Using GNU Fortran

For gcc version 13.1.0

(GCC)

The gfortran team

Published by the Free Software Foundation
51 Franklin Street, Fifth Floor
Boston, MA 02110-1301, USA

Copyright c 1999-2023 Free Software Foundation, Inc.

Permission is granted to copy, distribute and/or modify this document under the terms of
the GNU Free Documentation License, Version 1.3 or any later version published by the
Free Software Foundation; with the Invariant Sections being “Funding Free Software”, the
Front-Cover Texts being (a) (see below), and with the Back-Cover Texts being (b) (see
below). A copy of the license is included in the section entitled “GNU Free Documentation
License”.
(a) The FSF’s Front-Cover Text is:
A GNU Manual
(b) The FSF’s Back-Cover Text is:
You have freedom to copy and modify this GNU Manual, like GNU software. Copies
published by the Free Software Foundation raise funds for GNU development.
 i

Short Contents
1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

Invoking GNU Fortran

2 GNU Fortran Command Options . . . . . . . . . . . . . . . . . . . . . . . . . 7
3 Runtime: Influencing runtime behavior with
environment variables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35

Language Reference
4 Compiler Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
5 Extensions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
6 Mixed-Language Programming . . . . . . . . . . . . . . . . . . . . . . . . . . 69
7 Coarray Programming . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
8 Intrinsic Procedures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111
9 Intrinsic Modules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293

Contributing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 301
GNU General Public License . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 303
GNU Free Documentation License . . . . . . . . . . . . . . . . . . . . . . . . . . 315
Funding Free Software . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 323
Option Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 325
Keyword Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 327
 iii

Table of Contents

1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1 About GNU Fortran . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 GNU Fortran and GCC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.3 Standards . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.3.1 Fortran 95 status . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.3.2 Fortran 2003 status . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.3.3 Fortran 2008 status . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.3.4 Fortran 2018 status . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

Part I: Invoking GNU Fortran . . . . . . . . . . . . . . . . . . . . . 5

2 GNU Fortran Command Options . . . . . . . . . . . . . . 7

2.1 Option summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.2 Options controlling Fortran dialect . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
2.3 Enable and customize preprocessing . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
2.4 Options to request or suppress errors and warnings . . . . . . . . . . . . . 17
2.5 Options for debugging your program or GNU Fortran . . . . . . . . . . 22
2.6 Options for directory search . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
2.7 Influencing the linking step . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
2.8 Influencing runtime behavior. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
2.9 Options for code generation conventions . . . . . . . . . . . . . . . . . . . . . . . 25
2.10 Options for interoperability with other languages . . . . . . . . . . . . . 32
2.11 Environment variables affecting gfortran. . . . . . . . . . . . . . . . . . . . . 33

3 Runtime: Influencing runtime behavior with

environment variables . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
3.1 TMPDIR—Directory for scratch files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
3.2 GFORTRAN_STDIN_UNIT—Unit number for standard input . . . . . . . 35
3.3 GFORTRAN_STDOUT_UNIT—Unit number for standard output . . . . 35
3.4 GFORTRAN_STDERR_UNIT—Unit number for standard error . . . . . . 35
3.5 GFORTRAN_UNBUFFERED_ALL—Do not buffer I/O on all units . . . . 35
3.6 GFORTRAN_UNBUFFERED_PRECONNECTED—Do not
buffer I/O on preconnected units . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
3.7 GFORTRAN_SHOW_LOCUS—Show location for runtime errors . . . . . . 35
3.8 GFORTRAN_OPTIONAL_PLUS—Print leading + where permitted . . 36
3.9 GFORTRAN_LIST_SEPARATOR—Separator for list output . . . . . . . . . 36
3.10 GFORTRAN_CONVERT_UNIT—Set conversion for unformatted I/O . . 36
3.11 GFORTRAN_ERROR_BACKTRACE—Show backtrace on run-time errors . . 37
3.12 GFORTRAN_FORMATTED_BUFFER_SIZE—Set
buffer size for formatted I/O. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
3.13 GFORTRAN_UNFORMATTED_BUFFER_SIZE—Set
buffer size for unformatted I/O . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
iv

Part II: Language Reference . . . . . . . . . . . . . . . . . . . . . . 39

4 Compiler Characteristics . . . . . . . . . . . . . . . . . . . . . . 41
4.1 KIND Type Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
4.2 Internal representation of LOGICAL variables . . . . . . . . . . . . . . . . . 41
4.3 Evaluation of logical expressions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
4.4 MAX and MIN intrinsics with REAL NaN arguments . . . . . . . . . . 42
4.5 Thread-safety of the runtime library . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
4.6 Data consistency and durability. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
4.7 Files opened without an explicit ACTION= specifier . . . . . . . . . . . 44
4.8 File operations on symbolic links. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
4.9 File format of unformatted sequential files . . . . . . . . . . . . . . . . . . . . . 44
4.10 Asynchronous I/O . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
4.11 Behavior on integer overflow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45

5 Extensions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
5.1 Extensions implemented in GNU Fortran . . . . . . . . . . . . . . . . . . . . . . 47
5.1.1 Old-style kind specifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
5.1.2 Old-style variable initialization . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
5.1.3 Extensions to namelist . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
5.1.4 X format descriptor without count field . . . . . . . . . . . . . . . . . . . . 49
5.1.5 Commas in FORMAT specifications . . . . . . . . . . . . . . . . . . . . . . . . . 49
5.1.6 Missing period in FORMAT specifications . . . . . . . . . . . . . . . . . . . 49
5.1.7 Default widths for F, G and I format descriptors . . . . . . . . . . . 49
5.1.8 I/O item lists . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
5.1.9 Q exponent-letter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
5.1.10 BOZ literal constants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
5.1.11 Real array indices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
5.1.12 Unary operators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
5.1.13 Implicitly convert LOGICAL and INTEGER values . . . . . . . . . . 50
5.1.14 Hollerith constants support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
5.1.15 Character conversion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
5.1.16 Cray pointers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
5.1.17 CONVERT specifier . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
5.1.18 OpenMP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
5.1.19 OpenACC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
5.1.20 Argument list functions %VAL, %REF and %LOC . . . . . . . . . . . . 55
5.1.21 Read/Write after EOF marker . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
5.1.22 STRUCTURE and RECORD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
5.1.23 UNION and MAP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
5.1.24 Type variants for integer intrinsics . . . . . . . . . . . . . . . . . . . . . . . 60
5.1.25 AUTOMATIC and STATIC attributes . . . . . . . . . . . . . . . . . . . . . . . . 61
5.1.26 Extended math intrinsics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
5.1.27 Form feed as whitespace . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
5.1.28 TYPE as an alias for PRINT . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
 v

5.1.29 %LOC as an rvalue . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63

5.1.30 .XOR. operator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
5.1.31 Bitwise logical operators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
5.1.32 Extended I/O specifiers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
5.1.33 Legacy PARAMETER statements . . . . . . . . . . . . . . . . . . . . . . . 65
5.1.34 Default exponents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
5.2 Extensions not implemented in GNU Fortran . . . . . . . . . . . . . . . . . . 65
5.2.1 ENCODE and DECODE statements . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
5.2.2 Variable FORMAT expressions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
5.2.3 Alternate complex function syntax . . . . . . . . . . . . . . . . . . . . . . . . 67
5.2.4 Volatile COMMON blocks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
5.2.5 OPEN( ... NAME=) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
5.2.6 Q edit descriptor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67

6 Mixed-Language Programming . . . . . . . . . . . . . . . 69
6.1 Interoperability with C . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
6.1.1 Intrinsic Types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
6.1.2 Derived Types and struct . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
6.1.3 Interoperable Global Variables . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
6.1.4 Interoperable Subroutines and Functions . . . . . . . . . . . . . . . . . . 70
6.1.5 Working with C Pointers. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72
6.1.6 Further Interoperability of Fortran with C . . . . . . . . . . . . . . . . 74
6.2 GNU Fortran Compiler Directives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
6.2.1 ATTRIBUTES directive . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
6.2.2 UNROLL directive . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
6.2.3 BUILTIN directive. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
6.2.4 IVDEP directive . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
6.2.5 VECTOR directive . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
6.2.6 NOVECTOR directive . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
6.3 Non-Fortran Main Program . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
6.3.1 _gfortran_set_args — Save command-line arguments . . . 77
6.3.2 _gfortran_set_options — Set library option flags . . . . . . . 77
6.3.3 _gfortran_set_convert — Set endian conversion . . . . . . . . 79
6.3.4 _gfortran_set_record_marker —
Set length of record markers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
6.3.5 _gfortran_set_fpe — Enable floating point exception traps . . 80
6.3.6 _gfortran_set_max_subrecord_
length — Set subrecord length . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
6.4 Naming and argument-passing conventions . . . . . . . . . . . . . . . . . . . . . 80
6.4.1 Naming conventions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
6.4.2 Argument passing conventions . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
vi

7 Coarray Programming . . . . . . . . . . . . . . . . . . . . . . . . . 85
7.1 Type and enum ABI Documentation . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
7.1.1 caf_token_t . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
7.1.2 caf_register_t . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
7.1.3 caf_deregister_t . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
7.1.4 caf_reference_t . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
7.1.5 caf_team_t . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
7.2 Function ABI Documentation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
7.2.1 _gfortran_caf_init — Initialiation function . . . . . . . . . . . . . 88
7.2.2 _gfortran_caf_finish — Finalization function . . . . . . . . . . 88
7.2.3 _gfortran_caf_this_image — Querying the image number . . 88
7.2.4 _gfortran_caf_num_images — Querying
the maximal number of images . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
7.2.5 _gfortran_caf_image_status —
Query the status of an image . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
7.2.6 _gfortran_caf_failed_images — Get an array
of the indexes of the failed images . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
7.2.7 _gfortran_caf_stopped_images — Get an array of
the indexes of the stopped images . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
7.2.8 _gfortran_caf_register — Registering coarrays . . . . . . . . 90
7.2.9 _gfortran_caf_deregister — Deregistering coarrays . . . . 91
7.2.10 _gfortran_caf_is_present — Query whether an allocatable or pointer
component in a derived type coarray is allocated . . . . . . . . . . . . . . . 92
7.2.11 _gfortran_caf_send — Sending data from
a local image to a remote image . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
7.2.12 _gfortran_caf_get — Getting data from a remote image . . 93
7.2.13 _gfortran_caf_sendget — Sending
data between remote images . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
7.2.14 _gfortran_caf_send_by_ref — Sending data from a local image to a
remote image with enhanced referencing options . . . . . . . . . . . . . . . 96
7.2.15 _gfortran_caf_get_by_ref — Getting data from a
remote image using enhanced references . . . . . . . . . . . . . . . . . . . . . . . 97
7.2.16 _gfortran_caf_sendget_by_ref — Sending data between remote
images using enhanced references on both sides . . . . . . . . . . . . . . . . 98
7.2.17 _gfortran_caf_lock — Locking a lock variable . . . . . . . . . 99
7.2.18 _gfortran_caf_lock — Unlocking a lock variable . . . . . . 100
7.2.19 _gfortran_caf_event_post — Post an event . . . . . . . . . . 100
7.2.20 _gfortran_caf_event_wait —
Wait that an event occurred . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
7.2.21 _gfortran_caf_event_query — Query event count . . . . 102
7.2.22 _gfortran_caf_sync_all — All-image barrier . . . . . . . . . 102
7.2.23 _gfortran_caf_sync_images —
Barrier for selected images . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
7.2.24 _gfortran_caf_sync_memory — Wait for
completion of segment-memory operations . . . . . . . . . . . . . . . . . . . . 103
 vii

7.2.25 _gfortran_caf_error_stop —
Error termination with exit code . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
7.2.26 _gfortran_caf_error_stop_str —
Error termination with string. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104
7.2.27 _gfortran_caf_fail_image — Mark the
image failed and end its execution . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104
7.2.28 _gfortran_caf_atomic_define —
Atomic variable assignment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104
7.2.29 _gfortran_caf_atomic_ref — Atomic variable reference . . 104
7.2.30 _gfortran_caf_atomic_cas — Atomic compare and swap . . 105
7.2.31 _gfortran_caf_atomic_op — Atomic operation . . . . . . . . 106
7.2.32 _gfortran_caf_co_broadcast —
Sending data to all images . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106
7.2.33 _gfortran_caf_co_max — Collective maximum reduction . . 107
7.2.34 _gfortran_caf_co_min — Collective minimum reduction . . 107
7.2.35 _gfortran_caf_co_sum — Collective summing reduction . . 108
7.2.36 _gfortran_caf_co_reduce — Generic collective reduction . . 108

8 Intrinsic Procedures . . . . . . . . . . . . . . . . . . . . . . . . . . 111

8.1 Introduction to intrinsic procedures . . . . . . . . . . . . . . . . . . . . . . . . . . . 111
8.2 ABORT — Abort the program . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111
8.3 ABS — Absolute value . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112
8.4 ACCESS — Checks file access modes . . . . . . . . . . . . . . . . . . . . . . . . . . . 112
8.5 ACHAR — Character in ASCII collating sequence . . . . . . . . . . . . . . . 113
8.6 ACOS — Arccosine function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114
8.7 ACOSD — Arccosine function, degrees . . . . . . . . . . . . . . . . . . . . . . . . . 114
8.8 ACOSH — Inverse hyperbolic cosine function . . . . . . . . . . . . . . . . . . . 115
8.9 ADJUSTL — Left adjust a string . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
8.10 ADJUSTR — Right adjust a string . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
8.11 AIMAG — Imaginary part of complex number . . . . . . . . . . . . . . . . . 117
8.12 AINT — Truncate to a whole number . . . . . . . . . . . . . . . . . . . . . . . . 117
8.13 ALARM — Execute a routine after a given delay . . . . . . . . . . . . . . . 118
8.14 ALL — All values in MASK along DIM are true . . . . . . . . . . . . . 119
8.15 ALLOCATED — Status of an allocatable entity . . . . . . . . . . . . . . . . . 120
8.16 AND — Bitwise logical AND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120
8.17 ANINT — Nearest whole number . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
8.18 ANY — Any value in MASK along DIM is true . . . . . . . . . . . . . . . 122
8.19 ASIN — Arcsine function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122
8.20 ASIND — Arcsine function, degrees . . . . . . . . . . . . . . . . . . . . . . . . . . 123
8.21 ASINH — Inverse hyperbolic sine function . . . . . . . . . . . . . . . . . . . . 124
8.22 ASSOCIATED — Status of a pointer or pointer/target pair. . . . . 124
8.23 ATAN — Arctangent function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
8.24 ATAND — Arctangent function, degrees . . . . . . . . . . . . . . . . . . . . . . . 126
8.25 ATAN2 — Arctangent function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
8.26 ATAN2D — Arctangent function, degrees. . . . . . . . . . . . . . . . . . . . . . 128
8.27 ATANH — Inverse hyperbolic tangent function . . . . . . . . . . . . . . . . 129
viii

8.28 ATOMIC_ADD — Atomic ADD operation . . . . . . . . . . . . . . . . . . . . . . 129

8.29 ATOMIC_AND — Atomic bitwise AND operation. . . . . . . . . . . . . . . 130
8.30 ATOMIC_CAS — Atomic compare and swap . . . . . . . . . . . . . . . . . . . 131
8.31 ATOMIC_DEFINE — Setting a variable atomically . . . . . . . . . . . . . 132
8.32 ATOMIC_FETCH_ADD — Atomic ADD operation with prior fetch . . 132
8.33 ATOMIC_FETCH_AND — Atomic bitwise AND
operation with prior fetch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133
8.34 ATOMIC_FETCH_OR — Atomic bitwise OR
operation with prior fetch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
8.35 ATOMIC_FETCH_XOR — Atomic bitwise XOR
operation with prior fetch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
8.36 ATOMIC_OR — Atomic bitwise OR operation . . . . . . . . . . . . . . . . . 135
8.37 ATOMIC_REF — Obtaining the value of a variable atomically . . 136
8.38 ATOMIC_XOR — Atomic bitwise OR operation . . . . . . . . . . . . . . . . 137
8.39 BACKTRACE — Show a backtrace . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138
8.40 BESSEL_J0 — Bessel function of the first kind of order 0 . . . . . 138
8.41 BESSEL_J1 — Bessel function of the first kind of order 1 . . . . . 139
8.42 BESSEL_JN — Bessel function of the first kind . . . . . . . . . . . . . . . 139
8.43 BESSEL_Y0 — Bessel function of the second kind of order 0 . . 140
8.44 BESSEL_Y1 — Bessel function of the second kind of order 1 . . 140
8.45 BESSEL_YN — Bessel function of the second kind . . . . . . . . . . . . . 141
8.46 BGE — Bitwise greater than or equal to . . . . . . . . . . . . . . . . . . . . . . 142
8.47 BGT — Bitwise greater than . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142
8.48 BIT_SIZE — Bit size inquiry function . . . . . . . . . . . . . . . . . . . . . . . . 142
8.49 BLE — Bitwise less than or equal to . . . . . . . . . . . . . . . . . . . . . . . . . 143
8.50 BLT — Bitwise less than . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143
8.51 BTEST — Bit test function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144
8.52 C_ASSOCIATED — Status of a C pointer . . . . . . . . . . . . . . . . . . . . . . 144
8.53 C_F_POINTER — Convert C into Fortran pointer . . . . . . . . . . . . . 145
8.54 C_F_PROCPOINTER — Convert C into Fortran procedure pointer . . 146
8.55 C_FUNLOC — Obtain the C address of a procedure . . . . . . . . . . . 146
8.56 C_LOC — Obtain the C address of an object . . . . . . . . . . . . . . . . . 147
8.57 C_SIZEOF — Size in bytes of an expression . . . . . . . . . . . . . . . . . . . 148
8.58 CEILING — Integer ceiling function . . . . . . . . . . . . . . . . . . . . . . . . . . 149
8.59 CHAR — Character conversion function . . . . . . . . . . . . . . . . . . . . . . . 149
8.60 CHDIR — Change working directory . . . . . . . . . . . . . . . . . . . . . . . . . . 150
8.61 CHMOD — Change access permissions of files . . . . . . . . . . . . . . . . . . 151
8.62 CMPLX — Complex conversion function . . . . . . . . . . . . . . . . . . . . . . . 152
8.63 CO_BROADCAST — Copy a value to all
images the current set of images . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152
8.64 CO_MAX — Maximal value on the current set of images . . . . . . . 153
8.65 CO_MIN — Minimal value on the current set of images . . . . . . . . 154
8.66 CO_REDUCE — Reduction of values on the current set of images . . 155
8.67 CO_SUM — Sum of values on the current set of images . . . . . . . . 156
8.68 COMMAND_ARGUMENT_COUNT — Get number
of command line arguments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157
8.69 COMPILER_OPTIONS — Options passed to the compiler . . . . . . . 157
 ix

8.70 COMPILER_VERSION — Compiler version string . . . . . . . . . . . . . . . 158

8.71 COMPLEX — Complex conversion function . . . . . . . . . . . . . . . . . . . . 158
8.72 CONJG — Complex conjugate function . . . . . . . . . . . . . . . . . . . . . . . 159
8.73 COS — Cosine function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160
8.74 COSD — Cosine function, degrees . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160
8.75 COSH — Hyperbolic cosine function . . . . . . . . . . . . . . . . . . . . . . . . . . 161
8.76 COTAN — Cotangent function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162
8.77 COTAND — Cotangent function, degrees . . . . . . . . . . . . . . . . . . . . . . 162
8.78 COUNT — Count function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163
8.79 CPU_TIME — CPU elapsed time in seconds . . . . . . . . . . . . . . . . . . . 164
8.80 CSHIFT — Circular shift elements of an array . . . . . . . . . . . . . . . . 164
8.81 CTIME — Convert a time into a string . . . . . . . . . . . . . . . . . . . . . . . 165
8.82 DATE_AND_TIME — Date and time subroutine . . . . . . . . . . . . . . . . 166
8.83 DBLE — Double conversion function . . . . . . . . . . . . . . . . . . . . . . . . . 167
8.84 DCMPLX — Double complex conversion function . . . . . . . . . . . . . . 167
8.85 DIGITS — Significant binary digits function. . . . . . . . . . . . . . . . . . 168
8.86 DIM — Positive difference . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
8.87 DOT_PRODUCT — Dot product function . . . . . . . . . . . . . . . . . . . . . . . 169
8.88 DPROD — Double product function . . . . . . . . . . . . . . . . . . . . . . . . . . . 170
8.89 DREAL — Double real part function . . . . . . . . . . . . . . . . . . . . . . . . . . 171
8.90 DSHIFTL — Combined left shift . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171
8.91 DSHIFTR — Combined right shift . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172
8.92 DTIME — Execution time subroutine (or function) . . . . . . . . . . . . 172
8.93 EOSHIFT — End-off shift elements of an array . . . . . . . . . . . . . . . . 173
8.94 EPSILON — Epsilon function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174
8.95 ERF — Error function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175
8.96 ERFC — Error function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175
8.97 ERFC_SCALED — Error function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 176
8.98 ETIME — Execution time subroutine (or function) . . . . . . . . . . . . 176
8.99 EVENT_QUERY — Query whether a coarray event has occurred . . 177
8.100 EXECUTE_COMMAND_LINE — Execute a shell command . . . . . . . 178
8.101 EXIT — Exit the program with status. . . . . . . . . . . . . . . . . . . . . . . 179
8.102 EXP — Exponential function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179
8.103 EXPONENT — Exponent function . . . . . . . . . . . . . . . . . . . . . . . . . . . . 180
8.104 EXTENDS_TYPE_OF — Query dynamic type for extension. . . . . 180
8.105 FDATE — Get the current time as a string . . . . . . . . . . . . . . . . . . 181
8.106 FGET — Read a single character in stream mode from stdin . . 182
8.107 FGETC — Read a single character in stream mode . . . . . . . . . . . 183
8.108 FINDLOC — Search an array for a value . . . . . . . . . . . . . . . . . . . . . 183
8.109 FLOOR — Integer floor function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184
8.110 FLUSH — Flush I/O unit(s) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185
8.111 FNUM — File number function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 186
8.112 FPUT — Write a single character in stream mode to stdout . . 186
8.113 FPUTC — Write a single character in stream mode . . . . . . . . . . 187
8.114 FRACTION — Fractional part of the model representation . . . . 188
8.115 FREE — Frees memory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 188
8.116 FSEEK — Low level file positioning subroutine . . . . . . . . . . . . . . 189
x

8.117 FSTAT — Get file status . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 190

8.118 FTELL — Current stream position . . . . . . . . . . . . . . . . . . . . . . . . . . 191
8.119 GAMMA — Gamma function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191
8.120 GERROR — Get last system error message . . . . . . . . . . . . . . . . . . . 192
8.121 GETARG — Get command line arguments . . . . . . . . . . . . . . . . . . . . 192
8.122 GET_COMMAND — Get the entire command line. . . . . . . . . . . . . . . 193
8.123 GET_COMMAND_ARGUMENT — Get command line arguments . . . 194
8.124 GETCWD — Get current working directory . . . . . . . . . . . . . . . . . . . 195
8.125 GETENV — Get an environmental variable . . . . . . . . . . . . . . . . . . . 195
8.126 GET_ENVIRONMENT_VARIABLE — Get an environmental variable . . 196
8.127 GETGID — Group ID function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197
8.128 GETLOG — Get login name . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197
8.129 GETPID — Process ID function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 198
8.130 GETUID — User ID function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 198
8.131 GMTIME — Convert time to GMT info . . . . . . . . . . . . . . . . . . . . . . 198
8.132 HOSTNM — Get system host name . . . . . . . . . . . . . . . . . . . . . . . . . . . 199
8.133 HUGE — Largest number of a kind . . . . . . . . . . . . . . . . . . . . . . . . . . 200
8.134 HYPOT — Euclidean distance function . . . . . . . . . . . . . . . . . . . . . . . 200
8.135 IACHAR — Code in ASCII collating sequence . . . . . . . . . . . . . . . . 201
8.136 IALL — Bitwise AND of array elements . . . . . . . . . . . . . . . . . . . . 201
8.137 IAND — Bitwise logical and. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 202
8.138 IANY — Bitwise OR of array elements . . . . . . . . . . . . . . . . . . . . . . 203
8.139 IARGC — Get the number of command line arguments . . . . . . 204
8.140 IBCLR — Clear bit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204
8.141 IBITS — Bit extraction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205
8.142 IBSET — Set bit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 206
8.143 ICHAR — Character-to-integer conversion function . . . . . . . . . . 206
8.144 IDATE — Get current local time subroutine (day/month/year) . . 207
8.145 IEOR — Bitwise logical exclusive or . . . . . . . . . . . . . . . . . . . . . . . . . 208
8.146 IERRNO — Get the last system error number . . . . . . . . . . . . . . . . 209
8.147 IMAGE_INDEX — Function that converts a
cosubscript to an image index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209
8.148 INDEX — Position of a substring within a string . . . . . . . . . . . . 209
8.149 INT — Convert to integer type . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210
8.150 INT2 — Convert to 16-bit integer type. . . . . . . . . . . . . . . . . . . . . . 211
8.151 INT8 — Convert to 64-bit integer type. . . . . . . . . . . . . . . . . . . . . . 211
8.152 IOR — Bitwise logical or . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212
8.153 IPARITY — Bitwise XOR of array elements . . . . . . . . . . . . . . . . . 212
8.154 IRAND — Integer pseudo-random number . . . . . . . . . . . . . . . . . . . 213
8.155 IS_CONTIGUOUS — Test whether an array is contiguous . . . . . 214
8.156 IS_IOSTAT_END — Test for end-of-file value . . . . . . . . . . . . . . . . . 214
8.157 IS_IOSTAT_EOR — Test for end-of-record value . . . . . . . . . . . . . 215
8.158 ISATTY — Whether a unit is a terminal device . . . . . . . . . . . . . . 215
8.159 ISHFT — Shift bits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 216
8.160 ISHFTC — Shift bits circularly . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217
8.161 ISNAN — Test for a NaN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217
 xi

8.162 ITIME — Get current local time

subroutine (hour/minutes/seconds) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 218
8.163 KILL — Send a signal to a process . . . . . . . . . . . . . . . . . . . . . . . . . . 218
8.164 KIND — Kind of an entity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 219
8.165 LBOUND — Lower dimension bounds of an array . . . . . . . . . . . . . 219
8.166 LCOBOUND — Lower codimension bounds of an array . . . . . . . . 220
8.167 LEADZ — Number of leading zero bits of an integer . . . . . . . . . 220
8.168 LEN — Length of a character entity . . . . . . . . . . . . . . . . . . . . . . . . . 221
8.169 LEN_TRIM — Length of a character entity
without trailing blank characters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 222
8.170 LGE — Lexical greater than or equal . . . . . . . . . . . . . . . . . . . . . . . . 222
8.171 LGT — Lexical greater than . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223
8.172 LINK — Create a hard link . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223
8.173 LLE — Lexical less than or equal . . . . . . . . . . . . . . . . . . . . . . . . . . . 224
8.174 LLT — Lexical less than . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 225
8.175 LNBLNK — Index of the last non-blank character in a string . . 225
8.176 LOC — Returns the address of a variable . . . . . . . . . . . . . . . . . . . . 226
8.177 LOG — Natural logarithm function . . . . . . . . . . . . . . . . . . . . . . . . . . 226
8.178 LOG10 — Base 10 logarithm function . . . . . . . . . . . . . . . . . . . . . . . 227
8.179 LOG_GAMMA — Logarithm of the Gamma function . . . . . . . . . . . 227
8.180 LOGICAL — Convert to logical type . . . . . . . . . . . . . . . . . . . . . . . . . 228
8.181 LSHIFT — Left shift bits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 228
8.182 LSTAT — Get file status . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229
8.183 LTIME — Convert time to local time info . . . . . . . . . . . . . . . . . . . 230
8.184 MALLOC — Allocate dynamic memory . . . . . . . . . . . . . . . . . . . . . . . 230
8.185 MASKL — Left justified mask. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231
8.186 MASKR — Right justified mask . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 232
8.187 MATMUL — matrix multiplication . . . . . . . . . . . . . . . . . . . . . . . . . . . . 232
8.188 MAX — Maximum value of an argument list . . . . . . . . . . . . . . . . . 232
8.189 MAXEXPONENT — Maximum exponent of a real kind . . . . . . . . . 233
8.190 MAXLOC — Location of the maximum value within an array . . 234
8.191 MAXVAL — Maximum value of an array . . . . . . . . . . . . . . . . . . . . . 235
8.192 MCLOCK — Time function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 235
8.193 MCLOCK8 — Time function (64-bit) . . . . . . . . . . . . . . . . . . . . . . . . . 236
8.194 MERGE — Merge variables. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 236
8.195 MERGE_BITS — Merge of bits under mask . . . . . . . . . . . . . . . . . . . 237
8.196 MIN — Minimum value of an argument list . . . . . . . . . . . . . . . . . 237
8.197 MINEXPONENT — Minimum exponent of a real kind . . . . . . . . . . 238
8.198 MINLOC — Location of the minimum value within an array . . 238
8.199 MINVAL — Minimum value of an array . . . . . . . . . . . . . . . . . . . . . . 239
8.200 MOD — Remainder function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 240
8.201 MODULO — Modulo function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241
8.202 MOVE_ALLOC — Move allocation from one object to another . . 242
8.203 MVBITS — Move bits from one integer to another . . . . . . . . . . . 242
8.204 NEAREST — Nearest representable number . . . . . . . . . . . . . . . . . . 243
8.205 NEW_LINE — New line character . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243
8.206 NINT — Nearest whole number . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 244
xii

8.207 NORM2 — Euclidean vector norms . . . . . . . . . . . . . . . . . . . . . . . . . . . 245

8.208 NOT — Logical negation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245
8.209 NULL — Function that returns an disassociated pointer . . . . . 246
8.210 NUM_IMAGES — Function that returns the number of images . . 246
8.211 OR — Bitwise logical OR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 247
8.212 PACK — Pack an array into an array of rank one . . . . . . . . . . . . 248
8.213 PARITY — Reduction with exclusive OR . . . . . . . . . . . . . . . . . . . . 249
8.214 PERROR — Print system error message . . . . . . . . . . . . . . . . . . . . . . 249
8.215 POPCNT — Number of bits set . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 250
8.216 POPPAR — Parity of the number of bits set. . . . . . . . . . . . . . . . . . 250
8.217 PRECISION — Decimal precision of a real kind . . . . . . . . . . . . . . 251
8.218 PRESENT — Determine whether an optional
dummy argument is specified . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 251
8.219 PRODUCT — Product of array elements . . . . . . . . . . . . . . . . . . . . . . 252
8.220 RADIX — Base of a model number . . . . . . . . . . . . . . . . . . . . . . . . . . 252
8.221 RAN — Real pseudo-random number . . . . . . . . . . . . . . . . . . . . . . . . 253
8.222 RAND — Real pseudo-random number . . . . . . . . . . . . . . . . . . . . . . . 253
8.223 RANDOM_INIT — Initialize a pseudo-random number generator . . 254
8.224 RANDOM_NUMBER — Pseudo-random number . . . . . . . . . . . . . . . . . 255
8.225 RANDOM_SEED — Initialize a pseudo-random number sequence . . 255
8.226 RANGE — Decimal exponent range . . . . . . . . . . . . . . . . . . . . . . . . . . 256
8.227 RANK — Rank of a data object. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 257
8.228 REAL — Convert to real type . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 257
8.229 RENAME — Rename a file . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 258
8.230 REPEAT — Repeated string concatenation . . . . . . . . . . . . . . . . . . . 258
8.231 RESHAPE — Function to reshape an array . . . . . . . . . . . . . . . . . . . 259
8.232 RRSPACING — Reciprocal of the relative spacing . . . . . . . . . . . . 259
8.233 RSHIFT — Right shift bits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 260
8.234 SAME_TYPE_AS — Query dynamic types for equality. . . . . . . . . 260
8.235 SCALE — Scale a real value . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 261
8.236 SCAN — Scan a string for the presence of a set of characters . . 261
8.237 SECNDS — Time function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 262
8.238 SECOND — CPU time function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 263
8.239 SELECTED_CHAR_KIND — Choose character kind . . . . . . . . . . . . . 263
8.240 SELECTED_INT_KIND — Choose integer kind . . . . . . . . . . . . . . . . 264
8.241 SELECTED_REAL_KIND — Choose real kind . . . . . . . . . . . . . . . . . . 264
8.242 SET_EXPONENT — Set the exponent of the model . . . . . . . . . . . . 265
8.243 SHAPE — Determine the shape of an array . . . . . . . . . . . . . . . . . . 266
8.244 SHIFTA — Right shift with fill . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267
8.245 SHIFTL — Left shift . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267
8.246 SHIFTR — Right shift . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 268
8.247 SIGN — Sign copying function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 268
8.248 SIGNAL — Signal handling subroutine (or function) . . . . . . . . . 269
8.249 SIN — Sine function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 269
8.250 SIND — Sine function, degrees . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 270
8.251 SINH — Hyperbolic sine function . . . . . . . . . . . . . . . . . . . . . . . . . . . 271
8.252 SIZE — Determine the size of an array . . . . . . . . . . . . . . . . . . . . . 271
 xiii

8.253 SIZEOF — Size in bytes of an expression . . . . . . . . . . . . . . . . . . . . 272

8.254 SLEEP — Sleep for the specified number of seconds . . . . . . . . . 273
8.255 SPACING — Smallest distance between
two numbers of a given type . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 273
8.256 SPREAD — Add a dimension to an array . . . . . . . . . . . . . . . . . . . . 273
8.257 SQRT — Square-root function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 274
8.258 SRAND — Reinitialize the random number generator. . . . . . . . . 275
8.259 STAT — Get file status . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 275
8.260 STORAGE_SIZE — Storage size in bits . . . . . . . . . . . . . . . . . . . . . . . 276
8.261 SUM — Sum of array elements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 277
8.262 SYMLNK — Create a symbolic link . . . . . . . . . . . . . . . . . . . . . . . . . . . 278
8.263 SYSTEM — Execute a shell command . . . . . . . . . . . . . . . . . . . . . . . . 278
8.264 SYSTEM_CLOCK — Time function . . . . . . . . . . . . . . . . . . . . . . . . . . . . 279
8.265 TAN — Tangent function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 280
8.266 TAND — Tangent function, degrees . . . . . . . . . . . . . . . . . . . . . . . . . . 280
8.267 TANH — Hyperbolic tangent function . . . . . . . . . . . . . . . . . . . . . . . 281
8.268 THIS_IMAGE — Function that returns the
cosubscript index of this image . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 281
8.269 TIME — Time function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 282
8.270 TIME8 — Time function (64-bit) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 283
8.271 TINY — Smallest positive number of a real kind . . . . . . . . . . . . 284
8.272 TRAILZ — Number of trailing zero bits of an integer . . . . . . . . 284
8.273 TRANSFER — Transfer bit patterns . . . . . . . . . . . . . . . . . . . . . . . . . . 285
8.274 TRANSPOSE — Transpose an array of rank two . . . . . . . . . . . . . . 285
8.275 TRIM — Remove trailing blank characters of a string . . . . . . . . 286
8.276 TTYNAM — Get the name of a terminal device . . . . . . . . . . . . . . . 286
8.277 UBOUND — Upper dimension bounds of an array . . . . . . . . . . . . . 287
8.278 UCOBOUND — Upper codimension bounds of an array . . . . . . . . 287
8.279 UMASK — Set the file creation mask . . . . . . . . . . . . . . . . . . . . . . . . . 288
8.280 UNLINK — Remove a file from the file system . . . . . . . . . . . . . . . 288
8.281 UNPACK — Unpack an array of rank one into an array . . . . . . . 289
8.282 VERIFY — Scan a string for characters not a given set . . . . . . 289
8.283 XOR — Bitwise logical exclusive OR. . . . . . . . . . . . . . . . . . . . . . . . . 290

9 Intrinsic Modules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293

9.1 ISO_FORTRAN_ENV . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293
9.2 ISO_C_BINDING . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 295
9.3 IEEE modules: IEEE_EXCEPTIONS,
IEEE_ARITHMETIC, and IEEE_FEATURES . . . . . . . . . . . . . . . . . . . . . . . . . . 296
9.4 OpenMP Modules OMP_LIB and OMP_LIB_KINDS . . . . . . . . . . . . . . 297
9.5 OpenACC Module OPENACC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 299

Contributing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 301
Contributors to GNU Fortran. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 301
Projects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 302
xiv

GNU General Public License . . . . . . . . . . . . . . . . . . . . 303

GNU Free Documentation License . . . . . . . . . . . . . . 315

ADDENDUM: How to use this License for your documents . . . . . . . . . 322

Funding Free Software . . . . . . . . . . . . . . . . . . . . . . . . . . . 323

Option Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 325

Keyword Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 327

 1

1 Introduction

This manual documents the use of gfortran, the GNU Fortran compiler. You can find in
this manual how to invoke gfortran, as well as its features and incompatibilities.

1.1 About GNU Fortran

The GNU Fortran compiler is the successor to g77, the Fortran 77 front end included in
GCC prior to version 4 (released in 2005). While it is backward-compatible with most g77
extensions and command-line options, gfortran is a completely new implemention designed
to support more modern dialects of Fortran. GNU Fortran implements the Fortran 77, 90
and 95 standards completely, most of the Fortran 2003 and 2008 standards, and some
features from the 2018 standard. It also implements several extensions including OpenMP
and OpenACC support for parallel programming.
The GNU Fortran compiler passes the NIST Fortran 77 Test Suite (http://www.
fortran-2000.com/ArnaudRecipes/fcvs21_f95.html), and produces acceptable results
on the LAPACK Test Suite (https://www.netlib.org/lapack/faq.html). It also
provides respectable performance on the Polyhedron Fortran compiler benchmarks
(https://polyhedron.com/?page_id=175) and the Livermore Fortran Kernels test
(https://www.netlib.org/benchmark/livermore). It has been used to compile a
number of large real-world programs, including the HARMONIE and HIRLAM weather
forecasting code (http://hirlam.org/) and the Tonto quantum chemistry package
(https://github.com/dylan-jayatilaka/tonto); see https://gcc.gnu.org/wiki/
GfortranApps for an extended list.
GNU Fortran provides the following functionality:
• Read a program, stored in a file and containing source code instructions written in
Fortran 77.
• Translate the program into instructions a computer can carry out more quickly than
it takes to translate the original Fortran instructions. The result after compilation of
a program is machine code, which is efficiently translated and processed by a machine
such as your computer. Humans usually are not as good writing machine code as they
are at writing Fortran (or C++, Ada, or Java), because it is easy to make tiny mistakes
writing machine code.
• Provide information about the reasons why the compiler may be unable to create a
binary from the source code, for example if the source code is flawed. The Fortran
language standards require that the compiler can point out mistakes in your code. An
incorrect usage of the language causes an error message.
The compiler also attempts to diagnose cases where your program contains a correct
usage of the language, but instructs the computer to do something questionable. This
kind of diagnostic message is called a warning message.
• Provide optional information about the translation passes from the source code to
machine code. This can help you to find the cause of certain bugs which may not be
obvious in the source code, but may be more easily found at a lower level compiler
output. It also helps developers to find bugs in the compiler itself.
2 The GNU Fortran Compiler

• Provide information in the generated machine code that can make it easier to find bugs
in the program (using a debugging tool, called a debugger, such as the GNU Debugger
gdb).
• Locate and gather machine code already generated to perform actions requested by
statements in the program. This machine code is organized into modules and is located
and linked to the user program.
The GNU Fortran compiler consists of several components:
• A version of the gcc command (which also might be installed as the system’s cc com-
mand) that also understands and accepts Fortran source code. The gcc command is
the driver program for all the languages in the GNU Compiler Collection (GCC); With
gcc, you can compile the source code of any language for which a front end is available
in GCC.
• The gfortran command itself, which also might be installed as the system’s f95 com-
mand. gfortran is just another driver program, but specifically for the Fortran com-
piler only. The primary difference between the gcc and gfortran commands is that
the latter automatically links the correct libraries to your program.
• A collection of run-time libraries. These libraries contain the machine code needed
to support capabilities of the Fortran language that are not directly provided by the
machine code generated by the gfortran compilation phase, such as intrinsic functions
and subroutines, and routines for interaction with files and the operating system.
• The Fortran compiler itself, (f951). This is the GNU Fortran parser and code generator,
linked to and interfaced with the GCC backend library. f951 “translates” the source
code to assembler code. You would typically not use this program directly; instead,
the gcc or gfortran driver programs call it for you.

1.2 GNU Fortran and GCC

GNU Fortran is a part of GCC, the GNU Compiler Collection. GCC consists of a collec-
tion of front ends for various languages, which translate the source code into a language-
independent form called GENERIC. This is then processed by a common middle end which
provides optimization, and then passed to one of a collection of back ends which generate
code for different computer architectures and operating systems.
Functionally, this is implemented with a driver program (gcc) which provides the
command-line interface for the compiler. It calls the relevant compiler front-end program
(e.g., f951 for Fortran) for each file in the source code, and then calls the assembler and
linker as appropriate to produce the compiled output. In a copy of GCC that has been
compiled with Fortran language support enabled, gcc recognizes files with .f, .for, .ftn,
.f90, .f95, .f03 and .f08 extensions as Fortran source code, and compiles it accordingly.
A gfortran driver program is also provided, which is identical to gcc except that it
automatically links the Fortran runtime libraries into the compiled program.
Source files with .f, .for, .fpp, .ftn, .F, .FOR, .FPP, and .FTN extensions are treated
as fixed form. Source files with .f90, .f95, .f03, .f08, .F90, .F95, .F03 and .F08 ex-
tensions are treated as free form. The capitalized versions of either form are run through
preprocessing. Source files with the lower case .fpp extension are also run through prepro-
cessing.
Chapter 1: Introduction 3

This manual specifically documents the Fortran front end, which handles the program-
ming language’s syntax and semantics. The aspects of GCC that relate to the optimization
passes and the back-end code generation are documented in the GCC manual; see Section
“Introduction” in Using the GNU Compiler Collection (GCC). The two manuals together
provide a complete reference for the GNU Fortran compiler.

1.3 Standards
Fortran is developed by the Working Group 5 of Sub-Committee 22 of the Joint Technical
Committee 1 of the International Organization for Standardization and the International
Electrotechnical Commission (IEC). This group is known as WG5 (http://www.nag.co.
uk/sc22wg5/). Official Fortran standard documents are available for purchase from ISO; a
collection of free documents (typically final drafts) are also available on the wiki (https://
gcc.gnu.org/wiki/GFortranStandards).
The GNU Fortran compiler implements ISO/IEC 1539:1997 (Fortran 95). As such, it
can also compile essentially all standard-compliant Fortran 90 and Fortran 77 programs. It
also supports the ISO/IEC TR-15581 enhancements to allocatable arrays.
GNU Fortran also supports almost all of ISO/IEC 1539-1:2004 (Fortran 2003) and
ISO/IEC 1539-1:2010 (Fortran 2008). It has partial support for features introduced in
ISO/IEC 1539:2018 (Fortran 2018), the most recent version of the Fortran language stan-
dard, including full support for the Technical Specification Further Interoperability of
Fortran with C (ISO/IEC TS 29113:2012). More details on support for these standards
can be found in the following sections of the documentation.
Additionally, the GNU Fortran compilers supports the OpenMP specification (version
4.5 and partial support of the features of the 5.0 version, https://openmp.org/
specifications/). There also is support for the OpenACC specification (targeting version
2.6, https://www.openacc.org/). See https://gcc.gnu.org/wiki/OpenACC for more in-
formation.

1.3.1 Fortran 95 status

The Fortran 95 standard specifies in Part 2 (ISO/IEC 1539-2:2000) varying length character
strings. While GNU Fortran currently does not support such strings directly, there exist
two Fortran implementations for them, which work with GNU Fortran. One can be found
at http://user.astro.wisc.edu/~townsend/static.php?ref=iso-varying-string.
Deferred-length character strings of Fortran 2003 supports part of the features of ISO_
VARYING_STRING and should be considered as replacement. (Namely, allocatable or pointers
of the type character(len=:).)
Part 3 of the Fortran 95 standard (ISO/IEC 1539-3:1998) defines Conditional Compila-
tion, which is not widely used and not directly supported by the GNU Fortran compiler.
You can use the program coco to preprocess such files (http://www.daniellnagle.com/
coco.html).

1.3.2 Fortran 2003 status

GNU Fortran implements the Fortran 2003 (ISO/IEC 1539-1:2004) standard except for fi-
nalization support, which is incomplete. See the wiki page (https://gcc.gnu.org/wiki/
4 The GNU Fortran Compiler

Fortran2003) for a full list of new features introduced by Fortran 2003 and their imple-
mentation status.

1.3.3 Fortran 2008 status

The GNU Fortran compiler supports almost all features of Fortran 2008; the wiki (https://
gcc.gnu.org/wiki/Fortran2008Status) has some information about the current imple-
mentation status. In particular, the following are not yet supported:
• DO CONCURRENT and FORALL do not recognize a type-spec in the loop header.
• The change to permit any constant expression in subscripts and nested implied-do
limits in a DATA statement has not been implemented.

1.3.4 Fortran 2018 status

Fortran 2018 (ISO/IEC 1539:2018) is the most recent version of the Fortran language stan-
dard. GNU Fortran implements some of the new features of this standard:
• All Fortran 2018 features derived from ISO/IEC TS 29113:2012, “Further Interoper-
ability of Fortran with C”, are supported by GNU Fortran. This includes assumed-type
and assumed-rank objects and the SELECT RANK construct as well as the parts relating
to BIND(C) functions. See also Section 6.1.6 [Further Interoperability of Fortran with
C], page 74.
• GNU Fortran supports a subset of features derived from ISO/IEC TS 18508:2015,
“Additional Parallel Features in Fortran”:
• The new atomic ADD, CAS, FETCH and ADD/OR/XOR, OR and XOR intrin-
sics.
• The CO_MIN and CO_MAX and SUM reduction intrinsics, and the CO_BROADCAST and
CO_REDUCE intrinsic, except that those do not support polymorphic types or types
with allocatable, pointer or polymorphic components.
• Events (EVENT POST, EVENT WAIT, EVENT_QUERY).
• Failed images (FAIL IMAGE, IMAGE_STATUS, FAILED_IMAGES, STOPPED_IMAGES).
• An ERROR STOP statement is permitted in a PURE procedure.
• GNU Fortran supports the IMPLICIT NONE statement with an implicit-none-spec-
list.
• The behavior of the INQUIRE statement with the RECL= specifier now conforms to
Fortran 2018.
Chapter 1: Introduction 5

Part I: Invoking GNU Fortran

 7

2 GNU Fortran Command Options

The gfortran command supports all the options supported by the gcc command. Only
options specific to GNU Fortran are documented here.
See Section “GCC Command Options” in Using the GNU Compiler Collection (GCC),
for information on the non-Fortran-specific aspects of the gcc command (and, therefore,
the gfortran command).
All GCC and GNU Fortran options are accepted both by gfortran and by gcc (as well
as any other drivers built at the same time, such as g++), since adding GNU Fortran to the
GCC distribution enables acceptance of GNU Fortran options by all of the relevant drivers.
In some cases, options have positive and negative forms; the negative form of -ffoo
would be -fno-foo. This manual documents only one of these two forms, whichever one is
not the default.

2.1 Option summary

Here is a summary of all the options specific to GNU Fortran, grouped by type. Explanations
are in the following sections.
Fortran Language Options
See Section 2.2 [Options controlling Fortran dialect], page 8.
-fall-intrinsics -fallow-argument-mismatch -fallow-invalid-boz
-fbackslash -fcray-pointer -fd-lines-as-code -fd-lines-as-comments
-fdec -fdec-char-conversions -fdec-structure -fdec-intrinsic-ints
-fdec-static -fdec-math -fdec-include -fdec-format-defaults
-fdec-blank-format-item -fdefault-double-8 -fdefault-integer-8
-fdefault-real-8 -fdefault-real-10 -fdefault-real-16 -fdollar-ok
-ffixed-line-length-n -ffixed-line-length-none -fpad-source
-ffree-form -ffree-line-length-n -ffree-line-length-none
-fimplicit-none -finteger-4-integer-8 -fmax-identifier-length
-fmodule-private -ffixed-form -fno-range-check -fopenacc -fopenmp
-freal-4-real-10 -freal-4-real-16 -freal-4-real-8 -freal-8-real-10
-freal-8-real-16 -freal-8-real-4 -std=std -ftest-forall-temp

Preprocessing Options
See Section 2.3 [Enable and customize preprocessing], page 14.
-A-question[=answer]
-Aquestion=answer -C -CC -Dmacro[=defn]
-H -P
-Umacro -cpp -dD -dI -dM -dN -dU -fworking-directory
-imultilib dir
-iprefix file -iquote -isysroot dir -isystem dir -nocpp
-nostdinc
-undef

Error and Warning Options

See Section 2.4 [Options to request or suppress errors and warnings], page 17.
-Waliasing -Wall -Wampersand -Warray-bounds
-Wc-binding-type -Wcharacter-truncation -Wconversion
-Wdo-subscript -Wfunction-elimination -Wimplicit-interface
-Wimplicit-procedure -Wintrinsic-shadow -Wuse-without-only
8 The GNU Fortran Compiler

-Wintrinsics-std -Wline-truncation -Wno-align-commons

-Wno-overwrite-recursive -Wno-tabs -Wreal-q-constant -Wsurprising
-Wunderflow -Wunused-parameter -Wrealloc-lhs -Wrealloc-lhs-all
-Wfrontend-loop-interchange -Wtarget-lifetime -fmax-errors=n
-fsyntax-only -pedantic
-pedantic-errors

Debugging Options
See Section 2.5 [Options for debugging your program or GNU Fortran], page 22.
-fbacktrace -fdump-fortran-optimized -fdump-fortran-original
-fdebug-aux-vars -fdump-fortran-global -fdump-parse-tree -ffpe-trap=list
-ffpe-summary=list

Directory Options
See Section 2.6 [Options for directory search], page 23.
-Idir -Jdir -fintrinsic-modules-path dir

Link Options
See Section 2.7 [Options for influencing the linking step], page 24.
-static-libgfortran -static-libquadmath

Runtime Options
See Section 2.8 [Options for influencing runtime behavior], page 24.
-fconvert=conversion -fmax-subrecord-length=length
-frecord-marker=length -fsign-zero

Interoperability Options
See Section 2.10 [Options for interoperability], page 32.
-fc-prototypes -fc-prototypes-external

Code Generation Options

See Section 2.9 [Options for code generation conventions], page 25.
-faggressive-function-elimination -fblas-matmul-limit=n
-fbounds-check -ftail-call-workaround -ftail-call-workaround=n
-fcheck-array-temporaries
-fcheck=<all|array-temps|bits|bounds|do|mem|pointer|recursion>
-fcoarray=<none|single|lib> -fexternal-blas -ff2c
-ffrontend-loop-interchange -ffrontend-optimize
-finit-character=n -finit-integer=n -finit-local-zero
-finit-derived -finit-logical=<true|false>
-finit-real=<zero|inf|-inf|nan|snan>
-finline-matmul-limit=n
-finline-arg-packing -fmax-array-constructor=n
-fmax-stack-var-size=n -fno-align-commons -fno-automatic
-fno-protect-parens -fno-underscoring -fsecond-underscore
-fpack-derived -frealloc-lhs -frecursive -frepack-arrays
-fshort-enums -fstack-arrays

2.2 Options controlling Fortran dialect

The following options control the details of the Fortran dialect accepted by the compiler:
Chapter 2: GNU Fortran Command Options 9

-ffree-form
-ffixed-form
Specify the layout used by the source file. The free form layout was introduced
in Fortran 90. Fixed form was traditionally used in older Fortran programs.
When neither option is specified, the source form is determined by the file
extension.
-fall-intrinsics
This option causes all intrinsic procedures (including the GNU-specific ex-
tensions) to be accepted. This can be useful with -std= to force standard-
compliance but get access to the full range of intrinsics available with gfortran.
As a consequence, -Wintrinsics-std will be ignored and no user-defined pro-
cedure with the same name as any intrinsic will be called except when it is
explicitly declared EXTERNAL.
-fallow-argument-mismatch
Some code contains calls to external procedures with mismatches between the
calls and the procedure definition, or with mismatches between different calls.
Such code is non-conforming, and will usually be flagged with an error. This
options degrades the error to a warning, which can only be disabled by disabling
all warnings via -w. Only a single occurrence per argument is flagged by this
warning. -fallow-argument-mismatch is implied by -std=legacy.
Using this option is strongly discouraged. It is possible to provide standard-
conforming code which allows different types of arguments by using an explicit
interface and TYPE(*).
-fallow-invalid-boz
A BOZ literal constant can occur in a limited number of contexts in standard
conforming Fortran. This option degrades an error condition to a warning,
and allows a BOZ literal constant to appear where the Fortran standard would
otherwise prohibit its use.
-fd-lines-as-code
-fd-lines-as-comments
Enable special treatment for lines beginning with d or D in fixed form sources. If
the -fd-lines-as-code option is given they are treated as if the first column
contained a blank. If the -fd-lines-as-comments option is given, they are
treated as comment lines.
-fdec DEC compatibility mode. Enables extensions and other features that mimic
the default behavior of older compilers (such as DEC). These features are non-
standard and should be avoided at all costs. For details on GNU Fortran’s
implementation of these extensions see the full documentation.
Other flags enabled by this switch are: -fdollar-ok -fcray-pointer
-fdec-char-conversions -fdec-structure -fdec-intrinsic-ints
-fdec-static -fdec-math -fdec-include -fdec-blank-format-item
-fdec-format-defaults
If -fd-lines-as-code/-fd-lines-as-comments are unset, then -fdec also
sets -fd-lines-as-comments.
10 The GNU Fortran Compiler

-fdec-char-conversions
Enable the use of character literals in assignments and DATA statements for
non-character variables.

-fdec-structure
Enable DEC STRUCTURE and RECORD as well as UNION, MAP, and dot (’.’) as a
member separator (in addition to ’%’). This is provided for compatibility only;
Fortran 90 derived types should be used instead where possible.

-fdec-intrinsic-ints
Enable B/I/J/K kind variants of existing integer functions (e.g. BIAND,
IIAND, JIAND, etc...). For a complete list of intrinsics see the full
documentation.

-fdec-math
Enable legacy math intrinsics such as COTAN and degree-valued trigonometric
functions (e.g. TAND, ATAND, etc...) for compatability with older code.

-fdec-static
Enable DEC-style STATIC and AUTOMATIC attributes to explicitly specify
the storage of variables and other objects.

-fdec-include
Enable parsing of INCLUDE as a statement in addition to parsing it as IN-
CLUDE line. When parsed as INCLUDE statement, INCLUDE does not have
to be on a single line and can use line continuations.

-fdec-format-defaults
Enable format specifiers F, G and I to be used without width specifiers, default
widths will be used instead.

-fdec-blank-format-item
Enable a blank format item at the end of a format specification i.e. nothing
following the final comma.

-fdollar-ok
Allow ‘$’ as a valid non-first character in a symbol name. Symbols that start
with ‘$’ are rejected since it is unclear which rules to apply to implicit typing as
different vendors implement different rules. Using ‘$’ in IMPLICIT statements
is also rejected.

-fbackslash
Change the interpretation of backslashes in string literals from a single back-
slash character to “C-style” escape characters. The following combinations are
expanded \a, \b, \f, \n, \r, \t, \v, \\, and \0 to the ASCII characters alert,
backspace, form feed, newline, carriage return, horizontal tab, vertical tab,
backslash, and NUL, respectively. Additionally, \xnn, \unnnn and \Unnnnnnnn
(where each n is a hexadecimal digit) are translated into the Unicode charac-
ters corresponding to the specified code points. All other combinations of a
character preceded by \ are unexpanded.
Chapter 2: GNU Fortran Command Options 11

-fmodule-private
Set the default accessibility of module entities to PRIVATE. Use-associated en-
tities will not be accessible unless they are explicitly declared as PUBLIC.
-ffixed-line-length-n
Set column after which characters are ignored in typical fixed-form lines in the
source file, and, unless -fno-pad-source, through which spaces are assumed
(as if padded to that length) after the ends of short fixed-form lines.
Popular values for n include 72 (the standard and the default), 80 (card im-
age), and 132 (corresponding to “extended-source” options in some popular
compilers). n may also be ‘none’, meaning that the entire line is meaningful
and that continued character constants never have implicit spaces appended to
them to fill out the line. -ffixed-line-length-0 means the same thing as
-ffixed-line-length-none.
-fno-pad-source
By default fixed-form lines have spaces assumed (as if padded to that length)
after the ends of short fixed-form lines. This is not done either if -ffixed-line-
length-0, -ffixed-line-length-none or if -fno-pad-source option is used.
With any of those options continued character constants never have implicit
spaces appended to them to fill out the line.
-ffree-line-length-n
Set column after which characters are ignored in typical free-form lines in the
source file. The default value is 132. n may be ‘none’, meaning that the entire
line is meaningful. -ffree-line-length-0 means the same thing as -ffree-
line-length-none.
-fmax-identifier-length=n
Specify the maximum allowed identifier length. Typical values are 31 (Fortran
95) and 63 (Fortran 2003 and later).
-fimplicit-none
Specify that no implicit typing is allowed, unless overridden by explicit
IMPLICIT statements. This is the equivalent of adding implicit none to the
start of every procedure.
-fcray-pointer
Enable the Cray pointer extension, which provides C-like pointer functionality.
-fopenacc
Enable the OpenACC extensions. This includes OpenACC !$acc directives in
free form and c$acc, *$acc and !$acc directives in fixed form, !$ conditional
compilation sentinels in free form and c$, *$ and !$ sentinels in fixed form,
and when linking arranges for the OpenACC runtime library to be linked in.
-fopenmp Enable the OpenMP extensions. This includes OpenMP !$omp directives in
free form and c$omp, *$omp and !$omp directives in fixed form, !$ conditional
compilation sentinels in free form and c$, *$ and !$ sentinels in fixed form, and
when linking arranges for the OpenMP runtime library to be linked in. The
option -fopenmp implies -frecursive.
12 The GNU Fortran Compiler

-fno-range-check
Disable range checking on results of simplification of constant expressions during
compilation. For example, GNU Fortran will give an error at compile time when
simplifying a = 1. / 0. With this option, no error will be given and a will be
assigned the value +Infinity. If an expression evaluates to a value outside of
the relevant range of [-HUGE():HUGE()], then the expression will be replaced by
-Inf or +Inf as appropriate. Similarly, DATA i/Z'FFFFFFFF'/ will result in an
integer overflow on most systems, but with -fno-range-check the value will
“wrap around” and i will be initialized to −1 instead.
-fdefault-integer-8
Set the default integer and logical types to an 8 byte wide type. This option also
affects the kind of integer constants like 42. Unlike -finteger-4-integer-8,
it does not promote variables with explicit kind declaration.
-fdefault-real-8
Set the default real type to an 8 byte wide type. This option also affects the kind
of non-double real constants like 1.0. This option promotes the default width
of DOUBLE PRECISION and double real constants like 1.d0 to 16 bytes if pos-
sible. If -fdefault-double-8 is given along with fdefault-real-8, DOUBLE
PRECISION and double real constants are not promoted. Unlike -freal-4-
real-8, fdefault-real-8 does not promote variables with explicit kind dec-
larations.
-fdefault-real-10
Set the default real type to an 10 byte wide type. This option also affects the
kind of non-double real constants like 1.0. This option promotes the default
width of DOUBLE PRECISION and double real constants like 1.d0 to 16 bytes
if possible. If -fdefault-double-8 is given along with fdefault-real-10,
DOUBLE PRECISION and double real constants are not promoted. Unlike -freal-
4-real-10, fdefault-real-10 does not promote variables with explicit kind
declarations.
-fdefault-real-16
Set the default real type to an 16 byte wide type. This option also affects the
kind of non-double real constants like 1.0. This option promotes the default
width of DOUBLE PRECISION and double real constants like 1.d0 to 16 bytes
if possible. If -fdefault-double-8 is given along with fdefault-real-16,
DOUBLE PRECISION and double real constants are not promoted. Unlike -freal-
4-real-16, fdefault-real-16 does not promote variables with explicit kind
declarations.
-fdefault-double-8
Set the DOUBLE PRECISION type and double real constants like 1.d0 to an 8
byte wide type. Do nothing if this is already the default. This option prevents
-fdefault-real-8, -fdefault-real-10, and -fdefault-real-16, from pro-
moting DOUBLE PRECISION and double real constants like 1.d0 to 16 bytes.
-finteger-4-integer-8
Promote all INTEGER(KIND=4) entities to an INTEGER(KIND=8) entities. If
KIND=8 is unavailable, then an error will be issued. This option should be
Chapter 2: GNU Fortran Command Options 13

used with care and may not be suitable for your codes. Areas of possible con-
cern include calls to external procedures, alignment in EQUIVALENCE and/or
COMMON, generic interfaces, BOZ literal constant conversion, and I/O. Inspec-
tion of the intermediate representation of the translated Fortran code, produced
by -fdump-tree-original, is suggested.

-freal-4-real-8
-freal-4-real-10
-freal-4-real-16
-freal-8-real-4
-freal-8-real-10
-freal-8-real-16
Promote all REAL(KIND=M) entities to REAL(KIND=N) entities. If REAL(KIND=N)
is unavailable, then an error will be issued. The -freal-4- flags also affect
the default real kind and the -freal-8- flags also the double-precision real
kind. All other real-kind types are unaffected by this option. The promotion
is also applied to real literal constants of default and double-precision kind
and a specified kind number of 4 or 8, respectively. However, -fdefault-
real-8, -fdefault-real-10, -fdefault-real-10, and -fdefault-double-8
take precedence for the default and double-precision real kinds, both for real
literal constants and for declarations without a kind number. Note that for
REAL(KIND=KIND(1.0)) the literal may get promoted and then the result may
get promoted again. These options should be used with care and may not be
suitable for your codes. Areas of possible concern include calls to external proce-
dures, alignment in EQUIVALENCE and/or COMMON, generic interfaces, BOZ literal
constant conversion, and I/O and calls to intrinsic procedures when passing a
value to the kind= dummy argument. Inspection of the intermediate represen-
tation of the translated Fortran code, produced by -fdump-fortran-original
or -fdump-tree-original, is suggested.

-std=std Specify the standard to which the program is expected to conform, which may
be one of ‘f95’, ‘f2003’, ‘f2008’, ‘f2018’, ‘gnu’, or ‘legacy’. The default value
for std is ‘gnu’, which specifies a superset of the latest Fortran standard that
includes all of the extensions supported by GNU Fortran, although warnings
will be given for obsolete extensions not recommended for use in new code. The
‘legacy’ value is equivalent but without the warnings for obsolete extensions,
and may be useful for old non-standard programs. The ‘f95’, ‘f2003’, ‘f2008’,
and ‘f2018’ values specify strict conformance to the Fortran 95, Fortran 2003,
Fortran 2008 and Fortran 2018 standards, respectively; errors are given for all
extensions beyond the relevant language standard, and warnings are given for
the Fortran 77 features that are permitted but obsolescent in later standards.
The deprecated option ‘-std=f2008ts’ acts as an alias for ‘-std=f2018’. It
is only present for backwards compatibility with earlier gfortran versions and
should not be used any more.

-ftest-forall-temp
Enhance test coverage by forcing most forall assignments to use temporary.
14 The GNU Fortran Compiler

2.3 Enable and customize preprocessing

Many Fortran compilers including GNU Fortran allow passing the source code through a
C preprocessor (CPP; sometimes also called the Fortran preprocessor, FPP) to allow for
conditional compilation. In the case of GNU Fortran, this is the GNU C Preprocessor
in the traditional mode. On systems with case-preserving file names, the preprocessor
is automatically invoked if the filename extension is .F, .FOR, .FTN, .fpp, .FPP, .F90,
.F95, .F03 or .F08. To manually invoke the preprocessor on any file, use -cpp, to disable
preprocessing on files where the preprocessor is run automatically, use -nocpp.
If a preprocessed file includes another file with the Fortran INCLUDE statement, the in-
cluded file is not preprocessed. To preprocess included files, use the equivalent preprocessor
statement #include.
If GNU Fortran invokes the preprocessor, __GFORTRAN__ is defined. The macros __GNUC_
_, __GNUC_MINOR__ and __GNUC_PATCHLEVEL__ can be used to determine the version of the
compiler. See Section “Overview” in The C Preprocessor for details.
GNU Fortran supports a number of INTEGER and REAL kind types in additional to the
kind types required by the Fortran standard. The availability of any given kind type is
architecture dependent. The following pre-defined preprocessor macros can be used to
conditionally include code for these additional kind types: __GFC_INT_1__, __GFC_INT_2_
_, __GFC_INT_8__, __GFC_INT_16__, __GFC_REAL_10__, and __GFC_REAL_16__.
While CPP is the de-facto standard for preprocessing Fortran code, Part 3 of the Fortran
95 standard (ISO/IEC 1539-3:1998) defines Conditional Compilation, which is not widely
used and not directly supported by the GNU Fortran compiler. You can use the program
coco to preprocess such files (http://www.daniellnagle.com/coco.html).
The following options control preprocessing of Fortran code:
-cpp
-nocpp Enable preprocessing. The preprocessor is automatically invoked if the file
extension is .fpp, .FPP, .F, .FOR, .FTN, .F90, .F95, .F03 or .F08. Use this
option to manually enable preprocessing of any kind of Fortran file.
To disable preprocessing of files with any of the above listed extensions, use the
negative form: -nocpp.
The preprocessor is run in traditional mode. Any restrictions of the file-format,
especially the limits on line length, apply for preprocessed output as well, so it
might be advisable to use the -ffree-line-length-none or -ffixed-line-
length-none options.
-dM Instead of the normal output, generate a list of '#define' directives for all the
macros defined during the execution of the preprocessor, including predefined
macros. This gives you a way of finding out what is predefined in your version
of the preprocessor. Assuming you have no file foo.f90, the command
touch foo.f90; gfortran -cpp -E -dM foo.f90
will show all the predefined macros.
-dD Like -dM except in two respects: it does not include the predefined macros, and
it outputs both the #define directives and the result of preprocessing. Both
kinds of output go to the standard output file.
Chapter 2: GNU Fortran Command Options 15

-dN Like -dD, but emit only the macro names, not their expansions.

-dU Like dD except that only macros that are expanded, or whose definedness is
tested in preprocessor directives, are output; the output is delayed until the use
or test of the macro; and '#undef' directives are also output for macros tested
but undefined at the time.

-dI Output '#include' directives in addition to the result of preprocessing.

-fworking-directory
Enable generation of linemarkers in the preprocessor output that will let the
compiler know the current working directory at the time of preprocessing. When
this option is enabled, the preprocessor will emit, after the initial linemarker, a
second linemarker with the current working directory followed by two slashes.
GCC will use this directory, when it is present in the preprocessed input, as
the directory emitted as the current working directory in some debugging in-
formation formats. This option is implicitly enabled if debugging information
is enabled, but this can be inhibited with the negated form -fno-working-
directory. If the -P flag is present in the command line, this option has no
effect, since no #line directives are emitted whatsoever.

-idirafter dir
Search dir for include files, but do it after all directories specified with -I and
the standard system directories have been exhausted. dir is treated as a system
include directory. If dir begins with =, then the = will be replaced by the sysroot
prefix; see --sysroot and -isysroot.

-imultilib dir
Use dir as a subdirectory of the directory containing target-specific C++ headers.

-iprefix prefix
Specify prefix as the prefix for subsequent -iwithprefix options. If the prefix
represents a directory, you should include the final '/'.

-isysroot dir
This option is like the --sysroot option, but applies only to header files. See
the --sysroot option for more information.

-iquote dir
Search dir only for header files requested with #include "file"; they are not
searched for #include <file>, before all directories specified by -I and before
the standard system directories. If dir begins with =, then the = will be replaced
by the sysroot prefix; see --sysroot and -isysroot.

-isystem dir
Search dir for header files, after all directories specified by -I but before the
standard system directories. Mark it as a system directory, so that it gets the
same special treatment as is applied to the standard system directories. If dir
begins with =, then the = will be replaced by the sysroot prefix; see --sysroot
and -isysroot.
16 The GNU Fortran Compiler

-nostdinc
Do not search the standard system directories for header files. Only the direc-
tories you have specified with -I options (and the directory of the current file,
if appropriate) are searched.
-undef Do not predefine any system-specific or GCC-specific macros. The standard
predefined macros remain defined.
-Apredicate=answer
Make an assertion with the predicate predicate and answer answer. This form
is preferred to the older form -A predicate(answer), which is still supported,
because it does not use shell special characters.
-A-predicate=answer
Cancel an assertion with the predicate predicate and answer answer.
-C Do not discard comments. All comments are passed through to the output file,
except for comments in processed directives, which are deleted along with the
directive.
You should be prepared for side effects when using -C; it causes the preproces-
sor to treat comments as tokens in their own right. For example, comments
appearing at the start of what would be a directive line have the effect of turn-
ing that line into an ordinary source line, since the first token on the line is no
longer a '#'.
Warning: this currently handles C-Style comments only. The preprocessor does
not yet recognize Fortran-style comments.
-CC Do not discard comments, including during macro expansion. This is like -C,
except that comments contained within macros are also passed through to the
output file where the macro is expanded.
In addition to the side-effects of the -C option, the -CC option causes all C++-
style comments inside a macro to be converted to C-style comments. This
is to prevent later use of that macro from inadvertently commenting out the
remainder of the source line. The -CC option is generally used to support lint
comments.
Warning: this currently handles C- and C++-Style comments only. The prepro-
cessor does not yet recognize Fortran-style comments.
-Dname Predefine name as a macro, with definition 1.
-Dname=definition
The contents of definition are tokenized and processed as if they appeared
during translation phase three in a '#define' directive. In particular, the
definition will be truncated by embedded newline characters.
If you are invoking the preprocessor from a shell or shell-like program you may
need to use the shell’s quoting syntax to protect characters such as spaces that
have a meaning in the shell syntax.
If you wish to define a function-like macro on the command line, write its
argument list with surrounding parentheses before the equals sign (if any).
Chapter 2: GNU Fortran Command Options 17

Parentheses are meaningful to most shells, so you will need to quote the option.
With sh and csh, -D'name(args...)=definition' works.
-D and -U options are processed in the order they are given on the command
line. All -imacros file and -include file options are processed after all -D and -U
options.
-H Print the name of each header file used, in addition to other normal activities.
Each name is indented to show how deep in the '#include' stack it is.
-P Inhibit generation of linemarkers in the output from the preprocessor. This
might be useful when running the preprocessor on something that is not C code,
and will be sent to a program which might be confused by the linemarkers.
-Uname Cancel any previous definition of name, either built in or provided with a -D
option.

2.4 Options to request or suppress errors and warnings

Errors are diagnostic messages that report that the GNU Fortran compiler cannot compile
the relevant piece of source code. The compiler will continue to process the program in an
attempt to report further errors to aid in debugging, but will not produce any compiled
output.
Warnings are diagnostic messages that report constructions which are not inherently
erroneous but which are risky or suggest there is likely to be a bug in the program. Unless
-Werror is specified, they do not prevent compilation of the program.
You can request many specific warnings with options beginning -W, for example
-Wimplicit to request warnings on implicit declarations. Each of these specific warning
options also has a negative form beginning -Wno- to turn off warnings; for example,
-Wno-implicit. This manual lists only one of the two forms, whichever is not the default.
These options control the amount and kinds of errors and warnings produced by GNU
Fortran:
-fmax-errors=n
Limits the maximum number of error messages to n, at which point GNU
Fortran bails out rather than attempting to continue processing the source
code. If n is 0, there is no limit on the number of error messages produced.
-fsyntax-only
Check the code for syntax errors, but do not actually compile it. This will
generate module files for each module present in the code, but no other output
file.
-Wpedantic
-pedantic
Issue warnings for uses of extensions to Fortran. -pedantic also applies to
C-language constructs where they occur in GNU Fortran source files, such as
use of ‘\e’ in a character constant within a directive like #include.
Valid Fortran programs should compile properly with or without this option.
However, without this option, certain GNU extensions and traditional Fortran
features are supported as well. With this option, many of them are rejected.
18 The GNU Fortran Compiler

Some users try to use -pedantic to check programs for conformance. They
soon find that it does not do quite what they want—it finds some nonstandard
practices, but not all. However, improvements to GNU Fortran in this area are
welcome.
This should be used in conjunction with -std=f95, -std=f2003, -std=f2008
or -std=f2018.
-pedantic-errors
Like -pedantic, except that errors are produced rather than warnings.
-Wall Enables commonly used warning options pertaining to usage that we
recommend avoiding and that we believe are easy to avoid. This currently
includes -Waliasing, -Wampersand, -Wconversion, -Wsurprising,
-Wc-binding-type, -Wintrinsics-std, -Wtabs, -Wintrinsic-shadow,
-Wline-truncation, -Wtarget-lifetime, -Winteger-division,
-Wreal-q-constant, -Wunused and -Wundefined-do-loop.
-Waliasing
Warn about possible aliasing of dummy arguments. Specifically, it warns if the
same actual argument is associated with a dummy argument with INTENT(IN)
and a dummy argument with INTENT(OUT) in a call with an explicit interface.
The following example will trigger the warning.
interface
subroutine bar(a,b)
integer, intent(in) :: a
integer, intent(out) :: b
end subroutine
end interface
integer :: a

call bar(a,a)

-Wampersand
Warn about missing ampersand in continued character constants. The warning
is given with -Wampersand, -pedantic, -std=f95, -std=f2003, -std=f2008
and -std=f2018. Note: With no ampersand given in a continued character
constant, GNU Fortran assumes continuation at the first non-comment, non-
whitespace character after the ampersand that initiated the continuation.
-Warray-temporaries
Warn about array temporaries generated by the compiler. The information
generated by this warning is sometimes useful in optimization, in order to avoid
such temporaries.
-Wc-binding-type
Warn if the a variable might not be C interoperable. In particular, warn if the
variable has been declared using an intrinsic type with default kind instead of
using a kind parameter defined for C interoperability in the intrinsic ISO_C_
Binding module. This option is implied by -Wall.
-Wcharacter-truncation
Warn when a character assignment will truncate the assigned string.
Chapter 2: GNU Fortran Command Options 19

-Wline-truncation
Warn when a source code line will be truncated. This option is implied by
-Wall. For free-form source code, the default is -Werror=line-truncation
such that truncations are reported as error.
-Wconversion
Warn about implicit conversions that are likely to change the value of the
expression after conversion. Implied by -Wall.
-Wconversion-extra
Warn about implicit conversions between different types and kinds. This option
does not imply -Wconversion.
-Wextra Enables some warning options for usages of language features which may be
problematic. This currently includes -Wcompare-reals, -Wunused-parameter
and -Wdo-subscript.
-Wfrontend-loop-interchange
Warn when using -ffrontend-loop-interchange for performing loop inter-
changes.
-Wimplicit-interface
Warn if a procedure is called without an explicit interface. Note this only
checks that an explicit interface is present. It does not check that the declared
interfaces are consistent across program units.
-Wimplicit-procedure
Warn if a procedure is called that has neither an explicit interface nor has been
declared as EXTERNAL.
-Winteger-division
Warn if a constant integer division truncates its result. As an example, 3/5
evaluates to 0.
-Wintrinsics-std
Warn if gfortran finds a procedure named like an intrinsic not available in the
currently selected standard (with -std) and treats it as EXTERNAL procedure
because of this. -fall-intrinsics can be used to never trigger this behavior
and always link to the intrinsic regardless of the selected standard.
-Wno-overwrite-recursive
Do not warn when -fno-automatic is used with -frecursive. Recursion will
be broken if the relevant local variables do not have the attribute AUTOMATIC
explicitly declared. This option can be used to suppress the warning when it
is known that recursion is not broken. Useful for build environments that use
-Werror.
-Wreal-q-constant
Produce a warning if a real-literal-constant contains a q exponent-letter.
-Wsurprising
Produce a warning when “suspicious” code constructs are encountered. While
technically legal these usually indicate that an error has been made.
20 The GNU Fortran Compiler

This currently produces a warning under the following circumstances:

• An INTEGER SELECT construct has a CASE that can never be matched
as its lower value is greater than its upper value.
• A LOGICAL SELECT construct has three CASE statements.
• A TRANSFER specifies a source that is shorter than the destination.
• The type of a function result is declared more than once with the same
type. If -pedantic or standard-conforming mode is enabled, this is an
error.
• A CHARACTER variable is declared with negative length.
• With -fopenmp, for fixed-form source code, when an omx vendor-extension
sentinel is encountered. (The equivalent ompx, used in free-form source
code, is diagnosed by default.)
-Wtabs By default, tabs are accepted as whitespace, but tabs are not members of the
Fortran Character Set. For continuation lines, a tab followed by a digit between
1 and 9 is supported. -Wtabs will cause a warning to be issued if a tab is
encountered. Note, -Wtabs is active for -pedantic, -std=f95, -std=f2003,
-std=f2008, -std=f2018 and -Wall.
-Wundefined-do-loop
Warn if a DO loop with step either 1 or -1 yields an underflow or an overflow
during iteration of an induction variable of the loop. This option is implied by
-Wall.
-Wunderflow
Produce a warning when numerical constant expressions are encountered, which
yield an UNDERFLOW during compilation. Enabled by default.
-Wintrinsic-shadow
Warn if a user-defined procedure or module procedure has the same name
as an intrinsic; in this case, an explicit interface or EXTERNAL or INTRINSIC
declaration might be needed to get calls later resolved to the desired intrin-
sic/procedure. This option is implied by -Wall.
-Wuse-without-only
Warn if a USE statement has no ONLY qualifier and thus implicitly imports all
public entities of the used module.
-Wunused-dummy-argument
Warn about unused dummy arguments. This option is implied by -Wall.
-Wunused-parameter
Contrary to gcc’s meaning of -Wunused-parameter, gfortran’s implemen-
tation of this option does not warn about unused dummy arguments (see
-Wunused-dummy-argument), but about unused PARAMETER values. -Wunused-
parameter is implied by -Wextra if also -Wunused or -Wall is used.
-Walign-commons
By default, gfortran warns about any occasion of variables being padded for
proper alignment inside a COMMON block. This warning can be turned off via
-Wno-align-commons. See also -falign-commons.
Chapter 2: GNU Fortran Command Options 21

-Wfunction-elimination
Warn if any calls to impure functions are eliminated by the optimizations en-
abled by the -ffrontend-optimize option. This option is implied by -Wextra.

-Wrealloc-lhs
Warn when the compiler might insert code to for allocation or reallocation of
an allocatable array variable of intrinsic type in intrinsic assignments. In hot
loops, the Fortran 2003 reallocation feature may reduce the performance. If
the array is already allocated with the correct shape, consider using a whole-
array array-spec (e.g. (:,:,:)) for the variable on the left-hand side to prevent
the reallocation check. Note that in some cases the warning is shown, even if
the compiler will optimize reallocation checks away. For instance, when the
right-hand side contains the same variable multiplied by a scalar. See also
-frealloc-lhs.

-Wrealloc-lhs-all
Warn when the compiler inserts code to for allocation or reallocation of an
allocatable variable; this includes scalars and derived types.

-Wcompare-reals
Warn when comparing real or complex types for equality or inequality. This
option is implied by -Wextra.

-Wtarget-lifetime
Warn if the pointer in a pointer assignment might be longer than the its target.
This option is implied by -Wall.

-Wzerotrip
Warn if a DO loop is known to execute zero times at compile time. This option
is implied by -Wall.

-Wdo-subscript
Warn if an array subscript inside a DO loop could lead to an out-of-bounds
access even if the compiler cannot prove that the statement is actually executed,
in cases like
real a(3)
do i=1,4
if (condition(i)) then
a(i) = 1.2
end if
end do

This option is implied by -Wextra.

-Werror Turns all warnings into errors.

See Section “Options to Request or Suppress Errors and Warnings” in Using the GNU
Compiler Collection (GCC), for information on more options offered by the GBE shared by
gfortran, gcc and other GNU compilers.
Some of these have no effect when compiling programs written in Fortran.
22 The GNU Fortran Compiler

2.5 Options for debugging your program or GNU Fortran

GNU Fortran has various special options that are used for debugging either your program
or the GNU Fortran compiler.
-fdump-fortran-original
Output the internal parse tree after translating the source program into internal
representation. This option is mostly useful for debugging the GNU Fortran
compiler itself. The output generated by this option might change between
releases. This option may also generate internal compiler errors for features
which have only recently been added.
-fdump-fortran-optimized
Output the parse tree after front-end optimization. Mostly useful for debugging
the GNU Fortran compiler itself. The output generated by this option might
change between releases. This option may also generate internal compiler errors
for features which have only recently been added.
-fdump-parse-tree
Output the internal parse tree after translating the source program into internal
representation. Mostly useful for debugging the GNU Fortran compiler itself.
The output generated by this option might change between releases. This option
may also generate internal compiler errors for features which have only recently
been added. This option is deprecated; use -fdump-fortran-original instead.
-fdebug-aux-vars
Renames internal variables created by the gfortran front end and makes them
accessible to a debugger. The name of the internal variables then start with
upper-case letters followed by an underscore. This option is useful for debug-
ging the compiler’s code generation together with -fdump-tree-original and
enabling debugging of the executable program by using -g or -ggdb3.
-fdump-fortran-global
Output a list of the global identifiers after translating into middle-end repre-
sentation. Mostly useful for debugging the GNU Fortran compiler itself. The
output generated by this option might change between releases. This option
may also generate internal compiler errors for features which have only recently
been added.
-ffpe-trap=list
Specify a list of floating point exception traps to enable. On most systems, if
a floating point exception occurs and the trap for that exception is enabled, a
SIGFPE signal will be sent and the program being aborted, producing a core
file useful for debugging. list is a (possibly empty) comma-separated list of
the following exceptions: ‘invalid’ (invalid floating point operation, such as
SQRT(-1.0)), ‘zero’ (division by zero), ‘overflow’ (overflow in a floating point
operation), ‘underflow’ (underflow in a floating point operation), ‘inexact’
(loss of precision during operation), and ‘denormal’ (operation performed on
a denormal value). The first five exceptions correspond to the five IEEE 754
exceptions, whereas the last one (‘denormal’) is not part of the IEEE 754
standard but is available on some common architectures such as x86.
Chapter 2: GNU Fortran Command Options 23

The first three exceptions (‘invalid’, ‘zero’, and ‘overflow’) often indicate
serious errors, and unless the program has provisions for dealing with these
exceptions, enabling traps for these three exceptions is probably a good idea.
If the option is used more than once in the command line, the lists will be joined:
’ffpe-trap=list1 ffpe-trap=list2’ is equivalent to ffpe-trap=list1,list2.
Note that once enabled an exception cannot be disabled (no negative form).
Many, if not most, floating point operations incur loss of precision due to round-
ing, and hence the ffpe-trap=inexact is likely to be uninteresting in practice.
By default no exception traps are enabled.
-ffpe-summary=list
Specify a list of floating-point exceptions, whose flag status is printed to ERROR_
UNIT when invoking STOP and ERROR STOP. list can be either ‘none’, ‘all’ or a
comma-separated list of the following exceptions: ‘invalid’, ‘zero’, ‘overflow’,
‘underflow’, ‘inexact’ and ‘denormal’. (See -ffpe-trap for a description of
the exceptions.)
If the option is used more than once in the command line, only the last one will
be used.
By default, a summary for all exceptions but ‘inexact’ is shown.
-fno-backtrace
When a serious runtime error is encountered or a deadly signal is emitted (seg-
mentation fault, illegal instruction, bus error, floating-point exception, and the
other POSIX signals that have the action ‘core’), the Fortran runtime library
tries to output a backtrace of the error. -fno-backtrace disables the backtrace
generation. This option only has influence for compilation of the Fortran main
program.
See Section “Options for Debugging Your Program or GCC” in Using the GNU Compiler
Collection (GCC), for more information on debugging options.

2.6 Options for directory search

These options affect how GNU Fortran searches for files specified by the INCLUDE directive
and where it searches for previously compiled modules.
It also affects the search paths used by cpp when used to preprocess Fortran source.
-Idir These affect interpretation of the INCLUDE directive (as well as of the #include
directive of the cpp preprocessor).
Also note that the general behavior of -I and INCLUDE is pretty much the same
as of -I with #include in the cpp preprocessor, with regard to looking for
header.gcc files and other such things.
This path is also used to search for .mod files when previously compiled modules
are required by a USE statement.
See Section “Options for Directory Search” in Using the GNU Compiler Col-
lection (GCC), for information on the -I option.
24 The GNU Fortran Compiler

-Jdir This option specifies where to put .mod files for compiled modules. It is also
added to the list of directories to searched by an USE statement.
The default is the current directory.
-fintrinsic-modules-path dir
This option specifies the location of pre-compiled intrinsic modules, if they are
not in the default location expected by the compiler.

2.7 Influencing the linking step

These options come into play when the compiler links object files into an executable output
file. They are meaningless if the compiler is not doing a link step.
-static-libgfortran
On systems that provide libgfortran as a shared and a static library, this
option forces the use of the static version. If no shared version of libgfortran
was built when the compiler was configured, this option has no effect.
-static-libquadmath
On systems that provide libquadmath as a shared and a static library, this
option forces the use of the static version. If no shared version of libquadmath
was built when the compiler was configured, this option has no effect.
Please note that the libquadmath runtime library is licensed under the GNU
Lesser General Public License (LGPL), and linking it statically introduces re-
quirements when redistributing the resulting binaries.

2.8 Influencing runtime behavior

These options affect the runtime behavior of programs compiled with GNU Fortran.
-fconvert=conversion
Specify the representation of data for unformatted files. Valid values for con-
version on most systems are: ‘native’, the default; ‘swap’, swap between big-
and little-endian; ‘big-endian’, use big-endian representation for unformatted
files; ‘little-endian’, use little-endian representation for unformatted files.
On POWER systems which suppport -mabi=ieeelongdouble, there are addi-
tional options, which can be combined with others with commas. Those are
-fconvert=r16_ieee Use IEEE 128-bit format for REAL(KIND=16).
-fconvert=r16_ibm Use IBM long double format for REAL(KIND=16).
This option has an effect only when used in the main program. The CONVERT
specifier and the GFORTRAN CONVERT UNIT environment variable over-
ride the default specified by -fconvert.
-frecord-marker=length
Specify the length of record markers for unformatted files. Valid values for
length are 4 and 8. Default is 4. This is different from previous versions of
gfortran, which specified a default record marker length of 8 on most systems.
If you want to read or write files compatible with earlier versions of gfortran,
use -frecord-marker=8.
Chapter 2: GNU Fortran Command Options 25

-fmax-subrecord-length=length
Specify the maximum length for a subrecord. The maximum permitted value
for length is 2147483639, which is also the default. Only really useful for use
by the gfortran testsuite.

-fsign-zero
When enabled, floating point numbers of value zero with the sign bit set are
written as negative number in formatted output and treated as negative in the
SIGN intrinsic. -fno-sign-zero does not print the negative sign of zero values
(or values rounded to zero for I/O) and regards zero as positive number in the
SIGN intrinsic for compatibility with Fortran 77. The default is -fsign-zero.

2.9 Options for code generation conventions

These machine-independent options control the interface conventions used in code genera-
tion.
Most of them have both positive and negative forms; the negative form of -ffoo would
be -fno-foo. In the table below, only one of the forms is listed—the one which is not the
default. You can figure out the other form by either removing no- or adding it.

-fno-automatic
Treat each program unit (except those marked as RECURSIVE) as if the SAVE
statement were specified for every local variable and array referenced in it. Does
not affect common blocks. (Some Fortran compilers provide this option under
the name -static or -save.) The default, which is -fautomatic, uses the stack
for local variables smaller than the value given by -fmax-stack-var-size. Use
the option -frecursive to use no static memory.
Local variables or arrays having an explicit SAVE attribute are silently ignored
unless the -pedantic option is added.

-ff2c Generate code designed to be compatible with code generated by g77 and f2c.
The calling conventions used by g77 (originally implemented in f2c) require
functions that return type default REAL to actually return the C type double,
and functions that return type COMPLEX to return the values via an extra ar-
gument in the calling sequence that points to where to store the return value.
Under the default GNU calling conventions, such functions simply return their
results as they would in GNU C—default REAL functions return the C type
float, and COMPLEX functions return the GNU C type complex. Additionally,
this option implies the -fsecond-underscore option, unless -fno-second-
underscore is explicitly requested.
This does not affect the generation of code that interfaces with the libgfortran
library.
Caution: It is not a good idea to mix Fortran code compiled with -ff2c with
code compiled with the default -fno-f2c calling conventions as, calling COMPLEX
or default REAL functions between program parts which were compiled with
different calling conventions will break at execution time.
26 The GNU Fortran Compiler

Caution: This will break code which passes intrinsic functions of type default
REAL or COMPLEX as actual arguments, as the library implementations use the
-fno-f2c calling conventions.
-fno-underscoring
Do not transform names of entities specified in the Fortran source file by ap-
pending underscores to them.
With -funderscoring in effect, GNU Fortran appends one underscore to ex-
ternal names. This is done to ensure compatibility with code produced by many
UNIX Fortran compilers.
Caution: The default behavior of GNU Fortran is incompatible with f2c and
g77, please use the -ff2c option if you want object files compiled with GNU
Fortran to be compatible with object code created with these tools.
Use of -fno-underscoring is not recommended unless you are experimenting
with issues such as integration of GNU Fortran into existing system environ-
ments (vis-à-vis existing libraries, tools, and so on).
For example, with -funderscoring, and assuming that j() and max_count()
are external functions while my_var and lvar are local variables, a statement
like
I = J() + MAX_COUNT (MY_VAR, LVAR)
is implemented as something akin to:
i = j_() + max_count_(&my_var, &lvar);
With -fno-underscoring, the same statement is implemented as:
i = j() + max_count(&my_var, &lvar);
Use of -fno-underscoring allows direct specification of user-defined names
while debugging and when interfacing GNU Fortran code with other languages.
Note that just because the names match does not mean that the interface
implemented by GNU Fortran for an external name matches the interface im-
plemented by some other language for that same name. That is, getting code
produced by GNU Fortran to link to code produced by some other compiler
using this or any other method can be only a small part of the overall solution—
getting the code generated by both compilers to agree on issues other than
naming can require significant effort, and, unlike naming disagreements, linkers
normally cannot detect disagreements in these other areas.
Also, note that with -fno-underscoring, the lack of appended underscores in-
troduces the very real possibility that a user-defined external name will conflict
with a name in a system library, which could make finding unresolved-reference
bugs quite difficult in some cases—they might occur at program run time, and
show up only as buggy behavior at run time.
In future versions of GNU Fortran we hope to improve naming and linking
issues so that debugging always involves using the names as they appear in the
source, even if the names as seen by the linker are mangled to prevent accidental
linking between procedures with incompatible interfaces.
-fsecond-underscore
By default, GNU Fortran appends an underscore to external names. If this
option is used GNU Fortran appends two underscores to names with underscores
Chapter 2: GNU Fortran Command Options 27

and one underscore to external names with no underscores. GNU Fortran also
appends two underscores to internal names with underscores to avoid naming
collisions with external names.
This option has no effect if -fno-underscoring is in effect. It is implied by
the -ff2c option.
Otherwise, with this option, an external name such as MAX_COUNT is imple-
mented as a reference to the link-time external symbol max_count__, instead
of max_count_. This is required for compatibility with g77 and f2c, and is
implied by use of the -ff2c option.
-fcoarray=<keyword>
‘none’ Disable coarray support; using coarray declarations and image-
control statements will produce a compile-time error. (Default)
‘single’ Single-image mode, i.e. num_images() is always one.
‘lib’ Library-based coarray parallelization; a suitable GNU Fortran coar-
ray library needs to be linked.
-fcheck=<keyword>
Enable the generation of run-time checks; the argument shall be a comma-
delimited list of the following keywords. Prefixing a check with no- disables it
if it was activated by a previous specification.
‘all’ Enable all run-time test of -fcheck.
‘array-temps’
Warns at run time when for passing an actual argument a tempo-
rary array had to be generated. The information generated by this
warning is sometimes useful in optimization, in order to avoid such
temporaries.
Note: The warning is only printed once per location.
‘bits’ Enable generation of run-time checks for invalid arguments to the
bit manipulation intrinsics.
‘bounds’ Enable generation of run-time checks for array subscripts and
against the declared minimum and maximum values. It also
checks array indices for assumed and deferred shape arrays against
the actual allocated bounds and ensures that all string lengths
are equal for character array constructors without an explicit
typespec.
Some checks require that -fcheck=bounds is set for the compilation
of the main program.
Note: In the future this may also include other forms of checking,
e.g., checking substring references.
‘do’ Enable generation of run-time checks for invalid modification of
loop iteration variables.
28 The GNU Fortran Compiler

‘mem’ Enable generation of run-time checks for memory allocation. Note:

This option does not affect explicit allocations using the ALLOCATE
statement, which will be always checked.

‘pointer’ Enable generation of run-time checks for pointers and allocatables.

‘recursion’
Enable generation of run-time checks for recursively called sub-
routines and functions which are not marked as recursive. See
also -frecursive. Note: This check does not work for OpenMP
programs and is disabled if used together with -frecursive and
-fopenmp.

Example: Assuming you have a file foo.f90, the command

gfortran -fcheck=all,no-array-temps foo.f90
will compile the file with all checks enabled as specified above except warnings
for generated array temporaries.

-fbounds-check
Deprecated alias for -fcheck=bounds.

-ftail-call-workaround
-ftail-call-workaround=n
Some C interfaces to Fortran codes violate the gfortran ABI by omitting the
hidden character length arguments as described in See Section 6.4.2 [Argument
passing conventions], page 81. This can lead to crashes because pushing argu-
ments for tail calls can overflow the stack.
To provide a workaround for existing binary packages, this option disables
tail call optimization for gfortran procedures with character arguments. With
-ftail-call-workaround=2 tail call optimization is disabled in all gfortran
procedures with character arguments, with -ftail-call-workaround=1
or equivalent -ftail-call-workaround only in gfortran procedures with
character arguments that call implicitly prototyped procedures.
Using this option can lead to problems including crashes due to insufficient
stack space.
It is very strongly recommended to fix the code in question. The
-fc-prototypes-external option can be used to generate prototypes which
conform to gfortran’s ABI, for inclusion in the source code.
Support for this option will likely be withdrawn in a future release of gfortran.
The negative form, -fno-tail-call-workaround or equivalent -ftail-call-
workaround=0, can be used to disable this option.
Default is currently -ftail-call-workaround, this will change in future re-
leases.

-fcheck-array-temporaries
Deprecated alias for -fcheck=array-temps.
Chapter 2: GNU Fortran Command Options 29

-fmax-array-constructor=n
This option can be used to increase the upper limit permitted in array con-
structors. The code below requires this option to expand the array at compile
time.
program test
implicit none
integer j
integer, parameter :: n = 100000
integer, parameter :: i(n) = (/ (2*j, j = 1, n) /)
print '(10(I0,1X))', i
end program test
Caution: This option can lead to long compile times and excessively large object
files.
The default value for n is 65535.
-fmax-stack-var-size=n
This option specifies the size in bytes of the largest array that will be put on
the stack; if the size is exceeded static memory is used (except in procedures
marked as RECURSIVE). Use the option -frecursive to allow for recursive
procedures which do not have a RECURSIVE attribute or for parallel programs.
Use -fno-automatic to never use the stack.
This option currently only affects local arrays declared with constant bounds,
and may not apply to all character variables. Future versions of GNU Fortran
may improve this behavior.
The default value for n is 65536.
-fstack-arrays
Adding this option will make the Fortran compiler put all arrays of unknown
size and array temporaries onto stack memory. If your program uses very large
local arrays it is possible that you will have to extend your runtime limits for
stack memory on some operating systems. This flag is enabled by default at
optimization level -Ofast unless -fmax-stack-var-size is specified.
-fpack-derived
This option tells GNU Fortran to pack derived type members as closely as
possible. Code compiled with this option is likely to be incompatible with code
compiled without this option, and may execute slower.
-frepack-arrays
In some circumstances GNU Fortran may pass assumed shape array sections
via a descriptor describing a noncontiguous area of memory. This option adds
code to the function prologue to repack the data into a contiguous block at
runtime.
This should result in faster accesses to the array. However it can introduce
significant overhead to the function call, especially when the passed data is
noncontiguous.
-fshort-enums
This option is provided for interoperability with C code that was compiled
with the -fshort-enums option. It will make GNU Fortran choose the smallest
30 The GNU Fortran Compiler

INTEGER kind a given enumerator set will fit in, and give all its enumerators
this kind.

-finline-arg-packing
When passing an assumed-shape argument of a procedure as actual argument
to an assumed-size or explicit size or as argument to a procedure that does not
have an explicit interface, the argument may have to be packed, that is put into
contiguous memory. An example is the call to foo in
subroutine foo(a)
real, dimension(*) :: a
end subroutine foo
subroutine bar(b)
real, dimension(:) :: b
call foo(b)
end subroutine bar
When -finline-arg-packing is in effect, this packing will be performed by
inline code. This allows for more optimization while increasing code size.
-finline-arg-packing is implied by any of the -O options except when opti-
mizing for size via -Os. If the code contains a very large number of argument
that have to be packed, code size and also compilation time may become ex-
cessive. If that is the case, it may be better to disable this option. Instances of
packing can be found by using -Warray-temporaries.

-fexternal-blas
This option will make gfortran generate calls to BLAS functions for some
matrix operations like MATMUL, instead of using our own algorithms, if the size
of the matrices involved is larger than a given limit (see -fblas-matmul-limit).
This may be profitable if an optimized vendor BLAS library is available. The
BLAS library will have to be specified at link time.

-fblas-matmul-limit=n
Only significant when -fexternal-blas is in effect. Matrix multiplication of
matrices with size larger than (or equal to) n will be performed by calls to
BLAS functions, while others will be handled by gfortran internal algorithms.
If the matrices involved are not square, the size comparison is performed using
the geometric mean of the dimensions of the argument and result matrices.
The default value for n is 30.

-finline-matmul-limit=n
When front-end optimization is active, some calls to the MATMUL intrinsic func-
tion will be inlined. This may result in code size increase if the size of the matrix
cannot be determined at compile time, as code for both cases is generated. Set-
ting -finline-matmul-limit=0 will disable inlining in all cases. Setting this
option with a value of n will produce inline code for matrices with size up to n.
If the matrices involved are not square, the size comparison is performed using
the geometric mean of the dimensions of the argument and result matrices.
The default value for n is 30. The -fblas-matmul-limit can be used to change
this value.
Chapter 2: GNU Fortran Command Options 31

-frecursive
Allow indirect recursion by forcing all local arrays to be allocated on the
stack. This flag cannot be used together with -fmax-stack-var-size= or
-fno-automatic.
-finit-local-zero
-finit-derived
-finit-integer=n
-finit-real=<zero|inf|-inf|nan|snan>
-finit-logical=<true|false>
-finit-character=n
The -finit-local-zero option instructs the compiler to initialize local
INTEGER, REAL, and COMPLEX variables to zero, LOGICAL variables to
false, and CHARACTER variables to a string of null bytes. Finer-grained
initialization options are provided by the -finit-integer=n, -finit-
real=<zero|inf|-inf|nan|snan> (which also initializes the real and
imaginary parts of local COMPLEX variables), -finit-logical=<true|false>,
and -finit-character=n (where n is an ASCII character value) options.
With -finit-derived, components of derived type variables will be initialized
according to these flags. Components whose type is not covered by an explicit
-finit-* flag will be treated as described above with -finit-local-zero.
These options do not initialize
• objects with the POINTER attribute
• allocatable arrays
• variables that appear in an EQUIVALENCE statement.
(These limitations may be removed in future releases).
Note that the -finit-real=nan option initializes REAL and COMPLEX variables
with a quiet NaN. For a signalling NaN use -finit-real=snan; note, however,
that compile-time optimizations may convert them into quiet NaN and that
trapping needs to be enabled (e.g. via -ffpe-trap).
The -finit-integer option will parse the value into an integer of type
INTEGER(kind=C_LONG) on the host. Said value is then assigned to the integer
variables in the Fortran code, which might result in wraparound if the value is
too large for the kind.
Finally, note that enabling any of the -finit-* options will silence warnings
that would have been emitted by -Wuninitialized for the affected local vari-
ables.
-falign-commons
By default, gfortran enforces proper alignment of all variables in a COMMON
block by padding them as needed. On certain platforms this is mandatory,
on others it increases performance. If a COMMON block is not declared with
consistent data types everywhere, this padding can cause trouble, and -fno-
align-commons can be used to disable automatic alignment. The same form
of this option should be used for all files that share a COMMON block. To avoid
32 The GNU Fortran Compiler

potential alignment issues in COMMON blocks, it is recommended to order objects

from largest to smallest.
-fno-protect-parens
By default the parentheses in expression are honored for all optimization levels
such that the compiler does not do any re-association. Using -fno-protect-
parens allows the compiler to reorder REAL and COMPLEX expressions to produce
faster code. Note that for the re-association optimization -fno-signed-zeros
and -fno-trapping-math need to be in effect. The parentheses protection is
enabled by default, unless -Ofast is given.
-frealloc-lhs
An allocatable left-hand side of an intrinsic assignment is automatically
(re)allocated if it is either unallocated or has a different shape. The option is
enabled by default except when -std=f95 is given. See also -Wrealloc-lhs.
-faggressive-function-elimination
Functions with identical argument lists are eliminated within statements, re-
gardless of whether these functions are marked PURE or not. For example, in
a = f(b,c) + f(b,c)
there will only be a single call to f. This option only works if -ffrontend-
optimize is in effect.
-ffrontend-optimize
This option performs front-end optimization, based on manipulating parts the
Fortran parse tree. Enabled by default by any -O option except -O0 and -Og.
Optimizations enabled by this option include:
• inlining calls to MATMUL,
• elimination of identical function calls within expressions,
• removing unnecessary calls to TRIM in comparisons and assignments,
• replacing TRIM(a) with a(1:LEN_TRIM(a)) and
• short-circuiting of logical operators (.AND. and .OR.).
It can be deselected by specifying -fno-frontend-optimize.
-ffrontend-loop-interchange
Attempt to interchange loops in the Fortran front end where profitable. Enabled
by default by any -O option. At the moment, this option only affects FORALL
and DO CONCURRENT statements with several forall triplets.
See Section “Options for Code Generation Conventions” in Using the GNU Compiler
Collection (GCC), for information on more options offered by the GBE shared by gfortran,
gcc, and other GNU compilers.

2.10 Options for interoperability with other languages

-fc-prototypes
This option will generate C prototypes from BIND(C) variable declarations,
types and procedure interfaces and writes them to standard output. ENUM is
not yet supported.
Chapter 2: GNU Fortran Command Options 33

The generated prototypes may need inclusion of an appropriate header, such

as <stdint.h> or <stdlib.h>. For types which are not specified using the
appropriate kind from the iso_c_binding module, a warning is added as a
comment to the code.
For function pointers, a pointer to a function returning int without an explicit
argument list is generated.
Example of use:
$ gfortran -fc-prototypes -fsyntax-only foo.f90 > foo.h
where the C code intended for interoperating with the Fortran code then uses
#include "foo.h".
-fc-prototypes-external
This option will generate C prototypes from external functions and subroutines
and write them to standard output. This may be useful for making sure that C
bindings to Fortran code are correct. This option does not generate prototypes
for BIND(C) procedures, use -fc-prototypes for that.
The generated prototypes may need inclusion of an appropriate header, such
as <stdint.h> or <stdlib.h>.
This is primarily meant for legacy code to ensure that existing C bindings match
what gfortran emits. The generated C prototypes should be correct for the
current version of the compiler, but may not match what other compilers or
earlier versions of gfortran need. For new developments, use of the BIND(C)
features is recommended.
Example of use:
$ gfortran -fc-prototypes-external -fsyntax-only foo.f > foo.h
where the C code intended for interoperating with the Fortran code then uses
#include "foo.h".

2.11 Environment variables affecting gfortran

The gfortran compiler currently does not make use of any environment variables to control
its operation above and beyond those that affect the operation of gcc.
See Section “Environment Variables Affecting GCC” in Using the GNU Compiler Col-
lection (GCC), for information on environment variables.
See Chapter 3 [Runtime], page 35, for environment variables that affect the run-time
behavior of programs compiled with GNU Fortran.
 35

3 Runtime: Influencing runtime behavior with

environment variables
The behavior of the gfortran can be influenced by environment variables.
Malformed environment variables are silently ignored.

3.1 TMPDIR—Directory for scratch files

When opening a file with STATUS='SCRATCH', GNU Fortran tries to create the file in one
of the potential directories by testing each directory in the order below.
1. The environment variable TMPDIR, if it exists.
2. On the MinGW target, the directory returned by the GetTempPath function. Alterna-
tively, on the Cygwin target, the TMP and TEMP environment variables, if they exist, in
that order.
3. The P_tmpdir macro if it is defined, otherwise the directory /tmp.

3.2 GFORTRAN_STDIN_UNIT—Unit number for standard input

This environment variable can be used to select the unit number preconnected to standard
input. This must be a positive integer. The default value is 5.

3.3 GFORTRAN_STDOUT_UNIT—Unit number for standard output

This environment variable can be used to select the unit number preconnected to standard
output. This must be a positive integer. The default value is 6.

3.4 GFORTRAN_STDERR_UNIT—Unit number for standard error

This environment variable can be used to select the unit number preconnected to standard
error. This must be a positive integer. The default value is 0.

3.5 GFORTRAN_UNBUFFERED_ALL—Do not buffer I/O on all units

This environment variable controls whether all I/O is unbuffered. If the first letter is ‘y’,
‘Y’ or ‘1’, all I/O is unbuffered. This will slow down small sequential reads and writes. If
the first letter is ‘n’, ‘N’ or ‘0’, I/O is buffered. This is the default.

3.6 GFORTRAN_UNBUFFERED_PRECONNECTED—Do not buffer I/O on

preconnected units
The environment variable named GFORTRAN_UNBUFFERED_PRECONNECTED controls whether
I/O on a preconnected unit (i.e. STDOUT or STDERR) is unbuffered. If the first letter is
‘y’, ‘Y’ or ‘1’, I/O is unbuffered. This will slow down small sequential reads and writes. If
the first letter is ‘n’, ‘N’ or ‘0’, I/O is buffered. This is the default.

3.7 GFORTRAN_SHOW_LOCUS—Show location for runtime errors

If the first letter is ‘y’, ‘Y’ or ‘1’, filename and line numbers for runtime errors are printed.
If the first letter is ‘n’, ‘N’ or ‘0’, do not print filename and line numbers for runtime errors.
The default is to print the location.
36 The GNU Fortran Compiler

3.8 GFORTRAN_OPTIONAL_PLUS—Print leading + where

permitted
If the first letter is ‘y’, ‘Y’ or ‘1’, a plus sign is printed where permitted by the Fortran
standard. If the first letter is ‘n’, ‘N’ or ‘0’, a plus sign is not printed in most cases. Default
is not to print plus signs.

3.9 GFORTRAN_LIST_SEPARATOR—Separator for list output

This environment variable specifies the separator when writing list-directed output. It may
contain any number of spaces and at most one comma. If you specify this on the command
line, be sure to quote spaces, as in
$ GFORTRAN_LIST_SEPARATOR=' , ' ./a.out
when a.out is the compiled Fortran program that you want to run. Default is a single
space.

3.10 GFORTRAN_CONVERT_UNIT—Set conversion for unformatted

I/O
By setting the GFORTRAN_CONVERT_UNIT variable, it is possible to change the representation
of data for unformatted files. The syntax for the GFORTRAN_CONVERT_UNIT variable for most
systems is:
GFORTRAN_CONVERT_UNIT: mode | mode ';' exception | exception ;
mode: 'native' | 'swap' | 'big_endian' | 'little_endian' ;
exception: mode ':' unit_list | unit_list ;
unit_list: unit_spec | unit_list unit_spec ;
unit_spec: INTEGER | INTEGER '-' INTEGER ;
The variable consists of an optional default mode, followed by a list of optional excep-
tions, which are separated by semicolons from the preceding default and each other. Each
exception consists of a format and a comma-separated list of units. Valid values for the
modes are the same as for the CONVERT specifier:
NATIVE Use the native format. This is the default.
SWAP Swap between little- and big-endian.
LITTLE_ENDIAN Use the little-endian format for unformatted files.
BIG_ENDIAN Use the big-endian format for unformatted files.
For POWER systems which support -mabi=ieeelongdouble, there are additional op-
tions, which can be combined with the others with commas. Those are
R16_IEEE Use IEEE 128-bit format for REAL(KIND=16).
R16_IBM Use IBM long double format for REAL(KIND=16).
A missing mode for an exception is taken to mean BIG_ENDIAN. Examples of values for
GFORTRAN_CONVERT_UNIT are:
'big_endian' Do all unformatted I/O in big endian mode.
'little_endian;native:10-20,25' Do all unformatted I/O in little endian mode,
except for units 10 to 20 and 25, which are in native format.
'10-20' Units 10 to 20 are big-endian, the rest is native.
'big_endian,r16_ibm' Do all unformatted I/O in big-endian mode and use IBM long
double for output of REAL(KIND=16) values.
Chapter 3: Runtime: Influencing runtime behavior with environment variables 37

Setting the environment variables should be done on the command line or via the export
command for sh-compatible shells and via setenv for csh-compatible shells.
Example for sh:
$ gfortran foo.f90
$ GFORTRAN_CONVERT_UNIT='big_endian;native:10-20' ./a.out
Example code for csh:
% gfortran foo.f90
% setenv GFORTRAN_CONVERT_UNIT 'big_endian;native:10-20'
% ./a.out
Using anything but the native representation for unformatted data carries a significant
speed overhead. If speed in this area matters to you, it is best if you use this only for data
that needs to be portable.
See Section 5.1.17 [CONVERT specifier], page 53, for an alternative way to specify the
data representation for unformatted files. See Section 2.8 [Runtime Options], page 24, for
setting a default data representation for the whole program. The CONVERT specifier overrides
the -fconvert compile options.
Note that the values specified via the GFORTRAN CONVERT UNIT environment vari-
able will override the CONVERT specifier in the open statement. This is to give control
over data formats to users who do not have the source code of their program available.

3.11 GFORTRAN_ERROR_BACKTRACE—Show backtrace on run-time

errors
If the GFORTRAN_ERROR_BACKTRACE variable is set to ‘y’, ‘Y’ or ‘1’ (only the first letter
is relevant) then a backtrace is printed when a serious run-time error occurs. To disable
the backtracing, set the variable to ‘n’, ‘N’, ‘0’. Default is to print a backtrace unless the
-fno-backtrace compile option was used.

3.12 GFORTRAN_FORMATTED_BUFFER_SIZE—Set buffer size for

formatted I/O
The GFORTRAN_FORMATTED_BUFFER_SIZE environment variable specifies buffer size in bytes
to be used for formatted output. The default value is 8192.

3.13 GFORTRAN_UNFORMATTED_BUFFER_SIZE—Set buffer size for

unformatted I/O
The GFORTRAN_UNFORMATTED_BUFFER_SIZE environment variable specifies buffer size in
bytes to be used for unformatted output. The default value is 131072.
Chapter 3: Runtime: Influencing runtime behavior with environment variables 39

Part II: Language Reference

 41

4 Compiler Characteristics
This chapter describes certain characteristics of the GNU Fortran compiler, that are not
specified by the Fortran standard, but which might in some way or another become visible
to the programmer.

4.1 KIND Type Parameters

The KIND type parameters supported by GNU Fortran for the primitive data types are:

INTEGER 1, 2, 4, 8*, 16*, default: 4**

LOGICAL 1, 2, 4, 8*, 16*, default: 4**
REAL 4, 8, 10*, 16*, default: 4***
COMPLEX 4, 8, 10*, 16*, default: 4***
DOUBLE PRECISION
4, 8, 10*, 16*, default: 8***
CHARACTER
1, 4, default: 1

* not available on all systems

** unless -fdefault-integer-8 is used
*** unless -fdefault-real-8 is used (see Section 2.2 [Fortran Dialect Options], page 8)
The KIND value matches the storage size in bytes, except for COMPLEX where the
storage size is twice as much (or both real and imaginary part are a real value of the
given size). It is recommended to use the Section 8.239 [SELECTED CHAR KIND],
page 263, Section 8.240 [SELECTED INT KIND], page 264, and Section 8.241
[SELECTED REAL KIND], page 264, intrinsics or the INT8, INT16, INT32, INT64,
REAL32, REAL64, and REAL128 parameters of the ISO_FORTRAN_ENV module instead of
the concrete values. The available kind parameters can be found in the constant arrays
CHARACTER_KINDS, INTEGER_KINDS, LOGICAL_KINDS and REAL_KINDS in the Section 9.1
[ISO FORTRAN ENV], page 293, module. For C interoperability, the kind parameters of
the Section 9.2 [ISO C BINDING], page 295, module should be used.

4.2 Internal representation of LOGICAL variables

The Fortran standard does not specify how variables of LOGICAL type are represented,
beyond requiring that LOGICAL variables of default kind have the same storage size as default
INTEGER and REAL variables. The GNU Fortran internal representation is as follows.
A LOGICAL(KIND=N) variable is represented as an INTEGER(KIND=N) variable, however,
with only two permissible values: 1 for .TRUE. and 0 for .FALSE.. Any other integer value
results in undefined behavior.
See also Section 6.4.2 [Argument passing conventions], page 81, and Section 6.1 [Inter-
operability with C], page 69.
42 The GNU Fortran Compiler

4.3 Evaluation of logical expressions

The Fortran standard does not require the compiler to evaluate all parts of an expression,
if they do not contribute to the final result. For logical expressions with .AND. or .OR.
operators, in particular, GNU Fortran will optimize out function calls (even to impure
functions) if the result of the expression can be established without them. However, since
not all compilers do that, and such an optimization can potentially modify the program
flow and subsequent results, GNU Fortran throws warnings for such situations with the
-Wfunction-elimination flag.

4.4 MAX and MIN intrinsics with REAL NaN arguments

The Fortran standard does not specify what the result of the MAX and MIN intrinsics are if
one of the arguments is a NaN. Accordingly, the GNU Fortran compiler does not specify that
either, as this allows for faster and more compact code to be generated. If the programmer
wishes to take some specific action in case one of the arguments is a NaN, it is necessary to
explicitly test the arguments before calling MAX or MIN, e.g. with the IEEE_IS_NAN function
from the intrinsic module IEEE_ARITHMETIC.

4.5 Thread-safety of the runtime library

GNU Fortran can be used in programs with multiple threads, e.g. by using OpenMP, by
calling OS thread handling functions via the ISO_C_BINDING facility, or by GNU Fortran
compiled library code being called from a multi-threaded program.
The GNU Fortran runtime library, (libgfortran), supports being called concurrently
from multiple threads with the following exceptions.
During library initialization, the C getenv function is used, which need not be thread-
safe. Similarly, the getenv function is used to implement the GET_ENVIRONMENT_VARIABLE
and GETENV intrinsics. It is the responsibility of the user to ensure that the environment is
not being updated concurrently when any of these actions are taking place.
The EXECUTE_COMMAND_LINE and SYSTEM intrinsics are implemented with the system
function, which need not be thread-safe. It is the responsibility of the user to ensure that
system is not called concurrently.
For platforms not supporting thread-safe POSIX functions, further functionality might
not be thread-safe. For details, please consult the documentation for your operating system.
The GNU Fortran runtime library uses various C library functions that depend on the
locale, such as strtod and snprintf. In order to work correctly in locale-aware programs
that set the locale using setlocale, the locale is reset to the default “C” locale while
executing a formatted READ or WRITE statement. On targets supporting the POSIX 2008
per-thread locale functions (e.g. newlocale, uselocale, freelocale), these are used and
thus the global locale set using setlocale or the per-thread locales in other threads are not
affected. However, on targets lacking this functionality, the global LC NUMERIC locale is
set to “C” during the formatted I/O. Thus, on such targets it’s not safe to call setlocale
concurrently from another thread while a Fortran formatted I/O operation is in progress.
Also, other threads doing something dependent on the LC NUMERIC locale might not
work correctly if a formatted I/O operation is in progress in another thread.
Chapter 4: Compiler Characteristics 43

4.6 Data consistency and durability

This section contains a brief overview of data and metadata consistency and durability
issues when doing I/O.
With respect to durability, GNU Fortran makes no effort to ensure that data is commit-
ted to stable storage. If this is required, the GNU Fortran programmer can use the intrinsic
FNUM to retrieve the low level file descriptor corresponding to an open Fortran unit. Then,
using e.g. the ISO_C_BINDING feature, one can call the underlying system call to flush
dirty data to stable storage, such as fsync on POSIX, _commit on MingW, or fcntl(fd,
F_FULLSYNC, 0) on Mac OS X. The following example shows how to call fsync:
! Declare the interface for POSIX fsync function
interface
function fsync (fd) bind(c,name="fsync")
use iso_c_binding, only: c_int
integer(c_int), value :: fd
integer(c_int) :: fsync
end function fsync
end interface

! Variable declaration
integer :: ret

! Opening unit 10
open (10,file="foo")

! ...
! Perform I/O on unit 10
! ...

! Flush and sync

flush(10)
ret = fsync(fnum(10))

! Handle possible error

if (ret /= 0) stop "Error calling FSYNC"
With respect to consistency, for regular files GNU Fortran uses buffered I/O in order
to improve performance. This buffer is flushed automatically when full and in some other
situations, e.g. when closing a unit. It can also be explicitly flushed with the FLUSH
statement. Also, the buffering can be turned off with the GFORTRAN_UNBUFFERED_ALL and
GFORTRAN_UNBUFFERED_PRECONNECTED environment variables. Special files, such as termi-
nals and pipes, are always unbuffered. Sometimes, however, further things may need to be
done in order to allow other processes to see data that GNU Fortran has written, as follows.
The Windows platform supports a relaxed metadata consistency model, where file meta-
data is written to the directory lazily. This means that, for instance, the dir command can
show a stale size for a file. One can force a directory metadata update by closing the unit,
or by calling _commit on the file descriptor. Note, though, that _commit will force all dirty
data to stable storage, which is often a very slow operation.
The Network File System (NFS) implements a relaxed consistency model called open-to-
close consistency. Closing a file forces dirty data and metadata to be flushed to the server,
and opening a file forces the client to contact the server in order to revalidate cached data.
fsync will also force a flush of dirty data and metadata to the server. Similar to open and
44 The GNU Fortran Compiler

close, acquiring and releasing fcntl file locks, if the server supports them, will also force
cache validation and flushing dirty data and metadata.

4.7 Files opened without an explicit ACTION= specifier

The Fortran standard says that if an OPEN statement is executed without an explicit ACTION=
specifier, the default value is processor dependent. GNU Fortran behaves as follows:
1. Attempt to open the file with ACTION='READWRITE'
2. If that fails, try to open with ACTION='READ'
3. If that fails, try to open with ACTION='WRITE'
4. If that fails, generate an error

4.8 File operations on symbolic links

This section documents the behavior of GNU Fortran for file operations on symbolic links,
on systems that support them.
• Results of INQUIRE statements of the “inquire by file” form will relate to the target
of the symbolic link. For example, INQUIRE(FILE="foo",EXIST=ex) will set ex to
.true. if foo is a symbolic link pointing to an existing file, and .false. if foo points to an
non-existing file (“dangling” symbolic link).
• Using the OPEN statement with a STATUS="NEW" specifier on a symbolic link will result
in an error condition, whether the symbolic link points to an existing target or is
dangling.
• If a symbolic link was connected, using the CLOSE statement with a STATUS="DELETE"
specifier will cause the symbolic link itself to be deleted, not its target.

4.9 File format of unformatted sequential files

Unformatted sequential files are stored as logical records using record markers. Each logical
record consists of one of more subrecords.
Each subrecord consists of a leading record marker, the data written by the user program,
and a trailing record marker. The record markers are four-byte integers by default, and
eight-byte integers if the -fmax-subrecord-length=8 option (which exists for backwards
compability only) is in effect.
The representation of the record markers is that of unformatted files given with the
-fconvert option, the Section 5.1.17 [CONVERT specifier], page 53, in an open statement
or the Section 3.10 [GFORTRAN CONVERT UNIT], page 36, environment variable.
The maximum number of bytes of user data in a subrecord is 2147483639 (2 GiB - 9) for
a four-byte record marker. This limit can be lowered with the -fmax-subrecord-length
option, although this is rarely useful. If the length of a logical record exceeds this limit, the
data is distributed among several subrecords.
The absolute of the number stored in the record markers is the number of bytes of user
data in the corresponding subrecord. If the leading record marker of a subrecord contains
a negative number, another subrecord follows the current one. If the trailing record marker
contains a negative number, then there is a preceding subrecord.
Chapter 4: Compiler Characteristics 45

In the most simple case, with only one subrecord per logical record, both record markers
contain the number of bytes of user data in the record.
The format for unformatted sequential data can be duplicated using unformatted stream,
as shown in the example program for an unformatted record containing a single subrecord:
program main
use iso_fortran_env, only: int32
implicit none
integer(int32) :: i
real, dimension(10) :: a, b
call random_number(a)
open (10,file='test.dat',form='unformatted',access='stream')
inquire (iolength=i) a
write (10) i, a, i
close (10)
open (10,file='test.dat',form='unformatted')
read (10) b
if (all (a == b)) print *,'success!'
end program main

4.10 Asynchronous I/O

Asynchronous I/O is supported if the program is linked against the POSIX thread library.
If that is not the case, all I/O is performed as synchronous. On systems which do not
support pthread condition variables, such as AIX, I/O is also performed as synchronous.
On some systems, such as Darwin or Solaris, the POSIX thread library is always linked
in, so asynchronous I/O is always performed. On other sytems, such as Linux, it is necessary
to specify -pthread, -lpthread or -fopenmp during the linking step.

4.11 Behavior on integer overflow

Integer overflow is prohibited by the Fortran standard. The behavior of gfortran on integer
overflow is undefined by default. Traditional code, like linear congruential pseudo-random
number generators in old programs that rely on specific, non-standard behavior may gen-
erate unexpected results. The -fsanitize=undefined option can be used to detect such
code at runtime.
It is recommended to use the intrinsic subroutine RANDOM_NUMBER for random number
generators or, if the old behavior is desired, to use the -fwrapv option. Note that this
option can impact performance.
 47

5 Extensions
The two sections below detail the extensions to standard Fortran that are implemented in
GNU Fortran, as well as some of the popular or historically important extensions that are
not (or not yet) implemented. For the latter case, we explain the alternatives available to
GNU Fortran users, including replacement by standard-conforming code or GNU extensions.

5.1 Extensions implemented in GNU Fortran

GNU Fortran implements a number of extensions over standard Fortran. This chapter
contains information on their syntax and meaning. There are currently two categories of
GNU Fortran extensions, those that provide functionality beyond that provided by any
standard, and those that are supported by GNU Fortran purely for backward compatibility
with legacy compilers. By default, -std=gnu allows the compiler to accept both types of
extensions, but to warn about the use of the latter. Specifying either -std=f95, -std=f2003,
-std=f2008, or -std=f2018 disables both types of extensions, and -std=legacy allows
both without warning. The special compile flag -fdec enables additional compatibility
extensions along with those enabled by -std=legacy.

5.1.1 Old-style kind specifications

GNU Fortran allows old-style kind specifications in declarations. These look like:
TYPESPEC*size x,y,z
where TYPESPEC is a basic type (INTEGER, REAL, etc.), and where size is a byte count
corresponding to the storage size of a valid kind for that type. (For COMPLEX variables,
size is the total size of the real and imaginary parts.) The statement then declares x, y
and z to be of type TYPESPEC with the appropriate kind. This is equivalent to the standard-
conforming declaration
TYPESPEC(k) x,y,z
where k is the kind parameter suitable for the intended precision. As kind parameters are
implementation-dependent, use the KIND, SELECTED_INT_KIND and SELECTED_REAL_KIND
intrinsics to retrieve the correct value, for instance REAL*8 x can be replaced by:
INTEGER, PARAMETER :: dbl = KIND(1.0d0)
REAL(KIND=dbl) :: x

5.1.2 Old-style variable initialization

GNU Fortran allows old-style initialization of variables of the form:
INTEGER i/1/,j/2/
REAL x(2,2) /3*0.,1./
The syntax for the initializers is as for the DATA statement, but unlike in a DATA state-
ment, an initializer only applies to the variable immediately preceding the initialization. In
other words, something like INTEGER I,J/2,3/ is not valid. This style of initialization is
only allowed in declarations without double colons (::); the double colons were introduced
in Fortran 90, which also introduced a standard syntax for initializing variables in type
declarations.
Examples of standard-conforming code equivalent to the above example are:
! Fortran 90
48 The GNU Fortran Compiler

INTEGER :: i = 1, j = 2
REAL :: x(2,2) = RESHAPE((/0.,0.,0.,1./),SHAPE(x))
! Fortran 77
INTEGER i, j
REAL x(2,2)
DATA i/1/, j/2/, x/3*0.,1./
Note that variables which are explicitly initialized in declarations or in DATA statements
automatically acquire the SAVE attribute.

5.1.3 Extensions to namelist

GNU Fortran fully supports the Fortran 95 standard for namelist I/O including array
qualifiers, substrings and fully qualified derived types. The output from a namelist write is
compatible with namelist read. The output has all names in upper case and indentation to
column 1 after the namelist name. Two extensions are permitted:
Old-style use of ‘$’ instead of ‘&’
$MYNML
X(:)%Y(2) = 1.0 2.0 3.0
CH(1:4) = "abcd"
$END
It should be noted that the default terminator is ‘/’ rather than ‘&END’.
Querying of the namelist when inputting from stdin. After at least one space, entering
‘?’ sends to stdout the namelist name and the names of the variables in the namelist:
?

&mynml
x
x%y
ch
&end
Entering ‘=?’ outputs the namelist to stdout, as if WRITE(*,NML = mynml) had been
called:
=?

&MYNML
X(1)%Y= 0.000000 , 1.000000 , 0.000000 ,
X(2)%Y= 0.000000 , 2.000000 , 0.000000 ,
X(3)%Y= 0.000000 , 3.000000 , 0.000000 ,
CH=abcd, /
To aid this dialog, when input is from stdin, errors send their messages to stderr and
execution continues, even if IOSTAT is set.
PRINT namelist is permitted. This causes an error if -std=f95 is used.
PROGRAM test_print
REAL, dimension (4) :: x = (/1.0, 2.0, 3.0, 4.0/)
NAMELIST /mynml/ x
PRINT mynml
END PROGRAM test_print
Expanded namelist reads are permitted. This causes an error if -std=f95 is used. In
the following example, the first element of the array will be given the value 0.00 and the
two succeeding elements will be given the values 1.00 and 2.00.
&MYNML
Chapter 5: Extensions 49

X(1,1) = 0.00 , 1.00 , 2.00

/
When writing a namelist, if no DELIM= is specified, by default a double quote is used to
delimit character strings. If -std=F95, F2003, or F2008, etc, the delim status is set to ’none’.
Defaulting to quotes ensures that namelists with character strings can be subsequently read
back in accurately.

5.1.4 X format descriptor without count field

To support legacy codes, GNU Fortran permits the count field of the X edit descriptor in
FORMAT statements to be omitted. When omitted, the count is implicitly assumed to be
one.
PRINT 10, 2, 3
10 FORMAT (I1, X, I1)

5.1.5 Commas in FORMAT specifications

To support legacy codes, GNU Fortran allows the comma separator to be omitted imme-
diately before and after character string edit descriptors in FORMAT statements. A comma
with no following format decriptor is permited if the -fdec-blank-format-item is given
on the command line. This is considered non-conforming code and is discouraged.
PRINT 10, 2, 3
10 FORMAT ('FOO='I1' BAR='I2)
print 20, 5, 6
20 FORMAT (I3, I3,)

5.1.6 Missing period in FORMAT specifications

To support legacy codes, GNU Fortran allows missing periods in format specifications if
and only if -std=legacy is given on the command line. This is considered non-conforming
code and is discouraged.
REAL :: value
READ(*,10) value
10 FORMAT ('F4')

5.1.7 Default widths for F, G and I format descriptors

To support legacy codes, GNU Fortran allows width to be omitted from format specifications
if and only if -fdec-format-defaults is given on the command line. Default widths will
be used. This is considered non-conforming code and is discouraged.
REAL :: value1
INTEGER :: value2
WRITE(*,10) value1, value1, value2
10 FORMAT ('F, G, I')

5.1.8 I/O item lists

To support legacy codes, GNU Fortran allows the input item list of the READ statement,
and the output item lists of the WRITE and PRINT statements, to start with a comma.

5.1.9 Q exponent-letter
GNU Fortran accepts real literal constants with an exponent-letter of Q, for example,
1.23Q45. The constant is interpreted as a REAL(16) entity on targets that support this
50 The GNU Fortran Compiler

type. If the target does not support REAL(16) but has a REAL(10) type, then the real-
literal-constant will be interpreted as a REAL(10) entity. In the absence of REAL(16) and
REAL(10), an error will occur.

5.1.10 BOZ literal constants

Besides decimal constants, Fortran also supports binary (b), octal (o) and hexadecimal (z)
integer constants. The syntax is: ‘prefix quote digits quote’, where the prefix is either
b, o or z, quote is either ' or " and the digits are 0 or 1 for binary, between 0 and 7 for
octal, and between 0 and F for hexadecimal. (Example: b'01011101'.)
Up to Fortran 95, BOZ literal constants were only allowed to initialize integer variables
in DATA statements. Since Fortran 2003 BOZ literal constants are also allowed as actual
arguments to the REAL, DBLE, INT and CMPLX intrinsic functions. The BOZ literal constant
is simply a string of bits, which is padded or truncated as needed, during conversion to a
numeric type. The Fortran standard states that the treatment of the sign bit is processor
dependent. Gfortran interprets the sign bit as a user would expect.
As a deprecated extension, GNU Fortran allows hexadecimal BOZ literal constants to
be specified using the X prefix. That the BOZ literal constant can also be specified by
adding a suffix to the string, for example, Z'ABC' and 'ABC'X are equivalent. Additionally,
as extension, BOZ literals are permitted in some contexts outside of DATA and the intrinsic
functions listed in the Fortran standard. Use -fallow-invalid-boz to enable the extension.

5.1.11 Real array indices

As an extension, GNU Fortran allows the use of REAL expressions or variables as array
indices.

5.1.12 Unary operators

As an extension, GNU Fortran allows unary plus and unary minus operators to appear as
the second operand of binary arithmetic operators without the need for parenthesis.
X = Y * -Z

5.1.13 Implicitly convert LOGICAL and INTEGER values

As an extension for backwards compatibility with other compilers, GNU Fortran allows the
implicit conversion of LOGICAL values to INTEGER values and vice versa. When converting
from a LOGICAL to an INTEGER, .FALSE. is interpreted as zero, and .TRUE. is interpreted as
one. When converting from INTEGER to LOGICAL, the value zero is interpreted as .FALSE.
and any nonzero value is interpreted as .TRUE..
LOGICAL :: l
l = 1
INTEGER :: i
i = .TRUE.
However, there is no implicit conversion of INTEGER values in if-statements, nor of
LOGICAL or INTEGER values in I/O operations.

5.1.14 Hollerith constants support

GNU Fortran supports Hollerith constants in assignments, DATA statements, function and
subroutine arguments. A Hollerith constant is written as a string of characters preceded
Chapter 5: Extensions 51

by an integer constant indicating the character count, and the letter H or h, and stored in
bytewise fashion in a numeric (INTEGER, REAL, or COMPLEX), LOGICAL or CHARACTER variable.
The constant will be padded with spaces or truncated to fit the size of the variable in which
it is stored.
Examples of valid uses of Hollerith constants:
complex*16 x(2)
data x /16Habcdefghijklmnop, 16Hqrstuvwxyz012345/
x(1) = 16HABCDEFGHIJKLMNOP
call foo (4h abc)

Examples of Hollerith constants:

integer*4 a
a = 0H ! Invalid, at least one character is needed.
a = 4HAB12 ! Valid
a = 8H12345678 ! Valid, but the Hollerith constant will be truncated.
a = 3Hxyz ! Valid, but the Hollerith constant will be padded.

In general, Hollerith constants were used to provide a rudimentary facility for handling
character strings in early Fortran compilers, prior to the introduction of CHARACTER variables
in Fortran 77; in those cases, the standard-compliant equivalent is to convert the program
to use proper character strings. On occasion, there may be a case where the intent is
specifically to initialize a numeric variable with a given byte sequence. In these cases, the
same result can be obtained by using the TRANSFER statement, as in this example.
integer(kind=4) :: a
a = transfer ("abcd", a) ! equivalent to: a = 4Habcd

The use of the -fdec option extends support of Hollerith constants to comparisons:
integer*4 a
a = 4hABCD
if (a .ne. 4habcd) then
write(*,*) "no match"
end if

Supported types are numeric (INTEGER, REAL, or COMPLEX), and CHARACTER.

5.1.15 Character conversion

Allowing character literals to be used in a similar way to Hollerith constants is a non-
standard extension. This feature is enabled using -fdec-char-conversions and only applies
to character literals of kind=1.
Character literals can be used in DATA statements and assignments with numeric
(INTEGER, REAL, or COMPLEX) or LOGICAL variables. Like Hollerith constants they are
copied byte-wise fashion. The constant will be padded with spaces or truncated to fit the
size of the variable in which it is stored.
Examples:
integer*4 x
data x / 'abcd' /

x = 'A' ! Will be padded.

x = 'ab1234' ! Will be truncated.
52 The GNU Fortran Compiler

5.1.16 Cray pointers

Cray pointers are part of a non-standard extension that provides a C-like pointer in Fortran.
This is accomplished through a pair of variables: an integer "pointer" that holds a memory
address, and a "pointee" that is used to dereference the pointer.
Pointer/pointee pairs are declared in statements of the form:
pointer ( <pointer> , <pointee> )
or,
pointer ( <pointer1> , <pointee1> ), ( <pointer2> , <pointee2> ), ...
The pointer is an integer that is intended to hold a memory address. The pointee may be
an array or scalar. If an assumed-size array is permitted within the scoping unit, a pointee
can be an assumed-size array. That is, the last dimension may be left unspecified by using
a * in place of a value. A pointee cannot be an assumed shape array. No space is allocated
for the pointee.
The pointee may have its type declared before or after the pointer statement, and its
array specification (if any) may be declared before, during, or after the pointer statement.
The pointer may be declared as an integer prior to the pointer statement. However, some
machines have default integer sizes that are different than the size of a pointer, and so the
following code is not portable:
integer ipt
pointer (ipt, iarr)
If a pointer is declared with a kind that is too small, the compiler will issue a warning;
the resulting binary will probably not work correctly, because the memory addresses stored
in the pointers may be truncated. It is safer to omit the first line of the above example;
if explicit declaration of ipt’s type is omitted, then the compiler will ensure that ipt is an
integer variable large enough to hold a pointer.
Pointer arithmetic is valid with Cray pointers, but it is not the same as C pointer arith-
metic. Cray pointers are just ordinary integers, so the user is responsible for determining
how many bytes to add to a pointer in order to increment it. Consider the following example:
real target(10)
real pointee(10)
pointer (ipt, pointee)
ipt = loc (target)
ipt = ipt + 1
The last statement does not set ipt to the address of target(1), as it would in C pointer
arithmetic. Adding 1 to ipt just adds one byte to the address stored in ipt.
Any expression involving the pointee will be translated to use the value stored in the
pointer as the base address.
To get the address of elements, this extension provides an intrinsic function LOC(). The
LOC() function is equivalent to the & operator in C, except the address is cast to an integer
type:
real ar(10)
pointer(ipt, arpte(10))
real arpte
ipt = loc(ar) ! Makes arpte is an alias for ar
arpte(1) = 1.0 ! Sets ar(1) to 1.0
Chapter 5: Extensions 53

The pointer can also be set by a call to the MALLOC intrinsic (see Section 8.184 [MALLOC],
page 230).
Cray pointees often are used to alias an existing variable. For example:
integer target(10)
integer iarr(10)
pointer (ipt, iarr)
ipt = loc(target)
As long as ipt remains unchanged, iarr is now an alias for target. The optimizer,
however, will not detect this aliasing, so it is unsafe to use iarr and target simultaneously.
Using a pointee in any way that violates the Fortran aliasing rules or assumptions is illegal.
It is the user’s responsibility to avoid doing this; the compiler works under the assumption
that no such aliasing occurs.
Cray pointers will work correctly when there is no aliasing (i.e., when they are used to
access a dynamically allocated block of memory), and also in any routine where a pointee
is used, but any variable with which it shares storage is not used. Code that violates these
rules may not run as the user intends. This is not a bug in the optimizer; any code that
violates the aliasing rules is illegal. (Note that this is not unique to GNU Fortran; any
Fortran compiler that supports Cray pointers will “incorrectly” optimize code with illegal
aliasing.)
There are a number of restrictions on the attributes that can be applied to Cray point-
ers and pointees. Pointees may not have the ALLOCATABLE, INTENT, OPTIONAL, DUMMY,
TARGET, INTRINSIC, or POINTER attributes. Pointers may not have the DIMENSION, POINTER,
TARGET, ALLOCATABLE, EXTERNAL, or INTRINSIC attributes, nor may they be function re-
sults. Pointees may not occur in more than one pointer statement. A pointee cannot be a
pointer. Pointees cannot occur in equivalence, common, or data statements.
A Cray pointer may also point to a function or a subroutine. For example, the following
excerpt is valid:
implicit none
external sub
pointer (subptr,subpte)
external subpte
subptr = loc(sub)
call subpte()
[...]
subroutine sub
[...]
end subroutine sub
A pointer may be modified during the course of a program, and this will change the
location to which the pointee refers. However, when pointees are passed as arguments, they
are treated as ordinary variables in the invoked function. Subsequent changes to the pointer
will not change the base address of the array that was passed.

5.1.17 CONVERT specifier

GNU Fortran allows the conversion of unformatted data between little- and big-endian
representation to facilitate moving of data between different systems. The conversion can
be indicated with the CONVERT specifier on the OPEN statement. See Section 3.10 [GFOR-
TRAN CONVERT UNIT], page 36, for an alternative way of specifying the data format
via an environment variable.
54 The GNU Fortran Compiler

Valid values for CONVERT on most systems are:

CONVERT='NATIVE' Use the native format. This is the default.
CONVERT='SWAP' Swap between little- and big-endian.
CONVERT='LITTLE_ENDIAN' Use the little-endian representation for unformatted files.
CONVERT='BIG_ENDIAN' Use the big-endian representation for unformatted files.
On POWER systems which support -mabi=ieeelongdouble, there are additional op-
tions, which can be combined with the others with commas. Those are
CONVERT='R16_IEEE' Use IEEE 128-bit format for REAL(KIND=16).
CONVERT='R16_IBM' Use IBM long double format for realREAL(KIND=16).
Using the option could look like this:
open(file='big.dat',form='unformatted',access='sequential', &
convert='big_endian')
The value of the conversion can be queried by using INQUIRE(CONVERT=ch). The values
returned are 'BIG_ENDIAN' and 'LITTLE_ENDIAN'.
CONVERT works between big- and little-endian for INTEGER values of all supported kinds
and for REAL on IEEE systems of kinds 4 and 8. Conversion between different “extended
double” types on different architectures such as m68k and x86 64, which GNU Fortran
supports as REAL(KIND=10) and REAL(KIND=16), will probably not work.
Note that the values specified via the GFORTRAN CONVERT UNIT environment vari-
able will override the CONVERT specifier in the open statement. This is to give control
over data formats to users who do not have the source code of their program available.
Using anything but the native representation for unformatted data carries a significant
speed overhead. If speed in this area matters to you, it is best if you use this only for data
that needs to be portable.

5.1.18 OpenMP
OpenMP (Open Multi-Processing) is an application programming interface (API) that sup-
ports multi-platform shared memory multiprocessing programming in C/C++ and Fortran
on many architectures, including Unix and Microsoft Windows platforms. It consists of a set
of compiler directives, library routines, and environment variables that influence run-time
behavior.
GNU Fortran strives to be compatible to the OpenMP Application Program Interface
v4.5 (https://openmp.org/specifications/).
To enable the processing of the OpenMP directive !$omp in free-form source code; the
c$omp, *$omp and !$omp directives in fixed form; the !$ conditional compilation sentinels in
free form; and the c$, *$ and !$ sentinels in fixed form, gfortran needs to be invoked with
the -fopenmp. This also arranges for automatic linking of the GNU Offloading and Multi
Processing Runtime Library Section “libgomp” in GNU Offloading and Multi Processing
Runtime Library.
The OpenMP Fortran runtime library routines are provided both in a form of a Fortran
90 module named omp_lib and in a form of a Fortran include file named omp_lib.h.
An example of a parallelized loop taken from Appendix A.1 of the OpenMP Application
Program Interface v2.5:
SUBROUTINE A1(N, A, B)
Chapter 5: Extensions 55

INTEGER I, N
REAL B(N), A(N)
!$OMP PARALLEL DO !I is private by default
DO I=2,N
B(I) = (A(I) + A(I-1)) / 2.0
ENDDO
!$OMP END PARALLEL DO
END SUBROUTINE A1
Please note:
• -fopenmp implies -frecursive, i.e., all local arrays will be allocated on the stack.
When porting existing code to OpenMP, this may lead to surprising results, especially
to segmentation faults if the stacksize is limited.
• On glibc-based systems, OpenMP enabled applications cannot be statically linked due
to limitations of the underlying pthreads-implementation. It might be possible to get
a working solution if -Wl,--whole-archive -lpthread -Wl,--no-whole-archive is
added to the command line. However, this is not supported by gcc and thus not
recommended.

5.1.19 OpenACC
OpenACC is an application programming interface (API) that supports offloading of code
to accelerator devices. It consists of a set of compiler directives, library routines, and
environment variables that influence run-time behavior.
GNU Fortran strives to be compatible to the OpenACC Application Programming In-
terface v2.6 (https://www.openacc.org/).
To enable the processing of the OpenACC directive !$acc in free-form source code; the
c$acc, *$acc and !$acc directives in fixed form; the !$ conditional compilation sentinels in
free form; and the c$, *$ and !$ sentinels in fixed form, gfortran needs to be invoked with
the -fopenacc. This also arranges for automatic linking of the GNU Offloading and Multi
Processing Runtime Library Section “libgomp” in GNU Offloading and Multi Processing
Runtime Library.
The OpenACC Fortran runtime library routines are provided both in a form of a Fortran
90 module named openacc and in a form of a Fortran include file named openacc_lib.h.

5.1.20 Argument list functions %VAL, %REF and %LOC

GNU Fortran supports argument list functions %VAL, %REF and %LOC statements, for back-
ward compatibility with g77. It is recommended that these should be used only for code
that is accessing facilities outside of GNU Fortran, such as operating system or windowing
facilities. It is best to constrain such uses to isolated portions of a program–portions that
deal specifically and exclusively with low-level, system-dependent facilities. Such portions
might well provide a portable interface for use by the program as a whole, but are them-
selves not portable, and should be thoroughly tested each time they are rebuilt using a new
compiler or version of a compiler.
%VAL passes a scalar argument by value, %REF passes it by reference and %LOC passes its
memory location. Since gfortran already passes scalar arguments by reference, %REF is in
effect a do-nothing. %LOC has the same effect as a Fortran pointer.
An example of passing an argument by value to a C subroutine foo.:
C
56 The GNU Fortran Compiler

C prototype void foo_ (float x);

C
external foo
real*4 x
x = 3.14159
call foo (%VAL (x))
end
For details refer to the g77 manual https://gcc.gnu.org/onlinedocs/gcc-3.4.6/
g77/index.html#Top.
Also, c_by_val.f and its partner c_by_val.c of the GNU Fortran testsuite are worth
a look.

5.1.21 Read/Write after EOF marker

Some legacy codes rely on allowing READ or WRITE after the EOF file marker in order to find
the end of a file. GNU Fortran normally rejects these codes with a run-time error message
and suggests the user consider BACKSPACE or REWIND to properly position the file before the
EOF marker. As an extension, the run-time error may be disabled using -std=legacy.

5.1.22 STRUCTURE and RECORD

Record structures are a pre-Fortran-90 vendor extension to create user-defined aggregate
data types. Support for record structures in GNU Fortran can be enabled with the -fdec-
structure compile flag. If you have a choice, you should instead use Fortran 90’s “derived
types”, which have a different syntax.
In many cases, record structures can easily be converted to derived types. To convert,
replace STRUCTURE /structure-name/ by TYPE type-name. Additionally, replace RECORD
/structure-name/ by TYPE(type-name). Finally, in the component access, replace the pe-
riod (.) by the percent sign (%).
Here is an example of code using the non portable record structure syntax:
! Declaring a structure named ``item'' and containing three fields:
! an integer ID, an description string and a floating-point price.
STRUCTURE /item/
INTEGER id
CHARACTER(LEN=200) description
REAL price
END STRUCTURE

! Define two variables, an single record of type ``item''

! named ``pear'', and an array of items named ``store_catalog''
RECORD /item/ pear, store_catalog(100)

! We can directly access the fields of both variables

pear.id = 92316
pear.description = "juicy D'Anjou pear"
pear.price = 0.15
store_catalog(7).id = 7831
store_catalog(7).description = "milk bottle"
store_catalog(7).price = 1.2
Chapter 5: Extensions 57

! We can also manipulate the whole structure

store_catalog(12) = pear
print *, store_catalog(12)
This code can easily be rewritten in the Fortran 90 syntax as following:
! ``STRUCTURE /name/ ... END STRUCTURE'' becomes
! ``TYPE name ... END TYPE''
TYPE item
INTEGER id
CHARACTER(LEN=200) description
REAL price
END TYPE

! ``RECORD /name/ variable'' becomes ``TYPE(name) variable''

TYPE(item) pear, store_catalog(100)

! Instead of using a dot (.) to access fields of a record, the

! standard syntax uses a percent sign (%)
pear%id = 92316
pear%description = "juicy D'Anjou pear"
pear%price = 0.15
store_catalog(7)%id = 7831
store_catalog(7)%description = "milk bottle"
store_catalog(7)%price = 1.2

! Assignments of a whole variable do not change

store_catalog(12) = pear
print *, store_catalog(12)
GNU Fortran implements STRUCTURES like derived types with the following rules and
exceptions:
• Structures act like derived types with the SEQUENCE attribute. Otherwise they may
contain no specifiers.
• Structures may contain a special field with the name %FILL. This will create an anony-
mous component which cannot be accessed but occupies space just as if a component of
the same type was declared in its place, useful for alignment purposes. As an example,
the following structure will consist of at least sixteen bytes:
structure /padded/
character(4) start
character(8) %FILL
character(4) end
end structure
• Structures may share names with other symbols. For example, the following is invalid
for derived types, but valid for structures:
structure /header/
! ...
end structure
record /header/ header
58 The GNU Fortran Compiler

• Structure types may be declared nested within another parent structure. The syntax
is:
structure /type-name/
...
structure [/<type-name>/] <field-list>
...

The type name may be ommitted, in which case the structure type itself is anonymous,
and other structures of the same type cannot be instantiated. The following shows
some examples:
structure /appointment/
! nested structure definition: app_time is an array of two 'time'
structure /time/ app_time (2)
integer(1) hour, minute
end structure
character(10) memo
end structure

! The 'time' structure is still usable

record /time/ now
now = time(5, 30)

...

structure /appointment/
! anonymous nested structure definition
structure start, end
integer(1) hour, minute
end structure
character(10) memo
end structure
• Structures may contain UNION blocks. For more detail see the section on Section 5.1.23
[UNION and MAP], page 58.
• Structures support old-style initialization of components, like those described in Sec-
tion 5.1.2 [Old-style variable initialization], page 47. For array initializers, an initial-
izer may contain a repeat specification of the form <literal-integer> * <constant-
initializer>. The value of the integer indicates the number of times to repeat the
constant initializer when expanding the initializer list.

5.1.23 UNION and MAP

Unions are an old vendor extension which were commonly used with the non-standard
Section 5.1.22 [STRUCTURE and RECORD], page 56, extensions. Use of UNION and MAP
is automatically enabled with -fdec-structure.
A UNION declaration occurs within a structure; within the definition of each union is a
number of MAP blocks. Each MAP shares storage with its sibling maps (in the same union),
and the size of the union is the size of the largest map within it, just as with unions in C.
Chapter 5: Extensions 59

The major difference is that component references do not indicate which union or map the
component is in (the compiler gets to figure that out).
Here is a small example:
structure /myunion/
union
map
character(2) w0, w1, w2
end map
map
character(6) long
end map
end union
end structure

record /myunion/ rec

! After this assignment...
rec.long = 'hello!'

! The following is true:

! rec.w0 === 'he'
! rec.w1 === 'll'
! rec.w2 === 'o!'
The two maps share memory, and the size of the union is ultimately six bytes:
0 1 2 3 4 5 6 Byte offset
-------------------------------
| | | | | | |
-------------------------------

^ W0 ^ W1 ^ W2 ^
\-------/ \-------/ \-------/

^ LONG ^
\---------------------------/
Following is an example mirroring the layout of an Intel x86 64 register:
structure /reg/
union ! U0 ! rax
map
character(16) rx
end map
map
character(8) rh ! rah
union ! U1
map
character(8) rl ! ral
end map
map
character(8) ex ! eax
end map
map
60 The GNU Fortran Compiler

character(4) eh ! eah
union ! U2
map
character(4) el ! eal
end map
map
character(4) x ! ax
end map
map
character(2) h ! ah
character(2) l ! al
end map
end union
end map
end union
end map
end union
end structure
record /reg/ a

! After this assignment...

a.rx = 'AAAAAAAA.BBB.C.D'

! The following is true:

a.rx === 'AAAAAAAA.BBB.C.D'
a.rh === 'AAAAAAAA'
a.rl === '.BBB.C.D'
a.ex === '.BBB.C.D'
a.eh === '.BBB'
a.el === '.C.D'
a.x === '.C.D'
a.h === '.C'
a.l === '.D'

5.1.24 Type variants for integer intrinsics

Similar to the D/C prefixes to real functions to specify the input/output types, GNU
Fortran offers B/I/J/K prefixes to integer functions for compatibility with DEC programs.
The types implied by each are:
B - INTEGER(kind=1)
I - INTEGER(kind=2)
J - INTEGER(kind=4)
K - INTEGER(kind=8)
GNU Fortran supports these with the flag -fdec-intrinsic-ints. Intrinsics for which
prefixed versions are available and in what form are noted in Chapter 8 [Intrinsic Proce-
dures], page 111. The complete list of supported intrinsics is here:
Chapter 5: Extensions 61

Intrinsic B I J K

Section 8.3 BABS IIABS JIABS KIABS

[ABS], page 112
Section 8.51 BBTEST BITEST BJTEST BKTEST
[BTEST],
page 144
Section 8.137 BIAND IIAND JIAND KIAND
[IAND],
page 202
Section 8.140 BBCLR IIBCLR JIBCLR KIBCLR
[IBCLR],
page 204
Section 8.141 BBITS IIBITS JIBITS KIBITS
[IBITS],
page 205
Section 8.142 BBSET IIBSET JIBSET KIBSET
[IBSET],
page 206
Section 8.145 BIEOR IIEOR JIEOR KIEOR
[IEOR],
page 208
Section 8.152 BIOR IIOR JIOR KIOR
[IOR], page 212
Section 8.159 BSHFT IISHFT JISHFT KISHFT
[ISHFT],
page 216
Section 8.160 BSHFTC IISHFTC JISHFTC KISHFTC
[ISHFTC],
page 217
Section 8.200 BMOD IMOD JMOD KMOD
[MOD], page 240
Section 8.208 BNOT INOT JNOT KNOT
[NOT], page 245
Section 8.228 -- FLOATI FLOATJ FLOATK
[REAL],
page 257

5.1.25 AUTOMATIC and STATIC attributes

With -fdec-static GNU Fortran supports the DEC extended attributes STATIC and
AUTOMATIC to provide explicit specification of entity storage. These follow the syntax of
the Fortran standard SAVE attribute.
STATIC is exactly equivalent to SAVE, and specifies that an entity should be allocated
in static memory. As an example, STATIC local variables will retain their values across
multiple calls to a function.
62 The GNU Fortran Compiler

Entities marked AUTOMATIC will be stack automatic whenever possible. AUTOMATIC is the
default for local variables smaller than -fmax-stack-var-size, unless -fno-automatic
is given. This attribute overrides -fno-automatic, -fmax-stack-var-size, and blanket
SAVE statements.
Examples:
subroutine f
integer, automatic :: i ! automatic variable
integer x, y ! static variables
save
...
endsubroutine
subroutine f
integer a, b, c, x, y, z
static :: x
save y
automatic z, c
! a, b, c, and z are automatic
! x and y are static
endsubroutine
! Compiled with -fno-automatic
subroutine f
integer a, b, c, d
automatic :: a
! a is automatic; b, c, and d are static
endsubroutine

5.1.26 Extended math intrinsics

GNU Fortran supports an extended list of mathematical intrinsics with the compile flag
-fdec-math for compatability with legacy code. These intrinsics are described fully in
Chapter 8 [Intrinsic Procedures], page 111, where it is noted that they are extensions and
should be avoided whenever possible.
Specifically, -fdec-math enables the Section 8.76 [COTAN], page 162, intrinsic, and
trigonometric intrinsics which accept or produce values in degrees instead of radians. Here
is a summary of the new intrinsics:
Radians Degrees
Section 8.6 [ACOS], page 114 Section 8.7 [ACOSD], page 114*
Section 8.19 [ASIN], page 122 Section 8.20 [ASIND], page 123*
Section 8.23 [ATAN], page 126 Section 8.24 [ATAND], page 126*
Section 8.25 [ATAN2], page 127 Section 8.26 [ATAN2D], page 128*
Section 8.73 [COS], page 160 Section 8.74 [COSD], page 160*
Section 8.76 [COTAN], page 162* Section 8.77 [COTAND], page 162*
Section 8.249 [SIN], page 269 Section 8.250 [SIND], page 270*
Section 8.265 [TAN], page 280 Section 8.266 [TAND], page 280*
* Enabled with -fdec-math.
Chapter 5: Extensions 63

For advanced users, it may be important to know the implementation of these functions.
They are simply wrappers around the standard radian functions, which have more accurate
builtin versions. These functions convert their arguments (or results) to degrees (or radians)
by taking the value modulus 360 (or 2*pi) and then multiplying it by a constant radian-to-
degree (or degree-to-radian) factor, as appropriate. The factor is computed at compile-time
as 180/pi (or pi/180).

5.1.27 Form feed as whitespace

Historically, legacy compilers allowed insertion of form feed characters (’\f’, ASCII 0xC) at
the beginning of lines for formatted output to line printers, though the Fortran standard
does not mention this. GNU Fortran supports the interpretation of form feed characters in
source as whitespace for compatibility.

5.1.28 TYPE as an alias for PRINT

For compatibility, GNU Fortran will interpret TYPE statements as PRINT statements with
the flag -fdec. With this flag asserted, the following two examples are equivalent:
TYPE *, 'hello world'
PRINT *, 'hello world'

5.1.29 %LOC as an rvalue

Normally %LOC is allowed only in parameter lists. However the intrinsic function LOC does
the same thing, and is usable as the right-hand-side of assignments. For compatibility, GNU
Fortran supports the use of %LOC as an alias for the builtin LOC with -std=legacy. With
this feature enabled the following two examples are equivalent:
integer :: i, l
l = %loc(i)
call sub(l)
integer :: i
call sub(%loc(i))

5.1.30 .XOR. operator

GNU Fortran supports .XOR. as a logical operator with -std=legacy for compatibility with
legacy code. .XOR. is equivalent to .NEQV.. That is, the output is true if and only if the
inputs differ.

5.1.31 Bitwise logical operators

With -fdec, GNU Fortran relaxes the type constraints on logical operators to allow inte-
ger operands, and performs the corresponding bitwise operation instead. This flag is for
compatibility only, and should be avoided in new code. Consider:
INTEGER :: i, j
i = z'33'
j = z'cc'
print *, i .AND. j
In this example, compiled with -fdec, GNU Fortran will replace the .AND. operation
with a call to the intrinsic Section 8.137 [IAND], page 202 function, yielding the bitwise-
and of i and j.
64 The GNU Fortran Compiler

Note that this conversion will occur if at least one operand is of integral type. As a result,
a logical operand will be converted to an integer when the other operand is an integer in a
logical operation. In this case, .TRUE. is converted to 1 and .FALSE. to 0.
Here is the mapping of logical operator to bitwise intrinsic used with -fdec:
Operator Intrinsic Bitwise operation
.NOT. Section 8.208 complement
[NOT], page 245
.AND. Section 8.137 intersection
[IAND], page 202
.OR. Section 8.152 union
[IOR], page 212
.NEQV. Section 8.145 exclusive or
[IEOR], page 208
.EQV. Section 8.208 complement of exclusive or
[NOT],
page 245(Section 8.145
[IEOR], page 208)

5.1.32 Extended I/O specifiers

GNU Fortran supports the additional legacy I/O specifiers CARRIAGECONTROL, READONLY,
and SHARE with the compile flag -fdec, for compatibility.
CARRIAGECONTROL
The CARRIAGECONTROL specifier allows a user to control line termination set-
tings between output records for an I/O unit. The specifier has no meaning for
readonly files. When CARRAIGECONTROL is specified upon opening a unit for for-
matted writing, the exact CARRIAGECONTROL setting determines what characters
to write between output records. The syntax is:
OPEN(..., CARRIAGECONTROL=cc)
Where cc is a character expression that evaluates to one of the following values:
'LIST' One line feed between records (default)
'FORTRAN' Legacy interpretation of the first character (see below)
'NONE' No separator between records
With CARRIAGECONTROL='FORTRAN', when a record is written, the first charac-
ter of the input record is not written, and instead determines the output record
separator as follows:
Leading character Meaning Output separating
character(s)
'+' Overprinting Carriage return only
'-' New line Line feed and carriage return
'0' Skip line Two line feeds and carriage
return
'1' New page Form feed and carriage
return
'$' Prompting Line feed (no carriage return)
Chapter 5: Extensions 65

CHAR(0) Overprinting (no None

advance)
READONLY The READONLY specifier may be given upon opening a unit, and is equivalent
to specifying ACTION='READ', except that the file may not be deleted on close
(i.e. CLOSE with STATUS="DELETE"). The syntax is:
OPEN(..., READONLY)

SHARE The SHARE specifier allows system-level locking on a unit upon opening it for
controlled access from multiple processes/threads. The SHARE specifier has
several forms:
OPEN(..., SHARE=sh)
OPEN(..., SHARED)
OPEN(..., NOSHARED)
Where sh in the first form is a character expression that evaluates to a value as
seen in the table below. The latter two forms are aliases for particular values
of sh:
Explicit form Short form Meaning
SHARE='DENYRW' NOSHARED Exclusive (write) lock
SHARE='DENYNONE' SHARED Shared (read) lock
In general only one process may hold an exclusive (write) lock for a given file
at a time, whereas many processes may hold shared (read) locks for the same
file.
The behavior of locking may vary with your operating system. On POSIX
systems, locking is implemented with fcntl. Consult your corresponding op-
erating system’s manual pages for further details. Locking via SHARE= is not
supported on other systems.

5.1.33 Legacy PARAMETER statements

For compatibility, GNU Fortran supports legacy PARAMETER statements without paren-
theses with -std=legacy. A warning is emitted if used with -std=gnu, and an error is
acknowledged with a real Fortran standard flag (-std=f95, etc...). These statements take
the following form:
implicit real (E)
parameter e = 2.718282
real c
parameter c = 3.0e8

5.1.34 Default exponents

For compatibility, GNU Fortran supports a default exponent of zero in real constants with
-fdec. For example, 9e would be interpreted as 9e0, rather than an error.

5.2 Extensions not implemented in GNU Fortran

The long history of the Fortran language, its wide use and broad userbase, the large num-
ber of different compiler vendors and the lack of some features crucial to users in the first
standards have lead to the existence of a number of important extensions to the language.
While some of the most useful or popular extensions are supported by the GNU Fortran
66 The GNU Fortran Compiler

compiler, not all existing extensions are supported. This section aims at listing these ex-
tensions and offering advice on how best make code that uses them running with the GNU
Fortran compiler.

5.2.1 ENCODE and DECODE statements

GNU Fortran does not support the ENCODE and DECODE statements. These statements are
best replaced by READ and WRITE statements involving internal files (CHARACTER variables
and arrays), which have been part of the Fortran standard since Fortran 77. For example,
replace a code fragment like
INTEGER*1 LINE(80)
REAL A, B, C
c ... Code that sets LINE
DECODE (80, 9000, LINE) A, B, C
9000 FORMAT (1X, 3(F10.5))
with the following:
CHARACTER(LEN=80) LINE
REAL A, B, C
c ... Code that sets LINE
READ (UNIT=LINE, FMT=9000) A, B, C
9000 FORMAT (1X, 3(F10.5))
Similarly, replace a code fragment like
INTEGER*1 LINE(80)
REAL A, B, C
c ... Code that sets A, B and C
ENCODE (80, 9000, LINE) A, B, C
9000 FORMAT (1X, 'OUTPUT IS ', 3(F10.5))
with the following:
CHARACTER(LEN=80) LINE
REAL A, B, C
c ... Code that sets A, B and C
WRITE (UNIT=LINE, FMT=9000) A, B, C
9000 FORMAT (1X, 'OUTPUT IS ', 3(F10.5))

5.2.2 Variable FORMAT expressions

A variable FORMAT expression is format statement which includes angle brackets enclosing a
Fortran expression: FORMAT(I<N>). GNU Fortran does not support this legacy extension.
The effect of variable format expressions can be reproduced by using the more powerful
(and standard) combination of internal output and string formats. For example, replace a
code fragment like this:
WRITE(6,20) INT1
20 FORMAT(I<N+1>)
with the following:
c Variable declaration
CHARACTER(LEN=20) FMT
c
c Other code here...
c
WRITE(FMT,'("(I", I0, ")")') N+1
WRITE(6,FMT) INT1
Chapter 5: Extensions 67

or with:
c Variable declaration
CHARACTER(LEN=20) FMT
c
c Other code here...
c
WRITE(FMT,*) N+1
WRITE(6,"(I" // ADJUSTL(FMT) // ")") INT1

5.2.3 Alternate complex function syntax

Some Fortran compilers, including g77, let the user declare complex functions with the
syntax COMPLEX FUNCTION name*16(), as well as COMPLEX*16 FUNCTION name(). Both are
non-standard, legacy extensions. gfortran accepts the latter form, which is more common,
but not the former.

5.2.4 Volatile COMMON blocks

Some Fortran compilers, including g77, let the user declare COMMON with the VOLATILE
attribute. This is invalid standard Fortran syntax and is not supported by gfortran. Note
that gfortran accepts VOLATILE variables in COMMON blocks since revision 4.3.

5.2.5 OPEN( ... NAME=)

Some Fortran compilers, including g77, let the user declare OPEN( ... NAME=). This is
invalid standard Fortran syntax and is not supported by gfortran. OPEN( ... NAME=)
should be replaced with OPEN( ... FILE=).

5.2.6 Q edit descriptor

Some Fortran compilers provide the Q edit descriptor, which transfers the number of char-
acters left within an input record into an integer variable.
A direct replacement of the Q edit descriptor is not available in gfortran. How to
replicate its functionality using standard-conforming code depends on what the intent of
the original code is.
Options to replace Q may be to read the whole line into a character variable and then
counting the number of non-blank characters left using LEN_TRIM. Another method may be
to use formatted stream, read the data up to the position where the Q descriptor occurred,
use INQUIRE to get the file position, count the characters up to the next NEW_LINE and then
start reading from the position marked previously.
 69

6 Mixed-Language Programming
This chapter is about mixed-language interoperability, but also applies if you link Fortran
code compiled by different compilers. In most cases, use of the C Binding features of the
Fortran 2003 and later standards is sufficient.
For example, it is possible to mix Fortran code with C++ code as well as C, if you declare
the interface functions as extern "C" on the C++ side and BIND(C) on the Fortran side,
and follow the rules for interoperability with C. Note that you cannot manipulate C++ class
objects in Fortran or vice versa except as opaque pointers.
You can use the gfortran command to link both Fortran and non-Fortran code into the
same program, or you can use gcc or g++ if you also add an explicit -lgfortran option to
link with the Fortran library. If your main program is written in C or some other language
instead of Fortran, see Section 6.3 [Non-Fortran Main Program], page 76, below.

6.1 Interoperability with C

Since Fortran 2003 (ISO/IEC 1539-1:2004(E)) there is a standardized way to generate
procedure and derived-type declarations and global variables that are interoperable with C
(ISO/IEC 9899:1999). The BIND(C) attribute has been added to inform the compiler that
a symbol shall be interoperable with C; also, some constraints are added. Note, however,
that not all C features have a Fortran equivalent or vice versa. For instance, neither C’s
unsigned integers nor C’s functions with variable number of arguments have an equivalent
in Fortran.
Note that array dimensions are reversely ordered in C and that arrays in C always start
with index 0 while in Fortran they start by default with 1. Thus, an array declaration
A(n,m) in Fortran matches A[m][n] in C and accessing the element A(i,j) matches A[j-
1][i-1]. The element following A(i,j) (C: A[j-1][i-1]; assuming i < n) in memory is
A(i+1,j) (C: A[j-1][i]).

6.1.1 Intrinsic Types

In order to ensure that exactly the same variable type and kind is used in C and Fortran, you
should use the named constants for kind parameters that are defined in the ISO_C_BINDING
intrinsic module. That module contains named constants of character type representing the
escaped special characters in C, such as newline. For a list of the constants, see Section 9.2
[ISO C BINDING], page 295.
For logical types, please note that the Fortran standard only guarantees interoperability
between C99’s _Bool and Fortran’s C_Bool-kind logicals and C99 defines that true has the
value 1 and false the value 0. Using any other integer value with GNU Fortran’s LOGICAL
(with any kind parameter) gives an undefined result. (Passing other integer values than 0
and 1 to GCC’s _Bool is also undefined, unless the integer is explicitly or implicitly casted
to _Bool.)

6.1.2 Derived Types and struct

For compatibility of derived types with struct, use the BIND(C) attribute in the type
declaration. For instance, the following type declaration
USE ISO_C_BINDING
70 The GNU Fortran Compiler

TYPE, BIND(C) :: myType

INTEGER(C_INT) :: i1, i2
INTEGER(C_SIGNED_CHAR) :: i3
REAL(C_DOUBLE) :: d1
COMPLEX(C_FLOAT_COMPLEX) :: c1
CHARACTER(KIND=C_CHAR) :: str(5)
END TYPE
matches the following struct declaration in C
struct {
int i1, i2;
/* Note: "char" might be signed or unsigned. */
signed char i3;
double d1;
float _Complex c1;
char str[5];
} myType;
Derived types with the C binding attribute shall not have the sequence attribute, type
parameters, the extends attribute, nor type-bound procedures. Every component must be
of interoperable type and kind and may not have the pointer or allocatable attribute.
The names of the components are irrelevant for interoperability.
As there exist no direct Fortran equivalents, neither unions nor structs with bit field or
variable-length array members are interoperable.

6.1.3 Interoperable Global Variables

Variables can be made accessible from C using the C binding attribute, optionally together
with specifying a binding name. Those variables have to be declared in the declaration part
of a MODULE, be of interoperable type, and have neither the pointer nor the allocatable
attribute.
MODULE m
USE myType_module
USE ISO_C_BINDING
integer(C_INT), bind(C, name="_MyProject_flags") :: global_flag
type(myType), bind(C) :: tp
END MODULE
Here, _MyProject_flags is the case-sensitive name of the variable as seen from C pro-
grams while global_flag is the case-insensitive name as seen from Fortran. If no binding
name is specified, as for tp, the C binding name is the (lowercase) Fortran binding name.
If a binding name is specified, only a single variable may be after the double colon. Note of
warning: You cannot use a global variable to access errno of the C library as the C standard
allows it to be a macro. Use the IERRNO intrinsic (GNU extension) instead.

6.1.4 Interoperable Subroutines and Functions

Subroutines and functions have to have the BIND(C) attribute to be compatible with C.
The dummy argument declaration is relatively straightforward. However, one needs to be
careful because C uses call-by-value by default while Fortran behaves usually similar to
call-by-reference. Furthermore, strings and pointers are handled differently.
To pass a variable by value, use the VALUE attribute. Thus, the following C prototype
int func(int i, int *j)
Chapter 6: Mixed-Language Programming 71

matches the Fortran declaration

integer(c_int) function func(i,j)
use iso_c_binding, only: c_int
integer(c_int), VALUE :: i
integer(c_int) :: j
Note that pointer arguments also frequently need the VALUE attribute, see Section 6.1.5
[Working with C Pointers], page 72.
Strings are handled quite differently in C and Fortran. In C a string is a NUL-terminated
array of characters while in Fortran each string has a length associated with it and is thus
not terminated (by e.g. NUL). For example, if you want to use the following C function,
#include <stdio.h>
void print_C(char *string) /* equivalent: char string[] */
{
printf("%s\n", string);
}
to print “Hello World” from Fortran, you can call it using
use iso_c_binding, only: C_CHAR, C_NULL_CHAR
interface
subroutine print_c(string) bind(C, name="print_C")
use iso_c_binding, only: c_char
character(kind=c_char) :: string(*)
end subroutine print_c
end interface
call print_c(C_CHAR_"Hello World"//C_NULL_CHAR)
As the example shows, you need to ensure that the string is NUL terminated. Addi-
tionally, the dummy argument string of print_C is a length-one assumed-size array; using
character(len=*) is not allowed. The example above uses c_char_"Hello World" to en-
sure the string literal has the right type; typically the default character kind and c_char are
the same and thus "Hello World" is equivalent. However, the standard does not guarantee
this.
The use of strings is now further illustrated using the C library function strncpy, whose
prototype is
char *strncpy(char *restrict s1, const char *restrict s2, size_t n);
The function strncpy copies at most n characters from string s2 to s1 and returns s1. In
the following example, we ignore the return value:
use iso_c_binding
implicit none
character(len=30) :: str,str2
interface
! Ignore the return value of strncpy -> subroutine
! "restrict" is always assumed if we do not pass a pointer
subroutine strncpy(dest, src, n) bind(C)
import
character(kind=c_char), intent(out) :: dest(*)
character(kind=c_char), intent(in) :: src(*)
integer(c_size_t), value, intent(in) :: n
end subroutine strncpy
end interface
str = repeat('X',30) ! Initialize whole string with 'X'
call strncpy(str, c_char_"Hello World"//C_NULL_CHAR, &
len(c_char_"Hello World",kind=c_size_t))
72 The GNU Fortran Compiler

print '(a)', str ! prints: "Hello WorldXXXXXXXXXXXXXXXXXXX"

end
The intrinsic procedures are described in Chapter 8 [Intrinsic Procedures], page 111.

6.1.5 Working with C Pointers

C pointers are represented in Fortran via the special opaque derived type type(c_ptr)
(with private components). C pointers are distinct from Fortran objects with the POINTER
attribute. Thus one needs to use intrinsic conversion procedures to convert from or to C
pointers. For some applications, using an assumed type (TYPE(*)) can be an alternative to
a C pointer, and you can also use library routines to access Fortran pointers from C. See
Section 6.1.6 [Further Interoperability of Fortran with C], page 74.
Here is an example of using C pointers in Fortran:
use iso_c_binding
type(c_ptr) :: cptr1, cptr2
integer, target :: array(7), scalar
integer, pointer :: pa(:), ps
cptr1 = c_loc(array(1)) ! The programmer needs to ensure that the
! array is contiguous if required by the C
! procedure
cptr2 = c_loc(scalar)
call c_f_pointer(cptr2, ps)
call c_f_pointer(cptr2, pa, shape=[7])
When converting C to Fortran arrays, the one-dimensional SHAPE argument has to be
passed.
If a pointer is a dummy argument of an interoperable procedure, it usually has to be
declared using the VALUE attribute. void* matches TYPE(C_PTR), VALUE, while TYPE(C_
PTR) alone matches void**.
Procedure pointers are handled analogously to pointers; the C type is TYPE(C_FUNPTR)
and the intrinsic conversion procedures are C_F_PROCPOINTER and C_FUNLOC.
Let us consider two examples of actually passing a procedure pointer from C to Fortran
and vice versa. Note that these examples are also very similar to passing ordinary pointers
between both languages. First, consider this code in C:
/* Procedure implemented in Fortran. */
void get_values (void (*)(double));

/* Call-back routine we want called from Fortran. */

void
print_it (double x)
{
printf ("Number is %f.\n", x);
}

/* Call Fortran routine and pass call-back to it. */

void
foobar ()
{
get_values (&print_it);
}
A matching implementation for get_values in Fortran, that correctly receives the pro-
cedure pointer from C and is able to call it, is given in the following MODULE:
MODULE m
Chapter 6: Mixed-Language Programming 73

IMPLICIT NONE

! Define interface of call-back routine.

ABSTRACT INTERFACE
SUBROUTINE callback (x)
USE, INTRINSIC :: ISO_C_BINDING
REAL(KIND=C_DOUBLE), INTENT(IN), VALUE :: x
END SUBROUTINE callback
END INTERFACE

CONTAINS

! Define C-bound procedure.

SUBROUTINE get_values (cproc) BIND(C)
USE, INTRINSIC :: ISO_C_BINDING
TYPE(C_FUNPTR), INTENT(IN), VALUE :: cproc

PROCEDURE(callback), POINTER :: proc

! Convert C to Fortran procedure pointer.

CALL C_F_PROCPOINTER (cproc, proc)

! Call it.
CALL proc (1.0_C_DOUBLE)
CALL proc (-42.0_C_DOUBLE)
CALL proc (18.12_C_DOUBLE)
END SUBROUTINE get_values

END MODULE m
Next, we want to call a C routine that expects a procedure pointer argument and pass
it a Fortran procedure (which clearly must be interoperable!). Again, the C function may
be:
int
call_it (int (*func)(int), int arg)
{
return func (arg);
}
It can be used as in the following Fortran code:
MODULE m
USE, INTRINSIC :: ISO_C_BINDING
IMPLICIT NONE

! Define interface of C function.

INTERFACE
INTEGER(KIND=C_INT) FUNCTION call_it (func, arg) BIND(C)
USE, INTRINSIC :: ISO_C_BINDING
TYPE(C_FUNPTR), INTENT(IN), VALUE :: func
INTEGER(KIND=C_INT), INTENT(IN), VALUE :: arg
END FUNCTION call_it
END INTERFACE

CONTAINS

! Define procedure passed to C function.

! It must be interoperable!
INTEGER(KIND=C_INT) FUNCTION double_it (arg) BIND(C)
74 The GNU Fortran Compiler

INTEGER(KIND=C_INT), INTENT(IN), VALUE :: arg

double_it = arg + arg
END FUNCTION double_it

! Call C function.
SUBROUTINE foobar ()
TYPE(C_FUNPTR) :: cproc
INTEGER(KIND=C_INT) :: i

! Get C procedure pointer.

cproc = C_FUNLOC (double_it)

! Use it.
DO i = 1_C_INT, 10_C_INT
PRINT *, call_it (cproc, i)
END DO
END SUBROUTINE foobar

END MODULE m

6.1.6 Further Interoperability of Fortran with C

GNU Fortran implements the Technical Specification ISO/IEC TS 29113:2012, which ex-
tends the interoperability support of Fortran 2003 and Fortran 2008 and is now part of the
2018 Fortran standard. Besides removing some restrictions and constraints, the Technical
Specification adds assumed-type (TYPE(*)) and assumed-rank (DIMENSION(..)) variables
and allows for interoperability of assumed-shape, assumed-rank, and deferred-shape arrays,
as well as allocatables and pointers. Objects of these types are passed to BIND(C) func-
tions as descriptors with a standard interface, declared in the header file <ISO_Fortran_
binding.h>.
Note: Currently, GNU Fortran does not use internally the array descriptor (dope vec-
tor) as specified in the Technical Specification, but uses an array descriptor with different
fields in functions without the BIND(C) attribute. Arguments to functions marked BIND(C)
are converted to the specified form. If you need to access GNU Fortran’s internal array de-
scriptor, you can use the Chasm Language Interoperability Tools, http://chasm-interop.
sourceforge.net/.

6.2 GNU Fortran Compiler Directives

6.2.1 ATTRIBUTES directive
The Fortran standard describes how a conforming program shall behave; however, the
exact implementation is not standardized. In order to allow the user to choose specific
implementation details, compiler directives can be used to set attributes of variables and
procedures which are not part of the standard. Whether a given attribute is supported and
its exact effects depend on both the operating system and on the processor; see Section “C
Extensions” in Using the GNU Compiler Collection (GCC) for details.
For procedures and procedure pointers, the following attributes can be used to change
the calling convention:
• CDECL – standard C calling convention
• STDCALL – convention where the called procedure pops the stack
Chapter 6: Mixed-Language Programming 75

• FASTCALL – part of the arguments are passed via registers instead using the stack
Besides changing the calling convention, the attributes also influence the decoration of
the symbol name, e.g., by a leading underscore or by a trailing at-sign followed by the
number of bytes on the stack. When assigning a procedure to a procedure pointer, both
should use the same calling convention.
On some systems, procedures and global variables (module variables and COMMON blocks)
need special handling to be accessible when they are in a shared library. The following
attributes are available:
• DLLEXPORT – provide a global pointer to a pointer in the DLL
• DLLIMPORT – reference the function or variable using a global pointer
For dummy arguments, the NO_ARG_CHECK attribute can be used; in other compilers, it
is also known as IGNORE_TKR. For dummy arguments with this attribute actual arguments
of any type and kind (similar to TYPE(*)), scalars and arrays of any rank (no equivalent in
Fortran standard) are accepted. As with TYPE(*), the argument is unlimited polymorphic
and no type information is available. Additionally, the argument may only be passed
to dummy arguments with the NO_ARG_CHECK attribute and as argument to the PRESENT
intrinsic function and to C_LOC of the ISO_C_BINDING module.
Variables with NO_ARG_CHECK attribute shall be of assumed-type (TYPE(*);
recommended) or of type INTEGER, LOGICAL, REAL or COMPLEX. They shall not have the
ALLOCATE, CODIMENSION, INTENT(OUT), POINTER or VALUE attribute; furthermore, they
shall be either scalar or of assumed-size (dimension(*)). As TYPE(*), the NO_ARG_CHECK
attribute requires an explicit interface.
• NO_ARG_CHECK – disable the type, kind and rank checking
• DEPRECATED – print a warning when using a such-tagged deprecated procedure, variable
or parameter; the warning can be suppressed with -Wno-deprecated-declarations.
• NOINLINE – prevent inlining given function.
• NORETURN – add a hint that a given function cannot return.
• WEAK – emit the declaration of an external symbol as a weak symbol rather than a
global. This is primarily useful in defining library functions that can be overridden in
user code, though it can also be used with non-function declarations. The overriding
symbol must have the same type as the weak symbol.
The attributes are specified using the syntax
!GCC$ ATTRIBUTES attribute-list :: variable-list
where in free-form source code only whitespace is allowed before !GCC$ and in fixed-form
source code !GCC$, cGCC$ or *GCC$ shall start in the first column.
For procedures, the compiler directives shall be placed into the body of the procedure; for
variables and procedure pointers, they shall be in the same declaration part as the variable
or procedure pointer.

6.2.2 UNROLL directive

The syntax of the directive is
!GCC$ unroll N
76 The GNU Fortran Compiler

You can use this directive to control how many times a loop should be unrolled. It must
be placed immediately before a DO loop and applies only to the loop that follows. N is an
integer constant specifying the unrolling factor. The values of 0 and 1 block any unrolling
of the loop.

6.2.3 BUILTIN directive

The syntax of the directive is
!GCC$ BUILTIN (B) attributes simd FLAGS IF('target')
You can use this directive to define which middle-end built-ins provide vector imple-
mentations. B is name of the middle-end built-in. FLAGS are optional and must be either
"(inbranch)" or "(notinbranch)". IF statement is optional and is used to filter multilib
ABIs for the built-in that should be vectorized. Example usage:
!GCC$ builtin (sinf) attributes simd (notinbranch) if('x86_64')
The purpose of the directive is to provide an API among the GCC compiler and the
GNU C Library which would define vector implementations of math routines.

6.2.4 IVDEP directive

The syntax of the directive is
!GCC$ ivdep
This directive tells the compiler to ignore vector dependencies in the following loop. It
must be placed immediately before a DO loop and applies only to the loop that follows.
Sometimes the compiler may not have sufficient information to decide whether a partic-
ular loop is vectorizable due to potential dependencies between iterations. The purpose of
the directive is to tell the compiler that vectorization is safe.
This directive is intended for annotation of existing code. For new code it is recom-
mended to consider OpenMP SIMD directives as potential alternative.

6.2.5 VECTOR directive

The syntax of the directive is
!GCC$ vector
This directive tells the compiler to vectorize the following loop. It must be placed
immediately before a DO loop and applies only to the loop that follows.

6.2.6 NOVECTOR directive

The syntax of the directive is
!GCC$ novector
This directive tells the compiler to not vectorize the following loop. It must be placed
immediately before a DO loop and applies only to the loop that follows.

6.3 Non-Fortran Main Program

Even if you are doing mixed-language programming, it is very likely that you do not need
to know or use the information in this section. Since it is about the internal structure of
GNU Fortran, it may also change in GCC minor releases.
Chapter 6: Mixed-Language Programming 77

When you compile a PROGRAM with GNU Fortran, a function with the name main (in
the symbol table of the object file) is generated, which initializes the libgfortran library
and then calls the actual program which uses the name MAIN__, for historic reasons. If
you link GNU Fortran compiled procedures to, e.g., a C or C++ program or to a Fortran
program compiled by a different compiler, the libgfortran library is not initialized and thus
a few intrinsic procedures do not work properly, e.g. those for obtaining the command-line
arguments.
Therefore, if your PROGRAM is not compiled with GNU Fortran and the GNU Fortran
compiled procedures require intrinsics relying on the library initialization, you need to
initialize the library yourself. Using the default options, gfortran calls _gfortran_set_
args and _gfortran_set_options. The initialization of the former is needed if the called
procedures access the command line (and for backtracing); the latter sets some flags based
on the standard chosen or to enable backtracing. In typical programs, it is not necessary
to call any initialization function.
If your PROGRAM is compiled with GNU Fortran, you shall not call any of the follow-
ing functions. The libgfortran initialization functions are shown in C syntax but using C
bindings they are also accessible from Fortran.

6.3.1 _gfortran_set_args — Save command-line arguments

Description:
_gfortran_set_args saves the command-line arguments; this initialization is
required if any of the command-line intrinsics is called. Additionally, it shall
be called if backtracing is enabled (see _gfortran_set_options).
Syntax : void _gfortran_set_args (int argc, char *argv[])
Arguments:
argc number of command line argument strings
argv the command-line argument strings; argv[0] is the
pathname of the executable itself.
Example:
int main (int argc, char *argv[])
{
/* Initialize libgfortran. */
_gfortran_set_args (argc, argv);
return 0;
}

6.3.2 _gfortran_set_options — Set library option flags

Description:
_gfortran_set_options sets several flags related to the Fortran standard to be
used, whether backtracing should be enabled and whether range checks should
be performed. The syntax allows for upward compatibility since the number of
passed flags is specified; for non-passed flags, the default value is used. See also
see Section 2.9 [Code Gen Options], page 25. Please note that not all flags are
actually used.
Syntax : void _gfortran_set_options (int num, int options[])
78 The GNU Fortran Compiler

Arguments:
num number of options passed
argv The list of flag values

option flag list:

option[0] Allowed standard; can give run-time errors if e.g.
an input-output edit descriptor is invalid in a
given standard. Possible values are (bitwise or-ed)
GFC_STD_F77 (1), GFC_STD_F95_OBS (2), GFC_STD_
F95_DEL (4), GFC_STD_F95 (8), GFC_STD_F2003
(16), GFC_STD_GNU (32), GFC_STD_LEGACY (64),
GFC_STD_F2008 (128), GFC_STD_F2008_OBS (256),
GFC_STD_F2008_TS (512), GFC_STD_F2018 (1024),
GFC_STD_F2018_OBS (2048), and GFC_STD=F2018_
DEL (4096). Default: GFC_STD_F95_OBS |
GFC_STD_F95_DEL | GFC_STD_F95 | GFC_STD_F2003
| GFC_STD_F2008 | GFC_STD_F2008_TS | GFC_
STD_F2008_OBS | GFC_STD_F77 | GFC_STD_F2018
| GFC_STD_F2018_OBS | GFC_STD_F2018_DEL |
GFC_STD_GNU | GFC_STD_LEGACY.
option[1] Standard-warning flag; prints a warning to stan-
dard error. Default: GFC_STD_F95_DEL | GFC_STD_
LEGACY.
option[2] If non zero, enable pedantic checking. Default: off.
option[3] Unused.
option[4] If non zero, enable backtracing on run-time errors.
Default: off. (Default in the compiler: on.) Note:
Installs a signal handler and requires command-line
initialization using _gfortran_set_args.
option[5] If non zero, supports signed zeros. Default: enabled.
option[6] Enables run-time checking. Possible values are
(bitwise or-ed): GFC RTCHECK BOUNDS
(1), GFC RTCHECK ARRAY TEMPS
(2), GFC RTCHECK RECURSION
(4), GFC RTCHECK DO (8),
GFC RTCHECK POINTER (16),
GFC RTCHECK MEM (32),
GFC RTCHECK BITS (64). Default:
disabled.
option[7] Unused.
Chapter 6: Mixed-Language Programming 79

option[8] Show a warning when invoking STOP and ERROR

STOP if a floating-point exception occurred. Possible
values are (bitwise or-ed) GFC_FPE_INVALID
(1), GFC_FPE_DENORMAL (2), GFC_FPE_ZERO (4),
GFC_FPE_OVERFLOW (8), GFC_FPE_UNDERFLOW (16),
GFC_FPE_INEXACT (32). Default: None (0). (Default
in the compiler: GFC_FPE_INVALID | GFC_FPE_
DENORMAL | GFC_FPE_ZERO | GFC_FPE_OVERFLOW |
GFC_FPE_UNDERFLOW.)
Example:
/* Use gfortran 4.9 default options. */
static int options[] = {68, 511, 0, 0, 1, 1, 0, 0, 31};
_gfortran_set_options (9, &options);

6.3.3 _gfortran_set_convert — Set endian conversion

Description:
_gfortran_set_convert set the representation of data for unformatted files.
Syntax : void _gfortran_set_convert (int conv)
Arguments:
conv Endian conversion, possible values:
GFC CONVERT NATIVE (0, default),
GFC CONVERT SWAP (1), GFC CONVERT BIG
(2), GFC CONVERT LITTLE (3).
Example:
int main (int argc, char *argv[])
{
/* Initialize libgfortran. */
_gfortran_set_args (argc, argv);
_gfortran_set_convert (1);
return 0;
}

6.3.4 _gfortran_set_record_marker — Set length of record markers

Description:
_gfortran_set_record_marker sets the length of record markers for unfor-
matted files.
Syntax : void _gfortran_set_record_marker (int val)
Arguments:
val Length of the record marker; valid values are 4 and
8. Default is 4.
Example:
int main (int argc, char *argv[])
{
/* Initialize libgfortran. */
_gfortran_set_args (argc, argv);
_gfortran_set_record_marker (8);
80 The GNU Fortran Compiler

return 0;
}

6.3.5 _gfortran_set_fpe — Enable floating point exception traps

Description:
_gfortran_set_fpe enables floating point exception traps for the specified ex-
ceptions. On most systems, this will result in a SIGFPE signal being sent and
the program being aborted.
Syntax : void _gfortran_set_fpe (int val)
Arguments:
option[0] IEEE exceptions. Possible values are (bitwise or-
ed) zero (0, default) no trapping, GFC_FPE_INVALID
(1), GFC_FPE_DENORMAL (2), GFC_FPE_ZERO (4), GFC_
FPE_OVERFLOW (8), GFC_FPE_UNDERFLOW (16), and
GFC_FPE_INEXACT (32).
Example:
int main (int argc, char *argv[])
{
/* Initialize libgfortran. */
_gfortran_set_args (argc, argv);
/* FPE for invalid operations such as SQRT(-1.0). */
_gfortran_set_fpe (1);
return 0;
}

6.3.6 _gfortran_set_max_subrecord_length — Set subrecord length

Description:
_gfortran_set_max_subrecord_length set the maximum length for a sub-
record. This option only makes sense for testing and debugging of unformatted
I/O.
Syntax : void _gfortran_set_max_subrecord_length (int val)
Arguments:
val the maximum length for a subrecord; the maximum
permitted value is 2147483639, which is also the
default.
Example:
int main (int argc, char *argv[])
{
/* Initialize libgfortran. */
_gfortran_set_args (argc, argv);
_gfortran_set_max_subrecord_length (8);
return 0;
}

6.4 Naming and argument-passing conventions

This section gives an overview about the naming convention of procedures and global vari-
ables and about the argument passing conventions used by GNU Fortran. If a C binding
Chapter 6: Mixed-Language Programming 81

has been specified, the naming convention and some of the argument-passing conventions
change. If possible, mixed-language and mixed-compiler projects should use the better de-
fined C binding for interoperability. See see Section 6.1 [Interoperability with C], page 69.

6.4.1 Naming conventions

According the Fortran standard, valid Fortran names consist of a letter between A to Z, a
to z, digits 0, 1 to 9 and underscores (_) with the restriction that names may only start
with a letter. As vendor extension, the dollar sign ($) is additionally permitted with the
option -fdollar-ok, but not as first character and only if the target system supports it.
By default, the procedure name is the lower-cased Fortran name with an appended
underscore (_); using -fno-underscoring no underscore is appended while -fsecond-
underscore appends two underscores. Depending on the target system and the calling
convention, the procedure might be additionally dressed; for instance, on 32bit Windows
with stdcall, an at-sign @ followed by an integer number is appended. For the changing
the calling convention, see see Section 6.2 [GNU Fortran Compiler Directives], page 74.
For common blocks, the same convention is used, i.e. by default an underscore is ap-
pended to the lower-cased Fortran name. Blank commons have the name __BLNK__.
For procedures and variables declared in the specification space of a module, the name
is formed by __, followed by the lower-cased module name, _MOD_, and the lower-cased
Fortran name. Note that no underscore is appended.

6.4.2 Argument passing conventions

Subroutines do not return a value (matching C99’s void) while functions either return a
value as specified in the platform ABI or the result variable is passed as hidden argument
to the function and no result is returned. A hidden result variable is used when the result
variable is an array or of type CHARACTER.
Arguments are passed according to the platform ABI. In particular, complex arguments
might not be compatible to a struct with two real components for the real and imaginary
part. The argument passing matches the one of C99’s _Complex. Functions with scalar
complex result variables return their value and do not use a by-reference argument. Note
that with the -ff2c option, the argument passing is modified and no longer completely
matches the platform ABI. Some other Fortran compilers use f2c semantic by default; this
might cause problems with interoperablility.
GNU Fortran passes most arguments by reference, i.e. by passing a pointer to the data.
Note that the compiler might use a temporary variable into which the actual argument has
been copied, if required semantically (copy-in/copy-out).
For arguments with ALLOCATABLE and POINTER attribute (including procedure pointers),
a pointer to the pointer is passed such that the pointer address can be modified in the
procedure.
For dummy arguments with the VALUE attribute: Scalar arguments of the type INTEGER,
LOGICAL, REAL and COMPLEX are passed by value according to the platform ABI. (As vendor
extension and not recommended, using %VAL() in the call to a procedure has the same
effect.) For TYPE(C_PTR) and procedure pointers, the pointer itself is passed such that it
can be modified without affecting the caller.
82 The GNU Fortran Compiler

For Boolean (LOGICAL) arguments, please note that GCC expects only the integer value
0 and 1. If a GNU Fortran LOGICAL variable contains another integer value, the result is
undefined. As some other Fortran compilers use −1 for .TRUE., extra care has to be taken –
such as passing the value as INTEGER. (The same value restriction also applies to other front
ends of GCC, e.g. to GCC’s C99 compiler for _Bool or GCC’s Ada compiler for Boolean.)
For arguments of CHARACTER type, the character length is passed as a hidden argument
at the end of the argument list. For deferred-length strings, the value is passed by reference,
otherwise by value. The character length has the C type size_t (or INTEGER(kind=C_SIZE_
T) in Fortran). Note that this is different to older versions of the GNU Fortran compiler,
where the type of the hidden character length argument was a C int. In order to retain
compatibility with older versions, one can e.g. for the following Fortran procedure
subroutine fstrlen (s, a)
character(len=*) :: s
integer :: a
print*, len(s)
end subroutine fstrlen
define the corresponding C prototype as follows:
#if __GNUC__ > 7
typedef size_t fortran_charlen_t;
#else
typedef int fortran_charlen_t;
#endif

void fstrlen_ (char*, int*, fortran_charlen_t);

In order to avoid such compiler-specific details, for new code it is instead recommended
to use the ISO C BINDING feature.
Note with C binding, CHARACTER(len=1) result variables are returned according to the
platform ABI and no hidden length argument is used for dummy arguments; with VALUE,
those variables are passed by value.
For OPTIONAL dummy arguments, an absent argument is denoted by a NULL pointer,
except for scalar dummy arguments of intrinsic type which have the VALUE attribute. For
those, a hidden Boolean argument (logical(kind=C_bool),value) is used to indicate
whether the argument is present.
Arguments which are assumed-shape, assumed-rank or deferred-rank arrays or, with
-fcoarray=lib, allocatable scalar coarrays use an array descriptor. All other arrays pass
the address of the first element of the array. With -fcoarray=lib, the token and the offset
belonging to nonallocatable coarrays dummy arguments are passed as hidden argument
along the character length hidden arguments. The token is an opaque pointer identifying
the coarray and the offset is a passed-by-value integer of kind C_PTRDIFF_T, denoting the
byte offset between the base address of the coarray and the passed scalar or first element
of the passed array.
The arguments are passed in the following order
• Result variable, when the function result is passed by reference
• Character length of the function result, if it is a of type CHARACTER and no C binding
is used
• The arguments in the order in which they appear in the Fortran declaration
Chapter 6: Mixed-Language Programming 83

• The present status for optional arguments with value attribute, which are internally
passed by value
• The character length and/or coarray token and offset for the first argument which is
a CHARACTER or a nonallocatable coarray dummy argument, followed by the hidden
arguments of the next dummy argument of such a type
 85

7 Coarray Programming

7.1 Type and enum ABI Documentation

7.1.1 caf_token_t
Typedef of type void * on the compiler side. Can be any data type on the library side.

7.1.2 caf_register_t
Indicates which kind of coarray variable should be registered.
typedef enum caf_register_t {
CAF_REGTYPE_COARRAY_STATIC,
CAF_REGTYPE_COARRAY_ALLOC,
CAF_REGTYPE_LOCK_STATIC,
CAF_REGTYPE_LOCK_ALLOC,
CAF_REGTYPE_CRITICAL,
CAF_REGTYPE_EVENT_STATIC,
CAF_REGTYPE_EVENT_ALLOC,
CAF_REGTYPE_COARRAY_ALLOC_REGISTER_ONLY,
CAF_REGTYPE_COARRAY_ALLOC_ALLOCATE_ONLY
}
caf_register_t;
The values CAF_REGTYPE_COARRAY_ALLOC_REGISTER_ONLY and CAF_REGTYPE_COARRAY_
ALLOC_ALLOCATE_ONLY are for allocatable components in derived type coarrays only. The
first one sets up the token without allocating memory for allocatable component. The latter
one only allocates the memory for an allocatable component in a derived type coarray.
The token needs to be setup previously by the REGISTER ONLY. This allows to have
allocatable components un-allocated on some images. The status whether an allocatable
component is allocated on a remote image can be queried by _caf_is_present which used
internally by the ALLOCATED intrinsic.

7.1.3 caf_deregister_t
typedef enum caf_deregister_t {
CAF_DEREGTYPE_COARRAY_DEREGISTER,
CAF_DEREGTYPE_COARRAY_DEALLOCATE_ONLY
}
caf_deregister_t;
Allows to specifiy the type of deregistration of a coarray object. The CAF_DEREGTYPE_
COARRAY_DEALLOCATE_ONLY flag is only allowed for allocatable components in derived type
coarrays.

7.1.4 caf_reference_t
The structure used for implementing arbitrary reference chains. A CAF_REFERENCE_T allows
to specify a component reference or any kind of array reference of any rank supported by
86 The GNU Fortran Compiler

gfortran. For array references all kinds as known by the compiler/Fortran standard are
supported indicated by a MODE.
typedef enum caf_ref_type_t {
/* Reference a component of a derived type, either regular one or an
allocatable or pointer type. For regular ones idx in caf_reference_t is
set to -1. */
CAF_REF_COMPONENT,
/* Reference an allocatable array. */
CAF_REF_ARRAY,
/* Reference a non-allocatable/non-pointer array. I.e., the coarray object
has no array descriptor associated and the addressing is done
completely using the ref. */
CAF_REF_STATIC_ARRAY
} caf_ref_type_t;
typedef enum caf_array_ref_t {
/* No array ref. This terminates the array ref. */
CAF_ARR_REF_NONE = 0,
/* Reference array elements given by a vector. Only for this mode
caf_reference_t.u.a.dim[i].v is valid. */
CAF_ARR_REF_VECTOR,
/* A full array ref (:). */
CAF_ARR_REF_FULL,
/* Reference a range on elements given by start, end and stride. */
CAF_ARR_REF_RANGE,
/* Only a single item is referenced given in the start member. */
CAF_ARR_REF_SINGLE,
/* An array ref of the kind (i:), where i is an arbitrary valid index in the
array. The index i is given in the start member. */
CAF_ARR_REF_OPEN_END,
/* An array ref of the kind (:i), where the lower bound of the array ref
is given by the remote side. The index i is given in the end member. */
CAF_ARR_REF_OPEN_START
} caf_array_ref_t;
/* References to remote components of a derived type. */
typedef struct caf_reference_t {
/* A pointer to the next ref or NULL. */
struct caf_reference_t *next;
/* The type of the reference. */
/* caf_ref_type_t, replaced by int to allow specification in fortran FE. */
int type;
/* The size of an item referenced in bytes. I.e. in an array ref this is
the factor to advance the array pointer with to get to the next item.
For component refs this gives just the size of the element referenced. */
size_t item_size;
union {
struct {
Chapter 7: Coarray Programming 87

/* The offset (in bytes) of the component in the derived type.

Unused for allocatable or pointer components. */
ptrdiff_t offset;
/* The offset (in bytes) to the caf_token associated with this
component. NULL, when not allocatable/pointer ref. */
ptrdiff_t caf_token_offset;
} c;
struct {
/* The mode of the array ref. See CAF_ARR_REF_*. */
/* caf_array_ref_t, replaced by unsigend char to allow specification in
fortran FE. */
unsigned char mode[GFC_MAX_DIMENSIONS];
/* The type of a static array. Unset for array's with descriptors. */
int static_array_type;
/* Subscript refs (s) or vector refs (v). */
union {
struct {
/* The start and end boundary of the ref and the stride. */
index_type start, end, stride;
} s;
struct {
/* nvec entries of kind giving the elements to reference. */
void *vector;
/* The number of entries in vector. */
size_t nvec;
/* The integer kind used for the elements in vector. */
int kind;
} v;
} dim[GFC_MAX_DIMENSIONS];
} a;
} u;
} caf_reference_t;
The references make up a single linked list of reference operations. The NEXT member
links to the next reference or NULL to indicate the end of the chain. Component and array
refs can be arbitrarily mixed as long as they comply to the Fortran standard.
NOTES The member STATIC_ARRAY_TYPE is used only when the TYPE is CAF_REF_
STATIC_ARRAY. The member gives the type of the data referenced. Because no array
descriptor is available for a descriptor-less array and type conversion still needs to take
place the type is transported here.
At the moment CAF_ARR_REF_VECTOR is not implemented in the front end for descriptor-
less arrays. The library caf single has untested support for it.

7.1.5 caf_team_t
Opaque pointer to represent a team-handle. This type is a stand-in for the future imple-
mentation of teams. It is about to change without further notice.
88 The GNU Fortran Compiler

7.2 Function ABI Documentation

7.2.1 _gfortran_caf_init — Initialiation function
Description:
This function is called at startup of the program before the Fortran main pro-
gram, if the latter has been compiled with -fcoarray=lib. It takes as argu-
ments the command-line arguments of the program. It is permitted to pass two
NULL pointers as argument; if non-NULL, the library is permitted to modify the
arguments.
Syntax : void _gfortran_caf_init (int *argc, char ***argv)
Arguments:
argc intent(inout) An integer pointer with the number of
arguments passed to the program or NULL.
argv intent(inout) A pointer to an array of strings with
the command-line arguments or NULL.
NOTES The function is modelled after the initialization function of the Message Passing
Interface (MPI) specification. Due to the way coarray registration works, it
might not be the first call to the library. If the main program is not written
in Fortran and only a library uses coarrays, it can happen that this function
is never called. Therefore, it is recommended that the library does not rely on
the passed arguments and whether the call has been done.

7.2.2 _gfortran_caf_finish — Finalization function

Description:
This function is called at the end of the Fortran main program, if it has been
compiled with the -fcoarray=lib option.
Syntax : void _gfortran_caf_finish (void)
NOTES For non-Fortran programs, it is recommended to call the function at the end of
the main program. To ensure that the shutdown is also performed for programs
where this function is not explicitly invoked, for instance non-Fortran programs
or calls to the system’s exit() function, the library can use a destructor function.
Note that programs can also be terminated using the STOP and ERROR STOP
statements; those use different library calls.

7.2.3 _gfortran_caf_this_image — Querying the image number

Description:
This function returns the current image number, which is a positive number.
Syntax : int _gfortran_caf_this_image (int distance)
Arguments:
distance As specified for the this_image intrinsic in TS18508.
Shall be a non-negative number.
NOTES If the Fortran intrinsic this_image is invoked without an argument, which is
the only permitted form in Fortran 2008, GCC passes 0 as first argument.
Chapter 7: Coarray Programming 89

7.2.4 _gfortran_caf_num_images — Querying the maximal number

of images
Description:
This function returns the number of images in the current team, if distance is 0
or the number of images in the parent team at the specified distance. If failed
is -1, the function returns the number of all images at the specified distance;
if it is 0, the function returns the number of nonfailed images, and if it is 1, it
returns the number of failed images.
Syntax : int _gfortran_caf_num_images(int distance, int failed)
Arguments:
distance the distance from this image to the ancestor. Shall
be positive.
failed shall be -1, 0, or 1
NOTES This function follows TS18508. If the num image intrinsic has no arguments,
then the compiler passes distance=0 and failed=-1 to the function.

7.2.5 _gfortran_caf_image_status — Query the status of an image

Description:
Get the status of the image given by the id image of the team given by
team. Valid results are zero, for image is ok, STAT_STOPPED_IMAGE from the
ISO FORTRAN ENV module to indicate that the image has been stopped
and STAT_FAILED_IMAGE also from ISO FORTRAN ENV to indicate that the
image has executed a FAIL IMAGE statement.
Syntax : int _gfortran_caf_image_status (int image, caf_team_t * team)
Arguments:
image the positive scalar id of the image in the current
TEAM.
team optional; team on the which the inquiry is to be
performed.
NOTES This function follows TS18508. Because team-functionality is not yet imple-
mented a null-pointer is passed for the team argument at the moment.
7.2.6 _gfortran_caf_failed_images — Get an array of the indexes
of the failed images
Description:
Get an array of image indexes in the current team that have failed. The array is
sorted ascendingly. When team is not provided the current team is to be used.
When kind is provided then the resulting array is of that integer kind else it
is of default integer kind. The returns an unallocated size zero array when no
images have failed.
Syntax : int _gfortran_caf_failed_images (caf_team_t * team, int * kind)
Arguments:
team optional; team on the which the inquiry is to be
performed.
90 The GNU Fortran Compiler

image optional; the kind of the resulting integer array.

NOTES This function follows TS18508. Because team-functionality is not yet imple-
mented a null-pointer is passed for the team argument at the moment.
7.2.7 _gfortran_caf_stopped_images — Get an array of the indexes
of the stopped images
Description:
Get an array of image indexes in the current team that have stopped. The
array is sorted ascendingly. When team is not provided the current team is to
be used. When kind is provided then the resulting array is of that integer kind
else it is of default integer kind. The returns an unallocated size zero array
when no images have failed.
Syntax : int _gfortran_caf_stopped_images (caf_team_t * team, int * kind)
Arguments:
team optional; team on the which the inquiry is to be
performed.
image optional; the kind of the resulting integer array.
NOTES This function follows TS18508. Because team-functionality is not yet imple-
mented a null-pointer is passed for the team argument at the moment.

7.2.8 _gfortran_caf_register — Registering coarrays

Description:
Registers memory for a coarray and creates a token to identify the coarray.
The routine is called for both coarrays with SAVE attribute and using an ex-
plicit ALLOCATE statement. If an error occurs and STAT is a NULL pointer,
the function shall abort with printing an error message and starting the error
termination. If no error occurs and STAT is present, it shall be set to zero.
Otherwise, it shall be set to a positive value and, if not-NULL, ERRMSG shall
be set to a string describing the failure. The routine shall register the mem-
ory provided in the DATA-component of the array descriptor DESC, when that
component is non-NULL, else it shall allocate sufficient memory and provide a
pointer to it in the DATA-component of DESC. The array descriptor has rank
zero, when a scalar object is to be registered and the array descriptor may be
invalid after the call to _gfortran_caf_register. When an array is to be
allocated the descriptor persists.
For CAF_REGTYPE_COARRAY_STATIC and CAF_REGTYPE_COARRAY_ALLOC, the
passed size is the byte size requested. For CAF_REGTYPE_LOCK_STATIC,
CAF_REGTYPE_LOCK_ALLOC and CAF_REGTYPE_CRITICAL it is the array size or
one for a scalar.
When CAF_REGTYPE_COARRAY_ALLOC_REGISTER_ONLY is used, then only a token
for an allocatable or pointer component is created. The SIZE parameter is not
used then. On the contrary when CAF_REGTYPE_COARRAY_ALLOC_ALLOCATE_
ONLY is specified, then the token needs to be registered by a previous call with
regtype CAF_REGTYPE_COARRAY_ALLOC_REGISTER_ONLY and either the memory
Chapter 7: Coarray Programming 91

specified in the DESC’s data-ptr is registered or allocate when the data-ptr is

NULL.

Syntax : void caf_register (size_t size, caf_register_t type, caf_token_t

*token, gfc_descriptor_t *desc, int *stat, char *errmsg, size_t
errmsg_len)

Arguments:
size For normal coarrays, the byte size of the coarray to be
allocated; for lock types and event types, the number
of elements.
type one of the caf register t types.
token intent(out) An opaque pointer identifying the
coarray.
desc intent(inout) The (pseudo) array descriptor.
stat intent(out) For allocatable coarrays, stores the
STAT=; may be NULL
errmsg intent(out) When an error occurs, this will be set to
an error message; may be NULL
errmsg len the buffer size of errmsg.

NOTES Nonallocatable coarrays have to be registered prior use from remote images. In
order to guarantee this, they have to be registered before the main program.
This can be achieved by creating constructor functions. That is what GCC does
such that also for nonallocatable coarrays the memory is allocated and no static
memory is used. The token permits to identify the coarray; to the processor,
the token is a nonaliasing pointer. The library can, for instance, store the base
address of the coarray in the token, some handle or a more complicated struct.
The library may also store the array descriptor DESC when its rank is non-zero.
For lock types, the value shall only be used for checking the allocation status.
Note that for critical blocks, the locking is only required on one image; in
the locking statement, the processor shall always pass an image index of one
for critical-block lock variables (CAF_REGTYPE_CRITICAL). For lock types and
critical-block variables, the initial value shall be unlocked (or, respectively, not
in critical section) such as the value false; for event types, the initial state should
be no event, e.g. zero.

7.2.9 _gfortran_caf_deregister — Deregistering coarrays

Description:
Called to free or deregister the memory of a coarray; the processor calls this
function for automatic and explicit deallocation. In case of an error, this func-
tion shall fail with an error message, unless the STAT variable is not null.
The library is only expected to free memory it allocated itself during a call to
_gfortran_caf_register.

Syntax : void caf_deregister (caf_token_t *token, caf_deregister_t type, int

*stat, char *errmsg, size_t errmsg_len)
92 The GNU Fortran Compiler

Arguments:
token the token to free.
type the type of action to take for the coarray. A CAF_
DEREGTYPE_COARRAY_DEALLOCATE_ONLY is allowed
only for allocatable or pointer components of derived
type coarrays. The action only deallocates the local
memory without deleting the token.
stat intent(out) Stores the STAT=; may be NULL
errmsg intent(out) When an error occurs, this will be set to
an error message; may be NULL
errmsg len the buffer size of errmsg.
NOTES For nonalloatable coarrays this function is never called. If a cleanup is required,
it has to be handled via the finish, stop and error stop functions, and via
destructors.
7.2.10 _gfortran_caf_is_present — Query whether an allocatable
or pointer component in a derived type coarray is allocated
Description:
Used to query the coarray library whether an allocatable component in a derived
type coarray is allocated on a remote image.
Syntax : void _gfortran_caf_is_present (caf_token_t token, int image_index,
gfc_reference_t *ref)
Arguments:
token An opaque pointer identifying the coarray.
image index The ID of the remote image; must be a positive
number.
ref A chain of references to address the allocatable or
pointer component in the derived type coarray. The
object reference needs to be a scalar or a full array
reference, respectively.

7.2.11 _gfortran_caf_send — Sending data from a local image to a

remote image
Description:
Called to send a scalar, an array section or a whole array from a local to a
remote image identified by the image index.
Syntax : void _gfortran_caf_send (caf_token_t token, size_t offset,
int image_index, gfc_descriptor_t *dest, caf_vector_t *dst_
vector, gfc_descriptor_t *src, int dst_kind, int src_kind, bool
may_require_tmp, int *stat)
Arguments:
token intent(in) An opaque pointer identifying the coarray.
offset intent(in) By which amount of bytes the actual data
is shifted compared to the base address of the coarray.
Chapter 7: Coarray Programming 93

image index intent(in) The ID of the remote image; must be a

positive number.
dest intent(in) Array descriptor for the remote image for
the bounds and the size. The base_addr shall not be
accessed.
dst vector intent(in) If not NULL, it contains the vector sub-
script of the destination array; the values are relative
to the dimension triplet of the dest argument.
src intent(in) Array descriptor of the local array to be
transferred to the remote image
dst kind intent(in) Kind of the destination argument
src kind intent(in) Kind of the source argument
may require tmp
intent(in) The variable is false when it is known at
compile time that the dest and src either cannot over-
lap or overlap (fully or partially) such that walking
src and dest in element wise element order (honor-
ing the stride value) will not lead to wrong results.
Otherwise, the value is true.
stat intent(out) when non-NULL give the result of the
operation, i.e., zero on success and non-zero on er-
ror. When NULL and an error occurs, then an error
message is printed and the program is terminated.
NOTES It is permitted to have image index equal the current image; the memory of
the send-to and the send-from might (partially) overlap in that case. The
implementation has to take care that it handles this case, e.g. using memmove
which handles (partially) overlapping memory. If may require tmp is true, the
library might additionally create a temporary variable, unless additional checks
show that this is not required (e.g. because walking backward is possible or
because both arrays are contiguous and memmove takes care of overlap issues).
Note that the assignment of a scalar to an array is permitted. In addition,
the library has to handle numeric-type conversion and for strings, padding and
different character kinds.

7.2.12 _gfortran_caf_get — Getting data from a remote image

Description:
Called to get an array section or a whole array from a remote, image identified
by the image index.
Syntax : void _gfortran_caf_get (caf_token_t token, size_t offset, int
image_index, gfc_descriptor_t *src, caf_vector_t *src_vector,
gfc_descriptor_t *dest, int src_kind, int dst_kind, bool may_
require_tmp, int *stat)
Arguments:
token intent(in) An opaque pointer identifying the coarray.
offset intent(in) By which amount of bytes the actual data
is shifted compared to the base address of the coarray.
94 The GNU Fortran Compiler

image index intent(in) The ID of the remote image; must be a

positive number.
dest intent(out) Array descriptor of the local array to
store the data retrieved from the remote image
src intent(in) Array descriptor for the remote image for
the bounds and the size. The base_addr shall not be
accessed.
src vector intent(in) If not NULL, it contains the vector sub-
script of the source array; the values are relative to
the dimension triplet of the src argument.
dst kind intent(in) Kind of the destination argument
src kind intent(in) Kind of the source argument
may require tmp
intent(in) The variable is false when it is known at
compile time that the dest and src either cannot over-
lap or overlap (fully or partially) such that walking
src and dest in element wise element order (honor-
ing the stride value) will not lead to wrong results.
Otherwise, the value is true.
stat intent(out) When non-NULL give the result of the
operation, i.e., zero on success and non-zero on er-
ror. When NULL and an error occurs, then an error
message is printed and the program is terminated.

NOTES It is permitted to have image index equal the current image; the memory of
the send-to and the send-from might (partially) overlap in that case. The
implementation has to take care that it handles this case, e.g. using memmove
which handles (partially) overlapping memory. If may require tmp is true, the
library might additionally create a temporary variable, unless additional checks
show that this is not required (e.g. because walking backward is possible or
because both arrays are contiguous and memmove takes care of overlap issues).
Note that the library has to handle numeric-type conversion and for strings,
padding and different character kinds.

7.2.13 _gfortran_caf_sendget — Sending data between remote

images
Description:
Called to send a scalar, an array section or a whole array from a remote
image identified by the src image index to a remote image identified by the
dst image index.

Syntax : void _gfortran_caf_sendget (caf_token_t dst_token, size_t dst_

offset, int dst_image_index, gfc_descriptor_t *dest, caf_vector_t
*dst_vector, caf_token_t src_token, size_t src_offset, int src_
image_index, gfc_descriptor_t *src, caf_vector_t *src_vector, int
dst_kind, int src_kind, bool may_require_tmp, int *stat)
Chapter 7: Coarray Programming 95

Arguments:
dst token intent(in) An opaque pointer identifying the destina-
tion coarray.
dst offset intent(in) By which amount of bytes the actual data
is shifted compared to the base address of the desti-
nation coarray.
dst image index
intent(in) The ID of the destination remote image;
must be a positive number.
dest intent(in) Array descriptor for the destination remote
image for the bounds and the size. The base_addr
shall not be accessed.
dst vector intent(int) If not NULL, it contains the vector sub-
script of the destination array; the values are relative
to the dimension triplet of the dest argument.
src token intent(in) An opaque pointer identifying the source
coarray.
src offset intent(in) By which amount of bytes the actual data
is shifted compared to the base address of the source
coarray.
src image index
intent(in) The ID of the source remote image; must
be a positive number.
src intent(in) Array descriptor of the local array to be
transferred to the remote image.
src vector intent(in) Array descriptor of the local array to be
transferred to the remote image
dst kind intent(in) Kind of the destination argument
src kind intent(in) Kind of the source argument
may require tmp
intent(in) The variable is false when it is known at
compile time that the dest and src either cannot over-
lap or overlap (fully or partially) such that walking
src and dest in element wise element order (honor-
ing the stride value) will not lead to wrong results.
Otherwise, the value is true.
stat intent(out) when non-NULL give the result of the
operation, i.e., zero on success and non-zero on er-
ror. When NULL and an error occurs, then an error
message is printed and the program is terminated.

NOTES It is permitted to have the same image index for both src image index and
dst image index; the memory of the send-to and the send-from might (par-
tially) overlap in that case. The implementation has to take care that it handles
this case, e.g. using memmove which handles (partially) overlapping memory. If
may require tmp is true, the library might additionally create a temporary
variable, unless additional checks show that this is not required (e.g. because
walking backward is possible or because both arrays are contiguous and memmove
takes care of overlap issues).
96 The GNU Fortran Compiler

Note that the assignment of a scalar to an array is permitted. In addition,

the library has to handle numeric-type conversion and for strings, padding and
different character kinds.
7.2.14 _gfortran_caf_send_by_ref — Sending data from a local
image to a remote image with enhanced referencing options
Description:
Called to send a scalar, an array section or a whole array from a local to a
remote image identified by the image index.
Syntax : void _gfortran_caf_send_by_ref (caf_token_t token, int image_index,
gfc_descriptor_t *src, caf_reference_t *refs, int dst_kind, int
src_kind, bool may_require_tmp, bool dst_reallocatable, int *stat,
int dst_type)
Arguments:
token intent(in) An opaque pointer identifying the coarray.
image index intent(in) The ID of the remote image; must be a
positive number.
src intent(in) Array descriptor of the local array to be
transferred to the remote image
refs intent(in) The references on the remote array to store
the data given by src. Guaranteed to have at least
one entry.
dst kind intent(in) Kind of the destination argument
src kind intent(in) Kind of the source argument
may require tmp
intent(in) The variable is false when it is known at
compile time that the dest and src either cannot over-
lap or overlap (fully or partially) such that walking
src and dest in element wise element order (honor-
ing the stride value) will not lead to wrong results.
Otherwise, the value is true.
dst reallocatable
intent(in) Set when the destination is of allocatable
or pointer type and the refs will allow reallocation,
i.e., the ref is a full array or component ref.
stat intent(out) When non-NULL give the result of the op-
eration, i.e., zero on success and non-zero on error.
When NULL and an error occurs, then an error mes-
sage is printed and the program is terminated.
dst type intent(in) Give the type of the destination. When
the destination is not an array, than the precise type,
e.g. of a component in a derived type, is not known,
but provided here.
NOTES It is permitted to have image index equal the current image; the memory of
the send-to and the send-from might (partially) overlap in that case. The
implementation has to take care that it handles this case, e.g. using memmove
which handles (partially) overlapping memory. If may require tmp is true, the
Chapter 7: Coarray Programming 97

library might additionally create a temporary variable, unless additional checks

show that this is not required (e.g. because walking backward is possible or
because both arrays are contiguous and memmove takes care of overlap issues).
Note that the assignment of a scalar to an array is permitted. In addition,
the library has to handle numeric-type conversion and for strings, padding and
different character kinds.
Because of the more complicated references possible some operations may be
unsupported by certain libraries. The library is expected to issue a precise error
message why the operation is not permitted.

7.2.15 _gfortran_caf_get_by_ref — Getting data from a remote

image using enhanced references
Description:
Called to get a scalar, an array section or a whole array from a remote image
identified by the image index.
Syntax : void _gfortran_caf_get_by_ref (caf_token_t token, int image_index,
caf_reference_t *refs, gfc_descriptor_t *dst, int dst_kind, int
src_kind, bool may_require_tmp, bool dst_reallocatable, int *stat,
int src_type)
Arguments:
token intent(in) An opaque pointer identifying the coarray.
image index intent(in) The ID of the remote image; must be a
positive number.
refs intent(in) The references to apply to the remote
structure to get the data.
dst intent(in) Array descriptor of the local array to
store the data transferred from the remote im-
age. May be reallocated where needed and when
DST REALLOCATABLE allows it.
dst kind intent(in) Kind of the destination argument
src kind intent(in) Kind of the source argument
may require tmp
intent(in) The variable is false when it is known at
compile time that the dest and src either cannot over-
lap or overlap (fully or partially) such that walking
src and dest in element wise element order (honor-
ing the stride value) will not lead to wrong results.
Otherwise, the value is true.
dst reallocatable
intent(in) Set when DST is of allocatable or pointer
type and its refs allow reallocation, i.e., the full array
or a component is referenced.
stat intent(out) When non-NULL give the result of the op-
eration, i.e., zero on success and non-zero on error.
When NULL and an error occurs, then an error mes-
sage is printed and the program is terminated.
98 The GNU Fortran Compiler

src type intent(in) Give the type of the source. When the
source is not an array, than the precise type, e.g. of
a component in a derived type, is not known, but
provided here.
NOTES It is permitted to have image_index equal the current image; the memory
of the send-to and the send-from might (partially) overlap in that case. The
implementation has to take care that it handles this case, e.g. using memmove
which handles (partially) overlapping memory. If may require tmp is true, the
library might additionally create a temporary variable, unless additional checks
show that this is not required (e.g. because walking backward is possible or
because both arrays are contiguous and memmove takes care of overlap issues).
Note that the library has to handle numeric-type conversion and for strings,
padding and different character kinds.
Because of the more complicated references possible some operations may be
unsupported by certain libraries. The library is expected to issue a precise error
message why the operation is not permitted.

7.2.16 _gfortran_caf_sendget_by_ref — Sending data between

remote images using enhanced references on both sides
Description:
Called to send a scalar, an array section or a whole array from a remote
image identified by the src image index to a remote image identified by the
dst image index.
Syntax : void _gfortran_caf_sendget_by_ref (caf_token_t dst_token, int
dst_image_index, caf_reference_t *dst_refs, caf_token_t src_token,
int src_image_index, caf_reference_t *src_refs, int dst_kind, int
src_kind, bool may_require_tmp, int *dst_stat, int *src_stat, int
dst_type, int src_type)
Arguments:
dst token intent(in) An opaque pointer identifying the destina-
tion coarray.
dst image index
intent(in) The ID of the destination remote image;
must be a positive number.
dst refs intent(in) The references on the remote array to store
the data given by the source. Guaranteed to have at
least one entry.
src token intent(in) An opaque pointer identifying the source
coarray.
src image index
intent(in) The ID of the source remote image; must
be a positive number.
src refs intent(in) The references to apply to the remote
structure to get the data.
dst kind intent(in) Kind of the destination argument
src kind intent(in) Kind of the source argument
Chapter 7: Coarray Programming 99

may require tmp

intent(in) The variable is false when it is known at
compile time that the dest and src either cannot over-
lap or overlap (fully or partially) such that walking
src and dest in element wise element order (honor-
ing the stride value) will not lead to wrong results.
Otherwise, the value is true.
dst stat intent(out) when non-NULL give the result of the send-
operation, i.e., zero on success and non-zero on error.
When NULL and an error occurs, then an error mes-
sage is printed and the program is terminated.
src stat intent(out) When non-NULL give the result of the get-
operation, i.e., zero on success and non-zero on error.
When NULL and an error occurs, then an error mes-
sage is printed and the program is terminated.
dst type intent(in) Give the type of the destination. When
the destination is not an array, than the precise type,
e.g. of a component in a derived type, is not known,
but provided here.
src type intent(in) Give the type of the source. When the
source is not an array, than the precise type, e.g. of
a component in a derived type, is not known, but
provided here.
NOTES It is permitted to have the same image index for both src image index and
dst image index; the memory of the send-to and the send-from might (par-
tially) overlap in that case. The implementation has to take care that it handles
this case, e.g. using memmove which handles (partially) overlapping memory. If
may require tmp is true, the library might additionally create a temporary
variable, unless additional checks show that this is not required (e.g. because
walking backward is possible or because both arrays are contiguous and memmove
takes care of overlap issues).
Note that the assignment of a scalar to an array is permitted. In addition,
the library has to handle numeric-type conversion and for strings, padding and
different character kinds.
Because of the more complicated references possible some operations may be
unsupported by certain libraries. The library is expected to issue a precise error
message why the operation is not permitted.

7.2.17 _gfortran_caf_lock — Locking a lock variable

Description:
Acquire a lock on the given image on a scalar locking variable or for the given
array element for an array-valued variable. If the acquired lock is NULL, the
function returns after having obtained the lock. If it is non-NULL, then ac-
quired lock is assigned the value true (one) when the lock could be obtained
and false (zero) otherwise. Locking a lock variable which has already been
locked by the same image is an error.
100 The GNU Fortran Compiler

Syntax : void _gfortran_caf_lock (caf_token_t token, size_t index, int

image_index, int *acquired_lock, int *stat, char *errmsg, size_t
errmsg_len)
Arguments:
token intent(in) An opaque pointer identifying the coarray.
index intent(in) Array index; first array index is 0. For
scalars, it is always 0.
image index intent(in) The ID of the remote image; must be a
positive number.
acquired lock intent(out) If not NULL, it returns whether lock
could be obtained.
stat intent(out) Stores the STAT=; may be NULL.
errmsg intent(out) When an error occurs, this will be set to
an error message; may be NULL.
errmsg len intent(in) the buffer size of errmsg
NOTES This function is also called for critical blocks; for those, the array index is always
zero and the image index is one. Libraries are permitted to use other images
for critical-block locking variables.

7.2.18 _gfortran_caf_lock — Unlocking a lock variable

Description:
Release a lock on the given image on a scalar locking variable or for the given
array element for an array-valued variable. Unlocking a lock variable which is
unlocked or has been locked by a different image is an error.
Syntax : void _gfortran_caf_unlock (caf_token_t token, size_t index, int
image_index, int *stat, char *errmsg, size_t errmsg_len)
Arguments:
token intent(in) An opaque pointer identifying the coarray.
index intent(in) Array index; first array index is 0. For
scalars, it is always 0.
image index intent(in) The ID of the remote image; must be a
positive number.
stat intent(out) For allocatable coarrays, stores the
STAT=; may be NULL.
errmsg intent(out) When an error occurs, this will be set to
an error message; may be NULL.
errmsg len intent(in) the buffer size of errmsg
NOTES This function is also called for critical block; for those, the array index is always
zero and the image index is one. Libraries are permitted to use other images
for critical-block locking variables.

7.2.19 _gfortran_caf_event_post — Post an event

Description:
Increment the event count of the specified event variable.
Chapter 7: Coarray Programming 101

Syntax : void _gfortran_caf_event_post (caf_token_t token, size_t index, int

image_index, int *stat, char *errmsg, size_t errmsg_len)

Arguments:
token intent(in) An opaque pointer identifying the coarray.
index intent(in) Array index; first array index is 0. For
scalars, it is always 0.
image index intent(in) The ID of the remote image; must be a
positive number; zero indicates the current image,
when accessed noncoindexed.
stat intent(out) Stores the STAT=; may be NULL.
errmsg intent(out) When an error occurs, this will be set to
an error message; may be NULL.
errmsg len intent(in) the buffer size of errmsg

NOTES This acts like an atomic add of one to the remote image’s event variable. The
statement is an image-control statement but does not imply sync memory. Still,
all preceeding push communications of this image to the specified remote image
have to be completed before event_wait on the remote image returns.

7.2.20 _gfortran_caf_event_wait — Wait that an event occurred

Description:
Wait until the event count has reached at least the specified until count; if so,
atomically decrement the event variable by this amount and return.

Syntax : void _gfortran_caf_event_wait (caf_token_t token, size_t index, int

until_count, int *stat, char *errmsg, size_t errmsg_len)

Arguments:
token intent(in) An opaque pointer identifying the coarray.
index intent(in) Array index; first array index is 0. For
scalars, it is always 0.
until count intent(in) The number of events which have to be
available before the function returns.
stat intent(out) Stores the STAT=; may be NULL.
errmsg intent(out) When an error occurs, this will be set to
an error message; may be NULL.
errmsg len intent(in) the buffer size of errmsg

NOTES This function only operates on a local coarray. It acts like a loop checking
atomically the value of the event variable, breaking if the value is greater or
equal the requested number of counts. Before the function returns, the event
variable has to be decremented by the requested until count value. A possible
implementation would be a busy loop for a certain number of spins (possibly
depending on the number of threads relative to the number of available cores)
followed by another waiting strategy such as a sleeping wait (possibly with an
increasing number of sleep time) or, if possible, a futex wait.
102 The GNU Fortran Compiler

The statement is an image-control statement but does not imply sync memory.
Still, all preceeding push communications of this image to the specified remote
image have to be completed before event_wait on the remote image returns.

7.2.21 _gfortran_caf_event_query — Query event count

Description:
Return the event count of the specified event variable.
Syntax : void _gfortran_caf_event_query (caf_token_t token, size_t index,
int image_index, int *count, int *stat)
Arguments:
token intent(in) An opaque pointer identifying the coarray.
index intent(in) Array index; first array index is 0. For
scalars, it is always 0.
image index intent(in) The ID of the remote image; must be a pos-
itive number; zero indicates the current image when
accessed noncoindexed.
count intent(out) The number of events currently posted to
the event variable.
stat intent(out) Stores the STAT=; may be NULL.
NOTES The typical use is to check the local event variable to only call event_wait
when the data is available. However, a coindexed variable is permitted; there
is no ordering or synchronization implied. It acts like an atomic fetch of the
value of the event variable.

7.2.22 _gfortran_caf_sync_all — All-image barrier

Description:
Synchronization of all images in the current team; the program only continues
on a given image after this function has been called on all images of the cur-
rent team. Additionally, it ensures that all pending data transfers of previous
segment have completed.
Syntax : void _gfortran_caf_sync_all (int *stat, char *errmsg, size_t
errmsg_len)
Arguments:
stat intent(out) Stores the status STAT= and may be
NULL.
errmsg intent(out) When an error occurs, this will be set to
an error message; may be NULL.
errmsg len intent(in) the buffer size of errmsg

7.2.23 _gfortran_caf_sync_images — Barrier for selected images

Description:
Synchronization between the specified images; the program only continues on
a given image after this function has been called on all images specified for
Chapter 7: Coarray Programming 103

that image. Note that one image can wait for all other images in the current
team (e.g. via sync images(*)) while those only wait for that specific image.
Additionally, sync images ensures that all pending data transfers of previous
segments have completed.
Syntax : void _gfortran_caf_sync_images (int count, int images[], int *stat,
char *errmsg, size_t errmsg_len)
Arguments:
count intent(in) The number of images which are provided
in the next argument. For a zero-sized array, the
value is zero. For sync images (*), the value is −1.
images intent(in) An array with the images provided by the
user. If count is zero, a NULL pointer is passed.
stat intent(out) Stores the status STAT= and may be
NULL.
errmsg intent(out) When an error occurs, this will be set to
an error message; may be NULL.
errmsg len intent(in) the buffer size of errmsg

7.2.24 _gfortran_caf_sync_memory — Wait for completion of

segment-memory operations
Description:
Acts as optimization barrier between different segments. It also ensures that
all pending memory operations of this image have been completed.
Syntax : void _gfortran_caf_sync_memory (int *stat, char *errmsg, size_t
errmsg_len)
Arguments:
stat intent(out) Stores the status STAT= and may be
NULL.
errmsg intent(out) When an error occurs, this will be set to
an error message; may be NULL.
errmsg len intent(in) the buffer size of errmsg
NOTE A simple implementation could be
__asm__ __volatile__ ("":::"memory") to prevent code movements.

7.2.25 _gfortran_caf_error_stop — Error termination with exit

code
Description:
Invoked for an ERROR STOP statement which has an integer argument. The
function should terminate the program with the specified exit code.
Syntax : void _gfortran_caf_error_stop (int error)
Arguments:
error intent(in) The exit status to be used.
104 The GNU Fortran Compiler

7.2.26 _gfortran_caf_error_stop_str — Error termination with

string
Description:
Invoked for an ERROR STOP statement which has a string as argument. The
function should terminate the program with a nonzero-exit code.
Syntax : void _gfortran_caf_error_stop (const char *string, size_t len)
Arguments:
string intent(in) the error message (not zero terminated)
len intent(in) the length of the string

7.2.27 _gfortran_caf_fail_image — Mark the image failed and end

its execution
Description:
Invoked for an FAIL IMAGE statement. The function should terminate the cur-
rent image.
Syntax : void _gfortran_caf_fail_image ()
NOTES This function follows TS18508.

7.2.28 _gfortran_caf_atomic_define — Atomic variable assignment

Description:
Assign atomically a value to an integer or logical variable.
Syntax : void _gfortran_caf_atomic_define (caf_token_t token, size_t offset,
int image_index, void *value, int *stat, int type, int kind)
Arguments:
token intent(in) An opaque pointer identifying the coarray.
offset intent(in) By which amount of bytes the actual data
is shifted compared to the base address of the coarray.
image index intent(in) The ID of the remote image; must be a pos-
itive number; zero indicates the current image when
used noncoindexed.
value intent(in) the value to be assigned, passed by
reference
stat intent(out) Stores the status STAT= and may be
NULL.
type intent(in) The data type, i.e. BT_INTEGER (1) or BT_
LOGICAL (2).
kind intent(in) The kind value (only 4; always int)

7.2.29 _gfortran_caf_atomic_ref — Atomic variable reference

Description:
Reference atomically a value of a kind-4 integer or logical variable.
Syntax : void _gfortran_caf_atomic_ref (caf_token_t token, size_t offset,
int image_index, void *value, int *stat, int type, int kind)
Chapter 7: Coarray Programming 105

Arguments:
token intent(in) An opaque pointer identifying the coarray.
offset intent(in) By which amount of bytes the actual data
is shifted compared to the base address of the coarray.
image index intent(in) The ID of the remote image; must be a pos-
itive number; zero indicates the current image when
used noncoindexed.
value intent(out) The variable assigned the atomically ref-
erenced variable.
stat intent(out) Stores the status STAT= and may be
NULL.
type the data type, i.e. BT_INTEGER (1) or BT_LOGICAL
(2).
kind The kind value (only 4; always int)

7.2.30 _gfortran_caf_atomic_cas — Atomic compare and swap

Description:
Atomic compare and swap of a kind-4 integer or logical variable. Assigns atom-
ically the specified value to the atomic variable, if the latter has the value
specified by the passed condition value.

Syntax : void _gfortran_caf_atomic_cas (caf_token_t token, size_t offset,

int image_index, void *old, void *compare, void *new_val, int *stat,
int type, int kind)

Arguments:
token intent(in) An opaque pointer identifying the coarray.
offset intent(in) By which amount of bytes the actual data
is shifted compared to the base address of the coarray.
image index intent(in) The ID of the remote image; must be a pos-
itive number; zero indicates the current image when
used noncoindexed.
old intent(out) The value which the atomic variable had
just before the cas operation.
compare intent(in) The value used for comparision.
new val intent(in) The new value for the atomic variable, as-
signed to the atomic variable, if compare equals the
value of the atomic variable.
stat intent(out) Stores the status STAT= and may be
NULL.
type intent(in) the data type, i.e. BT_INTEGER (1) or BT_
LOGICAL (2).
kind intent(in) The kind value (only 4; always int)
106 The GNU Fortran Compiler

7.2.31 _gfortran_caf_atomic_op — Atomic operation

Description:
Apply an operation atomically to an atomic integer or logical variable. After the
operation, old contains the value just before the operation, which, respectively,
adds (GFC CAF ATOMIC ADD) atomically the value to the atomic integer
variable or does a bitwise AND, OR or exclusive OR between the atomic variable
and value; the result is then stored in the atomic variable.
Syntax : void _gfortran_caf_atomic_op (int op, caf_token_t token, size_t
offset, int image_index, void *value, void *old, int *stat, int type,
int kind)
Arguments:
op intent(in) the operation to be performed; possible
values GFC_CAF_ATOMIC_ADD (1), GFC_CAF_ATOMIC_
AND (2), GFC_CAF_ATOMIC_OR (3), GFC_CAF_ATOMIC_
XOR (4).
token intent(in) An opaque pointer identifying the coarray.
offset intent(in) By which amount of bytes the actual data
is shifted compared to the base address of the coarray.
image index intent(in) The ID of the remote image; must be a pos-
itive number; zero indicates the current image when
used noncoindexed.
old intent(out) The value which the atomic variable had
just before the atomic operation.
val intent(in) The new value for the atomic variable, as-
signed to the atomic variable, if compare equals the
value of the atomic variable.
stat intent(out) Stores the status STAT= and may be
NULL.
type intent(in) the data type, i.e. BT_INTEGER (1) or BT_
LOGICAL (2)
kind intent(in) the kind value (only 4; always int)

7.2.32 _gfortran_caf_co_broadcast — Sending data to all images

Description:
Distribute a value from a given image to all other images in the team. Has to
be called collectively.
Syntax : void _gfortran_caf_co_broadcast (gfc_descriptor_t *a, int
source_image, int *stat, char *errmsg, size_t errmsg_len)
Arguments:
a intent(inout) An array descriptor with the data to
be broadcasted (on source image) or to be received
(other images).
source image intent(in) The ID of the image from which the data
should be broadcasted.
Chapter 7: Coarray Programming 107

stat intent(out) Stores the status STAT= and may be

NULL.
errmsg intent(out) When an error occurs, this will be set to
an error message; may be NULL.
errmsg len intent(in) the buffer size of errmsg.

7.2.33 _gfortran_caf_co_max — Collective maximum reduction

Description:
Calculates for each array element of the variable a the maximum value for that
element in the current team; if result image has the value 0, the result shall
be stored on all images, otherwise, only on the specified image. This function
operates on numeric values and character strings.
Syntax : void _gfortran_caf_co_max (gfc_descriptor_t *a, int result_image,
int *stat, char *errmsg, int a_len, size_t errmsg_len)
Arguments:
a intent(inout) An array descriptor for the data to be
processed. On the destination image(s) the result
overwrites the old content.
result image intent(in) The ID of the image to which the reduced
value should be copied to; if zero, it has to be copied
to all images.
stat intent(out) Stores the status STAT= and may be
NULL.
errmsg intent(out) When an error occurs, this will be set to
an error message; may be NULL.
a len intent(in) the string length of argument a
errmsg len intent(in) the buffer size of errmsg
NOTES If result image is nonzero, the data in the array descriptor a on all images
except of the specified one become undefined; hence, the library may make use
of this.

7.2.34 _gfortran_caf_co_min — Collective minimum reduction

Description:
Calculates for each array element of the variable a the minimum value for that
element in the current team; if result image has the value 0, the result shall
be stored on all images, otherwise, only on the specified image. This function
operates on numeric values and character strings.
Syntax : void _gfortran_caf_co_min (gfc_descriptor_t *a, int result_image,
int *stat, char *errmsg, int a_len, size_t errmsg_len)
Arguments:
a intent(inout) An array descriptor for the data to be
processed. On the destination image(s) the result
overwrites the old content.
108 The GNU Fortran Compiler

result image intent(in) The ID of the image to which the reduced

value should be copied to; if zero, it has to be copied
to all images.
stat intent(out) Stores the status STAT= and may be
NULL.
errmsg intent(out) When an error occurs, this will be set to
an error message; may be NULL.
a len intent(in) the string length of argument a
errmsg len intent(in) the buffer size of errmsg
NOTES If result image is nonzero, the data in the array descriptor a on all images
except of the specified one become undefined; hence, the library may make use
of this.

7.2.35 _gfortran_caf_co_sum — Collective summing reduction

Description:
Calculates for each array element of the variable a the sum of all values for that
element in the current team; if result image has the value 0, the result shall
be stored on all images, otherwise, only on the specified image. This function
operates on numeric values only.
Syntax : void _gfortran_caf_co_sum (gfc_descriptor_t *a, int result_image,
int *stat, char *errmsg, size_t errmsg_len)
Arguments:
a intent(inout) An array descriptor with the data to
be processed. On the destination image(s) the result
overwrites the old content.
result image intent(in) The ID of the image to which the reduced
value should be copied to; if zero, it has to be copied
to all images.
stat intent(out) Stores the status STAT= and may be
NULL.
errmsg intent(out) When an error occurs, this will be set to
an error message; may be NULL.
errmsg len intent(in) the buffer size of errmsg
NOTES If result image is nonzero, the data in the array descriptor a on all images
except of the specified one become undefined; hence, the library may make use
of this.

7.2.36 _gfortran_caf_co_reduce — Generic collective reduction

Description:
Calculates for each array element of the variable a the reduction value for that
element in the current team; if result image has the value 0, the result shall be
stored on all images, otherwise, only on the specified image. The opr is a pure
function doing a mathematically commutative and associative operation.
Chapter 7: Coarray Programming 109

The opr flags denote the following; the values are bitwise ored. GFC_CAF_
BYREF (1) if the result should be returned by reference; GFC_CAF_HIDDENLEN
(2) whether the result and argument string lengths shall be specified as hidden
arguments; GFC_CAF_ARG_VALUE (4) whether the arguments shall be passed
by value, GFC_CAF_ARG_DESC (8) whether the arguments shall be passed by
descriptor.
Syntax : void _gfortran_caf_co_reduce (gfc_descriptor_t *a, void * (*opr)
(void *, void *), int opr_flags, int result_image, int *stat, char
*errmsg, int a_len, size_t errmsg_len)
Arguments:
a intent(inout) An array descriptor with the data to
be processed. On the destination image(s) the result
overwrites the old content.
opr intent(in) Function pointer to the reduction function
opr flags intent(in) Flags regarding the reduction function
result image intent(in) The ID of the image to which the reduced
value should be copied to; if zero, it has to be copied
to all images.
stat intent(out) Stores the status STAT= and may be
NULL.
errmsg intent(out) When an error occurs, this will be set to
an error message; may be NULL.
a len intent(in) the string length of argument a
errmsg len intent(in) the buffer size of errmsg
NOTES If result image is nonzero, the data in the array descriptor a on all images
except of the specified one become undefined; hence, the library may make use
of this.
For character arguments, the result is passed as first argument, followed by
the result string length, next come the two string arguments, followed by the
two hidden string length arguments. With C binding, there are no hidden
arguments and by-reference passing and either only a single character is passed
or an array descriptor.
 111

8 Intrinsic Procedures

8.1 Introduction to intrinsic procedures

The intrinsic procedures provided by GNU Fortran include procedures required by the
Fortran 95 and later supported standards, and a set of intrinsic procedures for backwards
compatibility with G77. Any conflict between a description here and a description in the
Fortran standards is unintentional, and the standard(s) should be considered authoritative.
The enumeration of the KIND type parameter is processor defined in the Fortran
95 standard. GNU Fortran defines the default integer type and default real type by
INTEGER(KIND=4) and REAL(KIND=4), respectively. The standard mandates that both data
types shall have another kind, which have more precision. On typical target architectures
supported by gfortran, this kind type parameter is KIND=8. Hence, REAL(KIND=8) and
DOUBLE PRECISION are equivalent. In the description of generic intrinsic procedures,
the kind type parameter will be specified by KIND=*, and in the description of specific
names for an intrinsic procedure the kind type parameter will be explicitly given (e.g.,
REAL(KIND=4) or REAL(KIND=8)). Finally, for brevity the optional KIND= syntax will be
omitted.
Many of the intrinsic procedures take one or more optional arguments. This document
follows the convention used in the Fortran 95 standard, and denotes such arguments by
square brackets.
GNU Fortran offers the -std= command-line option, which can be used to restrict the
set of intrinsic procedures to a given standard. By default, gfortran sets the -std=gnu
option, and so all intrinsic procedures described here are accepted. There is one caveat. For
a select group of intrinsic procedures, g77 implemented both a function and a subroutine.
Both classes have been implemented in gfortran for backwards compatibility with g77. It is
noted here that these functions and subroutines cannot be intermixed in a given subprogram.
In the descriptions that follow, the applicable standard for each intrinsic procedure is noted.

8.2 ABORT — Abort the program

Description:
ABORT causes immediate termination of the program. On operating systems
that support a core dump, ABORT will produce a core dump. It will also print
a backtrace, unless -fno-backtrace is given.
Standard : GNU extension
Class: Subroutine
Syntax : CALL ABORT
Return value:
Does not return.
Example:
program test_abort
integer :: i = 1, j = 2
if (i /= j) call abort
end program test_abort
112 The GNU Fortran Compiler

See also: Section 8.101 [EXIT], page 179,

Section 8.163 [KILL], page 218,
Section 8.39 [BACKTRACE], page 138,

8.3 ABS — Absolute value

Description:
ABS(A) computes the absolute value of A.
Standard : Fortran 77 and later, has overloads that are GNU extensions
Class: Elemental function
Syntax : RESULT = ABS(A)
Arguments:
A The type of the argument shall be an INTEGER, REAL,
or COMPLEX.
Return value:
The return value is of the same type and kind as the argument except the return
value is REAL for a COMPLEX argument.
Example:
program test_abs
integer :: i = -1
real :: x = -1.e0
complex :: z = (-1.e0,0.e0)
i = abs(i)
x = abs(x)
x = abs(z)
end program test_abs

Specific names:
Name Argument Return type Standard
ABS(A) REAL(4) A REAL(4) Fortran 77 and later
CABS(A) COMPLEX(4) A REAL(4) Fortran 77 and later
DABS(A) REAL(8) A REAL(8) Fortran 77 and later
IABS(A) INTEGER(4) A INTEGER(4) Fortran 77 and later
BABS(A) INTEGER(1) A INTEGER(1) GNU extension
IIABS(A) INTEGER(2) A INTEGER(2) GNU extension
JIABS(A) INTEGER(4) A INTEGER(4) GNU extension
KIABS(A) INTEGER(8) A INTEGER(8) GNU extension
ZABS(A) COMPLEX(8) A REAL(8) GNU extension
CDABS(A) COMPLEX(8) A REAL(8) GNU extension

8.4 ACCESS — Checks file access modes

Description:
ACCESS(NAME, MODE) checks whether the file NAME exists, is readable, writable
or executable. Except for the executable check, ACCESS can be replaced by
Fortran 95’s INQUIRE.
Chapter 8: Intrinsic Procedures 113

Standard : GNU extension

Class: Inquiry function
Syntax : RESULT = ACCESS(NAME, MODE)
Arguments:
NAME Scalar CHARACTER of default kind with the file
name. Trailing blank are ignored unless the char-
acter achar(0) is present, then all characters up to
and excluding achar(0) are used as file name.
MODE Scalar CHARACTER of default kind with the file access
mode, may be any concatenation of "r" (readable),
"w" (writable) and "x" (executable), or " " to check
for existence.
Return value:
Returns a scalar INTEGER, which is 0 if the file is accessible in the given mode;
otherwise or if an invalid argument has been given for MODE the value 1 is
returned.
Example:
program access_test
implicit none
character(len=*), parameter :: file = 'test.dat'
character(len=*), parameter :: file2 = 'test.dat '//achar(0)
if(access(file,' ') == 0) print *, trim(file),' is exists'
if(access(file,'r') == 0) print *, trim(file),' is readable'
if(access(file,'w') == 0) print *, trim(file),' is writable'
if(access(file,'x') == 0) print *, trim(file),' is executable'
if(access(file2,'rwx') == 0) &
print *, trim(file2),' is readable, writable and executable'
end program access_test

8.5 ACHAR — Character in ASCII collating sequence

Description:
ACHAR(I) returns the character located at position I in the ASCII collating
sequence.
Standard : Fortran 77 and later, with KIND argument Fortran 2003 and later
Class: Elemental function
Syntax : RESULT = ACHAR(I [, KIND])
Arguments:
I The type shall be INTEGER.
KIND (Optional) A scalar INTEGER constant expression in-
dicating the kind parameter of the result.
Return value:
The return value is of type CHARACTER with a length of one. If the KIND
argument is present, the return value is of the specified kind and of the default
kind otherwise.
114 The GNU Fortran Compiler

Example:
program test_achar
character c
c = achar(32)
end program test_achar

Note: See Section 8.143 [ICHAR], page 206, for a discussion of converting between
numerical values and formatted string representations.
See also: Section 8.59 [CHAR], page 149,
Section 8.135 [IACHAR], page 201,
Section 8.143 [ICHAR], page 206,

8.6 ACOS — Arccosine function

Description:
ACOS(X) computes the arccosine of X (inverse of COS(X)).
Standard : Fortran 77 and later, for a complex argument Fortran 2008 or later
Class: Elemental function
Syntax : RESULT = ACOS(X)
Arguments:
X The type shall either be REAL with a magnitude that
is less than or equal to one - or the type shall be
COMPLEX.
Return value:
The return value is of the same type and kind as X. The real part of the result
is in radians and lies in the range 0 ≤ < acos(x) ≤ π.
Example:
program test_acos
real(8) :: x = 0.866_8
x = acos(x)
end program test_acos

Specific names:
Name Argument Return type Standard
ACOS(X) REAL(4) X REAL(4) Fortran 77 and later
DACOS(X) REAL(8) X REAL(8) Fortran 77 and later
See also: Inverse function:
Section 8.73 [COS], page 160,
Degrees function:
Section 8.7 [ACOSD], page 114,

8.7 ACOSD — Arccosine function, degrees

Description:
ACOSD(X) computes the arccosine of X in degrees (inverse of COSD(X)).
This function is for compatibility only and should be avoided in favor of stan-
dard constructs wherever possible.
Chapter 8: Intrinsic Procedures 115

Standard : GNU extension, enabled with -fdec-math

Class: Elemental function
Syntax : RESULT = ACOSD(X)
Arguments:
X The type shall either be REAL with a magnitude that
is less than or equal to one - or the type shall be
COMPLEX.
Return value:
The return value is of the same type and kind as X. The real part of the result
is in degrees and lies in the range 0 ≤ < acos(x) ≤ 180.
Example:
program test_acosd
real(8) :: x = 0.866_8
x = acosd(x)
end program test_acosd

Specific names:
Name Argument Return type Standard
ACOSD(X) REAL(4) X REAL(4) GNU extension
DACOSD(X) REAL(8) X REAL(8) GNU extension
See also: Inverse function:
Section 8.74 [COSD], page 160,
Radians function:
Section 8.6 [ACOS], page 114,

8.8 ACOSH — Inverse hyperbolic cosine function

Description:
ACOSH(X) computes the inverse hyperbolic cosine of X.
Standard : Fortran 2008 and later
Class: Elemental function
Syntax : RESULT = ACOSH(X)
Arguments:
X The type shall be REAL or COMPLEX.
Return value:
The return value has the same type and kind as X. If X is complex, the
imaginary part of the result is in radians and lies between 0 ≤ = acosh(x) ≤ π.
Example:
PROGRAM test_acosh
REAL(8), DIMENSION(3) :: x = (/ 1.0, 2.0, 3.0 /)
WRITE (*,*) ACOSH(x)
END PROGRAM
116 The GNU Fortran Compiler

Specific names:
Name Argument Return type Standard
DACOSH(X) REAL(8) X REAL(8) GNU extension
See also: Inverse function:
Section 8.75 [COSH], page 161,

8.9 ADJUSTL — Left adjust a string

Description:
ADJUSTL(STRING) will left adjust a string by removing leading spaces. Spaces
are inserted at the end of the string as needed.
Standard : Fortran 90 and later
Class: Elemental function
Syntax : RESULT = ADJUSTL(STRING)
Arguments:
STRING The type shall be CHARACTER.
Return value:
The return value is of type CHARACTER and of the same kind as STRING where
leading spaces are removed and the same number of spaces are inserted on the
end of STRING.
Example:
program test_adjustl
character(len=20) :: str = ' gfortran'
str = adjustl(str)
print *, str
end program test_adjustl

See also: Section 8.10 [ADJUSTR], page 116,

Section 8.275 [TRIM], page 286,

8.10 ADJUSTR — Right adjust a string

Description:
ADJUSTR(STRING) will right adjust a string by removing trailing spaces. Spaces
are inserted at the start of the string as needed.
Standard : Fortran 90 and later
Class: Elemental function
Syntax : RESULT = ADJUSTR(STRING)
Arguments:
STR The type shall be CHARACTER.
Return value:
The return value is of type CHARACTER and of the same kind as STRING where
trailing spaces are removed and the same number of spaces are inserted at the
start of STRING.
Chapter 8: Intrinsic Procedures 117

Example:
program test_adjustr
character(len=20) :: str = 'gfortran'
str = adjustr(str)
print *, str
end program test_adjustr

See also: Section 8.9 [ADJUSTL], page 116,

Section 8.275 [TRIM], page 286,

8.11 AIMAG — Imaginary part of complex number

Description:
AIMAG(Z) yields the imaginary part of complex argument Z. The IMAG(Z) and
IMAGPART(Z) intrinsic functions are provided for compatibility with g77, and
their use in new code is strongly discouraged.
Standard : Fortran 77 and later, has overloads that are GNU extensions
Class: Elemental function
Syntax : RESULT = AIMAG(Z)
Arguments:
Z The type of the argument shall be COMPLEX.
Return value:
The return value is of type REAL with the kind type parameter of the argument.
Example:
program test_aimag
complex(4) z4
complex(8) z8
z4 = cmplx(1.e0_4, 0.e0_4)
z8 = cmplx(0.e0_8, 1.e0_8)
print *, aimag(z4), dimag(z8)
end program test_aimag

Specific names:
Name Argument Return type Standard
AIMAG(Z) COMPLEX Z REAL Fortran 77 and later
DIMAG(Z) COMPLEX(8) Z REAL(8) GNU extension
IMAG(Z) COMPLEX Z REAL GNU extension
IMAGPART(Z) COMPLEX Z REAL GNU extension

8.12 AINT — Truncate to a whole number

Description:
AINT(A [, KIND]) truncates its argument to a whole number.
Standard : Fortran 77 and later
Class: Elemental function
Syntax : RESULT = AINT(A [, KIND])
118 The GNU Fortran Compiler

Arguments:
A The type of the argument shall be REAL.
KIND (Optional) A scalar INTEGER constant expression in-
dicating the kind parameter of the result.
Return value:
The return value is of type REAL with the kind type parameter of the argument
if the optional KIND is absent; otherwise, the kind type parameter will be given
by KIND. If the magnitude of X is less than one, AINT(X) returns zero. If the
magnitude is equal to or greater than one then it returns the largest whole
number that does not exceed its magnitude. The sign is the same as the sign
of X.
Example:
program test_aint
real(4) x4
real(8) x8
x4 = 1.234E0_4
x8 = 4.321_8
print *, aint(x4), dint(x8)
x8 = aint(x4,8)
end program test_aint

Specific names:
Name Argument Return type Standard
AINT(A) REAL(4) A REAL(4) Fortran 77 and later
DINT(A) REAL(8) A REAL(8) Fortran 77 and later

8.13 ALARM — Execute a routine after a given delay

Description:
ALARM(SECONDS, HANDLER [, STATUS]) causes external subroutine HANDLER
to be executed after a delay of SECONDS by using alarm(2) to set up a signal
and signal(2) to catch it. If STATUS is supplied, it will be returned with the
number of seconds remaining until any previously scheduled alarm was due to
be delivered, or zero if there was no previously scheduled alarm.
Standard : GNU extension
Class: Subroutine
Syntax : CALL ALARM(SECONDS, HANDLER [, STATUS])
Arguments:
SECONDS The type of the argument shall be a scalar INTEGER.
It is INTENT(IN).
HANDLER Signal handler (INTEGER FUNCTION or SUBROUTINE)
or dummy/global INTEGER scalar. The scalar values
may be either SIG_IGN=1 to ignore the alarm gen-
erated or SIG_DFL=0 to set the default action. It is
INTENT(IN).
STATUS (Optional) STATUS shall be a scalar variable of the
default INTEGER kind. It is INTENT(OUT).
Chapter 8: Intrinsic Procedures 119

Example:
program test_alarm
external handler_print
integer i
call alarm (3, handler_print, i)
print *, i
call sleep(10)
end program test_alarm
This will cause the external routine handler print to be called after 3 seconds.

8.14 ALL — All values in MASK along DIM are true

Description:
ALL(MASK [, DIM]) determines if all the values are true in MASK in the array
along dimension DIM.
Standard : Fortran 90 and later
Class: Transformational function
Syntax : RESULT = ALL(MASK [, DIM])
Arguments:
MASK The type of the argument shall be LOGICAL and it
shall not be scalar.
DIM (Optional) DIM shall be a scalar integer with a value
that lies between one and the rank of MASK.
Return value:
ALL(MASK) returns a scalar value of type LOGICAL where the kind type param-
eter is the same as the kind type parameter of MASK. If DIM is present, then
ALL(MASK, DIM) returns an array with the rank of MASK minus 1. The shape
is determined from the shape of MASK where the DIM dimension is elided.
(A) ALL(MASK) is true if all elements of MASK are true. It also is true
if MASK has zero size; otherwise, it is false.
(B) If the rank of MASK is one, then ALL(MASK,DIM) is equivalent to
ALL(MASK). If the rank is greater than one, then ALL(MASK,DIM)
is determined by applying ALL to the array sections.
Example:
program test_all
logical l
l = all((/.true., .true., .true./))
print *, l
call section
contains
subroutine section
integer a(2,3), b(2,3)
a = 1
b = 1
b(2,2) = 2
print *, all(a .eq. b, 1)
print *, all(a .eq. b, 2)
120 The GNU Fortran Compiler

end subroutine section

end program test_all

8.15 ALLOCATED — Status of an allocatable entity

Description:
ALLOCATED(ARRAY) and ALLOCATED(SCALAR) check the allocation status of AR-
RAY and SCALAR, respectively.
Standard : Fortran 90 and later. Note, the SCALAR= keyword and allocatable scalar entities
are available in Fortran 2003 and later.
Class: Inquiry function
Syntax :
RESULT = ALLOCATED(ARRAY)
RESULT = ALLOCATED(SCALAR)
Arguments:
ARRAY The argument shall be an ALLOCATABLE array.
SCALAR The argument shall be an ALLOCATABLE scalar.
Return value:
The return value is a scalar LOGICAL with the default logical kind type parame-
ter. If the argument is allocated, then the result is .TRUE.; otherwise, it returns
.FALSE.
Example:
program test_allocated
integer :: i = 4
real(4), allocatable :: x(:)
if (.not. allocated(x)) allocate(x(i))
end program test_allocated

8.16 AND — Bitwise logical AND

Description:
Bitwise logical AND.
This intrinsic routine is provided for backwards compatibility with GNU For-
tran 77. For integer arguments, programmers should consider the use of the
Section 8.137 [IAND], page 202, intrinsic defined by the Fortran standard.
Standard : GNU extension
Class: Function
Syntax : RESULT = AND(I, J)
Arguments:
I The type shall be either a scalar INTEGER type or a
scalar LOGICAL type or a boz-literal-constant.
J The type shall be the same as the type of I or a boz-
literal-constant. I and J shall not both be boz-literal-
constants. If either I or J is a boz-literal-constant,
then the other argument must be a scalar INTEGER.
Chapter 8: Intrinsic Procedures 121

Return value:
The return type is either a scalar INTEGER or a scalar LOGICAL. If the kind type
parameters differ, then the smaller kind type is implicitly converted to larger
kind, and the return has the larger kind. A boz-literal-constant is converted to
an INTEGER with the kind type parameter of the other argument as-if a call to
Section 8.149 [INT], page 210, occurred.
Example:
PROGRAM test_and
LOGICAL :: T = .TRUE., F = .FALSE.
INTEGER :: a, b
DATA a / Z'F' /, b / Z'3' /

WRITE (*,*) AND(T, T), AND(T, F), AND(F, T), AND(F, F)

WRITE (*,*) AND(a, b)
END PROGRAM
See also: Fortran 95 elemental function:
Section 8.137 [IAND], page 202,

8.17 ANINT — Nearest whole number

Description:
ANINT(A [, KIND]) rounds its argument to the nearest whole number.
Standard : Fortran 77 and later
Class: Elemental function
Syntax : RESULT = ANINT(A [, KIND])
Arguments:
A The type of the argument shall be REAL.
KIND (Optional) A scalar INTEGER constant expression in-
dicating the kind parameter of the result.
Return value:
The return value is of type real with the kind type parameter of the argument if
the optional KIND is absent; otherwise, the kind type parameter will be given
by KIND. If A is greater than zero, ANINT(A) returns AINT(X+0.5). If A is
less than or equal to zero then it returns AINT(X-0.5).
Example:
program test_anint
real(4) x4
real(8) x8
x4 = 1.234E0_4
x8 = 4.321_8
print *, anint(x4), dnint(x8)
x8 = anint(x4,8)
end program test_anint
Specific names:
Name Argument Return type Standard
ANINT(A) REAL(4) A REAL(4) Fortran 77 and later
DNINT(A) REAL(8) A REAL(8) Fortran 77 and later
122 The GNU Fortran Compiler

8.18 ANY — Any value in MASK along DIM is true

Description:
ANY(MASK [, DIM]) determines if any of the values in the logical array MASK
along dimension DIM are .TRUE..
Standard : Fortran 90 and later
Class: Transformational function
Syntax : RESULT = ANY(MASK [, DIM])
Arguments:
MASK The type of the argument shall be LOGICAL and it
shall not be scalar.
DIM (Optional) DIM shall be a scalar integer with a value
that lies between one and the rank of MASK.
Return value:
ANY(MASK) returns a scalar value of type LOGICAL where the kind type param-
eter is the same as the kind type parameter of MASK. If DIM is present, then
ANY(MASK, DIM) returns an array with the rank of MASK minus 1. The shape
is determined from the shape of MASK where the DIM dimension is elided.
(A) ANY(MASK) is true if any element of MASK is true; otherwise, it is
false. It also is false if MASK has zero size.
(B) If the rank of MASK is one, then ANY(MASK,DIM) is equivalent to
ANY(MASK). If the rank is greater than one, then ANY(MASK,DIM)
is determined by applying ANY to the array sections.
Example:
program test_any
logical l
l = any((/.true., .true., .true./))
print *, l
call section
contains
subroutine section
integer a(2,3), b(2,3)
a = 1
b = 1
b(2,2) = 2
print *, any(a .eq. b, 1)
print *, any(a .eq. b, 2)
end subroutine section
end program test_any

8.19 ASIN — Arcsine function

Description:
ASIN(X) computes the arcsine of its X (inverse of SIN(X)).
Standard : Fortran 77 and later, for a complex argument Fortran 2008 or later
Class: Elemental function
Chapter 8: Intrinsic Procedures 123

Syntax : RESULT = ASIN(X)

Arguments:
X The type shall be either REAL and a magnitude that
is less than or equal to one - or be COMPLEX.
Return value:
The return value is of the same type and kind as X. The real part of the result
is in radians and lies in the range −π/2 ≤ < asin(x) ≤ π/2.
Example:
program test_asin
real(8) :: x = 0.866_8
x = asin(x)
end program test_asin

Specific names:
Name Argument Return type Standard
ASIN(X) REAL(4) X REAL(4) Fortran 77 and later
DASIN(X) REAL(8) X REAL(8) Fortran 77 and later
See also: Inverse function:
Section 8.249 [SIN], page 269,
Degrees function:
Section 8.20 [ASIND], page 123,

8.20 ASIND — Arcsine function, degrees

Description:
ASIND(X) computes the arcsine of its X in degrees (inverse of SIND(X)).
This function is for compatibility only and should be avoided in favor of stan-
dard constructs wherever possible.
Standard : GNU extension, enabled with -fdec-math.
Class: Elemental function
Syntax : RESULT = ASIND(X)
Arguments:
X The type shall be either REAL and a magnitude that
is less than or equal to one - or be COMPLEX.
Return value:
The return value is of the same type and kind as X. The real part of the result
is in degrees and lies in the range −90 ≤ < asin(x) ≤ 90.
Example:
program test_asind
real(8) :: x = 0.866_8
x = asind(x)
end program test_asind

Specific names:
Name Argument Return type Standard
ASIND(X) REAL(4) X REAL(4) GNU extension
124 The GNU Fortran Compiler

DASIND(X) REAL(8) X REAL(8) GNU extension

See also: Inverse function:
Section 8.250 [SIND], page 270,
Radians function:
Section 8.19 [ASIN], page 122,

8.21 ASINH — Inverse hyperbolic sine function

Description:
ASINH(X) computes the inverse hyperbolic sine of X.
Standard : Fortran 2008 and later
Class: Elemental function
Syntax : RESULT = ASINH(X)
Arguments:
X The type shall be REAL or COMPLEX.
Return value:
The return value is of the same type and kind as X. If X is complex, the
imaginary part of the result is in radians and lies between −π/2 ≤ = asinh(x) ≤
π/2.
Example:
PROGRAM test_asinh
REAL(8), DIMENSION(3) :: x = (/ -1.0, 0.0, 1.0 /)
WRITE (*,*) ASINH(x)
END PROGRAM

Specific names:
Name Argument Return type Standard
DASINH(X) REAL(8) X REAL(8) GNU extension.
See also: Inverse function:
Section 8.251 [SINH], page 271,

8.22 ASSOCIATED — Status of a pointer or pointer/target pair

Description:
ASSOCIATED(POINTER [, TARGET]) determines the status of the pointer
POINTER or if POINTER is associated with the target TARGET.
Standard : Fortran 90 and later
Class: Inquiry function
Syntax : RESULT = ASSOCIATED(POINTER [, TARGET])
Arguments:
POINTER POINTER shall have the POINTER attribute and it
can be of any type.
Chapter 8: Intrinsic Procedures 125

TARGET (Optional) TARGET shall be a pointer or a target.

It must have the same type, kind type parameter,
and array rank as POINTER.
The association status of neither POINTER nor TARGET shall be undefined.

Return value:
ASSOCIATED(POINTER) returns a scalar value of type LOGICAL(4). There are
several cases:

(A) When the optional TARGET is not present then

ASSOCIATED(POINTER) is true if POINTER is associated with a
target; otherwise, it returns false.

(B) If TARGET is present and a scalar target, the result is true if

TARGET is not a zero-sized storage sequence and the target associ-
ated with POINTER occupies the same storage units. If POINTER
is disassociated, the result is false.

(C) If TARGET is present and an array target, the result is true if

TARGET and POINTER have the same shape, are not zero-sized
arrays, are arrays whose elements are not zero-sized storage se-
quences, and TARGET and POINTER occupy the same storage
units in array element order. As in case(B), the result is false, if
POINTER is disassociated.

(D) If TARGET is present and an scalar pointer, the result is true

if TARGET is associated with POINTER, the target associated
with TARGET are not zero-sized storage sequences and occupy
the same storage units. The result is false, if either TARGET or
POINTER is disassociated.

(E) If TARGET is present and an array pointer, the result is true if

target associated with POINTER and the target associated with
TARGET have the same shape, are not zero-sized arrays, are ar-
rays whose elements are not zero-sized storage sequences, and TAR-
GET and POINTER occupy the same storage units in array ele-
ment order. The result is false, if either TARGET or POINTER is
disassociated.

Example:
program test_associated
implicit none
real, target :: tgt(2) = (/1., 2./)
real, pointer :: ptr(:)
ptr => tgt
if (associated(ptr) .eqv. .false.) call abort
if (associated(ptr,tgt) .eqv. .false.) call abort
end program test_associated

See also: Section 8.209 [NULL], page 246,

126 The GNU Fortran Compiler

8.23 ATAN — Arctangent function

Description:
ATAN(X) computes the arctangent of X.
Standard : Fortran 77 and later, for a complex argument and for two arguments Fortran
2008 or later
Class: Elemental function
Syntax :
RESULT = ATAN(X)
RESULT = ATAN(Y, X)
Arguments:
X The type shall be REAL or COMPLEX; if Y is present,
X shall be REAL.
Y The type and kind type parameter shall be the same
as X.
Return value:
The return value is of the same type and kind as X. If Y is present, the result
is identical to ATAN2(Y,X). Otherwise, it the arcus tangent of X, where the real
part of the result is in radians and lies in the range −π/2 ≤ < atan(x) ≤ π/2.
Example:
program test_atan
real(8) :: x = 2.866_8
x = atan(x)
end program test_atan

Specific names:
Name Argument Return type Standard
ATAN(X) REAL(4) X REAL(4) Fortran 77 and later
DATAN(X) REAL(8) X REAL(8) Fortran 77 and later
See also: Inverse function:
Section 8.265 [TAN], page 280,
Degrees function:
Section 8.24 [ATAND], page 126,

8.24 ATAND — Arctangent function, degrees

Description:
ATAND(X) computes the arctangent of X in degrees (inverse of Section 8.266
[TAND], page 280).
This function is for compatibility only and should be avoided in favor of stan-
dard constructs wherever possible.
Standard : GNU extension, enabled with -fdec-math.
Class: Elemental function
Chapter 8: Intrinsic Procedures 127

Syntax :
RESULT = ATAND(X)
RESULT = ATAND(Y, X)
Arguments:
X The type shall be REAL or COMPLEX; if Y is present,
X shall be REAL.
Y The type and kind type parameter shall be the same
as X.
Return value:
The return value is of the same type and kind as X. If Y is present, the result
is identical to ATAND2(Y,X). Otherwise, it is the arcus tangent of X, where the
real part of the result is in degrees and lies in the range −90 ≤ < atand(x) ≤ 90.
Example:
program test_atand
real(8) :: x = 2.866_8
x = atand(x)
end program test_atand

Specific names:
Name Argument Return type Standard
ATAND(X) REAL(4) X REAL(4) GNU extension
DATAND(X) REAL(8) X REAL(8) GNU extension
See also: Inverse function:
Section 8.266 [TAND], page 280,
Radians function:
Section 8.23 [ATAN], page 126,

8.25 ATAN2 — Arctangent function

Description:
ATAN2(Y, X) computes the principal value of the argument function of the com-
plex number X +iY . This function can be used to transform from Cartesian into
polar coordinates and allows to determine the angle in the correct quadrant.
Standard : Fortran 77 and later
Class: Elemental function
Syntax : RESULT = ATAN2(Y, X)
Arguments:
Y The type shall be REAL.
X The type and kind type parameter shall be the same
as Y. If Y is zero, then X must be nonzero.
Return value:
The return value has the same type and kind type parameter as Y. It is the
principal value of the complex number X + iY . If X is nonzero, then it lies in
the range −π ≤ atan(x) ≤ π. The sign is positive if Y is positive. If Y is zero,
128 The GNU Fortran Compiler

then the return value is zero if X is strictly positive, π if X is negative and Y

is positive zero (or the processor does not handle signed zeros), and −π if X is
negative and Y is negative zero. Finally, if X is zero, then the magnitude of
the result is π/2.
Example:
program test_atan2
real(4) :: x = 1.e0_4, y = 0.5e0_4
x = atan2(y,x)
end program test_atan2

Specific names:
Name Argument Return type Standard
ATAN2(X, Y) REAL(4) X, Y REAL(4) Fortran 77 and later
DATAN2(X, Y) REAL(8) X, Y REAL(8) Fortran 77 and later
See also: Alias:
Section 8.23 [ATAN], page 126,
Degrees function:
Section 8.26 [ATAN2D], page 128,

8.26 ATAN2D — Arctangent function, degrees

Description:
ATAN2D(Y, X) computes the principal value of the argument function of the
complex number X + iY in degrees. This function can be used to transform
from Cartesian into polar coordinates and allows to determine the angle in the
correct quadrant.
This function is for compatibility only and should be avoided in favor of stan-
dard constructs wherever possible.
Standard : GNU extension, enabled with -fdec-math.
Class: Elemental function
Syntax : RESULT = ATAN2D(Y, X)
Arguments:
Y The type shall be REAL.
X The type and kind type parameter shall be the same
as Y. If Y is zero, then X must be nonzero.
Return value:
The return value has the same type and kind type parameter as Y. It is the
principal value of the complex number X + iY . If X is nonzero, then it lies
in the range −180 ≤ atan(x) ≤ 180. The sign is positive if Y is positive. If
Y is zero, then the return value is zero if X is strictly positive, 180 if X is
negative and Y is positive zero (or the processor does not handle signed zeros),
and −180 if X is negative and Y is negative zero. Finally, if X is zero, then
the magnitude of the result is 90.
Example:
program test_atan2d
Chapter 8: Intrinsic Procedures 129

real(4) :: x = 1.e0_4, y = 0.5e0_4

x = atan2d(y,x)
end program test_atan2d

Specific names:
Name Argument Return type Standard
ATAN2D(X, Y) REAL(4) X, Y REAL(4) GNU extension
DATAN2D(X, Y) REAL(8) X, Y REAL(8) GNU extension
See also: Alias:
Section 8.24 [ATAND], page 126,
Radians function:
Section 8.25 [ATAN2], page 127,

8.27 ATANH — Inverse hyperbolic tangent function

Description:
ATANH(X) computes the inverse hyperbolic tangent of X.
Standard : Fortran 2008 and later
Class: Elemental function
Syntax : RESULT = ATANH(X)
Arguments:
X The type shall be REAL or COMPLEX.
Return value:
The return value has same type and kind as X. If X is complex, the imaginary
part of the result is in radians and lies between −π/2 ≤ = atanh(x) ≤ π/2.
Example:
PROGRAM test_atanh
REAL, DIMENSION(3) :: x = (/ -1.0, 0.0, 1.0 /)
WRITE (*,*) ATANH(x)
END PROGRAM

Specific names:
Name Argument Return type Standard
DATANH(X) REAL(8) X REAL(8) GNU extension
See also: Inverse function:
Section 8.267 [TANH], page 281,

8.28 ATOMIC_ADD — Atomic ADD operation

Description:
ATOMIC_ADD(ATOM, VALUE) atomically adds the value of VALUE to the variable
ATOM. When STAT is present and the invocation was successful, it is assigned
the value 0. If it is present and the invocation has failed, it is assigned a positive
value; in particular, for a coindexed ATOM, if the remote image has stopped,
it is assigned the value of ISO_FORTRAN_ENV’s STAT_STOPPED_IMAGE and if the
remote image has failed, the value STAT_FAILED_IMAGE.
130 The GNU Fortran Compiler

Standard : TS 18508 or later

Class: Atomic subroutine
Syntax : CALL ATOMIC_ADD (ATOM, VALUE [, STAT])
Arguments:
ATOM Scalar coarray or coindexed variable of integer type
with ATOMIC_INT_KIND kind.
VALUE Scalar of the same type as ATOM. If the kind is
different, the value is converted to the kind of ATOM.
STAT (optional) Scalar default-kind integer variable.
Example:
program atomic
use iso_fortran_env
integer(atomic_int_kind) :: atom[*]
call atomic_add (atom[1], this_image())
end program atomic

See also: Section 8.31 [ATOMIC DEFINE], page 132,

Section 8.32 [ATOMIC FETCH ADD], page 132,
Section 9.1 [ISO FORTRAN ENV], page 293,
Section 8.29 [ATOMIC AND], page 130,
Section 8.36 [ATOMIC OR], page 135,
Section 8.38 [ATOMIC XOR], page 137,

8.29 ATOMIC_AND — Atomic bitwise AND operation

Description:
ATOMIC_AND(ATOM, VALUE) atomically defines ATOM with the bitwise AND
between the values of ATOM and VALUE. When STAT is present and the
invocation was successful, it is assigned the value 0. If it is present and the
invocation has failed, it is assigned a positive value; in particular, for a coin-
dexed ATOM, if the remote image has stopped, it is assigned the value of
ISO_FORTRAN_ENV’s STAT_STOPPED_IMAGE and if the remote image has failed,
the value STAT_FAILED_IMAGE.
Standard : TS 18508 or later
Class: Atomic subroutine
Syntax : CALL ATOMIC_AND (ATOM, VALUE [, STAT])
Arguments:
ATOM Scalar coarray or coindexed variable of integer type
with ATOMIC_INT_KIND kind.
VALUE Scalar of the same type as ATOM. If the kind is
different, the value is converted to the kind of ATOM.
STAT (optional) Scalar default-kind integer variable.
Example:
program atomic
Chapter 8: Intrinsic Procedures 131

use iso_fortran_env
integer(atomic_int_kind) :: atom[*]
call atomic_and (atom[1], int(b'10100011101'))
end program atomic

See also: Section 8.31 [ATOMIC DEFINE], page 132,

Section 8.33 [ATOMIC FETCH AND], page 133,
Section 9.1 [ISO FORTRAN ENV], page 293,
Section 8.28 [ATOMIC ADD], page 129,
Section 8.36 [ATOMIC OR], page 135,
Section 8.38 [ATOMIC XOR], page 137,

8.30 ATOMIC_CAS — Atomic compare and swap

Description:
ATOMIC_CAS compares the variable ATOM with the value of COMPARE; if the
value is the same, ATOM is set to the value of NEW. Additionally, OLD is set
to the value of ATOM that was used for the comparison. When STAT is present
and the invocation was successful, it is assigned the value 0. If it is present and
the invocation has failed, it is assigned a positive value; in particular, for a
coindexed ATOM, if the remote image has stopped, it is assigned the value of
ISO_FORTRAN_ENV’s STAT_STOPPED_IMAGE and if the remote image has failed,
the value STAT_FAILED_IMAGE.

Standard : TS 18508 or later

Class: Atomic subroutine

Syntax : CALL ATOMIC_CAS (ATOM, OLD, COMPARE, NEW [, STAT])

Arguments:
ATOM Scalar coarray or coindexed variable of either integer
type with ATOMIC_INT_KIND kind or logical type with
ATOMIC_LOGICAL_KIND kind.
OLD Scalar of the same type and kind as ATOM.
COMPARE Scalar variable of the same type and kind as ATOM.
NEW Scalar variable of the same type as ATOM. If kind is
different, the value is converted to the kind of ATOM.
STAT (optional) Scalar default-kind integer variable.

Example:
program atomic
use iso_fortran_env
logical(atomic_logical_kind) :: atom[*], prev
call atomic_cas (atom[1], prev, .false., .true.))
end program atomic

See also: Section 8.31 [ATOMIC DEFINE], page 132,

Section 8.37 [ATOMIC REF], page 136,
Section 9.1 [ISO FORTRAN ENV], page 293,
132 The GNU Fortran Compiler

8.31 ATOMIC_DEFINE — Setting a variable atomically

Description:
ATOMIC_DEFINE(ATOM, VALUE) defines the variable ATOM with the value
VALUE atomically. When STAT is present and the invocation was successful,
it is assigned the value 0. If it is present and the invocation has failed, it
is assigned a positive value; in particular, for a coindexed ATOM, if the
remote image has stopped, it is assigned the value of ISO_FORTRAN_ENV’s
STAT_STOPPED_IMAGE and if the remote image has failed, the value
STAT_FAILED_IMAGE.
Standard : Fortran 2008 and later; with STAT, TS 18508 or later
Class: Atomic subroutine
Syntax : CALL ATOMIC_DEFINE (ATOM, VALUE [, STAT])
Arguments:
ATOM Scalar coarray or coindexed variable of either integer
type with ATOMIC_INT_KIND kind or logical type with
ATOMIC_LOGICAL_KIND kind.

VALUE Scalar of the same type as ATOM. If the kind is

different, the value is converted to the kind of ATOM.
STAT (optional) Scalar default-kind integer variable.
Example:
program atomic
use iso_fortran_env
integer(atomic_int_kind) :: atom[*]
call atomic_define (atom[1], this_image())
end program atomic

See also: Section 8.37 [ATOMIC REF], page 136,

Section 8.30 [ATOMIC CAS], page 131,
Section 9.1 [ISO FORTRAN ENV], page 293,
Section 8.28 [ATOMIC ADD], page 129,
Section 8.29 [ATOMIC AND], page 130,
Section 8.36 [ATOMIC OR], page 135,
Section 8.38 [ATOMIC XOR], page 137,

8.32 ATOMIC_FETCH_ADD — Atomic ADD operation with prior

fetch
Description:
ATOMIC_FETCH_ADD(ATOM, VALUE, OLD) atomically stores the value of ATOM
in OLD and adds the value of VALUE to the variable ATOM. When STAT
is present and the invocation was successful, it is assigned the value 0. If it is
present and the invocation has failed, it is assigned a positive value; in particu-
lar, for a coindexed ATOM, if the remote image has stopped, it is assigned the
value of ISO_FORTRAN_ENV’s STAT_STOPPED_IMAGE and if the remote image has
failed, the value STAT_FAILED_IMAGE.
Chapter 8: Intrinsic Procedures 133

Standard : TS 18508 or later

Class: Atomic subroutine
Syntax : CALL ATOMIC_FETCH_ADD (ATOM, VALUE, old [, STAT])
Arguments:
ATOM Scalar coarray or coindexed variable of integer type
with ATOMIC_INT_KIND kind. ATOMIC_LOGICAL_KIND
kind.

VALUE Scalar of the same type as ATOM. If the kind is

different, the value is converted to the kind of ATOM.
OLD Scalar of the same type and kind as ATOM.
STAT (optional) Scalar default-kind integer variable.
Example:
program atomic
use iso_fortran_env
integer(atomic_int_kind) :: atom[*], old
call atomic_add (atom[1], this_image(), old)
end program atomic

See also: Section 8.31 [ATOMIC DEFINE], page 132,

Section 8.28 [ATOMIC ADD], page 129,
Section 9.1 [ISO FORTRAN ENV], page 293,
Section 8.33 [ATOMIC FETCH AND], page 133,
Section 8.34 [ATOMIC FETCH OR], page 134,
Section 8.35 [ATOMIC FETCH XOR], page 135,

8.33 ATOMIC_FETCH_AND — Atomic bitwise AND operation

with prior fetch
Description:
ATOMIC_AND(ATOM, VALUE) atomically stores the value of ATOM in OLD
and defines ATOM with the bitwise AND between the values of ATOM and
VALUE. When STAT is present and the invocation was successful, it is
assigned the value 0. If it is present and the invocation has failed, it is assigned
a positive value; in particular, for a coindexed ATOM, if the remote image has
stopped, it is assigned the value of ISO_FORTRAN_ENV’s STAT_STOPPED_IMAGE
and if the remote image has failed, the value STAT_FAILED_IMAGE.
Standard : TS 18508 or later
Class: Atomic subroutine
Syntax : CALL ATOMIC_FETCH_AND (ATOM, VALUE, OLD [, STAT])
Arguments:
ATOM Scalar coarray or coindexed variable of integer type
with ATOMIC_INT_KIND kind.
VALUE Scalar of the same type as ATOM. If the kind is
different, the value is converted to the kind of ATOM.
134 The GNU Fortran Compiler

OLD Scalar of the same type and kind as ATOM.

STAT (optional) Scalar default-kind integer variable.
Example:
program atomic
use iso_fortran_env
integer(atomic_int_kind) :: atom[*], old
call atomic_fetch_and (atom[1], int(b'10100011101'), old)
end program atomic

See also: Section 8.31 [ATOMIC DEFINE], page 132,

Section 8.29 [ATOMIC AND], page 130,
Section 9.1 [ISO FORTRAN ENV], page 293,
Section 8.32 [ATOMIC FETCH ADD], page 132,
Section 8.34 [ATOMIC FETCH OR], page 134,
Section 8.35 [ATOMIC FETCH XOR], page 135,

8.34 ATOMIC_FETCH_OR — Atomic bitwise OR operation with

prior fetch
Description:
ATOMIC_OR(ATOM, VALUE) atomically stores the value of ATOM in OLD and
defines ATOM with the bitwise OR between the values of ATOM and VALUE.
When STAT is present and the invocation was successful, it is assigned the
value 0. If it is present and the invocation has failed, it is assigned a positive
value; in particular, for a coindexed ATOM, if the remote image has stopped,
it is assigned the value of ISO_FORTRAN_ENV’s STAT_STOPPED_IMAGE and if the
remote image has failed, the value STAT_FAILED_IMAGE.
Standard : TS 18508 or later
Class: Atomic subroutine
Syntax : CALL ATOMIC_FETCH_OR (ATOM, VALUE, OLD [, STAT])
Arguments:
ATOM Scalar coarray or coindexed variable of integer type
with ATOMIC_INT_KIND kind.
VALUE Scalar of the same type as ATOM. If the kind is
different, the value is converted to the kind of ATOM.
OLD Scalar of the same type and kind as ATOM.
STAT (optional) Scalar default-kind integer variable.
Example:
program atomic
use iso_fortran_env
integer(atomic_int_kind) :: atom[*], old
call atomic_fetch_or (atom[1], int(b'10100011101'), old)
end program atomic

See also: Section 8.31 [ATOMIC DEFINE], page 132,

Section 8.36 [ATOMIC OR], page 135,
Section 9.1 [ISO FORTRAN ENV], page 293,
Chapter 8: Intrinsic Procedures 135

Section 8.32 [ATOMIC FETCH ADD], page 132,

Section 8.33 [ATOMIC FETCH AND], page 133,
Section 8.35 [ATOMIC FETCH XOR], page 135,

8.35 ATOMIC_FETCH_XOR — Atomic bitwise XOR operation

with prior fetch
Description:
ATOMIC_XOR(ATOM, VALUE) atomically stores the value of ATOM in OLD and
defines ATOM with the bitwise XOR between the values of ATOM and VALUE.
When STAT is present and the invocation was successful, it is assigned the
value 0. If it is present and the invocation has failed, it is assigned a positive
value; in particular, for a coindexed ATOM, if the remote image has stopped,
it is assigned the value of ISO_FORTRAN_ENV’s STAT_STOPPED_IMAGE and if the
remote image has failed, the value STAT_FAILED_IMAGE.
Standard : TS 18508 or later
Class: Atomic subroutine
Syntax : CALL ATOMIC_FETCH_XOR (ATOM, VALUE, OLD [, STAT])
Arguments:
ATOM Scalar coarray or coindexed variable of integer type
with ATOMIC_INT_KIND kind.
VALUE Scalar of the same type as ATOM. If the kind is
different, the value is converted to the kind of ATOM.
OLD Scalar of the same type and kind as ATOM.
STAT (optional) Scalar default-kind integer variable.
Example:
program atomic
use iso_fortran_env
integer(atomic_int_kind) :: atom[*], old
call atomic_fetch_xor (atom[1], int(b'10100011101'), old)
end program atomic
See also: Section 8.31 [ATOMIC DEFINE], page 132,
Section 8.38 [ATOMIC XOR], page 137,
Section 9.1 [ISO FORTRAN ENV], page 293,
Section 8.32 [ATOMIC FETCH ADD], page 132,
Section 8.33 [ATOMIC FETCH AND], page 133,
Section 8.34 [ATOMIC FETCH OR], page 134,

8.36 ATOMIC_OR — Atomic bitwise OR operation

Description:
ATOMIC_OR(ATOM, VALUE) atomically defines ATOM with the bitwise AND be-
tween the values of ATOM and VALUE. When STAT is present and the
invocation was successful, it is assigned the value 0. If it is present and the
invocation has failed, it is assigned a positive value; in particular, for a coin-
dexed ATOM, if the remote image has stopped, it is assigned the value of
136 The GNU Fortran Compiler

ISO_FORTRAN_ENV’s STAT_STOPPED_IMAGE and if the remote image has failed,

the value STAT_FAILED_IMAGE.
Standard : TS 18508 or later
Class: Atomic subroutine
Syntax : CALL ATOMIC_OR (ATOM, VALUE [, STAT])
Arguments:
ATOM Scalar coarray or coindexed variable of integer type
with ATOMIC_INT_KIND kind.
VALUE Scalar of the same type as ATOM. If the kind is
different, the value is converted to the kind of ATOM.
STAT (optional) Scalar default-kind integer variable.
Example:
program atomic
use iso_fortran_env
integer(atomic_int_kind) :: atom[*]
call atomic_or (atom[1], int(b'10100011101'))
end program atomic

See also: Section 8.31 [ATOMIC DEFINE], page 132,

Section 8.34 [ATOMIC FETCH OR], page 134,
Section 9.1 [ISO FORTRAN ENV], page 293,
Section 8.28 [ATOMIC ADD], page 129,
Section 8.36 [ATOMIC OR], page 135,
Section 8.38 [ATOMIC XOR], page 137,

8.37 ATOMIC_REF — Obtaining the value of a variable

atomically
Description:
ATOMIC_DEFINE(ATOM, VALUE) atomically assigns the value of the variable
ATOM to VALUE. When STAT is present and the invocation was successful,
it is assigned the value 0. If it is present and the invocation has failed, it
is assigned a positive value; in particular, for a coindexed ATOM, if the
remote image has stopped, it is assigned the value of ISO_FORTRAN_ENV’s
STAT_STOPPED_IMAGE and if the remote image has failed, the value
STAT_FAILED_IMAGE.
Standard : Fortran 2008 and later; with STAT, TS 18508 or later
Class: Atomic subroutine
Syntax : CALL ATOMIC_REF(VALUE, ATOM [, STAT])
Arguments:
VALUE Scalar of the same type as ATOM. If the kind is
different, the value is converted to the kind of ATOM.
ATOM Scalar coarray or coindexed variable of either integer
type with ATOMIC_INT_KIND kind or logical type with
ATOMIC_LOGICAL_KIND kind.
Chapter 8: Intrinsic Procedures 137

STAT (optional) Scalar default-kind integer variable.

Example:
program atomic
use iso_fortran_env
logical(atomic_logical_kind) :: atom[*]
logical :: val
call atomic_ref (atom, .false.)
! ...
call atomic_ref (atom, val)
if (val) then
print *, "Obtained"
end if
end program atomic

See also: Section 8.31 [ATOMIC DEFINE], page 132,

Section 8.30 [ATOMIC CAS], page 131,
Section 9.1 [ISO FORTRAN ENV], page 293,
Section 8.32 [ATOMIC FETCH ADD], page 132,
Section 8.33 [ATOMIC FETCH AND], page 133,
Section 8.34 [ATOMIC FETCH OR], page 134,
Section 8.35 [ATOMIC FETCH XOR], page 135,

8.38 ATOMIC_XOR — Atomic bitwise OR operation

Description:
ATOMIC_AND(ATOM, VALUE) atomically defines ATOM with the bitwise XOR
between the values of ATOM and VALUE. When STAT is present and the
invocation was successful, it is assigned the value 0. If it is present and the
invocation has failed, it is assigned a positive value; in particular, for a coin-
dexed ATOM, if the remote image has stopped, it is assigned the value of
ISO_FORTRAN_ENV’s STAT_STOPPED_IMAGE and if the remote image has failed,
the value STAT_FAILED_IMAGE.
Standard : TS 18508 or later
Class: Atomic subroutine
Syntax : CALL ATOMIC_XOR (ATOM, VALUE [, STAT])
Arguments:
ATOM Scalar coarray or coindexed variable of integer type
with ATOMIC_INT_KIND kind.
VALUE Scalar of the same type as ATOM. If the kind is
different, the value is converted to the kind of ATOM.
STAT (optional) Scalar default-kind integer variable.
Example:
program atomic
use iso_fortran_env
integer(atomic_int_kind) :: atom[*]
call atomic_xor (atom[1], int(b'10100011101'))
end program atomic
138 The GNU Fortran Compiler

See also: Section 8.31 [ATOMIC DEFINE], page 132,

Section 8.35 [ATOMIC FETCH XOR], page 135,
Section 9.1 [ISO FORTRAN ENV], page 293,
Section 8.28 [ATOMIC ADD], page 129,
Section 8.36 [ATOMIC OR], page 135,
Section 8.38 [ATOMIC XOR], page 137,

8.39 BACKTRACE — Show a backtrace

Description:
BACKTRACE shows a backtrace at an arbitrary place in user code. Program
execution continues normally afterwards. The backtrace information is printed
to the unit corresponding to ERROR_UNIT in ISO_FORTRAN_ENV.
Standard : GNU extension
Class: Subroutine
Syntax : CALL BACKTRACE
Arguments:
None
See also: Section 8.2 [ABORT], page 111,

8.40 BESSEL_J0 — Bessel function of the first kind of order 0

Description:
BESSEL_J0(X) computes the Bessel function of the first kind of order 0 of X.
This function is available under the name BESJ0 as a GNU extension.
Standard : Fortran 2008 and later
Class: Elemental function
Syntax : RESULT = BESSEL_J0(X)
Arguments:
X The type shall be REAL.
Return value:
The return value is of type REAL and lies in the range −0.4027... ≤
Bessel(0, x) ≤ 1. It has the same kind as X.
Example:
program test_besj0
real(8) :: x = 0.0_8
x = bessel_j0(x)
end program test_besj0

Specific names:
Name Argument Return type Standard
DBESJ0(X) REAL(8) X REAL(8) GNU extension
Chapter 8: Intrinsic Procedures 139

8.41 BESSEL_J1 — Bessel function of the first kind of order 1

Description:
BESSEL_J1(X) computes the Bessel function of the first kind of order 1 of X.
This function is available under the name BESJ1 as a GNU extension.
Standard : Fortran 2008
Class: Elemental function
Syntax : RESULT = BESSEL_J1(X)
Arguments:
X The type shall be REAL.
Return value:
The return value is of type REAL and lies in the range −0.5818... ≤
Bessel(0, x) ≤ 0.5818. It has the same kind as X.
Example:
program test_besj1
real(8) :: x = 1.0_8
x = bessel_j1(x)
end program test_besj1

Specific names:
Name Argument Return type Standard
DBESJ1(X) REAL(8) X REAL(8) GNU extension

8.42 BESSEL_JN — Bessel function of the first kind

Description:
BESSEL_JN(N, X) computes the Bessel function of the first kind of order N of
X. This function is available under the name BESJN as a GNU extension. If N
and X are arrays, their ranks and shapes shall conform.
BESSEL_JN(N1, N2, X) returns an array with the Bessel functions of the first
kind of the orders N1 to N2.
Standard : Fortran 2008 and later, negative N is allowed as GNU extension
Class: Elemental function, except for the transformational function BESSEL_JN(N1,
N2, X)
Syntax :
RESULT = BESSEL_JN(N, X)
RESULT = BESSEL_JN(N1, N2, X)
Arguments:
N Shall be a scalar or an array of type INTEGER.
N1 Shall be a non-negative scalar of type INTEGER.
N2 Shall be a non-negative scalar of type INTEGER.
X Shall be a scalar or an array of type REAL; for BESSEL_
JN(N1, N2, X) it shall be scalar.
140 The GNU Fortran Compiler

Return value:
The return value is a scalar of type REAL. It has the same kind as X.
Note: The transformational function uses a recurrence algorithm which might, for
some values of X, lead to different results than calls to the elemental function.
Example:
program test_besjn
real(8) :: x = 1.0_8
x = bessel_jn(5,x)
end program test_besjn

Specific names:
Name Argument Return type Standard
DBESJN(N, X) INTEGER N REAL(8) GNU extension
REAL(8) X

8.43 BESSEL_Y0 — Bessel function of the second kind of

order 0
Description:
BESSEL_Y0(X) computes the Bessel function of the second kind of order 0 of X.
This function is available under the name BESY0 as a GNU extension.
Standard : Fortran 2008 and later
Class: Elemental function
Syntax : RESULT = BESSEL_Y0(X)
Arguments:
X The type shall be REAL.
Return value:
The return value is of type REAL. It has the same kind as X.
Example:
program test_besy0
real(8) :: x = 0.0_8
x = bessel_y0(x)
end program test_besy0

Specific names:
Name Argument Return type Standard
DBESY0(X) REAL(8) X REAL(8) GNU extension

8.44 BESSEL_Y1 — Bessel function of the second kind of

order 1
Description:
BESSEL_Y1(X) computes the Bessel function of the second kind of order 1 of X.
This function is available under the name BESY1 as a GNU extension.
Standard : Fortran 2008 and later
Chapter 8: Intrinsic Procedures 141

Class: Elemental function

Syntax : RESULT = BESSEL_Y1(X)
Arguments:
X The type shall be REAL.
Return value:
The return value is of type REAL. It has the same kind as X.
Example:
program test_besy1
real(8) :: x = 1.0_8
x = bessel_y1(x)
end program test_besy1
Specific names:
Name Argument Return type Standard
DBESY1(X) REAL(8) X REAL(8) GNU extension

8.45 BESSEL_YN — Bessel function of the second kind

Description:
BESSEL_YN(N, X) computes the Bessel function of the second kind of order N
of X. This function is available under the name BESYN as a GNU extension. If
N and X are arrays, their ranks and shapes shall conform.
BESSEL_YN(N1, N2, X) returns an array with the Bessel functions of the first
kind of the orders N1 to N2.
Standard : Fortran 2008 and later, negative N is allowed as GNU extension
Class: Elemental function, except for the transformational function BESSEL_YN(N1,
N2, X)
Syntax :
RESULT = BESSEL_YN(N, X)
RESULT = BESSEL_YN(N1, N2, X)
Arguments:
N Shall be a scalar or an array of type INTEGER .
N1 Shall be a non-negative scalar of type INTEGER.
N2 Shall be a non-negative scalar of type INTEGER.
X Shall be a scalar or an array of type REAL; for BESSEL_
YN(N1, N2, X) it shall be scalar.
Return value:
The return value is a scalar of type REAL. It has the same kind as X.
Note: The transformational function uses a recurrence algorithm which might, for
some values of X, lead to different results than calls to the elemental function.
Example:
program test_besyn
real(8) :: x = 1.0_8
x = bessel_yn(5,x)
end program test_besyn
142 The GNU Fortran Compiler

Specific names:
Name Argument Return type Standard
DBESYN(N,X) INTEGER N REAL(8) GNU extension
REAL(8) X

8.46 BGE — Bitwise greater than or equal to

Description:
Determines whether an integral is a bitwise greater than or equal to another.
Standard : Fortran 2008 and later
Class: Elemental function
Syntax : RESULT = BGE(I, J)
Arguments:
I Shall be of INTEGER type.
J Shall be of INTEGER type, and of the same kind as I.
Return value:
The return value is of type LOGICAL and of the default kind.
See also: Section 8.47 [BGT], page 142,
Section 8.49 [BLE], page 143,
Section 8.50 [BLT], page 143,

8.47 BGT — Bitwise greater than

Description:
Determines whether an integral is a bitwise greater than another.
Standard : Fortran 2008 and later
Class: Elemental function
Syntax : RESULT = BGT(I, J)
Arguments:
I Shall be of INTEGER type.
J Shall be of INTEGER type, and of the same kind as I.
Return value:
The return value is of type LOGICAL and of the default kind.
See also: Section 8.46 [BGE], page 142,
Section 8.49 [BLE], page 143,
Section 8.50 [BLT], page 143,

8.48 BIT_SIZE — Bit size inquiry function

Description:
BIT_SIZE(I) returns the number of bits (integer precision plus sign bit) repre-
sented by the type of I. The result of BIT_SIZE(I) is independent of the actual
value of I.
Chapter 8: Intrinsic Procedures 143

Standard : Fortran 90 and later

Class: Inquiry function
Syntax : RESULT = BIT_SIZE(I)
Arguments:
I The type shall be INTEGER.
Return value:
The return value is of type INTEGER
Example:
program test_bit_size
integer :: i = 123
integer :: size
size = bit_size(i)
print *, size
end program test_bit_size

8.49 BLE — Bitwise less than or equal to

Description:
Determines whether an integral is a bitwise less than or equal to another.
Standard : Fortran 2008 and later
Class: Elemental function
Syntax : RESULT = BLE(I, J)
Arguments:
I Shall be of INTEGER type.
J Shall be of INTEGER type, and of the same kind as I.
Return value:
The return value is of type LOGICAL and of the default kind.
See also: Section 8.47 [BGT], page 142,
Section 8.46 [BGE], page 142,
Section 8.50 [BLT], page 143,

8.50 BLT — Bitwise less than

Description:
Determines whether an integral is a bitwise less than another.
Standard : Fortran 2008 and later
Class: Elemental function
Syntax : RESULT = BLT(I, J)
Arguments:
I Shall be of INTEGER type.
J Shall be of INTEGER type, and of the same kind as I.
144 The GNU Fortran Compiler

Return value:
The return value is of type LOGICAL and of the default kind.
See also: Section 8.46 [BGE], page 142,
Section 8.47 [BGT], page 142,
Section 8.49 [BLE], page 143,

8.51 BTEST — Bit test function

Description:
BTEST(I,POS) returns logical .TRUE. if the bit at POS in I is set. The counting
of the bits starts at 0.
Standard : Fortran 90 and later, has overloads that are GNU extensions
Class: Elemental function
Syntax : RESULT = BTEST(I, POS)
Arguments:
I The type shall be INTEGER.
POS The type shall be INTEGER.
Return value:
The return value is of type LOGICAL
Example:
program test_btest
integer :: i = 32768 + 1024 + 64
integer :: pos
logical :: bool
do pos=0,16
bool = btest(i, pos)
print *, pos, bool
end do
end program test_btest

Specific names:
Name Argument Return Standard
type
BTEST(I,POS) INTEGER I,POS LOGICAL Fortran 95 and later
BBTEST(I,POS) INTEGER(1) I,POS LOGICAL(1) GNU extension
BITEST(I,POS) INTEGER(2) I,POS LOGICAL(2) GNU extension
BJTEST(I,POS) INTEGER(4) I,POS LOGICAL(4) GNU extension
BKTEST(I,POS) INTEGER(8) I,POS LOGICAL(8) GNU extension

8.52 C_ASSOCIATED — Status of a C pointer

Description:
C_ASSOCIATED(c_ptr_1[, c_ptr_2]) determines the status of the C pointer
c ptr 1 or if c ptr 1 is associated with the target c ptr 2.
Standard : Fortran 2003 and later
Chapter 8: Intrinsic Procedures 145

Class: Inquiry function

Syntax : RESULT = C_ASSOCIATED(c_ptr_1[, c_ptr_2])
Arguments:
c ptr 1 Scalar of the type C_PTR or C_FUNPTR.
c ptr 2 (Optional) Scalar of the same type as c ptr 1.
Return value:
The return value is of type LOGICAL; it is .false. if either c ptr 1 is a C NULL
pointer or if c ptr1 and c ptr 2 point to different addresses.
Example:
subroutine association_test(a,b)
use iso_c_binding, only: c_associated, c_loc, c_ptr
implicit none
real, pointer :: a
type(c_ptr) :: b
if(c_associated(b, c_loc(a))) &
stop 'b and a do not point to same target'
end subroutine association_test

See also: Section 8.56 [C LOC], page 147,

Section 8.55 [C FUNLOC], page 146,

8.53 C_F_POINTER — Convert C into Fortran pointer

Description:
C_F_POINTER(CPTR, FPTR[, SHAPE]) assigns the target of the C pointer CPTR
to the Fortran pointer FPTR and specifies its shape.
Standard : Fortran 2003 and later
Class: Subroutine
Syntax : CALL C_F_POINTER(CPTR, FPTR[, SHAPE])
Arguments:
CPTR scalar of the type C_PTR. It is INTENT(IN).
FPTR pointer interoperable with cptr. It is INTENT(OUT).
SHAPE (Optional) Rank-one array of type INTEGER with
INTENT(IN). It shall be present if and only if fptr
is an array. The size must be equal to the rank of
fptr.
Example:
program main
use iso_c_binding
implicit none
interface
subroutine my_routine(p) bind(c,name='myC_func')
import :: c_ptr
type(c_ptr), intent(out) :: p
end subroutine
end interface
type(c_ptr) :: cptr
146 The GNU Fortran Compiler

real,pointer :: a(:)
call my_routine(cptr)
call c_f_pointer(cptr, a, [12])
end program main

See also: Section 8.56 [C LOC], page 147,

Section 8.54 [C F PROCPOINTER], page 146,

8.54 C_F_PROCPOINTER — Convert C into Fortran procedure

pointer
Description:
C_F_PROCPOINTER(CPTR, FPTR) Assign the target of the C function pointer
CPTR to the Fortran procedure pointer FPTR.
Standard : Fortran 2003 and later
Class: Subroutine
Syntax : CALL C_F_PROCPOINTER(cptr, fptr)
Arguments:
CPTR scalar of the type C_FUNPTR. It is INTENT(IN).
FPTR procedure pointer interoperable with cptr. It is
INTENT(OUT).
Example:
program main
use iso_c_binding
implicit none
abstract interface
function func(a)
import :: c_float
real(c_float), intent(in) :: a
real(c_float) :: func
end function
end interface
interface
function getIterFunc() bind(c,name="getIterFunc")
import :: c_funptr
type(c_funptr) :: getIterFunc
end function
end interface
type(c_funptr) :: cfunptr
procedure(func), pointer :: myFunc
cfunptr = getIterFunc()
call c_f_procpointer(cfunptr, myFunc)
end program main

See also: Section 8.56 [C LOC], page 147,

Section 8.53 [C F POINTER], page 145,

8.55 C_FUNLOC — Obtain the C address of a procedure

Description:
C_FUNLOC(x) determines the C address of the argument.
Chapter 8: Intrinsic Procedures 147

Standard : Fortran 2003 and later

Class: Inquiry function

Syntax : RESULT = C_FUNLOC(x)

Arguments:
x Interoperable function or pointer to such function.

Return value:
The return value is of type C_FUNPTR and contains the C address of the argu-
ment.

Example:
module x
use iso_c_binding
implicit none
contains
subroutine sub(a) bind(c)
real(c_float) :: a
a = sqrt(a)+5.0
end subroutine sub
end module x
program main
use iso_c_binding
use x
implicit none
interface
subroutine my_routine(p) bind(c,name='myC_func')
import :: c_funptr
type(c_funptr), intent(in) :: p
end subroutine
end interface
call my_routine(c_funloc(sub))
end program main

See also: Section 8.52 [C ASSOCIATED], page 144,

Section 8.56 [C LOC], page 147,
Section 8.53 [C F POINTER], page 145,
Section 8.54 [C F PROCPOINTER], page 146,

8.56 C_LOC — Obtain the C address of an object

Description:
C_LOC(X) determines the C address of the argument.

Standard : Fortran 2003 and later

Class: Inquiry function

Syntax : RESULT = C_LOC(X)

148 The GNU Fortran Compiler

Arguments:
X Shall have either the POINTER or TARGET attribute.
It shall not be a coindexed object. It shall either be a
variable with interoperable type and kind type param-
eters, or be a scalar, nonpolymorphic variable with no
length type parameters.

Return value:
The return value is of type C_PTR and contains the C address of the argument.
Example:
subroutine association_test(a,b)
use iso_c_binding, only: c_associated, c_loc, c_ptr
implicit none
real, pointer :: a
type(c_ptr) :: b
if(c_associated(b, c_loc(a))) &
stop 'b and a do not point to same target'
end subroutine association_test

See also: Section 8.52 [C ASSOCIATED], page 144,

Section 8.55 [C FUNLOC], page 146,
Section 8.53 [C F POINTER], page 145,
Section 8.54 [C F PROCPOINTER], page 146,

8.57 C_SIZEOF — Size in bytes of an expression

Description:
C_SIZEOF(X) calculates the number of bytes of storage the expression X occu-
pies.
Standard : Fortran 2008
Class: Inquiry function of the module ISO_C_BINDING
Syntax : N = C_SIZEOF(X)
Arguments:
X The argument shall be an interoperable data entity.
Return value:
The return value is of type integer and of the system-dependent kind C_SIZE_T
(from the ISO_C_BINDING module). Its value is the number of bytes occupied by
the argument. If the argument has the POINTER attribute, the number of bytes
of the storage area pointed to is returned. If the argument is of a derived type
with POINTER or ALLOCATABLE components, the return value does not account
for the sizes of the data pointed to by these components.
Example:
use iso_c_binding
integer(c_int) :: i
real(c_float) :: r, s(5)
print *, (c_sizeof(s)/c_sizeof(r) == 5)
Chapter 8: Intrinsic Procedures 149

end
The example will print T unless you are using a platform where default REAL
variables are unusually padded.
See also: Section 8.253 [SIZEOF], page 272,
Section 8.260 [STORAGE SIZE], page 276,

8.58 CEILING — Integer ceiling function

Description:
CEILING(A) returns the least integer greater than or equal to A.
Standard : Fortran 95 and later
Class: Elemental function
Syntax : RESULT = CEILING(A [, KIND])
Arguments:
A The type shall be REAL.
KIND (Optional) A scalar INTEGER constant expression in-
dicating the kind parameter of the result.
Return value:
The return value is of type INTEGER(KIND) if KIND is present and a default-
kind INTEGER otherwise.
Example:
program test_ceiling
real :: x = 63.29
real :: y = -63.59
print *, ceiling(x) ! returns 64
print *, ceiling(y) ! returns -63
end program test_ceiling

See also: Section 8.109 [FLOOR], page 184,

Section 8.206 [NINT], page 244,

8.59 CHAR — Character conversion function

Description:
CHAR(I [, KIND]) returns the character represented by the integer I.
Standard : Fortran 77 and later
Class: Elemental function
Syntax : RESULT = CHAR(I [, KIND])
Arguments:
I The type shall be INTEGER.
KIND (Optional) A scalar INTEGER constant expression in-
dicating the kind parameter of the result.
Return value:
The return value is of type CHARACTER(1)
150 The GNU Fortran Compiler

Example:
program test_char
integer :: i = 74
character(1) :: c
c = char(i)
print *, i, c ! returns 'J'
end program test_char

Specific names:
Name Argument Return type Standard
CHAR(I) INTEGER I CHARACTER(LEN=1) Fortran 77 and later
Note: See Section 8.143 [ICHAR], page 206, for a discussion of converting between
numerical values and formatted string representations.
See also: Section 8.5 [ACHAR], page 113,
Section 8.135 [IACHAR], page 201,
Section 8.143 [ICHAR], page 206,

8.60 CHDIR — Change working directory

Description:
Change current working directory to a specified path.
This intrinsic is provided in both subroutine and function forms; however, only
one form can be used in any given program unit.
Standard : GNU extension
Class: Subroutine, function
Syntax :
CALL CHDIR(NAME [, STATUS])
STATUS = CHDIR(NAME)
Arguments:
NAME The type shall be CHARACTER of default kind and shall
specify a valid path within the file system.
STATUS (Optional) INTEGER status flag of the default kind.
Returns 0 on success, and a system specific and
nonzero error code otherwise.
Example:
PROGRAM test_chdir
CHARACTER(len=255) :: path
CALL getcwd(path)
WRITE(*,*) TRIM(path)
CALL chdir("/tmp")
CALL getcwd(path)
WRITE(*,*) TRIM(path)
END PROGRAM

See also: Section 8.124 [GETCWD], page 195,

Chapter 8: Intrinsic Procedures 151

8.61 CHMOD — Change access permissions of files

Description:
CHMOD changes the permissions of a file.
This intrinsic is provided in both subroutine and function forms; however, only
one form can be used in any given program unit.

Standard : GNU extension

Class: Subroutine, function

Syntax :
CALL CHMOD(NAME, MODE[, STATUS])
STATUS = CHMOD(NAME, MODE)

Arguments:

NAME Scalar CHARACTER of default kind with the file name.

Trailing blanks are ignored unless the character
achar(0) is present, then all characters up to and
excluding achar(0) are used as the file name.

MODE Scalar CHARACTER of default kind giving the file per-

mission. MODE uses the same syntax as the chmod
utility as defined by the POSIX standard. The argu-
ment shall either be a string of a nonnegative octal
number or a symbolic mode.

STATUS (optional) scalar INTEGER, which is 0 on success and

nonzero otherwise.

Return value:
In either syntax, STATUS is set to 0 on success and nonzero otherwise.

Example: CHMOD as subroutine

program chmod_test
implicit none
integer :: status
call chmod('test.dat','u+x',status)
print *, 'Status: ', status
end program chmod_test

CHMOD as function:
program chmod_test
implicit none
integer :: status
status = chmod('test.dat','u+x')
print *, 'Status: ', status
end program chmod_test
152 The GNU Fortran Compiler

8.62 CMPLX — Complex conversion function

Description:
CMPLX(X [, Y [, KIND]]) returns a complex number where X is converted to
the real component. If Y is present it is converted to the imaginary component.
If Y is not present then the imaginary component is set to 0.0. If X is complex
then Y must not be present.
Standard : Fortran 77 and later
Class: Elemental function
Syntax : RESULT = CMPLX(X [, Y [, KIND]])
Arguments:
X The type may be INTEGER, REAL, or COMPLEX.
Y (Optional; only allowed if X is not COMPLEX.) May
be INTEGER or REAL.
KIND (Optional) A scalar INTEGER constant expression in-
dicating the kind parameter of the result.
Return value:
The return value is of COMPLEX type, with a kind equal to KIND if it is specified.
If KIND is not specified, the result is of the default COMPLEX kind, regardless
of the kinds of X and Y.
Example:
program test_cmplx
integer :: i = 42
real :: x = 3.14
complex :: z
z = cmplx(i, x)
print *, z, cmplx(x)
end program test_cmplx

See also: Section 8.71 [COMPLEX], page 158,

8.63 CO_BROADCAST — Copy a value to all images the current

set of images
Description:
CO_BROADCAST copies the value of argument A on the image with image index
SOURCE_IMAGE to all images in the current team. A becomes defined as if by
intrinsic assignment. If the execution was successful and STAT is present, it is
assigned the value zero. If the execution failed, STAT gets assigned a nonzero
value and, if present, ERRMSG gets assigned a value describing the occurred
error.
Standard : Technical Specification (TS) 18508 or later
Class: Collective subroutine
Syntax : CALL CO_BROADCAST(A, SOURCE_IMAGE [, STAT, ERRMSG])
Chapter 8: Intrinsic Procedures 153

Arguments:
A INTENT(INOUT) argument; shall have the
same dynamic type and type parameters on all
images of the current team. If it is an array, it
shall have the same shape on all images.
SOURCE IMAGEa scalar integer expression. It shall have the same
value on all images and refer to an image of the
current team.
STAT (optional) a scalar integer variable
ERRMSG (optional) a scalar character variable
Example:
program test
integer :: val(3)
if (this_image() == 1) then
val = [1, 5, 3]
end if
call co_broadcast (val, source_image=1)
print *, this_image, ":", val
end program test

See also: Section 8.64 [CO MAX], page 153,

Section 8.65 [CO MIN], page 154,
Section 8.67 [CO SUM], page 156,
Section 8.66 [CO REDUCE], page 155,

8.64 CO_MAX — Maximal value on the current set of images

Description:
CO_MAX determines element-wise the maximal value of A on all images of the
current team. If RESULT IMAGE is present, the maximum values are returned
in A on the specified image only and the value of A on the other images become
undefined. If RESULT IMAGE is not present, the value is returned on all
images. If the execution was successful and STAT is present, it is assigned the
value zero. If the execution failed, STAT gets assigned a nonzero value and, if
present, ERRMSG gets assigned a value describing the occurred error.
Standard : Technical Specification (TS) 18508 or later
Class: Collective subroutine
Syntax : CALL CO_MAX(A [, RESULT_IMAGE, STAT, ERRMSG])
Arguments:
A shall be an integer, real or character variable,
which has the same type and type parameters
on all images of the team.
RESULT IMAGE(optional) a scalar integer expression; if present,
it shall have the same value on all images and
refer to an image of the current team.
STAT (optional) a scalar integer variable
ERRMSG (optional) a scalar character variable
154 The GNU Fortran Compiler

Example:
program test
integer :: val
val = this_image ()
call co_max (val, result_image=1)
if (this_image() == 1) then
write(*,*) "Maximal value", val ! prints num_images()
end if
end program test

See also: Section 8.65 [CO MIN], page 154,

Section 8.67 [CO SUM], page 156,
Section 8.66 [CO REDUCE], page 155,
Section 8.63 [CO BROADCAST], page 152,

8.65 CO_MIN — Minimal value on the current set of images

Description:
CO_MIN determines element-wise the minimal value of A on all images of the
current team. If RESULT IMAGE is present, the minimal values are returned
in A on the specified image only and the value of A on the other images become
undefined. If RESULT IMAGE is not present, the value is returned on all
images. If the execution was successful and STAT is present, it is assigned the
value zero. If the execution failed, STAT gets assigned a nonzero value and, if
present, ERRMSG gets assigned a value describing the occurred error.
Standard : Technical Specification (TS) 18508 or later
Class: Collective subroutine
Syntax : CALL CO_MIN(A [, RESULT_IMAGE, STAT, ERRMSG])
Arguments:
A shall be an integer, real or character variable,
which has the same type and type parameters
on all images of the team.
RESULT IMAGE(optional) a scalar integer expression; if present,
it shall have the same value on all images and
refer to an image of the current team.
STAT (optional) a scalar integer variable
ERRMSG (optional) a scalar character variable
Example:
program test
integer :: val
val = this_image ()
call co_min (val, result_image=1)
if (this_image() == 1) then
write(*,*) "Minimal value", val ! prints 1
end if
end program test

See also: Section 8.64 [CO MAX], page 153,

Section 8.67 [CO SUM], page 156,
Chapter 8: Intrinsic Procedures 155

Section 8.66 [CO REDUCE], page 155,

Section 8.63 [CO BROADCAST], page 152,

8.66 CO_REDUCE — Reduction of values on the current set of

images
Description:
CO_REDUCE determines element-wise the reduction of the value of A on all images
of the current team. The pure function passed as OPERATION is used to
pairwise reduce the values of A by passing either the value of A of different
images or the result values of such a reduction as argument. If A is an array,
the deduction is done element wise. If RESULT IMAGE is present, the result
values are returned in A on the specified image only and the value of A on the
other images become undefined. If RESULT IMAGE is not present, the value
is returned on all images. If the execution was successful and STAT is present,
it is assigned the value zero. If the execution failed, STAT gets assigned a
nonzero value and, if present, ERRMSG gets assigned a value describing the
occurred error.
Standard : Technical Specification (TS) 18508 or later
Class: Collective subroutine
Syntax : CALL CO_REDUCE(A, OPERATION, [, RESULT_IMAGE, STAT, ERRMSG])
Arguments:
A is an INTENT(INOUT) argument and shall be non-
polymorphic. If it is allocatable, it shall be allo-
cated; if it is a pointer, it shall be associated. A
shall have the same type and type parameters on
all images of the team; if it is an array, it shall
have the same shape on all images.
OPERATION pure function with two scalar nonallocatable ar-
guments, which shall be nonpolymorphic and
have the same type and type parameters as A.
The function shall return a nonallocatable scalar
of the same type and type parameters as A. The
function shall be the same on all images and with
regards to the arguments mathematically com-
mutative and associative. Note that OPERA-
TION may not be an elemental function, unless
it is an intrisic function.
RESULT IMAGE(optional) a scalar integer expression; if present,
it shall have the same value on all images and
refer to an image of the current team.
STAT (optional) a scalar integer variable
ERRMSG (optional) a scalar character variable
Example:
program test
156 The GNU Fortran Compiler

integer :: val
val = this_image ()
call co_reduce (val, result_image=1, operation=myprod)
if (this_image() == 1) then
write(*,*) "Product value", val ! prints num_images() factorial
end if
contains
pure function myprod(a, b)
integer, value :: a, b
integer :: myprod
myprod = a * b
end function myprod
end program test

Note: While the rules permit in principle an intrinsic function, none of the intrinsics
in the standard fulfill the criteria of having a specific function, which takes two
arguments of the same type and returning that type as result.
See also: Section 8.65 [CO MIN], page 154,
Section 8.64 [CO MAX], page 153,
Section 8.67 [CO SUM], page 156,
Section 8.63 [CO BROADCAST], page 152,

8.67 CO_SUM — Sum of values on the current set of images

Description:
CO_SUM sums up the values of each element of A on all images of the current
team. If RESULT IMAGE is present, the summed-up values are returned in
A on the specified image only and the value of A on the other images become
undefined. If RESULT IMAGE is not present, the value is returned on all
images. If the execution was successful and STAT is present, it is assigned the
value zero. If the execution failed, STAT gets assigned a nonzero value and, if
present, ERRMSG gets assigned a value describing the occurred error.
Standard : Technical Specification (TS) 18508 or later
Class: Collective subroutine
Syntax : CALL CO_SUM(A [, RESULT_IMAGE, STAT, ERRMSG])
Arguments:
A shall be an integer, real or complex variable,
which has the same type and type parameters
on all images of the team.
RESULT IMAGE(optional) a scalar integer expression; if present,
it shall have the same value on all images and
refer to an image of the current team.
STAT (optional) a scalar integer variable
ERRMSG (optional) a scalar character variable
Example:
program test
integer :: val
val = this_image ()
Chapter 8: Intrinsic Procedures 157

call co_sum (val, result_image=1)

if (this_image() == 1) then
write(*,*) "The sum is ", val ! prints (n**2 + n)/2,
! with n = num_images()
end if
end program test

See also: Section 8.64 [CO MAX], page 153,

Section 8.65 [CO MIN], page 154,
Section 8.66 [CO REDUCE], page 155,
Section 8.63 [CO BROADCAST], page 152,

8.68 COMMAND_ARGUMENT_COUNT — Get number of command

line arguments
Description:
COMMAND_ARGUMENT_COUNT returns the number of arguments passed on the com-
mand line when the containing program was invoked.
Standard : Fortran 2003 and later
Class: Inquiry function
Syntax : RESULT = COMMAND_ARGUMENT_COUNT()
Arguments:
None
Return value:
The return value is an INTEGER of default kind.
Example:
program test_command_argument_count
integer :: count
count = command_argument_count()
print *, count
end program test_command_argument_count

See also: Section 8.122 [GET COMMAND], page 193,

Section 8.123 [GET COMMAND ARGUMENT], page 194,

8.69 COMPILER_OPTIONS — Options passed to the compiler

Description:
COMPILER_OPTIONS returns a string with the options used for compiling.
Standard : Fortran 2008
Class: Inquiry function of the module ISO_FORTRAN_ENV
Syntax : STR = COMPILER_OPTIONS()
Arguments:
None
158 The GNU Fortran Compiler

Return value:
The return value is a default-kind string with system-dependent length. It
contains the compiler flags used to compile the file, which called the COMPILER_
OPTIONS intrinsic.
Example:
use iso_fortran_env
print '(4a)', 'This file was compiled by ', &
compiler_version(), ' using the options ', &
compiler_options()
end

See also: Section 8.70 [COMPILER VERSION], page 158,

Section 9.1 [ISO FORTRAN ENV], page 293,

8.70 COMPILER_VERSION — Compiler version string

Description:
COMPILER_VERSION returns a string with the name and the version of the com-
piler.
Standard : Fortran 2008
Class: Inquiry function of the module ISO_FORTRAN_ENV
Syntax : STR = COMPILER_VERSION()
Arguments:
None
Return value:
The return value is a default-kind string with system-dependent length. It
contains the name of the compiler and its version number.
Example:
use iso_fortran_env
print '(4a)', 'This file was compiled by ', &
compiler_version(), ' using the options ', &
compiler_options()
end

See also: Section 8.69 [COMPILER OPTIONS], page 157,

Section 9.1 [ISO FORTRAN ENV], page 293,

8.71 COMPLEX — Complex conversion function

Description:
COMPLEX(X, Y) returns a complex number where X is converted to the real
component and Y is converted to the imaginary component.
Standard : GNU extension
Class: Elemental function
Syntax : RESULT = COMPLEX(X, Y)
Chapter 8: Intrinsic Procedures 159

Arguments:
X The type may be INTEGER or REAL.
Y The type may be INTEGER or REAL.

Return value:
If X and Y are both of INTEGER type, then the return value is of default COMPLEX
type.
If X and Y are of REAL type, or one is of REAL type and one is of INTEGER type,
then the return value is of COMPLEX type with a kind equal to that of the REAL
argument with the highest precision.

Example:
program test_complex
integer :: i = 42
real :: x = 3.14
print *, complex(i, x)
end program test_complex

See also: Section 8.62 [CMPLX], page 152,

8.72 CONJG — Complex conjugate function

Description:
CONJG(Z) returns the conjugate of Z. If Z is (x, y) then the result is (x, -y)

Standard : Fortran 77 and later, has an overload that is a GNU extension

Class: Elemental function

Syntax : Z = CONJG(Z)

Arguments:
Z The type shall be COMPLEX.

Return value:
The return value is of type COMPLEX.

Example:
program test_conjg
complex :: z = (2.0, 3.0)
complex(8) :: dz = (2.71_8, -3.14_8)
z= conjg(z)
print *, z
dz = dconjg(dz)
print *, dz
end program test_conjg

Specific names:
Name Argument Return type Standard
DCONJG(Z) COMPLEX(8) Z COMPLEX(8) GNU extension
160 The GNU Fortran Compiler

8.73 COS — Cosine function

Description:
COS(X) computes the cosine of X.
Standard : Fortran 77 and later, has overloads that are GNU extensions
Class: Elemental function
Syntax : RESULT = COS(X)
Arguments:
X The type shall be REAL or COMPLEX.
Return value:
The return value is of the same type and kind as X. The real part of the
result is in radians. If X is of the type REAL, the return value lies in the range
−1 ≤ cos(x) ≤ 1.
Example:
program test_cos
real :: x = 0.0
x = cos(x)
end program test_cos

Specific names:
Name Argument Return type Standard
COS(X) REAL(4) X REAL(4) Fortran 77 and later
DCOS(X) REAL(8) X REAL(8) Fortran 77 and later
CCOS(X) COMPLEX(4) X COMPLEX(4) Fortran 77 and later
ZCOS(X) COMPLEX(8) X COMPLEX(8) GNU extension
CDCOS(X) COMPLEX(8) X COMPLEX(8) GNU extension
See also: Inverse function:
Section 8.6 [ACOS], page 114,
Degrees function:
Section 8.74 [COSD], page 160,

8.74 COSD — Cosine function, degrees

Description:
COSD(X) computes the cosine of X in degrees.
This function is for compatibility only and should be avoided in favor of stan-
dard constructs wherever possible.
Standard : GNU extension, enabled with -fdec-math.
Class: Elemental function
Syntax : RESULT = COSD(X)
Arguments:
X The type shall be REAL or COMPLEX.
Chapter 8: Intrinsic Procedures 161

Return value:
The return value is of the same type and kind as X. The real part of the
result is in degrees. If X is of the type REAL, the return value lies in the range
−1 ≤ cosd(x) ≤ 1.
Example:
program test_cosd
real :: x = 0.0
x = cosd(x)
end program test_cosd
Specific names:
Name Argument Return type Standard
COSD(X) REAL(4) X REAL(4) GNU extension
DCOSD(X) REAL(8) X REAL(8) GNU extension
CCOSD(X) COMPLEX(4) X COMPLEX(4) GNU extension
ZCOSD(X) COMPLEX(8) X COMPLEX(8) GNU extension
CDCOSD(X) COMPLEX(8) X COMPLEX(8) GNU extension
See also: Inverse function:
Section 8.7 [ACOSD], page 114,
Radians function:
Section 8.73 [COS], page 160,

8.75 COSH — Hyperbolic cosine function

Description:
COSH(X) computes the hyperbolic cosine of X.
Standard : Fortran 77 and later, for a complex argument Fortran 2008 or later
Class: Elemental function
Syntax : X = COSH(X)
Arguments:
X The type shall be REAL or COMPLEX.
Return value:
The return value has same type and kind as X. If X is complex, the imaginary
part of the result is in radians. If X is REAL, the return value has a lower bound
of one, cosh(x) ≥ 1.
Example:
program test_cosh
real(8) :: x = 1.0_8
x = cosh(x)
end program test_cosh
Specific names:
Name Argument Return type Standard
COSH(X) REAL(4) X REAL(4) Fortran 77 and later
DCOSH(X) REAL(8) X REAL(8) Fortran 77 and later
See also: Inverse function:
Section 8.8 [ACOSH], page 115,
162 The GNU Fortran Compiler

8.76 COTAN — Cotangent function

Description:
COTAN(X) computes the cotangent of X. Equivalent to COS(x) divided by
SIN(x), or 1 / TAN(x).
This function is for compatibility only and should be avoided in favor of stan-
dard constructs wherever possible.
Standard : GNU extension, enabled with -fdec-math.
Class: Elemental function
Syntax : RESULT = COTAN(X)
Arguments:
X The type shall be REAL or COMPLEX.
Return value:
The return value has same type and kind as X, and its value is in radians.
Example:
program test_cotan
real(8) :: x = 0.165_8
x = cotan(x)
end program test_cotan

Specific names:
Name Argument Return type Standard
COTAN(X) REAL(4) X REAL(4) GNU extension
DCOTAN(X) REAL(8) X REAL(8) GNU extension
See also: Converse function:
Section 8.265 [TAN], page 280,
Degrees function:
Section 8.77 [COTAND], page 162,

8.77 COTAND — Cotangent function, degrees

Description:
COTAND(X) computes the cotangent of X in degrees. Equivalent to COSD(x)
divided by SIND(x), or 1 / TAND(x).
Standard : GNU extension, enabled with -fdec-math.
This function is for compatibility only and should be avoided in favor of stan-
dard constructs wherever possible.
Class: Elemental function
Syntax : RESULT = COTAND(X)
Arguments:
X The type shall be REAL or COMPLEX.
Return value:
The return value has same type and kind as X, and its value is in degrees.
Chapter 8: Intrinsic Procedures 163

Example:
program test_cotand
real(8) :: x = 0.165_8
x = cotand(x)
end program test_cotand

Specific names:
Name Argument Return type Standard
COTAND(X) REAL(4) X REAL(4) GNU extension
DCOTAND(X) REAL(8) X REAL(8) GNU extension
See also: Converse function:
Section 8.266 [TAND], page 280,
Radians function:
Section 8.76 [COTAN], page 162,

8.78 COUNT — Count function

Description:
Counts the number of .TRUE. elements in a logical MASK, or, if the DIM
argument is supplied, counts the number of elements along each row of the
array in the DIM direction. If the array has zero size, or all of the elements of
MASK are .FALSE., then the result is 0.
Standard : Fortran 90 and later, with KIND argument Fortran 2003 and later
Class: Transformational function
Syntax : RESULT = COUNT(MASK [, DIM, KIND])
Arguments:
MASK The type shall be LOGICAL.
DIM (Optional) The type shall be INTEGER.
KIND (Optional) A scalar INTEGER constant expression in-
dicating the kind parameter of the result.
Return value:
The return value is of type INTEGER and of kind KIND. If KIND is absent, the
return value is of default integer kind. If DIM is present, the result is an array
with a rank one less than the rank of ARRAY, and a size corresponding to the
shape of ARRAY with the DIM dimension removed.
Example:
program test_count
integer, dimension(2,3) :: a, b
logical, dimension(2,3) :: mask
a = reshape( (/ 1, 2, 3, 4, 5, 6 /), (/ 2, 3 /))
b = reshape( (/ 0, 7, 3, 4, 5, 8 /), (/ 2, 3 /))
print '(3i3)', a(1,:)
print '(3i3)', a(2,:)
print *
print '(3i3)', b(1,:)
print '(3i3)', b(2,:)
print *
164 The GNU Fortran Compiler

mask = a.ne.b
print '(3l3)', mask(1,:)
print '(3l3)', mask(2,:)
print *
print '(3i3)', count(mask)
print *
print '(3i3)', count(mask, 1)
print *
print '(3i3)', count(mask, 2)
end program test_count

8.79 CPU_TIME — CPU elapsed time in seconds

Description:
Returns a REAL value representing the elapsed CPU time in seconds. This is
useful for testing segments of code to determine execution time.
If a time source is available, time will be reported with microsecond resolution.
If no time source is available, TIME is set to -1.0.
Note that TIME may contain a, system dependent, arbitrary offset and may not
start with 0.0. For CPU_TIME, the absolute value is meaningless, only differences
between subsequent calls to this subroutine, as shown in the example below,
should be used.
Standard : Fortran 95 and later
Class: Subroutine
Syntax : CALL CPU_TIME(TIME)
Arguments:
TIME The type shall be REAL with INTENT(OUT).
Return value:
None
Example:
program test_cpu_time
real :: start, finish
call cpu_time(start)
! put code to test here
call cpu_time(finish)
print '("Time = ",f6.3," seconds.")',finish-start
end program test_cpu_time

See also: Section 8.264 [SYSTEM CLOCK], page 279,

Section 8.82 [DATE AND TIME], page 166,

8.80 CSHIFT — Circular shift elements of an array

Description:
CSHIFT(ARRAY, SHIFT [, DIM]) performs a circular shift on elements of AR-
RAY along the dimension of DIM. If DIM is omitted it is taken to be 1. DIM
is a scalar of type INTEGER in the range of 1 ≤ DIM ≤ n) where n is the rank
of ARRAY. If the rank of ARRAY is one, then all elements of ARRAY are
Chapter 8: Intrinsic Procedures 165

shifted by SHIFT places. If rank is greater than one, then all complete rank
one sections of ARRAY along the given dimension are shifted. Elements shifted
out one end of each rank one section are shifted back in the other end.
Standard : Fortran 90 and later
Class: Transformational function
Syntax : RESULT = CSHIFT(ARRAY, SHIFT [, DIM])
Arguments:
ARRAY Shall be an array of any type.
SHIFT The type shall be INTEGER.
DIM The type shall be INTEGER.
Return value:
Returns an array of same type and rank as the ARRAY argument.
Example:
program test_cshift
integer, dimension(3,3) :: a
a = reshape( (/ 1, 2, 3, 4, 5, 6, 7, 8, 9 /), (/ 3, 3 /))
print '(3i3)', a(1,:)
print '(3i3)', a(2,:)
print '(3i3)', a(3,:)
a = cshift(a, SHIFT=(/1, 2, -1/), DIM=2)
print *
print '(3i3)', a(1,:)
print '(3i3)', a(2,:)
print '(3i3)', a(3,:)
end program test_cshift

8.81 CTIME — Convert a time into a string

Description:
CTIME converts a system time value, such as returned by Section 8.270 [TIME8],
page 283, to a string. The output will be of the form ‘Sat Aug 19 18:13:14
1995’.
This intrinsic is provided in both subroutine and function forms; however, only
one form can be used in any given program unit.
Standard : GNU extension
Class: Subroutine, function
Syntax :
CALL CTIME(TIME, RESULT).
RESULT = CTIME(TIME).
Arguments:
TIME The type shall be of type INTEGER.
RESULT The type shall be of type CHARACTER and of default
kind. It is an INTENT(OUT) argument. If the length
of this variable is too short for the time and date
string to fit completely, it will be blank on procedure
return.
166 The GNU Fortran Compiler

Return value:
The converted date and time as a string.
Example:
program test_ctime
integer(8) :: i
character(len=30) :: date
i = time8()

! Do something, main part of the program

call ctime(i,date)
print *, 'Program was started on ', date
end program test_ctime

See Also: Section 8.82 [DATE AND TIME], page 166,

Section 8.131 [GMTIME], page 198,
Section 8.183 [LTIME], page 230,
Section 8.269 [TIME], page 282,
Section 8.270 [TIME8], page 283,

8.82 DATE_AND_TIME — Date and time subroutine

Description:
DATE_AND_TIME(DATE, TIME, ZONE, VALUES) gets the corresponding date and
time information from the real-time system clock. DATE is INTENT(OUT) and
has form ccyymmdd. TIME is INTENT(OUT) and has form hhmmss.sss. ZONE
is INTENT(OUT) and has form (+-)hhmm, representing the difference with re-
spect to Coordinated Universal Time (UTC). Unavailable time and date pa-
rameters return blanks.
VALUES is INTENT(OUT) and provides the following:
VALUE(1): The year
VALUE(2): The month
VALUE(3): The day of the month
VALUE(4): Time difference with UTC in minutes
VALUE(5): The hour of the day
VALUE(6): The minutes of the hour
VALUE(7): The seconds of the minute
VALUE(8): The milliseconds of the second
Standard : Fortran 90 and later
Class: Subroutine
Syntax : CALL DATE_AND_TIME([DATE, TIME, ZONE, VALUES])
Arguments:
DATE (Optional) The type shall be CHARACTER(LEN=8) or
larger, and of default kind.
TIME (Optional) The type shall be CHARACTER(LEN=10) or
larger, and of default kind.
Chapter 8: Intrinsic Procedures 167

ZONE (Optional) The type shall be CHARACTER(LEN=5) or

larger, and of default kind.
VALUES (Optional) The type shall be INTEGER(8).
Return value:
None
Example:
program test_time_and_date
character(8) :: date
character(10) :: time
character(5) :: zone
integer,dimension(8) :: values
! using keyword arguments
call date_and_time(date,time,zone,values)
call date_and_time(DATE=date,ZONE=zone)
call date_and_time(TIME=time)
call date_and_time(VALUES=values)
print '(a,2x,a,2x,a)', date, time, zone
print '(8i5)', values
end program test_time_and_date

See also: Section 8.79 [CPU TIME], page 164,

Section 8.264 [SYSTEM CLOCK], page 279,

8.83 DBLE — Double conversion function

Description:
DBLE(A) Converts A to double precision real type.
Standard : Fortran 77 and later
Class: Elemental function
Syntax : RESULT = DBLE(A)
Arguments:
A The type shall be INTEGER, REAL, or COMPLEX.
Return value:
The return value is of type double precision real.
Example:
program test_dble
real :: x = 2.18
integer :: i = 5
complex :: z = (2.3,1.14)
print *, dble(x), dble(i), dble(z)
end program test_dble

See also: Section 8.228 [REAL], page 257,

8.84 DCMPLX — Double complex conversion function

Description:
DCMPLX(X [,Y]) returns a double complex number where X is converted to the
real component. If Y is present it is converted to the imaginary component. If
168 The GNU Fortran Compiler

Y is not present then the imaginary component is set to 0.0. If X is complex

then Y must not be present.
Standard : GNU extension
Class: Elemental function
Syntax : RESULT = DCMPLX(X [, Y])
Arguments:
X The type may be INTEGER, REAL, or COMPLEX.
Y (Optional if X is not COMPLEX.) May be INTEGER or
REAL.
Return value:
The return value is of type COMPLEX(8)
Example:
program test_dcmplx
integer :: i = 42
real :: x = 3.14
complex :: z
z = cmplx(i, x)
print *, dcmplx(i)
print *, dcmplx(x)
print *, dcmplx(z)
print *, dcmplx(x,i)
end program test_dcmplx

8.85 DIGITS — Significant binary digits function

Description:
DIGITS(X) returns the number of significant binary digits of the internal model
representation of X. For example, on a system using a 32-bit floating point
representation, a default real number would likely return 24.
Standard : Fortran 90 and later
Class: Inquiry function
Syntax : RESULT = DIGITS(X)
Arguments:
X The type may be INTEGER or REAL.
Return value:
The return value is of type INTEGER.
Example:
program test_digits
integer :: i = 12345
real :: x = 3.143
real(8) :: y = 2.33
print *, digits(i)
print *, digits(x)
print *, digits(y)
end program test_digits
Chapter 8: Intrinsic Procedures 169

8.86 DIM — Positive difference

Description:
DIM(X,Y) returns the difference X-Y if the result is positive; otherwise returns
zero.
Standard : Fortran 77 and later
Class: Elemental function
Syntax : RESULT = DIM(X, Y)
Arguments:
X The type shall be INTEGER or REAL
Y The type shall be the same type and kind as X. (As
a GNU extension, arguments of different kinds are
permitted.)
Return value:
The return value is of type INTEGER or REAL. (As a GNU extension, kind is the
largest kind of the actual arguments.)
Example:
program test_dim
integer :: i
real(8) :: x
i = dim(4, 15)
x = dim(4.345_8, 2.111_8)
print *, i
print *, x
end program test_dim

Specific names:
Name Argument Return type Standard
DIM(X,Y) REAL(4) X, Y REAL(4) Fortran 77 and later
IDIM(X,Y) INTEGER(4) X, Y INTEGER(4) Fortran 77 and later
DDIM(X,Y) REAL(8) X, Y REAL(8) Fortran 77 and later

8.87 DOT_PRODUCT — Dot product function

Description:
DOT_PRODUCT(VECTOR_A, VECTOR_B) computes the dot product multiplication
of two vectors VECTOR A and VECTOR B. The two vectors may be either
numeric or logical and must be arrays of rank one and of equal size. If the
vectors are INTEGER or REAL, the result is SUM(VECTOR_A*VECTOR_B). If the
vectors are COMPLEX, the result is SUM(CONJG(VECTOR_A)*VECTOR_B). If the
vectors are LOGICAL, the result is ANY(VECTOR_A .AND. VECTOR_B).
Standard : Fortran 90 and later
Class: Transformational function
Syntax : RESULT = DOT_PRODUCT(VECTOR_A, VECTOR_B)
170 The GNU Fortran Compiler

Arguments:
VECTOR A The type shall be numeric or LOGICAL, rank 1.
VECTOR B The type shall be numeric if VECTOR A is of nu-
meric type or LOGICAL if VECTOR A is of type
LOGICAL. VECTOR B shall be a rank-one array.
Return value:
If the arguments are numeric, the return value is a scalar of numeric type,
INTEGER, REAL, or COMPLEX. If the arguments are LOGICAL, the return value is
.TRUE. or .FALSE..
Example:
program test_dot_prod
integer, dimension(3) :: a, b
a = (/ 1, 2, 3 /)
b = (/ 4, 5, 6 /)
print '(3i3)', a
print *
print '(3i3)', b
print *
print *, dot_product(a,b)
end program test_dot_prod

8.88 DPROD — Double product function

Description:
DPROD(X,Y) returns the product X*Y.
Standard : Fortran 77 and later
Class: Elemental function
Syntax : RESULT = DPROD(X, Y)
Arguments:
X The type shall be REAL.
Y The type shall be REAL.
Return value:
The return value is of type REAL(8).
Example:
program test_dprod
real :: x = 5.2
real :: y = 2.3
real(8) :: d
d = dprod(x,y)
print *, d
end program test_dprod

Specific names:
Name Argument Return type Standard
DPROD(X,Y) REAL(4) X, Y REAL(8) Fortran 77 and later
Chapter 8: Intrinsic Procedures 171

8.89 DREAL — Double real part function

Description:
DREAL(Z) returns the real part of complex variable Z.
Standard : GNU extension
Class: Elemental function
Syntax : RESULT = DREAL(A)
Arguments:
A The type shall be COMPLEX(8).
Return value:
The return value is of type REAL(8).
Example:
program test_dreal
complex(8) :: z = (1.3_8,7.2_8)
print *, dreal(z)
end program test_dreal

See also: Section 8.11 [AIMAG], page 117,

8.90 DSHIFTL — Combined left shift

Description:
DSHIFTL(I, J, SHIFT) combines bits of I and J. The rightmost SHIFT bits
of the result are the leftmost SHIFT bits of J, and the remaining bits are the
rightmost bits of I.
Standard : Fortran 2008 and later
Class: Elemental function
Syntax : RESULT = DSHIFTL(I, J, SHIFT)
Arguments:
I Shall be of type INTEGER or a BOZ constant.
J Shall be of type INTEGER or a BOZ constant. If both
I and J have integer type, then they shall have the
same kind type parameter. I and J shall not both be
BOZ constants.
SHIFT Shall be of type INTEGER. It shall be nonnegative. If I
is not a BOZ constant, then SHIFT shall be less than
or equal to BIT_SIZE(I); otherwise, SHIFT shall be
less than or equal to BIT_SIZE(J).
Return value:
If either I or J is a BOZ constant, it is first converted as if by the intrinsic
function INT to an integer type with the kind type parameter of the other.
See also: Section 8.91 [DSHIFTR], page 172,
172 The GNU Fortran Compiler

8.91 DSHIFTR — Combined right shift

Description:
DSHIFTR(I, J, SHIFT) combines bits of I and J. The leftmost SHIFT bits of
the result are the rightmost SHIFT bits of I, and the remaining bits are the
leftmost bits of J.
Standard : Fortran 2008 and later
Class: Elemental function
Syntax : RESULT = DSHIFTR(I, J, SHIFT)
Arguments:
I Shall be of type INTEGER or a BOZ constant.
J Shall be of type INTEGER or a BOZ constant. If both
I and J have integer type, then they shall have the
same kind type parameter. I and J shall not both be
BOZ constants.
SHIFT Shall be of type INTEGER. It shall be nonnegative. If I
is not a BOZ constant, then SHIFT shall be less than
or equal to BIT_SIZE(I); otherwise, SHIFT shall be
less than or equal to BIT_SIZE(J).
Return value:
If either I or J is a BOZ constant, it is first converted as if by the intrinsic
function INT to an integer type with the kind type parameter of the other.
See also: Section 8.90 [DSHIFTL], page 171,

8.92 DTIME — Execution time subroutine (or function)

Description:
DTIME(VALUES, TIME) initially returns the number of seconds of runtime since
the start of the process’s execution in TIME. VALUES returns the user and sys-
tem components of this time in VALUES(1) and VALUES(2) respectively. TIME
is equal to VALUES(1) + VALUES(2).
Subsequent invocations of DTIME return values accumulated since the previous
invocation.
On some systems, the underlying timings are represented using types with
sufficiently small limits that overflows (wrap around) are possible, such as 32-
bit types. Therefore, the values returned by this intrinsic might be, or become,
negative, or numerically less than previous values, during a single run of the
compiled program.
Please note, that this implementation is thread safe if used within OpenMP
directives, i.e., its state will be consistent while called from multiple threads.
However, if DTIME is called from multiple threads, the result is still the time
since the last invocation. This may not give the intended results. If possible,
use CPU_TIME instead.
Chapter 8: Intrinsic Procedures 173

This intrinsic is provided in both subroutine and function forms; however, only
one form can be used in any given program unit.
VALUES and TIME are INTENT(OUT) and provide the following:
VALUES(1): User time in seconds.
VALUES(2): System time in seconds.
TIME: Run time since start in seconds.
Standard : GNU extension
Class: Subroutine, function
Syntax :
CALL DTIME(VALUES, TIME).
TIME = DTIME(VALUES), (not recommended).
Arguments:
VALUES The type shall be REAL(4), DIMENSION(2).
TIME The type shall be REAL(4).
Return value:
Elapsed time in seconds since the last invocation or since the start of program
execution if not called before.
Example:
program test_dtime
integer(8) :: i, j
real, dimension(2) :: tarray
real :: result
call dtime(tarray, result)
print *, result
print *, tarray(1)
print *, tarray(2)
do i=1,100000000 ! Just a delay
j = i * i - i
end do
call dtime(tarray, result)
print *, result
print *, tarray(1)
print *, tarray(2)
end program test_dtime

See also: Section 8.79 [CPU TIME], page 164,

8.93 EOSHIFT — End-off shift elements of an array

Description:
EOSHIFT(ARRAY, SHIFT[, BOUNDARY, DIM]) performs an end-off shift on ele-
ments of ARRAY along the dimension of DIM. If DIM is omitted it is taken
to be 1. DIM is a scalar of type INTEGER in the range of 1 ≤ DIM ≤ n) where
n is the rank of ARRAY. If the rank of ARRAY is one, then all elements of
ARRAY are shifted by SHIFT places. If rank is greater than one, then all
complete rank one sections of ARRAY along the given dimension are shifted.
Elements shifted out one end of each rank one section are dropped. If BOUND-
ARY is present then the corresponding value of from BOUNDARY is copied
174 The GNU Fortran Compiler

back in the other end. If BOUNDARY is not present then the following are
copied in depending on the type of ARRAY.
Array Type Boundary Value
Numeric 0 of the type and kind of ARRAY.
Logical .FALSE..
Character(len)len blanks.
Standard : Fortran 90 and later
Class: Transformational function
Syntax : RESULT = EOSHIFT(ARRAY, SHIFT [, BOUNDARY, DIM])
Arguments:
ARRAY May be any type, not scalar.
SHIFT The type shall be INTEGER.
BOUNDARY Same type as ARRAY.
DIM The type shall be INTEGER.
Return value:
Returns an array of same type and rank as the ARRAY argument.
Example:
program test_eoshift
integer, dimension(3,3) :: a
a = reshape( (/ 1, 2, 3, 4, 5, 6, 7, 8, 9 /), (/ 3, 3 /))
print '(3i3)', a(1,:)
print '(3i3)', a(2,:)
print '(3i3)', a(3,:)
a = EOSHIFT(a, SHIFT=(/1, 2, 1/), BOUNDARY=-5, DIM=2)
print *
print '(3i3)', a(1,:)
print '(3i3)', a(2,:)
print '(3i3)', a(3,:)
end program test_eoshift

8.94 EPSILON — Epsilon function

Description:
EPSILON(X) returns the smallest number E of the same kind as X such that
1 + E > 1.
Standard : Fortran 90 and later
Class: Inquiry function
Syntax : RESULT = EPSILON(X)
Arguments:
X The type shall be REAL.
Return value:
The return value is of same type as the argument.
Example:
program test_epsilon
Chapter 8: Intrinsic Procedures 175

real :: x = 3.143
real(8) :: y = 2.33
print *, EPSILON(x)
print *, EPSILON(y)
end program test_epsilon

8.95 ERF — Error function

Description:
ERF(X) computes the error function of X.
Standard : Fortran 2008 and later
Class: Elemental function
Syntax : RESULT = ERF(X)
Arguments:
X The type shall be REAL.
Return value:
The return value is of type REAL, of the same kind as X and lies in the range
−1 ≤ erf (x) ≤ 1.
Example:
program test_erf
real(8) :: x = 0.17_8
x = erf(x)
end program test_erf

Specific names:
Name Argument Return type Standard
DERF(X) REAL(8) X REAL(8) GNU extension

8.96 ERFC — Error function

Description:
ERFC(X) computes the complementary error function of X.
Standard : Fortran 2008 and later
Class: Elemental function
Syntax : RESULT = ERFC(X)
Arguments:
X The type shall be REAL.
Return value:
The return value is of type REAL and of the same kind as X. It lies in the range
0 ≤ erf c(x) ≤ 2.
Example:
program test_erfc
real(8) :: x = 0.17_8
x = erfc(x)
end program test_erfc
176 The GNU Fortran Compiler

Specific names:
Name Argument Return type Standard
DERFC(X) REAL(8) X REAL(8) GNU extension

8.97 ERFC_SCALED — Error function

Description:
ERFC_SCALED(X) computes the exponentially-scaled complementary error func-
tion of X.
Standard : Fortran 2008 and later
Class: Elemental function
Syntax : RESULT = ERFC_SCALED(X)
Arguments:
X The type shall be REAL.
Return value:
The return value is of type REAL and of the same kind as X.
Example:
program test_erfc_scaled
real(8) :: x = 0.17_8
x = erfc_scaled(x)
end program test_erfc_scaled

8.98 ETIME — Execution time subroutine (or function)

Description:
ETIME(VALUES, TIME) returns the number of seconds of runtime since the start
of the process’s execution in TIME. VALUES returns the user and system
components of this time in VALUES(1) and VALUES(2) respectively. TIME is
equal to VALUES(1) + VALUES(2).
On some systems, the underlying timings are represented using types with
sufficiently small limits that overflows (wrap around) are possible, such as 32-
bit types. Therefore, the values returned by this intrinsic might be, or become,
negative, or numerically less than previous values, during a single run of the
compiled program.
This intrinsic is provided in both subroutine and function forms; however, only
one form can be used in any given program unit.
VALUES and TIME are INTENT(OUT) and provide the following:
VALUES(1): User time in seconds.
VALUES(2): System time in seconds.
TIME: Run time since start in seconds.
Standard : GNU extension
Class: Subroutine, function
Chapter 8: Intrinsic Procedures 177

Syntax :
CALL ETIME(VALUES, TIME).
TIME = ETIME(VALUES), (not recommended).
Arguments:
VALUES The type shall be REAL(4), DIMENSION(2).
TIME The type shall be REAL(4).
Return value:
Elapsed time in seconds since the start of program execution.
Example:
program test_etime
integer(8) :: i, j
real, dimension(2) :: tarray
real :: result
call ETIME(tarray, result)
print *, result
print *, tarray(1)
print *, tarray(2)
do i=1,100000000 ! Just a delay
j = i * i - i
end do
call ETIME(tarray, result)
print *, result
print *, tarray(1)
print *, tarray(2)
end program test_etime

See also: Section 8.79 [CPU TIME], page 164,

8.99 EVENT_QUERY — Query whether a coarray event has

occurred
Description:
EVENT_QUERY assignes the number of events to COUNT which have been posted
to the EVENT variable and not yet been removed by calling EVENT WAIT. When
STAT is present and the invocation was successful, it is assigned the value 0.
If it is present and the invocation has failed, it is assigned a positive value and
COUNT is assigned the value −1.
Standard : TS 18508 or later
Class: subroutine
Syntax : CALL EVENT_QUERY (EVENT, COUNT [, STAT])
Arguments:
EVENT (intent(IN)) Scalar of type EVENT_TYPE, defined in
ISO_FORTRAN_ENV; shall not be coindexed.
COUNT (intent(out))Scalar integer with at least the precision
of default integer.
STAT (optional) Scalar default-kind integer variable.
178 The GNU Fortran Compiler

Example:
program atomic
use iso_fortran_env
implicit none
type(event_type) :: event_value_has_been_set[*]
integer :: cnt
if (this_image() == 1) then
call event_query (event_value_has_been_set, cnt)
if (cnt > 0) write(*,*) "Value has been set"
elseif (this_image() == 2) then
event post (event_value_has_been_set[1])
end if
end program atomic

8.100 EXECUTE_COMMAND_LINE — Execute a shell command

Description:
EXECUTE_COMMAND_LINE runs a shell command, synchronously or
asynchronously.
The COMMAND argument is passed to the shell and executed (The shell is sh on
Unix systems, and cmd.exe on Windows.). If WAIT is present and has the value
false, the execution of the command is asynchronous if the system supports it;
otherwise, the command is executed synchronously using the C library’s system
call.
The three last arguments allow the user to get status information. After syn-
chronous execution, EXITSTAT contains the integer exit code of the command,
as returned by system. CMDSTAT is set to zero if the command line was executed
(whatever its exit status was). CMDMSG is assigned an error message if an error
has occurred.
Note that the system function need not be thread-safe. It is the responsibility
of the user to ensure that system is not called concurrently.
For asynchronous execution on supported targets, the POSIX posix_spawn
or fork functions are used. Also, a signal handler for the SIGCHLD signal is
installed.
Standard : Fortran 2008 and later
Class: Subroutine
Syntax : CALL EXECUTE_COMMAND_LINE(COMMAND [, WAIT, EXITSTAT, CMDSTAT,
CMDMSG ])
Arguments:
COMMAND Shall be a default CHARACTER scalar.
WAIT (Optional) Shall be a default LOGICAL scalar.
EXITSTAT (Optional) Shall be an INTEGER of the default kind.
CMDSTAT (Optional) Shall be an INTEGER of the default kind.
CMDMSG (Optional) Shall be an CHARACTER scalar of the de-
fault kind.
Example:
program test_exec
Chapter 8: Intrinsic Procedures 179

integer :: i

call execute_command_line ("external_prog.exe", exitstat=i)

print *, "Exit status of external_prog.exe was ", i

call execute_command_line ("reindex_files.exe", wait=.false.)

print *, "Now reindexing files in the background"

end program test_exec

Note:
Because this intrinsic is implemented in terms of the system function call, its
behavior with respect to signaling is processor dependent. In particular, on
POSIX-compliant systems, the SIGINT and SIGQUIT signals will be ignored,
and the SIGCHLD will be blocked. As such, if the parent process is terminated,
the child process might not be terminated alongside.
See also: Section 8.263 [SYSTEM], page 278,

8.101 EXIT — Exit the program with status.

Description:
EXIT causes immediate termination of the program with status. If status is
omitted it returns the canonical success for the system. All Fortran I/O units
are closed.
Standard : GNU extension
Class: Subroutine
Syntax : CALL EXIT([STATUS])
Arguments:
STATUS Shall be an INTEGER of the default kind.
Return value:
STATUS is passed to the parent process on exit.
Example:
program test_exit
integer :: STATUS = 0
print *, 'This program is going to exit.'
call EXIT(STATUS)
end program test_exit

See also: Section 8.2 [ABORT], page 111,

Section 8.163 [KILL], page 218,

8.102 EXP — Exponential function

Description:
EXP(X) computes the base e exponential of X.
Standard : Fortran 77 and later, has overloads that are GNU extensions
Class: Elemental function
180 The GNU Fortran Compiler

Syntax : RESULT = EXP(X)

Arguments:
X The type shall be REAL or COMPLEX.
Return value:
The return value has same type and kind as X.
Example:
program test_exp
real :: x = 1.0
x = exp(x)
end program test_exp

Specific names:
Name Argument Return type Standard
EXP(X) REAL(4) X REAL(4) Fortran 77 and later
DEXP(X) REAL(8) X REAL(8) Fortran 77 and later
CEXP(X) COMPLEX(4) X COMPLEX(4) Fortran 77 and later
ZEXP(X) COMPLEX(8) X COMPLEX(8) GNU extension
CDEXP(X) COMPLEX(8) X COMPLEX(8) GNU extension

8.103 EXPONENT — Exponent function

Description:
EXPONENT(X) returns the value of the exponent part of X. If X is zero the value
returned is zero.
Standard : Fortran 90 and later
Class: Elemental function
Syntax : RESULT = EXPONENT(X)
Arguments:
X The type shall be REAL.
Return value:
The return value is of type default INTEGER.
Example:
program test_exponent
real :: x = 1.0
integer :: i
i = exponent(x)
print *, i
print *, exponent(0.0)
end program test_exponent

8.104 EXTENDS_TYPE_OF — Query dynamic type for extension

Description:
Query dynamic type for extension.
Standard : Fortran 2003 and later
Chapter 8: Intrinsic Procedures 181

Class: Inquiry function

Syntax : RESULT = EXTENDS_TYPE_OF(A, MOLD)
Arguments:
A Shall be an object of extensible declared type or un-
limited polymorphic.
MOLD Shall be an object of extensible declared type or un-
limited polymorphic.
Return value:
The return value is a scalar of type default logical. It is true if and only if the
dynamic type of A is an extension type of the dynamic type of MOLD.
See also: Section 8.234 [SAME TYPE AS], page 260,

8.105 FDATE — Get the current time as a string

Description:
FDATE(DATE) returns the current date (using the same format as Section 8.81
[CTIME], page 165) in DATE. It is equivalent to CALL CTIME(DATE, TIME()).
This intrinsic is provided in both subroutine and function forms; however, only
one form can be used in any given program unit.
Standard : GNU extension
Class: Subroutine, function
Syntax :
CALL FDATE(DATE).
DATE = FDATE().
Arguments:
DATE The type shall be of type CHARACTER of the default
kind. It is an INTENT(OUT) argument. If the length
of this variable is too short for the date and time
string to fit completely, it will be blank on procedure
return.
Return value:
The current date and time as a string.
Example:
program test_fdate
integer(8) :: i, j
character(len=30) :: date
call fdate(date)
print *, 'Program started on ', date
do i = 1, 100000000 ! Just a delay
j = i * i - i
end do
call fdate(date)
print *, 'Program ended on ', date
end program test_fdate
182 The GNU Fortran Compiler

See also: Section 8.82 [DATE AND TIME], page 166,

Section 8.81 [CTIME], page 165,

8.106 FGET — Read a single character in stream mode from

stdin
Description:
Read a single character in stream mode from stdin by bypassing normal for-
matted output. Stream I/O should not be mixed with normal record-oriented
(formatted or unformatted) I/O on the same unit; the results are unpredictable.
This intrinsic is provided in both subroutine and function forms; however, only
one form can be used in any given program unit.
Note that the FGET intrinsic is provided for backwards compatibility with g77.
GNU Fortran provides the Fortran 2003 Stream facility. Programmers should
consider the use of new stream IO feature in new code for future portability.
See also Section 1.3.2 [Fortran 2003 status], page 3.

Standard : GNU extension

Class: Subroutine, function

Syntax :
CALL FGET(C [, STATUS])
STATUS = FGET(C)

Arguments:
C The type shall be CHARACTER and of default kind.
STATUS (Optional) status flag of type INTEGER. Returns 0
on success, -1 on end-of-file, and a system specific
positive error code otherwise.

Example:
PROGRAM test_fget
INTEGER, PARAMETER :: strlen = 100
INTEGER :: status, i = 1
CHARACTER(len=strlen) :: str = ""

WRITE (*,*) 'Enter text:'

DO
CALL fget(str(i:i), status)
if (status /= 0 .OR. i > strlen) exit
i = i + 1
END DO
WRITE (*,*) TRIM(str)
END PROGRAM

See also: Section 8.107 [FGETC], page 183,

Section 8.112 [FPUT], page 186,
Section 8.113 [FPUTC], page 187,
Chapter 8: Intrinsic Procedures 183

8.107 FGETC — Read a single character in stream mode

Description:
Read a single character in stream mode by bypassing normal formatted output.
Stream I/O should not be mixed with normal record-oriented (formatted or
unformatted) I/O on the same unit; the results are unpredictable.
This intrinsic is provided in both subroutine and function forms; however, only
one form can be used in any given program unit.
Note that the FGET intrinsic is provided for backwards compatibility with g77.
GNU Fortran provides the Fortran 2003 Stream facility. Programmers should
consider the use of new stream IO feature in new code for future portability.
See also Section 1.3.2 [Fortran 2003 status], page 3.
Standard : GNU extension
Class: Subroutine, function
Syntax :
CALL FGETC(UNIT, C [, STATUS])
STATUS = FGETC(UNIT, C)
Arguments:
UNIT The type shall be INTEGER.
C The type shall be CHARACTER and of default kind.
STATUS (Optional) status flag of type INTEGER. Returns 0 on
success, -1 on end-of-file and a system specific posi-
tive error code otherwise.
Example:
PROGRAM test_fgetc
INTEGER :: fd = 42, status
CHARACTER :: c

OPEN(UNIT=fd, FILE="/etc/passwd", ACTION="READ", STATUS = "OLD")

DO
CALL fgetc(fd, c, status)
IF (status /= 0) EXIT
call fput(c)
END DO
CLOSE(UNIT=fd)
END PROGRAM

See also: Section 8.106 [FGET], page 182,

Section 8.112 [FPUT], page 186,
Section 8.113 [FPUTC], page 187,

8.108 FINDLOC — Search an array for a value

Description:
Determines the location of the element in the array with the value given in
the VALUE argument, or, if the DIM argument is supplied, determines the
locations of the elements equal to the VALUE argument element along each
184 The GNU Fortran Compiler

row of the array in the DIM direction. If MASK is present, only the elements
for which MASK is .TRUE. are considered. If more than one element in the
array has the value VALUE, the location returned is that of the first such
element in array element order if the BACK is not present or if it is .FALSE..
If BACK is true, the location returned is that of the last such element. If the
array has zero size, or all of the elements of MASK are .FALSE., then the result
is an array of zeroes. Similarly, if DIM is supplied and all of the elements of
MASK along a given row are zero, the result value for that row is zero.
Standard : Fortran 2008 and later.
Class: Transformational function
Syntax :
RESULT = FINDLOC(ARRAY, VALUE, DIM [, MASK] [,KIND]
[,BACK])
RESULT = FINDLOC(ARRAY, VALUE, [, MASK] [,KIND]
[,BACK])
Arguments:
ARRAY Shall be an array of intrinsic type.
VALUE A scalar of intrinsic type which is in type confor-
mance with ARRAY.
DIM (Optional) Shall be a scalar of type INTEGER, with a
value between one and the rank of ARRAY, inclusive.
It may not be an optional dummy argument.
MASK (Optional) Shall be of type LOGICAL, and con-
formable with ARRAY.
KIND (Optional) A scalar INTEGER constant expression in-
dicating the kind parameter of the result.
BACK (Optional) A scalar of type LOGICAL.
Return value:
If DIM is absent, the result is a rank-one array with a length equal to the rank
of ARRAY. If DIM is present, the result is an array with a rank one less than
the rank of ARRAY, and a size corresponding to the size of ARRAY with the
DIM dimension removed. If DIM is present and ARRAY has a rank of one,
the result is a scalar. If the optional argument KIND is present, the result is
an integer of kind KIND, otherwise it is of default kind.
See also: Section 8.190 [MAXLOC], page 234,
Section 8.198 [MINLOC], page 238,

8.109 FLOOR — Integer floor function

Description:
FLOOR(A) returns the greatest integer less than or equal to A.
Standard : Fortran 95 and later
Class: Elemental function
Chapter 8: Intrinsic Procedures 185

Syntax : RESULT = FLOOR(A [, KIND])

Arguments:
A The type shall be REAL.
KIND (Optional) A scalar INTEGER constant expression in-
dicating the kind parameter of the result.
Return value:
The return value is of type INTEGER(KIND) if KIND is present and of default-
kind INTEGER otherwise.
Example:
program test_floor
real :: x = 63.29
real :: y = -63.59
print *, floor(x) ! returns 63
print *, floor(y) ! returns -64
end program test_floor

See also: Section 8.58 [CEILING], page 149,

Section 8.206 [NINT], page 244,

8.110 FLUSH — Flush I/O unit(s)

Description:
Flushes Fortran unit(s) currently open for output. Without the optional argu-
ment, all units are flushed, otherwise just the unit specified.
Standard : GNU extension
Class: Subroutine
Syntax : CALL FLUSH(UNIT)
Arguments:
UNIT (Optional) The type shall be INTEGER.
Note: Beginning with the Fortran 2003 standard, there is a FLUSH statement that
should be preferred over the FLUSH intrinsic.
The FLUSH intrinsic and the Fortran 2003 FLUSH statement have identical effect:
they flush the runtime library’s I/O buffer so that the data becomes visible to
other processes. This does not guarantee that the data is committed to disk.
On POSIX systems, you can request that all data is transferred to the storage
device by calling the fsync function, with the POSIX file descriptor of the I/O
unit as argument (retrieved with GNU intrinsic FNUM). The following example
shows how:
! Declare the interface for POSIX fsync function
interface
function fsync (fd) bind(c,name="fsync")
use iso_c_binding, only: c_int
integer(c_int), value :: fd
integer(c_int) :: fsync
end function fsync
end interface
186 The GNU Fortran Compiler

! Variable declaration
integer :: ret

! Opening unit 10
open (10,file="foo")

! ...
! Perform I/O on unit 10
! ...

! Flush and sync

flush(10)
ret = fsync(fnum(10))

! Handle possible error

if (ret /= 0) stop "Error calling FSYNC"

8.111 FNUM — File number function

Description:
FNUM(UNIT) returns the POSIX file descriptor number corresponding to the
open Fortran I/O unit UNIT.
Standard : GNU extension
Class: Function
Syntax : RESULT = FNUM(UNIT)
Arguments:
UNIT The type shall be INTEGER.
Return value:
The return value is of type INTEGER
Example:
program test_fnum
integer :: i
open (unit=10, status = "scratch")
i = fnum(10)
print *, i
close (10)
end program test_fnum

8.112 FPUT — Write a single character in stream mode to

stdout
Description:
Write a single character in stream mode to stdout by bypassing normal for-
matted output. Stream I/O should not be mixed with normal record-oriented
(formatted or unformatted) I/O on the same unit; the results are unpredictable.
This intrinsic is provided in both subroutine and function forms; however, only
one form can be used in any given program unit.
Chapter 8: Intrinsic Procedures 187

Note that the FGET intrinsic is provided for backwards compatibility with g77.
GNU Fortran provides the Fortran 2003 Stream facility. Programmers should
consider the use of new stream IO feature in new code for future portability.
See also Section 1.3.2 [Fortran 2003 status], page 3.
Standard : GNU extension
Class: Subroutine, function
Syntax :
CALL FPUT(C [, STATUS])
STATUS = FPUT(C)
Arguments:
C The type shall be CHARACTER and of default kind.
STATUS (Optional) status flag of type INTEGER. Returns 0 on
success, -1 on end-of-file and a system specific posi-
tive error code otherwise.
Example:
PROGRAM test_fput
CHARACTER(len=10) :: str = "gfortran"
INTEGER :: i
DO i = 1, len_trim(str)
CALL fput(str(i:i))
END DO
END PROGRAM

See also: Section 8.113 [FPUTC], page 187,

Section 8.106 [FGET], page 182,
Section 8.107 [FGETC], page 183,

8.113 FPUTC — Write a single character in stream mode

Description:
Write a single character in stream mode by bypassing normal formatted output.
Stream I/O should not be mixed with normal record-oriented (formatted or
unformatted) I/O on the same unit; the results are unpredictable.
This intrinsic is provided in both subroutine and function forms; however, only
one form can be used in any given program unit.
Note that the FGET intrinsic is provided for backwards compatibility with g77.
GNU Fortran provides the Fortran 2003 Stream facility. Programmers should
consider the use of new stream IO feature in new code for future portability.
See also Section 1.3.2 [Fortran 2003 status], page 3.
Standard : GNU extension
Class: Subroutine, function
Syntax :
CALL FPUTC(UNIT, C [, STATUS])
STATUS = FPUTC(UNIT, C)
188 The GNU Fortran Compiler

Arguments:
UNIT The type shall be INTEGER.
C The type shall be CHARACTER and of default kind.
STATUS (Optional) status flag of type INTEGER. Returns 0 on
success, -1 on end-of-file and a system specific posi-
tive error code otherwise.
Example:
PROGRAM test_fputc
CHARACTER(len=10) :: str = "gfortran"
INTEGER :: fd = 42, i

OPEN(UNIT = fd, FILE = "out", ACTION = "WRITE", STATUS="NEW")

DO i = 1, len_trim(str)
CALL fputc(fd, str(i:i))
END DO
CLOSE(fd)
END PROGRAM

See also: Section 8.112 [FPUT], page 186,

Section 8.106 [FGET], page 182,
Section 8.107 [FGETC], page 183,

8.114 FRACTION — Fractional part of the model

representation
Description:
FRACTION(X) returns the fractional part of the model representation of X.
Standard : Fortran 90 and later
Class: Elemental function
Syntax : Y = FRACTION(X)
Arguments:
X The type of the argument shall be a REAL.
Return value:
The return value is of the same type and kind as the argument. The frac-
tional part of the model representation of X is returned; it is X * RADIX(X)**(-
EXPONENT(X)).
Example:
program test_fraction
real :: x
x = 178.1387e-4
print *, fraction(x), x * radix(x)**(-exponent(x))
end program test_fraction

8.115 FREE — Frees memory

Description:
Frees memory previously allocated by MALLOC. The FREE intrinsic is an exten-
sion intended to be used with Cray pointers, and is provided in GNU Fortran
Chapter 8: Intrinsic Procedures 189

to allow user to compile legacy code. For new code using Fortran 95 pointers,
the memory de-allocation intrinsic is DEALLOCATE.
Standard : GNU extension
Class: Subroutine
Syntax : CALL FREE(PTR)
Arguments:
PTR The type shall be INTEGER. It represents the location
of the memory that should be de-allocated.
Return value:
None
Example: See MALLOC for an example.
See also: Section 8.184 [MALLOC], page 230,

8.116 FSEEK — Low level file positioning subroutine

Description:
Moves UNIT to the specified OFFSET. If WHENCE is set to 0, the OFFSET
is taken as an absolute value SEEK_SET, if set to 1, OFFSET is taken to be
relative to the current position SEEK_CUR, and if set to 2 relative to the end of
the file SEEK_END. On error, STATUS is set to a nonzero value. If STATUS
the seek fails silently.
This intrinsic routine is not fully backwards compatible with g77. In g77, the
FSEEK takes a statement label instead of a STATUS variable. If FSEEK is used
in old code, change
CALL FSEEK(UNIT, OFFSET, WHENCE, *label)
to
INTEGER :: status
CALL FSEEK(UNIT, OFFSET, WHENCE, status)
IF (status /= 0) GOTO label
Please note that GNU Fortran provides the Fortran 2003 Stream facility. Pro-
grammers should consider the use of new stream IO feature in new code for
future portability. See also Section 1.3.2 [Fortran 2003 status], page 3.
Standard : GNU extension
Class: Subroutine
Syntax : CALL FSEEK(UNIT, OFFSET, WHENCE[, STATUS])
Arguments:
UNIT Shall be a scalar of type INTEGER.
OFFSET Shall be a scalar of type INTEGER.
WHENCE Shall be a scalar of type INTEGER. Its value shall be
either 0, 1 or 2.
STATUS (Optional) shall be a scalar of type INTEGER(4).
190 The GNU Fortran Compiler

Example:
PROGRAM test_fseek
INTEGER, PARAMETER :: SEEK_SET = 0, SEEK_CUR = 1, SEEK_END = 2
INTEGER :: fd, offset, ierr

ierr = 0
offset = 5
fd = 10

OPEN(UNIT=fd, FILE="fseek.test")
CALL FSEEK(fd, offset, SEEK_SET, ierr) ! move to OFFSET
print *, FTELL(fd), ierr

CALL FSEEK(fd, 0, SEEK_END, ierr) ! move to end

print *, FTELL(fd), ierr

CALL FSEEK(fd, 0, SEEK_SET, ierr) ! move to beginning

print *, FTELL(fd), ierr

CLOSE(UNIT=fd)
END PROGRAM

See also: Section 8.118 [FTELL], page 191,

8.117 FSTAT — Get file status

Description:
FSTAT is identical to Section 8.259 [STAT], page 275, except that information
about an already opened file is obtained.
The elements in VALUES are the same as described by Section 8.259 [STAT],
page 275.
This intrinsic is provided in both subroutine and function forms; however, only
one form can be used in any given program unit.
Standard : GNU extension
Class: Subroutine, function
Syntax :
CALL FSTAT(UNIT, VALUES [, STATUS])
STATUS = FSTAT(UNIT, VALUES)
Arguments:
UNIT An open I/O unit number of type INTEGER.
VALUES The type shall be INTEGER(4), DIMENSION(13).
STATUS (Optional) status flag of type INTEGER(4). Returns 0
on success and a system specific error code otherwise.
Example: See Section 8.259 [STAT], page 275, for an example.
See also: To stat a link:
Section 8.182 [LSTAT], page 229,
To stat a file:
Section 8.259 [STAT], page 275,
Chapter 8: Intrinsic Procedures 191

8.118 FTELL — Current stream position

Description:
Retrieves the current position within an open file.
This intrinsic is provided in both subroutine and function forms; however, only
one form can be used in any given program unit.
Standard : GNU extension
Class: Subroutine, function
Syntax :
CALL FTELL(UNIT, OFFSET)
OFFSET = FTELL(UNIT)
Arguments:
OFFSET Shall of type INTEGER.
UNIT Shall of type INTEGER.
Return value:
In either syntax, OFFSET is set to the current offset of unit number UNIT, or
to −1 if the unit is not currently open.
Example:
PROGRAM test_ftell
INTEGER :: i
OPEN(10, FILE="temp.dat")
CALL ftell(10,i)
WRITE(*,*) i
END PROGRAM

See also: Section 8.116 [FSEEK], page 189,

8.119 GAMMA — Gamma function

Description:
GAMMA(X) computes Gamma (Γ) of X. For positive, integer values of X the
Gamma function simplifies to the factorial function Γ(x) = (x − 1)!.
Z ∞
Γ(x) = tx−1 e−t dt
0

Standard : Fortran 2008 and later

Class: Elemental function
Syntax : X = GAMMA(X)
Arguments:
X Shall be of type REAL and neither zero nor a negative
integer.
Return value:
The return value is of type REAL of the same kind as X.
192 The GNU Fortran Compiler

Example:
program test_gamma
real :: x = 1.0
x = gamma(x) ! returns 1.0
end program test_gamma

Specific names:
Name Argument Return type Standard
DGAMMA(X) REAL(8) X REAL(8) GNU extension
See also: Logarithm of the Gamma function:
Section 8.179 [LOG GAMMA], page 227,

8.120 GERROR — Get last system error message

Description:
Returns the system error message corresponding to the last system error. This
resembles the functionality of strerror(3) in C.
Standard : GNU extension
Class: Subroutine
Syntax : CALL GERROR(RESULT)
Arguments:
RESULT Shall be of type CHARACTER and of default kind.
Example:
PROGRAM test_gerror
CHARACTER(len=100) :: msg
CALL gerror(msg)
WRITE(*,*) msg
END PROGRAM

See also: Section 8.146 [IERRNO], page 209,

Section 8.214 [PERROR], page 249,

8.121 GETARG — Get command line arguments

Description:
Retrieve the POS-th argument that was passed on the command line when the
containing program was invoked.
This intrinsic routine is provided for backwards compatibility with GNU Fortran
77. In new code, programmers should consider the use of the Section 8.123
[GET COMMAND ARGUMENT], page 194, intrinsic defined by the Fortran
2003 standard.
Standard : GNU extension
Class: Subroutine
Syntax : CALL GETARG(POS, VALUE)
Chapter 8: Intrinsic Procedures 193

Arguments:
POS Shall be of type INTEGER and not wider than the
default integer kind; POS ≥ 0
VALUE Shall be of type CHARACTER and of default kind.
Return value:
After GETARG returns, the VALUE argument holds the POSth command line
argument. If VALUE cannot hold the argument, it is truncated to fit the length
of VALUE. If there are less than POS arguments specified at the command
line, VALUE will be filled with blanks. If POS = 0, VALUE is set to the name
of the program (on systems that support this feature).
Example:
PROGRAM test_getarg
INTEGER :: i
CHARACTER(len=32) :: arg

DO i = 1, iargc()
CALL getarg(i, arg)
WRITE (*,*) arg
END DO
END PROGRAM

See also: GNU Fortran 77 compatibility function:

Section 8.139 [IARGC], page 204,
Fortran 2003 functions and subroutines:
Section 8.122 [GET COMMAND], page 193,
Section 8.123 [GET COMMAND ARGUMENT], page 194,
Section 8.68 [COMMAND ARGUMENT COUNT], page 157,

8.122 GET_COMMAND — Get the entire command line

Description:
Retrieve the entire command line that was used to invoke the program.
Standard : Fortran 2003 and later
Class: Subroutine
Syntax : CALL GET_COMMAND([COMMAND, LENGTH, STATUS])
Arguments:
COMMAND (Optional) shall be of type CHARACTER and of default
kind.
LENGTH (Optional) Shall be of type INTEGER and of default
kind.
STATUS (Optional) Shall be of type INTEGER and of default
kind.
Return value:
If COMMAND is present, stores the entire command line that was used to
invoke the program in COMMAND. If LENGTH is present, it is assigned the
length of the command line. If STATUS is present, it is assigned 0 upon success
194 The GNU Fortran Compiler

of the command, -1 if COMMAND is too short to store the command line, or

a positive value in case of an error.
Example:
PROGRAM test_get_command
CHARACTER(len=255) :: cmd
CALL get_command(cmd)
WRITE (*,*) TRIM(cmd)
END PROGRAM

See also: Section 8.123 [GET COMMAND ARGUMENT], page 194,

Section 8.68 [COMMAND ARGUMENT COUNT], page 157,

8.123 GET_COMMAND_ARGUMENT — Get command line arguments

Description:
Retrieve the NUMBER-th argument that was passed on the command line when
the containing program was invoked.
Standard : Fortran 2003 and later
Class: Subroutine
Syntax : CALL GET_COMMAND_ARGUMENT(NUMBER [, VALUE, LENGTH, STATUS])
Arguments:
NUMBER Shall be a scalar of type INTEGER and of default kind,
NUMBER ≥ 0
VALUE (Optional) Shall be a scalar of type CHARACTER and
of default kind.
LENGTH (Optional) Shall be a scalar of type INTEGER and of
default kind.
STATUS (Optional) Shall be a scalar of type INTEGER and of
default kind.
Return value:
After GET_COMMAND_ARGUMENT returns, the VALUE argument holds the NUM-
BER-th command line argument. If VALUE cannot hold the argument, it is
truncated to fit the length of VALUE. If there are less than NUMBER ar-
guments specified at the command line, VALUE will be filled with blanks. If
NUMBER = 0, VALUE is set to the name of the program (on systems that
support this feature). The LENGTH argument contains the length of the NUM-
BER-th command line argument. If the argument retrieval fails, STATUS is
a positive number; if VALUE contains a truncated command line argument,
STATUS is -1; and otherwise the STATUS is zero.
Example:
PROGRAM test_get_command_argument
INTEGER :: i
CHARACTER(len=32) :: arg

i = 0
DO
Chapter 8: Intrinsic Procedures 195

CALL get_command_argument(i, arg)

IF (LEN_TRIM(arg) == 0) EXIT

WRITE (*,*) TRIM(arg)

i = i+1
END DO
END PROGRAM

See also: Section 8.122 [GET COMMAND], page 193,

Section 8.68 [COMMAND ARGUMENT COUNT], page 157,

8.124 GETCWD — Get current working directory

Description:
Get current working directory.
This intrinsic is provided in both subroutine and function forms; however, only
one form can be used in any given program unit.
Standard : GNU extension
Class: Subroutine, function
Syntax :
CALL GETCWD(C [, STATUS])
STATUS = GETCWD(C)
Arguments:
C The type shall be CHARACTER and of default kind.
STATUS (Optional) status flag. Returns 0 on success, a system
specific and nonzero error code otherwise.
Example:
PROGRAM test_getcwd
CHARACTER(len=255) :: cwd
CALL getcwd(cwd)
WRITE(*,*) TRIM(cwd)
END PROGRAM

See also: Section 8.60 [CHDIR], page 150,

8.125 GETENV — Get an environmental variable

Description:
Get the VALUE of the environmental variable NAME.
This intrinsic routine is provided for backwards compatibility with GNU For-
tran 77. In new code, programmers should consider the use of the Section 8.126
[GET ENVIRONMENT VARIABLE], page 196, intrinsic defined by the For-
tran 2003 standard.
Note that GETENV need not be thread-safe. It is the responsibility of the user
to ensure that the environment is not being updated concurrently with a call
to the GETENV intrinsic.
Standard : GNU extension
196 The GNU Fortran Compiler

Class: Subroutine
Syntax : CALL GETENV(NAME, VALUE)
Arguments:
NAME Shall be of type CHARACTER and of default kind.
VALUE Shall be of type CHARACTER and of default kind.
Return value:
Stores the value of NAME in VALUE. If VALUE is not large enough to hold
the data, it is truncated. If NAME is not set, VALUE will be filled with blanks.
Example:
PROGRAM test_getenv
CHARACTER(len=255) :: homedir
CALL getenv("HOME", homedir)
WRITE (*,*) TRIM(homedir)
END PROGRAM

See also: Section 8.126 [GET ENVIRONMENT VARIABLE], page 196,

8.126 GET_ENVIRONMENT_VARIABLE — Get an environmental

variable
Description:
Get the VALUE of the environmental variable NAME.
Note that GET_ENVIRONMENT_VARIABLE need not be thread-safe. It is the re-
sponsibility of the user to ensure that the environment is not being updated
concurrently with a call to the GET_ENVIRONMENT_VARIABLE intrinsic.
Standard : Fortran 2003 and later
Class: Subroutine
Syntax : CALL GET_ENVIRONMENT_VARIABLE(NAME[, VALUE, LENGTH, STATUS,
TRIM_NAME)
Arguments:
NAME Shall be a scalar of type CHARACTER and of default
kind.
VALUE (Optional) Shall be a scalar of type CHARACTER and
of default kind.
LENGTH (Optional) Shall be a scalar of type INTEGER and of
default kind.
STATUS (Optional) Shall be a scalar of type INTEGER and of
default kind.
TRIM NAME(Optional) Shall be a scalar of type LOGICAL and of
default kind.
Return value:
Stores the value of NAME in VALUE. If VALUE is not large enough to hold the
data, it is truncated. If NAME is not set, VALUE will be filled with blanks.
Argument LENGTH contains the length needed for storing the environment
Chapter 8: Intrinsic Procedures 197

variable NAME or zero if it is not present. STATUS is -1 if VALUE is present

but too short for the environment variable; it is 1 if the environment variable
does not exist and 2 if the processor does not support environment variables;
in all other cases STATUS is zero. If TRIM NAME is present with the value
.FALSE., the trailing blanks in NAME are significant; otherwise they are not
part of the environment variable name.
Example:
PROGRAM test_getenv
CHARACTER(len=255) :: homedir
CALL get_environment_variable("HOME", homedir)
WRITE (*,*) TRIM(homedir)
END PROGRAM

8.127 GETGID — Group ID function

Description:
Returns the numerical group ID of the current process.
Standard : GNU extension
Class: Function
Syntax : RESULT = GETGID()
Return value:
The return value of GETGID is an INTEGER of the default kind.
Example: See GETPID for an example.
See also: Section 8.129 [GETPID], page 198,
Section 8.130 [GETUID], page 198,

8.128 GETLOG — Get login name

Description:
Gets the username under which the program is running.
Standard : GNU extension
Class: Subroutine
Syntax : CALL GETLOG(C)
Arguments:
C Shall be of type CHARACTER and of default kind.
Return value:
Stores the current user name in C. (On systems where POSIX functions
geteuid and getpwuid are not available, and the getlogin function is not
implemented either, this will return a blank string.)
Example:
PROGRAM TEST_GETLOG
CHARACTER(32) :: login
198 The GNU Fortran Compiler

CALL GETLOG(login)
WRITE(*,*) login
END PROGRAM

See also: Section 8.130 [GETUID], page 198,

8.129 GETPID — Process ID function

Description:
Returns the numerical process identifier of the current process.
Standard : GNU extension
Class: Function
Syntax : RESULT = GETPID()
Return value:
The return value of GETPID is an INTEGER of the default kind.
Example:
program info
print *, "The current process ID is ", getpid()
print *, "Your numerical user ID is ", getuid()
print *, "Your numerical group ID is ", getgid()
end program info

See also: Section 8.127 [GETGID], page 197,

Section 8.130 [GETUID], page 198,

8.130 GETUID — User ID function

Description:
Returns the numerical user ID of the current process.
Standard : GNU extension
Class: Function
Syntax : RESULT = GETUID()
Return value:
The return value of GETUID is an INTEGER of the default kind.
Example: See GETPID for an example.
See also: Section 8.129 [GETPID], page 198,
Section 8.128 [GETLOG], page 197,

8.131 GMTIME — Convert time to GMT info

Description:
Given a system time value TIME (as provided by the Section 8.269 [TIME],
page 282, intrinsic), fills VALUES with values extracted from it appropriate to
the UTC time zone (Universal Coordinated Time, also known in some countries
as GMT, Greenwich Mean Time), using gmtime(3).
Chapter 8: Intrinsic Procedures 199

This intrinsic routine is provided for backwards compatibility with GNU Fortran
77. In new code, programmers should consider the use of the Section 8.82
[DATE AND TIME], page 166, intrinsic defined by the Fortran 95 standard.
Standard : GNU extension
Class: Subroutine
Syntax : CALL GMTIME(TIME, VALUES)
Arguments:
TIME An INTEGER scalar expression corresponding to a sys-
tem time, with INTENT(IN).
VALUES A default INTEGER array with 9 elements, with
INTENT(OUT).
Return value:
The elements of VALUES are assigned as follows:
1. Seconds after the minute, range 0–59 or 0–61 to allow for leap seconds
2. Minutes after the hour, range 0–59
3. Hours past midnight, range 0–23
4. Day of month, range 1–31
5. Number of months since January, range 0–11
6. Years since 1900
7. Number of days since Sunday, range 0–6
8. Days since January 1, range 0–365
9. Daylight savings indicator: positive if daylight savings is in effect, zero if
not, and negative if the information is not available.
See also: Section 8.82 [DATE AND TIME], page 166,
Section 8.81 [CTIME], page 165,
Section 8.183 [LTIME], page 230,
Section 8.269 [TIME], page 282,
Section 8.270 [TIME8], page 283,

8.132 HOSTNM — Get system host name

Description:
Retrieves the host name of the system on which the program is running.
This intrinsic is provided in both subroutine and function forms; however, only
one form can be used in any given program unit.
Standard : GNU extension
Class: Subroutine, function
Syntax :
CALL HOSTNM(C [, STATUS])
STATUS = HOSTNM(NAME)
200 The GNU Fortran Compiler

Arguments:
C Shall of type CHARACTER and of default kind.
STATUS (Optional) status flag of type INTEGER. Returns 0 on
success, or a system specific error code otherwise.
Return value:
In either syntax, NAME is set to the current hostname if it can be obtained,
or to a blank string otherwise.

8.133 HUGE — Largest number of a kind

Description:
HUGE(X) returns the largest number that is not an infinity in the model of the
type of X.
Standard : Fortran 90 and later
Class: Inquiry function
Syntax : RESULT = HUGE(X)
Arguments:
X Shall be of type REAL or INTEGER.
Return value:
The return value is of the same type and kind as X
Example:
program test_huge_tiny
print *, huge(0), huge(0.0), huge(0.0d0)
print *, tiny(0.0), tiny(0.0d0)
end program test_huge_tiny

8.134 HYPOT — Euclidean distance function

Description: √
HYPOT(X,Y) is the Euclidean distance function. It is equal to X 2 + Y 2 , with-
out undue underflow or overflow.
Standard : Fortran 2008 and later
Class: Elemental function
Syntax : RESULT = HYPOT(X, Y)
Arguments:
X The type shall be REAL.
Y The type and kind type parameter shall be the same
as X.
Return value:
The return value has the same type and kind type parameter as X.
Example:
program test_hypot
real(4) :: x = 1.e0_4, y = 0.5e0_4
x = hypot(x,y)
end program test_hypot
Chapter 8: Intrinsic Procedures 201

8.135 IACHAR — Code in ASCII collating sequence

Description:
IACHAR(C) returns the code for the ASCII character in the first character posi-
tion of C.
Standard : Fortran 95 and later, with KIND argument Fortran 2003 and later
Class: Elemental function
Syntax : RESULT = IACHAR(C [, KIND])
Arguments:
C Shall be a scalar CHARACTER, with INTENT(IN)
KIND (Optional) A scalar INTEGER constant expression in-
dicating the kind parameter of the result.
Return value:
The return value is of type INTEGER and of kind KIND. If KIND is absent, the
return value is of default integer kind.
Example:
program test_iachar
integer i
i = iachar(' ')
end program test_iachar

Note: See Section 8.143 [ICHAR], page 206, for a discussion of converting between
numerical values and formatted string representations.
See also: Section 8.5 [ACHAR], page 113,
Section 8.59 [CHAR], page 149,
Section 8.143 [ICHAR], page 206,

8.136 IALL — Bitwise AND of array elements

Description:
Reduces with bitwise AND the elements of ARRAY along dimension DIM if
the corresponding element in MASK is TRUE.
Standard : Fortran 2008 and later
Class: Transformational function
Syntax :
RESULT = IALL(ARRAY[, MASK])
RESULT = IALL(ARRAY, DIM[, MASK])
Arguments:
ARRAY Shall be an array of type INTEGER
DIM (Optional) shall be a scalar of type INTEGER with a
value in the range from 1 to n, where n equals the
rank of ARRAY.
MASK (Optional) shall be of type LOGICAL and either be a
scalar or an array of the same shape as ARRAY.
202 The GNU Fortran Compiler

Return value:
The result is of the same type as ARRAY.
If DIM is absent, a scalar with the bitwise ALL of all elements in ARRAY is
returned. Otherwise, an array of rank n-1, where n equals the rank of ARRAY,
and a shape similar to that of ARRAY with dimension DIM dropped is re-
turned.
Example:
PROGRAM test_iall
INTEGER(1) :: a(2)

a(1) = b'00100100'
a(2) = b'01101010'

! prints 00100000
PRINT '(b8.8)', IALL(a)
END PROGRAM

See also: Section 8.138 [IANY], page 203,

Section 8.153 [IPARITY], page 212,
Section 8.137 [IAND], page 202,

8.137 IAND — Bitwise logical and

Description:
Bitwise logical AND.
Standard : Fortran 90 and later, with boz-literal-constant Fortran 2008 and later, has over-
loads that are GNU extensions
Class: Elemental function
Syntax : RESULT = IAND(I, J)
Arguments:
I The type shall be INTEGER or a boz-literal-constant.
J The type shall be INTEGER with the same kind type
parameter as I or a boz-literal-constant. I and J shall
not both be boz-literal-constants.
Return value:
The return type is INTEGER with the kind type parameter of the arguments. A
boz-literal-constant is converted to an INTEGER with the kind type parameter
of the other argument as-if a call to Section 8.149 [INT], page 210, occurred.
Example:
PROGRAM test_iand
INTEGER :: a, b
DATA a / Z'F' /, b / Z'3' /
WRITE (*,*) IAND(a, b)
END PROGRAM

Specific names:
Name Argument Return type Standard
IAND(A) INTEGER A INTEGER Fortran 90 and later
Chapter 8: Intrinsic Procedures 203

BIAND(A) INTEGER(1) A INTEGER(1) GNU extension

IIAND(A) INTEGER(2) A INTEGER(2) GNU extension
JIAND(A) INTEGER(4) A INTEGER(4) GNU extension
KIAND(A) INTEGER(8) A INTEGER(8) GNU extension
See also: Section 8.152 [IOR], page 212,
Section 8.145 [IEOR], page 208,
Section 8.141 [IBITS], page 205,
Section 8.142 [IBSET], page 206,
Section 8.140 [IBCLR], page 204,
Section 8.208 [NOT], page 245,

8.138 IANY — Bitwise OR of array elements

Description:
Reduces with bitwise OR (inclusive or) the elements of ARRAY along dimen-
sion DIM if the corresponding element in MASK is TRUE.
Standard : Fortran 2008 and later
Class: Transformational function
Syntax :
RESULT = IANY(ARRAY[, MASK])
RESULT = IANY(ARRAY, DIM[, MASK])
Arguments:
ARRAY Shall be an array of type INTEGER
DIM (Optional) shall be a scalar of type INTEGER with a
value in the range from 1 to n, where n equals the
rank of ARRAY.
MASK (Optional) shall be of type LOGICAL and either be a
scalar or an array of the same shape as ARRAY.
Return value:
The result is of the same type as ARRAY.
If DIM is absent, a scalar with the bitwise OR of all elements in ARRAY is
returned. Otherwise, an array of rank n-1, where n equals the rank of ARRAY,
and a shape similar to that of ARRAY with dimension DIM dropped is re-
turned.
Example:
PROGRAM test_iany
INTEGER(1) :: a(2)

a(1) = b'00100100'
a(2) = b'01101010'

! prints 01101110
PRINT '(b8.8)', IANY(a)
END PROGRAM
204 The GNU Fortran Compiler

See also: Section 8.153 [IPARITY], page 212,

Section 8.136 [IALL], page 201,
Section 8.152 [IOR], page 212,

8.139 IARGC — Get the number of command line arguments

Description:
IARGC returns the number of arguments passed on the command line when the
containing program was invoked.
This intrinsic routine is provided for backwards compatibility with GNU For-
tran 77. In new code, programmers should consider the use of the Section 8.68
[COMMAND ARGUMENT COUNT], page 157, intrinsic defined by the For-
tran 2003 standard.
Standard : GNU extension
Class: Function
Syntax : RESULT = IARGC()
Arguments:
None
Return value:
The number of command line arguments, type INTEGER(4).
Example: See Section 8.121 [GETARG], page 192,
See also: GNU Fortran 77 compatibility subroutine:
Section 8.121 [GETARG], page 192,
Fortran 2003 functions and subroutines:
Section 8.122 [GET COMMAND], page 193,
Section 8.123 [GET COMMAND ARGUMENT], page 194,
Section 8.68 [COMMAND ARGUMENT COUNT], page 157,

8.140 IBCLR — Clear bit

Description:
IBCLR returns the value of I with the bit at position POS set to zero.
Standard : Fortran 90 and later, has overloads that are GNU extensions
Class: Elemental function
Syntax : RESULT = IBCLR(I, POS)
Arguments:
I The type shall be INTEGER.
POS The type shall be INTEGER.
Return value:
The return value is of type INTEGER and of the same kind as I.
Chapter 8: Intrinsic Procedures 205

Specific names:
Name Argument Return type Standard
IBCLR(A) INTEGER A INTEGER Fortran 90 and later
BBCLR(A) INTEGER(1) A INTEGER(1) GNU extension
IIBCLR(A) INTEGER(2) A INTEGER(2) GNU extension
JIBCLR(A) INTEGER(4) A INTEGER(4) GNU extension
KIBCLR(A) INTEGER(8) A INTEGER(8) GNU extension
See also: Section 8.141 [IBITS], page 205,
Section 8.142 [IBSET], page 206,
Section 8.137 [IAND], page 202,
Section 8.152 [IOR], page 212,
Section 8.145 [IEOR], page 208,
Section 8.203 [MVBITS], page 242,

8.141 IBITS — Bit extraction

Description:
IBITS extracts a field of length LEN from I, starting from bit position POS
and extending left for LEN bits. The result is right-justified and the remaining
bits are zeroed. The value of POS+LEN must be less than or equal to the value
BIT_SIZE(I).
Standard : Fortran 90 and later, has overloads that are GNU extensions
Class: Elemental function
Syntax : RESULT = IBITS(I, POS, LEN)
Arguments:
I The type shall be INTEGER.
POS The type shall be INTEGER.
LEN The type shall be INTEGER.
Return value:
The return value is of type INTEGER and of the same kind as I.
Specific names:
Name Argument Return type Standard
IBITS(A) INTEGER A INTEGER Fortran 90 and later
BBITS(A) INTEGER(1) A INTEGER(1) GNU extension
IIBITS(A) INTEGER(2) A INTEGER(2) GNU extension
JIBITS(A) INTEGER(4) A INTEGER(4) GNU extension
KIBITS(A) INTEGER(8) A INTEGER(8) GNU extension
See also: Section 8.48 [BIT SIZE], page 142,
Section 8.140 [IBCLR], page 204,
Section 8.142 [IBSET], page 206,
Section 8.137 [IAND], page 202,
Section 8.152 [IOR], page 212,
Section 8.145 [IEOR], page 208,
206 The GNU Fortran Compiler

8.142 IBSET — Set bit

Description:
IBSET returns the value of I with the bit at position POS set to one.
Standard : Fortran 90 and later, has overloads that are GNU extensions
Class: Elemental function
Syntax : RESULT = IBSET(I, POS)
Arguments:
I The type shall be INTEGER.
POS The type shall be INTEGER.
Return value:
The return value is of type INTEGER and of the same kind as I.
Specific names:
Name Argument Return type Standard
IBSET(A) INTEGER A INTEGER Fortran 90 and later
BBSET(A) INTEGER(1) A INTEGER(1) GNU extension
IIBSET(A) INTEGER(2) A INTEGER(2) GNU extension
JIBSET(A) INTEGER(4) A INTEGER(4) GNU extension
KIBSET(A) INTEGER(8) A INTEGER(8) GNU extension
See also: Section 8.140 [IBCLR], page 204,
Section 8.141 [IBITS], page 205,
Section 8.137 [IAND], page 202,
Section 8.152 [IOR], page 212,
Section 8.145 [IEOR], page 208,
Section 8.203 [MVBITS], page 242,

8.143 ICHAR — Character-to-integer conversion function

Description:
ICHAR(C) returns the code for the character in the first character position of
C in the system’s native character set. The correspondence between charac-
ters and their codes is not necessarily the same across different GNU Fortran
implementations.
Standard : Fortran 77 and later, with KIND argument Fortran 2003 and later
Class: Elemental function
Syntax : RESULT = ICHAR(C [, KIND])
Arguments:
C Shall be a scalar CHARACTER, with INTENT(IN)
KIND (Optional) A scalar INTEGER constant expression in-
dicating the kind parameter of the result.
Return value:
The return value is of type INTEGER and of kind KIND. If KIND is absent, the
return value is of default integer kind.
Chapter 8: Intrinsic Procedures 207

Example:
program test_ichar
integer i
i = ichar(' ')
end program test_ichar

Specific names:
Name Argument Return type Standard
ICHAR(C) CHARACTER C INTEGER(4) Fortran 77 and later
Note: No intrinsic exists to convert between a numeric value and a formatted character
string representation – for instance, given the CHARACTER value '154', obtaining
an INTEGER or REAL value with the value 154, or vice versa. Instead, this
functionality is provided by internal-file I/O, as in the following example:
program read_val
integer value
character(len=10) string, string2
string = '154'

! Convert a string to a numeric value

read (string,'(I10)') value
print *, value

! Convert a value to a formatted string

write (string2,'(I10)') value
print *, string2
end program read_val

See also: Section 8.5 [ACHAR], page 113,

Section 8.59 [CHAR], page 149,
Section 8.135 [IACHAR], page 201,

8.144 IDATE — Get current local time subroutine

(day/month/year)
Description:
IDATE(VALUES) Fills VALUES with the numerical values at the current local
time. The day (in the range 1-31), month (in the range 1-12), and year appear
in elements 1, 2, and 3 of VALUES, respectively. The year has four significant
digits.
This intrinsic routine is provided for backwards compatibility with GNU Fortran
77. In new code, programmers should consider the use of the Section 8.82
[DATE AND TIME], page 166, intrinsic defined by the Fortran 95 standard.
Standard : GNU extension
Class: Subroutine
Syntax : CALL IDATE(VALUES)
Arguments:
VALUES The type shall be INTEGER, DIMENSION(3) and the
kind shall be the default integer kind.
208 The GNU Fortran Compiler

Return value:
Does not return anything.
Example:
program test_idate
integer, dimension(3) :: tarray
call idate(tarray)
print *, tarray(1)
print *, tarray(2)
print *, tarray(3)
end program test_idate

See also: Section 8.82 [DATE AND TIME], page 166,

8.145 IEOR — Bitwise logical exclusive or

Description:
IEOR returns the bitwise Boolean exclusive-OR of I and J.
Standard : Fortran 90 and later, with boz-literal-constant Fortran 2008 and later, has over-
loads that are GNU extensions
Class: Elemental function
Syntax : RESULT = IEOR(I, J)
Arguments:
I The type shall be INTEGER or a boz-literal-constant.
J The type shall be INTEGER with the same kind type
parameter as I or a boz-literal-constant. I and J shall
not both be boz-literal-constants.
Return value:
The return type is INTEGER with the kind type parameter of the arguments. A
boz-literal-constant is converted to an INTEGER with the kind type parameter
of the other argument as-if a call to Section 8.149 [INT], page 210, occurred.
Specific names:
Name Argument Return type Standard
IEOR(A) INTEGER A INTEGER Fortran 90 and later
BIEOR(A) INTEGER(1) A INTEGER(1) GNU extension
IIEOR(A) INTEGER(2) A INTEGER(2) GNU extension
JIEOR(A) INTEGER(4) A INTEGER(4) GNU extension
KIEOR(A) INTEGER(8) A INTEGER(8) GNU extension
See also: Section 8.152 [IOR], page 212,
Section 8.137 [IAND], page 202,
Section 8.141 [IBITS], page 205,
Section 8.142 [IBSET], page 206,
Section 8.140 [IBCLR], page 204,
Section 8.208 [NOT], page 245,
Chapter 8: Intrinsic Procedures 209

8.146 IERRNO — Get the last system error number

Description:
Returns the last system error number, as given by the C errno variable.
Standard : GNU extension
Class: Function
Syntax : RESULT = IERRNO()
Arguments:
None
Return value:
The return value is of type INTEGER and of the default integer kind.
See also: Section 8.214 [PERROR], page 249,

8.147 IMAGE_INDEX — Function that converts a cosubscript to

an image index
Description:
Returns the image index belonging to a cosubscript.
Standard : Fortran 2008 and later
Class: Inquiry function.
Syntax : RESULT = IMAGE_INDEX(COARRAY, SUB)
Arguments:
COARRAY Coarray of any type.
SUB default integer rank-1 array of a size equal to the
corank of COARRAY.
Return value:
Scalar default integer with the value of the image index which corresponds to
the cosubscripts. For invalid cosubscripts the result is zero.
Example:
INTEGER :: array[2,-1:4,8,*]
! Writes 28 (or 0 if there are fewer than 28 images)
WRITE (*,*) IMAGE_INDEX (array, [2,0,3,1])

See also: Section 8.268 [THIS IMAGE], page 281,

Section 8.210 [NUM IMAGES], page 246,

8.148 INDEX — Position of a substring within a string

Description:
Returns the position of the start of the first occurrence of string SUBSTRING
as a substring in STRING, counting from one. If SUBSTRING is not present
in STRING, zero is returned. If the BACK argument is present and true, the
return value is the start of the last occurrence rather than the first.
210 The GNU Fortran Compiler

Standard : Fortran 77 and later, with KIND argument Fortran 2003 and later
Class: Elemental function
Syntax : RESULT = INDEX(STRING, SUBSTRING [, BACK [, KIND]])
Arguments:
STRING Shall be a scalar CHARACTER, with INTENT(IN)
SUBSTRING Shall be a scalar CHARACTER, with INTENT(IN)
BACK (Optional) Shall be a scalar LOGICAL, with
INTENT(IN)
KIND (Optional) A scalar INTEGER constant expression in-
dicating the kind parameter of the result.
Return value:
The return value is of type INTEGER and of kind KIND. If KIND is absent, the
return value is of default integer kind.
Specific names:
Name Argument Return Standard
type
INDEX(STRING,SUBSTRING) CHARACTER INTEGER(4) Fortran 77 and later
See also: Section 8.236 [SCAN], page 261,
Section 8.282 [VERIFY], page 289,

8.149 INT — Convert to integer type

Description:
Convert to integer type
Standard : Fortran 77 and later, with boz-literal-constant Fortran 2008 and later.
Class: Elemental function
Syntax : RESULT = INT(A [, KIND))
Arguments:
A Shall be of type INTEGER, REAL, or COMPLEX or a boz-
literal-constant.
KIND (Optional) A scalar INTEGER constant expression in-
dicating the kind parameter of the result.
Return value:
These functions return a INTEGER variable or array under the following rules:
(A) If A is of type INTEGER, INT(A) = A
(B) If A is of type REAL and |A| < 1, INT(A) equals 0. If |A| ≥ 1, then
INT(A) is the integer whose magnitude is the largest integer that
does not exceed the magnitude of A and whose sign is the same as
the sign of A.
(C) If A is of type COMPLEX, rule B is applied to the real part of A.
Chapter 8: Intrinsic Procedures 211

Example:
program test_int
integer :: i = 42
complex :: z = (-3.7, 1.0)
print *, int(i)
print *, int(z), int(z,8)
end program
Specific names:
Name Argument Return type Standard
INT(A) REAL(4) A INTEGER Fortran 77 and later
IFIX(A) REAL(4) A INTEGER Fortran 77 and later
IDINT(A) REAL(8) A INTEGER Fortran 77 and later

8.150 INT2 — Convert to 16-bit integer type

Description:
Convert to a KIND=2 integer type. This is equivalent to the standard INT intrin-
sic with an optional argument of KIND=2, and is only included for backwards
compatibility.
Standard : GNU extension
Class: Elemental function
Syntax : RESULT = INT2(A)
Arguments:
A Shall be of type INTEGER, REAL, or COMPLEX.
Return value:
The return value is a INTEGER(2) variable.
See also: Section 8.149 [INT], page 210,
Section 8.151 [INT8], page 211,

8.151 INT8 — Convert to 64-bit integer type

Description:
Convert to a KIND=8 integer type. This is equivalent to the standard INT intrin-
sic with an optional argument of KIND=8, and is only included for backwards
compatibility.
Standard : GNU extension
Class: Elemental function
Syntax : RESULT = INT8(A)
Arguments:
A Shall be of type INTEGER, REAL, or COMPLEX.
Return value:
The return value is a INTEGER(8) variable.
See also: Section 8.149 [INT], page 210,
Section 8.150 [INT2], page 211,
212 The GNU Fortran Compiler

8.152 IOR — Bitwise logical or

Description:
IOR returns the bitwise Boolean inclusive-OR of I and J.
Standard : Fortran 90 and later, with boz-literal-constant Fortran 2008 and later, has over-
loads that are GNU extensions
Class: Elemental function
Syntax : RESULT = IOR(I, J)
Arguments:
I The type shall be INTEGER or a boz-literal-constant.
J The type shall be INTEGER with the same kind type
parameter as I or a boz-literal-constant. I and J shall
not both be boz-literal-constants.
Return value:
The return type is INTEGER with the kind type parameter of the arguments. A
boz-literal-constant is converted to an INTEGER with the kind type parameter
of the other argument as-if a call to Section 8.149 [INT], page 210, occurred.
Specific names:
Name Argument Return type Standard
IOR(A) INTEGER A INTEGER Fortran 90 and later
BIOR(A) INTEGER(1) A INTEGER(1) GNU extension
IIOR(A) INTEGER(2) A INTEGER(2) GNU extension
JIOR(A) INTEGER(4) A INTEGER(4) GNU extension
KIOR(A) INTEGER(8) A INTEGER(8) GNU extension
See also: Section 8.145 [IEOR], page 208,
Section 8.137 [IAND], page 202,
Section 8.141 [IBITS], page 205,
Section 8.142 [IBSET], page 206,
Section 8.140 [IBCLR], page 204,
Section 8.208 [NOT], page 245,

8.153 IPARITY — Bitwise XOR of array elements

Description:
Reduces with bitwise XOR (exclusive or) the elements of ARRAY along dimen-
sion DIM if the corresponding element in MASK is TRUE.
Standard : Fortran 2008 and later
Class: Transformational function
Syntax :
RESULT = IPARITY(ARRAY[, MASK])
RESULT = IPARITY(ARRAY, DIM[, MASK])
Arguments:
ARRAY Shall be an array of type INTEGER
Chapter 8: Intrinsic Procedures 213

DIM (Optional) shall be a scalar of type INTEGER with a

value in the range from 1 to n, where n equals the
rank of ARRAY.
MASK (Optional) shall be of type LOGICAL and either be a
scalar or an array of the same shape as ARRAY.
Return value:
The result is of the same type as ARRAY.
If DIM is absent, a scalar with the bitwise XOR of all elements in ARRAY is
returned. Otherwise, an array of rank n-1, where n equals the rank of ARRAY,
and a shape similar to that of ARRAY with dimension DIM dropped is re-
turned.
Example:
PROGRAM test_iparity
INTEGER(1) :: a(2)

a(1) = int(b'00100100', 1)
a(2) = int(b'01101010', 1)

! prints 01001110
PRINT '(b8.8)', IPARITY(a)
END PROGRAM

See also: Section 8.138 [IANY], page 203,

Section 8.136 [IALL], page 201,
Section 8.145 [IEOR], page 208,
Section 8.213 [PARITY], page 249,

8.154 IRAND — Integer pseudo-random number

Description:
IRAND(FLAG) returns a pseudo-random number from a uniform distribution
between 0 and a system-dependent limit (which is in most cases 2147483647).
If FLAG is 0, the next number in the current sequence is returned; if FLAG is
1, the generator is restarted by CALL SRAND(0); if FLAG has any other value,
it is used as a new seed with SRAND.
This intrinsic routine is provided for backwards compatibility with GNU Fortran
77. It implements a simple modulo generator as provided by g77. For new code,
one should consider the use of Section 8.224 [RANDOM NUMBER], page 255,
as it implements a superior algorithm.
Standard : GNU extension
Class: Function
Syntax : RESULT = IRAND(I)
Arguments:
I Shall be a scalar INTEGER of kind 4.
Return value:
The return value is of INTEGER(kind=4) type.
214 The GNU Fortran Compiler

Example:
program test_irand
integer,parameter :: seed = 86456

call srand(seed)
print *, irand(), irand(), irand(), irand()
print *, irand(seed), irand(), irand(), irand()
end program test_irand

8.155 IS_CONTIGUOUS — Test whether an array is contiguous

Description:
IS_CONTIGUOUS tests whether an array is contiguous.
Standard : Fortran 2008 and later
Class: Inquiry function
Syntax : RESULT = IS_CONTIGUOUS(ARRAY)
Arguments:
ARRAY Shall be an array of any type.
Return value:
Returns a LOGICAL of the default kind, which .TRUE. if ARRAY is contiguous
and false otherwise.
Example:
program test
integer :: a(10)
a = [1,2,3,4,5,6,7,8,9,10]
call sub (a) ! every element, is contiguous
call sub (a(::2)) ! every other element, is noncontiguous
contains
subroutine sub (x)
integer :: x(:)
if (is_contiguous (x)) then
write (*,*) 'X is contiguous'
else
write (*,*) 'X is not contiguous'
end if
end subroutine sub
end program test

8.156 IS_IOSTAT_END — Test for end-of-file value

Description:
IS_IOSTAT_END tests whether an variable has the value of the I/O status “end of
file”. The function is equivalent to comparing the variable with the IOSTAT_END
parameter of the intrinsic module ISO_FORTRAN_ENV.
Standard : Fortran 2003 and later
Class: Elemental function
Syntax : RESULT = IS_IOSTAT_END(I)
Chapter 8: Intrinsic Procedures 215

Arguments:
I Shall be of the type INTEGER.
Return value:
Returns a LOGICAL of the default kind, which .TRUE. if I has the value which
indicates an end of file condition for IOSTAT= specifiers, and is .FALSE. other-
wise.
Example:
PROGRAM iostat
IMPLICIT NONE
INTEGER :: stat, i
OPEN(88, FILE='test.dat')
READ(88, *, IOSTAT=stat) i
IF(IS_IOSTAT_END(stat)) STOP 'END OF FILE'
END PROGRAM

8.157 IS_IOSTAT_EOR — Test for end-of-record value

Description:
IS_IOSTAT_EOR tests whether an variable has the value of the I/O status “end of
record”. The function is equivalent to comparing the variable with the IOSTAT_
EOR parameter of the intrinsic module ISO_FORTRAN_ENV.
Standard : Fortran 2003 and later
Class: Elemental function
Syntax : RESULT = IS_IOSTAT_EOR(I)
Arguments:
I Shall be of the type INTEGER.
Return value:
Returns a LOGICAL of the default kind, which .TRUE. if I has the value which
indicates an end of file condition for IOSTAT= specifiers, and is .FALSE. other-
wise.
Example:
PROGRAM iostat
IMPLICIT NONE
INTEGER :: stat, i(50)
OPEN(88, FILE='test.dat', FORM='UNFORMATTED')
READ(88, IOSTAT=stat) i
IF(IS_IOSTAT_EOR(stat)) STOP 'END OF RECORD'
END PROGRAM

8.158 ISATTY — Whether a unit is a terminal device

Description:
Determine whether a unit is connected to a terminal device.
Standard : GNU extension
Class: Function
216 The GNU Fortran Compiler

Syntax : RESULT = ISATTY(UNIT)

Arguments:
UNIT Shall be a scalar INTEGER.

Return value:
Returns .TRUE. if the UNIT is connected to a terminal device, .FALSE. other-
wise.

Example:
PROGRAM test_isatty
INTEGER(kind=1) :: unit
DO unit = 1, 10
write(*,*) isatty(unit=unit)
END DO
END PROGRAM

See also: Section 8.276 [TTYNAM], page 286,

8.159 ISHFT — Shift bits

Description:
ISHFT returns a value corresponding to I with all of the bits shifted SHIFT
places. A value of SHIFT greater than zero corresponds to a left shift, a value
of zero corresponds to no shift, and a value less than zero corresponds to a right
shift. If the absolute value of SHIFT is greater than BIT_SIZE(I), the value
is undefined. Bits shifted out from the left end or right end are lost; zeros are
shifted in from the opposite end.

Standard : Fortran 90 and later, has overloads that are GNU extensions

Class: Elemental function

Syntax : RESULT = ISHFT(I, SHIFT)

Arguments:
I The type shall be INTEGER.
SHIFT The type shall be INTEGER.

Return value:
The return value is of type INTEGER and of the same kind as I.

Specific names:
Name Argument Return type Standard
ISHFT(A) INTEGER A INTEGER Fortran 90 and later
BSHFT(A) INTEGER(1) A INTEGER(1) GNU extension
IISHFT(A) INTEGER(2) A INTEGER(2) GNU extension
JISHFT(A) INTEGER(4) A INTEGER(4) GNU extension
KISHFT(A) INTEGER(8) A INTEGER(8) GNU extension

See also: Section 8.160 [ISHFTC], page 217,

Chapter 8: Intrinsic Procedures 217

8.160 ISHFTC — Shift bits circularly

Description:
ISHFTC returns a value corresponding to I with the rightmost SIZE bits shifted
circularly SHIFT places; that is, bits shifted out one end are shifted into the
opposite end. A value of SHIFT greater than zero corresponds to a left shift,
a value of zero corresponds to no shift, and a value less than zero corresponds
to a right shift. The absolute value of SHIFT must be less than SIZE. If the
SIZE argument is omitted, it is taken to be equivalent to BIT_SIZE(I).
Standard : Fortran 90 and later, has overloads that are GNU extensions
Class: Elemental function
Syntax : RESULT = ISHFTC(I, SHIFT [, SIZE])
Arguments:
I The type shall be INTEGER.
SHIFT The type shall be INTEGER.
SIZE (Optional) The type shall be INTEGER; the value must
be greater than zero and less than or equal to BIT_
SIZE(I).
Return value:
The return value is of type INTEGER and of the same kind as I.
Specific names:
Name Argument Return type Standard
ISHFTC(A) INTEGER A INTEGER Fortran 90 and later
BSHFTC(A) INTEGER(1) A INTEGER(1) GNU extension
IISHFTC(A) INTEGER(2) A INTEGER(2) GNU extension
JISHFTC(A) INTEGER(4) A INTEGER(4) GNU extension
KISHFTC(A) INTEGER(8) A INTEGER(8) GNU extension
See also: Section 8.159 [ISHFT], page 216,

8.161 ISNAN — Test for a NaN

Description:
ISNAN tests whether a floating-point value is an IEEE Not-a-Number (NaN).
Standard : GNU extension
Class: Elemental function
Syntax : ISNAN(X)
Arguments:
X Variable of the type REAL.

Return value:
Returns a default-kind LOGICAL. The returned value is TRUE if X is a NaN and
FALSE otherwise.
218 The GNU Fortran Compiler

Example:
program test_nan
implicit none
real :: x
x = -1.0
x = sqrt(x)
if (isnan(x)) stop '"x" is a NaN'
end program test_nan

8.162 ITIME — Get current local time subroutine

(hour/minutes/seconds)
Description:
ITIME(VALUES) Fills VALUES with the numerical values at the current local
time. The hour (in the range 1-24), minute (in the range 1-60), and seconds (in
the range 1-60) appear in elements 1, 2, and 3 of VALUES, respectively.
This intrinsic routine is provided for backwards compatibility with GNU Fortran
77. In new code, programmers should consider the use of the Section 8.82
[DATE AND TIME], page 166, intrinsic defined by the Fortran 95 standard.
Standard : GNU extension
Class: Subroutine
Syntax : CALL ITIME(VALUES)
Arguments:
VALUES The type shall be INTEGER, DIMENSION(3) and the
kind shall be the default integer kind.
Return value:
Does not return anything.
Example:
program test_itime
integer, dimension(3) :: tarray
call itime(tarray)
print *, tarray(1)
print *, tarray(2)
print *, tarray(3)
end program test_itime

See also: Section 8.82 [DATE AND TIME], page 166,

8.163 KILL — Send a signal to a process

Description:
Sends the signal specified by SIG to the process PID. See kill(2).
This intrinsic is provided in both subroutine and function forms; however, only
one form can be used in any given program unit.
Standard : GNU extension
Standard : GNU extension
Chapter 8: Intrinsic Procedures 219

Class: Subroutine, function

Syntax :
CALL KILL(PID, SIG [, STATUS])
STATUS = KILL(PID, SIG)
Arguments:
PID Shall be a scalar INTEGER with INTENT(IN).
SIG Shall be a scalar INTEGER with INTENT(IN).
STATUS [Subroutine](Optional) Shall be a scalar INTEGER.
Returns 0 on success; otherwise a system-specific er-
ror code is returned.
STATUS [Function] The kind type parameter is that of pid.
Returns 0 on success; otherwise a system-specific er-
ror code is returned.
See also: Section 8.2 [ABORT], page 111,
Section 8.101 [EXIT], page 179,

8.164 KIND — Kind of an entity

Description:
KIND(X) returns the kind value of the entity X.
Standard : Fortran 95 and later
Class: Inquiry function
Syntax : K = KIND(X)
Arguments:
X Shall be of type LOGICAL, INTEGER, REAL, COMPLEX
or CHARACTER. It may be scalar or array valued.
Return value:
The return value is a scalar of type INTEGER and of the default integer kind.
Example:
program test_kind
integer,parameter :: kc = kind(' ')
integer,parameter :: kl = kind(.true.)

print *, "The default character kind is ", kc

print *, "The default logical kind is ", kl
end program test_kind

8.165 LBOUND — Lower dimension bounds of an array

Description:
Returns the lower bounds of an array, or a single lower bound along the DIM
dimension.
Standard : Fortran 90 and later, with KIND argument Fortran 2003 and later
220 The GNU Fortran Compiler

Class: Inquiry function

Syntax : RESULT = LBOUND(ARRAY [, DIM [, KIND]])
Arguments:
ARRAY Shall be an array, of any type.
DIM (Optional) Shall be a scalar INTEGER.
KIND (Optional) A scalar INTEGER constant expression in-
dicating the kind parameter of the result.
Return value:
The return value is of type INTEGER and of kind KIND. If KIND is absent,
the return value is of default integer kind. If DIM is absent, the result is an
array of the lower bounds of ARRAY. If DIM is present, the result is a scalar
corresponding to the lower bound of the array along that dimension. If ARRAY
is an expression rather than a whole array or array structure component, or if
it has a zero extent along the relevant dimension, the lower bound is taken to
be 1.
See also: Section 8.277 [UBOUND], page 287,
Section 8.166 [LCOBOUND], page 220,

8.166 LCOBOUND — Lower codimension bounds of an array

Description:
Returns the lower bounds of a coarray, or a single lower cobound along the
DIM codimension.
Standard : Fortran 2008 and later
Class: Inquiry function
Syntax : RESULT = LCOBOUND(COARRAY [, DIM [, KIND]])
Arguments:
ARRAY Shall be an coarray, of any type.
DIM (Optional) Shall be a scalar INTEGER.
KIND (Optional) A scalar INTEGER constant expression in-
dicating the kind parameter of the result.
Return value:
The return value is of type INTEGER and of kind KIND. If KIND is absent, the
return value is of default integer kind. If DIM is absent, the result is an array
of the lower cobounds of COARRAY. If DIM is present, the result is a scalar
corresponding to the lower cobound of the array along that codimension.
See also: Section 8.278 [UCOBOUND], page 287,
Section 8.165 [LBOUND], page 219,

8.167 LEADZ — Number of leading zero bits of an integer

Description:
LEADZ returns the number of leading zero bits of an integer.
Chapter 8: Intrinsic Procedures 221

Standard : Fortran 2008 and later

Class: Elemental function
Syntax : RESULT = LEADZ(I)
Arguments:
I Shall be of type INTEGER.
Return value:
The type of the return value is the default INTEGER. If all the bits of I are zero,
the result value is BIT_SIZE(I).
Example:
PROGRAM test_leadz
WRITE (*,*) BIT_SIZE(1) ! prints 32
WRITE (*,*) LEADZ(1) ! prints 31
END PROGRAM

See also: Section 8.48 [BIT SIZE], page 142,

Section 8.272 [TRAILZ], page 284,
Section 8.215 [POPCNT], page 250,
Section 8.216 [POPPAR], page 250,

8.168 LEN — Length of a character entity

Description:
Returns the length of a character string. If STRING is an array, the length of an
element of STRING is returned. Note that STRING need not be defined when
this intrinsic is invoked, since only the length, not the content, of STRING is
needed.
Standard : Fortran 77 and later, with KIND argument Fortran 2003 and later
Class: Inquiry function
Syntax : L = LEN(STRING [, KIND])
Arguments:
STRING Shall be a scalar or array of type CHARACTER, with
INTENT(IN)
KIND (Optional) A scalar INTEGER constant expression in-
dicating the kind parameter of the result.
Return value:
The return value is of type INTEGER and of kind KIND. If KIND is absent, the
return value is of default integer kind.
Specific names:
Name Argument Return type Standard
LEN(STRING) CHARACTER INTEGER Fortran 77 and later
See also: Section 8.169 [LEN TRIM], page 222,
Section 8.9 [ADJUSTL], page 116,
Section 8.10 [ADJUSTR], page 116,
222 The GNU Fortran Compiler

8.169 LEN_TRIM — Length of a character entity without

trailing blank characters
Description:
Returns the length of a character string, ignoring any trailing blanks.
Standard : Fortran 90 and later, with KIND argument Fortran 2003 and later
Class: Elemental function
Syntax : RESULT = LEN_TRIM(STRING [, KIND])
Arguments:
STRING Shall be a scalar of type CHARACTER, with INTENT(IN)
KIND (Optional) A scalar INTEGER constant expression in-
dicating the kind parameter of the result.
Return value:
The return value is of type INTEGER and of kind KIND. If KIND is absent, the
return value is of default integer kind.
See also: Section 8.168 [LEN], page 221,
Section 8.9 [ADJUSTL], page 116,
Section 8.10 [ADJUSTR], page 116,

8.170 LGE — Lexical greater than or equal

Description:
Determines whether one string is lexically greater than or equal to another
string, where the two strings are interpreted as containing ASCII character
codes. If the String A and String B are not the same length, the shorter is
compared as if spaces were appended to it to form a value that has the same
length as the longer.
In general, the lexical comparison intrinsics LGE, LGT, LLE, and LLT differ from
the corresponding intrinsic operators .GE., .GT., .LE., and .LT., in that the
latter use the processor’s character ordering (which is not ASCII on some tar-
gets), whereas the former always use the ASCII ordering.
Standard : Fortran 77 and later
Class: Elemental function
Syntax : RESULT = LGE(STRING_A, STRING_B)
Arguments:
STRING A Shall be of default CHARACTER type.
STRING B Shall be of default CHARACTER type.
Return value:
Returns .TRUE. if STRING_A >= STRING_B, and .FALSE. otherwise, based on
the ASCII ordering.
Specific names:
Name Argument Return Standard
type
LGE(STRING_A,STRING_B) CHARACTER LOGICAL Fortran 77 and later
Chapter 8: Intrinsic Procedures 223

See also: Section 8.171 [LGT], page 223,

Section 8.173 [LLE], page 224,
Section 8.174 [LLT], page 225,

8.171 LGT — Lexical greater than

Description:
Determines whether one string is lexically greater than another string, where
the two strings are interpreted as containing ASCII character codes. If the
String A and String B are not the same length, the shorter is compared as if
spaces were appended to it to form a value that has the same length as the
longer.
In general, the lexical comparison intrinsics LGE, LGT, LLE, and LLT differ from
the corresponding intrinsic operators .GE., .GT., .LE., and .LT., in that the
latter use the processor’s character ordering (which is not ASCII on some tar-
gets), whereas the former always use the ASCII ordering.
Standard : Fortran 77 and later
Class: Elemental function
Syntax : RESULT = LGT(STRING_A, STRING_B)
Arguments:
STRING A Shall be of default CHARACTER type.
STRING B Shall be of default CHARACTER type.
Return value:
Returns .TRUE. if STRING_A > STRING_B, and .FALSE. otherwise, based on the
ASCII ordering.
Specific names:
Name Argument Return Standard
type
LGT(STRING_A,STRING_B) CHARACTER LOGICAL Fortran 77 and later
See also: Section 8.170 [LGE], page 222,
Section 8.173 [LLE], page 224,
Section 8.174 [LLT], page 225,

8.172 LINK — Create a hard link

Description:
Makes a (hard) link from file PATH1 to PATH2. A null character (CHAR(0)) can
be used to mark the end of the names in PATH1 and PATH2; otherwise, trailing
blanks in the file names are ignored. If the STATUS argument is supplied, it
contains 0 on success or a nonzero error code upon return; see link(2).
This intrinsic is provided in both subroutine and function forms; however, only
one form can be used in any given program unit.
Standard : GNU extension
224 The GNU Fortran Compiler

Class: Subroutine, function

Syntax :
CALL LINK(PATH1, PATH2 [, STATUS])
STATUS = LINK(PATH1, PATH2)

Arguments:
PATH1 Shall be of default CHARACTER type.
PATH2 Shall be of default CHARACTER type.
STATUS (Optional) Shall be of default INTEGER type.

See also: Section 8.262 [SYMLNK], page 278,

Section 8.280 [UNLINK], page 288,

8.173 LLE — Lexical less than or equal

Description:
Determines whether one string is lexically less than or equal to another string,
where the two strings are interpreted as containing ASCII character codes. If
the String A and String B are not the same length, the shorter is compared as
if spaces were appended to it to form a value that has the same length as the
longer.
In general, the lexical comparison intrinsics LGE, LGT, LLE, and LLT differ from
the corresponding intrinsic operators .GE., .GT., .LE., and .LT., in that the
latter use the processor’s character ordering (which is not ASCII on some tar-
gets), whereas the former always use the ASCII ordering.

Standard : Fortran 77 and later

Class: Elemental function

Syntax : RESULT = LLE(STRING_A, STRING_B)

Arguments:
STRING A Shall be of default CHARACTER type.
STRING B Shall be of default CHARACTER type.

Return value:
Returns .TRUE. if STRING_A <= STRING_B, and .FALSE. otherwise, based on
the ASCII ordering.

Specific names:
Name Argument Return Standard
type
LLE(STRING_A,STRING_B) CHARACTER LOGICAL Fortran 77 and later

See also: Section 8.170 [LGE], page 222,

Section 8.171 [LGT], page 223,
Section 8.174 [LLT], page 225,
Chapter 8: Intrinsic Procedures 225

8.174 LLT — Lexical less than

Description:
Determines whether one string is lexically less than another string, where the
two strings are interpreted as containing ASCII character codes. If the String
A and String B are not the same length, the shorter is compared as if spaces
were appended to it to form a value that has the same length as the longer.
In general, the lexical comparison intrinsics LGE, LGT, LLE, and LLT differ from
the corresponding intrinsic operators .GE., .GT., .LE., and .LT., in that the
latter use the processor’s character ordering (which is not ASCII on some tar-
gets), whereas the former always use the ASCII ordering.
Standard : Fortran 77 and later
Class: Elemental function
Syntax : RESULT = LLT(STRING_A, STRING_B)
Arguments:
STRING A Shall be of default CHARACTER type.
STRING B Shall be of default CHARACTER type.
Return value:
Returns .TRUE. if STRING_A < STRING_B, and .FALSE. otherwise, based on the
ASCII ordering.
Specific names:
Name Argument Return Standard
type
LLT(STRING_A,STRING_B) CHARACTER LOGICAL Fortran 77 and later
See also: Section 8.170 [LGE], page 222,
Section 8.171 [LGT], page 223,
Section 8.173 [LLE], page 224,

8.175 LNBLNK — Index of the last non-blank character in a

string
Description:
Returns the length of a character string, ignoring any trailing blanks. This is
identical to the standard LEN_TRIM intrinsic, and is only included for backwards
compatibility.
Standard : GNU extension
Class: Elemental function
Syntax : RESULT = LNBLNK(STRING)
Arguments:
STRING Shall be a scalar of type CHARACTER, with INTENT(IN)
Return value:
The return value is of INTEGER(kind=4) type.
226 The GNU Fortran Compiler

See also: Section 8.148 [INDEX intrinsic], page 209,

Section 8.169 [LEN TRIM], page 222,

8.176 LOC — Returns the address of a variable

Description:
LOC(X) returns the address of X as an integer.
Standard : GNU extension
Class: Inquiry function
Syntax : RESULT = LOC(X)
Arguments:
X Variable of any type.
Return value:
The return value is of type INTEGER, with a KIND corresponding to the size (in
bytes) of a memory address on the target machine.
Example:
program test_loc
integer :: i
real :: r
i = loc(r)
print *, i
end program test_loc

8.177 LOG — Natural logarithm function

Description:
LOG(X) computes the natural logarithm of X, i.e. the logarithm to the base e.
Standard : Fortran 77 and later, has GNU extensions
Class: Elemental function
Syntax : RESULT = LOG(X)
Arguments:
X The type shall be REAL or COMPLEX.
Return value:
The return value is of type REAL or COMPLEX. The kind type parameter is the
same as X. If X is COMPLEX, the imaginary part ω is in the range −π < ω ≤ π.
Example:
program test_log
real(8) :: x = 2.7182818284590451_8
complex :: z = (1.0, 2.0)
x = log(x) ! will yield (approximately) 1
z = log(z)
end program test_log
Chapter 8: Intrinsic Procedures 227

Specific names:
Name Argument Return type Standard
ALOG(X) REAL(4) X REAL(4) Fortran 77 or later
DLOG(X) REAL(8) X REAL(8) Fortran 77 or later
CLOG(X) COMPLEX(4) X COMPLEX(4) Fortran 77 or later
ZLOG(X) COMPLEX(8) X COMPLEX(8) GNU extension
CDLOG(X) COMPLEX(8) X COMPLEX(8) GNU extension

8.178 LOG10 — Base 10 logarithm function

Description:
LOG10(X) computes the base 10 logarithm of X.
Standard : Fortran 77 and later
Class: Elemental function
Syntax : RESULT = LOG10(X)
Arguments:
X The type shall be REAL.
Return value:
The return value is of type REAL or COMPLEX. The kind type parameter is the
same as X.
Example:
program test_log10
real(8) :: x = 10.0_8
x = log10(x)
end program test_log10

Specific names:
Name Argument Return type Standard
ALOG10(X) REAL(4) X REAL(4) Fortran 77 and later
DLOG10(X) REAL(8) X REAL(8) Fortran 77 and later

8.179 LOG_GAMMA — Logarithm of the Gamma function

Description:
LOG_GAMMA(X) computes the natural logarithm of the absolute value of the
Gamma (Γ) function.
Standard : Fortran 2008 and later
Class: Elemental function
Syntax : X = LOG_GAMMA(X)
Arguments:
X Shall be of type REAL and neither zero nor a negative
integer.
Return value:
The return value is of type REAL of the same kind as X.
228 The GNU Fortran Compiler

Example:
program test_log_gamma
real :: x = 1.0
x = lgamma(x) ! returns 0.0
end program test_log_gamma

Specific names:
Name Argument Return type Standard
LGAMMA(X) REAL(4) X REAL(4) GNU extension
ALGAMA(X) REAL(4) X REAL(4) GNU extension
DLGAMA(X) REAL(8) X REAL(8) GNU extension
See also: Gamma function:
Section 8.119 [GAMMA], page 191,

8.180 LOGICAL — Convert to logical type

Description:
Converts one kind of LOGICAL variable to another.
Standard : Fortran 90 and later
Class: Elemental function
Syntax : RESULT = LOGICAL(L [, KIND])
Arguments:
L The type shall be LOGICAL.
KIND (Optional) A scalar INTEGER constant expression in-
dicating the kind parameter of the result.
Return value:
The return value is a LOGICAL value equal to L, with a kind corresponding to
KIND, or of the default logical kind if KIND is not given.
See also: Section 8.149 [INT], page 210,
Section 8.228 [REAL], page 257,
Section 8.62 [CMPLX], page 152,

8.181 LSHIFT — Left shift bits

Description:
LSHIFT returns a value corresponding to I with all of the bits shifted left by
SHIFT places. SHIFT shall be nonnegative and less than or equal to BIT_
SIZE(I), otherwise the result value is undefined. Bits shifted out from the left
end are lost; zeros are shifted in from the opposite end.
This function has been superseded by the ISHFT intrinsic, which is standard in
Fortran 95 and later, and the SHIFTL intrinsic, which is standard in Fortran
2008 and later.
Standard : GNU extension
Class: Elemental function
Chapter 8: Intrinsic Procedures 229

Syntax : RESULT = LSHIFT(I, SHIFT)

Arguments:
I The type shall be INTEGER.
SHIFT The type shall be INTEGER.

Return value:
The return value is of type INTEGER and of the same kind as I.

See also: Section 8.159 [ISHFT], page 216,

Section 8.160 [ISHFTC], page 217,
Section 8.233 [RSHIFT], page 260,
Section 8.244 [SHIFTA], page 267,
Section 8.245 [SHIFTL], page 267,
Section 8.246 [SHIFTR], page 268,

8.182 LSTAT — Get file status

Description:
LSTAT is identical to Section 8.259 [STAT], page 275, except that if path is a
symbolic link, then the link itself is statted, not the file that it refers to.
The elements in VALUES are the same as described by Section 8.259 [STAT],
page 275.
This intrinsic is provided in both subroutine and function forms; however, only
one form can be used in any given program unit.

Standard : GNU extension

Class: Subroutine, function

Syntax :
CALL LSTAT(NAME, VALUES [, STATUS])
STATUS = LSTAT(NAME, VALUES)

Arguments:
NAME The type shall be CHARACTER of the default kind, a
valid path within the file system.
VALUES The type shall be INTEGER(4), DIMENSION(13).
STATUS (Optional) status flag of type INTEGER(4). Returns 0
on success and a system specific error code otherwise.

Example: See Section 8.259 [STAT], page 275, for an example.

See also: To stat an open file:

Section 8.117 [FSTAT], page 190,
To stat a file:
Section 8.259 [STAT], page 275,
230 The GNU Fortran Compiler

8.183 LTIME — Convert time to local time info

Description:
Given a system time value TIME (as provided by the Section 8.269 [TIME],
page 282, intrinsic), fills VALUES with values extracted from it appropriate to
the local time zone using localtime(3).
This intrinsic routine is provided for backwards compatibility with GNU Fortran
77. In new code, programmers should consider the use of the Section 8.82
[DATE AND TIME], page 166, intrinsic defined by the Fortran 95 standard.
Standard : GNU extension
Class: Subroutine
Syntax : CALL LTIME(TIME, VALUES)
Arguments:
TIME An INTEGER scalar expression corresponding to a sys-
tem time, with INTENT(IN).
VALUES A default INTEGER array with 9 elements, with
INTENT(OUT).
Return value:
The elements of VALUES are assigned as follows:
1. Seconds after the minute, range 0–59 or 0–61 to allow for leap seconds
2. Minutes after the hour, range 0–59
3. Hours past midnight, range 0–23
4. Day of month, range 1–31
5. Number of months since January, range 0–11
6. Years since 1900
7. Number of days since Sunday, range 0–6
8. Days since January 1, range 0–365
9. Daylight savings indicator: positive if daylight savings is in effect, zero if
not, and negative if the information is not available.
See also: Section 8.82 [DATE AND TIME], page 166,
Section 8.81 [CTIME], page 165,
Section 8.131 [GMTIME], page 198,
Section 8.269 [TIME], page 282,
Section 8.270 [TIME8], page 283,

8.184 MALLOC — Allocate dynamic memory

Description:
MALLOC(SIZE) allocates SIZE bytes of dynamic memory and returns the address
of the allocated memory. The MALLOC intrinsic is an extension intended to be
used with Cray pointers, and is provided in GNU Fortran to allow the user
to compile legacy code. For new code using Fortran 95 pointers, the memory
allocation intrinsic is ALLOCATE.
Chapter 8: Intrinsic Procedures 231

Standard : GNU extension

Class: Function
Syntax : PTR = MALLOC(SIZE)
Arguments:
SIZE The type shall be INTEGER.
Return value:
The return value is of type INTEGER(K), with K such that variables of type
INTEGER(K) have the same size as C pointers (sizeof(void *)).
Example: The following example demonstrates the use of MALLOC and FREE with Cray
pointers.
program test_malloc
implicit none
integer i
real*8 x(*), z
pointer(ptr_x,x)

ptr_x = malloc(20*8)
do i = 1, 20
x(i) = sqrt(1.0d0 / i)
end do
z = 0
do i = 1, 20
z = z + x(i)
print *, z
end do
call free(ptr_x)
end program test_malloc

See also: Section 8.115 [FREE], page 188,

8.185 MASKL — Left justified mask

Description:
MASKL(I[, KIND]) has its leftmost I bits set to 1, and the remaining bits set
to 0.
Standard : Fortran 2008 and later
Class: Elemental function
Syntax : RESULT = MASKL(I[, KIND])
Arguments:
I Shall be of type INTEGER.
KIND Shall be a scalar constant expression of type INTEGER.
Return value:
The return value is of type INTEGER. If KIND is present, it specifies the kind
value of the return type; otherwise, it is of the default integer kind.
See also: Section 8.186 [MASKR], page 232,
232 The GNU Fortran Compiler

8.186 MASKR — Right justified mask

Description:
MASKL(I[, KIND]) has its rightmost I bits set to 1, and the remaining bits set
to 0.
Standard : Fortran 2008 and later
Class: Elemental function
Syntax : RESULT = MASKR(I[, KIND])
Arguments:
I Shall be of type INTEGER.
KIND Shall be a scalar constant expression of type INTEGER.
Return value:
The return value is of type INTEGER. If KIND is present, it specifies the kind
value of the return type; otherwise, it is of the default integer kind.
See also: Section 8.185 [MASKL], page 231,

8.187 MATMUL — matrix multiplication

Description:
Performs a matrix multiplication on numeric or logical arguments.
Standard : Fortran 90 and later
Class: Transformational function
Syntax : RESULT = MATMUL(MATRIX_A, MATRIX_B)
Arguments:
MATRIX A An array of INTEGER, REAL, COMPLEX, or LOGICAL
type, with a rank of one or two.
MATRIX B An array of INTEGER, REAL, or COMPLEX type if MA-
TRIX A is of a numeric type; otherwise, an array of
LOGICAL type. The rank shall be one or two, and
the first (or only) dimension of MATRIX B shall be
equal to the last (or only) dimension of MATRIX A.
MATRIX A and MATRIX B shall not both be rank
one arrays.
Return value:
The matrix product of MATRIX A and MATRIX B. The type and kind of the
result follow the usual type and kind promotion rules, as for the * or .AND.
operators.

8.188 MAX — Maximum value of an argument list

Description:
Returns the argument with the largest (most positive) value.
Chapter 8: Intrinsic Procedures 233

Standard : Fortran 77 and later

Class: Elemental function
Syntax : RESULT = MAX(A1, A2 [, A3 [, ...]])
Arguments:
A1 The type shall be INTEGER or REAL.
A2, A3, ... An expression of the same type and kind as A1. (As
a GNU extension, arguments of different kinds are
permitted.)
Return value:
The return value corresponds to the maximum value among the arguments, and
has the same type and kind as the first argument.
Specific names:
Name Argument Return type Standard
MAX0(A1) INTEGER(4) A1 INTEGER(4) Fortran 77 and later
AMAX0(A1) INTEGER(4) A1 REAL(MAX(X)) Fortran 77 and later
MAX1(A1) REAL A1 INT(MAX(X)) Fortran 77 and later
AMAX1(A1) REAL(4) A1 REAL(4) Fortran 77 and later
DMAX1(A1) REAL(8) A1 REAL(8) Fortran 77 and later
See also: Section 8.190 [MAXLOC], page 234,
Section 8.191 [MAXVAL], page 235,
Section 8.196 [MIN], page 237,

8.189 MAXEXPONENT — Maximum exponent of a real kind

Description:
MAXEXPONENT(X) returns the maximum exponent in the model of the type of X.
Standard : Fortran 90 and later
Class: Inquiry function
Syntax : RESULT = MAXEXPONENT(X)
Arguments:
X Shall be of type REAL.
Return value:
The return value is of type INTEGER and of the default integer kind.
Example:
program exponents
real(kind=4) :: x
real(kind=8) :: y

print *, minexponent(x), maxexponent(x)

print *, minexponent(y), maxexponent(y)
end program exponents
234 The GNU Fortran Compiler

8.190 MAXLOC — Location of the maximum value within an

array
Description:
Determines the location of the element in the array with the maximum value,
or, if the DIM argument is supplied, determines the locations of the maximum
element along each row of the array in the DIM direction. If MASK is present,
only the elements for which MASK is .TRUE. are considered. If more than one
element in the array has the maximum value, the location returned is that of
the first such element in array element order if the BACK is not present, or is
false; if BACK is true, the location returned is that of the last such element. If
the array has zero size, or all of the elements of MASK are .FALSE., then the
result is an array of zeroes. Similarly, if DIM is supplied and all of the elements
of MASK along a given row are zero, the result value for that row is zero.

Standard : Fortran 95 and later; ARRAY of CHARACTER and the KIND argument are avail-
able in Fortran 2003 and later. The BACK argument is available in Fortran
2008 and later.

Class: Transformational function

Syntax :
RESULT = MAXLOC(ARRAY, DIM [, MASK] [,KIND] [,BACK])
RESULT = MAXLOC(ARRAY [, MASK] [,KIND] [,BACK])

Arguments:
ARRAY Shall be an array of type INTEGER or REAL.
DIM (Optional) Shall be a scalar of type INTEGER, with a
value between one and the rank of ARRAY, inclusive.
It may not be an optional dummy argument.
MASK Shall be of type LOGICAL, and conformable with
ARRAY.
KIND (Optional) A scalar INTEGER constant expression in-
dicating the kind parameter of the result.
BACK (Optional) A scalar of type LOGICAL.

Return value:
If DIM is absent, the result is a rank-one array with a length equal to the rank
of ARRAY. If DIM is present, the result is an array with a rank one less than
the rank of ARRAY, and a size corresponding to the size of ARRAY with the
DIM dimension removed. If DIM is present and ARRAY has a rank of one,
the result is a scalar. If the optional argument KIND is present, the result is
an integer of kind KIND, otherwise it is of default kind.

See also: Section 8.108 [FINDLOC], page 183,

Section 8.188 [MAX], page 232,
Section 8.191 [MAXVAL], page 235,
Chapter 8: Intrinsic Procedures 235

8.191 MAXVAL — Maximum value of an array

Description:
Determines the maximum value of the elements in an array value, or, if the
DIM argument is supplied, determines the maximum value along each row of
the array in the DIM direction. If MASK is present, only the elements for
which MASK is .TRUE. are considered. If the array has zero size, or all of the
elements of MASK are .FALSE., then the result is -HUGE(ARRAY) if ARRAY is
numeric, or a string of nulls if ARRAY is of character type.
Standard : Fortran 90 and later
Class: Transformational function
Syntax :
RESULT = MAXVAL(ARRAY, DIM [, MASK])
RESULT = MAXVAL(ARRAY [, MASK])
Arguments:
ARRAY Shall be an array of type INTEGER or REAL.
DIM (Optional) Shall be a scalar of type INTEGER, with a
value between one and the rank of ARRAY, inclusive.
It may not be an optional dummy argument.
MASK (Optional) Shall be of type LOGICAL, and con-
formable with ARRAY.
Return value:
If DIM is absent, or if ARRAY has a rank of one, the result is a scalar. If DIM is
present, the result is an array with a rank one less than the rank of ARRAY, and
a size corresponding to the size of ARRAY with the DIM dimension removed.
In all cases, the result is of the same type and kind as ARRAY.
See also: Section 8.188 [MAX], page 232,
Section 8.190 [MAXLOC], page 234,

8.192 MCLOCK — Time function

Description:
Returns the number of clock ticks since the start of the process, based on the
function clock(3) in the C standard library.
This intrinsic is not fully portable, such as to systems with 32-bit INTEGER
types but supporting times wider than 32 bits. Therefore, the values returned
by this intrinsic might be, or become, negative, or numerically less than previous
values, during a single run of the compiled program.
Standard : GNU extension
Class: Function
Syntax : RESULT = MCLOCK()
236 The GNU Fortran Compiler

Return value:
The return value is a scalar of type INTEGER(4), equal to the number of clock
ticks since the start of the process, or -1 if the system does not support
clock(3).
See also: Section 8.81 [CTIME], page 165,
Section 8.131 [GMTIME], page 198,
Section 8.183 [LTIME], page 230,
Section 8.192 [MCLOCK], page 235,
Section 8.269 [TIME], page 282,

8.193 MCLOCK8 — Time function (64-bit)

Description:
Returns the number of clock ticks since the start of the process, based on the
function clock(3) in the C standard library.
Warning: this intrinsic does not increase the range of the timing values over
that returned by clock(3). On a system with a 32-bit clock(3), MCLOCK8 will
return a 32-bit value, even though it is converted to a 64-bit INTEGER(8) value.
That means overflows of the 32-bit value can still occur. Therefore, the values
returned by this intrinsic might be or become negative or numerically less than
previous values during a single run of the compiled program.
Standard : GNU extension
Class: Function
Syntax : RESULT = MCLOCK8()
Return value:
The return value is a scalar of type INTEGER(8), equal to the number of clock
ticks since the start of the process, or -1 if the system does not support
clock(3).
See also: Section 8.81 [CTIME], page 165,
Section 8.131 [GMTIME], page 198,
Section 8.183 [LTIME], page 230,
Section 8.192 [MCLOCK], page 235,
Section 8.270 [TIME8], page 283,

8.194 MERGE — Merge variables

Description:
Select values from two arrays according to a logical mask. The result is equal
to TSOURCE if MASK is .TRUE., or equal to FSOURCE if it is .FALSE..
Standard : Fortran 90 and later
Class: Elemental function
Syntax : RESULT = MERGE(TSOURCE, FSOURCE, MASK)
Chapter 8: Intrinsic Procedures 237

Arguments:
TSOURCE May be of any type.
FSOURCE Shall be of the same type and type parameters as
TSOURCE.
MASK Shall be of type LOGICAL.
Return value:
The result is of the same type and type parameters as TSOURCE.

8.195 MERGE_BITS — Merge of bits under mask

Description:
MERGE_BITS(I, J, MASK) merges the bits of I and J as determined by the mask.
The i-th bit of the result is equal to the i-th bit of I if the i-th bit of MASK is
1; it is equal to the i-th bit of J otherwise.
Standard : Fortran 2008 and later
Class: Elemental function
Syntax : RESULT = MERGE_BITS(I, J, MASK)
Arguments:
I Shall be of type INTEGER or a boz-literal-constant.
J Shall be of type INTEGER with the same kind type
parameter as I or a boz-literal-constant. I and J
shall not both be boz-literal-constants.
MASK Shall be of type INTEGER or a boz-literal-constant and
of the same kind as I.
Return value:
The result is of the same type and kind as I.

8.196 MIN — Minimum value of an argument list

Description:
Returns the argument with the smallest (most negative) value.
Standard : Fortran 77 and later
Class: Elemental function
Syntax : RESULT = MIN(A1, A2 [, A3, ...])
Arguments:
A1 The type shall be INTEGER or REAL.
A2, A3, ... An expression of the same type and kind as A1. (As
a GNU extension, arguments of different kinds are
permitted.)
Return value:
The return value corresponds to the minimum value among the arguments, and
has the same type and kind as the first argument.
238 The GNU Fortran Compiler

Specific names:
Name Argument Return type Standard
MIN0(A1) INTEGER(4) A1 INTEGER(4) Fortran 77 and later
AMIN0(A1) INTEGER(4) A1 REAL(4) Fortran 77 and later
MIN1(A1) REAL A1 INTEGER(4) Fortran 77 and later
AMIN1(A1) REAL(4) A1 REAL(4) Fortran 77 and later
DMIN1(A1) REAL(8) A1 REAL(8) Fortran 77 and later
See also: Section 8.188 [MAX], page 232,
Section 8.198 [MINLOC], page 238,
Section 8.199 [MINVAL], page 239,

8.197 MINEXPONENT — Minimum exponent of a real kind

Description:
MINEXPONENT(X) returns the minimum exponent in the model of the type of X.
Standard : Fortran 90 and later
Class: Inquiry function
Syntax : RESULT = MINEXPONENT(X)
Arguments:
X Shall be of type REAL.
Return value:
The return value is of type INTEGER and of the default integer kind.
Example: See MAXEXPONENT for an example.

8.198 MINLOC — Location of the minimum value within an

array
Description:
Determines the location of the element in the array with the minimum value,
or, if the DIM argument is supplied, determines the locations of the minimum
element along each row of the array in the DIM direction. If MASK is present,
only the elements for which MASK is .TRUE. are considered. If more than one
element in the array has the minimum value, the location returned is that of
the first such element in array element order if the BACK is not present, or is
false; if BACK is true, the location returned is that of the last such element. If
the array has zero size, or all of the elements of MASK are .FALSE., then the
result is an array of zeroes. Similarly, if DIM is supplied and all of the elements
of MASK along a given row are zero, the result value for that row is zero.
Standard : Fortran 90 and later; ARRAY of CHARACTER and the KIND argument are avail-
able in Fortran 2003 and later. The BACK argument is available in Fortran
2008 and later.
Class: Transformational function
Chapter 8: Intrinsic Procedures 239

Syntax :
RESULT = MINLOC(ARRAY, DIM [, MASK] [,KIND] [,BACK])
RESULT = MINLOC(ARRAY [, MASK], [,KIND] [,BACK])
Arguments:
ARRAY Shall be an array of type INTEGER, REAL or
CHARACTER.
DIM (Optional) Shall be a scalar of type INTEGER, with a
value between one and the rank of ARRAY, inclusive.
It may not be an optional dummy argument.
MASK Shall be of type LOGICAL, and conformable with
ARRAY.
KIND (Optional) A scalar INTEGER constant expression in-
dicating the kind parameter of the result.
BACK (Optional) A scalar of type LOGICAL.
Return value:
If DIM is absent, the result is a rank-one array with a length equal to the rank
of ARRAY. If DIM is present, the result is an array with a rank one less than
the rank of ARRAY, and a size corresponding to the size of ARRAY with the
DIM dimension removed. If DIM is present and ARRAY has a rank of one,
the result is a scalar. If the optional argument KIND is present, the result is
an integer of kind KIND, otherwise it is of default kind.
See also: Section 8.108 [FINDLOC], page 183,
Section 8.196 [MIN], page 237,
Section 8.199 [MINVAL], page 239,

8.199 MINVAL — Minimum value of an array

Description:
Determines the minimum value of the elements in an array value, or, if the
DIM argument is supplied, determines the minimum value along each row of
the array in the DIM direction. If MASK is present, only the elements for
which MASK is .TRUE. are considered. If the array has zero size, or all of the
elements of MASK are .FALSE., then the result is HUGE(ARRAY) if ARRAY is
numeric, or a string of CHAR(255) characters if ARRAY is of character type.
Standard : Fortran 90 and later
Class: Transformational function
Syntax :
RESULT = MINVAL(ARRAY, DIM [, MASK])
RESULT = MINVAL(ARRAY [, MASK])
Arguments:
ARRAY Shall be an array of type INTEGER or REAL.
DIM (Optional) Shall be a scalar of type INTEGER, with a
value between one and the rank of ARRAY, inclusive.
It may not be an optional dummy argument.
240 The GNU Fortran Compiler

MASK Shall be of type LOGICAL, and conformable with

ARRAY.
Return value:
If DIM is absent, or if ARRAY has a rank of one, the result is a scalar. If DIM is
present, the result is an array with a rank one less than the rank of ARRAY, and
a size corresponding to the size of ARRAY with the DIM dimension removed.
In all cases, the result is of the same type and kind as ARRAY.
See also: Section 8.196 [MIN], page 237,
Section 8.198 [MINLOC], page 238,

8.200 MOD — Remainder function

Description:
MOD(A,P) computes the remainder of the division of A by P.
Standard : Fortran 77 and later, has overloads that are GNU extensions
Class: Elemental function
Syntax : RESULT = MOD(A, P)
Arguments:
A Shall be a scalar of type INTEGER or REAL.
P Shall be a scalar of the same type and kind as A and
not equal to zero. (As a GNU extension, arguments
of different kinds are permitted.)
Return value:
The return value is the result of A - (INT(A/P) * P). The type and kind of
the return value is the same as that of the arguments. The returned value has
the same sign as A and a magnitude less than the magnitude of P. (As a GNU
extension, kind is the largest kind of the actual arguments.)
Example:
program test_mod
print *, mod(17,3)
print *, mod(17.5,5.5)
print *, mod(17.5d0,5.5)
print *, mod(17.5,5.5d0)

print *, mod(-17,3)
print *, mod(-17.5,5.5)
print *, mod(-17.5d0,5.5)
print *, mod(-17.5,5.5d0)

print *, mod(17,-3)
print *, mod(17.5,-5.5)
print *, mod(17.5d0,-5.5)
print *, mod(17.5,-5.5d0)
end program test_mod

Specific names:
Name Arguments Return type Standard
MOD(A,P) INTEGER A,P INTEGER Fortran 77 and later
Chapter 8: Intrinsic Procedures 241

AMOD(A,P) REAL(4) A,P REAL(4) Fortran 77 and later

DMOD(A,P) REAL(8) A,P REAL(8) Fortran 77 and later
BMOD(A,P) INTEGER(1) A,P INTEGER(1) GNU extension
IMOD(A,P) INTEGER(2) A,P INTEGER(2) GNU extension
JMOD(A,P) INTEGER(4) A,P INTEGER(4) GNU extension
KMOD(A,P) INTEGER(8) A,P INTEGER(8) GNU extension
See also: Section 8.201 [MODULO], page 241,

8.201 MODULO — Modulo function

Description:
MODULO(A,P) computes the A modulo P.
Standard : Fortran 95 and later
Class: Elemental function
Syntax : RESULT = MODULO(A, P)
Arguments:
A Shall be a scalar of type INTEGER or REAL.
P Shall be a scalar of the same type and kind as A. It
shall not be zero. (As a GNU extension, arguments
of different kinds are permitted.)
Return value:
The type and kind of the result are those of the arguments. (As a GNU exten-
sion, kind is the largest kind of the actual arguments.)
If A and P are of type INTEGER:
MODULO(A,P) has the value R such that A=Q*P+R, where Q is an
integer and R is between 0 (inclusive) and P (exclusive).
If A and P are of type REAL:
MODULO(A,P) has the value of A - FLOOR (A / P) * P.
The returned value has the same sign as P and a magnitude less than the
magnitude of P.
Example:
program test_modulo
print *, modulo(17,3)
print *, modulo(17.5,5.5)

print *, modulo(-17,3)
print *, modulo(-17.5,5.5)

print *, modulo(17,-3)
print *, modulo(17.5,-5.5)
end program

See also: Section 8.200 [MOD], page 240,

242 The GNU Fortran Compiler

8.202 MOVE_ALLOC — Move allocation from one object to

another
Description:
MOVE_ALLOC(FROM, TO) moves the allocation from FROM to TO. FROM will
become deallocated in the process.
Standard : Fortran 2003 and later
Class: Pure subroutine
Syntax : CALL MOVE_ALLOC(FROM, TO)
Arguments:
FROM ALLOCATABLE, INTENT(INOUT), may be of any type
and kind.
TO ALLOCATABLE, INTENT(OUT), shall be of the same
type, kind and rank as FROM.
Return value:
None
Example:
program test_move_alloc
integer, allocatable :: a(:), b(:)

allocate(a(3))
a = [ 1, 2, 3 ]
call move_alloc(a, b)
print *, allocated(a), allocated(b)
print *, b
end program test_move_alloc

8.203 MVBITS — Move bits from one integer to another

Description:
Moves LEN bits from positions FROMPOS through FROMPOS+LEN-1 of FROM
to positions TOPOS through TOPOS+LEN-1 of TO. The portion of argument TO
not affected by the movement of bits is unchanged. The values of FROMPOS+LEN-
1 and TOPOS+LEN-1 must be less than BIT_SIZE(FROM).
Standard : Fortran 90 and later, has overloads that are GNU extensions
Class: Elemental subroutine
Syntax : CALL MVBITS(FROM, FROMPOS, LEN, TO, TOPOS)
Arguments:
FROM The type shall be INTEGER.
FROMPOS The type shall be INTEGER.
LEN The type shall be INTEGER.
TO The type shall be INTEGER, of the same kind as
FROM.
TOPOS The type shall be INTEGER.
Chapter 8: Intrinsic Procedures 243

Specific names:
Name Argument Return type Standard
MVBITS(A) INTEGER A INTEGER Fortran 90 and later
BMVBITS(A) INTEGER(1) A INTEGER(1) GNU extension
IMVBITS(A) INTEGER(2) A INTEGER(2) GNU extension
JMVBITS(A) INTEGER(4) A INTEGER(4) GNU extension
KMVBITS(A) INTEGER(8) A INTEGER(8) GNU extension
See also: Section 8.140 [IBCLR], page 204,
Section 8.142 [IBSET], page 206,
Section 8.141 [IBITS], page 205,
Section 8.137 [IAND], page 202,
Section 8.152 [IOR], page 212,
Section 8.145 [IEOR], page 208,

8.204 NEAREST — Nearest representable number

Description:
NEAREST(X, S) returns the processor-representable number nearest to X in the
direction indicated by the sign of S.
Standard : Fortran 90 and later
Class: Elemental function
Syntax : RESULT = NEAREST(X, S)
Arguments:
X Shall be of type REAL.
S Shall be of type REAL and not equal to zero.
Return value:
The return value is of the same type as X. If S is positive, NEAREST returns
the processor-representable number greater than X and nearest to it. If S is
negative, NEAREST returns the processor-representable number smaller than X
and nearest to it.
Example:
program test_nearest
real :: x, y
x = nearest(42.0, 1.0)
y = nearest(42.0, -1.0)
write (*,"(3(G20.15))") x, y, x - y
end program test_nearest

8.205 NEW_LINE — New line character

Description:
NEW_LINE(C) returns the new-line character.
Standard : Fortran 2003 and later
Class: Inquiry function
244 The GNU Fortran Compiler

Syntax : RESULT = NEW_LINE(C)

Arguments:
C The argument shall be a scalar or array of the type
CHARACTER.
Return value:
Returns a CHARACTER scalar of length one with the new-line character of
the same kind as parameter C.
Example:
program newline
implicit none
write(*,'(A)') 'This is record 1.'//NEW_LINE('A')//'This is record 2.'
end program newline

8.206 NINT — Nearest whole number

Description:
NINT(A) rounds its argument to the nearest whole number.
Standard : Fortran 77 and later, with KIND argument Fortran 90 and later
Class: Elemental function
Syntax : RESULT = NINT(A [, KIND])
Arguments:
A The type of the argument shall be REAL.
KIND (Optional) A scalar INTEGER constant expression in-
dicating the kind parameter of the result.
Return value:
Returns A with the fractional portion of its magnitude eliminated by round-
ing to the nearest whole number and with its sign preserved, converted to an
INTEGER of the default kind.
Example:
program test_nint
real(4) x4
real(8) x8
x4 = 1.234E0_4
x8 = 4.321_8
print *, nint(x4), idnint(x8)
end program test_nint

Specific names:
Name Argument Return Type Standard
NINT(A) REAL(4) A INTEGER Fortran 77 and later
IDNINT(A) REAL(8) A INTEGER Fortran 77 and later
See also: Section 8.58 [CEILING], page 149,
Section 8.109 [FLOOR], page 184,
Chapter 8: Intrinsic Procedures 245

8.207 NORM2 — Euclidean vector norms

Description:
Calculates the Euclidean vector norm (L2 norm) of ARRAY along dimension
DIM.
Standard : Fortran 2008 and later
Class: Transformational function
Syntax :
RESULT = NORM2(ARRAY[, DIM])
Arguments:
ARRAY Shall be an array of type REAL
DIM (Optional) shall be a scalar of type INTEGER with a
value in the range from 1 to n, where n equals the
rank of ARRAY.
Return value:
The result is of the same type as ARRAY.
If DIM is absent, a scalar with the square root of the sum of all elements in
ARRAY squared is returned. Otherwise, an array of rank n − 1, where n equals
the rank of ARRAY, and a shape similar to that of ARRAY with dimension
DIM dropped is returned.
Example:
PROGRAM test_sum
REAL :: x(5) = [ real :: 1, 2, 3, 4, 5 ]
print *, NORM2(x) ! = sqrt(55.) ~ 7.416
END PROGRAM

8.208 NOT — Logical negation

Description:
NOT returns the bitwise Boolean inverse of I.
Standard : Fortran 90 and later, has overloads that are GNU extensions
Class: Elemental function
Syntax : RESULT = NOT(I)
Arguments:
I The type shall be INTEGER.
Return value:
The return type is INTEGER, of the same kind as the argument.
Specific names:
Name Argument Return type Standard
NOT(A) INTEGER A INTEGER Fortran 95 and later
BNOT(A) INTEGER(1) A INTEGER(1) GNU extension
INOT(A) INTEGER(2) A INTEGER(2) GNU extension
JNOT(A) INTEGER(4) A INTEGER(4) GNU extension
KNOT(A) INTEGER(8) A INTEGER(8) GNU extension
246 The GNU Fortran Compiler

See also: Section 8.137 [IAND], page 202,

Section 8.145 [IEOR], page 208,
Section 8.152 [IOR], page 212,
Section 8.141 [IBITS], page 205,
Section 8.142 [IBSET], page 206,
Section 8.140 [IBCLR], page 204,

8.209 NULL — Function that returns an disassociated pointer

Description:
Returns a disassociated pointer.
If MOLD is present, a disassociated pointer of the same type is returned, oth-
erwise the type is determined by context.
In Fortran 95, MOLD is optional. Please note that Fortran 2003 includes cases
where it is required.
Standard : Fortran 95 and later
Class: Transformational function
Syntax : PTR => NULL([MOLD])
Arguments:
MOLD (Optional) shall be a pointer of any association status
and of any type.
Return value:
A disassociated pointer.
Example:
REAL, POINTER, DIMENSION(:) :: VEC => NULL ()

See also: Section 8.22 [ASSOCIATED], page 124,

8.210 NUM_IMAGES — Function that returns the number of

images
Description:
Returns the number of images.
Standard : Fortran 2008 and later. With DISTANCE or FAILED argument, Technical
Specification (TS) 18508 or later
Class: Transformational function
Syntax : RESULT = NUM_IMAGES(DISTANCE, FAILED)
Arguments:
DISTANCE (optional, intent(in)) Nonnegative scalar integer
FAILED (optional, intent(in)) Scalar logical expression
Return value:
Scalar default-kind integer. If DISTANCE is not present or has value 0, the
number of images in the current team is returned. For values smaller or equal
Chapter 8: Intrinsic Procedures 247

distance to the initial team, it returns the number of images index on the
ancestor team which has a distance of DISTANCE from the invoking team. If
DISTANCE is larger than the distance to the initial team, the number of images
of the initial team is returned. If FAILED is not present the total number of
images is returned; if it has the value .TRUE., the number of failed images is
returned, otherwise, the number of images which do have not the failed status.
Example:
INTEGER :: value[*]
INTEGER :: i
value = THIS_IMAGE()
SYNC ALL
IF (THIS_IMAGE() == 1) THEN
DO i = 1, NUM_IMAGES()
WRITE(*,'(2(a,i0))') 'value[', i, '] is ', value[i]
END DO
END IF

See also: Section 8.268 [THIS IMAGE], page 281,

Section 8.147 [IMAGE INDEX], page 209,

8.211 OR — Bitwise logical OR

Description:
Bitwise logical OR.
This intrinsic routine is provided for backwards compatibility with GNU For-
tran 77. For integer arguments, programmers should consider the use of the
Section 8.152 [IOR], page 212, intrinsic defined by the Fortran standard.
Standard : GNU extension
Class: Function
Syntax : RESULT = OR(I, J)
Arguments:
I The type shall be either a scalar INTEGER type or a
scalar LOGICAL type or a boz-literal-constant.
J The type shall be the same as the type of I or a boz-
literal-constant. I and J shall not both be boz-literal-
constants. If either I and J is a boz-literal-constant,
then the other argument must be a scalar INTEGER.
Return value:
The return type is either a scalar INTEGER or a scalar LOGICAL. If the kind type
parameters differ, then the smaller kind type is implicitly converted to larger
kind, and the return has the larger kind. A boz-literal-constant is converted to
an INTEGER with the kind type parameter of the other argument as-if a call to
Section 8.149 [INT], page 210, occurred.
Example:
PROGRAM test_or
LOGICAL :: T = .TRUE., F = .FALSE.
248 The GNU Fortran Compiler

INTEGER :: a, b
DATA a / Z'F' /, b / Z'3' /

WRITE (*,*) OR(T, T), OR(T, F), OR(F, T), OR(F, F)

WRITE (*,*) OR(a, b)
END PROGRAM

See also: Fortran 95 elemental function:

Section 8.152 [IOR], page 212,

8.212 PACK — Pack an array into an array of rank one

Description:
Stores the elements of ARRAY in an array of rank one.
The beginning of the resulting array is made up of elements whose MASK equals
TRUE. Afterwards, positions are filled with elements taken from VECTOR.
Standard : Fortran 90 and later
Class: Transformational function
Syntax : RESULT = PACK(ARRAY, MASK[,VECTOR])
Arguments:
ARRAY Shall be an array of any type.
MASK Shall be an array of type LOGICAL and of the same
size as ARRAY. Alternatively, it may be a LOGICAL
scalar.
VECTOR (Optional) shall be an array of the same type as AR-
RAY and of rank one. If present, the number of ele-
ments in VECTOR shall be equal to or greater than
the number of true elements in MASK. If MASK is
scalar, the number of elements in VECTOR shall be
equal to or greater than the number of elements in
ARRAY.
Return value:
The result is an array of rank one and the same type as that of ARRAY. If
VECTOR is present, the result size is that of VECTOR, the number of TRUE
values in MASK otherwise.
Example: Gathering nonzero elements from an array:
PROGRAM test_pack_1
INTEGER :: m(6)
m = (/ 1, 0, 0, 0, 5, 0 /)
WRITE(*, FMT="(6(I0, ' '))") pack(m, m /= 0) ! "1 5"
END PROGRAM
Gathering nonzero elements from an array and appending elements from VEC-
TOR:
PROGRAM test_pack_2
INTEGER :: m(4)
m = (/ 1, 0, 0, 2 /)
! The following results in "1 2 3 4"
Chapter 8: Intrinsic Procedures 249

WRITE(*, FMT="(4(I0, ' '))") pack(m, m /= 0, (/ 0, 0, 3, 4 /))

END PROGRAM

See also: Section 8.281 [UNPACK], page 289,

8.213 PARITY — Reduction with exclusive OR

Description:
Calculates the parity, i.e. the reduction using .XOR., of MASK along dimension
DIM.
Standard : Fortran 2008 and later
Class: Transformational function
Syntax :
RESULT = PARITY(MASK[, DIM])
Arguments:
MASK Shall be an array of type LOGICAL
DIM (Optional) shall be a scalar of type INTEGER with a
value in the range from 1 to n, where n equals the
rank of MASK.
Return value:
The result is of the same type as MASK.
If DIM is absent, a scalar with the parity of all elements in MASK is returned,
i.e. true if an odd number of elements is .true. and false otherwise. If DIM
is present, an array of rank n − 1, where n equals the rank of ARRAY, and a
shape similar to that of MASK with dimension DIM dropped is returned.
Example:
PROGRAM test_sum
LOGICAL :: x(2) = [ .true., .false. ]
print *, PARITY(x) ! prints "T" (true).
END PROGRAM

8.214 PERROR — Print system error message

Description:
Prints (on the C stderr stream) a newline-terminated error message corre-
sponding to the last system error. This is prefixed by STRING, a colon and a
space. See perror(3).
Standard : GNU extension
Class: Subroutine
Syntax : CALL PERROR(STRING)
Arguments:
STRING A scalar of type CHARACTER and of the default kind.
See also: Section 8.146 [IERRNO], page 209,
250 The GNU Fortran Compiler

8.215 POPCNT — Number of bits set

Description:
POPCNT(I) returns the number of bits set (’1’ bits) in the binary representation
of I.
Standard : Fortran 2008 and later
Class: Elemental function
Syntax : RESULT = POPCNT(I)
Arguments:
I Shall be of type INTEGER.
Return value:
The return value is of type INTEGER and of the default integer kind.
Example:
program test_population
print *, popcnt(127), poppar(127)
print *, popcnt(huge(0_4)), poppar(huge(0_4))
print *, popcnt(huge(0_8)), poppar(huge(0_8))
end program test_population

See also: Section 8.216 [POPPAR], page 250,

Section 8.167 [LEADZ], page 220,
Section 8.272 [TRAILZ], page 284,

8.216 POPPAR — Parity of the number of bits set

Description:
POPPAR(I) returns parity of the integer I, i.e. the parity of the number of bits
set (’1’ bits) in the binary representation of I. It is equal to 0 if I has an even
number of bits set, and 1 for an odd number of ’1’ bits.
Standard : Fortran 2008 and later
Class: Elemental function
Syntax : RESULT = POPPAR(I)
Arguments:
I Shall be of type INTEGER.
Return value:
The return value is of type INTEGER and of the default integer kind.
Example:
program test_population
print *, popcnt(127), poppar(127)
print *, popcnt(huge(0_4)), poppar(huge(0_4))
print *, popcnt(huge(0_8)), poppar(huge(0_8))
end program test_population

See also: Section 8.215 [POPCNT], page 250,

Section 8.167 [LEADZ], page 220,
Section 8.272 [TRAILZ], page 284,
Chapter 8: Intrinsic Procedures 251

8.217 PRECISION — Decimal precision of a real kind

Description:
PRECISION(X) returns the decimal precision in the model of the type of X.
Standard : Fortran 90 and later
Class: Inquiry function
Syntax : RESULT = PRECISION(X)
Arguments:
X Shall be of type REAL or COMPLEX. It may be scalar
or valued.
Return value:
The return value is of type INTEGER and of the default integer kind.
Example:
program prec_and_range
real(kind=4) :: x(2)
complex(kind=8) :: y

print *, precision(x), range(x)

print *, precision(y), range(y)
end program prec_and_range

See also: Section 8.241 [SELECTED REAL KIND], page 264,

Section 8.226 [RANGE], page 256,

8.218 PRESENT — Determine whether an optional dummy

argument is specified
Description:
Determines whether an optional dummy argument is present.
Standard : Fortran 90 and later
Class: Inquiry function
Syntax : RESULT = PRESENT(A)
Arguments:
A May be of any type and may be a pointer, scalar
or array value, or a dummy procedure. It shall be
the name of an optional dummy argument accessible
within the current subroutine or function.
Return value:
Returns either TRUE if the optional argument A is present, or FALSE otherwise.
Example:
PROGRAM test_present
WRITE(*,*) f(), f(42) ! "F T"
CONTAINS
LOGICAL FUNCTION f(x)
252 The GNU Fortran Compiler

INTEGER, INTENT(IN), OPTIONAL :: x

f = PRESENT(x)
END FUNCTION
END PROGRAM

8.219 PRODUCT — Product of array elements

Description:
Multiplies the elements of ARRAY along dimension DIM if the corresponding
element in MASK is TRUE.
Standard : Fortran 90 and later
Class: Transformational function
Syntax :
RESULT = PRODUCT(ARRAY[, MASK])
RESULT = PRODUCT(ARRAY, DIM[, MASK])
Arguments:
ARRAY Shall be an array of type INTEGER, REAL or COMPLEX.
DIM (Optional) shall be a scalar of type INTEGER with a
value in the range from 1 to n, where n equals the
rank of ARRAY.
MASK (Optional) shall be of type LOGICAL and either be a
scalar or an array of the same shape as ARRAY.
Return value:
The result is of the same type as ARRAY.
If DIM is absent, a scalar with the product of all elements in ARRAY is re-
turned. Otherwise, an array of rank n-1, where n equals the rank of ARRAY,
and a shape similar to that of ARRAY with dimension DIM dropped is re-
turned.
Example:
PROGRAM test_product
INTEGER :: x(5) = (/ 1, 2, 3, 4 ,5 /)
print *, PRODUCT(x) ! all elements, product = 120
print *, PRODUCT(x, MASK=MOD(x, 2)==1) ! odd elements, product = 15
END PROGRAM

See also: Section 8.261 [SUM], page 277,

8.220 RADIX — Base of a model number

Description:
RADIX(X) returns the base of the model representing the entity X.
Standard : Fortran 90 and later
Class: Inquiry function
Syntax : RESULT = RADIX(X)
Chapter 8: Intrinsic Procedures 253

Arguments:
X Shall be of type INTEGER or REAL
Return value:
The return value is a scalar of type INTEGER and of the default integer kind.
Example:
program test_radix
print *, "The radix for the default integer kind is", radix(0)
print *, "The radix for the default real kind is", radix(0.0)
end program test_radix

See also: Section 8.241 [SELECTED REAL KIND], page 264,

8.221 RAN — Real pseudo-random number

Description:
For compatibility with HP FORTRAN 77/iX, the RAN intrinsic is provided as
an alias for RAND. See Section 8.222 [RAND], page 253, for complete documen-
tation.
Standard : GNU extension
Class: Function
See also: Section 8.222 [RAND], page 253,
Section 8.224 [RANDOM NUMBER], page 255,

8.222 RAND — Real pseudo-random number

Description:
RAND(FLAG) returns a pseudo-random number from a uniform distribution be-
tween 0 and 1. If FLAG is 0, the next number in the current sequence is
returned; if FLAG is 1, the generator is restarted by CALL SRAND(0); if FLAG
has any other value, it is used as a new seed with SRAND.
This intrinsic routine is provided for backwards compatibility with GNU Fortran
77. It implements a simple modulo generator as provided by g77. For new code,
one should consider the use of Section 8.224 [RANDOM NUMBER], page 255,
as it implements a superior algorithm.
Standard : GNU extension
Class: Function
Syntax : RESULT = RAND(I)
Arguments:
I Shall be a scalar INTEGER of kind 4.
Return value:
The return value is of REAL type and the default kind.
Example:
program test_rand
254 The GNU Fortran Compiler

integer,parameter :: seed = 86456

call srand(seed)
print *, rand(), rand(), rand(), rand()
print *, rand(seed), rand(), rand(), rand()
end program test_rand

See also: Section 8.258 [SRAND], page 275,

Section 8.224 [RANDOM NUMBER], page 255,

8.223 RANDOM_INIT — Initialize a pseudo-random number

generator
Description:
Initializes the state of the pseudorandom number generator used by RANDOM_
NUMBER.
Standard : Fortran 2018
Class: Subroutine
Syntax : CALL RANDOM_INIT(REPEATABLE, IMAGE_DISTINCT)
Arguments:
REPEATABLE Shall be a scalar with a LOGICAL type, and it is
INTENT(IN). If it is .true., the seed is set to a
processor-dependent value that is the same each time
RANDOM_INIT is called from the same image. The
term “same image” means a single instance of pro-
gram execution. The sequence of random numbers is
different for repeated execution of the program. If it
is .false., the seed is set to a processor-dependent
value.
IMAGE DISTINCT Shall be a scalar with a LOGICAL type, and it is
INTENT(IN). If it is .true., the seed is set to a
processor-dependent value that is distinct from th
seed set by a call to RANDOM_INIT in another image.
If it is .false., the seed is set to a value that does
depend which image called RANDOM_INIT.
Example:
program test_random_seed
implicit none
real x(3), y(3)
call random_init(.true., .true.)
call random_number(x)
call random_init(.true., .true.)
call random_number(y)
! x and y are the same sequence
if (any(x /= y)) call abort
end program test_random_seed

See also: Section 8.224 [RANDOM NUMBER], page 255,

Section 8.225 [RANDOM SEED], page 255,
Chapter 8: Intrinsic Procedures 255

8.224 RANDOM_NUMBER — Pseudo-random number

Description:
Returns a single pseudorandom number or an array of pseudorandom numbers
from the uniform distribution over the range 0 ≤ x < 1.
The runtime-library implements the xoshiro256** pseudorandom number gen-
erator (PRNG). This generator has a period of 2256 −1, and when using multiple
threads up to 2128 threads can each generate 2128 random numbers before any
aliasing occurs.
Note that in a multi-threaded program (e.g. using OpenMP directives), each
thread will have its own random number state. For details of the seeding
procedure, see the documentation for the RANDOM_SEED intrinsic.
Standard : Fortran 90 and later
Class: Subroutine
Syntax : CALL RANDOM_NUMBER(HARVEST)
Arguments:
HARVEST Shall be a scalar or an array of type REAL.
Example:
program test_random_number
REAL :: r(5,5)
CALL RANDOM_NUMBER(r)
end program

See also: Section 8.225 [RANDOM SEED], page 255,

Section 8.223 [RANDOM INIT], page 254,

8.225 RANDOM_SEED — Initialize a pseudo-random number

sequence
Description:
Restarts or queries the state of the pseudorandom number generator used by
RANDOM_NUMBER.
If RANDOM_SEED is called without arguments, it is seeded with random data
retrieved from the operating system.
As an extension to the Fortran standard, the GFortran RANDOM_NUMBER sup-
ports multiple threads. Each thread in a multi-threaded program has its own
seed. When RANDOM_SEED is called either without arguments or with the PUT
argument, the given seed is copied into a master seed as well as the seed of the
current thread. When a new thread uses RANDOM_NUMBER for the first time, the
seed is copied from the master seed, and forwarded N ∗ 2128 steps to guarantee
that the random stream does not alias any other stream in the system, where
N is the number of threads that have used RANDOM_NUMBER so far during the
program execution.
Standard : Fortran 90 and later
Class: Subroutine
256 The GNU Fortran Compiler

Syntax : CALL RANDOM_SEED([SIZE, PUT, GET])

Arguments:
SIZE (Optional) Shall be a scalar and of type default
INTEGER, with INTENT(OUT). It specifies the mini-
mum size of the arrays used with the PUT and GET
arguments.
PUT (Optional) Shall be an array of type default INTEGER
and rank one. It is INTENT(IN) and the size of the
array must be larger than or equal to the number
returned by the SIZE argument.
GET (Optional) Shall be an array of type default INTEGER
and rank one. It is INTENT(OUT) and the size of the
array must be larger than or equal to the number
returned by the SIZE argument.
Example:
program test_random_seed
implicit none
integer, allocatable :: seed(:)
integer :: n

call random_seed(size = n)
allocate(seed(n))
call random_seed(get=seed)
write (*, *) seed
end program test_random_seed

See also: Section 8.224 [RANDOM NUMBER], page 255,

Section 8.223 [RANDOM INIT], page 254,

8.226 RANGE — Decimal exponent range

Description:
RANGE(X) returns the decimal exponent range in the model of the type of X.
Standard : Fortran 90 and later
Class: Inquiry function
Syntax : RESULT = RANGE(X)
Arguments:
X Shall be of type INTEGER, REAL or COMPLEX.
Return value:
The return value is of type INTEGER and of the default integer kind.
Example: See PRECISION for an example.
See also: Section 8.241 [SELECTED REAL KIND], page 264,
Section 8.217 [PRECISION], page 251,
Chapter 8: Intrinsic Procedures 257

8.227 RANK — Rank of a data object

Description:
RANK(A) returns the rank of a scalar or array data object.
Standard : Technical Specification (TS) 29113
Class: Inquiry function
Syntax : RESULT = RANK(A)
Arguments:
A can be of any type
Return value:
The return value is of type INTEGER and of the default integer kind. For arrays,
their rank is returned; for scalars zero is returned.
Example:
program test_rank
integer :: a
real, allocatable :: b(:,:)

print *, rank(a), rank(b) ! Prints: 0 2

end program test_rank

8.228 REAL — Convert to real type

Description:
REAL(A [, KIND]) converts its argument A to a real type. The REALPART func-
tion is provided for compatibility with g77, and its use is strongly discouraged.
Standard : Fortran 77 and later, with KIND argument Fortran 90 and later, has GNU
extensions
Class: Elemental function
Syntax :
RESULT = REAL(A [, KIND])
RESULT = REALPART(Z)
Arguments:
A Shall be INTEGER, REAL, or COMPLEX.
KIND (Optional) A scalar INTEGER constant expression in-
dicating the kind parameter of the result.
Return value:
These functions return a REAL variable or array under the following rules:
(A) REAL(A) is converted to a default real type if A is an integer or real
variable.
(B) REAL(A) is converted to a real type with the kind type parameter
of A if A is a complex variable.
(C) REAL(A, KIND) is converted to a real type with kind type parameter
KIND if A is a complex, integer, or real variable.
258 The GNU Fortran Compiler

Example:
program test_real
complex :: x = (1.0, 2.0)
print *, real(x), real(x,8), realpart(x)
end program test_real

Specific names:
Name Argument Return type Standard
FLOAT(A) INTEGER(4) REAL(4) Fortran 77 and later
DFLOAT(A) INTEGER(4) REAL(8) GNU extension
FLOATI(A) INTEGER(2) REAL(4) GNU extension (-fdec)
FLOATJ(A) INTEGER(4) REAL(4) GNU extension (-fdec)
FLOATK(A) INTEGER(8) REAL(4) GNU extension (-fdec)
SNGL(A) REAL(8) REAL(4) Fortran 77 and later
See also: Section 8.83 [DBLE], page 167,

8.229 RENAME — Rename a file

Description:
Renames a file from file PATH1 to PATH2. A null character (CHAR(0)) can be
used to mark the end of the names in PATH1 and PATH2; otherwise, trailing
blanks in the file names are ignored. If the STATUS argument is supplied, it
contains 0 on success or a nonzero error code upon return; see rename(2).
This intrinsic is provided in both subroutine and function forms; however, only
one form can be used in any given program unit.
Standard : GNU extension
Class: Subroutine, function
Syntax :
CALL RENAME(PATH1, PATH2 [, STATUS])
STATUS = RENAME(PATH1, PATH2)
Arguments:
PATH1 Shall be of default CHARACTER type.
PATH2 Shall be of default CHARACTER type.
STATUS (Optional) Shall be of default INTEGER type.
See also: Section 8.172 [LINK], page 223,

8.230 REPEAT — Repeated string concatenation

Description:
Concatenates NCOPIES copies of a string.
Standard : Fortran 90 and later
Class: Transformational function
Syntax : RESULT = REPEAT(STRING, NCOPIES)
Chapter 8: Intrinsic Procedures 259

Arguments:
STRING Shall be scalar and of type CHARACTER.
NCOPIES Shall be scalar and of type INTEGER.
Return value:
A new scalar of type CHARACTER built up from NCOPIES copies of STRING.
Example:
program test_repeat
write(*,*) repeat("x", 5) ! "xxxxx"
end program

8.231 RESHAPE — Function to reshape an array

Description:
Reshapes SOURCE to correspond to SHAPE. If necessary, the new array may
be padded with elements from PAD or permuted as defined by ORDER.
Standard : Fortran 90 and later
Class: Transformational function
Syntax : RESULT = RESHAPE(SOURCE, SHAPE[, PAD, ORDER])
Arguments:
SOURCE Shall be an array of any type.
SHAPE Shall be of type INTEGER and an array of rank one.
Its values must be positive or zero.
PAD (Optional) shall be an array of the same type as
SOURCE.
ORDER (Optional) shall be of type INTEGER and an array
of the same shape as SHAPE. Its values shall be a
permutation of the numbers from 1 to n, where n is
the size of SHAPE. If ORDER is absent, the natural
ordering shall be assumed.
Return value:
The result is an array of shape SHAPE with the same type as SOURCE.
Example:
PROGRAM test_reshape
INTEGER, DIMENSION(4) :: x
WRITE(*,*) SHAPE(x) ! prints "4"
WRITE(*,*) SHAPE(RESHAPE(x, (/2, 2/))) ! prints "2 2"
END PROGRAM

See also: Section 8.243 [SHAPE], page 266,

8.232 RRSPACING — Reciprocal of the relative spacing

Description:
RRSPACING(X) returns the reciprocal of the relative spacing of model numbers
near X.
260 The GNU Fortran Compiler

Standard : Fortran 90 and later

Class: Elemental function
Syntax : RESULT = RRSPACING(X)
Arguments:
X Shall be of type REAL.
Return value:
The return value is of the same type and kind as X. The value returned is equal
to ABS(FRACTION(X)) * FLOAT(RADIX(X))**DIGITS(X).
See also: Section 8.255 [SPACING], page 273,

8.233 RSHIFT — Right shift bits

Description:
RSHIFT returns a value corresponding to I with all of the bits shifted right by
SHIFT places. SHIFT shall be nonnegative and less than or equal to BIT_
SIZE(I), otherwise the result value is undefined. Bits shifted out from the
right end are lost. The fill is arithmetic: the bits shifted in from the left end
are equal to the leftmost bit, which in two’s complement representation is the
sign bit.
This function has been superseded by the SHIFTA intrinsic, which is standard
in Fortran 2008 and later.
Standard : GNU extension
Class: Elemental function
Syntax : RESULT = RSHIFT(I, SHIFT)
Arguments:
I The type shall be INTEGER.
SHIFT The type shall be INTEGER.
Return value:
The return value is of type INTEGER and of the same kind as I.
See also: Section 8.159 [ISHFT], page 216,
Section 8.160 [ISHFTC], page 217,
Section 8.181 [LSHIFT], page 228,
Section 8.244 [SHIFTA], page 267,
Section 8.246 [SHIFTR], page 268,
Section 8.245 [SHIFTL], page 267,

8.234 SAME_TYPE_AS — Query dynamic types for equality

Description:
Query dynamic types for equality.
Standard : Fortran 2003 and later
Chapter 8: Intrinsic Procedures 261

Class: Inquiry function

Syntax : RESULT = SAME_TYPE_AS(A, B)
Arguments:
A Shall be an object of extensible declared type or un-
limited polymorphic.
B Shall be an object of extensible declared type or un-
limited polymorphic.
Return value:
The return value is a scalar of type default logical. It is true if and only if the
dynamic type of A is the same as the dynamic type of B.
See also: Section 8.104 [EXTENDS TYPE OF], page 180,

8.235 SCALE — Scale a real value

Description:
SCALE(X,I) returns X * RADIX(X)**I.
Standard : Fortran 90 and later
Class: Elemental function
Syntax : RESULT = SCALE(X, I)
Arguments:
X The type of the argument shall be a REAL.
I The type of the argument shall be a INTEGER.
Return value:
The return value is of the same type and kind as X. Its value is X *
RADIX(X)**I.
Example:
program test_scale
real :: x = 178.1387e-4
integer :: i = 5
print *, scale(x,i), x*radix(x)**i
end program test_scale

8.236 SCAN — Scan a string for the presence of a set of

characters
Description:
Scans a STRING for any of the characters in a SET of characters.
If BACK is either absent or equals FALSE, this function returns the position of
the leftmost character of STRING that is in SET. If BACK equals TRUE, the
rightmost position is returned. If no character of SET is found in STRING, the
result is zero.
Standard : Fortran 90 and later, with KIND argument Fortran 2003 and later
Class: Elemental function
262 The GNU Fortran Compiler

Syntax : RESULT = SCAN(STRING, SET[, BACK [, KIND]])

Arguments:
STRING Shall be of type CHARACTER.
SET Shall be of type CHARACTER.
BACK (Optional) shall be of type LOGICAL.
KIND (Optional) A scalar INTEGER constant expression in-
dicating the kind parameter of the result.
Return value:
The return value is of type INTEGER and of kind KIND. If KIND is absent, the
return value is of default integer kind.
Example:
PROGRAM test_scan
WRITE(*,*) SCAN("FORTRAN", "AO") ! 2, found 'O'
WRITE(*,*) SCAN("FORTRAN", "AO", .TRUE.) ! 6, found 'A'
WRITE(*,*) SCAN("FORTRAN", "C++") ! 0, found none
END PROGRAM

See also: Section 8.148 [INDEX intrinsic], page 209,

Section 8.282 [VERIFY], page 289,

8.237 SECNDS — Time function

Description:
SECNDS(X) gets the time in seconds from the real-time system clock. X is a
reference time, also in seconds. If this is zero, the time in seconds from midnight
is returned. This function is non-standard and its use is discouraged.
Standard : GNU extension
Class: Function
Syntax : RESULT = SECNDS (X)
Arguments:
T Shall be of type REAL(4).
X Shall be of type REAL(4).
Return value:
None
Example:
program test_secnds
integer :: i
real(4) :: t1, t2
print *, secnds (0.0) ! seconds since midnight
t1 = secnds (0.0) ! reference time
do i = 1, 10000000 ! do something
end do
t2 = secnds (t1) ! elapsed time
print *, "Something took ", t2, " seconds."
end program test_secnds
Chapter 8: Intrinsic Procedures 263

8.238 SECOND — CPU time function

Description:
Returns a REAL(4) value representing the elapsed CPU time in seconds. This
provides the same functionality as the standard CPU_TIME intrinsic, and is only
included for backwards compatibility.
This intrinsic is provided in both subroutine and function forms; however, only
one form can be used in any given program unit.
Standard : GNU extension
Class: Subroutine, function
Syntax :
CALL SECOND(TIME)
TIME = SECOND()
Arguments:
TIME Shall be of type REAL(4).
Return value:
In either syntax, TIME is set to the process’s current runtime in seconds.
See also: Section 8.79 [CPU TIME], page 164,

8.239 SELECTED_CHAR_KIND — Choose character kind

Description:
SELECTED_CHAR_KIND(NAME) returns the kind value for the character set named
NAME, if a character set with such a name is supported, or −1 otherwise.
Currently, supported character sets include “ASCII” and “DEFAULT”, which
are equivalent, and “ISO 10646” (Universal Character Set, UCS-4) which is
commonly known as Unicode.
Standard : Fortran 2003 and later
Class: Transformational function
Syntax : RESULT = SELECTED_CHAR_KIND(NAME)
Arguments:
NAME Shall be a scalar and of the default character type.
Example:
program character_kind
use iso_fortran_env
implicit none
integer, parameter :: ascii = selected_char_kind ("ascii")
integer, parameter :: ucs4 = selected_char_kind ('ISO_10646')

character(kind=ascii, len=26) :: alphabet

character(kind=ucs4, len=30) :: hello_world

alphabet = ascii_"abcdefghijklmnopqrstuvwxyz"
hello_world = ucs4_'Hello World and Ni Hao -- ' &
264 The GNU Fortran Compiler

// char (int (z'4F60'), ucs4) &

// char (int (z'597D'), ucs4)

write (*,*) alphabet

open (output_unit, encoding='UTF-8')

write (*,*) trim (hello_world)
end program character_kind

8.240 SELECTED_INT_KIND — Choose integer kind

Description:
SELECTED_INT_KIND(R) return the kind value of the smallest integer type that
can represent all values ranging from −10R (exclusive) to 10R (exclusive). If
there is no integer kind that accommodates this range, SELECTED_INT_KIND
returns −1.
Standard : Fortran 90 and later
Class: Transformational function
Syntax : RESULT = SELECTED_INT_KIND(R)
Arguments:
R Shall be a scalar and of type INTEGER.
Example:
program large_integers
integer,parameter :: k5 = selected_int_kind(5)
integer,parameter :: k15 = selected_int_kind(15)
integer(kind=k5) :: i5
integer(kind=k15) :: i15

print *, huge(i5), huge(i15)

! The following inequalities are always true

print *, huge(i5) >= 10_k5**5-1
print *, huge(i15) >= 10_k15**15-1
end program large_integers

8.241 SELECTED_REAL_KIND — Choose real kind

Description:
SELECTED_REAL_KIND(P,R) returns the kind value of a real data type with
decimal precision of at least P digits, exponent range of at least R, and with a
radix of RADIX.
Standard : Fortran 90 and later, with RADIX Fortran 2008 or later
Class: Transformational function
Syntax : RESULT = SELECTED_REAL_KIND([P, R, RADIX])
Arguments:
P (Optional) shall be a scalar and of type INTEGER.
R (Optional) shall be a scalar and of type INTEGER.
Chapter 8: Intrinsic Procedures 265

RADIX (Optional) shall be a scalar and of type INTEGER.

Before Fortran 2008, at least one of the arguments R or P shall be present;
since Fortran 2008, they are assumed to be zero if absent.
Return value:
SELECTED_REAL_KIND returns the value of the kind type parameter of a real
data type with decimal precision of at least P digits, a decimal exponent range
of at least R, and with the requested RADIX. If the RADIX parameter is absent,
real kinds with any radix can be returned. If more than one real data type
meet the criteria, the kind of the data type with the smallest decimal precision
is returned. If no real data type matches the criteria, the result is
-1 if the processor does not support a real data type with a
precision greater than or equal to P, but the R and RADIX require-
ments can be fulfilled
-2 if the processor does not support a real type with an exponent
range greater than or equal to R, but P and RADIX are fulfillable
-3 if RADIX but not P and R requirements
are fulfillable
-4 if RADIX and either P or R requirements
are fulfillable
-5 if there is no real type with the given RADIX
Example:
program real_kinds
integer,parameter :: p6 = selected_real_kind(6)
integer,parameter :: p10r100 = selected_real_kind(10,100)
integer,parameter :: r400 = selected_real_kind(r=400)
real(kind=p6) :: x
real(kind=p10r100) :: y
real(kind=r400) :: z

print *, precision(x), range(x)

print *, precision(y), range(y)
print *, precision(z), range(z)
end program real_kinds

See also: Section 8.217 [PRECISION], page 251,

Section 8.226 [RANGE], page 256,
Section 8.220 [RADIX], page 252,

8.242 SET_EXPONENT — Set the exponent of the model

Description:
SET_EXPONENT(X, I) returns the real number whose fractional part is that of
X and whose exponent part is I.
Standard : Fortran 90 and later
Class: Elemental function
Syntax : RESULT = SET_EXPONENT(X, I)
266 The GNU Fortran Compiler

Arguments:
X Shall be of type REAL.
I Shall be of type INTEGER.

Return value:
The return value is of the same type and kind as X. The real number whose
fractional part is that of X and whose exponent part if I is returned; it is
FRACTION(X) * RADIX(X)**I.

Example:
PROGRAM test_setexp
REAL :: x = 178.1387e-4
INTEGER :: i = 17
PRINT *, SET_EXPONENT(x, i), FRACTION(x) * RADIX(x)**i
END PROGRAM

8.243 SHAPE — Determine the shape of an array

Description:
Determines the shape of an array.

Standard : Fortran 90 and later, with KIND argument Fortran 2003 and later

Class: Inquiry function

Syntax : RESULT = SHAPE(SOURCE [, KIND])

Arguments:
SOURCE Shall be an array or scalar of any type. If SOURCE is
a pointer it must be associated and allocatable arrays
must be allocated.
KIND (Optional) A scalar INTEGER constant expression in-
dicating the kind parameter of the result.

Return value:
An INTEGER array of rank one with as many elements as SOURCE has dimen-
sions. The elements of the resulting array correspond to the extend of SOURCE
along the respective dimensions. If SOURCE is a scalar, the result is the rank
one array of size zero. If KIND is absent, the return value has the default
integer kind otherwise the specified kind.

Example:
PROGRAM test_shape
INTEGER, DIMENSION(-1:1, -1:2) :: A
WRITE(*,*) SHAPE(A) ! (/ 3, 4 /)
WRITE(*,*) SIZE(SHAPE(42)) ! (/ /)
END PROGRAM

See also: Section 8.231 [RESHAPE], page 259,

Section 8.252 [SIZE], page 271,
Chapter 8: Intrinsic Procedures 267

8.244 SHIFTA — Right shift with fill

Description:
SHIFTA returns a value corresponding to I with all of the bits shifted right by
SHIFT places. SHIFT that be nonnegative and less than or equal to BIT_
SIZE(I), otherwise the result value is undefined. Bits shifted out from the
right end are lost. The fill is arithmetic: the bits shifted in from the left end
are equal to the leftmost bit, which in two’s complement representation is the
sign bit.

Standard : Fortran 2008 and later

Class: Elemental function

Syntax : RESULT = SHIFTA(I, SHIFT)

Arguments:
I The type shall be INTEGER.
SHIFT The type shall be INTEGER.

Return value:
The return value is of type INTEGER and of the same kind as I.

See also: Section 8.245 [SHIFTL], page 267,

Section 8.246 [SHIFTR], page 268,

8.245 SHIFTL — Left shift

Description:
SHIFTL returns a value corresponding to I with all of the bits shifted left by
SHIFT places. SHIFT shall be nonnegative and less than or equal to BIT_
SIZE(I), otherwise the result value is undefined. Bits shifted out from the left
end are lost, and bits shifted in from the right end are set to 0.

Standard : Fortran 2008 and later

Class: Elemental function

Syntax : RESULT = SHIFTL(I, SHIFT)

Arguments:
I The type shall be INTEGER.
SHIFT The type shall be INTEGER.

Return value:
The return value is of type INTEGER and of the same kind as I.

See also: Section 8.244 [SHIFTA], page 267,

Section 8.246 [SHIFTR], page 268,
268 The GNU Fortran Compiler

8.246 SHIFTR — Right shift

Description:
SHIFTR returns a value corresponding to I with all of the bits shifted right by
SHIFT places. SHIFT shall be nonnegative and less than or equal to BIT_
SIZE(I), otherwise the result value is undefined. Bits shifted out from the
right end are lost, and bits shifted in from the left end are set to 0.
Standard : Fortran 2008 and later
Class: Elemental function
Syntax : RESULT = SHIFTR(I, SHIFT)
Arguments:
I The type shall be INTEGER.
SHIFT The type shall be INTEGER.
Return value:
The return value is of type INTEGER and of the same kind as I.
See also: Section 8.244 [SHIFTA], page 267,
Section 8.245 [SHIFTL], page 267,

8.247 SIGN — Sign copying function

Description:
SIGN(A,B) returns the value of A with the sign of B.
Standard : Fortran 77 and later
Class: Elemental function
Syntax : RESULT = SIGN(A, B)
Arguments:
A Shall be of type INTEGER or REAL
B Shall be of the same type and kind as A.
Return value:
The kind of the return value is that of A and B. If B ≥ 0 then the result is
ABS(A), else it is -ABS(A).
Example:
program test_sign
print *, sign(-12,1)
print *, sign(-12,0)
print *, sign(-12,-1)

print *, sign(-12.,1.)
print *, sign(-12.,0.)
print *, sign(-12.,-1.)
end program test_sign

Specific names:
Name Arguments Return type Standard
SIGN(A,B) REAL(4) A, B REAL(4) Fortran 77 and later
Chapter 8: Intrinsic Procedures 269

ISIGN(A,B) INTEGER(4) A, B INTEGER(4) Fortran 77 and later

DSIGN(A,B) REAL(8) A, B REAL(8) Fortran 77 and later

8.248 SIGNAL — Signal handling subroutine (or function)

Description:
SIGNAL(NUMBER, HANDLER [, STATUS]) causes external subroutine HANDLER
to be executed with a single integer argument when signal NUMBER occurs. If
HANDLER is an integer, it can be used to turn off handling of signal NUMBER
or revert to its default action. See signal(2).
If SIGNAL is called as a subroutine and the STATUS argument is supplied, it is
set to the value returned by signal(2).
Standard : GNU extension
Class: Subroutine, function
Syntax :
CALL SIGNAL(NUMBER, HANDLER [, STATUS])
STATUS = SIGNAL(NUMBER, HANDLER)
Arguments:
NUMBER Shall be a scalar integer, with INTENT(IN)
HANDLER Signal handler (INTEGER FUNCTION or SUBROUTINE)
or dummy/global INTEGER scalar. INTEGER. It is
INTENT(IN).
STATUS (Optional) STATUS shall be a scalar integer. It has
INTENT(OUT).
Return value:
The SIGNAL function returns the value returned by signal(2).
Example:
program test_signal
intrinsic signal
external handler_print

call signal (12, handler_print)

call signal (10, 1)

call sleep (30)

end program test_signal

8.249 SIN — Sine function

Description:
SIN(X) computes the sine of X.
Standard : Fortran 77 and later
Class: Elemental function
Syntax : RESULT = SIN(X)
270 The GNU Fortran Compiler

Arguments:
X The type shall be REAL or COMPLEX.
Return value:
The return value has same type and kind as X.
Example:
program test_sin
real :: x = 0.0
x = sin(x)
end program test_sin

Specific names:
Name Argument Return type Standard
SIN(X) REAL(4) X REAL(4) Fortran 77 and later
DSIN(X) REAL(8) X REAL(8) Fortran 77 and later
CSIN(X) COMPLEX(4) X COMPLEX(4) Fortran 77 and later
ZSIN(X) COMPLEX(8) X COMPLEX(8) GNU extension
CDSIN(X) COMPLEX(8) X COMPLEX(8) GNU extension
See also: Inverse function:
Section 8.19 [ASIN], page 122,
Degrees function:
Section 8.250 [SIND], page 270,

8.250 SIND — Sine function, degrees

Description:
SIND(X) computes the sine of X in degrees.
This function is for compatibility only and should be avoided in favor of stan-
dard constructs wherever possible.
Standard : GNU extension, enabled with -fdec-math.
Class: Elemental function
Syntax : RESULT = SIND(X)
Arguments:
X The type shall be REAL or COMPLEX.
Return value:
The return value has same type and kind as X, and its value is in degrees.
Example:
program test_sind
real :: x = 0.0
x = sind(x)
end program test_sind

Specific names:
Name Argument Return type Standard
SIND(X) REAL(4) X REAL(4) GNU extension
DSIND(X) REAL(8) X REAL(8) GNU extension
Chapter 8: Intrinsic Procedures 271

CSIND(X) COMPLEX(4) X COMPLEX(4) GNU extension

ZSIND(X) COMPLEX(8) X COMPLEX(8) GNU extension
CDSIND(X) COMPLEX(8) X COMPLEX(8) GNU extension
See also: Inverse function:
Section 8.20 [ASIND], page 123,
Radians function:
Section 8.249 [SIN], page 269,

8.251 SINH — Hyperbolic sine function

Description:
SINH(X) computes the hyperbolic sine of X.
Standard : Fortran 90 and later, for a complex argument Fortran 2008 or later, has a GNU
extension
Class: Elemental function
Syntax : RESULT = SINH(X)
Arguments:
X The type shall be REAL or COMPLEX.
Return value:
The return value has same type and kind as X.
Example:
program test_sinh
real(8) :: x = - 1.0_8
x = sinh(x)
end program test_sinh

Specific names:
Name Argument Return type Standard
DSINH(X) REAL(8) X REAL(8) Fortran 90 and later
See also: Section 8.21 [ASINH], page 124,

8.252 SIZE — Determine the size of an array

Description:
Determine the extent of ARRAY along a specified dimension DIM, or the total
number of elements in ARRAY if DIM is absent.
Standard : Fortran 90 and later, with KIND argument Fortran 2003 and later
Class: Inquiry function
Syntax : RESULT = SIZE(ARRAY[, DIM [, KIND]])
Arguments:
ARRAY Shall be an array of any type. If ARRAY is a pointer
it must be associated and allocatable arrays must be
allocated.
272 The GNU Fortran Compiler

DIM (Optional) shall be a scalar of type INTEGER and its

value shall be in the range from 1 to n, where n equals
the rank of ARRAY.
KIND (Optional) A scalar INTEGER constant expression in-
dicating the kind parameter of the result.
Return value:
The return value is of type INTEGER and of kind KIND. If KIND is absent, the
return value is of default integer kind.
Example:
PROGRAM test_size
WRITE(*,*) SIZE((/ 1, 2 /)) ! 2
END PROGRAM

See also: Section 8.243 [SHAPE], page 266,

Section 8.231 [RESHAPE], page 259,

8.253 SIZEOF — Size in bytes of an expression

Description:
SIZEOF(X) calculates the number of bytes of storage the expression X occupies.
Standard : GNU extension
Class: Inquiry function
Syntax : N = SIZEOF(X)
Arguments:
X The argument shall be of any type, rank or shape.
Return value:
The return value is of type integer and of the system-dependent kind C SIZE T
(from the ISO C BINDING module). Its value is the number of bytes occupied
by the argument. If the argument has the POINTER attribute, the number
of bytes of the storage area pointed to is returned. If the argument is of a
derived type with POINTER or ALLOCATABLE components, the return value does
not account for the sizes of the data pointed to by these components. If the
argument is polymorphic, the size according to the dynamic type is returned.
The argument may not be a procedure or procedure pointer. Note that the code
assumes for arrays that those are contiguous; for contiguous arrays, it returns
the storage or an array element multiplied by the size of the array.
Example:
integer :: i
real :: r, s(5)
print *, (sizeof(s)/sizeof(r) == 5)
end
The example will print .TRUE. unless you are using a platform where default
REAL variables are unusually padded.
See also: Section 8.57 [C SIZEOF], page 148,
Section 8.260 [STORAGE SIZE], page 276,
Chapter 8: Intrinsic Procedures 273

8.254 SLEEP — Sleep for the specified number of seconds

Description:
Calling this subroutine causes the process to pause for SECONDS seconds.
Standard : GNU extension
Class: Subroutine
Syntax : CALL SLEEP(SECONDS)
Arguments:
SECONDS The type shall be of default INTEGER.
Example:
program test_sleep
call sleep(5)
end

8.255 SPACING — Smallest distance between two numbers of

a given type
Description:
Determines the distance between the argument X and the nearest adjacent
number of the same type.
Standard : Fortran 90 and later
Class: Elemental function
Syntax : RESULT = SPACING(X)
Arguments:
X Shall be of type REAL.
Return value:
The result is of the same type as the input argument X.
Example:
PROGRAM test_spacing
INTEGER, PARAMETER :: SGL = SELECTED_REAL_KIND(p=6, r=37)
INTEGER, PARAMETER :: DBL = SELECTED_REAL_KIND(p=13, r=200)

WRITE(*,*) spacing(1.0_SGL) ! "1.1920929E-07" on i686

WRITE(*,*) spacing(1.0_DBL) ! "2.220446049250313E-016" on i686
END PROGRAM

See also: Section 8.232 [RRSPACING], page 259,

8.256 SPREAD — Add a dimension to an array

Description:
Replicates a SOURCE array NCOPIES times along a specified dimension DIM.
Standard : Fortran 90 and later
Class: Transformational function
274 The GNU Fortran Compiler

Syntax : RESULT = SPREAD(SOURCE, DIM, NCOPIES)

Arguments:
SOURCE Shall be a scalar or an array of any type and a rank
less than seven.
DIM Shall be a scalar of type INTEGER with a value in
the range from 1 to n+1, where n equals the rank of
SOURCE.
NCOPIES Shall be a scalar of type INTEGER.
Return value:
The result is an array of the same type as SOURCE and has rank n+1 where n
equals the rank of SOURCE.
Example:
PROGRAM test_spread
INTEGER :: a = 1, b(2) = (/ 1, 2 /)
WRITE(*,*) SPREAD(A, 1, 2) ! "1 1"
WRITE(*,*) SPREAD(B, 1, 2) ! "1 1 2 2"
END PROGRAM
See also: Section 8.281 [UNPACK], page 289,

8.257 SQRT — Square-root function

Description:
SQRT(X) computes the square root of X.
Standard : Fortran 77 and later
Class: Elemental function
Syntax : RESULT = SQRT(X)
Arguments:
X The type shall be REAL or COMPLEX.
Return value:
The return value is of type REAL or COMPLEX. The kind type parameter is the
same as X.
Example:
program test_sqrt
real(8) :: x = 2.0_8
complex :: z = (1.0, 2.0)
x = sqrt(x)
z = sqrt(z)
end program test_sqrt
Specific names:
Name Argument Return type Standard
SQRT(X) REAL(4) X REAL(4) Fortran 77 and later
DSQRT(X) REAL(8) X REAL(8) Fortran 77 and later
CSQRT(X) COMPLEX(4) X COMPLEX(4) Fortran 77 and later
ZSQRT(X) COMPLEX(8) X COMPLEX(8) GNU extension
CDSQRT(X) COMPLEX(8) X COMPLEX(8) GNU extension
Chapter 8: Intrinsic Procedures 275

8.258 SRAND — Reinitialize the random number generator

Description:
SRAND reinitializes the pseudo-random number generator called by RAND and
IRAND. The new seed used by the generator is specified by the required argument
SEED.
Standard : GNU extension
Class: Subroutine
Syntax : CALL SRAND(SEED)
Arguments:
SEED Shall be a scalar INTEGER(kind=4).
Return value:
Does not return anything.
Example: See RAND and IRAND for examples.
Notes: The Fortran standard specifies the intrinsic subroutines RANDOM_SEED to ini-
tialize the pseudo-random number generator and RANDOM_NUMBER to generate
pseudo-random numbers. These subroutines should be used in new codes.
Please note that in GNU Fortran, these two sets of intrinsics (RAND, IRAND and
SRAND on the one hand, RANDOM_NUMBER and RANDOM_SEED on the other hand)
access two independent pseudo-random number generators.
See also: Section 8.222 [RAND], page 253,
Section 8.225 [RANDOM SEED], page 255,
Section 8.224 [RANDOM NUMBER], page 255,

8.259 STAT — Get file status

Description:
This function returns information about a file. No permissions are required on
the file itself, but execute (search) permission is required on all of the directories
in path that lead to the file.
The elements that are obtained and stored in the array VALUES:
VALUES(1) Device ID
VALUES(2) Inode number
VALUES(3) File mode
VALUES(4) Number of links
VALUES(5) Owner’s uid
VALUES(6) Owner’s gid
VALUES(7) ID of device containing directory entry for file (0 if
not available)
VALUES(8) File size (bytes)
VALUES(9) Last access time
VALUES(10) Last modification time
VALUES(11) Last file status change time
276 The GNU Fortran Compiler

VALUES(12) Preferred I/O block size (-1 if not available)

VALUES(13) Number of blocks allocated (-1 if not available)
Not all these elements are relevant on all systems. If an element is not relevant,
it is returned as 0.
This intrinsic is provided in both subroutine and function forms; however, only
one form can be used in any given program unit.
Standard : GNU extension
Class: Subroutine, function
Syntax :
CALL STAT(NAME, VALUES [, STATUS])
STATUS = STAT(NAME, VALUES)
Arguments:
NAME The type shall be CHARACTER, of the default kind and
a valid path within the file system.
VALUES The type shall be INTEGER(4), DIMENSION(13).
STATUS (Optional) status flag of type INTEGER(4). Returns 0
on success and a system specific error code otherwise.
Example:
PROGRAM test_stat
INTEGER, DIMENSION(13) :: buff
INTEGER :: status

CALL STAT("/etc/passwd", buff, status)

IF (status == 0) THEN
WRITE (*, FMT="('Device ID:', T30, I19)") buff(1)
WRITE (*, FMT="('Inode number:', T30, I19)") buff(2)
WRITE (*, FMT="('File mode (octal):', T30, O19)") buff(3)
WRITE (*, FMT="('Number of links:', T30, I19)") buff(4)
WRITE (*, FMT="('Owner''s uid:', T30, I19)") buff(5)
WRITE (*, FMT="('Owner''s gid:', T30, I19)") buff(6)
WRITE (*, FMT="('Device where located:', T30, I19)") buff(7)
WRITE (*, FMT="('File size:', T30, I19)") buff(8)
WRITE (*, FMT="('Last access time:', T30, A19)") CTIME(buff(9))
WRITE (*, FMT="('Last modification time', T30, A19)") CTIME(buff(10))
WRITE (*, FMT="('Last status change time:', T30, A19)") CTIME(buff(11))
WRITE (*, FMT="('Preferred block size:', T30, I19)") buff(12)
WRITE (*, FMT="('No. of blocks allocated:', T30, I19)") buff(13)
END IF
END PROGRAM
See also: To stat an open file:
Section 8.117 [FSTAT], page 190,
To stat a link:
Section 8.182 [LSTAT], page 229,

8.260 STORAGE_SIZE — Storage size in bits

Description:
Returns the storage size of argument A in bits.
Chapter 8: Intrinsic Procedures 277

Standard : Fortran 2008 and later

Class: Inquiry function
Syntax : RESULT = STORAGE_SIZE(A [, KIND])
Arguments:
A Shall be a scalar or array of any type.
KIND (Optional) shall be a scalar integer constant
expression.
Return Value:
The result is a scalar integer with the kind type parameter specified by KIND (or
default integer type if KIND is missing). The result value is the size expressed in
bits for an element of an array that has the dynamic type and type parameters
of A.
See also: Section 8.57 [C SIZEOF], page 148,
Section 8.253 [SIZEOF], page 272,

8.261 SUM — Sum of array elements

Description:
Adds the elements of ARRAY along dimension DIM if the corresponding ele-
ment in MASK is TRUE.
Standard : Fortran 90 and later
Class: Transformational function
Syntax :
RESULT = SUM(ARRAY[, MASK])
RESULT = SUM(ARRAY, DIM[, MASK])
Arguments:
ARRAY Shall be an array of type INTEGER, REAL or COMPLEX.
DIM (Optional) shall be a scalar of type INTEGER with a
value in the range from 1 to n, where n equals the
rank of ARRAY.
MASK (Optional) shall be of type LOGICAL and either be a
scalar or an array of the same shape as ARRAY.
Return value:
The result is of the same type as ARRAY.
If DIM is absent, a scalar with the sum of all elements in ARRAY is returned.
Otherwise, an array of rank n-1, where n equals the rank of ARRAY, and a
shape similar to that of ARRAY with dimension DIM dropped is returned.
Example:
PROGRAM test_sum
INTEGER :: x(5) = (/ 1, 2, 3, 4 ,5 /)
print *, SUM(x) ! all elements, sum = 15
print *, SUM(x, MASK=MOD(x, 2)==1) ! odd elements, sum = 9
END PROGRAM
See also: Section 8.219 [PRODUCT], page 252,
278 The GNU Fortran Compiler

8.262 SYMLNK — Create a symbolic link

Description:
Makes a symbolic link from file PATH1 to PATH2. A null character (CHAR(0))
can be used to mark the end of the names in PATH1 and PATH2; other-
wise, trailing blanks in the file names are ignored. If the STATUS argument
is supplied, it contains 0 on success or a nonzero error code upon return; see
symlink(2). If the system does not supply symlink(2), ENOSYS is returned.
This intrinsic is provided in both subroutine and function forms; however, only
one form can be used in any given program unit.
Standard : GNU extension
Class: Subroutine, function
Syntax :
CALL SYMLNK(PATH1, PATH2 [, STATUS])
STATUS = SYMLNK(PATH1, PATH2)
Arguments:
PATH1 Shall be of default CHARACTER type.
PATH2 Shall be of default CHARACTER type.
STATUS (Optional) Shall be of default INTEGER type.
See also: Section 8.172 [LINK], page 223,
Section 8.280 [UNLINK], page 288,

8.263 SYSTEM — Execute a shell command

Description:
Passes the command COMMAND to a shell (see system(3)). If argument
STATUS is present, it contains the value returned by system(3), which is
presumably 0 if the shell command succeeded. Note that which shell is used to
invoke the command is system-dependent and environment-dependent.
This intrinsic is provided in both subroutine and function forms; however, only
one form can be used in any given program unit.
Note that the system function need not be thread-safe. It is the responsibility
of the user to ensure that system is not called concurrently.
Standard : GNU extension
Class: Subroutine, function
Syntax :
CALL SYSTEM(COMMAND [, STATUS])
STATUS = SYSTEM(COMMAND)
Arguments:
COMMAND Shall be of default CHARACTER type.
STATUS (Optional) Shall be of default INTEGER type.
See also: Section 8.100 [EXECUTE COMMAND LINE], page 178, which is part of the
Fortran 2008 standard and should considered in new code for future portability.
Chapter 8: Intrinsic Procedures 279

8.264 SYSTEM_CLOCK — Time function

Description:
Determines the COUNT of a processor clock since an unspecified time in the
past modulo COUNT MAX, COUNT RATE determines the number of clock
ticks per second. If the platform supports a monotonic clock, that clock is
used and can, depending on the platform clock implementation, provide up to
nanosecond resolution. If a monotonic clock is not available, the implementation
falls back to a realtime clock.
COUNT RATE is system dependent and can vary depending on the kind of
the arguments. For kind=4 arguments (and smaller integer kinds), COUNT
represents milliseconds, while for kind=8 arguments (and larger integer kinds),
COUNT typically represents micro- or nanoseconds depending on resolution
of the underlying platform clock. COUNT MAX usually equals HUGE(COUNT_
MAX). Note that the millisecond resolution of the kind=4 version implies that
the COUNT will wrap around in roughly 25 days. In order to avoid issues with
the wrap around and for more precise timing, please use the kind=8 version.
If there is no clock, or querying the clock fails, COUNT is set to -HUGE(COUNT),
and COUNT RATE and COUNT MAX are set to zero.
When running on a platform using the GNU C library (glibc) version 2.16 or
older, or a derivative thereof, the high resolution monotonic clock is available
only when linking with the rt library. This can be done explicitly by adding the
-lrt flag when linking the application, but is also done implicitly when using
OpenMP.
On the Windows platform, the version with kind=4 arguments
uses the GetTickCount function, whereas the kind=8 version uses
QueryPerformanceCounter and QueryPerformanceCounterFrequency.
For more information, and potential caveats, please see the platform
documentation.
Standard : Fortran 90 and later
Class: Subroutine
Syntax : CALL SYSTEM_CLOCK([COUNT, COUNT_RATE, COUNT_MAX])
Arguments:
COUNT (Optional) shall be a scalar of type INTEGER with
INTENT(OUT).
COUNT RATE (Optional) shall be a scalar of type INTEGER or
REAL, with INTENT(OUT).
COUNT MAX (Optional) shall be a scalar of type INTEGER with
INTENT(OUT).
Example:
PROGRAM test_system_clock
INTEGER :: count, count_rate, count_max
CALL SYSTEM_CLOCK(count, count_rate, count_max)
WRITE(*,*) count, count_rate, count_max
END PROGRAM
280 The GNU Fortran Compiler

See also: Section 8.82 [DATE AND TIME], page 166,

Section 8.79 [CPU TIME], page 164,

8.265 TAN — Tangent function

Description:
TAN(X) computes the tangent of X.
Standard : Fortran 77 and later, for a complex argument Fortran 2008 or later
Class: Elemental function
Syntax : RESULT = TAN(X)
Arguments:
X The type shall be REAL or COMPLEX.
Return value:
The return value has same type and kind as X, and its value is in radians.
Example:
program test_tan
real(8) :: x = 0.165_8
x = tan(x)
end program test_tan

Specific names:
Name Argument Return type Standard
TAN(X) REAL(4) X REAL(4) Fortran 77 and later
DTAN(X) REAL(8) X REAL(8) Fortran 77 and later
See also: Inverse function:
Section 8.23 [ATAN], page 126,
Degrees function:
Section 8.266 [TAND], page 280,

8.266 TAND — Tangent function, degrees

Description:
TAND(X) computes the tangent of X in degrees.
This function is for compatibility only and should be avoided in favor of stan-
dard constructs wherever possible.
Standard : GNU extension, enabled with -fdec-math.
Class: Elemental function
Syntax : RESULT = TAND(X)
Arguments:
X The type shall be REAL or COMPLEX.
Return value:
The return value has same type and kind as X, and its value is in degrees.
Chapter 8: Intrinsic Procedures 281

Example:
program test_tand
real(8) :: x = 0.165_8
x = tand(x)
end program test_tand

Specific names:
Name Argument Return type Standard
TAND(X) REAL(4) X REAL(4) GNU extension
DTAND(X) REAL(8) X REAL(8) GNU extension
See also: Inverse function:
Section 8.24 [ATAND], page 126,
Radians function:
Section 8.265 [TAN], page 280,

8.267 TANH — Hyperbolic tangent function

Description:
TANH(X) computes the hyperbolic tangent of X.
Standard : Fortran 77 and later, for a complex argument Fortran 2008 or later
Class: Elemental function
Syntax : X = TANH(X)
Arguments:
X The type shall be REAL or COMPLEX.
Return value:
The return value has same type and kind as X. If X is complex, the imaginary
part of the result is in radians. If X is REAL, the return value lies in the range
−1 ≤ tanh(x) ≤ 1.
Example:
program test_tanh
real(8) :: x = 2.1_8
x = tanh(x)
end program test_tanh

Specific names:
Name Argument Return type Standard
TANH(X) REAL(4) X REAL(4) Fortran 77 and later
DTANH(X) REAL(8) X REAL(8) Fortran 77 and later
See also: Section 8.27 [ATANH], page 129,

8.268 THIS_IMAGE — Function that returns the cosubscript

index of this image
Description:
Returns the cosubscript for this image.
282 The GNU Fortran Compiler

Standard : Fortran 2008 and later. With DISTANCE argument, Technical Specification
(TS) 18508 or later
Class: Transformational function
Syntax :
RESULT = THIS_IMAGE()
RESULT = THIS_IMAGE(DISTANCE)
RESULT = THIS_IMAGE(COARRAY [, DIM])
Arguments:
DISTANCE (optional, intent(in)) Nonnegative scalar integer (not
permitted together with COARRAY ).
COARRAY Coarray of any type (optional; if DIM present,
required).
DIM default integer scalar (optional). If present, DIM
shall be between one and the corank of COARRAY.
Return value:
Default integer. If COARRAY is not present, it is scalar; if DISTANCE is not
present or has value 0, its value is the image index on the invoking image for the
current team, for values smaller or equal distance to the initial team, it returns
the image index on the ancestor team which has a distance of DISTANCE
from the invoking team. If DISTANCE is larger than the distance to the
initial team, the image index of the initial team is returned. Otherwise when
the COARRAY is present, if DIM is not present, a rank-1 array with corank
elements is returned, containing the cosubscripts for COARRAY specifying the
invoking image. If DIM is present, a scalar is returned, with the value of the
DIM element of THIS_IMAGE(COARRAY).
Example:
INTEGER :: value[*]
INTEGER :: i
value = THIS_IMAGE()
SYNC ALL
IF (THIS_IMAGE() == 1) THEN
DO i = 1, NUM_IMAGES()
WRITE(*,'(2(a,i0))') 'value[', i, '] is ', value[i]
END DO
END IF

! Check whether the current image is the initial image

IF (THIS_IMAGE(HUGE(1)) /= THIS_IMAGE())
error stop "something is rotten here"

See also: Section 8.210 [NUM IMAGES], page 246,

Section 8.147 [IMAGE INDEX], page 209,

8.269 TIME — Time function

Description:
Returns the current time encoded as an integer (in the manner of the func-
tion time(3) in the C standard library). This value is suitable for passing
Chapter 8: Intrinsic Procedures 283

to Section 8.81 [CTIME], page 165, Section 8.131 [GMTIME], page 198, and
Section 8.183 [LTIME], page 230.
This intrinsic is not fully portable, such as to systems with 32-bit INTEGER
types but supporting times wider than 32 bits. Therefore, the values returned
by this intrinsic might be, or become, negative, or numerically less than previous
values, during a single run of the compiled program.
See Section 8.270 [TIME8], page 283, for information on a similar intrinsic that
might be portable to more GNU Fortran implementations, though to fewer
Fortran compilers.
Standard : GNU extension
Class: Function
Syntax : RESULT = TIME()
Return value:
The return value is a scalar of type INTEGER(4).
See also: Section 8.82 [DATE AND TIME], page 166,
Section 8.81 [CTIME], page 165,
Section 8.131 [GMTIME], page 198,
Section 8.183 [LTIME], page 230,
Section 8.192 [MCLOCK], page 235,
Section 8.270 [TIME8], page 283,

8.270 TIME8 — Time function (64-bit)

Description:
Returns the current time encoded as an integer (in the manner of the func-
tion time(3) in the C standard library). This value is suitable for passing
to Section 8.81 [CTIME], page 165, Section 8.131 [GMTIME], page 198, and
Section 8.183 [LTIME], page 230.
Warning: this intrinsic does not increase the range of the timing values over
that returned by time(3). On a system with a 32-bit time(3), TIME8 will
return a 32-bit value, even though it is converted to a 64-bit INTEGER(8) value.
That means overflows of the 32-bit value can still occur. Therefore, the values
returned by this intrinsic might be or become negative or numerically less than
previous values during a single run of the compiled program.
Standard : GNU extension
Class: Function
Syntax : RESULT = TIME8()
Return value:
The return value is a scalar of type INTEGER(8).
See also: Section 8.82 [DATE AND TIME], page 166,
Section 8.81 [CTIME], page 165,
Section 8.131 [GMTIME], page 198,
284 The GNU Fortran Compiler

Section 8.183 [LTIME], page 230,

Section 8.193 [MCLOCK8], page 236,
Section 8.269 [TIME], page 282,

8.271 TINY — Smallest positive number of a real kind

Description:
TINY(X) returns the smallest positive (non zero) number in the model of the
type of X.
Standard : Fortran 90 and later
Class: Inquiry function
Syntax : RESULT = TINY(X)
Arguments:
X Shall be of type REAL.
Return value:
The return value is of the same type and kind as X
Example: See HUGE for an example.

8.272 TRAILZ — Number of trailing zero bits of an integer

Description:
TRAILZ returns the number of trailing zero bits of an integer.
Standard : Fortran 2008 and later
Class: Elemental function
Syntax : RESULT = TRAILZ(I)
Arguments:
I Shall be of type INTEGER.
Return value:
The type of the return value is the default INTEGER. If all the bits of I are zero,
the result value is BIT_SIZE(I).
Example:
PROGRAM test_trailz
WRITE (*,*) TRAILZ(8) ! prints 3
END PROGRAM

See also: Section 8.48 [BIT SIZE], page 142,

Section 8.167 [LEADZ], page 220,
Section 8.216 [POPPAR], page 250,
Section 8.215 [POPCNT], page 250,
Chapter 8: Intrinsic Procedures 285

8.273 TRANSFER — Transfer bit patterns

Description:
Interprets the bitwise representation of SOURCE in memory as if it is the
representation of a variable or array of the same type and type parameters as
MOLD.
This is approximately equivalent to the C concept of casting one type to another.
Standard : Fortran 90 and later
Class: Transformational function
Syntax : RESULT = TRANSFER(SOURCE, MOLD[, SIZE])
Arguments:
SOURCE Shall be a scalar or an array of any type.
MOLD Shall be a scalar or an array of any type.
SIZE (Optional) shall be a scalar of type INTEGER.
Return value:
The result has the same type as MOLD, with the bit level representation of
SOURCE. If SIZE is present, the result is a one-dimensional array of length
SIZE. If SIZE is absent but MOLD is an array (of any size or shape), the
result is a one- dimensional array of the minimum length needed to contain
the entirety of the bitwise representation of SOURCE. If SIZE is absent and
MOLD is a scalar, the result is a scalar.
If the bitwise representation of the result is longer than that of SOURCE, then
the leading bits of the result correspond to those of SOURCE and any trailing
bits are filled arbitrarily.
When the resulting bit representation does not correspond to a valid represen-
tation of a variable of the same type as MOLD, the results are undefined, and
subsequent operations on the result cannot be guaranteed to produce sensible
behavior. For example, it is possible to create LOGICAL variables for which VAR
and .NOT.VAR both appear to be true.
Example:
PROGRAM test_transfer
integer :: x = 2143289344
print *, transfer(x, 1.0) ! prints "NaN" on i686
END PROGRAM

8.274 TRANSPOSE — Transpose an array of rank two

Description:
Transpose an array of rank two. Element (i, j) of the result has the value
MATRIX(j, i), for all i, j.
Standard : Fortran 90 and later
Class: Transformational function
Syntax : RESULT = TRANSPOSE(MATRIX)
286 The GNU Fortran Compiler

Arguments:
MATRIX Shall be an array of any type and have a rank of two.
Return value:
The result has the same type as MATRIX, and has shape (/ m, n /) if MA-
TRIX has shape (/ n, m /).

8.275 TRIM — Remove trailing blank characters of a string

Description:
Removes trailing blank characters of a string.
Standard : Fortran 90 and later
Class: Transformational function
Syntax : RESULT = TRIM(STRING)
Arguments:
STRING Shall be a scalar of type CHARACTER.
Return value:
A scalar of type CHARACTER which length is that of STRING less the number
of trailing blanks.
Example:
PROGRAM test_trim
CHARACTER(len=10), PARAMETER :: s = "GFORTRAN "
WRITE(*,*) LEN(s), LEN(TRIM(s)) ! "10 8", with/without trailing blanks
END PROGRAM

See also: Section 8.9 [ADJUSTL], page 116,

Section 8.10 [ADJUSTR], page 116,

8.276 TTYNAM — Get the name of a terminal device

Description:
Get the name of a terminal device. For more information, see ttyname(3).
This intrinsic is provided in both subroutine and function forms; however, only
one form can be used in any given program unit.
Standard : GNU extension
Class: Subroutine, function
Syntax :
CALL TTYNAM(UNIT, NAME)
NAME = TTYNAM(UNIT)
Arguments:
UNIT Shall be a scalar INTEGER.
NAME Shall be of type CHARACTER.
Chapter 8: Intrinsic Procedures 287

Example:
PROGRAM test_ttynam
INTEGER :: unit
DO unit = 1, 10
IF (isatty(unit=unit)) write(*,*) ttynam(unit)
END DO
END PROGRAM

See also: Section 8.158 [ISATTY], page 215,

8.277 UBOUND — Upper dimension bounds of an array

Description:
Returns the upper bounds of an array, or a single upper bound along the DIM
dimension.
Standard : Fortran 90 and later, with KIND argument Fortran 2003 and later
Class: Inquiry function
Syntax : RESULT = UBOUND(ARRAY [, DIM [, KIND]])
Arguments:
ARRAY Shall be an array, of any type.
DIM (Optional) Shall be a scalar INTEGER.
KIND (Optional) A scalar INTEGER constant expression in-
dicating the kind parameter of the result.
Return value:
The return value is of type INTEGER and of kind KIND. If KIND is absent,
the return value is of default integer kind. If DIM is absent, the result is an
array of the upper bounds of ARRAY. If DIM is present, the result is a scalar
corresponding to the upper bound of the array along that dimension. If ARRAY
is an expression rather than a whole array or array structure component, or if
it has a zero extent along the relevant dimension, the upper bound is taken to
be the number of elements along the relevant dimension.
See also: Section 8.165 [LBOUND], page 219,
Section 8.166 [LCOBOUND], page 220,

8.278 UCOBOUND — Upper codimension bounds of an array

Description:
Returns the upper cobounds of a coarray, or a single upper cobound along the
DIM codimension.
Standard : Fortran 2008 and later
Class: Inquiry function
Syntax : RESULT = UCOBOUND(COARRAY [, DIM [, KIND]])
Arguments:
ARRAY Shall be an coarray, of any type.
288 The GNU Fortran Compiler

DIM (Optional) Shall be a scalar INTEGER.

KIND (Optional) A scalar INTEGER constant expression in-
dicating the kind parameter of the result.
Return value:
The return value is of type INTEGER and of kind KIND. If KIND is absent, the
return value is of default integer kind. If DIM is absent, the result is an array
of the lower cobounds of COARRAY. If DIM is present, the result is a scalar
corresponding to the lower cobound of the array along that codimension.
See also: Section 8.166 [LCOBOUND], page 220,
Section 8.165 [LBOUND], page 219,

8.279 UMASK — Set the file creation mask

Description:
Sets the file creation mask to MASK. If called as a function, it returns the old
value. If called as a subroutine and argument OLD if it is supplied, it is set to
the old value. See umask(2).
Standard : GNU extension
Class: Subroutine, function
Syntax :
CALL UMASK(MASK [, OLD])
OLD = UMASK(MASK)
Arguments:
MASK Shall be a scalar of type INTEGER.
OLD (Optional) Shall be a scalar of type INTEGER.

8.280 UNLINK — Remove a file from the file system

Description:
Unlinks the file PATH. A null character (CHAR(0)) can be used to mark the end
of the name in PATH ; otherwise, trailing blanks in the file name are ignored. If
the STATUS argument is supplied, it contains 0 on success or a nonzero error
code upon return; see unlink(2).
This intrinsic is provided in both subroutine and function forms; however, only
one form can be used in any given program unit.
Standard : GNU extension
Class: Subroutine, function
Syntax :
CALL UNLINK(PATH [, STATUS])
STATUS = UNLINK(PATH)
Arguments:
PATH Shall be of default CHARACTER type.
STATUS (Optional) Shall be of default INTEGER type.
Chapter 8: Intrinsic Procedures 289

See also: Section 8.172 [LINK], page 223,

Section 8.262 [SYMLNK], page 278,

8.281 UNPACK — Unpack an array of rank one into an array

Description:
Store the elements of VECTOR in an array of higher rank.
Standard : Fortran 90 and later
Class: Transformational function
Syntax : RESULT = UNPACK(VECTOR, MASK, FIELD)
Arguments:
VECTOR Shall be an array of any type and rank one. It shall
have at least as many elements as MASK has TRUE
values.
MASK Shall be an array of type LOGICAL.
FIELD Shall be of the same type as VECTOR and have the
same shape as MASK.
Return value:
The resulting array corresponds to FIELD with TRUE elements of MASK re-
placed by values from VECTOR in array element order.
Example:
PROGRAM test_unpack
integer :: vector(2) = (/1,1/)
logical :: mask(4) = (/ .TRUE., .FALSE., .FALSE., .TRUE. /)
integer :: field(2,2) = 0, unity(2,2)

! result: unity matrix

unity = unpack(vector, reshape(mask, (/2,2/)), field)
END PROGRAM

See also: Section 8.212 [PACK], page 248,

Section 8.256 [SPREAD], page 273,

8.282 VERIFY — Scan a string for characters not a given set

Description:
Verifies that all the characters in STRING belong to the set of characters in
SET.
If BACK is either absent or equals FALSE, this function returns the position of
the leftmost character of STRING that is not in SET. If BACK equals TRUE,
the rightmost position is returned. If all characters of STRING are found in
SET, the result is zero.
Standard : Fortran 90 and later, with KIND argument Fortran 2003 and later
Class: Elemental function
Syntax : RESULT = VERIFY(STRING, SET[, BACK [, KIND]])
290 The GNU Fortran Compiler

Arguments:
STRING Shall be of type CHARACTER.
SET Shall be of type CHARACTER.
BACK (Optional) shall be of type LOGICAL.
KIND (Optional) A scalar INTEGER constant expression in-
dicating the kind parameter of the result.
Return value:
The return value is of type INTEGER and of kind KIND. If KIND is absent, the
return value is of default integer kind.
Example:
PROGRAM test_verify
WRITE(*,*) VERIFY("FORTRAN", "AO") ! 1, found 'F'
WRITE(*,*) VERIFY("FORTRAN", "FOO") ! 3, found 'R'
WRITE(*,*) VERIFY("FORTRAN", "C++") ! 1, found 'F'
WRITE(*,*) VERIFY("FORTRAN", "C++", .TRUE.) ! 7, found 'N'
WRITE(*,*) VERIFY("FORTRAN", "FORTRAN") ! 0' found none
END PROGRAM

See also: Section 8.236 [SCAN], page 261,

Section 8.148 [INDEX intrinsic], page 209,

8.283 XOR — Bitwise logical exclusive OR

Description:
Bitwise logical exclusive or.
This intrinsic routine is provided for backwards compatibility with GNU For-
tran 77. For integer arguments, programmers should consider the use of the
Section 8.145 [IEOR], page 208, intrinsic and for logical arguments the .NEQV.
operator, which are both defined by the Fortran standard.
Standard : GNU extension
Class: Function
Syntax : RESULT = XOR(I, J)
Arguments:
I The type shall be either a scalar INTEGER type or a
scalar LOGICAL type or a boz-literal-constant.
J The type shall be the same as the type of I or a boz-
literal-constant. I and J shall not both be boz-literal-
constants. If either I and J is a boz-literal-constant,
then the other argument must be a scalar INTEGER.
Return value:
The return type is either a scalar INTEGER or a scalar LOGICAL. If the kind type
parameters differ, then the smaller kind type is implicitly converted to larger
kind, and the return has the larger kind. A boz-literal-constant is converted to
an INTEGER with the kind type parameter of the other argument as-if a call to
Section 8.149 [INT], page 210, occurred.
Chapter 8: Intrinsic Procedures 291

Example:
PROGRAM test_xor
LOGICAL :: T = .TRUE., F = .FALSE.
INTEGER :: a, b
DATA a / Z'F' /, b / Z'3' /

WRITE (*,*) XOR(T, T), XOR(T, F), XOR(F, T), XOR(F, F)

WRITE (*,*) XOR(a, b)
END PROGRAM

See also: Fortran 95 elemental function:

Section 8.145 [IEOR], page 208,
 293

9 Intrinsic Modules

9.1 ISO_FORTRAN_ENV
Standard : Fortran 2003 and later, except when otherwise noted
The ISO_FORTRAN_ENV module provides the following scalar default-integer named con-
stants:
ATOMIC_INT_KIND:
Default-kind integer constant to be used as kind parameter when defining inte-
ger variables used in atomic operations. (Fortran 2008 or later.)
ATOMIC_LOGICAL_KIND:
Default-kind integer constant to be used as kind parameter when defining logical
variables used in atomic operations. (Fortran 2008 or later.)
CHARACTER_KINDS:
Default-kind integer constant array of rank one containing the supported kind
parameters of the CHARACTER type. (Fortran 2008 or later.)
CHARACTER_STORAGE_SIZE:
Size in bits of the character storage unit.
ERROR_UNIT:
Identifies the preconnected unit used for error reporting.
FILE_STORAGE_SIZE:
Size in bits of the file-storage unit.
INPUT_UNIT:
Identifies the preconnected unit identified by the asterisk (*) in READ statement.
INT8, INT16, INT32, INT64:
Kind type parameters to specify an INTEGER type with a storage size of 16, 32,
and 64 bits. It is negative if a target platform does not support the particular
kind. (Fortran 2008 or later.)
INTEGER_KINDS:
Default-kind integer constant array of rank one containing the supported kind
parameters of the INTEGER type. (Fortran 2008 or later.)
IOSTAT_END:
The value assigned to the variable passed to the IOSTAT= specifier of an in-
put/output statement if an end-of-file condition occurred.
IOSTAT_EOR:
The value assigned to the variable passed to the IOSTAT= specifier of an in-
put/output statement if an end-of-record condition occurred.
IOSTAT_INQUIRE_INTERNAL_UNIT:
Scalar default-integer constant, used by INQUIRE for the IOSTAT= specifier to
denote an that a unit number identifies an internal unit. (Fortran 2008 or later.)
294 The GNU Fortran Compiler

NUMERIC_STORAGE_SIZE:
The size in bits of the numeric storage unit.
LOGICAL_KINDS:
Default-kind integer constant array of rank one containing the supported kind
parameters of the LOGICAL type. (Fortran 2008 or later.)
OUTPUT_UNIT:
Identifies the preconnected unit identified by the asterisk (*) in WRITE state-
ment.
REAL32, REAL64, REAL128:
Kind type parameters to specify a REAL type with a storage size of 32, 64,
and 128 bits. It is negative if a target platform does not support the particular
kind. (Fortran 2008 or later.)
REAL_KINDS:
Default-kind integer constant array of rank one containing the supported kind
parameters of the REAL type. (Fortran 2008 or later.)
STAT_LOCKED:
Scalar default-integer constant used as STAT= return value by LOCK to denote
that the lock variable is locked by the executing image. (Fortran 2008 or later.)
STAT_LOCKED_OTHER_IMAGE:
Scalar default-integer constant used as STAT= return value by UNLOCK to denote
that the lock variable is locked by another image. (Fortran 2008 or later.)
STAT_STOPPED_IMAGE:
Positive, scalar default-integer constant used as STAT= return value if the
argument in the statement requires synchronisation with an image, which has
initiated the termination of the execution. (Fortran 2008 or later.)
STAT_FAILED_IMAGE:
Positive, scalar default-integer constant used as STAT= return value if the
argument in the statement requires communication with an image, which has
is in the failed state. (TS 18508 or later.)
STAT_UNLOCKED:
Scalar default-integer constant used as STAT= return value by UNLOCK to denote
that the lock variable is unlocked. (Fortran 2008 or later.)

The module provides the following derived type:

LOCK_TYPE:
Derived type with private components to be use with the LOCK and UNLOCK
statement. A variable of its type has to be always declared as coarray and may
not appear in a variable-definition context. (Fortran 2008 or later.)

The module also provides the following intrinsic procedures: Section 8.69
[COMPILER OPTIONS], page 157, and Section 8.70 [COMPILER VERSION], page 158.
Chapter 9: Intrinsic Modules 295

9.2 ISO_C_BINDING
Standard : Fortran 2003 and later, GNU extensions
The following intrinsic procedures are provided by the module; their definition can be
found in the section Intrinsic Procedures of this manual.
C_ASSOCIATED
C_F_POINTER
C_F_PROCPOINTER
C_FUNLOC
C_LOC
C_SIZEOF
The ISO_C_BINDING module provides the following named constants of type default
integer, which can be used as KIND type parameters.
In addition to the integer named constants required by the Fortran 2003 standard and
C_PTRDIFF_T of TS 29113, GNU Fortran provides as an extension named constants for
the 128-bit integer types supported by the C compiler: C_INT128_T, C_INT_LEAST128_T,
C_INT_FAST128_T. Furthermore, if _Float128 is supported in C, the named constants
C_FLOAT128 and C_FLOAT128_COMPLEX are defined.
Fortran Type Named constant C type Extension
INTEGER C_INT int
INTEGER C_SHORT short int
INTEGER C_LONG long int
INTEGER C_LONG_LONG long long int
INTEGER C_SIGNED_CHAR signed char/unsigned
char
INTEGER C_SIZE_T size_t
INTEGER C_INT8_T int8_t
INTEGER C_INT16_T int16_t
INTEGER C_INT32_T int32_t
INTEGER C_INT64_T int64_t
INTEGER C_INT128_T int128_t Ext.
INTEGER C_INT_LEAST8_T int_least8_t
INTEGER C_INT_LEAST16_T int_least16_t
INTEGER C_INT_LEAST32_T int_least32_t
INTEGER C_INT_LEAST64_T int_least64_t
INTEGER C_INT_LEAST128_T int_least128_t Ext.
INTEGER C_INT_FAST8_T int_fast8_t
INTEGER C_INT_FAST16_T int_fast16_t
INTEGER C_INT_FAST32_T int_fast32_t
INTEGER C_INT_FAST64_T int_fast64_t
INTEGER C_INT_FAST128_T int_fast128_t Ext.
INTEGER C_INTMAX_T intmax_t
INTEGER C_INTPTR_T intptr_t
INTEGER C_PTRDIFF_T ptrdiff_t TS 29113
296 The GNU Fortran Compiler

REAL C_FLOAT float

REAL C_DOUBLE double
REAL C_LONG_DOUBLE long double
REAL C_FLOAT128 _Float128 Ext.
COMPLEX C_FLOAT_COMPLEX float _Complex
COMPLEX C_DOUBLE_COMPLEX double _Complex
COMPLEX C_LONG_DOUBLE_COMPLEX long double _Complex
COMPLEX C_FLOAT128_COMPLEX _Float128 _Complex Ext.
LOGICAL C_BOOL _Bool
CHARACTER C_CHAR char
Additionally, the following parameters of type CHARACTER(KIND=C_CHAR) are defined.
Name C definition Value
C_NULL_CHAR null character '\0'
C_ALERT alert '\a'
C_BACKSPACE backspace '\b'
C_FORM_FEED form feed '\f'
C_NEW_LINE new line '\n'
C_CARRIAGE_ carriage return '\r'
RETURN
C_HORIZONTAL_ horizontal tab '\t'
TAB
C_VERTICAL_TAB vertical tab '\v'
Moreover, the following two named constants are defined:
Name Type
C_NULL_PTR C_PTR
C_NULL_FUNPTR C_FUNPTR
Both are equivalent to the value NULL in C.

9.3 IEEE modules: IEEE_EXCEPTIONS, IEEE_ARITHMETIC, and

IEEE_FEATURES
Standard : Fortran 2003 and later
The IEEE_EXCEPTIONS, IEEE_ARITHMETIC, and IEEE_FEATURES intrinsic modules pro-
vide support for exceptions and IEEE arithmetic, as defined in Fortran 2003 and later
standards, and the IEC 60559:1989 standard (Binary floating-point arithmetic for micro-
processor systems). These modules are only provided on the following supported platforms:
• i386 and x86 64 processors
• platforms which use the GNU C Library (glibc)
• platforms with support for SysV/386 routines for floating point interface (including
Solaris and BSDs)
• platforms with the AIX OS
For full compliance with the Fortran standards, code using the IEEE_EXCEPTIONS or
IEEE_ARITHMETIC modules should be compiled with the following options: -fno-unsafe-
math-optimizations -frounding-math -fsignaling-nans.
Chapter 9: Intrinsic Modules 297

9.4 OpenMP Modules OMP_LIB and OMP_LIB_KINDS

Standard : OpenMP Application Program Interface v4.5, OpenMP Application Program
Interface v5.0 (partially supported) and OpenMP Application Program Inter-
face v5.1 (partially supported).

The OpenMP Fortran runtime library routines are provided both in a form of two Fortran
modules, named OMP_LIB and OMP_LIB_KINDS, and in a form of a Fortran include file
named omp_lib.h. The procedures provided by OMP_LIB can be found in the Section
“Introduction” in GNU Offloading and Multi Processing Runtime Library manual, the
named constants defined in the modules are listed below.

For details refer to the actual OpenMP Application Program Interface v4.5
(https://www.openmp.org/wp-content/uploads/openmp-4.5.pdf) and OpenMP
Application Program Interface v5.0 (https://www.openmp.org/wp-content/uploads/
OpenMP-API-Specification-5.0.pdf).

OMP_LIB_KINDS provides the following scalar default-integer named constants:

omp_allocator_handle_kind
omp_alloctrait_key_kind
omp_alloctrait_val_kind
omp_depend_kind
omp_lock_kind
omp_lock_hint_kind
omp_nest_lock_kind
omp_pause_resource_kind
omp_memspace_handle_kind
omp_proc_bind_kind
omp_sched_kind
omp_sync_hint_kind

OMP_LIB provides the scalar default-integer named constant openmp_version with a

value of the form yyyymm, where yyyy is the year and mm the month of the OpenMP
version; for OpenMP v4.5 the value is 201511.

The following derived type:

omp_alloctrait

The following scalar integer named constants of the kind omp_sched_kind:

omp_sched_static
omp_sched_dynamic
omp_sched_guided
omp_sched_auto

And the following scalar integer named constants of the kind omp_proc_bind_kind:
298 The GNU Fortran Compiler

omp_proc_bind_false
omp_proc_bind_true
omp_proc_bind_primary
omp_proc_bind_master
omp_proc_bind_close
omp_proc_bind_spread

The following scalar integer named constants are of the kind omp_lock_hint_kind:

omp_lock_hint_none
omp_lock_hint_uncontended
omp_lock_hint_contended
omp_lock_hint_nonspeculative
omp_lock_hint_speculative
omp_sync_hint_none
omp_sync_hint_uncontended
omp_sync_hint_contended
omp_sync_hint_nonspeculative
omp_sync_hint_speculative

And the following two scalar integer named constants are of the kind omp_pause_
resource_kind:

omp_pause_soft
omp_pause_hard

The following scalar integer named constants are of the kind omp_alloctrait_key_kind:

omp_atk_sync_hint
omp_atk_alignment
omp_atk_access
omp_atk_pool_size
omp_atk_fallback
omp_atk_fb_data
omp_atk_pinned
omp_atk_partition

The following scalar integer named constants are of the kind omp_alloctrait_val_kind:

omp_alloctrait_key_kind:
Chapter 9: Intrinsic Modules 299

omp_atv_default
omp_atv_false
omp_atv_true
omp_atv_contended
omp_atv_uncontended
omp_atv_serialized
omp_atv_sequential
omp_atv_private
omp_atv_all
omp_atv_thread
omp_atv_pteam
omp_atv_cgroup
omp_atv_default_mem_fb
omp_atv_null_fb
omp_atv_abort_fb
omp_atv_allocator_fb
omp_atv_environment
omp_atv_nearest
omp_atv_blocked
The following scalar integer named constants are of the kind omp_allocator_handle_
kind:
omp_null_allocator
omp_default_mem_alloc
omp_large_cap_mem_alloc
omp_const_mem_alloc
omp_high_bw_mem_alloc
omp_low_lat_mem_alloc
omp_cgroup_mem_alloc
omp_pteam_mem_alloc
omp_thread_mem_alloc
The following scalar integer named constants are of the kind omp_memspace_handle_
kind:
omp_default_mem_space
omp_large_cap_mem_space
omp_const_mem_space
omp_high_bw_mem_space
omp_low_lat_mem_space

9.5 OpenACC Module OPENACC

Standard : OpenACC Application Programming Interface v2.6
The OpenACC Fortran runtime library routines are provided both in a form of a Fortran
90 module, named OPENACC, and in form of a Fortran include file named openacc_lib.h.
The procedures provided by OPENACC can be found in the Section “Introduction” in GNU
Offloading and Multi Processing Runtime Library manual, the named constants defined in
the modules are listed below.
300 The GNU Fortran Compiler

For details refer to the actual OpenACC Application Programming Interface v2.6
(https://www.openacc.org/).
OPENACC provides the scalar default-integer named constant openacc_version with a
value of the form yyyymm, where yyyy is the year and mm the month of the OpenACC
version; for OpenACC v2.6 the value is 201711.
 301

Contributing
Free software is only possible if people contribute to efforts to create it. We’re always in
need of more people helping out with ideas and comments, writing documentation and
contributing code.
If you want to contribute to GNU Fortran, have a look at the long lists of projects you
can take on. Some of these projects are small, some of them are large; some are completely
orthogonal to the rest of what is happening on GNU Fortran, but others are “mainstream”
projects in need of enthusiastic hackers. All of these projects are important! We will
eventually get around to the things here, but they are also things doable by someone who
is willing and able.

Contributors to GNU Fortran

Most of the parser was hand-crafted by Andy Vaught, who is also the initiator of the whole
project. Thanks Andy! Most of the interface with GCC was written by Paul Brook.
The following individuals have contributed code and/or ideas and significant help to the
GNU Fortran project (in alphabetical order):
− Janne Blomqvist
− Steven Bosscher
− Paul Brook
− Tobias Burnus
− François-Xavier Coudert
− Bud Davis
− Jerry DeLisle
− Erik Edelmann
− Bernhard Fischer
− Daniel Franke
− Richard Guenther
− Richard Henderson
− Katherine Holcomb
− Jakub Jelinek
− Niels Kristian Bech Jensen
− Steven Johnson
− Steven G. Kargl
− Thomas Koenig
− Asher Langton
− H. J. Lu
− Toon Moene
− Brooks Moses
− Andrew Pinski
302 The GNU Fortran Compiler

− Tim Prince
− Christopher D. Rickett
− Richard Sandiford
− Tobias Schlüter
− Roger Sayle
− Paul Thomas
− Andy Vaught
− Feng Wang
− Janus Weil
− Daniel Kraft
The following people have contributed bug reports, smaller or larger patches, and much
needed feedback and encouragement for the GNU Fortran project:
− Bill Clodius
− Dominique d’Humières
− Kate Hedstrom
− Erik Schnetter
− Gerhard Steinmetz
− Joost VandeVondele
Many other individuals have helped debug, test and improve the GNU Fortran compiler
over the past few years, and we welcome you to do the same! If you already have done so,
and you would like to see your name listed in the list above, please contact us.

Projects
Help build the test suite
Solicit more code for donation to the test suite: the more extensive the testsuite,
the smaller the risk of breaking things in the future! We can keep code private
on request.
Bug hunting/squishing
Find bugs and write more test cases! Test cases are especially very welcome, be-
cause it allows us to concentrate on fixing bugs instead of isolating them. Going
through the bugzilla database at https://gcc.gnu.org/bugzilla/ to reduce
testcases posted there and add more information (for example, for which version
does the testcase work, for which versions does it fail?) is also very helpful.
Missing features
For a larger project, consider working on the missing features required for For-
tran language standards compliance (see Section 1.3 [Standards], page 3), or
contributing to the implementation of extensions such as OpenMP (see Sec-
tion 5.1.18 [OpenMP], page 54) or OpenACC (see Section 5.1.19 [OpenACC],
page 55) that are under active development. Again, contributing test cases for
these features is useful too!
 303

GNU General Public License

Version 3, 29 June 2007
Copyright c 2007 Free Software Foundation, Inc. https://www.fsf.org

Everyone is permitted to copy and distribute verbatim copies of this

license document, but changing it is not allowed.

Preamble
The GNU General Public License is a free, copyleft license for software and other kinds of
works.
The licenses for most software and other practical works are designed to take away your
freedom to share and change the works. By contrast, the GNU General Public License is
intended to guarantee your freedom to share and change all versions of a program–to make
sure it remains free software for all its users. We, the Free Software Foundation, use the
GNU General Public License for most of our software; it applies also to any other work
released this way by its authors. You can apply it to your programs, too.
When we speak of free software, we are referring to freedom, not price. Our General
Public Licenses are designed to make sure that you have the freedom to distribute copies
of free software (and charge for them if you wish), that you receive source code or can get
it if you want it, that you can change the software or use pieces of it in new free programs,
and that you know you can do these things.
To protect your rights, we need to prevent others from denying you these rights or asking
you to surrender the rights. Therefore, you have certain responsibilities if you distribute
copies of the software, or if you modify it: responsibilities to respect the freedom of others.
For example, if you distribute copies of such a program, whether gratis or for a fee, you
must pass on to the recipients the same freedoms that you received. You must make sure
that they, too, receive or can get the source code. And you must show them these terms so
they know their rights.
Developers that use the GNU GPL protect your rights with two steps: (1) assert copy-
right on the software, and (2) offer you this License giving you legal permission to copy,
distribute and/or modify it.
For the developers’ and authors’ protection, the GPL clearly explains that there is no
warranty for this free software. For both users’ and authors’ sake, the GPL requires that
modified versions be marked as changed, so that their problems will not be attributed
erroneously to authors of previous versions.
Some devices are designed to deny users access to install or run modified versions of the
software inside them, although the manufacturer can do so. This is fundamentally incom-
patible with the aim of protecting users’ freedom to change the software. The systematic
pattern of such abuse occurs in the area of products for individuals to use, which is pre-
cisely where it is most unacceptable. Therefore, we have designed this version of the GPL
to prohibit the practice for those products. If such problems arise substantially in other
domains, we stand ready to extend this provision to those domains in future versions of the
GPL, as needed to protect the freedom of users.
304 The GNU Fortran Compiler

Finally, every program is threatened constantly by software patents. States should not
allow patents to restrict development and use of software on general-purpose computers, but
in those that do, we wish to avoid the special danger that patents applied to a free program
could make it effectively proprietary. To prevent this, the GPL assures that patents cannot
be used to render the program non-free.
The precise terms and conditions for copying, distribution and modification follow.

TERMS AND CONDITIONS

0. Definitions.
“This License” refers to version 3 of the GNU General Public License.
“Copyright” also means copyright-like laws that apply to other kinds of works, such as
semiconductor masks.
“The Program” refers to any copyrightable work licensed under this License. Each
licensee is addressed as “you”. “Licensees” and “recipients” may be individuals or
organizations.
To “modify” a work means to copy from or adapt all or part of the work in a fashion
requiring copyright permission, other than the making of an exact copy. The resulting
work is called a “modified version” of the earlier work or a work “based on” the earlier
work.
A “covered work” means either the unmodified Program or a work based on the Pro-
gram.
To “propagate” a work means to do anything with it that, without permission, would
make you directly or secondarily liable for infringement under applicable copyright law,
except executing it on a computer or modifying a private copy. Propagation includes
copying, distribution (with or without modification), making available to the public,
and in some countries other activities as well.
To “convey” a work means any kind of propagation that enables other parties to make
or receive copies. Mere interaction with a user through a computer network, with no
transfer of a copy, is not conveying.
An interactive user interface displays “Appropriate Legal Notices” to the extent that it
includes a convenient and prominently visible feature that (1) displays an appropriate
copyright notice, and (2) tells the user that there is no warranty for the work (except
to the extent that warranties are provided), that licensees may convey the work under
this License, and how to view a copy of this License. If the interface presents a list
of user commands or options, such as a menu, a prominent item in the list meets this
criterion.
1. Source Code.
The “source code” for a work means the preferred form of the work for making modi-
fications to it. “Object code” means any non-source form of a work.
A “Standard Interface” means an interface that either is an official standard defined
by a recognized standards body, or, in the case of interfaces specified for a particular
programming language, one that is widely used among developers working in that
language.
GNU General Public License 305

The “System Libraries” of an executable work include anything, other than the work as
a whole, that (a) is included in the normal form of packaging a Major Component, but
which is not part of that Major Component, and (b) serves only to enable use of the
work with that Major Component, or to implement a Standard Interface for which an
implementation is available to the public in source code form. A “Major Component”,
in this context, means a major essential component (kernel, window system, and so
on) of the specific operating system (if any) on which the executable work runs, or a
compiler used to produce the work, or an object code interpreter used to run it.
The “Corresponding Source” for a work in object code form means all the source code
needed to generate, install, and (for an executable work) run the object code and to
modify the work, including scripts to control those activities. However, it does not
include the work’s System Libraries, or general-purpose tools or generally available
free programs which are used unmodified in performing those activities but which are
not part of the work. For example, Corresponding Source includes interface definition
files associated with source files for the work, and the source code for shared libraries
and dynamically linked subprograms that the work is specifically designed to require,
such as by intimate data communication or control flow between those subprograms
and other parts of the work.
The Corresponding Source need not include anything that users can regenerate auto-
matically from other parts of the Corresponding Source.
The Corresponding Source for a work in source code form is that same work.
2. Basic Permissions.
All rights granted under this License are granted for the term of copyright on the
Program, and are irrevocable provided the stated conditions are met. This License ex-
plicitly affirms your unlimited permission to run the unmodified Program. The output
from running a covered work is covered by this License only if the output, given its
content, constitutes a covered work. This License acknowledges your rights of fair use
or other equivalent, as provided by copyright law.
You may make, run and propagate covered works that you do not convey, without
conditions so long as your license otherwise remains in force. You may convey covered
works to others for the sole purpose of having them make modifications exclusively
for you, or provide you with facilities for running those works, provided that you
comply with the terms of this License in conveying all material for which you do not
control copyright. Those thus making or running the covered works for you must do
so exclusively on your behalf, under your direction and control, on terms that prohibit
them from making any copies of your copyrighted material outside their relationship
with you.
Conveying under any other circumstances is permitted solely under the conditions
stated below. Sublicensing is not allowed; section 10 makes it unnecessary.
3. Protecting Users’ Legal Rights From Anti-Circumvention Law.
No covered work shall be deemed part of an effective technological measure under
any applicable law fulfilling obligations under article 11 of the WIPO copyright treaty
adopted on 20 December 1996, or similar laws prohibiting or restricting circumvention
of such measures.
306 The GNU Fortran Compiler

When you convey a covered work, you waive any legal power to forbid circumvention of
technological measures to the extent such circumvention is effected by exercising rights
under this License with respect to the covered work, and you disclaim any intention
to limit operation or modification of the work as a means of enforcing, against the
work’s users, your or third parties’ legal rights to forbid circumvention of technological
measures.
4. Conveying Verbatim Copies.
You may convey verbatim copies of the Program’s source code as you receive it, in any
medium, provided that you conspicuously and appropriately publish on each copy an
appropriate copyright notice; keep intact all notices stating that this License and any
non-permissive terms added in accord with section 7 apply to the code; keep intact all
notices of the absence of any warranty; and give all recipients a copy of this License
along with the Program.
You may charge any price or no price for each copy that you convey, and you may offer
support or warranty protection for a fee.
5. Conveying Modified Source Versions.
You may convey a work based on the Program, or the modifications to produce it from
the Program, in the form of source code under the terms of section 4, provided that
you also meet all of these conditions:
a. The work must carry prominent notices stating that you modified it, and giving a
relevant date.
b. The work must carry prominent notices stating that it is released under this Li-
cense and any conditions added under section 7. This requirement modifies the
requirement in section 4 to “keep intact all notices”.
c. You must license the entire work, as a whole, under this License to anyone who
comes into possession of a copy. This License will therefore apply, along with any
applicable section 7 additional terms, to the whole of the work, and all its parts,
regardless of how they are packaged. This License gives no permission to license
the work in any other way, but it does not invalidate such permission if you have
separately received it.
d. If the work has interactive user interfaces, each must display Appropriate Legal
Notices; however, if the Program has interactive interfaces that do not display
Appropriate Legal Notices, your work need not make them do so.
A compilation of a covered work with other separate and independent works, which
are not by their nature extensions of the covered work, and which are not combined
with it such as to form a larger program, in or on a volume of a storage or distribution
medium, is called an “aggregate” if the compilation and its resulting copyright are
not used to limit the access or legal rights of the compilation’s users beyond what the
individual works permit. Inclusion of a covered work in an aggregate does not cause
this License to apply to the other parts of the aggregate.
6. Conveying Non-Source Forms.
You may convey a covered work in object code form under the terms of sections 4 and
5, provided that you also convey the machine-readable Corresponding Source under
the terms of this License, in one of these ways:
GNU General Public License 307

a. Convey the object code in, or embodied in, a physical product (including a phys-
ical distribution medium), accompanied by the Corresponding Source fixed on a
durable physical medium customarily used for software interchange.
b. Convey the object code in, or embodied in, a physical product (including a physi-
cal distribution medium), accompanied by a written offer, valid for at least three
years and valid for as long as you offer spare parts or customer support for that
product model, to give anyone who possesses the object code either (1) a copy of
the Corresponding Source for all the software in the product that is covered by this
License, on a durable physical medium customarily used for software interchange,
for a price no more than your reasonable cost of physically performing this con-
veying of source, or (2) access to copy the Corresponding Source from a network
server at no charge.
c. Convey individual copies of the object code with a copy of the written offer to
provide the Corresponding Source. This alternative is allowed only occasionally
and noncommercially, and only if you received the object code with such an offer,
in accord with subsection 6b.
d. Convey the object code by offering access from a designated place (gratis or for
a charge), and offer equivalent access to the Corresponding Source in the same
way through the same place at no further charge. You need not require recipients
to copy the Corresponding Source along with the object code. If the place to
copy the object code is a network server, the Corresponding Source may be on
a different server (operated by you or a third party) that supports equivalent
copying facilities, provided you maintain clear directions next to the object code
saying where to find the Corresponding Source. Regardless of what server hosts
the Corresponding Source, you remain obligated to ensure that it is available for
as long as needed to satisfy these requirements.
e. Convey the object code using peer-to-peer transmission, provided you inform other
peers where the object code and Corresponding Source of the work are being offered
to the general public at no charge under subsection 6d.
A separable portion of the object code, whose source code is excluded from the Cor-
responding Source as a System Library, need not be included in conveying the object
code work.
A “User Product” is either (1) a “consumer product”, which means any tangible per-
sonal property which is normally used for personal, family, or household purposes, or
(2) anything designed or sold for incorporation into a dwelling. In determining whether
a product is a consumer product, doubtful cases shall be resolved in favor of coverage.
For a particular product received by a particular user, “normally used” refers to a
typical or common use of that class of product, regardless of the status of the par-
ticular user or of the way in which the particular user actually uses, or expects or is
expected to use, the product. A product is a consumer product regardless of whether
the product has substantial commercial, industrial or non-consumer uses, unless such
uses represent the only significant mode of use of the product.
“Installation Information” for a User Product means any methods, procedures, autho-
rization keys, or other information required to install and execute modified versions of a
covered work in that User Product from a modified version of its Corresponding Source.
308 The GNU Fortran Compiler

The information must suffice to ensure that the continued functioning of the modified
object code is in no case prevented or interfered with solely because modification has
been made.
If you convey an object code work under this section in, or with, or specifically for
use in, a User Product, and the conveying occurs as part of a transaction in which
the right of possession and use of the User Product is transferred to the recipient in
perpetuity or for a fixed term (regardless of how the transaction is characterized),
the Corresponding Source conveyed under this section must be accompanied by the
Installation Information. But this requirement does not apply if neither you nor any
third party retains the ability to install modified object code on the User Product (for
example, the work has been installed in ROM).
The requirement to provide Installation Information does not include a requirement
to continue to provide support service, warranty, or updates for a work that has been
modified or installed by the recipient, or for the User Product in which it has been
modified or installed. Access to a network may be denied when the modification itself
materially and adversely affects the operation of the network or violates the rules and
protocols for communication across the network.
Corresponding Source conveyed, and Installation Information provided, in accord with
this section must be in a format that is publicly documented (and with an implementa-
tion available to the public in source code form), and must require no special password
or key for unpacking, reading or copying.
7. Additional Terms.
“Additional permissions” are terms that supplement the terms of this License by mak-
ing exceptions from one or more of its conditions. Additional permissions that are
applicable to the entire Program shall be treated as though they were included in this
License, to the extent that they are valid under applicable law. If additional permis-
sions apply only to part of the Program, that part may be used separately under those
permissions, but the entire Program remains governed by this License without regard
to the additional permissions.
When you convey a copy of a covered work, you may at your option remove any
additional permissions from that copy, or from any part of it. (Additional permissions
may be written to require their own removal in certain cases when you modify the
work.) You may place additional permissions on material, added by you to a covered
work, for which you have or can give appropriate copyright permission.
Notwithstanding any other provision of this License, for material you add to a covered
work, you may (if authorized by the copyright holders of that material) supplement
the terms of this License with terms:
a. Disclaiming warranty or limiting liability differently from the terms of sections 15
and 16 of this License; or
b. Requiring preservation of specified reasonable legal notices or author attributions
in that material or in the Appropriate Legal Notices displayed by works containing
it; or
c. Prohibiting misrepresentation of the origin of that material, or requiring that mod-
ified versions of such material be marked in reasonable ways as different from the
original version; or
GNU General Public License 309

d. Limiting the use for publicity purposes of names of licensors or authors of the
material; or
e. Declining to grant rights under trademark law for use of some trade names, trade-
marks, or service marks; or
f. Requiring indemnification of licensors and authors of that material by anyone who
conveys the material (or modified versions of it) with contractual assumptions
of liability to the recipient, for any liability that these contractual assumptions
directly impose on those licensors and authors.
All other non-permissive additional terms are considered “further restrictions” within
the meaning of section 10. If the Program as you received it, or any part of it, con-
tains a notice stating that it is governed by this License along with a term that is a
further restriction, you may remove that term. If a license document contains a further
restriction but permits relicensing or conveying under this License, you may add to a
covered work material governed by the terms of that license document, provided that
the further restriction does not survive such relicensing or conveying.
If you add terms to a covered work in accord with this section, you must place, in the
relevant source files, a statement of the additional terms that apply to those files, or a
notice indicating where to find the applicable terms.
Additional terms, permissive or non-permissive, may be stated in the form of a sep-
arately written license, or stated as exceptions; the above requirements apply either
way.
8. Termination.
You may not propagate or modify a covered work except as expressly provided un-
der this License. Any attempt otherwise to propagate or modify it is void, and will
automatically terminate your rights under this License (including any patent licenses
granted under the third paragraph of section 11).
However, if you cease all violation of this License, then your license from a particular
copyright holder is reinstated (a) provisionally, unless and until the copyright holder
explicitly and finally terminates your license, and (b) permanently, if the copyright
holder fails to notify you of the violation by some reasonable means prior to 60 days
after the cessation.
Moreover, your license from a particular copyright holder is reinstated permanently if
the copyright holder notifies you of the violation by some reasonable means, this is the
first time you have received notice of violation of this License (for any work) from that
copyright holder, and you cure the violation prior to 30 days after your receipt of the
notice.
Termination of your rights under this section does not terminate the licenses of parties
who have received copies or rights from you under this License. If your rights have
been terminated and not permanently reinstated, you do not qualify to receive new
licenses for the same material under section 10.
9. Acceptance Not Required for Having Copies.
You are not required to accept this License in order to receive or run a copy of the
Program. Ancillary propagation of a covered work occurring solely as a consequence of
using peer-to-peer transmission to receive a copy likewise does not require acceptance.
310 The GNU Fortran Compiler

However, nothing other than this License grants you permission to propagate or modify
any covered work. These actions infringe copyright if you do not accept this License.
Therefore, by modifying or propagating a covered work, you indicate your acceptance
of this License to do so.
10. Automatic Licensing of Downstream Recipients.
Each time you convey a covered work, the recipient automatically receives a license
from the original licensors, to run, modify and propagate that work, subject to this
License. You are not responsible for enforcing compliance by third parties with this
License.
An “entity transaction” is a transaction transferring control of an organization, or
substantially all assets of one, or subdividing an organization, or merging organizations.
If propagation of a covered work results from an entity transaction, each party to that
transaction who receives a copy of the work also receives whatever licenses to the work
the party’s predecessor in interest had or could give under the previous paragraph, plus
a right to possession of the Corresponding Source of the work from the predecessor in
interest, if the predecessor has it or can get it with reasonable efforts.
You may not impose any further restrictions on the exercise of the rights granted or
affirmed under this License. For example, you may not impose a license fee, royalty, or
other charge for exercise of rights granted under this License, and you may not initiate
litigation (including a cross-claim or counterclaim in a lawsuit) alleging that any patent
claim is infringed by making, using, selling, offering for sale, or importing the Program
or any portion of it.
11. Patents.
A “contributor” is a copyright holder who authorizes use under this License of the
Program or a work on which the Program is based. The work thus licensed is called
the contributor’s “contributor version”.
A contributor’s “essential patent claims” are all patent claims owned or controlled by
the contributor, whether already acquired or hereafter acquired, that would be infringed
by some manner, permitted by this License, of making, using, or selling its contributor
version, but do not include claims that would be infringed only as a consequence of
further modification of the contributor version. For purposes of this definition, “con-
trol” includes the right to grant patent sublicenses in a manner consistent with the
requirements of this License.
Each contributor grants you a non-exclusive, worldwide, royalty-free patent license
under the contributor’s essential patent claims, to make, use, sell, offer for sale, import
and otherwise run, modify and propagate the contents of its contributor version.
In the following three paragraphs, a “patent license” is any express agreement or com-
mitment, however denominated, not to enforce a patent (such as an express permission
to practice a patent or covenant not to sue for patent infringement). To “grant” such
a patent license to a party means to make such an agreement or commitment not to
enforce a patent against the party.
If you convey a covered work, knowingly relying on a patent license, and the Corre-
sponding Source of the work is not available for anyone to copy, free of charge and under
the terms of this License, through a publicly available network server or other readily
accessible means, then you must either (1) cause the Corresponding Source to be so
GNU General Public License 311

available, or (2) arrange to deprive yourself of the benefit of the patent license for this
particular work, or (3) arrange, in a manner consistent with the requirements of this
License, to extend the patent license to downstream recipients. “Knowingly relying”
means you have actual knowledge that, but for the patent license, your conveying the
covered work in a country, or your recipient’s use of the covered work in a country,
would infringe one or more identifiable patents in that country that you have reason
to believe are valid.
If, pursuant to or in connection with a single transaction or arrangement, you convey,
or propagate by procuring conveyance of, a covered work, and grant a patent license
to some of the parties receiving the covered work authorizing them to use, propagate,
modify or convey a specific copy of the covered work, then the patent license you grant
is automatically extended to all recipients of the covered work and works based on it.
A patent license is “discriminatory” if it does not include within the scope of its cover-
age, prohibits the exercise of, or is conditioned on the non-exercise of one or more of the
rights that are specifically granted under this License. You may not convey a covered
work if you are a party to an arrangement with a third party that is in the business of
distributing software, under which you make payment to the third party based on the
extent of your activity of conveying the work, and under which the third party grants,
to any of the parties who would receive the covered work from you, a discriminatory
patent license (a) in connection with copies of the covered work conveyed by you (or
copies made from those copies), or (b) primarily for and in connection with specific
products or compilations that contain the covered work, unless you entered into that
arrangement, or that patent license was granted, prior to 28 March 2007.
Nothing in this License shall be construed as excluding or limiting any implied license or
other defenses to infringement that may otherwise be available to you under applicable
patent law.
12. No Surrender of Others’ Freedom.
If conditions are imposed on you (whether by court order, agreement or otherwise) that
contradict the conditions of this License, they do not excuse you from the conditions
of this License. If you cannot convey a covered work so as to satisfy simultaneously
your obligations under this License and any other pertinent obligations, then as a
consequence you may not convey it at all. For example, if you agree to terms that
obligate you to collect a royalty for further conveying from those to whom you convey
the Program, the only way you could satisfy both those terms and this License would
be to refrain entirely from conveying the Program.
13. Use with the GNU Affero General Public License.
Notwithstanding any other provision of this License, you have permission to link or
combine any covered work with a work licensed under version 3 of the GNU Affero
General Public License into a single combined work, and to convey the resulting work.
The terms of this License will continue to apply to the part which is the covered work,
but the special requirements of the GNU Affero General Public License, section 13,
concerning interaction through a network will apply to the combination as such.
14. Revised Versions of this License.
312 The GNU Fortran Compiler

The Free Software Foundation may publish revised and/or new versions of the GNU
General Public License from time to time. Such new versions will be similar in spirit
to the present version, but may differ in detail to address new problems or concerns.
Each version is given a distinguishing version number. If the Program specifies that
a certain numbered version of the GNU General Public License “or any later version”
applies to it, you have the option of following the terms and conditions either of that
numbered version or of any later version published by the Free Software Foundation.
If the Program does not specify a version number of the GNU General Public License,
you may choose any version ever published by the Free Software Foundation.
If the Program specifies that a proxy can decide which future versions of the GNU
General Public License can be used, that proxy’s public statement of acceptance of a
version permanently authorizes you to choose that version for the Program.
Later license versions may give you additional or different permissions. However, no
additional obligations are imposed on any author or copyright holder as a result of your
choosing to follow a later version.
15. Disclaimer of Warranty.
THERE IS NO WARRANTY FOR THE PROGRAM, TO THE EXTENT PER-
MITTED BY APPLICABLE LAW. EXCEPT WHEN OTHERWISE STATED IN
WRITING THE COPYRIGHT HOLDERS AND/OR OTHER PARTIES PROVIDE
THE PROGRAM “AS IS” WITHOUT WARRANTY OF ANY KIND, EITHER EX-
PRESSED OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
PURPOSE. THE ENTIRE RISK AS TO THE QUALITY AND PERFORMANCE
OF THE PROGRAM IS WITH YOU. SHOULD THE PROGRAM PROVE DEFEC-
TIVE, YOU ASSUME THE COST OF ALL NECESSARY SERVICING, REPAIR OR
CORRECTION.
16. Limitation of Liability.
IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN
WRITING WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO
MODIFIES AND/OR CONVEYS THE PROGRAM AS PERMITTED ABOVE, BE
LIABLE TO YOU FOR DAMAGES, INCLUDING ANY GENERAL, SPECIAL, IN-
CIDENTAL OR CONSEQUENTIAL DAMAGES ARISING OUT OF THE USE OR
INABILITY TO USE THE PROGRAM (INCLUDING BUT NOT LIMITED TO
LOSS OF DATA OR DATA BEING RENDERED INACCURATE OR LOSSES SUS-
TAINED BY YOU OR THIRD PARTIES OR A FAILURE OF THE PROGRAM
TO OPERATE WITH ANY OTHER PROGRAMS), EVEN IF SUCH HOLDER OR
OTHER PARTY HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAM-
AGES.
17. Interpretation of Sections 15 and 16.
If the disclaimer of warranty and limitation of liability provided above cannot be given
local legal effect according to their terms, reviewing courts shall apply local law that
most closely approximates an absolute waiver of all civil liability in connection with
the Program, unless a warranty or assumption of liability accompanies a copy of the
Program in return for a fee.
GNU General Public License 313

END OF TERMS AND CONDITIONS

How to Apply These Terms to Your New Programs

If you develop a new program, and you want it to be of the greatest possible use to the public,
the best way to achieve this is to make it free software which everyone can redistribute and
change under these terms.
To do so, attach the following notices to the program. It is safest to attach them to the
start of each source file to most effectively state the exclusion of warranty; and each file
should have at least the “copyright” line and a pointer to where the full notice is found.
one line to give the program's name and a brief idea of what it does.
Copyright (C) year name of author

This program is free software: you can redistribute it and/or modify

it under the terms of the GNU General Public License as published by
the Free Software Foundation, either version 3 of the License, or (at
your option) any later version.

This program is distributed in the hope that it will be useful, but

WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
General Public License for more details.

You should have received a copy of the GNU General Public License
along with this program. If not, see https://www.gnu.org/licenses/.
Also add information on how to contact you by electronic and paper mail.
If the program does terminal interaction, make it output a short notice like this when it
starts in an interactive mode:
program Copyright (C) year name of author
This program comes with ABSOLUTELY NO WARRANTY; for details type ‘show w’.
This is free software, and you are welcome to redistribute it
under certain conditions; type ‘show c’ for details.
The hypothetical commands ‘show w’ and ‘show c’ should show the appropriate parts of
the General Public License. Of course, your program’s commands might be different; for a
GUI interface, you would use an “about box”.
You should also get your employer (if you work as a programmer) or school, if any, to
sign a “copyright disclaimer” for the program, if necessary. For more information on this,
and how to apply and follow the GNU GPL, see https://www.gnu.org/licenses/.
The GNU General Public License does not permit incorporating your program into pro-
prietary programs. If your program is a subroutine library, you may consider it more useful
to permit linking proprietary applications with the library. If this is what you want to do,
use the GNU Lesser General Public License instead of this License. But first, please read
https://www.gnu.org/licenses/why-not-lgpl.html.
 315

GNU Free Documentation License

Version 1.3, 3 November 2008
Copyright c 2000, 2001, 2002, 2007, 2008 Free Software Foundation, Inc.
https://www.fsf.org

Everyone is permitted to copy and distribute verbatim copies

of this license document, but changing it is not allowed.
0. PREAMBLE
The purpose of this License is to make a manual, textbook, or other functional and
useful document free in the sense of freedom: to assure everyone the effective freedom
to copy and redistribute it, with or without modifying it, either commercially or non-
commercially. Secondarily, this License preserves for the author and publisher a way
to get credit for their work, while not being considered responsible for modifications
made by others.
This License is a kind of “copyleft”, which means that derivative works of the document
must themselves be free in the same sense. It complements the GNU General Public
License, which is a copyleft license designed for free software.
We have designed this License in order to use it for manuals for free software, because
free software needs free documentation: a free program should come with manuals
providing the same freedoms that the software does. But this License is not limited to
software manuals; it can be used for any textual work, regardless of subject matter or
whether it is published as a printed book. We recommend this License principally for
works whose purpose is instruction or reference.
1. APPLICABILITY AND DEFINITIONS
This License applies to any manual or other work, in any medium, that contains a
notice placed by the copyright holder saying it can be distributed under the terms
of this License. Such a notice grants a world-wide, royalty-free license, unlimited in
duration, to use that work under the conditions stated herein. The “Document”,
below, refers to any such manual or work. Any member of the public is a licensee, and
is addressed as “you”. You accept the license if you copy, modify or distribute the work
in a way requiring permission under copyright law.
A “Modified Version” of the Document means any work containing the Document or
a portion of it, either copied verbatim, or with modifications and/or translated into
another language.
A “Secondary Section” is a named appendix or a front-matter section of the Document
that deals exclusively with the relationship of the publishers or authors of the Document
to the Document’s overall subject (or to related matters) and contains nothing that
could fall directly within that overall subject. (Thus, if the Document is in part a
textbook of mathematics, a Secondary Section may not explain any mathematics.) The
relationship could be a matter of historical connection with the subject or with related
matters, or of legal, commercial, philosophical, ethical or political position regarding
them.
The “Invariant Sections” are certain Secondary Sections whose titles are designated, as
being those of Invariant Sections, in the notice that says that the Document is released
316 The GNU Fortran Compiler

under this License. If a section does not fit the above definition of Secondary then it is
not allowed to be designated as Invariant. The Document may contain zero Invariant
Sections. If the Document does not identify any Invariant Sections then there are none.
The “Cover Texts” are certain short passages of text that are listed, as Front-Cover
Texts or Back-Cover Texts, in the notice that says that the Document is released under
this License. A Front-Cover Text may be at most 5 words, and a Back-Cover Text may
be at most 25 words.
A “Transparent” copy of the Document means a machine-readable copy, represented
in a format whose specification is available to the general public, that is suitable for
revising the document straightforwardly with generic text editors or (for images com-
posed of pixels) generic paint programs or (for drawings) some widely available drawing
editor, and that is suitable for input to text formatters or for automatic translation to
a variety of formats suitable for input to text formatters. A copy made in an otherwise
Transparent file format whose markup, or absence of markup, has been arranged to
thwart or discourage subsequent modification by readers is not Transparent. An image
format is not Transparent if used for any substantial amount of text. A copy that is
not “Transparent” is called “Opaque”.
Examples of suitable formats for Transparent copies include plain ascii without
markup, Texinfo input format, LaTEX input format, SGML or XML using a publicly
available DTD, and standard-conforming simple HTML, PostScript or PDF designed
for human modification. Examples of transparent image formats include PNG, XCF
and JPG. Opaque formats include proprietary formats that can be read and edited
only by proprietary word processors, SGML or XML for which the DTD and/or
processing tools are not generally available, and the machine-generated HTML,
PostScript or PDF produced by some word processors for output purposes only.
The “Title Page” means, for a printed book, the title page itself, plus such following
pages as are needed to hold, legibly, the material this License requires to appear in the
title page. For works in formats which do not have any title page as such, “Title Page”
means the text near the most prominent appearance of the work’s title, preceding the
beginning of the body of the text.
The “publisher” means any person or entity that distributes copies of the Document
to the public.
A section “Entitled XYZ” means a named subunit of the Document whose title either
is precisely XYZ or contains XYZ in parentheses following text that translates XYZ in
another language. (Here XYZ stands for a specific section name mentioned below, such
as “Acknowledgements”, “Dedications”, “Endorsements”, or “History”.) To “Preserve
the Title” of such a section when you modify the Document means that it remains a
section “Entitled XYZ” according to this definition.
The Document may include Warranty Disclaimers next to the notice which states that
this License applies to the Document. These Warranty Disclaimers are considered to
be included by reference in this License, but only as regards disclaiming warranties:
any other implication that these Warranty Disclaimers may have is void and has no
effect on the meaning of this License.
2. VERBATIM COPYING
GNU Free Documentation License 317

You may copy and distribute the Document in any medium, either commercially or
noncommercially, provided that this License, the copyright notices, and the license
notice saying this License applies to the Document are reproduced in all copies, and
that you add no other conditions whatsoever to those of this License. You may not use
technical measures to obstruct or control the reading or further copying of the copies
you make or distribute. However, you may accept compensation in exchange for copies.
If you distribute a large enough number of copies you must also follow the conditions
in section 3.
You may also lend copies, under the same conditions stated above, and you may publicly
display copies.
3. COPYING IN QUANTITY
If you publish printed copies (or copies in media that commonly have printed covers) of
the Document, numbering more than 100, and the Document’s license notice requires
Cover Texts, you must enclose the copies in covers that carry, clearly and legibly, all
these Cover Texts: Front-Cover Texts on the front cover, and Back-Cover Texts on
the back cover. Both covers must also clearly and legibly identify you as the publisher
of these copies. The front cover must present the full title with all words of the title
equally prominent and visible. You may add other material on the covers in addition.
Copying with changes limited to the covers, as long as they preserve the title of the
Document and satisfy these conditions, can be treated as verbatim copying in other
respects.
If the required texts for either cover are too voluminous to fit legibly, you should put
the first ones listed (as many as fit reasonably) on the actual cover, and continue the
rest onto adjacent pages.
If you publish or distribute Opaque copies of the Document numbering more than 100,
you must either include a machine-readable Transparent copy along with each Opaque
copy, or state in or with each Opaque copy a computer-network location from which
the general network-using public has access to download using public-standard network
protocols a complete Transparent copy of the Document, free of added material. If
you use the latter option, you must take reasonably prudent steps, when you begin
distribution of Opaque copies in quantity, to ensure that this Transparent copy will
remain thus accessible at the stated location until at least one year after the last time
you distribute an Opaque copy (directly or through your agents or retailers) of that
edition to the public.
It is requested, but not required, that you contact the authors of the Document well
before redistributing any large number of copies, to give them a chance to provide you
with an updated version of the Document.
4. MODIFICATIONS
You may copy and distribute a Modified Version of the Document under the conditions
of sections 2 and 3 above, provided that you release the Modified Version under precisely
this License, with the Modified Version filling the role of the Document, thus licensing
distribution and modification of the Modified Version to whoever possesses a copy of
it. In addition, you must do these things in the Modified Version:
A. Use in the Title Page (and on the covers, if any) a title distinct from that of the
Document, and from those of previous versions (which should, if there were any,
318 The GNU Fortran Compiler

be listed in the History section of the Document). You may use the same title as
a previous version if the original publisher of that version gives permission.
B. List on the Title Page, as authors, one or more persons or entities responsible for
authorship of the modifications in the Modified Version, together with at least five
of the principal authors of the Document (all of its principal authors, if it has fewer
than five), unless they release you from this requirement.
C. State on the Title page the name of the publisher of the Modified Version, as the
publisher.
D. Preserve all the copyright notices of the Document.
E. Add an appropriate copyright notice for your modifications adjacent to the other
copyright notices.
F. Include, immediately after the copyright notices, a license notice giving the public
permission to use the Modified Version under the terms of this License, in the form
shown in the Addendum below.
G. Preserve in that license notice the full lists of Invariant Sections and required Cover
Texts given in the Document’s license notice.
H. Include an unaltered copy of this License.
I. Preserve the section Entitled “History”, Preserve its Title, and add to it an item
stating at least the title, year, new authors, and publisher of the Modified Version
as given on the Title Page. If there is no section Entitled “History” in the Docu-
ment, create one stating the title, year, authors, and publisher of the Document
as given on its Title Page, then add an item describing the Modified Version as
stated in the previous sentence.
J. Preserve the network location, if any, given in the Document for public access to
a Transparent copy of the Document, and likewise the network locations given in
the Document for previous versions it was based on. These may be placed in the
“History” section. You may omit a network location for a work that was published
at least four years before the Document itself, or if the original publisher of the
version it refers to gives permission.
K. For any section Entitled “Acknowledgements” or “Dedications”, Preserve the Title
of the section, and preserve in the section all the substance and tone of each of the
contributor acknowledgements and/or dedications given therein.
L. Preserve all the Invariant Sections of the Document, unaltered in their text and
in their titles. Section numbers or the equivalent are not considered part of the
section titles.
M. Delete any section Entitled “Endorsements”. Such a section may not be included
in the Modified Version.
N. Do not retitle any existing section to be Entitled “Endorsements” or to conflict in
title with any Invariant Section.
O. Preserve any Warranty Disclaimers.
If the Modified Version includes new front-matter sections or appendices that qualify
as Secondary Sections and contain no material copied from the Document, you may at
your option designate some or all of these sections as invariant. To do this, add their
GNU Free Documentation License 319

titles to the list of Invariant Sections in the Modified Version’s license notice. These
titles must be distinct from any other section titles.
You may add a section Entitled “Endorsements”, provided it contains nothing but
endorsements of your Modified Version by various parties—for example, statements of
peer review or that the text has been approved by an organization as the authoritative
definition of a standard.
You may add a passage of up to five words as a Front-Cover Text, and a passage of up
to 25 words as a Back-Cover Text, to the end of the list of Cover Texts in the Modified
Version. Only one passage of Front-Cover Text and one of Back-Cover Text may be
added by (or through arrangements made by) any one entity. If the Document already
includes a cover text for the same cover, previously added by you or by arrangement
made by the same entity you are acting on behalf of, you may not add another; but
you may replace the old one, on explicit permission from the previous publisher that
added the old one.
The author(s) and publisher(s) of the Document do not by this License give permission
to use their names for publicity for or to assert or imply endorsement of any Modified
Version.
5. COMBINING DOCUMENTS
You may combine the Document with other documents released under this License,
under the terms defined in section 4 above for modified versions, provided that you
include in the combination all of the Invariant Sections of all of the original documents,
unmodified, and list them all as Invariant Sections of your combined work in its license
notice, and that you preserve all their Warranty Disclaimers.
The combined work need only contain one copy of this License, and multiple identical
Invariant Sections may be replaced with a single copy. If there are multiple Invariant
Sections with the same name but different contents, make the title of each such section
unique by adding at the end of it, in parentheses, the name of the original author or
publisher of that section if known, or else a unique number. Make the same adjustment
to the section titles in the list of Invariant Sections in the license notice of the combined
work.
In the combination, you must combine any sections Entitled “History” in the vari-
ous original documents, forming one section Entitled “History”; likewise combine any
sections Entitled “Acknowledgements”, and any sections Entitled “Dedications”. You
must delete all sections Entitled “Endorsements.”
6. COLLECTIONS OF DOCUMENTS
You may make a collection consisting of the Document and other documents released
under this License, and replace the individual copies of this License in the various
documents with a single copy that is included in the collection, provided that you
follow the rules of this License for verbatim copying of each of the documents in all
other respects.
You may extract a single document from such a collection, and distribute it individu-
ally under this License, provided you insert a copy of this License into the extracted
document, and follow this License in all other respects regarding verbatim copying of
that document.
320 The GNU Fortran Compiler

7. AGGREGATION WITH INDEPENDENT WORKS

A compilation of the Document or its derivatives with other separate and independent
documents or works, in or on a volume of a storage or distribution medium, is called
an “aggregate” if the copyright resulting from the compilation is not used to limit the
legal rights of the compilation’s users beyond what the individual works permit. When
the Document is included in an aggregate, this License does not apply to the other
works in the aggregate which are not themselves derivative works of the Document.
If the Cover Text requirement of section 3 is applicable to these copies of the Document,
then if the Document is less than one half of the entire aggregate, the Document’s Cover
Texts may be placed on covers that bracket the Document within the aggregate, or the
electronic equivalent of covers if the Document is in electronic form. Otherwise they
must appear on printed covers that bracket the whole aggregate.
8. TRANSLATION
Translation is considered a kind of modification, so you may distribute translations
of the Document under the terms of section 4. Replacing Invariant Sections with
translations requires special permission from their copyright holders, but you may
include translations of some or all Invariant Sections in addition to the original versions
of these Invariant Sections. You may include a translation of this License, and all the
license notices in the Document, and any Warranty Disclaimers, provided that you
also include the original English version of this License and the original versions of
those notices and disclaimers. In case of a disagreement between the translation and
the original version of this License or a notice or disclaimer, the original version will
prevail.
If a section in the Document is Entitled “Acknowledgements”, “Dedications”, or “His-
tory”, the requirement (section 4) to Preserve its Title (section 1) will typically require
changing the actual title.
9. TERMINATION
You may not copy, modify, sublicense, or distribute the Document except as expressly
provided under this License. Any attempt otherwise to copy, modify, sublicense, or
distribute it is void, and will automatically terminate your rights under this License.
However, if you cease all violation of this License, then your license from a particular
copyright holder is reinstated (a) provisionally, unless and until the copyright holder
explicitly and finally terminates your license, and (b) permanently, if the copyright
holder fails to notify you of the violation by some reasonable means prior to 60 days
after the cessation.
Moreover, your license from a particular copyright holder is reinstated permanently if
the copyright holder notifies you of the violation by some reasonable means, this is the
first time you have received notice of violation of this License (for any work) from that
copyright holder, and you cure the violation prior to 30 days after your receipt of the
notice.
Termination of your rights under this section does not terminate the licenses of parties
who have received copies or rights from you under this License. If your rights have
been terminated and not permanently reinstated, receipt of a copy of some or all of the
same material does not give you any rights to use it.
GNU Free Documentation License 321

10. FUTURE REVISIONS OF THIS LICENSE

The Free Software Foundation may publish new, revised versions of the GNU Free
Documentation License from time to time. Such new versions will be similar in spirit
to the present version, but may differ in detail to address new problems or concerns.
See https://www.gnu.org/copyleft/.
Each version of the License is given a distinguishing version number. If the Document
specifies that a particular numbered version of this License “or any later version”
applies to it, you have the option of following the terms and conditions either of that
specified version or of any later version that has been published (not as a draft) by
the Free Software Foundation. If the Document does not specify a version number of
this License, you may choose any version ever published (not as a draft) by the Free
Software Foundation. If the Document specifies that a proxy can decide which future
versions of this License can be used, that proxy’s public statement of acceptance of a
version permanently authorizes you to choose that version for the Document.
11. RELICENSING
“Massive Multiauthor Collaboration Site” (or “MMC Site”) means any World Wide
Web server that publishes copyrightable works and also provides prominent facilities
for anybody to edit those works. A public wiki that anybody can edit is an example of
such a server. A “Massive Multiauthor Collaboration” (or “MMC”) contained in the
site means any set of copyrightable works thus published on the MMC site.
“CC-BY-SA” means the Creative Commons Attribution-Share Alike 3.0 license pub-
lished by Creative Commons Corporation, a not-for-profit corporation with a principal
place of business in San Francisco, California, as well as future copyleft versions of that
license published by that same organization.
“Incorporate” means to publish or republish a Document, in whole or in part, as part
of another Document.
An MMC is “eligible for relicensing” if it is licensed under this License, and if all works
that were first published under this License somewhere other than this MMC, and
subsequently incorporated in whole or in part into the MMC, (1) had no cover texts
or invariant sections, and (2) were thus incorporated prior to November 1, 2008.
The operator of an MMC Site may republish an MMC contained in the site under
CC-BY-SA on the same site at any time before August 1, 2009, provided the MMC is
eligible for relicensing.
322 The GNU Fortran Compiler

ADDENDUM: How to use this License for your documents

To use this License in a document you have written, include a copy of the License in the
document and put the following copyright and license notices just after the title page:
Copyright (C) year your name.
Permission is granted to copy, distribute and/or modify this document
under the terms of the GNU Free Documentation License, Version 1.3
or any later version published by the Free Software Foundation;
with no Invariant Sections, no Front-Cover Texts, and no Back-Cover
Texts. A copy of the license is included in the section entitled ``GNU
Free Documentation License''.
If you have Invariant Sections, Front-Cover Texts and Back-Cover Texts, replace the
“with...Texts.” line with this:
with the Invariant Sections being list their titles, with
the Front-Cover Texts being list, and with the Back-Cover Texts
being list.
If you have Invariant Sections without Cover Texts, or some other combination of the
three, merge those two alternatives to suit the situation.
If your document contains nontrivial examples of program code, we recommend releasing
these examples in parallel under your choice of free software license, such as the GNU
General Public License, to permit their use in free software.
 323

Funding Free Software

If you want to have more free software a few years from now, it makes sense for you to
help encourage people to contribute funds for its development. The most effective approach
known is to encourage commercial redistributors to donate.
Users of free software systems can boost the pace of development by encouraging for-a-
fee distributors to donate part of their selling price to free software developers—the Free
Software Foundation, and others.
The way to convince distributors to do this is to demand it and expect it from them. So
when you compare distributors, judge them partly by how much they give to free software
development. Show distributors they must compete to be the one who gives the most.
To make this approach work, you must insist on numbers that you can compare, such as,
“We will donate ten dollars to the Frobnitz project for each disk sold.” Don’t be satisfied
with a vague promise, such as “A portion of the profits are donated,” since it doesn’t give
a basis for comparison.
Even a precise fraction “of the profits from this disk” is not very meaningful, since
creative accounting and unrelated business decisions can greatly alter what fraction of the
sales price counts as profit. If the price you pay is $50, ten percent of the profit is probably
less than a dollar; it might be a few cents, or nothing at all.
Some redistributors do development work themselves. This is useful too; but to keep
everyone honest, you need to inquire how much they do, and what kind. Some kinds of
development make much more long-term difference than others. For example, maintaining
a separate version of a program contributes very little; maintaining the standard version
of a program for the whole community contributes much. Easy new ports contribute little,
since someone else would surely do them; difficult ports such as adding a new CPU to the
GNU Compiler Collection contribute more; major new features or packages contribute the
most.
By establishing the idea that supporting further development is “the proper thing to
do” when distributing free software for a fee, we can assure a steady flow of resources into
making more free software.
Copyright c 1994 Free Software Foundation, Inc.
Verbatim copying and redistribution of this section is permitted
without royalty; alteration is not permitted.
 325

Option Index
gfortran’s command line options are indexed here without any initial ‘-’ or ‘--’. Where
an option has both positive and negative forms (such as -foption and -fno-option), relevant
entries in the manual are indexed under the most appropriate form; it may sometimes be
useful to look up both forms.

A fdec-math . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
A-predicate=answer . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 fdec-static . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
allow-invalid-boz . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 fdec-structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Apredicate=answer . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 fdefault-double-8 . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
fdefault-integer-8 . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
fdefault-real-10 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
B fdefault-real-16 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
fdefault-real-8 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
backslash . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
fdollar-ok . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
fdump-fortran-global . . . . . . . . . . . . . . . . . . . . . . . . 22
C fdump-fortran-optimized . . . . . . . . . . . . . . . . . . . . . 22
fdump-fortran-original . . . . . . . . . . . . . . . . . . . . . . 22
c-prototypes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
fdump-parse-tree . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
c-prototypes-external . . . . . . . . . . . . . . . . . . . . . . . 33
cpp . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 fexternal-blas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
C . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 ff2c . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
CC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 ffixed-form . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
ffixed-line-length-n . . . . . . . . . . . . . . . . . . . . . . . . 11
ffpe-summary=list . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
D ffpe-trap=list . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
dD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 ffree-form . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
dI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 ffree-line-length-n. . . . . . . . . . . . . . . . . . . . . . . . . . 11
dM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 fimplicit-none . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Dname . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 finit-character . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
Dname=definition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 finit-derived . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
dN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 finit-integer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
dU . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 finit-local-zero . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
finit-logical . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
finit-real . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
F finline-arg-packing. . . . . . . . . . . . . . . . . . . . . . . . . . 30
faggressive-function-elimination . . . . . . . . . . . 32 finline-matmul-limit . . . . . . . . . . . . . . . . . . . . . . . . 30
falign-commons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 finteger-4-integer-8 . . . . . . . . . . . . . . . . . . . . . . . . 12
fall-intrinsics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 fintrinsic-modules-path dir . . . . . . . . . . . . . . . . . 24
fallow-argument-mismatch . . . . . . . . . . . . . . . . . . . . . 9 fmax-array-constructor . . . . . . . . . . . . . . . . . . . . . . 28
fblas-matmul-limit . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 fmax-errors=n . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
fbounds-check . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 fmax-identifier-length=n . . . . . . . . . . . . . . . . . . . . 11
fcheck. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 fmax-stack-var-size. . . . . . . . . . . . . . . . . . . . . . . . . . 29
fcheck-array-temporaries . . . . . . . . . . . . . . . . . . . . 28 fmax-subrecord-length=length . . . . . . . . . . . . . . . 24
fcoarray . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 fmodule-private . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
fconvert=conversion. . . . . . . . . . . . . . . . . . . . . . . . . . 24 fno-automatic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
fcray-pointer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 fno-backtrace . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
fd-lines-as-code . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 fno-protect-parens . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
fd-lines-as-comments. . . . . . . . . . . . . . . . . . . . . . . . . . 9 fno-underscoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
fdebug-aux-vars . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 fopenacc . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
fdec . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 fopenmp . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
fdec-blank-format-item . . . . . . . . . . . . . . . . . . . . . . 10 fpack-derived . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
fdec-char-conversions . . . . . . . . . . . . . . . . . . . . . . . . 9 fpad-source . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
fdec-format-defaults . . . . . . . . . . . . . . . . . . . . . . . . 10 fpp . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
fdec-include . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 frange-check . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
fdec-intrinsic-ints. . . . . . . . . . . . . . . . . . . . . . . . . . 10 freal-4-real-10 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
326 The GNU Fortran Compiler

freal-4-real-16 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 S
freal-4-real-8 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 static-libgfortran . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
freal-8-real-10 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 static-libquadmath . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
freal-8-real-16 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 std=std option . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
freal-8-real-4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
frealloc-lhs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
frecord-marker=length . . . . . . . . . . . . . . . . . . . . . . . 24 T
frecursive . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
tail-call-workaround . . . . . . . . . . . . . . . . . . . . . . . . 28
frepack-arrays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
frontend-loop-interchange . . . . . . . . . . . . . . . . . . . 32
frontend-optimize . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 U
fsecond-underscore . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
fshort-enums . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 Uname . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
fsign-zero . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 undef . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
fstack-arrays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
fsyntax-only . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 W
ftest-forall-temp . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
fworking-directory . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 Waliasing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
Walign-commons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
Wall . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
Wampersand . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
H Warray-temporaries . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
H . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 Wc-binding-type . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
Wcharacter-truncation . . . . . . . . . . . . . . . . . . . . . . . 18
Wcompare-reals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
I Wconversion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
Idir . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 Wconversion-extra . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
idirafter dir . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 Wdo-subscript . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
imultilib dir . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 Werror. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
iprefix prefix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 Wextra. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
iquote dir . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 Wfrontend-loop-interchange. . . . . . . . . . . . . . . . . . 19
isysroot dir . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 Wfunction-elimination . . . . . . . . . . . . . . . . . . . . . . . 20
isystem dir . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 Wimplicit-interface. . . . . . . . . . . . . . . . . . . . . . . . . . 19
Wimplicit-procedure. . . . . . . . . . . . . . . . . . . . . . . . . . 19
Winteger-division . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
J Wintrinsic-shadow . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
Jdir . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 Wintrinsics-std . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
Wline-truncation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
Woverwrite-recursive . . . . . . . . . . . . . . . . . . . . . . . . 19
M Wpedantic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
Wreal-q-constant . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
Mdir . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 Wrealloc-lhs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
Wrealloc-lhs-all . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
Wsurprising . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
N Wtabs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
nostdinc . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 Wtargt-lifetime . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
Wundefined-do-loop . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
Wunderflow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
P Wunused-dummy-argument . . . . . . . . . . . . . . . . . . . . . . 20
pedantic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 Wunused-parameter . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
pedantic-errors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 Wuse-without-only . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
P . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 Wzerotrip . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
 327

Keyword Index

$ ANINT. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
$ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 ANY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122
area hyperbolic cosine . . . . . . . . . . . . . . . . . . . . . . . . 115
area hyperbolic sine . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
% area hyperbolic tangent . . . . . . . . . . . . . . . . . . . . . . . 129
argument list functions . . . . . . . . . . . . . . . . . . . . . . . . . 55
%LOC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
%REF . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 arguments, to program . . . . . 157, 192, 193, 194, 204
%VAL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 array descriptor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
array, add elements . . . . . . . . . . . . . . . . . . . . . . . . . . . 277
array, AND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201
& array, apply condition . . . . . . . . . . . . . . . . . . . . 119, 122
array, bounds checking . . . . . . . . . . . . . . . . . . . . . . . . . 27
& . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
array, change dimensions . . . . . . . . . . . . . . . . . . . . . . 259
array, combine arrays . . . . . . . . . . . . . . . . . . . . . . . . . 236
array, condition testing . . . . . . . . . . . . . . . . . . . 119, 122
_gfortran_set_args . . . . . . . . . . . . . . . . . . . . . . . . . . . 77 array, conditionally add elements . . . . . . . . . . . . . . 277
_gfortran_set_convert . . . . . . . . . . . . . . . . . . . . . . . 79 array, conditionally count elements . . . . . . . . . . . . 163
_gfortran_set_fpe . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80 array, conditionally multiply elements . . . . . . . . . 252
_gfortran_set_max_subrecord_length . . . . . . . . 80 array, contiguity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 214
_gfortran_set_options . . . . . . . . . . . . . . . . . . . . . . . 77 array, count elements . . . . . . . . . . . . . . . . . . . . . . . . . 271
_gfortran_set_record_marker . . . . . . . . . . . . . . . . 79 array, duplicate dimensions . . . . . . . . . . . . . . . . . . . 273
array, duplicate elements . . . . . . . . . . . . . . . . . . . . . . 273
array, element counting . . . . . . . . . . . . . . . . . . . . . . . 163
A array, gather elements . . . . . . . . . . . . . . . . . . . . . . . . 248
array, increase dimension . . . . . . . . . . . . . . . . . 273, 289
ABORT. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111
array, indices of type real . . . . . . . . . . . . . . . . . . . . . . 50
absolute value . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112
array, location of maximum element . . . . . . . . . . . 234
ABS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112
array, location of minimum element . . . . . . . . . . . 238
ACCESS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112
ACHAR. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113 array, lower bound . . . . . . . . . . . . . . . . . . . . . . . . . . . . 219
ACOS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114 array, maximum value . . . . . . . . . . . . . . . . . . . . . . . . 235
ACOSD. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114 array, merge arrays . . . . . . . . . . . . . . . . . . . . . . . . . . . 236
ACOSH. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115 array, minimum value . . . . . . . . . . . . . . . . . . . . . . . . . 239
adjust string . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116 array, multiply elements . . . . . . . . . . . . . . . . . . . . . . 252
ADJUSTL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116 array, number of elements . . . . . . . . . . . . . . . . 163, 271
ADJUSTR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116 array, OR. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203
AIMAG. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117 array, packing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 248
AINT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117 array, parity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212
ALARM. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118 array, permutation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164
ALGAMA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 227 array, product . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 252
aliasing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 array, reduce dimension . . . . . . . . . . . . . . . . . . . . . . . 248
alignment of COMMON blocks . . . . . . . . . . . . . . . . . 20, 31 array, rotate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164
all warnings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 array, scatter elements . . . . . . . . . . . . . . . . . . . . . . . . 289
ALL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119 array, shape . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 266
ALLOCATED . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120 array, shift . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173
allocation, moving . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242 array, shift circularly . . . . . . . . . . . . . . . . . . . . . . . . . . 164
allocation, status . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120 array, size . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 271
ALOG . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 226 array, sum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 277
ALOG10 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 227 array, transmogrify . . . . . . . . . . . . . . . . . . . . . . . . . . . 259
AMAX0. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 232 array, transpose . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285
AMAX1. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 232 array, unpacking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 289
AMIN0. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237 array, upper bound . . . . . . . . . . . . . . . . . . . . . . . . . . . 287
AMIN1. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237 array, XOR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212
AMOD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 240 ASCII collating sequence . . . . . . . . . . . . . . . . . 113, 201
AND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120 ASIN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122
328 The GNU Fortran Compiler

ASIND. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123 BESSEL_Y1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140

ASINH. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124 BESSEL_YN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141
ASSOCIATED . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124 BESY0. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140
association status . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124 BESY1. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140
association status, C pointer . . . . . . . . . . . . . . . . . . 144 BESYN. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141
assumed-rank . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74 BGE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142
assumed-type . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74 BGT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142
asynchronous I/O . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 BIAND. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 202
ATAN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
BIEOR. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208
ATAN2. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
binary representation . . . . . . . . . . . . . . . . . . . . . . . . . 250
ATAN2D . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128
BIOR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212
ATAND. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
ATANH. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129 bit intrinsics checking . . . . . . . . . . . . . . . . . . . . . . . . . . 27
Atomic subroutine, add . . . . . . . . . . . . . . . . . . . . . . . 129 BIT_SIZE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142
Atomic subroutine, ADD with fetch . . . . . . . . . . . 132 BITEST . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144
Atomic subroutine, AND. . . . . . . . . . . . . . . . . . . . . . 130 bits set . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 250
Atomic subroutine, AND with fetch . . . . . . . . . . . 133 bits, AND of array elements. . . . . . . . . . . . . . . . . . . 201
Atomic subroutine, compare and swap . . . . . . . . 131 bits, clear . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204
Atomic subroutine, define . . . . . . . . . . . . . . . . . . . . . 132 bits, extract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205
Atomic subroutine, OR . . . . . . . . . . . . . . . . . . . . . . . 135 bits, get . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205
Atomic subroutine, OR with fetch . . . . . . . . . . . . . 134 bits, merge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237
Atomic subroutine, reference . . . . . . . . . . . . . . . . . . 136 bits, move . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242, 285
Atomic subroutine, XOR . . . . . . . . . . . . . . . . . . . . . . 137 bits, negate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245
Atomic subroutine, XOR with fetch . . . . . . . . . . . 135 bits, number of . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142
ATOMIC_ADD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129 bits, OR of array elements . . . . . . . . . . . . . . . . . . . . 203
ATOMIC_AND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130 bits, set . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 206
ATOMIC_DEFINE . . . . . . . . . . . . . . . . . . . . . . . . . . . 131, 132 bits, shift . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 216
ATOMIC_FETCH_ADD . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132 bits, shift circular . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217
ATOMIC_FETCH_AND . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133
bits, shift left . . . . . . . . . . . . . . . . . . . . . . . . . . . . 228, 267
ATOMIC_FETCH_OR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
bits, shift right . . . . . . . . . . . . . . . . . . . . . . 260, 267, 268
ATOMIC_FETCH_XOR . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
bits, testing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144
ATOMIC_OR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
ATOMIC_REF . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136 bits, unset . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204
ATOMIC_XOR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137 bits, XOR of array elements . . . . . . . . . . . . . . . . . . . 212
Authors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 301 bitwise comparison . . . . . . . . . . . . . . . . . . . . . . . 142, 143
AUTOMATIC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 bitwise logical and . . . . . . . . . . . . . . . . . . . . . . . 120, 202
bitwise logical exclusive or . . . . . . . . . . . . . . . 208, 290
bitwise logical not . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245
B bitwise logical or . . . . . . . . . . . . . . . . . . . . . . . . . 212, 247
BABS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112 BJTEST . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144
backslash . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 BKTEST . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144
BACKSPACE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56 BLE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143
backtrace . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23, 138 BLT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143
BACKTRACE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138 BMOD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 240
base 10 logarithm function . . . . . . . . . . . . . . . . . . . . 227 BMVBITS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242
BBCLR. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204 BNOT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245
BBITS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205 bounds checking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
BBSET. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 206 BOZ literal constants . . . . . . . . . . . . . . . . . . . . . . . . . . 50
BBTEST . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144 BSHFT. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 216
BESJ0. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138 BSHFTC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217
BESJ1. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139 BTEST. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144
BESJN. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
Bessel function, first kind . . . . . . . . . . . . . . . . 138, 139
Bessel function, second kind . . . . . . . . . . . . . . 140, 141
BESSEL_J0 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138
BESSEL_J1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
BESSEL_JN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
BESSEL_Y0 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140
Keyword Index 329

C Coarray, gfortran caf event wait . . . . . . . . . . . . . 101

C derived type and struct interoperability . . . . . . 69 Coarray, gfortran caf fail image . . . . . . . . . . . . . . 104
C interoperability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69 Coarray, gfortran caf failed images . . . . . . . . . . . . 89
C intrinsic type interoperability . . . . . . . . . . . . . . . . 69 Coarray, gfortran caf finish . . . . . . . . . . . . . . . . . . . 88
C pointers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72 Coarray, gfortran caf get . . . . . . . . . . . . . . . . . . . . . . 93
C procedure interoperability . . . . . . . . . . . . . . . . . . . 70 Coarray, gfortran caf get by ref . . . . . . . . . . . . . . . 97
C variable interoperability . . . . . . . . . . . . . . . . . . . . . 70 Coarray, gfortran caf image status . . . . . . . . . . . . 89
C_ASSOCIATED . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144 Coarray, gfortran caf init . . . . . . . . . . . . . . . . . . . . . 88
C_F_POINTER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145 Coarray, gfortran caf is present . . . . . . . . . . . . . . . 92
C_F_PROCPOINTER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146 Coarray, gfortran caf lock . . . . . . . . . . . . . . . . . . . . . 99
C_FUNLOC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146 Coarray, gfortran caf num images . . . . . . . . . . . . . 89
C_LOC. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147 Coarray, gfortran caf register . . . . . . . . . . . . . . . . . 90
C_SIZEOF . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148 Coarray, gfortran caf send . . . . . . . . . . . . . . . . . . . . 92
CABS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112 Coarray, gfortran caf send by ref . . . . . . . . . . . . . 96
calling convention . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 Coarray, gfortran caf sendget . . . . . . . . . . . . . . . . . 94
CARRIAGECONTROL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 Coarray, gfortran caf sendget by ref . . . . . . . . . . 98
CCOS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160 Coarray, gfortran caf stopped images . . . . . . . . . 90
CCOSD. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160 Coarray, gfortran caf sync all . . . . . . . . . . . . . . . . 102
CDABS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112 Coarray, gfortran caf sync images. . . . . . . . . . . . 102
CDCOS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160 Coarray, gfortran caf sync memory . . . . . . . . . . 103
CDCOSD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160 Coarray, gfortran caf this image . . . . . . . . . . . . . . 88
CDEXP. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179 Coarray, gfortran caf unlock . . . . . . . . . . . . . . . . . 100
CDLOG. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 226 coarray, IMAGE_INDEX . . . . . . . . . . . . . . . . . . . . . . . . . 209
CDSIN. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 269 coarray, lower bound . . . . . . . . . . . . . . . . . . . . . . . . . . 220
CDSIND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 270 coarray, NUM_IMAGES . . . . . . . . . . . . . . . . . . . . . . . . . . 246
CDSQRT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 274 coarray, THIS_IMAGE . . . . . . . . . . . . . . . . . . . . . . . . . . 281
ceiling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121, 149 coarray, upper bound . . . . . . . . . . . . . . . . . . . . . . . . . 287
CEILING . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149 coarrays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
CEXP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179 Coarrays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
character kind . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 263 code generation, conventions . . . . . . . . . . . . . . . . . . . 25
character set . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 collating sequence, ASCII . . . . . . . . . . . . . . . . . 113, 201
CHAR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149 Collectives, generic reduction . . . . . . . . . . . . . . . . . 155
CHDIR. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150 Collectives, maximal value . . . . . . . . . . . . . . . . . . . . 153
checking array temporaries . . . . . . . . . . . . . . . . . . . . . 27 Collectives, minimal value . . . . . . . . . . . . . . . . . . . . . 154
checking subscripts. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 Collectives, sum of values . . . . . . . . . . . . . . . . . . . . . 156
CHMOD. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151 Collectives, value broadcasting . . . . . . . . . . . . . . . . 152
clock ticks . . . . . . . . . . . . . . . . . . . . . . . . . . . 235, 236, 279 command line. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 178
CLOG . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 226 command options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
CMPLX. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152 command-line arguments . . . 157, 192, 193, 194, 204
CO_BROADCAST . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152 command-line arguments, number of . . . . . 157, 204
CO_MAX . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153 COMMAND_ARGUMENT_COUNT . . . . . . . . . . . . . . . . . . . . . 157
CO_MIN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154 COMMON. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
CO_REDUCE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155 compiler flags inquiry function . . . . . . . . . . . . . . . . 157
CO_SUM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156 compiler, name and version . . . . . . . . . . . . . . . . . . . 158
Coarray, gfortran caf atomic cas . . . . . . . . . . . . . 105 COMPILER_OPTIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157
Coarray, gfortran caf atomic define . . . . . . . . . . 104 COMPILER_VERSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158
Coarray, gfortran caf atomic op . . . . . . . . . . . . . 106 complex conjugate . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159
Coarray, gfortran caf atomic ref . . . . . . . . . . . . . 104 Complex function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
Coarray, gfortran caf co broadcast . . . . . . . . . . . 106 complex numbers, conversion to . . . . . 152, 158, 167
Coarray, gfortran caf co max . . . . . . . . . . . . . . . . 107 complex numbers, imaginary part . . . . . . . . . . . . . 117
Coarray, gfortran caf co min . . . . . . . . . . . . . . . . . 107 complex numbers, real part . . . . . . . . . . . . . . 171, 257
Coarray, gfortran caf co reduce . . . . . . . . . . . . . . 108 COMPLEX . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158
Coarray, gfortran caf co sum . . . . . . . . . . . . . . . . 108 Conditional compilation . . . . . . . . . . . . . . . . . . . . . . . . 14
Coarray, gfortran caf deregister . . . . . . . . . . . . . . . 91 conditional compilation . . . . . . . . . . . . . . . . . . . . . . . . . 3
Coarray, gfortran caf error stop . . . . . . . . . . . . . . 103 CONJG. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159
Coarray, gfortran caf error stop str . . . . . . . . . . 104 consistency, durability . . . . . . . . . . . . . . . . . . . . . . . . . 43
Coarray, gfortran caf event post . . . . . . . . . . . . . 100 Contributing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 301
Coarray, gfortran caf event query . . . . . . . . . . . . 102 Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 301
330 The GNU Fortran Compiler

conversion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 DCONJG . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159

conversion, to character . . . . . . . . . . . . . . . . . . . 51, 149 DCOS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160
conversion, to complex . . . . . . . . . . . . . . 152, 158, 167 DCOSD. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160
conversion, to integer . . . . . . . 50, 201, 206, 210, 211 DCOSH. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161
conversion, to logical . . . . . . . . . . . . . . . . . . . . . . 50, 228 DCOTAN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162
conversion, to real. . . . . . . . . . . . . . . . . . . . . . . . 167, 257 DCOTAND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162
conversion, to string . . . . . . . . . . . . . . . . . . . . . . . . . . 165 DDIM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
CONVERT specifier . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 debugging information options . . . . . . . . . . . . . . . . . 22
core, dump . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111
debugging, preprocessor . . . . . . . . . . . . . . . . . . . . 14, 15
COS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160
DECODE. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
COSD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160
delayed execution . . . . . . . . . . . . . . . . . . . . . . . . 118, 273
COSH . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161
cosine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160 derived type interoperability with C . . . . . . . . . . . . 69
cosine, degrees . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160 DEXP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179
cosine, hyperbolic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161 DFLOAT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 257
cosine, hyperbolic, inverse . . . . . . . . . . . . . . . . . . . . . 115 DGAMMA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191
cosine, inverse . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114 dialect options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
cosine, inverse, degrees . . . . . . . . . . . . . . . . . . . . . . . . 114 DIGITS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 168
COTAND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162 DIM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
cotangent . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162 DIMAG. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
cotangent, degrees . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162 DINT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
COTAN. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162 directive, INCLUDE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
COUNT. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163 directory, options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
CPP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 directory, search paths for inclusion . . . . . . . . . . . . 23
CPU_TIME . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164 division, modulo . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241
Credits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 301 division, remainder . . . . . . . . . . . . . . . . . . . . . . . . . . . 240
CSHIFT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164 DLGAMA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 227
CSIN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 269 DLOG . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 226
CSIND. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 270
DLOG10 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 227
CSQRT. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 274
DMAX1. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 232
CTIME. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165
current date . . . . . . . . . . . . . . . . . . . . . . . . . 166, 181, 207 DMIN1. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237
current time . . . . . . . . . . . . . . . 166, 181, 218, 282, 283 DMOD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 240
DNINT. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
dope vector . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
D dot product . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
DOT_PRODUCT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
DABS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112
DPROD. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 170
DACOS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114
DACOSD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114 DREAL. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171
DACOSH . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115 DSHIFTL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171
DASIN. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122 DSHIFTR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172
DASIND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123 DSIGN. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 268
DASINH . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124 DSIN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 269
DATAN. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126 DSIND. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 270
DATAN2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127 DSINH. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 271
DATAN2D . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128 DSQRT. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 274
DATAND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126 DTAN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 280
DATANH . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129 DTAND. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 280
date, current . . . . . . . . . . . . . . . . . . . . . . . . 166, 181, 207 DTANH. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 281
DATE_AND_TIME . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 166 DTIME. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172
DBESJ0 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138 dummy argument, unused . . . . . . . . . . . . . . . . . . . . . . 20
DBESJ1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
DBESJN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
DBESY0 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140
DBESY1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140
DBESYN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141
DBLE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167
DCMPLX . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167
Keyword Index 331

E file system, hard link . . . . . . . . . . . . . . . . . . . . . . . . . . 223

elapsed time. . . . . . . . . . . . . . . . . . . . . . . . . 172, 262, 263 file system, remove file . . . . . . . . . . . . . . . . . . . . . . . . 288
Elimination of functions with file system, rename file . . . . . . . . . . . . . . . . . . . . . . . . 258
identical argument lists . . . . . . . . . . . . . . . . . . . . . . 32 file system, soft link. . . . . . . . . . . . . . . . . . . . . . . . . . . 278
ENCODE. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66 file, symbolic link . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
environment variable . . . . . . . . . . . . . . 33, 35, 195, 196 file, unformatted sequential . . . . . . . . . . . . . . . . . . . . 44
EOF . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56 findloc . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183
EOSHIFT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173 FINDLOC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183
EPSILON . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174 flags inquiry function . . . . . . . . . . . . . . . . . . . . . . . . . 157
ERF . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175 floating point, exponent . . . . . . . . . . . . . . . . . . . . . . . 180
ERFC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175 floating point, fraction . . . . . . . . . . . . . . . . . . . . . . . . 188
ERFC_SCALED . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 176 floating point, nearest different . . . . . . . . . . . . . . . . 243
error function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175 floating point, relative spacing . . . . . . . . . . . . 259, 273
error function, complementary . . . . . . . . . . . . . . . . 175 floating point, scale . . . . . . . . . . . . . . . . . . . . . . . . . . . 261
error function, complementary, floating point, set exponent . . . . . . . . . . . . . . . . . . . 265
exponentially-scaled . . . . . . . . . . . . . . . . . . . . . . . . 176 FLOAT. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 257
errors, limiting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 FLOATI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 257
escape characters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 FLOATJ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 257
ETIME. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 176 FLOATK . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 257
Euclidean distance . . . . . . . . . . . . . . . . . . . . . . . . . . . . 200 floor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117, 184
Euclidean vector norm . . . . . . . . . . . . . . . . . . . . . . . . 245 FLOOR. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184
EVENT_QUERY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177 FLUSH. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185
Events, EVENT QUERY . . . . . . . . . . . . . . . . . . . . . 177 FNUM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 186
EXECUTE_COMMAND_LINE . . . . . . . . . . . . . . . . . . . . . . . 178 form feed whitespace . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
EXIT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179 FORMAT. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
exponent . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 FPP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
exponential function . . . . . . . . . . . . . . . . . . . . . . . . . . 179 FPUT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 186
exponential function, inverse . . . . . . . . . . . . . 226, 227 FPUTC. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187
expression size . . . . . . . . . . . . . . . . . . . . . . . . . . . 148, 272 FRACTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 188
EXP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179 FREE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 188
EXPONENT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 180 Front-end optimization . . . . . . . . . . . . . . . . . . . . . . . . . 32
EXTENDS_TYPE_OF . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 180 FSEEK. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189
extensions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 FSTAT. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 190
extensions, implemented . . . . . . . . . . . . . . . . . . . . . . . 47 FTELL. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191
extensions, not implemented . . . . . . . . . . . . . . . . . . . 65 function elimination . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
extra warnings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 function interoperability with C . . . . . . . . . . . . . . . . 70
Further Interoperability of Fortran with C . . . . . . 74

F
f2c calling convention . . . . . . . . . . . . . . . . . . . . . . 25, 26 G
Factorial function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191 g77 calling convention . . . . . . . . . . . . . . . . . . . . . . 25, 26
FDATE. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181 Gamma function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191
FDL, GNU Free Documentation License . . . . . . 315 Gamma function, logarithm of . . . . . . . . . . . . . . . . 227
FGET . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 182 GAMMA. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191
FGETC. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183 GCC. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
file format, fixed . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9, 11 Generating C prototypes from
file format, free . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9, 11 external procedures . . . . . . . . . . . . . . . . . . . . . . . . . . 33
file operation, file number . . . . . . . . . . . . . . . . . . . . . 186 Generating C prototypes from Fortran
file operation, flush . . . . . . . . . . . . . . . . . . . . . . . . . . . 185 BIND(C) enteties . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
file operation, position . . . . . . . . . . . . . . . . . . . 189, 191 GERROR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192
file operation, read character . . . . . . . . . . . . . 182, 183 GET_COMMAND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193
file operation, seek . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189 GET_COMMAND_ARGUMENT . . . . . . . . . . . . . . . . . . . . . . . 194
file operation, write character . . . . . . . . . . . . 186, 187 GET_ENVIRONMENT_VARIABLE . . . . . . . . . . . . . . . . . . . 196
file system, access mode . . . . . . . . . . . . . . . . . . . . . . . 112 GETARG . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192
file system, change access mode . . . . . . . . . . . . . . . 151 GETCWD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195
file system, create link . . . . . . . . . . . . . . . . . . . 223, 278 GETENV . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195
file system, file creation mask . . . . . . . . . . . . . . . . . 288 GETGID . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197
file system, file status. . . . . . . . . . . . . . . . 190, 229, 275 GETLOG . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197
332 The GNU Fortran Compiler

GETPID . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 198 images, number of . . . . . . . . . . . . . . . . . . . . . . . . . . . . 246

GETUID . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 198 IMAGPART . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
GMTIME . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 198 IMOD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 240
GNU Compiler Collection . . . . . . . . . . . . . . . . . . . . . . . 2 IMVBITS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242
GNU Fortran command options . . . . . . . . . . . . . . . . . 7 INCLUDE directive . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
inclusion, directory search paths for . . . . . . . . . . . . 23
INDEX. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209
H INOT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245
Hollerith constants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50 input/output, asynchronous . . . . . . . . . . . . . . . . . . . . 45
HOSTNM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199 INT2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211
HUGE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 200 INT8 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211
hyperbolic cosine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161 integer kind. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 264
hyperbolic function, cosine . . . . . . . . . . . . . . . . . . . . 161 integer overflow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
hyperbolic function, cosine, inverse . . . . . . . . . . . . 115 Interoperability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
hyperbolic function, sine . . . . . . . . . . . . . . . . . . . . . . 271 interoperability with C . . . . . . . . . . . . . . . . . . . . . . . . . 69
hyperbolic function, sine, inverse . . . . . . . . . . . . . . 124 interoperability, derived type and struct . . . . . . . . 69
hyperbolic function, tangent . . . . . . . . . . . . . . . . . . 281 interoperability, intrinsic type . . . . . . . . . . . . . . . . . . 69
hyperbolic function, tangent, inverse . . . . . . . . . . 129 interoperability, subroutine and function . . . . . . . 70
hyperbolic sine. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 271 interoperability, variable . . . . . . . . . . . . . . . . . . . . . . . 70
hyperbolic tangent . . . . . . . . . . . . . . . . . . . . . . . . . . . . 281 intrinsic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
HYPOT. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 200 intrinsic Modules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293
intrinsic procedures . . . . . . . . . . . . . . . . . . . . . . . . . . . 111
intrinsic type interoperability with C . . . . . . . . . . . 69
I intrinsics, integer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
intrinsics, math . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
I/O item lists . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
intrinsics, trigonometric functions . . . . . . . . . . . . . . 62
I/O specifiers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
INT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210
IABS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112
inverse hyperbolic cosine . . . . . . . . . . . . . . . . . . . . . . 115
IACHAR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201
inverse hyperbolic sine . . . . . . . . . . . . . . . . . . . . . . . . 124
IALL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201
inverse hyperbolic tangent . . . . . . . . . . . . . . . . . . . . 129
IAND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 202
IOR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212
IANY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203
IOSTAT, end of file . . . . . . . . . . . . . . . . . . . . . . . . . . . . 214
IARGC. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204
IOSTAT, end of record . . . . . . . . . . . . . . . . . . . . . . . . . 215
IBCLR. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204
IPARITY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212
IBITS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205
IRAND. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213
IBSET. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 206
IS_IOSTAT_END . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 214
ICHAR. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 206
IS_IOSTAT_EOR . . . . . . . . . . . . . . . . . . . . . . . . . . . 214, 215
IDATE. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207
ISATTY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215
IDIM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
ISHFT. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 216
IDINT. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210
ISHFTC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217
IDNINT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 244
ISIGN. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 268
IEEE, ISNAN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217
ISNAN. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217
IEOR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208
ITIME. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 218
IERRNO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209
IFIX . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210
IIABS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112 J
IIAND. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 202
JIABS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112
IIBCLR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204
JIAND. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 202
IIBITS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205
JIBCLR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204
IIBSET . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 206
JIBITS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205
IIEOR. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208
JIBSET . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 206
IIOR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212
JIEOR. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208
IISHFT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 216
JIOR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212
IISHFTC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217
JISHFT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 216
IMAG . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
JISHFTC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217
IMAGE_INDEX . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209
JMOD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 240
images, cosubscript to image
JMVBITS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242
index conversion . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209
JNOT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245
images, index of this image . . . . . . . . . . . . . . . . . . . 281
Keyword Index 333

K logical exclusive or, bitwise . . . . . . . . . . . . . . . 208, 290

KIABS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112 logical not, bitwise . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245
KIAND. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 202 logical or, bitwise . . . . . . . . . . . . . . . . . . . . . . . . 212, 247
KIBCLR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204 logical, bitwise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
KIBITS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205 logical, variable representation . . . . . . . . . . . . . . . . . 41
KIBSET . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 206 LOG . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 226
KIEOR. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208 LOGICAL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 228
KILL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 218 login name. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197
kind . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41, 219 loop interchange, Fortran . . . . . . . . . . . . . . . . . . . . . . 32
kind, character. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 263 loop interchange, warning . . . . . . . . . . . . . . . . . . . . . . 19
kind, integer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 264 LSHIFT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 228
kind, old-style . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 LSTAT. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229
kind, real . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 264 LTIME. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 230
KIND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 219
KIOR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212
KISHFT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 216 M
KISHFTC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217 MALLOC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 230
KMOD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 240 MAP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
KMVBITS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242 mask, left justified . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231
KNOT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245 mask, right justified. . . . . . . . . . . . . . . . . . . . . . . . . . . 232
MASKL. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231
MASKR. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 232
L MATMUL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 232
L2 vector norm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245 matrix multiplication . . . . . . . . . . . . . . . . . . . . . . . . . 232
language, dialect options . . . . . . . . . . . . . . . . . . . . . . . . 8 matrix, transpose . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285
LBOUND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 219 MAX, MIN, NaN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
LCOBOUND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220 MAX0 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 232
LEADZ. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220 MAX1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 232
left shift, combined . . . . . . . . . . . . . . . . . . . . . . . . . . . 171 MAXEXPONENT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233
LEN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 221 maximum value . . . . . . . . . . . . . . . . . . . . . . . . . . 232, 235
LEN_TRIM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 222 MAX . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 232
lexical comparison of strings . . . . 222, 223, 224, 225 MAXLOC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 234
LGAMMA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 227 MAXVAL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 235
LGE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 222 MCLOCK . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 235
LGT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223 MCLOCK8 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 236
libf2c calling convention . . . . . . . . . . . . . . . . . . . . 25, 26 memory checking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
libgfortran initialization, set args . . . . . . . . . . . . . . . 77 MERGE. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 236
libgfortran initialization, set convert. . . . . . . . . . . . 79 MERGE_BITS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237
libgfortran initialization, set fpe . . . . . . . . . . . . . . . . 80 messages, error . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
libgfortran initialization, messages, warning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
set max subrecord length . . . . . . . . . . . . . . . . . . . . 80 MIN0 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237
libgfortran initialization, set options. . . . . . . . . . . . 77 MIN1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237
libgfortran initialization, set record marker . . . . . 79 MINEXPONENT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 238
limits, largest number . . . . . . . . . . . . . . . . . . . . . . . . . 200 minimum value . . . . . . . . . . . . . . . . . . . . . . . . . . 237, 239
limits, smallest number . . . . . . . . . . . . . . . . . . . . . . . 284 MIN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237
linking, static . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 MINLOC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 238
LINK . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223 MINVAL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 239
LLE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 224 Mixed-language programming . . . . . . . . . . . . . . . . . . 69
LLT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 225 model representation, base . . . . . . . . . . . . . . . . . . . . 252
LNBLNK . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 225 model representation, epsilon . . . . . . . . . . . . . . . . . 174
location of a variable in memory . . . . . . . . . . . . . . 226 model representation, largest number. . . . . . . . . . 200
LOC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63, 226 model representation, maximum exponent . . . . . 233
LOG_GAMMA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 227 model representation, minimum exponent . . . . . 238
LOG10. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 227 model representation, precision . . . . . . . . . . . . . . . . 251
logarithm function . . . . . . . . . . . . . . . . . . . . . . . . . . . . 226 model representation, radix . . . . . . . . . . . . . . . . . . . 252
logarithm function with base 10 . . . . . . . . . . . . . . . 227 model representation, range . . . . . . . . . . . . . . . . . . . 256
logarithm function, inverse . . . . . . . . . . . . . . . . . . . . 179 model representation, significant digits . . . . . . . . 168
logical and, bitwise . . . . . . . . . . . . . . . . . . . . . . . 120, 202 model representation, smallest number . . . . . . . . 284
334 The GNU Fortran Compiler

module entities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 P
module search path. . . . . . . . . . . . . . . . . . . . . . . . . 23, 24 PACK . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 248
modulo . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241 PARAMETER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
MOD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 240 Parity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249
MODULO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241 parity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 250
MOVE_ALLOC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242 PARITY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249
moving allocation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242 paths, search . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23, 24
multiply array elements . . . . . . . . . . . . . . . . . . . . . . . 252 PERROR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249
MVBITS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242
pointer checking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
pointer, C address of pointers . . . . . . . . . . . . . . . . . 146
pointer, C address of procedures . . . . . . . . . . . . . . 146
N pointer, C association status . . . . . . . . . . . . . . . . . . 144
Namelist . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 pointer, convert C to Fortran . . . . . . . . . . . . . . . . . 145
NAME . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67 pointer, Cray . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
natural logarithm function . . . . . . . . . . . . . . . . . . . . 226 pointer, cray . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 188, 230
NEAREST . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243 pointer, disassociated . . . . . . . . . . . . . . . . . . . . . . . . . 246
NEW_LINE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243 pointer, status . . . . . . . . . . . . . . . . . . . . . . . . . . . 124, 246
newline . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243 pointers, C. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72
NINT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 244 POPCNT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 250
norm, Euclidean . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245 POPPAR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 250
NORM2. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245 positive difference . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
NOSHARED . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 PRECISION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 251
NOT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245 Preprocessing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
NULL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 246 preprocessing, assertion . . . . . . . . . . . . . . . . . . . . . . . . 16
NUM_IMAGES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 246 preprocessing, define macros . . . . . . . . . . . . . . . . . . . 16
preprocessing, include path. . . . . . . . . . . . . . . . . . . . . 15
preprocessing, keep comments . . . . . . . . . . . . . . . . . . 16
preprocessing, no linemarkers . . . . . . . . . . . . . . . . . . 17
O preprocessing, undefine macros . . . . . . . . . . . . . . . . . 17
open, action . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 preprocessor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
OpenACC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11, 55 preprocessor, debugging . . . . . . . . . . . . . . . . . . . . 14, 15
OpenMP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11, 54 preprocessor, disable . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
operators, unary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50 preprocessor, enable. . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
operators, xor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63 preprocessor, include file handling . . . . . . . . . . . . . . 14
options inquiry function . . . . . . . . . . . . . . . . . . . . . . . 157 preprocessor, working directory . . . . . . . . . . . . . . . . 15
options, code generation . . . . . . . . . . . . . . . . . . . . . . . 25 PRESENT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 251
options, debugging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 private . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
options, dialect. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 procedure interoperability with C . . . . . . . . . . . . . . 70
options, directory search . . . . . . . . . . . . . . . . . . . . . . . 23 procedure pointer, convert C to Fortran . . . . . . . 147
options, errors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 process ID . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 198
options, Fortran dialect . . . . . . . . . . . . . . . . . . . . . . . . . 9 product, double-precision . . . . . . . . . . . . . . . . . . . . . 170
options, gfortran command . . . . . . . . . . . . . . . . . . . . . 7 product, matrix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 232
options, linking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 product, vector . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
options, negative forms . . . . . . . . . . . . . . . . . . . . . . . . . . 7 PRODUCT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 252
options, preprocessor . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 program termination . . . . . . . . . . . . . . . . . . . . . . . . . . 179
options, real kind type promotion . . . . . . . . . . . . . . 13 program termination, with core dump . . . . . . . . . 111
options, run-time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
options, runtime . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
options, warnings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
OR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 247
Q
output, newline . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243 Q edit descriptor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
overflow handling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 Q exponent-letter. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
Keyword Index 335

R seeding a random number generator . . . . . . 255, 275

radix, real . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 264 SELECTED_CHAR_KIND . . . . . . . . . . . . . . . . . . . . . . . . . . 263
RADIX. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 252 SELECTED_INT_KIND . . . . . . . . . . . . . . . . . . . . . . . . . . . 264
random number generation . . . . . . . . . . 213, 253, 255 SELECTED_REAL_KIND . . . . . . . . . . . . . . . . . . . . . . . . . . 264
random number generation, initialization . . . . . . 254 sequential, unformatted . . . . . . . . . . . . . . . . . . . . . . . . 44
random number generation, seeding . . . . . . 255, 275 SET_EXPONENT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 265
range checking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 SHAPE. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 266
rank . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 257 SHARE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
RAN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 253 SHARED. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
RAND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 253 shift, left . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171, 267
RANDOM_INIT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 254 shift, right . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172, 268
RANDOM_NUMBER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 255 shift, right with fill. . . . . . . . . . . . . . . . . . . . . . . . . . . . 267
RANDOM_SEED . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 255 SHIFTA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267
RANGE. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 256 SHIFTL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267
RANK . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 257 SHIFTR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 268
re-association of parenthesized expressions . . . . . . 32 sign copying . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 268
read character, stream mode . . . . . . . . . . . . . 182, 183 SIGN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 268
READONLY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 SIGNAL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 269
real kind . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 264 SIND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 270
real number, exponent . . . . . . . . . . . . . . . . . . . . . . . . 180 sine. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 269
real number, fraction. . . . . . . . . . . . . . . . . . . . . . . . . . 188 sine, degrees . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 270
real number, nearest different . . . . . . . . . . . . . . . . . 243 sine, hyperbolic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 271
real number, relative spacing . . . . . . . . . . . . . 259, 273 sine, hyperbolic, inverse . . . . . . . . . . . . . . . . . . . . . . . 124
real number, scale . . . . . . . . . . . . . . . . . . . . . . . . . . . . 261 sine, inverse . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122
real number, set exponent . . . . . . . . . . . . . . . . . . . . . 265 sine, inverse, degrees . . . . . . . . . . . . . . . . . . . . . . . . . . 123
Reallocate the LHS in assignments . . . . . . . . . . . . . 32 SIN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 269
Reallocate the LHS in SINH . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 271
assignments, notification . . . . . . . . . . . . . . . . . . . . . 21 size of a variable, in bits . . . . . . . . . . . . . . . . . . . . . . 142
REAL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 257 size of an expression . . . . . . . . . . . . . . . . . . . . . 148, 272
REALPART . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 257 SIZE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 271
record marker . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 SIZEOF . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 272
RECORD. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56 SLEEP. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 273
Reduction, XOR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249 SNGL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 257
remainder . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 240 SPACING . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 273
RENAME . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 258 SPREAD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 273
repacking arrays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 SQRT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 274
REPEAT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 258 square-root . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 274
RESHAPE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 259 SRAND. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 275
REWIND. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56 Standards . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
right shift, combined . . . . . . . . . . . . . . . . . . . . . . . . . . 172 statement, SAVE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
root . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 274 STAT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 275
rounding, ceiling . . . . . . . . . . . . . . . . . . . . . . . . . 121, 149 STATIC. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
rounding, floor . . . . . . . . . . . . . . . . . . . . . . . . . . . 117, 184 storage size . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 276
rounding, nearest whole number. . . . . . . . . . . . . . . 244 STORAGE_SIZE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 276
RRSPACING . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 259 stream mode, read character . . . . . . . . . . . . . 182, 183
RSHIFT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 260 stream mode, write character . . . . . . . . . . . . . 186, 187
run-time checking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 string, adjust left . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
string, adjust right . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
string, comparison . . . . . . . . . . . . . . 222, 223, 224, 225
S string, concatenate . . . . . . . . . . . . . . . . . . . . . . . . . . . . 258
SAME_TYPE_AS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 260 string, find missing set . . . . . . . . . . . . . . . . . . . . . . . . 289
SAVE statement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 string, find non-blank character . . . . . . . . . . . . . . . 225
SCALE. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 261 string, find subset . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 261
SCAN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 261 string, find substring . . . . . . . . . . . . . . . . . . . . . . . . . . 209
search path . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 string, length . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 221
search paths, for included files . . . . . . . . . . . . . . . . . . 23 string, length, without trailing whitespace . . . . . 222
SECNDS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 262 string, remove trailing whitespace . . . . . . . . . . . . . 286
SECOND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 263 string, repeat . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 258
336 The GNU Fortran Compiler

strings, varying length . . . . . . . . . . . . . . . . . . . . . . . . . . 3 TRANSPOSE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285

structure packing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 trigonometric function, cosine . . . . . . . . . . . . . . . . . 160
STRUCTURE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56 trigonometric function, cosine, degrees . . . . . . . . 160
subrecord . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 trigonometric function, cosine, inverse . . . . . . . . . 114
subroutine interoperability with C . . . . . . . . . . . . . . 70 trigonometric function, cosine,
subscript checking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 inverse, degrees . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114
substring position . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209 trigonometric function, cotangent . . . . . . . . . . . . . 162
sum array elements . . . . . . . . . . . . . . . . . . . . . . . . . . . 277 trigonometric function, cotangent, degrees . . . . . 162
SUM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 277 trigonometric function, sine . . . . . . . . . . . . . . . . . . . 269
suppressing warnings . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
trigonometric function, sine, degrees. . . . . . . . . . . 270
symbol names. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
trigonometric function, sine, inverse . . . . . . . . . . . 122
symbol names, transforming . . . . . . . . . . . . . . . . . . . . 26
trigonometric function, sine, inverse, degrees . . 123
symbol names, underscores . . . . . . . . . . . . . . . . . . . . . 26
SYMLNK . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 278 trigonometric function, tangent . . . . . . . . . . . . . . . 280
syntax checking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 trigonometric function, tangent, degrees . . . . . . . 280
system, error handling . . . . . . . . . . . . . . . 192, 209, 249 trigonometric function, tangent, inverse. . . 126, 127
system, group ID . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197 trigonometric function, tangent,
system, host name . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199 inverse, degrees . . . . . . . . . . . . . . . . . . . . . . . . 126, 128
system, login name . . . . . . . . . . . . . . . . . . . . . . . . . . . 197 TRIM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 286
system, process ID . . . . . . . . . . . . . . . . . . . . . . . . . . . . 198 TS 29113 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
system, signal handling . . . . . . . . . . . . . . . . . . . . . . . 269 TTYNAM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 286
system, system call . . . . . . . . . . . . . . . . . . . . . . . 178, 278 type alias print . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
system, terminal . . . . . . . . . . . . . . . . . . . . . . . . . 215, 286 type cast . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285
system, user ID . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 198
system, working directory . . . . . . . . . . . . . . . . 150, 195
SYSTEM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 278
SYSTEM_CLOCK . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 279 U
UBOUND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 287
T UCOBOUND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 287
UMASK. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 288
tabulators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
underflow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
tangent . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 280
underscore . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
tangent, degrees . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 280
tangent, hyperbolic . . . . . . . . . . . . . . . . . . . . . . . . . . . 281 unformatted sequential . . . . . . . . . . . . . . . . . . . . . . . . . 44
tangent, hyperbolic, inverse . . . . . . . . . . . . . . . . . . . 129 UNION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
tangent, inverse . . . . . . . . . . . . . . . . . . . . . . . . . . 126, 127 UNLINK . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 288
tangent, inverse, degrees . . . . . . . . . . . . . . . . . 126, 128 UNPACK . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 289
TAN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 280 unused dummy argument . . . . . . . . . . . . . . . . . . . . . . 20
TAND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 280 unused parameter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
TANH . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 281 user id . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 198
terminate program . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179
terminate program, with core dump . . . . . . . . . . . 111
THIS_IMAGE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 281
thread-safety, threads . . . . . . . . . . . . . . . . . . . . . . . . . . 42 V
time, clock ticks . . . . . . . . . . . . . . . . . . . . . 235, 236, 279 variable attributes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
time, conversion to GMT info . . . . . . . . . . . . . . . . . 198
variable interoperability with C . . . . . . . . . . . . . . . . 70
time, conversion to local time info . . . . . . . . . . . . . 230
Varying length strings . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
time, conversion to string . . . . . . . . . . . . . . . . . . . . . 165
vector product . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
time, current . . . . . . . . . . . . . . 166, 181, 218, 282, 283
time, elapsed . . . . . . . . . . . . . . 164, 172, 176, 262, 263 VERIFY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 289
TIME . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 282 version of the compiler . . . . . . . . . . . . . . . . . . . . . . . . 158
TIME8. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 283 VOLATILE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
TINY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 284
trace. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
TRAILZ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 284
TRANSFER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285
transforming symbol names . . . . . . . . . . . . . . . . . . . . 26
transpose . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285
Keyword Index 337

W warnings, tabs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
warnings, to errors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
warning, C binding type . . . . . . . . . . . . . . . . . . . . . . . 18
warnings, undefined do loop . . . . . . . . . . . . . . . . . . . . 20
warnings, aliasing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
warnings, underflow . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
warnings, alignment of COMMON blocks . . . . . . . . . . . 20
warnings, unused dummy argument . . . . . . . . . . . . 20
warnings, all . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
warnings, unused parameter . . . . . . . . . . . . . . . . . . . . 20
warnings, ampersand . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 warnings, use statements . . . . . . . . . . . . . . . . . . . . . . . 20
warnings, array temporaries . . . . . . . . . . . . . . . . . . . . 18 write character, stream mode . . . . . . . . . . . . . 186, 187
warnings, character truncation . . . . . . . . . . . . . . . . . 18
warnings, conversion . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
warnings, division of integers . . . . . . . . . . . . . . . . . . . 19 X
warnings, extra . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
XOR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 290
warnings, function elimination. . . . . . . . . . . . . . . . . . 20 XOR reduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249
warnings, implicit interface . . . . . . . . . . . . . . . . . . . . . 19
warnings, implicit procedure . . . . . . . . . . . . . . . . . . . 19
warnings, integer division . . . . . . . . . . . . . . . . . . . . . . 19 Z
warnings, intrinsic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 ZABS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112
warnings, intrinsics of other standards . . . . . . . . . . 19 ZCOS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160
warnings, line truncation . . . . . . . . . . . . . . . . . . . . . . . 18 ZCOSD. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160
warnings, loop interchange . . . . . . . . . . . . . . . . . . . . . 19 zero bits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220, 284
warnings, non-standard intrinsics . . . . . . . . . . . . . . . 19 ZEXP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179
warnings, overwrite recursive . . . . . . . . . . . . . . . . . . . 19 ZLOG . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 226
warnings, q exponent-letter . . . . . . . . . . . . . . . . . . . . 19 ZSIN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 269
warnings, suppressing . . . . . . . . . . . . . . . . . . . . . . . . . . 17 ZSIND. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 270
warnings, suspicious code . . . . . . . . . . . . . . . . . . . . . . 19 ZSQRT. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 274