifically pad your data structures
to ensure proper cache alignment. You can use -Mcache_align for only the beginning
address of each common block to be cache-aligned.
The following examples show the results of compiling the sample code in Vector
operation using SIMD instructions both with and without the option -Mvect=simd.
User's Guide for x86-64 CPUs
Version 2017 | 34
Optimizing and Parallelizing
Vector operation using SIMD instructions
program vector_op
parameter (N = 9999)
real*4 x(N), y(N), z(N), W(N)
do i = 1, n
y(i) = i
z(i) = 2*i
w(i) = 4*i
enddo
do j = 1, 200000
call loop(x,y,z,w,1.0e0,N)
enddo
print *, x(1),x(771),x(3618),x(6498),x(9999)
end
subroutine loop(a,b,c,d,s,n)
integer i, n
real*4 a(n), b(n), c(n), d(n),s
do i = 1, n
a(i) = b(i) + c(i) - s * d(i)
enddo
end
Assume the preceding program is compiled as follows, where -Mvect=nosimd disables
x86 SSE vectorization:
% pgfortran -fast -Mvect=nosimd -Minfo vadd.f -Mfree -o vadd
vector_op:
4, Loop unrolled 16 times
Generates 1 prefetches in scalar loop
9, Loop not vectorized/parallelized: contains call
loop:
18, Loop unrolled 4 times
The following output shows a sample result if the generated executable is run and timed
on a x86 4 Core Intel Sandybridge 2.5 GHz system:
% /bin/time vadd
-1.000000 -771.000 -3618.000 -6498.00 -9999.00
1.35user 0.00system 0:01.35elapsed 99%CPU (0avgtext+0avgdata 3936maxresident)k
0inputs+0outputs (0major+290minor)pagefaults 0swaps
Now, recompile with x86 SSE vectorization enabled, and you see results similar to these:
% pgfortran -fast -Minfo vadd.f -Mfree -o vadd
vector_op:
4, Loop not vectorized: may not be beneficial
Unrolled inner loop 8 times
Residual loop unrolled 7 times (completely unrolled)
Generated 1 prefetches in scalar loop
9, Loop not vectorized/parallelized: contains call
loop:
17, Generated 4 alternate versions of the loop
Generated vector sse code for the loop
Generated 3 prefetch instructions for the loop
Notice the informational message for the loop at line 17.
‣
The first two lines of the message indicate that the loop was vectorized, x86 SSE
instructions were generated, and four alternate versions of the loop were also
generated. The loop count and alignments of the arrays determine which of these
versions is executed.
User's Guide for x86-64 CPUs
Version 2017 | 35
Optimizing and Parallelizing
‣
The last line of the informational message indicates that prefetch instructions have
been generated for three loads to minimize latency of data transfers from main
memory.
Executing again, you should see results similar to the following:
% /bin/time vadd
-1.000000 -771.000 -3618.00 -6498.00 -9999.0
0.60user 0.00system 0:00.61elapsed 99%CPU (0avgtext+0avgdata 3920maxresident)k
0inputs+0outputs (0major+289minor)pagefaults 0swaps
The SIMD result is 2.25 times faster than the equivalent non-SIMD version of the
program.
Speed-up realized by a given loop or program can vary widely based on a number of
factors:
‣
‣
‣
‣
When the vectors of data are resident in the data cache, performance improvement
using vector x86 SSE or SSE2 instructions is most effective.
If data is aligned properly, performance will be better in general than when using
vector x86 SSE operations on unaligned data.
If the compiler can guarantee that data is aligned properly, even more efficient
sequences of x86 SSE instructions can be generated.
The efficiency of loops that operate on single-precision data can be higher. x86 SSE2
vector instructions can operate on four single-precision elements concurrently, but
only two double-precision elements.
Compiling with -Mvect=simd can result in numerical differences from the executables
generated with less optimization. Certain vectorizable operations, for example dot
products, are sensitive to order of operations and the associative transformations
necessary to enable vectorization (or parallelization).
3.7. Auto-Parallelization using -Mconcur
With the -Mconcur option the compiler scans code searching for loops that are
candidates for auto-parallelization. -Mconcur must be used at both compile-time
and link-time. When the parallelizer finds opportunities for auto-parallelization, it
parallelizes loops and you are informed of the line or loop being parallelized if the
-Minfo option is present on the compile line. For a complete specification of -Mconcur,
refer to the ‘Optimization Controls’ section of the PGI Compiler Reference Manual.
A loop is considered parallelizable if it doesn't contain any cross-iteration data
dependencies. Cross-iteration dependencies from reductions and expandable scalars
are excluded from consideration, enabling more loops to be parallelizable. In general,
loops with calls are not parallelized due to unknown side effects. Also, loops with
low trip counts are not parallelized since the overhead in setting up and starting a
parallel loop will likely outweigh the potential benefits. In addition, the default is not
to parallelize innermost loops, since these often by definition are vectorizable and
User's Guide for x86-64 CPUs
Version 2017 | 36
Optimizing and Parallelizing
it is seldom profitable to both vectorize and parallelize the same loop, especially on
multicore processors. Compiler switches and directives are available to let you override
most of these restrictions on auto-parallelization.
3.7.1. Auto-Parallelization Sub-options
The parallelizer performs various operations that can be controlled by arguments to
the -Mconcur command line option. The following sections describe these arguments
that affect the operation of the parallelizer. In addition, these parallelizer operations can
be controlled from within code using directives and pragmas. For details on the use of
directives and pragmas, refer to Using Directives and Pragmas.
By default, -Mconcur without any sub-options is equivalent to:
-Mconcur=dist:block
This enables parallelization of loops with blocked iteration allocation across the available
threads of execution. These defaults may vary depending on the target system. The
following table lists and briefly describes some of the -Mconcur suboptions.
Table 9 -Mconcur Suboptions
Use this option ...
To instruct the parallelizer to do this...
-Mconcur=allcores
Use all available cores. Specify this option at link
time.
-Mconcur=[no]altcode
Generate [do not 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. Specifying noaltcode disables this
option and no alternate serial code is generated.
-Mconcur=[no]assoc
Enable [disable] parallelization of loops with
associative reductions.
-Mconcur=bind
Bind threads to cores. Specify this option at link
time.
-Mconcur=cncall
Specifies that it is safe to parallelize loops that
contain subroutine or function calls. By default,
such loops are excluded from consideration for
auto-parallelization. Also, no minimum loop count
threshold must be satisfied before parallelization
occurs, and last values of scalars are assumed to
be safe.
-Mconcur=dist:{block|cyclic}
Specifies whether to assign loop iterations to the
available threads in blocks or in a cyclic (roundrobin) fashion. Block distribution is the default.
If cyclic is specified, iterations are allocated to
processors cyclically. That is, processor 0 performs
iterations 0, 3, 6, etc.; processor 1 performs
User's Guide for x86-64 CPUs
Version 2017 | 37
Optimizing and Parallelizing
Use this option ...
To instruct the parallelizer to do this...
iterations 1, 4, 7, etc.; and processor 2 performs
iterations 2, 5, 8, etc.
-Mconcur=innermost
-Mconcur=levels:<n>
-Mconcur=[no]numa
Enable parallelization of innermost loops.
Parallelize loops nested at most n levels deep.
Use thread/processors affinity when
running on a NUMA architecture. Specifying
-Mconcur=nonuma disables this option.
The environment variable NCPUS is checked at runtime for a parallel program. If
NCPUS is set to 1, a parallel program runs serially, but will use the parallel routines
generated during compilation. If NCPUS is set to a value greater than 1, the specified
number of processors are used to execute the program. Setting NCPUS to a value
exceeding the number of physical processors can produce inefficient execution.
Executing a program on multiple processors in an environment where some of the
processors are being time-shared with another executing job can also result in inefficient
execution.
As with the vectorizer, the -Mconcur option can speed up code if it contains wellbehaved countable loops and/or computationally intensive nested loops that operate
on arrays. However, it is possible that some codes show a decrease in performance
on multi-processor systems when compiled with -Mconcur due to parallelization
overheads, memory bandwidth limitations in the target system, false-sharing of
cache lines, or other architectural or code-generation factors. For this reason, it is
recommended that you check carefully whether particular program units or loops show
improved performance when compiled using this option.
If the compiler is not able to successfully auto-parallelize your application, you should
refer to Using OpenMP. It is possible that insertion of explicit parallelization directives
or pragmas, and use of the compiler option -mp might enable the application to run in
parallel.
3.7.2. Loops That Fail to Parallelize
In spite of the sophisticated analysis and transformations performed by the compiler,
programmers may notice loops that are seemingly parallel, but are not parallelized. In
this subsection, we look at some examples of common situations where parallelization
does not occur.
Innermost Loops
As noted earlier in this section, the PGI compilers will not parallelize innermost loops by
default, because it is usually not profitable. However, you can override this default using
the -Mconcur=innermost command-line option.
User's Guide for x86-64 CPUs
Version 2017 | 38
Optimizing and Parallelizing
Timing Loops
Often, loops occur in programs that are similar to timing loops. The outer loop in the
following example is one such loop.
do j = 1, 2
do i = 1, n
1
a(i) = b(i) + c(i)
enddo
enddo
The outer loop in the preceding example is not parallelized because the compiler detects
a cross-iteration dependence in the assignment to a(i). Suppose the outer loop were
parallelized. Then both processors would simultaneously attempt to make assignments
into a(1:n). Now in general the values computed by each processor for a(1:n) will
differ, so that simultaneous assignment into a(1:n) will produce values different from
sequential execution of the loops.
In this example, values computed for a(1:n) don’t depend on j, so that simultaneous
assignment by both processors does not yield incorrect results. However, it is beyond
the scope of the compilers’ dependence analysis to determine that values computed
in one iteration of a loop don’t differ from values computed in another iteration. So
the worst case is assumed, and different iterations of the outer loop are assumed to
compute different values for a(1:n). Is this assumption too pessimistic? If j doesn’t
occur anywhere within a loop, the loop exists only to cause some delay, most probably
to improve timing resolution. It is not usually valid to parallelize timing loops; to do so
would distort the timing information for the inner loops.
Scalars
Quite often, scalars will inhibit parallelization of non-innermost loops. There are two
separate cases that present problems. In the first case, scalars appear to be expandable,
but appear in non-innermost loops, as in the following example.
do j = 1, n
x = b(j)
do i = 1, n
a(i,j) = x + c(i,j)
enddo
enddo
There are a number of technical problems to be resolved in order to recognize
expandable scalars in non-innermost loops. Until this generalization occurs, scalars
like x in the preceding code segment inhibit parallelization of loops in which they
are assigned. In the following example, scalar k is not expandable, and it is not an
accumulator for a reduction.
1
2
k = 1
do i = 1, n
do j = 1, n
a(j,i) = b(k) * x
enddo
k = i
if (i .gt. n/2) k = n - (i - n/2)
enddo
User's Guide for x86-64 CPUs
Version 2017 | 39
Optimizing and Parallelizing
If the outer loop is parallelized, conflicting values are stored into k by the various
processors. The variable k cannot be made local to each processor because its value
must remain coherent among the processors. It is possible the loop could be parallelized
if all assignments to k are placed in critical sections. However, it is not clear where
critical sections should be introduced because in general the value for k could depend on
another scalar (or on k itself), and code to obtain the value of other scalars must reside in
the same critical section.
In the previous example, the assignment to k within a conditional at label 2 prevents k
from being recognized as an induction variable. If the conditional statement at label 2 is
removed, k would be an induction variable whose value varies linearly with j, and the
loop could be parallelized.
Scalar Last Values
During parallelization, scalars within loops often need to be privatized; that is, each
execution thread has its own independent copy of the scalar. Problems can arise if a
privatized scalar is accessed outside the loop. For example, consider the following loops
in C/C++ and Fortran:
/* C/C++ version */
for (i = 1; i<N; i++){
if( x[i] > 5.0 )
t = x[i];
}
v = t;
f(v);
! Fortran version
do I = 1,N
if (x(I) > 5.0 ) then
t = x(I)
endif
enddo
v = t
call f(v)
The value of t may not be computed on the last iteration of the loop. Normally, if a
scalar is assigned within a loop and used following the loop, the PGI compilers save
the last value of the scalar. However, if the loop is parallelized and the scalar is not
assigned on every iteration, it may be difficult, without resorting to costly critical
sections, to determine on what iteration t is last assigned. Analysis allows the compiler
to determine that a scalar is assigned on each iteration and hence that the loop is safe
to parallelize if the scalar is used later, as illustrated in the following C/C++ and Fortran
examples.
User's Guide for x86-64 CPUs
Version 2017 | 40
Optimizing and Parallelizing
/* C/C++ version */
for (i=1;i<n;i++) {
if (x[i]>0.0) {
t=2.0;
}
else {
t=3.0;
y[i]=t;
}
}
v=t;
! Fortran version
do I = 1,N
if (x(I)>0.0) then
t=2.0
else
t=3.0
y(i)=t
endif
enddo
v=t
Notice that t is assigned on every iteration of the loop. However, there are cases where
a scalar may be privatizable, but if it is used after the loop, it is unsafe to parallelize.
Examine the following loops in which each use of t within the loop is reached by a
definition from the same iteration.
/* C/C++ Version */
for (i=1;i<N;i++){
if(x[i]>0.0){
t=x[i];
y[i]=t;
}
}
v=t;
f(v);
! Fortran Version
do I = 1,N
if (x(I)>0.0) then
t=x(I)
y(i)=t
endif
enddo
v=t
call f(v)
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 last
assigned. The compiler detects the previous cases. When a scalar is used after the loop
but is not defined on every iteration of the loop, parallelization does not occur.
When the programmer knows that the scalar is assigned on the last iteration of the loop,
the programmer may use a directive or pragma to let the compiler know the loop is safe
to parallelize. The directive or pragma safe_lastval informs the compiler that, for
User's Guide for x86-64 CPUs
Version 2017 | 41
Optimizing and Parallelizing
a given loop, all scalars are assigned in the last iteration of the loop; thus, it is safe to
parallelize the loop. We could add the following line to any of our previous examples.
!pgi$l safe_lastval
! Fortran Version
#pragma loop safe_lastval
/* C/C++ Version */
The resulting code looks similar to this:
/* C/C++ Version */
#pragma loop safe_lastval
...
for (i=1;i<N;i++){
if(x[i]>5.0 ) t=x[i];
}
v = t;
! Fortran Version
!pgi$l safe_lastv
...
do I = 1,N
if (x(I) > 5.0 ) then
t = x(I)
endif
enddo
v = t
In addition, a command-line option -Msafe_lastval provides this information for all
loops within the routines being compiled, which essentially provides global scope.
3.8. Processor-Specific Optimization and the
Unified Binary
Every brand of processor has differences, some subtle, in hardware features such as
instruction sets and cache size. The compilers make architecture-specific decisions
about things such as instruction selection, instruction scheduling, and vectorization. By
default, the PGI compilers produce code specifically targeted to the type of processor
on which the compilation is performed. That is, the default is to use all supported
instructions wherever possible when compiling on a given system. As a result,
executables created on a given system may not be usable on previous generation
systems. For example, executables created on a Pentium 4 may fail to execute on a
Pentium III or Pentium II.
All PGI compilers have the capability of generating unified binaries, which provide a lowoverhead means for generating a single executable that is compatible with and has good
performance on more than one hardware platform.
You can use the -tp option to control compilation behavior by specifying the processor
or processors with which the generated code is compatible. The compilers generate and
combine into one executable multiple binary code streams, each optimized for a specific
platform. At runtime, the executable senses the environment and dynamically selects
the appropriate code stream. For specific information on the -tp option, refer to the PGI
Compiler Reference Manual.
Executable size is automatically controlled via unified binary culling. Only those
functions and subroutines where the target affects the generated code have unique
User's Guide for x86-64 CPUs
Version 2017 | 42
Optimizing and Parallelizing
binary images, resulting in a code-size savings of 10% to 90% compared to generating
full copies of code for each target.
Programs can use the PGI Unified Binary even if all of the object files and libraries are
not compiled as unified binaries. Like any other object file, you can use PGI Unified
Binary object files to create programs or libraries. No special start up code is needed;
support is linked in from the PGI libraries.
The -Mipa option disables generation of PGI Unified Binary. Instead, the default target
auto-detect rules for the host are used to select the target processor.
3.9. Interprocedural Analysis and Optimization
using -Mipa
The PGI Fortran, C and C++ compilers use interprocedural analysis (IPA) that results in
minimal changes to makefiles and the standard edit-build-run application development
cycle. Other than adding -Mipa to the command line, no other changes are required. For
reference and background, the process of building a program without IPA is described
later in this section, followed by the minor modifications required to use IPA with the
PGI compilers. While the PGCC compiler is used here to show how IPA works, similar
capabilities apply to each of the PGI Fortran, C and C++ compilers.
The examples use Linux file naming conventions. As of the PGI 16.3 release, -Mipa
has been disabled on Windows.
3.9.1. Building a Program Without IPA – Single Step
Using the pgcc command-level compiler driver, multiple source files can be compiled
and linked into a single executable with one command. The following example compiles
and links three source files:
% pgcc -o a.out file1.c file2.c file3.c
In actuality, the pgcc driver executes several steps to produce the assembly code and
object files corresponding to each source file, and subsequently to link the object
files together into a single executable file. This command is roughly equivalent to the
following commands performed individually:
%
%
%
%
%
%
%
pgcc -S -o file1.s file1.c
as -o file1.o file1.s
pgcc -S -o file2.s file2.c
as -o file2.o file2.s
pgcc -S -o file3.s file3.c
as -o file3.o file3.s
pgcc -o a.out file1.o file2.o file3.o
User's Guide for x86-64 CPUs
Version 2017 | 43
Optimizing and Parallelizing
If any of the three source files is edited, the executable can be rebuilt with the same
command line:
% pgcc -o a.out file1.c file2.c file3.c
This always works as intended, but has the side-effect of recompiling all of the
source files, even if only one has changed. For applications with a large number of
source files, this can be time-consuming and inefficient.
3.9.2. Building a Program Without IPA – Several Steps
It is also possible to use individual pgcc commands to compile each source file into a
corresponding object file, and one to link the resulting object files into an executable:
%
%
%
%
pgcc
pgcc
pgcc
pgcc
-c
-c
-c
-o
file1.c
file2.c
file3.c
a.out file1.o file2.o file3.o
The pgcc driver invokes the compiler and assembler as required to process each source
file, and invokes the linker for the final link command. If you modify one of the source
files, the executable can be rebuilt by compiling just that file and then relinking:
% pgcc -c file1.c
% pgcc -o a.out file1.o file2.o file3.o
3.9.3. Building a Program Without IPA Using Make
The program compilation and linking process can be simplified greatly using the make
utility on systems where it is supported. Suppose you create a makefile containing the
following lines:
a.out: file1.o file2.o file3.o
pgcc $(OPT) -o a.out file1.o file2.o file3.o
file1.o: file1.c
pgcc $(OPT) -c file1.c
file2.o: file2.c
pgcc $(OPT) -c file2.c
file3.o: file3.c
pgcc $(OPT) -c file3.c
It is then possible to type a single make command:
% make
The make utility determines which object files are out of date with respect to their
corresponding source files, and invokes the compiler to recompile only those source
files and to relink the executable. If you subsequently edit one or more source files, the
executable can be rebuilt with the minimum number of recompilations using the same
single make command.
3.9.4. Building a Program with IPA
Interprocedural analysis and optimization (IPA) by the PGI compilers alters the
standard and make utility command-level interfaces as little as possible. IPA occurs in
three phases:
User's Guide for x86-64 CPUs
Version 2017 | 44
Optimizing and Parallelizing
‣
‣
‣
Collection: Create a summary of each function or procedure, collecting the useful
information for interprocedural optimizations. This is done during the compile
step if the -Mipa switch is present on the command line; summary information is
collected and stored in the object file.
Propagation: Process all the object files to propagate the interprocedural summary
information across function and file boundaries. This is done during the link step,
when all the object files are combined, if the -Mipa switch is present on the link
command line.
Recompile/Optimization: Recompile each of the object files with the propagated
interprocedural information, producing a specialized object file. This process is
also performed during the link step when the -Mipa switch is present on the link
command line.
When linking with -Mipa, the PGI compilers automatically regenerate IPA-optimized
versions of each object file, essentially recompiling each file. If there are IPA-optimized
objects from a previous build, the compilers will minimize the recompile time by reusing
those objects if they are still valid. They will still be valid if the IPA-optimized object is
newer than the original object file, and the propagated IPA information for that file has
not changed since it was optimized.
After each object file has been recompiled, the regular linker is invoked to build the
application with the IPA-optimized object files. The IPA-optimized object files are saved
in the same directory as the original object files, for use in subsequent program builds.
3.9.5. Building a Program with IPA – Single Step
By adding the -Mipa command line switch, several source files can be compiled and
linked with interprocedural optimizations with one command:
% pgcc -Mipa=fast -o a.out file1.c file2.c file3.c
Just like compiling without -Mipa, the driver executes several steps to produce the
assembly and object files to create the executable:
%
%
%
%
%
%
%
pgcc -Mipa=fast -S -o file1.s file1.c
as -o file1.o file1.s
pgcc -Mipa=fast -S -o file2.s file2.c
as -o file2.o file2.s
pgcc -Mipa=fast -S -o file3.s file3.c
as -o file3.o file3.s
pgcc -Mipa=fast -o a.out file1.o file2.o file3.o
In the last step, an IPA linker is invoked to read all the IPA summary information and
perform the interprocedural propagation. The IPA linker reinvokes the compiler on each
of the object files to recompile them with interprocedural information. This creates three
new objects with mangled names:
file1_ipa5_a.out.oo.o, file2_ipa5_a.out.oo.o, file3_ipa5_a.out.oo.o
The system linker is then invoked to link these IPA-optimized objects into the final
executable. Later, if one of the three source files is edited, the executable can be rebuilt
with the same command line:
% pgcc -Mipa=fast -o a.out file1.c file2.c file3.c
This works, but again has the side-effect of compiling each source file, and recompiling
each object file at link time.
User's Guide for x86-64 CPUs
Version 2017 | 45
Optimizing and Parallelizing
3.9.6. Building a Program with IPA – Several Steps
Just by adding the -Mipa command-line switch, it is possible to use individual pgcc
commands to compile each source file, followed by a command to link the resulting
object files into an executable:
%
%
%
%
pgcc
pgcc
pgcc
pgcc
-Mipa=fast
-Mipa=fast
-Mipa=fast
-Mipa=fast
-c
-c
-c
-o
file1.c
file2.c
file3.c
a.out file1.o file2.o file3.o
The pgcc driver invokes the compiler and assembler as required to process each source
file, and invokes the IPA linker for the final link command. If you modify one of the
source files, the executable can be rebuilt by compiling just that file and then relinking:
% pgcc -Mipa=fast -c file1.c
% pgcc -Mipa=fast -o a.out file1.o file2.o file3.o
When the IPA linker is invoked, it will determine that the IPA-optimized object
for file1.o (file1_ipa5_a.out.oo.o) is stale, since it is older than the object
file1.o; and hence it needs to be rebuilt, and reinvokes the compiler to generate it.
In addition, depending on the nature of the changes to the source file file1.c, the
interprocedural optimizations previously performed for file2 and file3 may now
be inaccurate. For instance, IPA may have propagated a constant argument value in a
call from a function in file1.c to a function in file2.c; if the value of the argument
has changed, any optimizations based on that constant value are invalid. The IPA linker
determines which, if any, of the previously created IPA-optimized objects need to be
regenerated; and, as appropriate, reinvokes the compiler to regenerate them. Only those
objects that are stale or which have new or different IPA information are regenerated.
This approach saves compile time.
3.9.7. Building a Program with IPA Using Make
As shown earlier, programs can be built with IPA using the make utility. Just add the
command-line switch -Mipa, as shown here:
OPT=-Mipa=fast
a.out: file1.o file2.o file3.o
pgcc $(OPT) -o a.out file1.o file2.o file3.o
file1.o: file1.c
pgcc $(OPT) -c file1.c
file2.o: file2.c
pgcc $(OPT) -c file2.c
file3.o: file3.c
pgcc $(OPT) -c file3.c
Using the single make command invokes the compiler to generate any of the object files
that are out-of-date, then invokes pgcc to link the objects into the executable. At link
time, pgcc calls the IPA linker to regenerate any stale or invalid IPA-optimized objects.
% make
3.9.8. Questions about IPA
Question: Why is the object file so large?
User's Guide for x86-64 CPUs
Version 2017 | 46
Optimizing and Parallelizing
Answer: An object file created with -Mipa contains several additional sections. One is
the summary information used to drive the interprocedural analysis. In addition, the
object file contains the compiler internal representation of the source file, so the file can
be recompiled at link time with interprocedural optimizations. There may be additional
information when inlining is enabled. The total size of the object file may be 5-10 times
its original size. The extra sections are not added to the final executable.
Question: What if I compile with -Mipa and link without -Mipa?
Answer: The PGI compilers generate a legal object file, even when the source file
is compiled with -Mipa. If you compile with -Mipa and link without -Mipa, the
linker is invoked on the original object files. A legal executable is generated. While
this executable does not have the benefit of interprocedural optimizations, any other
optimizations do apply.
Question: What if I compile without -Mipa and link with -Mipa?
Answer: At link time, the IPA linker must have summary information about all the
functions or routines used in the program. This information is created only when
a file is compiled with -Mipa. If you compile a file without -Mipa and then try to
get interprocedural optimizations by linking with -Mipa, the IPA linker will issue a
message that some routines have no IPA summary information, and will proceed to run
the system linker using the original object files. If some files were compiled with -Mipa
and others were not, it will determine the safest approximation of the IPA summary
information for those files not compiled with -Mipa, and use that to recompile the other
files using interprocedural optimizations.
Question: Can I build multiple applications in the same directory with -Mipa?
Answer: Yes. Suppose you have three source files: main1.c, main2.c, and sub.c,
where sub.c is shared between the two applications. Suppose you build the first
application with -Mipa, using this command:
% pgcc -Mipa=fast -o app1 main1.c sub.c
The IPA linker creates two IPA-optimized object files and uses them to build the first
application.
main1_ipa4_app1.oo sub_ipa4_app1.oo
Now suppose you build the second application using this command:
% pgcc -Mipa=fast -o app2 main2.c sub.c
The IPA linker creates two more IPA-optimized object files:
main2_ipa4_app2.oo sub_ipa4_app2.oo
There are now three object files for sub.c: the original sub.o, and two IPAoptimized objects, one for each application in which it appears.
Question: How is the mangled name for the IPA-optimized object files generated?
Answer: The mangled name has ‘_ipa’ appended, followed by the decimal number of
the length of the executable file name, followed by an underscore and the executable
file name itself. The suffix is changed to .oo so that linking *.o does not pull in the IPAoptimized objects. If the IPA linker determines that the file would not benefit from any
User's Guide for x86-64 CPUs
Version 2017 | 47
Optimizing and Parallelizing
interprocedural optimizations, it does not have to recompile the file at link time, and
uses the original object.
Question: Can I use parallel make environments (e.g., pmake) with IPA?
Answer: No. IPA is not compatible with parallel make environments.
3.10. Profile-Feedback Optimization using -Mpfi/Mpfo
The PGI compilers support many common profile-feedback optimizations, including
semi-invariant value optimizations and block placement. These are performed under
control of the -Mpfi/-Mpfo command-line options.
When invoked with the -Mpfi option, the PGI compilers instrument the generated
executable for collection of profile and data feedback information. This information can
be used in subsequent compilations that include the -Mpfo optimization option. -Mpfi
must be used at both compile-time and link-time. Programs compiled with -Mpfi
include extra code to collect runtime statistics and write them out to a trace file. When
the resulting program is executed, a profile feedback trace file pgfi.out is generated in
the current working directory.
Programs compiled and linked with -Mpfi execute more slowly due to the
instrumentation and data collection overhead. You should use executables compiled
with -Mpfi only for execution of training runs.
When invoked with the -Mpfo option, the PGI compilers use data from a pgfi.out
profile feedback tracefile to enable or enhance certain performance optimizations. Use
of this option requires the presence of a pgfi.out trace file in the current working
directory.
3.11. Default Optimization Levels
The following table shows the interaction between the -O<level>, -g, and -M<opt>
options. In the table, level can be 0, 1, 2, 3 or 4, and <opt> can be vect, concur, unroll
or ipa. The default optimization level is dependent upon these command-line options.
Table 10 Optimization and -O, -g and -M<opt> Options
Optimize Option
Debug Option
-M<opt> Option
Optimization Level
none
none
none
1
none
none
-M<opt>
2
none
-g
none
0
-O
none or -g
none
2
-O<level>
none or -g
none
level
User's Guide for x86-64 CPUs
Version 2017 | 48
Optimizing and Parallelizing
Optimize Option
Debug Option
-M<opt> Option
Optimization Level
-O<level> <= 2
none or -g
-M<opt>
2
Code that is not optimized yet compiled using the option -O0 can be significantly slower
than code generated at other optimization levels. The -M<opt> option, where <opt>
is vect, concur, unroll or ipa, sets the optimization level to 2 if no -O options
are supplied. The -fast option sets the optimization level to a target-dependent
optimization level if no -O options are supplied.
3.12. Local Optimization Using Directives and
Pragmas
Command-line options let you specify optimizations for an entire source file. Directives
supplied within a Fortran source file and pragmas supplied within a C or C++ source
file provide information to the compiler and alter the effects of certain command-line
options or the default behavior of the compiler. (Many directives have a corresponding
command-line option.)
While a command line option affects the entire source file that is being compiled,
directives and pragmas let you do the following:
‣
‣
‣
Apply, or disable, the effects of a particular command-line option to selected
subprograms or to selected loops in the source file (for example, an optimization).
Globally override command-line options.
Tune selected routines or loops based on your knowledge or on information
obtained through profiling.
Using Directives and Pragmas provides details on how to add directives and pragmas to
your source files.
3.13. Execution Timing and Instruction Counting
As this chapter describes, once you have a program that compiles, executes and gives
correct results, you may optimize your code for execution efficiency.
Selecting the correct optimization level requires some thought and may require that you
compare several optimization levels before arriving at the best solution. To compare
optimization levels, you need to measure the execution time for your program. There are
several approaches you can take for timing execution.
‣
‣
‣
You can use shell commands that provide execution time statistics.
You can include function calls in your code that provide timing information.
You can profile sections of code.
Timing functions available with the PGI compilers include these:
‣
‣
3F timing routines.
The SECNDS pre-declared function in PGF77, PGF95, or PGFORTRAN.
User's Guide for x86-64 CPUs
Version 2017 | 49
Optimizing and Parallelizing
‣
The SYSTEM_CLOCK or CPU_CLOCK intrinsics in PGF95 or PGFORTRAN.
In general, when timing a program, you should try to eliminate or reduce the amount of
system level activities such as I/O, program loading, and task switching.
The following example shows a fragment that indicates how to use SYSTEM_CLOCK
effectively within a Fortran program unit.
Using SYSTEM_CLOCK code fragment
integer :: nprocs, hz, clock0, clock1
real :: time
call system_clock (count_rate=hz)
call system_clock(count=clock0)
< do work>
call system_clock(count=clock1)
t = (clock1 - clock0)
time = real (t) / real(hz)
Or you can use the F90 cpu_time subroutine:
real :: t1, t2, time
call cpu_time(t1)
< do work>
call cpu_time(t2)
time = t2 - t1
3.14. Portability of Multi-Threaded Programs on
Linux
PGI created the library libnuma to handle the variations between various
implementations of Linux.
Some older versions of Linux are lacking certain features that support multi-processor
and multicore systems; in particular, the system call 'sched_setaffinity' and the numa
library libnuma. The PGI runtime library uses these features to implement some
-Mconcur and -mp operations.
These variations led to the creation of the PGI library: libnuma, which is used on all 64bit Linux systems, but is not needed on Windows or macOS.
When a program is linked with the system libnuma library, the program depends on
that library to run. On systems without a libnuma library, the PGI version of libnuma
provides the required stubs so that the program links and executes properly. If the
program is linked with libnuma, the differences between systems is masked by the
different versions of libnuma.
When a program is deployed to the target system, the proper set of libraries, real or stub,
should be deployed with the program.
This facility requires that the program be dynamically linked with libnuma.
User's Guide for x86-64 CPUs
Version 2017 | 50
Optimizing and Parallelizing
3.14.1. libnuma
Not all systems have libnuma. Typically, only numa systems have this library. PGI
supplies a stub version of libnuma which satisfies the calls from the PGI runtime to
libnuma. libnuma is a shared library that is linked dynamically at runtime.
The reason to have a numa library on all systems is to allow multi-threaded programs,
such as programs compiled with -Mconcur or -mp, to be compiled, linked, and
executed without regard to whether the host or target systems has a numa library. When
the numa library is not available, a multi-threaded program still runs because the calls to
the numa library are satisfied by the PGI stub library.
During installation, the installation procedure checks for the existence of a real libnuma
among the system libraries. If the real library is not found, the PGI stub version is
substituted.
User's Guide for x86-64 CPUs
Version 2017 | 51
Chapter 4.
USING FUNCTION INLINING
Function inlining replaces a call to a function or a subroutine with the body of the
function or subroutine. This can speed up execution by eliminating parameter passing
and function/subroutine call and return overhead. It also allows the compiler to
optimize the function with the rest of the code. Note that using function inlining
indiscriminately can result in much larger code size and no increase in execution speed.
The PGI compilers provide two categories of inlining:
‣
‣
Automatic function inlining – In C/C++, you can inline static functions with the
inline keyword by using the -Mautoinline option, which is included with
-fast.
Function inlining – You can inline functions which were extracted to the inline
libraries in C/Fortran/C++. There are two ways of enabling function inlining:
with and without the lib suboption. For the latter, you create inline libraries, for
example using the pgfortran compiler driver and the -o and -Mextract options.
There are important restrictions on inlining. Inlining only applies to certain types
of functions. Refer to Restrictions on Inlining for more details on function inlining
limitations.
This section describes how to use the following options related to function inlining:
-Mautoinline
-Mextract
-Minline
-Mrecursive
4.1. Automatic function inlining in C/C++
To enable automatic function inlining in C/C++ for static functions with the inline
keyword, use the -Mautoinline option (included in -fast). Use -Mnoautoinline
to disable it.
Several -Mautoinline suboptions let you determine the selection criteria. These
suboptions are:
User's Guide for x86-64 CPUs
Version 2017 | 52
Using Function Inlining
maxsize:n
Automatically inline functions size n and less
totalsize:n
Limit automatic inlining to total size of n
4.2. Invoking Function Inlining
To invoke the function inliner, use the -Minline option. If you do not specify an
inline library, the compiler performs a special prepass on all source files named on the
compiler command line before it compiles any of them. This pass extracts functions that
meet the requirements for inlining and puts them in a temporary inline library for use
by the compilation pass.
Several -Minline suboptions let you determine the selection criteria for functions to be
inlined. These suboptions include:
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.
[size:]n
Inlines functions with a statement count less than or equal to n, the specified size.
The size n may not exactly equal the number of statements in a selected
function; the size parameter is merely a rough gauge.
levels:n
Inlines n level of function calling levels. The default number is one (1). Using a
level greater than one indicates that function calls within inlined functions may be
replaced with inlined code. This approach allows the function inliner to automatically
perform a sequence of inline and extract processes.
[lib:]file.ext
Instructs the inliner to inline the functions within the library file file.ext. If no
inline library is specified, functions are extracted from a temporary library created
during an extract prepass.
Tip Create the library file using the -Mextract option.
If you specify both a function name and a size n, the compiler inlines functions that
match the function name or have n or fewer statements.
If a name is used without a keyword, then a name with a period is assumed to be
an inline library and a name without a period is assumed to be a function name. If a
number is used without a keyword, the number is assumed to be a size.
User's Guide for x86-64 CPUs
Version 2017 | 53
Using Function Inlining
In the following example, the compiler inlines functions with fewer than approximately
100 statements in the source file myprog.f and writes the executable code in the default
output file a.out.
$ pgfortran -Minline=size:100 myprog.f
Refer to ‘-M Options by Category’ in the PGI Compiler’s Reference Guide. For more
information on the -Minline options, refer to ‘-M Options by Category’ section of the
PGI Compiler’s Reference Guide.
4.3. Using an Inline Library
If you specify one or more inline libraries on the command line with the -Minline
option, the compiler does not perform an initial extract pass. The compiler selects
functions to inline from the specified inline library. If you also specify a size or function
name, all functions in the inline library meeting the selection criteria are selected for
inline expansion at points in the source text where they are called.
If you do not specify a function name or a size limitation for the -Minline option, the
compiler inlines every function in the inline library that matches a function in the source
text.
In the following example, the compiler inlines the function proc from the inline library
lib.il and writes the executable code in the default output file a.out.
$ pgfortran -Minline=name:proc,lib:lib.il myprog.f
The following command line is equivalent to the preceding line, with the exception that
in the following example does not use the keywords name: and lib:. You typically use
keywords to avoid name conflicts when you use an inline library name that does not
contain a period. Otherwise, without the keywords, a period informs the compiler that
the file on the command line is an inline library.
$ pgfortran -Minline=proc,lib.il myprog.f
4.4. Creating an Inline Library
You can create or update an inline library using the -Mextract command-line option. If
you do not specify selection criteria with the -Mextract option, the compiler attempts
to extract all subprograms.
Several -Mextract options let you determine the selection criteria for creating or
updating an inline library. These selection criteria include:
func
Extracts the function func. you can use a comma-separated list to specify multiple
functions.
[name:]func
Extracts the functions whose name matches func, a function in the source text.
User's Guide for x86-64 CPUs
Version 2017 | 54
Using Function Inlining
[size:]n
Limits the size of the extracted functions to functions with a statement count less than
or equal to n, the specified size.
The size n may not exactly equal the number of statements in a selected function;
the size parameter is merely a rough gauge.
[lib:]ext.lib
Stores the extracted information in the library directory ext.lib.
If no inline library is specified, functions are extracted to a temporary library created
during an extract prepass for use during the compilation stage.
When you use the -Mextract option, only the extract phase is performed; the compile
and link phases are not performed. The output of an extract pass is a library of functions
available for inlining. This output is placed in the inline library file specified on
the command line with the -o filename specification. If the library file exists, new
information is appended to it. If the file does not exist, it is created. You can use a
command similar to the following:
$ pgfortran -Mextract=lib:lib.il myfunc.f
You can use the -Minline option with the -Mextract option. In this case, the
extracted library of functions can have other functions inlined into the library. Using
both options enables you to obtain more than one level of inlining. In this situation, if
you do not specify a library with the -Minline option, the inline process consists of two
extract passes. The first pass is a hidden pass implied by the -Minline option, during
which the compiler extracts functions and places them into a temporary library. The
second pass uses the results of the first pass but puts its results into the library that you
specify with the -o option.
4.4.1. Working with Inline Libraries
An inline library is implemented as a directory with each inline function in the library
stored as a file using an encoded form of the inlinable function.
A special file named TOC in the inline library directory serves as a table of contents
for the inline library. This is a printable, ASCII file which you can examine to locate
information about the library contents, such as names and sizes of functions, the source
file from which they were extracted, the version number of the extractor which created
the entry, and so on.
Libraries and their elements can be manipulated using ordinary system commands.
‣
‣
‣
Inline libraries can be copied or renamed.
Elements of libraries can be deleted or copied from one library to another.
The ls or dir command can be used to determine the last-change date of a library
entry.
User's Guide for x86-64 CPUs
Version 2017 | 55
Using Function Inlining
4.4.2. Dependencies
When a library is created or updated using one of the PGI compilers, the last-change
date of the library directory is updated. This allows a library to be listed as a dependence
in a makefile and ensures that the necessary compilations are performed when a library
is changed.
4.4.3. Updating Inline Libraries – Makefiles
If you use inline libraries you must be certain that they remain up-to-date with the
source files into which they are inlined. One way to assure inline libraries are updated is
to include them in a makefile.
The makefile fragment in the following example assumes the file utils.f contains a
number of small functions used in the files parser.f and alloc.f.
This portion of the makefile:
‣
‣
‣
Maintains the inline library utils.il.
Updates the library whenever you change utils.f or one of the include files it
uses.
Compiles parser.f and alloc.f whenever you update the library.
Sample Makefile
SRC = mydir
FC = pgfortran
FFLAGS = -O2
main.o: $(SRC)/main.f $(SRC)/global.h
$(FC) $(FFLAGS) -c $(SRC)/main.f
utils.o: $(SRC)/utils.f $(SRC)/global.h $(SRC)/utils.h
$(FC) $(FFLAGS) -c $(SRC)/utils.f
utils.il: $(SRC)/utils.f $(SRC)/global.h $(SRC)/utils.h
$(FC) $(FFLAGS) -Mextract=15 -o utils.il $(SRC)/utils.f
parser.o: $(SRC)/parser.f $(SRC)/global.h utils.il
$(FC) $(FFLAGS) -Minline=utils.il -c $(SRC)/parser.f
alloc.o: $(SRC)/alloc.f $(SRC)/global.h utils.il
$(FC) $(FFLAGS) -Minline=utils.il -c $(SRC)/alloc.f
myprog: main.o utils.o parser.o alloc.o
$(FC) -o myprog main.o utils.o parser.o alloc.o
4.5. Error Detection during Inlining
You can specify the -Minfo=inline option to request inlining information from the
compiler when you invoke the inliner. For example:
$ pgfortran -Minline=mylib.il -Minfo=inline myext.f
User's Guide for x86-64 CPUs
Version 2017 | 56
Using Function Inlining
4.6. Examples
Assume the program dhry consists of a single source file dhry.f. The following
command line builds an executable file for dhry in which proc7 is inlined wherever it is
called:
$ pgfortran dhry.f -Minline=proc7
The following command lines build an executable file for dhry in which proc7 plus any
functions of approximately 10 or fewer statements are inlined (one level only).
The specified functions are inlined only if they are previously placed in the inline
library, temp.il, during the extract phase.
$ pgfortran dhry.f -Mextract=lib:temp.il
$ pgfortran dhry.f -Minline=10,proc7,temp.il
Using the same source file dhry.f, the following example builds an executable for
dhry in which all functions of roughly ten or fewer statements are inlined. Two levels of
inlining are performed. This means that if function A calls function B, and B calls C, and
both B and C are inlinable, then the version of B which is inlined into A will have had C
inlined into it.
$ pgfortran dhry.f -Minline=size:10,levels:2
4.7. Restrictions on Inlining
The following Fortran subprograms cannot be extracted:
‣
‣
‣
‣
Main or BLOCK DATA programs.
Subprograms containing alternate return, assigned GO TO, DATA, SAVE, or
EQUIVALENCE statements.
Subprograms containing FORMAT statements.
Subprograms containing multiple entries.
A Fortran subprogram is not inlined if any of the following applies:
‣
‣
‣
‣
It is referenced in a statement function.
A common block mismatch exists; in other words, the caller must contain all
common blocks specified in the callee, and elements of the common blocks must
agree in name, order, and type (except that the caller's common block can have
additional members appended to the end of the common block).
An argument mismatch exists; in other words, the number and type (size) of actual
and formal parameters must be equal.
A name clash exists, such as a call to subroutine xyz in the extracted subprogram
and a variable named xyz in the caller.
The following types of C and C++ functions cannot be inlined:
‣
Functions containing switch statements
User's Guide for x86-64 CPUs
Version 2017 | 57
Using Function Inlining
‣
‣
Functions which reference a static variable whose definition is nested within the
function
Functions which accept a variable number of arguments
Certain C/C++ functions can only be inlined into the file that contains their definition:
‣
‣
‣
Static functions
Functions which call a static function
Functions which reference a static variable
User's Guide for x86-64 CPUs
Version 2017 | 58
Chapter 5.
USING OPENMP
The PGF77 and PGFORTRAN Fortran compilers support the OpenMP Fortran
Application Program Interface. The PGCC and PGC++ compilers support the OpenMP
C/C++ Application Program Interface.
OpenMP is a specification for a set of compiler directives, an applications programming
interface (API), and a set of environment variables that can be used to specify shared
memory parallelism in FORTRAN and C/C++ programs. OpenMP may be used to obtain
most of the parallel performance you can expect from your code, or it may serve as a
stepping stone to parallelizing an entire application with MPI.
This section provides information on OpenMP as it is supported by PGI compilers.
Currently, all PGI compilers support the version 3.1 OpenMP specification.
Use the -mp compiler switch to enable processing of the OpenMP pragmas listed in this
section. As of the PGI 2011 Release, the OpenMP runtime library is linked by default.
Note that GNU pthreads are not completely interoperable with OpenMP threads.
When using pgc++ on Linux or macOS, the GNU STL is thread-safe to the extent listed
in the GNU documentation as required by the C++11 standard. If an STL thread-safe
issue is suspected, the suspect code can be run sequentially inside of an OpenMP
region using #pragma omp critical sections.
This section describes how to use the following option supporting OpenMP: -mp
5.1. OpenMP Overview
Let’s look at the OpenMP shared-memory parallel programming model and some
common OpenMP terminology.
User's Guide for x86-64 CPUs
Version 2017 | 59
Using OpenMP
5.1.1. OpenMP Shared-Memory Parallel Programming
Model
The OpenMP shared-memory programming model is a collection of compiler directives
or pragmas, library routines, and environment variables that can be used to specify
shared-memory parallelism in Fortran and in C/C++ programs.
Fortran directives and C/C++ pragmas
Allow users to mark sections of code that can be executed in parallel when the code is
compiled using the -mp switch.
When this switch is not present, the compiler ignores these directives and pragmas.
Fixed-form Fortran OpenMP directives begin with !$OMP, C$OMP, or *$OMP,
beginning in column 1. Free-form Fortran OpenMP pragmas begin with !$OMP.
OpenMP pragmas for C/C++ begin with #pragma omp. This format allows the user
to have a single source code file for use with or without the -mp switch, as these lines
are then merely viewed as comments when -mp is not present.
These directives and pragmas allow the user to create task, loop, and parallel section
work-sharing constructs and synchronization constructs. They also allow the user to
define how data is shared or copied between parallel threads of execution.
The data environment is controlled either by using clauses on the directives or
pragmas, or with additional directives or pragmas.
Runtime library routines
Are available to query the parallel runtime environment, for example to determine
how many threads are participating in execution of a parallel region.
Environment variables
Are available to control the execution behavior of parallel programs. For more
information on OpenMP, see www.openmp.org.
Macro substitution
C and C++ pragmas are subject to macro replacement after #pragma omp.
5.1.2. Terminology
For OpenMP 3.1 there are a number of terms for which it is useful to have common
definitions.
Thread
An execution entity with a stack and associated static memory, called threadprivate
memory.
‣
‣
An OpenMP thread is a thread that is managed by the OpenMP runtime system.
A thread-safe routine is a routine that performs the intended function even when
executed concurrently, that is, by more than one thread.
User's Guide for x86-64 CPUs
Version 2017 | 60
Using OpenMP
Region
All code encountered during a specific instance of the execution of a given construct
or of an OpenMP library routine. A region includes any code in called routines as
well as any implicit code introduced by the OpenMP implementation.
Regions are nested if one region is (dynamically) enclosed by another region, that is,
a region is encountered during the execution of another region. PGI supports both
lexically and non-lexically nested parallel regions.
Parallel region
In OpenMP 3.1 there is a distinction between a parallel region and an active parallel
region. A parallel region can be either inactive or active.
‣
‣
An inactive parallel region is executed by a single thread.
An active parallel region is a parallel region that is executed by a team consisting
of more than one thread.
The definition of an active parallel region changed between OpenMP 2.5 and
OpenMP 3.1. In OpenMP 2.5, the definition was a parallel region whose IF
clause evaluates to true. To examine the significance of this change, look at
the following example:
program test
logical omp_in_parallel
!$omp parallel
print *, omp_in_parallel()
!$omp end parallel
stop
end
Suppose we run this program with OMP_NUM_THREADS set to one. In OpenMP
2.5, this program yields T while in OpenMP 3.1, the program yields F. In
OpenMP 3.1, execution is not occurring by more than one thread. Therefore,
change in this definition may mean previous programs require modification.
Task
A specific instance of executable code and its data environment, generated when a
thread encounters a task construct or a parallel construct.
5.1.3. OpenMP Example
Look at the following simple OpenMP example involving loops.
User's Guide for x86-64 CPUs
Version 2017 | 61
Using OpenMP
OpenMP Loop Example
PROGRAM MAIN
INTEGER I, N, OMP_GET_THREAD_NUM
REAL*8 V(1000), GSUM, LSUM
GSUM = 0.0D0
N = 1000
DO I = 1, N
V(I) = DBLE(I)
ENDDO
!$OMP PARALLEL PRIVATE(I,LSUM) SHARED(V,GSUM,N)
LSUM = 0.0D0
!$OMP DO
DO I = 1, N
LSUM = LSUM + V(I)
ENDDO
!$OMP END DO
!$OMP CRITICAL
print *, "Thread ",OMP_GET_THREAD_NUM()," local sum: ",LSUM
GSUM = GSUM + LSUM
!$OMP END CRITICAL
!$OMP END PARALLEL
PRINT *, "Global Sum: ",GSUM
STOP
END
If you execute this example with the environment variable OMP_NUM_THREADS set to
4, then the output looks similar to this:
Thread
Thread
Thread
Thread
Global Sum:
FORTRAN STOP
0 local sum:
1 local sum:
2 local sum:
3 local sum:
500500.0000000000
31375.00000000000
93875.00000000000
156375.0000000000
218875.0000000000
5.2. Task Overview
Every part of an OpenMP program is part of a task. A task, whose execution can be
performed immediately or delayed, has these characteristics:
‣
‣
‣
Code to execute
A data environment – that is, it owns its data
An assigned thread that executes the code and uses the data.
There are two activities associated with tasks: packaging and execution.
‣
‣
Packaging: Each encountering thread packages a new instance of a task – code and
data.
Execution: Some thread in the team executes the task at some later time.
In the following sections, we use this terminology:
Task
The package of code and instructions for allocating data created when a thread
encounters a task construct. A task can be implicit or explicit.
‣
An explicit task is a task generated when a task construct is encountered during
execution.
User's Guide for x86-64 CPUs
Version 2017 | 62
Using OpenMP
‣
An implicit task is a task generated by the implicit parallel region or generated
when a parallel construct is encountered during execution.
Task construct
A task directive or pragma plus a structured block.
Task region
The dynamic sequence of instructions produced by the execution of a task by a
thread.
5.3. Fortran Parallelization Directives
Parallelization directives are comments in a program that are interpreted by the PGI
Fortran compilers when the option -mp is specified on the command line. The form of a
parallelization directive is:
sentinel directive_name
[clauses]
With the exception of the SGI-compatible DOACROSS directive, the sentinel must
comply with these rules:
‣
‣
‣
‣
Be one of these: !$OMP, C$OMP, or *$OMP.
Must start in column 1 (one) for free-form code.
Must appear as a single word without embedded white space.
The sentinel marking a DOACROSS directive is C$.
The directive_name can be any of the directives listed in Directive and Pragma Summary
Table. The valid clauses depend on the directive. Directive and Pragma Clauses provides
a list of clauses, the directives and pragmas to which they apply, and their functionality.
In addition to the sentinel rules, the directive must also comply with these rules:
‣
‣
‣
‣
Standard Fortran syntax restrictions, such as line length, case insensitivity, and so
on, apply to the directive line.
Initial directive lines must have a space or zero in column six.
Continuation directive lines must have a character other than a space or a zero in
column six. Continuation lines for C$DOACROSS directives are specified using the
C$& sentinel.
Directives which are presented in pairs must be used in pairs.
Clauses associated with directives have these characteristics:
‣
‣
‣
The order in which clauses appear in the parallelization directives is not significant.
Commas separate clauses within the directives, but commas are not allowed
between the directive name and the first clause.
Clauses on directives may be repeated as needed, subject to the restrictions listed in
the description of each clause.
User's Guide for x86-64 CPUs
Version 2017 | 63
Using OpenMP
5.4. C/C++ Parallelization Pragmas
Parallelization pragmas are #pragma statements in a C or C++ program that are
interpreted by the PGI C and C++ compilers when the option -mp is specified on the
command line. The form of a parallelization pragma is:
#pragma omp
pragma_name
[clauses]
The format for pragmas include these rules:
‣
‣
‣
‣
‣
‣
The pragmas follow the conventions of the C and C++ rules.
Whitespace can appear before and after the #.
Preprocessing tokens following the #pragma omp are subject to macro replacement.
The order in which clauses appear in the parallelization pragmas is not significant.
Spaces separate clauses within the pragmas.
Clauses on pragmas may be repeated as needed subject to the restrictions listed in
the description of each clause.
For the purposes of the OpenMP pragmas, a C/C++ structured block is defined to be
a statement or compound statement (a sequence of statements beginning with { and
ending with }) that has a single entry and a single exit. No statement or compound
statement is a C/C++ structured block if there is a jump into or out of that statement.
5.5. Directive and Pragma Recognition
The compiler option -mp enables recognition of the parallelization directives and
pragmas.
The use of this option also implies:
-Miomutex
For directives, critical sections are generated around Fortran I/O statements.
For pragmas, calls to I/O library functions are system-dependent and are not necessarily
guaranteed to be thread-safe. I/O library calls within parallel regions should be
protected by critical regions to ensure they function correctly on all systems.
5.6. Directive and Pragma Summary Table
The following table provides a brief summary of the directives and pragmas that PGI
supports.
In the table, the values in uppercase letters are Fortran directives while the names in
lowercase letters are C/C++ pragmas.
User's Guide for x86-64 CPUs
Version 2017 | 64
Using OpenMP
5.6.1. Directive and Pragma Summary Table
Table 11 Directive and Pragma Summary Table
Fortran Directive and C++
Pragma
Description
ATOMIC [TYPE] ... END ATOMIC
and atomic
Semantically equivalent to enclosing a single statement in the
CRITCIAL...END CRITICAL directive or critical pragma.
TYPE may be empty or one of the following: UPDATE, READ, WRITE,
or CAPTURE. The END ATOMIC directive is only allowed when ending
ATOMIC CAPTURE regions.
Only certain statements are allowed.
BARRIER and barrier
Synchronizes all threads at a specific point in a program so that all
threads complete work to that point before any thread continues.
CRITICAL ... END CRITICAL
and critical
Defines a subsection of code within a parallel region, a critical section,
which is executed one thread at a time.
DO...END DO and for
Provides a mechanism for distribution of loop iterations across the
available threads in a parallel region.
C$DOACROSS
Specifies that the compiler should parallelize the loop to which it
applies, even though that loop is not contained within a parallel
region.
FLUSH and flush
When this appears, all processor-visible data items, or, when a list is
present (FLUSH [list]), only those specified in the list, are written to
memory, thus ensuring that all the threads in a team have a consistent
view of certain objects in memory.
MASTER ... END MASTER and
master
Designates code that executes on the master thread and that is
skipped by the other threads.
ORDERED and ordered
Defines a code block that is executed by only one thread at a time,
and in the order of the loop iterations; this makes the ordered code
block sequential, while allowing parallel execution of statements
outside the code block.
PARALLEL DO and parallel for
Enables you to specify which loops the compiler should parallelize.
PARALLEL ... END PARALLEL
and parallel
Supports a fork/join execution model in which a single thread executes
all statements until a parallel region is encountered.
PARALLEL SECTIONS and
parallel sections
Defines a non-iterative work-sharing construct without the need to
define an enclosing parallel region.
PARALLEL WORKSHARE ... END
PARALLEL WORKSHARE
Provides a short form method for including a WORKSHARE directive
inside a PARALLEL construct.
SECTIONS ... END SECTIONS
and sections
Defines a non-iterative work-sharing construct within a parallel region.
SINGLE ... END SINGLE and
single
Designates code that executes on a single thread and that is skipped by
the other threads.
TASK and task
Defines an explicit task.
User's Guide for x86-64 CPUs
Version 2017 | 65
Using OpenMP
Fortran Directive and C++
Pragma
Description
TASKYIELD and taskyield
Specifies a scheduling point for a task where the currently executing
task may be yielded, and a different deferred task may be executed.
TASKWAIT and taskwait
Specifies a wait on the completion of child tasks generated since the
beginning of the current task.
THREADPRIVATE and
threadprivate
When a common block or variable that is initialized appears in this
directive or pragma, each thread’s copy is initialized once prior to its
first use.
WORKSHARE ... END
WORKSHARE
Provides a mechanism to effect parallel execution of non-iterative but
implicitly data parallel constructs.
5.7. Directive and Pragma Clauses
Some directives and pragmas accept clauses that further allow a user to control
the scope attributes of variables for the duration of the directive or pragma. Not all
clauses are allowed on all directives, so the clauses that are valid are included with the
description of the directive and pragma.
The following table provides a brief summary of the clauses associated with OPENMP
directives and pragmas that PGI supports.
For complete information on these clauses, refer to the OpenMP documentation
available on the World Wide Web.
Table 12 Directive and Pragma Summary Table
Applies to this
directive
Applies to this
pragma
CAPTURE
ATOMIC
atomic
Specifies that the atomic action
is reading and updating, or
writing and updating a value,
capturing the intermediate
state.
COLLAPSE (n)
DO...END DO
PARALLEL DO
PARALLEL
WORKSHARE
parallel for
Specifies how many loops
are associated with the loop
construct.
COPYIN (list)
PARALLEL
PARALLEL DO
PARALLEL SECTIONS
PARALLEL
WORKSHARE
parallel
parallel for
Allows threads to access the
master thread's value, for
a threadprivate variable.
You assign the same value to
threadprivate variables for each
thread in the team executing
the parallel region. Then, for
each variable specified, the
value of the variable in the
master thread of the team is
copied to the threadprivate
This clause...
User's Guide for x86-64 CPUs
Has this functionality
Version 2017 | 66
Using OpenMP
This clause...
Applies to this
directive
Applies to this
pragma
Has this functionality
copies at the beginning of the
parallel region.
COPYPRIVATE(list)
SINGLE
single
Specifies that one or more
variables should be shared
among all threads. This clause
provides a mechanism to use a
private variable to broadcast
a value from one member of a
team to the other members.
DEFAULT
PARALLEL
PARALLEL DO
PARALLEL SECTIONS
PARALLEL
WORKSHARE
parallel
parallel for
Specifies the behavior of
unscoped variables in a parallel
region, such as the data-sharing
attributes of variables.
FINAL
TASK
task
Specifies that all subtasks
of this task will be run
immediately.
FIRSTPRIVATE(list)
DO
PARALLEL
PARALLEL DO
PARALLEL SECTIONS
PARALLEL
WORKSHARE
SECTIONS
SINGLE
for
parallel
parallel for
sections
single
Specifies that each thread
should have its own instance
of a variable, and that each
variable in the list should be
initialized with the value of
the original variable, because
it exists before the parallel
construct.
IF()
PARALLEL ... END
PARALLEL
PARALLEL DO ...
END PARALLEL DO
PARALLEL
SECTIONS ...
END PARALLEL
SECTIONS
PARALLEL
WORKSHARE
parallel
parallel for
parallel
sections
Specifies whether a loop should
be executed in parallel or in
serial.
LASTPRIVATE(list)
DO
PARALLEL DO ...
END PARALLEL DO
PARALLEL
SECTIONS ...
END PARALLEL
SECTIONS
SECTIONS
parallel
parallel for
parallel
sections
sections
Specifies that the enclosing
context's version of the variable
is set equal to the private
version of whichever thread
executes the final iteration of a
loop construct or last section of
an OpenMP section.
MERGEABLE
TASK
task
Specifies that this task will
run with the same data
environment, including OpenMP
User's Guide for x86-64 CPUs
Version 2017 | 67
Using OpenMP
This clause...
Applies to this
directive
Applies to this
pragma
Has this functionality
internal control variables, as
when it is encountered.
NOWAIT
DO ... END DO
SECTIONS
SINGLE
WORKSHARE ...
END WORKSHARE
for
sections
single
Eliminates the barrier implicit
at the end of a parallel region.
NUM_THREADS
PARALLEL
PARALLEL DO ...
END PARALLEL DO
PARALLEL
SECTIONS ...
END PARALLEL
SECTIONS
PARALLEL
WORKSHARE
parallel
parallel for
parallel
sections
Sets the number of threads in a
thread team.
ORDERED
DO...END DO
PARALLEL DO...
END PARALLEL DO
parallel for
Specifies that this block within
the parallel DO or FOR region
needs to be execute serially in
the same order indicated by the
enclosing loop.
PRIVATE
DO
PARALLEL
PARALLEL DO ...
END PARALLEL DO
PARALLEL
SECTIONS ...
END PARALLEL
SECTIONS
PARALLEL
WORKSHARE
SECTIONS
SINGLE
for
parallel
parallel for
parallel
sections
sections
single
Specifies that each thread
should have its own instance of
a variable.
READ
ATOMIC
atomic
Specifies that the atomic action
is reading a value.
REDUCTION
DO
PARALLEL
PARALLEL DO ...
END PARALLEL DO
PARALLEL
SECTIONS ...
END PARALLEL
SECTIONS
PARALLEL
WORKSHARE
SECTIONS
for
parallel
parallel for
parallel
sections
sections
Specifies that one or more
variables in list that are
private to each thread are
the subject of a reduction
operation at the end of the
parallel region.
({operator
| intrinsic }
:
list)
User's Guide for x86-64 CPUs
Version 2017 | 68
Using OpenMP
This clause...
Applies to this
directive
Applies to this
pragma
Has this functionality
DO ... END DO
PARALLEL DO...
END PARALLEL DO
for
parallel for
Applies to the looping directive,
allowing the user to specify
the chunking method for
parallelization. Work is assigned
to threads in different manners
depending on the scheduling
type or chunk size used.
SHARED
PARALLEL
PARALLEL DO ...
END PARALLEL DO
PARALLEL
SECTIONS ...
END PARALLEL
SECTIONS
PARALLEL
WORKSHARE
parallel
parallel for
parallel
sections
Specifies that one or more
variables should be shared
among all threads. All threads
within a team access the
same storage area for shared
variables.
UNTIED
TASK
TASKWAIT
task
taskwait
Specifies that any thread in
the team can resume the task
region after a suspension.
SCHEDULE
(type[
,chunk])
UPDATE
ATOMIC
atomic
Specifies that the atomic action
is updating a value.
WRITE
ATOMIC
atomic
Specifies that the atomic action
is writing a value.
5.8. Runtime Library Routines
User-callable functions are available to the programmer to query and alter the parallel
execution environment.
Any C/C++ program unit that invokes these functions should include the statement
#include <omp.h>. The omp.h header file contains definitions for each of the C/
C++ library routines and the required type definitions. For example, to use the
omp_get_num_threads function, use this syntax:
#include <omp.h>
int omp_get_num_threads(void);
Unlimited OpenMP thread counts are available in all PGI configurations. The number
of threads is unlicensed in the OpenMP runtime libraries – up to the hard limit of 256
threads. The OpenPOWER compiler relies on the llvm OpenMP runtime, which has a
31
maximum of 2 threads.
The following table summarizes the runtime library calls.
The Fortran call is shown first followed by the equivalent C/C++ call.
User's Guide for x86-64 CPUs
Version 2017 | 69
Using OpenMP
Table 13 Runtime Library Routines Summary
Runtime Library Routines with Examples
omp_get_num_threads
Returns the number of threads in the team executing the parallel region from which it is called.
When called from a serial region, this function returns 1. A nested parallel region is the same as
a single parallel region. By default, the value returned by this function is equal to the value of
the environment variable OMP_NUM_THREADS or to the value set by the last previous call to
omp_set_num_threads().
Fortran
integer function omp_get_num_threads()
C/C++
int omp_get_num_threads(void);
omp_set_num_threads
Sets the number of threads to use for the next parallel region.
This subroutine or function can only be called from a serial region of code. If it is called from
within a parallel region, or from within a subroutine or function that is called from within a parallel
region, the results are undefined. Further, this subroutine or function has precedence over the
OMP_NUM_THREADS environment variable.
Fortran
subroutine omp_set_num_threads(scalar_integer_exp)
C/C++
void omp_set_num_threads(int num_threads);
omp_get_thread_num
Returns the thread number within the team. The thread number lies between 0 and
omp_get_num_threads()-1. When called from a serial region, this function returns 0. A nested
parallel region is the same as a single parallel region.
Fortran
integer function omp_get_thread_num()
C/C++
int omp_get_thread_num(void);
omp_get_ancestor_thread_num
Returns, for a given nested level of the current thread, the thread number of the ancestor.
Fortran
integer function omp_get_ancestor_thread_num(level)
integer level
C/C++
int omp_get_ancestor_thread_num(int level);
omp_get_active_level
Returns the number of enclosing active parallel regions enclosing the task that contains the call.
Fortran
integer function omp_get_active_level()
C/C++
int omp_get_active_level(void);
omp_get_level
Returns the number of parallel regions enclosing the task that contains the call.
Fortran
integer function omp_get_level()
C/C++
int omp_get_level(void);
User's Guide for x86-64 CPUs
Version 2017 | 70
Using OpenMP
Runtime Library Routines with Examples
omp_get_max_threads
Returns the maximum value that can be returned by calls to omp_get_num_threads().
If omp_set_num_threads() is used to change the number of processors, subsequent calls to
omp_get_max_threads() return the new value. Further, this function returns the maximum value
whether executing from a parallel or serial region of code.
Fortran
integer function omp_get_max_threads()
C/C++
int omp_get_max_threads(void);
omp_get_num_procs
Returns the number of processors that are available to the program
Fortran
integer function omp_get_num_procs()
C/C++
int omp_get_num_procs(void);
omp_get_stack_size
Returns the value of the OpenMP internal control variable that specifies the size that is used to create a
stack for a newly created thread.
This value may not be the size of the stack of the current thread.
Fortran
!omp_get_stack_size interface
function omp_get_stack_size ()
use omp_lib_kinds
integer ( kind=OMP_STACK_SIZE_KIND )
:: omp_get_stack_size
end function omp_get_stack_size
end interface
C/C++
size_t omp_get_stack_size(void);
omp_set_stack_size
Changes the value of the OpenMP internal control variable that specifies the size to be used to create a
stack for a newly created thread.
The integer argument specifies the stack size in kilobytes. The size of the stack of the current thread
cannot be changed. In the PGI implementation, all OpenMP or auto-parallelization threads are created
just prior to the first parallel region; therefore, only calls to omp_set_stack_size() that occur
prior to the first region have an effect.
Fortran
subroutine omp_set_stack_size(integer(KIND=OMP_STACK_SIZE_KIND))
C/C++
void omp_set_stack_size(size_t stack_size);
omp_get_team_size
Returns, for a given nested level of the current thread, the size of the thread team to which the
ancestor belongs.
Fortran
integer function omp_get_team_size (level)
integer level
C/C++
int omp_get_team_size(int level);
omp_in_final
Returns whether or not the call is within a final task.
User's Guide for x86-64 CPUs
Version 2017 | 71
Using OpenMP
Runtime Library Routines with Examples
Returns
.TRUE. for directives and non-zero for pragmas if called from within a final task region.
Fortran
integer function omp_in_final()
C/C++
int omp_in_final(void);
omp_in_parallel
Returns whether or not the call is within a parallel region.
.TRUE. for directives and non-zero for pragmas if called from within a parallel region and
.FALSE. for directives and zero for pragmas if called outside of a parallel region. When called
Returns
from within a parallel region that is serialized, for example in the presence of an IF clause evaluating
.FALSE. for directives and zero for pragmas, the function returns .FALSE. for directives and zero
for pragmas.
Fortran
logical function omp_in_parallel()
C/C++
int omp_in_parallel(void);
omp_set_dynamic
Allows automatic dynamic adjustment of the number of threads used for execution of parallel regions.
This function is recognized, but currently has no effect.
Fortran
subroutine omp_set_dynamic(scalar_logical_exp)
C/C++
void omp_set_dynamic(int dynamic_threads);
omp_get_dynamic
Allows the user to query whether automatic dynamic adjustment of the number of threads used for
execution of parallel regions is enabled.
This function is recognized, but currently always returns
pragmas.
.FALSE. for directives and zero for
Fortran
logical function omp_get_dynamic()
C/C++
void omp_get_dynamic(void);
omp_set_nested
Allows enabling/disabling of nested parallel regions.
Fortran
subroutine omp_set_nested(nested)
logical nested
C/C++
void omp_set_nested(int nested);
omp_get_nested
Allows the user to query whether dynamic adjustment of the number of threads available for execution
of parallel regions is enabled.
Fortran
logical function omp_get_nested()
C/C++
int omp_get_nested(void);
omp_set_schedule
User's Guide for x86-64 CPUs
Version 2017 | 72
Using OpenMP
Runtime Library Routines with Examples
Set the value of the run_sched_var.
Fortran
subroutine
include
integer
integer
omp_set_schedule(kind, modifier)
'omp_lib_kinds.h'
(kind=omp_sched_kind) kind
modifier
C/C++
void omp_set_schedule(omp_sched_t kind, int chunk_size);
omp_get_schedule
Retrieve the value of the run_sched_var.
Fortran
subroutine
include
integer
integer
omp_get_schedule(kind, modifier)
'omp_lib_kinds.h'
(kind=omp_sched_kind) kind
modifier
C/C++
void omp_get_schedule(omp_sched_t *kind, int *chunk_size);
omp_get_wtime
Returns the elapsed wall clock time, in seconds, as a DOUBLE PRECISION value for directives and as a
floating-point double value for pragmas.
Times returned are per-thread times, and are not necessarily globally consistent across all threads.
Fortran
double precision function omp_get_wtime()
C/C++
double omp_get_wtime(void);
omp_get_wtick
Returns the resolution of omp_get_wtime(), in seconds, as a DOUBLE PRECISION value for Fortran
directives and as a floating-point double value for C/C++ pragmas.
Fortran
double precision function omp_get_wtick()
C/C++
double omp_get_wtick();
omp_init_lock
Initializes a lock associated with the variable lock for use in subsequent calls to lock routines.
The initial state of the lock is unlocked. If the variable is already associated with a lock, it is illegal to
make a call to this routine.
Fortran
subroutine omp_init_lock(lock)
include 'omp_lib_kinds.h'
integer(kind=omp_lock_kind) lock
C/C++
void omp_init_lock(omp_lock_t *lock);
void omp_init_nest_lock(omp_nest_lock_t *lock);
omp_destroy_lock
User's Guide for x86-64 CPUs
Version 2017 | 73
Using OpenMP
Runtime Library Routines with Examples
Disassociates a lock associated with the variable.
Fortran
subroutine omp_destroy_lock(lock)
include 'omp_lib_kinds.h'
integer(kind=omp_lock_kind) lock
C/C++
void omp_destroy_lock(omp_lock_t *lock);
void omp_destroy_nest_lock(omp_nest_lock_t *lock);
omp_set_lock
Causes the calling thread to wait until the specified lock is available.
The thread gains ownership of the lock when it is available. If the variable is not already associated
with a lock, it is illegal to make a call to this routine.
Fortran
subroutine omp_set_lock(lock)
include 'omp_lib_kinds.h'
integer(kind=omp_lock_kind) lock
C/C++
void omp_set_lock(omp_lock_t *lock);
void omp_set_nest_lock(omp_nest_lock_t *lock);
omp_unset_lock
Causes the calling thread to release ownership of the lock associated with integer_var.
If the variable is not already associated with a lock, it is illegal to make a call to this routine.
Fortran
subroutine omp_unset_lock(lock)
include 'omp_lib_kinds.h'
integer(kind=omp_lock_kind) lock
C/C++
void omp_unset_lock(omp_lock_t *lock);
void omp_unset_nest_lock(omp_nest_lock_t *lock);
omp_test_lock
Causes the calling thread to try to gain ownership of the lock associated with the variable.
The function returns .TRUE. for directives and non-zero for pragmas if the thread gains ownership of
the lock; otherwise, it returns .FALSE. for directives and zero for pragmas.
If the variable is not already associated with a lock, it is illegal to make a call to this routine.
Fortran
logical function omp_test_lock(lock)
include 'omp_lib_kinds.h'
integer(kind=omp_lock_kind) lock
C/C++
int omp_test_lock(omp_lock_t *lock);
int omp_test_nest_lock(omp_nest_lock_t *lock);
5.9. Environment Variables
You can use OpenMP environment variables to control the behavior of OpenMP
programs. These environment variables allow you to set and pass information that can
alter the behavior of directives and pragmas.
The following summary table is a quick reference for the OpenMP environment
variables that PGI uses.
User's Guide for x86-64 CPUs
Version 2017 | 74
Using OpenMP
Table 14 OpenMP-related Environment Variable Summary Table
Environment Variable
Default
Description
OMP_DYNAMIC
FALSE
Currently has no effect.
Typically enables (TRUE) or disables (FALSE) the dynamic
adjustment of the number of threads.
OMP_MAX_ACTIVE_LEVELS
1
Specifies the maximum number of nested parallel
regions.
OMP_NESTED
FALSE
Enables (TRUE) or disables (FALSE) nested parallelism.
OMP_NUM_THREADS
1
Specifies the number of threads to use during execution
of parallel regions at the corresponding nested level.
For example, OMP_NUM_THREADS=4,2 uses 4 threads at
the first nested parallel level, and 2 at the next nested
parallel level.
OMP_SCHEDULE
STATIC with
Specifies the type of iteration scheduling and optionally
chunk size of the chunk size to use for omp for and omp parallel for
1
loops that include the runtime schedule clause. The
supported schedule types, which can be specified in
upper- or lower-case are static, dynamic, guided, and
auto.
OMP_PROC_BIND
FALSE
OMP_STACKSIZE
Specifies whether executing threads should be bound to
a core during execution. Allowable values are "true" and
"false".
Overrides the default stack size for a newly created
thread.
OMP_THREAD_LIMIT
64
Specifies the absolute maximum number of threads that
can be used in a program.
OMP_WAIT_POLICY
ACTIVE
Sets the behavior of idle threads, defining whether
they spin or sleep when idle. The values are ACTIVE and
PASSIVE.
User's Guide for x86-64 CPUs
Version 2017 | 75
Chapter 6.
USING MPI
Message Passing Interface (MPI) is an industry-standard application programming
interface designed for rapid data exchange between processors in a cluster application.
MPI is computer software used in computer clusters that allows the processes of a
parallel application to communicate with one another.
PGI provides MPI support with PGI compilers and tools. PGI compilers provide explicit
support to build MPI applications on Windows using Microsoft’s implementation of
MPI, MS-MPI, on macOS using MPICH, and on Linux using MPICH, MVAPICH2, and
Open MPI. Of course, you may always build using an arbitrary version of MPI; to do
this, use the -I, -L, and -l option.
PGI products for Linux include Open MPI, PGI products for macOS includes MPICH,
and PGI products for Windows includes MS-MPI. PGI users with permanent licenses
can download PGI-built versions of MVAPICH2 and MPICH from pgroup.com. This
section describes how to use the MPI capabilities of PGI compilers and how to configure
PGI compilers so these capabilities can be used with custom MPI installations.
The debugger is enabled to support MPI applications running locally with a limited
number of processes. The PGI Debugger Guide describes the MPI-enabled tools in detail.
6.1. MPI Overview
This section contains general information applicable to various MPI distributions. For
distribution-specific information, refer to the sections later in this section.
MPI is a set of function calls and libraries that are used to send messages between
multiple processes. These processes can be located on the same system or on a collection
of distributed servers. Unlike OpenMP, the distributed nature of MPI allows it to work in
almost any parallel environment.
6.2. Compiling and Linking MPI Applications
The PGI compilers provide an option, -Mmpi, to make building MPI applications
with some MPI distributions more convenient by adding the MPI include and library
User's Guide for x86-64 CPUs
Version 2017 | 76
Using MPI
directories to the compiler's include and library search paths. The compiler determines
the location of these directories using various mechanisms described in the MPI
distribution-specific sections later in this section.
Table 15 lists the sub-options supported by -Mmpi.
Table 15 MPI Distribution Options
This MPI
implementation...
Requires compiling and linking with this option...
MPICH1
Deprecated.
MPICH2
Deprecated.
MPICH v3
-Mmpi=mpich
-Mmpi=mpich1
-Mmpi=mpich2
MS-MPI
-Mmpi=msmpi
MVAPICH1
Deprecated.
MVAPICH2
Use MVAPICH2 compiler wrappers.
Open MPI
Use Open MPI compiler wrappers.
SGI MPI
-Mmpi=sgimpi
-Mmpi=mvapich1
For distributions of MPI that are not supported by the -Mmpi compiler option, use the
MPI-distribution-supplied compiler wrappers mpicc, mpic++, mpif77, or mpif90 to
compile and link.
6.3. Debugging MPI Applications
The PGI debugger provides support for symbolic debugging of MPI applications. The
number and location of processes that can be debugged is limited by your license. PGI
node-locked licenses limit processes to a single system whereas PGI network floating
licenses support general development on clusters.
For all distributions of MPI except MPICH1, support for which has been deprecated, and
SGI MPI, you can initiate an MPI debugging session from either the command line or
from within the debugger. For MPICH1 and SGI MPI, debugging must be initiated at
the command line. For specific information on how to initiate a debugging session for a
particular version of MPI, refer to the PGI Debugger Guide.
The debugger can display the contents of message queues for instances of MPI that have
been configured to support that feature. The version of MPICH1 provided with PGI
products for Linux and macOS is configured with this support. At this time, MS-MPI
does not support displaying message queue contents.
For more information on MPI and displaying message queues, refer to the
documentation for your specific distribution of MPI.
User's Guide for x86-64 CPUs
Version 2017 | 77
Using MPI
6.4. Using MPICH on Linux and Apple macOS
PGI products for Linux and macOS include MPI header files, libraries and tools required
to compile, execute and debug MPI programs. PGI compilers can be installed on a single
system, and that system can be treated as if it is a small cluster by using this version of
MPI.
Example
MPI Hello World Example
The following MPI example program uses Open MPI.
$
$
$
$
$
cd my_example_dir
cp -r $PGI/linux86-64/17.5/EXAMPLES/MPI/mpihello .
cd mpihello
export PATH=$PGI/linux86-64/2017/mpi/openmpi/bin:$PATH
mpif90 mpihello.f -o mpihello
$ mpiexec mpihello
Hello world! I'm node 0
$ mpiexec -np 4 mpihello
Hello world! I'm node 0
Hello world! I'm node 2
Hello world! I'm node 1
Hello world! I'm node 3
6.5. Using MPICH1, MPICH2, and MVAPICH1 on
Linux
The -Mmpi sub-options mpich1, mpich2 and mvapich1 have been deprecated in the PGI
2017 release and will be disabled in a future release. To use these deprecated options,
you must first set the environment variable MPIDIR to the root of the MPI installation
directory you intend to use, that is, the directory that contains bin, include, lib and so on.
If you do not set MPIDIR, you will see an error like the following:
$ pgfortran -Mmpi=mpich2 pi.f90 -o fpi
-Mmpi=mpich2 has been deprecated and will be removed in a future release
pgfortran-Error-The environment variable $MPIDIR must be set to use MPICH2
Once MPIDIR is set, you can use -Mmpi with the mpich1, mpich2 and mvapich1 suboptions as you have in previous releases.
6.6. Using MVAPICH2 on Linux
PGI users with permanent licenses can download a PGI-built version of MVAPICH2 for
Linux that includes everything required to compile, execute and debug MPI programs
using MVAPICH2.
To build an application using MVAPICH2, use the MVAPICH2 compiler wrappers:
mpicc, mpic++, mpif77, and mpif90. These wrappers autonmatically set up the compiler
User's Guide for x86-64 CPUs
Version 2017 | 78
Using MPI
commands with the correct include file search paths, library directories, and link
libraries.
To build an application using MVAPICH2 for debugging, add -g to the compiler
wrapper command line arguments.
6.7. Using MS-MPI on Windows
PGI products on Windows include a version of Microsoft’s MPI. You can compile, run,
and debug locally on your system using this instance of MS-MPI.
To compile the application, use the -Mmpi=msmpi option. This option automatically sets
up the appropriate include and library paths to use the MS-MPI headers and libraries.
To compile for debugging, add -g.
6.8. Using Open MPI on Linux
PGI products for Linux ship with a PGI-built version of Open MPI that includes
everything required to compile, execute and debug MPI programs using Open MPI.
To build an application using Open MPI, use the Open MPI compiler wrappers: mpicc,
mpic++, mpif77, and mpif90. These wrappers automatically set up the compiler
commands with the correct include file search paths, library directories, and link
libraries.
To build an application using Open MPI for debugging, add -g to the compiler wrapper
command line arguments.
6.9. Using SGI MPI on Linux
PGI compilers and tools support SGI’s version of MPI. If you want to build your
MPI application using an instance of SGI MPI, you must first set either MPIDIR or
MPI_ROOT to the root of the SGI MPI installation directory that you want to use, that
is, the directory that contains bin, include, lib, and so on. Then you can compile with the
-Mmpi=sgimpi option. Add -g for debugging. To run programs built with SGI MPI,
you must include the SGI MPI lib directory in the LD_LIBRARY_PATH environment
variable.
6.10. Using MPI Compiler Wrappers
When you use MPI compiler wrappers to build with the -fpic or -mcmodel=medium
options, then you must specify -shlib to link with the correct libraries. Here are a few
examples:
For a static link to the MPI libraries, use this command:
% mpicc hello.f
User's Guide for x86-64 CPUs
Version 2017 | 79
Using MPI
For a dynamic link to the MPI libraries, use this command:
% mpicc hello.f -shlib
To compile with -fpic, which, by default, invokes dynamic linking, use this command:
% mpicc -fpic -shlib hello.f
To compile with -mcmodel=medium, use this command:
% mpicc -mcmodel=medium -shlib hello.f
6.11. Limitations
The Open Source Cluster utilities, in particular the MPICH and ScaLAPACK libraries,
are provided with support necessary to build and define their proper use. However, use
of these libraries on linux86-64 systems is subject to the following limitations:
‣
‣
‣
MPI libraries are limited to Messages of length < 2GB, and integer arguments are
INTEGER*4 in FORTRAN, and int in C.
Integer arguments for ScaLAPACK libraries are INTEGER*4 in FORTRAN, and int
in C.
Arrays passed must be < 2GB in size.
6.12. Testing and Benchmarking
The Examples directory contains various benchmarks and tests. Copy this directory
into a local working directory by issuing the following command:
% cp -r $PGI/linux86/17.5/EXAMPLES/MPI .
NAS Parallel Benchmarks
The NPB2.3 subdirectory contains version 2.3 of the NAS Parallel Benchmarks in MPI.
Issue the following commands to run the BT benchmark on four nodes of your cluster:
%
%
%
%
cd MPI/NPB2.3/BT
make BT NPROCS=4 CLASS=W
cd ../bin
mpirun -np 4 bt.W.4
There are several other NAS parallel benchmarks available in this directory. Similar
commands are used to build and run each of them. If you want to run a larger problem,
try building the Class A version of BT by substituting "A" for "W" in the previous
commands.
ScaLAPACK
The ScaLAPACK test times execution of the 3D PBLAS (parallel BLAS) on your cluster.
To run this test, execute the following commands:
% cd scalapack
% make
% mpirun -np 4 pdbla3tim
User's Guide for x86-64 CPUs
Version 2017 | 80
Chapter 7.
USING AN ACCELERATOR
An accelerator is a special-purpose co-processor attached to a CPU and to which the
CPU can offload data and executable kernels to perform compute-intensive calculations.
This section describes a collection of compiler directives used to specify regions of
code in Fortran and C programs that can be offloaded from a host CPU to an attached
accelerator.
7.1. Overview
The programming model and directives described in this section allow programmers
to create high-level host+accelerator programs without the need to explicitly initialize
the accelerator, manage data or program transfers between the host and accelerator,
or initiate accelerator startup and shutdown. Rather, all of these details are implicit in
the programming model and are managed by the PGI Fortran , C, and C++ accelerator
compilers.
The method described provides a model for accelerator programming that is portable
across operating systems and various host CPUs and accelerators. The directives
allow a programmer to migrate applications incrementally to accelerator targets using
standards-compliant Fortran or C.
This programming model allows the programmer to augment information available to
the compilers, including specification of data local to an accelerator region, guidance on
mapping of loops onto an accelerator, and similar performance-related details.
7.1.1. Components
The PGI Accelerator compiler technology includes the following components:
‣
‣
‣
‣
‣
PGFORTRAN™ native OpenMP and OpenACC Fortran 2003 compiler.
PGCC® native OpenMP and OpenACC ANSI C11 and K&R C compiler.
PGC++® native OpenMP and OpenACC ANSI C++14 compiler.
PGI profiler for OpenACC, CUDA, OpenMP, and multi-thread programs.
NVIDIA CUDA Toolkit components
User's Guide for x86-64 CPUs
Version 2017 | 81
Using an Accelerator
‣
A simple command-line tool to detect whether the system has an appropriate GPU
or accelerator card
No accelerator-enabled debugger is included with this release
7.1.2. Availability
The PGI 17.5 Fortran & C Accelerator compilers are available on x86-64 processorbased workstations and servers with an attached NVIDIA CUDA-enabled GPU or Tesla
card. These compilers target all platforms that PGI supports. All examples included in
this section are developed and presented on such a platform. For a list of supported
GPUs, refer to the Accelerator Installation and Supported Platforms list in the latest PGI
Release Notes.
7.1.3. User-directed Accelerator Programming
In user-directed accelerator programming the user specifies the regions of a host
program to be targeted for offloading to an accelerator device. The bulk of a user’s
program, as well as regions containing constructs that are not supported on the targeted
accelerator, are executed on the host. This section concentrates on specification of loops
and regions of code to be offloaded to an accelerator.
7.1.4. Features Not Covered or Implemented
This section does not describe features or limitations of the host programming
environment as a whole. Further, it does not cover automatic detection and offloading
of regions of code to an accelerator by a compiler or other tool. While future versions
of the PGI compilers may allow for automatic offloading, this feature is not currently
supported.
7.2. Terminology
Clear and consistent terminology is important in describing any programming model.
This section provides definitions of the terms required for you to effectively use this
section and the associated programming model.
Accelerator
a special-purpose co-processor attached to a CPU and to which the CPU can offload
data and executable kernels to perform compute-intensive calculations.
Compute intensity
for a given loop, region, or program unit, the ratio of the number of arithmetic
operations performed on computed data divided by the number of memory transfers
required to move that data between two levels of a memory hierarchy.
Compute region
a structure block defined by an OpenACC compute construct. A compute construct is a
structured block containing loops which are compiled for the accelerator. A compute
region may require device memory to be allocated and data to be copied from host
to device upon region entry, and data to be copied from device to host memory and
device memory deallocated upon exit. The dynamic range of a compute construct,
User's Guide for x86-64 CPUs
Version 2017 | 82
Using an Accelerator
including any code in procedures called from within the construct, is the compute
region. In this release, compute regions may not contain other compute regions or
data regions.
Construct
a structured block identified by the programmer or implicitly defined by the
language. Certain actions may occur when program execution reaches the start and
end of a construct, such as device memory allocation or data movement between the
host and device memory. Loops in a compute construct are targeted for execution on
the accelerator. The dynamic range of a construct including any code in procedures
called from within the construct, is called a region.
CUDA
stands for Compute Unified Device Architecture; the CUDA environment from
NVIDIA is a C-like programming environment used to explicitly control and
program an NVIDIA GPU.
Data region
a region defined by an OpenACC data construct, or an implicit data region for
a function or subroutine containing OpenACC directives. Data regions typically
require device memory to be allocated and data to be copied from host to device
memory upon entry, and data to be copied from device to host memory and device
memory deallocated upon exit. Data regions may contain other data regions and
compute regions.
Device
a general reference to any type of accelerator.
Device memory
memory attached to an accelerator which is physically separate from the host
memory.
Directive
in C, a #pragma, or in Fortran, a specially formatted comment statement that is
interpreted by a compiler to augment information about or specify the behavior of the
program.
DMA
Direct Memory Access, a method to move data between physically separate
memories; this is typically performed by a DMA engine, separate from the host CPU,
that can access the host physical memory as well as an IO device or GPU physical
memory.
GPU
a Graphics Processing Unit; one type of accelerator device.
GPGPU
General Purpose computation on Graphics Processing Units.
Host
the main CPU that in this context has an attached accelerator device. The host CPU
controls the program regions and data loaded into and executed on the device.
Loop trip count
the number of times a particular loop executes.
OpenACC
a parallel programming standard describing a set of compiler directives which can be
applied to standard C, C++, and Fortran to specify regions of code for offloading from
a host CPU to an attached accelerator.
User's Guide for x86-64 CPUs
Version 2017 | 83
Using an Accelerator
Private data
with respect to an iterative loop, data which is used only during a particular loop
iteration. With respect to a more general region of code, data which is used within the
region but is not initialized prior to the region and is re-initialized prior to any use
after the region.
Region
the dynamic range of a construct, including any procedures invoked from within the
construct.
Structured block
in C, an executable statement, possibly compound, with a single entry at the top and
a single exit at the bottom. In Fortran, a block of executable statements with a single
entry at the top and a single exit at the bottom.
Vector operation
a single operation or sequence of operations applied uniformly to each element of an
array.
Visible device copy
a copy of a variable, array, or subarray allocated in device memory, that is visible to
the program unit being compiled.
7.3. System Requirements
For NVIDIA GPUs
To use the PGI Accelerator compiler features on NVIDIA GPUs, you must install the
NVIDIA drivers. You may download these components from the NVIDIA website at
www.nvidia.com/cuda
These are not PGI products. They are licensed and supported by NVIDIA.
You must be using an operating system that is supported by both the current PGI
release and by the CUDA software and drivers.
7.4. Supported Processors and GPUs
This PGI Accelerator compiler release supports all AMD64 and Intel 64 host processors.
Use the -tp=<target> flag as documented in the release to specify the target
processor.
Use the -acc flag to enable OpenACC directives and the -ta=<target> flag to target
NVIDIA GPUs. You can then use the generated code on any supported system with
CUDA installed that has a CUDA-enabled GeForce, Quadro, or Tesla card.
For more information on these flags as they relate to accelerator technology, refer to
Applicable Command-line Options.
For a complete list of supported CUDA GPUs, refer to the NVIDIA website
at:www.nvidia.com/object/cuda_learn_products.html
User's Guide for x86-64 CPUs
Version 2017 | 84
Using an Accelerator
You can detect whether the system has CUDA properly installed and has an attached
GPU by running the pgaccelinfo command, which is delivered as part of the PGI
Accelerator compilers software package.
7.5. Installation and Licensing
The PGI Accelerator compilers have a different license key than the -x64 only version
of the PGI products.
7.5.1. Required Files
If you are installing on Windows, the required files are built for you.
The default NVIDIA Compute Capability for generated code in this release is cc2+ or
fermi+, enabling code generation for NVIDIA Fermi and Kepler GPUs. You can use the
-ta flag to specify other compute capabilities.
You can also change the default to one or more of the supported compute capabilities
by adding a line similar to the following one to the sitenvrc file. This example sets
the compute capability to enable code generation for all of the supported compute
capabilities. Notice that the compute capabilities are separated by a space.
set COMPUTECAP=20 30 35 50;
Place the sitenvrc file in the following directory, where $PGI is the PGI installation
directory, which is typically /opt/pgi or /usr/pgi.
$PGI/linux86-64/17.5/bin/
7.5.2. Command-line Flag
After acquiring the PGI Accelerator compilers license key, you can use the -acc or -ta
option with the pgfortran, pgcc or pgc++ commands.
For more information on the -ta flag and the suboptions that relate to the target
accelerators, refer to Applicable Command-line Options.
The compiler automatically invokes the necessary CUDA software tools to create the
kernel code and embeds the kernels in the Linux object file.
To access the accelerator libraries, you must link an accelerator program with the
-ta flag as well.
7.6. Execution Model
The execution model targeted by the PGI Accelerator compilers is host-directed
execution with an attached accelerator device, such as a GPU. The bulk of a user
User's Guide for x86-64 CPUs
Version 2017 | 85
Using an Accelerator
application executes on the host. Compute intensive regions are offloaded to the
accelerator device under control of the host. The accelerator device executes kernels,
which may be as simple as a tightly-nested loop, or as complex as a subroutine,
depending on the accelerator hardware.
7.6.1. Host Functions
Even in accelerator-targeted regions, the host must orchestrate the execution; it
‣
‣
‣
‣
‣
‣
‣
‣
allocates memory on the accelerator device
initiates data transfer
sends the kernel code to the accelerator
passes kernel arguments
queues the kernel
waits for completion
transfers results back to the host
deallocates memory
In most cases, the host can queue a sequence of kernels to be executed on the
device, one after the other.
7.6.2. Levels of Parallelism
Most current GPUs support two levels of parallelism:
‣
‣
an outer doall (fully parallel) loop level
an inner synchronous (SIMD or vector) loop level
Each level can be multidimensional with 2 or 3 dimensions, but the domain must be
strictly rectangular. The synchronous level may not be fully implemented with SIMD or
vector operations, so explicit synchronization is supported and required across this level.
No synchronization is supported between parallel threads across the doall level.
The execution model on the device side exposes these two levels of parallelism and
the programmer is required to understand the difference between, for example, a
fully parallel loop and a loop that is vectorizable but requires synchronization across
iterations. All fully parallel loops can be scheduled for either doall or synchronous parallel
execution, but by definition SIMD vector loops that require synchronization can only be
scheduled for synchronous parallel execution.
7.7. Memory Model
The most significant difference between a host-only program and a host+accelerator
program is that the memory on the accelerator can be completely separate from host
memory, which is the case on most current GPUs. For example:
‣
The host cannot read or write accelerator memory by reference because it is not
mapped into the virtual memory space of the host.
User's Guide for x86-64 CPUs
Version 2017 | 86
Using an Accelerator
‣
‣
All data movement between host memory and accelerator memory must be
performed by the host through runtime library calls that explicitly move data
between the separate memories.
It is not valid to assume the accelerator can read or write host memory, though this
may be supported by accelerators in the future.
7.7.1. Separate Host and Accelerator Memory
Considerations
The programmer must be aware of the potentially separate memories for many reasons,
including but not limited to:
‣
‣
Memory bandwidth between host memory and accelerator memory determines the
compute intensity required to effectively accelerate a given region of code.
Limited size of accelerator memory may prohibit offloading of regions of code that
operate on very large amounts of data.
7.7.2. Accelerator Memory
On the accelerator side, current GPUs implement a weak memory model. In particular,
they do not support memory coherence between threads unless those threads are
parallel only at the synchronous level and the memory operations are separated by
an explicit barrier. Otherwise, if one thread updates a memory location and another
reads the same location, or two threads store a value to the same location, the hardware
does not guarantee the results. While the results of running such a program might be
inconsistent, it is not accurate to say that the results are incorrect. By definition, such
programs are defined as being in error. While a compiler can detect some potential
errors of this nature, it is nonetheless possible to write an accelerator region that
produces inconsistent numerical results.
7.7.3. Cache Management
Some current GPUs have a software-managed cache, some have hardware-managed
caches, and most have hardware caches that can be used only in certain situations
and are limited to read-only data. In low-level programming models such as CUDA,
it is up to the programmer to manage these caches. However, in the PGI Accelerator
programming model, the compiler manages these caches using hints from the
programmer in the form of directives.
7.8. Running an Accelerator Program
Running a program that has accelerator directives and was compiled and linked with
the -ta flag is the same as running the program compiled without the -ta flag.
‣
When running programs on NVIDIA GPUs, the program looks for and dynamically
loads the CUDA libraries. If the libraries are not available, or if they are in a different
directory than they were when the program was compiled, you may need to append
User's Guide for x86-64 CPUs
Version 2017 | 87
Using an Accelerator
‣
‣
‣
the appropriate library directory to your LD_LIBRARY_PATH environment variable
on Linux or to the PATH environment variable on Windows.
On Linux, if you have no server running on your NVIDIA GPU, when your program
reaches its first accelerator region, there may be a 0.5 to 1.5 second pause to warm
up the GPU from a power-off audience. You can avoid this delay by running the
pgcudainit program in the background, which keeps the GPU powered on.
If you compile a program for a particular accelerator type, then run the program
on a system without that accelerator, or on a system where the target libraries are
not in a directory where the runtime library can find them, the program may fail at
runtime with an error message.
If you set the environment variable PGI_ACC_NOTIFY to a nonzero integer value,
the runtime library prints a line to standard error every time it launches a kernel on
the accelerator.
7.9. Accelerator Directives
This section provides an overview of the Fortran and C directives used to delineate
accelerator regions and to augment information available to the compiler for scheduling
of loops and classification of data.
7.9.1. Enable Accelerator Directives
PGI Accelerator compilers enable accelerator directives with the -acc and -ta
command line option. For more information on this option as it relates to the
Accelerator, refer to Applicable Command-line Options.
The syntax used to define directives allows compilers to ignore accelerator directives
if support is disabled or not provided.
_ACCEL macro
The _ACCEL macro name is defined to have a value yyyymm where yyyy is the year and
mm is the month designation of the version of the Accelerator directives supported by the
implementation. For example, the version for May, 2009 is 200905. The PGI compilers
define this macro when accelerator directives are enabled.
_OPENACC macro
The _OPENACC macro name is defined to have a value yyyymm where yyyy is the year
and mm is the month designation of the version of the OpenACC directives supported
by the implementation. For example, the version for June, 2013 is 201306. All OpenACC
compilers define this macro when OpenACC directives are enabled.
User's Guide for x86-64 CPUs
Version 2017 | 88
Using an Accelerator
7.9.2. Format
The specific format of the directive depends on the language and the format or form of
the source.
Directives include a name and clauses, and the format of the directive depends on the
type:
‣
‣
‣
C directives, described in ‘C Directives’
Free-form Fortran directives, described in ‘Free-Form Fortran Directives’
Fixed-form Fortran directives, described in ‘Fixed-Form Fortran Directives’
This document uses free form for all PGI Accelerator compiler Fortran directive
examples.
Rules
The following rules apply to all OpenACC compiler directives:
‣
‣
‣
‣
Only one directive-name can be specified per directive.
The order in which clauses appear is not significant.
Clauses may be repeated unless otherwise specified.
For clauses that have a list argument, a list is a comma-separated list of variable
names, array names, or, in some cases, subarrays with subscript ranges.
7.9.3. C Directives
In C, OpenACC compiler directives are specified using #pragma
Syntax
The syntax of an OpenACC compiler directive is:
#pragma acc directive-name [clause [,clause]...] new-line
Rules
In addition to the general directive rules, the following rules apply to OpenACC C
directives:
‣
‣
‣
‣
‣
Each directive starts with #pragma acc.
The remainder of the directive follows the C conventions for pragmas.
White space may be used before and after the #; white space may be required to
separate words in a directive.
Preprocessing tokens following the #pragma acc are subject to macro
replacement.
C directives are case sensitive.
User's Guide for x86-64 CPUs
Version 2017 | 89
Using an Accelerator
‣
An Accelerator directive applies to the immediately following structured block or
loop.
7.9.4. Free-Form Fortran Directives
OpenACC Fortran directives can be either Free-Form or Fixed-Form directives. FreeForm Accelerator directives are specified with the !$acc mechanism.
Syntax
The syntax of directives in free-form source files is:
!$acc directive-name [clause [,clause]...]
Rules
In addition to the general directive rules, the following rules apply to OpenACC FreeForm Fortran directives:
‣
‣
‣
‣
‣
‣
‣
‣
‣
‣
The comment prefix (!) may appear in any column, but may only be preceded by
white space (spaces and tabs).
The sentinel (!$acc) must appear as a single word, with no intervening white space.
Line length, white space, and continuation rules apply to the directive line.
Initial directive lines must have a space after the sentinel.
Continued directive lines must have an ampersand (&) as the last nonblank
character on the line, prior to any comment placed in the directive.
Comments may appear on the same line as the directive, starting with an
exclamation point and extending to the end of the line.
If the first nonblank character after the sentinel is an exclamation point, the line is
ignored.
Directives are case-insensitive.
Directives cannot be embedded within continued statements.
Statements must not be embedded within continued directives.
7.9.5. Fixed-Form Fortran Directives
Fixed-Form Accelerator directives are specified using one of three formats.
Syntax
The syntax of directives in fixed-form source files is one these three formats:
!$acc directive-name [clause [,clause]...]
c$acc directive-name [clause [,clause]...]
*$acc directive-name [clause [,clause]...]
User's Guide for x86-64 CPUs
Version 2017 | 90
Using an Accelerator
Rules
In addition to the general directive rules, the following rules apply to Accelerator FixedForm Fortran directives:
‣
‣
‣
‣
‣
‣
‣
The sentinel (!$acc, c$acc, or *$acc) must occupy columns 1-5.
Fixed form line length, white space, continuation, and column rules apply to the
directive line.
Initial directive lines must have a space or zero in column 6, and continuation
directive lines must have a character other than a space or zero in column 6.
Comments may appear on the same line as a directive, starting with an exclamation
point on or after column 7 and continuing to the end of the line.
Directives are case-insensitive.
Directives cannot be embedded within continued statements.
Statements must not be embedded within continued directives.
7.9.6. OpenACC Directive Summary
PGI currently supports these types of accelerator directives:
Parallel Directive
Kernels Directive
Loop Directive
Combined Directive
Data Directive
Enter Data and Exit Data Directives
Host_Data Directive
Cache Directive
Declare Directive
Update Directive
Routine Directive
Wait Directive
Table 16 lists and briefly describes each of the accelerator directives that PGI currently
supports. For a complete description of each directive, refer to ‘PGI Accelerator
Directives’ in the PGI Compiler’s Reference Guide.
Table 16 PGI Accelerator Directive Summary Table
This directive...
Accepts these clauses...
Parallel Directive
User's Guide for x86-64 CPUs
async [( int-expr )]
wait [( int-expr-list )]
num_gangs ( int-expr )
num_workers( int-expr
)
vector_length( int-expr
)
Has this functionality...
Defines the region of the program that should
be compiled for parallel execution on the
accelerator device.
Version 2017 | 91
Using an Accelerator
This directive...
Accepts these clauses...
Has this functionality...
if( condition )
reduction( operator :
list )
copy ( list )
copyin( list )
copyout( list )
create( list )
present( list )
deviceptr( list )
private( list )
firstprivate( list )
C Syntax
#pragma acc parallel [clause [, clause]...] new-line
structured block
Fortran Syntax
!$acc parallel [clause [, clause]...]
structured block
!$acc end parallel
Kernels Directive
async [( int-expr )]
wait [( int-expr-list )]
if( condition )
copy ( list )
copyin( list )
copyout( list )
create( list )
present( list )
deviceptr( list )
Defines the region of the program that should
be compiled into a sequence of kernels for
execution on the accelerator device.
C Syntax
#pragma acc kernels [clause [, clause]...] new-line
structured block
Fortran Syntax
!$acc kernels [clause [, clause]...]
structured block
!$acc end kernels
Data Directive
if( condition )
copy ( list )
copyin( list )
copyout( list )
create( list )
deviceptr( list )
Defines data, typically arrays, that should be
allocated in the device memory for the duration
of the data region, whether data should be
copied from the host to the device memory upon
region entry, and copied from the device to host
memory upon region exit.
C Syntax
#pragma acc data [clause [, clause]...] new-line
structured block
User's Guide for x86-64 CPUs
Version 2017 | 92
Using an Accelerator
This directive...
Accepts these clauses...
Has this functionality...
Fortran Syntax
!$acc data [clause [, clause]...]
structured block
!$acc end data
Enter Data Directive
if( condition )
async [( int-expr )]
wait [( int-expr-list )]
copyin( list )
create( list )
Defines data, typically arrays, that should be
allocated in the device memory for the duration
of the program or until an exit data
regdirective that deallocates the data, and
whether data should be copied from the host
to the device memory at the enter data
directive.
C Syntax
#pragma acc enter data [clause [, clause]...] new-line
FortranSyntax
!$acc enter data [clause [, clause]...]
Exit Data Directive
if( condition )
async [( int-expr )]
wait [( int-expr-list )]
copyout( list )
delete( list )
Defines data, typically arrays, that should be
allocated in the device memory, and whether
data should be copied from the device to the
host memory.
C Syntax
#pragma acc exit data [clause [, clause]...] new-line
Fortran Syntax
!$acc exit data [clause [, clause]...]
Host_Data Directive
use_device( list )
Makes the address of the device data available
on the host.
C Syntax
#pragma acc host_data [clause [, clause]...] new-line
structured block
Fortran Syntax
!$acc host_data [clause [, clause]...]
structured block
!$acc end host_data
Loop Directive
User's Guide for x86-64 CPUs
collapse( n )
gang [( gang-arg-list )]
worker [( [num:] intexpr )]
vector [( [length:] intexpr )]
seq
independent
private( list )
reduction( operator :
list )
Describes what type of parallelism to use to
execute the loop and declare loop-private
variables and arrays and reduction operations.
Applies to a loop which must appear on the
following line.
Version 2017 | 93
Using an Accelerator
This directive...
Accepts these clauses...
Has this functionality...
shortloop
C Syntax
#pragma acc loop [clause [,clause]...] new-line
for loop
Fortran Syntax
!$acc loop [clause [,clause]...]
do loop
Combined Parallel
Loop Directive
Any clause that is allowed
on a parallel directive or
a loop directive is allowed
on a combined parallel loop
directive.
Is a shortcut for specifying a loop directive
nested immediately inside an accelerator parallel
directive. The meaning is identical to explicitly
specifying a parallel construct containing a loop
directive.
C Syntax
#pragma acc parallel loop [clause [, clause]...] new-line
for loop
Fortran Syntax
!$acc parallel loop [clause [, clause]...]
do loop
Combined Kernels
Loop Directive
Any clause that is allowed
on a kernels directive or a
loop directive is allowed
on a combined kernels loop
directive.
Is a shortcut for specifying a loop directive
nested immediately inside an accelerator kernels
directive. The meaning is identical to explicitly
specifying a kernels construct containing a loop
directive.
C Syntax
#pragma acc kernels loop [clause [, clause]...] new-line
for loop
Fortran Syntax
!$acc kernels loop [clause [, clause]...]
do loop
Cache Directive
Specifies array elements or subarrays that should
be fetched into the highest level of the cache
for the body of a loop. Must appear at the top of
(inside of) the loop.
C Syntax
#pragma acc cache ( list )... new-line
Fortran Syntax
!$acc cache ( list )
Declare Directive
User's Guide for x86-64 CPUs
copy ( list )
copyin( list )
copyout( list )
create( list )
deviceptr( list )
device_resident( list )
Specifies that an array or arrays are to be
allocated in the device memory for the duration
of the implicit data region of a function,
subroutine, or program.
Specifies whether the data values are to be
transferred from the host to the device memory
upon entry to the implicit data region, and from
Version 2017 | 94
Using an Accelerator
This directive...
Accepts these clauses...
Has this functionality...
the device to the host memory upon exit from
the implicit data region.
Creates a visible device copy of the variable or
array.
C Syntax
#pragma acc declare clause [,clause]... new-line
FortranSyntax
!$acc declare clause [,clause]...
Update Directive
async [( int-expr )]
wait [( int-expr )]
device-type ( devtypelist )
if ( condition )
self ( list )
host ( list )
device( list )
Used during the lifetime of accelerator data
to update all or part of a host memory array
with values from the corresponding array in
device memory, or to update all or part of a
device memory array with values from the
corresponding array in host memory.
C Syntax
#pragma acc update clause [,clause]... new-line
Fortran Syntax
!$acc update clause [,clause]...
Routine Directive
gang
worker
vector
seq
Used to tell the compiler to compile a given
procedure for an accelerator as well as the host.
In a file or routine with a procedure call the
routine directive tells the implementation
the attributes of the procedure when called on
the accelerator.
C Syntax
#pragma acc routine clause [,clause]... new-line
or
#pragma acc routine ( name ) clause [,clause]... new-line
Fortran Syntax
!$acc routine clause [,clause]...
or
!$acc routine( name ) clause [,clause]...
Wait Directive
if( condition )
device( list )
Specifies to wait until all operations on a specific
device async queue or all async queues are
complete.
C Syntax
#pragma acc wait [(int-expr-list)] [clause [,clause]...] new-line
Fortran Syntax
!$acc wait [(int-expr-list)] [clause [,clause]...]
User's Guide for x86-64 CPUs
Version 2017 | 95
Using an Accelerator
7.10. Accelerator Directive Clauses
Table 17 provides an alphabetical listing and brief description of each clause that is
applicable for the various Accelerator directives. The table also indicates for which
directives the clause is applicable.
Table 17 Directive Clauses Summary
Use this clause...
In these
directives...
To do this...
async [(int-expr)]
Parallel
Kernels
Data
Enter Data
Exit Data
Update
The parallel or kernels region or data operations may
be processed asynchronously while the local thread
continues with the code following the construct or
directive. (The async clause on the data construct is a
PGI extension)
collapse (n)
Loop
Specifies how many tightly nested loops are associated
with the loop construct.
copy (list)
Parallel
Kernels
Data Region
Declare
At execution time, the implementation will test
whether each of the items in the list are already
present in the device memory. If not, corresponding
device memory will be allocated and the data will
be copied to the device at entry to the region,
and data will be copied back to the host and the
memory deallocated at exit from the region. For
compatibility with OpenACC 2.0, this may also be
spelled present_or_copy or pcopy.
copyin (list)
Parallel
Kernels
Data
Declare
Enter Data
At execution time, the implementation will test
whether each of the items in the list are already
present in the device memory. If not, corresponding
device memory will be allocated and the data will
be copied to the device at entry to the region, and
the memory deallocated at exit from the region. For
compatibility with OpenACC 2.0, this may also be
spelled present_or_copyin or pcopyin.
copyout (list)
Parallel
Kernels
Data
Declare
Exit Data
At execution time, the implementation will test
whether each of the items in the list are already
present in the device memory. If not, corresponding
device memory will be allocated at entry to the
region, and data will be copied back to the host and
the memory deallocated at exit from the region. For
compatibility with OpenACC 2.0, this may also be
spelled present_or_copyout or pcopyout.
create (list)
Parallel
Kernels
Data
Declare
Enter Data
At execution time, the implementation will test
whether each of the items in the list are already
present in the device memory. If not, corresponding
device memory will be allocated at entry to the region,
and the memory deallocated at exit from the region.
User's Guide for x86-64 CPUs
Version 2017 | 96
Using an Accelerator
Use this clause...
In these
directives...
To do this...
For compatibility with OpenACC 2.0, this may also be
spelled present_or_create or pcreate.
delete (list)
Exit Data
At execution time, the implementation will deallocate
the corresponding device memory.
device (list)
Update
Copies the variables, arrays, or subarrays in the list
argument from host memory to the visible device copy
of the variables, arrays, or subarrays in device memory.
device_resident (list)
Declare
Specifies that the memory of the named variables
should be allocated in the accelerator device memory
and not in the host memory.
deviceptr (list)
Parallel
Kernels
Data
Declare
Declares that the items in the list are device pointers,
so the data need not be allocated or moved between
the host and device..
firstprivate (list)
Parallel
Declares that a copy of each item on the list will be
created for each parallel gang, and that the copy will
be initialized with the value of that item on the host
when the parallel construct is encountered.
gang [(gang-arg-list)]
Loop
Routine
Specifies that the iterations of the associate loop or
loops are to be executed in parallel by distributing
the iterations among the gangs created by the parallel
construct.
host (list)
Update
Copies the visible device copies of the variables, arrays,
or subarrays in the list argument to the associated host
memory locations. The copy occurs after completion of
the compute or data region.
if (condition)
Parallel
Kernels
When present, tells the compiler to generate two copies
of the region – one for the accelerator, one for the
host – and to generate code to decide which copy to
execute.
if (condition)
Data Region
Enter Data
Exit Data
Update
Conditionally allocate memory on, and move data to
and/or from the device.
independent
Loop
Tells the compiler that the iterations of this loop are
data-independent of each other, thus allowing the
compiler to generate code to execute the iterations in
parallel, without synchronization.
num_gangs (int-expr)
Parallel
Defines the number of parallel gangs that will execute
the region.
num_workers (intexpr)
Parallel
Defines the number of workers within each gang that
will be active after a gang transitions from workersingle mode to worker-partitioned mode.
present (list)
Parallel
Kernels
Tells the implementation that the items in the list are
already present in device memory.
User's Guide for x86-64 CPUs
Version 2017 | 97
Using an Accelerator
Use this clause...
In these
directives...
To do this...
Data
Declare
private (list)
Loop
Specifies that a copy of each item in the list will be
created for each thread that executes one or more
iterations of the associated loop or loops.
private (list)
Parallel
Declares that a copy of each item on the list will be
created for each parallel gang.
reduction (operator:
list)
Loop
For each variable in the list, a private copy is created
for each thread that executes iterations of the
associated loop or loops and initialized for the operator.
At the end of the loop, the values for each thread are
combined using the reduction operator, and the result
combined with the value of the original variable and
stored in the original variable.
reduction (operator:
list)
Parallel
For each variable in the list, a private copy is created
for each parallel gang and initialized for the operator.
At the end of the region, the values for each gang are
combined using the reduction operator, and the result
combined with the value of the original variable and
stored in the original variable.
self (list)
Update
Specifies that the items in the list are to be copied from
the accelerator device memory to the local memory.
The self clause is a synonym for the host clause.
seq
Loop
Tells the compiler to execute this loop sequentially
on the accelerator. There is no maximum number of
iterations for a seq schedule.
shortloop
Loop
If the loop has the vector clause, this tells the compiler
that the trip count for the loop is less than or equal to
the number of vector lanes created for this kernel, as
specified by the vector_length on the parallel construct
or vector clause in a kernels construct. (PGI extension)
use_device (list)
Host_Data
Tells the compiler to use the device address of any item
in the list in code within the construct.
vector [( [length:]
int-expr)]
Loop
Tells the compiler to execute this loop in vector or SIMD
mode on the accelerator.
vector_length (intexpr)
Parallel
Defines the number of vector lanes that will be active
after a worker transitions from vector-single mode to
vector-partitioned mode.
wait [( int-expr-list )]
Parallel
Kernels
Enter Data
Exit Data
Update
The compute, data or update operation may not be
launched or executed until all operations enqueued
up to this point by this thread on the associated
asynchronous device activity queues have completed.
worker [( [num:] intexpr)]
Loop
Specifies that the iterations of the associated loop or
loops are to be executed in parallel by distributing the
iterations among the multiple workers within a single
gang.
User's Guide for x86-64 CPUs
Version 2017 | 98
Using an Accelerator
7.11. OpenAcc Runtime Libraries
This section provides an overview of the user-callable functions and library routines that
are available for use by programmers to query the accelerator features and to control
behavior of accelerator-enabled programs at runtime.
In Fortran, none of the OpenACC runtime library routines may be called from a PURE
or ELEMENTAL procedure.
7.11.1. Runtime Library Definitions
There are separate runtime library files for C and for Fortran.
C Runtime Library Files
In C, prototypes for the runtime library routines are available in a header file named
accel.h. All the library routines are extern functions with ‘C’ linkage. This file
defines:
‣
‣
The prototypes of all routines in this section.
Any data types used in those prototypes, including an enumeration type to describe
types of accelerators.
Fortran Runtime Library Files
In Fortran, interface declarations are provided in a Fortran include file named
accel_lib.h and in a Fortran module named accel_lib. These files define:
‣
‣
‣
‣
Interfaces for all routines in this section.
Integer parameters to define integer kinds for arguments to those routines.
Integer parameters to describe types of accelerators.
The integer parameter accel_version with a value yyyymm where yyyy and mm
are the year and month designations of the version of the Accelerator programming
model supported. This value matches the value of the preprocessor variable
_ACCEL.
7.11.2. Runtime Library Routines
Table 18 lists and briefly describes the runtime library routines supported by PGI in
addition to the standard OpenACC runtine API routines.
User's Guide for x86-64 CPUs
Version 2017 | 99
Using an Accelerator
Table 18 Accelerator Runtime Library Routines
This Runtime Library
Routine...
Does this...
acc_allocs
Returns the number of arrays allocated in data or compute regions.
acc_bytesalloc
Returns the total bytes allocated by data or compute regions.
acc_bytesin
Returns the total bytes copied in to the accelerator by data or compute
regions.
acc_bytesout
Returns the total bytes copied out from the accelerator by data or compute
regions.
acc_copyins
Returns the number of arrays copied in to the accelerator by data or
compute regions.
acc_copyouts
Returns the number of arrays copied out from the accelerator by data or
compute regions.
acc_disable_time
Tells the runtime to stop profiling accelerator regions and kernels.
acc_enable_time
Tells the runtime to start profiling accelerator regions and kernels, if it is
not already doing so.
acc_exec_time
Returns the number of microseconds spent on the accelerator executing
kernels.
acc_frees
Returns the number of arrays freed or deallocated in data or compute
regions.
acc_get_device
Returns the type of accelerator device used to run the next accelerator
region, if one is selected.
acc_get_device_num
Returns the number of the device being used to execute an accelerator
region.
acc_get_free_memory
Returns the total available free memory on the attached accelerator
device.
acc_get_memory
Returns the total memory on the attached accelerator device.
acc_get_num_devices
Returns the number of accelerator devices of the given type attached to
the host.
acc_kernels
Returns the number of accelerator kernels launched since the start of the
program.
acc_present_dump
Summarizes all data present on the current device.
acc_present_dump_all
Summarizes all data present on all devices.
acc_regions
Returns the number of accelerator regions entered since the start of the
program.
acc_total_time
Returns the number of microseconds spent in accelerator compute regions
and in moving data for accelerator data regions.
7.12. Environment Variables
PGI supports environment variables that modify the behavior of accelerator regions.
This section defines the user-setable environment variables used to control behavior of
User's Guide for x86-64 CPUs
Version 2017 | 100
Using an Accelerator
accelerator-enabled programs at execution. These environment variables must comply
with these rules:
‣
‣
‣
The names of the environment variables must be upper case.
The values assigned environment variables are case insensitive and may have
leading and trailing white space.
The behavior is implementation-defined if the values of the environment variables
change after the program has started, even if the program itself modifies the values.
Table 19 lists and briefly describes the Accelerator environment variables that PGI
supports.
Table 19 Accelerator Environment Variables
This environment variable... Does this...
ACC_DEVICE_TYPE
Controls which accelerator device to use when executing accelerator
regions, if the program has been compiled to use more than one
different type of device. The value of this environment variable is
implementation-defined, and currently may be the string NVIDIA, or
HOST.
ACC_DEVICE_NUM
Controls the default device number to use when executing
accelerator regions. The value of this environment variable must
be a nonnegative integer between zero and the number of devices
attached to the host.
PGI_ACC_NOTIFY
When set to an integer value, the value is used as a bit mask to print
information about kernel launches (value 1), data transfers (value
2), wait operations or synchronizations with the device (value 4),
region entry/exit (value 8), and data allocate/free (value 16).
PGI_ACC_TIME
Enables a lightweight profiler to measure data movement and
accelerator kernel execution time and print a summary at the end of
program execution.
PGI_ACC_BUFFERSIZE
For NVIDIA CUDA devices, this defines the size of the pinned buffer
used to transfer data between host and device.
PGI_ACC_GANGLIMIT
For NVIDIA CUDA devices, this defines the maximum number of gangs
(CUDA thread blocks) that will be launched by a kernel.
7.13. Applicable Command-line Options
The following command line options are applicable specifically when working with
accelerators.
-ta
Use this option to enable recognition of the !$ACC directives in Fortran, and
#pragma acc directives in C.
-tp
Use this option to specify the target host processor architecture.
-acc
Use this option to enable OpenACC directives. You can use the -acc suboptions
to specify loop autoparallelization, how the compiler reports compute regions
User's Guide for x86-64 CPUs
Version 2017 | 101
Using an Accelerator
failures to accelerate, and whether to issue a warning or an error for non-OpenACC
accelerator directives.
-Minfo or -Minfo=accel
Use this option to see messages about the success or failure of the compiler in
translating the accelerator region into GPU kernels.
The -ta flag has the following accelerator-related suboptions:
tesla
Select NVIDIA accelerator target. This option has a number of suboptions:
cc20, cc30, cc35,
cc50, cc60
Generate code for compute capability 2.0, 3.0, 3.5, 5.0 or 6.0 respectively;
multiple selections are valid.
cuda7.5 or 7.5
Specify the CUDA 7.5 version of the toolkit. This is the default.
cuda8.0 or 8.0
Specify the CUDA 8.0 version of the toolkit.
fastmath
Use routines from the fast math library.
fermi
Generate code for Fermi Architecture equivalent to NVIDIA compute capability
2.x.
[no]flushz
Control flush-to-zero mode for floating point computations in the GPU code.
keep
Keep the kernel files.
kepler
Generate code for Kepler Architecture equivalent to NVIDIA compute
capability 3.x.
maxregcount:n
Specify the maximum number of registers to use on the GPU.
Leaving this blank indicates no limit.
[no]fma
Control generation of fused multiply-add instructions.
noL1
Prevent the use of L1 hardware data cache to cache global variables.
[no]rdc
Control generation of relocatable device code (default on).
host
Generate host code for accelerator regions.
The compiler automatically invokes the necessary software tools to create the kernel
code and embeds the kernels in the object file.
To access accelerator libraries, you must link an accelerator program with the -ta
flag.
User's Guide for x86-64 CPUs
Version 2017 | 102
Using an Accelerator
7.14. Profiling Accelerator Kernels
This release supports the environment variable PGI_ACC_TIME for 64-bits. Setting
this environment variable to a nonzero value enables collection and printing of simple
timing information about the accelerator regions and generated kernels.
Turn off all CUDA Profilers (NVIDIA's Visual Profiler, NVPROF, CUDA_PROFILE, etc)
when enabling PGI_ACC_TIME, they use the same library to gather performance data
and cannot be used concurently.
Windows Users: To ensure that all the performance information is collected we
recommend that 'acc_shutdown' is called before your application is finished and 'main'
exits.
Accelerator Kernel Timing Data
bb04.f90
s1
15: region entered 1 times
time(us): total=1490738
init=1489138 region=1600
kernels=155 data=1445
w/o init: total=1600 max=1600
min=1600 avg=1600
18: kernel launched 1 times
time(us): total=155 max=155 min=155 avg=155
In this example, a number of things are occurring:
‣
‣
‣
‣
For each accelerator region, the file name bb04.f90 and subroutine or function
name s1 is printed, with the line number of the accelerator region, which in the
example is 15.
The library counts how many times the region is entered (1 in the example) and
the microseconds spent in the region (in this example 1490738), which is split into
initialization time (in this example 1489138) and execution time (in this example
1600).
The execution time is then divided into kernel execution time and data transfer time
between the host and GPU.
For each kernel, the line number is given, (18 in the example), along with a count of
kernel launches, and the total, maximum, minimum, and average time spent in the
kernel, all of which are 155 in this example.
7.15. Related Accelerator Programming Tools
User's Guide for x86-64 CPUs
Version 2017 | 103
Using an Accelerator
7.15.1. NVIDIA CUDA Profile
You can use the NVIDIA CUDA Profiler with PGI-generated code for the NVIDIA
GPUs. You may download the CUDA Profiler from the same website as the CUDA
software:www.nvidia.com/cuda
Documentation and support is provided by NVIDIA.
7.15.2. TAU – Tuning and Analysis Utility
You can use the TAU (Tuning and Analysis Utility), version 2.18.1+, with PGI-generated
accelerator code. TAU instruments code at the function or loop level, and version 2.18.1
is enhanced with support to track performance in accelerator regions. TAU software and
documentation is available at this website: http://tau.uoregon.edu
7.16. Supported Intrinsics
An intrinsic is a function available in a given language whose implementation is
handled specifically by the compiler. Typically, an intrinsic substitutes a sequence of
automatically-generated instructions for the original function call. Since the compiler has
an intimate knowledge of the intrinsic function, it can better integrate it and optimize it
for the situation.
Intrinsics make the use of processor-specific enhancements easier because they provide
a language interface to assembly instructions. In doing so, the compiler manages things
that the user would normally have to be concerned with, such as register names, register
allocations, and memory locations of data.
This section contains an overview of the Fortran and C intrinsics that the accelerator
supports.
7.16.1. Supported Fortran Intrinsics Summary Table
Table 20 is an alphabetical summary of the supported Fortran intrinsics that the
accelerator supports. These functions are specific to Fortran 90/95 unless otherwise
specified.
For complete descriptions of these intrinsics, refer to ‘Fortran Intrinsics’ of the PGI
Fortran Reference Manual.
In most cases PGI provides support for all the data types for which the intrinsic is valid.
When support is available for only certain data types, the middle column of the table
specifies which ones, using the following codes:
I for integer
User's Guide for x86-64 CPUs
S for single precision real
C for single precision complex
D for double precision real
Z for double precision complex
Version 2017 | 104
Using an Accelerator
Table 20 Supported Fortran Intrinsics
This intrinsic
ABS
Returns this value ...
I,S,D
absolute value of the supplied argument.
ACOS
arccosine of the specified value.
AINT
truncation of the supplied value to a whole number.
ANINT
nearest whole number to the supplied argument.
ASIN
arcsine of the specified value.
ATAN
arctangent of the specified value.
ATAN2
arctangent of the specified value.
COS
S,D
COSH
DBLE
hyperbolic cosine of the specified value.
S,D
DPROD
EXP
cosine of the specified value.
conversion of the value to double precision real.
double precision real product.
S,D
exponential value of the argument.
IAND
result of a bit-by-bit logical AND on the arguments.
IEOR
result of a bit-by-bit logical exclusive OR on the arguments.
INT
I,S,D
IOR
LOG
conversion of the value to integer type.
result of a bit-by-bit logical OR on the arguments.
S,D
natural logarithm of the specified value.
LOG10
base-10 logarithm of the specified value.
MAX
maximum value of the supplied arguments.
MIN
minimum value of the supplied arguments.
MOD
I
remainder of the division.
NINT
nearest integer to the real argument.
NOT
result of a bit-by-bit logical complement on the argument.
REAL
I,S,D
SIGN
SIN
absolute value of A times the sign of B.
S,D
SINH
SQRT
conversion of the argument to real.
value of the sine of the argument.
hyperbolic sine of the argument.
S,D
square root of the argument.
TAN
tangent of the specified value.
TANH
hyperbolic tangent of the specified value.
User's Guide for x86-64 CPUs
Version 2017 | 105
Using an Accelerator
7.16.2. Supported C Intrinsics Summary Table
This section contains two alphabetical summaries – one for double functions and a
second for float functions. These lists contain only those C intrinsics that the accelerator
supports.
Table 21 Supported C Intrinsic Double Functions
This intrinsic
Returns this value ...
acos
arccosine of the specified value.
asin
arcsine of the specified value.
atan
arctangent of the specified value.
atan2
arctangent of y/x, where y is the first argument, x the second.
cos
cosine of the specified value.
cosh
hyperbolic cosine of the specified value.
exp
exponential value of the argument.
fabs
absolute value of the argument.
fmax
maximum value of the two supplied arguments
fmin
minimum value of the two supplied arguments
log
natural logarithm of the specified value.
log10
base-10 logarithm of the specified value.
pow
value of the first argument raised to the power of the second argument.
sin
value of the sine of the argument.
sinh
hyperbolic sine of the argument.
sqrt
square root of the argument.
tan
tangent of the specified value.
tanh
hyperbolic tangent of the specified value.
Table 22 Supported C Intrinsic Float Functions
This intrinsic
Returns this value ...
acosf
arccosine of the specified value.
asinf
arcsine of the specified value.
atanf
arctangent of the specified value.
atan2f
arctangent of y/x, where y is the first argument, x the second.
cosf
cosine of the specified value.
coshf
hyperbolic cosine of the specified value.
expf
exponential value of the floating-point argument.
User's Guide for x86-64 CPUs
Version 2017 | 106
Using an Accelerator
This intrinsic
Returns this value ...
fabsf
absolute value of the floating-point argument.
logf
natural logarithm of the specified value.
log10f
base-10 logarithm of the specified value.
powf
value of the first argument raised to the power of the second argument.
sinf
value of the sine of the argument.
sinhf
hyperbolic sine of the argument.
sqrtf
square root of the argument.
tanf
tangent of the specified value.
tanhf
hyperbolic tangent of the specified value.
7.17. References related to Accelerators
‣
‣
‣
‣
‣
‣
‣
‣
ISO/IEC 1539-1:1997, Information Technology – Programming Languages – Fortran,
Geneva, 1997 (Fortran 95).
American National Standard Programming Language C, ANSI X3.159-1989 (ANSI
C).
ISO/IEC 9899:1999, Information Technology – Programming Languages – C, Geneva,
1999 (C99).
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).
PGI Debugger User's Guide, The Portland Group. Available online at http://
www.pgroup.com/doc/pgdbg.pdf.
PGI Profiler User's Guide, The Portland Group. Available online at http://
www.pgroup.com/doc/pgprof.pdf.
PGI Fortran Reference, The Portland Group. Available online at http://
www.pgroup.com/doc/pgifortref.pdf.
User's Guide for x86-64 CPUs
Version 2017 | 107
Chapter 8.
ECLIPSE
This document explains how to install and use the PGI plugin for Eclipse CDT (C/C++
development tool). PGI Eclipse integration is only available on Linux.
8.1. Install Eclipse CDT
To install the Eclipse plugin for the PGI C and C++ compilers:
1. Download the Eclipse plugin from the PGI website at http://www.pgroup.com/
support/downloads.php.
2. Untar and install the PGI Eclipse plugin componennts.
$ tar zxf pgieclipse-2017.tar.gz
$ ./insall_compiilers
Note you will be asked to provide the location of your current PGI installation. The
default is /opt/pgi.
3. Before you install the actual PGI Eclipse plugin, from within Eclipse, check your CDT
version.
1. Go to Help -> About Eclipse
2. Click the Eclipse CDT button.
You might need to hover the mouse pointer on the button to see the hint.
3. Select Eclipse C/C++ Development Tools.
The first number in the feature version specifies which plugin version is selected.
4. Go to Help -> Install New software.
5. Click the Add button to add a new software repository.
6. In the Add Repository dialog box:
1. Click Local.
2. Select your PGI installation directory, such as /opt/pgi.
3. Browse inside 2017/eclipse and select the directory matching your CDT
version.
4. Click OK.
User's Guide for x86-64 CPUs
Version 2017 | 108
Eclipse
The Add Repository dialog should show the path to the local directory
containing the plugin for your CDT version. For example, if PGI compilers
are installed in /opt/pgi, then the CDT 7 plugin is located in /opt/pgi/<osversion>/2017/eclipse/cdt7; the CDT 8 plugin is in /opt/pgi/<osversion>/2017/eclipse/cdt8, and so on.
5. Click OK in the Add Repository dialog.
The install form now shows “PGI C/C++ Compiler Plugin” as an option to install.
7. Check the box next to PGI option and select Next to get to the Install Details view.
8. Click Next again.
9. Review and accept the End-User License agreement.
10.Click Finish.
You are prompted to restart. Select Restart to complete installation of the plugin.
8.2. Use Eclipse CDT
To use the Eclipse plugin for the PGI C and C++ compilers, the directory containing PGI
compilers and tools should be included in your PATH prior to starting the Eclipse IDE.
For details on how to include this directory in your PATH environment variable, refer to
Using Environment Variables, and specifically to PATH.
This plugin does not currently support the Code Analysis feature of Eclipse CDT. This
feature is not disabled by default for PGI projects and is the cause of spurious syntax
errors during pre-compilation. This feature can be disabled manually at either the
Project or Workspace level; any actual compilation or link errors are reported at build
time.
‣
‣
To disable Code Analysis at the Workspace level, select the menu item Window
| Preferences, then select C/C++, then Code Analysis; uncheck all category items,
which should deselect all sub-items.
To disable Code Analysis at the Project level, select the menu item Project |
Properties, then select C/C++ General, then Code Analysis; uncheck the top-level
categories to deselect everything.
The PGI plugin follows the same rules for creating, building, and running a project
as any other compiler supported by Eclipse. For more information, refer to Eclipse
documentation and tutorials at: http://www.eclipse.org/documentation/.
User's Guide for x86-64 CPUs
Version 2017 | 109
Chapter 9.
USING DIRECTIVES AND PRAGMAS
It is often useful to be able to alter the effects of certain command line options or default
behavior of the compiler. Fortran directives and C/C++ pragmas 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 and pragmas 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 and pragmas
to tune selected routines or loops.
9.1. PGI Proprietary Fortran Directives
PGI Fortran compilers support proprietary directives that may have any of the following
forms:
!pgi$g directive
!pgi$r directive
!pgi$l directive
!pgi$ directive
If the input is in fixed format, the comment character must begin in column 1 and
either * or C is allowed in place of !.
The scope indicator controls the scope of the directive. This indicator occurs after the $.
Some directives ignore the scope indicator.
The valid scopes, shown in the previous forms of the directive, are these:
g
(global) indicates the directive applies to the end of the source file.
r
(routine) indicates the directive applies to the next subprogram.
l
(loop) indicates the directive applies to the next loop, but not to any loop contained
within the loop body. Loop-scoped directives are only applied to DO loops.
blank
indicates that the default scope for the directive is applied.
User's Guide for x86-64 CPUs
Version 2017 | 110
Using Directives and Pragmas
The body of the directive may immediately follow the scope indicator. Alternatively, any
number of blanks may precede the name of the directive. Any names in the body of the
directive, including the directive name, may not contain embedded blanks. Blanks may
surround any special characters, such as a comma or an equal sign.
The directive name, including the directive prefix, may contain upper or lower case
letters, and the case is not significant. Case is significant for any variable names that
appear in the body of the directive if the command line option -Mupcase is selected. For
compatibility with other vendors’ directives, the prefix cpgi$ may be substituted with
cdir$ or cvd$.
9.2. PGI Proprietary C and C++ Pragmas
Pragmas may be supplied in a C/C++ source file to provide information to the compiler.
Many pragmas have a corresponding command-line option. Pragmas may also toggle an
option, selectively enabling and disabling the option.
The general syntax of a pragma is:
#pragma [ scope ] pragma-body
The optional scope field is an indicator for the scope of the pragma; some pragmas
ignore the scope indicator.
The valid scopes are:
global
indicates the pragma applies to the entire source file.
routine
indicates the pragma applies to the next function.
loop
indicates the pragma applies to the next loop (but not to any loop contained within
the loop body). Loop-scoped pragmas are only applied to for and while loops.
If a scope indicator is not present, the default scope, if any, is applied. Whitespace must
appear after the pragma keyword and between the scope indicator and the body of the
pragma. Whitespace may also surround any special characters, such as a comma or an
equal sign. Case is significant for the names of the pragmas and any variable names that
appear in the body of the pragma.
9.3. PGI Proprietary Optimization Directive and
Pragma Summary
The following table summarizes the supported Fortran directives and C/C++ pragmas.
The following terms are useful in understanding the table.
‣
Functionality is a brief summary of the way to use the directive or pragma. For a
complete description, refer to the ‘Directives and Pragmas Reference’ section of the
PGI Compiler’s Reference Guide.
User's Guide for x86-64 CPUs
Version 2017 | 111
Using Directives and Pragmas
‣
‣
Many of the directives and pragmas can be preceded by NO. The default entry
indicates the default for the directive or pragma. N/A appears if a default does not
apply.
The scope entry indicates the allowed scope indicators for each directive or pragma,
with L for loop, R for routine, and G for global. The default scope is surrounded by
parentheses and N/A appears if the directive or pragma is not available in the given
language.
The "*" in the scope indicates this:
For routine-scoped directive
The scope includes the code following the directive or pragma until the end of
the routine.
For globally-scoped directive
The scope includes the code following the directive or pragma until the end of
the file rather than for the entire file.
The name of a directive or pragma may also be prefixed with -M.
For example, you can use the directive -Mbounds, which is equivalent to the
directive bounds and you can use -Mopt, which is equivalent to opt. For pragmas,
you can use the directive -Mnoassoc, which is equivalent to the pragma noassoc,
and -Mvintr, which is equivalent to vintr.
Table 23 Proprietary Optimization-Related Fortran Directive and C/C++
Pragma Summary
Directive or
pragma
Functionality
Default
Fortran
Scope
C/C++
Scope
altcode
(noaltcode)
Do/don’t generate alternate code for
vectorized and parallelized loops.
altcode
(L)RG
(L)RG
assoc (noassoc)
Do/don’t perform associative
transformations.
assoc
(L)RG
(L)RG
bounds (nobounds) Do/don’t perform array bounds checking. nobounds
(R)G*
(R)G
cncall (nocncall)
Loops are considered for parallelization,
even if they contain calls to user-defined
subroutines or functions, or if their loop
counts do not exceed usual thresholds.
nocncall
(L)RG
(L)RG
concur (noconcur)
Do/don’t enable auto-concurrentization
of loops.
concur
(L)RG
(L)RG
depchk
(nodepchk)
Do/don’t ignore potential data
dependencies.
depchk
(L)RG
(L)RG
eqvchk (noeqvchk) Do/don’t check EQUIVALENCE for data
dependencies.
eqvchk
(L)RG
N/A
fcon (nofcon)
Do/don’t assume unsuffixed real
constants are single precision.
nofcon
N/A
(R)G
invarif (noinvarif)
Do/don’t remove invariant if constructs
from loops.
invarif
(L)RG
(L)RG
User's Guide for x86-64 CPUs
Version 2017 | 112
Using Directives and Pragmas
Directive or
pragma
Functionality
Default
Fortran
Scope
C/C++
Scope
ivdep
Ignore potential data dependencies.
ivdep
(L)RG
N/A
lstval (nolstval)
Do/don’t compute last values.
lstval
(L)RG
(L)RG
prefetch
Control how prefetch instructions are
emitted
opt
Select optimization level.
N/A
(R)G
(R)G
safe (nosafe)
Do/don’t treat pointer arguments as
safe.
safe
N/A
(R)G
safe_lastval
Parallelize when loop contains a scalar
used outside of loop.
not enabled
(L)
(L)
safeptr (nosafeptr) Do/don’t ignore potential data
dependencies to pointers.
nosafeptr
N/A
L(R)G
single (nosingle)
Do/don’t convert float parameters to
double.
nosingle
N/A
(R)G*
tp
Generate PGI Unified Binary code
optimized for specified targets.
N/A
(R)G
(R)G
unroll (nounroll)
Do/don’t unroll loops.
nounroll
(L)RG
(L)RG
vector (novector)
Do/don't perform vectorizations.
vector
(L)RG*
(L)RG
vintr (novintr)
Do/don’t recognize vector intrinsics.
vintr
(L)RG
(L)RG
9.4. Scope of Fortran Directives and CommandLine Options
During compilation the effect of a directive may be to either turn an option on, or
turn an option off. Directives apply to the section of code following the directive,
corresponding to the specified scope, which may include the following loop, the
following routine, or the rest of the program. This section presents several examples that
show the effect of directives as well as their scope. Consider the following Fortran code:
integer maxtime, time
parameter (n = 1000, maxtime = 10)
double precision a(n,n), b(n,n), c(n,n)
do time = 1, maxtime
do i = 1, n
do j = 1, n
c(i,j) = a(i,j) + b(i,j)
enddo
enddo
enddo
end
When compiled with -Mvect, both interior loops are interchanged with the outer loop.
$ pgfortran -Mvect dirvect1.f
User's Guide for x86-64 CPUs
Version 2017 | 113
Using Directives and Pragmas
Directives alter this behavior either globally or on a routine or loop by loop basis. To
assure that vectorization is not applied, use the novector directive with global scope.
!pgi$g novector
integer maxtime, time
parameter (n = 1000, maxtime = 10)
double precision a(n,n), b(n,n), c(n,n)
do time = 1, maxtime
do i = 1, n
do j = 1, n
c(i,j) = a(i,j) + b(i,j)
enddo
enddo
enddo
end
In this version, the compiler disables vectorization for the entire source file. Another use
of the directive scoping mechanism turns an option on or off locally, either for a specific
procedure or for a specific loop:
integer maxtime, time
parameter (n = 1000, maxtime = 10)
double precision a(n,n), b(n,n), c(n,n)
!pgi$l novector
do time = 1, maxtime
do i = 1, n
do j = 1, n
c(i,j) = a(i,j) + b(i,j)
enddo
enddo
enddo
end
Loop level scoping does not apply to nested loops. That is, the directive only applies to
the following loop. In this example, the directive turns off vector transformations for the
top-level loop. If the outer loop were a timing loop, this would be a practical use for a
loop-scoped directive.
9.5. Scope of C/C++ Pragmas and Command-Line
Options
During compilation a pragma either turns an option on or turns an option off. Pragmas
apply to the section of code corresponding to the specified scope – either the entire file,
the following loop, or the following or current routine. This section presents several
examples showing the effect of pragmas and the use of the pragma scope indicators.
In all cases, pragmas override a corresponding command-line option.
For pragmas that have only routine and global scope, there are two rules for
determining the scope of the pragma. We cover these special scope rules at the end of
this section.
User's Guide for x86-64 CPUs
Version 2017 | 114
Using Directives and Pragmas
Consider the following program:
main() {
float a[100][100], b[100][100], c[100][100];
int time, maxtime, n, i, j;
maxtime=10;
n=100;
for (time=0; time<maxtime;time++)
for (j=0; j<n;j++)
for (i=0; i<n;i++)
c[i][j] = a[i][j] + b[i][j];
}
When this is compiled using the -Mvect command-line option, both interior loops
are interchanged with the outer loop. Pragmas alter this behavior either globally or
on a routine or loop by loop basis. To ensure that vectorization is not applied, use the
novector pragma with global scope.
main() {
#pragma global novector
float a[100][100], b[100][100],c[100][100];
int time, maxtime, n, i, j;
maxtime=10;
n=100;
for (time=0; time<maxtime;time++)
for (j=0; j<n;j++)
for (i=0; i<n;i++)
c[i][j] = a[i][j] + b[i][j];
}
In this version, the compiler does not perform vectorization for the entire source file.
Another use of the pragma scoping mechanism turns an option on or off locally either
for a specific procedure or for a specific loop. The following example shows the use of a
loop-scoped pragma.
main() {
float a[100][100], b[100][100],c[100][100];
int time, maxtime, n, i, j;
maxtime=10;
n=100;
#pragma loop novector
for (time=0; time<maxtime;time++)
for (j=0; j<n;j++)
for (i=0; i<n;i++)
c[i][j] = a[i][j] + b[i][j];
}
Loop level scoping does not apply to nested loops. That is, the pragma only applies to
the following loop. In this example, the pragma turns off vector transformations for the
User's Guide for x86-64 CPUs
Version 2017 | 115
Using Directives and Pragmas
top-level loop. If the outer loop were a timing loop, this would be a practical use for a
loop-scoped pragma. The following example shows routine pragma scope:
#include "math.h"
func1() {
#pragma routine novector
float a[100][100], b[100][100];
float c[100][100], d[100][100];
int i,j;
for (i=0;i<100;i++)
for (j=0;j<100;j++)
a[i][j] = a[i][j] + b[i][j] * c[i][j];
c[i][j] = c[i][j] + b[i][j] * d[i][j];
}
func2() {
float a[200][200], b[200][200];
float c[200][200], d[200][200];
int i,j;
for (i=0;i<200;i++)
for (j=0;j<200;j++)
a[i][j] = a[i][j] + b[i][j] * c[i][j];
c[i][j] = c[i][j] + b[i][j] * d[i][j];
}
When this source is compiled using the -Mvect command-line option, func2 is
vectorized but func1 is not vectorized. In the following example, the global novector
pragma turns off vectorization for the entire file.
#include "math.h"
func1() {
#pragma global novector
float a[100][100], b[100][100];
float c[100][100], d[100][100];
int i,j;
for (i=0;i<100;i++)
for (j=0;j<100;j++)
a[i][j] = a[i][j] + b[i][j]
c[i][j] = c[i][j] + b[i][j]
}
func2() {
float a[200][200], b[200][200];
float c[200][200], d[200][200];
int i,j;
for (i=0;i<200;i++)
for (j=0;j<200;j++)
a[i][j] = a[i][j] + b[i][j]
c[i][j] = c[i][j] + b[i][j]
}
* c[i][j];
* d[i][j];
* c[i][j];
* d[i][j];
Special Scope Rules
Special rules apply for a pragma with loop, routine, and global scope. When the pragma
is placed within a routine, it applies to the routine from its point in the routine to the end
of the routine. The same rule applies for one of these pragmas with global scope.
However, there are several pragmas for which only routine and global scope applies and
which affect code immediately following the pragma:
‣
bounds and fcon – The bounds and fcon pragmas behave in a similar manner to
pragmas with loop scope. That is, they apply to the code following the pragma.
User's Guide for x86-64 CPUs
Version 2017 | 116
Using Directives and Pragmas
‣
opt and safe – When the opt or safe pragmas are placed within a routine, they apply
to the entire routine as if they had been placed at the beginning of the routine.
9.6. Prefetch Directives and Pragmas
Today’s processors are so fast that it is difficult to bring data into them quickly enough
to keep them busy. 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.
When vectorization is enabled using the -Mvect or -Mprefetch compiler options,
or an aggregate option such as -fast that incorporates -Mvect, the PGI compilers
selectively emit instructions to explicitly prefetch data into the data cache prior to first
use. You can control how these prefetch instructions are emitted by using prefetch
directives and pragmas.
For a list of processors that support prefetch instructions refer to the PGI Release Notes.
9.6.1. Prefetch Directive Syntax in Fortran
The syntax of a prefetch directive is as follows:
!$mem prefetch <var1>[,<var2>[,...]]
where <varn> is any valid variable, member, or array element reference.
9.6.2. Prefetch Directive Format Requirements
The sentinel for prefetch directives is !$mem, which is distinct from the !pgi$
sentinel used for optimization directives. Any prefetch directives that use the !pgi$
sentinel are ignored by the PGI compilers.
‣
‣
‣
‣
‣
The "c" must be in column 1 for fixed format.
Either * or ! is allowed in place of c for fixed format.
The scope indicators g, r and l used with the !pgi$ sentinel are not supported.
The directive name, including the directive prefix, may contain upper or lower case
letters and is case insensitive (case is not significant).
If the command line option -Mupcase is used, any variable names that appear in
the body of the directive are case sensitive.
User's Guide for x86-64 CPUs
Version 2017 | 117
Using Directives and Pragmas
9.6.3. Sample Usage of Prefetch Directive
Prefetch Directive Use
This example uses prefetch directives to prefetch data in a matrix multiplication inner
loop where a row of one source matrix has been gathered into a contiguous vector.
real*8 a(m,n), b(n,p), c(m,p), arow(n)
...
do j = 1, p
!$mem prefetch arow(1),b(1,j)
!$mem prefetch arow(5),b(5,j)
!$mem prefetch arow(9),b(9,j)
do k = 1, n, 4
!$mem prefetch arow(k+12),b(k+12,j)
c(i,j) = c(i,j) + arow(k) * b(k,j)
c(i,j) = c(i,j) + arow(k+1) * b(k+1,j)
c(i,j) = c(i,j) + arow(k+2) * b(k+2,j)
c(i,j) = c(i,j) + arow(k+3) * b(k+3,j)
enddo
enddo
This pattern of prefetch directives the compiler emits prefetch instructions whereby
elements of arow and b are fetched into the data cache starting four iterations prior to
first use. By varying the prefetch distance in this way, it is sometimes possible to reduce
the effects of main memory latency and improve performance.
9.6.4. Prefetch Pragma Syntax in C/C++
The syntax of a prefetch pragma is as follows:
#pragma mem prefetch <var1>[,<var2>[,...]]
where <varn> is any valid variable, member, or array element reference.
9.6.5. Sample Usage of Prefetch Pragma
Prefetch Pragma in C
This example uses the prefetch pragma to prefetch data from the source vector x for
eight iterations beyond the current iteration.
for (i=0; i<n; i++) {
#pragma mem prefetch x[i+8]
y[i] = y[i] + a*x[i];
}
9.7. !$PRAGMA C
When programs are compiled using one of the PGI Fortran compilers on Linux, Win64,
and macOS systems, an underscore is appended to Fortran global names, including
names of functions, subroutines, and common blocks. This mechanism distinguishes
Fortran name space from C/C++ name space.
User's Guide for x86-64 CPUs
Version 2017 | 118
Using Directives and Pragmas
You can use !$PRAGMA C in the Fortran program to call a C/C++ function from Fortran.
The statement would look similar to this:
!$PRAGMA C(name[,name]...)
This statement directs the compiler to recognize the routine 'name' as a C function,
thus preventing the Fortran compiler from appending an underscore to the routine
name.
9.8. IGNORE_TKR Directive
This directive indicates to the compiler to ignore the type, kind, and/or rank (/TKR/)
of the specified dummy arguments in an interface of a procedure. The compiler also
ignores the type, kind, and/or rank of the actual arguments when checking all the
specifics in a generic call for ambiguities.
9.8.1. IGNORE_TKR Directive Syntax
The syntax for the IGNORE_TKR directive is this:
!DIR$ 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.
9.8.2. 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 assumedshape arrays.
User's Guide for x86-64 CPUs
Version 2017 | 119
Using Directives and Pragmas
9.8.3. 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 24 indicates which rules are ignored for which dummy arguments in the
preceding sample subroutine fragment:
Table 24 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.
9.9. !DEC$ Directives
PGI Fortran compilers for Microsoft Windows support several de-facto standard Fortran
directives that help with inter-language calling and importing and exporting routines to
and from DLLs.
9.9.1. !DEC$ Directive Syntax
These directives all take the form:
!DEC$ directive
9.9.2. Format Requirements
You must follow the following format requirements for the directive to be recognized in
your program:
‣
‣
‣
‣
‣
The directive must begin in column 1 when the file is fixed format or compiled with
-Mfixed.
The directive prefix !DEC$ requires a space between the prefix and the directive
keyword, such as ATTRIBUTES.
The ! must begin the prefix when compiling Fortran 90/95 free-form format.
The characters C or * can be used in place of ! in either form of the prefix when
compiling F77-style fixed-form format.
The directives are completely case insensitive.
User's Guide for x86-64 CPUs
Version 2017 | 120
Using Directives and Pragmas
9.9.3. Summary Table
The following table summarizes the supported !DEC$ directives. For a complete
description of each directive, refer to the ‘!DEC$ Directives’ section of the ‘Directives and
Pragmas Reference’ section in the PGI Compiler’s Reference Guide.
Table 25 !DEC$ Directives Summary Table
Directive
Functionality
ALIAS
Specifies an alternative name with which to resolve a routine.
ATTRIBUTES
Lets you specify properties for data objects and procedures.
DECORATE
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. This directive has no effect if ALIAS is not
specified.
DISTRIBUTE
Tells the compiler at what point within a loop to split into two loops.
User's Guide for x86-64 CPUs
Version 2017 | 121
Chapter 10.
CREATING AND USING LIBRARIES
A library is a collection of functions or subprograms that are grouped for reference and
ease of linking. This section discusses issues related to PGI-supplied compiler libraries.
Specifically, it addresses the use of C/C++ builtin functions in place of the corresponding
libc routines, creation of dynamically linked libraries, known as shared objects or shared
libraries, and math libraries.
This section does not duplicate material related to using libraries for inlining,
described in Creating an Inline Library or information related to runtime library
routines available to OpenMP programmers, described in Runtime Library Routines.
PGI provides libraries that export C interfaces by using Fortran modules. On Windows,
PGI also provides additions to the supported library functionality for runtime functions
included in DFLIB.
This section has examples that include the following options related to creating and
using libraries.
-Bdynamic
-def<file>
-implib <file>
-Mmakeimplib
-c
-fpic
-Mmakedll
-shared
-Bstatic
-dynamiclib
-l
-o
10.1. Using builtin Math Functions in C/C++
The name of the math header file is math.h. Include the math header file in all of your
source files that use a math library routine as in the following example, which calculates
the inverse cosine of 3.5.
#include <math.h>
#include<stdio.h>
#define PI 3.1415926535
void main()
{
double x, y;
x = PI/3.0;
y = acos(0.5);
printf(‘%f %f\n’,x,y);
}
User's Guide for x86-64 CPUs
Version 2017 | 122
Creating and Using Libraries
Including math.h causes PGI C and C++ to use builtin functions, which are much more
efficient than library calls. In particular, if you include math.h, the following intrinsics
calls are processed using builtins:
abs
acosf
asinf
atan
atan2
atan2f
atanf
cos
cosf
exp
expf
fabs
fabsf
fmax
fmaxf
fmin
fminf
log
log10
log10f
logf
pow
powf
sin
sinf
sqrt
sqrtf
tan
tanf
10.2. Using System Library Routines
Release 17.5 of the PGI runtime libraries makes use of Linux system libraries to
implement, for example, OpenMP and Fortran I/O. The PGI runtime libraries make use
of several additional system library routines.
On 64-bit Linux systems, the system library routines that PGI supports include these:
aio_error
aio_write
pthread_mutex_init
aio_read
calloc
pthread_mutex_lock
aio_return
getrlimit
pthread_mutex_unlock
aio_suspend
pthread_attr_init
setrlimit
sleep
10.3. Creating and Using Shared Object Files on
Linux
All of the PGI Fortran, C, and C++ compilers support creation of shared object files.
Unlike statically-linked object and library files, shared object files link and resolve
references with an executable at runtime via a dynamic linker supplied with your
operating system. The PGI compilers must generate position independent code to
support creation of shared objects by the linker. However, this is not the default. You
must create object files with position independent code and shared object files that will
include them.
10.3.1. Procedure to create a use a shared object file
The following steps describe how to create and use a shared object file.
1. Create an object file with position independent code.
To do this, compile your code with the appropriate PGI compiler using the -fpic
option, or one of the equivalent options, such as -fPIC, -Kpic, and -KPIC, which
are supported for compatibility with other systems. For example, use the following
command to create an object file with position independent code using pgfortran:
% pgfortran -c -fpic tobeshared.f
2. Produce a shared object file.
User's Guide for x86-64 CPUs
Version 2017 | 123
Creating and Using Libraries
To do this, use the appropriate PGI compiler to invoke the linker supplied with your
system. It is customary to name such files using a .so filename extension. On Linux,
you do this by passing the -shared option to the linker:
% pgfortran -shared -o tobeshared.so tobeshared.o
Compilation and generation of the shared object can be performed in one step
using both the -fpic option and the appropriate option for generation of a
shared object file.
3. Use a shared object file.
To do this, use the appropriate PGI compiler to compile and link the program which
will reference functions or subroutines in the shared object file, and list the shared
object on the link line, as shown here:
% pgfortran -o myprog myprog.f tobeshared.so
4. Make the executable available.
You now have an executable myprog which does not include any code from functions
or subroutines in tobeshared.so, but which can be executed and dynamically
linked to that code. By default, when the program is linked to produce myprog, no
assumptions are made on the location of tobeshared.so. Therefore, for myprog to
execute correctly, you must initialize the environment variable LD_LIBRARY_PATH
to include the directory containing tobeshared.so. If LD_LIBRARY_PATH is
already initialized, it is important not to overwrite its contents. If you have placed
tobeshared.so in directory /home/myusername/bin, you can initialize
LD_LIBRARY_PATH to include that directory and preserve its existing contents, as
shown in the following:
% setenv LD_LIBRARY_PATH "$LD_LIBRARY_PATH":/home/myusername/bin
If you know that tobeshared.so always resides in a specific directory, you can
create the executable myprog in a form that assumes this directory by using the -R
link-time option. For example, you can link as follows:
% pgfortran -o myprog myprof.f tobeshared.so -R/home/myusername/bin
As with the -L option, there is no space between -R and the directory name. If
the -R option is used, it is not necessary to initialize LD_LIBRARY_PATH.
In the previous example, the dynamic linker always looks in /home/myusername/
bin to resolve references to tobeshared.so. By default, if the LD_LIBRARY_PATH
environment variable is not set, the linker only searches /usr/lib and /lib for
shared objects.
10.3.2. ldd Command
The ldd command is a useful tool when working with shared object files and
executables that reference them. When applied to an executable, as shown in the
following example, ldd lists all shared object files referenced in the executable along
with the pathname of the directory from which they will be extracted.
% ldd myprog
User's Guide for x86-64 CPUs
Version 2017 | 124
Creating and Using Libraries
If the pathname is not hard-coded using the-R option, and if LD_LIBRARY_PATH is
not initialized, the pathname is listed as "not found". For more information on ldd, its
options and usage, see the online man page for ldd.
10.4. Creating and Using Dynamic Libraries on
Apple macOX
PGI compilers for macOS do not support static linking on user executables. Apple
only ships dynamic versions of its system libraries – not static versions. You can create
static libraries; however, you cannot create 100% static executables.
10.5. PGI Runtime Libraries on Windows
Both statically- and dynamically-linked library (DLL) versions are available with the PGI
runtime libraries on Windows. The static libraries are used by default.
‣
‣
You can use the dynamically-linked version of the runtime by specifying
-Bdynamic at both compile and link time.
You can explicitly specify static linking, the default, by using -Bstatic at compile
and link time.
For details on why you might choose one type of linking over another type, refer to
Creating and Using Dynamic-Link Libraries on Windows.
10.6. Creating and Using Static Libraries on
Windows
The Microsoft Library Manager (LIB.EXE) is the tool that is typically used to create
and manage a static library of object files on Windows. LIB is provided with the PGI
compilers as part of the Microsoft Open Tools. Refer to www.msdn2.com for a complete
LIB reference – search for LIB.EXE. For a list of available options, invoke LIB with the
/? switch.
For compatibility with legacy makefiles, PGI provides a wrapper for LIB and LINK
called ar. This version of ar is compatible with Windows and object-file formats.
PGI also provides ranlib as a placeholder for legacy makefile support.
10.6.1. ar command
The ar command is a legacy archive wrapper that interprets legacy ar command line
options and translates these to LINK/LIB options. You can use it to create libraries of
object files.
User's Guide for x86-64 CPUs
Version 2017 | 125
Creating and Using Libraries
Syntax
The syntax for the ar command is this:
ar [options] [archive] [object file].
Where:
‣
‣
‣
‣
The first argument must be a command line switch, and the leading dash on the first
option is optional.
The single character options, such as -d and -v, may be combined into one option,
such as -dv.
Thus, ar dv, ar -dv, and ar -d -v all mean the same thing.
The first non-switch argument must be the library name.
Exactly one of -d, -r, -t, or -x must appear on the command line.
Options
The options available for the ar command are these:
-c
-d
-r
-t
-v
-x
This switch is for compatibility; it is ignored.
Deletes the named object files from the library.
Replaces in or adds the named object files to the library.
Writes a table of contents of the library to standard out.
Writes a verbose file-by-file description of the making of the new library to standard
out.
Extracts the named files by copying them into the current directory.
10.6.2. ranlib command
The ranlib command is a wrapper that allows use of legacy scripts and makefiles that
use the ranlib command. The command actually does nothing; it merely exists for
compatibility.
Syntax
The syntax for the ranlib command is this:
ranlib [options] [archive]
User's Guide for x86-64 CPUs
Version 2017 | 126
Creating and Using Libraries
Options
The options available for the ar command are these:
-help
Short help information is printed out.
-V
Version information is printed out.
10.7. Creating and Using Dynamic-Link Libraries
on Windows
There are several differences between static- and dynamic-link libraries on Windows.
Libraries of either type are used when resolving external references for linking an
executable, but the process differs for each type of library. When linking with a static
library, the code needed from the library is incorporated into the executable. When
linking with a DLL, external references are resolved using the DLL's import library, not
the DLL itself. The code in the DLL associated with the external references does not
become a part of the executable. The DLL is loaded when the executable that needs it is
run. For the DLL to be loaded in this manner, the DLL must be in your path.
Static libraries and DLLs also handle global data differently. Global data in static
libraries is automatically accessible to other objects linked into an executable. Global
data in a DLL can only be accessed from outside the DLL if the DLL exports the data and
the image that uses the data imports it.
To access global data, the C compilers support the Microsoft storage class extensions:
__declspec(dllimport) and __declspec(dllexport). These extensions may
appear as storage class modifiers and enable functions and data to be imported and
exported:
extern int __declspec(dllimport) intfunc();
float __declspec(dllexport) fdata;
The PGI Fortran compilers support the DEC$ ATTRIBUTES extensions DLLIMPORT and
DLLEXPORT:
cDEC$ ATTRIBUTES DLLEXPORT :: object [,object] ...
cDEC$ ATTRIBUTES DLLIMPORT :: object [,object] ...
Here c is one of C, c, !, or *. object is the name of the subprogram or common block
that is exported or imported. Further, common block names are enclosed within slashes
(/), as shown here:
cDEC$ ATTRIBUTES DLLIMPORT :: intfunc
!DEC$ ATTRIBUTES DLLEXPORT :: /fdata/
For more information on these extensions, refer to !DEC$ Directives.
The examples in this section further illustrate the use of these extensions.
To create a DLL from the command line, use the -Mmakedll option.
The following switches apply to making and using DLLs with the PGI compilers:
User's Guide for x86-64 CPUs
Version 2017 | 127
Creating and Using Libraries
-Bdynamic
Compile for and link to the DLL version of the PGI runtime libraries. This flag
is required when linking with any DLL built by the PGI compilers. This flag
corresponds to the /MD flag used by Microsoft’s cl compilers.
When you use the PGI compiler flag -Bdynamic to create an executable that links to
the DLL form of the runtime, the executable built is smaller than one built without
-Bdynamic. The PGI runtime DLLs, however, must be available on the system where
the executable is run. You must use the -Bdynamic flag when linking an executable
against a DLL built by the PGI compilers.
-Bstatic
Compile for and link to the static version of the PGI runtime libraries. This flag
corresponds to the /MT flag used by Microsoft’s cl compilers.
On Windows, you must use-Bstatic for both compiling and linking.
-Mmakedll
Generate a dynamic-link library or DLL. Implies -Bdynamic.
-Mmakeimplib
Generate an import library without generating a DLL. Use this flag when you want
to generate an import library for a DLL but are not yet ready to build the DLL itself.
This situation might arise, for example, when building DLLs with mutual imports, as
shown in Build DLLs Containing Mutual Imports: Fortran.
-o <file>
Passed to the linker. Name the DLL or import library <file>.
-def <file>
When used with -Mmakedll, this flag is passed to the linker and a .def file named
<file> is generated for the DLL. The .def file contains the symbols exported by the
DLL. Generating a .def file is not required when building a DLL but can be a useful
debugging tool if the DLL does not contain the symbols that you expect it to contain.
When used with -Mmakeimplib, this flag is passed to lib which requires a .def
file to create an import library. The .def file can be empty if the list of symbols to
export are passed to lib on the command line or explicitly marked as DLLEXPORT in
the source code.
-implib <file>
Passed to the colinker. Generate an import library named <file> for the DLL. A DLL’s
import library is the interface used when linking an executable that depends on
routines in a DLL.
To use the PGI compilers to create an executable that links to the DLL form of the
runtime, use the compiler flag -Bdynamic. The executable built will be smaller than
one built without -Bdynamic; the PGI runtime DLLs, 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 DLL built by the PGI compilers.
The following examples outline how to use -Bdynamic, -Mmakedll and
-Mmakeimplib to build and use DLLs with the PGI compilers.
User's Guide for x86-64 CPUs
Version 2017 | 128
Creating and Using Libraries
10.7.1. Build a DLL: Fortran
This example builds a DLL from a single source file, object1.f, which exports data
and a subroutine using DLLEXPORT. The source file, prog1.f, uses DLLIMPORT to
import the data and subroutine from the DLL.
object1.f
subroutine sub1(i)
!DEC$ ATTRIBUTES DLLEXPORT :: sub1
integer i
common /acommon/ adata
integer adata
!DEC$ ATTRIBUTES DLLEXPORT :: /acommon/
print *, "sub1 adata", adata
print *, "sub1 i ", i
adata = i
end
prog1.f
program prog1
common /acommon/ adata
integer adata
external sub1
!DEC$ ATTRIBUTES DLLIMPORT:: sub1, /acommon/
adata = 11
call sub1(12)
print *, "main adata", adata
end
1. 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
2. Compile the main program:
% pgfortran -Bdynamic -o prog1 prog1.f -defaultlib:obj1
The -Bdynamic and -Mmakedll switches cause the compiler to link against the
PGI runtime DLLs instead of the PGI runtime static libraries. The -Bdynamic switch
is required when linking against any PGI-compiled DLL, such as obj1.dll. The defaultlib: switch specifies that obj1.lib, the DLL’s import library, should be used
to resolve imports.
3. Ensure that obj1.dll is in your path, then run the executable prog1 to determine if
the DLL was successfully created and linked:
% prog1
sub1 adata 11
sub1 i 12
main adata 12
Should you wish to change obj1.dll without changing the subroutine or function
interfaces, no rebuilding of prog1 is necessary. Just recreate obj1.dll and the new
obj1.dll is loaded at runtime.
User's Guide for x86-64 CPUs
Version 2017 | 129
Creating and Using Libraries
10.7.2. Build a DLL: C
In this example, we build a DLL out of a single source file, object2.c, which exports
data and a subroutine using __declspec(dllexport). The main source file, prog2.c,
uses __declspec(dllimport) to import the data and subroutine from the DLL.
object2.c
int __declspec(dllexport) data;
void __declspec(dllexport)
func2(int i)
{
printf("func2: data == %d\n", data);
printf("func2: i == %d\n", i);
data = i;
}
prog2.c
int __declspec(dllimport) data;
void __declspec(dllimport) func2(int);
int
main()
{
data = 11;
func2(12);
printf("main: data == %d\n",data);
return 0;
}
1. Step 1: Create the DLL obj2.dll and its import library obj2.lib using the
following series of commands:
% pgcc -Bdynamic -c object2.c
% pgcc -Mmakedll object2.obj -o obj2.dll
2. Step 2: Compile the main program:
% pgcc -Bdynamic -o prog2 prog2.c -defaultlib:obj2
The -Bdynamic switch causes the compiler to link against the PGI runtime DLLs
instead of the PGI runtime static libraries. The -Bdynamic switch is required when
linking against any PGI-compiled DLL such as obj2.dll. The -defaultlib:
switch specifies that obj2.lib, the DLL’s import library, should be used to resolve
the imported data and subroutine in prog2.c.
3. Step 3: Ensure that obj2.dll is in your path, then run the executable prog2 to
determine if the DLL was successfully created and linked:
% prog2
func2: data == 11
func2: i == 12
main: data == 12
Should you wish to change obj2.dll without changing the subroutine or function
interfaces, no rebuilding of prog2 is necessary. Just recreate obj2.dll and the new
obj2.dll is loaded at runtime.
User's Guide for x86-64 CPUs
Version 2017 | 130
Creating and Using Libraries
10.7.3. Build DLLs Containing Circular Mutual Imports: C
In this example we build two DLLs, obj3.dll and obj4.dll, each of which imports
a routine that is exported by the other. To link the first DLL, the import library for
the second DLL must be available. Usually an import library is created when a DLL
is linked. In this case, however, the second DLL cannot be linked without the import
library for the first DLL. When such circular imports exist, an import library for one of
the DLLs must be created in a separate step without creating the DLL. The PGI drivers
call the Microsoft lib tool to create import libraries in this situation. Once the DLLs are
built, we can use them to build the main program.
/* object3.c */
void __declspec(dllimport) func_4b(void);
void __declspec(dllexport)
func_3a(void)
{
printf("func_3a, calling a routine in obj4.dll\n");
func_4b();
}
void __declspec(dllexport)
func_3b(void)
{
printf("func_3b\n");
}
/* object4.c */
void __declspec(dllimport) func_3b(void);
void __declspec(dllexport)
func_4a(void)
{
printf("func_4a, calling a routine in obj3.dll\n");
func_3b();
}
void __declspec(dllexport)
func_4b(void)
{
printf("func_4b\n");
}
/* prog3.c */
void __declspec(dllimport) func_3a(void);
void __declspec(dllimport) func_4a(void);
int
main()
{
func_3a();
func_4a();
return 0;
}
1. Step 1: Use -Mmakeimplib with the PGI compilers to build an import library for the
first DLL without building the DLL itself.
% pgcc -Bdynamic -c object3.c
% pgcc -Mmakeimplib -o obj3.lib object3.obj
Tip The -def=<deffile> option can also be used with -Mmakeimplib.
Use a .def file when you need to export additional symbols from the DLL. A
User's Guide for x86-64 CPUs
Version 2017 | 131
Creating and Using Libraries
.def file is not needed in this example because all symbols are exported using
__declspec(dllexport).
2. Step 2: Use the import library, obj3.lib, created in Step 1, to link the second DLL.
% pgcc -Bdynamic -c object4.c
% pgcc -Mmakedll -o obj4.dll object4.obj -defaultlib:obj3
3. Step 3: Use the import library, obj4.lib, created in Step 2, to link the first DLL.
% pgcc -Mmakedll -o obj3.dll object3.obj -defaultlib:obj4
4. Step 4: Compile the main program and link against the import libraries for the two
DLLs
% pgcc -Bdynamic prog3.c -o prog3 -defaultlib:obj3 -defaultlib:obj4
5. Step 5: Execute prog3.exe to ensure that the DLLs were create properly.
% prog3
func_3a, calling a routine in obj4.dll
func_4b
func_4a, calling a routine in obj3.dll
func_3b
10.7.4. Build DLLs Containing Mutual Imports: Fortran
In this example we build two DLLs when each DLL is dependent on the other, and use
them to build the main program.
In the following source files, object2.f95 makes calls to routines defined in
object3.f95, and vice versa. This situation of mutual imports requires two steps to
build each DLL.
To link the first DLL, the import library for the second DLL must be available. Usually
an import library is created when a DLL is linked. In this case, however, the second
DLL cannot be linked without the import library for the first DLL. When such circular
imports exist, an import library for one of the DLLs must be created in a separate step
without creating the DLL. The PGI drivers call the Microsoft lib tool to create import
libraries in this situation.
Once the DLLs are built, we can use them to build the main program.
object2.f95
subroutine func_2a
external func_3b
!DEC$ ATTRIBUTES DLLEXPORT :: func_2a
!DEC$ ATTRIBUTES DLLIMPORT :: func_3b
print*,"func_2a, calling a routine in obj3.dll"
call func_3b() end subroutine
subroutine func_2b
!DEC$ ATTRIBUTES DLLEXPORT :: func_2b
print*,"func_2b"
end subroutine
User's Guide for x86-64 CPUs
Version 2017 | 132
Creating and Using Libraries
object3.f95
subroutine func_3a
external func_2b
!DEC$ ATTRIBUTES DLLEXPORT :: func_3a
!DEC$ ATTRIBUTES DLLIMPORT :: func_2b
print*,"func_3a, calling a routine in obj2.dll"
call func_2b() end subroutine
subroutine func_3b
!DEC$ ATTRIBUTES DLLEXPORT :: func_3b
print*,"func_3b"
end subroutine
prog2.f95
program prog2
external func_2a
external func_3a
!DEC$ ATTRIBUTES DLLIMPORT :: func_2a
!DEC$ ATTRIBUTES DLLIMPORT :: func_3a
call func_2a()
call func_3a()
end program
1. Use -Mmakeimplib with the PGI compilers to build an import library for the first
DLL without building the DLL itself.
% pgfortran -Bdynamic -c object2.f95
% pgfortran -Mmakeimplib -o obj2.lib object2.obj
Tip The -def=<deffile> option can also be used with -Mmakeimplib.
Use a .def file when you need to export additional symbols from the DLL. A
.def file is not needed in this example because all symbols are exported using
DLLEXPORT.
2. Use the import library, obj2.lib, created in Step 1, to link the second DLL.
% pgfortran -Bdynamic -c object3.f95
% pgfortran -Mmakedll -o obj3.dll object3.obj -defaultlib:obj2
3. Use the import library, obj3.lib, created in Step 2, to link the first DLL.
% pgfortran -Mmakedll -o obj2.dll object2.obj -defaultlib:obj3
4. Compile the main program and link against the import libraries for the two DLLs.
% pgfortran -Bdynamic prog2.f95 -o prog2 -defaultlib:obj2 -defaultlib:obj3
5. Execute prog2 to ensure that the DLLs were create properly.
% prog2
func_2a, calling a routine in obj3.dll
func_3b
func_3a, calling a routine in obj2.dll
func_2b
10.7.5. Import a Fortran module from a DLL
In this example we import a Fortran module from a DLL. We use the source file
defmod.f90 to create a DLL containing a Fortran module. We then use the source
file use_mod.f90 to build a program that imports and uses the Fortran module from
defmod.f90.
User's Guide for x86-64 CPUs
Version 2017 | 133
Creating and Using Libraries
defmod.f90
module testm
type a_type
integer :: an_int
end type a_type
type(a_type) :: a, b
!DEC$ ATTRIBUTES DLLEXPORT :: a,b
contains
subroutine print_a
!DEC$ ATTRIBUTES DLLEXPORT :: print_a
write(*,*) a%an_int
end subroutine
subroutine print_b
!DEC$ ATTRIBUTES DLLEXPORT :: print_b
write(*,*) b%an_int
end subroutine
end module
usemod.f90
use testm
a%an_int = 1
b%an_int = 2
call print_a
call print_b
end
1. Create the DLL.
% pgf90 -Mmakedll -o defmod.dll defmod.f90
Creating library defmod.lib and object defmod.exp
2. Create the exe and link against the import library for the imported DLL.
% pgf90 -Bdynamic -o usemod usemod.f90 -defaultlib:defmod.lib
3. Run the exe to ensure that the module was imported from the DLL properly.
% usemod
1
2
10.8. Using LIB3F
The PGI Fortran compilers include complete support for the de facto standard LIB3F
library routines on both Linux and Windows operating systems. See the PGI Fortran
Language Reference manual for a complete list of available routines in the PGI
implementation of LIB3F.
10.9. LAPACK, BLAS and FFTs
All PGI products now include a BLAS and LAPACK library based on the customized
OpenBLAS project source and built with PGI compilers. The LAPACK library is
called liblapack.a or on Windows, liblapack.lib. The BLAS library is called
libblas.a or on Windows, libblas.lib. These libraries are installed to $PGI/
<target>/lib, where <target> is replaced with the appropriate target name
(linux86-64, osx86-64, or win64).
User's Guide for x86-64 CPUs
Version 2017 | 134
Creating and Using Libraries
To use these libraries, simply link them in using the -l option when linking your main
program:
% pgfortran myprog.f -llapack -lblas
10.10. Linking with ScaLAPACK
The ScaLAPACK libraries are automatically installed with each MPI library version
which accompanies a PGI installation. You can link with the ScaLAPACK libraries by
specifying -Mscalapack on any of the PGI compiler command lines. For example:
% mpif90 myprog.f -Mscalapack
A pre-built version of the BLAS library is automatically added when the -Mscalapack
switch is specified. If you wish to use a different BLAS library, and still use the
-Mscalapack switch, then you can list the set of libraries explicitly on your link line.
Alternately, you can copy your BLAS library into $PGI/linux86-64/17.5/lib/
libblas.a.
10.11. The C++ Standard Template Library
On Linux and macOS, the GNU-compatible pgc++ compiler uses the GNU g++ header
files and Standard Template Library (STL) directly. The versions used are dependent
on the version of the GNU compilers installed on your system, or specified when
makelocalrc was run during installation of the PGI compilers.
User's Guide for x86-64 CPUs
Version 2017 | 135
Chapter 11.
USING ENVIRONMENT VARIABLES
Environment variables allow you to set and pass information that can alter the default
behavior of the PGI compilers and the executables which they generate. This section
includes explanations of the environment variables specific to PGI compilers. Other
environment variables are referenced and documented in other sections of this User’s
Guide, the accompanying Reference Manual, the PGI Debugger User’s Guide and the
PGI Profiler User’s Guide.
‣
‣
You use OpenMP environment variables to control the behavior of OpenMP
programs. For consistency related to the OpenMP environment, the details of
the OpenMP-related environment variables are included in OpenMP section:
Environment Variables.
You can use environment variables to control the behavior of the PGI debugger or
PGI profiler. For a description of environment variables that affect these tools, refer
to the PGI Debugger User’s Guide and PGI Profiler User’s Guide, respectively.
11.1. Setting Environment Variables
Before we look at the environment variables that you might use with the PGI compilers
and tools, let’s take a look at how to set environment variables. To illustrate how to set
these variables in various environments, let’s look at how a user might initialize the shell
environment prior to using the PGI compilers and tools.
11.1.1. Setting Environment Variables on Linux
Let’s assume that you want access to the PGI products when you log in. Let’s further
assume that you installed the PGI compilers in /opt/pgi and that the license file is
in /opt/pgi/license.dat. For access at startup, you can add the following lines to
your startup file.
In csh, use these commands:
%
%
%
%
setenv PGI /opt/pgi
setenv MANPATH "$MANPATH":$PGI/linux86-64/17.5/man
setenv LM_LICENSE_FILE $PGI/license.dat
set path = ($PGI/linux86-64/17.5/bin $path)
User's Guide for x86-64 CPUs
Version 2017 | 136
Using Environment Variables
In bash, sh, zsh, or ksh, use these commands:
$
$
$
$
PGI=/opt/pgi; export PGI
MANPATH=$MANPATH:$PGI/linux86-64/17.5/man; export MANPATH
LM_LICENSE_FILE=$PGI/license.dat; export LM_LICENSE_FILE
PATH=$PGI/linux86-64/17.5/bin:$PATH; export PATH
11.1.2. Setting Environment Variables on Windows
In Windows, when you access PGI 17.5 (for example, using Start | ALL Programs | PGI |
Command Shells 17.5), you have options that PGI provides for setting your environment
variables – either the DOS command environment or the Cygwin Bash environment.
When you open either of these shells available to you, the default environment variables
are already set and available to you.
You may want to use other environment variables, such as the OpenMP ones. This
section explains how to do that.
Suppose that your home directory is C:\tmp. The following examples show how you
might set the temporary directory to your home directory, and then verify that it is set.
Command prompt:
Once you have launched a command shell, enter the following:
DOS> set TMPDIR=C:\tmp
DOS> echo %TMPDIR%
C:\tmp
DOS>
Cygwin Bash prompt:
From the PGI 17.5 menu item, select PGI option and at the Cygwin Bash prompt, enter
the following
PGI$ export TMPDIR=C:\tmp
PGI$ echo $TMPDIR
C:\tmp
PGI$
11.1.3. Setting Environment Variables on Apple macOS
Let’s assume that you want access to the PGI products when you log in. Let’s further
assume that you installed the PGI compilers in /opt/pgi and that the license file is
in /opt/pgi/license.dat. For access at startup, you can add the following lines to
your startup file.
For x86-64 osx86-64 in a csh:
% set path = (/opt/pgi/osx86-64/17.5/bin $path)
For x86-64 osx86-64 in a bash, sh, zsh, or ksh:
$ PATH=/opt/pgi/osx86-64/17.5/bin:$PATH; export PATH
User's Guide for x86-64 CPUs
Version 2017 | 137
Using Environment Variables
11.2. PGI-Related Environment Variables
For easy reference, the following table provides a quick listing of some OpenMP and
all PGI compiler-related environment variables. This section provides more detailed
descriptions of the environment variables specific to PGI compilers and the executables
they generate. For information specific to OpenMP environment variables, refer to Table
14 and to the complete descriptions in ‘OpenMP Environment Variables’ in the PGI
Compiler’s Reference Manual.
Table 26 PGI-Related Environment Variable Summary
Environment Variable
Description
FLEXLM_BATCH
(Windows only) When set to 1, prevents interactive pop-ups from
appearing by sending all licensing errors and warnings to standard
out rather than to a pop-up window.
FORTRANOPT
Allows the user to specify that the PGI Fortran compilers user VAX I/
O conventions.
GMON_OUT_PREFIX
Specifies the name of the output file for programs that are compiled
and linked with the -pg option.
LD_LIBRARY_PATH
Specifies a colon-separated set of directories where libraries should
first be searched, prior to searching the standard set of directories.
LM_LICENSE_FILE
Specifies the full path of the license file that is required for running
the PGI software. On Windows, LM_LICENSE_FILE does not
need to be set.
MANPATH
Sets the directories that are searched for manual pages associated
with the command that the user types. On macOS, MANPATH does
not need to be set.
MPSTKZ
Increases the size of the stacks used by threads executing in parallel
regions. The value should be an integer <n> concatenated with M or
m to specify stack sizes of n megabytes.
MP_BIND
Specifies whether to bind processes or threads executing in a
parallel region to a physical processor.
MP_BLIST
When MP_BIND is yes, this variable specifically defines the
thread-CPU relationship, overriding the default values.
MP_SPIN
Specifies the number of times to check a semaphore before calling
sched_yield() (on Linux or macOS) or _sleep() (on Windows).
MP_WARN
Allows you to eliminate certain default warning messages.
NCPUS
Sets the number of processes or threads used in parallel regions.
NCPUS_MAX
Limits the maximum number of processors or threads that can be
used in a parallel region.
NO_STOP_MESSAGE
If used, the execution of a plain STOP statement does not produce
the message FORTRAN STOP.
OMP_DYNAMIC
User's Guide for x86-64 CPUs
Currently has no effect. Enables (TRUE) or disables (FALSE) the
dynamic adjustment of the number of threads. The default is FALSE.
Version 2017 | 138
Using Environment Variables
Environment Variable
Description
OMP_MAX_ACTIVE_LEVELS
Specifies the maximum number of nested parallel regions.
OMP_NESTED
Currently has no effect. Enables (TRUE) or disables (FALSE)
nested parallelism. The default is FALSE.
OMP_NUM_THREADS
Specifies the number of threads to use during execution of parallel
regions. Default is 1.
OMP_SCHEDULE
Specifies the type of iteration scheduling and, optionally, the chunk
size to use for omp for and omp parallel for loops that include the
runtime schedule clause. The default is STATIC with chunk size = 1.
OMP_STACKSIZE
Overrides the default stack size for a newly created thread.
OMP_WAIT_POLICY
Sets the behavior of idle threads, defining whether they spin or
sleep when idle. The values are ACTIVE and PASSIVE. The
default is ACTIVE.
PATH
Determines which locations are searched for commands the user
may type.
PGI
Specifies, at compile-time, the root directory where the PGI
compilers and tools are installed.
PGI_CONTINUE
If set, when a program compiled with-Mchkfpstk is executed,
the stack is automatically cleaned up and execution then continues.
PGI_OBJSUFFIX
(Windows only) Allows you to control the suffix on generated object
files.
PGI_STACK_USAGE
(Windows only) Allows you to explicitly set stack properties for your
program.
PGI_TERM
Controls the stack traceback and just-in-time debugging
functionality.
PGI_TERM_DEBUG
Overrides the default behavior when
PGROUPD_LICENSE_FILE
Specifies the location of the PGI license. This variable is set in the
registry on Windows machines, and is specific to PGI products. On
Windows, PGROUPD_LICENSE_FILE does not need to be set.
PWD
Allows you to display the current directory.
STATIC_RANDOM_SEED
Forces the seed returned by
TMP
Sets the directory to use for temporary files created during
execution of the PGI compilers and tools; interchangeable with
TMPDIR.
TMPDIR
Sets the directory to use for temporary files created during
execution of the PGI compilers and tools.
PGI_TERM is set to debug.
RANDOM_SEED to be constant.
11.3. PGI Environment Variables
You use the environment variables listed in Table 26 to alter the default behavior of the
PGI compilers and the executables which they generate. This section provides more
detailed descriptions about the variables in this table that are not OpenMP environment
variables.
User's Guide for x86-64 CPUs
Version 2017 | 139
Using Environment Variables
11.3.1. FLEXLM_BATCH
By default, on Windows the license server creates interactive pop-up messages to issue
warning and errors. You can use the environment variable FLEXLM_BATCH to prevent
interactive pop-up windows. To do this, set the environment variable FLEXLM_BATCH to
1.
The following csh example prevents interactive pop-up messages for licensing warnings
and errors:
% set FLEXLM_BATCH = 1;
11.3.2. FORTRANOPT
FORTRANOPT allows the user to adjust the behavior of the PGI Fortran compilers.
‣
‣
‣
If FORTRANOPT exists and contains the value vaxio, the record length in the open
statement is in units of 4-byte words, and the $ edit descriptor only has an effect for
lines beginning with a space or a plus sign (+).
If FORTRANOPT exists and contains the value format_relaxed, an I/O item
corresponding to a numerical edit descriptor (such as F, E, I, and so on) is not
required to be a type implied by the descriptor.
In a non-Windows environment, if FORTRANOPT exists and contains the value crif,
a sequential formatted or list-directed record is allowed to be terminated with the
character sequence \r\n (carriage return, newline). This approach is useful when
reading records from a file produced on a Window’s system.
The following example causes the PGI Fortran compilers to use VAX I/O conventions:
% setenv FORTRANOPT vaxio
11.3.3. GMON_OUT_PREFIX
GMON_OUT_PREFIX specifies the name of the output file for programs that are compiled
and linked with the -pg option. The default name is gmon.out.
If GMON_OUT_PREFIX is set, the name of the output file has GMON_OUT_PREFIX as a
prefix. Further, the suffix is the pid of the running process. The prefix and suffix are
separated by a dot. For example, if the output file is mygmon, then the full filename may
look something similar to this: mygmon.0012348567.
The following example causes the PGI Fortran compilers to use pgout as the output file
for programs compiled and linked with the -pg option.
% setenv GMON_OUT_PREFIX pgout
11.3.4. LD_LIBRARY_PATH
The LD_LIBRARY_PATH variable is a colon-separated set of directories specifying where
libraries should first be searched, prior to searching the standard set of directories. This
variable is useful when debugging a new library or using a nonstandard library for
special purposes.
User's Guide for x86-64 CPUs
Version 2017 | 140
Using Environment Variables
The following csh example adds the current directory to your LD_LIBRARY_PATH
variable.
% setenv LD_LIBRARY_PATH "$LD_LIBRARY_PATH":"./"
11.3.5. LM_LICENSE_FILE
The LM_LICENSE_FILE variable specifies the full path of the license file that is required
for running the PGI software.
For example, once the license file is in place, you can execute the following csh
commands to make the products you have purchased accessible and to initialize your
environment for use of FLEXlm. These commands assume that you use the default
installation directory: /opt/pgi
% setenv PGI /opt/pgi
% setenv LM_LICENSE_FILE "$LM_LICENSE_FILE":/opt/pgi/license.dat
To set the environment variable LM_LICENSE_FILE to the full path of the license key
file, do this:
1. Open the System Properties dialog: Start | Control Panel | System.
2. Select the Advanced tab.
3. Click the Environment Variables button.
‣
‣
If LM_LICENSE_FILE is not already an environment variable, create a new
system variable for it. Set its value to the full path, including the name of the
license key file, license.dat.
If LM_LICENSE_FILE already exists as an environment variable, append the
path to the license file to the variable’s current value using a semi-colon to
separate entries.
11.3.6. MANPATH
The MANPATH variable sets the directories that are searched for manual pages associated
with the commands that the user types. When using PGI products, it is important that
you set your PATH to include the location of the PGI products and then set the MANPATH
variable to include the man pages associated with the products. MANPATH is not required
for macOS.
The following csh example targets the linux86-64 version of the compilers and tool s and
allows the user access to the manual pages associated with them.
% set path = (/opt/pgi/linux86-64/17.5/bin $path
% setenv MANPATH "$MANPATH":/opt/pgi/linux86-64/17.5/man
11.3.7. MPSTKZ
MPSTKZ increases the size of the stacks used by threads executing in parallel regions.
You typically use this variable with programs that utilize large amounts of thread-local
storage in the form of private variables or local variables in functions or subroutines
called within parallel regions. The value should be an integer <n> concatenated with M or
m to specify stack sizes of n megabytes.
User's Guide for x86-64 CPUs
Version 2017 | 141
Using Environment Variables
For example, the following setting specifies a stack size of 8 megabytes.
% setenv MPSTKZ 8M
11.3.8. MP_BIND
You can set MP_BIND to yes or y to bind processes or threads executing in a parallel
region to physical processor. Set it to no or n to disable such binding. The default is
to not bind processes to processors. This variable is an execution-time environment
variable interpreted by the PGI runtime support libraries. It does not affect the behavior
of the PGI compilers in any way.
The MP_BIND environment variable is not supported on all platforms.
% setenv MP_BIND y
11.3.9. MP_BLIST
MP_BLIST allows you to specifically define the thread-CPU relationship.
This variable is only in effect when MP_BIND is yes.
While the MP_BIND variable binds processors or threads to a physical processor,
MP_BLIST allows you to specifically define which thread is associated with which
processor. The list defines the processor-thread relationship order, beginning with
thread 0. This list overrides the default binding.
For example, the following setting for MP_BLIST maps CPUs 3, 2, 1 and 0 to threads 0, 1,
2 and 3 respectively.
% setenv MP_BLIST=3,2,1,0
11.3.10. MP_SPIN
When a thread executing in a parallel region enters a barrier, it spins on a semaphore.
You can use MP_SPIN to specify the number of times it checks the semaphore before
calling sched_yield() (on Linux or macOS) or _sleep() (on Windows). These calls
cause the thread to be re-scheduled, allowing other processes to run. The default value is
1000000.
% setenv MP_SPIN 200
11.3.11. MP_WARN
MP_WARN allows you to eliminate certain default warning messages.
By default, a warning is printed to standard error if you execute an OpenMP or autoparallelized program with NCPUS or OMP_NUM_THREADS set to a value larger than the
number of physical processors in the system.
User's Guide for x86-64 CPUs
Version 2017 | 142
Using Environment Variables
For example, if you produce a parallelized executable a.out and execute as follows on a
system with only one processor, you get a warning message.
% setenv OMP_NUM_THREADS 2
% a.out
Warning: OMP_NUM_THREADS or NCPUS (2) greater than available cpus (1)
FORTRAN STOP
Setting MP_WARN to NO eliminates these warning messages.
11.3.12. NCPUS
You can use the NCPUS environment variable to set the number of processes or threads
used in parallel regions. The default is to use only one process or thread, which is
known as serial mode.
OMP_NUM_THREADS has the same functionality as NCPUS. For historical reasons,
PGI supports the environment variable NCPUS. If both OMP_NUM_THREADS and
NCPUS are set, the value of OMP_NUM_THREADS takes precedence.
Setting NCPUS to a value larger than the number of physical processors or cores in your
system can cause parallel programs to run very slowly.
11.3.13. NCPUS_MAX
You can use the NCPUS_MAX environment variable to limit the maximum number
of processes or threads used in a parallel program. Attempts to dynamically
set the number of processes or threads to a higher value, for example using
set_omp_num_threads(), will cause the number of processes or threads to be set at the
value of NCPUS_MAX rather than the value specified in the function call.
11.3.14. NO_STOP_MESSAGE
If the NO_STOP_MESSAGE variable exists, the execution of a plain STOP statement does
not produce the message FORTRAN STOP. The default behavior of the PGI Fortran
compilers is to issue this message.
11.3.15. PATH
The PATH variable sets the directories that are searched for commands that the user
types. When using PGI products, it is important that you set your PATH to include the
location of the PGI products.
You can also use this variable to specify that you want to use only the linux86 version of
the compilers and tools, or to target linux86 as the default.
The following csh example targets x64 linux86-64 version of the compilers and tools.
% set path = (/opt/pgi/linux86-64/17.5/bin $path)
User's Guide for x86-64 CPUs
Version 2017 | 143
Using Environment Variables
11.3.16. PGI
The PGI environment variable specifies the root directory where the PGI compilers and
tools are installed. This variable is recognized at compile-time. If it is not set, the default
value depends on your system as well as which compilers are installed:
‣
‣
‣
On Linux, the default value of this variable is /opt/pgi.
On Windows, the default value is C:\Program Files\PGI, where C represents
the system drive.
On macOS, the default value of this variable is /opt/pgi.
In most cases, if the PGI environment variable is not set, the PGI compilers and tools
dynamically determine the location of this root directory based on the instance of the
compiler or tool that was invoked. However, there are still some dependencies on the
PGI environment variable, and you can use it as a convenience when initializing your
environment for use of the PGI compilers and tools.
For example, assuming you use csh and want the 64-bit linux86-64 versions of the PGI
compilers and tools to be the default, you would use this syntax:
%
%
%
%
setenv PGI /opt/pgi
setenv MANPATH "$MANPATH":$PGI/linux86-64/17.5/man
setenv LM_LICENSE_FILE $PGI/license.dat
set path = ($PGI/linux86-64/17.5/bin $path)
11.3.17. PGI_CONTINUE
You set the PGI_CONTINUE variable to specify the actions to take before continuing
with execution. For example, if the PGI_CONTINUE environment variable is set and then
a program that is compiled with -Mchkfpstk is executed, the stack is automatically
cleaned up and execution then continues. If PGI_CONTINUE is set to verbose, the stack
is automatically cleaned up, a warning message is printed, and then execution continues.
There is a performance penalty associated with the stack cleanup.
11.3.18. PGI_OBJSUFFIX
You can set the PGI_OBJSUFFIX environment variable to generate object files that have
a specific suffix. For example, if you set PGI_OBJSUFFIX to .o, the object files have a
suffix of .o rather than .obj.
11.3.19. PGI_STACK_USAGE
(Windows only) The PGI_STACK_USAGE variable allows you to explicitly set stack
properties for your program. When the user compiles a program with the -Mchkstk
option and sets the PGI_STACK_USAGE environment variable to any value, the
program displays the stack space allocated and used after the program exits. You might
see something similar to the following message:
thread 0 stack: max 8180KB, used 48KB
User's Guide for x86-64 CPUs
Version 2017 | 144
Using Environment Variables
This message indicates that the program used 48KB of a 8180KB allocated stack. For
more information on the -Mchkstk option, refer to ‘-Mchkstk’ in the PGI Compiler’s
Reference Manual.
11.3.20. PGI_TERM
The PGI_TERM environment variable controls the stack traceback and just-in-time
debugging functionality. The runtime libraries use the value of PGI_TERM to determine
what action to take when a program abnormally terminates.
The value of PGI_TERM is a comma-separated list of options. The commands for setting
the environment variable follow.
‣
‣
‣
In csh:
% setenv PGI_TERM option[,option...]
In bash, sh, zsh, or ksh:
$ PGI_TERM=option[,option...]
$ export PGI_TERM
In the Windows Command Prompt:
C:\> set PGI_TERM=option[,option...]
Table 27 lists the supported values for option. Following the table is a complete
description of each option that indicates specifically how you might apply the option.
By default, all of these options are disabled.
Table 27 Supported PGI_TERM Values
[no]debug
Enables/disables just-in-time debugging (debugging invoked on error)
[no]trace
Enables/disables stack traceback on error
[no]signal
Enables/disables establishment of signal handlers for common signals that cause
program termination
[no]abort
Enables/disables calling the system termination routine abort()
[no]debug
This enables/disables just-in-time debugging. The default is nodebug.
When PGI_TERM is set to debug, the following command is invoked on error, unless
you use PGI_TERM_DEBUG to override this default.
pgdbg -text -attach <pid>
<pid> is the process ID of the process being debugged.
The PGI_TERM_DEBUG environment variable may be set to override the default
setting. For more information, refer to PGI_TERM_DEBUG.
[no]trace
This enables/disables stack traceback on error.
User's Guide for x86-64 CPUs
Version 2017 | 145
Using Environment Variables
[no]signal
This enables/disables establishing signal handlers for the most common signals that
cause program termination. The default is nosignal. Setting trace and debug
automatically enables signal. Specifically setting nosignal allows you to override
this behavior.
[no]abort
This enables/disables calling the system termination routine abort(). The default is
noabort. When noabort is in effect the process terminates by calling _exit(127).
On Linux, when abort is in effect, the abort routine creates a core file and exits with
code 127.
On Windows, when abort is in effect, the abort routine exits with the status of the
exception received. For example, if the program receives an access violation, abort() exits
with status 0xC0000005.
A few runtime errors just print an error message and call exit(127), regardless of the
status of PGI_TERM. These are mainly errors such as specifying an invalid environment
variable value where a traceback would not be useful.
If it appears that abort() does not generate core files on a Linux system, be sure to
unlimit the coredumpsize. You can do this in these ways:
‣
‣
Using csh:
% limit coredumpsize unlimited
% setenv PGI_TERM abort
Using bash, sh, zsh, or ksh:
$ ulimit -c unlimited
$ export PGI_TERM=abort
To debug a core file with pgdbg, start pgdbg with the -core option. For example, to view
a core file named "core" for a program named "a.out":
$ pgdbg -core core a.out
For more information on why to use this variable, refer to Stack Traceback and JIT
Debugging.
11.3.21. PGI_TERM_DEBUG
The PGI_TERM_DEBUG variable may be set to override the default behavior when
PGI_TERM is set to debug.
The value of PGI_TERM_DEBUG should be set to the command line used to invoke the
program. For example:
gdb --quiet --pid %d
User's Guide for x86-64 CPUs
Version 2017 | 146
Using Environment Variables
The first occurrence of %d in the PGI_TERM_DEBUG string is replaced by the process id.
The program named in the PGI_TERM_DEBUG string must be found on the current PATH
or specified with a full path name.
11.3.22. PGROUPD_LICENSE_FILE
You can use the PGROUPD_LICENSE_FILE to specifies the location of the PGI license.
This variable is set in the registry on Windows machines, and is specific to PGI products.
The system environment variable PGROUPD_LICENSE_FILE is not required by PGI
products on Windows but you can use it to override the default location that is searched
for the license.dat file.
To use the system environment variable PGROUPD_LICENSE_FILE, set it to the full path
of the license keys file. To do this, follow these steps:
1. Open the System Properties dialog from Control Panel | System.
2. Select the ‘Advanced’ tab.
3. Click the ‘Environment Variables’ button.
‣
‣
If PGROUPD_LICENSE_FILE is not already an environment variable, create a
new system variable for it. Set its value to the full path, including the name of
the file, for the license keys file.
If PGROUPD_LICENSE_FILE already exists as an environment variable, append
the path to the license file to the variable’s current value using a semi-colon to
separate entries.
11.3.23. PWD
The PWD variable allows you to display the current directory.
11.3.24. STATIC_RANDOM_SEED
You can use STATIC_RANDOM_SEED to force the seed returned by the Fortran 90/95
RANDOM_SEED intrinsic to be constant. The first call to RANDOM_SEED without
arguments resets the random seed to a default value, then advances the seed by a
variable amount based on time. Subsequent calls to RANDOM_SEED without arguments
reset the random seed to the same initial value as the first call. Unless the time is
exactly the same, each time a program is run a different random number sequence is
generated. Setting the environment variable STATIC_RANDOM_SEED to YES forces the
seed returned by RANDOM_SEED to be constant, thereby generating the same sequence of
random numbers at each execution of the program.
11.3.25. TMP
You can use TMP to specify the directory to use for placement of any temporary files
created during execution of the PGI compilers and tools. This variable is interchangeable
with TMPDIR.
User's Guide for x86-64 CPUs
Version 2017 | 147
Using Environment Variables
11.3.26. TMPDIR
You can use TMPDIR to specify the directory to use for placement of any temporary files
created during execution of the PGI compilers and tools.
11.4. Using Environment Modules on Linux
On Linux, if you use the Environment Modules package, that is, the module load
command, PGI includes a script to set up the appropriate module files.
Assuming your installation base directory is /opt/pgi, and your MODULEPATH
environment variable is /usr/local/Modules/modulefiles, execute this
command:
% /opt/pgi/linux86-64/17.5/etc/modulefiles/pgi.module.install \
-all -install /usr/local/Modules/modulefiles
This command creates module files for all installed versions of the PGI compilers.
You must have write permission to the modulefiles directory to enable the module
commands:
% module load pgi64/17.5
% module load pgi/17.5
where "pgi/17.5" also uses the 64-bit compilers.
To see what versions are available, use this command:
% module avail pgi
The module load command sets or modifies the environment variables as indicated in
the following table.
This Environment Variable...
Is set or modified by the module load command
CC
Full path to
CPP
Full path to
CXX
Path to
FC
Full path to
F77
Full path to
F90
Full path to
LD_LIBRARY_PATH
Prepends the PGI library directory
MANPATH
Prepends the PGI man page directory
PATH
Prepends the PGI compiler and tools
User's Guide for x86-64 CPUs
pgcc
pgprepro
pgc++
pgfortran
pgf77
pgf90
bin directory
Version 2017 | 148
Using Environment Variables
This Environment Variable...
Is set or modified by the module load command
PGI
The base installation directory
PGI does not provide support for the Environment Modules package. For more
information about the package, go to: http://modules.sourceforge.net.
11.5. Stack Traceback and JIT Debugging
When a programming error results in a runtime error message or an application
exception, a program will usually exit, perhaps with an error message. The PGI runtime
library includes a mechanism to override this default action and instead print a stack
traceback, start a debugger, or, on Linux, create a core file for post-mortem debugging.
The stack traceback and just-in-time debugging functionality is controlled by an
environment variable, PGI_TERM, described in PGI_TERM. The runtime libraries use
the value of PGI_TERM to determine what action to take when a program abnormally
terminates.
When the PGI runtime library detects an error or catches a signal, it calls the routine
pgi_stop_here() prior to generating a stack traceback or starting the debugger. The
pgi_stop_here() routine is a convenient spot to set a breakpoint when debugging a
program.
User's Guide for x86-64 CPUs
Version 2017 | 149
Chapter 12.
DISTRIBUTING FILES – DEPLOYMENT
Once you have successfully built, debugged and tuned your application, you may want
to distribute it to users who need to run it on a variety of systems. This section addresses
how to effectively distribute applications built using PGI compilers and tools. The
application must be installed in such a way that it executes accurately on a system other
than the one on which it was built, and which may be configured differently.
12.1. Deploying Applications on Linux
To successfully deploy your application on Linux, some of the issues to consider include:
‣
‣
‣
‣
Runtime Libraries
64-bit Linux Systems
Redistribution of Files
Licensing
12.1.1. Runtime Library Considerations
On Linux systems, the system runtime libraries can be linked to an application either
statically or dynamically. For example, for the C runtime library, libc, you can use
either the static version libc.a or the shared object version libc.so. If the application
is intended to run on Linux systems other than the one on which it was built, it is
generally safer to use the shared object version of the library. This approach ensures that
the application uses a version of the library that is compatible with the system on which
the application is running. Further, it works best when the application is linked on a
system that has an equivalent or earlier version of the system software than the system
on which the application will be run.
Building on a newer system and running the application on an older system may not
produce the desired output.
To use the shared object version of a library, the application must also link to shared
object versions of the PGI runtime libraries. To execute an application built in such a
way on a system on which PGI compilers are not installed, those shared objects must
User's Guide for x86-64 CPUs
Version 2017 | 150
Distributing Files – Deployment
be available.To build using the shared object versions of the runtime libraries, use the
-Bdynamic option, as shown here:
$ pgf90 -Bdynamic myprog.f90
12.1.2. 64-bit Linux Considerations
On 64-bit Linux systems, 64-bit applications that use the -mcmodel=medium option
sometimes cannot be successfully linked statically. Therefore, users with executables
built with the -mcmodel=medium option may need to use shared libraries, linking
dynamically. Also, runtime libraries built using the -fpic option use 32-bit offsets,
so they sometimes need to reside near other runtime libs in a shared area of Linux
program memory.
If your application is linked dynamically using shared objects, then the shared object
versions of the PGI runtime are required.
12.1.3. Linux Redistributable Files
The method for installing the shared object versions of the runtime libraries required for
applications built with PGI compilers and tools is manual distribution.
When the PGI compilers are installed, there are directories that have a name that begins
with REDIST; these directories contain the redistributed shared object libraries. These
may be redistributed by licensed PGI customers under the terms of the End-User License
Agreement.
12.1.4. Restrictions on Linux Portability
You cannot expect to be able to run an executable on any given Linux machine.
Portability depends on the system you build on as well as how much your program
uses system routines that may have changed from Linux release to Linux release.
For example, an area of significant change between some versions of Linux is in
libpthread.so and libnuma.so. PGI compilers use these dynamically linked
libraries for the options -Mconcur (auto-parallel), -mp (OpenMP), and -acc ta=multicore (OpenACC). Statically linking these libraries may not be possible, or
may result in failure at execution.
Typically, portability is supported for forward execution, meaning running a program
on the same or a later version of Linux. But not for backward compatibility, that is,
running on a prior release. For example, a user who compiles and links a program under
RHEL 7.2 should not expect the program to run without incident on a RHEL 5.2 system,
an earlier Linux version. It may run, but it is less likely. Developers might consider
building applications on earlier Linux versions for wider usage. Dynamic linking of
Linux and gcc system routines on the platform executing the program can also reduce
problems.
User's Guide for x86-64 CPUs
Version 2017 | 151
Distributing Files – Deployment
12.1.5. Licensing for Redistributable Files
The files in the REDIST directories may be redistributed under the terms of the End-User
License Agreement for the product in which they were included.
12.2. Deploying Applications on Windows
Windows programs may be linked statically or dynamically.
‣
‣
A statically linked program is completely self-contained, created by linking to static
versions of the PGI and Microsoft runtime libraries.
A dynamically linked program depends on separate dynamically-linked libraries
(DLLs) that must be installed on a system for the application to run on that system.
Although it may be simpler to install a statically linked executable, there are advantages
to using the DLL versions of the runtime, including:
‣
‣
‣
Executable binary file size is smaller.
Multiple processes can use DLLs at once, saving system resources.
New versions of the runtime can be installed and used by the application without
rebuilding the application.
Dynamically-linked Windows programs built with PGI compilers depend on dynamic
runtime library files (DLLs). These DLLs must be distributed with such programs to
enable them to execute on systems where the PGI compilers are not installed. These
redistributable libraries include both PGI runtime libraries and Microsoft runtime
libraries.
12.2.1. PGI Redistributables
PGI redistributable directories contain all of the PGI Linux runtime library shared object
files or Windows dynamically-linked libraries that can be re-distributed by PGI 17.5
licensees under the terms of the PGI End-User License Agreement (EULA).
12.2.2. Microsoft Redistributables
The PGI products on Windows include Microsoft Open Tools. The Microsoft Open Tools
directory contains a subdirectory named redist. PGI licensees may redistribute the
files contained in this directory in accordance with the terms of the End-User License
Agreement.
Microsoft supplies installation packages, vcredist_x86.exe and
vcredist_x64.exe, containing these runtime files. These files are available in the
redist directory.
User's Guide for x86-64 CPUs
Version 2017 | 152
Distributing Files – Deployment
12.3. Code Generation and Processor Architecture
The PGI compilers can generate much more efficient code if they know the specific
x86-64 processor architecture on which the program will run. When preparing to deploy
your application, you should determine whether you want the application to run on
the widest possible set of x86-64 processors, or if you want to restrict the application to
run on a specific processor or set of processors. The restricted approach allows you to
optimize performance for that set of processors.
Different processors have differences, some subtle, in hardware features, such as
instruction sets and cache size. The compilers make architecture-specific decisions such
as instruction selection, instruction scheduling, and vectorization, all of which can have
a profound effect on the performance of applications.
Processor-specific code generation is controlled by the -tp option, described in the
section ‘-tp <target> [,target...]’ of the PGI Compiler Reference Manual. When an
application is compiled without any -tp options, the compiler generates code for the
type of processor on which the compiler is run.
12.3.1. Generating Generic x86-64 Code
To generate generic x86-64 code, use one of the following forms of the-tp option on
your command line:
-tp px ! generate code for any x86-64 cpu type
-tp px-64 ! generate code for any x86-64 cpu type
Both of these examples are good choices for portable execution.
12.3.2. Generating Code for a Specific Processor
You can use the -tp option to request that the compiler generate code optimized for a
specific processor. The PGI Release Notes contains a list of supported processors or you
can look at the -tp entry in the compiler output generated by using the -help option,
described in ‘-help’ in the PGI Compiler Reference Manual.
12.4. Generating One Executable for Multiple
Types of Processors
PGI unified binaries provide a low-overhead method for a single program to run well on
a number of hardware platforms.
All 64-bit PGI compilers can produce PGI Unified Binary programs that contain code
streams fully optimized and supported for both AMD64 and Intel 64 processors using
the -tp target option.
User's Guide for x86-64 CPUs
Version 2017 | 153
Distributing Files – Deployment
The compilers generate and combine multiple binary code streams into one executable,
where each stream is optimized for a specific platform. At runtime, this one executable
senses the environment and dynamically selects the appropriate code stream.
Executable size is automatically controlled via unified binary culling. Only those
functions and subroutines where the target affects the generated code have unique
binary images, resulting in a code-size savings of 10–90% compared to generating full
copies of code for each target.
Programs can use PGI Unified Binary technology even if all of the object files and
libraries are not compiled as unified binaries. Like any other object file, you can use PGI
Unified Binary object files to create programs or libraries. No special start up code is
needed; support is linked in from the PGI libraries.
The -Mpfi option disables generation of PGI Unified Binary object files. Instead, the
default target auto-detect rules for the host are used to select the target processor.
12.4.1. PGI Unified Binary Command-line Switches
The PGI Unified Binary command-line switch is an extension of the target processor
switch, -tp, which may be applied to individual files during compilation.
The target processor switch, -tp, accepts a comma-separated list of 64-bit targets and
generates code optimized for each listed target.
The following example generates optimized code for three targets:
-tp k8-64,p7-64,core2-64
A special target switch, -tp x64, is the same as -tp k8-64, p7-64.
12.4.2. PGI Unified Binary Directives and Pragmas
PGI Unified binary directives and pragmas may be applied to functions, subroutines,
or whole files. The directives and pragmas cause the compiler to generate PGI Unified
Binary code optimized for one or more targets. No special command line options are
needed for these pragmas and directives to take effect.
The syntax of the Fortran directive is:
pgi$[g|r| ] pgi tp [target]...
where the scope is g (global), r (routine) or blank. The default is r, routine.
For example, the following syntax indicates that the whole file, represented by g, should
be optimized for both k8_64 and p7_64.
pgi$g pgi tp k8_64 p7_64
The syntax of the C/C++ pragma is:
#pragma [global|routine|] tp [target]...
where the scope is global, routine, or blank. The default is routine.
For example, the following syntax indicates that the next function should be optimized
for k8_64, p7_64, and core2_64.
#pragma routine tp k8_64 p7_64 core2_64
User's Guide for x86-64 CPUs
Version 2017 | 154
Chapter 13.
INTER-LANGUAGE CALLING
This section describes inter-language calling conventions for C, C++, and Fortran
programs using the PGI compilers. Fortran 2003 ISO_C_Binding provides a mechanism
to support the interoperability with C. This includes the ISO_C_Binding intrinsic
module, binding labels, and the BIND attribute. In the absence of this mechanism, the
following sections describe how to call a Fortran function or subroutine from a C or C+
+ program and how to call a C or C++ function from a Fortran program. For information
on calling assembly language programs, refer to the ‘‘Runtime Environment’’ section of
the PGI Compilers Reference Guide.
This section provides examples that use the following options related to inter-language
calling. For more information on these options, refer to the ‘‘Command-Line Options
Reference’’ section of the PGI Compiler Reference Guide.
-c
-Mnomain
-Miface
-Mupcase
13.1. Overview of Calling Conventions
This section includes information on the following topics:
‣
‣
‣
‣
‣
Functions and subroutines in Fortran, C, and C++
Naming and case conversion conventions
Compatible data types
Argument passing and special return values
Arrays and indexes
The sections Inter-language Calling Considerations through Example – C++ Calling
Fortran describe how to perform inter-language calling using the Linux, macOS, or
Win64 convention.
13.2. Inter-language Calling Considerations
In general, when argument data types and function return values agree, you can call a
C or C++ function from Fortran as well as call a Fortran function from C or C++. When
User's Guide for x86-64 CPUs
Version 2017 | 155
Inter-language Calling
data types for arguments do not agree, you may need to develop custom mechanisms
to handle them. For example, the Fortran COMPLEX type has a matching type in C99
but does not have a matching type in C89; however, it is still possible to provide interlanguage calls but there are no general calling conventions for such cases.
‣
‣
‣
‣
If a C++ function contains objects with constructors and destructors, calling such
a function from either C or Fortran is not possible unless the initialization in
the main program is performed from a C + + program in which constructors and
destructors are properly initialized.
In general, you can call a C or Fortran function from C++ without problems
as long as you use the extern "C" keyword to declare the function in the C++
program. This declaration prevents name mangling for the C function name.
If you want to call a C++ function from C or Fortran, you also have to use the
extern "C" keyword to declare the C++ function. This keeps the C++ compiler from
mangling the name of the function.
You can use the __cplusplus macro to allow a program or header file to work for
both C and C++. For example, the following defines in the header file stdio.h
allow this file to work for both C and C++.
#ifndef _STDIO_H
#define _STDIO_H
#ifdef __cplusplus
extern "C" {
#endif /* __cplusplus */
.
. /* Functions and data types defined... */
.
#ifdef __cplusplus
}
#endif /* __cplusplus */
#endif
C++ member functions cannot be declared extern, since their names will
always be mangled. Therefore, C++ member functions cannot be called from C or
Fortran.
13.3. Functions and Subroutines
Fortran, C, and C++ define functions and subroutines differently.
For a Fortran program calling a C or C++ function, observe the following return value
convention:
‣
‣
When a C or C++ function returns a value, call it from Fortran as a function.
When a C or C++ function does not return a value, call it as a subroutine.
For a C/C++ program calling a Fortran function, the call should return a similar type.
Table 28, Fortran and C/C++ Data Type Compatibility, lists compatible types. If the call is
to a Fortran subroutine, a Fortran CHARACTER function, or a Fortran COMPLEX function,
call it from C/C++ as a function that returns void. The exception to this convention is
when a Fortran subroutine has alternate returns; call such a subroutine from C/C++ as
a function returning int whose value is the value of the integer expression specified in
the alternate RETURN statement.
User's Guide for x86-64 CPUs
Version 2017 | 156
Inter-language Calling
13.4. Upper and Lower Case Conventions,
Underscores
By default on Linux, Win64, and macOS systems, all Fortran symbol names are
converted to lower case. C and C++ are case sensitive, so upper-case function names
stay upper-case. When you use inter-language calling, you can either name your C/C+
+ functions with lower-case names, or invoke the Fortran compiler command with the
option -Mupcase, in which case it will not convert symbol names to lower-case.
When programs are compiled using one of the PGI Fortran compilers on Linux, Win64,
and macOS systems, an underscore is appended to Fortran global names (names of
functions, subroutines and common blocks). This mechanism distinguishes Fortran
name space from C/C++ name space. Use these naming conventions:
‣
‣
If you call a C/C++ function from Fortran, you should rename the C/C++ function by
appending an underscore or use C$PRAGMA C in the Fortran program. For more
information on C$PRAGMA C, refer to !$PRAGMA C.
If you call a Fortran function from C/C++, you should append an underscore to the
Fortran function name in the calling program.
13.5. Compatible Data Types
Table 28 shows compatible data types between Fortran and C/C++. Table 29, Fortran
and C/C++ Representation of the COMPLEX Type shows how the Fortran COMPLEX type
may be represented in C/C++.
Tip If you can make your function/subroutine parameters as well as your return values
match types, you should be able to use inter-language calling.
Table 28 Fortran and C/C++ Data Type Compatibility
Fortran Type (lower case)
C/C++ Type
Size (bytes)
character x
char x
1
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
double x
8
integer x
int x
4
integer*1 x
signed char x
1
integer*2 x
short x
2
User's Guide for x86-64 CPUs
Version 2017 | 157
Inter-language Calling
Fortran Type (lower case)
C/C++ Type
Size (bytes)
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
int x
4
logical*8
long x
8
Table 29 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;
float complex x;
8
8
complex*8 x
struct {float r,i;} x;
float complex x;
8
8
double complex x
struct {double dr,di;} x;
double complex x;
16
16
complex *16 x
struct {double dr,di;} x;
double complex x;
16
16
For C/C++, the complex type implies C99 or later.
13.5.1. Fortran Named 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. For example, here is a Fortran common block:
INTEGER I
COMPLEX C
DOUBLE COMPLEX CD
DOUBLE PRECISION D
COMMON /COM/ i, c, cd, d
This Fortran Common Block is represented in C with the following equivalent:
extern struct {
int i;
struct {float real, imag;} c;
struct {double real, imag;} cd;
double d;
} com_;
User's Guide for x86-64 CPUs
Version 2017 | 158
Inter-language Calling
This same Fortran Common Block is represented in C++ with the following equivalent:
extern "C" struct {
int i;
struct {float real, imag;} c;
struct {double real, imag;} cd;
double d;
} com_;
Tip For global or external data sharing, extern "C" is not required.
13.6. Argument Passing and Return Values
In Fortran, arguments are passed by reference, that is, the address of the argument is
passed, rather than the argument itself. In C/C++, arguments are passed by value, except
for strings and arrays, which are passed by reference. Due to the flexibility provided
in C/C++, you can work around these differences. Solving the parameter passing
differences generally involves intelligent use of the & and * operators in argument
passing when C/C++ calls Fortran and in argument declarations when Fortran calls C/C+
+.
For strings declared in Fortran as type CHARACTER, an argument representing the length
of the string is also passed to a calling function.
On the following systems, the compiler places the length argument(s) at the end of the
parameter list, following the other formal arguments:
‣
‣
On Linux and macOS systems
On Win64 systems, except when using the option -Miface=cref
The length argument is passed by value, not by reference.
13.6.1. Passing by Value (%VAL)
When passing parameters from a Fortran subprogram to a C/C++ function, it is possible
to pass by value using the %VAL function. If you enclose a Fortran parameter with
%VAL(), the parameter is passed by value. For example, the following call passes the
integer i and the logical bvar by value.
integer*1 i
logical*1 bvar
call cvalue (%VAL(i), %VAL(bvar))
13.6.2. Character Return Values
Functions and Subroutines describes the general rules for return values for C/C++
and Fortran inter-language calling. There is a special return value to consider. When a
Fortran function returns a character, two arguments need to be added at the beginning
of the C/C++ calling function’s argument list:
‣
‣
The address of the return character or characters
The length of the return character
User's Guide for x86-64 CPUs
Version 2017 | 159
Inter-language Calling
The following example illustrates the extra parameters, tmp and 10, supplied by the
caller:
Character Return Parameters
! Fortran function returns a character
CHARACTER*(*) FUNCTION CHF(C1,I)
CHARACTER*(*) C1
INTEGER I
END
/* C declaration of Fortran function */
extern void chf_();
char tmp[10];
char c1[9];
int i;
chf_(tmp, 10, c1, &i, 9);
If the 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 must still be supplied, but is not used.
The value of the character function is not automatically NULL-terminated.
13.6.3. Complex Return Values
When a Fortran function returns a complex value, an argument needs to be added at
the beginning of the C/C++ calling function’s argument list; this argument is the address
of the complex return value. COMPLEX Return Values illustrates the extra parameter,
cplx, supplied by the caller.
COMPLEX Return Values
COMPLEX FUNCTION CF(C, I)
INTEGER I
. . .
END
extern void cf_();
typedef struct {float real, imag;} cplx;
cplx c1;
int i;
cf_(&c1, &i);
13.7. Array Indices
C/C++ arrays and Fortran arrays use different default initial array index values. By
default, arrays in C/C++ start at 0 and arrqays in Fortran start at 1. If you adjust your
array comparisons so that a Fortran second element is compared to a C/C++ first
element, and adjust similarly for other elements, you should not have problems working
with this difference. If this is not satisfactory, you can declare your Fortran arrays to start
at zero.
Another difference between Fortran and C/C++ arrays is the storage method used.
Fortran uses column-major order and C/C++ uses row-major order. For one-dimensional
User's Guide for x86-64 CPUs
Version 2017 | 160
Inter-language Calling
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. For
arrays other than single dimensional arrays, and square two-dimensional arrays, interlanguage function mixing is not recommended.
13.8. Examples
This section contains examples that illustrate inter-language calling.
13.8.1. Example – Fortran Calling C
There are other solutions to calling C from Fortran than the one presented in this
section. For example, you can use the iso_c_binding intrinsic module which PGI
does support. For more information on this module and for examples of how to use it,
search the web using the keyword iso_c_binding.
C function f2c_func_ shows a C function that is called by the Fortran main program
shown in Fortran Main Program f2c_main.f. Notice that each argument is defined as a
pointer, since Fortran passes by reference. Also notice that the C function name uses all
lower-case and a trailing "_".
Fortran Main Program f2c_main.f
logical*1 bool1
character letter1
integer*4 numint1, numint2
real numfloat1
double precision numdoub1
integer*2 numshor1
external f2c_func
call f2c_func(bool1, letter1, numint1, numint2, numfloat1, numdoub1, numshor1)
write( *, "(L2, A2, I5, I5, F6.1, F6.1, I5)")
+ bool1, letter1, numint1, numint2, numfloat1,numdoub1, numshor1
end
C function f2c_func_
#define TRUE 0xff
#define FALSE 0
void f2c_func_( bool1, letter1, numint1, numint2, numfloat1,\
numdoub1, numshor1, len_letter1)
char *bool1, *letter1;
int *numint1, *numint2;
float *numfloat1;
double *numdoub1;
short *numshor1;
int len_letter1;
{
*bool1 = TRUE; *letter1 = 'v';
*numint1 = 11; *numint2 = -44;
*numfloat1 = 39.6 ;
*numdoub1 = 39.2;
*numshor1 = 981;
}
User's Guide for x86-64 CPUs
Version 2017 | 161
Inter-language Calling
Compile and execute the program f2c_main.f with the call to f2c_func_ using the
following command lines:
$ pgcc -c f2c_func.c
$ pgfortran f2c_func.o f2c_main.f
Executing the a.out file should produce the following output:
T v 11 -44 39.6 39.2 981
13.8.2. Example – C Calling Fortran
The example C Main Program c2f_main.c shows a C main program that calls the Fortran
subroutine shown in Fortran Subroutine c2f_sub.f.
‣
‣
Each call uses the & operator to pass by reference.
The call to the Fortran subroutine uses all lower-case and a trailing "_".
C Main Program c2f_main.c
void main () {
char bool1, letter1;
int numint1, numint2;
float numfloat1;
double numdoub1;
short numshor1;
extern void c2f_func_();
c2f_sub_(&bool1,&letter1,&numint1,&numint2,&numfloat1,&numdoub1,&numshor1, 1);
printf(" %s %c %d %d %3.1f %.0f %d\n",
bool1?"TRUE":"FALSE", letter1, numint1, numint2,
numfloat1, numdoub1, numshor1);
}
Fortran Subroutine c2f_sub.f
subroutine c2f_func ( bool1, letter1, numint1, numint2,
+ numfloat1, numdoub1, numshor1)
logical*1 bool1
character letter1
integer numint1, numint2
double precision numdoub1
real numfloat1
integer*2 numshor1
bool1 = .true.
letter1 = "v"
numint1 = 11
numint2 = -44
numdoub1 = 902
numfloat1 = 39.6
numshor1 = 299
return
end
To compile this Fortran subroutine and C program, use the following commands:
$ pgcc -c c2f_main.c
$ pgfortran -Mnomain c2f_main.o c2_sub.f
Executing the resulting a.out file should produce the following output:
TRUE v 11 -44 39.6 902 299
User's Guide for x86-64 CPUs
Version 2017 | 162
Inter-language Calling
13.8.3. Example – C++ Calling C
C++ Main Program cp2c_main.C Calling a C Function shows a C++ main program that
calls the C function shown in Simple C Function c2cp_func.c.
C++ Main Program cp2c_main.C Calling a C Function
extern "C" void cp2c_func(int n, int m, int *p);
#include <iostream>
main()
{
int a,b,c;
a=8;
b=2;
c=0;
cout << "main: a = "<<a<<" b = "<<b<<"ptr c = "<<hex<<&c<< endl;
cp2c_func(a,b,&c);
cout << "main: res = "<<c<<endl;
}
Simple C Function c2cp_func.c
void cp2c_func(num1, num2, res)
int num1, num2, *res;
{
printf("func: a = %d b = %d ptr c = %x\n",num1,num2,res);
*res=num1/num2;
printf("func: res = %d\n",*res);
}
To compile this C function and C++ main program, use the following commands:
$ pgcc -c cp2c_func.c
$ pgc++ cp2c_main.C cp2c_func.o
Executing the resulting a.out file should produce the following output:
main:
func:
func:
main:
a =
a =
res
res
8
8
=
=
b = 2 ptr c = 0xbffffb94
b = 2 ptr c = bffffb94
4
4
13.8.4. Example – C Calling C ++
The example in C Main Program c2cp_main.c Calling a C++ Function shows a C main
program that calls the C++ function shown in Simple C++ Function c2cp_func.C with
Extern C.
C Main Program c2cp_main.c Calling a C++ Function
extern void c2cp_func(int a, int b, int *c);
#include <stdio.h>
main() {
int a,b,c;
a=8; b=2;
printf("main: a = %d b = %d ptr c = %x\n",a,b,&c);
c2cp_func(a,b,&c);
printf("main: res = %d\n",c);
}
User's Guide for x86-64 CPUs
Version 2017 | 163
Inter-language Calling
Simple C++ Function c2cp_func.C with Extern C
#include <iostream>
extern "C" void c2cp_func(int num1,int num2,int *res)
{
cout << "func: a = "<<num1<<" b = "<<num2<<"ptr c ="<<res<<endl;
*res=num1/num2;
cout << "func: res = "<<res<<endl;
}
To compile this C function and C++ main program, use the following commands:
$ pgcc -c c2cp_main.c
$ pgc++ c2cp_main.o c2cp_func.C
Executing the resulting a.out file should produce the following output:
main:
func:
func:
main:
a =
a =
res
res
8
8
=
=
b = 2 ptr c = 0xbffffb94
b = 2 ptr c = bffffb94
4
4
You cannot use the extern "C" form of declaration for an object’s member functions.
13.8.5. Example – Fortran Calling C++
The Fortran main program shown in Fortran Main Program f2cp_main.f calling a C++
function calls the C++ function shown in C++ function f2cp_func.C .
Notice:
‣
‣
Each argument is defined as a pointer in the C++ function, since Fortran passes by
reference.
The C++ function name uses all lower-case and a trailing "_":
Fortran Main Program f2cp_main.f calling a C++ function
logical*1 bool1
character letter1
integer*4 numint1, numint2
real numfloat1
double precision numdoub1
integer*2 numshor1
external f2cpfunc
call f2cp_func (bool1, letter1, numint1,
+ numint2, numfloat1, numdoub1, numshor1)
write( *, "(L2, A2, I5, I5, F6.1, F6.1, I5)")
+ bool1, letter1, numint1, numint2, numfloat1,
+ numdoub1, numshor1
end
User's Guide for x86-64 CPUs
Version 2017 | 164
Inter-language Calling
C++ function f2cp_func.C
#define TRUE 0xff
#define FALSE 0
extern "C"
{
extern void f2cp_func_ (
char *bool1, *letter1,
int *numint1, *numint2,
float *numfloat1,
double *numdoub1,
short *numshort1,
int len_letter1)
{
*bool1 = TRUE;
*letter1 = 'v';
*numint1 = 11;
*numint2 = -44;
*numfloat1 = 39.6; *numdoub1 = 39.2;
}
}
*numshort1 = 981;
Assuming the Fortran program is in a file fmain.f, and the C++ function is in a file
cpfunc.C, create an executable, using the following command lines:
$ pgc++ -c f2cp_func.C
$ pgfortran f2cp_func.o f2cp_main.f -pgc++libs
Executing the a.out file should produce the following output:
T v 11 -44 39.6 39.2 981
13.8.6. Example – C++ Calling Fortran
Fortran Subroutine cp2f_func.f shows a Fortran subroutine called by the C++ main
program shown in C++ main program cp2f_main.C. Notice that each call uses the &
operator to pass by reference. Also notice that the call to the Fortran subroutine uses all
lower-case and a trailing "_":
C++ main program cp2f_main.C
#include <iostream>
extern "C" { extern void cp2f_func_(char *,char *,int *,int *,
float *,double *,short *); }
main ()
{
char bool1, letter1;
int numint1, numint2;
float numfloat1;
double numdoub1;
short numshor1;
cp2f_func(&bool1,&letter1,&numint1,&numint2,&numfloat1, &numdoub1,&numshor1);
cout << " bool1 = ";
bool1?cout << "TRUE ":cout << "FALSE "; cout <<endl;
cout << " letter1 = " << letter1 <<endl;
cout << " numint1 = " << numint1 <<endl;
cout << " numint2 = " << numint2 <<endl;
cout << " numfloat1 = " << numfloat1 <<endl;
cout << " numdoub1 = " << numdoub1 <<endl;
cout << " numshor1 = " << numshor1 <<endl;
}
User's Guide for x86-64 CPUs
Version 2017 | 165
Inter-language Calling
Fortran Subroutine cp2f_func.f
subroutine cp2f_func ( bool1, letter1, numint1,
+ numint2, numfloat1, numdoub1, numshor1)
logical*1 bool1
character letter1
integer numint1, numint2
double precision numdoub1
real numfloat1
integer*2 numshor1
bool1 = .true. ; letter1 = "v"
numint1 = 11 ; numint2 = -44
numdoub1 = 902 ; numfloat1 = 39.6 ; numshor1 = 299
return
end
To compile this Fortran subroutine and C++ program, use the following command lines:
$ pgfortran -c cp2f_func.f
$ pgc++ cp2f_func.o cp2f_main.C -pgf90libs
Executing this C++ main should produce the following output:
bool1 = TRUE letter1 = v numint1 = 11 numint2 = -44 numfloat1 = 39.6 numdoub1 = 902
numshor1 = 299
..
You must explicitly link in the PGFORTRAN runtime support libraries when linking
pgfortran-compiled program units into C or C++ main programs. When linking pgf77compiled program units into C or C++ main programs, you need only link in -lpgftnrtl.
User's Guide for x86-64 CPUs
Version 2017 | 166
Chapter 14.
PROGRAMMING CONSIDERATIONS FOR
64-BIT ENVIRONMENTS
PGI provides 64-bit compilers for the 64-bit Linux, Windows, and macOS operating
systems running on the x64 architecture. You can use these compilers to create
programs that use 64-bit memory addresses. However, there are limitations to how this
capability can be applied. With the exception of Linux86-64, the object file formats on
all of the operating systems limit the total cumulative size of code plus static data to
2GB. This limit includes the code and statically declared data in the program and in
system and user object libraries. Linux86-64 implements a mechanism that overcomes
this limitations, as described in Large Static Data in Linux. This section describes the
specifics of how to use the PGI compilers to make use of 64-bit memory addressing.
The 64-bit PGI compilers are 64-bit applications which cannot run on anything but 64bit CPUs running 64-bit Operating Systems.
This section describes how to use the following options related to 64-bit programming.
-fPIC
-i8
-mcmodel=medium
-Mlargeaddressaware
-Mlarge_arrays
-tp
14.1. Data Types in the 64-Bit Environment
The size of some data types can be different in a 64-bit environment. This section
describes the major differences. For detailed information, refer to the ‘Fortran, C, and C+
+ Data Types’ section of the PGI Compiler’s Reference Guide.
14.1.1. C/C++ Data Types
On 64-bit windows, the size of an int is 4 bytes, a long is 4 bytes, and a pointer is 8 bytes.
On the 64-bit Linux and macOS operating systems, the size of an int is 4 bytes, a long is 8
bytes, and a pointer is 8 bytes.
User's Guide for x86-64 CPUs
Version 2017 | 167
Programming Considerations for 64-Bit Environments
14.1.2. Fortran Data Types
In Fortran, the default size of the INTEGER type is 4 bytes. The -i8 compiler option
may be used to make the default size of all INTEGER data in the program 8 bytes.
When using the -Mlarge_arrays option, described in 64-Bit Array Indexing, any
4-byte INTEGER variables that are used to index arrays are silently promoted by the
compiler to 8 bytes. This promotion can lead to unexpected consequences, so 8-byte
INTEGER variables are recommended for array indexing when using the option Mlarge_arrays.
14.2. Large Static Data in Linux
Linux86-64 operating systems support two different memory models. The default model
used by PGI compilers is the small memory model, which can be specified using mcmodel=small. This is the 32-bit model, which limits the size of code plus statically
allocated data, including system and user libraries, to 2GB. The medium memory model,
specified by -mcmodel=medium, allows combined code and static data areas (.text
and .bss sections) larger than 2GB. The -mcmodel=medium option must be used on both
the compile command and the link command in order to take effect.
The Win64 and 64-bit macOS operating systems do not have any support for large static
data declarations.
There are two drawbacks to using -mcmodel=medium. First, there is increased
addressing overhead to support the large data range. This can affect performance,
though the compilers seek to minimize the added overhead through careful instruction
generation. Second, -mcmodel=medium cannot be used for objects in shared libraries,
because there is no OS support for 64-bit dynamic linkage.
14.3. Large Dynamically Allocated Data
Dynamically allocated data objects in programs compiled by the 64-bit PGI compilers
can be larger than 2GB. No special compiler options are required to enable this
functionality. The size of the allocation is only limited by the system. However, to
correctly access dynamically allocated arrays with more than 2G elements you should
use the -Mlarge_arrays option, described in the following section.
User's Guide for x86-64 CPUs
Version 2017 | 168
Programming Considerations for 64-Bit Environments
14.4. 64-Bit Array Indexing
The 64-bit PGI compilers provide an option, -Mlarge_arrays, that enables 64bit indexing of arrays. This means that, as necessary, 64-bit INTEGER constants and
variables are used to index arrays.
In the presence of -Mlarge_arrays, the compiler may silently promote 32-bit
integers to 64 bits, which can have unexpected side effects.
On Linux86-64, the -Mlarge_arrays option also enables single static data objects
larger than 2 GB. This option is the default in the presence of -mcmodel=medium.
On Win64, static data may not be larger than 2GB.
14.5. Compiler Options for 64-bit Programming
The usual switches that apply to 64-bit programmers seeking to increase the data range
of their applications are in the following table.
Table 30 64-bit Compiler Options
Option
Purpose
Considerations
-mcmodel=medium
Enlarge object size;
Allow for declared
data the size of larger
than 2GB
Linux86-64 only. Slower execution. Cannot be used
with -fPIC. Objects cannot be put into shared libraries.
-Mlargeaddressaware
[Win64 only]
Generates code that
allows for addresses
greater than 2GB,
using RIP-relative
addressing.
Use -Mlargeaddressaware=no for a direct addressing
mechanism that restricts the total addressable
memory. This is not applicable if the object file is
placed in a DLL. Further, if an object file is compiled
with this option, it must also be used when linking.
-Mlarge_arrays
Perform all arraylocation-to-address
calculations using 64bit integer arithmetic.
Slightly slower execution. Is implicit with
-mcmodel=medium. Can be used with option
-mcmodel=small. Win64 does not support
-Mlarge_arrays for static objects larger than 2GB.
-fPIC
Position independent
code. Necessary for
shared libraries.
Dynamic linking restricted to a 32-bit offset. External
symbol references should refer to other shared lib
routines, rather than the program calling them.
-i8
All INTEGER functions, Users should take care to explicitly declare INTEGER
data, and constants
functions as INTEGER*4.
not explicitly
declared INTEGER*4
are assumed to be
INTEGER*8.
User's Guide for x86-64 CPUs
Version 2017 | 169
Programming Considerations for 64-Bit Environments
The following table summarizes the limits of these programming models under the
specified conditions. The compiler options you use vary by processor.
Table 31 Effects of Options on Memory and Array Sizes
Addr. Math
Max Size Gbytes
A
I
AS
DS
TS
64
32
2
2
2
-fpic incompatible with -mcmodel=medium
64
32
2
2
2
Enable full support for 64-bit data addressing
64
64
>2
>2
>2
Condition
64-bit addr limited by option
-mcmodel=small
A
Address Type – size in bits of data used for address calculations, 64-bits.
I
Index Arithmetic -bit-size of data used to index into arrays and other aggregate data structures.
If 32-bit, total range of any single data object is limited to 2GB.
AS
Maximum Array Size- the maximum size in gigabytes of any single data object.
DS
- max size in gigabytes combined of all data objects in .bss
TS
Maximum Total Size- max size in gigabytes, in aggregate, of all executable code and data
objects in a running program.
14.6. Practical Limitations of Large Array
Programming
The 64-bit addressing capability of the Linux86-64 and Win64 environments can
cause unexpected issues when data sizes are enlarged significantly. The following
table describes the most common occurrences of practical limitations of large array
programming.
Table 32 64-Bit Limitations
array initialization
Initializing a large array with a data statement may result in very large assembly
and object files, where a line of assembler source is required for each element in
the initialized array. Compilation and linking can be very time consuming as well.
To avoid this issue, consider initializing large arrays in a loop at runtime rather
than in a data statement.
stack space
Stack space can be a problem for data that is stack-based. In Win64, stack space
can be increased by using this link-time switch, where N is the desired stack
size:-Wl,-stack:N
In linux86-64, stack size is increased in the environment.
Setting stacksize to unlimited often is not large enough.
limit stacksize new_size ! in csh
ulimit -s new_size ! in bash
User's Guide for x86-64 CPUs
Version 2017 | 170
Programming Considerations for 64-Bit Environments
page swapping
If your executable is much larger than the physical size of memory, page swapping
can cause it to run dramatically slower; it may even fail. This is not a compiler
problem. Try smaller data sets to determine whether or not a problem is due to
page thrashing.
configured space
Be sure your linux86-64 system is configured with swap space sufficiently large to
support the data sets used in your application(s). If your memory+swap space is
not sufficiently large, your application will likely encounter a segmentation fault
at runtime.
support for large
address offsets in
object file format
Arrays that are not dynamically allocated are limited by how the compiler
can express the ‘distance’ between them when generating code. A field in
the object file stores this ‘distance’ value, which is limited to 32-bits on
Win64 and linux86-64 with -mcmodel=small. It is 64-bits on linux86-64 with
-mcmodel=medium.
Without the 64-bit offset support in the object file format,
large arrays cannot be declared statically or locally stackbased.
14.7. Medium Memory Model and Large Array in C
Consider the following example, where the aggregate size of the arrays exceeds 2GB.
Medium Memory Model and Large Array in C
% cat bigadd.c
#include <stdio.h>
#define SIZE 600000000 /* > 2GB/4 */
static float a[SIZE], b[SIZE];
int
main()
{
long long i, n, m;
float c[SIZE]; /* goes on stack */
n = SIZE;
m = 0;
for (i = 0; i < n; i += 10000) {
a[i] = i + 1;
b[i] = 2.0 * (i + 1);
c[i] = a[i] + b[i];
m = i;
}
printf("a[0]=%g b[0]=%g c[0]=%g\n", a[0], b[0], c[0]);
printf("m=%lld a[%lld]=%g b[%lld]=%gc[%lld]=%g\n",m,m,a[m],m,b[m],m,c[m]);
return 0;
}
% pgcc -mcmodel=medium -o bigadd bigadd.c
When SIZE is greater than 2G/4, and the arrays are of type float with 4 bytes
per element, the size of each array is greater than 2GB. With pgcc, using the
-mcmodel=medium switch, a static data object can now be > 2GB in size. If you execute
with these settings in your environment, you may see the following:
% bigadd
Segmentation fault
User's Guide for x86-64 CPUs
Version 2017 | 171
Programming Considerations for 64-Bit Environments
Execution fails because the stack size is not large enough. You can most likely correct
this error by using the limit stacksize command to reset the stack size in your
environment:
% limit stacksize 3000M
The command limit stacksize unlimited probably does not provide as large a
stack as we are using in the this example.
% bigadd
a[0]=1 b[0]=2 c[0]=3
n=599990000 a[599990000]=5.9999e+08 b[599990000]=1.19998e+09
c[599990000]=1.79997e+09
14.8. Medium Memory Model and Large Array in
Fortran
The following example works with the PGFORTRAN, PGF95, and PGF77 compilers
included in Release 2017. These compilers use 64-bit addresses and index arithmetic
when the -mcmodel=medium option is used.
Consider the following example:
User's Guide for x86-64 CPUs
Version 2017 | 172
Programming Considerations for 64-Bit Environments
Medium Memory Model and Large Array in Fortran
% cat mat.f
program mat
integer i, j, k, size, l, m, n
parameter (size=16000) ! >2GB
parameter (m=size,n=size)
real*8 a(m,n),b(m,n),c(m,n),d
do i = 1, m
do j = 1, n
a(i,j)=10000.0D0*dble(i)+dble(j)
b(i,j)=20000.0D0*dble(i)+dble(j)
enddo
enddo
!$omp parallel
!$omp do
do i = 1, m
do j = 1, n
c(i,j) = a(i,j) + b(i,j)
enddo
enddo
!$omp do
do i=1,m
do j = 1, n
d = 30000.0D0*dble(i)+dble(j)+dble(j)
if (d .ne. c(i,j)) then
print *,"err i=",i,"j=",j
print *,"c(i,j)=",c(i,j)
print *,"d=",d
stop
endif
enddo
enddo
!$omp end parallel
print *, "M =",M,", N =",N
print *, "c(M,N) = ", c(m,n)
end
When compiled with the PGFORTRAN compiler using -mcmodel=medium:
% pgfortran -Mfree -mp -o mat mat.f -i8 -mcmodel=medium
% setenv OMP_NUM_THREADS 2
% mat
M = 16000 , N = 16000
c(M,N) = 480032000.0000000
14.9. Large Array and Small Memory Model in
Fortran
The following example uses large, dynamically-allocated arrays. The code is divided
into a main and subroutine so you could put the subroutine into a shared library.
Dynamic allocation of large arrays saves space in the size of executable and saves time
initializing data.
User's Guide for x86-64 CPUs
Version 2017 | 173
Programming Considerations for 64-Bit Environments
Large Array and Small Memory Model in Fortran
% cat mat_allo.f90
program mat_allo
integer i, j
integer size, m, n
parameter (size=16000)
parameter (m=size,n=size)
double precision, allocatable::a(:,:),b(:,:),c(:,:)
allocate(a(m,n), b(m,n), c(m,n))
do i = 100, m, 1
do j = 100, n, 1
a(i,j) = 10000.0D0 * dble(i) + dble(j)
b(i,j) = 20000.0D0 * dble(i) + dble(j)
enddo
enddo
call mat_add(a,b,c,m,n)
print *, "M =",m,",N =",n
print *, "c(M,N) = ", c(m,n)
end
subroutine mat_add(a,b,c,m,n)
integer m, n, i, j
double precision a(m,n),b(m,n),c(m,n)
do i = 1, m
do j = 1, n
c(i,j) = a(i,j) + b(i,j)
enddo
enddo
return
end
% pgfortran -o mat_allo mat_allo.f90 -i8 -Mlarge_arrays -mp -fast
User's Guide for x86-64 CPUs
Version 2017 | 174
Chapter 15.
C/C++ INLINE ASSEMBLY AND
INTRINSICS
15.1. Inline Assembly
Inline Assembly lets you specify machine instructions inside a "C" function. The format
for an inline assembly instruction is this:
{ asm | __asm__ } ("string");
The asm statement begins with the asm or __asm__ keyword. The __asm__ keyword is
typically used in header files that may be included in ISO "C" programs.
string is one or more machine specific instructions separated with a semi-colon (;) or
newline (\n) character. These instructions are inserted directly into the compiler’s
assembly-language output for the enclosing function.
Some simple asm statements are:
asm ("cli");
asm ("sti");
These asm statements disable and enable system interrupts respectively.
In the following example, the eax register is set to zero.
asm( "pushl %eax\n\t" "movl $0, %eax\n\t" "popl %eax");
Notice that eax is pushed on the stack so that it is it not clobbered. When the statement is
done with eax, it is restored with the popl instruction.
Typically a program uses macros that enclose asm statements. The following two
examples use the interrupt constructs created previously in this section:
#define disableInt __asm__ ("cli");
#define enableInt __asm__ ("sti");
User's Guide for x86-64 CPUs
Version 2017 | 175
C/C++ Inline Assembly and Intrinsics
15.2. Extended Inline Assembly
Inline Assembly explains how to use inline assembly to specify machine specific
instructions inside a "C" function. This approach works well for simple machine
operations such as disabling and enabling system interrupts. However, inline assembly
has three distinct limitations:
1. The programmer must choose the registers required by the inline assembly.
2. To prevent register clobbering, the inline assembly must include push and pop code
for registers that get modified by the inline assembly.
3. There is no easy way to access stack variables in an inline assembly statement.
Extended Inline Assembly was created to address these limitations. The format for
extended inline assembly, also known as extended asm, is as follows:
{ asm | __asm__ } [ volatile | __volatile__ ]
("string" [: [output operands]] [: [input operands]] [: [clobberlist]]);
‣
‣
‣
‣
Extended asm statements begin with the asm or __asm__ keyword. Typically the
__asm__ keyword is used in header files that may be included by ISO "C" programs.
An optional volatile or __volatile__ keyword may appear after the asm keyword. This
keyword instructs the compiler not to delete, move significantly, or combine with
any other asm statement. Like __asm__, the __volatile__ keyword is typically used
with header files that may be included by ISO "C" programs.
"string" is one or more machine specific instructions separated with a semi-colon
(;) or newline (\n) character. The string can also contain operands specified in the
[output operands], [input operands], and [clobber list]. The instructions are inserted
directly into the compiler's assembly-language output for the enclosing function.
The [output operands], [input operands], and [clobber list] items each describe the effect
of the instruction for the compiler. For example:
asm( "movl %1, %%eax\n" "movl %%eax, %0":"=r" (x) : "r" (y) : "%eax" );
where "=r" (x) is an output operand.
"r" (y) is an input operand.
"%eax" is the clobber list consisting of one register, "%eax".
The notation for the output and input operands is a constraint string surrounded by
quotes, followed by an expression, and surrounded by parentheses. The constraint
string describes how the input and output operands are used in the asm "string".
For example, "r" tells the compiler that the operand is a register. The "=" tells the
compiler that the operand is write only, which means that a value is stored in an
output operand's expression at the end of the asm statement.
Each operand is referenced in the asm "string" by a percent "%" and its number. The
first operand is number 0, the second is number 1, the third is number 2, and so on.
In the preceding example, "%0" references the output operand, and "%1" references
the input operand. The asm "string" also contains "%%eax", which references
machine register "%eax". Hard coded registers like "%eax" should be specified in the
clobber list to prevent conflicts with other instructions in the compiler's assemblylanguage output.
User's Guide for x86-64 CPUs
Version 2017 | 176
C/C++ Inline Assembly and Intrinsics
[output operands], [input operands], and [clobber list] items are described in more detail
in the following sections.
15.2.1. Output Operands
The [output operands] are an optional list of output constraint and expression pairs that
specify the result(s) of the asm statement. An output constraint is a string that specifies
how a result is delivered to the expression. For example, "=r" (x) says the output operand
is a write-only register that stores its value in the "C" variable x at the end of the asm
statement. An example follows:
int x;
void example()
{
asm( "movl $0, %0" : "=r" (x) );
}
The previous example assigns 0 to the "C" variable x. For the function in this example,
the compiler produces the following assembly. If you want to produce an assembly
listing, compile the example with the pgcc -S compiler option:
example:
..Dcfb0:
pushq %rbp
..Dcfi0:
movq %rsp, %rbp
..Dcfi1:
..EN1:
## lineno: 8
movl $0, %eax
movl %eax, x(%rip)
## lineno: 0
popq %rbp
ret
In the generated assembly shown, notice that the compiler generated two statements
for the asm statement at line number 5. The compiler generated "movl $0, %eax" from
the asm "string". Also notice that %eax appears in place of "%0" because the compiler
assigned the %eax register to variable x. Since item 0 is an output operand, the result
must be stored in its expression (x).
In addition to write-only output operands, there are read/write output operands
designated with a "+" instead of a "=". For example, "+r" (x) tells the compiler to initialize
the output operand with variable x at the beginning of the asm statement.
To illustrate this point, the following example increments variable x by 1:
int x=1;
void example2()
{
asm( "addl $1, %0" : "+r" (x) );
}
To perform the increment, the output operand must be initialized with variable x.
The read/write constraint modifier ("+") instructs the compiler to initialize the output
User's Guide for x86-64 CPUs
Version 2017 | 177
C/C++ Inline Assembly and Intrinsics
operand with its expression. The compiler generates the following assembly code for the
example2() function:
example2:
..Dcfb0:
pushq %rbp
..Dcfi0:
movq %rsp, %rbp
..Dcfi1:
..EN1:
## lineno: 5
movl x(%rip), %eax
addl $1, %eax
movl %eax, x(%rip)
## lineno: 0
popq %rbp
ret
From the example2() code, two extraneous moves are generated in the assembly: one
movl for initializing the output register and a second movl to write it to variable x. To
eliminate these moves, use a memory constraint type instead of a register constraint
type, as shown in the following example:
int x=1;
void example2()
{
asm( "addl $1, %0" : "+m" (x) );
}
The compiler generates a memory reference in place of a memory constraint. This
eliminates the two extraneous moves. Because the assembly uses a memory reference to
variable x, it does not have to move x into a register prior to the asm statement; nor does
it need to store the result after the asm statement. Additional constraint types are found
in Additional Constraints.
example2:
..Dcfb0:
pushq %rbp
..Dcfi0:
movq %rsp, %rbp
..Dcfi1:
..EN1:
## lineno: 5
addl $1, x(%rip)
## lineno: 0
popq %rbp
ret
The examples thus far have used only one output operand. Because extended asm
accepts a list of output operands, asm statements can have more than one result, as
shown in the following example:
void example4()
{
int x=1; int y=2;
asm( "addl $1, %1\n" "addl %1, %0": "+r" (x), "+m" (y) );
}
This example increments variable y by 1 then adds it to variable x. Multiple output
operands are separated with a comma. The first output operand is item 0 ("%0") and the
second is item 1 ("%1") in the asm "string". The resulting values for x and y are 4 and 3
respectively.
User's Guide for x86-64 CPUs
Version 2017 | 178
C/C++ Inline Assembly and Intrinsics
15.2.2. Input Operands
The [input operands] are an optional list of input constraint and expression pairs that
specify what "C" values are needed by the asm statement. The input constraints specify
how the data is delivered to the asm statement. For example, "r" (x) says that the input
operand is a register that has a copy of the value stored in "C" variable x. Another
example is "m" (x) which says that the input item is the memory location associated with
variable x. Other constraint types are discussed in Additional Constraints. An example
follows:
void example5()
{
int x=1;
int y=2;
int z=3;
asm( "addl %2, %1\n" "addl %2, %0" : "+r" (x), "+m" (y) : "r" (z) );
}
The previous example adds variable z, item 2, to variable x and variable y. The resulting
values for x and y are 4 and 5 respectively.
Another type of input constraint worth mentioning here is the matching constraint. A
matching constraint is used to specify an operand that fills both an input as well as an
output role. An example follows:
int x=1;
void example6()
{
asm( "addl $1, %1"
: "=r" (x)
: "0" (x) );
}
The previous example is equivalent to the example2() function shown in Output
Operands. The constraint/expression pair, "0" (x), tells the compiler to initialize output
item 0 with variable x at the beginning of the asm statement. The resulting value for x is
2. Also note that "%1" in the asm "string" means the same thing as "%0" in this case. That
is because there is only one operand with both an input and an output role.
Matching constraints are very similar to the read/write output operands mentioned
in Output Operands. However, there is one key difference between read/write output
operands and matching constraints. The matching constraint can have an input expression
that differs from its output expression.
The following example uses different values for the input and output roles:
int x;
int y=2;
void example7()
{
asm( "addl $1, %1"
: "=r" (x)
: "0" (y) );
}
User's Guide for x86-64 CPUs
Version 2017 | 179
C/C++ Inline Assembly and Intrinsics
The compiler generates the following assembly for example7():
example7:
..Dcfb0:
pushq %rbp
..Dcfi0:
movq %rsp, %rbp
..Dcfi1:
..EN1:
## lineno: 8
movl y(%rip), %eax
addl $1, %eax
movl %eax, x(%rip)
## lineno: 0
popq %rbp
ret
Variable x gets initialized with the value stored in y, which is 2. After adding 1, the
resulting value for variable x is 3.
Because matching constraints perform an input role for an output operand, it does not
make sense for the output operand to have the read/write ("+") modifier. In fact, the
compiler disallows matching constraints with read/write output operands. The output
operand must have a write only ("=") modifier.
15.2.3. Clobber List
The [clobber list] is an optional list of strings that hold machine registers used in the asm
"string". Essentially, these strings tell the compiler which registers may be clobbered by
the asm statement. By placing registers in this list, the programmer does not have to
explicitly save and restore them as required in traditional inline assembly (described in
Inline Assembly). The compiler takes care of any required saving and restoring of the
registers in this list.
Each machine register in the [clobber list] is a string separated by a comma. The leading
'%' is optional in the register name. For example, "%eax" is equivalent to "eax". When
specifying the register inside the asm "string", you must include two leading '%'
characters in front of the name (for example., "%%eax"). Otherwise, the compiler will
behave as if a bad input/output operand was specified and generate an error message.
An example follows:
void example8()
{
int x;
int y=2;
asm( "movl %1, %%eax\n"
"movl %1, %%edx\n"
"addl %%edx, %%eax\n"
"addl %%eax, %0"
: "=r" (x)
: "0" (y)
: "eax", "edx" );
}
This code uses two hard-coded registers, eax and edx. It performs the equivalent of 3*y
and assigns it to x, producing a result of 6.
In addition to machine registers, the clobber list may contain the following special flags:
User's Guide for x86-64 CPUs
Version 2017 | 180
C/C++ Inline Assembly and Intrinsics
"cc"
The asm statement may alter the control code register.
"memory"
The asm statement may modify memory in an unpredictable fashion.
When the "memory" flag is present, the compiler does not keep memory values cached
in registers across the asm statement and does not optimize stores or loads to that
memory. For example:
asm("call MyFunc":::"memory");
This asm statement contains a "memory" flag because it contains a call. The callee may
otherwise clobber registers in use by the caller without the "memory" flag.
The following function uses extended asm and the "cc" flag to compute a power of 2 that
is less than or equal to the input parameter n.
#pragma noinline
int asmDivideConquer(int n)
{
int ax = 0;
int bx = 1;
asm (
"LogLoop:n"
"cmp %2, %1n"
"jnle Donen"
"inc %0n"
"add %1,%1n"
"jmp LogLoopn"
"Done:n"
"dec %0n"
:"+r" (ax), "+r" (bx) : "r" (n) : "cc");
return ax;
}
The ‘cc’ flag is used because the asm statement contains some control flow that may alter
the control code register. The #pragma noinline statement prevents the compiler from
inlining the asmDivideConquer() function. If the compiler inlines asmDivideConquer(),
then it may illegally duplicate the labels LogLoop and Done in the generated assembly.
15.2.4. Additional Constraints
Operand constraints can be divided into four main categories:
‣
‣
‣
‣
Simple Constraints
Machine Constraints
Multiple Alternative Constraints
Constraint Modifiers
15.2.5. Simple Constraints
The simplest kind of constraint is a string of letters or characters, known as Simple
Constraints, such as the "r" and "m" constraints introduced in Output Operands. Table 33
describes these constraints.
User's Guide for x86-64 CPUs
Version 2017 | 181
C/C++ Inline Assembly and Intrinsics
Table 33 Simple Constraints
Constraint
Description
whitespace
Whitespace characters are ignored.
E
An immediate floating point operand.
F
Same as "E".
g
Any general purpose register, memory, or immediate integer operand is allowed.
i
An immediate integer operand.
m
A memory operand. Any address supported by the machine is allowed.
n
Same as "i".
o
Same as "m".
p
An operand that is a valid memory address. The expression associated with the
constraint is expected to evaluate to an address (for example, "p" (&x) ).
r
A general purpose register operand.
X
Same as "g".
0,1,2,..9
Matching Constraint. See Output Operands for a description.
The following example uses the general or "g" constraint, which allows the compiler to
pick an appropriate constraint type for the operand; the compiler chooses from a general
purpose register, memory, or immediate operand. This code lets the compiler choose the
constraint type for "y".
void example9()
{
int x, y=2;
asm( "movl %1, %0\n" : "=r"
(x) : "g" (y) );
}
This technique can result in more efficient code. For example, when compiling
example9() the compiler replaces the load and store of y with a constant 2. The compiler
can then generate an immediate 2 for the y operand in the example. The assembly
generated by pgcc for our example is as follows:
example9:
..Dcfb0:
pushq %rbp
..Dcfi0:
movq %rsp, %rbp
..Dcfi1:
..EN1:
## lineno: 3
movl $2, %eax
## lineno: 6
popq %rbp
ret
In this example, notice the use of $2 for the "y" operand.
User's Guide for x86-64 CPUs
Version 2017 | 182
C/C++ Inline Assembly and Intrinsics
Of course, if y is always 2, then the immediate value may be used instead of the variable
with the "i" constraint, as shown here:
void example10()
{
int x;
asm( "movl %1, %0\n"
: "=r" (x)
: "i" (2) );
}
Compiling example10() with pgcc produces assembly similar to that produced for
example9().
15.2.6. Machine Constraints
Another category of constraints is Machine Constraints. The x86_64 architectures has
several classes of registers. To choose a particular class of register, you can use the
x86_64 machine constraints described in Table 34.
Table 34 x86_64 Machine Constraints
Constraint
Description
a
a register (e.g., %al, %ax, %eax, %rax)
A
Specifies a or d registers. The d register holds the most significant bits and the a
register holds the least significant bits.
b
b register (e.g, %bl, %bx, %ebx, %rbx)
c
c register (e.g., %cl, %cx, %ecx, %rcx)
C
Not supported.
d
d register (e.g., %dl, %dx, %edx, %rdx)
D
di register (e.g., %dil, %di, %edi, %rdi)
e
Constant in range of 0xffffffff to 0x7fffffff
f
Not supported.
G
Floating point constant in range of 0.0 to 1.0.
I
Constant in range of 0 to 31 (e.g., for 32-bit shifts).
J
Constant in range of 0 to 63 (e.g., for 64-bit shifts)
K
Constant in range of 0to 127.
L
Constant in range of 0 to 65535.
M
Constant in range of 0 to 3 constant (e.g., shifts for lea instruction).
N
Constant in range of 0 to 255 (e.g., for out instruction).
q
Same as "r" simple constraint.
Q
Same as "r" simple constraint.
R
Same as "r" simple constraint.
S
si register (e.g., %sil, %si, %edi, %rsi)
User's Guide for x86-64 CPUs
Version 2017 | 183
C/C++ Inline Assembly and Intrinsics
Constraint
Description
t
Not supported.
u
Not supported.
x
XMM SSE register
y
Not supported.
Z
Constant in range of 0 to 0x7fffffff.
The following example uses the "x" or XMM register constraint to subtract c from b and
store the result in a.
double example11()
{
double a;
double b = 400.99;
double c = 300.98;
asm ( "subpd %2, %0;"
:"=x" (a)
: "0" (b), "x" (c)
);
return a;
}
The generated assembly for this example is this:
example11:
..Dcfb0:
pushq %rbp
..Dcfi0:
movq %rsp, %rbp
..Dcfi1:
..EN1:
## lineno: 4
movsd .C00128(%rip), %xmm1
movsd .C00130(%rip), %xmm2
movapd %xmm1, %xmm0
subpd %xmm2, %xmm0;
## lineno: 10
## lineno: 11
popq %rbp
ret
If a specified register is not available, the pgcc and pgc++ compilers issue an error
message.
15.2.7. Multiple Alternative Constraints
Sometimes a single instruction can take a variety of operand types. For example, the
x86-64 permits register-to-memory and memory-to-register operations. To allow this
flexibility in inline assembly, use multiple alternative constraints. An alternative is a series
of constraints for each operand.
To specify multiple alternatives, separate each alternative with a comma.
User's Guide for x86-64 CPUs
Version 2017 | 184
C/C++ Inline Assembly and Intrinsics
Table 35 Multiple Alternative Constraints
Constraint
Description
,
Separates each alternative for a particular operand.
?
Ignored
!
Ignored
The following example uses multiple alternatives for an add operation.
void example13()
{
int x=1;
int y=1;
asm( "addl %1, %0\n"
: "+ab,cd" (x)
: "db,cam" (y) );
}
The preceding example13() has two alternatives for each operand: "ab,cd" for the output
operand and "db,cam" for the input operand. Each operand must have the same number
of alternatives; however, each alternative can have any number of constraints (for
example, the output operand in example13() has two constraints for its second alternative
and the input operand has three for its second alternative).
The compiler first tries to satisfy the left-most alternative of the first operand (for
example, the output operand in example13()). When satisfying the operand, the compiler
starts with the left-most constraint. If the compiler cannot satisfy an alternative with
this constraint (for example, if the desired register is not available), it tries to use any
subsequent constraints. If the compiler runs out of constraints, it moves on to the next
alternative. If the compiler runs out of alternatives, it issues an error similar to the
one mentioned in example12(). If an alternative is found, the compiler uses the same
alternative for subsequent operands. For example, if the compiler chooses the "c" register
for the output operand in example13(), then it will use either the "a" or "m" constraint for
the input operand.
15.2.8. Constraint Modifiers
Characters that affect the compiler's interpretation of a constraint are known as
Constraint Modifiers. Two constraint modifiers, the "=" and the "+", were introduced in
Output Operands. The following table summarizes each constraint modifier.
Table 36 Constraint Modifier Characters
Constraint
Modifier
Description
=
This operand is write-only. It is valid for output operands only. If specified, the "=" must
appear as the first character of the constraint string.
+
This operand is both read and written by the instruction. It is valid for output operands
only. The output operand is initialized with its expression before the first instruction
in the asm statement. If specified, the "+" must appear as the first character of the
constraint string.
User's Guide for x86-64 CPUs
Version 2017 | 185
C/C++ Inline Assembly and Intrinsics
Constraint
Modifier
Description
&
A constraint or an alternative constraint, as defined in Multiple Alternative Constraints,
containing an "&" indicates that the output operand is an early clobber operand. This
type operand is an output operand that may be modified before the asm statement
finishes using all of the input operands. The compiler will not place this operand in a
register that may be used as an input operand or part of any memory address.
%
Ignored.
Characters following a "" up to the first comma (if present) are to be ignored in the
constraint.
*
The character that follows the "*" is to be ignored in the constraint.
The "=" and "+" modifiers apply to the operand, regardless of the number of alternatives
in the constraint string. For example, the "+" in the output operand of example13()
appears once and applies to both alternatives in the constraint string. The "&", "#", and
"*" modifiers apply only to the alternative in which they appear.
Normally, the compiler assumes that input operands are used before assigning results to
the output operands. This assumption lets the compiler reuse registers as needed inside
the asm statement. However, if the asm statement does not follow this convention,
the compiler may indiscriminately clobber a result register with an input operand. To
prevent this behavior, apply the early clobber "&" modifier. An example follows:
void example15()
{
int w=1;
int z;
asm( "movl $1, %0\n"
"addl %2, %0\n"
"movl %2, %1"
: "=a" (w), "=r" (z) : "r" (w) );
}
The previous code example presents an interesting ambiguity because "w" appears both
as an output and as an input operand. So, the value of "z" can be either 1 or 2, depending
on whether the compiler uses the same register for operand 0 and operand 2. The use of
constraint "r" for operand 2 allows the compiler to pick any general purpose register, so
it may (or may not) pick register "a" for operand 2. This ambiguity can be eliminated by
changing the constraint for operand 2 from "r" to "a" so the value of "z" will be 2, or by
adding an early clobber "&" modifier so that "z" will be 1. The following example shows
the same function with an early clobber "&" modifier:
void example16()
{
int w=1;
int z;
asm( "movl $1, %0\n"
"addl %2, %0\n"
"movl %2, %1"
: "=&a" (w), "=r" (z) : "r" (w) );
}
Adding the early clobber "&" forces the compiler not to use the "a" register for anything
other than operand 0. Operand 2 will therefore get its own register with its own copy of
"w". The result for "z" in example16() is 1.
User's Guide for x86-64 CPUs
Version 2017 | 186
C/C++ Inline Assembly and Intrinsics
15.3. Operand Aliases
Extended asm specifies operands in assembly strings with a percent '%' followed by the
operand number. For example, "%0" references operand 0 or the output item "=&a" (w)
in function example16() in the previous example. Extended asm also supports operand
aliasing, which allows use of a symbolic name instead of a number for specifying
operands, as illustrated in this example:
void example17()
{
int w=1, z=0;
asm( "movl $1, %[output1]\n"
"addl %[input], %[output1]\n"
"movl %[input], %[output2]"
: [output1] "=&a" (w), [output2] "=r"
(z)
: [input] "r" (w));
}
In example18(), "%0" and "%[output1]" both represent the output operand.
15.4. Assembly String Modifiers
Special character sequences in the assembly string affect the way the assembly is
generated by the compiler. For example, the "%" is an escape sequence for specifying an
operand, "%%" produces a percent for hard coded registers, and "\n" specifies a new
line. Table 37 summarizes these modifiers, known as Assembly String Modifiers.
Table 37 Assembly String Modifier Characters
Modifier
Description
\
Same as \ in printf format strings.
%*
Adds a '*' in the assembly string.
%%
Adds a '%' in the assembly string.
%A
Adds a '*' in front of an operand in the assembly string. (For example, %A0 adds a '*' in
front of operand 0 in the assembly output.)
%B
Produces the byte op code suffix for this operand. (For example, %b0 produces 'b' on
x86-64.)
%L
Produces the word op code suffix for this operand. (For example, %L0 produces 'l' on
x86-64.)
%P
If producing Position Independent Code (PIC), the compiler adds the PIC suffix for this
operand. (For example, %P0 produces @PLT on x86-64.)
%Q
Produces a quad word op code suffix for this operand if it is supported by the target.
Otherwise, it produces a word op code suffix. (For example, %Q0 produces 'q' on
x86-64.)
%S
Produces 's' suffix for this operand. (For example, %S0 produces 's' on x86-64.)
%T
Produces 't' suffix for this operand. (For example, %S0 produces 't' on x86-64.)
User's Guide for x86-64 CPUs
Version 2017 | 187
C/C++ Inline Assembly and Intrinsics
Modifier
Description
%W
Produces the half word op code suffix for this operand. (For example, %W0 produces 'w'
on x86-64.)
%a
Adds open and close parentheses ( ) around the operand.
%b
Produces the byte register name for an operand. (For example, if operand 0 is in
register 'a', then %b0 will produce '%al'.)
%c
Cuts the '$' character from an immediate operand.
%k
Produces the word register name for an operand. (For example, if operand 0 is in
register 'a', then %k0 will produce '%eax'.)
%q
Produces the quad word register name for an operand if the target supports quad word.
Otherwise, it produces a word register name. (For example, if operand 0 is in register
'a', then %q0 produces %rax on x86-64.)
%w
Produces the half word register name for an operand. (For example, if operand 0 is in
register 'a', then %w0 will produce '%ax'.)
%z
Produces an op code suffix based on the size of an operand. (For example, 'b' for byte,
'w' for half word, 'l' for word, and 'q' for quad word.)
%+ %C %D %F %O %X %f %h %l %n %s %y are not supported.
These modifiers begin with either a backslash "\" or a percent "%".
The modifiers that begin with a backslash "\" (e.g., "\n") have the same effect as they do
in a printf format string. The modifiers that are preceded with a "%" are used to modify a
particular operand.
These modifiers begin with either a backslash "\" or a percent "%" For example, "%b0"
means, "produce the byte or 8 bit version of operand 0". If operand 0 is a register, it will
produce a byte register such as %al, %bl, %cl, and so on.
15.5. Extended Asm Macros
As with traditional inline assembly, described in Inline Assembly, extended asm can be
used in a macro. For example, you can use the following macro to access the runtime
stack pointer.
#define GET_SP(x) \
asm("mov %%sp, %0": "=m" (##x):: "%sp" );
void example20()
{
void * stack_pointer;
GET_SP(stack_pointer);
}
User's Guide for x86-64 CPUs
Version 2017 | 188
C/C++ Inline Assembly and Intrinsics
The GET_SP macro assigns the value of the stack pointer to whatever is inserted in
its argument (for example, stack_pointer). Another "C" extension known as statement
expressions is used to write the GET_SP macro another way:
#define GET_SP2 ({ \
void *my_stack_ptr; \
asm("mov %%sp, %0": "=m" (my_stack_ptr) :: "%sp" ); \
my_stack_ptr; \
})
void example21()
{
void * stack_pointer = GET_SP2;
}
The statement expression allows a body of code to evaluate to a single value. This value
is specified as the last instruction in the statement expression. In this case, the value is
the result of the asm statement, my_stack_ptr. By writing an asm macro with a statement
expression, the asm result may be assigned directly to another variable (for example,
void * stack_pointer = GET_SP2) or included in a larger expression, such as: void *
stack_pointer = GET_SP2 - sizeof(long).
Which style of macro to use depends on the application. If the asm statement needs to
be a part of an expression, then a macro with a statement expression is a good approach.
Otherwise, a traditional macro, like GET_SP(x), will probably suffice.
15.6. Intrinsics
Inline intrinsic functions map to actual x86-64 machine instructions. Intrinsics are
inserted inline to avoid the overhead of a function call. The compiler has special
knowledge of intrinsics, so with use of intrinsics, better code may be generated as
compared to extended inline assembly code.
The PGI version 7.0 or higher compiler intrinsics library implements MMX, SSE, SS2,
SSE3, SSSE3, SSE4a, ABM, and AVX instructions. The intrinsic functions are available
to C and C++ programs on Linux and C programs on Windows. Unlike most functions
which are in libraries, intrinsics are implemented internally by the compiler. A program
can call the intrinsic functions from C/C++ source code after including the corresponding
header file.
The intrinsics are divided into header files as follows:
Table 38 Intrinsic Header File Organization
Instructions
Header File
Instructions
Header File
ABM
intrin.h
SSE2
emmintrin.h
AVX
immintrin.h
SSE3
pmmintrin.h
MMX
mmintrin.h
SSSE3
tmmintrin.h
SSE
xmmintrin.h
SSE4a
ammintrin.h
User's Guide for x86-64 CPUs
Version 2017 | 189
C/C++ Inline Assembly and Intrinsics
The following is a simple example program that calls XMM intrinsics.
#include <xmmintrin.h>
int main(){
__m128 __A, __B, result;
__A = _mm_set_ps(23.3, 43.7, 234.234, 98.746);
__B = _mm_set_ps(15.4, 34.3, 4.1, 8.6);
result = _mm_add_ps(__A,__B);
return 0;
}
User's Guide for x86-64 CPUs
Version 2017 | 190
Chapter 16.
CONTACT INFORMATION
You can contact PGI at:
20400 NW Amberwood Drive Suite 100
Beaverton, OR 97006
Or electronically using any of the following means:
Fax: +1-503-682-2637
Sales: [email protected]
WWW: http://www.pgroup.com
The PGI User Forum is monitored by members of the PGI engineering and support
teams as well as other PGI customers. The forum newsgroups may contain answers to
commonly asked questions. Log in to the PGI website to access the forum:
http://www.pgroup.com/userforum/index.php
Many questions and problems can be resolved by following instructions and the
information available at our frequently asked questions (FAQ) site:
http://www.pgroup.com/support/faq.htm
Submit technical support requests through the online form at:
https://www.pgroup.com/support/support_request.php
PGI documentation is available at http://www.pgroup.com/resources/docs.htm.
User's Guide for x86-64 CPUs
Version 2017 | 191
Notice
ALL NVIDIA DESIGN SPECIFICATIONS, REFERENCE BOARDS, FILES, DRAWINGS,
DIAGNOSTICS, LISTS, AND OTHER DOCUMENTS (TOGETHER AND SEPARATELY,
"MATERIALS") ARE BEING PROVIDED "AS IS." NVIDIA MAKES NO WARRANTIES,
EXPRESSED, IMPLIED, STATUTORY, OR OTHERWISE WITH RESPECT TO THE
MATERIALS, AND EXPRESSLY DISCLAIMS ALL IMPLIED WARRANTIES OF
NONINFRINGEMENT, MERCHANTABILITY, AND FITNESS FOR A PARTICULAR
PURPOSE.
Information furnished is believed to be accurate and reliable. However, NVIDIA
Corporation assumes no responsibility for the consequences of use of such
information or for any infringement of patents or other rights of third parties
that may result from its use. No license is granted by implication of otherwise
under any patent rights of NVIDIA Corporation. Specifications mentioned in this
publication are subject to change without notice. This publication supersedes and
replaces all other information previously supplied. NVIDIA Corporation products
are not authorized as critical components in life support devices or systems
without express written approval of NVIDIA Corporation.
Trademarks
PGI Workstation, PGI Server, PGI Accelerator, PGF95, PGF90, PGFORTRAN, and
PGI Unified Binary are trademarks; and PGI, PGHPF, PGF77, PGCC, PGC++, PGI
Visual Fortran, PVF, PGI CDK, Cluster Development Kit, PGPROF, PGDBG, and The
Portland Group are registered trademarks of NVIDIA Corporation in the U.S. and
other countries. Other company and product names may be trademarks of the
respective companies with which they are associated.
Copyright
© 2013–2017 NVIDIA Corporation. All rights reserved.
PGI Compilers and Tools
">